1
0
mirror of https://github.com/RPCS3/llvm-mirror.git synced 2024-10-24 21:42:54 +02:00
llvm-mirror/test/CodeGen
Craig Topper f41154b488 [AVX-512] Teach two address instruction pass to replace masked move instructions with blendm instructions when its beneficial.
Isel now selects masked move instructions for vselect instead of blendm. But sometimes it beneficial to register allocation to remove the tied register constraint by using blendm instructions.

This also picks up cases where the masked move was created due to a masked load intrinsic.

Differential Revision: https://reviews.llvm.org/D28454

llvm-svn: 292005
2017-01-14 07:50:52 +00:00
..
AArch64 [GlobalISel] track predecessor mapping during switch lowering. 2017-01-13 23:11:37 +00:00
AMDGPU [AMDGPU] Implement f16 fcopysign and fcopysign(f32, f64) 2017-01-13 19:49:25 +00:00
ARM Revert "CodeGen: Allow small copyable blocks to "break" the CFG." 2017-01-11 19:55:19 +00:00
AVR [AVR] Implement TargetLoweing::getRegisterByName 2017-01-07 23:39:47 +00:00
BPF
Generic Track validity of pass results 2017-01-13 06:09:54 +00:00
Hexagon
Inputs
Lanai
Mips Track validity of pass results 2017-01-13 06:09:54 +00:00
MIR
MSP430
NVPTX [NVPTX] Added support for half-precision floating point. 2017-01-13 20:56:17 +00:00
PowerPC Revert "CodeGen: Allow small copyable blocks to "break" the CFG." 2017-01-11 19:55:19 +00:00
SPARC Check for register clobbers when merging a vreg live range with a 2017-01-13 19:08:36 +00:00
SystemZ Revert "CodeGen: Allow small copyable blocks to "break" the CFG." 2017-01-11 19:55:19 +00:00
Thumb Revert "CodeGen: Allow small copyable blocks to "break" the CFG." 2017-01-11 19:55:19 +00:00
Thumb2 ARM: match GCC's behaviour for builtins 2017-01-13 16:25:33 +00:00
WebAssembly Revert "CodeGen: Allow small copyable blocks to "break" the CFG." 2017-01-11 19:55:19 +00:00
WinEH
X86 [AVX-512] Teach two address instruction pass to replace masked move instructions with blendm instructions when its beneficial. 2017-01-14 07:50:52 +00:00
XCore