Charlie Turner 7a016a152d [ARM] Don't pessimize i32 vselect. ...

The underlying issues surrounding codegen for 32-bit vselects have been resolved. The pessimistic costs for 64-bit vselects remain due to the bad
scalarization that is still happening there.

I tested this on A57 in T32, A32 and A64 modes. I saw no regressions, and some improvements.

From my benchmarks, I saw these improvements in A57 (T32)
spec.cpu2000.ref.177_mesa 5.95%
lnt.SingleSource/Benchmarks/Shootout/strcat 12.93%
lnt.MultiSource/Benchmarks/MiBench/telecomm-CRC32/telecomm-CRC32 11.89%

I also measured A57 A32, A53 T32 and A9 T32 and found no performance regressions. I see much bigger wins in third-party benchmarks with this change

Differential Revision: http://reviews.llvm.org/D14743



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@253349 91177308-0d34-0410-b5e6-96231b3b80d8

2015-11-17 17:25:15 +00:00

..

AArch64

[CostModel][AArch64] Remove amortization factor for some of the vector select instructions

2015-09-09 15:35:02 +00:00

ARM

[ARM] Don't pessimize i32 vselect.

2015-11-17 17:25:15 +00:00

PowerPC

[PowerPC] Include the permutation cost for unaligned vector loads

2015-09-03 21:23:18 +00:00

X86

[CostModel] Fixed AVX integer shift costs

2015-10-17 13:23:38 +00:00

no_info.ll

Roll forward r243250

2015-07-26 19:10:03 +00:00