maxjhandsome
/
llvm-project
forked from OSchip/llvm-project
1
0
Fork 0
Code Issues Pull Requests Packages Releases Wiki Activity
llvm-project/llvm/test/CodeGen/X86/fma-scalar-memfold.ll

402 lines
16 KiB
LLVM
Raw Normal View History

[AVX512] Add AVX512 run lines to some tests for scalar fma/add/sub/mul/div and regenerate. Follow up commits will bring AVX512 code up to the same quality as AVX/SSE. llvm-svn: 277118
2016-07-29 14:05:58 +08:00
; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
; RUN: llc < %s -mcpu=core-avx2 | FileCheck %s --check-prefix=CHECK --check-prefix=AVX2
; RUN: llc < %s -mcpu=skx | FileCheck %s --check-prefix=CHECK --check-prefix=AVX512
X86-FMA3: Improved/enabled the memory folding optimization for scalar loads generated for _mm_losd_s{s,d}() intrinsics and used in scalar FMAs generated for FMA intrinsics _mm_f{madd,msub,nmadd,nmsub}_s{s,d}(). Reviewer: David Kreitzer Differential Revision: http://reviews.llvm.org/D14762 llvm-svn: 254140
2015-11-26 15:45:30 +08:00
[AVX512] Add AVX512 run lines to some tests for scalar fma/add/sub/mul/div and regenerate. Follow up commits will bring AVX512 code up to the same quality as AVX/SSE. llvm-svn: 277118
2016-07-29 14:05:58 +08:00
target triple = "x86_64-unknown-unknown"
X86-FMA3: Improved/enabled the memory folding optimization for scalar loads generated for _mm_losd_s{s,d}() intrinsics and used in scalar FMAs generated for FMA intrinsics _mm_f{madd,msub,nmadd,nmsub}_s{s,d}(). Reviewer: David Kreitzer Differential Revision: http://reviews.llvm.org/D14762 llvm-svn: 254140
2015-11-26 15:45:30 +08:00
declare <4 x float> @llvm.x86.fma.vfmadd.ss(<4 x float>, <4 x float>, <4 x float>)
declare <4 x float> @llvm.x86.fma.vfmsub.ss(<4 x float>, <4 x float>, <4 x float>)
declare <4 x float> @llvm.x86.fma.vfnmadd.ss(<4 x float>, <4 x float>, <4 x float>)
declare <4 x float> @llvm.x86.fma.vfnmsub.ss(<4 x float>, <4 x float>, <4 x float>)
declare <2 x double> @llvm.x86.fma.vfmadd.sd(<2 x double>, <2 x double>, <2 x double>)
declare <2 x double> @llvm.x86.fma.vfmsub.sd(<2 x double>, <2 x double>, <2 x double>)
declare <2 x double> @llvm.x86.fma.vfnmadd.sd(<2 x double>, <2 x double>, <2 x double>)
declare <2 x double> @llvm.x86.fma.vfnmsub.sd(<2 x double>, <2 x double>, <2 x double>)
[AVX512] Add AVX512 run lines to some tests for scalar fma/add/sub/mul/div and regenerate. Follow up commits will bring AVX512 code up to the same quality as AVX/SSE. llvm-svn: 277118
2016-07-29 14:05:58 +08:00
define void @fmadd_aab_ss(float* %a, float* %b) {
[AVX512] Mark EVEX VMOVSSrm and VMOVSDrm as canFoldAsLoad and isReMaterializable. llvm-svn: 277120
2016-07-29 14:06:04 +08:00
; CHECK-LABEL: fmadd_aab_ss:
[CodeGen] Unify MBB reference format in both MIR and debug output As part of the unification of the debug format and the MIR format, print MBB references as '%bb.5'. The MIR printer prints the IR name of a MBB only for block definitions. * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)->getNumber\(\)/" << printMBBReference(*\1)/g' * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)\.getNumber\(\)/" << printMBBReference(\1)/g' * find . \( -name "*.txt" -o -name "*.s" -o -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#([0-9]+)/%bb.\1/g' * grep -nr 'BB#' and fix Differential Revision: https://reviews.llvm.org/D40422 llvm-svn: 319665
2017-12-05 01:18:51 +08:00
; CHECK: # %bb.0:
[AVX512] Mark EVEX VMOVSSrm and VMOVSDrm as canFoldAsLoad and isReMaterializable. llvm-svn: 277120
2016-07-29 14:06:04 +08:00
; CHECK-NEXT: vmovss {{.*#+}} xmm0 = mem[0],zero,zero,zero
; CHECK-NEXT: vfmadd213ss (%rsi), %xmm0, %xmm0
; CHECK-NEXT: vmovss %xmm0, (%rdi)
; CHECK-NEXT: retq
X86-FMA3: Improved/enabled the memory folding optimization for scalar loads generated for _mm_losd_s{s,d}() intrinsics and used in scalar FMAs generated for FMA intrinsics _mm_f{madd,msub,nmadd,nmsub}_s{s,d}(). Reviewer: David Kreitzer Differential Revision: http://reviews.llvm.org/D14762 llvm-svn: 254140
2015-11-26 15:45:30 +08:00
%a.val = load float, float* %a
%av0 = insertelement <4 x float> undef, float %a.val, i32 0
%av1 = insertelement <4 x float> %av0, float 0.000000e+00, i32 1
%av2 = insertelement <4 x float> %av1, float 0.000000e+00, i32 2
%av = insertelement <4 x float> %av2, float 0.000000e+00, i32 3
%b.val = load float, float* %b
%bv0 = insertelement <4 x float> undef, float %b.val, i32 0
%bv1 = insertelement <4 x float> %bv0, float 0.000000e+00, i32 1
%bv2 = insertelement <4 x float> %bv1, float 0.000000e+00, i32 2
%bv = insertelement <4 x float> %bv2, float 0.000000e+00, i32 3
%vr = call <4 x float> @llvm.x86.fma.vfmadd.ss(<4 x float> %av, <4 x float> %av, <4 x float> %bv)
%sr = extractelement <4 x float> %vr, i32 0
store float %sr, float* %a
ret void
}
[AVX512] Add AVX512 run lines to some tests for scalar fma/add/sub/mul/div and regenerate. Follow up commits will bring AVX512 code up to the same quality as AVX/SSE. llvm-svn: 277118
2016-07-29 14:05:58 +08:00
define void @fmadd_aba_ss(float* %a, float* %b) {
