[llvm] 08ac661 - [test][msan] Relax order of param shadow
Vitaly Buka via llvm-commits
llvm-commits at lists.llvm.org
Sat May 7 21:19:41 PDT 2022
Author: Vitaly Buka
Date: 2022-05-07T21:17:44-07:00
New Revision: 08ac66124874d70dab63c731da0244f9e29ef168
URL: https://github.com/llvm/llvm-project/commit/08ac66124874d70dab63c731da0244f9e29ef168
DIFF: https://github.com/llvm/llvm-project/commit/08ac66124874d70dab63c731da0244f9e29ef168.diff
LOG: [test][msan] Relax order of param shadow
Looks like different bots have them in a different order.
Added:
Modified:
llvm/test/Instrumentation/MemorySanitizer/avx-intrinsics-x86.ll
llvm/test/Instrumentation/MemorySanitizer/avx2-intrinsics-x86.ll
llvm/test/Instrumentation/MemorySanitizer/sse-intrinsics-x86.ll
llvm/test/Instrumentation/MemorySanitizer/sse2-intrinsics-x86.ll
llvm/test/Instrumentation/MemorySanitizer/sse41-intrinsics-x86.ll
Removed:
################################################################################
diff --git a/llvm/test/Instrumentation/MemorySanitizer/avx-intrinsics-x86.ll b/llvm/test/Instrumentation/MemorySanitizer/avx-intrinsics-x86.ll
index 70f4d6f237a4..8661f29d80fd 100644
--- a/llvm/test/Instrumentation/MemorySanitizer/avx-intrinsics-x86.ll
+++ b/llvm/test/Instrumentation/MemorySanitizer/avx-intrinsics-x86.ll
@@ -6,8 +6,8 @@ target triple = "x86_64-unknown-linux-gnu"
define <4 x double> @test_x86_avx_addsub_pd_256(<4 x double> %a0, <4 x double> %a1) #0 {
; CHECK-LABEL: @test_x86_avx_addsub_pd_256(
-; CHECK-NEXT: [[TMP1:%.*]] = load <4 x i64>, <4 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i64>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <4 x i64>, <4 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <4 x i64>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <4 x i64>, <4 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i64>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <4 x i64>, <4 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <4 x i64>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[_MSPROP:%.*]] = or <4 x i64> [[TMP1]], [[TMP2]]
; CHECK-NEXT: [[RES:%.*]] = call <4 x double> @llvm.x86.avx.addsub.pd.256(<4 x double> [[A0:%.*]], <4 x double> [[A1:%.*]])
@@ -22,8 +22,8 @@ declare <4 x double> @llvm.x86.avx.addsub.pd.256(<4 x double>, <4 x double>) nou
define <8 x float> @test_x86_avx_addsub_ps_256(<8 x float> %a0, <8 x float> %a1) #0 {
; CHECK-LABEL: @test_x86_avx_addsub_ps_256(
-; CHECK-NEXT: [[TMP1:%.*]] = load <8 x i32>, <8 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i32>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <8 x i32>, <8 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <8 x i32>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <8 x i32>, <8 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i32>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <8 x i32>, <8 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <8 x i32>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[_MSPROP:%.*]] = or <8 x i32> [[TMP1]], [[TMP2]]
; CHECK-NEXT: [[RES:%.*]] = call <8 x float> @llvm.x86.avx.addsub.ps.256(<8 x float> [[A0:%.*]], <8 x float> [[A1:%.*]])
@@ -38,9 +38,9 @@ declare <8 x float> @llvm.x86.avx.addsub.ps.256(<8 x float>, <8 x float>) nounwi
define <4 x double> @test_x86_avx_blendv_pd_256(<4 x double> %a0, <4 x double> %a1, <4 x double> %a2) #0 {
; CHECK-LABEL: @test_x86_avx_blendv_pd_256(
-; CHECK-NEXT: [[TMP1:%.*]] = load <4 x i64>, <4 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i64>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <4 x i64>, <4 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <4 x i64>*), align 8
-; CHECK-NEXT: [[TMP3:%.*]] = load <4 x i64>, <4 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 64) to <4 x i64>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <4 x i64>, <4 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i64>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <4 x i64>, <4 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <4 x i64>*), align 8
+; CHECK-DAG: [[TMP3:%.*]] = load <4 x i64>, <4 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 64) to <4 x i64>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[_MSPROP:%.*]] = or <4 x i64> [[TMP1]], [[TMP2]]
; CHECK-NEXT: [[_MSPROP1:%.*]] = or <4 x i64> [[_MSPROP]], [[TMP3]]
@@ -56,9 +56,9 @@ declare <4 x double> @llvm.x86.avx.blendv.pd.256(<4 x double>, <4 x double>, <4
define <8 x float> @test_x86_avx_blendv_ps_256(<8 x float> %a0, <8 x float> %a1, <8 x float> %a2) #0 {
; CHECK-LABEL: @test_x86_avx_blendv_ps_256(
-; CHECK-NEXT: [[TMP1:%.*]] = load <8 x i32>, <8 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i32>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <8 x i32>, <8 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <8 x i32>*), align 8
-; CHECK-NEXT: [[TMP3:%.*]] = load <8 x i32>, <8 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 64) to <8 x i32>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <8 x i32>, <8 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i32>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <8 x i32>, <8 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <8 x i32>*), align 8
+; CHECK-DAG: [[TMP3:%.*]] = load <8 x i32>, <8 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 64) to <8 x i32>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[_MSPROP:%.*]] = or <8 x i32> [[TMP1]], [[TMP2]]
; CHECK-NEXT: [[_MSPROP1:%.*]] = or <8 x i32> [[_MSPROP]], [[TMP3]]
@@ -74,8 +74,8 @@ declare <8 x float> @llvm.x86.avx.blendv.ps.256(<8 x float>, <8 x float>, <8 x f
define <4 x double> @test_x86_avx_cmp_pd_256(<4 x double> %a0, <4 x double> %a1) #0 {
; CHECK-LABEL: @test_x86_avx_cmp_pd_256(
-; CHECK-NEXT: [[TMP1:%.*]] = load <4 x i64>, <4 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i64>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <4 x i64>, <4 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <4 x i64>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <4 x i64>, <4 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i64>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <4 x i64>, <4 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <4 x i64>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP3:%.*]] = bitcast <4 x i64> [[TMP1]] to i256
; CHECK-NEXT: [[_MSCMP:%.*]] = icmp ne i256 [[TMP3]], 0
@@ -103,8 +103,8 @@ declare <4 x double> @llvm.x86.avx.cmp.pd.256(<4 x double>, <4 x double>, i8) no
define <8 x float> @test_x86_avx_cmp_ps_256(<8 x float> %a0, <8 x float> %a1) #0 {
; CHECK-LABEL: @test_x86_avx_cmp_ps_256(
-; CHECK-NEXT: [[TMP1:%.*]] = load <8 x i32>, <8 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i32>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <8 x i32>, <8 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <8 x i32>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <8 x i32>, <8 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i32>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <8 x i32>, <8 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <8 x i32>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP3:%.*]] = bitcast <8 x i32> [[TMP1]] to i256
; CHECK-NEXT: [[_MSCMP:%.*]] = icmp ne i256 [[TMP3]], 0
@@ -130,8 +130,8 @@ define <8 x float> @test_x86_avx_cmp_ps_256(<8 x float> %a0, <8 x float> %a1) #0
define <8 x float> @test_x86_avx_cmp_ps_256_pseudo_op(<8 x float> %a0, <8 x float> %a1) #0 {
; CHECK-LABEL: @test_x86_avx_cmp_ps_256_pseudo_op(
-; CHECK-NEXT: [[TMP1:%.*]] = load <8 x i32>, <8 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i32>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <8 x i32>, <8 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <8 x i32>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <8 x i32>, <8 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i32>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <8 x i32>, <8 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <8 x i32>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP3:%.*]] = bitcast <8 x i32> [[TMP1]] to i256
; CHECK-NEXT: [[_MSCMP:%.*]] = icmp ne i256 [[TMP3]], 0
@@ -438,7 +438,7 @@ declare <8 x float> @llvm.x86.avx.cmp.ps.256(<8 x float>, <8 x float>, i8) nounw
define <4 x float> @test_x86_avx_cvt_pd2_ps_256(<4 x double> %a0) #0 {
; CHECK-LABEL: @test_x86_avx_cvt_pd2_ps_256(
-; CHECK-NEXT: [[TMP1:%.*]] = load <4 x i64>, <4 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i64>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <4 x i64>, <4 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i64>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP2:%.*]] = bitcast <4 x i64> [[TMP1]] to i256
; CHECK-NEXT: [[_MSCMP:%.*]] = icmp ne i256 [[TMP2]], 0
@@ -459,7 +459,7 @@ declare <4 x float> @llvm.x86.avx.cvt.pd2.ps.256(<4 x double>) nounwind readnone
define <4 x i32> @test_x86_avx_cvt_pd2dq_256(<4 x double> %a0) #0 {
; CHECK-LABEL: @test_x86_avx_cvt_pd2dq_256(
-; CHECK-NEXT: [[TMP1:%.*]] = load <4 x i64>, <4 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i64>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <4 x i64>, <4 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i64>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP2:%.*]] = bitcast <4 x i64> [[TMP1]] to i256
; CHECK-NEXT: [[_MSCMP:%.*]] = icmp ne i256 [[TMP2]], 0
@@ -480,7 +480,7 @@ declare <4 x i32> @llvm.x86.avx.cvt.pd2dq.256(<4 x double>) nounwind readnone
define <8 x i32> @test_x86_avx_cvt_ps2dq_256(<8 x float> %a0) #0 {
; CHECK-LABEL: @test_x86_avx_cvt_ps2dq_256(
-; CHECK-NEXT: [[TMP1:%.*]] = load <8 x i32>, <8 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i32>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <8 x i32>, <8 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i32>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP2:%.*]] = bitcast <8 x i32> [[TMP1]] to i256
; CHECK-NEXT: [[_MSCMP:%.*]] = icmp ne i256 [[TMP2]], 0
@@ -501,7 +501,7 @@ declare <8 x i32> @llvm.x86.avx.cvt.ps2dq.256(<8 x float>) nounwind readnone
define <4 x i32> @test_x86_avx_cvtt_pd2dq_256(<4 x double> %a0) #0 {
; CHECK-LABEL: @test_x86_avx_cvtt_pd2dq_256(
-; CHECK-NEXT: [[TMP1:%.*]] = load <4 x i64>, <4 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i64>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <4 x i64>, <4 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i64>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP2:%.*]] = bitcast <4 x i64> [[TMP1]] to i256
; CHECK-NEXT: [[_MSCMP:%.*]] = icmp ne i256 [[TMP2]], 0
@@ -522,7 +522,7 @@ declare <4 x i32> @llvm.x86.avx.cvtt.pd2dq.256(<4 x double>) nounwind readnone
define <8 x i32> @test_x86_avx_cvtt_ps2dq_256(<8 x float> %a0) #0 {
; CHECK-LABEL: @test_x86_avx_cvtt_ps2dq_256(
-; CHECK-NEXT: [[TMP1:%.*]] = load <8 x i32>, <8 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i32>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <8 x i32>, <8 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i32>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP2:%.*]] = bitcast <8 x i32> [[TMP1]] to i256
; CHECK-NEXT: [[_MSCMP:%.*]] = icmp ne i256 [[TMP2]], 0
@@ -543,8 +543,8 @@ declare <8 x i32> @llvm.x86.avx.cvtt.ps2dq.256(<8 x float>) nounwind readnone
define <8 x float> @test_x86_avx_dp_ps_256(<8 x float> %a0, <8 x float> %a1) #0 {
; CHECK-LABEL: @test_x86_avx_dp_ps_256(
-; CHECK-NEXT: [[TMP1:%.*]] = load <8 x i32>, <8 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i32>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <8 x i32>, <8 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <8 x i32>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <8 x i32>, <8 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i32>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <8 x i32>, <8 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <8 x i32>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP3:%.*]] = bitcast <8 x i32> [[TMP1]] to i256
; CHECK-NEXT: [[_MSCMP:%.*]] = icmp ne i256 [[TMP3]], 0
@@ -572,8 +572,8 @@ declare <8 x float> @llvm.x86.avx.dp.ps.256(<8 x float>, <8 x float>, i8) nounwi
define <4 x double> @test_x86_avx_hadd_pd_256(<4 x double> %a0, <4 x double> %a1) #0 {
; CHECK-LABEL: @test_x86_avx_hadd_pd_256(
-; CHECK-NEXT: [[TMP1:%.*]] = load <4 x i64>, <4 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i64>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <4 x i64>, <4 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <4 x i64>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <4 x i64>, <4 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i64>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <4 x i64>, <4 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <4 x i64>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[_MSPROP:%.*]] = or <4 x i64> [[TMP1]], [[TMP2]]
; CHECK-NEXT: [[RES:%.*]] = call <4 x double> @llvm.x86.avx.hadd.pd.256(<4 x double> [[A0:%.*]], <4 x double> [[A1:%.*]])
@@ -588,8 +588,8 @@ declare <4 x double> @llvm.x86.avx.hadd.pd.256(<4 x double>, <4 x double>) nounw
define <8 x float> @test_x86_avx_hadd_ps_256(<8 x float> %a0, <8 x float> %a1) #0 {
; CHECK-LABEL: @test_x86_avx_hadd_ps_256(
-; CHECK-NEXT: [[TMP1:%.*]] = load <8 x i32>, <8 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i32>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <8 x i32>, <8 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <8 x i32>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <8 x i32>, <8 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i32>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <8 x i32>, <8 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <8 x i32>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[_MSPROP:%.*]] = or <8 x i32> [[TMP1]], [[TMP2]]
; CHECK-NEXT: [[RES:%.*]] = call <8 x float> @llvm.x86.avx.hadd.ps.256(<8 x float> [[A0:%.*]], <8 x float> [[A1:%.*]])
@@ -604,8 +604,8 @@ declare <8 x float> @llvm.x86.avx.hadd.ps.256(<8 x float>, <8 x float>) nounwind
define <4 x double> @test_x86_avx_hsub_pd_256(<4 x double> %a0, <4 x double> %a1) #0 {
; CHECK-LABEL: @test_x86_avx_hsub_pd_256(
-; CHECK-NEXT: [[TMP1:%.*]] = load <4 x i64>, <4 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i64>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <4 x i64>, <4 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <4 x i64>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <4 x i64>, <4 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i64>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <4 x i64>, <4 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <4 x i64>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[_MSPROP:%.*]] = or <4 x i64> [[TMP1]], [[TMP2]]
; CHECK-NEXT: [[RES:%.*]] = call <4 x double> @llvm.x86.avx.hsub.pd.256(<4 x double> [[A0:%.*]], <4 x double> [[A1:%.*]])
@@ -620,8 +620,8 @@ declare <4 x double> @llvm.x86.avx.hsub.pd.256(<4 x double>, <4 x double>) nounw
define <8 x float> @test_x86_avx_hsub_ps_256(<8 x float> %a0, <8 x float> %a1) #0 {
; CHECK-LABEL: @test_x86_avx_hsub_ps_256(
-; CHECK-NEXT: [[TMP1:%.*]] = load <8 x i32>, <8 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i32>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <8 x i32>, <8 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <8 x i32>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <8 x i32>, <8 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i32>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <8 x i32>, <8 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <8 x i32>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[_MSPROP:%.*]] = or <8 x i32> [[TMP1]], [[TMP2]]
; CHECK-NEXT: [[RES:%.*]] = call <8 x float> @llvm.x86.avx.hsub.ps.256(<8 x float> [[A0:%.*]], <8 x float> [[A1:%.*]])
@@ -636,7 +636,7 @@ declare <8 x float> @llvm.x86.avx.hsub.ps.256(<8 x float>, <8 x float>) nounwind
define <32 x i8> @test_x86_avx_ldu_dq_256(i8* %a0) #0 {
; CHECK-LABEL: @test_x86_avx_ldu_dq_256(
-; CHECK-NEXT: [[TMP1:%.*]] = load i64, i64* getelementptr inbounds ([100 x i64], [100 x i64]* @__msan_param_tls, i32 0, i32 0), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load i64, i64* getelementptr inbounds ([100 x i64], [100 x i64]* @__msan_param_tls, i32 0, i32 0), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP2:%.*]] = ptrtoint i8* [[A0:%.*]] to i64
; CHECK-NEXT: [[TMP3:%.*]] = xor i64 [[TMP2]], 87960930222080
@@ -660,8 +660,8 @@ declare <32 x i8> @llvm.x86.avx.ldu.dq.256(i8*) nounwind readonly
define <2 x double> @test_x86_avx_maskload_pd(i8* %a0, <2 x i64> %mask) #0 {
; CHECK-LABEL: @test_x86_avx_maskload_pd(
-; CHECK-NEXT: [[TMP1:%.*]] = load i64, i64* getelementptr inbounds ([100 x i64], [100 x i64]* @__msan_param_tls, i32 0, i32 0), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <2 x i64>, <2 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 8) to <2 x i64>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load i64, i64* getelementptr inbounds ([100 x i64], [100 x i64]* @__msan_param_tls, i32 0, i32 0), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <2 x i64>, <2 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 8) to <2 x i64>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[_MSCMP:%.*]] = icmp ne i64 [[TMP1]], 0
; CHECK-NEXT: br i1 [[_MSCMP]], label [[TMP3:%.*]], label [[TMP4:%.*]], !prof [[PROF0]]
@@ -688,8 +688,8 @@ declare <2 x double> @llvm.x86.avx.maskload.pd(i8*, <2 x i64>) nounwind readonly
define <4 x double> @test_x86_avx_maskload_pd_256(i8* %a0, <4 x i64> %mask) #0 {
; CHECK-LABEL: @test_x86_avx_maskload_pd_256(
-; CHECK-NEXT: [[TMP1:%.*]] = load i64, i64* getelementptr inbounds ([100 x i64], [100 x i64]* @__msan_param_tls, i32 0, i32 0), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <4 x i64>, <4 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 8) to <4 x i64>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load i64, i64* getelementptr inbounds ([100 x i64], [100 x i64]* @__msan_param_tls, i32 0, i32 0), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <4 x i64>, <4 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 8) to <4 x i64>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[_MSCMP:%.*]] = icmp ne i64 [[TMP1]], 0
; CHECK-NEXT: br i1 [[_MSCMP]], label [[TMP3:%.*]], label [[TMP4:%.*]], !prof [[PROF0]]
@@ -716,8 +716,8 @@ declare <4 x double> @llvm.x86.avx.maskload.pd.256(i8*, <4 x i64>) nounwind read
define <4 x float> @test_x86_avx_maskload_ps(i8* %a0, <4 x i32> %mask) #0 {
; CHECK-LABEL: @test_x86_avx_maskload_ps(
-; CHECK-NEXT: [[TMP1:%.*]] = load i64, i64* getelementptr inbounds ([100 x i64], [100 x i64]* @__msan_param_tls, i32 0, i32 0), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <4 x i32>, <4 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 8) to <4 x i32>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load i64, i64* getelementptr inbounds ([100 x i64], [100 x i64]* @__msan_param_tls, i32 0, i32 0), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <4 x i32>, <4 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 8) to <4 x i32>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[_MSCMP:%.*]] = icmp ne i64 [[TMP1]], 0
; CHECK-NEXT: br i1 [[_MSCMP]], label [[TMP3:%.*]], label [[TMP4:%.*]], !prof [[PROF0]]
@@ -744,8 +744,8 @@ declare <4 x float> @llvm.x86.avx.maskload.ps(i8*, <4 x i32>) nounwind readonly
define <8 x float> @test_x86_avx_maskload_ps_256(i8* %a0, <8 x i32> %mask) #0 {
; CHECK-LABEL: @test_x86_avx_maskload_ps_256(
-; CHECK-NEXT: [[TMP1:%.*]] = load i64, i64* getelementptr inbounds ([100 x i64], [100 x i64]* @__msan_param_tls, i32 0, i32 0), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <8 x i32>, <8 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 8) to <8 x i32>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load i64, i64* getelementptr inbounds ([100 x i64], [100 x i64]* @__msan_param_tls, i32 0, i32 0), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <8 x i32>, <8 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 8) to <8 x i32>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[_MSCMP:%.*]] = icmp ne i64 [[TMP1]], 0
; CHECK-NEXT: br i1 [[_MSCMP]], label [[TMP3:%.*]], label [[TMP4:%.*]], !prof [[PROF0]]
@@ -772,9 +772,9 @@ declare <8 x float> @llvm.x86.avx.maskload.ps.256(i8*, <8 x i32>) nounwind reado
define void @test_x86_avx_maskstore_pd(i8* %a0, <2 x i64> %mask, <2 x double> %a2) #0 {
; CHECK-LABEL: @test_x86_avx_maskstore_pd(
-; CHECK-NEXT: [[TMP1:%.*]] = load i64, i64* getelementptr inbounds ([100 x i64], [100 x i64]* @__msan_param_tls, i32 0, i32 0), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <2 x i64>, <2 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 8) to <2 x i64>*), align 8
-; CHECK-NEXT: [[TMP3:%.*]] = load <2 x i64>, <2 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 24) to <2 x i64>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load i64, i64* getelementptr inbounds ([100 x i64], [100 x i64]* @__msan_param_tls, i32 0, i32 0), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <2 x i64>, <2 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 8) to <2 x i64>*), align 8
+; CHECK-DAG: [[TMP3:%.*]] = load <2 x i64>, <2 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 24) to <2 x i64>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[_MSCMP:%.*]] = icmp ne i64 [[TMP1]], 0
; CHECK-NEXT: br i1 [[_MSCMP]], label [[TMP4:%.*]], label [[TMP5:%.*]], !prof [[PROF0]]
@@ -807,9 +807,9 @@ declare void @llvm.x86.avx.maskstore.pd(i8*, <2 x i64>, <2 x double>) nounwind
define void @test_x86_avx_maskstore_pd_256(i8* %a0, <4 x i64> %mask, <4 x double> %a2) #0 {
; CHECK-LABEL: @test_x86_avx_maskstore_pd_256(
-; CHECK-NEXT: [[TMP1:%.*]] = load i64, i64* getelementptr inbounds ([100 x i64], [100 x i64]* @__msan_param_tls, i32 0, i32 0), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <4 x i64>, <4 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 8) to <4 x i64>*), align 8
-; CHECK-NEXT: [[TMP3:%.*]] = load <4 x i64>, <4 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 40) to <4 x i64>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load i64, i64* getelementptr inbounds ([100 x i64], [100 x i64]* @__msan_param_tls, i32 0, i32 0), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <4 x i64>, <4 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 8) to <4 x i64>*), align 8
+; CHECK-DAG: [[TMP3:%.*]] = load <4 x i64>, <4 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 40) to <4 x i64>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[_MSCMP:%.*]] = icmp ne i64 [[TMP1]], 0
; CHECK-NEXT: br i1 [[_MSCMP]], label [[TMP4:%.*]], label [[TMP5:%.*]], !prof [[PROF0]]
@@ -842,9 +842,9 @@ declare void @llvm.x86.avx.maskstore.pd.256(i8*, <4 x i64>, <4 x double>) nounwi
define void @test_x86_avx_maskstore_ps(i8* %a0, <4 x i32> %mask, <4 x float> %a2) #0 {
; CHECK-LABEL: @test_x86_avx_maskstore_ps(
-; CHECK-NEXT: [[TMP1:%.*]] = load i64, i64* getelementptr inbounds ([100 x i64], [100 x i64]* @__msan_param_tls, i32 0, i32 0), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <4 x i32>, <4 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 8) to <4 x i32>*), align 8
-; CHECK-NEXT: [[TMP3:%.*]] = load <4 x i32>, <4 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 24) to <4 x i32>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load i64, i64* getelementptr inbounds ([100 x i64], [100 x i64]* @__msan_param_tls, i32 0, i32 0), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <4 x i32>, <4 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 8) to <4 x i32>*), align 8
+; CHECK-DAG: [[TMP3:%.*]] = load <4 x i32>, <4 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 24) to <4 x i32>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[_MSCMP:%.*]] = icmp ne i64 [[TMP1]], 0
; CHECK-NEXT: br i1 [[_MSCMP]], label [[TMP4:%.*]], label [[TMP5:%.*]], !prof [[PROF0]]
@@ -877,9 +877,9 @@ declare void @llvm.x86.avx.maskstore.ps(i8*, <4 x i32>, <4 x float>) nounwind
define void @test_x86_avx_maskstore_ps_256(i8* %a0, <8 x i32> %mask, <8 x float> %a2) #0 {
; CHECK-LABEL: @test_x86_avx_maskstore_ps_256(
-; CHECK-NEXT: [[TMP1:%.*]] = load i64, i64* getelementptr inbounds ([100 x i64], [100 x i64]* @__msan_param_tls, i32 0, i32 0), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <8 x i32>, <8 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 8) to <8 x i32>*), align 8
-; CHECK-NEXT: [[TMP3:%.*]] = load <8 x i32>, <8 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 40) to <8 x i32>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load i64, i64* getelementptr inbounds ([100 x i64], [100 x i64]* @__msan_param_tls, i32 0, i32 0), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <8 x i32>, <8 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 8) to <8 x i32>*), align 8
+; CHECK-DAG: [[TMP3:%.*]] = load <8 x i32>, <8 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 40) to <8 x i32>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[_MSCMP:%.*]] = icmp ne i64 [[TMP1]], 0
; CHECK-NEXT: br i1 [[_MSCMP]], label [[TMP4:%.*]], label [[TMP5:%.*]], !prof [[PROF0]]
@@ -912,8 +912,8 @@ declare void @llvm.x86.avx.maskstore.ps.256(i8*, <8 x i32>, <8 x float>) nounwin
define <4 x double> @test_x86_avx_max_pd_256(<4 x double> %a0, <4 x double> %a1) #0 {
; CHECK-LABEL: @test_x86_avx_max_pd_256(
-; CHECK-NEXT: [[TMP1:%.*]] = load <4 x i64>, <4 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i64>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <4 x i64>, <4 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <4 x i64>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <4 x i64>, <4 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i64>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <4 x i64>, <4 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <4 x i64>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[_MSPROP:%.*]] = or <4 x i64> [[TMP1]], [[TMP2]]
; CHECK-NEXT: [[RES:%.*]] = call <4 x double> @llvm.x86.avx.max.pd.256(<4 x double> [[A0:%.*]], <4 x double> [[A1:%.*]])
