Hi All.
(Sorry had to write to this list, as the subscription to
[email protected] seemingly went successful, but never got the
confirmation email. Even sending emails get bounced).
I have been trying to analyze a core, but unable to receive the
complete backtrace due to following error :
#######################################
#0 0x0088ad6c in __libc_signal_restore_set (set=0xf48f2a78) at
../sysdeps/unix/sysv/linux/nptl-signals.h:79
79 ../sysdeps/unix/sysv/linux/nptl-signals.h: No such file or directory.
#######################################
Following is the (incomplete) backtrace :
#######################################
[Current thread is 1 (LWP 7491)]
(gdb) bt
#0 0x0088ad6c in __libc_signal_restore_set (set=0xf48f2a78) at
../sysdeps/unix/sysv/linux/nptl-signals.h:79
#1 __GI_raise (sig=sig@entry=6) at ../sysdeps/unix/sysv/linux/raise.c:48
#2 0x0088cb78 in __GI_abort () at abort.c:89
#3 <frame with modified names> 0x049b3a54 in func_1 (ffp=<optimized
out>, base_name=<optimized out>, max_size=<optimized out>,
file_type=<optimized out>) at a/b/c/d/e:58
#4 0x00000270 in ?? ()
(gdb)
#######################################
Since GDB is able to find raise.c, I did a full base-machine search
for it, via :
#######################################
find / -name raise.c 2> /dev/null
#######################################
but could not find raise.c either on the base-machine filesystem.
So, following are my queries :
1.
Where are raise.c, nptl-signals.h etc. supposed to reside?
2.
How can we bring back the missing nptl-signals.h file?
Will be grateful for pointers.
Thanks and Regards,
Ajay
+add some kernel gdb people + list
On 3/11/23 08:00, Ajay Garg wrote:
> Hi All.
>
> (Sorry had to write to this list, as the subscription to
> [email protected] seemingly went successful, but never got the
Is that @sourceware.org ?
> confirmation email. Even sending emails get bounced).
>
>
> I have been trying to analyze a core, but unable to receive the
> complete backtrace due to following error :
>
> #######################################
> #0 0x0088ad6c in __libc_signal_restore_set (set=0xf48f2a78) at
> ../sysdeps/unix/sysv/linux/nptl-signals.h:79
> 79 ../sysdeps/unix/sysv/linux/nptl-signals.h: No such file or directory.
> #######################################
>
>
> Following is the (incomplete) backtrace :
>
> #######################################
> [Current thread is 1 (LWP 7491)]
> (gdb) bt
> #0 0x0088ad6c in __libc_signal_restore_set (set=0xf48f2a78) at
> ../sysdeps/unix/sysv/linux/nptl-signals.h:79
> #1 __GI_raise (sig=sig@entry=6) at ../sysdeps/unix/sysv/linux/raise.c:48
> #2 0x0088cb78 in __GI_abort () at abort.c:89
> #3 <frame with modified names> 0x049b3a54 in func_1 (ffp=<optimized
> out>, base_name=<optimized out>, max_size=<optimized out>,
> file_type=<optimized out>) at a/b/c/d/e:58
> #4 0x00000270 in ?? ()
> (gdb)
> #######################################
>
>
> Since GDB is able to find raise.c, I did a full base-machine search
> for it, via :
>
> #######################################
> find / -name raise.c 2> /dev/null
> #######################################
>
> but could not find raise.c either on the base-machine filesystem.
>
>
> So, following are my queries :
>
> 1.
> Where are raise.c, nptl-signals.h etc. supposed to reside?
>
> 2.
> How can we bring back the missing nptl-signals.h file?
>
>
> Will be grateful for pointers.
>
>
> Thanks and Regards,
> Ajay
--
~Randy
On Sat, Mar 11, 2023 at 08:52:53AM -0800, Randy Dunlap wrote:
> +add some kernel gdb people + list
This looks like a userspace backtrace to me, so rather off-topic for
LKML but since I've read it already...
> On 3/11/23 08:00, Ajay Garg wrote:
> > Hi All.
> >
> > (Sorry had to write to this list, as the subscription to
> > [email protected] seemingly went successful, but never got the
>
> Is that @sourceware.org ?
>
> > confirmation email. Even sending emails get bounced).
> >
> >
> > I have been trying to analyze a core, but unable to receive the
> > complete backtrace due to following error :
> >
> > #######################################
> > #0 0x0088ad6c in __libc_signal_restore_set (set=0xf48f2a78) at
> > ../sysdeps/unix/sysv/linux/nptl-signals.h:79
> > 79 ../sysdeps/unix/sysv/linux/nptl-signals.h: No such file or directory.
> > #######################################
This is not why you are getting an incomplete backtrace!
> > Following is the (incomplete) backtrace :
> >
> > #######################################
> > [Current thread is 1 (LWP 7491)]
> > (gdb) bt
> > #0 0x0088ad6c in __libc_signal_restore_set (set=0xf48f2a78) at
> > ../sysdeps/unix/sysv/linux/nptl-signals.h:79
> > #1 __GI_raise (sig=sig@entry=6) at ../sysdeps/unix/sysv/linux/raise.c:48
> > #2 0x0088cb78 in __GI_abort () at abort.c:89
Frames 0-2 tell use your program called abort() (e.g. you got a core
dump because whatever you were running *asked* for a core dump to be
made)..
> > #3 <frame with modified names> 0x049b3a54 in func_1 (ffp=<optimized
> > out>, base_name=<optimized out>, max_size=<optimized out>,
> > file_type=<optimized out>) at a/b/c/d/e:58
Not sure if this is good or bad but...
> > #4 0x00000270 in ?? ()
... based on the address here I'd say that this stack from is for a
function provided by your application... and you forgot to compile the
application with debug information!
Daniel.