projects / oota-llvm.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: 587fb1d) | patch
AVX: Move the ZEXT/ANYEXT DAGCo optimizations to the lowering of these optimizations...
authorNadav Rotem <nrotem@apple.com>
Fri, 28 Dec 2012 05:45:24 +0000 (05:45 +0000)
committerNadav Rotem <nrotem@apple.com>
Fri, 28 Dec 2012 05:45:24 +0000 (05:45 +0000)
commit0509db27386f5cafffd364618365ecda741cf0bd
tree2ef5af4e9f96faee5e55a274cbd6aede69de6229tree | snapshot
parent587fb1dd30b73afb3c83a1e88d9ea101a0b28ab2commit | diff
AVX: Move the ZEXT/ANYEXT DAGCo optimizations to the lowering of these optimizations. The old test cases still cover all of these lowering/optimizations. The single change that we have is that now anyext does not need to zero a register, because it does not use the exact code path as the zero_extend.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@171178 91177308-0d34-0410-b5e6-96231b3b80d8
lib/Target/X86/X86ISelLowering.cpp diff | blob | history
lib/Target/X86/X86ISelLowering.h diff | blob | history
test/CodeGen/X86/avx-zext.ll diff | blob | history
C/C++ LLVM-based compilers that forbids OOTA behaviors