2.4 series IDE troubles

2.4 series IDE troubles

[j]

Hello,

IM not sure this is the way to go about this (obviously) but doing
searchs at google for the error message I get just leads to the painful realization that there are many questions out there but 
few answers. Anyhoo when I try adding a compact flash card as an 
IDE drive to a system running under 2.4 I get kernel panics:

   VFS: Cannot open root device "307" or 03:07
   Please append a corect "root=" boot option
   Kernel panic: VFS: Unable to mount root

Just on a lark I decided to go back to a 2.2 series kernel (since
I had gotten flash recognized b4 when using a Mandrake distro)
 Eureka! working flash drive that is seen and formatable with no
troubles. PLEASE have the developers take a closer look at the IDE
drivers for 2.4 as it seems they cause a great deal of grief, at 
least from what I see in searchs on these errors. Usually the problem
is blamed on the user and they are asked hundreds of irritating questions about hardware and configuration.

Regards,

John Sims

Re: 2.4 series IDE troubles

[Miguel A. Bolanos]

Greetings John,
Please read /usr/src/linux-2.4/REPORTING-BUGS

Providing the info requested there, you will help the developers find a
solutions to this sooner.

Thanks

Mike


On Fri, 2003-02-21 at 07:01, j wrote:
> Hello,
> 
> IM not sure this is the way to go about this (obviously) but doing
> searchs at google for the error message I get just leads to the painful realization that there are many questions out there but 
> few answers. Anyhoo when I try adding a compact flash card as an 
> IDE drive to a system running under 2.4 I get kernel panics:
> 
>    VFS: Cannot open root device "307" or 03:07
>    Please append a corect "root=" boot option
>    Kernel panic: VFS: Unable to mount root
> 
> Just on a lark I decided to go back to a 2.2 series kernel (since
> I had gotten flash recognized b4 when using a Mandrake distro)
>  Eureka! working flash drive that is seen and formatable with no
> troubles. PLEASE have the developers take a closer look at the IDE
> drivers for 2.4 as it seems they cause a great deal of grief, at 
> least from what I see in searchs on these errors. Usually the problem
> is blamed on the user and they are asked hundreds of irritating questions about hardware and configuration.
> 
> Regards,
> 
> John Sims
> -
> 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/
> 

Re: 2.4 series IDE troubles

[Alan Cox]

On Fri, 2003-02-21 at 13:01, j wrote:
> least from what I see in searchs on these errors. Usually the problem
> is blamed on the user and they are asked hundreds of irritating questions about hardware and configuration.
> 

And without that information there is no way to fix it. At a first guess 
you've stuck an IDE master and a flash slave via an adapter on the same
cable. 

Re: 2.4 series IDE troubles

[Bill Davidsen]

On 21 Feb 2003, Alan Cox wrote:

> On Fri, 2003-02-21 at 13:01, j wrote:
> > least from what I see in searchs on these errors. Usually the problem
> > is blamed on the user and they are asked hundreds of irritating questions about hardware and configuration.
> > 
> 
> And without that information there is no way to fix it. At a first guess 
> you've stuck an IDE master and a flash slave via an adapter on the same
> cable. 

Didn't he say it worked in 2.2? If that's true then perhaps it should in
2.4 and later.

I've given up a bought a USB flash adaptor, my ISA bus PCMCIA adaptor
hasn't worked for flash since about 2.4.16 or so. Not a complaint, but
there may be issues in that support, I just didn't have time to fight with
the problem.

-- 
bill davidsen <davidsen@tmr.com>
  CTO, TMR Associates, Inc
Doing interesting things with little computers since 1979.

Re: 2.4 series IDE troubles

[Alan Cox]

> And without that information there is no way to fix it. At a first guess 
> > you've stuck an IDE master and a flash slave via an adapter on the same
> > cable. 
> 
> Didn't he say it worked in 2.2? If that's true then perhaps it should in
> 2.4 and later.

Did I say otherwise ? But if he isnt using 2.4.21pre-ac then it wont


> I've given up a bought a USB flash adaptor, my ISA bus PCMCIA adaptor
> hasn't worked for flash since about 2.4.16 or so. Not a complaint, but
> there may be issues in that support, I just didn't have time to fight with
> the problem.

With 2.4.21pre (the firs 2.4 IDE I hacked on seriously) pcmcia flash works on
my test setups, and gets used fairly hard for digital cameras

Re: 2.4 series IDE troubles

[Bill Davidsen]

On 21 Feb 2003, Alan Cox wrote:

> > And without that information there is no way to fix it. At a first guess 
> > > you've stuck an IDE master and a flash slave via an adapter on the same
> > > cable. 
> > 
> > Didn't he say it worked in 2.2? If that's true then perhaps it should in
> > 2.4 and later.
> 
> Did I say otherwise ? But if he isnt using 2.4.21pre-ac then it wont

Will this not migrate into mainline? Or is there some objection to it?

