Roman Lebedev 049799c311

[X86][Costmodel] getReplicationShuffleCost(): promote 1 bit-wide elements to 8 bit when have AVX512BW+AVX512VBMI ...

If in addition to AVX512BW (that provides `{k}<->{i8,i16}` casts and i16 shuffles),
we have AVX512VBMI, which provides i8 shuffles, we are in an optimal situation.

Reviewed By: RKSimon

Differential Revision: https://reviews.llvm.org/D114071

2021-11-19 15:58:10 +03:00

..

AArch64

[AArch64] Add some fp16 cast cost-model tests.

2021-11-11 18:21:44 +00:00

AMDGPU

[NFC][AMDGPU] Improve cost model tests coverage.

2021-09-30 18:13:17 +03:00

ARM

[ARM] Add datalayout to costmodel tests. NFC

2021-11-16 09:49:42 +00:00

PowerPC

Recommit "[Matrix] Overload stride arg in matrix.columnwise.load/store."

2021-08-12 18:31:57 +01:00

RISCV

[RISCV] Expand unaligned fixed-length vector memory accesses

2021-06-02 09:27:44 +01:00

SystemZ

Revert "[NFC] remove explicit default value for strboolattr attribute in tests"

2021-05-24 19:43:40 +02:00

X86

[X86][Costmodel] getReplicationShuffleCost(): promote 1 bit-wide elements to 8 bit when have AVX512BW+AVX512VBMI

2021-11-19 15:58:10 +03:00

free-intrinsics-datalayout.ll

[Tests] Fix incorrect noalias metadata

2021-09-18 20:51:00 +02:00

free-intrinsics-no_info.ll

[Tests] Fix incorrect noalias metadata

2021-09-18 20:51:00 +02:00

no_info.ll

…