From 46e57fd0735b2f57cd19a40eefcaa750b0d32dcc Mon Sep 17 00:00:00 2001
From: Simon Pilgrim <llvm-dev@redking.me.uk>
Date: Tue, 1 Mar 2016 21:38:41 +0000
Subject: [PATCH] [X86][AVX2] Regenerated intrinsics tests

llvm-svn: 262401
---
 llvm/test/CodeGen/X86/avx2-intrinsics-x86.ll | 707 ++++++++++++++++++++-------
 1 file changed, 542 insertions(+), 165 deletions(-)

diff --git a/llvm/test/CodeGen/X86/avx2-intrinsics-x86.ll b/llvm/test/CodeGen/X86/avx2-intrinsics-x86.ll
index 606aca9..130acfe 100644
--- a/llvm/test/CodeGen/X86/avx2-intrinsics-x86.ll
+++ b/llvm/test/CodeGen/X86/avx2-intrinsics-x86.ll
@@ -1,7 +1,11 @@
-; RUN: llc < %s -mtriple=x86_64-apple-darwin -march=x86 -mattr=avx2 | FileCheck %s
+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
+; RUN: llc < %s -mtriple=i686-apple-darwin -mattr=avx2 | FileCheck %s
 
 define <16 x i16> @test_x86_avx2_packssdw(<8 x i32> %a0, <8 x i32> %a1) {
-  ; CHECK: vpackssdw
+; CHECK-LABEL: test_x86_avx2_packssdw:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    vpackssdw %ymm1, %ymm0, %ymm0
+; CHECK-NEXT:    retl
   %res = call <16 x i16> @llvm.x86.avx2.packssdw(<8 x i32> %a0, <8 x i32> %a1) ; <<16 x i16>> [#uses=1]
   ret <16 x i16> %res
 }
@@ -9,7 +13,10 @@ declare <16 x i16> @llvm.x86.avx2.packssdw(<8 x i32>, <8 x i32>) nounwind readno
 
 
 define <32 x i8> @test_x86_avx2_packsswb(<16 x i16> %a0, <16 x i16> %a1) {
-  ; CHECK: vpacksswb
+; CHECK-LABEL: test_x86_avx2_packsswb:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    vpacksswb %ymm1, %ymm0, %ymm0
+; CHECK-NEXT:    retl
   %res = call <32 x i8> @llvm.x86.avx2.packsswb(<16 x i16> %a0, <16 x i16> %a1) ; <<32 x i8>> [#uses=1]
   ret <32 x i8> %res
 }
@@ -17,7 +24,10 @@ declare <32 x i8> @llvm.x86.avx2.packsswb(<16 x i16>, <16 x i16>) nounwind readn
 
 
 define <32 x i8> @test_x86_avx2_packuswb(<16 x i16> %a0, <16 x i16> %a1) {
-  ; CHECK: vpackuswb
+; CHECK-LABEL: test_x86_avx2_packuswb:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    vpackuswb %ymm1, %ymm0, %ymm0
+; CHECK-NEXT:    retl
   %res = call <32 x i8> @llvm.x86.avx2.packuswb(<16 x i16> %a0, <16 x i16> %a1) ; <<32 x i8>> [#uses=1]
   ret <32 x i8> %res
 }
@@ -25,7 +35,10 @@ declare <32 x i8> @llvm.x86.avx2.packuswb(<16 x i16>, <16 x i16>) nounwind readn
 
 
 define <32 x i8> @test_x86_avx2_padds_b(<32 x i8> %a0, <32 x i8> %a1) {
-  ; CHECK: vpaddsb
+; CHECK-LABEL: test_x86_avx2_padds_b:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    vpaddsb %ymm1, %ymm0, %ymm0
+; CHECK-NEXT:    retl
   %res = call <32 x i8> @llvm.x86.avx2.padds.b(<32 x i8> %a0, <32 x i8> %a1) ; <<32 x i8>> [#uses=1]
   ret <32 x i8> %res
 }
@@ -33,7 +46,10 @@ declare <32 x i8> @llvm.x86.avx2.padds.b(<32 x i8>, <32 x i8>) nounwind readnone
 
 
 define <16 x i16> @test_x86_avx2_padds_w(<16 x i16> %a0, <16 x i16> %a1) {
-  ; CHECK: vpaddsw
+; CHECK-LABEL: test_x86_avx2_padds_w:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    vpaddsw %ymm1, %ymm0, %ymm0
+; CHECK-NEXT:    retl
   %res = call <16 x i16> @llvm.x86.avx2.padds.w(<16 x i16> %a0, <16 x i16> %a1) ; <<16 x i16>> [#uses=1]
   ret <16 x i16> %res
 }
@@ -41,7 +57,10 @@ declare <16 x i16> @llvm.x86.avx2.padds.w(<16 x i16>, <16 x i16>) nounwind readn
 
 
 define <32 x i8> @test_x86_avx2_paddus_b(<32 x i8> %a0, <32 x i8> %a1) {
-  ; CHECK: vpaddusb
+; CHECK-LABEL: test_x86_avx2_paddus_b:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    vpaddusb %ymm1, %ymm0, %ymm0
+; CHECK-NEXT:    retl
   %res = call <32 x i8> @llvm.x86.avx2.paddus.b(<32 x i8> %a0, <32 x i8> %a1) ; <<32 x i8>> [#uses=1]
   ret <32 x i8> %res
 }
@@ -49,7 +68,10 @@ declare <32 x i8> @llvm.x86.avx2.paddus.b(<32 x i8>, <32 x i8>) nounwind readnon
 
 
 define <16 x i16> @test_x86_avx2_paddus_w(<16 x i16> %a0, <16 x i16> %a1) {
-  ; CHECK: vpaddusw
+; CHECK-LABEL: test_x86_avx2_paddus_w:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    vpaddusw %ymm1, %ymm0, %ymm0
+; CHECK-NEXT:    retl
   %res = call <16 x i16> @llvm.x86.avx2.paddus.w(<16 x i16> %a0, <16 x i16> %a1) ; <<16 x i16>> [#uses=1]
   ret <16 x i16> %res
 }
@@ -57,7 +79,10 @@ declare <16 x i16> @llvm.x86.avx2.paddus.w(<16 x i16>, <16 x i16>) nounwind read
 
 
 define <32 x i8> @test_x86_avx2_pavg_b(<32 x i8> %a0, <32 x i8> %a1) {
-  ; CHECK: vpavgb
+; CHECK-LABEL: test_x86_avx2_pavg_b:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    vpavgb %ymm1, %ymm0, %ymm0
+; CHECK-NEXT:    retl
   %res = call <32 x i8> @llvm.x86.avx2.pavg.b(<32 x i8> %a0, <32 x i8> %a1) ; <<32 x i8>> [#uses=1]
   ret <32 x i8> %res
 }
@@ -65,7 +90,10 @@ declare <32 x i8> @llvm.x86.avx2.pavg.b(<32 x i8>, <32 x i8>) nounwind readnone
 
 
 define <16 x i16> @test_x86_avx2_pavg_w(<16 x i16> %a0, <16 x i16> %a1) {
-  ; CHECK: vpavgw
+; CHECK-LABEL: test_x86_avx2_pavg_w:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    vpavgw %ymm1, %ymm0, %ymm0
+; CHECK-NEXT:    retl
   %res = call <16 x i16> @llvm.x86.avx2.pavg.w(<16 x i16> %a0, <16 x i16> %a1) ; <<16 x i16>> [#uses=1]
   ret <16 x i16> %res
 }
@@ -73,7 +101,10 @@ declare <16 x i16> @llvm.x86.avx2.pavg.w(<16 x i16>, <16 x i16>) nounwind readno
 
 
 define <8 x i32> @test_x86_avx2_pmadd_wd(<16 x i16> %a0, <16 x i16> %a1) {
-  ; CHECK: vpmaddwd
+; CHECK-LABEL: test_x86_avx2_pmadd_wd:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    vpmaddwd %ymm1, %ymm0, %ymm0
+; CHECK-NEXT:    retl
   %res = call <8 x i32> @llvm.x86.avx2.pmadd.wd(<16 x i16> %a0, <16 x i16> %a1) ; <<8 x i32>> [#uses=1]
   ret <8 x i32> %res
 }
@@ -81,7 +112,10 @@ declare <8 x i32> @llvm.x86.avx2.pmadd.wd(<16 x i16>, <16 x i16>) nounwind readn
 
 
 define <16 x i16> @test_x86_avx2_pmaxs_w(<16 x i16> %a0, <16 x i16> %a1) {
-  ; CHECK: vpmaxsw
+; CHECK-LABEL: test_x86_avx2_pmaxs_w:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    vpmaxsw %ymm1, %ymm0, %ymm0
+; CHECK-NEXT:    retl
   %res = call <16 x i16> @llvm.x86.avx2.pmaxs.w(<16 x i16> %a0, <16 x i16> %a1) ; <<16 x i16>> [#uses=1]
   ret <16 x i16> %res
 }
@@ -89,7 +123,10 @@ declare <16 x i16> @llvm.x86.avx2.pmaxs.w(<16 x i16>, <16 x i16>) nounwind readn
 
 
 define <32 x i8> @test_x86_avx2_pmaxu_b(<32 x i8> %a0, <32 x i8> %a1) {
-  ; CHECK: vpmaxub
+; CHECK-LABEL: test_x86_avx2_pmaxu_b:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    vpmaxub %ymm1, %ymm0, %ymm0
+; CHECK-NEXT:    retl
   %res = call <32 x i8> @llvm.x86.avx2.pmaxu.b(<32 x i8> %a0, <32 x i8> %a1) ; <<32 x i8>> [#uses=1]
   ret <32 x i8> %res
 }
@@ -97,7 +134,10 @@ declare <32 x i8> @llvm.x86.avx2.pmaxu.b(<32 x i8>, <32 x i8>) nounwind readnone
 
 
 define <16 x i16> @test_x86_avx2_pmins_w(<16 x i16> %a0, <16 x i16> %a1) {
-  ; CHECK: vpminsw
+; CHECK-LABEL: test_x86_avx2_pmins_w:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    vpminsw %ymm1, %ymm0, %ymm0
+; CHECK-NEXT:    retl
   %res = call <16 x i16> @llvm.x86.avx2.pmins.w(<16 x i16> %a0, <16 x i16> %a1) ; <<16 x i16>> [#uses=1]
   ret <16 x i16> %res
 }
@@ -105,7 +145,10 @@ declare <16 x i16> @llvm.x86.avx2.pmins.w(<16 x i16>, <16 x i16>) nounwind readn
 
 
 define <32 x i8> @test_x86_avx2_pminu_b(<32 x i8> %a0, <32 x i8> %a1) {
-  ; CHECK: vpminub
+; CHECK-LABEL: test_x86_avx2_pminu_b:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    vpminub %ymm1, %ymm0, %ymm0
+; CHECK-NEXT:    retl
   %res = call <32 x i8> @llvm.x86.avx2.pminu.b(<32 x i8> %a0, <32 x i8> %a1) ; <<32 x i8>> [#uses=1]
   ret <32 x i8> %res
 }
@@ -113,7 +156,11 @@ declare <32 x i8> @llvm.x86.avx2.pminu.b(<32 x i8>, <32 x i8>) nounwind readnone
 
 
 define i32 @test_x86_avx2_pmovmskb(<32 x i8> %a0) {
-  ; CHECK: vpmovmskb
+; CHECK-LABEL: test_x86_avx2_pmovmskb:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    vpmovmskb %ymm0, %eax
+; CHECK-NEXT:    vzeroupper
+; CHECK-NEXT:    retl
   %res = call i32 @llvm.x86.avx2.pmovmskb(<32 x i8> %a0) ; <i32> [#uses=1]
   ret i32 %res
 }
@@ -121,7 +168,10 @@ declare i32 @llvm.x86.avx2.pmovmskb(<32 x i8>) nounwind readnone
 
 
 define <16 x i16> @test_x86_avx2_pmulh_w(<16 x i16> %a0, <16 x i16> %a1) {
-  ; CHECK: vpmulhw
+; CHECK-LABEL: test_x86_avx2_pmulh_w:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    vpmulhw %ymm1, %ymm0, %ymm0
+; CHECK-NEXT:    retl
   %res = call <16 x i16> @llvm.x86.avx2.pmulh.w(<16 x i16> %a0, <16 x i16> %a1) ; <<16 x i16>> [#uses=1]
   ret <16 x i16> %res
 }
@@ -129,7 +179,10 @@ declare <16 x i16> @llvm.x86.avx2.pmulh.w(<16 x i16>, <16 x i16>) nounwind readn
 
 
 define <16 x i16> @test_x86_avx2_pmulhu_w(<16 x i16> %a0, <16 x i16> %a1) {
-  ; CHECK: vpmulhuw
+; CHECK-LABEL: test_x86_avx2_pmulhu_w:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    vpmulhuw %ymm1, %ymm0, %ymm0
+; CHECK-NEXT:    retl
   %res = call <16 x i16> @llvm.x86.avx2.pmulhu.w(<16 x i16> %a0, <16 x i16> %a1) ; <<16 x i16>> [#uses=1]
   ret <16 x i16> %res
 }
@@ -137,7 +190,10 @@ declare <16 x i16> @llvm.x86.avx2.pmulhu.w(<16 x i16>, <16 x i16>) nounwind read
 
 
 define <4 x i64> @test_x86_avx2_pmulu_dq(<8 x i32> %a0, <8 x i32> %a1) {
-  ; CHECK: vpmuludq
+; CHECK-LABEL: test_x86_avx2_pmulu_dq:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    vpmuludq %ymm1, %ymm0, %ymm0
+; CHECK-NEXT:    retl
   %res = call <4 x i64> @llvm.x86.avx2.pmulu.dq(<8 x i32> %a0, <8 x i32> %a1) ; <<4 x i64>> [#uses=1]
   ret <4 x i64> %res
 }
@@ -145,7 +201,10 @@ declare <4 x i64> @llvm.x86.avx2.pmulu.dq(<8 x i32>, <8 x i32>) nounwind readnon
 
 
 define <4 x i64> @test_x86_avx2_psad_bw(<32 x i8> %a0, <32 x i8> %a1) {
-  ; CHECK: vpsadbw
+; CHECK-LABEL: test_x86_avx2_psad_bw:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    vpsadbw %ymm1, %ymm0, %ymm0
+; CHECK-NEXT:    retl
   %res = call <4 x i64> @llvm.x86.avx2.psad.bw(<32 x i8> %a0, <32 x i8> %a1) ; <<4 x i64>> [#uses=1]
   ret <4 x i64> %res
 }
@@ -153,7 +212,10 @@ declare <4 x i64> @llvm.x86.avx2.psad.bw(<32 x i8>, <32 x i8>) nounwind readnone
 
 
 define <8 x i32> @test_x86_avx2_psll_d(<8 x i32> %a0, <4 x i32> %a1) {
-  ; CHECK: vpslld
+; CHECK-LABEL: test_x86_avx2_psll_d:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    vpslld %xmm1, %ymm0, %ymm0
+; CHECK-NEXT:    retl
   %res = call <8 x i32> @llvm.x86.avx2.psll.d(<8 x i32> %a0, <4 x i32> %a1) ; <<8 x i32>> [#uses=1]
   ret <8 x i32> %res
 }
@@ -161,7 +223,10 @@ declare <8 x i32> @llvm.x86.avx2.psll.d(<8 x i32>, <4 x i32>) nounwind readnone
 
 
 define <4 x i64> @test_x86_avx2_psll_q(<4 x i64> %a0, <2 x i64> %a1) {
-  ; CHECK: vpsllq
+; CHECK-LABEL: test_x86_avx2_psll_q:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    vpsllq %xmm1, %ymm0, %ymm0
+; CHECK-NEXT:    retl
   %res = call <4 x i64> @llvm.x86.avx2.psll.q(<4 x i64> %a0, <2 x i64> %a1) ; <<4 x i64>> [#uses=1]
   ret <4 x i64> %res
 }
@@ -169,7 +234,10 @@ declare <4 x i64> @llvm.x86.avx2.psll.q(<4 x i64>, <2 x i64>) nounwind readnone
 
 
 define <16 x i16> @test_x86_avx2_psll_w(<16 x i16> %a0, <8 x i16> %a1) {
-  ; CHECK: vpsllw
+; CHECK-LABEL: test_x86_avx2_psll_w:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    vpsllw %xmm1, %ymm0, %ymm0
+; CHECK-NEXT:    retl
   %res = call <16 x i16> @llvm.x86.avx2.psll.w(<16 x i16> %a0, <8 x i16> %a1) ; <<16 x i16>> [#uses=1]
   ret <16 x i16> %res
 }
@@ -177,7 +245,10 @@ declare <16 x i16> @llvm.x86.avx2.psll.w(<16 x i16>, <8 x i16>) nounwind readnon
 
 
 define <8 x i32> @test_x86_avx2_pslli_d(<8 x i32> %a0) {
-  ; CHECK: vpslld
+; CHECK-LABEL: test_x86_avx2_pslli_d:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    vpslld $7, %ymm0, %ymm0
+; CHECK-NEXT:    retl
   %res = call <8 x i32> @llvm.x86.avx2.pslli.d(<8 x i32> %a0, i32 7) ; <<8 x i32>> [#uses=1]
   ret <8 x i32> %res
 }
@@ -185,7 +256,10 @@ declare <8 x i32> @llvm.x86.avx2.pslli.d(<8 x i32>, i32) nounwind readnone
 
 
 define <4 x i64> @test_x86_avx2_pslli_q(<4 x i64> %a0) {
-  ; CHECK: vpsllq
+; CHECK-LABEL: test_x86_avx2_pslli_q:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    vpsllq $7, %ymm0, %ymm0
+; CHECK-NEXT:    retl
   %res = call <4 x i64> @llvm.x86.avx2.pslli.q(<4 x i64> %a0, i32 7) ; <<4 x i64>> [#uses=1]
   ret <4 x i64> %res
 }
@@ -193,7 +267,10 @@ declare <4 x i64> @llvm.x86.avx2.pslli.q(<4 x i64>, i32) nounwind readnone
 
 
 define <16 x i16> @test_x86_avx2_pslli_w(<16 x i16> %a0) {
-  ; CHECK: vpsllw
+; CHECK-LABEL: test_x86_avx2_pslli_w:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    vpsllw $7, %ymm0, %ymm0
+; CHECK-NEXT:    retl
   %res = call <16 x i16> @llvm.x86.avx2.pslli.w(<16 x i16> %a0, i32 7) ; <<16 x i16>> [#uses=1]
   ret <16 x i16> %res
 }
@@ -201,7 +278,10 @@ declare <16 x i16> @llvm.x86.avx2.pslli.w(<16 x i16>, i32) nounwind readnone
 
 
 define <8 x i32> @test_x86_avx2_psra_d(<8 x i32> %a0, <4 x i32> %a1) {
-  ; CHECK: vpsrad
+; CHECK-LABEL: test_x86_avx2_psra_d:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    vpsrad %xmm1, %ymm0, %ymm0
+; CHECK-NEXT:    retl
   %res = call <8 x i32> @llvm.x86.avx2.psra.d(<8 x i32> %a0, <4 x i32> %a1) ; <<8 x i32>> [#uses=1]
   ret <8 x i32> %res
 }
@@ -209,7 +289,10 @@ declare <8 x i32> @llvm.x86.avx2.psra.d(<8 x i32>, <4 x i32>) nounwind readnone
 
 
 define <16 x i16> @test_x86_avx2_psra_w(<16 x i16> %a0, <8 x i16> %a1) {
-  ; CHECK: vpsraw
+; CHECK-LABEL: test_x86_avx2_psra_w:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    vpsraw %xmm1, %ymm0, %ymm0
+; CHECK-NEXT:    retl
   %res = call <16 x i16> @llvm.x86.avx2.psra.w(<16 x i16> %a0, <8 x i16> %a1) ; <<16 x i16>> [#uses=1]
   ret <16 x i16> %res
 }
@@ -217,7 +300,10 @@ declare <16 x i16> @llvm.x86.avx2.psra.w(<16 x i16>, <8 x i16>) nounwind readnon
 
 
 define <8 x i32> @test_x86_avx2_psrai_d(<8 x i32> %a0) {
-  ; CHECK: vpsrad
+; CHECK-LABEL: test_x86_avx2_psrai_d:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    vpsrad $7, %ymm0, %ymm0
+; CHECK-NEXT:    retl
   %res = call <8 x i32> @llvm.x86.avx2.psrai.d(<8 x i32> %a0, i32 7) ; <<8 x i32>> [#uses=1]
   ret <8 x i32> %res
 }
@@ -225,7 +311,10 @@ declare <8 x i32> @llvm.x86.avx2.psrai.d(<8 x i32>, i32) nounwind readnone
 
 
 define <16 x i16> @test_x86_avx2_psrai_w(<16 x i16> %a0) {
-  ; CHECK: vpsraw
+; CHECK-LABEL: test_x86_avx2_psrai_w:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    vpsraw $7, %ymm0, %ymm0
+; CHECK-NEXT:    retl
   %res = call <16 x i16> @llvm.x86.avx2.psrai.w(<16 x i16> %a0, i32 7) ; <<16 x i16>> [#uses=1]
   ret <16 x i16> %res
 }
@@ -233,7 +322,10 @@ declare <16 x i16> @llvm.x86.avx2.psrai.w(<16 x i16>, i32) nounwind readnone
 
 
 define <8 x i32> @test_x86_avx2_psrl_d(<8 x i32> %a0, <4 x i32> %a1) {
-  ; CHECK: vpsrld
+; CHECK-LABEL: test_x86_avx2_psrl_d:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    vpsrld %xmm1, %ymm0, %ymm0
+; CHECK-NEXT:    retl
   %res = call <8 x i32> @llvm.x86.avx2.psrl.d(<8 x i32> %a0, <4 x i32> %a1) ; <<8 x i32>> [#uses=1]
   ret <8 x i32> %res
 }
@@ -241,7 +333,10 @@ declare <8 x i32> @llvm.x86.avx2.psrl.d(<8 x i32>, <4 x i32>) nounwind readnone
 
 
 define <4 x i64> @test_x86_avx2_psrl_q(<4 x i64> %a0, <2 x i64> %a1) {
-  ; CHECK: vpsrlq
+; CHECK-LABEL: test_x86_avx2_psrl_q:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    vpsrlq %xmm1, %ymm0, %ymm0
+; CHECK-NEXT:    retl
   %res = call <4 x i64> @llvm.x86.avx2.psrl.q(<4 x i64> %a0, <2 x i64> %a1) ; <<4 x i64>> [#uses=1]
   ret <4 x i64> %res
 }
@@ -249,7 +344,10 @@ declare <4 x i64> @llvm.x86.avx2.psrl.q(<4 x i64>, <2 x i64>) nounwind readnone
 
 
 define <16 x i16> @test_x86_avx2_psrl_w(<16 x i16> %a0, <8 x i16> %a1) {
-  ; CHECK: vpsrlw
+; CHECK-LABEL: test_x86_avx2_psrl_w:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    vpsrlw %xmm1, %ymm0, %ymm0
+; CHECK-NEXT:    retl
   %res = call <16 x i16> @llvm.x86.avx2.psrl.w(<16 x i16> %a0, <8 x i16> %a1) ; <<16 x i16>> [#uses=1]
   ret <16 x i16> %res
 }
@@ -257,7 +355,10 @@ declare <16 x i16> @llvm.x86.avx2.psrl.w(<16 x i16>, <8 x i16>) nounwind readnon
 
 
 define <8 x i32> @test_x86_avx2_psrli_d(<8 x i32> %a0) {
-  ; CHECK: vpsrld
+; CHECK-LABEL: test_x86_avx2_psrli_d:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    vpsrld $7, %ymm0, %ymm0
+; CHECK-NEXT:    retl
   %res = call <8 x i32> @llvm.x86.avx2.psrli.d(<8 x i32> %a0, i32 7) ; <<8 x i32>> [#uses=1]
   ret <8 x i32> %res
 }
@@ -265,7 +366,10 @@ declare <8 x i32> @llvm.x86.avx2.psrli.d(<8 x i32>, i32) nounwind readnone
 
