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 <andr...@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 <chrchang...@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 <andr...@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@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 >