@@ -928,8 +928,8 @@ declare <4 x double> @llvm.x86.avx.max.pd.256(<4 x double>, <4 x double>) nounwi
define <8 x float> @test_x86_avx_max_ps_256(<8 x float> %a0, <8 x float> %a1) #0 {
; CHECK-LABEL: @test_x86_avx_max_ps_256(
-; CHECK-NEXT: [[TMP1:%.*]] = load <8 x i32>, <8 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i32>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <8 x i32>, <8 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <8 x i32>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <8 x i32>, <8 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i32>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <8 x i32>, <8 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <8 x i32>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[_MSPROP:%.*]] = or <8 x i32> [[TMP1]], [[TMP2]]
; CHECK-NEXT: [[RES:%.*]] = call <8 x float> @llvm.x86.avx.max.ps.256(<8 x float> [[A0:%.*]], <8 x float> [[A1:%.*]])
@@ -944,8 +944,8 @@ declare <8 x float> @llvm.x86.avx.max.ps.256(<8 x float>, <8 x float>) nounwind
define <4 x double> @test_x86_avx_min_pd_256(<4 x double> %a0, <4 x double> %a1) #0 {
; CHECK-LABEL: @test_x86_avx_min_pd_256(
-; CHECK-NEXT: [[TMP1:%.*]] = load <4 x i64>, <4 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i64>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <4 x i64>, <4 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <4 x i64>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <4 x i64>, <4 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i64>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <4 x i64>, <4 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <4 x i64>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[_MSPROP:%.*]] = or <4 x i64> [[TMP1]], [[TMP2]]
; CHECK-NEXT: [[RES:%.*]] = call <4 x double> @llvm.x86.avx.min.pd.256(<4 x double> [[A0:%.*]], <4 x double> [[A1:%.*]])
@@ -960,8 +960,8 @@ declare <4 x double> @llvm.x86.avx.min.pd.256(<4 x double>, <4 x double>) nounwi
define <8 x float> @test_x86_avx_min_ps_256(<8 x float> %a0, <8 x float> %a1) #0 {
; CHECK-LABEL: @test_x86_avx_min_ps_256(
-; CHECK-NEXT: [[TMP1:%.*]] = load <8 x i32>, <8 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i32>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <8 x i32>, <8 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <8 x i32>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <8 x i32>, <8 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i32>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <8 x i32>, <8 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <8 x i32>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[_MSPROP:%.*]] = or <8 x i32> [[TMP1]], [[TMP2]]
; CHECK-NEXT: [[RES:%.*]] = call <8 x float> @llvm.x86.avx.min.ps.256(<8 x float> [[A0:%.*]], <8 x float> [[A1:%.*]])
@@ -976,7 +976,7 @@ declare <8 x float> @llvm.x86.avx.min.ps.256(<8 x float>, <8 x float>) nounwind
define i32 @test_x86_avx_movmsk_pd_256(<4 x double> %a0) #0 {
; CHECK-LABEL: @test_x86_avx_movmsk_pd_256(
-; CHECK-NEXT: [[TMP1:%.*]] = load <4 x i64>, <4 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i64>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <4 x i64>, <4 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i64>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP2:%.*]] = bitcast <4 x i64> [[TMP1]] to i256
; CHECK-NEXT: [[_MSCMP:%.*]] = icmp ne i256 [[TMP2]], 0
@@ -997,7 +997,7 @@ declare i32 @llvm.x86.avx.movmsk.pd.256(<4 x double>) nounwind readnone
define i32 @test_x86_avx_movmsk_ps_256(<8 x float> %a0) #0 {
; CHECK-LABEL: @test_x86_avx_movmsk_ps_256(
-; CHECK-NEXT: [[TMP1:%.*]] = load <8 x i32>, <8 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i32>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <8 x i32>, <8 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i32>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP2:%.*]] = bitcast <8 x i32> [[TMP1]] to i256
; CHECK-NEXT: [[_MSCMP:%.*]] = icmp ne i256 [[TMP2]], 0
@@ -1018,8 +1018,8 @@ declare i32 @llvm.x86.avx.movmsk.ps.256(<8 x float>) nounwind readnone
define i32 @test_x86_avx_ptestc_256(<4 x i64> %a0, <4 x i64> %a1) #0 {
; CHECK-LABEL: @test_x86_avx_ptestc_256(
-; CHECK-NEXT: [[TMP1:%.*]] = load <4 x i64>, <4 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i64>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <4 x i64>, <4 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <4 x i64>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <4 x i64>, <4 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i64>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <4 x i64>, <4 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <4 x i64>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP3:%.*]] = bitcast <4 x i64> [[TMP1]] to i256
; CHECK-NEXT: [[_MSCMP:%.*]] = icmp ne i256 [[TMP3]], 0
@@ -1047,8 +1047,8 @@ declare i32 @llvm.x86.avx.ptestc.256(<4 x i64>, <4 x i64>) nounwind readnone
define i32 @test_x86_avx_ptestnzc_256(<4 x i64> %a0, <4 x i64> %a1) #0 {
; CHECK-LABEL: @test_x86_avx_ptestnzc_256(
-; CHECK-NEXT: [[TMP1:%.*]] = load <4 x i64>, <4 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i64>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <4 x i64>, <4 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <4 x i64>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <4 x i64>, <4 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i64>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <4 x i64>, <4 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <4 x i64>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP3:%.*]] = bitcast <4 x i64> [[TMP1]] to i256
; CHECK-NEXT: [[_MSCMP:%.*]] = icmp ne i256 [[TMP3]], 0
@@ -1076,8 +1076,8 @@ declare i32 @llvm.x86.avx.ptestnzc.256(<4 x i64>, <4 x i64>) nounwind readnone
define i32 @test_x86_avx_ptestz_256(<4 x i64> %a0, <4 x i64> %a1) #0 {
; CHECK-LABEL: @test_x86_avx_ptestz_256(
-; CHECK-NEXT: [[TMP1:%.*]] = load <4 x i64>, <4 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i64>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <4 x i64>, <4 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <4 x i64>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <4 x i64>, <4 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i64>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <4 x i64>, <4 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <4 x i64>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP3:%.*]] = bitcast <4 x i64> [[TMP1]] to i256
; CHECK-NEXT: [[_MSCMP:%.*]] = icmp ne i256 [[TMP3]], 0
@@ -1105,7 +1105,7 @@ declare i32 @llvm.x86.avx.ptestz.256(<4 x i64>, <4 x i64>) nounwind readnone
define <8 x float> @test_x86_avx_rcp_ps_256(<8 x float> %a0) #0 {
; CHECK-LABEL: @test_x86_avx_rcp_ps_256(
-; CHECK-NEXT: [[TMP1:%.*]] = load <8 x i32>, <8 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i32>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <8 x i32>, <8 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i32>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[RES:%.*]] = call <8 x float> @llvm.x86.avx.rcp.ps.256(<8 x float> [[A0:%.*]])
; CHECK-NEXT: store <8 x i32> [[TMP1]], <8 x i32>* bitcast ([100 x i64]* @__msan_retval_tls to <8 x i32>*), align 8
@@ -1119,7 +1119,7 @@ declare <8 x float> @llvm.x86.avx.rcp.ps.256(<8 x float>) nounwind readnone
define <4 x double> @test_x86_avx_round_pd_256(<4 x double> %a0) #0 {
; CHECK-LABEL: @test_x86_avx_round_pd_256(
-; CHECK-NEXT: [[TMP1:%.*]] = load <4 x i64>, <4 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i64>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <4 x i64>, <4 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i64>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP2:%.*]] = bitcast <4 x i64> [[TMP1]] to i256
; CHECK-NEXT: [[_MSCMP:%.*]] = icmp ne i256 [[TMP2]], 0
@@ -1140,7 +1140,7 @@ declare <4 x double> @llvm.x86.avx.round.pd.256(<4 x double>, i32) nounwind read
define <8 x float> @test_x86_avx_round_ps_256(<8 x float> %a0) #0 {
; CHECK-LABEL: @test_x86_avx_round_ps_256(
-; CHECK-NEXT: [[TMP1:%.*]] = load <8 x i32>, <8 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i32>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <8 x i32>, <8 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i32>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP2:%.*]] = bitcast <8 x i32> [[TMP1]] to i256
; CHECK-NEXT: [[_MSCMP:%.*]] = icmp ne i256 [[TMP2]], 0
@@ -1161,7 +1161,7 @@ declare <8 x float> @llvm.x86.avx.round.ps.256(<8 x float>, i32) nounwind readno
define <8 x float> @test_x86_avx_rsqrt_ps_256(<8 x float> %a0) #0 {
; CHECK-LABEL: @test_x86_avx_rsqrt_ps_256(
-; CHECK-NEXT: [[TMP1:%.*]] = load <8 x i32>, <8 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i32>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <8 x i32>, <8 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i32>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[RES:%.*]] = call <8 x float> @llvm.x86.avx.rsqrt.ps.256(<8 x float> [[A0:%.*]])
; CHECK-NEXT: store <8 x i32> [[TMP1]], <8 x i32>* bitcast ([100 x i64]* @__msan_retval_tls to <8 x i32>*), align 8
@@ -1174,8 +1174,8 @@ declare <8 x float> @llvm.x86.avx.rsqrt.ps.256(<8 x float>) nounwind readnone
define <2 x double> @test_x86_avx_vpermilvar_pd(<2 x double> %a0, <2 x i64> %a1) #0 {
; CHECK-LABEL: @test_x86_avx_vpermilvar_pd(
-; CHECK-NEXT: [[TMP1:%.*]] = load <2 x i64>, <2 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <2 x i64>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <2 x i64>, <2 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <2 x i64>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <2 x i64>, <2 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <2 x i64>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <2 x i64>, <2 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <2 x i64>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP3:%.*]] = bitcast <2 x i64> [[TMP1]] to i128
; CHECK-NEXT: [[_MSCMP:%.*]] = icmp ne i128 [[TMP3]], 0
@@ -1203,8 +1203,8 @@ declare <2 x double> @llvm.x86.avx.vpermilvar.pd(<2 x double>, <2 x i64>) nounwi
define <4 x double> @test_x86_avx_vpermilvar_pd_256(<4 x double> %a0, <4 x i64> %a1) #0 {
; CHECK-LABEL: @test_x86_avx_vpermilvar_pd_256(
-; CHECK-NEXT: [[TMP1:%.*]] = load <4 x i64>, <4 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i64>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <4 x i64>, <4 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <4 x i64>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <4 x i64>, <4 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i64>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <4 x i64>, <4 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <4 x i64>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP3:%.*]] = bitcast <4 x i64> [[TMP1]] to i256
; CHECK-NEXT: [[_MSCMP:%.*]] = icmp ne i256 [[TMP3]], 0
@@ -1231,7 +1231,7 @@ declare <4 x double> @llvm.x86.avx.vpermilvar.pd.256(<4 x double>, <4 x i64>) no
define <4 x double> @test_x86_avx_vpermilvar_pd_256_2(<4 x double> %a0) #0 {
; CHECK-LABEL: @test_x86_avx_vpermilvar_pd_256_2(
-; CHECK-NEXT: [[TMP1:%.*]] = load <4 x i64>, <4 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i64>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <4 x i64>, <4 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i64>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP2:%.*]] = bitcast <4 x i64> [[TMP1]] to i256
; CHECK-NEXT: [[_MSCMP:%.*]] = icmp ne i256 [[TMP2]], 0
@@ -1250,8 +1250,8 @@ define <4 x double> @test_x86_avx_vpermilvar_pd_256_2(<4 x double> %a0) #0 {
define <4 x float> @test_x86_avx_vpermilvar_ps(<4 x float> %a0, <4 x i32> %a1) #0 {
; CHECK-LABEL: @test_x86_avx_vpermilvar_ps(
-; CHECK-NEXT: [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i32>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <4 x i32>, <4 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <4 x i32>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i32>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <4 x i32>, <4 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <4 x i32>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP3:%.*]] = bitcast <4 x i32> [[TMP1]] to i128
; CHECK-NEXT: [[_MSCMP:%.*]] = icmp ne i128 [[TMP3]], 0
@@ -1276,8 +1276,8 @@ define <4 x float> @test_x86_avx_vpermilvar_ps(<4 x float> %a0, <4 x i32> %a1) #
}
define <4 x float> @test_x86_avx_vpermilvar_ps_load(<4 x float> %a0, <4 x i32>* %a1) #0 {
; CHECK-LABEL: @test_x86_avx_vpermilvar_ps_load(
-; CHECK-NEXT: [[TMP1:%.*]] = load i64, i64* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to i64*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i32>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load i64, i64* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to i64*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i32>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[_MSCMP:%.*]] = icmp ne i64 [[TMP1]], 0
; CHECK-NEXT: br i1 [[_MSCMP]], label [[TMP3:%.*]], label [[TMP4:%.*]], !prof [[PROF0]]
@@ -1317,8 +1317,8 @@ declare <4 x float> @llvm.x86.avx.vpermilvar.ps(<4 x float>, <4 x i32>) nounwind
define <8 x float> @test_x86_avx_vpermilvar_ps_256(<8 x float> %a0, <8 x i32> %a1) #0 {
; CHECK-LABEL: @test_x86_avx_vpermilvar_ps_256(
-; CHECK-NEXT: [[TMP1:%.*]] = load <8 x i32>, <8 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i32>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <8 x i32>, <8 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <8 x i32>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <8 x i32>, <8 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i32>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <8 x i32>, <8 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <8 x i32>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP3:%.*]] = bitcast <8 x i32> [[TMP1]] to i256
; CHECK-NEXT: [[_MSCMP:%.*]] = icmp ne i256 [[TMP3]], 0
@@ -1346,8 +1346,8 @@ declare <8 x float> @llvm.x86.avx.vpermilvar.ps.256(<8 x float>, <8 x i32>) noun
define i32 @test_x86_avx_vtestc_pd(<2 x double> %a0, <2 x double> %a1) #0 {
; CHECK-LABEL: @test_x86_avx_vtestc_pd(
-; CHECK-NEXT: [[TMP1:%.*]] = load <2 x i64>, <2 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <2 x i64>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <2 x i64>, <2 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <2 x i64>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <2 x i64>, <2 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <2 x i64>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <2 x i64>, <2 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <2 x i64>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP3:%.*]] = bitcast <2 x i64> [[TMP1]] to i128
; CHECK-NEXT: [[_MSCMP:%.*]] = icmp ne i128 [[TMP3]], 0
@@ -1375,8 +1375,8 @@ declare i32 @llvm.x86.avx.vtestc.pd(<2 x double>, <2 x double>) nounwind readnon
define i32 @test_x86_avx_vtestc_pd_256(<4 x double> %a0, <4 x double> %a1) #0 {
; CHECK-LABEL: @test_x86_avx_vtestc_pd_256(
-; CHECK-NEXT: [[TMP1:%.*]] = load <4 x i64>, <4 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i64>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <4 x i64>, <4 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <4 x i64>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <4 x i64>, <4 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i64>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <4 x i64>, <4 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <4 x i64>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP3:%.*]] = bitcast <4 x i64> [[TMP1]] to i256
; CHECK-NEXT: [[_MSCMP:%.*]] = icmp ne i256 [[TMP3]], 0
@@ -1404,8 +1404,8 @@ declare i32 @llvm.x86.avx.vtestc.pd.256(<4 x double>, <4 x double>) nounwind rea
define i32 @test_x86_avx_vtestc_ps(<4 x float> %a0, <4 x float> %a1) #0 {
; CHECK-LABEL: @test_x86_avx_vtestc_ps(
-; CHECK-NEXT: [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i32>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <4 x i32>, <4 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <4 x i32>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i32>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <4 x i32>, <4 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <4 x i32>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP3:%.*]] = bitcast <4 x i32> [[TMP1]] to i128
; CHECK-NEXT: [[_MSCMP:%.*]] = icmp ne i128 [[TMP3]], 0
@@ -1433,8 +1433,8 @@ declare i32 @llvm.x86.avx.vtestc.ps(<4 x float>, <4 x float>) nounwind readnone
define i32 @test_x86_avx_vtestc_ps_256(<8 x float> %a0, <8 x float> %a1) #0 {
; CHECK-LABEL: @test_x86_avx_vtestc_ps_256(
-; CHECK-NEXT: [[TMP1:%.*]] = load <8 x i32>, <8 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i32>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <8 x i32>, <8 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <8 x i32>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <8 x i32>, <8 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i32>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <8 x i32>, <8 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <8 x i32>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP3:%.*]] = bitcast <8 x i32> [[TMP1]] to i256
; CHECK-NEXT: [[_MSCMP:%.*]] = icmp ne i256 [[TMP3]], 0
@@ -1462,8 +1462,8 @@ declare i32 @llvm.x86.avx.vtestc.ps.256(<8 x float>, <8 x float>) nounwind readn
define i32 @test_x86_avx_vtestnzc_pd(<2 x double> %a0, <2 x double> %a1) #0 {
; CHECK-LABEL: @test_x86_avx_vtestnzc_pd(
-; CHECK-NEXT: [[TMP1:%.*]] = load <2 x i64>, <2 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <2 x i64>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <2 x i64>, <2 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <2 x i64>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <2 x i64>, <2 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <2 x i64>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <2 x i64>, <2 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <2 x i64>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP3:%.*]] = bitcast <2 x i64> [[TMP1]] to i128
; CHECK-NEXT: [[_MSCMP:%.*]] = icmp ne i128 [[TMP3]], 0
@@ -1491,8 +1491,8 @@ declare i32 @llvm.x86.avx.vtestnzc.pd(<2 x double>, <2 x double>) nounwind readn
define i32 @test_x86_avx_vtestnzc_pd_256(<4 x double> %a0, <4 x double> %a1) #0 {
; CHECK-LABEL: @test_x86_avx_vtestnzc_pd_256(
-; CHECK-NEXT: [[TMP1:%.*]] = load <4 x i64>, <4 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i64>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <4 x i64>, <4 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <4 x i64>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <4 x i64>, <4 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i64>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <4 x i64>, <4 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <4 x i64>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP3:%.*]] = bitcast <4 x i64> [[TMP1]] to i256
; CHECK-NEXT: [[_MSCMP:%.*]] = icmp ne i256 [[TMP3]], 0
@@ -1520,8 +1520,8 @@ declare i32 @llvm.x86.avx.vtestnzc.pd.256(<4 x double>, <4 x double>) nounwind r
define i32 @test_x86_avx_vtestnzc_ps(<4 x float> %a0, <4 x float> %a1) #0 {
; CHECK-LABEL: @test_x86_avx_vtestnzc_ps(
-; CHECK-NEXT: [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i32>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <4 x i32>, <4 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <4 x i32>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i32>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <4 x i32>, <4 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <4 x i32>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP3:%.*]] = bitcast <4 x i32> [[TMP1]] to i128
; CHECK-NEXT: [[_MSCMP:%.*]] = icmp ne i128 [[TMP3]], 0
@@ -1549,8 +1549,8 @@ declare i32 @llvm.x86.avx.vtestnzc.ps(<4 x float>, <4 x float>) nounwind readnon
define i32 @test_x86_avx_vtestnzc_ps_256(<8 x float> %a0, <8 x float> %a1) #0 {
; CHECK-LABEL: @test_x86_avx_vtestnzc_ps_256(
-; CHECK-NEXT: [[TMP1:%.*]] = load <8 x i32>, <8 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i32>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <8 x i32>, <8 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <8 x i32>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <8 x i32>, <8 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i32>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <8 x i32>, <8 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <8 x i32>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP3:%.*]] = bitcast <8 x i32> [[TMP1]] to i256
; CHECK-NEXT: [[_MSCMP:%.*]] = icmp ne i256 [[TMP3]], 0
@@ -1578,8 +1578,8 @@ declare i32 @llvm.x86.avx.vtestnzc.ps.256(<8 x float>, <8 x float>) nounwind rea
define i32 @test_x86_avx_vtestz_pd(<2 x double> %a0, <2 x double> %a1) #0 {
; CHECK-LABEL: @test_x86_avx_vtestz_pd(
-; CHECK-NEXT: [[TMP1:%.*]] = load <2 x i64>, <2 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <2 x i64>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <2 x i64>, <2 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <2 x i64>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <2 x i64>, <2 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <2 x i64>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <2 x i64>, <2 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <2 x i64>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP3:%.*]] = bitcast <2 x i64> [[TMP1]] to i128
; CHECK-NEXT: [[_MSCMP:%.*]] = icmp ne i128 [[TMP3]], 0
@@ -1607,8 +1607,8 @@ declare i32 @llvm.x86.avx.vtestz.pd(<2 x double>, <2 x double>) nounwind readnon
define i32 @test_x86_avx_vtestz_pd_256(<4 x double> %a0, <4 x double> %a1) #0 {
; CHECK-LABEL: @test_x86_avx_vtestz_pd_256(
-; CHECK-NEXT: [[TMP1:%.*]] = load <4 x i64>, <4 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i64>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <4 x i64>, <4 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <4 x i64>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <4 x i64>, <4 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i64>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <4 x i64>, <4 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <4 x i64>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP3:%.*]] = bitcast <4 x i64> [[TMP1]] to i256
; CHECK-NEXT: [[_MSCMP:%.*]] = icmp ne i256 [[TMP3]], 0
@@ -1636,8 +1636,8 @@ declare i32 @llvm.x86.avx.vtestz.pd.256(<4 x double>, <4 x double>) nounwind rea
define i32 @test_x86_avx_vtestz_ps(<4 x float> %a0, <4 x float> %a1) #0 {
; CHECK-LABEL: @test_x86_avx_vtestz_ps(
-; CHECK-NEXT: [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i32>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <4 x i32>, <4 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <4 x i32>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i32>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <4 x i32>, <4 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <4 x i32>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP3:%.*]] = bitcast <4 x i32> [[TMP1]] to i128
; CHECK-NEXT: [[_MSCMP:%.*]] = icmp ne i128 [[TMP3]], 0
@@ -1665,8 +1665,8 @@ declare i32 @llvm.x86.avx.vtestz.ps(<4 x float>, <4 x float>) nounwind readnone
define i32 @test_x86_avx_vtestz_ps_256(<8 x float> %a0, <8 x float> %a1) #0 {
; CHECK-LABEL: @test_x86_avx_vtestz_ps_256(
-; CHECK-NEXT: [[TMP1:%.*]] = load <8 x i32>, <8 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i32>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <8 x i32>, <8 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <8 x i32>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <8 x i32>, <8 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i32>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <8 x i32>, <8 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <8 x i32>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP3:%.*]] = bitcast <8 x i32> [[TMP1]] to i256
; CHECK-NEXT: [[_MSCMP:%.*]] = icmp ne i256 [[TMP3]], 0
@@ -1717,8 +1717,8 @@ declare void @llvm.x86.avx.vzeroupper() nounwind
define void @movnt_dq(i8* %p, <2 x i64> %a1) nounwind #0 {
; CHECK-LABEL: @movnt_dq(
-; CHECK-NEXT: [[TMP1:%.*]] = load <2 x i64>, <2 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 8) to <2 x i64>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load i64, i64* getelementptr inbounds ([100 x i64], [100 x i64]* @__msan_param_tls, i32 0, i32 0), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <2 x i64>, <2 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 8) to <2 x i64>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load i64, i64* getelementptr inbounds ([100 x i64], [100 x i64]* @__msan_param_tls, i32 0, i32 0), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[_MSPROP:%.*]] = or <2 x i64> [[TMP1]], zeroinitializer
; CHECK-NEXT: [[A2:%.*]] = add <2 x i64> [[A1:%.*]], <i64 1, i64 1>
@@ -1747,8 +1747,8 @@ declare void @llvm.x86.avx.movnt.dq.256(i8*, <4 x i64>) nounwind
define void @movnt_ps(i8* %p, <8 x float> %a) nounwind #0 {
; CHECK-LABEL: @movnt_ps(
-; CHECK-NEXT: [[TMP1:%.*]] = load i64, i64* getelementptr inbounds ([100 x i64], [100 x i64]* @__msan_param_tls, i32 0, i32 0), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <8 x i32>, <8 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 8) to <8 x i32>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load i64, i64* getelementptr inbounds ([100 x i64], [100 x i64]* @__msan_param_tls, i32 0, i32 0), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <8 x i32>, <8 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 8) to <8 x i32>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[CAST:%.*]] = bitcast i8* [[P:%.*]] to <8 x float>*
; CHECK-NEXT: [[_MSCMP:%.*]] = icmp ne i64 [[TMP1]], 0
@@ -1772,8 +1772,8 @@ declare void @llvm.x86.avx.movnt.ps.256(i8*, <8 x float>) nounwind
define void @movnt_pd(i8* %p, <4 x double> %a1) nounwind #0 {
; add operation forces the execution domain.
