Re: Kernel crash in interrupt handler: nested interrupt breaks saved %eip?

[Nikita V. Youshchenko]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

> > EIP:    0060:[<c02262a2>]    Tainted: P
>                                ^^^^^^^^^^
> What was that? Your lsmod please?

That's caused by vmware bodules; I don't think they affect the situation.

> > So I think that an interrupt happened at that time, and %eip was
> > broken inside the handler.
>
> You can istrument your kernel to check for that on every interrupt
> and printk a KERN_NOTICE message if eip was changed.

What's the correct place for those? do_IRQ()? Or probably some other places 
also (APIC timer interrupt handler? exception handler?)

Unpleasant situation is that the problem is on a production server, which 
should be operational 24/7 ...  However, probably I'll try to install an 
instrumented kernel in the next kernel upgrade.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.4 (GNU/Linux)

iD8DBQFA7oLLsTbPknTfAB4RAhGvAJ9n9246V7cMntgPnwmKBI/pcR0XGACeMnSY
fuIaVh1yCssaJC0PWdHd7sE=
=/8fI
-----END PGP SIGNATURE-----
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/