1
0
mirror of https://github.com/c64scene-ar/llvm-6502.git synced 2025-03-29 13:37:15 +00:00
Elena Demikhovsky 2785766bc8 Fixed a bug in type legalizer for masked load/store intrinsics.
The problem occurs when after vectorization we have type
<2 x i32>. This type is promoted to <2 x i64> and then requires
additional efforts for expanding loads and truncating stores.
I added EXPAND / TRUNCATE attributes to the masked load/store
SDNodes. The code now contains additional shuffles.
I've prepared changes in the cost estimation for masked memory
operations, it will be submitted separately.



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@226808 91177308-0d34-0410-b5e6-96231b3b80d8
2015-01-22 12:07:59 +00:00
..
2014-10-16 20:00:02 +00:00
2014-12-15 22:34:58 +00:00
2015-01-10 23:41:24 +00:00
2014-12-16 20:15:45 +00:00
2015-01-19 15:16:06 +00:00
2015-01-19 15:16:06 +00:00
2014-10-16 20:00:02 +00:00
2014-10-16 20:00:02 +00:00
2014-10-21 23:01:01 +00:00
2014-10-21 23:01:01 +00:00
2015-01-19 15:16:06 +00:00
2015-01-08 07:57:27 +00:00
2014-10-16 20:00:02 +00:00
2015-01-19 15:16:06 +00:00
2014-11-11 20:49:16 +00:00
2014-12-03 02:08:38 +00:00
2014-12-03 02:08:38 +00:00
2015-01-11 07:29:51 +00:00
2014-12-16 20:15:45 +00:00
2014-07-16 15:37:08 +00:00