Nadav Rotem f80a894bf0 [AVX] When joining two XMM registers into a YMM register, make sure that the ...

lower XMM register gets in first. This will allow the SUBREG pattern to
elliminate the first vector insertion. 



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@137310 91177308-0d34-0410-b5e6-96231b3b80d8

2011-08-11 16:49:36 +00:00

..

Analysis

Reapplying r136844.

2011-08-10 23:22:57 +00:00

Archive

Rewrite the CMake build to use explicit dependencies between libraries,

2011-07-29 00:14:25 +00:00

AsmParser

Representation of 'atomic load' and 'atomic store' in IR.

2011-08-09 23:02:53 +00:00

Bitcode

Representation of 'atomic load' and 'atomic store' in IR.

2011-08-09 23:02:53 +00:00

CodeGen

[AVX] When joining two XMM registers into a YMM register, make sure that the

2011-08-11 16:49:36 +00:00

CompilerDriver

…

ExecutionEngine

Rewrite the CMake build to use explicit dependencies between libraries,

2011-07-29 00:14:25 +00:00

Linker

Linke NamedMDNodes after linking global values as comment suggests.

2011-08-04 19:44:28 +00:00

MC

Don't truncate MachO addresses.

2011-08-09 22:12:37 +00:00

Object

Rewrite the CMake build to use explicit dependencies between libraries,

2011-07-29 00:14:25 +00:00

Support

Clarify a comment.

2011-08-10 05:02:22 +00:00

Target

[AVX] If the data which is going to be saved is already in two XMM registers

2011-08-11 16:41:21 +00:00

Transforms

Distinguish between two copies of one inlined variable. Take 2.

2011-08-10 21:50:54 +00:00

VMCore

Whitespace.

2011-08-10 17:39:11 +00:00

CMakeLists.txt

…

Makefile

…