|
1 |
| -; RUN: opt -cost-model -analyze -mtriple=amdgcn-unknown-amdhsa -mcpu=gfx900 %s | FileCheck -check-prefixes=GFX9,GCN %s |
2 |
| -; RUN: opt -cost-model -analyze -mtriple=amdgcn-unknown-amdhsa -mcpu=fiji %s | FileCheck -check-prefixes=VI,GCN %s |
3 |
| -; RUN: opt -cost-model -cost-kind=code-size -analyze -mtriple=amdgcn-unknown-amdhsa -mcpu=gfx900 %s | FileCheck -check-prefixes=GFX9,GCN %s |
4 |
| -; RUN: opt -cost-model -cost-kind=code-size -analyze -mtriple=amdgcn-unknown-amdhsa -mcpu=fiji %s | FileCheck -check-prefixes=VI,GCN %s |
| 1 | +; RUN: opt -cost-model -analyze -mtriple=amdgcn-unknown-amdhsa -mcpu=gfx900 %s | FileCheck -check-prefixes=GFX9,GCN,TPT %s |
| 2 | +; RUN: opt -cost-model -analyze -mtriple=amdgcn-unknown-amdhsa -mcpu=fiji %s | FileCheck -check-prefixes=VI,GCN,TPT %s |
| 3 | +; RUN: opt -cost-model -cost-kind=code-size -analyze -mtriple=amdgcn-unknown-amdhsa -mcpu=gfx900 %s | FileCheck -check-prefixes=GFX9,GCN,CS %s |
| 4 | +; RUN: opt -cost-model -cost-kind=code-size -analyze -mtriple=amdgcn-unknown-amdhsa -mcpu=fiji %s | FileCheck -check-prefixes=VI,GCN,CS %s |
5 | 5 |
|
6 | 6 | ; GCN-LABEL: 'shufflevector_00_v2i16'
|
7 | 7 | ; GFX9: estimated cost of 0 for {{.*}} shufflevector <2 x i16> %vec, <2 x i16> undef, <2 x i32> zeroinitializer
|
@@ -51,3 +51,14 @@ define amdgpu_kernel void @shufflevector_02_v2i16(<2 x i16> addrspace(1)* %out,
|
51 | 51 | store <2 x i16> %shuf, <2 x i16> addrspace(1)* %out
|
52 | 52 | ret void
|
53 | 53 | }
|
| 54 | + |
| 55 | +; GCN-LABEL: 'shufflevector_xxx' |
| 56 | +; TPT: Unknown cost for {{.*}} shufflevector <2 x i8> %vec, <2 x i8> undef, <4 x i32> <i32 0, i32 1, i32 undef, i32 undef> |
| 57 | +; CS: estimated cost of 1 for {{.*}} shufflevector <2 x i8> %vec, <2 x i8> undef, <4 x i32> <i32 0, i32 1, i32 undef, i32 undef> |
| 58 | +; Should not assert |
| 59 | +define amdgpu_kernel void @shufflevector_xxx(<4 x i8> addrspace(1)* %out, <2 x i8> addrspace(1)* %vaddr) { |
| 60 | + %vec = load <2 x i8>, <2 x i8> addrspace(1)* %vaddr |
| 61 | + %shuf = shufflevector <2 x i8> %vec, <2 x i8> undef, <4 x i32> <i32 0, i32 1, i32 undef, i32 undef> |
| 62 | + store <4 x i8> %shuf, <4 x i8> addrspace(1)* %out |
| 63 | + ret void |
| 64 | +} |
0 commit comments