projects / platform / upstream / llvm.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: 616eeb8) | patch
[X86] Add the mem-reg form of CMP to the macro fusion shouldScheduleAdjacent.
authorCraig Topper <craig.topper@intel.com>
Sat, 11 Aug 2018 06:42:50 +0000 (06:42 +0000)
committerCraig Topper <craig.topper@intel.com>
Sat, 11 Aug 2018 06:42:50 +0000 (06:42 +0000)
commitc6cf16994097d5c8759968940589cdc1b6bbd591
tree99019b048218d6fd8fe0f5b44a904d5f9a832628tree | snapshot
parent616eeb827d35e420e9ab8cd6e01cd6974065a1e2commit | diff
[X86] Add the mem-reg form of CMP to the macro fusion shouldScheduleAdjacent.

Unlike the other arithmetic instructions the mem-reg form of compare is just a load and not a RMW operation. According to the Intel optimization manual, this form is also supported by macro fusion.

llvm-svn: 339498
llvm/lib/Target/X86/X86MacroFusion.cpp diff | blob | history
Domain: System / Toolchain;