Skip Menu |
 

This queue is for tickets about the Coro-ProcessPool CPAN distribution.

Report information
The Basics
Id: 122565
Status: rejected
Priority: 0/
Queue: Coro-ProcessPool

People
Owner: Nobody in particular
Requestors: ANDK [...] cpan.org
Cc:
AdminCc:

Bug Information
Severity: (no value)
Broken in: 0.26
Fixed in: (no value)



Subject: Core dumps during testing with recent perls
Download (untitled) / with headers
text/plain 1.7k
Some of my smokers already sent fail reports with a SEGV indication. Here is a stack trace that I just observed with v5.27.2: Program terminated with signal SIGSEGV, Segmentation fault. #0 coro_semaphore_adjust (av=0x0, adjust=adjust@entry=1) at State.xs:3099 3099 State.xs: No such file or directory. (gdb) bt #0 coro_semaphore_adjust (av=0x0, adjust=adjust@entry=1) at State.xs:3099 #1 0x00007fb6c661b440 in XS_Coro__Semaphore_up (cv=<optimized out>) at State.xs:4359 #2 0x000055d8a485e158 in Perl_pp_entersub () at pp_hot.c:4235 #3 0x000055d8a4856233 in Perl_runops_standard () at run.c:41 #4 0x000055d8a47dbf36 in Perl_call_sv (sv=0x55d8a52c0ec0, flags=flags@entry=45) at perl.c:2872 #5 0x000055d8a48629f6 in S_curse (sv=sv@entry=0x55d8a60a6410, check_refcnt=check_refcnt@entry=true) at sv.c:6980 #6 0x000055d8a48633dd in Perl_sv_clear (orig_sv=orig_sv@entry=0x55d8a60a6410) at sv.c:6584 #7 0x000055d8a48636de in Perl_sv_free2 (sv=0x55d8a60a6410, rc=<optimized out>) at sv.c:7081 #8 0x000055d8a48618d7 in S_visit (f=f@entry=0x55d8a48638c0 <do_clean_objs>, flags=flags@entry=2048, mask=mask@entry=2048) at sv.c:476 #9 0x000055d8a4863a3e in Perl_sv_clean_objs () at sv.c:627 #10 0x000055d8a47deb86 in perl_destruct (my_perl=<optimized out>) at perl.c:883 #11 0x00007fb6c6613c9e in perlish_exit () at State.xs:1520 #12 0x00007fb6c6616e93 in cctx_run (arg=<optimized out>) at State.xs:1571 #13 0x00007fb6c6615f35 in coro_init () at libcoro/coro.c:97 #14 0x00007fb6c71e3290 in ?? () from /lib/x86_64-linux-gnu/libc.so.6 #15 0x0000000000000000 in ?? () Anoter v5.27.2, compiled with -DDEBUGGING, fails with this assertion: perl: pp_hot.c:2489: Perl_pp_multideref: Assertion `isGV_with_GP(sv)' failed. Let me know whether I can provide more info. HTH&&Thanks,
Download (untitled) / with headers
text/plain 2.4k
Thank you for the report. Marc Lehmann just released Coro 6.513, which has experimental support for newer perls, triggering a flurry of test failures in modules using Coro, especially those using AnyEvent::Impl::Perl in testing (which is useful, since most, if not all, of the other AnyEvent event loop plugins have external dependencies, such as libevent). I am working on a fix for these errors, or at least a temporary work-around to prevent testing on modern perls until Coro support for them stabilizes. On Sat Jul 22 15:36:08 2017, ANDK wrote: Show quoted text
> Some of my smokers already sent fail reports with a SEGV indication. > Here is a stack trace that I just observed with v5.27.2: > > Program terminated with signal SIGSEGV, Segmentation fault. > #0 coro_semaphore_adjust (av=0x0, adjust=adjust@entry=1) at > State.xs:3099 > 3099 State.xs: No such file or directory. > (gdb) bt > #0 coro_semaphore_adjust (av=0x0, adjust=adjust@entry=1) at > State.xs:3099 > #1 0x00007fb6c661b440 in XS_Coro__Semaphore_up (cv=<optimized out>) > at State.xs:4359 > #2 0x000055d8a485e158 in Perl_pp_entersub () at pp_hot.c:4235 > #3 0x000055d8a4856233 in Perl_runops_standard () at run.c:41 > #4 0x000055d8a47dbf36 in Perl_call_sv (sv=0x55d8a52c0ec0, > flags=flags@entry=45) > at perl.c:2872 > #5 0x000055d8a48629f6 in S_curse (sv=sv@entry=0x55d8a60a6410, > check_refcnt=check_refcnt@entry=true) at sv.c:6980 > #6 0x000055d8a48633dd in Perl_sv_clear > (orig_sv=orig_sv@entry=0x55d8a60a6410) > at sv.c:6584 > #7 0x000055d8a48636de in Perl_sv_free2 (sv=0x55d8a60a6410, > rc=<optimized out>) > at sv.c:7081 > #8 0x000055d8a48618d7 in S_visit (f=f@entry=0x55d8a48638c0 > <do_clean_objs>, > flags=flags@entry=2048, mask=mask@entry=2048) at sv.c:476 > #9 0x000055d8a4863a3e in Perl_sv_clean_objs () at sv.c:627 > #10 0x000055d8a47deb86 in perl_destruct (my_perl=<optimized out>) at > perl.c:883 > #11 0x00007fb6c6613c9e in perlish_exit () at State.xs:1520 > #12 0x00007fb6c6616e93 in cctx_run (arg=<optimized out>) at > State.xs:1571 > #13 0x00007fb6c6615f35 in coro_init () at libcoro/coro.c:97 > #14 0x00007fb6c71e3290 in ?? () from /lib/x86_64-linux-gnu/libc.so.6 > #15 0x0000000000000000 in ?? () > > Anoter v5.27.2, compiled with -DDEBUGGING, fails with this assertion: > > perl: pp_hot.c:2489: Perl_pp_multideref: Assertion `isGV_with_GP(sv)' > failed. > > Let me know whether I can provide more info. > > HTH&&Thanks,
-- -Jeff Ober
Download (untitled) / with headers
text/plain 2.7k
It seems that there's no improvement with Coro-ProcessPool-0.26_003. I see SIGBUS (FreeBSD) and SIGSEGV (Linux) crashes in the test scripts, for various perl versions. On 2017-07-25 08:19:13, JEFFOBER wrote: Show quoted text
> Thank you for the report. Marc Lehmann just released Coro 6.513, which > has experimental support for newer perls, triggering a flurry of test > failures in modules using Coro, especially those using > AnyEvent::Impl::Perl in testing (which is useful, since most, if not > all, of the other AnyEvent event loop plugins have external > dependencies, such as libevent). > > I am working on a fix for these errors, or at least a temporary work- > around to prevent testing on modern perls until Coro support for them > stabilizes. > > On Sat Jul 22 15:36:08 2017, ANDK wrote:
> > Some of my smokers already sent fail reports with a SEGV indication. > > Here is a stack trace that I just observed with v5.27.2: > > > > Program terminated with signal SIGSEGV, Segmentation fault. > > #0 coro_semaphore_adjust (av=0x0, adjust=adjust@entry=1) at > > State.xs:3099 > > 3099 State.xs: No such file or directory. > > (gdb) bt > > #0 coro_semaphore_adjust (av=0x0, adjust=adjust@entry=1) at > > State.xs:3099 > > #1 0x00007fb6c661b440 in XS_Coro__Semaphore_up (cv=<optimized out>) > > at State.xs:4359 > > #2 0x000055d8a485e158 in Perl_pp_entersub () at pp_hot.c:4235 > > #3 0x000055d8a4856233 in Perl_runops_standard () at run.c:41 > > #4 0x000055d8a47dbf36 in Perl_call_sv (sv=0x55d8a52c0ec0, > > flags=flags@entry=45) > > at perl.c:2872 > > #5 0x000055d8a48629f6 in S_curse (sv=sv@entry=0x55d8a60a6410, > > check_refcnt=check_refcnt@entry=true) at sv.c:6980 > > #6 0x000055d8a48633dd in Perl_sv_clear > > (orig_sv=orig_sv@entry=0x55d8a60a6410) > > at sv.c:6584 > > #7 0x000055d8a48636de in Perl_sv_free2 (sv=0x55d8a60a6410, > > rc=<optimized out>) > > at sv.c:7081 > > #8 0x000055d8a48618d7 in S_visit (f=f@entry=0x55d8a48638c0 > > <do_clean_objs>, > > flags=flags@entry=2048, mask=mask@entry=2048) at sv.c:476 > > #9 0x000055d8a4863a3e in Perl_sv_clean_objs () at sv.c:627 > > #10 0x000055d8a47deb86 in perl_destruct (my_perl=<optimized out>) at > > perl.c:883 > > #11 0x00007fb6c6613c9e in perlish_exit () at State.xs:1520 > > #12 0x00007fb6c6616e93 in cctx_run (arg=<optimized out>) at > > State.xs:1571 > > #13 0x00007fb6c6615f35 in coro_init () at libcoro/coro.c:97 > > #14 0x00007fb6c71e3290 in ?? () from /lib/x86_64-linux-gnu/libc.so.6 > > #15 0x0000000000000000 in ?? () > > > > Anoter v5.27.2, compiled with -DDEBUGGING, fails with this assertion: > > > > perl: pp_hot.c:2489: Perl_pp_multideref: Assertion `isGV_with_GP(sv)' > > failed. > > > > Let me know whether I can provide more info. > > > > HTH&&Thanks,
