derek.holloway
7502bf64a4
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
160 lines
6.4 KiB
LLVM
160 lines
6.4 KiB
LLVM
; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py UTC_ARGS: --version 5
|
|
; RUN: llc < %s -mtriple=x86_64-linux | FileCheck %s --check-prefix=X64
|
|
|
|
; Verify that we can look through a ZERO_EXTEND/ANY_EXTEND when doing
|
|
; ComputeNumSignBits for SHL.
|
|
; We use the (sshlsat x, c) -> (shl x, c) fold as verification.
|
|
; That fold should happen if c is less than the number of sign bits in x
|
|
|
|
define void @computeNumSignBits_shl_zext_1(i8 %x, ptr %p) nounwind {
|
|
; X64-LABEL: computeNumSignBits_shl_zext_1:
|
|
; X64: # %bb.0:
|
|
; X64-NEXT: sarb $5, %dil
|
|
; X64-NEXT: movzbl %dil, %eax
|
|
; X64-NEXT: movl %eax, %ecx
|
|
; X64-NEXT: shll $11, %ecx
|
|
; X64-NEXT: movw %cx, (%rsi)
|
|
; X64-NEXT: movl %eax, %ecx
|
|
; X64-NEXT: shll $12, %ecx
|
|
; X64-NEXT: movw %cx, (%rsi)
|
|
; X64-NEXT: shll $13, %eax
|
|
; X64-NEXT: movw %ax, (%rsi)
|
|
; X64-NEXT: retq
|
|
%ashr = ashr i8 %x, 5
|
|
%zext = zext i8 %ashr to i16
|
|
%nsb4 = shl i16 %zext, 10
|
|
; Expecting (sshlsat x, c) -> (shl x, c) fold.
|
|
%tmp1 = call i16 @llvm.sshl.sat.i16(i16 %nsb4, i16 1)
|
|
store volatile i16 %tmp1, ptr %p
|
|
; Expecting (sshlsat x, c) -> (shl x, c) fold.
|
|
%tmp2 = call i16 @llvm.sshl.sat.i16(i16 %nsb4, i16 2)
|
|
store volatile i16 %tmp2, ptr %p
|
|
; Expecting (sshlsat x, c) -> (shl x, c) fold.
|
|
%tmp3 = call i16 @llvm.sshl.sat.i16(i16 %nsb4, i16 3)
|
|
store volatile i16 %tmp3, ptr %p
|
|
ret void
|
|
}
|
|
|
|
define void @computeNumSignBits_shl_zext_2(i8 %x, ptr %p) nounwind {
|
|
; X64-LABEL: computeNumSignBits_shl_zext_2:
|
|
; X64: # %bb.0:
|
|
; X64-NEXT: sarb $5, %dil
|
|
; X64-NEXT: movzbl %dil, %eax
|
|
; X64-NEXT: movl %eax, %ecx
|
|
; X64-NEXT: shll $10, %ecx
|
|
; X64-NEXT: xorl %edx, %edx
|
|
; X64-NEXT: testw %cx, %cx
|
|
; X64-NEXT: sets %dl
|
|
; X64-NEXT: addl $32767, %edx # imm = 0x7FFF
|
|
; X64-NEXT: shll $14, %eax
|
|
; X64-NEXT: movswl %ax, %edi
|
|
; X64-NEXT: shrl $4, %edi
|
|
; X64-NEXT: cmpw %di, %cx
|
|
; X64-NEXT: cmovnel %edx, %eax
|
|
; X64-NEXT: movw %ax, (%rsi)
|
|
; X64-NEXT: retq
|
|
%ashr = ashr i8 %x, 5
|
|
%zext = zext i8 %ashr to i16
|
|
%nsb4 = shl i16 %zext, 10
|
|
; 4 sign bits. Not expecting (sshlsat x, c) -> (shl x, c) fold.
