CodeQL / Analyze (actions) (push) Canceled after 0s
CodeQL / Analyze (python) (push) Canceled after 0s
pre-commit / pre-commit (push) Canceled after 0s
Multi-Arch CI / setup (push) Canceled after 0s
Publish build_manylinux_x86_64 images / publish_build_manylinux_x86_64 (push) Canceled after 0s
Setup (Multi-Arch) / setup (push) Canceled after 0s
Multi-Arch CI / Linux::${{ fromJSON(needs.setup.outputs.linux_build_config || '{}').build_variant_label || 'skip' }} (push) Canceled after 0s
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Publish no_rocm_image_ubi10 image / publish_no_rocm_image_ubi10 (push) Canceled after 0s
Multi-Arch CI / Windows::${{ fromJSON(needs.setup.outputs.windows_build_config || '{}').build_variant_label || 'skip' }} (push) Canceled after 0s
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Multi-Arch CI / Manifest Diff (push) Canceled after 0s
TheRock Manifest Diff Report / Generate Manifest Diff Report (push) Canceled after 0s
Multi-Arch CI / CI Summary (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04 (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04_ocl_rt images / publish_no_rocm_image_ubuntu24_04_ocl_rt (push) Canceled after 0s
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04_rocgdb (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04_rocgdb images / publish_no_rocm_image_ubuntu24_04_rocgdb (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04_ocl_rt (push) Canceled after 0s
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Unit Tests / Unit Tests :: ubuntu-24.04 (push) Canceled after 0s
Unit Tests / Unit Tests :: windows-2022 (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04_media (push) Canceled after 0s
Unit Tests / Unit Tests Summary (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04_openmpi (push) Canceled after 0s
2041 lines
144 KiB
YAML
2041 lines
144 KiB
YAML
# NOTE: Assertions have been autogenerated by utils/update_mir_test_checks.py UTC_ARGS: --version 5
|
|
# RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=gfx942 -start-before=greedy,2 -stop-after=virtregrewriter,2 -verify-regalloc -verify-machineinstrs -o - %s | FileCheck %s
|
|
|
|
# There aren't enough VGPRs for %0 to stay in a VGPR. %0 should be
|
|
# split and inflated to AV_512. The VGPR version of the instruction
|
|
# should be replaced with the AGPR version, and the copies eliminated.
|
|
|
|
---
|
|
name: inflate_result_to_agpr__V_MFMA_F32_32X32X8F16_mac_vgprcd_e64
|
|
tracksRegLiveness: true
|
|
machineFunctionInfo:
|
|
isEntryFunction: true
|
|
stackPtrOffsetReg: '$sgpr32'
|
|
occupancy: 10
|
|
sgprForEXECCopy: '$sgpr100_sgpr101'
|
|
body: |
|
|
; CHECK-LABEL: name: inflate_result_to_agpr__V_MFMA_F32_32X32X8F16_mac_vgprcd_e64
|
|
; CHECK: bb.0:
|
|
; CHECK-NEXT: successors: %bb.1(0x80000000)
|
|
; CHECK-NEXT: {{ $}}
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $agpr0
|
|
; CHECK-NEXT: renamable $sgpr0 = S_MOV_B32 0
|
|
; CHECK-NEXT: renamable $vgpr8 = V_MOV_B32_e32 0, implicit $exec
|
|
; CHECK-NEXT: renamable $sgpr1 = COPY renamable $sgpr0
|
|
; CHECK-NEXT: renamable $vgpr0_vgpr1 = COPY killed renamable $sgpr0_sgpr1
|
|
; CHECK-NEXT: renamable $vcc = S_AND_B64 $exec, -1, implicit-def dead $scc
|
|
; CHECK-NEXT: dead renamable $vgpr9 = COPY renamable $vgpr8
|
|
; CHECK-NEXT: {{ $}}
|
|
; CHECK-NEXT: bb.1:
|
|
; CHECK-NEXT: successors: %bb.1(0x40000000), %bb.2(0x40000000)
|
|
; CHECK-NEXT: liveins: $vcc, $vgpr0_vgpr1
|
|
; CHECK-NEXT: {{ $}}
|
|
; CHECK-NEXT: renamable $agpr0_agpr1 = GLOBAL_LOAD_DWORDX2 undef renamable $vgpr0_vgpr1, 0, 0, implicit $exec :: (load (s64), addrspace 1)
|
|
; CHECK-NEXT: renamable $agpr0_agpr1_agpr2_agpr3_agpr4_agpr5_agpr6_agpr7_agpr8_agpr9_agpr10_agpr11_agpr12_agpr13_agpr14_agpr15 = V_MFMA_F32_32X32X8F16_mac_e64 $vgpr0_vgpr1, $vgpr0_vgpr1, $agpr0_agpr1_agpr2_agpr3_agpr4_agpr5_agpr6_agpr7_agpr8_agpr9_agpr10_agpr11_agpr12_agpr13_agpr14_agpr15, 0, 0, 0, implicit $mode, implicit $exec
|
|
; CHECK-NEXT: S_CBRANCH_VCCNZ %bb.1, implicit $vcc
|
|
; CHECK-NEXT: S_BRANCH %bb.2
|
|
; CHECK-NEXT: {{ $}}
|
|
; CHECK-NEXT: bb.2:
|
|
; CHECK-NEXT: liveins: $agpr0_agpr1_agpr2_agpr3_agpr4_agpr5_agpr6_agpr7_agpr8_agpr9_agpr10_agpr11_agpr12_agpr13_agpr14_agpr15:0x00000000FFFFFFFF
|
|
; CHECK-NEXT: {{ $}}
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr0_vgpr1_vgpr2_vgpr3_vgpr4_vgpr5_vgpr6_vgpr7
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr8_vgpr9_vgpr10_vgpr11_vgpr12_vgpr13_vgpr14_vgpr15
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr16_vgpr17_vgpr18_vgpr19_vgpr20_vgpr21_vgpr22_vgpr23
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr24_vgpr25_vgpr26_vgpr27_vgpr28_vgpr29_vgpr30_vgpr31
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr32_vgpr33_vgpr34_vgpr35_vgpr36_vgpr37_vgpr38_vgpr39
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr40_vgpr41_vgpr42_vgpr43_vgpr44_vgpr45_vgpr46_vgpr47
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr48_vgpr49_vgpr50_vgpr51_vgpr52_vgpr53_vgpr54_vgpr55
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr56_vgpr57_vgpr58_vgpr59_vgpr60_vgpr61_vgpr62_vgpr63
|
|
; CHECK-NEXT: renamable $vgpr0 = V_MOV_B32_e32 0, implicit $exec
|
|
; CHECK-NEXT: GLOBAL_STORE_DWORDX4_SADDR renamable $vgpr0, renamable $agpr8_agpr9_agpr10_agpr11, undef $sgpr0_sgpr1, 32, 0, implicit $exec :: (store (s128), align 32, addrspace 1)
|
|
; CHECK-NEXT: GLOBAL_STORE_DWORDX4_SADDR renamable $vgpr0, renamable $agpr12_agpr13_agpr14_agpr15, undef $sgpr0_sgpr1, 48, 0, implicit $exec :: (store (s128), addrspace 1)
|
|
; CHECK-NEXT: GLOBAL_STORE_DWORDX4_SADDR renamable $vgpr0, renamable $agpr0_agpr1_agpr2_agpr3, undef $sgpr0_sgpr1, 0, 0, implicit $exec :: (store (s128), align 128, addrspace 1)
|
|
; CHECK-NEXT: GLOBAL_STORE_DWORDX4_SADDR killed renamable $vgpr0, killed renamable $agpr4_agpr5_agpr6_agpr7, killed undef $sgpr0_sgpr1, 16, 0, implicit $exec :: (store (s128), addrspace 1)
|
|
; CHECK-NEXT: S_ENDPGM 0
|
|
bb.0:
|
|
S_NOP 0, implicit-def $agpr0
|
|
renamable $sgpr0 = S_MOV_B32 0
|
|
undef %0.sub8:vreg_512_align2 = V_MOV_B32_e32 0, implicit $exec
|
|
renamable $sgpr1 = COPY renamable $sgpr0
|
|
%1:vreg_64_align2 = COPY killed renamable $sgpr0_sgpr1
|
|
renamable $vcc = S_AND_B64 $exec, -1, implicit-def dead $scc
|
|
%0.sub9:vreg_512_align2 = COPY %0.sub8
|
|
|
|
bb.1:
|
|
liveins: $vcc
|
|
|
|
undef %0.sub0_sub1:vreg_512_align2 = GLOBAL_LOAD_DWORDX2 undef %3:vreg_64_align2, 0, 0, implicit $exec :: (load (s64), addrspace 1)
|
|
%0:vreg_512_align2 = V_MFMA_F32_32X32X8F16_mac_vgprcd_e64 %1, %1, %0, 0, 0, 0, implicit $mode, implicit $exec
|
|
S_CBRANCH_VCCNZ %bb.1, implicit $vcc
|
|
S_BRANCH %bb.2
|
|
|
|
bb.2:
|
|
; No VGPRs available for %0
|
|
S_NOP 0, implicit-def $vgpr0_vgpr1_vgpr2_vgpr3_vgpr4_vgpr5_vgpr6_vgpr7
|
|
S_NOP 0, implicit-def $vgpr8_vgpr9_vgpr10_vgpr11_vgpr12_vgpr13_vgpr14_vgpr15
|
|
S_NOP 0, implicit-def $vgpr16_vgpr17_vgpr18_vgpr19_vgpr20_vgpr21_vgpr22_vgpr23
|
|
S_NOP 0, implicit-def $vgpr24_vgpr25_vgpr26_vgpr27_vgpr28_vgpr29_vgpr30_vgpr31
|
|
S_NOP 0, implicit-def $vgpr32_vgpr33_vgpr34_vgpr35_vgpr36_vgpr37_vgpr38_vgpr39
|
|
S_NOP 0, implicit-def $vgpr40_vgpr41_vgpr42_vgpr43_vgpr44_vgpr45_vgpr46_vgpr47
|
|
S_NOP 0, implicit-def $vgpr48_vgpr49_vgpr50_vgpr51_vgpr52_vgpr53_vgpr54_vgpr55
|
|
S_NOP 0, implicit-def $vgpr56_vgpr57_vgpr58_vgpr59_vgpr60_vgpr61_vgpr62_vgpr63
|
|
%2:vgpr_32 = V_MOV_B32_e32 0, implicit $exec
|
|
GLOBAL_STORE_DWORDX4_SADDR %2, %0.sub8_sub9_sub10_sub11, undef $sgpr0_sgpr1, 32, 0, implicit $exec :: (store (s128), align 32, addrspace 1)
|
|
GLOBAL_STORE_DWORDX4_SADDR %2, %0.sub12_sub13_sub14_sub15, undef $sgpr0_sgpr1, 48, 0, implicit $exec :: (store (s128), addrspace 1)
|
|
GLOBAL_STORE_DWORDX4_SADDR %2, %0.sub0_sub1_sub2_sub3, undef $sgpr0_sgpr1, 0, 0, implicit $exec :: (store (s128), align 128, addrspace 1)
|
|
GLOBAL_STORE_DWORDX4_SADDR %2, %0.sub4_sub5_sub6_sub7, killed undef $sgpr0_sgpr1, 16, 0, implicit $exec :: (store (s128), addrspace 1)
|
|
S_ENDPGM 0
|
|
|
|
...
|
|
|
|
# Same, except spam some other uses of the register that we shouldn't
|
|
# break the liveness of
|
|
|
|
---
|
|
name: inflate_result_to_agpr__V_MFMA_F32_32X32X8F16_mac_vgprcd_e64_multi_load
|
|
tracksRegLiveness: true
|
|
machineFunctionInfo:
|
|
isEntryFunction: true
|
|
stackPtrOffsetReg: '$sgpr32'
|
|
occupancy: 10
|
|
sgprForEXECCopy: '$sgpr100_sgpr101'
|
|
body: |
|
|
; CHECK-LABEL: name: inflate_result_to_agpr__V_MFMA_F32_32X32X8F16_mac_vgprcd_e64_multi_load
|
|
; CHECK: bb.0:
|
|
; CHECK-NEXT: successors: %bb.1(0x80000000)
|
|
; CHECK-NEXT: {{ $}}
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $agpr0
|
|
; CHECK-NEXT: renamable $sgpr0 = S_MOV_B32 0
|
|
; CHECK-NEXT: renamable $vgpr8 = V_MOV_B32_e32 0, implicit $exec
|
|
; CHECK-NEXT: renamable $sgpr1 = COPY renamable $sgpr0
|
|
; CHECK-NEXT: renamable $vgpr0_vgpr1 = COPY killed renamable $sgpr0_sgpr1
|
|
; CHECK-NEXT: renamable $vcc = S_AND_B64 $exec, -1, implicit-def dead $scc
|
|
; CHECK-NEXT: dead renamable $vgpr9 = COPY renamable $vgpr8
|
|
; CHECK-NEXT: {{ $}}
|
|
; CHECK-NEXT: bb.1:
|
|
; CHECK-NEXT: successors: %bb.1(0x40000000), %bb.2(0x40000000)
|
|
; CHECK-NEXT: liveins: $vcc, $vgpr0_vgpr1
|
|
; CHECK-NEXT: {{ $}}
|
|
; CHECK-NEXT: renamable $agpr0_agpr1 = GLOBAL_LOAD_DWORDX2 undef renamable $vgpr0_vgpr1, 0, 0, implicit $exec :: (load (s64), addrspace 1)
|
|
; CHECK-NEXT: renamable $agpr2_agpr3 = GLOBAL_LOAD_DWORDX2 undef renamable $vgpr0_vgpr1, 8, 0, implicit $exec :: (load (s64), addrspace 1)
|
|
; CHECK-NEXT: renamable $agpr4_agpr5_agpr6 = GLOBAL_LOAD_DWORDX3 undef renamable $vgpr0_vgpr1, 16, 0, implicit $exec :: (load (s96), align 16, addrspace 1)
|
|
; CHECK-NEXT: renamable $agpr0_agpr1_agpr2_agpr3_agpr4_agpr5_agpr6_agpr7_agpr8_agpr9_agpr10_agpr11_agpr12_agpr13_agpr14_agpr15 = V_MFMA_F32_32X32X8F16_mac_e64 $vgpr0_vgpr1, $vgpr0_vgpr1, $agpr0_agpr1_agpr2_agpr3_agpr4_agpr5_agpr6_agpr7_agpr8_agpr9_agpr10_agpr11_agpr12_agpr13_agpr14_agpr15, 28, 0, 0, implicit $mode, implicit $exec
|
|
; CHECK-NEXT: S_NOP 0, implicit renamable $agpr0_agpr1_agpr2_agpr3_agpr4_agpr5_agpr6_agpr7_agpr8_agpr9_agpr10_agpr11_agpr12_agpr13_agpr14_agpr15
|
|
; CHECK-NEXT: GLOBAL_STORE_DWORDX4_SADDR undef renamable $vgpr0, renamable $agpr4_agpr5_agpr6_agpr7, killed undef $sgpr0_sgpr1, 16, 0, implicit $exec :: (store (s128), addrspace 1)
|
|
; CHECK-NEXT: S_CBRANCH_VCCNZ %bb.1, implicit $vcc
|
|
; CHECK-NEXT: S_BRANCH %bb.2
|
|
; CHECK-NEXT: {{ $}}
|
|
; CHECK-NEXT: bb.2:
|
|
; CHECK-NEXT: liveins: $agpr0_agpr1_agpr2_agpr3_agpr4_agpr5_agpr6_agpr7_agpr8_agpr9_agpr10_agpr11_agpr12_agpr13_agpr14_agpr15:0x00000000FFFFFFFF
|
|
; CHECK-NEXT: {{ $}}
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr0_vgpr1_vgpr2_vgpr3_vgpr4_vgpr5_vgpr6_vgpr7
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr8_vgpr9_vgpr10_vgpr11_vgpr12_vgpr13_vgpr14_vgpr15
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr16_vgpr17_vgpr18_vgpr19_vgpr20_vgpr21_vgpr22_vgpr23
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr24_vgpr25_vgpr26_vgpr27_vgpr28_vgpr29_vgpr30_vgpr31
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr32_vgpr33_vgpr34_vgpr35_vgpr36_vgpr37_vgpr38_vgpr39
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr40_vgpr41_vgpr42_vgpr43_vgpr44_vgpr45_vgpr46_vgpr47
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr48_vgpr49_vgpr50_vgpr51_vgpr52_vgpr53_vgpr54_vgpr55
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr56_vgpr57_vgpr58_vgpr59_vgpr60_vgpr61_vgpr62_vgpr63
|
|
; CHECK-NEXT: renamable $vgpr0 = V_MOV_B32_e32 0, implicit $exec
|
|
; CHECK-NEXT: GLOBAL_STORE_DWORDX4_SADDR renamable $vgpr0, renamable $agpr8_agpr9_agpr10_agpr11, undef $sgpr0_sgpr1, 32, 0, implicit $exec :: (store (s128), align 32, addrspace 1)
|
|
; CHECK-NEXT: GLOBAL_STORE_DWORDX4_SADDR renamable $vgpr0, renamable $agpr12_agpr13_agpr14_agpr15, undef $sgpr0_sgpr1, 48, 0, implicit $exec :: (store (s128), addrspace 1)
|
|
; CHECK-NEXT: GLOBAL_STORE_DWORDX4_SADDR renamable $vgpr0, renamable $agpr0_agpr1_agpr2_agpr3, undef $sgpr0_sgpr1, 0, 0, implicit $exec :: (store (s128), align 128, addrspace 1)
|
|
; CHECK-NEXT: GLOBAL_STORE_DWORDX4_SADDR killed renamable $vgpr0, killed renamable $agpr4_agpr5_agpr6_agpr7, killed undef $sgpr0_sgpr1, 16, 0, implicit $exec :: (store (s128), addrspace 1)
|
|
; CHECK-NEXT: S_ENDPGM 0
|
|
bb.0:
|
|
S_NOP 0, implicit-def $agpr0
|
|
renamable $sgpr0 = S_MOV_B32 0
|
|
undef %0.sub8:vreg_512_align2 = V_MOV_B32_e32 0, implicit $exec
|
|
renamable $sgpr1 = COPY renamable $sgpr0
|
|
%1:vreg_64_align2 = COPY killed renamable $sgpr0_sgpr1
|
|
renamable $vcc = S_AND_B64 $exec, -1, implicit-def dead $scc
|
|
%0.sub9:vreg_512_align2 = COPY %0.sub8
|
|
|
|
bb.1:
|
|
liveins: $vcc
|
|
|
|
undef %0.sub0_sub1:vreg_512_align2 = GLOBAL_LOAD_DWORDX2 undef %3:vreg_64_align2, 0, 0, implicit $exec :: (load (s64), addrspace 1)
|
|
%0.sub2_sub3:vreg_512_align2 = GLOBAL_LOAD_DWORDX2 undef %3:vreg_64_align2, 8, 0, implicit $exec :: (load (s64), addrspace 1)
|
|
%0.sub4_sub5_sub6:vreg_512_align2 = GLOBAL_LOAD_DWORDX3 undef %3:vreg_64_align2, 16, 0, implicit $exec :: (load (s96), addrspace 1)
|
|
%0:vreg_512_align2 = V_MFMA_F32_32X32X8F16_mac_vgprcd_e64 %1, %1, %0, 28, 0, 0, implicit $mode, implicit $exec
|
|
S_NOP 0, implicit %0
|
|
GLOBAL_STORE_DWORDX4_SADDR undef %4:vgpr_32, %0.sub4_sub5_sub6_sub7, killed undef $sgpr0_sgpr1, 16, 0, implicit $exec :: (store (s128), addrspace 1)
|
|
S_CBRANCH_VCCNZ %bb.1, implicit $vcc
|
|
S_BRANCH %bb.2
|
|
|
|
bb.2:
|
|
; No VGPRs available for %0
|
|
S_NOP 0, implicit-def $vgpr0_vgpr1_vgpr2_vgpr3_vgpr4_vgpr5_vgpr6_vgpr7
|
|
S_NOP 0, implicit-def $vgpr8_vgpr9_vgpr10_vgpr11_vgpr12_vgpr13_vgpr14_vgpr15
|
|
S_NOP 0, implicit-def $vgpr16_vgpr17_vgpr18_vgpr19_vgpr20_vgpr21_vgpr22_vgpr23
|
|
S_NOP 0, implicit-def $vgpr24_vgpr25_vgpr26_vgpr27_vgpr28_vgpr29_vgpr30_vgpr31
|
|
S_NOP 0, implicit-def $vgpr32_vgpr33_vgpr34_vgpr35_vgpr36_vgpr37_vgpr38_vgpr39
|
|
S_NOP 0, implicit-def $vgpr40_vgpr41_vgpr42_vgpr43_vgpr44_vgpr45_vgpr46_vgpr47
|
|
S_NOP 0, implicit-def $vgpr48_vgpr49_vgpr50_vgpr51_vgpr52_vgpr53_vgpr54_vgpr55
|
|
S_NOP 0, implicit-def $vgpr56_vgpr57_vgpr58_vgpr59_vgpr60_vgpr61_vgpr62_vgpr63
|
|
%2:vgpr_32 = V_MOV_B32_e32 0, implicit $exec
|
|
GLOBAL_STORE_DWORDX4_SADDR %2, %0.sub8_sub9_sub10_sub11, undef $sgpr0_sgpr1, 32, 0, implicit $exec :: (store (s128), align 32, addrspace 1)
|
|
GLOBAL_STORE_DWORDX4_SADDR %2, %0.sub12_sub13_sub14_sub15, undef $sgpr0_sgpr1, 48, 0, implicit $exec :: (store (s128), addrspace 1)
|
|
GLOBAL_STORE_DWORDX4_SADDR %2, %0.sub0_sub1_sub2_sub3, undef $sgpr0_sgpr1, 0, 0, implicit $exec :: (store (s128), align 128, addrspace 1)
|
|
GLOBAL_STORE_DWORDX4_SADDR %2, %0.sub4_sub5_sub6_sub7, killed undef $sgpr0_sgpr1, 16, 0, implicit $exec :: (store (s128), addrspace 1)
|
|
S_ENDPGM 0
|
|
|
|
...
|
|
|
|
# Non-mac variant, src2 is other VGPR. We cannot replace this case
|
|
# unless we also try to reassign src2 to an AGPR.
|
|
---
|
|
name: inflate_result_to_agpr__V_MFMA_F32_32X32X8F16_vgprcd_e64
|
|
tracksRegLiveness: true
|
|
machineFunctionInfo:
|
|
isEntryFunction: true
|
|
stackPtrOffsetReg: '$sgpr32'
|
|
occupancy: 10
|
|
sgprForEXECCopy: '$sgpr100_sgpr101'
|
|
body: |
|
|
; CHECK-LABEL: name: inflate_result_to_agpr__V_MFMA_F32_32X32X8F16_vgprcd_e64
|
|
; CHECK: bb.0:
|
|
; CHECK-NEXT: successors: %bb.1(0x80000000)
|
|
; CHECK-NEXT: {{ $}}
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $agpr0
|
|
; CHECK-NEXT: renamable $sgpr0 = S_MOV_B32 0
|
|
; CHECK-NEXT: renamable $vgpr8 = V_MOV_B32_e32 0, implicit $exec
|
|
; CHECK-NEXT: renamable $sgpr1 = COPY renamable $sgpr0
|
|
; CHECK-NEXT: renamable $vgpr2_vgpr3 = COPY killed renamable $sgpr0_sgpr1
|
|
; CHECK-NEXT: renamable $vcc = S_AND_B64 $exec, -1, implicit-def dead $scc
|
|
; CHECK-NEXT: dead renamable $vgpr9 = COPY renamable $vgpr8
|
|
; CHECK-NEXT: {{ $}}
|
|
; CHECK-NEXT: bb.1:
|
|
; CHECK-NEXT: successors: %bb.1(0x40000000), %bb.2(0x40000000)
|
|
; CHECK-NEXT: liveins: $vcc, $vgpr2_vgpr3
|
|
; CHECK-NEXT: {{ $}}
|
|
; CHECK-NEXT: renamable $agpr16_agpr17 = GLOBAL_LOAD_DWORDX2 undef renamable $vgpr0_vgpr1, 0, 0, implicit $exec :: (load (s64), addrspace 1)
|
|
; CHECK-NEXT: early-clobber renamable $agpr0_agpr1_agpr2_agpr3_agpr4_agpr5_agpr6_agpr7_agpr8_agpr9_agpr10_agpr11_agpr12_agpr13_agpr14_agpr15 = V_MFMA_F32_32X32X8F16_e64 $vgpr2_vgpr3, $vgpr2_vgpr3, $agpr16_agpr17_agpr18_agpr19_agpr20_agpr21_agpr22_agpr23_agpr24_agpr25_agpr26_agpr27_agpr28_agpr29_agpr30_agpr31, 0, 0, 0, implicit $mode, implicit $exec
|
|
; CHECK-NEXT: S_CBRANCH_VCCNZ %bb.1, implicit $vcc
|
|
; CHECK-NEXT: S_BRANCH %bb.2
|
|
; CHECK-NEXT: {{ $}}
|
|
; CHECK-NEXT: bb.2:
|
|
; CHECK-NEXT: liveins: $agpr0_agpr1_agpr2_agpr3_agpr4_agpr5_agpr6_agpr7_agpr8_agpr9_agpr10_agpr11_agpr12_agpr13_agpr14_agpr15:0x00000000FFFFFFFF
|
|
; CHECK-NEXT: {{ $}}
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr0_vgpr1_vgpr2_vgpr3_vgpr4_vgpr5_vgpr6_vgpr7
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr8_vgpr9_vgpr10_vgpr11_vgpr12_vgpr13_vgpr14_vgpr15
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr16_vgpr17_vgpr18_vgpr19_vgpr20_vgpr21_vgpr22_vgpr23
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr24_vgpr25_vgpr26_vgpr27_vgpr28_vgpr29_vgpr30_vgpr31
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr32_vgpr33_vgpr34_vgpr35_vgpr36_vgpr37_vgpr38_vgpr39
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr40_vgpr41_vgpr42_vgpr43_vgpr44_vgpr45_vgpr46_vgpr47
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr48_vgpr49_vgpr50_vgpr51_vgpr52_vgpr53_vgpr54_vgpr55
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr56_vgpr57_vgpr58_vgpr59_vgpr60_vgpr61_vgpr62_vgpr63
|
|
; CHECK-NEXT: renamable $vgpr0 = V_MOV_B32_e32 0, implicit $exec
|
|
; CHECK-NEXT: GLOBAL_STORE_DWORDX4_SADDR renamable $vgpr0, renamable $agpr8_agpr9_agpr10_agpr11, undef $sgpr0_sgpr1, 32, 0, implicit $exec :: (store (s128), align 32, addrspace 1)
|
|
; CHECK-NEXT: GLOBAL_STORE_DWORDX4_SADDR renamable $vgpr0, renamable $agpr12_agpr13_agpr14_agpr15, undef $sgpr0_sgpr1, 48, 0, implicit $exec :: (store (s128), addrspace 1)
|
|
; CHECK-NEXT: GLOBAL_STORE_DWORDX4_SADDR renamable $vgpr0, renamable $agpr0_agpr1_agpr2_agpr3, undef $sgpr0_sgpr1, 0, 0, implicit $exec :: (store (s128), align 128, addrspace 1)
|
|
; CHECK-NEXT: GLOBAL_STORE_DWORDX4_SADDR killed renamable $vgpr0, killed renamable $agpr4_agpr5_agpr6_agpr7, killed undef $sgpr0_sgpr1, 16, 0, implicit $exec :: (store (s128), addrspace 1)
|
|
; CHECK-NEXT: S_ENDPGM 0
|
|
bb.0:
|
|
S_NOP 0, implicit-def $agpr0
|
|
renamable $sgpr0 = S_MOV_B32 0
|
|
undef %0.sub8:vreg_512_align2 = V_MOV_B32_e32 0, implicit $exec
|
|
renamable $sgpr1 = COPY renamable $sgpr0
|
|
%1:vreg_64_align2 = COPY killed renamable $sgpr0_sgpr1
|
|
renamable $vcc = S_AND_B64 $exec, -1, implicit-def dead $scc
|
|
%0.sub9:vreg_512_align2 = COPY %0.sub8
|
|
|
|
bb.1:
|
|
liveins: $vcc
|
|
|
|
undef %4.sub0_sub1:vreg_512_align2 = GLOBAL_LOAD_DWORDX2 undef %3:vreg_64_align2, 0, 0, implicit $exec :: (load (s64), addrspace 1)
|
|
%0:vreg_512_align2 = V_MFMA_F32_32X32X8F16_vgprcd_e64 %1, %1, %4, 0, 0, 0, implicit $mode, implicit $exec
|
|
S_CBRANCH_VCCNZ %bb.1, implicit $vcc
|
|
S_BRANCH %bb.2
|
|
|
|
bb.2:
|
|
; No VGPRs available for %0
|
|
S_NOP 0, implicit-def $vgpr0_vgpr1_vgpr2_vgpr3_vgpr4_vgpr5_vgpr6_vgpr7
|
|
S_NOP 0, implicit-def $vgpr8_vgpr9_vgpr10_vgpr11_vgpr12_vgpr13_vgpr14_vgpr15
|
|
S_NOP 0, implicit-def $vgpr16_vgpr17_vgpr18_vgpr19_vgpr20_vgpr21_vgpr22_vgpr23
|
|
S_NOP 0, implicit-def $vgpr24_vgpr25_vgpr26_vgpr27_vgpr28_vgpr29_vgpr30_vgpr31
|
|
S_NOP 0, implicit-def $vgpr32_vgpr33_vgpr34_vgpr35_vgpr36_vgpr37_vgpr38_vgpr39
|
|
S_NOP 0, implicit-def $vgpr40_vgpr41_vgpr42_vgpr43_vgpr44_vgpr45_vgpr46_vgpr47
|
|
S_NOP 0, implicit-def $vgpr48_vgpr49_vgpr50_vgpr51_vgpr52_vgpr53_vgpr54_vgpr55
|
|
S_NOP 0, implicit-def $vgpr56_vgpr57_vgpr58_vgpr59_vgpr60_vgpr61_vgpr62_vgpr63
|
|
%2:vgpr_32 = V_MOV_B32_e32 0, implicit $exec
|
|
GLOBAL_STORE_DWORDX4_SADDR %2, %0.sub8_sub9_sub10_sub11, undef $sgpr0_sgpr1, 32, 0, implicit $exec :: (store (s128), align 32, addrspace 1)
|
|
GLOBAL_STORE_DWORDX4_SADDR %2, %0.sub12_sub13_sub14_sub15, undef $sgpr0_sgpr1, 48, 0, implicit $exec :: (store (s128), addrspace 1)
|
|
GLOBAL_STORE_DWORDX4_SADDR %2, %0.sub0_sub1_sub2_sub3, undef $sgpr0_sgpr1, 0, 0, implicit $exec :: (store (s128), align 128, addrspace 1)
|
|
GLOBAL_STORE_DWORDX4_SADDR %2, %0.sub4_sub5_sub6_sub7, killed undef $sgpr0_sgpr1, 16, 0, implicit $exec :: (store (s128), addrspace 1)
|
|
S_ENDPGM 0
|
|
|
|
...