 
 define <4 x i64> @test_x86_avx2_psrli_q(<4 x i64> %a0) {
-  ; CHECK: vpsrlq
+; CHECK-LABEL: test_x86_avx2_psrli_q:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    vpsrlq $7, %ymm0, %ymm0
+; CHECK-NEXT:    retl
   %res = call <4 x i64> @llvm.x86.avx2.psrli.q(<4 x i64> %a0, i32 7) ; <<4 x i64>> [#uses=1]
   ret <4 x i64> %res
 }
@@ -273,7 +377,10 @@ declare <4 x i64> @llvm.x86.avx2.psrli.q(<4 x i64>, i32) nounwind readnone
 
 
 define <16 x i16> @test_x86_avx2_psrli_w(<16 x i16> %a0) {
-  ; CHECK: vpsrlw
+; CHECK-LABEL: test_x86_avx2_psrli_w:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    vpsrlw $7, %ymm0, %ymm0
+; CHECK-NEXT:    retl
   %res = call <16 x i16> @llvm.x86.avx2.psrli.w(<16 x i16> %a0, i32 7) ; <<16 x i16>> [#uses=1]
   ret <16 x i16> %res
 }
@@ -281,7 +388,10 @@ declare <16 x i16> @llvm.x86.avx2.psrli.w(<16 x i16>, i32) nounwind readnone
 
 
 define <32 x i8> @test_x86_avx2_psubs_b(<32 x i8> %a0, <32 x i8> %a1) {
-  ; CHECK: vpsubsb
+; CHECK-LABEL: test_x86_avx2_psubs_b:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    vpsubsb %ymm1, %ymm0, %ymm0
+; CHECK-NEXT:    retl
   %res = call <32 x i8> @llvm.x86.avx2.psubs.b(<32 x i8> %a0, <32 x i8> %a1) ; <<32 x i8>> [#uses=1]
   ret <32 x i8> %res
 }
@@ -289,7 +399,10 @@ declare <32 x i8> @llvm.x86.avx2.psubs.b(<32 x i8>, <32 x i8>) nounwind readnone
 
 
 define <16 x i16> @test_x86_avx2_psubs_w(<16 x i16> %a0, <16 x i16> %a1) {
-  ; CHECK: vpsubsw
+; CHECK-LABEL: test_x86_avx2_psubs_w:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    vpsubsw %ymm1, %ymm0, %ymm0
+; CHECK-NEXT:    retl
   %res = call <16 x i16> @llvm.x86.avx2.psubs.w(<16 x i16> %a0, <16 x i16> %a1) ; <<16 x i16>> [#uses=1]
   ret <16 x i16> %res
 }
@@ -297,7 +410,10 @@ declare <16 x i16> @llvm.x86.avx2.psubs.w(<16 x i16>, <16 x i16>) nounwind readn
 
 
 define <32 x i8> @test_x86_avx2_psubus_b(<32 x i8> %a0, <32 x i8> %a1) {
-  ; CHECK: vpsubusb
+; CHECK-LABEL: test_x86_avx2_psubus_b:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    vpsubusb %ymm1, %ymm0, %ymm0
+; CHECK-NEXT:    retl
   %res = call <32 x i8> @llvm.x86.avx2.psubus.b(<32 x i8> %a0, <32 x i8> %a1) ; <<32 x i8>> [#uses=1]
   ret <32 x i8> %res
 }
@@ -305,7 +421,10 @@ declare <32 x i8> @llvm.x86.avx2.psubus.b(<32 x i8>, <32 x i8>) nounwind readnon
 
 
 define <16 x i16> @test_x86_avx2_psubus_w(<16 x i16> %a0, <16 x i16> %a1) {
-  ; CHECK: vpsubusw
+; CHECK-LABEL: test_x86_avx2_psubus_w:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    vpsubusw %ymm1, %ymm0, %ymm0
+; CHECK-NEXT:    retl
   %res = call <16 x i16> @llvm.x86.avx2.psubus.w(<16 x i16> %a0, <16 x i16> %a1) ; <<16 x i16>> [#uses=1]
   ret <16 x i16> %res
 }
@@ -313,7 +432,10 @@ declare <16 x i16> @llvm.x86.avx2.psubus.w(<16 x i16>, <16 x i16>) nounwind read
 
 
 define <32 x i8> @test_x86_avx2_pabs_b(<32 x i8> %a0) {
-  ; CHECK: vpabsb
+; CHECK-LABEL: test_x86_avx2_pabs_b:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    vpabsb %ymm0, %ymm0
+; CHECK-NEXT:    retl
   %res = call <32 x i8> @llvm.x86.avx2.pabs.b(<32 x i8> %a0) ; <<32 x i8>> [#uses=1]
   ret <32 x i8> %res
 }
@@ -321,7 +443,10 @@ declare <32 x i8> @llvm.x86.avx2.pabs.b(<32 x i8>) nounwind readnone
 
 
 define <8 x i32> @test_x86_avx2_pabs_d(<8 x i32> %a0) {
-  ; CHECK: vpabsd
+; CHECK-LABEL: test_x86_avx2_pabs_d:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    vpabsd %ymm0, %ymm0
+; CHECK-NEXT:    retl
   %res = call <8 x i32> @llvm.x86.avx2.pabs.d(<8 x i32> %a0) ; <<8 x i32>> [#uses=1]
   ret <8 x i32> %res
 }
@@ -329,7 +454,10 @@ declare <8 x i32> @llvm.x86.avx2.pabs.d(<8 x i32>) nounwind readnone
 
 
 define <16 x i16> @test_x86_avx2_pabs_w(<16 x i16> %a0) {
-  ; CHECK: vpabsw
+; CHECK-LABEL: test_x86_avx2_pabs_w:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    vpabsw %ymm0, %ymm0
+; CHECK-NEXT:    retl
   %res = call <16 x i16> @llvm.x86.avx2.pabs.w(<16 x i16> %a0) ; <<16 x i16>> [#uses=1]
   ret <16 x i16> %res
 }
@@ -337,7 +465,10 @@ declare <16 x i16> @llvm.x86.avx2.pabs.w(<16 x i16>) nounwind readnone
 
 
 define <8 x i32> @test_x86_avx2_phadd_d(<8 x i32> %a0, <8 x i32> %a1) {
-  ; CHECK: vphaddd
+; CHECK-LABEL: test_x86_avx2_phadd_d:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    vphaddd %ymm1, %ymm0, %ymm0
+; CHECK-NEXT:    retl
   %res = call <8 x i32> @llvm.x86.avx2.phadd.d(<8 x i32> %a0, <8 x i32> %a1) ; <<8 x i32>> [#uses=1]
   ret <8 x i32> %res
 }
@@ -345,7 +476,10 @@ declare <8 x i32> @llvm.x86.avx2.phadd.d(<8 x i32>, <8 x i32>) nounwind readnone
 
 
 define <16 x i16> @test_x86_avx2_phadd_sw(<16 x i16> %a0, <16 x i16> %a1) {
-  ; CHECK: vphaddsw
+; CHECK-LABEL: test_x86_avx2_phadd_sw:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    vphaddsw %ymm1, %ymm0, %ymm0
+; CHECK-NEXT:    retl
   %res = call <16 x i16> @llvm.x86.avx2.phadd.sw(<16 x i16> %a0, <16 x i16> %a1) ; <<16 x i16>> [#uses=1]
   ret <16 x i16> %res
 }
@@ -353,7 +487,10 @@ declare <16 x i16> @llvm.x86.avx2.phadd.sw(<16 x i16>, <16 x i16>) nounwind read
 
 
 define <16 x i16> @test_x86_avx2_phadd_w(<16 x i16> %a0, <16 x i16> %a1) {
-  ; CHECK: vphaddw
+; CHECK-LABEL: test_x86_avx2_phadd_w:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    vphaddw %ymm1, %ymm0, %ymm0
+; CHECK-NEXT:    retl
   %res = call <16 x i16> @llvm.x86.avx2.phadd.w(<16 x i16> %a0, <16 x i16> %a1) ; <<16 x i16>> [#uses=1]
   ret <16 x i16> %res
 }
@@ -361,7 +498,10 @@ declare <16 x i16> @llvm.x86.avx2.phadd.w(<16 x i16>, <16 x i16>) nounwind readn
 
 
 define <8 x i32> @test_x86_avx2_phsub_d(<8 x i32> %a0, <8 x i32> %a1) {
-  ; CHECK: vphsubd
+; CHECK-LABEL: test_x86_avx2_phsub_d:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    vphsubd %ymm1, %ymm0, %ymm0
+; CHECK-NEXT:    retl
   %res = call <8 x i32> @llvm.x86.avx2.phsub.d(<8 x i32> %a0, <8 x i32> %a1) ; <<8 x i32>> [#uses=1]
   ret <8 x i32> %res
 }
@@ -369,7 +509,10 @@ declare <8 x i32> @llvm.x86.avx2.phsub.d(<8 x i32>, <8 x i32>) nounwind readnone
 
 
 define <16 x i16> @test_x86_avx2_phsub_sw(<16 x i16> %a0, <16 x i16> %a1) {
-  ; CHECK: vphsubsw
+; CHECK-LABEL: test_x86_avx2_phsub_sw:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    vphsubsw %ymm1, %ymm0, %ymm0
+; CHECK-NEXT:    retl
   %res = call <16 x i16> @llvm.x86.avx2.phsub.sw(<16 x i16> %a0, <16 x i16> %a1) ; <<16 x i16>> [#uses=1]
   ret <16 x i16> %res
 }
@@ -377,7 +520,10 @@ declare <16 x i16> @llvm.x86.avx2.phsub.sw(<16 x i16>, <16 x i16>) nounwind read
 
 
 define <16 x i16> @test_x86_avx2_phsub_w(<16 x i16> %a0, <16 x i16> %a1) {
-  ; CHECK: vphsubw
+; CHECK-LABEL: test_x86_avx2_phsub_w:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    vphsubw %ymm1, %ymm0, %ymm0
+; CHECK-NEXT:    retl
   %res = call <16 x i16> @llvm.x86.avx2.phsub.w(<16 x i16> %a0, <16 x i16> %a1) ; <<16 x i16>> [#uses=1]
   ret <16 x i16> %res
 }
@@ -385,7 +531,10 @@ declare <16 x i16> @llvm.x86.avx2.phsub.w(<16 x i16>, <16 x i16>) nounwind readn
 
 
 define <16 x i16> @test_x86_avx2_pmadd_ub_sw(<32 x i8> %a0, <32 x i8> %a1) {
-  ; CHECK: vpmaddubsw
+; CHECK-LABEL: test_x86_avx2_pmadd_ub_sw:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    vpmaddubsw %ymm1, %ymm0, %ymm0
+; CHECK-NEXT:    retl
   %res = call <16 x i16> @llvm.x86.avx2.pmadd.ub.sw(<32 x i8> %a0, <32 x i8> %a1) ; <<16 x i16>> [#uses=1]
   ret <16 x i16> %res
 }
@@ -393,7 +542,10 @@ declare <16 x i16> @llvm.x86.avx2.pmadd.ub.sw(<32 x i8>, <32 x i8>) nounwind rea
 
 
 define <16 x i16> @test_x86_avx2_pmul_hr_sw(<16 x i16> %a0, <16 x i16> %a1) {
-  ; CHECK: vpmulhrsw
+; CHECK-LABEL: test_x86_avx2_pmul_hr_sw:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    vpmulhrsw %ymm1, %ymm0, %ymm0
+; CHECK-NEXT:    retl
   %res = call <16 x i16> @llvm.x86.avx2.pmul.hr.sw(<16 x i16> %a0, <16 x i16> %a1) ; <<16 x i16>> [#uses=1]
   ret <16 x i16> %res
 }
@@ -401,7 +553,10 @@ declare <16 x i16> @llvm.x86.avx2.pmul.hr.sw(<16 x i16>, <16 x i16>) nounwind re
 
 
 define <32 x i8> @test_x86_avx2_pshuf_b(<32 x i8> %a0, <32 x i8> %a1) {
-  ; CHECK: vpshufb
+; CHECK-LABEL: test_x86_avx2_pshuf_b:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    vpshufb %ymm1, %ymm0, %ymm0
+; CHECK-NEXT:    retl
   %res = call <32 x i8> @llvm.x86.avx2.pshuf.b(<32 x i8> %a0, <32 x i8> %a1) ; <<16 x i8>> [#uses=1]
   ret <32 x i8> %res
 }
@@ -409,7 +564,10 @@ declare <32 x i8> @llvm.x86.avx2.pshuf.b(<32 x i8>, <32 x i8>) nounwind readnone
 
