maxjhandsome
/
llvm-project
forked from OSchip/llvm-project
1
0
Fork 0
Code Issues Pull Requests Packages Releases Wiki Activity

[X86][AVX] matchShuffleAsBlend - use isElementEquivalent to help match broadcast/repeated elements 

Extend matchShuffleAsBlend to not only match against known in-place elements for BLEND shuffles, but use isElementEquivalent to determine if the shuffle mask's referenced element is the same as the in-place element.

This allows us to replace a number of insertps instructions with more general blendps instructions (better opportunities for commutation, concatenation etc.).
This commit is contained in:
Simon Pilgrim 2021-08-22 15:26:17 +01:00
parent 96fb3eef66
commit 352df10a23
4 changed files with 55 additions and 51 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

9
llvm/lib/Target/X86/X86ISelLowering.cpp
View File

@ -12248,10 +12248,15 @@ static bool matchShuffleAsBlend(SDValue V1, SDValue V2,
int M = Mask[i];
if (M == SM_SentinelUndef)
continue;
if (M == i)
if (M == i ||
(0 <= M && M < Size && IsElementEquivalent(Size, V1, V1, M, i))) {
Mask[i] = i;
continue;
if (M == i + Size) {
}
if (M == (i + Size) ||
(Size <= M && IsElementEquivalent(Size, V2, V2, M - Size, i))) {
BlendMask |= 1ull << i;
Mask[i] = i + Size;
continue;
}
if (Zeroable[i]) {

17
llvm/test/CodeGen/X86/avx.ll
View File

@ -138,18 +138,17 @@ define <4 x float> @insertps_from_broadcast_loadv4f32(<4 x float> %a, <4 x float
ret <4 x float> %7
}
;; FIXME: We're emitting an extraneous pshufd/vbroadcast.
define <4 x float> @insertps_from_broadcast_multiple_use(<4 x float> %a, <4 x float> %b, <4 x float> %c, <4 x float> %d, float* nocapture readonly %fb, i64 %index) {
; X86-LABEL: insertps_from_broadcast_multiple_use:
; X86: ## %bb.0:
; X86-NEXT: movl {{[0-9]+}}(%esp), %eax
; X86-NEXT: movl {{[0-9]+}}(%esp), %ecx
; X86-NEXT: vbroadcastss (%ecx,%eax,4), %xmm4
; X86-NEXT: vinsertps {{.*#+}} xmm0 = xmm0[0,1,2],xmm4[0]
; X86-NEXT: vinsertps {{.*#+}} xmm1 = xmm1[0,1,2],xmm4[0]
; X86-NEXT: vblendps {{.*#+}} xmm0 = xmm0[0,1,2],xmm4[3]
; X86-NEXT: vblendps {{.*#+}} xmm1 = xmm1[0,1,2],xmm4[3]
; X86-NEXT: vaddps %xmm1, %xmm0, %xmm0
; X86-NEXT: vinsertps {{.*#+}} xmm1 = xmm2[0,1,2],xmm4[0]
; X86-NEXT: vinsertps {{.*#+}} xmm2 = xmm3[0,1,2],xmm4[0]
; X86-NEXT: vblendps {{.*#+}} xmm1 = xmm2[0,1,2],xmm4[3]
; X86-NEXT: vblendps {{.*#+}} xmm2 = xmm3[0,1,2],xmm4[3]
; X86-NEXT: vaddps %xmm2, %xmm1, %xmm1
; X86-NEXT: vaddps %xmm1, %xmm0, %xmm0
; X86-NEXT: retl
@ -157,11 +156,11 @@ define <4 x float> @insertps_from_broadcast_multiple_use(<4 x float> %a, <4 x fl
; X64-LABEL: insertps_from_broadcast_multiple_use:
; X64: ## %bb.0:
; X64-NEXT: vbroadcastss (%rdi,%rsi,4), %xmm4
; X64-NEXT: vinsertps {{.*#+}} xmm0 = xmm0[0,1,2],xmm4[0]
; X64-NEXT: vinsertps {{.*#+}} xmm1 = xmm1[0,1,2],xmm4[0]
; X64-NEXT: vblendps {{.*#+}} xmm0 = xmm0[0,1,2],xmm4[3]
; X64-NEXT: vblendps {{.*#+}} xmm1 = xmm1[0,1,2],xmm4[3]
; X64-NEXT: vaddps %xmm1, %xmm0, %xmm0
; X64-NEXT: vinsertps {{.*#+}} xmm1 = xmm2[0,1,2],xmm4[0]
; X64-NEXT: vinsertps {{.*#+}} xmm2 = xmm3[0,1,2],xmm4[0]
; X64-NEXT: vblendps {{.*#+}} xmm1 = xmm2[0,1,2],xmm4[3]
; X64-NEXT: vblendps {{.*#+}} xmm2 = xmm3[0,1,2],xmm4[3]
; X64-NEXT: vaddps %xmm2, %xmm1, %xmm1
; X64-NEXT: vaddps %xmm1, %xmm0, %xmm0
; X64-NEXT: retq

