[PATCH] [X86] Provide execution domains for scalar floating point operations (mk2)
Simon Pilgrim
llvm-dev at redking.me.uk
Sat Apr 18 05:42:54 PDT 2015
Hi chandlerc, qcolombet, spatel, delena,
This is an updated version of Chandler's patch D7402 that got accepted but never committed, and has bit-rotted a bit since.
I've updated the execution domain declarations to match the approach of the packed templates and also added some extra scalar unary tests.
Note that the extra tests demonstrate that scalar unary ops aren't aware of the 'pass through' nature of the remaining vector lanes - this can be fixed in a future patch with explicit patterns like for scalar binary instructions.
Including Sanjay + Elena as they have both worked on the scalar operations recently.
REPOSITORY
rL LLVM
http://reviews.llvm.org/D9095
Files:
lib/Target/X86/X86InstrSSE.td
test/CodeGen/X86/sink-hoist.ll
test/CodeGen/X86/sse-minmax.ll
test/CodeGen/X86/sse-scalar-fp-arith.ll
EMAIL PREFERENCES
http://reviews.llvm.org/settings/panel/emailpreferences/
-------------- next part --------------
A non-text attachment was scrubbed...
Name: D9095.23989.patch
Type: text/x-patch
Size: 22020 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20150418/ff836b1a/attachment.bin>
More information about the llvm-commits
mailing list