[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
Fri Feb 18 19:49:29 GMT 2022 

What compiler version are you using?  This implies that the pgl_malloc
inline function is not being compiled to the expected code; there is an
existing non-inlined version that is used for very old gcc versions, but it
looks like it may also be needed here.

On Fri, Feb 18, 2022 at 11:40 AM Andreas Tille <andreas at an3as.eu> wrote:

> Hi again,
>
> I applied this patch and now I get:
>
> (gdb) run
> Starting program: /usr/lib/plink2/plink2-sse2 --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 4060797)]
> [New Thread 0x7fffec4c6640 (LWP 4060798)]
> [New Thread 0x7fffebcc5640 (LWP 4060799)]
> PLINK v2.00a3 64-bit (29 Jan 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: Fri Feb 18 19:06:45 2022
> 31998 MiB RAM detected; reserving 15999 MiB for main workspace.
> Using up to 4 compute threads.
> [New Thread 0x7ffff7fc5640 (LWP 4060800)]
> sizeof(PhenoCol): 40  pheno_cols: 0
> --debug: setting pheno_cols[0].nonmiss. = nullptr
>
> Thread 1 "plink2-sse2" received signal SIGSEGV, Segmentation fault.
> 0x00005555556fb82e in plink2::LoadPsam (psamname=psamname at entry=0x7fffffffbe70
> "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=0x7fffffff8880, 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:615
> warning: Source file is more recent than executable.
> 615             pheno_cols[pheno_idx].nonmiss = nullptr;
>
> Kind regards
>
>       Andreas.
>
> Am Fri, Feb 18, 2022 at 08:45:12AM -0800 schrieb Chris Chang:
> > Ok, I don't know why that particular line would fail, but I've added
> > another debug-print before it on GitHub.
> >
> > On Fri, Feb 18, 2022 at 4:24 AM Andreas Tille <andreas at fam-tille.de>
> wrote:
> >
> > > Hi Chris,
> > >
> > > Am Thu, Feb 17, 2022 at 07:13:49PM -0800 schrieb Chris Chang:
> > > > I was unable to replicate this issue on a Debian EC2 instance.
> However,
> > > > there are very few things that happen between printing "End time:"
> and
> > > > program exit, so I have added a bunch of debug-prints (active when
> the
> > > > --debug flag is passed in) to the latest GitHub commit that should
> reveal
> > > > which of those few things is triggering the segfault; let me know if
> you
> > > > are able to run this build.
> > >
> > > I think the issue is a bit more complex.  Debian provides a wrapper
> > > which calls the best / most performant plink2.  The issue seems to
> > > occure for SFX=avx.  First I do:
> > >
> > >
> > >    /usr/lib/plink2/plink2-avx --debug --dummy 33 65537 0.1
> dosage-freq=0.1
> > > --out tmp_data
> > >
> > > This works.  In the next step I fire up gdb then which results in
> > >
> > >
> > > (gdb) run
> > > Starting program: /usr/lib/plink2/plink2-avx --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 2931408)]
> > > [New Thread 0x7ffff44c6640 (LWP 2931409)]
> > > [New Thread 0x7fffebcc5640 (LWP 2931411)]
> > > PLINK v2.00a3 SSE4.2 (29 Jan 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: Fri Feb 18 11:58:49 2022
> > > 31998 MiB RAM detected; reserving 15999 MiB for main workspace.
> > > Using up to 4 compute threads.
> > > [New Thread 0x7ffff7fc5640 (LWP 2931412)]
> > >
> > > Thread 1 "plink2-avx" received signal SIGSEGV, Segmentation fault.
> > > plink2::LoadPsam (psamname=psamname at entry=0x7fffffffbe70
> "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=0x7fffffff8880, sample_include_ptr=0x7fffffff87a0,
> > > founder_info_ptr=0x7fffffff87b8, sex_nm_ptr=0x7fffffff87a8,
> > >     sex_male_ptr=0x7fffffff87b0, pheno_cols_ptr=0x7fffffff8780,
> > > pheno_names_ptr=0x7fffffff8790, raw_sample_ct_ptr=0x7fffffff8738,
> > > pheno_ct_ptr=0x7fffffff8730,
> > >     max_pheno_name_blen_ptr=0x7fffffff87c0) at ../plink2_psam.cc:611
> > > warning: Source file is more recent than executable.
> > > 611             pheno_cols[pheno_idx].nonmiss = nullptr;
> > >
> > >
> > > I also added some more debug lines in a patch[1].
> > >
> > > It seems that there is actually the weak part of the code since the
> > > output turns to
> > >
> > > ...
> > > Start time: Fri Feb 18 13:19:13 2022
> > > 31998 MiB RAM detected; reserving 15999 MiB for main workspace.
> > > Using up to 4 compute threads.
> > > [New Thread 0x7ffff7fc5640 (LWP 3957711)]
> > > --debug: setting pheno_cols[0].nonmiss. = nullptr
> > >
> > > Thread 1 "plink2-sse2" received signal SIGSEGV, Segmentation fault.
> > > 0x00005555556fb6ff in plink2::LoadPsam (psamname=psamname at entry
> =0x7fffffffbe70
> > > "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=0x7fffffff8880, sample_include_ptr=0x7fffffff87a0,
> > > founder_info_ptr=0x7fffffff87b8, sex_nm_ptr=0x7fffffff87a8,
> > >     sex_male_ptr=0x7fffffff87b0, pheno_cols_ptr=0x7fffffff8780,
> > > pheno_names_ptr=0x7fffffff8790, raw_sample_ct_ptr=0x7fffffff8738,
> > > pheno_ct_ptr=0x7fffffff8730,
> > >     max_pheno_name_blen_ptr=0x7fffffff87c0) at ../plink2_psam.cc:614
> > > warning: Source file is more recent than executable.
> > > 614             pheno_cols[pheno_idx].nonmiss = nullptr;
> > >
> > >
> > > I hope this might help a bit to track down the issue
> > >
> > >     Andreas.
> > >
> > >
> > >
> > > [1]
> > >
> https://salsa.debian.org/med-team/plink2/-/blob/master/debian/patches/debug2.patch
> > >
> > > --
> > > 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/20220218/8740077e/attachment-0001.htm>

More information about the Debian-med-packaging mailing list