; CHECK-LABEL: @movnt_pd(
-; CHECK-NEXT: [[TMP1:%.*]] = load <4 x i64>, <4 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 8) to <4 x i64>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load i64, i64* getelementptr inbounds ([100 x i64], [100 x i64]* @__msan_param_tls, i32 0, i32 0), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <4 x i64>, <4 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 8) to <4 x i64>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load i64, i64* getelementptr inbounds ([100 x i64], [100 x i64]* @__msan_param_tls, i32 0, i32 0), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[_MSPROP:%.*]] = or <4 x i64> [[TMP1]], zeroinitializer
; CHECK-NEXT: [[A2:%.*]] = fadd <4 x double> [[A1:%.*]], zeroinitializer
@@ -1800,8 +1800,8 @@ declare void @llvm.x86.avx.movnt.pd.256(i8*, <4 x double>) nounwind
define <2 x i64> @test_x86_pclmulqdq(<2 x i64> %a0, <2 x i64> %a1) #0 {
; CHECK-LABEL: @test_x86_pclmulqdq(
-; CHECK-NEXT: [[TMP1:%.*]] = load <2 x i64>, <2 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <2 x i64>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <2 x i64>, <2 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <2 x i64>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <2 x i64>, <2 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <2 x i64>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <2 x i64>, <2 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <2 x i64>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP3:%.*]] = shufflevector <2 x i64> [[TMP1]], <2 x i64> poison, <2 x i32> zeroinitializer
; CHECK-NEXT: [[TMP4:%.*]] = shufflevector <2 x i64> [[TMP2]], <2 x i64> poison, <2 x i32> zeroinitializer
diff --git a/llvm/test/Instrumentation/MemorySanitizer/avx2-intrinsics-x86.ll b/llvm/test/Instrumentation/MemorySanitizer/avx2-intrinsics-x86.ll
index 7c896c6ff008..31b9ff926467 100644
--- a/llvm/test/Instrumentation/MemorySanitizer/avx2-intrinsics-x86.ll
+++ b/llvm/test/Instrumentation/MemorySanitizer/avx2-intrinsics-x86.ll
@@ -6,8 +6,8 @@ target triple = "x86_64-unknown-linux-gnu"
define <16 x i16> @test_x86_avx2_packssdw(<8 x i32> %a0, <8 x i32> %a1) #0 {
; CHECK-LABEL: @test_x86_avx2_packssdw(
-; CHECK-NEXT: [[TMP1:%.*]] = load <8 x i32>, <8 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i32>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <8 x i32>, <8 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <8 x i32>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <8 x i32>, <8 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i32>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <8 x i32>, <8 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <8 x i32>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP3:%.*]] = icmp ne <8 x i32> [[TMP1]], zeroinitializer
; CHECK-NEXT: [[TMP4:%.*]] = sext <8 x i1> [[TMP3]] to <8 x i32>
@@ -39,8 +39,8 @@ define <16 x i16> @test_x86_avx2_packssdw_fold() #0 {
define <32 x i8> @test_x86_avx2_packsswb(<16 x i16> %a0, <16 x i16> %a1) #0 {
; CHECK-LABEL: @test_x86_avx2_packsswb(
-; CHECK-NEXT: [[TMP1:%.*]] = load <16 x i16>, <16 x i16>* bitcast ([100 x i64]* @__msan_param_tls to <16 x i16>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <16 x i16>, <16 x i16>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <16 x i16>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <16 x i16>, <16 x i16>* bitcast ([100 x i64]* @__msan_param_tls to <16 x i16>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <16 x i16>, <16 x i16>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <16 x i16>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP3:%.*]] = icmp ne <16 x i16> [[TMP1]], zeroinitializer
; CHECK-NEXT: [[TMP4:%.*]] = sext <16 x i1> [[TMP3]] to <16 x i16>
@@ -72,8 +72,8 @@ define <32 x i8> @test_x86_avx2_packsswb_fold() #0 {
define <32 x i8> @test_x86_avx2_packuswb(<16 x i16> %a0, <16 x i16> %a1) #0 {
; CHECK-LABEL: @test_x86_avx2_packuswb(
-; CHECK-NEXT: [[TMP1:%.*]] = load <16 x i16>, <16 x i16>* bitcast ([100 x i64]* @__msan_param_tls to <16 x i16>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <16 x i16>, <16 x i16>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <16 x i16>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <16 x i16>, <16 x i16>* bitcast ([100 x i64]* @__msan_param_tls to <16 x i16>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <16 x i16>, <16 x i16>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <16 x i16>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP3:%.*]] = icmp ne <16 x i16> [[TMP1]], zeroinitializer
; CHECK-NEXT: [[TMP4:%.*]] = sext <16 x i1> [[TMP3]] to <16 x i16>
@@ -105,8 +105,8 @@ define <32 x i8> @test_x86_avx2_packuswb_fold() #0 {
define <32 x i8> @test_x86_avx2_pavg_b(<32 x i8> %a0, <32 x i8> %a1) #0 {
; CHECK-LABEL: @test_x86_avx2_pavg_b(
-; CHECK-NEXT: [[TMP1:%.*]] = load <32 x i8>, <32 x i8>* bitcast ([100 x i64]* @__msan_param_tls to <32 x i8>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <32 x i8>, <32 x i8>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <32 x i8>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <32 x i8>, <32 x i8>* bitcast ([100 x i64]* @__msan_param_tls to <32 x i8>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <32 x i8>, <32 x i8>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <32 x i8>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[_MSPROP:%.*]] = or <32 x i8> [[TMP1]], [[TMP2]]
; CHECK-NEXT: [[RES:%.*]] = call <32 x i8> @llvm.x86.avx2.pavg.b(<32 x i8> [[A0:%.*]], <32 x i8> [[A1:%.*]])
@@ -121,8 +121,8 @@ 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) #0 {
; CHECK-LABEL: @test_x86_avx2_pavg_w(
-; CHECK-NEXT: [[TMP1:%.*]] = load <16 x i16>, <16 x i16>* bitcast ([100 x i64]* @__msan_param_tls to <16 x i16>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <16 x i16>, <16 x i16>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <16 x i16>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <16 x i16>, <16 x i16>* bitcast ([100 x i64]* @__msan_param_tls to <16 x i16>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <16 x i16>, <16 x i16>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <16 x i16>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[_MSPROP:%.*]] = or <16 x i16> [[TMP1]], [[TMP2]]
; CHECK-NEXT: [[RES:%.*]] = call <16 x i16> @llvm.x86.avx2.pavg.w(<16 x i16> [[A0:%.*]], <16 x i16> [[A1:%.*]])
@@ -137,8 +137,8 @@ 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) #0 {
; CHECK-LABEL: @test_x86_avx2_pmadd_wd(
-; CHECK-NEXT: [[TMP1:%.*]] = load <16 x i16>, <16 x i16>* bitcast ([100 x i64]* @__msan_param_tls to <16 x i16>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <16 x i16>, <16 x i16>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <16 x i16>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <16 x i16>, <16 x i16>* bitcast ([100 x i64]* @__msan_param_tls to <16 x i16>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <16 x i16>, <16 x i16>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <16 x i16>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP3:%.*]] = or <16 x i16> [[TMP1]], [[TMP2]]
; CHECK-NEXT: [[TMP4:%.*]] = bitcast <16 x i16> [[TMP3]] to <8 x i32>
@@ -156,7 +156,7 @@ declare <8 x i32> @llvm.x86.avx2.pmadd.wd(<16 x i16>, <16 x i16>) nounwind readn
define i32 @test_x86_avx2_pmovmskb(<32 x i8> %a0) #0 {
; CHECK-LABEL: @test_x86_avx2_pmovmskb(
-; CHECK-NEXT: [[TMP1:%.*]] = load <32 x i8>, <32 x i8>* bitcast ([100 x i64]* @__msan_param_tls to <32 x i8>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <32 x i8>, <32 x i8>* bitcast ([100 x i64]* @__msan_param_tls to <32 x i8>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP2:%.*]] = bitcast <32 x i8> [[TMP1]] to i256
; CHECK-NEXT: [[_MSCMP:%.*]] = icmp ne i256 [[TMP2]], 0
@@ -177,8 +177,8 @@ 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) #0 {
; CHECK-LABEL: @test_x86_avx2_pmulh_w(
-; CHECK-NEXT: [[TMP1:%.*]] = load <16 x i16>, <16 x i16>* bitcast ([100 x i64]* @__msan_param_tls to <16 x i16>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <16 x i16>, <16 x i16>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <16 x i16>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <16 x i16>, <16 x i16>* bitcast ([100 x i64]* @__msan_param_tls to <16 x i16>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <16 x i16>, <16 x i16>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <16 x i16>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[_MSPROP:%.*]] = or <16 x i16> [[TMP1]], [[TMP2]]
; CHECK-NEXT: [[RES:%.*]] = call <16 x i16> @llvm.x86.avx2.pmulh.w(<16 x i16> [[A0:%.*]], <16 x i16> [[A1:%.*]])
@@ -193,8 +193,8 @@ 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) #0 {
; CHECK-LABEL: @test_x86_avx2_pmulhu_w(
-; CHECK-NEXT: [[TMP1:%.*]] = load <16 x i16>, <16 x i16>* bitcast ([100 x i64]* @__msan_param_tls to <16 x i16>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <16 x i16>, <16 x i16>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <16 x i16>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <16 x i16>, <16 x i16>* bitcast ([100 x i64]* @__msan_param_tls to <16 x i16>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <16 x i16>, <16 x i16>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <16 x i16>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[_MSPROP:%.*]] = or <16 x i16> [[TMP1]], [[TMP2]]
; CHECK-NEXT: [[RES:%.*]] = call <16 x i16> @llvm.x86.avx2.pmulhu.w(<16 x i16> [[A0:%.*]], <16 x i16> [[A1:%.*]])
@@ -209,8 +209,8 @@ declare <16 x i16> @llvm.x86.avx2.pmulhu.w(<16 x i16>, <16 x i16>) nounwind read
define <4 x i64> @test_x86_avx2_psad_bw(<32 x i8> %a0, <32 x i8> %a1) #0 {
; CHECK-LABEL: @test_x86_avx2_psad_bw(
-; CHECK-NEXT: [[TMP1:%.*]] = load <32 x i8>, <32 x i8>* bitcast ([100 x i64]* @__msan_param_tls to <32 x i8>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <32 x i8>, <32 x i8>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <32 x i8>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <32 x i8>, <32 x i8>* bitcast ([100 x i64]* @__msan_param_tls to <32 x i8>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <32 x i8>, <32 x i8>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <32 x i8>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP3:%.*]] = or <32 x i8> [[TMP1]], [[TMP2]]
; CHECK-NEXT: [[TMP4:%.*]] = bitcast <32 x i8> [[TMP3]] to <4 x i64>
@@ -229,8 +229,8 @@ 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) #0 {
; CHECK-LABEL: @test_x86_avx2_psll_d(
-; CHECK-NEXT: [[TMP1:%.*]] = load <8 x i32>, <8 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i32>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <4 x i32>, <4 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <4 x i32>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <8 x i32>, <8 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i32>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <4 x i32>, <4 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <4 x i32>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP3:%.*]] = bitcast <4 x i32> [[TMP2]] to i128
; CHECK-NEXT: [[TMP4:%.*]] = trunc i128 [[TMP3]] to i64
@@ -251,8 +251,8 @@ 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) #0 {
; CHECK-LABEL: @test_x86_avx2_psll_q(
-; CHECK-NEXT: [[TMP1:%.*]] = load <4 x i64>, <4 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i64>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <2 x i64>, <2 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <2 x i64>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <4 x i64>, <4 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i64>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <2 x i64>, <2 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <2 x i64>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP3:%.*]] = bitcast <2 x i64> [[TMP2]] to i128
; CHECK-NEXT: [[TMP4:%.*]] = trunc i128 [[TMP3]] to i64
@@ -273,8 +273,8 @@ 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) #0 {
; CHECK-LABEL: @test_x86_avx2_psll_w(
-; CHECK-NEXT: [[TMP1:%.*]] = load <16 x i16>, <16 x i16>* bitcast ([100 x i64]* @__msan_param_tls to <16 x i16>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <8 x i16>, <8 x i16>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <8 x i16>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <16 x i16>, <16 x i16>* bitcast ([100 x i64]* @__msan_param_tls to <16 x i16>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <8 x i16>, <8 x i16>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <8 x i16>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP3:%.*]] = bitcast <8 x i16> [[TMP2]] to i128
; CHECK-NEXT: [[TMP4:%.*]] = trunc i128 [[TMP3]] to i64
@@ -295,7 +295,7 @@ 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) #0 {
; CHECK-LABEL: @test_x86_avx2_pslli_d(
-; CHECK-NEXT: [[TMP1:%.*]] = load <8 x i32>, <8 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i32>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <8 x i32>, <8 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i32>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP2:%.*]] = call <8 x i32> @llvm.x86.avx2.pslli.d(<8 x i32> [[TMP1]], i32 7)
; CHECK-NEXT: [[TMP3:%.*]] = or <8 x i32> [[TMP2]], zeroinitializer
@@ -311,7 +311,7 @@ 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) #0 {
; CHECK-LABEL: @test_x86_avx2_pslli_q(
-; CHECK-NEXT: [[TMP1:%.*]] = load <4 x i64>, <4 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i64>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <4 x i64>, <4 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i64>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP2:%.*]] = call <4 x i64> @llvm.x86.avx2.pslli.q(<4 x i64> [[TMP1]], i32 7)
; CHECK-NEXT: [[TMP3:%.*]] = or <4 x i64> [[TMP2]], zeroinitializer
@@ -327,7 +327,7 @@ 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) #0 {
; CHECK-LABEL: @test_x86_avx2_pslli_w(
-; CHECK-NEXT: [[TMP1:%.*]] = load <16 x i16>, <16 x i16>* bitcast ([100 x i64]* @__msan_param_tls to <16 x i16>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <16 x i16>, <16 x i16>* bitcast ([100 x i64]* @__msan_param_tls to <16 x i16>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP2:%.*]] = call <16 x i16> @llvm.x86.avx2.pslli.w(<16 x i16> [[TMP1]], i32 7)
; CHECK-NEXT: [[TMP3:%.*]] = or <16 x i16> [[TMP2]], zeroinitializer
@@ -343,8 +343,8 @@ 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) #0 {
; CHECK-LABEL: @test_x86_avx2_psra_d(
-; CHECK-NEXT: [[TMP1:%.*]] = load <8 x i32>, <8 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i32>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <4 x i32>, <4 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <4 x i32>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <8 x i32>, <8 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i32>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <4 x i32>, <4 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <4 x i32>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP3:%.*]] = bitcast <4 x i32> [[TMP2]] to i128
; CHECK-NEXT: [[TMP4:%.*]] = trunc i128 [[TMP3]] to i64
@@ -365,8 +365,8 @@ 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) #0 {
; CHECK-LABEL: @test_x86_avx2_psra_w(
-; CHECK-NEXT: [[TMP1:%.*]] = load <16 x i16>, <16 x i16>* bitcast ([100 x i64]* @__msan_param_tls to <16 x i16>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <8 x i16>, <8 x i16>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <8 x i16>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <16 x i16>, <16 x i16>* bitcast ([100 x i64]* @__msan_param_tls to <16 x i16>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <8 x i16>, <8 x i16>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <8 x i16>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP3:%.*]] = bitcast <8 x i16> [[TMP2]] to i128
; CHECK-NEXT: [[TMP4:%.*]] = trunc i128 [[TMP3]] to i64
@@ -387,7 +387,7 @@ 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) #0 {
; CHECK-LABEL: @test_x86_avx2_psrai_d(
-; CHECK-NEXT: [[TMP1:%.*]] = load <8 x i32>, <8 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i32>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <8 x i32>, <8 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i32>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP2:%.*]] = call <8 x i32> @llvm.x86.avx2.psrai.d(<8 x i32> [[TMP1]], i32 7)
; CHECK-NEXT: [[TMP3:%.*]] = or <8 x i32> [[TMP2]], zeroinitializer
@@ -403,7 +403,7 @@ 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) #0 {
; CHECK-LABEL: @test_x86_avx2_psrai_w(
-; CHECK-NEXT: [[TMP1:%.*]] = load <16 x i16>, <16 x i16>* bitcast ([100 x i64]* @__msan_param_tls to <16 x i16>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <16 x i16>, <16 x i16>* bitcast ([100 x i64]* @__msan_param_tls to <16 x i16>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP2:%.*]] = call <16 x i16> @llvm.x86.avx2.psrai.w(<16 x i16> [[TMP1]], i32 7)
; CHECK-NEXT: [[TMP3:%.*]] = or <16 x i16> [[TMP2]], zeroinitializer
@@ -419,8 +419,8 @@ 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) #0 {
; CHECK-LABEL: @test_x86_avx2_psrl_d(
-; CHECK-NEXT: [[TMP1:%.*]] = load <8 x i32>, <8 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i32>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <4 x i32>, <4 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <4 x i32>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <8 x i32>, <8 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i32>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <4 x i32>, <4 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <4 x i32>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP3:%.*]] = bitcast <4 x i32> [[TMP2]] to i128
; CHECK-NEXT: [[TMP4:%.*]] = trunc i128 [[TMP3]] to i64
@@ -441,8 +441,8 @@ 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) #0 {
; CHECK-LABEL: @test_x86_avx2_psrl_q(
-; CHECK-NEXT: [[TMP1:%.*]] = load <4 x i64>, <4 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i64>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <2 x i64>, <2 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <2 x i64>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <4 x i64>, <4 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i64>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <2 x i64>, <2 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <2 x i64>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP3:%.*]] = bitcast <2 x i64> [[TMP2]] to i128
; CHECK-NEXT: [[TMP4:%.*]] = trunc i128 [[TMP3]] to i64
@@ -463,8 +463,8 @@ 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) #0 {
; CHECK-LABEL: @test_x86_avx2_psrl_w(
-; CHECK-NEXT: [[TMP1:%.*]] = load <16 x i16>, <16 x i16>* bitcast ([100 x i64]* @__msan_param_tls to <16 x i16>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <8 x i16>, <8 x i16>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <8 x i16>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <16 x i16>, <16 x i16>* bitcast ([100 x i64]* @__msan_param_tls to <16 x i16>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <8 x i16>, <8 x i16>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <8 x i16>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP3:%.*]] = bitcast <8 x i16> [[TMP2]] to i128
; CHECK-NEXT: [[TMP4:%.*]] = trunc i128 [[TMP3]] to i64
@@ -485,8 +485,8 @@ declare <16 x i16> @llvm.x86.avx2.psrl.w(<16 x i16>, <8 x i16>) nounwind readnon
define <16 x i16> @test_x86_avx2_psrl_w_load(<16 x i16> %a0, <8 x i16>* %p) #0 {
; CHECK-LABEL: @test_x86_avx2_psrl_w_load(
-; CHECK-NEXT: [[TMP1:%.*]] = load i64, i64* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to i64*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <16 x i16>, <16 x i16>* bitcast ([100 x i64]* @__msan_param_tls to <16 x i16>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load i64, i64* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to i64*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <16 x i16>, <16 x i16>* bitcast ([100 x i64]* @__msan_param_tls to <16 x i16>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[_MSCMP:%.*]] = icmp ne i64 [[TMP1]], 0
; CHECK-NEXT: br i1 [[_MSCMP]], label [[TMP3:%.*]], label [[TMP4:%.*]], !prof [[PROF0]]
@@ -518,7 +518,7 @@ define <16 x i16> @test_x86_avx2_psrl_w_load(<16 x i16> %a0, <8 x i16>* %p) #0 {
define <8 x i32> @test_x86_avx2_psrli_d(<8 x i32> %a0) #0 {
; CHECK-LABEL: @test_x86_avx2_psrli_d(
-; CHECK-NEXT: [[TMP1:%.*]] = load <8 x i32>, <8 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i32>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <8 x i32>, <8 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i32>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP2:%.*]] = call <8 x i32> @llvm.x86.avx2.psrli.d(<8 x i32> [[TMP1]], i32 7)
; CHECK-NEXT: [[TMP3:%.*]] = or <8 x i32> [[TMP2]], zeroinitializer
@@ -534,7 +534,7 @@ 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) #0 {
; CHECK-LABEL: @test_x86_avx2_psrli_q(
-; CHECK-NEXT: [[TMP1:%.*]] = load <4 x i64>, <4 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i64>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <4 x i64>, <4 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i64>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP2:%.*]] = call <4 x i64> @llvm.x86.avx2.psrli.q(<4 x i64> [[TMP1]], i32 7)
; CHECK-NEXT: [[TMP3:%.*]] = or <4 x i64> [[TMP2]], zeroinitializer
@@ -550,7 +550,7 @@ 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) #0 {
; CHECK-LABEL: @test_x86_avx2_psrli_w(
-; CHECK-NEXT: [[TMP1:%.*]] = load <16 x i16>, <16 x i16>* bitcast ([100 x i64]* @__msan_param_tls to <16 x i16>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <16 x i16>, <16 x i16>* bitcast ([100 x i64]* @__msan_param_tls to <16 x i16>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP2:%.*]] = call <16 x i16> @llvm.x86.avx2.psrli.w(<16 x i16> [[TMP1]], i32 7)
; CHECK-NEXT: [[TMP3:%.*]] = or <16 x i16> [[TMP2]], zeroinitializer
@@ -566,8 +566,8 @@ declare <16 x i16> @llvm.x86.avx2.psrli.w(<16 x i16>, i32) nounwind readnone
define <8 x i32> @test_x86_avx2_phadd_d(<8 x i32> %a0, <8 x i32> %a1) #0 {
; CHECK-LABEL: @test_x86_avx2_phadd_d(
-; CHECK-NEXT: [[TMP1:%.*]] = load <8 x i32>, <8 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i32>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <8 x i32>, <8 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <8 x i32>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <8 x i32>, <8 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i32>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <8 x i32>, <8 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <8 x i32>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[_MSPROP:%.*]] = or <8 x i32> [[TMP1]], [[TMP2]]
; CHECK-NEXT: [[RES:%.*]] = call <8 x i32> @llvm.x86.avx2.phadd.d(<8 x i32> [[A0:%.*]], <8 x i32> [[A1:%.*]])
@@ -582,8 +582,8 @@ 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) #0 {
; CHECK-LABEL: @test_x86_avx2_phadd_sw(
-; CHECK-NEXT: [[TMP1:%.*]] = load <16 x i16>, <16 x i16>* bitcast ([100 x i64]* @__msan_param_tls to <16 x i16>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <16 x i16>, <16 x i16>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <16 x i16>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <16 x i16>, <16 x i16>* bitcast ([100 x i64]* @__msan_param_tls to <16 x i16>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <16 x i16>, <16 x i16>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <16 x i16>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[_MSPROP:%.*]] = or <16 x i16> [[TMP1]], [[TMP2]]
; CHECK-NEXT: [[RES:%.*]] = call <16 x i16> @llvm.x86.avx2.phadd.sw(<16 x i16> [[A0:%.*]], <16 x i16> [[A1:%.*]])
@@ -598,8 +598,8 @@ 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) #0 {
; CHECK-LABEL: @test_x86_avx2_phadd_w(
-; CHECK-NEXT: [[TMP1:%.*]] = load <16 x i16>, <16 x i16>* bitcast ([100 x i64]* @__msan_param_tls to <16 x i16>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <16 x i16>, <16 x i16>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <16 x i16>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <16 x i16>, <16 x i16>* bitcast ([100 x i64]* @__msan_param_tls to <16 x i16>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <16 x i16>, <16 x i16>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <16 x i16>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[_MSPROP:%.*]] = or <16 x i16> [[TMP1]], [[TMP2]]
; CHECK-NEXT: [[RES:%.*]] = call <16 x i16> @llvm.x86.avx2.phadd.w(<16 x i16> [[A0:%.*]], <16 x i16> [[A1:%.*]])
@@ -614,8 +614,8 @@ 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) #0 {
; CHECK-LABEL: @test_x86_avx2_phsub_d(
-; CHECK-NEXT: [[TMP1:%.*]] = load <8 x i32>, <8 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i32>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <8 x i32>, <8 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <8 x i32>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <8 x i32>, <8 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i32>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <8 x i32>, <8 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <8 x i32>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[_MSPROP:%.*]] = or <8 x i32> [[TMP1]], [[TMP2]]
; CHECK-NEXT: [[RES:%.*]] = call <8 x i32> @llvm.x86.avx2.phsub.d(<8 x i32> [[A0:%.*]], <8 x i32> [[A1:%.*]])
@@ -630,8 +630,8 @@ 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) #0 {
; CHECK-LABEL: @test_x86_avx2_phsub_sw(
-; CHECK-NEXT: [[TMP1:%.*]] = load <16 x i16>, <16 x i16>* bitcast ([100 x i64]* @__msan_param_tls to <16 x i16>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <16 x i16>, <16 x i16>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <16 x i16>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <16 x i16>, <16 x i16>* bitcast ([100 x i64]* @__msan_param_tls to <16 x i16>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <16 x i16>, <16 x i16>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <16 x i16>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[_MSPROP:%.*]] = or <16 x i16> [[TMP1]], [[TMP2]]
; CHECK-NEXT: [[RES:%.*]] = call <16 x i16> @llvm.x86.avx2.phsub.sw(<16 x i16> [[A0:%.*]], <16 x i16> [[A1:%.*]])
@@ -646,8 +646,8 @@ 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) #0 {
; CHECK-LABEL: @test_x86_avx2_phsub_w(
-; CHECK-NEXT: [[TMP1:%.*]] = load <16 x i16>, <16 x i16>* bitcast ([100 x i64]* @__msan_param_tls to <16 x i16>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <16 x i16>, <16 x i16>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <16 x i16>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <16 x i16>, <16 x i16>* bitcast ([100 x i64]* @__msan_param_tls to <16 x i16>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <16 x i16>, <16 x i16>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <16 x i16>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[_MSPROP:%.*]] = or <16 x i16> [[TMP1]], [[TMP2]]
; CHECK-NEXT: [[RES:%.*]] = call <16 x i16> @llvm.x86.avx2.phsub.w(<16 x i16> [[A0:%.*]], <16 x i16> [[A1:%.*]])
@@ -662,8 +662,8 @@ 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) #0 {
; CHECK-LABEL: @test_x86_avx2_pmadd_ub_sw(
-; CHECK-NEXT: [[TMP1:%.*]] = load <32 x i8>, <32 x i8>* bitcast ([100 x i64]* @__msan_param_tls to <32 x i8>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <32 x i8>, <32 x i8>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <32 x i8>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <32 x i8>, <32 x i8>* bitcast ([100 x i64]* @__msan_param_tls to <32 x i8>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <32 x i8>, <32 x i8>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <32 x i8>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP3:%.*]] = or <32 x i8> [[TMP1]], [[TMP2]]
