[Debian-med-packaging] Bug#1004037: Segmentation fault in plink2 (Was: src:plink2: fails to migrate to testing for too long: autopkgtest regression)

Chris Chang chrchang523 at gmail.com
Sun Feb 20 17:43:10 GMT 2022


Hmm.  Ordinarily, that download link would be fine.  I had not updated it
this time because the issue only affects gcc 11+, so it did not affect any
of the precompiled binaries and I wanted to keep the posted source
identical to that used to compile those binaries.

On the other hand, the only difference (in the non-gcc-11+ case) between
the binaries and the updated source is the version date.  I guess this is
insignificant enough that a temporary discordance is less disruptive than
making you change your packing process, so I've updated the website source
link to point to the newer code; let me know if you still run into any
problems.

On Sat, Feb 19, 2022 at 11:53 PM Andreas Tille <andreas at an3as.eu> wrote:

> Hi Chris,
>
> Am Sat, Feb 19, 2022 at 11:37:34PM -0800 schrieb Chris Chang:
> > To elaborate: when I look at
> > https://salsa.debian.org/med-team/plink2/-/tree/master , I can see that
> > plink2.cc, include/plink2_base.h, and include/plink2_base.cc don't have
> the
> > same contents as the v2.00a3-20220218 GitHub tag/release.  So I don't
> know
> > where you grabbed the code from this time, but it does not appear to be
> > from GitHub.
>
> That's true since on Github the source code of plink1.9 and plink2 is not
> separated.  Thus I download the tarball from
>
>    https://www.cog-genomics.org/plink/2.0/
>
> which points to
>
>    https://www.cog-genomics.org/static/bin/plink2_src_220217.zip
>
> I confirm that the tag on Github says v2.00a3-20220218 which seems to be
> later than 220217.  In principle I could point the packaging metadata to
> Github and exclude everything that I do not need for plink2 packaging (so
> remove 1.9 directory and code copies of libdeflate, simde and zstd in 2.0)
>
> In case you consider this the safer way to download plink2 code I'll do so.
>
> Kind regards and thanks a lot for comparing the code base
>
>      Andreas.
>
> > On Sat, Feb 19, 2022 at 7:54 AM Chris Chang <chrchang523 at gmail.com>
> wrote:
> >
> > > If you build off the current head, the version string should say "18
> Feb
> > > 2022", not "17 Feb 2022".
> > >
> > > On Sat, Feb 19, 2022 at 7:02 AM Andreas Tille <andreas at an3as.eu>
> wrote:
> > >
> > >> Hi again,
> > >>
> > >> Am Sat, Feb 19, 2022 at 12:28:21AM -0800 schrieb Chris Chang:
> > >> > Ok, new release/tag created.
> > >>
> > >> I've built the new version now.  The bad news is that it keeps on
> > >> SEGFAULTing for me:
> > >>
> > >> ...
> > >> (gdb) run
> > >> Starting program: /usr/lib/plink2/plink2-sse3 --debug --pfile tmp_data
> > >> --export vcf vcf-dosage=DS --out tmp_data2
> > >> [Thread debugging using libthread_db enabled]
> > >> Using host libthread_db library
> "/lib/x86_64-linux-gnu/libthread_db.so.1".
> > >> [New Thread 0x7ffff4cc7640 (LWP 985019)]
> > >> [New Thread 0x7ffff44c6640 (LWP 985020)]
> > >> [New Thread 0x7fffebcc5640 (LWP 985021)]
> > >> PLINK v2.00a3 64-bit (17 Feb 2022)
> > >> www.cog-genomics.org/plink/2.0/
> > >> (C) 2005-2022 Shaun Purcell, Christopher Chang   GNU General Public
> > >> License v3
> > >> Logging to tmp_data2.log.
> > >> Options in effect:
> > >>   --debug
> > >>   --export vcf vcf-dosage=DS
> > >>   --out tmp_data2
> > >>   --pfile tmp_data
> > >>
> > >> Start time: Sat Feb 19 16:00:00 2022
> > >> 31998 MiB RAM detected; reserving 15999 MiB for main workspace.
> > >> Using up to 4 compute threads.
> > >> [New Thread 0x7ffff7fc5640 (LWP 985022)]
> > >>
> > >> Thread 1 "plink2-sse3" received signal SIGSEGV, Segmentation fault.
> > >> plink2::LoadPsam (psamname=psamname at entry=0x7fffffffbe60
> > >> "tmp_data.psam", pheno_range_list_ptr=<optimized out>, fam_cols=...,
> > >> pheno_ct_max=<optimized out>,
> > >>     missing_pheno=<optimized out>, affection_01=0, max_thread_ct=4,
> > >> piip=0x7fffffff8870, sample_include_ptr=0x7fffffff8790,
> > >> founder_info_ptr=0x7fffffff87a8, sex_nm_ptr=0x7fffffff8798,
> > >>     sex_male_ptr=0x7fffffff87a0, pheno_cols_ptr=0x7fffffff8770,
> > >> pheno_names_ptr=0x7fffffff8780, raw_sample_ct_ptr=0x7fffffff8728,
> > >> pheno_ct_ptr=0x7fffffff8720,
> > >>     max_pheno_name_blen_ptr=0x7fffffff87b0) at ../plink2_psam.cc:611
> > >> 611             pheno_cols[pheno_idx].nonmiss = nullptr;
> > >>
> > >>
> > >> Please let me know if I can do further checks.
> > >>
> > >> Kind regards
> > >>
> > >>       Andreas.
> > >>
> > >> --
> > >> http://fam-tille.de
> > >>
> > >
>
> --
> http://fam-tille.de
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://alioth-lists.debian.net/pipermail/debian-med-packaging/attachments/20220220/26c02b74/attachment-0001.htm>


More information about the Debian-med-packaging mailing list