Re: netgear fa311 NIC driver
On Fri, 24 Aug 2001, Noah Meyerhans wrote:
> On Fri, Aug 24, 2001 at 05:50:25PM -0700, bedlam@alumni.rice.edu wrote:
> > I have upgraded to kernel 2.4.9 so I would have support for the
> > netgear fa311 NIC, specifically the natsemi driver.
You didn't need to upgrade your kernel.
http://www.scyld.com/network/ethercard.html
ftp://www.scyld.com/pub/network/natsemi.c
> However, it is
> > behaving strangely. I repeatedly get the error message:
> >
> > eth1: Something Wicked Happened! 18000
Hmmm, the Rx status steck overflowed.
I don't see this with my driver release.
This is very unusual -- either the driver misconfigured the chip, or the
chip couldn't get any PCI bus bandwidth to transfer Rx descriptors.
> I can't tell if it is a bug in the National Semiconductor chip on the
> card or with the driver. From what I was able to determine, it has
> something to do with IRQ sharing, though I'm not totally sure what.
That's pretty unlikely, unless the interrupt is blocked for a really
long time. That's a problem with another driver on the system, not with
sharing IRQs.
> Many current BIOSes set up multiple devices on a single IRQ. The Linux
> kernel supports this on a higher level, but perhaps there's support
> needed in the driver that the natsemi driver doesn't have.
The driver does need to be aware of sharing, but that's not difficult to
get right.
What many drivers get wrong is that they fail teo limit the work that
they will do during an interrupt. Even worse are drivers, typically
SCSI, that register with SA_INTERRUPT and then do unlimited work.
> Between this bug and the natsemi driver's broken multicast support,
It was broken on my initial version, but I fixed it. The broken version
was apparently ported to 2.4.
> I've spent a good deal of time looking at the driver. I was never able to
> solve the problem, though, and eventually stopped trying as it was
> easier for me to just swap in another NIC. I'm cc'ing Donald Becker
> becker@scyld.com with this mail, as he's the primary author of the
> natsemi driver and may be able to shed some light on the situation.
As usual, my primary reply is that I don't support modified drivers. If
you touch it, you should be willing to test and maintain the whole
driver.
I will debug problems that you can reproduce on an original driver.
Donald Becker becker@scyld.com
Scyld Computing Corporation http://www.scyld.com
410 Severn Ave. Suite 210 Second Generation Beowulf Clusters
Annapolis MD 21403 410-990-9993
Reply to: