linux-arm-kernel.lists.infradead.org archive mirror
 help / color / mirror / Atom feed
* Problem in USB on the Sheevaplug, 2.6.31
@ 2009-09-23  8:43 Lluís Batlle
  2009-09-23  9:35 ` Matthias Kaehlcke
  0 siblings, 1 reply; 4+ messages in thread
From: Lluís Batlle @ 2009-09-23  8:43 UTC (permalink / raw)
  To: linux-arm-kernel

Hello all,

after using a USB storage device (a Kingston DataTraveler 8GB flash
stick) in the sheevaplug for some hours, at one point the device
stopped working until I replugged it in.

The system kept stable, while all this happened.

Here comes the 'dmesg' part:

EXT3 FS on sda1, internal journal
EXT3-fs: mounted filesystem with writeback data mode.
#------------ some hours passed ----------
usb 1-1: reset high speed USB device using orion-ehci and address 2
------------[ cut here ]------------
WARNING: at arch/arm/kernel/process.c:171 cpu_idle+0x74/0xbc()
Modules linked in: sg
[<c002f85c>] (unwind_backtrace+0x0/0xdc) from [<c003ecb0>]
(warn_slowpath_common+0x4c/0x80)
[<c003ecb0>] (warn_slowpath_common+0x4c/0x80) from [<c002b974>]
(cpu_idle+0x74/0xbc)
[<c002b974>] (cpu_idle+0x74/0xbc) from [<c0008b04>] (start_kernel+0x270/0x2d4)
[<c0008b04>] (start_kernel+0x270/0x2d4) from [<00008034>] (0x8034)
---[ end trace 81d90711559dbd0e ]---
usb 1-1: reset high speed USB device using orion-ehci and address 2
usb 1-1: device descriptor read/64, error -110
usb 1-1: device descriptor read/64, error -32
usb 1-1: reset high speed USB device using orion-ehci and address 2
usb 1-1: device descriptor read/64, error -32
usb 1-1: device descriptor read/64, error -32
usb 1-1: reset high speed USB device using orion-ehci and address 2
usb 1-1: device not accepting address 2, error -32
usb 1-1: reset high speed USB device using orion-ehci and address 2
usb 1-1: device not accepting address 2, error -32
sd 0:0:0:0: [sda] Unhandled error code
sd 0:0:0:0: [sda] Result: hostbyte=0x07 driverbyte=0x00
end_request: I/O error, dev sda, sector 15466566
usb 1-1: USB disconnect, address 2
EXT3-fs error (device sda1): read_inode_bitmap: Cannot read inode
bitmap - block_group = 59, inode_bitmap = 1933313
EXT3-fs error (device sda1) in ext3_new_inode: IO failure
------------[ cut here ]------------
WARNING: at fs/buffer.c:1152 mark_buffer_dirty+0x34/0xe8()
Modules linked in: sg
[<c002f85c>] (unwind_backtrace+0x0/0xdc) from [<c003ecb0>]
(warn_slowpath_common+0x4c/0x80)
[<c003ecb0>] (warn_slowpath_common+0x4c/0x80) from [<c00d67d4>]
(mark_buffer_dirty+0x34/0xe8)
[<c00d67d4>] (mark_buffer_dirty+0x34/0xe8) from [<c010b3ac>]
(ext3_commit_super+0x4c/0x70)
[<c010b3ac>] (ext3_commit_super+0x4c/0x70) from [<c010c5a0>]
(ext3_handle_error+0x98/0xc4)
[<c010c5a0>] (ext3_handle_error+0x98/0xc4) from [<c010c634>]
(__ext3_std_error+0x68/0x74)
[<c010c634>] (__ext3_std_error+0x68/0x74) from [<c0102e60>]
(ext3_new_inode+0x898/0x91c)
[<c0102e60>] (ext3_new_inode+0x898/0x91c) from [<c0109898>]
(ext3_mkdir+0x8c/0x2cc)
[<c0109898>] (ext3_mkdir+0x8c/0x2cc) from [<c00bc5ac>] (vfs_mkdir+0x6c/0xc0)
[<c00bc5ac>] (vfs_mkdir+0x6c/0xc0) from [<c00be928>] (sys_mkdirat+0x84/0xd0)
[<c00be928>] (sys_mkdirat+0x84/0xd0) from [<c002a9c0>]
(ret_fast_syscall+0x0/0x2c)
---[ end trace 81d90711559dbd0f ]---
Aborting journal on device sda1.
journal commit I/O error
ext3_abort called.
EXT3-fs error (device sda1): ext3_journal_start_sb: Detected aborted journal
Remounting filesystem read-only