|
|
|
|
# TODO: The two MFMAs are both rewritable as a set, but if we look at
|
|
# one at a time the register class constraints of the other block the
|
|
# transform.
|
|
---
|
|
name: inflate_result_to_agpr__V_MFMA_F32_32X32X8F16_mac_vgprcd_e64_two_chained_uses
|
|
tracksRegLiveness: true
|
|
machineFunctionInfo:
|
|
isEntryFunction: true
|
|
stackPtrOffsetReg: '$sgpr32'
|
|
occupancy: 10
|
|
sgprForEXECCopy: '$sgpr100_sgpr101'
|
|
body: |
|
|
; CHECK-LABEL: name: inflate_result_to_agpr__V_MFMA_F32_32X32X8F16_mac_vgprcd_e64_two_chained_uses
|
|
; CHECK: bb.0:
|
|
; CHECK-NEXT: successors: %bb.1(0x80000000)
|
|
; CHECK-NEXT: {{ $}}
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $agpr0
|
|
; CHECK-NEXT: renamable $sgpr0 = S_MOV_B32 0
|
|
; CHECK-NEXT: renamable $vgpr8 = V_MOV_B32_e32 0, implicit $exec
|
|
; CHECK-NEXT: renamable $sgpr1 = COPY renamable $sgpr0
|
|
; CHECK-NEXT: renamable $vgpr0_vgpr1 = COPY killed renamable $sgpr0_sgpr1
|
|
; CHECK-NEXT: renamable $vcc = S_AND_B64 $exec, -1, implicit-def dead $scc
|
|
; CHECK-NEXT: dead renamable $vgpr9 = COPY renamable $vgpr8
|
|
; CHECK-NEXT: {{ $}}
|
|
; CHECK-NEXT: bb.1:
|
|
; CHECK-NEXT: successors: %bb.1(0x40000000), %bb.2(0x40000000)
|
|
; CHECK-NEXT: liveins: $vcc, $vgpr0_vgpr1
|
|
; CHECK-NEXT: {{ $}}
|
|
; CHECK-NEXT: renamable $agpr0_agpr1 = GLOBAL_LOAD_DWORDX2 undef renamable $vgpr0_vgpr1, 0, 0, implicit $exec :: (load (s64), addrspace 1)
|
|
; CHECK-NEXT: renamable $agpr0_agpr1_agpr2_agpr3_agpr4_agpr5_agpr6_agpr7_agpr8_agpr9_agpr10_agpr11_agpr12_agpr13_agpr14_agpr15 = V_MFMA_F32_32X32X8F16_mac_e64 $vgpr0_vgpr1, $vgpr0_vgpr1, $agpr0_agpr1_agpr2_agpr3_agpr4_agpr5_agpr6_agpr7_agpr8_agpr9_agpr10_agpr11_agpr12_agpr13_agpr14_agpr15, 0, 0, 0, implicit $mode, implicit $exec
|
|
; CHECK-NEXT: renamable $agpr0_agpr1_agpr2_agpr3_agpr4_agpr5_agpr6_agpr7_agpr8_agpr9_agpr10_agpr11_agpr12_agpr13_agpr14_agpr15 = V_MFMA_F32_32X32X8F16_mac_e64 $vgpr0_vgpr1, $vgpr0_vgpr1, killed $agpr0_agpr1_agpr2_agpr3_agpr4_agpr5_agpr6_agpr7_agpr8_agpr9_agpr10_agpr11_agpr12_agpr13_agpr14_agpr15, 0, 0, 0, implicit $mode, implicit $exec
|
|
; CHECK-NEXT: S_CBRANCH_VCCNZ %bb.1, implicit $vcc
|
|
; CHECK-NEXT: S_BRANCH %bb.2
|
|
; CHECK-NEXT: {{ $}}
|
|
; CHECK-NEXT: bb.2:
|
|
; CHECK-NEXT: liveins: $agpr0_agpr1_agpr2_agpr3_agpr4_agpr5_agpr6_agpr7_agpr8_agpr9_agpr10_agpr11_agpr12_agpr13_agpr14_agpr15:0x00000000FFFFFFFF
|
|
; CHECK-NEXT: {{ $}}
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr0_vgpr1_vgpr2_vgpr3_vgpr4_vgpr5_vgpr6_vgpr7
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr8_vgpr9_vgpr10_vgpr11_vgpr12_vgpr13_vgpr14_vgpr15
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr16_vgpr17_vgpr18_vgpr19_vgpr20_vgpr21_vgpr22_vgpr23
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr24_vgpr25_vgpr26_vgpr27_vgpr28_vgpr29_vgpr30_vgpr31
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr32_vgpr33_vgpr34_vgpr35_vgpr36_vgpr37_vgpr38_vgpr39
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr40_vgpr41_vgpr42_vgpr43_vgpr44_vgpr45_vgpr46_vgpr47
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr48_vgpr49_vgpr50_vgpr51_vgpr52_vgpr53_vgpr54_vgpr55
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr56_vgpr57_vgpr58_vgpr59_vgpr60_vgpr61_vgpr62_vgpr63
|
|
; CHECK-NEXT: renamable $vgpr0 = V_MOV_B32_e32 0, implicit $exec
|
|
; CHECK-NEXT: GLOBAL_STORE_DWORDX4_SADDR renamable $vgpr0, renamable $agpr8_agpr9_agpr10_agpr11, undef $sgpr0_sgpr1, 32, 0, implicit $exec :: (store (s128), align 32, addrspace 1)
|
|
; CHECK-NEXT: GLOBAL_STORE_DWORDX4_SADDR renamable $vgpr0, renamable $agpr12_agpr13_agpr14_agpr15, undef $sgpr0_sgpr1, 48, 0, implicit $exec :: (store (s128), addrspace 1)
|
|
; CHECK-NEXT: GLOBAL_STORE_DWORDX4_SADDR renamable $vgpr0, renamable $agpr0_agpr1_agpr2_agpr3, undef $sgpr0_sgpr1, 0, 0, implicit $exec :: (store (s128), align 128, addrspace 1)
|
|
; CHECK-NEXT: GLOBAL_STORE_DWORDX4_SADDR killed renamable $vgpr0, killed renamable $agpr4_agpr5_agpr6_agpr7, killed undef $sgpr0_sgpr1, 16, 0, implicit $exec :: (store (s128), addrspace 1)
|
|
; CHECK-NEXT: S_ENDPGM 0
|
|
bb.0:
|
|
S_NOP 0, implicit-def $agpr0
|
|
renamable $sgpr0 = S_MOV_B32 0
|
|
undef %0.sub8:vreg_512_align2 = V_MOV_B32_e32 0, implicit $exec
|
|
renamable $sgpr1 = COPY renamable $sgpr0
|
|
%1:vreg_64_align2 = COPY killed renamable $sgpr0_sgpr1
|
|
renamable $vcc = S_AND_B64 $exec, -1, implicit-def dead $scc
|
|
%0.sub9:vreg_512_align2 = COPY %0.sub8
|
|
|
|
bb.1:
|
|
liveins: $vcc
|
|
|
|
undef %0.sub0_sub1:vreg_512_align2 = GLOBAL_LOAD_DWORDX2 undef %3:vreg_64_align2, 0, 0, implicit $exec :: (load (s64), addrspace 1)
|
|
%0:vreg_512_align2 = V_MFMA_F32_32X32X8F16_mac_vgprcd_e64 %1, %1, %0, 0, 0, 0, implicit $mode, implicit $exec
|
|
%0:vreg_512_align2 = V_MFMA_F32_32X32X8F16_mac_vgprcd_e64 %1, %1, %0, 0, 0, 0, implicit $mode, implicit $exec
|
|
S_CBRANCH_VCCNZ %bb.1, implicit $vcc
|
|
S_BRANCH %bb.2
|
|
|
|
bb.2:
|
|
; No VGPRs available for %0
|
|
S_NOP 0, implicit-def $vgpr0_vgpr1_vgpr2_vgpr3_vgpr4_vgpr5_vgpr6_vgpr7
|
|
S_NOP 0, implicit-def $vgpr8_vgpr9_vgpr10_vgpr11_vgpr12_vgpr13_vgpr14_vgpr15
|
|
S_NOP 0, implicit-def $vgpr16_vgpr17_vgpr18_vgpr19_vgpr20_vgpr21_vgpr22_vgpr23
|
|
S_NOP 0, implicit-def $vgpr24_vgpr25_vgpr26_vgpr27_vgpr28_vgpr29_vgpr30_vgpr31
|
|
S_NOP 0, implicit-def $vgpr32_vgpr33_vgpr34_vgpr35_vgpr36_vgpr37_vgpr38_vgpr39
|
|
S_NOP 0, implicit-def $vgpr40_vgpr41_vgpr42_vgpr43_vgpr44_vgpr45_vgpr46_vgpr47
|
|
S_NOP 0, implicit-def $vgpr48_vgpr49_vgpr50_vgpr51_vgpr52_vgpr53_vgpr54_vgpr55
|
|
S_NOP 0, implicit-def $vgpr56_vgpr57_vgpr58_vgpr59_vgpr60_vgpr61_vgpr62_vgpr63
|
|
%2:vgpr_32 = V_MOV_B32_e32 0, implicit $exec
|
|
GLOBAL_STORE_DWORDX4_SADDR %2, %0.sub8_sub9_sub10_sub11, undef $sgpr0_sgpr1, 32, 0, implicit $exec :: (store (s128), align 32, addrspace 1)
|
|
GLOBAL_STORE_DWORDX4_SADDR %2, %0.sub12_sub13_sub14_sub15, undef $sgpr0_sgpr1, 48, 0, implicit $exec :: (store (s128), addrspace 1)
|
|
GLOBAL_STORE_DWORDX4_SADDR %2, %0.sub0_sub1_sub2_sub3, undef $sgpr0_sgpr1, 0, 0, implicit $exec :: (store (s128), align 128, addrspace 1)
|
|
GLOBAL_STORE_DWORDX4_SADDR %2, %0.sub4_sub5_sub6_sub7, killed undef $sgpr0_sgpr1, 16, 0, implicit $exec :: (store (s128), addrspace 1)
|
|
S_ENDPGM 0
|
|
|
|
...
|
|
|
|
# The second MFMA has additional references to the rewritten register
|
|
# %0 in src0/src1, not just src2.
|
|
---
|
|
name: inflate_result_to_agpr__V_MFMA_F32_32X32X8F16_mac_vgprcd_e64_two_chained_uses_repeated_operand
|
|
tracksRegLiveness: true
|
|
machineFunctionInfo:
|
|
isEntryFunction: true
|
|
stackPtrOffsetReg: '$sgpr32'
|
|
occupancy: 10
|
|
sgprForEXECCopy: '$sgpr100_sgpr101'
|
|
body: |
|
|
; CHECK-LABEL: name: inflate_result_to_agpr__V_MFMA_F32_32X32X8F16_mac_vgprcd_e64_two_chained_uses_repeated_operand
|
|
; CHECK: bb.0:
|
|
; CHECK-NEXT: successors: %bb.1(0x80000000)
|
|
; CHECK-NEXT: {{ $}}
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $agpr0
|
|
; CHECK-NEXT: renamable $sgpr0 = S_MOV_B32 0
|
|
; CHECK-NEXT: renamable $vgpr8 = V_MOV_B32_e32 0, implicit $exec
|
|
; CHECK-NEXT: renamable $sgpr1 = COPY renamable $sgpr0
|
|
; CHECK-NEXT: renamable $vgpr0_vgpr1 = COPY killed renamable $sgpr0_sgpr1
|
|
; CHECK-NEXT: renamable $vcc = S_AND_B64 $exec, -1, implicit-def dead $scc
|
|
; CHECK-NEXT: dead renamable $vgpr9 = COPY renamable $vgpr8
|
|
; CHECK-NEXT: {{ $}}
|
|
; CHECK-NEXT: bb.1:
|
|
; CHECK-NEXT: successors: %bb.1(0x40000000), %bb.2(0x40000000)
|
|
; CHECK-NEXT: liveins: $vcc, $vgpr0_vgpr1
|
|
; CHECK-NEXT: {{ $}}
|
|
; CHECK-NEXT: renamable $agpr0_agpr1 = GLOBAL_LOAD_DWORDX2 undef renamable $vgpr0_vgpr1, 0, 0, implicit $exec :: (load (s64), addrspace 1)
|
|
; CHECK-NEXT: renamable $agpr0_agpr1_agpr2_agpr3_agpr4_agpr5_agpr6_agpr7_agpr8_agpr9_agpr10_agpr11_agpr12_agpr13_agpr14_agpr15 = V_MFMA_F32_32X32X8F16_mac_e64 $vgpr0_vgpr1, $vgpr0_vgpr1, $agpr0_agpr1_agpr2_agpr3_agpr4_agpr5_agpr6_agpr7_agpr8_agpr9_agpr10_agpr11_agpr12_agpr13_agpr14_agpr15, 0, 0, 0, implicit $mode, implicit $exec
|
|
; CHECK-NEXT: renamable $agpr0_agpr1_agpr2_agpr3_agpr4_agpr5_agpr6_agpr7_agpr8_agpr9_agpr10_agpr11_agpr12_agpr13_agpr14_agpr15 = V_MFMA_F32_32X32X8F16_mac_e64 killed $agpr2_agpr3, $agpr0_agpr1, $agpr0_agpr1_agpr2_agpr3_agpr4_agpr5_agpr6_agpr7_agpr8_agpr9_agpr10_agpr11_agpr12_agpr13_agpr14_agpr15, 0, 0, 0, implicit $mode, implicit $exec
|
|
; CHECK-NEXT: S_CBRANCH_VCCNZ %bb.1, implicit $vcc
|
|
; CHECK-NEXT: S_BRANCH %bb.2
|
|
; CHECK-NEXT: {{ $}}
|
|
; CHECK-NEXT: bb.2:
|
|
; CHECK-NEXT: liveins: $agpr0_agpr1_agpr2_agpr3_agpr4_agpr5_agpr6_agpr7_agpr8_agpr9_agpr10_agpr11_agpr12_agpr13_agpr14_agpr15:0x00000000FFFFFFFF
|
|
; CHECK-NEXT: {{ $}}
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr0_vgpr1_vgpr2_vgpr3_vgpr4_vgpr5_vgpr6_vgpr7
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr8_vgpr9_vgpr10_vgpr11_vgpr12_vgpr13_vgpr14_vgpr15
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr16_vgpr17_vgpr18_vgpr19_vgpr20_vgpr21_vgpr22_vgpr23
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr24_vgpr25_vgpr26_vgpr27_vgpr28_vgpr29_vgpr30_vgpr31
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr32_vgpr33_vgpr34_vgpr35_vgpr36_vgpr37_vgpr38_vgpr39
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr40_vgpr41_vgpr42_vgpr43_vgpr44_vgpr45_vgpr46_vgpr47
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr48_vgpr49_vgpr50_vgpr51_vgpr52_vgpr53_vgpr54_vgpr55
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr56_vgpr57_vgpr58_vgpr59_vgpr60_vgpr61_vgpr62_vgpr63
|
|
; CHECK-NEXT: renamable $vgpr0 = V_MOV_B32_e32 0, implicit $exec
|
|
; CHECK-NEXT: GLOBAL_STORE_DWORDX4_SADDR renamable $vgpr0, renamable $agpr8_agpr9_agpr10_agpr11, undef $sgpr0_sgpr1, 32, 0, implicit $exec :: (store (s128), align 32, addrspace 1)
|
|
; CHECK-NEXT: GLOBAL_STORE_DWORDX4_SADDR renamable $vgpr0, renamable $agpr12_agpr13_agpr14_agpr15, undef $sgpr0_sgpr1, 48, 0, implicit $exec :: (store (s128), addrspace 1)
|
|
; CHECK-NEXT: GLOBAL_STORE_DWORDX4_SADDR renamable $vgpr0, renamable $agpr0_agpr1_agpr2_agpr3, undef $sgpr0_sgpr1, 0, 0, implicit $exec :: (store (s128), align 128, addrspace 1)
|
|
; CHECK-NEXT: GLOBAL_STORE_DWORDX4_SADDR killed renamable $vgpr0, killed renamable $agpr4_agpr5_agpr6_agpr7, killed undef $sgpr0_sgpr1, 16, 0, implicit $exec :: (store (s128), addrspace 1)
|
|
; CHECK-NEXT: S_ENDPGM 0
|
|
bb.0:
|
|
S_NOP 0, implicit-def $agpr0
|
|
renamable $sgpr0 = S_MOV_B32 0
|
|
undef %0.sub8:vreg_512_align2 = V_MOV_B32_e32 0, implicit $exec
|
|
renamable $sgpr1 = COPY renamable $sgpr0
|
|
%1:vreg_64_align2 = COPY killed renamable $sgpr0_sgpr1
|
|
renamable $vcc = S_AND_B64 $exec, -1, implicit-def dead $scc
|
|
%0.sub9:vreg_512_align2 = COPY %0.sub8
|
|
|
|
bb.1:
|
|
liveins: $vcc
|
|
|
|
undef %0.sub0_sub1:vreg_512_align2 = GLOBAL_LOAD_DWORDX2 undef %3:vreg_64_align2, 0, 0, implicit $exec :: (load (s64), addrspace 1)
|
|
%0:vreg_512_align2 = V_MFMA_F32_32X32X8F16_mac_vgprcd_e64 %1, %1, %0, 0, 0, 0, implicit $mode, implicit $exec
|
|
%0:vreg_512_align2 = V_MFMA_F32_32X32X8F16_mac_vgprcd_e64 %0.sub2_sub3, %0.sub0_sub1, %0, 0, 0, 0, implicit $mode, implicit $exec
|
|
S_CBRANCH_VCCNZ %bb.1, implicit $vcc
|
|
S_BRANCH %bb.2
|
|
|
|
bb.2:
|
|
; No VGPRs available for %0
|
|
S_NOP 0, implicit-def $vgpr0_vgpr1_vgpr2_vgpr3_vgpr4_vgpr5_vgpr6_vgpr7
|
|
S_NOP 0, implicit-def $vgpr8_vgpr9_vgpr10_vgpr11_vgpr12_vgpr13_vgpr14_vgpr15
|
|
S_NOP 0, implicit-def $vgpr16_vgpr17_vgpr18_vgpr19_vgpr20_vgpr21_vgpr22_vgpr23
|
|
S_NOP 0, implicit-def $vgpr24_vgpr25_vgpr26_vgpr27_vgpr28_vgpr29_vgpr30_vgpr31
|
|
S_NOP 0, implicit-def $vgpr32_vgpr33_vgpr34_vgpr35_vgpr36_vgpr37_vgpr38_vgpr39
|
|
S_NOP 0, implicit-def $vgpr40_vgpr41_vgpr42_vgpr43_vgpr44_vgpr45_vgpr46_vgpr47
|
|
S_NOP 0, implicit-def $vgpr48_vgpr49_vgpr50_vgpr51_vgpr52_vgpr53_vgpr54_vgpr55
|
|
S_NOP 0, implicit-def $vgpr56_vgpr57_vgpr58_vgpr59_vgpr60_vgpr61_vgpr62_vgpr63
|
|
%2:vgpr_32 = V_MOV_B32_e32 0, implicit $exec
|
|
GLOBAL_STORE_DWORDX4_SADDR %2, %0.sub8_sub9_sub10_sub11, undef $sgpr0_sgpr1, 32, 0, implicit $exec :: (store (s128), align 32, addrspace 1)
|
|
GLOBAL_STORE_DWORDX4_SADDR %2, %0.sub12_sub13_sub14_sub15, undef $sgpr0_sgpr1, 48, 0, implicit $exec :: (store (s128), addrspace 1)
|
|
GLOBAL_STORE_DWORDX4_SADDR %2, %0.sub0_sub1_sub2_sub3, undef $sgpr0_sgpr1, 0, 0, implicit $exec :: (store (s128), align 128, addrspace 1)
|
|
GLOBAL_STORE_DWORDX4_SADDR %2, %0.sub4_sub5_sub6_sub7, killed undef $sgpr0_sgpr1, 16, 0, implicit $exec :: (store (s128), addrspace 1)
|
|
S_ENDPGM 0
|
|
|
|
...
|
|
|
|
|
|
---
|
|
name: inflate_result_to_agpr__V_MFMA_F32_32X32X8F16_mac_vgprcd_e64_two_chained_uses_cannot_rewrite_final_use
|
|
tracksRegLiveness: true
|
|
machineFunctionInfo:
|
|
isEntryFunction: true
|
|
stackPtrOffsetReg: '$sgpr32'
|
|
occupancy: 10
|
|
sgprForEXECCopy: '$sgpr100_sgpr101'
|
|
body: |
|
|
; CHECK-LABEL: name: inflate_result_to_agpr__V_MFMA_F32_32X32X8F16_mac_vgprcd_e64_two_chained_uses_cannot_rewrite_final_use
|
|
; CHECK: bb.0:
|
|
; CHECK-NEXT: successors: %bb.1(0x80000000)
|
|
; CHECK-NEXT: {{ $}}
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $agpr0
|
|
; CHECK-NEXT: renamable $sgpr0 = S_MOV_B32 0
|
|
; CHECK-NEXT: renamable $vgpr8 = V_MOV_B32_e32 0, implicit $exec
|
|
; CHECK-NEXT: renamable $sgpr1 = COPY renamable $sgpr0
|
|
; CHECK-NEXT: renamable $vgpr0_vgpr1 = COPY killed renamable $sgpr0_sgpr1
|
|
; CHECK-NEXT: renamable $vcc = S_AND_B64 $exec, -1, implicit-def dead $scc
|
|
; CHECK-NEXT: dead renamable $vgpr9 = COPY renamable $vgpr8
|
|
; CHECK-NEXT: {{ $}}
|
|
; CHECK-NEXT: bb.1:
|
|
; CHECK-NEXT: successors: %bb.1(0x40000000), %bb.2(0x40000000)
|
|
; CHECK-NEXT: liveins: $vcc, $vgpr0_vgpr1
|
|
; CHECK-NEXT: {{ $}}
|
|
; CHECK-NEXT: renamable $agpr0_agpr1 = GLOBAL_LOAD_DWORDX2 undef renamable $vgpr0_vgpr1, 0, 0, implicit $exec :: (load (s64), addrspace 1)
|
|
; CHECK-NEXT: renamable $agpr0_agpr1_agpr2_agpr3_agpr4_agpr5_agpr6_agpr7_agpr8_agpr9_agpr10_agpr11_agpr12_agpr13_agpr14_agpr15 = V_MFMA_F32_32X32X8F16_mac_e64 $vgpr0_vgpr1, $vgpr0_vgpr1, $agpr0_agpr1_agpr2_agpr3_agpr4_agpr5_agpr6_agpr7_agpr8_agpr9_agpr10_agpr11_agpr12_agpr13_agpr14_agpr15, 0, 0, 0, implicit $mode, implicit $exec
|
|
; CHECK-NEXT: renamable $agpr0_agpr1_agpr2_agpr3_agpr4_agpr5_agpr6_agpr7_agpr8_agpr9_agpr10_agpr11_agpr12_agpr13_agpr14_agpr15 = V_MFMA_F32_32X32X8F16_mac_e64 $vgpr0_vgpr1, $vgpr0_vgpr1, killed $agpr0_agpr1_agpr2_agpr3_agpr4_agpr5_agpr6_agpr7_agpr8_agpr9_agpr10_agpr11_agpr12_agpr13_agpr14_agpr15, 0, 0, 0, implicit $mode, implicit $exec
|
|
; CHECK-NEXT: S_CBRANCH_VCCNZ %bb.1, implicit $vcc
|
|
; CHECK-NEXT: S_BRANCH %bb.2
|
|
; CHECK-NEXT: {{ $}}
|
|
; CHECK-NEXT: bb.2:
|
|
; CHECK-NEXT: liveins: $agpr0_agpr1_agpr2_agpr3_agpr4_agpr5_agpr6_agpr7_agpr8_agpr9_agpr10_agpr11_agpr12_agpr13_agpr14_agpr15:0x00000000FFFFFFFF
|
|
; CHECK-NEXT: {{ $}}
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr0_vgpr1_vgpr2_vgpr3_vgpr4_vgpr5_vgpr6_vgpr7
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr8_vgpr9_vgpr10_vgpr11_vgpr12_vgpr13_vgpr14_vgpr15
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr16_vgpr17_vgpr18_vgpr19_vgpr20_vgpr21_vgpr22_vgpr23
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr24_vgpr25_vgpr26_vgpr27_vgpr28_vgpr29_vgpr30_vgpr31
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr32_vgpr33_vgpr34_vgpr35_vgpr36_vgpr37_vgpr38_vgpr39
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr40_vgpr41_vgpr42_vgpr43_vgpr44_vgpr45_vgpr46_vgpr47
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr48_vgpr49_vgpr50_vgpr51_vgpr52_vgpr53_vgpr54_vgpr55
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr56_vgpr57_vgpr58_vgpr59_vgpr60_vgpr61_vgpr62_vgpr63
|
|
; CHECK-NEXT: renamable $vgpr0_vgpr1_vgpr2_vgpr3_vgpr4_vgpr5_vgpr6_vgpr7_vgpr8_vgpr9_vgpr10_vgpr11_vgpr12_vgpr13_vgpr14_vgpr15 = COPY killed renamable $agpr0_agpr1_agpr2_agpr3_agpr4_agpr5_agpr6_agpr7_agpr8_agpr9_agpr10_agpr11_agpr12_agpr13_agpr14_agpr15
|
|
; CHECK-NEXT: INLINEASM &"; use $0 ", sideeffect attdialect, reguse:VReg_512_Align2, killed renamable $vgpr0_vgpr1_vgpr2_vgpr3_vgpr4_vgpr5_vgpr6_vgpr7_vgpr8_vgpr9_vgpr10_vgpr11_vgpr12_vgpr13_vgpr14_vgpr15
|
|
; CHECK-NEXT: S_ENDPGM 0
|
|
bb.0:
|
|
S_NOP 0, implicit-def $agpr0
|
|
renamable $sgpr0 = S_MOV_B32 0
|
|
undef %0.sub8:vreg_512_align2 = V_MOV_B32_e32 0, implicit $exec
|
|
renamable $sgpr1 = COPY renamable $sgpr0
|
|
%1:vreg_64_align2 = COPY killed renamable $sgpr0_sgpr1
|
|
renamable $vcc = S_AND_B64 $exec, -1, implicit-def dead $scc
|
|
%0.sub9:vreg_512_align2 = COPY %0.sub8
|
|
|
|
bb.1:
|
|
liveins: $vcc
|
|
|
|
undef %0.sub0_sub1:vreg_512_align2 = GLOBAL_LOAD_DWORDX2 undef %3:vreg_64_align2, 0, 0, implicit $exec :: (load (s64), addrspace 1)
|
|
%0:vreg_512_align2 = V_MFMA_F32_32X32X8F16_mac_vgprcd_e64 %1, %1, %0, 0, 0, 0, implicit $mode, implicit $exec
|
|
%0:vreg_512_align2 = V_MFMA_F32_32X32X8F16_mac_vgprcd_e64 %1, %1, %0, 0, 0, 0, implicit $mode, implicit $exec
|
|
S_CBRANCH_VCCNZ %bb.1, implicit $vcc
|
|
S_BRANCH %bb.2
|
|
|
|
bb.2:
|
|
; No VGPRs available for %0
|
|
S_NOP 0, implicit-def $vgpr0_vgpr1_vgpr2_vgpr3_vgpr4_vgpr5_vgpr6_vgpr7
|
|
S_NOP 0, implicit-def $vgpr8_vgpr9_vgpr10_vgpr11_vgpr12_vgpr13_vgpr14_vgpr15
|
|
S_NOP 0, implicit-def $vgpr16_vgpr17_vgpr18_vgpr19_vgpr20_vgpr21_vgpr22_vgpr23
|
|
S_NOP 0, implicit-def $vgpr24_vgpr25_vgpr26_vgpr27_vgpr28_vgpr29_vgpr30_vgpr31
|
|
S_NOP 0, implicit-def $vgpr32_vgpr33_vgpr34_vgpr35_vgpr36_vgpr37_vgpr38_vgpr39
|
|
S_NOP 0, implicit-def $vgpr40_vgpr41_vgpr42_vgpr43_vgpr44_vgpr45_vgpr46_vgpr47
|
|
S_NOP 0, implicit-def $vgpr48_vgpr49_vgpr50_vgpr51_vgpr52_vgpr53_vgpr54_vgpr55
|
|
S_NOP 0, implicit-def $vgpr56_vgpr57_vgpr58_vgpr59_vgpr60_vgpr61_vgpr62_vgpr63
|
|
INLINEASM &"; use $0 ", sideeffect attdialect, reguse:VReg_512_Align2, %0:vreg_512_align2
|
|
S_ENDPGM 0
|
|
|
|
...
