[PATCH] D114893: [CostModel][AMDGPU] Fix instructions costs estimation for vector types.
Daniil Fukalov via Phabricator via llvm-commits
llvm-commits at lists.llvm.org
Wed Dec 1 16:11:59 PST 2021
dfukalov added inline comments.
================
Comment at: llvm/test/Analysis/CostModel/AMDGPU/add-sub.ll:75
%v5i64 = add <5 x i64> undef, undef
- %v6i64 = add <6 x i64> undef, undef
- %v7i64 = add <7 x i64> undef, undef
----------------
rampitec wrote:
> Why dropping this?
Actually tests for long vectors are needed to check the penalty is added in `getTypeLegalizationCost()`. But it is about vectors sized >256bits so we just need to check that cost for v5i64 is not a linear dependency with v4i64 and smaller vectors.
Repository:
rG LLVM Github Monorepo
CHANGES SINCE LAST ACTION
https://reviews.llvm.org/D114893/new/
https://reviews.llvm.org/D114893
More information about the llvm-commits
mailing list