Messages in this thread | | | Date | Fri, 11 Nov 2005 17:47:32 -0800 | From | Andrew Morton <> | Subject | Re: 2.6.14-mm1 |
| |
Michal Piotrowski <michal.k.k.piotrowski@gmail.com> wrote: > > On 12/11/05, Andrew Morton <akpm@osdl.org> wrote: > > Crap. This is one of those crashes where the sound people, the PCI people, > > the ACPI people and the PM people all earnestly hope that it's the other > > guy's bug and you and I are left with a mess on our hands. > > > > Possibly the card didn't get powered up. I know there's a way to get all > > those snd_printk()'s to print something, but I never have much success > > finding the right value for the right /proc file to make it happen. > > > > So can you add this please? > > > > --- devel/sound/pci/intel8x0.c~a 2005-11-11 16:47:00.000000000 -0800 > > +++ devel-akpm/sound/pci/intel8x0.c 2005-11-11 16:48:13.000000000 -0800 > > @@ -779,6 +779,7 @@ static irqreturn_t snd_intel8x0_interrup > > unsigned int i; > > > > status = igetdword(chip, chip->int_sta_reg); > > + printk("status: 0x%8x\n", status); > > if (status == 0xffffffff) /* we are not yet resumed */ > > return IRQ_NONE; > > > > _ > > > > and let us know what it says? > > > > Also, it would be useful if you could disable the sound driver in config > > and see if you can get it booted. If so, then generate the `dmesg -s > > 1000000' output for good and bad kernels and let's see what they look like. > > > > Thanks. > > > > I will try to reproduce it, but according to > http://klive.cpushare.com/2.6.14-mm1/?order_by=kernel_group&where_machine=all&branch=mm&scheduler=all&smp=all&live=all&ip=all > I have been using 2.6.14-mm1 about 48 hours with 12 reboots and this > problem appeared only once.
ah-hah. This sounds rather like Reuben Farrelly's e100 failure - something seems to be making PCI initialisation go stupid if CONFIG_PREEMPT is enabled.
It would be interesting if you could reboot sufficiently often to work out whether disabling CONFIG_PREEMPT fixes things up.
- 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/
| |