Download (untitled) / with headers
text/plain 3.2k
Which versions? I note that t/15_semaphore.t (which I believe is related to the stack trace in the original bug report) in the Coro distribution is failing for multiple 5.27s but is working with 5.26 and earlier. Also, are there versions on which it is passing without error or is this across the board? On Wed Nov 08 13:31:03 2017, SREZIC wrote: Show quoted text
> It seems that there's no improvement with Coro-ProcessPool-0.26_003. I > see SIGBUS (FreeBSD) and SIGSEGV (Linux) crashes in the test scripts, > for various perl versions. > > On 2017-07-25 08:19:13, JEFFOBER wrote:
> > Thank you for the report. Marc Lehmann just released Coro 6.513, > > which > > has experimental support for newer perls, triggering a flurry of test > > failures in modules using Coro, especially those using > > AnyEvent::Impl::Perl in testing (which is useful, since most, if not > > all, of the other AnyEvent event loop plugins have external > > dependencies, such as libevent). > > > > I am working on a fix for these errors, or at least a temporary work- > > around to prevent testing on modern perls until Coro support for them > > stabilizes. > > > > On Sat Jul 22 15:36:08 2017, ANDK wrote:
> > > Some of my smokers already sent fail reports with a SEGV > > > indication. > > > Here is a stack trace that I just observed with v5.27.2: > > > > > > Program terminated with signal SIGSEGV, Segmentation fault. > > > #0 coro_semaphore_adjust (av=0x0, adjust=adjust@entry=1) at > > > State.xs:3099 > > > 3099 State.xs: No such file or directory. > > > (gdb) bt > > > #0 coro_semaphore_adjust (av=0x0, adjust=adjust@entry=1) at > > > State.xs:3099 > > > #1 0x00007fb6c661b440 in XS_Coro__Semaphore_up (cv=<optimized
> > > out>)
> > > at State.xs:4359 > > > #2 0x000055d8a485e158 in Perl_pp_entersub () at pp_hot.c:4235 > > > #3 0x000055d8a4856233 in Perl_runops_standard () at run.c:41 > > > #4 0x000055d8a47dbf36 in Perl_call_sv (sv=0x55d8a52c0ec0, > > > flags=flags@entry=45) > > > at perl.c:2872 > > > #5 0x000055d8a48629f6 in S_curse (sv=sv@entry=0x55d8a60a6410, > > > check_refcnt=check_refcnt@entry=true) at sv.c:6980 > > > #6 0x000055d8a48633dd in Perl_sv_clear > > > (orig_sv=orig_sv@entry=0x55d8a60a6410) > > > at sv.c:6584 > > > #7 0x000055d8a48636de in Perl_sv_free2 (sv=0x55d8a60a6410, > > > rc=<optimized out>) > > > at sv.c:7081 > > > #8 0x000055d8a48618d7 in S_visit (f=f@entry=0x55d8a48638c0 > > > <do_clean_objs>, > > > flags=flags@entry=2048, mask=mask@entry=2048) at sv.c:476 > > > #9 0x000055d8a4863a3e in Perl_sv_clean_objs () at sv.c:627 > > > #10 0x000055d8a47deb86 in perl_destruct (my_perl=<optimized out>) > > > at > > > perl.c:883 > > > #11 0x00007fb6c6613c9e in perlish_exit () at State.xs:1520 > > > #12 0x00007fb6c6616e93 in cctx_run (arg=<optimized out>) at > > > State.xs:1571 > > > #13 0x00007fb6c6615f35 in coro_init () at libcoro/coro.c:97 > > > #14 0x00007fb6c71e3290 in ?? () from /lib/x86_64-linux- > > > gnu/libc.so.6 > > > #15 0x0000000000000000 in ?? () > > > > > > Anoter v5.27.2, compiled with -DDEBUGGING, fails with this > > > assertion: > > > > > > perl: pp_hot.c:2489: Perl_pp_multideref: Assertion > > > `isGV_with_GP(sv)' > > > failed. > > > > > > Let me know whether I can provide more info. > > > > > > HTH&&Thanks,