[AVX512] Mark EVEX VMOVSSrm and VMOVSDrm as canFoldAsLoad and isReMaterializable. llvm-svn: 277120
2016-07-29 14:06:04 +08:00
; CHECK-LABEL: fmadd_aba_ss:
[CodeGen] Unify MBB reference format in both MIR and debug output As part of the unification of the debug format and the MIR format, print MBB references as '%bb.5'. The MIR printer prints the IR name of a MBB only for block definitions. * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)->getNumber\(\)/" << printMBBReference(*\1)/g' * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)\.getNumber\(\)/" << printMBBReference(\1)/g' * find . \( -name "*.txt" -o -name "*.s" -o -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#([0-9]+)/%bb.\1/g' * grep -nr 'BB#' and fix Differential Revision: https://reviews.llvm.org/D40422 llvm-svn: 319665
2017-12-05 01:18:51 +08:00
; CHECK: # %bb.0:
[AVX512] Mark EVEX VMOVSSrm and VMOVSDrm as canFoldAsLoad and isReMaterializable. llvm-svn: 277120
2016-07-29 14:06:04 +08:00
; CHECK-NEXT: vmovss {{.*#+}} xmm0 = mem[0],zero,zero,zero
; CHECK-NEXT: vfmadd132ss (%rsi), %xmm0, %xmm0
; CHECK-NEXT: vmovss %xmm0, (%rdi)
; CHECK-NEXT: retq
X86-FMA3: Improved/enabled the memory folding optimization for scalar loads generated for _mm_losd_s{s,d}() intrinsics and used in scalar FMAs generated for FMA intrinsics _mm_f{madd,msub,nmadd,nmsub}_s{s,d}(). Reviewer: David Kreitzer Differential Revision: http://reviews.llvm.org/D14762 llvm-svn: 254140
2015-11-26 15:45:30 +08:00
%a.val = load float, float* %a
%av0 = insertelement <4 x float> undef, float %a.val, i32 0
%av1 = insertelement <4 x float> %av0, float 0.000000e+00, i32 1
%av2 = insertelement <4 x float> %av1, float 0.000000e+00, i32 2
%av = insertelement <4 x float> %av2, float 0.000000e+00, i32 3
%b.val = load float, float* %b
%bv0 = insertelement <4 x float> undef, float %b.val, i32 0
%bv1 = insertelement <4 x float> %bv0, float 0.000000e+00, i32 1
%bv2 = insertelement <4 x float> %bv1, float 0.000000e+00, i32 2
%bv = insertelement <4 x float> %bv2, float 0.000000e+00, i32 3
%vr = call <4 x float> @llvm.x86.fma.vfmadd.ss(<4 x float> %av, <4 x float> %bv, <4 x float> %av)
%sr = extractelement <4 x float> %vr, i32 0
store float %sr, float* %a
ret void
}
[AVX512] Add AVX512 run lines to some tests for scalar fma/add/sub/mul/div and regenerate. Follow up commits will bring AVX512 code up to the same quality as AVX/SSE. llvm-svn: 277118
2016-07-29 14:05:58 +08:00
define void @fmsub_aab_ss(float* %a, float* %b) {
[AVX512] Mark EVEX VMOVSSrm and VMOVSDrm as canFoldAsLoad and isReMaterializable. llvm-svn: 277120
2016-07-29 14:06:04 +08:00
; CHECK-LABEL: fmsub_aab_ss:
[CodeGen] Unify MBB reference format in both MIR and debug output As part of the unification of the debug format and the MIR format, print MBB references as '%bb.5'. The MIR printer prints the IR name of a MBB only for block definitions. * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)->getNumber\(\)/" << printMBBReference(*\1)/g' * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)\.getNumber\(\)/" << printMBBReference(\1)/g' * find . \( -name "*.txt" -o -name "*.s" -o -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#([0-9]+)/%bb.\1/g' * grep -nr 'BB#' and fix Differential Revision: https://reviews.llvm.org/D40422 llvm-svn: 319665
2017-12-05 01:18:51 +08:00
; CHECK: # %bb.0:
[AVX512] Mark EVEX VMOVSSrm and VMOVSDrm as canFoldAsLoad and isReMaterializable. llvm-svn: 277120
2016-07-29 14:06:04 +08:00
; CHECK-NEXT: vmovss {{.*#+}} xmm0 = mem[0],zero,zero,zero
; CHECK-NEXT: vfmsub213ss (%rsi), %xmm0, %xmm0
; CHECK-NEXT: vmovss %xmm0, (%rdi)
; CHECK-NEXT: retq
X86-FMA3: Improved/enabled the memory folding optimization for scalar loads generated for _mm_losd_s{s,d}() intrinsics and used in scalar FMAs generated for FMA intrinsics _mm_f{madd,msub,nmadd,nmsub}_s{s,d}(). Reviewer: David Kreitzer Differential Revision: http://reviews.llvm.org/D14762 llvm-svn: 254140
2015-11-26 15:45:30 +08:00
%a.val = load float, float* %a
%av0 = insertelement <4 x float> undef, float %a.val, i32 0
%av1 = insertelement <4 x float> %av0, float 0.000000e+00, i32 1
%av2 = insertelement <4 x float> %av1, float 0.000000e+00, i32 2
%av = insertelement <4 x float> %av2, float 0.000000e+00, i32 3
%b.val = load float, float* %b
%bv0 = insertelement <4 x float> undef, float %b.val, i32 0
%bv1 = insertelement <4 x float> %bv0, float 0.000000e+00, i32 1
%bv2 = insertelement <4 x float> %bv1, float 0.000000e+00, i32 2
%bv = insertelement <4 x float> %bv2, float 0.000000e+00, i32 3
%vr = call <4 x float> @llvm.x86.fma.vfmsub.ss(<4 x float> %av, <4 x float> %av, <4 x float> %bv)
%sr = extractelement <4 x float> %vr, i32 0
store float %sr, float* %a
ret void
}
[AVX512] Add AVX512 run lines to some tests for scalar fma/add/sub/mul/div and regenerate. Follow up commits will bring AVX512 code up to the same quality as AVX/SSE. llvm-svn: 277118
2016-07-29 14:05:58 +08:00
define void @fmsub_aba_ss(float* %a, float* %b) {
[AVX512] Mark EVEX VMOVSSrm and VMOVSDrm as canFoldAsLoad and isReMaterializable. llvm-svn: 277120
