siimage, 2.5.74 and irq 19: nobody cared!

siimage, 2.5.74 and irq 19: nobody cared!

[Miquel van Smoorenburg]

I was running 2.5.72-mm2 on our transit usenet news server
(700 GB in/day and 1 TB out/day) which ran just fine, until I
had some ext3 corruption on the /news partition. I remember
having seen something about this in the -mm changelogs.

So I tried 2.5.74 and 2.5.74-mm2, but with those kernels the
siimage.c driver doesn't work. The card is detected, but a bit
later in the boot process its IRQ is disabled and it won't work.
Log is below. As said, it worked fine with 2.5.72-mm2 (OK, needed
to enabled UDMA with hdparm, but other than that, no problems):

Jul  8 12:23:09 quantum kernel: irq 19: nobody cared!
Jul  8 12:23:09 quantum kernel: Call Trace:
Jul  8 12:23:09 quantum kernel:  [__report_bad_irq+50/144] __report_bad_irq+0x32/0x90
Jul  8 12:23:09 quantum kernel:  [note_interrupt+80/120] note_interrupt+0x50/0x78
Jul  8 12:23:09 quantum kernel:  [do_IRQ+179/280] do_IRQ+0xb3/0x118
Jul  8 12:23:09 quantum kernel:  [default_idle+0/52] default_idle+0x0/0x34
Jul  8 12:23:09 quantum kernel:  [rest_init+0/72] _stext+0x0/0x48
Jul  8 12:23:09 quantum kernel:  [common_interrupt+24/32] common_interrupt+0x18/0x20
Jul  8 12:23:09 quantum kernel:  [default_idle+0/52] default_idle+0x0/0x34
Jul  8 12:23:09 quantum kernel:  [rest_init+0/72] _stext+0x0/0x48
Jul  8 12:23:09 quantum kernel:  [default_idle+41/52] default_idle+0x29/0x34
Jul  8 12:23:09 quantum kernel:  [cpu_idle+55/72] cpu_idle+0x37/0x48
Jul  8 12:23:09 quantum kernel:  [rest_init+69/72] _stext+0x45/0x48
Jul  8 12:23:09 quantum kernel:  [start_kernel+322/328] start_kernel+0x142/0x148Jul  8 12:23:09 quantum kernel:
Jul  8 12:23:09 quantum kernel: handlers:
Jul  8 12:23:09 quantum kernel: [ide_intr+0/352] (ide_intr+0x0/0x160)
Jul  8 12:23:09 quantum kernel: [ide_intr+0/352] (ide_intr+0x0/0x160)
Jul  8 12:23:09 quantum kernel: Disabling IRQ #19

Mike.

Re: siimage, 2.5.74 and irq 19: nobody cared!

[Bartlomiej Zolnierkiewicz]

Hi,

Please send dmesg from 2.5.62-mm2 and 'lspci -vvv' output.
--
Bartlomiej

On Tue, 8 Jul 2003, Miquel van Smoorenburg wrote:

> I was running 2.5.72-mm2 on our transit usenet news server
> (700 GB in/day and 1 TB out/day) which ran just fine, until I
> had some ext3 corruption on the /news partition. I remember
> having seen something about this in the -mm changelogs.
>
> So I tried 2.5.74 and 2.5.74-mm2, but with those kernels the
> siimage.c driver doesn't work. The card is detected, but a bit
> later in the boot process its IRQ is disabled and it won't work.
> Log is below. As said, it worked fine with 2.5.72-mm2 (OK, needed
> to enabled UDMA with hdparm, but other than that, no problems):
>
> Jul  8 12:23:09 quantum kernel: irq 19: nobody cared!
> Jul  8 12:23:09 quantum kernel: Call Trace:
> Jul  8 12:23:09 quantum kernel:  [__report_bad_irq+50/144] __report_bad_irq+0x32/0x90
> Jul  8 12:23:09 quantum kernel:  [note_interrupt+80/120] note_interrupt+0x50/0x78
> Jul  8 12:23:09 quantum kernel:  [do_IRQ+179/280] do_IRQ+0xb3/0x118
> Jul  8 12:23:09 quantum kernel:  [default_idle+0/52] default_idle+0x0/0x34
> Jul  8 12:23:09 quantum kernel:  [rest_init+0/72] _stext+0x0/0x48
> Jul  8 12:23:09 quantum kernel:  [common_interrupt+24/32] common_interrupt+0x18/0x20
> Jul  8 12:23:09 quantum kernel:  [default_idle+0/52] default_idle+0x0/0x34
> Jul  8 12:23:09 quantum kernel:  [rest_init+0/72] _stext+0x0/0x48
> Jul  8 12:23:09 quantum kernel:  [default_idle+41/52] default_idle+0x29/0x34
> Jul  8 12:23:09 quantum kernel:  [cpu_idle+55/72] cpu_idle+0x37/0x48
> Jul  8 12:23:09 quantum kernel:  [rest_init+69/72] _stext+0x45/0x48
> Jul  8 12:23:09 quantum kernel:  [start_kernel+322/328] start_kernel+0x142/0x148Jul  8 12:23:09 quantum kernel:
> Jul  8 12:23:09 quantum kernel: handlers:
> Jul  8 12:23:09 quantum kernel: [ide_intr+0/352] (ide_intr+0x0/0x160)
> Jul  8 12:23:09 quantum kernel: [ide_intr+0/352] (ide_intr+0x0/0x160)
> Jul  8 12:23:09 quantum kernel: Disabling IRQ #19
>
> Mike.