12
llvm/test/CodeGen/X86/horizontal-sum.ll
View File

@ -39,7 +39,7 @@ define <4 x float> @pair_sum_v4f32_v4f32(<4 x float> %0, <4 x float> %1, <4 x fl
; AVX1-SLOW-NEXT: vshufps {{.*#+}} xmm2 = xmm0[0,2],xmm1[0,1]
; AVX1-SLOW-NEXT: vshufps {{.*#+}} xmm0 = xmm0[1,3],xmm1[1,1]
; AVX1-SLOW-NEXT: vhaddps %xmm3, %xmm3, %xmm1
; AVX1-SLOW-NEXT: vinsertps {{.*#+}} xmm0 = xmm0[0,1,2],xmm1[1]
; AVX1-SLOW-NEXT: vblendps {{.*#+}} xmm0 = xmm0[0,1,2],xmm1[3]
; AVX1-SLOW-NEXT: vinsertps {{.*#+}} xmm1 = xmm2[0,1,2],xmm1[0]
; AVX1-SLOW-NEXT: vaddps %xmm0, %xmm1, %xmm0
; AVX1-SLOW-NEXT: retq
@ -58,7 +58,7 @@ define <4 x float> @pair_sum_v4f32_v4f32(<4 x float> %0, <4 x float> %1, <4 x fl
; AVX2-SLOW-NEXT: vshufps {{.*#+}} xmm2 = xmm0[0,2],xmm1[0,3]
; AVX2-SLOW-NEXT: vshufps {{.*#+}} xmm0 = xmm0[1,3],xmm1[1,1]
; AVX2-SLOW-NEXT: vhaddps %xmm3, %xmm3, %xmm1
; AVX2-SLOW-NEXT: vinsertps {{.*#+}} xmm0 = xmm0[0,1,2],xmm1[1]
; AVX2-SLOW-NEXT: vblendps {{.*#+}} xmm0 = xmm0[0,1,2],xmm1[3]
; AVX2-SLOW-NEXT: vinsertps {{.*#+}} xmm1 = xmm2[0,1,2],xmm1[0]
; AVX2-SLOW-NEXT: vaddps %xmm0, %xmm1, %xmm0
; AVX2-SLOW-NEXT: retq
@ -227,7 +227,7 @@ define <8 x float> @pair_sum_v8f32_v4f32(<4 x float> %0, <4 x float> %1, <4 x fl
; AVX1-SLOW-NEXT: vshufps {{.*#+}} xmm3 = xmm2[0,2],xmm1[0,1]
; AVX1-SLOW-NEXT: vinsertps {{.*#+}} xmm3 = xmm3[0,1,2],xmm4[0]
; AVX1-SLOW-NEXT: vshufps {{.*#+}} xmm1 = xmm2[1,3],xmm1[1,3]
; AVX1-SLOW-NEXT: vinsertps {{.*#+}} xmm1 = xmm1[0,1,2],xmm4[1]
; AVX1-SLOW-NEXT: vblendps {{.*#+}} xmm1 = xmm1[0,1,2],xmm4[3]
; AVX1-SLOW-NEXT: vaddps %xmm1, %xmm3, %xmm1
; AVX1-SLOW-NEXT: vmovlhps {{.*#+}} xmm0 = xmm0[0],xmm1[0]
; AVX1-SLOW-NEXT: vpermilpd {{.*#+}} xmm1 = xmm1[1,0]