2016-07-29 14:06:04 +08:00
; CHECK-LABEL: fmsub_aba_ss:
[CodeGen] Unify MBB reference format in both MIR and debug output As part of the unification of the debug format and the MIR format, print MBB references as '%bb.5'. The MIR printer prints the IR name of a MBB only for block definitions. * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)->getNumber\(\)/" << printMBBReference(*\1)/g' * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)\.getNumber\(\)/" << printMBBReference(\1)/g' * find . \( -name "*.txt" -o -name "*.s" -o -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#([0-9]+)/%bb.\1/g' * grep -nr 'BB#' and fix Differential Revision: https://reviews.llvm.org/D40422 llvm-svn: 319665
2017-12-05 01:18:51 +08:00
; CHECK: # %bb.0:
[AVX512] Mark EVEX VMOVSSrm and VMOVSDrm as canFoldAsLoad and isReMaterializable. llvm-svn: 277120
2016-07-29 14:06:04 +08:00
; CHECK-NEXT: vmovss {{.*#+}} xmm0 = mem[0],zero,zero,zero
; CHECK-NEXT: vfmsub132ss (%rsi), %xmm0, %xmm0
; CHECK-NEXT: vmovss %xmm0, (%rdi)
; CHECK-NEXT: retq
X86-FMA3: Improved/enabled the memory folding optimization for scalar loads generated for _mm_losd_s{s,d}() intrinsics and used in scalar FMAs generated for FMA intrinsics _mm_f{madd,msub,nmadd,nmsub}_s{s,d}(). Reviewer: David Kreitzer Differential Revision: http://reviews.llvm.org/D14762 llvm-svn: 254140
2015-11-26 15:45:30 +08:00
%a.val = load float, float* %a
%av0 = insertelement <4 x float> undef, float %a.val, i32 0
%av1 = insertelement <4 x float> %av0, float 0.000000e+00, i32 1
%av2 = insertelement <4 x float> %av1, float 0.000000e+00, i32 2
%av = insertelement <4 x float> %av2, float 0.000000e+00, i32 3
%b.val = load float, float* %b
%bv0 = insertelement <4 x float> undef, float %b.val, i32 0
%bv1 = insertelement <4 x float> %bv0, float 0.000000e+00, i32 1
%bv2 = insertelement <4 x float> %bv1, float 0.000000e+00, i32 2
%bv = insertelement <4 x float> %bv2, float 0.000000e+00, i32 3
%vr = call <4 x float> @llvm.x86.fma.vfmsub.ss(<4 x float> %av, <4 x float> %bv, <4 x float> %av)
%sr = extractelement <4 x float> %vr, i32 0
store float %sr, float* %a
ret void
}
[AVX512] Add AVX512 run lines to some tests for scalar fma/add/sub/mul/div and regenerate. Follow up commits will bring AVX512 code up to the same quality as AVX/SSE. llvm-svn: 277118
2016-07-29 14:05:58 +08:00
define void @fnmadd_aab_ss(float* %a, float* %b) {
[AVX512] Mark EVEX VMOVSSrm and VMOVSDrm as canFoldAsLoad and isReMaterializable. llvm-svn: 277120
2016-07-29 14:06:04 +08:00
; CHECK-LABEL: fnmadd_aab_ss:
[CodeGen] Unify MBB reference format in both MIR and debug output As part of the unification of the debug format and the MIR format, print MBB references as '%bb.5'. The MIR printer prints the IR name of a MBB only for block definitions. * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)->getNumber\(\)/" << printMBBReference(*\1)/g' * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)\.getNumber\(\)/" << printMBBReference(\1)/g' * find . \( -name "*.txt" -o -name "*.s" -o -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#([0-9]+)/%bb.\1/g' * grep -nr 'BB#' and fix Differential Revision: https://reviews.llvm.org/D40422 llvm-svn: 319665
2017-12-05 01:18:51 +08:00
; CHECK: # %bb.0:
[AVX512] Mark EVEX VMOVSSrm and VMOVSDrm as canFoldAsLoad and isReMaterializable. llvm-svn: 277120
2016-07-29 14:06:04 +08:00
; CHECK-NEXT: vmovss {{.*#+}} xmm0 = mem[0],zero,zero,zero
; CHECK-NEXT: vfnmadd213ss (%rsi), %xmm0, %xmm0
; CHECK-NEXT: vmovss %xmm0, (%rdi)
; CHECK-NEXT: retq
X86-FMA3: Improved/enabled the memory folding optimization for scalar loads generated for _mm_losd_s{s,d}() intrinsics and used in scalar FMAs generated for FMA intrinsics _mm_f{madd,msub,nmadd,nmsub}_s{s,d}(). Reviewer: David Kreitzer Differential Revision: http://reviews.llvm.org/D14762 llvm-svn: 254140
2015-11-26 15:45:30 +08:00
%a.val = load float, float* %a
%av0 = insertelement <4 x float> undef, float %a.val, i32 0
%av1 = insertelement <4 x float> %av0, float 0.000000e+00, i32 1
%av2 = insertelement <4 x float> %av1, float 0.000000e+00, i32 2
%av = insertelement <4 x float> %av2, float 0.000000e+00, i32 3
%b.val = load float, float* %b
%bv0 = insertelement <4 x float> undef, float %b.val, i32 0
%bv1 = insertelement <4 x float> %bv0, float 0.000000e+00, i32 1
%bv2 = insertelement <4 x float> %bv1, float 0.000000e+00, i32 2
%bv = insertelement <4 x float> %bv2, float 0.000000e+00, i32 3
%vr = call <4 x float> @llvm.x86.fma.vfnmadd.ss(<4 x float> %av, <4 x float> %av, <4 x float> %bv)
%sr = extractelement <4 x float> %vr, i32 0
store float %sr, float* %a
ret void
}
[AVX512] Add AVX512 run lines to some tests for scalar fma/add/sub/mul/div and regenerate. Follow up commits will bring AVX512 code up to the same quality as AVX/SSE. llvm-svn: 277118
2016-07-29 14:05:58 +08:00
define void @fnmadd_aba_ss(float* %a, float* %b) {
[AVX512] Mark EVEX VMOVSSrm and VMOVSDrm as canFoldAsLoad and isReMaterializable. llvm-svn: 277120
