diff options
| author | Simon Pilgrim <llvm-dev@redking.me.uk> | 2020-06-11 10:22:35 +0100 |
|---|---|---|
| committer | Simon Pilgrim <llvm-dev@redking.me.uk> | 2020-06-11 10:22:55 +0100 |
| commit | 5cca9828ff1ccdb0a7bf99633afc836032dcf393 (patch) | |
| tree | 7a74a941ff991681eb287628fbe22a509a755a3f /llvm/tools/llvm-objdump/llvm-objdump.cpp | |
| parent | b651f67db8ccc9fcbb128e66987f99bb63fd3cd8 (diff) | |
| download | llvm-5cca9828ff1ccdb0a7bf99633afc836032dcf393.zip llvm-5cca9828ff1ccdb0a7bf99633afc836032dcf393.tar.gz llvm-5cca9828ff1ccdb0a7bf99633afc836032dcf393.tar.bz2 | |
[X86][AVX512] Avoid bitcasts between scalar and vXi1 bool vectors
AVX512 mask types are often bitcasted to scalar integers for various ops before being bitcast back to be used as a predicate. In many cases we can avoid these KMASK<->GPR transfers and perform equivalent operations on the mask unit.
If the destination mask type is legal, and we can confirm that the scalar op originally came from a mask/vector/float/double type then we should try to avoid the scalar entirely.
This avoids some codegen issues noticed while working on PTEST/MOVMSK improvements.
Partially fixes PR32547 - we don't create a KUNPCK yet, but OR(X,KSHIFTL(Y)) can be handled in a separate patch.
Differential Revision: https://reviews.llvm.org/D81548
Diffstat (limited to 'llvm/tools/llvm-objdump/llvm-objdump.cpp')
0 files changed, 0 insertions, 0 deletions