[llvm-commits] [Patch] Loads used during ARM vector widening

[Tim Northover]

Hi,

This patch improves the performance of vector-extending loads on ARM.

Previously VLDR was used followed by VMOVLs. Unfortunately this mixes VFP and 
NEON instructions which can be bad for performance on some cores (Cortex-A9 in 
particular).

This patch changes this pattern to use a single-lane NEON load instead:
   vld1 {d0[0]}, [wherever]

Could someone review and commit if OK?

Thanks.

Tim.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: extload-with-lane.diff
Type: text/x-patch
Size: 5037 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20120424/6f366a02/attachment.bin>