 
 define <32 x i8> @test_x86_avx2_psign_b(<32 x i8> %a0, <32 x i8> %a1) {
-  ; CHECK: vpsignb
+; CHECK-LABEL: test_x86_avx2_psign_b:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    vpsignb %ymm1, %ymm0, %ymm0
+; CHECK-NEXT:    retl
   %res = call <32 x i8> @llvm.x86.avx2.psign.b(<32 x i8> %a0, <32 x i8> %a1) ; <<32 x i8>> [#uses=1]
   ret <32 x i8> %res
 }
@@ -417,7 +575,10 @@ declare <32 x i8> @llvm.x86.avx2.psign.b(<32 x i8>, <32 x i8>) nounwind readnone
 
 
 define <8 x i32> @test_x86_avx2_psign_d(<8 x i32> %a0, <8 x i32> %a1) {
-  ; CHECK: vpsignd
+; CHECK-LABEL: test_x86_avx2_psign_d:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    vpsignd %ymm1, %ymm0, %ymm0
+; CHECK-NEXT:    retl
   %res = call <8 x i32> @llvm.x86.avx2.psign.d(<8 x i32> %a0, <8 x i32> %a1) ; <<4 x i32>> [#uses=1]
   ret <8 x i32> %res
 }
@@ -425,7 +586,10 @@ declare <8 x i32> @llvm.x86.avx2.psign.d(<8 x i32>, <8 x i32>) nounwind readnone
 
 
 define <16 x i16> @test_x86_avx2_psign_w(<16 x i16> %a0, <16 x i16> %a1) {
-  ; CHECK: vpsignw
+; CHECK-LABEL: test_x86_avx2_psign_w:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    vpsignw %ymm1, %ymm0, %ymm0
+; CHECK-NEXT:    retl
   %res = call <16 x i16> @llvm.x86.avx2.psign.w(<16 x i16> %a0, <16 x i16> %a1) ; <<16 x i16>> [#uses=1]
   ret <16 x i16> %res
 }
@@ -433,8 +597,11 @@ declare <16 x i16> @llvm.x86.avx2.psign.w(<16 x i16>, <16 x i16>) nounwind readn
 
 
 define <4 x i64> @test_x86_avx2_movntdqa(i8* %a0) {
-  ; CHECK: movl
-  ; CHECK: vmovntdqa
+; CHECK-LABEL: test_x86_avx2_movntdqa:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    movl {{[0-9]+}}(%esp), %eax
+; CHECK-NEXT:    vmovntdqa (%eax), %ymm0
+; CHECK-NEXT:    retl
   %res = call <4 x i64> @llvm.x86.avx2.movntdqa(i8* %a0) ; <<4 x i64>> [#uses=1]
   ret <4 x i64> %res
 }
@@ -442,7 +609,10 @@ declare <4 x i64> @llvm.x86.avx2.movntdqa(i8*) nounwind readonly
 
 
 define <16 x i16> @test_x86_avx2_mpsadbw(<32 x i8> %a0, <32 x i8> %a1) {
-  ; CHECK: vmpsadbw
+; CHECK-LABEL: test_x86_avx2_mpsadbw:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    vmpsadbw $7, %ymm1, %ymm0, %ymm0
+; CHECK-NEXT:    retl
   %res = call <16 x i16> @llvm.x86.avx2.mpsadbw(<32 x i8> %a0, <32 x i8> %a1, i8 7) ; <<16 x i16>> [#uses=1]
   ret <16 x i16> %res
 }
@@ -450,7 +620,10 @@ declare <16 x i16> @llvm.x86.avx2.mpsadbw(<32 x i8>, <32 x i8>, i8) nounwind rea
 
 
 define <16 x i16> @test_x86_avx2_packusdw(<8 x i32> %a0, <8 x i32> %a1) {
-  ; CHECK: vpackusdw
+; CHECK-LABEL: test_x86_avx2_packusdw:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    vpackusdw %ymm1, %ymm0, %ymm0
+; CHECK-NEXT:    retl
   %res = call <16 x i16> @llvm.x86.avx2.packusdw(<8 x i32> %a0, <8 x i32> %a1) ; <<16 x i16>> [#uses=1]
   ret <16 x i16> %res
 }
@@ -458,7 +631,10 @@ declare <16 x i16> @llvm.x86.avx2.packusdw(<8 x i32>, <8 x i32>) nounwind readno
 
 
 define <32 x i8> @test_x86_avx2_pblendvb(<32 x i8> %a0, <32 x i8> %a1, <32 x i8> %a2) {
-  ; CHECK: vpblendvb
+; CHECK-LABEL: test_x86_avx2_pblendvb:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    vpblendvb %ymm2, %ymm1, %ymm0, %ymm0
+; CHECK-NEXT:    retl
   %res = call <32 x i8> @llvm.x86.avx2.pblendvb(<32 x i8> %a0, <32 x i8> %a1, <32 x i8> %a2) ; <<32 x i8>> [#uses=1]
   ret <32 x i8> %res
 }
@@ -466,7 +642,10 @@ declare <32 x i8> @llvm.x86.avx2.pblendvb(<32 x i8>, <32 x i8>, <32 x i8>) nounw
 
 
 define <16 x i16> @test_x86_avx2_pblendw(<16 x i16> %a0, <16 x i16> %a1) {
-  ; CHECK: vpblendw
+; CHECK-LABEL: test_x86_avx2_pblendw:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    vpblendw {{.*#+}} ymm0 = ymm1[0,1,2],ymm0[3,4,5,6,7],ymm1[8,9,10],ymm0[11,12,13,14,15]
+; CHECK-NEXT:    retl
   %res = call <16 x i16> @llvm.x86.avx2.pblendw(<16 x i16> %a0, <16 x i16> %a1, i8 7) ; <<16 x i16>> [#uses=1]
   ret <16 x i16> %res
 }
@@ -474,7 +653,10 @@ declare <16 x i16> @llvm.x86.avx2.pblendw(<16 x i16>, <16 x i16>, i8) nounwind r
 
 
 define <32 x i8> @test_x86_avx2_pmaxsb(<32 x i8> %a0, <32 x i8> %a1) {
-  ; CHECK: vpmaxsb
+; CHECK-LABEL: test_x86_avx2_pmaxsb:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    vpmaxsb %ymm1, %ymm0, %ymm0
+; CHECK-NEXT:    retl
   %res = call <32 x i8> @llvm.x86.avx2.pmaxs.b(<32 x i8> %a0, <32 x i8> %a1) ; <<32 x i8>> [#uses=1]
   ret <32 x i8> %res
 }
@@ -482,7 +664,10 @@ declare <32 x i8> @llvm.x86.avx2.pmaxs.b(<32 x i8>, <32 x i8>) nounwind readnone
 
 
 define <8 x i32> @test_x86_avx2_pmaxsd(<8 x i32> %a0, <8 x i32> %a1) {
-  ; CHECK: vpmaxsd
+; CHECK-LABEL: test_x86_avx2_pmaxsd:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    vpmaxsd %ymm1, %ymm0, %ymm0
+; CHECK-NEXT:    retl
   %res = call <8 x i32> @llvm.x86.avx2.pmaxs.d(<8 x i32> %a0, <8 x i32> %a1) ; <<8 x i32>> [#uses=1]
   ret <8 x i32> %res
 }
@@ -490,7 +675,10 @@ declare <8 x i32> @llvm.x86.avx2.pmaxs.d(<8 x i32>, <8 x i32>) nounwind readnone
 
 
 define <8 x i32> @test_x86_avx2_pmaxud(<8 x i32> %a0, <8 x i32> %a1) {
-  ; CHECK: vpmaxud
+; CHECK-LABEL: test_x86_avx2_pmaxud:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    vpmaxud %ymm1, %ymm0, %ymm0
+; CHECK-NEXT:    retl
   %res = call <8 x i32> @llvm.x86.avx2.pmaxu.d(<8 x i32> %a0, <8 x i32> %a1) ; <<8 x i32>> [#uses=1]
   ret <8 x i32> %res
 }
@@ -498,7 +686,10 @@ declare <8 x i32> @llvm.x86.avx2.pmaxu.d(<8 x i32>, <8 x i32>) nounwind readnone
 
 
 define <16 x i16> @test_x86_avx2_pmaxuw(<16 x i16> %a0, <16 x i16> %a1) {
-  ; CHECK: vpmaxuw
+; CHECK-LABEL: test_x86_avx2_pmaxuw:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    vpmaxuw %ymm1, %ymm0, %ymm0
+; CHECK-NEXT:    retl
   %res = call <16 x i16> @llvm.x86.avx2.pmaxu.w(<16 x i16> %a0, <16 x i16> %a1) ; <<16 x i16>> [#uses=1]
   ret <16 x i16> %res
 }
@@ -506,7 +697,10 @@ declare <16 x i16> @llvm.x86.avx2.pmaxu.w(<16 x i16>, <16 x i16>) nounwind readn
 
 
 define <32 x i8> @test_x86_avx2_pminsb(<32 x i8> %a0, <32 x i8> %a1) {
-  ; CHECK: vpminsb
+; CHECK-LABEL: test_x86_avx2_pminsb:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    vpminsb %ymm1, %ymm0, %ymm0
+; CHECK-NEXT:    retl
   %res = call <32 x i8> @llvm.x86.avx2.pmins.b(<32 x i8> %a0, <32 x i8> %a1) ; <<32 x i8>> [#uses=1]
   ret <32 x i8> %res
 }
@@ -514,7 +708,10 @@ declare <32 x i8> @llvm.x86.avx2.pmins.b(<32 x i8>, <32 x i8>) nounwind readnone
 
 
 define <8 x i32> @test_x86_avx2_pminsd(<8 x i32> %a0, <8 x i32> %a1) {
-  ; CHECK: vpminsd
+; CHECK-LABEL: test_x86_avx2_pminsd:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    vpminsd %ymm1, %ymm0, %ymm0
+; CHECK-NEXT:    retl
   %res = call <8 x i32> @llvm.x86.avx2.pmins.d(<8 x i32> %a0, <8 x i32> %a1) ; <<8 x i32>> [#uses=1]
   ret <8 x i32> %res
 }
@@ -522,7 +719,10 @@ declare <8 x i32> @llvm.x86.avx2.pmins.d(<8 x i32>, <8 x i32>) nounwind readnone
 
 
 define <8 x i32> @test_x86_avx2_pminud(<8 x i32> %a0, <8 x i32> %a1) {
-  ; CHECK: vpminud
+; CHECK-LABEL: test_x86_avx2_pminud:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    vpminud %ymm1, %ymm0, %ymm0
+; CHECK-NEXT:    retl
   %res = call <8 x i32> @llvm.x86.avx2.pminu.d(<8 x i32> %a0, <8 x i32> %a1) ; <<8 x i32>> [#uses=1]
   ret <8 x i32> %res
 }
@@ -530,7 +730,10 @@ declare <8 x i32> @llvm.x86.avx2.pminu.d(<8 x i32>, <8 x i32>) nounwind readnone
 
 
 define <16 x i16> @test_x86_avx2_pminuw(<16 x i16> %a0, <16 x i16> %a1) {
-  ; CHECK: vpminuw
+; CHECK-LABEL: test_x86_avx2_pminuw:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    vpminuw %ymm1, %ymm0, %ymm0
+; CHECK-NEXT:    retl
   %res = call <16 x i16> @llvm.x86.avx2.pminu.w(<16 x i16> %a0, <16 x i16> %a1) ; <<16 x i16>> [#uses=1]
   ret <16 x i16> %res
 }
@@ -538,7 +741,10 @@ declare <16 x i16> @llvm.x86.avx2.pminu.w(<16 x i16>, <16 x i16>) nounwind readn
 
 
 define <8 x i32> @test_x86_avx2_pmovsxbd(<16 x i8> %a0) {
-  ; CHECK: vpmovsxbd
+; CHECK-LABEL: test_x86_avx2_pmovsxbd:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    vpmovsxbd %xmm0, %ymm0
+; CHECK-NEXT:    retl
   %res = call <8 x i32> @llvm.x86.avx2.pmovsxbd(<16 x i8> %a0) ; <<8 x i32>> [#uses=1]
   ret <8 x i32> %res
 }
@@ -546,7 +752,10 @@ declare <8 x i32> @llvm.x86.avx2.pmovsxbd(<16 x i8>) nounwind readnone
 
 
 define <4 x i64> @test_x86_avx2_pmovsxbq(<16 x i8> %a0) {
-  ; CHECK: vpmovsxbq
+; CHECK-LABEL: test_x86_avx2_pmovsxbq:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    vpmovsxbq %xmm0, %ymm0
+; CHECK-NEXT:    retl
   %res = call <4 x i64> @llvm.x86.avx2.pmovsxbq(<16 x i8> %a0) ; <<4 x i64>> [#uses=1]
   ret <4 x i64> %res
 }
@@ -554,7 +763,10 @@ declare <4 x i64> @llvm.x86.avx2.pmovsxbq(<16 x i8>) nounwind readnone
 
 
 define <16 x i16> @test_x86_avx2_pmovsxbw(<16 x i8> %a0) {
-  ; CHECK: vpmovsxbw
+; CHECK-LABEL: test_x86_avx2_pmovsxbw:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    vpmovsxbw %xmm0, %ymm0
+; CHECK-NEXT:    retl
   %res = call <16 x i16> @llvm.x86.avx2.pmovsxbw(<16 x i8> %a0) ; <<8 x i16>> [#uses=1]
   ret <16 x i16> %res
 }
@@ -562,7 +774,10 @@ declare <16 x i16> @llvm.x86.avx2.pmovsxbw(<16 x i8>) nounwind readnone
 
 
 define <4 x i64> @test_x86_avx2_pmovsxdq(<4 x i32> %a0) {
-  ; CHECK: vpmovsxdq
+; CHECK-LABEL: test_x86_avx2_pmovsxdq:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    vpmovsxdq %xmm0, %ymm0
+; CHECK-NEXT:    retl
   %res = call <4 x i64> @llvm.x86.avx2.pmovsxdq(<4 x i32> %a0) ; <<4 x i64>> [#uses=1]
   ret <4 x i64> %res
 }
@@ -570,7 +785,10 @@ declare <4 x i64> @llvm.x86.avx2.pmovsxdq(<4 x i32>) nounwind readnone
 
 
 define <8 x i32> @test_x86_avx2_pmovsxwd(<8 x i16> %a0) {
-  ; CHECK: vpmovsxwd
+; CHECK-LABEL: test_x86_avx2_pmovsxwd:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    vpmovsxwd %xmm0, %ymm0
+; CHECK-NEXT:    retl
   %res = call <8 x i32> @llvm.x86.avx2.pmovsxwd(<8 x i16> %a0) ; <<8 x i32>> [#uses=1]
   ret <8 x i32> %res
 }
@@ -578,7 +796,10 @@ declare <8 x i32> @llvm.x86.avx2.pmovsxwd(<8 x i16>) nounwind readnone
 
 
 define <4 x i64> @test_x86_avx2_pmovsxwq(<8 x i16> %a0) {
-  ; CHECK: vpmovsxwq
+; CHECK-LABEL: test_x86_avx2_pmovsxwq:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    vpmovsxwq %xmm0, %ymm0
+; CHECK-NEXT:    retl
   %res = call <4 x i64> @llvm.x86.avx2.pmovsxwq(<8 x i16> %a0) ; <<4 x i64>> [#uses=1]
   ret <4 x i64> %res
 }
@@ -586,7 +807,10 @@ declare <4 x i64> @llvm.x86.avx2.pmovsxwq(<8 x i16>) nounwind readnone
 
 
 define <8 x i32> @test_x86_avx2_pmovzxbd(<16 x i8> %a0) {
-  ; CHECK: vpmovzxbd
+; CHECK-LABEL: test_x86_avx2_pmovzxbd:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    vpmovzxbd {{.*#+}} ymm0 = xmm0[0],zero,zero,zero,xmm0[1],zero,zero,zero,xmm0[2],zero,zero,zero,xmm0[3],zero,zero,zero,xmm0[4],zero,zero,zero,xmm0[5],zero,zero,zero,xmm0[6],zero,zero,zero,xmm0[7],zero,zero,zero
+; CHECK-NEXT:    retl
   %res = call <8 x i32> @llvm.x86.avx2.pmovzxbd(<16 x i8> %a0) ; <<8 x i32>> [#uses=1]
   ret <8 x i32> %res
 }
@@ -594,7 +818,10 @@ declare <8 x i32> @llvm.x86.avx2.pmovzxbd(<16 x i8>) nounwind readnone
 
 
 define <4 x i64> @test_x86_avx2_pmovzxbq(<16 x i8> %a0) {
-  ; CHECK: vpmovzxbq
+; CHECK-LABEL: test_x86_avx2_pmovzxbq:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    vpmovzxbq {{.*#+}} ymm0 = xmm0[0],zero,zero,zero,zero,zero,zero,zero,xmm0[1],zero,zero,zero,zero,zero,zero,zero,xmm0[2],zero,zero,zero,zero,zero,zero,zero,xmm0[3],zero,zero,zero,zero,zero,zero,zero
+; CHECK-NEXT:    retl
   %res = call <4 x i64> @llvm.x86.avx2.pmovzxbq(<16 x i8> %a0) ; <<4 x i64>> [#uses=1]
   ret <4 x i64> %res
 }
@@ -602,7 +829,10 @@ declare <4 x i64> @llvm.x86.avx2.pmovzxbq(<16 x i8>) nounwind readnone
 
 
 define <16 x i16> @test_x86_avx2_pmovzxbw(<16 x i8> %a0) {
-  ; CHECK: vpmovzxbw
+; CHECK-LABEL: test_x86_avx2_pmovzxbw:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    vpmovzxbw {{.*#+}} ymm0 = xmm0[0],zero,xmm0[1],zero,xmm0[2],zero,xmm0[3],zero,xmm0[4],zero,xmm0[5],zero,xmm0[6],zero,xmm0[7],zero,xmm0[8],zero,xmm0[9],zero,xmm0[10],zero,xmm0[11],zero,xmm0[12],zero,xmm0[13],zero,xmm0[14],zero,xmm0[15],zero
+; CHECK-NEXT:    retl
   %res = call <16 x i16> @llvm.x86.avx2.pmovzxbw(<16 x i8> %a0) ; <<16 x i16>> [#uses=1]
   ret <16 x i16> %res
 }
@@ -610,7 +840,10 @@ declare <16 x i16> @llvm.x86.avx2.pmovzxbw(<16 x i8>) nounwind readnone
 
 
 define <4 x i64> @test_x86_avx2_pmovzxdq(<4 x i32> %a0) {
-  ; CHECK: vpmovzxdq
+; CHECK-LABEL: test_x86_avx2_pmovzxdq:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    vpmovzxdq {{.*#+}} ymm0 = xmm0[0],zero,xmm0[1],zero,xmm0[2],zero,xmm0[3],zero
+; CHECK-NEXT:    retl
   %res = call <4 x i64> @llvm.x86.avx2.pmovzxdq(<4 x i32> %a0) ; <<4 x i64>> [#uses=1]
   ret <4 x i64> %res
 }
@@ -618,7 +851,10 @@ declare <4 x i64> @llvm.x86.avx2.pmovzxdq(<4 x i32>) nounwind readnone
 
 
 define <8 x i32> @test_x86_avx2_pmovzxwd(<8 x i16> %a0) {
-  ; CHECK: vpmovzxwd
+; CHECK-LABEL: test_x86_avx2_pmovzxwd:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    vpmovzxwd {{.*#+}} ymm0 = xmm0[0],zero,xmm0[1],zero,xmm0[2],zero,xmm0[3],zero,xmm0[4],zero,xmm0[5],zero,xmm0[6],zero,xmm0[7],zero
+; CHECK-NEXT:    retl
   %res = call <8 x i32> @llvm.x86.avx2.pmovzxwd(<8 x i16> %a0) ; <<8 x i32>> [#uses=1]
   ret <8 x i32> %res
 }
@@ -626,7 +862,10 @@ declare <8 x i32> @llvm.x86.avx2.pmovzxwd(<8 x i16>) nounwind readnone
 
 
 define <4 x i64> @test_x86_avx2_pmovzxwq(<8 x i16> %a0) {
-  ; CHECK: vpmovzxwq
+; CHECK-LABEL: test_x86_avx2_pmovzxwq:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    vpmovzxwq {{.*#+}} ymm0 = xmm0[0],zero,zero,zero,xmm0[1],zero,zero,zero,xmm0[2],zero,zero,zero,xmm0[3],zero,zero,zero
+; CHECK-NEXT:    retl
   %res = call <4 x i64> @llvm.x86.avx2.pmovzxwq(<8 x i16> %a0) ; <<4 x i64>> [#uses=1]
   ret <4 x i64> %res
 }
@@ -642,7 +881,10 @@ declare <4 x i64> @llvm.x86.avx2.pmul.dq(<8 x i32>, <8 x i32>) nounwind readnone
 
 
 define <4 x i32> @test_x86_avx2_pblendd_128(<4 x i32> %a0, <4 x i32> %a1) {
-  ; CHECK: vpblendd
+; CHECK-LABEL: test_x86_avx2_pblendd_128:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    vpblendd {{.*#+}} xmm0 = xmm1[0,1,2],xmm0[3]
+; CHECK-NEXT:    retl
   %res = call <4 x i32> @llvm.x86.avx2.pblendd.128(<4 x i32> %a0, <4 x i32> %a1, i8 7) ; <<4 x i32>> [#uses=1]
   ret <4 x i32> %res
 }
@@ -650,29 +892,38 @@ declare <4 x i32> @llvm.x86.avx2.pblendd.128(<4 x i32>, <4 x i32>, i8) nounwind
 
 
 define <8 x i32> @test_x86_avx2_pblendd_256(<8 x i32> %a0, <8 x i32> %a1) {
-  ; CHECK: vpblendd
+; CHECK-LABEL: test_x86_avx2_pblendd_256:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    vpblendd {{.*#+}} ymm0 = ymm1[0,1,2],ymm0[3,4,5,6,7]
+; CHECK-NEXT:    retl
   %res = call <8 x i32> @llvm.x86.avx2.pblendd.256(<8 x i32> %a0, <8 x i32> %a1, i8 7) ; <<8 x i32>> [#uses=1]
   ret <8 x i32> %res
 }
 declare <8 x i32> @llvm.x86.avx2.pblendd.256(<8 x i32>, <8 x i32>, i8) nounwind readnone
 
 
+; Check that the arguments are swapped between the intrinsic definition
+; and its lowering. Indeed, the offsets are the first source in
+; the instruction.
 define <8 x i32> @test_x86_avx2_permd(<8 x i32> %a0, <8 x i32> %a1) {
-  ; Check that the arguments are swapped between the intrinsic definition
-  ; and its lowering. Indeed, the offsets are the first source in
-  ; the instruction.
-  ; CHECK: vpermd %ymm0, %ymm1, %ymm0
+; CHECK-LABEL: test_x86_avx2_permd:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    vpermd %ymm0, %ymm1, %ymm0
+; CHECK-NEXT:    retl
   %res = call <8 x i32> @llvm.x86.avx2.permd(<8 x i32> %a0, <8 x i32> %a1) ; <<8 x i32>> [#uses=1]
   ret <8 x i32> %res
 }
 declare <8 x i32> @llvm.x86.avx2.permd(<8 x i32>, <8 x i32>) nounwind readonly
 