@ -248,7 +248,7 @@ define <8 x float> @pair_sum_v8f32_v4f32(<4 x float> %0, <4 x float> %1, <4 x fl
; AVX1-FAST-NEXT: vshufps {{.*#+}} xmm3 = xmm2[0,2],xmm1[0,1]
; AVX1-FAST-NEXT: vinsertps {{.*#+}} xmm3 = xmm3[0,1,2],xmm4[0]
; AVX1-FAST-NEXT: vshufps {{.*#+}} xmm1 = xmm2[1,3],xmm1[1,3]
; AVX1-FAST-NEXT: vinsertps {{.*#+}} xmm1 = xmm1[0,1,2],xmm4[1]
; AVX1-FAST-NEXT: vblendps {{.*#+}} xmm1 = xmm1[0,1,2],xmm4[3]
; AVX1-FAST-NEXT: vaddps %xmm1, %xmm3, %xmm1
; AVX1-FAST-NEXT: vmovlhps {{.*#+}} xmm0 = xmm0[0],xmm1[0]
; AVX1-FAST-NEXT: vpermilpd {{.*#+}} xmm1 = xmm1[1,0]
@ -271,7 +271,7 @@ define <8 x float> @pair_sum_v8f32_v4f32(<4 x float> %0, <4 x float> %1, <4 x fl
; AVX2-SLOW-NEXT: vshufps {{.*#+}} xmm3 = xmm2[0,2],xmm1[0,3]
; AVX2-SLOW-NEXT: vinsertps {{.*#+}} xmm3 = xmm3[0,1,2],xmm4[0]
; AVX2-SLOW-NEXT: vshufps {{.*#+}} xmm1 = xmm2[1,3],xmm1[1,3]
; AVX2-SLOW-NEXT: vinsertps {{.*#+}} xmm1 = xmm1[0,1,2],xmm4[1]
; AVX2-SLOW-NEXT: vblendps {{.*#+}} xmm1 = xmm1[0,1,2],xmm4[3]
; AVX2-SLOW-NEXT: vaddps %xmm1, %xmm3, %xmm1
; AVX2-SLOW-NEXT: vmovlhps {{.*#+}} xmm0 = xmm0[0],xmm1[0]
; AVX2-SLOW-NEXT: vpermilpd {{.*#+}} xmm1 = xmm1[1,0]
@ -292,7 +292,7 @@ define <8 x float> @pair_sum_v8f32_v4f32(<4 x float> %0, <4 x float> %1, <4 x fl
; AVX2-FAST-NEXT: vshufps {{.*#+}} xmm3 = xmm2[0,2],xmm1[0,3]
; AVX2-FAST-NEXT: vinsertps {{.*#+}} xmm3 = xmm3[0,1,2],xmm4[0]
; AVX2-FAST-NEXT: vshufps {{.*#+}} xmm1 = xmm2[1,3],xmm1[1,3]
; AVX2-FAST-NEXT: vinsertps {{.*#+}} xmm1 = xmm1[0,1,2],xmm4[1]
; AVX2-FAST-NEXT: vblendps {{.*#+}} xmm1 = xmm1[0,1,2],xmm4[3]
; AVX2-FAST-NEXT: vaddps %xmm1, %xmm3, %xmm1
; AVX2-FAST-NEXT: vmovlhps {{.*#+}} xmm0 = xmm0[0],xmm1[0]
; AVX2-FAST-NEXT: vpermilpd {{.*#+}} xmm1 = xmm1[1,0]

