projects / oota-llvm.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: 7d0805d) | patch
A lo/hi mul has higher latency than an imul r,ri, e.g. 5 cycles compared to 3
authorCameron Zwarich <zwarich@apple.com>
Mon, 21 Feb 2011 01:29:32 +0000 (01:29 +0000)
committerCameron Zwarich <zwarich@apple.com>
Mon, 21 Feb 2011 01:29:32 +0000 (01:29 +0000)
commitad48a4fc321adc58cd2843da64b2e206441c7e96
tree45fbf89eae6f0cf390c3ba682d68ed78ec4cf39etree | snapshot
parent7d0805dcb82e9ba1d90ce8d702169683b9caded7commit | diff
A lo/hi mul has higher latency than an imul r,ri, e.g. 5 cycles compared to 3
on Core 2 and Nehalem, so the code we generate is better than GCC's here.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@126100 91177308-0d34-0410-b5e6-96231b3b80d8
lib/Target/X86/README.txt diff | blob | history
C/C++ LLVM-based compilers that forbids OOTA behaviors