2016-07-29 14:06:04 +08:00
; CHECK-LABEL: fnmadd_aba_ss:
[CodeGen] Unify MBB reference format in both MIR and debug output As part of the unification of the debug format and the MIR format, print MBB references as '%bb.5'. The MIR printer prints the IR name of a MBB only for block definitions. * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)->getNumber\(\)/" << printMBBReference(*\1)/g' * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)\.getNumber\(\)/" << printMBBReference(\1)/g' * find . \( -name "*.txt" -o -name "*.s" -o -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#([0-9]+)/%bb.\1/g' * grep -nr 'BB#' and fix Differential Revision: https://reviews.llvm.org/D40422 llvm-svn: 319665
2017-12-05 01:18:51 +08:00
; CHECK: # %bb.0:
[AVX512] Mark EVEX VMOVSSrm and VMOVSDrm as canFoldAsLoad and isReMaterializable. llvm-svn: 277120
2016-07-29 14:06:04 +08:00
; CHECK-NEXT: vmovss {{.*#+}} xmm0 = mem[0],zero,zero,zero
; CHECK-NEXT: vfnmadd132ss (%rsi), %xmm0, %xmm0
; CHECK-NEXT: vmovss %xmm0, (%rdi)
; CHECK-NEXT: retq
X86-FMA3: Improved/enabled the memory folding optimization for scalar loads generated for _mm_losd_s{s,d}() intrinsics and used in scalar FMAs generated for FMA intrinsics _mm_f{madd,msub,nmadd,nmsub}_s{s,d}(). Reviewer: David Kreitzer Differential Revision: http://reviews.llvm.org/D14762 llvm-svn: 254140
2015-11-26 15:45:30 +08:00
%a.val = load float, float* %a
%av0 = insertelement <4 x float> undef, float %a.val, i32 0
%av1 = insertelement <4 x float> %av0, float 0.000000e+00, i32 1
%av2 = insertelement <4 x float> %av1, float 0.000000e+00, i32 2
%av = insertelement <4 x float> %av2, float 0.000000e+00, i32 3
%b.val = load float, float* %b
%bv0 = insertelement <4 x float> undef, float %b.val, i32 0
%bv1 = insertelement <4 x float> %bv0, float 0.000000e+00, i32 1
%bv2 = insertelement <4 x float> %bv1, float 0.000000e+00, i32 2
%bv = insertelement <4 x float> %bv2, float 0.000000e+00, i32 3
%vr = call <4 x float> @llvm.x86.fma.vfnmadd.ss(<4 x float> %av, <4 x float> %bv, <4 x float> %av)
%sr = extractelement <4 x float> %vr, i32 0
store float %sr, float* %a
ret void
}
[AVX512] Add AVX512 run lines to some tests for scalar fma/add/sub/mul/div and regenerate. Follow up commits will bring AVX512 code up to the same quality as AVX/SSE. llvm-svn: 277118
2016-07-29 14:05:58 +08:00
define void @fnmsub_aab_ss(float* %a, float* %b) {
[AVX512] Mark EVEX VMOVSSrm and VMOVSDrm as canFoldAsLoad and isReMaterializable. llvm-svn: 277120
2016-07-29 14:06:04 +08:00
; CHECK-LABEL: fnmsub_aab_ss:
[CodeGen] Unify MBB reference format in both MIR and debug output As part of the unification of the debug format and the MIR format, print MBB references as '%bb.5'. The MIR printer prints the IR name of a MBB only for block definitions. * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)->getNumber\(\)/" << printMBBReference(*\1)/g' * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)\.getNumber\(\)/" << printMBBReference(\1)/g' * find . \( -name "*.txt" -o -name "*.s" -o -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#([0-9]+)/%bb.\1/g' * grep -nr 'BB#' and fix Differential Revision: https://reviews.llvm.org/D40422 llvm-svn: 319665
2017-12-05 01:18:51 +08:00
; CHECK: # %bb.0:
[AVX512] Mark EVEX VMOVSSrm and VMOVSDrm as canFoldAsLoad and isReMaterializable. llvm-svn: 277120
2016-07-29 14:06:04 +08:00
; CHECK-NEXT: vmovss {{.*#+}} xmm0 = mem[0],zero,zero,zero
; CHECK-NEXT: vfnmsub213ss (%rsi), %xmm0, %xmm0
; CHECK-NEXT: vmovss %xmm0, (%rdi)
; CHECK-NEXT: retq
X86-FMA3: Improved/enabled the memory folding optimization for scalar loads generated for _mm_losd_s{s,d}() intrinsics and used in scalar FMAs generated for FMA intrinsics _mm_f{madd,msub,nmadd,nmsub}_s{s,d}(). Reviewer: David Kreitzer Differential Revision: http://reviews.llvm.org/D14762 llvm-svn: 254140
2015-11-26 15:45:30 +08:00
%a.val = load float, float* %a
%av0 = insertelement <4 x float> undef, float %a.val, i32 0
%av1 = insertelement <4 x float> %av0, float 0.000000e+00, i32 1
%av2 = insertelement <4 x float> %av1, float 0.000000e+00, i32 2
%av = insertelement <4 x float> %av2, float 0.000000e+00, i32 3
%b.val = load float, float* %b
%bv0 = insertelement <4 x float> undef, float %b.val, i32 0
%bv1 = insertelement <4 x float> %bv0, float 0.000000e+00, i32 1
%bv2 = insertelement <4 x float> %bv1, float 0.000000e+00, i32 2
%bv = insertelement <4 x float> %bv2, float 0.000000e+00, i32 3
%vr = call <4 x float> @llvm.x86.fma.vfnmsub.ss(<4 x float> %av, <4 x float> %av, <4 x float> %bv)
%sr = extractelement <4 x float> %vr, i32 0
store float %sr, float* %a
ret void
}
[AVX512] Add AVX512 run lines to some tests for scalar fma/add/sub/mul/div and regenerate. Follow up commits will bring AVX512 code up to the same quality as AVX/SSE. llvm-svn: 277118
2016-07-29 14:05:58 +08:00
define void @fnmsub_aba_ss(float* %a, float* %b) {
[AVX512] Mark EVEX VMOVSSrm and VMOVSDrm as canFoldAsLoad and isReMaterializable. llvm-svn: 277120
2016-07-29 14:06:04 +08:00
; CHECK-LABEL: fnmsub_aba_ss:
