aboutsummaryrefslogtreecommitdiffstats
path: root/test/Transforms/SeparateConstOffsetFromGEP/NVPTX/split-gep-and-gvn.ll
diff options
context:
space:
mode:
Diffstat (limited to 'test/Transforms/SeparateConstOffsetFromGEP/NVPTX/split-gep-and-gvn.ll')
-rw-r--r--test/Transforms/SeparateConstOffsetFromGEP/NVPTX/split-gep-and-gvn.ll96
1 files changed, 48 insertions, 48 deletions
diff --git a/test/Transforms/SeparateConstOffsetFromGEP/NVPTX/split-gep-and-gvn.ll b/test/Transforms/SeparateConstOffsetFromGEP/NVPTX/split-gep-and-gvn.ll
index d054a3b..9ee492d 100644
--- a/test/Transforms/SeparateConstOffsetFromGEP/NVPTX/split-gep-and-gvn.ll
+++ b/test/Transforms/SeparateConstOffsetFromGEP/NVPTX/split-gep-and-gvn.ll
@@ -21,25 +21,25 @@ define void @sum_of_array(i32 %x, i32 %y, float* nocapture %output) {
.preheader:
%0 = sext i32 %y to i64
%1 = sext i32 %x to i64
- %2 = getelementptr inbounds [32 x [32 x float]] addrspace(3)* @array, i64 0, i64 %1, i64 %0
+ %2 = getelementptr inbounds [32 x [32 x float]], [32 x [32 x float]] addrspace(3)* @array, i64 0, i64 %1, i64 %0
%3 = addrspacecast float addrspace(3)* %2 to float*
- %4 = load float* %3, align 4
+ %4 = load float, float* %3, align 4
%5 = fadd float %4, 0.000000e+00
%6 = add i32 %y, 1
%7 = sext i32 %6 to i64
- %8 = getelementptr inbounds [32 x [32 x float]] addrspace(3)* @array, i64 0, i64 %1, i64 %7
+ %8 = getelementptr inbounds [32 x [32 x float]], [32 x [32 x float]] addrspace(3)* @array, i64 0, i64 %1, i64 %7
%9 = addrspacecast float addrspace(3)* %8 to float*
- %10 = load float* %9, align 4
+ %10 = load float, float* %9, align 4
%11 = fadd float %5, %10
%12 = add i32 %x, 1
%13 = sext i32 %12 to i64
- %14 = getelementptr inbounds [32 x [32 x float]] addrspace(3)* @array, i64 0, i64 %13, i64 %0
+ %14 = getelementptr inbounds [32 x [32 x float]], [32 x [32 x float]] addrspace(3)* @array, i64 0, i64 %13, i64 %0
%15 = addrspacecast float addrspace(3)* %14 to float*
- %16 = load float* %15, align 4
+ %16 = load float, float* %15, align 4
%17 = fadd float %11, %16
- %18 = getelementptr inbounds [32 x [32 x float]] addrspace(3)* @array, i64 0, i64 %13, i64 %7
+ %18 = getelementptr inbounds [32 x [32 x float]], [32 x [32 x float]] addrspace(3)* @array, i64 0, i64 %13, i64 %7
%19 = addrspacecast float addrspace(3)* %18 to float*
- %20 = load float* %19, align 4
+ %20 = load float, float* %19, align 4
%21 = fadd float %17, %20
store float %21, float* %output, align 4
ret void
@@ -51,10 +51,10 @@ define void @sum_of_array(i32 %x, i32 %y, float* nocapture %output) {
; PTX: ld.shared.f32 {{%f[0-9]+}}, {{\[}}[[BASE_REG]]+132{{\]}}
; IR-LABEL: @sum_of_array(
-; IR: [[BASE_PTR:%[a-zA-Z0-9]+]] = getelementptr inbounds [32 x [32 x float]] addrspace(3)* @array, i64 0, i64 %{{[a-zA-Z0-9]+}}, i64 %{{[a-zA-Z0-9]+}}
-; IR: getelementptr float addrspace(3)* [[BASE_PTR]], i64 1
