From: Artem Bityutskiy <dedekind1@gmail.com>
To: Atlant Schmidt <aschmidt@dekaresearch.com>
Cc: Andrew Coll <acoll@dekaresearch.com>,
"'linux-mtd@lists.infradead.org'" <linux-mtd@lists.infradead.org>,
Cale Surgen <csurgen@dekaresearch.com>
Subject: Re: Where is the "scrubbing" threshold set in UBI? Can it be easily varied?
Date: Wed, 16 Jul 2014 09:05:55 +0300 [thread overview]
Message-ID: <1405490755.1829.16.camel@sauron.fi.intel.com> (raw)
In-Reply-To: <0A40042D85E7C84DB443060EC44B3FD36D986A2F14@dekaexchange07.deka.local>
On Thu, 2014-07-03 at 10:07 -0400, Atlant Schmidt wrote:
> Folks:
>
> Within the UBI layer, where is the threshold set that
> determines how many bit-flips must occur on a read
> before the block is "scrubbed" (stored into another
> location)?
Are you now talking about the MTD layer, and the amount of bits
corrected by an ECC? In that case, see what Pekon answered. There is a
configurable threshold and MTD won't tell UBI that there were bit-flips
unless the amount of bits corrected in an ECC "step" is greater or
equivalent to the threshold. This is documented.
> And is this done on a page-by-page basis or on a
> whole-LEB/PEB basis?
On the MTD/NAND driver level, there is a notion of ECC step. One NAND
page may consist of a single or multiple ECC steps...
UBI has no idea about ECC steps, though. It just schedules a PEB for
scrubbing if it gets -EUNCLEAN from MTD when it reads the PEB.
> That is, if a single page has
> more than threshold number of bit flips, will just
> that page be relocated or will the entire LEB be
> relocated to another PEB?
Entire PEB will be copied to another PEB, and the LEB will be re-mapped
from the old PEB to the new PEB, and the old PEB will be erased.
--
Best Regards,
Artem Bityutskiy
prev parent reply other threads:[~2014-07-16 6:06 UTC|newest]
Thread overview: 3+ messages / expand[flat|nested] mbox.gz Atom feed top
2014-07-03 14:07 Where is the "scrubbing" threshold set in UBI? Can it be easily varied? Atlant Schmidt
2014-07-04 4:40 ` Gupta, Pekon
2014-07-16 6:05 ` Artem Bityutskiy [this message]
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=1405490755.1829.16.camel@sauron.fi.intel.com \
--to=dedekind1@gmail.com \
--cc=acoll@dekaresearch.com \
--cc=aschmidt@dekaresearch.com \
--cc=csurgen@dekaresearch.com \
--cc=linux-mtd@lists.infradead.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox