[llvm] r239115 - [InstCombine] Don't miscompile safe increment idiom

Daniel Jasper djasper at google.com
Fri Jun 5 02:35:28 PDT 2015


Reverted in r239139.

On Fri, Jun 5, 2015 at 11:33 AM, Daniel Jasper <djasper at google.com> wrote:

> I have verified that this is the root cause. From a brief look, I'd guess
> that the return needs to go into the "if (Simplified) {...}" block, but I
> don't understand this enough. I'll revert this for now so that David can
> take a look.
>
> On Fri, Jun 5, 2015 at 10:25 AM, Renato Golin <renato.golin at linaro.org>
> wrote:
>
>> On 5 June 2015 at 00:11, David Majnemer <david.majnemer at gmail.com> wrote:
>> > Author: majnemer
>> > Date: Thu Jun  4 18:11:30 2015
>> > New Revision: 239115
>> >
>> > URL: http://llvm.org/viewvc/llvm-project?rev=239115&view=rev
>> > Log:
>> > [InstCombine] Don't miscompile safe increment idiom
>>
>> Hi David,
>>
>> I haven't bisected yet, but it's likely that this commit is the one
>> responsible for this failure:
>>
>>
>> http://lab.llvm.org:8011/builders/clang-x86_64-linux-selfhost-abi-test/builds/5543
>>
>> And all self-hosting + test-suite failures in ARM. :)
>>
>> The effect is the same, times out, gets killed.
>>
>> cheers,
>> --renato
>> _______________________________________________
>> llvm-commits mailing list
>> llvm-commits at cs.uiuc.edu
>> http://lists.cs.uiuc.edu/mailman/listinfo/llvm-commits
>>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20150605/eb7a358c/attachment.html>


More information about the llvm-commits mailing list