Prev: vc: Add support for hiding the cursor when creating VTs
Next: [PATCH] forcedeth: mac address fix
From: Matthew Garrett on 13 Nov 2009 21:40
http://www.engadget.com/2009/10/31/iphone-and-windows-7-dont-play-nice-intel-p55-chipset-to-blame/
and similar stories elsewhere - some systems seem to have received BIOS
updates that deal with the issue. It'd be nice to know what those
updates do.

--
Matthew Garrett | mjg59(a)srcf.ucam.org
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo(a)vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
From: Alexander Holler on 13 Nov 2009 22:00
Hello,

Am 14.11.2009 00:35, schrieb Alexander Holler:
> Am 14.11.2009 00:30, schrieb Oliver Neukum:
>> Am Freitag, 13. November 2009 23:30:14 schrieb Alexander Holler:
>>> (And here comes the problem)
>>>
>>> Nov 13 19:47:34 krabat kernel: [ 2655.981980] usb 9-7: reset high speed
>>> USB device using ehci_hcd and address 3
>>> Nov 13 19:47:39 krabat kernel: [ 2661.120397] usb 9-7: device firmware
>>> changed
>>
>> You cut too early. This shows a failed reset but not the cause.
>
> There are no relevant messages in the log before:
>
> --------
> Nov 13 19:06:23 krabat kernel: [ 190.790549] EXT4-fs (dm-1): mounted
> filesystem with ordered data mode
> Nov 13 19:08:05 krabat ntpd[3381]: synchronized to 192.168.207.1, stratum 2
> Nov 13 19:08:05 krabat ntpd[3381]: kernel time sync status change 2001
> Nov 13 19:47:34 krabat kernel: [ 2655.981980] usb 9-7: reset high speed
> USB device using ehci_hcd and address 3
> ----------------------------------------------------

it might be that my assumption is totally wrong an the usb-disconnect is
unrelated to the time the usb-drives are going to sleep. What I have
done inbetween that mount and the disconnect was a rsync -c -n (dry-run)
between two of those usb-discs (I'm using them for backups). Both drives
are containing the same files, some large (up to 5-10 gb) and small
files. Looking at the leds of those usb-drives I just assumed the
problem happens when one of those was inactive for some time (because
rsync computes the hash of one of the large files on the other drive).
In the excerpt of the log I've posted before, I've cut out the
disconnect of the other drive, which has happend almost at the same time
(and with the same consequences) as the one shown. In fact I've cut out
one line from the log (sorry for that):
---------
Nov 13 19:08:05 krabat ntpd[3381]: kernel time sync status change 2001
Nov 13 19:47:34 krabat kernel: [ 2655.981980] usb 9-7: reset high speed
USB device using ehci_hcd and address 3
Nov 13 19:47:35 krabat kernel: [ 2656.233278] usb 9-1: reset high speed
USB device using ehci_hcd and address 4
Nov 13 19:47:39 krabat kernel: [ 2661.120397] usb 9-7: device firmware
changed
Nov 13 19:47:39 krabat kernel: [ 2661.120489] usb 9-7: USB disconnect,
address 3
---------
Afterwards the messages for the second drive where the same as for the
one I've posted.

This is reproducible, at least through starting the rsync. I will turn
on some debugging knobs and will see if I can see some more hints in the
resulting logs.

Kind regards,

Alexander Holler
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo(a)vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
From: Oliver Neukum on 14 Nov 2009 05:10
Am Samstag, 14. November 2009 03:50:53 schrieb Alexander Holler:
> In the excerpt of the log I've posted before, I've cut out the
> disconnect of the other drive, which has happend almost at the same time
> (and with the same consequences) as the one shown. In fact I've cut out
> one line from the log (sorry for that):
> ---------
> Nov 13 19:08:05 krabat ntpd[3381]: kernel time sync status change 2001
> Nov 13 19:47:34 krabat kernel: [ 2655.981980] usb 9-7: reset high speed
> USB device using ehci_hcd and address 3

Reset.

> Nov 13 19:47:35 krabat kernel: [ 2656.233278] usb 9-1: reset high speed
> USB device using ehci_hcd and address 4
> Nov 13 19:47:39 krabat kernel: [ 2661.120397] usb 9-7: device firmware
> changed

Descriptors changed

> Nov 13 19:47:39 krabat kernel: [ 2661.120489] usb 9-7: USB disconnect,
> address 3

Forced unbind

> ---------
> Afterwards the messages for the second drive where the same as for the
> one I've posted.
>
> This is reproducible, at least through starting the rsync. I will turn
> on some debugging knobs and will see if I can see some more hints in the
> resulting logs.

You can recompile with CONFIG_USB_DEBUG and CONFIG_USB_STORAGE_DEBUG.

Regards
Oliver
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo(a)vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
From: Alexander Holler on 14 Nov 2009 12:40
Hello,

Am 14.11.2009 11:06, schrieb Oliver Neukum:
> You can recompile with CONFIG_USB_DEBUG and CONFIG_USB_STORAGE_DEBUG.

