[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: new branch in gm2 git repro for 10.0.2
From: |
Gaius Mulley |
Subject: |
Re: new branch in gm2 git repro for 10.0.2 |
Date: |
Mon, 10 Aug 2020 18:05:38 +0100 |
User-agent: |
Gnus/5.13 (Gnus v5.13) Emacs/24.5 (gnu/linux) |
Dragiša Durić <dragisha@m3w.org> writes:
> Hi Gaius,
>
> Built from Savannah... It looks like not only RTentity's thread-safety
> is problematic. This is what happens mst often.
>
> (gdb) r
> The program being debugged has been started already.
> Start it from the beginning? (y or n) y
> Starting program: /home/dragisha/bluewd-m2/gm2.lib0/a.out
> [Thread debugging using libthread_db enabled]
> Using host libthread_db library "/lib/x86_64-linux-gnu/libthread_db.so.1".
> [New Thread 0x7ffff5ce5700 (LWP 16752)]
>
> Thread 1 "a.out" received signal SIGSEGV, Segmentation fault.
> _int_malloc (av=av@entry=0x7ffff6d24c40, bytes=bytes@entry=16385) at
> malloc.c:4100
> 4100 malloc.c: No such file or directory.
> (gdb) bt
> #0 _int_malloc (av=av@entry=0x7ffff6d24c40, bytes=bytes@entry=16385)
> at malloc.c:4100
> #1 0x00007ffff69d035d in __GI___libc_malloc (bytes=16385) at malloc.c:3065
> #2 0x0000000000475974 in Storage_ALLOCATE (addr=@0x7fffffffde08:
> 0x7fff00000029, amount=16385)
> at
> ../../../../gcc-versionno/libgm2/libiso/../../gcc/gm2/gm2-libs-iso/Storage.mod:36
> #3 0x000000000041ecd5 in DoFormat (res=..., args=0x725410) at hStrIO.mod:205
> #4 0x000000000041e70c in hStrIO_PrintF (res=..., args=0x725410) at
> hStrIO.mod:476
> #5 0x000000000046ad2f in AdHoc_Init (getyear=1, top=1) at AdHoc.mod:84
> #6 0x0000000000472144 in _M2_test5_init () at test5.mod:41
> #7 0x0000000000405ff1 in init (argc=1, argv=0x7fffffffe148) at
> test5_m2.cpp:282
> #8 0x0000000000406249 in main (argc=1, argv=0x7fffffffe148) at
> test5_m2.cpp:392
>
>
> Regards,
> Dragiša
Hi Dragiša,
many thanks for unearthing these bugs. Is there any chance test code
could be emailed - I'd not look at it - just really want to reproduce
the bug and debug the libraries? Failing that a brief description of
the calls the program is performing.
I'm extending the testsuite with a timer based example to stress
preemption - which hopefully will stress these libraries further
regards,
Gaius