Copyright (c) 2025 Advanced Micro Devices, Inc., or its affiliates. AMD Public Use. MIT 2025-07-24 1.1.1 AMD CDNA 4 3 ENC_SOP1 32 11111111100000001111111100000000 10111110100000000000000000000000 10111110100000000000000100000000 10111110100000000000001000000000 10111110100000000000001100000000 10111110100000000000010000000000 10111110100000000000010100000000 10111110100000000000011000000000 10111110100000000000011100000000 10111110100000000000100000000000 10111110100000000000100100000000 10111110100000000000101000000000 10111110100000000000101100000000 10111110100000000000110000000000 10111110100000000000110100000000 10111110100000000000111000000000 10111110100000000000111100000000 10111110100000000001000000000000 10111110100000000001000100000000 10111110100000000001001000000000 10111110100000000001001100000000 10111110100000000001010000000000 10111110100000000001010100000000 10111110100000000001011000000000 10111110100000000001011100000000 10111110100000000001100000000000 10111110100000000001100100000000 10111110100000000001101000000000 10111110100000000001101100000000 10111110100000000001110000000000 10111110100000000001110100000000 10111110100000000001111000000000 10111110100000000001111100000000 10111110100000000010000000000000 10111110100000000010000100000000 10111110100000000010001000000000 10111110100000000010001100000000 10111110100000000010010000000000 10111110100000000010010100000000 10111110100000000010011000000000 10111110100000000010011100000000 10111110100000000010100000000000 10111110100000000010100100000000 10111110100000000010101000000000 10111110100000000010101100000000 10111110100000000010110000000000 10111110100000000010110100000000 10111110100000000010111000000000 10111110100000000011000000000000 10111110100000000011001000000000 10111110100000000011001100000000 10111110100000000011010000000000 10111110100000000011010100000000 10111110100000000011011000000000 10111110100000000011011100000000 default != .fieldderef .fieldderef INST SOP1 SSRC0 I 32 255 I 32 U 1 Scalar ALU operations with one destination and one source. ENCODING Encoding. 9 23 OP Opcode. 8 8 SDST Destination for instruction. 7 16 SSRC0 Operand for instruction. 8 0 ENC_SOPC 32 11111111111111110000000000000000 10111111000000000000000000000000 10111111000000010000000000000000 10111111000000100000000000000000 10111111000000110000000000000000 10111111000001000000000000000000 10111111000001010000000000000000 10111111000001100000000000000000 10111111000001110000000000000000 10111111000010000000000000000000 10111111000010010000000000000000 10111111000010100000000000000000 10111111000010110000000000000000 10111111000011000000000000000000 10111111000011010000000000000000 10111111000011100000000000000000 10111111000011110000000000000000 10111111000100000000000000000000 10111111000100010000000000000000 10111111000100100000000000000000 10111111000100110000000000000000 default && != .fieldderef .fieldderef INST SOPC SSRC0 I 32 255 I 32 U 1 != .fieldderef .fieldderef INST SOPC SSRC1 I 32 255 I 32 U 1 U 1 Scalar ALU comparison operations with two sources. ENCODING Encoding. 9 23 OP Opcode. 7 16 SSRC0 First operand for instruction. 8 0 SSRC1 Second operand for instruction. 8 8 ENC_SOPP 32 11111111111111110000000000000000 10111111100000000000000000000000 10111111100000010000000000000000 10111111100000100000000000000000 10111111100000110000000000000000 10111111100001000000000000000000 10111111100001010000000000000000 10111111100001100000000000000000 10111111100001110000000000000000 10111111100010000000000000000000 10111111100010010000000000000000 10111111100010100000000000000000 10111111100010110000000000000000 10111111100011000000000000000000 10111111100011010000000000000000 10111111100011100000000000000000 10111111100011110000000000000000 10111111100100000000000000000000 10111111100100010000000000000000 10111111100100100000000000000000 10111111100100110000000000000000 10111111100101000000000000000000 10111111100101010000000000000000 10111111100101100000000000000000 10111111100101110000000000000000 10111111100110000000000000000000 10111111100110010000000000000000 10111111100110100000000000000000 10111111100110110000000000000000 10111111100111000000000000000000 10111111100111010000000000000000 10111111100111100000000000000000 10111111100111110000000000000000 default U 1 Scalar ALU operations for control flow. ENCODING Encoding. 9 23 OP Opcode. 7 16 SIMM16 16-bit integer input for opcode. Signedness is determined by opcode. 16 0 ENC_SOPK 32 11111111100000000000000000000000 10110000000000000000000000000000 10110000100000000000000000000000 10110001000000000000000000000000 10110001100000000000000000000000 10110010000000000000000000000000 10110010100000000000000000000000 10110011000000000000000000000000 10110011100000000000000000000000 10110100000000000000000000000000 10110100100000000000000000000000 10110101000000000000000000000000 10110101100000000000000000000000 10110110000000000000000000000000 10110110100000000000000000000000 10110111000000000000000000000000 10110111100000000000000000000000 10111000000000000000000000000000 10111000100000000000000000000000 10111001000000000000000000000000 10111010100000000000000000000000 default U 1 Scalar ALU operations with one destination and one 16-bit constant source. ENCODING Encoding. 4 28 OP Opcode. 5 23 SDST Destination for instruction. 7 16 SIMM16 16-bit integer input for opcode. Signedness is determined by opcode. 16 0 ENC_SOP2 32 11111111100000000000000000000000 10000000000000000000000000000000 10000000100000000000000000000000 10000001000000000000000000000000 10000001100000000000000000000000 10000010000000000000000000000000 10000010100000000000000000000000 10000011000000000000000000000000 10000011100000000000000000000000 10000100000000000000000000000000 10000100100000000000000000000000 10000101000000000000000000000000 10000101100000000000000000000000 10000110000000000000000000000000 10000110100000000000000000000000 10000111000000000000000000000000 10000111100000000000000000000000 10001000000000000000000000000000 10001000100000000000000000000000 10001001000000000000000000000000 10001001100000000000000000000000 10001010000000000000000000000000 10001010100000000000000000000000 10001011000000000000000000000000 10001011100000000000000000000000 10001100000000000000000000000000 10001100100000000000000000000000 10001101000000000000000000000000 10001101100000000000000000000000 10001110000000000000000000000000 10001110100000000000000000000000 10001111000000000000000000000000 10001111100000000000000000000000 10010000000000000000000000000000 10010000100000000000000000000000 10010001000000000000000000000000 10010001100000000000000000000000 10010010000000000000000000000000 10010010100000000000000000000000 10010011000000000000000000000000 10010011100000000000000000000000 10010100000000000000000000000000 10010100100000000000000000000000 10010101000000000000000000000000 10010101100000000000000000000000 10010110000000000000000000000000 10010110100000000000000000000000 10010111000000000000000000000000 10010111100000000000000000000000 10011000000000000000000000000000 10011000100000000000000000000000 10011001000000000000000000000000 10011001100000000000000000000000 10011010000000000000000000000000 default && != .fieldderef .fieldderef INST SOP2 SSRC0 I 32 255 I 32 U 1 != .fieldderef .fieldderef INST SOP2 SSRC1 I 32 255 I 32 U 1 U 1 Scalar ALU operations with one destination and two sources. ENCODING Encoding. 2 30 OP Opcode. 7 23 SDST Destination for instruction. 7 16 SSRC0 First operand for instruction. 8 0 SSRC1 Second operand for instruction. 8 8 ENC_SMEM 64 0000000000000000000000000000000011111111111111000000000000000000 0000000000000000000000000000000011000000000000000000000000000000 0000000000000000000000000000000011000000000001000000000000000000 0000000000000000000000000000000011000000000010000000000000000000 0000000000000000000000000000000011000000000011000000000000000000 0000000000000000000000000000000011000000000100000000000000000000 0000000000000000000000000000000011000000000101000000000000000000 0000000000000000000000000000000011000000000110000000000000000000 0000000000000000000000000000000011000000000111000000000000000000 0000000000000000000000000000000011000000001000000000000000000000 0000000000000000000000000000000011000000001001000000000000000000 0000000000000000000000000000000011000000001010000000000000000000 0000000000000000000000000000000011000000001011000000000000000000 0000000000000000000000000000000011000000001100000000000000000000 0000000000000000000000000000000011000000010000000000000000000000 0000000000000000000000000000000011000000010001000000000000000000 0000000000000000000000000000000011000000010010000000000000000000 0000000000000000000000000000000011000000010101000000000000000000 0000000000000000000000000000000011000000010110000000000000000000 0000000000000000000000000000000011000000010111000000000000000000 0000000000000000000000000000000011000000011000000000000000000000 0000000000000000000000000000000011000000011001000000000000000000 0000000000000000000000000000000011000000011010000000000000000000 0000000000000000000000000000000011000000100000000000000000000000 0000000000000000000000000000000011000000100001000000000000000000 0000000000000000000000000000000011000000100010000000000000000000 0000000000000000000000000000000011000000100011000000000000000000 0000000000000000000000000000000011000000100100000000000000000000 0000000000000000000000000000000011000000100101000000000000000000 0000000000000000000000000000000011000000100110000000000000000000 0000000000000000000000000000000011000000100111000000000000000000 0000000000000000000000000000000011000000101000000000000000000000 0000000000000000000000000000000011000000101001000000000000000000 0000000000000000000000000000000011000001000000000000000000000000 0000000000000000000000000000000011000001000001000000000000000000 0000000000000000000000000000000011000001000010000000000000000000 0000000000000000000000000000000011000001000011000000000000000000 0000000000000000000000000000000011000001000100000000000000000000 0000000000000000000000000000000011000001000101000000000000000000 0000000000000000000000000000000011000001000110000000000000000000 0000000000000000000000000000000011000001000111000000000000000000 0000000000000000000000000000000011000001001000000000000000000000 0000000000000000000000000000000011000001001001000000000000000000 0000000000000000000000000000000011000001001010000000000000000000 0000000000000000000000000000000011000001001011000000000000000000 0000000000000000000000000000000011000001001100000000000000000000 0000000000000000000000000000000011000001100000000000000000000000 0000000000000000000000000000000011000001100001000000000000000000 0000000000000000000000000000000011000001100010000000000000000000 0000000000000000000000000000000011000001100011000000000000000000 0000000000000000000000000000000011000001100100000000000000000000 0000000000000000000000000000000011000001100101000000000000000000 0000000000000000000000000000000011000001100110000000000000000000 0000000000000000000000000000000011000001100111000000000000000000 0000000000000000000000000000000011000001101000000000000000000000 0000000000000000000000000000000011000001101001000000000000000000 0000000000000000000000000000000011000001101010000000000000000000 0000000000000000000000000000000011000001101011000000000000000000 0000000000000000000000000000000011000001101100000000000000000000 0000000000000000000000000000000011000010000000000000000000000000 0000000000000000000000000000000011000010000001000000000000000000 0000000000000000000000000000000011000010000010000000000000000000 0000000000000000000000000000000011000010000011000000000000000000 0000000000000000000000000000000011000010000100000000000000000000 0000000000000000000000000000000011000010000101000000000000000000 0000000000000000000000000000000011000010000110000000000000000000 0000000000000000000000000000000011000010000111000000000000000000 0000000000000000000000000000000011000010001000000000000000000000 0000000000000000000000000000000011000010001001000000000000000000 0000000000000000000000000000000011000010001010000000000000000000 0000000000000000000000000000000011000010001011000000000000000000 0000000000000000000000000000000011000010001100000000000000000000 0000000000000000000000000000000011000010100000000000000000000000 0000000000000000000000000000000011000010100001000000000000000000 0000000000000000000000000000000011000010100010000000000000000000 0000000000000000000000000000000011000010100011000000000000000000 0000000000000000000000000000000011000010100100000000000000000000 0000000000000000000000000000000011000010100101000000000000000000 0000000000000000000000000000000011000010100110000000000000000000 0000000000000000000000000000000011000010100111000000000000000000 0000000000000000000000000000000011000010101000000000000000000000 0000000000000000000000000000000011000010101001000000000000000000 0000000000000000000000000000000011000010101010000000000000000000 0000000000000000000000000000000011000010101011000000000000000000 0000000000000000000000000000000011000010101100000000000000000000 default U 1 Scalar memory operations. ENCODING Encoding. 6 26 GLC If set, operation is globally coherent. 1 16 IMM Boolean. Specifies whether OFFSET field specifies a SGPR (false) or an inline constant offset (true). 1 17 NV If set, operation operates on non-volatile data. 1 15 OFFSET If IMM = 0 and SOFFSET_EN = 0: Specifies an SGPR address that supplies an UNSIGNED BYTE offset aligned to one DWORD for the memory operation (see enumeration). If IMM = 1: specifies a 21-bit SIGNED BYTE offset aligned to one DWORD. 21 32 OP Opcode. 8 18 SBASE Bits [6:1] of an aligned pair of SGPRs specifying {size[16], base[48]}, where base and size are in dword units. The low-order bits are in the first SGPR. 6 0 1 0 SDATA Destination for instruction. 7 6 SOFFSET If SOFFSET_EN = 1: Specifies an SGPR address that supplies an UNSIGNED BYTE offset aligned to one DWORD for the memory operation (see enumeration). 7 57 SOFFSET_EN If set, SOFFSET field is used. 1 14 ENC_VOP1 32 11111110000000011111111000000000 01111110000000000000000000000000 01111110000000000000001000000000 01111110000000000000010000000000 01111110000000000000011000000000 01111110000000000000100000000000 01111110000000000000101000000000 01111110000000000000110000000000 01111110000000000000111000000000 01111110000000000001000000000000 01111110000000000001010000000000 01111110000000000001011000000000 01111110000000000001100000000000 01111110000000000001101000000000 01111110000000000001110000000000 01111110000000000001111000000000 01111110000000000010000000000000 01111110000000000010001000000000 01111110000000000010010000000000 01111110000000000010011000000000 01111110000000000010100000000000 01111110000000000010101000000000 01111110000000000010110000000000 01111110000000000010111000000000 01111110000000000011000000000000 01111110000000000011001000000000 01111110000000000011010000000000 01111110000000000011011000000000 01111110000000000011100000000000 01111110000000000011101000000000 01111110000000000011110000000000 01111110000000000011111000000000 01111110000000000100000000000000 01111110000000000100001000000000 01111110000000000100010000000000 01111110000000000100011000000000 01111110000000000100100000000000 01111110000000000100101000000000 01111110000000000100110000000000 01111110000000000100111000000000 01111110000000000101000000000000 01111110000000000101001000000000 01111110000000000101010000000000 01111110000000000101011000000000 01111110000000000101100000000000 01111110000000000101101000000000 01111110000000000101110000000000 01111110000000000101111000000000 01111110000000000110000000000000 01111110000000000110001000000000 01111110000000000110010000000000 01111110000000000110011000000000 01111110000000000110100000000000 01111110000000000110101000000000 01111110000000000110111000000000 01111110000000000111000000000000 01111110000000000111001000000000 01111110000000000111010000000000 01111110000000000111011000000000 01111110000000000111100000000000 01111110000000000111101000000000 01111110000000000111110000000000 01111110000000000111111000000000 01111110000000001000000000000000 01111110000000001000001000000000 01111110000000001000010000000000 01111110000000001000011000000000 01111110000000001000100000000000 01111110000000001000101000000000 01111110000000001000110000000000 01111110000000001000111000000000 01111110000000001001000000000000 01111110000000001001001000000000 01111110000000001001010000000000 01111110000000001001011000000000 01111110000000001001100000000000 01111110000000001001101000000000 01111110000000001001110000000000 01111110000000001001111000000000 01111110000000001010001000000000 01111110000000001010010000000000 01111110000000001010100000000000 01111110000000001010101000000000 01111110000000001010110000000000 01111110000000001010111000000000 01111110000000001011000000000000 01111110000000001011001000000000 01111110000000001011010000000000 01111110000000001011011000000000 default && && != .fieldderef .fieldderef INST VOP1 SRC0 I 32 250 I 32 U 1 != .fieldderef .fieldderef INST VOP1 SRC0 I 32 255 I 32 U 1 U 1 != .fieldderef .fieldderef INST VOP1 SRC0 I 32 249 I 32 U 1 U 1 Vector ALU operations with one destination and one source. Instructions in this encoding may be promoted to VOP3 unless otherwise noted. Instructions in this encoding may use SDWA or DPP modes unless otherwise noted. ENCODING Encoding. 7 25 OP Opcode. 8 9 SRC0 First operand for instruction. 9 0 VDST Destination for instruction. 8 17 ENC_VOPC 32 11111111111111100000000000000000 01111100001000000000000000000000 01111100001000100000000000000000 01111100001001000000000000000000 01111100001001100000000000000000 01111100001010000000000000000000 01111100001010100000000000000000 01111100010000000000000000000000 01111100010000100000000000000000 01111100010001000000000000000000 01111100010001100000000000000000 01111100010010000000000000000000 01111100010010100000000000000000 01111100010011000000000000000000 01111100010011100000000000000000 01111100010100000000000000000000 01111100010100100000000000000000 01111100010101000000000000000000 01111100010101100000000000000000 01111100010110000000000000000000 01111100010110100000000000000000 01111100010111000000000000000000 01111100010111100000000000000000 01111100011000000000000000000000 01111100011000100000000000000000 01111100011001000000000000000000 01111100011001100000000000000000 01111100011010000000000000000000 01111100011010100000000000000000 01111100011011000000000000000000 01111100011011100000000000000000 01111100011100000000000000000000 01111100011100100000000000000000 01111100011101000000000000000000 01111100011101100000000000000000 01111100011110000000000000000000 01111100011110100000000000000000 01111100011111000000000000000000 01111100011111100000000000000000 01111100100000000000000000000000 01111100100000100000000000000000 01111100100001000000000000000000 01111100100001100000000000000000 01111100100010000000000000000000 01111100100010100000000000000000 01111100100011000000000000000000 01111100100011100000000000000000 01111100100100000000000000000000 01111100100100100000000000000000 01111100100101000000000000000000 01111100100101100000000000000000 01111100100110000000000000000000 01111100100110100000000000000000 01111100100111000000000000000000 01111100100111100000000000000000 01111100101000000000000000000000 01111100101000100000000000000000 01111100101001000000000000000000 01111100101001100000000000000000 01111100101010000000000000000000 01111100101010100000000000000000 01111100101011000000000000000000 01111100101011100000000000000000 01111100101100000000000000000000 01111100101100100000000000000000 01111100101101000000000000000000 01111100101101100000000000000000 01111100101110000000000000000000 01111100101110100000000000000000 01111100101111000000000000000000 01111100101111100000000000000000 01111100110000000000000000000000 01111100110000100000000000000000 01111100110001000000000000000000 01111100110001100000000000000000 01111100110010000000000000000000 01111100110010100000000000000000 01111100110011000000000000000000 01111100110011100000000000000000 01111100110100000000000000000000 01111100110100100000000000000000 01111100110101000000000000000000 01111100110101100000000000000000 01111100110110000000000000000000 01111100110110100000000000000000 01111100110111000000000000000000 01111100110111100000000000000000 01111100111000000000000000000000 01111100111000100000000000000000 01111100111001000000000000000000 01111100111001100000000000000000 01111100111010000000000000000000 01111100111010100000000000000000 01111100111011000000000000000000 01111100111011100000000000000000 01111100111100000000000000000000 01111100111100100000000000000000 01111100111101000000000000000000 01111100111101100000000000000000 01111100111110000000000000000000 01111100111110100000000000000000 01111100111111000000000000000000 01111100111111100000000000000000 01111101010000000000000000000000 01111101010000100000000000000000 01111101010001000000000000000000 01111101010001100000000000000000 01111101010010000000000000000000 01111101010010100000000000000000 01111101010011000000000000000000 01111101010011100000000000000000 01111101010100000000000000000000 01111101010100100000000000000000 01111101010101000000000000000000 01111101010101100000000000000000 01111101010110000000000000000000 01111101010110100000000000000000 01111101010111000000000000000000 01111101010111100000000000000000 01111101011000000000000000000000 01111101011000100000000000000000 01111101011001000000000000000000 01111101011001100000000000000000 01111101011010000000000000000000 01111101011010100000000000000000 01111101011011000000000000000000 01111101011011100000000000000000 01111101011100000000000000000000 01111101011100100000000000000000 01111101011101000000000000000000 01111101011101100000000000000000 01111101011110000000000000000000 01111101011110100000000000000000 01111101011111000000000000000000 01111101011111100000000000000000 01111101100000000000000000000000 01111101100000100000000000000000 01111101100001000000000000000000 01111101100001100000000000000000 01111101100010000000000000000000 01111101100010100000000000000000 01111101100011000000000000000000 01111101100011100000000000000000 01111101100100000000000000000000 01111101100100100000000000000000 01111101100101000000000000000000 01111101100101100000000000000000 01111101100110000000000000000000 01111101100110100000000000000000 01111101100111000000000000000000 01111101100111100000000000000000 01111101101000000000000000000000 01111101101000100000000000000000 01111101101001000000000000000000 01111101101001100000000000000000 01111101101010000000000000000000 01111101101010100000000000000000 01111101101011000000000000000000 01111101101011100000000000000000 01111101101100000000000000000000 01111101101100100000000000000000 01111101101101000000000000000000 01111101101101100000000000000000 01111101101110000000000000000000 01111101101110100000000000000000 01111101101111000000000000000000 01111101101111100000000000000000 01111101110000000000000000000000 01111101110000100000000000000000 01111101110001000000000000000000 01111101110001100000000000000000 01111101110010000000000000000000 01111101110010100000000000000000 01111101110011000000000000000000 01111101110011100000000000000000 01111101110100000000000000000000 01111101110100100000000000000000 01111101110101000000000000000000 01111101110101100000000000000000 01111101110110000000000000000000 01111101110110100000000000000000 01111101110111000000000000000000 01111101110111100000000000000000 01111101111000000000000000000000 01111101111000100000000000000000 01111101111001000000000000000000 01111101111001100000000000000000 01111101111010000000000000000000 01111101111010100000000000000000 01111101111011000000000000000000 01111101111011100000000000000000 01111101111100000000000000000000 01111101111100100000000000000000 01111101111101000000000000000000 01111101111101100000000000000000 01111101111110000000000000000000 01111101111110100000000000000000 01111101111111000000000000000000 01111101111111100000000000000000 default && && != .fieldderef .fieldderef INST VOP2 SRC0 I 32 250 I 32 U 1 != .fieldderef .fieldderef INST VOP2 SRC0 I 32 255 I 32 U 1 U 1 != .fieldderef .fieldderef INST VOP2 SRC0 I 32 249 I 32 U 1 U 1 Vector ALU comparison operations with two sources. Instructions in this encoding may be promoted to VOP3 unless otherwise noted. Note that vector compare instructions respect the EXEC mask for computing the result in each thread; if EXEC[laneId] is zero, then the comparison for that thread is not performed and the result of VCC[laneId] is zero. ENCODING Encoding. 7 25 OP Opcode. 8 17 SRC0 First operand for instruction. 9 0 VSRC1 Second operand for instruction. 8 9 ENC_VOP2 32 11111110000000000000000000000000 00000000000000000000000000000000 00000010000000000000000000000000 00000100000000000000000000000000 00000110000000000000000000000000 00001000000000000000000000000000 00001010000000000000000000000000 00001100000000000000000000000000 00001110000000000000000000000000 00010000000000000000000000000000 00010010000000000000000000000000 00010100000000000000000000000000 00010110000000000000000000000000 00011000000000000000000000000000 00011010000000000000000000000000 00011100000000000000000000000000 00011110000000000000000000000000 00100000000000000000000000000000 00100010000000000000000000000000 00100100000000000000000000000000 00100110000000000000000000000000 00101000000000000000000000000000 00101010000000000000000000000000 00101100000000000000000000000000 00110010000000000000000000000000 00110100000000000000000000000000 00110110000000000000000000000000 00111000000000000000000000000000 00111010000000000000000000000000 00111100000000000000000000000000 00111110000000000000000000000000 01000000000000000000000000000000 01000010000000000000000000000000 01000100000000000000000000000000 01000110000000000000000000000000 01001100000000000000000000000000 01001110000000000000000000000000 01010000000000000000000000000000 01010010000000000000000000000000 01010100000000000000000000000000 01010110000000000000000000000000 01011000000000000000000000000000 01011010000000000000000000000000 01011100000000000000000000000000 01011110000000000000000000000000 01100000000000000000000000000000 01100010000000000000000000000000 01100100000000000000000000000000 01100110000000000000000000000000 01101000000000000000000000000000 01101010000000000000000000000000 01101100000000000000000000000000 01101110000000000000000000000000 01110000000000000000000000000000 01110010000000000000000000000000 01110100000000000000000000000000 01110110000000000000000000000000 01111000000000000000000000000000 01111010000000000000000000000000 default && && != .fieldderef .fieldderef INST VOP2 SRC0 I 32 250 I 32 U 1 != .fieldderef .fieldderef INST VOP2 SRC0 I 32 255 I 32 U 1 U 1 != .fieldderef .fieldderef INST VOP2 SRC0 I 32 249 I 32 U 1 U 1 Vector ALU operations with one destination and two sources. Instructions in this encoding may be promoted to VOP3 unless otherwise noted. ENCODING Encoding. 1 31 OP Opcode. 6 25 SRC0 First operand for instruction. 9 0 VDST Destination for instruction. 8 17 VSRC1 Second operand for instruction. 8 9 ENC_VOP3PX2 128 00000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000011111111111111110000000000000000 00000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000011010011101011010000000000000000 00000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000011010011101011100000000000000000 default U 1 Paired vector ALU operations. ABID CBSZ and ABID are for A matrix swizzling, it is independent with BLGP, and with much more fine grain level control. 4 75 ACC ACC field, specifies whether srcA, srcB use an accumulator VGPR. 2 123 ACC_CD ACC field, specifies whether srcC and VDST use an accumulator VGPR (both operands share the same config bit). For instructions with OPF_ACC_CD_ONLY_D flag set, ACC_CD only controls the destination and the srcC must be an ArchVGPR. 1 79 BLGP BLGP control the Matrix B input swizzling, 16 lanes are one group and swizzling are between those 4 groups. 3 125 CBSZ CBSZ and ABID are for A matrix swizzling, it is independent with BLGP, and with much more fine grain level control. 3 72 ENCODING Encoding. 9 87 NEG If NEG[N] set, take the floating-point negation of the N'th input low opcode. This is applied after absolute value. 3 61 NEG_HI If NEG_HI[N] set, take the floating-point negation of the N'th input high opcode. This is applied after absolute value. 3 8 OP Opcode. 7 80 OP_SEL Input modifier for low opcode src0, src1 and src2. If 0 select src[0:15]. If 1 select src[16:31]. 3 11 OP_SEL_HI Input modifier for high opcode src0 and src1. If 0 select src[0:15]. If 1 select src[16:31]. 2 59 1 14 SCALE_SRC0 First operand for instruction. 9 32 SCALE_SRC1 Second operand for instruction. 9 41 SRC0 First operand for instruction. 9 96 SRC1 Second operand for instruction. 9 105 SRC2 Third operand for instruction. 9 114 VDST Destination for instruction. 8 64 X2ENCODING Encoding. 16 16 ENC_VOP3P 64 0000000000000000000000000000000011111111111111110000000000000000 0000000000000000000000000000000011010011100000000000000000000000 0000000000000000000000000000000011010011100000010000000000000000 0000000000000000000000000000000011010011100000100000000000000000 0000000000000000000000000000000011010011100000110000000000000000 0000000000000000000000000000000011010011100001000000000000000000 0000000000000000000000000000000011010011100001010000000000000000 0000000000000000000000000000000011010011100001100000000000000000 0000000000000000000000000000000011010011100001110000000000000000 0000000000000000000000000000000011010011100010000000000000000000 0000000000000000000000000000000011010011100010010000000000000000 0000000000000000000000000000000011010011100010100000000000000000 0000000000000000000000000000000011010011100010110000000000000000 0000000000000000000000000000000011010011100011000000000000000000 0000000000000000000000000000000011010011100011010000000000000000 0000000000000000000000000000000011010011100011100000000000000000 0000000000000000000000000000000011010011100011110000000000000000 0000000000000000000000000000000011010011100100000000000000000000 0000000000000000000000000000000011010011100100010000000000000000 0000000000000000000000000000000011010011100100100000000000000000 0000000000000000000000000000000011010011100110100000000000000000 0000000000000000000000000000000011010011100110110000000000000000 0000000000000000000000000000000011010011100111000000000000000000 0000000000000000000000000000000011010011101000000000000000000000 0000000000000000000000000000000011010011101000010000000000000000 0000000000000000000000000000000011010011101000100000000000000000 0000000000000000000000000000000011010011101000110000000000000000 0000000000000000000000000000000011010011101001100000000000000000 0000000000000000000000000000000011010011101001110000000000000000 0000000000000000000000000000000011010011101010000000000000000000 0000000000000000000000000000000011010011101010010000000000000000 0000000000000000000000000000000011010011101010100000000000000000 0000000000000000000000000000000011010011101010110000000000000000 0000000000000000000000000000000011010011101100000000000000000000 0000000000000000000000000000000011010011101100010000000000000000 0000000000000000000000000000000011010011101100100000000000000000 0000000000000000000000000000000011010011101100110000000000000000 0000000000000000000000000000000011010011110110000000000000000000 0000000000000000000000000000000011010011110110010000000000000000 default U 1 Vector ALU operations requiring two or three sources and performing two 16 bit ops in one instruction. CLAMP If set, clamp output to [0.0, 1.0]. Applied after output modifier. 1 15 ENCODING Encoding. 9 23 NEG If NEG[N] set, take the floating-point negation of the N'th input low opcode. This is applied after absolute value. 3 61 NEG_HI If NEG_HI[N] set, take the floating-point negation of the N'th input high opcode. This is applied after absolute value. 3 8 OP Opcode. 7 16 OP_SEL Input modifier for low opcode src0,src1 and src2. If 0 select src[0:15]. If 1 select src[16:31] 3 11 OP_SEL_HI Input modifier for high opcode src0 and src1. If 0 select src[0:15]. If 1 select src[16:31]. 2 59 1 14 SRC0 First operand for instruction. 9 32 SRC1 Second operand for instruction. 9 41 SRC2 Third operand for instruction. 9 50 VDST Destination for instruction. 8 0 ENC_VOP3 64 0000000000000000000000000000000011111111111111110000000000000000 0000000000000000000000000000000011010001010000000000000000000000 0000000000000000000000000000000011010001010000010000000000000000 0000000000000000000000000000000011010001010000100000000000000000 0000000000000000000000000000000011010001010000110000000000000000 0000000000000000000000000000000011010001010001000000000000000000 0000000000000000000000000000000011010001010001010000000000000000 0000000000000000000000000000000011010001010001100000000000000000 0000000000000000000000000000000011010001010001110000000000000000 0000000000000000000000000000000011010001010010000000000000000000 0000000000000000000000000000000011010001010010100000000000000000 0000000000000000000000000000000011010001010010110000000000000000 0000000000000000000000000000000011010001010011000000000000000000 0000000000000000000000000000000011010001010011010000000000000000 0000000000000000000000000000000011010001010011100000000000000000 0000000000000000000000000000000011010001010011110000000000000000 0000000000000000000000000000000011010001010100000000000000000000 0000000000000000000000000000000011010001010100010000000000000000 0000000000000000000000000000000011010001010100100000000000000000 0000000000000000000000000000000011010001010100110000000000000000 0000000000000000000000000000000011010001010101000000000000000000 0000000000000000000000000000000011010001010101010000000000000000 0000000000000000000000000000000011010001010101100000000000000000 0000000000000000000000000000000011010001010101110000000000000000 0000000000000000000000000000000011010001010110000000000000000000 0000000000000000000000000000000011010001010110010000000000000000 0000000000000000000000000000000011010001010110100000000000000000 0000000000000000000000000000000011010001010110110000000000000000 0000000000000000000000000000000011010001010111000000000000000000 0000000000000000000000000000000011010001010111010000000000000000 0000000000000000000000000000000011010001010111100000000000000000 0000000000000000000000000000000011010001010111110000000000000000 0000000000000000000000000000000011010001011000000000000000000000 0000000000000000000000000000000011010001011000010000000000000000 0000000000000000000000000000000011010001011000100000000000000000 0000000000000000000000000000000011010001011000110000000000000000 0000000000000000000000000000000011010001011001000000000000000000 0000000000000000000000000000000011010001011001010000000000000000 0000000000000000000000000000000011010001011001100000000000000000 0000000000000000000000000000000011010001011001110000000000000000 0000000000000000000000000000000011010001011010000000000000000000 0000000000000000000000000000000011010001011010010000000000000000 0000000000000000000000000000000011010001011010100000000000000000 0000000000000000000000000000000011010001011010110000000000000000 0000000000000000000000000000000011010001011011000000000000000000 0000000000000000000000000000000011010001011011010000000000000000 0000000000000000000000000000000011010001011011100000000000000000 0000000000000000000000000000000011010001011011110000000000000000 0000000000000000000000000000000011010001011100000000000000000000 0000000000000000000000000000000011010001011100010000000000000000 0000000000000000000000000000000011010001011100100000000000000000 0000000000000000000000000000000011010001011100110000000000000000 0000000000000000000000000000000011010001011101000000000000000000 0000000000000000000000000000000011010001011101010000000000000000 0000000000000000000000000000000011010001011101110000000000000000 0000000000000000000000000000000011010001011110000000000000000000 0000000000000000000000000000000011010001011110010000000000000000 0000000000000000000000000000000011010001011110100000000000000000 0000000000000000000000000000000011010001011110110000000000000000 0000000000000000000000000000000011010001011111000000000000000000 0000000000000000000000000000000011010001011111010000000000000000 0000000000000000000000000000000011010001011111100000000000000000 0000000000000000000000000000000011010001011111110000000000000000 0000000000000000000000000000000011010001100000000000000000000000 0000000000000000000000000000000011010001100000010000000000000000 0000000000000000000000000000000011010001100000100000000000000000 0000000000000000000000000000000011010001100000110000000000000000 0000000000000000000000000000000011010001100001000000000000000000 0000000000000000000000000000000011010001100001010000000000000000 0000000000000000000000000000000011010001100001100000000000000000 0000000000000000000000000000000011010001100001110000000000000000 0000000000000000000000000000000011010001100010000000000000000000 0000000000000000000000000000000011010001100010010000000000000000 0000000000000000000000000000000011010001100010100000000000000000 0000000000000000000000000000000011010001100010110000000000000000 0000000000000000000000000000000011010001100011000000000000000000 0000000000000000000000000000000011010001100011010000000000000000 0000000000000000000000000000000011010001100011100000000000000000 0000000000000000000000000000000011010001100011110000000000000000 0000000000000000000000000000000011010001100100010000000000000000 0000000000000000000000000000000011010001100100100000000000000000 0000000000000000000000000000000011010001100101000000000000000000 0000000000000000000000000000000011010001100101010000000000000000 0000000000000000000000000000000011010001100101100000000000000000 0000000000000000000000000000000011010001100101110000000000000000 0000000000000000000000000000000011010001100110000000000000000000 0000000000000000000000000000000011010001100110010000000000000000 0000000000000000000000000000000011010001100110100000000000000000 0000000000000000000000000000000011010001100110110000000000000000 0000000000000000000000000000000011010001000000000000000000000000 0000000000000000000000000000000011010001000000010000000000000000 0000000000000000000000000000000011010001000000100000000000000000 0000000000000000000000000000000011010001000000110000000000000000 0000000000000000000000000000000011010001000001000000000000000000 0000000000000000000000000000000011010001000001010000000000000000 0000000000000000000000000000000011010001000001100000000000000000 0000000000000000000000000000000011010001000001110000000000000000 0000000000000000000000000000000011010001000010000000000000000000 0000000000000000000000000000000011010001000010010000000000000000 0000000000000000000000000000000011010001000010100000000000000000 0000000000000000000000000000000011010001000010110000000000000000 0000000000000000000000000000000011010001000011000000000000000000 0000000000000000000000000000000011010001000011010000000000000000 0000000000000000000000000000000011010001000011100000000000000000 0000000000000000000000000000000011010001000011110000000000000000 0000000000000000000000000000000011010001000100000000000000000000 0000000000000000000000000000000011010001000100010000000000000000 0000000000000000000000000000000011010001000100100000000000000000 0000000000000000000000000000000011010001000100110000000000000000 0000000000000000000000000000000011010001000101000000000000000000 0000000000000000000000000000000011010001000101010000000000000000 0000000000000000000000000000000011010001000101100000000000000000 0000000000000000000000000000000011010001000111110000000000000000 0000000000000000000000000000000011010001001000000000000000000000 0000000000000000000000000000000011010001001000010000000000000000 0000000000000000000000000000000011010001001000100000000000000000 0000000000000000000000000000000011010001001000110000000000000000 0000000000000000000000000000000011010001001001100000000000000000 0000000000000000000000000000000011010001001001110000000000000000 0000000000000000000000000000000011010001001010000000000000000000 0000000000000000000000000000000011010001001010010000000000000000 0000000000000000000000000000000011010001001010100000000000000000 0000000000000000000000000000000011010001001010110000000000000000 0000000000000000000000000000000011010001001011000000000000000000 0000000000000000000000000000000011010001001011010000000000000000 0000000000000000000000000000000011010001001011100000000000000000 0000000000000000000000000000000011010001001011110000000000000000 0000000000000000000000000000000011010001001100000000000000000000 0000000000000000000000000000000011010001001100010000000000000000 0000000000000000000000000000000011010001001100100000000000000000 0000000000000000000000000000000011010001001100110000000000000000 0000000000000000000000000000000011010001001101000000000000000000 0000000000000000000000000000000011010001001101010000000000000000 0000000000000000000000000000000011010001001101100000000000000000 0000000000000000000000000000000011010001001101110000000000000000 0000000000000000000000000000000011010001001110000000000000000000 0000000000000000000000000000000011010001001110010000000000000000 0000000000000000000000000000000011010001001110100000000000000000 0000000000000000000000000000000011010001001110110000000000000000 0000000000000000000000000000000011010001001111000000000000000000 0000000000000000000000000000000011010001001111010000000000000000 0000000000000000000000000000000011010001110000100000000000000000 0000000000000000000000000000000011010001110000110000000000000000 0000000000000000000000000000000011010001110001000000000000000000 0000000000000000000000000000000011010001110001010000000000000000 0000000000000000000000000000000011010001110001100000000000000000 0000000000000000000000000000000011010001110001110000000000000000 0000000000000000000000000000000011010001110010000000000000000000 0000000000000000000000000000000011010001110010010000000000000000 0000000000000000000000000000000011010001110010100000000000000000 0000000000000000000000000000000011010001110010110000000000000000 0000000000000000000000000000000011010001110011000000000000000000 0000000000000000000000000000000011010001110011010000000000000000 0000000000000000000000000000000011010001110011100000000000000000 0000000000000000000000000000000011010001110011110000000000000000 0000000000000000000000000000000011010001110100000000000000000000 0000000000000000000000000000000011010001110100010000000000000000 0000000000000000000000000000000011010001110100100000000000000000 0000000000000000000000000000000011010001110100110000000000000000 0000000000000000000000000000000011010001110101000000000000000000 0000000000000000000000000000000011010001110101010000000000000000 0000000000000000000000000000000011010001110101100000000000000000 0000000000000000000000000000000011010001110101110000000000000000 0000000000000000000000000000000011010001110110000000000000000000 0000000000000000000000000000000011010001110110010000000000000000 0000000000000000000000000000000011010001110110100000000000000000 0000000000000000000000000000000011010001110110110000000000000000 0000000000000000000000000000000011010001110111000000000000000000 0000000000000000000000000000000011010001110111010000000000000000 0000000000000000000000000000000011010001110111100000000000000000 0000000000000000000000000000000011010001110111110000000000000000 0000000000000000000000000000000011010001111000100000000000000000 0000000000000000000000000000000011010001111000110000000000000000 0000000000000000000000000000000011010001111001000000000000000000 0000000000000000000000000000000011010001111001010000000000000000 0000000000000000000000000000000011010001111001100000000000000000 0000000000000000000000000000000011010001111001110000000000000000 0000000000000000000000000000000011010001111010100000000000000000 0000000000000000000000000000000011010001111010110000000000000000 0000000000000000000000000000000011010001111011000000000000000000 0000000000000000000000000000000011010001111011010000000000000000 0000000000000000000000000000000011010001111011100000000000000000 0000000000000000000000000000000011010001111011110000000000000000 0000000000000000000000000000000011010001111100000000000000000000 0000000000000000000000000000000011010001111100010000000000000000 0000000000000000000000000000000011010001111100100000000000000000 0000000000000000000000000000000011010001111100110000000000000000 0000000000000000000000000000000011010001111101000000000000000000 0000000000000000000000000000000011010001111101010000000000000000 0000000000000000000000000000000011010001111101100000000000000000 0000000000000000000000000000000011010001111101110000000000000000 0000000000000000000000000000000011010001111110000000000000000000 0000000000000000000000000000000011010001111110010000000000000000 0000000000000000000000000000000011010001111110100000000000000000 0000000000000000000000000000000011010001111110110000000000000000 0000000000000000000000000000000011010001111111000000000000000000 0000000000000000000000000000000011010001111111010000000000000000 0000000000000000000000000000000011010001111111100000000000000000 0000000000000000000000000000000011010001111111110000000000000000 0000000000000000000000000000000011010010000000000000000000000000 0000000000000000000000000000000011010010000000010000000000000000 0000000000000000000000000000000011010010000000100000000000000000 0000000000000000000000000000000011010010000000110000000000000000 0000000000000000000000000000000011010010000001000000000000000000 0000000000000000000000000000000011010010000001010000000000000000 0000000000000000000000000000000011010010000001100000000000000000 0000000000000000000000000000000011010010000001110000000000000000 0000000000000000000000000000000011010010000010000000000000000000 0000000000000000000000000000000011010010001100110000000000000000 0000000000000000000000000000000011010010001101000000000000000000 0000000000000000000000000000000011010010001101010000000000000000 0000000000000000000000000000000011010010001101100000000000000000 0000000000000000000000000000000011010010001101110000000000000000 0000000000000000000000000000000011010010001110000000000000000000 0000000000000000000000000000000011010010001110010000000000000000 0000000000000000000000000000000011010010001110100000000000000000 0000000000000000000000000000000011010010001110110000000000000000 0000000000000000000000000000000011010010001111000000000000000000 0000000000000000000000000000000011010010001111010000000000000000 0000000000000000000000000000000011010010001111100000000000000000 0000000000000000000000000000000011010010001111110000000000000000 0000000000000000000000000000000011010010010000000000000000000000 0000000000000000000000000000000011010010010000010000000000000000 0000000000000000000000000000000011010010010000100000000000000000 0000000000000000000000000000000011010010010000110000000000000000 0000000000000000000000000000000011010010010001000000000000000000 0000000000000000000000000000000011010010010001010000000000000000 0000000000000000000000000000000011010010010001100000000000000000 0000000000000000000000000000000011010010010001110000000000000000 0000000000000000000000000000000011010010010010000000000000000000 0000000000000000000000000000000011010010010010010000000000000000 0000000000000000000000000000000011010010010010100000000000000000 0000000000000000000000000000000011010010010010110000000000000000 0000000000000000000000000000000011010010010011000000000000000000 0000000000000000000000000000000011010010010011010000000000000000 0000000000000000000000000000000011010010010011100000000000000000 0000000000000000000000000000000011010010010011110000000000000000 0000000000000000000000000000000011010010010100000000000000000000 0000000000000000000000000000000011010010010100010000000000000000 0000000000000000000000000000000011010010010100100000000000000000 0000000000000000000000000000000011010010010100110000000000000000 0000000000000000000000000000000011010010010101000000000000000000 0000000000000000000000000000000011010010010101010000000000000000 0000000000000000000000000000000011010010010101100000000000000000 0000000000000000000000000000000011010010010101110000000000000000 0000000000000000000000000000000011010010010110000000000000000000 0000000000000000000000000000000011010010010110010000000000000000 0000000000000000000000000000000011010010010110100000000000000000 0000000000000000000000000000000011010010010110110000000000000000 0000000000000000000000000000000011010010010111000000000000000000 0000000000000000000000000000000011010010010111010000000000000000 0000000000000000000000000000000011010010010111100000000000000000 0000000000000000000000000000000011010010010111110000000000000000 0000000000000000000000000000000011010010011000000000000000000000 0000000000000000000000000000000011010010011000010000000000000000 0000000000000000000000000000000011010010011000100000000000000000 0000000000000000000000000000000011010010011000110000000000000000 0000000000000000000000000000000011010010011001010000000000000000 0000000000000000000000000000000011010010011001100000000000000000 0000000000000000000000000000000011010010011001110000000000000000 0000000000000000000000000000000011010010011010000000000000000000 0000000000000000000000000000000011010010011010010000000000000000 0000000000000000000000000000000011010010011010100000000000000000 0000000000000000000000000000000011010010100000000000000000000000 0000000000000000000000000000000011010010100000010000000000000000 0000000000000000000000000000000011010010100000100000000000000000 0000000000000000000000000000000011010010100000110000000000000000 0000000000000000000000000000000011010010100001000000000000000000 0000000000000000000000000000000011010010100001010000000000000000 0000000000000000000000000000000011010010100001100000000000000000 0000000000000000000000000000000011010010100001110000000000000000 0000000000000000000000000000000011010010100010000000000000000000 0000000000000000000000000000000011010010100010010000000000000000 0000000000000000000000000000000011010010100010100000000000000000 0000000000000000000000000000000011010010100010110000000000000000 0000000000000000000000000000000011010010100011000000000000000000 0000000000000000000000000000000011010010100011010000000000000000 0000000000000000000000000000000011010010100011110000000000000000 0000000000000000000000000000000011010010100100000000000000000000 0000000000000000000000000000000011010010100100010000000000000000 0000000000000000000000000000000011010010100100100000000000000000 0000000000000000000000000000000011010010100100110000000000000000 0000000000000000000000000000000011010010100101000000000000000000 0000000000000000000000000000000011010010100101010000000000000000 0000000000000000000000000000000011010010100101100000000000000000 0000000000000000000000000000000011010010100101110000000000000000 0000000000000000000000000000000011010010100110000000000000000000 0000000000000000000000000000000011010010100110010000000000000000 0000000000000000000000000000000011010010100110100000000000000000 0000000000000000000000000000000011010010100111000000000000000000 0000000000000000000000000000000011010010100111010000000000000000 0000000000000000000000000000000011010010100111100000000000000000 0000000000000000000000000000000011010010100111110000000000000000 0000000000000000000000000000000011010010101000000000000000000000 0000000000000000000000000000000011010010101000010000000000000000 0000000000000000000000000000000011010010101000100000000000000000 0000000000000000000000000000000011010010101000110000000000000000 0000000000000000000000000000000011010010101001000000000000000000 0000000000000000000000000000000011010010101001010000000000000000 0000000000000000000000000000000011010010101001100000000000000000 0000000000000000000000000000000011010010101001110000000000000000 0000000000000000000000000000000011010010101010000000000000000000 0000000000000000000000000000000011010010101010010000000000000000 0000000000000000000000000000000011010000000100000000000000000000 0000000000000000000000000000000011010000000100010000000000000000 0000000000000000000000000000000011010000000100100000000000000000 0000000000000000000000000000000011010000000100110000000000000000 0000000000000000000000000000000011010000000101000000000000000000 0000000000000000000000000000000011010000000101010000000000000000 0000000000000000000000000000000011010000001000000000000000000000 0000000000000000000000000000000011010000001000010000000000000000 0000000000000000000000000000000011010000001000100000000000000000 0000000000000000000000000000000011010000001000110000000000000000 0000000000000000000000000000000011010000001001000000000000000000 0000000000000000000000000000000011010000001001010000000000000000 0000000000000000000000000000000011010000001001100000000000000000 0000000000000000000000000000000011010000001001110000000000000000 0000000000000000000000000000000011010000001010000000000000000000 0000000000000000000000000000000011010000001010010000000000000000 0000000000000000000000000000000011010000001010100000000000000000 0000000000000000000000000000000011010000001010110000000000000000 0000000000000000000000000000000011010000001011000000000000000000 0000000000000000000000000000000011010000001011010000000000000000 0000000000000000000000000000000011010000001011100000000000000000 0000000000000000000000000000000011010000001011110000000000000000 0000000000000000000000000000000011010000001100000000000000000000 0000000000000000000000000000000011010000001100010000000000000000 0000000000000000000000000000000011010000001100100000000000000000 0000000000000000000000000000000011010000001100110000000000000000 0000000000000000000000000000000011010000001101000000000000000000 0000000000000000000000000000000011010000001101010000000000000000 0000000000000000000000000000000011010000001101100000000000000000 0000000000000000000000000000000011010000001101110000000000000000 0000000000000000000000000000000011010000001110000000000000000000 0000000000000000000000000000000011010000001110010000000000000000 0000000000000000000000000000000011010000001110100000000000000000 0000000000000000000000000000000011010000001110110000000000000000 0000000000000000000000000000000011010000001111000000000000000000 0000000000000000000000000000000011010000001111010000000000000000 0000000000000000000000000000000011010000001111100000000000000000 0000000000000000000000000000000011010000001111110000000000000000 0000000000000000000000000000000011010000010000000000000000000000 0000000000000000000000000000000011010000010000010000000000000000 0000000000000000000000000000000011010000010000100000000000000000 0000000000000000000000000000000011010000010000110000000000000000 0000000000000000000000000000000011010000010001000000000000000000 0000000000000000000000000000000011010000010001010000000000000000 0000000000000000000000000000000011010000010001100000000000000000 0000000000000000000000000000000011010000010001110000000000000000 0000000000000000000000000000000011010000010010000000000000000000 0000000000000000000000000000000011010000010010010000000000000000 0000000000000000000000000000000011010000010010100000000000000000 0000000000000000000000000000000011010000010010110000000000000000 0000000000000000000000000000000011010000010011000000000000000000 0000000000000000000000000000000011010000010011010000000000000000 0000000000000000000000000000000011010000010011100000000000000000 0000000000000000000000000000000011010000010011110000000000000000 0000000000000000000000000000000011010000010100000000000000000000 0000000000000000000000000000000011010000010100010000000000000000 0000000000000000000000000000000011010000010100100000000000000000 0000000000000000000000000000000011010000010100110000000000000000 0000000000000000000000000000000011010000010101000000000000000000 0000000000000000000000000000000011010000010101010000000000000000 0000000000000000000000000000000011010000010101100000000000000000 0000000000000000000000000000000011010000010101110000000000000000 0000000000000000000000000000000011010000010110000000000000000000 0000000000000000000000000000000011010000010110010000000000000000 0000000000000000000000000000000011010000010110100000000000000000 0000000000000000000000000000000011010000010110110000000000000000 0000000000000000000000000000000011010000010111000000000000000000 0000000000000000000000000000000011010000010111010000000000000000 0000000000000000000000000000000011010000010111100000000000000000 0000000000000000000000000000000011010000010111110000000000000000 0000000000000000000000000000000011010000011000000000000000000000 0000000000000000000000000000000011010000011000010000000000000000 0000000000000000000000000000000011010000011000100000000000000000 0000000000000000000000000000000011010000011000110000000000000000 0000000000000000000000000000000011010000011001000000000000000000 0000000000000000000000000000000011010000011001010000000000000000 0000000000000000000000000000000011010000011001100000000000000000 0000000000000000000000000000000011010000011001110000000000000000 0000000000000000000000000000000011010000011010000000000000000000 0000000000000000000000000000000011010000011010010000000000000000 0000000000000000000000000000000011010000011010100000000000000000 0000000000000000000000000000000011010000011010110000000000000000 0000000000000000000000000000000011010000011011000000000000000000 0000000000000000000000000000000011010000011011010000000000000000 0000000000000000000000000000000011010000011011100000000000000000 0000000000000000000000000000000011010000011011110000000000000000 0000000000000000000000000000000011010000011100000000000000000000 0000000000000000000000000000000011010000011100010000000000000000 0000000000000000000000000000000011010000011100100000000000000000 0000000000000000000000000000000011010000011100110000000000000000 0000000000000000000000000000000011010000011101000000000000000000 0000000000000000000000000000000011010000011101010000000000000000 0000000000000000000000000000000011010000011101100000000000000000 0000000000000000000000000000000011010000011101110000000000000000 0000000000000000000000000000000011010000011110000000000000000000 0000000000000000000000000000000011010000011110010000000000000000 0000000000000000000000000000000011010000011110100000000000000000 0000000000000000000000000000000011010000011110110000000000000000 0000000000000000000000000000000011010000011111000000000000000000 0000000000000000000000000000000011010000011111010000000000000000 0000000000000000000000000000000011010000011111100000000000000000 0000000000000000000000000000000011010000011111110000000000000000 0000000000000000000000000000000011010000101000000000000000000000 0000000000000000000000000000000011010000101000010000000000000000 0000000000000000000000000000000011010000101000100000000000000000 0000000000000000000000000000000011010000101000110000000000000000 0000000000000000000000000000000011010000101001000000000000000000 0000000000000000000000000000000011010000101001010000000000000000 0000000000000000000000000000000011010000101001100000000000000000 0000000000000000000000000000000011010000101001110000000000000000 0000000000000000000000000000000011010000101010000000000000000000 0000000000000000000000000000000011010000101010010000000000000000 0000000000000000000000000000000011010000101010100000000000000000 0000000000000000000000000000000011010000101010110000000000000000 0000000000000000000000000000000011010000101011000000000000000000 0000000000000000000000000000000011010000101011010000000000000000 0000000000000000000000000000000011010000101011100000000000000000 0000000000000000000000000000000011010000101011110000000000000000 0000000000000000000000000000000011010000101100000000000000000000 0000000000000000000000000000000011010000101100010000000000000000 0000000000000000000000000000000011010000101100100000000000000000 0000000000000000000000000000000011010000101100110000000000000000 0000000000000000000000000000000011010000101101000000000000000000 0000000000000000000000000000000011010000101101010000000000000000 0000000000000000000000000000000011010000101101100000000000000000 0000000000000000000000000000000011010000101101110000000000000000 0000000000000000000000000000000011010000101110000000000000000000 0000000000000000000000000000000011010000101110010000000000000000 0000000000000000000000000000000011010000101110100000000000000000 0000000000000000000000000000000011010000101110110000000000000000 0000000000000000000000000000000011010000101111000000000000000000 0000000000000000000000000000000011010000101111010000000000000000 0000000000000000000000000000000011010000101111100000000000000000 0000000000000000000000000000000011010000101111110000000000000000 0000000000000000000000000000000011010000110000000000000000000000 0000000000000000000000000000000011010000110000010000000000000000 0000000000000000000000000000000011010000110000100000000000000000 0000000000000000000000000000000011010000110000110000000000000000 0000000000000000000000000000000011010000110001000000000000000000 0000000000000000000000000000000011010000110001010000000000000000 0000000000000000000000000000000011010000110001100000000000000000 0000000000000000000000000000000011010000110001110000000000000000 0000000000000000000000000000000011010000110010000000000000000000 0000000000000000000000000000000011010000110010010000000000000000 0000000000000000000000000000000011010000110010100000000000000000 0000000000000000000000000000000011010000110010110000000000000000 0000000000000000000000000000000011010000110011000000000000000000 0000000000000000000000000000000011010000110011010000000000000000 0000000000000000000000000000000011010000110011100000000000000000 0000000000000000000000000000000011010000110011110000000000000000 0000000000000000000000000000000011010000110100000000000000000000 0000000000000000000000000000000011010000110100010000000000000000 0000000000000000000000000000000011010000110100100000000000000000 0000000000000000000000000000000011010000110100110000000000000000 0000000000000000000000000000000011010000110101000000000000000000 0000000000000000000000000000000011010000110101010000000000000000 0000000000000000000000000000000011010000110101100000000000000000 0000000000000000000000000000000011010000110101110000000000000000 0000000000000000000000000000000011010000110110000000000000000000 0000000000000000000000000000000011010000110110010000000000000000 0000000000000000000000000000000011010000110110100000000000000000 0000000000000000000000000000000011010000110110110000000000000000 0000000000000000000000000000000011010000110111000000000000000000 0000000000000000000000000000000011010000110111010000000000000000 0000000000000000000000000000000011010000110111100000000000000000 0000000000000000000000000000000011010000110111110000000000000000 0000000000000000000000000000000011010000111000000000000000000000 0000000000000000000000000000000011010000111000010000000000000000 0000000000000000000000000000000011010000111000100000000000000000 0000000000000000000000000000000011010000111000110000000000000000 0000000000000000000000000000000011010000111001000000000000000000 0000000000000000000000000000000011010000111001010000000000000000 0000000000000000000000000000000011010000111001100000000000000000 0000000000000000000000000000000011010000111001110000000000000000 0000000000000000000000000000000011010000111010000000000000000000 0000000000000000000000000000000011010000111010010000000000000000 0000000000000000000000000000000011010000111010100000000000000000 0000000000000000000000000000000011010000111010110000000000000000 0000000000000000000000000000000011010000111011000000000000000000 0000000000000000000000000000000011010000111011010000000000000000 0000000000000000000000000000000011010000111011100000000000000000 0000000000000000000000000000000011010000111011110000000000000000 0000000000000000000000000000000011010000111100000000000000000000 0000000000000000000000000000000011010000111100010000000000000000 0000000000000000000000000000000011010000111100100000000000000000 0000000000000000000000000000000011010000111100110000000000000000 0000000000000000000000000000000011010000111101000000000000000000 0000000000000000000000000000000011010000111101010000000000000000 0000000000000000000000000000000011010000111101100000000000000000 0000000000000000000000000000000011010000111101110000000000000000 0000000000000000000000000000000011010000111110000000000000000000 0000000000000000000000000000000011010000111110010000000000000000 0000000000000000000000000000000011010000111110100000000000000000 0000000000000000000000000000000011010000111110110000000000000000 0000000000000000000000000000000011010000111111000000000000000000 0000000000000000000000000000000011010000111111010000000000000000 0000000000000000000000000000000011010000111111100000000000000000 0000000000000000000000000000000011010000111111110000000000000000 default U 1 Vector ALU operations requiring three sources or modifiers; also some obscure opcodes. Almost all VOP1, VOP2, VOPC and VINTRP opcodes can be promoted to VOP3 to use input and output modifiers. ABS If ABS[N] set, take the floating-point absolute value of the N'th input operand. This is applied before negation. 3 8 CLAMP If set, clamp output to [0.0, 1.0]. Applied after output modifier. 1 15 ENCODING Encoding. 6 26 NEG If NEG[N] set, take the floating-point negation of the N'th input operand. This is applied after absolute value. 3 61 OMOD Output modifier for instruction. Applied before clamping. 2 59 OP Opcode. 10 16 OP_SEL OP_SEL [0,1,2] to select src0,src1,src2 lower[0:15] or higher order[16:31] bits. OP_SEL[3] to select destination. If 0 select [0:15] If 1 select [16:31].For destination 0 sets higher order bits to zero and [0:15] is result. If 1 result in higher order bits and bits preserved in [0:15]. 4 11 SRC0 First operand for instruction. 9 32 SRC1 Second operand for instruction. 9 41 SRC2 Third operand for instruction. 9 50 VDST Destination for instruction. 8 0 ENC_DS 64 0000000000000000000000000000000011111101111111100000000000000000 0000000000000000000000000000000011011000000000000000000000000000 0000000000000000000000000000000011011000000000100000000000000000 0000000000000000000000000000000011011000000001000000000000000000 0000000000000000000000000000000011011000000001100000000000000000 0000000000000000000000000000000011011000000010000000000000000000 0000000000000000000000000000000011011000000010100000000000000000 0000000000000000000000000000000011011000000011000000000000000000 0000000000000000000000000000000011011000000011100000000000000000 0000000000000000000000000000000011011000000100000000000000000000 0000000000000000000000000000000011011000000100100000000000000000 0000000000000000000000000000000011011000000101000000000000000000 0000000000000000000000000000000011011000000101100000000000000000 0000000000000000000000000000000011011000000110000000000000000000 0000000000000000000000000000000011011000000110100000000000000000 0000000000000000000000000000000011011000000111000000000000000000 0000000000000000000000000000000011011000000111100000000000000000 0000000000000000000000000000000011011000001000000000000000000000 0000000000000000000000000000000011011000001000100000000000000000 0000000000000000000000000000000011011000001001000000000000000000 0000000000000000000000000000000011011000001001100000000000000000 0000000000000000000000000000000011011000001010000000000000000000 0000000000000000000000000000000011011000001010100000000000000000 0000000000000000000000000000000011011000001011100000000000000000 0000000000000000000000000000000011011000001100000000000000000000 0000000000000000000000000000000011011000001110100000000000000000 0000000000000000000000000000000011011000001111000000000000000000 0000000000000000000000000000000011011000001111100000000000000000 0000000000000000000000000000000011011000010000000000000000000000 0000000000000000000000000000000011011000010000100000000000000000 0000000000000000000000000000000011011000010001000000000000000000 0000000000000000000000000000000011011000010001100000000000000000 0000000000000000000000000000000011011000010010000000000000000000 0000000000000000000000000000000011011000010010100000000000000000 0000000000000000000000000000000011011000010011000000000000000000 0000000000000000000000000000000011011000010011100000000000000000 0000000000000000000000000000000011011000010100000000000000000000 0000000000000000000000000000000011011000010100100000000000000000 0000000000000000000000000000000011011000010101000000000000000000 0000000000000000000000000000000011011000010101100000000000000000 0000000000000000000000000000000011011000010110000000000000000000 0000000000000000000000000000000011011000010110100000000000000000 0000000000000000000000000000000011011000010111000000000000000000 0000000000000000000000000000000011011000010111100000000000000000 0000000000000000000000000000000011011000011000000000000000000000 0000000000000000000000000000000011011000011000100000000000000000 0000000000000000000000000000000011011000011001000000000000000000 0000000000000000000000000000000011011000011001100000000000000000 0000000000000000000000000000000011011000011010000000000000000000 0000000000000000000000000000000011011000011010100000000000000000 0000000000000000000000000000000011011000011011000000000000000000 0000000000000000000000000000000011011000011011100000000000000000 0000000000000000000000000000000011011000011100000000000000000000 0000000000000000000000000000000011011000011100100000000000000000 0000000000000000000000000000000011011000011101000000000000000000 0000000000000000000000000000000011011000011101100000000000000000 0000000000000000000000000000000011011000011110000000000000000000 0000000000000000000000000000000011011000011110100000000000000000 0000000000000000000000000000000011011000011111000000000000000000 0000000000000000000000000000000011011000011111100000000000000000 0000000000000000000000000000000011011000100000000000000000000000 0000000000000000000000000000000011011000100000100000000000000000 0000000000000000000000000000000011011000100001000000000000000000 0000000000000000000000000000000011011000100001100000000000000000 0000000000000000000000000000000011011000100010000000000000000000 0000000000000000000000000000000011011000100010100000000000000000 0000000000000000000000000000000011011000100011000000000000000000 0000000000000000000000000000000011011000100011100000000000000000 0000000000000000000000000000000011011000100100000000000000000000 0000000000000000000000000000000011011000100100100000000000000000 0000000000000000000000000000000011011000100101000000000000000000 0000000000000000000000000000000011011000100101100000000000000000 0000000000000000000000000000000011011000100110000000000000000000 0000000000000000000000000000000011011000100110100000000000000000 0000000000000000000000000000000011011000100111000000000000000000 0000000000000000000000000000000011011000100111100000000000000000 0000000000000000000000000000000011011000101000000000000000000000 0000000000000000000000000000000011011000101000100000000000000000 0000000000000000000000000000000011011000101001000000000000000000 0000000000000000000000000000000011011000101001100000000000000000 0000000000000000000000000000000011011000101010000000000000000000 0000000000000000000000000000000011011000101010100000000000000000 0000000000000000000000000000000011011000101011000000000000000000 0000000000000000000000000000000011011000101011100000000000000000 0000000000000000000000000000000011011000101100000000000000000000 0000000000000000000000000000000011011000101100100000000000000000 0000000000000000000000000000000011011000101101000000000000000000 0000000000000000000000000000000011011000101101100000000000000000 0000000000000000000000000000000011011000101110000000000000000000 0000000000000000000000000000000011011000110000000000000000000000 0000000000000000000000000000000011011000110000100000000000000000 0000000000000000000000000000000011011000110001000000000000000000 0000000000000000000000000000000011011000110001100000000000000000 0000000000000000000000000000000011011000110010000000000000000000 0000000000000000000000000000000011011000110010100000000000000000 0000000000000000000000000000000011011000110011000000000000000000 0000000000000000000000000000000011011000110011100000000000000000 0000000000000000000000000000000011011000110100000000000000000000 0000000000000000000000000000000011011000110100100000000000000000 0000000000000000000000000000000011011000110101000000000000000000 0000000000000000000000000000000011011000110101100000000000000000 0000000000000000000000000000000011011000110110000000000000000000 0000000000000000000000000000000011011000110110100000000000000000 0000000000000000000000000000000011011000110111000000000000000000 0000000000000000000000000000000011011000110111100000000000000000 0000000000000000000000000000000011011000111000000000000000000000 0000000000000000000000000000000011011000111000100000000000000000 0000000000000000000000000000000011011000111001000000000000000000 0000000000000000000000000000000011011000111001100000000000000000 0000000000000000000000000000000011011000111011000000000000000000 0000000000000000000000000000000011011000111011100000000000000000 0000000000000000000000000000000011011000111100000000000000000000 0000000000000000000000000000000011011000111110000000000000000000 0000000000000000000000000000000011011000111111000000000000000000 0000000000000000000000000000000011011001011011000000000000000000 0000000000000000000000000000000011011001011011100000000000000000 0000000000000000000000000000000011011001011100000000000000000000 0000000000000000000000000000000011011001011110100000000000000000 0000000000000000000000000000000011011001011111000000000000000000 0000000000000000000000000000000011011001101111000000000000000000 0000000000000000000000000000000011011001101111100000000000000000 0000000000000000000000000000000011011001110000000000000000000000 0000000000000000000000000000000011011001110000100000000000000000 0000000000000000000000000000000011011001110001000000000000000000 0000000000000000000000000000000011011001110001100000000000000000 0000000000000000000000000000000011011001111111000000000000000000 0000000000000000000000000000000011011001111111100000000000000000 default U 1 Local and global data share operations. ACC ACC field, specifies whether VDST use an accumulator VGPR 1 25 ADDR source LDS address VGPR 8 32 DATA0 source data 0 8 40 DATA1 source data 1 8 48 ENCODING Encoding. 6 26 GDS 1=GDS, 0=LDS 1 16 OFFSET0 First DS offset. See individual instructions for how this field is used. 8 0 OFFSET1 Second DS offset. See individual instructions for how this field is used. 8 8 OP Opcode. 8 17 VDST destination VGPR 8 56 ENC_MUBUF 64 0000000000000000000000000000000011111101111111000000000000000000 0000000000000000000000000000000011100000000000000000000000000000 0000000000000000000000000000000011100000000001000000000000000000 0000000000000000000000000000000011100000000010000000000000000000 0000000000000000000000000000000011100000000011000000000000000000 0000000000000000000000000000000011100000000100000000000000000000 0000000000000000000000000000000011100000000101000000000000000000 0000000000000000000000000000000011100000000110000000000000000000 0000000000000000000000000000000011100000000111000000000000000000 0000000000000000000000000000000011100000001000000000000000000000 0000000000000000000000000000000011100000001001000000000000000000 0000000000000000000000000000000011100000001010000000000000000000 0000000000000000000000000000000011100000001011000000000000000000 0000000000000000000000000000000011100000001100000000000000000000 0000000000000000000000000000000011100000001101000000000000000000 0000000000000000000000000000000011100000001110000000000000000000 0000000000000000000000000000000011100000001111000000000000000000 0000000000000000000000000000000011100000010000000000000000000000 0000000000000000000000000000000011100000010001000000000000000000 0000000000000000000000000000000011100000010010000000000000000000 0000000000000000000000000000000011100000010011000000000000000000 0000000000000000000000000000000011100000010100000000000000000000 0000000000000000000000000000000011100000010101000000000000000000 0000000000000000000000000000000011100000010110000000000000000000 0000000000000000000000000000000011100000010111000000000000000000 0000000000000000000000000000000011100000011000000000000000000000 0000000000000000000000000000000011100000011001000000000000000000 0000000000000000000000000000000011100000011010000000000000000000 0000000000000000000000000000000011100000011011000000000000000000 0000000000000000000000000000000011100000011100000000000000000000 0000000000000000000000000000000011100000011101000000000000000000 0000000000000000000000000000000011100000011110000000000000000000 0000000000000000000000000000000011100000011111000000000000000000 0000000000000000000000000000000011100000100000000000000000000000 0000000000000000000000000000000011100000100001000000000000000000 0000000000000000000000000000000011100000100010000000000000000000 0000000000000000000000000000000011100000100011000000000000000000 0000000000000000000000000000000011100000100100000000000000000000 0000000000000000000000000000000011100000100101000000000000000000 0000000000000000000000000000000011100000100110000000000000000000 0000000000000000000000000000000011100000100111000000000000000000 0000000000000000000000000000000011100000101000000000000000000000 0000000000000000000000000000000011100000101001000000000000000000 0000000000000000000000000000000011100001000000000000000000000000 0000000000000000000000000000000011100001000001000000000000000000 0000000000000000000000000000000011100001000010000000000000000000 0000000000000000000000000000000011100001000011000000000000000000 0000000000000000000000000000000011100001000100000000000000000000 0000000000000000000000000000000011100001000101000000000000000000 0000000000000000000000000000000011100001000110000000000000000000 0000000000000000000000000000000011100001000111000000000000000000 0000000000000000000000000000000011100001001000000000000000000000 0000000000000000000000000000000011100001001001000000000000000000 0000000000000000000000000000000011100001001010000000000000000000 0000000000000000000000000000000011100001001011000000000000000000 0000000000000000000000000000000011100001001100000000000000000000 0000000000000000000000000000000011100001001101000000000000000000 0000000000000000000000000000000011100001001110000000000000000000 0000000000000000000000000000000011100001001111000000000000000000 0000000000000000000000000000000011100001010000000000000000000000 0000000000000000000000000000000011100001010001000000000000000000 0000000000000000000000000000000011100001010010000000000000000000 0000000000000000000000000000000011100001100000000000000000000000 0000000000000000000000000000000011100001100001000000000000000000 0000000000000000000000000000000011100001100010000000000000000000 0000000000000000000000000000000011100001100011000000000000000000 0000000000000000000000000000000011100001100100000000000000000000 0000000000000000000000000000000011100001100101000000000000000000 0000000000000000000000000000000011100001100110000000000000000000 0000000000000000000000000000000011100001100111000000000000000000 0000000000000000000000000000000011100001101000000000000000000000 0000000000000000000000000000000011100001101001000000000000000000 0000000000000000000000000000000011100001101010000000000000000000 0000000000000000000000000000000011100001101011000000000000000000 0000000000000000000000000000000011100001101100000000000000000000 default U 1 Untyped vector memory buffer operations. ACC ACC field, specifies whether VDATA use an accumulator VGPR 1 55 ENCODING Encoding. 6 26 IDXEN If set, send VADDR as an index. If unset, treat the index as zero. 1 13 LDS If set, data is written to LDS memory. If unset, data is read from/written to a VGPR. 1 16 NT Non-Temporal: data is streamed or uncached. 1 17 OFFEN If set, send VADDR as an offset. If unset, send the instruction offset stored in OFFSET. Only one of these offsets may be sent. 1 12 OFFSET Unsigned byte offset 12 0 OP Opcode. 7 18 SC0 For regular load and store operations, SC0 and SC1 indicate memory scope such as wave, group, device and system. For atomics, SC0 means Atomic with return. 1 = Atomic with return, 0 = Atomic with no return. 1 14 SC1 For regular load and store operations, SC0 and SC1 indicate memory scope such as wave, group, device and system. 1 15 SOFFSET Scalar or constant GPR containing the base offset. 8 56 SRSRC Scalar GPR that specifies the resource constant, in units of 4 SGPRs. 5 48 2 00 VADDR Address source - may carry an offset or an index. 8 32 VDATA Vector GPR to read/write result to. 8 40 ENC_MTBUF 64 0000000000000000000000000000000011111100000001111000000000000000 0000000000000000000000000000000011101000000000000000000000000000 0000000000000000000000000000000011101000000000001000000000000000 0000000000000000000000000000000011101000000000010000000000000000 0000000000000000000000000000000011101000000000011000000000000000 0000000000000000000000000000000011101000000000100000000000000000 0000000000000000000000000000000011101000000000101000000000000000 0000000000000000000000000000000011101000000000110000000000000000 0000000000000000000000000000000011101000000000111000000000000000 0000000000000000000000000000000011101000000001000000000000000000 0000000000000000000000000000000011101000000001001000000000000000 0000000000000000000000000000000011101000000001010000000000000000 0000000000000000000000000000000011101000000001011000000000000000 0000000000000000000000000000000011101000000001100000000000000000 0000000000000000000000000000000011101000000001101000000000000000 0000000000000000000000000000000011101000000001110000000000000000 0000000000000000000000000000000011101000000001111000000000000000 default U 1 Typed vector memory buffer operations. ACC ACC field, specifies whether VDATA use an accumulator VGPR 1 55 DFMT Data format for typed buffer. 4 19 ENCODING Encoding. 6 26 IDXEN If set, send VADDR as an index. If unset, treat the index as zero. 1 13 NFMT Number format for typed buffer. 3 23 NT Non-Temporal: data is streamed or uncached. 1 54 OFFEN If set, send VADDR as an offset. If unset, send the instruction offset stored in OFFSET. Only one of these offsets may be sent. 1 12 OFFSET Unsigned byte offset. 12 0 OP opcode: read/write and x,xy,xyz,xyzw 4 15 SC0 For regular load and store operations, SC0 and SC1 indicate memory scope such as wave, group, device and system. For atomics, SC0 means Atomic with return. 1 = Atomic with return, 0 = Atomic with no return. 1 14 SC1 For regular load and store operations, SC0 and SC1 indicate memory scope such as wave, group, device and system. 1 53 SOFFSET Scalar GPR or constant containing the base offset. 8 56 SRSRC Scalar GPR that specifies the resource constant, in units of 4 SGPRs. 5 48 2 00 VADDR Address source - may carry an offset or an index. 8 32 VDATA Vector GPR to read/write result to. 8 40 ENC_FLAT 64 0000000000000000000000000000000011111101111111001100000000000000 0000000000000000000000000000000011011100010000000000000000000000 0000000000000000000000000000000011011100010001000000000000000000 0000000000000000000000000000000011011100010010000000000000000000 0000000000000000000000000000000011011100010011000000000000000000 0000000000000000000000000000000011011100010100000000000000000000 0000000000000000000000000000000011011100010101000000000000000000 0000000000000000000000000000000011011100010110000000000000000000 0000000000000000000000000000000011011100010111000000000000000000 0000000000000000000000000000000011011100011000000000000000000000 0000000000000000000000000000000011011100011001000000000000000000 0000000000000000000000000000000011011100011010000000000000000000 0000000000000000000000000000000011011100011011000000000000000000 0000000000000000000000000000000011011100011100000000000000000000 0000000000000000000000000000000011011100011101000000000000000000 0000000000000000000000000000000011011100011110000000000000000000 0000000000000000000000000000000011011100011111000000000000000000 0000000000000000000000000000000011011100100000000000000000000000 0000000000000000000000000000000011011100100001000000000000000000 0000000000000000000000000000000011011100100010000000000000000000 0000000000000000000000000000000011011100100011000000000000000000 0000000000000000000000000000000011011100100100000000000000000000 0000000000000000000000000000000011011100100101000000000000000000 0000000000000000000000000000000011011101000000000000000000000000 0000000000000000000000000000000011011101000001000000000000000000 0000000000000000000000000000000011011101000010000000000000000000 0000000000000000000000000000000011011101000011000000000000000000 0000000000000000000000000000000011011101000100000000000000000000 0000000000000000000000000000000011011101000101000000000000000000 0000000000000000000000000000000011011101000110000000000000000000 0000000000000000000000000000000011011101000111000000000000000000 0000000000000000000000000000000011011101001000000000000000000000 0000000000000000000000000000000011011101001001000000000000000000 0000000000000000000000000000000011011101001010000000000000000000 0000000000000000000000000000000011011101001011000000000000000000 0000000000000000000000000000000011011101001100000000000000000000 0000000000000000000000000000000011011101001101000000000000000000 0000000000000000000000000000000011011101001110000000000000000000 0000000000000000000000000000000011011101001111000000000000000000 0000000000000000000000000000000011011101010000000000000000000000 0000000000000000000000000000000011011101010001000000000000000000 0000000000000000000000000000000011011101010010000000000000000000 0000000000000000000000000000000011011101100000000000000000000000 0000000000000000000000000000000011011101100001000000000000000000 0000000000000000000000000000000011011101100010000000000000000000 0000000000000000000000000000000011011101100011000000000000000000 0000000000000000000000000000000011011101100100000000000000000000 0000000000000000000000000000000011011101100101000000000000000000 0000000000000000000000000000000011011101100110000000000000000000 0000000000000000000000000000000011011101100111000000000000000000 0000000000000000000000000000000011011101101000000000000000000000 0000000000000000000000000000000011011101101001000000000000000000 0000000000000000000000000000000011011101101010000000000000000000 0000000000000000000000000000000011011101101011000000000000000000 0000000000000000000000000000000011011101101100000000000000000000 default U 1 default U 1 default U 1 Flat (DUA) addressing instructions. ACC ACC field, specifies whether VDST use an accumulator VGPR 1 55 ADDR source flat address VGPR 8 32 DATA source data 8 40 ENCODING Encoding. 6 26 NT Non-Temporal: data is streamed or uncached. 1 17 OFFSET 12-bit unsigned offset 12 0 OP Opcode. 7 18 SADDR addr/offset SGPR 7 48 SC0 For regular load and store operations, SC0 and SC1 indicate memory scope such as wave, group, device and system. For atomics, SC0 means Atomic with return. 1 = Atomic with return, 0 = Atomic with no return. 1 16 SC1 For regular load and store operations, SC0 and SC1 indicate memory scope such as wave, group, device and system. 1 25 SEG FLAT=00 2 14 SVE SCRATCH VGPR ENABLE 1 13 VDST destination VGPR 8 56 ENC_FLAT_GLBL 64 0000000000000000000000000000000011111101111111001100000000000000 0000000000000000000000000000000011011100010000001000000000000000 0000000000000000000000000000000011011100010001001000000000000000 0000000000000000000000000000000011011100010010001000000000000000 0000000000000000000000000000000011011100010011001000000000000000 0000000000000000000000000000000011011100010100001000000000000000 0000000000000000000000000000000011011100010101001000000000000000 0000000000000000000000000000000011011100010110001000000000000000 0000000000000000000000000000000011011100010111001000000000000000 0000000000000000000000000000000011011100011000001000000000000000 0000000000000000000000000000000011011100011001001000000000000000 0000000000000000000000000000000011011100011010001000000000000000 0000000000000000000000000000000011011100011011001000000000000000 0000000000000000000000000000000011011100011100001000000000000000 0000000000000000000000000000000011011100011101001000000000000000 0000000000000000000000000000000011011100011110001000000000000000 0000000000000000000000000000000011011100011111001000000000000000 0000000000000000000000000000000011011100100000001000000000000000 0000000000000000000000000000000011011100100001001000000000000000 0000000000000000000000000000000011011100100010001000000000000000 0000000000000000000000000000000011011100100011001000000000000000 0000000000000000000000000000000011011100100100001000000000000000 0000000000000000000000000000000011011100100101001000000000000000 0000000000000000000000000000000011011100100110001000000000000000 0000000000000000000000000000000011011100100111001000000000000000 0000000000000000000000000000000011011100101000001000000000000000 0000000000000000000000000000000011011100101001001000000000000000 0000000000000000000000000000000011011100101010001000000000000000 0000000000000000000000000000000011011101000000001000000000000000 0000000000000000000000000000000011011101000001001000000000000000 0000000000000000000000000000000011011101000010001000000000000000 0000000000000000000000000000000011011101000011001000000000000000 0000000000000000000000000000000011011101000100001000000000000000 0000000000000000000000000000000011011101000101001000000000000000 0000000000000000000000000000000011011101000110001000000000000000 0000000000000000000000000000000011011101000111001000000000000000 0000000000000000000000000000000011011101001000001000000000000000 0000000000000000000000000000000011011101001001001000000000000000 0000000000000000000000000000000011011101001010001000000000000000 0000000000000000000000000000000011011101001011001000000000000000 0000000000000000000000000000000011011101001100001000000000000000 0000000000000000000000000000000011011101001101001000000000000000 0000000000000000000000000000000011011101001110001000000000000000 0000000000000000000000000000000011011101001111001000000000000000 0000000000000000000000000000000011011101010000001000000000000000 0000000000000000000000000000000011011101010001001000000000000000 0000000000000000000000000000000011011101010010001000000000000000 0000000000000000000000000000000011011101100000001000000000000000 0000000000000000000000000000000011011101100001001000000000000000 0000000000000000000000000000000011011101100010001000000000000000 0000000000000000000000000000000011011101100011001000000000000000 0000000000000000000000000000000011011101100100001000000000000000 0000000000000000000000000000000011011101100101001000000000000000 0000000000000000000000000000000011011101100110001000000000000000 0000000000000000000000000000000011011101100111001000000000000000 0000000000000000000000000000000011011101101000001000000000000000 0000000000000000000000000000000011011101101001001000000000000000 0000000000000000000000000000000011011101101010001000000000000000 0000000000000000000000000000000011011101101011001000000000000000 0000000000000000000000000000000011011101101100001000000000000000 0000000000000000000000000000000011011101111101001000000000000000 0000000000000000000000000000000011011101111110001000000000000000 See ENC_FLAT. ACC ACC field, specifies whether VDST use an accumulator VGPR 1 55 ADDR source flat address VGPR 8 32 DATA source data 8 40 ENCODING Encoding. 6 26 NT Non-Temporal: data is streamed or uncached. 1 17 OFFSET 13-bit signed offset 13 0 OP Opcode. 7 18 SADDR addr/offset SGPR 7 48 SC0 For regular load and store operations, SC0 and SC1 indicate memory scope such as wave, group, device and system. For atomics, SC0 means Atomic with return. 1 = Atomic with return, 0 = Atomic with no return. 1 16 SC1 For regular load and store operations, SC0 and SC1 indicate memory scope such as wave, group, device and system. 1 25 SEG GLOBAL=10 2 14 SVE SCRATCH VGPR ENABLE 1 13 VDST destination VGPR 8 56 ENC_FLAT_SCRATCH 64 0000000000000000000000000000000011111101111111001100000000000000 0000000000000000000000000000000011011100010000000100000000000000 0000000000000000000000000000000011011100010001000100000000000000 0000000000000000000000000000000011011100010010000100000000000000 0000000000000000000000000000000011011100010011000100000000000000 0000000000000000000000000000000011011100010100000100000000000000 0000000000000000000000000000000011011100010101000100000000000000 0000000000000000000000000000000011011100010110000100000000000000 0000000000000000000000000000000011011100010111000100000000000000 0000000000000000000000000000000011011100011000000100000000000000 0000000000000000000000000000000011011100011001000100000000000000 0000000000000000000000000000000011011100011010000100000000000000 0000000000000000000000000000000011011100011011000100000000000000 0000000000000000000000000000000011011100011100000100000000000000 0000000000000000000000000000000011011100011101000100000000000000 0000000000000000000000000000000011011100011110000100000000000000 0000000000000000000000000000000011011100011111000100000000000000 0000000000000000000000000000000011011100100000000100000000000000 0000000000000000000000000000000011011100100001000100000000000000 0000000000000000000000000000000011011100100010000100000000000000 0000000000000000000000000000000011011100100011000100000000000000 0000000000000000000000000000000011011100100100000100000000000000 0000000000000000000000000000000011011100100101000100000000000000 0000000000000000000000000000000011011100100110000100000000000000 0000000000000000000000000000000011011100100111000100000000000000 0000000000000000000000000000000011011100101000000100000000000000 0000000000000000000000000000000011011100101001000100000000000000 0000000000000000000000000000000011011100101010000100000000000000 See ENC_FLAT. ACC ACC field, specifies whether VDST use an accumulator VGPR 1 55 ADDR source flat address VGPR 8 32 DATA source data 8 40 ENCODING Encoding. 6 26 NT Non-Temporal: data is streamed or uncached. 1 17 OFFSET 13-bit signed offset 13 0 OP Opcode. 7 18 SADDR addr/offset SGPR 7 48 SC0 For regular load and store operations, SC0 and SC1 indicate memory scope such as wave, group, device and system. For atomics, SC0 means Atomic with return. 1 = Atomic with return, 0 = Atomic with no return. 1 16 SC1 For regular load and store operations, SC0 and SC1 indicate memory scope such as wave, group, device and system. 1 25 SEG SCRATCH=01 2 14 SVE SCRATCH VGPR ENABLE 1 13 VDST destination VGPR 8 56 SOP1_INST_LITERAL 64 0000000000000000000000000000000011111111100000001111111100000000 0000000000000000000000000000000010111110100000000000000000000000 0000000000000000000000000000000010111110100000000000000100000000 0000000000000000000000000000000010111110100000000000001000000000 0000000000000000000000000000000010111110100000000000001100000000 0000000000000000000000000000000010111110100000000000010000000000 0000000000000000000000000000000010111110100000000000010100000000 0000000000000000000000000000000010111110100000000000011000000000 0000000000000000000000000000000010111110100000000000011100000000 0000000000000000000000000000000010111110100000000000100000000000 0000000000000000000000000000000010111110100000000000100100000000 0000000000000000000000000000000010111110100000000000101000000000 0000000000000000000000000000000010111110100000000000101100000000 0000000000000000000000000000000010111110100000000000110000000000 0000000000000000000000000000000010111110100000000000110100000000 0000000000000000000000000000000010111110100000000000111000000000 0000000000000000000000000000000010111110100000000000111100000000 0000000000000000000000000000000010111110100000000001000000000000 0000000000000000000000000000000010111110100000000001000100000000 0000000000000000000000000000000010111110100000000001001000000000 0000000000000000000000000000000010111110100000000001001100000000 0000000000000000000000000000000010111110100000000001010000000000 0000000000000000000000000000000010111110100000000001010100000000 0000000000000000000000000000000010111110100000000001011000000000 0000000000000000000000000000000010111110100000000001011100000000 0000000000000000000000000000000010111110100000000001100000000000 0000000000000000000000000000000010111110100000000001100100000000 0000000000000000000000000000000010111110100000000001101000000000 0000000000000000000000000000000010111110100000000001101100000000 0000000000000000000000000000000010111110100000000010000000000000 0000000000000000000000000000000010111110100000000010000100000000 0000000000000000000000000000000010111110100000000010001000000000 0000000000000000000000000000000010111110100000000010001100000000 0000000000000000000000000000000010111110100000000010010000000000 0000000000000000000000000000000010111110100000000010010100000000 0000000000000000000000000000000010111110100000000010011000000000 0000000000000000000000000000000010111110100000000010011100000000 0000000000000000000000000000000010111110100000000010100000000000 0000000000000000000000000000000010111110100000000010100100000000 0000000000000000000000000000000010111110100000000010110000000000 0000000000000000000000000000000010111110100000000010110100000000 0000000000000000000000000000000010111110100000000010111000000000 0000000000000000000000000000000010111110100000000011000000000000 0000000000000000000000000000000010111110100000000011001000000000 0000000000000000000000000000000010111110100000000011001100000000 0000000000000000000000000000000010111110100000000011010000000000 0000000000000000000000000000000010111110100000000011010100000000 0000000000000000000000000000000010111110100000000011011000000000 0000000000000000000000000000000010111110100000000011011100000000 has_lit_0 == .fieldderef .fieldderef INST SOP1 SSRC0 I 32 255 I 32 U 1 Scalar ALU operations with one destination and one source. ENCODING Encoding. 9 23 OP Opcode. 8 8 SDST Destination for instruction. 7 16 SIMM32 32-bit literal constant. 32 32 SSRC0 Operand for instruction. 8 0 SOP2_INST_LITERAL 64 0000000000000000000000000000000011111111100000000000000000000000 0000000000000000000000000000000010000000000000000000000000000000 0000000000000000000000000000000010000000100000000000000000000000 0000000000000000000000000000000010000001000000000000000000000000 0000000000000000000000000000000010000001100000000000000000000000 0000000000000000000000000000000010000010000000000000000000000000 0000000000000000000000000000000010000010100000000000000000000000 0000000000000000000000000000000010000011000000000000000000000000 0000000000000000000000000000000010000011100000000000000000000000 0000000000000000000000000000000010000100000000000000000000000000 0000000000000000000000000000000010000100100000000000000000000000 0000000000000000000000000000000010000101000000000000000000000000 0000000000000000000000000000000010000101100000000000000000000000 0000000000000000000000000000000010000110000000000000000000000000 0000000000000000000000000000000010000110100000000000000000000000 0000000000000000000000000000000010000111000000000000000000000000 0000000000000000000000000000000010000111100000000000000000000000 0000000000000000000000000000000010001000000000000000000000000000 0000000000000000000000000000000010001000100000000000000000000000 0000000000000000000000000000000010001001000000000000000000000000 0000000000000000000000000000000010001001100000000000000000000000 0000000000000000000000000000000010001010000000000000000000000000 0000000000000000000000000000000010001010100000000000000000000000 0000000000000000000000000000000010001011000000000000000000000000 0000000000000000000000000000000010001011100000000000000000000000 0000000000000000000000000000000010001100000000000000000000000000 0000000000000000000000000000000010001100100000000000000000000000 0000000000000000000000000000000010001101000000000000000000000000 0000000000000000000000000000000010001101100000000000000000000000 0000000000000000000000000000000010001110000000000000000000000000 0000000000000000000000000000000010001110100000000000000000000000 0000000000000000000000000000000010001111000000000000000000000000 0000000000000000000000000000000010001111100000000000000000000000 0000000000000000000000000000000010010000000000000000000000000000 0000000000000000000000000000000010010000100000000000000000000000 0000000000000000000000000000000010010001000000000000000000000000 0000000000000000000000000000000010010001100000000000000000000000 0000000000000000000000000000000010010010000000000000000000000000 0000000000000000000000000000000010010010100000000000000000000000 0000000000000000000000000000000010010011000000000000000000000000 0000000000000000000000000000000010010011100000000000000000000000 0000000000000000000000000000000010010100000000000000000000000000 0000000000000000000000000000000010010100100000000000000000000000 0000000000000000000000000000000010010101000000000000000000000000 0000000000000000000000000000000010010110000000000000000000000000 0000000000000000000000000000000010010110100000000000000000000000 0000000000000000000000000000000010010111000000000000000000000000 0000000000000000000000000000000010010111100000000000000000000000 0000000000000000000000000000000010011000000000000000000000000000 0000000000000000000000000000000010011000100000000000000000000000 0000000000000000000000000000000010011001000000000000000000000000 0000000000000000000000000000000010011001100000000000000000000000 0000000000000000000000000000000010011010000000000000000000000000 has_lit_0 && == .fieldderef .fieldderef INST SOP2 SSRC0 I 32 255 I 32 U 1 != .fieldderef .fieldderef INST SOP2 SSRC1 I 32 255 I 32 U 1 U 1 has_lit_1 && != .fieldderef .fieldderef INST SOP2 SSRC0 I 32 255 I 32 U 1 == .fieldderef .fieldderef INST SOP2 SSRC1 I 32 255 I 32 U 1 U 1 has_lit_0_has_lit_1 && == .fieldderef .fieldderef INST SOP2 SSRC0 I 32 255 I 32 U 1 == .fieldderef .fieldderef INST SOP2 SSRC1 I 32 255 I 32 U 1 U 1 Scalar ALU operations with one destination and two sources. ENCODING Encoding. 2 30 OP Opcode. 7 23 SDST Destination for instruction. 7 16 SIMM32 32-bit literal constant. 32 32 SSRC0 First operand for instruction. 8 0 SSRC1 Second operand for instruction. 8 8 SOPC_INST_LITERAL 64 0000000000000000000000000000000011111111111111110000000000000000 0000000000000000000000000000000010111111000000000000000000000000 0000000000000000000000000000000010111111000000010000000000000000 0000000000000000000000000000000010111111000000100000000000000000 0000000000000000000000000000000010111111000000110000000000000000 0000000000000000000000000000000010111111000001000000000000000000 0000000000000000000000000000000010111111000001010000000000000000 0000000000000000000000000000000010111111000001100000000000000000 0000000000000000000000000000000010111111000001110000000000000000 0000000000000000000000000000000010111111000010000000000000000000 0000000000000000000000000000000010111111000010010000000000000000 0000000000000000000000000000000010111111000010100000000000000000 0000000000000000000000000000000010111111000010110000000000000000 0000000000000000000000000000000010111111000011000000000000000000 0000000000000000000000000000000010111111000011010000000000000000 0000000000000000000000000000000010111111000011100000000000000000 0000000000000000000000000000000010111111000011110000000000000000 0000000000000000000000000000000010111111000100000000000000000000 0000000000000000000000000000000010111111000100010000000000000000 0000000000000000000000000000000010111111000100100000000000000000 0000000000000000000000000000000010111111000100110000000000000000 has_lit_0 && == .fieldderef .fieldderef INST SOPC SSRC0 I 32 255 I 32 U 1 != .fieldderef .fieldderef INST SOPC SSRC1 I 32 255 I 32 U 1 U 1 has_lit_1 && != .fieldderef .fieldderef INST SOPC SSRC0 I 32 255 I 32 U 1 == .fieldderef .fieldderef INST SOPC SSRC1 I 32 255 I 32 U 1 U 1 has_lit_0_has_lit_1 && == .fieldderef .fieldderef INST SOPC SSRC0 I 32 255 I 32 U 1 == .fieldderef .fieldderef INST SOPC SSRC1 I 32 255 I 32 U 1 U 1 Scalar ALU comparison operations with two sources. ENCODING Encoding. 9 23 OP Opcode. 7 16 SIMM32 32-bit literal constant. 32 32 SSRC0 First operand for instruction. 8 0 SSRC1 Second operand for instruction. 8 8 SOPK_INST_LITERAL 64 0000000000000000000000000000000011111111100000000000000000000000 0000000000000000000000000000000010111010000000000000000000000000 default U 1 Scalar ALU operations with one destination and one 16-bit constant source. ENCODING Encoding. 4 28 OP Opcode. 5 23 SDST Destination for instruction. 7 16 SIMM16 16-bit integer input for opcode. Signedness is determined by opcode. 16 0 SIMM32 32-bit literal constant. 32 32 VOP1_INST_LITERAL 64 0000000000000000000000000000000011111110000000011111111000000000 0000000000000000000000000000000001111110000000000000000000000000 0000000000000000000000000000000001111110000000000000001000000000 0000000000000000000000000000000001111110000000000000011000000000 0000000000000000000000000000000001111110000000000000100000000000 0000000000000000000000000000000001111110000000000000101000000000 0000000000000000000000000000000001111110000000000000110000000000 0000000000000000000000000000000001111110000000000000111000000000 0000000000000000000000000000000001111110000000000001000000000000 0000000000000000000000000000000001111110000000000001010000000000 0000000000000000000000000000000001111110000000000001011000000000 0000000000000000000000000000000001111110000000000001100000000000 0000000000000000000000000000000001111110000000000001101000000000 0000000000000000000000000000000001111110000000000001110000000000 0000000000000000000000000000000001111110000000000001111000000000 0000000000000000000000000000000001111110000000000010000000000000 0000000000000000000000000000000001111110000000000010001000000000 0000000000000000000000000000000001111110000000000010010000000000 0000000000000000000000000000000001111110000000000010011000000000 0000000000000000000000000000000001111110000000000010100000000000 0000000000000000000000000000000001111110000000000010101000000000 0000000000000000000000000000000001111110000000000010110000000000 0000000000000000000000000000000001111110000000000010111000000000 0000000000000000000000000000000001111110000000000011000000000000 0000000000000000000000000000000001111110000000000011001000000000 0000000000000000000000000000000001111110000000000011010000000000 0000000000000000000000000000000001111110000000000011011000000000 0000000000000000000000000000000001111110000000000011100000000000 0000000000000000000000000000000001111110000000000011101000000000 0000000000000000000000000000000001111110000000000011110000000000 0000000000000000000000000000000001111110000000000011111000000000 0000000000000000000000000000000001111110000000000100000000000000 0000000000000000000000000000000001111110000000000100001000000000 0000000000000000000000000000000001111110000000000100010000000000 0000000000000000000000000000000001111110000000000100011000000000 0000000000000000000000000000000001111110000000000100100000000000 0000000000000000000000000000000001111110000000000100101000000000 0000000000000000000000000000000001111110000000000100110000000000 0000000000000000000000000000000001111110000000000100111000000000 0000000000000000000000000000000001111110000000000101000000000000 0000000000000000000000000000000001111110000000000101001000000000 0000000000000000000000000000000001111110000000000101010000000000 0000000000000000000000000000000001111110000000000101011000000000 0000000000000000000000000000000001111110000000000101100000000000 0000000000000000000000000000000001111110000000000101101000000000 0000000000000000000000000000000001111110000000000101110000000000 0000000000000000000000000000000001111110000000000101111000000000 0000000000000000000000000000000001111110000000000110000000000000 0000000000000000000000000000000001111110000000000110001000000000 0000000000000000000000000000000001111110000000000110010000000000 0000000000000000000000000000000001111110000000000110011000000000 0000000000000000000000000000000001111110000000000110100000000000 0000000000000000000000000000000001111110000000000110101000000000 0000000000000000000000000000000001111110000000000110111000000000 0000000000000000000000000000000001111110000000000111000000000000 0000000000000000000000000000000001111110000000000111001000000000 0000000000000000000000000000000001111110000000000111010000000000 0000000000000000000000000000000001111110000000000111011000000000 0000000000000000000000000000000001111110000000000111100000000000 0000000000000000000000000000000001111110000000000111101000000000 0000000000000000000000000000000001111110000000000111110000000000 0000000000000000000000000000000001111110000000000111111000000000 0000000000000000000000000000000001111110000000001000000000000000 0000000000000000000000000000000001111110000000001000001000000000 0000000000000000000000000000000001111110000000001000010000000000 0000000000000000000000000000000001111110000000001000011000000000 0000000000000000000000000000000001111110000000001000100000000000 0000000000000000000000000000000001111110000000001000101000000000 0000000000000000000000000000000001111110000000001000110000000000 0000000000000000000000000000000001111110000000001000111000000000 0000000000000000000000000000000001111110000000001001000000000000 0000000000000000000000000000000001111110000000001001001000000000 0000000000000000000000000000000001111110000000001001010000000000 0000000000000000000000000000000001111110000000001001011000000000 0000000000000000000000000000000001111110000000001001100000000000 0000000000000000000000000000000001111110000000001001101000000000 0000000000000000000000000000000001111110000000001001110000000000 0000000000000000000000000000000001111110000000001001111000000000 0000000000000000000000000000000001111110000000001010010000000000 0000000000000000000000000000000001111110000000001010100000000000 0000000000000000000000000000000001111110000000001010101000000000 0000000000000000000000000000000001111110000000001010110000000000 0000000000000000000000000000000001111110000000001010111000000000 0000000000000000000000000000000001111110000000001011000000000000 0000000000000000000000000000000001111110000000001011011000000000 has_lit == .fieldderef .fieldderef INST VOP1 SRC0 I 32 255 I 32 U 1 Vector ALU operations with one destination and one source. Instructions in this encoding may be promoted to VOP3 unless otherwise noted. Instructions in this encoding may use SDWA or DPP modes unless otherwise noted. ENCODING Encoding. 7 25 OP Opcode. 8 9 SIMM32 32-bit literal constant. 32 32 SRC0 First operand for instruction. 9 0 VDST Destination for instruction. 8 17 VOP1_VOP_DPP 64 0000000000000000000000000000000011111110000000011111111000000000 0000000000000000000000000000000001111110000000000000000000000000 0000000000000000000000000000000001111110000000000000001000000000 0000000000000000000000000000000001111110000000000000011000000000 0000000000000000000000000000000001111110000000000000100000000000 0000000000000000000000000000000001111110000000000000101000000000 0000000000000000000000000000000001111110000000000000110000000000 0000000000000000000000000000000001111110000000000000111000000000 0000000000000000000000000000000001111110000000000001000000000000 0000000000000000000000000000000001111110000000000001010000000000 0000000000000000000000000000000001111110000000000001011000000000 0000000000000000000000000000000001111110000000000001100000000000 0000000000000000000000000000000001111110000000000001101000000000 0000000000000000000000000000000001111110000000000001110000000000 0000000000000000000000000000000001111110000000000001111000000000 0000000000000000000000000000000001111110000000000010001000000000 0000000000000000000000000000000001111110000000000010010000000000 0000000000000000000000000000000001111110000000000010011000000000 0000000000000000000000000000000001111110000000000010100000000000 0000000000000000000000000000000001111110000000000010101000000000 0000000000000000000000000000000001111110000000000010110000000000 0000000000000000000000000000000001111110000000000010111000000000 0000000000000000000000000000000001111110000000000011000000000000 0000000000000000000000000000000001111110000000000011010000000000 0000000000000000000000000000000001111110000000000011011000000000 0000000000000000000000000000000001111110000000000011100000000000 0000000000000000000000000000000001111110000000000011101000000000 0000000000000000000000000000000001111110000000000011110000000000 0000000000000000000000000000000001111110000000000011111000000000 0000000000000000000000000000000001111110000000000100000000000000 0000000000000000000000000000000001111110000000000100001000000000 0000000000000000000000000000000001111110000000000100010000000000 0000000000000000000000000000000001111110000000000100011000000000 0000000000000000000000000000000001111110000000000100100000000000 0000000000000000000000000000000001111110000000000100111000000000 0000000000000000000000000000000001111110000000000101001000000000 0000000000000000000000000000000001111110000000000101010000000000 0000000000000000000000000000000001111110000000000101011000000000 0000000000000000000000000000000001111110000000000101100000000000 0000000000000000000000000000000001111110000000000101101000000000 0000000000000000000000000000000001111110000000000101110000000000 0000000000000000000000000000000001111110000000000101111000000000 0000000000000000000000000000000001111110000000000110000000000000 0000000000000000000000000000000001111110000000000110001000000000 0000000000000000000000000000000001111110000000000110010000000000 0000000000000000000000000000000001111110000000000110011000000000 0000000000000000000000000000000001111110000000000110100000000000 0000000000000000000000000000000001111110000000000110111000000000 0000000000000000000000000000000001111110000000000111000000000000 0000000000000000000000000000000001111110000000000111001000000000 0000000000000000000000000000000001111110000000000111010000000000 0000000000000000000000000000000001111110000000000111011000000000 0000000000000000000000000000000001111110000000000111100000000000 0000000000000000000000000000000001111110000000000111101000000000 0000000000000000000000000000000001111110000000000111110000000000 0000000000000000000000000000000001111110000000000111111000000000 0000000000000000000000000000000001111110000000001000000000000000 0000000000000000000000000000000001111110000000001000001000000000 0000000000000000000000000000000001111110000000001000010000000000 0000000000000000000000000000000001111110000000001000011000000000 0000000000000000000000000000000001111110000000001000100000000000 0000000000000000000000000000000001111110000000001000101000000000 0000000000000000000000000000000001111110000000001000110000000000 0000000000000000000000000000000001111110000000001000111000000000 0000000000000000000000000000000001111110000000001001000000000000 0000000000000000000000000000000001111110000000001001001000000000 0000000000000000000000000000000001111110000000001001010000000000 0000000000000000000000000000000001111110000000001001011000000000 0000000000000000000000000000000001111110000000001001100000000000 0000000000000000000000000000000001111110000000001001101000000000 0000000000000000000000000000000001111110000000001001110000000000 0000000000000000000000000000000001111110000000001001111000000000 0000000000000000000000000000000001111110000000001010010000000000 0000000000000000000000000000000001111110000000001010100000000000 0000000000000000000000000000000001111110000000001010101000000000 0000000000000000000000000000000001111110000000001010110000000000 0000000000000000000000000000000001111110000000001010111000000000 0000000000000000000000000000000001111110000000001011000000000000 0000000000000000000000000000000001111110000000001011011000000000 has_dpp == .fieldderef .fieldderef INST VOP1 SRC0 I 32 250 I 32 U 1 Vector ALU operations with one destination and one source. Instructions in this encoding may be promoted to VOP3 unless otherwise noted. Instructions in this encoding may use SDWA or DPP modes unless otherwise noted. BANK_MASK Bank enable mask. If bit N is set, then lanes[4*N:4*N+3, 4*N+16:4*N+19, 4*N+32:4*N+35, 4*N+48:4*N+51] are enabled. 4 56 BOUND_CTRL Specifies behaviour when shared data is invalid. 0: set write-enable to zero. 1: set source data to zero. 1 51 DPP_CTRL Data-parallel primitive control. 9 40 ENCODING Encoding. 7 25 OP Opcode. 8 9 ROW_MASK Row enable mask. If bit N is set, then lanes[16*N:16*N+15] are enabled. 4 60 SRC0 First operand for instruction. 9 0 SRC0_ABS If true, take the floating-point absolute value of src0. 1 53 SRC0_NEG If true, take the floating-point negation of src0. 1 52 SRC1_ABS If true, take the floating-point absolute value of src1. 1 55 SRC1_NEG If true, take the floating-point negation of src1. 1 54 VDST Destination for instruction. 8 17 VSRC0 VGPR address for src0. 8 32 VOP1_VOP_SDWA 64 0000000000000000000000000000000011111110000000011111111000000000 0000000000000000000000000000000001111110000000000000000000000000 0000000000000000000000000000000001111110000000000000001000000000 0000000000000000000000000000000001111110000000000000101000000000 0000000000000000000000000000000001111110000000000000110000000000 0000000000000000000000000000000001111110000000000000111000000000 0000000000000000000000000000000001111110000000000001000000000000 0000000000000000000000000000000001111110000000000001010000000000 0000000000000000000000000000000001111110000000000001011000000000 0000000000000000000000000000000001111110000000000001100000000000 0000000000000000000000000000000001111110000000000001101000000000 0000000000000000000000000000000001111110000000000001110000000000 0000000000000000000000000000000001111110000000000010001000000000 0000000000000000000000000000000001111110000000000010010000000000 0000000000000000000000000000000001111110000000000010011000000000 0000000000000000000000000000000001111110000000000010100000000000 0000000000000000000000000000000001111110000000000011011000000000 0000000000000000000000000000000001111110000000000011100000000000 0000000000000000000000000000000001111110000000000011101000000000 0000000000000000000000000000000001111110000000000011110000000000 0000000000000000000000000000000001111110000000000011111000000000 0000000000000000000000000000000001111110000000000100000000000000 0000000000000000000000000000000001111110000000000100001000000000 0000000000000000000000000000000001111110000000000100010000000000 0000000000000000000000000000000001111110000000000100011000000000 0000000000000000000000000000000001111110000000000100100000000000 0000000000000000000000000000000001111110000000000100111000000000 0000000000000000000000000000000001111110000000000101001000000000 0000000000000000000000000000000001111110000000000101010000000000 0000000000000000000000000000000001111110000000000101011000000000 0000000000000000000000000000000001111110000000000101100000000000 0000000000000000000000000000000001111110000000000101101000000000 0000000000000000000000000000000001111110000000000101110000000000 0000000000000000000000000000000001111110000000000101111000000000 0000000000000000000000000000000001111110000000000110011000000000 0000000000000000000000000000000001111110000000000110100000000000 0000000000000000000000000000000001111110000000000110111000000000 0000000000000000000000000000000001111110000000000111000000000000 0000000000000000000000000000000001111110000000000111001000000000 0000000000000000000000000000000001111110000000000111010000000000 0000000000000000000000000000000001111110000000000111011000000000 0000000000000000000000000000000001111110000000000111100000000000 0000000000000000000000000000000001111110000000000111101000000000 0000000000000000000000000000000001111110000000000111110000000000 0000000000000000000000000000000001111110000000000111111000000000 0000000000000000000000000000000001111110000000001000000000000000 0000000000000000000000000000000001111110000000001000001000000000 0000000000000000000000000000000001111110000000001000010000000000 0000000000000000000000000000000001111110000000001000011000000000 0000000000000000000000000000000001111110000000001000100000000000 0000000000000000000000000000000001111110000000001000101000000000 0000000000000000000000000000000001111110000000001000110000000000 0000000000000000000000000000000001111110000000001000111000000000 0000000000000000000000000000000001111110000000001001000000000000 0000000000000000000000000000000001111110000000001001001000000000 0000000000000000000000000000000001111110000000001001010000000000 0000000000000000000000000000000001111110000000001001011000000000 0000000000000000000000000000000001111110000000001001100000000000 0000000000000000000000000000000001111110000000001001101000000000 0000000000000000000000000000000001111110000000001001110000000000 0000000000000000000000000000000001111110000000001001111000000000 0000000000000000000000000000000001111110000000001010010000000000 0000000000000000000000000000000001111110000000001010100000000000 0000000000000000000000000000000001111110000000001010101000000000 0000000000000000000000000000000001111110000000001010110000000000 0000000000000000000000000000000001111110000000001010111000000000 0000000000000000000000000000000001111110000000001011000000000000 0000000000000000000000000000000001111110000000001011011000000000 has_sdwa == .fieldderef .fieldderef INST VOP1 SRC0 I 32 249 I 32 U 1 Vector ALU operations with one destination and one source. Instructions in this encoding may be promoted to VOP3 unless otherwise noted. Instructions in this encoding may use SDWA or DPP modes unless otherwise noted. CLAMP If true, clamp output. 