diff options
| author | Thomas Lively <tlively@google.com> | 2020-05-11 09:55:43 -0700 |
|---|---|---|
| committer | Thomas Lively <tlively@google.com> | 2020-05-11 10:01:55 -0700 |
| commit | 8e3e56f2a36701480eeb65e426701d5a9025cc59 (patch) | |
| tree | 61a56e15cd31e4230be3d8d6cf7386aed0bfe177 /clang/lib/Frontend/CompilerInvocation.cpp | |
| parent | fe69dfebcfa007e23bf528fff8b96ac63e593e34 (diff) | |
| download | llvm-8e3e56f2a36701480eeb65e426701d5a9025cc59.zip llvm-8e3e56f2a36701480eeb65e426701d5a9025cc59.tar.gz llvm-8e3e56f2a36701480eeb65e426701d5a9025cc59.tar.bz2 | |
[WebAssembly] Add wasm-specific vector shuffle builtin and intrinsic
Summary:
Although using `__builtin_shufflevector` and the `shufflevector`
instruction works fine, they are not opaque to the optimizer. As a
result, DAGCombine can potentially reduce the number of shuffles and
change the shuffle masks. This is unexpected behavior for users of the
WebAssembly SIMD intrinsics who have crafted their shuffles to
optimize the code generated by engines. This patch solves the problem
by adding a new shuffle intrinsic that is opaque to the optimizers in
line with the decision of the WebAssembly SIMD contributors at
https://github.com/WebAssembly/simd/issues/196#issuecomment-622494748. In
the future we may implement custom DAG combines to properly optimize
shuffles and replace this solution.
Reviewers: aheejin, dschuff
Subscribers: sbc100, jgravelle-google, hiraditya, sunfish, cfe-commits, llvm-commits
Tags: #clang, #llvm
Differential Revision: https://reviews.llvm.org/D66983
Diffstat (limited to 'clang/lib/Frontend/CompilerInvocation.cpp')
0 files changed, 0 insertions, 0 deletions