|
|
|
|
# There is a rewrite candidate, but it is used by another MFMA which
|
|
# does not have a tied result.
|
|
---
|
|
name: inflate_result_to_agpr__V_MFMA_F32_32X32X8F16_mac_vgprcd_e64_2nd_use_not_tied
|
|
tracksRegLiveness: true
|
|
machineFunctionInfo:
|
|
isEntryFunction: true
|
|
stackPtrOffsetReg: '$sgpr32'
|
|
occupancy: 10
|
|
sgprForEXECCopy: '$sgpr100_sgpr101'
|
|
body: |
|
|
; CHECK-LABEL: name: inflate_result_to_agpr__V_MFMA_F32_32X32X8F16_mac_vgprcd_e64_2nd_use_not_tied
|
|
; CHECK: bb.0:
|
|
; CHECK-NEXT: successors: %bb.1(0x80000000)
|
|
; CHECK-NEXT: {{ $}}
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $agpr0
|
|
; CHECK-NEXT: renamable $sgpr0 = S_MOV_B32 0
|
|
; CHECK-NEXT: renamable $vgpr8 = V_MOV_B32_e32 0, implicit $exec
|
|
; CHECK-NEXT: renamable $sgpr1 = COPY renamable $sgpr0
|
|
; CHECK-NEXT: renamable $vgpr16_vgpr17 = COPY killed renamable $sgpr0_sgpr1
|
|
; CHECK-NEXT: renamable $vcc = S_AND_B64 $exec, -1, implicit-def dead $scc
|
|
; CHECK-NEXT: dead renamable $vgpr9 = COPY renamable $vgpr8
|
|
; CHECK-NEXT: {{ $}}
|
|
; CHECK-NEXT: bb.1:
|
|
; CHECK-NEXT: successors: %bb.1(0x40000000), %bb.2(0x40000000)
|
|
; CHECK-NEXT: liveins: $vcc, $vgpr16_vgpr17
|
|
; CHECK-NEXT: {{ $}}
|
|
; CHECK-NEXT: renamable $agpr16_agpr17 = GLOBAL_LOAD_DWORDX2 undef renamable $vgpr0_vgpr1, 0, 0, implicit $exec :: (load (s64), addrspace 1)
|
|
; CHECK-NEXT: renamable $agpr16_agpr17_agpr18_agpr19_agpr20_agpr21_agpr22_agpr23_agpr24_agpr25_agpr26_agpr27_agpr28_agpr29_agpr30_agpr31 = V_MFMA_F32_32X32X8F16_mac_e64 $vgpr16_vgpr17, $vgpr16_vgpr17, $agpr16_agpr17_agpr18_agpr19_agpr20_agpr21_agpr22_agpr23_agpr24_agpr25_agpr26_agpr27_agpr28_agpr29_agpr30_agpr31, 0, 0, 0, implicit $mode, implicit $exec
|
|
; CHECK-NEXT: early-clobber renamable $agpr0_agpr1_agpr2_agpr3_agpr4_agpr5_agpr6_agpr7_agpr8_agpr9_agpr10_agpr11_agpr12_agpr13_agpr14_agpr15 = V_MFMA_F32_32X32X8F16_e64 $vgpr16_vgpr17, $vgpr16_vgpr17, killed $agpr16_agpr17_agpr18_agpr19_agpr20_agpr21_agpr22_agpr23_agpr24_agpr25_agpr26_agpr27_agpr28_agpr29_agpr30_agpr31, 0, 0, 0, implicit $mode, implicit $exec
|
|
; CHECK-NEXT: S_CBRANCH_VCCNZ %bb.1, implicit $vcc
|
|
; CHECK-NEXT: S_BRANCH %bb.2
|
|
; CHECK-NEXT: {{ $}}
|
|
; CHECK-NEXT: bb.2:
|
|
; CHECK-NEXT: liveins: $agpr0_agpr1_agpr2_agpr3_agpr4_agpr5_agpr6_agpr7_agpr8_agpr9_agpr10_agpr11_agpr12_agpr13_agpr14_agpr15:0x00000000FFFFFFFF
|
|
; CHECK-NEXT: {{ $}}
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr0_vgpr1_vgpr2_vgpr3_vgpr4_vgpr5_vgpr6_vgpr7
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr8_vgpr9_vgpr10_vgpr11_vgpr12_vgpr13_vgpr14_vgpr15
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr16_vgpr17_vgpr18_vgpr19_vgpr20_vgpr21_vgpr22_vgpr23
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr24_vgpr25_vgpr26_vgpr27_vgpr28_vgpr29_vgpr30_vgpr31
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr32_vgpr33_vgpr34_vgpr35_vgpr36_vgpr37_vgpr38_vgpr39
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr40_vgpr41_vgpr42_vgpr43_vgpr44_vgpr45_vgpr46_vgpr47
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr48_vgpr49_vgpr50_vgpr51_vgpr52_vgpr53_vgpr54_vgpr55
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr56_vgpr57_vgpr58_vgpr59_vgpr60_vgpr61_vgpr62_vgpr63
|
|
; CHECK-NEXT: renamable $vgpr0 = V_MOV_B32_e32 0, implicit $exec
|
|
; CHECK-NEXT: GLOBAL_STORE_DWORDX4_SADDR renamable $vgpr0, renamable $agpr8_agpr9_agpr10_agpr11, undef $sgpr0_sgpr1, 32, 0, implicit $exec :: (store (s128), align 32, addrspace 1)
|
|
; CHECK-NEXT: GLOBAL_STORE_DWORDX4_SADDR renamable $vgpr0, renamable $agpr12_agpr13_agpr14_agpr15, undef $sgpr0_sgpr1, 48, 0, implicit $exec :: (store (s128), addrspace 1)
|
|
; CHECK-NEXT: GLOBAL_STORE_DWORDX4_SADDR renamable $vgpr0, renamable $agpr0_agpr1_agpr2_agpr3, undef $sgpr0_sgpr1, 0, 0, implicit $exec :: (store (s128), align 128, addrspace 1)
|
|
; CHECK-NEXT: GLOBAL_STORE_DWORDX4_SADDR killed renamable $vgpr0, killed renamable $agpr4_agpr5_agpr6_agpr7, killed undef $sgpr0_sgpr1, 16, 0, implicit $exec :: (store (s128), addrspace 1)
|
|
; CHECK-NEXT: S_ENDPGM 0
|
|
bb.0:
|
|
S_NOP 0, implicit-def $agpr0
|
|
renamable $sgpr0 = S_MOV_B32 0
|
|
undef %0.sub8:vreg_512_align2 = V_MOV_B32_e32 0, implicit $exec
|
|
renamable $sgpr1 = COPY renamable $sgpr0
|
|
%1:vreg_64_align2 = COPY killed renamable $sgpr0_sgpr1
|
|
renamable $vcc = S_AND_B64 $exec, -1, implicit-def dead $scc
|
|
%0.sub9:vreg_512_align2 = COPY %0.sub8
|
|
|
|
bb.1:
|
|
liveins: $vcc
|
|
|
|
undef %0.sub0_sub1:vreg_512_align2 = GLOBAL_LOAD_DWORDX2 undef %2:vreg_64_align2, 0, 0, implicit $exec :: (load (s64), addrspace 1)
|
|
%0:vreg_512_align2 = V_MFMA_F32_32X32X8F16_mac_vgprcd_e64 %1, %1, %0, 0, 0, 0, implicit $mode, implicit $exec
|
|
%3:vreg_512_align2 = V_MFMA_F32_32X32X8F16_vgprcd_e64 %1, %1, %0, 0, 0, 0, implicit $mode, implicit $exec
|
|
S_CBRANCH_VCCNZ %bb.1, implicit $vcc
|
|
S_BRANCH %bb.2
|
|
|
|
bb.2:
|
|
; No VGPRs available for %0 or %3
|
|
S_NOP 0, implicit-def $vgpr0_vgpr1_vgpr2_vgpr3_vgpr4_vgpr5_vgpr6_vgpr7
|
|
S_NOP 0, implicit-def $vgpr8_vgpr9_vgpr10_vgpr11_vgpr12_vgpr13_vgpr14_vgpr15
|
|
S_NOP 0, implicit-def $vgpr16_vgpr17_vgpr18_vgpr19_vgpr20_vgpr21_vgpr22_vgpr23
|
|
S_NOP 0, implicit-def $vgpr24_vgpr25_vgpr26_vgpr27_vgpr28_vgpr29_vgpr30_vgpr31
|
|
S_NOP 0, implicit-def $vgpr32_vgpr33_vgpr34_vgpr35_vgpr36_vgpr37_vgpr38_vgpr39
|
|
S_NOP 0, implicit-def $vgpr40_vgpr41_vgpr42_vgpr43_vgpr44_vgpr45_vgpr46_vgpr47
|
|
S_NOP 0, implicit-def $vgpr48_vgpr49_vgpr50_vgpr51_vgpr52_vgpr53_vgpr54_vgpr55
|
|
S_NOP 0, implicit-def $vgpr56_vgpr57_vgpr58_vgpr59_vgpr60_vgpr61_vgpr62_vgpr63
|
|
%4:vgpr_32 = V_MOV_B32_e32 0, implicit $exec
|
|
GLOBAL_STORE_DWORDX4_SADDR %4, %3.sub8_sub9_sub10_sub11, undef $sgpr0_sgpr1, 32, 0, implicit $exec :: (store (s128), align 32, addrspace 1)
|
|
GLOBAL_STORE_DWORDX4_SADDR %4, %3.sub12_sub13_sub14_sub15, undef $sgpr0_sgpr1, 48, 0, implicit $exec :: (store (s128), addrspace 1)
|
|
GLOBAL_STORE_DWORDX4_SADDR %4, %3.sub0_sub1_sub2_sub3, undef $sgpr0_sgpr1, 0, 0, implicit $exec :: (store (s128), align 128, addrspace 1)
|
|
GLOBAL_STORE_DWORDX4_SADDR %4, %3.sub4_sub5_sub6_sub7, killed undef $sgpr0_sgpr1, 16, 0, implicit $exec :: (store (s128), addrspace 1)
|
|
S_ENDPGM 0
|
|
|
|
...
|
|
|
|
# Chain of 2 untied cases
|
|
---
|
|
name: inflate_result_to_agpr__V_MFMA_F32_32X32X8F16_vgprcd_e64_chain
|
|
tracksRegLiveness: true
|
|
machineFunctionInfo:
|
|
isEntryFunction: true
|
|
stackPtrOffsetReg: '$sgpr32'
|
|
occupancy: 10
|
|
sgprForEXECCopy: '$sgpr100_sgpr101'
|
|
body: |
|
|
; CHECK-LABEL: name: inflate_result_to_agpr__V_MFMA_F32_32X32X8F16_vgprcd_e64_chain
|
|
; CHECK: bb.0:
|
|
; CHECK-NEXT: successors: %bb.1(0x80000000)
|
|
; CHECK-NEXT: {{ $}}
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $agpr0
|
|
; CHECK-NEXT: renamable $sgpr0 = S_MOV_B32 0
|
|
; CHECK-NEXT: renamable $vgpr8 = V_MOV_B32_e32 0, implicit $exec
|
|
; CHECK-NEXT: renamable $sgpr1 = COPY renamable $sgpr0
|
|
; CHECK-NEXT: renamable $vgpr18_vgpr19 = COPY killed renamable $sgpr0_sgpr1
|
|
; CHECK-NEXT: renamable $vcc = S_AND_B64 $exec, -1, implicit-def dead $scc
|
|
; CHECK-NEXT: dead renamable $vgpr9 = COPY renamable $vgpr8
|
|
; CHECK-NEXT: {{ $}}
|
|
; CHECK-NEXT: bb.1:
|
|
; CHECK-NEXT: successors: %bb.1(0x40000000), %bb.2(0x40000000)
|
|
; CHECK-NEXT: liveins: $vcc, $vgpr18_vgpr19
|
|
; CHECK-NEXT: {{ $}}
|
|
; CHECK-NEXT: renamable $agpr0_agpr1 = GLOBAL_LOAD_DWORDX2 undef renamable $vgpr0_vgpr1, 0, 0, implicit $exec :: (load (s64), addrspace 1)
|
|
; CHECK-NEXT: early-clobber renamable $agpr16_agpr17_agpr18_agpr19_agpr20_agpr21_agpr22_agpr23_agpr24_agpr25_agpr26_agpr27_agpr28_agpr29_agpr30_agpr31 = V_MFMA_F32_32X32X8F16_e64 $vgpr18_vgpr19, $vgpr18_vgpr19, $agpr0_agpr1_agpr2_agpr3_agpr4_agpr5_agpr6_agpr7_agpr8_agpr9_agpr10_agpr11_agpr12_agpr13_agpr14_agpr15, 0, 0, 0, implicit $mode, implicit $exec
|
|
; CHECK-NEXT: early-clobber renamable $agpr0_agpr1_agpr2_agpr3_agpr4_agpr5_agpr6_agpr7_agpr8_agpr9_agpr10_agpr11_agpr12_agpr13_agpr14_agpr15 = V_MFMA_F32_32X32X8F16_e64 $vgpr18_vgpr19, $vgpr18_vgpr19, killed $agpr16_agpr17_agpr18_agpr19_agpr20_agpr21_agpr22_agpr23_agpr24_agpr25_agpr26_agpr27_agpr28_agpr29_agpr30_agpr31, 0, 0, 0, implicit $mode, implicit $exec
|
|
; CHECK-NEXT: S_CBRANCH_VCCNZ %bb.1, implicit $vcc
|
|
; CHECK-NEXT: S_BRANCH %bb.2
|
|
; CHECK-NEXT: {{ $}}
|
|
; CHECK-NEXT: bb.2:
|
|
; CHECK-NEXT: liveins: $agpr0_agpr1_agpr2_agpr3_agpr4_agpr5_agpr6_agpr7_agpr8_agpr9_agpr10_agpr11_agpr12_agpr13_agpr14_agpr15:0x00000000FFFFFFFF
|
|
; CHECK-NEXT: {{ $}}
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr0_vgpr1_vgpr2_vgpr3_vgpr4_vgpr5_vgpr6_vgpr7
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr8_vgpr9_vgpr10_vgpr11_vgpr12_vgpr13_vgpr14_vgpr15
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr16_vgpr17_vgpr18_vgpr19_vgpr20_vgpr21_vgpr22_vgpr23
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr24_vgpr25_vgpr26_vgpr27_vgpr28_vgpr29_vgpr30_vgpr31
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr32_vgpr33_vgpr34_vgpr35_vgpr36_vgpr37_vgpr38_vgpr39
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr40_vgpr41_vgpr42_vgpr43_vgpr44_vgpr45_vgpr46_vgpr47
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr48_vgpr49_vgpr50_vgpr51_vgpr52_vgpr53_vgpr54_vgpr55
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr56_vgpr57_vgpr58_vgpr59_vgpr60_vgpr61_vgpr62_vgpr63
|
|
; CHECK-NEXT: renamable $vgpr0 = V_MOV_B32_e32 0, implicit $exec
|
|
; CHECK-NEXT: GLOBAL_STORE_DWORDX4_SADDR renamable $vgpr0, renamable $agpr8_agpr9_agpr10_agpr11, undef $sgpr0_sgpr1, 32, 0, implicit $exec :: (store (s128), align 32, addrspace 1)
|
|
; CHECK-NEXT: GLOBAL_STORE_DWORDX4_SADDR renamable $vgpr0, renamable $agpr12_agpr13_agpr14_agpr15, undef $sgpr0_sgpr1, 48, 0, implicit $exec :: (store (s128), addrspace 1)
|
|
; CHECK-NEXT: GLOBAL_STORE_DWORDX4_SADDR renamable $vgpr0, renamable $agpr0_agpr1_agpr2_agpr3, undef $sgpr0_sgpr1, 0, 0, implicit $exec :: (store (s128), align 128, addrspace 1)
|
|
; CHECK-NEXT: GLOBAL_STORE_DWORDX4_SADDR killed renamable $vgpr0, killed renamable $agpr4_agpr5_agpr6_agpr7, killed undef $sgpr0_sgpr1, 16, 0, implicit $exec :: (store (s128), addrspace 1)
|
|
; CHECK-NEXT: S_ENDPGM 0
|
|
bb.0:
|
|
S_NOP 0, implicit-def $agpr0
|
|
renamable $sgpr0 = S_MOV_B32 0
|
|
undef %0.sub8:vreg_512_align2 = V_MOV_B32_e32 0, implicit $exec
|
|
renamable $sgpr1 = COPY renamable $sgpr0
|
|
%1:vreg_64_align2 = COPY killed renamable $sgpr0_sgpr1
|
|
renamable $vcc = S_AND_B64 $exec, -1, implicit-def dead $scc
|
|
%0.sub9:vreg_512_align2 = COPY %0.sub8
|
|
|
|
bb.1:
|
|
liveins: $vcc
|
|
|
|
undef %0.sub0_sub1:vreg_512_align2 = GLOBAL_LOAD_DWORDX2 undef %2:vreg_64_align2, 0, 0, implicit $exec :: (load (s64), addrspace 1)
|
|
%3:vreg_512_align2 = V_MFMA_F32_32X32X8F16_vgprcd_e64 %1, %1, %0, 0, 0, 0, implicit $mode, implicit $exec
|
|
%4:vreg_512_align2 = V_MFMA_F32_32X32X8F16_vgprcd_e64 %1, %1, %3, 0, 0, 0, implicit $mode, implicit $exec
|
|
S_CBRANCH_VCCNZ %bb.1, implicit $vcc
|
|
S_BRANCH %bb.2
|
|
|
|
bb.2:
|
|
; No VGPRs available for %0 or %4
|
|
S_NOP 0, implicit-def $vgpr0_vgpr1_vgpr2_vgpr3_vgpr4_vgpr5_vgpr6_vgpr7
|
|
S_NOP 0, implicit-def $vgpr8_vgpr9_vgpr10_vgpr11_vgpr12_vgpr13_vgpr14_vgpr15
|
|
S_NOP 0, implicit-def $vgpr16_vgpr17_vgpr18_vgpr19_vgpr20_vgpr21_vgpr22_vgpr23
|
|
S_NOP 0, implicit-def $vgpr24_vgpr25_vgpr26_vgpr27_vgpr28_vgpr29_vgpr30_vgpr31
|
|
S_NOP 0, implicit-def $vgpr32_vgpr33_vgpr34_vgpr35_vgpr36_vgpr37_vgpr38_vgpr39
|
|
S_NOP 0, implicit-def $vgpr40_vgpr41_vgpr42_vgpr43_vgpr44_vgpr45_vgpr46_vgpr47
|
|
S_NOP 0, implicit-def $vgpr48_vgpr49_vgpr50_vgpr51_vgpr52_vgpr53_vgpr54_vgpr55
|
|
S_NOP 0, implicit-def $vgpr56_vgpr57_vgpr58_vgpr59_vgpr60_vgpr61_vgpr62_vgpr63
|
|
%5:vgpr_32 = V_MOV_B32_e32 0, implicit $exec
|
|
GLOBAL_STORE_DWORDX4_SADDR %5, %4.sub8_sub9_sub10_sub11, undef $sgpr0_sgpr1, 32, 0, implicit $exec :: (store (s128), align 32, addrspace 1)
|
|
GLOBAL_STORE_DWORDX4_SADDR %5, %4.sub12_sub13_sub14_sub15, undef $sgpr0_sgpr1, 48, 0, implicit $exec :: (store (s128), addrspace 1)
|
|
GLOBAL_STORE_DWORDX4_SADDR %5, %4.sub0_sub1_sub2_sub3, undef $sgpr0_sgpr1, 0, 0, implicit $exec :: (store (s128), align 128, addrspace 1)
|
|
GLOBAL_STORE_DWORDX4_SADDR %5, %4.sub4_sub5_sub6_sub7, killed undef $sgpr0_sgpr1, 16, 0, implicit $exec :: (store (s128), addrspace 1)
|
|
S_ENDPGM 0
|
|
|
|
...
|
|
# Chain of 2 untied cases, but the use isn't in src2.
|
|
---
|
|
name: inflate_result_to_agpr__V_MFMA_F32_32X32X8F16_vgprcd_e64_chain_non_src2
|
|
tracksRegLiveness: true
|
|
machineFunctionInfo:
|
|
isEntryFunction: true
|
|
stackPtrOffsetReg: '$sgpr32'
|
|
occupancy: 10
|
|
sgprForEXECCopy: '$sgpr100_sgpr101'
|
|
body: |
|
|
; CHECK-LABEL: name: inflate_result_to_agpr__V_MFMA_F32_32X32X8F16_vgprcd_e64_chain_non_src2
|
|
; CHECK: bb.0:
|
|
; CHECK-NEXT: successors: %bb.1(0x80000000)
|
|
; CHECK-NEXT: {{ $}}
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $agpr0
|
|
; CHECK-NEXT: renamable $sgpr0 = S_MOV_B32 0
|
|
; CHECK-NEXT: renamable $vgpr8 = V_MOV_B32_e32 0, implicit $exec
|
|
; CHECK-NEXT: renamable $sgpr1 = COPY renamable $sgpr0
|
|
; CHECK-NEXT: renamable $vgpr18_vgpr19 = COPY killed renamable $sgpr0_sgpr1
|
|
; CHECK-NEXT: renamable $vcc = S_AND_B64 $exec, -1, implicit-def dead $scc
|
|
; CHECK-NEXT: dead renamable $vgpr9 = COPY renamable $vgpr8
|
|
; CHECK-NEXT: {{ $}}
|
|
; CHECK-NEXT: bb.1:
|
|
; CHECK-NEXT: successors: %bb.1(0x40000000), %bb.2(0x40000000)
|
|
; CHECK-NEXT: liveins: $vcc, $vgpr18_vgpr19
|
|
; CHECK-NEXT: {{ $}}
|
|
; CHECK-NEXT: renamable $vgpr16_vgpr17 = GLOBAL_LOAD_DWORDX2 undef renamable $vgpr0_vgpr1, 0, 0, implicit $exec :: (load (s64), addrspace 1)
|
|
; CHECK-NEXT: early-clobber renamable $vgpr0_vgpr1_vgpr2_vgpr3_vgpr4_vgpr5_vgpr6_vgpr7_vgpr8_vgpr9_vgpr10_vgpr11_vgpr12_vgpr13_vgpr14_vgpr15 = V_MFMA_F32_32X32X8F16_vgprcd_e64 $vgpr18_vgpr19, $vgpr18_vgpr19, $vgpr16_vgpr17_vgpr18_vgpr19_vgpr20_vgpr21_vgpr22_vgpr23_vgpr24_vgpr25_vgpr26_vgpr27_vgpr28_vgpr29_vgpr30_vgpr31, 0, 0, 0, implicit $mode, implicit $exec
|
|
; CHECK-NEXT: early-clobber renamable $agpr0_agpr1_agpr2_agpr3_agpr4_agpr5_agpr6_agpr7_agpr8_agpr9_agpr10_agpr11_agpr12_agpr13_agpr14_agpr15 = V_MFMA_F32_32X32X8F16_e64 killed $vgpr4_vgpr5, $vgpr8_vgpr9, undef $agpr0_agpr1_agpr2_agpr3_agpr4_agpr5_agpr6_agpr7_agpr8_agpr9_agpr10_agpr11_agpr12_agpr13_agpr14_agpr15, 0, 0, 0, implicit $mode, implicit $exec
|
|
; CHECK-NEXT: S_CBRANCH_VCCNZ %bb.1, implicit $vcc
|
|
; CHECK-NEXT: S_BRANCH %bb.2
|
|
; CHECK-NEXT: {{ $}}
|
|
; CHECK-NEXT: bb.2:
|
|
; CHECK-NEXT: liveins: $agpr0_agpr1_agpr2_agpr3_agpr4_agpr5_agpr6_agpr7_agpr8_agpr9_agpr10_agpr11_agpr12_agpr13_agpr14_agpr15:0x00000000FFFFFFFF
|
|
; CHECK-NEXT: {{ $}}
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr0_vgpr1_vgpr2_vgpr3_vgpr4_vgpr5_vgpr6_vgpr7
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr8_vgpr9_vgpr10_vgpr11_vgpr12_vgpr13_vgpr14_vgpr15
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr16_vgpr17_vgpr18_vgpr19_vgpr20_vgpr21_vgpr22_vgpr23
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr24_vgpr25_vgpr26_vgpr27_vgpr28_vgpr29_vgpr30_vgpr31
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr32_vgpr33_vgpr34_vgpr35_vgpr36_vgpr37_vgpr38_vgpr39
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr40_vgpr41_vgpr42_vgpr43_vgpr44_vgpr45_vgpr46_vgpr47
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr48_vgpr49_vgpr50_vgpr51_vgpr52_vgpr53_vgpr54_vgpr55
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr56_vgpr57_vgpr58_vgpr59_vgpr60_vgpr61_vgpr62_vgpr63
|
|
; CHECK-NEXT: renamable $vgpr0 = V_MOV_B32_e32 0, implicit $exec
|
|
; CHECK-NEXT: GLOBAL_STORE_DWORDX4_SADDR renamable $vgpr0, renamable $agpr8_agpr9_agpr10_agpr11, undef $sgpr0_sgpr1, 32, 0, implicit $exec :: (store (s128), align 32, addrspace 1)
|
|
; CHECK-NEXT: GLOBAL_STORE_DWORDX4_SADDR renamable $vgpr0, renamable $agpr12_agpr13_agpr14_agpr15, undef $sgpr0_sgpr1, 48, 0, implicit $exec :: (store (s128), addrspace 1)
|
|
; CHECK-NEXT: GLOBAL_STORE_DWORDX4_SADDR renamable $vgpr0, renamable $agpr0_agpr1_agpr2_agpr3, undef $sgpr0_sgpr1, 0, 0, implicit $exec :: (store (s128), align 128, addrspace 1)
|
|
; CHECK-NEXT: GLOBAL_STORE_DWORDX4_SADDR killed renamable $vgpr0, killed renamable $agpr4_agpr5_agpr6_agpr7, killed undef $sgpr0_sgpr1, 16, 0, implicit $exec :: (store (s128), addrspace 1)
|
|
; CHECK-NEXT: S_ENDPGM 0
|
|
bb.0:
|
|
S_NOP 0, implicit-def $agpr0
|
|
renamable $sgpr0 = S_MOV_B32 0
|
|
undef %0.sub8:vreg_512_align2 = V_MOV_B32_e32 0, implicit $exec
|
|
renamable $sgpr1 = COPY renamable $sgpr0
|
|
%1:vreg_64_align2 = COPY killed renamable $sgpr0_sgpr1
|
|
renamable $vcc = S_AND_B64 $exec, -1, implicit-def dead $scc
|
|
%0.sub9:vreg_512_align2 = COPY %0.sub8
|
|
|
|
bb.1:
|
|
liveins: $vcc
|
|
|
|
undef %0.sub0_sub1:vreg_512_align2 = GLOBAL_LOAD_DWORDX2 undef %2:vreg_64_align2, 0, 0, implicit $exec :: (load (s64), addrspace 1)
|
|
%3:vreg_512_align2 = V_MFMA_F32_32X32X8F16_vgprcd_e64 %1, %1, %0, 0, 0, 0, implicit $mode, implicit $exec
|
|
%4:vreg_512_align2 = V_MFMA_F32_32X32X8F16_vgprcd_e64 %3.sub4_sub5, %3.sub8_sub9, undef %6:vreg_512_align2, 0, 0, 0, implicit $mode, implicit $exec
|
|
S_CBRANCH_VCCNZ %bb.1, implicit $vcc
|
|
S_BRANCH %bb.2
|
|
|
|
bb.2:
|
|
; No VGPRs available for %0 or %4
|
|
S_NOP 0, implicit-def $vgpr0_vgpr1_vgpr2_vgpr3_vgpr4_vgpr5_vgpr6_vgpr7
|
|
S_NOP 0, implicit-def $vgpr8_vgpr9_vgpr10_vgpr11_vgpr12_vgpr13_vgpr14_vgpr15
|
|
S_NOP 0, implicit-def $vgpr16_vgpr17_vgpr18_vgpr19_vgpr20_vgpr21_vgpr22_vgpr23
|
|
S_NOP 0, implicit-def $vgpr24_vgpr25_vgpr26_vgpr27_vgpr28_vgpr29_vgpr30_vgpr31
|
|
S_NOP 0, implicit-def $vgpr32_vgpr33_vgpr34_vgpr35_vgpr36_vgpr37_vgpr38_vgpr39
|
|
S_NOP 0, implicit-def $vgpr40_vgpr41_vgpr42_vgpr43_vgpr44_vgpr45_vgpr46_vgpr47
|
|
S_NOP 0, implicit-def $vgpr48_vgpr49_vgpr50_vgpr51_vgpr52_vgpr53_vgpr54_vgpr55
|
|
S_NOP 0, implicit-def $vgpr56_vgpr57_vgpr58_vgpr59_vgpr60_vgpr61_vgpr62_vgpr63
|
|
%5:vgpr_32 = V_MOV_B32_e32 0, implicit $exec
|
|
GLOBAL_STORE_DWORDX4_SADDR %5, %4.sub8_sub9_sub10_sub11, undef $sgpr0_sgpr1, 32, 0, implicit $exec :: (store (s128), align 32, addrspace 1)
|
|
GLOBAL_STORE_DWORDX4_SADDR %5, %4.sub12_sub13_sub14_sub15, undef $sgpr0_sgpr1, 48, 0, implicit $exec :: (store (s128), addrspace 1)
|
|
GLOBAL_STORE_DWORDX4_SADDR %5, %4.sub0_sub1_sub2_sub3, undef $sgpr0_sgpr1, 0, 0, implicit $exec :: (store (s128), align 128, addrspace 1)
|
|
GLOBAL_STORE_DWORDX4_SADDR %5, %4.sub4_sub5_sub6_sub7, killed undef $sgpr0_sgpr1, 16, 0, implicit $exec :: (store (s128), addrspace 1)
|
|
S_ENDPGM 0
|
|
|
|
...
