eepro100 question: why SCBCmd byte is 0x80?

eepro100 question: why SCBCmd byte is 0x80?

[Jun Sun]

I am trying to get netgear card working on a new (read as potentially buggy
hardware) MIPS board.

The eepro100 driver basically works fine.  It is just after a little while
(usually 2 sec to 15 sec) network communication suddenly stops and I start see
error message like "eepro100: wait_for_cmd_done timeout!".

I looked into this, and it appears that the SCBCmd byte in the command word
has value 0x80 instead of the expected 0.  I looked at the Intel manual, and
it says nothing about the value being 0x80.

Does anybody have a clue here?  I suspect some timing is wrong or a buggy PCI
controller.

Please cc your reply to my email address.  Thanks.

Jun

Re: eepro100 question: why SCBCmd byte is 0x80?

[christophe barbe]

Which kernel are you using.

I've had a similar problem with 2.2.18.
I've backported 2.2.19pre changes to it. 
(i.e. apply on 2.2.18 a diff of the file drivers/net/eepro100.c made between 2.2.18 and the last 2.2.19pre)
And since I've never seen this problem again.

Christophe

On jeu, 22 mar 2001 22:04:45 Jun Sun wrote:
> 
> I am trying to get netgear card working on a new (read as potentially buggy
> hardware) MIPS board.
> 
> The eepro100 driver basically works fine.  It is just after a little while
> (usually 2 sec to 15 sec) network communication suddenly stops and I start see
> error message like "eepro100: wait_for_cmd_done timeout!".
> 
> I looked into this, and it appears that the SCBCmd byte in the command word
> has value 0x80 instead of the expected 0.  I looked at the Intel manual, and
> it says nothing about the value being 0x80.
> 
> Does anybody have a clue here?  I suspect some timing is wrong or a buggy PCI
> controller.
> 
> Please cc your reply to my email address.  Thanks.
> 
> Jun
> -
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at  http://www.tux.org/lkml/
> 
-- 
Christophe Barbé
Software Engineer
Lineo High Availability Group
42-46, rue Médéric
92110 Clichy - France
phone (33).1.41.40.02.12
fax (33).1.41.40.02.01
www.lineo.com

Re: eepro100 question: why SCBCmd byte is 0x80?

[Jun Sun]

christophe barbe wrote:
> 
> Which kernel are you using.
> 
> I've had a similar problem with 2.2.18.
> I've backported 2.2.19pre changes to it.
> (i.e. apply on 2.2.18 a diff of the file drivers/net/eepro100.c made between 2.2.18 and the last 2.2.19pre)
> And since I've never seen this problem again.
> 
> Christophe
> 

Kernel is 2.4.2.  It is a MIPS machine.

I don't really think it is a driver problem, because the same dirver works
fine on many other boards (including MIPS boards).  In addition, I also tested
with tulip cards and I had the same symptom.  I am convinced it is a low-level
problem (bus timing, PCI setting, buggy hardware, etc).

On the other hand, it could be a driver problem which is only exposed in this
particular board, although very unlikely.

BTW, does the eepro100 patch for 2.2.19pre apply to 2.4.2?  Or it is already
in it?

Thanks.

Jun

> On jeu, 22 mar 2001 22:04:45 Jun Sun wrote:
> >
> > I am trying to get netgear card working on a new (read as potentially buggy
> > hardware) MIPS board.
> >
> > The eepro100 driver basically works fine.  It is just after a little while
> > (usually 2 sec to 15 sec) network communication suddenly stops and I start see
> > error message like "eepro100: wait_for_cmd_done timeout!".
> >
> > I looked into this, and it appears that the SCBCmd byte in the command word
> > has value 0x80 instead of the expected 0.  I looked at the Intel manual, and
> > it says nothing about the value being 0x80.
> >
> > Does anybody have a clue here?  I suspect some timing is wrong or a buggy PCI
> > controller.
> >
> > Please cc your reply to my email address.  Thanks.
> >
> > Jun
> > -
> > To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> > the body of a message to majordomo@vger.kernel.org
> > More majordomo info at  http://vger.kernel.org/majordomo-info.html
> > Please read the FAQ at  http://www.tux.org/lkml/
> >
> --
> Christophe Barbé
> Software Engineer
> Lineo High Availability Group
> 42-46, rue Médéric
> 92110 Clichy - France
> phone (33).1.41.40.02.12
> fax (33).1.41.40.02.01
> www.lineo.com

