Evandro Menezes d4b737007e [AArch64] Consider all vector types for FeatureSlowMisaligned128Store ...

The original code considered only v2i64 as slow for this feature. This patch
consider all 128-bit long vector types as slow candidates.

In internal tests, extending this feature to all 128-bit vector types
resulted in an overall improvement of 1% on Exynos M1.

Differential revision: https://reviews.llvm.org/D27998

llvm-svn: 291616

2017-01-10 23:42:21 +00:00

..

bswap.ll

[AArch64][CostModel] Add coverage for bswap intrinsics.

2017-01-05 16:55:32 +00:00

falkor.ll

[AArch64] Reduce vector insert/extract cost for Falkor.

2017-01-06 18:03:26 +00:00

gep.ll

[AArch64] Remove mcpu option as this test is not target specific. NFC.

2017-01-05 15:05:03 +00:00

kryo.ll

[AArch64] Reduce vector insert/extract cost for Kryo

2016-02-18 18:35:45 +00:00

lit.local.cfg

…

select.ll

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

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

store.ll

[AArch64] Consider all vector types for FeatureSlowMisaligned128Store

2017-01-10 23:42:21 +00:00