projects / platform / upstream / llvm.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: 9bf1761) | patch
[mlir] Support masks in TransferOpReduceRank and TransferReadPermutationLowering
authorMatthias Springer <springerm@google.com>
Thu, 13 May 2021 06:04:40 +0000 (15:04 +0900)
committerMatthias Springer <springerm@google.com>
Thu, 13 May 2021 06:08:08 +0000 (15:08 +0900)
commit60da33c2d4b2cd744c70259088b2ab89eb858f33
tree00b33b80c17db3b2bed7a8e936eb2fc7ed7fe420tree | snapshot
parent9bf17619750494ee971c66782039d4102d805c07commit | diff
[mlir] Support masks in TransferOpReduceRank and TransferReadPermutationLowering

These two patterns allow for more efficient codegen in VectorToSCF.

Differential Revision: https://reviews.llvm.org/D102222
mlir/lib/Dialect/Vector/VectorTransforms.cpp diff | blob | history
mlir/test/Dialect/Vector/vector-transfer-lowering.mlir diff | blob | history
Domain: System / Toolchain;