-; IR: getelementptr float addrspace(3)* [[BASE_PTR]], i64 32
-; IR: getelementptr float addrspace(3)* [[BASE_PTR]], i64 33
+; IR: [[BASE_PTR:%[a-zA-Z0-9]+]] = getelementptr inbounds [32 x [32 x float]], [32 x [32 x float]] addrspace(3)* @array, i64 0, i64 %{{[a-zA-Z0-9]+}}, i64 %{{[a-zA-Z0-9]+}}
+; IR: getelementptr float, float addrspace(3)* [[BASE_PTR]], i64 1
+; IR: getelementptr float, float addrspace(3)* [[BASE_PTR]], i64 32
+; IR: getelementptr float, float addrspace(3)* [[BASE_PTR]], i64 33
; @sum_of_array2 is very similar to @sum_of_array. The only difference is in
; the order of "sext" and "add" when computing the array indices. @sum_of_array
@@ -66,23 +66,23 @@ define void @sum_of_array2(i32 %x, i32 %y, float* nocapture %output) {
.preheader:
%0 = sext i32 %y to i64
%1 = sext i32 %x to i64
- %2 = getelementptr inbounds [32 x [32 x float]] addrspace(3)* @array, i64 0, i64 %1, i64 %0
+ %2 = getelementptr inbounds [32 x [32 x float]], [32 x [32 x float]] addrspace(3)* @array, i64 0, i64 %1, i64 %0
%3 = addrspacecast float addrspace(3)* %2 to float*
- %4 = load float* %3, align 4
+ %4 = load float, float* %3, align 4
%5 = fadd float %4, 0.000000e+00
%6 = add i64 %0, 1
- %7 = getelementptr inbounds [32 x [32 x float]] addrspace(3)* @array, i64 0, i64 %1, i64 %6
+ %7 = getelementptr inbounds [32 x [32 x float]], [32 x [32 x float]] addrspace(3)* @array, i64 0, i64 %1, i64 %6
%8 = addrspacecast float addrspace(3)* %7 to float*
- %9 = load float* %8, align 4
+ %9 = load float, float* %8, align 4
%10 = fadd float %5, %9
%11 = add i64 %1, 1
- %12 = getelementptr inbounds [32 x [32 x float]] addrspace(3)* @array, i64 0, i64 %11, i64 %0
+ %12 = getelementptr inbounds [32 x [32 x float]], [32 x [32 x float]] addrspace(3)* @array, i64 0, i64 %11, i64 %0
%13 = addrspacecast float addrspace(3)* %12 to float*
- %14 = load float* %13, align 4
+ %14 = load float, float* %13, align 4
%15 = fadd float %10, %14
- %16 = getelementptr inbounds [32 x [32 x float]] addrspace(3)* @array, i64 0, i64 %11, i64 %6
+ %16 = getelementptr inbounds [32 x [32 x float]], [32 x [32 x float]] addrspace(3)* @array, i64 0, i64 %11, i64 %6
%17 = addrspacecast float addrspace(3)* %16 to float*
- %18 = load float* %17, align 4
+ %18 = load float, float* %17, align 4
%19 = fadd float %15, %18
store float %19, float* %output, align 4
ret void
@@ -94,10 +94,10 @@ define void @sum_of_array2(i32 %x, i32 %y, float* nocapture %output) {
; PTX: ld.shared.f32 {{%f[0-9]+}}, {{\[}}[[BASE_REG]]+132{{\]}}
; IR-LABEL: @sum_of_array2(
-; IR: [[BASE_PTR:%[a-zA-Z0-9]+]] = getelementptr inbounds [32 x [32 x float]] addrspace(3)* @array, i64 0, i64 %{{[a-zA-Z0-9]+}}, i64 %{{[a-zA-Z0-9]+}}
-; IR: getelementptr float addrspace(3)* [[BASE_PTR]], i64 1
-; IR: getelementptr float addrspace(3)* [[BASE_PTR]], i64 32
-; IR: getelementptr float addrspace(3)* [[BASE_PTR]], i64 33