; CHECK-NEXT: [[TMP4:%.*]] = bitcast <32 x i8> [[TMP3]] to <16 x i16>
@@ -680,8 +680,8 @@ 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_pmadd_ub_sw_load_op0(<32 x i8>* %ptr, <32 x i8> %a1) #0 {
; CHECK-LABEL: @test_x86_avx2_pmadd_ub_sw_load_op0(
-; CHECK-NEXT: [[TMP1:%.*]] = load i64, i64* getelementptr inbounds ([100 x i64], [100 x i64]* @__msan_param_tls, i32 0, i32 0), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <32 x i8>, <32 x i8>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 8) to <32 x i8>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load i64, i64* getelementptr inbounds ([100 x i64], [100 x i64]* @__msan_param_tls, i32 0, i32 0), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <32 x i8>, <32 x i8>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 8) to <32 x i8>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[_MSCMP:%.*]] = icmp ne i64 [[TMP1]], 0
; CHECK-NEXT: br i1 [[_MSCMP]], label [[TMP3:%.*]], label [[TMP4:%.*]], !prof [[PROF0]]
@@ -709,8 +709,8 @@ define <16 x i16> @test_x86_avx2_pmadd_ub_sw_load_op0(<32 x i8>* %ptr, <32 x i8>
define <16 x i16> @test_x86_avx2_pmul_hr_sw(<16 x i16> %a0, <16 x i16> %a1) #0 {
; CHECK-LABEL: @test_x86_avx2_pmul_hr_sw(
-; CHECK-NEXT: [[TMP1:%.*]] = load <16 x i16>, <16 x i16>* bitcast ([100 x i64]* @__msan_param_tls to <16 x i16>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <16 x i16>, <16 x i16>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <16 x i16>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <16 x i16>, <16 x i16>* bitcast ([100 x i64]* @__msan_param_tls to <16 x i16>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <16 x i16>, <16 x i16>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <16 x i16>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[_MSPROP:%.*]] = or <16 x i16> [[TMP1]], [[TMP2]]
; CHECK-NEXT: [[RES:%.*]] = call <16 x i16> @llvm.x86.avx2.pmul.hr.sw(<16 x i16> [[A0:%.*]], <16 x i16> [[A1:%.*]])
@@ -725,8 +725,8 @@ 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) #0 {
; CHECK-LABEL: @test_x86_avx2_pshuf_b(
-; CHECK-NEXT: [[TMP1:%.*]] = load <32 x i8>, <32 x i8>* bitcast ([100 x i64]* @__msan_param_tls to <32 x i8>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <32 x i8>, <32 x i8>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <32 x i8>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <32 x i8>, <32 x i8>* bitcast ([100 x i64]* @__msan_param_tls to <32 x i8>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <32 x i8>, <32 x i8>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <32 x i8>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[_MSPROP:%.*]] = or <32 x i8> [[TMP1]], [[TMP2]]
; CHECK-NEXT: [[RES:%.*]] = call <32 x i8> @llvm.x86.avx2.pshuf.b(<32 x i8> [[A0:%.*]], <32 x i8> [[A1:%.*]])
@@ -741,8 +741,8 @@ 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) #0 {
; CHECK-LABEL: @test_x86_avx2_psign_b(
-; CHECK-NEXT: [[TMP1:%.*]] = load <32 x i8>, <32 x i8>* bitcast ([100 x i64]* @__msan_param_tls to <32 x i8>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <32 x i8>, <32 x i8>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <32 x i8>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <32 x i8>, <32 x i8>* bitcast ([100 x i64]* @__msan_param_tls to <32 x i8>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <32 x i8>, <32 x i8>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <32 x i8>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[_MSPROP:%.*]] = or <32 x i8> [[TMP1]], [[TMP2]]
; CHECK-NEXT: [[RES:%.*]] = call <32 x i8> @llvm.x86.avx2.psign.b(<32 x i8> [[A0:%.*]], <32 x i8> [[A1:%.*]])
@@ -757,8 +757,8 @@ 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) #0 {
; CHECK-LABEL: @test_x86_avx2_psign_d(
-; CHECK-NEXT: [[TMP1:%.*]] = load <8 x i32>, <8 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i32>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <8 x i32>, <8 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <8 x i32>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <8 x i32>, <8 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i32>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <8 x i32>, <8 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <8 x i32>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[_MSPROP:%.*]] = or <8 x i32> [[TMP1]], [[TMP2]]
; CHECK-NEXT: [[RES:%.*]] = call <8 x i32> @llvm.x86.avx2.psign.d(<8 x i32> [[A0:%.*]], <8 x i32> [[A1:%.*]])
@@ -773,8 +773,8 @@ 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) #0 {
; CHECK-LABEL: @test_x86_avx2_psign_w(
-; CHECK-NEXT: [[TMP1:%.*]] = load <16 x i16>, <16 x i16>* bitcast ([100 x i64]* @__msan_param_tls to <16 x i16>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <16 x i16>, <16 x i16>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <16 x i16>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <16 x i16>, <16 x i16>* bitcast ([100 x i64]* @__msan_param_tls to <16 x i16>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <16 x i16>, <16 x i16>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <16 x i16>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[_MSPROP:%.*]] = or <16 x i16> [[TMP1]], [[TMP2]]
; CHECK-NEXT: [[RES:%.*]] = call <16 x i16> @llvm.x86.avx2.psign.w(<16 x i16> [[A0:%.*]], <16 x i16> [[A1:%.*]])
@@ -789,8 +789,8 @@ declare <16 x i16> @llvm.x86.avx2.psign.w(<16 x i16>, <16 x i16>) nounwind readn
define <16 x i16> @test_x86_avx2_mpsadbw(<32 x i8> %a0, <32 x i8> %a1) #0 {
; CHECK-LABEL: @test_x86_avx2_mpsadbw(
-; CHECK-NEXT: [[TMP1:%.*]] = load <32 x i8>, <32 x i8>* bitcast ([100 x i64]* @__msan_param_tls to <32 x i8>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <32 x i8>, <32 x i8>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <32 x i8>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <32 x i8>, <32 x i8>* bitcast ([100 x i64]* @__msan_param_tls to <32 x i8>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <32 x i8>, <32 x i8>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <32 x i8>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP3:%.*]] = bitcast <32 x i8> [[TMP1]] to i256
; CHECK-NEXT: [[_MSCMP:%.*]] = icmp ne i256 [[TMP3]], 0
@@ -817,8 +817,8 @@ declare <16 x i16> @llvm.x86.avx2.mpsadbw(<32 x i8>, <32 x i8>, i8) nounwind rea
define <16 x i16> @test_x86_avx2_mpsadbw_load_op0(<32 x i8>* %ptr, <32 x i8> %a1) #0 {
; CHECK-LABEL: @test_x86_avx2_mpsadbw_load_op0(
-; CHECK-NEXT: [[TMP1:%.*]] = load i64, i64* getelementptr inbounds ([100 x i64], [100 x i64]* @__msan_param_tls, i32 0, i32 0), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <32 x i8>, <32 x i8>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 8) to <32 x i8>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load i64, i64* getelementptr inbounds ([100 x i64], [100 x i64]* @__msan_param_tls, i32 0, i32 0), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <32 x i8>, <32 x i8>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 8) to <32 x i8>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[_MSCMP:%.*]] = icmp ne i64 [[TMP1]], 0
; CHECK-NEXT: br i1 [[_MSCMP]], label [[TMP3:%.*]], label [[TMP4:%.*]], !prof [[PROF0]]
@@ -856,8 +856,8 @@ define <16 x i16> @test_x86_avx2_mpsadbw_load_op0(<32 x i8>* %ptr, <32 x i8> %a1
define <16 x i16> @test_x86_avx2_packusdw(<8 x i32> %a0, <8 x i32> %a1) #0 {
; CHECK-LABEL: @test_x86_avx2_packusdw(
-; CHECK-NEXT: [[TMP1:%.*]] = load <8 x i32>, <8 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i32>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <8 x i32>, <8 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <8 x i32>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <8 x i32>, <8 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i32>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <8 x i32>, <8 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <8 x i32>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP3:%.*]] = icmp ne <8 x i32> [[TMP1]], zeroinitializer
; CHECK-NEXT: [[TMP4:%.*]] = sext <8 x i1> [[TMP3]] to <8 x i32>
@@ -889,9 +889,9 @@ define <16 x i16> @test_x86_avx2_packusdw_fold() #0 {
define <32 x i8> @test_x86_avx2_pblendvb(<32 x i8> %a0, <32 x i8> %a1, <32 x i8> %a2) #0 {
; CHECK-LABEL: @test_x86_avx2_pblendvb(
-; CHECK-NEXT: [[TMP1:%.*]] = load <32 x i8>, <32 x i8>* bitcast ([100 x i64]* @__msan_param_tls to <32 x i8>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <32 x i8>, <32 x i8>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <32 x i8>*), align 8
-; CHECK-NEXT: [[TMP3:%.*]] = load <32 x i8>, <32 x i8>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 64) to <32 x i8>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <32 x i8>, <32 x i8>* bitcast ([100 x i64]* @__msan_param_tls to <32 x i8>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <32 x i8>, <32 x i8>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <32 x i8>*), align 8
+; CHECK-DAG: [[TMP3:%.*]] = load <32 x i8>, <32 x i8>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 64) to <32 x i8>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[_MSPROP:%.*]] = or <32 x i8> [[TMP1]], [[TMP2]]
; CHECK-NEXT: [[_MSPROP1:%.*]] = or <32 x i8> [[_MSPROP]], [[TMP3]]
@@ -907,8 +907,8 @@ 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) #0 {
; CHECK-LABEL: @test_x86_avx2_pblendw(
-; CHECK-NEXT: [[TMP1:%.*]] = load <16 x i16>, <16 x i16>* bitcast ([100 x i64]* @__msan_param_tls to <16 x i16>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <16 x i16>, <16 x i16>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <16 x i16>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <16 x i16>, <16 x i16>* bitcast ([100 x i64]* @__msan_param_tls to <16 x i16>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <16 x i16>, <16 x i16>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <16 x i16>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[_MSPROP:%.*]] = shufflevector <16 x i16> [[TMP1]], <16 x i16> [[TMP2]], <16 x i32> <i32 16, i32 17, i32 18, i32 3, i32 4, i32 5, i32 6, i32 7, i32 24, i32 25, i32 26, i32 11, i32 12, i32 13, i32 14, i32 15>
; CHECK-NEXT: [[TMP3:%.*]] = shufflevector <16 x i16> [[A0:%.*]], <16 x i16> [[A1:%.*]], <16 x i32> <i32 16, i32 17, i32 18, i32 3, i32 4, i32 5, i32 6, i32 7, i32 24, i32 25, i32 26, i32 11, i32 12, i32 13, i32 14, i32 15>
@@ -923,8 +923,8 @@ declare <16 x i16> @llvm.x86.avx2.pblendw(<16 x i16>, <16 x i16>, i8) nounwind r
define <4 x i32> @test_x86_avx2_pblendd_128(<4 x i32> %a0, <4 x i32> %a1) #0 {
; CHECK-LABEL: @test_x86_avx2_pblendd_128(
-; CHECK-NEXT: [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i32>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <4 x i32>, <4 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <4 x i32>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i32>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <4 x i32>, <4 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <4 x i32>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[_MSPROP:%.*]] = shufflevector <4 x i32> [[TMP1]], <4 x i32> [[TMP2]], <4 x i32> <i32 4, i32 5, i32 6, i32 3>
; CHECK-NEXT: [[TMP3:%.*]] = shufflevector <4 x i32> [[A0:%.*]], <4 x i32> [[A1:%.*]], <4 x i32> <i32 4, i32 5, i32 6, i32 3>
@@ -939,8 +939,8 @@ 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) #0 {
; CHECK-LABEL: @test_x86_avx2_pblendd_256(
-; CHECK-NEXT: [[TMP1:%.*]] = load <8 x i32>, <8 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i32>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <8 x i32>, <8 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <8 x i32>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <8 x i32>, <8 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i32>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <8 x i32>, <8 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <8 x i32>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[_MSPROP:%.*]] = shufflevector <8 x i32> [[TMP1]], <8 x i32> [[TMP2]], <8 x i32> <i32 8, i32 9, i32 10, i32 3, i32 4, i32 5, i32 6, i32 7>
; CHECK-NEXT: [[TMP3:%.*]] = shufflevector <8 x i32> [[A0:%.*]], <8 x i32> [[A1:%.*]], <8 x i32> <i32 8, i32 9, i32 10, i32 3, i32 4, i32 5, i32 6, i32 7>
@@ -955,8 +955,8 @@ declare <8 x i32> @llvm.x86.avx2.pblendd.256(<8 x i32>, <8 x i32>, i8) nounwind
define <8 x i32> @test_x86_avx2_permd(<8 x i32> %a0, <8 x i32> %a1) #0 {
; CHECK-LABEL: @test_x86_avx2_permd(
-; CHECK-NEXT: [[TMP1:%.*]] = load <8 x i32>, <8 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i32>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <8 x i32>, <8 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <8 x i32>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <8 x i32>, <8 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i32>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <8 x i32>, <8 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <8 x i32>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[_MSPROP:%.*]] = or <8 x i32> [[TMP1]], [[TMP2]]
; CHECK-NEXT: [[RES:%.*]] = call <8 x i32> @llvm.x86.avx2.permd(<8 x i32> [[A0:%.*]], <8 x i32> [[A1:%.*]])
@@ -971,8 +971,8 @@ declare <8 x i32> @llvm.x86.avx2.permd(<8 x i32>, <8 x i32>) nounwind readonly
define <8 x float> @test_x86_avx2_permps(<8 x float> %a0, <8 x i32> %a1) #0 {
; CHECK-LABEL: @test_x86_avx2_permps(
-; CHECK-NEXT: [[TMP1:%.*]] = load <8 x i32>, <8 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i32>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <8 x i32>, <8 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <8 x i32>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <8 x i32>, <8 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i32>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <8 x i32>, <8 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <8 x i32>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP3:%.*]] = bitcast <8 x i32> [[TMP1]] to i256
; CHECK-NEXT: [[_MSCMP:%.*]] = icmp ne i256 [[TMP3]], 0
@@ -1000,8 +1000,8 @@ declare <8 x float> @llvm.x86.avx2.permps(<8 x float>, <8 x i32>) nounwind reado
define <2 x i64> @test_x86_avx2_maskload_q(i8* %a0, <2 x i64> %a1) #0 {
; CHECK-LABEL: @test_x86_avx2_maskload_q(
-; CHECK-NEXT: [[TMP1:%.*]] = load i64, i64* getelementptr inbounds ([100 x i64], [100 x i64]* @__msan_param_tls, i32 0, i32 0), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <2 x i64>, <2 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 8) to <2 x i64>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load i64, i64* getelementptr inbounds ([100 x i64], [100 x i64]* @__msan_param_tls, i32 0, i32 0), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <2 x i64>, <2 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 8) to <2 x i64>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[_MSCMP:%.*]] = icmp ne i64 [[TMP1]], 0
; CHECK-NEXT: br i1 [[_MSCMP]], label [[TMP3:%.*]], label [[TMP4:%.*]], !prof [[PROF0]]
@@ -1028,8 +1028,8 @@ 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) #0 {
; CHECK-LABEL: @test_x86_avx2_maskload_q_256(
-; CHECK-NEXT: [[TMP1:%.*]] = load i64, i64* getelementptr inbounds ([100 x i64], [100 x i64]* @__msan_param_tls, i32 0, i32 0), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <4 x i64>, <4 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 8) to <4 x i64>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load i64, i64* getelementptr inbounds ([100 x i64], [100 x i64]* @__msan_param_tls, i32 0, i32 0), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <4 x i64>, <4 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 8) to <4 x i64>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[_MSCMP:%.*]] = icmp ne i64 [[TMP1]], 0
; CHECK-NEXT: br i1 [[_MSCMP]], label [[TMP3:%.*]], label [[TMP4:%.*]], !prof [[PROF0]]
@@ -1056,8 +1056,8 @@ 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) #0 {
; CHECK-LABEL: @test_x86_avx2_maskload_d(
-; CHECK-NEXT: [[TMP1:%.*]] = load i64, i64* getelementptr inbounds ([100 x i64], [100 x i64]* @__msan_param_tls, i32 0, i32 0), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <4 x i32>, <4 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 8) to <4 x i32>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load i64, i64* getelementptr inbounds ([100 x i64], [100 x i64]* @__msan_param_tls, i32 0, i32 0), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <4 x i32>, <4 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 8) to <4 x i32>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[_MSCMP:%.*]] = icmp ne i64 [[TMP1]], 0
; CHECK-NEXT: br i1 [[_MSCMP]], label [[TMP3:%.*]], label [[TMP4:%.*]], !prof [[PROF0]]
@@ -1084,8 +1084,8 @@ 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) #0 {
; CHECK-LABEL: @test_x86_avx2_maskload_d_256(
-; CHECK-NEXT: [[TMP1:%.*]] = load i64, i64* getelementptr inbounds ([100 x i64], [100 x i64]* @__msan_param_tls, i32 0, i32 0), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <8 x i32>, <8 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 8) to <8 x i32>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load i64, i64* getelementptr inbounds ([100 x i64], [100 x i64]* @__msan_param_tls, i32 0, i32 0), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <8 x i32>, <8 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 8) to <8 x i32>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[_MSCMP:%.*]] = icmp ne i64 [[TMP1]], 0
; CHECK-NEXT: br i1 [[_MSCMP]], label [[TMP3:%.*]], label [[TMP4:%.*]], !prof [[PROF0]]
@@ -1112,9 +1112,9 @@ 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) #0 {
; CHECK-LABEL: @test_x86_avx2_maskstore_q(
-; CHECK-NEXT: [[TMP1:%.*]] = load i64, i64* getelementptr inbounds ([100 x i64], [100 x i64]* @__msan_param_tls, i32 0, i32 0), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <2 x i64>, <2 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 8) to <2 x i64>*), align 8
-; CHECK-NEXT: [[TMP3:%.*]] = load <2 x i64>, <2 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 24) to <2 x i64>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load i64, i64* getelementptr inbounds ([100 x i64], [100 x i64]* @__msan_param_tls, i32 0, i32 0), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <2 x i64>, <2 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 8) to <2 x i64>*), align 8
+; CHECK-DAG: [[TMP3:%.*]] = load <2 x i64>, <2 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 24) to <2 x i64>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[_MSCMP:%.*]] = icmp ne i64 [[TMP1]], 0
; CHECK-NEXT: br i1 [[_MSCMP]], label [[TMP4:%.*]], label [[TMP5:%.*]], !prof [[PROF0]]
@@ -1147,9 +1147,9 @@ 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) #0 {
; CHECK-LABEL: @test_x86_avx2_maskstore_q_256(
-; CHECK-NEXT: [[TMP1:%.*]] = load i64, i64* getelementptr inbounds ([100 x i64], [100 x i64]* @__msan_param_tls, i32 0, i32 0), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <4 x i64>, <4 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 8) to <4 x i64>*), align 8
-; CHECK-NEXT: [[TMP3:%.*]] = load <4 x i64>, <4 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 40) to <4 x i64>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load i64, i64* getelementptr inbounds ([100 x i64], [100 x i64]* @__msan_param_tls, i32 0, i32 0), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <4 x i64>, <4 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 8) to <4 x i64>*), align 8
+; CHECK-DAG: [[TMP3:%.*]] = load <4 x i64>, <4 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 40) to <4 x i64>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[_MSCMP:%.*]] = icmp ne i64 [[TMP1]], 0
; CHECK-NEXT: br i1 [[_MSCMP]], label [[TMP4:%.*]], label [[TMP5:%.*]], !prof [[PROF0]]
@@ -1182,9 +1182,9 @@ 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) #0 {
; CHECK-LABEL: @test_x86_avx2_maskstore_d(
-; CHECK-NEXT: [[TMP1:%.*]] = load i64, i64* getelementptr inbounds ([100 x i64], [100 x i64]* @__msan_param_tls, i32 0, i32 0), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <4 x i32>, <4 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 8) to <4 x i32>*), align 8
-; CHECK-NEXT: [[TMP3:%.*]] = load <4 x i32>, <4 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 24) to <4 x i32>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load i64, i64* getelementptr inbounds ([100 x i64], [100 x i64]* @__msan_param_tls, i32 0, i32 0), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <4 x i32>, <4 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 8) to <4 x i32>*), align 8
+; CHECK-DAG: [[TMP3:%.*]] = load <4 x i32>, <4 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 24) to <4 x i32>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[_MSCMP:%.*]] = icmp ne i64 [[TMP1]], 0
; CHECK-NEXT: br i1 [[_MSCMP]], label [[TMP4:%.*]], label [[TMP5:%.*]], !prof [[PROF0]]
@@ -1217,9 +1217,9 @@ 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) #0 {
; CHECK-LABEL: @test_x86_avx2_maskstore_d_256(
-; CHECK-NEXT: [[TMP1:%.*]] = load i64, i64* getelementptr inbounds ([100 x i64], [100 x i64]* @__msan_param_tls, i32 0, i32 0), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <8 x i32>, <8 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 8) to <8 x i32>*), align 8
-; CHECK-NEXT: [[TMP3:%.*]] = load <8 x i32>, <8 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 40) to <8 x i32>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load i64, i64* getelementptr inbounds ([100 x i64], [100 x i64]* @__msan_param_tls, i32 0, i32 0), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <8 x i32>, <8 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 8) to <8 x i32>*), align 8
+; CHECK-DAG: [[TMP3:%.*]] = load <8 x i32>, <8 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 40) to <8 x i32>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[_MSCMP:%.*]] = icmp ne i64 [[TMP1]], 0
; CHECK-NEXT: br i1 [[_MSCMP]], label [[TMP4:%.*]], label [[TMP5:%.*]], !prof [[PROF0]]
@@ -1252,8 +1252,8 @@ 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) #0 {
; CHECK-LABEL: @test_x86_avx2_psllv_d(
-; CHECK-NEXT: [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i32>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <4 x i32>, <4 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <4 x i32>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i32>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <4 x i32>, <4 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <4 x i32>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP3:%.*]] = icmp ne <4 x i32> [[TMP2]], zeroinitializer
; CHECK-NEXT: [[TMP4:%.*]] = sext <4 x i1> [[TMP3]] to <4 x i32>
@@ -1291,8 +1291,8 @@ 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) #0 {
; CHECK-LABEL: @test_x86_avx2_psllv_d_256(
-; CHECK-NEXT: [[TMP1:%.*]] = load <8 x i32>, <8 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i32>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <8 x i32>, <8 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <8 x i32>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <8 x i32>, <8 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i32>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <8 x i32>, <8 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <8 x i32>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP3:%.*]] = icmp ne <8 x i32> [[TMP2]], zeroinitializer
; CHECK-NEXT: [[TMP4:%.*]] = sext <8 x i1> [[TMP3]] to <8 x i32>
@@ -1330,8 +1330,8 @@ 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) #0 {
; CHECK-LABEL: @test_x86_avx2_psllv_q(
-; CHECK-NEXT: [[TMP1:%.*]] = load <2 x i64>, <2 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <2 x i64>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <2 x i64>, <2 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <2 x i64>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <2 x i64>, <2 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <2 x i64>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <2 x i64>, <2 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <2 x i64>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP3:%.*]] = icmp ne <2 x i64> [[TMP2]], zeroinitializer
; CHECK-NEXT: [[TMP4:%.*]] = sext <2 x i1> [[TMP3]] to <2 x i64>
@@ -1361,8 +1361,8 @@ 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) #0 {
; CHECK-LABEL: @test_x86_avx2_psllv_q_256(
-; CHECK-NEXT: [[TMP1:%.*]] = load <4 x i64>, <4 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i64>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <4 x i64>, <4 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <4 x i64>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <4 x i64>, <4 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i64>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <4 x i64>, <4 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <4 x i64>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP3:%.*]] = icmp ne <4 x i64> [[TMP2]], zeroinitializer
; CHECK-NEXT: [[TMP4:%.*]] = sext <4 x i1> [[TMP3]] to <4 x i64>
@@ -1393,8 +1393,8 @@ 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) #0 {
; CHECK-LABEL: @test_x86_avx2_psrlv_d(
-; CHECK-NEXT: [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i32>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <4 x i32>, <4 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <4 x i32>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i32>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <4 x i32>, <4 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <4 x i32>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP3:%.*]] = icmp ne <4 x i32> [[TMP2]], zeroinitializer
; CHECK-NEXT: [[TMP4:%.*]] = sext <4 x i1> [[TMP3]] to <4 x i32>
@@ -1432,8 +1432,8 @@ 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) #0 {
; CHECK-LABEL: @test_x86_avx2_psrlv_d_256(
-; CHECK-NEXT: [[TMP1:%.*]] = load <8 x i32>, <8 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i32>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <8 x i32>, <8 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <8 x i32>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <8 x i32>, <8 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i32>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <8 x i32>, <8 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <8 x i32>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP3:%.*]] = icmp ne <8 x i32> [[TMP2]], zeroinitializer
; CHECK-NEXT: [[TMP4:%.*]] = sext <8 x i1> [[TMP3]] to <8 x i32>
@@ -1471,8 +1471,8 @@ 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) #0 {
; CHECK-LABEL: @test_x86_avx2_psrlv_q(
-; CHECK-NEXT: [[TMP1:%.*]] = load <2 x i64>, <2 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <2 x i64>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <2 x i64>, <2 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <2 x i64>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <2 x i64>, <2 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <2 x i64>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <2 x i64>, <2 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <2 x i64>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP3:%.*]] = icmp ne <2 x i64> [[TMP2]], zeroinitializer
; CHECK-NEXT: [[TMP4:%.*]] = sext <2 x i1> [[TMP3]] to <2 x i64>
@@ -1503,8 +1503,8 @@ 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) #0 {
