projects / platform / upstream / llvm.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: 966fac1) | patch
[ARM] Enable MVE masked loads and stores
authorDavid Green <david.green@arm.com>
Sun, 8 Dec 2019 16:10:01 +0000 (16:10 +0000)
committerDavid Green <david.green@arm.com>
Mon, 9 Dec 2019 11:37:34 +0000 (11:37 +0000)
commitb1aba0378e52be51cfb7fb6f03417ebf408d66cc
tree5ba5ef72595fbb2714705299a29c61af4321b84ftree | snapshot
parent966fac1941ea99e076a7654d229b27e1e6e4ad17commit | diff
[ARM] Enable MVE masked loads and stores

With the extra optimisations we have done, these should now be fine to
enable by default. Which is what this patch does.

Differential Revision: https://reviews.llvm.org/D70968
16 files changed:
llvm/lib/Target/ARM/ARMTargetTransformInfo.cpp diff | blob | history
llvm/test/CodeGen/Thumb2/LowOverheadLoops/cond-vector-reduce-mve-codegen.ll diff | blob | history
llvm/test/CodeGen/Thumb2/LowOverheadLoops/fast-fp-loops.ll diff | blob | history
llvm/test/CodeGen/Thumb2/LowOverheadLoops/mve-tail-data-types.ll diff | blob | history
llvm/test/CodeGen/Thumb2/LowOverheadLoops/vector-arith-codegen.ll diff | blob | history
llvm/test/CodeGen/Thumb2/mve-intrinsics/load-store.ll diff | blob | history
llvm/test/CodeGen/Thumb2/mve-masked-ldst-offset.ll diff | blob | history
llvm/test/CodeGen/Thumb2/mve-masked-ldst-postinc.ll diff | blob | history
llvm/test/CodeGen/Thumb2/mve-masked-ldst-preinc.ll diff | blob | history
llvm/test/CodeGen/Thumb2/mve-masked-ldst.ll diff | blob | history
llvm/test/CodeGen/Thumb2/mve-masked-load.ll diff | blob | history
llvm/test/CodeGen/Thumb2/mve-masked-store.ll diff | blob | history
llvm/test/Transforms/LoopVectorize/ARM/mve-maskedldst.ll diff | blob | history
llvm/test/Transforms/LoopVectorize/ARM/mve-shiftcost.ll diff | blob | history
llvm/test/Transforms/LoopVectorize/ARM/prefer-tail-loop-folding.ll diff | blob | history
llvm/test/Transforms/LoopVectorize/ARM/tail-loop-folding.ll diff | blob | history
Domain: System / Toolchain;