Chandler Carruth 29720a4bad [x86] Teach the new vector shuffle lowering of v4f64 to prefer a direct ...

VBLENDPD over using VSHUFPD. While the 256-bit variant of VBLENDPD slows
down to the same speed as VSHUFPD on Sandy Bridge CPUs, it has twice the
reciprocal throughput on Ivy Bridge CPUs much like it does everywhere
for 128-bits. There isn't a downside, so just eagerly use this
instruction when it suffices.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@218208 91177308-0d34-0410-b5e6-96231b3b80d8

2014-09-21 11:17:55 +00:00

..

AArch64

[FastIsel][AArch64] Fix a think-o in address computation.

2014-09-19 22:23:46 +00:00

ARM

[ARM] Do not perform a tail call when the caller returns several values.

2014-09-18 21:17:50 +00:00

CPP

…

Generic

Fix crash with an insertvalue that produces an empty object.

2014-09-20 00:10:47 +00:00

Hexagon

…

Inputs

…

Mips

Add mips32 r1 to the list of supported targets for Mips fast-isel

2014-09-15 20:30:25 +00:00

MSP430

…

NVPTX

…

PowerPC

Optionally enable more-aggressive FMA formation in DAGCombine

2014-09-19 11:42:56 +00:00

R600

R600: Un-xfail a test which passes with pass disabled

2014-09-19 23:02:20 +00:00

SPARC

Add back tests for empty function in SPARC and PowerPC.

2014-09-15 22:11:07 +00:00

SystemZ

…

Thumb

…

Thumb2

…

X86

[x86] Teach the new vector shuffle lowering of v4f64 to prefer a direct

2014-09-21 11:17:55 +00:00

XCore

…