I'm running 2.6.14-rc4-rt6 and trying the skge driver instead of the
sk98lin and I'm getting these warnings in my logs (this is probably not
related to the rt patch):
network driver disabled interrupts: skge_xmit_frame+0x0/0x320 [skge]
No other relevant messages around that I can see. Is this a bug? Any
information I could supply to help debug it?
-- Fernando
Fernando Lopez-Lezcano wrote:
> I'm running 2.6.14-rc4-rt6 and trying the skge driver instead of the
> sk98lin and I'm getting these warnings in my logs (this is probably not
> related to the rt patch):
>
> network driver disabled interrupts: skge_xmit_frame+0x0/0x320 [skge]
>
> No other relevant messages around that I can see. Is this a bug? Any
> information I could supply to help debug it?
This is a bogus message added by the -rt patch. It is not a bug.
The trylock scheme in some newer net drivers (grep for NETDEV_TX_LOCKED)
uses local_irq_save/restore because there is no
spin_trylock_irqsave/spin_trylock_failed_irqrestore API.
Jeff
On Mon, 2005-10-17 at 21:57 -0400, Jeff Garzik wrote:
> Fernando Lopez-Lezcano wrote:
> > I'm running 2.6.14-rc4-rt6 and trying the skge driver instead of the
> > sk98lin and I'm getting these warnings in my logs (this is probably not
> > related to the rt patch):
> >
> > network driver disabled interrupts: skge_xmit_frame+0x0/0x320 [skge]
> >
> > No other relevant messages around that I can see. Is this a bug? Any
> > information I could supply to help debug it?
>
> This is a bogus message added by the -rt patch. It is not a bug.
>
> The trylock scheme in some newer net drivers (grep for NETDEV_TX_LOCKED)
> uses local_irq_save/restore because there is no
> spin_trylock_irqsave/spin_trylock_failed_irqrestore API.
Would it have any undesirable effect to find this and comment it out?
There are quite a few messages in the logs. Knowing it is not a bug I
may try the driver a bit more (I rebooted into sk98lin just in case ;-)
Thanks for the info.
-- Fernando
On Mon, 17 Oct 2005 19:02:33 -0700
Fernando Lopez-Lezcano <[email protected]> wrote:
> On Mon, 2005-10-17 at 21:57 -0400, Jeff Garzik wrote:
> > Fernando Lopez-Lezcano wrote:
> > > I'm running 2.6.14-rc4-rt6 and trying the skge driver instead of the
> > > sk98lin and I'm getting these warnings in my logs (this is probably not
> > > related to the rt patch):
> > >
> > > network driver disabled interrupts: skge_xmit_frame+0x0/0x320 [skge]
> > >
> > > No other relevant messages around that I can see. Is this a bug? Any
> > > information I could supply to help debug it?
> >
> > This is a bogus message added by the -rt patch. It is not a bug.
> >
> > The trylock scheme in some newer net drivers (grep for NETDEV_TX_LOCKED)
> > uses local_irq_save/restore because there is no
> > spin_trylock_irqsave/spin_trylock_failed_irqrestore API.
>
> Would it have any undesirable effect to find this and comment it out?
> There are quite a few messages in the logs. Knowing it is not a bug I
> may try the driver a bit more (I rebooted into sk98lin just in case ;-)
>
> Thanks for the info.
> -- Fernando
Or get the -rt folks to come with a better way.
> -
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to [email protected]
> More majordomo info at http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at http://www.tux.org/lkml/
--
Stephen Hemminger <[email protected]>
OSDL http://developer.osdl.org/~shemminger
On Tue, 18 Oct 2005, Stephen Hemminger wrote:
> On Mon, 17 Oct 2005 19:02:33 -0700
> Fernando Lopez-Lezcano <[email protected]> wrote:
>
> > On Mon, 2005-10-17 at 21:57 -0400, Jeff Garzik wrote:
> > > Fernando Lopez-Lezcano wrote:
> > > > I'm running 2.6.14-rc4-rt6 and trying the skge driver instead of the
> > > > sk98lin and I'm getting these warnings in my logs (this is probably not
> > > > related to the rt patch):
> > > >
> > > > network driver disabled interrupts: skge_xmit_frame+0x0/0x320 [skge]
> > > >
> > > > No other relevant messages around that I can see. Is this a bug? Any
> > > > information I could supply to help debug it?
> > >
> > > This is a bogus message added by the -rt patch. It is not a bug.
> > >
> > > The trylock scheme in some newer net drivers (grep for NETDEV_TX_LOCKED)
> > > uses local_irq_save/restore because there is no
> > > spin_trylock_irqsave/spin_trylock_failed_irqrestore API.
> >
> > Would it have any undesirable effect to find this and comment it out?
> > There are quite a few messages in the logs. Knowing it is not a bug I
> > may try the driver a bit more (I rebooted into sk98lin just in case ;-)
> >
> > Thanks for the info.
> > -- Fernando
>
> Or get the -rt folks to come with a better way.
>
Like adding a spin_trylock_irqsave/spin_trylock_failed_irqrestore API?
-- Steve
* Steven Rostedt <[email protected]> wrote:
> Like adding a spin_trylock_irqsave/spin_trylock_failed_irqrestore API?
there's spin_trylock_irqsave() already, which can be used just fine.
but in any case, this is something for the -rt tree only, with which we
dont want to pester upstream ...
Ingo