NAND problems with at91 kernels

NAND problems with at91 kernels

[Savin Zlobec]

Hi all,

I'am running kernel 2.6.17 on AT91RM9200 board with Samsung 32MiB NAND.
I've been experiencing a lot of NAND flash errors (running JFFS2 or 
YAFFS)  which
traced down to busy delay handling. Asking on linux-arm-kernel list I've 
got a link to
this message thread:

http://lists.infradead.org/pipermail/linux-mtd/2005-September/013625.html

I've patched nand_base:nand_read_oob with the patch attached to the 
above mail and
it seems to solve my problems - I'am going to run some test for the next 
couple of days
just to be shure.

Is there any reason why this patch was not accepted and more importantly 
can I expect
more delay related problems which apperantly show only on some platforms.

Thanks,

savin

Re: NAND problems with at91 kernels

[Thomas Gleixner]

On Mon, 2006-06-19 at 15:06 +0200, Savin Zlobec wrote:
> Hi all,
> 
> I'am running kernel 2.6.17 on AT91RM9200 board with Samsung 32MiB NAND.
> I've been experiencing a lot of NAND flash errors (running JFFS2 or 
> YAFFS)  which
> traced down to busy delay handling. Asking on linux-arm-kernel list I've 
> got a link to
> this message thread:
> 
> http://lists.infradead.org/pipermail/linux-mtd/2005-September/013625.html
> 
> I've patched nand_base:nand_read_oob with the patch attached to the 
> above mail and
> it seems to solve my problems - I'am going to run some test for the next 
> couple of days
> just to be shure.
> 
> Is there any reason why this patch was not accepted and more importantly 
> can I expect
> more delay related problems which apperantly show only on some platforms.

I probably missed that patch and nobody yelled loud enough. :)

I looked at the code in question and I think this is actually a problem
for FLASH chips which have the "autoincrement" feature. So after reading
one chunk, it leaves the chip in the autoread mode. Can you please check
if this still happens with the current mtd-git code ? 

You need update the at91 nand code to the recent changes and when you
are at it you might resubmit the patch to MTD mailinglist so we can
include it into the official tree.

Thanks,

	tglx