On Thu, Oct 14, 2010 at 3:14 PM, Tejun Heo hte...@gmail.com wrote:
Hello,
Can you please try this one then? It seems to work here but I can't
reproduce the original problem reliably so I'm not really sure.
Thanks.
diff --git a/arch/um/drivers/ubd_kern.c b/arch/um/drivers/ubd_kern.c
index
Hello,
On 10/14/2010 04:20 PM, richard -rw- weinberger wrote:
It does not work for me.
But the error is a different one. :-)
Without your patch I've never got this kernel trace.
[ 59.85] kworker/0:1: page allocation failure. order:0, mode:0x20
Hmm... you're seeing out of memory
On Thu, Oct 14, 2010 at 03:14:28PM +0200, Tejun Heo wrote:
Hello,
Can you please try this one then? It seems to work here but I can't
reproduce the original problem reliably so I'm not really sure.
Thanks.
diff --git a/arch/um/drivers/ubd_kern.c b/arch/um/drivers/ubd_kern.c
index
Hello,
Can you please try this one then? It seems to work here but I can't
reproduce the original problem reliably so I'm not really sure.
Thanks.
diff --git a/arch/um/drivers/ubd_kern.c b/arch/um/drivers/ubd_kern.c
index 1bcd208..9734994 100644
--- a/arch/um/drivers/ubd_kern.c
+++
On 2010-10-05 22:31, Chris Frey wrote:
On Tue, Oct 05, 2010 at 10:23:19AM +0200, Tejun Heo wrote:
H, can you please give a shot at the following one? Thank you.
I applied this patch on top of stock 2.6.35.5 as usual (no other patches)
and tested on my maverick image as before. I ran a
On Thu, Oct 07, 2010 at 09:58:16AM +0200, Jens Axboe wrote:
So how about this? Note that I haven't even compiled this. The request
handling logic really should be fixed in there, it's horribly
inefficient.
Thanks. I fixed the compile error with:
+ .rq_offset 0, \
to
+
On Mon, Oct 04, 2010 at 06:37:36PM +0200, Tejun Heo wrote:
Hello, sorry about chiming in later. I was off last week.
No problem, I'm eager to test patches to fix this.
I think we're on the right track. The problem with Jens' patch was
that it didn't consider the fact that blk_end_request()
On 10/04/2010 09:51 PM, Chris Frey wrote:
On Mon, Oct 04, 2010 at 06:37:36PM +0200, Tejun Heo wrote:
Hello, sorry about chiming in later. I was off last week.
No problem, I'm eager to test patches to fix this.
I think we're on the right track. The problem with Jens' patch was
that it
On Wed, Sep 29, 2010 at 7:21 AM, Jens Axboe jax...@fusionio.com wrote:
I think we need to find the real fix here, just disabling merging
is not a fix (it's just a nasty work-around for the real bug).
Jens,
Do you have an idea which parts of the code are buggy?
--
Cheers,
//richard
Hello, sorry about chiming in later. I was off last week.
On 09/29/2010 08:34 AM, Chris Frey wrote:
On Wed, Sep 29, 2010 at 02:21:07PM +0900, Jens Axboe wrote:
This seems to imply that the original commit pin pointed is not
the only issue we have in that code atm.
I think we need to find
On Wed, Sep 29, 2010 at 12:13:10AM +0200, Richard Weinberger wrote:
Am Mittwoch 29 September 2010, 00:00:00 schrieb Andrew Morton:
This is a workaround, I think? Do we know what the actual bug is?
From the comment it appears to be a regression?
Yes, it is a workaround.
For more details
On Tue, 28 Sep 2010 23:47:36 +0200
Richard Weinberger rich...@nod.at wrote:
Under high load the file system gets corrupted.
This patch fixes the issue.
Many thanks to Janjaap Bos janj...@bos.nl!
LKML-Reference: AANLkTi=PTp7YW_eYxtF-H2QSxgei3whWH59wU0C9oCkz () mail !
gmail ! com
On 2010-09-29 07:52, Chris Frey wrote:
On Wed, Sep 29, 2010 at 12:13:10AM +0200, Richard Weinberger wrote:
Am Mittwoch 29 September 2010, 00:00:00 schrieb Andrew Morton:
This is a workaround, I think? Do we know what the actual bug is?
From the comment it appears to be a regression?
Yes, it
On Wed, 2010-09-29 at 08:10 +0900, Jens Axboe wrote:
It looks like that if we need to restart the requeue, then
we use the initial position and not the current index. Does
this help?
diff --git a/arch/um/drivers/ubd_kern.c b/arch/um/drivers/ubd_kern.c
index 1bcd208..81ee063 100644
---
On Wed, Sep 29, 2010 at 12:13:10AM +0200, Richard Weinberger wrote:
Am Mittwoch 29 September 2010, 00:00:00 schrieb Andrew Morton:
This is a workaround, I think? Do we know what the actual bug is?
From the comment it appears to be a regression?
Yes, it is a workaround.
For more details
On 2010-09-29 10:29, Chris Frey wrote:
On Wed, Sep 29, 2010 at 08:10:06AM +0900, Jens Axboe wrote:
It looks like that if we need to restart the requeue, then
we use the initial position and not the current index. Does
this help?
diff --git a/arch/um/drivers/ubd_kern.c
On Wed, Sep 29, 2010 at 02:21:07PM +0900, Jens Axboe wrote:
This seems to imply that the original commit pin pointed is not
the only issue we have in that code atm.
I think we need to find the real fix here, just disabling merging
is not a fix (it's just a nasty work-around for the real
17 matches
Mail list logo