-- -Jeff Ober
Subject: Re: [rt.cpan.org #122565] Core dumps during testing with recent perls
Date: Wed, 08 Nov 2017 16:43:01 -0500
To: "Slaven_Rezic via RT" <bug-Coro-ProcessPool [...] rt.cpan.org>
From: Jeff Ober <sysread [...] fastmail.fm>
Download (untitled) / with headers
text/plain 3.1k
Also, is the latest Coro failing in those same environments as well? -- Jeff Ober sysread@fastmail.fm Old programmers don't die; they just parse on... On Wed, Nov 8, 2017, at 01:31 PM, Slaven_Rezic via RT wrote: Show quoted text
> Queue: Coro-ProcessPool > Ticket <URL: https://rt.cpan.org/Ticket/Display.html?id=122565 > > > It seems that there's no improvement with Coro-ProcessPool-0.26_003. I > see SIGBUS (FreeBSD) and SIGSEGV (Linux) crashes in the test scripts, for > various perl versions. > > On 2017-07-25 08:19:13, JEFFOBER wrote:
> > Thank you for the report. Marc Lehmann just released Coro 6.513, which > > has experimental support for newer perls, triggering a flurry of test > > failures in modules using Coro, especially those using > > AnyEvent::Impl::Perl in testing (which is useful, since most, if not > > all, of the other AnyEvent event loop plugins have external > > dependencies, such as libevent). > > > > I am working on a fix for these errors, or at least a temporary work- > > around to prevent testing on modern perls until Coro support for them > > stabilizes. > > > > On Sat Jul 22 15:36:08 2017, ANDK wrote:
> > > Some of my smokers already sent fail reports with a SEGV indication. > > > Here is a stack trace that I just observed with v5.27.2: > > > > > > Program terminated with signal SIGSEGV, Segmentation fault. > > > #0 coro_semaphore_adjust (av=0x0, adjust=adjust@entry=1) at > > > State.xs:3099 > > > 3099 State.xs: No such file or directory. > > > (gdb) bt > > > #0 coro_semaphore_adjust (av=0x0, adjust=adjust@entry=1) at > > > State.xs:3099 > > > #1 0x00007fb6c661b440 in XS_Coro__Semaphore_up (cv=<optimized out>) > > > at State.xs:4359 > > > #2 0x000055d8a485e158 in Perl_pp_entersub () at pp_hot.c:4235 > > > #3 0x000055d8a4856233 in Perl_runops_standard () at run.c:41 > > > #4 0x000055d8a47dbf36 in Perl_call_sv (sv=0x55d8a52c0ec0, > > > flags=flags@entry=45) > > > at perl.c:2872 > > > #5 0x000055d8a48629f6 in S_curse (sv=sv@entry=0x55d8a60a6410, > > > check_refcnt=check_refcnt@entry=true) at sv.c:6980 > > > #6 0x000055d8a48633dd in Perl_sv_clear > > > (orig_sv=orig_sv@entry=0x55d8a60a6410) > > > at sv.c:6584 > > > #7 0x000055d8a48636de in Perl_sv_free2 (sv=0x55d8a60a6410, > > > rc=<optimized out>) > > > at sv.c:7081 > > > #8 0x000055d8a48618d7 in S_visit (f=f@entry=0x55d8a48638c0 > > > <do_clean_objs>, > > > flags=flags@entry=2048, mask=mask@entry=2048) at sv.c:476 > > > #9 0x000055d8a4863a3e in Perl_sv_clean_objs () at sv.c:627 > > > #10 0x000055d8a47deb86 in perl_destruct (my_perl=<optimized out>) at > > > perl.c:883 > > > #11 0x00007fb6c6613c9e in perlish_exit () at State.xs:1520 > > > #12 0x00007fb6c6616e93 in cctx_run (arg=<optimized out>) at > > > State.xs:1571 > > > #13 0x00007fb6c6615f35 in coro_init () at libcoro/coro.c:97 > > > #14 0x00007fb6c71e3290 in ?? () from /lib/x86_64-linux-gnu/libc.so.6 > > > #15 0x0000000000000000 in ?? () > > > > > > Anoter v5.27.2, compiled with -DDEBUGGING, fails with this assertion: > > > > > > perl: pp_hot.c:2489: Perl_pp_multideref: Assertion `isGV_with_GP(sv)' > > > failed. > > > > > > Let me know whether I can provide more info. > > > > > > HTH&&Thanks,
Download (untitled) / with headers
text/plain 3.3k