[CodeGen] Unify MBB reference format in both MIR and debug output As part of the unification of the debug format and the MIR format, print MBB references as '%bb.5'. The MIR printer prints the IR name of a MBB only for block definitions. * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)->getNumber\(\)/" << printMBBReference(*\1)/g' * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)\.getNumber\(\)/" << printMBBReference(\1)/g' * find . \( -name "*.txt" -o -name "*.s" -o -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#([0-9]+)/%bb.\1/g' * grep -nr 'BB#' and fix Differential Revision: https://reviews.llvm.org/D40422 llvm-svn: 319665
2017-12-05 01:18:51 +08:00
; CHECK: # %bb.0:
[AVX512] Mark EVEX VMOVSSrm and VMOVSDrm as canFoldAsLoad and isReMaterializable. llvm-svn: 277120
2016-07-29 14:06:04 +08:00
; CHECK-NEXT: vmovss {{.*#+}} xmm0 = mem[0],zero,zero,zero
; CHECK-NEXT: vfnmsub132ss (%rsi), %xmm0, %xmm0
; CHECK-NEXT: vmovss %xmm0, (%rdi)
; CHECK-NEXT: retq
X86-FMA3: Improved/enabled the memory folding optimization for scalar loads generated for _mm_losd_s{s,d}() intrinsics and used in scalar FMAs generated for FMA intrinsics _mm_f{madd,msub,nmadd,nmsub}_s{s,d}(). Reviewer: David Kreitzer Differential Revision: http://reviews.llvm.org/D14762 llvm-svn: 254140
2015-11-26 15:45:30 +08:00
%a.val = load float, float* %a
%av0 = insertelement <4 x float> undef, float %a.val, i32 0
%av1 = insertelement <4 x float> %av0, float 0.000000e+00, i32 1
%av2 = insertelement <4 x float> %av1, float 0.000000e+00, i32 2
%av = insertelement <4 x float> %av2, float 0.000000e+00, i32 3
%b.val = load float, float* %b
%bv0 = insertelement <4 x float> undef, float %b.val, i32 0
%bv1 = insertelement <4 x float> %bv0, float 0.000000e+00, i32 1
%bv2 = insertelement <4 x float> %bv1, float 0.000000e+00, i32 2
%bv = insertelement <4 x float> %bv2, float 0.000000e+00, i32 3
%vr = call <4 x float> @llvm.x86.fma.vfnmsub.ss(<4 x float> %av, <4 x float> %bv, <4 x float> %av)
%sr = extractelement <4 x float> %vr, i32 0
store float %sr, float* %a
ret void
}
[AVX512] Add AVX512 run lines to some tests for scalar fma/add/sub/mul/div and regenerate. Follow up commits will bring AVX512 code up to the same quality as AVX/SSE. llvm-svn: 277118
2016-07-29 14:05:58 +08:00
define void @fmadd_aab_sd(double* %a, double* %b) {
[AVX512] Mark EVEX VMOVSSrm and VMOVSDrm as canFoldAsLoad and isReMaterializable. llvm-svn: 277120
2016-07-29 14:06:04 +08:00
; CHECK-LABEL: fmadd_aab_sd:
[CodeGen] Unify MBB reference format in both MIR and debug output As part of the unification of the debug format and the MIR format, print MBB references as '%bb.5'. The MIR printer prints the IR name of a MBB only for block definitions. * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)->getNumber\(\)/" << printMBBReference(*\1)/g' * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)\.getNumber\(\)/" << printMBBReference(\1)/g' * find . \( -name "*.txt" -o -name "*.s" -o -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#([0-9]+)/%bb.\1/g' * grep -nr 'BB#' and fix Differential Revision: https://reviews.llvm.org/D40422 llvm-svn: 319665
2017-12-05 01:18:51 +08:00
; CHECK: # %bb.0:
[AVX512] Mark EVEX VMOVSSrm and VMOVSDrm as canFoldAsLoad and isReMaterializable. llvm-svn: 277120
2016-07-29 14:06:04 +08:00
; CHECK-NEXT: vmovsd {{.*#+}} xmm0 = mem[0],zero
; CHECK-NEXT: vfmadd213sd (%rsi), %xmm0, %xmm0
; CHECK-NEXT: vmovlpd %xmm0, (%rdi)
; CHECK-NEXT: retq
X86-FMA3: Improved/enabled the memory folding optimization for scalar loads generated for _mm_losd_s{s,d}() intrinsics and used in scalar FMAs generated for FMA intrinsics _mm_f{madd,msub,nmadd,nmsub}_s{s,d}(). Reviewer: David Kreitzer Differential Revision: http://reviews.llvm.org/D14762 llvm-svn: 254140
2015-11-26 15:45:30 +08:00
%a.val = load double, double* %a
%av0 = insertelement <2 x double> undef, double %a.val, i32 0
%av = insertelement <2 x double> %av0, double 0.000000e+00, i32 1
%b.val = load double, double* %b
%bv0 = insertelement <2 x double> undef, double %b.val, i32 0
%bv = insertelement <2 x double> %bv0, double 0.000000e+00, i32 1
%vr = call <2 x double> @llvm.x86.fma.vfmadd.sd(<2 x double> %av, <2 x double> %av, <2 x double> %bv)
%sr = extractelement <2 x double> %vr, i32 0
store double %sr, double* %a
ret void
}
[AVX512] Add AVX512 run lines to some tests for scalar fma/add/sub/mul/div and regenerate. Follow up commits will bring AVX512 code up to the same quality as AVX/SSE. llvm-svn: 277118
2016-07-29 14:05:58 +08:00
define void @fmadd_aba_sd(double* %a, double* %b) {
[AVX512] Mark EVEX VMOVSSrm and VMOVSDrm as canFoldAsLoad and isReMaterializable. llvm-svn: 277120
2016-07-29 14:06:04 +08:00
; CHECK-LABEL: fmadd_aba_sd:
