Kernel 2.6.0-test9, deadlock using usb-storage, eventually memory allocation bug

Kernel 2.6.0-test9, deadlock using usb-storage, eventually memory allocation bug

[Maximilian Mehnert]

Hello, everybody!

So far I have spoken to Jari Ruusu <jariruusu@users.sourceforge.net> who
is maintaining loop-AES and to Matthew Dharm <mdharm-usb@one-eyed-alien.net>
who
seems to be maintaining usb-storage at the moment.

Jari Ruusu helped me by locating the whereabouts of the following bug by
reading
my syslogs with debugging output from usb-storage and by providing some kernel
patches to isolate the error:
I am using a harddisk attached via a cardbus usb 2.0 card. On it I have an
encrypted partition which I access via loop-AES.
I can easily reproduce a complete deadlock in the usb-storage system by
mounting
my encrypted partition or copying files to or from it (depends on
configuration).

That's what Jari Ruusu wrote on Sun, 09 Nov 2003:
> It was very similar memory allocation failure again: usb storage RAM alloc
> waited for pages to be freed, and all freeable pages were waiting to be
> written out to your usb device. Same thing, just different place.


On Sun, Nov 16, 2003 at 12:48:42PM -0800, Matthew Dharm wrote:
> You should take this up with linux-kernel@vger.kernel.org -- it's a memory
> allocation problem, not a usb-storage problem.  I can't fix it.

I did without posting my kernel config and my syslog files with the usb-storage
messages as according to the FAQ this would be overkill.

If anybody is willing to help me I  would gratefully send her/him all my logs
and my previous correspondence on this topic :)

Greetings from Berlin && excuses for my bad English,

Maximilian

-- 
Maximilian Mehnert <mmehnert at gmx dot net>
http://members.lycos.co.uk/endofuniverse/gpg.html
Fingerprint: 387F 5AA6 5856 2A49 C88C  DA86 FBA8 5122 817B C60E


----------------------------------------------------------------
This message was sent using IMP, the Internet Messaging Program.

Re: Kernel 2.6.0-test9, deadlock using usb-storage, eventually memory allocation bug

[Vedran Rodic]

On Mon, 17 Nov 2003 20:25:45 +0100, Maximilian Mehnert wrote:

> If anybody is willing to help me I  would gratefully send her/him all my logs
> and my previous correspondence on this topic :)


I think you may be hitting the same problem I did some time ago.

Please see
http://www.mail-archive.com/linux-usb-devel%40lists.sourceforge.net/msg17080.html


Try the attached patch Alan Stern made. You'll have to add a "," to the
end of "+      .max_sectors =                  240" line. 

Alan, this patch should be in queue for 2.6.0, right?



Vedran Rodic

Re: Kernel 2.6.0-test9, deadlock using usb-storage, eventually memory allocation bug

[Maximilian Mehnert]

Zitat von Vedran Rodic <vedran@quark.fsb.hr>:

> I think you may be hitting the same problem I did some time ago.
> 
> Please see
>
http://www.mail-archive.com/linux-usb-devel%40lists.sourceforge.net/msg17080.html
> 
That was very fast! Thanks a lot!
You were absolutely right. My usb disk is working fine now.


Maximilian

-- 
Maximilian Mehnert <mmehnert at gmx dot net>
http://members.lycos.co.uk/endofuniverse/gpg.html
Fingerprint: 387F 5AA6 5856 2A49 C88C  DA86 FBA8 5122 817B C60E