|
|
|
|
# Chain of 2 untied cases, but the second mfma is a different size and
|
|
# uses a subregister.
|
|
---
|
|
name: inflate_result_to_agpr__V_MFMA_F32_32X32X8F16_vgprcd_e64_chain_subreg
|
|
tracksRegLiveness: true
|
|
machineFunctionInfo:
|
|
isEntryFunction: true
|
|
stackPtrOffsetReg: '$sgpr32'
|
|
occupancy: 10
|
|
sgprForEXECCopy: '$sgpr100_sgpr101'
|
|
body: |
|
|
; CHECK-LABEL: name: inflate_result_to_agpr__V_MFMA_F32_32X32X8F16_vgprcd_e64_chain_subreg
|
|
; CHECK: bb.0:
|
|
; CHECK-NEXT: successors: %bb.1(0x80000000)
|
|
; CHECK-NEXT: {{ $}}
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $agpr0
|
|
; CHECK-NEXT: renamable $sgpr0 = S_MOV_B32 0
|
|
; CHECK-NEXT: renamable $vgpr8 = V_MOV_B32_e32 0, implicit $exec
|
|
; CHECK-NEXT: renamable $sgpr1 = COPY renamable $sgpr0
|
|
; CHECK-NEXT: renamable $vgpr18_vgpr19 = COPY killed renamable $sgpr0_sgpr1
|
|
; CHECK-NEXT: renamable $vcc = S_AND_B64 $exec, -1, implicit-def dead $scc
|
|
; CHECK-NEXT: dead renamable $vgpr9 = COPY renamable $vgpr8
|
|
; CHECK-NEXT: {{ $}}
|
|
; CHECK-NEXT: bb.1:
|
|
; CHECK-NEXT: successors: %bb.1(0x40000000), %bb.2(0x40000000)
|
|
; CHECK-NEXT: liveins: $vcc, $vgpr18_vgpr19
|
|
; CHECK-NEXT: {{ $}}
|
|
; CHECK-NEXT: renamable $agpr16_agpr17 = GLOBAL_LOAD_DWORDX2 undef renamable $vgpr0_vgpr1, 0, 0, implicit $exec :: (load (s64), addrspace 1)
|
|
; CHECK-NEXT: early-clobber renamable $agpr0_agpr1_agpr2_agpr3_agpr4_agpr5_agpr6_agpr7_agpr8_agpr9_agpr10_agpr11_agpr12_agpr13_agpr14_agpr15 = V_MFMA_F32_32X32X8F16_e64 $vgpr18_vgpr19, $vgpr18_vgpr19, $agpr16_agpr17_agpr18_agpr19_agpr20_agpr21_agpr22_agpr23_agpr24_agpr25_agpr26_agpr27_agpr28_agpr29_agpr30_agpr31, 0, 0, 0, implicit $mode, implicit $exec
|
|
; CHECK-NEXT: renamable $agpr0_agpr1_agpr2_agpr3 = V_MFMA_F32_16X16X16F16_e64 $vgpr18_vgpr19, $vgpr18_vgpr19, killed $agpr2_agpr3_agpr4_agpr5, 0, 0, 0, implicit $mode, implicit $exec
|
|
; CHECK-NEXT: S_CBRANCH_VCCNZ %bb.1, implicit $vcc
|
|
; CHECK-NEXT: S_BRANCH %bb.2
|
|
; CHECK-NEXT: {{ $}}
|
|
; CHECK-NEXT: bb.2:
|
|
; CHECK-NEXT: liveins: $agpr0_agpr1_agpr2_agpr3
|
|
; CHECK-NEXT: {{ $}}
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr0_vgpr1_vgpr2_vgpr3_vgpr4_vgpr5_vgpr6_vgpr7
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr8_vgpr9_vgpr10_vgpr11_vgpr12_vgpr13_vgpr14_vgpr15
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr16_vgpr17_vgpr18_vgpr19_vgpr20_vgpr21_vgpr22_vgpr23
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr24_vgpr25_vgpr26_vgpr27_vgpr28_vgpr29_vgpr30_vgpr31
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr32_vgpr33_vgpr34_vgpr35_vgpr36_vgpr37_vgpr38_vgpr39
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr40_vgpr41_vgpr42_vgpr43_vgpr44_vgpr45_vgpr46_vgpr47
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr48_vgpr49_vgpr50_vgpr51_vgpr52_vgpr53_vgpr54_vgpr55
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr56_vgpr57_vgpr58_vgpr59_vgpr60_vgpr61_vgpr62_vgpr63
|
|
; CHECK-NEXT: renamable $vgpr0 = V_MOV_B32_e32 0, implicit $exec
|
|
; CHECK-NEXT: GLOBAL_STORE_DWORDX4_SADDR killed renamable $vgpr0, killed renamable $agpr0_agpr1_agpr2_agpr3, undef $sgpr0_sgpr1, 32, 0, implicit $exec :: (store (s128), align 32, addrspace 1)
|
|
; CHECK-NEXT: S_ENDPGM 0
|
|
bb.0:
|
|
S_NOP 0, implicit-def $agpr0
|
|
renamable $sgpr0 = S_MOV_B32 0
|
|
undef %0.sub8:vreg_512_align2 = V_MOV_B32_e32 0, implicit $exec
|
|
renamable $sgpr1 = COPY renamable $sgpr0
|
|
%1:vreg_64_align2 = COPY killed renamable $sgpr0_sgpr1
|
|
renamable $vcc = S_AND_B64 $exec, -1, implicit-def dead $scc
|
|
%0.sub9:vreg_512_align2 = COPY %0.sub8
|
|
|
|
bb.1:
|
|
liveins: $vcc
|
|
|
|
undef %0.sub0_sub1:vreg_512_align2 = GLOBAL_LOAD_DWORDX2 undef %2:vreg_64_align2, 0, 0, implicit $exec :: (load (s64), addrspace 1)
|
|
%3:vreg_512_align2 = V_MFMA_F32_32X32X8F16_vgprcd_e64 %1, %1, %0, 0, 0, 0, implicit $mode, implicit $exec
|
|
%4:vreg_128_align2 = V_MFMA_F32_16X16X16F16_vgprcd_e64 %1, %1, %3.sub2_sub3_sub4_sub5, 0, 0, 0, implicit $mode, implicit $exec
|
|
S_CBRANCH_VCCNZ %bb.1, implicit $vcc
|
|
S_BRANCH %bb.2
|
|
|
|
bb.2:
|
|
; No VGPRs available for %0 or %4
|
|
S_NOP 0, implicit-def $vgpr0_vgpr1_vgpr2_vgpr3_vgpr4_vgpr5_vgpr6_vgpr7
|
|
S_NOP 0, implicit-def $vgpr8_vgpr9_vgpr10_vgpr11_vgpr12_vgpr13_vgpr14_vgpr15
|
|
S_NOP 0, implicit-def $vgpr16_vgpr17_vgpr18_vgpr19_vgpr20_vgpr21_vgpr22_vgpr23
|
|
S_NOP 0, implicit-def $vgpr24_vgpr25_vgpr26_vgpr27_vgpr28_vgpr29_vgpr30_vgpr31
|
|
S_NOP 0, implicit-def $vgpr32_vgpr33_vgpr34_vgpr35_vgpr36_vgpr37_vgpr38_vgpr39
|
|
S_NOP 0, implicit-def $vgpr40_vgpr41_vgpr42_vgpr43_vgpr44_vgpr45_vgpr46_vgpr47
|
|
S_NOP 0, implicit-def $vgpr48_vgpr49_vgpr50_vgpr51_vgpr52_vgpr53_vgpr54_vgpr55
|
|
S_NOP 0, implicit-def $vgpr56_vgpr57_vgpr58_vgpr59_vgpr60_vgpr61_vgpr62_vgpr63
|
|
%5:vgpr_32 = V_MOV_B32_e32 0, implicit $exec
|
|
GLOBAL_STORE_DWORDX4_SADDR %5, %4, undef $sgpr0_sgpr1, 32, 0, implicit $exec :: (store (s128), align 32, addrspace 1)
|
|
S_ENDPGM 0
|
|
|
|
...
|
|
|
|
---
|
|
name: inflate_result_to_agpr__V_MFMA_F32_32X32X8F16_mac_vgprcd_e64_local_split
|
|
tracksRegLiveness: true
|
|
machineFunctionInfo:
|
|
isEntryFunction: true
|
|
stackPtrOffsetReg: '$sgpr32'
|
|
occupancy: 10
|
|
sgprForEXECCopy: '$sgpr100_sgpr101'
|
|
body: |
|
|
bb.0:
|
|
; CHECK-LABEL: name: inflate_result_to_agpr__V_MFMA_F32_32X32X8F16_mac_vgprcd_e64_local_split
|
|
; CHECK: S_NOP 0, implicit-def $agpr0
|
|
; CHECK-NEXT: renamable $sgpr0 = S_MOV_B32 0
|
|
; CHECK-NEXT: renamable $vgpr8 = V_MOV_B32_e32 0, implicit $exec
|
|
; CHECK-NEXT: renamable $sgpr1 = COPY renamable $sgpr0
|
|
; CHECK-NEXT: renamable $vgpr0_vgpr1 = COPY killed renamable $sgpr0_sgpr1
|
|
; CHECK-NEXT: renamable $vcc = S_AND_B64 $exec, -1, implicit-def dead $scc
|
|
; CHECK-NEXT: dead renamable $vgpr9 = COPY renamable $vgpr8
|
|
; CHECK-NEXT: renamable $agpr0_agpr1 = GLOBAL_LOAD_DWORDX2 undef renamable $vgpr0_vgpr1, 0, 0, implicit $exec :: (load (s64), addrspace 1)
|
|
; CHECK-NEXT: renamable $agpr0_agpr1_agpr2_agpr3_agpr4_agpr5_agpr6_agpr7_agpr8_agpr9_agpr10_agpr11_agpr12_agpr13_agpr14_agpr15 = V_MFMA_F32_32X32X8F16_mac_e64 killed $vgpr0_vgpr1, $vgpr0_vgpr1, $agpr0_agpr1_agpr2_agpr3_agpr4_agpr5_agpr6_agpr7_agpr8_agpr9_agpr10_agpr11_agpr12_agpr13_agpr14_agpr15, 0, 0, 0, implicit $mode, implicit $exec
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr0_vgpr1_vgpr2_vgpr3_vgpr4_vgpr5_vgpr6_vgpr7
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr8_vgpr9_vgpr10_vgpr11_vgpr12_vgpr13_vgpr14_vgpr15
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr16_vgpr17_vgpr18_vgpr19_vgpr20_vgpr21_vgpr22_vgpr23
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr24_vgpr25_vgpr26_vgpr27_vgpr28_vgpr29_vgpr30_vgpr31
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr32_vgpr33_vgpr34_vgpr35_vgpr36_vgpr37_vgpr38_vgpr39
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr40_vgpr41_vgpr42_vgpr43_vgpr44_vgpr45_vgpr46_vgpr47
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr48_vgpr49_vgpr50_vgpr51_vgpr52_vgpr53_vgpr54_vgpr55
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr56_vgpr57_vgpr58_vgpr59_vgpr60_vgpr61_vgpr62_vgpr63
|
|
; CHECK-NEXT: renamable $vgpr0 = V_MOV_B32_e32 0, implicit $exec
|
|
; CHECK-NEXT: renamable $vgpr2_vgpr3_vgpr4_vgpr5_vgpr6_vgpr7_vgpr8_vgpr9_vgpr10_vgpr11_vgpr12_vgpr13_vgpr14_vgpr15_vgpr16_vgpr17 = COPY killed renamable $agpr0_agpr1_agpr2_agpr3_agpr4_agpr5_agpr6_agpr7_agpr8_agpr9_agpr10_agpr11_agpr12_agpr13_agpr14_agpr15
|
|
; CHECK-NEXT: GLOBAL_STORE_DWORDX4_SADDR renamable $vgpr0, renamable $vgpr10_vgpr11_vgpr12_vgpr13, undef $sgpr0_sgpr1, 32, 0, implicit $exec :: (store (s128), align 32, addrspace 1)
|
|
; CHECK-NEXT: GLOBAL_STORE_DWORDX4_SADDR renamable $vgpr0, renamable $vgpr14_vgpr15_vgpr16_vgpr17, undef $sgpr0_sgpr1, 48, 0, implicit $exec :: (store (s128), addrspace 1)
|
|
; CHECK-NEXT: GLOBAL_STORE_DWORDX4_SADDR renamable $vgpr0, renamable $vgpr2_vgpr3_vgpr4_vgpr5, undef $sgpr0_sgpr1, 0, 0, implicit $exec :: (store (s128), align 128, addrspace 1)
|
|
; CHECK-NEXT: GLOBAL_STORE_DWORDX4_SADDR killed renamable $vgpr0, killed renamable $vgpr6_vgpr7_vgpr8_vgpr9, killed undef $sgpr0_sgpr1, 16, 0, implicit $exec :: (store (s128), addrspace 1)
|
|
; CHECK-NEXT: S_ENDPGM 0
|
|
S_NOP 0, implicit-def $agpr0
|
|
renamable $sgpr0 = S_MOV_B32 0
|
|
undef %0.sub8:vreg_512_align2 = V_MOV_B32_e32 0, implicit $exec
|
|
renamable $sgpr1 = COPY renamable $sgpr0
|
|
%1:vreg_64_align2 = COPY killed renamable $sgpr0_sgpr1
|
|
renamable $vcc = S_AND_B64 $exec, -1, implicit-def dead $scc
|
|
%0.sub9:vreg_512_align2 = COPY %0.sub8
|
|
undef %0.sub0_sub1:vreg_512_align2 = GLOBAL_LOAD_DWORDX2 undef %3:vreg_64_align2, 0, 0, implicit $exec :: (load (s64), addrspace 1)
|
|
%0:vreg_512_align2 = V_MFMA_F32_32X32X8F16_mac_vgprcd_e64 %1, %1, %0, 0, 0, 0, implicit $mode, implicit $exec
|
|
; No VGPRs available for %0
|
|
S_NOP 0, implicit-def $vgpr0_vgpr1_vgpr2_vgpr3_vgpr4_vgpr5_vgpr6_vgpr7
|
|
S_NOP 0, implicit-def $vgpr8_vgpr9_vgpr10_vgpr11_vgpr12_vgpr13_vgpr14_vgpr15
|
|
S_NOP 0, implicit-def $vgpr16_vgpr17_vgpr18_vgpr19_vgpr20_vgpr21_vgpr22_vgpr23
|
|
S_NOP 0, implicit-def $vgpr24_vgpr25_vgpr26_vgpr27_vgpr28_vgpr29_vgpr30_vgpr31
|
|
S_NOP 0, implicit-def $vgpr32_vgpr33_vgpr34_vgpr35_vgpr36_vgpr37_vgpr38_vgpr39
|
|
S_NOP 0, implicit-def $vgpr40_vgpr41_vgpr42_vgpr43_vgpr44_vgpr45_vgpr46_vgpr47
|
|
S_NOP 0, implicit-def $vgpr48_vgpr49_vgpr50_vgpr51_vgpr52_vgpr53_vgpr54_vgpr55
|
|
S_NOP 0, implicit-def $vgpr56_vgpr57_vgpr58_vgpr59_vgpr60_vgpr61_vgpr62_vgpr63
|
|
%2:vgpr_32 = V_MOV_B32_e32 0, implicit $exec
|
|
GLOBAL_STORE_DWORDX4_SADDR %2, %0.sub8_sub9_sub10_sub11, undef $sgpr0_sgpr1, 32, 0, implicit $exec :: (store (s128), align 32, addrspace 1)
|
|
GLOBAL_STORE_DWORDX4_SADDR %2, %0.sub12_sub13_sub14_sub15, undef $sgpr0_sgpr1, 48, 0, implicit $exec :: (store (s128), addrspace 1)
|
|
GLOBAL_STORE_DWORDX4_SADDR %2, %0.sub0_sub1_sub2_sub3, undef $sgpr0_sgpr1, 0, 0, implicit $exec :: (store (s128), align 128, addrspace 1)
|
|
GLOBAL_STORE_DWORDX4_SADDR %2, %0.sub4_sub5_sub6_sub7, killed undef $sgpr0_sgpr1, 16, 0, implicit $exec :: (store (s128), addrspace 1)
|
|
S_ENDPGM 0
|
|
|
|
...
|
|
|
|
# Performs a split and inflate around the single instruction
|
|
---
|
|
name: inflate_result_to_agpr__V_MFMA_F32_32X32X8F16_mac_vgprcd_e64_instruction_split
|
|
tracksRegLiveness: true
|
|
machineFunctionInfo:
|
|
isEntryFunction: true
|
|
stackPtrOffsetReg: '$sgpr32'
|
|
occupancy: 10
|
|
sgprForEXECCopy: '$sgpr100_sgpr101'
|
|
body: |
|
|
bb.0:
|
|
; CHECK-LABEL: name: inflate_result_to_agpr__V_MFMA_F32_32X32X8F16_mac_vgprcd_e64_instruction_split
|
|
; CHECK: S_NOP 0, implicit-def $agpr0
|
|
; CHECK-NEXT: renamable $sgpr0 = S_MOV_B32 0
|
|
; CHECK-NEXT: renamable $vgpr8 = V_MOV_B32_e32 0, implicit $exec
|
|
; CHECK-NEXT: renamable $sgpr1 = COPY renamable $sgpr0
|
|
; CHECK-NEXT: renamable $vgpr0_vgpr1 = COPY killed renamable $sgpr0_sgpr1
|
|
; CHECK-NEXT: SI_SPILL_AV64_SAVE killed $vgpr0_vgpr1, %stack.0, $sgpr32, 0, implicit $exec :: (store (s64) into %stack.0, align 4, addrspace 5)
|
|
; CHECK-NEXT: renamable $vcc = S_AND_B64 $exec, -1, implicit-def dead $scc
|
|
; CHECK-NEXT: dead renamable $vgpr9 = COPY renamable $vgpr8
|
|
; CHECK-NEXT: renamable $agpr0_agpr1 = GLOBAL_LOAD_DWORDX2 undef renamable $vgpr0_vgpr1, 0, 0, implicit $exec :: (load (s64), addrspace 1)
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr0_vgpr1_vgpr2_vgpr3_vgpr4_vgpr5_vgpr6_vgpr7
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr8_vgpr9_vgpr10_vgpr11_vgpr12_vgpr13_vgpr14_vgpr15
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr16_vgpr17_vgpr18_vgpr19_vgpr20_vgpr21_vgpr22_vgpr23
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr24_vgpr25_vgpr26_vgpr27_vgpr28_vgpr29_vgpr30_vgpr31
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr32_vgpr33_vgpr34_vgpr35_vgpr36_vgpr37_vgpr38_vgpr39
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr40_vgpr41_vgpr42_vgpr43_vgpr44_vgpr45_vgpr46_vgpr47
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr48_vgpr49_vgpr50_vgpr51_vgpr52_vgpr53_vgpr54_vgpr55
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr56_vgpr57_vgpr58_vgpr59_vgpr60_vgpr61_vgpr62_vgpr63
|
|
; CHECK-NEXT: renamable $vgpr2_vgpr3 = SI_SPILL_AV64_RESTORE %stack.0, $sgpr32, 0, implicit $exec :: (load (s64) from %stack.0, align 4, addrspace 5)
|
|
; CHECK-NEXT: renamable $agpr0_agpr1_agpr2_agpr3_agpr4_agpr5_agpr6_agpr7_agpr8_agpr9_agpr10_agpr11_agpr12_agpr13_agpr14_agpr15 = V_MFMA_F32_32X32X8F16_mac_e64 killed $vgpr2_vgpr3, $vgpr2_vgpr3, $agpr0_agpr1_agpr2_agpr3_agpr4_agpr5_agpr6_agpr7_agpr8_agpr9_agpr10_agpr11_agpr12_agpr13_agpr14_agpr15, 0, 0, 0, implicit $mode, implicit $exec
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr0_vgpr1_vgpr2_vgpr3_vgpr4_vgpr5_vgpr6_vgpr7
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr8_vgpr9_vgpr10_vgpr11_vgpr12_vgpr13_vgpr14_vgpr15
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr16_vgpr17_vgpr18_vgpr19_vgpr20_vgpr21_vgpr22_vgpr23
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr24_vgpr25_vgpr26_vgpr27_vgpr28_vgpr29_vgpr30_vgpr31
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr32_vgpr33_vgpr34_vgpr35_vgpr36_vgpr37_vgpr38_vgpr39
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr40_vgpr41_vgpr42_vgpr43_vgpr44_vgpr45_vgpr46_vgpr47
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr48_vgpr49_vgpr50_vgpr51_vgpr52_vgpr53_vgpr54_vgpr55
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr56_vgpr57_vgpr58_vgpr59_vgpr60_vgpr61_vgpr62_vgpr63
|
|
; CHECK-NEXT: renamable $vgpr0 = V_MOV_B32_e32 0, implicit $exec
|
|
; CHECK-NEXT: renamable $vgpr2_vgpr3_vgpr4_vgpr5_vgpr6_vgpr7_vgpr8_vgpr9_vgpr10_vgpr11_vgpr12_vgpr13_vgpr14_vgpr15_vgpr16_vgpr17 = COPY killed renamable $agpr0_agpr1_agpr2_agpr3_agpr4_agpr5_agpr6_agpr7_agpr8_agpr9_agpr10_agpr11_agpr12_agpr13_agpr14_agpr15
|
|
; CHECK-NEXT: GLOBAL_STORE_DWORDX4_SADDR renamable $vgpr0, renamable $vgpr10_vgpr11_vgpr12_vgpr13, undef $sgpr0_sgpr1, 32, 0, implicit $exec :: (store (s128), align 32, addrspace 1)
|
|
; CHECK-NEXT: GLOBAL_STORE_DWORDX4_SADDR renamable $vgpr0, renamable $vgpr14_vgpr15_vgpr16_vgpr17, undef $sgpr0_sgpr1, 48, 0, implicit $exec :: (store (s128), addrspace 1)
|
|
; CHECK-NEXT: GLOBAL_STORE_DWORDX4_SADDR renamable $vgpr0, renamable $vgpr2_vgpr3_vgpr4_vgpr5, undef $sgpr0_sgpr1, 0, 0, implicit $exec :: (store (s128), align 128, addrspace 1)
|
|
; CHECK-NEXT: GLOBAL_STORE_DWORDX4_SADDR killed renamable $vgpr0, killed renamable $vgpr6_vgpr7_vgpr8_vgpr9, killed undef $sgpr0_sgpr1, 16, 0, implicit $exec :: (store (s128), addrspace 1)
|
|
; CHECK-NEXT: S_ENDPGM 0
|
|
S_NOP 0, implicit-def $agpr0
|
|
renamable $sgpr0 = S_MOV_B32 0
|
|
undef %0.sub8:vreg_512_align2 = V_MOV_B32_e32 0, implicit $exec
|
|
renamable $sgpr1 = COPY renamable $sgpr0
|
|
%1:vreg_64_align2 = COPY killed renamable $sgpr0_sgpr1
|
|
renamable $vcc = S_AND_B64 $exec, -1, implicit-def dead $scc
|
|
%0.sub9:vreg_512_align2 = COPY %0.sub8
|
|
undef %0.sub0_sub1:vreg_512_align2 = GLOBAL_LOAD_DWORDX2 undef %3:vreg_64_align2, 0, 0, implicit $exec :: (load (s64), addrspace 1)
|
|
S_NOP 0, implicit-def $vgpr0_vgpr1_vgpr2_vgpr3_vgpr4_vgpr5_vgpr6_vgpr7
|
|
S_NOP 0, implicit-def $vgpr8_vgpr9_vgpr10_vgpr11_vgpr12_vgpr13_vgpr14_vgpr15
|
|
S_NOP 0, implicit-def $vgpr16_vgpr17_vgpr18_vgpr19_vgpr20_vgpr21_vgpr22_vgpr23
|
|
S_NOP 0, implicit-def $vgpr24_vgpr25_vgpr26_vgpr27_vgpr28_vgpr29_vgpr30_vgpr31
|
|
S_NOP 0, implicit-def $vgpr32_vgpr33_vgpr34_vgpr35_vgpr36_vgpr37_vgpr38_vgpr39
|
|
S_NOP 0, implicit-def $vgpr40_vgpr41_vgpr42_vgpr43_vgpr44_vgpr45_vgpr46_vgpr47
|
|
S_NOP 0, implicit-def $vgpr48_vgpr49_vgpr50_vgpr51_vgpr52_vgpr53_vgpr54_vgpr55
|
|
S_NOP 0, implicit-def $vgpr56_vgpr57_vgpr58_vgpr59_vgpr60_vgpr61_vgpr62_vgpr63
|
|
%0:vreg_512_align2 = V_MFMA_F32_32X32X8F16_mac_vgprcd_e64 %1, %1, %0, 0, 0, 0, implicit $mode, implicit $exec
|
|
; No VGPRs available for %0
|
|
S_NOP 0, implicit-def $vgpr0_vgpr1_vgpr2_vgpr3_vgpr4_vgpr5_vgpr6_vgpr7
|
|
S_NOP 0, implicit-def $vgpr8_vgpr9_vgpr10_vgpr11_vgpr12_vgpr13_vgpr14_vgpr15
|
|
S_NOP 0, implicit-def $vgpr16_vgpr17_vgpr18_vgpr19_vgpr20_vgpr21_vgpr22_vgpr23
|
|
S_NOP 0, implicit-def $vgpr24_vgpr25_vgpr26_vgpr27_vgpr28_vgpr29_vgpr30_vgpr31
|
|
S_NOP 0, implicit-def $vgpr32_vgpr33_vgpr34_vgpr35_vgpr36_vgpr37_vgpr38_vgpr39
|
|
S_NOP 0, implicit-def $vgpr40_vgpr41_vgpr42_vgpr43_vgpr44_vgpr45_vgpr46_vgpr47
|
|
S_NOP 0, implicit-def $vgpr48_vgpr49_vgpr50_vgpr51_vgpr52_vgpr53_vgpr54_vgpr55
|
|
S_NOP 0, implicit-def $vgpr56_vgpr57_vgpr58_vgpr59_vgpr60_vgpr61_vgpr62_vgpr63
|
|
%2:vgpr_32 = V_MOV_B32_e32 0, implicit $exec
|
|
GLOBAL_STORE_DWORDX4_SADDR %2, %0.sub8_sub9_sub10_sub11, undef $sgpr0_sgpr1, 32, 0, implicit $exec :: (store (s128), align 32, addrspace 1)
|
|
GLOBAL_STORE_DWORDX4_SADDR %2, %0.sub12_sub13_sub14_sub15, undef $sgpr0_sgpr1, 48, 0, implicit $exec :: (store (s128), addrspace 1)
|
|
GLOBAL_STORE_DWORDX4_SADDR %2, %0.sub0_sub1_sub2_sub3, undef $sgpr0_sgpr1, 0, 0, implicit $exec :: (store (s128), align 128, addrspace 1)
|
|
GLOBAL_STORE_DWORDX4_SADDR %2, %0.sub4_sub5_sub6_sub7, killed undef $sgpr0_sgpr1, 16, 0, implicit $exec :: (store (s128), addrspace 1)
|
|
S_ENDPGM 0
|
|
|
|
...
