Re: [Bug #14925] sky2 panic under load

[Jarek Poplawski <jarkao2@gmail.com>]

On Sun, Jan 10, 2010 at 05:36:46PM -0700, Berck E. Nash wrote:
> Rafael J. Wysocki wrote:
> > This message has been generated automatically as a part of a report
> > of recent regressions.
> > 
> > The following bug entry is on the current list of known regressions
> > from 2.6.32.  Please verify if it still should be listed and let me know
> > (either way).

BTW, I don't know why Berck didn't experience such a panic before
2.6.32, but seems not a regression to me. There might be new/more sky2
TX timeouts which trigger this panic and would make a real regression.

Cheers,
Jarek P.

> > 
> > 
> > Bug-Entry	: http://bugzilla.kernel.org/show_bug.cgi?id=14925
> > Subject		: sky2 panic under load
> > Submitter	: Berck E. Nash <flyboy@gmail.com>
> > Date		: 2009-12-21 23:52 (21 days old)
> > References	: http://marc.info/?l=linux-kernel&m=126143955730347&w=4
> > 		  http://marc.info/?l=linux-kernel&m=126160893126548&w=4
> > Handled-By	: Stephen Hemminger <shemminger@vyatta.com>
> 
> The patch attached to the bug report did not fix the problem, but I'm
> fairly certain that this one from Jarek P. did:
> 
> During TX timeout procedure dev could be awoken too early, e.g. by
> sky2_complete_tx() called from sky2_down(). Then sky2_xmit_frame()
> can run while buffers are freed causing an oops. This patch fixes it
> by adding netif_device_present() test in sky2_tx_complete().
> 
> Fixes: http://bugzilla.kernel.org/show_bug.cgi?id=14925
> 
> With debugging by: Mike McCormack <mikem@ring3k.org>
> 
> Reported-by: Berck E. Nash <flyboy@gmail.com>
> Tested-by: Berck E. Nash <flyboy@gmail.com>
> Signed-off-by: Jarek Poplawski <jarkao2@gmail.com>
> 
> ---
> 
>  drivers/net/sky2.c |    3 ++-
>  1 files changed, 2 insertions(+), 1 deletions(-)
> 
> diff --git a/drivers/net/sky2.c b/drivers/net/sky2.c
> index 1c01b96..7650f73 100644
> --- a/drivers/net/sky2.c
> +++ b/drivers/net/sky2.c
> @@ -1844,7 +1844,8 @@ static void sky2_tx_complete(struct sky2_port
> *sky2, u16 done)
>  	sky2->tx_cons = idx;
>  	smp_mb();
> 
> -	if (tx_avail(sky2) > MAX_SKB_TX_LE + 4)
> +	/* Wake unless it's detached, and called e.g. from sky2_down() */
> +	if (tx_avail(sky2) > MAX_SKB_TX_LE + 4 && netif_device_present(dev))
>  		netif_wake_queue(dev);
>  }
>