Re: eepro100 question: why SCBCmd byte is 0x80?

[Steven Walter]

I'm having a similar problem with the onboard network card of a Sony
Vaio Laptop.  I haven't tracked it down as far as you can; how can I
confirm its the same problem as yours?

On Fri, Mar 23, 2001 at 09:34:36AM -0800, Jun Sun wrote:
> christophe barbe wrote:
> > 
> > Which kernel are you using.
> > 
> > I've had a similar problem with 2.2.18.
> > I've backported 2.2.19pre changes to it.
> > (i.e. apply on 2.2.18 a diff of the file drivers/net/eepro100.c made between 2.2.18 and the last 2.2.19pre)
> > And since I've never seen this problem again.
> > 
> > Christophe
> > 
> 
> Kernel is 2.4.2.  It is a MIPS machine.
> 
> I don't really think it is a driver problem, because the same dirver works
> fine on many other boards (including MIPS boards).  In addition, I also tested
> with tulip cards and I had the same symptom.  I am convinced it is a low-level
> problem (bus timing, PCI setting, buggy hardware, etc).
> 
> On the other hand, it could be a driver problem which is only exposed in this
> particular board, although very unlikely.
> 
> BTW, does the eepro100 patch for 2.2.19pre apply to 2.4.2?  Or it is already
> in it?
> 
> Thanks.
> 
> Jun
> 
> > On jeu, 22 mar 2001 22:04:45 Jun Sun wrote:
> > >
> > > I am trying to get netgear card working on a new (read as potentially buggy
> > > hardware) MIPS board.
> > >
> > > The eepro100 driver basically works fine.  It is just after a little while
> > > (usually 2 sec to 15 sec) network communication suddenly stops and I start see
> > > error message like "eepro100: wait_for_cmd_done timeout!".
> > >
> > > I looked into this, and it appears that the SCBCmd byte in the command word
> > > has value 0x80 instead of the expected 0.  I looked at the Intel manual, and
> > > it says nothing about the value being 0x80.
> > >
> > > Does anybody have a clue here?  I suspect some timing is wrong or a buggy PCI
> > > controller.
> > >
> > > Please cc your reply to my email address.  Thanks.
> > >
> > > Jun
> > > -
> > > To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> > > the body of a message to majordomo@vger.kernel.org
> > > More majordomo info at  http://vger.kernel.org/majordomo-info.html
> > > Please read the FAQ at  http://www.tux.org/lkml/
> > >
> > --
> > Christophe Barbé
> > Software Engineer
> > Lineo High Availability Group
> > 42-46, rue Médéric
> > 92110 Clichy - France
> > phone (33).1.41.40.02.12
> > fax (33).1.41.40.02.01
> > www.lineo.com
> -
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at  http://www.tux.org/lkml/

-- 
-Steven
Freedom is the freedom to say that two plus two equals four.

Re: eepro100 question: why SCBCmd byte is 0x80?

[Ion Badulescu]

On Fri, 23 Mar 2001 09:34:36 -0800, Jun Sun <jsun@mvista.com> wrote:

> BTW, does the eepro100 patch for 2.2.19pre apply to 2.4.2?  Or it is already
> in it?

It was backported from 2.4.1, so yes, it's already in.

Ion

-- 
  It is better to keep your mouth shut and be thought a fool,
            than to open it and remove all doubt.