Arnold Schwaighofer 2e750c12e9 ARM NEON: Merge a f32 bitcast of a v2i32 extractelt
A vectorized sitfp on doubles will get scalarized to a sequence of an
extract_element of <2 x i32>, a bitcast to f32 and a sitofp.
Due to the the extract_element, and the bitcast we will uneccessarily generate
moves between scalar and vector registers.

The patch fixes this by using a COPY_TO_REGCLASS and a EXTRACT_SUBREG to extract
the element from the vector instead.

radar://13191881

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@175520 91177308-0d34-0410-b5e6-96231b3b80d8
2013-02-19 15:27:05 +00:00
..
2012-09-29 21:43:49 +00:00
2012-10-16 07:16:47 +00:00
2012-10-16 07:16:47 +00:00
2012-10-19 08:23:06 +00:00
2012-04-10 03:15:42 +00:00
2012-06-11 08:07:26 +00:00
2012-02-28 02:08:50 +00:00
2012-09-29 21:43:49 +00:00
2012-12-06 21:24:47 +00:00
2012-12-06 02:29:29 +00:00
2012-09-29 21:43:49 +00:00
2012-09-29 21:43:49 +00:00
2012-09-29 21:43:49 +00:00
2012-09-29 21:43:49 +00:00
2012-06-15 21:32:12 +00:00
2012-09-29 21:43:49 +00:00
2011-12-19 23:26:44 +00:00
2012-09-29 21:43:49 +00:00
2012-09-29 21:43:49 +00:00
2012-09-29 21:43:49 +00:00
2012-09-29 21:43:49 +00:00
2012-09-29 21:43:49 +00:00
2011-11-29 23:09:16 +00:00
2012-09-29 21:43:49 +00:00
2012-09-29 21:43:49 +00:00
2011-03-15 01:49:08 +00:00
2011-04-01 03:36:33 +00:00
2011-07-26 20:49:44 +00:00
2010-11-29 22:39:38 +00:00
2012-10-12 22:59:21 +00:00