replica
/
hanchenye-llvm-project
2
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

[AMDGPU] Re-enabled 128bit wide-vector generation for local addr space by default. 

Summary: Bug reported here https://bugs.freedesktop.org/show_bug.cgi?id=105464 found
         to be resolved by some other fixes.

Author: FarhanaAleen
llvm-svn: 333380
This commit is contained in:
Farhana Aleen 2018-05-28 18:15:11 +00:00
parent 07ac63f89e
commit eacb1020aa
3 changed files with 5 additions and 9 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

8
llvm/lib/Target/AMDGPU/AMDGPUTargetTransformInfo.cpp
View File

@ -267,12 +267,10 @@ unsigned AMDGPUTTIImpl::getLoadStoreVecRegBitWidth(unsigned AddrSpace) const {
return 512; return 512;
} }
if (AddrSpace == AS.FLAT_ADDRESS) if (AddrSpace == AS.FLAT_ADDRESS ||
return 128; AddrSpace == AS.LOCAL_ADDRESS ||
if (AddrSpace == AS.LOCAL_ADDRESS ||
AddrSpace == AS.REGION_ADDRESS) AddrSpace == AS.REGION_ADDRESS)
return ST->useDS128() ? 128 : 64; return 128;
if (AddrSpace == AS.PRIVATE_ADDRESS) if (AddrSpace == AS.PRIVATE_ADDRESS)
return 8 * ST->getMaxPrivateElementSize(); return 8 * ST->getMaxPrivateElementSize();

3
llvm/test/Transforms/LoadStoreVectorizer/AMDGPU/merge-stores.ll
View File

@ -504,8 +504,7 @@ define amdgpu_kernel void @merge_local_store_2_constants_i32_align_2(i32 addrspa
} }
; CHECK-LABEL: @merge_local_store_4_constants_i32 ; CHECK-LABEL: @merge_local_store_4_constants_i32
; CHECK: store <2 x i32> <i32 456, i32 333>, <2 x i32> addrspace(3)* ; CHECK: store <4 x i32> <i32 1234, i32 123, i32 456, i32 333>, <4 x i32> addrspace(3)*
; CHECK: store <2 x i32> <i32 1234, i32 123>, <2 x i32> addrspace(3)*
define amdgpu_kernel void @merge_local_store_4_constants_i32(i32 addrspace(3)* %out) #0 { define amdgpu_kernel void @merge_local_store_4_constants_i32(i32 addrspace(3)* %out) #0 {
%out.gep.1 = getelementptr i32, i32 addrspace(3)* %out, i32 1 %out.gep.1 = getelementptr i32, i32 addrspace(3)* %out, i32 1
%out.gep.2 = getelementptr i32, i32 addrspace(3)* %out, i32 2 %out.gep.2 = getelementptr i32, i32 addrspace(3)* %out, i32 2

3
llvm/test/Transforms/LoadStoreVectorizer/AMDGPU/multiple_tails.ll
View File

@ -29,11 +29,10 @@ define amdgpu_kernel void @no_crash(i32 %arg) {
; longest chain vectorized ; longest chain vectorized
; CHECK-LABEL: @interleave_get_longest ; CHECK-LABEL: @interleave_get_longest
; CHECK: load <2 x i32> ; CHECK: load <4 x i32>
; CHECK: load i32 ; CHECK: load i32
; CHECK: store <2 x i32> zeroinitializer ; CHECK: store <2 x i32> zeroinitializer
; CHECK: load i32 ; CHECK: load i32
; CHECK: load <2 x i32>
; CHECK: load i32 ; CHECK: load i32
; CHECK: load i32 ; CHECK: load i32