[X86][AVX] Added some shuffle combine from load tests

As discussed on D19198 - we need to check what happens when we shuffle with different value type to the load

llvm-svn: 269068

2 files changed, 80 insertions, 0 deletions

diff --git a/llvm/test/CodeGen/X86/vector-shuffle-combining-avx.ll b/llvm/test/CodeGen/X86/vector-shuffle-combining-avx.ll
index cbc4fc0dee7..8f00703cf37 100644
--- a/llvm/test/CodeGen/X86/vector-shuffle-combining-avx.ll
+++ b/llvm/test/CodeGen/X86/vector-shuffle-combining-avx.ll
@@ -36,6 +36,15 @@ define <4 x float> @combine_vpermilvar_4f32_movddup(<4 x float> %a0) {
   %1 = tail call <4 x float> @llvm.x86.avx.vpermilvar.ps(<4 x float> %a0, <4 x i32> <i32 0, i32 1, i32 0, i32 1>)
   ret <4 x float> %1
 }
+define <4 x float> @combine_vpermilvar_4f32_movddup_load(<4 x float> *%a0) {
+; ALL-LABEL: combine_vpermilvar_4f32_movddup_load:
+; ALL:       # BB#0:
+; ALL-NEXT:    vmovddup {{.*#+}} xmm0 = mem[0,0]
+; ALL-NEXT:    retq
+  %1 = load <4 x float>, <4 x float> *%a0
+  %2 = tail call <4 x float> @llvm.x86.avx.vpermilvar.ps(<4 x float> %1, <4 x i32> <i32 0, i32 1, i32 0, i32 1>)
+  ret <4 x float> %2
+}
 
 define <4 x float> @combine_vpermilvar_4f32_movshdup(<4 x float> %a0) {
 ; ALL-LABEL: combine_vpermilvar_4f32_movshdup:
@@ -90,6 +99,16 @@ define <8 x float> @combine_vpermilvar_8f32_movddup(<8 x float> %a0) {
   %1 = tail call <8 x float> @llvm.x86.avx.vpermilvar.ps.256(<8 x float> %a0, <8 x i32> <i32 0, i32 1, i32 0, i32 1, i32 4, i32 5, i32 4, i32 5>)
   ret <8 x float> %1
 }
+define <8 x float> @combine_vpermilvar_8f32_movddup_load(<8 x float> *%a0) {
+; ALL-LABEL: combine_vpermilvar_8f32_movddup_load:
+; ALL:       # BB#0:
+; ALL-NEXT:    vmovaps (%rdi), %ymm0
+; ALL-NEXT:    vpermilps {{.*#+}} ymm0 = ymm0[0,1,0,1,4,5,4,5]
+; ALL-NEXT:    retq
+  %1 = load <8 x float>, <8 x float> *%a0
+  %2 = tail call <8 x float> @llvm.x86.avx.vpermilvar.ps.256(<8 x float> %1, <8 x i32> <i32 0, i32 1, i32 0, i32 1, i32 4, i32 5, i32 4, i32 5>)
+  ret <8 x float> %2
+}
 
