maxjhandsome
/
llvm-project
forked from OSchip/llvm-project
1
0
Fork 0
Code Issues Pull Requests Packages Releases Wiki Activity

AMDGPU: Add support for amdgpu-unsafe-fp-atomics attribute 

If amdgpu-unsafe-fp-atomics is specified, allow {flat|global}_atomic_add_f32 even if atomic modes don't match.

Differential Revision: https://reviews.llvm.org/D95391
This commit is contained in:
Konstantin Zhuravlyov 2021-02-04 08:08:44 -05:00
parent 83e2710eb0
commit 6054a456da
3 changed files with 5 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
llvm/lib/Target/AMDGPU/SIISelLowering.cpp
View File

@ -11773,7 +11773,9 @@ SITargetLowering::shouldExpandAtomicRMWInIR(AtomicRMWInst *RMW) const {
unsigned AS = RMW->getPointerAddressSpace();
if (AS == AMDGPUAS::GLOBAL_ADDRESS && Subtarget->hasAtomicFaddInsts()) {
if (!fpModeMatchesGlobalFPAtomicMode(RMW))
if (!fpModeMatchesGlobalFPAtomicMode(RMW) ||
RMW->getFunction()->getFnAttribute("amdgpu-unsafe-fp-atomics")
.getValueAsString() != "true")
return AtomicExpansionKind::CmpXChg;
return RMW->use_empty() ? AtomicExpansionKind::None :

2
llvm/test/CodeGen/AMDGPU/global-atomics-fp.ll
View File

@ -61,5 +61,5 @@ define amdgpu_kernel void @global_atomic_fadd_noret_f32_wrong_subtarget(float ad
ret void
}
attributes #0 = { "denormal-fp-math-f32"="preserve-sign,preserve-sign"}
attributes #0 = { "denormal-fp-math-f32"="preserve-sign,preserve-sign" "amdgpu-unsafe-fp-atomics"="true" }
attributes #1 = { "denormal-fp-math-f32"="preserve-sign,preserve-sign" "target-cpu"="gfx803" "target-features"="+atomic-fadd-insts" }

2
llvm/test/Transforms/AtomicExpand/AMDGPU/expand-atomic-rmw-fadd.ll
View File

@ -463,4 +463,4 @@ define double @test_atomicrmw_fadd_f64_local(double addrspace(3)* %ptr, double %
ret double %res
}
attributes #0 = { "denormal-fp-math-f32"="preserve-sign,preserve-sign" }
attributes #0 = { "denormal-fp-math-f32"="preserve-sign,preserve-sign" "amdgpu-unsafe-fp-atomics"="true" }