llvm-project

History

[NFC][Costmodel][X86] Replication shuffle: AVX512F can promote i1 to i32.

As the added codegen test coverage shows,
there isn't that much difference between AVX512DQI and
baseline AVX512F codegen, DQI added `vpmovm2d`/`vpmovd2m`,
but with just the Foundation we can use `vpternlogd`/`vptestmd`
to do the same.

2022-12-13 21:21:07 +03:00

AArch64

[AARCH64][CostModel] Modified the cost of mask vector load/store

2022-12-09 14:11:21 +08:00

AMDGPU

[AMDGPU] Add support for new LLVM vector types

2022-11-29 17:02:04 +01:00

ARM

[ARM] Fix for MVE i128 vector icmp costs.

2022-10-14 18:49:25 +01:00

PowerPC

[CostModel] Update RUN -passes=* to double quotes to appease update scripts on windows

2022-08-10 17:54:06 +01:00

RISCV

[VP][RISCV] Recommit "Add vp.fshl/fshr and RISC-V support."

2022-12-07 15:58:12 +08:00

SystemZ

[CostModel] Replace getUserCost with getInstructionCost

2022-08-18 11:55:23 +01:00

X86

[NFC][Costmodel][X86] Replication shuffle: AVX512F can promote i1 to i32.

2022-12-13 21:21:07 +03:00

free-intrinsics-datalayout.ll

Overload all llvm.annotation intrinsics for globals argument

2022-12-07 18:29:18 +00:00

free-intrinsics-no_info.ll

Overload all llvm.annotation intrinsics for globals argument

2022-12-07 18:29:18 +00:00

no_info.ll

[CostModel] Update RUN -passes=* to double quotes to appease update scripts on windows

2022-08-10 17:54:06 +01:00