[PATCH 8/8] mtd: nand: use ECC, if present, when scanning OOB

[Artem Bityutskiy]

Hi Shmulik,

On Wed, 2012-08-15 at 17:31 +0300, Shmulik Ladkani wrote:
> > Amended the commit message to satisfy Smulik's request, and pushed to
> > l2-mtd.git, thanks Brian!
> 
> I had some second thoughts WRT the move from MTD_OPS_RAW to
> MTD_OPS_PLACE_OOB.
> 
> When reading BBT content, MTD_OPS_PLACE_OOB is preferred.
> 
> But when scanning the OOBs for BBMs (when creating the table), I guess
> MTD_OPS_RAW should be used - since manufacturer bad blocks are not
> protected by ECC (as Matthieu noted in [1], and I elaborated in [2]).
> 
> Can you please review and comment regarding the ideas discussed there?
> 
> Thanks,
> Shmulik
> 
> [1]
> http://thread.gmane.org/gmane.linux.drivers.mtd/42243/focus=42365
> [2]
> http://thread.gmane.org/gmane.linux.drivers.mtd/42243/focus=42669

I cannot spend much time reading the threads carefully, so I went
through the discussion only very briefly. Apologies. But I took the
patches because:

1. They do not break existing (in-tree) systems which do not have OOB
protected with ECC. This means the patches are harmless and while useful
because they clean up the code.

2. For systems like Brian has where OOB is protected with ECC, the BBM
should just not be covered by the OOB ECC. And the OOB scanning code
should not read non-BBM bytes, and we are done. This seems to be the
only logical setup for me.

If it is not the case, I just consider it is Brian's problem, he'll deal
with it and send us another set of patches.

Or to put it differently: he sent a good clean-up, which does not seem t
break anything, why would we not take it?

If you object merging these patches, I'd appreciate if you could
elaborate why they would be _harmful_ to have.

Thanks!

-- 
Best Regards,
Artem Bityutskiy
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 836 bytes
Desc: This is a digitally signed message part
URL: <http://lists.infradead.org/pipermail/linux-mtd/attachments/20120816/f566f261/attachment.sig>