Re: 2.6.15-mm2

[Reuben Farrelly]

On 8/01/2006 10:31 a.m., Andrew Morton wrote:
> Reuben Farrelly <reuben-lkml@xxxxxxxx> wrote:
> > ...
> > QLogic Fibre Channel HBA Driver
> > ahci: probe of 0000:00:1f.2 failed with error -12
>
> It's odd that the ahci driver returned -EBUSY.  Maybe this is due to "we
> have legacy mode, but all ports are unavailable" in ata_pci_init_one().

I've now removed this driver from my .config via menuconfig, I certainly don't 
have the hardware and have no idea whatsoever how it came to be built in. 
Although I guess it shouldn't be blowing up even if that is the case?

> > ata1: SATA max UDMA/133 cmd 0x0 ctl 0x2 bmdma 0x0 irq 0
> > ata2: SATA max UDMA/133 cmd 0x0 ctl 0x2 bmdma 0x8 irq 0
> > Unable to handle kernel NULL pointer dereference at virtual address 00000000
> >   printing eip:
> > c0234702
> > *pde = 00000000
> > Oops: 0000 [#1]
> > SMP
> > last sysfs file:
> > Modules linked in:
> > CPU:    1
> > EIP:    0060:[<c0234702>]    Not tainted VLI
> > EFLAGS: 00010206   (2.6.15-mm2)
> > EIP is at make_class_name+0x28/0x8d
> > eax: 00000000   ebx: ffffffff   ecx: ffffffff   edx: c19d9224
> > esi: 00000009   edi: 00000000   ebp: 00000000   esp: c1921d9c
> > ds: 007b   es: 007b   ss: 0068
> > Process swapper (pid: 1, threadinfo=c1921000 task=c1920a70)
> > Stack: <0>c19d9224 c03a9158 c19d9224 c03a9158 c03a9160 c0234925 c03a90e0 00000000
> >         <0>00000246 c19d9224 c19d9000 c19d9030 00000002 c02349db c19d90e4 c0253218
> >         <0>c19d92c0 00000000 00000000 c0276693 00000000 c0279391 c035749f c1961640
> > Call Trace:
> >   [<c0234925>] class_device_del+0x9f/0x14d
> >   [<c02349db>] class_device_unregister+0x8/0x10
> >   [<c0253218>] scsi_remove_host+0xb8/0xf8
> >   [<c0276693>] ata_host_remove+0xe/0x18
> >   [<c0279391>] ata_device_add+0x2d3/0xb99
> >   [<c02b6fb0>] pci_mmcfg_write+0xd3/0x103
> >   [<c01eb713>] pci_bus_write_config_byte+0x4e/0x58
> >   [<c02b67d3>] pcibios_set_master+0x74/0x8c
> >   [<c027a2e5>] ata_pci_init_one+0x32c/0x38e
> >   [<c01eb7ea>] pci_bus_read_config_word+0x62/0x6c
> >   [<c01ef8bd>] pci_get_subsys+0x6c/0xe0
> >   [<c027e334>] piix_init_one+0x18e/0x33a
> >   [<c01ef259>] pci_device_probe+0x40/0x5b
> >   [<c0233ed7>] driver_probe_device+0x35/0x98
> >   [<c0234038>] __driver_attach+0x8a/0x8c
> >   [<c02339a7>] bus_for_each_dev+0x39/0x57
> >   [<c0233e4c>] driver_attach+0x16/0x1a
> >   [<c0233fae>] __driver_attach+0x0/0x8c
> >   [<c023365b>] bus_add_driver+0x6f/0x126
> >   [<c01ef3f1>] __pci_register_driver+0x7d/0xac
> >   [<c04023e9>] piix_init+0xc/0x1e
> >   [<c01003c8>] init+0xff/0x324
> >   [<c01002c9>] init+0x0/0x324
> >   [<c0100d35>] kernel_thread_helper+0x5/0xb
> > Code: 89 c8 c3 55 57 56 53 83 ec 04 89 04 24 89 c2 8b 40 48 8b 38 31 ed bb ff ff 
> > ff ff 89 d9 89 e8 f2 ae f7 d1 49 89 ce 8b 7a 08 89 d9 <f2> ae f7 d1 49 89 ca 8d 
> > 4e 02 8d 04 0a ba d0 00 00 00 e8 22 cf
>
> ata_device_add() has given up, has called ata_host_remove() and then we
> presumably oopsed over incompletely initialised class stuff.  It's likely
> that this oops is a second bug - a consequence of the -EBUSY.
>
> > 2. Notice above how the sky2 driver is being bailed out:
> >
> > ACPI: PCI Interrupt 0000:04:00.0[A] -> GSI 17 (level, low) -> IRQ 177
> > sky2 Cannot find PowerManagement capability, aborting.
> > sky2: probe of 0000:04:00.0 failed with error -5
> >
> > This has happened a number of times in the last few days, and I suspect is 
> > unrelated to the oops that followed above.
> >
> > This driver worked fine in 2.6.15-rc5-mm3, and seems to work OK when built as a 
> > module.  But most of the time (not all the time) it doesn't like being 
> > statically built in and fails with the above error.  Changes to this driver have 
> > been fairly small lately so I'm not sure if it's the driver or something else 
> > like ACPI that is the root cause.
>
> Could be acpi, yes.
>
> Parenthetically, I wouldn't have thought that this error should be fatal
> for the driver.

lspci -vv shows that when the driver fails we see this:

Status: Cap+ 66MHz- UDF- FastB2B- ParErr- DEVSEL=fast >TAbort- <TAbort- <MAbort- 
>SERR+ <PERR-

and when it works we see this:

Status: Cap+ 66MHz- UDF- FastB2B- ParErr- DEVSEL=fast >TAbort- <TAbort- <MAbort- 
>SERR- <PERR-

I'm not sure if that's a consequence of the fail, or the cause of it.

> > 3.  The boot up process with -mm2 was pretty lengthy, I had two periods of time 
> > when the whole system just came to a crawl, first time was when starting cups, 
> > and it came back to life and continued booting about 30s later.  Next when 
> > starting hpijs it didn't come to life at all and I had to reboot.  No output to 
> > the console for either, unfortunately.
>
> Don't know, sorry.  But this kernel had oopsed, hadn't it?

I reloaded multiple times, the oopsing only occurred till I did a full cold 
boot, and then it came right (but until then I had the oops twice in a row 
across a warm reboot).

If I have time to play later on today I'll see if I can get more info.

reuben


-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/