 
+; Check that the arguments are swapped between the intrinsic definition
+; and its lowering. Indeed, the offsets are the first source in
+; the instruction.
 define <8 x float> @test_x86_avx2_permps(<8 x float> %a0, <8 x i32> %a1) {
-  ; Check that the arguments are swapped between the intrinsic definition
-  ; and its lowering. Indeed, the offsets are the first source in
-  ; the instruction.
-  ; CHECK: vpermps %ymm0, %ymm1, %ymm0
+; CHECK-LABEL: test_x86_avx2_permps:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    vpermps %ymm0, %ymm1, %ymm0
+; CHECK-NEXT:    retl
   %res = call <8 x float> @llvm.x86.avx2.permps(<8 x float> %a0, <8 x i32> %a1) ; <<8 x float>> [#uses=1]
   ret <8 x float> %res
 }
@@ -680,7 +931,10 @@ declare <8 x float> @llvm.x86.avx2.permps(<8 x float>, <8 x i32>) nounwind reado
 
 
 define <4 x i64> @test_x86_avx2_vperm2i128(<4 x i64> %a0, <4 x i64> %a1) {
-  ; CHECK: vperm2i128
+; CHECK-LABEL: test_x86_avx2_vperm2i128:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    vperm2f128 {{.*#+}} ymm0 = ymm0[2,3,0,1]
+; CHECK-NEXT:    retl
   %res = call <4 x i64> @llvm.x86.avx2.vperm2i128(<4 x i64> %a0, <4 x i64> %a1, i8 1) ; <<4 x i64>> [#uses=1]
   ret <4 x i64> %res
 }
@@ -688,7 +942,11 @@ declare <4 x i64> @llvm.x86.avx2.vperm2i128(<4 x i64>, <4 x i64>, i8) nounwind r
 
 
 define <2 x i64> @test_x86_avx2_maskload_q(i8* %a0, <2 x i64> %a1) {
-  ; CHECK: vpmaskmovq
+; CHECK-LABEL: test_x86_avx2_maskload_q:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    movl {{[0-9]+}}(%esp), %eax
+; CHECK-NEXT:    vpmaskmovq (%eax), %xmm0, %xmm0
+; CHECK-NEXT:    retl
   %res = call <2 x i64> @llvm.x86.avx2.maskload.q(i8* %a0, <2 x i64> %a1) ; <<2 x i64>> [#uses=1]
   ret <2 x i64> %res
 }
@@ -696,7 +954,11 @@ declare <2 x i64> @llvm.x86.avx2.maskload.q(i8*, <2 x i64>) nounwind readonly
 
 
 define <4 x i64> @test_x86_avx2_maskload_q_256(i8* %a0, <4 x i64> %a1) {
-  ; CHECK: vpmaskmovq
+; CHECK-LABEL: test_x86_avx2_maskload_q_256:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    movl {{[0-9]+}}(%esp), %eax
+; CHECK-NEXT:    vpmaskmovq (%eax), %ymm0, %ymm0
+; CHECK-NEXT:    retl
   %res = call <4 x i64> @llvm.x86.avx2.maskload.q.256(i8* %a0, <4 x i64> %a1) ; <<4 x i64>> [#uses=1]
   ret <4 x i64> %res
 }
@@ -704,7 +966,11 @@ declare <4 x i64> @llvm.x86.avx2.maskload.q.256(i8*, <4 x i64>) nounwind readonl
 
 
 define <4 x i32> @test_x86_avx2_maskload_d(i8* %a0, <4 x i32> %a1) {
-  ; CHECK: vpmaskmovd
+; CHECK-LABEL: test_x86_avx2_maskload_d:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    movl {{[0-9]+}}(%esp), %eax
+; CHECK-NEXT:    vpmaskmovd (%eax), %xmm0, %xmm0
+; CHECK-NEXT:    retl
   %res = call <4 x i32> @llvm.x86.avx2.maskload.d(i8* %a0, <4 x i32> %a1) ; <<4 x i32>> [#uses=1]
   ret <4 x i32> %res
 }
@@ -712,7 +978,11 @@ declare <4 x i32> @llvm.x86.avx2.maskload.d(i8*, <4 x i32>) nounwind readonly
 
 
 define <8 x i32> @test_x86_avx2_maskload_d_256(i8* %a0, <8 x i32> %a1) {
-  ; CHECK: vpmaskmovd
+; CHECK-LABEL: test_x86_avx2_maskload_d_256:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    movl {{[0-9]+}}(%esp), %eax
+; CHECK-NEXT:    vpmaskmovd (%eax), %ymm0, %ymm0
+; CHECK-NEXT:    retl
   %res = call <8 x i32> @llvm.x86.avx2.maskload.d.256(i8* %a0, <8 x i32> %a1) ; <<8 x i32>> [#uses=1]
   ret <8 x i32> %res
 }
@@ -720,7 +990,11 @@ declare <8 x i32> @llvm.x86.avx2.maskload.d.256(i8*, <8 x i32>) nounwind readonl
 
 
 define void @test_x86_avx2_maskstore_q(i8* %a0, <2 x i64> %a1, <2 x i64> %a2) {
-  ; CHECK: vpmaskmovq
+; CHECK-LABEL: test_x86_avx2_maskstore_q:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    movl {{[0-9]+}}(%esp), %eax
+; CHECK-NEXT:    vpmaskmovq %xmm1, %xmm0, (%eax)
+; CHECK-NEXT:    retl
   call void @llvm.x86.avx2.maskstore.q(i8* %a0, <2 x i64> %a1, <2 x i64> %a2)
   ret void
 }
@@ -728,7 +1002,12 @@ declare void @llvm.x86.avx2.maskstore.q(i8*, <2 x i64>, <2 x i64>) nounwind
 
 
 define void @test_x86_avx2_maskstore_q_256(i8* %a0, <4 x i64> %a1, <4 x i64> %a2) {
-  ; CHECK: vpmaskmovq
+; CHECK-LABEL: test_x86_avx2_maskstore_q_256:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    movl {{[0-9]+}}(%esp), %eax
+; CHECK-NEXT:    vpmaskmovq %ymm1, %ymm0, (%eax)
+; CHECK-NEXT:    vzeroupper
+; CHECK-NEXT:    retl
   call void @llvm.x86.avx2.maskstore.q.256(i8* %a0, <4 x i64> %a1, <4 x i64> %a2)
   ret void
 }
@@ -736,7 +1015,11 @@ declare void @llvm.x86.avx2.maskstore.q.256(i8*, <4 x i64>, <4 x i64>) nounwind
 
 
 define void @test_x86_avx2_maskstore_d(i8* %a0, <4 x i32> %a1, <4 x i32> %a2) {
-  ; CHECK: vpmaskmovd
+; CHECK-LABEL: test_x86_avx2_maskstore_d:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    movl {{[0-9]+}}(%esp), %eax
+; CHECK-NEXT:    vpmaskmovd %xmm1, %xmm0, (%eax)
+; CHECK-NEXT:    retl
   call void @llvm.x86.avx2.maskstore.d(i8* %a0, <4 x i32> %a1, <4 x i32> %a2)
   ret void
 }
@@ -744,7 +1027,12 @@ declare void @llvm.x86.avx2.maskstore.d(i8*, <4 x i32>, <4 x i32>) nounwind
 
 
 define void @test_x86_avx2_maskstore_d_256(i8* %a0, <8 x i32> %a1, <8 x i32> %a2) {
-  ; CHECK: vpmaskmovd
+; CHECK-LABEL: test_x86_avx2_maskstore_d_256:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    movl {{[0-9]+}}(%esp), %eax
+; CHECK-NEXT:    vpmaskmovd %ymm1, %ymm0, (%eax)
+; CHECK-NEXT:    vzeroupper
+; CHECK-NEXT:    retl
   call void @llvm.x86.avx2.maskstore.d.256(i8* %a0, <8 x i32> %a1, <8 x i32> %a2)
   ret void
 }
@@ -752,7 +1040,10 @@ declare void @llvm.x86.avx2.maskstore.d.256(i8*, <8 x i32>, <8 x i32>) nounwind
 
 
 define <4 x i32> @test_x86_avx2_psllv_d(<4 x i32> %a0, <4 x i32> %a1) {
-  ; CHECK: vpsllvd
+; CHECK-LABEL: test_x86_avx2_psllv_d:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    vpsllvd %xmm1, %xmm0, %xmm0
+; CHECK-NEXT:    retl
   %res = call <4 x i32> @llvm.x86.avx2.psllv.d(<4 x i32> %a0, <4 x i32> %a1) ; <<4 x i32>> [#uses=1]
   ret <4 x i32> %res
 }
@@ -760,7 +1051,10 @@ declare <4 x i32> @llvm.x86.avx2.psllv.d(<4 x i32>, <4 x i32>) nounwind readnone
 
 
 define <8 x i32> @test_x86_avx2_psllv_d_256(<8 x i32> %a0, <8 x i32> %a1) {
-  ; CHECK: vpsllvd
+; CHECK-LABEL: test_x86_avx2_psllv_d_256:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    vpsllvd %ymm1, %ymm0, %ymm0
+; CHECK-NEXT:    retl
   %res = call <8 x i32> @llvm.x86.avx2.psllv.d.256(<8 x i32> %a0, <8 x i32> %a1) ; <<8 x i32>> [#uses=1]
   ret <8 x i32> %res
 }
@@ -768,7 +1062,10 @@ declare <8 x i32> @llvm.x86.avx2.psllv.d.256(<8 x i32>, <8 x i32>) nounwind read
 
 
 define <2 x i64> @test_x86_avx2_psllv_q(<2 x i64> %a0, <2 x i64> %a1) {
-  ; CHECK: vpsllvq
+; CHECK-LABEL: test_x86_avx2_psllv_q:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    vpsllvq %xmm1, %xmm0, %xmm0
+; CHECK-NEXT:    retl
   %res = call <2 x i64> @llvm.x86.avx2.psllv.q(<2 x i64> %a0, <2 x i64> %a1) ; <<2 x i64>> [#uses=1]
   ret <2 x i64> %res
 }
@@ -776,7 +1073,10 @@ declare <2 x i64> @llvm.x86.avx2.psllv.q(<2 x i64>, <2 x i64>) nounwind readnone
 
 
 define <4 x i64> @test_x86_avx2_psllv_q_256(<4 x i64> %a0, <4 x i64> %a1) {
-  ; CHECK: vpsllvq
+; CHECK-LABEL: test_x86_avx2_psllv_q_256:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    vpsllvq %ymm1, %ymm0, %ymm0
+; CHECK-NEXT:    retl
   %res = call <4 x i64> @llvm.x86.avx2.psllv.q.256(<4 x i64> %a0, <4 x i64> %a1) ; <<4 x i64>> [#uses=1]
   ret <4 x i64> %res
 }
@@ -784,7 +1084,10 @@ declare <4 x i64> @llvm.x86.avx2.psllv.q.256(<4 x i64>, <4 x i64>) nounwind read
 
 
 define <4 x i32> @test_x86_avx2_psrlv_d(<4 x i32> %a0, <4 x i32> %a1) {
-  ; CHECK: vpsrlvd
+; CHECK-LABEL: test_x86_avx2_psrlv_d:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    vpsrlvd %xmm1, %xmm0, %xmm0
+; CHECK-NEXT:    retl
   %res = call <4 x i32> @llvm.x86.avx2.psrlv.d(<4 x i32> %a0, <4 x i32> %a1) ; <<4 x i32>> [#uses=1]
   ret <4 x i32> %res
 }
@@ -792,7 +1095,10 @@ declare <4 x i32> @llvm.x86.avx2.psrlv.d(<4 x i32>, <4 x i32>) nounwind readnone
 
 
 define <8 x i32> @test_x86_avx2_psrlv_d_256(<8 x i32> %a0, <8 x i32> %a1) {
-  ; CHECK: vpsrlvd
+; CHECK-LABEL: test_x86_avx2_psrlv_d_256:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    vpsrlvd %ymm1, %ymm0, %ymm0
+; CHECK-NEXT:    retl
   %res = call <8 x i32> @llvm.x86.avx2.psrlv.d.256(<8 x i32> %a0, <8 x i32> %a1) ; <<8 x i32>> [#uses=1]
   ret <8 x i32> %res
 }
@@ -800,7 +1106,10 @@ declare <8 x i32> @llvm.x86.avx2.psrlv.d.256(<8 x i32>, <8 x i32>) nounwind read
 
 
 define <2 x i64> @test_x86_avx2_psrlv_q(<2 x i64> %a0, <2 x i64> %a1) {
-  ; CHECK: vpsrlvq
+; CHECK-LABEL: test_x86_avx2_psrlv_q:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    vpsrlvq %xmm1, %xmm0, %xmm0
+; CHECK-NEXT:    retl
   %res = call <2 x i64> @llvm.x86.avx2.psrlv.q(<2 x i64> %a0, <2 x i64> %a1) ; <<2 x i64>> [#uses=1]
   ret <2 x i64> %res
 }
@@ -808,7 +1117,10 @@ declare <2 x i64> @llvm.x86.avx2.psrlv.q(<2 x i64>, <2 x i64>) nounwind readnone
 
