diff options
| author | Simon Pilgrim <llvm-dev@redking.me.uk> | 2019-04-30 10:18:25 +0000 |
|---|---|---|
| committer | Simon Pilgrim <llvm-dev@redking.me.uk> | 2019-04-30 10:18:25 +0000 |
| commit | 22641cc19417a938f1511d0fe3686d2508ebd009 (patch) | |
| tree | 4eb9535a18b40e2d1048712259adde1e1151190a /llvm/test/Bitcode/bitwiseInstructions.3.2.ll | |
| parent | 59b6889238a61b864bb0a28cf02608be1ae3c324 (diff) | |
| download | bcm5719-llvm-22641cc19417a938f1511d0fe3686d2508ebd009.tar.gz bcm5719-llvm-22641cc19417a938f1511d0fe3686d2508ebd009.zip | |
Fix for bug 41512: lower INSERT_VECTOR_ELT(ZeroVec, 0, Elt) to SCALAR_TO_VECTOR(Elt) for all SSE flavors
Current LLVM uses pxor+pinsrb on SSE4+ for INSERT_VECTOR_ELT(ZeroVec, 0, Elt) insead of much simpler movd.
INSERT_VECTOR_ELT(ZeroVec, 0, Elt) is idiomatic construct which is used e.g. for _mm_cvtsi32_si128(Elt) and for lowest element initialization in _mm_set_epi32.
So such inefficient lowering leads to significant performance digradations in ceratin cases switching from SSSE3 to SSE4.
https://bugs.llvm.org/show_bug.cgi?id=41512
Here INSERT_VECTOR_ELT(ZeroVec, 0, Elt) is simply converted to SCALAR_TO_VECTOR(Elt) when applicable since latter is closer match to desired behavior and always efficiently lowered to movd and alike.
Committed on behalf of @Serge_Preis (Serge Preis)
Differential Revision: https://reviews.llvm.org/D60852
llvm-svn: 359545
Diffstat (limited to 'llvm/test/Bitcode/bitwiseInstructions.3.2.ll')
0 files changed, 0 insertions, 0 deletions