68
llvm/test/CodeGen/X86/sse41.ll
View File

@ -1661,15 +1661,15 @@ define <4 x float> @insertps_from_broadcast_multiple_use(<4 x float> %a, <4 x fl
; X86-AVX1-NEXT: movl {{[0-9]+}}(%esp), %eax ## encoding: [0x8b,0x44,0x24,0x08]
; X86-AVX1-NEXT: movl {{[0-9]+}}(%esp), %ecx ## encoding: [0x8b,0x4c,0x24,0x04]
; X86-AVX1-NEXT: vbroadcastss (%ecx,%eax,4), %xmm4 ## encoding: [0xc4,0xe2,0x79,0x18,0x24,0x81]
; X86-AVX1-NEXT: vinsertps $48, %xmm4, %xmm0, %xmm0 ## encoding: [0xc4,0xe3,0x79,0x21,0xc4,0x30]
; X86-AVX1-NEXT: ## xmm0 = xmm0[0,1,2],xmm4[0]
; X86-AVX1-NEXT: vinsertps $48, %xmm4, %xmm1, %xmm1 ## encoding: [0xc4,0xe3,0x71,0x21,0xcc,0x30]
; X86-AVX1-NEXT: ## xmm1 = xmm1[0,1,2],xmm4[0]
; X86-AVX1-NEXT: vblendps $8, %xmm4, %xmm0, %xmm0 ## encoding: [0xc4,0xe3,0x79,0x0c,0xc4,0x08]
; X86-AVX1-NEXT: ## xmm0 = xmm0[0,1,2],xmm4[3]
; X86-AVX1-NEXT: vblendps $8, %xmm4, %xmm1, %xmm1 ## encoding: [0xc4,0xe3,0x71,0x0c,0xcc,0x08]
; X86-AVX1-NEXT: ## xmm1 = xmm1[0,1,2],xmm4[3]
; X86-AVX1-NEXT: vaddps %xmm1, %xmm0, %xmm0 ## encoding: [0xc5,0xf8,0x58,0xc1]
; X86-AVX1-NEXT: vinsertps $48, %xmm4, %xmm2, %xmm1 ## encoding: [0xc4,0xe3,0x69,0x21,0xcc,0x30]
; X86-AVX1-NEXT: ## xmm1 = xmm2[0,1,2],xmm4[0]
; X86-AVX1-NEXT: vinsertps $48, %xmm4, %xmm3, %xmm2 ## encoding: [0xc4,0xe3,0x61,0x21,0xd4,0x30]
; X86-AVX1-NEXT: ## xmm2 = xmm3[0,1,2],xmm4[0]
; X86-AVX1-NEXT: vblendps $8, %xmm4, %xmm2, %xmm1 ## encoding: [0xc4,0xe3,0x69,0x0c,0xcc,0x08]
; X86-AVX1-NEXT: ## xmm1 = xmm2[0,1,2],xmm4[3]
; X86-AVX1-NEXT: vblendps $8, %xmm4, %xmm3, %xmm2 ## encoding: [0xc4,0xe3,0x61,0x0c,0xd4,0x08]
; X86-AVX1-NEXT: ## xmm2 = xmm3[0,1,2],xmm4[3]
; X86-AVX1-NEXT: vaddps %xmm2, %xmm1, %xmm1 ## encoding: [0xc5,0xf0,0x58,0xca]
; X86-AVX1-NEXT: vaddps %xmm1, %xmm0, %xmm0 ## encoding: [0xc5,0xf8,0x58,0xc1]
; X86-AVX1-NEXT: retl ## encoding: [0xc3]
@ -1679,16 +1679,16 @@ define <4 x float> @insertps_from_broadcast_multiple_use(<4 x float> %a, <4 x fl
; X86-AVX512-NEXT: movl {{[0-9]+}}(%esp), %eax ## encoding: [0x8b,0x44,0x24,0x08]
; X86-AVX512-NEXT: movl {{[0-9]+}}(%esp), %ecx ## encoding: [0x8b,0x4c,0x24,0x04]
; X86-AVX512-NEXT: vbroadcastss (%ecx,%eax,4), %xmm4 ## EVEX TO VEX Compression encoding: [0xc4,0xe2,0x79,0x18,0x24,0x81]
; X86-AVX512-NEXT: vinsertps $48, %xmm4, %xmm0, %xmm0 ## EVEX TO VEX Compression encoding: [0xc4,0xe3,0x79,0x21,0xc4,0x30]
; X86-AVX512-NEXT: ## xmm0 = xmm0[0,1,2],xmm4[0]
; X86-AVX512-NEXT: vinsertps $48, %xmm4, %xmm1, %xmm1 ## EVEX TO VEX Compression encoding: [0xc4,0xe3,0x71,0x21,0xcc,0x30]
; X86-AVX512-NEXT: ## xmm1 = xmm1[0,1,2],xmm4[0]