; CHECK-LABEL: @test_x86_avx2_psrlv_q_256(
-; CHECK-NEXT: [[TMP1:%.*]] = load <4 x i64>, <4 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i64>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <4 x i64>, <4 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <4 x i64>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <4 x i64>, <4 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i64>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <4 x i64>, <4 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <4 x i64>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP3:%.*]] = icmp ne <4 x i64> [[TMP2]], zeroinitializer
; CHECK-NEXT: [[TMP4:%.*]] = sext <4 x i1> [[TMP3]] to <4 x i64>
@@ -1536,8 +1536,8 @@ 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) #0 {
; CHECK-LABEL: @test_x86_avx2_psrav_d(
-; CHECK-NEXT: [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i32>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <4 x i32>, <4 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <4 x i32>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i32>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <4 x i32>, <4 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <4 x i32>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP3:%.*]] = icmp ne <4 x i32> [[TMP2]], zeroinitializer
; CHECK-NEXT: [[TMP4:%.*]] = sext <4 x i1> [[TMP3]] to <4 x i32>
@@ -1567,8 +1567,8 @@ 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) #0 {
; CHECK-LABEL: @test_x86_avx2_psrav_d_256(
-; CHECK-NEXT: [[TMP1:%.*]] = load <8 x i32>, <8 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i32>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <8 x i32>, <8 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <8 x i32>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <8 x i32>, <8 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i32>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <8 x i32>, <8 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <8 x i32>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP3:%.*]] = icmp ne <8 x i32> [[TMP2]], zeroinitializer
; CHECK-NEXT: [[TMP4:%.*]] = sext <8 x i1> [[TMP3]] to <8 x i32>
@@ -1598,10 +1598,10 @@ declare <8 x i32> @llvm.x86.avx2.psrav.d.256(<8 x i32>, <8 x i32>) nounwind read
define <2 x double> @test_x86_avx2_gather_d_pd(<2 x double> %a0, i8* %a1, <4 x i32> %idx, <2 x double> %mask) #0 {
; CHECK-LABEL: @test_x86_avx2_gather_d_pd(
-; CHECK-NEXT: [[TMP1:%.*]] = load <2 x i64>, <2 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <2 x i64>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load i64, i64* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to i64*), align 8
-; CHECK-NEXT: [[TMP3:%.*]] = load <4 x i32>, <4 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 24) to <4 x i32>*), align 8
-; CHECK-NEXT: [[TMP4:%.*]] = load <2 x i64>, <2 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 40) to <2 x i64>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <2 x i64>, <2 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <2 x i64>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load i64, i64* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to i64*), align 8
+; CHECK-DAG: [[TMP3:%.*]] = load <4 x i32>, <4 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 24) to <4 x i32>*), align 8
+; CHECK-DAG: [[TMP4:%.*]] = load <2 x i64>, <2 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 40) to <2 x i64>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP5:%.*]] = bitcast <2 x i64> [[TMP1]] to i128
; CHECK-NEXT: [[_MSCMP:%.*]] = icmp ne i128 [[TMP5]], 0
@@ -1643,10 +1643,10 @@ declare <2 x double> @llvm.x86.avx2.gather.d.pd(<2 x double>, i8*,
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) #0 {
; CHECK-LABEL: @test_x86_avx2_gather_d_pd_256(
-; CHECK-NEXT: [[TMP1:%.*]] = load <4 x i64>, <4 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i64>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load i64, i64* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to i64*), align 8
-; CHECK-NEXT: [[TMP3:%.*]] = load <4 x i32>, <4 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 40) to <4 x i32>*), align 8
-; CHECK-NEXT: [[TMP4:%.*]] = load <4 x i64>, <4 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 56) to <4 x i64>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <4 x i64>, <4 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i64>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load i64, i64* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to i64*), align 8
+; CHECK-DAG: [[TMP3:%.*]] = load <4 x i32>, <4 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 40) to <4 x i32>*), align 8
+; CHECK-DAG: [[TMP4:%.*]] = load <4 x i64>, <4 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 56) to <4 x i64>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP5:%.*]] = bitcast <4 x i64> [[TMP1]] to i256
; CHECK-NEXT: [[_MSCMP:%.*]] = icmp ne i256 [[TMP5]], 0
@@ -1688,10 +1688,10 @@ declare <4 x double> @llvm.x86.avx2.gather.d.pd.256(<4 x double>, i8*,
define <2 x double> @test_x86_avx2_gather_q_pd(<2 x double> %a0, i8* %a1, <2 x i64> %idx, <2 x double> %mask) #0 {
; CHECK-LABEL: @test_x86_avx2_gather_q_pd(
-; CHECK-NEXT: [[TMP1:%.*]] = load <2 x i64>, <2 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <2 x i64>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load i64, i64* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to i64*), align 8
-; CHECK-NEXT: [[TMP3:%.*]] = load <2 x i64>, <2 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 24) to <2 x i64>*), align 8
-; CHECK-NEXT: [[TMP4:%.*]] = load <2 x i64>, <2 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 40) to <2 x i64>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <2 x i64>, <2 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <2 x i64>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load i64, i64* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to i64*), align 8
+; CHECK-DAG: [[TMP3:%.*]] = load <2 x i64>, <2 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 24) to <2 x i64>*), align 8
+; CHECK-DAG: [[TMP4:%.*]] = load <2 x i64>, <2 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 40) to <2 x i64>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP5:%.*]] = bitcast <2 x i64> [[TMP1]] to i128
; CHECK-NEXT: [[_MSCMP:%.*]] = icmp ne i128 [[TMP5]], 0
@@ -1733,10 +1733,10 @@ declare <2 x double> @llvm.x86.avx2.gather.q.pd(<2 x double>, i8*,
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) #0 {
; CHECK-LABEL: @test_x86_avx2_gather_q_pd_256(
-; CHECK-NEXT: [[TMP1:%.*]] = load <4 x i64>, <4 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i64>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load i64, i64* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to i64*), align 8
-; CHECK-NEXT: [[TMP3:%.*]] = load <4 x i64>, <4 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 40) to <4 x i64>*), align 8
-; CHECK-NEXT: [[TMP4:%.*]] = load <4 x i64>, <4 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 72) to <4 x i64>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <4 x i64>, <4 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i64>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load i64, i64* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to i64*), align 8
+; CHECK-DAG: [[TMP3:%.*]] = load <4 x i64>, <4 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 40) to <4 x i64>*), align 8
+; CHECK-DAG: [[TMP4:%.*]] = load <4 x i64>, <4 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 72) to <4 x i64>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP5:%.*]] = bitcast <4 x i64> [[TMP1]] to i256
; CHECK-NEXT: [[_MSCMP:%.*]] = icmp ne i256 [[TMP5]], 0
@@ -1778,10 +1778,10 @@ declare <4 x double> @llvm.x86.avx2.gather.q.pd.256(<4 x double>, i8*,
define <4 x float> @test_x86_avx2_gather_d_ps(<4 x float> %a0, i8* %a1, <4 x i32> %idx, <4 x float> %mask) #0 {
; CHECK-LABEL: @test_x86_avx2_gather_d_ps(
-; CHECK-NEXT: [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i32>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load i64, i64* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to i64*), align 8
-; CHECK-NEXT: [[TMP3:%.*]] = load <4 x i32>, <4 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 24) to <4 x i32>*), align 8
-; CHECK-NEXT: [[TMP4:%.*]] = load <4 x i32>, <4 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 40) to <4 x i32>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i32>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load i64, i64* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to i64*), align 8
+; CHECK-DAG: [[TMP3:%.*]] = load <4 x i32>, <4 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 24) to <4 x i32>*), align 8
+; CHECK-DAG: [[TMP4:%.*]] = load <4 x i32>, <4 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 40) to <4 x i32>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP5:%.*]] = bitcast <4 x i32> [[TMP1]] to i128
; CHECK-NEXT: [[_MSCMP:%.*]] = icmp ne i128 [[TMP5]], 0
@@ -1823,10 +1823,10 @@ declare <4 x float> @llvm.x86.avx2.gather.d.ps(<4 x float>, i8*,
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) #0 {
; CHECK-LABEL: @test_x86_avx2_gather_d_ps_256(
-; CHECK-NEXT: [[TMP1:%.*]] = load <8 x i32>, <8 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i32>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load i64, i64* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to i64*), align 8
-; CHECK-NEXT: [[TMP3:%.*]] = load <8 x i32>, <8 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 40) to <8 x i32>*), align 8
-; CHECK-NEXT: [[TMP4:%.*]] = load <8 x i32>, <8 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 72) to <8 x i32>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <8 x i32>, <8 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i32>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load i64, i64* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to i64*), align 8
+; CHECK-DAG: [[TMP3:%.*]] = load <8 x i32>, <8 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 40) to <8 x i32>*), align 8
+; CHECK-DAG: [[TMP4:%.*]] = load <8 x i32>, <8 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 72) to <8 x i32>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP5:%.*]] = bitcast <8 x i32> [[TMP1]] to i256
; CHECK-NEXT: [[_MSCMP:%.*]] = icmp ne i256 [[TMP5]], 0
@@ -1868,10 +1868,10 @@ declare <8 x float> @llvm.x86.avx2.gather.d.ps.256(<8 x float>, i8*,
define <4 x float> @test_x86_avx2_gather_q_ps(<4 x float> %a0, i8* %a1, <2 x i64> %idx, <4 x float> %mask) #0 {
; CHECK-LABEL: @test_x86_avx2_gather_q_ps(
-; CHECK-NEXT: [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i32>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load i64, i64* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to i64*), align 8
-; CHECK-NEXT: [[TMP3:%.*]] = load <2 x i64>, <2 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 24) to <2 x i64>*), align 8
-; CHECK-NEXT: [[TMP4:%.*]] = load <4 x i32>, <4 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 40) to <4 x i32>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i32>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load i64, i64* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to i64*), align 8
+; CHECK-DAG: [[TMP3:%.*]] = load <2 x i64>, <2 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 24) to <2 x i64>*), align 8
+; CHECK-DAG: [[TMP4:%.*]] = load <4 x i32>, <4 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 40) to <4 x i32>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP5:%.*]] = bitcast <4 x i32> [[TMP1]] to i128
; CHECK-NEXT: [[_MSCMP:%.*]] = icmp ne i128 [[TMP5]], 0
@@ -1913,10 +1913,10 @@ declare <4 x float> @llvm.x86.avx2.gather.q.ps(<4 x float>, i8*,
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) #0 {
; CHECK-LABEL: @test_x86_avx2_gather_q_ps_256(
-; CHECK-NEXT: [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i32>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load i64, i64* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to i64*), align 8
-; CHECK-NEXT: [[TMP3:%.*]] = load <4 x i64>, <4 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 24) to <4 x i64>*), align 8
-; CHECK-NEXT: [[TMP4:%.*]] = load <4 x i32>, <4 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 56) to <4 x i32>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i32>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load i64, i64* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to i64*), align 8
+; CHECK-DAG: [[TMP3:%.*]] = load <4 x i64>, <4 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 24) to <4 x i64>*), align 8
+; CHECK-DAG: [[TMP4:%.*]] = load <4 x i32>, <4 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 56) to <4 x i32>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP5:%.*]] = bitcast <4 x i32> [[TMP1]] to i128
; CHECK-NEXT: [[_MSCMP:%.*]] = icmp ne i128 [[TMP5]], 0
@@ -1958,10 +1958,10 @@ declare <4 x float> @llvm.x86.avx2.gather.q.ps.256(<4 x float>, i8*,
define <2 x i64> @test_x86_avx2_gather_d_q(<2 x i64> %a0, i8* %a1, <4 x i32> %idx, <2 x i64> %mask) #0 {
; CHECK-LABEL: @test_x86_avx2_gather_d_q(
-; CHECK-NEXT: [[TMP1:%.*]] = load <2 x i64>, <2 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <2 x i64>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load i64, i64* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to i64*), align 8
-; CHECK-NEXT: [[TMP3:%.*]] = load <4 x i32>, <4 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 24) to <4 x i32>*), align 8
-; CHECK-NEXT: [[TMP4:%.*]] = load <2 x i64>, <2 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 40) to <2 x i64>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <2 x i64>, <2 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <2 x i64>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load i64, i64* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to i64*), align 8
+; CHECK-DAG: [[TMP3:%.*]] = load <4 x i32>, <4 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 24) to <4 x i32>*), align 8
+; CHECK-DAG: [[TMP4:%.*]] = load <2 x i64>, <2 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 40) to <2 x i64>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP5:%.*]] = bitcast <2 x i64> [[TMP1]] to i128
; CHECK-NEXT: [[_MSCMP:%.*]] = icmp ne i128 [[TMP5]], 0
@@ -2003,10 +2003,10 @@ declare <2 x i64> @llvm.x86.avx2.gather.d.q(<2 x i64>, i8*,
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) #0 {
; CHECK-LABEL: @test_x86_avx2_gather_d_q_256(
-; CHECK-NEXT: [[TMP1:%.*]] = load <4 x i64>, <4 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i64>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load i64, i64* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to i64*), align 8
-; CHECK-NEXT: [[TMP3:%.*]] = load <4 x i32>, <4 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 40) to <4 x i32>*), align 8
-; CHECK-NEXT: [[TMP4:%.*]] = load <4 x i64>, <4 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 56) to <4 x i64>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <4 x i64>, <4 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i64>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load i64, i64* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to i64*), align 8
+; CHECK-DAG: [[TMP3:%.*]] = load <4 x i32>, <4 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 40) to <4 x i32>*), align 8
+; CHECK-DAG: [[TMP4:%.*]] = load <4 x i64>, <4 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 56) to <4 x i64>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP5:%.*]] = bitcast <4 x i64> [[TMP1]] to i256
; CHECK-NEXT: [[_MSCMP:%.*]] = icmp ne i256 [[TMP5]], 0
@@ -2048,10 +2048,10 @@ declare <4 x i64> @llvm.x86.avx2.gather.d.q.256(<4 x i64>, i8*,
define <2 x i64> @test_x86_avx2_gather_q_q(<2 x i64> %a0, i8* %a1, <2 x i64> %idx, <2 x i64> %mask) #0 {
; CHECK-LABEL: @test_x86_avx2_gather_q_q(
-; CHECK-NEXT: [[TMP1:%.*]] = load <2 x i64>, <2 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <2 x i64>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load i64, i64* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to i64*), align 8
-; CHECK-NEXT: [[TMP3:%.*]] = load <2 x i64>, <2 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 24) to <2 x i64>*), align 8
-; CHECK-NEXT: [[TMP4:%.*]] = load <2 x i64>, <2 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 40) to <2 x i64>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <2 x i64>, <2 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <2 x i64>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load i64, i64* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to i64*), align 8
+; CHECK-DAG: [[TMP3:%.*]] = load <2 x i64>, <2 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 24) to <2 x i64>*), align 8
+; CHECK-DAG: [[TMP4:%.*]] = load <2 x i64>, <2 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 40) to <2 x i64>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP5:%.*]] = bitcast <2 x i64> [[TMP1]] to i128
; CHECK-NEXT: [[_MSCMP:%.*]] = icmp ne i128 [[TMP5]], 0
@@ -2093,10 +2093,10 @@ declare <2 x i64> @llvm.x86.avx2.gather.q.q(<2 x i64>, i8*,
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) #0 {
; CHECK-LABEL: @test_x86_avx2_gather_q_q_256(
-; CHECK-NEXT: [[TMP1:%.*]] = load <4 x i64>, <4 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i64>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load i64, i64* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to i64*), align 8
-; CHECK-NEXT: [[TMP3:%.*]] = load <4 x i64>, <4 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 40) to <4 x i64>*), align 8
-; CHECK-NEXT: [[TMP4:%.*]] = load <4 x i64>, <4 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 72) to <4 x i64>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <4 x i64>, <4 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i64>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load i64, i64* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to i64*), align 8
+; CHECK-DAG: [[TMP3:%.*]] = load <4 x i64>, <4 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 40) to <4 x i64>*), align 8
+; CHECK-DAG: [[TMP4:%.*]] = load <4 x i64>, <4 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 72) to <4 x i64>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP5:%.*]] = bitcast <4 x i64> [[TMP1]] to i256
; CHECK-NEXT: [[_MSCMP:%.*]] = icmp ne i256 [[TMP5]], 0
@@ -2138,10 +2138,10 @@ declare <4 x i64> @llvm.x86.avx2.gather.q.q.256(<4 x i64>, i8*,
define <4 x i32> @test_x86_avx2_gather_d_d(<4 x i32> %a0, i8* %a1, <4 x i32> %idx, <4 x i32> %mask) #0 {
; CHECK-LABEL: @test_x86_avx2_gather_d_d(
-; CHECK-NEXT: [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i32>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load i64, i64* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to i64*), align 8
-; CHECK-NEXT: [[TMP3:%.*]] = load <4 x i32>, <4 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 24) to <4 x i32>*), align 8
-; CHECK-NEXT: [[TMP4:%.*]] = load <4 x i32>, <4 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 40) to <4 x i32>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i32>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load i64, i64* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to i64*), align 8
+; CHECK-DAG: [[TMP3:%.*]] = load <4 x i32>, <4 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 24) to <4 x i32>*), align 8
+; CHECK-DAG: [[TMP4:%.*]] = load <4 x i32>, <4 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 40) to <4 x i32>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP5:%.*]] = bitcast <4 x i32> [[TMP1]] to i128
; CHECK-NEXT: [[_MSCMP:%.*]] = icmp ne i128 [[TMP5]], 0
@@ -2183,10 +2183,10 @@ declare <4 x i32> @llvm.x86.avx2.gather.d.d(<4 x i32>, i8*,
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) #0 {
; CHECK-LABEL: @test_x86_avx2_gather_d_d_256(
-; CHECK-NEXT: [[TMP1:%.*]] = load <8 x i32>, <8 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i32>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load i64, i64* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to i64*), align 8
-; CHECK-NEXT: [[TMP3:%.*]] = load <8 x i32>, <8 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 40) to <8 x i32>*), align 8
-; CHECK-NEXT: [[TMP4:%.*]] = load <8 x i32>, <8 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 72) to <8 x i32>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <8 x i32>, <8 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i32>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load i64, i64* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to i64*), align 8
+; CHECK-DAG: [[TMP3:%.*]] = load <8 x i32>, <8 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 40) to <8 x i32>*), align 8
+; CHECK-DAG: [[TMP4:%.*]] = load <8 x i32>, <8 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 72) to <8 x i32>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP5:%.*]] = bitcast <8 x i32> [[TMP1]] to i256
; CHECK-NEXT: [[_MSCMP:%.*]] = icmp ne i256 [[TMP5]], 0
@@ -2228,10 +2228,10 @@ declare <8 x i32> @llvm.x86.avx2.gather.d.d.256(<8 x i32>, i8*,
define <4 x i32> @test_x86_avx2_gather_q_d(<4 x i32> %a0, i8* %a1, <2 x i64> %idx, <4 x i32> %mask) #0 {
; CHECK-LABEL: @test_x86_avx2_gather_q_d(
-; CHECK-NEXT: [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i32>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load i64, i64* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to i64*), align 8
-; CHECK-NEXT: [[TMP3:%.*]] = load <2 x i64>, <2 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 24) to <2 x i64>*), align 8
-; CHECK-NEXT: [[TMP4:%.*]] = load <4 x i32>, <4 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 40) to <4 x i32>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i32>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load i64, i64* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to i64*), align 8
+; CHECK-DAG: [[TMP3:%.*]] = load <2 x i64>, <2 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 24) to <2 x i64>*), align 8
+; CHECK-DAG: [[TMP4:%.*]] = load <4 x i32>, <4 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 40) to <4 x i32>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP5:%.*]] = bitcast <4 x i32> [[TMP1]] to i128
; CHECK-NEXT: [[_MSCMP:%.*]] = icmp ne i128 [[TMP5]], 0
@@ -2273,10 +2273,10 @@ declare <4 x i32> @llvm.x86.avx2.gather.q.d(<4 x i32>, i8*,
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) #0 {
; CHECK-LABEL: @test_x86_avx2_gather_q_d_256(
-; CHECK-NEXT: [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i32>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load i64, i64* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to i64*), align 8
-; CHECK-NEXT: [[TMP3:%.*]] = load <4 x i64>, <4 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 24) to <4 x i64>*), align 8
-; CHECK-NEXT: [[TMP4:%.*]] = load <4 x i32>, <4 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 56) to <4 x i32>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i32>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load i64, i64* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to i64*), align 8
+; CHECK-DAG: [[TMP3:%.*]] = load <4 x i64>, <4 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 24) to <4 x i64>*), align 8
+; CHECK-DAG: [[TMP4:%.*]] = load <4 x i32>, <4 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 56) to <4 x i32>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP5:%.*]] = bitcast <4 x i32> [[TMP1]] to i128
; CHECK-NEXT: [[_MSCMP:%.*]] = icmp ne i128 [[TMP5]], 0
@@ -2318,11 +2318,11 @@ declare <4 x i32> @llvm.x86.avx2.gather.q.d.256(<4 x i32>, i8*,
define <8 x float> @test_gather_mask(<8 x float> %a0, float* %a, <8 x i32> %idx, <8 x float> %mask, float* nocapture %out) #0 {
; CHECK-LABEL: @test_gather_mask(
-; CHECK-NEXT: [[TMP1:%.*]] = load i64, i64* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to i64*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <8 x i32>, <8 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i32>*), align 8
-; CHECK-NEXT: [[TMP3:%.*]] = load <8 x i32>, <8 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 40) to <8 x i32>*), align 8
-; CHECK-NEXT: [[TMP4:%.*]] = load <8 x i32>, <8 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 72) to <8 x i32>*), align 8
-; CHECK-NEXT: [[TMP5:%.*]] = load i64, i64* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 104) to i64*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load i64, i64* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to i64*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <8 x i32>, <8 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i32>*), align 8
+; CHECK-DAG: [[TMP3:%.*]] = load <8 x i32>, <8 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 40) to <8 x i32>*), align 8
+; CHECK-DAG: [[TMP4:%.*]] = load <8 x i32>, <8 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 72) to <8 x i32>*), align 8
+; CHECK-DAG: [[TMP5:%.*]] = load i64, i64* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 104) to i64*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[A_I8:%.*]] = bitcast float* [[A:%.*]] to i8*
; CHECK-NEXT: [[TMP6:%.*]] = bitcast <8 x i32> [[TMP2]] to i256
@@ -2380,10 +2380,10 @@ define <8 x float> @test_gather_mask(<8 x float> %a0, float* %a, <8 x i32> %idx
define <2 x i64> @test_mask_demanded_bits(<2 x i64> %a0, i8* %a1, <2 x i64> %idx, <2 x i1> %mask) #0 {
; CHECK-LABEL: @test_mask_demanded_bits(
-; CHECK-NEXT: [[TMP1:%.*]] = load <2 x i1>, <2 x i1>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 40) to <2 x i1>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <2 x i64>, <2 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <2 x i64>*), align 8
-; CHECK-NEXT: [[TMP3:%.*]] = load i64, i64* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to i64*), align 8
-; CHECK-NEXT: [[TMP4:%.*]] = load <2 x i64>, <2 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 24) to <2 x i64>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <2 x i1>, <2 x i1>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 40) to <2 x i1>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <2 x i64>, <2 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <2 x i64>*), align 8
+; CHECK-DAG: [[TMP3:%.*]] = load i64, i64* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to i64*), align 8
+; CHECK-DAG: [[TMP4:%.*]] = load <2 x i64>, <2 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 24) to <2 x i64>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[_MSPROP:%.*]] = sext <2 x i1> [[TMP1]] to <2 x i64>
; CHECK-NEXT: [[MASK1:%.*]] = sext <2 x i1> [[MASK:%.*]] to <2 x i64>
diff --git a/llvm/test/Instrumentation/MemorySanitizer/sse-intrinsics-x86.ll b/llvm/test/Instrumentation/MemorySanitizer/sse-intrinsics-x86.ll
index c3b98628649e..2cfaaccd32eb 100644
--- a/llvm/test/Instrumentation/MemorySanitizer/sse-intrinsics-x86.ll
+++ b/llvm/test/Instrumentation/MemorySanitizer/sse-intrinsics-x86.ll
@@ -6,8 +6,8 @@ target triple = "x86_64-unknown-linux-gnu"
define <4 x float> @test_x86_sse_cmp_ps(<4 x float> %a0, <4 x float> %a1) #0 {
; CHECK-LABEL: @test_x86_sse_cmp_ps(
-; CHECK-NEXT: [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i32>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <4 x i32>, <4 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <4 x i32>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i32>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <4 x i32>, <4 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <4 x i32>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP3:%.*]] = or <4 x i32> [[TMP1]], [[TMP2]]
