On Wed, 2005-08-24 at 15:52 -0400, Jeff Garzik wrote:

> Three comments:
> 
> 1) Since you may have to deal with the SMP case, don't you need to add 
> wmb() or use atomic bit tests?

I don't think there will be any nasty races between bnx2_reset_task()
and bnx2_close(). If they are racing, the reset task will see that
netif_running() is zero and will just return.

> 
> 2) Would prefer to follow the generic net stack and other areas of the 
> kernel, for the last piece of quoted code.  net stack used to loop on 
> schedule_timeout(1) in dev_close() [net/core/dev.c], which has now been 
> updated to loop on msleep(1).  As the code comment there notes, we're 
> not in a hurry here.
> 

I see other drivers doing yield() also. But if you prefer, I can change
it to msleep().

> 3) Once bnx2 is fixed, any chance you could be talked into proposing 
> patches for the other drivers with this problem?  If you don't have 
> access to hardware to test, that's not a big deal.  Just note that in 
> the patch description.

Yeah, I think so. It may take a while to get to it though.

> 
> Otherwise, patch is OK.
> 
>       Jeff
> 
> 
> 

-
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to [EMAIL PROTECTED]
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Reply via email to