|
|
|
|
# Performs a split and inflate around the single instruction, non-tied case
|
|
---
|
|
name: inflate_result_to_agpr__V_MFMA_F32_32X32X8F16_vgprcd_e64_instruction_split
|
|
tracksRegLiveness: true
|
|
machineFunctionInfo:
|
|
isEntryFunction: true
|
|
stackPtrOffsetReg: '$sgpr32'
|
|
occupancy: 10
|
|
sgprForEXECCopy: '$sgpr100_sgpr101'
|
|
body: |
|
|
bb.0:
|
|
; CHECK-LABEL: name: inflate_result_to_agpr__V_MFMA_F32_32X32X8F16_vgprcd_e64_instruction_split
|
|
; CHECK: S_NOP 0, implicit-def $agpr0
|
|
; CHECK-NEXT: renamable $sgpr0 = S_MOV_B32 0
|
|
; CHECK-NEXT: renamable $vgpr8 = V_MOV_B32_e32 0, implicit $exec
|
|
; CHECK-NEXT: renamable $sgpr1 = COPY renamable $sgpr0
|
|
; CHECK-NEXT: renamable $vgpr0_vgpr1 = COPY killed renamable $sgpr0_sgpr1
|
|
; CHECK-NEXT: SI_SPILL_AV64_SAVE killed $vgpr0_vgpr1, %stack.0, $sgpr32, 0, implicit $exec :: (store (s64) into %stack.0, align 4, addrspace 5)
|
|
; CHECK-NEXT: renamable $vcc = S_AND_B64 $exec, -1, implicit-def dead $scc
|
|
; CHECK-NEXT: dead renamable $vgpr9 = COPY renamable $vgpr8
|
|
; CHECK-NEXT: renamable $agpr0_agpr1 = GLOBAL_LOAD_DWORDX2 undef renamable $vgpr0_vgpr1, 0, 0, implicit $exec :: (load (s64), addrspace 1)
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr0_vgpr1_vgpr2_vgpr3_vgpr4_vgpr5_vgpr6_vgpr7
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr8_vgpr9_vgpr10_vgpr11_vgpr12_vgpr13_vgpr14_vgpr15
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr16_vgpr17_vgpr18_vgpr19_vgpr20_vgpr21_vgpr22_vgpr23
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr24_vgpr25_vgpr26_vgpr27_vgpr28_vgpr29_vgpr30_vgpr31
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr32_vgpr33_vgpr34_vgpr35_vgpr36_vgpr37_vgpr38_vgpr39
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr40_vgpr41_vgpr42_vgpr43_vgpr44_vgpr45_vgpr46_vgpr47
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr48_vgpr49_vgpr50_vgpr51_vgpr52_vgpr53_vgpr54_vgpr55
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr56_vgpr57_vgpr58_vgpr59_vgpr60_vgpr61_vgpr62_vgpr63
|
|
; CHECK-NEXT: renamable $vgpr0_vgpr1 = SI_SPILL_AV64_RESTORE %stack.0, $sgpr32, 0, implicit $exec :: (load (s64) from %stack.0, align 4, addrspace 5)
|
|
; CHECK-NEXT: renamable $agpr16_agpr17 = COPY killed renamable $agpr0_agpr1
|
|
; CHECK-NEXT: early-clobber renamable $agpr0_agpr1_agpr2_agpr3_agpr4_agpr5_agpr6_agpr7_agpr8_agpr9_agpr10_agpr11_agpr12_agpr13_agpr14_agpr15 = V_MFMA_F32_32X32X8F16_e64 killed $vgpr0_vgpr1, $vgpr0_vgpr1, $agpr16_agpr17_agpr18_agpr19_agpr20_agpr21_agpr22_agpr23_agpr24_agpr25_agpr26_agpr27_agpr28_agpr29_agpr30_agpr31, 0, 0, 0, implicit $mode, implicit $exec
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr0_vgpr1_vgpr2_vgpr3_vgpr4_vgpr5_vgpr6_vgpr7
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr8_vgpr9_vgpr10_vgpr11_vgpr12_vgpr13_vgpr14_vgpr15
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr16_vgpr17_vgpr18_vgpr19_vgpr20_vgpr21_vgpr22_vgpr23
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr24_vgpr25_vgpr26_vgpr27_vgpr28_vgpr29_vgpr30_vgpr31
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr32_vgpr33_vgpr34_vgpr35_vgpr36_vgpr37_vgpr38_vgpr39
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr40_vgpr41_vgpr42_vgpr43_vgpr44_vgpr45_vgpr46_vgpr47
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr48_vgpr49_vgpr50_vgpr51_vgpr52_vgpr53_vgpr54_vgpr55
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr56_vgpr57_vgpr58_vgpr59_vgpr60_vgpr61_vgpr62_vgpr63
|
|
; CHECK-NEXT: renamable $vgpr0 = V_MOV_B32_e32 0, implicit $exec
|
|
; CHECK-NEXT: renamable $vgpr2_vgpr3_vgpr4_vgpr5_vgpr6_vgpr7_vgpr8_vgpr9_vgpr10_vgpr11_vgpr12_vgpr13_vgpr14_vgpr15_vgpr16_vgpr17 = COPY killed renamable $agpr0_agpr1_agpr2_agpr3_agpr4_agpr5_agpr6_agpr7_agpr8_agpr9_agpr10_agpr11_agpr12_agpr13_agpr14_agpr15
|
|
; CHECK-NEXT: GLOBAL_STORE_DWORDX4_SADDR renamable $vgpr0, renamable $vgpr10_vgpr11_vgpr12_vgpr13, undef $sgpr0_sgpr1, 32, 0, implicit $exec :: (store (s128), align 32, addrspace 1)
|
|
; CHECK-NEXT: GLOBAL_STORE_DWORDX4_SADDR renamable $vgpr0, renamable $vgpr14_vgpr15_vgpr16_vgpr17, undef $sgpr0_sgpr1, 48, 0, implicit $exec :: (store (s128), addrspace 1)
|
|
; CHECK-NEXT: GLOBAL_STORE_DWORDX4_SADDR renamable $vgpr0, renamable $vgpr2_vgpr3_vgpr4_vgpr5, undef $sgpr0_sgpr1, 0, 0, implicit $exec :: (store (s128), align 128, addrspace 1)
|
|
; CHECK-NEXT: GLOBAL_STORE_DWORDX4_SADDR killed renamable $vgpr0, killed renamable $vgpr6_vgpr7_vgpr8_vgpr9, killed undef $sgpr0_sgpr1, 16, 0, implicit $exec :: (store (s128), addrspace 1)
|
|
; CHECK-NEXT: S_ENDPGM 0
|
|
S_NOP 0, implicit-def $agpr0
|
|
renamable $sgpr0 = S_MOV_B32 0
|
|
undef %0.sub8:vreg_512_align2 = V_MOV_B32_e32 0, implicit $exec
|
|
renamable $sgpr1 = COPY renamable $sgpr0
|
|
%1:vreg_64_align2 = COPY killed renamable $sgpr0_sgpr1
|
|
renamable $vcc = S_AND_B64 $exec, -1, implicit-def dead $scc
|
|
%0.sub9:vreg_512_align2 = COPY %0.sub8
|
|
undef %0.sub0_sub1:vreg_512_align2 = GLOBAL_LOAD_DWORDX2 undef %3:vreg_64_align2, 0, 0, implicit $exec :: (load (s64), addrspace 1)
|
|
S_NOP 0, implicit-def $vgpr0_vgpr1_vgpr2_vgpr3_vgpr4_vgpr5_vgpr6_vgpr7
|
|
S_NOP 0, implicit-def $vgpr8_vgpr9_vgpr10_vgpr11_vgpr12_vgpr13_vgpr14_vgpr15
|
|
S_NOP 0, implicit-def $vgpr16_vgpr17_vgpr18_vgpr19_vgpr20_vgpr21_vgpr22_vgpr23
|
|
S_NOP 0, implicit-def $vgpr24_vgpr25_vgpr26_vgpr27_vgpr28_vgpr29_vgpr30_vgpr31
|
|
S_NOP 0, implicit-def $vgpr32_vgpr33_vgpr34_vgpr35_vgpr36_vgpr37_vgpr38_vgpr39
|
|
S_NOP 0, implicit-def $vgpr40_vgpr41_vgpr42_vgpr43_vgpr44_vgpr45_vgpr46_vgpr47
|
|
S_NOP 0, implicit-def $vgpr48_vgpr49_vgpr50_vgpr51_vgpr52_vgpr53_vgpr54_vgpr55
|
|
S_NOP 0, implicit-def $vgpr56_vgpr57_vgpr58_vgpr59_vgpr60_vgpr61_vgpr62_vgpr63
|
|
%4:vreg_512_align2 = V_MFMA_F32_32X32X8F16_vgprcd_e64 %1, %1, %0, 0, 0, 0, implicit $mode, implicit $exec
|
|
; No VGPRs available for %0
|
|
S_NOP 0, implicit-def $vgpr0_vgpr1_vgpr2_vgpr3_vgpr4_vgpr5_vgpr6_vgpr7
|
|
S_NOP 0, implicit-def $vgpr8_vgpr9_vgpr10_vgpr11_vgpr12_vgpr13_vgpr14_vgpr15
|
|
S_NOP 0, implicit-def $vgpr16_vgpr17_vgpr18_vgpr19_vgpr20_vgpr21_vgpr22_vgpr23
|
|
S_NOP 0, implicit-def $vgpr24_vgpr25_vgpr26_vgpr27_vgpr28_vgpr29_vgpr30_vgpr31
|
|
S_NOP 0, implicit-def $vgpr32_vgpr33_vgpr34_vgpr35_vgpr36_vgpr37_vgpr38_vgpr39
|
|
S_NOP 0, implicit-def $vgpr40_vgpr41_vgpr42_vgpr43_vgpr44_vgpr45_vgpr46_vgpr47
|
|
S_NOP 0, implicit-def $vgpr48_vgpr49_vgpr50_vgpr51_vgpr52_vgpr53_vgpr54_vgpr55
|
|
S_NOP 0, implicit-def $vgpr56_vgpr57_vgpr58_vgpr59_vgpr60_vgpr61_vgpr62_vgpr63
|
|
%5:vgpr_32 = V_MOV_B32_e32 0, implicit $exec
|
|
GLOBAL_STORE_DWORDX4_SADDR %5, %4.sub8_sub9_sub10_sub11, undef $sgpr0_sgpr1, 32, 0, implicit $exec :: (store (s128), align 32, addrspace 1)
|
|
GLOBAL_STORE_DWORDX4_SADDR %5, %4.sub12_sub13_sub14_sub15, undef $sgpr0_sgpr1, 48, 0, implicit $exec :: (store (s128), addrspace 1)
|
|
GLOBAL_STORE_DWORDX4_SADDR %5, %4.sub0_sub1_sub2_sub3, undef $sgpr0_sgpr1, 0, 0, implicit $exec :: (store (s128), align 128, addrspace 1)
|
|
GLOBAL_STORE_DWORDX4_SADDR %5, %4.sub4_sub5_sub6_sub7, killed undef $sgpr0_sgpr1, 16, 0, implicit $exec :: (store (s128), addrspace 1)
|
|
S_ENDPGM 0
|
|
|
|
...
|
|
|
|
# This case does not fully use %0 after the MFMA. As a result,
|
|
# SplitKits insert a copy bundle for the subset of used lanes instead
|
|
# of a simple copy.
|
|
---
|
|
name: inflate_result_to_agpr__V_MFMA_F32_32X32X8F16_mac_vgprcd_e64_instruction_split_partial_uses_only
|
|
tracksRegLiveness: true
|
|
machineFunctionInfo:
|
|
isEntryFunction: true
|
|
stackPtrOffsetReg: '$sgpr32'
|
|
occupancy: 10
|
|
sgprForEXECCopy: '$sgpr100_sgpr101'
|
|
body: |
|
|
bb.0:
|
|
; CHECK-LABEL: name: inflate_result_to_agpr__V_MFMA_F32_32X32X8F16_mac_vgprcd_e64_instruction_split_partial_uses_only
|
|
; CHECK: S_NOP 0, implicit-def $agpr0
|
|
; CHECK-NEXT: renamable $sgpr0 = S_MOV_B32 0
|
|
; CHECK-NEXT: renamable $vgpr8 = V_MOV_B32_e32 0, implicit $exec
|
|
; CHECK-NEXT: renamable $sgpr1 = COPY renamable $sgpr0
|
|
; CHECK-NEXT: renamable $vgpr0_vgpr1 = COPY killed renamable $sgpr0_sgpr1
|
|
; CHECK-NEXT: SI_SPILL_AV64_SAVE killed $vgpr0_vgpr1, %stack.0, $sgpr32, 0, implicit $exec :: (store (s64) into %stack.0, align 4, addrspace 5)
|
|
; CHECK-NEXT: renamable $vcc = S_AND_B64 $exec, -1, implicit-def dead $scc
|
|
; CHECK-NEXT: dead renamable $vgpr9 = COPY renamable $vgpr8
|
|
; CHECK-NEXT: renamable $agpr0_agpr1 = GLOBAL_LOAD_DWORDX2 undef renamable $vgpr0_vgpr1, 0, 0, implicit $exec :: (load (s64), addrspace 1)
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr0_vgpr1_vgpr2_vgpr3_vgpr4_vgpr5_vgpr6_vgpr7
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr8_vgpr9_vgpr10_vgpr11_vgpr12_vgpr13_vgpr14_vgpr15
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr16_vgpr17_vgpr18_vgpr19_vgpr20_vgpr21_vgpr22_vgpr23
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr24_vgpr25_vgpr26_vgpr27_vgpr28_vgpr29_vgpr30_vgpr31
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr32_vgpr33_vgpr34_vgpr35_vgpr36_vgpr37_vgpr38_vgpr39
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr40_vgpr41_vgpr42_vgpr43_vgpr44_vgpr45_vgpr46_vgpr47
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr48_vgpr49_vgpr50_vgpr51_vgpr52_vgpr53_vgpr54_vgpr55
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr56_vgpr57_vgpr58_vgpr59_vgpr60_vgpr61_vgpr62_vgpr63
|
|
; CHECK-NEXT: renamable $vgpr2_vgpr3 = SI_SPILL_AV64_RESTORE %stack.0, $sgpr32, 0, implicit $exec :: (load (s64) from %stack.0, align 4, addrspace 5)
|
|
; CHECK-NEXT: renamable $agpr0_agpr1_agpr2_agpr3_agpr4_agpr5_agpr6_agpr7_agpr8_agpr9_agpr10_agpr11_agpr12_agpr13_agpr14_agpr15 = V_MFMA_F32_32X32X8F16_mac_e64 killed $vgpr2_vgpr3, $vgpr2_vgpr3, $agpr0_agpr1_agpr2_agpr3_agpr4_agpr5_agpr6_agpr7_agpr8_agpr9_agpr10_agpr11_agpr12_agpr13_agpr14_agpr15, 0, 0, 0, implicit $mode, implicit $exec
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr0_vgpr1_vgpr2_vgpr3_vgpr4_vgpr5_vgpr6_vgpr7
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr8_vgpr9_vgpr10_vgpr11_vgpr12_vgpr13_vgpr14_vgpr15
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr16_vgpr17_vgpr18_vgpr19_vgpr20_vgpr21_vgpr22_vgpr23
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr24_vgpr25_vgpr26_vgpr27_vgpr28_vgpr29_vgpr30_vgpr31
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr32_vgpr33_vgpr34_vgpr35_vgpr36_vgpr37_vgpr38_vgpr39
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr40_vgpr41_vgpr42_vgpr43_vgpr44_vgpr45_vgpr46_vgpr47
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr48_vgpr49_vgpr50_vgpr51_vgpr52_vgpr53_vgpr54_vgpr55
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr56_vgpr57_vgpr58_vgpr59_vgpr60_vgpr61_vgpr62_vgpr63
|
|
; CHECK-NEXT: renamable $vgpr0 = V_MOV_B32_e32 0, implicit $exec
|
|
; CHECK-NEXT: renamable $vgpr10_vgpr11_vgpr12_vgpr13_vgpr14_vgpr15_vgpr16_vgpr17 = COPY killed renamable $agpr8_agpr9_agpr10_agpr11_agpr12_agpr13_agpr14_agpr15
|
|
; CHECK-NEXT: renamable $vgpr2_vgpr3_vgpr4_vgpr5 = COPY renamable $agpr0_agpr1_agpr2_agpr3
|
|
; CHECK-NEXT: GLOBAL_STORE_DWORDX4_SADDR renamable $vgpr0, renamable $vgpr10_vgpr11_vgpr12_vgpr13, undef $sgpr0_sgpr1, 32, 0, implicit $exec :: (store (s128), align 32, addrspace 1)
|
|
; CHECK-NEXT: GLOBAL_STORE_DWORDX4_SADDR renamable $vgpr0, renamable $vgpr14_vgpr15_vgpr16_vgpr17, undef $sgpr0_sgpr1, 48, 0, implicit $exec :: (store (s128), addrspace 1)
|
|
; CHECK-NEXT: GLOBAL_STORE_DWORDX4_SADDR killed renamable $vgpr0, killed renamable $vgpr2_vgpr3_vgpr4_vgpr5, undef $sgpr0_sgpr1, 0, 0, implicit $exec :: (store (s128), align 128, addrspace 1)
|
|
; CHECK-NEXT: S_ENDPGM 0
|
|
S_NOP 0, implicit-def $agpr0
|
|
renamable $sgpr0 = S_MOV_B32 0
|
|
undef %0.sub8:vreg_512_align2 = V_MOV_B32_e32 0, implicit $exec
|
|
renamable $sgpr1 = COPY renamable $sgpr0
|
|
%1:vreg_64_align2 = COPY killed renamable $sgpr0_sgpr1
|
|
renamable $vcc = S_AND_B64 $exec, -1, implicit-def dead $scc
|
|
%0.sub9:vreg_512_align2 = COPY %0.sub8
|
|
undef %0.sub0_sub1:vreg_512_align2 = GLOBAL_LOAD_DWORDX2 undef %3:vreg_64_align2, 0, 0, implicit $exec :: (load (s64), addrspace 1)
|
|
S_NOP 0, implicit-def $vgpr0_vgpr1_vgpr2_vgpr3_vgpr4_vgpr5_vgpr6_vgpr7
|
|
S_NOP 0, implicit-def $vgpr8_vgpr9_vgpr10_vgpr11_vgpr12_vgpr13_vgpr14_vgpr15
|
|
S_NOP 0, implicit-def $vgpr16_vgpr17_vgpr18_vgpr19_vgpr20_vgpr21_vgpr22_vgpr23
|
|
S_NOP 0, implicit-def $vgpr24_vgpr25_vgpr26_vgpr27_vgpr28_vgpr29_vgpr30_vgpr31
|
|
S_NOP 0, implicit-def $vgpr32_vgpr33_vgpr34_vgpr35_vgpr36_vgpr37_vgpr38_vgpr39
|
|
S_NOP 0, implicit-def $vgpr40_vgpr41_vgpr42_vgpr43_vgpr44_vgpr45_vgpr46_vgpr47
|
|
S_NOP 0, implicit-def $vgpr48_vgpr49_vgpr50_vgpr51_vgpr52_vgpr53_vgpr54_vgpr55
|
|
S_NOP 0, implicit-def $vgpr56_vgpr57_vgpr58_vgpr59_vgpr60_vgpr61_vgpr62_vgpr63
|
|
%0:vreg_512_align2 = V_MFMA_F32_32X32X8F16_mac_vgprcd_e64 %1, %1, %0, 0, 0, 0, implicit $mode, implicit $exec
|
|
; No VGPRs available for %0
|
|
S_NOP 0, implicit-def $vgpr0_vgpr1_vgpr2_vgpr3_vgpr4_vgpr5_vgpr6_vgpr7
|
|
S_NOP 0, implicit-def $vgpr8_vgpr9_vgpr10_vgpr11_vgpr12_vgpr13_vgpr14_vgpr15
|
|
S_NOP 0, implicit-def $vgpr16_vgpr17_vgpr18_vgpr19_vgpr20_vgpr21_vgpr22_vgpr23
|
|
S_NOP 0, implicit-def $vgpr24_vgpr25_vgpr26_vgpr27_vgpr28_vgpr29_vgpr30_vgpr31
|
|
S_NOP 0, implicit-def $vgpr32_vgpr33_vgpr34_vgpr35_vgpr36_vgpr37_vgpr38_vgpr39
|
|
S_NOP 0, implicit-def $vgpr40_vgpr41_vgpr42_vgpr43_vgpr44_vgpr45_vgpr46_vgpr47
|
|
S_NOP 0, implicit-def $vgpr48_vgpr49_vgpr50_vgpr51_vgpr52_vgpr53_vgpr54_vgpr55
|
|
S_NOP 0, implicit-def $vgpr56_vgpr57_vgpr58_vgpr59_vgpr60_vgpr61_vgpr62_vgpr63
|
|
%2:vgpr_32 = V_MOV_B32_e32 0, implicit $exec
|
|
GLOBAL_STORE_DWORDX4_SADDR %2, %0.sub8_sub9_sub10_sub11, undef $sgpr0_sgpr1, 32, 0, implicit $exec :: (store (s128), align 32, addrspace 1)
|
|
GLOBAL_STORE_DWORDX4_SADDR %2, %0.sub12_sub13_sub14_sub15, undef $sgpr0_sgpr1, 48, 0, implicit $exec :: (store (s128), addrspace 1)
|
|
GLOBAL_STORE_DWORDX4_SADDR %2, %0.sub0_sub1_sub2_sub3, undef $sgpr0_sgpr1, 0, 0, implicit $exec :: (store (s128), align 128, addrspace 1)
|
|
S_ENDPGM 0
|
|
|
|
...
|
|
|
|
# Untied version of previous. This case does not fully use %4 after
|
|
# the MFMA. As a result, SplitKits insert a copy bundle for the subset
|
|
# of used lanes instead of a simple copy,
|
|
---
|
|
name: inflate_result_to_agpr__V_MFMA_F32_32X32X8F16_vgprcd_e64_instruction_split_partial_uses_only
|
|
tracksRegLiveness: true
|
|
machineFunctionInfo:
|
|
isEntryFunction: true
|
|
stackPtrOffsetReg: '$sgpr32'
|
|
occupancy: 10
|
|
sgprForEXECCopy: '$sgpr100_sgpr101'
|
|
body: |
|
|
bb.0:
|
|
; CHECK-LABEL: name: inflate_result_to_agpr__V_MFMA_F32_32X32X8F16_vgprcd_e64_instruction_split_partial_uses_only
|
|
; CHECK: S_NOP 0, implicit-def $agpr0
|
|
; CHECK-NEXT: renamable $sgpr0 = S_MOV_B32 0
|
|
; CHECK-NEXT: renamable $vgpr8 = V_MOV_B32_e32 0, implicit $exec
|
|
; CHECK-NEXT: renamable $sgpr1 = COPY renamable $sgpr0
|
|
; CHECK-NEXT: renamable $vgpr0_vgpr1 = COPY killed renamable $sgpr0_sgpr1
|
|
; CHECK-NEXT: SI_SPILL_AV64_SAVE killed $vgpr0_vgpr1, %stack.0, $sgpr32, 0, implicit $exec :: (store (s64) into %stack.0, align 4, addrspace 5)
|
|
; CHECK-NEXT: renamable $vcc = S_AND_B64 $exec, -1, implicit-def dead $scc
|
|
; CHECK-NEXT: dead renamable $vgpr9 = COPY renamable $vgpr8
|
|
; CHECK-NEXT: renamable $agpr0_agpr1 = GLOBAL_LOAD_DWORDX2 undef renamable $vgpr0_vgpr1, 0, 0, implicit $exec :: (load (s64), addrspace 1)
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr0_vgpr1_vgpr2_vgpr3_vgpr4_vgpr5_vgpr6_vgpr7
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr8_vgpr9_vgpr10_vgpr11_vgpr12_vgpr13_vgpr14_vgpr15
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr16_vgpr17_vgpr18_vgpr19_vgpr20_vgpr21_vgpr22_vgpr23
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr24_vgpr25_vgpr26_vgpr27_vgpr28_vgpr29_vgpr30_vgpr31
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr32_vgpr33_vgpr34_vgpr35_vgpr36_vgpr37_vgpr38_vgpr39
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr40_vgpr41_vgpr42_vgpr43_vgpr44_vgpr45_vgpr46_vgpr47
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr48_vgpr49_vgpr50_vgpr51_vgpr52_vgpr53_vgpr54_vgpr55
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr56_vgpr57_vgpr58_vgpr59_vgpr60_vgpr61_vgpr62_vgpr63
|
|
; CHECK-NEXT: renamable $vgpr0_vgpr1 = SI_SPILL_AV64_RESTORE %stack.0, $sgpr32, 0, implicit $exec :: (load (s64) from %stack.0, align 4, addrspace 5)
|
|
; CHECK-NEXT: renamable $agpr16_agpr17 = COPY killed renamable $agpr0_agpr1
|
|
; CHECK-NEXT: early-clobber renamable $agpr0_agpr1_agpr2_agpr3_agpr4_agpr5_agpr6_agpr7_agpr8_agpr9_agpr10_agpr11_agpr12_agpr13_agpr14_agpr15 = V_MFMA_F32_32X32X8F16_e64 killed $vgpr0_vgpr1, $vgpr0_vgpr1, $agpr16_agpr17_agpr18_agpr19_agpr20_agpr21_agpr22_agpr23_agpr24_agpr25_agpr26_agpr27_agpr28_agpr29_agpr30_agpr31, 0, 0, 0, implicit $mode, implicit $exec
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr0_vgpr1_vgpr2_vgpr3_vgpr4_vgpr5_vgpr6_vgpr7
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr8_vgpr9_vgpr10_vgpr11_vgpr12_vgpr13_vgpr14_vgpr15
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr16_vgpr17_vgpr18_vgpr19_vgpr20_vgpr21_vgpr22_vgpr23
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr24_vgpr25_vgpr26_vgpr27_vgpr28_vgpr29_vgpr30_vgpr31
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr32_vgpr33_vgpr34_vgpr35_vgpr36_vgpr37_vgpr38_vgpr39
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr40_vgpr41_vgpr42_vgpr43_vgpr44_vgpr45_vgpr46_vgpr47
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr48_vgpr49_vgpr50_vgpr51_vgpr52_vgpr53_vgpr54_vgpr55
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr56_vgpr57_vgpr58_vgpr59_vgpr60_vgpr61_vgpr62_vgpr63
|
|
; CHECK-NEXT: renamable $vgpr0 = V_MOV_B32_e32 0, implicit $exec
|
|
; CHECK-NEXT: renamable $vgpr10_vgpr11_vgpr12_vgpr13_vgpr14_vgpr15_vgpr16_vgpr17 = COPY killed renamable $agpr8_agpr9_agpr10_agpr11_agpr12_agpr13_agpr14_agpr15
|
|
; CHECK-NEXT: renamable $vgpr2_vgpr3_vgpr4_vgpr5 = COPY renamable $agpr0_agpr1_agpr2_agpr3
|
|
; CHECK-NEXT: GLOBAL_STORE_DWORDX4_SADDR renamable $vgpr0, renamable $vgpr10_vgpr11_vgpr12_vgpr13, undef $sgpr0_sgpr1, 32, 0, implicit $exec :: (store (s128), align 32, addrspace 1)
|
|
; CHECK-NEXT: GLOBAL_STORE_DWORDX4_SADDR renamable $vgpr0, renamable $vgpr14_vgpr15_vgpr16_vgpr17, undef $sgpr0_sgpr1, 48, 0, implicit $exec :: (store (s128), addrspace 1)
|
|
; CHECK-NEXT: GLOBAL_STORE_DWORDX4_SADDR killed renamable $vgpr0, killed renamable $vgpr2_vgpr3_vgpr4_vgpr5, undef $sgpr0_sgpr1, 0, 0, implicit $exec :: (store (s128), align 128, addrspace 1)
|
|
; CHECK-NEXT: S_ENDPGM 0
|
|
S_NOP 0, implicit-def $agpr0
|
|
renamable $sgpr0 = S_MOV_B32 0
|
|
undef %0.sub8:vreg_512_align2 = V_MOV_B32_e32 0, implicit $exec
|
|
renamable $sgpr1 = COPY renamable $sgpr0
|
|
%1:vreg_64_align2 = COPY killed renamable $sgpr0_sgpr1
|
|
renamable $vcc = S_AND_B64 $exec, -1, implicit-def dead $scc
|
|
%0.sub9:vreg_512_align2 = COPY %0.sub8
|
|
undef %0.sub0_sub1:vreg_512_align2 = GLOBAL_LOAD_DWORDX2 undef %3:vreg_64_align2, 0, 0, implicit $exec :: (load (s64), addrspace 1)
|
|
S_NOP 0, implicit-def $vgpr0_vgpr1_vgpr2_vgpr3_vgpr4_vgpr5_vgpr6_vgpr7
|
|
S_NOP 0, implicit-def $vgpr8_vgpr9_vgpr10_vgpr11_vgpr12_vgpr13_vgpr14_vgpr15
|
|
S_NOP 0, implicit-def $vgpr16_vgpr17_vgpr18_vgpr19_vgpr20_vgpr21_vgpr22_vgpr23
|
|
S_NOP 0, implicit-def $vgpr24_vgpr25_vgpr26_vgpr27_vgpr28_vgpr29_vgpr30_vgpr31
|
|
S_NOP 0, implicit-def $vgpr32_vgpr33_vgpr34_vgpr35_vgpr36_vgpr37_vgpr38_vgpr39
|
|
S_NOP 0, implicit-def $vgpr40_vgpr41_vgpr42_vgpr43_vgpr44_vgpr45_vgpr46_vgpr47
|
|
S_NOP 0, implicit-def $vgpr48_vgpr49_vgpr50_vgpr51_vgpr52_vgpr53_vgpr54_vgpr55
|
|
S_NOP 0, implicit-def $vgpr56_vgpr57_vgpr58_vgpr59_vgpr60_vgpr61_vgpr62_vgpr63
|
|
%4:vreg_512_align2 = V_MFMA_F32_32X32X8F16_vgprcd_e64 %1, %1, %0, 0, 0, 0, implicit $mode, implicit $exec
|
|
; No VGPRs available for %4
|
|
S_NOP 0, implicit-def $vgpr0_vgpr1_vgpr2_vgpr3_vgpr4_vgpr5_vgpr6_vgpr7
|
|
S_NOP 0, implicit-def $vgpr8_vgpr9_vgpr10_vgpr11_vgpr12_vgpr13_vgpr14_vgpr15
|
|
S_NOP 0, implicit-def $vgpr16_vgpr17_vgpr18_vgpr19_vgpr20_vgpr21_vgpr22_vgpr23
|
|
S_NOP 0, implicit-def $vgpr24_vgpr25_vgpr26_vgpr27_vgpr28_vgpr29_vgpr30_vgpr31
|
|
S_NOP 0, implicit-def $vgpr32_vgpr33_vgpr34_vgpr35_vgpr36_vgpr37_vgpr38_vgpr39
|
|
S_NOP 0, implicit-def $vgpr40_vgpr41_vgpr42_vgpr43_vgpr44_vgpr45_vgpr46_vgpr47
|
|
S_NOP 0, implicit-def $vgpr48_vgpr49_vgpr50_vgpr51_vgpr52_vgpr53_vgpr54_vgpr55
|
|
S_NOP 0, implicit-def $vgpr56_vgpr57_vgpr58_vgpr59_vgpr60_vgpr61_vgpr62_vgpr63
|
|
%5:vgpr_32 = V_MOV_B32_e32 0, implicit $exec
|
|
GLOBAL_STORE_DWORDX4_SADDR %5, %4.sub8_sub9_sub10_sub11, undef $sgpr0_sgpr1, 32, 0, implicit $exec :: (store (s128), align 32, addrspace 1)
|
|
GLOBAL_STORE_DWORDX4_SADDR %5, %4.sub12_sub13_sub14_sub15, undef $sgpr0_sgpr1, 48, 0, implicit $exec :: (store (s128), addrspace 1)
|
|
GLOBAL_STORE_DWORDX4_SADDR %5, %4.sub0_sub1_sub2_sub3, undef $sgpr0_sgpr1, 0, 0, implicit $exec :: (store (s128), align 128, addrspace 1)
|
|
S_ENDPGM 0
|
|
|
|
...
