I confirm its gcc-11. I'll check tomorrow. Thanks a lot for your quick and helpful responses, Andreas.
Am Fri, Feb 18, 2022 at 12:53:58PM -0800 schrieb Chris Chang: > I have posted an update under the provisional assumption that it's gcc 11's > new ipa-modref pass that is causing this code to fail, since it does seem > to break some similar code. > > On Fri, Feb 18, 2022 at 11:49 AM Chris Chang <chrchang...@gmail.com> wrote: > > > 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 <andr...@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@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 <andr...@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@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@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 > >> > > -- http://fam-tille.de