This Coro::ProcessPool test failure: http://www.cpantesters.org/cpan/report/84f15dfe-c4b1-11e7-bca4-9afccc55d78c ...fails identically to this one for Coro under 5.27: http://www.cpantesters.org/cpan/report/e6737314-909a-11e7-ad69-bb03d9a57ab1 I suspect this is not a bug in Coro::ProcessPool but yet another compatibility issue between Coro and perl's internals. I would not expect stable support until the next stable perl release has been finalized. On Wed Nov 08 13:31:03 2017, SREZIC wrote: Show quoted text
> It seems that there's no improvement with Coro-ProcessPool-0.26_003. I > see SIGBUS (FreeBSD) and SIGSEGV (Linux) crashes in the test scripts, > for various perl versions. > > On 2017-07-25 08:19:13, JEFFOBER wrote:
> > Thank you for the report. Marc Lehmann just released Coro 6.513, > > which > > has experimental support for newer perls, triggering a flurry of test > > failures in modules using Coro, especially those using > > AnyEvent::Impl::Perl in testing (which is useful, since most, if not > > all, of the other AnyEvent event loop plugins have external > > dependencies, such as libevent). > > > > I am working on a fix for these errors, or at least a temporary work- > > around to prevent testing on modern perls until Coro support for them > > stabilizes. > > > > On Sat Jul 22 15:36:08 2017, ANDK wrote:
> > > Some of my smokers already sent fail reports with a SEGV > > > indication. > > > Here is a stack trace that I just observed with v5.27.2: > > > > > > Program terminated with signal SIGSEGV, Segmentation fault. > > > #0 coro_semaphore_adjust (av=0x0, adjust=adjust@entry=1) at > > > State.xs:3099 > > > 3099 State.xs: No such file or directory. > > > (gdb) bt > > > #0 coro_semaphore_adjust (av=0x0, adjust=adjust@entry=1) at > > > State.xs:3099 > > > #1 0x00007fb6c661b440 in XS_Coro__Semaphore_up (cv=<optimized
> > > out>)
> > > at State.xs:4359 > > > #2 0x000055d8a485e158 in Perl_pp_entersub () at pp_hot.c:4235 > > > #3 0x000055d8a4856233 in Perl_runops_standard () at run.c:41 > > > #4 0x000055d8a47dbf36 in Perl_call_sv (sv=0x55d8a52c0ec0, > > > flags=flags@entry=45) > > > at perl.c:2872 > > > #5 0x000055d8a48629f6 in S_curse (sv=sv@entry=0x55d8a60a6410, > > > check_refcnt=check_refcnt@entry=true) at sv.c:6980 > > > #6 0x000055d8a48633dd in Perl_sv_clear > > > (orig_sv=orig_sv@entry=0x55d8a60a6410) > > > at sv.c:6584 > > > #7 0x000055d8a48636de in Perl_sv_free2 (sv=0x55d8a60a6410, > > > rc=<optimized out>) > > > at sv.c:7081 > > > #8 0x000055d8a48618d7 in S_visit (f=f@entry=0x55d8a48638c0 > > > <do_clean_objs>, > > > flags=flags@entry=2048, mask=mask@entry=2048) at sv.c:476 > > > #9 0x000055d8a4863a3e in Perl_sv_clean_objs () at sv.c:627 > > > #10 0x000055d8a47deb86 in perl_destruct (my_perl=<optimized out>) > > > at > > > perl.c:883 > > > #11 0x00007fb6c6613c9e in perlish_exit () at State.xs:1520 > > > #12 0x00007fb6c6616e93 in cctx_run (arg=<optimized out>) at > > > State.xs:1571 > > > #13 0x00007fb6c6615f35 in coro_init () at libcoro/coro.c:97 > > > #14 0x00007fb6c71e3290 in ?? () from /lib/x86_64-linux- > > > gnu/libc.so.6 > > > #15 0x0000000000000000 in ?? () > > > > > > Anoter v5.27.2, compiled with -DDEBUGGING, fails with this > > > assertion: > > > > > > perl: pp_hot.c:2489: Perl_pp_multideref: Assertion > > > `isGV_with_GP(sv)' > > > failed. > > > > > > Let me know whether I can provide more info. > > > > > > HTH&&Thanks,
-- -Jeff Ober


This service is sponsored and maintained by Best Practical Solutions and runs on Perl.org infrastructure.

Please report any issues with rt.cpan.org to rt-cpan-admin@bestpractical.com.