[CodeGen] Unify MBB reference format in both MIR and debug output As part of the unification of the debug format and the MIR format, print MBB references as '%bb.5'. The MIR printer prints the IR name of a MBB only for block definitions. * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)->getNumber\(\)/" << printMBBReference(*\1)/g' * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)\.getNumber\(\)/" << printMBBReference(\1)/g' * find . \( -name "*.txt" -o -name "*.s" -o -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#([0-9]+)/%bb.\1/g' * grep -nr 'BB#' and fix Differential Revision: https://reviews.llvm.org/D40422 llvm-svn: 319665
2017-12-05 01:18:51 +08:00
; CHECK: # %bb.0:
[AVX512] Mark EVEX VMOVSSrm and VMOVSDrm as canFoldAsLoad and isReMaterializable. llvm-svn: 277120
2016-07-29 14:06:04 +08:00
; CHECK-NEXT: vmovsd {{.*#+}} xmm0 = mem[0],zero
; CHECK-NEXT: vfmadd132sd (%rsi), %xmm0, %xmm0
; CHECK-NEXT: vmovlpd %xmm0, (%rdi)
; CHECK-NEXT: retq
X86-FMA3: Improved/enabled the memory folding optimization for scalar loads generated for _mm_losd_s{s,d}() intrinsics and used in scalar FMAs generated for FMA intrinsics _mm_f{madd,msub,nmadd,nmsub}_s{s,d}(). Reviewer: David Kreitzer Differential Revision: http://reviews.llvm.org/D14762 llvm-svn: 254140
2015-11-26 15:45:30 +08:00
%a.val = load double, double* %a
%av0 = insertelement <2 x double> undef, double %a.val, i32 0
%av = insertelement <2 x double> %av0, double 0.000000e+00, i32 1
%b.val = load double, double* %b
%bv0 = insertelement <2 x double> undef, double %b.val, i32 0
%bv = insertelement <2 x double> %bv0, double 0.000000e+00, i32 1
%vr = call <2 x double> @llvm.x86.fma.vfmadd.sd(<2 x double> %av, <2 x double> %bv, <2 x double> %av)
%sr = extractelement <2 x double> %vr, i32 0
store double %sr, double* %a
ret void
}
[AVX512] Add AVX512 run lines to some tests for scalar fma/add/sub/mul/div and regenerate. Follow up commits will bring AVX512 code up to the same quality as AVX/SSE. llvm-svn: 277118
2016-07-29 14:05:58 +08:00
define void @fmsub_aab_sd(double* %a, double* %b) {
[AVX512] Mark EVEX VMOVSSrm and VMOVSDrm as canFoldAsLoad and isReMaterializable. llvm-svn: 277120
2016-07-29 14:06:04 +08:00
; CHECK-LABEL: fmsub_aab_sd:
[CodeGen] Unify MBB reference format in both MIR and debug output As part of the unification of the debug format and the MIR format, print MBB references as '%bb.5'. The MIR printer prints the IR name of a MBB only for block definitions. * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)->getNumber\(\)/" << printMBBReference(*\1)/g' * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)\.getNumber\(\)/" << printMBBReference(\1)/g' * find . \( -name "*.txt" -o -name "*.s" -o -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#([0-9]+)/%bb.\1/g' * grep -nr 'BB#' and fix Differential Revision: https://reviews.llvm.org/D40422 llvm-svn: 319665
2017-12-05 01:18:51 +08:00
; CHECK: # %bb.0:
[AVX512] Mark EVEX VMOVSSrm and VMOVSDrm as canFoldAsLoad and isReMaterializable. llvm-svn: 277120
2016-07-29 14:06:04 +08:00
; CHECK-NEXT: vmovsd {{.*#+}} xmm0 = mem[0],zero
; CHECK-NEXT: vfmsub213sd (%rsi), %xmm0, %xmm0
; CHECK-NEXT: vmovlpd %xmm0, (%rdi)
; CHECK-NEXT: retq
X86-FMA3: Improved/enabled the memory folding optimization for scalar loads generated for _mm_losd_s{s,d}() intrinsics and used in scalar FMAs generated for FMA intrinsics _mm_f{madd,msub,nmadd,nmsub}_s{s,d}(). Reviewer: David Kreitzer Differential Revision: http://reviews.llvm.org/D14762 llvm-svn: 254140
2015-11-26 15:45:30 +08:00
%a.val = load double, double* %a
%av0 = insertelement <2 x double> undef, double %a.val, i32 0
%av = insertelement <2 x double> %av0, double 0.000000e+00, i32 1
%b.val = load double, double* %b
%bv0 = insertelement <2 x double> undef, double %b.val, i32 0
%bv = insertelement <2 x double> %bv0, double 0.000000e+00, i32 1
%vr = call <2 x double> @llvm.x86.fma.vfmsub.sd(<2 x double> %av, <2 x double> %av, <2 x double> %bv)
%sr = extractelement <2 x double> %vr, i32 0
store double %sr, double* %a
ret void
}
[AVX512] Add AVX512 run lines to some tests for scalar fma/add/sub/mul/div and regenerate. Follow up commits will bring AVX512 code up to the same quality as AVX/SSE. llvm-svn: 277118
2016-07-29 14:05:58 +08:00
define void @fmsub_aba_sd(double* %a, double* %b) {
[AVX512] Mark EVEX VMOVSSrm and VMOVSDrm as canFoldAsLoad and isReMaterializable. llvm-svn: 277120
2016-07-29 14:06:04 +08:00
; CHECK-LABEL: fmsub_aba_sd:
[CodeGen] Unify MBB reference format in both MIR and debug output As part of the unification of the debug format and the MIR format, print MBB references as '%bb.5'. The MIR printer prints the IR name of a MBB only for block definitions. * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)->getNumber\(\)/" << printMBBReference(*\1)/g' * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)\.getNumber\(\)/" << printMBBReference(\1)/g' * find . \( -name "*.txt" -o -name "*.s" -o -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#([0-9]+)/%bb.\1/g' * grep -nr 'BB#' and fix Differential Revision: https://reviews.llvm.org/D40422 llvm-svn: 319665
2017-12-05 01:18:51 +08:00
; CHECK: # %bb.0:
[AVX512] Mark EVEX VMOVSSrm and VMOVSDrm as canFoldAsLoad and isReMaterializable. llvm-svn: 277120
2016-07-29 14:06:04 +08:00
