[X86][FMA] More thorough FMA tests

Added FMADD/FMSUB/FNMADD/FNMSUB tests for all types

Added load folding tests for 512-bit vectors

NOTE: Many of the AVX512 FMA instructions don't yet commute/fold correctly

As discussed on D14909

llvm-svn: 254232

1 files changed, 152 insertions, 37 deletions

diff --git a/llvm/test/CodeGen/X86/fma_patterns_wide.ll b/llvm/test/CodeGen/X86/fma_patterns_wide.ll
index 5541e05196e..4a6a22aebd2 100644
--- a/llvm/test/CodeGen/X86/fma_patterns_wide.ll
+++ b/llvm/test/CodeGen/X86/fma_patterns_wide.ll
@@ -4,20 +4,24 @@
 ; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx,+fma4 -fp-contract=fast | FileCheck %s --check-prefix=FMA4
 ; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx512f -fp-contract=fast | FileCheck %s --check-prefix=AVX512
 
-define <16 x float> @test_x86_fmadd_ps_y_wide(<16 x float> %a0, <16 x float> %a1, <16 x float> %a2) {
-; FMA-LABEL: test_x86_fmadd_ps_y_wide:
+;
+; Pattern: (fadd (fmul x, y), z) -> (fmadd x,y,z)
+;
+
+define <16 x float> @test_16f32_fmadd(<16 x float> %a0, <16 x float> %a1, <16 x float> %a2) {
+; FMA-LABEL: test_16f32_fmadd:
 ; FMA:       # BB#0:
 ; FMA-NEXT:    vfmadd213ps %ymm4, %ymm2, %ymm0
 ; FMA-NEXT:    vfmadd213ps %ymm5, %ymm3, %ymm1
 ; FMA-NEXT:    retq
 ;
-; FMA4-LABEL: test_x86_fmadd_ps_y_wide:
+; FMA4-LABEL: test_16f32_fmadd:
 ; FMA4:       # BB#0:
 ; FMA4-NEXT:    vfmaddps %ymm4, %ymm2, %ymm0, %ymm0
 ; FMA4-NEXT:    vfmaddps %ymm5, %ymm3, %ymm1, %ymm1
 ; FMA4-NEXT:    retq
 ;
-; AVX512-LABEL: test_x86_fmadd_ps_y_wide:
+; AVX512-LABEL: test_16f32_fmadd:
 ; AVX512:       # BB#0:
 ; AVX512-NEXT:    vfmadd213ps %zmm2, %zmm1, %zmm0
 ; AVX512-NEXT:    retq
@@ -26,20 +30,46 @@ define <16 x float> @test_x86_fmadd_ps_y_wide(<16 x float> %a0, <16 x float> %a1
   ret <16 x float> %res
 }
 
-define <16 x float> @test_x86_fmsub_ps_y_wide(<16 x float> %a0, <16 x float> %a1, <16 x float> %a2) {
-; FMA-LABEL: test_x86_fmsub_ps_y_wide:
+define <8 x double> @test_8f64_fmadd(<8 x double> %a0, <8 x double> %a1, <8 x double> %a2) {
+; FMA-LABEL: test_8f64_fmadd:
+; FMA:       # BB#0:
+; FMA-NEXT:    vfmadd213pd %ymm4, %ymm2, %ymm0
+; FMA-NEXT:    vfmadd213pd %ymm5, %ymm3, %ymm1
+; FMA-NEXT:    retq
+;
+; FMA4-LABEL: test_8f64_fmadd:
+; FMA4:       # BB#0:
+; FMA4-NEXT:    vfmaddpd %ymm4, %ymm2, %ymm0, %ymm0
+; FMA4-NEXT:    vfmaddpd %ymm5, %ymm3, %ymm1, %ymm1
+; FMA4-NEXT:    retq
+;
+; AVX512-LABEL: test_8f64_fmadd:
+; AVX512:       # BB#0:
+; AVX512-NEXT:    vfmadd213pd %zmm2, %zmm1, %zmm0
+; AVX512-NEXT:    retq
+  %x = fmul <8 x double> %a0, %a1
+  %res = fadd <8 x double> %x, %a2
+  ret <8 x double> %res
+}
+
+;
+; Pattern: (fsub (fmul x, y), z) -> (fmsub x, y, z)
+;
+
+define <16 x float> @test_16f32_fmsub(<16 x float> %a0, <16 x float> %a1, <16 x float> %a2) {
+; FMA-LABEL: test_16f32_fmsub:
 ; FMA:       # BB#0:
 ; FMA-NEXT:    vfmsub213ps %ymm4, %ymm2, %ymm0
 ; FMA-NEXT:    vfmsub213ps %ymm5, %ymm3, %ymm1
 ; FMA-NEXT:    retq
 ;
-; FMA4-LABEL: test_x86_fmsub_ps_y_wide:
+; FMA4-LABEL: test_16f32_fmsub:
 ; FMA4:       # BB#0:
 ; FMA4-NEXT:    vfmsubps %ymm4, %ymm2, %ymm0, %ymm0
 ; FMA4-NEXT:    vfmsubps %ymm5, %ymm3, %ymm1, %ymm1
 ; FMA4-NEXT:    retq
 ;
-; AVX512-LABEL: test_x86_fmsub_ps_y_wide:
+; AVX512-LABEL: test_16f32_fmsub:
 ; AVX512:       # BB#0:
 ; AVX512-NEXT:    vfmsub213ps %zmm2, %zmm1, %zmm0
 ; AVX512-NEXT:    retq
@@ -48,20 +78,46 @@ define <16 x float> @test_x86_fmsub_ps_y_wide(<16 x float> %a0, <16 x float> %a1
   ret <16 x float> %res
 }
 
