Re: new branch in gm2 git repro for 10.0.2

[Dragiša Durić]

Hi Gaius,

Not much time for Modula-2, but still problems... I build from
floopsie, changes from Aug 23 included.

Thread 1 "a.out" received signal SIGSEGV, Segmentation fault.
findChildAndParent (t=t@entry=0x70e760, a=a@entry=0x73af60, 
child=@0x7fffffffdeb8: 0xc400c400c400c4, parent=@0x7fffffffdeb0: 0x73af30)
    at 
../../../../../../gm2-floppsie/libgm2/libm2iso/../../gcc/m2/gm2-libs-iso/RTentity.mod:256
256           UNTIL (child = NIL) OR (a = child^.entity)
Warning: the current language does not match this frame.
(gdb) bt
#0  findChildAndParent (t=t@entry=0x70e760, a=a@entry=0x73af60, 
child=@0x7fffffffdeb8: 0xc400c400c400c4, parent=@0x7fffffffdeb0: 0x73af30)
    at 
../../../../../../gm2-floppsie/libgm2/libm2iso/../../gcc/m2/gm2-libs-iso/RTentity.mod:256
#1  0x0000000000475aee in RTentity_PutKey (g=0x70e760, a=0x73af60, key=168) at 
../../../../../../gm2-floppsie/libgm2/libm2iso/../../gcc/m2/gm2-libs-iso/RTentity.mod:110
#2  0x000000000044e2e5 in Windows_Define (u=@0x6e38c8: 0x73af60, params=0) at 
Windows.mod:837
#3  0x0000000000471751 in AdHoc_Init (getyear=1, top=1) at AdHoc.mod:87
#4  0x000000000047224d in _M2_test5_init () at test5.mod:41
#5  0x0000000000406116 in init (argc=1, argv=0x7fffffffe128) at 
a-test5_m2.cpp:285
#6  0x0000000000406371 in main (argc=1, argv=0x7fffffffe128) at 
a-test5_m2.cpp:396


Relevant code is a lot of it, I will try to make smaller example with
reproducible problem, as soon as I can.

Regards,
Dragisa

On Mon, 2020-08-10 at 18:05 +0100, Gaius Mulley wrote:
> 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