; CHECK-NEXT: vmovsd {{.*#+}} xmm0 = mem[0],zero
; CHECK-NEXT: vfmsub132sd (%rsi), %xmm0, %xmm0
; CHECK-NEXT: vmovlpd %xmm0, (%rdi)
; CHECK-NEXT: retq
X86-FMA3: Improved/enabled the memory folding optimization for scalar loads generated for _mm_losd_s{s,d}() intrinsics and used in scalar FMAs generated for FMA intrinsics _mm_f{madd,msub,nmadd,nmsub}_s{s,d}(). Reviewer: David Kreitzer Differential Revision: http://reviews.llvm.org/D14762 llvm-svn: 254140
2015-11-26 15:45:30 +08:00
%a.val = load double, double* %a
%av0 = insertelement <2 x double> undef, double %a.val, i32 0
%av = insertelement <2 x double> %av0, double 0.000000e+00, i32 1
%b.val = load double, double* %b
%bv0 = insertelement <2 x double> undef, double %b.val, i32 0
%bv = insertelement <2 x double> %bv0, double 0.000000e+00, i32 1
%vr = call <2 x double> @llvm.x86.fma.vfmsub.sd(<2 x double> %av, <2 x double> %bv, <2 x double> %av)
%sr = extractelement <2 x double> %vr, i32 0
store double %sr, double* %a
ret void
}
[AVX512] Add AVX512 run lines to some tests for scalar fma/add/sub/mul/div and regenerate. Follow up commits will bring AVX512 code up to the same quality as AVX/SSE. llvm-svn: 277118
2016-07-29 14:05:58 +08:00
define void @fnmadd_aab_sd(double* %a, double* %b) {
[AVX512] Mark EVEX VMOVSSrm and VMOVSDrm as canFoldAsLoad and isReMaterializable. llvm-svn: 277120
2016-07-29 14:06:04 +08:00
; CHECK-LABEL: fnmadd_aab_sd:
[CodeGen] Unify MBB reference format in both MIR and debug output As part of the unification of the debug format and the MIR format, print MBB references as '%bb.5'. The MIR printer prints the IR name of a MBB only for block definitions. * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)->getNumber\(\)/" << printMBBReference(*\1)/g' * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)\.getNumber\(\)/" << printMBBReference(\1)/g' * find . \( -name "*.txt" -o -name "*.s" -o -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#([0-9]+)/%bb.\1/g' * grep -nr 'BB#' and fix Differential Revision: https://reviews.llvm.org/D40422 llvm-svn: 319665
2017-12-05 01:18:51 +08:00
; CHECK: # %bb.0:
[AVX512] Mark EVEX VMOVSSrm and VMOVSDrm as canFoldAsLoad and isReMaterializable. llvm-svn: 277120
2016-07-29 14:06:04 +08:00
; CHECK-NEXT: vmovsd {{.*#+}} xmm0 = mem[0],zero
; CHECK-NEXT: vfnmadd213sd (%rsi), %xmm0, %xmm0
; CHECK-NEXT: vmovlpd %xmm0, (%rdi)
; CHECK-NEXT: retq
X86-FMA3: Improved/enabled the memory folding optimization for scalar loads generated for _mm_losd_s{s,d}() intrinsics and used in scalar FMAs generated for FMA intrinsics _mm_f{madd,msub,nmadd,nmsub}_s{s,d}(). Reviewer: David Kreitzer Differential Revision: http://reviews.llvm.org/D14762 llvm-svn: 254140
2015-11-26 15:45:30 +08:00
%a.val = load double, double* %a
%av0 = insertelement <2 x double> undef, double %a.val, i32 0
%av = insertelement <2 x double> %av0, double 0.000000e+00, i32 1
%b.val = load double, double* %b
%bv0 = insertelement <2 x double> undef, double %b.val, i32 0
%bv = insertelement <2 x double> %bv0, double 0.000000e+00, i32 1
%vr = call <2 x double> @llvm.x86.fma.vfnmadd.sd(<2 x double> %av, <2 x double> %av, <2 x double> %bv)
%sr = extractelement <2 x double> %vr, i32 0
store double %sr, double* %a
ret void
}
[AVX512] Add AVX512 run lines to some tests for scalar fma/add/sub/mul/div and regenerate. Follow up commits will bring AVX512 code up to the same quality as AVX/SSE. llvm-svn: 277118
2016-07-29 14:05:58 +08:00
define void @fnmadd_aba_sd(double* %a, double* %b) {
[AVX512] Mark EVEX VMOVSSrm and VMOVSDrm as canFoldAsLoad and isReMaterializable. llvm-svn: 277120
2016-07-29 14:06:04 +08:00
; CHECK-LABEL: fnmadd_aba_sd:
[CodeGen] Unify MBB reference format in both MIR and debug output As part of the unification of the debug format and the MIR format, print MBB references as '%bb.5'. The MIR printer prints the IR name of a MBB only for block definitions. * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)->getNumber\(\)/" << printMBBReference(*\1)/g' * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)\.getNumber\(\)/" << printMBBReference(\1)/g' * find . \( -name "*.txt" -o -name "*.s" -o -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#([0-9]+)/%bb.\1/g' * grep -nr 'BB#' and fix Differential Revision: https://reviews.llvm.org/D40422 llvm-svn: 319665
2017-12-05 01:18:51 +08:00
; CHECK: # %bb.0:
[AVX512] Mark EVEX VMOVSSrm and VMOVSDrm as canFoldAsLoad and isReMaterializable. llvm-svn: 277120
2016-07-29 14:06:04 +08:00
; CHECK-NEXT: vmovsd {{.*#+}} xmm0 = mem[0],zero
; CHECK-NEXT: vfnmadd132sd (%rsi), %xmm0, %xmm0
; CHECK-NEXT: vmovlpd %xmm0, (%rdi)
; CHECK-NEXT: retq
X86-FMA3: Improved/enabled the memory folding optimization for scalar loads generated for _mm_losd_s{s,d}() intrinsics and used in scalar FMAs generated for FMA intrinsics _mm_f{madd,msub,nmadd,nmsub}_s{s,d}(). Reviewer: David Kreitzer Differential Revision: http://reviews.llvm.org/D14762 llvm-svn: 254140
2015-11-26 15:45:30 +08:00
%a.val = load double, double* %a