It's a vanilla kernel built with gcc 4.3.4 . If you need more
information, I can compile special versions of that kernel to get some
additional information.

Regards,
Llu?s.

^ permalink raw reply	[flat|nested] 4+ messages in thread

* Problem in USB on the Sheevaplug, 2.6.31
  2009-09-23  8:43 Problem in USB on the Sheevaplug, 2.6.31 Lluís Batlle
@ 2009-09-23  9:35 ` Matthias Kaehlcke
  2009-09-23  9:47   ` Lluís Batlle
  0 siblings, 1 reply; 4+ messages in thread
From: Matthias Kaehlcke @ 2009-09-23  9:35 UTC (permalink / raw)
  To: linux-arm-kernel

hi lluis,

El Wed, Sep 23, 2009 at 10:43:56AM +0200 Llu?s Batlle ha dit:

> after using a USB storage device (a Kingston DataTraveler 8GB flash
> stick) in the sheevaplug for some hours, at one point the device
> stopped working until I replugged it in.
> 
> The system kept stable, while all this happened.
> 
> Here comes the 'dmesg' part:
> 
> EXT3 FS on sda1, internal journal
> EXT3-fs: mounted filesystem with writeback data mode.
> #------------ some hours passed ----------
> usb 1-1: reset high speed USB device using orion-ehci and address 2
> ------------[ cut here ]------------
> WARNING: at arch/arm/kernel/process.c:171 cpu_idle+0x74/0xbc()
> Modules linked in: sg
> [<c002f85c>] (unwind_backtrace+0x0/0xdc) from [<c003ecb0>]
> (warn_slowpath_common+0x4c/0x80)
> [<c003ecb0>] (warn_slowpath_common+0x4c/0x80) from [<c002b974>]
> (cpu_idle+0x74/0xbc)
> [<c002b974>] (cpu_idle+0x74/0xbc) from [<c0008b04>] (start_kernel+0x270/0x2d4)
> [<c0008b04>] (start_kernel+0x270/0x2d4) from [<00008034>] (0x8034)
> ---[ end trace 81d90711559dbd0e ]---
> usb 1-1: reset high speed USB device using orion-ehci and address 2
> usb 1-1: device descriptor read/64, error -110
> usb 1-1: device descriptor read/64, error -32
> usb 1-1: reset high speed USB device using orion-ehci and address 2
> usb 1-1: device descriptor read/64, error -32
> usb 1-1: device descriptor read/64, error -32
> usb 1-1: reset high speed USB device using orion-ehci and address 2
> usb 1-1: device not accepting address 2, error -32
> usb 1-1: reset high speed USB device using orion-ehci and address 2
> usb 1-1: device not accepting address 2, error -32
> sd 0:0:0:0: [sda] Unhandled error code
> sd 0:0:0:0: [sda] Result: hostbyte=0x07 driverbyte=0x00
> end_request: I/O error, dev sda, sector 15466566
> usb 1-1: USB disconnect, address 2
> EXT3-fs error (device sda1): read_inode_bitmap: Cannot read inode
> bitmap - block_group = 59, inode_bitmap = 1933313
> EXT3-fs error (device sda1) in ext3_new_inode: IO failure
> ------------[ cut here ]------------
> WARNING: at fs/buffer.c:1152 mark_buffer_dirty+0x34/0xe8()
> Modules linked in: sg
> [<c002f85c>] (unwind_backtrace+0x0/0xdc) from [<c003ecb0>]
> (warn_slowpath_common+0x4c/0x80)
> [<c003ecb0>] (warn_slowpath_common+0x4c/0x80) from [<c00d67d4>]
> (mark_buffer_dirty+0x34/0xe8)
> [<c00d67d4>] (mark_buffer_dirty+0x34/0xe8) from [<c010b3ac>]
> (ext3_commit_super+0x4c/0x70)
> [<c010b3ac>] (ext3_commit_super+0x4c/0x70) from [<c010c5a0>]
> (ext3_handle_error+0x98/0xc4)
> [<c010c5a0>] (ext3_handle_error+0x98/0xc4) from [<c010c634>]
> (__ext3_std_error+0x68/0x74)
> [<c010c634>] (__ext3_std_error+0x68/0x74) from [<c0102e60>]
> (ext3_new_inode+0x898/0x91c)
> [<c0102e60>] (ext3_new_inode+0x898/0x91c) from [<c0109898>]
> (ext3_mkdir+0x8c/0x2cc)
> [<c0109898>] (ext3_mkdir+0x8c/0x2cc) from [<c00bc5ac>] (vfs_mkdir+0x6c/0xc0)
> [<c00bc5ac>] (vfs_mkdir+0x6c/0xc0) from [<c00be928>] (sys_mkdirat+0x84/0xd0)
> [<c00be928>] (sys_mkdirat+0x84/0xd0) from [<c002a9c0>]
> (ret_fast_syscall+0x0/0x2c)
> ---[ end trace 81d90711559dbd0f ]---
> Aborting journal on device sda1.
> journal commit I/O error
> ext3_abort called.
> EXT3-fs error (device sda1): ext3_journal_start_sb: Detected aborted journal
> Remounting filesystem read-only
> 
> It's a vanilla kernel built with gcc 4.3.4 . If you need more
> information, I can compile special versions of that kernel to get some
> additional information.