|
|
%tmp4 = call i16 @llvm.sshl.sat.i16(i16 %nsb4, i16 4)
|
|
store volatile i16 %tmp4, ptr %p
|
|
ret void
|
|
}
|
|
|
|
define void @computeNumSignBits_shl_zext_vec_1(<2 x i8> %x, ptr %p) nounwind {
|
|
; X64-LABEL: computeNumSignBits_shl_zext_vec_1:
|
|
; X64: # %bb.0:
|
|
; X64-NEXT: psrlw $5, %xmm0
|
|
; X64-NEXT: pand {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0
|
|
; X64-NEXT: movdqa {{.*#+}} xmm1 = [4,4,4,4,4,4,4,4,4,4,4,4,4,4,4,4]
|
|
; X64-NEXT: pxor %xmm1, %xmm0
|
|
; X64-NEXT: psubb %xmm1, %xmm0
|
|
; X64-NEXT: pxor %xmm1, %xmm1
|
|
; X64-NEXT: punpcklbw {{.*#+}} xmm0 = xmm0[0],xmm1[0],xmm0[1],xmm1[1],xmm0[2],xmm1[2],xmm0[3],xmm1[3],xmm0[4],xmm1[4],xmm0[5],xmm1[5],xmm0[6],xmm1[6],xmm0[7],xmm1[7]
|
|
; X64-NEXT: pmullw {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0 # [2048,8192,u,u,u,u,u,u]
|
|
; X64-NEXT: movd %xmm0, (%rdi)
|
|
; X64-NEXT: retq
|
|
%ashr = ashr <2 x i8> %x, <i8 5, i8 5>
|
|
%zext = zext <2 x i8> %ashr to <2 x i16>
|
|
%nsb4_2 = shl <2 x i16> %zext, <i16 10, i16 12>
|
|
; Expecting (sshlsat x, c) -> (shl x, c) fold.
|
|
%tmp1 = call <2 x i16> @llvm.sshl.sat.v2i16(<2 x i16> %nsb4_2, <2 x i16> <i16 1, i16 1>)
|
|
store volatile <2 x i16> %tmp1, ptr %p
|
|
ret void
|
|
}
|
|
|
|
define void @computeNumSignBits_shl_zext_vec_2(<2 x i8> %x, ptr %p) nounwind {
|
|
; X64-LABEL: computeNumSignBits_shl_zext_vec_2:
|
|
; X64: # %bb.0:
|
|
; X64-NEXT: psrlw $5, %xmm0
|
|
; X64-NEXT: pand {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0
|
|
; X64-NEXT: movdqa {{.*#+}} xmm1 = [4,4,4,4,4,4,4,4,4,4,4,4,4,4,4,4]
|
|
; X64-NEXT: pxor %xmm1, %xmm0
|
|
; X64-NEXT: psubb %xmm1, %xmm0
|
|
; X64-NEXT: pxor %xmm1, %xmm1
|
|
; X64-NEXT: punpcklbw {{.*#+}} xmm0 = xmm0[0],xmm1[0],xmm0[1],xmm1[1],xmm0[2],xmm1[2],xmm0[3],xmm1[3],xmm0[4],xmm1[4],xmm0[5],xmm1[5],xmm0[6],xmm1[6],xmm0[7],xmm1[7]
|
|
; X64-NEXT: pmullw {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0 # [1024,4096,u,u,u,u,u,u]
|
|
; X64-NEXT: movdqa {{.*#+}} xmm2 = [32768,32768,32768,32768,32768,32768,32768,32768]
|
|
; X64-NEXT: pand %xmm0, %xmm2
|
|
; X64-NEXT: pcmpgtw %xmm0, %xmm1
|
|
; X64-NEXT: pandn {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm1
|
|
; X64-NEXT: por %xmm2, %xmm1
|
|
; X64-NEXT: movdqa %xmm0, %xmm2
|
|
; X64-NEXT: psllw $2, %xmm2
|
|
; X64-NEXT: movdqa %xmm2, %xmm3
|
|
; X64-NEXT: psraw $2, %xmm3
|
|
; X64-NEXT: pcmpeqw %xmm0, %xmm3
|
|
; X64-NEXT: movdqa %xmm3, %xmm0
|
|
; X64-NEXT: pandn %xmm1, %xmm0
|
|
; X64-NEXT: pand %xmm2, %xmm3
|
|
; X64-NEXT: por %xmm0, %xmm3
|
|
; X64-NEXT: movd %xmm3, (%rdi)
|
|
; X64-NEXT: retq
|
|
%ashr = ashr <2 x i8> %x, <i8 5, i8 5>
|
|
%zext = zext <2 x i8> %ashr to <2 x i16>
|
|
%nsb4_2 = shl <2 x i16> %zext, <i16 10, i16 12>
|
|
; Not expecting (sshlsat x, c) -> (shl x, c) fold.
