RPCS3/llvm-mirror
1
0
Fork 0
mirror of https://github.com/RPCS3/llvm-mirror.git synced 2024-11-22 10:42:39 +01:00
Code Issues Projects Releases Wiki Activity

AMDGPU: Fix crash with invariant markers

Browse Source 
The promote alloca pass didn't handle these intrinsics and crashed.
These intrinsics should accept any address space, but for now just
erase them to avoid breaking.

llvm-svn: 258537
This commit is contained in:
Matt Arsenault 2016-01-22 19:47:54 +00:00
parent 90a1a65026
commit 2b88adb9bd
2 changed files with 33 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

8
lib/Target/AMDGPU/AMDGPUPromoteAlloca.cpp
View File

@ -412,6 +412,14 @@ void AMDGPUPromoteAlloca::visitAlloca(AllocaInst &I) {
Intr->eraseFromParent();
continue;
}
case Intrinsic::invariant_start:
case Intrinsic::invariant_end:
case Intrinsic::invariant_group_barrier:
Intr->eraseFromParent();
// FIXME: I think the invariant marker should still theoretically apply,
// but the intrinsics need to be changed to accept pointers with any
// address space.
continue;
default:
Intr->dump();
llvm_unreachable("Don't know how to promote alloca intrinsic use.");

25
test/CodeGen/AMDGPU/promote-alloca-invariant-markers.ll Normal file
View File

@ -0,0 +1,25 @@
; RUN: llc -march=amdgcn -mattr=+promote-alloca -verify-machineinstrs < %s | FileCheck -check-prefix=GCN %s
declare {}* @llvm.invariant.start(i64, i8* nocapture) #0
declare void @llvm.invariant.end({}*, i64, i8* nocapture) #0
declare i8* @llvm.invariant.group.barrier(i8*) #1
; GCN-LABEL: {{^}}use_invariant_promotable_lds:
; GCN: buffer_load_dword
; GCN: ds_write_b32
define void @use_invariant_promotable_lds(i32 addrspace(1)* %arg) #2 {
bb:
%tmp = alloca i32, align 4
%tmp1 = bitcast i32* %tmp to i8*
%tmp2 = getelementptr inbounds i32, i32 addrspace(1)* %arg, i64 1
%tmp3 = load i32, i32 addrspace(1)* %tmp2
store i32 %tmp3, i32* %tmp
%tmp4 = call {}* @llvm.invariant.start(i64 4, i8* %tmp1) #0
call void @llvm.invariant.end({}* %tmp4, i64 4, i8* %tmp1) #0
%tmp5 = call i8* @llvm.invariant.group.barrier(i8* %tmp1) #1
ret void
}
attributes #0 = { argmemonly nounwind }
attributes #1 = { nounwind readnone }
attributes #2 = { nounwind }