[llvm] r226808 - Fixed a bug in type legalizer for masked load/store intrinsics.

Hans Wennborg hans at chromium.org
Tue Feb 17 14:00:36 PST 2015


On Thu, Jan 22, 2015 at 4:07 AM, Elena Demikhovsky
<elena.demikhovsky at intel.com> wrote:
> Author: delena
> Date: Thu Jan 22 06:07:59 2015
> New Revision: 226808
>
> URL: http://llvm.org/viewvc/llvm-project?rev=226808&view=rev
> Log:
> 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.

Merged to 3.6 in r229561.



More information about the llvm-commits mailing list