Hi Richard,

Felipe W Damasio wrote:
>     This is second version of the patch (against 2.6-test4) to fix SMP 
> support on the 3c527 net driver, by removing cli/sti calls and replacing 
> them with local locks.
> 
>     This one includes Manfred's idea of using wait_event instead of 
> sleep_on, and releasing the lock to prevent the driver from going 
> bezerk..it also uses spinlocks, since on some situations we are in 
> bottom half context.

	Please try this patch instead.

	This one holds the device lock before doing "finish_wait", which 
seems to be the Right Way to do it.

	Thanks.

Felipe