1 45 DST_SEL Destination data select. 3 40 DST_UNUSED Format for unused destination bits. 2 43 ENCODING Encoding. 7 25 OMOD Output modifier for instruction. 2 46 OP Opcode. 8 9 S0 If 0 vsrc0 is VGPR address. If 1 vsrc0 is SGPR or built in inline constant. 1 55 S1 If 0 vsrc1 in VOP2/VOPC is VGPR address. If 1 vsrc1 is SGPR or built in inline constant. 1 63 SRC0 First operand for instruction. 9 0 SRC0_ABS If true, take the floating-point absolute value of src0. 1 53 SRC0_NEG If true, take the floating-point negation of src0. 1 52 SRC0_SEL Source data select for src0. 3 48 SRC0_SEXT If true, sign-extend data for src0. If false, zero-extend. 1 51 SRC1_ABS If true, take the floating-point absolute value of src1. 1 61 SRC1_NEG If true, take the floating-point negation of src1. 1 60 SRC1_SEL Source data select for src1. 3 56 SRC1_SEXT If true, sign-extend data for src1. If false, zero-extend. 1 59 VDST Destination for instruction. 8 17 VSRC0 VGPR address for src0. 8 32 VOP2_INST_LITERAL 64 0000000000000000000000000000000011111110000000000000000000000000 0000000000000000000000000000000000000000000000000000000000000000 0000000000000000000000000000000000000010000000000000000000000000 0000000000000000000000000000000000000100000000000000000000000000 0000000000000000000000000000000000000110000000000000000000000000 0000000000000000000000000000000000001000000000000000000000000000 0000000000000000000000000000000000001010000000000000000000000000 0000000000000000000000000000000000001100000000000000000000000000 0000000000000000000000000000000000001110000000000000000000000000 0000000000000000000000000000000000010000000000000000000000000000 0000000000000000000000000000000000010010000000000000000000000000 0000000000000000000000000000000000010100000000000000000000000000 0000000000000000000000000000000000010110000000000000000000000000 0000000000000000000000000000000000011000000000000000000000000000 0000000000000000000000000000000000011010000000000000000000000000 0000000000000000000000000000000000011100000000000000000000000000 0000000000000000000000000000000000011110000000000000000000000000 0000000000000000000000000000000000100000000000000000000000000000 0000000000000000000000000000000000100010000000000000000000000000 0000000000000000000000000000000000100100000000000000000000000000 0000000000000000000000000000000000100110000000000000000000000000 0000000000000000000000000000000000101000000000000000000000000000 0000000000000000000000000000000000101010000000000000000000000000 0000000000000000000000000000000000101100000000000000000000000000 0000000000000000000000000000000000101110000000000000000000000000 0000000000000000000000000000000000110000000000000000000000000000 0000000000000000000000000000000000110010000000000000000000000000 0000000000000000000000000000000000110100000000000000000000000000 0000000000000000000000000000000000110110000000000000000000000000 0000000000000000000000000000000000111000000000000000000000000000 0000000000000000000000000000000000111010000000000000000000000000 0000000000000000000000000000000000111100000000000000000000000000 0000000000000000000000000000000000111110000000000000000000000000 0000000000000000000000000000000001000000000000000000000000000000 0000000000000000000000000000000001000010000000000000000000000000 0000000000000000000000000000000001000100000000000000000000000000 0000000000000000000000000000000001000110000000000000000000000000 0000000000000000000000000000000001001000000000000000000000000000 0000000000000000000000000000000001001010000000000000000000000000 0000000000000000000000000000000001001100000000000000000000000000 0000000000000000000000000000000001001110000000000000000000000000 0000000000000000000000000000000001010000000000000000000000000000 0000000000000000000000000000000001010010000000000000000000000000 0000000000000000000000000000000001010100000000000000000000000000 0000000000000000000000000000000001010110000000000000000000000000 0000000000000000000000000000000001011000000000000000000000000000 0000000000000000000000000000000001011010000000000000000000000000 0000000000000000000000000000000001011100000000000000000000000000 0000000000000000000000000000000001011110000000000000000000000000 0000000000000000000000000000000001100000000000000000000000000000 0000000000000000000000000000000001100010000000000000000000000000 0000000000000000000000000000000001100100000000000000000000000000 0000000000000000000000000000000001100110000000000000000000000000 0000000000000000000000000000000001101000000000000000000000000000 0000000000000000000000000000000001101010000000000000000000000000 0000000000000000000000000000000001101100000000000000000000000000 0000000000000000000000000000000001101110000000000000000000000000 0000000000000000000000000000000001110000000000000000000000000000 0000000000000000000000000000000001110010000000000000000000000000 0000000000000000000000000000000001110100000000000000000000000000 0000000000000000000000000000000001110110000000000000000000000000 0000000000000000000000000000000001111000000000000000000000000000 0000000000000000000000000000000001111010000000000000000000000000 has_lit == .fieldderef .fieldderef INST VOP2 SRC0 I 32 255 I 32 U 1 default && && != .fieldderef .fieldderef INST VOP2 SRC0 I 32 250 I 32 U 1 != .fieldderef .fieldderef INST VOP2 SRC0 I 32 255 I 32 U 1 U 1 != .fieldderef .fieldderef INST VOP2 SRC0 I 32 249 I 32 U 1 U 1 Vector ALU operations with one destination and two sources. Instructions in this encoding may be promoted to VOP3 unless otherwise noted. ENCODING Encoding. 1 31 OP Opcode. 6 25 SIMM32 32-bit literal constant. 32 32 SRC0 First operand for instruction. 9 0 VDST Destination for instruction. 8 17 VSRC1 Second operand for instruction. 8 9 VOP2_VOP_DPP 64 0000000000000000000000000000000011111110000000000000000000000000 0000000000000000000000000000000000000000000000000000000000000000 0000000000000000000000000000000000000010000000000000000000000000 0000000000000000000000000000000000000100000000000000000000000000 0000000000000000000000000000000000000110000000000000000000000000 0000000000000000000000000000000000001000000000000000000000000000 0000000000000000000000000000000000001010000000000000000000000000 0000000000000000000000000000000000001100000000000000000000000000 0000000000000000000000000000000000001110000000000000000000000000 0000000000000000000000000000000000010000000000000000000000000000 0000000000000000000000000000000000010010000000000000000000000000 0000000000000000000000000000000000010100000000000000000000000000 0000000000000000000000000000000000010110000000000000000000000000 0000000000000000000000000000000000011000000000000000000000000000 0000000000000000000000000000000000011010000000000000000000000000 0000000000000000000000000000000000011100000000000000000000000000 0000000000000000000000000000000000011110000000000000000000000000 0000000000000000000000000000000000100000000000000000000000000000 0000000000000000000000000000000000100010000000000000000000000000 0000000000000000000000000000000000100100000000000000000000000000 0000000000000000000000000000000000100110000000000000000000000000 0000000000000000000000000000000000101000000000000000000000000000 0000000000000000000000000000000000101010000000000000000000000000 0000000000000000000000000000000000101100000000000000000000000000 0000000000000000000000000000000000110010000000000000000000000000 0000000000000000000000000000000000110100000000000000000000000000 0000000000000000000000000000000000110110000000000000000000000000 0000000000000000000000000000000000111000000000000000000000000000 0000000000000000000000000000000000111010000000000000000000000000 0000000000000000000000000000000000111100000000000000000000000000 0000000000000000000000000000000000111110000000000000000000000000 0000000000000000000000000000000001000000000000000000000000000000 0000000000000000000000000000000001000010000000000000000000000000 0000000000000000000000000000000001000100000000000000000000000000 0000000000000000000000000000000001000110000000000000000000000000 0000000000000000000000000000000001001100000000000000000000000000 0000000000000000000000000000000001001110000000000000000000000000 0000000000000000000000000000000001010000000000000000000000000000 0000000000000000000000000000000001010010000000000000000000000000 0000000000000000000000000000000001010100000000000000000000000000 0000000000000000000000000000000001010110000000000000000000000000 0000000000000000000000000000000001011000000000000000000000000000 0000000000000000000000000000000001011010000000000000000000000000 0000000000000000000000000000000001011100000000000000000000000000 0000000000000000000000000000000001011110000000000000000000000000 0000000000000000000000000000000001100000000000000000000000000000 0000000000000000000000000000000001100010000000000000000000000000 0000000000000000000000000000000001100100000000000000000000000000 0000000000000000000000000000000001100110000000000000000000000000 0000000000000000000000000000000001101000000000000000000000000000 0000000000000000000000000000000001101010000000000000000000000000 0000000000000000000000000000000001101100000000000000000000000000 0000000000000000000000000000000001101110000000000000000000000000 0000000000000000000000000000000001110000000000000000000000000000 0000000000000000000000000000000001110010000000000000000000000000 0000000000000000000000000000000001110100000000000000000000000000 0000000000000000000000000000000001110110000000000000000000000000 0000000000000000000000000000000001111000000000000000000000000000 0000000000000000000000000000000001111010000000000000000000000000 has_dpp == .fieldderef .fieldderef INST VOP2 SRC0 I 32 250 I 32 U 1 Vector ALU operations with one destination and two sources. Instructions in this encoding may be promoted to VOP3 unless otherwise noted. BANK_MASK Bank enable mask. If bit N is set, then lanes[4*N:4*N+3, 4*N+16:4*N+19, 4*N+32:4*N+35, 4*N+48:4*N+51] are enabled. 4 56 BOUND_CTRL Specifies behaviour when shared data is invalid. 0: set write-enable to zero. 1: set source data to zero. 1 51 DPP_CTRL Data-parallel primitive control. 9 40 ENCODING Encoding. 1 31 OP Opcode. 6 25 ROW_MASK Row enable mask. If bit N is set, then lanes[16*N:16*N+15] are enabled. 4 60 SRC0 First operand for instruction. 9 0 SRC0_ABS If true, take the floating-point absolute value of src0. 1 53 SRC0_NEG If true, take the floating-point negation of src0. 1 52 SRC1_ABS If true, take the floating-point absolute value of src1. 1 55 SRC1_NEG If true, take the floating-point negation of src1. 1 54 VDST Destination for instruction. 8 17 VSRC0 VGPR address for src0. 8 32 VSRC1 Second operand for instruction. 8 9 VOP2_VOP_SDWA 64 0000000000000000000000000000000011111110000000000000000000000000 0000000000000000000000000000000000000000000000000000000000000000 0000000000000000000000000000000000000010000000000000000000000000 0000000000000000000000000000000000000100000000000000000000000000 0000000000000000000000000000000000000110000000000000000000000000 0000000000000000000000000000000000001010000000000000000000000000 0000000000000000000000000000000000001100000000000000000000000000 0000000000000000000000000000000000001110000000000000000000000000 0000000000000000000000000000000000010000000000000000000000000000 0000000000000000000000000000000000010010000000000000000000000000 0000000000000000000000000000000000010100000000000000000000000000 0000000000000000000000000000000000010110000000000000000000000000 0000000000000000000000000000000000011000000000000000000000000000 0000000000000000000000000000000000011010000000000000000000000000 0000000000000000000000000000000000011100000000000000000000000000 0000000000000000000000000000000000011110000000000000000000000000 0000000000000000000000000000000000100000000000000000000000000000 0000000000000000000000000000000000100010000000000000000000000000 0000000000000000000000000000000000100100000000000000000000000000 0000000000000000000000000000000000100110000000000000000000000000 0000000000000000000000000000000000101000000000000000000000000000 0000000000000000000000000000000000101010000000000000000000000000 0000000000000000000000000000000000111110000000000000000000000000 0000000000000000000000000000000001000000000000000000000000000000 0000000000000000000000000000000001000010000000000000000000000000 0000000000000000000000000000000001000100000000000000000000000000 0000000000000000000000000000000001001100000000000000000000000000 0000000000000000000000000000000001001110000000000000000000000000 0000000000000000000000000000000001010000000000000000000000000000 0000000000000000000000000000000001010010000000000000000000000000 0000000000000000000000000000000001010100000000000000000000000000 0000000000000000000000000000000001010110000000000000000000000000 0000000000000000000000000000000001011000000000000000000000000000 0000000000000000000000000000000001011010000000000000000000000000 0000000000000000000000000000000001011100000000000000000000000000 0000000000000000000000000000000001011110000000000000000000000000 0000000000000000000000000000000001100000000000000000000000000000 0000000000000000000000000000000001100010000000000000000000000000 0000000000000000000000000000000001100100000000000000000000000000 0000000000000000000000000000000001100110000000000000000000000000 0000000000000000000000000000000001101000000000000000000000000000 0000000000000000000000000000000001101010000000000000000000000000 0000000000000000000000000000000001101100000000000000000000000000 0000000000000000000000000000000001111010000000000000000000000000 has_sdwa == .fieldderef .fieldderef INST VOP2 SRC0 I 32 249 I 32 U 1 Vector ALU operations with one destination and two sources. Instructions in this encoding may be promoted to VOP3 unless otherwise noted. CLAMP If true, clamp output. 1 45 DST_SEL Destination data select. 3 40 DST_UNUSED Format for unused destination bits. 2 43 ENCODING Encoding. 1 31 OMOD Output modifier for instruction. 2 46 OP Opcode. 6 25 S0 If 0 vsrc0 is VGPR address. If 1 vsrc0 is SGPR or built in inline constant. 1 55 S1 If 0 vsrc1 in VOP2/VOPC is VGPR address. If 1 vsrc1 is SGPR or built in inline constant. 1 63 SRC0 First operand for instruction. 9 0 SRC0_ABS If true, take the floating-point absolute value of src0. 1 53 SRC0_NEG If true, take the floating-point negation of src0. 1 52 SRC0_SEL Source data select for src0. 3 48 SRC0_SEXT If true, sign-extend data for src0. If false, zero-extend. 1 51 SRC1_ABS If true, take the floating-point absolute value of src1. 1 61 SRC1_NEG If true, take the floating-point negation of src1. 1 60 SRC1_SEL Source data select for src1. 3 56 SRC1_SEXT If true, sign-extend data for src1. If false, zero-extend. 1 59 VDST Destination for instruction. 8 17 VSRC0 VGPR address for src0. 8 32 VSRC1 Second operand for instruction. 8 9 VOP2_VOP_SDWA_SDST_ENC 64 0000000000000000000000000000000011111110000000000000000000000000 0000000000000000000000000000000000110010000000000000000000000000 0000000000000000000000000000000000110100000000000000000000000000 0000000000000000000000000000000000110110000000000000000000000000 0000000000000000000000000000000000111000000000000000000000000000 0000000000000000000000000000000000111010000000000000000000000000 0000000000000000000000000000000000111100000000000000000000000000 has_sdwa == .fieldderef .fieldderef INST VOP2 SRC0 I 32 249 I 32 U 1 Vector ALU operations with one destination and two sources. Instructions in this encoding may be promoted to VOP3 unless otherwise noted. ENCODING Encoding. 1 31 OP Opcode. 6 25 S0 If 0 vsrc0 is VGPR address. If 1 vsrc0 is SGPR or built in inline constant. 1 55 S1 If 0 vsrc1 in VOP2/VOPC is VGPR address. If 1 vsrc1 is SGPR or built in inline constant. 1 63 SD If 1 SDST used. If 0 bits 8:15 used as SDWA destination control. 1 47 SDST Destination for compare result. 7 40 SRC0 First operand for instruction. 9 0 SRC0_ABS If true, take the floating-point absolute value of src0. 1 53 SRC0_NEG If true, take the floating-point negation of src0. 1 52 SRC0_SEL Source data select for src0. 3 48 SRC0_SEXT If true, sign-extend data for src0. If false, zero-extend. 1 51 SRC1_ABS If true, take the floating-point absolute value of src1. 1 61 SRC1_NEG If true, take the floating-point negation of src1. 1 60 SRC1_SEL Source data select for src1. 3 56 SRC1_SEXT If true, sign-extend data for src1. If false, zero-extend. 1 59 VDST Destination for instruction. 8 17 VSRC0 VGPR address for src0. 8 32 VSRC1 Second operand for instruction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default U 1 Vector ALU operations requiring three sources or modifiers; also some obscure opcodes. Almost all VOP1, VOP2, VOPC and VINTRP opcodes can be promoted to VOP3 to use input and output modifiers. CLAMP If set, clamp output to [0.0, 1.0]. Applied after output modifier. 1 15 ENCODING Encoding. 6 26 NEG If NEG[N] set, take the floating-point negation of the N'th input operand. This is applied after absolute value. 3 61 OMOD Output modifier for instruction. Applied before clamping. 2 59 OP Opcode. 10 16 SDST Destination for compare result. 7 8 SRC0 First operand for instruction. 9 32 SRC1 Second operand for instruction. 9 41 SRC2 Third operand for instruction. 9 50 VDST Destination for instruction. 8 0 VOP3P_MFMA 64 0000000000000000000000000000000011111111111111110000000000000000 0000000000000000000000000000000011010011101011010000000000000000 0000000000000000000000000000000011010011101011100000000000000000 0000000000000000000000000000000011010011101101010000000000000000 0000000000000000000000000000000011010011101101100000000000000000 0000000000000000000000000000000011010011101101110000000000000000 0000000000000000000000000000000011010011101110000000000000000000 0000000000000000000000000000000011010011101110010000000000000000 0000000000000000000000000000000011010011101110100000000000000000 0000000000000000000000000000000011010011101110110000000000000000 0000000000000000000000000000000011010011101111000000000000000000 0000000000000000000000000000000011010011101111010000000000000000 0000000000000000000000000000000011010011110000000000000000000000 0000000000000000000000000000000011010011110000010000000000000000 0000000000000000000000000000000011010011110000100000000000000000 0000000000000000000000000000000011010011110000110000000000000000 0000000000000000000000000000000011010011110001000000000000000000 0000000000000000000000000000000011010011110001010000000000000000 0000000000000000000000000000000011010011110001100000000000000000 0000000000000000000000000000000011010011110001110000000000000000 0000000000000000000000000000000011010011110010000000000000000000 0000000000000000000000000000000011010011110010010000000000000000 0000000000000000000000000000000011010011110010100000000000000000 0000000000000000000000000000000011010011110010110000000000000000 0000000000000000000000000000000011010011110011000000000000000000 0000000000000000000000000000000011010011110011010000000000000000 0000000000000000000000000000000011010011110011100000000000000000 0000000000000000000000000000000011010011110011110000000000000000 0000000000000000000000000000000011010011110100000000000000000000 0000000000000000000000000000000011010011110100010000000000000000 0000000000000000000000000000000011010011110100100000000000000000 0000000000000000000000000000000011010011110100110000000000000000 0000000000000000000000000000000011010011110101000000000000000000 0000000000000000000000000000000011010011110101010000000000000000 0000000000000000000000000000000011010011110101100000000000000000 0000000000000000000000000000000011010011110101110000000000000000 0000000000000000000000000000000011010011110110100000000000000000 0000000000000000000000000000000011010011110110110000000000000000 0000000000000000000000000000000011010011110111010000000000000000 0000000000000000000000000000000011010011110111100000000000000000 0000000000000000000000000000000011010011110111110000000000000000 0000000000000000000000000000000011010011111000000000000000000000 0000000000000000000000000000000011010011111000010000000000000000 0000000000000000000000000000000011010011111000100000000000000000 0000000000000000000000000000000011010011111001000000000000000000 0000000000000000000000000000000011010011111001100000000000000000 0000000000000000000000000000000011010011111010000000000000000000 0000000000000000000000000000000011010011111010100000000000000000 0000000000000000000000000000000011010011111011000000000000000000 0000000000000000000000000000000011010011111011100000000000000000 0000000000000000000000000000000011010011111011110000000000000000 0000000000000000000000000000000011010011111100000000000000000000 0000000000000000000000000000000011010011111100010000000000000000 0000000000000000000000000000000011010011111100100000000000000000 0000000000000000000000000000000011010011111100110000000000000000 0000000000000000000000000000000011010011111101000000000000000000 0000000000000000000000000000000011010011111101010000000000000000 0000000000000000000000000000000011010011111101100000000000000000 0000000000000000000000000000000011010011111101110000000000000000 0000000000000000000000000000000011010011111110000000000000000000 0000000000000000000000000000000011010011111110010000000000000000 0000000000000000000000000000000011010011111110100000000000000000 0000000000000000000000000000000011010011111110110000000000000000 0000000000000000000000000000000011010011111111000000000000000000 0000000000000000000000000000000011010011111111010000000000000000 0000000000000000000000000000000011010011111111100000000000000000 0000000000000000000000000000000011010011111111110000000000000000 default U 1 Vector ALU operations requiring two or three sources and performing two 16 bit ops in one instruction. ABID CBSZ and ABID are for A matrix swizzling, it is independent with BLGP, and with much more fine grain level control. 4 11 ACC ACC field, specifies whether srcA, srcB use an accumulator VGPR. 2 59 ACC_CD ACC field, specifies whether srcC and VDST use an accumulator VGPR (both operands share the same config bit). For instructions with OPF_ACC_CD_ONLY_D flag set, ACC_CD only controls the destination and the srcC must be an ArchVGPR. 1 15 BLGP BLGP control the Matrix B input swizzling, 16 lanes are one group and swizzling are between those 4 groups. 3 61 CBSZ CBSZ and ABID are for A matrix swizzling, it is independent with BLGP, and with much more fine grain level control. 3 8 ENCODING Encoding. 9 23 OP Opcode. 7 16 SRC0 First operand for instruction. 9 32 SRC1 Second operand for instruction. 9 41 SRC2 Third operand for instruction. 9 50 VDST Destination for instruction. 8 0 VOPC_INST_LITERAL 64 0000000000000000000000000000000011111111111111100000000000000000 0000000000000000000000000000000001111100001000000000000000000000 0000000000000000000000000000000001111100001000100000000000000000 0000000000000000000000000000000001111100001001000000000000000000 0000000000000000000000000000000001111100001001100000000000000000 0000000000000000000000000000000001111100001010000000000000000000 0000000000000000000000000000000001111100001010100000000000000000 0000000000000000000000000000000001111100010000000000000000000000 0000000000000000000000000000000001111100010000100000000000000000 0000000000000000000000000000000001111100010001000000000000000000 0000000000000000000000000000000001111100010001100000000000000000 0000000000000000000000000000000001111100010010000000000000000000 0000000000000000000000000000000001111100010010100000000000000000 0000000000000000000000000000000001111100010011000000000000000000 0000000000000000000000000000000001111100010011100000000000000000 0000000000000000000000000000000001111100010100000000000000000000 0000000000000000000000000000000001111100010100100000000000000000 0000000000000000000000000000000001111100010101000000000000000000 0000000000000000000000000000000001111100010101100000000000000000 0000000000000000000000000000000001111100010110000000000000000000 0000000000000000000000000000000001111100010110100000000000000000 0000000000000000000000000000000001111100010111000000000000000000 0000000000000000000000000000000001111100010111100000000000000000 0000000000000000000000000000000001111100011000000000000000000000 0000000000000000000000000000000001111100011000100000000000000000 0000000000000000000000000000000001111100011001000000000000000000 0000000000000000000000000000000001111100011001100000000000000000 0000000000000000000000000000000001111100011010000000000000000000 0000000000000000000000000000000001111100011010100000000000000000 0000000000000000000000000000000001111100011011000000000000000000 0000000000000000000000000000000001111100011011100000000000000000 0000000000000000000000000000000001111100011100000000000000000000 0000000000000000000000000000000001111100011100100000000000000000 0000000000000000000000000000000001111100011101000000000000000000 0000000000000000000000000000000001111100011101100000000000000000 0000000000000000000000000000000001111100011110000000000000000000 0000000000000000000000000000000001111100011110100000000000000000 0000000000000000000000000000000001111100011111000000000000000000 0000000000000000000000000000000001111100011111100000000000000000 0000000000000000000000000000000001111100100000000000000000000000 0000000000000000000000000000000001111100100000100000000000000000 0000000000000000000000000000000001111100100001000000000000000000 0000000000000000000000000000000001111100100001100000000000000000 0000000000000000000000000000000001111100100010000000000000000000 0000000000000000000000000000000001111100100010100000000000000000 0000000000000000000000000000000001111100100011000000000000000000 0000000000000000000000000000000001111100100011100000000000000000 0000000000000000000000000000000001111100100100000000000000000000 0000000000000000000000000000000001111100100100100000000000000000 0000000000000000000000000000000001111100100101000000000000000000 0000000000000000000000000000000001111100100101100000000000000000 0000000000000000000000000000000001111100100110000000000000000000 0000000000000000000000000000000001111100100110100000000000000000 0000000000000000000000000000000001111100100111000000000000000000 0000000000000000000000000000000001111100100111100000000000000000 0000000000000000000000000000000001111100101000000000000000000000 0000000000000000000000000000000001111100101000100000000000000000 0000000000000000000000000000000001111100101001000000000000000000 0000000000000000000000000000000001111100101001100000000000000000 0000000000000000000000000000000001111100101010000000000000000000 0000000000000000000000000000000001111100101010100000000000000000 0000000000000000000000000000000001111100101011000000000000000000 0000000000000000000000000000000001111100101011100000000000000000 0000000000000000000000000000000001111100101100000000000000000000 0000000000000000000000000000000001111100101100100000000000000000 0000000000000000000000000000000001111100101101000000000000000000 0000000000000000000000000000000001111100101101100000000000000000 0000000000000000000000000000000001111100101110000000000000000000 0000000000000000000000000000000001111100101110100000000000000000 0000000000000000000000000000000001111100101111000000000000000000 0000000000000000000000000000000001111100101111100000000000000000 0000000000000000000000000000000001111100110000000000000000000000 0000000000000000000000000000000001111100110000100000000000000000 0000000000000000000000000000000001111100110001000000000000000000 0000000000000000000000000000000001111100110001100000000000000000 0000000000000000000000000000000001111100110010000000000000000000 0000000000000000000000000000000001111100110010100000000000000000 0000000000000000000000000000000001111100110011000000000000000000 0000000000000000000000000000000001111100110011100000000000000000 0000000000000000000000000000000001111100110100000000000000000000 0000000000000000000000000000000001111100110100100000000000000000 0000000000000000000000000000000001111100110101000000000000000000 0000000000000000000000000000000001111100110101100000000000000000 0000000000000000000000000000000001111100110110000000000000000000 0000000000000000000000000000000001111100110110100000000000000000 0000000000000000000000000000000001111100110111000000000000000000 0000000000000000000000000000000001111100110111100000000000000000 0000000000000000000000000000000001111100111000000000000000000000 0000000000000000000000000000000001111100111000100000000000000000 0000000000000000000000000000000001111100111001000000000000000000 0000000000000000000000000000000001111100111001100000000000000000 0000000000000000000000000000000001111100111010000000000000000000 0000000000000000000000000000000001111100111010100000000000000000 0000000000000000000000000000000001111100111011000000000000000000 0000000000000000000000000000000001111100111011100000000000000000 0000000000000000000000000000000001111100111100000000000000000000 0000000000000000000000000000000001111100111100100000000000000000 0000000000000000000000000000000001111100111101000000000000000000 0000000000000000000000000000000001111100111101100000000000000000 0000000000000000000000000000000001111100111110000000000000000000 0000000000000000000000000000000001111100111110100000000000000000 0000000000000000000000000000000001111100111111000000000000000000 0000000000000000000000000000000001111100111111100000000000000000 0000000000000000000000000000000001111101010000000000000000000000 0000000000000000000000000000000001111101010000100000000000000000 0000000000000000000000000000000001111101010001000000000000000000 0000000000000000000000000000000001111101010001100000000000000000 0000000000000000000000000000000001111101010010000000000000000000 0000000000000000000000000000000001111101010010100000000000000000 0000000000000000000000000000000001111101010011000000000000000000 0000000000000000000000000000000001111101010011100000000000000000 0000000000000000000000000000000001111101010100000000000000000000 0000000000000000000000000000000001111101010100100000000000000000 0000000000000000000000000000000001111101010101000000000000000000 0000000000000000000000000000000001111101010101100000000000000000 0000000000000000000000000000000001111101010110000000000000000000 0000000000000000000000000000000001111101010110100000000000000000 0000000000000000000000000000000001111101010111000000000000000000 0000000000000000000000000000000001111101010111100000000000000000 0000000000000000000000000000000001111101011000000000000000000000 0000000000000000000000000000000001111101011000100000000000000000 0000000000000000000000000000000001111101011001000000000000000000 0000000000000000000000000000000001111101011001100000000000000000 0000000000000000000000000000000001111101011010000000000000000000 0000000000000000000000000000000001111101011010100000000000000000 0000000000000000000000000000000001111101011011000000000000000000 0000000000000000000000000000000001111101011011100000000000000000 0000000000000000000000000000000001111101011100000000000000000000 0000000000000000000000000000000001111101011100100000000000000000 0000000000000000000000000000000001111101011101000000000000000000 0000000000000000000000000000000001111101011101100000000000000000 0000000000000000000000000000000001111101011110000000000000000000 0000000000000000000000000000000001111101011110100000000000000000 0000000000000000000000000000000001111101011111000000000000000000 0000000000000000000000000000000001111101011111100000000000000000 0000000000000000000000000000000001111101100000000000000000000000 0000000000000000000000000000000001111101100000100000000000000000 0000000000000000000000000000000001111101100001000000000000000000 0000000000000000000000000000000001111101100001100000000000000000 0000000000000000000000000000000001111101100010000000000000000000 0000000000000000000000000000000001111101100010100000000000000000 0000000000000000000000000000000001111101100011000000000000000000 0000000000000000000000000000000001111101100011100000000000000000 0000000000000000000000000000000001111101100100000000000000000000 0000000000000000000000000000000001111101100100100000000000000000 0000000000000000000000000000000001111101100101000000000000000000 0000000000000000000000000000000001111101100101100000000000000000 0000000000000000000000000000000001111101100110000000000000000000 0000000000000000000000000000000001111101100110100000000000000000 0000000000000000000000000000000001111101100111000000000000000000 0000000000000000000000000000000001111101100111100000000000000000 0000000000000000000000000000000001111101101000000000000000000000 0000000000000000000000000000000001111101101000100000000000000000 0000000000000000000000000000000001111101101001000000000000000000 0000000000000000000000000000000001111101101001100000000000000000 0000000000000000000000000000000001111101101010000000000000000000 0000000000000000000000000000000001111101101010100000000000000000 0000000000000000000000000000000001111101101011000000000000000000 0000000000000000000000000000000001111101101011100000000000000000 0000000000000000000000000000000001111101101100000000000000000000 0000000000000000000000000000000001111101101100100000000000000000 0000000000000000000000000000000001111101101101000000000000000000 0000000000000000000000000000000001111101101101100000000000000000 0000000000000000000000000000000001111101101110000000000000000000 0000000000000000000000000000000001111101101110100000000000000000 0000000000000000000000000000000001111101101111000000000000000000 0000000000000000000000000000000001111101101111100000000000000000 0000000000000000000000000000000001111101110000000000000000000000 0000000000000000000000000000000001111101110000100000000000000000 0000000000000000000000000000000001111101110001000000000000000000 0000000000000000000000000000000001111101110001100000000000000000 0000000000000000000000000000000001111101110010000000000000000000 0000000000000000000000000000000001111101110010100000000000000000 0000000000000000000000000000000001111101110011000000000000000000 0000000000000000000000000000000001111101110011100000000000000000 0000000000000000000000000000000001111101110100000000000000000000 0000000000000000000000000000000001111101110100100000000000000000 0000000000000000000000000000000001111101110101000000000000000000 0000000000000000000000000000000001111101110101100000000000000000 0000000000000000000000000000000001111101110110000000000000000000 0000000000000000000000000000000001111101110110100000000000000000 0000000000000000000000000000000001111101110111000000000000000000 0000000000000000000000000000000001111101110111100000000000000000 0000000000000000000000000000000001111101111000000000000000000000 0000000000000000000000000000000001111101111000100000000000000000 0000000000000000000000000000000001111101111001000000000000000000 0000000000000000000000000000000001111101111001100000000000000000 0000000000000000000000000000000001111101111010000000000000000000 0000000000000000000000000000000001111101111010100000000000000000 0000000000000000000000000000000001111101111011000000000000000000 0000000000000000000000000000000001111101111011100000000000000000 0000000000000000000000000000000001111101111100000000000000000000 0000000000000000000000000000000001111101111100100000000000000000 0000000000000000000000000000000001111101111101000000000000000000 0000000000000000000000000000000001111101111101100000000000000000 0000000000000000000000000000000001111101111110000000000000000000 0000000000000000000000000000000001111101111110100000000000000000 0000000000000000000000000000000001111101111111000000000000000000 0000000000000000000000000000000001111101111111100000000000000000 has_lit == .fieldderef .fieldderef INST VOP2 SRC0 I 32 255 I 32 U 1 Vector ALU comparison operations with two sources. Instructions in this encoding may be promoted to VOP3 unless otherwise noted. Note that vector compare instructions respect the EXEC mask for computing the result in each thread; if EXEC[laneId] is zero, then the comparison for that thread is not performed and the result of VCC[laneId] is zero. ENCODING Encoding. 7 25 OP Opcode. 8 17 SIMM32 32-bit literal constant. 32 32 SRC0 First operand for instruction. 9 0 VSRC1 Second operand for instruction. 8 9 VOPC_VOP_SDWA_SDST_ENC 64 0000000000000000000000000000000011111111111111100000000000000000 0000000000000000000000000000000001111100001000000000000000000000 0000000000000000000000000000000001111100001000100000000000000000 0000000000000000000000000000000001111100001010000000000000000000 0000000000000000000000000000000001111100001010100000000000000000 0000000000000000000000000000000001111100010000000000000000000000 0000000000000000000000000000000001111100010000100000000000000000 0000000000000000000000000000000001111100010001000000000000000000 0000000000000000000000000000000001111100010001100000000000000000 0000000000000000000000000000000001111100010010000000000000000000 0000000000000000000000000000000001111100010010100000000000000000 0000000000000000000000000000000001111100010011000000000000000000 0000000000000000000000000000000001111100010011100000000000000000 0000000000000000000000000000000001111100010100000000000000000000 0000000000000000000000000000000001111100010100100000000000000000 0000000000000000000000000000000001111100010101000000000000000000 0000000000000000000000000000000001111100010101100000000000000000 0000000000000000000000000000000001111100010110000000000000000000 0000000000000000000000000000000001111100010110100000000000000000 0000000000000000000000000000000001111100010111000000000000000000 0000000000000000000000000000000001111100010111100000000000000000 0000000000000000000000000000000001111100011000000000000000000000 0000000000000000000000000000000001111100011000100000000000000000 0000000000000000000000000000000001111100011001000000000000000000 0000000000000000000000000000000001111100011001100000000000000000 0000000000000000000000000000000001111100011010000000000000000000 0000000000000000000000000000000001111100011010100000000000000000 0000000000000000000000000000000001111100011011000000000000000000 0000000000000000000000000000000001111100011011100000000000000000 0000000000000000000000000000000001111100011100000000000000000000 0000000000000000000000000000000001111100011100100000000000000000 0000000000000000000000000000000001111100011101000000000000000000 0000000000000000000000000000000001111100011101100000000000000000 0000000000000000000000000000000001111100011110000000000000000000 0000000000000000000000000000000001111100011110100000000000000000 0000000000000000000000000000000001111100011111000000000000000000 0000000000000000000000000000000001111100011111100000000000000000 0000000000000000000000000000000001111100100000000000000000000000 0000000000000000000000000000000001111100100000100000000000000000 0000000000000000000000000000000001111100100001000000000000000000 0000000000000000000000000000000001111100100001100000000000000000 0000000000000000000000000000000001111100100010000000000000000000 0000000000000000000000000000000001111100100010100000000000000000 0000000000000000000000000000000001111100100011000000000000000000 0000000000000000000000000000000001111100100011100000000000000000 0000000000000000000000000000000001111100100100000000000000000000 0000000000000000000000000000000001111100100100100000000000000000 0000000000000000000000000000000001111100100101000000000000000000 0000000000000000000000000000000001111100100101100000000000000000 0000000000000000000000000000000001111100100110000000000000000000 0000000000000000000000000000000001111100100110100000000000000000 0000000000000000000000000000000001111100100111000000000000000000 0000000000000000000000000000000001111100100111100000000000000000 0000000000000000000000000000000001111100101000000000000000000000 0000000000000000000000000000000001111100101000100000000000000000 0000000000000000000000000000000001111100101001000000000000000000 0000000000000000000000000000000001111100101001100000000000000000 0000000000000000000000000000000001111100101010000000000000000000 0000000000000000000000000000000001111100101010100000000000000000 0000000000000000000000000000000001111100101011000000000000000000 0000000000000000000000000000000001111100101011100000000000000000 0000000000000000000000000000000001111100101100000000000000000000 0000000000000000000000000000000001111100101100100000000000000000 0000000000000000000000000000000001111100101101000000000000000000 0000000000000000000000000000000001111100101101100000000000000000 0000000000000000000000000000000001111100101110000000000000000000 0000000000000000000000000000000001111100101110100000000000000000 0000000000000000000000000000000001111100101111000000000000000000 0000000000000000000000000000000001111100101111100000000000000000 0000000000000000000000000000000001111101010000000000000000000000 0000000000000000000000000000000001111101010000100000000000000000 0000000000000000000000000000000001111101010001000000000000000000 0000000000000000000000000000000001111101010001100000000000000000 0000000000000000000000000000000001111101010010000000000000000000 0000000000000000000000000000000001111101010010100000000000000000 0000000000000000000000000000000001111101010011000000000000000000 0000000000000000000000000000000001111101010011100000000000000000 0000000000000000000000000000000001111101010100000000000000000000 0000000000000000000000000000000001111101010100100000000000000000 0000000000000000000000000000000001111101010101000000000000000000 0000000000000000000000000000000001111101010101100000000000000000 0000000000000000000000000000000001111101010110000000000000000000 0000000000000000000000000000000001111101010110100000000000000000 0000000000000000000000000000000001111101010111000000000000000000 0000000000000000000000000000000001111101010111100000000000000000 0000000000000000000000000000000001111101011000000000000000000000 0000000000000000000000000000000001111101011000100000000000000000 0000000000000000000000000000000001111101011001000000000000000000 0000000000000000000000000000000001111101011001100000000000000000 0000000000000000000000000000000001111101011010000000000000000000 0000000000000000000000000000000001111101011010100000000000000000 0000000000000000000000000000000001111101011011000000000000000000 0000000000000000000000000000000001111101011011100000000000000000 0000000000000000000000000000000001111101011100000000000000000000 0000000000000000000000000000000001111101011100100000000000000000 0000000000000000000000000000000001111101011101000000000000000000 0000000000000000000000000000000001111101011101100000000000000000 0000000000000000000000000000000001111101011110000000000000000000 0000000000000000000000000000000001111101011110100000000000000000 0000000000000000000000000000000001111101011111000000000000000000 0000000000000000000000000000000001111101011111100000000000000000 0000000000000000000000000000000001111101100000000000000000000000 0000000000000000000000000000000001111101100000100000000000000000 0000000000000000000000000000000001111101100001000000000000000000 0000000000000000000000000000000001111101100001100000000000000000 0000000000000000000000000000000001111101100010000000000000000000 0000000000000000000000000000000001111101100010100000000000000000 0000000000000000000000000000000001111101100011000000000000000000 0000000000000000000000000000000001111101100011100000000000000000 0000000000000000000000000000000001111101100100000000000000000000 0000000000000000000000000000000001111101100100100000000000000000 0000000000000000000000000000000001111101100101000000000000000000 0000000000000000000000000000000001111101100101100000000000000000 0000000000000000000000000000000001111101100110000000000000000000 0000000000000000000000000000000001111101100110100000000000000000 0000000000000000000000000000000001111101100111000000000000000000 0000000000000000000000000000000001111101100111100000000000000000 0000000000000000000000000000000001111101101000000000000000000000 0000000000000000000000000000000001111101101000100000000000000000 0000000000000000000000000000000001111101101001000000000000000000 0000000000000000000000000000000001111101101001100000000000000000 0000000000000000000000000000000001111101101010000000000000000000 0000000000000000000000000000000001111101101010100000000000000000 0000000000000000000000000000000001111101101011000000000000000000 0000000000000000000000000000000001111101101011100000000000000000 0000000000000000000000000000000001111101101100000000000000000000 0000000000000000000000000000000001111101101100100000000000000000 0000000000000000000000000000000001111101101101000000000000000000 0000000000000000000000000000000001111101101101100000000000000000 0000000000000000000000000000000001111101101110000000000000000000 0000000000000000000000000000000001111101101110100000000000000000 0000000000000000000000000000000001111101101111000000000000000000 0000000000000000000000000000000001111101101111100000000000000000 has_sdwa == .fieldderef .fieldderef INST VOP2 SRC0 I 32 249 I 32 U 1 Vector ALU comparison operations with two sources. Instructions in this encoding may be promoted to VOP3 unless otherwise noted. Note that vector compare instructions respect the EXEC mask for computing the result in each thread; if EXEC[laneId] is zero, then the comparison for that thread is not performed and the result of VCC[laneId] is zero. ENCODING Encoding. 7 25 OP Opcode. 8 17 S0 If 0 vsrc0 is VGPR address. If 1 vsrc0 is SGPR or built in inline constant. 1 55 S1 If 0 vsrc1 in VOP2/VOPC is VGPR address. If 1 vsrc1 is SGPR or built in inline constant. 1 63 SD If 1 SDST used. If 0 bits 8:15 used as SDWA destination control. 1 47 SDST Destination for compare result. 7 40 SRC0 First operand for instruction. 9 0 SRC0_ABS If true, take the floating-point absolute value of src0. 1 53 SRC0_NEG If true, take the floating-point negation of src0. 1 52 SRC0_SEL Source data select for src0. 3 48 SRC0_SEXT If true, sign-extend data for src0. If false, zero-extend. 1 51 SRC1_ABS If true, take the floating-point absolute value of src1. 1 61 SRC1_NEG If true, take the floating-point negation of src1. 1 60 SRC1_SEL Source data select for src1. 3 56 SRC1_SEXT If true, sign-extend data for src1. If false, zero-extend. 1 59 VSRC0 VGPR address for src0. 8 32 VSRC1 Second operand for instruction. 8 9 FALSE FALSE FALSE FALSE FALSE DS_ADD_U32 Add two unsigned 32-bit integer values stored in the data register and a location in a data share. ENC_DS default 0 ADDR FMT_NUM_B32 OPR_VGPR 32 DATA0 FMT_NUM_U32 OPR_VGPR_OR_ACCVGPR 32 FMT_NUM_U32 OPR_DSMEM 32 FMT_NUM_U32 OPR_DSMEM 32 VMEM ATOMIC DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_SUB_U32 Subtract an unsigned 32-bit integer value stored in the data register from a value stored in a location in a data share. ENC_DS default 1 ADDR FMT_NUM_B32 OPR_VGPR 32 DATA0 FMT_NUM_U32 OPR_VGPR_OR_ACCVGPR 32 FMT_NUM_U32 OPR_DSMEM 32 FMT_NUM_U32 OPR_DSMEM 32 VMEM ATOMIC DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_RSUB_U32 Subtract an unsigned 32-bit integer value stored in a location in a data share from a value stored in the data register. ENC_DS default 2 ADDR FMT_NUM_B32 OPR_VGPR 32 DATA0 FMT_NUM_U32 OPR_VGPR_OR_ACCVGPR 32 FMT_NUM_U32 OPR_DSMEM 32 FMT_NUM_U32 OPR_DSMEM 32 VMEM ATOMIC DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_INC_U32 Increment an unsigned 32-bit integer value from a location in a data share with wraparound to 0 if the value exceeds a value in the data register. ENC_DS default 3 ADDR FMT_NUM_B32 OPR_VGPR 32 DATA0 FMT_NUM_U32 OPR_VGPR_OR_ACCVGPR 32 FMT_NUM_U32 OPR_DSMEM 32 FMT_NUM_U32 OPR_DSMEM 32 VMEM ATOMIC DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_DEC_U32 Decrement an unsigned 32-bit integer value from a location in a data share with wraparound to a value in the data register if the decrement yields a negative value. ENC_DS default 4 ADDR FMT_NUM_B32 OPR_VGPR 32 DATA0 FMT_NUM_U32 OPR_VGPR_OR_ACCVGPR 32 FMT_NUM_U32 OPR_DSMEM 32 FMT_NUM_U32 OPR_DSMEM 32 VMEM ATOMIC DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_MIN_I32 Select the minimum of two signed 32-bit integer inputs, given two values stored in the data register and a location in a data share. Update the data share with the selected value. ENC_DS default 5 ADDR FMT_NUM_B32 OPR_VGPR 32 DATA0 FMT_NUM_I32 OPR_VGPR_OR_ACCVGPR 32 FMT_NUM_I32 OPR_DSMEM 32 FMT_NUM_I32 OPR_DSMEM 32 VMEM ATOMIC DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_MAX_I32 Select the maximum of two signed 32-bit integer inputs, given two values stored in the data register and a location in a data share. Update the data share with the selected value. ENC_DS default 6 ADDR FMT_NUM_B32 OPR_VGPR 32 DATA0 FMT_NUM_I32 OPR_VGPR_OR_ACCVGPR 32 FMT_NUM_I32 OPR_DSMEM 32 FMT_NUM_I32 OPR_DSMEM 32 VMEM ATOMIC DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_MIN_U32 Select the minimum of two unsigned 32-bit integer inputs, given two values stored in the data register and a location in a data share. Update the data share with the selected value. ENC_DS default 7 ADDR FMT_NUM_B32 OPR_VGPR 32 DATA0 FMT_NUM_U32 OPR_VGPR_OR_ACCVGPR 32 FMT_NUM_U32 OPR_DSMEM 32 FMT_NUM_U32 OPR_DSMEM 32 VMEM ATOMIC DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_MAX_U32 Select the maximum of two unsigned 32-bit integer inputs, given two values stored in the data register and a location in a data share. Update the data share with the selected value. ENC_DS default 8 ADDR FMT_NUM_B32 OPR_VGPR 32 DATA0 FMT_NUM_U32 OPR_VGPR_OR_ACCVGPR 32 FMT_NUM_U32 OPR_DSMEM 32 FMT_NUM_U32 OPR_DSMEM 32 VMEM ATOMIC DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_AND_B32 Calculate bitwise AND given two unsigned 32-bit integer values stored in the data register and a location in a data share. ENC_DS default 9 ADDR FMT_NUM_B32 OPR_VGPR 32 DATA0 FMT_NUM_B32 OPR_VGPR_OR_ACCVGPR 32 FMT_NUM_B32 OPR_DSMEM 32 FMT_NUM_B32 OPR_DSMEM 32 VMEM ATOMIC DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_OR_B32 Calculate bitwise OR given two unsigned 32-bit integer values stored in the data register and a location in a data share. ENC_DS default 10 ADDR FMT_NUM_B32 OPR_VGPR 32 DATA0 FMT_NUM_B32 OPR_VGPR_OR_ACCVGPR 32 FMT_NUM_B32 OPR_DSMEM 32 FMT_NUM_B32 OPR_DSMEM 32 VMEM ATOMIC DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_XOR_B32 Calculate bitwise XOR given two unsigned 32-bit integer values stored in the data register and a location in a data share. ENC_DS default 11 ADDR FMT_NUM_B32 OPR_VGPR 32 DATA0 FMT_NUM_B32 OPR_VGPR_OR_ACCVGPR 32 FMT_NUM_B32 OPR_DSMEM 32 FMT_NUM_B32 OPR_DSMEM 32 VMEM ATOMIC DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_MSKOR_B32 Calculate masked bitwise OR on an unsigned 32-bit integer location in a data share, given mask value and bits to OR in the data registers. ENC_DS default 12 ADDR FMT_NUM_B32 OPR_VGPR 32 DATA0 FMT_NUM_B32 OPR_VGPR_OR_ACCVGPR 32 DATA1 FMT_NUM_B32 OPR_VGPR_OR_ACCVGPR 32 FMT_NUM_B32 OPR_DSMEM 32 FMT_NUM_B32 OPR_DSMEM 32 VMEM ATOMIC DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_WRITE_B32 Store 32 bits of data from a vector input register into a data share. ENC_DS default 13 ADDR FMT_NUM_B32 OPR_VGPR 32 DATA0 FMT_NUM_B32 OPR_VGPR_OR_ACCVGPR 32 FMT_NUM_B32 OPR_DSMEM 32 VMEM DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_WRITE2_B32 Store 32 bits of data from one vector input register and then 32 bits of data from a second vector input register into a data share. ENC_DS default 14 ADDR FMT_NUM_B32 OPR_VGPR 32 DATA0 FMT_NUM_B32 OPR_VGPR_OR_ACCVGPR 32 DATA1 FMT_NUM_B32 OPR_VGPR_OR_ACCVGPR 32 FMT_NUM_B32 OPR_DSMEM 32 VMEM DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_WRITE2ST64_B32 Store 32 bits of data from one vector input register and then 32 bits of data from a second vector input register into a data share. Treat each offset as an index and multiply by a stride of 64 elements (256 bytes) to generate an offset for each DS address. ENC_DS default 15 ADDR FMT_NUM_B32 OPR_VGPR 32 DATA0 FMT_NUM_B32 OPR_VGPR_OR_ACCVGPR 32 DATA1 FMT_NUM_B32 OPR_VGPR_OR_ACCVGPR 32 FMT_NUM_B32 OPR_DSMEM 32 VMEM DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_CMPST_B32 Compare an unsigned 32-bit integer value in the data comparison register with a location in a data share, and modify the memory location with a value in the data source register if the comparison is equal. ENC_DS default 16 ADDR FMT_NUM_B32 OPR_VGPR 32 DATA0 FMT_NUM_B32 OPR_VGPR_OR_ACCVGPR 32 DATA1 FMT_NUM_B32 OPR_VGPR_OR_ACCVGPR 32 FMT_NUM_B32 OPR_DSMEM 32 FMT_NUM_B32 OPR_DSMEM 32 VMEM ATOMIC DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_CMPST_F32 Compare a single-precision float value in the data comparison register with a location in a data share, and modify the memory location with a value in the data source register if the comparison is equal. ENC_DS default 17 ADDR FMT_NUM_B32 OPR_VGPR 32 DATA0 FMT_NUM_F32 OPR_VGPR_OR_ACCVGPR 32 DATA1 FMT_NUM_F32 OPR_VGPR_OR_ACCVGPR 32 FMT_NUM_F32 OPR_DSMEM 32 FMT_NUM_F32 OPR_DSMEM 32 VMEM ATOMIC DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_MIN_F32 Select the minimum of two single-precision float inputs, given two values stored in the data register and a location in a data share. Update the data share with the selected value. ENC_DS default 18 ADDR FMT_NUM_B32 OPR_VGPR 32 DATA0 FMT_NUM_F32 OPR_VGPR_OR_ACCVGPR 32 FMT_NUM_F32 OPR_DSMEM 32 FMT_NUM_F32 OPR_DSMEM 32 VMEM ATOMIC DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_MAX_F32 Select the maximum of two single-precision float inputs, given two values stored in the data register and a location in a data share. Update the data share with the selected value. ENC_DS default 19 ADDR FMT_NUM_B32 OPR_VGPR 32 DATA0 FMT_NUM_F32 OPR_VGPR_OR_ACCVGPR 32 FMT_NUM_F32 OPR_DSMEM 32 FMT_NUM_F32 OPR_DSMEM 32 VMEM ATOMIC DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_NOP Do nothing. ENC_DS default 20 VMEM DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_ADD_F32 Add two single-precision float values stored in the data register and a location in a data share. ENC_DS default 21 ADDR FMT_NUM_B32 OPR_VGPR 32 DATA0 FMT_NUM_F32 OPR_VGPR_OR_ACCVGPR 32 FMT_NUM_F32 OPR_DSMEM 32 FMT_NUM_F32 OPR_DSMEM 32 VMEM ATOMIC DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_PK_ADD_F16 Add a packed 2-component half-precision float value in the data register to a location in a data share. ENC_DS default 23 ADDR FMT_NUM_B32 OPR_VGPR 32 DATA0 FMT_NUM_PK2_F16 OPR_VGPR_OR_ACCVGPR 32 FMT_NUM_PK2_F16 OPR_DSMEM 32 FMT_NUM_PK2_F16 OPR_DSMEM 32 VMEM ATOMIC DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_PK_ADD_BF16 Add a packed 2-component BF16 float value in the data register to a location in a data share. ENC_DS default 24 ADDR FMT_NUM_B32 OPR_VGPR 32 DATA0 FMT_NUM_PK2_BF16 OPR_VGPR_OR_ACCVGPR 32 FMT_NUM_PK2_BF16 OPR_DSMEM 32 FMT_NUM_PK2_BF16 OPR_DSMEM 32 VMEM ATOMIC DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_WRITE_ADDTID_B32 Store 32 bits of data from a vector input register into a data share. The memory base address is provided as an immediate value and the lane ID is used as an offset. ENC_DS default 29 DATA0 FMT_NUM_B32 OPR_VGPR_OR_ACCVGPR 32 FMT_NUM_B32 OPR_DSMEM 32 FMT_NUM_B32 OPR_SDST_M0 32 VMEM DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_WRITE_B8 Store 8 bits of data from a vector register into a data share. ENC_DS default 30 ADDR FMT_NUM_B32 OPR_VGPR 32 DATA0 FMT_NUM_B8 OPR_VGPR_OR_ACCVGPR 8 FMT_NUM_B8 OPR_DSMEM 8 VMEM DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_WRITE_B16 Store 16 bits of data from a vector register into a data share. ENC_DS default 31 ADDR FMT_NUM_B32 OPR_VGPR 32 DATA0 FMT_NUM_B16 OPR_VGPR_OR_ACCVGPR 16 FMT_NUM_B16 OPR_DSMEM 16 VMEM DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_ADD_RTN_U32 Add two unsigned 32-bit integer values stored in the data register and a location in a data share. Store the original value from data share into a vector register. ENC_DS default 32 VDST FMT_NUM_U32 OPR_VGPR_OR_ACCVGPR 32 ADDR FMT_NUM_B32 OPR_VGPR 32 DATA0 FMT_NUM_U32 OPR_VGPR_OR_ACCVGPR 32 FMT_NUM_U32 OPR_DSMEM 32 FMT_NUM_U32 OPR_DSMEM 32 VMEM ATOMIC DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_SUB_RTN_U32 Subtract an unsigned 32-bit integer value stored in the data register from a value stored in a location in a data share. Store the original value from data share into a vector register. ENC_DS default 33 VDST FMT_NUM_U32 OPR_VGPR_OR_ACCVGPR 32 ADDR FMT_NUM_B32 OPR_VGPR 32 DATA0 FMT_NUM_U32 OPR_VGPR_OR_ACCVGPR 32 FMT_NUM_U32 OPR_DSMEM 32 FMT_NUM_U32 OPR_DSMEM 32 VMEM ATOMIC DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_RSUB_RTN_U32 Subtract an unsigned 32-bit integer value stored in a location in a data share from a value stored in the data register. Store the original value from data share into a vector register. ENC_DS default 34 VDST FMT_NUM_U32 OPR_VGPR_OR_ACCVGPR 32 ADDR FMT_NUM_B32 OPR_VGPR 32 DATA0 FMT_NUM_U32 OPR_VGPR_OR_ACCVGPR 32 FMT_NUM_U32 OPR_DSMEM 32 FMT_NUM_U32 OPR_DSMEM 32 VMEM ATOMIC DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_INC_RTN_U32 Increment an unsigned 32-bit integer value from a location in a data share with wraparound to 0 if the value exceeds a value in the data register. Store the original value from data share into a vector register. ENC_DS default 35 VDST FMT_NUM_U32 OPR_VGPR_OR_ACCVGPR 32 ADDR FMT_NUM_B32 OPR_VGPR 32 DATA0 FMT_NUM_U32 OPR_VGPR_OR_ACCVGPR 32 FMT_NUM_U32 OPR_DSMEM 32 FMT_NUM_U32 OPR_DSMEM 32 VMEM ATOMIC DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_DEC_RTN_U32 Decrement an unsigned 32-bit integer value from a location in a data share with wraparound to a value in the data register if the decrement yields a negative value. Store the original value from data share into a vector register. ENC_DS default 36 VDST FMT_NUM_U32 OPR_VGPR_OR_ACCVGPR 32 ADDR FMT_NUM_B32 OPR_VGPR 32 DATA0 FMT_NUM_U32 OPR_VGPR_OR_ACCVGPR 32 FMT_NUM_U32 OPR_DSMEM 32 FMT_NUM_U32 OPR_DSMEM 32 VMEM ATOMIC DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_MIN_RTN_I32 Select the minimum of two signed 32-bit integer inputs, given two values stored in the data register and a location in a data share. Update the data share with the selected value. Store the original value from data share into a vector register. ENC_DS default 37 VDST FMT_NUM_I32 OPR_VGPR_OR_ACCVGPR 32 ADDR FMT_NUM_B32 OPR_VGPR 32 DATA0 FMT_NUM_I32 OPR_VGPR_OR_ACCVGPR 32 FMT_NUM_I32 OPR_DSMEM 32 FMT_NUM_I32 OPR_DSMEM 32 VMEM ATOMIC DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_MAX_RTN_I32 Select the maximum of two signed 32-bit integer inputs, given two values stored in the data register and a location in a data share. Update the data share with the selected value. Store the original value from data share into a vector register. ENC_DS default 38 VDST FMT_NUM_I32 OPR_VGPR_OR_ACCVGPR 32 ADDR FMT_NUM_B32 OPR_VGPR 32 DATA0 FMT_NUM_I32 OPR_VGPR_OR_ACCVGPR 32 FMT_NUM_I32 OPR_DSMEM 32 FMT_NUM_I32 OPR_DSMEM 32 VMEM ATOMIC DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_MIN_RTN_U32 Select the minimum of two unsigned 32-bit integer inputs, given two values stored in the data register and a location in a data share. Update the data share with the selected value. Store the original value from data share into a vector register. ENC_DS default 39 VDST FMT_NUM_U32 OPR_VGPR_OR_ACCVGPR 32 ADDR FMT_NUM_B32 OPR_VGPR 32 DATA0 FMT_NUM_U32 OPR_VGPR_OR_ACCVGPR 32 FMT_NUM_U32 OPR_DSMEM 32 FMT_NUM_U32 OPR_DSMEM 32 VMEM ATOMIC DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_MAX_RTN_U32 Select the maximum of two unsigned 32-bit integer inputs, given two values stored in the data register and a location in a data share. Update the data share with the selected value. Store the original value from data share into a vector register. ENC_DS default 40 VDST FMT_NUM_U32 OPR_VGPR_OR_ACCVGPR 32 ADDR FMT_NUM_B32 OPR_VGPR 32 DATA0 FMT_NUM_U32 OPR_VGPR_OR_ACCVGPR 32 FMT_NUM_U32 OPR_DSMEM 32 FMT_NUM_U32 OPR_DSMEM 32 VMEM ATOMIC DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_AND_RTN_B32 Calculate bitwise AND given two unsigned 32-bit integer values stored in the data register and a location in a data share. Store the original value from data share into a vector register. ENC_DS default 41 VDST FMT_NUM_B32 OPR_VGPR_OR_ACCVGPR 32 ADDR FMT_NUM_B32 OPR_VGPR 32 DATA0 FMT_NUM_B32 OPR_VGPR_OR_ACCVGPR 32 FMT_NUM_B32 OPR_DSMEM 32 FMT_NUM_B32 OPR_DSMEM 32 VMEM ATOMIC DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_OR_RTN_B32 Calculate bitwise OR given two unsigned 32-bit integer values stored in the data register and a location in a data share. Store the original value from data share into a vector register. ENC_DS default 42 VDST FMT_NUM_B32 OPR_VGPR_OR_ACCVGPR 32 ADDR FMT_NUM_B32 OPR_VGPR 32 DATA0 FMT_NUM_B32 OPR_VGPR_OR_ACCVGPR 32 FMT_NUM_B32 OPR_DSMEM 32 FMT_NUM_B32 OPR_DSMEM 32 VMEM ATOMIC DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_XOR_RTN_B32 Calculate bitwise XOR given two unsigned 32-bit integer values stored in the data register and a location in a data share. Store the original value from data share into a vector register. ENC_DS default 43 VDST FMT_NUM_B32 OPR_VGPR_OR_ACCVGPR 32 ADDR FMT_NUM_B32 OPR_VGPR 32 DATA0 FMT_NUM_B32 OPR_VGPR_OR_ACCVGPR 32 FMT_NUM_B32 OPR_DSMEM 32 FMT_NUM_B32 OPR_DSMEM 32 VMEM ATOMIC DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_MSKOR_RTN_B32 Calculate masked bitwise OR on an unsigned 32-bit integer location in a data share, given mask value and bits to OR in the data registers. ENC_DS default 44 VDST FMT_NUM_B32 OPR_VGPR_OR_ACCVGPR 32 ADDR FMT_NUM_B32 OPR_VGPR 32 DATA0 FMT_NUM_B32 OPR_VGPR_OR_ACCVGPR 32 DATA1 FMT_NUM_B32 OPR_VGPR_OR_ACCVGPR 32 FMT_NUM_B32 OPR_DSMEM 32 FMT_NUM_B32 OPR_DSMEM 32 VMEM ATOMIC DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_WRXCHG_RTN_B32 Swap an unsigned 32-bit integer value in the data register with a location in a data share. ENC_DS default 45 VDST FMT_NUM_B32 OPR_VGPR_OR_ACCVGPR 32 ADDR FMT_NUM_B32 OPR_VGPR 32 DATA0 FMT_NUM_B32 OPR_VGPR_OR_ACCVGPR 32 FMT_NUM_B32 OPR_DSMEM 32 FMT_NUM_B32 OPR_DSMEM 32 VMEM ATOMIC DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_WRXCHG2_RTN_B32 Swap two unsigned 32-bit integer values in the data registers with two locations in a data share. ENC_DS default 46 VDST FMT_NUM_PK2_B32 OPR_VGPR_OR_ACCVGPR 64 ADDR FMT_NUM_B32 OPR_VGPR 32 DATA0 FMT_NUM_B32 OPR_VGPR_OR_ACCVGPR 32 DATA1 FMT_NUM_B32 OPR_VGPR_OR_ACCVGPR 32 FMT_NUM_B32 OPR_DSMEM 32 FMT_NUM_B32 OPR_DSMEM 32 VMEM ATOMIC DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_WRXCHG2ST64_RTN_B32 Swap two unsigned 32-bit integer values in the data registers with two locations in a data share. Treat each offset as an index and multiply by a stride of 64 elements (256 bytes) to generate an offset for each DS address. ENC_DS default 47 VDST FMT_NUM_PK2_B32 OPR_VGPR_OR_ACCVGPR 64 ADDR FMT_NUM_B32 OPR_VGPR 32 DATA0 FMT_NUM_B32 OPR_VGPR_OR_ACCVGPR 32 DATA1 FMT_NUM_B32 OPR_VGPR_OR_ACCVGPR 32 FMT_NUM_B32 OPR_DSMEM 32 FMT_NUM_B32 OPR_DSMEM 32 VMEM ATOMIC DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_CMPST_RTN_B32 Compare an unsigned 32-bit integer value in the data comparison register with a location in a data share, and modify the memory location with a value in the data source register if the comparison is equal. ENC_DS default 48 VDST FMT_NUM_B32 OPR_VGPR_OR_ACCVGPR 32 ADDR FMT_NUM_B32 OPR_VGPR 32 DATA0 FMT_NUM_B32 OPR_VGPR_OR_ACCVGPR 32 DATA1 FMT_NUM_B32 OPR_VGPR_OR_ACCVGPR 32 FMT_NUM_B32 OPR_DSMEM 32 FMT_NUM_B32 OPR_DSMEM 32 VMEM ATOMIC DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_CMPST_RTN_F32 Compare a single-precision float value in the data comparison register with a location in a data share, and modify the memory location with a value in the data source register if the comparison is equal. ENC_DS default 49 VDST FMT_NUM_F32 OPR_VGPR_OR_ACCVGPR 32 ADDR FMT_NUM_B32 OPR_VGPR 32 DATA0 FMT_NUM_F32 OPR_VGPR_OR_ACCVGPR 32 DATA1 FMT_NUM_F32 OPR_VGPR_OR_ACCVGPR 32 FMT_NUM_F32 OPR_DSMEM 32 FMT_NUM_F32 OPR_DSMEM 32 VMEM ATOMIC DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_MIN_RTN_F32 Select the minimum of two single-precision float inputs, given two values stored in the data register and a location in a data share. Update the data share with the selected value. Store the original value from data share into a vector register. ENC_DS default 50 VDST FMT_NUM_F32 OPR_VGPR_OR_ACCVGPR 32 ADDR FMT_NUM_B32 OPR_VGPR 32 DATA0 FMT_NUM_F32 OPR_VGPR_OR_ACCVGPR 32 FMT_NUM_F32 OPR_DSMEM 32 FMT_NUM_F32 OPR_DSMEM 32 VMEM ATOMIC DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_MAX_RTN_F32 Select the maximum of two single-precision float inputs, given two values stored in the data register and a location in a data share. Update the data share with the selected value. Store the original value from data share into a vector register. ENC_DS default 51 VDST FMT_NUM_F32 OPR_VGPR_OR_ACCVGPR 32 ADDR FMT_NUM_B32 OPR_VGPR 32 DATA0 FMT_NUM_F32 OPR_VGPR_OR_ACCVGPR 32 FMT_NUM_F32 OPR_DSMEM 32 FMT_NUM_F32 OPR_DSMEM 32 VMEM ATOMIC DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_WRAP_RTN_B32 Given a minuend from a location in data share and a subtrahend from a vector register, subtract the two values iff the result is nonnegative; otherwise add a value from a second vector register to the memory location. ENC_DS default 52 VDST FMT_NUM_B32 OPR_VGPR_OR_ACCVGPR 32 ADDR FMT_NUM_B32 OPR_VGPR 32 DATA0 FMT_NUM_B32 OPR_VGPR_OR_ACCVGPR 32 DATA1 FMT_NUM_B32 OPR_VGPR_OR_ACCVGPR 32 FMT_NUM_B32 OPR_DSMEM 32 FMT_NUM_B32 OPR_DSMEM 32 VMEM ATOMIC DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_ADD_RTN_F32 Add two single-precision float values stored in the data register and a location in a data share. Store the original value from data share into a vector register. ENC_DS default 53 VDST FMT_NUM_F32 OPR_VGPR_OR_ACCVGPR 32 ADDR FMT_NUM_B32 OPR_VGPR 32 DATA0 FMT_NUM_F32 OPR_VGPR_OR_ACCVGPR 32 FMT_NUM_F32 OPR_DSMEM 32 FMT_NUM_F32 OPR_DSMEM 32 VMEM ATOMIC DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_READ_B32 Load 32 bits of data from a data share into a vector register. ENC_DS default 54 VDST FMT_NUM_B32 OPR_VGPR_OR_ACCVGPR 32 ADDR FMT_NUM_B32 OPR_VGPR 32 FMT_NUM_B32 OPR_DSMEM 32 VMEM DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_READ2_B32 Load 32 bits of data from one location in a data share and then 32 bits of data from a second location in a data share and store the results into a 64-bit vector register. ENC_DS default 55 VDST FMT_NUM_PK2_B32 OPR_VGPR_OR_ACCVGPR 64 ADDR FMT_NUM_B32 OPR_VGPR 32 FMT_NUM_B32 OPR_DSMEM 32 VMEM DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_READ2ST64_B32 Load 32 bits of data from one location in a data share and then 32 bits of data from a second location in a data share and store the results into a 64-bit vector register. Treat each offset as an index and multiply by a stride of 64 elements (256 bytes) to generate an offset for each DS address. ENC_DS default 56 VDST FMT_NUM_PK2_B32 OPR_VGPR_OR_ACCVGPR 64 ADDR FMT_NUM_B32 OPR_VGPR 32 FMT_NUM_B32 OPR_DSMEM 32 VMEM DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_READ_I8 Load 8 bits of signed data from a data share, sign extend to 32 bits and store the result into a vector register. ENC_DS default 57 VDST FMT_NUM_I32 OPR_VGPR_OR_ACCVGPR 32 ADDR FMT_NUM_B32 OPR_VGPR 32 FMT_NUM_B8 OPR_DSMEM 8 VMEM DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_READ_U8 Load 8 bits of unsigned data from a data share, zero extend to 32 bits and store the result into a vector register. ENC_DS default 58 VDST FMT_NUM_U32 OPR_VGPR_OR_ACCVGPR 32 ADDR FMT_NUM_B32 OPR_VGPR 32 FMT_NUM_B8 OPR_DSMEM 8 VMEM DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_READ_I16 Load 16 bits of signed data from a data share, sign extend to 32 bits and store the result into a vector register. ENC_DS default 59 VDST FMT_NUM_I32 OPR_VGPR_OR_ACCVGPR 32 ADDR FMT_NUM_B32 OPR_VGPR 32 FMT_NUM_B16 OPR_DSMEM 16 VMEM DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_READ_U16 Load 16 bits of unsigned data from a data share, zero extend to 32 bits and store the result into a vector register. ENC_DS default 60 VDST FMT_NUM_U32 OPR_VGPR_OR_ACCVGPR 32 ADDR FMT_NUM_B32 OPR_VGPR 32 FMT_NUM_B16 OPR_DSMEM 16 VMEM DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_SWIZZLE_B32 Dword swizzle, no data is written to LDS memory. ENC_DS default 61 VDST FMT_NUM_B32 OPR_VGPR_OR_ACCVGPR 32 ADDR FMT_NUM_B32 OPR_VGPR 32 VMEM ATOMIC DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_PERMUTE_B32 Forward permute. This does not access LDS memory and may be called even if no LDS memory is allocated to the wave. It uses LDS to implement an arbitrary swizzle across threads in a wavefront. ENC_DS default 62 VDST FMT_NUM_B32 OPR_VGPR_OR_ACCVGPR 32 ADDR FMT_NUM_B32 OPR_VGPR 32 DATA0 FMT_NUM_B32 OPR_VGPR_OR_ACCVGPR 32 VMEM ATOMIC DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_BPERMUTE_B32 Backward permute. This does not access LDS memory and may be called even if no LDS memory is allocated to the wave. It uses LDS hardware to implement an arbitrary swizzle across threads in a wavefront. ENC_DS default 63 VDST FMT_NUM_B32 OPR_VGPR_OR_ACCVGPR 32 ADDR FMT_NUM_B32 OPR_VGPR 32 DATA0 FMT_NUM_B32 OPR_VGPR_OR_ACCVGPR 32 VMEM ATOMIC DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_ADD_U64 Add two unsigned 64-bit integer values stored in the data register and a location in a data share. ENC_DS default 64 ADDR FMT_NUM_B32 OPR_VGPR 32 DATA0 FMT_NUM_U64 OPR_VGPR_OR_ACCVGPR 64 FMT_NUM_U64 OPR_DSMEM 64 FMT_NUM_U64 OPR_DSMEM 64 VMEM ATOMIC DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_SUB_U64 Subtract an unsigned 64-bit integer value stored in the data register from a value stored in a location in a data share. ENC_DS default 65 ADDR FMT_NUM_B32 OPR_VGPR 32 DATA0 FMT_NUM_U64 OPR_VGPR_OR_ACCVGPR 64 FMT_NUM_U64 OPR_DSMEM 64 FMT_NUM_U64 OPR_DSMEM 64 VMEM ATOMIC DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_RSUB_U64 Subtract an unsigned 64-bit integer value stored in a location in a data share from a value stored in the data register. ENC_DS default 66 ADDR FMT_NUM_B32 OPR_VGPR 32 DATA0 FMT_NUM_U64 OPR_VGPR_OR_ACCVGPR 64 FMT_NUM_U64 OPR_DSMEM 64 FMT_NUM_U64 OPR_DSMEM 64 VMEM ATOMIC DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_INC_U64 Increment an unsigned 64-bit integer value from a location in a data share with wraparound to 0 if the value exceeds a value in the data register. ENC_DS default 67 ADDR FMT_NUM_B32 OPR_VGPR 32 DATA0 FMT_NUM_U64 OPR_VGPR_OR_ACCVGPR 64 FMT_NUM_U64 OPR_DSMEM 64 FMT_NUM_U64 OPR_DSMEM 64 VMEM ATOMIC DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_DEC_U64 Decrement an unsigned 64-bit integer value from a location in a data share with wraparound to a value in the data register if the decrement yields a negative value. ENC_DS default 68 ADDR FMT_NUM_B32 OPR_VGPR 32 DATA0 FMT_NUM_U64 OPR_VGPR_OR_ACCVGPR 64 FMT_NUM_U64 OPR_DSMEM 64 FMT_NUM_U64 OPR_DSMEM 64 VMEM ATOMIC DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_MIN_I64 Select the minimum of two signed 64-bit integer inputs, given two values stored in the data register and a location in a data share. Update the data share with the selected value. ENC_DS default 69 ADDR FMT_NUM_B32 OPR_VGPR 32 DATA0 FMT_NUM_I64 OPR_VGPR_OR_ACCVGPR 64 FMT_NUM_I64 OPR_DSMEM 64 FMT_NUM_I64 OPR_DSMEM 64 VMEM ATOMIC DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_MAX_I64 Select the maximum of two signed 64-bit integer inputs, given two values stored in the data register and a location in a data share. Update the data share with the selected value. ENC_DS default 70 ADDR FMT_NUM_B32 OPR_VGPR 32 DATA0 FMT_NUM_I64 OPR_VGPR_OR_ACCVGPR 64 FMT_NUM_I64 OPR_DSMEM 64 FMT_NUM_I64 OPR_DSMEM 64 VMEM ATOMIC DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_MIN_U64 Select the minimum of two unsigned 64-bit integer inputs, given two values stored in the data register and a location in a data share. Update the data share with the selected value. ENC_DS default 71 ADDR FMT_NUM_B32 OPR_VGPR 32 DATA0 FMT_NUM_U64 OPR_VGPR_OR_ACCVGPR 64 FMT_NUM_U64 OPR_DSMEM 64 FMT_NUM_U64 OPR_DSMEM 64 VMEM ATOMIC DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_MAX_U64 Select the maximum of two unsigned 64-bit integer inputs, given two values stored in the data register and a location in a data share. Update the data share with the selected value. ENC_DS default 72 ADDR FMT_NUM_B32 OPR_VGPR 32 DATA0 FMT_NUM_U64 OPR_VGPR_OR_ACCVGPR 64 FMT_NUM_U64 OPR_DSMEM 64 FMT_NUM_U64 OPR_DSMEM 64 VMEM ATOMIC DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_AND_B64 Calculate bitwise AND given two unsigned 64-bit integer values stored in the data register and a location in a data share. ENC_DS default 73 ADDR FMT_NUM_B32 OPR_VGPR 32 DATA0 FMT_NUM_B64 OPR_VGPR_OR_ACCVGPR 64 FMT_NUM_B64 OPR_DSMEM 64 FMT_NUM_B64 OPR_DSMEM 64 VMEM ATOMIC DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_OR_B64 Calculate bitwise OR given two unsigned 64-bit integer values stored in the data register and a location in a data share. ENC_DS default 74 ADDR FMT_NUM_B32 OPR_VGPR 32 DATA0 FMT_NUM_B64 OPR_VGPR_OR_ACCVGPR 64 FMT_NUM_B64 OPR_DSMEM 64 FMT_NUM_B64 OPR_DSMEM 64 VMEM ATOMIC DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_XOR_B64 Calculate bitwise XOR given two unsigned 64-bit integer values stored in the data register and a location in a data share. ENC_DS default 75 ADDR FMT_NUM_B32 OPR_VGPR 32 DATA0 FMT_NUM_B64 OPR_VGPR_OR_ACCVGPR 64 FMT_NUM_B64 OPR_DSMEM 64 FMT_NUM_B64 OPR_DSMEM 64 VMEM ATOMIC DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_MSKOR_B64 Calculate masked bitwise OR on an unsigned 64-bit integer location in a data share, given mask value and bits to OR in the data registers. ENC_DS default 76 ADDR FMT_NUM_B32 OPR_VGPR 32 DATA0 FMT_NUM_B64 OPR_VGPR_OR_ACCVGPR 64 DATA1 FMT_NUM_B64 OPR_VGPR_OR_ACCVGPR 64 FMT_NUM_B64 OPR_DSMEM 64 FMT_NUM_B64 OPR_DSMEM 64 VMEM ATOMIC DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_WRITE_B64 Store 64 bits of data from a vector input register into a data share. ENC_DS default 77 ADDR FMT_NUM_B32 OPR_VGPR 32 DATA0 FMT_NUM_B64 OPR_VGPR_OR_ACCVGPR 64 FMT_NUM_B64 OPR_DSMEM 64 VMEM DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_WRITE2_B64 Store 64 bits of data from one vector input register and then 64 bits of data from a second vector input register into a data share. ENC_DS default 78 ADDR FMT_NUM_B32 OPR_VGPR 32 DATA0 FMT_NUM_B64 OPR_VGPR_OR_ACCVGPR 64 DATA1 FMT_NUM_B64 OPR_VGPR_OR_ACCVGPR 64 FMT_NUM_B64 OPR_DSMEM 64 VMEM DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_WRITE2ST64_B64 Store 64 bits of data from one vector input register and then 64 bits of data from a second vector input register into a data share. Treat each offset as an index and multiply by a stride of 64 elements (256 bytes) to generate an offset for each DS address. ENC_DS default 79 ADDR FMT_NUM_B32 OPR_VGPR 32 DATA0 FMT_NUM_B64 OPR_VGPR_OR_ACCVGPR 64 DATA1 FMT_NUM_B64 OPR_VGPR_OR_ACCVGPR 64 FMT_NUM_B64 OPR_DSMEM 64 VMEM DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_CMPST_B64 Compare an unsigned 64-bit integer value in the data comparison register with a location in a data share, and modify the memory location with a value in the data source register if the comparison is equal. ENC_DS default 80 ADDR FMT_NUM_B32 OPR_VGPR 32 DATA0 FMT_NUM_B64 OPR_VGPR_OR_ACCVGPR 64 DATA1 FMT_NUM_B64 OPR_VGPR_OR_ACCVGPR 64 FMT_NUM_B64 OPR_DSMEM 64 FMT_NUM_B64 OPR_DSMEM 64 VMEM ATOMIC DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_CMPST_F64 Compare a double-precision float value in the data comparison register with a location in a data share, and modify the memory location with a value in the data source register if the comparison is equal. ENC_DS default 81 ADDR FMT_NUM_B32 OPR_VGPR 32 DATA0 FMT_NUM_F64 OPR_VGPR_OR_ACCVGPR 64 DATA1 FMT_NUM_F64 OPR_VGPR_OR_ACCVGPR 64 FMT_NUM_F64 OPR_DSMEM 64 FMT_NUM_F64 OPR_DSMEM 64 VMEM ATOMIC DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_MIN_F64 Select the minimum of two double-precision float inputs, given two values stored in the data register and a location in a data share. Update the data share with the selected value. ENC_DS default 82 ADDR FMT_NUM_B32 OPR_VGPR 32 DATA0 FMT_NUM_F64 OPR_VGPR_OR_ACCVGPR 64 FMT_NUM_F64 OPR_DSMEM 64 FMT_NUM_F64 OPR_DSMEM 64 VMEM ATOMIC DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_MAX_F64 Select the maximum of two double-precision float inputs, given two values stored in the data register and a location in a data share. Update the data share with the selected value. ENC_DS default 83 ADDR FMT_NUM_B32 OPR_VGPR 32 DATA0 FMT_NUM_F64 OPR_VGPR_OR_ACCVGPR 64 FMT_NUM_F64 OPR_DSMEM 64 FMT_NUM_F64 OPR_DSMEM 64 VMEM ATOMIC DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_WRITE_B8_D16_HI Store 8 bits of data from the high bits of a vector register into a data share. ENC_DS default 84 ADDR FMT_NUM_B32 OPR_VGPR 32 DATA0 FMT_NUM_PK4_B8 OPR_VGPR_OR_ACCVGPR 32 FMT_NUM_B8 OPR_DSMEM 8 VMEM DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_WRITE_B16_D16_HI Store 16 bits of data from the high bits of a vector register into a data share. ENC_DS default 85 ADDR FMT_NUM_B32 OPR_VGPR 32 DATA0 FMT_NUM_PK2_B16 OPR_VGPR_OR_ACCVGPR 32 FMT_NUM_B16 OPR_DSMEM 16 VMEM DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_READ_U8_D16 Load 8 bits of unsigned data from a data share, zero extend to 16 bits and store the result into the low 16 bits of a vector register. ENC_DS default 86 VDST FMT_NUM_PK2_U16 OPR_VGPR_OR_ACCVGPR 32 ADDR FMT_NUM_B32 OPR_VGPR 32 FMT_NUM_B8 OPR_DSMEM 8 VMEM DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_READ_U8_D16_HI Load 8 bits of unsigned data from a data share, zero extend to 16 bits and store the result into the high 16 bits of a vector register. ENC_DS default 87 VDST FMT_NUM_PK2_U16 OPR_VGPR_OR_ACCVGPR 32 ADDR FMT_NUM_B32 OPR_VGPR 32 FMT_NUM_B8 OPR_DSMEM 8 VMEM DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_READ_I8_D16 Load 8 bits of signed data from a data share, sign extend to 16 bits and store the result into the low 16 bits of a vector register. ENC_DS default 88 VDST FMT_NUM_PK2_I16 OPR_VGPR_OR_ACCVGPR 32 ADDR FMT_NUM_B32 OPR_VGPR 32 FMT_NUM_B8 OPR_DSMEM 8 VMEM DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_READ_I8_D16_HI Load 8 bits of signed data from a data share, sign extend to 16 bits and store the result into the high 16 bits of a vector register. ENC_DS default 89 VDST FMT_NUM_PK2_I16 OPR_VGPR_OR_ACCVGPR 32 ADDR FMT_NUM_B32 OPR_VGPR 32 FMT_NUM_B8 OPR_DSMEM 8 VMEM DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_READ_U16_D16 Load 16 bits of unsigned data from a data share and store the result into the low 16 bits of a vector register. ENC_DS default 90 VDST FMT_NUM_PK2_U16 OPR_VGPR_OR_ACCVGPR 32 ADDR FMT_NUM_B32 OPR_VGPR 32 FMT_NUM_B16 OPR_DSMEM 16 VMEM DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_READ_U16_D16_HI Load 16 bits of unsigned data from a data share and store the result into the high 16 bits of a vector register. ENC_DS default 91 VDST FMT_NUM_PK2_U16 OPR_VGPR_OR_ACCVGPR 32 ADDR FMT_NUM_B32 OPR_VGPR 32 FMT_NUM_B16 OPR_DSMEM 16 VMEM DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_ADD_F64 Add a double-precision float value in the data register to a location in a data share. ENC_DS default 92 ADDR FMT_NUM_B32 OPR_VGPR 32 DATA0 FMT_NUM_F64 OPR_VGPR_OR_ACCVGPR 64 FMT_NUM_F64 OPR_DSMEM 64 FMT_NUM_F64 OPR_DSMEM 64 VMEM ATOMIC DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_ADD_RTN_U64 Add two unsigned 64-bit integer values stored in the data register and a location in a data share. Store the original value from data share into a vector register. ENC_DS default 96 VDST FMT_NUM_U64 OPR_VGPR_OR_ACCVGPR 64 ADDR FMT_NUM_B32 OPR_VGPR 32 DATA0 FMT_NUM_U64 OPR_VGPR_OR_ACCVGPR 64 FMT_NUM_U64 OPR_DSMEM 64 FMT_NUM_U64 OPR_DSMEM 64 VMEM ATOMIC DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_SUB_RTN_U64 Subtract an unsigned 64-bit integer value stored in the data register from a value stored in a location in a data share. Store the original value from data share into a vector register. ENC_DS default 97 VDST FMT_NUM_U64 OPR_VGPR_OR_ACCVGPR 64 ADDR FMT_NUM_B32 OPR_VGPR 32 DATA0 FMT_NUM_U64 OPR_VGPR_OR_ACCVGPR 64 FMT_NUM_U64 OPR_DSMEM 64 FMT_NUM_U64 OPR_DSMEM 64 VMEM ATOMIC DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_RSUB_RTN_U64 Subtract an unsigned 64-bit integer value stored in a location in a data share from a value stored in the data register. Store the original value from data share into a vector register. ENC_DS default 98 VDST FMT_NUM_U64 OPR_VGPR_OR_ACCVGPR 64 ADDR FMT_NUM_B32 OPR_VGPR 32 DATA0 FMT_NUM_U64 OPR_VGPR_OR_ACCVGPR 64 FMT_NUM_U64 OPR_DSMEM 64 FMT_NUM_U64 OPR_DSMEM 64 VMEM ATOMIC DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_INC_RTN_U64 Increment an unsigned 64-bit integer value from a location in a data share with wraparound to 0 if the value exceeds a value in the data register. Store the original value from data share into a vector register. ENC_DS default 99 VDST FMT_NUM_U64 OPR_VGPR_OR_ACCVGPR 64 ADDR FMT_NUM_B32 OPR_VGPR 32 DATA0 FMT_NUM_U64 OPR_VGPR_OR_ACCVGPR 64 FMT_NUM_U64 OPR_DSMEM 64 FMT_NUM_U64 OPR_DSMEM 64 VMEM ATOMIC DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_DEC_RTN_U64 Decrement an unsigned 64-bit integer value from a location in a data share with wraparound to a value in the data register if the decrement yields a negative value. Store the original value from data share into a vector register. ENC_DS default 100 VDST FMT_NUM_U64 OPR_VGPR_OR_ACCVGPR 64 ADDR FMT_NUM_B32 OPR_VGPR 32 DATA0 FMT_NUM_U64 OPR_VGPR_OR_ACCVGPR 64 FMT_NUM_U64 OPR_DSMEM 64 FMT_NUM_U64 OPR_DSMEM 64 VMEM ATOMIC DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_MIN_RTN_I64 Select the minimum of two signed 64-bit integer inputs, given two values stored in the data register and a location in a data share. Update the data share with the selected value. Store the original value from data share into a vector register. ENC_DS default 101 VDST FMT_NUM_I64 OPR_VGPR_OR_ACCVGPR 64 ADDR FMT_NUM_B32 OPR_VGPR 32 DATA0 FMT_NUM_I64 OPR_VGPR_OR_ACCVGPR 64 FMT_NUM_I64 OPR_DSMEM 64 FMT_NUM_I64 OPR_DSMEM 64 VMEM ATOMIC DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_MAX_RTN_I64 Select the maximum of two signed 64-bit integer inputs, given two values stored in the data register and a location in a data share. Update the data share with the selected value. Store the original value from data share into a vector register. ENC_DS default 102 VDST FMT_NUM_I64 OPR_VGPR_OR_ACCVGPR 64 ADDR FMT_NUM_B32 OPR_VGPR 32 DATA0 FMT_NUM_I64 OPR_VGPR_OR_ACCVGPR 64 FMT_NUM_I64 OPR_DSMEM 64 FMT_NUM_I64 OPR_DSMEM 64 VMEM ATOMIC DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_MIN_RTN_U64 Select the minimum of two unsigned 64-bit integer inputs, given two values stored in the data register and a location in a data share. Update the data share with the selected value. Store the original value from data share into a vector register. ENC_DS default 103 VDST FMT_NUM_U64 OPR_VGPR_OR_ACCVGPR 64 ADDR FMT_NUM_B32 OPR_VGPR 32 DATA0 FMT_NUM_U64 OPR_VGPR_OR_ACCVGPR 64 FMT_NUM_U64 OPR_DSMEM 64 FMT_NUM_U64 OPR_DSMEM 64 VMEM ATOMIC DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_MAX_RTN_U64 Select the maximum of two unsigned 64-bit integer inputs, given two values stored in the data register and a location in a data share. Update the data share with the selected value. Store the original value from data share into a vector register. ENC_DS default 104 VDST FMT_NUM_U64 OPR_VGPR_OR_ACCVGPR 64 ADDR FMT_NUM_B32 OPR_VGPR 32 DATA0 FMT_NUM_U64 OPR_VGPR_OR_ACCVGPR 64 FMT_NUM_U64 OPR_DSMEM 64 FMT_NUM_U64 OPR_DSMEM 64 VMEM ATOMIC DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_AND_RTN_B64 Calculate bitwise AND given two unsigned 64-bit integer values stored in the data register and a location in a data share. Store the original value from data share into a vector register. ENC_DS default 105 VDST FMT_NUM_B64 OPR_VGPR_OR_ACCVGPR 64 ADDR FMT_NUM_B32 OPR_VGPR 32 DATA0 FMT_NUM_B64 OPR_VGPR_OR_ACCVGPR 64 FMT_NUM_B64 OPR_DSMEM 64 FMT_NUM_B64 OPR_DSMEM 64 VMEM ATOMIC DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_OR_RTN_B64 Calculate bitwise OR given two unsigned 64-bit integer values stored in the data register and a location in a data share. Store the original value from data share into a vector register. ENC_DS default 106 VDST FMT_NUM_B64 OPR_VGPR_OR_ACCVGPR 64 ADDR FMT_NUM_B32 OPR_VGPR 32 DATA0 FMT_NUM_B64 OPR_VGPR_OR_ACCVGPR 64 FMT_NUM_B64 OPR_DSMEM 64 FMT_NUM_B64 OPR_DSMEM 64 VMEM ATOMIC DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_XOR_RTN_B64 Calculate bitwise XOR given two unsigned 64-bit integer values stored in the data register and a location in a data share. Store the original value from data share into a vector register. ENC_DS default 107 VDST FMT_NUM_B64 OPR_VGPR_OR_ACCVGPR 64 ADDR FMT_NUM_B32 OPR_VGPR 32 DATA0 FMT_NUM_B64 OPR_VGPR_OR_ACCVGPR 64 FMT_NUM_B64 OPR_DSMEM 64 FMT_NUM_B64 OPR_DSMEM 64 VMEM ATOMIC DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_MSKOR_RTN_B64 Calculate masked bitwise OR on an unsigned 64-bit integer location in a data share, given mask value and bits to OR in the data registers. ENC_DS default 108 VDST FMT_NUM_B64 OPR_VGPR_OR_ACCVGPR 64 ADDR FMT_NUM_B32 OPR_VGPR 32 DATA0 FMT_NUM_B64 OPR_VGPR_OR_ACCVGPR 64 DATA1 FMT_NUM_B64 OPR_VGPR_OR_ACCVGPR 64 FMT_NUM_B64 OPR_DSMEM 64 FMT_NUM_B64 OPR_DSMEM 64 VMEM ATOMIC DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_WRXCHG_RTN_B64 Swap an unsigned 64-bit integer value in the data register with a location in a data share. ENC_DS default 109 VDST FMT_NUM_B64 OPR_VGPR_OR_ACCVGPR 64 ADDR FMT_NUM_B32 OPR_VGPR 32 DATA0 FMT_NUM_B64 OPR_VGPR_OR_ACCVGPR 64 FMT_NUM_B64 OPR_DSMEM 64 FMT_NUM_B64 OPR_DSMEM 64 VMEM ATOMIC DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_WRXCHG2_RTN_B64 Swap two unsigned 64-bit integer values in the data registers with two locations in a data share. ENC_DS default 110 VDST FMT_NUM_PK2_B64 OPR_VGPR_OR_ACCVGPR 128 ADDR FMT_NUM_B32 OPR_VGPR 32 DATA0 FMT_NUM_B64 OPR_VGPR_OR_ACCVGPR 64 DATA1 FMT_NUM_B64 OPR_VGPR_OR_ACCVGPR 64 FMT_NUM_B64 OPR_DSMEM 64 FMT_NUM_B64 OPR_DSMEM 64 VMEM ATOMIC DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_WRXCHG2ST64_RTN_B64 Swap two unsigned 64-bit integer values in the data registers with two locations in a data share. Treat each offset as an index and multiply by a stride of 64 elements (256 bytes) to generate an offset for each DS address. ENC_DS default 111 VDST FMT_NUM_PK2_B64 OPR_VGPR_OR_ACCVGPR 128 ADDR FMT_NUM_B32 OPR_VGPR 32 DATA0 FMT_NUM_B64 OPR_VGPR_OR_ACCVGPR 64 DATA1 FMT_NUM_B64 OPR_VGPR_OR_ACCVGPR 64 FMT_NUM_B64 OPR_DSMEM 64 FMT_NUM_B64 OPR_DSMEM 64 VMEM ATOMIC DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_CMPST_RTN_B64 Compare an unsigned 64-bit integer value in the data comparison register with a location in a data share, and modify the memory location with a value in the data source register if the comparison is equal. ENC_DS default 112 VDST FMT_NUM_B64 OPR_VGPR_OR_ACCVGPR 64 ADDR FMT_NUM_B32 OPR_VGPR 32 DATA0 FMT_NUM_B64 OPR_VGPR_OR_ACCVGPR 64 DATA1 FMT_NUM_B64 OPR_VGPR_OR_ACCVGPR 64 FMT_NUM_B64 OPR_DSMEM 64 FMT_NUM_B64 OPR_DSMEM 64 VMEM ATOMIC DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_CMPST_RTN_F64 Compare a double-precision float value in the data comparison register with a location in a data share, and modify the memory location with a value in the data source register if the comparison is equal. ENC_DS default 113 VDST FMT_NUM_F64 OPR_VGPR_OR_ACCVGPR 64 ADDR FMT_NUM_B32 OPR_VGPR 32 DATA0 FMT_NUM_F64 OPR_VGPR_OR_ACCVGPR 64 DATA1 FMT_NUM_F64 OPR_VGPR_OR_ACCVGPR 64 FMT_NUM_F64 OPR_DSMEM 64 FMT_NUM_F64 OPR_DSMEM 64 VMEM ATOMIC DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_MIN_RTN_F64 Select the minimum of two double-precision float inputs, given two values stored in the data register and a location in a data share. Update the data share with the selected value. Store the original value from data share into a vector register. ENC_DS default 114 VDST FMT_NUM_F64 OPR_VGPR_OR_ACCVGPR 64 ADDR FMT_NUM_B32 OPR_VGPR 32 DATA0 FMT_NUM_F64 OPR_VGPR_OR_ACCVGPR 64 FMT_NUM_F64 OPR_DSMEM 64 FMT_NUM_F64 OPR_DSMEM 64 VMEM ATOMIC DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_MAX_RTN_F64 Select the maximum of two double-precision float inputs, given two values stored in the data register and a location in a data share. Update the data share with the selected value. Store the original value from data share into a vector register. ENC_DS default 115 VDST FMT_NUM_F64 OPR_VGPR_OR_ACCVGPR 64 ADDR FMT_NUM_B32 OPR_VGPR 32 DATA0 FMT_NUM_F64 OPR_VGPR_OR_ACCVGPR 64 FMT_NUM_F64 OPR_DSMEM 64 FMT_NUM_F64 OPR_DSMEM 64 VMEM ATOMIC DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_READ_B64 Load 64 bits of data from a data share into a vector register. ENC_DS default 118 VDST FMT_NUM_B64 OPR_VGPR_OR_ACCVGPR 64 ADDR FMT_NUM_B32 OPR_VGPR 32 FMT_NUM_B64 OPR_DSMEM 64 VMEM DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_READ2_B64 Load 64 bits of data from one location in a data share and then 64 bits of data from a second location in a data share and store the results into a 128-bit vector register. ENC_DS default 119 VDST FMT_NUM_PK2_B64 OPR_VGPR_OR_ACCVGPR 128 ADDR FMT_NUM_B32 OPR_VGPR 32 FMT_NUM_B64 OPR_DSMEM 64 VMEM DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_READ2ST64_B64 Load 64 bits of data from one location in a data share and then 64 bits of data from a second location in a data share and store the results into a 128-bit vector register. Treat each offset as an index and multiply by a stride of 64 elements (256 bytes) to generate an offset for each DS address. ENC_DS default 120 VDST FMT_NUM_PK2_B64 OPR_VGPR_OR_ACCVGPR 128 ADDR FMT_NUM_B32 OPR_VGPR 32 FMT_NUM_B64 OPR_DSMEM 64 VMEM DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_ADD_RTN_F64 Add a double-precision float value in the data register to a location in a data share. Store the original value from data share into a vector register. ENC_DS default 124 VDST FMT_NUM_F64 OPR_VGPR_OR_ACCVGPR 64 ADDR FMT_NUM_B32 OPR_VGPR 32 DATA0 FMT_NUM_F64 OPR_VGPR_OR_ACCVGPR 64 FMT_NUM_F64 OPR_DSMEM 64 FMT_NUM_F64 OPR_DSMEM 64 VMEM ATOMIC DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_CONDXCHG32_RTN_B64 Perform 2 conditional write exchanges, where each conditional write exchange writes a 32 bit value from a data register to a location in data share iff the most significant bit of the data value is set. ENC_DS default 126 VDST FMT_NUM_B64 OPR_VGPR_OR_ACCVGPR 64 ADDR FMT_NUM_B32 OPR_VGPR 32 DATA0 FMT_NUM_B64 OPR_VGPR_OR_ACCVGPR 64 FMT_NUM_B64 OPR_DSMEM 64 FMT_NUM_B64 OPR_DSMEM 64 VMEM ATOMIC DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_READ_ADDTID_B32 Load 32 bits of data from a data share into a vector register. The memory base address is provided as an immediate value and the lane ID is used as an offset. ENC_DS default 182 VDST FMT_NUM_B32 OPR_VGPR_OR_ACCVGPR 32 FMT_NUM_B32 OPR_DSMEM 32 FMT_NUM_B32 OPR_SDST_M0 32 VMEM DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_PK_ADD_RTN_F16 Add a packed 2-component half-precision float value in the data register to a location in a data share. Store the original value from data share into a vector register. ENC_DS default 183 VDST FMT_NUM_PK2_F16 OPR_VGPR_OR_ACCVGPR 32 ADDR FMT_NUM_B32 OPR_VGPR 32 DATA0 FMT_NUM_PK2_F16 OPR_VGPR_OR_ACCVGPR 32 FMT_NUM_PK2_F16 OPR_DSMEM 32 FMT_NUM_PK2_F16 OPR_DSMEM 32 VMEM ATOMIC DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_PK_ADD_RTN_BF16 Add a packed 2-component BF16 float value in the data register to a location in a data share. Store the original value from data share into a vector register. ENC_DS default 184 VDST FMT_NUM_PK2_BF16 OPR_VGPR_OR_ACCVGPR 32 ADDR FMT_NUM_B32 OPR_VGPR 32 DATA0 FMT_NUM_PK2_BF16 OPR_VGPR_OR_ACCVGPR 32 FMT_NUM_PK2_BF16 OPR_DSMEM 32 FMT_NUM_PK2_BF16 OPR_DSMEM 32 VMEM ATOMIC DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_CONSUME Subtract (count_bits(exec_mask)) from the value stored in DS memory at (M0.base + instr_offset) if GDS, or at instr_offset if LDS. Return the pre-operation value to VGPRs. ENC_DS default 189 VDST FMT_NUM_B32 OPR_VGPR_OR_ACCVGPR 32 FMT_NUM_B32 OPR_DSMEM 32 FMT_NUM_B32 OPR_DSMEM 32 VMEM ATOMIC DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_APPEND Add (count_bits(exec_mask)) to the value stored in DS memory at (M0.base + instr_offset) if GDS, or at instr_offset if LDS. Return the pre-operation value to VGPRs. ENC_DS default 190 VDST FMT_NUM_B32 OPR_VGPR_OR_ACCVGPR 32 FMT_NUM_B32 OPR_DSMEM 32 FMT_NUM_B32 OPR_DSMEM 32 VMEM ATOMIC DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_WRITE_B96 Store 96 bits of data from a vector input register into a data share. ENC_DS default 222 ADDR FMT_NUM_B32 OPR_VGPR 32 DATA0 FMT_NUM_B96 OPR_VGPR_OR_ACCVGPR 96 FMT_NUM_B96 OPR_DSMEM 96 VMEM DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_WRITE_B128 Store 128 bits of data from a vector input register into a data share. ENC_DS default 223 ADDR FMT_NUM_B32 OPR_VGPR 32 DATA0 FMT_NUM_B128 OPR_VGPR_OR_ACCVGPR 128 FMT_NUM_B128 OPR_DSMEM 128 VMEM DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_READ_B64_TR_B4 Read 64 bits of data per lane from data share. Interpret the data as a matrix with 4 bit elements and transpose the matrix. Store the result into vector registers. ENC_DS default 224 VDST FMT_NUM_B64 OPR_VGPR_OR_ACCVGPR 64 ADDR FMT_NUM_B32 OPR_VGPR 32 FMT_NUM_B64 OPR_DSMEM 64 VMEM DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_READ_B96_TR_B6 DS_READ_B128_TR_B6 Read 96 bits of data per lane from data share. Interpret the data as a matrix with 6 bit elements and transpose the matrix. Store the result into vector registers. ENC_DS default 225 VDST FMT_NUM_B96 OPR_VGPR_OR_ACCVGPR 96 ADDR FMT_NUM_B32 OPR_VGPR 32 FMT_NUM_B96 OPR_DSMEM 96 VMEM DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_READ_B64_TR_B8 Read 64 bits of data per lane from data share. Interpret the data as a matrix with 8 bit elements and transpose the matrix. Store the result into vector registers. ENC_DS default 226 VDST FMT_NUM_B64 OPR_VGPR_OR_ACCVGPR 64 ADDR FMT_NUM_B32 OPR_VGPR 32 FMT_NUM_B64 OPR_DSMEM 64 VMEM DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_READ_B64_TR_B16 Read 64 bits of data per lane from data share. Interpret the data as a matrix with 16 bit elements and transpose the matrix. Store the result into vector registers. ENC_DS default 227 VDST FMT_NUM_B64 OPR_VGPR_OR_ACCVGPR 64 ADDR FMT_NUM_B32 OPR_VGPR 32 FMT_NUM_B64 OPR_DSMEM 64 VMEM DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_READ_B96 Load 96 bits of data from a data share into a vector register. ENC_DS default 254 VDST FMT_NUM_B96 OPR_VGPR_OR_ACCVGPR 96 ADDR FMT_NUM_B32 OPR_VGPR 32 FMT_NUM_B96 OPR_DSMEM 96 VMEM DATA_SHARE FALSE FALSE FALSE FALSE FALSE DS_READ_B128 Load 128 bits of data from a data share into a vector register. ENC_DS default 255 VDST FMT_NUM_B128 OPR_VGPR_OR_ACCVGPR 128 ADDR FMT_NUM_B32 OPR_VGPR 32 FMT_NUM_B128 OPR_DSMEM 128 VMEM DATA_SHARE FALSE FALSE FALSE FALSE FALSE FLAT_LOAD_UBYTE Load 8 bits of unsigned data from the flat aperture, zero extend to 32 bits and store the result into a vector register. ENC_FLAT default 16 VDST FMT_ANY OPR_VGPR_OR_ACCVGPR 32 ADDR FMT_ANY OPR_VGPR 64 FMT_NUM_B64 OPR_FLAT_SCRATCH 64 FMT_NUM_U8 OPR_GPUMEM 8 FMT_NUM_B32 OPR_SDST_M0 32 VMEM FLAT FALSE FALSE FALSE FALSE FALSE FLAT_LOAD_SBYTE Load 8 bits of signed data from the flat aperture, sign extend to 32 bits and store the result into a vector register. ENC_FLAT default 17 VDST FMT_ANY OPR_VGPR_OR_ACCVGPR 32 ADDR FMT_ANY OPR_VGPR 64 FMT_NUM_B64 OPR_FLAT_SCRATCH 64 FMT_NUM_I8 OPR_GPUMEM 8 FMT_NUM_B32 OPR_SDST_M0 32 VMEM FLAT FALSE FALSE FALSE FALSE FALSE FLAT_LOAD_USHORT Load 16 bits of unsigned data from the flat aperture, zero extend to 32 bits and store the result into a vector register. ENC_FLAT default 18 VDST FMT_ANY OPR_VGPR_OR_ACCVGPR 32 ADDR FMT_ANY OPR_VGPR 64 FMT_NUM_B64 OPR_FLAT_SCRATCH 64 FMT_NUM_U16 OPR_GPUMEM 16 FMT_NUM_B32 OPR_SDST_M0 32 VMEM FLAT FALSE FALSE FALSE FALSE FALSE FLAT_LOAD_SSHORT Load 16 bits of signed data from the flat aperture, sign extend to 32 bits and store the result into a vector register. ENC_FLAT default 19 VDST FMT_ANY OPR_VGPR_OR_ACCVGPR 32 ADDR FMT_ANY OPR_VGPR 64 FMT_NUM_B64 OPR_FLAT_SCRATCH 64 FMT_NUM_I16 OPR_GPUMEM 16 FMT_NUM_B32 OPR_SDST_M0 32 VMEM FLAT FALSE FALSE FALSE FALSE FALSE FLAT_LOAD_DWORD Load 32 bits of data from the flat aperture into a vector register. ENC_FLAT default 20 VDST FMT_ANY OPR_VGPR_OR_ACCVGPR 32 ADDR FMT_ANY OPR_VGPR 64 FMT_NUM_B64 OPR_FLAT_SCRATCH 64 FMT_NUM_B32 OPR_GPUMEM 32 FMT_NUM_B32 OPR_SDST_M0 32 VMEM FLAT FALSE FALSE FALSE FALSE FALSE FLAT_LOAD_DWORDX2 Load 64 bits of data from the flat aperture into a vector register. ENC_FLAT default 21 VDST FMT_ANY OPR_VGPR_OR_ACCVGPR 64 ADDR FMT_ANY OPR_VGPR 64 FMT_NUM_B64 OPR_FLAT_SCRATCH 64 FMT_NUM_B64 OPR_GPUMEM 64 FMT_NUM_B32 OPR_SDST_M0 32 VMEM FLAT FALSE FALSE FALSE FALSE FALSE FLAT_LOAD_DWORDX3 Load 96 bits of data from the flat aperture into a vector register. ENC_FLAT default 22 VDST FMT_ANY OPR_VGPR_OR_ACCVGPR 96 ADDR FMT_ANY OPR_VGPR 64 FMT_NUM_B64 OPR_FLAT_SCRATCH 64 FMT_NUM_B96 OPR_GPUMEM 96 FMT_NUM_B32 OPR_SDST_M0 32 VMEM FLAT FALSE FALSE FALSE FALSE FALSE FLAT_LOAD_DWORDX4 Load 128 bits of data from the flat aperture into a vector register. ENC_FLAT default 23 VDST FMT_ANY OPR_VGPR_OR_ACCVGPR 128 ADDR FMT_ANY OPR_VGPR 64 FMT_NUM_B64 OPR_FLAT_SCRATCH 64 FMT_NUM_B128 OPR_GPUMEM 128 FMT_NUM_B32 OPR_SDST_M0 32 VMEM FLAT FALSE FALSE FALSE FALSE FALSE FLAT_STORE_BYTE Store 8 bits of data from a vector register into the flat aperture. ENC_FLAT default 24 ADDR FMT_ANY OPR_VGPR 64 DATA FMT_ANY OPR_VGPR_OR_ACCVGPR 32 FMT_NUM_B8 OPR_GPUMEM 8 FMT_NUM_B64 OPR_FLAT_SCRATCH 64 FMT_NUM_B32 OPR_SDST_M0 32 VMEM FLAT FALSE FALSE FALSE FALSE FALSE FLAT_STORE_BYTE_D16_HI Store 8 bits of data from the high 16 bits of a 32-bit vector register into the flat aperture. ENC_FLAT default 25 ADDR FMT_ANY OPR_VGPR 64 DATA FMT_ANY OPR_VGPR_OR_ACCVGPR 32 FMT_NUM_B8 OPR_GPUMEM 8 FMT_NUM_B64 OPR_FLAT_SCRATCH 64 FMT_NUM_B32 OPR_SDST_M0 32 VMEM FLAT FALSE FALSE FALSE FALSE FALSE FLAT_STORE_SHORT Store 16 bits of data from a vector register into the flat aperture. ENC_FLAT default 26 ADDR FMT_ANY OPR_VGPR 64 DATA FMT_ANY OPR_VGPR_OR_ACCVGPR 32 FMT_NUM_B16 OPR_GPUMEM 16 FMT_NUM_B64 OPR_FLAT_SCRATCH 64 FMT_NUM_B32 OPR_SDST_M0 32 VMEM FLAT FALSE FALSE FALSE FALSE FALSE FLAT_STORE_SHORT_D16_HI Store 16 bits of data from the high 16 bits of a 32-bit vector register into the flat aperture. ENC_FLAT default 27 ADDR FMT_ANY OPR_VGPR 64 DATA FMT_ANY OPR_VGPR_OR_ACCVGPR 32 FMT_NUM_B16 OPR_GPUMEM 16 FMT_NUM_B64 OPR_FLAT_SCRATCH 64 FMT_NUM_B32 OPR_SDST_M0 32 VMEM FLAT FALSE FALSE FALSE FALSE FALSE FLAT_STORE_DWORD Store 32 bits of data from vector input registers into the flat aperture. ENC_FLAT default 28 ADDR FMT_ANY OPR_VGPR 64 DATA FMT_ANY OPR_VGPR_OR_ACCVGPR 32 FMT_NUM_B32 OPR_GPUMEM 32 FMT_NUM_B64 OPR_FLAT_SCRATCH 64 FMT_NUM_B32 OPR_SDST_M0 32 VMEM FLAT FALSE FALSE FALSE FALSE FALSE FLAT_STORE_DWORDX2 Store 64 bits of data from vector input registers into the flat aperture. ENC_FLAT default 29 ADDR FMT_ANY OPR_VGPR 64 DATA FMT_ANY OPR_VGPR_OR_ACCVGPR 64 FMT_NUM_B64 OPR_GPUMEM 64 FMT_NUM_B64 OPR_FLAT_SCRATCH 64 FMT_NUM_B32 OPR_SDST_M0 32 VMEM FLAT FALSE FALSE FALSE FALSE FALSE FLAT_STORE_DWORDX3 Store 96 bits of data from vector input registers into the flat aperture. ENC_FLAT default 30 ADDR FMT_ANY OPR_VGPR 64 DATA FMT_ANY OPR_VGPR_OR_ACCVGPR 96 FMT_NUM_B96 OPR_GPUMEM 96 FMT_NUM_B64 OPR_FLAT_SCRATCH 64 FMT_NUM_B32 OPR_SDST_M0 32 VMEM FLAT FALSE FALSE FALSE FALSE FALSE FLAT_STORE_DWORDX4 Store 128 bits of data from vector input registers into the flat aperture. ENC_FLAT default 31 ADDR FMT_ANY OPR_VGPR 64 DATA FMT_ANY OPR_VGPR_OR_ACCVGPR 128 FMT_NUM_B128 OPR_GPUMEM 128 FMT_NUM_B64 OPR_FLAT_SCRATCH 64 FMT_NUM_B32 OPR_SDST_M0 32 VMEM FLAT FALSE FALSE FALSE FALSE FALSE FLAT_LOAD_UBYTE_D16 Load 8 bits of unsigned data from the flat aperture, zero extend to 16 bits and store the result into the low 16 bits of a 32-bit vector register. ENC_FLAT default 32 VDST FMT_ANY OPR_VGPR_OR_ACCVGPR 32 ADDR FMT_ANY OPR_VGPR 64 FMT_NUM_B64 OPR_FLAT_SCRATCH 64 FMT_NUM_U8 OPR_GPUMEM 8 FMT_NUM_B32 OPR_SDST_M0 32 VMEM FLAT FALSE FALSE FALSE FALSE FALSE FLAT_LOAD_UBYTE_D16_HI Load 8 bits of unsigned data from the flat aperture, zero extend to 16 bits and store the result into the high 16 bits of a 32-bit vector register. ENC_FLAT default 33 VDST FMT_ANY OPR_VGPR_OR_ACCVGPR 32 ADDR FMT_ANY OPR_VGPR 64 FMT_NUM_B64 OPR_FLAT_SCRATCH 64 FMT_NUM_U8 OPR_GPUMEM 8 FMT_NUM_B32 OPR_SDST_M0 32 VMEM FLAT FALSE FALSE FALSE FALSE FALSE FLAT_LOAD_SBYTE_D16 Load 8 bits of signed data from the flat aperture, sign extend to 16 bits and store the result into the low 16 bits of a 32-bit vector register. ENC_FLAT default 34 VDST FMT_ANY OPR_VGPR_OR_ACCVGPR 32 ADDR FMT_ANY OPR_VGPR 64 FMT_NUM_B64 OPR_FLAT_SCRATCH 64 FMT_NUM_I8 OPR_GPUMEM 8 FMT_NUM_B32 OPR_SDST_M0 32 VMEM FLAT FALSE FALSE FALSE FALSE FALSE FLAT_LOAD_SBYTE_D16_HI Load 8 bits of signed data from the flat aperture, sign extend to 16 bits and store the result into the high 16 bits of a 32-bit vector register. ENC_FLAT default 35 VDST FMT_ANY OPR_VGPR_OR_ACCVGPR 32 ADDR FMT_ANY OPR_VGPR 64 FMT_NUM_B64 OPR_FLAT_SCRATCH 64 FMT_NUM_I8 OPR_GPUMEM 8 FMT_NUM_B32 OPR_SDST_M0 32 VMEM FLAT FALSE FALSE FALSE FALSE FALSE FLAT_LOAD_SHORT_D16 Load 16 bits of unsigned data from the flat aperture and store the result into the low 16 bits of a 32-bit vector register. ENC_FLAT default 36 VDST FMT_ANY OPR_VGPR_OR_ACCVGPR 32 ADDR FMT_ANY OPR_VGPR 64 FMT_NUM_B64 OPR_FLAT_SCRATCH 64 FMT_NUM_B16 OPR_GPUMEM 16 FMT_NUM_B32 OPR_SDST_M0 32 VMEM FLAT FALSE FALSE FALSE FALSE FALSE FLAT_LOAD_SHORT_D16_HI Load 16 bits of unsigned data from the flat aperture and store the result into the high 16 bits of a 32-bit vector register. ENC_FLAT default 37 VDST FMT_ANY OPR_VGPR_OR_ACCVGPR 32 ADDR FMT_ANY OPR_VGPR 64 FMT_NUM_B64 OPR_FLAT_SCRATCH 64 FMT_NUM_B16 OPR_GPUMEM 16 FMT_NUM_B32 OPR_SDST_M0 32 VMEM FLAT FALSE FALSE FALSE FALSE FALSE FLAT_ATOMIC_SWAP Swap an unsigned 32-bit integer value in the data register with a location in the flat aperture. Store the original value from flat aperture into a vector register iff the SC0 bit is set. ENC_FLAT default 64 VDST FMT_ANY OPR_VGPR_OR_ACCVGPR 32 ADDR FMT_ANY OPR_VGPR 64 DATA FMT_ANY OPR_VGPR_OR_ACCVGPR 32 FMT_NUM_B32 OPR_GPUMEM 32 FMT_NUM_B64 OPR_FLAT_SCRATCH 64 FMT_NUM_B32 OPR_GPUMEM 32 FMT_NUM_B32 OPR_SDST_M0 32 VMEM ATOMIC FLAT FALSE FALSE FALSE FALSE FALSE FLAT_ATOMIC_CMPSWAP Compare two unsigned 32-bit integer values stored in the data comparison register and a location in the flat aperture. Modify the memory location with a value in the data source register iff the comparison is equal. Store the original value from flat aperture into a vector register iff the SC0 bit is set. ENC_FLAT default 65 VDST FMT_NUM_U32 OPR_VGPR_OR_ACCVGPR 32 ADDR FMT_ANY OPR_VGPR 64 DATA FMT_ANY OPR_VGPR_OR_ACCVGPR 64 FMT_NUM_U32 OPR_GPUMEM 32 FMT_NUM_B64 OPR_FLAT_SCRATCH 64 FMT_NUM_U32 OPR_GPUMEM 32 FMT_NUM_B32 OPR_SDST_M0 32 VMEM ATOMIC FLAT FALSE FALSE FALSE FALSE FALSE FLAT_ATOMIC_ADD Add two unsigned 32-bit integer values stored in the data register and a location in the flat aperture. Store the original value from flat aperture into a vector register iff the SC0 bit is set. ENC_FLAT default 66 VDST FMT_ANY OPR_VGPR_OR_ACCVGPR 32 ADDR FMT_ANY OPR_VGPR 64 DATA FMT_ANY OPR_VGPR_OR_ACCVGPR 32 FMT_NUM_U32 OPR_GPUMEM 32 FMT_NUM_B64 OPR_FLAT_SCRATCH 64 FMT_NUM_U32 OPR_GPUMEM 32 FMT_NUM_B32 OPR_SDST_M0 32 VMEM ATOMIC FLAT FALSE FALSE FALSE FALSE FALSE FLAT_ATOMIC_SUB Subtract an unsigned 32-bit integer value stored in the data register from a value stored in a location in the flat aperture. Store the original value from flat aperture into a vector register iff the SC0 bit is set. ENC_FLAT default 67 VDST FMT_ANY OPR_VGPR_OR_ACCVGPR 32 ADDR FMT_ANY OPR_VGPR 64 DATA FMT_ANY OPR_VGPR_OR_ACCVGPR 32 FMT_NUM_U32 OPR_GPUMEM 32 FMT_NUM_B64 OPR_FLAT_SCRATCH 64 FMT_NUM_U32 OPR_GPUMEM 32 FMT_NUM_B32 OPR_SDST_M0 32 VMEM ATOMIC FLAT FALSE FALSE FALSE FALSE FALSE FLAT_ATOMIC_SMIN Select the minimum of two signed 32-bit integer inputs, given two values stored in the data register and a location in the flat aperture. Update the flat aperture with the selected value. Store the original value from flat aperture into a vector register iff the SC0 bit is set. ENC_FLAT default 68 VDST FMT_ANY OPR_VGPR_OR_ACCVGPR 32 ADDR FMT_ANY OPR_VGPR 64 DATA FMT_ANY OPR_VGPR_OR_ACCVGPR 32 FMT_NUM_I32 OPR_GPUMEM 32 FMT_NUM_B64 OPR_FLAT_SCRATCH 64 FMT_NUM_I32 OPR_GPUMEM 32 FMT_NUM_B32 OPR_SDST_M0 32 VMEM ATOMIC FLAT FALSE FALSE FALSE FALSE FALSE FLAT_ATOMIC_UMIN Select the minimum of two unsigned 32-bit integer inputs, given two values stored in the data register and a location in the flat aperture. Update the flat aperture with the selected value. Store the original value from flat aperture into a vector register iff the SC0 bit is set. ENC_FLAT default 69 VDST FMT_ANY OPR_VGPR_OR_ACCVGPR 32 ADDR FMT_ANY OPR_VGPR 64 DATA FMT_ANY OPR_VGPR_OR_ACCVGPR 32 FMT_NUM_U32 OPR_GPUMEM 32 FMT_NUM_B64 OPR_FLAT_SCRATCH 64 FMT_NUM_U32 OPR_GPUMEM 32 FMT_NUM_B32 OPR_SDST_M0 32 VMEM ATOMIC FLAT FALSE FALSE FALSE FALSE FALSE FLAT_ATOMIC_SMAX Select the maximum of two signed 32-bit integer inputs, given two values stored in the data register and a location in the flat aperture. Update the flat aperture with the selected value. Store the original value from flat aperture into a vector register iff the SC0 bit is set. ENC_FLAT default 70 VDST FMT_ANY OPR_VGPR_OR_ACCVGPR 32 ADDR FMT_ANY OPR_VGPR 64 DATA FMT_ANY OPR_VGPR_OR_ACCVGPR 32 FMT_NUM_I32 OPR_GPUMEM 32 FMT_NUM_B64 OPR_FLAT_SCRATCH 64 FMT_NUM_I32 OPR_GPUMEM 32 FMT_NUM_B32 OPR_SDST_M0 32 VMEM ATOMIC FLAT FALSE FALSE FALSE FALSE FALSE FLAT_ATOMIC_UMAX Select the maximum of two unsigned 32-bit integer inputs, given two values stored in the data register and a location in the flat aperture. Update the flat aperture with the selected value. Store the original value from flat aperture into a vector register iff the SC0 bit is set. ENC_FLAT default 71 VDST FMT_ANY OPR_VGPR_OR_ACCVGPR 32 ADDR FMT_ANY OPR_VGPR 64 DATA FMT_ANY OPR_VGPR_OR_ACCVGPR 32 FMT_NUM_U32 OPR_GPUMEM 32 FMT_NUM_B64 OPR_FLAT_SCRATCH 64 FMT_NUM_U32 OPR_GPUMEM 32 FMT_NUM_B32 OPR_SDST_M0 32 VMEM ATOMIC FLAT FALSE FALSE FALSE FALSE FALSE FLAT_ATOMIC_AND Calculate bitwise AND given two unsigned 32-bit integer values stored in the data register and a location in the flat aperture. Store the original value from flat aperture into a vector register iff the SC0 bit is set. ENC_FLAT default 72 VDST FMT_ANY OPR_VGPR_OR_ACCVGPR 32 ADDR FMT_ANY OPR_VGPR 64 DATA FMT_ANY OPR_VGPR_OR_ACCVGPR 32 FMT_NUM_B32 OPR_GPUMEM 32 FMT_NUM_B64 OPR_FLAT_SCRATCH 64 FMT_NUM_B32 OPR_GPUMEM 32 FMT_NUM_B32 OPR_SDST_M0 32 VMEM ATOMIC FLAT FALSE FALSE FALSE FALSE FALSE FLAT_ATOMIC_OR Calculate bitwise OR given two unsigned 32-bit integer values stored in the data register and a location in the flat aperture. Store the original value from flat aperture into a vector register iff the SC0 bit is set. ENC_FLAT default 73 VDST FMT_ANY OPR_VGPR_OR_ACCVGPR 32 ADDR FMT_ANY OPR_VGPR 64 DATA FMT_ANY OPR_VGPR_OR_ACCVGPR 32 FMT_NUM_B32 OPR_GPUMEM 32 FMT_NUM_B64 OPR_FLAT_SCRATCH 64 FMT_NUM_B32 OPR_GPUMEM 32 FMT_NUM_B32 OPR_SDST_M0 32 VMEM ATOMIC FLAT FALSE FALSE FALSE FALSE FALSE FLAT_ATOMIC_XOR Calculate bitwise XOR given two unsigned 32-bit integer values stored in the data register and a location in the flat aperture. Store the original value from flat aperture into a vector register iff the SC0 bit is set. ENC_FLAT default 74 VDST FMT_ANY OPR_VGPR_OR_ACCVGPR 32 ADDR FMT_ANY OPR_VGPR 64 DATA FMT_ANY OPR_VGPR_OR_ACCVGPR 32 FMT_NUM_B32 OPR_GPUMEM 32 FMT_NUM_B64 OPR_FLAT_SCRATCH 64 FMT_NUM_B32 OPR_GPUMEM 32 FMT_NUM_B32 OPR_SDST_M0 32 VMEM ATOMIC FLAT FALSE FALSE FALSE FALSE FALSE FLAT_ATOMIC_INC Increment an unsigned 32-bit integer value from a location in the flat aperture with wraparound to 0 if the value exceeds a value in the data register. Store the original value from flat aperture into a vector register iff the SC0 bit is set. ENC_FLAT default 75 VDST FMT_ANY OPR_VGPR_OR_ACCVGPR 32 ADDR FMT_ANY OPR_VGPR 64 DATA FMT_ANY OPR_VGPR_OR_ACCVGPR 32 FMT_NUM_U32 OPR_GPUMEM 32 FMT_NUM_B64 OPR_FLAT_SCRATCH 64 FMT_NUM_U32 OPR_GPUMEM 32 FMT_NUM_B32 OPR_SDST_M0 32 VMEM ATOMIC FLAT FALSE FALSE FALSE FALSE FALSE FLAT_ATOMIC_DEC Decrement an unsigned 32-bit integer value from a location in the flat aperture with wraparound to a value in the data register if the decrement yields a negative value. Store the original value from flat aperture into a vector register iff the SC0 bit is set. ENC_FLAT default 76 VDST FMT_ANY OPR_VGPR_OR_ACCVGPR 32 ADDR FMT_ANY OPR_VGPR 64 DATA FMT_ANY OPR_VGPR_OR_ACCVGPR 32 FMT_NUM_U32 OPR_GPUMEM 32 FMT_NUM_B64 OPR_FLAT_SCRATCH 64 FMT_NUM_U32 OPR_GPUMEM 32 FMT_NUM_B32 OPR_SDST_M0 32 VMEM ATOMIC FLAT FALSE FALSE FALSE FALSE FALSE FLAT_ATOMIC_ADD_F32 Add a single-precision float value in the data register to a location in the flat aperture. Store the original value from flat aperture into a vector register iff the SC0 bit is set. ENC_FLAT default 77 VDST FMT_ANY OPR_VGPR_OR_ACCVGPR 32 ADDR FMT_ANY OPR_VGPR 64 DATA FMT_ANY OPR_VGPR_OR_ACCVGPR 32 FMT_NUM_F32 OPR_GPUMEM 32 FMT_NUM_B64 OPR_FLAT_SCRATCH 64 FMT_NUM_F32 OPR_GPUMEM 32 FMT_NUM_B32 OPR_SDST_M0 32 VMEM ATOMIC FLAT FALSE FALSE FALSE FALSE FALSE FLAT_ATOMIC_PK_ADD_F16 Add a packed 2-component half-precision float value in the data register to a location in the flat aperture. Store the original value from flat aperture into a vector register iff the SC0 bit is set. ENC_FLAT default 78 VDST FMT_ANY OPR_VGPR_OR_ACCVGPR 32 ADDR FMT_ANY OPR_VGPR 64 DATA FMT_ANY OPR_VGPR_OR_ACCVGPR 32 FMT_NUM_PK2_F16 OPR_GPUMEM 32 FMT_NUM_B64 OPR_FLAT_SCRATCH 64 FMT_NUM_PK2_F16 OPR_GPUMEM 32 FMT_NUM_B32 OPR_SDST_M0 32 VMEM ATOMIC FLAT FALSE FALSE FALSE FALSE FALSE FLAT_ATOMIC_ADD_F64 Add a double-precision float value in the data register to a location in the flat aperture. Store the original value from flat aperture into a vector register iff the SC0 bit is set. ENC_FLAT default 79 VDST FMT_ANY OPR_VGPR_OR_ACCVGPR 64 ADDR FMT_ANY OPR_VGPR 64 DATA FMT_ANY OPR_VGPR_OR_ACCVGPR 64 FMT_NUM_F64 OPR_GPUMEM 64 FMT_NUM_B64 OPR_FLAT_SCRATCH 64 FMT_NUM_F64 OPR_GPUMEM 64 FMT_NUM_B32 OPR_SDST_M0 32 VMEM ATOMIC FLAT FALSE FALSE FALSE FALSE FALSE FLAT_ATOMIC_MIN_F64 Select the minimum of two double-precision float inputs, given two values stored in the data register and a location in the flat aperture. Update the flat aperture with the selected value. Store the original value from flat aperture into a vector register iff the SC0 bit is set. ENC_FLAT default 80 VDST FMT_ANY OPR_VGPR_OR_ACCVGPR 64 ADDR FMT_ANY OPR_VGPR 64 DATA FMT_ANY OPR_VGPR_OR_ACCVGPR 64 FMT_NUM_F64 OPR_GPUMEM 64 FMT_NUM_B64 OPR_FLAT_SCRATCH 64 FMT_NUM_F64 OPR_GPUMEM 64 FMT_NUM_B32 OPR_SDST_M0 32 VMEM ATOMIC FLAT FALSE FALSE FALSE FALSE FALSE FLAT_ATOMIC_MAX_F64 Select the maximum of two double-precision float inputs, given two values stored in the data register and a location in the flat aperture. Update the flat aperture with the selected value. Store the original value from flat aperture into a vector register iff the SC0 bit is set. ENC_FLAT default 81 VDST FMT_ANY OPR_VGPR_OR_ACCVGPR 64 ADDR FMT_ANY OPR_VGPR 64 DATA FMT_ANY OPR_VGPR_OR_ACCVGPR 64 FMT_NUM_F64 OPR_GPUMEM 64 FMT_NUM_B64 OPR_FLAT_SCRATCH 64 FMT_NUM_F64 OPR_GPUMEM 64 FMT_NUM_B32 OPR_SDST_M0 32 VMEM ATOMIC FLAT FALSE FALSE FALSE FALSE FALSE FLAT_ATOMIC_PK_ADD_BF16 Add a packed 2-component BF16 float value in the data register to a location in the flat aperture. Store the original value from flat aperture into a vector register iff the SC0 bit is set. ENC_FLAT default 82 VDST FMT_ANY OPR_VGPR_OR_ACCVGPR 32 ADDR FMT_ANY OPR_VGPR 64 DATA FMT_ANY OPR_VGPR_OR_ACCVGPR 32 FMT_NUM_PK2_BF16 OPR_GPUMEM 32 FMT_NUM_B64 OPR_FLAT_SCRATCH 64 FMT_NUM_PK2_BF16 OPR_GPUMEM 32 FMT_NUM_B32 OPR_SDST_M0 32 VMEM ATOMIC FLAT FALSE FALSE FALSE FALSE FALSE FLAT_ATOMIC_SWAP_X2 Swap an unsigned 64-bit integer value in the data register with a location in the flat aperture. Store the original value from flat aperture into a vector register iff the SC0 bit is set. ENC_FLAT default 96 VDST FMT_ANY OPR_VGPR_OR_ACCVGPR 64 ADDR FMT_ANY OPR_VGPR 64 DATA FMT_ANY OPR_VGPR_OR_ACCVGPR 64 FMT_NUM_B64 OPR_GPUMEM 64 FMT_NUM_B64 OPR_FLAT_SCRATCH 64 FMT_NUM_B64 OPR_GPUMEM 64 FMT_NUM_B32 OPR_SDST_M0 32 VMEM ATOMIC FLAT FALSE FALSE FALSE FALSE FALSE FLAT_ATOMIC_CMPSWAP_X2 Compare two unsigned 64-bit integer values stored in the data comparison register and a location in the flat aperture. Modify the memory location with a value in the data source register iff the comparison is equal. Store the original value from flat aperture into a vector register iff the SC0 bit is set. ENC_FLAT default 97 VDST FMT_NUM_U64 OPR_VGPR_OR_ACCVGPR 64 ADDR FMT_ANY OPR_VGPR 64 DATA FMT_ANY OPR_VGPR_OR_ACCVGPR 128 FMT_NUM_U64 OPR_GPUMEM 64 FMT_NUM_B64 OPR_FLAT_SCRATCH 64 FMT_NUM_U64 OPR_GPUMEM 64 FMT_NUM_B32 OPR_SDST_M0 32 VMEM ATOMIC FLAT FALSE FALSE FALSE FALSE FALSE FLAT_ATOMIC_ADD_X2 Add two unsigned 64-bit integer values stored in the data register and a location in the flat aperture. Store the original value from flat aperture into a vector register iff the SC0 bit is set. ENC_FLAT default 98 VDST FMT_ANY OPR_VGPR_OR_ACCVGPR 64 ADDR FMT_ANY OPR_VGPR 64 DATA FMT_ANY OPR_VGPR_OR_ACCVGPR 64 FMT_NUM_U64 OPR_GPUMEM 64 FMT_NUM_B64 OPR_FLAT_SCRATCH 64 FMT_NUM_U64 OPR_GPUMEM 64 FMT_NUM_B32 OPR_SDST_M0 32 VMEM ATOMIC FLAT FALSE FALSE FALSE FALSE FALSE FLAT_ATOMIC_SUB_X2 Subtract an unsigned 64-bit integer value stored in the data register from a value stored in a location in the flat aperture. Store the original value from flat aperture into a vector register iff the SC0 bit is set. ENC_FLAT default 99 VDST FMT_ANY OPR_VGPR_OR_ACCVGPR 64 ADDR FMT_ANY OPR_VGPR 64 DATA FMT_ANY OPR_VGPR_OR_ACCVGPR 64 FMT_NUM_U64 OPR_GPUMEM 64 FMT_NUM_B64 OPR_FLAT_SCRATCH 64 FMT_NUM_U64 OPR_GPUMEM 64 FMT_NUM_B32 OPR_SDST_M0 32 VMEM ATOMIC FLAT FALSE FALSE FALSE FALSE FALSE FLAT_ATOMIC_SMIN_X2 Select the minimum of two signed 64-bit integer inputs, given two values stored in the data register and a location in the flat aperture. Update the flat aperture with the selected value. Store the original value from flat aperture into a vector register iff the SC0 bit is set. ENC_FLAT default 100 VDST FMT_ANY OPR_VGPR_OR_ACCVGPR 64 ADDR FMT_ANY OPR_VGPR 64 DATA FMT_ANY OPR_VGPR_OR_ACCVGPR 64 FMT_NUM_I64 OPR_GPUMEM 64 FMT_NUM_B64 OPR_FLAT_SCRATCH 64 FMT_NUM_I64 OPR_GPUMEM 64 FMT_NUM_B32 OPR_SDST_M0 32 VMEM ATOMIC FLAT FALSE FALSE FALSE FALSE FALSE FLAT_ATOMIC_UMIN_X2 Select the minimum of two unsigned 64-bit integer inputs, given two values stored in the data register and a location in the flat aperture. Update the flat aperture with the selected value. Store the original value from flat aperture into a vector register iff the SC0 bit is set. ENC_FLAT default 101 VDST FMT_ANY OPR_VGPR_OR_ACCVGPR 64 ADDR FMT_ANY OPR_VGPR 64 DATA FMT_ANY OPR_VGPR_OR_ACCVGPR 64 FMT_NUM_U64 OPR_GPUMEM 64 FMT_NUM_B64 OPR_FLAT_SCRATCH 64 FMT_NUM_U64 OPR_GPUMEM 64 FMT_NUM_B32 OPR_SDST_M0 32 VMEM ATOMIC FLAT FALSE FALSE FALSE FALSE FALSE FLAT_ATOMIC_SMAX_X2 Select the maximum of two signed 64-bit integer inputs, given two values stored in the data register and a location in the flat aperture. Update the flat aperture with the selected value. Store the original value from flat aperture into a vector register iff the SC0 bit is set. ENC_FLAT default 102 VDST FMT_ANY OPR_VGPR_OR_ACCVGPR 64 ADDR FMT_ANY OPR_VGPR 64 DATA FMT_ANY OPR_VGPR_OR_ACCVGPR 64 FMT_NUM_I64 OPR_GPUMEM 64 FMT_NUM_B64 OPR_FLAT_SCRATCH 64 FMT_NUM_I64 OPR_GPUMEM 64 FMT_NUM_B32 OPR_SDST_M0 32 VMEM ATOMIC FLAT FALSE FALSE FALSE FALSE FALSE FLAT_ATOMIC_UMAX_X2 Select the maximum of two unsigned 64-bit integer inputs, given two values stored in the data register and a location in the flat aperture. Update the flat aperture with the selected value. Store the original value from flat aperture into a vector register iff the SC0 bit is set. ENC_FLAT default 103 VDST FMT_ANY OPR_VGPR_OR_ACCVGPR 64 ADDR FMT_ANY OPR_VGPR 64 DATA FMT_ANY OPR_VGPR_OR_ACCVGPR 64 FMT_NUM_U64 OPR_GPUMEM 64 FMT_NUM_B64 OPR_FLAT_SCRATCH 64 FMT_NUM_U64 OPR_GPUMEM 64 FMT_NUM_B32 OPR_SDST_M0 32 VMEM ATOMIC FLAT FALSE FALSE FALSE FALSE FALSE FLAT_ATOMIC_AND_X2 Calculate bitwise AND given two unsigned 64-bit integer values stored in the data register and a location in the flat aperture. Store the original value from flat aperture into a vector register iff the SC0 bit is set. ENC_FLAT default 104 VDST FMT_ANY OPR_VGPR_OR_ACCVGPR 64 ADDR FMT_ANY OPR_VGPR 64 DATA FMT_ANY OPR_VGPR_OR_ACCVGPR 64 FMT_NUM_B64 OPR_GPUMEM 64 FMT_NUM_B64 OPR_FLAT_SCRATCH 64 FMT_NUM_B64 OPR_GPUMEM 64 FMT_NUM_B32 OPR_SDST_M0 32 VMEM ATOMIC FLAT FALSE FALSE FALSE FALSE FALSE FLAT_ATOMIC_OR_X2 Calculate bitwise OR given two unsigned 64-bit integer values stored in the data register and a location in the flat aperture. Store the original value from flat aperture into a vector register iff the SC0 bit is set. ENC_FLAT default 105 VDST FMT_ANY OPR_VGPR_OR_ACCVGPR 64 ADDR FMT_ANY OPR_VGPR 64 DATA FMT_ANY OPR_VGPR_OR_ACCVGPR 64 FMT_NUM_B64 OPR_GPUMEM 64 FMT_NUM_B64 OPR_FLAT_SCRATCH 64 FMT_NUM_B64 OPR_GPUMEM 64 FMT_NUM_B32 OPR_SDST_M0 32 VMEM ATOMIC FLAT FALSE FALSE FALSE FALSE FALSE FLAT_ATOMIC_XOR_X2 Calculate bitwise XOR given two unsigned 64-bit integer values stored in the data register and a location in the flat aperture. Store the original value from flat aperture into a vector register iff the SC0 bit is set. ENC_FLAT default 106 VDST FMT_ANY OPR_VGPR_OR_ACCVGPR 64 ADDR FMT_ANY OPR_VGPR 64 DATA FMT_ANY OPR_VGPR_OR_ACCVGPR 64 FMT_NUM_B64 OPR_GPUMEM 64 FMT_NUM_B64 OPR_FLAT_SCRATCH 64 FMT_NUM_B64 OPR_GPUMEM 64 FMT_NUM_B32 OPR_SDST_M0 32 VMEM ATOMIC FLAT FALSE FALSE FALSE FALSE FALSE FLAT_ATOMIC_INC_X2 Increment an unsigned 64-bit integer value from a location in the flat aperture with wraparound to 0 if the value exceeds a value in the data register. Store the original value from flat aperture into a vector register iff the SC0 bit is set. ENC_FLAT default 107 VDST FMT_ANY OPR_VGPR_OR_ACCVGPR 64 ADDR FMT_ANY OPR_VGPR 64 DATA FMT_ANY OPR_VGPR_OR_ACCVGPR 64 FMT_NUM_U64 OPR_GPUMEM 64 FMT_NUM_B64 OPR_FLAT_SCRATCH 64 FMT_NUM_U64 OPR_GPUMEM 64 FMT_NUM_B32 OPR_SDST_M0 32 VMEM ATOMIC FLAT FALSE FALSE FALSE FALSE FALSE FLAT_ATOMIC_DEC_X2 Decrement an unsigned 64-bit integer value from a location in the flat aperture with wraparound to a value in the data register if the decrement yields a negative value. Store the original value from flat aperture into a vector register iff the SC0 bit is set. ENC_FLAT default 108 VDST FMT_ANY OPR_VGPR_OR_ACCVGPR 64 ADDR FMT_ANY OPR_VGPR 64 DATA FMT_ANY OPR_VGPR_OR_ACCVGPR 64 FMT_NUM_U64 OPR_GPUMEM 64 FMT_NUM_B64 OPR_FLAT_SCRATCH 64 FMT_NUM_U64 OPR_GPUMEM 64 FMT_NUM_B32 OPR_SDST_M0 32 VMEM ATOMIC FLAT FALSE FALSE FALSE FALSE FALSE GLOBAL_LOAD_UBYTE Load 8 bits of unsigned data from the global aperture, zero extend to 32 bits and store the result into a vector register. ENC_FLAT_GLBL default 16 VDST FMT_ANY OPR_VGPR_OR_ACCVGPR 32 ADDR FMT_ANY OPR_VGPR 64 SADDR FMT_ANY OPR_SREG 64 FMT_NUM_U8 OPR_GPUMEM 8 FMT_NUM_B32 OPR_SDST_M0 32 VMEM FLAT FALSE FALSE FALSE FALSE FALSE GLOBAL_LOAD_SBYTE Load 8 bits of signed data from the global aperture, sign extend to 32 bits and store the result into a vector register. ENC_FLAT_GLBL default 17 VDST FMT_ANY OPR_VGPR_OR_ACCVGPR 32 ADDR FMT_ANY OPR_VGPR 64 SADDR FMT_ANY OPR_SREG 64 FMT_NUM_I8 OPR_GPUMEM 8 FMT_NUM_B32 OPR_SDST_M0 32 VMEM FLAT FALSE FALSE FALSE FALSE FALSE GLOBAL_LOAD_USHORT Load 16 bits of unsigned data from the global aperture, zero extend to 32 bits and store the result into a vector register. ENC_FLAT_GLBL default 18 VDST FMT_ANY OPR_VGPR_OR_ACCVGPR 32 ADDR FMT_ANY OPR_VGPR 64 SADDR FMT_ANY OPR_SREG 64 FMT_NUM_U16 OPR_GPUMEM 16 FMT_NUM_B32 OPR_SDST_M0 32 VMEM FLAT FALSE FALSE FALSE FALSE FALSE GLOBAL_LOAD_SSHORT Load 16 bits of signed data from the global aperture, sign extend to 32 bits and store the result into a vector register. ENC_FLAT_GLBL default 19 VDST FMT_ANY OPR_VGPR_OR_ACCVGPR 32 ADDR FMT_ANY OPR_VGPR 64 SADDR FMT_ANY OPR_SREG 64 FMT_NUM_I16 OPR_GPUMEM 16 FMT_NUM_B32 OPR_SDST_M0 32 VMEM FLAT FALSE FALSE FALSE FALSE FALSE GLOBAL_LOAD_DWORD Load 32 bits of data from the global aperture into a vector register. ENC_FLAT_GLBL default 20 VDST FMT_ANY OPR_VGPR_OR_ACCVGPR 32 ADDR FMT_ANY OPR_VGPR 64 SADDR FMT_ANY OPR_SREG 64 FMT_NUM_B32 OPR_GPUMEM 32 FMT_NUM_B32 OPR_SDST_M0 32 VMEM FLAT FALSE FALSE FALSE FALSE FALSE GLOBAL_LOAD_DWORDX2 Load 64 bits of data from the global aperture into a vector register. ENC_FLAT_GLBL default 21 VDST FMT_ANY OPR_VGPR_OR_ACCVGPR 64 ADDR FMT_ANY OPR_VGPR 64 SADDR FMT_ANY OPR_SREG 64 FMT_NUM_B64 OPR_GPUMEM 64 FMT_NUM_B32 OPR_SDST_M0 32 VMEM FLAT FALSE FALSE FALSE FALSE FALSE GLOBAL_LOAD_DWORDX3 Load 96 bits of data from the global aperture into a vector register. ENC_FLAT_GLBL default 22 VDST FMT_ANY OPR_VGPR_OR_ACCVGPR 96 ADDR FMT_ANY OPR_VGPR 64 SADDR FMT_ANY OPR_SREG 64 FMT_NUM_B96 OPR_GPUMEM 96 FMT_NUM_B32 OPR_SDST_M0 32 VMEM FLAT FALSE FALSE FALSE FALSE FALSE GLOBAL_LOAD_DWORDX4 Load 128 bits of data from the global aperture into a vector register. ENC_FLAT_GLBL default 23 VDST FMT_ANY OPR_VGPR_OR_ACCVGPR 128 ADDR FMT_ANY OPR_VGPR 64 SADDR FMT_ANY OPR_SREG 64 FMT_NUM_B128 OPR_GPUMEM 128 FMT_NUM_B32 OPR_SDST_M0 32 VMEM FLAT FALSE FALSE FALSE FALSE FALSE GLOBAL_STORE_BYTE Store 8 bits