Bug#942193: dwz: dh_dwz freecad build regression: write_die: Assertion `value && refdcu->cu_kind != CU_ALT' failed.

2019-10-15 Thread Tom de Vries
On 15-10-2019 14:33, Matthias Klose wrote:
> On 14.10.19 10:10, Tom de Vries wrote:
>> On 13-10-2019 21:34, Kurt Kremitzki wrote:
>>> Hi Matthias,
>>>
>>> On Saturday, October 12, 2019 12:34:30 PM CDT Matthias Klose wrote:
>>>> Control: tags -1 + moreinfo
>>>> Control: severity -1 grave
>>>>
>>>> please could you attach the binary, or put it somewhere on the web?
>>>>
>>>
>>> Sorry, which binary are you referring to?
>>>
>>
>> Hi,
>>
>> in order to reproduce a dwz problem, we need:
>> - the dwz command line that triggered the assertion, and
>> - the files that were used as arguments. [ And since dwz modifies files
>>    in place, we need the files as they were before dwz was run. ]
>>
>>> Here are a few links in the meantime that might help, buildd logs for
>>> the
>>> i386/s390x/mipsel failures:
>>>
>>> https://buildd.debian.org/status/fetch.php?
>>> pkg=freecad=i386=0.18.3%2Bdfsg1-3=1568751036=0
>>>
>>> https://buildd.debian.org/status/fetch.php?
>>> pkg=freecad=mipsel=0.18.3%2Bdfsg1-3=1568784049=0
>>>
>>> https://buildd.debian.org/status/fetch.php?
>>> pkg=freecad=s390x=0.18.3%2Bdfsg1-3=1569764936=0
>>>
>>
>> I think it would be the easiest for me to reproduce on i386, so if you
>> could provide the reproduction information for that one, that would be
>> great.
> 
> https://people.debian.org/~doko/tmp/dwz-freecad.tar.xz
> https://people.debian.org/~doko/tmp/dwz-freecad.sh

Reproduced it, thanks.

Filed as: https://sourceware.org/bugzilla/show_bug.cgi?id=25109

- Tom



Bug#942193: dwz: dh_dwz freecad build regression: write_die: Assertion `value && refdcu->cu_kind != CU_ALT' failed.

2019-10-14 Thread Tom de Vries
On 13-10-2019 21:34, Kurt Kremitzki wrote:
> Hi Matthias,
> 
> On Saturday, October 12, 2019 12:34:30 PM CDT Matthias Klose wrote:
>> Control: tags -1 + moreinfo
>> Control: severity -1 grave
>>
>> please could you attach the binary, or put it somewhere on the web?
>>
> 
> Sorry, which binary are you referring to?
> 

Hi,

in order to reproduce a dwz problem, we need:
- the dwz command line that triggered the assertion, and
- the files that were used as arguments. [ And since dwz modifies files
  in place, we need the files as they were before dwz was run. ]

> Here are a few links in the meantime that might help, buildd logs for the 
> i386/s390x/mipsel failures:
> 
> https://buildd.debian.org/status/fetch.php?
> pkg=freecad=i386=0.18.3%2Bdfsg1-3=1568751036=0
> 
> https://buildd.debian.org/status/fetch.php?
> pkg=freecad=mipsel=0.18.3%2Bdfsg1-3=1568784049=0
> 
> https://buildd.debian.org/status/fetch.php?
> pkg=freecad=s390x=0.18.3%2Bdfsg1-3=1569764936=0
> 

I think it would be the easiest for me to reproduce on i386, so if you
could provide the reproduction information for that one, that would be
great.

Thanks,
- Tom