In D121898 we restrict parameter alignment size in IR since DAGISel
only have 4 bits to hold the alignment value. However intrinsics
won't be lowered to call instruction, so we can remove the constrain
for intrinsics.
Details
Details
Diff Detail
Diff Detail
- Repository
- rG LLVM Github Monorepo
Event Timeline
Comment Actions
Thanks for the patch. Could it use a simpler intrinsic for the test? Maybe a llvm.fshr.v8192i32?
llvm/lib/IR/Verifier.cpp | ||
---|---|---|
3228–3230 | Why do we need this check? Is there a LIT test for it? |
llvm/lib/IR/Verifier.cpp | ||
---|---|---|
3228–3230 | This is the original code, and I just move it above VerifyTypeAlign check. See the left line 3240 ~ 3245. |
Why do we need this check? Is there a LIT test for it?