the USB FAQ suggests it might be a problem of signal quality, try if it
works when removing ehci-hcd:
http://www.linux-usb.org/FAQ.html#ts6

is the file system subject to a lot of write accesses? in this case it
is likely you will encounter problems of corrupt sectors due to flash
wear out. be sure not to mount locations like /var (especially
/var/log) on the flash device. you also should mount the device with
the options noatime and nodiratime if you aren't doing it yet

<ca>per cert: que tal el sheeva plug? porto temps pensant de
comprar-ne un</ca>

salutacions

-- 
Matthias Kaehlcke
Embedded Linux Engineer
Barcelona

     Nationalism is an infantile disease. It is the measles of mankind
                         (Albert Einstein)
                                                                 .''`.
    using free software / Debian GNU/Linux | http://debian.org  : :'  :
                                                                `. `'`
gpg --keyserver pgp.mit.edu --recv-keys 47D8E5D4                  `-

^ permalink raw reply	[flat|nested] 4+ messages in thread

* Problem in USB on the Sheevaplug, 2.6.31
  2009-09-23  9:35 ` Matthias Kaehlcke
@ 2009-09-23  9:47   ` Lluís Batlle
  2009-09-23 10:12     ` Matthias Kaehlcke
  0 siblings, 1 reply; 4+ messages in thread
From: Lluís Batlle @ 2009-09-23  9:47 UTC (permalink / raw)
  To: linux-arm-kernel

Hello Matthias,
glad to find you here. :)

2009/9/23, Matthias Kaehlcke <matthias@kaehlcke.net>:
> hi lluis,
>
> the USB FAQ suggests it might be a problem of signal quality, try if it
>  works when removing ehci-hcd:
>  http://www.linux-usb.org/FAQ.html#ts6
I have no cable between the board USB connector and the stick, and the
Sheevaplug people claim to have good USB 2.0 support... I will see if
it happens again.

>  is the file system subject to a lot of write accesses? in this case it
Yes, I'm using it as /tmp, where I do builds. Kernel builds for example.
>  is likely you will encounter problems of corrupt sectors due to flash
>  wear out. be sure not to mount locations like /var (especially
I barely used that stick... and I trust Kingston so I think this is
not the case.
>  /var/log) on the flash device. you also should mount the device with
>  the options noatime and nodiratime if you aren't doing it yet
I use noatime, but I didn't know of nodiratime.
>
>  <ca>per cert: que tal el sheeva plug? porto temps pensant de
>  comprar-ne un</ca>
<ca>Doncs n'estic ben content. Pel preu crec que est? molt b?. Hi he
estat portant el NixOS, i no he fet cap compilaci? creuada, aix? que
s'ho ha menjat tot el Sheevaplug. Per? fa nom?s un dia que utilitzo
l'stick USB, abans ho feia tot amb nfsroot.</ca>

Salutacions,
Llu?s.

