llvm-project

History

[AArch64] Fix costs of float vector compare/selects pairs.

The current cost-model overestimates the cost of vector compares &
selects for ordered floating point compares. This patch fixes that by
extending the existing logic for integer predicates.

Reviewed By: dmgreen

Differential Revision: https://reviews.llvm.org/D118256

2022-01-31 10:18:29 +00:00

AArch64

[AArch64] Fix costs of float vector compare/selects pairs.

2022-01-31 10:18:29 +00:00

AMDGPU

[AMDGPU] Make v8i16/v8f16 legal

2022-01-24 11:51:08 -08:00

ARM

[CostModel] Use cost of target trunc type when only it is the only use of a non-register sized load

2022-01-12 18:03:50 -06:00

PowerPC

[CostModel] Use cost of target trunc type when only it is the only use of a non-register sized load