 
 define <4 x i64> @test_x86_avx2_psrlv_q_256(<4 x i64> %a0, <4 x i64> %a1) {
-  ; CHECK: vpsrlvq
+; CHECK-LABEL: test_x86_avx2_psrlv_q_256:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    vpsrlvq %ymm1, %ymm0, %ymm0
+; CHECK-NEXT:    retl
   %res = call <4 x i64> @llvm.x86.avx2.psrlv.q.256(<4 x i64> %a0, <4 x i64> %a1) ; <<4 x i64>> [#uses=1]
   ret <4 x i64> %res
 }
@@ -816,7 +1128,10 @@ declare <4 x i64> @llvm.x86.avx2.psrlv.q.256(<4 x i64>, <4 x i64>) nounwind read
 
 
 define <4 x i32> @test_x86_avx2_psrav_d(<4 x i32> %a0, <4 x i32> %a1) {
-  ; CHECK: vpsravd
+; CHECK-LABEL: test_x86_avx2_psrav_d:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    vpsravd %xmm1, %xmm0, %xmm0
+; CHECK-NEXT:    retl
   %res = call <4 x i32> @llvm.x86.avx2.psrav.d(<4 x i32> %a0, <4 x i32> %a1) ; <<4 x i32>> [#uses=1]
   ret <4 x i32> %res
 }
@@ -824,7 +1139,10 @@ declare <4 x i32> @llvm.x86.avx2.psrav.d(<4 x i32>, <4 x i32>) nounwind readnone
 
 
 define <8 x i32> @test_x86_avx2_psrav_d_256(<8 x i32> %a0, <8 x i32> %a1) {
-  ; CHECK: vpsravd
+; CHECK-LABEL: test_x86_avx2_psrav_d_256:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    vpsravd %ymm1, %ymm0, %ymm0
+; CHECK-NEXT:    retl
   %res = call <8 x i32> @llvm.x86.avx2.psrav.d.256(<8 x i32> %a0, <8 x i32> %a1) ; <<8 x i32>> [#uses=1]
   ret <8 x i32> %res
 }
@@ -832,7 +1150,13 @@ declare <8 x i32> @llvm.x86.avx2.psrav.d.256(<8 x i32>, <8 x i32>) nounwind read
 
 ; This is checked here because the execution dependency fix pass makes it hard to test in AVX mode since we don't have 256-bit integer instructions
 define void @test_x86_avx_storeu_dq_256(i8* %a0, <32 x i8> %a1) {
-  ; CHECK: vmovdqu
+; CHECK-LABEL: test_x86_avx_storeu_dq_256:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    movl {{[0-9]+}}(%esp), %eax
+; CHECK-NEXT:    vpaddb LCPI103_0, %ymm0, %ymm0
+; CHECK-NEXT:    vmovdqu %ymm0, (%eax)
+; CHECK-NEXT:    vzeroupper
+; CHECK-NEXT:    retl
   ; add operation forces the execution domain.
   %a2 = add <32 x i8> %a1, <i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1>
   call void @llvm.x86.avx.storeu.dq.256(i8* %a0, <32 x i8> %a2)
@@ -840,9 +1164,12 @@ define void @test_x86_avx_storeu_dq_256(i8* %a0, <32 x i8> %a1) {
 }
 declare void @llvm.x86.avx.storeu.dq.256(i8*, <32 x i8>) nounwind
 
-define <2 x double> @test_x86_avx2_gather_d_pd(<2 x double> %a0, i8* %a1,
-                     <4 x i32> %idx, <2 x double> %mask) {
-  ; CHECK: vgatherdpd
+define <2 x double> @test_x86_avx2_gather_d_pd(<2 x double> %a0, i8* %a1, <4 x i32> %idx, <2 x double> %mask) {
+; CHECK-LABEL: test_x86_avx2_gather_d_pd:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    movl {{[0-9]+}}(%esp), %eax
+; CHECK-NEXT:    vgatherdpd %xmm2, (%eax,%xmm1,2), %xmm0
+; CHECK-NEXT:    retl
   %res = call <2 x double> @llvm.x86.avx2.gather.d.pd(<2 x double> %a0,
                             i8* %a1, <4 x i32> %idx, <2 x double> %mask, i8 2) ;
   ret <2 x double> %res
@@ -850,9 +1177,12 @@ define <2 x double> @test_x86_avx2_gather_d_pd(<2 x double> %a0, i8* %a1,
 declare <2 x double> @llvm.x86.avx2.gather.d.pd(<2 x double>, i8*,
                       <4 x i32>, <2 x double>, i8) nounwind readonly
 
-define <4 x double> @test_x86_avx2_gather_d_pd_256(<4 x double> %a0, i8* %a1,
-                     <4 x i32> %idx, <4 x double> %mask) {
-  ; CHECK: vgatherdpd
+define <4 x double> @test_x86_avx2_gather_d_pd_256(<4 x double> %a0, i8* %a1, <4 x i32> %idx, <4 x double> %mask) {
+; CHECK-LABEL: test_x86_avx2_gather_d_pd_256:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    movl {{[0-9]+}}(%esp), %eax
+; CHECK-NEXT:    vgatherdpd %ymm2, (%eax,%xmm1,2), %ymm0
+; CHECK-NEXT:    retl
   %res = call <4 x double> @llvm.x86.avx2.gather.d.pd.256(<4 x double> %a0,
                             i8* %a1, <4 x i32> %idx, <4 x double> %mask, i8 2) ;
   ret <4 x double> %res
@@ -860,9 +1190,12 @@ define <4 x double> @test_x86_avx2_gather_d_pd_256(<4 x double> %a0, i8* %a1,
 declare <4 x double> @llvm.x86.avx2.gather.d.pd.256(<4 x double>, i8*,
                       <4 x i32>, <4 x double>, i8) nounwind readonly
 
-define <2 x double> @test_x86_avx2_gather_q_pd(<2 x double> %a0, i8* %a1,
-                     <2 x i64> %idx, <2 x double> %mask) {
-  ; CHECK: vgatherqpd
+define <2 x double> @test_x86_avx2_gather_q_pd(<2 x double> %a0, i8* %a1, <2 x i64> %idx, <2 x double> %mask) {
+; CHECK-LABEL: test_x86_avx2_gather_q_pd:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    movl {{[0-9]+}}(%esp), %eax
+; CHECK-NEXT:    vgatherqpd %xmm2, (%eax,%xmm1,2), %xmm0
+; CHECK-NEXT:    retl
   %res = call <2 x double> @llvm.x86.avx2.gather.q.pd(<2 x double> %a0,
                             i8* %a1, <2 x i64> %idx, <2 x double> %mask, i8 2) ;
   ret <2 x double> %res
@@ -870,9 +1203,12 @@ define <2 x double> @test_x86_avx2_gather_q_pd(<2 x double> %a0, i8* %a1,
 declare <2 x double> @llvm.x86.avx2.gather.q.pd(<2 x double>, i8*,
                       <2 x i64>, <2 x double>, i8) nounwind readonly
 
-define <4 x double> @test_x86_avx2_gather_q_pd_256(<4 x double> %a0, i8* %a1,
-                     <4 x i64> %idx, <4 x double> %mask) {
-  ; CHECK: vgatherqpd
+define <4 x double> @test_x86_avx2_gather_q_pd_256(<4 x double> %a0, i8* %a1, <4 x i64> %idx, <4 x double> %mask) {
+; CHECK-LABEL: test_x86_avx2_gather_q_pd_256:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    movl {{[0-9]+}}(%esp), %eax
+; CHECK-NEXT:    vgatherqpd %ymm2, (%eax,%ymm1,2), %ymm0
+; CHECK-NEXT:    retl
   %res = call <4 x double> @llvm.x86.avx2.gather.q.pd.256(<4 x double> %a0,
                             i8* %a1, <4 x i64> %idx, <4 x double> %mask, i8 2) ;
   ret <4 x double> %res
@@ -880,9 +1216,12 @@ define <4 x double> @test_x86_avx2_gather_q_pd_256(<4 x double> %a0, i8* %a1,
 declare <4 x double> @llvm.x86.avx2.gather.q.pd.256(<4 x double>, i8*,
                       <4 x i64>, <4 x double>, i8) nounwind readonly
 
-define <4 x float> @test_x86_avx2_gather_d_ps(<4 x float> %a0, i8* %a1,
-                     <4 x i32> %idx, <4 x float> %mask) {
-  ; CHECK: vgatherdps
+define <4 x float> @test_x86_avx2_gather_d_ps(<4 x float> %a0, i8* %a1, <4 x i32> %idx, <4 x float> %mask) {
+; CHECK-LABEL: test_x86_avx2_gather_d_ps:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    movl {{[0-9]+}}(%esp), %eax
+; CHECK-NEXT:    vgatherdps %xmm2, (%eax,%xmm1,2), %xmm0
+; CHECK-NEXT:    retl
   %res = call <4 x float> @llvm.x86.avx2.gather.d.ps(<4 x float> %a0,
                             i8* %a1, <4 x i32> %idx, <4 x float> %mask, i8 2) ;
   ret <4 x float> %res
@@ -890,9 +1229,12 @@ define <4 x float> @test_x86_avx2_gather_d_ps(<4 x float> %a0, i8* %a1,
 declare <4 x float> @llvm.x86.avx2.gather.d.ps(<4 x float>, i8*,
                       <4 x i32>, <4 x float>, i8) nounwind readonly
 
-define <8 x float> @test_x86_avx2_gather_d_ps_256(<8 x float> %a0, i8* %a1,
-                     <8 x i32> %idx, <8 x float> %mask) {
-  ; CHECK: vgatherdps
+define <8 x float> @test_x86_avx2_gather_d_ps_256(<8 x float> %a0, i8* %a1, <8 x i32> %idx, <8 x float> %mask) {
+; CHECK-LABEL: test_x86_avx2_gather_d_ps_256:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    movl {{[0-9]+}}(%esp), %eax
+; CHECK-NEXT:    vgatherdps %ymm2, (%eax,%ymm1,2), %ymm0
+; CHECK-NEXT:    retl
   %res = call <8 x float> @llvm.x86.avx2.gather.d.ps.256(<8 x float> %a0,
                             i8* %a1, <8 x i32> %idx, <8 x float> %mask, i8 2) ;
   ret <8 x float> %res
@@ -900,9 +1242,12 @@ define <8 x float> @test_x86_avx2_gather_d_ps_256(<8 x float> %a0, i8* %a1,
 declare <8 x float> @llvm.x86.avx2.gather.d.ps.256(<8 x float>, i8*,
                       <8 x i32>, <8 x float>, i8) nounwind readonly
 
-define <4 x float> @test_x86_avx2_gather_q_ps(<4 x float> %a0, i8* %a1,
-                     <2 x i64> %idx, <4 x float> %mask) {
-  ; CHECK: vgatherqps
+define <4 x float> @test_x86_avx2_gather_q_ps(<4 x float> %a0, i8* %a1, <2 x i64> %idx, <4 x float> %mask) {
+; CHECK-LABEL: test_x86_avx2_gather_q_ps:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    movl {{[0-9]+}}(%esp), %eax
+; CHECK-NEXT:    vgatherqps %xmm2, (%eax,%xmm1,2), %xmm0
+; CHECK-NEXT:    retl
   %res = call <4 x float> @llvm.x86.avx2.gather.q.ps(<4 x float> %a0,
                             i8* %a1, <2 x i64> %idx, <4 x float> %mask, i8 2) ;
   ret <4 x float> %res
@@ -910,9 +1255,13 @@ define <4 x float> @test_x86_avx2_gather_q_ps(<4 x float> %a0, i8* %a1,
 declare <4 x float> @llvm.x86.avx2.gather.q.ps(<4 x float>, i8*,
                       <2 x i64>, <4 x float>, i8) nounwind readonly
 
