[llvm] r257164 - Re-commit r257064, this time with a fixed assert

[Michael Zolotukhin via llvm-commits]

Hi Silviu,

I haven’t seen it failing again after the revert, but I’m ok with recommitting the change as I also hope that it might be the same issue you fixed. Please go ahead, and I’ll let you know if I see anything.

Thanks,
Michael

> On Jan 14, 2016, at 7:43 AM, Silviu Baranga <silviu.baranga at arm.com> wrote:
> 
> Hi Michael,
> 
> Do you have some more data regarding this?
> FWIW I did several more runs of 483.xalancbmk - and still couldn’t trigger that issue.
> 
> I'd like to re-commit this (with fixes for the issues seen so far). Would that be ok with you?
> 
> Thanks,
> Silviu
> 
> 
>>> -----Original Message-----
>>> From: mzolotukhin at apple.com [mailto:mzolotukhin at apple.com]
>>> Sent: 11 January 2016 17:41
>>> To: Silviu Baranga
>>> Cc: Joerg Sonnenberger; llvm-commits at lists.llvm.org; nd
>>> Subject: Re: [llvm] r257164 - Re-commit r257064, this time with a fixed assert
>>> 
>>> Hi Silviu,
>>> 
>>> I’m not 100% sure that this commit is causing it, as the fail isn’t always
>>> reproducible. However, in several local runs where I reverted it, I didn’t see
>>> a failure occurring again. I’ll let you know if the fail will happen after the
>>> revert.
>>> 
>>> Thanks,
>>> Michael
>>> 
>>>> On Jan 11, 2016, at 9:27 AM, Silviu Baranga <Silviu.Baranga at arm.com>
>>> wrote:
>>>> 
>>>> FYI, reverted in r257340.
>>>> 
>>>> Thanks,
>>>> Silviu
>>>> 
>>>>>> -----Original Message-----
>>>>>> From: Silviu Baranga
>>>>>> Sent: 11 January 2016 15:41
>>>>>> To: 'mzolotukhin at apple.com'
>>>>>> Cc: Joerg Sonnenberger; llvm-commits at lists.llvm.org; nd
>>>>>> Subject: RE: [llvm] r257164 - Re-commit r257064, this time with a
>>>>>> fixed assert
>>>>>> 
>>>>>> Hi Michael,
>>>>>> 
>>>>>> I can't see the spec failure in my testing. Does it go away if you
>>>>>> revert the commit?
>>>>>> 
>>>>>> Thanks,
>>>>>> Silviu
>>>>>> 
>>>>>>>> -----Original Message-----
>>>>>>>> From: mzolotukhin at apple.com [mailto:mzolotukhin at apple.com]
>>>>>>>> Sent: 10 January 2016 19:25
>>>>>>>> To: Silviu Baranga
>>>>>>>> Cc: Joerg Sonnenberger; llvm-commits at lists.llvm.org
>>>>>>>> Subject: Re: [llvm] r257164 - Re-commit r257064, this time with a
>>>>>>>> fixed assert
>>>>>>>> 
>>>>>>>> BTW, I saw a segfault on 483.xalancbmk after this commit, but it
>>>>>>>> wasn’t always reproducible (and thus I had hard times bisecting it).
>>>>>>>> I suspect it might be the same issue.
>>>>>>>> 
>>>>>>>> Michael
>>>>>>>> 
>>>>>>>>> On Jan 10, 2016, at 4:05 AM, Silviu Baranga via llvm-commits
>>>>>>>>> <llvm-
>>>>>>>> commits at lists.llvm.org> wrote:
>>>>>>>>> 
>>>>>>>>> Hi Joerg,
>>>>>>>>> 
>>>>>>>>> Thanks for reporting! Would it be ok if I looked at this tomorrow?
>>>>>>>>> 
>>>>>>>>> -Silviu
>>>>>>>>> 
>>>>>>>>> ________________________________________
>>>>>>>>> From: Joerg Sonnenberger <joerg at britannica.bec.de>
>>>>>>>>> Sent: 10 January 2016 03:36
>>>>>>>>> To: Silviu Baranga
>>>>>>>>> Cc: llvm-commits at lists.llvm.org
>>>>>>>>> Subject: Re: [llvm] r257164 - Re-commit r257064, this time with a
>>>>>>>>> fixed assert
>>>>>>>>> 
>>>>>>>>> On Fri, Jan 08, 2016 at 11:11:05AM -0000, Silviu Baranga via
>>>>>>>>> llvm-commits
>>>>>>>> wrote:
>>>>>>>>>> Author: sbaranga
>>>>>>>>>> Date: Fri Jan  8 05:11:04 2016
>>>>>>>>>> New Revision: 257164
>>>>>>>>>> 
>>>>>>>>>> URL: http://llvm.org/viewvc/llvm-project?rev=257164&view=rev
>>>>>>>>>> Log:
>>>>>>>>>> Re-commit r257064, this time with a fixed assert
>>>>>>>>> 
>>>>>>>>> This results in segmentation faults when compiling the attached
>>>>>>>>> input samples with -O2.
>>>>>>>>> 
>>>>>>>>> Joerg
>>>>>>>>> IMPORTANT NOTICE: The contents of this email and any
>>> attachments
>>>>>>>>> are
>>>>>>>> confidential and may also be privileged. If you are not the
>>>>>>>> intended recipient, please notify the sender immediately and do
>>>>>>>> not disclose the contents to any other person, use it for any
>>>>>>>> purpose, or store or copy the information in any medium. Thank you.
>>>>>>>>> 
>>>>>>>>> _______________________________________________
>>>>>>>>> llvm-commits mailing list
>>>>>>>>> llvm-commits at lists.llvm.org
>>>>>>>>> http://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-commits
>>>> 
>