|
|
|
|
---
|
|
name: inflate_result_to_agpr__V_MFMA_F32_32X32X8F16_mac_vgprcd_e64_same_subreg
|
|
tracksRegLiveness: true
|
|
machineFunctionInfo:
|
|
isEntryFunction: true
|
|
stackPtrOffsetReg: '$sgpr32'
|
|
occupancy: 10
|
|
sgprForEXECCopy: '$sgpr100_sgpr101'
|
|
body: |
|
|
; CHECK-LABEL: name: inflate_result_to_agpr__V_MFMA_F32_32X32X8F16_mac_vgprcd_e64_same_subreg
|
|
; CHECK: bb.0:
|
|
; CHECK-NEXT: successors: %bb.1(0x80000000)
|
|
; CHECK-NEXT: {{ $}}
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $agpr0
|
|
; CHECK-NEXT: renamable $sgpr0 = S_MOV_B32 0
|
|
; CHECK-NEXT: renamable $vgpr10 = V_MOV_B32_e32 0, implicit $exec
|
|
; CHECK-NEXT: renamable $sgpr1 = COPY renamable $sgpr0
|
|
; CHECK-NEXT: renamable $vgpr0_vgpr1 = COPY killed renamable $sgpr0_sgpr1
|
|
; CHECK-NEXT: renamable $vcc = S_AND_B64 $exec, -1, implicit-def dead $scc
|
|
; CHECK-NEXT: renamable $vgpr11 = COPY renamable $vgpr10
|
|
; CHECK-NEXT: {{ $}}
|
|
; CHECK-NEXT: bb.1:
|
|
; CHECK-NEXT: successors: %bb.1(0x40000000), %bb.2(0x40000000)
|
|
; CHECK-NEXT: liveins: $vcc, $vgpr0_vgpr1
|
|
; CHECK-NEXT: {{ $}}
|
|
; CHECK-NEXT: renamable $vgpr2_vgpr3_vgpr4_vgpr5 = GLOBAL_LOAD_DWORDX4 undef renamable $vgpr0_vgpr1, 0, 0, implicit $exec :: (load (s128), addrspace 1)
|
|
; CHECK-NEXT: renamable $vgpr6_vgpr7_vgpr8_vgpr9 = GLOBAL_LOAD_DWORDX4 undef renamable $vgpr0_vgpr1, 0, 0, implicit $exec :: (load (s128), addrspace 1)
|
|
; CHECK-NEXT: renamable $vgpr10_vgpr11_vgpr12_vgpr13 = GLOBAL_LOAD_DWORDX4 undef renamable $vgpr0_vgpr1, 0, 0, implicit $exec :: (load (s128), addrspace 1)
|
|
; CHECK-NEXT: renamable $vgpr14_vgpr15_vgpr16_vgpr17 = GLOBAL_LOAD_DWORDX4 undef renamable $vgpr0_vgpr1, 0, 0, implicit $exec :: (load (s128), addrspace 1)
|
|
; CHECK-NEXT: renamable $vgpr2_vgpr3_vgpr4_vgpr5_vgpr6_vgpr7_vgpr8_vgpr9_vgpr10_vgpr11_vgpr12_vgpr13_vgpr14_vgpr15_vgpr16_vgpr17 = V_MFMA_F32_32X32X8F16_mac_vgprcd_e64 $vgpr0_vgpr1, $vgpr0_vgpr1, $vgpr2_vgpr3_vgpr4_vgpr5_vgpr6_vgpr7_vgpr8_vgpr9_vgpr10_vgpr11_vgpr12_vgpr13_vgpr14_vgpr15_vgpr16_vgpr17, 0, 0, 0, implicit $mode, implicit $exec
|
|
; CHECK-NEXT: S_CBRANCH_VCCNZ %bb.1, implicit $vcc
|
|
; CHECK-NEXT: S_BRANCH %bb.2
|
|
; CHECK-NEXT: {{ $}}
|
|
; CHECK-NEXT: bb.2:
|
|
; CHECK-NEXT: liveins: $vgpr2_vgpr3_vgpr4_vgpr5_vgpr6_vgpr7_vgpr8_vgpr9_vgpr10_vgpr11_vgpr12_vgpr13_vgpr14_vgpr15_vgpr16_vgpr17_vgpr18_vgpr19_vgpr20_vgpr21_vgpr22_vgpr23_vgpr24_vgpr25_vgpr26_vgpr27_vgpr28_vgpr29_vgpr30_vgpr31_vgpr32_vgpr33:0x00000000FFFFFFFF
|
|
; CHECK-NEXT: {{ $}}
|
|
; CHECK-NEXT: renamable $agpr0_agpr1_agpr2_agpr3_agpr4_agpr5_agpr6_agpr7_agpr8_agpr9_agpr10_agpr11_agpr12_agpr13_agpr14_agpr15 = COPY killed renamable $vgpr2_vgpr3_vgpr4_vgpr5_vgpr6_vgpr7_vgpr8_vgpr9_vgpr10_vgpr11_vgpr12_vgpr13_vgpr14_vgpr15_vgpr16_vgpr17
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr0_vgpr1_vgpr2_vgpr3_vgpr4_vgpr5_vgpr6_vgpr7
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr8_vgpr9_vgpr10_vgpr11_vgpr12_vgpr13_vgpr14_vgpr15
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr16_vgpr17_vgpr18_vgpr19_vgpr20_vgpr21_vgpr22_vgpr23
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr24_vgpr25_vgpr26_vgpr27_vgpr28_vgpr29_vgpr30_vgpr31
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr32_vgpr33_vgpr34_vgpr35_vgpr36_vgpr37_vgpr38_vgpr39
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr40_vgpr41_vgpr42_vgpr43_vgpr44_vgpr45_vgpr46_vgpr47
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr48_vgpr49_vgpr50_vgpr51_vgpr52_vgpr53_vgpr54_vgpr55
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr56_vgpr57_vgpr58_vgpr59_vgpr60_vgpr61_vgpr62_vgpr63
|
|
; CHECK-NEXT: renamable $vgpr0 = V_MOV_B32_e32 0, implicit $exec
|
|
; CHECK-NEXT: GLOBAL_STORE_DWORDX4_SADDR renamable $vgpr0, renamable $agpr8_agpr9_agpr10_agpr11, undef $sgpr0_sgpr1, 32, 0, implicit $exec :: (store (s128), align 32, addrspace 1)
|
|
; CHECK-NEXT: GLOBAL_STORE_DWORDX4_SADDR renamable $vgpr0, renamable $agpr12_agpr13_agpr14_agpr15, undef $sgpr0_sgpr1, 48, 0, implicit $exec :: (store (s128), addrspace 1)
|
|
; CHECK-NEXT: GLOBAL_STORE_DWORDX4_SADDR renamable $vgpr0, renamable $agpr0_agpr1_agpr2_agpr3, undef $sgpr0_sgpr1, 0, 0, implicit $exec :: (store (s128), align 128, addrspace 1)
|
|
; CHECK-NEXT: GLOBAL_STORE_DWORDX4_SADDR killed renamable $vgpr0, killed renamable $agpr4_agpr5_agpr6_agpr7, killed undef $sgpr0_sgpr1, 16, 0, implicit $exec :: (store (s128), addrspace 1)
|
|
; CHECK-NEXT: S_ENDPGM 0
|
|
bb.0:
|
|
S_NOP 0, implicit-def $agpr0
|
|
renamable $sgpr0 = S_MOV_B32 0
|
|
undef %0.sub8:vreg_1024_align2 = V_MOV_B32_e32 0, implicit $exec
|
|
renamable $sgpr1 = COPY renamable $sgpr0
|
|
%1:vreg_64_align2 = COPY killed renamable $sgpr0_sgpr1
|
|
renamable $vcc = S_AND_B64 $exec, -1, implicit-def dead $scc
|
|
%0.sub9:vreg_1024_align2 = COPY %0.sub8
|
|
|
|
bb.1:
|
|
liveins: $vcc
|
|
|
|
%0.sub0_sub1_sub2_sub3:vreg_1024_align2 = GLOBAL_LOAD_DWORDX4 undef %3:vreg_64_align2, 0, 0, implicit $exec :: (load (s128), addrspace 1)
|
|
%0.sub4_sub5_sub6_sub7:vreg_1024_align2 = GLOBAL_LOAD_DWORDX4 undef %3:vreg_64_align2, 0, 0, implicit $exec :: (load (s128), addrspace 1)
|
|
%0.sub8_sub9_sub10_sub11:vreg_1024_align2 = GLOBAL_LOAD_DWORDX4 undef %3:vreg_64_align2, 0, 0, implicit $exec :: (load (s128), addrspace 1)
|
|
%0.sub12_sub13_sub14_sub15:vreg_1024_align2 = GLOBAL_LOAD_DWORDX4 undef %3:vreg_64_align2, 0, 0, implicit $exec :: (load (s128), addrspace 1)
|
|
%0.sub0_sub1_sub2_sub3_sub4_sub5_sub6_sub7_sub8_sub9_sub10_sub11_sub12_sub13_sub14_sub15:vreg_1024_align2 = V_MFMA_F32_32X32X8F16_mac_vgprcd_e64 %1, %1, %0.sub0_sub1_sub2_sub3_sub4_sub5_sub6_sub7_sub8_sub9_sub10_sub11_sub12_sub13_sub14_sub15, 0, 0, 0, implicit $mode, implicit $exec
|
|
S_CBRANCH_VCCNZ %bb.1, implicit $vcc
|
|
S_BRANCH %bb.2
|
|
|
|
bb.2:
|
|
; No VGPRs available for %0
|
|
S_NOP 0, implicit-def $vgpr0_vgpr1_vgpr2_vgpr3_vgpr4_vgpr5_vgpr6_vgpr7
|
|
S_NOP 0, implicit-def $vgpr8_vgpr9_vgpr10_vgpr11_vgpr12_vgpr13_vgpr14_vgpr15
|
|
S_NOP 0, implicit-def $vgpr16_vgpr17_vgpr18_vgpr19_vgpr20_vgpr21_vgpr22_vgpr23
|
|
S_NOP 0, implicit-def $vgpr24_vgpr25_vgpr26_vgpr27_vgpr28_vgpr29_vgpr30_vgpr31
|
|
S_NOP 0, implicit-def $vgpr32_vgpr33_vgpr34_vgpr35_vgpr36_vgpr37_vgpr38_vgpr39
|
|
S_NOP 0, implicit-def $vgpr40_vgpr41_vgpr42_vgpr43_vgpr44_vgpr45_vgpr46_vgpr47
|
|
S_NOP 0, implicit-def $vgpr48_vgpr49_vgpr50_vgpr51_vgpr52_vgpr53_vgpr54_vgpr55
|
|
S_NOP 0, implicit-def $vgpr56_vgpr57_vgpr58_vgpr59_vgpr60_vgpr61_vgpr62_vgpr63
|
|
%2:vgpr_32 = V_MOV_B32_e32 0, implicit $exec
|
|
GLOBAL_STORE_DWORDX4_SADDR %2, %0.sub8_sub9_sub10_sub11, undef $sgpr0_sgpr1, 32, 0, implicit $exec :: (store (s128), align 32, addrspace 1)
|
|
GLOBAL_STORE_DWORDX4_SADDR %2, %0.sub12_sub13_sub14_sub15, undef $sgpr0_sgpr1, 48, 0, implicit $exec :: (store (s128), addrspace 1)
|
|
GLOBAL_STORE_DWORDX4_SADDR %2, %0.sub0_sub1_sub2_sub3, undef $sgpr0_sgpr1, 0, 0, implicit $exec :: (store (s128), align 128, addrspace 1)
|
|
GLOBAL_STORE_DWORDX4_SADDR %2, %0.sub4_sub5_sub6_sub7, killed undef $sgpr0_sgpr1, 16, 0, implicit $exec :: (store (s128), addrspace 1)
|
|
S_ENDPGM 0
|
|
|
|
...
|
|
|
|
---
|
|
name: chained_mfma_dst_user_is_vgpr
|
|
tracksRegLiveness: true
|
|
machineFunctionInfo:
|
|
isEntryFunction: true
|
|
stackPtrOffsetReg: '$sgpr32'
|
|
occupancy: 10
|
|
sgprForEXECCopy: '$sgpr100_sgpr101'
|
|
body: |
|
|
; CHECK-LABEL: name: chained_mfma_dst_user_is_vgpr
|
|
; CHECK: bb.0:
|
|
; CHECK-NEXT: successors: %bb.1(0x80000000)
|
|
; CHECK-NEXT: {{ $}}
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $agpr0
|
|
; CHECK-NEXT: renamable $sgpr0 = S_MOV_B32 0
|
|
; CHECK-NEXT: renamable $vgpr8 = V_MOV_B32_e32 0, implicit $exec
|
|
; CHECK-NEXT: renamable $sgpr1 = COPY renamable $sgpr0
|
|
; CHECK-NEXT: renamable $vgpr16_vgpr17 = COPY killed renamable $sgpr0_sgpr1
|
|
; CHECK-NEXT: renamable $vcc = S_AND_B64 $exec, -1, implicit-def dead $scc
|
|
; CHECK-NEXT: dead renamable $vgpr9 = COPY renamable $vgpr8
|
|
; CHECK-NEXT: {{ $}}
|
|
; CHECK-NEXT: bb.1:
|
|
; CHECK-NEXT: successors: %bb.1(0x40000000), %bb.2(0x40000000)
|
|
; CHECK-NEXT: liveins: $vcc, $vgpr16_vgpr17
|
|
; CHECK-NEXT: {{ $}}
|
|
; CHECK-NEXT: renamable $vgpr0_vgpr1 = GLOBAL_LOAD_DWORDX2 undef renamable $vgpr0_vgpr1, 0, 0, implicit $exec :: (load (s64), addrspace 1)
|
|
; CHECK-NEXT: early-clobber renamable $vgpr18_vgpr19_vgpr20_vgpr21_vgpr22_vgpr23_vgpr24_vgpr25_vgpr26_vgpr27_vgpr28_vgpr29_vgpr30_vgpr31_vgpr32_vgpr33 = V_MFMA_F32_32X32X8F16_vgprcd_e64 $vgpr16_vgpr17, $vgpr16_vgpr17, $vgpr0_vgpr1_vgpr2_vgpr3_vgpr4_vgpr5_vgpr6_vgpr7_vgpr8_vgpr9_vgpr10_vgpr11_vgpr12_vgpr13_vgpr14_vgpr15, 0, 0, 0, implicit $mode, implicit $exec
|
|
; CHECK-NEXT: early-clobber renamable $vgpr0_vgpr1_vgpr2_vgpr3_vgpr4_vgpr5_vgpr6_vgpr7_vgpr8_vgpr9_vgpr10_vgpr11_vgpr12_vgpr13_vgpr14_vgpr15 = V_MFMA_F32_32X32X8F16_vgprcd_e64 $vgpr16_vgpr17, $vgpr16_vgpr17, $vgpr18_vgpr19_vgpr20_vgpr21_vgpr22_vgpr23_vgpr24_vgpr25_vgpr26_vgpr27_vgpr28_vgpr29_vgpr30_vgpr31_vgpr32_vgpr33, 0, 0, 0, implicit $mode, implicit $exec
|
|
; CHECK-NEXT: INLINEASM &"; use $0", sideeffect attdialect, reguse:VReg_512_Align2, killed renamable $vgpr0_vgpr1_vgpr2_vgpr3_vgpr4_vgpr5_vgpr6_vgpr7_vgpr8_vgpr9_vgpr10_vgpr11_vgpr12_vgpr13_vgpr14_vgpr15
|
|
; CHECK-NEXT: S_CBRANCH_VCCNZ %bb.1, implicit $vcc
|
|
; CHECK-NEXT: S_BRANCH %bb.2
|
|
; CHECK-NEXT: {{ $}}
|
|
; CHECK-NEXT: bb.2:
|
|
; CHECK-NEXT: liveins: $vgpr18_vgpr19_vgpr20_vgpr21_vgpr22_vgpr23_vgpr24_vgpr25_vgpr26_vgpr27_vgpr28_vgpr29_vgpr30_vgpr31_vgpr32_vgpr33:0x00000000FFFFFFFF
|
|
; CHECK-NEXT: {{ $}}
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr0_vgpr1_vgpr2_vgpr3_vgpr4_vgpr5_vgpr6_vgpr7
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr8_vgpr9_vgpr10_vgpr11_vgpr12_vgpr13_vgpr14_vgpr15
|
|
; CHECK-NEXT: renamable $agpr0_agpr1_agpr2_agpr3_agpr4_agpr5_agpr6_agpr7_agpr8_agpr9_agpr10_agpr11_agpr12_agpr13_agpr14_agpr15 = COPY killed renamable $vgpr18_vgpr19_vgpr20_vgpr21_vgpr22_vgpr23_vgpr24_vgpr25_vgpr26_vgpr27_vgpr28_vgpr29_vgpr30_vgpr31_vgpr32_vgpr33
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr16_vgpr17_vgpr18_vgpr19_vgpr20_vgpr21_vgpr22_vgpr23
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr24_vgpr25_vgpr26_vgpr27_vgpr28_vgpr29_vgpr30_vgpr31
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr32_vgpr33_vgpr34_vgpr35_vgpr36_vgpr37_vgpr38_vgpr39
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr40_vgpr41_vgpr42_vgpr43_vgpr44_vgpr45_vgpr46_vgpr47
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr48_vgpr49_vgpr50_vgpr51_vgpr52_vgpr53_vgpr54_vgpr55
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr56_vgpr57_vgpr58_vgpr59_vgpr60_vgpr61_vgpr62_vgpr63
|
|
; CHECK-NEXT: renamable $vgpr0 = V_MOV_B32_e32 0, implicit $exec
|
|
; CHECK-NEXT: GLOBAL_STORE_DWORDX4_SADDR renamable $vgpr0, renamable $agpr8_agpr9_agpr10_agpr11, undef $sgpr0_sgpr1, 32, 0, implicit $exec :: (store (s128), align 32, addrspace 1)
|
|
; CHECK-NEXT: GLOBAL_STORE_DWORDX4_SADDR renamable $vgpr0, renamable $agpr12_agpr13_agpr14_agpr15, undef $sgpr0_sgpr1, 48, 0, implicit $exec :: (store (s128), addrspace 1)
|
|
; CHECK-NEXT: GLOBAL_STORE_DWORDX4_SADDR renamable $vgpr0, renamable $agpr0_agpr1_agpr2_agpr3, undef $sgpr0_sgpr1, 0, 0, implicit $exec :: (store (s128), align 128, addrspace 1)
|
|
; CHECK-NEXT: GLOBAL_STORE_DWORDX4_SADDR killed renamable $vgpr0, killed renamable $agpr4_agpr5_agpr6_agpr7, killed undef $sgpr0_sgpr1, 16, 0, implicit $exec :: (store (s128), addrspace 1)
|
|
; CHECK-NEXT: S_ENDPGM 0
|
|
bb.0:
|
|
successors: %bb.1(0x80000000)
|
|
|
|
S_NOP 0, implicit-def $agpr0
|
|
renamable $sgpr0 = S_MOV_B32 0
|
|
undef %0.sub8:vreg_512_align2 = V_MOV_B32_e32 0, implicit $exec
|
|
renamable $sgpr1 = COPY renamable $sgpr0
|
|
%1:vreg_64_align2 = COPY killed renamable $sgpr0_sgpr1
|
|
renamable $vcc = S_AND_B64 $exec, -1, implicit-def dead $scc
|
|
%0.sub9:vreg_512_align2 = COPY %0.sub8
|
|
|
|
bb.1:
|
|
successors: %bb.1(0x40000000), %bb.2(0x40000000)
|
|
liveins: $vcc
|
|
|
|
undef %2.sub0_sub1:vreg_512_align2 = GLOBAL_LOAD_DWORDX2 undef %3:vreg_64_align2, 0, 0, implicit $exec :: (load (s64), addrspace 1)
|
|
early-clobber %0:vreg_512_align2 = V_MFMA_F32_32X32X8F16_vgprcd_e64 %1, %1, %2, 0, 0, 0, implicit $mode, implicit $exec
|
|
early-clobber %4:vreg_512_align2 = V_MFMA_F32_32X32X8F16_vgprcd_e64 %1, %1, %0, 0, 0, 0, implicit $mode, implicit $exec
|
|
INLINEASM &"; use $0", sideeffect attdialect, reguse:VReg_512_Align2, %4
|
|
S_CBRANCH_VCCNZ %bb.1, implicit $vcc
|
|
S_BRANCH %bb.2
|
|
|
|
bb.2:
|
|
S_NOP 0, implicit-def $vgpr0_vgpr1_vgpr2_vgpr3_vgpr4_vgpr5_vgpr6_vgpr7
|
|
S_NOP 0, implicit-def $vgpr8_vgpr9_vgpr10_vgpr11_vgpr12_vgpr13_vgpr14_vgpr15
|
|
S_NOP 0, implicit-def $vgpr16_vgpr17_vgpr18_vgpr19_vgpr20_vgpr21_vgpr22_vgpr23
|
|
S_NOP 0, implicit-def $vgpr24_vgpr25_vgpr26_vgpr27_vgpr28_vgpr29_vgpr30_vgpr31
|
|
S_NOP 0, implicit-def $vgpr32_vgpr33_vgpr34_vgpr35_vgpr36_vgpr37_vgpr38_vgpr39
|
|
S_NOP 0, implicit-def $vgpr40_vgpr41_vgpr42_vgpr43_vgpr44_vgpr45_vgpr46_vgpr47
|
|
S_NOP 0, implicit-def $vgpr48_vgpr49_vgpr50_vgpr51_vgpr52_vgpr53_vgpr54_vgpr55
|
|
S_NOP 0, implicit-def $vgpr56_vgpr57_vgpr58_vgpr59_vgpr60_vgpr61_vgpr62_vgpr63
|
|
%6:vgpr_32 = V_MOV_B32_e32 0, implicit $exec
|
|
GLOBAL_STORE_DWORDX4_SADDR %6, %0.sub8_sub9_sub10_sub11, undef $sgpr0_sgpr1, 32, 0, implicit $exec :: (store (s128), align 32, addrspace 1)
|
|
GLOBAL_STORE_DWORDX4_SADDR %6, %0.sub12_sub13_sub14_sub15, undef $sgpr0_sgpr1, 48, 0, implicit $exec :: (store (s128), addrspace 1)
|
|
GLOBAL_STORE_DWORDX4_SADDR %6, %0.sub0_sub1_sub2_sub3, undef $sgpr0_sgpr1, 0, 0, implicit $exec :: (store (s128), align 128, addrspace 1)
|
|
GLOBAL_STORE_DWORDX4_SADDR %6, %0.sub4_sub5_sub6_sub7, killed undef $sgpr0_sgpr1, 16, 0, implicit $exec :: (store (s128), addrspace 1)
|
|
S_ENDPGM 0
|
|
|
|
...
|
|
|
|
# TODO: In this trivial case, the single copy required is cheaper than
|
|
# the tuple copy.
|
|
---
|
|
name: chained_mfma_dst_user_is_vgpr_small_subreg
|
|
tracksRegLiveness: true
|
|
machineFunctionInfo:
|
|
isEntryFunction: true
|
|
stackPtrOffsetReg: '$sgpr32'
|
|
occupancy: 10
|
|
sgprForEXECCopy: '$sgpr100_sgpr101'
|
|
body: |
|
|
; CHECK-LABEL: name: chained_mfma_dst_user_is_vgpr_small_subreg
|
|
; CHECK: bb.0:
|
|
; CHECK-NEXT: successors: %bb.1(0x80000000)
|
|
; CHECK-NEXT: {{ $}}
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $agpr0
|
|
; CHECK-NEXT: renamable $sgpr0 = S_MOV_B32 0
|
|
; CHECK-NEXT: renamable $vgpr8 = V_MOV_B32_e32 0, implicit $exec
|
|
; CHECK-NEXT: renamable $sgpr1 = COPY renamable $sgpr0
|
|
; CHECK-NEXT: renamable $vgpr16_vgpr17 = COPY killed renamable $sgpr0_sgpr1
|
|
; CHECK-NEXT: renamable $vcc = S_AND_B64 $exec, -1, implicit-def dead $scc
|
|
; CHECK-NEXT: dead renamable $vgpr9 = COPY renamable $vgpr8
|
|
; CHECK-NEXT: {{ $}}
|
|
; CHECK-NEXT: bb.1:
|
|
; CHECK-NEXT: successors: %bb.1(0x40000000), %bb.2(0x40000000)
|
|
; CHECK-NEXT: liveins: $vcc, $vgpr16_vgpr17
|
|
; CHECK-NEXT: {{ $}}
|
|
; CHECK-NEXT: renamable $vgpr0_vgpr1 = GLOBAL_LOAD_DWORDX2 undef renamable $vgpr0_vgpr1, 0, 0, implicit $exec :: (load (s64), addrspace 1)
|
|
; CHECK-NEXT: early-clobber renamable $vgpr18_vgpr19_vgpr20_vgpr21_vgpr22_vgpr23_vgpr24_vgpr25_vgpr26_vgpr27_vgpr28_vgpr29_vgpr30_vgpr31_vgpr32_vgpr33 = V_MFMA_F32_32X32X8F16_vgprcd_e64 $vgpr16_vgpr17, $vgpr16_vgpr17, $vgpr0_vgpr1_vgpr2_vgpr3_vgpr4_vgpr5_vgpr6_vgpr7_vgpr8_vgpr9_vgpr10_vgpr11_vgpr12_vgpr13_vgpr14_vgpr15, 0, 0, 0, implicit $mode, implicit $exec
|
|
; CHECK-NEXT: early-clobber renamable $vgpr0_vgpr1_vgpr2_vgpr3_vgpr4_vgpr5_vgpr6_vgpr7_vgpr8_vgpr9_vgpr10_vgpr11_vgpr12_vgpr13_vgpr14_vgpr15 = V_MFMA_F32_32X32X8F16_vgprcd_e64 $vgpr16_vgpr17, $vgpr16_vgpr17, $vgpr18_vgpr19_vgpr20_vgpr21_vgpr22_vgpr23_vgpr24_vgpr25_vgpr26_vgpr27_vgpr28_vgpr29_vgpr30_vgpr31_vgpr32_vgpr33, 0, 0, 0, implicit $mode, implicit $exec
|
|
; CHECK-NEXT: dead renamable $vgpr0 = nofpexcept V_CVT_F16_F32_e32 killed $vgpr0, implicit $mode, implicit $exec
|
|
; CHECK-NEXT: S_CBRANCH_VCCNZ %bb.1, implicit $vcc
|
|
; CHECK-NEXT: S_BRANCH %bb.2
|
|
; CHECK-NEXT: {{ $}}
|
|
; CHECK-NEXT: bb.2:
|
|
; CHECK-NEXT: liveins: $vgpr18_vgpr19_vgpr20_vgpr21_vgpr22_vgpr23_vgpr24_vgpr25_vgpr26_vgpr27_vgpr28_vgpr29_vgpr30_vgpr31_vgpr32_vgpr33:0x00000000FFFFFFFF
|
|
; CHECK-NEXT: {{ $}}
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr0_vgpr1_vgpr2_vgpr3_vgpr4_vgpr5_vgpr6_vgpr7
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr8_vgpr9_vgpr10_vgpr11_vgpr12_vgpr13_vgpr14_vgpr15
|
|
; CHECK-NEXT: renamable $agpr0_agpr1_agpr2_agpr3_agpr4_agpr5_agpr6_agpr7_agpr8_agpr9_agpr10_agpr11_agpr12_agpr13_agpr14_agpr15 = COPY killed renamable $vgpr18_vgpr19_vgpr20_vgpr21_vgpr22_vgpr23_vgpr24_vgpr25_vgpr26_vgpr27_vgpr28_vgpr29_vgpr30_vgpr31_vgpr32_vgpr33
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr16_vgpr17_vgpr18_vgpr19_vgpr20_vgpr21_vgpr22_vgpr23
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr24_vgpr25_vgpr26_vgpr27_vgpr28_vgpr29_vgpr30_vgpr31
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr32_vgpr33_vgpr34_vgpr35_vgpr36_vgpr37_vgpr38_vgpr39
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr40_vgpr41_vgpr42_vgpr43_vgpr44_vgpr45_vgpr46_vgpr47
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr48_vgpr49_vgpr50_vgpr51_vgpr52_vgpr53_vgpr54_vgpr55
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr56_vgpr57_vgpr58_vgpr59_vgpr60_vgpr61_vgpr62_vgpr63
|
|
; CHECK-NEXT: renamable $vgpr0 = V_MOV_B32_e32 0, implicit $exec
|
|
; CHECK-NEXT: GLOBAL_STORE_DWORDX4_SADDR renamable $vgpr0, renamable $agpr8_agpr9_agpr10_agpr11, undef $sgpr0_sgpr1, 32, 0, implicit $exec :: (store (s128), align 32, addrspace 1)
|
|
; CHECK-NEXT: GLOBAL_STORE_DWORDX4_SADDR renamable $vgpr0, renamable $agpr12_agpr13_agpr14_agpr15, undef $sgpr0_sgpr1, 48, 0, implicit $exec :: (store (s128), addrspace 1)
|
|
; CHECK-NEXT: GLOBAL_STORE_DWORDX4_SADDR renamable $vgpr0, renamable $agpr0_agpr1_agpr2_agpr3, undef $sgpr0_sgpr1, 0, 0, implicit $exec :: (store (s128), align 128, addrspace 1)
|
|
; CHECK-NEXT: GLOBAL_STORE_DWORDX4_SADDR killed renamable $vgpr0, killed renamable $agpr4_agpr5_agpr6_agpr7, killed undef $sgpr0_sgpr1, 16, 0, implicit $exec :: (store (s128), addrspace 1)
|
|
; CHECK-NEXT: S_ENDPGM 0
|
|
bb.0:
|
|
successors: %bb.1(0x80000000)
|
|
|
|
S_NOP 0, implicit-def $agpr0
|
|
renamable $sgpr0 = S_MOV_B32 0
|
|
undef %0.sub8:vreg_512_align2 = V_MOV_B32_e32 0, implicit $exec
|
|
renamable $sgpr1 = COPY renamable $sgpr0
|
|
%1:vreg_64_align2 = COPY killed renamable $sgpr0_sgpr1
|
|
renamable $vcc = S_AND_B64 $exec, -1, implicit-def dead $scc
|
|
%0.sub9:vreg_512_align2 = COPY %0.sub8
|
|
|
|
bb.1:
|
|
successors: %bb.1(0x40000000), %bb.2(0x40000000)
|
|
liveins: $vcc
|
|
|
|
undef %2.sub0_sub1:vreg_512_align2 = GLOBAL_LOAD_DWORDX2 undef %3:vreg_64_align2, 0, 0, implicit $exec :: (load (s64), addrspace 1)
|
|
early-clobber %0:vreg_512_align2 = V_MFMA_F32_32X32X8F16_vgprcd_e64 %1, %1, %2, 0, 0, 0, implicit $mode, implicit $exec
|
|
early-clobber %4:vreg_512_align2 = V_MFMA_F32_32X32X8F16_vgprcd_e64 %1, %1, %0, 0, 0, 0, implicit $mode, implicit $exec
|
|
%5:vgpr_32 = nofpexcept V_CVT_F16_F32_e32 %4.sub0, implicit $mode, implicit $exec
|
|
S_CBRANCH_VCCNZ %bb.1, implicit $vcc
|
|
S_BRANCH %bb.2
|
|
|
|
bb.2:
|
|
S_NOP 0, implicit-def $vgpr0_vgpr1_vgpr2_vgpr3_vgpr4_vgpr5_vgpr6_vgpr7
|
|
S_NOP 0, implicit-def $vgpr8_vgpr9_vgpr10_vgpr11_vgpr12_vgpr13_vgpr14_vgpr15
|
|
S_NOP 0, implicit-def $vgpr16_vgpr17_vgpr18_vgpr19_vgpr20_vgpr21_vgpr22_vgpr23
|
|
S_NOP 0, implicit-def $vgpr24_vgpr25_vgpr26_vgpr27_vgpr28_vgpr29_vgpr30_vgpr31
|
|
S_NOP 0, implicit-def $vgpr32_vgpr33_vgpr34_vgpr35_vgpr36_vgpr37_vgpr38_vgpr39
|
|
S_NOP 0, implicit-def $vgpr40_vgpr41_vgpr42_vgpr43_vgpr44_vgpr45_vgpr46_vgpr47
|
|
S_NOP 0, implicit-def $vgpr48_vgpr49_vgpr50_vgpr51_vgpr52_vgpr53_vgpr54_vgpr55
|
|
S_NOP 0, implicit-def $vgpr56_vgpr57_vgpr58_vgpr59_vgpr60_vgpr61_vgpr62_vgpr63
|
|
%6:vgpr_32 = V_MOV_B32_e32 0, implicit $exec
|
|
GLOBAL_STORE_DWORDX4_SADDR %6, %0.sub8_sub9_sub10_sub11, undef $sgpr0_sgpr1, 32, 0, implicit $exec :: (store (s128), align 32, addrspace 1)
|
|
GLOBAL_STORE_DWORDX4_SADDR %6, %0.sub12_sub13_sub14_sub15, undef $sgpr0_sgpr1, 48, 0, implicit $exec :: (store (s128), addrspace 1)
|
|
GLOBAL_STORE_DWORDX4_SADDR %6, %0.sub0_sub1_sub2_sub3, undef $sgpr0_sgpr1, 0, 0, implicit $exec :: (store (s128), align 128, addrspace 1)
|
|
GLOBAL_STORE_DWORDX4_SADDR %6, %0.sub4_sub5_sub6_sub7, killed undef $sgpr0_sgpr1, 16, 0, implicit $exec :: (store (s128), addrspace 1)
|
|
S_ENDPGM 0
|
|
|
|
...
