2 Commits

Author SHA1 Message Date
Gheorghe-Teodor Bercea
3df36a2b18
[AMDGPU] Enable vectorization of i8 values. (#134934)
This patch adjusts the cost model to account for the ability of the
AMDGPU optimizer to group together i8 values into i32 values.

Co-authored-by: Erich Keane <ekeane@nvidia.com>
2025-06-26 19:15:31 -04:00
Gheorghe-Teodor Bercea
25a031947a
[AMDGPU][NFC] Add tests in preparation for i8 vectorization (#138801)
Precommit tests for PR: https://github.com/llvm/llvm-project/pull/134934
2025-05-09 10:32:49 -04:00