%av0 = insertelement <2 x double> undef, double %a.val, i32 0
%av = insertelement <2 x double> %av0, double 0.000000e+00, i32 1
%b.val = load double, double* %b
%bv0 = insertelement <2 x double> undef, double %b.val, i32 0
%bv = insertelement <2 x double> %bv0, double 0.000000e+00, i32 1
%vr = call <2 x double> @llvm.x86.fma.vfnmadd.sd(<2 x double> %av, <2 x double> %bv, <2 x double> %av)
%sr = extractelement <2 x double> %vr, i32 0
store double %sr, double* %a
ret void
}
[AVX512] Add AVX512 run lines to some tests for scalar fma/add/sub/mul/div and regenerate. Follow up commits will bring AVX512 code up to the same quality as AVX/SSE. llvm-svn: 277118
2016-07-29 14:05:58 +08:00
define void @fnmsub_aab_sd(double* %a, double* %b) {
[AVX512] Mark EVEX VMOVSSrm and VMOVSDrm as canFoldAsLoad and isReMaterializable. llvm-svn: 277120
2016-07-29 14:06:04 +08:00
; CHECK-LABEL: fnmsub_aab_sd:
[CodeGen] Unify MBB reference format in both MIR and debug output As part of the unification of the debug format and the MIR format, print MBB references as '%bb.5'. The MIR printer prints the IR name of a MBB only for block definitions. * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)->getNumber\(\)/" << printMBBReference(*\1)/g' * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)\.getNumber\(\)/" << printMBBReference(\1)/g' * find . \( -name "*.txt" -o -name "*.s" -o -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#([0-9]+)/%bb.\1/g' * grep -nr 'BB#' and fix Differential Revision: https://reviews.llvm.org/D40422 llvm-svn: 319665
2017-12-05 01:18:51 +08:00
; CHECK: # %bb.0:
[AVX512] Mark EVEX VMOVSSrm and VMOVSDrm as canFoldAsLoad and isReMaterializable. llvm-svn: 277120
2016-07-29 14:06:04 +08:00
; CHECK-NEXT: vmovsd {{.*#+}} xmm0 = mem[0],zero
; CHECK-NEXT: vfnmsub213sd (%rsi), %xmm0, %xmm0
; CHECK-NEXT: vmovlpd %xmm0, (%rdi)
; CHECK-NEXT: retq
X86-FMA3: Improved/enabled the memory folding optimization for scalar loads generated for _mm_losd_s{s,d}() intrinsics and used in scalar FMAs generated for FMA intrinsics _mm_f{madd,msub,nmadd,nmsub}_s{s,d}(). Reviewer: David Kreitzer Differential Revision: http://reviews.llvm.org/D14762 llvm-svn: 254140
2015-11-26 15:45:30 +08:00
%a.val = load double, double* %a
%av0 = insertelement <2 x double> undef, double %a.val, i32 0
%av = insertelement <2 x double> %av0, double 0.000000e+00, i32 1
%b.val = load double, double* %b
%bv0 = insertelement <2 x double> undef, double %b.val, i32 0
%bv = insertelement <2 x double> %bv0, double 0.000000e+00, i32 1
%vr = call <2 x double> @llvm.x86.fma.vfnmsub.sd(<2 x double> %av, <2 x double> %av, <2 x double> %bv)
%sr = extractelement <2 x double> %vr, i32 0
store double %sr, double* %a
ret void
}
[AVX512] Add AVX512 run lines to some tests for scalar fma/add/sub/mul/div and regenerate. Follow up commits will bring AVX512 code up to the same quality as AVX/SSE. llvm-svn: 277118
2016-07-29 14:05:58 +08:00
define void @fnmsub_aba_sd(double* %a, double* %b) {
[AVX512] Mark EVEX VMOVSSrm and VMOVSDrm as canFoldAsLoad and isReMaterializable. llvm-svn: 277120
2016-07-29 14:06:04 +08:00
; CHECK-LABEL: fnmsub_aba_sd:
[CodeGen] Unify MBB reference format in both MIR and debug output As part of the unification of the debug format and the MIR format, print MBB references as '%bb.5'. The MIR printer prints the IR name of a MBB only for block definitions. * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)->getNumber\(\)/" << printMBBReference(*\1)/g' * find . \( -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#" << ([a-zA-Z0-9_]+)\.getNumber\(\)/" << printMBBReference(\1)/g' * find . \( -name "*.txt" -o -name "*.s" -o -name "*.mir" -o -name "*.cpp" -o -name "*.h" -o -name "*.ll" \) -type f -print0 | xargs -0 sed -i '' -E 's/BB#([0-9]+)/%bb.\1/g' * grep -nr 'BB#' and fix Differential Revision: https://reviews.llvm.org/D40422 llvm-svn: 319665
2017-12-05 01:18:51 +08:00
; CHECK: # %bb.0:
[AVX512] Mark EVEX VMOVSSrm and VMOVSDrm as canFoldAsLoad and isReMaterializable. llvm-svn: 277120
2016-07-29 14:06:04 +08:00
; CHECK-NEXT: vmovsd {{.*#+}} xmm0 = mem[0],zero
; CHECK-NEXT: vfnmsub132sd (%rsi), %xmm0, %xmm0
; CHECK-NEXT: vmovlpd %xmm0, (%rdi)
; CHECK-NEXT: retq
X86-FMA3: Improved/enabled the memory folding optimization for scalar loads generated for _mm_losd_s{s,d}() intrinsics and used in scalar FMAs generated for FMA intrinsics _mm_f{madd,msub,nmadd,nmsub}_s{s,d}(). Reviewer: David Kreitzer Differential Revision: http://reviews.llvm.org/D14762 llvm-svn: 254140
2015-11-26 15:45:30 +08:00
%a.val = load double, double* %a
%av0 = insertelement <2 x double> undef, double %a.val, i32 0
%av = insertelement <2 x double> %av0, double 0.000000e+00, i32 1
%b.val = load double, double* %b
%bv0 = insertelement <2 x double> undef, double %b.val, i32 0
%bv = insertelement <2 x double> %bv0, double 0.000000e+00, i32 1
%vr = call <2 x double> @llvm.x86.fma.vfnmsub.sd(<2 x double> %av, <2 x double> %bv, <2 x double> %av)
%sr = extractelement <2 x double> %vr, i32 0
store double %sr, double* %a
ret void
}