; X86-AVX512-NEXT: vblendps $8, %xmm4, %xmm0, %xmm0 ## encoding: [0xc4,0xe3,0x79,0x0c,0xc4,0x08]
; X86-AVX512-NEXT: ## xmm0 = xmm0[0,1,2],xmm4[3]
; X86-AVX512-NEXT: vblendps $8, %xmm4, %xmm1, %xmm1 ## encoding: [0xc4,0xe3,0x71,0x0c,0xcc,0x08]
; X86-AVX512-NEXT: ## xmm1 = xmm1[0,1,2],xmm4[3]
; X86-AVX512-NEXT: vblendps $8, %xmm4, %xmm2, %xmm2 ## encoding: [0xc4,0xe3,0x69,0x0c,0xd4,0x08]
; X86-AVX512-NEXT: ## xmm2 = xmm2[0,1,2],xmm4[3]
; X86-AVX512-NEXT: vblendps $8, %xmm4, %xmm3, %xmm3 ## encoding: [0xc4,0xe3,0x61,0x0c,0xdc,0x08]
; X86-AVX512-NEXT: ## xmm3 = xmm3[0,1,2],xmm4[3]
; X86-AVX512-NEXT: vaddps %xmm1, %xmm0, %xmm0 ## EVEX TO VEX Compression encoding: [0xc5,0xf8,0x58,0xc1]
; X86-AVX512-NEXT: vinsertps $48, %xmm4, %xmm2, %xmm1 ## EVEX TO VEX Compression encoding: [0xc4,0xe3,0x69,0x21,0xcc,0x30]
; X86-AVX512-NEXT: ## xmm1 = xmm2[0,1,2],xmm4[0]
; X86-AVX512-NEXT: vinsertps $48, %xmm4, %xmm3, %xmm2 ## EVEX TO VEX Compression encoding: [0xc4,0xe3,0x61,0x21,0xd4,0x30]
; X86-AVX512-NEXT: ## xmm2 = xmm3[0,1,2],xmm4[0]
; X86-AVX512-NEXT: vaddps %xmm2, %xmm1, %xmm1 ## EVEX TO VEX Compression encoding: [0xc5,0xf0,0x58,0xca]
; X86-AVX512-NEXT: vaddps %xmm3, %xmm2, %xmm1 ## EVEX TO VEX Compression encoding: [0xc5,0xe8,0x58,0xcb]
; X86-AVX512-NEXT: vaddps %xmm1, %xmm0, %xmm0 ## EVEX TO VEX Compression encoding: [0xc5,0xf8,0x58,0xc1]
; X86-AVX512-NEXT: retl ## encoding: [0xc3]
;
@ -1712,15 +1712,15 @@ define <4 x float> @insertps_from_broadcast_multiple_use(<4 x float> %a, <4 x fl
; X64-AVX1-LABEL: insertps_from_broadcast_multiple_use:
; X64-AVX1: ## %bb.0:
; X64-AVX1-NEXT: vbroadcastss (%rdi,%rsi,4), %xmm4 ## encoding: [0xc4,0xe2,0x79,0x18,0x24,0xb7]
; X64-AVX1-NEXT: vinsertps $48, %xmm4, %xmm0, %xmm0 ## encoding: [0xc4,0xe3,0x79,0x21,0xc4,0x30]
; X64-AVX1-NEXT: ## xmm0 = xmm0[0,1,2],xmm4[0]
; X64-AVX1-NEXT: vinsertps $48, %xmm4, %xmm1, %xmm1 ## encoding: [0xc4,0xe3,0x71,0x21,0xcc,0x30]
; X64-AVX1-NEXT: ## xmm1 = xmm1[0,1,2],xmm4[0]
; X64-AVX1-NEXT: vblendps $8, %xmm4, %xmm0, %xmm0 ## encoding: [0xc4,0xe3,0x79,0x0c,0xc4,0x08]
; X64-AVX1-NEXT: ## xmm0 = xmm0[0,1,2],xmm4[3]
; X64-AVX1-NEXT: vblendps $8, %xmm4, %xmm1, %xmm1 ## encoding: [0xc4,0xe3,0x71,0x0c,0xcc,0x08]
; X64-AVX1-NEXT: ## xmm1 = xmm1[0,1,2],xmm4[3]
; X64-AVX1-NEXT: vaddps %xmm1, %xmm0, %xmm0 ## encoding: [0xc5,0xf8,0x58,0xc1]
; X64-AVX1-NEXT: vinsertps $48, %xmm4, %xmm2, %xmm1 ## encoding: [0xc4,0xe3,0x69,0x21,0xcc,0x30]
; X64-AVX1-NEXT: ## xmm1 = xmm2[0,1,2],xmm4[0]
; X64-AVX1-NEXT: vinsertps $48, %xmm4, %xmm3, %xmm2 ## encoding: [0xc4,0xe3,0x61,0x21,0xd4,0x30]
; X64-AVX1-NEXT: ## xmm2 = xmm3[0,1,2],xmm4[0]
