projects / oota-llvm.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: d71cebf) | patch
Based on the testcase for pr3120, running on my MacPro with Xeon processors,
authorBob Wilson <bob.wilson@apple.com>
Wed, 25 Nov 2009 17:27:53 +0000 (17:27 +0000)
committerBob Wilson <bob.wilson@apple.com>
Wed, 25 Nov 2009 17:27:53 +0000 (17:27 +0000)
commit4a66e5bfc2a18a11b64c45b82862fb0f10109a98
tree4d4862eb9675bea27f56ca16f467008381ba3f3dtree | snapshot
parentd71cebf57591e2dc27930d3002848dddc76c3f92commit | diff
Based on the testcase for pr3120, running on my MacPro with Xeon processors,
it is definitely profitable to tail duplicate indirect branches for x86.
This is likely to be true to various degrees for all modern x86 processors.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@89865 91177308-0d34-0410-b5e6-96231b3b80d8
lib/Target/X86/X86InstrInfo.h diff | blob | history
C/C++ LLVM-based compilers that forbids OOTA behaviors