[llvm-commits] [llvm] r158537 - /llvm/trunk/lib/Transforms/Scalar/LoopStrengthReduce.cpp

Chandler Carruth chandlerc at google.com
Fri Jun 15 16:29:13 PDT 2012


On Fri, Jun 15, 2012 at 4:12 PM, Andrew Trick <atrick at apple.com> wrote:

>
> On Jun 15, 2012, at 1:30 PM, Chandler Carruth <chandlerc at google.com>
> wrote:
>
> On Fri, Jun 15, 2012 at 1:07 PM, Andrew Trick <atrick at apple.com> wrote:
>
>> Author: atrick
>> Date: Fri Jun 15 15:07:29 2012
>> New Revision: 158537
>>
>> URL: http://llvm.org/viewvc/llvm-project?rev=158537&view=rev
>> Log:
>> LSR: fix expansion of scaled reg in non-address type formulae.
>>
>> For non-address users, Base and Scaled registers are not specially
>> associated to fit an address mode, so SCEVExpander should apply normal
>> expansion rules. Otherwise we may sink computation into inner loops
>> that have already been optimized.
>>
>
> Test case possible?
>
>
> Not yet. Or at least I can't think of a way.
>
> There are things we can do to improve the situation a bit:
> llvm.org/pr13122 Need unit test driver for codegen IR passes.
>

Yea, I suspect that this would require the unittest framework, just wanted
to be sure I understood the problems correctly. Thanks.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20120615/f2252bc2/attachment.html>


More information about the llvm-commits mailing list