of data from a vector register into the global aperture. ENC_FLAT_GLBL default 24 ADDR FMT_ANY OPR_VGPR 64 DATA FMT_ANY OPR_VGPR_OR_ACCVGPR 32 SADDR FMT_ANY OPR_SREG 64 FMT_NUM_B8 OPR_GPUMEM 8 FMT_NUM_B32 OPR_SDST_M0 32 VMEM FLAT FALSE FALSE FALSE FALSE FALSE GLOBAL_STORE_BYTE_D16_HI Store 8 bits of data from the high 16 bits of a 32-bit vector register into the global aperture. ENC_FLAT_GLBL default 25 ADDR FMT_ANY OPR_VGPR 64 DATA FMT_ANY OPR_VGPR_OR_ACCVGPR 32 SADDR FMT_ANY OPR_SREG 64 FMT_NUM_B8 OPR_GPUMEM 8 FMT_NUM_B32 OPR_SDST_M0 32 VMEM FLAT FALSE FALSE FALSE FALSE FALSE GLOBAL_STORE_SHORT Store 16 bits of data from a vector register into the global aperture. ENC_FLAT_GLBL default 26 ADDR FMT_ANY OPR_VGPR 64 DATA FMT_ANY OPR_VGPR_OR_ACCVGPR 32 SADDR FMT_ANY OPR_SREG 64 FMT_NUM_B16 OPR_GPUMEM 16 FMT_NUM_B32 OPR_SDST_M0 32 VMEM FLAT FALSE FALSE FALSE FALSE FALSE GLOBAL_STORE_SHORT_D16_HI Store 16 bits of data from the high 16 bits of a 32-bit vector register into the global aperture. ENC_FLAT_GLBL default 27 ADDR FMT_ANY OPR_VGPR 64 DATA FMT_ANY OPR_VGPR_OR_ACCVGPR 32 SADDR FMT_ANY OPR_SREG 64 FMT_NUM_B16 OPR_GPUMEM 16 FMT_NUM_B32 OPR_SDST_M0 32 VMEM FLAT FALSE FALSE FALSE FALSE FALSE GLOBAL_STORE_DWORD Store 32 bits of data from vector input registers into the global aperture. ENC_FLAT_GLBL default 28 ADDR FMT_ANY OPR_VGPR 64 DATA FMT_ANY OPR_VGPR_OR_ACCVGPR 32 SADDR FMT_ANY OPR_SREG 64 FMT_NUM_B32 OPR_GPUMEM 32 FMT_NUM_B32 OPR_SDST_M0 32 VMEM FLAT FALSE FALSE FALSE FALSE FALSE GLOBAL_STORE_DWORDX2 Store 64 bits of data from vector input registers into the global aperture. ENC_FLAT_GLBL default 29 ADDR FMT_ANY OPR_VGPR 64 DATA FMT_ANY OPR_VGPR_OR_ACCVGPR 64 SADDR FMT_ANY OPR_SREG 64 FMT_NUM_B64 OPR_GPUMEM 64 FMT_NUM_B32 OPR_SDST_M0 32 VMEM FLAT FALSE FALSE FALSE FALSE FALSE GLOBAL_STORE_DWORDX3 Store 96 bits of data from vector input registers into the global aperture. ENC_FLAT_GLBL default 30 ADDR FMT_ANY OPR_VGPR 64 DATA FMT_ANY OPR_VGPR_OR_ACCVGPR 96 SADDR FMT_ANY OPR_SREG 64 FMT_NUM_B96 OPR_GPUMEM 96 FMT_NUM_B32 OPR_SDST_M0 32 VMEM FLAT FALSE FALSE FALSE FALSE FALSE GLOBAL_STORE_DWORDX4 Store 128 bits of data from vector input registers into the global aperture. ENC_FLAT_GLBL default 31 ADDR FMT_ANY OPR_VGPR 64 DATA FMT_ANY OPR_VGPR_OR_ACCVGPR 128 SADDR FMT_ANY OPR_SREG 64 FMT_NUM_B128 OPR_GPUMEM 128 FMT_NUM_B32 OPR_SDST_M0 32 VMEM FLAT FALSE FALSE FALSE FALSE FALSE GLOBAL_LOAD_UBYTE_D16 Load 8 bits of unsigned data from the global aperture, zero extend to 16 bits and store the result into the low 16 bits of a 32-bit vector register. ENC_FLAT_GLBL default 32 VDST FMT_ANY OPR_VGPR_OR_ACCVGPR 32 ADDR FMT_ANY OPR_VGPR 64 SADDR FMT_ANY OPR_SREG 64 FMT_NUM_U8 OPR_GPUMEM 8 FMT_NUM_B32 OPR_SDST_M0 32 VMEM FLAT FALSE FALSE FALSE FALSE FALSE GLOBAL_LOAD_UBYTE_D16_HI Load 8 bits of unsigned data from the global aperture, zero extend to 16 bits and store the result into the high 16 bits of a 32-bit vector register. ENC_FLAT_GLBL default 33 VDST FMT_ANY OPR_VGPR_OR_ACCVGPR 32 ADDR FMT_ANY OPR_VGPR 64 SADDR FMT_ANY OPR_SREG 64 FMT_NUM_U8 OPR_GPUMEM 8 FMT_NUM_B32 OPR_SDST_M0 32 VMEM FLAT FALSE FALSE FALSE FALSE FALSE GLOBAL_LOAD_SBYTE_D16 Load 8 bits of signed data from the global aperture, sign extend to 16 bits and store the result into the low 16 bits of a 32-bit vector register. ENC_FLAT_GLBL default 34 VDST FMT_ANY OPR_VGPR_OR_ACCVGPR 32 ADDR FMT_ANY OPR_VGPR 64 SADDR FMT_ANY OPR_SREG 64 FMT_NUM_I8 OPR_GPUMEM 8 FMT_NUM_B32 OPR_SDST_M0 32 VMEM FLAT FALSE FALSE FALSE FALSE FALSE GLOBAL_LOAD_SBYTE_D16_HI Load 8 bits of signed data from the global aperture, sign extend to 16 bits and store the result into the high 16 bits of a 32-bit vector register. ENC_FLAT_GLBL default 35 VDST FMT_ANY OPR_VGPR_OR_ACCVGPR 32 ADDR FMT_ANY OPR_VGPR 64 SADDR FMT_ANY OPR_SREG 64 FMT_NUM_I8 OPR_GPUMEM 8 FMT_NUM_B32 OPR_SDST_M0 32 VMEM FLAT FALSE FALSE FALSE FALSE FALSE GLOBAL_LOAD_SHORT_D16 Load 16 bits of unsigned data from the global aperture and store the result into the low 16 bits of a 32-bit vector register. ENC_FLAT_GLBL default 36 VDST FMT_ANY OPR_VGPR_OR_ACCVGPR 32 ADDR FMT_ANY OPR_VGPR 64 SADDR FMT_ANY OPR_SREG 64 FMT_NUM_B16 OPR_GPUMEM 16 FMT_NUM_B32 OPR_SDST_M0 32 VMEM FLAT FALSE FALSE FALSE FALSE FALSE GLOBAL_LOAD_SHORT_D16_HI Load 16 bits of unsigned data from the global aperture and store the result into the high 16 bits of a 32-bit vector register. ENC_FLAT_GLBL default 37 VDST FMT_ANY OPR_VGPR_OR_ACCVGPR 32 ADDR FMT_ANY OPR_VGPR 64 SADDR FMT_ANY OPR_SREG 64 FMT_NUM_B16 OPR_GPUMEM 16 FMT_NUM_B32 OPR_SDST_M0 32 VMEM FLAT FALSE FALSE FALSE FALSE FALSE GLOBAL_LOAD_LDS_UBYTE Load 8 bits of untyped data from the global aperture, zero extend to 32 bits and store the result into a data share. ENC_FLAT_GLBL default 38 VDST FMT_ANY OPR_VGPR_OR_ACCVGPR 32 ADDR FMT_ANY OPR_VGPR 64 SADDR FMT_ANY OPR_SREG 64 FMT_NUM_B32 OPR_SDST_M0 32 VMEM FLAT FALSE FALSE FALSE FALSE FALSE GLOBAL_LOAD_LDS_SBYTE Load 8 bits of untyped data from the global aperture, sign extend to 32 bits and store the result into a data share. ENC_FLAT_GLBL default 39 VDST FMT_ANY OPR_VGPR_OR_ACCVGPR 32 ADDR FMT_ANY OPR_VGPR 64 SADDR FMT_ANY OPR_SREG 64 FMT_NUM_B32 OPR_SDST_M0 32 VMEM FLAT FALSE FALSE FALSE FALSE FALSE GLOBAL_LOAD_LDS_USHORT Load 16 bits of untyped data from the global aperture, zero extend to 32 bits and store the result into a data share. ENC_FLAT_GLBL default 40 VDST FMT_ANY OPR_VGPR_OR_ACCVGPR 32 ADDR FMT_ANY OPR_VGPR 64 SADDR FMT_ANY OPR_SREG 64 FMT_NUM_B32 OPR_SDST_M0 32 VMEM FLAT FALSE FALSE FALSE FALSE FALSE GLOBAL_LOAD_LDS_SSHORT Load 16 bits of untyped data from the global aperture, sign extend to 32 bits and store the result into a data share. ENC_FLAT_GLBL default 41 VDST FMT_ANY OPR_VGPR_OR_ACCVGPR 32 ADDR FMT_ANY OPR_VGPR 64 SADDR FMT_ANY OPR_SREG 64 FMT_NUM_B32 OPR_SDST_M0 32 VMEM FLAT FALSE FALSE FALSE FALSE FALSE GLOBAL_LOAD_LDS_DWORD Load 32 bits of untyped data from the global aperture and store the result into a data share. ENC_FLAT_GLBL default 42 VDST FMT_ANY OPR_VGPR_OR_ACCVGPR 32 ADDR FMT_ANY OPR_VGPR 64 SADDR FMT_ANY OPR_SREG 64 FMT_NUM_B32 OPR_SDST_M0 32 VMEM FLAT FALSE FALSE FALSE FALSE FALSE GLOBAL_ATOMIC_SWAP Swap an unsigned 32-bit integer value in the data register with a location in the global aperture. Store the original value from global aperture into a vector register iff the SC0 bit is set. ENC_FLAT_GLBL default 64 VDST FMT_ANY OPR_VGPR_OR_ACCVGPR 32 ADDR FMT_ANY OPR_VGPR 64 DATA FMT_ANY OPR_VGPR_OR_ACCVGPR 32 SADDR FMT_ANY OPR_SREG 64 FMT_NUM_B32 OPR_GPUMEM 32 FMT_NUM_B32 OPR_GPUMEM 32 FMT_NUM_B32 OPR_SDST_M0 32 VMEM ATOMIC FLAT FALSE FALSE FALSE FALSE FALSE GLOBAL_ATOMIC_CMPSWAP Compare two unsigned 32-bit integer values stored in the data comparison register and a location in the global aperture. Modify the memory location with a value in the data source register iff the comparison is equal. Store the original value from global aperture into a vector register iff the SC0 bit is set. ENC_FLAT_GLBL default 65 VDST FMT_NUM_U32 OPR_VGPR_OR_ACCVGPR 32 ADDR FMT_ANY OPR_VGPR 64 DATA FMT_ANY OPR_VGPR_OR_ACCVGPR 64 SADDR FMT_ANY OPR_SREG 64 FMT_NUM_U32 OPR_GPUMEM 32 FMT_NUM_U32 OPR_GPUMEM 32 FMT_NUM_B32 OPR_SDST_M0 32 VMEM ATOMIC FLAT FALSE FALSE FALSE FALSE FALSE GLOBAL_ATOMIC_ADD Add two unsigned 32-bit integer values stored in the data register and a location in the global aperture. Store the original value from global aperture into a vector register iff the SC0 bit is set. ENC_FLAT_GLBL default 66 VDST FMT_ANY OPR_VGPR_OR_ACCVGPR 32 ADDR FMT_ANY OPR_VGPR 64 DATA FMT_ANY OPR_VGPR_OR_ACCVGPR 32 SADDR FMT_ANY OPR_SREG 64 FMT_NUM_U32 OPR_GPUMEM 32 FMT_NUM_U32 OPR_GPUMEM 32 FMT_NUM_B32 OPR_SDST_M0 32 VMEM ATOMIC FLAT FALSE FALSE FALSE FALSE FALSE GLOBAL_ATOMIC_SUB Subtract an unsigned 32-bit integer value stored in the data register from a value stored in a location in the global aperture. Store the original value from global aperture into a vector register iff the SC0 bit is set. ENC_FLAT_GLBL default 67 VDST FMT_ANY OPR_VGPR_OR_ACCVGPR 32 ADDR FMT_ANY OPR_VGPR 64 DATA FMT_ANY OPR_VGPR_OR_ACCVGPR 32 SADDR FMT_ANY OPR_SREG 64 FMT_NUM_U32 OPR_GPUMEM 32 FMT_NUM_U32 OPR_GPUMEM 32 FMT_NUM_B32 OPR_SDST_M0 32 VMEM ATOMIC FLAT FALSE FALSE FALSE FALSE FALSE GLOBAL_ATOMIC_SMIN Select the minimum of two signed 32-bit integer inputs, given two values stored in the data register and a location in the global aperture. Update the global aperture with the selected value. Store the original value from global aperture into a vector register iff the SC0 bit is set. ENC_FLAT_GLBL default 68 VDST FMT_ANY OPR_VGPR_OR_ACCVGPR 32 ADDR FMT_ANY OPR_VGPR 64 DATA FMT_ANY OPR_VGPR_OR_ACCVGPR 32 SADDR FMT_ANY OPR_SREG 64 FMT_NUM_I32 OPR_GPUMEM 32 FMT_NUM_I32 OPR_GPUMEM 32 FMT_NUM_B32 OPR_SDST_M0 32 VMEM ATOMIC FLAT FALSE FALSE FALSE FALSE FALSE GLOBAL_ATOMIC_UMIN Select the minimum of two unsigned 32-bit integer inputs, given two values stored in the data register and a location in the global aperture. Update the global aperture with the selected value. Store the original value from global aperture into a vector register iff the SC0 bit is set. ENC_FLAT_GLBL default 69 VDST FMT_ANY OPR_VGPR_OR_ACCVGPR 32 ADDR FMT_ANY OPR_VGPR 64 DATA FMT_ANY OPR_VGPR_OR_ACCVGPR 32 SADDR FMT_ANY OPR_SREG 64 FMT_NUM_U32 OPR_GPUMEM 32 FMT_NUM_U32 OPR_GPUMEM 32 FMT_NUM_B32 OPR_SDST_M0 32 VMEM ATOMIC FLAT FALSE FALSE FALSE FALSE FALSE GLOBAL_ATOMIC_SMAX Select the maximum of two signed 32-bit integer inputs, given two values stored in the data register and a location in the global aperture. Update the global aperture with the selected value. Store the original value from global aperture into a vector register iff the SC0 bit is set. ENC_FLAT_GLBL default 70 VDST FMT_ANY OPR_VGPR_OR_ACCVGPR 32 ADDR FMT_ANY OPR_VGPR 64 DATA FMT_ANY OPR_VGPR_OR_ACCVGPR 32 SADDR FMT_ANY OPR_SREG 64 FMT_NUM_I32 OPR_GPUMEM 32 FMT_NUM_I32 OPR_GPUMEM 32 FMT_NUM_B32 OPR_SDST_M0 32 VMEM ATOMIC FLAT FALSE FALSE FALSE FALSE FALSE GLOBAL_ATOMIC_UMAX Select the maximum of two unsigned 32-bit integer inputs, given two values stored in the data register and a location in the global aperture. Update the global aperture with the selected value. Store the original value from global aperture into a vector register iff the SC0 bit is set. ENC_FLAT_GLBL default 71 VDST FMT_ANY OPR_VGPR_OR_ACCVGPR 32 ADDR FMT_ANY OPR_VGPR 64 DATA FMT_ANY OPR_VGPR_OR_ACCVGPR 32 SADDR FMT_ANY OPR_SREG 64 FMT_NUM_U32 OPR_GPUMEM 32 FMT_NUM_U32 OPR_GPUMEM 32 FMT_NUM_B32 OPR_SDST_M0 32 VMEM ATOMIC FLAT FALSE FALSE FALSE FALSE FALSE GLOBAL_ATOMIC_AND Calculate bitwise AND given two unsigned 32-bit integer values stored in the data register and a location in the global aperture. Store the original value from global aperture into a vector register iff the SC0 bit is set. ENC_FLAT_GLBL default 72 VDST FMT_ANY OPR_VGPR_OR_ACCVGPR 32 ADDR FMT_ANY OPR_VGPR 64 DATA FMT_ANY OPR_VGPR_OR_ACCVGPR 32 SADDR FMT_ANY OPR_SREG 64 FMT_NUM_B32 OPR_GPUMEM 32 FMT_NUM_B32 OPR_GPUMEM 32 FMT_NUM_B32 OPR_SDST_M0 32 VMEM ATOMIC FLAT FALSE FALSE FALSE FALSE FALSE GLOBAL_ATOMIC_OR Calculate bitwise OR given two unsigned 32-bit integer values stored in the data register and a location in the global aperture. Store the original value from global aperture into a vector register iff the SC0 bit is set. ENC_FLAT_GLBL default 73 VDST FMT_ANY OPR_VGPR_OR_ACCVGPR 32 ADDR FMT_ANY OPR_VGPR 64 DATA FMT_ANY OPR_VGPR_OR_ACCVGPR 32 SADDR FMT_ANY OPR_SREG 64 FMT_NUM_B32 OPR_GPUMEM 32 FMT_NUM_B32 OPR_GPUMEM 32 FMT_NUM_B32 OPR_SDST_M0 32 VMEM ATOMIC FLAT FALSE FALSE FALSE FALSE FALSE GLOBAL_ATOMIC_XOR Calculate bitwise XOR given two unsigned 32-bit integer values stored in the data register and a location in the global aperture. Store the original value from global aperture into a vector register iff the SC0 bit is set. ENC_FLAT_GLBL default 74 VDST FMT_ANY OPR_VGPR_OR_ACCVGPR 32 ADDR FMT_ANY OPR_VGPR 64 DATA FMT_ANY OPR_VGPR_OR_ACCVGPR 32 SADDR FMT_ANY OPR_SREG 64 FMT_NUM_B32 OPR_GPUMEM 32 FMT_NUM_B32 OPR_GPUMEM 32 FMT_NUM_B32 OPR_SDST_M0 32 VMEM ATOMIC FLAT FALSE FALSE FALSE FALSE FALSE GLOBAL_ATOMIC_INC Increment an unsigned 32-bit integer value from a location in the global aperture with wraparound to 0 if the value exceeds a value in the data register. Store the original value from global aperture into a vector register iff the SC0 bit is set. ENC_FLAT_GLBL default 75 VDST FMT_ANY OPR_VGPR_OR_ACCVGPR 32 ADDR FMT_ANY OPR_VGPR 64 DATA FMT_ANY OPR_VGPR_OR_ACCVGPR 32 SADDR FMT_ANY OPR_SREG 64 FMT_NUM_U32 OPR_GPUMEM 32 FMT_NUM_U32 OPR_GPUMEM 32 FMT_NUM_B32 OPR_SDST_M0 32 VMEM ATOMIC FLAT FALSE FALSE FALSE FALSE FALSE GLOBAL_ATOMIC_DEC Decrement an unsigned 32-bit integer value from a location in the global aperture with wraparound to a value in the data register if the decrement yields a negative value. Store the original value from global aperture into a vector register iff the SC0 bit is set. ENC_FLAT_GLBL default 76 VDST FMT_ANY OPR_VGPR_OR_ACCVGPR 32 ADDR FMT_ANY OPR_VGPR 64 DATA FMT_ANY OPR_VGPR_OR_ACCVGPR 32 SADDR FMT_ANY OPR_SREG 64 FMT_NUM_U32 OPR_GPUMEM 32 FMT_NUM_U32 OPR_GPUMEM 32 FMT_NUM_B32 OPR_SDST_M0 32 VMEM ATOMIC FLAT FALSE FALSE FALSE FALSE FALSE GLOBAL_ATOMIC_ADD_F32 Add a single-precision float value in the data register to a location in the global aperture. Store the original value from global aperture into a vector register iff the SC0 bit is set. ENC_FLAT_GLBL default 77 VDST FMT_ANY OPR_VGPR_OR_ACCVGPR 32 ADDR FMT_ANY OPR_VGPR 64 DATA FMT_ANY OPR_VGPR_OR_ACCVGPR 32 SADDR FMT_ANY OPR_SREG 64 FMT_NUM_F32 OPR_GPUMEM 32 FMT_NUM_F32 OPR_GPUMEM 32 FMT_NUM_B32 OPR_SDST_M0 32 VMEM ATOMIC FLAT FALSE FALSE FALSE FALSE FALSE GLOBAL_ATOMIC_PK_ADD_F16 Add a packed 2-component half-precision float value in the data register to a location in the global aperture. Store the original value from global aperture into a vector register iff the SC0 bit is set. ENC_FLAT_GLBL default 78 VDST FMT_ANY OPR_VGPR_OR_ACCVGPR 32 ADDR FMT_ANY OPR_VGPR 64 DATA FMT_ANY OPR_VGPR_OR_ACCVGPR 32 SADDR FMT_ANY OPR_SREG 64 FMT_NUM_PK2_F16 OPR_GPUMEM 32 FMT_NUM_PK2_F16 OPR_GPUMEM 32 FMT_NUM_B32 OPR_SDST_M0 32 VMEM ATOMIC FLAT FALSE FALSE FALSE FALSE FALSE GLOBAL_ATOMIC_ADD_F64 Add a double-precision float value in the data register to a location in the global aperture. Store the original value from global aperture into a vector register iff the SC0 bit is set. ENC_FLAT_GLBL default 79 VDST FMT_ANY OPR_VGPR_OR_ACCVGPR 64 ADDR FMT_ANY OPR_VGPR 64 DATA FMT_ANY OPR_VGPR_OR_ACCVGPR 64 SADDR FMT_ANY OPR_SREG 64 FMT_NUM_F64 OPR_GPUMEM 64 FMT_NUM_F64 OPR_GPUMEM 64 FMT_NUM_B32 OPR_SDST_M0 32 VMEM ATOMIC FLAT FALSE FALSE FALSE FALSE FALSE GLOBAL_ATOMIC_MIN_F64 Select the minimum of two double-precision float inputs, given two values stored in the data register and a location in the global aperture. Update the global aperture with the selected value. Store the original value from global aperture into a vector register iff the SC0 bit is set. ENC_FLAT_GLBL default 80 VDST FMT_ANY OPR_VGPR_OR_ACCVGPR 64 ADDR FMT_ANY OPR_VGPR 64 DATA FMT_ANY OPR_VGPR_OR_ACCVGPR 64 SADDR FMT_ANY OPR_SREG 64 FMT_NUM_F64 OPR_GPUMEM 64 FMT_NUM_F64 OPR_GPUMEM 64 FMT_NUM_B32 OPR_SDST_M0 32 VMEM ATOMIC FLAT FALSE FALSE FALSE FALSE FALSE GLOBAL_ATOMIC_MAX_F64 Select the maximum of two double-precision float inputs, given two values stored in the data register and a location in the global aperture. Update the global aperture with the selected value. Store the original value from global aperture into a vector register iff the SC0 bit is set. ENC_FLAT_GLBL default 81 VDST FMT_ANY OPR_VGPR_OR_ACCVGPR 64 ADDR FMT_ANY OPR_VGPR 64 DATA FMT_ANY OPR_VGPR_OR_ACCVGPR 64 SADDR FMT_ANY OPR_SREG 64 FMT_NUM_F64 OPR_GPUMEM 64 FMT_NUM_F64 OPR_GPUMEM 64 FMT_NUM_B32 OPR_SDST_M0 32 VMEM ATOMIC FLAT FALSE FALSE FALSE FALSE FALSE GLOBAL_ATOMIC_PK_ADD_BF16 Add a packed 2-component BF16 float value in the data register to a location in the global aperture. Store the original value from global aperture into a vector register iff the SC0 bit is set. ENC_FLAT_GLBL default 82 VDST FMT_ANY OPR_VGPR_OR_ACCVGPR 32 ADDR FMT_ANY OPR_VGPR 64 DATA FMT_ANY OPR_VGPR_OR_ACCVGPR 32 SADDR FMT_ANY OPR_SREG 64 FMT_NUM_PK2_BF16 OPR_GPUMEM 32 FMT_NUM_PK2_BF16 OPR_GPUMEM 32 FMT_NUM_B32 OPR_SDST_M0 32 VMEM ATOMIC FLAT FALSE FALSE FALSE FALSE FALSE GLOBAL_ATOMIC_SWAP_X2 Swap an unsigned 64-bit integer value in the data register with a location in the global aperture. Store the original value from global aperture into a vector register iff the SC0 bit is set. ENC_FLAT_GLBL default 96 VDST FMT_ANY OPR_VGPR_OR_ACCVGPR 64 ADDR FMT_ANY OPR_VGPR 64 DATA FMT_ANY OPR_VGPR_OR_ACCVGPR 64 SADDR FMT_ANY OPR_SREG 64 FMT_NUM_B64 OPR_GPUMEM 64 FMT_NUM_B64 OPR_GPUMEM 64 FMT_NUM_B32 OPR_SDST_M0 32 VMEM ATOMIC FLAT FALSE FALSE FALSE FALSE FALSE GLOBAL_ATOMIC_CMPSWAP_X2 Compare two unsigned 64-bit integer values stored in the data comparison register and a location in the global aperture. Modify the memory location with a value in the data source register iff the comparison is equal. Store the original value from global aperture into a vector register iff the SC0 bit is set. ENC_FLAT_GLBL default 97 VDST FMT_NUM_U64 OPR_VGPR_OR_ACCVGPR 64 ADDR FMT_ANY OPR_VGPR 64 DATA FMT_ANY OPR_VGPR_OR_ACCVGPR 128 SADDR FMT_ANY OPR_SREG 64 FMT_NUM_U64 OPR_GPUMEM 64 FMT_NUM_U64 OPR_GPUMEM 64 FMT_NUM_B32 OPR_SDST_M0 32 VMEM ATOMIC FLAT FALSE FALSE FALSE FALSE FALSE GLOBAL_ATOMIC_ADD_X2 Add two unsigned 64-bit integer values stored in the data register and a location in the global aperture. Store the original value from global aperture into a vector register iff the SC0 bit is set. ENC_FLAT_GLBL default 98 VDST FMT_ANY OPR_VGPR_OR_ACCVGPR 64 ADDR FMT_ANY OPR_VGPR 64 DATA FMT_ANY OPR_VGPR_OR_ACCVGPR 64 SADDR FMT_ANY OPR_SREG 64 FMT_NUM_U64 OPR_GPUMEM 64 FMT_NUM_U64 OPR_GPUMEM 64 FMT_NUM_B32 OPR_SDST_M0 32 VMEM ATOMIC FLAT FALSE FALSE FALSE FALSE FALSE GLOBAL_ATOMIC_SUB_X2 Subtract an unsigned 64-bit integer value stored in the data register from a value stored in a location in the global aperture. Store the original value from global aperture into a vector register iff the SC0 bit is set. ENC_FLAT_GLBL default 99 VDST FMT_ANY OPR_VGPR_OR_ACCVGPR 64 ADDR FMT_ANY OPR_VGPR 64 DATA FMT_ANY OPR_VGPR_OR_ACCVGPR 64 SADDR FMT_ANY OPR_SREG 64 FMT_NUM_U64 OPR_GPUMEM 64 FMT_NUM_U64 OPR_GPUMEM 64 FMT_NUM_B32 OPR_SDST_M0 32 VMEM ATOMIC FLAT FALSE FALSE FALSE FALSE FALSE GLOBAL_ATOMIC_SMIN_X2 Select the minimum of two signed 64-bit integer inputs, given two values stored in the data register and a location in the global aperture. Update the global aperture with the selected value. Store the original value from global aperture into a vector register iff the SC0 bit is set. ENC_FLAT_GLBL default 100 VDST FMT_ANY OPR_VGPR_OR_ACCVGPR 64 ADDR FMT_ANY OPR_VGPR 64 DATA FMT_ANY OPR_VGPR_OR_ACCVGPR 64 SADDR FMT_ANY OPR_SREG 64 FMT_NUM_I64 OPR_GPUMEM 64 FMT_NUM_I64 OPR_GPUMEM 64 FMT_NUM_B32 OPR_SDST_M0 32 VMEM ATOMIC FLAT FALSE FALSE FALSE FALSE FALSE GLOBAL_ATOMIC_UMIN_X2 Select the minimum of two unsigned 64-bit integer inputs, given two values stored in the data register and a location in the global aperture. Update the global aperture with the selected value. Store the original value from global aperture into a vector register iff the SC0 bit is set. ENC_FLAT_GLBL default 101 VDST FMT_ANY OPR_VGPR_OR_ACCVGPR 64 ADDR FMT_ANY OPR_VGPR 64 DATA FMT_ANY OPR_VGPR_OR_ACCVGPR 64 SADDR FMT_ANY OPR_SREG 64 FMT_NUM_U64 OPR_GPUMEM 64 FMT_NUM_U64 OPR_GPUMEM 64 FMT_NUM_B32 OPR_SDST_M0 32 VMEM ATOMIC FLAT FALSE FALSE FALSE FALSE FALSE GLOBAL_ATOMIC_SMAX_X2 Select the maximum of two signed 64-bit integer inputs, given two values stored in the data register and a location in the global aperture. Update the global aperture with the selected value. Store the original value from global aperture into a vector register iff the SC0 bit is set. ENC_FLAT_GLBL default 102 VDST FMT_ANY OPR_VGPR_OR_ACCVGPR 64 ADDR FMT_ANY OPR_VGPR 64 DATA FMT_ANY OPR_VGPR_OR_ACCVGPR 64 SADDR FMT_ANY OPR_SREG 64 FMT_NUM_I64 OPR_GPUMEM 64 FMT_NUM_I64 OPR_GPUMEM 64 FMT_NUM_B32 OPR_SDST_M0 32 VMEM ATOMIC FLAT FALSE FALSE FALSE FALSE FALSE GLOBAL_ATOMIC_UMAX_X2 Select the maximum of two unsigned 64-bit integer inputs, given two values stored in the data register and a location in the global aperture. Update the global aperture with the selected value. Store the original value from global aperture into a vector register iff the SC0 bit is set. ENC_FLAT_GLBL default 103 VDST FMT_ANY OPR_VGPR_OR_ACCVGPR 64 ADDR FMT_ANY OPR_VGPR 64 DATA FMT_ANY OPR_VGPR_OR_ACCVGPR 64 SADDR FMT_ANY OPR_SREG 64 FMT_NUM_U64 OPR_GPUMEM 64 FMT_NUM_U64 OPR_GPUMEM 64 FMT_NUM_B32 OPR_SDST_M0 32 VMEM ATOMIC FLAT FALSE FALSE FALSE FALSE FALSE GLOBAL_ATOMIC_AND_X2 Calculate bitwise AND given two unsigned 64-bit integer values stored in the data register and a location in the global aperture. Store the original value from global aperture into a vector register iff the SC0 bit is set. ENC_FLAT_GLBL default 104 VDST FMT_ANY OPR_VGPR_OR_ACCVGPR 64 ADDR FMT_ANY OPR_VGPR 64 DATA FMT_ANY OPR_VGPR_OR_ACCVGPR 64 SADDR FMT_ANY OPR_SREG 64 FMT_NUM_B64 OPR_GPUMEM 64 FMT_NUM_B64 OPR_GPUMEM 64 FMT_NUM_B32 OPR_SDST_M0 32 VMEM ATOMIC FLAT FALSE FALSE FALSE FALSE FALSE GLOBAL_ATOMIC_OR_X2 Calculate bitwise OR given two unsigned 64-bit integer values stored in the data register and a location in the global aperture. Store the original value from global aperture into a vector register iff the SC0 bit is set. ENC_FLAT_GLBL default 105 VDST FMT_ANY OPR_VGPR_OR_ACCVGPR 64 ADDR FMT_ANY OPR_VGPR 64 DATA FMT_ANY OPR_VGPR_OR_ACCVGPR 64 SADDR FMT_ANY OPR_SREG 64 FMT_NUM_B64 OPR_GPUMEM 64 FMT_NUM_B64 OPR_GPUMEM 64 FMT_NUM_B32 OPR_SDST_M0 32 VMEM ATOMIC FLAT FALSE FALSE FALSE FALSE FALSE GLOBAL_ATOMIC_XOR_X2 Calculate bitwise XOR given two unsigned 64-bit integer values stored in the data register and a location in the global aperture. Store the original value from global aperture into a vector register iff the SC0 bit is set. ENC_FLAT_GLBL default 106 VDST FMT_ANY OPR_VGPR_OR_ACCVGPR 64 ADDR FMT_ANY OPR_VGPR 64 DATA FMT_ANY OPR_VGPR_OR_ACCVGPR 64 SADDR FMT_ANY OPR_SREG 64 FMT_NUM_B64 OPR_GPUMEM 64 FMT_NUM_B64 OPR_GPUMEM 64 FMT_NUM_B32 OPR_SDST_M0 32 VMEM ATOMIC FLAT FALSE FALSE FALSE FALSE FALSE GLOBAL_ATOMIC_INC_X2 Increment an unsigned 64-bit integer value from a location in the global aperture with wraparound to 0 if the value exceeds a value in the data register. Store the original value from global aperture into a vector register iff the SC0 bit is set. ENC_FLAT_GLBL default 107 VDST FMT_ANY OPR_VGPR_OR_ACCVGPR 64 ADDR FMT_ANY OPR_VGPR 64 DATA FMT_ANY OPR_VGPR_OR_ACCVGPR 64 SADDR FMT_ANY OPR_SREG 64 FMT_NUM_U64 OPR_GPUMEM 64 FMT_NUM_U64 OPR_GPUMEM 64 FMT_NUM_B32 OPR_SDST_M0 32 VMEM ATOMIC FLAT FALSE FALSE FALSE FALSE FALSE GLOBAL_ATOMIC_DEC_X2 Decrement an unsigned 64-bit integer value from a location in the global aperture with wraparound to a value in the data register if the decrement yields a negative value. Store the original value from global aperture into a vector register iff the SC0 bit is set. ENC_FLAT_GLBL default 108 VDST FMT_ANY OPR_VGPR_OR_ACCVGPR 64 ADDR FMT_ANY OPR_VGPR 64 DATA FMT_ANY OPR_VGPR_OR_ACCVGPR 64 SADDR FMT_ANY OPR_SREG 64 FMT_NUM_U64 OPR_GPUMEM 64 FMT_NUM_U64 OPR_GPUMEM 64 FMT_NUM_B32 OPR_SDST_M0 32 VMEM ATOMIC FLAT FALSE FALSE FALSE FALSE FALSE GLOBAL_LOAD_LDS_DWORDX4 Untyped buffer load 4 dwords, store result into data share. ENC_FLAT_GLBL default 125 VDST FMT_ANY OPR_VGPR_OR_ACCVGPR 128 ADDR FMT_ANY OPR_VGPR 64 SADDR FMT_ANY OPR_SREG 64 FMT_NUM_B32 OPR_SDST_M0 32 VMEM FLAT FALSE FALSE FALSE FALSE FALSE GLOBAL_LOAD_LDS_DWORDX3 Untyped buffer load 3 dwords, store result into data share. ENC_FLAT_GLBL default 126 VDST FMT_ANY OPR_VGPR_OR_ACCVGPR 96 ADDR FMT_ANY OPR_VGPR 64 SADDR FMT_ANY OPR_SREG 64 FMT_NUM_B32 OPR_SDST_M0 32 VMEM FLAT FALSE FALSE FALSE FALSE FALSE SCRATCH_LOAD_UBYTE Load 8 bits of unsigned data from the scratch aperture, zero extend to 32 bits and store the result into a vector register. ENC_FLAT_SCRATCH default 16 VDST FMT_ANY OPR_VGPR_OR_ACCVGPR 32 ADDR FMT_ANY OPR_VGPR 32 SADDR FMT_ANY OPR_SREG 32 FMT_NUM_B64 OPR_FLAT_SCRATCH 64 FMT_NUM_U8 OPR_GPUMEM 8 FMT_NUM_B32 OPR_SDST_M0 32 VMEM FLAT FALSE FALSE FALSE FALSE FALSE SCRATCH_LOAD_SBYTE Load 8 bits of signed data from the scratch aperture, sign extend to 32 bits and store the result into a vector register. ENC_FLAT_SCRATCH default 17 VDST FMT_ANY OPR_VGPR_OR_ACCVGPR 32 ADDR FMT_ANY OPR_VGPR 32 SADDR FMT_ANY OPR_SREG 32 FMT_NUM_B64 OPR_FLAT_SCRATCH 64 FMT_NUM_I8 OPR_GPUMEM 8 FMT_NUM_B32 OPR_SDST_M0 32 VMEM FLAT FALSE FALSE FALSE FALSE FALSE SCRATCH_LOAD_USHORT Load 16 bits of unsigned data from the scratch aperture, zero extend to 32 bits and store the result into a vector register. ENC_FLAT_SCRATCH default 18 VDST FMT_ANY OPR_VGPR_OR_ACCVGPR 32 ADDR FMT_ANY OPR_VGPR 32 SADDR FMT_ANY OPR_SREG 32 FMT_NUM_B64 OPR_FLAT_SCRATCH 64 FMT_NUM_U16 OPR_GPUMEM 16 FMT_NUM_B32 OPR_SDST_M0 32 VMEM FLAT FALSE FALSE FALSE FALSE FALSE SCRATCH_LOAD_SSHORT Load 16 bits of signed data from the scratch aperture, sign extend to 32 bits and store the result into a vector register. ENC_FLAT_SCRATCH default 19 VDST FMT_ANY OPR_VGPR_OR_ACCVGPR 32 ADDR FMT_ANY OPR_VGPR 32 SADDR FMT_ANY OPR_SREG 32 FMT_NUM_B64 OPR_FLAT_SCRATCH 64 FMT_NUM_I16 OPR_GPUMEM 16 FMT_NUM_B32 OPR_SDST_M0 32 VMEM FLAT FALSE FALSE FALSE FALSE FALSE SCRATCH_LOAD_DWORD Load 32 bits of data from the scratch aperture into a vector register. ENC_FLAT_SCRATCH default 20 VDST FMT_ANY OPR_VGPR_OR_ACCVGPR 32 ADDR FMT_ANY OPR_VGPR 32 SADDR FMT_ANY OPR_SREG 32 FMT_NUM_B64 OPR_FLAT_SCRATCH 64 FMT_NUM_B32 OPR_GPUMEM 32 FMT_NUM_B32 OPR_SDST_M0 32 VMEM FLAT FALSE FALSE FALSE FALSE FALSE SCRATCH_LOAD_DWORDX2 Load 64 bits of data from the scratch aperture into a vector register. ENC_FLAT_SCRATCH default 21 VDST FMT_ANY OPR_VGPR_OR_ACCVGPR 64 ADDR FMT_ANY OPR_VGPR 32 SADDR FMT_ANY OPR_SREG 32 FMT_NUM_B64 OPR_FLAT_SCRATCH 64 FMT_NUM_B64 OPR_GPUMEM 64 FMT_NUM_B32 OPR_SDST_M0 32 VMEM FLAT FALSE FALSE FALSE FALSE FALSE SCRATCH_LOAD_DWORDX3 Load 96 bits of data from the scratch aperture into a vector register. ENC_FLAT_SCRATCH default 22 VDST FMT_ANY OPR_VGPR_OR_ACCVGPR 96 ADDR FMT_ANY OPR_VGPR 32 SADDR FMT_ANY OPR_SREG 32 FMT_NUM_B64 OPR_FLAT_SCRATCH 64 FMT_NUM_B96 OPR_GPUMEM 96 FMT_NUM_B32 OPR_SDST_M0 32 VMEM FLAT FALSE FALSE FALSE FALSE FALSE SCRATCH_LOAD_DWORDX4 Load 128 bits of data from the scratch aperture into a vector register. ENC_FLAT_SCRATCH default 23 VDST FMT_ANY OPR_VGPR_OR_ACCVGPR 128 ADDR FMT_ANY OPR_VGPR 32 SADDR FMT_ANY OPR_SREG 32 FMT_NUM_B64 OPR_FLAT_SCRATCH 64 FMT_NUM_B128 OPR_GPUMEM 128 FMT_NUM_B32 OPR_SDST_M0 32 VMEM FLAT FALSE FALSE FALSE FALSE FALSE SCRATCH_STORE_BYTE Store 8 bits of data from a vector register into the scratch aperture. ENC_FLAT_SCRATCH default 24 ADDR FMT_ANY OPR_VGPR 32 DATA FMT_ANY OPR_VGPR_OR_ACCVGPR 32 SADDR FMT_ANY OPR_SREG 32 FMT_NUM_B8 OPR_GPUMEM 8 FMT_NUM_B64 OPR_FLAT_SCRATCH 64 FMT_NUM_B32 OPR_SDST_M0 32 VMEM FLAT FALSE FALSE FALSE FALSE FALSE SCRATCH_STORE_BYTE_D16_HI Store 8 bits of data from the high 16 bits of a 32-bit vector register into the scratch aperture. ENC_FLAT_SCRATCH default 25 ADDR FMT_ANY OPR_VGPR 32 DATA FMT_ANY OPR_VGPR_OR_ACCVGPR 32 SADDR FMT_ANY OPR_SREG 32 FMT_NUM_B8 OPR_GPUMEM 8 FMT_NUM_B64 OPR_FLAT_SCRATCH 64 FMT_NUM_B32 OPR_SDST_M0 32 VMEM FLAT FALSE FALSE FALSE FALSE FALSE SCRATCH_STORE_SHORT Store 16 bits of data from a vector register into the scratch aperture. ENC_FLAT_SCRATCH default 26 ADDR FMT_ANY OPR_VGPR 32 DATA FMT_ANY OPR_VGPR_OR_ACCVGPR 32 SADDR FMT_ANY OPR_SREG 32 FMT_NUM_B16 OPR_GPUMEM 16 FMT_NUM_B64 OPR_FLAT_SCRATCH 64 FMT_NUM_B32 OPR_SDST_M0 32 VMEM FLAT FALSE FALSE FALSE FALSE FALSE SCRATCH_STORE_SHORT_D16_HI Store 16 bits of data from the high 16 bits of a 32-bit vector register into the scratch aperture. ENC_FLAT_SCRATCH default 27 ADDR FMT_ANY OPR_VGPR 32 DATA FMT_ANY OPR_VGPR_OR_ACCVGPR 32 SADDR FMT_ANY OPR_SREG 32 FMT_NUM_B16 OPR_GPUMEM 16 FMT_NUM_B64 OPR_FLAT_SCRATCH 64 FMT_NUM_B32 OPR_SDST_M0 32 VMEM FLAT FALSE FALSE FALSE FALSE FALSE SCRATCH_STORE_DWORD Store 32 bits of data from vector input registers into the scratch aperture. ENC_FLAT_SCRATCH default 28 ADDR FMT_ANY OPR_VGPR 32 DATA FMT_ANY OPR_VGPR_OR_ACCVGPR 32 SADDR FMT_ANY OPR_SREG 32 FMT_NUM_B32 OPR_GPUMEM 32 FMT_NUM_B64 OPR_FLAT_SCRATCH 64 FMT_NUM_B32 OPR_SDST_M0 32 VMEM FLAT FALSE FALSE FALSE FALSE FALSE SCRATCH_STORE_DWORDX2 Store 64 bits of data from vector input registers into the scratch aperture. ENC_FLAT_SCRATCH default 29 ADDR FMT_ANY OPR_VGPR 32 DATA FMT_ANY OPR_VGPR_OR_ACCVGPR 64 SADDR FMT_ANY OPR_SREG 32 FMT_NUM_B64 OPR_GPUMEM 64 FMT_NUM_B64 OPR_FLAT_SCRATCH 64 FMT_NUM_B32 OPR_SDST_M0 32 VMEM FLAT FALSE FALSE FALSE FALSE FALSE SCRATCH_STORE_DWORDX3 Store 96 bits of data from vector input registers into the scratch aperture. ENC_FLAT_SCRATCH default 30 ADDR FMT_ANY OPR_VGPR 32 DATA FMT_ANY OPR_VGPR_OR_ACCVGPR 96 SADDR FMT_ANY OPR_SREG 32 FMT_NUM_B96 OPR_GPUMEM 96 FMT_NUM_B64 OPR_FLAT_SCRATCH 64 FMT_NUM_B32 OPR_SDST_M0 32 VMEM FLAT FALSE FALSE FALSE FALSE FALSE SCRATCH_STORE_DWORDX4 Store 128 bits of data from vector input registers into the scratch aperture. ENC_FLAT_SCRATCH default 31 ADDR FMT_ANY OPR_VGPR 32 DATA FMT_ANY OPR_VGPR_OR_ACCVGPR 128 SADDR FMT_ANY OPR_SREG 32 FMT_NUM_B128 OPR_GPUMEM 128 FMT_NUM_B64 OPR_FLAT_SCRATCH 64 FMT_NUM_B32 OPR_SDST_M0 32 VMEM FLAT FALSE FALSE FALSE FALSE FALSE SCRATCH_LOAD_UBYTE_D16 Load 8 bits of unsigned data from the scratch aperture, zero extend to 16 bits and store the result into the low 16 bits of a 32-bit vector register. ENC_FLAT_SCRATCH default 32 VDST FMT_ANY OPR_VGPR_OR_ACCVGPR 32 ADDR FMT_ANY OPR_VGPR 32 SADDR FMT_ANY OPR_SREG 32 FMT_NUM_B64 OPR_FLAT_SCRATCH 64 FMT_NUM_U8 OPR_GPUMEM 8 FMT_NUM_B32 OPR_SDST_M0 32 VMEM FLAT FALSE FALSE FALSE FALSE FALSE SCRATCH_LOAD_UBYTE_D16_HI Load 8 bits of unsigned data from the scratch aperture, zero extend to 16 bits and store the result into the high 16 bits of a 32-bit vector register. ENC_FLAT_SCRATCH default 33 VDST FMT_ANY OPR_VGPR_OR_ACCVGPR 32 ADDR FMT_ANY OPR_VGPR 32 SADDR FMT_ANY OPR_SREG 32 FMT_NUM_B64 OPR_FLAT_SCRATCH 64 FMT_NUM_U8 OPR_GPUMEM 8 FMT_NUM_B32 OPR_SDST_M0 32 VMEM FLAT FALSE FALSE FALSE FALSE FALSE SCRATCH_LOAD_SBYTE_D16 Load 8 bits of signed data from the scratch aperture, sign extend to 16 bits and store the result into the low 16 bits of a 32-bit vector register. ENC_FLAT_SCRATCH default 34 VDST FMT_ANY OPR_VGPR_OR_ACCVGPR 32 ADDR FMT_ANY OPR_VGPR 32 SADDR FMT_ANY OPR_SREG 32 FMT_NUM_B64 OPR_FLAT_SCRATCH 64 FMT_NUM_I8 OPR_GPUMEM 8 FMT_NUM_B32 OPR_SDST_M0 32 VMEM FLAT FALSE FALSE FALSE FALSE FALSE SCRATCH_LOAD_SBYTE_D16_HI Load 8 bits of signed data from the scratch aperture, sign extend to 16 bits and store the result into the high 16 bits of a 32-bit vector register. ENC_FLAT_SCRATCH default 35 VDST FMT_ANY OPR_VGPR_OR_ACCVGPR 32 ADDR FMT_ANY OPR_VGPR 32 SADDR FMT_ANY OPR_SREG 32 FMT_NUM_B64 OPR_FLAT_SCRATCH 64 FMT_NUM_I8 OPR_GPUMEM 8 FMT_NUM_B32 OPR_SDST_M0 32 VMEM FLAT FALSE FALSE FALSE FALSE FALSE SCRATCH_LOAD_SHORT_D16 Load 16 bits of unsigned data from the scratch aperture and store the result into the low 16 bits of a 32-bit vector register. ENC_FLAT_SCRATCH default 36 VDST FMT_ANY OPR_VGPR_OR_ACCVGPR 32 ADDR FMT_ANY OPR_VGPR 32 SADDR FMT_ANY OPR_SREG 32 FMT_NUM_B64 OPR_FLAT_SCRATCH 64 FMT_NUM_B16 OPR_GPUMEM 16 FMT_NUM_B32 OPR_SDST_M0 32 VMEM FLAT FALSE FALSE FALSE FALSE FALSE SCRATCH_LOAD_SHORT_D16_HI Load 16 bits of unsigned data from the scratch aperture and store the result into the high 16 bits of a 32-bit vector register. ENC_FLAT_SCRATCH default 37 VDST FMT_ANY OPR_VGPR_OR_ACCVGPR 32 ADDR FMT_ANY OPR_VGPR 32 SADDR FMT_ANY OPR_SREG 32 FMT_NUM_B64 OPR_FLAT_SCRATCH 64 FMT_NUM_B16 OPR_GPUMEM 16 FMT_NUM_B32 OPR_SDST_M0 32 VMEM FLAT FALSE FALSE FALSE FALSE FALSE SCRATCH_LOAD_LDS_UBYTE Load 8 bits of untyped data from the scratch aperture, zero extend to 32 bits and store the result into a data share. ENC_FLAT_SCRATCH default 38 VDST FMT_ANY OPR_VGPR_OR_ACCVGPR 32 ADDR FMT_ANY OPR_VGPR 32 SADDR FMT_ANY OPR_SREG 32 FMT_NUM_B64 OPR_FLAT_SCRATCH 64 FMT_NUM_B32 OPR_SDST_M0 32 VMEM FLAT FALSE FALSE FALSE FALSE FALSE SCRATCH_LOAD_LDS_SBYTE Load 8 bits of untyped data from the scratch aperture, sign extend to 32 bits and store the result into a data share. ENC_FLAT_SCRATCH default 39 VDST FMT_ANY OPR_VGPR_OR_ACCVGPR 32 ADDR FMT_ANY OPR_VGPR 32 SADDR FMT_ANY OPR_SREG 32 FMT_NUM_B64 OPR_FLAT_SCRATCH 64 FMT_NUM_B32 OPR_SDST_M0 32 VMEM FLAT FALSE FALSE FALSE FALSE FALSE SCRATCH_LOAD_LDS_USHORT Load 16 bits of untyped data from the scratch aperture, zero extend to 32 bits and store the result into a data share. ENC_FLAT_SCRATCH default 40 VDST FMT_ANY OPR_VGPR_OR_ACCVGPR 32 ADDR FMT_ANY OPR_VGPR 32 SADDR FMT_ANY OPR_SREG 32 FMT_NUM_B64 OPR_FLAT_SCRATCH 64 FMT_NUM_B32 OPR_SDST_M0 32 VMEM FLAT FALSE FALSE FALSE FALSE FALSE SCRATCH_LOAD_LDS_SSHORT Load 16 bits of untyped data from the scratch aperture, sign extend to 32 bits and store the result into a data share. ENC_FLAT_SCRATCH default 41 VDST FMT_ANY OPR_VGPR_OR_ACCVGPR 32 ADDR FMT_ANY OPR_VGPR 32 SADDR FMT_ANY OPR_SREG 32 FMT_NUM_B64 OPR_FLAT_SCRATCH 64 FMT_NUM_B32 OPR_SDST_M0 32 VMEM FLAT FALSE FALSE FALSE FALSE FALSE SCRATCH_LOAD_LDS_DWORD Load 32 bits of untyped data from the scratch aperture and store the result into a data share. ENC_FLAT_SCRATCH default 42 VDST FMT_ANY OPR_VGPR_OR_ACCVGPR 32 ADDR FMT_ANY OPR_VGPR 32 SADDR FMT_ANY OPR_SREG 32 FMT_NUM_B64 OPR_FLAT_SCRATCH 64 FMT_NUM_B32 OPR_SDST_M0 32 VMEM FLAT FALSE FALSE FALSE FALSE FALSE TBUFFER_LOAD_FORMAT_X Load 1-component formatted data from a buffer surface, convert the data to 32 bit integral or floating point format, then store the result into a vector register. The instruction specifies the data format of the surface, overriding the resource descriptor. ENC_MTBUF default 0 VDATA FMT_ANY OPR_VGPR_OR_ACCVGPR 32 VADDR FMT_ANY OPR_VGPR 64 SRSRC FMT_RSRC_TYPED OPR_SREG 128 SOFFSET FMT_ANY OPR_SSRC_NOLIT 32 FMT_NUM_B32 OPR_GPUMEM 32 VMEM TEXTURE LOAD FALSE FALSE FALSE FALSE FALSE TBUFFER_LOAD_FORMAT_XY Load 2-component formatted data from a buffer surface, convert the data to 32 bit integral or floating point format, then store the result into a vector register. The instruction specifies the data format of the surface, overriding the resource descriptor. ENC_MTBUF default 1 VDATA FMT_ANY OPR_VGPR_OR_ACCVGPR 64 VADDR FMT_ANY OPR_VGPR 64 SRSRC FMT_RSRC_TYPED OPR_SREG 128 SOFFSET FMT_ANY OPR_SSRC_NOLIT 32 FMT_NUM_B64 OPR_GPUMEM 64 VMEM TEXTURE LOAD FALSE FALSE FALSE FALSE FALSE TBUFFER_LOAD_FORMAT_XYZ Load 3-component formatted data from a buffer surface, convert the data to 32 bit integral or floating point format, then store the result into a vector register. The instruction specifies the data format of the surface, overriding the resource descriptor. ENC_MTBUF default 2 VDATA FMT_ANY OPR_VGPR_OR_ACCVGPR 96 VADDR FMT_ANY OPR_VGPR 64 SRSRC FMT_RSRC_TYPED OPR_SREG 128 SOFFSET FMT_ANY OPR_SSRC_NOLIT 32 FMT_NUM_B96 OPR_GPUMEM 96 VMEM TEXTURE LOAD FALSE FALSE FALSE FALSE FALSE TBUFFER_LOAD_FORMAT_XYZW Load 4-component formatted data from a buffer surface, convert the data to 32 bit integral or floating point format, then store the result into a vector register. The instruction specifies the data format of the surface, overriding the resource descriptor. ENC_MTBUF default 3 VDATA FMT_ANY OPR_VGPR_OR_ACCVGPR 128 VADDR FMT_ANY OPR_VGPR 64 SRSRC FMT_RSRC_TYPED OPR_SREG 128 SOFFSET FMT_ANY OPR_SSRC_NOLIT 32 FMT_NUM_B128 OPR_GPUMEM 128 VMEM TEXTURE LOAD FALSE FALSE FALSE FALSE FALSE TBUFFER_STORE_FORMAT_X Convert 32 bits of data from vector input registers into 1-component formatted data and store the data into a buffer surface. The instruction specifies the data format of the surface, overriding the resource descriptor. ENC_MTBUF default 4 VDATA FMT_ANY OPR_VGPR_OR_ACCVGPR 32 VADDR FMT_ANY OPR_VGPR 64 SRSRC FMT_RSRC_TYPED OPR_SREG 128 SOFFSET FMT_ANY OPR_SSRC_NOLIT 32 FMT_NUM_B32 OPR_GPUMEM 32 VMEM TEXTURE STORE FALSE FALSE FALSE FALSE FALSE TBUFFER_STORE_FORMAT_XY Convert 64 bits of data from vector input registers into 2-component formatted data and store the data into a buffer surface. The instruction specifies the data format of the surface, overriding the resource descriptor. ENC_MTBUF default 5 VDATA FMT_ANY OPR_VGPR_OR_ACCVGPR 64 VADDR FMT_ANY OPR_VGPR 64 SRSRC FMT_RSRC_TYPED OPR_SREG 128 SOFFSET FMT_ANY OPR_SSRC_NOLIT 32 FMT_NUM_B64 OPR_GPUMEM 64 VMEM TEXTURE STORE FALSE FALSE FALSE FALSE FALSE TBUFFER_STORE_FORMAT_XYZ Convert 96 bits of data from vector input registers into 3-component formatted data and store the data into a buffer surface. The instruction specifies the data format of the surface, overriding the resource descriptor. ENC_MTBUF default 6 VDATA FMT_ANY OPR_VGPR_OR_ACCVGPR 96 VADDR FMT_ANY OPR_VGPR 64 SRSRC FMT_RSRC_TYPED OPR_SREG 128 SOFFSET FMT_ANY OPR_SSRC_NOLIT 32 FMT_NUM_B96 OPR_GPUMEM 96 VMEM TEXTURE STORE FALSE FALSE FALSE FALSE FALSE TBUFFER_STORE_FORMAT_XYZW Convert 128 bits of data from vector input registers into 4-component formatted data and store the data into a buffer surface. The instruction specifies the data format of the surface, overriding the resource descriptor. ENC_MTBUF default 7 VDATA FMT_ANY OPR_VGPR_OR_ACCVGPR 128 VADDR FMT_ANY OPR_VGPR 64 SRSRC FMT_RSRC_TYPED OPR_SREG 128 SOFFSET FMT_ANY OPR_SSRC_NOLIT 32 FMT_NUM_B128 OPR_GPUMEM 128 VMEM TEXTURE STORE FALSE FALSE FALSE FALSE FALSE TBUFFER_LOAD_FORMAT_D16_X Load 1-component formatted data from a buffer surface, convert the data to packed 16 bit integral or floating point format, then store the result into a vector register. The instruction specifies the data format of the surface, overriding the resource descriptor. ENC_MTBUF default 8 VDATA FMT_ANY OPR_VGPR_OR_ACCVGPR 32 VADDR FMT_ANY OPR_VGPR 64 SRSRC FMT_RSRC_TYPED OPR_SREG 128 SOFFSET FMT_ANY OPR_SSRC_NOLIT 32 FMT_NUM_B32 OPR_GPUMEM 32 VMEM TEXTURE LOAD FALSE FALSE FALSE FALSE FALSE TBUFFER_LOAD_FORMAT_D16_XY Load 2-component formatted data from a buffer surface, convert the data to packed 16 bit integral or floating point format, then store the result into a vector register. The instruction specifies the data format of the surface, overriding the resource descriptor. ENC_MTBUF default 9 VDATA FMT_ANY OPR_VGPR_OR_ACCVGPR 32 VADDR FMT_ANY OPR_VGPR 64 SRSRC FMT_RSRC_TYPED OPR_SREG 128 SOFFSET FMT_ANY OPR_SSRC_NOLIT 32 FMT_NUM_B64 OPR_GPUMEM 64 VMEM TEXTURE LOAD FALSE FALSE FALSE FALSE FALSE TBUFFER_LOAD_FORMAT_D16_XYZ Load 3-component formatted data from a buffer surface, convert the data to packed 16 bit integral or floating point format, then store the result into a vector register. The instruction specifies the data format of the surface, overriding the resource descriptor. ENC_MTBUF default 10 VDATA FMT_ANY OPR_VGPR_OR_ACCVGPR 64 VADDR FMT_ANY OPR_VGPR 64 SRSRC FMT_RSRC_TYPED OPR_SREG 128 SOFFSET FMT_ANY OPR_SSRC_NOLIT 32 FMT_NUM_B96 OPR_GPUMEM 96 VMEM TEXTURE LOAD FALSE FALSE FALSE FALSE FALSE TBUFFER_LOAD_FORMAT_D16_XYZW Load 4-component formatted data from a buffer surface, convert the data to packed 16 bit integral or floating point format, then store the result into a vector register. The instruction specifies the data format of the surface, overriding the resource descriptor. ENC_MTBUF default 11 VDATA FMT_ANY OPR_VGPR_OR_ACCVGPR 64 VADDR FMT_ANY OPR_VGPR 64 SRSRC FMT_RSRC_TYPED OPR_SREG 128 SOFFSET FMT_ANY OPR_SSRC_NOLIT 32 FMT_NUM_B128 OPR_GPUMEM 128 VMEM TEXTURE LOAD FALSE FALSE FALSE FALSE FALSE TBUFFER_STORE_FORMAT_D16_X Convert 16 bits of data from vector input registers into 1-component formatted data and store the data into a buffer surface. The instruction specifies the data format of the surface, overriding the resource descriptor. ENC_MTBUF default 12 VDATA FMT_ANY OPR_VGPR_OR_ACCVGPR 32 VADDR FMT_ANY OPR_VGPR 64 SRSRC FMT_RSRC_TYPED OPR_SREG 128 SOFFSET FMT_ANY OPR_SSRC_NOLIT 32 FMT_NUM_B32 OPR_GPUMEM 32 VMEM TEXTURE STORE FALSE FALSE FALSE FALSE FALSE TBUFFER_STORE_FORMAT_D16_XY Convert 32 bits of data from vector input registers into 2-component formatted data and store the data into a buffer surface. The instruction specifies the data format of the surface, overriding the resource descriptor. ENC_MTBUF default 13 VDATA FMT_ANY OPR_VGPR_OR_ACCVGPR 32 VADDR FMT_ANY OPR_VGPR 64 SRSRC FMT_RSRC_TYPED OPR_SREG 128 SOFFSET FMT_ANY OPR_SSRC_NOLIT 32 FMT_NUM_B64 OPR_GPUMEM 64 VMEM TEXTURE STORE FALSE FALSE FALSE FALSE FALSE TBUFFER_STORE_FORMAT_D16_XYZ Convert 48 bits of data from vector input registers into 3-component formatted data and store the data into a buffer surface. The instruction specifies the data format of the surface, overriding the resource descriptor. ENC_MTBUF default 14 VDATA FMT_ANY OPR_VGPR_OR_ACCVGPR 64 VADDR FMT_ANY OPR_VGPR 64 SRSRC FMT_RSRC_TYPED OPR_SREG 128 SOFFSET FMT_ANY OPR_SSRC_NOLIT 32 FMT_NUM_B96 OPR_GPUMEM 96 VMEM TEXTURE STORE FALSE FALSE FALSE FALSE FALSE TBUFFER_STORE_FORMAT_D16_XYZW Convert 64 bits of data from vector input registers into 4-component formatted data and store the data into a buffer surface. The instruction specifies the data format of the surface, overriding the resource descriptor. ENC_MTBUF default 15 VDATA FMT_ANY OPR_VGPR_OR_ACCVGPR 64 VADDR FMT_ANY OPR_VGPR 64 SRSRC FMT_RSRC_TYPED OPR_SREG 128 SOFFSET FMT_ANY OPR_SSRC_NOLIT 32 FMT_NUM_B128 OPR_GPUMEM 128 VMEM TEXTURE STORE FALSE FALSE FALSE FALSE FALSE BUFFER_LOAD_FORMAT_X Load 1-component formatted data from a buffer surface, convert the data to 32 bit integral or floating point format, then store the result into a vector register. The resource descriptor specifies the data format of the surface. ENC_MUBUF default 0 VDATA FMT_ANY OPR_VGPR_OR_ACCVGPR 32 VADDR FMT_ANY OPR_VGPR 64 SRSRC FMT_RSRC_TYPED OPR_SREG 128 SOFFSET FMT_ANY OPR_SSRC_NOLIT 32 FMT_NUM_B32 OPR_GPUMEM 32 VMEM TEXTURE LOAD FALSE FALSE FALSE FALSE FALSE BUFFER_LOAD_FORMAT_XY Load 2-component formatted data from a buffer surface, convert the data to 32 bit integral or floating point format, then store the result into a vector register. The resource descriptor specifies the data format of the surface. ENC_MUBUF default 1 VDATA FMT_ANY OPR_VGPR_OR_ACCVGPR 64 VADDR FMT_ANY OPR_VGPR 64 SRSRC FMT_RSRC_TYPED OPR_SREG 128 SOFFSET FMT_ANY OPR_SSRC_NOLIT 32 FMT_NUM_B64 OPR_GPUMEM 64 VMEM TEXTURE LOAD FALSE FALSE FALSE FALSE FALSE BUFFER_LOAD_FORMAT_XYZ Load 3-component formatted data from a buffer surface, convert the data to 32 bit integral or floating point format, then store the result into a vector register. The resource descriptor specifies the data format of the surface. ENC_MUBUF default 2 VDATA FMT_ANY OPR_VGPR_OR_ACCVGPR 96 VADDR FMT_ANY OPR_VGPR 64 SRSRC FMT_RSRC_TYPED OPR_SREG 128 SOFFSET FMT_ANY OPR_SSRC_NOLIT 32 FMT_NUM_B96 OPR_GPUMEM 96 VMEM TEXTURE LOAD FALSE FALSE FALSE FALSE FALSE BUFFER_LOAD_FORMAT_XYZW Load 4-component formatted data from a buffer surface, convert the data to 32 bit integral or floating point format, then store the result into a vector register. The resource descriptor specifies the data format of the surface. ENC_MUBUF default 3 VDATA FMT_ANY OPR_VGPR_OR_ACCVGPR 128 VADDR FMT_ANY OPR_VGPR 64 SRSRC FMT_RSRC_TYPED OPR_SREG 128 SOFFSET FMT_ANY OPR_SSRC_NOLIT 32 FMT_NUM_B128 OPR_GPUMEM 128 VMEM TEXTURE LOAD FALSE FALSE FALSE FALSE FALSE BUFFER_STORE_FORMAT_X Convert 32 bits of data from vector input registers into 1-component formatted data and store the data into a buffer surface. The instruction specifies the data format of the surface, overriding the resource descriptor. ENC_MUBUF default 4 VDATA FMT_ANY OPR_VGPR_OR_ACCVGPR 32 VADDR FMT_ANY OPR_VGPR 64 SRSRC FMT_RSRC_TYPED OPR_SREG 128 SOFFSET FMT_ANY OPR_SSRC_NOLIT 32 FMT_NUM_B32 OPR_GPUMEM 32 VMEM TEXTURE STORE FALSE FALSE FALSE FALSE FALSE BUFFER_STORE_FORMAT_XY Convert 64 bits of data from vector input registers into 2-component formatted data and store the data into a buffer surface. The instruction specifies the data format of the surface, overriding the resource descriptor. ENC_MUBUF default 5 VDATA FMT_ANY OPR_VGPR_OR_ACCVGPR 64 VADDR FMT_ANY OPR_VGPR 64 SRSRC FMT_RSRC_TYPED OPR_SREG 128 SOFFSET FMT_ANY OPR_SSRC_NOLIT 32 FMT_NUM_B64 OPR_GPUMEM 64 VMEM TEXTURE STORE FALSE FALSE FALSE FALSE FALSE BUFFER_STORE_FORMAT_XYZ Convert 96 bits of data from vector input registers into 3-component formatted data and store the data into a buffer surface. The instruction specifies the data format of the surface, overriding the resource descriptor. ENC_MUBUF default 6 VDATA FMT_ANY OPR_VGPR_OR_ACCVGPR 96 VADDR FMT_ANY OPR_VGPR 64 SRSRC FMT_RSRC_TYPED OPR_SREG 128 SOFFSET FMT_ANY OPR_SSRC_NOLIT 32 FMT_NUM_B96 OPR_GPUMEM 96 VMEM TEXTURE STORE FALSE FALSE FALSE FALSE FALSE BUFFER_STORE_FORMAT_XYZW Convert 128 bits of data from vector input registers into 4-component formatted data and store the data into a buffer surface. The instruction specifies the data format of the surface, overriding the resource descriptor. ENC_MUBUF default 7 VDATA FMT_ANY OPR_VGPR_OR_ACCVGPR 128 VADDR FMT_ANY OPR_VGPR 64 SRSRC FMT_RSRC_TYPED OPR_SREG 128 SOFFSET FMT_ANY OPR_SSRC_NOLIT 32 FMT_NUM_B128 OPR_GPUMEM 128 VMEM TEXTURE STORE FALSE FALSE FALSE FALSE FALSE BUFFER_LOAD_FORMAT_D16_X Load 1-component formatted data from a buffer surface, convert the data to packed 16 bit integral or floating point format, then store the result into the low 16 bits of a 32-bit vector register. The resource descriptor specifies the data format of the surface. ENC_MUBUF default 8 VDATA FMT_ANY OPR_VGPR_OR_ACCVGPR 32 VADDR FMT_ANY OPR_VGPR 64 SRSRC FMT_RSRC_TYPED OPR_SREG 128 SOFFSET FMT_ANY OPR_SSRC_NOLIT 32 FMT_NUM_B32 OPR_GPUMEM 32 VMEM TEXTURE LOAD FALSE FALSE FALSE FALSE FALSE BUFFER_LOAD_FORMAT_D16_XY Load 2-component formatted data from a buffer surface, convert the data to packed 16 bit integral or floating point format, then store the result into a vector register. The resource descriptor specifies the data format of the surface. ENC_MUBUF default 9 VDATA FMT_ANY OPR_VGPR_OR_ACCVGPR 32 VADDR FMT_ANY OPR_VGPR 64 SRSRC FMT_RSRC_TYPED OPR_SREG 128 SOFFSET FMT_ANY OPR_SSRC_NOLIT 32 FMT_NUM_B64 OPR_GPUMEM 64 VMEM TEXTURE LOAD FALSE FALSE FALSE FALSE FALSE BUFFER_LOAD_FORMAT_D16_XYZ Load 3-component formatted data from a buffer surface, convert the data to packed 16 bit integral or floating point format, then store the result into a vector register. The resource descriptor specifies the data format of the surface. ENC_MUBUF default 10 VDATA FMT_ANY OPR_VGPR_OR_ACCVGPR 64 VADDR FMT_ANY OPR_VGPR 64 SRSRC FMT_RSRC_TYPED OPR_SREG 128 SOFFSET FMT_ANY OPR_SSRC_NOLIT 32 FMT_NUM_B96 OPR_GPUMEM 96 VMEM TEXTURE LOAD FALSE FALSE FALSE FALSE FALSE BUFFER_LOAD_FORMAT_D16_XYZW Load 4-component formatted data from a buffer surface, convert the data to packed 16 bit integral or floating point format, then store the result into a vector register. The resource descriptor specifies the data format of the surface. ENC_MUBUF default 11 VDATA FMT_ANY OPR_VGPR_OR_ACCVGPR 64 VADDR FMT_ANY OPR_VGPR 64 SRSRC FMT_RSRC_TYPED OPR_SREG 128 SOFFSET FMT_ANY OPR_SSRC_NOLIT 32 FMT_NUM_B128 OPR_GPUMEM 128 VMEM TEXTURE LOAD FALSE FALSE FALSE FALSE FALSE BUFFER_STORE_FORMAT_D16_X Convert 16 bits of data from the low 16 bits of a 32-bit vector input register into 1-component formatted data and store the data into a buffer surface. The instruction specifies the data format of the surface, overriding the resource descriptor. ENC_MUBUF default 12 VDATA FMT_ANY OPR_VGPR_OR_ACCVGPR 32 VADDR FMT_ANY OPR_VGPR 64 SRSRC FMT_RSRC_TYPED OPR_SREG 128 SOFFSET FMT_ANY OPR_SSRC_NOLIT 32 FMT_NUM_B32 OPR_GPUMEM 32 VMEM TEXTURE STORE FALSE FALSE FALSE FALSE FALSE BUFFER_STORE_FORMAT_D16_XY Convert 32 bits of data from vector input registers into 2-component formatted data and store the data into a buffer surface. The instruction specifies the data format of the surface, overriding the resource descriptor. ENC_MUBUF default 13 VDATA FMT_ANY OPR_VGPR_OR_ACCVGPR 32 VADDR FMT_ANY OPR_VGPR 64 SRSRC FMT_RSRC_TYPED OPR_SREG 128 SOFFSET FMT_ANY OPR_SSRC_NOLIT 32 FMT_NUM_B64 OPR_GPUMEM 64 VMEM TEXTURE STORE FALSE FALSE FALSE FALSE FALSE BUFFER_STORE_FORMAT_D16_XYZ Convert 48 bits of data from vector input registers into 3-component formatted data and store the data into a buffer surface. The instruction specifies the data format of the surface, overriding the resource descriptor. ENC_MUBUF default 14 VDATA FMT_ANY OPR_VGPR_OR_ACCVGPR 64 VADDR FMT_ANY OPR_VGPR 64 SRSRC FMT_RSRC_TYPED OPR_SREG 128 SOFFSET FMT_ANY OPR_SSRC_NOLIT 32 FMT_NUM_B96 OPR_GPUMEM 96 VMEM TEXTURE STORE FALSE FALSE FALSE FALSE FALSE BUFFER_STORE_FORMAT_D16_XYZW Convert 64 bits of data from vector input registers into 4-component formatted data and store the data into a buffer surface. The instruction specifies the data format of the surface, overriding the resource descriptor. ENC_MUBUF default 15 VDATA FMT_ANY OPR_VGPR_OR_ACCVGPR 64 VADDR FMT_ANY OPR_VGPR 64 SRSRC FMT_RSRC_TYPED OPR_SREG 128 SOFFSET FMT_ANY OPR_SSRC_NOLIT 32 FMT_NUM_B128 OPR_GPUMEM 128 VMEM TEXTURE STORE FALSE FALSE FALSE FALSE FALSE BUFFER_LOAD_UBYTE Load 8 bits of unsigned data from a buffer surface, zero extend to 32 bits and store the result into a vector register. ENC_MUBUF default 16 VDATA FMT_ANY OPR_VGPR_OR_ACCVGPR 32 VADDR FMT_ANY OPR_VGPR 64 SRSRC FMT_RSRC_SCRATCH OPR_SREG 128 SOFFSET FMT_ANY OPR_SSRC_NOLIT 32 FMT_NUM_U8 OPR_GPUMEM 8 VMEM TEXTURE FALSE FALSE FALSE FALSE FALSE BUFFER_LOAD_SBYTE Load 8 bits of signed data from a buffer surface, sign extend to 32 bits and store the result into a vector register. ENC_MUBUF default 17 VDATA FMT_ANY OPR_VGPR_OR_ACCVGPR 32 VADDR FMT_ANY OPR_VGPR 64 SRSRC FMT_RSRC_SCRATCH OPR_SREG 128 SOFFSET FMT_ANY OPR_SSRC_NOLIT 32 FMT_NUM_I8 OPR_GPUMEM 8 VMEM TEXTURE FALSE FALSE FALSE FALSE FALSE BUFFER_LOAD_USHORT Load 16 bits of unsigned data from a buffer surface, zero extend to 32 bits and store the result into a vector register. ENC_MUBUF default 18 VDATA FMT_ANY OPR_VGPR_OR_ACCVGPR 32 VADDR FMT_ANY OPR_VGPR 64 SRSRC FMT_RSRC_SCRATCH OPR_SREG 128 SOFFSET FMT_ANY OPR_SSRC_NOLIT 32 FMT_NUM_U16 OPR_GPUMEM 16 VMEM TEXTURE FALSE FALSE FALSE FALSE FALSE BUFFER_LOAD_SSHORT Load 16 bits of signed data from a buffer surface, sign extend to 32 bits and store the result into a vector register. ENC_MUBUF default 19 VDATA FMT_ANY OPR_VGPR_OR_ACCVGPR 32 VADDR FMT_ANY OPR_VGPR 64 SRSRC FMT_RSRC_SCRATCH OPR_SREG 128 SOFFSET FMT_ANY OPR_SSRC_NOLIT 32 FMT_NUM_I16 OPR_GPUMEM 16 VMEM TEXTURE FALSE FALSE FALSE FALSE FALSE BUFFER_LOAD_DWORD Load 32 bits of data from a buffer surface into a vector register. ENC_MUBUF default 20 VDATA FMT_ANY OPR_VGPR_OR_ACCVGPR 32 VADDR FMT_ANY OPR_VGPR 64 SRSRC FMT_RSRC_SCRATCH OPR_SREG 128 SOFFSET FMT_ANY OPR_SSRC_NOLIT 32 FMT_NUM_B32 OPR_GPUMEM 32 VMEM TEXTURE FALSE FALSE FALSE FALSE FALSE BUFFER_LOAD_DWORDX2 Load 64 bits of data from a buffer surface into a vector register. ENC_MUBUF default 21 VDATA FMT_ANY OPR_VGPR_OR_ACCVGPR 64 VADDR FMT_ANY OPR_VGPR 64 SRSRC FMT_RSRC_SCRATCH OPR_SREG 128 SOFFSET FMT_ANY OPR_SSRC_NOLIT 32 FMT_NUM_B64 OPR_GPUMEM 64 VMEM TEXTURE FALSE FALSE FALSE FALSE FALSE BUFFER_LOAD_DWORDX3 Load 96 bits of data from a buffer surface into a vector register. ENC_MUBUF default 22 VDATA FMT_ANY OPR_VGPR_OR_ACCVGPR 96 VADDR FMT_ANY OPR_VGPR 64 SRSRC FMT_RSRC_SCRATCH OPR_SREG 128 SOFFSET FMT_ANY OPR_SSRC_NOLIT 32 FMT_NUM_B96 OPR_GPUMEM 96 VMEM TEXTURE FALSE FALSE FALSE FALSE FALSE BUFFER_LOAD_DWORDX4 Load 128 bits of data from a buffer surface into a vector register. ENC_MUBUF default 23 VDATA FMT_ANY OPR_VGPR_OR_ACCVGPR 128 VADDR FMT_ANY OPR_VGPR 64 SRSRC FMT_RSRC_SCRATCH OPR_SREG 128 SOFFSET FMT_ANY OPR_SSRC_NOLIT 32 FMT_NUM_B128 OPR_GPUMEM 128 VMEM TEXTURE FALSE FALSE FALSE FALSE FALSE BUFFER_STORE_BYTE Store 8 bits of data from a vector register into a buffer surface. ENC_MUBUF default 24 VDATA FMT_ANY OPR_VGPR_OR_ACCVGPR 32 VADDR FMT_ANY OPR_VGPR 64 SRSRC FMT_RSRC_SCRATCH OPR_SREG 128 SOFFSET FMT_ANY OPR_SSRC_NOLIT 32 FMT_NUM_B8 OPR_GPUMEM 8 VMEM TEXTURE FALSE FALSE FALSE FALSE FALSE BUFFER_STORE_BYTE_D16_HI Store 8 bits of data from the high 16 bits of a 32-bit vector register into a buffer surface. ENC_MUBUF default 25 VDATA FMT_ANY OPR_VGPR_OR_ACCVGPR 32 VADDR FMT_ANY OPR_VGPR 64 SRSRC FMT_RSRC_SCRATCH OPR_SREG 128 SOFFSET FMT_ANY OPR_SSRC_NOLIT 32 FMT_NUM_B8 OPR_GPUMEM 8 VMEM TEXTURE FALSE FALSE FALSE FALSE FALSE BUFFER_STORE_SHORT Store 16 bits of data from a vector register into a buffer surface. ENC_MUBUF default 26 VDATA FMT_ANY OPR_VGPR_OR_ACCVGPR 32 VADDR FMT_ANY OPR_VGPR 64 SRSRC FMT_RSRC_SCRATCH OPR_SREG 128 SOFFSET FMT_ANY OPR_SSRC_NOLIT 32 FMT_NUM_B16 OPR_GPUMEM 16 VMEM TEXTURE FALSE FALSE FALSE FALSE FALSE BUFFER_STORE_SHORT_D16_HI Store 16 bits of data from the high 16 bits of a 32-bit vector register into a buffer surface. ENC_MUBUF default 27 VDATA FMT_ANY OPR_VGPR_OR_ACCVGPR 32 VADDR FMT_ANY OPR_VGPR 64 SRSRC FMT_RSRC_SCRATCH OPR_SREG 128 SOFFSET FMT_ANY OPR_SSRC_NOLIT 32 FMT_NUM_B16 OPR_GPUMEM 16 VMEM TEXTURE FALSE FALSE FALSE FALSE FALSE BUFFER_STORE_DWORD Store 32 bits of data from vector input registers into a buffer surface. ENC_MUBUF default 28 VDATA FMT_ANY OPR_VGPR_OR_ACCVGPR 32 VADDR FMT_ANY OPR_VGPR 64 SRSRC FMT_RSRC_SCRATCH OPR_SREG 128 SOFFSET FMT_ANY OPR_SSRC_NOLIT 32 FMT_NUM_B32 OPR_GPUMEM 32 VMEM TEXTURE FALSE FALSE FALSE FALSE FALSE BUFFER_STORE_DWORDX2 Store 64 bits of data from vector input registers into a buffer surface. ENC_MUBUF default 29 VDATA FMT_ANY OPR_VGPR_OR_ACCVGPR 64 VADDR FMT_ANY OPR_VGPR 64 SRSRC FMT_RSRC_SCRATCH OPR_SREG 128 SOFFSET FMT_ANY OPR_SSRC_NOLIT 32 FMT_NUM_B64 OPR_GPUMEM 64 VMEM TEXTURE FALSE FALSE FALSE FALSE FALSE BUFFER_STORE_DWORDX3 Store 96 bits of data from vector input registers into a buffer surface. ENC_MUBUF default 30 VDATA FMT_ANY OPR_VGPR_OR_ACCVGPR 96 VADDR FMT_ANY OPR_VGPR 64 SRSRC FMT_RSRC_SCRATCH OPR_SREG 128 SOFFSET FMT_ANY OPR_SSRC_NOLIT 32 FMT_NUM_B96 OPR_GPUMEM 96 VMEM TEXTURE FALSE FALSE FALSE FALSE FALSE BUFFER_STORE_DWORDX4 Store 128 bits of data from vector input registers into a buffer surface. ENC_MUBUF default 31 VDATA FMT_ANY OPR_VGPR_OR_ACCVGPR 128 VADDR FMT_ANY OPR_VGPR 64 SRSRC FMT_RSRC_SCRATCH OPR_SREG 128 SOFFSET FMT_ANY OPR_SSRC_NOLIT 32 FMT_NUM_B128 OPR_GPUMEM 128 VMEM TEXTURE FALSE FALSE FALSE FALSE FALSE BUFFER_LOAD_UBYTE_D16 Load 8 bits of unsigned data from a buffer surface, zero extend to 16 bits and store the result into the low 16 bits of a 32-bit vector register. ENC_MUBUF default 32 VDATA FMT_ANY OPR_VGPR_OR_ACCVGPR 32 VADDR FMT_ANY OPR_VGPR 64 SRSRC FMT_RSRC_SCRATCH OPR_SREG 128 SOFFSET FMT_ANY OPR_SSRC_NOLIT 32 FMT_NUM_U8 OPR_GPUMEM 8 VMEM TEXTURE FALSE FALSE FALSE FALSE FALSE BUFFER_LOAD_UBYTE_D16_HI Load 8 bits of unsigned data from a buffer surface, zero extend to 16 bits and store the result into the high 16 bits of a 32-bit vector register. ENC_MUBUF default 33 VDATA FMT_ANY OPR_VGPR_OR_ACCVGPR 32 VADDR FMT_ANY OPR_VGPR 64 SRSRC FMT_RSRC_SCRATCH OPR_SREG 128 SOFFSET FMT_ANY OPR_SSRC_NOLIT 32 FMT_NUM_U8 OPR_GPUMEM 8 VMEM TEXTURE FALSE FALSE FALSE FALSE FALSE BUFFER_LOAD_SBYTE_D16 Load 8 bits of signed data from a buffer surface, sign extend to 16 bits and store the result into the low 16 bits of a 32-bit vector register. ENC_MUBUF default 34 VDATA FMT_ANY OPR_VGPR_OR_ACCVGPR 32 VADDR FMT_ANY OPR_VGPR 64 SRSRC FMT_RSRC_SCRATCH OPR_SREG 128 SOFFSET FMT_ANY OPR_SSRC_NOLIT 32 FMT_NUM_I8 OPR_GPUMEM 8 VMEM TEXTURE FALSE FALSE FALSE FALSE FALSE BUFFER_LOAD_SBYTE_D16_HI Load 8 bits of signed data from a buffer surface, sign extend to 16 bits and store the result into the high 16 bits of a 32-bit vector register. ENC_MUBUF default 35 VDATA FMT_ANY OPR_VGPR_OR_ACCVGPR 32 VADDR FMT_ANY OPR_VGPR 64 SRSRC FMT_RSRC_SCRATCH OPR_SREG 128 SOFFSET FMT_ANY OPR_SSRC_NOLIT 32 FMT_NUM_I8 OPR_GPUMEM 8 VMEM TEXTURE FALSE FALSE FALSE FALSE FALSE BUFFER_LOAD_SHORT_D16 Load 16 bits of unsigned data from a buffer surface and store the result into the low 16 bits of a 32-bit vector register. ENC_MUBUF default 36 VDATA FMT_ANY OPR_VGPR_OR_ACCVGPR 32 VADDR FMT_ANY OPR_VGPR 64 SRSRC FMT_RSRC_SCRATCH OPR_SREG 128 SOFFSET FMT_ANY OPR_SSRC_NOLIT 32 FMT_NUM_B16 OPR_GPUMEM 16 VMEM TEXTURE FALSE FALSE FALSE FALSE FALSE BUFFER_LOAD_SHORT_D16_HI Load 16 bits of unsigned data from a buffer surface and store the result into the high 16 bits of a 32-bit vector register. ENC_MUBUF default 37 VDATA FMT_ANY OPR_VGPR_OR_ACCVGPR 32 VADDR FMT_ANY OPR_VGPR 64 SRSRC FMT_RSRC_SCRATCH OPR_SREG 128 SOFFSET FMT_ANY OPR_SSRC_NOLIT 32 FMT_NUM_B16 OPR_GPUMEM 16 VMEM TEXTURE FALSE FALSE FALSE FALSE FALSE BUFFER_LOAD_FORMAT_D16_HI_X Load 1-component formatted data from a buffer surface, convert the data to packed 16 bit integral or floating point format, then store the result into the high 16 bits of a 32-bit vector register. The resource descriptor specifies the data format of the surface. ENC_MUBUF default 38 VDATA FMT_ANY OPR_VGPR_OR_ACCVGPR 32 VADDR FMT_ANY OPR_VGPR 64 SRSRC FMT_RSRC_TYPED OPR_SREG 128 SOFFSET FMT_ANY OPR_SSRC_NOLIT 32 FMT_NUM_B32 OPR_GPUMEM 32 VMEM TEXTURE LOAD FALSE FALSE FALSE FALSE FALSE BUFFER_STORE_FORMAT_D16_HI_X Convert 16 bits of data from the high 16 bits of a 32-bit vector input register into 1-component formatted data and store the data into a buffer surface. The instruction specifies the data format of the surface, overriding the resource descriptor. ENC_MUBUF default 39 VDATA FMT_ANY OPR_VGPR_OR_ACCVGPR 32 VADDR FMT_ANY OPR_VGPR 64 SRSRC FMT_RSRC_TYPED OPR_SREG 128 SOFFSET FMT_ANY OPR_SSRC_NOLIT 32 FMT_NUM_B32 OPR_GPUMEM 32 VMEM TEXTURE STORE FALSE FALSE FALSE FALSE FALSE BUFFER_WBL2 Write back L2 cache. Returns ACK to shader. ENC_MUBUF default 40 VMEM TEXTURE FALSE FALSE FALSE FALSE FALSE BUFFER_INV BUFFER_INVL2 Invalidate CU and/or L2 cache depending on sc0 and sc1 bits. Returns ACK to shader. ENC_MUBUF default 41 VMEM TEXTURE FALSE FALSE FALSE FALSE FALSE BUFFER_ATOMIC_SWAP Swap an unsigned 32-bit integer value in the data register with a location in a buffer surface. Store the original value from buffer surface into a vector register iff the SC0 bit is set. ENC_MUBUF default 64 VDATA FMT_ANY OPR_VGPR_OR_ACCVGPR 32 VADDR FMT_ANY OPR_VGPR 64 SRSRC FMT_RSRC_SCRATCH OPR_SREG 128 SOFFSET FMT_ANY OPR_SSRC_NOLIT 32 FMT_NUM_B32 OPR_GPUMEM 32 FMT_NUM_B32 OPR_GPUMEM 32 VMEM TEXTURE ATOMIC FALSE FALSE FALSE FALSE FALSE BUFFER_ATOMIC_CMPSWAP Compare two unsigned 32-bit integer values stored in the data comparison register and a location in a buffer surface. Modify the memory location with a value in the data source register iff the comparison is equal. Store the original value from buffer surface into a vector register iff the SC0 bit is set. ENC_MUBUF default 65 VDATA FMT_ANY OPR_VGPR_OR_ACCVGPR 64 VADDR FMT_ANY OPR_VGPR 64 SRSRC FMT_RSRC_SCRATCH OPR_SREG 128 SOFFSET FMT_ANY OPR_SSRC_NOLIT 32 FMT_NUM_U32 OPR_GPUMEM 32 FMT_NUM_U32 OPR_GPUMEM 32 VMEM TEXTURE ATOMIC FALSE FALSE FALSE FALSE FALSE BUFFER_ATOMIC_ADD Add two unsigned 32-bit integer values stored in the data register and a location in a buffer surface. Store the original value from buffer surface into a vector register iff the SC0 bit is set. ENC_MUBUF default 66 VDATA FMT_ANY OPR_VGPR_OR_ACCVGPR 32 VADDR FMT_ANY OPR_VGPR 64 SRSRC FMT_RSRC_SCRATCH OPR_SREG 128 SOFFSET FMT_ANY OPR_SSRC_NOLIT 32 FMT_NUM_U32 OPR_GPUMEM 32 FMT_NUM_U32 OPR_GPUMEM 32 VMEM TEXTURE ATOMIC FALSE FALSE FALSE FALSE FALSE BUFFER_ATOMIC_SUB Subtract an unsigned 32-bit integer value stored in the data register from a value stored in a location in a buffer surface. Store the original value from buffer surface into a vector register iff the SC0 bit is set. ENC_MUBUF default 67 VDATA FMT_ANY OPR_VGPR_OR_ACCVGPR 32 VADDR FMT_ANY OPR_VGPR 64 SRSRC FMT_RSRC_SCRATCH OPR_SREG 128 SOFFSET FMT_ANY OPR_SSRC_NOLIT 32 FMT_NUM_U32 OPR_GPUMEM 32 FMT_NUM_U32 OPR_GPUMEM 32 VMEM TEXTURE ATOMIC FALSE FALSE FALSE FALSE FALSE BUFFER_ATOMIC_SMIN Select the minimum of two signed 32-bit integer inputs, given two values stored in the data register and a location in a buffer surface. Update the buffer surface with the selected value. Store the original value from buffer surface into a vector register iff the SC0 bit is set. ENC_MUBUF default 68 VDATA FMT_ANY OPR_VGPR_OR_ACCVGPR 32 VADDR FMT_ANY OPR_VGPR 64 SRSRC FMT_RSRC_SCRATCH OPR_SREG 128 SOFFSET FMT_ANY OPR_SSRC_NOLIT 32 FMT_NUM_I32 OPR_GPUMEM 32 FMT_NUM_I32 OPR_GPUMEM 32 VMEM TEXTURE ATOMIC FALSE FALSE FALSE FALSE FALSE BUFFER_ATOMIC_UMIN Select the minimum of two unsigned 32-bit integer inputs, given two values stored in the data register and a location in a buffer surface. Update the buffer surface with the selected value. Store the original value from buffer surface into a vector register iff the SC0 bit is set. ENC_MUBUF default 69 VDATA FMT_ANY OPR_VGPR_OR_ACCVGPR 32 VADDR FMT_ANY OPR_VGPR 64 SRSRC FMT_RSRC_SCRATCH OPR_SREG 128 SOFFSET FMT_ANY OPR_SSRC_NOLIT 32 FMT_NUM_U32 OPR_GPUMEM 32 FMT_NUM_U32 OPR_GPUMEM 32 VMEM TEXTURE ATOMIC FALSE FALSE FALSE FALSE FALSE BUFFER_ATOMIC_SMAX Select the maximum of two signed 32-bit integer inputs, given two values stored in the data register and a location in a buffer surface. Update the buffer surface with the selected value. Store the original value from buffer surface into a vector register iff the SC0 bit is set. ENC_MUBUF default 70 VDATA FMT_ANY OPR_VGPR_OR_ACCVGPR 32 VADDR FMT_ANY OPR_VGPR 64 SRSRC FMT_RSRC_SCRATCH OPR_SREG 128 SOFFSET FMT_ANY OPR_SSRC_NOLIT 32 FMT_NUM_I32 OPR_GPUMEM 32 FMT_NUM_I32 OPR_GPUMEM 32 VMEM TEXTURE ATOMIC FALSE FALSE FALSE FALSE FALSE BUFFER_ATOMIC_UMAX Select the maximum of two unsigned 32-bit integer inputs, given two values stored in the data register and a location in a buffer surface. Update the buffer surface with the selected value. Store the original value from buffer surface into a vector register iff the SC0 bit is set. ENC_MUBUF default 71 VDATA FMT_ANY OPR_VGPR_OR_ACCVGPR 32 VADDR FMT_ANY OPR_VGPR 64 SRSRC FMT_RSRC_SCRATCH OPR_SREG 128 SOFFSET FMT_ANY OPR_SSRC_NOLIT 32 FMT_NUM_U32 OPR_GPUMEM 32 FMT_NUM_U32 OPR_GPUMEM 32 VMEM TEXTURE ATOMIC FALSE FALSE FALSE FALSE FALSE BUFFER_ATOMIC_AND Calculate bitwise AND given two unsigned 32-bit integer values stored in the data register and a location in a buffer surface. Store the original value from buffer surface into a vector register iff the SC0 bit is set. ENC_MUBUF default 72 VDATA FMT_ANY OPR_VGPR_OR_ACCVGPR 32 VADDR FMT_ANY OPR_VGPR 64 SRSRC FMT_RSRC_SCRATCH OPR_SREG 128 SOFFSET FMT_ANY OPR_SSRC_NOLIT 32 FMT_NUM_B32 OPR_GPUMEM 32 FMT_NUM_B32 OPR_GPUMEM 32 VMEM TEXTURE ATOMIC FALSE FALSE FALSE FALSE FALSE BUFFER_ATOMIC_OR Calculate bitwise OR given two unsigned 32-bit integer values stored in the data register and a location in a buffer surface. Store the original value from buffer surface into a vector register iff the SC0 bit is set. ENC_MUBUF default 73 VDATA FMT_ANY OPR_VGPR_OR_ACCVGPR 32 VADDR FMT_ANY OPR_VGPR 64 SRSRC FMT_RSRC_SCRATCH OPR_SREG 128 SOFFSET FMT_ANY OPR_SSRC_NOLIT 32 FMT_NUM_B32 OPR_GPUMEM 32 FMT_NUM_B32 OPR_GPUMEM 32 VMEM TEXTURE ATOMIC FALSE FALSE FALSE FALSE FALSE BUFFER_ATOMIC_XOR Calculate bitwise XOR given two unsigned 32-bit integer values stored in the data register and a location in a buffer surface. Store the original value from buffer surface into a vector register iff the SC0 bit is set. ENC_MUBUF default 74 VDATA FMT_ANY OPR_VGPR_OR_ACCVGPR 32 VADDR FMT_ANY OPR_VGPR 64 SRSRC FMT_RSRC_SCRATCH OPR_SREG 128 SOFFSET FMT_ANY OPR_SSRC_NOLIT 32 FMT_NUM_B32 OPR_GPUMEM 32 FMT_NUM_B32 OPR_GPUMEM 32 VMEM TEXTURE ATOMIC FALSE FALSE FALSE FALSE FALSE BUFFER_ATOMIC_INC Increment an unsigned 32-bit integer value from a location in a buffer surface with wraparound to 0 if the value exceeds a value in the data register. Store the original value from buffer surface into a vector register iff the SC0 bit is set. ENC_MUBUF default 75 VDATA FMT_ANY OPR_VGPR_OR_ACCVGPR 32 VADDR FMT_ANY OPR_VGPR 64 SRSRC FMT_RSRC_SCRATCH OPR_SREG 128 SOFFSET FMT_ANY OPR_SSRC_NOLIT 32 FMT_NUM_U32 OPR_GPUMEM 32 FMT_NUM_U32 OPR_GPUMEM 32 VMEM TEXTURE ATOMIC FALSE FALSE FALSE FALSE FALSE BUFFER_ATOMIC_DEC Decrement an unsigned 32-bit integer value from a location in a buffer surface with wraparound to a value in the data register if the decrement yields a negative value. Store the original value from buffer surface into a vector register iff the SC0 bit is set. ENC_MUBUF default 76 VDATA FMT_ANY OPR_VGPR_OR_ACCVGPR 32 VADDR FMT_ANY OPR_VGPR 64 SRSRC FMT_RSRC_SCRATCH OPR_SREG 128 SOFFSET FMT_ANY OPR_SSRC_NOLIT 32 FMT_NUM_U32 OPR_GPUMEM 32 FMT_NUM_U32 OPR_GPUMEM 32 VMEM TEXTURE ATOMIC FALSE FALSE FALSE FALSE FALSE BUFFER_ATOMIC_ADD_F32 Add a single-precision float value in the data register to a location in a buffer surface. Store the original value from buffer surface into a vector register iff the SC0 bit is set. ENC_MUBUF default 77 VDATA FMT_ANY OPR_VGPR_OR_ACCVGPR 32 VADDR FMT_ANY OPR_VGPR 64 SRSRC FMT_RSRC_SCRATCH OPR_SREG 128 SOFFSET FMT_ANY OPR_SSRC_NOLIT 32 FMT_NUM_F32 OPR_GPUMEM 32 FMT_NUM_F32 OPR_GPUMEM 32 VMEM TEXTURE ATOMIC FALSE FALSE FALSE FALSE FALSE BUFFER_ATOMIC_PK_ADD_F16 Add a packed 2-component half-precision float value in the data register to a location in a buffer surface. Store the original value from buffer surface into a vector register iff the SC0 bit is set. ENC_MUBUF default 78 VDATA FMT_ANY OPR_VGPR_OR_ACCVGPR 32 VADDR FMT_ANY OPR_VGPR 64 SRSRC FMT_RSRC_SCRATCH OPR_SREG 128 SOFFSET FMT_ANY OPR_SSRC_NOLIT 32 FMT_NUM_PK2_F16 OPR_GPUMEM 32 FMT_NUM_PK2_F16 OPR_GPUMEM 32 VMEM TEXTURE ATOMIC FALSE FALSE FALSE FALSE FALSE BUFFER_ATOMIC_ADD_F64 Add a double-precision float value in the data register to a location in a buffer surface. Store the original value from buffer surface into a vector register iff the SC0 bit is set. ENC_MUBUF default 79 VDATA FMT_ANY OPR_VGPR_OR_ACCVGPR 64 VADDR FMT_ANY OPR_VGPR 64 SRSRC FMT_RSRC_SCRATCH OPR_SREG 128 SOFFSET FMT_ANY OPR_SSRC_NOLIT 32 FMT_NUM_F64 OPR_GPUMEM 64 FMT_NUM_F64 OPR_GPUMEM 64 VMEM TEXTURE ATOMIC FALSE FALSE FALSE FALSE FALSE BUFFER_ATOMIC_MIN_F64 Select the minimum of two double-precision float inputs, given two values stored in the data register and a location in a buffer surface. Update the buffer surface with the selected value. Store the original value from buffer surface into a vector register iff the SC0 bit is set. ENC_MUBUF default 80 VDATA FMT_ANY OPR_VGPR_OR_ACCVGPR 64 VADDR FMT_ANY OPR_VGPR 64 SRSRC FMT_RSRC_SCRATCH OPR_SREG 128 SOFFSET FMT_ANY OPR_SSRC_NOLIT 32 FMT_NUM_F64 OPR_GPUMEM 64 FMT_NUM_F64 OPR_GPUMEM 64 VMEM TEXTURE ATOMIC FALSE FALSE FALSE FALSE FALSE BUFFER_ATOMIC_MAX_F64 Select the maximum of two double-precision float inputs, given two values stored in the data register and a location in a buffer surface. Update the buffer surface with the selected value. Store the original value from buffer surface into a vector register iff the SC0 bit is set. ENC_MUBUF default 81 VDATA FMT_ANY OPR_VGPR_OR_ACCVGPR 64 VADDR FMT_ANY OPR_VGPR 64 SRSRC FMT_RSRC_SCRATCH OPR_SREG 128 SOFFSET FMT_ANY OPR_SSRC_NOLIT 32 FMT_NUM_F64 OPR_GPUMEM 64 FMT_NUM_F64 OPR_GPUMEM 64 VMEM TEXTURE ATOMIC FALSE FALSE FALSE FALSE FALSE BUFFER_ATOMIC_PK_ADD_BF16 Add a packed 2-component BF16 float value in the data register to a location in a buffer surface. Store the original value from buffer surface into a vector register iff the SC0 bit is set. ENC_MUBUF default 82 VDATA FMT_ANY OPR_VGPR_OR_ACCVGPR 32 VADDR FMT_ANY OPR_VGPR 64 SRSRC FMT_RSRC_SCRATCH OPR_SREG 128 SOFFSET FMT_ANY OPR_SSRC_NOLIT 32 FMT_NUM_PK2_BF16 OPR_GPUMEM 32 FMT_NUM_PK2_BF16 OPR_GPUMEM 32 VMEM TEXTURE ATOMIC FALSE FALSE FALSE FALSE FALSE BUFFER_ATOMIC_SWAP_X2 Swap an unsigned 64-bit integer value in the data register with a location in a buffer surface. Store the original value from buffer surface into a vector register iff the SC0 bit is set. ENC_MUBUF default 96 VDATA FMT_ANY OPR_VGPR_OR_ACCVGPR 64 VADDR FMT_ANY OPR_VGPR 64 SRSRC FMT_RSRC_SCRATCH OPR_SREG 128 SOFFSET FMT_ANY OPR_SSRC_NOLIT 32 FMT_NUM_B64 OPR_GPUMEM 64 FMT_NUM_B64 OPR_GPUMEM 64 VMEM TEXTURE ATOMIC FALSE FALSE FALSE FALSE FALSE BUFFER_ATOMIC_CMPSWAP_X2 Compare two unsigned 64-bit integer values stored in the data comparison register and a location in a buffer surface. Modify the memory location with a value in the data source register iff the comparison is equal. Store the original value from buffer surface into a vector register iff the SC0 bit is set. ENC_MUBUF default 97 VDATA FMT_ANY OPR_VGPR_OR_ACCVGPR 128 VADDR FMT_ANY OPR_VGPR 64 SRSRC FMT_RSRC_SCRATCH OPR_SREG 128 SOFFSET FMT_ANY OPR_SSRC_NOLIT 32 FMT_NUM_U64 OPR_GPUMEM 64 FMT_NUM_U64 OPR_GPUMEM 64 VMEM TEXTURE ATOMIC FALSE FALSE FALSE FALSE FALSE BUFFER_ATOMIC_ADD_X2 Add two unsigned 64-bit integer values stored in the data register and a location in a buffer surface. Store the original value from buffer surface into a vector register iff the SC0 bit is set. ENC_MUBUF default 98 VDATA FMT_ANY OPR_VGPR_OR_ACCVGPR 64 VADDR FMT_ANY OPR_VGPR 64 SRSRC FMT_RSRC_SCRATCH OPR_SREG 128 SOFFSET FMT_ANY OPR_SSRC_NOLIT 32 FMT_NUM_U64 OPR_GPUMEM 64 FMT_NUM_U64 OPR_GPUMEM 64 VMEM TEXTURE ATOMIC FALSE FALSE FALSE FALSE FALSE BUFFER_ATOMIC_SUB_X2 Subtract an unsigned 64-bit integer value stored in the data register from a value stored in a location in a buffer surface. Store the original value from buffer surface into a vector register iff the SC0 bit is set. ENC_MUBUF default 99 VDATA FMT_ANY OPR_VGPR_OR_ACCVGPR 64 VADDR FMT_ANY OPR_VGPR 64 SRSRC FMT_RSRC_SCRATCH OPR_SREG 128 SOFFSET FMT_ANY OPR_SSRC_NOLIT 32 FMT_NUM_U64 OPR_GPUMEM 64 FMT_NUM_U64 OPR_GPUMEM 64 VMEM TEXTURE ATOMIC FALSE FALSE FALSE FALSE FALSE BUFFER_ATOMIC_SMIN_X2 Select the minimum of two signed 64-bit integer inputs, given two values stored in the data register and a location in a buffer surface. Update the buffer surface with the selected value. Store the original value from buffer surface into a vector register iff the SC0 bit is set. ENC_MUBUF default 100 VDATA FMT_ANY OPR_VGPR_OR_ACCVGPR 64 VADDR FMT_ANY OPR_VGPR 64 SRSRC FMT_RSRC_SCRATCH OPR_SREG 128 SOFFSET FMT_ANY OPR_SSRC_NOLIT 32 FMT_NUM_I64 OPR_GPUMEM 64 FMT_NUM_I64 OPR_GPUMEM 64 VMEM TEXTURE ATOMIC FALSE FALSE FALSE FALSE FALSE BUFFER_ATOMIC_UMIN_X2 Select the minimum of two unsigned 64-bit integer inputs, given two values stored in the data register and a location in a buffer surface. Update the buffer surface with the selected value. Store the original value from buffer surface into a vector register iff the SC0 bit is set. ENC_MUBUF default 101 VDATA FMT_ANY OPR_VGPR_OR_ACCVGPR 64 VADDR FMT_ANY OPR_VGPR 64 SRSRC FMT_RSRC_SCRATCH OPR_SREG 128 SOFFSET FMT_ANY OPR_SSRC_NOLIT 32 FMT_NUM_U64 OPR_GPUMEM 64 FMT_NUM_U64 OPR_GPUMEM 64 VMEM TEXTURE ATOMIC FALSE FALSE FALSE FALSE FALSE BUFFER_ATOMIC_SMAX_X2 Select the maximum of two signed 64-bit integer inputs, given two values stored in the data register and a location in a buffer surface. Update the buffer surface with the selected value. Store the original value from buffer surface into a vector register iff the SC0 bit is set. ENC_MUBUF default 102 VDATA FMT_ANY OPR_VGPR_OR_ACCVGPR 64 VADDR FMT_ANY OPR_VGPR 64 SRSRC FMT_RSRC_SCRATCH OPR_SREG 128 SOFFSET FMT_ANY OPR_SSRC_NOLIT 32 FMT_NUM_I64 OPR_GPUMEM 64 FMT_NUM_I64 OPR_GPUMEM 64 VMEM TEXTURE ATOMIC FALSE FALSE FALSE FALSE FALSE BUFFER_ATOMIC_UMAX_X2 Select the maximum of two unsigned 64-bit integer inputs, given two values stored in the data register and a location in a buffer surface. Update the buffer surface with the selected value. Store the original value from buffer surface into a vector register iff the SC0 bit is set. ENC_MUBUF default 103 VDATA FMT_ANY OPR_VGPR_OR_ACCVGPR 64 VADDR FMT_ANY OPR_VGPR 64 SRSRC FMT_RSRC_SCRATCH OPR_SREG 128 SOFFSET FMT_ANY OPR_SSRC_NOLIT 32 FMT_NUM_U64 OPR_GPUMEM 64 FMT_NUM_U64 OPR_GPUMEM 64 VMEM TEXTURE ATOMIC FALSE FALSE FALSE FALSE FALSE BUFFER_ATOMIC_AND_X2 Calculate bitwise AND given two unsigned 64-bit integer values stored in the data register and a location in a buffer surface. Store the original value from buffer surface into a vector register iff the SC0 bit is set. ENC_MUBUF default 104 VDATA FMT_ANY OPR_VGPR_OR_ACCVGPR 64 VADDR FMT_ANY OPR_VGPR 64 SRSRC FMT_RSRC_SCRATCH OPR_SREG 128 SOFFSET FMT_ANY OPR_SSRC_NOLIT 32 FMT_NUM_B64 OPR_GPUMEM 64 FMT_NUM_B64 OPR_GPUMEM 64 VMEM TEXTURE ATOMIC FALSE FALSE FALSE FALSE FALSE BUFFER_ATOMIC_OR_X2 Calculate bitwise OR given two unsigned 64-bit integer values stored in the data register and a location in a buffer surface. Store the original value from buffer surface into a vector register iff the SC0 bit is set. ENC_MUBUF default 105 VDATA FMT_ANY OPR_VGPR_OR_ACCVGPR 64 VADDR FMT_ANY OPR_VGPR 64 SRSRC FMT_RSRC_SCRATCH OPR_SREG 128 SOFFSET FMT_ANY OPR_SSRC_NOLIT 32 FMT_NUM_B64 OPR_GPUMEM 64 FMT_NUM_B64 OPR_GPUMEM 64 VMEM TEXTURE ATOMIC FALSE FALSE FALSE FALSE FALSE BUFFER_ATOMIC_XOR_X2 Calculate bitwise XOR given two unsigned 64-bit integer values stored in the data register and a location in a buffer surface. Store the original value from buffer surface into a vector register iff the SC0 bit is set. ENC_MUBUF default 106 VDATA FMT_ANY OPR_VGPR_OR_ACCVGPR 64 VADDR FMT_ANY OPR_VGPR 64 SRSRC FMT_RSRC_SCRATCH OPR_SREG 128 SOFFSET FMT_ANY OPR_SSRC_NOLIT 32 FMT_NUM_B64 OPR_GPUMEM 64 FMT_NUM_B64 OPR_GPUMEM 64 VMEM TEXTURE ATOMIC FALSE FALSE FALSE FALSE FALSE BUFFER_ATOMIC_INC_X2 Increment an unsigned 64-bit integer value from a location in a buffer surface with wraparound to 0 if the value exceeds a value in the data register. Store the original value from buffer surface into a vector register iff the SC0 bit is set. ENC_MUBUF default 107 VDATA FMT_ANY OPR_VGPR_OR_ACCVGPR 64 VADDR FMT_ANY OPR_VGPR 64 SRSRC FMT_RSRC_SCRATCH OPR_SREG 128 SOFFSET FMT_ANY OPR_SSRC_NOLIT 32 FMT_NUM_U64 OPR_GPUMEM 64 FMT_NUM_U64 OPR_GPUMEM 64 VMEM TEXTURE ATOMIC FALSE FALSE FALSE FALSE FALSE BUFFER_ATOMIC_DEC_X2 Decrement an unsigned 64-bit integer value from a location in a buffer surface with wraparound to a value in the data register if the decrement yields a negative value. Store the original value from buffer surface into a vector register iff the SC0 bit is set. ENC_MUBUF default 108 VDATA FMT_ANY OPR_VGPR_OR_ACCVGPR 64 VADDR FMT_ANY OPR_VGPR 64 SRSRC FMT_RSRC_SCRATCH OPR_SREG 128 SOFFSET FMT_ANY OPR_SSRC_NOLIT 32 FMT_NUM_U64 OPR_GPUMEM 64 FMT_NUM_U64 OPR_GPUMEM 64 VMEM TEXTURE ATOMIC FALSE FALSE FALSE FALSE FALSE S_LOAD_DWORD Load 32 bits of data from the scalar memory into a scalar register. ENC_SMEM default 0 SDATA FMT_ANY OPR_SREG 32 SBASE FMT_BUF OPR_SREG 64 SOFFSET FMT_NUM_B32 OPR_SMEM_OFFSET 32 FMT_NUM_B32 OPR_GPUMEM 32 SMEM NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_LOAD_DWORDX2 Load 64 bits of data from the scalar memory into a scalar register. ENC_SMEM default 1 SDATA FMT_ANY OPR_SREG 64 SBASE FMT_BUF OPR_SREG 64 SOFFSET FMT_NUM_B32 OPR_SMEM_OFFSET 32 FMT_NUM_B64 OPR_GPUMEM 64 SMEM NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_LOAD_DWORDX4 Load 128 bits of data from the scalar memory into a scalar register. ENC_SMEM default 2 SDATA FMT_ANY OPR_SREG 128 SBASE FMT_BUF OPR_SREG 64 SOFFSET FMT_NUM_B32 OPR_SMEM_OFFSET 32 FMT_NUM_B128 OPR_GPUMEM 128 SMEM NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_LOAD_DWORDX8 Load 256 bits of data from the scalar memory into a scalar register. ENC_SMEM default 3 SDATA FMT_ANY OPR_SREG 256 SBASE FMT_BUF OPR_SREG 64 SOFFSET FMT_NUM_B32 OPR_SMEM_OFFSET 32 FMT_NUM_B256 OPR_GPUMEM 256 SMEM NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_LOAD_DWORDX16 Load 512 bits of data from the scalar memory into a scalar register. ENC_SMEM default 4 SDATA FMT_ANY OPR_SREG 512 SBASE FMT_BUF OPR_SREG 64 SOFFSET FMT_NUM_B32 OPR_SMEM_OFFSET 32 FMT_NUM_B512 OPR_GPUMEM 512 SMEM NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_SCRATCH_LOAD_DWORD Load 32 bits of data from the scalar scratch aperture into a scalar register. ENC_SMEM default 5 SDATA FMT_ANY OPR_SREG 32 SBASE FMT_BUF OPR_SREG 64 SOFFSET FMT_NUM_B32 OPR_SMEM_OFFSET 32 FMT_NUM_B32 OPR_GPUMEM 32 SMEM NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_SCRATCH_LOAD_DWORDX2 Load 64 bits of data from the scalar scratch aperture into a scalar register. ENC_SMEM default 6 SDATA FMT_ANY OPR_SREG 64 SBASE FMT_BUF OPR_SREG 64 SOFFSET FMT_NUM_B32 OPR_SMEM_OFFSET 32 FMT_NUM_B64 OPR_GPUMEM 64 SMEM NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_SCRATCH_LOAD_DWORDX4 Load 128 bits of data from the scalar scratch aperture into a scalar register. ENC_SMEM default 7 SDATA FMT_ANY OPR_SREG 128 SBASE FMT_BUF OPR_SREG 64 SOFFSET FMT_NUM_B32 OPR_SMEM_OFFSET 32 FMT_NUM_B128 OPR_GPUMEM 128 SMEM NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_BUFFER_LOAD_DWORD Load 32 bits of data from a scalar buffer surface into a scalar register. ENC_SMEM default 8 SDATA FMT_ANY OPR_SREG 32 SBASE FMT_RSRC_SCALAR OPR_SREG 128 SOFFSET FMT_NUM_B32 OPR_SMEM_OFFSET 32 FMT_NUM_B32 OPR_GPUMEM 32 SMEM NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_BUFFER_LOAD_DWORDX2 Load 64 bits of data from a scalar buffer surface into a scalar register. ENC_SMEM default 9 SDATA FMT_ANY OPR_SREG 64 SBASE FMT_RSRC_SCALAR OPR_SREG 128 SOFFSET FMT_NUM_B32 OPR_SMEM_OFFSET 32 FMT_NUM_B64 OPR_GPUMEM 64 SMEM NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_BUFFER_LOAD_DWORDX4 Load 128 bits of data from a scalar buffer surface into a scalar register. ENC_SMEM default 10 SDATA FMT_ANY OPR_SREG 128 SBASE FMT_RSRC_SCALAR OPR_SREG 128 SOFFSET FMT_NUM_B32 OPR_SMEM_OFFSET 32 FMT_NUM_B128 OPR_GPUMEM 128 SMEM NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_BUFFER_LOAD_DWORDX8 Load 256 bits of data from a scalar buffer surface into a scalar register. ENC_SMEM default 11 SDATA FMT_ANY OPR_SREG 256 SBASE FMT_RSRC_SCALAR OPR_SREG 128 SOFFSET FMT_NUM_B32 OPR_SMEM_OFFSET 32 FMT_NUM_B256 OPR_GPUMEM 256 SMEM NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_BUFFER_LOAD_DWORDX16 Load 512 bits of data from a scalar buffer surface into a scalar register. ENC_SMEM default 12 SDATA FMT_ANY OPR_SREG 512 SBASE FMT_RSRC_SCALAR OPR_SREG 128 SOFFSET FMT_NUM_B32 OPR_SMEM_OFFSET 32 FMT_NUM_B512 OPR_GPUMEM 512 SMEM NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_STORE_DWORD Store 32 bits of data from a scalar register into the scalar memory. ENC_SMEM default 16 SDATA FMT_ANY OPR_SREG 32 SBASE FMT_BUF OPR_SREG 64 SOFFSET FMT_NUM_B32 OPR_SMEM_OFFSET 32 FMT_NUM_B32 OPR_GPUMEM 32 SMEM NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_STORE_DWORDX2 Store 64 bits of data from a scalar register into the scalar memory. ENC_SMEM default 17 SDATA FMT_ANY OPR_SREG 64 SBASE FMT_BUF OPR_SREG 64 SOFFSET FMT_NUM_B32 OPR_SMEM_OFFSET 32 FMT_NUM_B64 OPR_GPUMEM 64 SMEM NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_STORE_DWORDX4 Store 128 bits of data from a scalar register into the scalar memory. ENC_SMEM default 18 SDATA FMT_ANY OPR_SREG 128 SBASE FMT_BUF OPR_SREG 64 SOFFSET FMT_NUM_B32 OPR_SMEM_OFFSET 32 FMT_NUM_B128 OPR_GPUMEM 128 SMEM NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_SCRATCH_STORE_DWORD Store 32 bits of data from a scalar register into the scalar scratch aperture. ENC_SMEM default 21 SDATA FMT_ANY OPR_SREG 32 SBASE FMT_BUF OPR_SREG 64 SOFFSET FMT_NUM_B32 OPR_SMEM_OFFSET 32 FMT_NUM_B32 OPR_GPUMEM 32 SMEM NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_SCRATCH_STORE_DWORDX2 Store 64 bits of data from a scalar register into the scalar scratch aperture. ENC_SMEM default 22 SDATA FMT_ANY OPR_SREG 64 SBASE FMT_BUF OPR_SREG 64 SOFFSET FMT_NUM_B32 OPR_SMEM_OFFSET 32 FMT_NUM_B64 OPR_GPUMEM 64 SMEM NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_SCRATCH_STORE_DWORDX4 Store 128 bits of data from a scalar register into the scalar scratch aperture. ENC_SMEM default 23 SDATA FMT_ANY OPR_SREG 128 SBASE FMT_BUF OPR_SREG 64 SOFFSET FMT_NUM_B32 OPR_SMEM_OFFSET 32 FMT_NUM_B128 OPR_GPUMEM 128 SMEM NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_BUFFER_STORE_DWORD Store 32 bits of data from a scalar register into a scalar buffer surface. ENC_SMEM default 24 SDATA FMT_ANY OPR_SREG 32 SBASE FMT_RSRC_SCALAR OPR_SREG 128 SOFFSET FMT_NUM_B32 OPR_SMEM_OFFSET 32 FMT_NUM_B32 OPR_GPUMEM 32 SMEM NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_BUFFER_STORE_DWORDX2 Store 64 bits of data from a scalar register into a scalar buffer surface. ENC_SMEM default 25 SDATA FMT_ANY OPR_SREG 64 SBASE FMT_RSRC_SCALAR OPR_SREG 128 SOFFSET FMT_NUM_B32 OPR_SMEM_OFFSET 32 FMT_NUM_B64 OPR_GPUMEM 64 SMEM NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_BUFFER_STORE_DWORDX4 Store 128 bits of data from a scalar register into a scalar buffer surface. ENC_SMEM default 26 SDATA FMT_ANY OPR_SREG 128 SBASE FMT_RSRC_SCALAR OPR_SREG 128 SOFFSET FMT_NUM_B32 OPR_SMEM_OFFSET 32 FMT_NUM_B128 OPR_GPUMEM 128 SMEM NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_DCACHE_INV Invalidate the scalar data L0 cache. ENC_SMEM default 32 SMEM NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_DCACHE_WB Write back dirty data in the scalar data L0 cache. ENC_SMEM default 33 SMEM NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_DCACHE_INV_VOL Invalidate the scalar data L0 cache volatile lines. ENC_SMEM default 34 SMEM NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_DCACHE_WB_VOL Write back dirty data in the scalar data L0 cache volatile lines. ENC_SMEM default 35 SMEM NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_MEMTIME Return current 64-bit timestamp. ENC_SMEM default 36 SDATA FMT_NUM_B64 OPR_SREG 64 SMEM NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_MEMREALTIME Return current 64-bit RTC. ENC_SMEM default 37 SDATA FMT_NUM_B64 OPR_SREG 64 SMEM NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_ATC_PROBE Probe or prefetch an address into the scalar data cache. ENC_SMEM default 38 SDATA FMT_NUM_B8 OPR_SIMM8 8 SBASE FMT_BUF OPR_SREG 64 SOFFSET FMT_NUM_B32 OPR_SMEM_OFFSET 32 SMEM NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_ATC_PROBE_BUFFER Probe or prefetch an address into the scalar data cache. ENC_SMEM default 39 SDATA FMT_NUM_B8 OPR_SIMM8 8 SBASE FMT_RSRC_SCALAR OPR_SREG 128 SOFFSET FMT_NUM_B32 OPR_SMEM_OFFSET 32 SMEM NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_DCACHE_DISCARD Discard one dirty scalar data L0 cache line. A cache line is 64 bytes. ENC_SMEM default 40 SBASE FMT_BUF OPR_SREG 64 SOFFSET FMT_NUM_B32 OPR_SMEM_OFFSET 32 SMEM NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_DCACHE_DISCARD_X2 Discard two consecutive dirty scalar data L0 cache lines. A cache line is 64 bytes. ENC_SMEM default 41 SBASE FMT_BUF OPR_SREG 64 SOFFSET FMT_NUM_B32 OPR_SMEM_OFFSET 32 SMEM NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_BUFFER_ATOMIC_SWAP Swap an unsigned 32-bit integer value in the data register with a location in a scalar buffer surface. ENC_SMEM default 64 SDATA FMT_ANY OPR_SREG 32 SBASE FMT_RSRC_SCALAR OPR_SREG 128 SOFFSET FMT_NUM_B32 OPR_SMEM_OFFSET 32 SMEM ATOMIC FALSE FALSE FALSE FALSE FALSE S_BUFFER_ATOMIC_CMPSWAP Compare two unsigned 32-bit integer values stored in the data comparison register and a location in a scalar buffer surface. Modify the memory location with a value in the data source register iff the comparison is equal. ENC_SMEM default 65 SDATA FMT_ANY OPR_SREG 64 SBASE FMT_RSRC_SCALAR OPR_SREG 128 SOFFSET FMT_NUM_B32 OPR_SMEM_OFFSET 32 SMEM ATOMIC FALSE FALSE FALSE FALSE FALSE S_BUFFER_ATOMIC_ADD Add two unsigned 32-bit integer values stored in the data register and a location in a scalar buffer surface. ENC_SMEM default 66 SDATA FMT_ANY OPR_SREG 32 SBASE FMT_RSRC_SCALAR OPR_SREG 128 SOFFSET FMT_NUM_B32 OPR_SMEM_OFFSET 32 SMEM ATOMIC FALSE FALSE FALSE FALSE FALSE S_BUFFER_ATOMIC_SUB Subtract an unsigned 32-bit integer value stored in the data register from a value stored in a location in a scalar buffer surface. ENC_SMEM default 67 SDATA FMT_ANY OPR_SREG 32 SBASE FMT_RSRC_SCALAR OPR_SREG 128 SOFFSET FMT_NUM_B32 OPR_SMEM_OFFSET 32 SMEM ATOMIC FALSE FALSE FALSE FALSE FALSE S_BUFFER_ATOMIC_SMIN Select the minimum of two signed 32-bit integer inputs, given two values stored in the data register and a location in a scalar buffer surface. Update the scalar buffer with the selected value. ENC_SMEM default 68 SDATA FMT_ANY OPR_SREG 32 SBASE FMT_RSRC_SCALAR OPR_SREG 128 SOFFSET FMT_NUM_B32 OPR_SMEM_OFFSET 32 SMEM ATOMIC FALSE FALSE FALSE FALSE FALSE S_BUFFER_ATOMIC_UMIN Select the minimum of two unsigned 32-bit integer inputs, given two values stored in the data register and a location in a scalar buffer surface. Update the scalar buffer with the selected value. ENC_SMEM default 69 SDATA FMT_ANY OPR_SREG 32 SBASE FMT_RSRC_SCALAR OPR_SREG 128 SOFFSET FMT_NUM_B32 OPR_SMEM_OFFSET 32 SMEM ATOMIC FALSE FALSE FALSE FALSE FALSE S_BUFFER_ATOMIC_SMAX Select the maximum of two signed 32-bit integer inputs, given two values stored in the data register and a location in a scalar buffer surface. Update the scalar buffer with the selected value. ENC_SMEM default 70 SDATA FMT_ANY OPR_SREG 32 SBASE FMT_RSRC_SCALAR OPR_SREG 128 SOFFSET FMT_NUM_B32 OPR_SMEM_OFFSET 32 SMEM ATOMIC FALSE FALSE FALSE FALSE FALSE S_BUFFER_ATOMIC_UMAX Select the maximum of two unsigned 32-bit integer inputs, given two values stored in the data register and a location in a scalar buffer surface. Update the scalar buffer with the selected value. ENC_SMEM default 71 SDATA FMT_ANY OPR_SREG 32 SBASE FMT_RSRC_SCALAR OPR_SREG 128 SOFFSET FMT_NUM_B32 OPR_SMEM_OFFSET 32 SMEM ATOMIC FALSE FALSE FALSE FALSE FALSE S_BUFFER_ATOMIC_AND Calculate bitwise AND given two unsigned 32-bit integer values stored in the data register and a location in a scalar buffer surface. ENC_SMEM default 72 SDATA FMT_ANY OPR_SREG 32 SBASE FMT_RSRC_SCALAR OPR_SREG 128 SOFFSET FMT_NUM_B32 OPR_SMEM_OFFSET 32 SMEM ATOMIC FALSE FALSE FALSE FALSE FALSE S_BUFFER_ATOMIC_OR Calculate bitwise OR given two unsigned 32-bit integer values stored in the data register and a location in a scalar buffer surface. ENC_SMEM default 73 SDATA FMT_ANY OPR_SREG 32 SBASE FMT_RSRC_SCALAR OPR_SREG 128 SOFFSET FMT_NUM_B32 OPR_SMEM_OFFSET 32 SMEM ATOMIC FALSE FALSE FALSE FALSE FALSE S_BUFFER_ATOMIC_XOR Calculate bitwise XOR given two unsigned 32-bit integer values stored in the data register and a location in a scalar buffer surface. ENC_SMEM default 74 SDATA FMT_ANY OPR_SREG 32 SBASE FMT_RSRC_SCALAR OPR_SREG 128 SOFFSET FMT_NUM_B32 OPR_SMEM_OFFSET 32 SMEM ATOMIC FALSE FALSE FALSE FALSE FALSE S_BUFFER_ATOMIC_INC Increment an unsigned 32-bit integer value from a location in a scalar buffer surface with wraparound to 0 if the value exceeds a value in the data register. ENC_SMEM default 75 SDATA FMT_ANY OPR_SREG 32 SBASE FMT_RSRC_SCALAR OPR_SREG 128 SOFFSET FMT_NUM_B32 OPR_SMEM_OFFSET 32 SMEM ATOMIC FALSE FALSE FALSE FALSE FALSE S_BUFFER_ATOMIC_DEC Decrement an unsigned 32-bit integer value from a location in a scalar buffer surface with wraparound to a value in the data register if the decrement yields a negative value. ENC_SMEM default 76 SDATA FMT_ANY OPR_SREG 32 SBASE FMT_RSRC_SCALAR OPR_SREG 128 SOFFSET FMT_NUM_B32 OPR_SMEM_OFFSET 32 SMEM ATOMIC FALSE FALSE FALSE FALSE FALSE S_BUFFER_ATOMIC_SWAP_X2 Swap an unsigned 64-bit integer value in the data register with a location in a scalar buffer surface. ENC_SMEM default 96 SDATA FMT_ANY OPR_SREG 64 SBASE FMT_RSRC_SCALAR OPR_SREG 128 SOFFSET FMT_NUM_B32 OPR_SMEM_OFFSET 32 SMEM ATOMIC FALSE FALSE FALSE FALSE FALSE S_BUFFER_ATOMIC_CMPSWAP_X2 Compare two unsigned 64-bit integer values stored in the data comparison register and a location in a scalar buffer surface. Modify the memory location with a value in the data source register iff the comparison is equal. ENC_SMEM default 97 SDATA FMT_ANY OPR_SREG 128 SBASE FMT_RSRC_SCALAR OPR_SREG 128 SOFFSET FMT_NUM_B32 OPR_SMEM_OFFSET 32 SMEM ATOMIC FALSE FALSE FALSE FALSE FALSE S_BUFFER_ATOMIC_ADD_X2 Add two unsigned 64-bit integer values stored in the data register and a location in a scalar buffer surface. ENC_SMEM default 98 SDATA FMT_ANY OPR_SREG 64 SBASE FMT_RSRC_SCALAR OPR_SREG 128 SOFFSET FMT_NUM_B32 OPR_SMEM_OFFSET 32 SMEM ATOMIC FALSE FALSE FALSE FALSE FALSE S_BUFFER_ATOMIC_SUB_X2 Subtract an unsigned 64-bit integer value stored in the data register from a value stored in a location in a scalar buffer surface. ENC_SMEM default 99 SDATA FMT_ANY OPR_SREG 64 SBASE FMT_RSRC_SCALAR OPR_SREG 128 SOFFSET FMT_NUM_B32 OPR_SMEM_OFFSET 32 SMEM ATOMIC FALSE FALSE FALSE FALSE FALSE S_BUFFER_ATOMIC_SMIN_X2 Select the minimum of two signed 64-bit integer inputs, given two values stored in the data register and a location in a scalar buffer surface. Update the scalar buffer with the selected value. ENC_SMEM default 100 SDATA FMT_ANY OPR_SREG 64 SBASE FMT_RSRC_SCALAR OPR_SREG 128 SOFFSET FMT_NUM_B32 OPR_SMEM_OFFSET 32 SMEM ATOMIC FALSE FALSE FALSE FALSE FALSE S_BUFFER_ATOMIC_UMIN_X2 Select the minimum of two unsigned 64-bit integer inputs, given two values stored in the data register and a location in a scalar buffer surface. Update the scalar buffer with the selected value. ENC_SMEM default 101 SDATA FMT_ANY OPR_SREG 64 SBASE FMT_RSRC_SCALAR OPR_SREG 128 SOFFSET FMT_NUM_B32 OPR_SMEM_OFFSET 32 SMEM ATOMIC FALSE FALSE FALSE FALSE FALSE S_BUFFER_ATOMIC_SMAX_X2 Select the maximum of two signed 64-bit integer inputs, given two values stored in the data register and a location in a scalar buffer surface. Update the scalar buffer with the selected value. ENC_SMEM default 102 SDATA FMT_ANY OPR_SREG 64 SBASE FMT_RSRC_SCALAR OPR_SREG 128 SOFFSET FMT_NUM_B32 OPR_SMEM_OFFSET 32 SMEM ATOMIC FALSE FALSE FALSE FALSE FALSE S_BUFFER_ATOMIC_UMAX_X2 Select the maximum of two unsigned 64-bit integer inputs, given two values stored in the data register and a location in a scalar buffer surface. Update the scalar buffer with the selected value. ENC_SMEM default 103 SDATA FMT_ANY OPR_SREG 64 SBASE FMT_RSRC_SCALAR OPR_SREG 128 SOFFSET FMT_NUM_B32 OPR_SMEM_OFFSET 32 SMEM ATOMIC FALSE FALSE FALSE FALSE FALSE S_BUFFER_ATOMIC_AND_X2 Calculate bitwise AND given two unsigned 64-bit integer values stored in the data register and a location in a scalar buffer surface. ENC_SMEM default 104 SDATA FMT_ANY OPR_SREG 64 SBASE FMT_RSRC_SCALAR OPR_SREG 128 SOFFSET FMT_NUM_B32 OPR_SMEM_OFFSET 32 SMEM ATOMIC FALSE FALSE FALSE FALSE FALSE S_BUFFER_ATOMIC_OR_X2 Calculate bitwise OR given two unsigned 64-bit integer values stored in the data register and a location in a scalar buffer surface. ENC_SMEM default 105 SDATA FMT_ANY OPR_SREG 64 SBASE FMT_RSRC_SCALAR OPR_SREG 128 SOFFSET FMT_NUM_B32 OPR_SMEM_OFFSET 32 SMEM ATOMIC FALSE FALSE FALSE FALSE FALSE S_BUFFER_ATOMIC_XOR_X2 Calculate bitwise XOR given two unsigned 64-bit integer values stored in the data register and a location in a scalar buffer surface. ENC_SMEM default 106 SDATA FMT_ANY OPR_SREG 64 SBASE FMT_RSRC_SCALAR OPR_SREG 128 SOFFSET FMT_NUM_B32 OPR_SMEM_OFFSET 32 SMEM ATOMIC FALSE FALSE FALSE FALSE FALSE S_BUFFER_ATOMIC_INC_X2 Increment an unsigned 64-bit integer value from a location in a scalar buffer surface with wraparound to 0 if the value exceeds a value in the data register. ENC_SMEM default 107 SDATA FMT_ANY OPR_SREG 64 SBASE FMT_RSRC_SCALAR OPR_SREG 128 SOFFSET FMT_NUM_B32 OPR_SMEM_OFFSET 32 SMEM ATOMIC FALSE FALSE FALSE FALSE FALSE S_BUFFER_ATOMIC_DEC_X2 Decrement an unsigned 64-bit integer value from a location in a scalar buffer surface with wraparound to a value in the data register if the decrement yields a negative value. ENC_SMEM default 108 SDATA FMT_ANY OPR_SREG 64 SBASE FMT_RSRC_SCALAR OPR_SREG 128 SOFFSET FMT_NUM_B32 OPR_SMEM_OFFSET 32 SMEM ATOMIC FALSE FALSE FALSE FALSE FALSE S_ATOMIC_SWAP Swap an unsigned 32-bit integer value in the data register with a location in the scalar memory. ENC_SMEM default 128 SDATA FMT_ANY OPR_SREG 32 SBASE FMT_BUF OPR_SREG 64 SOFFSET FMT_NUM_B32 OPR_SMEM_OFFSET 32 SMEM ATOMIC FALSE FALSE FALSE FALSE FALSE S_ATOMIC_CMPSWAP Compare two unsigned 32-bit integer values stored in the data comparison register and a location in the scalar memory. Modify the memory location with a value in the data source register iff the comparison is equal. ENC_SMEM default 129 SDATA FMT_ANY OPR_SREG 64 SBASE FMT_BUF OPR_SREG 64 SOFFSET FMT_NUM_B32 OPR_SMEM_OFFSET 32 SMEM ATOMIC FALSE FALSE FALSE FALSE FALSE S_ATOMIC_ADD Add two unsigned 32-bit integer values stored in the data register and a location in the scalar memory. ENC_SMEM default 130 SDATA FMT_ANY OPR_SREG 32 SBASE FMT_BUF OPR_SREG 64 SOFFSET FMT_NUM_B32 OPR_SMEM_OFFSET 32 SMEM ATOMIC FALSE FALSE FALSE FALSE FALSE S_ATOMIC_SUB Subtract an unsigned 32-bit integer value stored in the data register from a value stored in a location in the scalar memory. ENC_SMEM default 131 SDATA FMT_ANY OPR_SREG 32 SBASE FMT_BUF OPR_SREG 64 SOFFSET FMT_NUM_B32 OPR_SMEM_OFFSET 32 SMEM ATOMIC FALSE FALSE FALSE FALSE FALSE S_ATOMIC_SMIN Select the minimum of two signed 32-bit integer inputs, given two values stored in the data register and a location in the scalar memory. Update the scalar memory with the selected value. ENC_SMEM default 132 SDATA FMT_ANY OPR_SREG 32 SBASE FMT_BUF OPR_SREG 64 SOFFSET FMT_NUM_B32 OPR_SMEM_OFFSET 32 SMEM ATOMIC FALSE FALSE FALSE FALSE FALSE S_ATOMIC_UMIN Select the minimum of two unsigned 32-bit integer inputs, given two values stored in the data register and a location in the scalar memory. Update the scalar memory with the selected value. ENC_SMEM default 133 SDATA FMT_ANY OPR_SREG 32 SBASE FMT_BUF OPR_SREG 64 SOFFSET FMT_NUM_B32 OPR_SMEM_OFFSET 32 SMEM ATOMIC FALSE FALSE FALSE FALSE FALSE S_ATOMIC_SMAX Select the maximum of two signed 32-bit integer inputs, given two values stored in the data register and a location in the scalar memory. Update the scalar memory with the selected value. ENC_SMEM default 134 SDATA FMT_ANY OPR_SREG 32 SBASE FMT_BUF OPR_SREG 64 SOFFSET FMT_NUM_B32 OPR_SMEM_OFFSET 32 SMEM ATOMIC FALSE FALSE FALSE FALSE FALSE S_ATOMIC_UMAX Select the maximum of two unsigned 32-bit integer inputs, given two values stored in the data register and a location in the scalar memory. Update the scalar memory with the selected value. ENC_SMEM default 135 SDATA FMT_ANY OPR_SREG 32 SBASE FMT_BUF OPR_SREG 64 SOFFSET FMT_NUM_B32 OPR_SMEM_OFFSET 32 SMEM ATOMIC FALSE FALSE FALSE FALSE FALSE S_ATOMIC_AND Calculate bitwise AND given two unsigned 32-bit integer values stored in the data register and a location in the scalar memory. ENC_SMEM default 136 SDATA FMT_ANY OPR_SREG 32 SBASE FMT_BUF OPR_SREG 64 SOFFSET FMT_NUM_B32 OPR_SMEM_OFFSET 32 SMEM ATOMIC FALSE FALSE FALSE FALSE FALSE S_ATOMIC_OR Calculate bitwise OR given two unsigned 32-bit integer values stored in the data register and a location in the scalar memory. ENC_SMEM default 137 SDATA FMT_ANY OPR_SREG 32 SBASE FMT_BUF OPR_SREG 64 SOFFSET FMT_NUM_B32 OPR_SMEM_OFFSET 32 SMEM ATOMIC FALSE FALSE FALSE FALSE FALSE S_ATOMIC_XOR Calculate bitwise XOR given two unsigned 32-bit integer values stored in the data register and a location in the scalar memory. ENC_SMEM default 138 SDATA FMT_ANY OPR_SREG 32 SBASE FMT_BUF OPR_SREG 64 SOFFSET FMT_NUM_B32 OPR_SMEM_OFFSET 32 SMEM ATOMIC FALSE FALSE FALSE FALSE FALSE S_ATOMIC_INC Increment an unsigned 32-bit integer value from a location in the scalar memory with wraparound to 0 if the value exceeds a value in the data register. ENC_SMEM default 139 SDATA FMT_ANY OPR_SREG 32 SBASE FMT_BUF OPR_SREG 64 SOFFSET FMT_NUM_B32 OPR_SMEM_OFFSET 32 SMEM ATOMIC FALSE FALSE FALSE FALSE FALSE S_ATOMIC_DEC Decrement an unsigned 32-bit integer value from a location in the scalar memory with wraparound to a value in the data register if the decrement yields a negative value. ENC_SMEM default 140 SDATA FMT_ANY OPR_SREG 32 SBASE FMT_BUF OPR_SREG 64 SOFFSET FMT_NUM_B32 OPR_SMEM_OFFSET 32 SMEM ATOMIC FALSE FALSE FALSE FALSE FALSE S_ATOMIC_SWAP_X2 Swap an unsigned 64-bit integer value in the data register with a location in the scalar memory. ENC_SMEM default 160 SDATA FMT_ANY OPR_SREG 64 SBASE FMT_BUF OPR_SREG 64 SOFFSET FMT_NUM_B32 OPR_SMEM_OFFSET 32 SMEM ATOMIC FALSE FALSE FALSE FALSE FALSE S_ATOMIC_CMPSWAP_X2 Compare two unsigned 64-bit integer values stored in the data comparison register and a location in the scalar memory. Modify the memory location with a value in the data source register iff the comparison is equal. ENC_SMEM default 161 SDATA FMT_ANY OPR_SREG 128 SBASE FMT_BUF OPR_SREG 64 SOFFSET FMT_NUM_B32 OPR_SMEM_OFFSET 32 SMEM ATOMIC FALSE FALSE FALSE FALSE FALSE S_ATOMIC_ADD_X2 Add two unsigned 64-bit integer values stored in the data register and a location in the scalar memory. ENC_SMEM default 162 SDATA FMT_ANY OPR_SREG 64 SBASE FMT_BUF OPR_SREG 64 SOFFSET FMT_NUM_B32 OPR_SMEM_OFFSET 32 SMEM ATOMIC FALSE FALSE FALSE FALSE FALSE S_ATOMIC_SUB_X2 Subtract an unsigned 64-bit integer value stored in the data register from a value stored in a location in the scalar memory. ENC_SMEM default 163 SDATA FMT_ANY OPR_SREG 64 SBASE FMT_BUF OPR_SREG 64 SOFFSET FMT_NUM_B32 OPR_SMEM_OFFSET 32 SMEM ATOMIC FALSE FALSE FALSE FALSE FALSE S_ATOMIC_SMIN_X2 Select the minimum of two signed 64-bit integer inputs, given two values stored in the data register and a location in the scalar memory. Update the scalar memory with the selected value. ENC_SMEM default 164 SDATA FMT_ANY OPR_SREG 64 SBASE FMT_BUF OPR_SREG 64 SOFFSET FMT_NUM_B32 OPR_SMEM_OFFSET 32 SMEM ATOMIC FALSE FALSE FALSE FALSE FALSE S_ATOMIC_UMIN_X2 Select the minimum of two unsigned 64-bit integer inputs, given two values stored in the data register and a location in the scalar memory. Update the scalar memory with the selected value. ENC_SMEM default 165 SDATA FMT_ANY OPR_SREG 64 SBASE FMT_BUF OPR_SREG 64 SOFFSET FMT_NUM_B32 OPR_SMEM_OFFSET 32 SMEM ATOMIC FALSE FALSE FALSE FALSE FALSE S_ATOMIC_SMAX_X2 Select the maximum of two signed 64-bit integer inputs, given two values stored in the data register and a location in the scalar memory. Update the scalar memory with the selected value. ENC_SMEM default 166 SDATA FMT_ANY OPR_SREG 64 SBASE FMT_BUF OPR_SREG 64 SOFFSET FMT_NUM_B32 OPR_SMEM_OFFSET 32 SMEM ATOMIC FALSE FALSE FALSE FALSE FALSE S_ATOMIC_UMAX_X2 Select the maximum of two unsigned 64-bit integer inputs, given two values stored in the data register and a location in the scalar memory. Update the scalar memory with the selected value. ENC_SMEM default 167 SDATA FMT_ANY OPR_SREG 64 SBASE FMT_BUF OPR_SREG 64 SOFFSET FMT_NUM_B32 OPR_SMEM_OFFSET 32 SMEM ATOMIC FALSE FALSE FALSE FALSE FALSE S_ATOMIC_AND_X2 Calculate bitwise AND given two unsigned 64-bit integer values stored in the data register and a location in the scalar memory. ENC_SMEM default 168 SDATA FMT_ANY OPR_SREG 64 SBASE FMT_BUF OPR_SREG 64 SOFFSET FMT_NUM_B32 OPR_SMEM_OFFSET 32 SMEM ATOMIC FALSE FALSE FALSE FALSE FALSE S_ATOMIC_OR_X2 Calculate bitwise OR given two unsigned 64-bit integer values stored in the data register and a location in the scalar memory. ENC_SMEM default 169 SDATA FMT_ANY OPR_SREG 64 SBASE FMT_BUF OPR_SREG 64 SOFFSET FMT_NUM_B32 OPR_SMEM_OFFSET 32 SMEM ATOMIC FALSE FALSE FALSE FALSE FALSE S_ATOMIC_XOR_X2 Calculate bitwise XOR given two unsigned 64-bit integer values stored in the data register and a location in the scalar memory. ENC_SMEM default 170 SDATA FMT_ANY OPR_SREG 64 SBASE FMT_BUF OPR_SREG 64 SOFFSET FMT_NUM_B32 OPR_SMEM_OFFSET 32 SMEM ATOMIC FALSE FALSE FALSE FALSE FALSE S_ATOMIC_INC_X2 Increment an unsigned 64-bit integer value from a location in the scalar memory with wraparound to 0 if the value exceeds a value in the data register. ENC_SMEM default 171 SDATA FMT_ANY OPR_SREG 64 SBASE FMT_BUF OPR_SREG 64 SOFFSET FMT_NUM_B32 OPR_SMEM_OFFSET 32 SMEM ATOMIC FALSE FALSE FALSE FALSE FALSE S_ATOMIC_DEC_X2 Decrement an unsigned 64-bit integer value from a location in the scalar memory with wraparound to a value in the data register if the decrement yields a negative value. ENC_SMEM default 172 SDATA FMT_ANY OPR_SREG 64 SBASE FMT_BUF OPR_SREG 64 SOFFSET FMT_NUM_B32 OPR_SMEM_OFFSET 32 SMEM ATOMIC FALSE FALSE FALSE FALSE FALSE S_MOV_B32 Move scalar input into a scalar register. ENC_SOP1 default 0 SDST FMT_NUM_B32 OPR_SDST 32 SSRC0 FMT_NUM_B32 OPR_SSRC 32 SOP1_INST_LITERAL has_lit_0 0 SDST FMT_NUM_B32 OPR_SDST 32 SIMM32 FMT_NUM_B32 OPR_SSRC 32 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_MOV_B64 Move scalar input into a scalar register. ENC_SOP1 default 1 SDST FMT_NUM_B64 OPR_SDST 64 SSRC0 FMT_NUM_B64 OPR_SSRC 64 SOP1_INST_LITERAL has_lit_0 1 SDST FMT_NUM_B64 OPR_SDST 64 SIMM32 FMT_NUM_B64 OPR_SSRC 64 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_CMOV_B32 Move scalar input into a scalar register iff SCC is nonzero. ENC_SOP1 default 2 SDST FMT_NUM_B32 OPR_SDST 32 SSRC0 FMT_NUM_B32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP1_INST_LITERAL has_lit_0 2 SDST FMT_NUM_B32 OPR_SDST 32 SIMM32 FMT_NUM_B32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_CMOV_B64 Move scalar input into a scalar register iff SCC is nonzero. ENC_SOP1 default 3 SDST FMT_NUM_B64 OPR_SDST 64 SSRC0 FMT_NUM_B64 OPR_SSRC 64 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP1_INST_LITERAL has_lit_0 3 SDST FMT_NUM_B64 OPR_SDST 64 SIMM32 FMT_NUM_B64 OPR_SSRC 64 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_NOT_B32 Calculate bitwise negation on a scalar input, store the result into a scalar register and set SCC iff the result is nonzero. ENC_SOP1 default 4 SDST FMT_NUM_U32 OPR_SDST 32 SSRC0 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP1_INST_LITERAL has_lit_0 4 SDST FMT_NUM_U32 OPR_SDST 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_NOT_B64 Calculate bitwise negation on a scalar input, store the result into a scalar register and set SCC iff the result is nonzero. ENC_SOP1 default 5 SDST FMT_NUM_U64 OPR_SDST 64 SSRC0 FMT_NUM_U64 OPR_SSRC 64 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP1_INST_LITERAL has_lit_0 5 SDST FMT_NUM_U64 OPR_SDST 64 SIMM32 FMT_NUM_U64 OPR_SSRC 64 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_WQM_B32 Given an active pixel mask in a scalar input, calculate whole quad mode mask for that input, store the result into a scalar register and set SCC iff the result is nonzero. ENC_SOP1 default 6 SDST FMT_NUM_U32 OPR_SDST 32 SSRC0 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP1_INST_LITERAL has_lit_0 6 SDST FMT_NUM_U32 OPR_SDST 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_WQM_B64 Given an active pixel mask in a scalar input, calculate whole quad mode mask for that input, store the result into a scalar register and set SCC iff the result is nonzero. ENC_SOP1 default 7 SDST FMT_NUM_U64 OPR_SDST 64 SSRC0 FMT_NUM_U64 OPR_SSRC 64 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP1_INST_LITERAL has_lit_0 7 SDST FMT_NUM_U64 OPR_SDST 64 SIMM32 FMT_NUM_U64 OPR_SSRC 64 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_BREV_B32 Reverse the order of bits in a scalar input and store the result into a scalar register. ENC_SOP1 default 8 SDST FMT_NUM_U32 OPR_SDST 32 SSRC0 FMT_NUM_U32 OPR_SSRC 32 SOP1_INST_LITERAL has_lit_0 8 SDST FMT_NUM_U32 OPR_SDST 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_BREV_B64 Reverse the order of bits in a scalar input and store the result into a scalar register. ENC_SOP1 default 9 SDST FMT_NUM_U64 OPR_SDST 64 SSRC0 FMT_NUM_U64 OPR_SSRC 64 SOP1_INST_LITERAL has_lit_0 9 SDST FMT_NUM_U64 OPR_SDST 64 SIMM32 FMT_NUM_U64 OPR_SSRC 64 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_BCNT0_I32_B32 Count the number of "0" bits in a scalar input, store the result into a scalar register and set SCC iff the result is nonzero. ENC_SOP1 default 10 SDST FMT_NUM_I32 OPR_SDST 32 SSRC0 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP1_INST_LITERAL has_lit_0 10 SDST FMT_NUM_I32 OPR_SDST 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_BCNT0_I32_B64 Count the number of "0" bits in a scalar input, store the result into a scalar register and set SCC iff the result is nonzero. ENC_SOP1 default 11 SDST FMT_NUM_I32 OPR_SDST 32 SSRC0 FMT_NUM_U64 OPR_SSRC 64 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP1_INST_LITERAL has_lit_0 11 SDST FMT_NUM_I32 OPR_SDST 32 SIMM32 FMT_NUM_U64 OPR_SSRC 64 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_BCNT1_I32_B32 Count the number of "1" bits in a scalar input, store the result into a scalar register and set SCC iff the result is nonzero. ENC_SOP1 default 12 SDST FMT_NUM_I32 OPR_SDST 32 SSRC0 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP1_INST_LITERAL has_lit_0 12 SDST FMT_NUM_I32 OPR_SDST 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_BCNT1_I32_B64 Count the number of "1" bits in a scalar input, store the result into a scalar register and set SCC iff the result is nonzero. ENC_SOP1 default 13 SDST FMT_NUM_I32 OPR_SDST 32 SSRC0 FMT_NUM_U64 OPR_SSRC 64 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP1_INST_LITERAL has_lit_0 13 SDST FMT_NUM_I32 OPR_SDST 32 SIMM32 FMT_NUM_U64 OPR_SSRC 64 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_FF0_I32_B32 Count the number of trailing "1" bits before the first "0" in a scalar input and store the result into a scalar register. Store -1 if there are no "0" bits in the input. ENC_SOP1 default 14 SDST FMT_NUM_I32 OPR_SDST 32 SSRC0 FMT_NUM_U32 OPR_SSRC 32 SOP1_INST_LITERAL has_lit_0 14 SDST FMT_NUM_I32 OPR_SDST 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_FF0_I32_B64 Count the number of trailing "1" bits before the first "0" in a scalar input and store the result into a scalar register. Store -1 if there are no "0" bits in the input. ENC_SOP1 default 15 SDST FMT_NUM_I32 OPR_SDST 32 SSRC0 FMT_NUM_U64 OPR_SSRC 64 SOP1_INST_LITERAL has_lit_0 15 SDST FMT_NUM_I32 OPR_SDST 32 SIMM32 FMT_NUM_U64 OPR_SSRC 64 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_FF1_I32_B32 Count the number of trailing "0" bits before the first "1" in a scalar input and store the result into a scalar register. Store -1 if there are no "1" bits in the input. ENC_SOP1 default 16 SDST FMT_NUM_I32 OPR_SDST 32 SSRC0 FMT_NUM_U32 OPR_SSRC 32 SOP1_INST_LITERAL has_lit_0 16 SDST FMT_NUM_I32 OPR_SDST 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_FF1_I32_B64 Count the number of trailing "0" bits before the first "1" in a scalar input and store the result into a scalar register. Store -1 if there are no "1" bits in the input. ENC_SOP1 default 17 SDST FMT_NUM_I32 OPR_SDST 32 SSRC0 FMT_NUM_U64 OPR_SSRC 64 SOP1_INST_LITERAL has_lit_0 17 SDST FMT_NUM_I32 OPR_SDST 32 SIMM32 FMT_NUM_U64 OPR_SSRC 64 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_FLBIT_I32_B32 Count the number of leading "0" bits before the first "1" in a scalar input and store the result into a scalar register. Store -1 if there are no "1" bits. ENC_SOP1 default 18 SDST FMT_NUM_I32 OPR_SDST 32 SSRC0 FMT_NUM_U32 OPR_SSRC 32 SOP1_INST_LITERAL has_lit_0 18 SDST FMT_NUM_I32 OPR_SDST 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_FLBIT_I32_B64 Count the number of leading "0" bits before the first "1" in a scalar input and store the result into a scalar register. Store -1 if there are no "1" bits. ENC_SOP1 default 19 SDST FMT_NUM_I32 OPR_SDST 32 SSRC0 FMT_NUM_U64 OPR_SSRC 64 SOP1_INST_LITERAL has_lit_0 19 SDST FMT_NUM_I32 OPR_SDST 32 SIMM32 FMT_NUM_U64 OPR_SSRC 64 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_FLBIT_I32 Count the number of leading bits that are the same as the sign bit of a scalar input and store the result into a scalar register. Store -1 if all input bits are the same. ENC_SOP1 default 20 SDST FMT_NUM_I32 OPR_SDST 32 SSRC0 FMT_NUM_I32 OPR_SSRC 32 SOP1_INST_LITERAL has_lit_0 20 SDST FMT_NUM_I32 OPR_SDST 32 SIMM32 FMT_NUM_I32 OPR_SSRC 32 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_FLBIT_I32_I64 Count the number of leading bits that are the same as the sign bit of a scalar input and store the result into a scalar register. Store -1 if all input bits are the same. ENC_SOP1 default 21 SDST FMT_NUM_I32 OPR_SDST 32 SSRC0 FMT_NUM_I64 OPR_SSRC 64 SOP1_INST_LITERAL has_lit_0 21 SDST FMT_NUM_I32 OPR_SDST 32 SIMM32 FMT_NUM_I64 OPR_SSRC 64 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_SEXT_I32_I8 Sign extend a signed 8 bit scalar input to 32 bits and store the result into a scalar register. ENC_SOP1 default 22 SDST FMT_NUM_I32 OPR_SDST 32 SSRC0 FMT_NUM_I8 OPR_SSRC 8 SOP1_INST_LITERAL has_lit_0 22 SDST FMT_NUM_I32 OPR_SDST 32 SIMM32 FMT_NUM_I8 OPR_SSRC 8 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_SEXT_I32_I16 Sign extend a signed 16 bit scalar input to 32 bits and store the result into a scalar register. ENC_SOP1 default 23 SDST FMT_NUM_I32 OPR_SDST 32 SSRC0 FMT_NUM_I16 OPR_SSRC 16 SOP1_INST_LITERAL has_lit_0 23 SDST FMT_NUM_I32 OPR_SDST 32 SIMM32 FMT_NUM_I16 OPR_SSRC 16 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_BITSET0_B32 Given a bit offset in a scalar input, set the indicated bit in the destination scalar register to 0. ENC_SOP1 default 24 SDST FMT_NUM_B32 OPR_SDST 32 SSRC0 FMT_NUM_U32 OPR_SSRC 32 SOP1_INST_LITERAL has_lit_0 24 SDST FMT_NUM_B32 OPR_SDST 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_BITSET0_B64 Given a bit offset in a scalar input, set the indicated bit in the destination scalar register to 0. ENC_SOP1 default 25 SDST FMT_NUM_B64 OPR_SDST 64 SSRC0 FMT_NUM_U32 OPR_SSRC 32 SOP1_INST_LITERAL has_lit_0 25 SDST FMT_NUM_B64 OPR_SDST 64 SIMM32 FMT_NUM_U32 OPR_SSRC 32 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_BITSET1_B32 Given a bit offset in a scalar input, set the indicated bit in the destination scalar register to 1. ENC_SOP1 default 26 SDST FMT_NUM_B32 OPR_SDST 32 SSRC0 FMT_NUM_U32 OPR_SSRC 32 SOP1_INST_LITERAL has_lit_0 26 SDST FMT_NUM_B32 OPR_SDST 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_BITSET1_B64 Given a bit offset in a scalar input, set the indicated bit in the destination scalar register to 1. ENC_SOP1 default 27 SDST FMT_NUM_B64 OPR_SDST 64 SSRC0 FMT_NUM_U32 OPR_SSRC 32 SOP1_INST_LITERAL has_lit_0 27 SDST FMT_NUM_B64 OPR_SDST 64 SIMM32 FMT_NUM_U32 OPR_SSRC 32 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_GETPC_B64 Store the address of the next instruction to a scalar register. ENC_SOP1 default 28 SDST FMT_NUM_B64 OPR_SDST 64 FMT_NUM_B64 OPR_PC 64 SALU NOT_ASSIGNED TRUE FALSE TRUE FALSE FALSE S_SETPC_B64 Jump to an address specified in a scalar register. ENC_SOP1 default 29 SSRC0 FMT_NUM_B64 OPR_SREG 64 FMT_NUM_B64 OPR_PC 64 SALU NOT_ASSIGNED TRUE FALSE TRUE FALSE FALSE S_SWAPPC_B64 Store the address of the next instruction to a scalar register and then jump to an address specified in the scalar input. ENC_SOP1 default 30 SDST FMT_NUM_B64 OPR_SDST 64 SSRC0 FMT_NUM_B64 OPR_SREG 64 FMT_NUM_B64 OPR_PC 64 FMT_NUM_B64 OPR_PC 64 SALU NOT_ASSIGNED TRUE FALSE TRUE FALSE FALSE S_RFE_B64 Return from the exception handler. Clear the wave's PRIV bit and then jump to an address specified by the scalar input. ENC_SOP1 default 31 SSRC0 FMT_NUM_B64 OPR_SREG 64 FMT_NUM_B64 OPR_PC 64 TRAP NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_AND_SAVEEXEC_B64 Calculate bitwise AND on the scalar input and the EXEC mask, store the calculated result into the EXEC mask, set SCC iff the calculated result is nonzero and store the original value of the EXEC mask into the scalar destination register. ENC_SOP1 default 32 SDST FMT_NUM_U64 OPR_SREG 64 SSRC0 FMT_NUM_U64 OPR_SSRC 64 FMT_NUM_M64 OPR_SDST_EXEC 64 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 FMT_NUM_M64 OPR_SDST_EXEC 64 SOP1_INST_LITERAL has_lit_0 32 SDST FMT_NUM_U64 OPR_SREG 64 SIMM32 FMT_NUM_U64 OPR_SSRC 64 FMT_NUM_M64 OPR_SDST_EXEC 64 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 FMT_NUM_M64 OPR_SDST_EXEC 64 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_OR_SAVEEXEC_B64 Calculate bitwise OR on the scalar input and the EXEC mask, store the calculated result into the EXEC mask, set SCC iff the calculated result is nonzero and store the original value of the EXEC mask into the scalar destination register. ENC_SOP1 default 33 SDST FMT_NUM_U64 OPR_SREG 64 SSRC0 FMT_NUM_U64 OPR_SSRC 64 FMT_NUM_M64 OPR_SDST_EXEC 64 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 FMT_NUM_M64 OPR_SDST_EXEC 64 SOP1_INST_LITERAL has_lit_0 33 SDST FMT_NUM_U64 OPR_SREG 64 SIMM32 FMT_NUM_U64 OPR_SSRC 64 FMT_NUM_M64 OPR_SDST_EXEC 64 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 FMT_NUM_M64 OPR_SDST_EXEC 64 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_XOR_SAVEEXEC_B64 Calculate bitwise XOR on the scalar input and the EXEC mask, store the calculated result into the EXEC mask, set SCC iff the calculated result is nonzero and store the original value of the EXEC mask into the scalar destination register. ENC_SOP1 default 34 SDST FMT_NUM_U64 OPR_SREG 64 SSRC0 FMT_NUM_U64 OPR_SSRC 64 FMT_NUM_M64 OPR_SDST_EXEC 64 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 FMT_NUM_M64 OPR_SDST_EXEC 64 SOP1_INST_LITERAL has_lit_0 34 SDST FMT_NUM_U64 OPR_SREG 64 SIMM32 FMT_NUM_U64 OPR_SSRC 64 FMT_NUM_M64 OPR_SDST_EXEC 64 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 FMT_NUM_M64 OPR_SDST_EXEC 64 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_ANDN2_SAVEEXEC_B64 Calculate bitwise AND on the scalar input and the negation of the EXEC mask, store the calculated result into the EXEC mask, set SCC iff the calculated result is nonzero and store the original value of the EXEC mask into the scalar destination register. ENC_SOP1 default 35 SDST FMT_NUM_U64 OPR_SREG 64 SSRC0 FMT_NUM_U64 OPR_SSRC 64 FMT_NUM_M64 OPR_SDST_EXEC 64 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 FMT_NUM_M64 OPR_SDST_EXEC 64 SOP1_INST_LITERAL has_lit_0 35 SDST FMT_NUM_U64 OPR_SREG 64 SIMM32 FMT_NUM_U64 OPR_SSRC 64 FMT_NUM_M64 OPR_SDST_EXEC 64 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 FMT_NUM_M64 OPR_SDST_EXEC 64 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_ORN2_SAVEEXEC_B64 Calculate bitwise OR on the scalar input and the negation of the EXEC mask, store the calculated result into the EXEC mask, set SCC iff the calculated result is nonzero and store the original value of the EXEC mask into the scalar destination register. ENC_SOP1 default 36 SDST FMT_NUM_U64 OPR_SREG 64 SSRC0 FMT_NUM_U64 OPR_SSRC 64 FMT_NUM_M64 OPR_SDST_EXEC 64 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 FMT_NUM_M64 OPR_SDST_EXEC 64 SOP1_INST_LITERAL has_lit_0 36 SDST FMT_NUM_U64 OPR_SREG 64 SIMM32 FMT_NUM_U64 OPR_SSRC 64 FMT_NUM_M64 OPR_SDST_EXEC 64 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 FMT_NUM_M64 OPR_SDST_EXEC 64 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_NAND_SAVEEXEC_B64 Calculate bitwise NAND on the scalar input and the EXEC mask, store the calculated result into the EXEC mask, set SCC iff the calculated result is nonzero and store the original value of the EXEC mask into the scalar destination register. ENC_SOP1 default 37 SDST FMT_NUM_U64 OPR_SREG 64 SSRC0 FMT_NUM_U64 OPR_SSRC 64 FMT_NUM_M64 OPR_SDST_EXEC 64 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 FMT_NUM_M64 OPR_SDST_EXEC 64 SOP1_INST_LITERAL has_lit_0 37 SDST FMT_NUM_U64 OPR_SREG 64 SIMM32 FMT_NUM_U64 OPR_SSRC 64 FMT_NUM_M64 OPR_SDST_EXEC 64 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 FMT_NUM_M64 OPR_SDST_EXEC 64 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_NOR_SAVEEXEC_B64 Calculate bitwise NOR on the scalar input and the EXEC mask, store the calculated result into the EXEC mask, set SCC iff the calculated result is nonzero and store the original value of the EXEC mask into the scalar destination register. ENC_SOP1 default 38 SDST FMT_NUM_U64 OPR_SREG 64 SSRC0 FMT_NUM_U64 OPR_SSRC 64 FMT_NUM_M64 OPR_SDST_EXEC 64 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 FMT_NUM_M64 OPR_SDST_EXEC 64 SOP1_INST_LITERAL has_lit_0 38 SDST FMT_NUM_U64 OPR_SREG 64 SIMM32 FMT_NUM_U64 OPR_SSRC 64 FMT_NUM_M64 OPR_SDST_EXEC 64 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 FMT_NUM_M64 OPR_SDST_EXEC 64 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_XNOR_SAVEEXEC_B64 Calculate bitwise XNOR on the scalar input and the EXEC mask, store the calculated result into the EXEC mask, set SCC iff the calculated result is nonzero and store the original value of the EXEC mask into the scalar destination register. ENC_SOP1 default 39 SDST FMT_NUM_U64 OPR_SREG 64 SSRC0 FMT_NUM_U64 OPR_SSRC 64 FMT_NUM_M64 OPR_SDST_EXEC 64 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 FMT_NUM_M64 OPR_SDST_EXEC 64 SOP1_INST_LITERAL has_lit_0 39 SDST FMT_NUM_U64 OPR_SREG 64 SIMM32 FMT_NUM_U64 OPR_SSRC 64 FMT_NUM_M64 OPR_SDST_EXEC 64 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 FMT_NUM_M64 OPR_SDST_EXEC 64 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_QUADMASK_B32 Reduce a pixel mask from the scalar input into a quad mask, store the result in a scalar register and set SCC iff the result is nonzero. ENC_SOP1 default 40 SDST FMT_NUM_U32 OPR_SDST 32 SSRC0 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP1_INST_LITERAL has_lit_0 40 SDST FMT_NUM_U32 OPR_SDST 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_QUADMASK_B64 Reduce a pixel mask from the scalar input into a quad mask, store the result in a scalar register and set SCC iff the result is nonzero. ENC_SOP1 default 41 SDST FMT_NUM_U64 OPR_SDST 64 SSRC0 FMT_NUM_U64 OPR_SSRC 64 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP1_INST_LITERAL has_lit_0 41 SDST FMT_NUM_U64 OPR_SDST 64 SIMM32 FMT_NUM_U64 OPR_SSRC 64 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_MOVRELS_B32 Move data from a relatively-indexed scalar register into another scalar register. ENC_SOP1 default 42 SDST FMT_NUM_B32 OPR_SDST 32 SSRC0 FMT_NUM_B32 OPR_SREG 32 FMT_NUM_B32 OPR_SDST_M0 32 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_MOVRELS_B64 Move data from a relatively-indexed scalar register into another scalar register. ENC_SOP1 default 43 SDST FMT_NUM_B64 OPR_SDST 64 SSRC0 FMT_NUM_B64 OPR_SREG 64 FMT_NUM_B32 OPR_SDST_M0 32 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_MOVRELD_B32 Move data from a scalar input into a relatively-indexed scalar register. ENC_SOP1 default 44 SDST FMT_NUM_B32 OPR_SREG 32 SSRC0 FMT_NUM_B32 OPR_SSRC 32 FMT_NUM_B32 OPR_SDST_M0 32 SOP1_INST_LITERAL has_lit_0 44 SDST FMT_NUM_B32 OPR_SREG 32 SIMM32 FMT_NUM_B32 OPR_SSRC 32 FMT_NUM_B32 OPR_SDST_M0 32 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_MOVRELD_B64 Move data from a scalar input into a relatively-indexed scalar register. ENC_SOP1 default 45 SDST FMT_NUM_B64 OPR_SREG 64 SSRC0 FMT_NUM_B64 OPR_SSRC 64 FMT_NUM_B32 OPR_SDST_M0 32 SOP1_INST_LITERAL has_lit_0 45 SDST FMT_NUM_B64 OPR_SREG 64 SIMM32 FMT_NUM_B64 OPR_SSRC 64 FMT_NUM_B32 OPR_SDST_M0 32 SALU NOT_ASSIGNED TRUE TRUE TRUE FALSE FALSE S_CBRANCH_JOIN Conditional branch join point (end of conditional branch block). ENC_SOP1 default 46 SSRC0 FMT_NUM_B32 OPR_SREG 32 FMT_NUM_M64 OPR_SDST_EXEC 64 FMT_NUM_B64 OPR_PC 64 SOP1_INST_LITERAL has_lit_0 46 SIMM32 FMT_NUM_B32 OPR_SREG 32 FMT_NUM_M64 OPR_SDST_EXEC 64 FMT_NUM_B64 OPR_PC 64 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_ABS_I32 Compute the absolute value of a scalar input, store the result into a scalar register and set SCC iff the result is nonzero. ENC_SOP1 default 48 SDST FMT_NUM_I32 OPR_SDST 32 SSRC0 FMT_NUM_I32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP1_INST_LITERAL has_lit_0 48 SDST FMT_NUM_I32 OPR_SDST 32 SIMM32 FMT_NUM_I32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_SET_GPR_IDX_IDX Set the index used in vector GPR indexing. ENC_SOP1 default 50 SSRC0 FMT_NUM_B32 OPR_SSRC 32 FMT_NUM_B32 OPR_SDST_M0 32 FMT_NUM_B32 OPR_SDST_M0 32 SOP1_INST_LITERAL has_lit_0 50 SIMM32 FMT_NUM_B32 OPR_SSRC 32 FMT_NUM_B32 OPR_SDST_M0 32 FMT_NUM_B32 OPR_SDST_M0 32 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_ANDN1_SAVEEXEC_B64 Calculate bitwise AND on the EXEC mask and the negation of the scalar input, store the calculated result into the EXEC mask, set SCC iff the calculated result is nonzero and store the original value of the EXEC mask into the scalar destination register. ENC_SOP1 default 51 SDST FMT_NUM_U64 OPR_SREG 64 SSRC0 FMT_NUM_U64 OPR_SSRC 64 FMT_NUM_M64 OPR_SDST_EXEC 64 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 FMT_NUM_M64 OPR_SDST_EXEC 64 SOP1_INST_LITERAL has_lit_0 51 SDST FMT_NUM_U64 OPR_SREG 64 SIMM32 FMT_NUM_U64 OPR_SSRC 64 FMT_NUM_M64 OPR_SDST_EXEC 64 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 FMT_NUM_M64 OPR_SDST_EXEC 64 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_ORN1_SAVEEXEC_B64 Calculate bitwise OR on the EXEC mask and the negation of the scalar input, store the calculated result into the EXEC mask, set SCC iff the calculated result is nonzero and store the original value of the EXEC mask into the scalar destination register. ENC_SOP1 default 52 SDST FMT_NUM_U64 OPR_SREG 64 SSRC0 FMT_NUM_U64 OPR_SSRC 64 FMT_NUM_M64 OPR_SDST_EXEC 64 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 FMT_NUM_M64 OPR_SDST_EXEC 64 SOP1_INST_LITERAL has_lit_0 52 SDST FMT_NUM_U64 OPR_SREG 64 SIMM32 FMT_NUM_U64 OPR_SSRC 64 FMT_NUM_M64 OPR_SDST_EXEC 64 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 FMT_NUM_M64 OPR_SDST_EXEC 64 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_ANDN1_WREXEC_B64 Calculate bitwise AND on the EXEC mask and the negation of the scalar input, store the calculated result into the EXEC mask and also into the scalar destination register, and set SCC iff the calculated result is nonzero. ENC_SOP1 default 53 SDST FMT_NUM_U64 OPR_SREG 64 SSRC0 FMT_NUM_U64 OPR_SSRC 64 