+; IR: [[BASE_PTR:%[a-zA-Z0-9]+]] = getelementptr inbounds [32 x [32 x float]], [32 x [32 x float]] addrspace(3)* @array, i64 0, i64 %{{[a-zA-Z0-9]+}}, i64 %{{[a-zA-Z0-9]+}}
+; IR: getelementptr float, float addrspace(3)* [[BASE_PTR]], i64 1
+; IR: getelementptr float, float addrspace(3)* [[BASE_PTR]], i64 32
+; IR: getelementptr float, float addrspace(3)* [[BASE_PTR]], i64 33
; This function loads
@@ -114,25 +114,25 @@ define void @sum_of_array3(i32 %x, i32 %y, float* nocapture %output) {
.preheader:
%0 = zext i32 %y to i64
%1 = zext i32 %x to i64
- %2 = getelementptr inbounds [32 x [32 x float]] addrspace(3)* @array, i64 0, i64 %1, i64 %0
+ %2 = getelementptr inbounds [32 x [32 x float]], [32 x [32 x float]] addrspace(3)* @array, i64 0, i64 %1, i64 %0
%3 = addrspacecast float addrspace(3)* %2 to float*
- %4 = load float* %3, align 4
+ %4 = load float, float* %3, align 4
%5 = fadd float %4, 0.000000e+00
%6 = add nuw i32 %y, 1
%7 = zext i32 %6 to i64
- %8 = getelementptr inbounds [32 x [32 x float]] addrspace(3)* @array, i64 0, i64 %1, i64 %7
+ %8 = getelementptr inbounds [32 x [32 x float]], [32 x [32 x float]] addrspace(3)* @array, i64 0, i64 %1, i64 %7
%9 = addrspacecast float addrspace(3)* %8 to float*
- %10 = load float* %9, align 4
+ %10 = load float, float* %9, align 4
%11 = fadd float %5, %10
%12 = add nuw i32 %x, 1
%13 = zext i32 %12 to i64
- %14 = getelementptr inbounds [32 x [32 x float]] addrspace(3)* @array, i64 0, i64 %13, i64 %0
+ %14 = getelementptr inbounds [32 x [32 x float]], [32 x [32 x float]] addrspace(3)* @array, i64 0, i64 %13, i64 %0
%15 = addrspacecast float addrspace(3)* %14 to float*
- %16 = load float* %15, align 4
+ %16 = load float, float* %15, align 4
%17 = fadd float %11, %16
- %18 = getelementptr inbounds [32 x [32 x float]] addrspace(3)* @array, i64 0, i64 %13, i64 %7
+ %18 = getelementptr inbounds [32 x [32 x float]], [32 x [32 x float]] addrspace(3)* @array, i64 0, i64 %13, i64 %7
%19 = addrspacecast float addrspace(3)* %18 to float*
- %20 = load float* %19, align 4
+ %20 = load float, float* %19, align 4
%21 = fadd float %17, %20
store float %21, float* %output, align 4
ret void
@@ -144,10 +144,10 @@ define void @sum_of_array3(i32 %x, i32 %y, float* nocapture %output) {
; PTX: ld.shared.f32 {{%f[0-9]+}}, {{\[}}[[BASE_REG]]+132{{\]}}
; IR-LABEL: @sum_of_array3(
-; IR: [[BASE_PTR:%[a-zA-Z0-9]+]] = getelementptr inbounds [32 x [32 x float]] addrspace(3)* @array, i64 0, i64 %{{[a-zA-Z0-9]+}}, i64 %{{[a-zA-Z0-9]+}}
-; IR: getelementptr float addrspace(3)* [[BASE_PTR]], i64 1
-; IR: getelementptr float addrspace(3)* [[BASE_PTR]], i64 32
-; IR: getelementptr float addrspace(3)* [[BASE_PTR]], i64 33
+; IR: [[BASE_PTR:%[a-zA-Z0-9]+]] = getelementptr inbounds [32 x [32 x float]], [32 x [32 x float]] addrspace(3)* @array, i64 0, i64 %{{[a-zA-Z0-9]+}}, i64 %{{[a-zA-Z0-9]+}}
