diff options
| author | Simon Pilgrim <llvm-dev@redking.me.uk> | 2021-03-24 10:57:50 +0000 |
|---|---|---|
| committer | Simon Pilgrim <llvm-dev@redking.me.uk> | 2021-03-24 11:14:38 +0000 |
| commit | 69b71d245f27ff2538043710bb4207d2d1748099 (patch) | |
| tree | 16d93bf7c79a4121bcbed4ecd793dd67b89c549c /clang/lib/Frontend/CompilerInvocation.cpp | |
| parent | dccf83acf95530da0425ab6293addb410fb062fd (diff) | |
| download | llvm-69b71d245f27ff2538043710bb4207d2d1748099.zip llvm-69b71d245f27ff2538043710bb4207d2d1748099.tar.gz llvm-69b71d245f27ff2538043710bb4207d2d1748099.tar.bz2 | |
[X86][AVX] Cleanup gather_v8i32_v8i32 special test case
Cleanup the gather_v8i32_v8i32 IR to more closely match how the middle-end will optimise the vector geps (exposing more splats).
This helps the gather scalarization case a lot, but shows a missed opportunity for AVX512 gathers to recognise uniform-constant indices.
And none of the cases realise that some of the gathers are really blended broadcasts....
Diffstat (limited to 'clang/lib/Frontend/CompilerInvocation.cpp')
0 files changed, 0 insertions, 0 deletions