Assert in linux_nat_resume

Assert in linux_nat_resume

[Marc Ferland]

Hi,

I've been having the following problem with GDB version 6.8 for a while now:

[New LWP 20436]
linux-nat.c:1152: internal-error: linux_nat_resume: Assertion `lp != 
NULL' failed.
A problem internal to GDB has been detected,
further debugging may prove unreliable.
Quit this debugging session? (y or n) y
linux-nat.c:1152: internal-error: linux_nat_resume: Assertion `lp != 
NULL' failed.
A problem internal to GDB has been detected,
further debugging may prove unreliable.
Create a core file of GDB? (y or n) n


I'm debugging a program with many threads and this happens often when I 
close(pthread_exit()) and start threads (pthread_create). I really don't 
know what is wrong.

I'm pretty sure it has something to do with my program but I don't know 
where to start (Debugging with printf() is a rather tedious task!). Any 
help would be greatly appreciated.

Regards,

Marc

Re: Assert in linux_nat_resume

[Daniel Jacobowitz]

On Tue, Aug 12, 2008 at 03:21:40PM -0400, Marc Ferland wrote:
> I'm pretty sure it has something to do with my program but I don't know  
> where to start (Debugging with printf() is a rather tedious task!). Any  
> help would be greatly appreciated.

The first thing to try is definitely a snapshot of current CVS GDB.

-- 
Daniel Jacobowitz
CodeSourcery

Re: Assert in linux_nat_resume

[Marc Ferland]

Daniel Jacobowitz wrote:
> On Tue, Aug 12, 2008 at 03:21:40PM -0400, Marc Ferland wrote:
>> I'm pretty sure it has something to do with my program but I don't know  
>> where to start (Debugging with printf() is a rather tedious task!). Any  
>> help would be greatly appreciated.
> 
> The first thing to try is definitely a snapshot of current CVS GDB.
> 
I tried with 6.8.50.20080812-cvs seems to have a similar problem.


[New LWP 29169]
linux-nat.c:1669: internal-error: linux_nat_resume: Assertion `lp != 
NULL' failed.
A problem internal to GDB has been detected,
further debugging may prove unreliable.

After further testing, this seems to happen only when I have a 
breakpoint somewhere inside a thread. The first time I hit the 
breakpoint everything is fine. But if I exit a thread and recreate it, I 
get the previously mentioned message.

I also found a workaround:
- The first time I hit the breakpoint, I delete it;
- I place a breakpoint on the pthread_join() call for this thread;
- When I break on the pthread_join() call, I recreate the breakpoint 
inside the thread;

This way, I'm able to break inside the thread every time I create it 
(pthread_create()).

Marc