2003-03-14 21:02:52

by Daniel Egger

[permalink] [raw]
Subject: 2.5.64-ac3: Crash in ide_init_queue

Hija,

my new motherboard with KT400 chipset oopses in ide_init_queue on bootup
with the follow output:

Oops: 0000
CPU: 0
EIP: 0060:[<00000000>] Not tainted
EFLAGS: 00010282
EIP is at 0x0
eax: c044af44 ebx: c044aff0 ecx: 00000000 edx: 00000000
esi: c044b000 edi: c044af44 ebp: c151dee8 esp: c151ded0
ds: 007b es: 007b ss: 0068
Process swapper (pid: 1, threadinfo=c151c000 task=dff8e040)
Stack: c0263dff c044aff0 00000001 c03cd660 c044aff0 c1527ef4 c151df30 c0264006
c044aff0 c0262b10
Call Trace:
[<c0263dff>] ide_init_queue+0x9f/0xb0
[<c0264006>] init_irq+0x1f6/0x3d0
[<c0262b10>] ide_intr+0x0/0x180
[<c0264658>] hwif_init+0xe8/0x290
[<c0264924>] ideprobe_init+0xf4/0x110
[<c02209d1>] driver_register+0x31/0x40
[<c010507d>] init+0x3d/0x160
[<c0105040>] init+0x0/0x160
[<c010740d>] kernel_thread_helper+0x5/0x18

Code: Bad EIP value.

The system doesn't have a harddrive, both channels are enabled,
the second bus has a TOSHIBA Combo DVD/CD-R SD-R1002 conntected
as master device. Disabling the secondary channel channel fixes
the problem.

Should I bugzill it?

--
Servus,
Daniel


Attachments:
signature.asc (189.00 B)
Dies ist ein digital signierter Nachrichtenteil

2003-03-14 21:14:09

by Jens Axboe

[permalink] [raw]
Subject: Re: 2.5.64-ac3: Crash in ide_init_queue

On Fri, Mar 14 2003, Daniel Egger wrote:
> Hija,
>
> my new motherboard with KT400 chipset oopses in ide_init_queue on bootup
> with the follow output:

using ide tcq?

--
Jens Axboe

2003-03-15 15:14:41

by Daniel Egger

[permalink] [raw]
Subject: Re: 2.5.64-ac3: Crash in ide_init_queue

Am Fre, 2003-03-14 um 22.25 schrieb Jens Axboe:

> using ide tcq?

It's compiled into the kernel but unused since there's no harddrive in
the machine. I'll remove it from the config and retry.

--
Servus,
Daniel


Attachments:
signature.asc (189.00 B)
Dies ist ein digital signierter Nachrichtenteil

2003-03-15 15:22:34

by Daniel Egger

[permalink] [raw]
Subject: Re: 2.5.64-ac3: Crash in ide_init_queue

Am Sam, 2003-03-15 um 16.25 schrieb Daniel Egger:

> > using ide tcq?

> It's compiled into the kernel but unused since there's no harddrive in
> the machine. I'll remove it from the config and retry.

Nope, same problem without tcq.

--
Servus,
Daniel


Attachments:
signature.asc (189.00 B)
Dies ist ein digital signierter Nachrichtenteil

2003-03-15 16:20:48

by Jens Axboe

[permalink] [raw]
Subject: Re: 2.5.64-ac3: Crash in ide_init_queue

On Sat, Mar 15 2003, Daniel Egger wrote:
> Am Sam, 2003-03-15 um 16.25 schrieb Daniel Egger:
>
> > > using ide tcq?
>
> > It's compiled into the kernel but unused since there's no harddrive in
> > the machine. I'll remove it from the config and retry.
>
> Nope, same problem without tcq.

Please double check, TCQ is the only way that ide_init_queue() would end
up with NULL EIP.

You could make the ide_dma_queued_on() conditional, ala:

#ifdef CONFIG_BLK_DEV_IDE_TCQ_DEFAULT
if (HWIF(drive)->ide_dma_queued_on)
HWIF(drive)->ide_dma_queued_on(drive);
#endif


--
Jens Axboe

2003-03-15 17:21:06

by Daniel Egger

[permalink] [raw]
Subject: Re: 2.5.64-ac3: Crash in ide_init_queue

Am Sam, 2003-03-15 um 17.31 schrieb Jens Axboe:

> Please double check, TCQ is the only way that ide_init_queue() would end
> up with NULL EIP.

Yep this was a nobrainer of mine. As stated in a followup mail, this
problem went away and a different IDE problem stepped up.

--
Servus,
Daniel


Attachments:
signature.asc (189.00 B)
Dies ist ein digital signierter Nachrichtenteil