|
|
; Because only 2 sign bits in element 1.
|
|
%tmp1 = call <2 x i16> @llvm.sshl.sat.v2i16(<2 x i16> %nsb4_2, <2 x i16> <i16 2, i16 2>)
|
|
store volatile <2 x i16> %tmp1, ptr %p
|
|
ret void
|
|
}
|
|
|
|
define void @computeNumSignBits_shl_zext_vec_3(<2 x i8> %x, ptr %p) nounwind {
|
|
; X64-LABEL: computeNumSignBits_shl_zext_vec_3:
|
|
; X64: # %bb.0:
|
|
; X64-NEXT: psrlw $5, %xmm0
|
|
; X64-NEXT: pand {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0
|
|
; X64-NEXT: movdqa {{.*#+}} xmm1 = [4,4,4,4,4,4,4,4,4,4,4,4,4,4,4,4]
|
|
; X64-NEXT: pxor %xmm1, %xmm0
|
|
; X64-NEXT: psubb %xmm1, %xmm0
|
|
; X64-NEXT: pxor %xmm1, %xmm1
|
|
; X64-NEXT: punpcklbw {{.*#+}} xmm0 = xmm0[0],xmm1[0],xmm0[1],xmm1[1],xmm0[2],xmm1[2],xmm0[3],xmm1[3],xmm0[4],xmm1[4],xmm0[5],xmm1[5],xmm0[6],xmm1[6],xmm0[7],xmm1[7]
|
|
; X64-NEXT: pmullw {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0 # [16384,4096,u,u,u,u,u,u]
|
|
; X64-NEXT: movdqa {{.*#+}} xmm2 = [32768,32768,32768,32768,32768,32768,32768,32768]
|
|
; X64-NEXT: pand %xmm0, %xmm2
|
|
; X64-NEXT: pcmpgtw %xmm0, %xmm1
|
|
; X64-NEXT: pandn {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm1
|
|
; X64-NEXT: por %xmm2, %xmm1
|
|
; X64-NEXT: movdqa %xmm0, %xmm2
|
|
; X64-NEXT: paddw %xmm2, %xmm2
|
|
; X64-NEXT: movdqa %xmm2, %xmm3
|
|
; X64-NEXT: psraw $1, %xmm3
|
|
; X64-NEXT: pcmpeqw %xmm0, %xmm3
|
|
; X64-NEXT: movdqa %xmm3, %xmm0
|
|
; X64-NEXT: pandn %xmm1, %xmm0
|
|
; X64-NEXT: pand %xmm2, %xmm3
|
|
; X64-NEXT: por %xmm0, %xmm3
|
|
; X64-NEXT: movd %xmm3, (%rdi)
|
|
; X64-NEXT: retq
|
|
%ashr = ashr <2 x i8> %x, <i8 5, i8 5>
|
|
%zext = zext <2 x i8> %ashr to <2 x i16>
|
|
%nsb1_2 = shl <2 x i16> %zext, <i16 14, i16 12>
|
|
; Not expecting (sshlsat x, c) -> (shl x, c) fold.
|
|
; Because all sign bits shifted out for element 0
|
|
%tmp1 = call <2 x i16> @llvm.sshl.sat.v2i16(<2 x i16> %nsb1_2, <2 x i16> <i16 1, i16 1>)
|
|
store volatile <2 x i16> %tmp1, ptr %p
|
|
ret void
|
|
}
|