; CHECK-NEXT: [[TMP4:%.*]] = icmp ne <4 x i32> [[TMP3]], zeroinitializer
@@ -24,8 +24,8 @@ declare <4 x float> @llvm.x86.sse.cmp.ps(<4 x float>, <4 x float>, i8) nounwind
define <4 x float> @test_x86_sse_cmp_ss(<4 x float> %a0, <4 x float> %a1) #0 {
; CHECK-LABEL: @test_x86_sse_cmp_ss(
-; CHECK-NEXT: [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i32>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <4 x i32>, <4 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <4 x i32>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i32>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <4 x i32>, <4 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <4 x i32>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP3:%.*]] = or <4 x i32> [[TMP1]], [[TMP2]]
; CHECK-NEXT: [[TMP4:%.*]] = extractelement <4 x i32> [[TMP3]], i64 0
@@ -44,8 +44,8 @@ declare <4 x float> @llvm.x86.sse.cmp.ss(<4 x float>, <4 x float>, i8) nounwind
define i32 @test_x86_sse_comieq_ss(<4 x float> %a0, <4 x float> %a1) #0 {
; CHECK-LABEL: @test_x86_sse_comieq_ss(
-; CHECK-NEXT: [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i32>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <4 x i32>, <4 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <4 x i32>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i32>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <4 x i32>, <4 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <4 x i32>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP3:%.*]] = or <4 x i32> [[TMP1]], [[TMP2]]
; CHECK-NEXT: [[TMP4:%.*]] = extractelement <4 x i32> [[TMP3]], i64 0
@@ -63,8 +63,8 @@ declare i32 @llvm.x86.sse.comieq.ss(<4 x float>, <4 x float>) nounwind readnone
define i32 @test_x86_sse_comige_ss(<4 x float> %a0, <4 x float> %a1) #0 {
; CHECK-LABEL: @test_x86_sse_comige_ss(
-; CHECK-NEXT: [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i32>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <4 x i32>, <4 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <4 x i32>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i32>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <4 x i32>, <4 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <4 x i32>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP3:%.*]] = or <4 x i32> [[TMP1]], [[TMP2]]
; CHECK-NEXT: [[TMP4:%.*]] = extractelement <4 x i32> [[TMP3]], i64 0
@@ -82,8 +82,8 @@ declare i32 @llvm.x86.sse.comige.ss(<4 x float>, <4 x float>) nounwind readnone
define i32 @test_x86_sse_comigt_ss(<4 x float> %a0, <4 x float> %a1) #0 {
; CHECK-LABEL: @test_x86_sse_comigt_ss(
-; CHECK-NEXT: [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i32>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <4 x i32>, <4 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <4 x i32>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i32>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <4 x i32>, <4 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <4 x i32>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP3:%.*]] = or <4 x i32> [[TMP1]], [[TMP2]]
; CHECK-NEXT: [[TMP4:%.*]] = extractelement <4 x i32> [[TMP3]], i64 0
@@ -101,8 +101,8 @@ declare i32 @llvm.x86.sse.comigt.ss(<4 x float>, <4 x float>) nounwind readnone
define i32 @test_x86_sse_comile_ss(<4 x float> %a0, <4 x float> %a1) #0 {
; CHECK-LABEL: @test_x86_sse_comile_ss(
-; CHECK-NEXT: [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i32>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <4 x i32>, <4 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <4 x i32>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i32>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <4 x i32>, <4 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <4 x i32>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP3:%.*]] = or <4 x i32> [[TMP1]], [[TMP2]]
; CHECK-NEXT: [[TMP4:%.*]] = extractelement <4 x i32> [[TMP3]], i64 0
@@ -120,8 +120,8 @@ declare i32 @llvm.x86.sse.comile.ss(<4 x float>, <4 x float>) nounwind readnone
define i32 @test_x86_sse_comilt_ss(<4 x float> %a0, <4 x float> %a1) #0 {
; CHECK-LABEL: @test_x86_sse_comilt_ss(
-; CHECK-NEXT: [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i32>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <4 x i32>, <4 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <4 x i32>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i32>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <4 x i32>, <4 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <4 x i32>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP3:%.*]] = or <4 x i32> [[TMP1]], [[TMP2]]
; CHECK-NEXT: [[TMP4:%.*]] = extractelement <4 x i32> [[TMP3]], i64 0
@@ -139,8 +139,8 @@ declare i32 @llvm.x86.sse.comilt.ss(<4 x float>, <4 x float>) nounwind readnone
define i32 @test_x86_sse_comineq_ss(<4 x float> %a0, <4 x float> %a1) #0 {
; CHECK-LABEL: @test_x86_sse_comineq_ss(
-; CHECK-NEXT: [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i32>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <4 x i32>, <4 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <4 x i32>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i32>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <4 x i32>, <4 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <4 x i32>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP3:%.*]] = or <4 x i32> [[TMP1]], [[TMP2]]
; CHECK-NEXT: [[TMP4:%.*]] = extractelement <4 x i32> [[TMP3]], i64 0
@@ -158,7 +158,7 @@ declare i32 @llvm.x86.sse.comineq.ss(<4 x float>, <4 x float>) nounwind readnone
define i32 @test_x86_sse_cvtss2si(<4 x float> %a0) #0 {
; CHECK-LABEL: @test_x86_sse_cvtss2si(
-; CHECK-NEXT: [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i32>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i32>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP2:%.*]] = extractelement <4 x i32> [[TMP1]], i32 0
; CHECK-NEXT: [[_MSCMP:%.*]] = icmp ne i32 [[TMP2]], 0
@@ -179,7 +179,7 @@ declare i32 @llvm.x86.sse.cvtss2si(<4 x float>) nounwind readnone
define i32 @test_x86_sse_cvttss2si(<4 x float> %a0) #0 {
; CHECK-LABEL: @test_x86_sse_cvttss2si(
-; CHECK-NEXT: [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i32>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i32>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP2:%.*]] = extractelement <4 x i32> [[TMP1]], i32 0
; CHECK-NEXT: [[_MSCMP:%.*]] = icmp ne i32 [[TMP2]], 0
@@ -200,7 +200,7 @@ declare i32 @llvm.x86.sse.cvttss2si(<4 x float>) nounwind readnone
define void @test_x86_sse_ldmxcsr(i8* %a0) #0 {
; CHECK-LABEL: @test_x86_sse_ldmxcsr(
-; CHECK-NEXT: [[TMP1:%.*]] = load i64, i64* getelementptr inbounds ([100 x i64], [100 x i64]* @__msan_param_tls, i32 0, i32 0), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load i64, i64* getelementptr inbounds ([100 x i64], [100 x i64]* @__msan_param_tls, i32 0, i32 0), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP2:%.*]] = ptrtoint i8* [[A0:%.*]] to i64
; CHECK-NEXT: [[TMP3:%.*]] = xor i64 [[TMP2]], 87960930222080
@@ -230,8 +230,8 @@ declare void @llvm.x86.sse.ldmxcsr(i8*) nounwind
define <4 x float> @test_x86_sse_max_ps(<4 x float> %a0, <4 x float> %a1) #0 {
; CHECK-LABEL: @test_x86_sse_max_ps(
-; CHECK-NEXT: [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i32>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <4 x i32>, <4 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <4 x i32>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i32>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <4 x i32>, <4 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <4 x i32>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[_MSPROP:%.*]] = or <4 x i32> [[TMP1]], [[TMP2]]
; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.x86.sse.max.ps(<4 x float> [[A0:%.*]], <4 x float> [[A1:%.*]])
@@ -246,8 +246,8 @@ declare <4 x float> @llvm.x86.sse.max.ps(<4 x float>, <4 x float>) nounwind read
define <4 x float> @test_x86_sse_max_ss(<4 x float> %a0, <4 x float> %a1) #0 {
; CHECK-LABEL: @test_x86_sse_max_ss(
-; CHECK-NEXT: [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i32>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <4 x i32>, <4 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <4 x i32>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i32>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <4 x i32>, <4 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <4 x i32>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[_MSPROP:%.*]] = or <4 x i32> [[TMP1]], [[TMP2]]
; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.x86.sse.max.ss(<4 x float> [[A0:%.*]], <4 x float> [[A1:%.*]])
@@ -262,8 +262,8 @@ declare <4 x float> @llvm.x86.sse.max.ss(<4 x float>, <4 x float>) nounwind read
define <4 x float> @test_x86_sse_min_ps(<4 x float> %a0, <4 x float> %a1) #0 {
; CHECK-LABEL: @test_x86_sse_min_ps(
-; CHECK-NEXT: [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i32>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <4 x i32>, <4 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <4 x i32>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i32>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <4 x i32>, <4 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <4 x i32>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[_MSPROP:%.*]] = or <4 x i32> [[TMP1]], [[TMP2]]
; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.x86.sse.min.ps(<4 x float> [[A0:%.*]], <4 x float> [[A1:%.*]])
@@ -278,8 +278,8 @@ declare <4 x float> @llvm.x86.sse.min.ps(<4 x float>, <4 x float>) nounwind read
define <4 x float> @test_x86_sse_min_ss(<4 x float> %a0, <4 x float> %a1) #0 {
; CHECK-LABEL: @test_x86_sse_min_ss(
-; CHECK-NEXT: [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i32>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <4 x i32>, <4 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <4 x i32>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i32>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <4 x i32>, <4 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <4 x i32>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[_MSPROP:%.*]] = or <4 x i32> [[TMP1]], [[TMP2]]
; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.x86.sse.min.ss(<4 x float> [[A0:%.*]], <4 x float> [[A1:%.*]])
@@ -294,7 +294,7 @@ declare <4 x float> @llvm.x86.sse.min.ss(<4 x float>, <4 x float>) nounwind read
define i32 @test_x86_sse_movmsk_ps(<4 x float> %a0) #0 {
; CHECK-LABEL: @test_x86_sse_movmsk_ps(
-; CHECK-NEXT: [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i32>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i32>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP2:%.*]] = bitcast <4 x i32> [[TMP1]] to i128
; CHECK-NEXT: [[_MSCMP:%.*]] = icmp ne i128 [[TMP2]], 0
@@ -316,7 +316,7 @@ declare i32 @llvm.x86.sse.movmsk.ps(<4 x float>) nounwind readnone
define <4 x float> @test_x86_sse_rcp_ps(<4 x float> %a0) #0 {
; CHECK-LABEL: @test_x86_sse_rcp_ps(
-; CHECK-NEXT: [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i32>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i32>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.x86.sse.rcp.ps(<4 x float> [[A0:%.*]])
; CHECK-NEXT: store <4 x i32> [[TMP1]], <4 x i32>* bitcast ([100 x i64]* @__msan_retval_tls to <4 x i32>*), align 8
@@ -330,7 +330,7 @@ declare <4 x float> @llvm.x86.sse.rcp.ps(<4 x float>) nounwind readnone
define <4 x float> @test_x86_sse_rcp_ss(<4 x float> %a0) #0 {
; CHECK-LABEL: @test_x86_sse_rcp_ss(
-; CHECK-NEXT: [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i32>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i32>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.x86.sse.rcp.ss(<4 x float> [[A0:%.*]])
; CHECK-NEXT: store <4 x i32> [[TMP1]], <4 x i32>* bitcast ([100 x i64]* @__msan_retval_tls to <4 x i32>*), align 8
@@ -344,7 +344,7 @@ declare <4 x float> @llvm.x86.sse.rcp.ss(<4 x float>) nounwind readnone
define <4 x float> @test_x86_sse_rsqrt_ps(<4 x float> %a0) #0 {
; CHECK-LABEL: @test_x86_sse_rsqrt_ps(
-; CHECK-NEXT: [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i32>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i32>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.x86.sse.rsqrt.ps(<4 x float> [[A0:%.*]])
; CHECK-NEXT: store <4 x i32> [[TMP1]], <4 x i32>* bitcast ([100 x i64]* @__msan_retval_tls to <4 x i32>*), align 8
@@ -358,7 +358,7 @@ declare <4 x float> @llvm.x86.sse.rsqrt.ps(<4 x float>) nounwind readnone
define <4 x float> @test_x86_sse_rsqrt_ss(<4 x float> %a0) #0 {
; CHECK-LABEL: @test_x86_sse_rsqrt_ss(
-; CHECK-NEXT: [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i32>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i32>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[RES:%.*]] = call <4 x float> @llvm.x86.sse.rsqrt.ss(<4 x float> [[A0:%.*]])
; CHECK-NEXT: store <4 x i32> [[TMP1]], <4 x i32>* bitcast ([100 x i64]* @__msan_retval_tls to <4 x i32>*), align 8
@@ -372,7 +372,7 @@ declare <4 x float> @llvm.x86.sse.rsqrt.ss(<4 x float>) nounwind readnone
define void @test_x86_sse_stmxcsr(i8* %a0) #0 {
; CHECK-LABEL: @test_x86_sse_stmxcsr(
-; CHECK-NEXT: [[TMP1:%.*]] = load i64, i64* getelementptr inbounds ([100 x i64], [100 x i64]* @__msan_param_tls, i32 0, i32 0), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load i64, i64* getelementptr inbounds ([100 x i64], [100 x i64]* @__msan_param_tls, i32 0, i32 0), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP2:%.*]] = ptrtoint i8* [[A0:%.*]] to i64
; CHECK-NEXT: [[TMP3:%.*]] = xor i64 [[TMP2]], 87960930222080
@@ -395,8 +395,8 @@ declare void @llvm.x86.sse.stmxcsr(i8*) nounwind
define i32 @test_x86_sse_ucomieq_ss(<4 x float> %a0, <4 x float> %a1) #0 {
; CHECK-LABEL: @test_x86_sse_ucomieq_ss(
-; CHECK-NEXT: [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i32>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <4 x i32>, <4 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <4 x i32>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i32>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <4 x i32>, <4 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <4 x i32>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP3:%.*]] = or <4 x i32> [[TMP1]], [[TMP2]]
; CHECK-NEXT: [[TMP4:%.*]] = extractelement <4 x i32> [[TMP3]], i64 0
@@ -414,8 +414,8 @@ declare i32 @llvm.x86.sse.ucomieq.ss(<4 x float>, <4 x float>) nounwind readnone
define i32 @test_x86_sse_ucomige_ss(<4 x float> %a0, <4 x float> %a1) #0 {
; CHECK-LABEL: @test_x86_sse_ucomige_ss(
-; CHECK-NEXT: [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i32>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <4 x i32>, <4 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <4 x i32>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i32>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <4 x i32>, <4 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <4 x i32>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP3:%.*]] = or <4 x i32> [[TMP1]], [[TMP2]]
; CHECK-NEXT: [[TMP4:%.*]] = extractelement <4 x i32> [[TMP3]], i64 0
@@ -433,8 +433,8 @@ declare i32 @llvm.x86.sse.ucomige.ss(<4 x float>, <4 x float>) nounwind readnone
define i32 @test_x86_sse_ucomigt_ss(<4 x float> %a0, <4 x float> %a1) #0 {
; CHECK-LABEL: @test_x86_sse_ucomigt_ss(
-; CHECK-NEXT: [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i32>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <4 x i32>, <4 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <4 x i32>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i32>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <4 x i32>, <4 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <4 x i32>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP3:%.*]] = or <4 x i32> [[TMP1]], [[TMP2]]
; CHECK-NEXT: [[TMP4:%.*]] = extractelement <4 x i32> [[TMP3]], i64 0
@@ -452,8 +452,8 @@ declare i32 @llvm.x86.sse.ucomigt.ss(<4 x float>, <4 x float>) nounwind readnone
define i32 @test_x86_sse_ucomile_ss(<4 x float> %a0, <4 x float> %a1) #0 {
; CHECK-LABEL: @test_x86_sse_ucomile_ss(
-; CHECK-NEXT: [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i32>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <4 x i32>, <4 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <4 x i32>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i32>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <4 x i32>, <4 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <4 x i32>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP3:%.*]] = or <4 x i32> [[TMP1]], [[TMP2]]
; CHECK-NEXT: [[TMP4:%.*]] = extractelement <4 x i32> [[TMP3]], i64 0
@@ -471,8 +471,8 @@ declare i32 @llvm.x86.sse.ucomile.ss(<4 x float>, <4 x float>) nounwind readnone
define i32 @test_x86_sse_ucomilt_ss(<4 x float> %a0, <4 x float> %a1) #0 {
; CHECK-LABEL: @test_x86_sse_ucomilt_ss(
-; CHECK-NEXT: [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i32>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <4 x i32>, <4 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <4 x i32>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i32>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <4 x i32>, <4 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <4 x i32>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP3:%.*]] = or <4 x i32> [[TMP1]], [[TMP2]]
; CHECK-NEXT: [[TMP4:%.*]] = extractelement <4 x i32> [[TMP3]], i64 0
@@ -490,8 +490,8 @@ declare i32 @llvm.x86.sse.ucomilt.ss(<4 x float>, <4 x float>) nounwind readnone
define i32 @test_x86_sse_ucomineq_ss(<4 x float> %a0, <4 x float> %a1) #0 {
; CHECK-LABEL: @test_x86_sse_ucomineq_ss(
-; CHECK-NEXT: [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i32>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <4 x i32>, <4 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <4 x i32>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i32>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <4 x i32>, <4 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <4 x i32>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP3:%.*]] = or <4 x i32> [[TMP1]], [[TMP2]]
; CHECK-NEXT: [[TMP4:%.*]] = extractelement <4 x i32> [[TMP3]], i64 0
diff --git a/llvm/test/Instrumentation/MemorySanitizer/sse2-intrinsics-x86.ll b/llvm/test/Instrumentation/MemorySanitizer/sse2-intrinsics-x86.ll
index ec03c7bf6a57..74083b49c875 100644
--- a/llvm/test/Instrumentation/MemorySanitizer/sse2-intrinsics-x86.ll
+++ b/llvm/test/Instrumentation/MemorySanitizer/sse2-intrinsics-x86.ll
@@ -6,8 +6,8 @@ target triple = "x86_64-unknown-linux-gnu"
define <2 x double> @test_x86_sse2_cmp_pd(<2 x double> %a0, <2 x double> %a1) #0 {
; CHECK-LABEL: @test_x86_sse2_cmp_pd(
-; CHECK-NEXT: [[TMP1:%.*]] = load <2 x i64>, <2 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <2 x i64>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <2 x i64>, <2 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <2 x i64>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <2 x i64>, <2 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <2 x i64>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <2 x i64>, <2 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <2 x i64>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP3:%.*]] = or <2 x i64> [[TMP1]], [[TMP2]]
; CHECK-NEXT: [[TMP4:%.*]] = icmp ne <2 x i64> [[TMP3]], zeroinitializer
@@ -24,8 +24,8 @@ declare <2 x double> @llvm.x86.sse2.cmp.pd(<2 x double>, <2 x double>, i8) nounw
define <2 x double> @test_x86_sse2_cmp_sd(<2 x double> %a0, <2 x double> %a1) #0 {
; CHECK-LABEL: @test_x86_sse2_cmp_sd(
-; CHECK-NEXT: [[TMP1:%.*]] = load <2 x i64>, <2 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <2 x i64>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <2 x i64>, <2 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <2 x i64>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <2 x i64>, <2 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <2 x i64>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <2 x i64>, <2 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <2 x i64>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP3:%.*]] = or <2 x i64> [[TMP1]], [[TMP2]]
; CHECK-NEXT: [[TMP4:%.*]] = extractelement <2 x i64> [[TMP3]], i64 0
@@ -44,8 +44,8 @@ declare <2 x double> @llvm.x86.sse2.cmp.sd(<2 x double>, <2 x double>, i8) nounw
define i32 @test_x86_sse2_comieq_sd(<2 x double> %a0, <2 x double> %a1) #0 {
; CHECK-LABEL: @test_x86_sse2_comieq_sd(
-; CHECK-NEXT: [[TMP1:%.*]] = load <2 x i64>, <2 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <2 x i64>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <2 x i64>, <2 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <2 x i64>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <2 x i64>, <2 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <2 x i64>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <2 x i64>, <2 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <2 x i64>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP3:%.*]] = or <2 x i64> [[TMP1]], [[TMP2]]
; CHECK-NEXT: [[TMP4:%.*]] = extractelement <2 x i64> [[TMP3]], i64 0
@@ -63,8 +63,8 @@ declare i32 @llvm.x86.sse2.comieq.sd(<2 x double>, <2 x double>) nounwind readno
define i32 @test_x86_sse2_comige_sd(<2 x double> %a0, <2 x double> %a1) #0 {
; CHECK-LABEL: @test_x86_sse2_comige_sd(
-; CHECK-NEXT: [[TMP1:%.*]] = load <2 x i64>, <2 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <2 x i64>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <2 x i64>, <2 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <2 x i64>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <2 x i64>, <2 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <2 x i64>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <2 x i64>, <2 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <2 x i64>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP3:%.*]] = or <2 x i64> [[TMP1]], [[TMP2]]
; CHECK-NEXT: [[TMP4:%.*]] = extractelement <2 x i64> [[TMP3]], i64 0
@@ -82,8 +82,8 @@ declare i32 @llvm.x86.sse2.comige.sd(<2 x double>, <2 x double>) nounwind readno
define i32 @test_x86_sse2_comigt_sd(<2 x double> %a0, <2 x double> %a1) #0 {
; CHECK-LABEL: @test_x86_sse2_comigt_sd(
-; CHECK-NEXT: [[TMP1:%.*]] = load <2 x i64>, <2 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <2 x i64>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <2 x i64>, <2 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <2 x i64>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <2 x i64>, <2 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <2 x i64>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <2 x i64>, <2 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <2 x i64>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP3:%.*]] = or <2 x i64> [[TMP1]], [[TMP2]]
; CHECK-NEXT: [[TMP4:%.*]] = extractelement <2 x i64> [[TMP3]], i64 0
@@ -101,8 +101,8 @@ declare i32 @llvm.x86.sse2.comigt.sd(<2 x double>, <2 x double>) nounwind readno
define i32 @test_x86_sse2_comile_sd(<2 x double> %a0, <2 x double> %a1) #0 {
; CHECK-LABEL: @test_x86_sse2_comile_sd(
-; CHECK-NEXT: [[TMP1:%.*]] = load <2 x i64>, <2 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <2 x i64>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <2 x i64>, <2 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <2 x i64>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <2 x i64>, <2 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <2 x i64>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <2 x i64>, <2 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <2 x i64>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP3:%.*]] = or <2 x i64> [[TMP1]], [[TMP2]]
; CHECK-NEXT: [[TMP4:%.*]] = extractelement <2 x i64> [[TMP3]], i64 0
@@ -120,8 +120,8 @@ declare i32 @llvm.x86.sse2.comile.sd(<2 x double>, <2 x double>) nounwind readno
define i32 @test_x86_sse2_comilt_sd(<2 x double> %a0, <2 x double> %a1) #0 {
; CHECK-LABEL: @test_x86_sse2_comilt_sd(
-; CHECK-NEXT: [[TMP1:%.*]] = load <2 x i64>, <2 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <2 x i64>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <2 x i64>, <2 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <2 x i64>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <2 x i64>, <2 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <2 x i64>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <2 x i64>, <2 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <2 x i64>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP3:%.*]] = or <2 x i64> [[TMP1]], [[TMP2]]
; CHECK-NEXT: [[TMP4:%.*]] = extractelement <2 x i64> [[TMP3]], i64 0
@@ -139,8 +139,8 @@ declare i32 @llvm.x86.sse2.comilt.sd(<2 x double>, <2 x double>) nounwind readno
define i32 @test_x86_sse2_comineq_sd(<2 x double> %a0, <2 x double> %a1) #0 {
; CHECK-LABEL: @test_x86_sse2_comineq_sd(
-; CHECK-NEXT: [[TMP1:%.*]] = load <2 x i64>, <2 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <2 x i64>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <2 x i64>, <2 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <2 x i64>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <2 x i64>, <2 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <2 x i64>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <2 x i64>, <2 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <2 x i64>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP3:%.*]] = or <2 x i64> [[TMP1]], [[TMP2]]
; CHECK-NEXT: [[TMP4:%.*]] = extractelement <2 x i64> [[TMP3]], i64 0
@@ -158,7 +158,7 @@ declare i32 @llvm.x86.sse2.comineq.sd(<2 x double>, <2 x double>) nounwind readn
define <4 x i32> @test_x86_sse2_cvtpd2dq(<2 x double> %a0) #0 {
; CHECK-LABEL: @test_x86_sse2_cvtpd2dq(
-; CHECK-NEXT: [[TMP1:%.*]] = load <2 x i64>, <2 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <2 x i64>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <2 x i64>, <2 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <2 x i64>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP2:%.*]] = bitcast <2 x i64> [[TMP1]] to i128
; CHECK-NEXT: [[_MSCMP:%.*]] = icmp ne i128 [[TMP2]], 0
@@ -179,7 +179,7 @@ declare <4 x i32> @llvm.x86.sse2.cvtpd2dq(<2 x double>) nounwind readnone
define <2 x i64> @test_mm_cvtpd_epi32_zext(<2 x double> %a0) nounwind #0 {
; CHECK-LABEL: @test_mm_cvtpd_epi32_zext(
-; CHECK-NEXT: [[TMP1:%.*]] = load <2 x i64>, <2 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <2 x i64>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <2 x i64>, <2 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <2 x i64>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP2:%.*]] = bitcast <2 x i64> [[TMP1]] to i128
; CHECK-NEXT: [[_MSCMP:%.*]] = icmp ne i128 [[TMP2]], 0
@@ -203,7 +203,7 @@ define <2 x i64> @test_mm_cvtpd_epi32_zext(<2 x double> %a0) nounwind #0 {
define <2 x i64> @test_mm_cvtpd_epi32_zext_load(<2 x double>* %p0) nounwind #0 {
; CHECK-LABEL: @test_mm_cvtpd_epi32_zext_load(
-; CHECK-NEXT: [[TMP1:%.*]] = load i64, i64* getelementptr inbounds ([100 x i64], [100 x i64]* @__msan_param_tls, i32 0, i32 0), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load i64, i64* getelementptr inbounds ([100 x i64], [100 x i64]* @__msan_param_tls, i32 0, i32 0), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[_MSCMP:%.*]] = icmp ne i64 [[TMP1]], 0
; CHECK-NEXT: br i1 [[_MSCMP]], label [[TMP2:%.*]], label [[TMP3:%.*]], !prof [[PROF0]]