; X64-AVX1-NEXT: vblendps $8, %xmm4, %xmm2, %xmm1 ## encoding: [0xc4,0xe3,0x69,0x0c,0xcc,0x08]
; X64-AVX1-NEXT: ## xmm1 = xmm2[0,1,2],xmm4[3]
; X64-AVX1-NEXT: vblendps $8, %xmm4, %xmm3, %xmm2 ## encoding: [0xc4,0xe3,0x61,0x0c,0xd4,0x08]
; X64-AVX1-NEXT: ## xmm2 = xmm3[0,1,2],xmm4[3]
; X64-AVX1-NEXT: vaddps %xmm2, %xmm1, %xmm1 ## encoding: [0xc5,0xf0,0x58,0xca]
; X64-AVX1-NEXT: vaddps %xmm1, %xmm0, %xmm0 ## encoding: [0xc5,0xf8,0x58,0xc1]
; X64-AVX1-NEXT: retq ## encoding: [0xc3]
@ -1728,16 +1728,16 @@ define <4 x float> @insertps_from_broadcast_multiple_use(<4 x float> %a, <4 x fl
; X64-AVX512-LABEL: insertps_from_broadcast_multiple_use:
; X64-AVX512: ## %bb.0:
; X64-AVX512-NEXT: vbroadcastss (%rdi,%rsi,4), %xmm4 ## EVEX TO VEX Compression encoding: [0xc4,0xe2,0x79,0x18,0x24,0xb7]
; X64-AVX512-NEXT: vinsertps $48, %xmm4, %xmm0, %xmm0 ## EVEX TO VEX Compression encoding: [0xc4,0xe3,0x79,0x21,0xc4,0x30]
; X64-AVX512-NEXT: ## xmm0 = xmm0[0,1,2],xmm4[0]
; X64-AVX512-NEXT: vinsertps $48, %xmm4, %xmm1, %xmm1 ## EVEX TO VEX Compression encoding: [0xc4,0xe3,0x71,0x21,0xcc,0x30]
; X64-AVX512-NEXT: ## xmm1 = xmm1[0,1,2],xmm4[0]
; X64-AVX512-NEXT: vblendps $8, %xmm4, %xmm0, %xmm0 ## encoding: [0xc4,0xe3,0x79,0x0c,0xc4,0x08]
; X64-AVX512-NEXT: ## xmm0 = xmm0[0,1,2],xmm4[3]
; X64-AVX512-NEXT: vblendps $8, %xmm4, %xmm1, %xmm1 ## encoding: [0xc4,0xe3,0x71,0x0c,0xcc,0x08]
; X64-AVX512-NEXT: ## xmm1 = xmm1[0,1,2],xmm4[3]
; X64-AVX512-NEXT: vblendps $8, %xmm4, %xmm2, %xmm2 ## encoding: [0xc4,0xe3,0x69,0x0c,0xd4,0x08]
; X64-AVX512-NEXT: ## xmm2 = xmm2[0,1,2],xmm4[3]
; X64-AVX512-NEXT: vblendps $8, %xmm4, %xmm3, %xmm3 ## encoding: [0xc4,0xe3,0x61,0x0c,0xdc,0x08]
; X64-AVX512-NEXT: ## xmm3 = xmm3[0,1,2],xmm4[3]
; X64-AVX512-NEXT: vaddps %xmm1, %xmm0, %xmm0 ## EVEX TO VEX Compression encoding: [0xc5,0xf8,0x58,0xc1]
; X64-AVX512-NEXT: vinsertps $48, %xmm4, %xmm2, %xmm1 ## EVEX TO VEX Compression encoding: [0xc4,0xe3,0x69,0x21,0xcc,0x30]
; X64-AVX512-NEXT: ## xmm1 = xmm2[0,1,2],xmm4[0]
; X64-AVX512-NEXT: vinsertps $48, %xmm4, %xmm3, %xmm2 ## EVEX TO VEX Compression encoding: [0xc4,0xe3,0x61,0x21,0xd4,0x30]
; X64-AVX512-NEXT: ## xmm2 = xmm3[0,1,2],xmm4[0]
; X64-AVX512-NEXT: vaddps %xmm2, %xmm1, %xmm1 ## EVEX TO VEX Compression encoding: [0xc5,0xf0,0x58,0xca]
; X64-AVX512-NEXT: vaddps %xmm3, %xmm2, %xmm1 ## EVEX TO VEX Compression encoding: [0xc5,0xe8,0x58,0xcb]
; X64-AVX512-NEXT: vaddps %xmm1, %xmm0, %xmm0 ## EVEX TO VEX Compression encoding: [0xc5,0xf8,0x58,0xc1]
; X64-AVX512-NEXT: retq ## encoding: [0xc3]
%1 = getelementptr inbounds float, float* %fb, i64 %index