FMT_NUM_M64 OPR_SDST_EXEC 64 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 FMT_NUM_M64 OPR_SDST_EXEC 64 SOP1_INST_LITERAL has_lit_0 53 SDST FMT_NUM_U64 OPR_SREG 64 SIMM32 FMT_NUM_U64 OPR_SSRC 64 FMT_NUM_M64 OPR_SDST_EXEC 64 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 FMT_NUM_M64 OPR_SDST_EXEC 64 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_ANDN2_WREXEC_B64 Calculate bitwise AND on the scalar input and the negation of the EXEC mask, store the calculated result into the EXEC mask and also into the scalar destination register, and set SCC iff the calculated result is nonzero. ENC_SOP1 default 54 SDST FMT_NUM_U64 OPR_SREG 64 SSRC0 FMT_NUM_U64 OPR_SSRC 64 FMT_NUM_M64 OPR_SDST_EXEC 64 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 FMT_NUM_M64 OPR_SDST_EXEC 64 SOP1_INST_LITERAL has_lit_0 54 SDST FMT_NUM_U64 OPR_SREG 64 SIMM32 FMT_NUM_U64 OPR_SSRC 64 FMT_NUM_M64 OPR_SDST_EXEC 64 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 FMT_NUM_M64 OPR_SDST_EXEC 64 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_BITREPLICATE_B64_B32 Substitute each bit of a 32 bit scalar input with two instances of itself and store the result into a 64 bit scalar register. ENC_SOP1 default 55 SDST FMT_NUM_U64 OPR_SDST 64 SSRC0 FMT_NUM_U32 OPR_SSRC 32 SOP1_INST_LITERAL has_lit_0 55 SDST FMT_NUM_U64 OPR_SDST 64 SIMM32 FMT_NUM_U32 OPR_SSRC 32 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_ADD_U32 Add two unsigned 32-bit integer inputs, store the result into a scalar register and store the carry-out bit into SCC. ENC_SOP2 default 0 SDST FMT_NUM_U32 OPR_SDST 32 SSRC0 FMT_NUM_U32 OPR_SSRC 32 SSRC1 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_0 0 SDST FMT_NUM_U32 OPR_SDST 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 SSRC1 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_1 0 SDST FMT_NUM_U32 OPR_SDST 32 SSRC0 FMT_NUM_U32 OPR_SSRC 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_0_has_lit_1 0 SDST FMT_NUM_U32 OPR_SDST 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_SUB_U32 Subtract the second unsigned 32-bit integer input from the first input, store the result into a scalar register and store the carry-out bit into SCC. ENC_SOP2 default 1 SDST FMT_NUM_U32 OPR_SDST 32 SSRC0 FMT_NUM_U32 OPR_SSRC 32 SSRC1 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_0 1 SDST FMT_NUM_U32 OPR_SDST 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 SSRC1 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_1 1 SDST FMT_NUM_U32 OPR_SDST 32 SSRC0 FMT_NUM_U32 OPR_SSRC 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_0_has_lit_1 1 SDST FMT_NUM_U32 OPR_SDST 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_ADD_I32 Add two signed 32-bit integer inputs, store the result into a scalar register and store the carry-out bit into SCC. ENC_SOP2 default 2 SDST FMT_NUM_I32 OPR_SDST 32 SSRC0 FMT_NUM_I32 OPR_SSRC 32 SSRC1 FMT_NUM_I32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_0 2 SDST FMT_NUM_I32 OPR_SDST 32 SIMM32 FMT_NUM_I32 OPR_SSRC 32 SSRC1 FMT_NUM_I32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_1 2 SDST FMT_NUM_I32 OPR_SDST 32 SSRC0 FMT_NUM_I32 OPR_SSRC 32 SIMM32 FMT_NUM_I32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_0_has_lit_1 2 SDST FMT_NUM_I32 OPR_SDST 32 SIMM32 FMT_NUM_I32 OPR_SSRC 32 SIMM32 FMT_NUM_I32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_SUB_I32 Subtract the second signed 32-bit integer input from the first input, store the result into a scalar register and store the carry-out bit into SCC. ENC_SOP2 default 3 SDST FMT_NUM_I32 OPR_SDST 32 SSRC0 FMT_NUM_I32 OPR_SSRC 32 SSRC1 FMT_NUM_I32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_0 3 SDST FMT_NUM_I32 OPR_SDST 32 SIMM32 FMT_NUM_I32 OPR_SSRC 32 SSRC1 FMT_NUM_I32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_1 3 SDST FMT_NUM_I32 OPR_SDST 32 SSRC0 FMT_NUM_I32 OPR_SSRC 32 SIMM32 FMT_NUM_I32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_0_has_lit_1 3 SDST FMT_NUM_I32 OPR_SDST 32 SIMM32 FMT_NUM_I32 OPR_SSRC 32 SIMM32 FMT_NUM_I32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_ADDC_U32 Add two unsigned 32-bit integer inputs and a carry-in bit from SCC, store the result into a scalar register and store the carry-out bit into SCC. ENC_SOP2 default 4 SDST FMT_NUM_U32 OPR_SDST 32 SSRC0 FMT_NUM_U32 OPR_SSRC 32 SSRC1 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_0 4 SDST FMT_NUM_U32 OPR_SDST 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 SSRC1 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_1 4 SDST FMT_NUM_U32 OPR_SDST 32 SSRC0 FMT_NUM_U32 OPR_SSRC 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_0_has_lit_1 4 SDST FMT_NUM_U32 OPR_SDST 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_SUBB_U32 Subtract the second unsigned 32-bit integer input from the first input, subtract the carry-in bit, store the result into a scalar register and store the carry-out bit into SCC. ENC_SOP2 default 5 SDST FMT_NUM_U32 OPR_SDST 32 SSRC0 FMT_NUM_U32 OPR_SSRC 32 SSRC1 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_0 5 SDST FMT_NUM_U32 OPR_SDST 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 SSRC1 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_1 5 SDST FMT_NUM_U32 OPR_SDST 32 SSRC0 FMT_NUM_U32 OPR_SSRC 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_0_has_lit_1 5 SDST FMT_NUM_U32 OPR_SDST 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_MIN_I32 Select the minimum of two signed 32-bit integer inputs, store the selected value into a scalar register and set SCC iff the first value is selected. ENC_SOP2 default 6 SDST FMT_NUM_I32 OPR_SDST 32 SSRC0 FMT_NUM_I32 OPR_SSRC 32 SSRC1 FMT_NUM_I32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_0 6 SDST FMT_NUM_I32 OPR_SDST 32 SIMM32 FMT_NUM_I32 OPR_SSRC 32 SSRC1 FMT_NUM_I32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_1 6 SDST FMT_NUM_I32 OPR_SDST 32 SSRC0 FMT_NUM_I32 OPR_SSRC 32 SIMM32 FMT_NUM_I32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_0_has_lit_1 6 SDST FMT_NUM_I32 OPR_SDST 32 SIMM32 FMT_NUM_I32 OPR_SSRC 32 SIMM32 FMT_NUM_I32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_MIN_U32 Select the minimum of two unsigned 32-bit integer inputs, store the selected value into a scalar register and set SCC iff the first value is selected. ENC_SOP2 default 7 SDST FMT_NUM_U32 OPR_SDST 32 SSRC0 FMT_NUM_U32 OPR_SSRC 32 SSRC1 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_0 7 SDST FMT_NUM_U32 OPR_SDST 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 SSRC1 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_1 7 SDST FMT_NUM_U32 OPR_SDST 32 SSRC0 FMT_NUM_U32 OPR_SSRC 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_0_has_lit_1 7 SDST FMT_NUM_U32 OPR_SDST 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_MAX_I32 Select the maximum of two signed 32-bit integer inputs, store the selected value into a scalar register and set SCC iff the first value is selected. ENC_SOP2 default 8 SDST FMT_NUM_I32 OPR_SDST 32 SSRC0 FMT_NUM_I32 OPR_SSRC 32 SSRC1 FMT_NUM_I32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_0 8 SDST FMT_NUM_I32 OPR_SDST 32 SIMM32 FMT_NUM_I32 OPR_SSRC 32 SSRC1 FMT_NUM_I32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_1 8 SDST FMT_NUM_I32 OPR_SDST 32 SSRC0 FMT_NUM_I32 OPR_SSRC 32 SIMM32 FMT_NUM_I32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_0_has_lit_1 8 SDST FMT_NUM_I32 OPR_SDST 32 SIMM32 FMT_NUM_I32 OPR_SSRC 32 SIMM32 FMT_NUM_I32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_MAX_U32 Select the maximum of two unsigned 32-bit integer inputs, store the selected value into a scalar register and set SCC iff the first value is selected. ENC_SOP2 default 9 SDST FMT_NUM_U32 OPR_SDST 32 SSRC0 FMT_NUM_U32 OPR_SSRC 32 SSRC1 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_0 9 SDST FMT_NUM_U32 OPR_SDST 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 SSRC1 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_1 9 SDST FMT_NUM_U32 OPR_SDST 32 SSRC0 FMT_NUM_U32 OPR_SSRC 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_0_has_lit_1 9 SDST FMT_NUM_U32 OPR_SDST 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_CSELECT_B32 Select the first input if SCC is true otherwise select the second input, then store the selected input into a scalar register. ENC_SOP2 default 10 SDST FMT_NUM_B32 OPR_SDST 32 SSRC0 FMT_NUM_B32 OPR_SSRC 32 SSRC1 FMT_NUM_B32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_0 10 SDST FMT_NUM_B32 OPR_SDST 32 SIMM32 FMT_NUM_B32 OPR_SSRC 32 SSRC1 FMT_NUM_B32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_1 10 SDST FMT_NUM_B32 OPR_SDST 32 SSRC0 FMT_NUM_B32 OPR_SSRC 32 SIMM32 FMT_NUM_B32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_0_has_lit_1 10 SDST FMT_NUM_B32 OPR_SDST 32 SIMM32 FMT_NUM_B32 OPR_SSRC 32 SIMM32 FMT_NUM_B32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_CSELECT_B64 Select the first input if SCC is true otherwise select the second input, then store the selected input into a scalar register. ENC_SOP2 default 11 SDST FMT_NUM_B64 OPR_SDST 64 SSRC0 FMT_NUM_B64 OPR_SSRC 64 SSRC1 FMT_NUM_B64 OPR_SSRC 64 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_0 11 SDST FMT_NUM_B64 OPR_SDST 64 SIMM32 FMT_NUM_B64 OPR_SSRC 64 SSRC1 FMT_NUM_B64 OPR_SSRC 64 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_1 11 SDST FMT_NUM_B64 OPR_SDST 64 SSRC0 FMT_NUM_B64 OPR_SSRC 64 SIMM32 FMT_NUM_B64 OPR_SSRC 64 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_0_has_lit_1 11 SDST FMT_NUM_B64 OPR_SDST 64 SIMM32 FMT_NUM_B64 OPR_SSRC 64 SIMM32 FMT_NUM_B64 OPR_SSRC 64 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_AND_B32 Calculate bitwise AND on two scalar inputs, store the result into a scalar register and set SCC iff the result is nonzero. ENC_SOP2 default 12 SDST FMT_NUM_U32 OPR_SDST 32 SSRC0 FMT_NUM_U32 OPR_SSRC 32 SSRC1 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_0 12 SDST FMT_NUM_U32 OPR_SDST 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 SSRC1 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_1 12 SDST FMT_NUM_U32 OPR_SDST 32 SSRC0 FMT_NUM_U32 OPR_SSRC 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_0_has_lit_1 12 SDST FMT_NUM_U32 OPR_SDST 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_AND_B64 Calculate bitwise AND on two scalar inputs, store the result into a scalar register and set SCC iff the result is nonzero. ENC_SOP2 default 13 SDST FMT_NUM_U64 OPR_SDST 64 SSRC0 FMT_NUM_U64 OPR_SSRC 64 SSRC1 FMT_NUM_U64 OPR_SSRC 64 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_0 13 SDST FMT_NUM_U64 OPR_SDST 64 SIMM32 FMT_NUM_U64 OPR_SSRC 64 SSRC1 FMT_NUM_U64 OPR_SSRC 64 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_1 13 SDST FMT_NUM_U64 OPR_SDST 64 SSRC0 FMT_NUM_U64 OPR_SSRC 64 SIMM32 FMT_NUM_U64 OPR_SSRC 64 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_0_has_lit_1 13 SDST FMT_NUM_U64 OPR_SDST 64 SIMM32 FMT_NUM_U64 OPR_SSRC 64 SIMM32 FMT_NUM_U64 OPR_SSRC 64 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_OR_B32 Calculate bitwise OR on two scalar inputs, store the result into a scalar register and set SCC iff the result is nonzero. ENC_SOP2 default 14 SDST FMT_NUM_U32 OPR_SDST 32 SSRC0 FMT_NUM_U32 OPR_SSRC 32 SSRC1 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_0 14 SDST FMT_NUM_U32 OPR_SDST 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 SSRC1 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_1 14 SDST FMT_NUM_U32 OPR_SDST 32 SSRC0 FMT_NUM_U32 OPR_SSRC 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_0_has_lit_1 14 SDST FMT_NUM_U32 OPR_SDST 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_OR_B64 Calculate bitwise OR on two scalar inputs, store the result into a scalar register and set SCC iff the result is nonzero. ENC_SOP2 default 15 SDST FMT_NUM_U64 OPR_SDST 64 SSRC0 FMT_NUM_U64 OPR_SSRC 64 SSRC1 FMT_NUM_U64 OPR_SSRC 64 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_0 15 SDST FMT_NUM_U64 OPR_SDST 64 SIMM32 FMT_NUM_U64 OPR_SSRC 64 SSRC1 FMT_NUM_U64 OPR_SSRC 64 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_1 15 SDST FMT_NUM_U64 OPR_SDST 64 SSRC0 FMT_NUM_U64 OPR_SSRC 64 SIMM32 FMT_NUM_U64 OPR_SSRC 64 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_0_has_lit_1 15 SDST FMT_NUM_U64 OPR_SDST 64 SIMM32 FMT_NUM_U64 OPR_SSRC 64 SIMM32 FMT_NUM_U64 OPR_SSRC 64 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_XOR_B32 Calculate bitwise XOR on two scalar inputs, store the result into a scalar register and set SCC iff the result is nonzero. ENC_SOP2 default 16 SDST FMT_NUM_U32 OPR_SDST 32 SSRC0 FMT_NUM_U32 OPR_SSRC 32 SSRC1 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_0 16 SDST FMT_NUM_U32 OPR_SDST 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 SSRC1 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_1 16 SDST FMT_NUM_U32 OPR_SDST 32 SSRC0 FMT_NUM_U32 OPR_SSRC 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_0_has_lit_1 16 SDST FMT_NUM_U32 OPR_SDST 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_XOR_B64 Calculate bitwise XOR on two scalar inputs, store the result into a scalar register and set SCC iff the result is nonzero. ENC_SOP2 default 17 SDST FMT_NUM_U64 OPR_SDST 64 SSRC0 FMT_NUM_U64 OPR_SSRC 64 SSRC1 FMT_NUM_U64 OPR_SSRC 64 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_0 17 SDST FMT_NUM_U64 OPR_SDST 64 SIMM32 FMT_NUM_U64 OPR_SSRC 64 SSRC1 FMT_NUM_U64 OPR_SSRC 64 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_1 17 SDST FMT_NUM_U64 OPR_SDST 64 SSRC0 FMT_NUM_U64 OPR_SSRC 64 SIMM32 FMT_NUM_U64 OPR_SSRC 64 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_0_has_lit_1 17 SDST FMT_NUM_U64 OPR_SDST 64 SIMM32 FMT_NUM_U64 OPR_SSRC 64 SIMM32 FMT_NUM_U64 OPR_SSRC 64 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_ANDN2_B32 Calculate bitwise AND with the first input and the negation of the second input, store the result into a scalar register and set SCC if the result is nonzero. ENC_SOP2 default 18 SDST FMT_NUM_U32 OPR_SDST 32 SSRC0 FMT_NUM_U32 OPR_SSRC 32 SSRC1 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_0 18 SDST FMT_NUM_U32 OPR_SDST 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 SSRC1 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_1 18 SDST FMT_NUM_U32 OPR_SDST 32 SSRC0 FMT_NUM_U32 OPR_SSRC 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_0_has_lit_1 18 SDST FMT_NUM_U32 OPR_SDST 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_ANDN2_B64 Calculate bitwise AND with the first input and the negation of the second input, store the result into a scalar register and set SCC if the result is nonzero. ENC_SOP2 default 19 SDST FMT_NUM_U64 OPR_SDST 64 SSRC0 FMT_NUM_U64 OPR_SSRC 64 SSRC1 FMT_NUM_U64 OPR_SSRC 64 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_0 19 SDST FMT_NUM_U64 OPR_SDST 64 SIMM32 FMT_NUM_U64 OPR_SSRC 64 SSRC1 FMT_NUM_U64 OPR_SSRC 64 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_1 19 SDST FMT_NUM_U64 OPR_SDST 64 SSRC0 FMT_NUM_U64 OPR_SSRC 64 SIMM32 FMT_NUM_U64 OPR_SSRC 64 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_0_has_lit_1 19 SDST FMT_NUM_U64 OPR_SDST 64 SIMM32 FMT_NUM_U64 OPR_SSRC 64 SIMM32 FMT_NUM_U64 OPR_SSRC 64 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_ORN2_B32 Calculate bitwise OR with the first input and the negation of the second input, store the result into a scalar register and set SCC if the result is nonzero. ENC_SOP2 default 20 SDST FMT_NUM_U32 OPR_SDST 32 SSRC0 FMT_NUM_U32 OPR_SSRC 32 SSRC1 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_0 20 SDST FMT_NUM_U32 OPR_SDST 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 SSRC1 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_1 20 SDST FMT_NUM_U32 OPR_SDST 32 SSRC0 FMT_NUM_U32 OPR_SSRC 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_0_has_lit_1 20 SDST FMT_NUM_U32 OPR_SDST 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_ORN2_B64 Calculate bitwise OR with the first input and the negation of the second input, store the result into a scalar register and set SCC if the result is nonzero. ENC_SOP2 default 21 SDST FMT_NUM_U64 OPR_SDST 64 SSRC0 FMT_NUM_U64 OPR_SSRC 64 SSRC1 FMT_NUM_U64 OPR_SSRC 64 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_0 21 SDST FMT_NUM_U64 OPR_SDST 64 SIMM32 FMT_NUM_U64 OPR_SSRC 64 SSRC1 FMT_NUM_U64 OPR_SSRC 64 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_1 21 SDST FMT_NUM_U64 OPR_SDST 64 SSRC0 FMT_NUM_U64 OPR_SSRC 64 SIMM32 FMT_NUM_U64 OPR_SSRC 64 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_0_has_lit_1 21 SDST FMT_NUM_U64 OPR_SDST 64 SIMM32 FMT_NUM_U64 OPR_SSRC 64 SIMM32 FMT_NUM_U64 OPR_SSRC 64 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_NAND_B32 Calculate bitwise NAND on two scalar inputs, store the result into a scalar register and set SCC if the result is nonzero. ENC_SOP2 default 22 SDST FMT_NUM_U32 OPR_SDST 32 SSRC0 FMT_NUM_U32 OPR_SSRC 32 SSRC1 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_0 22 SDST FMT_NUM_U32 OPR_SDST 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 SSRC1 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_1 22 SDST FMT_NUM_U32 OPR_SDST 32 SSRC0 FMT_NUM_U32 OPR_SSRC 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_0_has_lit_1 22 SDST FMT_NUM_U32 OPR_SDST 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_NAND_B64 Calculate bitwise NAND on two scalar inputs, store the result into a scalar register and set SCC if the result is nonzero. ENC_SOP2 default 23 SDST FMT_NUM_U64 OPR_SDST 64 SSRC0 FMT_NUM_U64 OPR_SSRC 64 SSRC1 FMT_NUM_U64 OPR_SSRC 64 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_0 23 SDST FMT_NUM_U64 OPR_SDST 64 SIMM32 FMT_NUM_U64 OPR_SSRC 64 SSRC1 FMT_NUM_U64 OPR_SSRC 64 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_1 23 SDST FMT_NUM_U64 OPR_SDST 64 SSRC0 FMT_NUM_U64 OPR_SSRC 64 SIMM32 FMT_NUM_U64 OPR_SSRC 64 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_0_has_lit_1 23 SDST FMT_NUM_U64 OPR_SDST 64 SIMM32 FMT_NUM_U64 OPR_SSRC 64 SIMM32 FMT_NUM_U64 OPR_SSRC 64 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_NOR_B32 Calculate bitwise NOR on two scalar inputs, store the result into a scalar register and set SCC if the result is nonzero. ENC_SOP2 default 24 SDST FMT_NUM_U32 OPR_SDST 32 SSRC0 FMT_NUM_U32 OPR_SSRC 32 SSRC1 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_0 24 SDST FMT_NUM_U32 OPR_SDST 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 SSRC1 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_1 24 SDST FMT_NUM_U32 OPR_SDST 32 SSRC0 FMT_NUM_U32 OPR_SSRC 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_0_has_lit_1 24 SDST FMT_NUM_U32 OPR_SDST 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_NOR_B64 Calculate bitwise NOR on two scalar inputs, store the result into a scalar register and set SCC if the result is nonzero. ENC_SOP2 default 25 SDST FMT_NUM_U64 OPR_SDST 64 SSRC0 FMT_NUM_U64 OPR_SSRC 64 SSRC1 FMT_NUM_U64 OPR_SSRC 64 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_0 25 SDST FMT_NUM_U64 OPR_SDST 64 SIMM32 FMT_NUM_U64 OPR_SSRC 64 SSRC1 FMT_NUM_U64 OPR_SSRC 64 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_1 25 SDST FMT_NUM_U64 OPR_SDST 64 SSRC0 FMT_NUM_U64 OPR_SSRC 64 SIMM32 FMT_NUM_U64 OPR_SSRC 64 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_0_has_lit_1 25 SDST FMT_NUM_U64 OPR_SDST 64 SIMM32 FMT_NUM_U64 OPR_SSRC 64 SIMM32 FMT_NUM_U64 OPR_SSRC 64 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_XNOR_B32 Calculate bitwise XNOR on two scalar inputs, store the result into a scalar register and set SCC if the result is nonzero. ENC_SOP2 default 26 SDST FMT_NUM_U32 OPR_SDST 32 SSRC0 FMT_NUM_U32 OPR_SSRC 32 SSRC1 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_0 26 SDST FMT_NUM_U32 OPR_SDST 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 SSRC1 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_1 26 SDST FMT_NUM_U32 OPR_SDST 32 SSRC0 FMT_NUM_U32 OPR_SSRC 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_0_has_lit_1 26 SDST FMT_NUM_U32 OPR_SDST 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_XNOR_B64 Calculate bitwise XNOR on two scalar inputs, store the result into a scalar register and set SCC if the result is nonzero. ENC_SOP2 default 27 SDST FMT_NUM_U64 OPR_SDST 64 SSRC0 FMT_NUM_U64 OPR_SSRC 64 SSRC1 FMT_NUM_U64 OPR_SSRC 64 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_0 27 SDST FMT_NUM_U64 OPR_SDST 64 SIMM32 FMT_NUM_U64 OPR_SSRC 64 SSRC1 FMT_NUM_U64 OPR_SSRC 64 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_1 27 SDST FMT_NUM_U64 OPR_SDST 64 SSRC0 FMT_NUM_U64 OPR_SSRC 64 SIMM32 FMT_NUM_U64 OPR_SSRC 64 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_0_has_lit_1 27 SDST FMT_NUM_U64 OPR_SDST 64 SIMM32 FMT_NUM_U64 OPR_SSRC 64 SIMM32 FMT_NUM_U64 OPR_SSRC 64 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_LSHL_B32 Given a shift count in the second scalar input, calculate the logical shift left of the first scalar input, store the result into a scalar register and set SCC iff the result is nonzero. ENC_SOP2 default 28 SDST FMT_NUM_U32 OPR_SDST 32 SSRC0 FMT_NUM_U32 OPR_SSRC 32 SSRC1 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_0 28 SDST FMT_NUM_U32 OPR_SDST 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 SSRC1 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_1 28 SDST FMT_NUM_U32 OPR_SDST 32 SSRC0 FMT_NUM_U32 OPR_SSRC 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_0_has_lit_1 28 SDST FMT_NUM_U32 OPR_SDST 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_LSHL_B64 Given a shift count in the second scalar input, calculate the logical shift left of the first scalar input, store the result into a scalar register and set SCC iff the result is nonzero. ENC_SOP2 default 29 SDST FMT_NUM_U64 OPR_SDST 64 SSRC0 FMT_NUM_U64 OPR_SSRC 64 SSRC1 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_0 29 SDST FMT_NUM_U64 OPR_SDST 64 SIMM32 FMT_NUM_U64 OPR_SSRC 64 SSRC1 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_1 29 SDST FMT_NUM_U64 OPR_SDST 64 SSRC0 FMT_NUM_U64 OPR_SSRC 64 SIMM32 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_0_has_lit_1 29 SDST FMT_NUM_U64 OPR_SDST 64 SIMM32 FMT_NUM_U64 OPR_SSRC 64 SIMM32 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_LSHR_B32 Given a shift count in the second scalar input, calculate the logical shift right of the first scalar input, store the result into a scalar register and set SCC iff the result is nonzero. ENC_SOP2 default 30 SDST FMT_NUM_U32 OPR_SDST 32 SSRC0 FMT_NUM_U32 OPR_SSRC 32 SSRC1 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_0 30 SDST FMT_NUM_U32 OPR_SDST 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 SSRC1 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_1 30 SDST FMT_NUM_U32 OPR_SDST 32 SSRC0 FMT_NUM_U32 OPR_SSRC 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_0_has_lit_1 30 SDST FMT_NUM_U32 OPR_SDST 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_LSHR_B64 Given a shift count in the second scalar input, calculate the logical shift right of the first scalar input, store the result into a scalar register and set SCC iff the result is nonzero. ENC_SOP2 default 31 SDST FMT_NUM_U64 OPR_SDST 64 SSRC0 FMT_NUM_U64 OPR_SSRC 64 SSRC1 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_0 31 SDST FMT_NUM_U64 OPR_SDST 64 SIMM32 FMT_NUM_U64 OPR_SSRC 64 SSRC1 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_1 31 SDST FMT_NUM_U64 OPR_SDST 64 SSRC0 FMT_NUM_U64 OPR_SSRC 64 SIMM32 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_0_has_lit_1 31 SDST FMT_NUM_U64 OPR_SDST 64 SIMM32 FMT_NUM_U64 OPR_SSRC 64 SIMM32 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_ASHR_I32 Given a shift count in the second scalar input, calculate the arithmetic shift right (preserving sign bit) of the first scalar input, store the result into a scalar register and set SCC iff the result is nonzero. ENC_SOP2 default 32 SDST FMT_NUM_I32 OPR_SDST 32 SSRC0 FMT_NUM_I32 OPR_SSRC 32 SSRC1 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_0 32 SDST FMT_NUM_I32 OPR_SDST 32 SIMM32 FMT_NUM_I32 OPR_SSRC 32 SSRC1 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_1 32 SDST FMT_NUM_I32 OPR_SDST 32 SSRC0 FMT_NUM_I32 OPR_SSRC 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_0_has_lit_1 32 SDST FMT_NUM_I32 OPR_SDST 32 SIMM32 FMT_NUM_I32 OPR_SSRC 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_ASHR_I64 Given a shift count in the second scalar input, calculate the arithmetic shift right (preserving sign bit) of the first scalar input, store the result into a scalar register and set SCC iff the result is nonzero. ENC_SOP2 default 33 SDST FMT_NUM_I64 OPR_SDST 64 SSRC0 FMT_NUM_I64 OPR_SSRC 64 SSRC1 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_0 33 SDST FMT_NUM_I64 OPR_SDST 64 SIMM32 FMT_NUM_I64 OPR_SSRC 64 SSRC1 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_1 33 SDST FMT_NUM_I64 OPR_SDST 64 SSRC0 FMT_NUM_I64 OPR_SSRC 64 SIMM32 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_0_has_lit_1 33 SDST FMT_NUM_I64 OPR_SDST 64 SIMM32 FMT_NUM_I64 OPR_SSRC 64 SIMM32 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_BFM_B32 Calculate a bitfield mask given a field offset and size and store the result in a scalar register. ENC_SOP2 default 34 SDST FMT_NUM_U32 OPR_SDST 32 SSRC0 FMT_NUM_U32 OPR_SSRC 32 SSRC1 FMT_NUM_U32 OPR_SSRC 32 SOP2_INST_LITERAL has_lit_0 34 SDST FMT_NUM_U32 OPR_SDST 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 SSRC1 FMT_NUM_U32 OPR_SSRC 32 SOP2_INST_LITERAL has_lit_1 34 SDST FMT_NUM_U32 OPR_SDST 32 SSRC0 FMT_NUM_U32 OPR_SSRC 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 SOP2_INST_LITERAL has_lit_0_has_lit_1 34 SDST FMT_NUM_U32 OPR_SDST 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_BFM_B64 Calculate a bitfield mask given a field offset and size and store the result in a scalar register. ENC_SOP2 default 35 SDST FMT_NUM_U64 OPR_SDST 64 SSRC0 FMT_NUM_U32 OPR_SSRC 32 SSRC1 FMT_NUM_U32 OPR_SSRC 32 SOP2_INST_LITERAL has_lit_0 35 SDST FMT_NUM_U64 OPR_SDST 64 SIMM32 FMT_NUM_U32 OPR_SSRC 32 SSRC1 FMT_NUM_U32 OPR_SSRC 32 SOP2_INST_LITERAL has_lit_1 35 SDST FMT_NUM_U64 OPR_SDST 64 SSRC0 FMT_NUM_U32 OPR_SSRC 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 SOP2_INST_LITERAL has_lit_0_has_lit_1 35 SDST FMT_NUM_U64 OPR_SDST 64 SIMM32 FMT_NUM_U32 OPR_SSRC 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_MUL_I32 Multiply two signed 32-bit integer inputs and store the result into a scalar register. ENC_SOP2 default 36 SDST FMT_NUM_I32 OPR_SDST 32 SSRC0 FMT_NUM_I32 OPR_SSRC 32 SSRC1 FMT_NUM_I32 OPR_SSRC 32 SOP2_INST_LITERAL has_lit_0 36 SDST FMT_NUM_I32 OPR_SDST 32 SIMM32 FMT_NUM_I32 OPR_SSRC 32 SSRC1 FMT_NUM_I32 OPR_SSRC 32 SOP2_INST_LITERAL has_lit_1 36 SDST FMT_NUM_I32 OPR_SDST 32 SSRC0 FMT_NUM_I32 OPR_SSRC 32 SIMM32 FMT_NUM_I32 OPR_SSRC 32 SOP2_INST_LITERAL has_lit_0_has_lit_1 36 SDST FMT_NUM_I32 OPR_SDST 32 SIMM32 FMT_NUM_I32 OPR_SSRC 32 SIMM32 FMT_NUM_I32 OPR_SSRC 32 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_BFE_U32 Extract an unsigned bitfield from the first input using field offset and size encoded in the second input, store the result into a scalar register and set SCC iff the result is nonzero. ENC_SOP2 default 37 SDST FMT_NUM_U32 OPR_SDST 32 SSRC0 FMT_NUM_U32 OPR_SSRC 32 SSRC1 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_0 37 SDST FMT_NUM_U32 OPR_SDST 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 SSRC1 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_1 37 SDST FMT_NUM_U32 OPR_SDST 32 SSRC0 FMT_NUM_U32 OPR_SSRC 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_0_has_lit_1 37 SDST FMT_NUM_U32 OPR_SDST 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_BFE_I32 Extract a signed bitfield from the first input using field offset and size encoded in the second input, store the result into a scalar register and set SCC iff the result is nonzero. ENC_SOP2 default 38 SDST FMT_NUM_I32 OPR_SDST 32 SSRC0 FMT_NUM_I32 OPR_SSRC 32 SSRC1 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_0 38 SDST FMT_NUM_I32 OPR_SDST 32 SIMM32 FMT_NUM_I32 OPR_SSRC 32 SSRC1 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_1 38 SDST FMT_NUM_I32 OPR_SDST 32 SSRC0 FMT_NUM_I32 OPR_SSRC 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_0_has_lit_1 38 SDST FMT_NUM_I32 OPR_SDST 32 SIMM32 FMT_NUM_I32 OPR_SSRC 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_BFE_U64 Extract an unsigned bitfield from the first input using field offset and size encoded in the second input, store the result into a scalar register and set SCC iff the result is nonzero. ENC_SOP2 default 39 SDST FMT_NUM_U64 OPR_SDST 64 SSRC0 FMT_NUM_U64 OPR_SSRC 64 SSRC1 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_0 39 SDST FMT_NUM_U64 OPR_SDST 64 SIMM32 FMT_NUM_U64 OPR_SSRC 64 SSRC1 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_1 39 SDST FMT_NUM_U64 OPR_SDST 64 SSRC0 FMT_NUM_U64 OPR_SSRC 64 SIMM32 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_0_has_lit_1 39 SDST FMT_NUM_U64 OPR_SDST 64 SIMM32 FMT_NUM_U64 OPR_SSRC 64 SIMM32 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_BFE_I64 Extract a signed bitfield from the first input using field offset and size encoded in the second input, store the result into a scalar register and set SCC iff the result is nonzero. ENC_SOP2 default 40 SDST FMT_NUM_I64 OPR_SDST 64 SSRC0 FMT_NUM_I64 OPR_SSRC 64 SSRC1 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_0 40 SDST FMT_NUM_I64 OPR_SDST 64 SIMM32 FMT_NUM_I64 OPR_SSRC 64 SSRC1 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_1 40 SDST FMT_NUM_I64 OPR_SDST 64 SSRC0 FMT_NUM_I64 OPR_SSRC 64 SIMM32 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_0_has_lit_1 40 SDST FMT_NUM_I64 OPR_SDST 64 SIMM32 FMT_NUM_I64 OPR_SSRC 64 SIMM32 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SALU NOT_ASSIGNED TRUE TRUE TRUE FALSE FALSE S_CBRANCH_G_FORK Conditional branch using branch-stack. ENC_SOP2 default 41 SSRC0 FMT_NUM_B64 OPR_SSRC_NOLIT 64 SSRC1 FMT_NUM_B64 OPR_SSRC_NOLIT 64 FMT_NUM_B64 OPR_PC 64 SOP2_INST_LITERAL has_lit_0 41 SIMM32 FMT_NUM_B64 OPR_SSRC_NOLIT 64 SSRC1 FMT_NUM_B64 OPR_SSRC_NOLIT 64 FMT_NUM_B64 OPR_PC 64 SOP2_INST_LITERAL has_lit_1 41 SSRC0 FMT_NUM_B64 OPR_SSRC_NOLIT 64 SIMM32 FMT_NUM_B64 OPR_SSRC_NOLIT 64 FMT_NUM_B64 OPR_PC 64 SOP2_INST_LITERAL has_lit_0_has_lit_1 41 SIMM32 FMT_NUM_B64 OPR_SSRC_NOLIT 64 SIMM32 FMT_NUM_B64 OPR_SSRC_NOLIT 64 FMT_NUM_B64 OPR_PC 64 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_ABSDIFF_I32 Calculate the absolute value of difference between two scalar inputs, store the result into a scalar register and set SCC iff the result is nonzero. ENC_SOP2 default 42 SDST FMT_NUM_I32 OPR_SDST 32 SSRC0 FMT_NUM_I32 OPR_SSRC 32 SSRC1 FMT_NUM_I32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_0 42 SDST FMT_NUM_I32 OPR_SDST 32 SIMM32 FMT_NUM_I32 OPR_SSRC 32 SSRC1 FMT_NUM_I32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_1 42 SDST FMT_NUM_I32 OPR_SDST 32 SSRC0 FMT_NUM_I32 OPR_SSRC 32 SIMM32 FMT_NUM_I32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_0_has_lit_1 42 SDST FMT_NUM_I32 OPR_SDST 32 SIMM32 FMT_NUM_I32 OPR_SSRC 32 SIMM32 FMT_NUM_I32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SALU NOT_ASSIGNED TRUE FALSE TRUE FALSE FALSE S_RFE_RESTORE_B64 Return from exception handler and continue. This instruction may only be used within a trap handler. ENC_SOP2 default 43 SSRC0 FMT_NUM_B64 OPR_SSRC 64 SSRC1 FMT_NUM_B32 OPR_SSRC 32 FMT_NUM_B64 OPR_PC 64 TRAP NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_MUL_HI_U32 Multiply two unsigned integers and store the high 32 bits of the result into a scalar register. ENC_SOP2 default 44 SDST FMT_NUM_U32 OPR_SDST 32 SSRC0 FMT_NUM_U32 OPR_SSRC 32 SSRC1 FMT_NUM_U32 OPR_SSRC 32 SOP2_INST_LITERAL has_lit_0 44 SDST FMT_NUM_U32 OPR_SDST 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 SSRC1 FMT_NUM_U32 OPR_SSRC 32 SOP2_INST_LITERAL has_lit_1 44 SDST FMT_NUM_U32 OPR_SDST 32 SSRC0 FMT_NUM_U32 OPR_SSRC 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 SOP2_INST_LITERAL has_lit_0_has_lit_1 44 SDST FMT_NUM_U32 OPR_SDST 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_MUL_HI_I32 Multiply two signed integers and store the high 32 bits of the result into a scalar register. ENC_SOP2 default 45 SDST FMT_NUM_I32 OPR_SDST 32 SSRC0 FMT_NUM_I32 OPR_SSRC 32 SSRC1 FMT_NUM_I32 OPR_SSRC 32 SOP2_INST_LITERAL has_lit_0 45 SDST FMT_NUM_I32 OPR_SDST 32 SIMM32 FMT_NUM_I32 OPR_SSRC 32 SSRC1 FMT_NUM_I32 OPR_SSRC 32 SOP2_INST_LITERAL has_lit_1 45 SDST FMT_NUM_I32 OPR_SDST 32 SSRC0 FMT_NUM_I32 OPR_SSRC 32 SIMM32 FMT_NUM_I32 OPR_SSRC 32 SOP2_INST_LITERAL has_lit_0_has_lit_1 45 SDST FMT_NUM_I32 OPR_SDST 32 SIMM32 FMT_NUM_I32 OPR_SSRC 32 SIMM32 FMT_NUM_I32 OPR_SSRC 32 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_LSHL1_ADD_U32 Calculate the logical shift left of the first input by 1, then add the second input, store the result into a scalar register and set SCC iff the summation results in an unsigned overflow. ENC_SOP2 default 46 SDST FMT_NUM_U32 OPR_SDST 32 SSRC0 FMT_NUM_U32 OPR_SSRC 32 SSRC1 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_0 46 SDST FMT_NUM_U32 OPR_SDST 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 SSRC1 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_1 46 SDST FMT_NUM_U32 OPR_SDST 32 SSRC0 FMT_NUM_U32 OPR_SSRC 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_0_has_lit_1 46 SDST FMT_NUM_U32 OPR_SDST 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_LSHL2_ADD_U32 Calculate the logical shift left of the first input by 2, then add the second input, store the result into a scalar register and set SCC iff the summation results in an unsigned overflow. ENC_SOP2 default 47 SDST FMT_NUM_U32 OPR_SDST 32 SSRC0 FMT_NUM_U32 OPR_SSRC 32 SSRC1 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_0 47 SDST FMT_NUM_U32 OPR_SDST 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 SSRC1 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_1 47 SDST FMT_NUM_U32 OPR_SDST 32 SSRC0 FMT_NUM_U32 OPR_SSRC 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_0_has_lit_1 47 SDST FMT_NUM_U32 OPR_SDST 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_LSHL3_ADD_U32 Calculate the logical shift left of the first input by 3, then add the second input, store the result into a scalar register and set SCC iff the summation results in an unsigned overflow. ENC_SOP2 default 48 SDST FMT_NUM_U32 OPR_SDST 32 SSRC0 FMT_NUM_U32 OPR_SSRC 32 SSRC1 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_0 48 SDST FMT_NUM_U32 OPR_SDST 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 SSRC1 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_1 48 SDST FMT_NUM_U32 OPR_SDST 32 SSRC0 FMT_NUM_U32 OPR_SSRC 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_0_has_lit_1 48 SDST FMT_NUM_U32 OPR_SDST 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_LSHL4_ADD_U32 Calculate the logical shift left of the first input by 4, then add the second input, store the result into a scalar register and set SCC iff the summation results in an unsigned overflow. ENC_SOP2 default 49 SDST FMT_NUM_U32 OPR_SDST 32 SSRC0 FMT_NUM_U32 OPR_SSRC 32 SSRC1 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_0 49 SDST FMT_NUM_U32 OPR_SDST 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 SSRC1 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_1 49 SDST FMT_NUM_U32 OPR_SDST 32 SSRC0 FMT_NUM_U32 OPR_SSRC 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOP2_INST_LITERAL has_lit_0_has_lit_1 49 SDST FMT_NUM_U32 OPR_SDST 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_PACK_LL_B32_B16 Pack two 16-bit scalar values into a scalar register. ENC_SOP2 default 50 SDST FMT_NUM_B32 OPR_SDST 32 SSRC0 FMT_NUM_B16 OPR_SSRC 16 SSRC1 FMT_NUM_B16 OPR_SSRC 16 SOP2_INST_LITERAL has_lit_0 50 SDST FMT_NUM_B32 OPR_SDST 32 SIMM32 FMT_NUM_B16 OPR_SSRC 16 SSRC1 FMT_NUM_B16 OPR_SSRC 16 SOP2_INST_LITERAL has_lit_1 50 SDST FMT_NUM_B32 OPR_SDST 32 SSRC0 FMT_NUM_B16 OPR_SSRC 16 SIMM32 FMT_NUM_B16 OPR_SSRC 16 SOP2_INST_LITERAL has_lit_0_has_lit_1 50 SDST FMT_NUM_B32 OPR_SDST 32 SIMM32 FMT_NUM_B16 OPR_SSRC 16 SIMM32 FMT_NUM_B16 OPR_SSRC 16 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_PACK_LH_B32_B16 Pack two 16-bit scalar values into a scalar register. ENC_SOP2 default 51 SDST FMT_NUM_B32 OPR_SDST 32 SSRC0 FMT_NUM_B16 OPR_SSRC 16 SSRC1 FMT_NUM_B32 OPR_SSRC 32 SOP2_INST_LITERAL has_lit_0 51 SDST FMT_NUM_B32 OPR_SDST 32 SIMM32 FMT_NUM_B16 OPR_SSRC 16 SSRC1 FMT_NUM_B32 OPR_SSRC 32 SOP2_INST_LITERAL has_lit_1 51 SDST FMT_NUM_B32 OPR_SDST 32 SSRC0 FMT_NUM_B16 OPR_SSRC 16 SIMM32 FMT_NUM_B32 OPR_SSRC 32 SOP2_INST_LITERAL has_lit_0_has_lit_1 51 SDST FMT_NUM_B32 OPR_SDST 32 SIMM32 FMT_NUM_B16 OPR_SSRC 16 SIMM32 FMT_NUM_B32 OPR_SSRC 32 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_PACK_HH_B32_B16 Pack two 16-bit scalar values into a scalar register. ENC_SOP2 default 52 SDST FMT_NUM_B32 OPR_SDST 32 SSRC0 FMT_NUM_B32 OPR_SSRC 32 SSRC1 FMT_NUM_B32 OPR_SSRC 32 SOP2_INST_LITERAL has_lit_0 52 SDST FMT_NUM_B32 OPR_SDST 32 SIMM32 FMT_NUM_B32 OPR_SSRC 32 SSRC1 FMT_NUM_B32 OPR_SSRC 32 SOP2_INST_LITERAL has_lit_1 52 SDST FMT_NUM_B32 OPR_SDST 32 SSRC0 FMT_NUM_B32 OPR_SSRC 32 SIMM32 FMT_NUM_B32 OPR_SSRC 32 SOP2_INST_LITERAL has_lit_0_has_lit_1 52 SDST FMT_NUM_B32 OPR_SDST 32 SIMM32 FMT_NUM_B32 OPR_SSRC 32 SIMM32 FMT_NUM_B32 OPR_SSRC 32 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_CMP_EQ_I32 Set SCC to 1 iff the first scalar input is equal to the second scalar input. ENC_SOPC default 0 SSRC0 FMT_NUM_I32 OPR_SSRC 32 SSRC1 FMT_NUM_I32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOPC_INST_LITERAL has_lit_0 0 SIMM32 FMT_NUM_I32 OPR_SSRC 32 SSRC1 FMT_NUM_I32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOPC_INST_LITERAL has_lit_1 0 SSRC0 FMT_NUM_I32 OPR_SSRC 32 SIMM32 FMT_NUM_I32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOPC_INST_LITERAL has_lit_0_has_lit_1 0 SIMM32 FMT_NUM_I32 OPR_SSRC 32 SIMM32 FMT_NUM_I32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_CMP_LG_I32 Set SCC to 1 iff the first scalar input is less than or greater than the second scalar input. ENC_SOPC default 1 SSRC0 FMT_NUM_I32 OPR_SSRC 32 SSRC1 FMT_NUM_I32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOPC_INST_LITERAL has_lit_0 1 SIMM32 FMT_NUM_I32 OPR_SSRC 32 SSRC1 FMT_NUM_I32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOPC_INST_LITERAL has_lit_1 1 SSRC0 FMT_NUM_I32 OPR_SSRC 32 SIMM32 FMT_NUM_I32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOPC_INST_LITERAL has_lit_0_has_lit_1 1 SIMM32 FMT_NUM_I32 OPR_SSRC 32 SIMM32 FMT_NUM_I32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_CMP_GT_I32 Set SCC to 1 iff the first scalar input is greater than the second scalar input. ENC_SOPC default 2 SSRC0 FMT_NUM_I32 OPR_SSRC 32 SSRC1 FMT_NUM_I32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOPC_INST_LITERAL has_lit_0 2 SIMM32 FMT_NUM_I32 OPR_SSRC 32 SSRC1 FMT_NUM_I32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOPC_INST_LITERAL has_lit_1 2 SSRC0 FMT_NUM_I32 OPR_SSRC 32 SIMM32 FMT_NUM_I32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOPC_INST_LITERAL has_lit_0_has_lit_1 2 SIMM32 FMT_NUM_I32 OPR_SSRC 32 SIMM32 FMT_NUM_I32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_CMP_GE_I32 Set SCC to 1 iff the first scalar input is greater than or equal to the second scalar input. ENC_SOPC default 3 SSRC0 FMT_NUM_I32 OPR_SSRC 32 SSRC1 FMT_NUM_I32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOPC_INST_LITERAL has_lit_0 3 SIMM32 FMT_NUM_I32 OPR_SSRC 32 SSRC1 FMT_NUM_I32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOPC_INST_LITERAL has_lit_1 3 SSRC0 FMT_NUM_I32 OPR_SSRC 32 SIMM32 FMT_NUM_I32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOPC_INST_LITERAL has_lit_0_has_lit_1 3 SIMM32 FMT_NUM_I32 OPR_SSRC 32 SIMM32 FMT_NUM_I32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_CMP_LT_I32 Set SCC to 1 iff the first scalar input is less than the second scalar input. ENC_SOPC default 4 SSRC0 FMT_NUM_I32 OPR_SSRC 32 SSRC1 FMT_NUM_I32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOPC_INST_LITERAL has_lit_0 4 SIMM32 FMT_NUM_I32 OPR_SSRC 32 SSRC1 FMT_NUM_I32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOPC_INST_LITERAL has_lit_1 4 SSRC0 FMT_NUM_I32 OPR_SSRC 32 SIMM32 FMT_NUM_I32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOPC_INST_LITERAL has_lit_0_has_lit_1 4 SIMM32 FMT_NUM_I32 OPR_SSRC 32 SIMM32 FMT_NUM_I32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_CMP_LE_I32 Set SCC to 1 iff the first scalar input is less than or equal to the second scalar input. ENC_SOPC default 5 SSRC0 FMT_NUM_I32 OPR_SSRC 32 SSRC1 FMT_NUM_I32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOPC_INST_LITERAL has_lit_0 5 SIMM32 FMT_NUM_I32 OPR_SSRC 32 SSRC1 FMT_NUM_I32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOPC_INST_LITERAL has_lit_1 5 SSRC0 FMT_NUM_I32 OPR_SSRC 32 SIMM32 FMT_NUM_I32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOPC_INST_LITERAL has_lit_0_has_lit_1 5 SIMM32 FMT_NUM_I32 OPR_SSRC 32 SIMM32 FMT_NUM_I32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_CMP_EQ_U32 Set SCC to 1 iff the first scalar input is equal to the second scalar input. ENC_SOPC default 6 SSRC0 FMT_NUM_U32 OPR_SSRC 32 SSRC1 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOPC_INST_LITERAL has_lit_0 6 SIMM32 FMT_NUM_U32 OPR_SSRC 32 SSRC1 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOPC_INST_LITERAL has_lit_1 6 SSRC0 FMT_NUM_U32 OPR_SSRC 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOPC_INST_LITERAL has_lit_0_has_lit_1 6 SIMM32 FMT_NUM_U32 OPR_SSRC 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_CMP_LG_U32 Set SCC to 1 iff the first scalar input is less than or greater than the second scalar input. ENC_SOPC default 7 SSRC0 FMT_NUM_U32 OPR_SSRC 32 SSRC1 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOPC_INST_LITERAL has_lit_0 7 SIMM32 FMT_NUM_U32 OPR_SSRC 32 SSRC1 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOPC_INST_LITERAL has_lit_1 7 SSRC0 FMT_NUM_U32 OPR_SSRC 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOPC_INST_LITERAL has_lit_0_has_lit_1 7 SIMM32 FMT_NUM_U32 OPR_SSRC 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_CMP_GT_U32 Set SCC to 1 iff the first scalar input is greater than the second scalar input. ENC_SOPC default 8 SSRC0 FMT_NUM_U32 OPR_SSRC 32 SSRC1 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOPC_INST_LITERAL has_lit_0 8 SIMM32 FMT_NUM_U32 OPR_SSRC 32 SSRC1 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOPC_INST_LITERAL has_lit_1 8 SSRC0 FMT_NUM_U32 OPR_SSRC 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOPC_INST_LITERAL has_lit_0_has_lit_1 8 SIMM32 FMT_NUM_U32 OPR_SSRC 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_CMP_GE_U32 Set SCC to 1 iff the first scalar input is greater than or equal to the second scalar input. ENC_SOPC default 9 SSRC0 FMT_NUM_U32 OPR_SSRC 32 SSRC1 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOPC_INST_LITERAL has_lit_0 9 SIMM32 FMT_NUM_U32 OPR_SSRC 32 SSRC1 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOPC_INST_LITERAL has_lit_1 9 SSRC0 FMT_NUM_U32 OPR_SSRC 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOPC_INST_LITERAL has_lit_0_has_lit_1 9 SIMM32 FMT_NUM_U32 OPR_SSRC 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_CMP_LT_U32 Set SCC to 1 iff the first scalar input is less than the second scalar input. ENC_SOPC default 10 SSRC0 FMT_NUM_U32 OPR_SSRC 32 SSRC1 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOPC_INST_LITERAL has_lit_0 10 SIMM32 FMT_NUM_U32 OPR_SSRC 32 SSRC1 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOPC_INST_LITERAL has_lit_1 10 SSRC0 FMT_NUM_U32 OPR_SSRC 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOPC_INST_LITERAL has_lit_0_has_lit_1 10 SIMM32 FMT_NUM_U32 OPR_SSRC 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_CMP_LE_U32 Set SCC to 1 iff the first scalar input is less than or equal to the second scalar input. ENC_SOPC default 11 SSRC0 FMT_NUM_U32 OPR_SSRC 32 SSRC1 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOPC_INST_LITERAL has_lit_0 11 SIMM32 FMT_NUM_U32 OPR_SSRC 32 SSRC1 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOPC_INST_LITERAL has_lit_1 11 SSRC0 FMT_NUM_U32 OPR_SSRC 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOPC_INST_LITERAL has_lit_0_has_lit_1 11 SIMM32 FMT_NUM_U32 OPR_SSRC 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_BITCMP0_B32 Extract a bit from the first scalar input based on an index in the second scalar input, and set SCC to 1 iff the extracted bit is equal to 0. ENC_SOPC default 12 SSRC0 FMT_NUM_U32 OPR_SSRC 32 SSRC1 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOPC_INST_LITERAL has_lit_0 12 SIMM32 FMT_NUM_U32 OPR_SSRC 32 SSRC1 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOPC_INST_LITERAL has_lit_1 12 SSRC0 FMT_NUM_U32 OPR_SSRC 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOPC_INST_LITERAL has_lit_0_has_lit_1 12 SIMM32 FMT_NUM_U32 OPR_SSRC 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_BITCMP1_B32 Extract a bit from the first scalar input based on an index in the second scalar input, and set SCC to 1 iff the extracted bit is equal to 1. ENC_SOPC default 13 SSRC0 FMT_NUM_U32 OPR_SSRC 32 SSRC1 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOPC_INST_LITERAL has_lit_0 13 SIMM32 FMT_NUM_U32 OPR_SSRC 32 SSRC1 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOPC_INST_LITERAL has_lit_1 13 SSRC0 FMT_NUM_U32 OPR_SSRC 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOPC_INST_LITERAL has_lit_0_has_lit_1 13 SIMM32 FMT_NUM_U32 OPR_SSRC 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_BITCMP0_B64 Extract a bit from the first scalar input based on an index in the second scalar input, and set SCC to 1 iff the extracted bit is equal to 0. ENC_SOPC default 14 SSRC0 FMT_NUM_U64 OPR_SSRC 64 SSRC1 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOPC_INST_LITERAL has_lit_0 14 SIMM32 FMT_NUM_U64 OPR_SSRC 64 SSRC1 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOPC_INST_LITERAL has_lit_1 14 SSRC0 FMT_NUM_U64 OPR_SSRC 64 SIMM32 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOPC_INST_LITERAL has_lit_0_has_lit_1 14 SIMM32 FMT_NUM_U64 OPR_SSRC 64 SIMM32 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_BITCMP1_B64 Extract a bit from the first scalar input based on an index in the second scalar input, and set SCC to 1 iff the extracted bit is equal to 1. ENC_SOPC default 15 SSRC0 FMT_NUM_U64 OPR_SSRC 64 SSRC1 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOPC_INST_LITERAL has_lit_0 15 SIMM32 FMT_NUM_U64 OPR_SSRC 64 SSRC1 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOPC_INST_LITERAL has_lit_1 15 SSRC0 FMT_NUM_U64 OPR_SSRC 64 SIMM32 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOPC_INST_LITERAL has_lit_0_has_lit_1 15 SIMM32 FMT_NUM_U64 OPR_SSRC 64 SIMM32 FMT_NUM_U32 OPR_SSRC 32 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE TRUE S_SETVSKIP Enables or disables VSKIP mode. ENC_SOPC default 16 SSRC0 FMT_NUM_U32 OPR_SSRC 32 SSRC1 FMT_NUM_U32 OPR_SSRC 32 SOPC_INST_LITERAL has_lit_0 16 SIMM32 FMT_NUM_U32 OPR_SSRC 32 SSRC1 FMT_NUM_U32 OPR_SSRC 32 SOPC_INST_LITERAL has_lit_1 16 SSRC0 FMT_NUM_U32 OPR_SSRC 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 SOPC_INST_LITERAL has_lit_0_has_lit_1 16 SIMM32 FMT_NUM_U32 OPR_SSRC 32 SIMM32 FMT_NUM_U32 OPR_SSRC 32 WAVE_CONTROL NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_SET_GPR_IDX_ON Enable GPR indexing mode. ENC_SOPC default 17 SSRC0 FMT_NUM_B32 OPR_SSRC 32 SSRC1 FMT_NUM_B32 OPR_SIMM4 32 FMT_NUM_B32 OPR_SDST_M0 32 FMT_NUM_B32 OPR_SDST_M0 32 SOPC_INST_LITERAL has_lit_0 17 SIMM32 FMT_NUM_B32 OPR_SSRC 32 SSRC1 FMT_NUM_B32 OPR_SIMM4 32 FMT_NUM_B32 OPR_SDST_M0 32 FMT_NUM_B32 OPR_SDST_M0 32 SOPC_INST_LITERAL has_lit_1 17 SSRC0 FMT_NUM_B32 OPR_SSRC 32 SIMM32 FMT_NUM_B32 OPR_SIMM4 32 FMT_NUM_B32 OPR_SDST_M0 32 FMT_NUM_B32 OPR_SDST_M0 32 SOPC_INST_LITERAL has_lit_0_has_lit_1 17 SIMM32 FMT_NUM_B32 OPR_SSRC 32 SIMM32 FMT_NUM_B32 OPR_SIMM4 32 FMT_NUM_B32 OPR_SDST_M0 32 FMT_NUM_B32 OPR_SDST_M0 32 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_CMP_EQ_U64 Set SCC to 1 iff the first scalar input is equal to the second scalar input. ENC_SOPC default 18 SSRC0 FMT_NUM_U64 OPR_SSRC 64 SSRC1 FMT_NUM_U64 OPR_SSRC 64 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOPC_INST_LITERAL has_lit_0 18 SIMM32 FMT_NUM_U64 OPR_SSRC 64 SSRC1 FMT_NUM_U64 OPR_SSRC 64 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOPC_INST_LITERAL has_lit_1 18 SSRC0 FMT_NUM_U64 OPR_SSRC 64 SIMM32 FMT_NUM_U64 OPR_SSRC 64 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOPC_INST_LITERAL has_lit_0_has_lit_1 18 SIMM32 FMT_NUM_U64 OPR_SSRC 64 SIMM32 FMT_NUM_U64 OPR_SSRC 64 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_CMP_LG_U64 Set SCC to 1 iff the first scalar input is less than or greater than the second scalar input. ENC_SOPC default 19 SSRC0 FMT_NUM_U64 OPR_SSRC 64 SSRC1 FMT_NUM_U64 OPR_SSRC 64 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOPC_INST_LITERAL has_lit_0 19 SIMM32 FMT_NUM_U64 OPR_SSRC 64 SSRC1 FMT_NUM_U64 OPR_SSRC 64 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOPC_INST_LITERAL has_lit_1 19 SSRC0 FMT_NUM_U64 OPR_SSRC 64 SIMM32 FMT_NUM_U64 OPR_SSRC 64 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SOPC_INST_LITERAL has_lit_0_has_lit_1 19 SIMM32 FMT_NUM_U64 OPR_SSRC 64 SIMM32 FMT_NUM_U64 OPR_SSRC 64 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_MOVK_I32 Sign extend a literal 16-bit constant and store the result into a scalar register. ENC_SOPK default 0 SDST FMT_NUM_I32 OPR_SDST 32 SIMM16 FMT_NUM_I16 OPR_SIMM16 16 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_CMOVK_I32 Move the sign extension of a literal 16-bit constant into a scalar register iff SCC is nonzero. ENC_SOPK default 1 SDST FMT_NUM_I32 OPR_SDST 32 SIMM16 FMT_NUM_I16 OPR_SIMM16 16 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_CMPK_EQ_I32 Set SCC to 1 iff scalar input is equal to the sign extension of a literal 16-bit constant. ENC_SOPK default 2 SDST FMT_NUM_I32 OPR_SDST 32 SIMM16 FMT_NUM_I16 OPR_SIMM16 16 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_CMPK_LG_I32 Set SCC to 1 iff scalar input is less than or greater than the sign extension of a literal 16-bit constant. ENC_SOPK default 3 SDST FMT_NUM_I32 OPR_SDST 32 SIMM16 FMT_NUM_I16 OPR_SIMM16 16 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_CMPK_GT_I32 Set SCC to 1 iff scalar input is greater than the sign extension of a literal 16-bit constant. ENC_SOPK default 4 SDST FMT_NUM_I32 OPR_SDST 32 SIMM16 FMT_NUM_I16 OPR_SIMM16 16 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_CMPK_GE_I32 Set SCC to 1 iff scalar input is greater than or equal to the sign extension of a literal 16-bit constant. ENC_SOPK default 5 SDST FMT_NUM_I32 OPR_SDST 32 SIMM16 FMT_NUM_I16 OPR_SIMM16 16 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_CMPK_LT_I32 Set SCC to 1 iff scalar input is less than the sign extension of a literal 16-bit constant. ENC_SOPK default 6 SDST FMT_NUM_I32 OPR_SDST 32 SIMM16 FMT_NUM_I16 OPR_SIMM16 16 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_CMPK_LE_I32 Set SCC to 1 iff scalar input is less than or equal to the sign extension of a literal 16-bit constant. ENC_SOPK default 7 SDST FMT_NUM_I32 OPR_SDST 32 SIMM16 FMT_NUM_I16 OPR_SIMM16 16 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_CMPK_EQ_U32 Set SCC to 1 iff scalar input is equal to the zero extension of a literal 16-bit constant. ENC_SOPK default 8 SDST FMT_NUM_U32 OPR_SDST 32 SIMM16 FMT_NUM_U16 OPR_SIMM16 16 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_CMPK_LG_U32 Set SCC to 1 iff scalar input is less than or greater than the zero extension of a literal 16-bit constant. ENC_SOPK default 9 SDST FMT_NUM_U32 OPR_SDST 32 SIMM16 FMT_NUM_U16 OPR_SIMM16 16 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_CMPK_GT_U32 Set SCC to 1 iff scalar input is greater than the zero extension of a literal 16-bit constant. ENC_SOPK default 10 SDST FMT_NUM_U32 OPR_SDST 32 SIMM16 FMT_NUM_U16 OPR_SIMM16 16 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_CMPK_GE_U32 Set SCC to 1 iff scalar input is greater than or equal to the zero extension of a literal 16-bit constant. ENC_SOPK default 11 SDST FMT_NUM_U32 OPR_SDST 32 SIMM16 FMT_NUM_U16 OPR_SIMM16 16 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_CMPK_LT_U32 Set SCC to 1 iff scalar input is less than the zero extension of a literal 16-bit constant. ENC_SOPK default 12 SDST FMT_NUM_U32 OPR_SDST 32 SIMM16 FMT_NUM_U16 OPR_SIMM16 16 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_CMPK_LE_U32 Set SCC to 1 iff scalar input is less than or equal to the zero extension of a literal 16-bit constant. ENC_SOPK default 13 SDST FMT_NUM_U32 OPR_SDST 32 SIMM16 FMT_NUM_U16 OPR_SIMM16 16 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_ADDK_I32 Add a scalar input and the sign extension of a literal 16-bit constant, store the result into a scalar register and store the carry-out bit into SCC. ENC_SOPK default 14 SDST FMT_NUM_I32 OPR_SDST 32 SIMM16 FMT_NUM_I16 OPR_SIMM16 16 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_MULK_I32 Multiply a scalar input with the sign extension of a literal 16-bit constant and store the result into a scalar register. ENC_SOPK default 15 SDST FMT_NUM_I32 OPR_SDST 32 SIMM16 FMT_NUM_I16 OPR_SIMM16 16 SALU NOT_ASSIGNED TRUE TRUE FALSE FALSE FALSE S_CBRANCH_I_FORK Conditional branch using branch-stack. ENC_SOPK default 16 SDST FMT_NUM_B64 OPR_SDST 64 SIMM16 FMT_NUM_I16 OPR_LABEL 16 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_GETREG_B32 Read some or all of a hardware register into the LSBs of destination. ENC_SOPK default 17 SDST FMT_NUM_B32 OPR_SDST 32 SIMM16 FMT_NUM_B16 OPR_HWREG 16 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_SETREG_B32 Write some or all of the LSBs of source argument into a hardware register. ENC_SOPK default 18 SIMM16 FMT_NUM_B16 OPR_HWREG 16 SDST FMT_NUM_B32 OPR_SDST 32 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_SETREG_IMM32_B32 Write some or all of the LSBs of a 32-bit literal constant into a hardware register; this instruction requires a 32-bit literal constant. SOPK_INST_LITERAL default 20 SIMM16 FMT_NUM_B16 OPR_HWREG 16 FMT_NUM_B32 OPR_SIMM32 32 SALU NOT_ASSIGNED TRUE FALSE FALSE FALSE FALSE S_CALL_B64 Store the address of the next instruction to a scalar register and then jump to a constant offset relative to the current PC. ENC_SOPK default 21 SDST FMT_NUM_B64 OPR_SDST 64 SIMM16 FMT_NUM_I16 OPR_LABEL 16 FMT_NUM_B64 OPR_PC 64 FMT_NUM_B64 OPR_PC 64 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE TRUE S_NOP Do nothing. Delay issue of next instruction by a small, fixed amount. ENC_SOPP default 0 SIMM16 FMT_NUM_B16 OPR_SIMM16 16 WAVE_CONTROL NOT_ASSIGNED FALSE FALSE FALSE TRUE FALSE S_ENDPGM End of program; terminate wavefront. ENC_SOPP default 1 MESSAGE NOT_ASSIGNED TRUE FALSE FALSE FALSE FALSE S_BRANCH Jump to a constant offset relative to the current PC. ENC_SOPP default 2 SIMM16 FMT_NUM_I16 OPR_LABEL 16 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_WAKEUP Allow a wave to 'ping' all the other waves in its threadgroup to force them to wake up early from an S_SLEEP instruction. ENC_SOPP default 3 SALU NOT_ASSIGNED TRUE TRUE FALSE FALSE FALSE S_CBRANCH_SCC0 If SCC is 0 then jump to a constant offset relative to the current PC. ENC_SOPP default 4 SIMM16 FMT_NUM_I16 OPR_LABEL 16 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SALU NOT_ASSIGNED TRUE TRUE FALSE FALSE FALSE S_CBRANCH_SCC1 If SCC is 1 then jump to a constant offset relative to the current PC. ENC_SOPP default 5 SIMM16 FMT_NUM_I16 OPR_LABEL 16 FMT_NUM_B1 OPR_SSRC_SPECIAL_SCC 1 SALU NOT_ASSIGNED TRUE TRUE FALSE FALSE FALSE S_CBRANCH_VCCZ If VCCZ is 1 then jump to a constant offset relative to the current PC. ENC_SOPP default 6 SIMM16 FMT_NUM_I16 OPR_LABEL 16 FMT_NUM_M64 OPR_VCC 64 SALU NOT_ASSIGNED TRUE TRUE FALSE FALSE FALSE S_CBRANCH_VCCNZ If VCCZ is 0 then jump to a constant offset relative to the current PC. ENC_SOPP default 7 SIMM16 FMT_NUM_I16 OPR_LABEL 16 FMT_NUM_M64 OPR_VCC 64 SALU NOT_ASSIGNED TRUE TRUE FALSE FALSE FALSE S_CBRANCH_EXECZ If EXECZ is 1 then jump to a constant offset relative to the current PC. ENC_SOPP default 8 SIMM16 FMT_NUM_I16 OPR_LABEL 16 FMT_NUM_M64 OPR_SDST_EXEC 64 SALU NOT_ASSIGNED TRUE TRUE FALSE FALSE FALSE S_CBRANCH_EXECNZ If EXECZ is 0 then jump to a constant offset relative to the current PC. ENC_SOPP default 9 SIMM16 FMT_NUM_I16 OPR_LABEL 16 FMT_NUM_M64 OPR_SDST_EXEC 64 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_BARRIER Synchronize waves within a threadgroup. ENC_SOPP default 10 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE TRUE S_SETKILL Kill this wave if the least significant bit of the immediate constant is 1. ENC_SOPP default 11 SIMM16 FMT_NUM_B16 OPR_SIMM16 16 WAVE_CONTROL NOT_ASSIGNED FALSE FALSE FALSE FALSE TRUE S_WAITCNT Wait for the counts of outstanding local data share, vector memory and export instructions to be at or below the specified levels. ENC_SOPP default 12 SIMM16 FMT_NUM_B16 OPR_WAITCNT 16 WAVE_CONTROL NOT_ASSIGNED FALSE FALSE FALSE FALSE TRUE S_SETHALT Set or clear the HALT status bit. ENC_SOPP default 13 SIMM16 FMT_NUM_B16 OPR_SIMM16 16 WAVE_CONTROL NOT_ASSIGNED FALSE FALSE FALSE FALSE TRUE S_SLEEP Cause a wave to sleep for up to ~8000 clocks. ENC_SOPP default 14 SIMM16 FMT_NUM_B16 OPR_SIMM16 16 WAVE_CONTROL NOT_ASSIGNED FALSE FALSE FALSE FALSE TRUE S_SETPRIO Change wave user priority. ENC_SOPP default 15 SIMM16 FMT_NUM_B16 OPR_SIMM16 16 WAVE_CONTROL NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_SENDMSG Send a message upstream to graphics control hardware. ENC_SOPP default 16 SIMM16 FMT_NUM_B16 OPR_SENDMSG 16 FMT_NUM_B32 OPR_SDST_M0 32 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_SENDMSGHALT Send a message to upstream control hardware and then HALT the wavefront; see S_SENDMSG for details. ENC_SOPP default 17 SIMM16 FMT_NUM_B16 OPR_SENDMSG 16 FMT_NUM_B32 OPR_SDST_M0 32 SALU NOT_ASSIGNED TRUE FALSE TRUE FALSE FALSE S_TRAP Enter the trap handler. ENC_SOPP default 18 SIMM16 FMT_NUM_B16 OPR_SIMM16 16 TRAP NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_ICACHE_INV Invalidate entire first level instruction cache. ENC_SOPP default 19 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_INCPERFLEVEL Increment performance counter specified in SIMM16[3:0] by 1. ENC_SOPP default 20 SIMM16 FMT_NUM_B16 OPR_SIMM16 16 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_DECPERFLEVEL Decrement performance counter specified in SIMM16[3:0] by 1. ENC_SOPP default 21 SIMM16 FMT_NUM_B16 OPR_SIMM16 16 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_TTRACEDATA Send M0 as user data to the thread trace stream. ENC_SOPP default 22 FMT_NUM_B32 OPR_SDST_M0 32 SALU NOT_ASSIGNED TRUE TRUE FALSE FALSE FALSE S_CBRANCH_CDBGSYS If the system debug flag is set then jump to a constant offset relative to the current PC. ENC_SOPP default 23 SIMM16 FMT_NUM_I16 OPR_LABEL 16 SALU NOT_ASSIGNED TRUE TRUE FALSE FALSE FALSE S_CBRANCH_CDBGUSER If the user debug flag is set then jump to a constant offset relative to the current PC. ENC_SOPP default 24 SIMM16 FMT_NUM_I16 OPR_LABEL 16 SALU NOT_ASSIGNED TRUE TRUE FALSE FALSE FALSE S_CBRANCH_CDBGSYS_OR_USER If either the system debug flag or the user debug flag is set then jump to a constant offset relative to the current PC. ENC_SOPP default 25 SIMM16 FMT_NUM_I16 OPR_LABEL 16 SALU NOT_ASSIGNED TRUE TRUE FALSE FALSE FALSE S_CBRANCH_CDBGSYS_AND_USER If both the system debug flag and the user debug flag are set then jump to a constant offset relative to the current PC. ENC_SOPP default 26 SIMM16 FMT_NUM_I16 OPR_LABEL 16 SALU NOT_ASSIGNED FALSE FALSE FALSE TRUE FALSE S_ENDPGM_SAVED End of program; signal that a wave has been saved by the context-switch trap handler and terminate wavefront. ENC_SOPP default 27 MESSAGE NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_SET_GPR_IDX_OFF Clear GPR indexing mode. ENC_SOPP default 28 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_SET_GPR_IDX_MODE Modify the mode used for vector GPR indexing. ENC_SOPP default 29 SIMM16 FMT_NUM_B16 OPR_SIMM16 16 FMT_NUM_B32 OPR_SDST_M0 32 FMT_NUM_B32 OPR_SDST_M0 32 SALU NOT_ASSIGNED FALSE FALSE FALSE TRUE FALSE S_ENDPGM_ORDERED_PS_DONE End of program; signal that a wave has exited its POPS critical section and terminate wavefront. ENC_SOPP default 30 MESSAGE NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE S_SET_VALU_COEXEC_MODE Set the vector ALU co-execution mode to the value encoded in SIMM16[1:0] for the next VALU instruction. The co-execution mode is cleared to zero after the next VALU instruction is issued. ENC_SOPP default 31 SIMM16 FMT_NUM_B16 OPR_SIMM16 16 SALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_NOP Do nothing. ENC_VOP1 default 0 VOP1_INST_LITERAL has_lit 0 VOP1_VOP_DPP has_dpp 0 VOP1_VOP_SDWA has_sdwa 0 ENC_VOP3 default 320 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_MOV_B32 Move 32-bit data from a vector input into a vector register. ENC_VOP1 default 1 VDST FMT_NUM_B32 OPR_VGPR 32 SRC0 FMT_NUM_B32 OPR_SRC 32 VOP1_INST_LITERAL has_lit 1 VDST FMT_NUM_B32 OPR_VGPR 32 SIMM32 FMT_NUM_B32 OPR_SRC 32 VOP1_VOP_DPP has_dpp 1 VDST FMT_NUM_B32 OPR_VGPR 32 VSRC0 FMT_NUM_B32 OPR_SRC 32 VOP1_VOP_SDWA has_sdwa 1 VDST FMT_NUM_B32 OPR_VGPR 32 VSRC0 FMT_NUM_B32 OPR_SRC 32 ENC_VOP3 default 321 VDST FMT_NUM_B32 OPR_VGPR 32 SRC0 FMT_NUM_B32 OPR_SRC_NOLIT 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_READFIRSTLANE_B32 Read the scalar value in the lowest active lane of the input vector register and store it into a scalar register. ENC_VOP1 default 2 VDST FMT_NUM_B32 OPR_SREG_NOVCC 32 SRC0 FMT_NUM_B32 OPR_VGPR_OR_LDS 32 ENC_VOP3 default 322 VDST FMT_NUM_B32 OPR_SREG_NOVCC 32 SRC0 FMT_NUM_B32 OPR_VGPR_OR_LDS 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CVT_I32_F64 Convert from a double-precision float input to a signed 32-bit integer value and store the result into a vector register. ENC_VOP1 default 3 VDST FMT_NUM_I32 OPR_VGPR 32 SRC0 FMT_NUM_F64 OPR_SRC 64 VOP1_INST_LITERAL has_lit 3 VDST FMT_NUM_I32 OPR_VGPR 32 SIMM32 FMT_NUM_F64 OPR_SRC 64 VOP1_VOP_DPP has_dpp 3 VDST FMT_NUM_I32 OPR_VGPR 32 VSRC0 FMT_NUM_F64 OPR_SRC 64 ENC_VOP3 default 323 VDST FMT_NUM_I32 OPR_VGPR 32 SRC0 FMT_NUM_F64 OPR_SRC_NOLIT 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CVT_F64_I32 Convert from a signed 32-bit integer input to a double-precision float value and store the result into a vector register. ENC_VOP1 default 4 VDST FMT_NUM_F64 OPR_VGPR 64 SRC0 FMT_NUM_I32 OPR_SRC 32 VOP1_INST_LITERAL has_lit 4 VDST FMT_NUM_F64 OPR_VGPR 64 SIMM32 FMT_NUM_I32 OPR_SRC 32 VOP1_VOP_DPP has_dpp 4 VDST FMT_NUM_F64 OPR_VGPR 64 VSRC0 FMT_NUM_I32 OPR_SRC 32 ENC_VOP3 default 324 VDST FMT_NUM_F64 OPR_VGPR 64 SRC0 FMT_NUM_I32 OPR_SRC_NOLIT 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CVT_F32_I32 Convert from a signed 32-bit integer input to a single-precision float value and store the result into a vector register. ENC_VOP1 default 5 VDST FMT_NUM_F32 OPR_VGPR 32 SRC0 FMT_NUM_I32 OPR_SRC 32 VOP1_INST_LITERAL has_lit 5 VDST FMT_NUM_F32 OPR_VGPR 32 SIMM32 FMT_NUM_I32 OPR_SRC 32 VOP1_VOP_DPP has_dpp 5 VDST FMT_NUM_F32 OPR_VGPR 32 VSRC0 FMT_NUM_I32 OPR_SRC 32 VOP1_VOP_SDWA has_sdwa 5 VDST FMT_NUM_F32 OPR_VGPR 32 VSRC0 FMT_NUM_I32 OPR_SRC 32 ENC_VOP3 default 325 VDST FMT_NUM_F32 OPR_VGPR 32 SRC0 FMT_NUM_I32 OPR_SRC_NOLIT 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CVT_F32_U32 Convert from an unsigned 32-bit integer input to a single-precision float value and store the result into a vector register. ENC_VOP1 default 6 VDST FMT_NUM_F32 OPR_VGPR 32 SRC0 FMT_NUM_U32 OPR_SRC 32 VOP1_INST_LITERAL has_lit 6 VDST FMT_NUM_F32 OPR_VGPR 32 SIMM32 FMT_NUM_U32 OPR_SRC 32 VOP1_VOP_DPP has_dpp 6 VDST FMT_NUM_F32 OPR_VGPR 32 VSRC0 FMT_NUM_U32 OPR_SRC 32 VOP1_VOP_SDWA has_sdwa 6 VDST FMT_NUM_F32 OPR_VGPR 32 VSRC0 FMT_NUM_U32 OPR_SRC 32 ENC_VOP3 default 326 VDST FMT_NUM_F32 OPR_VGPR 32 SRC0 FMT_NUM_U32 OPR_SRC_NOLIT 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CVT_U32_F32 Convert from a single-precision float input to an unsigned 32-bit integer value and store the result into a vector register. ENC_VOP1 default 7 VDST FMT_NUM_U32 OPR_VGPR 32 SRC0 FMT_NUM_F32 OPR_SRC 32 VOP1_INST_LITERAL has_lit 7 VDST FMT_NUM_U32 OPR_VGPR 32 SIMM32 FMT_NUM_F32 OPR_SRC 32 VOP1_VOP_DPP has_dpp 7 VDST FMT_NUM_U32 OPR_VGPR 32 VSRC0 FMT_NUM_F32 OPR_SRC 32 VOP1_VOP_SDWA has_sdwa 7 VDST FMT_NUM_U32 OPR_VGPR 32 VSRC0 FMT_NUM_F32 OPR_SRC 32 ENC_VOP3 default 327 VDST FMT_NUM_U32 OPR_VGPR 32 SRC0 FMT_NUM_F32 OPR_SRC_NOLIT 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CVT_I32_F32 Convert from a single-precision float input to a signed 32-bit integer value and store the result into a vector register. ENC_VOP1 default 8 VDST FMT_NUM_I32 OPR_VGPR 32 SRC0 FMT_NUM_F32 OPR_SRC 32 VOP1_INST_LITERAL has_lit 8 VDST FMT_NUM_I32 OPR_VGPR 32 SIMM32 FMT_NUM_F32 OPR_SRC 32 VOP1_VOP_DPP has_dpp 8 VDST FMT_NUM_I32 OPR_VGPR 32 VSRC0 FMT_NUM_F32 OPR_SRC 32 VOP1_VOP_SDWA has_sdwa 8 VDST FMT_NUM_I32 OPR_VGPR 32 VSRC0 FMT_NUM_F32 OPR_SRC 32 ENC_VOP3 default 328 VDST FMT_NUM_I32 OPR_VGPR 32 SRC0 FMT_NUM_F32 OPR_SRC_NOLIT 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CVT_F16_F32 Convert from a single-precision float input to a half-precision float value and store the result into a vector register. ENC_VOP1 default 10 VDST FMT_NUM_F16 OPR_VGPR 16 SRC0 FMT_NUM_F32 OPR_SRC 32 VOP1_INST_LITERAL has_lit 10 VDST FMT_NUM_F16 OPR_VGPR 16 SIMM32 FMT_NUM_F32 OPR_SRC 32 VOP1_VOP_DPP has_dpp 10 VDST FMT_NUM_F16 OPR_VGPR 16 VSRC0 FMT_NUM_F32 OPR_SRC 32 VOP1_VOP_SDWA has_sdwa 10 VDST FMT_NUM_F16 OPR_VGPR 16 VSRC0 FMT_NUM_F32 OPR_SRC 32 ENC_VOP3 default 330 VDST FMT_NUM_F16 OPR_VGPR 16 SRC0 FMT_NUM_F32 OPR_SRC_NOLIT 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CVT_F32_F16 Convert from a half-precision float input to a single-precision float value and store the result into a vector register. ENC_VOP1 default 11 VDST FMT_NUM_F32 OPR_VGPR 32 SRC0 FMT_NUM_F16 OPR_SRC 16 VOP1_INST_LITERAL has_lit 11 VDST FMT_NUM_F32 OPR_VGPR 32 SIMM32 FMT_NUM_F16 OPR_SRC 16 VOP1_VOP_DPP has_dpp 11 VDST FMT_NUM_F32 OPR_VGPR 32 VSRC0 FMT_NUM_F16 OPR_SRC 16 VOP1_VOP_SDWA has_sdwa 11 VDST FMT_NUM_F32 OPR_VGPR 32 VSRC0 FMT_NUM_F16 OPR_SRC 16 ENC_VOP3 default 331 VDST FMT_NUM_F32 OPR_VGPR 32 SRC0 FMT_NUM_F16 OPR_SRC_NOLIT 16 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CVT_RPI_I32_F32 Convert from a single-precision float input to a signed 32-bit integer value using round to nearest integer semantics (ignore the default rounding mode) and store the result into a vector register. ENC_VOP1 default 12 VDST FMT_NUM_I32 OPR_VGPR 32 SRC0 FMT_NUM_F32 OPR_SRC 32 VOP1_INST_LITERAL has_lit 12 VDST FMT_NUM_I32 OPR_VGPR 32 SIMM32 FMT_NUM_F32 OPR_SRC 32 VOP1_VOP_DPP has_dpp 12 VDST FMT_NUM_I32 OPR_VGPR 32 VSRC0 FMT_NUM_F32 OPR_SRC 32 VOP1_VOP_SDWA has_sdwa 12 VDST FMT_NUM_I32 OPR_VGPR 32 VSRC0 FMT_NUM_F32 OPR_SRC 32 ENC_VOP3 default 332 VDST FMT_NUM_I32 OPR_VGPR 32 SRC0 FMT_NUM_F32 OPR_SRC_NOLIT 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CVT_FLR_I32_F32 Convert from a single-precision float input to a signed 32-bit integer value using round-down semantics (ignore the default rounding mode) and store the result into a vector register. ENC_VOP1 default 13 VDST FMT_NUM_I32 OPR_VGPR 32 SRC0 FMT_NUM_F32 OPR_SRC 32 VOP1_INST_LITERAL has_lit 13 VDST FMT_NUM_I32 OPR_VGPR 32 SIMM32 FMT_NUM_F32 OPR_SRC 32 VOP1_VOP_DPP has_dpp 13 VDST FMT_NUM_I32 OPR_VGPR 32 VSRC0 FMT_NUM_F32 OPR_SRC 32 VOP1_VOP_SDWA has_sdwa 13 VDST FMT_NUM_I32 OPR_VGPR 32 VSRC0 FMT_NUM_F32 OPR_SRC 32 ENC_VOP3 default 333 VDST FMT_NUM_I32 OPR_VGPR 32 SRC0 FMT_NUM_F32 OPR_SRC_NOLIT 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CVT_OFF_F32_I4 Convert from a signed 4-bit integer input to a single-precision float value using an offset table and store the result into a vector register. ENC_VOP1 default 14 VDST FMT_NUM_F32 OPR_VGPR 32 SRC0 FMT_NUM_I32 OPR_SRC 32 VOP1_INST_LITERAL has_lit 14 VDST FMT_NUM_F32 OPR_VGPR 32 SIMM32 FMT_NUM_I32 OPR_SRC 32 VOP1_VOP_DPP has_dpp 14 VDST FMT_NUM_F32 OPR_VGPR 32 VSRC0 FMT_NUM_I32 OPR_SRC 32 VOP1_VOP_SDWA has_sdwa 14 VDST FMT_NUM_F32 OPR_VGPR 32 VSRC0 FMT_NUM_I32 OPR_SRC 32 ENC_VOP3 default 334 VDST FMT_NUM_F32 OPR_VGPR 32 SRC0 FMT_NUM_I32 OPR_SRC_NOLIT 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CVT_F32_F64 Convert from a double-precision float input to a single-precision float value and store the result into a vector register. ENC_VOP1 default 15 VDST FMT_NUM_F32 OPR_VGPR 32 SRC0 FMT_NUM_F64 OPR_SRC 64 VOP1_INST_LITERAL has_lit 15 VDST FMT_NUM_F32 OPR_VGPR 32 SIMM32 FMT_NUM_F64 OPR_SRC 64 VOP1_VOP_DPP has_dpp 15 VDST FMT_NUM_F32 OPR_VGPR 32 VSRC0 FMT_NUM_F64 OPR_SRC 64 ENC_VOP3 default 335 VDST FMT_NUM_F32 OPR_VGPR 32 SRC0 FMT_NUM_F64 OPR_SRC_NOLIT 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CVT_F64_F32 Convert from a single-precision float input to a double-precision float value and store the result into a vector register. ENC_VOP1 default 16 VDST FMT_NUM_F64 OPR_VGPR 64 SRC0 FMT_NUM_F32 OPR_SRC 32 VOP1_INST_LITERAL has_lit 16 VDST FMT_NUM_F64 OPR_VGPR 64 SIMM32 FMT_NUM_F32 OPR_SRC 32 ENC_VOP3 default 336 VDST FMT_NUM_F64 OPR_VGPR 64 SRC0 FMT_NUM_F32 OPR_SRC_NOLIT 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CVT_F32_UBYTE0 Convert an unsigned byte in byte 0 of the input to a single-precision float value and store the result into a vector register. ENC_VOP1 default 17 VDST FMT_NUM_F32 OPR_VGPR 32 SRC0 FMT_NUM_U32 OPR_SRC 32 VOP1_INST_LITERAL has_lit 17 VDST FMT_NUM_F32 OPR_VGPR 32 SIMM32 FMT_NUM_U32 OPR_SRC 32 VOP1_VOP_DPP has_dpp 17 VDST FMT_NUM_F32 OPR_VGPR 32 VSRC0 FMT_NUM_U32 OPR_SRC 32 VOP1_VOP_SDWA has_sdwa 17 VDST FMT_NUM_F32 OPR_VGPR 32 VSRC0 FMT_NUM_U32 OPR_SRC 32 ENC_VOP3 default 337 VDST FMT_NUM_F32 OPR_VGPR 32 SRC0 FMT_NUM_U32 OPR_SRC_NOLIT 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CVT_F32_UBYTE1 Convert an unsigned byte in byte 1 of the input to a single-precision float value and store the result into a vector register. ENC_VOP1 default 18 VDST FMT_NUM_F32 OPR_VGPR 32 SRC0 FMT_NUM_U32 OPR_SRC 32 VOP1_INST_LITERAL has_lit 18 VDST FMT_NUM_F32 OPR_VGPR 32 SIMM32 FMT_NUM_U32 OPR_SRC 32 VOP1_VOP_DPP has_dpp 18 VDST FMT_NUM_F32 OPR_VGPR 32 VSRC0 FMT_NUM_U32 OPR_SRC 32 VOP1_VOP_SDWA has_sdwa 18 VDST FMT_NUM_F32 OPR_VGPR 32 VSRC0 FMT_NUM_U32 OPR_SRC 32 ENC_VOP3 default 338 VDST FMT_NUM_F32 OPR_VGPR 32 SRC0 FMT_NUM_U32 OPR_SRC_NOLIT 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CVT_F32_UBYTE2 Convert an unsigned byte in byte 2 of the input to a single-precision float value and store the result into a vector register. ENC_VOP1 default 19 VDST FMT_NUM_F32 OPR_VGPR 32 SRC0 FMT_NUM_U32 OPR_SRC 32 VOP1_INST_LITERAL has_lit 19 VDST FMT_NUM_F32 OPR_VGPR 32 SIMM32 FMT_NUM_U32 OPR_SRC 32 VOP1_VOP_DPP has_dpp 19 VDST FMT_NUM_F32 OPR_VGPR 32 VSRC0 FMT_NUM_U32 OPR_SRC 32 VOP1_VOP_SDWA has_sdwa 19 VDST FMT_NUM_F32 OPR_VGPR 32 VSRC0 FMT_NUM_U32 OPR_SRC 32 ENC_VOP3 default 339 VDST FMT_NUM_F32 OPR_VGPR 32 SRC0 FMT_NUM_U32 OPR_SRC_NOLIT 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CVT_F32_UBYTE3 Convert an unsigned byte in byte 3 of the input to a single-precision float value and store the result into a vector register. ENC_VOP1 default 20 VDST FMT_NUM_F32 OPR_VGPR 32 SRC0 FMT_NUM_U32 OPR_SRC 32 VOP1_INST_LITERAL has_lit 20 VDST FMT_NUM_F32 OPR_VGPR 32 SIMM32 FMT_NUM_U32 OPR_SRC 32 VOP1_VOP_DPP has_dpp 20 VDST FMT_NUM_F32 OPR_VGPR 32 VSRC0 FMT_NUM_U32 OPR_SRC 32 VOP1_VOP_SDWA has_sdwa 20 VDST FMT_NUM_F32 OPR_VGPR 32 VSRC0 FMT_NUM_U32 OPR_SRC 32 ENC_VOP3 default 340 VDST FMT_NUM_F32 OPR_VGPR 32 SRC0 FMT_NUM_U32 OPR_SRC_NOLIT 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CVT_U32_F64 Convert from a double-precision float input to an unsigned 32-bit integer value and store the result into a vector register. ENC_VOP1 default 21 VDST FMT_NUM_U32 OPR_VGPR 32 SRC0 FMT_NUM_F64 OPR_SRC 64 VOP1_INST_LITERAL has_lit 21 VDST FMT_NUM_U32 OPR_VGPR 32 SIMM32 FMT_NUM_F64 OPR_SRC 64 VOP1_VOP_DPP has_dpp 21 VDST FMT_NUM_U32 OPR_VGPR 32 VSRC0 FMT_NUM_F64 OPR_SRC 64 ENC_VOP3 default 341 VDST FMT_NUM_U32 OPR_VGPR 32 SRC0 FMT_NUM_F64 OPR_SRC_NOLIT 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CVT_F64_U32 Convert from an unsigned 32-bit integer input to a double-precision float value and store the result into a vector register. ENC_VOP1 default 22 VDST FMT_NUM_F64 OPR_VGPR 64 SRC0 FMT_NUM_U32 OPR_SRC 32 VOP1_INST_LITERAL has_lit 22 VDST FMT_NUM_F64 OPR_VGPR 64 SIMM32 FMT_NUM_U32 OPR_SRC 32 VOP1_VOP_DPP has_dpp 22 VDST FMT_NUM_F64 OPR_VGPR 64 VSRC0 FMT_NUM_U32 OPR_SRC 32 ENC_VOP3 default 342 VDST FMT_NUM_F64 OPR_VGPR 64 SRC0 FMT_NUM_U32 OPR_SRC_NOLIT 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_TRUNC_F64 Compute the integer part of a double-precision float input using round toward zero semantics and store the result in floating point format into a vector register. ENC_VOP1 default 23 VDST FMT_NUM_F64 OPR_VGPR 64 SRC0 FMT_NUM_F64 OPR_SRC 64 VOP1_INST_LITERAL has_lit 23 VDST FMT_NUM_F64 OPR_VGPR 64 SIMM32 FMT_NUM_F64 OPR_SRC 64 VOP1_VOP_DPP has_dpp 23 VDST FMT_NUM_F64 OPR_VGPR 64 VSRC0 FMT_NUM_F64 OPR_SRC 64 ENC_VOP3 default 343 VDST FMT_NUM_F64 OPR_VGPR 64 SRC0 FMT_NUM_F64 OPR_SRC_NOLIT 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CEIL_F64 Round the double-precision float input up to next integer and store the result in floating point format into a vector register. ENC_VOP1 default 24 VDST FMT_NUM_F64 OPR_VGPR 64 SRC0 FMT_NUM_F64 OPR_SRC 64 VOP1_INST_LITERAL has_lit 24 VDST FMT_NUM_F64 OPR_VGPR 64 SIMM32 FMT_NUM_F64 OPR_SRC 64 VOP1_VOP_DPP has_dpp 24 VDST FMT_NUM_F64 OPR_VGPR 64 VSRC0 FMT_NUM_F64 OPR_SRC 64 ENC_VOP3 default 344 VDST FMT_NUM_F64 OPR_VGPR 64 SRC0 FMT_NUM_F64 OPR_SRC_NOLIT 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_RNDNE_F64 Round the double-precision float input to the nearest even integer and store the result in floating point format into a vector register. ENC_VOP1 default 25 VDST FMT_NUM_F64 OPR_VGPR 64 SRC0 FMT_NUM_F64 OPR_SRC 64 VOP1_INST_LITERAL has_lit 25 VDST FMT_NUM_F64 OPR_VGPR 64 SIMM32 FMT_NUM_F64 OPR_SRC 64 ENC_VOP3 default 345 VDST FMT_NUM_F64 OPR_VGPR 64 SRC0 FMT_NUM_F64 OPR_SRC_NOLIT 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_FLOOR_F64 Round the double-precision float input down to previous integer and store the result in floating point format into a vector register. ENC_VOP1 default 26 VDST FMT_NUM_F64 OPR_VGPR 64 SRC0 FMT_NUM_F64 OPR_SRC 64 VOP1_INST_LITERAL has_lit 26 VDST FMT_NUM_F64 OPR_VGPR 64 SIMM32 FMT_NUM_F64 OPR_SRC 64 VOP1_VOP_DPP has_dpp 26 VDST FMT_NUM_F64 OPR_VGPR 64 VSRC0 FMT_NUM_F64 OPR_SRC 64 ENC_VOP3 default 346 VDST FMT_NUM_F64 OPR_VGPR 64 SRC0 FMT_NUM_F64 OPR_SRC_NOLIT 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_FRACT_F32 Compute the fractional portion of a single-precision float input and store the result in floating point format into a vector register. ENC_VOP1 default 27 VDST FMT_NUM_F32 OPR_VGPR 32 SRC0 FMT_NUM_F32 OPR_SRC 32 VOP1_INST_LITERAL has_lit 27 VDST FMT_NUM_F32 OPR_VGPR 32 SIMM32 FMT_NUM_F32 OPR_SRC 32 VOP1_VOP_DPP has_dpp 27 VDST FMT_NUM_F32 OPR_VGPR 32 VSRC0 FMT_NUM_F32 OPR_SRC 32 VOP1_VOP_SDWA has_sdwa 27 VDST FMT_NUM_F32 OPR_VGPR 32 VSRC0 FMT_NUM_F32 OPR_SRC 32 ENC_VOP3 default 347 VDST FMT_NUM_F32 OPR_VGPR 32 SRC0 FMT_NUM_F32 OPR_SRC_NOLIT 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_TRUNC_F32 Compute the integer part of a single-precision float input using round toward zero semantics and store the result in floating point format into a vector register. ENC_VOP1 default 28 VDST FMT_NUM_F32 OPR_VGPR 32 SRC0 FMT_NUM_F32 OPR_SRC 32 VOP1_INST_LITERAL has_lit 28 VDST FMT_NUM_F32 OPR_VGPR 32 SIMM32 FMT_NUM_F32 OPR_SRC 32 VOP1_VOP_DPP has_dpp 28 VDST FMT_NUM_F32 OPR_VGPR 32 VSRC0 FMT_NUM_F32 OPR_SRC 32 VOP1_VOP_SDWA has_sdwa 28 VDST FMT_NUM_F32 OPR_VGPR 32 VSRC0 FMT_NUM_F32 OPR_SRC 32 ENC_VOP3 default 348 VDST FMT_NUM_F32 OPR_VGPR 32 SRC0 FMT_NUM_F32 OPR_SRC_NOLIT 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CEIL_F32 Round the single-precision float input up to next integer and store the result in floating point format into a vector register. ENC_VOP1 default 29 VDST FMT_NUM_F32 OPR_VGPR 32 SRC0 FMT_NUM_F32 OPR_SRC 32 VOP1_INST_LITERAL has_lit 29 VDST FMT_NUM_F32 OPR_VGPR 32 SIMM32 FMT_NUM_F32 OPR_SRC 32 VOP1_VOP_DPP has_dpp 29 VDST FMT_NUM_F32 OPR_VGPR 32 VSRC0 FMT_NUM_F32 OPR_SRC 32 VOP1_VOP_SDWA has_sdwa 29 VDST FMT_NUM_F32 OPR_VGPR 32 VSRC0 FMT_NUM_F32 OPR_SRC 32 ENC_VOP3 default 349 VDST FMT_NUM_F32 OPR_VGPR 32 SRC0 FMT_NUM_F32 OPR_SRC_NOLIT 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_RNDNE_F32 Round the single-precision float input to the nearest even integer and store the result in floating point format into a vector register. ENC_VOP1 default 30 VDST FMT_NUM_F32 OPR_VGPR 32 SRC0 FMT_NUM_F32 OPR_SRC 32 VOP1_INST_LITERAL has_lit 30 VDST FMT_NUM_F32 OPR_VGPR 32 SIMM32 FMT_NUM_F32 OPR_SRC 32 VOP1_VOP_DPP has_dpp 30 VDST FMT_NUM_F32 OPR_VGPR 32 VSRC0 FMT_NUM_F32 OPR_SRC 32 VOP1_VOP_SDWA has_sdwa 30 VDST FMT_NUM_F32 OPR_VGPR 32 VSRC0 FMT_NUM_F32 OPR_SRC 32 ENC_VOP3 default 350 VDST FMT_NUM_F32 OPR_VGPR 32 SRC0 FMT_NUM_F32 OPR_SRC_NOLIT 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_FLOOR_F32 Round the single-precision float input down to previous integer and store the result in floating point format into a vector register. ENC_VOP1 default 31 VDST FMT_NUM_F32 OPR_VGPR 32 SRC0 FMT_NUM_F32 OPR_SRC 32 VOP1_INST_LITERAL has_lit 31 VDST FMT_NUM_F32 OPR_VGPR 32 SIMM32 FMT_NUM_F32 OPR_SRC 32 VOP1_VOP_DPP has_dpp 31 VDST FMT_NUM_F32 OPR_VGPR 32 VSRC0 FMT_NUM_F32 OPR_SRC 32 VOP1_VOP_SDWA has_sdwa 31 VDST FMT_NUM_F32 OPR_VGPR 32 VSRC0 FMT_NUM_F32 OPR_SRC 32 ENC_VOP3 default 351 VDST FMT_NUM_F32 OPR_VGPR 32 SRC0 FMT_NUM_F32 OPR_SRC_NOLIT 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_EXP_F32 Calculate 2 raised to the power of the single-precision float input and store the result into a vector register. ENC_VOP1 default 32 VDST FMT_NUM_F32 OPR_VGPR 32 SRC0 FMT_NUM_F32 OPR_SRC 32 VOP1_INST_LITERAL has_lit 32 VDST FMT_NUM_F32 OPR_VGPR 32 SIMM32 FMT_NUM_F32 OPR_SRC 32 VOP1_VOP_DPP has_dpp 32 VDST FMT_NUM_F32 OPR_VGPR 32 VSRC0 FMT_NUM_F32 OPR_SRC 32 VOP1_VOP_SDWA has_sdwa 32 VDST FMT_NUM_F32 OPR_VGPR 32 VSRC0 FMT_NUM_F32 OPR_SRC 32 ENC_VOP3 default 352 VDST FMT_NUM_F32 OPR_VGPR 32 SRC0 FMT_NUM_F32 OPR_SRC_NOLIT 32 VALU TRANSCENDENTAL FALSE FALSE FALSE FALSE FALSE V_LOG_F32 Calculate the base 2 logarithm of the single-precision float input and store the result into a vector register. ENC_VOP1 default 33 VDST FMT_NUM_F32 OPR_VGPR 32 SRC0 FMT_NUM_F32 OPR_SRC 32 VOP1_INST_LITERAL has_lit 33 VDST FMT_NUM_F32 OPR_VGPR 32 SIMM32 FMT_NUM_F32 OPR_SRC 32 VOP1_VOP_DPP has_dpp 33 VDST FMT_NUM_F32 OPR_VGPR 32 VSRC0 FMT_NUM_F32 OPR_SRC 32 VOP1_VOP_SDWA has_sdwa 33 VDST FMT_NUM_F32 OPR_VGPR 32 VSRC0 FMT_NUM_F32 OPR_SRC 32 ENC_VOP3 default 353 VDST FMT_NUM_F32 OPR_VGPR 32 SRC0 FMT_NUM_F32 OPR_SRC_NOLIT 32 VALU TRANSCENDENTAL FALSE FALSE FALSE FALSE FALSE V_RCP_F32 Calculate the reciprocal of the single-precision float input using IEEE rules and store the result into a vector register. ENC_VOP1 default 34 VDST FMT_NUM_F32 OPR_VGPR 32 SRC0 FMT_NUM_F32 OPR_SRC 32 VOP1_INST_LITERAL has_lit 34 VDST FMT_NUM_F32 OPR_VGPR 32 SIMM32 FMT_NUM_F32 OPR_SRC 32 VOP1_VOP_DPP has_dpp 34 VDST FMT_NUM_F32 OPR_VGPR 32 VSRC0 FMT_NUM_F32 OPR_SRC 32 VOP1_VOP_SDWA has_sdwa 34 VDST FMT_NUM_F32 OPR_VGPR 32 VSRC0 FMT_NUM_F32 OPR_SRC 32 ENC_VOP3 default 354 VDST FMT_NUM_F32 OPR_VGPR 32 SRC0 FMT_NUM_F32 OPR_SRC_NOLIT 32 VALU TRANSCENDENTAL FALSE FALSE FALSE FALSE FALSE V_RCP_IFLAG_F32 Calculate the reciprocal of the vector float input in a manner suitable for integer division and store the result into a vector register. This opcode is intended for use as part of an integer division macro. ENC_VOP1 default 35 VDST FMT_NUM_F32 OPR_VGPR 32 SRC0 FMT_NUM_F32 OPR_SRC 32 VOP1_INST_LITERAL has_lit 35 VDST FMT_NUM_F32 OPR_VGPR 32 SIMM32 FMT_NUM_F32 OPR_SRC 32 VOP1_VOP_DPP has_dpp 35 VDST FMT_NUM_F32 OPR_VGPR 32 VSRC0 FMT_NUM_F32 OPR_SRC 32 VOP1_VOP_SDWA has_sdwa 35 VDST FMT_NUM_F32 OPR_VGPR 32 VSRC0 FMT_NUM_F32 OPR_SRC 32 ENC_VOP3 default 355 VDST FMT_NUM_F32 OPR_VGPR 32 SRC0 FMT_NUM_F32 OPR_SRC_NOLIT 32 VALU TRANSCENDENTAL FALSE FALSE FALSE FALSE FALSE V_RSQ_F32 Calculate the reciprocal of the square root of the single-precision float input using IEEE rules and store the result into a vector register. ENC_VOP1 default 36 VDST FMT_NUM_F32 OPR_VGPR 32 SRC0 FMT_NUM_F32 OPR_SRC 32 VOP1_INST_LITERAL has_lit 36 VDST FMT_NUM_F32 OPR_VGPR 32 SIMM32 FMT_NUM_F32 OPR_SRC 32 VOP1_VOP_DPP has_dpp 36 VDST FMT_NUM_F32 OPR_VGPR 32 VSRC0 FMT_NUM_F32 OPR_SRC 32 VOP1_VOP_SDWA has_sdwa 36 VDST FMT_NUM_F32 OPR_VGPR 32 VSRC0 FMT_NUM_F32 OPR_SRC 32 ENC_VOP3 default 356 VDST FMT_NUM_F32 OPR_VGPR 32 SRC0 FMT_NUM_F32 OPR_SRC_NOLIT 32 VALU TRANSCENDENTAL FALSE FALSE FALSE FALSE FALSE V_RCP_F64 Calculate the reciprocal of the double-precision float input using IEEE rules and store the result into a vector register. ENC_VOP1 default 37 VDST FMT_NUM_F64 OPR_VGPR 64 SRC0 FMT_NUM_F64 OPR_SRC 64 VOP1_INST_LITERAL has_lit 37 VDST FMT_NUM_F64 OPR_VGPR 64 SIMM32 FMT_NUM_F64 OPR_SRC 64 ENC_VOP3 default 357 VDST FMT_NUM_F64 OPR_VGPR 64 SRC0 FMT_NUM_F64 OPR_SRC_NOLIT 64 VALU TRANSCENDENTAL FALSE FALSE FALSE FALSE FALSE V_RSQ_F64 Calculate the reciprocal of the square root of the double-precision float input using IEEE rules and store the result into a vector register. ENC_VOP1 default 38 VDST FMT_NUM_F64 OPR_VGPR 64 SRC0 FMT_NUM_F64 OPR_SRC 64 VOP1_INST_LITERAL has_lit 38 VDST FMT_NUM_F64 OPR_VGPR 64 SIMM32 FMT_NUM_F64 OPR_SRC 64 ENC_VOP3 default 358 VDST FMT_NUM_F64 OPR_VGPR 64 SRC0 FMT_NUM_F64 OPR_SRC_NOLIT 64 VALU TRANSCENDENTAL FALSE FALSE FALSE FALSE FALSE V_SQRT_F32 Calculate the square root of the single-precision float input using IEEE rules and store the result into a vector register. ENC_VOP1 default 39 VDST FMT_NUM_F32 OPR_VGPR 32 SRC0 FMT_NUM_F32 OPR_SRC 32 VOP1_INST_LITERAL has_lit 39 VDST FMT_NUM_F32 OPR_VGPR 32 SIMM32 FMT_NUM_F32 OPR_SRC 32 VOP1_VOP_DPP has_dpp 39 VDST FMT_NUM_F32 OPR_VGPR 32 VSRC0 FMT_NUM_F32 OPR_SRC 32 VOP1_VOP_SDWA has_sdwa 39 VDST FMT_NUM_F32 OPR_VGPR 32 VSRC0 FMT_NUM_F32 OPR_SRC 32 ENC_VOP3 default 359 VDST FMT_NUM_F32 OPR_VGPR 32 SRC0 FMT_NUM_F32 OPR_SRC_NOLIT 32 VALU TRANSCENDENTAL FALSE FALSE FALSE FALSE FALSE V_SQRT_F64 Calculate the square root of the double-precision float input using IEEE rules and store the result into a vector register. ENC_VOP1 default 40 VDST FMT_NUM_F64 OPR_VGPR 64 SRC0 FMT_NUM_F64 OPR_SRC 64 VOP1_INST_LITERAL has_lit 40 VDST FMT_NUM_F64 OPR_VGPR 64 SIMM32 FMT_NUM_F64 OPR_SRC 64 ENC_VOP3 default 360 VDST FMT_NUM_F64 OPR_VGPR 64 SRC0 FMT_NUM_F64 OPR_SRC_NOLIT 64 VALU TRANSCENDENTAL FALSE FALSE FALSE FALSE FALSE V_SIN_F32 Calculate the trigonometric sine of a single-precision float value using IEEE rules and store the result into a vector register. The operand is calculated by scaling the vector input by 2 PI. ENC_VOP1 default 41 VDST FMT_NUM_F32 OPR_VGPR 32 SRC0 FMT_NUM_F32 OPR_SRC 32 VOP1_INST_LITERAL has_lit 41 VDST FMT_NUM_F32 OPR_VGPR 32 SIMM32 FMT_NUM_F32 OPR_SRC 32 VOP1_VOP_DPP has_dpp 41 VDST FMT_NUM_F32 OPR_VGPR 32 VSRC0 FMT_NUM_F32 OPR_SRC 32 VOP1_VOP_SDWA has_sdwa 41 VDST FMT_NUM_F32 OPR_VGPR 32 VSRC0 FMT_NUM_F32 OPR_SRC 32 ENC_VOP3 default 361 VDST FMT_NUM_F32 OPR_VGPR 32 SRC0 FMT_NUM_F32 OPR_SRC_NOLIT 32 VALU TRANSCENDENTAL FALSE FALSE FALSE FALSE FALSE V_COS_F32 Calculate the trigonometric cosine of a single-precision float value using IEEE rules and store the result into a vector register. The operand is calculated by scaling the vector input by 2 PI. ENC_VOP1 default 42 VDST FMT_NUM_F32 OPR_VGPR 32 SRC0 FMT_NUM_F32 OPR_SRC 32 VOP1_INST_LITERAL has_lit 42 VDST FMT_NUM_F32 OPR_VGPR 32 SIMM32 FMT_NUM_F32 OPR_SRC 32 VOP1_VOP_DPP has_dpp 42 VDST FMT_NUM_F32 OPR_VGPR 32 VSRC0 FMT_NUM_F32 OPR_SRC 32 VOP1_VOP_SDWA has_sdwa 42 VDST FMT_NUM_F32 OPR_VGPR 32 VSRC0 FMT_NUM_F32 OPR_SRC 32 ENC_VOP3 default 362 VDST FMT_NUM_F32 OPR_VGPR 32 SRC0 FMT_NUM_F32 OPR_SRC_NOLIT 32 VALU TRANSCENDENTAL FALSE FALSE FALSE FALSE FALSE V_NOT_B32 Calculate bitwise negation on a vector input and store the result into a vector register. ENC_VOP1 default 43 VDST FMT_NUM_U32 OPR_VGPR 32 SRC0 FMT_NUM_U32 OPR_SRC 32 VOP1_INST_LITERAL has_lit 43 VDST FMT_NUM_U32 OPR_VGPR 32 SIMM32 FMT_NUM_U32 OPR_SRC 32 VOP1_VOP_DPP has_dpp 43 VDST FMT_NUM_U32 OPR_VGPR 32 VSRC0 FMT_NUM_U32 OPR_SRC 32 VOP1_VOP_SDWA has_sdwa 43 VDST FMT_NUM_U32 OPR_VGPR 32 VSRC0 FMT_NUM_U32 OPR_SRC 32 ENC_VOP3 default 363 VDST FMT_NUM_U32 OPR_VGPR 32 SRC0 FMT_NUM_U32 OPR_SRC_NOLIT 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_BFREV_B32 Reverse the order of bits in a vector input and store the result into a vector register. ENC_VOP1 default 44 VDST FMT_NUM_B32 OPR_VGPR 32 SRC0 FMT_NUM_B32 OPR_SRC 32 VOP1_INST_LITERAL has_lit 44 VDST FMT_NUM_B32 OPR_VGPR 32 SIMM32 FMT_NUM_B32 OPR_SRC 32 VOP1_VOP_DPP has_dpp 44 VDST FMT_NUM_B32 OPR_VGPR 32 VSRC0 FMT_NUM_B32 OPR_SRC 32 VOP1_VOP_SDWA has_sdwa 44 VDST FMT_NUM_B32 OPR_VGPR 32 VSRC0 FMT_NUM_B32 OPR_SRC 32 ENC_VOP3 default 364 VDST FMT_NUM_B32 OPR_VGPR 32 SRC0 FMT_NUM_B32 OPR_SRC_NOLIT 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_FFBH_U32 Count the number of leading "0" bits before the first "1" in a vector input and store the result into a vector register. Store -1 if there are no "1" bits. ENC_VOP1 default 45 VDST FMT_NUM_I32 OPR_VGPR 32 SRC0 FMT_NUM_U32 OPR_SRC 32 VOP1_INST_LITERAL has_lit 45 VDST FMT_NUM_I32 OPR_VGPR 32 SIMM32 FMT_NUM_U32 OPR_SRC 32 VOP1_VOP_DPP has_dpp 45 VDST FMT_NUM_I32 OPR_VGPR 32 VSRC0 FMT_NUM_U32 OPR_SRC 32 VOP1_VOP_SDWA has_sdwa 45 VDST FMT_NUM_I32 OPR_VGPR 32 VSRC0 FMT_NUM_U32 OPR_SRC 32 ENC_VOP3 default 365 VDST FMT_NUM_I32 OPR_VGPR 32 SRC0 FMT_NUM_U32 OPR_SRC_NOLIT 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_FFBL_B32 Count the number of trailing "0" bits before the first "1" in a vector input and store the result into a vector register. Store -1 if there are no "1" bits in the input. ENC_VOP1 default 46 VDST FMT_NUM_I32 OPR_VGPR 32 SRC0 FMT_NUM_U32 OPR_SRC 32 VOP1_INST_LITERAL has_lit 46 VDST FMT_NUM_I32 OPR_VGPR 32 SIMM32 FMT_NUM_U32 OPR_SRC 32 VOP1_VOP_DPP has_dpp 46 VDST FMT_NUM_I32 OPR_VGPR 32 VSRC0 FMT_NUM_U32 OPR_SRC 32 VOP1_VOP_SDWA has_sdwa 46 VDST FMT_NUM_I32 OPR_VGPR 32 VSRC0 FMT_NUM_U32 OPR_SRC 32 ENC_VOP3 default 366 VDST FMT_NUM_I32 OPR_VGPR 32 SRC0 FMT_NUM_U32 OPR_SRC_NOLIT 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_FFBH_I32 Count the number of leading bits that are the same as the sign bit of a vector input and store the result into a vector register. Store -1 if all input bits are the same. ENC_VOP1 default 47 VDST FMT_NUM_I32 OPR_VGPR 32 SRC0 FMT_NUM_I32 OPR_SRC 32 VOP1_INST_LITERAL has_lit 47 VDST FMT_NUM_I32 OPR_VGPR 32 SIMM32 FMT_NUM_I32 OPR_SRC 32 VOP1_VOP_DPP has_dpp 47 VDST FMT_NUM_I32 OPR_VGPR 32 VSRC0 FMT_NUM_I32 OPR_SRC 32 VOP1_VOP_SDWA has_sdwa 47 VDST FMT_NUM_I32 OPR_VGPR 32 VSRC0 FMT_NUM_I32 OPR_SRC 32 ENC_VOP3 default 367 VDST FMT_NUM_I32 OPR_VGPR 32 SRC0 FMT_NUM_I32 OPR_SRC_NOLIT 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_FREXP_EXP_I32_F64 Extract the exponent of a double-precision float input and store the result as a signed 32-bit integer into a vector register. ENC_VOP1 default 48 VDST FMT_NUM_I32 OPR_VGPR 32 SRC0 FMT_NUM_F64 OPR_SRC 64 VOP1_INST_LITERAL has_lit 48 VDST FMT_NUM_I32 OPR_VGPR 32 SIMM32 FMT_NUM_F64 OPR_SRC 64 VOP1_VOP_DPP has_dpp 48 VDST FMT_NUM_I32 OPR_VGPR 32 VSRC0 FMT_NUM_F64 OPR_SRC 64 ENC_VOP3 default 368 VDST FMT_NUM_I32 OPR_VGPR 32 SRC0 FMT_NUM_F64 OPR_SRC_NOLIT 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_FREXP_MANT_F64 Extract the binary significand, or mantissa, of a double-precision float input and store the result as a double-precision float into a vector register. ENC_VOP1 default 49 VDST FMT_NUM_F64 OPR_VGPR 64 SRC0 FMT_NUM_F64 OPR_SRC 64 VOP1_INST_LITERAL has_lit 49 VDST FMT_NUM_F64 OPR_VGPR 64 SIMM32 FMT_NUM_F64 OPR_SRC 64 VOP1_VOP_DPP has_dpp 49 VDST FMT_NUM_F64 OPR_VGPR 64 VSRC0 FMT_NUM_F64 OPR_SRC 64 ENC_VOP3 default 369 VDST FMT_NUM_F64 OPR_VGPR 64 SRC0 FMT_NUM_F64 OPR_SRC_NOLIT 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_FRACT_F64 Compute the fractional portion of a double-precision float input and store the result in floating point format into a vector register. ENC_VOP1 default 50 VDST FMT_NUM_F64 OPR_VGPR 64 SRC0 FMT_NUM_F64 OPR_SRC 64 VOP1_INST_LITERAL has_lit 50 VDST FMT_NUM_F64 OPR_VGPR 64 SIMM32 FMT_NUM_F64 OPR_SRC 64 VOP1_VOP_DPP has_dpp 50 VDST FMT_NUM_F64 OPR_VGPR 64 VSRC0 FMT_NUM_F64 OPR_SRC 64 ENC_VOP3 default 370 VDST FMT_NUM_F64 OPR_VGPR 64 SRC0 FMT_NUM_F64 OPR_SRC_NOLIT 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_FREXP_EXP_I32_F32 Extract the exponent of a single-precision float input and store the result as a signed 32-bit integer into a vector register. ENC_VOP1 default 51 VDST FMT_NUM_I32 OPR_VGPR 32 SRC0 FMT_NUM_F32 OPR_SRC 32 VOP1_INST_LITERAL has_lit 51 VDST FMT_NUM_I32 OPR_VGPR 32 SIMM32 FMT_NUM_F32 OPR_SRC 32 VOP1_VOP_DPP has_dpp 51 VDST FMT_NUM_I32 OPR_VGPR 32 VSRC0 FMT_NUM_F32 OPR_SRC 32 VOP1_VOP_SDWA has_sdwa 51 VDST FMT_NUM_I32 OPR_VGPR 32 VSRC0 FMT_NUM_F32 OPR_SRC 32 ENC_VOP3 default 371 VDST FMT_NUM_I32 OPR_VGPR 32 SRC0 FMT_NUM_F32 OPR_SRC_NOLIT 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_FREXP_MANT_F32 Extract the binary significand, or mantissa, of a single-precision float input and store the result as a single-precision float into a vector register. ENC_VOP1 default 52 VDST FMT_NUM_F32 OPR_VGPR 32 SRC0 FMT_NUM_F32 OPR_SRC 32 VOP1_INST_LITERAL has_lit 52 VDST FMT_NUM_F32 OPR_VGPR 32 SIMM32 FMT_NUM_F32 OPR_SRC 32 VOP1_VOP_DPP has_dpp 52 VDST FMT_NUM_F32 OPR_VGPR 32 VSRC0 FMT_NUM_F32 OPR_SRC 32 VOP1_VOP_SDWA has_sdwa 52 VDST FMT_NUM_F32 OPR_VGPR 32 VSRC0 FMT_NUM_F32 OPR_SRC 32 ENC_VOP3 default 372 VDST FMT_NUM_F32 OPR_VGPR 32 SRC0 FMT_NUM_F32 OPR_SRC_NOLIT 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CLREXCP Clear this wave's exception state in the vector ALU. ENC_VOP1 default 53 VOP1_INST_LITERAL has_lit 53 ENC_VOP3 default 373 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_SCREEN_PARTITION_4SE_B32 4SE version of LUT instruction for screen partitioning/filtering. This opcode is intended to help accelerate screen partitioning in the 4SE case only. 2SE and 1SE cases use normal ALU instructions. ENC_VOP1 default 55 VDST FMT_NUM_B32 OPR_VGPR 32 SRC0 FMT_NUM_B32 OPR_SRC 32 VOP1_INST_LITERAL has_lit 55 VDST FMT_NUM_B32 OPR_VGPR 32 SIMM32 FMT_NUM_B32 OPR_SRC 32 VOP1_VOP_DPP has_dpp 55 VDST FMT_NUM_B32 OPR_VGPR 32 VSRC0 FMT_NUM_B32 OPR_SRC 32 VOP1_VOP_SDWA has_sdwa 55 VDST FMT_NUM_B32 OPR_VGPR 32 VSRC0 FMT_NUM_B32 OPR_SRC 32 ENC_VOP3 default 375 VDST FMT_NUM_B32 OPR_VGPR 32 SRC0 FMT_NUM_B32 OPR_SRC_NOLIT 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_MOV_B64 Move data from a 64-bit vector input into a vector register. ENC_VOP1 default 56 VDST FMT_NUM_U64 OPR_VGPR 64 SRC0 FMT_NUM_U64 OPR_SRC 64 VOP1_INST_LITERAL has_lit 56 VDST FMT_NUM_U64 OPR_VGPR 64 SIMM32 FMT_NUM_U64 OPR_SRC 64 VOP1_VOP_DPP has_dpp 56 VDST FMT_NUM_U64 OPR_VGPR 64 VSRC0 FMT_NUM_U64 OPR_SRC 64 VOP1_VOP_SDWA has_sdwa 56 VDST FMT_NUM_U64 OPR_VGPR 64 VSRC0 FMT_NUM_U64 OPR_SRC 64 ENC_VOP3 default 376 VDST FMT_NUM_U64 OPR_VGPR 64 SRC0 FMT_NUM_U64 OPR_SRC_NOLIT 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CVT_F16_U16 Convert from an unsigned 16-bit integer input to a half-precision float value and store the result into a vector register. ENC_VOP1 default 57 VDST FMT_NUM_F16 OPR_VGPR 16 SRC0 FMT_NUM_U16 OPR_SRC 16 VOP1_INST_LITERAL has_lit 57 VDST FMT_NUM_F16 OPR_VGPR 16 SIMM32 FMT_NUM_U16 OPR_SRC 16 VOP1_VOP_DPP has_dpp 57 VDST FMT_NUM_F16 OPR_VGPR 16 VSRC0 FMT_NUM_U16 OPR_SRC 16 VOP1_VOP_SDWA has_sdwa 57 VDST FMT_NUM_F16 OPR_VGPR 16 VSRC0 FMT_NUM_U16 OPR_SRC 16 ENC_VOP3 default 377 VDST FMT_NUM_F16 OPR_VGPR 16 SRC0 FMT_NUM_U16 OPR_SRC_NOLIT 16 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CVT_F16_I16 Convert from a signed 16-bit integer input to a half-precision float value and store the result into a vector register. ENC_VOP1 default 58 VDST FMT_NUM_F16 OPR_VGPR 16 SRC0 FMT_NUM_I16 OPR_SRC 16 VOP1_INST_LITERAL has_lit 58 VDST FMT_NUM_F16 OPR_VGPR 16 SIMM32 FMT_NUM_I16 OPR_SRC 16 VOP1_VOP_DPP has_dpp 58 VDST FMT_NUM_F16 OPR_VGPR 16 VSRC0 FMT_NUM_I16 OPR_SRC 16 VOP1_VOP_SDWA has_sdwa 58 VDST FMT_NUM_F16 OPR_VGPR 16 VSRC0 FMT_NUM_I16 OPR_SRC 16 ENC_VOP3 default 378 VDST FMT_NUM_F16 OPR_VGPR 16 SRC0 FMT_NUM_I16 OPR_SRC_NOLIT 16 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CVT_U16_F16 Convert from a half-precision float input to an unsigned 16-bit integer value and store the result into a vector register. ENC_VOP1 default 59 VDST FMT_NUM_U16 OPR_VGPR 16 SRC0 FMT_NUM_F16 OPR_SRC 16 VOP1_INST_LITERAL has_lit 59 VDST FMT_NUM_U16 OPR_VGPR 16 SIMM32 FMT_NUM_F16 OPR_SRC 16 VOP1_VOP_DPP has_dpp 59 VDST FMT_NUM_U16 OPR_VGPR 16 VSRC0 FMT_NUM_F16 OPR_SRC 16 VOP1_VOP_SDWA has_sdwa 59 VDST FMT_NUM_U16 OPR_VGPR 16 VSRC0 FMT_NUM_F16 OPR_SRC 16 ENC_VOP3 default 379 VDST FMT_NUM_U16 OPR_VGPR 16 SRC0 FMT_NUM_F16 OPR_SRC_NOLIT 16 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CVT_I16_F16 Convert from a half-precision float input to a signed 16-bit integer value and store the result into a vector register. ENC_VOP1 default 60 VDST FMT_NUM_I16 OPR_VGPR 16 SRC0 FMT_NUM_F16 OPR_SRC 16 VOP1_INST_LITERAL has_lit 60 VDST FMT_NUM_I16 OPR_VGPR 16 SIMM32 FMT_NUM_F16 OPR_SRC 16 VOP1_VOP_DPP has_dpp 60 VDST FMT_NUM_I16 OPR_VGPR 16 VSRC0 FMT_NUM_F16 OPR_SRC 16 VOP1_VOP_SDWA has_sdwa 60 VDST FMT_NUM_I16 OPR_VGPR 16 VSRC0 FMT_NUM_F16 OPR_SRC 16 ENC_VOP3 default 380 VDST FMT_NUM_I16 OPR_VGPR 16 SRC0 FMT_NUM_F16 OPR_SRC_NOLIT 16 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_RCP_F16 Calculate the reciprocal of the half-precision float input using IEEE rules and store the result into a vector register. ENC_VOP1 default 61 VDST FMT_NUM_F16 OPR_VGPR 16 SRC0 FMT_NUM_F16 OPR_SRC 16 VOP1_INST_LITERAL has_lit 61 VDST FMT_NUM_F16 OPR_VGPR 16 SIMM32 FMT_NUM_F16 OPR_SRC 16 VOP1_VOP_DPP has_dpp 61 VDST FMT_NUM_F16 OPR_VGPR 16 VSRC0 FMT_NUM_F16 OPR_SRC 16 VOP1_VOP_SDWA has_sdwa 61 VDST FMT_NUM_F16 OPR_VGPR 16 VSRC0 FMT_NUM_F16 OPR_SRC 16 ENC_VOP3 default 381 VDST FMT_NUM_F16 OPR_VGPR 16 SRC0 FMT_NUM_F16 OPR_SRC_NOLIT 16 VALU TRANSCENDENTAL FALSE FALSE FALSE FALSE FALSE V_SQRT_F16 Calculate the square root of the half-precision float input using IEEE rules and store the result into a vector register. ENC_VOP1 default 62 VDST FMT_NUM_F16 OPR_VGPR 16 SRC0 FMT_NUM_F16 OPR_SRC 16 VOP1_INST_LITERAL has_lit 62 VDST FMT_NUM_F16 OPR_VGPR 16 SIMM32 FMT_NUM_F16 OPR_SRC 16 VOP1_VOP_DPP has_dpp 62 VDST FMT_NUM_F16 OPR_VGPR 16 VSRC0 FMT_NUM_F16 OPR_SRC 16 VOP1_VOP_SDWA has_sdwa 62 VDST FMT_NUM_F16 OPR_VGPR 16 VSRC0 FMT_NUM_F16 OPR_SRC 16 ENC_VOP3 default 382 VDST FMT_NUM_F16 OPR_VGPR 16 SRC0 FMT_NUM_F16 OPR_SRC_NOLIT 16 VALU TRANSCENDENTAL FALSE FALSE FALSE FALSE FALSE V_RSQ_F16 Calculate the reciprocal of the square root of the half-precision float input using IEEE rules and store the result into a vector register. ENC_VOP1 default 63 VDST FMT_NUM_F16 OPR_VGPR 16 SRC0 FMT_NUM_F16 OPR_SRC 16 VOP1_INST_LITERAL has_lit 63 VDST FMT_NUM_F16 OPR_VGPR 16 SIMM32 FMT_NUM_F16 OPR_SRC 16 VOP1_VOP_DPP has_dpp 63 VDST FMT_NUM_F16 OPR_VGPR 16 VSRC0 FMT_NUM_F16 OPR_SRC 16 VOP1_VOP_SDWA has_sdwa 63 VDST FMT_NUM_F16 OPR_VGPR 16 VSRC0 FMT_NUM_F16 OPR_SRC 16 ENC_VOP3 default 383 VDST FMT_NUM_F16 OPR_VGPR 16 SRC0 FMT_NUM_F16 OPR_SRC_NOLIT 16 VALU TRANSCENDENTAL FALSE FALSE FALSE FALSE FALSE V_LOG_F16 Calculate the base 2 logarithm of the half-precision float input and store the result into a vector register. ENC_VOP1 default 64 VDST FMT_NUM_F16 OPR_VGPR 16 SRC0 FMT_NUM_F16 OPR_SRC 16 VOP1_INST_LITERAL has_lit 64 VDST FMT_NUM_F16 OPR_VGPR 16 SIMM32 FMT_NUM_F16 OPR_SRC 16 VOP1_VOP_DPP has_dpp 64 VDST FMT_NUM_F16 OPR_VGPR 16 VSRC0 FMT_NUM_F16 OPR_SRC 16 VOP1_VOP_SDWA has_sdwa 64 VDST FMT_NUM_F16 OPR_VGPR 16 VSRC0 FMT_NUM_F16 OPR_SRC 16 ENC_VOP3 default 384 VDST FMT_NUM_F16 OPR_VGPR 16 SRC0 FMT_NUM_F16 OPR_SRC_NOLIT 16 VALU TRANSCENDENTAL FALSE FALSE FALSE FALSE FALSE V_EXP_F16 Calculate 2 raised to the power of the half-precision float input and store the result into a vector register. ENC_VOP1 default 65 VDST FMT_NUM_F16 OPR_VGPR 16 SRC0 FMT_NUM_F16 OPR_SRC 16 VOP1_INST_LITERAL has_lit 65 VDST FMT_NUM_F16 OPR_VGPR 16 SIMM32 FMT_NUM_F16 OPR_SRC 16 VOP1_VOP_DPP has_dpp 65 VDST FMT_NUM_F16 OPR_VGPR 16 VSRC0 FMT_NUM_F16 OPR_SRC 16 VOP1_VOP_SDWA has_sdwa 65 VDST FMT_NUM_F16 OPR_VGPR 16 VSRC0 FMT_NUM_F16 OPR_SRC 16 ENC_VOP3 default 385 VDST FMT_NUM_F16 OPR_VGPR 16 SRC0 FMT_NUM_F16 OPR_SRC_NOLIT 16 VALU TRANSCENDENTAL FALSE FALSE FALSE FALSE FALSE V_FREXP_MANT_F16 Extract the binary significand, or mantissa, of a half-precision float input and store the result as a half-precision float into a vector register. ENC_VOP1 default 66 VDST FMT_NUM_F16 OPR_VGPR 16 SRC0 FMT_NUM_F16 OPR_SRC 16 VOP1_INST_LITERAL has_lit 66 VDST FMT_NUM_F16 OPR_VGPR 16 SIMM32 FMT_NUM_F16 OPR_SRC 16 VOP1_VOP_DPP has_dpp 66 VDST FMT_NUM_F16 OPR_VGPR 16 VSRC0 FMT_NUM_F16 OPR_SRC 16 VOP1_VOP_SDWA has_sdwa 66 VDST FMT_NUM_F16 OPR_VGPR 16 VSRC0 FMT_NUM_F16 OPR_SRC 16 ENC_VOP3 default 386 VDST FMT_NUM_F16 OPR_VGPR 16 SRC0 FMT_NUM_F16 OPR_SRC_NOLIT 16 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_FREXP_EXP_I16_F16 Extract the exponent of a half-precision float input and store the result as a signed 16-bit integer into a vector register. ENC_VOP1 default 67 VDST FMT_NUM_I16 OPR_VGPR 16 SRC0 FMT_NUM_F16 OPR_SRC 16 VOP1_INST_LITERAL has_lit 67 VDST FMT_NUM_I16 OPR_VGPR 16 SIMM32 FMT_NUM_F16 OPR_SRC 16 VOP1_VOP_DPP has_dpp 67 VDST FMT_NUM_I16 OPR_VGPR 16 VSRC0 FMT_NUM_F16 OPR_SRC 16 VOP1_VOP_SDWA has_sdwa 67 VDST FMT_NUM_I16 OPR_VGPR 16 VSRC0 FMT_NUM_F16 OPR_SRC 16 ENC_VOP3 default 387 VDST FMT_NUM_I16 OPR_VGPR 16 SRC0 FMT_NUM_F16 OPR_SRC_NOLIT 16 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_FLOOR_F16 Round the half-precision float input down to previous integer and store the result in floating point format into a vector register. ENC_VOP1 default 68 VDST FMT_NUM_F16 OPR_VGPR 16 SRC0 FMT_NUM_F16 OPR_SRC 16 VOP1_INST_LITERAL has_lit 68 VDST FMT_NUM_F16 OPR_VGPR 16 SIMM32 FMT_NUM_F16 OPR_SRC 16 VOP1_VOP_DPP has_dpp 68 VDST FMT_NUM_F16 OPR_VGPR 16 VSRC0 FMT_NUM_F16 OPR_SRC 16 VOP1_VOP_SDWA has_sdwa 68 VDST FMT_NUM_F16 OPR_VGPR 16 VSRC0 FMT_NUM_F16 OPR_SRC 16 ENC_VOP3 default 388 VDST FMT_NUM_F16 OPR_VGPR 16 SRC0 FMT_NUM_F16 OPR_SRC_NOLIT 16 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CEIL_F16 Round the half-precision float input up to next integer and store the result in floating point format into a vector register. ENC_VOP1 default 69 VDST FMT_NUM_F16 OPR_VGPR 16 SRC0 FMT_NUM_F16 OPR_SRC 16 VOP1_INST_LITERAL has_lit 69 VDST FMT_NUM_F16 OPR_VGPR 16 SIMM32 FMT_NUM_F16 OPR_SRC 16 VOP1_VOP_DPP has_dpp 69 VDST FMT_NUM_F16 OPR_VGPR 16 VSRC0 FMT_NUM_F16 OPR_SRC 16 VOP1_VOP_SDWA has_sdwa 69 VDST FMT_NUM_F16 OPR_VGPR 16 VSRC0 FMT_NUM_F16 OPR_SRC 16 ENC_VOP3 default 389 VDST FMT_NUM_F16 OPR_VGPR 16 SRC0 FMT_NUM_F16 OPR_SRC_NOLIT 16 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_TRUNC_F16 Compute the integer part of a half-precision float input using round toward zero semantics and store the result in floating point format into a vector register. ENC_VOP1 default 70 VDST FMT_NUM_F16 OPR_VGPR 16 SRC0 FMT_NUM_F16 OPR_SRC 16 VOP1_INST_LITERAL has_lit 70 VDST FMT_NUM_F16 OPR_VGPR 16 SIMM32 FMT_NUM_F16 OPR_SRC 16 VOP1_VOP_DPP has_dpp 70 VDST FMT_NUM_F16 OPR_VGPR 16 VSRC0 FMT_NUM_F16 OPR_SRC 16 VOP1_VOP_SDWA has_sdwa 70 VDST FMT_NUM_F16 OPR_VGPR 16 VSRC0 FMT_NUM_F16 OPR_SRC 16 ENC_VOP3 default 390 VDST FMT_NUM_F16 OPR_VGPR 16 SRC0 FMT_NUM_F16 OPR_SRC_NOLIT 16 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_RNDNE_F16 Round the half-precision float input to the nearest even integer and store the result in floating point format into a vector register. ENC_VOP1 default 71 VDST FMT_NUM_F16 OPR_VGPR 16 SRC0 FMT_NUM_F16 OPR_SRC 16 VOP1_INST_LITERAL has_lit 71 VDST FMT_NUM_F16 OPR_VGPR 16 SIMM32 FMT_NUM_F16 OPR_SRC 16 VOP1_VOP_DPP has_dpp 71 VDST FMT_NUM_F16 OPR_VGPR 16 VSRC0 FMT_NUM_F16 OPR_SRC 16 VOP1_VOP_SDWA has_sdwa 71 VDST FMT_NUM_F16 OPR_VGPR 16 VSRC0 FMT_NUM_F16 OPR_SRC 16 ENC_VOP3 default 391 VDST FMT_NUM_F16 OPR_VGPR 16 SRC0 FMT_NUM_F16 OPR_SRC_NOLIT 16 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_FRACT_F16 Compute the fractional portion of a half-precision float input and store the result in floating point format into a vector register. ENC_VOP1 default 72 VDST FMT_NUM_F16 OPR_VGPR 16 SRC0 FMT_NUM_F16 OPR_SRC 16 VOP1_INST_LITERAL has_lit 72 VDST FMT_NUM_F16 OPR_VGPR 16 SIMM32 FMT_NUM_F16 OPR_SRC 16 VOP1_VOP_DPP has_dpp 72 VDST FMT_NUM_F16 OPR_VGPR 16 VSRC0 FMT_NUM_F16 OPR_SRC 16 VOP1_VOP_SDWA has_sdwa 72 VDST FMT_NUM_F16 OPR_VGPR 16 VSRC0 FMT_NUM_F16 OPR_SRC 16 ENC_VOP3 default 392 VDST FMT_NUM_F16 OPR_VGPR 16 SRC0 FMT_NUM_F16 OPR_SRC_NOLIT 16 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_SIN_F16 Calculate the trigonometric sine of a half-precision float value using IEEE rules and store the result into a vector register. The operand is calculated by scaling the vector input by 2 PI. ENC_VOP1 default 73 VDST FMT_NUM_F16 OPR_VGPR 16 SRC0 FMT_NUM_F16 OPR_SRC 16 VOP1_INST_LITERAL has_lit 73 VDST FMT_NUM_F16 OPR_VGPR 16 SIMM32 FMT_NUM_F16 OPR_SRC 16 VOP1_VOP_DPP has_dpp 73 VDST FMT_NUM_F16 OPR_VGPR 16 VSRC0 FMT_NUM_F16 OPR_SRC 16 VOP1_VOP_SDWA has_sdwa 73 VDST FMT_NUM_F16 OPR_VGPR 16 VSRC0 FMT_NUM_F16 OPR_SRC 16 ENC_VOP3 default 393 VDST FMT_NUM_F16 OPR_VGPR 16 SRC0 FMT_NUM_F16 OPR_SRC_NOLIT 16 VALU TRANSCENDENTAL FALSE FALSE FALSE FALSE FALSE V_COS_F16 Calculate the trigonometric cosine of a half-precision float value using IEEE rules and store the result into a vector register. The operand is calculated by scaling the vector input by 2 PI. ENC_VOP1 default 74 VDST FMT_NUM_F16 OPR_VGPR 16 SRC0 FMT_NUM_F16 OPR_SRC 16 VOP1_INST_LITERAL has_lit 74 VDST FMT_NUM_F16 OPR_VGPR 16 SIMM32 FMT_NUM_F16 OPR_SRC 16 VOP1_VOP_DPP has_dpp 74 VDST FMT_NUM_F16 OPR_VGPR 16 VSRC0 FMT_NUM_F16 OPR_SRC 16 VOP1_VOP_SDWA has_sdwa 74 VDST FMT_NUM_F16 OPR_VGPR 16 VSRC0 FMT_NUM_F16 OPR_SRC 16 ENC_VOP3 default 394 VDST FMT_NUM_F16 OPR_VGPR 16 SRC0 FMT_NUM_F16 OPR_SRC_NOLIT 16 VALU TRANSCENDENTAL FALSE FALSE FALSE FALSE FALSE V_EXP_LEGACY_F32 Power with legacy semantics. ENC_VOP1 default 75 VDST FMT_NUM_F32 OPR_VGPR 32 SRC0 FMT_NUM_F32 OPR_SRC 32 VOP1_INST_LITERAL has_lit 75 VDST FMT_NUM_F32 OPR_VGPR 32 SIMM32 FMT_NUM_F32 OPR_SRC 32 VOP1_VOP_DPP has_dpp 75 VDST FMT_NUM_F32 OPR_VGPR 32 VSRC0 FMT_NUM_F32 OPR_SRC 32 VOP1_VOP_SDWA has_sdwa 75 VDST FMT_NUM_F32 OPR_VGPR 32 VSRC0 FMT_NUM_F32 OPR_SRC 32 ENC_VOP3 default 395 VDST FMT_NUM_F32 OPR_VGPR 32 SRC0 FMT_NUM_F32 OPR_SRC_NOLIT 32 VALU TRANSCENDENTAL FALSE FALSE FALSE FALSE FALSE V_LOG_LEGACY_F32 Base 2 logarithm with legacy semantics. ENC_VOP1 default 76 VDST FMT_NUM_F32 OPR_VGPR 32 SRC0 FMT_NUM_F32 OPR_SRC 32 VOP1_INST_LITERAL has_lit 76 VDST FMT_NUM_F32 OPR_VGPR 32 SIMM32 FMT_NUM_F32 OPR_SRC 32 VOP1_VOP_DPP has_dpp 76 VDST FMT_NUM_F32 OPR_VGPR 32 VSRC0 FMT_NUM_F32 OPR_SRC 32 VOP1_VOP_SDWA has_sdwa 76 VDST FMT_NUM_F32 OPR_VGPR 32 VSRC0 FMT_NUM_F32 OPR_SRC 32 ENC_VOP3 default 396 VDST FMT_NUM_F32 OPR_VGPR 32 SRC0 FMT_NUM_F32 OPR_SRC_NOLIT 32 VALU TRANSCENDENTAL FALSE FALSE FALSE FALSE FALSE V_CVT_NORM_I16_F16 Convert from a half-precision float input to a signed normalized short and store the result into a vector register. ENC_VOP1 default 77 VDST FMT_NUM_I16 OPR_VGPR 16 SRC0 FMT_NUM_F16 OPR_SRC 16 VOP1_INST_LITERAL has_lit 77 VDST FMT_NUM_I16 OPR_VGPR 16 SIMM32 FMT_NUM_F16 OPR_SRC 16 VOP1_VOP_DPP has_dpp 77 VDST FMT_NUM_I16 OPR_VGPR 16 VSRC0 FMT_NUM_F16 OPR_SRC 16 VOP1_VOP_SDWA has_sdwa 77 VDST FMT_NUM_I16 OPR_VGPR 16 VSRC0 FMT_NUM_F16 OPR_SRC 16 ENC_VOP3 default 397 VDST FMT_NUM_I16 OPR_VGPR 16 SRC0 FMT_NUM_F16 OPR_SRC_NOLIT 16 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CVT_NORM_U16_F16 Convert from a half-precision float input to an unsigned normalized short and store the result into a vector register. ENC_VOP1 default 78 VDST FMT_NUM_U16 OPR_VGPR 16 SRC0 FMT_NUM_F16 OPR_SRC 16 VOP1_INST_LITERAL has_lit 78 VDST FMT_NUM_U16 OPR_VGPR 16 SIMM32 FMT_NUM_F16 OPR_SRC 16 VOP1_VOP_DPP has_dpp 78 VDST FMT_NUM_U16 OPR_VGPR 16 VSRC0 FMT_NUM_F16 OPR_SRC 16 VOP1_VOP_SDWA has_sdwa 78 VDST FMT_NUM_U16 OPR_VGPR 16 VSRC0 FMT_NUM_F16 OPR_SRC 16 ENC_VOP3 default 398 VDST FMT_NUM_U16 OPR_VGPR 16 SRC0 FMT_NUM_F16 OPR_SRC_NOLIT 16 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_SAT_PK_U8_I16 Given 2 signed 16-bit integer inputs, saturate each input over an unsigned 8-bit integer range, pack the resulting values into a packed 16-bit value and store the result into a vector register. ENC_VOP1 default 79 VDST FMT_NUM_PK2_U8 OPR_VGPR 16 SRC0 FMT_NUM_PK2_I16 OPR_SRC 32 VOP1_INST_LITERAL has_lit 79 VDST FMT_NUM_PK2_U8 OPR_VGPR 16 SIMM32 FMT_NUM_PK2_I16 OPR_SRC 32 VOP1_VOP_DPP has_dpp 79 VDST FMT_NUM_PK2_U8 OPR_VGPR 16 VSRC0 FMT_NUM_PK2_I16 OPR_SRC 32 VOP1_VOP_SDWA has_sdwa 79 VDST FMT_NUM_PK2_U8 OPR_VGPR 16 VSRC0 FMT_NUM_PK2_I16 OPR_SRC 32 ENC_VOP3 default 399 VDST FMT_NUM_PK2_U8 OPR_VGPR 16 SRC0 FMT_NUM_PK2_I16 OPR_SRC_NOLIT 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_SWAP_B32 Swap the values in two vector registers. ENC_VOP1 default 81 VDST FMT_NUM_B32 OPR_VGPR 32 SRC0 FMT_NUM_B32 OPR_SRC_VGPR 32 ENC_VOP3 default 401 VDST FMT_NUM_B32 OPR_VGPR 32 SRC0 FMT_NUM_B32 OPR_SRC_VGPR 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_ACCVGPR_MOV_B32 Move data from one accumulator register to another accumulator register. ENC_VOP1 default 82 VDST FMT_NUM_B32 OPR_ACCVGPR 32 SRC0 FMT_NUM_B32 OPR_SRC_ACCVGPR 32 VOP1_INST_LITERAL has_lit 82 VDST FMT_NUM_B32 OPR_ACCVGPR 32 SIMM32 FMT_NUM_B32 OPR_SRC_ACCVGPR 32 VOP1_VOP_DPP has_dpp 82 VDST FMT_NUM_B32 OPR_ACCVGPR 32 VSRC0 FMT_NUM_B32 OPR_SRC_ACCVGPR 32 VOP1_VOP_SDWA has_sdwa 82 VDST FMT_NUM_B32 OPR_ACCVGPR 32 VSRC0 FMT_NUM_B32 OPR_SRC_ACCVGPR 32 ENC_VOP3 default 402 VDST FMT_NUM_B32 OPR_ACCVGPR 32 SRC0 FMT_NUM_B32 OPR_SRC_ACCVGPR 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CVT_F32_FP8 Convert from an FP8 float input to a single-precision float value and store the result into a vector register. ENC_VOP1 default 84 VDST FMT_NUM_F32 OPR_VGPR 32 SRC0 FMT_NUM_FP8 OPR_SRC 8 VOP1_INST_LITERAL has_lit 84 VDST FMT_NUM_F32 OPR_VGPR 32 SIMM32 FMT_NUM_FP8 OPR_SRC 8 VOP1_VOP_DPP has_dpp 84 VDST FMT_NUM_F32 OPR_VGPR 32 VSRC0 FMT_NUM_FP8 OPR_SRC 8 VOP1_VOP_SDWA has_sdwa 84 VDST FMT_NUM_F32 OPR_VGPR 32 VSRC0 FMT_NUM_FP8 OPR_SRC 8 ENC_VOP3 default 404 VDST FMT_NUM_F32 OPR_VGPR 32 SRC0 FMT_NUM_FP8 OPR_SRC_NOLIT 8 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CVT_F32_BF8 Convert from a BF8 float input to a single-precision float value and store the result into a vector register. ENC_VOP1 default 85 VDST FMT_NUM_F32 OPR_VGPR 32 SRC0 FMT_NUM_BF8 OPR_SRC 8 VOP1_INST_LITERAL has_lit 85 VDST FMT_NUM_F32 OPR_VGPR 32 SIMM32 FMT_NUM_BF8 OPR_SRC 8 VOP1_VOP_DPP has_dpp 85 VDST FMT_NUM_F32 OPR_VGPR 32 VSRC0 FMT_NUM_BF8 OPR_SRC 8 VOP1_VOP_SDWA has_sdwa 85 VDST FMT_NUM_F32 OPR_VGPR 32 VSRC0 FMT_NUM_BF8 OPR_SRC 8 ENC_VOP3 default 405 VDST FMT_NUM_F32 OPR_VGPR 32 SRC0 FMT_NUM_BF8 OPR_SRC_NOLIT 8 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CVT_PK_F32_FP8 Convert from a packed 2-component FP8 float input to a packed single-precision float value and store the result into a vector register. ENC_VOP1 default 86 VDST FMT_NUM_PK2_F32 OPR_VGPR 64 SRC0 FMT_NUM_PK2_FP8 OPR_SRC 16 VOP1_INST_LITERAL has_lit 86 VDST FMT_NUM_PK2_F32 OPR_VGPR 64 SIMM32 FMT_NUM_PK2_FP8 OPR_SRC 16 VOP1_VOP_DPP has_dpp 86 VDST FMT_NUM_PK2_F32 OPR_VGPR 64 VSRC0 FMT_NUM_PK2_FP8 OPR_SRC 16 VOP1_VOP_SDWA has_sdwa 86 VDST FMT_NUM_PK2_F32 OPR_VGPR 64 VSRC0 FMT_NUM_PK2_FP8 OPR_SRC 16 ENC_VOP3 default 406 VDST FMT_NUM_PK2_F32 OPR_VGPR 64 SRC0 FMT_NUM_PK2_FP8 OPR_SRC_NOLIT 16 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CVT_PK_F32_BF8 Convert from a packed 2-component BF8 float input to a packed single-precision float value and store the result into a vector register. ENC_VOP1 default 87 VDST FMT_NUM_PK2_F32 OPR_VGPR 64 SRC0 FMT_NUM_PK2_BF8 OPR_SRC 16 VOP1_INST_LITERAL has_lit 87 VDST FMT_NUM_PK2_F32 OPR_VGPR 64 SIMM32 FMT_NUM_PK2_BF8 OPR_SRC 16 VOP1_VOP_DPP has_dpp 87 VDST FMT_NUM_PK2_F32 OPR_VGPR 64 VSRC0 FMT_NUM_PK2_BF8 OPR_SRC 16 VOP1_VOP_SDWA has_sdwa 87 VDST FMT_NUM_PK2_F32 OPR_VGPR 64 VSRC0 FMT_NUM_PK2_BF8 OPR_SRC 16 ENC_VOP3 default 407 VDST FMT_NUM_PK2_F32 OPR_VGPR 64 SRC0 FMT_NUM_PK2_BF8 OPR_SRC_NOLIT 16 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_PRNG_B32 Generate a pseudorandom number using an LFSR (linear feedback shift register) seeded with the vector input, then store the result into a vector register. ENC_VOP1 default 88 VDST FMT_NUM_B32 OPR_VGPR 32 SRC0 FMT_NUM_B32 OPR_SRC 32 VOP1_INST_LITERAL has_lit 88 VDST FMT_NUM_B32 OPR_VGPR 32 SIMM32 FMT_NUM_B32 OPR_SRC 32 VOP1_VOP_DPP has_dpp 88 VDST FMT_NUM_B32 OPR_VGPR 32 VSRC0 FMT_NUM_B32 OPR_SRC 32 VOP1_VOP_SDWA has_sdwa 88 VDST FMT_NUM_B32 OPR_VGPR 32 VSRC0 FMT_NUM_B32 OPR_SRC 32 ENC_VOP3 default 408 VDST FMT_NUM_B32 OPR_VGPR 32 SRC0 FMT_NUM_B32 OPR_SRC_NOLIT 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_PERMLANE16_SWAP_B32 Swap data between two vector registers. Odd rows of the first operand are swapped with even rows of the second operand (one row is 16 lanes). ENC_VOP1 default 89 VDST FMT_NUM_B32 OPR_VGPR 32 SRC0 FMT_NUM_B32 OPR_SRC_VGPR 32 ENC_VOP3 default 409 VDST FMT_NUM_B32 OPR_VGPR 32 SRC0 FMT_NUM_B32 OPR_SRC_VGPR 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_PERMLANE32_SWAP_B32 Swap data between two vector registers. Rows 2 and 3 of the first operand are swapped with rows 0 and 1 of the second operand (one row is 16 lanes). ENC_VOP1 default 90 VDST FMT_NUM_B32 OPR_VGPR 32 SRC0 FMT_NUM_B32 OPR_SRC_VGPR 32 ENC_VOP3 default 410 VDST FMT_NUM_B32 OPR_VGPR 32 SRC0 FMT_NUM_B32 OPR_SRC_VGPR 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CVT_F32_BF16 Convert from a BF16 float input to a single-precision float value and store the result into a vector register. ENC_VOP1 default 91 VDST FMT_NUM_F32 OPR_VGPR 32 SRC0 FMT_NUM_BF16 OPR_SRC 16 VOP1_INST_LITERAL has_lit 91 VDST FMT_NUM_F32 OPR_VGPR 32 SIMM32 FMT_NUM_BF16 OPR_SRC 16 VOP1_VOP_DPP has_dpp 91 VDST FMT_NUM_F32 OPR_VGPR 32 VSRC0 FMT_NUM_BF16 OPR_SRC 16 VOP1_VOP_SDWA has_sdwa 91 VDST FMT_NUM_F32 OPR_VGPR 32 VSRC0 FMT_NUM_BF16 OPR_SRC 16 ENC_VOP3 default 411 VDST FMT_NUM_F32 OPR_VGPR 32 SRC0 FMT_NUM_BF16 OPR_SRC_NOLIT 16 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CNDMASK_B32 Copy data from one of two inputs based on the per-lane condition code and store the result into a vector register. ENC_VOP2 default 0 VDST FMT_NUM_B32 OPR_VGPR 32 SRC0 FMT_NUM_B32 OPR_SRC 32 VSRC1 FMT_NUM_B32 OPR_VGPR 32 FMT_NUM_M64 OPR_VCC 64 VOP2_INST_LITERAL has_lit 0 VDST FMT_NUM_B32 OPR_VGPR 32 SIMM32 FMT_NUM_B32 OPR_SRC 32 VSRC1 FMT_NUM_B32 OPR_VGPR 32 FMT_NUM_M64 OPR_VCC 64 VOP2_VOP_DPP has_dpp 0 VDST FMT_NUM_B32 OPR_VGPR 32 VSRC0 FMT_NUM_B32 OPR_SRC 32 VSRC1 FMT_NUM_B32 OPR_VGPR 32 FMT_NUM_M64 OPR_VCC 64 VOP2_VOP_SDWA has_sdwa 0 VDST FMT_NUM_B32 OPR_VGPR 32 VSRC0 FMT_NUM_B32 OPR_SRC 32 VSRC1 FMT_NUM_B32 OPR_VGPR 32 FMT_NUM_M64 OPR_VCC 64 ENC_VOP3 default 256 VDST FMT_NUM_B32 OPR_VGPR 32 SRC0 FMT_NUM_B32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_B32 OPR_SRC_SIMPLE 32 SRC2 FMT_NUM_M64 OPR_SREG 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_ADD_F32 Add two floating point inputs and store the result into a vector register. ENC_VOP2 default 1 VDST FMT_NUM_F32 OPR_VGPR 32 SRC0 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 VOP2_INST_LITERAL has_lit 1 VDST FMT_NUM_F32 OPR_VGPR 32 SIMM32 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 VOP2_VOP_DPP has_dpp 1 VDST FMT_NUM_F32 OPR_VGPR 32 VSRC0 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 VOP2_VOP_SDWA has_sdwa 1 VDST FMT_NUM_F32 OPR_VGPR 32 VSRC0 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 ENC_VOP3 default 257 VDST FMT_NUM_F32 OPR_VGPR 32 SRC0 FMT_NUM_F32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_F32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_SUB_F32 Subtract the second floating point input from the first input and store the result into a vector register. ENC_VOP2 default 2 VDST FMT_NUM_F32 OPR_VGPR 32 SRC0 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 VOP2_INST_LITERAL has_lit 2 VDST FMT_NUM_F32 OPR_VGPR 32 SIMM32 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 VOP2_VOP_DPP has_dpp 2 VDST FMT_NUM_F32 OPR_VGPR 32 VSRC0 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 VOP2_VOP_SDWA has_sdwa 2 VDST FMT_NUM_F32 OPR_VGPR 32 VSRC0 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 ENC_VOP3 default 258 VDST FMT_NUM_F32 OPR_VGPR 32 SRC0 FMT_NUM_F32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_F32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_SUBREV_F32 Subtract the first floating point input from the second input and store the result into a vector register. ENC_VOP2 default 3 VDST FMT_NUM_F32 OPR_VGPR 32 SRC0 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 VOP2_INST_LITERAL has_lit 3 VDST FMT_NUM_F32 OPR_VGPR 32 SIMM32 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 VOP2_VOP_DPP has_dpp 3 VDST FMT_NUM_F32 OPR_VGPR 32 VSRC0 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 VOP2_VOP_SDWA has_sdwa 3 VDST FMT_NUM_F32 OPR_VGPR 32 VSRC0 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 ENC_VOP3 default 259 VDST FMT_NUM_F32 OPR_VGPR 32 SRC0 FMT_NUM_F32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_F32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_FMAC_F64 Multiply two floating point inputs and accumulate the result into the destination register using fused multiply add. ENC_VOP2 default 4 VDST FMT_NUM_F64 OPR_VGPR 64 SRC0 FMT_NUM_F64 OPR_SRC 64 VSRC1 FMT_NUM_F64 OPR_VGPR 64 VOP2_INST_LITERAL has_lit 4 VDST FMT_NUM_F64 OPR_VGPR 64 SIMM32 FMT_NUM_F64 OPR_SRC 64 VSRC1 FMT_NUM_F64 OPR_VGPR 64 VOP2_VOP_DPP has_dpp 4 VDST FMT_NUM_F64 OPR_VGPR 64 VSRC0 FMT_NUM_F64 OPR_SRC 64 VSRC1 FMT_NUM_F64 OPR_VGPR 64 ENC_VOP3 default 260 VDST FMT_NUM_F64 OPR_VGPR 64 SRC0 FMT_NUM_F64 OPR_SRC_NOLIT 64 SRC1 FMT_NUM_F64 OPR_SRC_SIMPLE 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_MUL_F32 Multiply two floating point inputs and store the result into a vector register. ENC_VOP2 default 5 VDST FMT_NUM_F32 OPR_VGPR 32 SRC0 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 VOP2_INST_LITERAL has_lit 5 VDST FMT_NUM_F32 OPR_VGPR 32 SIMM32 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 VOP2_VOP_DPP has_dpp 5 VDST FMT_NUM_F32 OPR_VGPR 32 VSRC0 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 VOP2_VOP_SDWA has_sdwa 5 VDST FMT_NUM_F32 OPR_VGPR 32 VSRC0 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 ENC_VOP3 default 261 VDST FMT_NUM_F32 OPR_VGPR 32 SRC0 FMT_NUM_F32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_F32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_MUL_I32_I24 Multiply two signed 24-bit integer inputs and store the result as a signed 32-bit integer into a vector register. ENC_VOP2 default 6 VDST FMT_NUM_I32 OPR_VGPR 32 SRC0 FMT_NUM_I32 OPR_SRC 32 VSRC1 FMT_NUM_I32 OPR_VGPR 32 VOP2_INST_LITERAL has_lit 6 VDST FMT_NUM_I32 OPR_VGPR 32 SIMM32 FMT_NUM_I32 OPR_SRC 32 VSRC1 FMT_NUM_I32 OPR_VGPR 32 VOP2_VOP_DPP has_dpp 6 VDST FMT_NUM_I32 OPR_VGPR 32 VSRC0 FMT_NUM_I32 OPR_SRC 32 VSRC1 FMT_NUM_I32 OPR_VGPR 32 VOP2_VOP_SDWA has_sdwa 6 VDST FMT_NUM_I32 OPR_VGPR 32 VSRC0 FMT_NUM_I32 OPR_SRC 32 VSRC1 FMT_NUM_I32 OPR_VGPR 32 ENC_VOP3 default 262 VDST FMT_NUM_I32 OPR_VGPR 32 SRC0 FMT_NUM_I32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_I32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_MUL_HI_I32_I24 Multiply two signed 24-bit integer inputs and store the high 32 bits of the result as a signed 32-bit integer into a vector register. ENC_VOP2 default 7 VDST FMT_NUM_I32 OPR_VGPR 32 SRC0 FMT_NUM_I32 OPR_SRC 32 VSRC1 FMT_NUM_I32 OPR_VGPR 32 VOP2_INST_LITERAL has_lit 7 VDST FMT_NUM_I32 OPR_VGPR 32 SIMM32 FMT_NUM_I32 OPR_SRC 32 VSRC1 FMT_NUM_I32 OPR_VGPR 32 VOP2_VOP_DPP has_dpp 7 VDST FMT_NUM_I32 OPR_VGPR 32 VSRC0 FMT_NUM_I32 OPR_SRC 32 VSRC1 FMT_NUM_I32 OPR_VGPR 32 VOP2_VOP_SDWA has_sdwa 7 VDST FMT_NUM_I32 OPR_VGPR 32 VSRC0 FMT_NUM_I32 OPR_SRC 32 VSRC1 FMT_NUM_I32 OPR_VGPR 32 ENC_VOP3 default 263 VDST FMT_NUM_I32 OPR_VGPR 32 SRC0 FMT_NUM_I32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_I32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_MUL_U32_U24 Multiply two unsigned 24-bit integer inputs and store the result as an unsigned 32-bit integer into a vector register. ENC_VOP2 default 8 VDST FMT_NUM_U32 OPR_VGPR 32 SRC0 FMT_NUM_U32 OPR_SRC 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 VOP2_INST_LITERAL has_lit 8 VDST FMT_NUM_U32 OPR_VGPR 32 SIMM32 FMT_NUM_U32 OPR_SRC 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 VOP2_VOP_DPP has_dpp 8 VDST FMT_NUM_U32 OPR_VGPR 32 VSRC0 FMT_NUM_U32 OPR_SRC 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 VOP2_VOP_SDWA has_sdwa 8 VDST FMT_NUM_U32 OPR_VGPR 32 VSRC0 FMT_NUM_U32 OPR_SRC 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 ENC_VOP3 default 264 VDST FMT_NUM_U32 OPR_VGPR 32 SRC0 FMT_NUM_U32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_U32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_MUL_HI_U32_U24 Multiply two unsigned 24-bit integer inputs and store the high 32 bits of the result as an unsigned 32-bit integer into a vector register. ENC_VOP2 default 9 VDST FMT_NUM_U32 OPR_VGPR 32 SRC0 FMT_NUM_U32 OPR_SRC 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 VOP2_INST_LITERAL has_lit 9 VDST FMT_NUM_U32 OPR_VGPR 32 SIMM32 FMT_NUM_U32 OPR_SRC 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 VOP2_VOP_DPP has_dpp 9 VDST FMT_NUM_U32 OPR_VGPR 32 VSRC0 FMT_NUM_U32 OPR_SRC 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 VOP2_VOP_SDWA has_sdwa 9 VDST FMT_NUM_U32 OPR_VGPR 32 VSRC0 FMT_NUM_U32 OPR_SRC 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 ENC_VOP3 default 265 VDST FMT_NUM_U32 OPR_VGPR 32 SRC0 FMT_NUM_U32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_U32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_MIN_F32 Select the minimum of two single-precision float inputs and store the selected value into a vector register. ENC_VOP2 default 10 VDST FMT_NUM_F32 OPR_VGPR 32 SRC0 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 VOP2_INST_LITERAL has_lit 10 VDST FMT_NUM_F32 OPR_VGPR 32 SIMM32 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 VOP2_VOP_DPP has_dpp 10 VDST FMT_NUM_F32 OPR_VGPR 32 VSRC0 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 VOP2_VOP_SDWA has_sdwa 10 VDST FMT_NUM_F32 OPR_VGPR 32 VSRC0 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 ENC_VOP3 default 266 VDST FMT_NUM_F32 OPR_VGPR 32 SRC0 FMT_NUM_F32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_F32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_MAX_F32 Select the maximum of two single-precision float inputs and store the selected value into a vector register. ENC_VOP2 default 11 VDST FMT_NUM_F32 OPR_VGPR 32 SRC0 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 VOP2_INST_LITERAL has_lit 11 VDST FMT_NUM_F32 OPR_VGPR 32 SIMM32 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 VOP2_VOP_DPP has_dpp 11 VDST FMT_NUM_F32 OPR_VGPR 32 VSRC0 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 VOP2_VOP_SDWA has_sdwa 11 VDST FMT_NUM_F32 OPR_VGPR 32 VSRC0 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 ENC_VOP3 default 267 VDST FMT_NUM_F32 OPR_VGPR 32 SRC0 FMT_NUM_F32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_F32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_MIN_I32 Select the minimum of two signed 32-bit integer inputs and store the selected value into a vector register. ENC_VOP2 default 12 VDST FMT_NUM_I32 OPR_VGPR 32 SRC0 FMT_NUM_I32 OPR_SRC 32 VSRC1 FMT_NUM_I32 OPR_VGPR 32 VOP2_INST_LITERAL has_lit 12 VDST FMT_NUM_I32 OPR_VGPR 32 SIMM32 FMT_NUM_I32 OPR_SRC 32 VSRC1 FMT_NUM_I32 OPR_VGPR 32 VOP2_VOP_DPP has_dpp 12 VDST FMT_NUM_I32 OPR_VGPR 32 VSRC0 FMT_NUM_I32 OPR_SRC 32 VSRC1 FMT_NUM_I32 OPR_VGPR 32 VOP2_VOP_SDWA has_sdwa 12 VDST FMT_NUM_I32 OPR_VGPR 32 VSRC0 FMT_NUM_I32 OPR_SRC 32 VSRC1 FMT_NUM_I32 OPR_VGPR 32 ENC_VOP3 default 268 VDST FMT_NUM_I32 OPR_VGPR 32 SRC0 FMT_NUM_I32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_I32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_MAX_I32 Select the maximum of two signed 32-bit integer inputs and store the selected value into a vector register. ENC_VOP2 default 13 VDST FMT_NUM_I32 OPR_VGPR 32 SRC0 FMT_NUM_I32 OPR_SRC 32 VSRC1 FMT_NUM_I32 OPR_VGPR 32 VOP2_INST_LITERAL has_lit 13 VDST FMT_NUM_I32 OPR_VGPR 32 SIMM32 FMT_NUM_I32 OPR_SRC 32 VSRC1 FMT_NUM_I32 OPR_VGPR 32 VOP2_VOP_DPP has_dpp 13 VDST FMT_NUM_I32 OPR_VGPR 32 VSRC0 FMT_NUM_I32 OPR_SRC 32 VSRC1 FMT_NUM_I32 OPR_VGPR 32 VOP2_VOP_SDWA has_sdwa 13 VDST FMT_NUM_I32 OPR_VGPR 32 VSRC0 FMT_NUM_I32 OPR_SRC 32 VSRC1 FMT_NUM_I32 OPR_VGPR 32 ENC_VOP3 default 269 VDST FMT_NUM_I32 OPR_VGPR 32 SRC0 FMT_NUM_I32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_I32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_MIN_U32 Select the minimum of two unsigned 32-bit integer inputs and store the selected value into a vector register. ENC_VOP2 default 14 VDST FMT_NUM_U32 OPR_VGPR 32 SRC0 FMT_NUM_U32 OPR_SRC 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 