Re: siimage, 2.5.74 and irq 19: nobody cared!

[Miquel van Smoorenburg]

On 2003.07.08 14:16, Bartlomiej Zolnierkiewicz wrote:
> On Tue, 8 Jul 2003, Miquel van Smoorenburg wrote:
> 
> > I was running 2.5.72-mm2 on our transit usenet news server
> > (700 GB in/day and 1 TB out/day) which ran just fine, until I
> > had some ext3 corruption on the /news partition. I remember
> > having seen something about this in the -mm changelogs.
> >
> > So I tried 2.5.74 and 2.5.74-mm2, but with those kernels the
> > siimage.c driver doesn't work. The card is detected, but a bit
> > later in the boot process its IRQ is disabled and it won't work.
> > Log is below. As said, it worked fine with 2.5.72-mm2 (OK, needed
> > to enabled UDMA with hdparm, but other than that, no problems):
> 
> Hi,
> 
> Please send dmesg from 2.5.62-mm2 and 'lspci -vvv' output.

You mean dmesg and lspci -vvv output of both 2.5.72-mm2 and 2.5.74 ?
Will do, as soon as I have a chance to play with the system again.
It's a newsfeeder, a few minutes downtime is not a problem, but I
had some serious problems and it was down for 2 hours - now it needs
several hours to "catch up" before I can take it down again.

Mike.

Re: siimage, 2.5.74 and irq 19: nobody cared!

[Bartlomiej Zolnierkiewicz]

On Tue, 8 Jul 2003, Miquel van Smoorenburg wrote:

> > Hi,
> >
> > Please send dmesg from 2.5.62-mm2 and 'lspci -vvv' output.
>
> You mean dmesg and lspci -vvv output of both 2.5.72-mm2 and 2.5.74 ?
> Will do, as soon as I have a chance to play with the system again.
> It's a newsfeeder, a few minutes downtime is not a problem, but I
> had some serious problems and it was down for 2 hours - now it needs
> several hours to "catch up" before I can take it down again.
>
> Mike.

lspci -vvv from any kernel version.
dmesg from 2.5.72-mm2 and from 2.5.74.

You can extract dmesgs from logs if you still have them...
--
Bartlomiej

Re: siimage, 2.5.74 and irq 19: nobody cared!

[Linus Torvalds]

Miquel van Smoorenburg wrote:
>
> I was running 2.5.72-mm2 on our transit usenet news server
> (700 GB in/day and 1 TB out/day) which ran just fine, until I
> had some ext3 corruption on the /news partition. I remember
> having seen something about this in the -mm changelogs.
> 
> So I tried 2.5.74 and 2.5.74-mm2, but with those kernels the
> siimage.c driver doesn't work. The card is detected, but a bit
> later in the boot process its IRQ is disabled and it won't work.

Ok. Can you send me the "lspci -vxx" output for your IDE chip?

The most likely reason for the breakage is that the siimage thing claims it
isn't a proper IDE storage device in legacy mode, and that means that newer
kernels won't try to probe for interrupts: they will just use the PCI
interrupt directly. That helps on machines with shared interrupts where
probing really doesn't work that well, but it can cause problems if the
PCI IDE controller is confused (and tries to implement a legacy IDE device,
but does it wrong).

If this is indeed the problem, then you could try fixing it by adding these
two lines to the top of  init_chipset_siimage():

        /* Mark it as a IDE device in legacy mode! */
        dev->class = (PCI_CLASS_STORAGE_IDE << 8) | 0;

which just tells the IDE layer that it's not a regular PCI device and might
be using the legacy ISA interrupts - so that the code will know to probe
for them.

                Linus