Re: Linux hangs during IDE initialization at boot for 30 sec

Re: Linux hangs during IDE initialization at boot for 30 sec

[Benjamin Herrenschmidt]

On Tue, 2005-02-01 at 20:22 +0000, Richard Hughes wrote:
> On Tue, 01 Feb 2005 12:57:33 +0100, Michael Brade wrote:
> 
> > Hi,
> > 
> > since at least kernel 2.6.9 I'm having a problem booting linux - it hangs 
> > after this
> > 
> > Probing IDE interface ide0...
> > hda: HITACHI_DK23DA-30, ATA DISK drive
> > elevator: using anticipatory as default io scheduler
> > ide0 at 0x1f0-0x1f7,0x3f6 on irq 14
> > Probing IDE interface ide1...
> > hdc: TOSHIBA DVD-ROM SD-R2212, ATAPI CD/DVD-ROM drive
> > ide1 at 0x170-0x177,0x376 on irq 15
> > 
> > After about 30 seconds everything continues fine with
> > 
> > hda: max request size: 128KiB
> > 
> > I found additional lines in the log just before the line above:
> 
> Same here on 2.6.11-rc2-bk3 using a *Toshiba* Satellite Pro A10.
> 
> messages can be found here:
> 
> http://hughsie.no-ip.com/write/kernel/messages

This looks like bogus HW, or bogus list of IDE interfaces ...

The IDE layer waits up to 30 seconds for a device to drop it's busy bit,
which is necessary for some drives that aren't fully initialized yet.

I suspect in your case, it's reading "ff", which indicates either that
there is no hardware where the kernel tries to probe, or that there is
bogus IDE interfaces which don't properly have the D7 line pulled low so
that BUSY appears not set in absence of a drive.

I'm not sure how the list of intefaces is probed on this machine, that's
probably where the problem is.

Ben.

Re: Linux hangs during IDE initialization at boot for 30 sec

[Alan Cox]

On Maw, 2005-02-01 at 23:18, Benjamin Herrenschmidt wrote:
> On Tue, 2005-02-01 at 20:22 +0000, Richard Hughes wrote:
> > On Tue, 01 Feb 2005 12:57:33 +0100, Michael Brade wrote:
> I suspect in your case, it's reading "ff", which indicates either that
> there is no hardware where the kernel tries to probe, or that there is
> bogus IDE interfaces which don't properly have the D7 line pulled low so
> that BUSY appears not set in absence of a drive.
> 
> I'm not sure how the list of intefaces is probed on this machine, that's
> probably where the problem is.

Known, fixed

There is a patch that allows probing for ISA ide4,5,6 etc only if there
is
no PCI bus

Re: Linux hangs during IDE initialization at boot for 30 sec

[Rolf Offermanns]

On Wednesday 02 February 2005 00:18, you wrote:

>
> The IDE layer waits up to 30 seconds for a device to drop it's busy bit,
> which is necessary for some drives that aren't fully initialized yet.
>
> I suspect in your case, it's reading "ff", which indicates either that
> there is no hardware where the kernel tries to probe, or that there is
> bogus IDE interfaces which don't properly have the D7 line pulled low so
> that BUSY appears not set in absence of a drive.
>
I have the same kind of bogus hardware here. What would be the right place to 
workaround this? (unfortunately it's not possible to fix/change the 
hardware).

-Rolf
-- 
Rolf Offermanns <roffermanns@sysgo.com>
SYSGO AG     Tel.: +49-6136-9948-0
Am Pfaffenstein 14   Fax: +49-6136-9948-10
55270 Klein-Winternheim  http://www.sysgo.com

Re: Linux hangs during IDE initialization at boot for 30 sec

[Benjamin Herrenschmidt]

On Thu, 2005-02-03 at 13:58 +0100, Rolf Offermanns wrote:
> On Wednesday 02 February 2005 00:18, you wrote:
> 
> >
> > The IDE layer waits up to 30 seconds for a device to drop it's busy bit,
> > which is necessary for some drives that aren't fully initialized yet.
> >
> > I suspect in your case, it's reading "ff", which indicates either that
> > there is no hardware where the kernel tries to probe, or that there is
> > bogus IDE interfaces which don't properly have the D7 line pulled low so
> > that BUSY appears not set in absence of a drive.
> >
> I have the same kind of bogus hardware here. What would be the right place to 
> workaround this? (unfortunately it's not possible to fix/change the 
> hardware).

Do not register interfaces that you know aren't populated...

One thing is, the wait code does explicitely check for a value of 0xff
and exists in this case (assuming this means no HW at all is wired
there). So for some reason, you interface is returning something else,
it would be interesting to "see" what value you are getting.

Ben.

Re: Linux hangs during IDE initialization at boot for 30 sec

[Rolf Offermanns]

On Thursday 03 February 2005 23:17, Benjamin Herrenschmidt wrote:
>
> Do not register interfaces that you know aren't populated...
>
> One thing is, the wait code does explicitely check for a value of 0xff
> and exists in this case (assuming this means no HW at all is wired
> there). So for some reason, you interface is returning something else,
> it would be interesting to "see" what value you are getting.
It's 0xff like you guessed in the first reply. As soon, as I connect a drive 
(no need to turn it on actually), everything is fine.

-Rolf
-- 
Rolf Offermanns <roffermanns@sysgo.com>
SYSGO AG     Tel.: +49-6136-9948-0
Am Pfaffenstein 14   Fax: +49-6136-9948-10
55270 Klein-Winternheim  http://www.sysgo.com

Re: Linux hangs during IDE initialization at boot for 30 sec

[Benjamin Herrenschmidt]

On Fri, 2005-02-04 at 11:22 +0100, Rolf Offermanns wrote:
> On Thursday 03 February 2005 23:17, Benjamin Herrenschmidt wrote:
> >
> > Do not register interfaces that you know aren't populated...
> >
> > One thing is, the wait code does explicitely check for a value of 0xff
> > and exists in this case (assuming this means no HW at all is wired
> > there). So for some reason, you interface is returning something else,
> > it would be interesting to "see" what value you are getting.

> It's 0xff like you guessed in the first reply. As soon, as I connect a drive 
> (no need to turn it on actually), everything is fine.

Well, that's strange then, since the code is explicitely returning
-ENODEV when it encounters that, maybe the caller isn't properly testing
for the error, I'll have to double check. (Can't today)

Ben.