VOP2_INST_LITERAL has_lit 14 VDST FMT_NUM_U32 OPR_VGPR 32 SIMM32 FMT_NUM_U32 OPR_SRC 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 VOP2_VOP_DPP has_dpp 14 VDST FMT_NUM_U32 OPR_VGPR 32 VSRC0 FMT_NUM_U32 OPR_SRC 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 VOP2_VOP_SDWA has_sdwa 14 VDST FMT_NUM_U32 OPR_VGPR 32 VSRC0 FMT_NUM_U32 OPR_SRC 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 ENC_VOP3 default 270 VDST FMT_NUM_U32 OPR_VGPR 32 SRC0 FMT_NUM_U32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_U32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_MAX_U32 Select the maximum of two unsigned 32-bit integer inputs and store the selected value into a vector register. ENC_VOP2 default 15 VDST FMT_NUM_U32 OPR_VGPR 32 SRC0 FMT_NUM_U32 OPR_SRC 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 VOP2_INST_LITERAL has_lit 15 VDST FMT_NUM_U32 OPR_VGPR 32 SIMM32 FMT_NUM_U32 OPR_SRC 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 VOP2_VOP_DPP has_dpp 15 VDST FMT_NUM_U32 OPR_VGPR 32 VSRC0 FMT_NUM_U32 OPR_SRC 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 VOP2_VOP_SDWA has_sdwa 15 VDST FMT_NUM_U32 OPR_VGPR 32 VSRC0 FMT_NUM_U32 OPR_SRC 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 ENC_VOP3 default 271 VDST FMT_NUM_U32 OPR_VGPR 32 SRC0 FMT_NUM_U32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_U32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_LSHRREV_B32 Given a shift count in the first vector input, calculate the logical shift right of the second vector input and store the result into a vector register. ENC_VOP2 default 16 VDST FMT_NUM_U32 OPR_VGPR 32 SRC0 FMT_NUM_U32 OPR_SRC_NOLDS 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 VOP2_INST_LITERAL has_lit 16 VDST FMT_NUM_U32 OPR_VGPR 32 SIMM32 FMT_NUM_U32 OPR_SRC_NOLDS 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 VOP2_VOP_DPP has_dpp 16 VDST FMT_NUM_U32 OPR_VGPR 32 VSRC0 FMT_NUM_U32 OPR_SRC_NOLDS 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 VOP2_VOP_SDWA has_sdwa 16 VDST FMT_NUM_U32 OPR_VGPR 32 VSRC0 FMT_NUM_U32 OPR_SRC_NOLDS 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 ENC_VOP3 default 272 VDST FMT_NUM_U32 OPR_VGPR 32 SRC0 FMT_NUM_U32 OPR_SRC_SIMPLE 32 SRC1 FMT_NUM_U32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_ASHRREV_I32 Given a shift count in the first vector input, calculate the arithmetic shift right (preserving sign bit) of the second vector input and store the result into a vector register. ENC_VOP2 default 17 VDST FMT_NUM_I32 OPR_VGPR 32 SRC0 FMT_NUM_U32 OPR_SRC_NOLDS 32 VSRC1 FMT_NUM_I32 OPR_VGPR 32 VOP2_INST_LITERAL has_lit 17 VDST FMT_NUM_I32 OPR_VGPR 32 SIMM32 FMT_NUM_U32 OPR_SRC_NOLDS 32 VSRC1 FMT_NUM_I32 OPR_VGPR 32 VOP2_VOP_DPP has_dpp 17 VDST FMT_NUM_I32 OPR_VGPR 32 VSRC0 FMT_NUM_U32 OPR_SRC_NOLDS 32 VSRC1 FMT_NUM_I32 OPR_VGPR 32 VOP2_VOP_SDWA has_sdwa 17 VDST FMT_NUM_I32 OPR_VGPR 32 VSRC0 FMT_NUM_U32 OPR_SRC_NOLDS 32 VSRC1 FMT_NUM_I32 OPR_VGPR 32 ENC_VOP3 default 273 VDST FMT_NUM_I32 OPR_VGPR 32 SRC0 FMT_NUM_U32 OPR_SRC_SIMPLE 32 SRC1 FMT_NUM_I32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_LSHLREV_B32 Given a shift count in the first vector input, calculate the logical shift left of the second vector input and store the result into a vector register. ENC_VOP2 default 18 VDST FMT_NUM_U32 OPR_VGPR 32 SRC0 FMT_NUM_U32 OPR_SRC_NOLDS 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 VOP2_INST_LITERAL has_lit 18 VDST FMT_NUM_U32 OPR_VGPR 32 SIMM32 FMT_NUM_U32 OPR_SRC_NOLDS 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 VOP2_VOP_DPP has_dpp 18 VDST FMT_NUM_U32 OPR_VGPR 32 VSRC0 FMT_NUM_U32 OPR_SRC_NOLDS 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 VOP2_VOP_SDWA has_sdwa 18 VDST FMT_NUM_U32 OPR_VGPR 32 VSRC0 FMT_NUM_U32 OPR_SRC_NOLDS 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 ENC_VOP3 default 274 VDST FMT_NUM_U32 OPR_VGPR 32 SRC0 FMT_NUM_U32 OPR_SRC_SIMPLE 32 SRC1 FMT_NUM_U32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_AND_B32 Calculate bitwise AND on two vector inputs and store the result into a vector register. ENC_VOP2 default 19 VDST FMT_NUM_U32 OPR_VGPR 32 SRC0 FMT_NUM_U32 OPR_SRC 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 VOP2_INST_LITERAL has_lit 19 VDST FMT_NUM_U32 OPR_VGPR 32 SIMM32 FMT_NUM_U32 OPR_SRC 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 VOP2_VOP_DPP has_dpp 19 VDST FMT_NUM_U32 OPR_VGPR 32 VSRC0 FMT_NUM_U32 OPR_SRC 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 VOP2_VOP_SDWA has_sdwa 19 VDST FMT_NUM_U32 OPR_VGPR 32 VSRC0 FMT_NUM_U32 OPR_SRC 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 ENC_VOP3 default 275 VDST FMT_NUM_U32 OPR_VGPR 32 SRC0 FMT_NUM_U32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_U32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_OR_B32 Calculate bitwise OR on two vector inputs and store the result into a vector register. ENC_VOP2 default 20 VDST FMT_NUM_U32 OPR_VGPR 32 SRC0 FMT_NUM_U32 OPR_SRC 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 VOP2_INST_LITERAL has_lit 20 VDST FMT_NUM_U32 OPR_VGPR 32 SIMM32 FMT_NUM_U32 OPR_SRC 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 VOP2_VOP_DPP has_dpp 20 VDST FMT_NUM_U32 OPR_VGPR 32 VSRC0 FMT_NUM_U32 OPR_SRC 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 VOP2_VOP_SDWA has_sdwa 20 VDST FMT_NUM_U32 OPR_VGPR 32 VSRC0 FMT_NUM_U32 OPR_SRC 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 ENC_VOP3 default 276 VDST FMT_NUM_U32 OPR_VGPR 32 SRC0 FMT_NUM_U32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_U32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_XOR_B32 Calculate bitwise XOR on two vector inputs and store the result into a vector register. ENC_VOP2 default 21 VDST FMT_NUM_U32 OPR_VGPR 32 SRC0 FMT_NUM_U32 OPR_SRC 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 VOP2_INST_LITERAL has_lit 21 VDST FMT_NUM_U32 OPR_VGPR 32 SIMM32 FMT_NUM_U32 OPR_SRC 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 VOP2_VOP_DPP has_dpp 21 VDST FMT_NUM_U32 OPR_VGPR 32 VSRC0 FMT_NUM_U32 OPR_SRC 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 VOP2_VOP_SDWA has_sdwa 21 VDST FMT_NUM_U32 OPR_VGPR 32 VSRC0 FMT_NUM_U32 OPR_SRC 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 ENC_VOP3 default 277 VDST FMT_NUM_U32 OPR_VGPR 32 SRC0 FMT_NUM_U32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_U32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_DOT2C_F32_BF16 Compute the dot product of two packed 2-D BF16 float inputs in the single-precision float domain and accumulate with the single-precision float value in the destination register. ENC_VOP2 default 22 VDST FMT_NUM_F32 OPR_VGPR 32 SRC0 FMT_NUM_PK2_BF16 OPR_SRC 32 VSRC1 FMT_NUM_PK2_BF16 OPR_VGPR 32 VOP2_INST_LITERAL has_lit 22 VDST FMT_NUM_F32 OPR_VGPR 32 SIMM32 FMT_NUM_PK2_BF16 OPR_SRC 32 VSRC1 FMT_NUM_PK2_BF16 OPR_VGPR 32 VOP2_VOP_DPP has_dpp 22 VDST FMT_NUM_F32 OPR_VGPR 32 VSRC0 FMT_NUM_PK2_BF16 OPR_SRC 32 VSRC1 FMT_NUM_PK2_BF16 OPR_VGPR 32 ENC_VOP3 default 278 VDST FMT_NUM_F32 OPR_VGPR 32 SRC0 FMT_NUM_PK2_BF16 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_PK2_BF16 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_FMAMK_F32 Multiply a single-precision float input with a literal constant and add a second single-precision float input using fused multiply add, and store the result into a vector register. VOP2_INST_LITERAL default 23 VDST FMT_NUM_F32 OPR_VGPR 32 SRC0 FMT_NUM_F32 OPR_SRC 32 FMT_NUM_F32 OPR_SIMM32 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 VOP2_INST_LITERAL has_lit 23 VDST FMT_NUM_F32 OPR_VGPR 32 SIMM32 FMT_NUM_F32 OPR_SRC 32 FMT_NUM_F32 OPR_SIMM32 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_FMAAK_F32 Multiply two single-precision float inputs and add a literal constant using fused multiply add, and store the result into a vector register. VOP2_INST_LITERAL default 24 VDST FMT_NUM_F32 OPR_VGPR 32 SRC0 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 FMT_NUM_F32 OPR_SIMM32 32 VOP2_INST_LITERAL has_lit 24 VDST FMT_NUM_F32 OPR_VGPR 32 SIMM32 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 FMT_NUM_F32 OPR_SIMM32 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_ADD_CO_U32 Add two unsigned 32-bit integer inputs, store the result into a vector register and store the carry-out mask into a scalar register. ENC_VOP2 default 25 VDST FMT_NUM_U32 OPR_VGPR 32 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_U32 OPR_SRC 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 VOP2_INST_LITERAL has_lit 25 VDST FMT_NUM_U32 OPR_VGPR 32 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_U32 OPR_SRC 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 VOP2_VOP_DPP has_dpp 25 VDST FMT_NUM_U32 OPR_VGPR 32 FMT_NUM_M64 OPR_VCC 64 VSRC0 FMT_NUM_U32 OPR_SRC 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 VOP2_VOP_SDWA_SDST_ENC has_sdwa 25 VDST FMT_NUM_U32 OPR_VGPR 32 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_U32 OPR_SRC 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 VOP3_SDST_ENC default 281 VDST FMT_NUM_U32 OPR_VGPR 32 SDST FMT_NUM_M64 OPR_SREG 64 SRC0 FMT_NUM_U32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_U32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_SUB_CO_U32 Subtract the second unsigned 32-bit integer input from the first input, store the result into a vector register and store the carry-out mask into a scalar register. ENC_VOP2 default 26 VDST FMT_NUM_U32 OPR_VGPR 32 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_U32 OPR_SRC 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 VOP2_INST_LITERAL has_lit 26 VDST FMT_NUM_U32 OPR_VGPR 32 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_U32 OPR_SRC 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 VOP2_VOP_DPP has_dpp 26 VDST FMT_NUM_U32 OPR_VGPR 32 FMT_NUM_M64 OPR_VCC 64 VSRC0 FMT_NUM_U32 OPR_SRC 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 VOP2_VOP_SDWA_SDST_ENC has_sdwa 26 VDST FMT_NUM_U32 OPR_VGPR 32 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_U32 OPR_SRC 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 VOP3_SDST_ENC default 282 VDST FMT_NUM_U32 OPR_VGPR 32 SDST FMT_NUM_M64 OPR_SREG 64 SRC0 FMT_NUM_U32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_U32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_SUBREV_CO_U32 Subtract the first unsigned 32-bit integer input from the second input, store the result into a vector register and store the carry-out mask into a scalar register. ENC_VOP2 default 27 VDST FMT_NUM_U32 OPR_VGPR 32 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_U32 OPR_SRC_NOLDS 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 VOP2_INST_LITERAL has_lit 27 VDST FMT_NUM_U32 OPR_VGPR 32 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_U32 OPR_SRC_NOLDS 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 VOP2_VOP_DPP has_dpp 27 VDST FMT_NUM_U32 OPR_VGPR 32 FMT_NUM_M64 OPR_VCC 64 VSRC0 FMT_NUM_U32 OPR_SRC_NOLDS 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 VOP2_VOP_SDWA_SDST_ENC has_sdwa 27 VDST FMT_NUM_U32 OPR_VGPR 32 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_U32 OPR_SRC_NOLDS 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 VOP3_SDST_ENC default 283 VDST FMT_NUM_U32 OPR_VGPR 32 SDST FMT_NUM_M64 OPR_SREG 64 SRC0 FMT_NUM_U32 OPR_SRC_SIMPLE 32 SRC1 FMT_NUM_U32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_ADDC_CO_U32 Add two unsigned 32-bit integer inputs and a bit from a carry-in mask, store the result into a vector register and store the carry-out mask into a scalar register. ENC_VOP2 default 28 VDST FMT_NUM_U32 OPR_VGPR 32 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_U32 OPR_SRC 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 FMT_NUM_M64 OPR_VCC 64 VOP2_INST_LITERAL has_lit 28 VDST FMT_NUM_U32 OPR_VGPR 32 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_U32 OPR_SRC 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 FMT_NUM_M64 OPR_VCC 64 VOP2_VOP_DPP has_dpp 28 VDST FMT_NUM_U32 OPR_VGPR 32 FMT_NUM_M64 OPR_VCC 64 VSRC0 FMT_NUM_U32 OPR_SRC 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 FMT_NUM_M64 OPR_VCC 64 VOP2_VOP_SDWA_SDST_ENC has_sdwa 28 VDST FMT_NUM_U32 OPR_VGPR 32 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_U32 OPR_SRC 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 FMT_NUM_M64 OPR_VCC 64 VOP3_SDST_ENC default 284 VDST FMT_NUM_U32 OPR_VGPR 32 SDST FMT_NUM_M64 OPR_SREG 64 SRC0 FMT_NUM_U32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_U32 OPR_SRC_SIMPLE 32 SRC2 FMT_NUM_M64 OPR_SREG 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_SUBB_CO_U32 Subtract the second unsigned 32-bit integer input from the first input, subtract a bit from the carry-in mask, store the result into a vector register and store the carry-out mask into a scalar register. ENC_VOP2 default 29 VDST FMT_NUM_U32 OPR_VGPR 32 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_U32 OPR_SRC 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 FMT_NUM_M64 OPR_VCC 64 VOP2_INST_LITERAL has_lit 29 VDST FMT_NUM_U32 OPR_VGPR 32 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_U32 OPR_SRC 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 FMT_NUM_M64 OPR_VCC 64 VOP2_VOP_DPP has_dpp 29 VDST FMT_NUM_U32 OPR_VGPR 32 FMT_NUM_M64 OPR_VCC 64 VSRC0 FMT_NUM_U32 OPR_SRC 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 FMT_NUM_M64 OPR_VCC 64 VOP2_VOP_SDWA_SDST_ENC has_sdwa 29 VDST FMT_NUM_U32 OPR_VGPR 32 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_U32 OPR_SRC 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 FMT_NUM_M64 OPR_VCC 64 VOP3_SDST_ENC default 285 VDST FMT_NUM_U32 OPR_VGPR 32 SDST FMT_NUM_M64 OPR_SREG 64 SRC0 FMT_NUM_U32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_U32 OPR_SRC_SIMPLE 32 SRC2 FMT_NUM_M64 OPR_SREG 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_SUBBREV_CO_U32 Subtract the first unsigned 32-bit integer input from the second input, subtract a bit from the carry-in mask, store the result into a vector register and store the carry-out mask into a scalar register. ENC_VOP2 default 30 VDST FMT_NUM_U32 OPR_VGPR 32 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_U32 OPR_SRC_NOLDS 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 FMT_NUM_M64 OPR_VCC 64 VOP2_INST_LITERAL has_lit 30 VDST FMT_NUM_U32 OPR_VGPR 32 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_U32 OPR_SRC_NOLDS 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 FMT_NUM_M64 OPR_VCC 64 VOP2_VOP_DPP has_dpp 30 VDST FMT_NUM_U32 OPR_VGPR 32 FMT_NUM_M64 OPR_VCC 64 VSRC0 FMT_NUM_U32 OPR_SRC_NOLDS 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 FMT_NUM_M64 OPR_VCC 64 VOP2_VOP_SDWA_SDST_ENC has_sdwa 30 VDST FMT_NUM_U32 OPR_VGPR 32 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_U32 OPR_SRC_NOLDS 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 FMT_NUM_M64 OPR_VCC 64 VOP3_SDST_ENC default 286 VDST FMT_NUM_U32 OPR_VGPR 32 SDST FMT_NUM_M64 OPR_SREG 64 SRC0 FMT_NUM_U32 OPR_SRC_SIMPLE 32 SRC1 FMT_NUM_U32 OPR_SRC_SIMPLE 32 SRC2 FMT_NUM_M64 OPR_SREG 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_ADD_F16 Add two floating point inputs and store the result into a vector register. ENC_VOP2 default 31 VDST FMT_NUM_F16 OPR_VGPR 16 SRC0 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 VOP2_INST_LITERAL has_lit 31 VDST FMT_NUM_F16 OPR_VGPR 16 SIMM32 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 VOP2_VOP_DPP has_dpp 31 VDST FMT_NUM_F16 OPR_VGPR 16 VSRC0 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 VOP2_VOP_SDWA has_sdwa 31 VDST FMT_NUM_F16 OPR_VGPR 16 VSRC0 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 ENC_VOP3 default 287 VDST FMT_NUM_F16 OPR_VGPR 16 SRC0 FMT_NUM_F16 OPR_SRC_NOLIT 16 SRC1 FMT_NUM_F16 OPR_SRC_SIMPLE 16 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_SUB_F16 Subtract the second floating point input from the first input and store the result into a vector register. ENC_VOP2 default 32 VDST FMT_NUM_F16 OPR_VGPR 16 SRC0 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 VOP2_INST_LITERAL has_lit 32 VDST FMT_NUM_F16 OPR_VGPR 16 SIMM32 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 VOP2_VOP_DPP has_dpp 32 VDST FMT_NUM_F16 OPR_VGPR 16 VSRC0 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 VOP2_VOP_SDWA has_sdwa 32 VDST FMT_NUM_F16 OPR_VGPR 16 VSRC0 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 ENC_VOP3 default 288 VDST FMT_NUM_F16 OPR_VGPR 16 SRC0 FMT_NUM_F16 OPR_SRC_NOLIT 16 SRC1 FMT_NUM_F16 OPR_SRC_SIMPLE 16 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_SUBREV_F16 Subtract the first floating point input from the second input and store the result into a vector register. ENC_VOP2 default 33 VDST FMT_NUM_F16 OPR_VGPR 16 SRC0 FMT_NUM_F16 OPR_SRC_NOLDS 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 VOP2_INST_LITERAL has_lit 33 VDST FMT_NUM_F16 OPR_VGPR 16 SIMM32 FMT_NUM_F16 OPR_SRC_NOLDS 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 VOP2_VOP_DPP has_dpp 33 VDST FMT_NUM_F16 OPR_VGPR 16 VSRC0 FMT_NUM_F16 OPR_SRC_NOLDS 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 VOP2_VOP_SDWA has_sdwa 33 VDST FMT_NUM_F16 OPR_VGPR 16 VSRC0 FMT_NUM_F16 OPR_SRC_NOLDS 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 ENC_VOP3 default 289 VDST FMT_NUM_F16 OPR_VGPR 16 SRC0 FMT_NUM_F16 OPR_SRC_SIMPLE 16 SRC1 FMT_NUM_F16 OPR_SRC_SIMPLE 16 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_MUL_F16 Multiply two floating point inputs and store the result into a vector register. ENC_VOP2 default 34 VDST FMT_NUM_F16 OPR_VGPR 16 SRC0 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 VOP2_INST_LITERAL has_lit 34 VDST FMT_NUM_F16 OPR_VGPR 16 SIMM32 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 VOP2_VOP_DPP has_dpp 34 VDST FMT_NUM_F16 OPR_VGPR 16 VSRC0 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 VOP2_VOP_SDWA has_sdwa 34 VDST FMT_NUM_F16 OPR_VGPR 16 VSRC0 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 ENC_VOP3 default 290 VDST FMT_NUM_F16 OPR_VGPR 16 SRC0 FMT_NUM_F16 OPR_SRC_NOLIT 16 SRC1 FMT_NUM_F16 OPR_SRC_SIMPLE 16 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_MAC_F16 Multiply two floating point inputs and accumulate the result into the destination register. Implements IEEE rules and non-standard rule for OPSEL. ENC_VOP2 default 35 VDST FMT_NUM_F16 OPR_VGPR 16 SRC0 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 VOP2_INST_LITERAL has_lit 35 VDST FMT_NUM_F16 OPR_VGPR 16 SIMM32 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 VOP2_VOP_DPP has_dpp 35 VDST FMT_NUM_F16 OPR_VGPR 16 VSRC0 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 ENC_VOP3 default 291 VDST FMT_NUM_F16 OPR_VGPR 16 SRC0 FMT_NUM_F16 OPR_SRC_NOLIT 16 SRC1 FMT_NUM_F16 OPR_SRC_SIMPLE 16 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_MADMK_F16 Multiply a floating point input with a literal constant and add a second floating point input, and store the result into a vector register. Implements IEEE rules. VOP2_INST_LITERAL default 36 VDST FMT_NUM_F16 OPR_VGPR 16 SRC0 FMT_NUM_F16 OPR_SRC 16 FMT_NUM_F16 OPR_SIMM32 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 VOP2_INST_LITERAL has_lit 36 VDST FMT_NUM_F16 OPR_VGPR 16 SIMM32 FMT_NUM_F16 OPR_SRC 16 FMT_NUM_F16 OPR_SIMM32 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_MADAK_F16 Multiply two floating point inputs and add a literal constant, and store the result into a vector register. Implements IEEE rules. VOP2_INST_LITERAL default 37 VDST FMT_NUM_F16 OPR_VGPR 16 SRC0 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 FMT_NUM_F16 OPR_SIMM32 16 VOP2_INST_LITERAL has_lit 37 VDST FMT_NUM_F16 OPR_VGPR 16 SIMM32 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 FMT_NUM_F16 OPR_SIMM32 16 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_ADD_U16 Add two unsigned 16-bit integer inputs and store the result into a vector register. No carry-in or carry-out support. ENC_VOP2 default 38 VDST FMT_NUM_U16 OPR_VGPR 16 SRC0 FMT_NUM_U16 OPR_SRC 16 VSRC1 FMT_NUM_U16 OPR_VGPR 16 VOP2_INST_LITERAL has_lit 38 VDST FMT_NUM_U16 OPR_VGPR 16 SIMM32 FMT_NUM_U16 OPR_SRC 16 VSRC1 FMT_NUM_U16 OPR_VGPR 16 VOP2_VOP_DPP has_dpp 38 VDST FMT_NUM_U16 OPR_VGPR 16 VSRC0 FMT_NUM_U16 OPR_SRC 16 VSRC1 FMT_NUM_U16 OPR_VGPR 16 VOP2_VOP_SDWA has_sdwa 38 VDST FMT_NUM_U16 OPR_VGPR 16 VSRC0 FMT_NUM_U16 OPR_SRC 16 VSRC1 FMT_NUM_U16 OPR_VGPR 16 ENC_VOP3 default 294 VDST FMT_NUM_U16 OPR_VGPR 16 SRC0 FMT_NUM_U16 OPR_SRC_NOLIT 16 SRC1 FMT_NUM_U16 OPR_SRC_SIMPLE 16 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_SUB_U16 Subtract the second unsigned 16-bit integer input from the first input and store the result into a vector register. No carry-in or carry-out support. ENC_VOP2 default 39 VDST FMT_NUM_U16 OPR_VGPR 16 SRC0 FMT_NUM_U16 OPR_SRC 16 VSRC1 FMT_NUM_U16 OPR_VGPR 16 VOP2_INST_LITERAL has_lit 39 VDST FMT_NUM_U16 OPR_VGPR 16 SIMM32 FMT_NUM_U16 OPR_SRC 16 VSRC1 FMT_NUM_U16 OPR_VGPR 16 VOP2_VOP_DPP has_dpp 39 VDST FMT_NUM_U16 OPR_VGPR 16 VSRC0 FMT_NUM_U16 OPR_SRC 16 VSRC1 FMT_NUM_U16 OPR_VGPR 16 VOP2_VOP_SDWA has_sdwa 39 VDST FMT_NUM_U16 OPR_VGPR 16 VSRC0 FMT_NUM_U16 OPR_SRC 16 VSRC1 FMT_NUM_U16 OPR_VGPR 16 ENC_VOP3 default 295 VDST FMT_NUM_U16 OPR_VGPR 16 SRC0 FMT_NUM_U16 OPR_SRC_NOLIT 16 SRC1 FMT_NUM_U16 OPR_SRC_SIMPLE 16 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_SUBREV_U16 Subtract the first unsigned 16-bit integer input from the second input and store the result into a vector register. No carry-in or carry-out support. ENC_VOP2 default 40 VDST FMT_NUM_U16 OPR_VGPR 16 SRC0 FMT_NUM_U16 OPR_SRC_NOLDS 16 VSRC1 FMT_NUM_U16 OPR_VGPR 16 VOP2_INST_LITERAL has_lit 40 VDST FMT_NUM_U16 OPR_VGPR 16 SIMM32 FMT_NUM_U16 OPR_SRC_NOLDS 16 VSRC1 FMT_NUM_U16 OPR_VGPR 16 VOP2_VOP_DPP has_dpp 40 VDST FMT_NUM_U16 OPR_VGPR 16 VSRC0 FMT_NUM_U16 OPR_SRC_NOLDS 16 VSRC1 FMT_NUM_U16 OPR_VGPR 16 VOP2_VOP_SDWA has_sdwa 40 VDST FMT_NUM_U16 OPR_VGPR 16 VSRC0 FMT_NUM_U16 OPR_SRC_NOLDS 16 VSRC1 FMT_NUM_U16 OPR_VGPR 16 ENC_VOP3 default 296 VDST FMT_NUM_U16 OPR_VGPR 16 SRC0 FMT_NUM_U16 OPR_SRC_SIMPLE 16 SRC1 FMT_NUM_U16 OPR_SRC_SIMPLE 16 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_MUL_LO_U16 Multiply two unsigned 16-bit integer inputs and store the low bits of the result into a vector register. ENC_VOP2 default 41 VDST FMT_NUM_U16 OPR_VGPR 16 SRC0 FMT_NUM_U16 OPR_SRC 16 VSRC1 FMT_NUM_U16 OPR_VGPR 16 VOP2_INST_LITERAL has_lit 41 VDST FMT_NUM_U16 OPR_VGPR 16 SIMM32 FMT_NUM_U16 OPR_SRC 16 VSRC1 FMT_NUM_U16 OPR_VGPR 16 VOP2_VOP_DPP has_dpp 41 VDST FMT_NUM_U16 OPR_VGPR 16 VSRC0 FMT_NUM_U16 OPR_SRC 16 VSRC1 FMT_NUM_U16 OPR_VGPR 16 VOP2_VOP_SDWA has_sdwa 41 VDST FMT_NUM_U16 OPR_VGPR 16 VSRC0 FMT_NUM_U16 OPR_SRC 16 VSRC1 FMT_NUM_U16 OPR_VGPR 16 ENC_VOP3 default 297 VDST FMT_NUM_U16 OPR_VGPR 16 SRC0 FMT_NUM_U16 OPR_SRC_NOLIT 16 SRC1 FMT_NUM_U16 OPR_SRC_SIMPLE 16 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_LSHLREV_B16 Given a shift count in the first vector input, calculate the logical shift left of the second vector input and store the result into a vector register. ENC_VOP2 default 42 VDST FMT_NUM_U16 OPR_VGPR 16 SRC0 FMT_NUM_U16 OPR_SRC_NOLDS 16 VSRC1 FMT_NUM_U16 OPR_VGPR 16 VOP2_INST_LITERAL has_lit 42 VDST FMT_NUM_U16 OPR_VGPR 16 SIMM32 FMT_NUM_U16 OPR_SRC_NOLDS 16 VSRC1 FMT_NUM_U16 OPR_VGPR 16 VOP2_VOP_DPP has_dpp 42 VDST FMT_NUM_U16 OPR_VGPR 16 VSRC0 FMT_NUM_U16 OPR_SRC_NOLDS 16 VSRC1 FMT_NUM_U16 OPR_VGPR 16 VOP2_VOP_SDWA has_sdwa 42 VDST FMT_NUM_U16 OPR_VGPR 16 VSRC0 FMT_NUM_U16 OPR_SRC_NOLDS 16 VSRC1 FMT_NUM_U16 OPR_VGPR 16 ENC_VOP3 default 298 VDST FMT_NUM_U16 OPR_VGPR 16 SRC0 FMT_NUM_U16 OPR_SRC_SIMPLE 16 SRC1 FMT_NUM_U16 OPR_SRC_SIMPLE 16 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_LSHRREV_B16 Given a shift count in the first vector input, calculate the logical shift right of the second vector input and store the result into a vector register. ENC_VOP2 default 43 VDST FMT_NUM_U16 OPR_VGPR 16 SRC0 FMT_NUM_U16 OPR_SRC_NOLDS 16 VSRC1 FMT_NUM_U16 OPR_VGPR 16 VOP2_INST_LITERAL has_lit 43 VDST FMT_NUM_U16 OPR_VGPR 16 SIMM32 FMT_NUM_U16 OPR_SRC_NOLDS 16 VSRC1 FMT_NUM_U16 OPR_VGPR 16 VOP2_VOP_DPP has_dpp 43 VDST FMT_NUM_U16 OPR_VGPR 16 VSRC0 FMT_NUM_U16 OPR_SRC_NOLDS 16 VSRC1 FMT_NUM_U16 OPR_VGPR 16 VOP2_VOP_SDWA has_sdwa 43 VDST FMT_NUM_U16 OPR_VGPR 16 VSRC0 FMT_NUM_U16 OPR_SRC_NOLDS 16 VSRC1 FMT_NUM_U16 OPR_VGPR 16 ENC_VOP3 default 299 VDST FMT_NUM_U16 OPR_VGPR 16 SRC0 FMT_NUM_U16 OPR_SRC_SIMPLE 16 SRC1 FMT_NUM_U16 OPR_SRC_SIMPLE 16 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_ASHRREV_I16 Given a shift count in the first vector input, calculate the arithmetic shift right (preserving sign bit) of the second vector input and store the result into a vector register. ENC_VOP2 default 44 VDST FMT_NUM_I16 OPR_VGPR 16 SRC0 FMT_NUM_U16 OPR_SRC_NOLDS 16 VSRC1 FMT_NUM_I16 OPR_VGPR 16 VOP2_INST_LITERAL has_lit 44 VDST FMT_NUM_I16 OPR_VGPR 16 SIMM32 FMT_NUM_U16 OPR_SRC_NOLDS 16 VSRC1 FMT_NUM_I16 OPR_VGPR 16 VOP2_VOP_DPP has_dpp 44 VDST FMT_NUM_I16 OPR_VGPR 16 VSRC0 FMT_NUM_U16 OPR_SRC_NOLDS 16 VSRC1 FMT_NUM_I16 OPR_VGPR 16 VOP2_VOP_SDWA has_sdwa 44 VDST FMT_NUM_I16 OPR_VGPR 16 VSRC0 FMT_NUM_U16 OPR_SRC_NOLDS 16 VSRC1 FMT_NUM_I16 OPR_VGPR 16 ENC_VOP3 default 300 VDST FMT_NUM_I16 OPR_VGPR 16 SRC0 FMT_NUM_U16 OPR_SRC_SIMPLE 16 SRC1 FMT_NUM_I16 OPR_SRC_SIMPLE 16 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_MAX_F16 Select the maximum of two half-precision float inputs and store the selected value into a vector register. ENC_VOP2 default 45 VDST FMT_NUM_F16 OPR_VGPR 16 SRC0 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 VOP2_INST_LITERAL has_lit 45 VDST FMT_NUM_F16 OPR_VGPR 16 SIMM32 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 VOP2_VOP_DPP has_dpp 45 VDST FMT_NUM_F16 OPR_VGPR 16 VSRC0 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 VOP2_VOP_SDWA has_sdwa 45 VDST FMT_NUM_F16 OPR_VGPR 16 VSRC0 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 ENC_VOP3 default 301 VDST FMT_NUM_F16 OPR_VGPR 16 SRC0 FMT_NUM_F16 OPR_SRC_NOLIT 16 SRC1 FMT_NUM_F16 OPR_SRC_SIMPLE 16 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_MIN_F16 Select the minimum of two half-precision float inputs and store the selected value into a vector register. ENC_VOP2 default 46 VDST FMT_NUM_F16 OPR_VGPR 16 SRC0 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 VOP2_INST_LITERAL has_lit 46 VDST FMT_NUM_F16 OPR_VGPR 16 SIMM32 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 VOP2_VOP_DPP has_dpp 46 VDST FMT_NUM_F16 OPR_VGPR 16 VSRC0 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 VOP2_VOP_SDWA has_sdwa 46 VDST FMT_NUM_F16 OPR_VGPR 16 VSRC0 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 ENC_VOP3 default 302 VDST FMT_NUM_F16 OPR_VGPR 16 SRC0 FMT_NUM_F16 OPR_SRC_NOLIT 16 SRC1 FMT_NUM_F16 OPR_SRC_SIMPLE 16 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_MAX_U16 Select the maximum of two unsigned 16-bit integer inputs and store the selected value into a vector register. ENC_VOP2 default 47 VDST FMT_NUM_U16 OPR_VGPR 16 SRC0 FMT_NUM_U16 OPR_SRC 16 VSRC1 FMT_NUM_U16 OPR_VGPR 16 VOP2_INST_LITERAL has_lit 47 VDST FMT_NUM_U16 OPR_VGPR 16 SIMM32 FMT_NUM_U16 OPR_SRC 16 VSRC1 FMT_NUM_U16 OPR_VGPR 16 VOP2_VOP_DPP has_dpp 47 VDST FMT_NUM_U16 OPR_VGPR 16 VSRC0 FMT_NUM_U16 OPR_SRC 16 VSRC1 FMT_NUM_U16 OPR_VGPR 16 VOP2_VOP_SDWA has_sdwa 47 VDST FMT_NUM_U16 OPR_VGPR 16 VSRC0 FMT_NUM_U16 OPR_SRC 16 VSRC1 FMT_NUM_U16 OPR_VGPR 16 ENC_VOP3 default 303 VDST FMT_NUM_U16 OPR_VGPR 16 SRC0 FMT_NUM_U16 OPR_SRC_NOLIT 16 SRC1 FMT_NUM_U16 OPR_SRC_SIMPLE 16 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_MAX_I16 Select the maximum of two signed 16-bit integer inputs and store the selected value into a vector register. ENC_VOP2 default 48 VDST FMT_NUM_I16 OPR_VGPR 16 SRC0 FMT_NUM_I16 OPR_SRC 16 VSRC1 FMT_NUM_I16 OPR_VGPR 16 VOP2_INST_LITERAL has_lit 48 VDST FMT_NUM_I16 OPR_VGPR 16 SIMM32 FMT_NUM_I16 OPR_SRC 16 VSRC1 FMT_NUM_I16 OPR_VGPR 16 VOP2_VOP_DPP has_dpp 48 VDST FMT_NUM_I16 OPR_VGPR 16 VSRC0 FMT_NUM_I16 OPR_SRC 16 VSRC1 FMT_NUM_I16 OPR_VGPR 16 VOP2_VOP_SDWA has_sdwa 48 VDST FMT_NUM_I16 OPR_VGPR 16 VSRC0 FMT_NUM_I16 OPR_SRC 16 VSRC1 FMT_NUM_I16 OPR_VGPR 16 ENC_VOP3 default 304 VDST FMT_NUM_I16 OPR_VGPR 16 SRC0 FMT_NUM_I16 OPR_SRC_NOLIT 16 SRC1 FMT_NUM_I16 OPR_SRC_SIMPLE 16 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_MIN_U16 Select the minimum of two unsigned 16-bit integer inputs and store the selected value into a vector register. ENC_VOP2 default 49 VDST FMT_NUM_U16 OPR_VGPR 16 SRC0 FMT_NUM_U16 OPR_SRC 16 VSRC1 FMT_NUM_U16 OPR_VGPR 16 VOP2_INST_LITERAL has_lit 49 VDST FMT_NUM_U16 OPR_VGPR 16 SIMM32 FMT_NUM_U16 OPR_SRC 16 VSRC1 FMT_NUM_U16 OPR_VGPR 16 VOP2_VOP_DPP has_dpp 49 VDST FMT_NUM_U16 OPR_VGPR 16 VSRC0 FMT_NUM_U16 OPR_SRC 16 VSRC1 FMT_NUM_U16 OPR_VGPR 16 VOP2_VOP_SDWA has_sdwa 49 VDST FMT_NUM_U16 OPR_VGPR 16 VSRC0 FMT_NUM_U16 OPR_SRC 16 VSRC1 FMT_NUM_U16 OPR_VGPR 16 ENC_VOP3 default 305 VDST FMT_NUM_U16 OPR_VGPR 16 SRC0 FMT_NUM_U16 OPR_SRC_NOLIT 16 SRC1 FMT_NUM_U16 OPR_SRC_SIMPLE 16 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_MIN_I16 Select the minimum of two signed 16-bit integer inputs and store the selected value into a vector register. ENC_VOP2 default 50 VDST FMT_NUM_I16 OPR_VGPR 16 SRC0 FMT_NUM_I16 OPR_SRC 16 VSRC1 FMT_NUM_I16 OPR_VGPR 16 VOP2_INST_LITERAL has_lit 50 VDST FMT_NUM_I16 OPR_VGPR 16 SIMM32 FMT_NUM_I16 OPR_SRC 16 VSRC1 FMT_NUM_I16 OPR_VGPR 16 VOP2_VOP_DPP has_dpp 50 VDST FMT_NUM_I16 OPR_VGPR 16 VSRC0 FMT_NUM_I16 OPR_SRC 16 VSRC1 FMT_NUM_I16 OPR_VGPR 16 VOP2_VOP_SDWA has_sdwa 50 VDST FMT_NUM_I16 OPR_VGPR 16 VSRC0 FMT_NUM_I16 OPR_SRC 16 VSRC1 FMT_NUM_I16 OPR_VGPR 16 ENC_VOP3 default 306 VDST FMT_NUM_I16 OPR_VGPR 16 SRC0 FMT_NUM_I16 OPR_SRC_NOLIT 16 SRC1 FMT_NUM_I16 OPR_SRC_SIMPLE 16 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_LDEXP_F16 Multiply the first input, a floating point value, by an integral power of 2 specified in the second input, a signed integer value, and store the floating point result into a vector register. ENC_VOP2 default 51 VDST FMT_NUM_F16 OPR_VGPR 16 SRC0 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 VOP2_INST_LITERAL has_lit 51 VDST FMT_NUM_F16 OPR_VGPR 16 SIMM32 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 VOP2_VOP_DPP has_dpp 51 VDST FMT_NUM_F16 OPR_VGPR 16 VSRC0 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 VOP2_VOP_SDWA has_sdwa 51 VDST FMT_NUM_F16 OPR_VGPR 16 VSRC0 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 ENC_VOP3 default 307 VDST FMT_NUM_F16 OPR_VGPR 16 SRC0 FMT_NUM_F16 OPR_SRC_NOLIT 16 SRC1 FMT_NUM_F16 OPR_SRC_SIMPLE 16 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_ADD_U32 Add two unsigned 32-bit integer inputs and store the result into a vector register. No carry-in or carry-out support. ENC_VOP2 default 52 VDST FMT_NUM_U32 OPR_VGPR 32 SRC0 FMT_NUM_U32 OPR_SRC 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 VOP2_INST_LITERAL has_lit 52 VDST FMT_NUM_U32 OPR_VGPR 32 SIMM32 FMT_NUM_U32 OPR_SRC 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 VOP2_VOP_DPP has_dpp 52 VDST FMT_NUM_U32 OPR_VGPR 32 VSRC0 FMT_NUM_U32 OPR_SRC 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 VOP2_VOP_SDWA has_sdwa 52 VDST FMT_NUM_U32 OPR_VGPR 32 VSRC0 FMT_NUM_U32 OPR_SRC 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 ENC_VOP3 default 308 VDST FMT_NUM_U32 OPR_VGPR 32 SRC0 FMT_NUM_U32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_U32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_SUB_U32 Subtract the second unsigned 32-bit integer input from the first input and store the result into a vector register. No carry-in or carry-out support. ENC_VOP2 default 53 VDST FMT_NUM_U32 OPR_VGPR 32 SRC0 FMT_NUM_U32 OPR_SRC 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 VOP2_INST_LITERAL has_lit 53 VDST FMT_NUM_U32 OPR_VGPR 32 SIMM32 FMT_NUM_U32 OPR_SRC 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 VOP2_VOP_DPP has_dpp 53 VDST FMT_NUM_U32 OPR_VGPR 32 VSRC0 FMT_NUM_U32 OPR_SRC 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 VOP2_VOP_SDWA has_sdwa 53 VDST FMT_NUM_U32 OPR_VGPR 32 VSRC0 FMT_NUM_U32 OPR_SRC 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 ENC_VOP3 default 309 VDST FMT_NUM_U32 OPR_VGPR 32 SRC0 FMT_NUM_U32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_U32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_SUBREV_U32 Subtract the first unsigned 32-bit integer input from the second input and store the result into a vector register. No carry-in or carry-out support. ENC_VOP2 default 54 VDST FMT_NUM_U32 OPR_VGPR 32 SRC0 FMT_NUM_U32 OPR_SRC_NOLDS 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 VOP2_INST_LITERAL has_lit 54 VDST FMT_NUM_U32 OPR_VGPR 32 SIMM32 FMT_NUM_U32 OPR_SRC_NOLDS 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 VOP2_VOP_DPP has_dpp 54 VDST FMT_NUM_U32 OPR_VGPR 32 VSRC0 FMT_NUM_U32 OPR_SRC_NOLDS 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 VOP2_VOP_SDWA has_sdwa 54 VDST FMT_NUM_U32 OPR_VGPR 32 VSRC0 FMT_NUM_U32 OPR_SRC_NOLDS 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 ENC_VOP3 default 310 VDST FMT_NUM_U32 OPR_VGPR 32 SRC0 FMT_NUM_U32 OPR_SRC_SIMPLE 32 SRC1 FMT_NUM_U32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_DOT2C_F32_F16 Compute the dot product of two packed 2-D half-precision float inputs in the single-precision float domain and accumulate with the single-precision float value in the destination register. ENC_VOP2 default 55 VDST FMT_NUM_F32 OPR_VGPR 32 SRC0 FMT_NUM_PK2_F16 OPR_SRC 32 VSRC1 FMT_NUM_PK2_F16 OPR_VGPR 32 VOP2_INST_LITERAL has_lit 55 VDST FMT_NUM_F32 OPR_VGPR 32 SIMM32 FMT_NUM_PK2_F16 OPR_SRC 32 VSRC1 FMT_NUM_PK2_F16 OPR_VGPR 32 VOP2_VOP_DPP has_dpp 55 VDST FMT_NUM_F32 OPR_VGPR 32 VSRC0 FMT_NUM_PK2_F16 OPR_SRC 32 VSRC1 FMT_NUM_PK2_F16 OPR_VGPR 32 ENC_VOP3 default 311 VDST FMT_NUM_F32 OPR_VGPR 32 SRC0 FMT_NUM_PK2_F16 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_PK2_F16 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_DOT2C_I32_I16 Compute the dot product of two packed 2-D signed 16-bit integer inputs in the signed 32-bit integer domain and accumulate with the signed 32-bit integer value in the destination register. ENC_VOP2 default 56 VDST FMT_NUM_I32 OPR_VGPR 32 SRC0 FMT_NUM_PK2_I16 OPR_SRC 32 VSRC1 FMT_NUM_PK2_I16 OPR_VGPR 32 VOP2_INST_LITERAL has_lit 56 VDST FMT_NUM_I32 OPR_VGPR 32 SIMM32 FMT_NUM_PK2_I16 OPR_SRC 32 VSRC1 FMT_NUM_PK2_I16 OPR_VGPR 32 VOP2_VOP_DPP has_dpp 56 VDST FMT_NUM_I32 OPR_VGPR 32 VSRC0 FMT_NUM_PK2_I16 OPR_SRC 32 VSRC1 FMT_NUM_PK2_I16 OPR_VGPR 32 ENC_VOP3 default 312 VDST FMT_NUM_I32 OPR_VGPR 32 SRC0 FMT_NUM_PK2_I16 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_PK2_I16 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_DOT4C_I32_I8 Compute the dot product of two packed 4-D signed 8-bit integer inputs in the signed 32-bit integer domain and accumulate with the signed 32-bit integer value in the destination register. ENC_VOP2 default 57 VDST FMT_NUM_I32 OPR_VGPR 32 SRC0 FMT_NUM_PK4_I8 OPR_SRC 32 VSRC1 FMT_NUM_PK4_I8 OPR_VGPR 32 VOP2_INST_LITERAL has_lit 57 VDST FMT_NUM_I32 OPR_VGPR 32 SIMM32 FMT_NUM_PK4_I8 OPR_SRC 32 VSRC1 FMT_NUM_PK4_I8 OPR_VGPR 32 VOP2_VOP_DPP has_dpp 57 VDST FMT_NUM_I32 OPR_VGPR 32 VSRC0 FMT_NUM_PK4_I8 OPR_SRC 32 VSRC1 FMT_NUM_PK4_I8 OPR_VGPR 32 ENC_VOP3 default 313 VDST FMT_NUM_I32 OPR_VGPR 32 SRC0 FMT_NUM_PK4_I8 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_PK4_I8 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_DOT8C_I32_I4 Compute the dot product of two packed 8-D signed 4-bit integer inputs in the signed 32-bit integer domain and accumulate with the signed 32-bit integer value in the destination register. ENC_VOP2 default 58 VDST FMT_NUM_I32 OPR_VGPR 32 SRC0 FMT_NUM_PK8_I4 OPR_SRC 32 VSRC1 FMT_NUM_PK8_I4 OPR_VGPR 32 VOP2_INST_LITERAL has_lit 58 VDST FMT_NUM_I32 OPR_VGPR 32 SIMM32 FMT_NUM_PK8_I4 OPR_SRC 32 VSRC1 FMT_NUM_PK8_I4 OPR_VGPR 32 VOP2_VOP_DPP has_dpp 58 VDST FMT_NUM_I32 OPR_VGPR 32 VSRC0 FMT_NUM_PK8_I4 OPR_SRC 32 VSRC1 FMT_NUM_PK8_I4 OPR_VGPR 32 ENC_VOP3 default 314 VDST FMT_NUM_I32 OPR_VGPR 32 SRC0 FMT_NUM_PK8_I4 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_PK8_I4 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_FMAC_F32 Multiply two floating point inputs and accumulate the result into the destination register using fused multiply add. ENC_VOP2 default 59 VDST FMT_NUM_F32 OPR_VGPR 32 SRC0 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 VOP2_INST_LITERAL has_lit 59 VDST FMT_NUM_F32 OPR_VGPR 32 SIMM32 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 VOP2_VOP_DPP has_dpp 59 VDST FMT_NUM_F32 OPR_VGPR 32 VSRC0 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 ENC_VOP3 default 315 VDST FMT_NUM_F32 OPR_VGPR 32 SRC0 FMT_NUM_F32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_F32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_PK_FMAC_F16 Multiply two packed half-precision float inputs component-wise and accumulate the result into the destination register using fused multiply add. ENC_VOP2 default 60 VDST FMT_NUM_PK2_F16 OPR_VGPR 32 SRC0 FMT_NUM_PK2_F16 OPR_SRC 32 VSRC1 FMT_NUM_PK2_F16 OPR_VGPR 32 VOP2_INST_LITERAL has_lit 60 VDST FMT_NUM_PK2_F16 OPR_VGPR 32 SIMM32 FMT_NUM_PK2_F16 OPR_SRC 32 VSRC1 FMT_NUM_PK2_F16 OPR_VGPR 32 VOP2_VOP_DPP has_dpp 60 VDST FMT_NUM_PK2_F16 OPR_VGPR 32 VSRC0 FMT_NUM_PK2_F16 OPR_SRC 32 VSRC1 FMT_NUM_PK2_F16 OPR_VGPR 32 ENC_VOP3 default 316 VDST FMT_NUM_PK2_F16 OPR_VGPR 32 SRC0 FMT_NUM_PK2_F16 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_PK2_F16 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_XNOR_B32 Calculate bitwise XNOR on two vector inputs and store the result into a vector register. ENC_VOP2 default 61 VDST FMT_NUM_U32 OPR_VGPR 32 SRC0 FMT_NUM_U32 OPR_SRC 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 VOP2_INST_LITERAL has_lit 61 VDST FMT_NUM_U32 OPR_VGPR 32 SIMM32 FMT_NUM_U32 OPR_SRC 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 VOP2_VOP_DPP has_dpp 61 VDST FMT_NUM_U32 OPR_VGPR 32 VSRC0 FMT_NUM_U32 OPR_SRC 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 VOP2_VOP_SDWA has_sdwa 61 VDST FMT_NUM_U32 OPR_VGPR 32 VSRC0 FMT_NUM_U32 OPR_SRC 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 ENC_VOP3 default 317 VDST FMT_NUM_U32 OPR_VGPR 32 SRC0 FMT_NUM_U32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_U32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_MAD_I32_I24 Multiply two signed 24-bit integer inputs in the signed 32-bit integer domain, add a signed 32-bit integer value from a third input, and store the result as a signed 32-bit integer into a vector register. ENC_VOP3 default 450 VDST FMT_NUM_I32 OPR_VGPR 32 SRC0 FMT_NUM_I24 OPR_SRC_NOLIT 24 SRC1 FMT_NUM_I24 OPR_SRC_SIMPLE 24 SRC2 FMT_NUM_I32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_MAD_U32_U24 Multiply two unsigned 24-bit integer inputs in the unsigned 32-bit integer domain, add a unsigned 32-bit integer value from a third input, and store the result as an unsigned 32-bit integer into a vector register. ENC_VOP3 default 451 VDST FMT_NUM_U32 OPR_VGPR 32 SRC0 FMT_NUM_U24 OPR_SRC_NOLIT 24 SRC1 FMT_NUM_U24 OPR_SRC_SIMPLE 24 SRC2 FMT_NUM_U32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CUBEID_F32 Compute the cubemap face ID of a 3D coordinate specified as three single-precision float inputs. Store the result in single-precision float format into a vector register. ENC_VOP3 default 452 VDST FMT_NUM_F32 OPR_VGPR 32 SRC0 FMT_NUM_F32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_F32 OPR_SRC_SIMPLE 32 SRC2 FMT_NUM_F32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CUBESC_F32 Compute the cubemap S coordinate of a 3D coordinate specified as three single-precision float inputs. Store the result in single-precision float format into a vector register. ENC_VOP3 default 453 VDST FMT_NUM_F32 OPR_VGPR 32 SRC0 FMT_NUM_F32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_F32 OPR_SRC_SIMPLE 32 SRC2 FMT_NUM_F32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CUBETC_F32 Compute the cubemap T coordinate of a 3D coordinate specified as three single-precision float inputs. Store the result in single-precision float format into a vector register. ENC_VOP3 default 454 VDST FMT_NUM_F32 OPR_VGPR 32 SRC0 FMT_NUM_F32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_F32 OPR_SRC_SIMPLE 32 SRC2 FMT_NUM_F32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CUBEMA_F32 Compute the cubemap major axis of a 3D coordinate specified as three single-precision float inputs. Store the result in single-precision float format into a vector register. ENC_VOP3 default 455 VDST FMT_NUM_F32 OPR_VGPR 32 SRC0 FMT_NUM_F32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_F32 OPR_SRC_SIMPLE 32 SRC2 FMT_NUM_F32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_BFE_U32 Extract an unsigned bitfield from the first input using field offset from the second input and size from the third input, then store the result into a vector register. ENC_VOP3 default 456 VDST FMT_NUM_U32 OPR_VGPR 32 SRC0 FMT_NUM_U32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_U32 OPR_SRC_SIMPLE 32 SRC2 FMT_NUM_U32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_BFE_I32 Extract a signed bitfield from the first input using field offset from the second input and size from the third input, then store the result into a vector register. ENC_VOP3 default 457 VDST FMT_NUM_I32 OPR_VGPR 32 SRC0 FMT_NUM_I32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_U32 OPR_SRC_SIMPLE 32 SRC2 FMT_NUM_U32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_BFI_B32 Overwrite a bitfield in the third input with a bitfield from the second input using a mask from the first input, then store the result into a vector register. ENC_VOP3 default 458 VDST FMT_NUM_U32 OPR_VGPR 32 SRC0 FMT_NUM_U32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_U32 OPR_SRC_SIMPLE 32 SRC2 FMT_NUM_U32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_FMA_F32 Multiply two single-precision float inputs and add a third input using fused multiply add, and store the result into a vector register. ENC_VOP3 default 459 VDST FMT_NUM_F32 OPR_VGPR 32 SRC0 FMT_NUM_F32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_F32 OPR_SRC_SIMPLE 32 SRC2 FMT_NUM_F32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_FMA_F64 Multiply two double-precision float inputs and add a third input using fused multiply add, and store the result into a vector register. ENC_VOP3 default 460 VDST FMT_NUM_F64 OPR_VGPR 64 SRC0 FMT_NUM_F64 OPR_SRC_NOLIT 64 SRC1 FMT_NUM_F64 OPR_SRC_SIMPLE 64 SRC2 FMT_NUM_F64 OPR_SRC_SIMPLE 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_LERP_U8 Average two 4-D vectors stored as packed bytes in the first two inputs with rounding control provided by the third input, then store the result into a vector register. Each byte in the third input acts as a rounding mode for the corresponding element; if the LSB is set then 0.5 rounds up, otherwise 0.5 truncates. ENC_VOP3 default 461 VDST FMT_NUM_U32 OPR_VGPR 32 SRC0 FMT_NUM_B32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_B32 OPR_SRC_SIMPLE 32 SRC2 FMT_NUM_B32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_ALIGNBIT_B32 Align a 64-bit value encoded in the first two inputs to a bit position specified in the third input, then store the result into a 32-bit vector register. ENC_VOP3 default 462 VDST FMT_NUM_U32 OPR_VGPR 32 SRC0 FMT_NUM_U32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_U32 OPR_SRC_SIMPLE 32 SRC2 FMT_NUM_U8 OPR_SRC_SIMPLE 8 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_ALIGNBYTE_B32 Align a 64-bit value encoded in the first two inputs to a byte position specified in the third input, then store the result into a 32-bit vector register. ENC_VOP3 default 463 VDST FMT_NUM_U32 OPR_VGPR 32 SRC0 FMT_NUM_U32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_U32 OPR_SRC_SIMPLE 32 SRC2 FMT_NUM_U8 OPR_SRC_SIMPLE 8 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_MIN3_F32 Select the minimum of three single-precision float inputs and store the selected value into a vector register. ENC_VOP3 default 464 VDST FMT_NUM_F32 OPR_VGPR 32 SRC0 FMT_NUM_F32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_F32 OPR_SRC_SIMPLE 32 SRC2 FMT_NUM_F32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_MIN3_I32 Select the minimum of three signed 32-bit integer inputs and store the selected value into a vector register. ENC_VOP3 default 465 VDST FMT_NUM_I32 OPR_VGPR 32 SRC0 FMT_NUM_I32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_I32 OPR_SRC_SIMPLE 32 SRC2 FMT_NUM_I32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_MIN3_U32 Select the minimum of three unsigned 32-bit integer inputs and store the selected value into a vector register. ENC_VOP3 default 466 VDST FMT_NUM_U32 OPR_VGPR 32 SRC0 FMT_NUM_U32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_U32 OPR_SRC_SIMPLE 32 SRC2 FMT_NUM_U32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_MAX3_F32 Select the maximum of three single-precision float inputs and store the selected value into a vector register. ENC_VOP3 default 467 VDST FMT_NUM_F32 OPR_VGPR 32 SRC0 FMT_NUM_F32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_F32 OPR_SRC_SIMPLE 32 SRC2 FMT_NUM_F32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_MAX3_I32 Select the maximum of three signed 32-bit integer inputs and store the selected value into a vector register. ENC_VOP3 default 468 VDST FMT_NUM_I32 OPR_VGPR 32 SRC0 FMT_NUM_I32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_I32 OPR_SRC_SIMPLE 32 SRC2 FMT_NUM_I32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_MAX3_U32 Select the maximum of three unsigned 32-bit integer inputs and store the selected value into a vector register. ENC_VOP3 default 469 VDST FMT_NUM_U32 OPR_VGPR 32 SRC0 FMT_NUM_U32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_U32 OPR_SRC_SIMPLE 32 SRC2 FMT_NUM_U32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_MED3_F32 Select the median of three single-precision float inputs and store the selected value into a vector register. ENC_VOP3 default 470 VDST FMT_NUM_F32 OPR_VGPR 32 SRC0 FMT_NUM_F32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_F32 OPR_SRC_SIMPLE 32 SRC2 FMT_NUM_F32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_MED3_I32 Select the median of three signed 32-bit integer inputs and store the selected value into a vector register. ENC_VOP3 default 471 VDST FMT_NUM_I32 OPR_VGPR 32 SRC0 FMT_NUM_I32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_I32 OPR_SRC_SIMPLE 32 SRC2 FMT_NUM_I32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_MED3_U32 Select the median of three unsigned 32-bit integer inputs and store the selected value into a vector register. ENC_VOP3 default 472 VDST FMT_NUM_U32 OPR_VGPR 32 SRC0 FMT_NUM_U32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_U32 OPR_SRC_SIMPLE 32 SRC2 FMT_NUM_U32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_SAD_U8 Calculate the sum of absolute differences of elements in two packed 4-component unsigned 8-bit integer inputs, add an unsigned 32-bit integer value from the third input and store the result into a vector register. ENC_VOP3 default 473 VDST FMT_NUM_U32 OPR_VGPR 32 SRC0 FMT_NUM_PK4_U8 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_PK4_U8 OPR_SRC_SIMPLE 32 SRC2 FMT_NUM_U32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_SAD_HI_U8 Calculate the sum of absolute differences of elements in two packed 4-component unsigned 8-bit integer inputs, shift the sum left by 16 bits, add an unsigned 32-bit integer value from the third input and store the result into a vector register. ENC_VOP3 default 474 VDST FMT_NUM_U32 OPR_VGPR 32 SRC0 FMT_NUM_PK4_U8 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_PK4_U8 OPR_SRC_SIMPLE 32 SRC2 FMT_NUM_U32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_SAD_U16 Calculate the sum of absolute differences of elements in two packed 2-component unsigned 16-bit integer inputs, add an unsigned 32-bit integer value from the third input and store the result into a vector register. ENC_VOP3 default 475 VDST FMT_NUM_U32 OPR_VGPR 32 SRC0 FMT_NUM_PK2_U16 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_PK2_U16 OPR_SRC_SIMPLE 32 SRC2 FMT_NUM_U32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_SAD_U32 Calculate the absolute difference of two unsigned 32-bit integer inputs, add an unsigned 32-bit integer value from the third input and store the result into a vector register. ENC_VOP3 default 476 VDST FMT_NUM_U32 OPR_VGPR 32 SRC0 FMT_NUM_U32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_U32 OPR_SRC_SIMPLE 32 SRC2 FMT_NUM_U32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CVT_PK_U8_F32 Convert a single-precision float value from the first input to an unsigned 8-bit integer value and pack the result into one byte of the third input using the second input as a byte select. Store the result into a vector register. ENC_VOP3 default 477 VDST FMT_NUM_PK4_U8 OPR_VGPR 32 SRC0 FMT_NUM_F32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_U32 OPR_SRC_SIMPLE 32 SRC2 FMT_NUM_PK4_U8 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_DIV_FIXUP_F32 Given a single-precision float quotient in the first input, a denominator in the second input and a numerator in the third input, detect and apply corner cases related to division, including divide by zero, NaN inputs and overflow, and modify the quotient accordingly. Generate any invalid, denormal and divide-by-zero exceptions that are a result of the division. Store the modified quotient into a vector register. ENC_VOP3 default 478 VDST FMT_NUM_F32 OPR_VGPR 32 SRC0 FMT_NUM_F32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_F32 OPR_SRC_SIMPLE 32 SRC2 FMT_NUM_F32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_DIV_FIXUP_F64 Given a double-precision float quotient in the first input, a denominator in the second input and a numerator in the third input, detect and apply corner cases related to division, including divide by zero, NaN inputs and overflow, and modify the quotient accordingly. Generate any invalid, denormal and divide-by-zero exceptions that are a result of the division. Store the modified quotient into a vector register. ENC_VOP3 default 479 VDST FMT_NUM_F64 OPR_VGPR 64 SRC0 FMT_NUM_F64 OPR_SRC_NOLIT 64 SRC1 FMT_NUM_F64 OPR_SRC_SIMPLE 64 SRC2 FMT_NUM_F64 OPR_SRC_SIMPLE 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_DIV_SCALE_F32 Given a single-precision float value to scale in the first input, a denominator in the second input and a numerator in the third input, scale the first input for division if required to avoid subnormal terms appearing during application of the Newton-Raphson correction method. Store the scaled result into a vector register and set the vector condition code iff post-scaling is required. VOP3_SDST_ENC default 480 VDST FMT_NUM_F32 OPR_VGPR 32 SDST FMT_NUM_M64 OPR_SDST 64 SRC0 FMT_NUM_F32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_F32 OPR_SRC_SIMPLE 32 SRC2 FMT_NUM_F32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_DIV_SCALE_F64 Given a double-precision float value to scale in the first input, a denominator in the second input and a numerator in the third input, scale the first input for division if required to avoid subnormal terms appearing during application of the Newton-Raphson correction method. Store the scaled result into a vector register and set the vector condition code iff post-scaling is required. VOP3_SDST_ENC default 481 VDST FMT_NUM_F64 OPR_VGPR 64 SDST FMT_NUM_M64 OPR_SDST 64 SRC0 FMT_NUM_F64 OPR_SRC_NOLIT 64 SRC1 FMT_NUM_F64 OPR_SRC_SIMPLE 64 SRC2 FMT_NUM_F64 OPR_SRC_SIMPLE 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_DIV_FMAS_F32 Multiply two single-precision float inputs and add a third input using fused multiply add, then scale the exponent of the result by a fixed factor if the vector condition code is set. Store the result into a vector register. ENC_VOP3 default 482 VDST FMT_NUM_F32 OPR_VGPR 32 SRC0 FMT_NUM_F32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_F32 OPR_SRC_SIMPLE 32 SRC2 FMT_NUM_F32 OPR_SRC_SIMPLE 32 FMT_NUM_M64 OPR_VCC 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_DIV_FMAS_F64 Multiply two double-precision float inputs and add a third input using fused multiply add, then scale the exponent of the result by a fixed factor if the vector condition code is set. Store the result into a vector register. ENC_VOP3 default 483 VDST FMT_NUM_F64 OPR_VGPR 64 SRC0 FMT_NUM_F64 OPR_SRC_NOLIT 64 SRC1 FMT_NUM_F64 OPR_SRC_SIMPLE 64 SRC2 FMT_NUM_F64 OPR_SRC_SIMPLE 64 FMT_NUM_M64 OPR_VCC 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_MSAD_U8 Calculate the sum of absolute differences of elements in two packed 4-component unsigned 8-bit integer inputs, except that elements where the second input (known as the reference input) is zero are not included in the sum. Add an unsigned 32-bit integer value from the third input and store the result into a vector register. ENC_VOP3 default 484 VDST FMT_NUM_U32 OPR_VGPR 32 SRC0 FMT_NUM_B32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_B32 OPR_SRC_SIMPLE 32 SRC2 FMT_NUM_B32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_QSAD_PK_U16_U8 Perform the V_SAD_U8 operation four times using different slices of the first array, all entries of the second array and each entry of the third array. Truncate each result to 16 bits, pack the values into a 4-entry array and store the array into a vector register. The first input is an 8-entry array of unsigned 8-bit integers, the second input is a 4-entry array of unsigned 8-bit integers and the third input is a 4-entry array of unsigned 16-bit integers. ENC_VOP3 default 485 VDST FMT_NUM_B64 OPR_VGPR 64 SRC0 FMT_NUM_B64 OPR_SRC_NOLIT 64 SRC1 FMT_NUM_B32 OPR_SRC_SIMPLE 32 SRC2 FMT_NUM_B64 OPR_SRC_SIMPLE 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_MQSAD_PK_U16_U8 Perform the V_MSAD_U8 operation four times using different slices of the first array, all entries of the second array and each entry of the third array. Truncate each result to 16 bits, pack the values into a 4-entry array and store the array into a vector register. The first input is an 8-entry array of unsigned 8-bit integers, the second input is a 4-entry array of unsigned 8-bit integers and the third input is a 4-entry array of unsigned 16-bit integers. ENC_VOP3 default 486 VDST FMT_NUM_B64 OPR_VGPR 64 SRC0 FMT_NUM_B64 OPR_SRC_NOLIT 64 SRC1 FMT_NUM_B32 OPR_SRC_SIMPLE 32 SRC2 FMT_NUM_B64 OPR_SRC_SIMPLE 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_MQSAD_U32_U8 Perform the V_MSAD_U8 operation four times using different slices of the first array, all entries of the second array and each entry of the third array. Pack each 32-bit value into a 4-entry array and store the array into a vector register. The first input is an 8-entry array of unsigned 8-bit integers, the second input is a 4-entry array of unsigned 8-bit integers and the third input is a 4-entry array of unsigned 32-bit integers. ENC_VOP3 default 487 VDST FMT_NUM_B128 OPR_VGPR 128 SRC0 FMT_NUM_B64 OPR_SRC_NOLIT 64 SRC1 FMT_NUM_B32 OPR_SRC_SIMPLE 32 SRC2 FMT_NUM_B128 OPR_SRC_VGPR 128 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_MAD_U64_U32 Multiply two unsigned integer inputs, add a third unsigned integer input, store the result into a 64-bit vector register and store the overflow/carryout into a scalar mask register. VOP3_SDST_ENC default 488 VDST FMT_NUM_U64 OPR_VGPR 64 SDST FMT_NUM_M64 OPR_SREG 64 SRC0 FMT_NUM_U32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_U32 OPR_SRC_SIMPLE 32 SRC2 FMT_NUM_U64 OPR_SRC_SIMPLE 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_MAD_I64_I32 Multiply two signed integer inputs, add a third signed integer input, store the result into a 64-bit vector register and store the overflow/carryout into a scalar mask register. VOP3_SDST_ENC default 489 VDST FMT_NUM_I64 OPR_VGPR 64 SDST FMT_NUM_M64 OPR_SREG 64 SRC0 FMT_NUM_I32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_I32 OPR_SRC_SIMPLE 32 SRC2 FMT_NUM_I64 OPR_SRC_SIMPLE 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_MAD_LEGACY_F16 Multiply add of FP16 values. Implements IEEE rules and non-standard rule for OPSEL. ENC_VOP3 default 490 VDST FMT_NUM_F16 OPR_VGPR 16 SRC0 FMT_NUM_F16 OPR_SRC_NOLIT 16 SRC1 FMT_NUM_F16 OPR_SRC_SIMPLE 16 SRC2 FMT_NUM_F16 OPR_SRC_SIMPLE 16 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_MAD_LEGACY_U16 Multiply add of unsigned short values. Has non-standard rule for OPSEL. ENC_VOP3 default 491 VDST FMT_NUM_B16 OPR_VGPR 16 SRC0 FMT_NUM_B16 OPR_SRC_NOLIT 16 SRC1 FMT_NUM_B16 OPR_SRC_SIMPLE 16 SRC2 FMT_NUM_B16 OPR_SRC_SIMPLE 16 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_MAD_LEGACY_I16 Multiply add of signed short values. Has non-standard rule for OPSEL. ENC_VOP3 default 492 VDST FMT_NUM_B16 OPR_VGPR 16 SRC0 FMT_NUM_B16 OPR_SRC_NOLIT 16 SRC1 FMT_NUM_B16 OPR_SRC_SIMPLE 16 SRC2 FMT_NUM_B16 OPR_SRC_SIMPLE 16 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_PERM_B32 Permute a 64-bit value constructed from two vector inputs (most significant bits come from the first input) using a per-lane selector from the third input. The lane selector allows each byte of the result to choose from any of the 8 input bytes, perform sign extension or pad with 0/1 bits. Store the result into a vector register. ENC_VOP3 default 493 VDST FMT_NUM_PK4_B8 OPR_VGPR 32 SRC0 FMT_NUM_B32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_B32 OPR_SRC_SIMPLE 32 SRC2 FMT_NUM_PK4_U8 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_FMA_LEGACY_F16 Fused half precision multiply add. Implements IEEE rules and non-standard rule for OPSEL. ENC_VOP3 default 494 VDST FMT_NUM_F16 OPR_VGPR 16 SRC0 FMT_NUM_F16 OPR_SRC_NOLIT 16 SRC1 FMT_NUM_F16 OPR_SRC_SIMPLE 16 SRC2 FMT_NUM_F16 OPR_SRC_SIMPLE 16 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_DIV_FIXUP_LEGACY_F16 Half precision division fixup. Has non-standard rule for OPSEL. ENC_VOP3 default 495 VDST FMT_NUM_F16 OPR_VGPR 16 SRC0 FMT_NUM_F16 OPR_SRC_NOLIT 16 SRC1 FMT_NUM_F16 OPR_SRC_SIMPLE 16 SRC2 FMT_NUM_F16 OPR_SRC_SIMPLE 16 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CVT_PKACCUM_U8_F32 Convert a single-precision float value in the first input to an unsigned 8-bit integer value and store the result into one byte of the destination register using the second input as a byte select. ENC_VOP3 default 496 VDST FMT_NUM_PK4_U8 OPR_VGPR 32 SRC0 FMT_NUM_F32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_U32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_MAD_U32_U16 Multiply two unsigned 16-bit integer inputs in the unsigned 32-bit integer domain, add an unsigned 32-bit integer value from a third input, and store the result as an unsigned 32-bit integer into a vector register. ENC_VOP3 default 497 VDST FMT_NUM_U32 OPR_VGPR 32 SRC0 FMT_NUM_U16 OPR_SRC_NOLIT 16 SRC1 FMT_NUM_U16 OPR_SRC_SIMPLE 16 SRC2 FMT_NUM_U32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_MAD_I32_I16 Multiply two signed 16-bit integer inputs in the signed 32-bit integer domain, add a signed 32-bit integer value from a third input, and store the result as a signed 32-bit integer into a vector register. ENC_VOP3 default 498 VDST FMT_NUM_I32 OPR_VGPR 32 SRC0 FMT_NUM_I16 OPR_SRC_NOLIT 16 SRC1 FMT_NUM_I16 OPR_SRC_SIMPLE 16 SRC2 FMT_NUM_I32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_XAD_U32 Calculate bitwise XOR of the first two vector inputs, then add the third vector input to the intermediate result, then store the final result into a vector register. ENC_VOP3 default 499 VDST FMT_NUM_U32 OPR_VGPR 32 SRC0 FMT_NUM_U32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_U32 OPR_SRC_SIMPLE 32 SRC2 FMT_NUM_U32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_MIN3_F16 Select the minimum of three half-precision float inputs and store the selected value into a vector register. ENC_VOP3 default 500 VDST FMT_NUM_F16 OPR_VGPR 16 SRC0 FMT_NUM_F16 OPR_SRC_NOLIT 16 SRC1 FMT_NUM_F16 OPR_SRC_SIMPLE 16 SRC2 FMT_NUM_F16 OPR_SRC_SIMPLE 16 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_MIN3_I16 Select the minimum of three signed 16-bit integer inputs and store the selected value into a vector register. ENC_VOP3 default 501 VDST FMT_NUM_I16 OPR_VGPR 16 SRC0 FMT_NUM_I16 OPR_SRC_NOLIT 16 SRC1 FMT_NUM_I16 OPR_SRC_SIMPLE 16 SRC2 FMT_NUM_I16 OPR_SRC_SIMPLE 16 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_MIN3_U16 Select the minimum of three unsigned 16-bit integer inputs and store the selected value into a vector register. ENC_VOP3 default 502 VDST FMT_NUM_U16 OPR_VGPR 16 SRC0 FMT_NUM_U16 OPR_SRC_NOLIT 16 SRC1 FMT_NUM_U16 OPR_SRC_SIMPLE 16 SRC2 FMT_NUM_U16 OPR_SRC_SIMPLE 16 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_MAX3_F16 Select the maximum of three half-precision float inputs and store the selected value into a vector register. ENC_VOP3 default 503 VDST FMT_NUM_F16 OPR_VGPR 16 SRC0 FMT_NUM_F16 OPR_SRC_NOLIT 16 SRC1 FMT_NUM_F16 OPR_SRC_SIMPLE 16 SRC2 FMT_NUM_F16 OPR_SRC_SIMPLE 16 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_MAX3_I16 Select the maximum of three signed 16-bit integer inputs and store the selected value into a vector register. ENC_VOP3 default 504 VDST FMT_NUM_I16 OPR_VGPR 16 SRC0 FMT_NUM_I16 OPR_SRC_NOLIT 16 SRC1 FMT_NUM_I16 OPR_SRC_SIMPLE 16 SRC2 FMT_NUM_I16 OPR_SRC_SIMPLE 16 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_MAX3_U16 Select the maximum of three unsigned 16-bit integer inputs and store the selected value into a vector register. ENC_VOP3 default 505 VDST FMT_NUM_U16 OPR_VGPR 16 SRC0 FMT_NUM_U16 OPR_SRC_NOLIT 16 SRC1 FMT_NUM_U16 OPR_SRC_SIMPLE 16 SRC2 FMT_NUM_U16 OPR_SRC_SIMPLE 16 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_MED3_F16 Select the median of three half-precision float inputs and store the selected value into a vector register. ENC_VOP3 default 506 VDST FMT_NUM_F16 OPR_VGPR 16 SRC0 FMT_NUM_F16 OPR_SRC_NOLIT 16 SRC1 FMT_NUM_F16 OPR_SRC_SIMPLE 16 SRC2 FMT_NUM_F16 OPR_SRC_SIMPLE 16 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_MED3_I16 Select the median of three signed 16-bit integer inputs and store the selected value into a vector register. ENC_VOP3 default 507 VDST FMT_NUM_I16 OPR_VGPR 16 SRC0 FMT_NUM_I16 OPR_SRC_NOLIT 16 SRC1 FMT_NUM_I16 OPR_SRC_SIMPLE 16 SRC2 FMT_NUM_I16 OPR_SRC_SIMPLE 16 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_MED3_U16 Select the median of three unsigned 16-bit integer inputs and store the selected value into a vector register. ENC_VOP3 default 508 VDST FMT_NUM_U16 OPR_VGPR 16 SRC0 FMT_NUM_U16 OPR_SRC_NOLIT 16 SRC1 FMT_NUM_U16 OPR_SRC_SIMPLE 16 SRC2 FMT_NUM_U16 OPR_SRC_SIMPLE 16 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_LSHL_ADD_U32 Given a shift count in the second input, calculate the logical shift left of the first input, then add the third input to the intermediate result, then store the final result into a vector register. ENC_VOP3 default 509 VDST FMT_NUM_U32 OPR_VGPR 32 SRC0 FMT_NUM_U32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_U32 OPR_SRC_SIMPLE 32 SRC2 FMT_NUM_U32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_ADD_LSHL_U32 Add the first two integer inputs, then given a shift count in the third input, calculate the logical shift left of the intermediate result, then store the final result into a vector register. ENC_VOP3 default 510 VDST FMT_NUM_U32 OPR_VGPR 32 SRC0 FMT_NUM_U32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_U32 OPR_SRC_SIMPLE 32 SRC2 FMT_NUM_U32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_ADD3_U32 Add three unsigned inputs and store the result into a vector register. No carry-in or carry-out support. ENC_VOP3 default 511 VDST FMT_NUM_U32 OPR_VGPR 32 SRC0 FMT_NUM_U32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_U32 OPR_SRC_SIMPLE 32 SRC2 FMT_NUM_U32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_LSHL_OR_B32 Given a shift count in the second input, calculate the logical shift left of the first input, then calculate the bitwise OR of the intermediate result and the third input, then store the final result into a vector register. ENC_VOP3 default 512 VDST FMT_NUM_U32 OPR_VGPR 32 SRC0 FMT_NUM_U32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_U32 OPR_SRC_SIMPLE 32 SRC2 FMT_NUM_U32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_AND_OR_B32 Calculate bitwise AND on the first two vector inputs, then compute the bitwise OR of the intermediate result and the third vector input, then store the final result into a vector register. ENC_VOP3 default 513 VDST FMT_NUM_U32 OPR_VGPR 32 SRC0 FMT_NUM_U32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_U32 OPR_SRC_SIMPLE 32 SRC2 FMT_NUM_U32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_OR3_B32 Calculate the bitwise OR of three vector inputs and store the result into a vector register. ENC_VOP3 default 514 VDST FMT_NUM_U32 OPR_VGPR 32 SRC0 FMT_NUM_U32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_U32 OPR_SRC_SIMPLE 32 SRC2 FMT_NUM_U32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_MAD_F16 Multiply two half-precision float inputs and add a third input, and store the result into a vector register. ENC_VOP3 default 515 VDST FMT_NUM_F16 OPR_VGPR 16 SRC0 FMT_NUM_F16 OPR_SRC_NOLIT 16 SRC1 FMT_NUM_F16 OPR_SRC_SIMPLE 16 SRC2 FMT_NUM_F16 OPR_SRC_SIMPLE 16 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_MAD_U16 Multiply two unsigned 16-bit integer inputs, add an unsigned 16-bit integer value from a third input, and store the result into a vector register. ENC_VOP3 default 516 VDST FMT_NUM_U16 OPR_VGPR 16 SRC0 FMT_NUM_U16 OPR_SRC_NOLIT 16 SRC1 FMT_NUM_U16 OPR_SRC_SIMPLE 16 SRC2 FMT_NUM_U16 OPR_SRC_SIMPLE 16 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_MAD_I16 Multiply two signed 16-bit integer inputs, add a signed 16-bit integer value from a third input, and store the result into a vector register. ENC_VOP3 default 517 VDST FMT_NUM_I16 OPR_VGPR 16 SRC0 FMT_NUM_I16 OPR_SRC_NOLIT 16 SRC1 FMT_NUM_I16 OPR_SRC_SIMPLE 16 SRC2 FMT_NUM_I16 OPR_SRC_SIMPLE 16 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_FMA_F16 Multiply two half-precision float inputs and add a third input using fused multiply add, and store the result into a vector register. ENC_VOP3 default 518 VDST FMT_NUM_F16 OPR_VGPR 16 SRC0 FMT_NUM_F16 OPR_SRC_NOLIT 16 SRC1 FMT_NUM_F16 OPR_SRC_SIMPLE 16 SRC2 FMT_NUM_F16 OPR_SRC_SIMPLE 16 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_DIV_FIXUP_F16 Given a half-precision float quotient in the first input, a denominator in the second input and a numerator in the third input, detect and apply corner cases related to division, including divide by zero, NaN inputs and overflow, and modify the quotient accordingly. Generate any invalid, denormal and divide-by-zero exceptions that are a result of the division. Store the modified quotient into a vector register. ENC_VOP3 default 519 VDST FMT_NUM_F16 OPR_VGPR 16 SRC0 FMT_NUM_F16 OPR_SRC_NOLIT 16 SRC1 FMT_NUM_F16 OPR_SRC_SIMPLE 16 SRC2 FMT_NUM_F16 OPR_SRC_SIMPLE 16 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_LSHL_ADD_U64 Given a shift count in the second input, calculate the logical shift left of the first input, then add the third input to the intermediate result, then store the final result into a vector register. ENC_VOP3 default 520 VDST FMT_NUM_U64 OPR_VGPR 64 SRC0 FMT_NUM_U64 OPR_SRC_NOLIT 64 SRC1 FMT_NUM_U32 OPR_SRC_SIMPLE 32 SRC2 FMT_NUM_U64 OPR_SRC_SIMPLE 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_BITOP3_B16 Calculate the generic bitwise operation of three 16-bit vector inputs using a truth table encoded in the instruction and store the result into a vector register. ENC_VOP3 default 563 VDST FMT_NUM_B16 OPR_VGPR 16 SRC0 FMT_NUM_B16 OPR_SRC_NOLIT 16 SRC1 FMT_NUM_B16 OPR_SRC_SIMPLE 16 SRC2 FMT_NUM_B16 OPR_SRC_SIMPLE 16 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_BITOP3_B32 Calculate the generic bitwise operation of three 32-bit vector inputs using a truth table encoded in the instruction and store the result into a vector register. ENC_VOP3 default 564 VDST FMT_NUM_B32 OPR_VGPR 32 SRC0 FMT_NUM_B32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_B32 OPR_SRC_SIMPLE 32 SRC2 FMT_NUM_B32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CVT_SCALEF32_PK_FP8_F32 V_CVT_SCALE_PK_FP8_F32 Scale two single-precision float inputs using the exponent provided by the third single-precision float input, then convert the values to a packed FP8 float value with round toward nearest even semantics. Store the result into 16 bits of a vector register using OPSEL. ENC_VOP3 default 565 VDST FMT_NUM_PK2_FP8 OPR_VGPR 16 SRC0 FMT_NUM_F32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_F32 OPR_SRC_SIMPLE 32 SRC2 FMT_NUM_F32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CVT_SCALEF32_PK_BF8_F32 V_CVT_SCALE_PK_BF8_F32 Scale two single-precision float inputs using the exponent provided by the third single-precision float input, then convert the values to a packed BF8 float value with round toward nearest even semantics. Store the result into 16 bits of a vector register using OPSEL. ENC_VOP3 default 566 VDST FMT_NUM_PK2_BF8 OPR_VGPR 16 SRC0 FMT_NUM_F32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_F32 OPR_SRC_SIMPLE 32 SRC2 FMT_NUM_F32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CVT_SCALEF32_SR_FP8_F32 V_CVT_SCALE_SR_FP8_F32 Scale a single-precision float input using the exponent provided by the third single-precision float input, then convert the values to an FP8 float value with stochastic rounding using seed data from the second input. Store the result into 8 bits of a vector register using OPSEL[3:2] to determine which byte of the destination to overwrite. ENC_VOP3 default 567 VDST FMT_NUM_FP8 OPR_VGPR 8 SRC0 FMT_NUM_F32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_U32 OPR_SRC_SIMPLE 32 SRC2 FMT_NUM_F32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CVT_SCALEF32_SR_BF8_F32 V_CVT_SCALE_SR_BF8_F32 Scale a single-precision float input using the exponent provided by the third single-precision float input, then convert the values to a BF8 float value with stochastic rounding using seed data from the second input. Store the result into 8 bits of a vector register using OPSEL[3:2] to determine which byte of the destination to overwrite. ENC_VOP3 default 568 VDST FMT_NUM_BF8 OPR_VGPR 8 SRC0 FMT_NUM_F32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_U32 OPR_SRC_SIMPLE 32 SRC2 FMT_NUM_F32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CVT_SCALEF32_PK_F32_FP8 V_CVT_SCALE_PK_F32_FP8 Convert from a packed 2-component FP8 float input to a packed single-precision float value, then scale the packed values using the exponent provided by the second single-precision float input. Store the result into a vector register. ENC_VOP3 default 569 VDST FMT_NUM_PK2_F32 OPR_VGPR 64 SRC0 FMT_NUM_PK2_FP8 OPR_SRC_NOLIT 16 SRC1 FMT_NUM_F32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CVT_SCALEF32_PK_F32_BF8 V_CVT_SCALE_PK_F32_BF8 Convert from a packed 2-component BF8 float input to a packed single-precision float value, then scale the packed values using the exponent provided by the second single-precision float input. Store the result into a vector register. ENC_VOP3 default 570 VDST FMT_NUM_PK2_F32 OPR_VGPR 64 SRC0 FMT_NUM_PK2_BF8 OPR_SRC_NOLIT 16 SRC1 FMT_NUM_F32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CVT_SCALEF32_F32_FP8 V_CVT_SCALE_F32_FP8 Convert from an FP8 float input to a single-precision float value, then scale the value using the exponent provided by the second single-precision float input. Store the result into a vector register. The value to convert is loaded from 8 bits of the input using OPSEL[1:0] to determine which byte to read. ENC_VOP3 default 571 VDST FMT_NUM_F32 OPR_VGPR 32 SRC0 FMT_NUM_FP8 OPR_SRC_NOLIT 8 SRC1 FMT_NUM_F32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CVT_SCALEF32_F32_BF8 V_CVT_SCALE_F32_BF8 Convert from a BF8 float input to a single-precision float value, then scale the value using the exponent provided by the second single-precision float input. Store the result into a vector register. The value to convert is loaded from 8 bits of the input using OPSEL[1:0] to determine which byte to read. ENC_VOP3 default 572 VDST FMT_NUM_F32 OPR_VGPR 32 SRC0 FMT_NUM_BF8 OPR_SRC_NOLIT 8 SRC1 FMT_NUM_F32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CVT_SCALEF32_PK_FP4_F32 V_CVT_SCALE_PK_FP4_F32 Scale two single-precision float inputs using the exponent provided by the third single-precision float input, then convert the values to a packed FP4 float value with round toward nearest even semantics. Store the result into 8 bits of a vector register using OPSEL[3:2] to determine which byte of the destination to overwrite. ENC_VOP3 default 573 VDST FMT_NUM_PK2_FP4 OPR_VGPR 8 SRC0 FMT_NUM_F32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_F32 OPR_SRC_SIMPLE 32 SRC2 FMT_NUM_F32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CVT_SCALEF32_SR_PK_FP4_F32 V_CVT_SCALE_SR_PK_FP4_F32 Scale a packed 2-component single-precision float input using the exponent provided by the third single-precision float input, then convert the values to a packed FP4 float value with stochastic rounding using seed data from the second input. Store the result into 8 bits of a vector register using OPSEL[3:2] to determine which byte of the destination to overwrite. ENC_VOP3 default 574 VDST FMT_NUM_PK2_FP4 OPR_VGPR 8 SRC0 FMT_NUM_PK2_F32 OPR_SRC_VGPR 64 SRC1 FMT_NUM_U32 OPR_SRC_SIMPLE 32 SRC2 FMT_NUM_F32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CVT_SCALEF32_PK_F32_FP4 V_CVT_SCALE_PK_F32_FP4 Convert from a packed 2-component FP4 float input to a packed single-precision float value, then scale the packed values using the exponent provided by the second single-precision float input. Store the result into a vector register. The value to convert is loaded from 8 bits of the input using OPSEL[1:0] to determine which byte to read. ENC_VOP3 default 575 VDST FMT_NUM_PK2_F32 OPR_VGPR 64 SRC0 FMT_NUM_PK2_FP4 OPR_SRC_NOLIT 8 SRC1 FMT_NUM_F32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CVT_SCALEF32_PK_FP8_F16 V_CVT_SCALE_PK_FP8_F16 Scale a packed 2-component half-precision float input using the exponent provided by the second single-precision float input, then convert the values to a packed FP8 float value with round toward nearest even semantics. Store the result into 16 bits of a vector register using OPSEL. ENC_VOP3 default 576 VDST FMT_NUM_PK2_FP8 OPR_VGPR 16 SRC0 FMT_NUM_PK2_F16 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_F32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CVT_SCALEF32_PK_BF8_F16 V_CVT_SCALE_PK_BF8_F16 Scale a packed 2-component half-precision float input using the exponent provided by the second single-precision float input, then convert the values to a packed BF8 float value with round toward nearest even semantics. Store the result into 16 bits of a vector register using OPSEL. ENC_VOP3 default 577 VDST FMT_NUM_PK2_BF8 OPR_VGPR 16 SRC0 FMT_NUM_PK2_F16 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_F32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CVT_SCALEF32_SR_FP8_F16 V_CVT_SCALE_SR_FP8_F16 Scale a half-precision float input using the exponent provided by the third single-precision float input, then convert the values to an FP8 float value with stochastic rounding using seed data from the second input. Store the result into 8 bits of a vector register using OPSEL[3:2] to determine which byte of the destination to overwrite. ENC_VOP3 default 578 VDST FMT_NUM_FP8 OPR_VGPR 8 SRC0 FMT_NUM_F16 OPR_SRC_NOLIT 16 SRC1 FMT_NUM_U32 OPR_SRC_SIMPLE 32 SRC2 FMT_NUM_F32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CVT_SCALEF32_SR_BF8_F16 V_CVT_SCALE_SR_BF8_F16 Scale a half-precision float input using the exponent provided by the third single-precision float input, then convert the values to a BF8 float value with stochastic rounding using seed data from the second input. Store the result into 8 bits of a vector register using OPSEL[3:2] to determine which byte of the destination to overwrite. ENC_VOP3 default 579 VDST FMT_NUM_BF8 OPR_VGPR 8 SRC0 FMT_NUM_F16 OPR_SRC_NOLIT 16 SRC1 FMT_NUM_U32 OPR_SRC_SIMPLE 32 SRC2 FMT_NUM_F32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CVT_SCALEF32_PK_FP8_BF16 V_CVT_SCALE_PK_FP8_BF16 Scale a packed 2-component BF16 float input using the exponent provided by the second single-precision float input, then convert the values to a packed FP8 float value with round toward nearest even semantics. Store the result into 16 bits of a vector register using OPSEL. ENC_VOP3 default 580 VDST FMT_NUM_PK2_FP8 OPR_VGPR 16 SRC0 FMT_NUM_PK2_BF16 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_F32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CVT_SCALEF32_PK_BF8_BF16 V_CVT_SCALE_PK_BF8_BF16 Scale a packed 2-component BF16 float input using the exponent provided by the second single-precision float input, then convert the values to a packed BF8 float value with round toward nearest even semantics. Store the result into 16 bits of a vector register using OPSEL. ENC_VOP3 default 581 VDST FMT_NUM_PK2_BF8 OPR_VGPR 16 SRC0 FMT_NUM_PK2_BF16 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_F32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CVT_SCALEF32_SR_FP8_BF16 V_CVT_SCALE_SR_FP8_BF16 Scale a BF16 float input using the exponent provided by the third single-precision float input, then convert the values to an FP8 float value with stochastic rounding using seed data from the second input. Store the result into 8 bits of a vector register using OPSEL[3:2] to determine which byte of the destination to overwrite. ENC_VOP3 default 582 VDST FMT_NUM_FP8 OPR_VGPR 8 SRC0 FMT_NUM_BF16 OPR_SRC_NOLIT 16 SRC1 FMT_NUM_U32 OPR_SRC_SIMPLE 32 SRC2 FMT_NUM_F32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CVT_SCALEF32_SR_BF8_BF16 V_CVT_SCALE_SR_BF8_BF16 Scale a BF16 float input using the exponent provided by the third single-precision float input, then convert the values to a BF8 float value with stochastic rounding using seed data from the second input. Store the result into 8 bits of a vector register using OPSEL[3:2] to determine which byte of the destination to overwrite. ENC_VOP3 default 583 VDST FMT_NUM_BF8 OPR_VGPR 8 SRC0 FMT_NUM_BF16 OPR_SRC_NOLIT 16 SRC1 FMT_NUM_U32 OPR_SRC_SIMPLE 32 SRC2 FMT_NUM_F32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CVT_SCALEF32_PK_F16_FP8 V_CVT_SCALE_PK_F16_FP8 Convert from a packed 2-component FP8 float input to a packed half-precision float value, then scale the packed values using the exponent provided by the second single-precision float input. Store the result into a vector register. ENC_VOP3 default 584 VDST FMT_NUM_PK2_F16 OPR_VGPR 32 SRC0 FMT_NUM_PK2_FP8 OPR_SRC_NOLIT 16 SRC1 FMT_NUM_F32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CVT_SCALEF32_PK_F16_BF8 V_CVT_SCALE_PK_F16_BF8 Convert from a packed 2-component BF8 float input to a packed half-precision float value, then scale the packed values using the exponent provided by the second single-precision float input. Store the result into a vector register. ENC_VOP3 default 585 VDST FMT_NUM_PK2_F16 OPR_VGPR 32 SRC0 FMT_NUM_PK2_BF8 OPR_SRC_NOLIT 16 SRC1 FMT_NUM_F32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CVT_SCALEF32_F16_FP8 V_CVT_SCALE_F16_FP8 Convert from an FP8 float input to a half-precision float value, then scale the value using the exponent provided by the second single-precision float input. Store the result into a vector register. The value to convert is loaded from 8 bits of the input using OPSEL[1:0] to determine which byte to read. ENC_VOP3 default 586 VDST FMT_NUM_F16 OPR_VGPR 16 SRC0 FMT_NUM_FP8 OPR_SRC_NOLIT 8 SRC1 FMT_NUM_F32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CVT_SCALEF32_F16_BF8 V_CVT_SCALE_F16_BF8 Convert from a BF8 float input to a half-precision float value, then scale the value using the exponent provided by the second single-precision float input. Store the result into a vector register. The value to convert is loaded from 8 bits of the input using OPSEL[1:0] to determine which byte to read. ENC_VOP3 default 587 VDST FMT_NUM_F16 OPR_VGPR 16 SRC0 FMT_NUM_BF8 OPR_SRC_NOLIT 8 SRC1 FMT_NUM_F32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CVT_SCALEF32_PK_FP4_F16 V_CVT_SCALE_PK_FP4_F16 Scale a packed 2-component half-precision float input using the exponent provided by the second single-precision float input, then convert the values to a packed FP4 float value with round toward nearest even semantics. Store the result into 8 bits of a vector register using OPSEL[3:2] to determine which byte of the destination to overwrite. ENC_VOP3 default 588 VDST FMT_NUM_PK2_FP4 OPR_VGPR 8 SRC0 FMT_NUM_PK2_F16 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_F32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CVT_SCALEF32_PK_FP4_BF16 V_CVT_SCALE_PK_FP4_BF16 Scale a packed 2-component BF16 float input using the exponent provided by the second single-precision float input, then convert the values to a packed FP4 float value with round toward nearest even semantics. Store the result into 8 bits of a vector register using OPSEL[3:2] to determine which byte of the destination to overwrite. ENC_VOP3 default 589 VDST FMT_NUM_PK2_FP4 OPR_VGPR 8 SRC0 FMT_NUM_PK2_BF16 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_F32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CVT_SCALEF32_SR_PK_FP4_F16 V_CVT_SCALE_SR_PK_FP4_F16 Scale a packed 2-component half-precision float input using the exponent provided by the third single-precision float input, then convert the values to a packed FP4 float value with stochastic rounding using seed data from the second input. Store the result into 8 bits of a vector register using OPSEL[3:2] to determine which byte of the destination to overwrite. ENC_VOP3 default 590 VDST FMT_NUM_PK2_FP4 OPR_VGPR 8 SRC0 FMT_NUM_PK2_F16 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_U32 OPR_SRC_SIMPLE 32 SRC2 FMT_NUM_F32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CVT_SCALEF32_SR_PK_FP4_BF16 V_CVT_SCALE_SR_PK_FP4_BF16 Scale a packed 2-component BF16 float input using the exponent provided by the third single-precision float input, then convert the values to a packed FP4 float value with stochastic rounding using seed data from the second input. Store the result into 8 bits of a vector register using OPSEL[3:2] to determine which byte of the destination to overwrite. ENC_VOP3 default 591 VDST FMT_NUM_PK2_FP4 OPR_VGPR 8 SRC0 FMT_NUM_PK2_BF16 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_U32 OPR_SRC_SIMPLE 32 SRC2 FMT_NUM_F32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CVT_SCALEF32_PK_F16_FP4 V_CVT_SCALE_PK_F16_FP4 Convert from a packed 2-component FP4 float input to a packed half-precision float value, then scale the packed values using the exponent provided by the second single-precision float input. Store the result into a vector register. The value to convert is loaded from 8 bits of the input using OPSEL[1:0] to determine which byte to read. ENC_VOP3 default 592 VDST FMT_NUM_PK2_F16 OPR_VGPR 32 SRC0 FMT_NUM_PK2_FP4 OPR_SRC_NOLIT 8 SRC1 FMT_NUM_F32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CVT_SCALEF32_PK_BF16_FP4 V_CVT_SCALE_PK_BF16_FP4 Convert from a packed 2-component FP4 float input to a packed BF16 float value, then scale the packed values using the exponent provided by the second single-precision float input. Store the result into a vector register. The value to convert is loaded from 8 bits of the input using OPSEL[1:0] to determine which byte to read. ENC_VOP3 default 593 VDST FMT_NUM_PK2_BF16 OPR_VGPR 32 SRC0 FMT_NUM_PK2_FP4 OPR_SRC_NOLIT 8 SRC1 FMT_NUM_F32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CVT_SCALEF32_2XPK16_FP6_F32 V_CVT_SCALE_PK_FP6_F32 Scale packed 16-component single-precision float vectors from two source inputs using the exponent provided by the third single-precision float input, then convert the values to a packed 32-component FP6 float value. Store the result into a vector register. ENC_VOP3 default 594 VDST FMT_NUM_PK32_FP6 OPR_VGPR 192 SRC0 FMT_NUM_PK16_F32 OPR_SRC_VGPR 512 SRC1 FMT_NUM_PK16_F32 OPR_SRC_VGPR 512 SRC2 FMT_NUM_F32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CVT_SCALEF32_2XPK16_BF6_F32 V_CVT_SCALE_PK_BF6_F32 Scale packed 16-component single-precision float vectors from two source inputs using the exponent provided by the third single-precision float input, then convert the values to a packed 32-component BF6 float value. Store the result into a vector register. ENC_VOP3 default 595 VDST FMT_NUM_PK32_BF6 OPR_VGPR 192 SRC0 FMT_NUM_PK16_F32 OPR_SRC_VGPR 512 SRC1 FMT_NUM_PK16_F32 OPR_SRC_VGPR 512 SRC2 FMT_NUM_F32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CVT_SCALEF32_SR_PK32_FP6_F32 V_CVT_SCALE_SR_PK_FP6_F32 Scale a packed 32-component single-precision float input using the exponent provided by the third single-precision float input, then convert the values to a packed 32-component FP6 float value with stochastic rounding using seed data from the second input. Store the result into a vector register. ENC_VOP3 default 596 VDST FMT_NUM_PK32_FP6 OPR_VGPR 192 SRC0 FMT_NUM_PK32_F32 OPR_SRC_VGPR 1024 SRC1 FMT_NUM_U32 OPR_SRC_SIMPLE 32 SRC2 FMT_NUM_F32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CVT_SCALEF32_SR_PK32_BF6_F32 V_CVT_SCALE_SR_PK_BF6_F32 Scale a packed 32-component single-precision float input using the exponent provided by the third single-precision float input, then convert the values to a packed 32-component BF6 float value with stochastic rounding using seed data from the second input. Store the result into a vector register. ENC_VOP3 default 597 VDST FMT_NUM_PK32_BF6 OPR_VGPR 192 SRC0 FMT_NUM_PK32_F32 OPR_SRC_VGPR 1024 SRC1 FMT_NUM_U32 OPR_SRC_SIMPLE 32 SRC2 FMT_NUM_F32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CVT_SCALEF32_PK32_F32_FP6 V_CVT_SCALE_PK_F32_FP6 Convert from a packed 32-component FP6 float input to a packed single-precision float value, then scale the packed values using the exponent provided by the second single-precision float input. Store the result into a vector register. ENC_VOP3 default 598 VDST FMT_NUM_PK32_F32 OPR_VGPR 1024 SRC0 FMT_NUM_PK32_FP6 OPR_SRC_VGPR 192 SRC1 FMT_NUM_F32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CVT_SCALEF32_PK32_F32_BF6 V_CVT_SCALE_PK_F32_BF6 Convert from a packed 32-component BF6 float input to a packed single-precision float value, then scale the packed values using the exponent provided by the second single-precision float input. Store the result into a vector register. ENC_VOP3 default 599 VDST FMT_NUM_PK32_F32 OPR_VGPR 1024 SRC0 FMT_NUM_PK32_BF6 OPR_SRC_VGPR 192 SRC1 FMT_NUM_F32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CVT_SCALEF32_PK32_FP6_F16 V_CVT_SCALE_PK_FP6_F16 Scale a packed 32-component half-precision float input using the exponent provided by the second single-precision float input, then convert the values to a packed 32-component FP6 float value. Store the result into a vector register. ENC_VOP3 default 600 VDST FMT_NUM_PK32_FP6 OPR_VGPR 192 SRC0 FMT_NUM_PK32_F16 OPR_SRC_VGPR 512 SRC1 FMT_NUM_F32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CVT_SCALEF32_PK32_FP6_BF16 V_CVT_SCALE_PK_FP6_BF16 Scale a packed 32-component BF16 float input using the exponent provided by the second single-precision float input, then convert the values to a packed 32-component FP6 float value. Store the result into a vector register. ENC_VOP3 default 601 VDST FMT_NUM_PK32_FP6 OPR_VGPR 192 SRC0 FMT_NUM_PK32_BF16 OPR_SRC_VGPR 512 SRC1 FMT_NUM_F32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CVT_SCALEF32_PK32_BF6_F16 V_CVT_SCALE_PK_BF6_F16 Scale a packed 32-component half-precision float input using the exponent provided by the second single-precision float input, then convert the values to a packed 32-component BF6 float value. Store the result into a vector register. ENC_VOP3 default 602 VDST FMT_NUM_PK32_BF6 OPR_VGPR 192 SRC0 FMT_NUM_PK32_F16 OPR_SRC_VGPR 512 SRC1 FMT_NUM_F32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CVT_SCALEF32_PK32_BF6_BF16 V_CVT_SCALE_PK_BF6_BF16 Scale a packed 32-component BF16 float input using the exponent provided by the second single-precision float input, then convert the values to a packed 32-component BF6 float value. Store the result into a vector register. ENC_VOP3 default 603 VDST FMT_NUM_PK32_BF6 OPR_VGPR 192 SRC0 FMT_NUM_PK32_BF16 OPR_SRC_VGPR 512 SRC1 FMT_NUM_F32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CVT_SCALEF32_SR_PK32_FP6_F16 V_CVT_SCALE_SR_PK_FP6_F16 Scale a packed 32-component half-precision float input using the exponent provided by the third single-precision float input, then convert the values to a packed 32-component FP6 float value with stochastic rounding using seed data from the second input. Store the result into a vector register. ENC_VOP3 default 604 VDST FMT_NUM_PK32_FP6 OPR_VGPR 192 SRC0 FMT_NUM_PK32_F16 OPR_SRC_VGPR 512 SRC1 FMT_NUM_U32 OPR_SRC_SIMPLE 32 SRC2 FMT_NUM_F32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CVT_SCALEF32_SR_PK32_FP6_BF16 V_CVT_SCALE_SR_PK_FP6_BF16 Scale a packed 32-component BF16 float input using the exponent provided by the third single-precision float input, then convert the values to a packed 32-component FP6 float value with stochastic rounding using seed data from the second input. Store the result into a vector register. ENC_VOP3 default 605 VDST FMT_NUM_PK32_FP6 OPR_VGPR 192 SRC0 FMT_NUM_PK32_BF16 OPR_SRC_VGPR 512 SRC1 FMT_NUM_U32 OPR_SRC_SIMPLE 32 SRC2 FMT_NUM_F32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CVT_SCALEF32_SR_PK32_BF6_F16 V_CVT_SCALE_SR_PK_BF6_F16 Scale a packed 32-component half-precision float input using the exponent provided by the third single-precision float input, then convert the values to a packed 32-component BF6 float value with stochastic rounding using seed data from the second input. Store the result into a vector register. ENC_VOP3 default 606 VDST FMT_NUM_PK32_BF6 OPR_VGPR 192 SRC0 FMT_NUM_PK32_F16 OPR_SRC_VGPR 512 SRC1 FMT_NUM_U32 OPR_SRC_SIMPLE 32 SRC2 FMT_NUM_F32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CVT_SCALEF32_SR_PK32_BF6_BF16 V_CVT_SCALE_SR_PK_BF6_BF16 Scale a packed 32-component BF16 float input using the exponent provided by the third single-precision float input, then convert the values to a packed 32-component BF6 float value with stochastic rounding using seed data from the second input. Store the result into a vector register. ENC_VOP3 default 607 VDST FMT_NUM_PK32_BF6 OPR_VGPR 192 SRC0 FMT_NUM_PK32_BF16 OPR_SRC_VGPR 512 SRC1 FMT_NUM_U32 OPR_SRC_SIMPLE 32 SRC2 FMT_NUM_F32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CVT_SCALEF32_PK32_F16_FP6 V_CVT_SCALE_PK_F16_FP6 Convert from a packed 32-component FP6 float input to a packed half-precision float value, then scale the packed values using the exponent provided by the second single-precision float input. Store the result into a vector register. ENC_VOP3 default 608 VDST FMT_NUM_PK32_F16 OPR_VGPR 512 SRC0 FMT_NUM_PK32_FP6 OPR_SRC_VGPR 192 SRC1 FMT_NUM_F32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CVT_SCALEF32_PK32_BF16_FP6 V_CVT_SCALE_PK_BF16_FP6 Convert from a packed 32-component FP6 float input to a packed BF16 float value, then scale the packed values using the exponent provided by the second single-precision float input. Store the result into a vector register. ENC_VOP3 default 609 VDST FMT_NUM_PK32_BF16 OPR_VGPR 512 SRC0 FMT_NUM_PK32_FP6 OPR_SRC_VGPR 192 SRC1 FMT_NUM_F32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CVT_SCALEF32_PK32_F16_BF6 V_CVT_SCALE_PK_F16_BF6 Convert from a packed 32-component BF6 float input to a packed half-precision float value, then scale the packed values using the exponent provided by the second single-precision float input. Store the result into a vector register. ENC_VOP3 default 610 VDST FMT_NUM_PK32_F16 OPR_VGPR 512 SRC0 FMT_NUM_PK32_BF6 OPR_SRC_VGPR 192 SRC1 FMT_NUM_F32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CVT_SCALEF32_PK32_BF16_BF6 V_CVT_SCALE_PK_BF16_BF6 Convert from a packed 32-component BF6 float input to a packed BF16 float value, then scale the packed values using the exponent provided by the second single-precision float input. Store the result into a vector register. ENC_VOP3 default 611 VDST FMT_NUM_PK32_BF16 OPR_VGPR 512 SRC0 FMT_NUM_PK32_BF6 OPR_SRC_VGPR 192 SRC1 FMT_NUM_F32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_ASHR_PK_I8_I32 Given two signed 32-bit integers and a shift count, calculate the arithmetic shift right (preserving sign bit) of the two integers, saturate the two results in the signed 8-bit interval [-128, 127], pack the bytes and store the result into a vector register. ENC_VOP3 default 613 VDST FMT_NUM_PK2_I8 OPR_VGPR 16 SRC0 FMT_NUM_I32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_I32 OPR_SRC_SIMPLE 32 SRC2 FMT_NUM_U32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_ASHR_PK_U8_I32 Given two signed 32-bit integers and a shift count, calculate the arithmetic shift right (preserving sign bit) of the two integers, saturate the two results in the unsigned 8-bit interval [0, 255], pack the bytes and store the result into a vector register. ENC_VOP3 default 614 VDST FMT_NUM_PK2_U8 OPR_VGPR 16 SRC0 FMT_NUM_I32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_I32 OPR_SRC_SIMPLE 32 SRC2 FMT_NUM_U32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CVT_PK_F16_F32 Convert from two single-precision float inputs to a packed half-precision value and store the result into a vector register. ENC_VOP3 default 615 VDST FMT_NUM_PK2_F16 OPR_VGPR 32 SRC0 FMT_NUM_F32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_F32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CVT_PK_BF16_F32 Convert from two single-precision float inputs to a packed BF16 value and store the result into a vector register. ENC_VOP3 default 616 VDST FMT_NUM_PK2_BF16 OPR_VGPR 32 SRC0 FMT_NUM_F32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_F32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CVT_SCALEF32_PK_BF16_FP8 V_CVT_SCALE_PK_BF16_FP8 Convert from a packed 2-component FP8 float input to a packed BF16 float value, then scale the packed values using the exponent provided by the second single-precision float input. Store the result into a vector register. ENC_VOP3 default 617 VDST FMT_NUM_PK2_BF16 OPR_VGPR 32 SRC0 FMT_NUM_PK2_FP8 OPR_SRC_NOLIT 16 SRC1 FMT_NUM_F32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CVT_SCALEF32_PK_BF16_BF8 V_CVT_SCALE_PK_BF16_BF8 Convert from a packed 2-component BF8 float input to a packed BF16 float value, then scale the packed values using the exponent provided by the second single-precision float input. Store the result into a vector register. ENC_VOP3 default 618 VDST FMT_NUM_PK2_BF16 OPR_VGPR 32 SRC0 FMT_NUM_PK2_BF8 OPR_SRC_NOLIT 16 SRC1 FMT_NUM_F32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_ADD_F64 Add two floating point inputs and store the result into a vector register. ENC_VOP3 default 640 VDST FMT_NUM_F64 OPR_VGPR 64 SRC0 FMT_NUM_F64 OPR_SRC_NOLIT 64 SRC1 FMT_NUM_F64 OPR_SRC_SIMPLE 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_MUL_F64 Multiply two floating point inputs and store the result into a vector register. ENC_VOP3 default 641 VDST FMT_NUM_F64 OPR_VGPR 64 SRC0 FMT_NUM_F64 OPR_SRC_NOLIT 64 SRC1 FMT_NUM_F64 OPR_SRC_SIMPLE 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_MIN_F64 Select the minimum of two double-precision float inputs and store the selected value into a vector register. ENC_VOP3 default 642 VDST FMT_NUM_F64 OPR_VGPR 64 SRC0 FMT_NUM_F64 OPR_SRC_NOLIT 64 SRC1 FMT_NUM_F64 OPR_SRC_SIMPLE 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_MAX_F64 Select the maximum of two double-precision float inputs and store the selected value into a vector register. ENC_VOP3 default 643 VDST FMT_NUM_F64 OPR_VGPR 64 SRC0 FMT_NUM_F64 OPR_SRC_NOLIT 64 SRC1 FMT_NUM_F64 OPR_SRC_SIMPLE 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_LDEXP_F64 Multiply the first input, a floating point value, by an integral power of 2 specified in the second input, a signed integer value, and store the floating point result into a vector register. ENC_VOP3 default 644 VDST FMT_NUM_F64 OPR_VGPR 64 SRC0 FMT_NUM_F64 OPR_SRC_NOLIT 64 SRC1 FMT_NUM_I32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_MUL_LO_U32 Multiply two unsigned 32-bit integer inputs and store the result into a vector register. ENC_VOP3 default 645 VDST FMT_NUM_U32 OPR_VGPR 32 SRC0 FMT_NUM_U32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_U32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_MUL_HI_U32 Multiply two unsigned 32-bit integer inputs and store the high 32 bits of the result into a vector register. ENC_VOP3 default 646 VDST FMT_NUM_U32 OPR_VGPR 32 SRC0 FMT_NUM_U32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_U32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_MUL_HI_I32 Multiply two signed 32-bit integer inputs and store the high 32 bits of the result into a vector register. ENC_VOP3 default 647 VDST FMT_NUM_I32 OPR_VGPR 32 SRC0 FMT_NUM_I32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_I32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_LDEXP_F32 Multiply the first input, a floating point value, by an integral power of 2 specified in the second input, a signed integer value, and store the floating point result into a vector register. ENC_VOP3 default 648 VDST FMT_NUM_F32 OPR_VGPR 32 SRC0 FMT_NUM_F32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_I32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_READLANE_B32 Read the scalar value in the specified lane of the first input where the lane select is in the second input. Store the result into a scalar register. ENC_VOP3 default 649 VDST FMT_NUM_B32 OPR_SREG_NOVCC 32 SRC0 FMT_NUM_B32 OPR_VGPR_OR_LDS 32 SRC1 FMT_NUM_B32 OPR_SSRC_LANESEL 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_WRITELANE_B32 Write the scalar value in the first input into the specified lane of a vector register where the lane select is in the second input. ENC_VOP3 default 650 VDST FMT_NUM_B32 OPR_VGPR 32 SRC0 FMT_NUM_B32 OPR_SSRC_NOLIT 32 SRC1 FMT_NUM_B32 OPR_SSRC_LANESEL 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_BCNT_U32_B32 Count the number of "1" bits in the vector input and store the result into a vector register. ENC_VOP3 default 651 VDST FMT_NUM_U32 OPR_VGPR 32 SRC0 FMT_NUM_U32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_U32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_MBCNT_LO_U32_B32 For each lane 0 <= N < 32, examine the N least significant bits of the first input and count how many of those bits are "1". For each lane 32 <= N < 64, all "1" bits in the first input are counted. Add this count to the value in the second input and store the result into a vector register. ENC_VOP3 default 652 VDST FMT_NUM_U32 OPR_VGPR 32 SRC0 FMT_NUM_U32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_U32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_MBCNT_HI_U32_B32 For each lane 32 <= N < 64, examine the N least significant bits of the first input and count how many of those bits are "1". For lane positions 0 <= N < 32 no bits are examined and the count is zero. Add this count to the value in the second input and store the result into a vector register. ENC_VOP3 default 653 VDST FMT_NUM_U32 OPR_VGPR 32 SRC0 FMT_NUM_U32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_U32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_LSHLREV_B64 Given a shift count in the first vector input, calculate the logical shift left of the second vector input and store the result into a vector register. ENC_VOP3 default 655 VDST FMT_NUM_U64 OPR_VGPR 64 SRC0 FMT_NUM_U32 OPR_SRC_SIMPLE 32 SRC1 FMT_NUM_U64 OPR_SRC_SIMPLE 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_LSHRREV_B64 Given a shift count in the first vector input, calculate the logical shift right of the second vector input and store the result into a vector register. ENC_VOP3 default 656 VDST FMT_NUM_U64 OPR_VGPR 64 SRC0 FMT_NUM_U32 OPR_SRC_SIMPLE 32 SRC1 FMT_NUM_U64 OPR_SRC_SIMPLE 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_ASHRREV_I64 Given a shift count in the first vector input, calculate the arithmetic shift right (preserving sign bit) of the second vector input and store the result into a vector register. ENC_VOP3 default 657 VDST FMT_NUM_I64 OPR_VGPR 64 SRC0 FMT_NUM_U32 OPR_SRC_SIMPLE 32 SRC1 FMT_NUM_I64 OPR_SRC_SIMPLE 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_TRIG_PREOP_F64 Look up a 53-bit segment of 2/PI using an integer segment select in the second input. Scale the intermediate result by the exponent from the first double-precision float input and store the double-precision float result into a vector register. ENC_VOP3 default 658 VDST FMT_NUM_F64 OPR_VGPR 64 SRC0 FMT_NUM_F64 OPR_SRC_NOLIT 64 SRC1 FMT_NUM_B32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_BFM_B32 Calculate a bitfield mask given a field offset and size and store the result into a vector register. ENC_VOP3 default 659 VDST FMT_NUM_U32 OPR_VGPR 32 SRC0 FMT_NUM_U32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_U32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CVT_PKNORM_I16_F32 Convert from two single-precision float inputs to a packed signed normalized short and store the result into a vector register. ENC_VOP3 default 660 VDST FMT_NUM_PK2_I16 OPR_VGPR 32 SRC0 FMT_NUM_F32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_F32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CVT_PKNORM_U16_F32 Convert from two single-precision float inputs to a packed unsigned normalized short and store the result into a vector register. ENC_VOP3 default 661 VDST FMT_NUM_PK2_U16 OPR_VGPR 32 SRC0 FMT_NUM_F32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_F32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CVT_PKRTZ_F16_F32 Convert two single-precision float inputs to a packed half-precision float value using round toward zero semantics (ignore the current rounding mode), and store the result into a vector register. ENC_VOP3 default 662 VDST FMT_NUM_PK2_F16 OPR_VGPR 32 SRC0 FMT_NUM_F32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_F32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CVT_PK_U16_U32 Convert from two unsigned 32-bit integer inputs to a packed unsigned 16-bit integer value and store the result into a vector register. ENC_VOP3 default 663 VDST FMT_NUM_PK2_U16 OPR_VGPR 32 SRC0 FMT_NUM_U32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_U32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CVT_PK_I16_I32 Convert from two signed 32-bit integer inputs to a packed signed 16-bit integer value and store the result into a vector register. ENC_VOP3 default 664 VDST FMT_NUM_PK2_I16 OPR_VGPR 32 SRC0 FMT_NUM_I32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_I32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CVT_PKNORM_I16_F16 Convert from two half-precision float inputs to a packed signed normalized short and store the result into a vector register. ENC_VOP3 default 665 VDST FMT_NUM_PK2_I16 OPR_VGPR 32 SRC0 FMT_NUM_F16 OPR_SRC_NOLIT 16 SRC1 FMT_NUM_F16 OPR_SRC_SIMPLE 16 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CVT_PKNORM_U16_F16 Convert from two half-precision float inputs to a packed unsigned normalized short and store the result into a vector register. ENC_VOP3 default 666 VDST FMT_NUM_PK2_U16 OPR_VGPR 32 SRC0 FMT_NUM_F16 OPR_SRC_NOLIT 16 SRC1 FMT_NUM_F16 OPR_SRC_SIMPLE 16 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_ADD_I32 Add two signed 32-bit integer inputs and store the result into a vector register. No carry-in or carry-out support. ENC_VOP3 default 668 VDST FMT_NUM_I32 OPR_VGPR 32 SRC0 FMT_NUM_I32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_I32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_SUB_I32 Subtract the second signed 32-bit integer input from the first input and store the result into a vector register. No carry-in or carry-out support. ENC_VOP3 default 669 VDST FMT_NUM_I32 OPR_VGPR 32 SRC0 FMT_NUM_I32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_I32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_ADD_I16 Add two signed 16-bit integer inputs and store the result into a vector register. No carry-in or carry-out support. ENC_VOP3 default 670 VDST FMT_NUM_I16 OPR_VGPR 16 SRC0 FMT_NUM_I16 OPR_SRC_NOLIT 16 SRC1 FMT_NUM_I16 OPR_SRC_SIMPLE 16 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_SUB_I16 Subtract the second signed 16-bit integer input from the first input and store the result into a vector register. No carry-in or carry-out support. ENC_VOP3 default 671 VDST FMT_NUM_I16 OPR_VGPR 16 SRC0 FMT_NUM_I16 OPR_SRC_NOLIT 16 SRC1 FMT_NUM_I16 OPR_SRC_SIMPLE 16 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_PACK_B32_F16 Pack two half-precision float values into a single 32-bit value and store the result into a vector register. ENC_VOP3 default 672 VDST FMT_NUM_B32 OPR_VGPR 32 SRC0 FMT_NUM_F16 OPR_SRC_NOLIT 16 SRC1 FMT_NUM_F16 OPR_SRC_SIMPLE 16 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_MUL_LEGACY_F32 Multiply two floating point inputs and store the result into a vector register. Follows DX9 rules where 0.0 times anything produces 0.0 (this differs from other APIs when the other input is infinity or NaN). ENC_VOP3 default 673 VDST FMT_NUM_F32 OPR_VGPR 32 SRC0 FMT_NUM_F32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_F32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CVT_PK_FP8_F32 Convert from two single-precision float inputs to a packed FP8 float value with round to nearest even semantics and store the result into 16 bits of a vector register using OPSEL. ENC_VOP3 default 674 VDST FMT_NUM_PK2_FP8 OPR_VGPR 16 SRC0 FMT_NUM_F32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_F32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CVT_PK_BF8_F32 Convert from two single-precision float inputs to a packed BF8 float value with round to nearest even semantics and store the result into 16 bits of a vector register using OPSEL. ENC_VOP3 default 675 VDST FMT_NUM_PK2_BF8 OPR_VGPR 16 SRC0 FMT_NUM_F32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_F32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CVT_SR_FP8_F32 Convert from a single-precision float input to an FP8 value with stochastic rounding using seed data from the second input. Store the result into 8 bits of a vector register using OPSEL to determine which byte of the destination to overwrite. ENC_VOP3 default 676 VDST FMT_NUM_FP8 OPR_VGPR 8 SRC0 FMT_NUM_F32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_U32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CVT_SR_BF8_F32 Convert from a single-precision float input to a BF8 value with stochastic rounding using seed data from the second input. Store the result into 8 bits of a vector register using OPSEL to determine which byte of the destination to overwrite. ENC_VOP3 default 677 VDST FMT_NUM_BF8 OPR_VGPR 8 SRC0 FMT_NUM_F32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_U32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CVT_SR_F16_F32 Convert from a single-precision float input to a half-precision value with stochastic rounding using seed data from the second input. Store the result into 16 bits of a vector register using OPSEL to determine which word of the destination to overwrite. ENC_VOP3 default 678 VDST FMT_NUM_F16 OPR_VGPR 16 SRC0 FMT_NUM_F32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_U32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CVT_SR_BF16_F32 Convert from a single-precision float input to a BF16 value with stochastic rounding using seed data from the second input. Store the result into 16 bits of a vector register using OPSEL to determine which word of the destination to overwrite. ENC_VOP3 default 679 VDST FMT_NUM_BF16 OPR_VGPR 16 SRC0 FMT_NUM_F32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_U32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_MINIMUM3_F32 Select the IEEE minimum() of three single-precision float inputs and store the result into a vector register. ENC_VOP3 default 680 VDST FMT_NUM_F32 OPR_VGPR 32 SRC0 FMT_NUM_F32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_F32 OPR_SRC_SIMPLE 32 SRC2 FMT_NUM_F32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_MAXIMUM3_F32 Select the IEEE maximum() of three single-precision float inputs and store the result into a vector register. ENC_VOP3 default 681 VDST FMT_NUM_F32 OPR_VGPR 32 SRC0 FMT_NUM_F32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_F32 OPR_SRC_SIMPLE 32 SRC2 FMT_NUM_F32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_PK_MAD_I16 Multiply two packed signed 16-bit integer inputs component-wise, add a packed signed 16-bit integer value from a third input component-wise, and store the result into a vector register. ENC_VOP3P default 0 VDST FMT_NUM_PK2_I16 OPR_VGPR 32 SRC0 FMT_NUM_PK2_I16 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_PK2_I16 OPR_SRC_SIMPLE 32 SRC2 FMT_NUM_PK2_I16 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_PK_MUL_LO_U16 Multiply two packed unsigned 16-bit integer inputs component-wise and store the low bits of each resulting component into a vector register. ENC_VOP3P default 1 VDST FMT_NUM_PK2_U16 OPR_VGPR 32 SRC0 FMT_NUM_PK2_U16 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_PK2_U16 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_PK_ADD_I16 Add two packed signed 16-bit integer inputs component-wise and store the result into a vector register. No carry-in or carry-out support. ENC_VOP3P default 2 VDST FMT_NUM_PK2_I16 OPR_VGPR 32 SRC0 FMT_NUM_PK2_I16 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_PK2_I16 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_PK_SUB_I16 Subtract the second packed signed 16-bit integer input from the first input component-wise and store the result into a vector register. No carry-in or carry-out support. ENC_VOP3P default 3 VDST FMT_NUM_PK2_I16 OPR_VGPR 32 SRC0 FMT_NUM_PK2_I16 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_PK2_I16 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_PK_LSHLREV_B16 Given a packed shift count in the first vector input, calculate the component-wise logical shift left of the second packed vector input and store the result into a vector register. ENC_VOP3P default 4 VDST FMT_NUM_PK2_U16 OPR_VGPR 32 SRC0 FMT_NUM_PK2_U16 OPR_SRC_SIMPLE 32 SRC1 FMT_NUM_PK2_U16 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_PK_LSHRREV_B16 Given a packed shift count in the first vector input, calculate the component-wise logical shift right of the second packed vector input and store the result into a vector register. ENC_VOP3P default 5 VDST FMT_NUM_PK2_U16 OPR_VGPR 32 SRC0 FMT_NUM_PK2_U16 OPR_SRC_SIMPLE 32 SRC1 FMT_NUM_PK2_U16 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_PK_ASHRREV_I16 Given a packed shift count in the first vector input, calculate the component-wise arithmetic shift right (preserving sign bit) of the second packed vector input and store the result into a vector register. ENC_VOP3P default 6 VDST FMT_NUM_PK2_I16 OPR_VGPR 32 SRC0 FMT_NUM_PK2_U16 OPR_SRC_SIMPLE 32 SRC1 FMT_NUM_PK2_I16 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_PK_MAX_I16 Select the component-wise maximum of two packed signed 16-bit integer inputs and store the selected values into a vector register. ENC_VOP3P default 7 VDST FMT_NUM_PK2_I16 OPR_VGPR 32 SRC0 FMT_NUM_PK2_I16 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_PK2_I16 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_PK_MIN_I16 Select the component-wise minimum of two packed signed 16-bit integer inputs and store the selected values into a vector register. ENC_VOP3P default 8 VDST FMT_NUM_PK2_I16 OPR_VGPR 32 SRC0 FMT_NUM_PK2_I16 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_PK2_I16 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_PK_MAD_U16 Multiply two packed unsigned 16-bit integer inputs component-wise, add a packed unsigned 16-bit integer value from a third input component-wise, and store the result into a vector register. ENC_VOP3P default 9 VDST FMT_NUM_PK2_U16 OPR_VGPR 32 SRC0 FMT_NUM_PK2_U16 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_PK2_U16 OPR_SRC_SIMPLE 32 SRC2 FMT_NUM_PK2_U16 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_PK_ADD_U16 Add two packed unsigned 16-bit integer inputs component-wise and store the result into a vector register. No carry-in or carry-out support. ENC_VOP3P default 10 VDST FMT_NUM_PK2_U16 OPR_VGPR 32 SRC0 FMT_NUM_PK2_U16 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_PK2_U16 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_PK_SUB_U16 Subtract the second packed unsigned 16-bit integer input from the first input component-wise and store the result into a vector register. No carry-in or carry-out support. ENC_VOP3P default 11 VDST FMT_NUM_PK2_U16 OPR_VGPR 32 SRC0 FMT_NUM_PK2_U16 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_PK2_U16 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_PK_MAX_U16 Select the component-wise maximum of two packed unsigned 16-bit integer inputs and store the selected values into a vector register. ENC_VOP3P default 12 VDST FMT_NUM_PK2_U16 OPR_VGPR 32 SRC0 FMT_NUM_PK2_U16 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_PK2_U16 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_PK_MIN_U16 Select the component-wise minimum of two packed unsigned 16-bit integer inputs and store the selected values into a vector register. ENC_VOP3P default 13 VDST FMT_NUM_PK2_U16 OPR_VGPR 32 SRC0 FMT_NUM_PK2_U16 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_PK2_U16 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_PK_FMA_F16 Multiply two packed half-precision float inputs component-wise and add a third input component-wise using fused multiply add, and store the result into a vector register. ENC_VOP3P default 14 VDST FMT_NUM_PK2_F16 OPR_VGPR 32 SRC0 FMT_NUM_PK2_F16 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_PK2_F16 OPR_SRC_SIMPLE 32 SRC2 FMT_NUM_PK2_F16 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_PK_ADD_F16 Add two packed half-precision float inputs component-wise and store the result into a vector register. No carry-in or carry-out support. ENC_VOP3P default 15 VDST FMT_NUM_PK2_F16 OPR_VGPR 32 SRC0 FMT_NUM_PK2_F16 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_PK2_F16 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_PK_MUL_F16 Multiply two packed half-precision float inputs component-wise and store the result into a vector register. ENC_VOP3P default 16 VDST FMT_NUM_PK2_F16 OPR_VGPR 32 SRC0 FMT_NUM_PK2_F16 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_PK2_F16 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_PK_MIN_F16 Select the component-wise minimum of two packed half-precision float inputs and store the selected values into a vector register. ENC_VOP3P default 17 VDST FMT_NUM_PK2_F16 OPR_VGPR 32 SRC0 FMT_NUM_PK2_F16 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_PK2_F16 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_PK_MAX_F16 Select the component-wise maximum of two packed half-precision float inputs and store the selected values into a vector register. ENC_VOP3P default 18 VDST FMT_NUM_PK2_F16 OPR_VGPR 32 SRC0 FMT_NUM_PK2_F16 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_PK2_F16 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_DOT2_F32_BF16 Calculate the dot product of BF16 float 2-vectors from the first and second inputs, convert the product to single-precision float format, add the third input and store the result into a vector register. ENC_VOP3P default 26 VDST FMT_NUM_F32 OPR_VGPR 32 SRC0 FMT_NUM_PK2_BF16 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_PK2_BF16 OPR_SRC_SIMPLE 32 SRC2 FMT_NUM_F32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_PK_MINIMUM3_F16 Select the component-wise IEEE minimum() of three half-precision float inputs and store the result into a vector register. ENC_VOP3P default 27 VDST FMT_NUM_PK2_F16 OPR_VGPR 32 SRC0 FMT_NUM_PK2_F16 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_PK2_F16 OPR_SRC_SIMPLE 32 SRC2 FMT_NUM_PK2_F16 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_PK_MAXIMUM3_F16 Select the component-wise IEEE maximum() of three half-precision float inputs and store the result into a vector register. ENC_VOP3P default 28 VDST FMT_NUM_PK2_F16 OPR_VGPR 32 SRC0 FMT_NUM_PK2_F16 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_PK2_F16 OPR_SRC_SIMPLE 32 SRC2 FMT_NUM_PK2_F16 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_MAD_MIX_F32 Multiply two inputs and add a third input where the inputs are a mix of half-precision float and single-precision float values. Store the result into a vector register. ENC_VOP3P default 32 VDST FMT_NUM_F32 OPR_VGPR 32 SRC0 FMT_NUM_B32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_B32 OPR_SRC_SIMPLE 32 SRC2 FMT_NUM_B32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_MAD_MIXLO_F16 Multiply two inputs and add a third input where the inputs are a mix of half-precision float and single-precision float values. Convert the result to a half-precision float. Store the result into the low bits of a vector register. ENC_VOP3P default 33 VDST FMT_NUM_F16 OPR_VGPR 16 SRC0 FMT_NUM_B32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_B32 OPR_SRC_SIMPLE 32 SRC2 FMT_NUM_B32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_MAD_MIXHI_F16 Multiply two inputs and add a third input where the inputs are a mix of half-precision float and single-precision float values. Convert the result to a half-precision float. Store the result into the high bits of a vector register. ENC_VOP3P default 34 VDST FMT_NUM_F16 OPR_VGPR 16 SRC0 FMT_NUM_B32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_B32 OPR_SRC_SIMPLE 32 SRC2 FMT_NUM_B32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_DOT2_F32_F16 Compute the dot product of two packed 2-D half-precision float inputs in the single-precision float domain, add a single-precision float value from the third input and store the result into a vector register. ENC_VOP3P default 35 VDST FMT_NUM_F32 OPR_VGPR 32 SRC0 FMT_NUM_PK2_F16 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_PK2_F16 OPR_SRC_SIMPLE 32 SRC2 FMT_NUM_F32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_DOT2_I32_I16 Compute the dot product of two packed 2-D signed 16-bit integer inputs in the signed 32-bit integer domain, add a signed 32-bit integer value from the third input and store the result into a vector register. ENC_VOP3P default 38 VDST FMT_NUM_I32 OPR_VGPR 32 SRC0 FMT_NUM_PK2_I16 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_PK2_I16 OPR_SRC_SIMPLE 32 SRC2 FMT_NUM_I32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_DOT2_U32_U16 Compute the dot product of two packed 2-D unsigned 16-bit integer inputs in the unsigned 32-bit integer domain, add an unsigned 32-bit integer value from the third input and store the result into a vector register. ENC_VOP3P default 39 VDST FMT_NUM_U32 OPR_VGPR 32 SRC0 FMT_NUM_PK2_U16 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_PK2_U16 OPR_SRC_SIMPLE 32 SRC2 FMT_NUM_U32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_DOT4_I32_I8 Compute the dot product of two packed 4-D signed 8-bit integer inputs in the signed 32-bit integer domain, add a signed 32-bit integer value from the third input and store the result into a vector register. ENC_VOP3P default 40 VDST FMT_NUM_I32 OPR_VGPR 32 SRC0 FMT_NUM_PK4_I8 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_PK4_I8 OPR_SRC_SIMPLE 32 SRC2 FMT_NUM_I32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_DOT4_U32_U8 Compute the dot product of two packed 4-D unsigned 8-bit integer inputs in the unsigned 32-bit integer domain, add an unsigned 32-bit integer value from the third input and store the result into a vector register. ENC_VOP3P default 41 VDST FMT_NUM_U32 OPR_VGPR 32 SRC0 FMT_NUM_PK4_U8 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_PK4_U8 OPR_SRC_SIMPLE 32 SRC2 FMT_NUM_U32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_DOT8_I32_I4 Compute the dot product of two packed 8-D signed 4-bit integer inputs in the signed 32-bit integer domain, add a signed 32-bit integer value from the third input and store the result into a vector register. ENC_VOP3P default 42 VDST FMT_NUM_I32 OPR_VGPR 32 SRC0 FMT_NUM_PK8_I4 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_PK8_I4 OPR_SRC_SIMPLE 32 SRC2 FMT_NUM_I32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_DOT8_U32_U4 Compute the dot product of two packed 8-D unsigned 4-bit integer inputs in the unsigned 32-bit integer domain, add an unsigned 32-bit integer value from the third input and store the result into a vector register. ENC_VOP3P default 43 VDST FMT_NUM_U32 OPR_VGPR 32 SRC0 FMT_NUM_PK8_U4 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_PK8_U4 OPR_SRC_SIMPLE 32 SRC2 FMT_NUM_U32 OPR_SRC_SIMPLE 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_MFMA_F32_16X16X128_F8F6F4 Multiply the 16x128 matrix in the first input by the 128x16 matrix in the second input and add the 16x16 matrix in the third input using fused multiply add. Store the resulting matrix into vector registers. VOP3P_MFMA default 45 VDST FMT_NUM_PK4_F32 OPR_VGPR_OR_ACCVGPR 128 SRC0 FMT_NUM_PK8_B32 OPR_SRC_VGPR_OR_ACCVGPR 256 SRC1 FMT_NUM_PK8_B32 OPR_SRC_VGPR_OR_ACCVGPR 256 SRC2 FMT_NUM_PK4_F32 OPR_SRC_VGPR_OR_ACCVGPR_OR_CONST 128 VALU MFMA FALSE FALSE FALSE FALSE FALSE V_MFMA_F32_32X32X64_F8F6F4 Multiply the 32x64 matrix in the first input by the 64x32 matrix in the second input and add the 32x32 matrix in the third input