-define <16 x float> @test_x86_fnmadd_ps_y_wide(<16 x float> %a0, <16 x float> %a1, <16 x float> %a2) {
-; FMA-LABEL: test_x86_fnmadd_ps_y_wide:
+define <8 x double> @test_8f64_fmsub(<8 x double> %a0, <8 x double> %a1, <8 x double> %a2) {
+; FMA-LABEL: test_8f64_fmsub:
+; FMA:       # BB#0:
+; FMA-NEXT:    vfmsub213pd %ymm4, %ymm2, %ymm0
+; FMA-NEXT:    vfmsub213pd %ymm5, %ymm3, %ymm1
+; FMA-NEXT:    retq
+;
+; FMA4-LABEL: test_8f64_fmsub:
+; FMA4:       # BB#0:
+; FMA4-NEXT:    vfmsubpd %ymm4, %ymm2, %ymm0, %ymm0
+; FMA4-NEXT:    vfmsubpd %ymm5, %ymm3, %ymm1, %ymm1
+; FMA4-NEXT:    retq
+;
+; AVX512-LABEL: test_8f64_fmsub:
+; AVX512:       # BB#0:
+; AVX512-NEXT:    vfmsub213pd %zmm2, %zmm1, %zmm0
+; AVX512-NEXT:    retq
+  %x = fmul <8 x double> %a0, %a1
+  %res = fsub <8 x double> %x, %a2
+  ret <8 x double> %res
+}
+
+;
+; Pattern: (fsub z, (fmul x, y)) -> (fnmadd x, y, z)
+;
+
+define <16 x float> @test_16f32_fnmadd(<16 x float> %a0, <16 x float> %a1, <16 x float> %a2) {
+; FMA-LABEL: test_16f32_fnmadd:
 ; FMA:       # BB#0:
 ; FMA-NEXT:    vfnmadd213ps %ymm4, %ymm2, %ymm0
 ; FMA-NEXT:    vfnmadd213ps %ymm5, %ymm3, %ymm1
 ; FMA-NEXT:    retq
 ;
-; FMA4-LABEL: test_x86_fnmadd_ps_y_wide:
+; FMA4-LABEL: test_16f32_fnmadd:
 ; FMA4:       # BB#0:
 ; FMA4-NEXT:    vfnmaddps %ymm4, %ymm2, %ymm0, %ymm0
 ; FMA4-NEXT:    vfnmaddps %ymm5, %ymm3, %ymm1, %ymm1
 ; FMA4-NEXT:    retq
 ;
-; AVX512-LABEL: test_x86_fnmadd_ps_y_wide:
+; AVX512-LABEL: test_16f32_fnmadd:
 ; AVX512:       # BB#0:
 ; AVX512-NEXT:    vfnmadd213ps %zmm2, %zmm1, %zmm0
 ; AVX512-NEXT:    retq
@@ -70,20 +126,46 @@ define <16 x float> @test_x86_fnmadd_ps_y_wide(<16 x float> %a0, <16 x float> %a
   ret <16 x float> %res
 }
 
-define <16 x float> @test_x86_fnmsub_ps_y_wide(<16 x float> %a0, <16 x float> %a1, <16 x float> %a2) {
-; FMA-LABEL: test_x86_fnmsub_ps_y_wide:
+define <8 x double> @test_8f64_fnmadd(<8 x double> %a0, <8 x double> %a1, <8 x double> %a2) {
+; FMA-LABEL: test_8f64_fnmadd:
+; FMA:       # BB#0:
+; FMA-NEXT:    vfnmadd213pd %ymm4, %ymm2, %ymm0
+; FMA-NEXT:    vfnmadd213pd %ymm5, %ymm3, %ymm1
+; FMA-NEXT:    retq
+;
+; FMA4-LABEL: test_8f64_fnmadd:
+; FMA4:       # BB#0:
+; FMA4-NEXT:    vfnmaddpd %ymm4, %ymm2, %ymm0, %ymm0
+; FMA4-NEXT:    vfnmaddpd %ymm5, %ymm3, %ymm1, %ymm1
+; FMA4-NEXT:    retq
+;
+; AVX512-LABEL: test_8f64_fnmadd:
+; AVX512:       # BB#0:
+; AVX512-NEXT:    vfnmadd213pd %zmm2, %zmm1, %zmm0
+; AVX512-NEXT:    retq
+  %x = fmul <8 x double> %a0, %a1
+  %res = fsub <8 x double> %a2, %x
+  ret <8 x double> %res
+}
+
+;
+; Pattern: (fsub (fneg (fmul x, y)), z) -> (fnmsub x, y, z)
+;
+
+define <16 x float> @test_16f32_fnmsub(<16 x float> %a0, <16 x float> %a1, <16 x float> %a2) {
+; FMA-LABEL: test_16f32_fnmsub:
 ; FMA:       # BB#0:
 ; FMA-NEXT:    vfnmsub213ps %ymm4, %ymm2, %ymm0
 ; FMA-NEXT:    vfnmsub213ps %ymm5, %ymm3, %ymm1
 ; FMA-NEXT:    retq
 ;
-; FMA4-LABEL: test_x86_fnmsub_ps_y_wide:
+; FMA4-LABEL: test_16f32_fnmsub:
 ; FMA4:       # BB#0:
 ; FMA4-NEXT:    vfnmsubps %ymm4, %ymm2, %ymm0, %ymm0
 ; FMA4-NEXT:    vfnmsubps %ymm5, %ymm3, %ymm1, %ymm1
 ; FMA4-NEXT:    retq
 ;
-; AVX512-LABEL: test_x86_fnmsub_ps_y_wide:
+; AVX512-LABEL: test_16f32_fnmsub:
 ; AVX512:       # BB#0:
 ; AVX512-NEXT:    vfnmsub213ps %zmm2, %zmm1, %zmm0
 ; AVX512-NEXT:    retq
@@ -93,46 +175,79 @@ define <16 x float> @test_x86_fnmsub_ps_y_wide(<16 x float> %a0, <16 x float> %a
   ret <16 x float> %res
 }
 
-define <8 x double> @test_x86_fmadd_pd_y_wide(<8 x double> %a0, <8 x double> %a1, <8 x double> %a2) {
-; FMA-LABEL: test_x86_fmadd_pd_y_wide:
+define <8 x double> @test_8f64_fnmsub(<8 x double> %a0, <8 x double> %a1, <8 x double> %a2) {
+; FMA-LABEL: test_8f64_fnmsub:
 ; FMA:       # BB#0:
-; FMA-NEXT:    vfmadd213pd %ymm4, %ymm2, %ymm0
-; FMA-NEXT:    vfmadd213pd %ymm5, %ymm3, %ymm1
+; FMA-NEXT:    vfnmsub213pd %ymm4, %ymm2, %ymm0
+; FMA-NEXT:    vfnmsub213pd %ymm5, %ymm3, %ymm1
 ; FMA-NEXT:    retq
 ;
-; FMA4-LABEL: test_x86_fmadd_pd_y_wide:
+; FMA4-LABEL: test_8f64_fnmsub:
 ; FMA4:       # BB#0:
-; FMA4-NEXT:    vfmaddpd %ymm4, %ymm2, %ymm0, %ymm0
-; FMA4-NEXT:    vfmaddpd %ymm5, %ymm3, %ymm1, %ymm1
+; FMA4-NEXT:    vfnmsubpd %ymm4, %ymm2, %ymm0, %ymm0
+; FMA4-NEXT:    vfnmsubpd %ymm5, %ymm3, %ymm1, %ymm1
 ; FMA4-NEXT:    retq
 ;
-; AVX512-LABEL: test_x86_fmadd_pd_y_wide:
+; AVX512-LABEL: test_8f64_fnmsub:
 ; AVX512:       # BB#0:
-; AVX512-NEXT:    vfmadd213pd %zmm2, %zmm1, %zmm0
+; AVX512-NEXT:    vfnmsub213pd %zmm2, %zmm1, %zmm0
 ; AVX512-NEXT:    retq
   %x = fmul <8 x double> %a0, %a1
-  %res = fadd <8 x double> %x, %a2
+  %y = fsub <8 x double> <double -0.000000e+00, double -0.000000e+00, double -0.000000e+00, double -0.000000e+00, double -0.000000e+00, double -0.000000e+00, double -0.000000e+00, double -0.000000e+00>, %x
+  %res = fsub <8 x double> %y, %a2
   ret <8 x double> %res
 }
 