 define <8 x float> @combine_vpermilvar_8f32_movshdup(<8 x float> %a0) {
 ; ALL-LABEL: combine_vpermilvar_8f32_movshdup:
diff --git a/llvm/test/CodeGen/X86/vector-shuffle-combining-avx512bw.ll b/llvm/test/CodeGen/X86/vector-shuffle-combining-avx512bw.ll
index 0268d9ef50a..23e1896f778 100644
--- a/llvm/test/CodeGen/X86/vector-shuffle-combining-avx512bw.ll
+++ b/llvm/test/CodeGen/X86/vector-shuffle-combining-avx512bw.ll
@@ -41,6 +41,18 @@ define <8 x double> @combine_vpermt2var_8f64_movddup(<8 x double> %x0, <8 x doub
   %res0 = call <8 x double> @llvm.x86.avx512.maskz.vpermt2var.pd.512(<8 x i64> <i64 0, i64 0, i64 2, i64 2, i64 4, i64 4, i64 6, i64 6>, <8 x double> %x0, <8 x double> %x1, i8 -1)
   ret <8 x double> %res0
 }
+define <8 x double> @combine_vpermt2var_8f64_movddup_load(<8 x double> *%p0, <8 x double> %x1) {
+; CHECK-LABEL: combine_vpermt2var_8f64_movddup_load:
+; CHECK:       # BB#0:
+; CHECK-NEXT:    vmovapd (%rdi), %zmm1
+; CHECK-NEXT:    vmovdqa64 {{.*#+}} zmm2 = [0,0,2,2,4,4,6,6]
+; CHECK-NEXT:    vpermt2pd %zmm0, %zmm2, %zmm1
+; CHECK-NEXT:    vmovaps %zmm1, %zmm0
+; CHECK-NEXT:    retq
+  %x0 = load <8 x double>, <8 x double> *%p0
+  %res0 = call <8 x double> @llvm.x86.avx512.maskz.vpermt2var.pd.512(<8 x i64> <i64 0, i64 0, i64 2, i64 2, i64 4, i64 4, i64 6, i64 6>, <8 x double> %x0, <8 x double> %x1, i8 -1)
+  ret <8 x double> %res0
+}
 define <8 x double> @combine_vpermt2var_8f64_movddup_mask(<8 x double> %x0, <8 x double> %x1, i8 %m) {
 ; CHECK-LABEL: combine_vpermt2var_8f64_movddup_mask:
 ; CHECK:       # BB#0:
@@ -105,6 +117,18 @@ define <16 x float> @combine_vpermt2var_16f32_vmovddup(<16 x float> %x0, <16 x f
   %res0 = call <16 x float> @llvm.x86.avx512.maskz.vpermt2var.ps.512(<16 x i32> <i32 0, i32 1, i32 0, i32 1, i32 4, i32 5, i32 4, i32 5, i32 8, i32 9, i32 8, i32 9, i32 12, i32 13, i32 12, i32 13>, <16 x float> %x0, <16 x float> %x1, i16 -1)
   ret <16 x float> %res0
 }
+define <16 x float> @combine_vpermt2var_16f32_vmovddup_load(<16 x float> *%p0, <16 x float> %x1) {
+; CHECK-LABEL: combine_vpermt2var_16f32_vmovddup_load:
+; CHECK:       # BB#0:
+; CHECK-NEXT:    vmovaps (%rdi), %zmm1
+; CHECK-NEXT:    vmovdqa32 {{.*#+}} zmm2 = [0,1,0,1,4,5,4,5,8,9,8,9,12,13,12,13]
+; CHECK-NEXT:    vpermt2ps %zmm0, %zmm2, %zmm1
+; CHECK-NEXT:    vmovaps %zmm1, %zmm0
+; CHECK-NEXT:    retq
+  %x0 = load <16 x float>, <16 x float> *%p0
+  %res0 = call <16 x float> @llvm.x86.avx512.maskz.vpermt2var.ps.512(<16 x i32> <i32 0, i32 1, i32 0, i32 1, i32 4, i32 5, i32 4, i32 5, i32 8, i32 9, i32 8, i32 9, i32 12, i32 13, i32 12, i32 13>, <16 x float> %x0, <16 x float> %x1, i16 -1)
+  ret <16 x float> %res0
+}
 define <16 x float> @combine_vpermt2var_16f32_vmovddup_mask(<16 x float> %x0, <16 x float> %x1, i16 %m) {
 ; CHECK-LABEL: combine_vpermt2var_16f32_vmovddup_mask:
 ; CHECK:       # BB#0:
@@ -125,6 +149,18 @@ define <16 x float> @combine_vpermt2var_16f32_vmovshdup(<16 x float> %x0, <16 x
   %res0 = call <16 x float> @llvm.x86.avx512.maskz.vpermt2var.ps.512(<16 x i32> <i32 1, i32 1, i32 3, i32 3, i32 5, i32 5, i32 7, i32 7, i32 9, i32 9, i32 11, i32 11, i32 13, i32 13, i32 15, i32 15>, <16 x float> %x0, <16 x float> %x1, i16 -1)
   ret <16 x float> %res0
 }
+define <16 x float> @combine_vpermt2var_16f32_vmovshdup_load(<16 x float> *%p0, <16 x float> %x1) {
+; CHECK-LABEL: combine_vpermt2var_16f32_vmovshdup_load:
