271010bf50
The loop in this test was supposed to have temporal divergence but this was broken by r367221. Fix it.
24 lines
732 B
LLVM
24 lines
732 B
LLVM
; RUN: llc -march=amdgcn -mcpu=gfx900 -amdgpu-atomic-optimizations=true < %s | FileCheck %s
|
|
; RUN: llc -march=amdgcn -mcpu=gfx900 -amdgpu-atomic-optimizations=true < %s -use-gpu-divergence-analysis | FileCheck %s
|
|
|
|
@local = addrspace(3) global i32 undef
|
|
|
|
define amdgpu_kernel void @reducible() {
|
|
; CHECK-LABEL: reducible:
|
|
; CHECK-NOT: dpp
|
|
entry:
|
|
%x = call i32 @llvm.amdgcn.workitem.id.x()
|
|
br label %loop
|
|
loop:
|
|
%i = phi i32 [ 0, %entry ], [ %i1, %loop ]
|
|
%gep = getelementptr i32, ptr addrspace(3) @local, i32 %i
|
|
%cond = icmp ult i32 %i, %x
|
|
%i1 = add i32 %i, 1
|
|
br i1 %cond, label %loop, label %exit
|
|
exit:
|
|
%old = atomicrmw add ptr addrspace(3) %gep, i32 %x acq_rel
|
|
ret void
|
|
}
|
|
|
|
declare i32 @llvm.amdgcn.workitem.id.x()
|