using fused multiply add. Store the resulting matrix into vector registers. VOP3P_MFMA default 46 VDST FMT_NUM_PK16_F32 OPR_VGPR_OR_ACCVGPR 512 SRC0 FMT_NUM_PK8_B32 OPR_SRC_VGPR_OR_ACCVGPR 256 SRC1 FMT_NUM_PK8_B32 OPR_SRC_VGPR_OR_ACCVGPR 256 SRC2 FMT_NUM_PK16_F32 OPR_SRC_VGPR_OR_ACCVGPR_OR_CONST 512 VALU MFMA FALSE FALSE FALSE FALSE FALSE V_PK_FMA_F32 Multiply two packed single-precision float inputs component-wise and add a third input component-wise using fused multiply add, and store the result into a vector register. ENC_VOP3P default 48 VDST FMT_NUM_PK2_F32 OPR_VGPR 64 SRC0 FMT_NUM_PK2_F32 OPR_SRC_NOLIT 64 SRC1 FMT_NUM_PK2_F32 OPR_SRC_SIMPLE 64 SRC2 FMT_NUM_PK2_F32 OPR_SRC_SIMPLE 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_PK_MUL_F32 Multiply two packed single-precision float inputs component-wise and store the result into a vector register. ENC_VOP3P default 49 VDST FMT_NUM_PK2_F32 OPR_VGPR 64 SRC0 FMT_NUM_PK2_F32 OPR_SRC_NOLIT 64 SRC1 FMT_NUM_PK2_F32 OPR_SRC_SIMPLE 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_PK_ADD_F32 Add two packed single-precision float inputs component-wise and store the result into a vector register. No carry-in or carry-out support. ENC_VOP3P default 50 VDST FMT_NUM_PK2_F32 OPR_VGPR 64 SRC0 FMT_NUM_PK2_F32 OPR_SRC_NOLIT 64 SRC1 FMT_NUM_PK2_F32 OPR_SRC_SIMPLE 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_PK_MOV_B32 Move data from two vector inputs into two vector registers. ENC_VOP3P default 51 VDST FMT_NUM_PK2_B32 OPR_VGPR 64 SRC0 FMT_NUM_PK2_B32 OPR_SRC_NOLIT 64 SRC1 FMT_NUM_PK2_B32 OPR_SRC_SIMPLE 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_MFMA_F32_16X16X32_BF16 Multiply the 16x32 matrix in the first input by the 32x16 matrix in the second input and add the 16x16 matrix in the third input using fused multiply add. Store the resulting matrix into vector registers. VOP3P_MFMA default 53 VDST FMT_NUM_PK4_F32 OPR_VGPR_OR_ACCVGPR 128 SRC0 FMT_NUM_PK8_BF16 OPR_SRC_VGPR_OR_ACCVGPR 128 SRC1 FMT_NUM_PK8_BF16 OPR_SRC_VGPR_OR_ACCVGPR 128 SRC2 FMT_NUM_PK4_F32 OPR_SRC_VGPR_OR_ACCVGPR_OR_CONST 128 VALU MFMA FALSE FALSE FALSE FALSE FALSE V_MFMA_I32_16X16X64_I8 Multiply the 16x64 matrix in the first input by the 64x16 matrix in the second input and add the 16x16 matrix in the third input using fused multiply add. Store the resulting matrix into vector registers. VOP3P_MFMA default 54 VDST FMT_NUM_PK4_I32 OPR_VGPR_OR_ACCVGPR 128 SRC0 FMT_NUM_PK16_I8 OPR_SRC_VGPR_OR_ACCVGPR 128 SRC1 FMT_NUM_PK16_I8 OPR_SRC_VGPR_OR_ACCVGPR 128 SRC2 FMT_NUM_PK4_I32 OPR_SRC_VGPR_OR_ACCVGPR_OR_CONST 128 VALU MFMA FALSE FALSE FALSE FALSE FALSE V_MFMA_F32_32X32X16_BF16 Multiply the 32x16 matrix in the first input by the 16x32 matrix in the second input and add the 32x32 matrix in the third input using fused multiply add. Store the resulting matrix into vector registers. VOP3P_MFMA default 55 VDST FMT_NUM_PK16_F32 OPR_VGPR_OR_ACCVGPR 512 SRC0 FMT_NUM_PK8_BF16 OPR_SRC_VGPR_OR_ACCVGPR 128 SRC1 FMT_NUM_PK8_BF16 OPR_SRC_VGPR_OR_ACCVGPR 128 SRC2 FMT_NUM_PK16_F32 OPR_SRC_VGPR_OR_ACCVGPR_OR_CONST 512 VALU MFMA FALSE FALSE FALSE FALSE FALSE V_MFMA_I32_32X32X32_I8 Multiply the 32x32 matrix in the first input by the 32x32 matrix in the second input and add the 32x32 matrix in the third input using fused multiply add. Store the resulting matrix into vector registers. VOP3P_MFMA default 56 VDST FMT_NUM_PK16_I32 OPR_VGPR_OR_ACCVGPR 512 SRC0 FMT_NUM_PK16_I8 OPR_SRC_VGPR_OR_ACCVGPR 128 SRC1 FMT_NUM_PK16_I8 OPR_SRC_VGPR_OR_ACCVGPR 128 SRC2 FMT_NUM_PK16_I32 OPR_SRC_VGPR_OR_ACCVGPR_OR_CONST 512 VALU MFMA FALSE FALSE FALSE FALSE FALSE V_SMFMAC_F32_16X16X64_BF16 Multiply the 16x64 sparse matrix in the first input by the 64x16 matrix in the second input and accumulate the result into the 16x16 matrix stored in the destination registers using fused multiply add. Sparse indexes for the first matrix are given in the third input. VOP3P_MFMA default 57 VDST FMT_NUM_PK4_F32 OPR_VGPR_OR_ACCVGPR 128 SRC0 FMT_NUM_PK8_BF16 OPR_SRC_VGPR_OR_ACCVGPR 128 SRC1 FMT_NUM_PK16_BF16 OPR_SRC_VGPR_OR_ACCVGPR 256 SRC2 FMT_NUM_B32 OPR_SRC_VGPR 32 VALU MFMA FALSE FALSE FALSE FALSE FALSE V_SMFMAC_I32_16X16X128_I8 Multiply the 16x128 sparse matrix in the first input by the 128x16 matrix in the second input and accumulate the result into the 16x16 matrix stored in the destination registers using fused multiply add. Sparse indexes for the first matrix are given in the third input. VOP3P_MFMA default 58 VDST FMT_NUM_PK4_I32 OPR_VGPR_OR_ACCVGPR 128 SRC0 FMT_NUM_PK16_I8 OPR_SRC_VGPR_OR_ACCVGPR 128 SRC1 FMT_NUM_PK32_I8 OPR_SRC_VGPR_OR_ACCVGPR 256 SRC2 FMT_NUM_B32 OPR_SRC_VGPR 32 VALU MFMA FALSE FALSE FALSE FALSE FALSE V_SMFMAC_F32_16X16X128_BF8_BF8 Multiply the 16x128 sparse matrix in the first input by the 128x16 matrix in the second input and accumulate the result into the 16x16 matrix stored in the destination registers using fused multiply add. Sparse indexes for the first matrix are given in the third input. VOP3P_MFMA default 59 VDST FMT_NUM_PK4_F32 OPR_VGPR_OR_ACCVGPR 128 SRC0 FMT_NUM_PK16_BF8 OPR_SRC_VGPR_OR_ACCVGPR 128 SRC1 FMT_NUM_PK32_BF8 OPR_SRC_VGPR_OR_ACCVGPR 256 SRC2 FMT_NUM_B32 OPR_SRC_VGPR 32 VALU MFMA FALSE FALSE FALSE FALSE FALSE V_SMFMAC_F32_16X16X128_BF8_FP8 Multiply the 16x128 sparse matrix in the first input by the 128x16 matrix in the second input and accumulate the result into the 16x16 matrix stored in the destination registers using fused multiply add. Sparse indexes for the first matrix are given in the third input. VOP3P_MFMA default 60 VDST FMT_NUM_PK4_F32 OPR_VGPR_OR_ACCVGPR 128 SRC0 FMT_NUM_PK16_BF8 OPR_SRC_VGPR_OR_ACCVGPR 128 SRC1 FMT_NUM_PK32_FP8 OPR_SRC_VGPR_OR_ACCVGPR 256 SRC2 FMT_NUM_B32 OPR_SRC_VGPR 32 VALU MFMA FALSE FALSE FALSE FALSE FALSE V_SMFMAC_F32_16X16X128_FP8_BF8 Multiply the 16x128 sparse matrix in the first input by the 128x16 matrix in the second input and accumulate the result into the 16x16 matrix stored in the destination registers using fused multiply add. Sparse indexes for the first matrix are given in the third input. VOP3P_MFMA default 61 VDST FMT_NUM_PK4_F32 OPR_VGPR_OR_ACCVGPR 128 SRC0 FMT_NUM_PK16_FP8 OPR_SRC_VGPR_OR_ACCVGPR 128 SRC1 FMT_NUM_PK32_BF8 OPR_SRC_VGPR_OR_ACCVGPR 256 SRC2 FMT_NUM_B32 OPR_SRC_VGPR 32 VALU MFMA FALSE FALSE FALSE FALSE FALSE V_MFMA_F32_32X32X1_2B_F32 V_MFMA_F32_32X32X1F32 Multiply the 32x1 matrix in the first input by the 1x32 matrix in the second input and add the 32x32 matrix in the third input using fused multiply add. Store the resulting matrix into vector registers. VOP3P_MFMA default 64 VDST FMT_NUM_PK32_F32 OPR_VGPR_OR_ACCVGPR 1024 SRC0 FMT_NUM_F32 OPR_SRC_VGPR_OR_ACCVGPR 32 SRC1 FMT_NUM_F32 OPR_SRC_VGPR_OR_ACCVGPR 32 SRC2 FMT_NUM_PK32_F32 OPR_SRC_VGPR_OR_ACCVGPR_OR_CONST 1024 VALU MFMA FALSE FALSE FALSE FALSE FALSE V_MFMA_F32_16X16X1_4B_F32 V_MFMA_F32_16X16X1F32 Multiply the 16x1 matrix in the first input by the 1x16 matrix in the second input and add the 16x16 matrix in the third input using fused multiply add. Store the resulting matrix into vector registers. VOP3P_MFMA default 65 VDST FMT_NUM_PK16_F32 OPR_VGPR_OR_ACCVGPR 512 SRC0 FMT_NUM_F32 OPR_SRC_VGPR_OR_ACCVGPR 32 SRC1 FMT_NUM_F32 OPR_SRC_VGPR_OR_ACCVGPR 32 SRC2 FMT_NUM_PK16_F32 OPR_SRC_VGPR_OR_ACCVGPR_OR_CONST 512 VALU MFMA FALSE FALSE FALSE FALSE FALSE V_MFMA_F32_4X4X1_16B_F32 V_MFMA_F32_4X4X1F32 Multiply the 4x1 matrix in the first input by the 1x4 matrix in the second input and add the 4x4 matrix in the third input using fused multiply add. Store the resulting matrix into vector registers. VOP3P_MFMA default 66 VDST FMT_NUM_PK4_F32 OPR_VGPR_OR_ACCVGPR 128 SRC0 FMT_NUM_F32 OPR_SRC_VGPR_OR_ACCVGPR 32 SRC1 FMT_NUM_F32 OPR_SRC_VGPR_OR_ACCVGPR 32 SRC2 FMT_NUM_PK4_F32 OPR_SRC_VGPR_OR_ACCVGPR_OR_CONST 128 VALU MFMA FALSE FALSE FALSE FALSE FALSE V_SMFMAC_F32_16X16X128_FP8_FP8 Multiply the 16x128 sparse matrix in the first input by the 128x16 matrix in the second input and accumulate the result into the 16x16 matrix stored in the destination registers using fused multiply add. Sparse indexes for the first matrix are given in the third input. VOP3P_MFMA default 67 VDST FMT_NUM_PK4_F32 OPR_VGPR_OR_ACCVGPR 128 SRC0 FMT_NUM_PK16_FP8 OPR_SRC_VGPR_OR_ACCVGPR 128 SRC1 FMT_NUM_PK32_FP8 OPR_SRC_VGPR_OR_ACCVGPR 256 SRC2 FMT_NUM_B32 OPR_SRC_VGPR 32 VALU MFMA FALSE FALSE FALSE FALSE FALSE V_MFMA_F32_32X32X2_F32 V_MFMA_F32_32X32X2F32 Multiply the 32x2 matrix in the first input by the 2x32 matrix in the second input and add the 32x32 matrix in the third input using fused multiply add. Store the resulting matrix into vector registers. VOP3P_MFMA default 68 VDST FMT_NUM_PK16_F32 OPR_VGPR_OR_ACCVGPR 512 SRC0 FMT_NUM_F32 OPR_SRC_VGPR_OR_ACCVGPR 32 SRC1 FMT_NUM_F32 OPR_SRC_VGPR_OR_ACCVGPR 32 SRC2 FMT_NUM_PK16_F32 OPR_SRC_VGPR_OR_ACCVGPR_OR_CONST 512 VALU MFMA FALSE FALSE FALSE FALSE FALSE V_MFMA_F32_16X16X4_F32 V_MFMA_F32_16X16X4F32 Multiply the 16x4 matrix in the first input by the 4x16 matrix in the second input and add the 16x16 matrix in the third input using fused multiply add. Store the resulting matrix into vector registers. VOP3P_MFMA default 69 VDST FMT_NUM_PK4_F32 OPR_VGPR_OR_ACCVGPR 128 SRC0 FMT_NUM_F32 OPR_SRC_VGPR_OR_ACCVGPR 32 SRC1 FMT_NUM_F32 OPR_SRC_VGPR_OR_ACCVGPR 32 SRC2 FMT_NUM_PK4_F32 OPR_SRC_VGPR_OR_ACCVGPR_OR_CONST 128 VALU MFMA FALSE FALSE FALSE FALSE FALSE V_SMFMAC_F32_32X32X32_BF16 Multiply the 32x32 sparse matrix in the first input by the 32x32 matrix in the second input and accumulate the result into the 32x32 matrix stored in the destination registers using fused multiply add. Sparse indexes for the first matrix are given in the third input. VOP3P_MFMA default 70 VDST FMT_NUM_PK16_F32 OPR_VGPR_OR_ACCVGPR 512 SRC0 FMT_NUM_PK8_BF16 OPR_SRC_VGPR_OR_ACCVGPR 128 SRC1 FMT_NUM_PK16_BF16 OPR_SRC_VGPR_OR_ACCVGPR 256 SRC2 FMT_NUM_B32 OPR_SRC_VGPR 32 VALU MFMA FALSE FALSE FALSE FALSE FALSE V_SMFMAC_I32_32X32X64_I8 Multiply the 32x64 sparse matrix in the first input by the 64x32 matrix in the second input and accumulate the result into the 32x32 matrix stored in the destination registers using fused multiply add. Sparse indexes for the first matrix are given in the third input. VOP3P_MFMA default 71 VDST FMT_NUM_PK16_I32 OPR_VGPR_OR_ACCVGPR 512 SRC0 FMT_NUM_PK16_I8 OPR_SRC_VGPR_OR_ACCVGPR 128 SRC1 FMT_NUM_PK32_I8 OPR_SRC_VGPR_OR_ACCVGPR 256 SRC2 FMT_NUM_B32 OPR_SRC_VGPR 32 VALU MFMA FALSE FALSE FALSE FALSE FALSE V_MFMA_F32_32X32X4_2B_F16 V_MFMA_F32_32X32X4F16 Multiply the 32x4 matrix in the first input by the 4x32 matrix in the second input and add the 32x32 matrix in the third input using fused multiply add. Store the resulting matrix into vector registers. VOP3P_MFMA default 72 VDST FMT_NUM_PK32_F32 OPR_VGPR_OR_ACCVGPR 1024 SRC0 FMT_NUM_PK4_F16 OPR_SRC_VGPR_OR_ACCVGPR 64 SRC1 FMT_NUM_PK4_F16 OPR_SRC_VGPR_OR_ACCVGPR 64 SRC2 FMT_NUM_PK32_F32 OPR_SRC_VGPR_OR_ACCVGPR_OR_CONST 1024 VALU MFMA FALSE FALSE FALSE FALSE FALSE V_MFMA_F32_16X16X4_4B_F16 V_MFMA_F32_16X16X4F16 Multiply the 16x4 matrix in the first input by the 4x16 matrix in the second input and add the 16x16 matrix in the third input using fused multiply add. Store the resulting matrix into vector registers. VOP3P_MFMA default 73 VDST FMT_NUM_PK16_F32 OPR_VGPR_OR_ACCVGPR 512 SRC0 FMT_NUM_PK4_F16 OPR_SRC_VGPR_OR_ACCVGPR 64 SRC1 FMT_NUM_PK4_F16 OPR_SRC_VGPR_OR_ACCVGPR 64 SRC2 FMT_NUM_PK16_F32 OPR_SRC_VGPR_OR_ACCVGPR_OR_CONST 512 VALU MFMA FALSE FALSE FALSE FALSE FALSE V_MFMA_F32_4X4X4_16B_F16 V_MFMA_F32_4X4X4F16 Multiply the 4x4 matrix in the first input by the 4x4 matrix in the second input and add the 4x4 matrix in the third input using fused multiply add. Store the resulting matrix into vector registers. VOP3P_MFMA default 74 VDST FMT_NUM_PK4_F32 OPR_VGPR_OR_ACCVGPR 128 SRC0 FMT_NUM_PK4_F16 OPR_SRC_VGPR_OR_ACCVGPR 64 SRC1 FMT_NUM_PK4_F16 OPR_SRC_VGPR_OR_ACCVGPR 64 SRC2 FMT_NUM_PK4_F32 OPR_SRC_VGPR_OR_ACCVGPR_OR_CONST 128 VALU MFMA FALSE FALSE FALSE FALSE FALSE V_SMFMAC_F32_32X32X64_BF8_BF8 Multiply the 32x64 sparse matrix in the first input by the 64x32 matrix in the second input and accumulate the result into the 32x32 matrix stored in the destination registers using fused multiply add. Sparse indexes for the first matrix are given in the third input. VOP3P_MFMA default 75 VDST FMT_NUM_PK16_F32 OPR_VGPR_OR_ACCVGPR 512 SRC0 FMT_NUM_PK16_BF8 OPR_SRC_VGPR_OR_ACCVGPR 128 SRC1 FMT_NUM_PK32_BF8 OPR_SRC_VGPR_OR_ACCVGPR 256 SRC2 FMT_NUM_B32 OPR_SRC_VGPR 32 VALU MFMA FALSE FALSE FALSE FALSE FALSE V_MFMA_F32_32X32X8_F16 V_MFMA_F32_32X32X8F16 Multiply the 32x8 matrix in the first input by the 8x32 matrix in the second input and add the 32x32 matrix in the third input using fused multiply add. Store the resulting matrix into vector registers. VOP3P_MFMA default 76 VDST FMT_NUM_PK16_F32 OPR_VGPR_OR_ACCVGPR 512 SRC0 FMT_NUM_PK4_F16 OPR_SRC_VGPR_OR_ACCVGPR 64 SRC1 FMT_NUM_PK4_F16 OPR_SRC_VGPR_OR_ACCVGPR 64 SRC2 FMT_NUM_PK16_F32 OPR_SRC_VGPR_OR_ACCVGPR_OR_CONST 512 VALU MFMA FALSE FALSE FALSE FALSE FALSE V_MFMA_F32_16X16X16_F16 V_MFMA_F32_16X16X16F16 Multiply the 16x16 matrix in the first input by the 16x16 matrix in the second input and add the 16x16 matrix in the third input using fused multiply add. Store the resulting matrix into vector registers. VOP3P_MFMA default 77 VDST FMT_NUM_PK4_F32 OPR_VGPR_OR_ACCVGPR 128 SRC0 FMT_NUM_PK4_F16 OPR_SRC_VGPR_OR_ACCVGPR 64 SRC1 FMT_NUM_PK4_F16 OPR_SRC_VGPR_OR_ACCVGPR 64 SRC2 FMT_NUM_PK4_F32 OPR_SRC_VGPR_OR_ACCVGPR_OR_CONST 128 VALU MFMA FALSE FALSE FALSE FALSE FALSE V_SMFMAC_F32_32X32X64_BF8_FP8 Multiply the 32x64 sparse matrix in the first input by the 64x32 matrix in the second input and accumulate the result into the 32x32 matrix stored in the destination registers using fused multiply add. Sparse indexes for the first matrix are given in the third input. VOP3P_MFMA default 78 VDST FMT_NUM_PK16_F32 OPR_VGPR_OR_ACCVGPR 512 SRC0 FMT_NUM_PK16_BF8 OPR_SRC_VGPR_OR_ACCVGPR 128 SRC1 FMT_NUM_PK32_FP8 OPR_SRC_VGPR_OR_ACCVGPR 256 SRC2 FMT_NUM_B32 OPR_SRC_VGPR 32 VALU MFMA FALSE FALSE FALSE FALSE FALSE V_SMFMAC_F32_32X32X64_FP8_BF8 Multiply the 32x64 sparse matrix in the first input by the 64x32 matrix in the second input and accumulate the result into the 32x32 matrix stored in the destination registers using fused multiply add. Sparse indexes for the first matrix are given in the third input. VOP3P_MFMA default 79 VDST FMT_NUM_PK16_F32 OPR_VGPR_OR_ACCVGPR 512 SRC0 FMT_NUM_PK16_FP8 OPR_SRC_VGPR_OR_ACCVGPR 128 SRC1 FMT_NUM_PK32_BF8 OPR_SRC_VGPR_OR_ACCVGPR 256 SRC2 FMT_NUM_B32 OPR_SRC_VGPR 32 VALU MFMA FALSE FALSE FALSE FALSE FALSE V_MFMA_I32_32X32X4_2B_I8 V_MFMA_I32_32X32X4I8 Multiply the 32x4 matrix in the first input by the 4x32 matrix in the second input and add the 32x32 matrix in the third input using fused multiply add. Store the resulting matrix into vector registers. VOP3P_MFMA default 80 VDST FMT_NUM_PK32_I32 OPR_VGPR_OR_ACCVGPR 1024 SRC0 FMT_NUM_PK4_I8 OPR_SRC_VGPR_OR_ACCVGPR 32 SRC1 FMT_NUM_PK4_I8 OPR_SRC_VGPR_OR_ACCVGPR 32 SRC2 FMT_NUM_PK32_I32 OPR_SRC_VGPR_OR_ACCVGPR_OR_CONST 1024 VALU MFMA FALSE FALSE FALSE FALSE FALSE V_MFMA_I32_16X16X4_4B_I8 V_MFMA_I32_16X16X4I8 Multiply the 16x4 matrix in the first input by the 4x16 matrix in the second input and add the 16x16 matrix in the third input using fused multiply add. Store the resulting matrix into vector registers. VOP3P_MFMA default 81 VDST FMT_NUM_PK16_I32 OPR_VGPR_OR_ACCVGPR 512 SRC0 FMT_NUM_PK4_I8 OPR_SRC_VGPR_OR_ACCVGPR 32 SRC1 FMT_NUM_PK4_I8 OPR_SRC_VGPR_OR_ACCVGPR 32 SRC2 FMT_NUM_PK16_I32 OPR_SRC_VGPR_OR_ACCVGPR_OR_CONST 512 VALU MFMA FALSE FALSE FALSE FALSE FALSE V_MFMA_I32_4X4X4_16B_I8 V_MFMA_I32_4X4X4I8 Multiply the 4x4 matrix in the first input by the 4x4 matrix in the second input and add the 4x4 matrix in the third input using fused multiply add. Store the resulting matrix into vector registers. VOP3P_MFMA default 82 VDST FMT_NUM_PK4_I32 OPR_VGPR_OR_ACCVGPR 128 SRC0 FMT_NUM_PK4_I8 OPR_SRC_VGPR_OR_ACCVGPR 32 SRC1 FMT_NUM_PK4_I8 OPR_SRC_VGPR_OR_ACCVGPR 32 SRC2 FMT_NUM_PK4_I32 OPR_SRC_VGPR_OR_ACCVGPR_OR_CONST 128 VALU MFMA FALSE FALSE FALSE FALSE FALSE V_SMFMAC_F32_32X32X64_FP8_FP8 Multiply the 32x64 sparse matrix in the first input by the 64x32 matrix in the second input and accumulate the result into the 32x32 matrix stored in the destination registers using fused multiply add. Sparse indexes for the first matrix are given in the third input. VOP3P_MFMA default 83 VDST FMT_NUM_PK16_F32 OPR_VGPR_OR_ACCVGPR 512 SRC0 FMT_NUM_PK16_FP8 OPR_SRC_VGPR_OR_ACCVGPR 128 SRC1 FMT_NUM_PK32_FP8 OPR_SRC_VGPR_OR_ACCVGPR 256 SRC2 FMT_NUM_B32 OPR_SRC_VGPR 32 VALU MFMA FALSE FALSE FALSE FALSE FALSE V_MFMA_F32_16X16X32_F16 Multiply the 16x32 matrix in the first input by the 32x16 matrix in the second input and add the 16x16 matrix in the third input using fused multiply add. Store the resulting matrix into vector registers. VOP3P_MFMA default 84 VDST FMT_NUM_PK4_F32 OPR_VGPR_OR_ACCVGPR 128 SRC0 FMT_NUM_PK8_F16 OPR_SRC_VGPR_OR_ACCVGPR 128 SRC1 FMT_NUM_PK8_F16 OPR_SRC_VGPR_OR_ACCVGPR 128 SRC2 FMT_NUM_PK4_F32 OPR_SRC_VGPR_OR_ACCVGPR_OR_CONST 128 VALU MFMA FALSE FALSE FALSE FALSE FALSE V_MFMA_F32_32X32X16_F16 Multiply the 32x16 matrix in the first input by the 16x32 matrix in the second input and add the 32x32 matrix in the third input using fused multiply add. Store the resulting matrix into vector registers. VOP3P_MFMA default 85 VDST FMT_NUM_PK16_F32 OPR_VGPR_OR_ACCVGPR 512 SRC0 FMT_NUM_PK8_F16 OPR_SRC_VGPR_OR_ACCVGPR 128 SRC1 FMT_NUM_PK8_F16 OPR_SRC_VGPR_OR_ACCVGPR 128 SRC2 FMT_NUM_PK16_F32 OPR_SRC_VGPR_OR_ACCVGPR_OR_CONST 512 VALU MFMA FALSE FALSE FALSE FALSE FALSE V_MFMA_I32_32X32X16_I8 V_MFMA_I32_32X32X16I8 Multiply the 32x16 matrix in the first input by the 16x32 matrix in the second input and add the 32x32 matrix in the third input using fused multiply add. Store the resulting matrix into vector registers. VOP3P_MFMA default 86 VDST FMT_NUM_PK16_I32 OPR_VGPR_OR_ACCVGPR 512 SRC0 FMT_NUM_PK8_I8 OPR_SRC_VGPR_OR_ACCVGPR 64 SRC1 FMT_NUM_PK8_I8 OPR_SRC_VGPR_OR_ACCVGPR 64 SRC2 FMT_NUM_PK16_I32 OPR_SRC_VGPR_OR_ACCVGPR_OR_CONST 512 VALU MFMA FALSE FALSE FALSE FALSE FALSE V_MFMA_I32_16X16X32_I8 V_MFMA_I32_16X16X32I8 Multiply the 16x32 matrix in the first input by the 32x16 matrix in the second input and add the 16x16 matrix in the third input using fused multiply add. Store the resulting matrix into vector registers. VOP3P_MFMA default 87 VDST FMT_NUM_PK4_I32 OPR_VGPR_OR_ACCVGPR 128 SRC0 FMT_NUM_PK8_I8 OPR_SRC_VGPR_OR_ACCVGPR 64 SRC1 FMT_NUM_PK8_I8 OPR_SRC_VGPR_OR_ACCVGPR 64 SRC2 FMT_NUM_PK4_I32 OPR_SRC_VGPR_OR_ACCVGPR_OR_CONST 128 VALU MFMA FALSE FALSE FALSE FALSE FALSE V_ACCVGPR_READ V_ACCVGPR_READ_B32 Move 32 bits of data from an accumulator vector register into an architectural vector register. ENC_VOP3P default 88 VDST FMT_NUM_B32 OPR_VGPR 32 SRC0 FMT_NUM_B32 OPR_SRC_ACCVGPR 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_ACCVGPR_WRITE V_ACCVGPR_WRITE_B32 Move 32 bits of data from an architectural vector register into an accumulator vector register. ENC_VOP3P default 89 VDST FMT_NUM_B32 OPR_ACCVGPR 32 SRC0 FMT_NUM_B32 OPR_SRC_NOLIT 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_SMFMAC_F32_16X16X64_F16 Multiply the 16x64 sparse matrix in the first input by the 64x16 matrix in the second input and accumulate the result into the 16x16 matrix stored in the destination registers using fused multiply add. Sparse indexes for the first matrix are given in the third input. VOP3P_MFMA default 90 VDST FMT_NUM_PK4_F32 OPR_VGPR_OR_ACCVGPR 128 SRC0 FMT_NUM_PK8_F16 OPR_SRC_VGPR_OR_ACCVGPR 128 SRC1 FMT_NUM_PK16_F16 OPR_SRC_VGPR_OR_ACCVGPR 256 SRC2 FMT_NUM_B32 OPR_SRC_VGPR 32 VALU MFMA FALSE FALSE FALSE FALSE FALSE V_SMFMAC_F32_32X32X32_F16 Multiply the 32x32 sparse matrix in the first input by the 32x32 matrix in the second input and accumulate the result into the 32x32 matrix stored in the destination registers using fused multiply add. Sparse indexes for the first matrix are given in the third input. VOP3P_MFMA default 91 VDST FMT_NUM_PK16_F32 OPR_VGPR_OR_ACCVGPR 512 SRC0 FMT_NUM_PK8_F16 OPR_SRC_VGPR_OR_ACCVGPR 128 SRC1 FMT_NUM_PK16_F16 OPR_SRC_VGPR_OR_ACCVGPR 256 SRC2 FMT_NUM_B32 OPR_SRC_VGPR 32 VALU MFMA FALSE FALSE FALSE FALSE FALSE V_MFMA_F32_32X32X4_2B_BF16 V_MFMA_F32_32X32X4BF16_1K Multiply the 32x4 matrix in the first input by the 4x32 matrix in the second input and add the 32x32 matrix in the third input using fused multiply add. Store the resulting matrix into vector registers. VOP3P_MFMA default 93 VDST FMT_NUM_PK32_F32 OPR_VGPR_OR_ACCVGPR 1024 SRC0 FMT_NUM_PK4_BF16 OPR_SRC_VGPR_OR_ACCVGPR 64 SRC1 FMT_NUM_PK4_BF16 OPR_SRC_VGPR_OR_ACCVGPR 64 SRC2 FMT_NUM_PK32_F32 OPR_SRC_VGPR_OR_ACCVGPR_OR_CONST 1024 VALU MFMA FALSE FALSE FALSE FALSE FALSE V_MFMA_F32_16X16X4_4B_BF16 V_MFMA_F32_16X16X4BF16 V_MFMA_F32_16X16X4BF16_1K Multiply the 16x4 matrix in the first input by the 4x16 matrix in the second input and add the 16x16 matrix in the third input using fused multiply add. Store the resulting matrix into vector registers. VOP3P_MFMA default 94 VDST FMT_NUM_PK16_F32 OPR_VGPR_OR_ACCVGPR 512 SRC0 FMT_NUM_PK4_BF16 OPR_SRC_VGPR_OR_ACCVGPR 64 SRC1 FMT_NUM_PK4_BF16 OPR_SRC_VGPR_OR_ACCVGPR 64 SRC2 FMT_NUM_PK16_F32 OPR_SRC_VGPR_OR_ACCVGPR_OR_CONST 512 VALU MFMA FALSE FALSE FALSE FALSE FALSE V_MFMA_F32_4X4X4_16B_BF16 V_MFMA_F32_4X4X4BF16 V_MFMA_F32_4X4X4BF16_1K Multiply the 4x4 matrix in the first input by the 4x4 matrix in the second input and add the 4x4 matrix in the third input using fused multiply add. Store the resulting matrix into vector registers. VOP3P_MFMA default 95 VDST FMT_NUM_PK4_F32 OPR_VGPR_OR_ACCVGPR 128 SRC0 FMT_NUM_PK4_BF16 OPR_SRC_VGPR_OR_ACCVGPR 64 SRC1 FMT_NUM_PK4_BF16 OPR_SRC_VGPR_OR_ACCVGPR 64 SRC2 FMT_NUM_PK4_F32 OPR_SRC_VGPR_OR_ACCVGPR_OR_CONST 128 VALU MFMA FALSE FALSE FALSE FALSE FALSE V_MFMA_F32_32X32X8_BF16 V_MFMA_F32_32X32X8BF16 V_MFMA_F32_32X32X8BF16_1K Multiply the 32x8 matrix in the first input by the 8x32 matrix in the second input and add the 32x32 matrix in the third input using fused multiply add. Store the resulting matrix into vector registers. VOP3P_MFMA default 96 VDST FMT_NUM_PK16_F32 OPR_VGPR_OR_ACCVGPR 512 SRC0 FMT_NUM_PK4_BF16 OPR_SRC_VGPR_OR_ACCVGPR 64 SRC1 FMT_NUM_PK4_BF16 OPR_SRC_VGPR_OR_ACCVGPR 64 SRC2 FMT_NUM_PK16_F32 OPR_SRC_VGPR_OR_ACCVGPR_OR_CONST 512 VALU MFMA FALSE FALSE FALSE FALSE FALSE V_MFMA_F32_16X16X16_BF16 V_MFMA_F32_16X16X16BF16 V_MFMA_F32_16X16X16BF16_1K Multiply the 16x16 matrix in the first input by the 16x16 matrix in the second input and add the 16x16 matrix in the third input using fused multiply add. Store the resulting matrix into vector registers. VOP3P_MFMA default 97 VDST FMT_NUM_PK4_F32 OPR_VGPR_OR_ACCVGPR 128 SRC0 FMT_NUM_PK4_BF16 OPR_SRC_VGPR_OR_ACCVGPR 64 SRC1 FMT_NUM_PK4_BF16 OPR_SRC_VGPR_OR_ACCVGPR 64 SRC2 FMT_NUM_PK4_F32 OPR_SRC_VGPR_OR_ACCVGPR_OR_CONST 128 VALU MFMA FALSE FALSE FALSE FALSE FALSE V_SMFMAC_F32_16X16X32_F16 V_SMFMAC_F32_16X16X32F16 Multiply the 16x32 sparse matrix in the first input by the 32x16 matrix in the second input and accumulate the result into the 16x16 matrix stored in the destination registers using fused multiply add. Sparse indexes for the first matrix are given in the third input. VOP3P_MFMA default 98 VDST FMT_NUM_PK4_F32 OPR_VGPR_OR_ACCVGPR 128 SRC0 FMT_NUM_PK4_F16 OPR_SRC_VGPR_OR_ACCVGPR 64 SRC1 FMT_NUM_PK8_F16 OPR_SRC_VGPR_OR_ACCVGPR 128 SRC2 FMT_NUM_B32 OPR_SRC_VGPR 32 VALU MFMA FALSE FALSE FALSE FALSE FALSE V_SMFMAC_F32_32X32X16_F16 V_SMFMAC_F32_32X32X16F16 Multiply the 32x16 sparse matrix in the first input by the 16x32 matrix in the second input and accumulate the result into the 32x32 matrix stored in the destination registers using fused multiply add. Sparse indexes for the first matrix are given in the third input. VOP3P_MFMA default 100 VDST FMT_NUM_PK16_F32 OPR_VGPR_OR_ACCVGPR 512 SRC0 FMT_NUM_PK4_F16 OPR_SRC_VGPR_OR_ACCVGPR 64 SRC1 FMT_NUM_PK8_F16 OPR_SRC_VGPR_OR_ACCVGPR 128 SRC2 FMT_NUM_B32 OPR_SRC_VGPR 32 VALU MFMA FALSE FALSE FALSE FALSE FALSE V_SMFMAC_F32_16X16X32_BF16 V_SMFMAC_F32_16X16X32BF16 Multiply the 16x32 sparse matrix in the first input by the 32x16 matrix in the second input and accumulate the result into the 16x16 matrix stored in the destination registers using fused multiply add. Sparse indexes for the first matrix are given in the third input. VOP3P_MFMA default 102 VDST FMT_NUM_PK4_F32 OPR_VGPR_OR_ACCVGPR 128 SRC0 FMT_NUM_PK4_BF16 OPR_SRC_VGPR_OR_ACCVGPR 64 SRC1 FMT_NUM_PK8_BF16 OPR_SRC_VGPR_OR_ACCVGPR 128 SRC2 FMT_NUM_B32 OPR_SRC_VGPR 32 VALU MFMA FALSE FALSE FALSE FALSE FALSE V_SMFMAC_F32_32X32X16_BF16 V_SMFMAC_F32_32X32X16BF16 Multiply the 32x16 sparse matrix in the first input by the 16x32 matrix in the second input and accumulate the result into the 32x32 matrix stored in the destination registers using fused multiply add. Sparse indexes for the first matrix are given in the third input. VOP3P_MFMA default 104 VDST FMT_NUM_PK16_F32 OPR_VGPR_OR_ACCVGPR 512 SRC0 FMT_NUM_PK4_BF16 OPR_SRC_VGPR_OR_ACCVGPR 64 SRC1 FMT_NUM_PK8_BF16 OPR_SRC_VGPR_OR_ACCVGPR 128 SRC2 FMT_NUM_B32 OPR_SRC_VGPR 32 VALU MFMA FALSE FALSE FALSE FALSE FALSE V_SMFMAC_I32_16X16X64_I8 V_SMFMAC_I32_16X16X64I8 Multiply the 16x64 sparse matrix in the first input by the 64x16 matrix in the second input and accumulate the result into the 16x16 matrix stored in the destination registers using fused multiply add. Sparse indexes for the first matrix are given in the third input. VOP3P_MFMA default 106 VDST FMT_NUM_PK4_I32 OPR_VGPR_OR_ACCVGPR 128 SRC0 FMT_NUM_PK8_I8 OPR_SRC_VGPR_OR_ACCVGPR 64 SRC1 FMT_NUM_PK16_I8 OPR_SRC_VGPR_OR_ACCVGPR 128 SRC2 FMT_NUM_B32 OPR_SRC_VGPR 32 VALU MFMA FALSE FALSE FALSE FALSE FALSE V_SMFMAC_I32_32X32X32_I8 V_SMFMAC_I32_32X32X32I8 Multiply the 32x32 sparse matrix in the first input by the 32x32 matrix in the second input and accumulate the result into the 32x32 matrix stored in the destination registers using fused multiply add. Sparse indexes for the first matrix are given in the third input. VOP3P_MFMA default 108 VDST FMT_NUM_PK16_I32 OPR_VGPR_OR_ACCVGPR 512 SRC0 FMT_NUM_PK8_I8 OPR_SRC_VGPR_OR_ACCVGPR 64 SRC1 FMT_NUM_PK16_I8 OPR_SRC_VGPR_OR_ACCVGPR 128 SRC2 FMT_NUM_B32 OPR_SRC_VGPR 32 VALU MFMA FALSE FALSE FALSE FALSE FALSE V_MFMA_F64_16X16X4_F64 V_MFMA_F64_16X16X4F64 Multiply the 16x4 matrix in the first input by the 4x16 matrix in the second input and add the 16x16 matrix in the third input using fused multiply add. Store the resulting matrix into vector registers. VOP3P_MFMA default 110 VDST FMT_NUM_PK4_F64 OPR_VGPR_OR_ACCVGPR 256 SRC0 FMT_NUM_F64 OPR_SRC_VGPR_OR_ACCVGPR 64 SRC1 FMT_NUM_F64 OPR_SRC_VGPR_OR_ACCVGPR 64 SRC2 FMT_NUM_PK4_F64 OPR_SRC_VGPR_OR_ACCVGPR_OR_CONST 256 VALU MFMA FALSE FALSE FALSE FALSE FALSE V_MFMA_F64_4X4X4_4B_F64 V_MFMA_F64_4X4X4F64 Multiply the 4x4 matrix in the first input by the 4x4 matrix in the second input and add the 4x4 matrix in the third input using fused multiply add. Store the resulting matrix into vector registers. VOP3P_MFMA default 111 VDST FMT_NUM_F64 OPR_VGPR_OR_ACCVGPR 64 SRC0 FMT_NUM_F64 OPR_SRC_VGPR_OR_ACCVGPR 64 SRC1 FMT_NUM_F64 OPR_SRC_VGPR_OR_ACCVGPR 64 SRC2 FMT_NUM_F64 OPR_SRC_VGPR_OR_ACCVGPR_OR_CONST 64 VALU MFMA FALSE FALSE FALSE FALSE FALSE V_MFMA_F32_16X16X32_BF8_BF8 V_MFMA_F32_16X16X32BF8BF8 Multiply the 16x32 matrix in the first input by the 32x16 matrix in the second input and add the 16x16 matrix in the third input using fused multiply add. Store the resulting matrix into vector registers. VOP3P_MFMA default 112 VDST FMT_NUM_PK4_F32 OPR_VGPR_OR_ACCVGPR 128 SRC0 FMT_NUM_PK8_BF8 OPR_SRC_VGPR_OR_ACCVGPR 64 SRC1 FMT_NUM_PK8_BF8 OPR_SRC_VGPR_OR_ACCVGPR 64 SRC2 FMT_NUM_PK4_F32 OPR_SRC_VGPR_OR_ACCVGPR_OR_CONST 128 VALU MFMA FALSE FALSE FALSE FALSE FALSE V_MFMA_F32_16X16X32_BF8_FP8 V_MFMA_F32_16X16X32BF8FP8 Multiply the 16x32 matrix in the first input by the 32x16 matrix in the second input and add the 16x16 matrix in the third input using fused multiply add. Store the resulting matrix into vector registers. VOP3P_MFMA default 113 VDST FMT_NUM_PK4_F32 OPR_VGPR_OR_ACCVGPR 128 SRC0 FMT_NUM_PK8_BF8 OPR_SRC_VGPR_OR_ACCVGPR 64 SRC1 FMT_NUM_PK8_FP8 OPR_SRC_VGPR_OR_ACCVGPR 64 SRC2 FMT_NUM_PK4_F32 OPR_SRC_VGPR_OR_ACCVGPR_OR_CONST 128 VALU MFMA FALSE FALSE FALSE FALSE FALSE V_MFMA_F32_16X16X32_FP8_BF8 V_MFMA_F32_16X16X32FP8BF8 Multiply the 16x32 matrix in the first input by the 32x16 matrix in the second input and add the 16x16 matrix in the third input using fused multiply add. Store the resulting matrix into vector registers. VOP3P_MFMA default 114 VDST FMT_NUM_PK4_F32 OPR_VGPR_OR_ACCVGPR 128 SRC0 FMT_NUM_PK8_FP8 OPR_SRC_VGPR_OR_ACCVGPR 64 SRC1 FMT_NUM_PK8_BF8 OPR_SRC_VGPR_OR_ACCVGPR 64 SRC2 FMT_NUM_PK4_F32 OPR_SRC_VGPR_OR_ACCVGPR_OR_CONST 128 VALU MFMA FALSE FALSE FALSE FALSE FALSE V_MFMA_F32_16X16X32_FP8_FP8 V_MFMA_F32_16X16X32FP8FP8 Multiply the 16x32 matrix in the first input by the 32x16 matrix in the second input and add the 16x16 matrix in the third input using fused multiply add. Store the resulting matrix into vector registers. VOP3P_MFMA default 115 VDST FMT_NUM_PK4_F32 OPR_VGPR_OR_ACCVGPR 128 SRC0 FMT_NUM_PK8_FP8 OPR_SRC_VGPR_OR_ACCVGPR 64 SRC1 FMT_NUM_PK8_FP8 OPR_SRC_VGPR_OR_ACCVGPR 64 SRC2 FMT_NUM_PK4_F32 OPR_SRC_VGPR_OR_ACCVGPR_OR_CONST 128 VALU MFMA FALSE FALSE FALSE FALSE FALSE V_MFMA_F32_32X32X16_BF8_BF8 V_MFMA_F32_32X32X16BF8BF8 Multiply the 32x16 matrix in the first input by the 16x32 matrix in the second input and add the 32x32 matrix in the third input using fused multiply add. Store the resulting matrix into vector registers. VOP3P_MFMA default 116 VDST FMT_NUM_PK16_F32 OPR_VGPR_OR_ACCVGPR 512 SRC0 FMT_NUM_PK8_BF8 OPR_SRC_VGPR_OR_ACCVGPR 64 SRC1 FMT_NUM_PK8_BF8 OPR_SRC_VGPR_OR_ACCVGPR 64 SRC2 FMT_NUM_PK16_F32 OPR_SRC_VGPR_OR_ACCVGPR_OR_CONST 512 VALU MFMA FALSE FALSE FALSE FALSE FALSE V_MFMA_F32_32X32X16_BF8_FP8 V_MFMA_F32_32X32X16BF8FP8 Multiply the 32x16 matrix in the first input by the 16x32 matrix in the second input and add the 32x32 matrix in the third input using fused multiply add. Store the resulting matrix into vector registers. VOP3P_MFMA default 117 VDST FMT_NUM_PK16_F32 OPR_VGPR_OR_ACCVGPR 512 SRC0 FMT_NUM_PK8_BF8 OPR_SRC_VGPR_OR_ACCVGPR 64 SRC1 FMT_NUM_PK8_FP8 OPR_SRC_VGPR_OR_ACCVGPR 64 SRC2 FMT_NUM_PK16_F32 OPR_SRC_VGPR_OR_ACCVGPR_OR_CONST 512 VALU MFMA FALSE FALSE FALSE FALSE FALSE V_MFMA_F32_32X32X16_FP8_BF8 V_MFMA_F32_32X32X16FP8BF8 Multiply the 32x16 matrix in the first input by the 16x32 matrix in the second input and add the 32x32 matrix in the third input using fused multiply add. Store the resulting matrix into vector registers. VOP3P_MFMA default 118 VDST FMT_NUM_PK16_F32 OPR_VGPR_OR_ACCVGPR 512 SRC0 FMT_NUM_PK8_FP8 OPR_SRC_VGPR_OR_ACCVGPR 64 SRC1 FMT_NUM_PK8_BF8 OPR_SRC_VGPR_OR_ACCVGPR 64 SRC2 FMT_NUM_PK16_F32 OPR_SRC_VGPR_OR_ACCVGPR_OR_CONST 512 VALU MFMA FALSE FALSE FALSE FALSE FALSE V_MFMA_F32_32X32X16_FP8_FP8 V_MFMA_F32_32X32X16FP8FP8 Multiply the 32x16 matrix in the first input by the 16x32 matrix in the second input and add the 32x32 matrix in the third input using fused multiply add. Store the resulting matrix into vector registers. VOP3P_MFMA default 119 VDST FMT_NUM_PK16_F32 OPR_VGPR_OR_ACCVGPR 512 SRC0 FMT_NUM_PK8_FP8 OPR_SRC_VGPR_OR_ACCVGPR 64 SRC1 FMT_NUM_PK8_FP8 OPR_SRC_VGPR_OR_ACCVGPR 64 SRC2 FMT_NUM_PK16_F32 OPR_SRC_VGPR_OR_ACCVGPR_OR_CONST 512 VALU MFMA FALSE FALSE FALSE FALSE FALSE V_SMFMAC_F32_16X16X64_BF8_BF8 V_SMFMAC_F32_16X16X64BF8BF8 Multiply the 16x64 sparse matrix in the first input by the 64x16 matrix in the second input and accumulate the result into the 16x16 matrix stored in the destination registers using fused multiply add. Sparse indexes for the first matrix are given in the third input. VOP3P_MFMA default 120 VDST FMT_NUM_PK4_F32 OPR_VGPR_OR_ACCVGPR 128 SRC0 FMT_NUM_PK8_BF8 OPR_SRC_VGPR_OR_ACCVGPR 64 SRC1 FMT_NUM_PK16_BF8 OPR_SRC_VGPR_OR_ACCVGPR 128 SRC2 FMT_NUM_B32 OPR_SRC_VGPR 32 VALU MFMA FALSE FALSE FALSE FALSE FALSE V_SMFMAC_F32_16X16X64_BF8_FP8 V_SMFMAC_F32_16X16X64BF8FP8 Multiply the 16x64 sparse matrix in the first input by the 64x16 matrix in the second input and accumulate the result into the 16x16 matrix stored in the destination registers using fused multiply add. Sparse indexes for the first matrix are given in the third input. VOP3P_MFMA default 121 VDST FMT_NUM_PK4_F32 OPR_VGPR_OR_ACCVGPR 128 SRC0 FMT_NUM_PK8_BF8 OPR_SRC_VGPR_OR_ACCVGPR 64 SRC1 FMT_NUM_PK16_FP8 OPR_SRC_VGPR_OR_ACCVGPR 128 SRC2 FMT_NUM_B32 OPR_SRC_VGPR 32 VALU MFMA FALSE FALSE FALSE FALSE FALSE V_SMFMAC_F32_16X16X64_FP8_BF8 V_SMFMAC_F32_16X16X64FP8BF8 Multiply the 16x64 sparse matrix in the first input by the 64x16 matrix in the second input and accumulate the result into the 16x16 matrix stored in the destination registers using fused multiply add. Sparse indexes for the first matrix are given in the third input. VOP3P_MFMA default 122 VDST FMT_NUM_PK4_F32 OPR_VGPR_OR_ACCVGPR 128 SRC0 FMT_NUM_PK8_FP8 OPR_SRC_VGPR_OR_ACCVGPR 64 SRC1 FMT_NUM_PK16_BF8 OPR_SRC_VGPR_OR_ACCVGPR 128 SRC2 FMT_NUM_B32 OPR_SRC_VGPR 32 VALU MFMA FALSE FALSE FALSE FALSE FALSE V_SMFMAC_F32_16X16X64_FP8_FP8 V_SMFMAC_F32_16X16X64FP8FP8 Multiply the 16x64 sparse matrix in the first input by the 64x16 matrix in the second input and accumulate the result into the 16x16 matrix stored in the destination registers using fused multiply add. Sparse indexes for the first matrix are given in the third input. VOP3P_MFMA default 123 VDST FMT_NUM_PK4_F32 OPR_VGPR_OR_ACCVGPR 128 SRC0 FMT_NUM_PK8_FP8 OPR_SRC_VGPR_OR_ACCVGPR 64 SRC1 FMT_NUM_PK16_FP8 OPR_SRC_VGPR_OR_ACCVGPR 128 SRC2 FMT_NUM_B32 OPR_SRC_VGPR 32 VALU MFMA FALSE FALSE FALSE FALSE FALSE V_SMFMAC_F32_32X32X32_BF8_BF8 V_SMFMAC_F32_32X32X32BF8BF8 Multiply the 32x32 sparse matrix in the first input by the 32x32 matrix in the second input and accumulate the result into the 32x32 matrix stored in the destination registers using fused multiply add. Sparse indexes for the first matrix are given in the third input. VOP3P_MFMA default 124 VDST FMT_NUM_PK16_F32 OPR_VGPR_OR_ACCVGPR 512 SRC0 FMT_NUM_PK8_BF8 OPR_SRC_VGPR_OR_ACCVGPR 64 SRC1 FMT_NUM_PK16_BF8 OPR_SRC_VGPR_OR_ACCVGPR 128 SRC2 FMT_NUM_B32 OPR_SRC_VGPR 32 VALU MFMA FALSE FALSE FALSE FALSE FALSE V_SMFMAC_F32_32X32X32_BF8_FP8 V_SMFMAC_F32_32X32X32BF8FP8 Multiply the 32x32 sparse matrix in the first input by the 32x32 matrix in the second input and accumulate the result into the 32x32 matrix stored in the destination registers using fused multiply add. Sparse indexes for the first matrix are given in the third input. VOP3P_MFMA default 125 VDST FMT_NUM_PK16_F32 OPR_VGPR_OR_ACCVGPR 512 SRC0 FMT_NUM_PK8_BF8 OPR_SRC_VGPR_OR_ACCVGPR 64 SRC1 FMT_NUM_PK16_FP8 OPR_SRC_VGPR_OR_ACCVGPR 128 SRC2 FMT_NUM_B32 OPR_SRC_VGPR 32 VALU MFMA FALSE FALSE FALSE FALSE FALSE V_SMFMAC_F32_32X32X32_FP8_BF8 V_SMFMAC_F32_32X32X32FP8BF8 Multiply the 32x32 sparse matrix in the first input by the 32x32 matrix in the second input and accumulate the result into the 32x32 matrix stored in the destination registers using fused multiply add. Sparse indexes for the first matrix are given in the third input. VOP3P_MFMA default 126 VDST FMT_NUM_PK16_F32 OPR_VGPR_OR_ACCVGPR 512 SRC0 FMT_NUM_PK8_FP8 OPR_SRC_VGPR_OR_ACCVGPR 64 SRC1 FMT_NUM_PK16_BF8 OPR_SRC_VGPR_OR_ACCVGPR 128 SRC2 FMT_NUM_B32 OPR_SRC_VGPR 32 VALU MFMA FALSE FALSE FALSE FALSE FALSE V_SMFMAC_F32_32X32X32_FP8_FP8 V_SMFMAC_F32_32X32X32FP8FP8 Multiply the 32x32 sparse matrix in the first input by the 32x32 matrix in the second input and accumulate the result into the 32x32 matrix stored in the destination registers using fused multiply add. Sparse indexes for the first matrix are given in the third input. VOP3P_MFMA default 127 VDST FMT_NUM_PK16_F32 OPR_VGPR_OR_ACCVGPR 512 SRC0 FMT_NUM_PK8_FP8 OPR_SRC_VGPR_OR_ACCVGPR 64 SRC1 FMT_NUM_PK16_FP8 OPR_SRC_VGPR_OR_ACCVGPR 128 SRC2 FMT_NUM_B32 OPR_SRC_VGPR 32 VALU MFMA FALSE FALSE FALSE FALSE FALSE V_MFMA_SCALE_F32_16X16X128_F8F6F4 Multiply the 16x128 matrix in the first input by the 128x16 matrix in the second input and add the 16x16 matrix in the third input using fused multiply add. Store the resulting matrix into vector registers. ENC_VOP3PX2 default 45 VDST FMT_NUM_PK4_F32 OPR_VGPR_OR_ACCVGPR 128 SRC0 FMT_NUM_PK8_B32 OPR_SRC_VGPR_OR_ACCVGPR 256 SRC1 FMT_NUM_PK8_B32 OPR_SRC_VGPR_OR_ACCVGPR 256 SRC2 FMT_NUM_PK4_F32 OPR_SRC_VGPR_OR_ACCVGPR_OR_CONST 128 SCALE_SRC0 FMT_NUM_B32 OPR_SRC_SIMPLE 32 SCALE_SRC1 FMT_NUM_B32 OPR_SRC_SIMPLE 32 VALU MFMA FALSE FALSE FALSE FALSE FALSE V_MFMA_SCALE_F32_32X32X64_F8F6F4 Multiply the 32x64 matrix in the first input by the 64x32 matrix in the second input and add the 32x32 matrix in the third input using fused multiply add. Store the resulting matrix into vector registers. ENC_VOP3PX2 default 46 VDST FMT_NUM_PK16_F32 OPR_VGPR_OR_ACCVGPR 512 SRC0 FMT_NUM_PK8_B32 OPR_SRC_VGPR_OR_ACCVGPR 256 SRC1 FMT_NUM_PK8_B32 OPR_SRC_VGPR_OR_ACCVGPR 256 SRC2 FMT_NUM_PK16_F32 OPR_SRC_VGPR_OR_ACCVGPR_OR_CONST 512 SCALE_SRC0 FMT_NUM_B32 OPR_SRC_SIMPLE 32 SCALE_SRC1 FMT_NUM_B32 OPR_SRC_SIMPLE 32 VALU MFMA FALSE FALSE FALSE FALSE FALSE V_CMP_CLASS_F32 Evaluate the IEEE numeric class function specified as a 10 bit mask in the second input on the first input, a single-precision float, and set the per-lane condition code to the result. Store the result into VCC or a scalar register. ENC_VOP3 default 16 VDST FMT_NUM_B64 OPR_SREG 64 SRC0 FMT_NUM_F32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_B32 OPR_SRC_SIMPLE 32 ENC_VOPC default 16 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_B32 OPR_VGPR 32 VOPC_INST_LITERAL has_lit 16 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_B32 OPR_VGPR 32 VOPC_VOP_SDWA_SDST_ENC has_sdwa 16 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_B32 OPR_VGPR 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMPX_CLASS_F32 Evaluate the IEEE numeric class function specified as a 10 bit mask in the second input on the first input, a single-precision float, and set the per-lane condition code to the result. Store the result into the EXEC mask and to VCC or a scalar register. ENC_VOP3 default 17 VDST FMT_NUM_B64 OPR_SDST 64 SRC0 FMT_NUM_F32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_B32 OPR_SRC_SIMPLE 32 FMT_NUM_M64 OPR_SDST_EXEC 64 ENC_VOPC default 17 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_B32 OPR_VGPR 32 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_INST_LITERAL has_lit 17 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_B32 OPR_VGPR 32 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_VOP_SDWA_SDST_ENC has_sdwa 17 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_B32 OPR_VGPR 32 FMT_NUM_M64 OPR_SDST_EXEC 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMP_CLASS_F64 Evaluate the IEEE numeric class function specified as a 10 bit mask in the second input on the first input, a double-precision float, and set the per-lane condition code to the result. Store the result into VCC or a scalar register. ENC_VOP3 default 18 VDST FMT_NUM_B64 OPR_SREG 64 SRC0 FMT_NUM_F64 OPR_SRC_NOLIT 64 SRC1 FMT_NUM_B32 OPR_SRC_SIMPLE 32 ENC_VOPC default 18 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_F64 OPR_SRC 64 VSRC1 FMT_NUM_B32 OPR_VGPR 32 VOPC_INST_LITERAL has_lit 18 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_F64 OPR_SRC 64 VSRC1 FMT_NUM_B32 OPR_VGPR 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMPX_CLASS_F64 Evaluate the IEEE numeric class function specified as a 10 bit mask in the second input on the first input, a double-precision float, and set the per-lane condition code to the result. Store the result into the EXEC mask and to VCC or a scalar register. ENC_VOP3 default 19 VDST FMT_NUM_B64 OPR_SDST 64 SRC0 FMT_NUM_F64 OPR_SRC_NOLIT 64 SRC1 FMT_NUM_B32 OPR_SRC_SIMPLE 32 FMT_NUM_M64 OPR_SDST_EXEC 64 ENC_VOPC default 19 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_F64 OPR_SRC 64 VSRC1 FMT_NUM_B32 OPR_VGPR 32 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_INST_LITERAL has_lit 19 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_F64 OPR_SRC 64 VSRC1 FMT_NUM_B32 OPR_VGPR 32 FMT_NUM_M64 OPR_SDST_EXEC 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMP_CLASS_F16 Evaluate the IEEE numeric class function specified as a 10 bit mask in the second input on the first input, a half-precision float, and set the per-lane condition code to the result. Store the result into VCC or a scalar register. ENC_VOP3 default 20 VDST FMT_NUM_B64 OPR_SREG 64 SRC0 FMT_NUM_F16 OPR_SRC_NOLIT 16 SRC1 FMT_NUM_F16 OPR_SRC_SIMPLE 16 ENC_VOPC default 20 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 VOPC_INST_LITERAL has_lit 20 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 VOPC_VOP_SDWA_SDST_ENC has_sdwa 20 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMPX_CLASS_F16 Evaluate the IEEE numeric class function specified as a 10 bit mask in the second input on the first input, a half-precision float, and set the per-lane condition code to the result. Store the result into the EXEC mask and to VCC or a scalar register. ENC_VOP3 default 21 VDST FMT_NUM_B64 OPR_SDST 64 SRC0 FMT_NUM_F16 OPR_SRC_NOLIT 16 SRC1 FMT_NUM_F16 OPR_SRC_SIMPLE 16 FMT_NUM_M64 OPR_SDST_EXEC 64 ENC_VOPC default 21 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_INST_LITERAL has_lit 21 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_VOP_SDWA_SDST_ENC has_sdwa 21 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 FMT_NUM_M64 OPR_SDST_EXEC 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMP_F_F16 Set the per-lane condition code to 0. Store the result into VCC or a scalar register. ENC_VOP3 default 32 VDST FMT_NUM_B64 OPR_SREG 64 SRC0 FMT_NUM_F16 OPR_SRC_NOLIT 16 SRC1 FMT_NUM_F16 OPR_SRC_SIMPLE 16 ENC_VOPC default 32 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 VOPC_INST_LITERAL has_lit 32 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 VOPC_VOP_SDWA_SDST_ENC has_sdwa 32 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMP_LT_F16 Set the per-lane condition code to 1 iff the first input is less than the second input. Store the result into VCC or a scalar register. ENC_VOP3 default 33 VDST FMT_NUM_B64 OPR_SREG 64 SRC0 FMT_NUM_F16 OPR_SRC_NOLIT 16 SRC1 FMT_NUM_F16 OPR_SRC_SIMPLE 16 ENC_VOPC default 33 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 VOPC_INST_LITERAL has_lit 33 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 VOPC_VOP_SDWA_SDST_ENC has_sdwa 33 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMP_EQ_F16 Set the per-lane condition code to 1 iff the first input is equal to the second input. Store the result into VCC or a scalar register. ENC_VOP3 default 34 VDST FMT_NUM_B64 OPR_SREG 64 SRC0 FMT_NUM_F16 OPR_SRC_NOLIT 16 SRC1 FMT_NUM_F16 OPR_SRC_SIMPLE 16 ENC_VOPC default 34 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 VOPC_INST_LITERAL has_lit 34 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 VOPC_VOP_SDWA_SDST_ENC has_sdwa 34 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMP_LE_F16 Set the per-lane condition code to 1 iff the first input is less than or equal to the second input. Store the result into VCC or a scalar register. ENC_VOP3 default 35 VDST FMT_NUM_B64 OPR_SREG 64 SRC0 FMT_NUM_F16 OPR_SRC_NOLIT 16 SRC1 FMT_NUM_F16 OPR_SRC_SIMPLE 16 ENC_VOPC default 35 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 VOPC_INST_LITERAL has_lit 35 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 VOPC_VOP_SDWA_SDST_ENC has_sdwa 35 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMP_GT_F16 Set the per-lane condition code to 1 iff the first input is greater than the second input. Store the result into VCC or a scalar register. ENC_VOP3 default 36 VDST FMT_NUM_B64 OPR_SREG 64 SRC0 FMT_NUM_F16 OPR_SRC_NOLIT 16 SRC1 FMT_NUM_F16 OPR_SRC_SIMPLE 16 ENC_VOPC default 36 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 VOPC_INST_LITERAL has_lit 36 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 VOPC_VOP_SDWA_SDST_ENC has_sdwa 36 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMP_LG_F16 Set the per-lane condition code to 1 iff the first input is less than or greater than the second input. Store the result into VCC or a scalar register. ENC_VOP3 default 37 VDST FMT_NUM_B64 OPR_SREG 64 SRC0 FMT_NUM_F16 OPR_SRC_NOLIT 16 SRC1 FMT_NUM_F16 OPR_SRC_SIMPLE 16 ENC_VOPC default 37 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 VOPC_INST_LITERAL has_lit 37 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 VOPC_VOP_SDWA_SDST_ENC has_sdwa 37 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMP_GE_F16 Set the per-lane condition code to 1 iff the first input is greater than or equal to the second input. Store the result into VCC or a scalar register. ENC_VOP3 default 38 VDST FMT_NUM_B64 OPR_SREG 64 SRC0 FMT_NUM_F16 OPR_SRC_NOLIT 16 SRC1 FMT_NUM_F16 OPR_SRC_SIMPLE 16 ENC_VOPC default 38 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 VOPC_INST_LITERAL has_lit 38 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 VOPC_VOP_SDWA_SDST_ENC has_sdwa 38 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMP_O_F16 Set the per-lane condition code to 1 iff the first input is orderable to the second input. Store the result into VCC or a scalar register. ENC_VOP3 default 39 VDST FMT_NUM_B64 OPR_SREG 64 SRC0 FMT_NUM_F16 OPR_SRC_NOLIT 16 SRC1 FMT_NUM_F16 OPR_SRC_SIMPLE 16 ENC_VOPC default 39 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 VOPC_INST_LITERAL has_lit 39 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 VOPC_VOP_SDWA_SDST_ENC has_sdwa 39 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMP_U_F16 Set the per-lane condition code to 1 iff the first input is not orderable to the second input. Store the result into VCC or a scalar register. ENC_VOP3 default 40 VDST FMT_NUM_B64 OPR_SREG 64 SRC0 FMT_NUM_F16 OPR_SRC_NOLIT 16 SRC1 FMT_NUM_F16 OPR_SRC_SIMPLE 16 ENC_VOPC default 40 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 VOPC_INST_LITERAL has_lit 40 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 VOPC_VOP_SDWA_SDST_ENC has_sdwa 40 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMP_NGE_F16 Set the per-lane condition code to 1 iff the first input is not greater than or equal to the second input. Store the result into VCC or a scalar register. ENC_VOP3 default 41 VDST FMT_NUM_B64 OPR_SREG 64 SRC0 FMT_NUM_F16 OPR_SRC_NOLIT 16 SRC1 FMT_NUM_F16 OPR_SRC_SIMPLE 16 ENC_VOPC default 41 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 VOPC_INST_LITERAL has_lit 41 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 VOPC_VOP_SDWA_SDST_ENC has_sdwa 41 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMP_NLG_F16 Set the per-lane condition code to 1 iff the first input is not less than or greater than the second input. Store the result into VCC or a scalar register. ENC_VOP3 default 42 VDST FMT_NUM_B64 OPR_SREG 64 SRC0 FMT_NUM_F16 OPR_SRC_NOLIT 16 SRC1 FMT_NUM_F16 OPR_SRC_SIMPLE 16 ENC_VOPC default 42 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 VOPC_INST_LITERAL has_lit 42 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 VOPC_VOP_SDWA_SDST_ENC has_sdwa 42 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMP_NGT_F16 Set the per-lane condition code to 1 iff the first input is not greater than the second input. Store the result into VCC or a scalar register. ENC_VOP3 default 43 VDST FMT_NUM_B64 OPR_SREG 64 SRC0 FMT_NUM_F16 OPR_SRC_NOLIT 16 SRC1 FMT_NUM_F16 OPR_SRC_SIMPLE 16 ENC_VOPC default 43 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 VOPC_INST_LITERAL has_lit 43 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 VOPC_VOP_SDWA_SDST_ENC has_sdwa 43 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMP_NLE_F16 Set the per-lane condition code to 1 iff the first input is not less than or equal to the second input. Store the result into VCC or a scalar register. ENC_VOP3 default 44 VDST FMT_NUM_B64 OPR_SREG 64 SRC0 FMT_NUM_F16 OPR_SRC_NOLIT 16 SRC1 FMT_NUM_F16 OPR_SRC_SIMPLE 16 ENC_VOPC default 44 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 VOPC_INST_LITERAL has_lit 44 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 VOPC_VOP_SDWA_SDST_ENC has_sdwa 44 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMP_NEQ_F16 Set the per-lane condition code to 1 iff the first input is not equal to the second input. Store the result into VCC or a scalar register. ENC_VOP3 default 45 VDST FMT_NUM_B64 OPR_SREG 64 SRC0 FMT_NUM_F16 OPR_SRC_NOLIT 16 SRC1 FMT_NUM_F16 OPR_SRC_SIMPLE 16 ENC_VOPC default 45 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 VOPC_INST_LITERAL has_lit 45 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 VOPC_VOP_SDWA_SDST_ENC has_sdwa 45 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMP_NLT_F16 Set the per-lane condition code to 1 iff the first input is not less than the second input. Store the result into VCC or a scalar register. ENC_VOP3 default 46 VDST FMT_NUM_B64 OPR_SREG 64 SRC0 FMT_NUM_F16 OPR_SRC_NOLIT 16 SRC1 FMT_NUM_F16 OPR_SRC_SIMPLE 16 ENC_VOPC default 46 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 VOPC_INST_LITERAL has_lit 46 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 VOPC_VOP_SDWA_SDST_ENC has_sdwa 46 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMP_TRU_F16 Set the per-lane condition code to 1. Store the result into VCC or a scalar register. ENC_VOP3 default 47 VDST FMT_NUM_B64 OPR_SREG 64 SRC0 FMT_NUM_F16 OPR_SRC_NOLIT 16 SRC1 FMT_NUM_F16 OPR_SRC_SIMPLE 16 ENC_VOPC default 47 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 VOPC_INST_LITERAL has_lit 47 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 VOPC_VOP_SDWA_SDST_ENC has_sdwa 47 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMPX_F_F16 Set the per-lane condition code to 0. Store the result into the EXEC mask and to VCC or a scalar register. ENC_VOP3 default 48 VDST FMT_NUM_B64 OPR_SDST 64 SRC0 FMT_NUM_F16 OPR_SRC_NOLIT 16 SRC1 FMT_NUM_F16 OPR_SRC_SIMPLE 16 FMT_NUM_M64 OPR_SDST_EXEC 64 ENC_VOPC default 48 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_INST_LITERAL has_lit 48 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_VOP_SDWA_SDST_ENC has_sdwa 48 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 FMT_NUM_M64 OPR_SDST_EXEC 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMPX_LT_F16 Set the per-lane condition code to 1 iff the first input is less than the second input. Store the result into the EXEC mask and to VCC or a scalar register. ENC_VOP3 default 49 VDST FMT_NUM_B64 OPR_SDST 64 SRC0 FMT_NUM_F16 OPR_SRC_NOLIT 16 SRC1 FMT_NUM_F16 OPR_SRC_SIMPLE 16 FMT_NUM_M64 OPR_SDST_EXEC 64 ENC_VOPC default 49 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_INST_LITERAL has_lit 49 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_VOP_SDWA_SDST_ENC has_sdwa 49 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 FMT_NUM_M64 OPR_SDST_EXEC 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMPX_EQ_F16 Set the per-lane condition code to 1 iff the first input is equal to the second input. Store the result into the EXEC mask and to VCC or a scalar register. ENC_VOP3 default 50 VDST FMT_NUM_B64 OPR_SDST 64 SRC0 FMT_NUM_F16 OPR_SRC_NOLIT 16 SRC1 FMT_NUM_F16 OPR_SRC_SIMPLE 16 FMT_NUM_M64 OPR_SDST_EXEC 64 ENC_VOPC default 50 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_INST_LITERAL has_lit 50 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_VOP_SDWA_SDST_ENC has_sdwa 50 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 FMT_NUM_M64 OPR_SDST_EXEC 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMPX_LE_F16 Set the per-lane condition code to 1 iff the first input is less than or equal to the second input. Store the result into the EXEC mask and to VCC or a scalar register. ENC_VOP3 default 51 VDST FMT_NUM_B64 OPR_SDST 64 SRC0 FMT_NUM_F16 OPR_SRC_NOLIT 16 SRC1 FMT_NUM_F16 OPR_SRC_SIMPLE 16 FMT_NUM_M64 OPR_SDST_EXEC 64 ENC_VOPC default 51 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_INST_LITERAL has_lit 51 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_VOP_SDWA_SDST_ENC has_sdwa 51 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 FMT_NUM_M64 OPR_SDST_EXEC 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMPX_GT_F16 Set the per-lane condition code to 1 iff the first input is greater than the second input. Store the result into the EXEC mask and to VCC or a scalar register. ENC_VOP3 default 52 VDST FMT_NUM_B64 OPR_SDST 64 SRC0 FMT_NUM_F16 OPR_SRC_NOLIT 16 SRC1 FMT_NUM_F16 OPR_SRC_SIMPLE 16 FMT_NUM_M64 OPR_SDST_EXEC 64 ENC_VOPC default 52 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_INST_LITERAL has_lit 52 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_VOP_SDWA_SDST_ENC has_sdwa 52 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 FMT_NUM_M64 OPR_SDST_EXEC 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMPX_LG_F16 Set the per-lane condition code to 1 iff the first input is less than or greater than the second input. Store the result into the EXEC mask and to VCC or a scalar register. ENC_VOP3 default 53 VDST FMT_NUM_B64 OPR_SDST 64 SRC0 FMT_NUM_F16 OPR_SRC_NOLIT 16 SRC1 FMT_NUM_F16 OPR_SRC_SIMPLE 16 FMT_NUM_M64 OPR_SDST_EXEC 64 ENC_VOPC default 53 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_INST_LITERAL has_lit 53 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_VOP_SDWA_SDST_ENC has_sdwa 53 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 FMT_NUM_M64 OPR_SDST_EXEC 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMPX_GE_F16 Set the per-lane condition code to 1 iff the first input is greater than or equal to the second input. Store the result into the EXEC mask and to VCC or a scalar register. ENC_VOP3 default 54 VDST FMT_NUM_B64 OPR_SDST 64 SRC0 FMT_NUM_F16 OPR_SRC_NOLIT 16 SRC1 FMT_NUM_F16 OPR_SRC_SIMPLE 16 FMT_NUM_M64 OPR_SDST_EXEC 64 ENC_VOPC default 54 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_INST_LITERAL has_lit 54 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_VOP_SDWA_SDST_ENC has_sdwa 54 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 FMT_NUM_M64 OPR_SDST_EXEC 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMPX_O_F16 Set the per-lane condition code to 1 iff the first input is orderable to the second input. Store the result into the EXEC mask and to VCC or a scalar register. ENC_VOP3 default 55 VDST FMT_NUM_B64 OPR_SDST 64 SRC0 FMT_NUM_F16 OPR_SRC_NOLIT 16 SRC1 FMT_NUM_F16 OPR_SRC_SIMPLE 16 FMT_NUM_M64 OPR_SDST_EXEC 64 ENC_VOPC default 55 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_INST_LITERAL has_lit 55 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_VOP_SDWA_SDST_ENC has_sdwa 55 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 FMT_NUM_M64 OPR_SDST_EXEC 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMPX_U_F16 Set the per-lane condition code to 1 iff the first input is not orderable to the second input. Store the result into the EXEC mask and to VCC or a scalar register. ENC_VOP3 default 56 VDST FMT_NUM_B64 OPR_SDST 64 SRC0 FMT_NUM_F16 OPR_SRC_NOLIT 16 SRC1 FMT_NUM_F16 OPR_SRC_SIMPLE 16 FMT_NUM_M64 OPR_SDST_EXEC 64 ENC_VOPC default 56 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_INST_LITERAL has_lit 56 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_VOP_SDWA_SDST_ENC has_sdwa 56 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 FMT_NUM_M64 OPR_SDST_EXEC 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMPX_NGE_F16 Set the per-lane condition code to 1 iff the first input is not greater than or equal to the second input. Store the result into the EXEC mask and to VCC or a scalar register. ENC_VOP3 default 57 VDST FMT_NUM_B64 OPR_SDST 64 SRC0 FMT_NUM_F16 OPR_SRC_NOLIT 16 SRC1 FMT_NUM_F16 OPR_SRC_SIMPLE 16 FMT_NUM_M64 OPR_SDST_EXEC 64 ENC_VOPC default 57 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_INST_LITERAL has_lit 57 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_VOP_SDWA_SDST_ENC has_sdwa 57 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 FMT_NUM_M64 OPR_SDST_EXEC 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMPX_NLG_F16 Set the per-lane condition code to 1 iff the first input is not less than or greater than the second input. Store the result into the EXEC mask and to VCC or a scalar register. ENC_VOP3 default 58 VDST FMT_NUM_B64 OPR_SDST 64 SRC0 FMT_NUM_F16 OPR_SRC_NOLIT 16 SRC1 FMT_NUM_F16 OPR_SRC_SIMPLE 16 FMT_NUM_M64 OPR_SDST_EXEC 64 ENC_VOPC default 58 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_INST_LITERAL has_lit 58 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_VOP_SDWA_SDST_ENC has_sdwa 58 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 FMT_NUM_M64 OPR_SDST_EXEC 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMPX_NGT_F16 Set the per-lane condition code to 1 iff the first input is not greater than the second input. Store the result into the EXEC mask and to VCC or a scalar register. ENC_VOP3 default 59 VDST FMT_NUM_B64 OPR_SDST 64 SRC0 FMT_NUM_F16 OPR_SRC_NOLIT 16 SRC1 FMT_NUM_F16 OPR_SRC_SIMPLE 16 FMT_NUM_M64 OPR_SDST_EXEC 64 ENC_VOPC default 59 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_INST_LITERAL has_lit 59 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_VOP_SDWA_SDST_ENC has_sdwa 59 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 FMT_NUM_M64 OPR_SDST_EXEC 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMPX_NLE_F16 Set the per-lane condition code to 1 iff the first input is not less than or equal to the second input. Store the result into the EXEC mask and to VCC or a scalar register. ENC_VOP3 default 60 VDST FMT_NUM_B64 OPR_SDST 64 SRC0 FMT_NUM_F16 OPR_SRC_NOLIT 16 SRC1 FMT_NUM_F16 OPR_SRC_SIMPLE 16 FMT_NUM_M64 OPR_SDST_EXEC 64 ENC_VOPC default 60 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_INST_LITERAL has_lit 60 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_VOP_SDWA_SDST_ENC has_sdwa 60 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 FMT_NUM_M64 OPR_SDST_EXEC 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMPX_NEQ_F16 Set the per-lane condition code to 1 iff the first input is not equal to the second input. Store the result into the EXEC mask and to VCC or a scalar register. ENC_VOP3 default 61 VDST FMT_NUM_B64 OPR_SDST 64 SRC0 FMT_NUM_F16 OPR_SRC_NOLIT 16 SRC1 FMT_NUM_F16 OPR_SRC_SIMPLE 16 FMT_NUM_M64 OPR_SDST_EXEC 64 ENC_VOPC default 61 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_INST_LITERAL has_lit 61 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_VOP_SDWA_SDST_ENC has_sdwa 61 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 FMT_NUM_M64 OPR_SDST_EXEC 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMPX_NLT_F16 Set the per-lane condition code to 1 iff the first input is not less than the second input. Store the result into the EXEC mask and to VCC or a scalar register. ENC_VOP3 default 62 VDST FMT_NUM_B64 OPR_SDST 64 SRC0 FMT_NUM_F16 OPR_SRC_NOLIT 16 SRC1 FMT_NUM_F16 OPR_SRC_SIMPLE 16 FMT_NUM_M64 OPR_SDST_EXEC 64 ENC_VOPC default 62 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_INST_LITERAL has_lit 62 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_VOP_SDWA_SDST_ENC has_sdwa 62 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 FMT_NUM_M64 OPR_SDST_EXEC 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMPX_TRU_F16 Set the per-lane condition code to 1. Store the result into the EXEC mask and to VCC or a scalar register. ENC_VOP3 default 63 VDST FMT_NUM_B64 OPR_SDST 64 SRC0 FMT_NUM_F16 OPR_SRC_NOLIT 16 SRC1 FMT_NUM_F16 OPR_SRC_SIMPLE 16 FMT_NUM_M64 OPR_SDST_EXEC 64 ENC_VOPC default 63 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_INST_LITERAL has_lit 63 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_VOP_SDWA_SDST_ENC has_sdwa 63 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_F16 OPR_SRC 16 VSRC1 FMT_NUM_F16 OPR_VGPR 16 FMT_NUM_M64 OPR_SDST_EXEC 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMP_F_F32 Set the per-lane condition code to 0. Store the result into VCC or a scalar register. ENC_VOP3 default 64 VDST FMT_NUM_B64 OPR_SREG 64 SRC0 FMT_NUM_F32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_F32 OPR_SRC_SIMPLE 32 ENC_VOPC default 64 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 VOPC_INST_LITERAL has_lit 64 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 VOPC_VOP_SDWA_SDST_ENC has_sdwa 64 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMP_LT_F32 Set the per-lane condition code to 1 iff the first input is less than the second input. Store the result into VCC or a scalar register. ENC_VOP3 default 65 VDST FMT_NUM_B64 OPR_SREG 64 SRC0 FMT_NUM_F32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_F32 OPR_SRC_SIMPLE 32 ENC_VOPC default 65 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 VOPC_INST_LITERAL has_lit 65 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 VOPC_VOP_SDWA_SDST_ENC has_sdwa 65 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMP_EQ_F32 Set the per-lane condition code to 1 iff the first input is equal to the second input. Store the result into VCC or a scalar register. ENC_VOP3 default 66 VDST FMT_NUM_B64 OPR_SREG 64 SRC0 FMT_NUM_F32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_F32 OPR_SRC_SIMPLE 32 ENC_VOPC default 66 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 VOPC_INST_LITERAL has_lit 66 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 VOPC_VOP_SDWA_SDST_ENC has_sdwa 66 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMP_LE_F32 Set the per-lane condition code to 1 iff the first input is less than or equal to the second input. Store the result into VCC or a scalar register. ENC_VOP3 default 67 VDST FMT_NUM_B64 OPR_SREG 64 SRC0 FMT_NUM_F32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_F32 OPR_SRC_SIMPLE 32 ENC_VOPC default 67 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 VOPC_INST_LITERAL has_lit 67 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 VOPC_VOP_SDWA_SDST_ENC has_sdwa 67 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMP_GT_F32 Set the per-lane condition code to 1 iff the first input is greater than the second input. Store the result into VCC or a scalar register. ENC_VOP3 default 68 VDST FMT_NUM_B64 OPR_SREG 64 SRC0 FMT_NUM_F32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_F32 OPR_SRC_SIMPLE 32 ENC_VOPC default 68 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 VOPC_INST_LITERAL has_lit 68 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 VOPC_VOP_SDWA_SDST_ENC has_sdwa 68 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMP_LG_F32 Set the per-lane condition code to 1 iff the first input is less than or greater than the second input. Store the result into VCC or a scalar register. ENC_VOP3 default 69 VDST FMT_NUM_B64 OPR_SREG 64 SRC0 FMT_NUM_F32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_F32 OPR_SRC_SIMPLE 32 ENC_VOPC default 69 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 VOPC_INST_LITERAL has_lit 69 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 VOPC_VOP_SDWA_SDST_ENC has_sdwa 69 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMP_GE_F32 Set the per-lane condition code to 1 iff the first input is greater than or equal to the second input. Store the result into VCC or a scalar register. ENC_VOP3 default 70 VDST FMT_NUM_B64 OPR_SREG 64 SRC0 FMT_NUM_F32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_F32 OPR_SRC_SIMPLE 32 ENC_VOPC default 70 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 VOPC_INST_LITERAL has_lit 70 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 VOPC_VOP_SDWA_SDST_ENC has_sdwa 70 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMP_O_F32 Set the per-lane condition code to 1 iff the first input is orderable to the second input. Store the result into VCC or a scalar register. ENC_VOP3 default 71 VDST FMT_NUM_B64 OPR_SREG 64 SRC0 FMT_NUM_F32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_F32 OPR_SRC_SIMPLE 32 ENC_VOPC default 71 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 VOPC_INST_LITERAL has_lit 71 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 VOPC_VOP_SDWA_SDST_ENC has_sdwa 71 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMP_U_F32 Set the per-lane condition code to 1 iff the first input is not orderable to the second input. Store the result into VCC or a scalar register. ENC_VOP3 default 72 VDST FMT_NUM_B64 OPR_SREG 64 SRC0 FMT_NUM_F32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_F32 OPR_SRC_SIMPLE 32 ENC_VOPC default 72 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 VOPC_INST_LITERAL has_lit 72 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 VOPC_VOP_SDWA_SDST_ENC has_sdwa 72 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMP_NGE_F32 Set the per-lane condition code to 1 iff the first input is not greater than or equal to the second input. Store the result into VCC or a scalar register. ENC_VOP3 default 73 VDST FMT_NUM_B64 OPR_SREG 64 SRC0 FMT_NUM_F32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_F32 OPR_SRC_SIMPLE 32 ENC_VOPC default 73 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 VOPC_INST_LITERAL has_lit 73 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 VOPC_VOP_SDWA_SDST_ENC has_sdwa 73 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMP_NLG_F32 Set the per-lane condition code to 1 iff the first input is not less than or greater than the second input. Store the result into VCC or a scalar register. ENC_VOP3 default 74 VDST FMT_NUM_B64 OPR_SREG 64 SRC0 FMT_NUM_F32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_F32 OPR_SRC_SIMPLE 32 ENC_VOPC default 74 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 VOPC_INST_LITERAL has_lit 74 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 VOPC_VOP_SDWA_SDST_ENC has_sdwa 74 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMP_NGT_F32 Set the per-lane condition code to 1 iff the first input is not greater than the second input. Store the result into VCC or a scalar register. ENC_VOP3 default 75 VDST FMT_NUM_B64 OPR_SREG 64 SRC0 FMT_NUM_F32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_F32 OPR_SRC_SIMPLE 32 ENC_VOPC default 75 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 VOPC_INST_LITERAL has_lit 75 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 VOPC_VOP_SDWA_SDST_ENC has_sdwa 75 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMP_NLE_F32 Set the per-lane condition code to 1 iff the first input is not less than or equal to the second input. Store the result into VCC or a scalar register. ENC_VOP3 default 76 VDST FMT_NUM_B64 OPR_SREG 64 SRC0 FMT_NUM_F32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_F32 OPR_SRC_SIMPLE 32 ENC_VOPC default 76 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 VOPC_INST_LITERAL has_lit 76 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 VOPC_VOP_SDWA_SDST_ENC has_sdwa 76 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMP_NEQ_F32 Set the per-lane condition code to 1 iff the first input is not equal to the second input. Store the result into VCC or a scalar register. ENC_VOP3 default 77 VDST FMT_NUM_B64 OPR_SREG 64 SRC0 FMT_NUM_F32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_F32 OPR_SRC_SIMPLE 32 ENC_VOPC default 77 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 VOPC_INST_LITERAL has_lit 77 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 VOPC_VOP_SDWA_SDST_ENC has_sdwa 77 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMP_NLT_F32 Set the per-lane condition code to 1 iff the first input is not less than the second input. Store the result into VCC or a scalar register. ENC_VOP3 default 78 VDST FMT_NUM_B64 OPR_SREG 64 SRC0 FMT_NUM_F32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_F32 OPR_SRC_SIMPLE 32 ENC_VOPC default 78 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 VOPC_INST_LITERAL has_lit 78 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 VOPC_VOP_SDWA_SDST_ENC has_sdwa 78 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMP_TRU_F32 Set the per-lane condition code to 1. Store the result into VCC or a scalar register. ENC_VOP3 default 79 VDST FMT_NUM_B64 OPR_SREG 64 SRC0 FMT_NUM_F32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_F32 OPR_SRC_SIMPLE 32 ENC_VOPC default 79 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 VOPC_INST_LITERAL has_lit 79 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 VOPC_VOP_SDWA_SDST_ENC has_sdwa 79 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMPX_F_F32 Set the per-lane condition code to 0. Store the result into the EXEC mask and to VCC or a scalar register. ENC_VOP3 default 80 VDST FMT_NUM_B64 OPR_SDST 64 SRC0 FMT_NUM_F32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_F32 OPR_SRC_SIMPLE 32 FMT_NUM_M64 OPR_SDST_EXEC 64 ENC_VOPC default 80 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_INST_LITERAL has_lit 80 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_VOP_SDWA_SDST_ENC has_sdwa 80 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 FMT_NUM_M64 OPR_SDST_EXEC 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMPX_LT_F32 Set the per-lane condition code to 1 iff the first input is less than the second input. Store the result into the EXEC mask and to VCC or a scalar register. ENC_VOP3 default 81 VDST FMT_NUM_B64 OPR_SDST 64 SRC0 FMT_NUM_F32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_F32 OPR_SRC_SIMPLE 32 FMT_NUM_M64 OPR_SDST_EXEC 64 ENC_VOPC default 81 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_INST_LITERAL has_lit 81 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_VOP_SDWA_SDST_ENC has_sdwa 81 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 FMT_NUM_M64 OPR_SDST_EXEC 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMPX_EQ_F32 Set the per-lane condition code to 1 iff the first input is equal to the second input. Store the result into the EXEC mask and to VCC or a scalar register. ENC_VOP3 default 82 VDST FMT_NUM_B64 OPR_SDST 64 SRC0 FMT_NUM_F32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_F32 OPR_SRC_SIMPLE 32 FMT_NUM_M64 OPR_SDST_EXEC 64 ENC_VOPC default 82 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_INST_LITERAL has_lit 82 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_VOP_SDWA_SDST_ENC has_sdwa 82 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 FMT_NUM_M64 OPR_SDST_EXEC 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMPX_LE_F32 Set the per-lane condition code to 1 iff the first input is less than or equal to the second input. Store the result into the EXEC mask and to VCC or a scalar register. ENC_VOP3 default 83 VDST FMT_NUM_B64 OPR_SDST 64 SRC0 FMT_NUM_F32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_F32 OPR_SRC_SIMPLE 32 FMT_NUM_M64 OPR_SDST_EXEC 64 ENC_VOPC default 83 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_INST_LITERAL has_lit 83 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_VOP_SDWA_SDST_ENC has_sdwa 83 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 FMT_NUM_M64 OPR_SDST_EXEC 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMPX_GT_F32 Set the per-lane condition code to 1 iff the first input is greater than the second input. Store the result into the EXEC mask and to VCC or a scalar register. ENC_VOP3 default 84 VDST FMT_NUM_B64 OPR_SDST 64 SRC0 FMT_NUM_F32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_F32 OPR_SRC_SIMPLE 32 FMT_NUM_M64 OPR_SDST_EXEC 64 ENC_VOPC default 84 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_INST_LITERAL has_lit 84 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_VOP_SDWA_SDST_ENC has_sdwa 84 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 FMT_NUM_M64 OPR_SDST_EXEC 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMPX_LG_F32 Set the per-lane condition code to 1 iff the first input is less than or greater than the second input. Store the result into the EXEC mask and to VCC or a scalar register. ENC_VOP3 default 85 VDST FMT_NUM_B64 OPR_SDST 64 SRC0 FMT_NUM_F32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_F32 OPR_SRC_SIMPLE 32 FMT_NUM_M64 OPR_SDST_EXEC 64 ENC_VOPC default 85 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_INST_LITERAL has_lit 85 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_VOP_SDWA_SDST_ENC has_sdwa 85 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 FMT_NUM_M64 OPR_SDST_EXEC 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMPX_GE_F32 Set the per-lane condition code to 1 iff the first input is greater than or equal to the second input. Store the result into the EXEC mask and to VCC or a scalar register. ENC_VOP3 default 86 VDST FMT_NUM_B64 OPR_SDST 64 SRC0 FMT_NUM_F32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_F32 OPR_SRC_SIMPLE 32 FMT_NUM_M64 OPR_SDST_EXEC 64 ENC_VOPC default 86 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_INST_LITERAL has_lit 86 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_VOP_SDWA_SDST_ENC has_sdwa 86 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 FMT_NUM_M64 OPR_SDST_EXEC 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMPX_O_F32 Set the per-lane condition code to 1 iff the first input is orderable to the second input. Store the result into the EXEC mask and to VCC or a scalar register. ENC_VOP3 default 87 VDST FMT_NUM_B64 OPR_SDST 64 SRC0 FMT_NUM_F32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_F32 OPR_SRC_SIMPLE 32 FMT_NUM_M64 OPR_SDST_EXEC 64 ENC_VOPC default 87 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_INST_LITERAL has_lit 87 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_VOP_SDWA_SDST_ENC has_sdwa 87 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 FMT_NUM_M64 OPR_SDST_EXEC 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMPX_U_F32 Set the per-lane condition code to 1 iff the first input is not orderable to the second input. Store the result into the EXEC mask and to VCC or a scalar register. ENC_VOP3 default 88 VDST FMT_NUM_B64 OPR_SDST 64 SRC0 FMT_NUM_F32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_F32 OPR_SRC_SIMPLE 32 FMT_NUM_M64 OPR_SDST_EXEC 64 ENC_VOPC default 88 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_INST_LITERAL has_lit 88 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_VOP_SDWA_SDST_ENC has_sdwa 88 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 FMT_NUM_M64 OPR_SDST_EXEC 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMPX_NGE_F32 Set the per-lane condition code to 1 iff the first input is not greater than or equal to the second input. Store the result into the EXEC mask and to VCC or a scalar register. ENC_VOP3 default 89 VDST FMT_NUM_B64 OPR_SDST 64 SRC0 FMT_NUM_F32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_F32 OPR_SRC_SIMPLE 32 FMT_NUM_M64 OPR_SDST_EXEC 64 ENC_VOPC default 89 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_INST_LITERAL has_lit 89 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_VOP_SDWA_SDST_ENC has_sdwa 89 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 FMT_NUM_M64 OPR_SDST_EXEC 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMPX_NLG_F32 Set the per-lane condition code to 1 iff the first input is not less than or greater than the second input. Store the result into the EXEC mask and to VCC or a scalar register. ENC_VOP3 default 90 VDST FMT_NUM_B64 OPR_SDST 64 SRC0 FMT_NUM_F32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_F32 OPR_SRC_SIMPLE 32 FMT_NUM_M64 OPR_SDST_EXEC 64 ENC_VOPC default 90 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_INST_LITERAL has_lit 90 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_VOP_SDWA_SDST_ENC has_sdwa 90 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 FMT_NUM_M64 OPR_SDST_EXEC 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMPX_NGT_F32 Set the per-lane condition code to 1 iff the first input is not greater than the second input. Store the result into the EXEC mask and to VCC or a scalar register. ENC_VOP3 default 91 VDST FMT_NUM_B64 OPR_SDST 64 SRC0 FMT_NUM_F32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_F32 OPR_SRC_SIMPLE 32 FMT_NUM_M64 OPR_SDST_EXEC 64 ENC_VOPC default 91 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_INST_LITERAL has_lit 91 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_VOP_SDWA_SDST_ENC has_sdwa 91 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 FMT_NUM_M64 OPR_SDST_EXEC 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMPX_NLE_F32 Set the per-lane condition code to 1 iff the first input is not less than or equal to the second input. Store the result into the EXEC mask and to VCC or a scalar register. ENC_VOP3 default 92 VDST FMT_NUM_B64 OPR_SDST 64 SRC0 FMT_NUM_F32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_F32 OPR_SRC_SIMPLE 32 FMT_NUM_M64 OPR_SDST_EXEC 64 ENC_VOPC default 92 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_INST_LITERAL has_lit 92 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_VOP_SDWA_SDST_ENC has_sdwa 92 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 FMT_NUM_M64 OPR_SDST_EXEC 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMPX_NEQ_F32 Set the per-lane condition code to 1 iff the first input is not equal to the second input. Store the result into the EXEC mask and to VCC or a scalar register. ENC_VOP3 default 93 VDST FMT_NUM_B64 OPR_SDST 64 SRC0 FMT_NUM_F32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_F32 OPR_SRC_SIMPLE 32 FMT_NUM_M64 OPR_SDST_EXEC 64 ENC_VOPC default 93 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_INST_LITERAL has_lit 93 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_VOP_SDWA_SDST_ENC has_sdwa 93 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 FMT_NUM_M64 OPR_SDST_EXEC 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMPX_NLT_F32 Set the per-lane condition code to 1 iff the first input is not less than the second input. Store the result into the EXEC mask and to VCC or a scalar register. ENC_VOP3 default 94 VDST FMT_NUM_B64 OPR_SDST 64 SRC0 FMT_NUM_F32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_F32 OPR_SRC_SIMPLE 32 FMT_NUM_M64 OPR_SDST_EXEC 64 ENC_VOPC default 94 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_INST_LITERAL has_lit 94 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_VOP_SDWA_SDST_ENC has_sdwa 94 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 FMT_NUM_M64 OPR_SDST_EXEC 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMPX_TRU_F32 Set the per-lane condition code to 1. Store the result into the EXEC mask and to VCC or a scalar register. ENC_VOP3 default 95 VDST FMT_NUM_B64 OPR_SDST 64 SRC0 FMT_NUM_F32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_F32 OPR_SRC_SIMPLE 32 FMT_NUM_M64 OPR_SDST_EXEC 64 ENC_VOPC default 95 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_INST_LITERAL has_lit 95 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_VOP_SDWA_SDST_ENC has_sdwa 95 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_F32 OPR_SRC 32 VSRC1 FMT_NUM_F32 OPR_VGPR 32 FMT_NUM_M64 OPR_SDST_EXEC 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMP_F_F64 Set the per-lane condition code to 0. Store the result into VCC or a scalar register. ENC_VOP3 default 96 VDST FMT_NUM_B64 OPR_SREG 64 SRC0 FMT_NUM_F64 OPR_SRC_NOLIT 64 SRC1 FMT_NUM_F64 OPR_SRC_SIMPLE 64 ENC_VOPC default 96 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_F64 OPR_SRC 64 VSRC1 FMT_NUM_F64 OPR_VGPR 64 VOPC_INST_LITERAL has_lit 96 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_F64 OPR_SRC 64 VSRC1 FMT_NUM_F64 OPR_VGPR 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMP_LT_F64 Set the per-lane condition code to 1 iff the first input is less than the second input. Store the result into VCC or a scalar register. ENC_VOP3 default 97 VDST FMT_NUM_B64 OPR_SREG 64 SRC0 FMT_NUM_F64 OPR_SRC_NOLIT 64 SRC1 FMT_NUM_F64 OPR_SRC_SIMPLE 64 ENC_VOPC default 97 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_F64 OPR_SRC 64 VSRC1 FMT_NUM_F64 OPR_VGPR 64 VOPC_INST_LITERAL has_lit 97 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_F64 OPR_SRC 64 VSRC1 FMT_NUM_F64 OPR_VGPR 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMP_EQ_F64 Set the per-lane condition code to 1 iff the first input is equal to the second input. Store the result into VCC or a scalar register. ENC_VOP3 default 98 VDST FMT_NUM_B64 OPR_SREG 64 SRC0 FMT_NUM_F64 OPR_SRC_NOLIT 64 SRC1 FMT_NUM_F64 OPR_SRC_SIMPLE 64 ENC_VOPC default 98 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_F64 OPR_SRC 64 VSRC1 FMT_NUM_F64 OPR_VGPR 64 VOPC_INST_LITERAL has_lit 98 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_F64 OPR_SRC 64 VSRC1 FMT_NUM_F64 OPR_VGPR 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMP_LE_F64 Set the per-lane condition code to 1 iff the first input is less than or equal to the second input. Store the result into VCC or a scalar register. ENC_VOP3 default 99 VDST FMT_NUM_B64 OPR_SREG 64 SRC0 FMT_NUM_F64 OPR_SRC_NOLIT 64 SRC1 FMT_NUM_F64 OPR_SRC_SIMPLE 64 ENC_VOPC default 99 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_F64 OPR_SRC 64 VSRC1 FMT_NUM_F64 OPR_VGPR 64 VOPC_INST_LITERAL has_lit 99 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_F64 OPR_SRC 64 VSRC1 FMT_NUM_F64 OPR_VGPR 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMP_GT_F64 Set the per-lane condition code to 1 iff the first input is greater than the second input. Store the result into VCC or a scalar register. ENC_VOP3 default 100 VDST FMT_NUM_B64 OPR_SREG 64 SRC0 FMT_NUM_F64 OPR_SRC_NOLIT 64 SRC1 FMT_NUM_F64 OPR_SRC_SIMPLE 64 ENC_VOPC default 100 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_F64 OPR_SRC 64 VSRC1 FMT_NUM_F64 OPR_VGPR 64 VOPC_INST_LITERAL has_lit 100 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_F64 OPR_SRC 64 VSRC1 FMT_NUM_F64 OPR_VGPR 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMP_LG_F64 Set the per-lane condition code to 1 iff the first input is less than or greater than the second input. Store the result into VCC or a scalar register. ENC_VOP3 default 101 VDST FMT_NUM_B64 OPR_SREG 64 SRC0 FMT_NUM_F64 OPR_SRC_NOLIT 64 SRC1 FMT_NUM_F64 OPR_SRC_SIMPLE 64 ENC_VOPC default 101 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_F64 OPR_SRC 64 VSRC1 FMT_NUM_F64 OPR_VGPR 64 VOPC_INST_LITERAL has_lit 101 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_F64 OPR_SRC 64 VSRC1 FMT_NUM_F64 OPR_VGPR 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMP_GE_F64 Set the per-lane condition code to 1 iff the first input is greater than or equal to the second input. Store the result into VCC or a scalar register. ENC_VOP3 default 102 VDST FMT_NUM_B64 OPR_SREG 64 SRC0 FMT_NUM_F64 OPR_SRC_NOLIT 64 SRC1 FMT_NUM_F64 OPR_SRC_SIMPLE 64 ENC_VOPC default 102 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_F64 OPR_SRC 64 VSRC1 FMT_NUM_F64 OPR_VGPR 64 VOPC_INST_LITERAL has_lit 102 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_F64 OPR_SRC 64 VSRC1 FMT_NUM_F64 OPR_VGPR 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMP_O_F64 Set the per-lane condition code to 1 iff the first input is orderable to the second input. Store the result into VCC or a scalar register. ENC_VOP3 default 103 VDST FMT_NUM_B64 OPR_SREG 64 SRC0 FMT_NUM_F64 OPR_SRC_NOLIT 64 SRC1 FMT_NUM_F64 OPR_SRC_SIMPLE 64 ENC_VOPC default 103 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_F64 OPR_SRC 64 VSRC1 FMT_NUM_F64 OPR_VGPR 64 VOPC_INST_LITERAL has_lit 103 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_F64 OPR_SRC 64 VSRC1 FMT_NUM_F64 OPR_VGPR 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMP_U_F64 Set the per-lane condition code to 1 iff the first input is not orderable to the second input. Store the result into VCC or a scalar register. ENC_VOP3 default 104 VDST FMT_NUM_B64 OPR_SREG 64 SRC0 FMT_NUM_F64 OPR_SRC_NOLIT 64 SRC1 FMT_NUM_F64 OPR_SRC_SIMPLE 64 ENC_VOPC default 104 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_F64 OPR_SRC 64 VSRC1 FMT_NUM_F64 OPR_VGPR 64 VOPC_INST_LITERAL has_lit 104 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_F64 OPR_SRC 64 VSRC1 FMT_NUM_F64 OPR_VGPR 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMP_NGE_F64 Set the per-lane condition code to 1 iff the first input is not greater than or equal to the second input. Store the result into VCC or a scalar register. ENC_VOP3 default 105 VDST FMT_NUM_B64 OPR_SREG 64 SRC0 FMT_NUM_F64 OPR_SRC_NOLIT 64 SRC1 FMT_NUM_F64 OPR_SRC_SIMPLE 64 ENC_VOPC default 105 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_F64 OPR_SRC 64 VSRC1 FMT_NUM_F64 OPR_VGPR 64 VOPC_INST_LITERAL has_lit 105 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_F64 OPR_SRC 64 VSRC1 FMT_NUM_F64 OPR_VGPR 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMP_NLG_F64 Set the per-lane condition code to 1 iff the first input is not less than or greater than the second input. Store the result into VCC or a scalar register. ENC_VOP3 default 106 VDST FMT_NUM_B64 OPR_SREG 64 SRC0 FMT_NUM_F64 OPR_SRC_NOLIT 64 SRC1 FMT_NUM_F64 OPR_SRC_SIMPLE 64 ENC_VOPC default 106 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_F64 OPR_SRC 64 VSRC1 FMT_NUM_F64 OPR_VGPR 64 VOPC_INST_LITERAL has_lit 106 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_F64 OPR_SRC 64 VSRC1 FMT_NUM_F64 OPR_VGPR 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMP_NGT_F64 Set the per-lane condition code to 1 iff the first input is not greater than the second input. Store the result into VCC or a scalar register. ENC_VOP3 default 107 VDST FMT_NUM_B64 OPR_SREG 64 SRC0 FMT_NUM_F64 OPR_SRC_NOLIT 64 SRC1 FMT_NUM_F64 OPR_SRC_SIMPLE 64 ENC_VOPC default 107 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_F64 OPR_SRC 64 VSRC1 FMT_NUM_F64 OPR_VGPR 64 VOPC_INST_LITERAL has_lit 107 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_F64 OPR_SRC 64 VSRC1 FMT_NUM_F64 OPR_VGPR 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMP_NLE_F64 Set the per-lane condition code to 1 iff the first input is not less than or equal to the second input. Store the result into VCC or a scalar register. ENC_VOP3 default 108 VDST FMT_NUM_B64 OPR_SREG 64 SRC0 FMT_NUM_F64 OPR_SRC_NOLIT 64 SRC1 FMT_NUM_F64 OPR_SRC_SIMPLE 64 ENC_VOPC default 108 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_F64 OPR_SRC 64 VSRC1 FMT_NUM_F64 OPR_VGPR 64 VOPC_INST_LITERAL has_lit 108 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_F64 OPR_SRC 64 VSRC1 FMT_NUM_F64 OPR_VGPR 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMP_NEQ_F64 Set the per-lane condition code to 1 iff the first input is not equal to the second input. Store the result into VCC or a scalar register. ENC_VOP3 default 109 VDST FMT_NUM_B64 OPR_SREG 64 SRC0 FMT_NUM_F64 OPR_SRC_NOLIT 64 SRC1 FMT_NUM_F64 OPR_SRC_SIMPLE 64 ENC_VOPC default 109 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_F64 OPR_SRC 64 VSRC1 FMT_NUM_F64 OPR_VGPR 64 VOPC_INST_LITERAL has_lit 109 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_F64 OPR_SRC 64 VSRC1 FMT_NUM_F64 OPR_VGPR 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMP_NLT_F64 Set the per-lane condition code to 1 iff the first input is not less than the second input. Store the result into VCC or a scalar register. ENC_VOP3 default 110 VDST FMT_NUM_B64 OPR_SREG 64 SRC0 FMT_NUM_F64 OPR_SRC_NOLIT 64 SRC1 FMT_NUM_F64 OPR_SRC_SIMPLE 64 ENC_VOPC default 110 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_F64 OPR_SRC 64 VSRC1 FMT_NUM_F64 OPR_VGPR 64 VOPC_INST_LITERAL has_lit 110 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_F64 OPR_SRC 64 VSRC1 FMT_NUM_F64 OPR_VGPR 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMP_TRU_F64 Set the per-lane condition code to 1. Store the result into VCC or a scalar register. ENC_VOP3 default 111 VDST FMT_NUM_B64 OPR_SREG 64 SRC0 FMT_NUM_F64 OPR_SRC_NOLIT 64 SRC1 FMT_NUM_F64 OPR_SRC_SIMPLE 64 ENC_VOPC default 111 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_F64 OPR_SRC 64 VSRC1 FMT_NUM_F64 OPR_VGPR 64 VOPC_INST_LITERAL has_lit 111 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_F64 OPR_SRC 64 VSRC1 FMT_NUM_F64 OPR_VGPR 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMPX_F_F64 Set the per-lane condition code to 0. Store the result into the EXEC mask and to VCC or a scalar register. ENC_VOP3 default 112 VDST FMT_NUM_B64 OPR_SDST 64 SRC0 FMT_NUM_F64 OPR_SRC_NOLIT 64 SRC1 FMT_NUM_F64 OPR_SRC_SIMPLE 64 FMT_NUM_M64 OPR_SDST_EXEC 64 ENC_VOPC default 112 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_F64 OPR_SRC 64 VSRC1 FMT_NUM_F64 OPR_VGPR 64 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_INST_LITERAL has_lit 112 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_F64 OPR_SRC 64 VSRC1 FMT_NUM_F64 OPR_VGPR 64 FMT_NUM_M64 OPR_SDST_EXEC 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMPX_LT_F64 Set the per-lane condition code to 1 iff the first input is less than the second input. Store the result into the EXEC mask and to VCC or a scalar register. ENC_VOP3 default 113 VDST FMT_NUM_B64 OPR_SDST 64 SRC0 FMT_NUM_F64 OPR_SRC_NOLIT 64 SRC1 FMT_NUM_F64 OPR_SRC_SIMPLE 64 FMT_NUM_M64 OPR_SDST_EXEC 64 ENC_VOPC default 113 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_F64 OPR_SRC 64 VSRC1 FMT_NUM_F64 OPR_VGPR 64 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_INST_LITERAL has_lit 113 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_F64 OPR_SRC 64 VSRC1 FMT_NUM_F64 OPR_VGPR 64 FMT_NUM_M64 OPR_SDST_EXEC 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMPX_EQ_F64 Set the per-lane condition code to 1 iff the first input is equal to the second input. Store the result into the EXEC mask and to VCC or a scalar register. ENC_VOP3 default 114 VDST FMT_NUM_B64 OPR_SDST 64 SRC0 FMT_NUM_F64 OPR_SRC_NOLIT 64 SRC1 FMT_NUM_F64 OPR_SRC_SIMPLE 64 FMT_NUM_M64 OPR_SDST_EXEC 64 ENC_VOPC default 114 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_F64 OPR_SRC 64 VSRC1 FMT_NUM_F64 OPR_VGPR 64 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_INST_LITERAL has_lit 114 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_F64 OPR_SRC 64 VSRC1 FMT_NUM_F64 OPR_VGPR 64 FMT_NUM_M64 OPR_SDST_EXEC 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMPX_LE_F64 Set the per-lane condition code to 1 iff the first input is less than or equal to the second input. Store the result into the EXEC mask and to VCC or a scalar register. ENC_VOP3 default 115 VDST FMT_NUM_B64 OPR_SDST 64 SRC0 FMT_NUM_F64 OPR_SRC_NOLIT 64 SRC1 FMT_NUM_F64 OPR_SRC_SIMPLE 64 FMT_NUM_M64 OPR_SDST_EXEC 64 ENC_VOPC default 115 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_F64 OPR_SRC 64 VSRC1 FMT_NUM_F64 OPR_VGPR 64 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_INST_LITERAL has_lit 115 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_F64 OPR_SRC 64 VSRC1 FMT_NUM_F64 OPR_VGPR 64 FMT_NUM_M64 OPR_SDST_EXEC 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMPX_GT_F64 Set the per-lane condition code to 1 iff the first input is greater than the second input. Store the result into the EXEC mask and to VCC or a scalar register. ENC_VOP3 default 116 VDST FMT_NUM_B64 OPR_SDST 64 SRC0 FMT_NUM_F64 OPR_SRC_NOLIT 64 SRC1 FMT_NUM_F64 OPR_SRC_SIMPLE 64 FMT_NUM_M64 OPR_SDST_EXEC 64 ENC_VOPC default 116 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_F64 OPR_SRC 64 VSRC1 FMT_NUM_F64 OPR_VGPR 64 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_INST_LITERAL has_lit 116 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_F64 OPR_SRC 64 VSRC1 FMT_NUM_F64 OPR_VGPR 64 FMT_NUM_M64 OPR_SDST_EXEC 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMPX_LG_F64 Set the per-lane condition code to 1 iff the first input is less than or greater than the second input. Store the result into the EXEC mask and to VCC or a scalar register. ENC_VOP3 default 117 VDST FMT_NUM_B64 OPR_SDST 64 SRC0 FMT_NUM_F64 OPR_SRC_NOLIT 64 SRC1 FMT_NUM_F64 OPR_SRC_SIMPLE 64 FMT_NUM_M64 OPR_SDST_EXEC 64 ENC_VOPC default 117 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_F64 OPR_SRC 64 VSRC1 FMT_NUM_F64 OPR_VGPR 64 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_INST_LITERAL has_lit 117 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_F64 OPR_SRC 64 VSRC1 FMT_NUM_F64 OPR_VGPR 64 FMT_NUM_M64 OPR_SDST_EXEC 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMPX_GE_F64 Set the per-lane condition code to 1 iff the first input is greater than or equal to the second input. Store the result into the EXEC mask and to VCC or a scalar register. ENC_VOP3 default 118 VDST FMT_NUM_B64 OPR_SDST 64 SRC0 FMT_NUM_F64 OPR_SRC_NOLIT 64 SRC1 FMT_NUM_F64 OPR_SRC_SIMPLE 64 FMT_NUM_M64 OPR_SDST_EXEC 64 ENC_VOPC default 118 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_F64 OPR_SRC 64 VSRC1 FMT_NUM_F64 OPR_VGPR 64 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_INST_LITERAL has_lit 118 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_F64 OPR_SRC 64 VSRC1 FMT_NUM_F64 OPR_VGPR 64 FMT_NUM_M64 OPR_SDST_EXEC 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMPX_O_F64 Set the per-lane condition code to 1 iff the first input is orderable to the second input. Store the result into the EXEC mask and to VCC or a scalar register. ENC_VOP3 default 119 VDST FMT_NUM_B64 OPR_SDST 64 SRC0 FMT_NUM_F64 OPR_SRC_NOLIT 64 SRC1 FMT_NUM_F64 OPR_SRC_SIMPLE 64 FMT_NUM_M64 OPR_SDST_EXEC 64 ENC_VOPC default 119 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_F64 OPR_SRC 64 VSRC1 FMT_NUM_F64 OPR_VGPR 64 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_INST_LITERAL has_lit 119 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_F64 OPR_SRC 64 VSRC1 FMT_NUM_F64 OPR_VGPR 64 FMT_NUM_M64 OPR_SDST_EXEC 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMPX_U_F64 Set the per-lane condition code to 1 iff the first input is not orderable to the second input. Store the result into the EXEC mask and to VCC or a scalar register. ENC_VOP3 default 120 VDST FMT_NUM_B64 OPR_SDST 64 SRC0 FMT_NUM_F64 OPR_SRC_NOLIT 64 SRC1 FMT_NUM_F64 OPR_SRC_SIMPLE 64 FMT_NUM_M64 OPR_SDST_EXEC 64 ENC_VOPC default 120 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_F64 OPR_SRC 64 VSRC1 FMT_NUM_F64 OPR_VGPR 64 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_INST_LITERAL has_lit 120 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_F64 OPR_SRC 64 VSRC1 FMT_NUM_F64 OPR_VGPR 64 FMT_NUM_M64 OPR_SDST_EXEC 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMPX_NGE_F64 Set the per-lane condition code to 1 iff the first input is not greater than or equal to the second input. Store the result into the EXEC mask and to VCC or a scalar register. ENC_VOP3 default 121 VDST FMT_NUM_B64 OPR_SDST 64 SRC0 FMT_NUM_F64 OPR_SRC_NOLIT 64 SRC1 FMT_NUM_F64 OPR_SRC_SIMPLE 64 FMT_NUM_M64 OPR_SDST_EXEC 64 ENC_VOPC default 121 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_F64 OPR_SRC 64 VSRC1 FMT_NUM_F64 OPR_VGPR 64 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_INST_LITERAL has_lit 121 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_F64 OPR_SRC 64 VSRC1 FMT_NUM_F64 OPR_VGPR 64 FMT_NUM_M64 OPR_SDST_EXEC 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMPX_NLG_F64 Set the per-lane condition code to 1 iff the first input is not less than or greater than the second input. Store the result into the EXEC mask and to VCC or a scalar register. ENC_VOP3 default 122 VDST FMT_NUM_B64 OPR_SDST 64 SRC0 FMT_NUM_F64 OPR_SRC_NOLIT 64 SRC1 FMT_NUM_F64 OPR_SRC_SIMPLE 64 FMT_NUM_M64 OPR_SDST_EXEC 64 ENC_VOPC default 122 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_F64 OPR_SRC 64 VSRC1 FMT_NUM_F64 OPR_VGPR 64 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_INST_LITERAL has_lit 122 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_F64 OPR_SRC 64 VSRC1 FMT_NUM_F64 OPR_VGPR 64 FMT_NUM_M64 OPR_SDST_EXEC 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMPX_NGT_F64 Set the per-lane condition code to 1 iff the first input is not greater than the second input. Store the result into the EXEC mask and to VCC or a scalar register. ENC_VOP3 default 123 VDST FMT_NUM_B64 OPR_SDST 64 SRC0 FMT_NUM_F64 OPR_SRC_NOLIT 64 SRC1 FMT_NUM_F64 OPR_SRC_SIMPLE 64 FMT_NUM_M64 OPR_SDST_EXEC 64 ENC_VOPC default 123 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_F64 OPR_SRC 64 VSRC1 FMT_NUM_F64 OPR_VGPR 64 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_INST_LITERAL has_lit 123 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_F64 OPR_SRC 64 VSRC1 FMT_NUM_F64 OPR_VGPR 64 FMT_NUM_M64 OPR_SDST_EXEC 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMPX_NLE_F64 Set the per-lane condition code to 1 iff the first input is not less than or equal to the second input. Store the result into the EXEC mask and to VCC or a scalar register. ENC_VOP3 default 124 VDST FMT_NUM_B64 OPR_SDST 64 SRC0 FMT_NUM_F64 OPR_SRC_NOLIT 64 SRC1 FMT_NUM_F64 OPR_SRC_SIMPLE 64 FMT_NUM_M64 OPR_SDST_EXEC 64 ENC_VOPC default 124 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_F64 OPR_SRC 64 VSRC1 FMT_NUM_F64 OPR_VGPR 64 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_INST_LITERAL has_lit 124 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_F64 OPR_SRC 64 VSRC1 FMT_NUM_F64 OPR_VGPR 64 FMT_NUM_M64 OPR_SDST_EXEC 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMPX_NEQ_F64 Set the per-lane condition code to 1 iff the first input is not equal to the second input. Store the result into the EXEC mask and to VCC or a scalar register. ENC_VOP3 default 125 VDST FMT_NUM_B64 OPR_SDST 64 SRC0 FMT_NUM_F64 OPR_SRC_NOLIT 64 SRC1 FMT_NUM_F64 OPR_SRC_SIMPLE 64 FMT_NUM_M64 OPR_SDST_EXEC 64 ENC_VOPC default 125 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_F64 OPR_SRC 64 VSRC1 FMT_NUM_F64 OPR_VGPR 64 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_INST_LITERAL has_lit 125 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_F64 OPR_SRC 64 VSRC1 FMT_NUM_F64 OPR_VGPR 64 FMT_NUM_M64 OPR_SDST_EXEC 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMPX_NLT_F64 Set the per-lane condition code to 1 iff the first input is not less than the second input. Store the result into the EXEC mask and to VCC or a scalar register. ENC_VOP3 default 126 VDST FMT_NUM_B64 OPR_SDST 64 SRC0 FMT_NUM_F64 OPR_SRC_NOLIT 64 SRC1 FMT_NUM_F64 OPR_SRC_SIMPLE 64 FMT_NUM_M64 OPR_SDST_EXEC 64 ENC_VOPC default 126 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_F64 OPR_SRC 64 VSRC1 FMT_NUM_F64 OPR_VGPR 64 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_INST_LITERAL has_lit 126 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_F64 OPR_SRC 64 VSRC1 FMT_NUM_F64 OPR_VGPR 64 FMT_NUM_M64 OPR_SDST_EXEC 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMPX_TRU_F64 Set the per-lane condition code to 1. Store the result into the EXEC mask and to VCC or a scalar register. ENC_VOP3 default 127 VDST FMT_NUM_B64 OPR_SDST 64 SRC0 FMT_NUM_F64 OPR_SRC_NOLIT 64 SRC1 FMT_NUM_F64 OPR_SRC_SIMPLE 64 FMT_NUM_M64 OPR_SDST_EXEC 64 ENC_VOPC default 127 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_F64 OPR_SRC 64 VSRC1 FMT_NUM_F64 OPR_VGPR 64 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_INST_LITERAL has_lit 127 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_F64 OPR_SRC 64 VSRC1 FMT_NUM_F64 OPR_VGPR 64 FMT_NUM_M64 OPR_SDST_EXEC 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMP_F_I16 Set the per-lane condition code to 0. Store the result into VCC or a scalar register. ENC_VOP3 default 160 VDST FMT_NUM_B64 OPR_SREG 64 SRC0 FMT_NUM_I16 OPR_SRC_NOLIT 16 SRC1 FMT_NUM_I16 OPR_SRC_SIMPLE 16 ENC_VOPC default 160 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_I16 OPR_SRC 16 VSRC1 FMT_NUM_I16 OPR_VGPR 16 VOPC_INST_LITERAL has_lit 160 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_I16 OPR_SRC 16 VSRC1 FMT_NUM_I16 OPR_VGPR 16 VOPC_VOP_SDWA_SDST_ENC has_sdwa 160 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_I16 OPR_SRC 16 VSRC1 FMT_NUM_I16 OPR_VGPR 16 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMP_LT_I16 Set the per-lane condition code to 1 iff the first input is less than the second input. Store the result into VCC or a scalar register. ENC_VOP3 default 161 VDST FMT_NUM_B64 OPR_SREG 64 SRC0 FMT_NUM_I16 OPR_SRC_NOLIT 16 SRC1 FMT_NUM_I16 OPR_SRC_SIMPLE 16 ENC_VOPC default 161 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_I16 OPR_SRC 16 VSRC1 FMT_NUM_I16 OPR_VGPR 16 VOPC_INST_LITERAL has_lit 161 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_I16 OPR_SRC 16 VSRC1 FMT_NUM_I16 OPR_VGPR 16 VOPC_VOP_SDWA_SDST_ENC has_sdwa 161 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_I16 OPR_SRC 16 VSRC1 FMT_NUM_I16 OPR_VGPR 16 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMP_EQ_I16 Set the per-lane condition code to 1 iff the first input is equal to the second input. Store the result into VCC or a scalar register. ENC_VOP3 default 162 VDST FMT_NUM_B64 OPR_SREG 64 SRC0 FMT_NUM_I16 OPR_SRC_NOLIT 16 SRC1 FMT_NUM_I16 OPR_SRC_SIMPLE 16 ENC_VOPC default 162 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_I16 OPR_SRC 16 VSRC1 FMT_NUM_I16 OPR_VGPR 16 VOPC_INST_LITERAL has_lit 162 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_I16 OPR_SRC 16 VSRC1 FMT_NUM_I16 OPR_VGPR 16 VOPC_VOP_SDWA_SDST_ENC has_sdwa 162 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_I16 OPR_SRC 16 VSRC1 FMT_NUM_I16 OPR_VGPR 16 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMP_LE_I16 Set the per-lane condition code to 1 iff the first input is less than or equal to the second input. Store the result into VCC or a scalar register. ENC_VOP3 default 163 VDST FMT_NUM_B64 OPR_SREG 64 SRC0 FMT_NUM_I16 OPR_SRC_NOLIT 16 SRC1 FMT_NUM_I16 OPR_SRC_SIMPLE 16 ENC_VOPC default 163 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_I16 OPR_SRC 16 VSRC1 FMT_NUM_I16 OPR_VGPR 16 VOPC_INST_LITERAL has_lit 163 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_I16 OPR_SRC 16 VSRC1 FMT_NUM_I16 OPR_VGPR 16 VOPC_VOP_SDWA_SDST_ENC has_sdwa 163 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_I16 OPR_SRC 16 VSRC1 FMT_NUM_I16 OPR_VGPR 16 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMP_GT_I16 Set the per-lane condition code to 1 iff the first input is greater than the second input. Store the result into VCC or a scalar register. ENC_VOP3 default 164 VDST FMT_NUM_B64 OPR_SREG 64 SRC0 FMT_NUM_I16 OPR_SRC_NOLIT 16 SRC1 FMT_NUM_I16 OPR_SRC_SIMPLE 16 ENC_VOPC default 164 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_I16 OPR_SRC 16 VSRC1 FMT_NUM_I16 OPR_VGPR 16 VOPC_INST_LITERAL has_lit 164 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_I16 OPR_SRC 16 VSRC1 FMT_NUM_I16 OPR_VGPR 16 VOPC_VOP_SDWA_SDST_ENC has_sdwa 164 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_I16 OPR_SRC 16 VSRC1 FMT_NUM_I16 OPR_VGPR 16 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMP_NE_I16 Set the per-lane condition code to 1 iff the first input is not equal to the second input. Store the result into VCC or a scalar register. ENC_VOP3 default 165 VDST FMT_NUM_B64 OPR_SREG 64 SRC0 FMT_NUM_I16 OPR_SRC_NOLIT 16 SRC1 FMT_NUM_I16 OPR_SRC_SIMPLE 16 ENC_VOPC default 165 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_I16 OPR_SRC 16 VSRC1 FMT_NUM_I16 OPR_VGPR 16 VOPC_INST_LITERAL has_lit 165 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_I16 OPR_SRC 16 VSRC1 FMT_NUM_I16 OPR_VGPR 16 VOPC_VOP_SDWA_SDST_ENC has_sdwa 165 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_I16 OPR_SRC 16 VSRC1 FMT_NUM_I16 OPR_VGPR 16 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMP_GE_I16 Set the per-lane condition code to 1 iff the first input is greater than or equal to the second input. Store the result into VCC or a scalar register. ENC_VOP3 default 166 VDST FMT_NUM_B64 OPR_SREG 64 SRC0 FMT_NUM_I16 OPR_SRC_NOLIT 16 SRC1 FMT_NUM_I16 OPR_SRC_SIMPLE 16 ENC_VOPC default 166 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_I16 OPR_SRC 16 VSRC1 FMT_NUM_I16 OPR_VGPR 16 VOPC_INST_LITERAL has_lit 166 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_I16 OPR_SRC 16 VSRC1 FMT_NUM_I16 OPR_VGPR 16 VOPC_VOP_SDWA_SDST_ENC has_sdwa 166 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_I16 OPR_SRC 16 VSRC1 FMT_NUM_I16 OPR_VGPR 16 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMP_T_I16 Set the per-lane condition code to 1. Store the result into VCC or a scalar register. ENC_VOP3 default 167 VDST FMT_NUM_B64 OPR_SREG 64 SRC0 FMT_NUM_I16 OPR_SRC_NOLIT 16 SRC1 FMT_NUM_I16 OPR_SRC_SIMPLE 16 ENC_VOPC default 167 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_I16 OPR_SRC 16 VSRC1 FMT_NUM_I16 OPR_VGPR 16 VOPC_INST_LITERAL has_lit 167 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_I16 OPR_SRC 16 VSRC1 FMT_NUM_I16 OPR_VGPR 16 VOPC_VOP_SDWA_SDST_ENC has_sdwa 167 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_I16 OPR_SRC 16 VSRC1 FMT_NUM_I16 OPR_VGPR 16 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMP_F_U16 Set the per-lane condition code to 0. Store the result into VCC or a scalar register. ENC_VOP3 default 168 VDST FMT_NUM_B64 OPR_SREG 64 SRC0 FMT_NUM_U16 OPR_SRC_NOLIT 16 SRC1 FMT_NUM_U16 OPR_SRC_SIMPLE 16 ENC_VOPC default 168 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_U16 OPR_SRC 16 VSRC1 FMT_NUM_U16 OPR_VGPR 16 VOPC_INST_LITERAL has_lit 168 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_U16 OPR_SRC 16 VSRC1 FMT_NUM_U16 OPR_VGPR 16 VOPC_VOP_SDWA_SDST_ENC has_sdwa 168 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_U16 OPR_SRC 16 VSRC1 FMT_NUM_U16 OPR_VGPR 16 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMP_LT_U16 Set the per-lane condition code to 1 iff the first input is less than the second input. Store the result into VCC or a scalar register. ENC_VOP3 default 169 VDST FMT_NUM_B64 OPR_SREG 64 SRC0 FMT_NUM_U16 OPR_SRC_NOLIT 16 SRC1 FMT_NUM_U16 OPR_SRC_SIMPLE 16 ENC_VOPC default 169 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_U16 OPR_SRC 16 VSRC1 FMT_NUM_U16 OPR_VGPR 16 VOPC_INST_LITERAL has_lit 169 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_U16 OPR_SRC 16 VSRC1 FMT_NUM_U16 OPR_VGPR 16 VOPC_VOP_SDWA_SDST_ENC has_sdwa 169 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_U16 OPR_SRC 16 VSRC1 FMT_NUM_U16 OPR_VGPR 16 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMP_EQ_U16 Set the per-lane condition code to 1 iff the first input is equal to the second input. Store the result into VCC or a scalar register. ENC_VOP3 default 170 VDST FMT_NUM_B64 OPR_SREG 64 SRC0 FMT_NUM_U16 OPR_SRC_NOLIT 16 SRC1 FMT_NUM_U16 OPR_SRC_SIMPLE 16 ENC_VOPC default 170 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_U16 OPR_SRC 16 VSRC1 FMT_NUM_U16 OPR_VGPR 16 VOPC_INST_LITERAL has_lit 170 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_U16 OPR_SRC 16 VSRC1 FMT_NUM_U16 OPR_VGPR 16 VOPC_VOP_SDWA_SDST_ENC has_sdwa 170 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_U16 OPR_SRC 16 VSRC1 FMT_NUM_U16 OPR_VGPR 16 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMP_LE_U16 Set the per-lane condition code to 1 iff the first input is less than or equal to the second input. Store the result into VCC or a scalar register. ENC_VOP3 default 171 VDST FMT_NUM_B64 OPR_SREG 64 SRC0 FMT_NUM_U16 OPR_SRC_NOLIT 16 SRC1 FMT_NUM_U16 OPR_SRC_SIMPLE 16 ENC_VOPC default 171 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_U16 OPR_SRC 16 VSRC1 FMT_NUM_U16 OPR_VGPR 16 VOPC_INST_LITERAL has_lit 171 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_U16 OPR_SRC 16 VSRC1 FMT_NUM_U16 OPR_VGPR 16 VOPC_VOP_SDWA_SDST_ENC has_sdwa 171 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_U16 OPR_SRC 16 VSRC1 FMT_NUM_U16 OPR_VGPR 16 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMP_GT_U16 Set the per-lane condition code to 1 iff the first input is greater than the second input. Store the result into VCC or a scalar register. ENC_VOP3 default 172 VDST FMT_NUM_B64 OPR_SREG 64 SRC0 FMT_NUM_U16 OPR_SRC_NOLIT 16 SRC1 FMT_NUM_U16 OPR_SRC_SIMPLE 16 ENC_VOPC default 172 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_U16 OPR_SRC 16 VSRC1 FMT_NUM_U16 OPR_VGPR 16 VOPC_INST_LITERAL has_lit 172 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_U16 OPR_SRC 16 VSRC1 FMT_NUM_U16 OPR_VGPR 16 VOPC_VOP_SDWA_SDST_ENC has_sdwa 172 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_U16 OPR_SRC 16 VSRC1 FMT_NUM_U16 OPR_VGPR 16 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMP_NE_U16 Set the per-lane condition code to 1 iff the first input is not equal to the second input. Store the result into VCC or a scalar register. ENC_VOP3 default 173 VDST FMT_NUM_B64 OPR_SREG 64 SRC0 FMT_NUM_U16 OPR_SRC_NOLIT 16 SRC1 FMT_NUM_U16 OPR_SRC_SIMPLE 16 ENC_VOPC default 173 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_U16 OPR_SRC 16 VSRC1 FMT_NUM_U16 OPR_VGPR 16 VOPC_INST_LITERAL has_lit 173 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_U16 OPR_SRC 16 VSRC1 FMT_NUM_U16 OPR_VGPR 16 VOPC_VOP_SDWA_SDST_ENC has_sdwa 173 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_U16 OPR_SRC 16 VSRC1 FMT_NUM_U16 OPR_VGPR 16 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMP_GE_U16 Set the per-lane condition code to 1 iff the first input is greater than or equal to the second input. Store the result into VCC or a scalar register. ENC_VOP3 default 174 VDST FMT_NUM_B64 OPR_SREG 64 SRC0 FMT_NUM_U16 OPR_SRC_NOLIT 16 SRC1 FMT_NUM_U16 OPR_SRC_SIMPLE 16 ENC_VOPC default 174 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_U16 OPR_SRC 16 VSRC1 FMT_NUM_U16 OPR_VGPR 16 VOPC_INST_LITERAL has_lit 174 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_U16 OPR_SRC 16 VSRC1 FMT_NUM_U16 OPR_VGPR 16 VOPC_VOP_SDWA_SDST_ENC has_sdwa 174 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_U16 OPR_SRC 16 VSRC1 FMT_NUM_U16 OPR_VGPR 16 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMP_T_U16 Set the per-lane condition code to 1. Store the result into VCC or a scalar register. ENC_VOP3 default 175 VDST FMT_NUM_B64 OPR_SREG 64 SRC0 FMT_NUM_U16 OPR_SRC_NOLIT 16 SRC1 FMT_NUM_U16 OPR_SRC_SIMPLE 16 ENC_VOPC default 175 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_U16 OPR_SRC 16 VSRC1 FMT_NUM_U16 OPR_VGPR 16 VOPC_INST_LITERAL has_lit 175 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_U16 OPR_SRC 16 VSRC1 FMT_NUM_U16 OPR_VGPR 16 VOPC_VOP_SDWA_SDST_ENC has_sdwa 175 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_U16 OPR_SRC 16 VSRC1 FMT_NUM_U16 OPR_VGPR 16 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMPX_F_I16 Set the per-lane condition code to 0. Store the result into the EXEC mask and to VCC or a scalar register. ENC_VOP3 default 176 VDST FMT_NUM_B64 OPR_SDST 64 SRC0 FMT_NUM_I16 OPR_SRC_NOLIT 16 SRC1 FMT_NUM_I16 OPR_SRC_SIMPLE 16 FMT_NUM_M64 OPR_SDST_EXEC 64 ENC_VOPC default 176 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_I16 OPR_SRC 16 VSRC1 FMT_NUM_I16 OPR_VGPR 16 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_INST_LITERAL has_lit 176 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_I16 OPR_SRC 16 VSRC1 FMT_NUM_I16 OPR_VGPR 16 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_VOP_SDWA_SDST_ENC has_sdwa 176 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_I16 OPR_SRC 16 VSRC1 FMT_NUM_I16 OPR_VGPR 16 FMT_NUM_M64 OPR_SDST_EXEC 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMPX_LT_I16 Set the per-lane condition code to 1 iff the first input is less than the second input. Store the result into the EXEC mask and to VCC or a scalar register. ENC_VOP3 default 177 VDST FMT_NUM_B64 OPR_SDST 64 SRC0 FMT_NUM_I16 OPR_SRC_NOLIT 16 SRC1 FMT_NUM_I16 OPR_SRC_SIMPLE 16 FMT_NUM_M64 OPR_SDST_EXEC 64 ENC_VOPC default 177 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_I16 OPR_SRC 16 VSRC1 FMT_NUM_I16 OPR_VGPR 16 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_INST_LITERAL has_lit 177 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_I16 OPR_SRC 16 VSRC1 FMT_NUM_I16 OPR_VGPR 16 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_VOP_SDWA_SDST_ENC has_sdwa 177 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_I16 OPR_SRC 16 VSRC1 FMT_NUM_I16 OPR_VGPR 16 FMT_NUM_M64 OPR_SDST_EXEC 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMPX_EQ_I16 Set the per-lane condition code to 1 iff the first input is equal to the second input. Store the result into the EXEC mask and to VCC or a scalar register. ENC_VOP3 default 178 VDST FMT_NUM_B64 OPR_SDST 64 SRC0 FMT_NUM_I16 OPR_SRC_NOLIT 16 SRC1 FMT_NUM_I16 OPR_SRC_SIMPLE 16 FMT_NUM_M64 OPR_SDST_EXEC 64 ENC_VOPC default 178 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_I16 OPR_SRC 16 VSRC1 FMT_NUM_I16 OPR_VGPR 16 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_INST_LITERAL has_lit 178 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_I16 OPR_SRC 16 VSRC1 FMT_NUM_I16 OPR_VGPR 16 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_VOP_SDWA_SDST_ENC has_sdwa 178 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_I16 OPR_SRC 16 VSRC1 FMT_NUM_I16 OPR_VGPR 16 FMT_NUM_M64 OPR_SDST_EXEC 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMPX_LE_I16 Set the per-lane condition code to 1 iff the first input is less than or equal to the second input. Store the result into the EXEC mask and to VCC or a scalar register. ENC_VOP3 default 179 VDST FMT_NUM_B64 OPR_SDST 64 SRC0 FMT_NUM_I16 OPR_SRC_NOLIT 16 SRC1 FMT_NUM_I16 OPR_SRC_SIMPLE 16 FMT_NUM_M64 OPR_SDST_EXEC 64 ENC_VOPC default 179 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_I16 OPR_SRC 16 VSRC1 FMT_NUM_I16 OPR_VGPR 16 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_INST_LITERAL has_lit 179 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_I16 OPR_SRC 16 VSRC1 FMT_NUM_I16 OPR_VGPR 16 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_VOP_SDWA_SDST_ENC has_sdwa 179 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_I16 OPR_SRC 16 VSRC1 FMT_NUM_I16 OPR_VGPR 16 FMT_NUM_M64 OPR_SDST_EXEC 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMPX_GT_I16 Set the per-lane condition code to 1 iff the first input is greater than the second input. Store the result into the EXEC mask and to VCC or a scalar register. ENC_VOP3 default 180 VDST FMT_NUM_B64 OPR_SDST 64 SRC0 FMT_NUM_I16 OPR_SRC_NOLIT 16 SRC1 FMT_NUM_I16 OPR_SRC_SIMPLE 16 FMT_NUM_M64 OPR_SDST_EXEC 64 ENC_VOPC default 180 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_I16 OPR_SRC 16 VSRC1 FMT_NUM_I16 OPR_VGPR 16 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_INST_LITERAL has_lit 180 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_I16 OPR_SRC 16 VSRC1 FMT_NUM_I16 OPR_VGPR 16 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_VOP_SDWA_SDST_ENC has_sdwa 180 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_I16 OPR_SRC 16 VSRC1 FMT_NUM_I16 OPR_VGPR 16 FMT_NUM_M64 OPR_SDST_EXEC 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMPX_NE_I16 Set the per-lane condition code to 1 iff the first input is not equal to the second input. Store the result into the EXEC mask and to VCC or a scalar register. ENC_VOP3 default 181 VDST FMT_NUM_B64 OPR_SDST 64 SRC0 FMT_NUM_I16 OPR_SRC_NOLIT 16 SRC1 FMT_NUM_I16 OPR_SRC_SIMPLE 16 FMT_NUM_M64 OPR_SDST_EXEC 64 ENC_VOPC default 181 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_I16 OPR_SRC 16 VSRC1 FMT_NUM_I16 OPR_VGPR 16 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_INST_LITERAL has_lit 181 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_I16 OPR_SRC 16 VSRC1 FMT_NUM_I16 OPR_VGPR 16 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_VOP_SDWA_SDST_ENC has_sdwa 181 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_I16 OPR_SRC 16 VSRC1 FMT_NUM_I16 OPR_VGPR 16 FMT_NUM_M64 OPR_SDST_EXEC 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMPX_GE_I16 Set the per-lane condition code to 1 iff the first input is greater than or equal to the second input. Store the result into the EXEC mask and to VCC or a scalar register. ENC_VOP3 default 182 VDST FMT_NUM_B64 OPR_SDST 64 SRC0 FMT_NUM_I16 OPR_SRC_NOLIT 16 SRC1 FMT_NUM_I16 OPR_SRC_SIMPLE 16 FMT_NUM_M64 OPR_SDST_EXEC 64 ENC_VOPC default 182 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_I16 OPR_SRC 16 VSRC1 FMT_NUM_I16 OPR_VGPR 16 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_INST_LITERAL has_lit 182 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_I16 OPR_SRC 16 VSRC1 FMT_NUM_I16 OPR_VGPR 16 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_VOP_SDWA_SDST_ENC has_sdwa 182 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_I16 OPR_SRC 16 VSRC1 FMT_NUM_I16 OPR_VGPR 16 FMT_NUM_M64 OPR_SDST_EXEC 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMPX_T_I16 Set the per-lane condition code to 1. Store the result into the EXEC mask and to VCC or a scalar register. ENC_VOP3 default 183 VDST FMT_NUM_B64 OPR_SDST 64 SRC0 FMT_NUM_I16 OPR_SRC_NOLIT 16 SRC1 FMT_NUM_I16 OPR_SRC_SIMPLE 16 FMT_NUM_M64 OPR_SDST_EXEC 64 ENC_VOPC default 183 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_I16 OPR_SRC 16 VSRC1 FMT_NUM_I16 OPR_VGPR 16 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_INST_LITERAL has_lit 183 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_I16 OPR_SRC 16 VSRC1 FMT_NUM_I16 OPR_VGPR 16 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_VOP_SDWA_SDST_ENC has_sdwa 183 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_I16 OPR_SRC 16 VSRC1 FMT_NUM_I16 OPR_VGPR 16 FMT_NUM_M64 OPR_SDST_EXEC 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMPX_F_U16 Set the per-lane condition code to 0. Store the result into the EXEC mask and to VCC or a scalar register. ENC_VOP3 default 184 VDST FMT_NUM_B64 OPR_SDST 64 SRC0 FMT_NUM_U16 OPR_SRC_NOLIT 16 SRC1 FMT_NUM_U16 OPR_SRC_SIMPLE 16 FMT_NUM_M64 OPR_SDST_EXEC 64 ENC_VOPC default 184 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_U16 OPR_SRC 16 VSRC1 FMT_NUM_U16 OPR_VGPR 16 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_INST_LITERAL has_lit 184 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_U16 OPR_SRC 16 VSRC1 FMT_NUM_U16 OPR_VGPR 16 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_VOP_SDWA_SDST_ENC has_sdwa 184 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_U16 OPR_SRC 16 VSRC1 FMT_NUM_U16 OPR_VGPR 16 FMT_NUM_M64 OPR_SDST_EXEC 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMPX_LT_U16 Set the per-lane condition code to 1 iff the first input is less than the second input. Store the result into the EXEC mask and to VCC or a scalar register. ENC_VOP3 default 185 VDST FMT_NUM_B64 OPR_SDST 64 SRC0 FMT_NUM_U16 OPR_SRC_NOLIT 16 SRC1 FMT_NUM_U16 OPR_SRC_SIMPLE 16 FMT_NUM_M64 OPR_SDST_EXEC 64 ENC_VOPC default 185 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_U16 OPR_SRC 16 VSRC1 FMT_NUM_U16 OPR_VGPR 16 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_INST_LITERAL has_lit 185 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_U16 OPR_SRC 16 VSRC1 FMT_NUM_U16 OPR_VGPR 16 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_VOP_SDWA_SDST_ENC has_sdwa 185 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_U16 OPR_SRC 16 VSRC1 FMT_NUM_U16 OPR_VGPR 16 FMT_NUM_M64 OPR_SDST_EXEC 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMPX_EQ_U16 Set the per-lane condition code to 1 iff the first input is equal to the second input. Store the result into the EXEC mask and to VCC or a scalar register. ENC_VOP3 default 186 VDST FMT_NUM_B64 OPR_SDST 64 SRC0 FMT_NUM_U16 OPR_SRC_NOLIT 16 SRC1 FMT_NUM_U16 OPR_SRC_SIMPLE 16 FMT_NUM_M64 OPR_SDST_EXEC 64 ENC_VOPC default 186 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_U16 OPR_SRC 16 VSRC1 FMT_NUM_U16 OPR_VGPR 16 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_INST_LITERAL has_lit 186 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_U16 OPR_SRC 16 VSRC1 FMT_NUM_U16 OPR_VGPR 16 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_VOP_SDWA_SDST_ENC has_sdwa 186 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_U16 OPR_SRC 16 VSRC1 FMT_NUM_U16 OPR_VGPR 16 FMT_NUM_M64 OPR_SDST_EXEC 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMPX_LE_U16 Set the per-lane condition code to 1 iff the first input is less than or equal to the second input. Store the result into the EXEC mask and to VCC or a scalar register. ENC_VOP3 default 187 VDST FMT_NUM_B64 OPR_SDST 64 SRC0 FMT_NUM_U16 OPR_SRC_NOLIT 16 SRC1 FMT_NUM_U16 OPR_SRC_SIMPLE 16 FMT_NUM_M64 OPR_SDST_EXEC 64 ENC_VOPC default 187 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_U16 OPR_SRC 16 VSRC1 FMT_NUM_U16 OPR_VGPR 16 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_INST_LITERAL has_lit 187 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_U16 OPR_SRC 16 VSRC1 FMT_NUM_U16 OPR_VGPR 16 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_VOP_SDWA_SDST_ENC has_sdwa 187 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_U16 OPR_SRC 16 VSRC1 FMT_NUM_U16 OPR_VGPR 16 FMT_NUM_M64 OPR_SDST_EXEC 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMPX_GT_U16 Set the per-lane condition code to 1 iff the first input is greater than the second input. Store the result into the EXEC mask and to VCC or a scalar register. ENC_VOP3 default 188 VDST FMT_NUM_B64 OPR_SDST 64 SRC0 FMT_NUM_U16 OPR_SRC_NOLIT 16 SRC1 FMT_NUM_U16 OPR_SRC_SIMPLE 16 FMT_NUM_M64 OPR_SDST_EXEC 64 ENC_VOPC default 188 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_U16 OPR_SRC 16 VSRC1 FMT_NUM_U16 OPR_VGPR 16 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_INST_LITERAL has_lit 188 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_U16 OPR_SRC 16 VSRC1 FMT_NUM_U16 OPR_VGPR 16 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_VOP_SDWA_SDST_ENC has_sdwa 188 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_U16 OPR_SRC 16 VSRC1 FMT_NUM_U16 OPR_VGPR 16 FMT_NUM_M64 OPR_SDST_EXEC 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMPX_NE_U16 Set the per-lane condition code to 1 iff the first input is not equal to the second input. Store the result into the EXEC mask and to VCC or a scalar register. ENC_VOP3 default 189 VDST FMT_NUM_B64 OPR_SDST 64 SRC0 FMT_NUM_U16 OPR_SRC_NOLIT 16 SRC1 FMT_NUM_U16 OPR_SRC_SIMPLE 16 FMT_NUM_M64 OPR_SDST_EXEC 64 ENC_VOPC default 189 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_U16 OPR_SRC 16 VSRC1 FMT_NUM_U16 OPR_VGPR 16 