[openib-general] Re: LLTX and netif_stop_queue

[jamal]

On Sat, 2004-12-18 at 10:35, Roland Dreier wrote:

> By the way, am I correct in thinking that the use of xmit_lock_owner
> in qdisc_restart() is racy?

No.

>     if (!spin_trylock(&dev->xmit_lock)) {
>     /* get the lock */
> 
>                                             if (!spin_trylock(&dev->xmit_lock)) {
>                                             /* fail */
>                                                 if (dev->xmit_lock_owner == smp_processor_id()) {
>                                                 /* test the wrong value */
> 
>     /* set the value too late */
>     dev->xmit_lock_owner = smp_processor_id();
> 

The setting is protected by the queue lock. So no race.

cheers,
jamal