[Pkg-julia-devel] Bug#971570: Bug#971570: Bug#971570: julia: libgit2 1.0 transition

Ximin Luo infinity0 at debian.org
Thu Dec 3 14:23:20 GMT 2020


Hi Norbert,

In October I did a build which failed with SEGV:

~~~~
OPENBLAS_NUM_THREADS=1 OMP_NUM_THREADS=1 ./sblat1
 Real BLAS Test Program Results


 Test of subprogram number  1             SDOT 

Program received signal SIGSEGV: Segmentation fault - invalid memory reference.

Backtrace for this error:
#0  0x7f2040487bd0 in ???
#1  0x7f2040486e25 in ???
#2  0x7f2040178e2f in ???
#3  0x563650b31334 in sdot_k_HASWELL
	at ../kernel/x86_64/sdot.c:141
#4  0x56364fd452cb in sdot_
	at ./deps/scratch/openblas-6a79b36d1bf73584a513139806d226f9189d621e/interface/dot.c:65
#5  0x56364fd4411e in check2_
	at ./deps/scratch/openblas-6a79b36d1bf73584a513139806d226f9189d621e/test/sblat1.f:619
#6  0x56364fd44e9a in sblat1
	at ./deps/scratch/openblas-6a79b36d1bf73584a513139806d226f9189d621e/test/sblat1.f:83
#7  0x56364fd4304e in main
	at ./deps/scratch/openblas-6a79b36d1bf73584a513139806d226f9189d621e/test/sblat1.f:94
Segmentation fault
~~~~

Now that I look at it closer, perhaps this is unrelated to libgit2 and the problem was already fixed in the meantime. If not, the full log is attached for you to investigate. In the meantime I will let the release team know not to block the libgit2 transition on julia.

X

Norbert Preining:
> severity 971570 normal
> retitle 971570 check whether binNMU transition to libgit2 1.0 is successful
> thanks
> 
> Ajusting severity according to what I wrote. I keep the bug open to
> remind us that we should check that the binNMU facility actually worked.
> 
> Nothing else should be necessary.
> 
> Norbert
> 
> On Wed, 02 Dec 2020, Norbert Preining wrote:
>> Hi Ximin,
>>
>> On Tue, 01 Dec 2020, Ximin Luo wrote:
>>> Control: severity -1 serious
>>
>> I am very doubtful what you are doing here.
>>
>> Have you tested julia with libgit2 1.0? Are there build errors?
>>
>> I don't see any, the package builds fine with libgit2-dev 1.0.1+dfsg.1-1
>> from experimental, I just did a test build.
>>
>> That means, the transition would require a simple binary rebuild,
>> without any action on our side.
>>
>> I tend to close this bug as incorrect.
>>
>> Best
>>
>> Norbert
> 
> --
> PREINING Norbert                              https://www.preining.info
> Accelia Inc. + IFMGA ProGuide + TU Wien + JAIST + TeX Live + Debian Dev
> GPG: 0x860CDC13   fp: F7D8 A928 26E3 16A1 9FA0 ACF0 6CAC A448 860C DC13
> 


-- 
GPG: ed25519/56034877E1F87C35
https://github.com/infinity0/pubkeys.git
-------------- next part --------------
A non-text attachment was scrubbed...
Name: julia_amd64-2020-10-02T00:00:26Z.build.gz
Type: application/gzip
Size: 379937 bytes
Desc: not available
URL: <http://alioth-lists.debian.net/pipermail/pkg-julia-devel/attachments/20201203/36fef99f/attachment-0001.gz>


More information about the Pkg-julia-devel mailing list