diff options
| author | Diego Caballero <diegocaballero@google.com> | 2022-11-16 00:43:50 +0000 |
|---|---|---|
| committer | Diego Caballero <diegocaballero@google.com> | 2022-11-18 01:05:42 +0000 |
| commit | 6c59c5cd08c879c9d1cfa653711613244a7c39bf (patch) | |
| tree | 4773078e39225096403962a372f936410a84d1b0 /clang/lib/FrontendTool/ExecuteCompilerInvocation.cpp | |
| parent | 46202543c9ba4dd2aca355b3b553ce9caeed9a80 (diff) | |
| download | llvm-6c59c5cd08c879c9d1cfa653711613244a7c39bf.zip llvm-6c59c5cd08c879c9d1cfa653711613244a7c39bf.tar.gz llvm-6c59c5cd08c879c9d1cfa653711613244a7c39bf.tar.bz2 | |
[mlir][Vector] Re-define masking semantics in vector.transfer ops
Masking hasn't been widely used in vector transfer ops and the semantics
of the mask operand were a bit loose. This patch states that the mask
operand in a vector transfer op is applied to the read/write part of the
operation and, therefore, its shape should match the shape of the
elements read/written from/into the memref/tensor regardless of any
permutation/broadcasting also applied by the transfer operation.
Reviewers: nicolasvasilache
Differential Revision: https://reviews.llvm.org/D138079
Diffstat (limited to 'clang/lib/FrontendTool/ExecuteCompilerInvocation.cpp')
0 files changed, 0 insertions, 0 deletions