> On Fri, 25 Apr 2008, PJ Waskiewicz wrote:
quoted text
> > This bug was introduced in the 2.6.24 i386/x86_64 tree merge, where
> 
> Can you please explain what exactly caused the bug. 
> Definitely not the move from arch/i386 to arch/x86 as the 
> code there was not changed at all and has not be changed since then.
> 
> CC'ed Eric as well.

Eric replied with the actual commit during the 2.6.24 merge window that
introduced this bug.  The io_apic.c code from the i386 tree did not stay
completely static when it was merged into the x86 io_apic_32.c code.
Here is the commit that Eric identified that introduced the defect:

In particular commit dbeb2be21d678c49a8d8bbf774903df15dd55474
Author: Rusty Russell <rusty@rustcorp.com.au>
Date:   Fri Oct 19 20:35:03 2007 +0200

    i386: introduce "used_vectors" bitmap which can be used to reserve
vectors.
    
    This simplifies the io_apic.c __assign_irq_vector() logic and
removes
    the explicit SYSCALL_VECTOR check, and also allows for vectors to be
    reserved by other mechanisms (ie. lguest).
    
    [ tglx: arch/x86 adaptation ]
    
    Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
    Signed-off-by: Andi Kleen <ak@suse.de>
    Signed-off-by: Ingo Molnar <mingo@elte.hu>
    Signed-off-by: Thomas Gleixner <tglx@linutronix.de>

Basically the code introduced the test_and_set_bit() on the used_vectors
bitmap, but it didn't have a corresponding clear_bit() on IRQ
destruction.

Cheers,
-PJ Waskiewicz
--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html