Re: [patch 3/4] 2.5.31 i2c updates

[Russell King]

On Mon, Aug 12, 2002 at 09:57:50AM -0400, Albert Cranford wrote:
> @@ -121,12 +118,12 @@
>         int timeout = 2;
>  
>         if (irq > 0) {
> -               cli();
> +               spin_lock_irq(&irq_driver_lock);
>                 if (pcf_pending == 0) {
>                         interruptible_sleep_on_timeout(&pcf_wait, timeout*HZ );
>                 } else
>                         pcf_pending = 0;
> -               sti();
> +               spin_unlock_irq(&irq_driver_lock);

In case you didn't get my previous message, sleeping with spinlocks held
is the perfect path to deadlock.  Also, scheduling with interrupts disabled
is not a good idea.  Unfortunately the code above does both.

-- 
Russell King ([e-mail])                The developer of ARM Linux
             http://www.arm.linux.org.uk/personal/aboutme.html