Nadav Rotem d6fb53adb1 On AVX/AVX2 the type v8i1 is legalized to v8i16, which is an XMM sized ...

register. In most cases we actually compare or select YMM-sized registers
and mixing the two types creates horrible code. This commit optimizes
some of the transition sequences.

PR14657.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@171148 91177308-0d34-0410-b5e6-96231b3b80d8

2012-12-27 08:15:45 +00:00

..

Analysis

We are not ready to estimate the cost of integer expansions based on the number of parts. This test is too noisy.

2012-12-23 09:11:07 +00:00

Archive

…

Assembler

…

Bindings/Ocaml

…

Bitcode

…

BugPoint

…

CodeGen

On AVX/AVX2 the type v8i1 is legalized to v8i16, which is an XMM sized

2012-12-27 08:15:45 +00:00

DebugInfo

For the dwarf5 split debug info code split out the string section

2012-12-27 02:14:01 +00:00

ExecutionEngine

…

Feature

…

FileCheck

…

Instrumentation

[msan] Raise alignment of origin stores/loads when possible.

2012-12-26 11:55:09 +00:00

Integer

…

JitListener

…

Linker

…

MC

[mips] Fix encoding of BAL instruction. Also, fix assembler test case which

2012-12-21 23:13:59 +00:00

Object

Teach MachO which sections contain code

2012-12-17 17:59:32 +00:00

Other

…

Scripts

…

TableGen

Fix typo, which prevent test from being check.

2012-12-12 21:10:56 +00:00

tools

…

Transforms

If all of the write objects are identified then we can vectorize the loop even if the read objects are unidentified.

2012-12-26 23:30:53 +00:00

Unit

…

Verifier

…

YAMLParser

…

CMakeLists.txt

…

lit.cfg

…

lit.site.cfg.in

…

Makefile

…

Makefile.tests

…

TestRunner.sh

…