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_INST_LITERAL has_lit 189 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_U16 OPR_SRC 16 VSRC1 FMT_NUM_U16 OPR_VGPR 16 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_VOP_SDWA_SDST_ENC has_sdwa 189 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_U16 OPR_SRC 16 VSRC1 FMT_NUM_U16 OPR_VGPR 16 FMT_NUM_M64 OPR_SDST_EXEC 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMPX_GE_U16 Set the per-lane condition code to 1 iff the first input is greater than or equal to the second input. Store the result into the EXEC mask and to VCC or a scalar register. ENC_VOP3 default 190 VDST FMT_NUM_B64 OPR_SDST 64 SRC0 FMT_NUM_U16 OPR_SRC_NOLIT 16 SRC1 FMT_NUM_U16 OPR_SRC_SIMPLE 16 FMT_NUM_M64 OPR_SDST_EXEC 64 ENC_VOPC default 190 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_U16 OPR_SRC 16 VSRC1 FMT_NUM_U16 OPR_VGPR 16 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_INST_LITERAL has_lit 190 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_U16 OPR_SRC 16 VSRC1 FMT_NUM_U16 OPR_VGPR 16 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_VOP_SDWA_SDST_ENC has_sdwa 190 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_U16 OPR_SRC 16 VSRC1 FMT_NUM_U16 OPR_VGPR 16 FMT_NUM_M64 OPR_SDST_EXEC 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMPX_T_U16 Set the per-lane condition code to 1. Store the result into the EXEC mask and to VCC or a scalar register. ENC_VOP3 default 191 VDST FMT_NUM_B64 OPR_SDST 64 SRC0 FMT_NUM_U16 OPR_SRC_NOLIT 16 SRC1 FMT_NUM_U16 OPR_SRC_SIMPLE 16 FMT_NUM_M64 OPR_SDST_EXEC 64 ENC_VOPC default 191 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_U16 OPR_SRC 16 VSRC1 FMT_NUM_U16 OPR_VGPR 16 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_INST_LITERAL has_lit 191 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_U16 OPR_SRC 16 VSRC1 FMT_NUM_U16 OPR_VGPR 16 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_VOP_SDWA_SDST_ENC has_sdwa 191 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_U16 OPR_SRC 16 VSRC1 FMT_NUM_U16 OPR_VGPR 16 FMT_NUM_M64 OPR_SDST_EXEC 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMP_F_I32 Set the per-lane condition code to 0. Store the result into VCC or a scalar register. ENC_VOP3 default 192 VDST FMT_NUM_B64 OPR_SREG 64 SRC0 FMT_NUM_I32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_I32 OPR_SRC_SIMPLE 32 ENC_VOPC default 192 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_I32 OPR_SRC 32 VSRC1 FMT_NUM_I32 OPR_VGPR 32 VOPC_INST_LITERAL has_lit 192 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_I32 OPR_SRC 32 VSRC1 FMT_NUM_I32 OPR_VGPR 32 VOPC_VOP_SDWA_SDST_ENC has_sdwa 192 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_I32 OPR_SRC 32 VSRC1 FMT_NUM_I32 OPR_VGPR 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMP_LT_I32 Set the per-lane condition code to 1 iff the first input is less than the second input. Store the result into VCC or a scalar register. ENC_VOP3 default 193 VDST FMT_NUM_B64 OPR_SREG 64 SRC0 FMT_NUM_I32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_I32 OPR_SRC_SIMPLE 32 ENC_VOPC default 193 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_I32 OPR_SRC 32 VSRC1 FMT_NUM_I32 OPR_VGPR 32 VOPC_INST_LITERAL has_lit 193 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_I32 OPR_SRC 32 VSRC1 FMT_NUM_I32 OPR_VGPR 32 VOPC_VOP_SDWA_SDST_ENC has_sdwa 193 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_I32 OPR_SRC 32 VSRC1 FMT_NUM_I32 OPR_VGPR 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMP_EQ_I32 Set the per-lane condition code to 1 iff the first input is equal to the second input. Store the result into VCC or a scalar register. ENC_VOP3 default 194 VDST FMT_NUM_B64 OPR_SREG 64 SRC0 FMT_NUM_I32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_I32 OPR_SRC_SIMPLE 32 ENC_VOPC default 194 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_I32 OPR_SRC 32 VSRC1 FMT_NUM_I32 OPR_VGPR 32 VOPC_INST_LITERAL has_lit 194 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_I32 OPR_SRC 32 VSRC1 FMT_NUM_I32 OPR_VGPR 32 VOPC_VOP_SDWA_SDST_ENC has_sdwa 194 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_I32 OPR_SRC 32 VSRC1 FMT_NUM_I32 OPR_VGPR 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMP_LE_I32 Set the per-lane condition code to 1 iff the first input is less than or equal to the second input. Store the result into VCC or a scalar register. ENC_VOP3 default 195 VDST FMT_NUM_B64 OPR_SREG 64 SRC0 FMT_NUM_I32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_I32 OPR_SRC_SIMPLE 32 ENC_VOPC default 195 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_I32 OPR_SRC 32 VSRC1 FMT_NUM_I32 OPR_VGPR 32 VOPC_INST_LITERAL has_lit 195 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_I32 OPR_SRC 32 VSRC1 FMT_NUM_I32 OPR_VGPR 32 VOPC_VOP_SDWA_SDST_ENC has_sdwa 195 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_I32 OPR_SRC 32 VSRC1 FMT_NUM_I32 OPR_VGPR 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMP_GT_I32 Set the per-lane condition code to 1 iff the first input is greater than the second input. Store the result into VCC or a scalar register. ENC_VOP3 default 196 VDST FMT_NUM_B64 OPR_SREG 64 SRC0 FMT_NUM_I32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_I32 OPR_SRC_SIMPLE 32 ENC_VOPC default 196 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_I32 OPR_SRC 32 VSRC1 FMT_NUM_I32 OPR_VGPR 32 VOPC_INST_LITERAL has_lit 196 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_I32 OPR_SRC 32 VSRC1 FMT_NUM_I32 OPR_VGPR 32 VOPC_VOP_SDWA_SDST_ENC has_sdwa 196 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_I32 OPR_SRC 32 VSRC1 FMT_NUM_I32 OPR_VGPR 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMP_NE_I32 Set the per-lane condition code to 1 iff the first input is not equal to the second input. Store the result into VCC or a scalar register. ENC_VOP3 default 197 VDST FMT_NUM_B64 OPR_SREG 64 SRC0 FMT_NUM_I32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_I32 OPR_SRC_SIMPLE 32 ENC_VOPC default 197 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_I32 OPR_SRC 32 VSRC1 FMT_NUM_I32 OPR_VGPR 32 VOPC_INST_LITERAL has_lit 197 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_I32 OPR_SRC 32 VSRC1 FMT_NUM_I32 OPR_VGPR 32 VOPC_VOP_SDWA_SDST_ENC has_sdwa 197 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_I32 OPR_SRC 32 VSRC1 FMT_NUM_I32 OPR_VGPR 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMP_GE_I32 Set the per-lane condition code to 1 iff the first input is greater than or equal to the second input. Store the result into VCC or a scalar register. ENC_VOP3 default 198 VDST FMT_NUM_B64 OPR_SREG 64 SRC0 FMT_NUM_I32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_I32 OPR_SRC_SIMPLE 32 ENC_VOPC default 198 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_I32 OPR_SRC 32 VSRC1 FMT_NUM_I32 OPR_VGPR 32 VOPC_INST_LITERAL has_lit 198 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_I32 OPR_SRC 32 VSRC1 FMT_NUM_I32 OPR_VGPR 32 VOPC_VOP_SDWA_SDST_ENC has_sdwa 198 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_I32 OPR_SRC 32 VSRC1 FMT_NUM_I32 OPR_VGPR 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMP_T_I32 Set the per-lane condition code to 1. Store the result into VCC or a scalar register. ENC_VOP3 default 199 VDST FMT_NUM_B64 OPR_SREG 64 SRC0 FMT_NUM_I32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_I32 OPR_SRC_SIMPLE 32 ENC_VOPC default 199 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_I32 OPR_SRC 32 VSRC1 FMT_NUM_I32 OPR_VGPR 32 VOPC_INST_LITERAL has_lit 199 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_I32 OPR_SRC 32 VSRC1 FMT_NUM_I32 OPR_VGPR 32 VOPC_VOP_SDWA_SDST_ENC has_sdwa 199 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_I32 OPR_SRC 32 VSRC1 FMT_NUM_I32 OPR_VGPR 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMP_F_U32 Set the per-lane condition code to 0. Store the result into VCC or a scalar register. ENC_VOP3 default 200 VDST FMT_NUM_B64 OPR_SREG 64 SRC0 FMT_NUM_U32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_U32 OPR_SRC_SIMPLE 32 ENC_VOPC default 200 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_U32 OPR_SRC 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 VOPC_INST_LITERAL has_lit 200 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_U32 OPR_SRC 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 VOPC_VOP_SDWA_SDST_ENC has_sdwa 200 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_U32 OPR_SRC 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMP_LT_U32 Set the per-lane condition code to 1 iff the first input is less than the second input. Store the result into VCC or a scalar register. ENC_VOP3 default 201 VDST FMT_NUM_B64 OPR_SREG 64 SRC0 FMT_NUM_U32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_U32 OPR_SRC_SIMPLE 32 ENC_VOPC default 201 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_U32 OPR_SRC 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 VOPC_INST_LITERAL has_lit 201 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_U32 OPR_SRC 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 VOPC_VOP_SDWA_SDST_ENC has_sdwa 201 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_U32 OPR_SRC 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMP_EQ_U32 Set the per-lane condition code to 1 iff the first input is equal to the second input. Store the result into VCC or a scalar register. ENC_VOP3 default 202 VDST FMT_NUM_B64 OPR_SREG 64 SRC0 FMT_NUM_U32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_U32 OPR_SRC_SIMPLE 32 ENC_VOPC default 202 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_U32 OPR_SRC 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 VOPC_INST_LITERAL has_lit 202 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_U32 OPR_SRC 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 VOPC_VOP_SDWA_SDST_ENC has_sdwa 202 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_U32 OPR_SRC 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMP_LE_U32 Set the per-lane condition code to 1 iff the first input is less than or equal to the second input. Store the result into VCC or a scalar register. ENC_VOP3 default 203 VDST FMT_NUM_B64 OPR_SREG 64 SRC0 FMT_NUM_U32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_U32 OPR_SRC_SIMPLE 32 ENC_VOPC default 203 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_U32 OPR_SRC 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 VOPC_INST_LITERAL has_lit 203 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_U32 OPR_SRC 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 VOPC_VOP_SDWA_SDST_ENC has_sdwa 203 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_U32 OPR_SRC 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMP_GT_U32 Set the per-lane condition code to 1 iff the first input is greater than the second input. Store the result into VCC or a scalar register. ENC_VOP3 default 204 VDST FMT_NUM_B64 OPR_SREG 64 SRC0 FMT_NUM_U32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_U32 OPR_SRC_SIMPLE 32 ENC_VOPC default 204 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_U32 OPR_SRC 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 VOPC_INST_LITERAL has_lit 204 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_U32 OPR_SRC 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 VOPC_VOP_SDWA_SDST_ENC has_sdwa 204 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_U32 OPR_SRC 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMP_NE_U32 Set the per-lane condition code to 1 iff the first input is not equal to the second input. Store the result into VCC or a scalar register. ENC_VOP3 default 205 VDST FMT_NUM_B64 OPR_SREG 64 SRC0 FMT_NUM_U32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_U32 OPR_SRC_SIMPLE 32 ENC_VOPC default 205 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_U32 OPR_SRC 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 VOPC_INST_LITERAL has_lit 205 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_U32 OPR_SRC 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 VOPC_VOP_SDWA_SDST_ENC has_sdwa 205 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_U32 OPR_SRC 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMP_GE_U32 Set the per-lane condition code to 1 iff the first input is greater than or equal to the second input. Store the result into VCC or a scalar register. ENC_VOP3 default 206 VDST FMT_NUM_B64 OPR_SREG 64 SRC0 FMT_NUM_U32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_U32 OPR_SRC_SIMPLE 32 ENC_VOPC default 206 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_U32 OPR_SRC 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 VOPC_INST_LITERAL has_lit 206 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_U32 OPR_SRC 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 VOPC_VOP_SDWA_SDST_ENC has_sdwa 206 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_U32 OPR_SRC 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMP_T_U32 Set the per-lane condition code to 1. Store the result into VCC or a scalar register. ENC_VOP3 default 207 VDST FMT_NUM_B64 OPR_SREG 64 SRC0 FMT_NUM_U32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_U32 OPR_SRC_SIMPLE 32 ENC_VOPC default 207 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_U32 OPR_SRC 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 VOPC_INST_LITERAL has_lit 207 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_U32 OPR_SRC 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 VOPC_VOP_SDWA_SDST_ENC has_sdwa 207 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_U32 OPR_SRC 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMPX_F_I32 Set the per-lane condition code to 0. Store the result into the EXEC mask and to VCC or a scalar register. ENC_VOP3 default 208 VDST FMT_NUM_B64 OPR_SDST 64 SRC0 FMT_NUM_I32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_I32 OPR_SRC_SIMPLE 32 FMT_NUM_M64 OPR_SDST_EXEC 64 ENC_VOPC default 208 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_I32 OPR_SRC 32 VSRC1 FMT_NUM_I32 OPR_VGPR 32 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_INST_LITERAL has_lit 208 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_I32 OPR_SRC 32 VSRC1 FMT_NUM_I32 OPR_VGPR 32 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_VOP_SDWA_SDST_ENC has_sdwa 208 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_I32 OPR_SRC 32 VSRC1 FMT_NUM_I32 OPR_VGPR 32 FMT_NUM_M64 OPR_SDST_EXEC 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMPX_LT_I32 Set the per-lane condition code to 1 iff the first input is less than the second input. Store the result into the EXEC mask and to VCC or a scalar register. ENC_VOP3 default 209 VDST FMT_NUM_B64 OPR_SDST 64 SRC0 FMT_NUM_I32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_I32 OPR_SRC_SIMPLE 32 FMT_NUM_M64 OPR_SDST_EXEC 64 ENC_VOPC default 209 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_I32 OPR_SRC 32 VSRC1 FMT_NUM_I32 OPR_VGPR 32 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_INST_LITERAL has_lit 209 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_I32 OPR_SRC 32 VSRC1 FMT_NUM_I32 OPR_VGPR 32 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_VOP_SDWA_SDST_ENC has_sdwa 209 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_I32 OPR_SRC 32 VSRC1 FMT_NUM_I32 OPR_VGPR 32 FMT_NUM_M64 OPR_SDST_EXEC 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMPX_EQ_I32 Set the per-lane condition code to 1 iff the first input is equal to the second input. Store the result into the EXEC mask and to VCC or a scalar register. ENC_VOP3 default 210 VDST FMT_NUM_B64 OPR_SDST 64 SRC0 FMT_NUM_I32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_I32 OPR_SRC_SIMPLE 32 FMT_NUM_M64 OPR_SDST_EXEC 64 ENC_VOPC default 210 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_I32 OPR_SRC 32 VSRC1 FMT_NUM_I32 OPR_VGPR 32 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_INST_LITERAL has_lit 210 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_I32 OPR_SRC 32 VSRC1 FMT_NUM_I32 OPR_VGPR 32 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_VOP_SDWA_SDST_ENC has_sdwa 210 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_I32 OPR_SRC 32 VSRC1 FMT_NUM_I32 OPR_VGPR 32 FMT_NUM_M64 OPR_SDST_EXEC 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMPX_LE_I32 Set the per-lane condition code to 1 iff the first input is less than or equal to the second input. Store the result into the EXEC mask and to VCC or a scalar register. ENC_VOP3 default 211 VDST FMT_NUM_B64 OPR_SDST 64 SRC0 FMT_NUM_I32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_I32 OPR_SRC_SIMPLE 32 FMT_NUM_M64 OPR_SDST_EXEC 64 ENC_VOPC default 211 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_I32 OPR_SRC 32 VSRC1 FMT_NUM_I32 OPR_VGPR 32 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_INST_LITERAL has_lit 211 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_I32 OPR_SRC 32 VSRC1 FMT_NUM_I32 OPR_VGPR 32 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_VOP_SDWA_SDST_ENC has_sdwa 211 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_I32 OPR_SRC 32 VSRC1 FMT_NUM_I32 OPR_VGPR 32 FMT_NUM_M64 OPR_SDST_EXEC 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMPX_GT_I32 Set the per-lane condition code to 1 iff the first input is greater than the second input. Store the result into the EXEC mask and to VCC or a scalar register. ENC_VOP3 default 212 VDST FMT_NUM_B64 OPR_SDST 64 SRC0 FMT_NUM_I32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_I32 OPR_SRC_SIMPLE 32 FMT_NUM_M64 OPR_SDST_EXEC 64 ENC_VOPC default 212 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_I32 OPR_SRC 32 VSRC1 FMT_NUM_I32 OPR_VGPR 32 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_INST_LITERAL has_lit 212 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_I32 OPR_SRC 32 VSRC1 FMT_NUM_I32 OPR_VGPR 32 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_VOP_SDWA_SDST_ENC has_sdwa 212 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_I32 OPR_SRC 32 VSRC1 FMT_NUM_I32 OPR_VGPR 32 FMT_NUM_M64 OPR_SDST_EXEC 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMPX_NE_I32 Set the per-lane condition code to 1 iff the first input is not equal to the second input. Store the result into the EXEC mask and to VCC or a scalar register. ENC_VOP3 default 213 VDST FMT_NUM_B64 OPR_SDST 64 SRC0 FMT_NUM_I32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_I32 OPR_SRC_SIMPLE 32 FMT_NUM_M64 OPR_SDST_EXEC 64 ENC_VOPC default 213 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_I32 OPR_SRC 32 VSRC1 FMT_NUM_I32 OPR_VGPR 32 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_INST_LITERAL has_lit 213 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_I32 OPR_SRC 32 VSRC1 FMT_NUM_I32 OPR_VGPR 32 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_VOP_SDWA_SDST_ENC has_sdwa 213 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_I32 OPR_SRC 32 VSRC1 FMT_NUM_I32 OPR_VGPR 32 FMT_NUM_M64 OPR_SDST_EXEC 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMPX_GE_I32 Set the per-lane condition code to 1 iff the first input is greater than or equal to the second input. Store the result into the EXEC mask and to VCC or a scalar register. ENC_VOP3 default 214 VDST FMT_NUM_B64 OPR_SDST 64 SRC0 FMT_NUM_I32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_I32 OPR_SRC_SIMPLE 32 FMT_NUM_M64 OPR_SDST_EXEC 64 ENC_VOPC default 214 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_I32 OPR_SRC 32 VSRC1 FMT_NUM_I32 OPR_VGPR 32 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_INST_LITERAL has_lit 214 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_I32 OPR_SRC 32 VSRC1 FMT_NUM_I32 OPR_VGPR 32 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_VOP_SDWA_SDST_ENC has_sdwa 214 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_I32 OPR_SRC 32 VSRC1 FMT_NUM_I32 OPR_VGPR 32 FMT_NUM_M64 OPR_SDST_EXEC 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMPX_T_I32 Set the per-lane condition code to 1. Store the result into the EXEC mask and to VCC or a scalar register. ENC_VOP3 default 215 VDST FMT_NUM_B64 OPR_SDST 64 SRC0 FMT_NUM_I32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_I32 OPR_SRC_SIMPLE 32 FMT_NUM_M64 OPR_SDST_EXEC 64 ENC_VOPC default 215 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_I32 OPR_SRC 32 VSRC1 FMT_NUM_I32 OPR_VGPR 32 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_INST_LITERAL has_lit 215 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_I32 OPR_SRC 32 VSRC1 FMT_NUM_I32 OPR_VGPR 32 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_VOP_SDWA_SDST_ENC has_sdwa 215 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_I32 OPR_SRC 32 VSRC1 FMT_NUM_I32 OPR_VGPR 32 FMT_NUM_M64 OPR_SDST_EXEC 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMPX_F_U32 Set the per-lane condition code to 0. Store the result into the EXEC mask and to VCC or a scalar register. ENC_VOP3 default 216 VDST FMT_NUM_B64 OPR_SDST 64 SRC0 FMT_NUM_U32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_U32 OPR_SRC_SIMPLE 32 FMT_NUM_M64 OPR_SDST_EXEC 64 ENC_VOPC default 216 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_U32 OPR_SRC 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_INST_LITERAL has_lit 216 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_U32 OPR_SRC 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_VOP_SDWA_SDST_ENC has_sdwa 216 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_U32 OPR_SRC 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 FMT_NUM_M64 OPR_SDST_EXEC 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMPX_LT_U32 Set the per-lane condition code to 1 iff the first input is less than the second input. Store the result into the EXEC mask and to VCC or a scalar register. ENC_VOP3 default 217 VDST FMT_NUM_B64 OPR_SDST 64 SRC0 FMT_NUM_U32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_U32 OPR_SRC_SIMPLE 32 FMT_NUM_M64 OPR_SDST_EXEC 64 ENC_VOPC default 217 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_U32 OPR_SRC 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_INST_LITERAL has_lit 217 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_U32 OPR_SRC 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_VOP_SDWA_SDST_ENC has_sdwa 217 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_U32 OPR_SRC 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 FMT_NUM_M64 OPR_SDST_EXEC 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMPX_EQ_U32 Set the per-lane condition code to 1 iff the first input is equal to the second input. Store the result into the EXEC mask and to VCC or a scalar register. ENC_VOP3 default 218 VDST FMT_NUM_B64 OPR_SDST 64 SRC0 FMT_NUM_U32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_U32 OPR_SRC_SIMPLE 32 FMT_NUM_M64 OPR_SDST_EXEC 64 ENC_VOPC default 218 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_U32 OPR_SRC 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_INST_LITERAL has_lit 218 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_U32 OPR_SRC 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_VOP_SDWA_SDST_ENC has_sdwa 218 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_U32 OPR_SRC 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 FMT_NUM_M64 OPR_SDST_EXEC 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMPX_LE_U32 Set the per-lane condition code to 1 iff the first input is less than or equal to the second input. Store the result into the EXEC mask and to VCC or a scalar register. ENC_VOP3 default 219 VDST FMT_NUM_B64 OPR_SDST 64 SRC0 FMT_NUM_U32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_U32 OPR_SRC_SIMPLE 32 FMT_NUM_M64 OPR_SDST_EXEC 64 ENC_VOPC default 219 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_U32 OPR_SRC 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_INST_LITERAL has_lit 219 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_U32 OPR_SRC 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_VOP_SDWA_SDST_ENC has_sdwa 219 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_U32 OPR_SRC 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 FMT_NUM_M64 OPR_SDST_EXEC 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMPX_GT_U32 Set the per-lane condition code to 1 iff the first input is greater than the second input. Store the result into the EXEC mask and to VCC or a scalar register. ENC_VOP3 default 220 VDST FMT_NUM_B64 OPR_SDST 64 SRC0 FMT_NUM_U32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_U32 OPR_SRC_SIMPLE 32 FMT_NUM_M64 OPR_SDST_EXEC 64 ENC_VOPC default 220 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_U32 OPR_SRC 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_INST_LITERAL has_lit 220 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_U32 OPR_SRC 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_VOP_SDWA_SDST_ENC has_sdwa 220 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_U32 OPR_SRC 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 FMT_NUM_M64 OPR_SDST_EXEC 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMPX_NE_U32 Set the per-lane condition code to 1 iff the first input is not equal to the second input. Store the result into the EXEC mask and to VCC or a scalar register. ENC_VOP3 default 221 VDST FMT_NUM_B64 OPR_SDST 64 SRC0 FMT_NUM_U32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_U32 OPR_SRC_SIMPLE 32 FMT_NUM_M64 OPR_SDST_EXEC 64 ENC_VOPC default 221 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_U32 OPR_SRC 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_INST_LITERAL has_lit 221 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_U32 OPR_SRC 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_VOP_SDWA_SDST_ENC has_sdwa 221 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_U32 OPR_SRC 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 FMT_NUM_M64 OPR_SDST_EXEC 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMPX_GE_U32 Set the per-lane condition code to 1 iff the first input is greater than or equal to the second input. Store the result into the EXEC mask and to VCC or a scalar register. ENC_VOP3 default 222 VDST FMT_NUM_B64 OPR_SDST 64 SRC0 FMT_NUM_U32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_U32 OPR_SRC_SIMPLE 32 FMT_NUM_M64 OPR_SDST_EXEC 64 ENC_VOPC default 222 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_U32 OPR_SRC 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_INST_LITERAL has_lit 222 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_U32 OPR_SRC 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_VOP_SDWA_SDST_ENC has_sdwa 222 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_U32 OPR_SRC 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 FMT_NUM_M64 OPR_SDST_EXEC 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMPX_T_U32 Set the per-lane condition code to 1. Store the result into the EXEC mask and to VCC or a scalar register. ENC_VOP3 default 223 VDST FMT_NUM_B64 OPR_SDST 64 SRC0 FMT_NUM_U32 OPR_SRC_NOLIT 32 SRC1 FMT_NUM_U32 OPR_SRC_SIMPLE 32 FMT_NUM_M64 OPR_SDST_EXEC 64 ENC_VOPC default 223 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_U32 OPR_SRC 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_INST_LITERAL has_lit 223 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_U32 OPR_SRC 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_VOP_SDWA_SDST_ENC has_sdwa 223 SDST FMT_NUM_M64 OPR_SREG 64 VSRC0 FMT_NUM_U32 OPR_SRC 32 VSRC1 FMT_NUM_U32 OPR_VGPR 32 FMT_NUM_M64 OPR_SDST_EXEC 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMP_F_I64 Set the per-lane condition code to 0. Store the result into VCC or a scalar register. ENC_VOP3 default 224 VDST FMT_NUM_B64 OPR_SREG 64 SRC0 FMT_NUM_I64 OPR_SRC_NOLIT 64 SRC1 FMT_NUM_I64 OPR_SRC_SIMPLE 64 ENC_VOPC default 224 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_I64 OPR_SRC 64 VSRC1 FMT_NUM_I64 OPR_VGPR 64 VOPC_INST_LITERAL has_lit 224 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_I64 OPR_SRC 64 VSRC1 FMT_NUM_I64 OPR_VGPR 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMP_LT_I64 Set the per-lane condition code to 1 iff the first input is less than the second input. Store the result into VCC or a scalar register. ENC_VOP3 default 225 VDST FMT_NUM_B64 OPR_SREG 64 SRC0 FMT_NUM_I64 OPR_SRC_NOLIT 64 SRC1 FMT_NUM_I64 OPR_SRC_SIMPLE 64 ENC_VOPC default 225 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_I64 OPR_SRC 64 VSRC1 FMT_NUM_I64 OPR_VGPR 64 VOPC_INST_LITERAL has_lit 225 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_I64 OPR_SRC 64 VSRC1 FMT_NUM_I64 OPR_VGPR 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMP_EQ_I64 Set the per-lane condition code to 1 iff the first input is equal to the second input. Store the result into VCC or a scalar register. ENC_VOP3 default 226 VDST FMT_NUM_B64 OPR_SREG 64 SRC0 FMT_NUM_I64 OPR_SRC_NOLIT 64 SRC1 FMT_NUM_I64 OPR_SRC_SIMPLE 64 ENC_VOPC default 226 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_I64 OPR_SRC 64 VSRC1 FMT_NUM_I64 OPR_VGPR 64 VOPC_INST_LITERAL has_lit 226 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_I64 OPR_SRC 64 VSRC1 FMT_NUM_I64 OPR_VGPR 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMP_LE_I64 Set the per-lane condition code to 1 iff the first input is less than or equal to the second input. Store the result into VCC or a scalar register. ENC_VOP3 default 227 VDST FMT_NUM_B64 OPR_SREG 64 SRC0 FMT_NUM_I64 OPR_SRC_NOLIT 64 SRC1 FMT_NUM_I64 OPR_SRC_SIMPLE 64 ENC_VOPC default 227 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_I64 OPR_SRC 64 VSRC1 FMT_NUM_I64 OPR_VGPR 64 VOPC_INST_LITERAL has_lit 227 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_I64 OPR_SRC 64 VSRC1 FMT_NUM_I64 OPR_VGPR 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMP_GT_I64 Set the per-lane condition code to 1 iff the first input is greater than the second input. Store the result into VCC or a scalar register. ENC_VOP3 default 228 VDST FMT_NUM_B64 OPR_SREG 64 SRC0 FMT_NUM_I64 OPR_SRC_NOLIT 64 SRC1 FMT_NUM_I64 OPR_SRC_SIMPLE 64 ENC_VOPC default 228 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_I64 OPR_SRC 64 VSRC1 FMT_NUM_I64 OPR_VGPR 64 VOPC_INST_LITERAL has_lit 228 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_I64 OPR_SRC 64 VSRC1 FMT_NUM_I64 OPR_VGPR 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMP_NE_I64 Set the per-lane condition code to 1 iff the first input is not equal to the second input. Store the result into VCC or a scalar register. ENC_VOP3 default 229 VDST FMT_NUM_B64 OPR_SREG 64 SRC0 FMT_NUM_I64 OPR_SRC_NOLIT 64 SRC1 FMT_NUM_I64 OPR_SRC_SIMPLE 64 ENC_VOPC default 229 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_I64 OPR_SRC 64 VSRC1 FMT_NUM_I64 OPR_VGPR 64 VOPC_INST_LITERAL has_lit 229 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_I64 OPR_SRC 64 VSRC1 FMT_NUM_I64 OPR_VGPR 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMP_GE_I64 Set the per-lane condition code to 1 iff the first input is greater than or equal to the second input. Store the result into VCC or a scalar register. ENC_VOP3 default 230 VDST FMT_NUM_B64 OPR_SREG 64 SRC0 FMT_NUM_I64 OPR_SRC_NOLIT 64 SRC1 FMT_NUM_I64 OPR_SRC_SIMPLE 64 ENC_VOPC default 230 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_I64 OPR_SRC 64 VSRC1 FMT_NUM_I64 OPR_VGPR 64 VOPC_INST_LITERAL has_lit 230 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_I64 OPR_SRC 64 VSRC1 FMT_NUM_I64 OPR_VGPR 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMP_T_I64 Set the per-lane condition code to 1. Store the result into VCC or a scalar register. ENC_VOP3 default 231 VDST FMT_NUM_B64 OPR_SREG 64 SRC0 FMT_NUM_I64 OPR_SRC_NOLIT 64 SRC1 FMT_NUM_I64 OPR_SRC_SIMPLE 64 ENC_VOPC default 231 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_I64 OPR_SRC 64 VSRC1 FMT_NUM_I64 OPR_VGPR 64 VOPC_INST_LITERAL has_lit 231 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_I64 OPR_SRC 64 VSRC1 FMT_NUM_I64 OPR_VGPR 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMP_F_U64 Set the per-lane condition code to 0. Store the result into VCC or a scalar register. ENC_VOP3 default 232 VDST FMT_NUM_B64 OPR_SREG 64 SRC0 FMT_NUM_U64 OPR_SRC_NOLIT 64 SRC1 FMT_NUM_U64 OPR_SRC_SIMPLE 64 ENC_VOPC default 232 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_U64 OPR_SRC 64 VSRC1 FMT_NUM_U64 OPR_VGPR 64 VOPC_INST_LITERAL has_lit 232 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_U64 OPR_SRC 64 VSRC1 FMT_NUM_U64 OPR_VGPR 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMP_LT_U64 Set the per-lane condition code to 1 iff the first input is less than the second input. Store the result into VCC or a scalar register. ENC_VOP3 default 233 VDST FMT_NUM_B64 OPR_SREG 64 SRC0 FMT_NUM_U64 OPR_SRC_NOLIT 64 SRC1 FMT_NUM_U64 OPR_SRC_SIMPLE 64 ENC_VOPC default 233 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_U64 OPR_SRC 64 VSRC1 FMT_NUM_U64 OPR_VGPR 64 VOPC_INST_LITERAL has_lit 233 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_U64 OPR_SRC 64 VSRC1 FMT_NUM_U64 OPR_VGPR 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMP_EQ_U64 Set the per-lane condition code to 1 iff the first input is equal to the second input. Store the result into VCC or a scalar register. ENC_VOP3 default 234 VDST FMT_NUM_B64 OPR_SREG 64 SRC0 FMT_NUM_U64 OPR_SRC_NOLIT 64 SRC1 FMT_NUM_U64 OPR_SRC_SIMPLE 64 ENC_VOPC default 234 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_U64 OPR_SRC 64 VSRC1 FMT_NUM_U64 OPR_VGPR 64 VOPC_INST_LITERAL has_lit 234 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_U64 OPR_SRC 64 VSRC1 FMT_NUM_U64 OPR_VGPR 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMP_LE_U64 Set the per-lane condition code to 1 iff the first input is less than or equal to the second input. Store the result into VCC or a scalar register. ENC_VOP3 default 235 VDST FMT_NUM_B64 OPR_SREG 64 SRC0 FMT_NUM_U64 OPR_SRC_NOLIT 64 SRC1 FMT_NUM_U64 OPR_SRC_SIMPLE 64 ENC_VOPC default 235 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_U64 OPR_SRC 64 VSRC1 FMT_NUM_U64 OPR_VGPR 64 VOPC_INST_LITERAL has_lit 235 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_U64 OPR_SRC 64 VSRC1 FMT_NUM_U64 OPR_VGPR 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMP_GT_U64 Set the per-lane condition code to 1 iff the first input is greater than the second input. Store the result into VCC or a scalar register. ENC_VOP3 default 236 VDST FMT_NUM_B64 OPR_SREG 64 SRC0 FMT_NUM_U64 OPR_SRC_NOLIT 64 SRC1 FMT_NUM_U64 OPR_SRC_SIMPLE 64 ENC_VOPC default 236 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_U64 OPR_SRC 64 VSRC1 FMT_NUM_U64 OPR_VGPR 64 VOPC_INST_LITERAL has_lit 236 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_U64 OPR_SRC 64 VSRC1 FMT_NUM_U64 OPR_VGPR 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMP_NE_U64 Set the per-lane condition code to 1 iff the first input is not equal to the second input. Store the result into VCC or a scalar register. ENC_VOP3 default 237 VDST FMT_NUM_B64 OPR_SREG 64 SRC0 FMT_NUM_U64 OPR_SRC_NOLIT 64 SRC1 FMT_NUM_U64 OPR_SRC_SIMPLE 64 ENC_VOPC default 237 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_U64 OPR_SRC 64 VSRC1 FMT_NUM_U64 OPR_VGPR 64 VOPC_INST_LITERAL has_lit 237 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_U64 OPR_SRC 64 VSRC1 FMT_NUM_U64 OPR_VGPR 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMP_GE_U64 Set the per-lane condition code to 1 iff the first input is greater than or equal to the second input. Store the result into VCC or a scalar register. ENC_VOP3 default 238 VDST FMT_NUM_B64 OPR_SREG 64 SRC0 FMT_NUM_U64 OPR_SRC_NOLIT 64 SRC1 FMT_NUM_U64 OPR_SRC_SIMPLE 64 ENC_VOPC default 238 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_U64 OPR_SRC 64 VSRC1 FMT_NUM_U64 OPR_VGPR 64 VOPC_INST_LITERAL has_lit 238 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_U64 OPR_SRC 64 VSRC1 FMT_NUM_U64 OPR_VGPR 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMP_T_U64 Set the per-lane condition code to 1. Store the result into VCC or a scalar register. ENC_VOP3 default 239 VDST FMT_NUM_B64 OPR_SREG 64 SRC0 FMT_NUM_U64 OPR_SRC_NOLIT 64 SRC1 FMT_NUM_U64 OPR_SRC_SIMPLE 64 ENC_VOPC default 239 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_U64 OPR_SRC 64 VSRC1 FMT_NUM_U64 OPR_VGPR 64 VOPC_INST_LITERAL has_lit 239 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_U64 OPR_SRC 64 VSRC1 FMT_NUM_U64 OPR_VGPR 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMPX_F_I64 Set the per-lane condition code to 0. Store the result into the EXEC mask and to VCC or a scalar register. ENC_VOP3 default 240 VDST FMT_NUM_B64 OPR_SDST 64 SRC0 FMT_NUM_I64 OPR_SRC_NOLIT 64 SRC1 FMT_NUM_I64 OPR_SRC_SIMPLE 64 FMT_NUM_M64 OPR_SDST_EXEC 64 ENC_VOPC default 240 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_I64 OPR_SRC 64 VSRC1 FMT_NUM_I64 OPR_VGPR 64 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_INST_LITERAL has_lit 240 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_I64 OPR_SRC 64 VSRC1 FMT_NUM_I64 OPR_VGPR 64 FMT_NUM_M64 OPR_SDST_EXEC 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMPX_LT_I64 Set the per-lane condition code to 1 iff the first input is less than the second input. Store the result into the EXEC mask and to VCC or a scalar register. ENC_VOP3 default 241 VDST FMT_NUM_B64 OPR_SDST 64 SRC0 FMT_NUM_I64 OPR_SRC_NOLIT 64 SRC1 FMT_NUM_I64 OPR_SRC_SIMPLE 64 FMT_NUM_M64 OPR_SDST_EXEC 64 ENC_VOPC default 241 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_I64 OPR_SRC 64 VSRC1 FMT_NUM_I64 OPR_VGPR 64 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_INST_LITERAL has_lit 241 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_I64 OPR_SRC 64 VSRC1 FMT_NUM_I64 OPR_VGPR 64 FMT_NUM_M64 OPR_SDST_EXEC 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMPX_EQ_I64 Set the per-lane condition code to 1 iff the first input is equal to the second input. Store the result into the EXEC mask and to VCC or a scalar register. ENC_VOP3 default 242 VDST FMT_NUM_B64 OPR_SDST 64 SRC0 FMT_NUM_I64 OPR_SRC_NOLIT 64 SRC1 FMT_NUM_I64 OPR_SRC_SIMPLE 64 FMT_NUM_M64 OPR_SDST_EXEC 64 ENC_VOPC default 242 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_I64 OPR_SRC 64 VSRC1 FMT_NUM_I64 OPR_VGPR 64 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_INST_LITERAL has_lit 242 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_I64 OPR_SRC 64 VSRC1 FMT_NUM_I64 OPR_VGPR 64 FMT_NUM_M64 OPR_SDST_EXEC 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMPX_LE_I64 Set the per-lane condition code to 1 iff the first input is less than or equal to the second input. Store the result into the EXEC mask and to VCC or a scalar register. ENC_VOP3 default 243 VDST FMT_NUM_B64 OPR_SDST 64 SRC0 FMT_NUM_I64 OPR_SRC_NOLIT 64 SRC1 FMT_NUM_I64 OPR_SRC_SIMPLE 64 FMT_NUM_M64 OPR_SDST_EXEC 64 ENC_VOPC default 243 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_I64 OPR_SRC 64 VSRC1 FMT_NUM_I64 OPR_VGPR 64 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_INST_LITERAL has_lit 243 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_I64 OPR_SRC 64 VSRC1 FMT_NUM_I64 OPR_VGPR 64 FMT_NUM_M64 OPR_SDST_EXEC 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMPX_GT_I64 Set the per-lane condition code to 1 iff the first input is greater than the second input. Store the result into the EXEC mask and to VCC or a scalar register. ENC_VOP3 default 244 VDST FMT_NUM_B64 OPR_SDST 64 SRC0 FMT_NUM_I64 OPR_SRC_NOLIT 64 SRC1 FMT_NUM_I64 OPR_SRC_SIMPLE 64 FMT_NUM_M64 OPR_SDST_EXEC 64 ENC_VOPC default 244 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_I64 OPR_SRC 64 VSRC1 FMT_NUM_I64 OPR_VGPR 64 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_INST_LITERAL has_lit 244 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_I64 OPR_SRC 64 VSRC1 FMT_NUM_I64 OPR_VGPR 64 FMT_NUM_M64 OPR_SDST_EXEC 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMPX_NE_I64 Set the per-lane condition code to 1 iff the first input is not equal to the second input. Store the result into the EXEC mask and to VCC or a scalar register. ENC_VOP3 default 245 VDST FMT_NUM_B64 OPR_SDST 64 SRC0 FMT_NUM_I64 OPR_SRC_NOLIT 64 SRC1 FMT_NUM_I64 OPR_SRC_SIMPLE 64 FMT_NUM_M64 OPR_SDST_EXEC 64 ENC_VOPC default 245 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_I64 OPR_SRC 64 VSRC1 FMT_NUM_I64 OPR_VGPR 64 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_INST_LITERAL has_lit 245 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_I64 OPR_SRC 64 VSRC1 FMT_NUM_I64 OPR_VGPR 64 FMT_NUM_M64 OPR_SDST_EXEC 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMPX_GE_I64 Set the per-lane condition code to 1 iff the first input is greater than or equal to the second input. Store the result into the EXEC mask and to VCC or a scalar register. ENC_VOP3 default 246 VDST FMT_NUM_B64 OPR_SDST 64 SRC0 FMT_NUM_I64 OPR_SRC_NOLIT 64 SRC1 FMT_NUM_I64 OPR_SRC_SIMPLE 64 FMT_NUM_M64 OPR_SDST_EXEC 64 ENC_VOPC default 246 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_I64 OPR_SRC 64 VSRC1 FMT_NUM_I64 OPR_VGPR 64 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_INST_LITERAL has_lit 246 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_I64 OPR_SRC 64 VSRC1 FMT_NUM_I64 OPR_VGPR 64 FMT_NUM_M64 OPR_SDST_EXEC 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMPX_T_I64 Set the per-lane condition code to 1. Store the result into the EXEC mask and to VCC or a scalar register. ENC_VOP3 default 247 VDST FMT_NUM_B64 OPR_SDST 64 SRC0 FMT_NUM_I64 OPR_SRC_NOLIT 64 SRC1 FMT_NUM_I64 OPR_SRC_SIMPLE 64 FMT_NUM_M64 OPR_SDST_EXEC 64 ENC_VOPC default 247 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_I64 OPR_SRC 64 VSRC1 FMT_NUM_I64 OPR_VGPR 64 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_INST_LITERAL has_lit 247 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_I64 OPR_SRC 64 VSRC1 FMT_NUM_I64 OPR_VGPR 64 FMT_NUM_M64 OPR_SDST_EXEC 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMPX_F_U64 Set the per-lane condition code to 0. Store the result into the EXEC mask and to VCC or a scalar register. ENC_VOP3 default 248 VDST FMT_NUM_B64 OPR_SDST 64 SRC0 FMT_NUM_U64 OPR_SRC_NOLIT 64 SRC1 FMT_NUM_U64 OPR_SRC_SIMPLE 64 FMT_NUM_M64 OPR_SDST_EXEC 64 ENC_VOPC default 248 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_U64 OPR_SRC 64 VSRC1 FMT_NUM_U64 OPR_VGPR 64 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_INST_LITERAL has_lit 248 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_U64 OPR_SRC 64 VSRC1 FMT_NUM_U64 OPR_VGPR 64 FMT_NUM_M64 OPR_SDST_EXEC 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMPX_LT_U64 Set the per-lane condition code to 1 iff the first input is less than the second input. Store the result into the EXEC mask and to VCC or a scalar register. ENC_VOP3 default 249 VDST FMT_NUM_B64 OPR_SDST 64 SRC0 FMT_NUM_U64 OPR_SRC_NOLIT 64 SRC1 FMT_NUM_U64 OPR_SRC_SIMPLE 64 FMT_NUM_M64 OPR_SDST_EXEC 64 ENC_VOPC default 249 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_U64 OPR_SRC 64 VSRC1 FMT_NUM_U64 OPR_VGPR 64 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_INST_LITERAL has_lit 249 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_U64 OPR_SRC 64 VSRC1 FMT_NUM_U64 OPR_VGPR 64 FMT_NUM_M64 OPR_SDST_EXEC 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMPX_EQ_U64 Set the per-lane condition code to 1 iff the first input is equal to the second input. Store the result into the EXEC mask and to VCC or a scalar register. ENC_VOP3 default 250 VDST FMT_NUM_B64 OPR_SDST 64 SRC0 FMT_NUM_U64 OPR_SRC_NOLIT 64 SRC1 FMT_NUM_U64 OPR_SRC_SIMPLE 64 FMT_NUM_M64 OPR_SDST_EXEC 64 ENC_VOPC default 250 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_U64 OPR_SRC 64 VSRC1 FMT_NUM_U64 OPR_VGPR 64 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_INST_LITERAL has_lit 250 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_U64 OPR_SRC 64 VSRC1 FMT_NUM_U64 OPR_VGPR 64 FMT_NUM_M64 OPR_SDST_EXEC 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMPX_LE_U64 Set the per-lane condition code to 1 iff the first input is less than or equal to the second input. Store the result into the EXEC mask and to VCC or a scalar register. ENC_VOP3 default 251 VDST FMT_NUM_B64 OPR_SDST 64 SRC0 FMT_NUM_U64 OPR_SRC_NOLIT 64 SRC1 FMT_NUM_U64 OPR_SRC_SIMPLE 64 FMT_NUM_M64 OPR_SDST_EXEC 64 ENC_VOPC default 251 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_U64 OPR_SRC 64 VSRC1 FMT_NUM_U64 OPR_VGPR 64 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_INST_LITERAL has_lit 251 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_U64 OPR_SRC 64 VSRC1 FMT_NUM_U64 OPR_VGPR 64 FMT_NUM_M64 OPR_SDST_EXEC 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMPX_GT_U64 Set the per-lane condition code to 1 iff the first input is greater than the second input. Store the result into the EXEC mask and to VCC or a scalar register. ENC_VOP3 default 252 VDST FMT_NUM_B64 OPR_SDST 64 SRC0 FMT_NUM_U64 OPR_SRC_NOLIT 64 SRC1 FMT_NUM_U64 OPR_SRC_SIMPLE 64 FMT_NUM_M64 OPR_SDST_EXEC 64 ENC_VOPC default 252 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_U64 OPR_SRC 64 VSRC1 FMT_NUM_U64 OPR_VGPR 64 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_INST_LITERAL has_lit 252 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_U64 OPR_SRC 64 VSRC1 FMT_NUM_U64 OPR_VGPR 64 FMT_NUM_M64 OPR_SDST_EXEC 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMPX_NE_U64 Set the per-lane condition code to 1 iff the first input is not equal to the second input. Store the result into the EXEC mask and to VCC or a scalar register. ENC_VOP3 default 253 VDST FMT_NUM_B64 OPR_SDST 64 SRC0 FMT_NUM_U64 OPR_SRC_NOLIT 64 SRC1 FMT_NUM_U64 OPR_SRC_SIMPLE 64 FMT_NUM_M64 OPR_SDST_EXEC 64 ENC_VOPC default 253 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_U64 OPR_SRC 64 VSRC1 FMT_NUM_U64 OPR_VGPR 64 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_INST_LITERAL has_lit 253 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_U64 OPR_SRC 64 VSRC1 FMT_NUM_U64 OPR_VGPR 64 FMT_NUM_M64 OPR_SDST_EXEC 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMPX_GE_U64 Set the per-lane condition code to 1 iff the first input is greater than or equal to the second input. Store the result into the EXEC mask and to VCC or a scalar register. ENC_VOP3 default 254 VDST FMT_NUM_B64 OPR_SDST 64 SRC0 FMT_NUM_U64 OPR_SRC_NOLIT 64 SRC1 FMT_NUM_U64 OPR_SRC_SIMPLE 64 FMT_NUM_M64 OPR_SDST_EXEC 64 ENC_VOPC default 254 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_U64 OPR_SRC 64 VSRC1 FMT_NUM_U64 OPR_VGPR 64 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_INST_LITERAL has_lit 254 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_U64 OPR_SRC 64 VSRC1 FMT_NUM_U64 OPR_VGPR 64 FMT_NUM_M64 OPR_SDST_EXEC 64 VALU NOT_ASSIGNED FALSE FALSE FALSE FALSE FALSE V_CMPX_T_U64 Set the per-lane condition code to 1. Store the result into the EXEC mask and to VCC or a scalar register. ENC_VOP3 default 255 VDST FMT_NUM_B64 OPR_SDST 64 SRC0 FMT_NUM_U64 OPR_SRC_NOLIT 64 SRC1 FMT_NUM_U64 OPR_SRC_SIMPLE 64 FMT_NUM_M64 OPR_SDST_EXEC 64 ENC_VOPC default 255 FMT_NUM_M64 OPR_VCC 64 SRC0 FMT_NUM_U64 OPR_SRC 64 VSRC1 FMT_NUM_U64 OPR_VGPR 64 FMT_NUM_M64 OPR_SDST_EXEC 64 VOPC_INST_LITERAL has_lit 255 FMT_NUM_M64 OPR_VCC 64 SIMM32 FMT_NUM_U64 OPR_SRC 64 VSRC1 FMT_NUM_U64 OPR_VGPR 64 FMT_NUM_M64 OPR_SDST_EXEC 64 VALU NOT_ASSIGNED FMT_ANY Value can be anything. bits 1 1 Bits 1 0 FMT_BUF 64-bit virtual memory address. descriptor 64 1 Descriptor 64 0 FMT_IMG 256-bit image resource constant. descriptor 256 1 Descriptor 256 0 FMT_NUM_B1 N/A bits 1 1 Bits 1 0 FMT_NUM_B128 N/A bits 128 1 Bits 128 0 FMT_NUM_B16 N/A bits 16 1 Bits 16 0 FMT_NUM_B256 N/A bits 256 1 Bits 256 0 FMT_NUM_B32 N/A bits 32 1 Bits 32 0 FMT_NUM_B512 N/A bits 512 1 Bits 512 0 FMT_NUM_B64 N/A bits 64 1 Bits 64 0 FMT_NUM_B8 N/A bits 8 1 Bits 8 0 FMT_NUM_B96 N/A bits 96 1 Bits 96 0 FMT_NUM_BF16 N/A float 16 1 Mantissa 7 0 Exponent 8 7 Sign 1 15 FMT_NUM_BF6 N/A float 6 1 Mantissa 2 0 Exponent 3 2 Sign 1 5 FMT_NUM_BF8 N/A float 8 1 Mantissa 2 0 Exponent 5 2 Sign 1 7 FMT_NUM_F16 N/A float 16 1 Mantissa 10 0 Exponent 5 10 Sign 1 15 FMT_NUM_F32 N/A float 32 1 Mantissa 23 0 Exponent 8 23 Sign 1 31 FMT_NUM_F64 N/A float 64 1 Mantissa 52 0 Exponent 11 52 Sign 1 63 FMT_NUM_FP4 N/A float 4 1 Mantissa 1 0 Exponent 2 1 Sign 1 3 FMT_NUM_FP6 N/A float 6 1 Mantissa 3 0 Exponent 2 3 Sign 1 5 FMT_NUM_FP8 N/A float 8 1 Mantissa 3 0 Exponent 4 3 Sign 1 7 FMT_NUM_I16 N/A integer 16 1 Integer 16 0 FMT_NUM_I24 N/A integer 24 1 Integer 24 0 FMT_NUM_I32 N/A integer 32 1 Integer 32 0 FMT_NUM_I4 N/A integer 4 1 Integer 4 0 FMT_NUM_I64 N/A integer 64 1 Integer 64 0 FMT_NUM_I8 N/A integer 8 1 Integer 8 0 FMT_NUM_M64 N/A bits 64 1 Bits 64 0 FMT_NUM_PK16_BF16 N/A float 256 16 Mantissa 7 0 Exponent 8 7 Sign 1 15 Mantissa 7 16 Exponent 8 23 Sign 1 31 Mantissa 7 32 Exponent 8 39 Sign 1 47 Mantissa 7 48 Exponent 8 55 Sign 1 63 Mantissa 7 64 Exponent 8 71 Sign 1 79 Mantissa 7 80 Exponent 8 87 Sign 1 95 Mantissa 7 96 Exponent 8 103 Sign 1 111 Mantissa 7 112 Exponent 8 119 Sign 1 127 Mantissa 7 128 Exponent 8 135 Sign 1 143 Mantissa 7 144 Exponent 8 151 Sign 1 159 Mantissa 7 160 Exponent 8 167 Sign 1 175 Mantissa 7 176 Exponent 8 183 Sign 1 191 Mantissa 7 192 Exponent 8 199 Sign 1 207 Mantissa 7 208 Exponent 8 215 Sign 1 223 Mantissa 7 224 Exponent 8 231 Sign 1 239 Mantissa 7 240 Exponent 8 247 Sign 1 255 FMT_NUM_PK16_BF8 N/A float 128 16 Mantissa 2 0 Exponent 5 2 Sign 1 7 Mantissa 2 8 Exponent 5 10 Sign 1 15 Mantissa 2 16 Exponent 5 18 Sign 1 23 Mantissa 2 24 Exponent 5 26 Sign 1 31 Mantissa 2 32 Exponent 5 34 Sign 1 39 Mantissa 2 40 Exponent 5 42 Sign 1 47 Mantissa 2 48 Exponent 5 50 Sign 1 55 Mantissa 2 56 Exponent 5 58 Sign 1 63 Mantissa 2 64 Exponent 5 66 Sign 1 71 Mantissa 2 72 Exponent 5 74 Sign 1 79 Mantissa 2 80 Exponent 5 82 Sign 1 87 Mantissa 2 88 Exponent 5 90 Sign 1 95 Mantissa 2 96 Exponent 5 98 Sign 1 103 Mantissa 2 104 Exponent 5 106 Sign 1 111 Mantissa 2 112 Exponent 5 114 Sign 1 119 Mantissa 2 120 Exponent 5 122 Sign 1 127 FMT_NUM_PK16_F16 N/A float 256 16 Mantissa 10 0 Exponent 5 10 Sign 1 15 Mantissa 10 16 Exponent 5 26 Sign 1 31 Mantissa 10 32 Exponent 5 42 Sign 1 47 Mantissa 10 48 Exponent 5 58 Sign 1 63 Mantissa 10 64 Exponent 5 74 Sign 1 79 Mantissa 10 80 Exponent 5 90 Sign 1 95 Mantissa 10 96 Exponent 5 106 Sign 1 111 Mantissa 10 112 Exponent 5 122 Sign 1 127 Mantissa 10 128 Exponent 5 138 Sign 1 143 Mantissa 10 144 Exponent 5 154 Sign 1 159 Mantissa 10 160 Exponent 5 170 Sign 1 175 Mantissa 10 176 Exponent 5 186 Sign 1 191 Mantissa 10 192 Exponent 5 202 Sign 1 207 Mantissa 10 208 Exponent 5 218 Sign 1 223 Mantissa 10 224 Exponent 5 234 Sign 1 239 Mantissa 10 240 Exponent 5 250 Sign 1 255 FMT_NUM_PK16_F32 N/A float 512 16 Mantissa 23 0 Exponent 8 23 Sign 1 31 Mantissa 23 32 Exponent 8 55 Sign 1 63 Mantissa 23 64 Exponent 8 87 Sign 1 95 Mantissa 23 96 Exponent 8 119 Sign 1 127 Mantissa 23 128 Exponent 8 151 Sign 1 159 Mantissa 23 160 Exponent 8 183 Sign 1 191 Mantissa 23 192 Exponent 8 215 Sign 1 223 Mantissa 23 224 Exponent 8 247 Sign 1 255 Mantissa 23 256 Exponent 8 279 Sign 1 287 Mantissa 23 288 Exponent 8 311 Sign 1 319 Mantissa 23 320 Exponent 8 343 Sign 1 351 Mantissa 23 352 Exponent 8 375 Sign 1 383 Mantissa 23 384 Exponent 8 407 Sign 1 415 Mantissa 23 416 Exponent 8 439 Sign 1 447 Mantissa 23 448 Exponent 8 471 Sign 1 479 Mantissa 23 480 Exponent 8 503 Sign 1 511 FMT_NUM_PK16_FP8 N/A float 128 16 Mantissa 3 0 Exponent 4 3 Sign 1 7 Mantissa 3 8 Exponent 4 11 Sign 1 15 Mantissa 3 16 Exponent 4 19 Sign 1 23 Mantissa 3 24 Exponent 4 27 Sign 1 31 Mantissa 3 32 Exponent 4 35 Sign 1 39 Mantissa 3 40 Exponent 4 43 Sign 1 47 Mantissa 3 48 Exponent 4 51 Sign 1 55 Mantissa 3 56 Exponent 4 59 Sign 1 63 Mantissa 3 64 Exponent 4 67 Sign 1 71 Mantissa 3 72 Exponent 4 75 Sign 1 79 Mantissa 3 80 Exponent 4 83 Sign 1 87 Mantissa 3 88 Exponent 4 91 Sign 1 95 Mantissa 3 96 Exponent 4 99 Sign 1 103 Mantissa 3 104 Exponent 4 107 Sign 1 111 Mantissa 3 112 Exponent 4 115 Sign 1 119 Mantissa 3 120 Exponent 4 123 Sign 1 127 FMT_NUM_PK16_I32 N/A integer 512 16 Integer 32 0 Integer 32 32 Integer 32 64 Integer 32 96 Integer 32 128 Integer 32 160 Integer 32 192 Integer 32 224 Integer 32 256 Integer 32 288 Integer 32 320 Integer 32 352 Integer 32 384 Integer 32 416 Integer 32 448 Integer 32 480 FMT_NUM_PK16_I8 N/A integer 128 16 Integer 8 0 Integer 8 8 Integer 8 16 Integer 8 24 Integer 8 32 Integer 8 40 Integer 8 48 Integer 8 56 Integer 8 64 Integer 8 72 Integer 8 80 Integer 8 88 Integer 8 96 Integer 8 104 Integer 8 112 Integer 8 120 FMT_NUM_PK2_B16 N/A bits 32 2 Bits 16 0 Bits 16 16 FMT_NUM_PK2_B32 N/A bits 64 2 Bits 32 0 Bits 32 32 FMT_NUM_PK2_B64 N/A bits 128 2 Bits 64 0 Bits 64 64 FMT_NUM_PK2_BF16 N/A float 32 2 Mantissa 7 0 Exponent 8 7 Sign 1 15 Mantissa 7 16 Exponent 8 23 Sign 1 31 FMT_NUM_PK2_BF8 N/A float 16 2 Mantissa 2 0 Exponent 5 2 Sign 1 7 Mantissa 2 8 Exponent 5 10 Sign 1 15 FMT_NUM_PK2_F16 N/A float 32 2 Mantissa 10 0 Exponent 5 10 Sign 1 15 Mantissa 10 16 Exponent 5 26 Sign 1 31 FMT_NUM_PK2_F32 N/A float 64 2 Mantissa 23 0 Exponent 8 23 Sign 1 31 Mantissa 23 32 Exponent 8 55 Sign 1 63 FMT_NUM_PK2_FP4 N/A float 8 2 Mantissa 1 0 Exponent 2 1 Sign 1 3 Mantissa 1 4 Exponent 2 5 Sign 1 7 FMT_NUM_PK2_FP8 N/A float 16 2 Mantissa 3 0 Exponent 4 3 Sign 1 7 Mantissa 3 8 Exponent 4 11 Sign 1 15 FMT_NUM_PK2_I16 N/A integer 32 2 Integer 16 0 Integer 16 16 FMT_NUM_PK2_I8 N/A integer 16 2 Integer 8 0 Integer 8 8 FMT_NUM_PK2_U16 N/A integer 32 2 Integer 16 0 Integer 16 16 FMT_NUM_PK2_U8 N/A integer 16 2 Integer 8 0 Integer 8 8 FMT_NUM_PK32_BF16 N/A float 512 32 Mantissa 7 0 Exponent 8 7 Sign 1 15 Mantissa 7 16 Exponent 8 23 Sign 1 31 Mantissa 7 32 Exponent 8 39 Sign 1 47 Mantissa 7 48 Exponent 8 55 Sign 1 63 Mantissa 7 64 Exponent 8 71 Sign 1 79 Mantissa 7 80 Exponent 8 87 Sign 1 95 Mantissa 7 96 Exponent 8 103 Sign 1 111 Mantissa 7 112 Exponent 8 119 Sign 1 127 Mantissa 7 128 Exponent 8 135 Sign 1 143 Mantissa 7 144 Exponent 8 151 Sign 1 159 Mantissa 7 160 Exponent 8 167 Sign 1 175 Mantissa 7 176 Exponent 8 183 Sign 1 191 Mantissa 7 192 Exponent 8 199 Sign 1 207 Mantissa 7 208 Exponent 8 215 Sign 1 223 Mantissa 7 224 Exponent 8 231 Sign 1 239 Mantissa 7 240 Exponent 8 247 Sign 1 255 Mantissa 7 256 Exponent 8 263 Sign 1 271 Mantissa 7 272 Exponent 8 279 Sign 1 287 Mantissa 7 288 Exponent 8 295 Sign 1 303 Mantissa 7 304 Exponent 8 311 Sign 1 319 Mantissa 7 320 Exponent 8 327 Sign 1 335 Mantissa 7 336 Exponent 8 343 Sign 1 351 Mantissa 7 352 Exponent 8 359 Sign 1 367 Mantissa 7 368 Exponent 8 375 Sign 1 383 Mantissa 7 384 Exponent 8 391 Sign 1 399 Mantissa 7 400 Exponent 8 407 Sign 1 415 Mantissa 7 416 Exponent 8 423 Sign 1 431 Mantissa 7 432 Exponent 8 439 Sign 1 447 Mantissa 7 448 Exponent 8 455 Sign 1 463 Mantissa 7 464 Exponent 8 471 Sign 1 479 Mantissa 7 480 Exponent 8 487 Sign 1 495 Mantissa 7 496 Exponent 8 503 Sign 1 511 FMT_NUM_PK32_BF6 N/A float 192 32 Mantissa 2 0 Exponent 3 2 Sign 1 5 Mantissa 2 6 Exponent 3 8 Sign 1 11 Mantissa 2 12 Exponent 3 14 Sign 1 17 Mantissa 2 18 Exponent 3 20 Sign 1 23 Mantissa 2 24 Exponent 3 26 Sign 1 29 Mantissa 2 30 Exponent 3 32 Sign 1 35 Mantissa 2 36 Exponent 3 38 Sign 1 41 Mantissa 2 42 Exponent 3 44 Sign 1 47 Mantissa 2 48 Exponent 3 50 Sign 1 53 Mantissa 2 54 Exponent 3 56 Sign 1 59 Mantissa 2 60 Exponent 3 62 Sign 1 65 Mantissa 2 66 Exponent 3 68 Sign 1 71 Mantissa 2 72 Exponent 3 74 Sign 1 77 Mantissa 2 78 Exponent 3 80 Sign 1 83 Mantissa 2 84 Exponent 3 86 Sign 1 89 Mantissa 2 90 Exponent 3 92 Sign 1 95 Mantissa 2 96 Exponent 3 98 Sign 1 101 Mantissa 2 102 Exponent 3 104 Sign 1 107 Mantissa 2 108 Exponent 3 110 Sign 1 113 Mantissa 2 114 Exponent 3 116 Sign 1 119 Mantissa 2 120 Exponent 3 122 Sign 1 125 Mantissa 2 126 Exponent 3 128 Sign 1 131 Mantissa 2 132 Exponent 3 134 Sign 1 137 Mantissa 2 138 Exponent 3 140 Sign 1 143 Mantissa 2 144 Exponent 3 146 Sign 1 149 Mantissa 2 150 Exponent 3 152 Sign 1 155 Mantissa 2 156 Exponent 3 158 Sign 1 161 Mantissa 2 162 Exponent 3 164 Sign 1 167 Mantissa 2 168 Exponent 3 170 Sign 1 173 Mantissa 2 174 Exponent 3 176 Sign 1 179 Mantissa 2 180 Exponent 3 182 Sign 1 185 Mantissa 2 186 Exponent 3 188 Sign 1 191 FMT_NUM_PK32_BF8 N/A float 256 32 Mantissa 2 0 Exponent 5 2 Sign 1 7 Mantissa 2 8 Exponent 5 10 Sign 1 15 Mantissa 2 16 Exponent 5 18 Sign 1 23 Mantissa 2 24 Exponent 5 26 Sign 1 31 Mantissa 2 32 Exponent 5 34 Sign 1 39 Mantissa 2 40 Exponent 5 42 Sign 1 47 Mantissa 2 48 Exponent 5 50 Sign 1 55 Mantissa 2 56 Exponent 5 58 Sign 1 63 Mantissa 2 64 Exponent 5 66 Sign 1 71 Mantissa 2 72 Exponent 5 74 Sign 1 79 Mantissa 2 80 Exponent 5 82 Sign 1 87 Mantissa 2 88 Exponent 5 90 Sign 1 95 Mantissa 2 96 Exponent 5 98 Sign 1 103 Mantissa 2 104 Exponent 5 106 Sign 1 111 Mantissa 2 112 Exponent 5 114 Sign 1 119 Mantissa 2 120 Exponent 5 122 Sign 1 127 Mantissa 2 128 Exponent 5 130 Sign 1 135 Mantissa 2 136 Exponent 5 138 Sign 1 143 Mantissa 2 144 Exponent 5 146 Sign 1 151 Mantissa 2 152 Exponent 5 154 Sign 1 159 Mantissa 2 160 Exponent 5 162 Sign 1 167 Mantissa 2 168 Exponent 5 170 Sign 1 175 Mantissa 2 176 Exponent 5 178 Sign 1 183 Mantissa 2 184 Exponent 5 186 Sign 1 191 Mantissa 2 192 Exponent 5 194 Sign 1 199 Mantissa 2 200 Exponent 5 202 Sign 1 207 Mantissa 2 208 Exponent 5 210 Sign 1 215 Mantissa 2 216 Exponent 5 218 Sign 1 223 Mantissa 2 224 Exponent 5 226 Sign 1 231 Mantissa 2 232 Exponent 5 234 Sign 1 239 Mantissa 2 240 Exponent 5 242 Sign 1 247 Mantissa 2 248 Exponent 5 250 Sign 1 255 FMT_NUM_PK32_F16 N/A float 512 32 Mantissa 10 0 Exponent 5 10 Sign 1 15 Mantissa 10 16 Exponent 5 26 Sign 1 31 Mantissa 10 32 Exponent 5 42 Sign 1 47 Mantissa 10 48 Exponent 5 58 Sign 1 63 Mantissa 10 64 Exponent 5 74 Sign 1 79 Mantissa 10 80 Exponent 5 90 Sign 1 95 Mantissa 10 96 Exponent 5 106 Sign 1 111 Mantissa 10 112 Exponent 5 122 Sign 1 127 Mantissa 10 128 Exponent 5 138 Sign 1 143 Mantissa 10 144 Exponent 5 154 Sign 1 159 Mantissa 10 160 Exponent 5 170 Sign 1 175 Mantissa 10 176 Exponent 5 186 Sign 1 191 Mantissa 10 192 Exponent 5 202 Sign 1 207 Mantissa 10 208 Exponent 5 218 Sign 1 223 Mantissa 10 224 Exponent 5 234 Sign 1 239 Mantissa 10 240 Exponent 5 250 Sign 1 255 Mantissa 10 256 Exponent 5 266 Sign 1 271 Mantissa 10 272 Exponent 5 282 Sign 1 287 Mantissa 10 288 Exponent 5 298 Sign 1 303 Mantissa 10 304 Exponent 5 314 Sign 1 319 Mantissa 10 320 Exponent 5 330 Sign 1 335 Mantissa 10 336 Exponent 5 346 Sign 1 351 Mantissa 10 352 Exponent 5 362 Sign 1 367 Mantissa 10 368 Exponent 5 378 Sign 1 383 Mantissa 10 384 Exponent 5 394 Sign 1 399 Mantissa 10 400 Exponent 5 410 Sign 1 415 Mantissa 10 416 Exponent 5 426 Sign 1 431 Mantissa 10 432 Exponent 5 442 Sign 1 447 Mantissa 10 448 Exponent 5 458 Sign 1 463 Mantissa 10 464 Exponent 5 474 Sign 1 479 Mantissa 10 480 Exponent 5 490 Sign 1 495 Mantissa 10 496 Exponent 5 506 Sign 1 511 FMT_NUM_PK32_F32 N/A float 1024 32 Mantissa 23 0 Exponent 8 23 Sign 1 31 Mantissa 23 32 Exponent 8 55 Sign 1 63 Mantissa 23 64 Exponent 8 87 Sign 1 95 Mantissa 23 96 Exponent 8 119 Sign 1 127 Mantissa 23 128 Exponent 8 151 Sign 1 159 Mantissa 23 160 Exponent 8 183 Sign 1 191 Mantissa 23 192 Exponent 8 215 Sign 1 223 Mantissa 23 224 Exponent 8 247 Sign 1 255 Mantissa 23 256 Exponent 8 279 Sign 1 287 Mantissa 23 288 Exponent 8 311 Sign 1 319 Mantissa 23 320 Exponent 8 343 Sign 1 351 Mantissa 23 352 Exponent 8 375 Sign 1 383 Mantissa 23 384 Exponent 8 407 Sign 1 415 Mantissa 23 416 Exponent 8 439 Sign 1 447 Mantissa 23 448 Exponent 8 471 Sign 1 479 Mantissa 23 480 Exponent 8 503 Sign 1 511 Mantissa 23 512 Exponent 8 535 Sign 1 543 Mantissa 23 544 Exponent 8 567 Sign 1 575 Mantissa 23 576 Exponent 8 599 Sign 1 607 Mantissa 23 608 Exponent 8 631 Sign 1 639 Mantissa 23 640 Exponent 8 663 Sign 1 671 Mantissa 23 672 Exponent 8 695 Sign 1 703 Mantissa 23 704 Exponent 8 727 Sign 1 735 Mantissa 23 736 Exponent 8 759 Sign 1 767 Mantissa 23 768 Exponent 8 791 Sign 1 799 Mantissa 23 800 Exponent 8 823 Sign 1 831 Mantissa 23 832 Exponent 8 855 Sign 1 863 Mantissa 23 864 Exponent 8 887 Sign 1 895 Mantissa 23 896 Exponent 8 919 Sign 1 927 Mantissa 23 928 Exponent 8 951 Sign 1 959 Mantissa 23 960 Exponent 8 983 Sign 1 991 Mantissa 23 992 Exponent 8 1015 Sign 1 1023 FMT_NUM_PK32_FP6 N/A float 192 32 Mantissa 3 0 Exponent 2 3 Sign 1 5 Mantissa 3 6 Exponent 2 9 Sign 1 11 Mantissa 3 12 Exponent 2 15 Sign 1 17 Mantissa 3 18 Exponent 2 21 Sign 1 23 Mantissa 3 24 Exponent 2 27 Sign 1 29 Mantissa 3 30 Exponent 2 33 Sign 1 35 Mantissa 3 36 Exponent 2 39 Sign 1 41 Mantissa 3 42 Exponent 2 45 Sign 1 47 Mantissa 3 48 Exponent 2 51 Sign 1 53 Mantissa 3 54 Exponent 2 57 Sign 1 59 Mantissa 3 60 Exponent 2 63 Sign 1 65 Mantissa 3 66 Exponent 2 69 Sign 1 71 Mantissa 3 72 Exponent 2 75 Sign 1 77 Mantissa 3 78 Exponent 2 81 Sign 1 83 Mantissa 3 84 Exponent 2 87 Sign 1 89 Mantissa 3 90 Exponent 2 93 Sign 1 95 Mantissa 3 96 Exponent 2 99 Sign 1 101 Mantissa 3 102 Exponent 2 105 Sign 1 107 Mantissa 3 108 Exponent 2 111 Sign 1 113 Mantissa 3 114 Exponent 2 117 Sign 1 119 Mantissa 3 120 Exponent 2 123 Sign 1 125 Mantissa 3 126 Exponent 2 129 Sign 1 131 Mantissa 3 132 Exponent 2 135 Sign 1 137 Mantissa 3 138 Exponent 2 141 Sign 1 143 Mantissa 3 144 Exponent 2 147 Sign 1 149 Mantissa 3 150 Exponent 2 153 Sign 1 155 Mantissa 3 156 Exponent 2 159 Sign 1 161 Mantissa 3 162 Exponent 2 165 Sign 1 167 Mantissa 3 168 Exponent 2 171 Sign 1 173 Mantissa 3 174 Exponent 2 177 Sign 1 179 Mantissa 3 180 Exponent 2 183 Sign 1 185 Mantissa 3 186 Exponent 2 189 Sign 1 191 FMT_NUM_PK32_FP8 N/A float 256 32 Mantissa 3 0 Exponent 4 3 Sign 1 7 Mantissa 3 8 Exponent 4 11 Sign 1 15 Mantissa 3 16 Exponent 4 19 Sign 1 23 Mantissa 3 24 Exponent 4 27 Sign 1 31 Mantissa 3 32 Exponent 4 35 Sign 1 39 Mantissa 3 40 Exponent 4 43 Sign 1 47 Mantissa 3 48 Exponent 4 51 Sign 1 55 Mantissa 3 56 Exponent 4 59 Sign 1 63 Mantissa 3 64 Exponent 4 67 Sign 1 71 Mantissa 3 72 Exponent 4 75 Sign 1 79 Mantissa 3 80 Exponent 4 83 Sign 1 87 Mantissa 3 88 Exponent 4 91 Sign 1 95 Mantissa 3 96 Exponent 4 99 Sign 1 103 Mantissa 3 104 Exponent 4 107 Sign 1 111 Mantissa 3 112 Exponent 4 115 Sign 1 119 Mantissa 3 120 Exponent 4 123 Sign 1 127 Mantissa 3 128 Exponent 4 131 Sign 1 135 Mantissa 3 136 Exponent 4 139 Sign 1 143 Mantissa 3 144 Exponent 4 147 Sign 1 151 Mantissa 3 152 Exponent 4 155 Sign 1 159 Mantissa 3 160 Exponent 4 163 Sign 1 167 Mantissa 3 168 Exponent 4 171 Sign 1 175 Mantissa 3 176 Exponent 4 179 Sign 1 183 Mantissa 3 184 Exponent 4 187 Sign 1 191 Mantissa 3 192 Exponent 4 195 Sign 1 199 Mantissa 3 200 Exponent 4 203 Sign 1 207 Mantissa 3 208 Exponent 4 211 Sign 1 215 Mantissa 3 216 Exponent 4 219 Sign 1 223 Mantissa 3 224 Exponent 4 227 Sign 1 231 Mantissa 3 232 Exponent 4 235 Sign 1 239 Mantissa 3 240 Exponent 4 243 Sign 1 247 Mantissa 3 248 Exponent 4 251 Sign 1 255 FMT_NUM_PK32_I32 N/A integer 1024 32 Integer 32 0 Integer 32 32 Integer 32 64 Integer 32 96 Integer 32 128 Integer 32 160 Integer 32 192 Integer 32 224 Integer 32 256 Integer 32 288 Integer 32 320 Integer 32 352 Integer 32 384 Integer 32 416 Integer 32 448 Integer 32 480 Integer 32 512 Integer 32 544 Integer 32 576 Integer 32 608 Integer 32 640 Integer 32 672 Integer 32 704 Integer 32 736 Integer 32 768 Integer 32 800 Integer 32 832 Integer 32 864 Integer 32 896 Integer 32 928 Integer 32 960 Integer 32 992 FMT_NUM_PK32_I8 N/A integer 256 32 Integer 8 0 Integer 8 8 Integer 8 16 Integer 8 24 Integer 8 32 Integer 8 40 Integer 8 48 Integer 8 56 Integer 8 64 Integer 8 72 Integer 8 80 Integer 8 88 Integer 8 96 Integer 8 104 Integer 8 112 Integer 8 120 Integer 8 128 Integer 8 136 Integer 8 144 Integer 8 152 Integer 8 160 Integer 8 168 Integer 8 176 Integer 8 184 Integer 8 192 Integer 8 200 Integer 8 208 Integer 8 216 Integer 8 224 Integer 8 232 Integer 8 240 Integer 8 248 FMT_NUM_PK4_B8 N/A bits 32 4 Bits 8 0 Bits 8 8 Bits 8 16 Bits 8 24 FMT_NUM_PK4_BF16 N/A float 64 4 Mantissa 7 0 Exponent 8 7 Sign 1 15 Mantissa 7 16 Exponent 8 23 Sign 1 31 Mantissa 7 32 Exponent 8 39 Sign 1 47 Mantissa 7 48 Exponent 8 55 Sign 1 63 FMT_NUM_PK4_F16 N/A float 64 4 Mantissa 10 0 Exponent 5 10 Sign 1 15 Mantissa 10 16 Exponent 5 26 Sign 1 31 Mantissa 10 32 Exponent 5 42 Sign 1 47 Mantissa 10 48 Exponent 5 58 Sign 1 63 FMT_NUM_PK4_F32 N/A float 128 4 Mantissa 23 0 Exponent 8 23 Sign 1 31 Mantissa 23 32 Exponent 8 55 Sign 1 63 Mantissa 23 64 Exponent 8 87 Sign 1 95 Mantissa 23 96 Exponent 8 119 Sign 1 127 FMT_NUM_PK4_F64 N/A float 256 4 Mantissa 52 0 Exponent 11 52 Sign 1 63 Mantissa 52 64 Exponent 11 116 Sign 1 127 Mantissa 52 128 Exponent 11 180 Sign 1 191 Mantissa 52 192 Exponent 11 244 Sign 1 255 FMT_NUM_PK4_I32 N/A integer 128 4 Integer 32 0 Integer 32 32 Integer 32 64 Integer 32 96 FMT_NUM_PK4_I8 N/A integer 32 4 Integer 8 0 Integer 8 8 Integer 8 16 Integer 8 24 FMT_NUM_PK4_U8 N/A integer 32 4 Integer 8 0 Integer 8 8 Integer 8 16 Integer 8 24 FMT_NUM_PK8_B32 N/A bits 256 8 Bits 32 0 Bits 32 32 Bits 32 64 Bits 32 96 Bits 32 128 Bits 32 160 Bits 32 192 Bits 32 224 FMT_NUM_PK8_BF16 N/A float 128 8 Mantissa 7 0 Exponent 8 7 Sign 1 15 Mantissa 7 16 Exponent 8 23 Sign 1 31 Mantissa 7 32 Exponent 8 39 Sign 1 47 Mantissa 7 48 Exponent 8 55 Sign 1 63 Mantissa 7 64 Exponent 8 71 Sign 1 79 Mantissa 7 80 Exponent 8 87 Sign 1 95 Mantissa 7 96 Exponent 8 103 Sign 1 111 Mantissa 7 112 Exponent 8 119 Sign 1 127 FMT_NUM_PK8_BF8 N/A float 64 8 Mantissa 2 0 Exponent 5 2 Sign 1 7 Mantissa 2 8 Exponent 5 10 Sign 1 15 Mantissa 2 16 Exponent 5 18 Sign 1 23 Mantissa 2 24 Exponent 5 26 Sign 1 31 Mantissa 2 32 Exponent 5 34 Sign 1 39 Mantissa 2 40 Exponent 5 42 Sign 1 47 Mantissa 2 48 Exponent 5 50 Sign 1 55 Mantissa 2 56 Exponent 5 58 Sign 1 63 FMT_NUM_PK8_F16 N/A float 128 8 Mantissa 10 0 Exponent 5 10 Sign 1 15 Mantissa 10 16 Exponent 5 26 Sign 1 31 Mantissa 10 32 Exponent 5 42 Sign 1 47 Mantissa 10 48 Exponent 5 58 Sign 1 63 Mantissa 10 64 Exponent 5 74 Sign 1 79 Mantissa 10 80 Exponent 5 90 Sign 1 95 Mantissa 10 96 Exponent 5 106 Sign 1 111 Mantissa 10 112 Exponent 5 122 Sign 1 127 FMT_NUM_PK8_FP8 N/A float 64 8 Mantissa 3 0 Exponent 4 3 Sign 1 7 Mantissa 3 8 Exponent 4 11 Sign 1 15 Mantissa 3 16 Exponent 4 19 Sign 1 23 Mantissa 3 24 Exponent 4 27 Sign 1 31 Mantissa 3 32 Exponent 4 35 Sign 1 39 Mantissa 3 40 Exponent 4 43 Sign 1 47 Mantissa 3 48 Exponent 4 51 Sign 1 55 Mantissa 3 56 Exponent 4 59 Sign 1 63 FMT_NUM_PK8_I4 N/A integer 32 8 Integer 4 0 Integer 4 4 Integer 4 8 Integer 4 12 Integer 4 16 Integer 4 20 Integer 4 24 Integer 4 28 FMT_NUM_PK8_I8 N/A integer 64 8 Integer 8 0 Integer 8 8 Integer 8 16 Integer 8 24 Integer 8 32 Integer 8 40 Integer 8 48 Integer 8 56 FMT_NUM_PK8_U4 N/A integer 32 8 Integer 4 0 Integer 4 4 Integer 4 8 Integer 4 12 Integer 4 16 Integer 4 20 Integer 4 24 Integer 4 28 FMT_NUM_PK8_U8 N/A integer 64 8 Integer 8 0 Integer 8 8 Integer 8 16 Integer 8 24 Integer 8 32 Integer 8 40 Integer 8 48 Integer 8 56 FMT_NUM_PK_F16 N/A float 32 2 Mantissa 10 0 Exponent 5 10 Sign 1 15 Mantissa 10 16 Exponent 5 26 Sign 1 31 FMT_NUM_PK_I16 N/A integer 32 2 Integer 16 0 Integer 16 16 FMT_NUM_PK_I8 N/A integer 32 4 Integer 8 0 Integer 8 8 Integer 8 16 Integer 8 24 FMT_NUM_PK_U16 N/A integer 32 2 Integer 16 0 Integer 16 16 FMT_NUM_PK_U8 N/A integer 32 4 Integer 8 0 Integer 8 8 Integer 8 16 Integer 8 24 FMT_NUM_U16 N/A integer 16 1 Integer 16 0 FMT_NUM_U24 N/A integer 24 1 Integer 24 0 FMT_NUM_U32 N/A integer 32 1 Integer 32 0 FMT_NUM_U4 N/A integer 4 1 Integer 4 0 FMT_NUM_U64 N/A integer 64 1 Integer 64 0 FMT_NUM_U8 N/A integer 8 1 Integer 8 0 FMT_RSRC 128-bit buffer resource constant. descriptor 128 1 Descriptor 128 0 FMT_RSRC_SCALAR 128-bit buffer resource constant. descriptor 128 1 Descriptor 128 0 FMT_RSRC_SCRATCH 128-bit buffer resource constant. descriptor 128 1 Descriptor 128 0 FMT_RSRC_SCRATCH_BYTE 128-bit buffer resource constant. descriptor 128 1 Descriptor 128 0 FMT_RSRC_SCRATCH_STRIDE 128-bit buffer resource constant. descriptor 128 1 Descriptor 128 0 FMT_RSRC_TYPED 128-bit buffer resource constant. descriptor 128 1 Descriptor 128 0 FMT_RSRC_TYPED_BYTE 128-bit buffer resource constant. descriptor 128 1 Descriptor 128 0 FMT_RSRC_TYPED_SCRATCH 128-bit buffer resource constant. descriptor 128 1 Descriptor 128 0 FMT_RSRC_TYPED_STRIDE 128-bit buffer resource constant. descriptor 128 1 Descriptor 128 0 FMT_RSRC_VECTOR 128-bit buffer resource constant. descriptor 128 1 Descriptor 128 0 FMT_RSRC_VECTOR_BYTE 128-bit buffer resource constant. descriptor 128 1 Descriptor 128 0 FMT_RSRC_VECTOR_STRIDE 128-bit buffer resource constant. descriptor 128 1 Descriptor 128 0 FMT_SAMP 128-bit image sampler constant. descriptor 128 1 Descriptor 128 0 OPR_ACCVGPR N/A acc0 Accumulator VGPR 0. Increment from here for additional GPRs. There are NUM_ACCVGPR ACCVGPRs in total. Alignment restriction: unless otherwise noted, 64-bit and wider ACCVGPR values must be even-aligned (first ACCVGPR address is even). 512 acc1 <p>See above.</p> 513 acc2 <p>See above.</p> 514 acc3 <p>See above.</p> 515 acc4 <p>See above.</p> 516 acc5 <p>See above.</p> 517 acc6 <p>See above.</p> 518 acc7 <p>See above.</p> 519 acc8 <p>See above.</p> 520 acc9 <p>See above.</p> 521 acc10 <p>See above.</p> 522 acc11 <p>See above.</p> 523 acc12 <p>See above.</p> 524 acc13 <p>See above.</p> 525 acc14 <p>See above.</p> 526 acc15 <p>See above.</p> 527 acc16 <p>See above.</p> 528 acc17 <p>See above.</p> 529 acc18 <p>See above.</p> 530 acc19 <p>See above.</p> 531 acc20 <p>See above.</p> 532 acc21 <p>See above.</p> 533 acc22 <p>See above.</p> 534 acc23 <p>See above.</p> 535 acc24 <p>See above.</p> 536 acc25 <p>See above.</p> 537 acc26 <p>See above.</p> 538 acc27 <p>See above.</p> 539 acc28 <p>See above.</p> 540 acc29 <p>See above.</p> 541 acc30 <p>See above.</p> 542 acc31 <p>See above.</p> 543 acc32 <p>See above.</p> 544 acc33 <p>See above.</p> 545 acc34 <p>See above.</p> 546 acc35 <p>See above.</p> 547 acc36 <p>See above.</p> 548 acc37 <p>See above.</p> 549 acc38 <p>See above.</p> 550 acc39 <p>See above.</p> 551 acc40 <p>See above.</p> 552 acc41 <p>See above.</p> 553 acc42 <p>See above.</p> 554 acc43 <p>See above.</p> 555 acc44 <p>See above.</p> 556 acc45 <p>See above.</p> 557 acc46 <p>See above.</p> 558 acc47 <p>See above.</p> 559 acc48 <p>See above.</p> 560 acc49 <p>See above.</p> 561 acc50 <p>See above.</p> 562 acc51 <p>See above.</p> 563 acc52 <p>See above.</p> 564 acc53 <p>See above.</p> 565 acc54 <p>See above.</p> 566 acc55 <p>See above.</p> 567 acc56 <p>See above.</p> 568 acc57 <p>See above.</p> 569 acc58 <p>See above.</p> 570 acc59 <p>See above.</p> 571 acc60 <p>See above.</p> 572 acc61 <p>See above.</p> 573 acc62 <p>See above.</p> 574 acc63 <p>See above.</p> 575 acc64 <p>See above.</p> 576 acc65 <p>See above.</p> 577 acc66 <p>See above.</p> 578 acc67 <p>See above.</p> 579 acc68 <p>See above.</p> 580 acc69 <p>See above.</p> 581 acc70 <p>See above.</p> 582 acc71 <p>See above.</p> 583 acc72 <p>See above.</p> 584 acc73 <p>See above.</p> 585 acc74 <p>See above.</p> 586 acc75 <p>See above.</p> 587 acc76 <p>See above.</p> 588 acc77 <p>See above.</p> 589 acc78 <p>See above.</p> 590 acc79 <p>See above.</p> 591 acc80 <p>See above.</p> 592 acc81 <p>See above.</p> 593 acc82 <p>See above.</p> 594 acc83 <p>See above.</p> 595 acc84 <p>See above.</p> 596 acc85 <p>See above.</p> 597 acc86 <p>See above.</p> 598 acc87 <p>See above.</p> 599 acc88 <p>See above.</p> 600 acc89 <p>See above.</p> 601 acc90 <p>See above.</p> 602 acc91 <p>See above.</p> 603 acc92 <p>See above.</p> 604 acc93 <p>See above.</p> 605 acc94 <p>See above.</p> 606 acc95 <p>See above.</p> 607 acc96 <p>See above.</p> 608 acc97 <p>See above.</p> 609 acc98 <p>See above.</p> 610 acc99 <p>See above.</p> 611 acc100 <p>See above.</p> 612 acc101 <p>See above.</p> 613 acc102 <p>See above.</p> 614 acc103 <p>See above.</p> 615 acc104 <p>See above.</p> 616 acc105 <p>See above.</p> 617 acc106 <p>See above.</p> 618 acc107 <p>See above.</p> 619 acc108 <p>See above.</p> 620 acc109 <p>See above.</p> 621 acc110 <p>See above.</p> 622 acc111 <p>See above.</p> 623 acc112 <p>See above.</p> 624 acc113 <p>See above.</p> 625 acc114 <p>See above.</p> 626 acc115 <p>See above.</p> 627 acc116 <p>See above.</p> 628 acc117 <p>See above.</p> 629 acc118 <p>See above.</p> 630 acc119 <p>See above.</p> 631 acc120 <p>See above.</p> 632 acc121 <p>See above.</p> 633 acc122 <p>See above.</p> 634 acc123 <p>See above.</p> 635 acc124 <p>See above.</p> 636 acc125 <p>See above.</p> 637 acc126 <p>See above.</p> 638 acc127 <p>See above.</p> 639 acc128 <p>See above.</p> 640 acc129 <p>See above.</p> 641 acc130 <p>See above.</p> 642 acc131 <p>See above.</p> 643 acc132 <p>See above.</p> 644 acc133 <p>See above.</p> 645 acc134 <p>See above.</p> 646 acc135 <p>See above.</p> 647 acc136 <p>See above.</p> 648 acc137 <p>See above.</p> 649 acc138 <p>See above.</p> 650 acc139 <p>See above.</p> 651 acc140 <p>See above.</p> 652 acc141 <p>See above.</p> 653 acc142 <p>See above.</p> 654 acc143 <p>See above.</p> 655 acc144 <p>See above.</p> 656 acc145 <p>See above.</p> 657 acc146 <p>See above.</p> 658 acc147 <p>See above.</p> 659 acc148 <p>See above.</p> 660 acc149 <p>See above.</p> 661 acc150 <p>See above.</p> 662 acc151 <p>See above.</p> 663 acc152 <p>See above.</p> 664 acc153 <p>See above.</p> 665 acc154 <p>See above.</p> 666 acc155 <p>See above.</p> 667 acc156 <p>See above.</p> 668 acc157 <p>See above.</p> 669 acc158 <p>See above.</p> 670 acc159 <p>See above.</p> 671 acc160 <p>See above.</p> 672 acc161 <p>See above.</p> 673 acc162 <p>See above.</p> 674 acc163 <p>See above.</p> 675 acc164 <p>See above.</p> 676 acc165 <p>See above.</p> 677 acc166 <p>See above.</p> 678 acc167 <p>See above.</p> 679 acc168 <p>See above.</p> 680 acc169 <p>See above.</p> 681 acc170 <p>See above.</p> 682 acc171 <p>See above.</p> 683 acc172 <p>See above.</p> 684 acc173 <p>See above.</p> 685 acc174 <p>See above.</p> 686 acc175 <p>See above.</p> 687 acc176 <p>See above.</p> 688 acc177 <p>See above.</p> 689 acc178 <p>See above.</p> 690 acc179 <p>See above.</p> 691 acc180 <p>See above.</p> 692 acc181 <p>See above.</p> 693 acc182 <p>See above.</p> 694 acc183 <p>See above.</p> 695 acc184 <p>See above.</p> 696 acc185 <p>See above.</p> 697 acc186 <p>See above.</p> 698 acc187 <p>See above.</p> 699 acc188 <p>See above.</p> 700 acc189 <p>See above.</p> 701 acc190 <p>See above.</p> 702 acc191 <p>See above.</p> 703 acc192 <p>See above.</p> 704 acc193 <p>See above.</p> 705 acc194 <p>See above.</p> 706 acc195 <p>See above.</p> 707 acc196 <p>See above.</p> 708 acc197 <p>See above.</p> 709 acc198 <p>See above.</p> 710 acc199 <p>See above.</p> 711 acc200 <p>See above.</p> 712 acc201 <p>See above.</p> 713 acc202 <p>See above.</p> 714 acc203 <p>See above.</p> 715 acc204 <p>See above.</p> 716 acc205 <p>See above.</p> 717 acc206 <p>See above.</p> 718 acc207 <p>See above.</p> 719 acc208 <p>See above.</p> 720 acc209 <p>See above.</p> 721 acc210 <p>See above.</p> 722 acc211 <p>See above.</p> 723 acc212 <p>See above.</p> 724 acc213 <p>See above.</p> 725 acc214 <p>See above.</p> 726 acc215 <p>See above.</p> 727 acc216 <p>See above.</p> 728 acc217 <p>See above.</p> 729 acc218 <p>See above.</p> 730 acc219 <p>See above.</p> 731 acc220 <p>See above.</p> 732 acc221 <p>See above.</p> 733 acc222 <p>See above.</p> 734 acc223 <p>See above.</p> 735 acc224 <p>See above.</p> 736 acc225 <p>See above.</p> 737 acc226 <p>See above.</p> 738 acc227 <p>See above.</p> 739 acc228 <p>See above.</p> 740 acc229 <p>See above.</p> 741 acc230 <p>See above.</p> 742 acc231 <p>See above.</p> 743 acc232 <p>See above.</p> 744 acc233 <p>See above.</p> 745 acc234 <p>See above.</p> 746 acc235 <p>See above.</p> 747 acc236 <p>See above.</p> 748 acc237 <p>See above.</p> 749 acc238 <p>See above.</p> 750 acc239 <p>See above.</p> 751 acc240 <p>See above.</p> 752 acc241 <p>See above.</p> 753 acc242 <p>See above.</p> 754 acc243 <p>See above.</p> 755 acc244 <p>See above.</p> 756 acc245 <p>See above.</p> 757 acc246 <p>See above.</p> 758 acc247 <p>See above.</p> 759 acc248 <p>See above.</p> 760 acc249 <p>See above.</p> 761 acc250 <p>See above.</p> 762 acc251 <p>See above.</p> 763 acc252 <p>See above.</p> 764 acc253 <p>See above.</p> 765 acc254 <p>See above.</p> 766 acc255 <p>See above.</p> 767 OPR_DSMEM Describes an operation that reads or writes either GDS or LDS memory. dsmem I/O to DS memory 0 OPR_FLAT_SCRATCH N/A FLAT_SCRATCH_ALL flat_scratch[63:0] 0 OPR_GPUMEM Describes an operation that reads or writes GPU memory. gpumem I/O to GPU memory 0 OPR_HWREG N/A ID N/A 6 0 HW_REG_MODE N/A 1 HW_REG_STATUS N/A 2 HW_REG_TRAPSTS N/A 3 HW_REG_HW_ID N/A 4 HW_REG_GPR_ALLOC N/A 5 HW_REG_LDS_ALLOC N/A 6 HW_REG_IB_STS N/A 7 HW_REG_PC_LO N/A 8 HW_REG_PC_HI N/A 9 HW_REG_INST_DW0 N/A 10 HW_REG_INST_DW1 N/A 11 HW_REG_IB_DBG0 N/A 12 HW_REG_IB_DBG1 N/A 13 HW_REG_FLUSH_IB N/A 14 HW_REG_SH_MEM_BASES N/A 15 HW_REG_SQ_SHADER_TBA_LO N/A 16 HW_REG_SQ_SHADER_TBA_HI N/A 17 HW_REG_SQ_SHADER_TMA_LO N/A 18 HW_REG_SQ_SHADER_TMA_HI N/A 19 HW_REG_XCC_ID N/A 20 HW_REG_SQ_PERF_SNAPSHOT_DATA N/A 21 HW_REG_SQ_PERF_SNAPSHOT_DATA1 N/A 22 HW_REG_SQ_PERF_SNAPSHOT_PC_LO N/A 23 HW_REG_SQ_PERF_SNAPSHOT_PC_HI N/A 24 OFFSET N/A 5 6 SIZE N/A 5 11 OPR_LABEL A relative jump offset. TARGET Singed jump target relative to PC+4, in dwords. 16 0 OPR_PC N/A PC_ALL pc[63:0] 0 OPR_SDST N/A OPR_SDST_EXEC OPR_SDST_M0 OPR_SREG OPR_SREG_NOVCC s0 Scalar GPR 0. Increment from here for additional GPRs. There are NUM_SGPR SGPRs in total. Alignment restriction: unless otherwise noted, 64-bit SGPR values must be even-aligned (first SGPR address is even); 128-bit and wider SGPR values must be aligned to a multiple of 4. 0 s1 <p>See above.</p> 1 s2 <p>See above.</p> 2 s3 <p>See above.</p> 3 s4 <p>See above.</p> 4 s5 <p>See above.</p> 5 s6 <p>See above.</p> 6 s7 <p>See above.</p> 7 s8 <p>See above.</p> 8 s9 <p>See above.</p> 9 s10 <p>See above.</p> 10 s11 <p>See above.</p> 11 s12 <p>See above.</p> 12 s13 <p>See above.</p> 13 s14 <p>See above.</p> 14 s15 <p>See above.</p> 15 s16 <p>See above.</p> 16 s17 <p>See above.</p> 17 s18 <p>See above.</p> 18 s19 <p>See above.</p> 19 s20 <p>See above.</p> 20 s21 <p>See above.</p> 21 s22 <p>See above.</p> 22 s23 <p>See above.</p> 23 s24 <p>See above.</p> 24 s25 <p>See above.</p> 25 s26 <p>See above.</p> 26 s27 <p>See above.</p> 27 s28 <p>See above.</p> 28 s29 <p>See above.</p> 29 s30 <p>See above.</p> 30 s31 <p>See above.</p> 31 s32 <p>See above.</p> 32 s33 <p>See above.</p> 33 s34 <p>See above.</p> 34 s35 <p>See above.</p> 35 s36 <p>See above.</p> 36 s37 <p>See above.</p> 37 s38 <p>See above.</p> 38 s39 <p>See above.</p> 39 s40 <p>See above.</p> 40 s41 <p>See above.</p> 41 s42 <p>See above.</p> 42 s43 <p>See above.</p> 43 s44 <p>See above.</p> 44 s45 <p>See above.</p> 45 s46 <p>See above.</p> 46 s47 <p>See above.</p> 47 s48 <p>See above.</p> 48 s49 <p>See above.</p> 49 s50 <p>See above.</p> 50 s51 <p>See above.</p> 51 s52 <p>See above.</p> 52 s53 <p>See above.</p> 53 s54 <p>See above.</p> 54 s55 <p>See above.</p> 55 s56 <p>See above.</p> 56 s57 <p>See above.</p> 57 s58 <p>See above.</p> 58 s59 <p>See above.</p> 59 s60 <p>See above.</p> 60 s61 <p>See above.</p> 61 s62 <p>See above.</p> 62 s63 <p>See above.</p> 63 s64 <p>See above.</p> 64 s65 <p>See above.</p> 65 s66 <p>See above.</p> 66 s67 <p>See above.</p> 67 s68 <p>See above.</p> 68 s69 <p>See above.</p> 69 s70 <p>See above.</p> 70 s71 <p>See above.</p> 71 s72 <p>See above.</p> 72 s73 <p>See above.</p> 73 s74 <p>See above.</p> 74 s75 <p>See above.</p> 75 s76 <p>See above.</p> 76 s77 <p>See above.</p> 77 s78 <p>See above.</p> 78 s79 <p>See above.</p> 79 s80 <p>See above.</p> 80 s81 <p>See above.</p> 81 s82 <p>See above.</p> 82 s83 <p>See above.</p> 83 s84 <p>See above.</p> 84 s85 <p>See above.</p> 85 s86 <p>See above.</p> 86 s87 <p>See above.</p> 87 s88 <p>See above.</p> 88 s89 <p>See above.</p> 89 s90 <p>See above.</p> 90 s91 <p>See above.</p> 91 s92 <p>See above.</p> 92 s93 <p>See above.</p> 93 s94 <p>See above.</p> 94 s95 <p>See above.</p> 95 s96 <p>See above.</p> 96 s97 <p>See above.</p> 97 s98 <p>See above.</p> 98 s99 <p>See above.</p> 99 s100 <p>See above.</p> 100 s101 <p>See above.</p> 101 XNACK_MASK_LO xnack_mask[31:0], see XNACK replay mechanism for details. 104 XNACK_MASK_HI xnack_mask[63:32], see XNACK replay mechanism for details. 105 FLAT_SCRATCH_LO {13'd0, size[18:0]} 102 FLAT_SCRATCH_HI {8'd0, offset[31:8]} 103 TTMP0 Trap handler temps (privileged). Increment from here for additional TTMPs. There are NUM_TTMP TTMPs in total. {TTMP1,TTMP0} = PC_save{hi,lo}. Alignment restriction: unless otherwise noted, 64-bit TTMP values must be even-aligned (first TTMP address is even); 128-bit and wider TTMP values must be aligned to a multiple of 4. 108 TTMP1 Trap handler temps (privileged). 109 TTMP2 Trap handler temps (privileged). 110 TTMP3 Trap handler temps (privileged). 111 TTMP4 Trap handler temps (privileged). 112 TTMP5 Trap handler temps (privileged). 113 TTMP6 Trap handler temps (privileged). 114 TTMP7 Trap handler temps (privileged). 115 TTMP8 Trap handler temps (privileged). 116 TTMP9 Trap handler temps (privileged). 117 TTMP10 Trap handler temps (privileged). 118 TTMP11 Trap handler temps (privileged). 119 TTMP12 Trap handler temps (privileged). 120 TTMP13 Trap handler temps (privileged). 121 TTMP14 Trap handler temps (privileged). 122 TTMP15 Trap handler temps (privileged). 123 VCC_LO VCC[31:0] 106 VCC_HI VCC[63:32] 107 M0 Special register used to hold LDS/GDS addresses, relative indices, and send-message values. 124 EXEC_LO exec[31:0] 126 EXEC_HI exec[63:32] 127 OPR_SDST_EXEC N/A EXEC_LO exec[31:0] 126 EXEC_HI exec[63:32] 127 OPR_SDST_M0 N/A M0 Special register used to hold LDS/GDS addresses, relative indices, and send-message values. 124 OPR_SENDMSG N/A GSOP N/A 2 4 GS_OP_NOP N/A 0 GS_OP_CUT N/A 1 GS_OP_EMIT N/A 2 GS_OP_EMIT_CUT N/A 3 MSG N/A 4 0 MSG_INTERRUPT N/A 1 MSG_GS N/A 2 MSG_GS_DONE N/A 3 MSG_SAVEWAVE N/A 4 MSG_STALL_WAVE_GEN N/A 5 MSG_HALT_WAVES N/A 6 MSG_ORDERED_PS_DONE N/A 7 MSG_EARLY_PRIM_DEALLOC N/A 8 MSG_GS_ALLOC_REQ N/A 9 MSG_GET_DOORBELL N/A 10 MSG_SYSMSG N/A 15 STREAMID N/A 2 8 OPR_SIMM16 16-bit integer constant. VALUE Integer constant value. 16 0 OPR_SIMM32 32-bit integer constant. VALUE Integer constant value. 32 0 OPR_SIMM4 4-bit integer constant. VALUE Integer constant value. 4 0 OPR_SIMM8 8-bit integer constant. VALUE Integer constant value. 8 0 OPR_SMEM_OFFSET N/A OPR_SDST_M0 OPR_SREG OPR_SREG_NOVCC s0 Scalar GPR 0. Increment from here for additional GPRs. There are NUM_SGPR SGPRs in total. Alignment restriction: unless otherwise noted, 64-bit SGPR values must be even-aligned (first SGPR address is even); 128-bit and wider SGPR values must be aligned to a multiple of 4. 0 s1 <p>See above.</p> 1 s2 <p>See above.</p> 2 s3 <p>See above.</p> 3 s4 <p>See above.</p> 4 s5 <p>See above.</p> 5 s6 <p>See above.</p> 6 s7 <p>See above.</p> 7 s8 <p>See above.</p> 8 s9 <p>See above.</p> 9 s10 <p>See above.</p> 10 s11 <p>See above.</p> 11 s12 <p>See above.</p> 12 s13 <p>See above.</p> 13 s14 <p>See above.</p> 14 s15 <p>See above.</p> 15 s16 <p>See above.</p> 16 s17 <p>See above.</p> 17 s18 <p>See above.</p> 18 s19 <p>See above.</p> 19 s20 <p>See above.</p> 20 s21 <p>See above.</p> 21 s22 <p>See above.</p> 22 s23 <p>See above.</p> 23 s24 <p>See above.</p> 24 s25 <p>See above.</p> 25 s26 <p>See above.</p> 26 s27 <p>See above.</p> 27 s28 <p>See above.</p> 28 s29 <p>See above.</p> 29 s30 <p>See above.</p> 30 s31 <p>See above.</p> 31 s32 <p>See above.</p> 32 s33 <p>See above.</p> 33 s34 <p>See above.</p> 34 s35 <p>See above.</p> 35 s36 <p>See above.</p> 36 s37 <p>See above.</p> 37 s38 <p>See above.</p> 38 s39 <p>See above.</p> 39 s40 <p>See above.</p> 40 s41 <p>See above.</p> 41 s42 <p>See above.</p> 42 s43 <p>See above.</p> 43 s44 <p>See above.</p> 44 s45 <p>See above.</p> 45 s46 <p>See above.</p> 46 s47 <p>See above.</p> 47 s48 <p>See above.</p> 48 s49 <p>See above.</p> 49 s50 <p>See above.</p> 50 s51 <p>See above.</p> 51 s52 <p>See above.</p> 52 s53 <p>See above.</p> 53 s54 <p>See above.</p> 54 s55 <p>See above.</p> 55 s56 <p>See above.</p> 56 s57 <p>See above.</p> 57 s58 <p>See above.</p> 58 s59 <p>See above.</p> 59 s60 <p>See above.</p> 60 s61 <p>See above.</p> 61 s62 <p>See above.</p> 62 s63 <p>See above.</p> 63 s64 <p>See above.</p> 64 s65 <p>See above.</p> 65 s66 <p>See above.</p> 66 s67 <p>See above.</p> 67 s68 <p>See above.</p> 68 s69 <p>See above.</p> 69 s70 <p>See above.</p> 70 s71 <p>See above.</p> 71 s72 <p>See above.</p> 72 s73 <p>See above.</p> 73 s74 <p>See above.</p> 74 s75 <p>See above.</p> 75 s76 <p>See above.</p> 76 s77 <p>See above.</p> 77 s78 <p>See above.</p> 78 s79 <p>See above.</p> 79 s80 <p>See above.</p> 80 s81 <p>See above.</p> 81 s82 <p>See above.</p> 82 s83 <p>See above.</p> 83 s84 <p>See above.</p> 84 s85 <p>See above.</p> 85 s86 <p>See above.</p> 86 s87 <p>See above.</p> 87 s88 <p>See above.</p> 88 s89 <p>See above.</p> 89 s90 <p>See above.</p> 90 s91 <p>See above.</p> 91 s92 <p>See above.</p> 92 s93 <p>See above.</p> 93 s94 <p>See above.</p> 94 s95 <p>See above.</p> 95 s96 <p>See above.</p> 96 s97 <p>See above.</p> 97 s98 <p>See above.</p> 98 s99 <p>See above.</p> 99 s100 <p>See above.</p> 100 s101 <p>See above.</p> 101 XNACK_MASK_LO xnack_mask[31:0], see XNACK replay mechanism for details. 104 XNACK_MASK_HI xnack_mask[63:32], see XNACK replay mechanism for details. 105 FLAT_SCRATCH_LO {13'd0, size[18:0]} 102 FLAT_SCRATCH_HI {8'd0, offset[31:8]} 103 TTMP0 Trap handler temps (privileged). Increment from here for additional TTMPs. There are NUM_TTMP TTMPs in total. {TTMP1,TTMP0} = PC_save{hi,lo}. Alignment restriction: unless otherwise noted, 64-bit TTMP values must be even-aligned (first TTMP address is even); 128-bit and wider TTMP values must be aligned to a multiple of 4. 108 TTMP1 Trap handler temps (privileged). 109 TTMP2 Trap handler temps (privileged). 110 TTMP3 Trap handler temps (privileged). 111 TTMP4 Trap handler temps (privileged). 112 TTMP5 Trap handler temps (privileged). 113 TTMP6 Trap handler temps (privileged). 114 TTMP7 Trap handler temps (privileged). 115 TTMP8 Trap handler temps (privileged). 116 TTMP9 Trap handler temps (privileged). 117 TTMP10 Trap handler temps (privileged). 118 TTMP11 Trap handler temps (privileged). 119 TTMP12 Trap handler temps (privileged). 120 TTMP13 Trap handler temps (privileged). 121 TTMP14 Trap handler temps (privileged). 122 TTMP15 Trap handler temps (privileged). 123 VCC_LO VCC[31:0] 106 VCC_HI VCC[63:32] 107 M0 Special register used to hold LDS/GDS addresses, relative indices, and send-message values. 124 OPR_SRC N/A OPR_SDST OPR_SDST_EXEC OPR_SDST_M0 OPR_SRC_NOLIT OPR_SRC_SIMPLE OPR_SRC_VGPR OPR_SREG OPR_SREG_NOVCC OPR_SSRC_NOLIT OPR_SSRC_SPECIAL_SCC s0 Scalar GPR 0. Increment from here for additional GPRs. There are NUM_SGPR SGPRs in total. Alignment restriction: unless otherwise noted, 64-bit SGPR values must be even-aligned (first SGPR address is even); 128-bit and wider SGPR values must be aligned to a multiple of 4. 0 s1 <p>See above.</p> 1 s2 <p>See above.</p> 2 s3 <p>See above.</p> 3 s4 <p>See above.</p> 4 s5 <p>See above.</p> 5 s6 <p>See above.</p> 6 s7 <p>See above.</p> 7 s8 <p>See above.</p> 8 s9 <p>See above.</p> 9 s10 <p>See above.</p> 10 s11 <p>See above.</p> 11 s12 <p>See above.</p> 12 s13 <p>See above.</p> 13 s14 <p>See above.</p> 14 s15 <p>See above.</p> 15 s16 <p>See above.</p> 16 s17 <p>See above.</p> 17 s18 <p>See above.</p> 18 s19 <p>See above.</p> 19 s20 <p>See above.</p> 20 s21 <p>See above.</p> 21 s22 <p>See above.</p> 22 s23 <p>See above.</p> 23 s24 <p>See above.</p> 24 s25 <p>See above.</p> 25 s26 <p>See above.</p> 26 s27 <p>See above.</p> 27 s28 <p>See above.</p> 28 s29 <p>See above.</p> 29 s30 <p>See above.</p> 30 s31 <p>See above.</p> 31 s32 <p>See above.</p> 32 s33 <p>See above.</p> 33 s34 <p>See above.</p> 34 s35 <p>See above.</p> 35 s36 <p>See above.</p> 36 s37 <p>See above.</p> 37 s38 <p>See above.</p> 38 s39 <p>See above.</p> 39 s40 <p>See above.</p> 40 s41 <p>See above.</p> 41 s42 <p>See above.</p> 42 s43 <p>See above.</p> 43 s44 <p>See above.</p> 44 s45 <p>See above.</p> 45 s46 <p>See above.</p> 46 s47 <p>See above.</p> 47 s48 <p>See above.</p> 48 s49 <p>See above.</p> 49 s50 <p>See above.</p> 50 s51 <p>See above.</p> 51 s52 <p>See above.</p> 52 s53 <p>See above.</p> 53 s54 <p>See above.</p> 54 s55 <p>See above.</p> 55 s56 <p>See above.</p> 56 s57 <p>See above.</p> 57 s58 <p>See above.</p> 58 s59 <p>See above.</p> 59 s60 <p>See above.</p> 60 s61 <p>See above.</p> 61 s62 <p>See above.</p> 62 s63 <p>See above.</p> 63 s64 <p>See above.</p> 64 s65 <p>See above.</p> 65 s66 <p>See above.</p> 66 s67 <p>See above.</p> 67 s68 <p>See above.</p> 68 s69 <p>See above.</p> 69 s70 <p>See above.</p> 70 s71 <p>See above.</p> 71 s72 <p>See above.</p> 72 s73 <p>See above.</p> 73 s74 <p>See above.</p> 74 s75 <p>See above.</p> 75 s76 <p>See above.</p> 76 s77 <p>See above.</p> 77 s78 <p>See above.</p> 78 s79 <p>See above.</p> 79 s80 <p>See above.</p> 80 s81 <p>See above.</p> 81 s82 <p>See above.</p> 82 s83 <p>See above.</p> 83 s84 <p>See above.</p> 84 s85 <p>See above.</p> 85 s86 <p>See above.</p> 86 s87 <p>See above.</p> 87 s88 <p>See above.</p> 88 s89 <p>See above.</p> 89 s90 <p>See above.</p> 90 s91 <p>See above.</p> 91 s92 <p>See above.</p> 92 s93 <p>See above.</p> 93 s94 <p>See above.</p> 94 s95 <p>See above.</p> 95 s96 <p>See above.</p> 96 s97 <p>See above.</p> 97 s98 <p>See above.</p> 98 s99 <p>See above.</p> 99 s100 <p>See above.</p> 100 s101 <p>See above.</p> 101 XNACK_MASK_LO xnack_mask[31:0], see XNACK replay mechanism for details. 104 XNACK_MASK_HI xnack_mask[63:32], see XNACK replay mechanism for details. 105 FLAT_SCRATCH_LO {13'd0, size[18:0]} 102 FLAT_SCRATCH_HI {8'd0, offset[31:8]} 103 TTMP0 Trap handler temps (privileged). Increment from here for additional TTMPs. There are NUM_TTMP TTMPs in total. {TTMP1,TTMP0} = PC_save{hi,lo}. Alignment restriction: unless otherwise noted, 64-bit TTMP values must be even-aligned (first TTMP address is even); 128-bit and wider TTMP values must be aligned to a multiple of 4. 108 TTMP1 Trap handler temps (privileged). 109 TTMP2 Trap handler temps (privileged). 110 TTMP3 Trap handler temps (privileged). 111 TTMP4 Trap handler temps (privileged). 112 TTMP5 Trap handler temps (privileged). 113 TTMP6 Trap handler temps (privileged). 114 TTMP7 Trap handler temps (privileged). 115 TTMP8 Trap handler temps (privileged). 116 TTMP9 Trap handler temps (privileged). 117 TTMP10 Trap handler temps (privileged). 118 TTMP11 Trap handler temps (privileged). 119 TTMP12 Trap handler temps (privileged). 120 TTMP13 Trap handler temps (privileged). 121 TTMP14 Trap handler temps (privileged). 122 TTMP15 Trap handler temps (privileged). 123 VCC_LO VCC[31:0] 106 VCC_HI VCC[63:32] 107 M0 Special register used to hold LDS/GDS addresses, relative indices, and send-message values. 124 EXEC_LO exec[31:0] 126 EXEC_HI exec[63:32] 127 0 0 128 1 1 (integer) 129 2 2 (integer) 130 3 3 (integer) 131 4 4 (integer) 132 5 5 (integer) 133 6 6 (integer) 134 7 7 (integer) 135 8 8 (integer) 136 9 9 (integer) 137 10 10 (integer) 138 11 11 (integer) 139 12 12 (integer) 140 13 13 (integer) 141 14 14 (integer) 142 15 15 (integer) 143 16 16 (integer) 144 17 17 (integer) 145 18 18 (integer) 146 19 19 (integer) 147 20 20 (integer) 148 21 21 (integer) 149 22 22 (integer) 150 23 23 (integer) 151 24 24 (integer) 152 25 25 (integer) 153 26 26 (integer) 154 27 27 (integer) 155 28 28 (integer) 156 29 29 (integer) 157 30 30 (integer) 158 31 31 (integer) 159 32 32 (integer) 160 33 33 (integer) 161 34 34 (integer) 162 35 35 (integer) 163 36 36 (integer) 164 37 37 (integer) 165 38 38 (integer) 166 39 39 (integer) 167 40 40 (integer) 168 41 41 (integer) 169 42 42 (integer) 170 43 43 (integer) 171 44 44 (integer) 172 45 45 (integer) 173 46 46 (integer) 174 47 47 (integer) 175 48 48 (integer) 176 49 49 (integer) 177 50 50 (integer) 178 51 51 (integer) 179 52 52 (integer) 180 53 53 (integer) 181 54 54 (integer) 182 55 55 (integer) 183 56 56 (integer) 184 57 57 (integer) 185 58 58 (integer) 186 59 59 (integer) 187 60 60 (integer) 188 61 61 (integer) 189 62 62 (integer) 190 63 63 (integer) 191 64 64 (integer) 192 -1 -1 (integer) 193 -2 -2 (integer) 194 -3 -3 (integer) 195 -4 -4 (integer) 196 -5 -5 (integer) 197 -6 -6 (integer) 198 -7 -7 (integer) 199 -8 -8 (integer) 200 -9 -9 (integer) 201 -10 -10 (integer) 202 -11 -11 (integer) 203 -12 -12 (integer) 204 -13 -13 (integer) 205 -14 -14 (integer) 206 -15 -15 (integer) 207 -16 -16 (integer) 208 0.5 0.5 240 -0.5 -0.5 241 1.0 1.0 242 -1.0 -1.0 243 2.0 2.0 244 -2.0 -2.0 245 4.0 4.0 246 -4.0 -4.0 247 0.15915494 1 / (2 * PI). 16-bit: 0x3118, 32-bit: 0x3e22f983, 64-bit: 0x3fc45f30_6dc9c882. 248 SRC_VCCZ True if vector condition code register is zero. 251 SRC_EXECZ True if execute mask is zero. 252 SRC_SCC Scalar condition code. 253 SRC_SHARED_BASE N/A 235 SRC_SHARED_LIMIT N/A 236 SRC_PRIVATE_BASE N/A 237 SRC_PRIVATE_LIMIT N/A 238 SRC_POPS_EXITING_WAVE_ID POPS counter (12b). Returns the ID of the most recent wave to complete in-order. 239 v0 Vector GPR 0. Increment from here for additional GPRs. There are NUM_VGPR VGPRs in total. Alignment restriction: unless otherwise noted, 64-bit and wider VGPR values must be even-aligned (first VGPR address is even). 256 v1 <p>See above.</p> 257 v2 <p>See above.</p> 258 v3 <p>See above.</p> 259 v4 <p>See above.</p> 260 v5 <p>See above.</p> 261 v6 <p>See above.</p> 262 v7 <p>See above.</p> 263 v8 <p>See above.</p> 264 v9 <p>See above.</p> 265 v10 <p>See above.</p> 266 v11 <p>See above.</p> 267 v12 <p>See above.</p> 268 v13 <p>See above.</p> 269 v14 <p>See above.</p> 270 v15 <p>See above.</p> 271 v16 <p>See above.</p> 272 v17 <p>See above.</p> 273 v18 <p>See above.</p> 274 v19 <p>See above.</p> 275 v20 <p>See above.</p> 276 v21 <p>See above.</p> 277 v22 <p>See above.</p> 278 v23 <p>See above.</p> 279 v24 <p>See above.</p> 280 v25 <p>See above.</p> 281 v26 <p>See above.</p> 282 v27 <p>See above.</p> 283 v28 <p>See above.</p> 284 v29 <p>See above.</p> 285 v30 <p>See above.</p> 286 v31 <p>See above.</p> 287 v32 <p>See above.</p> 288 v33 <p>See above.</p> 289 v34 <p>See above.</p> 290 v35 <p>See above.</p> 291 v36 <p>See above.</p> 292 v37 <p>See above.</p> 293 v38 <p>See above.</p> 294 v39 <p>See above.</p> 295 v40 <p>See above.</p> 296 v41 <p>See above.</p> 297 v42 <p>See above.</p> 298 v43 <p>See above.</p> 299 v44 <p>See above.</p> 300 v45 <p>See above.</p> 301 v46 <p>See above.</p> 302 v47 <p>See above.</p> 303 v48 <p>See above.</p> 304 v49 <p>See above.</p> 305 v50 <p>See above.</p> 306 v51 <p>See above.</p> 307 v52 <p>See above.</p> 308 v53 <p>See above.</p> 309 v54 <p>See above.</p> 310 v55 <p>See above.</p> 311 v56 <p>See above.</p> 312 v57 <p>See above.</p> 313 v58 <p>See above.</p> 314 v59 <p>See above.</p> 315 v60 <p>See above.</p> 316 v61 <p>See above.</p> 317 v62 <p>See above.</p> 318 v63 <p>See above.</p> 319 v64 <p>See above.</p> 320 v65 <p>See above.</p> 321 v66 <p>See above.</p> 322 v67 <p>See above.</p> 323 v68 <p>See above.</p> 324 v69 <p>See above.</p> 325 v70 <p>See above.</p> 326 v71 <p>See above.</p> 327 v72 <p>See above.</p> 328 v73 <p>See above.</p> 329 v74 <p>See above.</p> 330 v75 <p>See above.</p> 331 v76 <p>See above.</p> 332 v77 <p>See above.</p> 333 v78 <p>See above.</p> 334 v79 <p>See above.</p> 335 v80 <p>See above.</p> 336 v81 <p>See above.</p> 337 v82 <p>See above.</p> 338 v83 <p>See above.</p> 339 v84 <p>See above.</p> 340 v85 <p>See above.</p> 341 v86 <p>See above.</p> 342 v87 <p>See above.</p> 343 v88 <p>See above.</p> 344 v89 <p>See above.</p> 345 v90 <p>See above.</p> 346 v91 <p>See above.</p> 347 v92 <p>See above.</p> 348 v93 <p>See above.</p> 349 v94 <p>See above.</p> 350 v95 <p>See above.</p> 351 v96 <p>See above.</p> 352 v97 <p>See above.</p> 353 v98 <p>See above.</p> 354 v99 <p>See above.</p> 355 v100 <p>See above.</p> 356 v101 <p>See above.</p> 357 v102 <p>See above.</p> 358 v103 <p>See above.</p> 359 v104 <p>See above.</p> 360 v105 <p>See above.</p> 361 v106 <p>See above.</p> 362 v107 <p>See above.</p> 363 v108 <p>See above.</p> 364 v109 <p>See above.</p> 365 v110 <p>See above.</p> 366 v111 <p>See above.</p> 367 v112 <p>See above.</p> 368 v113 <p>See above.</p> 369 v114 <p>See above.</p> 370 v115 <p>See above.</p> 371 v116 <p>See above.</p> 372 v117 <p>See above.</p> 373 v118 <p>See above.</p> 374 v119 <p>See above.</p> 375 v120 <p>See above.</p> 376 v121 <p>See above.</p> 377 v122 <p>See above.</p> 378 v123 <p>See above.</p> 379 v124 <p>See above.</p> 380 v125 <p>See above.</p> 381 v126 <p>See above.</p> 382 v127 <p>See above.</p> 383 v128 <p>See above.</p> 384 v129 <p>See above.</p> 385 v130 <p>See above.</p> 386 v131 <p>See above.</p> 387 v132 <p>See above.</p> 388 v133 <p>See above.</p> 389 v134 <p>See above.</p> 390 v135 <p>See above.</p> 391 v136 <p>See above.</p> 392 v137 <p>See above.</p> 393 v138 <p>See above.</p> 394 v139 <p>See above.</p> 395 v140 <p>See above.</p> 396 v141 <p>See above.</p> 397 v142 <p>See above.</p> 398 v143 <p>See above.</p> 399 v144 <p>See above.</p> 400 v145 <p>See above.</p> 401 v146 <p>See above.</p> 402 v147 <p>See above.</p> 403 v148 <p>See above.</p> 404 v149 <p>See above.</p> 405 v150 <p>See above.</p> 406 v151 <p>See above.</p> 407 v152 <p>See above.</p> 408 v153 <p>See above.</p> 409 v154 <p>See above.</p> 410 v155 <p>See above.</p> 411 v156 <p>See above.</p> 412 v157 <p>See above.</p> 413 v158 <p>See above.</p> 414 v159 <p>See above.</p> 415 v160 <p>See above.</p> 416 v161 <p>See above.</p> 417 v162 <p>See above.</p> 418 v163 <p>See above.</p> 419 v164 <p>See above.</p> 420 v165 <p>See above.</p> 421 v166 <p>See above.</p> 422 v167 <p>See above.</p> 423 v168 <p>See above.</p> 424 v169 <p>See above.</p> 425 v170 <p>See above.</p> 426 v171 <p>See above.</p> 427 v172 <p>See above.</p> 428 v173 <p>See above.</p> 429 v174 <p>See above.</p> 430 v175 <p>See above.</p> 431 v176 <p>See above.</p> 432 v177 <p>See above.</p> 433 v178 <p>See above.</p> 434 v179 <p>See above.</p> 435 v180 <p>See above.</p> 436 v181 <p>See above.</p> 437 v182 <p>See above.</p> 438 v183 <p>See above.</p> 439 v184 <p>See above.</p> 440 v185 <p>See above.</p> 441 v186 <p>See above.</p> 442 v187 <p>See above.</p> 443 v188 <p>See above.</p> 444 v189 <p>See above.</p> 445 v190 <p>See above.</p> 446 v191 <p>See above.</p> 447 v192 <p>See above.</p> 448 v193 <p>See above.</p> 449 v194 <p>See above.</p> 450 v195 <p>See above.</p> 451 v196 <p>See above.</p> 452 v197 <p>See above.</p> 453 v198 <p>See above.</p> 454 v199 <p>See above.</p> 455 v200 <p>See above.</p> 456 v201 <p>See above.</p> 457 v202 <p>See above.</p> 458 v203 <p>See above.</p> 459 v204 <p>See above.</p> 460 v205 <p>See above.</p> 461 v206 <p>See above.</p> 462 v207 <p>See above.</p> 463 v208 <p>See above.</p> 464 v209 <p>See above.</p> 465 v210 <p>See above.</p> 466 v211 <p>See above.</p> 467 v212 <p>See above.</p> 468 v213 <p>See above.</p> 469 v214 <p>See above.</p> 470 v215 <p>See above.</p> 471 v216 <p>See above.</p> 472 v217 <p>See above.</p> 473 v218 <p>See above.</p> 474 v219 <p>See above.</p> 475 v220 <p>See above.</p> 476 v221 <p>See above.</p> 477 v222 <p>See above.</p> 478 v223 <p>See above.</p> 479 v224 <p>See above.</p> 480 v225 <p>See above.</p> 481 v226 <p>See above.</p> 482 v227 <p>See above.</p> 483 v228 <p>See above.</p> 484 v229 <p>See above.</p> 485 v230 <p>See above.</p> 486 v231 <p>See above.</p> 487 v232 <p>See above.</p> 488 v233 <p>See above.</p> 489 v234 <p>See above.</p> 490 v235 <p>See above.</p> 491 v236 <p>See above.</p> 492 v237 <p>See above.</p> 493 v238 <p>See above.</p> 494 v239 <p>See above.</p> 495 v240 <p>See above.</p> 496 v241 <p>See above.</p> 497 v242 <p>See above.</p> 498 v243 <p>See above.</p> 499 v244 <p>See above.</p> 500 v245 <p>See above.</p> 501 v246 <p>See above.</p> 502 v247 <p>See above.</p> 503 v248 <p>See above.</p> 504 v249 <p>See above.</p> 505 v250 <p>See above.</p> 506 v251 <p>See above.</p> 507 v252 <p>See above.</p> 508 v253 <p>See above.</p> 509 v254 <p>See above.</p> 510 v255 <p>See above.</p> 511 SRC_LITERAL 32-bit literal constant follows this instruction. 16-bit operands use only the 16 LSBs of this constant. 64-bit unsigned integer operands zero-extend the constant. 64-bit signed integer operands sign-extend the constant. 64-bit floating-point operands use the constant as the high 32 MSBs of a double-precision floating point value (1 sign bit, 11 exponent bits and 20 mantissa bits). 255 OPR_SRC_ACCVGPR N/A acc0 Accumulator VGPR 0. Increment from here for additional GPRs. There are NUM_ACCVGPR ACCVGPRs in total. Alignment restriction: unless otherwise noted, 64-bit and wider ACCVGPR values must be even-aligned (first ACCVGPR address is even). 768 acc1 <p>See above.</p> 769 acc2 <p>See above.</p> 770 acc3 <p>See above.</p> 771 acc4 <p>See above.</p> 772 acc5 <p>See above.</p> 773 acc6 <p>See above.</p> 774 acc7 <p>See above.</p> 775 acc8 <p>See above.</p> 776 acc9 <p>See above.</p> 777 acc10 <p>See above.</p> 778 acc11 <p>See above.</p> 779 acc12 <p>See above.</p> 780 acc13 <p>See above.</p> 781 acc14 <p>See above.</p> 782 acc15 <p>See above.</p> 783 acc16 <p>See above.</p> 784 acc17 <p>See above.</p> 785 acc18 <p>See above.</p> 786 acc19 <p>See above.</p> 787 acc20 <p>See above.</p> 788 acc21 <p>See above.</p> 789 acc22 <p>See above.</p> 790 acc23 <p>See above.</p> 791 acc24 <p>See above.</p> 792 acc25 <p>See above.</p> 793 acc26 <p>See above.</p> 794 acc27 <p>See above.</p> 795 acc28 <p>See above.</p> 796 acc29 <p>See above.</p> 797 acc30 <p>See above.</p> 798 acc31 <p>See above.</p> 799 acc32 <p>See above.</p> 800 acc33 <p>See above.</p> 801 acc34 <p>See above.</p> 802 acc35 <p>See above.</p> 803 acc36 <p>See above.</p> 804 acc37 <p>See above.</p> 805 acc38 <p>See above.</p> 806 acc39 <p>See above.</p> 807 acc40 <p>See above.</p> 808 acc41 <p>See above.</p> 809 acc42 <p>See above.</p> 810 acc43 <p>See above.</p> 811 acc44 <p>See above.</p> 812 acc45 <p>See above.</p> 813 acc46 <p>See above.</p> 814 acc47 <p>See above.</p> 815 acc48 <p>See above.</p> 816 acc49 <p>See above.</p> 817 acc50 <p>See above.</p> 818 acc51 <p>See above.</p> 819 acc52 <p>See above.</p> 820 acc53 <p>See above.</p> 821 acc54 <p>See above.</p> 822 acc55 <p>See above.</p> 823 acc56 <p>See above.</p> 824 acc57 <p>See above.</p> 825 acc58 <p>See above.</p> 826 acc59 <p>See above.</p> 827 acc60 <p>See above.</p> 828 acc61 <p>See above.</p> 829 acc62 <p>See above.</p> 830 acc63 <p>See above.</p> 831 acc64 <p>See above.</p> 832 acc65 <p>See above.</p> 833 acc66 <p>See above.</p> 834 acc67 <p>See above.</p> 835 acc68 <p>See above.</p> 836 acc69 <p>See above.</p> 837 acc70 <p>See above.</p> 838 acc71 <p>See above.</p> 839 acc72 <p>See above.</p> 840 acc73 <p>See above.</p> 841 acc74 <p>See above.</p> 842 acc75 <p>See above.</p> 843 acc76 <p>See above.</p> 844 acc77 <p>See above.</p> 845 acc78 <p>See above.</p> 846 acc79 <p>See above.</p> 847 acc80 <p>See above.</p> 848 acc81 <p>See above.</p> 849 acc82 <p>See above.</p> 850 acc83 <p>See above.</p> 851 acc84 <p>See above.</p> 852 acc85 <p>See above.</p> 853 acc86 <p>See above.</p> 854 acc87 <p>See above.</p> 855 acc88 <p>See above.</p> 856 acc89 <p>See above.</p> 857 acc90 <p>See above.</p> 858 acc91 <p>See above.</p> 859 acc92 <p>See above.</p> 860 acc93 <p>See above.</p> 861 acc94 <p>See above.</p> 862 acc95 <p>See above.</p> 863 acc96 <p>See above.</p> 864 acc97 <p>See above.</p> 865 acc98 <p>See above.</p> 866 acc99 <p>See above.</p> 867 acc100 <p>See above.</p> 868 acc101 <p>See above.</p> 869 acc102 <p>See above.</p> 870 acc103 <p>See above.</p> 871 acc104 <p>See above.</p> 872 acc105 <p>See above.</p> 873 acc106 <p>See above.</p> 874 acc107 <p>See above.</p> 875 acc108 <p>See above.</p> 876 acc109 <p>See above.</p> 877 acc110 <p>See above.</p> 878 acc111 <p>See above.</p> 879 acc112 <p>See above.</p> 880 acc113 <p>See above.</p> 881 acc114 <p>See above.</p> 882 acc115 <p>See above.</p> 883 acc116 <p>See above.</p> 884 acc117 <p>See above.</p> 885 acc118 <p>See above.</p> 886 acc119 <p>See above.</p> 887 acc120 <p>See above.</p> 888 acc121 <p>See above.</p> 889 acc122 <p>See above.</p> 890 acc123 <p>See above.</p> 891 acc124 <p>See above.</p> 892 acc125 <p>See above.</p> 893 acc126 <p>See above.</p> 894 acc127 <p>See above.</p> 895 acc128 <p>See above.</p> 896 acc129 <p>See above.</p> 897 acc130 <p>See above.</p> 898 acc131 <p>See above.</p> 899 acc132 <p>See above.</p> 900 acc133 <p>See above.</p> 901 acc134 <p>See above.</p> 902 acc135 <p>See above.</p> 903 acc136 <p>See above.</p> 904 acc137 <p>See above.</p> 905 acc138 <p>See above.</p> 906 acc139 <p>See above.</p> 907 acc140 <p>See above.</p> 908 acc141 <p>See above.</p> 909 acc142 <p>See above.</p> 910 acc143 <p>See above.</p> 911 acc144 <p>See above.</p> 912 acc145 <p>See above.</p> 913 acc146 <p>See above.</p> 914 acc147 <p>See above.</p> 915 acc148 <p>See above.</p> 916 acc149 <p>See above.</p> 917 acc150 <p>See above.</p> 918 acc151 <p>See above.</p> 919 acc152 <p>See above.</p> 920 acc153 <p>See above.</p> 921 acc154 <p>See above.</p> 922 acc155 <p>See above.</p> 923 acc156 <p>See above.</p> 924 acc157 <p>See above.</p> 925 acc158 <p>See above.</p> 926 acc159 <p>See above.</p> 927 acc160 <p>See above.</p> 928 acc161 <p>See above.</p> 929 acc162 <p>See above.</p> 930 acc163 <p>See above.</p> 931 acc164 <p>See above.</p> 932 acc165 <p>See above.</p> 933 acc166 <p>See above.</p> 934 acc167 <p>See above.</p> 935 acc168 <p>See above.</p> 936 acc169 <p>See above.</p> 937 acc170 <p>See above.</p> 938 acc171 <p>See above.</p> 939 acc172 <p>See above.</p> 940 acc173 <p>See above.</p> 941 acc174 <p>See above.</p> 942 acc175 <p>See above.</p> 943 acc176 <p>See above.</p> 944 acc177 <p>See above.</p> 945 acc178 <p>See above.</p> 946 acc179 <p>See above.</p> 947 acc180 <p>See above.</p> 948 acc181 <p>See above.</p> 949 acc182 <p>See above.</p> 950 acc183 <p>See above.</p> 951 acc184 <p>See above.</p> 952 acc185 <p>See above.</p> 953 acc186 <p>See above.</p> 954 acc187 <p>See above.</p> 955 acc188 <p>See above.</p> 956 acc189 <p>See above.</p> 957 acc190 <p>See above.</p> 958 acc191 <p>See above.</p> 959 acc192 <p>See above.</p> 960 acc193 <p>See above.</p> 961 acc194 <p>See above.</p> 962 acc195 <p>See above.</p> 963 acc196 <p>See above.</p> 964 acc197 <p>See above.</p> 965 acc198 <p>See above.</p> 966 acc199 <p>See above.</p> 967 acc200 <p>See above.</p> 968 acc201 <p>See above.</p> 969 acc202 <p>See above.</p> 970 acc203 <p>See above.</p> 971 acc204 <p>See above.</p> 972 acc205 <p>See above.</p> 973 acc206 <p>See above.</p> 974 acc207 <p>See above.</p> 975 acc208 <p>See above.</p> 976 acc209 <p>See above.</p> 977 acc210 <p>See above.</p> 978 acc211 <p>See above.</p> 979 acc212 <p>See above.</p> 980 acc213 <p>See above.</p> 981 acc214 <p>See above.</p> 982 acc215 <p>See above.</p> 983 acc216 <p>See above.</p> 984 acc217 <p>See above.</p> 985 acc218 <p>See above.</p> 986 acc219 <p>See above.</p> 987 acc220 <p>See above.</p> 988 acc221 <p>See above.</p> 989 acc222 <p>See above.</p> 990 acc223 <p>See above.</p> 991 acc224 <p>See above.</p> 992 acc225 <p>See above.</p> 993 acc226 <p>See above.</p> 994 acc227 <p>See above.</p> 995 acc228 <p>See above.</p> 996 acc229 <p>See above.</p> 997 acc230 <p>See above.</p> 998 acc231 <p>See above.</p> 999 acc232 <p>See above.</p> 1000 acc233 <p>See above.</p> 1001 acc234 <p>See above.</p> 1002 acc235 <p>See above.</p> 1003 acc236 <p>See above.</p> 1004 acc237 <p>See above.</p> 1005 acc238 <p>See above.</p> 1006 acc239 <p>See above.</p> 1007 acc240 <p>See above.</p> 1008 acc241 <p>See above.</p> 1009 acc242 <p>See above.</p> 1010 acc243 <p>See above.</p> 1011 acc244 <p>See above.</p> 1012 acc245 <p>See above.</p> 1013 acc246 <p>See above.</p> 1014 acc247 <p>See above.</p> 1015 acc248 <p>See above.</p> 1016 acc249 <p>See above.</p> 1017 acc250 <p>See above.</p> 1018 acc251 <p>See above.</p> 1019 acc252 <p>See above.</p> 1020 acc253 <p>See above.</p> 1021 acc254 <p>See above.</p> 1022 acc255 <p>See above.</p> 1023 OPR_SRC_NOLDS N/A OPR_SDST OPR_SDST_EXEC OPR_SDST_M0 OPR_SRC_SIMPLE OPR_SRC_VGPR OPR_SREG OPR_SREG_NOVCC OPR_SSRC_NOLIT OPR_SSRC_SPECIAL_SCC s0 Scalar GPR 0. Increment from here for additional GPRs. There are NUM_SGPR SGPRs in total. Alignment restriction: unless otherwise noted, 64-bit SGPR values must be even-aligned (first SGPR address is even); 128-bit and wider SGPR values must be aligned to a multiple of 4. 0 s1 <p>See above.</p> 1 s2 <p>See above.</p> 2 s3 <p>See above.</p> 3 s4 <p>See above.</p> 4 s5 <p>See above.</p> 5 s6 <p>See above.</p> 6 s7 <p>See above.</p> 7 s8 <p>See above.</p> 8 s9 <p>See above.</p> 9 s10 <p>See above.</p> 10 s11 <p>See above.</p> 11 s12 <p>See above.</p> 12 s13 <p>See above.</p> 13 s14 <p>See above.</p> 14 s15 <p>See above.</p> 15 s16 <p>See above.</p> 16 s17 <p>See above.</p> 17 s18 <p>See above.</p> 18 s19 <p>See above.</p> 19 s20 <p>See above.</p> 20 s21 <p>See above.</p> 21 s22 <p>See above.</p> 22 s23 <p>See above.</p> 23 s24 <p>See above.</p> 24 s25 <p>See above.</p> 25 s26 <p>See above.</p> 26 s27 <p>See above.</p> 27 s28 <p>See above.</p> 28 s29 <p>See above.</p> 29 s30 <p>See above.</p> 30 s31 <p>See above.</p> 31 s32 <p>See above.</p> 32 s33 <p>See above.</p> 33 s34 <p>See above.</p> 34 s35 <p>See above.</p> 35 s36 <p>See above.</p> 36 s37 <p>See above.</p> 37 s38 <p>See above.</p> 38 s39 <p>See above.</p> 39 s40 <p>See above.</p> 40 s41 <p>See above.</p> 41 s42 <p>See above.</p> 42 s43 <p>See above.</p> 43 s44 <p>See above.</p> 44 s45 <p>See above.</p> 45 s46 <p>See above.</p> 46 s47 <p>See above.</p> 47 s48 <p>See above.</p> 48 s49 <p>See above.</p> 49 s50 <p>See above.</p> 50 s51 <p>See above.</p> 51 s52 <p>See above.</p> 52 s53 <p>See above.</p> 53 s54 <p>See above.</p> 54 s55 <p>See above.</p> 55 s56 <p>See above.</p> 56 s57 <p>See above.</p> 57 s58 <p>See above.</p> 58 s59 <p>See above.</p> 59 s60 <p>See above.</p> 60 s61 <p>See above.</p> 61 s62 <p>See above.</p> 62 s63 <p>See above.</p> 63 s64 <p>See above.</p> 64 s65 <p>See above.</p> 65 s66 <p>See above.</p> 66 s67 <p>See above.</p> 67 s68 <p>See above.</p> 68 s69 <p>See above.</p> 69 s70 <p>See above.</p> 70 s71 <p>See above.</p> 71 s72 <p>See above.</p> 72 s73 <p>See above.</p> 73 s74 <p>See above.</p> 74 s75 <p>See above.</p> 75 s76 <p>See above.</p> 76 s77 <p>See above.</p> 77 s78 <p>See above.</p> 78 s79 <p>See above.</p> 79 s80 <p>See above.</p> 80 s81 <p>See above.</p> 81 s82 <p>See above.</p> 82 s83 <p>See above.</p> 83 s84 <p>See above.</p> 84 s85 <p>See above.</p> 85 s86 <p>See above.</p> 86 s87 <p>See above.</p> 87 s88 <p>See above.</p> 88 s89 <p>See above.</p> 89 s90 <p>See above.</p> 90 s91 <p>See above.</p> 91 s92 <p>See above.</p> 92 s93 <p>See above.</p> 93 s94 <p>See above.</p> 94 s95 <p>See above.</p> 95 s96 <p>See above.</p> 96 s97 <p>See above.</p> 97 s98 <p>See above.</p> 98 s99 <p>See above.</p> 99 s100 <p>See above.</p> 100 s101 <p>See above.</p> 101 XNACK_MASK_LO xnack_mask[31:0], see XNACK replay mechanism for details. 104 XNACK_MASK_HI xnack_mask[63:32], see XNACK replay mechanism for details. 105 FLAT_SCRATCH_LO {13'd0, size[18:0]} 102 FLAT_SCRATCH_HI {8'd0, offset[31:8]} 103 TTMP0 Trap handler temps (privileged). Increment from here for additional TTMPs. There are NUM_TTMP TTMPs in total. {TTMP1,TTMP0} = PC_save{hi,lo}. Alignment restriction: unless otherwise noted, 64-bit TTMP values must be even-aligned (first TTMP address is even); 128-bit and wider TTMP values must be aligned to a multiple of 4. 108 TTMP1 Trap handler temps (privileged). 109 TTMP2 Trap handler temps (privileged). 110 TTMP3 Trap handler temps (privileged). 111 TTMP4 Trap handler temps (privileged). 112 TTMP5 Trap handler temps (privileged). 113 TTMP6 Trap handler temps (privileged). 114 TTMP7 Trap handler temps (privileged). 115 TTMP8 Trap handler temps (privileged). 116 TTMP9 Trap handler temps (privileged). 117 TTMP10 Trap handler temps (privileged). 118 TTMP11 Trap handler temps (privileged). 119 TTMP12 Trap handler temps (privileged). 120 TTMP13 Trap handler temps (privileged). 121 TTMP14 Trap handler temps (privileged). 122 TTMP15 Trap handler temps (privileged). 123 VCC_LO VCC[31:0] 106 VCC_HI VCC[63:32] 107 M0 Special register used to hold LDS/GDS addresses, relative indices, and send-message values. 124 EXEC_LO exec[31:0] 126 EXEC_HI exec[63:32] 127 0 0 128 1 1 (integer) 129 2 2 (integer) 130 3 3 (integer) 131 4 4 (integer) 132 5 5 (integer) 133 6 6 (integer) 134 7 7 (integer) 135 8 8 (integer) 136 9 9 (integer) 137 10 10 (integer) 138 11 11 (integer) 139 12 12 (integer) 140 13 13 (integer) 141 14 14 (integer) 142 15 15 (integer) 143 16 16 (integer) 144 17 17 (integer) 145 18 18 (integer) 146 19 19 (integer) 147 20 20 (integer) 148 21 21 (integer) 149 22 22 (integer) 150 23 23 (integer) 151 24 24 (integer) 152 25 25 (integer) 153 26 26 (integer) 154 27 27 (integer) 155 28 28 (integer) 156 29 29 (integer) 157 30 30 (integer) 158 31 31 (integer) 159 32 32 (integer) 160 33 33 (integer) 161 34 34 (integer) 162 35 35 (integer) 163 36 36 (integer) 164 37 37 (integer) 165 38 38 (integer) 166 39 39 (integer) 167 40 40 (integer) 168 41 41 (integer) 169 42 42 (integer) 170 43 43 (integer) 171 44 44 (integer) 172 45 45 (integer) 173 46 46 (integer) 174 47 47 (integer) 175 48 48 (integer) 176 49 49 (integer) 177 50 50 (integer) 178 51 51 (integer) 179 52 52 (integer) 180 53 53 (integer) 181 54 54 (integer) 182 55 55 (integer) 183 56 56 (integer) 184 57 57 (integer) 185 58 58 (integer) 186 59 59 (integer) 187 60 60 (integer) 188 61 61 (integer) 189 62 62 (integer) 190 63 63 (integer) 191 64 64 (integer) 192 -1 -1 (integer) 193 -2 -2 (integer) 194 -3 -3 (integer) 195 -4 -4 (integer) 196 -5 -5 (integer) 197 -6 -6 (integer) 198 -7 -7 (integer) 199 -8 -8 (integer) 200 -9 -9 (integer) 201 -10 -10 (integer) 202 -11 -11 (integer) 203 -12 -12 (integer) 204 -13 -13 (integer) 205 -14 -14 (integer) 206 -15 -15 (integer) 207 -16 -16 (integer) 208 0.5 0.5 240 -0.5 -0.5 241 1.0 1.0 242 -1.0 -1.0 243 2.0 2.0 244 -2.0 -2.0 245 4.0 4.0 246 -4.0 -4.0 247 0.15915494 1 / (2 * PI). 16-bit: 0x3118, 32-bit: 0x3e22f983, 64-bit: 0x3fc45f30_6dc9c882. 248 SRC_VCCZ True if vector condition code register is zero. 251 SRC_EXECZ True if execute mask is zero. 252 SRC_SCC Scalar condition code. 253 SRC_SHARED_BASE N/A 235 SRC_SHARED_LIMIT N/A 236 SRC_PRIVATE_BASE N/A 237 SRC_PRIVATE_LIMIT N/A 238 SRC_POPS_EXITING_WAVE_ID POPS counter (12b). Returns the ID of the most recent wave to complete in-order. 239 v0 Vector GPR 0. Increment from here for additional GPRs. There are NUM_VGPR VGPRs in total. Alignment restriction: unless otherwise noted, 64-bit and wider VGPR values must be even-aligned (first VGPR address is even). 256 v1 <p>See above.</p> 257 v2 <p>See above.</p> 258 v3 <p>See above.</p> 259 v4 <p>See above.</p> 260 v5 <p>See above.</p> 261 v6 <p>See above.</p> 262 v7 <p>See above.</p> 263 v8 <p>See above.</p> 264 v9 <p>See above.</p> 265 v10 <p>See above.</p> 266 v11 <p>See above.</p> 267 v12 <p>See above.</p> 268 v13 <p>See above.</p> 269 v14 <p>See above.</p> 270 v15 <p>See above.</p> 271 v16 <p>See above.</p> 272 v17 <p>See above.</p> 273 v18 <p>See above.</p> 274 v19 <p>See above.</p> 275 v20 <p>See above.</p> 276 v21 <p>See above.</p> 277 v22 <p>See above.</p> 278 v23 <p>See above.</p> 279 v24 <p>See above.</p> 280 v25 <p>See above.</p> 281 v26 <p>See above.</p> 282 v27 <p>See above.</p> 283 v28 <p>See above.</p> 284 v29 <p>See above.</p> 285 v30 <p>See above.</p> 286 v31 <p>See above.</p> 287 v32 <p>See above.</p> 288 v33 <p>See above.</p> 289 v34 <p>See above.</p> 290 v35 <p>See above.</p> 291 v36 <p>See above.</p> 292 v37 <p>See above.</p> 293 v38 <p>See above.</p> 294 v39 <p>See above.</p> 295 v40 <p>See above.</p> 296 v41 <p>See above.</p> 297 v42 <p>See above.</p> 298 v43 <p>See above.</p> 299 v44 <p>See above.</p> 300 v45 <p>See above.</p> 301 v46 <p>See above.</p> 302 v47 <p>See above.</p> 303 v48 <p>See above.</p> 304 v49 <p>See above.</p> 305 v50 <p>See above.</p> 306 v51 <p>See above.</p> 307 v52 <p>See above.</p> 308 v53 <p>See above.</p> 309 v54 <p>See above.</p> 310 v55 <p>See above.</p> 311 v56 <p>See above.</p> 312 v57 <p>See above.</p> 313 v58 <p>See above.</p> 314 v59 <p>See above.</p> 315 v60 <p>See above.</p> 316 v61 <p>See above.</p> 317 v62 <p>See above.</p> 318 v63 <p>See above.</p> 319 v64 <p>See above.</p> 320 v65 <p>See above.</p> 321 v66 <p>See above.</p> 322 v67 <p>See above.</p> 323 v68 <p>See above.</p> 324 v69 <p>See above.</p> 325 v70 <p>See above.</p> 326 v71 <p>See above.</p> 327 v72 <p>See above.</p> 328 v73 <p>See above.</p> 329 v74 <p>See above.</p> 330 v75 <p>See above.</p> 331 v76 <p>See above.</p> 332 v77 <p>See above.</p> 333 v78 <p>See above.</p> 334 v79 <p>See above.</p> 335 v80 <p>See above.</p> 336 v81 <p>See above.</p> 337 v82 <p>See above.</p> 338 v83 <p>See above.</p> 339 v84 <p>See above.</p> 340 v85 <p>See above.</p> 341 v86 <p>See above.</p> 342 v87 <p>See above.</p> 343 v88 <p>See above.</p> 344 v89 <p>See above.</p> 345 v90 <p>See above.</p> 346 v91 <p>See above.</p> 347 v92 <p>See above.</p> 348 v93 <p>See above.</p> 349 v94 <p>See above.</p> 350 v95 <p>See above.</p> 351 v96 <p>See above.</p> 352 v97 <p>See above.</p> 353 v98 <p>See above.</p> 354 v99 <p>See above.</p> 355 v100 <p>See above.</p> 356 v101 <p>See above.</p> 357 v102 <p>See above.</p> 358 v103 <p>See above.</p> 359 v104 <p>See above.</p> 360 v105 <p>See above.</p> 361 v106 <p>See above.</p> 362 v107 <p>See above.</p> 363 v108 <p>See above.</p> 364 v109 <p>See above.</p> 365 v110 <p>See above.</p> 366 v111 <p>See above.</p> 367 v112 <p>See above.</p> 368 v113 <p>See above.</p> 369 v114 <p>See above.</p> 370 v115 <p>See above.</p> 371 v116 <p>See above.</p> 372 v117 <p>See above.</p> 373 v118 <p>See above.</p> 374 v119 <p>See above.</p> 375 v120 <p>See above.</p> 376 v121 <p>See above.</p> 377 v122 <p>See above.</p> 378 v123 <p>See above.</p> 379 v124 <p>See above.</p> 380 v125 <p>See above.</p> 381 v126 <p>See above.</p> 382 v127 <p>See above.</p> 383 v128 <p>See above.</p> 384 v129 <p>See above.</p> 385 v130 <p>See above.</p> 386 v131 <p>See above.</p> 387 v132 <p>See above.</p> 388 v133 <p>See above.</p> 389 v134 <p>See above.</p> 390 v135 <p>See above.</p> 391 v136 <p>See above.</p> 392 v137 <p>See above.</p> 393 v138 <p>See above.</p> 394 v139 <p>See above.</p> 395 v140 <p>See above.</p> 396 v141 <p>See above.</p> 397 v142 <p>See above.</p> 398 v143 <p>See above.</p> 399 v144 <p>See above.</p> 400 v145 <p>See above.</p> 401 v146 <p>See above.</p> 402 v147 <p>See above.</p> 403 v148 <p>See above.</p> 404 v149 <p>See above.</p> 405 v150 <p>See above.</p> 406 v151 <p>See above.</p> 407 v152 <p>See above.</p> 408 v153 <p>See above.</p> 409 v154 <p>See above.</p> 410 v155 <p>See above.</p> 411 v156 <p>See above.</p> 412 v157 <p>See above.</p> 413 v158 <p>See above.</p> 414 v159 <p>See above.</p> 415 v160 <p>See above.</p> 416 v161 <p>See above.</p> 417 v162 <p>See above.</p> 418 v163 <p>See above.</p> 419 v164 <p>See above.</p> 420 v165 <p>See above.</p> 421 v166 <p>See above.</p> 422 v167 <p>See above.</p> 423 v168 <p>See above.</p> 424 v169 <p>See above.</p> 425 v170 <p>See above.</p> 426 v171 <p>See above.</p> 427 v172 <p>See above.</p> 428 v173 <p>See above.</p> 429 v174 <p>See above.</p> 430 v175 <p>See above.</p> 431 v176 <p>See above.</p> 432 v177 <p>See above.</p> 433 v178 <p>See above.</p> 434 v179 <p>See above.</p> 435 v180 <p>See above.</p> 436 v181 <p>See above.</p> 437 v182 <p>See above.</p> 438 v183 <p>See above.</p> 439 v184 <p>See above.</p> 440 v185 <p>See above.</p> 441 v186 <p>See above.</p> 442 v187 <p>See above.</p> 443 v188 <p>See above.</p> 444 v189 <p>See above.</p> 445 v190 <p>See above.</p> 446 v191 <p>See above.</p> 447 v192 <p>See above.</p> 448 v193 <p>See above.</p> 449 v194 <p>See above.</p> 450 v195 <p>See above.</p> 451 v196 <p>See above.</p> 452 v197 <p>See above.</p> 453 v198 <p>See above.</p> 454 v199 <p>See above.</p> 455 v200 <p>See above.</p> 456 v201 <p>See above.</p> 457 v202 <p>See above.</p> 458 v203 <p>See above.</p> 459 v204 <p>See above.</p> 460 v205 <p>See above.</p> 461 v206 <p>See above.</p> 462 v207 <p>See above.</p> 463 v208 <p>See above.</p> 464 v209 <p>See above.</p> 465 v210 <p>See above.</p> 466 v211 <p>See above.</p> 467 v212 <p>See above.</p> 468 v213 <p>See above.</p> 469 v214 <p>See above.</p> 470 v215 <p>See above.</p> 471 v216 <p>See above.</p> 472 v217 <p>See above.</p> 473 v218 <p>See above.</p> 474 v219 <p>See above.</p> 475 v220 <p>See above.</p> 476 v221 <p>See above.</p> 477 v222 <p>See above.</p> 478 v223 <p>See above.</p> 479 v224 <p>See above.</p> 480 v225 <p>See above.</p> 481 v226 <p>See above.</p> 482 v227 <p>See above.</p> 483 v228 <p>See above.</p> 484 v229 <p>See above.</p> 485 v230 <p>See above.</p> 486 v231 <p>See above.</p> 487 v232 <p>See above.</p> 488 v233 <p>See above.</p> 489 v234 <p>See above.</p> 490 v235 <p>See above.</p> 491 v236 <p>See above.</p> 492 v237 <p>See above.</p> 493 v238 <p>See above.</p> 494 v239 <p>See above.</p> 495 v240 <p>See above.</p> 496 v241 <p>See above.</p> 497 v242 <p>See above.</p> 498 v243 <p>See above.</p> 499 v244 <p>See above.</p> 500 v245 <p>See above.</p> 501 v246 <p>See above.</p> 502 v247 <p>See above.</p> 503 v248 <p>See above.</p> 504 v249 <p>See above.</p> 505 v250 <p>See above.</p> 506 v251 <p>See above.</p> 507 v252 <p>See above.</p> 508 v253 <p>See above.</p> 509 v254 <p>See above.</p> 510 v255 <p>See above.</p> 511 SRC_LITERAL 32-bit literal constant follows this instruction. 16-bit operands use only the 16 LSBs of this constant. 64-bit unsigned integer operands zero-extend the constant. 64-bit signed integer operands sign-extend the constant. 64-bit floating-point operands use the constant as the high 32 MSBs of a double-precision floating point value (1 sign bit, 11 exponent bits and 20 mantissa bits). 