+; CHECK:       # BB#0:
+; CHECK-NEXT:    vmovaps (%rdi), %zmm1
+; CHECK-NEXT:    vmovdqa32 {{.*#+}} zmm2 = [1,1,3,3,5,5,7,7,9,9,11,11,13,13,15,15]
+; CHECK-NEXT:    vpermt2ps %zmm0, %zmm2, %zmm1
+; CHECK-NEXT:    vmovaps %zmm1, %zmm0
+; CHECK-NEXT:    retq
+  %x0 = load <16 x float>, <16 x float> *%p0
+  %res0 = call <16 x float> @llvm.x86.avx512.maskz.vpermt2var.ps.512(<16 x i32> <i32 1, i32 1, i32 3, i32 3, i32 5, i32 5, i32 7, i32 7, i32 9, i32 9, i32 11, i32 11, i32 13, i32 13, i32 15, i32 15>, <16 x float> %x0, <16 x float> %x1, i16 -1)
+  ret <16 x float> %res0
+}
 define <16 x float> @combine_vpermt2var_16f32_vmovshdup_mask(<16 x float> %x0, <16 x float> %x1, i16 %m) {
 ; CHECK-LABEL: combine_vpermt2var_16f32_vmovshdup_mask:
 ; CHECK:       # BB#0:
@@ -145,6 +181,18 @@ define <16 x float> @combine_vpermt2var_16f32_vmovsldup(<16 x float> %x0, <16 x
   %res0 = call <16 x float> @llvm.x86.avx512.maskz.vpermt2var.ps.512(<16 x i32> <i32 0, i32 0, i32 2, i32 2, i32 4, i32 4, i32 6, i32 6, i32 8, i32 8, i32 10, i32 10, i32 12, i32 12, i32 14, i32 14>, <16 x float> %x0, <16 x float> %x1, i16 -1)
   ret <16 x float> %res0
 }
+define <16 x float> @combine_vpermt2var_16f32_vmovsldup_load(<16 x float> *%p0, <16 x float> %x1) {
+; CHECK-LABEL: combine_vpermt2var_16f32_vmovsldup_load:
+; CHECK:       # BB#0:
+; CHECK-NEXT:    vmovaps (%rdi), %zmm1
+; CHECK-NEXT:    vmovdqa32 {{.*#+}} zmm2 = [0,0,2,2,4,4,6,6,8,8,10,10,12,12,14,14]
+; CHECK-NEXT:    vpermt2ps %zmm0, %zmm2, %zmm1
+; CHECK-NEXT:    vmovaps %zmm1, %zmm0
+; CHECK-NEXT:    retq
+  %x0 = load <16 x float>, <16 x float> *%p0
+  %res0 = call <16 x float> @llvm.x86.avx512.maskz.vpermt2var.ps.512(<16 x i32> <i32 0, i32 0, i32 2, i32 2, i32 4, i32 4, i32 6, i32 6, i32 8, i32 8, i32 10, i32 10, i32 12, i32 12, i32 14, i32 14>, <16 x float> %x0, <16 x float> %x1, i16 -1)
+  ret <16 x float> %res0
+}
 define <16 x float> @combine_vpermt2var_16f32_vmovsldup_mask(<16 x float> %x0, <16 x float> %x1, i16 %m) {
 ; CHECK-LABEL: combine_vpermt2var_16f32_vmovsldup_mask:
 ; CHECK:       # BB#0:
@@ -155,6 +203,19 @@ define <16 x float> @combine_vpermt2var_16f32_vmovsldup_mask(<16 x float> %x0, <
   %res0 = call <16 x float> @llvm.x86.avx512.maskz.vpermt2var.ps.512(<16 x i32> <i32 0, i32 0, i32 2, i32 2, i32 4, i32 4, i32 6, i32 6, i32 8, i32 8, i32 10, i32 10, i32 12, i32 12, i32 14, i32 14>, <16 x float> %x0, <16 x float> %x1, i16 %m)
   ret <16 x float> %res0
 }
+define <16 x float> @combine_vpermt2var_16f32_vmovsldup_mask_load(<16 x float> *%p0, <16 x float> %x1, i16 %m) {
+; CHECK-LABEL: combine_vpermt2var_16f32_vmovsldup_mask_load:
+; CHECK:       # BB#0:
+; CHECK-NEXT:    kmovw %esi, %k1
+; CHECK-NEXT:    vmovaps (%rdi), %zmm1
+; CHECK-NEXT:    vmovdqa32 {{.*#+}} zmm2 = [0,0,2,2,4,4,6,6,8,8,10,10,12,12,14,14]
+; CHECK-NEXT:    vpermt2ps %zmm0, %zmm2, %zmm1 {%k1} {z}
+; CHECK-NEXT:    vmovaps %zmm1, %zmm0
+; CHECK-NEXT:    retq
+  %x0 = load <16 x float>, <16 x float> *%p0
+  %res0 = call <16 x float> @llvm.x86.avx512.maskz.vpermt2var.ps.512(<16 x i32> <i32 0, i32 0, i32 2, i32 2, i32 4, i32 4, i32 6, i32 6, i32 8, i32 8, i32 10, i32 10, i32 12, i32 12, i32 14, i32 14>, <16 x float> %x0, <16 x float> %x1, i16 %m)
+  ret <16 x float> %res0
+}
 
 define <16 x i32> @combine_vpermt2var_16i32_identity(<16 x i32> %x0, <16 x i32> %x1) {
 ; CHECK-LABEL: combine_vpermt2var_16i32_identity: