[PATCH -rt][RESEND] spin lock imbalance in ibm emac

[PATCH -rt][RESEND] spin lock imbalance in ibm emac

[Daniel Walker]

Sent this a long time ago, still exists. 

Signed-Off-By: Daniel Walker <dwalker@mvista.com>

Index: linux-2.6.17/drivers/net/ibm_emac/ibm_emac_core.c
===================================================================
--- linux-2.6.17.orig/drivers/net/ibm_emac/ibm_emac_core.c
+++ linux-2.6.17/drivers/net/ibm_emac/ibm_emac_core.c
@@ -1140,6 +1140,8 @@ static int emac_start_xmit_sg(struct sk_
 	if (likely(!nr_frags && len <= MAL_MAX_TX_SIZE))
 		return emac_start_xmit(skb, ndev);
 
+	spin_lock(&dev->tx_lock);
+
 	len -= skb->data_len;
 
 	/* Note, this is only an *estimation*, we can still run out of empty
@@ -1208,6 +1210,7 @@ static int emac_start_xmit_sg(struct sk_
       stop_queue:
 	netif_stop_queue(ndev);
 	DBG2("%d: stopped TX queue" NL, dev->def->index);
+	spin_unlock(&dev->tx_lock);
 	return 1;
 }
 #else
--

Re: [PATCH -rt][RESEND] spin lock imbalance in ibm emac

[Ingo Molnar]

* Daniel Walker <dwalker@mvista.com> wrote:

> Sent this a long time ago, still exists. 
> 
> Signed-Off-By: Daniel Walker <dwalker@mvista.com>

hm, what does this do, and why isnt it upstream?

	Ingo

Re: [PATCH -rt][RESEND] spin lock imbalance in ibm emac

[Daniel Walker]

On Mon, 2006-12-11 at 10:27 +0100, Ingo Molnar wrote:
> * Daniel Walker <dwalker@mvista.com> wrote:
> 
> > Sent this a long time ago, still exists. 
> > 
> > Signed-Off-By: Daniel Walker <dwalker@mvista.com>
> 
> hm, what does this do, and why isnt it upstream?


AFAIK, those locks are added in -rt . I'm not sure how they got in
there, but they fix that driver when running in a thread. The driver has
unsafe SMP locking, but the only system it runs on (PPC4xx) is
uniprocessor. So it's not broken upstream per se.

Dainel

Re: [PATCH -rt][RESEND] spin lock imbalance in ibm emac

[Ingo Molnar]

* Daniel Walker <dwalker@mvista.com> wrote:

> On Mon, 2006-12-11 at 10:27 +0100, Ingo Molnar wrote:
> > * Daniel Walker <dwalker@mvista.com> wrote:
> > 
> > > Sent this a long time ago, still exists. 
> > > 
> > > Signed-Off-By: Daniel Walker <dwalker@mvista.com>
> > 
> > hm, what does this do, and why isnt it upstream?
> 
> AFAIK, those locks are added in -rt . I'm not sure how they got in 
> there, but they fix that driver when running in a thread. The driver 
> has unsafe SMP locking, but the only system it runs on (PPC4xx) is 
> uniprocessor. So it's not broken upstream per se.

ok, i've applied your patch.

	Ingo

Re: [PATCH -rt][RESEND] spin lock imbalance in ibm emac

[Daniel Walker]

On Mon, 2006-12-11 at 16:59 +0100, Ingo Molnar wrote:
> * Daniel Walker <dwalker@mvista.com> wrote:
> 
> > On Mon, 2006-12-11 at 10:27 +0100, Ingo Molnar wrote:
> > > * Daniel Walker <dwalker@mvista.com> wrote:
> > > 
> > > > Sent this a long time ago, still exists. 
> > > > 
> > > > Signed-Off-By: Daniel Walker <dwalker@mvista.com>
> > > 
> > > hm, what does this do, and why isnt it upstream?
> > 
> > AFAIK, those locks are added in -rt . I'm not sure how they got in 
> > there, but they fix that driver when running in a thread. The driver 
> > has unsafe SMP locking, but the only system it runs on (PPC4xx) is 
> > uniprocessor. So it's not broken upstream per se.
> 
> ok, i've applied your patch.


What about the OMAP fix? that one's been sitting around for a while too.

Daniel