|
|
|
|
# Transitive user of the register is an MFMA with non-register src2
|
|
---
|
|
name: chained_mfma_dst_user_has_imm_src2
|
|
tracksRegLiveness: true
|
|
machineFunctionInfo:
|
|
isEntryFunction: true
|
|
stackPtrOffsetReg: '$sgpr32'
|
|
occupancy: 10
|
|
sgprForEXECCopy: '$sgpr100_sgpr101'
|
|
body: |
|
|
; CHECK-LABEL: name: chained_mfma_dst_user_has_imm_src2
|
|
; CHECK: bb.0:
|
|
; CHECK-NEXT: successors: %bb.1(0x80000000)
|
|
; CHECK-NEXT: {{ $}}
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $agpr0
|
|
; CHECK-NEXT: renamable $sgpr0 = S_MOV_B32 0
|
|
; CHECK-NEXT: renamable $vgpr8 = V_MOV_B32_e32 0, implicit $exec
|
|
; CHECK-NEXT: renamable $sgpr1 = COPY renamable $sgpr0
|
|
; CHECK-NEXT: renamable $vgpr18_vgpr19 = COPY killed renamable $sgpr0_sgpr1
|
|
; CHECK-NEXT: renamable $vcc = S_AND_B64 $exec, -1, implicit-def dead $scc
|
|
; CHECK-NEXT: dead renamable $vgpr9 = COPY renamable $vgpr8
|
|
; CHECK-NEXT: {{ $}}
|
|
; CHECK-NEXT: bb.1:
|
|
; CHECK-NEXT: successors: %bb.1(0x40000000), %bb.2(0x40000000)
|
|
; CHECK-NEXT: liveins: $vcc, $vgpr18_vgpr19
|
|
; CHECK-NEXT: {{ $}}
|
|
; CHECK-NEXT: renamable $agpr0_agpr1 = GLOBAL_LOAD_DWORDX2 undef renamable $vgpr0_vgpr1, 0, 0, implicit $exec :: (load (s64), addrspace 1)
|
|
; CHECK-NEXT: early-clobber renamable $agpr16_agpr17_agpr18_agpr19_agpr20_agpr21_agpr22_agpr23_agpr24_agpr25_agpr26_agpr27_agpr28_agpr29_agpr30_agpr31 = V_MFMA_F32_32X32X8F16_e64 $vgpr18_vgpr19, $vgpr18_vgpr19, $agpr0_agpr1_agpr2_agpr3_agpr4_agpr5_agpr6_agpr7_agpr8_agpr9_agpr10_agpr11_agpr12_agpr13_agpr14_agpr15, 0, 0, 0, implicit $mode, implicit $exec
|
|
; CHECK-NEXT: early-clobber renamable $agpr0_agpr1_agpr2_agpr3_agpr4_agpr5_agpr6_agpr7_agpr8_agpr9_agpr10_agpr11_agpr12_agpr13_agpr14_agpr15 = V_MFMA_F32_32X32X8F16_e64 $vgpr18_vgpr19, $vgpr18_vgpr19, killed $agpr16_agpr17_agpr18_agpr19_agpr20_agpr21_agpr22_agpr23_agpr24_agpr25_agpr26_agpr27_agpr28_agpr29_agpr30_agpr31, 0, 0, 0, implicit $mode, implicit $exec
|
|
; CHECK-NEXT: renamable $agpr0_agpr1_agpr2_agpr3 = V_MFMA_F32_4X4X4F16_e64 $agpr0_agpr1, $vgpr18_vgpr19, 0, 0, 0, 0, implicit $mode, implicit $exec
|
|
; CHECK-NEXT: S_CBRANCH_VCCNZ %bb.1, implicit $vcc
|
|
; CHECK-NEXT: S_BRANCH %bb.2
|
|
; CHECK-NEXT: {{ $}}
|
|
; CHECK-NEXT: bb.2:
|
|
; CHECK-NEXT: liveins: $agpr0_agpr1_agpr2_agpr3_agpr4_agpr5_agpr6_agpr7_agpr8_agpr9_agpr10_agpr11_agpr12_agpr13_agpr14_agpr15:0x00000000FFFFFFFF
|
|
; CHECK-NEXT: {{ $}}
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr0_vgpr1_vgpr2_vgpr3_vgpr4_vgpr5_vgpr6_vgpr7
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr8_vgpr9_vgpr10_vgpr11_vgpr12_vgpr13_vgpr14_vgpr15
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr16_vgpr17_vgpr18_vgpr19_vgpr20_vgpr21_vgpr22_vgpr23
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr24_vgpr25_vgpr26_vgpr27_vgpr28_vgpr29_vgpr30_vgpr31
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr32_vgpr33_vgpr34_vgpr35_vgpr36_vgpr37_vgpr38_vgpr39
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr40_vgpr41_vgpr42_vgpr43_vgpr44_vgpr45_vgpr46_vgpr47
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr48_vgpr49_vgpr50_vgpr51_vgpr52_vgpr53_vgpr54_vgpr55
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr56_vgpr57_vgpr58_vgpr59_vgpr60_vgpr61_vgpr62_vgpr63
|
|
; CHECK-NEXT: renamable $vgpr0 = V_MOV_B32_e32 0, implicit $exec
|
|
; CHECK-NEXT: GLOBAL_STORE_DWORDX4_SADDR renamable $vgpr0, renamable $agpr8_agpr9_agpr10_agpr11, undef $sgpr0_sgpr1, 32, 0, implicit $exec :: (store (s128), align 32, addrspace 1)
|
|
; CHECK-NEXT: GLOBAL_STORE_DWORDX4_SADDR renamable $vgpr0, renamable $agpr12_agpr13_agpr14_agpr15, undef $sgpr0_sgpr1, 48, 0, implicit $exec :: (store (s128), addrspace 1)
|
|
; CHECK-NEXT: GLOBAL_STORE_DWORDX4_SADDR renamable $vgpr0, renamable $agpr0_agpr1_agpr2_agpr3, undef $sgpr0_sgpr1, 0, 0, implicit $exec :: (store (s128), align 128, addrspace 1)
|
|
; CHECK-NEXT: GLOBAL_STORE_DWORDX4_SADDR killed renamable $vgpr0, killed renamable $agpr4_agpr5_agpr6_agpr7, killed undef $sgpr0_sgpr1, 16, 0, implicit $exec :: (store (s128), addrspace 1)
|
|
; CHECK-NEXT: S_ENDPGM 0
|
|
bb.0:
|
|
successors: %bb.1(0x80000000)
|
|
|
|
S_NOP 0, implicit-def $agpr0
|
|
renamable $sgpr0 = S_MOV_B32 0
|
|
undef %0.sub8:vreg_512_align2 = V_MOV_B32_e32 0, implicit $exec
|
|
renamable $sgpr1 = COPY renamable $sgpr0
|
|
%1:vreg_64_align2 = COPY killed renamable $sgpr0_sgpr1
|
|
renamable $vcc = S_AND_B64 $exec, -1, implicit-def dead $scc
|
|
%0.sub9:vreg_512_align2 = COPY %0.sub8
|
|
|
|
bb.1:
|
|
successors: %bb.1(0x40000000), %bb.2(0x40000000)
|
|
liveins: $vcc
|
|
|
|
undef %2.sub0_sub1:vreg_512_align2 = GLOBAL_LOAD_DWORDX2 undef %3:vreg_64_align2, 0, 0, implicit $exec :: (load (s64), addrspace 1)
|
|
early-clobber %0:vreg_512_align2 = V_MFMA_F32_32X32X8F16_vgprcd_e64 %1, %1, %2, 0, 0, 0, implicit $mode, implicit $exec
|
|
early-clobber %4:vreg_512_align2 = V_MFMA_F32_32X32X8F16_vgprcd_e64 %1, %1, %0, 0, 0, 0, implicit $mode, implicit $exec
|
|
%4.sub0_sub1_sub2_sub3:vreg_512_align2 = V_MFMA_F32_4X4X4F16_vgprcd_e64 %4.sub0_sub1, %1, 0, 0, 0, 0, implicit $mode, implicit $exec
|
|
|
|
S_CBRANCH_VCCNZ %bb.1, implicit $vcc
|
|
S_BRANCH %bb.2
|
|
|
|
bb.2:
|
|
S_NOP 0, implicit-def $vgpr0_vgpr1_vgpr2_vgpr3_vgpr4_vgpr5_vgpr6_vgpr7
|
|
S_NOP 0, implicit-def $vgpr8_vgpr9_vgpr10_vgpr11_vgpr12_vgpr13_vgpr14_vgpr15
|
|
S_NOP 0, implicit-def $vgpr16_vgpr17_vgpr18_vgpr19_vgpr20_vgpr21_vgpr22_vgpr23
|
|
S_NOP 0, implicit-def $vgpr24_vgpr25_vgpr26_vgpr27_vgpr28_vgpr29_vgpr30_vgpr31
|
|
S_NOP 0, implicit-def $vgpr32_vgpr33_vgpr34_vgpr35_vgpr36_vgpr37_vgpr38_vgpr39
|
|
S_NOP 0, implicit-def $vgpr40_vgpr41_vgpr42_vgpr43_vgpr44_vgpr45_vgpr46_vgpr47
|
|
S_NOP 0, implicit-def $vgpr48_vgpr49_vgpr50_vgpr51_vgpr52_vgpr53_vgpr54_vgpr55
|
|
S_NOP 0, implicit-def $vgpr56_vgpr57_vgpr58_vgpr59_vgpr60_vgpr61_vgpr62_vgpr63
|
|
%6:vgpr_32 = V_MOV_B32_e32 0, implicit $exec
|
|
GLOBAL_STORE_DWORDX4_SADDR %6, %4.sub8_sub9_sub10_sub11, undef $sgpr0_sgpr1, 32, 0, implicit $exec :: (store (s128), align 32, addrspace 1)
|
|
GLOBAL_STORE_DWORDX4_SADDR %6, %4.sub12_sub13_sub14_sub15, undef $sgpr0_sgpr1, 48, 0, implicit $exec :: (store (s128), addrspace 1)
|
|
GLOBAL_STORE_DWORDX4_SADDR %6, %4.sub0_sub1_sub2_sub3, undef $sgpr0_sgpr1, 0, 0, implicit $exec :: (store (s128), align 128, addrspace 1)
|
|
GLOBAL_STORE_DWORDX4_SADDR %6, %4.sub4_sub5_sub6_sub7, killed undef $sgpr0_sgpr1, 16, 0, implicit $exec :: (store (s128), addrspace 1)
|
|
S_ENDPGM 0
|
|
|
|
...
|
|
|
|
---
|
|
name: inflate_result_to_agpr__V_MFMA_F32_32X32X8F16_mac_vgprcd_e64_has_untied_user
|
|
tracksRegLiveness: true
|
|
machineFunctionInfo:
|
|
isEntryFunction: true
|
|
stackPtrOffsetReg: '$sgpr32'
|
|
occupancy: 10
|
|
sgprForEXECCopy: '$sgpr100_sgpr101'
|
|
body: |
|
|
; CHECK-LABEL: name: inflate_result_to_agpr__V_MFMA_F32_32X32X8F16_mac_vgprcd_e64_has_untied_user
|
|
; CHECK: bb.0:
|
|
; CHECK-NEXT: successors: %bb.1(0x80000000)
|
|
; CHECK-NEXT: {{ $}}
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $agpr0
|
|
; CHECK-NEXT: renamable $sgpr0 = S_MOV_B32 0
|
|
; CHECK-NEXT: renamable $vgpr8 = V_MOV_B32_e32 0, implicit $exec
|
|
; CHECK-NEXT: renamable $sgpr1 = COPY renamable $sgpr0
|
|
; CHECK-NEXT: renamable $vgpr16_vgpr17 = COPY killed renamable $sgpr0_sgpr1
|
|
; CHECK-NEXT: renamable $vcc = S_AND_B64 $exec, -1, implicit-def dead $scc
|
|
; CHECK-NEXT: dead renamable $vgpr9 = COPY renamable $vgpr8
|
|
; CHECK-NEXT: {{ $}}
|
|
; CHECK-NEXT: bb.1:
|
|
; CHECK-NEXT: successors: %bb.1(0x40000000), %bb.2(0x40000000)
|
|
; CHECK-NEXT: liveins: $vcc, $vgpr16_vgpr17
|
|
; CHECK-NEXT: {{ $}}
|
|
; CHECK-NEXT: renamable $agpr16_agpr17 = GLOBAL_LOAD_DWORDX2 undef renamable $vgpr0_vgpr1, 0, 0, implicit $exec :: (load (s64), addrspace 1)
|
|
; CHECK-NEXT: renamable $agpr16_agpr17_agpr18_agpr19_agpr20_agpr21_agpr22_agpr23_agpr24_agpr25_agpr26_agpr27_agpr28_agpr29_agpr30_agpr31 = V_MFMA_F32_32X32X8F16_mac_e64 $vgpr16_vgpr17, $vgpr16_vgpr17, $agpr16_agpr17_agpr18_agpr19_agpr20_agpr21_agpr22_agpr23_agpr24_agpr25_agpr26_agpr27_agpr28_agpr29_agpr30_agpr31, 0, 0, 0, implicit $mode, implicit $exec
|
|
; CHECK-NEXT: early-clobber renamable $agpr0_agpr1_agpr2_agpr3_agpr4_agpr5_agpr6_agpr7_agpr8_agpr9_agpr10_agpr11_agpr12_agpr13_agpr14_agpr15 = V_MFMA_F32_32X32X8F16_e64 $vgpr16_vgpr17, $vgpr16_vgpr17, killed $agpr16_agpr17_agpr18_agpr19_agpr20_agpr21_agpr22_agpr23_agpr24_agpr25_agpr26_agpr27_agpr28_agpr29_agpr30_agpr31, 0, 0, 0, implicit $mode, implicit $exec
|
|
; CHECK-NEXT: S_CBRANCH_VCCNZ %bb.1, implicit $vcc
|
|
; CHECK-NEXT: S_BRANCH %bb.2
|
|
; CHECK-NEXT: {{ $}}
|
|
; CHECK-NEXT: bb.2:
|
|
; CHECK-NEXT: liveins: $agpr0_agpr1_agpr2_agpr3_agpr4_agpr5_agpr6_agpr7_agpr8_agpr9_agpr10_agpr11_agpr12_agpr13_agpr14_agpr15:0x00000000FFFFFFFF
|
|
; CHECK-NEXT: {{ $}}
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr0_vgpr1_vgpr2_vgpr3_vgpr4_vgpr5_vgpr6_vgpr7
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr8_vgpr9_vgpr10_vgpr11_vgpr12_vgpr13_vgpr14_vgpr15
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr16_vgpr17_vgpr18_vgpr19_vgpr20_vgpr21_vgpr22_vgpr23
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr24_vgpr25_vgpr26_vgpr27_vgpr28_vgpr29_vgpr30_vgpr31
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr32_vgpr33_vgpr34_vgpr35_vgpr36_vgpr37_vgpr38_vgpr39
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr40_vgpr41_vgpr42_vgpr43_vgpr44_vgpr45_vgpr46_vgpr47
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr48_vgpr49_vgpr50_vgpr51_vgpr52_vgpr53_vgpr54_vgpr55
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr56_vgpr57_vgpr58_vgpr59_vgpr60_vgpr61_vgpr62_vgpr63
|
|
; CHECK-NEXT: renamable $vgpr0 = V_MOV_B32_e32 0, implicit $exec
|
|
; CHECK-NEXT: GLOBAL_STORE_DWORDX4_SADDR renamable $vgpr0, renamable $agpr8_agpr9_agpr10_agpr11, undef $sgpr0_sgpr1, 32, 0, implicit $exec :: (store (s128), align 32, addrspace 1)
|
|
; CHECK-NEXT: GLOBAL_STORE_DWORDX4_SADDR renamable $vgpr0, renamable $agpr12_agpr13_agpr14_agpr15, undef $sgpr0_sgpr1, 48, 0, implicit $exec :: (store (s128), addrspace 1)
|
|
; CHECK-NEXT: GLOBAL_STORE_DWORDX4_SADDR renamable $vgpr0, renamable $agpr0_agpr1_agpr2_agpr3, undef $sgpr0_sgpr1, 0, 0, implicit $exec :: (store (s128), align 128, addrspace 1)
|
|
; CHECK-NEXT: GLOBAL_STORE_DWORDX4_SADDR killed renamable $vgpr0, killed renamable $agpr4_agpr5_agpr6_agpr7, killed undef $sgpr0_sgpr1, 16, 0, implicit $exec :: (store (s128), addrspace 1)
|
|
; CHECK-NEXT: S_ENDPGM 0
|
|
bb.0:
|
|
S_NOP 0, implicit-def $agpr0
|
|
renamable $sgpr0 = S_MOV_B32 0
|
|
undef %0.sub8:vreg_512_align2 = V_MOV_B32_e32 0, implicit $exec
|
|
renamable $sgpr1 = COPY renamable $sgpr0
|
|
%1:vreg_64_align2 = COPY killed renamable $sgpr0_sgpr1
|
|
renamable $vcc = S_AND_B64 $exec, -1, implicit-def dead $scc
|
|
%0.sub9:vreg_512_align2 = COPY %0.sub8
|
|
|
|
bb.1:
|
|
liveins: $vcc
|
|
|
|
undef %0.sub0_sub1:vreg_512_align2 = GLOBAL_LOAD_DWORDX2 undef %3:vreg_64_align2, 0, 0, implicit $exec :: (load (s64), addrspace 1)
|
|
%0:vreg_512_align2 = V_MFMA_F32_32X32X8F16_mac_vgprcd_e64 %1, %1, %0, 0, 0, 0, implicit $mode, implicit $exec
|
|
%4:vreg_512_align2 = V_MFMA_F32_32X32X8F16_vgprcd_e64 %1, %1, %0, 0, 0, 0, implicit $mode, implicit $exec
|
|
S_CBRANCH_VCCNZ %bb.1, implicit $vcc
|
|
S_BRANCH %bb.2
|
|
|
|
bb.2:
|
|
; No VGPRs available for %0
|
|
S_NOP 0, implicit-def $vgpr0_vgpr1_vgpr2_vgpr3_vgpr4_vgpr5_vgpr6_vgpr7
|
|
S_NOP 0, implicit-def $vgpr8_vgpr9_vgpr10_vgpr11_vgpr12_vgpr13_vgpr14_vgpr15
|
|
S_NOP 0, implicit-def $vgpr16_vgpr17_vgpr18_vgpr19_vgpr20_vgpr21_vgpr22_vgpr23
|
|
S_NOP 0, implicit-def $vgpr24_vgpr25_vgpr26_vgpr27_vgpr28_vgpr29_vgpr30_vgpr31
|
|
S_NOP 0, implicit-def $vgpr32_vgpr33_vgpr34_vgpr35_vgpr36_vgpr37_vgpr38_vgpr39
|
|
S_NOP 0, implicit-def $vgpr40_vgpr41_vgpr42_vgpr43_vgpr44_vgpr45_vgpr46_vgpr47
|
|
S_NOP 0, implicit-def $vgpr48_vgpr49_vgpr50_vgpr51_vgpr52_vgpr53_vgpr54_vgpr55
|
|
S_NOP 0, implicit-def $vgpr56_vgpr57_vgpr58_vgpr59_vgpr60_vgpr61_vgpr62_vgpr63
|
|
%5:vgpr_32 = V_MOV_B32_e32 0, implicit $exec
|
|
GLOBAL_STORE_DWORDX4_SADDR %5, %4.sub8_sub9_sub10_sub11, undef $sgpr0_sgpr1, 32, 0, implicit $exec :: (store (s128), align 32, addrspace 1)
|
|
GLOBAL_STORE_DWORDX4_SADDR %5, %4.sub12_sub13_sub14_sub15, undef $sgpr0_sgpr1, 48, 0, implicit $exec :: (store (s128), addrspace 1)
|
|
GLOBAL_STORE_DWORDX4_SADDR %5, %4.sub0_sub1_sub2_sub3, undef $sgpr0_sgpr1, 0, 0, implicit $exec :: (store (s128), align 128, addrspace 1)
|
|
GLOBAL_STORE_DWORDX4_SADDR %5, %4.sub4_sub5_sub6_sub7, killed undef $sgpr0_sgpr1, 16, 0, implicit $exec :: (store (s128), addrspace 1)
|
|
S_ENDPGM 0
|
|
|
|
...
|
|
|
|
---
|
|
name: inflate_result_to_agpr__V_MFMA_F32_32X32X8F16_mac_vgprcd_e64_has_untied_user_with_vgpr_use
|
|
tracksRegLiveness: true
|
|
machineFunctionInfo:
|
|
isEntryFunction: true
|
|
stackPtrOffsetReg: '$sgpr32'
|
|
occupancy: 10
|
|
sgprForEXECCopy: '$sgpr100_sgpr101'
|
|
body: |
|
|
; CHECK-LABEL: name: inflate_result_to_agpr__V_MFMA_F32_32X32X8F16_mac_vgprcd_e64_has_untied_user_with_vgpr_use
|
|
; CHECK: bb.0:
|
|
; CHECK-NEXT: successors: %bb.1(0x80000000)
|
|
; CHECK-NEXT: {{ $}}
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $agpr0
|
|
; CHECK-NEXT: renamable $sgpr0 = S_MOV_B32 0
|
|
; CHECK-NEXT: renamable $vgpr8 = V_MOV_B32_e32 0, implicit $exec
|
|
; CHECK-NEXT: renamable $sgpr1 = COPY renamable $sgpr0
|
|
; CHECK-NEXT: renamable $vgpr16_vgpr17 = COPY killed renamable $sgpr0_sgpr1
|
|
; CHECK-NEXT: renamable $vcc = S_AND_B64 $exec, -1, implicit-def dead $scc
|
|
; CHECK-NEXT: dead renamable $vgpr9 = COPY renamable $vgpr8
|
|
; CHECK-NEXT: {{ $}}
|
|
; CHECK-NEXT: bb.1:
|
|
; CHECK-NEXT: successors: %bb.1(0x40000000), %bb.2(0x40000000)
|
|
; CHECK-NEXT: liveins: $vcc, $vgpr16_vgpr17
|
|
; CHECK-NEXT: {{ $}}
|
|
; CHECK-NEXT: renamable $vgpr0_vgpr1 = GLOBAL_LOAD_DWORDX2 undef renamable $vgpr0_vgpr1, 0, 0, implicit $exec :: (load (s64), addrspace 1)
|
|
; CHECK-NEXT: renamable $vgpr0_vgpr1_vgpr2_vgpr3_vgpr4_vgpr5_vgpr6_vgpr7_vgpr8_vgpr9_vgpr10_vgpr11_vgpr12_vgpr13_vgpr14_vgpr15 = V_MFMA_F32_32X32X8F16_mac_vgprcd_e64 $vgpr16_vgpr17, $vgpr16_vgpr17, $vgpr0_vgpr1_vgpr2_vgpr3_vgpr4_vgpr5_vgpr6_vgpr7_vgpr8_vgpr9_vgpr10_vgpr11_vgpr12_vgpr13_vgpr14_vgpr15, 0, 0, 0, implicit $mode, implicit $exec
|
|
; CHECK-NEXT: early-clobber renamable $vgpr18_vgpr19_vgpr20_vgpr21_vgpr22_vgpr23_vgpr24_vgpr25_vgpr26_vgpr27_vgpr28_vgpr29_vgpr30_vgpr31_vgpr32_vgpr33 = V_MFMA_F32_32X32X8F16_vgprcd_e64 $vgpr16_vgpr17, $vgpr16_vgpr17, killed $vgpr0_vgpr1_vgpr2_vgpr3_vgpr4_vgpr5_vgpr6_vgpr7_vgpr8_vgpr9_vgpr10_vgpr11_vgpr12_vgpr13_vgpr14_vgpr15, 0, 0, 0, implicit $mode, implicit $exec
|
|
; CHECK-NEXT: INLINEASM &"; use $0", sideeffect attdialect, reguse:VReg_512_Align2, renamable $vgpr18_vgpr19_vgpr20_vgpr21_vgpr22_vgpr23_vgpr24_vgpr25_vgpr26_vgpr27_vgpr28_vgpr29_vgpr30_vgpr31_vgpr32_vgpr33
|
|
; CHECK-NEXT: S_CBRANCH_VCCNZ %bb.1, implicit $vcc
|
|
; CHECK-NEXT: S_BRANCH %bb.2
|
|
; CHECK-NEXT: {{ $}}
|
|
; CHECK-NEXT: bb.2:
|
|
; CHECK-NEXT: liveins: $vgpr18_vgpr19_vgpr20_vgpr21_vgpr22_vgpr23_vgpr24_vgpr25_vgpr26_vgpr27_vgpr28_vgpr29_vgpr30_vgpr31_vgpr32_vgpr33:0x00000000FFFFFFFF
|
|
; CHECK-NEXT: {{ $}}
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr0_vgpr1_vgpr2_vgpr3_vgpr4_vgpr5_vgpr6_vgpr7
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr8_vgpr9_vgpr10_vgpr11_vgpr12_vgpr13_vgpr14_vgpr15
|
|
; CHECK-NEXT: renamable $agpr0_agpr1_agpr2_agpr3_agpr4_agpr5_agpr6_agpr7_agpr8_agpr9_agpr10_agpr11_agpr12_agpr13_agpr14_agpr15 = COPY killed renamable $vgpr18_vgpr19_vgpr20_vgpr21_vgpr22_vgpr23_vgpr24_vgpr25_vgpr26_vgpr27_vgpr28_vgpr29_vgpr30_vgpr31_vgpr32_vgpr33
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr16_vgpr17_vgpr18_vgpr19_vgpr20_vgpr21_vgpr22_vgpr23
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr24_vgpr25_vgpr26_vgpr27_vgpr28_vgpr29_vgpr30_vgpr31
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr32_vgpr33_vgpr34_vgpr35_vgpr36_vgpr37_vgpr38_vgpr39
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr40_vgpr41_vgpr42_vgpr43_vgpr44_vgpr45_vgpr46_vgpr47
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr48_vgpr49_vgpr50_vgpr51_vgpr52_vgpr53_vgpr54_vgpr55
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr56_vgpr57_vgpr58_vgpr59_vgpr60_vgpr61_vgpr62_vgpr63
|
|
; CHECK-NEXT: renamable $vgpr0 = V_MOV_B32_e32 0, implicit $exec
|
|
; CHECK-NEXT: GLOBAL_STORE_DWORDX4_SADDR renamable $vgpr0, renamable $agpr8_agpr9_agpr10_agpr11, undef $sgpr0_sgpr1, 32, 0, implicit $exec :: (store (s128), align 32, addrspace 1)
|
|
; CHECK-NEXT: GLOBAL_STORE_DWORDX4_SADDR renamable $vgpr0, renamable $agpr12_agpr13_agpr14_agpr15, undef $sgpr0_sgpr1, 48, 0, implicit $exec :: (store (s128), addrspace 1)
|
|
; CHECK-NEXT: GLOBAL_STORE_DWORDX4_SADDR renamable $vgpr0, renamable $agpr0_agpr1_agpr2_agpr3, undef $sgpr0_sgpr1, 0, 0, implicit $exec :: (store (s128), align 128, addrspace 1)
|
|
; CHECK-NEXT: GLOBAL_STORE_DWORDX4_SADDR killed renamable $vgpr0, killed renamable $agpr4_agpr5_agpr6_agpr7, killed undef $sgpr0_sgpr1, 16, 0, implicit $exec :: (store (s128), addrspace 1)
|
|
; CHECK-NEXT: S_ENDPGM 0
|
|
bb.0:
|
|
S_NOP 0, implicit-def $agpr0
|
|
renamable $sgpr0 = S_MOV_B32 0
|
|
undef %0.sub8:vreg_512_align2 = V_MOV_B32_e32 0, implicit $exec
|
|
renamable $sgpr1 = COPY renamable $sgpr0
|
|
%1:vreg_64_align2 = COPY killed renamable $sgpr0_sgpr1
|
|
renamable $vcc = S_AND_B64 $exec, -1, implicit-def dead $scc
|
|
%0.sub9:vreg_512_align2 = COPY %0.sub8
|
|
|
|
bb.1:
|
|
liveins: $vcc
|
|
|
|
undef %0.sub0_sub1:vreg_512_align2 = GLOBAL_LOAD_DWORDX2 undef %3:vreg_64_align2, 0, 0, implicit $exec :: (load (s64), addrspace 1)
|
|
%0:vreg_512_align2 = V_MFMA_F32_32X32X8F16_mac_vgprcd_e64 %1, %1, %0, 0, 0, 0, implicit $mode, implicit $exec
|
|
%4:vreg_512_align2 = V_MFMA_F32_32X32X8F16_vgprcd_e64 %1, %1, %0, 0, 0, 0, implicit $mode, implicit $exec
|
|
INLINEASM &"; use $0", sideeffect attdialect, reguse:VReg_512_Align2, %4
|
|
S_CBRANCH_VCCNZ %bb.1, implicit $vcc
|
|
S_BRANCH %bb.2
|
|
|
|
bb.2:
|
|
; No VGPRs available for %0
|
|
S_NOP 0, implicit-def $vgpr0_vgpr1_vgpr2_vgpr3_vgpr4_vgpr5_vgpr6_vgpr7
|
|
S_NOP 0, implicit-def $vgpr8_vgpr9_vgpr10_vgpr11_vgpr12_vgpr13_vgpr14_vgpr15
|
|
S_NOP 0, implicit-def $vgpr16_vgpr17_vgpr18_vgpr19_vgpr20_vgpr21_vgpr22_vgpr23
|
|
S_NOP 0, implicit-def $vgpr24_vgpr25_vgpr26_vgpr27_vgpr28_vgpr29_vgpr30_vgpr31
|
|
S_NOP 0, implicit-def $vgpr32_vgpr33_vgpr34_vgpr35_vgpr36_vgpr37_vgpr38_vgpr39
|
|
S_NOP 0, implicit-def $vgpr40_vgpr41_vgpr42_vgpr43_vgpr44_vgpr45_vgpr46_vgpr47
|
|
S_NOP 0, implicit-def $vgpr48_vgpr49_vgpr50_vgpr51_vgpr52_vgpr53_vgpr54_vgpr55
|
|
S_NOP 0, implicit-def $vgpr56_vgpr57_vgpr58_vgpr59_vgpr60_vgpr61_vgpr62_vgpr63
|
|
%5:vgpr_32 = V_MOV_B32_e32 0, implicit $exec
|
|
GLOBAL_STORE_DWORDX4_SADDR %5, %4.sub8_sub9_sub10_sub11, undef $sgpr0_sgpr1, 32, 0, implicit $exec :: (store (s128), align 32, addrspace 1)
|
|
GLOBAL_STORE_DWORDX4_SADDR %5, %4.sub12_sub13_sub14_sub15, undef $sgpr0_sgpr1, 48, 0, implicit $exec :: (store (s128), addrspace 1)
|
|
GLOBAL_STORE_DWORDX4_SADDR %5, %4.sub0_sub1_sub2_sub3, undef $sgpr0_sgpr1, 0, 0, implicit $exec :: (store (s128), align 128, addrspace 1)
|
|
GLOBAL_STORE_DWORDX4_SADDR %5, %4.sub4_sub5_sub6_sub7, killed undef $sgpr0_sgpr1, 16, 0, implicit $exec :: (store (s128), addrspace 1)
|
|
S_ENDPGM 0
|
|
|
|
...