-define <8 x double> @test_x86_fmsub_pd_y_wide(<8 x double> %a0, <8 x double> %a1, <8 x double> %a2) {
-; FMA-LABEL: test_x86_fmsub_pd_y_wide:
+;
+; Load Folding Patterns
+;
+
+define <16 x float> @test_16f32_fmadd_load(<16 x float>* %a0, <16 x float> %a1, <16 x float> %a2) {
+; FMA-LABEL: test_16f32_fmadd_load:
 ; FMA:       # BB#0:
-; FMA-NEXT:    vfmsub213pd %ymm4, %ymm2, %ymm0
-; FMA-NEXT:    vfmsub213pd %ymm5, %ymm3, %ymm1
+; FMA-NEXT:    vfmadd132ps (%rdi), %ymm2, %ymm0
+; FMA-NEXT:    vfmadd132ps 32(%rdi), %ymm3, %ymm1
 ; FMA-NEXT:    retq
 ;
-; FMA4-LABEL: test_x86_fmsub_pd_y_wide:
+; FMA4-LABEL: test_16f32_fmadd_load:
 ; FMA4:       # BB#0:
-; FMA4-NEXT:    vfmsubpd %ymm4, %ymm2, %ymm0, %ymm0
-; FMA4-NEXT:    vfmsubpd %ymm5, %ymm3, %ymm1, %ymm1
+; FMA4-NEXT:    vfmaddps %ymm2, (%rdi), %ymm0, %ymm0
+; FMA4-NEXT:    vfmaddps %ymm3, 32(%rdi), %ymm1, %ymm1
 ; FMA4-NEXT:    retq
 ;
-; AVX512-LABEL: test_x86_fmsub_pd_y_wide:
+; AVX512-LABEL: test_16f32_fmadd_load:
 ; AVX512:       # BB#0:
-; AVX512-NEXT:    vfmsub213pd %zmm2, %zmm1, %zmm0
+; AVX512-NEXT:    vmovaps (%rdi), %zmm2
+; AVX512-NEXT:    vfmadd213ps %zmm1, %zmm0, %zmm2
+; AVX512-NEXT:    vmovaps %zmm2, %zmm0
 ; AVX512-NEXT:    retq
-  %x = fmul <8 x double> %a0, %a1
-  %res = fsub <8 x double> %x, %a2
+  %x = load <16 x float>, <16 x float>* %a0
+  %y = fmul <16 x float> %x, %a1
+  %res = fadd <16 x float> %y, %a2
+  ret <16 x float> %res
+}
+
+define <8 x double> @test_8f64_fmsub_load(<8 x double>* %a0, <8 x double> %a1, <8 x double> %a2) {
+; FMA-LABEL: test_8f64_fmsub_load:
+; FMA:       # BB#0:
+; FMA-NEXT:    vfmsub132pd (%rdi), %ymm2, %ymm0
+; FMA-NEXT:    vfmsub132pd 32(%rdi), %ymm3, %ymm1
+; FMA-NEXT:    retq
+;
+; FMA4-LABEL: test_8f64_fmsub_load:
+; FMA4:       # BB#0:
+; FMA4-NEXT:    vfmsubpd %ymm2, (%rdi), %ymm0, %ymm0
+; FMA4-NEXT:    vfmsubpd %ymm3, 32(%rdi), %ymm1, %ymm1
+; FMA4-NEXT:    retq
+;
+; AVX512-LABEL: test_8f64_fmsub_load:
+; AVX512:       # BB#0:
+; AVX512-NEXT:    vmovapd (%rdi), %zmm2
+; AVX512-NEXT:    vfmsub213pd %zmm1, %zmm0, %zmm2
+; AVX512-NEXT:    vmovaps %zmm2, %zmm0
+; AVX512-NEXT:    retq
+  %x = load <8 x double>, <8 x double>* %a0
+  %y = fmul <8 x double> %x, %a1
+  %res = fsub <8 x double> %y, %a2
   ret <8 x double> %res
 }