[PATCH] D73179: [IR] masked gather/scatter alignment should be set
Guillaume Chatelet via Phabricator via llvm-commits
llvm-commits at lists.llvm.org
Thu Jan 23 01:42:44 PST 2020
gchatelet marked 2 inline comments as done.
gchatelet added inline comments.
================
Comment at: llvm/test/CodeGen/X86/avx2-masked-gather.ll:54
%ld = load <2 x i32*>, <2 x i32*>* %ptr
- %res = call <2 x i32> @llvm.masked.gather.v2i32(<2 x i32*> %ld, i32 0, <2 x i1> %masks, <2 x i32> %passthro)
+ %res = call <2 x i32> @llvm.masked.gather.v2i32(<2 x i32*> %ld, i32 1, <2 x i1> %masks, <2 x i32> %passthro)
ret <2 x i32> %res
----------------
delena wrote:
> I think 0 is also good. You can put alignment 0 in load/store.
It seems like it's not the case for **masked** load / store
[unittest](https://github.com/llvm/llvm-project/blob/master/llvm/test/Verifier/masked-load.ll)
[godbolt](https://godbolt.org/z/hdaTpm)
Alignment needs to be a non 0 power of two.
Any reasons why we would want it to be 0?
Repository:
rG LLVM Github Monorepo
CHANGES SINCE LAST ACTION
https://reviews.llvm.org/D73179/new/
https://reviews.llvm.org/D73179
More information about the llvm-commits
mailing list