I am not seeing anything written to oob area via my flash driver
because yaffs is using inband tags. So in that case you think that
page writes/read writes are not working correctly.


Also generic question is: Is the mtd nand tests, speedtest, pagetest,
subpagetest,oobtest, erase test are enough tests to say that a flash
driver is flawless and ready to support jffs2/yaffs2, when an mtd
flash driver passes those tests?

Basically I am looking for ways to debug a flash driver without
debugging/dealing issues with yaffs2/jffs2 bring-up. Any ideas are
apreciated.

Thanks,
-Subba

On Sat, May 22, 2010 at 12:36 PM, Charles Manning <cdhmanning@gmail.com> wrote:
> This is caused by the oob area not being stored correctly. The tags arw
> getting corrupted and lost.
>
> It might look like the file is being stored and read back but it is in the
> cache.  If you drop the cache then you will see the files has not actually
> been stored properly.
>
> On May 22, 2010 6:44 AM, "subbarao mungara" <mungara@gmail.com> wrote:
>
> Hi,
>
>  I am having issue with yaffs2 file system. Files in yaffs2 flash file
> system not retained across mount operations. Here is what I am
> noticing.
>
>  // erase all flash contents
>  $ flaserase-all /dev/mtd3
> // mount yaffs2 to /flash
>  $ mount -t yaffs2 /dev/mtdblock3 /flash
>
> // add hello.c to flash filesystem
>  $cd /flash
>  $ cp /hello.c .
>  $ls                       // ls lists both lost_found and hello.c as well.
>  $ cd /
> // unmount flash file system
>  $umount /flash
>
> // remount the flash file system
>  $ mount -t yaffs2 /dev/mtdblock3 /flash
>  $ls                          // ls lists only "lost+found"
>
>
> Could you explain what causes this behavior?
>
> Thanks,
> -Subba
>
> _______________________________________________
> yaffs mailing list
> yaffs@lists.aleph1.co.uk
> http://lists.aleph1.co.uk/cgi-bin/mailman/listinfo/yaffs
>