^ permalink raw reply	[flat|nested] 4+ messages in thread

* Problem in USB on the Sheevaplug, 2.6.31
  2009-09-23  9:47   ` Lluís Batlle
@ 2009-09-23 10:12     ` Matthias Kaehlcke
  0 siblings, 0 replies; 4+ messages in thread
From: Matthias Kaehlcke @ 2009-09-23 10:12 UTC (permalink / raw)
  To: linux-arm-kernel

El Wed, Sep 23, 2009 at 11:47:52AM +0200 Llu?s Batlle ha dit:

> 2009/9/23, Matthias Kaehlcke <matthias@kaehlcke.net>:
> > hi lluis,
> >
> > the USB FAQ suggests it might be a problem of signal quality, try if it
> >  works when removing ehci-hcd:
> >  http://www.linux-usb.org/FAQ.html#ts6
> I have no cable between the board USB connector and the stick, and the
> Sheevaplug people claim to have good USB 2.0 support... I will see if
> it happens again.

ok

> >  is the file system subject to a lot of write accesses? in this case it
> Yes, I'm using it as /tmp, where I do builds. Kernel builds for
> example.

ouch, i don't think that's a good idea. i'd recommend using a
hard drive for stuff like this

> >  is likely you will encounter problems of corrupt sectors due to flash
> >  wear out. be sure not to mount locations like /var (especially
> I barely used that stick... and I trust Kingston so I think this is
> not the case.

flash sticks aren't designed for usage profiles with very frequent
writes, like kernel builds. i'd bet even a kingston one will have
problems in a near future if you use it in that way. have a look at
http://free-electrons.com/doc/flash-filesystems.pdf P. 29ff for more
information about the issue.

it's also recommended to use ext2 instead of ext3 to reduce wear out
produced by journalling

> >  /var/log) on the flash device. you also should mount the device with
> >  the options noatime and nodiratime if you aren't doing it yet
> I use noatime, but I didn't know of nodiratime.
> >
> >  <ca>per cert: que tal el sheeva plug? porto temps pensant de
> >  comprar-ne un</ca>
> <ca>Doncs n'estic ben content. Pel preu crec que est? molt b?. Hi he
> estat portant el NixOS, i no he fet cap compilaci? creuada, aix? que
> s'ho ha menjat tot el Sheevaplug. Per? fa nom?s un dia que utilitzo
> l'stick USB, abans ho feia tot amb nfsroot.</ca>

<ca>molt be allo del nixos, ets un friki de debo ;) jo encara funciono
amb un nslu2, pero va una mica just de recursos, per aixo m'estic
plantetjant el sheevaplug. per cert: t'havia comentat que el teu task
spooler va ser presentat a la versio impresa del <de>linux
magazin</de>?</ca>

-- 
Matthias Kaehlcke
Embedded Linux Engineer
Barcelona

        They that can give up essential liberty to obtain a little
           temporary safety deserve neither liberty nor safety
                         (Benjamin Franklin)
                                                                 .''`.
    using free software / Debian GNU/Linux | http://debian.org  : :'  :
                                                                `. `'`
gpg --keyserver pgp.mit.edu --recv-keys 47D8E5D4                  `-

^ permalink raw reply	[flat|nested] 4+ messages in thread

end of thread, other threads:[~2009-09-23 10:12 UTC | newest]

Thread overview: 4+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2009-09-23  8:43 Problem in USB on the Sheevaplug, 2.6.31 Lluís Batlle
2009-09-23  9:35 ` Matthias Kaehlcke
2009-09-23  9:47   ` Lluís Batlle
2009-09-23 10:12     ` Matthias Kaehlcke

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).