@@ -239,7 +239,7 @@ define <2 x i64> @test_mm_cvtpd_epi32_zext_load(<2 x double>* %p0) nounwind #0 {
define <4 x float> @test_x86_sse2_cvtpd2ps(<2 x double> %a0) #0 {
; CHECK-LABEL: @test_x86_sse2_cvtpd2ps(
-; CHECK-NEXT: [[TMP1:%.*]] = load <2 x i64>, <2 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <2 x i64>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <2 x i64>, <2 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <2 x i64>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP2:%.*]] = bitcast <2 x i64> [[TMP1]] to i128
; CHECK-NEXT: [[_MSCMP:%.*]] = icmp ne i128 [[TMP2]], 0
@@ -259,7 +259,7 @@ declare <4 x float> @llvm.x86.sse2.cvtpd2ps(<2 x double>) nounwind readnone
define <4 x float> @test_x86_sse2_cvtpd2ps_zext(<2 x double> %a0) nounwind #0 {
; CHECK-LABEL: @test_x86_sse2_cvtpd2ps_zext(
-; CHECK-NEXT: [[TMP1:%.*]] = load <2 x i64>, <2 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <2 x i64>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <2 x i64>, <2 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <2 x i64>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP2:%.*]] = bitcast <2 x i64> [[TMP1]] to i128
; CHECK-NEXT: [[_MSCMP:%.*]] = icmp ne i128 [[TMP2]], 0
@@ -280,7 +280,7 @@ define <4 x float> @test_x86_sse2_cvtpd2ps_zext(<2 x double> %a0) nounwind #0 {
define <4 x float> @test_x86_sse2_cvtpd2ps_zext_load(<2 x double>* %p0) nounwind #0 {
; CHECK-LABEL: @test_x86_sse2_cvtpd2ps_zext_load(
-; CHECK-NEXT: [[TMP1:%.*]] = load i64, i64* getelementptr inbounds ([100 x i64], [100 x i64]* @__msan_param_tls, i32 0, i32 0), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load i64, i64* getelementptr inbounds ([100 x i64], [100 x i64]* @__msan_param_tls, i32 0, i32 0), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[_MSCMP:%.*]] = icmp ne i64 [[TMP1]], 0
; CHECK-NEXT: br i1 [[_MSCMP]], label [[TMP2:%.*]], label [[TMP3:%.*]], !prof [[PROF0]]
@@ -313,7 +313,7 @@ define <4 x float> @test_x86_sse2_cvtpd2ps_zext_load(<2 x double>* %p0) nounwind
define <4 x i32> @test_x86_sse2_cvtps2dq(<4 x float> %a0) #0 {
; CHECK-LABEL: @test_x86_sse2_cvtps2dq(
-; CHECK-NEXT: [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i32>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i32>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP2:%.*]] = bitcast <4 x i32> [[TMP1]] to i128
; CHECK-NEXT: [[_MSCMP:%.*]] = icmp ne i128 [[TMP2]], 0
@@ -334,7 +334,7 @@ declare <4 x i32> @llvm.x86.sse2.cvtps2dq(<4 x float>) nounwind readnone
define i32 @test_x86_sse2_cvtsd2si(<2 x double> %a0) #0 {
; CHECK-LABEL: @test_x86_sse2_cvtsd2si(
-; CHECK-NEXT: [[TMP1:%.*]] = load <2 x i64>, <2 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <2 x i64>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <2 x i64>, <2 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <2 x i64>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP2:%.*]] = extractelement <2 x i64> [[TMP1]], i32 0
; CHECK-NEXT: [[_MSCMP:%.*]] = icmp ne i64 [[TMP2]], 0
@@ -355,8 +355,8 @@ declare i32 @llvm.x86.sse2.cvtsd2si(<2 x double>) nounwind readnone
define <4 x float> @test_x86_sse2_cvtsd2ss(<4 x float> %a0, <2 x double> %a1) #0 {
; CHECK-LABEL: @test_x86_sse2_cvtsd2ss(
-; CHECK-NEXT: [[TMP1:%.*]] = load <2 x i64>, <2 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <2 x i64>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i32>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <2 x i64>, <2 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <2 x i64>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i32>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP3:%.*]] = extractelement <2 x i64> [[TMP1]], i32 0
; CHECK-NEXT: [[TMP4:%.*]] = insertelement <4 x i32> [[TMP2]], i32 0, i32 0
@@ -378,8 +378,8 @@ declare <4 x float> @llvm.x86.sse2.cvtsd2ss(<4 x float>, <2 x double>) nounwind
define <4 x float> @test_x86_sse2_cvtsd2ss_load(<4 x float> %a0, <2 x double>* %p1) #0 {
; CHECK-LABEL: @test_x86_sse2_cvtsd2ss_load(
-; CHECK-NEXT: [[TMP1:%.*]] = load i64, i64* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to i64*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i32>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load i64, i64* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to i64*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i32>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[_MSCMP:%.*]] = icmp ne i64 [[TMP1]], 0
; CHECK-NEXT: br i1 [[_MSCMP]], label [[TMP3:%.*]], label [[TMP4:%.*]], !prof [[PROF0]]
@@ -412,8 +412,8 @@ define <4 x float> @test_x86_sse2_cvtsd2ss_load(<4 x float> %a0, <2 x double>* %
define <4 x float> @test_x86_sse2_cvtsd2ss_load_optsize(<4 x float> %a0, <2 x double>* %p1) optsize #0 {
; CHECK-LABEL: @test_x86_sse2_cvtsd2ss_load_optsize(
-; CHECK-NEXT: [[TMP1:%.*]] = load i64, i64* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to i64*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i32>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load i64, i64* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to i64*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i32>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[_MSCMP:%.*]] = icmp ne i64 [[TMP1]], 0
; CHECK-NEXT: br i1 [[_MSCMP]], label [[TMP3:%.*]], label [[TMP4:%.*]], !prof [[PROF0]]
@@ -446,7 +446,7 @@ define <4 x float> @test_x86_sse2_cvtsd2ss_load_optsize(<4 x float> %a0, <2 x do
define <4 x i32> @test_x86_sse2_cvttpd2dq(<2 x double> %a0) #0 {
; CHECK-LABEL: @test_x86_sse2_cvttpd2dq(
-; CHECK-NEXT: [[TMP1:%.*]] = load <2 x i64>, <2 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <2 x i64>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <2 x i64>, <2 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <2 x i64>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP2:%.*]] = bitcast <2 x i64> [[TMP1]] to i128
; CHECK-NEXT: [[_MSCMP:%.*]] = icmp ne i128 [[TMP2]], 0
@@ -467,7 +467,7 @@ declare <4 x i32> @llvm.x86.sse2.cvttpd2dq(<2 x double>) nounwind readnone
define <2 x i64> @test_mm_cvttpd_epi32_zext(<2 x double> %a0) nounwind #0 {
; CHECK-LABEL: @test_mm_cvttpd_epi32_zext(
-; CHECK-NEXT: [[TMP1:%.*]] = load <2 x i64>, <2 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <2 x i64>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <2 x i64>, <2 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <2 x i64>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP2:%.*]] = bitcast <2 x i64> [[TMP1]] to i128
; CHECK-NEXT: [[_MSCMP:%.*]] = icmp ne i128 [[TMP2]], 0
@@ -491,7 +491,7 @@ define <2 x i64> @test_mm_cvttpd_epi32_zext(<2 x double> %a0) nounwind #0 {
define <2 x i64> @test_mm_cvttpd_epi32_zext_load(<2 x double>* %p0) nounwind #0 {
; CHECK-LABEL: @test_mm_cvttpd_epi32_zext_load(
-; CHECK-NEXT: [[TMP1:%.*]] = load i64, i64* getelementptr inbounds ([100 x i64], [100 x i64]* @__msan_param_tls, i32 0, i32 0), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load i64, i64* getelementptr inbounds ([100 x i64], [100 x i64]* @__msan_param_tls, i32 0, i32 0), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[_MSCMP:%.*]] = icmp ne i64 [[TMP1]], 0
; CHECK-NEXT: br i1 [[_MSCMP]], label [[TMP2:%.*]], label [[TMP3:%.*]], !prof [[PROF0]]
@@ -527,7 +527,7 @@ define <2 x i64> @test_mm_cvttpd_epi32_zext_load(<2 x double>* %p0) nounwind #0
define <4 x i32> @test_x86_sse2_cvttps2dq(<4 x float> %a0) #0 {
; CHECK-LABEL: @test_x86_sse2_cvttps2dq(
-; CHECK-NEXT: [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i32>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i32>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP2:%.*]] = bitcast <4 x i32> [[TMP1]] to i128
; CHECK-NEXT: [[_MSCMP:%.*]] = icmp ne i128 [[TMP2]], 0
@@ -548,7 +548,7 @@ declare <4 x i32> @llvm.x86.sse2.cvttps2dq(<4 x float>) nounwind readnone
define i32 @test_x86_sse2_cvttsd2si(<2 x double> %a0) #0 {
; CHECK-LABEL: @test_x86_sse2_cvttsd2si(
-; CHECK-NEXT: [[TMP1:%.*]] = load <2 x i64>, <2 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <2 x i64>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <2 x i64>, <2 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <2 x i64>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP2:%.*]] = extractelement <2 x i64> [[TMP1]], i32 0
; CHECK-NEXT: [[_MSCMP:%.*]] = icmp ne i64 [[TMP2]], 0
@@ -569,8 +569,8 @@ declare i32 @llvm.x86.sse2.cvttsd2si(<2 x double>) nounwind readnone
define <2 x double> @test_x86_sse2_max_pd(<2 x double> %a0, <2 x double> %a1) #0 {
; CHECK-LABEL: @test_x86_sse2_max_pd(
-; CHECK-NEXT: [[TMP1:%.*]] = load <2 x i64>, <2 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <2 x i64>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <2 x i64>, <2 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <2 x i64>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <2 x i64>, <2 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <2 x i64>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <2 x i64>, <2 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <2 x i64>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[_MSPROP:%.*]] = or <2 x i64> [[TMP1]], [[TMP2]]
; CHECK-NEXT: [[RES:%.*]] = call <2 x double> @llvm.x86.sse2.max.pd(<2 x double> [[A0:%.*]], <2 x double> [[A1:%.*]])
@@ -585,8 +585,8 @@ declare <2 x double> @llvm.x86.sse2.max.pd(<2 x double>, <2 x double>) nounwind
define <2 x double> @test_x86_sse2_max_sd(<2 x double> %a0, <2 x double> %a1) #0 {
; CHECK-LABEL: @test_x86_sse2_max_sd(
-; CHECK-NEXT: [[TMP1:%.*]] = load <2 x i64>, <2 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <2 x i64>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <2 x i64>, <2 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <2 x i64>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <2 x i64>, <2 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <2 x i64>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <2 x i64>, <2 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <2 x i64>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP3:%.*]] = or <2 x i64> [[TMP1]], [[TMP2]]
; CHECK-NEXT: [[TMP4:%.*]] = shufflevector <2 x i64> [[TMP1]], <2 x i64> [[TMP3]], <2 x i32> <i32 2, i32 1>
@@ -602,8 +602,8 @@ declare <2 x double> @llvm.x86.sse2.max.sd(<2 x double>, <2 x double>) nounwind
define <2 x double> @test_x86_sse2_min_pd(<2 x double> %a0, <2 x double> %a1) #0 {
; CHECK-LABEL: @test_x86_sse2_min_pd(
-; CHECK-NEXT: [[TMP1:%.*]] = load <2 x i64>, <2 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <2 x i64>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <2 x i64>, <2 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <2 x i64>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <2 x i64>, <2 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <2 x i64>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <2 x i64>, <2 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <2 x i64>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[_MSPROP:%.*]] = or <2 x i64> [[TMP1]], [[TMP2]]
; CHECK-NEXT: [[RES:%.*]] = call <2 x double> @llvm.x86.sse2.min.pd(<2 x double> [[A0:%.*]], <2 x double> [[A1:%.*]])
@@ -618,8 +618,8 @@ declare <2 x double> @llvm.x86.sse2.min.pd(<2 x double>, <2 x double>) nounwind
define <2 x double> @test_x86_sse2_min_sd(<2 x double> %a0, <2 x double> %a1) #0 {
; CHECK-LABEL: @test_x86_sse2_min_sd(
-; CHECK-NEXT: [[TMP1:%.*]] = load <2 x i64>, <2 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <2 x i64>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <2 x i64>, <2 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <2 x i64>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <2 x i64>, <2 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <2 x i64>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <2 x i64>, <2 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <2 x i64>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP3:%.*]] = or <2 x i64> [[TMP1]], [[TMP2]]
; CHECK-NEXT: [[TMP4:%.*]] = shufflevector <2 x i64> [[TMP1]], <2 x i64> [[TMP3]], <2 x i32> <i32 2, i32 1>
@@ -635,7 +635,7 @@ declare <2 x double> @llvm.x86.sse2.min.sd(<2 x double>, <2 x double>) nounwind
define i32 @test_x86_sse2_movmsk_pd(<2 x double> %a0) #0 {
; CHECK-LABEL: @test_x86_sse2_movmsk_pd(
-; CHECK-NEXT: [[TMP1:%.*]] = load <2 x i64>, <2 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <2 x i64>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <2 x i64>, <2 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <2 x i64>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP2:%.*]] = bitcast <2 x i64> [[TMP1]] to i128
; CHECK-NEXT: [[_MSCMP:%.*]] = icmp ne i128 [[TMP2]], 0
@@ -656,8 +656,8 @@ declare i32 @llvm.x86.sse2.movmsk.pd(<2 x double>) nounwind readnone
define <8 x i16> @test_x86_sse2_packssdw_128(<4 x i32> %a0, <4 x i32> %a1) #0 {
; CHECK-LABEL: @test_x86_sse2_packssdw_128(
-; CHECK-NEXT: [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i32>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <4 x i32>, <4 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <4 x i32>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i32>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <4 x i32>, <4 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <4 x i32>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP3:%.*]] = icmp ne <4 x i32> [[TMP1]], zeroinitializer
; CHECK-NEXT: [[TMP4:%.*]] = sext <4 x i1> [[TMP3]] to <4 x i32>
@@ -689,8 +689,8 @@ define <8 x i16> @test_x86_sse2_packssdw_128_fold() #0 {
define <16 x i8> @test_x86_sse2_packsswb_128(<8 x i16> %a0, <8 x i16> %a1) #0 {
; CHECK-LABEL: @test_x86_sse2_packsswb_128(
-; CHECK-NEXT: [[TMP1:%.*]] = load <8 x i16>, <8 x i16>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i16>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <8 x i16>, <8 x i16>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <8 x i16>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <8 x i16>, <8 x i16>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i16>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <8 x i16>, <8 x i16>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <8 x i16>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP3:%.*]] = icmp ne <8 x i16> [[TMP1]], zeroinitializer
; CHECK-NEXT: [[TMP4:%.*]] = sext <8 x i1> [[TMP3]] to <8 x i16>
@@ -722,8 +722,8 @@ define <16 x i8> @test_x86_sse2_packsswb_128_fold() #0 {
define <16 x i8> @test_x86_sse2_packuswb_128(<8 x i16> %a0, <8 x i16> %a1) #0 {
; CHECK-LABEL: @test_x86_sse2_packuswb_128(
-; CHECK-NEXT: [[TMP1:%.*]] = load <8 x i16>, <8 x i16>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i16>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <8 x i16>, <8 x i16>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <8 x i16>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <8 x i16>, <8 x i16>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i16>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <8 x i16>, <8 x i16>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <8 x i16>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP3:%.*]] = icmp ne <8 x i16> [[TMP1]], zeroinitializer
; CHECK-NEXT: [[TMP4:%.*]] = sext <8 x i1> [[TMP3]] to <8 x i16>
@@ -755,8 +755,8 @@ define <16 x i8> @test_x86_sse2_packuswb_128_fold() #0 {
define <16 x i8> @test_x86_sse2_pavg_b(<16 x i8> %a0, <16 x i8> %a1) #0 {
; CHECK-LABEL: @test_x86_sse2_pavg_b(
-; CHECK-NEXT: [[TMP1:%.*]] = load <16 x i8>, <16 x i8>* bitcast ([100 x i64]* @__msan_param_tls to <16 x i8>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <16 x i8>, <16 x i8>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <16 x i8>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <16 x i8>, <16 x i8>* bitcast ([100 x i64]* @__msan_param_tls to <16 x i8>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <16 x i8>, <16 x i8>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <16 x i8>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[_MSPROP:%.*]] = or <16 x i8> [[TMP1]], [[TMP2]]
; CHECK-NEXT: [[RES:%.*]] = call <16 x i8> @llvm.x86.sse2.pavg.b(<16 x i8> [[A0:%.*]], <16 x i8> [[A1:%.*]])
@@ -771,8 +771,8 @@ declare <16 x i8> @llvm.x86.sse2.pavg.b(<16 x i8>, <16 x i8>) nounwind readnone
define <8 x i16> @test_x86_sse2_pavg_w(<8 x i16> %a0, <8 x i16> %a1) #0 {
; CHECK-LABEL: @test_x86_sse2_pavg_w(
-; CHECK-NEXT: [[TMP1:%.*]] = load <8 x i16>, <8 x i16>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i16>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <8 x i16>, <8 x i16>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <8 x i16>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <8 x i16>, <8 x i16>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i16>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <8 x i16>, <8 x i16>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <8 x i16>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[_MSPROP:%.*]] = or <8 x i16> [[TMP1]], [[TMP2]]
; CHECK-NEXT: [[RES:%.*]] = call <8 x i16> @llvm.x86.sse2.pavg.w(<8 x i16> [[A0:%.*]], <8 x i16> [[A1:%.*]])
@@ -787,8 +787,8 @@ declare <8 x i16> @llvm.x86.sse2.pavg.w(<8 x i16>, <8 x i16>) nounwind readnone
define <4 x i32> @test_x86_sse2_pmadd_wd(<8 x i16> %a0, <8 x i16> %a1) #0 {
; CHECK-LABEL: @test_x86_sse2_pmadd_wd(
-; CHECK-NEXT: [[TMP1:%.*]] = load <8 x i16>, <8 x i16>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i16>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <8 x i16>, <8 x i16>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <8 x i16>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <8 x i16>, <8 x i16>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i16>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <8 x i16>, <8 x i16>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <8 x i16>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP3:%.*]] = or <8 x i16> [[TMP1]], [[TMP2]]
; CHECK-NEXT: [[TMP4:%.*]] = bitcast <8 x i16> [[TMP3]] to <4 x i32>
@@ -806,7 +806,7 @@ declare <4 x i32> @llvm.x86.sse2.pmadd.wd(<8 x i16>, <8 x i16>) nounwind readnon
define i32 @test_x86_sse2_pmovmskb_128(<16 x i8> %a0) #0 {
; CHECK-LABEL: @test_x86_sse2_pmovmskb_128(
-; CHECK-NEXT: [[TMP1:%.*]] = load <16 x i8>, <16 x i8>* bitcast ([100 x i64]* @__msan_param_tls to <16 x i8>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <16 x i8>, <16 x i8>* bitcast ([100 x i64]* @__msan_param_tls to <16 x i8>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP2:%.*]] = bitcast <16 x i8> [[TMP1]] to i128
; CHECK-NEXT: [[_MSCMP:%.*]] = icmp ne i128 [[TMP2]], 0
@@ -827,8 +827,8 @@ declare i32 @llvm.x86.sse2.pmovmskb.128(<16 x i8>) nounwind readnone
define <8 x i16> @test_x86_sse2_pmulh_w(<8 x i16> %a0, <8 x i16> %a1) #0 {
; CHECK-LABEL: @test_x86_sse2_pmulh_w(
-; CHECK-NEXT: [[TMP1:%.*]] = load <8 x i16>, <8 x i16>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i16>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <8 x i16>, <8 x i16>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <8 x i16>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <8 x i16>, <8 x i16>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i16>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <8 x i16>, <8 x i16>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <8 x i16>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[_MSPROP:%.*]] = or <8 x i16> [[TMP1]], [[TMP2]]
; CHECK-NEXT: [[RES:%.*]] = call <8 x i16> @llvm.x86.sse2.pmulh.w(<8 x i16> [[A0:%.*]], <8 x i16> [[A1:%.*]])
@@ -843,8 +843,8 @@ declare <8 x i16> @llvm.x86.sse2.pmulh.w(<8 x i16>, <8 x i16>) nounwind readnone
define <8 x i16> @test_x86_sse2_pmulhu_w(<8 x i16> %a0, <8 x i16> %a1) #0 {
; CHECK-LABEL: @test_x86_sse2_pmulhu_w(
-; CHECK-NEXT: [[TMP1:%.*]] = load <8 x i16>, <8 x i16>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i16>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <8 x i16>, <8 x i16>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <8 x i16>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <8 x i16>, <8 x i16>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i16>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <8 x i16>, <8 x i16>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <8 x i16>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[_MSPROP:%.*]] = or <8 x i16> [[TMP1]], [[TMP2]]
; CHECK-NEXT: [[RES:%.*]] = call <8 x i16> @llvm.x86.sse2.pmulhu.w(<8 x i16> [[A0:%.*]], <8 x i16> [[A1:%.*]])
@@ -859,8 +859,8 @@ declare <8 x i16> @llvm.x86.sse2.pmulhu.w(<8 x i16>, <8 x i16>) nounwind readnon
define <2 x i64> @test_x86_sse2_psad_bw(<16 x i8> %a0, <16 x i8> %a1) #0 {
; CHECK-LABEL: @test_x86_sse2_psad_bw(
-; CHECK-NEXT: [[TMP1:%.*]] = load <16 x i8>, <16 x i8>* bitcast ([100 x i64]* @__msan_param_tls to <16 x i8>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <16 x i8>, <16 x i8>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <16 x i8>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <16 x i8>, <16 x i8>* bitcast ([100 x i64]* @__msan_param_tls to <16 x i8>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <16 x i8>, <16 x i8>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <16 x i8>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP3:%.*]] = or <16 x i8> [[TMP1]], [[TMP2]]
; CHECK-NEXT: [[TMP4:%.*]] = bitcast <16 x i8> [[TMP3]] to <2 x i64>
@@ -879,8 +879,8 @@ declare <2 x i64> @llvm.x86.sse2.psad.bw(<16 x i8>, <16 x i8>) nounwind readnone
define <4 x i32> @test_x86_sse2_psll_d(<4 x i32> %a0, <4 x i32> %a1) #0 {
; CHECK-LABEL: @test_x86_sse2_psll_d(
-; CHECK-NEXT: [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i32>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <4 x i32>, <4 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <4 x i32>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i32>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <4 x i32>, <4 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <4 x i32>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP3:%.*]] = bitcast <4 x i32> [[TMP2]] to i128
; CHECK-NEXT: [[TMP4:%.*]] = trunc i128 [[TMP3]] to i64
@@ -901,8 +901,8 @@ declare <4 x i32> @llvm.x86.sse2.psll.d(<4 x i32>, <4 x i32>) nounwind readnone
define <2 x i64> @test_x86_sse2_psll_q(<2 x i64> %a0, <2 x i64> %a1) #0 {
; CHECK-LABEL: @test_x86_sse2_psll_q(
-; CHECK-NEXT: [[TMP1:%.*]] = load <2 x i64>, <2 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <2 x i64>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <2 x i64>, <2 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <2 x i64>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <2 x i64>, <2 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <2 x i64>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <2 x i64>, <2 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <2 x i64>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP3:%.*]] = bitcast <2 x i64> [[TMP2]] to i128
; CHECK-NEXT: [[TMP4:%.*]] = trunc i128 [[TMP3]] to i64
@@ -923,8 +923,8 @@ declare <2 x i64> @llvm.x86.sse2.psll.q(<2 x i64>, <2 x i64>) nounwind readnone
define <8 x i16> @test_x86_sse2_psll_w(<8 x i16> %a0, <8 x i16> %a1) #0 {
; CHECK-LABEL: @test_x86_sse2_psll_w(
-; CHECK-NEXT: [[TMP1:%.*]] = load <8 x i16>, <8 x i16>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i16>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <8 x i16>, <8 x i16>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <8 x i16>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <8 x i16>, <8 x i16>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i16>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <8 x i16>, <8 x i16>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <8 x i16>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP3:%.*]] = bitcast <8 x i16> [[TMP2]] to i128
; CHECK-NEXT: [[TMP4:%.*]] = trunc i128 [[TMP3]] to i64
@@ -945,7 +945,7 @@ declare <8 x i16> @llvm.x86.sse2.psll.w(<8 x i16>, <8 x i16>) nounwind readnone
define <4 x i32> @test_x86_sse2_pslli_d(<4 x i32> %a0) #0 {
; CHECK-LABEL: @test_x86_sse2_pslli_d(
-; CHECK-NEXT: [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i32>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i32>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP2:%.*]] = call <4 x i32> @llvm.x86.sse2.pslli.d(<4 x i32> [[TMP1]], i32 7)
; CHECK-NEXT: [[TMP3:%.*]] = or <4 x i32> [[TMP2]], zeroinitializer
@@ -961,7 +961,7 @@ declare <4 x i32> @llvm.x86.sse2.pslli.d(<4 x i32>, i32) nounwind readnone
define <2 x i64> @test_x86_sse2_pslli_q(<2 x i64> %a0) #0 {
; CHECK-LABEL: @test_x86_sse2_pslli_q(
-; CHECK-NEXT: [[TMP1:%.*]] = load <2 x i64>, <2 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <2 x i64>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <2 x i64>, <2 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <2 x i64>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP2:%.*]] = call <2 x i64> @llvm.x86.sse2.pslli.q(<2 x i64> [[TMP1]], i32 7)
; CHECK-NEXT: [[TMP3:%.*]] = or <2 x i64> [[TMP2]], zeroinitializer
@@ -977,7 +977,7 @@ declare <2 x i64> @llvm.x86.sse2.pslli.q(<2 x i64>, i32) nounwind readnone
define <8 x i16> @test_x86_sse2_pslli_w(<8 x i16> %a0) #0 {
; CHECK-LABEL: @test_x86_sse2_pslli_w(
-; CHECK-NEXT: [[TMP1:%.*]] = load <8 x i16>, <8 x i16>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i16>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <8 x i16>, <8 x i16>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i16>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP2:%.*]] = call <8 x i16> @llvm.x86.sse2.pslli.w(<8 x i16> [[TMP1]], i32 7)
; CHECK-NEXT: [[TMP3:%.*]] = or <8 x i16> [[TMP2]], zeroinitializer
@@ -993,8 +993,8 @@ declare <8 x i16> @llvm.x86.sse2.pslli.w(<8 x i16>, i32) nounwind readnone
define <4 x i32> @test_x86_sse2_psra_d(<4 x i32> %a0, <4 x i32> %a1) #0 {
; CHECK-LABEL: @test_x86_sse2_psra_d(
-; CHECK-NEXT: [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i32>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <4 x i32>, <4 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <4 x i32>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i32>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <4 x i32>, <4 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <4 x i32>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP3:%.*]] = bitcast <4 x i32> [[TMP2]] to i128
; CHECK-NEXT: [[TMP4:%.*]] = trunc i128 [[TMP3]] to i64
@@ -1015,8 +1015,8 @@ declare <4 x i32> @llvm.x86.sse2.psra.d(<4 x i32>, <4 x i32>) nounwind readnone
define <8 x i16> @test_x86_sse2_psra_w(<8 x i16> %a0, <8 x i16> %a1) #0 {
; CHECK-LABEL: @test_x86_sse2_psra_w(
-; CHECK-NEXT: [[TMP1:%.*]] = load <8 x i16>, <8 x i16>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i16>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <8 x i16>, <8 x i16>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <8 x i16>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <8 x i16>, <8 x i16>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i16>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <8 x i16>, <8 x i16>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <8 x i16>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP3:%.*]] = bitcast <8 x i16> [[TMP2]] to i128
; CHECK-NEXT: [[TMP4:%.*]] = trunc i128 [[TMP3]] to i64
@@ -1037,7 +1037,7 @@ declare <8 x i16> @llvm.x86.sse2.psra.w(<8 x i16>, <8 x i16>) nounwind readnone
define <4 x i32> @test_x86_sse2_psrai_d(<4 x i32> %a0) #0 {
; CHECK-LABEL: @test_x86_sse2_psrai_d(
-; CHECK-NEXT: [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i32>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i32>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP2:%.*]] = call <4 x i32> @llvm.x86.sse2.psrai.d(<4 x i32> [[TMP1]], i32 7)