|
|
|
|
---
|
|
name: inflate_result_to_agpr__V_MFMA_F32_32X32X8F16_vgprcd_e64_has_tied_user
|
|
tracksRegLiveness: true
|
|
machineFunctionInfo:
|
|
isEntryFunction: true
|
|
stackPtrOffsetReg: '$sgpr32'
|
|
occupancy: 10
|
|
sgprForEXECCopy: '$sgpr100_sgpr101'
|
|
body: |
|
|
; CHECK-LABEL: name: inflate_result_to_agpr__V_MFMA_F32_32X32X8F16_vgprcd_e64_has_tied_user
|
|
; CHECK: bb.0:
|
|
; CHECK-NEXT: successors: %bb.1(0x80000000)
|
|
; CHECK-NEXT: {{ $}}
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $agpr0
|
|
; CHECK-NEXT: renamable $sgpr0 = S_MOV_B32 0
|
|
; CHECK-NEXT: renamable $vgpr8 = V_MOV_B32_e32 0, implicit $exec
|
|
; CHECK-NEXT: renamable $sgpr1 = COPY renamable $sgpr0
|
|
; CHECK-NEXT: renamable $vgpr2_vgpr3 = COPY killed renamable $sgpr0_sgpr1
|
|
; CHECK-NEXT: renamable $vcc = S_AND_B64 $exec, -1, implicit-def dead $scc
|
|
; CHECK-NEXT: dead renamable $vgpr9 = COPY renamable $vgpr8
|
|
; CHECK-NEXT: {{ $}}
|
|
; CHECK-NEXT: bb.1:
|
|
; CHECK-NEXT: successors: %bb.1(0x40000000), %bb.2(0x40000000)
|
|
; CHECK-NEXT: liveins: $vcc, $vgpr2_vgpr3
|
|
; CHECK-NEXT: {{ $}}
|
|
; CHECK-NEXT: renamable $agpr16_agpr17 = GLOBAL_LOAD_DWORDX2 undef renamable $vgpr0_vgpr1, 0, 0, implicit $exec :: (load (s64), addrspace 1)
|
|
; CHECK-NEXT: early-clobber renamable $agpr0_agpr1_agpr2_agpr3_agpr4_agpr5_agpr6_agpr7_agpr8_agpr9_agpr10_agpr11_agpr12_agpr13_agpr14_agpr15 = V_MFMA_F32_32X32X8F16_e64 $vgpr2_vgpr3, $vgpr2_vgpr3, $agpr16_agpr17_agpr18_agpr19_agpr20_agpr21_agpr22_agpr23_agpr24_agpr25_agpr26_agpr27_agpr28_agpr29_agpr30_agpr31, 0, 0, 0, implicit $mode, implicit $exec
|
|
; CHECK-NEXT: renamable $agpr0_agpr1_agpr2_agpr3_agpr4_agpr5_agpr6_agpr7_agpr8_agpr9_agpr10_agpr11_agpr12_agpr13_agpr14_agpr15 = V_MFMA_F32_32X32X8F16_mac_e64 $vgpr2_vgpr3, $vgpr2_vgpr3, killed $agpr0_agpr1_agpr2_agpr3_agpr4_agpr5_agpr6_agpr7_agpr8_agpr9_agpr10_agpr11_agpr12_agpr13_agpr14_agpr15, 0, 0, 0, implicit $mode, implicit $exec
|
|
; CHECK-NEXT: S_CBRANCH_VCCNZ %bb.1, implicit $vcc
|
|
; CHECK-NEXT: S_BRANCH %bb.2
|
|
; CHECK-NEXT: {{ $}}
|
|
; CHECK-NEXT: bb.2:
|
|
; CHECK-NEXT: liveins: $agpr0_agpr1_agpr2_agpr3_agpr4_agpr5_agpr6_agpr7_agpr8_agpr9_agpr10_agpr11_agpr12_agpr13_agpr14_agpr15:0x00000000FFFFFFFF
|
|
; CHECK-NEXT: {{ $}}
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr0_vgpr1_vgpr2_vgpr3_vgpr4_vgpr5_vgpr6_vgpr7
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr8_vgpr9_vgpr10_vgpr11_vgpr12_vgpr13_vgpr14_vgpr15
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr16_vgpr17_vgpr18_vgpr19_vgpr20_vgpr21_vgpr22_vgpr23
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr24_vgpr25_vgpr26_vgpr27_vgpr28_vgpr29_vgpr30_vgpr31
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr32_vgpr33_vgpr34_vgpr35_vgpr36_vgpr37_vgpr38_vgpr39
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr40_vgpr41_vgpr42_vgpr43_vgpr44_vgpr45_vgpr46_vgpr47
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr48_vgpr49_vgpr50_vgpr51_vgpr52_vgpr53_vgpr54_vgpr55
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr56_vgpr57_vgpr58_vgpr59_vgpr60_vgpr61_vgpr62_vgpr63
|
|
; CHECK-NEXT: renamable $vgpr0 = V_MOV_B32_e32 0, implicit $exec
|
|
; CHECK-NEXT: GLOBAL_STORE_DWORDX4_SADDR renamable $vgpr0, renamable $agpr8_agpr9_agpr10_agpr11, undef $sgpr0_sgpr1, 32, 0, implicit $exec :: (store (s128), align 32, addrspace 1)
|
|
; CHECK-NEXT: GLOBAL_STORE_DWORDX4_SADDR renamable $vgpr0, renamable $agpr12_agpr13_agpr14_agpr15, undef $sgpr0_sgpr1, 48, 0, implicit $exec :: (store (s128), addrspace 1)
|
|
; CHECK-NEXT: GLOBAL_STORE_DWORDX4_SADDR renamable $vgpr0, renamable $agpr0_agpr1_agpr2_agpr3, undef $sgpr0_sgpr1, 0, 0, implicit $exec :: (store (s128), align 128, addrspace 1)
|
|
; CHECK-NEXT: GLOBAL_STORE_DWORDX4_SADDR killed renamable $vgpr0, killed renamable $agpr4_agpr5_agpr6_agpr7, killed undef $sgpr0_sgpr1, 16, 0, implicit $exec :: (store (s128), addrspace 1)
|
|
; CHECK-NEXT: S_ENDPGM 0
|
|
bb.0:
|
|
S_NOP 0, implicit-def $agpr0
|
|
renamable $sgpr0 = S_MOV_B32 0
|
|
undef %0.sub8:vreg_512_align2 = V_MOV_B32_e32 0, implicit $exec
|
|
renamable $sgpr1 = COPY renamable $sgpr0
|
|
%1:vreg_64_align2 = COPY killed renamable $sgpr0_sgpr1
|
|
renamable $vcc = S_AND_B64 $exec, -1, implicit-def dead $scc
|
|
%0.sub9:vreg_512_align2 = COPY %0.sub8
|
|
|
|
bb.1:
|
|
liveins: $vcc
|
|
|
|
undef %0.sub0_sub1:vreg_512_align2 = GLOBAL_LOAD_DWORDX2 undef %3:vreg_64_align2, 0, 0, implicit $exec :: (load (s64), addrspace 1)
|
|
%4:vreg_512_align2 = V_MFMA_F32_32X32X8F16_vgprcd_e64 %1, %1, %0, 0, 0, 0, implicit $mode, implicit $exec
|
|
%4:vreg_512_align2 = V_MFMA_F32_32X32X8F16_mac_vgprcd_e64 %1, %1, %4, 0, 0, 0, implicit $mode, implicit $exec
|
|
S_CBRANCH_VCCNZ %bb.1, implicit $vcc
|
|
S_BRANCH %bb.2
|
|
|
|
bb.2:
|
|
; No VGPRs available for %0
|
|
S_NOP 0, implicit-def $vgpr0_vgpr1_vgpr2_vgpr3_vgpr4_vgpr5_vgpr6_vgpr7
|
|
S_NOP 0, implicit-def $vgpr8_vgpr9_vgpr10_vgpr11_vgpr12_vgpr13_vgpr14_vgpr15
|
|
S_NOP 0, implicit-def $vgpr16_vgpr17_vgpr18_vgpr19_vgpr20_vgpr21_vgpr22_vgpr23
|
|
S_NOP 0, implicit-def $vgpr24_vgpr25_vgpr26_vgpr27_vgpr28_vgpr29_vgpr30_vgpr31
|
|
S_NOP 0, implicit-def $vgpr32_vgpr33_vgpr34_vgpr35_vgpr36_vgpr37_vgpr38_vgpr39
|
|
S_NOP 0, implicit-def $vgpr40_vgpr41_vgpr42_vgpr43_vgpr44_vgpr45_vgpr46_vgpr47
|
|
S_NOP 0, implicit-def $vgpr48_vgpr49_vgpr50_vgpr51_vgpr52_vgpr53_vgpr54_vgpr55
|
|
S_NOP 0, implicit-def $vgpr56_vgpr57_vgpr58_vgpr59_vgpr60_vgpr61_vgpr62_vgpr63
|
|
%5:vgpr_32 = V_MOV_B32_e32 0, implicit $exec
|
|
GLOBAL_STORE_DWORDX4_SADDR %5, %4.sub8_sub9_sub10_sub11, undef $sgpr0_sgpr1, 32, 0, implicit $exec :: (store (s128), align 32, addrspace 1)
|
|
GLOBAL_STORE_DWORDX4_SADDR %5, %4.sub12_sub13_sub14_sub15, undef $sgpr0_sgpr1, 48, 0, implicit $exec :: (store (s128), addrspace 1)
|
|
GLOBAL_STORE_DWORDX4_SADDR %5, %4.sub0_sub1_sub2_sub3, undef $sgpr0_sgpr1, 0, 0, implicit $exec :: (store (s128), align 128, addrspace 1)
|
|
GLOBAL_STORE_DWORDX4_SADDR %5, %4.sub4_sub5_sub6_sub7, killed undef $sgpr0_sgpr1, 16, 0, implicit $exec :: (store (s128), addrspace 1)
|
|
S_ENDPGM 0
|
|
|
|
...
|
|
|
|
# Non-mac variant, src2 is an immediate.
|
|
---
|
|
name: inflate_result_to_agpr__V_MFMA_F32_32X32X8F16_vgprcd_e64_imm_src2
|
|
tracksRegLiveness: true
|
|
machineFunctionInfo:
|
|
isEntryFunction: true
|
|
stackPtrOffsetReg: '$sgpr32'
|
|
occupancy: 10
|
|
sgprForEXECCopy: '$sgpr100_sgpr101'
|
|
body: |
|
|
; CHECK-LABEL: name: inflate_result_to_agpr__V_MFMA_F32_32X32X8F16_vgprcd_e64_imm_src2
|
|
; CHECK: bb.0:
|
|
; CHECK-NEXT: successors: %bb.1(0x80000000)
|
|
; CHECK-NEXT: {{ $}}
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $agpr0
|
|
; CHECK-NEXT: renamable $sgpr0 = S_MOV_B32 0
|
|
; CHECK-NEXT: renamable $vgpr8 = V_MOV_B32_e32 0, implicit $exec
|
|
; CHECK-NEXT: renamable $sgpr1 = COPY renamable $sgpr0
|
|
; CHECK-NEXT: renamable $vgpr0_vgpr1 = COPY killed renamable $sgpr0_sgpr1
|
|
; CHECK-NEXT: renamable $vcc = S_AND_B64 $exec, -1, implicit-def dead $scc
|
|
; CHECK-NEXT: dead renamable $vgpr9 = COPY renamable $vgpr8
|
|
; CHECK-NEXT: {{ $}}
|
|
; CHECK-NEXT: bb.1:
|
|
; CHECK-NEXT: successors: %bb.1(0x40000000), %bb.2(0x40000000)
|
|
; CHECK-NEXT: liveins: $vcc, $vgpr0_vgpr1
|
|
; CHECK-NEXT: {{ $}}
|
|
; CHECK-NEXT: early-clobber renamable $agpr0_agpr1_agpr2_agpr3_agpr4_agpr5_agpr6_agpr7_agpr8_agpr9_agpr10_agpr11_agpr12_agpr13_agpr14_agpr15 = V_MFMA_F32_32X32X8F16_e64 $vgpr0_vgpr1, $vgpr0_vgpr1, 0, 0, 0, 0, implicit $mode, implicit $exec
|
|
; CHECK-NEXT: S_CBRANCH_VCCNZ %bb.1, implicit $vcc
|
|
; CHECK-NEXT: S_BRANCH %bb.2
|
|
; CHECK-NEXT: {{ $}}
|
|
; CHECK-NEXT: bb.2:
|
|
; CHECK-NEXT: liveins: $agpr0_agpr1_agpr2_agpr3_agpr4_agpr5_agpr6_agpr7_agpr8_agpr9_agpr10_agpr11_agpr12_agpr13_agpr14_agpr15:0x00000000FFFFFFFF
|
|
; CHECK-NEXT: {{ $}}
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr0_vgpr1_vgpr2_vgpr3_vgpr4_vgpr5_vgpr6_vgpr7
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr8_vgpr9_vgpr10_vgpr11_vgpr12_vgpr13_vgpr14_vgpr15
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr16_vgpr17_vgpr18_vgpr19_vgpr20_vgpr21_vgpr22_vgpr23
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr24_vgpr25_vgpr26_vgpr27_vgpr28_vgpr29_vgpr30_vgpr31
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr32_vgpr33_vgpr34_vgpr35_vgpr36_vgpr37_vgpr38_vgpr39
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr40_vgpr41_vgpr42_vgpr43_vgpr44_vgpr45_vgpr46_vgpr47
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr48_vgpr49_vgpr50_vgpr51_vgpr52_vgpr53_vgpr54_vgpr55
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr56_vgpr57_vgpr58_vgpr59_vgpr60_vgpr61_vgpr62_vgpr63
|
|
; CHECK-NEXT: renamable $vgpr0 = V_MOV_B32_e32 0, implicit $exec
|
|
; CHECK-NEXT: GLOBAL_STORE_DWORDX4_SADDR renamable $vgpr0, renamable $agpr8_agpr9_agpr10_agpr11, undef $sgpr0_sgpr1, 32, 0, implicit $exec :: (store (s128), align 32, addrspace 1)
|
|
; CHECK-NEXT: GLOBAL_STORE_DWORDX4_SADDR renamable $vgpr0, renamable $agpr12_agpr13_agpr14_agpr15, undef $sgpr0_sgpr1, 48, 0, implicit $exec :: (store (s128), addrspace 1)
|
|
; CHECK-NEXT: GLOBAL_STORE_DWORDX4_SADDR renamable $vgpr0, renamable $agpr0_agpr1_agpr2_agpr3, undef $sgpr0_sgpr1, 0, 0, implicit $exec :: (store (s128), align 128, addrspace 1)
|
|
; CHECK-NEXT: GLOBAL_STORE_DWORDX4_SADDR killed renamable $vgpr0, killed renamable $agpr4_agpr5_agpr6_agpr7, killed undef $sgpr0_sgpr1, 16, 0, implicit $exec :: (store (s128), addrspace 1)
|
|
; CHECK-NEXT: S_ENDPGM 0
|
|
bb.0:
|
|
S_NOP 0, implicit-def $agpr0
|
|
renamable $sgpr0 = S_MOV_B32 0
|
|
undef %0.sub8:vreg_512_align2 = V_MOV_B32_e32 0, implicit $exec
|
|
renamable $sgpr1 = COPY renamable $sgpr0
|
|
%1:vreg_64_align2 = COPY killed renamable $sgpr0_sgpr1
|
|
renamable $vcc = S_AND_B64 $exec, -1, implicit-def dead $scc
|
|
%0.sub9:vreg_512_align2 = COPY %0.sub8
|
|
|
|
bb.1:
|
|
liveins: $vcc
|
|
|
|
%0:vreg_512_align2 = V_MFMA_F32_32X32X8F16_vgprcd_e64 %1, %1, 0, 0, 0, 0, implicit $mode, implicit $exec
|
|
S_CBRANCH_VCCNZ %bb.1, implicit $vcc
|
|
S_BRANCH %bb.2
|
|
|
|
bb.2:
|
|
; No VGPRs available for %0
|
|
S_NOP 0, implicit-def $vgpr0_vgpr1_vgpr2_vgpr3_vgpr4_vgpr5_vgpr6_vgpr7
|
|
S_NOP 0, implicit-def $vgpr8_vgpr9_vgpr10_vgpr11_vgpr12_vgpr13_vgpr14_vgpr15
|
|
S_NOP 0, implicit-def $vgpr16_vgpr17_vgpr18_vgpr19_vgpr20_vgpr21_vgpr22_vgpr23
|
|
S_NOP 0, implicit-def $vgpr24_vgpr25_vgpr26_vgpr27_vgpr28_vgpr29_vgpr30_vgpr31
|
|
S_NOP 0, implicit-def $vgpr32_vgpr33_vgpr34_vgpr35_vgpr36_vgpr37_vgpr38_vgpr39
|
|
S_NOP 0, implicit-def $vgpr40_vgpr41_vgpr42_vgpr43_vgpr44_vgpr45_vgpr46_vgpr47
|
|
S_NOP 0, implicit-def $vgpr48_vgpr49_vgpr50_vgpr51_vgpr52_vgpr53_vgpr54_vgpr55
|
|
S_NOP 0, implicit-def $vgpr56_vgpr57_vgpr58_vgpr59_vgpr60_vgpr61_vgpr62_vgpr63
|
|
%2:vgpr_32 = V_MOV_B32_e32 0, implicit $exec
|
|
GLOBAL_STORE_DWORDX4_SADDR %2, %0.sub8_sub9_sub10_sub11, undef $sgpr0_sgpr1, 32, 0, implicit $exec :: (store (s128), align 32, addrspace 1)
|
|
GLOBAL_STORE_DWORDX4_SADDR %2, %0.sub12_sub13_sub14_sub15, undef $sgpr0_sgpr1, 48, 0, implicit $exec :: (store (s128), addrspace 1)
|
|
GLOBAL_STORE_DWORDX4_SADDR %2, %0.sub0_sub1_sub2_sub3, undef $sgpr0_sgpr1, 0, 0, implicit $exec :: (store (s128), align 128, addrspace 1)
|
|
GLOBAL_STORE_DWORDX4_SADDR %2, %0.sub4_sub5_sub6_sub7, killed undef $sgpr0_sgpr1, 16, 0, implicit $exec :: (store (s128), addrspace 1)
|
|
S_ENDPGM 0
|
|
|
|
...
|
|
|
|
# Non-mac variant, src2 is the same VGPR, but a different subregister.
|
|
---
|
|
name: inflate_result_to_agpr__V_MFMA_F32_32X32X8F16_vgprcd_e64_src2_different_subreg
|
|
tracksRegLiveness: true
|
|
machineFunctionInfo:
|
|
isEntryFunction: true
|
|
stackPtrOffsetReg: '$sgpr32'
|
|
occupancy: 10
|
|
sgprForEXECCopy: '$sgpr100_sgpr101'
|
|
body: |
|
|
; CHECK-LABEL: name: inflate_result_to_agpr__V_MFMA_F32_32X32X8F16_vgprcd_e64_src2_different_subreg
|
|
; CHECK: bb.0:
|
|
; CHECK-NEXT: successors: %bb.1(0x80000000)
|
|
; CHECK-NEXT: {{ $}}
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $agpr0
|
|
; CHECK-NEXT: renamable $sgpr0 = S_MOV_B32 0
|
|
; CHECK-NEXT: renamable $vgpr8 = V_MOV_B32_e32 0, implicit $exec
|
|
; CHECK-NEXT: renamable $sgpr1 = COPY renamable $sgpr0
|
|
; CHECK-NEXT: renamable $vgpr0_vgpr1 = COPY killed renamable $sgpr0_sgpr1
|
|
; CHECK-NEXT: renamable $vcc = S_AND_B64 $exec, -1, implicit-def dead $scc
|
|
; CHECK-NEXT: dead renamable $vgpr9 = COPY renamable $vgpr8
|
|
; CHECK-NEXT: {{ $}}
|
|
; CHECK-NEXT: bb.1:
|
|
; CHECK-NEXT: successors: %bb.1(0x40000000), %bb.2(0x40000000)
|
|
; CHECK-NEXT: liveins: $vcc, $vgpr0_vgpr1
|
|
; CHECK-NEXT: {{ $}}
|
|
; CHECK-NEXT: renamable $agpr0_agpr1 = GLOBAL_LOAD_DWORDX2 undef renamable $vgpr0_vgpr1, 0, 0, implicit $exec :: (load (s64), addrspace 1)
|
|
; CHECK-NEXT: renamable $agpr16_agpr17 = GLOBAL_LOAD_DWORDX2 undef renamable $vgpr0_vgpr1, 0, 0, implicit $exec :: (load (s64), addrspace 1)
|
|
; CHECK-NEXT: early-clobber renamable $agpr0_agpr1_agpr2_agpr3_agpr4_agpr5_agpr6_agpr7_agpr8_agpr9_agpr10_agpr11_agpr12_agpr13_agpr14_agpr15 = V_MFMA_F32_32X32X8F16_e64 $vgpr0_vgpr1, $vgpr0_vgpr1, $agpr16_agpr17_agpr18_agpr19_agpr20_agpr21_agpr22_agpr23_agpr24_agpr25_agpr26_agpr27_agpr28_agpr29_agpr30_agpr31, 0, 0, 0, implicit $mode, implicit $exec
|
|
; CHECK-NEXT: S_CBRANCH_VCCNZ %bb.1, implicit $vcc
|
|
; CHECK-NEXT: S_BRANCH %bb.2
|
|
; CHECK-NEXT: {{ $}}
|
|
; CHECK-NEXT: bb.2:
|
|
; CHECK-NEXT: liveins: $agpr0_agpr1_agpr2_agpr3_agpr4_agpr5_agpr6_agpr7_agpr8_agpr9_agpr10_agpr11_agpr12_agpr13_agpr14_agpr15_agpr16_agpr17_agpr18_agpr19_agpr20_agpr21_agpr22_agpr23_agpr24_agpr25_agpr26_agpr27_agpr28_agpr29_agpr30_agpr31:0x00000000FFFFFFFF
|
|
; CHECK-NEXT: {{ $}}
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr0_vgpr1_vgpr2_vgpr3_vgpr4_vgpr5_vgpr6_vgpr7
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr8_vgpr9_vgpr10_vgpr11_vgpr12_vgpr13_vgpr14_vgpr15
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr16_vgpr17_vgpr18_vgpr19_vgpr20_vgpr21_vgpr22_vgpr23
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr24_vgpr25_vgpr26_vgpr27_vgpr28_vgpr29_vgpr30_vgpr31
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr32_vgpr33_vgpr34_vgpr35_vgpr36_vgpr37_vgpr38_vgpr39
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr40_vgpr41_vgpr42_vgpr43_vgpr44_vgpr45_vgpr46_vgpr47
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr48_vgpr49_vgpr50_vgpr51_vgpr52_vgpr53_vgpr54_vgpr55
|
|
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr56_vgpr57_vgpr58_vgpr59_vgpr60_vgpr61_vgpr62_vgpr63
|
|
; CHECK-NEXT: renamable $vgpr0 = V_MOV_B32_e32 0, implicit $exec
|
|
; CHECK-NEXT: GLOBAL_STORE_DWORDX4_SADDR renamable $vgpr0, renamable $agpr8_agpr9_agpr10_agpr11, undef $sgpr0_sgpr1, 32, 0, implicit $exec :: (store (s128), align 32, addrspace 1)
|
|
; CHECK-NEXT: GLOBAL_STORE_DWORDX4_SADDR renamable $vgpr0, renamable $agpr12_agpr13_agpr14_agpr15, undef $sgpr0_sgpr1, 48, 0, implicit $exec :: (store (s128), addrspace 1)
|
|
; CHECK-NEXT: GLOBAL_STORE_DWORDX4_SADDR renamable $vgpr0, renamable $agpr0_agpr1_agpr2_agpr3, undef $sgpr0_sgpr1, 0, 0, implicit $exec :: (store (s128), align 128, addrspace 1)
|
|
; CHECK-NEXT: GLOBAL_STORE_DWORDX4_SADDR killed renamable $vgpr0, killed renamable $agpr4_agpr5_agpr6_agpr7, killed undef $sgpr0_sgpr1, 16, 0, implicit $exec :: (store (s128), addrspace 1)
|
|
; CHECK-NEXT: S_ENDPGM 0
|
|
bb.0:
|
|
S_NOP 0, implicit-def $agpr0
|
|
renamable $sgpr0 = S_MOV_B32 0
|
|
undef %0.sub8:vreg_1024_align2 = V_MOV_B32_e32 0, implicit $exec
|
|
renamable $sgpr1 = COPY renamable $sgpr0
|
|
%1:vreg_64_align2 = COPY killed renamable $sgpr0_sgpr1
|
|
renamable $vcc = S_AND_B64 $exec, -1, implicit-def dead $scc
|
|
%0.sub9:vreg_1024_align2 = COPY %0.sub8
|
|
|
|
bb.1:
|
|
liveins: $vcc
|
|
|
|
undef %0.sub0_sub1:vreg_1024_align2 = GLOBAL_LOAD_DWORDX2 undef %3:vreg_64_align2, 0, 0, implicit $exec :: (load (s64), addrspace 1)
|
|
%0.sub16_sub17:vreg_1024_align2 = GLOBAL_LOAD_DWORDX2 undef %3:vreg_64_align2, 0, 0, implicit $exec :: (load (s64), addrspace 1)
|
|
%0.sub0_sub1_sub2_sub3_sub4_sub5_sub6_sub7_sub8_sub9_sub10_sub11_sub12_sub13_sub14_sub15:vreg_1024_align2 = V_MFMA_F32_32X32X8F16_vgprcd_e64 %1, %1, %0.sub16_sub17_sub18_sub19_sub20_sub21_sub22_sub23_sub24_sub25_sub26_sub27_sub28_sub29_sub30_sub31, 0, 0, 0, implicit $mode, implicit $exec
|
|
S_CBRANCH_VCCNZ %bb.1, implicit $vcc
|
|
S_BRANCH %bb.2
|
|
|
|
bb.2:
|
|
; No VGPRs available for %0
|
|
S_NOP 0, implicit-def $vgpr0_vgpr1_vgpr2_vgpr3_vgpr4_vgpr5_vgpr6_vgpr7
|
|
S_NOP 0, implicit-def $vgpr8_vgpr9_vgpr10_vgpr11_vgpr12_vgpr13_vgpr14_vgpr15
|
|
S_NOP 0, implicit-def $vgpr16_vgpr17_vgpr18_vgpr19_vgpr20_vgpr21_vgpr22_vgpr23
|
|
S_NOP 0, implicit-def $vgpr24_vgpr25_vgpr26_vgpr27_vgpr28_vgpr29_vgpr30_vgpr31
|
|
S_NOP 0, implicit-def $vgpr32_vgpr33_vgpr34_vgpr35_vgpr36_vgpr37_vgpr38_vgpr39
|
|
S_NOP 0, implicit-def $vgpr40_vgpr41_vgpr42_vgpr43_vgpr44_vgpr45_vgpr46_vgpr47
|
|
S_NOP 0, implicit-def $vgpr48_vgpr49_vgpr50_vgpr51_vgpr52_vgpr53_vgpr54_vgpr55
|
|
S_NOP 0, implicit-def $vgpr56_vgpr57_vgpr58_vgpr59_vgpr60_vgpr61_vgpr62_vgpr63
|
|
%2:vgpr_32 = V_MOV_B32_e32 0, implicit $exec
|
|
GLOBAL_STORE_DWORDX4_SADDR %2, %0.sub8_sub9_sub10_sub11, undef $sgpr0_sgpr1, 32, 0, implicit $exec :: (store (s128), align 32, addrspace 1)
|
|
GLOBAL_STORE_DWORDX4_SADDR %2, %0.sub12_sub13_sub14_sub15, undef $sgpr0_sgpr1, 48, 0, implicit $exec :: (store (s128), addrspace 1)
|
|
GLOBAL_STORE_DWORDX4_SADDR %2, %0.sub0_sub1_sub2_sub3, undef $sgpr0_sgpr1, 0, 0, implicit $exec :: (store (s128), align 128, addrspace 1)
|
|
GLOBAL_STORE_DWORDX4_SADDR %2, %0.sub4_sub5_sub6_sub7, killed undef $sgpr0_sgpr1, 16, 0, implicit $exec :: (store (s128), addrspace 1)
|
|
S_ENDPGM 0
|
|
|
|
...
|