255 OPR_SRC_NOLIT N/A OPR_SDST OPR_SDST_EXEC OPR_SDST_M0 OPR_SRC_SIMPLE OPR_SRC_VGPR OPR_SREG OPR_SREG_NOVCC OPR_SSRC_NOLIT OPR_SSRC_SPECIAL_SCC s0 Scalar GPR 0. Increment from here for additional GPRs. There are NUM_SGPR SGPRs in total. Alignment restriction: unless otherwise noted, 64-bit SGPR values must be even-aligned (first SGPR address is even); 128-bit and wider SGPR values must be aligned to a multiple of 4. 0 s1 <p>See above.</p> 1 s2 <p>See above.</p> 2 s3 <p>See above.</p> 3 s4 <p>See above.</p> 4 s5 <p>See above.</p> 5 s6 <p>See above.</p> 6 s7 <p>See above.</p> 7 s8 <p>See above.</p> 8 s9 <p>See above.</p> 9 s10 <p>See above.</p> 10 s11 <p>See above.</p> 11 s12 <p>See above.</p> 12 s13 <p>See above.</p> 13 s14 <p>See above.</p> 14 s15 <p>See above.</p> 15 s16 <p>See above.</p> 16 s17 <p>See above.</p> 17 s18 <p>See above.</p> 18 s19 <p>See above.</p> 19 s20 <p>See above.</p> 20 s21 <p>See above.</p> 21 s22 <p>See above.</p> 22 s23 <p>See above.</p> 23 s24 <p>See above.</p> 24 s25 <p>See above.</p> 25 s26 <p>See above.</p> 26 s27 <p>See above.</p> 27 s28 <p>See above.</p> 28 s29 <p>See above.</p> 29 s30 <p>See above.</p> 30 s31 <p>See above.</p> 31 s32 <p>See above.</p> 32 s33 <p>See above.</p> 33 s34 <p>See above.</p> 34 s35 <p>See above.</p> 35 s36 <p>See above.</p> 36 s37 <p>See above.</p> 37 s38 <p>See above.</p> 38 s39 <p>See above.</p> 39 s40 <p>See above.</p> 40 s41 <p>See above.</p> 41 s42 <p>See above.</p> 42 s43 <p>See above.</p> 43 s44 <p>See above.</p> 44 s45 <p>See above.</p> 45 s46 <p>See above.</p> 46 s47 <p>See above.</p> 47 s48 <p>See above.</p> 48 s49 <p>See above.</p> 49 s50 <p>See above.</p> 50 s51 <p>See above.</p> 51 s52 <p>See above.</p> 52 s53 <p>See above.</p> 53 s54 <p>See above.</p> 54 s55 <p>See above.</p> 55 s56 <p>See above.</p> 56 s57 <p>See above.</p> 57 s58 <p>See above.</p> 58 s59 <p>See above.</p> 59 s60 <p>See above.</p> 60 s61 <p>See above.</p> 61 s62 <p>See above.</p> 62 s63 <p>See above.</p> 63 s64 <p>See above.</p> 64 s65 <p>See above.</p> 65 s66 <p>See above.</p> 66 s67 <p>See above.</p> 67 s68 <p>See above.</p> 68 s69 <p>See above.</p> 69 s70 <p>See above.</p> 70 s71 <p>See above.</p> 71 s72 <p>See above.</p> 72 s73 <p>See above.</p> 73 s74 <p>See above.</p> 74 s75 <p>See above.</p> 75 s76 <p>See above.</p> 76 s77 <p>See above.</p> 77 s78 <p>See above.</p> 78 s79 <p>See above.</p> 79 s80 <p>See above.</p> 80 s81 <p>See above.</p> 81 s82 <p>See above.</p> 82 s83 <p>See above.</p> 83 s84 <p>See above.</p> 84 s85 <p>See above.</p> 85 s86 <p>See above.</p> 86 s87 <p>See above.</p> 87 s88 <p>See above.</p> 88 s89 <p>See above.</p> 89 s90 <p>See above.</p> 90 s91 <p>See above.</p> 91 s92 <p>See above.</p> 92 s93 <p>See above.</p> 93 s94 <p>See above.</p> 94 s95 <p>See above.</p> 95 s96 <p>See above.</p> 96 s97 <p>See above.</p> 97 s98 <p>See above.</p> 98 s99 <p>See above.</p> 99 s100 <p>See above.</p> 100 s101 <p>See above.</p> 101 XNACK_MASK_LO xnack_mask[31:0], see XNACK replay mechanism for details. 104 XNACK_MASK_HI xnack_mask[63:32], see XNACK replay mechanism for details. 105 FLAT_SCRATCH_LO {13'd0, size[18:0]} 102 FLAT_SCRATCH_HI {8'd0, offset[31:8]} 103 TTMP0 Trap handler temps (privileged). Increment from here for additional TTMPs. There are NUM_TTMP TTMPs in total. {TTMP1,TTMP0} = PC_save{hi,lo}. Alignment restriction: unless otherwise noted, 64-bit TTMP values must be even-aligned (first TTMP address is even); 128-bit and wider TTMP values must be aligned to a multiple of 4. 108 TTMP1 Trap handler temps (privileged). 109 TTMP2 Trap handler temps (privileged). 110 TTMP3 Trap handler temps (privileged). 111 TTMP4 Trap handler temps (privileged). 112 TTMP5 Trap handler temps (privileged). 113 TTMP6 Trap handler temps (privileged). 114 TTMP7 Trap handler temps (privileged). 115 TTMP8 Trap handler temps (privileged). 116 TTMP9 Trap handler temps (privileged). 117 TTMP10 Trap handler temps (privileged). 118 TTMP11 Trap handler temps (privileged). 119 TTMP12 Trap handler temps (privileged). 120 TTMP13 Trap handler temps (privileged). 121 TTMP14 Trap handler temps (privileged). 122 TTMP15 Trap handler temps (privileged). 123 VCC_LO VCC[31:0] 106 VCC_HI VCC[63:32] 107 M0 Special register used to hold LDS/GDS addresses, relative indices, and send-message values. 124 EXEC_LO exec[31:0] 126 EXEC_HI exec[63:32] 127 0 0 128 1 1 (integer) 129 2 2 (integer) 130 3 3 (integer) 131 4 4 (integer) 132 5 5 (integer) 133 6 6 (integer) 134 7 7 (integer) 135 8 8 (integer) 136 9 9 (integer) 137 10 10 (integer) 138 11 11 (integer) 139 12 12 (integer) 140 13 13 (integer) 141 14 14 (integer) 142 15 15 (integer) 143 16 16 (integer) 144 17 17 (integer) 145 18 18 (integer) 146 19 19 (integer) 147 20 20 (integer) 148 21 21 (integer) 149 22 22 (integer) 150 23 23 (integer) 151 24 24 (integer) 152 25 25 (integer) 153 26 26 (integer) 154 27 27 (integer) 155 28 28 (integer) 156 29 29 (integer) 157 30 30 (integer) 158 31 31 (integer) 159 32 32 (integer) 160 33 33 (integer) 161 34 34 (integer) 162 35 35 (integer) 163 36 36 (integer) 164 37 37 (integer) 165 38 38 (integer) 166 39 39 (integer) 167 40 40 (integer) 168 41 41 (integer) 169 42 42 (integer) 170 43 43 (integer) 171 44 44 (integer) 172 45 45 (integer) 173 46 46 (integer) 174 47 47 (integer) 175 48 48 (integer) 176 49 49 (integer) 177 50 50 (integer) 178 51 51 (integer) 179 52 52 (integer) 180 53 53 (integer) 181 54 54 (integer) 182 55 55 (integer) 183 56 56 (integer) 184 57 57 (integer) 185 58 58 (integer) 186 59 59 (integer) 187 60 60 (integer) 188 61 61 (integer) 189 62 62 (integer) 190 63 63 (integer) 191 64 64 (integer) 192 -1 -1 (integer) 193 -2 -2 (integer) 194 -3 -3 (integer) 195 -4 -4 (integer) 196 -5 -5 (integer) 197 -6 -6 (integer) 198 -7 -7 (integer) 199 -8 -8 (integer) 200 -9 -9 (integer) 201 -10 -10 (integer) 202 -11 -11 (integer) 203 -12 -12 (integer) 204 -13 -13 (integer) 205 -14 -14 (integer) 206 -15 -15 (integer) 207 -16 -16 (integer) 208 0.5 0.5 240 -0.5 -0.5 241 1.0 1.0 242 -1.0 -1.0 243 2.0 2.0 244 -2.0 -2.0 245 4.0 4.0 246 -4.0 -4.0 247 0.15915494 1 / (2 * PI). 16-bit: 0x3118, 32-bit: 0x3e22f983, 64-bit: 0x3fc45f30_6dc9c882. 248 SRC_VCCZ True if vector condition code register is zero. 251 SRC_EXECZ True if execute mask is zero. 252 SRC_SCC Scalar condition code. 253 SRC_SHARED_BASE N/A 235 SRC_SHARED_LIMIT N/A 236 SRC_PRIVATE_BASE N/A 237 SRC_PRIVATE_LIMIT N/A 238 SRC_POPS_EXITING_WAVE_ID POPS counter (12b). Returns the ID of the most recent wave to complete in-order. 239 v0 Vector GPR 0. Increment from here for additional GPRs. There are NUM_VGPR VGPRs in total. Alignment restriction: unless otherwise noted, 64-bit and wider VGPR values must be even-aligned (first VGPR address is even). 256 v1 <p>See above.</p> 257 v2 <p>See above.</p> 258 v3 <p>See above.</p> 259 v4 <p>See above.</p> 260 v5 <p>See above.</p> 261 v6 <p>See above.</p> 262 v7 <p>See above.</p> 263 v8 <p>See above.</p> 264 v9 <p>See above.</p> 265 v10 <p>See above.</p> 266 v11 <p>See above.</p> 267 v12 <p>See above.</p> 268 v13 <p>See above.</p> 269 v14 <p>See above.</p> 270 v15 <p>See above.</p> 271 v16 <p>See above.</p> 272 v17 <p>See above.</p> 273 v18 <p>See above.</p> 274 v19 <p>See above.</p> 275 v20 <p>See above.</p> 276 v21 <p>See above.</p> 277 v22 <p>See above.</p> 278 v23 <p>See above.</p> 279 v24 <p>See above.</p> 280 v25 <p>See above.</p> 281 v26 <p>See above.</p> 282 v27 <p>See above.</p> 283 v28 <p>See above.</p> 284 v29 <p>See above.</p> 285 v30 <p>See above.</p> 286 v31 <p>See above.</p> 287 v32 <p>See above.</p> 288 v33 <p>See above.</p> 289 v34 <p>See above.</p> 290 v35 <p>See above.</p> 291 v36 <p>See above.</p> 292 v37 <p>See above.</p> 293 v38 <p>See above.</p> 294 v39 <p>See above.</p> 295 v40 <p>See above.</p> 296 v41 <p>See above.</p> 297 v42 <p>See above.</p> 298 v43 <p>See above.</p> 299 v44 <p>See above.</p> 300 v45 <p>See above.</p> 301 v46 <p>See above.</p> 302 v47 <p>See above.</p> 303 v48 <p>See above.</p> 304 v49 <p>See above.</p> 305 v50 <p>See above.</p> 306 v51 <p>See above.</p> 307 v52 <p>See above.</p> 308 v53 <p>See above.</p> 309 v54 <p>See above.</p> 310 v55 <p>See above.</p> 311 v56 <p>See above.</p> 312 v57 <p>See above.</p> 313 v58 <p>See above.</p> 314 v59 <p>See above.</p> 315 v60 <p>See above.</p> 316 v61 <p>See above.</p> 317 v62 <p>See above.</p> 318 v63 <p>See above.</p> 319 v64 <p>See above.</p> 320 v65 <p>See above.</p> 321 v66 <p>See above.</p> 322 v67 <p>See above.</p> 323 v68 <p>See above.</p> 324 v69 <p>See above.</p> 325 v70 <p>See above.</p> 326 v71 <p>See above.</p> 327 v72 <p>See above.</p> 328 v73 <p>See above.</p> 329 v74 <p>See above.</p> 330 v75 <p>See above.</p> 331 v76 <p>See above.</p> 332 v77 <p>See above.</p> 333 v78 <p>See above.</p> 334 v79 <p>See above.</p> 335 v80 <p>See above.</p> 336 v81 <p>See above.</p> 337 v82 <p>See above.</p> 338 v83 <p>See above.</p> 339 v84 <p>See above.</p> 340 v85 <p>See above.</p> 341 v86 <p>See above.</p> 342 v87 <p>See above.</p> 343 v88 <p>See above.</p> 344 v89 <p>See above.</p> 345 v90 <p>See above.</p> 346 v91 <p>See above.</p> 347 v92 <p>See above.</p> 348 v93 <p>See above.</p> 349 v94 <p>See above.</p> 350 v95 <p>See above.</p> 351 v96 <p>See above.</p> 352 v97 <p>See above.</p> 353 v98 <p>See above.</p> 354 v99 <p>See above.</p> 355 v100 <p>See above.</p> 356 v101 <p>See above.</p> 357 v102 <p>See above.</p> 358 v103 <p>See above.</p> 359 v104 <p>See above.</p> 360 v105 <p>See above.</p> 361 v106 <p>See above.</p> 362 v107 <p>See above.</p> 363 v108 <p>See above.</p> 364 v109 <p>See above.</p> 365 v110 <p>See above.</p> 366 v111 <p>See above.</p> 367 v112 <p>See above.</p> 368 v113 <p>See above.</p> 369 v114 <p>See above.</p> 370 v115 <p>See above.</p> 371 v116 <p>See above.</p> 372 v117 <p>See above.</p> 373 v118 <p>See above.</p> 374 v119 <p>See above.</p> 375 v120 <p>See above.</p> 376 v121 <p>See above.</p> 377 v122 <p>See above.</p> 378 v123 <p>See above.</p> 379 v124 <p>See above.</p> 380 v125 <p>See above.</p> 381 v126 <p>See above.</p> 382 v127 <p>See above.</p> 383 v128 <p>See above.</p> 384 v129 <p>See above.</p> 385 v130 <p>See above.</p> 386 v131 <p>See above.</p> 387 v132 <p>See above.</p> 388 v133 <p>See above.</p> 389 v134 <p>See above.</p> 390 v135 <p>See above.</p> 391 v136 <p>See above.</p> 392 v137 <p>See above.</p> 393 v138 <p>See above.</p> 394 v139 <p>See above.</p> 395 v140 <p>See above.</p> 396 v141 <p>See above.</p> 397 v142 <p>See above.</p> 398 v143 <p>See above.</p> 399 v144 <p>See above.</p> 400 v145 <p>See above.</p> 401 v146 <p>See above.</p> 402 v147 <p>See above.</p> 403 v148 <p>See above.</p> 404 v149 <p>See above.</p> 405 v150 <p>See above.</p> 406 v151 <p>See above.</p> 407 v152 <p>See above.</p> 408 v153 <p>See above.</p> 409 v154 <p>See above.</p> 410 v155 <p>See above.</p> 411 v156 <p>See above.</p> 412 v157 <p>See above.</p> 413 v158 <p>See above.</p> 414 v159 <p>See above.</p> 415 v160 <p>See above.</p> 416 v161 <p>See above.</p> 417 v162 <p>See above.</p> 418 v163 <p>See above.</p> 419 v164 <p>See above.</p> 420 v165 <p>See above.</p> 421 v166 <p>See above.</p> 422 v167 <p>See above.</p> 423 v168 <p>See above.</p> 424 v169 <p>See above.</p> 425 v170 <p>See above.</p> 426 v171 <p>See above.</p> 427 v172 <p>See above.</p> 428 v173 <p>See above.</p> 429 v174 <p>See above.</p> 430 v175 <p>See above.</p> 431 v176 <p>See above.</p> 432 v177 <p>See above.</p> 433 v178 <p>See above.</p> 434 v179 <p>See above.</p> 435 v180 <p>See above.</p> 436 v181 <p>See above.</p> 437 v182 <p>See above.</p> 438 v183 <p>See above.</p> 439 v184 <p>See above.</p> 440 v185 <p>See above.</p> 441 v186 <p>See above.</p> 442 v187 <p>See above.</p> 443 v188 <p>See above.</p> 444 v189 <p>See above.</p> 445 v190 <p>See above.</p> 446 v191 <p>See above.</p> 447 v192 <p>See above.</p> 448 v193 <p>See above.</p> 449 v194 <p>See above.</p> 450 v195 <p>See above.</p> 451 v196 <p>See above.</p> 452 v197 <p>See above.</p> 453 v198 <p>See above.</p> 454 v199 <p>See above.</p> 455 v200 <p>See above.</p> 456 v201 <p>See above.</p> 457 v202 <p>See above.</p> 458 v203 <p>See above.</p> 459 v204 <p>See above.</p> 460 v205 <p>See above.</p> 461 v206 <p>See above.</p> 462 v207 <p>See above.</p> 463 v208 <p>See above.</p> 464 v209 <p>See above.</p> 465 v210 <p>See above.</p> 466 v211 <p>See above.</p> 467 v212 <p>See above.</p> 468 v213 <p>See above.</p> 469 v214 <p>See above.</p> 470 v215 <p>See above.</p> 471 v216 <p>See above.</p> 472 v217 <p>See above.</p> 473 v218 <p>See above.</p> 474 v219 <p>See above.</p> 475 v220 <p>See above.</p> 476 v221 <p>See above.</p> 477 v222 <p>See above.</p> 478 v223 <p>See above.</p> 479 v224 <p>See above.</p> 480 v225 <p>See above.</p> 481 v226 <p>See above.</p> 482 v227 <p>See above.</p> 483 v228 <p>See above.</p> 484 v229 <p>See above.</p> 485 v230 <p>See above.</p> 486 v231 <p>See above.</p> 487 v232 <p>See above.</p> 488 v233 <p>See above.</p> 489 v234 <p>See above.</p> 490 v235 <p>See above.</p> 491 v236 <p>See above.</p> 492 v237 <p>See above.</p> 493 v238 <p>See above.</p> 494 v239 <p>See above.</p> 495 v240 <p>See above.</p> 496 v241 <p>See above.</p> 497 v242 <p>See above.</p> 498 v243 <p>See above.</p> 499 v244 <p>See above.</p> 500 v245 <p>See above.</p> 501 v246 <p>See above.</p> 502 v247 <p>See above.</p> 503 v248 <p>See above.</p> 504 v249 <p>See above.</p> 505 v250 <p>See above.</p> 506 v251 <p>See above.</p> 507 v252 <p>See above.</p> 508 v253 <p>See above.</p> 509 v254 <p>See above.</p> 510 v255 <p>See above.</p> 511 OPR_SRC_SIMPLE N/A OPR_SDST OPR_SDST_EXEC OPR_SDST_M0 OPR_SRC_VGPR OPR_SREG OPR_SREG_NOVCC OPR_SSRC_NOLIT OPR_SSRC_SPECIAL_SCC s0 Scalar GPR 0. Increment from here for additional GPRs. There are NUM_SGPR SGPRs in total. Alignment restriction: unless otherwise noted, 64-bit SGPR values must be even-aligned (first SGPR address is even); 128-bit and wider SGPR values must be aligned to a multiple of 4. 0 s1 <p>See above.</p> 1 s2 <p>See above.</p> 2 s3 <p>See above.</p> 3 s4 <p>See above.</p> 4 s5 <p>See above.</p> 5 s6 <p>See above.</p> 6 s7 <p>See above.</p> 7 s8 <p>See above.</p> 8 s9 <p>See above.</p> 9 s10 <p>See above.</p> 10 s11 <p>See above.</p> 11 s12 <p>See above.</p> 12 s13 <p>See above.</p> 13 s14 <p>See above.</p> 14 s15 <p>See above.</p> 15 s16 <p>See above.</p> 16 s17 <p>See above.</p> 17 s18 <p>See above.</p> 18 s19 <p>See above.</p> 19 s20 <p>See above.</p> 20 s21 <p>See above.</p> 21 s22 <p>See above.</p> 22 s23 <p>See above.</p> 23 s24 <p>See above.</p> 24 s25 <p>See above.</p> 25 s26 <p>See above.</p> 26 s27 <p>See above.</p> 27 s28 <p>See above.</p> 28 s29 <p>See above.</p> 29 s30 <p>See above.</p> 30 s31 <p>See above.</p> 31 s32 <p>See above.</p> 32 s33 <p>See above.</p> 33 s34 <p>See above.</p> 34 s35 <p>See above.</p> 35 s36 <p>See above.</p> 36 s37 <p>See above.</p> 37 s38 <p>See above.</p> 38 s39 <p>See above.</p> 39 s40 <p>See above.</p> 40 s41 <p>See above.</p> 41 s42 <p>See above.</p> 42 s43 <p>See above.</p> 43 s44 <p>See above.</p> 44 s45 <p>See above.</p> 45 s46 <p>See above.</p> 46 s47 <p>See above.</p> 47 s48 <p>See above.</p> 48 s49 <p>See above.</p> 49 s50 <p>See above.</p> 50 s51 <p>See above.</p> 51 s52 <p>See above.</p> 52 s53 <p>See above.</p> 53 s54 <p>See above.</p> 54 s55 <p>See above.</p> 55 s56 <p>See above.</p> 56 s57 <p>See above.</p> 57 s58 <p>See above.</p> 58 s59 <p>See above.</p> 59 s60 <p>See above.</p> 60 s61 <p>See above.</p> 61 s62 <p>See above.</p> 62 s63 <p>See above.</p> 63 s64 <p>See above.</p> 64 s65 <p>See above.</p> 65 s66 <p>See above.</p> 66 s67 <p>See above.</p> 67 s68 <p>See above.</p> 68 s69 <p>See above.</p> 69 s70 <p>See above.</p> 70 s71 <p>See above.</p> 71 s72 <p>See above.</p> 72 s73 <p>See above.</p> 73 s74 <p>See above.</p> 74 s75 <p>See above.</p> 75 s76 <p>See above.</p> 76 s77 <p>See above.</p> 77 s78 <p>See above.</p> 78 s79 <p>See above.</p> 79 s80 <p>See above.</p> 80 s81 <p>See above.</p> 81 s82 <p>See above.</p> 82 s83 <p>See above.</p> 83 s84 <p>See above.</p> 84 s85 <p>See above.</p> 85 s86 <p>See above.</p> 86 s87 <p>See above.</p> 87 s88 <p>See above.</p> 88 s89 <p>See above.</p> 89 s90 <p>See above.</p> 90 s91 <p>See above.</p> 91 s92 <p>See above.</p> 92 s93 <p>See above.</p> 93 s94 <p>See above.</p> 94 s95 <p>See above.</p> 95 s96 <p>See above.</p> 96 s97 <p>See above.</p> 97 s98 <p>See above.</p> 98 s99 <p>See above.</p> 99 s100 <p>See above.</p> 100 s101 <p>See above.</p> 101 XNACK_MASK_LO xnack_mask[31:0], see XNACK replay mechanism for details. 104 XNACK_MASK_HI xnack_mask[63:32], see XNACK replay mechanism for details. 105 FLAT_SCRATCH_LO {13'd0, size[18:0]} 102 FLAT_SCRATCH_HI {8'd0, offset[31:8]} 103 TTMP0 Trap handler temps (privileged). Increment from here for additional TTMPs. There are NUM_TTMP TTMPs in total. {TTMP1,TTMP0} = PC_save{hi,lo}. Alignment restriction: unless otherwise noted, 64-bit TTMP values must be even-aligned (first TTMP address is even); 128-bit and wider TTMP values must be aligned to a multiple of 4. 108 TTMP1 Trap handler temps (privileged). 109 TTMP2 Trap handler temps (privileged). 110 TTMP3 Trap handler temps (privileged). 111 TTMP4 Trap handler temps (privileged). 112 TTMP5 Trap handler temps (privileged). 113 TTMP6 Trap handler temps (privileged). 114 TTMP7 Trap handler temps (privileged). 115 TTMP8 Trap handler temps (privileged). 116 TTMP9 Trap handler temps (privileged). 117 TTMP10 Trap handler temps (privileged). 118 TTMP11 Trap handler temps (privileged). 119 TTMP12 Trap handler temps (privileged). 120 TTMP13 Trap handler temps (privileged). 121 TTMP14 Trap handler temps (privileged). 122 TTMP15 Trap handler temps (privileged). 123 VCC_LO VCC[31:0] 106 VCC_HI VCC[63:32] 107 M0 Special register used to hold LDS/GDS addresses, relative indices, and send-message values. 124 EXEC_LO exec[31:0] 126 EXEC_HI exec[63:32] 127 0 0 128 1 1 (integer) 129 2 2 (integer) 130 3 3 (integer) 131 4 4 (integer) 132 5 5 (integer) 133 6 6 (integer) 134 7 7 (integer) 135 8 8 (integer) 136 9 9 (integer) 137 10 10 (integer) 138 11 11 (integer) 139 12 12 (integer) 140 13 13 (integer) 141 14 14 (integer) 142 15 15 (integer) 143 16 16 (integer) 144 17 17 (integer) 145 18 18 (integer) 146 19 19 (integer) 147 20 20 (integer) 148 21 21 (integer) 149 22 22 (integer) 150 23 23 (integer) 151 24 24 (integer) 152 25 25 (integer) 153 26 26 (integer) 154 27 27 (integer) 155 28 28 (integer) 156 29 29 (integer) 157 30 30 (integer) 158 31 31 (integer) 159 32 32 (integer) 160 33 33 (integer) 161 34 34 (integer) 162 35 35 (integer) 163 36 36 (integer) 164 37 37 (integer) 165 38 38 (integer) 166 39 39 (integer) 167 40 40 (integer) 168 41 41 (integer) 169 42 42 (integer) 170 43 43 (integer) 171 44 44 (integer) 172 45 45 (integer) 173 46 46 (integer) 174 47 47 (integer) 175 48 48 (integer) 176 49 49 (integer) 177 50 50 (integer) 178 51 51 (integer) 179 52 52 (integer) 180 53 53 (integer) 181 54 54 (integer) 182 55 55 (integer) 183 56 56 (integer) 184 57 57 (integer) 185 58 58 (integer) 186 59 59 (integer) 187 60 60 (integer) 188 61 61 (integer) 189 62 62 (integer) 190 63 63 (integer) 191 64 64 (integer) 192 -1 -1 (integer) 193 -2 -2 (integer) 194 -3 -3 (integer) 195 -4 -4 (integer) 196 -5 -5 (integer) 197 -6 -6 (integer) 198 -7 -7 (integer) 199 -8 -8 (integer) 200 -9 -9 (integer) 201 -10 -10 (integer) 202 -11 -11 (integer) 203 -12 -12 (integer) 204 -13 -13 (integer) 205 -14 -14 (integer) 206 -15 -15 (integer) 207 -16 -16 (integer) 208 0.5 0.5 240 -0.5 -0.5 241 1.0 1.0 242 -1.0 -1.0 243 2.0 2.0 244 -2.0 -2.0 245 4.0 4.0 246 -4.0 -4.0 247 0.15915494 1 / (2 * PI). 16-bit: 0x3118, 32-bit: 0x3e22f983, 64-bit: 0x3fc45f30_6dc9c882. 248 SRC_VCCZ True if vector condition code register is zero. 251 SRC_EXECZ True if execute mask is zero. 252 SRC_SCC Scalar condition code. 253 SRC_SHARED_BASE N/A 235 SRC_SHARED_LIMIT N/A 236 SRC_PRIVATE_BASE N/A 237 SRC_PRIVATE_LIMIT N/A 238 SRC_POPS_EXITING_WAVE_ID POPS counter (12b). Returns the ID of the most recent wave to complete in-order. 239 v0 Vector GPR 0. Increment from here for additional GPRs. There are NUM_VGPR VGPRs in total. Alignment restriction: unless otherwise noted, 64-bit and wider VGPR values must be even-aligned (first VGPR address is even). 256 v1 <p>See above.</p> 257 v2 <p>See above.</p> 258 v3 <p>See above.</p> 259 v4 <p>See above.</p> 260 v5 <p>See above.</p> 261 v6 <p>See above.</p> 262 v7 <p>See above.</p> 263 v8 <p>See above.</p> 264 v9 <p>See above.</p> 265 v10 <p>See above.</p> 266 v11 <p>See above.</p> 267 v12 <p>See above.</p> 268 v13 <p>See above.</p> 269 v14 <p>See above.</p> 270 v15 <p>See above.</p> 271 v16 <p>See above.</p> 272 v17 <p>See above.</p> 273 v18 <p>See above.</p> 274 v19 <p>See above.</p> 275 v20 <p>See above.</p> 276 v21 <p>See above.</p> 277 v22 <p>See above.</p> 278 v23 <p>See above.</p> 279 v24 <p>See above.</p> 280 v25 <p>See above.</p> 281 v26 <p>See above.</p> 282 v27 <p>See above.</p> 283 v28 <p>See above.</p> 284 v29 <p>See above.</p> 285 v30 <p>See above.</p> 286 v31 <p>See above.</p> 287 v32 <p>See above.</p> 288 v33 <p>See above.</p> 289 v34 <p>See above.</p> 290 v35 <p>See above.</p> 291 v36 <p>See above.</p> 292 v37 <p>See above.</p> 293 v38 <p>See above.</p> 294 v39 <p>See above.</p> 295 v40 <p>See above.</p> 296 v41 <p>See above.</p> 297 v42 <p>See above.</p> 298 v43 <p>See above.</p> 299 v44 <p>See above.</p> 300 v45 <p>See above.</p> 301 v46 <p>See above.</p> 302 v47 <p>See above.</p> 303 v48 <p>See above.</p> 304 v49 <p>See above.</p> 305 v50 <p>See above.</p> 306 v51 <p>See above.</p> 307 v52 <p>See above.</p> 308 v53 <p>See above.</p> 309 v54 <p>See above.</p> 310 v55 <p>See above.</p> 311 v56 <p>See above.</p> 312 v57 <p>See above.</p> 313 v58 <p>See above.</p> 314 v59 <p>See above.</p> 315 v60 <p>See above.</p> 316 v61 <p>See above.</p> 317 v62 <p>See above.</p> 318 v63 <p>See above.</p> 319 v64 <p>See above.</p> 320 v65 <p>See above.</p> 321 v66 <p>See above.</p> 322 v67 <p>See above.</p> 323 v68 <p>See above.</p> 324 v69 <p>See above.</p> 325 v70 <p>See above.</p> 326 v71 <p>See above.</p> 327 v72 <p>See above.</p> 328 v73 <p>See above.</p> 329 v74 <p>See above.</p> 330 v75 <p>See above.</p> 331 v76 <p>See above.</p> 332 v77 <p>See above.</p> 333 v78 <p>See above.</p> 334 v79 <p>See above.</p> 335 v80 <p>See above.</p> 336 v81 <p>See above.</p> 337 v82 <p>See above.</p> 338 v83 <p>See above.</p> 339 v84 <p>See above.</p> 340 v85 <p>See above.</p> 341 v86 <p>See above.</p> 342 v87 <p>See above.</p> 343 v88 <p>See above.</p> 344 v89 <p>See above.</p> 345 v90 <p>See above.</p> 346 v91 <p>See above.</p> 347 v92 <p>See above.</p> 348 v93 <p>See above.</p> 349 v94 <p>See above.</p> 350 v95 <p>See above.</p> 351 v96 <p>See above.</p> 352 v97 <p>See above.</p> 353 v98 <p>See above.</p> 354 v99 <p>See above.</p> 355 v100 <p>See above.</p> 356 v101 <p>See above.</p> 357 v102 <p>See above.</p> 358 v103 <p>See above.</p> 359 v104 <p>See above.</p> 360 v105 <p>See above.</p> 361 v106 <p>See above.</p> 362 v107 <p>See above.</p> 363 v108 <p>See above.</p> 364 v109 <p>See above.</p> 365 v110 <p>See above.</p> 366 v111 <p>See above.</p> 367 v112 <p>See above.</p> 368 v113 <p>See above.</p> 369 v114 <p>See above.</p> 370 v115 <p>See above.</p> 371 v116 <p>See above.</p> 372 v117 <p>See above.</p> 373 v118 <p>See above.</p> 374 v119 <p>See above.</p> 375 v120 <p>See above.</p> 376 v121 <p>See above.</p> 377 v122 <p>See above.</p> 378 v123 <p>See above.</p> 379 v124 <p>See above.</p> 380 v125 <p>See above.</p> 381 v126 <p>See above.</p> 382 v127 <p>See above.</p> 383 v128 <p>See above.</p> 384 v129 <p>See above.</p> 385 v130 <p>See above.</p> 386 v131 <p>See above.</p> 387 v132 <p>See above.</p> 388 v133 <p>See above.</p> 389 v134 <p>See above.</p> 390 v135 <p>See above.</p> 391 v136 <p>See above.</p> 392 v137 <p>See above.</p> 393 v138 <p>See above.</p> 394 v139 <p>See above.</p> 395 v140 <p>See above.</p> 396 v141 <p>See above.</p> 397 v142 <p>See above.</p> 398 v143 <p>See above.</p> 399 v144 <p>See above.</p> 400 v145 <p>See above.</p> 401 v146 <p>See above.</p> 402 v147 <p>See above.</p> 403 v148 <p>See above.</p> 404 v149 <p>See above.</p> 405 v150 <p>See above.</p> 406 v151 <p>See above.</p> 407 v152 <p>See above.</p> 408 v153 <p>See above.</p> 409 v154 <p>See above.</p> 410 v155 <p>See above.</p> 411 v156 <p>See above.</p> 412 v157 <p>See above.</p> 413 v158 <p>See above.</p> 414 v159 <p>See above.</p> 415 v160 <p>See above.</p> 416 v161 <p>See above.</p> 417 v162 <p>See above.</p> 418 v163 <p>See above.</p> 419 v164 <p>See above.</p> 420 v165 <p>See above.</p> 421 v166 <p>See above.</p> 422 v167 <p>See above.</p> 423 v168 <p>See above.</p> 424 v169 <p>See above.</p> 425 v170 <p>See above.</p> 426 v171 <p>See above.</p> 427 v172 <p>See above.</p> 428 v173 <p>See above.</p> 429 v174 <p>See above.</p> 430 v175 <p>See above.</p> 431 v176 <p>See above.</p> 432 v177 <p>See above.</p> 433 v178 <p>See above.</p> 434 v179 <p>See above.</p> 435 v180 <p>See above.</p> 436 v181 <p>See above.</p> 437 v182 <p>See above.</p> 438 v183 <p>See above.</p> 439 v184 <p>See above.</p> 440 v185 <p>See above.</p> 441 v186 <p>See above.</p> 442 v187 <p>See above.</p> 443 v188 <p>See above.</p> 444 v189 <p>See above.</p> 445 v190 <p>See above.</p> 446 v191 <p>See above.</p> 447 v192 <p>See above.</p> 448 v193 <p>See above.</p> 449 v194 <p>See above.</p> 450 v195 <p>See above.</p> 451 v196 <p>See above.</p> 452 v197 <p>See above.</p> 453 v198 <p>See above.</p> 454 v199 <p>See above.</p> 455 v200 <p>See above.</p> 456 v201 <p>See above.</p> 457 v202 <p>See above.</p> 458 v203 <p>See above.</p> 459 v204 <p>See above.</p> 460 v205 <p>See above.</p> 461 v206 <p>See above.</p> 462 v207 <p>See above.</p> 463 v208 <p>See above.</p> 464 v209 <p>See above.</p> 465 v210 <p>See above.</p> 466 v211 <p>See above.</p> 467 v212 <p>See above.</p> 468 v213 <p>See above.</p> 469 v214 <p>See above.</p> 470 v215 <p>See above.</p> 471 v216 <p>See above.</p> 472 v217 <p>See above.</p> 473 v218 <p>See above.</p> 474 v219 <p>See above.</p> 475 v220 <p>See above.</p> 476 v221 <p>See above.</p> 477 v222 <p>See above.</p> 478 v223 <p>See above.</p> 479 v224 <p>See above.</p> 480 v225 <p>See above.</p> 481 v226 <p>See above.</p> 482 v227 <p>See above.</p> 483 v228 <p>See above.</p> 484 v229 <p>See above.</p> 485 v230 <p>See above.</p> 486 v231 <p>See above.</p> 487 v232 <p>See above.</p> 488 v233 <p>See above.</p> 489 v234 <p>See above.</p> 490 v235 <p>See above.</p> 491 v236 <p>See above.</p> 492 v237 <p>See above.</p> 493 v238 <p>See above.</p> 494 v239 <p>See above.</p> 495 v240 <p>See above.</p> 496 v241 <p>See above.</p> 497 v242 <p>See above.</p> 498 v243 <p>See above.</p> 499 v244 <p>See above.</p> 500 v245 <p>See above.</p> 501 v246 <p>See above.</p> 502 v247 <p>See above.</p> 503 v248 <p>See above.</p> 504 v249 <p>See above.</p> 505 v250 <p>See above.</p> 506 v251 <p>See above.</p> 507 v252 <p>See above.</p> 508 v253 <p>See above.</p> 509 v254 <p>See above.</p> 510 v255 <p>See above.</p> 511 OPR_SRC_VGPR N/A v0 Vector GPR 0. Increment from here for additional GPRs. There are NUM_VGPR VGPRs in total. Alignment restriction: unless otherwise noted, 64-bit and wider VGPR values must be even-aligned (first VGPR address is even). 256 v1 <p>See above.</p> 257 v2 <p>See above.</p> 258 v3 <p>See above.</p> 259 v4 <p>See above.</p> 260 v5 <p>See above.</p> 261 v6 <p>See above.</p> 262 v7 <p>See above.</p> 263 v8 <p>See above.</p> 264 v9 <p>See above.</p> 265 v10 <p>See above.</p> 266 v11 <p>See above.</p> 267 v12 <p>See above.</p> 268 v13 <p>See above.</p> 269 v14 <p>See above.</p> 270 v15 <p>See above.</p> 271 v16 <p>See above.</p> 272 v17 <p>See above.</p> 273 v18 <p>See above.</p> 274 v19 <p>See above.</p> 275 v20 <p>See above.</p> 276 v21 <p>See above.</p> 277 v22 <p>See above.</p> 278 v23 <p>See above.</p> 279 v24 <p>See above.</p> 280 v25 <p>See above.</p> 281 v26 <p>See above.</p> 282 v27 <p>See above.</p> 283 v28 <p>See above.</p> 284 v29 <p>See above.</p> 285 v30 <p>See above.</p> 286 v31 <p>See above.</p> 287 v32 <p>See above.</p> 288 v33 <p>See above.</p> 289 v34 <p>See above.</p> 290 v35 <p>See above.</p> 291 v36 <p>See above.</p> 292 v37 <p>See above.</p> 293 v38 <p>See above.</p> 294 v39 <p>See above.</p> 295 v40 <p>See above.</p> 296 v41 <p>See above.</p> 297 v42 <p>See above.</p> 298 v43 <p>See above.</p> 299 v44 <p>See above.</p> 300 v45 <p>See above.</p> 301 v46 <p>See above.</p> 302 v47 <p>See above.</p> 303 v48 <p>See above.</p> 304 v49 <p>See above.</p> 305 v50 <p>See above.</p> 306 v51 <p>See above.</p> 307 v52 <p>See above.</p> 308 v53 <p>See above.</p> 309 v54 <p>See above.</p> 310 v55 <p>See above.</p> 311 v56 <p>See above.</p> 312 v57 <p>See above.</p> 313 v58 <p>See above.</p> 314 v59 <p>See above.</p> 315 v60 <p>See above.</p> 316 v61 <p>See above.</p> 317 v62 <p>See above.</p> 318 v63 <p>See above.</p> 319 v64 <p>See above.</p> 320 v65 <p>See above.</p> 321 v66 <p>See above.</p> 322 v67 <p>See above.</p> 323 v68 <p>See above.</p> 324 v69 <p>See above.</p> 325 v70 <p>See above.</p> 326 v71 <p>See above.</p> 327 v72 <p>See above.</p> 328 v73 <p>See above.</p> 329 v74 <p>See above.</p> 330 v75 <p>See above.</p> 331 v76 <p>See above.</p> 332 v77 <p>See above.</p> 333 v78 <p>See above.</p> 334 v79 <p>See above.</p> 335 v80 <p>See above.</p> 336 v81 <p>See above.</p> 337 v82 <p>See above.</p> 338 v83 <p>See above.</p> 339 v84 <p>See above.</p> 340 v85 <p>See above.</p> 341 v86 <p>See above.</p> 342 v87 <p>See above.</p> 343 v88 <p>See above.</p> 344 v89 <p>See above.</p> 345 v90 <p>See above.</p> 346 v91 <p>See above.</p> 347 v92 <p>See above.</p> 348 v93 <p>See above.</p> 349 v94 <p>See above.</p> 350 v95 <p>See above.</p> 351 v96 <p>See above.</p> 352 v97 <p>See above.</p> 353 v98 <p>See above.</p> 354 v99 <p>See above.</p> 355 v100 <p>See above.</p> 356 v101 <p>See above.</p> 357 v102 <p>See above.</p> 358 v103 <p>See above.</p> 359 v104 <p>See above.</p> 360 v105 <p>See above.</p> 361 v106 <p>See above.</p> 362 v107 <p>See above.</p> 363 v108 <p>See above.</p> 364 v109 <p>See above.</p> 365 v110 <p>See above.</p> 366 v111 <p>See above.</p> 367 v112 <p>See above.</p> 368 v113 <p>See above.</p> 369 v114 <p>See above.</p> 370 v115 <p>See above.</p> 371 v116 <p>See above.</p> 372 v117 <p>See above.</p> 373 v118 <p>See above.</p> 374 v119 <p>See above.</p> 375 v120 <p>See above.</p> 376 v121 <p>See above.</p> 377 v122 <p>See above.</p> 378 v123 <p>See above.</p> 379 v124 <p>See above.</p> 380 v125 <p>See above.</p> 381 v126 <p>See above.</p> 382 v127 <p>See above.</p> 383 v128 <p>See above.</p> 384 v129 <p>See above.</p> 385 v130 <p>See above.</p> 386 v131 <p>See above.</p> 387 v132 <p>See above.</p> 388 v133 <p>See above.</p> 389 v134 <p>See above.</p> 390 v135 <p>See above.</p> 391 v136 <p>See above.</p> 392 v137 <p>See above.</p> 393 v138 <p>See above.</p> 394 v139 <p>See above.</p> 395 v140 <p>See above.</p> 396 v141 <p>See above.</p> 397 v142 <p>See above.</p> 398 v143 <p>See above.</p> 399 v144 <p>See above.</p> 400 v145 <p>See above.</p> 401 v146 <p>See above.</p> 402 v147 <p>See above.</p> 403 v148 <p>See above.</p> 404 v149 <p>See above.</p> 405 v150 <p>See above.</p> 406 v151 <p>See above.</p> 407 v152 <p>See above.</p> 408 v153 <p>See above.</p> 409 v154 <p>See above.</p> 410 v155 <p>See above.</p> 411 v156 <p>See above.</p> 412 v157 <p>See above.</p> 413 v158 <p>See above.</p> 414 v159 <p>See above.</p> 415 v160 <p>See above.</p> 416 v161 <p>See above.</p> 417 v162 <p>See above.</p> 418 v163 <p>See above.</p> 419 v164 <p>See above.</p> 420 v165 <p>See above.</p> 421 v166 <p>See above.</p> 422 v167 <p>See above.</p> 423 v168 <p>See above.</p> 424 v169 <p>See above.</p> 425 v170 <p>See above.</p> 426 v171 <p>See above.</p> 427 v172 <p>See above.</p> 428 v173 <p>See above.</p> 429 v174 <p>See above.</p> 430 v175 <p>See above.</p> 431 v176 <p>See above.</p> 432 v177 <p>See above.</p> 433 v178 <p>See above.</p> 434 v179 <p>See above.</p> 435 v180 <p>See above.</p> 436 v181 <p>See above.</p> 437 v182 <p>See above.</p> 438 v183 <p>See above.</p> 439 v184 <p>See above.</p> 440 v185 <p>See above.</p> 441 v186 <p>See above.</p> 442 v187 <p>See above.</p> 443 v188 <p>See above.</p> 444 v189 <p>See above.</p> 445 v190 <p>See above.</p> 446 v191 <p>See above.</p> 447 v192 <p>See above.</p> 448 v193 <p>See above.</p> 449 v194 <p>See above.</p> 450 v195 <p>See above.</p> 451 v196 <p>See above.</p> 452 v197 <p>See above.</p> 453 v198 <p>See above.</p> 454 v199 <p>See above.</p> 455 v200 <p>See above.</p> 456 v201 <p>See above.</p> 457 v202 <p>See above.</p> 458 v203 <p>See above.</p> 459 v204 <p>See above.</p> 460 v205 <p>See above.</p> 461 v206 <p>See above.</p> 462 v207 <p>See above.</p> 463 v208 <p>See above.</p> 464 v209 <p>See above.</p> 465 v210 <p>See above.</p> 466 v211 <p>See above.</p> 467 v212 <p>See above.</p> 468 v213 <p>See above.</p> 469 v214 <p>See above.</p> 470 v215 <p>See above.</p> 471 v216 <p>See above.</p> 472 v217 <p>See above.</p> 473 v218 <p>See above.</p> 474 v219 <p>See above.</p> 475 v220 <p>See above.</p> 476 v221 <p>See above.</p> 477 v222 <p>See above.</p> 478 v223 <p>See above.</p> 479 v224 <p>See above.</p> 480 v225 <p>See above.</p> 481 v226 <p>See above.</p> 482 v227 <p>See above.</p> 483 v228 <p>See above.</p> 484 v229 <p>See above.</p> 485 v230 <p>See above.</p> 486 v231 <p>See above.</p> 487 v232 <p>See above.</p> 488 v233 <p>See above.</p> 489 v234 <p>See above.</p> 490 v235 <p>See above.</p> 491 v236 <p>See above.</p> 492 v237 <p>See above.</p> 493 v238 <p>See above.</p> 494 v239 <p>See above.</p> 495 v240 <p>See above.</p> 496 v241 <p>See above.</p> 497 v242 <p>See above.</p> 498 v243 <p>See above.</p> 499 v244 <p>See above.</p> 500 v245 <p>See above.</p> 501 v246 <p>See above.</p> 502 v247 <p>See above.</p> 503 v248 <p>See above.</p> 504 v249 <p>See above.</p> 505 v250 <p>See above.</p> 506 v251 <p>See above.</p> 507 v252 <p>See above.</p> 508 v253 <p>See above.</p> 509 v254 <p>See above.</p> 510 v255 <p>See above.</p> 511 OPR_SRC_VGPR_OR_ACCVGPR N/A OPR_SRC_ACCVGPR OPR_SRC_VGPR v0 Vector GPR 0. Increment from here for additional GPRs. There are NUM_VGPR VGPRs in total. Alignment restriction: unless otherwise noted, 64-bit and wider VGPR values must be even-aligned (first VGPR address is even). 256 v1 <p>See above.</p> 257 v2 <p>See above.</p> 258 v3 <p>See above.</p> 259 v4 <p>See above.</p> 260 v5 <p>See above.</p> 261 v6 <p>See above.</p> 262 v7 <p>See above.</p> 263 v8 <p>See above.</p> 264 v9 <p>See above.</p> 265 v10 <p>See above.</p> 266 v11 <p>See above.</p> 267 v12 <p>See above.</p> 268 v13 <p>See above.</p> 269 v14 <p>See above.</p> 270 v15 <p>See above.</p> 271 v16 <p>See above.</p> 272 v17 <p>See above.</p> 273 v18 <p>See above.</p> 274 v19 <p>See above.</p> 275 v20 <p>See above.</p> 276 v21 <p>See above.</p> 277 v22 <p>See above.</p> 278 v23 <p>See above.</p> 279 v24 <p>See above.</p> 280 v25 <p>See above.</p> 281 v26 <p>See above.</p> 282 v27 <p>See above.</p> 283 v28 <p>See above.</p> 284 v29 <p>See above.</p> 285 v30 <p>See above.</p> 286 v31 <p>See above.</p> 287 v32 <p>See above.</p> 288 v33 <p>See above.</p> 289 v34 <p>See above.</p> 290 v35 <p>See above.</p> 291 v36 <p>See above.</p> 292 v37 <p>See above.</p> 293 v38 <p>See above.</p> 294 v39 <p>See above.</p> 295 v40 <p>See above.</p> 296 v41 <p>See above.</p> 297 v42 <p>See above.</p> 298 v43 <p>See above.</p> 299 v44 <p>See above.</p> 300 v45 <p>See above.</p> 301 v46 <p>See above.</p> 302 v47 <p>See above.</p> 303 v48 <p>See above.</p> 304 v49 <p>See above.</p> 305 v50 <p>See above.</p> 306 v51 <p>See above.</p> 307 v52 <p>See above.</p> 308 v53 <p>See above.</p> 309 v54 <p>See above.</p> 310 v55 <p>See above.</p> 311 v56 <p>See above.</p> 312 v57 <p>See above.</p> 313 v58 <p>See above.</p> 314 v59 <p>See above.</p> 315 v60 <p>See above.</p> 316 v61 <p>See above.</p> 317 v62 <p>See above.</p> 318 v63 <p>See above.</p> 319 v64 <p>See above.</p> 320 v65 <p>See above.</p> 321 v66 <p>See above.</p> 322 v67 <p>See above.</p> 323 v68 <p>See above.</p> 324 v69 <p>See above.</p> 325 v70 <p>See above.</p> 326 v71 <p>See above.</p> 327 v72 <p>See above.</p> 328 v73 <p>See above.</p> 329 v74 <p>See above.</p> 330 v75 <p>See above.</p> 331 v76 <p>See above.</p> 332 v77 <p>See above.</p> 333 v78 <p>See above.</p> 334 v79 <p>See above.</p> 335 v80 <p>See above.</p> 336 v81 <p>See above.</p> 337 v82 <p>See above.</p> 338 v83 <p>See above.</p> 339 v84 <p>See above.</p> 340 v85 <p>See above.</p> 341 v86 <p>See above.</p> 342 v87 <p>See above.</p> 343 v88 <p>See above.</p> 344 v89 <p>See above.</p> 345 v90 <p>See above.</p> 346 v91 <p>See above.</p> 347 v92 <p>See above.</p> 348 v93 <p>See above.</p> 349 v94 <p>See above.</p> 350 v95 <p>See above.</p> 351 v96 <p>See above.</p> 352 v97 <p>See above.</p> 353 v98 <p>See above.</p> 354 v99 <p>See above.</p> 355 v100 <p>See above.</p> 356 v101 <p>See above.</p> 357 v102 <p>See above.</p> 358 v103 <p>See above.</p> 359 v104 <p>See above.</p> 360 v105 <p>See above.</p> 361 v106 <p>See above.</p> 362 v107 <p>See above.</p> 363 v108 <p>See above.</p> 364 v109 <p>See above.</p> 365 v110 <p>See above.</p> 366 v111 <p>See above.</p> 367 v112 <p>See above.</p> 368 v113 <p>See above.</p> 369 v114 <p>See above.</p> 370 v115 <p>See above.</p> 371 v116 <p>See above.</p> 372 v117 <p>See above.</p> 373 v118 <p>See above.</p> 374 v119 <p>See above.</p> 375 v120 <p>See above.</p> 376 v121 <p>See above.</p> 377 v122 <p>See above.</p> 378 v123 <p>See above.</p> 379 v124 <p>See above.</p> 380 v125 <p>See above.</p> 381 v126 <p>See above.</p> 382 v127 <p>See above.</p> 383 v128 <p>See above.</p> 384 v129 <p>See above.</p> 385 v130 <p>See above.</p> 386 v131 <p>See above.</p> 387 v132 <p>See above.</p> 388 v133 <p>See above.</p> 389 v134 <p>See above.</p> 390 v135 <p>See above.</p> 391 v136 <p>See above.</p> 392 v137 <p>See above.</p> 393 v138 <p>See above.</p> 394 v139 <p>See above.</p> 395 v140 <p>See above.</p> 396 v141 <p>See above.</p> 397 v142 <p>See above.</p> 398 v143 <p>See above.</p> 399 v144 <p>See above.</p> 400 v145 <p>See above.</p> 401 v146 <p>See above.</p> 402 v147 <p>See above.</p> 403 v148 <p>See above.</p> 404 v149 <p>See above.</p> 405 v150 <p>See above.</p> 406 v151 <p>See above.</p> 407 v152 <p>See above.</p> 408 v153 <p>See above.</p> 409 v154 <p>See above.</p> 410 v155 <p>See above.</p> 411 v156 <p>See above.</p> 412 v157 <p>See above.</p> 413 v158 <p>See above.</p> 414 v159 <p>See above.</p> 415 v160 <p>See above.</p> 416 v161 <p>See above.</p> 417 v162 <p>See above.</p> 418 v163 <p>See above.</p> 419 v164 <p>See above.</p> 420 v165 <p>See above.</p> 421 v166 <p>See above.</p> 422 v167 <p>See above.</p> 423 v168 <p>See above.</p> 424 v169 <p>See above.</p> 425 v170 <p>See above.</p> 426 v171 <p>See above.</p> 427 v172 <p>See above.</p> 428 v173 <p>See above.</p> 429 v174 <p>See above.</p> 430 v175 <p>See above.</p> 431 v176 <p>See above.</p> 432 v177 <p>See above.</p> 433 v178 <p>See above.</p> 434 v179 <p>See above.</p> 435 v180 <p>See above.</p> 436 v181 <p>See above.</p> 437 v182 <p>See above.</p> 438 v183 <p>See above.</p> 439 v184 <p>See above.</p> 440 v185 <p>See above.</p> 441 v186 <p>See above.</p> 442 v187 <p>See above.</p> 443 v188 <p>See above.</p> 444 v189 <p>See above.</p> 445 v190 <p>See above.</p> 446 v191 <p>See above.</p> 447 v192 <p>See above.</p> 448 v193 <p>See above.</p> 449 v194 <p>See above.</p> 450 v195 <p>See above.</p> 451 v196 <p>See above.</p> 452 v197 <p>See above.</p> 453 v198 <p>See above.</p> 454 v199 <p>See above.</p> 455 v200 <p>See above.</p> 456 v201 <p>See above.</p> 457 v202 <p>See above.</p> 458 v203 <p>See above.</p> 459 v204 <p>See above.</p> 460 v205 <p>See above.</p> 461 v206 <p>See above.</p> 462 v207 <p>See above.</p> 463 v208 <p>See above.</p> 464 v209 <p>See above.</p> 465 v210 <p>See above.</p> 466 v211 <p>See above.</p> 467 v212 <p>See above.</p> 468 v213 <p>See above.</p> 469 v214 <p>See above.</p> 470 v215 <p>See above.</p> 471 v216 <p>See above.</p> 472 v217 <p>See above.</p> 473 v218 <p>See above.</p> 474 v219 <p>See above.</p> 475 v220 <p>See above.</p> 476 v221 <p>See above.</p> 477 v222 <p>See above.</p> 478 v223 <p>See above.</p> 479 v224 <p>See above.</p> 480 v225 <p>See above.</p> 481 v226 <p>See above.</p> 482 v227 <p>See above.</p> 483 v228 <p>See above.</p> 484 v229 <p>See above.</p> 485 v230 <p>See above.</p> 486 v231 <p>See above.</p> 487 v232 <p>See above.</p> 488 v233 <p>See above.</p> 489 v234 <p>See above.</p> 490 v235 <p>See above.</p> 491 v236 <p>See above.</p> 492 v237 <p>See above.</p> 493 v238 <p>See above.</p> 494 v239 <p>See above.</p> 495 v240 <p>See above.</p> 496 v241 <p>See above.</p> 497 v242 <p>See above.</p> 498 v243 <p>See above.</p> 499 v244 <p>See above.</p> 500 v245 <p>See above.</p> 501 v246 <p>See above.</p> 502 v247 <p>See above.</p> 503 v248 <p>See above.</p> 504 v249 <p>See above.</p> 505 v250 <p>See above.</p> 506 v251 <p>See above.</p> 507 v252 <p>See above.</p> 508 v253 <p>See above.</p> 509 v254 <p>See above.</p> 510 v255 <p>See above.</p> 511 acc0 Accumulator VGPR 0. Increment from here for additional GPRs. There are NUM_ACCVGPR ACCVGPRs in total. Alignment restriction: unless otherwise noted, 64-bit and wider ACCVGPR values must be even-aligned (first ACCVGPR address is even). 768 acc1 <p>See above.</p> 769 acc2 <p>See above.</p> 770 acc3 <p>See above.</p> 771 acc4 <p>See above.</p> 772 acc5 <p>See above.</p> 773 acc6 <p>See above.</p> 774 acc7 <p>See above.</p> 775 acc8 <p>See above.</p> 776 acc9 <p>See above.</p> 777 acc10 <p>See above.</p> 778 acc11 <p>See above.</p> 779 acc12 <p>See above.</p> 780 acc13 <p>See above.</p> 781 acc14 <p>See above.</p> 782 acc15 <p>See above.</p> 783 acc16 <p>See above.</p> 784 acc17 <p>See above.</p> 785 acc18 <p>See above.</p> 786 acc19 <p>See above.</p> 787 acc20 <p>See above.</p> 788 acc21 <p>See above.</p> 789 acc22 <p>See above.</p> 790 acc23 <p>See above.</p> 791 acc24 <p>See above.</p> 792 acc25 <p>See above.</p> 793 acc26 <p>See above.</p> 794 acc27 <p>See above.</p> 795 acc28 <p>See above.</p> 796 acc29 <p>See above.</p> 797 acc30 <p>See above.</p> 798 acc31 <p>See above.</p> 799 acc32 <p>See above.</p> 800 acc33 <p>See above.</p> 801 acc34 <p>See above.</p> 802 acc35 <p>See above.</p> 803 acc36 <p>See above.</p> 804 acc37 <p>See above.</p> 805 acc38 <p>See above.</p> 806 acc39 <p>See above.</p> 807 acc40 <p>See above.</p> 808 acc41 <p>See above.</p> 809 acc42 <p>See above.</p> 810 acc43 <p>See above.</p> 811 acc44 <p>See above.</p> 812 acc45 <p>See above.</p> 813 acc46 <p>See above.</p> 814 acc47 <p>See above.</p> 815 acc48 <p>See above.</p> 816 acc49 <p>See above.</p> 817 acc50 <p>See above.</p> 818 acc51 <p>See above.</p> 819 acc52 <p>See above.</p> 820 acc53 <p>See above.</p> 821 acc54 <p>See above.</p> 822 acc55 <p>See above.</p> 823 acc56 <p>See above.</p> 824 acc57 <p>See above.</p> 825 acc58 <p>See above.</p> 826 acc59 <p>See above.</p> 827 acc60 <p>See above.</p> 828 acc61 <p>See above.</p> 829 acc62 <p>See above.</p> 830 acc63 <p>See above.</p> 831 acc64 <p>See above.</p> 832 acc65 <p>See above.</p> 833 acc66 <p>See above.</p> 834 acc67 <p>See above.</p> 835 acc68 <p>See above.</p> 836 acc69 <p>See above.</p> 837 acc70 <p>See above.</p> 838 acc71 <p>See above.</p> 839 acc72 <p>See above.</p> 840 acc73 <p>See above.</p> 841 acc74 <p>See above.</p> 842 acc75 <p>See above.</p> 843 acc76 <p>See above.</p> 844 acc77 <p>See above.</p> 845 acc78 <p>See above.</p> 846 acc79 <p>See above.</p> 847 acc80 <p>See above.</p> 848 acc81 <p>See above.</p> 849 acc82 <p>See above.</p> 850 acc83 <p>See above.</p> 851 acc84 <p>See above.</p> 852 acc85 <p>See above.</p> 853 acc86 <p>See above.</p> 854 acc87 <p>See above.</p> 855 acc88 <p>See above.</p> 856 acc89 <p>See above.</p> 857 acc90 <p>See above.</p> 858 acc91 <p>See above.</p> 859 acc92 <p>See above.</p> 860 acc93 <p>See above.</p> 861 acc94 <p>See above.</p> 862 acc95 <p>See above.</p> 863 acc96 <p>See above.</p> 864 acc97 <p>See above.</p> 865 acc98 <p>See above.</p> 866 acc99 <p>See above.</p> 867 acc100 <p>See above.</p> 868 acc101 <p>See above.</p> 869 acc102 <p>See above.</p> 870 acc103 <p>See above.</p> 871 acc104 <p>See above.</p> 872 acc105 <p>See above.</p> 873 acc106 <p>See above.</p> 874 acc107 <p>See above.</p> 875 acc108 <p>See above.</p> 876 acc109 <p>See above.</p> 877 acc110 <p>See above.</p> 878 acc111 <p>See above.</p> 879 acc112 <p>See above.</p> 880 acc113 <p>See above.</p> 881 acc114 <p>See above.</p> 882 acc115 <p>See above.</p> 883 acc116 <p>See above.</p> 884 acc117 <p>See above.</p> 885 acc118 <p>See above.</p> 886 acc119 <p>See above.</p> 887 acc120 <p>See above.</p> 888 acc121 <p>See above.</p> 889 acc122 <p>See above.</p> 890 acc123 <p>See above.</p> 891 acc124 <p>See above.</p> 892 acc125 <p>See above.</p> 893 acc126 <p>See above.</p> 894 acc127 <p>See above.</p> 895 acc128 <p>See above.</p> 896 acc129 <p>See above.</p> 897 acc130 <p>See above.</p> 898 acc131 <p>See above.</p> 899 acc132 <p>See above.</p> 900 acc133 <p>See above.</p> 901 acc134 <p>See above.</p> 902 acc135 <p>See above.</p> 903 acc136 <p>See above.</p> 904 acc137 <p>See above.</p> 905 acc138 <p>See above.</p> 906 acc139 <p>See above.</p> 907 acc140 <p>See above.</p> 908 acc141 <p>See above.</p> 909 acc142 <p>See above.</p> 910 acc143 <p>See above.</p> 911 acc144 <p>See above.</p> 912 acc145 <p>See above.</p> 913 acc146 <p>See above.</p> 914 acc147 <p>See above.</p> 915 acc148 <p>See above.</p> 916 acc149 <p>See above.</p> 917 acc150 <p>See above.</p> 918 acc151 <p>See above.</p> 919 acc152 <p>See above.</p> 920 acc153 <p>See above.</p> 921 acc154 <p>See above.</p> 922 acc155 <p>See above.</p> 923 acc156 <p>See above.</p> 924 acc157 <p>See above.</p> 925 acc158 <p>See above.</p> 926 acc159 <p>See above.</p> 927 acc160 <p>See above.</p> 928 acc161 <p>See above.</p> 929 acc162 <p>See above.</p> 930 acc163 <p>See above.</p> 931 acc164 <p>See above.</p> 932 acc165 <p>See above.</p> 933 acc166 <p>See above.</p> 934 acc167 <p>See above.</p> 935 acc168 <p>See above.</p> 936 acc169 <p>See above.</p> 937 acc170 <p>See above.</p> 938 acc171 <p>See above.</p> 939 acc172 <p>See above.</p> 940 acc173 <p>See above.</p> 941 acc174 <p>See above.</p> 942 acc175 <p>See above.</p> 943 acc176 <p>See above.</p> 944 acc177 <p>See above.</p> 945 acc178 <p>See above.</p> 946 acc179 <p>See above.</p> 947 acc180 <p>See above.</p> 948 acc181 <p>See above.</p> 949 acc182 <p>See above.</p> 950 acc183 <p>See above.</p> 951 acc184 <p>See above.</p> 952 acc185 <p>See above.</p> 953 acc186 <p>See above.</p> 954 acc187 <p>See above.</p> 955 acc188 <p>See above.</p> 956 acc189 <p>See above.</p> 957 acc190 <p>See above.</p> 958 acc191 <p>See above.</p> 959 acc192 <p>See above.</p> 960 acc193 <p>See above.</p> 961 acc194 <p>See above.</p> 962 acc195 <p>See above.</p> 963 acc196 <p>See above.</p> 964 acc197 <p>See above.</p> 965 acc198 <p>See above.</p> 966 acc199 <p>See above.</p> 967 acc200 <p>See above.</p> 968 acc201 <p>See above.</p> 969 acc202 <p>See above.</p> 970 acc203 <p>See above.</p> 971 acc204 <p>See above.</p> 972 acc205 <p>See above.</p> 973 acc206 <p>See above.</p> 974 acc207 <p>See above.</p> 975 acc208 <p>See above.</p> 976 acc209 <p>See above.</p> 977 acc210 <p>See above.</p> 978 acc211 <p>See above.</p> 979 acc212 <p>See above.</p> 980 acc213 <p>See above.</p> 981 acc214 <p>See above.</p> 982 acc215 <p>See above.</p> 983 acc216 <p>See above.</p> 984 acc217 <p>See above.</p> 985 acc218 <p>See above.</p> 986 acc219 <p>See above.</p> 987 acc220 <p>See above.</p> 988 acc221 <p>See above.</p> 989 acc222 <p>See above.</p> 990 acc223 <p>See above.</p> 991 acc224 <p>See above.</p> 992 acc225 <p>See above.</p> 993 acc226 <p>See above.</p> 994 acc227 <p>See above.</p> 995 acc228 <p>See above.</p> 996 acc229 <p>See above.</p> 997 acc230 <p>See above.</p> 998 acc231 <p>See above.</p> 999 acc232 <p>See above.</p> 1000 acc233 <p>See above.</p> 1001 acc234 <p>See above.</p> 1002 acc235 <p>See above.</p> 1003 acc236 <p>See above.</p> 1004 acc237 <p>See above.</p> 1005 acc238 <p>See above.</p> 1006 acc239 <p>See above.</p> 1007 acc240 <p>See above.</p> 1008 acc241 <p>See above.</p> 1009 acc242 <p>See above.</p> 1010 acc243 <p>See above.</p> 1011 acc244 <p>See above.</p> 1012 acc245 <p>See above.</p> 1013 acc246 <p>See above.</p> 1014 acc247 <p>See above.</p> 1015 acc248 <p>See above.</p> 1016 acc249 <p>See above.</p> 1017 acc250 <p>See above.</p> 1018 acc251 <p>See above.</p> 1019 acc252 <p>See above.</p> 1020 acc253 <p>See above.</p> 1021 acc254 <p>See above.</p> 1022 acc255 <p>See above.</p> 1023 OPR_SRC_VGPR_OR_ACCVGPR_OR_CONST N/A OPR_SRC_ACCVGPR OPR_SRC_VGPR OPR_SRC_VGPR_OR_ACCVGPR v0 Vector GPR 0. Increment from here for additional GPRs. There are NUM_VGPR VGPRs in total. Alignment restriction: unless otherwise noted, 64-bit and wider VGPR values must be even-aligned (first VGPR address is even). 256 v1 <p>See above.</p> 257 v2 <p>See above.</p> 258 v3 <p>See above.</p> 259 v4 <p>See above.</p> 260 v5 <p>See above.</p> 261 v6 <p>See above.</p> 262 v7 <p>See above.</p> 263 v8 <p>See above.</p> 264 v9 <p>See above.</p> 265 v10 <p>See above.</p> 266 v11 <p>See above.</p> 267 v12 <p>See above.</p> 268 v13 <p>See above.</p> 269 v14 <p>See above.</p> 270 v15 <p>See above.</p> 271 v16 <p>See above.</p> 272 v17 <p>See above.</p> 273 v18 <p>See above.</p> 274 v19 <p>See above.</p> 275 v20 <p>See above.</p> 276 v21 <p>See above.</p> 277 v22 <p>See above.</p> 278 v23 <p>See above.</p> 279 v24 <p>See above.</p> 280 v25 <p>See above.</p> 281 v26 <p>See above.</p> 282 v27 <p>See above.</p> 283 v28 <p>See above.</p> 284 v29 <p>See above.</p> 285 v30 <p>See above.</p> 286 v31 <p>See above.</p> 287 v32 <p>See above.</p> 288 v33 <p>See above.</p> 289 v34 <p>See above.</p> 290 v35 <p>See above.</p> 291 v36 <p>See above.</p> 292 v37 <p>See above.</p> 293 v38 <p>See above.</p> 294 v39 <p>See above.</p> 295 v40 <p>See above.</p> 296 v41 <p>See above.</p> 297 v42 <p>See above.</p> 298 v43 <p>See above.</p> 299 v44 <p>See above.</p> 300 v45 <p>See above.</p> 301 v46 <p>See above.</p> 302 v47 <p>See above.</p> 303 v48 <p>See above.</p> 304 v49 <p>See above.</p> 305 v50 <p>See above.</p> 306 v51 <p>See above.</p> 307 v52 <p>See above.</p> 308 v53 <p>See above.</p> 309 v54 <p>See above.</p> 310 v55 <p>See above.</p> 311 v56 <p>See above.</p> 312 v57 <p>See above.</p> 313 v58 <p>See above.</p> 314 v59 <p>See above.</p> 315 v60 <p>See above.</p> 316 v61 <p>See above.</p> 317 v62 <p>See above.</p> 318 v63 <p>See above.</p> 319 v64 <p>See above.</p> 320 v65 <p>See above.</p> 321 v66 <p>See above.</p> 322 v67 <p>See above.</p> 323 v68 <p>See above.</p> 324 v69 <p>See above.</p> 325 v70 <p>See above.</p> 326 v71 <p>See above.</p> 327 v72 <p>See above.</p> 328 v73 <p>See above.</p> 329 v74 <p>See above.</p> 330 v75 <p>See above.</p> 331 v76 <p>See above.</p> 332 v77 <p>See above.</p> 333 v78 <p>See above.</p> 334 v79 <p>See above.</p> 335 v80 <p>See above.</p> 336 v81 <p>See above.</p> 337 v82 <p>See above.</p> 338 v83 <p>See above.</p> 339 v84 <p>See above.</p> 340 v85 <p>See above.</p> 341 v86 <p>See above.</p> 342 v87 <p>See above.</p> 343 v88 <p>See above.</p> 344 v89 <p>See above.</p> 345 v90 <p>See above.</p> 346 v91 <p>See above.</p> 347 v92 <p>See above.</p> 348 v93 <p>See above.</p> 349 v94 <p>See above.</p> 350 v95 <p>See above.</p> 351 v96 <p>See above.</p> 352 v97 <p>See above.</p> 353 v98 <p>See above.</p> 354 v99 <p>See above.</p> 355 v100 <p>See above.</p> 356 v101 <p>See above.</p> 357 v102 <p>See above.</p> 358 v103 <p>See above.</p> 359 v104 <p>See above.</p> 360 v105 <p>See above.</p> 361 v106 <p>See above.</p> 362 v107 <p>See above.</p> 363 v108 <p>See above.</p> 364 v109 <p>See above.</p> 365 v110 <p>See above.</p> 366 v111 <p>See above.</p> 367 v112 <p>See above.</p> 368 v113 <p>See above.</p> 369 v114 <p>See above.</p> 370 v115 <p>See above.</p> 371 v116 <p>See above.</p> 372 v117 <p>See above.</p> 373 v118 <p>See above.</p> 374 v119 <p>See above.</p> 375 v120 <p>See above.</p> 376 v121 <p>See above.</p> 377 v122 <p>See above.</p> 378 v123 <p>See above.</p> 379 v124 <p>See above.</p> 380 v125 <p>See above.</p> 381 v126 <p>See above.</p> 382 v127 <p>See above.</p> 383 v128 <p>See above.</p> 384 v129 <p>See above.</p> 385 v130 <p>See above.</p> 386 v131 <p>See above.</p> 387 v132 <p>See above.</p> 388 v133 <p>See above.</p> 389 v134 <p>See above.</p> 390 v135 <p>See above.</p> 391 v136 <p>See above.</p> 392 v137 <p>See above.</p> 393 v138 <p>See above.</p> 394 v139 <p>See above.</p> 395 v140 <p>See above.</p> 396 v141 <p>See above.</p> 397 v142 <p>See above.</p> 398 v143 <p>See above.</p> 399 v144 <p>See above.</p> 400 v145 <p>See above.</p> 401 v146 <p>See above.</p> 402 v147 <p>See above.</p> 403 v148 <p>See above.</p> 404 v149 <p>See above.</p> 405 v150 <p>See above.</p> 406 v151 <p>See above.</p> 407 v152 <p>See above.</p> 408 v153 <p>See above.</p> 409 v154 <p>See above.</p> 410 v155 <p>See above.</p> 411 v156 <p>See above.</p> 412 v157 <p>See above.</p> 413 v158 <p>See above.</p> 414 v159 <p>See above.</p> 415 v160 <p>See above.</p> 416 v161 <p>See above.</p> 417 v162 <p>See above.</p> 418 v163 <p>See above.</p> 419 v164 <p>See above.</p> 420 v165 <p>See above.</p> 421 v166 <p>See above.</p> 422 v167 <p>See above.</p> 423 v168 <p>See above.</p> 424 v169 <p>See above.</p> 425 v170 <p>See above.</p> 426 v171 <p>See above.</p> 427 v172 <p>See above.</p> 428 v173 <p>See above.</p> 429 v174 <p>See above.</p> 430 v175 <p>See above.</p> 431 v176 <p>See above.</p> 432 v177 <p>See above.</p> 433 v178 <p>See above.</p> 434 v179 <p>See above.</p> 435 v180 <p>See above.</p> 436 v181 <p>See above.</p> 437 v182 <p>See above.</p> 438 v183 <p>See above.</p> 439 v184 <p>See above.</p> 440 v185 <p>See above.</p> 441 v186 <p>See above.</p> 442 v187 <p>See above.</p> 443 v188 <p>See above.</p> 444 v189 <p>See above.</p> 445 v190 <p>See above.</p> 446 v191 <p>See above.</p> 447 v192 <p>See above.</p> 448 v193 <p>See above.</p> 449 v194 <p>See above.</p> 450 v195 <p>See above.</p> 451 v196 <p>See above.</p> 452 v197 <p>See above.</p> 453 v198 <p>See above.</p> 454 v199 <p>See above.</p> 455 v200 <p>See above.</p> 456 v201 <p>See above.</p> 457 v202 <p>See above.</p> 458 v203 <p>See above.</p> 459 v204 <p>See above.</p> 460 v205 <p>See above.</p> 461 v206 <p>See above.</p> 462 v207 <p>See above.</p> 463 v208 <p>See above.</p> 464 v209 <p>See above.</p> 465 v210 <p>See above.</p> 466 v211 <p>See above.</p> 467 v212 <p>See above.</p> 468 v213 <p>See above.</p> 469 v214 <p>See above.</p> 470 v215 <p>See above.</p> 471 v216 <p>See above.</p> 472 v217 <p>See above.</p> 473 v218 <p>See above.</p> 474 v219 <p>See above.</p> 475 v220 <p>See above.</p> 476 v221 <p>See above.</p> 477 v222 <p>See above.</p> 478 v223 <p>See above.</p> 479 v224 <p>See above.</p> 480 v225 <p>See above.</p> 481 v226 <p>See above.</p> 482 v227 <p>See above.</p> 483 v228 <p>See above.</p> 484 v229 <p>See above.</p> 485 v230 <p>See above.</p> 486 v231 <p>See above.</p> 487 v232 <p>See above.</p> 488 v233 <p>See above.</p> 489 v234 <p>See above.</p> 490 v235 <p>See above.</p> 491 v236 <p>See above.</p> 492 v237 <p>See above.</p> 493 v238 <p>See above.</p> 494 v239 <p>See above.</p> 495 v240 <p>See above.</p> 496 v241 <p>See above.</p> 497 v242 <p>See above.</p> 498 v243 <p>See above.</p> 499 v244 <p>See above.</p> 500 v245 <p>See above.</p> 501 v246 <p>See above.</p> 502 v247 <p>See above.</p> 503 v248 <p>See above.</p> 504 v249 <p>See above.</p> 505 v250 <p>See above.</p> 506 v251 <p>See above.</p> 507 v252 <p>See above.</p> 508 v253 <p>See above.</p> 509 v254 <p>See above.</p> 510 v255 <p>See above.</p> 511 acc0 Accumulator VGPR 0. Increment from here for additional GPRs. There are NUM_ACCVGPR ACCVGPRs in total. Alignment restriction: unless otherwise noted, 64-bit and wider ACCVGPR values must be even-aligned (first ACCVGPR address is even). 768 acc1 <p>See above.</p> 769 acc2 <p>See above.</p> 770 acc3 <p>See above.</p> 771 acc4 <p>See above.</p> 772 acc5 <p>See above.</p> 773 acc6 <p>See above.</p> 774 acc7 <p>See above.</p> 775 acc8 <p>See above.</p> 776 acc9 <p>See above.</p> 777 acc10 <p>See above.</p> 778 acc11 <p>See above.</p> 779 acc12 <p>See above.</p> 780 acc13 <p>See above.</p> 781 acc14 <p>See above.</p> 782 acc15 <p>See above.</p> 783 acc16 <p>See above.</p> 784 acc17 <p>See above.</p> 785 acc18 <p>See above.</p> 786 acc19 <p>See above.</p> 787 acc20 <p>See above.</p> 788 acc21 <p>See above.</p> 789 acc22 <p>See above.</p> 790 acc23 <p>See above.</p> 791 acc24 <p>See above.</p> 792 acc25 <p>See above.</p> 793 acc26 <p>See above.</p> 794 acc27 <p>See above.</p> 795 acc28 <p>See above.</p> 796 acc29 <p>See above.</p> 797 acc30 <p>See above.</p> 798 acc31 <p>See above.</p> 799 acc32 <p>See above.</p> 800 acc33 <p>See above.</p> 801 acc34 <p>See above.</p> 802 acc35 <p>See above.</p> 803 acc36 <p>See above.</p> 804 acc37 <p>See above.</p> 805 acc38 <p>See above.</p> 806 acc39 <p>See above.</p> 807 acc40 <p>See above.</p> 808 acc41 <p>See above.</p> 809 acc42 <p>See above.</p> 810 acc43 <p>See above.</p> 811 acc44 <p>See above.</p> 812 acc45 <p>See above.</p> 813 acc46 <p>See above.</p> 814 acc47 <p>See above.</p> 815 acc48 <p>See above.</p> 816 acc49 <p>See above.</p> 817 acc50 <p>See above.</p> 818 acc51 <p>See above.</p> 819 acc52 <p>See above.</p> 820 acc53 <p>See above.</p> 821 acc54 <p>See above.</p> 822 acc55 <p>See above.</p> 823 acc56 <p>See above.</p> 824 acc57 <p>See above.</p> 825 acc58 <p>See above.</p> 826 acc59 <p>See above.</p> 827 acc60 <p>See above.</p> 828 acc61 <p>See above.</p> 829 acc62 <p>See above.</p> 830 acc63 <p>See above.</p> 831 acc64 <p>See above.</p> 832 acc65 <p>See above.</p> 833 acc66 <p>See above.</p> 834 acc67 <p>See above.</p> 835 acc68 <p>See above.</p> 836 acc69 <p>See above.</p> 837 acc70 <p>See above.</p> 838 acc71 <p>See above.</p> 839 acc72 <p>See above.</p> 840 acc73 <p>See above.</p> 841 acc74 <p>See above.</p> 842 acc75 <p>See above.</p> 843 acc76 <p>See above.</p> 844 acc77 <p>See above.</p> 845 acc78 <p>See above.</p> 846 acc79 <p>See above.</p> 847 acc80 <p>See above.</p> 848 acc81 <p>See above.</p> 849 acc82 <p>See above.</p> 850 acc83 <p>See above.</p> 851 acc84 <p>See above.</p> 852 acc85 <p>See above.</p> 853 acc86 <p>See above.</p> 854 acc87 <p>See above.</p> 855 acc88 <p>See above.</p> 856 acc89 <p>See above.</p> 857 acc90 <p>See above.</p> 858 acc91 <p>See above.</p> 859 acc92 <p>See above.</p> 860 acc93 <p>See above.</p> 861 acc94 <p>See above.</p> 862 acc95 <p>See above.</p> 863 acc96 <p>See above.</p> 864 acc97 <p>See above.</p> 865 acc98 <p>See above.</p> 866 acc99 <p>See above.</p> 867 acc100 <p>See above.</p> 868 acc101 <p>See above.</p> 869 acc102 <p>See above.</p> 870 acc103 <p>See above.</p> 871 acc104 <p>See above.</p> 872 acc105 <p>See above.</p> 873 acc106 <p>See above.</p> 874 acc107 <p>See above.</p> 875 acc108 <p>See above.</p> 876 acc109 <p>See above.</p> 877 acc110 <p>See above.</p> 878 acc111 <p>See above.</p> 879 acc112 <p>See above.</p> 880 acc113 <p>See above.</p> 881 acc114 <p>See above.</p> 882 acc115 <p>See above.</p> 883 acc116 <p>See above.</p> 884 acc117 <p>See above.</p> 885 acc118 <p>See above.</p> 886 acc119 <p>See above.</p> 887 acc120 <p>See above.</p> 888 acc121 <p>See above.</p> 889 acc122 <p>See above.</p> 890 acc123 <p>See above.</p> 891 acc124 <p>See above.</p> 892 acc125 <p>See above.</p> 893 acc126 <p>See above.</p> 894 acc127 <p>See above.</p> 895 acc128 <p>See above.</p> 896 acc129 <p>See above.</p> 897 acc130 <p>See above.</p> 898 acc131 <p>See above.</p> 899 acc132 <p>See above.</p> 900 acc133 <p>See above.</p> 901 acc134 <p>See above.</p> 902 acc135 <p>See above.</p> 903 acc136 <p>See above.</p> 904 acc137 <p>See above.</p> 905 acc138 <p>See above.</p> 906 acc139 <p>See above.</p> 907 acc140 <p>See above.</p> 908 acc141 <p>See above.</p> 909 acc142 <p>See above.</p> 910 acc143 <p>See above.</p> 911 acc144 <p>See above.</p> 912 acc145 <p>See above.</p> 913 acc146 <p>See above.</p> 914 acc147 <p>See above.</p> 915 acc148 <p>See above.</p> 916 acc149 <p>See above.</p> 917 acc150 <p>See above.</p> 918 acc151 <p>See above.</p> 919 acc152 <p>See above.</p> 920 acc153 <p>See above.</p> 921 acc154 <p>See above.</p> 922 acc155 <p>See above.</p> 923 acc156 <p>See above.</p> 924 acc157 <p>See above.</p> 925 acc158 <p>See above.</p> 926 acc159 <p>See above.</p> 927 acc160 <p>See above.</p> 928 acc161 <p>See above.</p> 929 acc162 <p>See above.</p> 930 acc163 <p>See above.</p> 931 acc164 <p>See above.</p> 932 acc165 <p>See above.</p> 933 acc166 <p>See above.</p> 934 acc167 <p>See above.</p> 935 acc168 <p>See above.</p> 936 acc169 <p>See above.</p> 937 acc170 <p>See above.</p> 938 acc171 <p>See above.</p> 939 acc172 <p>See above.</p> 940 acc173 <p>See above.</p> 941 acc174 <p>See above.</p> 942 acc175 <p>See above.</p> 943 acc176 <p>See above.</p> 944 acc177 <p>See above.</p> 945 acc178 <p>See above.</p> 946 acc179 <p>See above.</p> 947 acc180 <p>See above.</p> 948 acc181 <p>See above.</p> 949 acc182 <p>See above.</p> 950 acc183 <p>See above.</p> 951 acc184 <p>See above.</p> 952 acc185 <p>See above.</p> 953 acc186 <p>See above.</p> 954 acc187 <p>See above.</p> 955 acc188 <p>See above.</p> 956 acc189 <p>See above.</p> 957 acc190 <p>See above.</p> 958 acc191 <p>See above.</p> 959 acc192 <p>See above.</p> 960 acc193 <p>See above.</p> 961 acc194 <p>See above.</p> 962 acc195 <p>See above.</p> 963 acc196 <p>See above.</p> 964 acc197 <p>See above.</p> 965 acc198 <p>See above.</p> 966 acc199 <p>See above.</p> 967 acc200 <p>See above.</p> 968 acc201 <p>See above.</p> 969 acc202 <p>See above.</p> 970 acc203 <p>See above.</p> 971 acc204 <p>See above.</p> 972 acc205 <p>See above.</p> 973 acc206 <p>See above.</p> 974 acc207 <p>See above.</p> 975 acc208 <p>See above.</p> 976 acc209 <p>See above.</p> 977 acc210 <p>See above.</p> 978 acc211 <p>See above.</p> 979 acc212 <p>See above.</p> 980 acc213 <p>See above.</p> 981 acc214 <p>See above.</p> 982 acc215 <p>See above.</p> 983 acc216 <p>See above.</p> 984 acc217 <p>See above.</p> 985 acc218 <p>See above.</p> 986 acc219 <p>See above.</p> 987 acc220 <p>See above.</p> 988 acc221 <p>See above.</p> 989 acc222 <p>See above.</p> 990 acc223 <p>See above.</p> 991 acc224 <p>See above.</p> 992 acc225 <p>See above.</p> 993 acc226 <p>See above.</p> 994 acc227 <p>See above.</p> 995 acc228 <p>See above.</p> 996 acc229 <p>See above.</p> 997 acc230 <p>See above.</p> 998 acc231 <p>See above.</p> 999 acc232 <p>See above.</p> 1000 acc233 <p>See above.</p> 1001 acc234 <p>See above.</p> 1002 acc235 <p>See above.</p> 1003 acc236 <p>See above.</p> 1004 acc237 <p>See above.</p> 1005 acc238 <p>See above.</p> 1006 acc239 <p>See above.</p> 1007 acc240 <p>See above.</p> 1008 acc241 <p>See above.</p> 1009 acc242 <p>See above.</p> 1010 acc243 <p>See above.</p> 1011 acc244 <p>See above.</p> 1012 acc245 <p>See above.</p> 1013 acc246 <p>See above.</p> 1014 acc247 <p>See above.</p> 1015 acc248 <p>See above.</p> 1016 acc249 <p>See above.</p> 1017 acc250 <p>See above.</p> 1018 acc251 <p>See above.</p> 1019 acc252 <p>See above.</p> 1020 acc253 <p>See above.</p> 1021 acc254 <p>See above.</p> 1022 acc255 <p>See above.</p> 1023 0 0 128 1 1 (integer) 129 2 2 (integer) 130 3 3 (integer) 131 4 4 (integer) 132 5 5 (integer) 133 6 6 (integer) 134 7 7 (integer) 135 8 8 (integer) 136 9 9 (integer) 137 10 10 (integer) 138 11 11 (integer) 139 12 12 (integer) 140 13 13 (integer) 141 14 14 (integer) 142 15 15 (integer) 143 16 16 (integer) 144 17 17 (integer) 145 18 18 (integer) 146 19 19 (integer) 147 20 20 (integer) 148 21 21 (integer) 149 22 22 (integer) 150 23 23 (integer) 151 24 24 (integer) 152 25 25 (integer) 153 26 26 (integer) 154 27 27 (integer) 155 28 28 (integer) 156 29 29 (integer) 157 30 30 (integer) 158 31 31 (integer) 159 32 32 (integer) 160 33 33 (integer) 161 34 34 (integer) 162 35 35 (integer) 163 36 36 (integer) 164 37 37 (integer) 165 38 38 (integer) 166 39 39 (integer) 167 40 40 (integer) 168 41 41 (integer) 169 42 42 (integer) 170 43 43 (integer) 171 44 44 (integer) 172 45 45 (integer) 173 46 46 (integer) 174 47 47 (integer) 175 48 48 (integer) 176 49 49 (integer) 177 50 50 (integer) 178 51 51 (integer) 179 52 52 (integer) 180 53 53 (integer) 181 54 54 (integer) 182 55 55 (integer) 183 56 56 (integer) 184 57 57 (integer) 185 58 58 (integer) 186 59 59 (integer) 187 60 60 (integer) 188 61 61 (integer) 189 62 62 (integer) 190 63 63 (integer) 191 64 64 (integer) 192 -1 -1 (integer) 193 -2 -2 (integer) 194 -3 -3 (integer) 195 -4 -4 (integer) 196 -5 -5 (integer) 197 -6 -6 (integer) 198 -7 -7 (integer) 199 -8 -8 (integer) 200 -9 -9 (integer) 201 -10 -10 (integer) 202 -11 -11 (integer) 203 -12 -12 (integer) 204 -13 -13 (integer) 205 -14 -14 (integer) 206 -15 -15 (integer) 207 -16 -16 (integer) 208 0.5 0.5 240 -0.5 -0.5 241 1.0 1.0 242 -1.0 -1.0 243 2.0 2.0 244 -2.0 -2.0 245 4.0 4.0 246 -4.0 -4.0 247 0.15915494 1 / (2 * PI). 16-bit: 0x3118, 32-bit: 0x3e22f983, 64-bit: 0x3fc45f30_6dc9c882. 248 OPR_SREG N/A OPR_SREG_NOVCC s0 Scalar GPR 0. Increment from here for additional GPRs. There are NUM_SGPR SGPRs in total. Alignment restriction: unless otherwise noted, 64-bit SGPR values must be even-aligned (first SGPR address is even); 128-bit and wider SGPR values must be aligned to a multiple of 4. 0 s1 <p>See above.</p> 1 s2 <p>See above.</p> 2 s3 <p>See above.</p> 3 s4 <p>See above.</p> 4 s5 <p>See above.</p> 5 s6 <p>See above.</p> 6 s7 <p>See above.</p> 7 s8 <p>See above.</p> 8 s9 <p>See above.</p> 9 s10 <p>See above.</p> 10 s11 <p>See above.</p> 11 s12 <p>See above.</p> 12 s13 <p>See above.</p> 13 s14 <p>See above.</p> 14 s15 <p>See above.</p> 15 s16 <p>See above.</p> 16 s17 <p>See above.</p> 17 s18 <p>See above.</p> 18 s19 <p>See above.</p> 19 s20 <p>See above.</p> 20 s21 <p>See above.</p> 21 s22 <p>See above.</p> 22 s23 <p>See above.</p> 23 s24 <p>See above.</p> 24 s25 <p>See above.</p> 25 s26 <p>See above.</p> 26 s27 <p>See above.</p> 27 s28 <p>See above.</p> 28 s29 <p>See above.</p> 29 s30 <p>See above.</p> 30 s31 <p>See above.</p> 31 s32 <p>See above.</p> 32 s33 <p>See above.</p> 33 s34 <p>See above.</p> 34 s35 <p>See above.</p> 35 s36 <p>See above.</p> 36 s37 <p>See above.</p> 37 s38 <p>See above.</p> 38 s39 <p>See above.</p> 39 s40 <p>See above.</p> 40 s41 <p>See above.</p> 41 s42 <p>See above.</p> 42 s43 <p>See above.</p> 43 s44 <p>See above.</p> 44 s45 <p>See above.</p> 45 s46 <p>See above.</p> 46 s47 <p>See above.</p> 47 s48 <p>See above.</p> 48 s49 <p>See above.</p> 49 s50 <p>See above.</p> 50 s51 <p>See above.</p> 51 s52 <p>See above.</p> 52 s53 <p>See above.</p> 53 s54 <p>See above.</p> 54 s55 <p>See above.</p> 55 s56 <p>See above.</p> 56 s57 <p>See above.</p> 57 s58 <p>See above.</p> 58 s59 <p>See above.</p> 59 s60 <p>See above.</p> 60 s61 <p>See above.</p> 61 s62 <p>See above.</p> 62 s63 <p>See above.</p> 63 s64 <p>See above.</p> 64 s65 <p>See above.</p> 65 s66 <p>See above.</p> 66 s67 <p>See above.</p> 67 s68 <p>See above.</p> 68 s69 <p>See above.</p> 69 s70 <p>See above.</p> 70 s71 <p>See above.</p> 71 s72 <p>See above.</p> 72 s73 <p>See above.</p> 73 s74 <p>See above.</p> 74 s75 <p>See above.</p> 75 s76 <p>See above.</p> 76 s77 <p>See above.</p> 77 s78 <p>See above.</p> 78 s79 <p>See above.</p> 79 s80 <p>See above.</p> 80 s81 <p>See above.</p> 81 s82 <p>See above.</p> 82 s83 <p>See above.</p> 83 s84 <p>See above.</p> 84 s85 <p>See above.</p> 85 s86 <p>See above.</p> 86 s87 <p>See above.</p> 87 s88 <p>See above.</p> 88 s89 <p>See above.</p> 89 s90 <p>See above.</p> 90 s91 <p>See above.</p> 91 s92 <p>See above.</p> 92 s93 <p>See above.</p> 93 s94 <p>See above.</p> 94 s95 <p>See above.</p> 95 s96 <p>See above.</p> 96 s97 <p>See above.</p> 97 s98 <p>See above.</p> 98 s99 <p>See above.</p> 99 s100 <p>See above.</p> 100 s101 <p>See above.</p> 101 XNACK_MASK_LO xnack_mask[31:0], see XNACK replay mechanism for details. 104 XNACK_MASK_HI xnack_mask[63:32], see XNACK replay mechanism for details. 105 FLAT_SCRATCH_LO {13'd0, size[18:0]} 102 FLAT_SCRATCH_HI {8'd0, offset[31:8]} 103 TTMP0 Trap handler temps (privileged). Increment from here for additional TTMPs. There are NUM_TTMP TTMPs in total. {TTMP1,TTMP0} = PC_save{hi,lo}. Alignment restriction: unless otherwise noted, 64-bit TTMP values must be even-aligned (first TTMP address is even); 128-bit and wider TTMP values must be aligned to a multiple of 4. 108 TTMP1 Trap handler temps (privileged). 109 TTMP2 Trap handler temps (privileged). 110 TTMP3 Trap handler temps (privileged). 111 TTMP4 Trap handler temps (privileged). 112 TTMP5 Trap handler temps (privileged). 113 TTMP6 Trap handler temps (privileged). 114 TTMP7 Trap handler temps (privileged). 115 TTMP8 Trap handler temps (privileged). 116 TTMP9 Trap handler temps (privileged). 117 TTMP10 Trap handler temps (privileged). 118 TTMP11 Trap handler temps (privileged). 119 TTMP12 Trap handler temps (privileged). 120 TTMP13 Trap handler temps (privileged). 121 TTMP14 Trap handler temps (privileged). 122 TTMP15 Trap handler temps (privileged). 123 VCC_LO VCC[31:0] 106 VCC_HI VCC[63:32] 107 OPR_SREG_NOVCC N/A s0 Scalar GPR 0. Increment from here for additional GPRs. There are NUM_SGPR SGPRs in total. Alignment restriction: unless otherwise noted, 64-bit SGPR values must be even-aligned (first SGPR address is even); 128-bit and wider SGPR values must be aligned to a multiple of 4. 0 s1 <p>See above.</p> 1 s2 <p>See above.</p> 2 s3 <p>See above.</p> 3 s4 <p>See above.</p> 4 s5 <p>See above.</p> 5 s6 <p>See above.</p> 6 s7 <p>See above.</p> 7 s8 <p>See above.</p> 8 s9 <p>See above.</p> 9 s10 <p>See above.</p> 10 s11 <p>See above.</p> 11 s12 <p>See above.</p> 12 s13 <p>See above.</p> 13 s14 <p>See above.</p> 14 s15 <p>See above.</p> 15 s16 <p>See above.</p> 16 s17 <p>See above.</p> 17 s18 <p>See above.</p> 18 s19 <p>See above.</p> 19 s20 <p>See above.</p> 20 s21 <p>See above.</p> 21 s22 <p>See above.</p> 22 s23 <p>See above.</p> 23 s24 <p>See above.</p> 24 s25 <p>See above.</p> 25 s26 <p>See above.</p> 26 s27 <p>See above.</p> 27 s28 <p>See above.</p> 28 s29 <p>See above.</p> 29 s30 <p>See above.</p> 30 s31 <p>See above.</p> 31 s32 <p>See above.</p> 32 s33 <p>See above.</p> 33 s34 <p>See above.</p> 34 s35 <p>See above.</p> 35 s36 <p>See above.</p> 36 s37 <p>See above.</p> 37 s38 <p>See above.</p> 38 s39 <p>See above.</p> 39 s40 <p>See above.</p> 40 s41 <p>See above.</p> 41 s42 <p>See above.</p> 42 s43 <p>See above.</p> 43 s44 <p>See above.</p> 44 s45 <p>See above.</p> 45 s46 <p>See above.</p> 46 s47 <p>See above.</p> 47 s48 <p>See above.</p> 48 s49 <p>See above.</p> 49 s50 <p>See above.</p> 50 s51 <p>See above.</p> 51 s52 <p>See above.</p> 52 s53 <p>See above.</p> 53 s54 <p>See above.</p> 54 s55 <p>See above.</p> 55 s56 <p>See above.</p> 56 s57 <p>See above.</p> 57 s58 <p>See above.</p> 58 s59 <p>See above.</p> 59 s60 <p>See above.</p> 60 s61 <p>See above.</p> 61 s62 <p>See above.</p> 62 s63 <p>See above.</p> 63 s64 <p>See above.</p> 64 s65 <p>See above.</p> 65 s66 <p>See above.</p> 66 s67 <p>See above.</p> 67 s68 <p>See above.</p> 68 s69 <p>See above.</p> 69 s70 <p>See above.</p> 70 s71 <p>See above.</p> 71 s72 <p>See above.</p> 72 s73 <p>See above.</p> 73 s74 <p>See above.</p> 74 s75 <p>See above.</p> 75 s76 <p>See above.</p> 76 s77 <p>See above.</p> 77 s78 <p>See above.</p> 78 s79 <p>See above.</p> 79 s80 <p>See above.</p> 80 s81 <p>See above.</p> 81 s82 <p>See above.</p> 82 s83 <p>See above.</p> 83 s84 <p>See above.</p> 84 s85 <p>See above.</p> 85 s86 <p>See above.</p> 86 s87 <p>See above.</p> 87 s88 <p>See above.</p> 88 s89 <p>See above.</p> 89 s90 <p>See above.</p> 90 s91 <p>See above.</p> 91 s92 <p>See above.</p> 92 s93 <p>See above.</p> 93 s94 <p>See above.</p> 94 s95 <p>See above.</p> 95 s96 <p>See above.</p> 96 s97 <p>See above.</p> 97 s98 <p>See above.</p> 98 s99 <p>See above.</p> 99 s100 <p>See above.</p> 100 s101 <p>See above.</p> 101 XNACK_MASK_LO xnack_mask[31:0], see XNACK replay mechanism for details. 104 XNACK_MASK_HI xnack_mask[63:32], see XNACK replay mechanism for details. 105 FLAT_SCRATCH_LO {13'd0, size[18:0]} 102 FLAT_SCRATCH_HI {8'd0, offset[31:8]} 103 TTMP0 Trap handler temps (privileged). Increment from here for additional TTMPs. There are NUM_TTMP TTMPs in total. {TTMP1,TTMP0} = PC_save{hi,lo}. Alignment restriction: unless otherwise noted, 64-bit TTMP values must be even-aligned (first TTMP address is even); 128-bit and wider TTMP values must be aligned to a multiple of 4. 108 TTMP1 Trap handler temps (privileged). 109 TTMP2 Trap handler temps (privileged). 110 TTMP3 Trap handler temps (privileged). 111 TTMP4 Trap handler temps (privileged). 112 TTMP5 Trap handler temps (privileged). 113 TTMP6 Trap handler temps (privileged). 114 TTMP7 Trap handler temps (privileged). 115 TTMP8 Trap handler temps (privileged). 116 TTMP9 Trap handler temps (privileged). 117 TTMP10 Trap handler temps (privileged). 118 TTMP11 Trap handler temps (privileged). 119 TTMP12 Trap handler temps (privileged). 120 TTMP13 Trap handler temps (privileged). 121 TTMP14 Trap handler temps (privileged). 122 TTMP15 Trap handler temps (privileged). 123 OPR_SSRC N/A OPR_SDST OPR_SDST_EXEC OPR_SDST_M0 OPR_SREG OPR_SREG_NOVCC OPR_SSRC_NOLIT OPR_SSRC_SPECIAL_SCC s0 Scalar GPR 0. Increment from here for additional GPRs. There are NUM_SGPR SGPRs in total. Alignment restriction: unless otherwise noted, 64-bit SGPR values must be even-aligned (first SGPR address is even); 128-bit and wider SGPR values must be aligned to a multiple of 4. 0 s1 <p>See above.</p> 1 s2 <p>See above.</p> 2 s3 <p>See above.</p> 3 s4 <p>See above.</p> 4 s5 <p>See above.</p> 5 s6 <p>See above.</p> 6 s7 <p>See above.</p> 7 s8 <p>See above.</p> 8 s9 <p>See above.</p> 9 s10 <p>See above.</p> 10 s11 <p>See above.</p> 11 s12 <p>See above.</p> 12 s13 <p>See above.</p> 13 s14 <p>See above.</p> 14 s15 <p>See above.</p> 15 s16 <p>See above.</p> 16 s17 <p>See above.</p> 17 s18 <p>See above.</p> 18 s19 <p>See above.</p> 19 s20 <p>See above.</p> 20 s21 <p>See above.</p> 21 s22 <p>See above.</p> 22 s23 <p>See above.</p> 23 s24 <p>See above.</p> 24 s25 <p>See above.</p> 25 s26 <p>See above.</p> 26 s27 <p>See above.</p> 27 s28 <p>See above.</p> 28 s29 <p>See above.</p> 29 s30 <p>See above.</p> 30 s31 <p>See above.</p> 31 s32 <p>See above.</p> 32 s33 <p>See above.</p> 33 s34 <p>See above.</p> 34 s35 <p>See above.</p> 35 s36 <p>See above.</p> 36 s37 <p>See above.</p> 37 s38 <p>See above.</p> 38 s39 <p>See above.</p> 39 s40 <p>See above.</p> 40 s41 <p>See above.</p> 41 s42 <p>See above.</p> 42 s43 <p>See above.</p> 43 s44 <p>See above.</p> 44 s45 <p>See above.</p> 45 s46 <p>See above.</p> 46 s47 <p>See above.</p> 47 s48 <p>See above.</p> 48 s49 <p>See above.</p> 49 s50 <p>See above.</p> 50 s51 <p>See above.</p> 51 s52 <p>See above.</p> 52 s53 <p>See above.</p> 53 s54 <p>See above.</p> 54 s55 <p>See above.</p> 55 s56 <p>See above.</p> 56 s57 <p>See above.</p> 57 s58 <p>See above.</p> 58 s59 <p>See above.</p> 59 s60 <p>See above.</p> 60 s61 <p>See above.</p> 61 s62 <p>See above.</p> 62 s63 <p>See above.</p> 63 s64 <p>See above.</p> 64 s65 <p>See above.</p> 65 s66 <p>See above.</p> 66 s67 <p>See above.</p> 67 s68 <p>See above.</p> 68 s69 <p>See above.</p> 69 s70 <p>See above.</p> 70 s71 <p>See above.</p> 71 s72 <p>See above.</p> 72 s73 <p>See above.</p> 73 s74 <p>See above.</p> 74 s75 <p>See above.</p> 75 s76 <p>See above.</p> 76 s77 <p>See above.</p> 77 s78 <p>See above.</p> 78 s79 <p>See above.</p> 79 s80 <p>See above.</p> 80 s81 <p>See above.</p> 81 s82 <p>See above.</p> 82 s83 <p>See above.</p> 83 s84 <p>See above.</p> 84 s85 <p>See above.</p> 85 s86 <p>See above.</p> 86 s87 <p>See above.</p> 87 s88 <p>See above.</p> 88 s89 <p>See above.</p> 89 s90 <p>See above.</p> 90 s91 <p>See above.</p> 91 s92 <p>See above.</p> 92 s93 <p>See above.</p> 93 s94 <p>See above.</p> 94 s95 <p>See above.</p> 95 s96 <p>See above.</p> 96 s97 <p>See above.</p> 97 s98 <p>See above.</p> 98 s99 <p>See above.</p> 99 s100 <p>See above.</p> 100 s101 <p>See above.</p> 101 XNACK_MASK_LO xnack_mask[31:0], see XNACK replay mechanism for details. 104 XNACK_MASK_HI xnack_mask[63:32], see XNACK replay mechanism for details. 105 FLAT_SCRATCH_LO {13'd0, size[18:0]} 102 FLAT_SCRATCH_HI {8'd0, offset[31:8]} 103 TTMP0 Trap handler temps (privileged). Increment from here for additional TTMPs. There are NUM_TTMP TTMPs in total. {TTMP1,TTMP0} = PC_save{hi,lo}. Alignment restriction: unless otherwise noted, 64-bit TTMP values must be even-aligned (first TTMP address is even); 128-bit and wider TTMP values must be aligned to a multiple of 4. 108 TTMP1 Trap handler temps (privileged). 109 TTMP2 Trap handler temps (privileged). 110 TTMP3 Trap handler temps (privileged). 111 TTMP4 Trap handler temps (privileged). 112 TTMP5 Trap handler temps (privileged). 113 TTMP6 Trap handler temps (privileged). 114 TTMP7 Trap handler temps (privileged). 115 TTMP8 Trap handler temps (privileged). 116 TTMP9 Trap handler temps (privileged). 117 TTMP10 Trap handler temps (privileged). 118 TTMP11 Trap handler temps (privileged). 119 TTMP12 Trap handler temps (privileged). 120 TTMP13 Trap handler temps (privileged). 121 TTMP14 Trap handler temps (privileged). 122 TTMP15 Trap handler temps (privileged). 123 VCC_LO VCC[31:0] 106 VCC_HI VCC[63:32] 107 M0 Special register used to hold LDS/GDS addresses, relative indices, and send-message values. 124 EXEC_LO exec[31:0] 126 EXEC_HI exec[63:32] 127 0 0 128 1 1 (integer) 129 2 2 (integer) 130 3 3 (integer) 131 4 4 (integer) 132 5 5 (integer) 133 6 6 (integer) 134 7 7 (integer) 135 8 8 (integer) 136 9 9 (integer) 137 10 10 (integer) 138 11 11 (integer) 139 12 12 (integer) 140 13 13 (integer) 141 14 14 (integer) 142 15 15 (integer) 143 16 16 (integer) 144 17 17 (integer) 145 18 18 (integer) 146 19 19 (integer) 147 20 20 (integer) 148 21 21 (integer) 149 22 22 (integer) 150 23 23 (integer) 151 24 24 (integer) 152 25 25 (integer) 153 26 26 (integer) 154 27 27 (integer) 155 28 28 (integer) 156 29 29 (integer) 157 30 30 (integer) 158 31 31 (integer) 159 32 32 (integer) 160 33 33 (integer) 161 34 34 (integer) 162 35 35 (integer) 163 36 36 (integer) 164 37 37 (integer) 165 38 38 (integer) 166 39 39 (integer) 167 40 40 (integer) 168 41 41 (integer) 169 42 42 (integer) 170 43 43 (integer) 171 44 44 (integer) 172 45 45 (integer) 173 46 46 (integer) 174 47 47 (integer) 175 48 48 (integer) 176 49 49 (integer) 177 50 50 (integer) 178 51 51 (integer) 179 52 52 (integer) 180 53 53 (integer) 181 54 54 (integer) 182 55 55 (integer) 183 56 56 (integer) 184 57 57 (integer) 185 58 58 (integer) 186 59 59 (integer) 187 60 60 (integer) 188 61 61 (integer) 189 62 62 (integer) 190 63 63 (integer) 191 64 64 (integer) 192 -1 -1 (integer) 193 -2 -2 (integer) 194 -3 -3 (integer) 195 -4 -4 (integer) 196 -5 -5 (integer) 197 -6 -6 (integer) 198 -7 -7 (integer) 199 -8 -8 (integer) 200 -9 -9 (integer) 201 -10 -10 (integer) 202 -11 -11 (integer) 203 -12 -12 (integer) 204 -13 -13 (integer) 205 -14 -14 (integer) 206 -15 -15 (integer) 207 -16 -16 (integer) 208 0.5 0.5 240 -0.5 -0.5 241 1.0 1.0 242 -1.0 -1.0 243 2.0 2.0 244 -2.0 -2.0 245 4.0 4.0 246 -4.0 -4.0 247 0.15915494 1 / (2 * PI). 16-bit: 0x3118, 32-bit: 0x3e22f983, 64-bit: 0x3fc45f30_6dc9c882. 248 SRC_VCCZ True if vector condition code register is zero. 251 SRC_EXECZ True if execute mask is zero. 252 SRC_SCC Scalar condition code. 253 SRC_SHARED_BASE N/A 235 SRC_SHARED_LIMIT N/A 236 SRC_PRIVATE_BASE N/A 237 SRC_PRIVATE_LIMIT N/A 238 SRC_POPS_EXITING_WAVE_ID POPS counter (12b). Returns the ID of the most recent wave to complete in-order. 239 SRC_LITERAL 32-bit literal constant follows this instruction. 16-bit operands use only the 16 LSBs of this constant. 64-bit unsigned integer operands zero-extend the constant. 64-bit signed integer operands sign-extend the constant. 64-bit floating-point operands use the constant as the high 32 MSBs of a double-precision floating point value (1 sign bit, 11 exponent bits and 20 mantissa bits). 255 OPR_SSRC_LANESEL N/A OPR_SDST_M0 OPR_SREG OPR_SREG_NOVCC s0 Scalar GPR 0. Increment from here for additional GPRs. There are NUM_SGPR SGPRs in total. Alignment restriction: unless otherwise noted, 64-bit SGPR values must be even-aligned (first SGPR address is even); 128-bit and wider SGPR values must be aligned to a multiple of 4. 0 s1 <p>See above.</p> 1 s2 <p>See above.</p> 2 s3 <p>See above.</p> 3 s4 <p>See above.</p> 4 s5 <p>See above.</p> 5 s6 <p>See above.</p> 6 s7 <p>See above.</p> 7 s8 <p>See above.</p> 8 s9 <p>See above.</p> 9 s10 <p>See above.</p> 10 s11 <p>See above.</p> 11 s12 <p>See above.</p> 12 s13 <p>See above.</p> 13 s14 <p>See above.</p> 14 s15 <p>See above.</p> 15 s16 <p>See above.</p> 16 s17 <p>See above.</p> 17 s18 <p>See above.</p> 18 s19 <p>See above.</p> 19 s20 <p>See above.</p> 20 s21 <p>See above.</p> 21 s22 <p>See above.</p> 22 s23 <p>See above.</p> 23 s24 <p>See above.</p> 24 s25 <p>See above.</p> 25 s26 <p>See above.</p> 26 s27 <p>See above.</p> 27 s28 <p>See above.</p> 28 s29 <p>See above.</p> 29 s30 <p>See above.</p> 30 s31 <p>See above.</p> 31 s32 <p>See above.</p> 32 s33 <p>See above.</p> 33 s34 <p>See above.</p> 34 s35 <p>See above.</p> 35 s36 <p>See above.</p> 36 s37 <p>See above.</p> 37 s38 <p>See above.</p> 38 s39 <p>See above.</p> 39 s40 <p>See above.</p> 40 s41 <p>See above.</p> 41 s42 <p>See above.</p> 42 s43 <p>See above.</p> 43 s44 <p>See above.</p> 44 s45 <p>See above.</p> 45 s46 <p>See above.</p> 46 s47 <p>See above.</p> 47 s48 <p>See above.</p> 48 s49 <p>See above.</p> 49 s50 <p>See above.</p> 50 s51 <p>See above.</p> 51 s52 <p>See above.</p> 52 s53 <p>See above.</p> 53 s54 <p>See above.</p> 54 s55 <p>See above.</p> 55 s56 <p>See above.</p> 56 s57 <p>See above.</p> 57 s58 <p>See above.</p> 58 s59 <p>See above.</p> 59 s60 <p>See above.</p> 60 s61 <p>See above.</p> 61 s62 <p>See above.</p> 62 s63 <p>See above.</p> 63 s64 <p>See above.</p> 64 s65 <p>See above.</p> 65 s66 <p>See above.</p> 66 s67 <p>See above.</p> 67 s68 <p>See above.</p> 68 s69 <p>See above.</p> 69 s70 <p>See above.</p> 70 s71 <p>See above.</p> 71 s72 <p>See above.</p> 72 s73 <p>See above.</p> 73 s74 <p>See above.</p> 74 s75 <p>See above.</p> 75 s76 <p>See above.</p> 76 s77 <p>See above.</p> 77 s78 <p>See above.</p> 78 s79 <p>See above.</p> 79 s80 <p>See above.</p> 80 s81 <p>See above.</p> 81 s82 <p>See above.</p> 82 s83 <p>See above.</p> 83 s84 <p>See above.</p> 84 s85 <p>See above.</p> 85 s86 <p>See above.</p> 86 s87 <p>See above.</p> 87 s88 <p>See above.</p> 88 s89 <p>See above.</p> 89 s90 <p>See above.</p> 90 s91 <p>See above.</p> 91 s92 <p>See above.</p> 92 s93 <p>See above.</p> 93 s94 <p>See above.</p> 94 s95 <p>See above.</p> 95 s96 <p>See above.</p> 96 s97 <p>See above.</p> 97 s98 <p>See above.</p> 98 s99 <p>See above.</p> 99 s100 <p>See above.</p> 100 s101 <p>See above.</p> 101 XNACK_MASK_LO xnack_mask[31:0], see XNACK replay mechanism for details. 104 XNACK_MASK_HI xnack_mask[63:32], see XNACK replay mechanism for details. 105 FLAT_SCRATCH_LO {13'd0, size[18:0]} 102 FLAT_SCRATCH_HI {8'd0, offset[31:8]} 103 TTMP0 Trap handler temps (privileged). Increment from here for additional TTMPs. There are NUM_TTMP TTMPs in total. {TTMP1,TTMP0} = PC_save{hi,lo}. Alignment restriction: unless otherwise noted, 64-bit TTMP values must be even-aligned (first TTMP address is even); 128-bit and wider TTMP values must be aligned to a multiple of 4. 108 TTMP1 Trap handler temps (privileged). 109 TTMP2 Trap handler temps (privileged). 110 TTMP3 Trap handler temps (privileged). 111 TTMP4 Trap handler temps (privileged). 112 TTMP5 Trap handler temps (privileged). 113 TTMP6 Trap handler temps (privileged). 114 TTMP7 Trap handler temps (privileged). 115 TTMP8 Trap handler temps (privileged). 116 TTMP9 Trap handler temps (privileged). 117 TTMP10 Trap handler temps (privileged). 118 TTMP11 Trap handler temps (privileged). 119 TTMP12 Trap handler temps (privileged). 120 TTMP13 Trap handler temps (privileged). 121 TTMP14 Trap handler temps (privileged). 122 TTMP15 Trap handler temps (privileged). 123 VCC_LO VCC[31:0] 106 VCC_HI VCC[63:32] 107 M0 Special register used to hold LDS/GDS addresses, relative indices, and send-message values. 124 0 0 128 1 1 (integer) 129 2 2 (integer) 130 3 3 (integer) 131 4 4 (integer) 132 5 5 (integer) 133 6 6 (integer) 134 7 7 (integer) 135 8 8 (integer) 136 9 9 (integer) 137 10 10 (integer) 138 11 11 (integer) 139 12 12 (integer) 140 13 13 (integer) 141 14 14 (integer) 142 15 15 (integer) 143 16 16 (integer) 144 17 17 (integer) 145 18 18 (integer) 146 19 19 (integer) 147 20 20 (integer) 148 21 21 (integer) 149 22 22 (integer) 150 23 23 (integer) 151 24 24 (integer) 152 25 25 (integer) 153 26 26 (integer) 154 27 27 (integer) 155 28 28 (integer) 156 29 29 (integer) 157 30 30 (integer) 158 31 31 (integer) 159 32 32 (integer) 160 33 33 (integer) 161 34 34 (integer) 162 35 35 (integer) 163 36 36 (integer) 164 37 37 (integer) 165 38 38 (integer) 166 39 39 (integer) 167 40 40 (integer) 168 41 41 (integer) 169 42 42 (integer) 170 43 43 (integer) 171 44 44 (integer) 172 45 45 (integer) 173 46 46 (integer) 174 47 47 (integer) 175 48 48 (integer) 176 49 49 (integer) 177 50 50 (integer) 178 51 51 (integer) 179 52 52 (integer) 180 53 53 (integer) 181 54 54 (integer) 182 55 55 (integer) 183 56 56 (integer) 184 57 57 (integer) 185 58 58 (integer) 186 59 59 (integer) 187 60 60 (integer) 188 61 61 (integer) 189 62 62 (integer) 190 63 63 (integer) 191 OPR_SSRC_NOLIT N/A OPR_SDST OPR_SDST_EXEC OPR_SDST_M0 OPR_SREG OPR_SREG_NOVCC OPR_SSRC_SPECIAL_SCC s0 Scalar GPR 0. Increment from here for additional GPRs. There are NUM_SGPR SGPRs in total. Alignment restriction: unless otherwise noted, 64-bit SGPR values must be even-aligned (first SGPR address is even); 128-bit and wider SGPR values must be aligned to a multiple of 4. 0 s1 <p>See above.</p> 1 s2 <p>See above.</p> 2 s3 <p>See above.</p> 3 s4 <p>See above.</p> 4 s5 <p>See above.</p> 5 s6 <p>See above.</p> 6 s7 <p>See above.</p> 7 s8 <p>See above.</p> 8 s9 <p>See above.</p> 9 s10 <p>See above.</p> 10 s11 <p>See above.</p> 11 s12 <p>See above.</p> 12 s13 <p>See above.</p> 13 s14 <p>See above.</p> 14 s15 <p>See above.</p> 15 s16 <p>See above.</p> 16 s17 <p>See above.</p> 17 s18 <p>See above.</p> 18 s19 <p>See above.</p> 19 s20 <p>See above.</p> 20 s21 <p>See above.</p> 21 s22 <p>See above.</p> 22 s23 <p>See above.</p> 23 s24 <p>See above.</p> 24 s25 <p>See above.</p> 25 s26 <p>See above.</p> 26 s27 <p>See above.</p> 27 s28 <p>See above.</p> 28 s29 <p>See above.</p> 29 s30 <p>See above.</p> 30 s31 <p>See above.</p> 31 s32 <p>See above.</p> 32 s33 <p>See above.</p> 33 s34 <p>See above.</p> 34 s35 <p>See above.</p> 35 s36 <p>See above.</p> 36 s37 <p>See above.</p> 37 s38 <p>See above.</p> 38 s39 <p>See above.</p> 39 s40 <p>See above.</p> 40 s41 <p>See above.</p> 41 s42 <p>See above.</p> 42 s43 <p>See above.</p> 43 s44 <p>See above.</p> 44 s45 <p>See above.</p> 45 s46 <p>See above.</p> 46 s47 <p>See above.</p> 47 s48 <p>See above.</p> 48 s49 <p>See above.</p> 49 s50 <p>See above.</p> 50 s51 <p>See above.</p> 51 s52 <p>See above.</p> 52 s53 <p>See above.</p> 53 s54 <p>See above.</p> 54 s55 <p>See above.</p> 55 s56 <p>See above.</p> 56 s57 <p>See above.</p> 57 s58 <p>See above.</p> 58 s59 <p>See above.</p> 59 s60 <p>See above.</p> 60 s61 <p>See above.</p> 61 s62 <p>See above.</p> 62 s63 <p>See above.</p> 63 s64 <p>See above.</p> 64 s65 <p>See above.</p> 65 s66 <p>See above.</p> 66 s67 <p>See above.</p> 67 s68 <p>See above.</p> 68 s69 <p>See above.</p> 69 s70 <p>See above.</p> 70 s71 <p>See above.</p> 71 s72 <p>See above.</p> 72 s73 <p>See above.</p> 73 s74 <p>See above.</p> 74 s75 <p>See above.</p> 75 s76 <p>See above.</p> 76 s77 <p>See above.</p> 77 s78 <p>See above.</p> 78 s79 <p>See above.</p> 79 s80 <p>See above.</p> 80 s81 <p>See above.</p> 81 s82 <p>See above.</p> 82 s83 <p>See above.</p> 83 s84 <p>See above.</p> 84 s85 <p>See above.</p> 85 s86 <p>See above.</p> 86 s87 <p>See above.</p> 87 s88 <p>See above.</p> 88 s89 <p>See above.</p> 89 s90 <p>See above.</p> 90 s91 <p>See above.</p> 91 s92 <p>See above.</p> 92 s93 <p>See above.</p> 93 s94 <p>See above.</p> 94 s95 <p>See above.</p> 95 s96 <p>See above.</p> 96 s97 <p>See above.</p> 97 s98 <p>See above.</p> 98 s99 <p>See above.</p> 99 s100 <p>See above.</p> 100 s101 <p>See above.</p> 101 XNACK_MASK_LO xnack_mask[31:0], see XNACK replay mechanism for details. 104 XNACK_MASK_HI xnack_mask[63:32], see XNACK replay mechanism for details. 105 FLAT_SCRATCH_LO {13'd0, size[18:0]} 102 FLAT_SCRATCH_HI {8'd0, offset[31:8]} 103 TTMP0 Trap handler temps (privileged). Increment from here for additional TTMPs. There are NUM_TTMP TTMPs in total. {TTMP1,TTMP0} = PC_save{hi,lo}. Alignment restriction: unless otherwise noted, 64-bit TTMP values must be even-aligned (first TTMP address is even); 128-bit and wider TTMP values must be aligned to a multiple of 4. 108 TTMP1 Trap handler temps (privileged). 109 TTMP2 Trap handler temps (privileged). 110 TTMP3 Trap handler temps (privileged). 111 TTMP4 Trap handler temps (privileged). 112 TTMP5 Trap handler temps (privileged). 113 TTMP6 Trap handler temps (privileged). 114 TTMP7 Trap handler temps (privileged). 115 TTMP8 Trap handler temps (privileged). 116 TTMP9 Trap handler temps (privileged). 117 TTMP10 Trap handler temps (privileged). 118 TTMP11 Trap handler temps (privileged). 119 TTMP12 Trap handler temps (privileged). 120 TTMP13 Trap handler temps (privileged). 121 TTMP14 Trap handler temps (privileged). 122 TTMP15 Trap handler temps (privileged). 123 VCC_LO VCC[31:0] 106 VCC_HI VCC[63:32] 107 M0 Special register used to hold LDS/GDS addresses, relative indices, and send-message values. 124 EXEC_LO exec[31:0] 126 EXEC_HI exec[63:32] 127 0 0 128 1 1 (integer) 129 2 2 (integer) 130 3 3 (integer) 131 4 4 (integer) 132 5 5 (integer) 133 6 6 (integer) 134 7 7 (integer) 135 8 8 (integer) 136 9 9 (integer) 137 10 10 (integer) 138 11 11 (integer) 139 12 12 (integer) 140 13 13 (integer) 141 14 14 (integer) 142 15 15 (integer) 143 16 16 (integer) 144 17 17 (integer) 145 18 18 (integer) 146 19 19 (integer) 147 20 20 (integer) 148 21 21 (integer) 149 22 22 (integer) 150 23 23 (integer) 151 24 24 (integer) 152 25 25 (integer) 153 26 26 (integer) 154 27 27 (integer) 155 28 28 (integer) 156 29 29 (integer) 157 30 30 (integer) 158 31 31 (integer) 159 32 32 (integer) 160 33 33 (integer) 161 34 34 (integer) 162 35 35 (integer) 163 36 36 (integer) 164 37 37 (integer) 165 38 38 (integer) 166 39 39 (integer) 167 40 40 (integer) 168 41 41 (integer) 169 42 42 (integer) 170 43 43 (integer) 171 44 44 (integer) 172 45 45 (integer) 173 46 46 (integer) 174 47 47 (integer) 175 48 48 (integer) 176 49 49 (integer) 177 50 50 (integer) 178 51 51 (integer) 179 52 52 (integer) 180 53 53 (integer) 181 54 54 (integer) 182 55 55 (integer) 183 56 56 (integer) 184 57 57 (integer) 185 58 58 (integer) 186 59 59 (integer) 187 60 60 (integer) 188 61 61 (integer) 189 62 62 (integer) 190 63 63 (integer) 191 64 64 (integer) 192 -1 -1 (integer) 193 -2 -2 (integer) 194 -3 -3 (integer) 195 -4 -4 (integer) 196 -5 -5 (integer) 197 -6 -6 (integer) 198 -7 -7 (integer) 199 -8 -8 (integer) 200 -9 -9 (integer) 201 -10 -10 (integer) 202 -11 -11 (integer) 203 -12 -12 (integer) 204 -13 -13 (integer) 205 -14 -14 (integer) 206 -15 -15 (integer) 207 -16 -16 (integer) 208 0.5 0.5 240 -0.5 -0.5 241 1.0 1.0 242 -1.0 -1.0 243 2.0 2.0 244 -2.0 -2.0 245 4.0 4.0 246 -4.0 -4.0 247 0.15915494 1 / (2 * PI). 16-bit: 0x3118, 32-bit: 0x3e22f983, 64-bit: 0x3fc45f30_6dc9c882. 248 SRC_VCCZ True if vector condition code register is zero. 251 SRC_EXECZ True if execute mask is zero. 252 SRC_SCC Scalar condition code. 253 SRC_SHARED_BASE N/A 235 SRC_SHARED_LIMIT N/A 236 SRC_PRIVATE_BASE N/A 237 SRC_PRIVATE_LIMIT N/A 238 SRC_POPS_EXITING_WAVE_ID POPS counter (12b). Returns the ID of the most recent wave to complete in-order. 239 OPR_SSRC_SPECIAL_SCC N/A SRC_SCC Scalar condition code. 253 OPR_VCC N/A vcc VCC[63:0] 0 OPR_VGPR N/A v0 Vector GPR 0. Increment from here for additional GPRs. There are NUM_VGPR VGPRs in total. Alignment restriction: unless otherwise noted, 64-bit and wider VGPR values must be even-aligned (first VGPR address is even). 0 v1 <p>See above.</p> 1 v2 <p>See above.</p> 2 v3 <p>See above.</p> 3 v4 <p>See above.</p> 4 v5 <p>See above.</p> 5 v6 <p>See above.</p> 6 v7 <p>See above.</p> 7 v8 <p>See above.</p> 8 v9 <p>See above.</p> 9 v10 <p>See above.</p> 10 v11 <p>See above.</p> 11 v12 <p>See above.</p> 12 v13 <p>See above.</p> 13 v14 <p>See above.</p> 14 v15 <p>See above.</p> 15 v16 <p>See above.</p> 16 v17 <p>See above.</p> 17 v18 <p>See above.</p> 18 v19 <p>See above.</p> 19 v20 <p>See above.</p> 20 v21 <p>See above.</p> 21 v22 <p>See above.</p> 22 v23 <p>See above.</p> 23 v24 <p>See above.</p> 24 v25 <p>See above.</p> 25 v26 <p>See above.</p> 26 v27 <p>See above.</p> 27 v28 <p>See above.</p> 28 v29 <p>See above.</p> 29 v30 <p>See above.</p> 30 v31 <p>See above.</p> 31 v32 <p>See above.</p> 32 v33 <p>See above.</p> 33 v34 <p>See above.</p> 34 v35 <p>See above.</p> 35 v36 <p>See above.</p> 36 v37 <p>See above.</p> 37 v38 <p>See above.</p> 38 v39 <p>See above.</p> 39 v40 <p>See above.</p> 40 v41 <p>See above.</p> 41 v42 <p>See above.</p> 42 v43 <p>See above.</p> 43 v44 <p>See above.</p> 44 v45 <p>See above.</p> 45 v46 <p>See above.</p> 46 v47 <p>See above.</p> 47 v48 <p>See above.</p> 48 v49 <p>See above.</p> 49 v50 <p>See above.</p> 50 v51 <p>See above.</p> 51 v52 <p>See above.</p> 52 v53 <p>See above.</p> 53 v54 <p>See above.</p> 54 v55 <p>See above.</p> 55 v56 <p>See above.</p> 56 v57 <p>See above.</p> 57 v58 <p>See above.</p> 58 v59 <p>See above.</p> 59 v60 <p>See above.</p> 60 v61 <p>See above.</p> 61 v62 <p>See above.</p> 62 v63 <p>See above.</p> 63 v64 <p>See above.</p> 64 v65 <p>See above.</p> 65 v66 <p>See above.</p> 66 v67 <p>See above.</p> 67 v68 <p>See above.</p> 68 v69 <p>See above.</p> 69 v70 <p>See above.</p> 70 v71 <p>See above.</p> 71 v72 <p>See above.</p> 72 v73 <p>See above.</p> 73 v74 <p>See above.</p> 74 v75 <p>See above.</p> 75 v76 <p>See above.</p> 76 v77 <p>See above.</p> 77 v78 <p>See above.</p> 78 v79 <p>See above.</p> 79 v80 <p>See above.</p> 80 v81 <p>See above.</p> 81 v82 <p>See above.</p> 82 v83 <p>See above.</p> 83 v84 <p>See above.</p> 84 v85 <p>See above.</p> 85 v86 <p>See above.</p> 86 v87 <p>See above.</p> 87 v88 <p>See above.</p> 88 v89 <p>See above.</p> 89 v90 <p>See above.</p> 90 v91 <p>See above.</p> 91 v92 <p>See above.</p> 92 v93 <p>See above.</p> 93 v94 <p>See above.</p> 94 v95 <p>See above.</p> 95 v96 <p>See above.</p> 96 v97 <p>See above.</p> 97 v98 <p>See above.</p> 98 v99 <p>See above.</p> 99 v100 <p>See above.</p> 100 v101 <p>See above.</p> 101 v102 <p>See above.</p> 102 v103 <p>See above.</p> 103 v104 <p>See above.</p> 104 v105 <p>See above.</p> 105 v106 <p>See above.</p> 106 v107 <p>See above.</p> 107 v108 <p>See above.</p> 108 v109 <p>See above.</p> 109 v110 <p>See above.</p> 110 v111 <p>See above.</p> 111 v112 <p>See above.</p> 112 v113 <p>See above.</p> 113 v114 <p>See above.</p> 114 v115 <p>See above.</p> 115 v116 <p>See above.</p> 116 v117 <p>See above.</p> 117 v118 <p>See above.</p> 118 v119 <p>See above.</p> 119 v120 <p>See above.</p> 120 v121 <p>See above.</p> 121 v122 <p>See above.</p> 122 v123 <p>See above.</p> 123 v124 <p>See above.</p> 124 v125 <p>See above.</p> 125 v126 <p>See above.</p> 126 v127 <p>See above.</p> 127 v128 <p>See above.</p> 128 v129 <p>See above.</p> 129 v130 <p>See above.</p> 130 v131 <p>See above.</p> 131 v132 <p>See above.</p> 132 v133 <p>See above.</p> 133 v134 <p>See above.</p> 134 v135 <p>See above.</p> 135 v136 <p>See above.</p> 136 v137 <p>See above.</p> 137 v138 <p>See above.</p> 138 v139 <p>See above.</p> 139 v140 <p>See above.</p> 140 v141 <p>See above.</p> 141 v142 <p>See above.</p> 142 v143 <p>See above.</p> 143 v144 <p>See above.</p> 144 v145 <p>See above.</p> 145 v146 <p>See above.</p> 146 v147 <p>See above.</p> 147 v148 <p>See above.</p> 148 v149 <p>See above.</p> 149 v150 <p>See above.</p> 150 v151 <p>See above.</p> 151 v152 <p>See above.</p> 152 v153 <p>See above.</p> 153 v154 <p>See above.</p> 154 v155 <p>See above.</p> 155 v156 <p>See above.</p> 156 v157 <p>See above.</p> 157 v158 <p>See above.</p> 158 v159 <p>See above.</p> 159 v160 <p>See above.</p> 160 v161 <p>See above.</p> 161 v162 <p>See above.</p> 162 v163 <p>See above.</p> 163 v164 <p>See above.</p> 164 v165 <p>See above.</p> 165 v166 <p>See above.</p> 166 v167 <p>See above.</p> 167 v168 <p>See above.</p> 168 v169 <p>See above.</p> 169 v170 <p>See above.</p> 170 v171 <p>See above.</p> 171 v172 <p>See above.</p> 172 v173 <p>See above.</p> 173 v174 <p>See above.</p> 174 v175 <p>See above.</p> 175 v176 <p>See above.</p> 176 v177 <p>See above.</p> 177 v178 <p>See above.</p> 178 v179 <p>See above.</p> 179 v180 <p>See above.</p> 180 v181 <p>See above.</p> 181 v182 <p>See above.</p> 182 v183 <p>See above.</p> 183 v184 <p>See above.</p> 184 v185 <p>See above.</p> 185 v186 <p>See above.</p> 186 v187 <p>See above.</p> 187 v188 <p>See above.</p> 188 v189 <p>See above.</p> 189 v190 <p>See above.</p> 190 v191 <p>See above.</p> 191 v192 <p>See above.</p> 192 v193 <p>See above.</p> 193 v194 <p>See above.</p> 194 v195 <p>See above.</p> 195 v196 <p>See above.</p> 196 v197 <p>See above.</p> 197 v198 <p>See above.</p> 198 v199 <p>See above.</p> 199 v200 <p>See above.</p> 200 v201 <p>See above.</p> 201 v202 <p>See above.</p> 202 v203 <p>See above.</p> 203 v204 <p>See above.</p> 204 v205 <p>See above.</p> 205 v206 <p>See above.</p> 206 v207 <p>See above.</p> 207 v208 <p>See above.</p> 208 v209 <p>See above.</p> 209 v210 <p>See above.</p> 210 v211 <p>See above.</p> 211 v212 <p>See above.</p> 212 v213 <p>See above.</p> 213 v214 <p>See above.</p> 214 v215 <p>See above.</p> 215 v216 <p>See above.</p> 216 v217 <p>See above.</p> 217 v218 <p>See above.</p> 218 v219 <p>See above.</p> 219 v220 <p>See above.</p> 220 v221 <p>See above.</p> 221 v222 <p>See above.</p> 222 v223 <p>See above.</p> 223 v224 <p>See above.</p> 224 v225 <p>See above.</p> 225 v226 <p>See above.</p> 226 v227 <p>See above.</p> 227 v228 <p>See above.</p> 228 v229 <p>See above.</p> 229 v230 <p>See above.</p> 230 v231 <p>See above.</p> 231 v232 <p>See above.</p> 232 v233 <p>See above.</p> 233 v234 <p>See above.</p> 234 v235 <p>See above.</p> 235 v236 <p>See above.</p> 236 v237 <p>See above.</p> 237 v238 <p>See above.</p> 238 v239 <p>See above.</p> 239 v240 <p>See above.</p> 240 v241 <p>See above.</p> 241 v242 <p>See above.</p> 242 v243 <p>See above.</p> 243 v244 <p>See above.</p> 244 v245 <p>See above.</p> 245 v246 <p>See above.</p> 246 v247 <p>See above.</p> 247 v248 <p>See above.</p> 248 v249 <p>See above.</p> 249 v250 <p>See above.</p> 250 v251 <p>See above.</p> 251 v252 <p>See above.</p> 252 v253 <p>See above.</p> 253 v254 <p>See above.</p> 254 v255 <p>See above.</p> 255 OPR_VGPR_OR_ACCVGPR N/A OPR_ACCVGPR OPR_VGPR v0 Vector GPR 0. Increment from here for additional GPRs. There are NUM_VGPR VGPRs in total. Alignment restriction: unless otherwise noted, 64-bit and wider VGPR values must be even-aligned (first VGPR address is even). 0 v1 <p>See above.</p> 1 v2 <p>See above.</p> 2 v3 <p>See above.</p> 3 v4 <p>See above.</p> 4 v5 <p>See above.</p> 5 v6 <p>See above.</p> 6 v7 <p>See above.</p> 7 v8 <p>See above.</p> 8 v9 <p>See above.</p> 9 v10 <p>See above.</p> 10 v11 <p>See above.</p> 11 v12 <p>See above.</p> 12 v13 <p>See above.</p> 13 v14 <p>See above.</p> 14 v15 <p>See above.</p> 15 v16 <p>See above.</p> 16 v17 <p>See above.</p> 17 v18 <p>See above.</p> 18 v19 <p>See above.</p> 19 v20 <p>See above.</p> 20 v21 <p>See above.</p> 21 v22 <p>See above.</p> 22 v23 <p>See above.</p> 23 v24 <p>See above.</p> 24 v25 <p>See above.</p> 25 v26 <p>See above.</p> 26 v27 <p>See above.</p> 27 v28 <p>See above.</p> 28 v29 <p>See above.</p> 29 v30 <p>See above.</p> 30 v31 <p>See above.</p> 31 v32 <p>See above.</p> 32 v33 <p>See above.</p> 33 v34 <p>See above.</p> 34 v35 <p>See above.</p> 35 v36 <p>See above.</p> 36 v37 <p>See above.</p> 37 v38 <p>See above.</p> 38 v39 <p>See above.</p> 39 v40 <p>See above.</p> 40 v41 <p>See above.</p> 41 v42 <p>See above.</p> 42 v43 <p>See above.</p> 43 v44 <p>See above.</p> 44 v45 <p>See above.</p> 45 v46 <p>See above.</p> 46 v47 <p>See above.</p> 47 v48 <p>See above.</p> 48 v49 <p>See above.</p> 49 v50 <p>See above.</p> 50 v51 <p>See above.</p> 51 v52 <p>See above.</p> 52 v53 <p>See above.</p> 53 v54 <p>See above.</p> 54 v55 <p>See above.</p> 55 v56 <p>See above.</p> 56 v57 <p>See above.</p> 57 v58 <p>See above.</p> 58 v59 <p>See above.</p> 59 v60 <p>See above.</p> 60 v61 <p>See above.</p> 61 v62 <p>See above.</p> 62 v63 <p>See above.</p> 63 v64 <p>See above.</p> 64 v65 <p>See above.</p> 65 v66 <p>See above.</p> 66 v67 <p>See above.</p> 67 v68 <p>See above.</p> 68 v69 <p>See above.</p> 69 v70 <p>See above.</p> 70 v71 <p>See above.</p> 71 v72 <p>See above.</p> 72 v73 <p>See above.</p> 73 v74 <p>See above.</p> 74 v75 <p>See above.</p> 75 v76 <p>See above.</p> 76 v77 <p>See above.</p> 77 v78 <p>See above.</p> 78 v79 <p>See above.</p> 79 v80 <p>See above.</p> 80 v81 <p>See above.</p> 81 v82 <p>See above.</p> 82 v83 <p>See above.</p> 83 v84 <p>See above.</p> 84 v85 <p>See above.</p> 85 v86 <p>See above.</p> 86 v87 <p>See above.</p> 87 v88 <p>See above.</p> 88 v89 <p>See above.</p> 89 v90 <p>See above.</p> 90 v91 <p>See above.</p> 91 v92 <p>See above.</p> 92 v93 <p>See above.</p> 93 v94 <p>See above.</p> 94 v95 <p>See above.</p> 95 v96 <p>See above.</p> 96 v97 <p>See above.</p> 97 v98 <p>See above.</p> 98 v99 <p>See above.</p> 99 v100 <p>See above.</p> 100 v101 <p>See above.</p> 101 v102 <p>See above.</p> 102 v103 <p>See above.</p> 103 v104 <p>See above.</p> 104 v105 <p>See above.</p> 105 v106 <p>See above.</p> 106 v107 <p>See above.</p> 107 v108 <p>See above.</p> 108 v109 <p>See above.</p> 109 v110 <p>See above.</p> 110 v111 <p>See above.</p> 111 v112 <p>See above.</p> 112 v113 <p>See above.</p> 113 v114 <p>See above.</p> 114 v115 <p>See above.</p> 115 v116 <p>See above.</p> 116 v117 <p>See above.</p> 117 v118 <p>See above.</p> 118 v119 <p>See above.</p> 119 v120 <p>See above.</p> 120 v121 <p>See above.</p> 121 v122 <p>See above.</p> 122 v123 <p>See above.</p> 123 v124 <p>See above.</p> 124 v125 <p>See above.</p> 125 v126 <p>See above.</p> 126 v127 <p>See above.</p> 127 v128 <p>See above.</p> 128 v129 <p>See above.</p> 129 v130 <p>See above.</p> 130 v131 <p>See above.</p> 131 v132 <p>See above.</p> 132 v133 <p>See above.</p> 133 v134 <p>See above.</p> 134 v135 <p>See above.</p> 135 v136 <p>See above.</p> 136 v137 <p>See above.</p> 137 v138 <p>See above.</p> 138 v139 <p>See above.</p> 139 v140 <p>See above.</p> 140 v141 <p>See above.</p> 141 v142 <p>See above.</p> 142 v143 <p>See above.</p> 143 v144 <p>See above.</p> 144 v145 <p>See above.</p> 145 v146 <p>See above.</p> 146 v147 <p>See above.</p> 147 v148 <p>See above.</p> 148 v149 <p>See above.</p> 149 v150 <p>See above.</p> 150 v151 <p>See above.</p> 151 v152 <p>See above.</p> 152 v153 <p>See above.</p> 153 v154 <p>See above.</p> 154 v155 <p>See above.</p> 155 v156 <p>See above.</p> 156 v157 <p>See above.</p> 157 v158 <p>See above.</p> 158 v159 <p>See above.</p> 159 v160 <p>See above.</p> 160 v161 <p>See above.</p> 161 v162 <p>See above.</p> 162 v163 <p>See above.</p> 163 v164 <p>See above.</p> 164 v165 <p>See above.</p> 165 v166 <p>See above.</p> 166 v167 <p>See above.</p> 167 v168 <p>See above.</p> 168 v169 <p>See above.</p> 169 v170 <p>See above.</p> 170 v171 <p>See above.</p> 171 v172 <p>See above.</p> 172 v173 <p>See above.</p> 173 v174 <p>See above.</p> 174 v175 <p>See above.</p> 175 v176 <p>See above.</p> 176 v177 <p>See above.</p> 177 v178 <p>See above.</p> 178 v179 <p>See above.</p> 179 v180 <p>See above.</p> 180 v181 <p>See above.</p> 181 v182 <p>See above.</p> 182 v183 <p>See above.</p> 183 v184 <p>See above.</p> 184 v185 <p>See above.</p> 185 v186 <p>See above.</p> 186 v187 <p>See above.</p> 187 v188 <p>See above.</p> 188 v189 <p>See above.</p> 189 v190 <p>See above.</p> 190 v191 <p>See above.</p> 191 v192 <p>See above.</p> 192 v193 <p>See above.</p> 193 v194 <p>See above.</p> 194 v195 <p>See above.</p> 195 v196 <p>See above.</p> 196 v197 <p>See above.</p> 197 v198 <p>See above.</p> 198 v199 <p>See above.</p> 199 v200 <p>See above.</p> 200 v201 <p>See above.</p> 201 v202 <p>See above.</p> 202 v203 <p>See above.</p> 203 v204 <p>See above.</p> 204 v205 <p>See above.</p> 205 v206 <p>See above.</p> 206 v207 <p>See above.</p> 207 v208 <p>See above.</p> 208 v209 <p>See above.</p> 209 v210 <p>See above.</p> 210 v211 <p>See above.</p> 211 v212 <p>See above.</p> 212 v213 <p>See above.</p> 213 v214 <p>See above.</p> 214 v215 <p>See above.</p> 215 v216 <p>See above.</p> 216 v217 <p>See above.</p> 217 v218 <p>See above.</p> 218 v219 <p>See above.</p> 219 v220 <p>See above.</p> 220 v221 <p>See above.</p> 221 v222 <p>See above.</p> 222 v223 <p>See above.</p> 223 v224 <p>See above.</p> 224 v225 <p>See above.</p> 225 v226 <p>See above.</p> 226 v227 <p>See above.</p> 227 v228 <p>See above.</p> 228 v229 <p>See above.</p> 229 v230 <p>See above.</p> 230 v231 <p>See above.</p> 231 v232 <p>See above.</p> 232 v233 <p>See above.</p> 233 v234 <p>See above.</p> 234 v235 <p>See above.</p> 235 v236 <p>See above.</p> 236 v237 <p>See above.</p> 237 v238 <p>See above.</p> 238 v239 <p>See above.</p> 239 v240 <p>See above.</p> 240 v241 <p>See above.</p> 241 v242 <p>See above.</p> 242 v243 <p>See above.</p> 243 v244 <p>See above.</p> 244 v245 <p>See above.</p> 245 v246 <p>See above.</p> 246 v247 <p>See above.</p> 247 v248 <p>See above.</p> 248 v249 <p>See above.</p> 249 v250 <p>See above.</p> 250 v251 <p>See above.</p> 251 v252 <p>See above.</p> 252 v253 <p>See above.</p> 253 v254 <p>See above.</p> 254 v255 <p>See above.</p> 255 acc0 Accumulator VGPR 0. Increment from here for additional GPRs. There are NUM_ACCVGPR ACCVGPRs in total. Alignment restriction: unless otherwise noted, 64-bit and wider ACCVGPR values must be even-aligned (first ACCVGPR address is even). 512 acc1 <p>See above.</p> 513 acc2 <p>See above.</p> 514 acc3 <p>See above.</p> 515 acc4 <p>See above.</p> 516 acc5 <p>See above.</p> 517 acc6 <p>See above.</p> 518 acc7 <p>See above.</p> 519 acc8 <p>See above.</p> 520 acc9 <p>See above.</p> 521 acc10 <p>See above.</p> 522 acc11 <p>See above.</p> 523 acc12 <p>See above.</p> 524 acc13 <p>See above.</p> 525 acc14 <p>See above.</p> 526 acc15 <p>See above.</p> 527 acc16 <p>See above.</p> 528 acc17 <p>See above.</p> 529 acc18 <p>See above.</p> 530 acc19 <p>See above.</p> 531 acc20 <p>See above.</p> 532 acc21 <p>See above.</p> 533 acc22 <p>See above.</p> 534 acc23 <p>See above.</p> 535 acc24 <p>See above.</p> 536 acc25 <p>See above.</p> 537 acc26 <p>See above.</p> 538 acc27 <p>See above.</p> 539 acc28 <p>See above.</p> 540 acc29 <p>See above.</p> 541 acc30 <p>See above.</p> 542 acc31 <p>See above.</p> 543 acc32 <p>See above.</p> 544 acc33 <p>See above.</p> 545 acc34 <p>See above.</p> 546 acc35 <p>See above.</p> 547 acc36 <p>See above.</p> 548 acc37 <p>See above.</p> 549 acc38 <p>See above.</p> 550 acc39 <p>See above.</p> 551 acc40 <p>See above.</p> 552 acc41 <p>See above.</p> 553 acc42 <p>See above.</p> 554 acc43 <p>See above.</p> 555 acc44 <p>See above.</p> 556 acc45 <p>See above.</p> 557 acc46 <p>See above.</p> 558 acc47 <p>See above.</p> 559 acc48 <p>See above.</p> 560 acc49 <p>See above.</p> 561 acc50 <p>See above.</p> 562 acc51 <p>See above.</p> 563 acc52 <p>See above.</p> 564 acc53 <p>See above.</p> 565 acc54 <p>See above.</p> 566 acc55 <p>See above.</p> 567 acc56 <p>See above.</p> 568 acc57 <p>See above.</p> 569 acc58 <p>See above.</p> 570 acc59 <p>See above.</p> 571 acc60 <p>See above.</p> 572 acc61 <p>See above.</p> 573 acc62 <p>See above.</p> 574 acc63 <p>See above.</p> 575 acc64 <p>See above.</p> 576 acc65 <p>See above.</p> 577 acc66 <p>See above.</p> 578 acc67 <p>See above.</p> 579 acc68 <p>See above.</p> 580 acc69 <p>See above.</p> 581 acc70 <p>See above.</p> 582 acc71 <p>See above.</p> 583 acc72 <p>See above.</p> 584 acc73 <p>See above.</p> 585 acc74 <p>See above.</p> 586 acc75 <p>See above.</p> 587 acc76 <p>See above.</p> 588 acc77 <p>See above.</p> 589 acc78 <p>See above.</p> 590 acc79 <p>See above.</p> 591 acc80 <p>See above.</p> 592 acc81 <p>See above.</p> 593 acc82 <p>See above.</p> 594 acc83 <p>See above.</p> 595 acc84 <p>See above.</p> 596 acc85 <p>See above.</p> 597 acc86 <p>See above.</p> 598 acc87 <p>See above.</p> 599 acc88 <p>See above.</p> 600 acc89 <p>See above.</p> 601 acc90 <p>See above.</p> 602 acc91 <p>See above.</p> 603 acc92 <p>See above.</p> 604 acc93 <p>See above.</p> 605 acc94 <p>See above.</p> 606 acc95 <p>See above.</p> 607 acc96 <p>See above.</p> 608 acc97 <p>See above.</p> 609 acc98 <p>See above.</p> 610 acc99 <p>See above.</p> 611 acc100 <p>See above.</p> 612 acc101 <p>See above.</p> 613 acc102 <p>See above.</p> 614 acc103 <p>See above.</p> 615 acc104 <p>See above.</p> 616 acc105 <p>See above.</p> 617 acc106 <p>See above.</p> 618 acc107 <p>See above.</p> 619 acc108 <p>See above.</p> 620 acc109 <p>See above.</p> 621 acc110 <p>See above.</p> 622 acc111 <p>See above.</p> 623 acc112 <p>See above.</p> 624 acc113 <p>See above.</p> 625 acc114 <p>See above.</p> 626 acc115 <p>See above.</p> 627 acc116 <p>See above.</p> 628 acc117 <p>See above.</p> 629 acc118 <p>See above.</p> 630 acc119 <p>See above.</p> 631 acc120 <p>See above.</p> 632 acc121 <p>See above.</p> 633 acc122 <p>See above.</p> 634 acc123 <p>See above.</p> 635 acc124 <p>See above.</p> 636 acc125 <p>See above.</p> 637 acc126 <p>See above.</p> 638 acc127 <p>See above.</p> 639 acc128 <p>See above.</p> 640 acc129 <p>See above.</p> 641 acc130 <p>See above.</p> 642 acc131 <p>See above.</p> 643 acc132 <p>See above.</p> 644 acc133 <p>See above.</p> 645 acc134 <p>See above.</p> 646 acc135 <p>See above.</p> 647 acc136 <p>See above.</p> 648 acc137 <p>See above.</p> 649 acc138 <p>See above.</p> 650 acc139 <p>See above.</p> 651 acc140 <p>See above.</p> 652 acc141 <p>See above.</p> 653 acc142 <p>See above.</p> 654 acc143 <p>See above.</p> 655 acc144 <p>See above.</p> 656 acc145 <p>See above.</p> 657 acc146 <p>See above.</p> 658 acc147 <p>See above.</p> 659 acc148 <p>See above.</p> 660 acc149 <p>See above.</p> 661 acc150 <p>See above.</p> 662 acc151 <p>See above.</p> 663 acc152 <p>See above.</p> 664 acc153 <p>See above.</p> 665 acc154 <p>See above.</p> 666 acc155 <p>See above.</p> 667 acc156 <p>See above.</p> 668 acc157 <p>See above.</p> 669 acc158 <p>See above.</p> 670 acc159 <p>See above.</p> 671 acc160 <p>See above.</p> 672 acc161 <p>See above.</p> 673 acc162 <p>See above.</p> 674 acc163 <p>See above.</p> 675 acc164 <p>See above.</p> 676 acc165 <p>See above.</p> 677 acc166 <p>See above.</p> 678 acc167 <p>See above.</p> 679 acc168 <p>See above.</p> 680 acc169 <p>See above.</p> 681 acc170 <p>See above.</p> 682 acc171 <p>See above.</p> 683 acc172 <p>See above.</p> 684 acc173 <p>See above.</p> 685 acc174 <p>See above.</p> 686 acc175 <p>See above.</p> 687 acc176 <p>See above.</p> 688 acc177 <p>See above.</p> 689 acc178 <p>See above.</p> 690 acc179 <p>See above.</p> 691 acc180 <p>See above.</p> 692 acc181 <p>See above.</p> 693 acc182 <p>See above.</p> 694 acc183 <p>See above.</p> 695 acc184 <p>See above.</p> 696 acc185 <p>See above.</p> 697 acc186 <p>See above.</p> 698 acc187 <p>See above.</p> 699 acc188 <p>See above.</p> 700 acc189 <p>See above.</p> 701 acc190 <p>See above.</p> 702 acc191 <p>See above.</p> 703 acc192 <p>See above.</p> 704 acc193 <p>See above.</p> 705 acc194 <p>See above.</p> 706 acc195 <p>See above.</p> 707 acc196 <p>See above.</p> 708 acc197 <p>See above.</p> 709 acc198 <p>See above.</p> 710 acc199 <p>See above.</p> 711 acc200 <p>See above.</p> 712 acc201 <p>See above.</p> 713 acc202 <p>See above.</p> 714 acc203 <p>See above.</p> 715 acc204 <p>See above.</p> 716 acc205 <p>See above.</p> 717 acc206 <p>See above.</p> 718 acc207 <p>See above.</p> 719 acc208 <p>See above.</p> 720 acc209 <p>See above.</p> 721 acc210 <p>See above.</p> 722 acc211 <p>See above.</p> 723 acc212 <p>See above.</p> 724 acc213 <p>See above.</p> 725 acc214 <p>See above.</p> 726 acc215 <p>See above.</p> 727 acc216 <p>See above.</p> 728 acc217 <p>See above.</p> 729 acc218 <p>See above.</p> 730 acc219 <p>See above.</p> 731 acc220 <p>See above.</p> 732 acc221 <p>See above.</p> 733 acc222 <p>See above.</p> 734 acc223 <p>See above.</p> 735 acc224 <p>See above.</p> 736 acc225 <p>See above.</p> 737 acc226 <p>See above.</p> 738 acc227 <p>See above.</p> 739 acc228 <p>See above.</p> 740 acc229 <p>See above.</p> 741 acc230 <p>See above.</p> 742 acc231 <p>See above.</p> 743 acc232 <p>See above.</p> 744 acc233 <p>See above.</p> 745 acc234 <p>See above.</p> 746 acc235 <p>See above.</p> 747 acc236 <p>See above.</p> 748 acc237 <p>See above.</p> 749 acc238 <p>See above.</p> 750 acc239 <p>See above.</p> 751 acc240 <p>See above.</p> 752 acc241 <p>See above.</p> 753 acc242 <p>See above.</p> 754 acc243 <p>See above.</p> 755 acc244 <p>See above.</p> 756 acc245 <p>See above.</p> 757 acc246 <p>See above.</p> 758 acc247 <p>See above.</p> 759 acc248 <p>See above.</p> 760 acc249 <p>See above.</p> 761 acc250 <p>See above.</p> 762 acc251 <p>See above.</p> 763 acc252 <p>See above.</p> 764 acc253 <p>See above.</p> 765 acc254 <p>See above.</p> 766 acc255 <p>See above.</p> 767 OPR_VGPR_OR_LDS N/A OPR_SRC_VGPR v0 Vector GPR 0. Increment from here for additional GPRs. There are NUM_VGPR VGPRs in total. Alignment restriction: unless otherwise noted, 64-bit and wider VGPR values must be even-aligned (first VGPR address is even). 256 v1 <p>See above.</p> 257 v2 <p>See above.</p> 258 v3 <p>See above.</p> 259 v4 <p>See above.</p> 260 v5 <p>See above.</p> 261 v6 <p>See above.</p> 262 v7 <p>See above.</p> 263 v8 <p>See above.</p> 264 v9 <p>See above.</p> 265 v10 <p>See above.</p> 266 v11 <p>See above.</p> 267 v12 <p>See above.</p> 268 v13 <p>See above.</p> 269 v14 <p>See above.</p> 270 v15 <p>See above.</p> 271 v16 <p>See above.</p> 272 v17 <p>See above.</p> 273 v18 <p>See above.</p> 274 v19 <p>See above.</p> 275 v20 <p>See above.</p> 276 v21 <p>See above.</p> 277 v22 <p>See above.</p> 278 v23 <p>See above.</p> 279 v24 <p>See above.</p> 280 v25 <p>See above.</p> 281 v26 <p>See above.</p> 282 v27 <p>See above.</p> 283 v28 <p>See above.</p> 284 v29 <p>See above.</p> 285 v30 <p>See above.</p> 286 v31 <p>See above.</p> 287 v32 <p>See above.</p> 288 v33 <p>See above.</p> 289 v34 <p>See above.</p> 290 v35 <p>See above.</p> 291 v36 <p>See above.</p> 292 v37 <p>See above.</p> 293 v38 <p>See above.</p> 294 v39 <p>See above.</p> 295 v40 <p>See above.</p> 296 v41 <p>See above.</p> 297 v42 <p>See above.</p> 298 v43 <p>See above.</p> 299 v44 <p>See above.</p> 300 v45 <p>See above.</p> 301 v46 <p>See above.</p> 302 v47 <p>See above.</p> 303 v48 <p>See above.</p> 304 v49 <p>See above.</p> 305 v50 <p>See above.</p> 306 v51 <p>See above.</p> 307 v52 <p>See above.</p> 308 v53 <p>See above.</p> 309 v54 <p>See above.</p> 310 v55 <p>See above.</p> 311 v56 <p>See above.</p> 312 v57 <p>See above.</p> 313 v58 <p>See above.</p> 314 v59 <p>See above.</p> 315 v60 <p>See above.</p> 316 v61 <p>See above.</p> 317 v62 <p>See above.</p> 318 v63 <p>See above.</p> 319 v64 <p>See above.</p> 320 v65 <p>See above.</p> 321 v66 <p>See above.</p> 322 v67 <p>See above.</p> 323 v68 <p>See above.</p> 324 v69 <p>See above.</p> 325 v70 <p>See above.</p> 326 v71 <p>See above.</p> 327 v72 <p>See above.</p> 328 v73 <p>See above.</p> 329 v74 <p>See above.</p> 330 v75 <p>See above.</p> 331 v76 <p>See above.</p> 332 v77 <p>See above.</p> 333 v78 <p>See above.</p> 334 v79 <p>See above.</p> 335 v80 <p>See above.</p> 336 v81 <p>See above.</p> 337 v82 <p>See above.</p> 338 v83 <p>See above.</p> 339 v84 <p>See above.</p> 340 v85 <p>See above.</p> 341 v86 <p>See above.</p> 342 v87 <p>See above.</p> 343 v88 <p>See above.</p> 344 v89 <p>See above.</p> 345 v90 <p>See above.</p> 346 v91 <p>See above.</p> 347 v92 <p>See above.</p> 348 v93 <p>See above.</p> 349 v94 <p>See above.</p> 350 v95 <p>See above.</p> 351 v96 <p>See above.</p> 352 v97 <p>See above.</p> 353 v98 <p>See above.</p> 354 v99 <p>See above.</p> 355 v100 <p>See above.</p> 356 v101 <p>See above.</p> 357 v102 <p>See above.</p> 358 v103 <p>See above.</p> 359 v104 <p>See above.</p> 360 v105 <p>See above.</p> 361 v106 <p>See above.</p> 362 v107 <p>See above.</p> 363 v108 <p>See above.</p> 364 v109 <p>See above.</p> 365 v110 <p>See above.</p> 366 v111 <p>See above.</p> 367 v112 <p>See above.</p> 368 v113 <p>See above.</p> 369 v114 <p>See above.</p> 370 v115 <p>See above.</p> 371 v116 <p>See above.</p> 372 v117 <p>See above.</p> 373 v118 <p>See above.</p> 374 v119 <p>See above.</p> 375 v120 <p>See above.</p> 376 v121 <p>See above.</p> 377 v122 <p>See above.</p> 378 v123 <p>See above.</p> 379 v124 <p>See above.</p> 380 v125 <p>See above.</p> 381 v126 <p>See above.</p> 382 v127 <p>See above.</p> 383 v128 <p>See above.</p> 384 v129 <p>See above.</p> 385 v130 <p>See above.</p> 386 v131 <p>See above.</p> 387 v132 <p>See above.</p> 388 v133 <p>See above.</p> 389 v134 <p>See above.</p> 390 v135 <p>See above.</p> 391 v136 <p>See above.</p> 392 v137 <p>See above.</p> 393 v138 <p>See above.</p> 394 v139 <p>See above.</p> 395 v140 <p>See above.</p> 396 v141 <p>See above.</p> 397 v142 <p>See above.</p> 398 v143 <p>See above.</p> 399 v144 <p>See above.</p> 400 v145 <p>See above.</p> 401 v146 <p>See above.</p> 402 v147 <p>See above.</p> 403 v148 <p>See above.</p> 404 v149 <p>See above.</p> 405 v150 <p>See above.</p> 406 v151 <p>See above.</p> 407 v152 <p>See above.</p> 408 v153 <p>See above.</p> 409 v154 <p>See above.</p> 410 v155 <p>See above.</p> 411 v156 <p>See above.</p> 412 v157 <p>See above.</p> 413 v158 <p>See above.</p> 414 v159 <p>See above.</p> 415 v160 <p>See above.</p> 416 v161 <p>See above.</p> 417 v162 <p>See above.</p> 418 v163 <p>See above.</p> 419 v164 <p>See above.</p> 420 v165 <p>See above.</p> 421 v166 <p>See above.</p> 422 v167 <p>See above.</p> 423 v168 <p>See above.</p> 424 v169 <p>See above.</p> 425 v170 <p>See above.</p> 426 v171 <p>See above.</p> 427 v172 <p>See above.</p> 428 v173 <p>See above.</p> 429 v174 <p>See above.</p> 430 v175 <p>See above.</p> 431 v176 <p>See above.</p> 432 v177 <p>See above.</p> 433 v178 <p>See above.</p> 434 v179 <p>See above.</p> 435 v180 <p>See above.</p> 436 v181 <p>See above.</p> 437 v182 <p>See above.</p> 438 v183 <p>See above.</p> 439 v184 <p>See above.</p> 440 v185 <p>See above.</p> 441 v186 <p>See above.</p> 442 v187 <p>See above.</p> 443 v188 <p>See above.</p> 444 v189 <p>See above.</p> 445 v190 <p>See above.</p> 446 v191 <p>See above.</p> 447 v192 <p>See above.</p> 448 v193 <p>See above.</p> 449 v194 <p>See above.</p> 450 v195 <p>See above.</p> 451 v196 <p>See above.</p> 452 v197 <p>See above.</p> 453 v198 <p>See above.</p> 454 v199 <p>See above.</p> 455 v200 <p>See above.</p> 456 v201 <p>See above.</p> 457 v202 <p>See above.</p> 458 v203 <p>See above.</p> 459 v204 <p>See above.</p> 460 v205 <p>See above.</p> 461 v206 <p>See above.</p> 462 v207 <p>See above.</p> 463 v208 <p>See above.</p> 464 v209 <p>See above.</p> 465 v210 <p>See above.</p> 466 v211 <p>See above.</p> 467 v212 <p>See above.</p> 468 v213 <p>See above.</p> 469 v214 <p>See above.</p> 470 v215 <p>See above.</p> 471 v216 <p>See above.</p> 472 v217 <p>See above.</p> 473 v218 <p>See above.</p> 474 v219 <p>See above.</p> 475 v220 <p>See above.</p> 476 v221 <p>See above.</p> 477 v222 <p>See above.</p> 478 v223 <p>See above.</p> 479 v224 <p>See above.</p> 480 v225 <p>See above.</p> 481 v226 <p>See above.</p> 482 v227 <p>See above.</p> 483 v228 <p>See above.</p> 484 v229 <p>See above.</p> 485 v230 <p>See above.</p> 486 v231 <p>See above.</p> 487 v232 <p>See above.</p> 488 v233 <p>See above.</p> 489 v234 <p>See above.</p> 490 v235 <p>See above.</p> 491 v236 <p>See above.</p> 492 v237 <p>See above.</p> 493 v238 <p>See above.</p> 494 v239 <p>See above.</p> 495 v240 <p>See above.</p> 496 v241 <p>See above.</p> 497 v242 <p>See above.</p> 498 v243 <p>See above.</p> 499 v244 <p>See above.</p> 500 v245 <p>See above.</p> 501 v246 <p>See above.</p> 502 v247 <p>See above.</p> 503 v248 <p>See above.</p> 504 v249 <p>See above.</p> 505 v250 <p>See above.</p> 506 v251 <p>See above.</p> 507 v252 <p>See above.</p> 508 v253 <p>See above.</p> 509 v254 <p>See above.</p> 510 v255 <p>See above.</p> 511 OPR_WAITCNT A waitcnt operand specifying expcnt, lgkmcnt and vmcnt. EXP Export wait count. 0x7 means do not wait on EXPCNT. 3 4 LGKM LGKM wait count. 0xf means do not wait on LGKMCNT. 4 8 VM VM wait count. 0xf means do not wait on VMCNT. 4 0 2 14 SALU Scalar ALU (SALU) instructions operate on values that are common to all work-items in the wave. These operations consist of 32-bit integer or floating-point arithmetic, and 32 or 64-bit bitwise operations, and they store results in scalar registers. SMEM Scalar Memory (SMEM) instructions allow a shader program to load data from memory into scalar registers, making the data accessible to all work-items in the wave. VALU Vector ALU (VALU) instructions perform arithmetic or logical operations on data for each of the wave's work-item and write results back to vector registers, scalar registers or the EXEC mask. VMEM Vector-memory (VMEM) instructions transfer data to and from vector registers, making the data uniquely available for each work-item in the wave. EXPORT Export instructions copy data from a VGPR to one of the export buffers (position, color, or Z). BRANCH Branch instructions allow programs to jump - direct or indirect - to an instruction to be executed next, conditionally or unconditionally. MESSAGE Message instructions send a message to fixed function hardware or the host and can be used to request that a value be returned to the wave. WAVE_CONTROL Wave control instructions control a wave's execution these include dependency, delay, scheduling, and wave termination operations. TRAP Trap instructions transition into and out of the wave's trap handler FLOATING_POINT BUFFER TEXTURE LOAD STORE SAMPLE BVH ATOMIC FLAT DATA_SHARE STATIC MFMA WMMA TRANSCENDENTAL