Hi everybody,
I have reached a point where I am wondering if a call to 'malloc' with
the M_NOWAIT flag is not falling asleep !
In fact, I suspect that the interrupted context is somewhere during a
call to 'malloc' (I increment a counter just before calling malloc and
increment another just after and the difference is one !) while I have
called 'splnet' beforehand, thus normally preventing competing with any
network isr. I assume that this shouldnever occur unless the code is
somewhere calling 'sleep' and provoke acontext switch.
Is there anybody who can help on this ?
Best regards,
Xavier
P.S.: I am still looking how I could dump the kernel stack frame of an
interrupted process from the actual interrupt kernel stack ...
Xavier Galleri wrote:
> Hi everybody,
>
> I am currently working with FreeBSD 4.1 on i386 platform on a
> pseudo-device driver that interacts with the networking subsystem in
> the kernel. Basically, we are intercepting both incoming and outgoing
> IP-level trafficfor specific processing. Obviously, there is also some
> administrative processingthat is conducted from a user-land daemon.
>
> I have detected a problem with my driver occuring in the context of
> the network SWI, as shown by the following basic stack:
>
> #14 in ... /* private code */
> #15 in ip_input
> #16 in ipintr
>
> My problem comes from that I would like to get the stack frame of the
> interrupted execution process (everything that would be frame #17 and
> below) and I do not know how to do this with the GDB debugger.
>
> Is there anybody how could provide me with the steps to achieve this ?
>
> Regards,
>
> X. Galleri
>
To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-hackers" in the body of the message