diff options
author | Simon Pilgrim <llvm-dev@redking.me.uk> | 2018-03-15 14:45:30 +0000 |
---|---|---|
committer | Simon Pilgrim <llvm-dev@redking.me.uk> | 2018-03-15 14:45:30 +0000 |
commit | fb7aa57bf14268e229b4eab57e105b691f1c48b3 (patch) | |
tree | 1164f88eeb366330d5fc8a6e5bd3c2785aaf83cf /lldb/packages/Python/lldbsuite/__init__.py | |
parent | 9a54397f81c9882d9cd25bde666f4b0c50bf0654 (diff) | |
download | bcm5719-llvm-fb7aa57bf14268e229b4eab57e105b691f1c48b3.tar.gz bcm5719-llvm-fb7aa57bf14268e229b4eab57e105b691f1c48b3.zip |
[X86][SSE] Introduce Float/Vector WriteMove, WriteLoad and Writetore scheduler classes
As discussed on D44428 and PR36726, this patch splits off WriteFMove/WriteVecMove, WriteFLoad/WriteVecLoad and WriteFStore/WriteVecStore scheduler classes to permit vectors to be handled separately from gpr/scalar types.
I've minimised the diff here by only moving various basic SSE/AVX vector instructions across - we can fix the rest when called for. This does fix the MOVDQA vs MOVAPS/MOVAPD discrepancies mentioned on D44428.
Differential Revision: https://reviews.llvm.org/D44471
llvm-svn: 327630
Diffstat (limited to 'lldb/packages/Python/lldbsuite/__init__.py')
0 files changed, 0 insertions, 0 deletions