| Issue |
165642
|
| Summary |
[llvm][HLSL][SPIR-V] Incorrect transforms for nested loop with wave operations
|
| Labels |
llvm:transforms,
clang:HLSL:SPIRV
|
| Assignees |
luciechoi
|
| Reporter |
luciechoi
|
```
RWStructuredBuffer<uint> Out : register(u0);
[numthreads(8,8,1)]
void main(uint3 TID : SV_GroupThreadID) {
for (uint i = 0; i < 8; i++) {
for (uint j = 0; j < 8; j++) {
if (i == TID.x && j == TID.y) {
uint index = TID.x * 8 + TID.y;
Out[index] = WaveActiveMax(index);
break;
}
}
}
}
```
Found while working on https://github.com/llvm/llvm-project/issues/164496. Child of https://github.com/llvm/llvm-project/issues/136930.
The issues are potentially with `SimplifyCFG`, `JumpThreading`, and `GVN` passes, where instructions are folded
```
Loop intrinsic cannot be preceded by a convergent operation in the same basic block.
%4 = call token @llvm.experimental.convergence.loop() [ "convergencectrl"(token %0) ]
in function main
```
![]()
_______________________________________________
llvm-bugs mailing list
[email protected]
https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-bugs