-define <4 x float> @test_x86_avx2_gather_q_ps_256(<4 x float> %a0, i8* %a1,
-                     <4 x i64> %idx, <4 x float> %mask) {
-  ; CHECK: vgatherqps
+define <4 x float> @test_x86_avx2_gather_q_ps_256(<4 x float> %a0, i8* %a1, <4 x i64> %idx, <4 x float> %mask) {
+; CHECK-LABEL: test_x86_avx2_gather_q_ps_256:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    movl {{[0-9]+}}(%esp), %eax
+; CHECK-NEXT:    vgatherqps %xmm2, (%eax,%ymm1,2), %xmm0
+; CHECK-NEXT:    vzeroupper
+; CHECK-NEXT:    retl
   %res = call <4 x float> @llvm.x86.avx2.gather.q.ps.256(<4 x float> %a0,
                             i8* %a1, <4 x i64> %idx, <4 x float> %mask, i8 2) ;
   ret <4 x float> %res
@@ -920,9 +1269,12 @@ define <4 x float> @test_x86_avx2_gather_q_ps_256(<4 x float> %a0, i8* %a1,
 declare <4 x float> @llvm.x86.avx2.gather.q.ps.256(<4 x float>, i8*,
                       <4 x i64>, <4 x float>, i8) nounwind readonly
 
-define <2 x i64> @test_x86_avx2_gather_d_q(<2 x i64> %a0, i8* %a1,
-                     <4 x i32> %idx, <2 x i64> %mask) {
-  ; CHECK: vpgatherdq
+define <2 x i64> @test_x86_avx2_gather_d_q(<2 x i64> %a0, i8* %a1, <4 x i32> %idx, <2 x i64> %mask) {
+; CHECK-LABEL: test_x86_avx2_gather_d_q:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    movl {{[0-9]+}}(%esp), %eax
+; CHECK-NEXT:    vpgatherdq %xmm2, (%eax,%xmm1,2), %xmm0
+; CHECK-NEXT:    retl
   %res = call <2 x i64> @llvm.x86.avx2.gather.d.q(<2 x i64> %a0,
                             i8* %a1, <4 x i32> %idx, <2 x i64> %mask, i8 2) ;
   ret <2 x i64> %res
@@ -930,9 +1282,12 @@ define <2 x i64> @test_x86_avx2_gather_d_q(<2 x i64> %a0, i8* %a1,
 declare <2 x i64> @llvm.x86.avx2.gather.d.q(<2 x i64>, i8*,
                       <4 x i32>, <2 x i64>, i8) nounwind readonly
 
-define <4 x i64> @test_x86_avx2_gather_d_q_256(<4 x i64> %a0, i8* %a1,
-                     <4 x i32> %idx, <4 x i64> %mask) {
-  ; CHECK: vpgatherdq
+define <4 x i64> @test_x86_avx2_gather_d_q_256(<4 x i64> %a0, i8* %a1, <4 x i32> %idx, <4 x i64> %mask) {
+; CHECK-LABEL: test_x86_avx2_gather_d_q_256:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    movl {{[0-9]+}}(%esp), %eax
+; CHECK-NEXT:    vpgatherdq %ymm2, (%eax,%xmm1,2), %ymm0
+; CHECK-NEXT:    retl
   %res = call <4 x i64> @llvm.x86.avx2.gather.d.q.256(<4 x i64> %a0,
                             i8* %a1, <4 x i32> %idx, <4 x i64> %mask, i8 2) ;
   ret <4 x i64> %res
@@ -940,9 +1295,12 @@ define <4 x i64> @test_x86_avx2_gather_d_q_256(<4 x i64> %a0, i8* %a1,
 declare <4 x i64> @llvm.x86.avx2.gather.d.q.256(<4 x i64>, i8*,
                       <4 x i32>, <4 x i64>, i8) nounwind readonly
 
-define <2 x i64> @test_x86_avx2_gather_q_q(<2 x i64> %a0, i8* %a1,
-                     <2 x i64> %idx, <2 x i64> %mask) {
-  ; CHECK: vpgatherqq
+define <2 x i64> @test_x86_avx2_gather_q_q(<2 x i64> %a0, i8* %a1, <2 x i64> %idx, <2 x i64> %mask) {
+; CHECK-LABEL: test_x86_avx2_gather_q_q:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    movl {{[0-9]+}}(%esp), %eax
+; CHECK-NEXT:    vpgatherqq %xmm2, (%eax,%xmm1,2), %xmm0
+; CHECK-NEXT:    retl
   %res = call <2 x i64> @llvm.x86.avx2.gather.q.q(<2 x i64> %a0,
                             i8* %a1, <2 x i64> %idx, <2 x i64> %mask, i8 2) ;
   ret <2 x i64> %res
@@ -950,9 +1308,12 @@ define <2 x i64> @test_x86_avx2_gather_q_q(<2 x i64> %a0, i8* %a1,
 declare <2 x i64> @llvm.x86.avx2.gather.q.q(<2 x i64>, i8*,
                       <2 x i64>, <2 x i64>, i8) nounwind readonly
 
-define <4 x i64> @test_x86_avx2_gather_q_q_256(<4 x i64> %a0, i8* %a1,
-                     <4 x i64> %idx, <4 x i64> %mask) {
-  ; CHECK: vpgatherqq
+define <4 x i64> @test_x86_avx2_gather_q_q_256(<4 x i64> %a0, i8* %a1, <4 x i64> %idx, <4 x i64> %mask) {
+; CHECK-LABEL: test_x86_avx2_gather_q_q_256:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    movl {{[0-9]+}}(%esp), %eax
+; CHECK-NEXT:    vpgatherqq %ymm2, (%eax,%ymm1,2), %ymm0
+; CHECK-NEXT:    retl
   %res = call <4 x i64> @llvm.x86.avx2.gather.q.q.256(<4 x i64> %a0,
                             i8* %a1, <4 x i64> %idx, <4 x i64> %mask, i8 2) ;
   ret <4 x i64> %res
@@ -960,9 +1321,12 @@ define <4 x i64> @test_x86_avx2_gather_q_q_256(<4 x i64> %a0, i8* %a1,
 declare <4 x i64> @llvm.x86.avx2.gather.q.q.256(<4 x i64>, i8*,
                       <4 x i64>, <4 x i64>, i8) nounwind readonly
 
-define <4 x i32> @test_x86_avx2_gather_d_d(<4 x i32> %a0, i8* %a1,
-                     <4 x i32> %idx, <4 x i32> %mask) {
-  ; CHECK: vpgatherdd
+define <4 x i32> @test_x86_avx2_gather_d_d(<4 x i32> %a0, i8* %a1, <4 x i32> %idx, <4 x i32> %mask) {
+; CHECK-LABEL: test_x86_avx2_gather_d_d:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    movl {{[0-9]+}}(%esp), %eax
+; CHECK-NEXT:    vpgatherdd %xmm2, (%eax,%xmm1,2), %xmm0
+; CHECK-NEXT:    retl
   %res = call <4 x i32> @llvm.x86.avx2.gather.d.d(<4 x i32> %a0,
                             i8* %a1, <4 x i32> %idx, <4 x i32> %mask, i8 2) ;
   ret <4 x i32> %res
@@ -970,9 +1334,12 @@ define <4 x i32> @test_x86_avx2_gather_d_d(<4 x i32> %a0, i8* %a1,
 declare <4 x i32> @llvm.x86.avx2.gather.d.d(<4 x i32>, i8*,
                       <4 x i32>, <4 x i32>, i8) nounwind readonly
 
-define <8 x i32> @test_x86_avx2_gather_d_d_256(<8 x i32> %a0, i8* %a1,
-                     <8 x i32> %idx, <8 x i32> %mask) {
-  ; CHECK: vpgatherdd
+define <8 x i32> @test_x86_avx2_gather_d_d_256(<8 x i32> %a0, i8* %a1, <8 x i32> %idx, <8 x i32> %mask) {
+; CHECK-LABEL: test_x86_avx2_gather_d_d_256:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    movl {{[0-9]+}}(%esp), %eax
+; CHECK-NEXT:    vpgatherdd %ymm2, (%eax,%ymm1,2), %ymm0
+; CHECK-NEXT:    retl
   %res = call <8 x i32> @llvm.x86.avx2.gather.d.d.256(<8 x i32> %a0,
                             i8* %a1, <8 x i32> %idx, <8 x i32> %mask, i8 2) ;
   ret <8 x i32> %res
@@ -980,9 +1347,12 @@ define <8 x i32> @test_x86_avx2_gather_d_d_256(<8 x i32> %a0, i8* %a1,
 declare <8 x i32> @llvm.x86.avx2.gather.d.d.256(<8 x i32>, i8*,
                       <8 x i32>, <8 x i32>, i8) nounwind readonly
 
-define <4 x i32> @test_x86_avx2_gather_q_d(<4 x i32> %a0, i8* %a1,
-                     <2 x i64> %idx, <4 x i32> %mask) {
-  ; CHECK: vpgatherqd
+define <4 x i32> @test_x86_avx2_gather_q_d(<4 x i32> %a0, i8* %a1, <2 x i64> %idx, <4 x i32> %mask) {
+; CHECK-LABEL: test_x86_avx2_gather_q_d:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    movl {{[0-9]+}}(%esp), %eax
+; CHECK-NEXT:    vpgatherqd %xmm2, (%eax,%xmm1,2), %xmm0
+; CHECK-NEXT:    retl
   %res = call <4 x i32> @llvm.x86.avx2.gather.q.d(<4 x i32> %a0,
                             i8* %a1, <2 x i64> %idx, <4 x i32> %mask, i8 2) ;
   ret <4 x i32> %res
@@ -990,9 +1360,13 @@ define <4 x i32> @test_x86_avx2_gather_q_d(<4 x i32> %a0, i8* %a1,
 declare <4 x i32> @llvm.x86.avx2.gather.q.d(<4 x i32>, i8*,
                       <2 x i64>, <4 x i32>, i8) nounwind readonly
 
-define <4 x i32> @test_x86_avx2_gather_q_d_256(<4 x i32> %a0, i8* %a1,
-                     <4 x i64> %idx, <4 x i32> %mask) {
-  ; CHECK: vpgatherqd
+define <4 x i32> @test_x86_avx2_gather_q_d_256(<4 x i32> %a0, i8* %a1, <4 x i64> %idx, <4 x i32> %mask) {
+; CHECK-LABEL: test_x86_avx2_gather_q_d_256:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    movl {{[0-9]+}}(%esp), %eax
+; CHECK-NEXT:    vpgatherqd %xmm2, (%eax,%ymm1,2), %xmm0
+; CHECK-NEXT:    vzeroupper
+; CHECK-NEXT:    retl
   %res = call <4 x i32> @llvm.x86.avx2.gather.q.d.256(<4 x i32> %a0,
                             i8* %a1, <4 x i64> %idx, <4 x i32> %mask, i8 2) ;
   ret <4 x i32> %res
@@ -1001,12 +1375,15 @@ declare <4 x i32> @llvm.x86.avx2.gather.q.d.256(<4 x i32>, i8*,
                       <4 x i64>, <4 x i32>, i8) nounwind readonly
 
 ; PR13298
-define <8 x float>  @test_gather_mask(<8 x float> %a0, float* %a,
-                                      <8 x i32> %idx, <8 x float> %mask,
-                                      float* nocapture %out) {
-; CHECK: test_gather_mask
-; CHECK: vmovaps %ymm2, [[DEST:%.*]]
-; CHECK: vgatherdps [[DEST]]
+define <8 x float>  @test_gather_mask(<8 x float> %a0, float* %a, <8 x i32> %idx, <8 x float> %mask, float* nocapture %out) {
+; CHECK-LABEL: test_gather_mask:
+; CHECK:       ## BB#0:
+; CHECK-NEXT:    movl {{[0-9]+}}(%esp), %eax
+; CHECK-NEXT:    movl {{[0-9]+}}(%esp), %ecx
+; CHECK-NEXT:    vmovaps %ymm2, %ymm3
+; CHECK-NEXT:    vgatherdps %ymm3, (%ecx,%ymm1,4), %ymm0
+; CHECK-NEXT:    vmovups %ymm2, (%eax)
+; CHECK-NEXT:    retl
 ;; gather with mask
   %a_i8 = bitcast float* %a to i8*
   %res = call <8 x float> @llvm.x86.avx2.gather.d.ps.256(<8 x float> %a0,
-- 
2.7.4