[PATCH] D38374: [memcpyopt] Memcpy-memcpy dependence isn't detected across basic blocks

[Dan Gohman via llvm-commits]

Friendly ping. Thanks again for helping with this patch!

Dan


On Tue, Oct 17, 2017 at 1:28 PM, Dan Gohman <sunfish at mozilla.com> wrote:

> I haven't rebased the patch; I tested it on an older tree.
>
> Thanks for your help!
>
> Dan
>
>
> On Oct 17, 2017 1:02 PM, "Daniel Berlin" <dberlin at dberlin.org> wrote:
>
> Hey, sorry, missed your email.
> Did you rebase it already?
> If so, can you just post your rebase, and i'll shepherd it through
> review/test/fix it?
>
> If not, i'll rebase it.
>
>
> On Tue, Oct 17, 2017 at 10:27 AM, Dan Gohman via Phabricator <
> reviews at reviews.llvm.org> wrote:
>
>> sunfish added a comment.
>>
>> Friendly ping :-). I confirmed that the MemorySSA-based patch does
>> implement the optimization. In fact, it tries //not// to, however it
>> doesn't try hard enough, and the optimization happens anyway. It passes all
>> the tests here as-is.
>>
>>
>> Repository:
>>   rL LLVM
>>
>> https://reviews.llvm.org/D38374
>>
>>
>>
>>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20171108/6295c545/attachment.html>