> 
> > I've given up a bought a USB flash adaptor, my ISA bus PCMCIA adaptor
> > hasn't worked for flash since about 2.4.16 or so. Not a complaint, but
> > there may be issues in that support, I just didn't have time to fight with
> > the problem.
> 
> With 2.4.21pre (the firs 2.4 IDE I hacked on seriously) pcmcia flash works on
> my test setups, and gets used fairly hard for digital cameras

I suppose I should stick a media in again and try 2.4.21-pre4-ac5 and
2.5.61-ac1, which are what I have moderately working in other ways. The
digital camera is exactly what drive my use.

-- 
bill davidsen <davidsen@tmr.com>
  CTO, TMR Associates, Inc
Doing interesting things with little computers since 1979.

Re: 2.4 series IDE troubles

[Alan Cox]

On Sat, 2003-02-22 at 19:02, Bill Davidsen wrote:
> On 21 Feb 2003, Alan Cox wrote:
> 
> > > And without that information there is no way to fix it. At a first guess 
> > > > you've stuck an IDE master and a flash slave via an adapter on the same
> > > > cable. 
> > > 
> > > Didn't he say it worked in 2.2? If that's true then perhaps it should in
> > > 2.4 and later.
> > 
> > Did I say otherwise ? But if he isnt using 2.4.21pre-ac then it wont
> 
> Will this not migrate into mainline? Or is there some objection to it?

It should be in mainline very soon. Its one of a set of patches Im waiting for
the next Marcelo release to resynchronize.

Re: 2.4 series IDE troubles

[Dag Bakke]

On 21 Feb 2003 18:59:40 +0000, Alan Cox wrote:
> With 2.4.21pre (the firs 2.4 IDE I hacked on seriously) pcmcia flash 
> works on my test setups, and gets used fairly hard for digital cameras

Anyone tried booting a  recent Toshiba Laptop from PCMCIA?
I have, and it doesn't work. Not that this necessarily has anything to do
with the IDE code. I have tried both recent -ac and vanilla.     

In short, if I load the kernel from PCMCIA, the CardBus slots disappear
from the PCI bus. -> no root device -> boom!

If I load the kernel from any other medium (haven't tried PCMCIA netboot),
the CardBus slots are present.

Is this a Toshiba oddity for which there isn't a PCI quirk yet?
Any suggestions for debugging?

Dag B.

Re: 2.4 series IDE troubles

[Alan Cox]

On Fri, 2003-02-21 at 21:13, Dag Bakke wrote:
> On 21 Feb 2003 18:59:40 +0000, Alan Cox wrote:
> > With 2.4.21pre (the firs 2.4 IDE I hacked on seriously) pcmcia flash 
> > works on my test setups, and gets used fairly hard for digital cameras
> 
> Anyone tried booting a  recent Toshiba Laptop from PCMCIA?
> I have, and it doesn't work. Not that this necessarily has anything to do
> with the IDE code. I have tried both recent -ac and vanilla.     
> 
> In short, if I load the kernel from PCMCIA, the CardBus slots disappear
> from the PCI bus. -> no root device -> boom!

Some toshiba stuff seems to hide the cardbus/pcmcia and fake the attached
CD-ROM used for booting as a native IDE device. I assume this is for 
windows 95/98 installation. Vaio's do something similar but do not hide
the cardbus. When the cardbus is initialised on the vaio the magic IDE
mapping vanishes

Re: 2.4 series IDE troubles

[Dag Bakke]

On Fri, Feb 21, 2003 at 11:52:21PM +0000, Alan Cox wrote:
> On Fri, 2003-02-21 at 21:13, Dag Bakke wrote:
> > On 21 Feb 2003 18:59:40 +0000, Alan Cox wrote:
> > > With 2.4.21pre (the firs 2.4 IDE I hacked on seriously) pcmcia flash 
> > > works on my test setups, and gets used fairly hard for digital cameras
> > 
> > Anyone tried booting a  recent Toshiba Laptop from PCMCIA?
> > I have, and it doesn't work. Not that this necessarily has anything to do
> > with the IDE code. I have tried both recent -ac and vanilla.     
> > 
> > In short, if I load the kernel from PCMCIA, the CardBus slots disappear
> > from the PCI bus. -> no root device -> boom!
> 
> Some toshiba stuff seems to hide the cardbus/pcmcia and fake the attached
> CD-ROM used for booting as a native IDE device. I assume this is for 
> windows 95/98 installation. Vaio's do something similar but do not hide
> the cardbus. When the cardbus is initialised on the vaio the magic IDE
> mapping vanishes
> 

I boot from CF in a PCMCIA sleeve. Sorry for not being clear about that.
But in any case:

even if these toshibas you mention above "fake the attached CD-ROM used for 
booting as a native IDE device", that IDE device has to be visible
after the operating system has initialized? Pretty hard to install anything
from CD-ROM otherwise...?

The way it looks to me, this is more a question about the PCI subsystem
in the toshiba.  Of course, I could be totally off base.

Anyone got any great ideas about how to beat the Toshiba into submission
and make it admit that it has the cardbus slots it just booted from?

Perhaps some poking in the magic registers of these animals?
(http://www.buzzard.org.uk/toshiba/docs.html)
Not sure if Jonathan still maintains the Toshiba utilities/docs, though.

Dag B