projects / platform / upstream / llvm.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: 53c8b1b) | patch
[X86] Lower to SHLD/SHRD on slow machines for optsize
authorSimon Pilgrim <llvm-dev@redking.me.uk>
Sat, 15 Dec 2018 19:43:44 +0000 (19:43 +0000)
committerSimon Pilgrim <llvm-dev@redking.me.uk>
Sat, 15 Dec 2018 19:43:44 +0000 (19:43 +0000)
commitef7b5949e5c35fdeefb69b8f9a8eef9cfa012ed0
treedfb999193efcf682c3307347f2d3b831a9f49587tree | snapshot
parent53c8b1b6f742eeac78995f1257146f504b64314acommit | diff
[X86] Lower to SHLD/SHRD on slow machines for optsize

Use consistent rules for when to lower to SHLD/SHRD for slow machines - fixes a weird issue where funnel shift gets expanded but then X86ISelLowering's combineOr sees the optsize and combines to SHLD/SHRD, but now with the modulo amount guard......

llvm-svn: 349285
llvm/lib/Target/X86/X86ISelLowering.cpp diff | blob | history
llvm/test/CodeGen/X86/fshl.ll diff | blob | history
llvm/test/CodeGen/X86/fshr.ll diff | blob | history
Domain: System / Toolchain;