+; IR: getelementptr float, float addrspace(3)* [[BASE_PTR]], i64 1
+; IR: getelementptr float, float addrspace(3)* [[BASE_PTR]], i64 32
+; IR: getelementptr float, float addrspace(3)* [[BASE_PTR]], i64 33
; This function loads
@@ -162,23 +162,23 @@ define void @sum_of_array4(i32 %x, i32 %y, float* nocapture %output) {
.preheader:
%0 = zext i32 %y to i64
%1 = zext i32 %x to i64
- %2 = getelementptr inbounds [32 x [32 x float]] addrspace(3)* @array, i64 0, i64 %1, i64 %0
+ %2 = getelementptr inbounds [32 x [32 x float]], [32 x [32 x float]] addrspace(3)* @array, i64 0, i64 %1, i64 %0
%3 = addrspacecast float addrspace(3)* %2 to float*
- %4 = load float* %3, align 4
+ %4 = load float, float* %3, align 4
%5 = fadd float %4, 0.000000e+00
%6 = add i64 %0, 1
- %7 = getelementptr inbounds [32 x [32 x float]] addrspace(3)* @array, i64 0, i64 %1, i64 %6
+ %7 = getelementptr inbounds [32 x [32 x float]], [32 x [32 x float]] addrspace(3)* @array, i64 0, i64 %1, i64 %6
%8 = addrspacecast float addrspace(3)* %7 to float*
- %9 = load float* %8, align 4
+ %9 = load float, float* %8, align 4
%10 = fadd float %5, %9
%11 = add i64 %1, 1
- %12 = getelementptr inbounds [32 x [32 x float]] addrspace(3)* @array, i64 0, i64 %11, i64 %0
+ %12 = getelementptr inbounds [32 x [32 x float]], [32 x [32 x float]] addrspace(3)* @array, i64 0, i64 %11, i64 %0
%13 = addrspacecast float addrspace(3)* %12 to float*
- %14 = load float* %13, align 4
+ %14 = load float, float* %13, align 4
%15 = fadd float %10, %14
- %16 = getelementptr inbounds [32 x [32 x float]] addrspace(3)* @array, i64 0, i64 %11, i64 %6
+ %16 = getelementptr inbounds [32 x [32 x float]], [32 x [32 x float]] addrspace(3)* @array, i64 0, i64 %11, i64 %6
%17 = addrspacecast float addrspace(3)* %16 to float*
- %18 = load float* %17, align 4
+ %18 = load float, float* %17, align 4
%19 = fadd float %15, %18
store float %19, float* %output, align 4
ret void
@@ -190,7 +190,7 @@ define void @sum_of_array4(i32 %x, i32 %y, float* nocapture %output) {
; PTX: ld.shared.f32 {{%f[0-9]+}}, {{\[}}[[BASE_REG]]+132{{\]}}
; IR-LABEL: @sum_of_array4(
-; IR: [[BASE_PTR:%[a-zA-Z0-9]+]] = getelementptr inbounds [32 x [32 x float]] addrspace(3)* @array, i64 0, i64 %{{[a-zA-Z0-9]+}}, i64 %{{[a-zA-Z0-9]+}}
-; IR: getelementptr float addrspace(3)* [[BASE_PTR]], i64 1
-; IR: getelementptr float addrspace(3)* [[BASE_PTR]], i64 32
-; IR: getelementptr float addrspace(3)* [[BASE_PTR]], i64 33
+; IR: [[BASE_PTR:%[a-zA-Z0-9]+]] = getelementptr inbounds [32 x [32 x float]], [32 x [32 x float]] addrspace(3)* @array, i64 0, i64 %{{[a-zA-Z0-9]+}}, i64 %{{[a-zA-Z0-9]+}}
+; IR: getelementptr float, float addrspace(3)* [[BASE_PTR]], i64 1
+; IR: getelementptr float, float addrspace(3)* [[BASE_PTR]], i64 32
+; IR: getelementptr float, float addrspace(3)* [[BASE_PTR]], i64 33
generated by cgit v1.1 at 2024-12-22 10:50:24 +0000