Discussion related to cado-nfs

[Emmanuel Thomé <Emmanuel.Thome@inria.fr>]

This issue is now fixed.

https://gitlab.inria.fr/cado-nfs/cado-nfs/-/issues/30023

As it turned out, it was indeed related to #30012

E.

On Thu, Nov 11, 2021 at 06:18:33PM -0800, Emmanuel Thomé wrote:
> Hi,
> 
> Thanks for your report.
> 
> Issue tracked here: https://gitlab.inria.fr/cado-nfs/cado-nfs/-/issues/30023
> 
> I need to investigate, because it seems to be quite an intricate
> situation. I think that we're still seeing the fallout of bug #30012 and
> the different merge requests that followed.
> 
> Note that bkthresh1 should very much be production quality, given its
> major importance in our latest computational records. But yes, some
> recent modifications of the code apparently don't work well with it, and
> this must be fixed.
> 
> Best regards,
> 
> E.
> 
> On Tue, Nov 09, 2021 at 08:23:23PM +0000, Robert Balfour wrote:
> > Dear all,
> > 
> > The rarely used -bkthresh1 option in las leads to a crash in the latest
> > CADO revision, as long as the threshold used is high enough. The bug
> > appears to be relatively new, as I could not reproduce it with revision
> > 271ad1f8 from two months ago; possibly it was brought in with
> > reduce-plattice?
> > 
> > For example, with the attached polynomial file, running the command
> > 
> > las -poly ../../cadojobs/3_748+_clients/download/3_748plus.poly -q0
> > 100000000 -A 30 -nq 20 -lim0 500000000 -lim1 300000000 -lpb0 33 -lpb1 34
> > -mfb0 66 -mfb1 98 -ncurves0 25 -ncurves1 20 -bkthresh1 100000000 -fb1
> > ../../cadojobs/3_748+_clients/download/3_748plus.roots1.gz -out
> > ../../cadojobs/sievertest/3_748plus.100M.bkthresh1 -t 4 -stats-stderr
> > 
> > quickly leads to the following:
> > 
> > # Error, p = 6440821 does not divide at (N,x) = (256,16509)
> > *** Error: caught signal "Aborted"
> > ======= Backtrace: =========
> > las(+0x291074) [0x55f651618074]
> > /lib/x86_64-linux-gnu/libc.so.6(+0x46210) [0x7fc693131210]
> > /lib/x86_64-linux-gnu/libc.so.6(gsignal+0xcb) [0x7fc69313118b]
> > /lib/x86_64-linux-gnu/libc.so.6(abort+0x12b) [0x7fc693110859]
> > las(_Z19divide_known_primesRSt6vectorImSaImEER7cxx_mpzjjbP15bucket_primes_tP21bucket_array_completeRK13trialdiv_datalmRKN13fb_factorbase7slicingE+0xe7d)
> > [0x55f65152c3fd]
> > las(_ZN25process_bucket_region_run16cofactoring_syncERSt6vectorItSaItEE+0x1485)
> > [0x55f6514c8105]
> > las(_ZN25process_bucket_region_runclEv+0x4dd) [0x55f6514ca53d]
> > las(_ZN27process_bucket_region_spawnclEP13worker_threadi+0x146)
> > [0x55f6514ca946]
> > las(_ZN11thread_pool16call_shared_taskI27process_bucket_region_spawnEEP11task_resultP13worker_threadP15task_parametersi+0x1d)
> > [0x55f6514cf0fd]
> > las(_ZN11thread_pool20thread_work_on_tasksER13worker_thread+0x91)
> > [0x55f651607a61]
> > las(_ZN11thread_pool27thread_work_on_tasks_staticEPv+0x13) 
> > [0x55f651607b33]
> > /lib/x86_64-linux-gnu/libpthread.so.0(+0x9609) [0x7fc693938609]
> > /lib/x86_64-linux-gnu/libc.so.6(clone+0x43) [0x7fc69320d293]
> > Aborted (core dumped)
> > 
> > I have also attached the output file.
> > 
> > Best regards,
> > Robert
> 
> 
>