at91: ubi: UBIFS error (pid 1): ubifs_iget: failed to read inode 1, error -2

[Bryan Evenson]

Adding appropriate UBIFS maintainers and the mtd mailing list to the thread as this deals with the UBIFS.

> -----Original Message-----
> From: linux-arm-kernel [mailto:linux-arm-kernel-
> bounces at lists.infradead.org] On Behalf Of Bryan Evenson
> Sent: Tuesday, November 19, 2013 2:46 PM
> To: linux-arm-kernel at lists.infradead.org
> Subject: at91: ubi: UBIFS error (pid 1): ubifs_iget: failed to read
> inode 1, error -2
> 
> I am using an Atmel AT91SAM9G25 board with Micron's MT29F2G08ABAEAWP
> NAND flash.  I am on Atmel's variant of the 3.6.9 kernel.
> Specifically, I'm using https://github.com/linux4sam/linux-
> at91/tree/linux-3.6.9-at91 commit
> 5fe012d4aee98ef82df66e6763882c735902ed0f.   I have several boards in
> which I see the error message shown in the message subject when the
> device is reset due to a power cycle and the device fails to continue
> to load the filesystem.  On one board I erased and re-wrote the
> filesystem and the device is working just fine through multiple power
> cycles.  My questions are:
> 
> 1. Is this a known issue with the version of the kernel that I am
> using?  I've searched this mailing list and the latest mainline kernel
> commits, and I didn't see anything directly related to this issue.
> 2. If so, is there a fix in a later kernel version?
> 3. If not, what would be helpful in figuring out what is going on?  I
> still have a bad device that I haven't touched the flash on.
> 

I took a deeper look at the commit history of the kernel I am using, and the last UBI/UBIFS patches to my kernel that I found in the mainline kernel are:

UBIFS: fix mounting problems after power cuts
UBIFS: introduce categorized lprops counter

which were merged into the mainline kernel on 2012-11-15.  I don't see any UBI/UBIFS related patches since then that specifically mention the error I'm seeing.  However, there are a few notable bugs that have been fixed since then such as:

UBIFS: fix use of freed ubifs_orphan objects
UBIFS: fix double free of ubifs_orphan objects
ubifs: wait for page writeback to provide stable pages
UBIFS: prepare to fix a horrid bug
UBIFS: fix a horrid bug
UBI: Fix PEB leak in wear_leveling_worker()

Could not having these fixes in place lead to the issue I'm seeing if there is a power cut?  The kernel I am using does not yet have support for fastmap and I am using the default UBI/UBIFS configuration (4096 wear-leveling, 2% reserved eraseblocks for bad eraseblock handling).  The boards on which I am having an issue are test boards that have been in use long enough that wear-leveling should be occurring.  And as previously noted, I still have at least one board in its broken state if the flash image would be useful for someone (or if someone knows what I should be looking for to figure out what happened).

Thanks,
Bryan

> Thanks,
> Bryan
> 
> 
> 
> 
> _______________________________________________
> linux-arm-kernel mailing list
> linux-arm-kernel at lists.infradead.org
> http://lists.infradead.org/mailman/listinfo/linux-arm-kernel