; CHECK-NEXT: [[TMP3:%.*]] = or <4 x i32> [[TMP2]], zeroinitializer
@@ -1053,7 +1053,7 @@ declare <4 x i32> @llvm.x86.sse2.psrai.d(<4 x i32>, i32) nounwind readnone
define <8 x i16> @test_x86_sse2_psrai_w(<8 x i16> %a0) #0 {
; CHECK-LABEL: @test_x86_sse2_psrai_w(
-; CHECK-NEXT: [[TMP1:%.*]] = load <8 x i16>, <8 x i16>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i16>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <8 x i16>, <8 x i16>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i16>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP2:%.*]] = call <8 x i16> @llvm.x86.sse2.psrai.w(<8 x i16> [[TMP1]], i32 7)
; CHECK-NEXT: [[TMP3:%.*]] = or <8 x i16> [[TMP2]], zeroinitializer
@@ -1069,8 +1069,8 @@ declare <8 x i16> @llvm.x86.sse2.psrai.w(<8 x i16>, i32) nounwind readnone
define <4 x i32> @test_x86_sse2_psrl_d(<4 x i32> %a0, <4 x i32> %a1) #0 {
; CHECK-LABEL: @test_x86_sse2_psrl_d(
-; CHECK-NEXT: [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i32>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <4 x i32>, <4 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <4 x i32>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i32>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <4 x i32>, <4 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <4 x i32>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP3:%.*]] = bitcast <4 x i32> [[TMP2]] to i128
; CHECK-NEXT: [[TMP4:%.*]] = trunc i128 [[TMP3]] to i64
@@ -1091,8 +1091,8 @@ declare <4 x i32> @llvm.x86.sse2.psrl.d(<4 x i32>, <4 x i32>) nounwind readnone
define <2 x i64> @test_x86_sse2_psrl_q(<2 x i64> %a0, <2 x i64> %a1) #0 {
; CHECK-LABEL: @test_x86_sse2_psrl_q(
-; CHECK-NEXT: [[TMP1:%.*]] = load <2 x i64>, <2 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <2 x i64>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <2 x i64>, <2 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <2 x i64>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <2 x i64>, <2 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <2 x i64>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <2 x i64>, <2 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <2 x i64>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP3:%.*]] = bitcast <2 x i64> [[TMP2]] to i128
; CHECK-NEXT: [[TMP4:%.*]] = trunc i128 [[TMP3]] to i64
@@ -1113,8 +1113,8 @@ declare <2 x i64> @llvm.x86.sse2.psrl.q(<2 x i64>, <2 x i64>) nounwind readnone
define <8 x i16> @test_x86_sse2_psrl_w(<8 x i16> %a0, <8 x i16> %a1) #0 {
; CHECK-LABEL: @test_x86_sse2_psrl_w(
-; CHECK-NEXT: [[TMP1:%.*]] = load <8 x i16>, <8 x i16>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i16>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <8 x i16>, <8 x i16>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <8 x i16>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <8 x i16>, <8 x i16>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i16>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <8 x i16>, <8 x i16>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <8 x i16>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP3:%.*]] = bitcast <8 x i16> [[TMP2]] to i128
; CHECK-NEXT: [[TMP4:%.*]] = trunc i128 [[TMP3]] to i64
@@ -1135,8 +1135,8 @@ declare <8 x i16> @llvm.x86.sse2.psrl.w(<8 x i16>, <8 x i16>) nounwind readnone
define <8 x i16> @test_x86_sse2_psrl_w_load(<8 x i16> %a0, <8 x i16>* %p) #0 {
; CHECK-LABEL: @test_x86_sse2_psrl_w_load(
-; CHECK-NEXT: [[TMP1:%.*]] = load i64, i64* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to i64*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <8 x i16>, <8 x i16>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i16>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load i64, i64* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to i64*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <8 x i16>, <8 x i16>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i16>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[_MSCMP:%.*]] = icmp ne i64 [[TMP1]], 0
; CHECK-NEXT: br i1 [[_MSCMP]], label [[TMP3:%.*]], label [[TMP4:%.*]], !prof [[PROF0]]
@@ -1168,7 +1168,7 @@ define <8 x i16> @test_x86_sse2_psrl_w_load(<8 x i16> %a0, <8 x i16>* %p) #0 {
define <4 x i32> @test_x86_sse2_psrli_d(<4 x i32> %a0) #0 {
; CHECK-LABEL: @test_x86_sse2_psrli_d(
-; CHECK-NEXT: [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i32>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i32>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP2:%.*]] = call <4 x i32> @llvm.x86.sse2.psrli.d(<4 x i32> [[TMP1]], i32 7)
; CHECK-NEXT: [[TMP3:%.*]] = or <4 x i32> [[TMP2]], zeroinitializer
@@ -1184,7 +1184,7 @@ declare <4 x i32> @llvm.x86.sse2.psrli.d(<4 x i32>, i32) nounwind readnone
define <2 x i64> @test_x86_sse2_psrli_q(<2 x i64> %a0) #0 {
; CHECK-LABEL: @test_x86_sse2_psrli_q(
-; CHECK-NEXT: [[TMP1:%.*]] = load <2 x i64>, <2 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <2 x i64>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <2 x i64>, <2 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <2 x i64>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP2:%.*]] = call <2 x i64> @llvm.x86.sse2.psrli.q(<2 x i64> [[TMP1]], i32 7)
; CHECK-NEXT: [[TMP3:%.*]] = or <2 x i64> [[TMP2]], zeroinitializer
@@ -1200,7 +1200,7 @@ declare <2 x i64> @llvm.x86.sse2.psrli.q(<2 x i64>, i32) nounwind readnone
define <8 x i16> @test_x86_sse2_psrli_w(<8 x i16> %a0) #0 {
; CHECK-LABEL: @test_x86_sse2_psrli_w(
-; CHECK-NEXT: [[TMP1:%.*]] = load <8 x i16>, <8 x i16>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i16>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <8 x i16>, <8 x i16>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i16>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP2:%.*]] = call <8 x i16> @llvm.x86.sse2.psrli.w(<8 x i16> [[TMP1]], i32 7)
; CHECK-NEXT: [[TMP3:%.*]] = or <8 x i16> [[TMP2]], zeroinitializer
@@ -1216,8 +1216,8 @@ declare <8 x i16> @llvm.x86.sse2.psrli.w(<8 x i16>, i32) nounwind readnone
define i32 @test_x86_sse2_ucomieq_sd(<2 x double> %a0, <2 x double> %a1) #0 {
; CHECK-LABEL: @test_x86_sse2_ucomieq_sd(
-; CHECK-NEXT: [[TMP1:%.*]] = load <2 x i64>, <2 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <2 x i64>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <2 x i64>, <2 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <2 x i64>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <2 x i64>, <2 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <2 x i64>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <2 x i64>, <2 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <2 x i64>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP3:%.*]] = or <2 x i64> [[TMP1]], [[TMP2]]
; CHECK-NEXT: [[TMP4:%.*]] = extractelement <2 x i64> [[TMP3]], i64 0
@@ -1235,8 +1235,8 @@ declare i32 @llvm.x86.sse2.ucomieq.sd(<2 x double>, <2 x double>) nounwind readn
define i32 @test_x86_sse2_ucomige_sd(<2 x double> %a0, <2 x double> %a1) #0 {
; CHECK-LABEL: @test_x86_sse2_ucomige_sd(
-; CHECK-NEXT: [[TMP1:%.*]] = load <2 x i64>, <2 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <2 x i64>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <2 x i64>, <2 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <2 x i64>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <2 x i64>, <2 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <2 x i64>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <2 x i64>, <2 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <2 x i64>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP3:%.*]] = or <2 x i64> [[TMP1]], [[TMP2]]
; CHECK-NEXT: [[TMP4:%.*]] = extractelement <2 x i64> [[TMP3]], i64 0
@@ -1254,8 +1254,8 @@ declare i32 @llvm.x86.sse2.ucomige.sd(<2 x double>, <2 x double>) nounwind readn
define i32 @test_x86_sse2_ucomigt_sd(<2 x double> %a0, <2 x double> %a1) #0 {
; CHECK-LABEL: @test_x86_sse2_ucomigt_sd(
-; CHECK-NEXT: [[TMP1:%.*]] = load <2 x i64>, <2 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <2 x i64>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <2 x i64>, <2 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <2 x i64>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <2 x i64>, <2 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <2 x i64>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <2 x i64>, <2 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <2 x i64>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP3:%.*]] = or <2 x i64> [[TMP1]], [[TMP2]]
; CHECK-NEXT: [[TMP4:%.*]] = extractelement <2 x i64> [[TMP3]], i64 0
@@ -1273,8 +1273,8 @@ declare i32 @llvm.x86.sse2.ucomigt.sd(<2 x double>, <2 x double>) nounwind readn
define i32 @test_x86_sse2_ucomile_sd(<2 x double> %a0, <2 x double> %a1) #0 {
; CHECK-LABEL: @test_x86_sse2_ucomile_sd(
-; CHECK-NEXT: [[TMP1:%.*]] = load <2 x i64>, <2 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <2 x i64>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <2 x i64>, <2 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <2 x i64>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <2 x i64>, <2 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <2 x i64>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <2 x i64>, <2 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <2 x i64>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP3:%.*]] = or <2 x i64> [[TMP1]], [[TMP2]]
; CHECK-NEXT: [[TMP4:%.*]] = extractelement <2 x i64> [[TMP3]], i64 0
@@ -1292,8 +1292,8 @@ declare i32 @llvm.x86.sse2.ucomile.sd(<2 x double>, <2 x double>) nounwind readn
define i32 @test_x86_sse2_ucomilt_sd(<2 x double> %a0, <2 x double> %a1) #0 {
; CHECK-LABEL: @test_x86_sse2_ucomilt_sd(
-; CHECK-NEXT: [[TMP1:%.*]] = load <2 x i64>, <2 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <2 x i64>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <2 x i64>, <2 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <2 x i64>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <2 x i64>, <2 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <2 x i64>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <2 x i64>, <2 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <2 x i64>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP3:%.*]] = or <2 x i64> [[TMP1]], [[TMP2]]
; CHECK-NEXT: [[TMP4:%.*]] = extractelement <2 x i64> [[TMP3]], i64 0
@@ -1311,8 +1311,8 @@ declare i32 @llvm.x86.sse2.ucomilt.sd(<2 x double>, <2 x double>) nounwind readn
define i32 @test_x86_sse2_ucomineq_sd(<2 x double> %a0, <2 x double> %a1) #0 {
; CHECK-LABEL: @test_x86_sse2_ucomineq_sd(
-; CHECK-NEXT: [[TMP1:%.*]] = load <2 x i64>, <2 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <2 x i64>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <2 x i64>, <2 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <2 x i64>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <2 x i64>, <2 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <2 x i64>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <2 x i64>, <2 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <2 x i64>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP3:%.*]] = or <2 x i64> [[TMP1]], [[TMP2]]
; CHECK-NEXT: [[TMP4:%.*]] = extractelement <2 x i64> [[TMP3]], i64 0
@@ -1362,7 +1362,7 @@ declare void @llvm.x86.sse2.mfence() nounwind
define void @clflush(i8* %p) nounwind #0 {
; CHECK-LABEL: @clflush(
-; CHECK-NEXT: [[TMP1:%.*]] = load i64, i64* getelementptr inbounds ([100 x i64], [100 x i64]* @__msan_param_tls, i32 0, i32 0), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load i64, i64* getelementptr inbounds ([100 x i64], [100 x i64]* @__msan_param_tls, i32 0, i32 0), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[_MSCMP:%.*]] = icmp ne i64 [[TMP1]], 0
; CHECK-NEXT: br i1 [[_MSCMP]], label [[TMP2:%.*]], label [[TMP3:%.*]], !prof [[PROF0]]
diff --git a/llvm/test/Instrumentation/MemorySanitizer/sse41-intrinsics-x86.ll b/llvm/test/Instrumentation/MemorySanitizer/sse41-intrinsics-x86.ll
index 53f902ee695b..0fd304e35874 100644
--- a/llvm/test/Instrumentation/MemorySanitizer/sse41-intrinsics-x86.ll
+++ b/llvm/test/Instrumentation/MemorySanitizer/sse41-intrinsics-x86.ll
@@ -6,9 +6,9 @@ target triple = "x86_64-unknown-linux-gnu"
define <2 x double> @test_x86_sse41_blendvpd(<2 x double> %a0, <2 x double> %a1, <2 x double> %a2) #0 {
; CHECK-LABEL: @test_x86_sse41_blendvpd(
-; CHECK-NEXT: [[TMP1:%.*]] = load <2 x i64>, <2 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <2 x i64>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <2 x i64>, <2 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <2 x i64>*), align 8
-; CHECK-NEXT: [[TMP3:%.*]] = load <2 x i64>, <2 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <2 x i64>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <2 x i64>, <2 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <2 x i64>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <2 x i64>, <2 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <2 x i64>*), align 8
+; CHECK-DAG: [[TMP3:%.*]] = load <2 x i64>, <2 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <2 x i64>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[_MSPROP:%.*]] = or <2 x i64> [[TMP1]], [[TMP2]]
; CHECK-NEXT: [[_MSPROP1:%.*]] = or <2 x i64> [[_MSPROP]], [[TMP3]]
@@ -24,9 +24,9 @@ declare <2 x double> @llvm.x86.sse41.blendvpd(<2 x double>, <2 x double>, <2 x d
define <4 x float> @test_x86_sse41_blendvps(<4 x float> %a0, <4 x float> %a1, <4 x float> %a2) #0 {
; CHECK-LABEL: @test_x86_sse41_blendvps(
-; CHECK-NEXT: [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i32>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <4 x i32>, <4 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <4 x i32>*), align 8
-; CHECK-NEXT: [[TMP3:%.*]] = load <4 x i32>, <4 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <4 x i32>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i32>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <4 x i32>, <4 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <4 x i32>*), align 8
+; CHECK-DAG: [[TMP3:%.*]] = load <4 x i32>, <4 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <4 x i32>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[_MSPROP:%.*]] = or <4 x i32> [[TMP1]], [[TMP2]]
; CHECK-NEXT: [[_MSPROP1:%.*]] = or <4 x i32> [[_MSPROP]], [[TMP3]]
@@ -42,8 +42,8 @@ declare <4 x float> @llvm.x86.sse41.blendvps(<4 x float>, <4 x float>, <4 x floa
define <2 x double> @test_x86_sse41_dppd(<2 x double> %a0, <2 x double> %a1) #0 {
; CHECK-LABEL: @test_x86_sse41_dppd(
-; CHECK-NEXT: [[TMP1:%.*]] = load <2 x i64>, <2 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <2 x i64>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <2 x i64>, <2 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <2 x i64>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <2 x i64>, <2 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <2 x i64>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <2 x i64>, <2 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <2 x i64>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP3:%.*]] = bitcast <2 x i64> [[TMP1]] to i128
; CHECK-NEXT: [[_MSCMP:%.*]] = icmp ne i128 [[TMP3]], 0
@@ -71,8 +71,8 @@ declare <2 x double> @llvm.x86.sse41.dppd(<2 x double>, <2 x double>, i8) nounwi
define <4 x float> @test_x86_sse41_dpps(<4 x float> %a0, <4 x float> %a1) #0 {
; CHECK-LABEL: @test_x86_sse41_dpps(
-; CHECK-NEXT: [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i32>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <4 x i32>, <4 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <4 x i32>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i32>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <4 x i32>, <4 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <4 x i32>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP3:%.*]] = bitcast <4 x i32> [[TMP1]] to i128
; CHECK-NEXT: [[_MSCMP:%.*]] = icmp ne i128 [[TMP3]], 0
@@ -100,8 +100,8 @@ declare <4 x float> @llvm.x86.sse41.dpps(<4 x float>, <4 x float>, i8) nounwind
define <4 x float> @test_x86_sse41_insertps(<4 x float> %a0, <4 x float> %a1) #0 {
; CHECK-LABEL: @test_x86_sse41_insertps(
-; CHECK-NEXT: [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i32>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <4 x i32>, <4 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <4 x i32>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i32>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <4 x i32>, <4 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <4 x i32>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP3:%.*]] = bitcast <4 x i32> [[TMP1]] to i128
; CHECK-NEXT: [[_MSCMP:%.*]] = icmp ne i128 [[TMP3]], 0
@@ -130,8 +130,8 @@ declare <4 x float> @llvm.x86.sse41.insertps(<4 x float>, <4 x float>, i8) nounw
define <8 x i16> @test_x86_sse41_mpsadbw(<16 x i8> %a0, <16 x i8> %a1) #0 {
; CHECK-LABEL: @test_x86_sse41_mpsadbw(
-; CHECK-NEXT: [[TMP1:%.*]] = load <16 x i8>, <16 x i8>* bitcast ([100 x i64]* @__msan_param_tls to <16 x i8>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <16 x i8>, <16 x i8>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <16 x i8>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <16 x i8>, <16 x i8>* bitcast ([100 x i64]* @__msan_param_tls to <16 x i8>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <16 x i8>, <16 x i8>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <16 x i8>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP3:%.*]] = bitcast <16 x i8> [[TMP1]] to i128
; CHECK-NEXT: [[_MSCMP:%.*]] = icmp ne i128 [[TMP3]], 0
@@ -158,8 +158,8 @@ declare <8 x i16> @llvm.x86.sse41.mpsadbw(<16 x i8>, <16 x i8>, i8) nounwind rea
define <8 x i16> @test_x86_sse41_mpsadbw_load_op0(<16 x i8>* %ptr, <16 x i8> %a1) #0 {
; CHECK-LABEL: @test_x86_sse41_mpsadbw_load_op0(
-; CHECK-NEXT: [[TMP1:%.*]] = load i64, i64* getelementptr inbounds ([100 x i64], [100 x i64]* @__msan_param_tls, i32 0, i32 0), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <16 x i8>, <16 x i8>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 8) to <16 x i8>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load i64, i64* getelementptr inbounds ([100 x i64], [100 x i64]* @__msan_param_tls, i32 0, i32 0), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <16 x i8>, <16 x i8>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 8) to <16 x i8>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[_MSCMP:%.*]] = icmp ne i64 [[TMP1]], 0
; CHECK-NEXT: br i1 [[_MSCMP]], label [[TMP3:%.*]], label [[TMP4:%.*]], !prof [[PROF0]]
@@ -197,8 +197,8 @@ define <8 x i16> @test_x86_sse41_mpsadbw_load_op0(<16 x i8>* %ptr, <16 x i8> %a1
define <8 x i16> @test_x86_sse41_packusdw(<4 x i32> %a0, <4 x i32> %a1) #0 {
; CHECK-LABEL: @test_x86_sse41_packusdw(
-; CHECK-NEXT: [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i32>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <4 x i32>, <4 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <4 x i32>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i32>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <4 x i32>, <4 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <4 x i32>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP3:%.*]] = icmp ne <4 x i32> [[TMP1]], zeroinitializer
; CHECK-NEXT: [[TMP4:%.*]] = sext <4 x i1> [[TMP3]] to <4 x i32>
@@ -230,9 +230,9 @@ define <8 x i16> @test_x86_sse41_packusdw_fold() #0 {
define <16 x i8> @test_x86_sse41_pblendvb(<16 x i8> %a0, <16 x i8> %a1, <16 x i8> %a2) #0 {
; CHECK-LABEL: @test_x86_sse41_pblendvb(
-; CHECK-NEXT: [[TMP1:%.*]] = load <16 x i8>, <16 x i8>* bitcast ([100 x i64]* @__msan_param_tls to <16 x i8>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <16 x i8>, <16 x i8>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <16 x i8>*), align 8
-; CHECK-NEXT: [[TMP3:%.*]] = load <16 x i8>, <16 x i8>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <16 x i8>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <16 x i8>, <16 x i8>* bitcast ([100 x i64]* @__msan_param_tls to <16 x i8>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <16 x i8>, <16 x i8>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <16 x i8>*), align 8
+; CHECK-DAG: [[TMP3:%.*]] = load <16 x i8>, <16 x i8>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <16 x i8>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[_MSPROP:%.*]] = or <16 x i8> [[TMP1]], [[TMP2]]
; CHECK-NEXT: [[_MSPROP1:%.*]] = or <16 x i8> [[_MSPROP]], [[TMP3]]
@@ -248,7 +248,7 @@ declare <16 x i8> @llvm.x86.sse41.pblendvb(<16 x i8>, <16 x i8>, <16 x i8>) noun
define <8 x i16> @test_x86_sse41_phminposuw(<8 x i16> %a0) #0 {
; CHECK-LABEL: @test_x86_sse41_phminposuw(
-; CHECK-NEXT: [[TMP1:%.*]] = load <8 x i16>, <8 x i16>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i16>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <8 x i16>, <8 x i16>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i16>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[RES:%.*]] = call <8 x i16> @llvm.x86.sse41.phminposuw(<8 x i16> [[A0:%.*]])
; CHECK-NEXT: store <8 x i16> [[TMP1]], <8 x i16>* bitcast ([100 x i64]* @__msan_retval_tls to <8 x i16>*), align 8
@@ -262,8 +262,8 @@ declare <8 x i16> @llvm.x86.sse41.phminposuw(<8 x i16>) nounwind readnone
define i32 @test_x86_sse41_ptestc(<2 x i64> %a0, <2 x i64> %a1) #0 {
; CHECK-LABEL: @test_x86_sse41_ptestc(
-; CHECK-NEXT: [[TMP1:%.*]] = load <2 x i64>, <2 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <2 x i64>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <2 x i64>, <2 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <2 x i64>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <2 x i64>, <2 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <2 x i64>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <2 x i64>, <2 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <2 x i64>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP3:%.*]] = bitcast <2 x i64> [[TMP1]] to i128
; CHECK-NEXT: [[_MSCMP:%.*]] = icmp ne i128 [[TMP3]], 0
@@ -291,8 +291,8 @@ declare i32 @llvm.x86.sse41.ptestc(<2 x i64>, <2 x i64>) nounwind readnone
define i32 @test_x86_sse41_ptestnzc(<2 x i64> %a0, <2 x i64> %a1) #0 {
; CHECK-LABEL: @test_x86_sse41_ptestnzc(
-; CHECK-NEXT: [[TMP1:%.*]] = load <2 x i64>, <2 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <2 x i64>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <2 x i64>, <2 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <2 x i64>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <2 x i64>, <2 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <2 x i64>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <2 x i64>, <2 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <2 x i64>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP3:%.*]] = bitcast <2 x i64> [[TMP1]] to i128
; CHECK-NEXT: [[_MSCMP:%.*]] = icmp ne i128 [[TMP3]], 0
@@ -320,8 +320,8 @@ declare i32 @llvm.x86.sse41.ptestnzc(<2 x i64>, <2 x i64>) nounwind readnone
define i32 @test_x86_sse41_ptestz(<2 x i64> %a0, <2 x i64> %a1) #0 {
; CHECK-LABEL: @test_x86_sse41_ptestz(
-; CHECK-NEXT: [[TMP1:%.*]] = load <2 x i64>, <2 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <2 x i64>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <2 x i64>, <2 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <2 x i64>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <2 x i64>, <2 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <2 x i64>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <2 x i64>, <2 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <2 x i64>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP3:%.*]] = bitcast <2 x i64> [[TMP1]] to i128
; CHECK-NEXT: [[_MSCMP:%.*]] = icmp ne i128 [[TMP3]], 0
@@ -349,7 +349,7 @@ declare i32 @llvm.x86.sse41.ptestz(<2 x i64>, <2 x i64>) nounwind readnone
define <2 x double> @test_x86_sse41_round_pd(<2 x double> %a0) #0 {
; CHECK-LABEL: @test_x86_sse41_round_pd(
-; CHECK-NEXT: [[TMP1:%.*]] = load <2 x i64>, <2 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <2 x i64>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <2 x i64>, <2 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <2 x i64>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP2:%.*]] = bitcast <2 x i64> [[TMP1]] to i128
; CHECK-NEXT: [[_MSCMP:%.*]] = icmp ne i128 [[TMP2]], 0
@@ -370,7 +370,7 @@ declare <2 x double> @llvm.x86.sse41.round.pd(<2 x double>, i32) nounwind readno
define <4 x float> @test_x86_sse41_round_ps(<4 x float> %a0) #0 {
; CHECK-LABEL: @test_x86_sse41_round_ps(
-; CHECK-NEXT: [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i32>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i32>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP2:%.*]] = bitcast <4 x i32> [[TMP1]] to i128
; CHECK-NEXT: [[_MSCMP:%.*]] = icmp ne i128 [[TMP2]], 0
@@ -391,8 +391,8 @@ declare <4 x float> @llvm.x86.sse41.round.ps(<4 x float>, i32) nounwind readnone
define <2 x double> @test_x86_sse41_round_sd(<2 x double> %a0, <2 x double> %a1) #0 {
; CHECK-LABEL: @test_x86_sse41_round_sd(
-; CHECK-NEXT: [[TMP1:%.*]] = load <2 x i64>, <2 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <2 x i64>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <2 x i64>, <2 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <2 x i64>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <2 x i64>, <2 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <2 x i64>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <2 x i64>, <2 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <2 x i64>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP3:%.*]] = shufflevector <2 x i64> [[TMP1]], <2 x i64> [[TMP2]], <2 x i32> <i32 2, i32 1>
; CHECK-NEXT: [[RES:%.*]] = call <2 x double> @llvm.x86.sse41.round.sd(<2 x double> [[A0:%.*]], <2 x double> [[A1:%.*]], i32 7)
@@ -407,8 +407,8 @@ declare <2 x double> @llvm.x86.sse41.round.sd(<2 x double>, <2 x double>, i32) n
define <2 x double> @test_x86_sse41_round_sd_load(<2 x double> %a0, <2 x double>* %a1) #0 {
; CHECK-LABEL: @test_x86_sse41_round_sd_load(
-; CHECK-NEXT: [[TMP1:%.*]] = load i64, i64* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to i64*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <2 x i64>, <2 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <2 x i64>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load i64, i64* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to i64*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <2 x i64>, <2 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <2 x i64>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[_MSCMP:%.*]] = icmp ne i64 [[TMP1]], 0
; CHECK-NEXT: br i1 [[_MSCMP]], label [[TMP3:%.*]], label [[TMP4:%.*]], !prof [[PROF0]]
@@ -434,8 +434,8 @@ define <2 x double> @test_x86_sse41_round_sd_load(<2 x double> %a0, <2 x double>
define <4 x float> @test_x86_sse41_round_ss(<4 x float> %a0, <4 x float> %a1) #0 {
; CHECK-LABEL: @test_x86_sse41_round_ss(
-; CHECK-NEXT: [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i32>*), align 8
-; CHECK-NEXT: [[TMP2:%.*]] = load <4 x i32>, <4 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <4 x i32>*), align 8
+; CHECK-DAG: [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i32>*), align 8
+; CHECK-DAG: [[TMP2:%.*]] = load <4 x i32>, <4 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <4 x i32>*), align 8
; CHECK-NEXT: call void @llvm.donothing()
; CHECK-NEXT: [[TMP3:%.*]] = bitcast <4 x i32> [[TMP1]] to i128
; CHECK-NEXT: [[_MSCMP:%.*]] = icmp ne i128 [[TMP3]], 0
More information about the llvm-commits
mailing list