irqpoll causing some breakage?

irqpoll causing some breakage?

[Daniel Drake]

Hi,

I recently added the irqpoll patch to Gentoo's 2.6.12 kernels, using this patch:

http://dev.gentoo.org/~dsd/genpatches/trunk/2.6.12/2700_irqpoll.patch

Since then, I've had a few reports of minor breakage, but this is the first 
one I've been able to get full info on.

Hans-Christian Armingeon (on CC) owns a a combined USB keyboard-mouse, which 
is broken when the irqpoll patch is applied.

input: USB HID v1.00 Keyboard [Cherry GmbH Cherry Slim Line Trackball 
Keyboard] on usb-0000:00:10.0-1
input: USB HID v1.00 Mouse [Cherry GmbH Cherry Slim Line Trackball Keyboard] 
on usb-0000:00:10.0-1

After the irqpoll patch has been applied, the mouse does not work (the 
keyboard works fine..!). This is without the irqpoll/irqfixup parameters, 
although adding them does not help either. No errors appear, as far as I can see.

The problem also exists in an unpatched 2.6.13_rc6.

dmesg here: https://bugs.gentoo.org/attachment.cgi?id=65470
full bug: https://bugs.gentoo.org/show_bug.cgi?id=101463

I realise that this is probably not enough information to make any sense out 
of! Please let me know how we can help further.

Thanks,
Daniel

Re: irqpoll causing some breakage?

[Hans-Christian Armingeon]

Am Dienstag 09 August 2005 02:48 schrieb Daniel Drake:
> Hi,
> 
> I recently added the irqpoll patch to Gentoo's 2.6.12 kernels, using this patch:
> 
> http://dev.gentoo.org/~dsd/genpatches/trunk/2.6.12/2700_irqpoll.patch
> 
> Since then, I've had a few reports of minor breakage, but this is the first 
> one I've been able to get full info on.
> 
> Hans-Christian Armingeon (on CC) owns a a combined USB keyboard-mouse, which 
> is broken when the irqpoll patch is applied.
> 
> input: USB HID v1.00 Keyboard [Cherry GmbH Cherry Slim Line Trackball 
> Keyboard] on usb-0000:00:10.0-1
> input: USB HID v1.00 Mouse [Cherry GmbH Cherry Slim Line Trackball Keyboard] 
> on usb-0000:00:10.0-1
> 
> After the irqpoll patch has been applied, the mouse does not work (the 
> keyboard works fine..!). This is without the irqpoll/irqfixup parameters, 
> although adding them does not help either. No errors appear, as far as I can see.

I have two other mice, and they don't work, too.

My usb Setup_

PortA --- Keyboard with integrated mouse
PortB --- Hub
HubPortA --- Mouse
HubPortA --- Trackball

The mice don't work, when I plug them directly into Port A or B .

The keyboard works ervery time.

> 
> The problem also exists in an unpatched 2.6.13_rc6.
> 
> dmesg here: https://bugs.gentoo.org/attachment.cgi?id=65470
> full bug: https://bugs.gentoo.org/show_bug.cgi?id=101463
> 
> I realise that this is probably not enough information to make any sense out 
> of! Please let me know how we can help further.

Me too.

> 
> Thanks,
> Daniel
> 
> 

Thanks in advance,

Johnny

Re: irqpoll causing some breakage?

[Alan Cox]

On Maw, 2005-08-09 at 10:57 +0200, Hans-Christian Armingeon wrote:
> PortA --- Keyboard with integrated mouse
> PortB --- Hub
> HubPortA --- Mouse
> HubPortA --- Trackball
> 
> The mice don't work, when I plug them directly into Port A or B .
> 
> The keyboard works ervery time.

Then its not irqpoll related because the keyboard and mouse are using
the same IRQ in some of the cases you describe.

Re: irqpoll causing some breakage?

[Alan Cox]

On Maw, 2005-08-09 at 01:48 +0100, Daniel Drake wrote:
> After the irqpoll patch has been applied, the mouse does not work (the 
> keyboard works fine..!). This is without the irqpoll/irqfixup parameters, 
> although adding them does not help either. No errors appear, as far as I can see.

Without the parameters it has exactly zero effect on the operation of
the kernel, the algorithms and the behaviour. So something odd is afoot
if its causing gentoo breakages. Do you see this using a reasonably
trustworthy compiler suite as we've not seen it on Fedora ?

If you do does the box log any stuck IRQ's during boot without the
patch 

Re: irqpoll causing some breakage?

[Daniel Drake]

Alan Cox wrote:
> Without the parameters it has exactly zero effect on the operation of
> the kernel, the algorithms and the behaviour. So something odd is afoot
> if its causing gentoo breakages.

Thats what I thought, yet it seems to be the difference between mouse and no 
mouse in this case.

Strange. We'll try a different compiler.

Thanks.
Daniel

Re: irqpoll causing some breakage?

[Alan Cox]

On Maw, 2005-08-09 at 18:12 +0100, Daniel Drake wrote:
> Alan Cox wrote:
> > Without the parameters it has exactly zero effect on the operation of
> > the kernel, the algorithms and the behaviour. So something odd is afoot
> > if its causing gentoo breakages.
> 
> Thats what I thought, yet it seems to be the difference between mouse and no 
> mouse in this case.
> 
> Strange. We'll try a different compiler.

What do the other reports look like ?

Re: irqpoll causing some breakage?

[Daniel Drake]

Alan Cox wrote:
> What do the other reports look like ?
> 

Here's one:

http://forums.gentoo.org/viewtopic-t-361718-highlight-irqpoll.html

This possibly suggests that the irqpoll patch actually caused a "nobody cared" 
which wasn't there previously. (Now that I have looked closer at the patch, I 
realise how unlikely this is, but this was my reaction at the time!)

I had another report like that by email (another network adapter, "nobody 
cared" message appeared which wasn't there before). The revision difference 
was even smaller and again irqpoll was my suspect. But he never responded to 
my request to test reverting the irqpoll patch and file a bug. I'll dig up the 
email and send a reminder.

Given that I haven't been able to pinpoint irqpoll as the cause of these, I 
don't think you should worry about them at this stage. The only interesting 
one at the moment is the keyboard/mouse thing...

Daniel