Here it comes. I've used 2.6.31.6 and turned usb-(storage-)debug on.
After mounting the two large encrypted files (loop and dmcrypt) from the
two connected usb-drives, I've started the rsync -n to compare both
partitions:

--------------------------------
Nov 14 13:03:41 krabat kernel: [ 164.474820] EXT4-fs (dm-1): mounted
filesystem with ordered data mode
Nov 14 13:04:38 krabat kernel: [ 220.963256] 3c da 00 00 10 00
....
(many lines with hex-numbers, nothing else)
....
Nov 14 14:18:49 krabat kernel: [ 4672.953175] 28 00 04 e8 3d da 00 00
10 00
Nov 14 14:18:49 krabat kernel: [ 4672.960345] 00 f0 00

Nov 14 14:18:49 krabat kernel: [ 4672.966138] 00

Nov 14 14:18:50 krabat kernel: [ 4673.071643] 00 04 e8 50 da 00 00 10
00
Nov 14 14:18:50 krabat kernel: [ 4673.077825] 00

Nov 14 14:18:50 krabat kernel: [ 4673.077833] 10 00

Nov 14 14:18:50 krabat kernel: [ 4673.336765] 00 f0 00

Nov 14 14:18:50 krabat kernel: [ 4673.342205] 28 00 05 3b 40 da 00 00
10 00
Nov 14 14:18:50 krabat kernel: [ 4673.342230] 00

Nov 14 14:18:50 krabat kernel: [ 4673.342235]

Nov 14 14:18:51 krabat kernel: [ 4674.647846] 28 00 05 3c 15 da 00 00
10 00
Nov 14 14:18:51 krabat kernel: [ 4674.822258]

Nov 14 14:18:51 krabat kernel: [ 4674.828338] 28 00 04 e9 6e ea 00 00
f0 00
Nov 14 14:18:57 krabat kernel: [ 4680.042678] 54 da 00 00 10 00

Nov 14 14:18:59 krabat kernel: [ 4682.646818] 00 05 e2 0c da 00 00 10
00
Nov 14 14:19:00 krabat kernel: [ 4683.516686] 00 f0 00

Nov 14 14:19:00 krabat kernel: [ 4683.522288] 28 00 05 e6 9f da 00 00
10 00
Nov 14 14:19:07 krabat kernel: [ 4690.760331]

Nov 14 14:19:39 krabat kernel: [ 4722.339214] usb 9-1: reset high speed
USB device using ehci_hcd and address 4
Nov 14 14:19:39 krabat kernel: [ 4722.591235] usb 9-7: reset high speed
USB device using ehci_hcd and address 3
Nov 14 14:19:44 krabat kernel: [ 4727.491664] usb 9-1: device firmware
changed
Nov 14 14:19:44 krabat kernel: [ 4727.491718] usb 9-1: USB disconnect,
address 4
Nov 14 14:19:44 krabat kernel: [ 4727.493060] scsi 7:0:0:0: Device
offlined - not ready after error recovery
Nov 14 14:19:44 krabat kernel: [ 4727.493069] scsi 7:0:0:0: [sdc]
Unhandled error code
Nov 14 14:19:44 krabat kernel: [ 4727.493072] scsi 7:0:0:0: [sdc]
Result: hostbyte=DID_ABORT driverbyte=DRIVER_OK
Nov 14 14:19:44 krabat kernel: [ 4727.493077] end_request: I/O error,
dev sdc, sector 317945322
Nov 14 14:19:44 krabat kernel: [ 4727.493091] scsi 7:0:0:0: rejecting
I/O to offline device
Nov 14 14:19:44 krabat kernel: [ 4727.493108] scsi 7:0:0:0: [sdc]
Unhandled error code
Nov 14 14:19:44 krabat kernel: [ 4727.493110] scsi 7:0:0:0: [sdc]
Result: hostbyte=DID_NO_CONNECT driverbyte=DRIVER_OK
Nov 14 14:19:44 krabat kernel: [ 4727.493114] end_request: I/O error,
dev sdc, sector 317945562
Nov 14 14:19:44 krabat kernel: [ 4727.615743] EXT4-fs error (device
dm-1): ext4_find_entry: reading directory #1771 offset 0
Nov 14 14:19:44 krabat kernel: [ 4727.616071] Buffer I/O error on device
dm-1, logical block 0
....
(many more io, buffer and ext4 errors)
....
--------------------------------

Regards,

Alexander
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo(a)vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
From: Alexander Holler on 16 Nov 2009 19:30
Hello,

Alexander Holler schrieb:
> ...
> (many lines with hex-numbers, nothing else)
> ...

shame on me. Sorry, I wasn't aware the logger (with default
configuration) of the system in question doesn't log every kernel
message. I've just started another rsync after I've changed that.

I useful log will follow after I reproduced the failure while logging
every kernel-message.

Regards,

Alexander
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo(a)vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
First  |  Prev  |  Next  |  Last
Pages: 1 2 3
Prev: vc: Add support for hiding the cursor when creating VTs
Next: [PATCH] forcedeth: mac address fix