public inbox for linux-mtd@lists.infradead.org
 help / color / mirror / Atom feed
From: Estelle HAMMACHE <estelle.hammache@st.com>
To: Rudresh NB <getrudri@yahoo.com>
Cc: linux-mtd@lists.infradead.org
Subject: Re: Query
Date: Wed, 22 Dec 2004 10:26:14 +0100	[thread overview]
Message-ID: <41C93DB6.16162C19@st.com> (raw)
In-Reply-To: 20041222090557.21188.qmail@web13202.mail.yahoo.com

Rudresh NB wrote:
> 
> Hi,
>     Thanx for the reply.
> 
> I have couple of questions.
> 
> 1) How JFFS2 will take care of Updating the same page
> in the block.

JFFS2 will not _update_ a NAND Flash page. When the user
overwrites a file with new data, JFFS2 will NOT update the
previous data. Rather JFFS2 will write the new data at
another free place, increasing the version number. The
old data which is not needed anymore will disappear during
garbage collection. JFFS2 does not write twice to the same
physical page, unless the block was erased in the meantime.
I suggest that you read http://sources.redhat.com/jffs2/jffs2.pdf 
if you haven't already.

> 2) How does the logical addr to physical addr of block
>  no happens

JFFS2 is NOT an FTL. There are no logical block numbers
in JFFS2. I think the block adresses are relative to the
start of the partition, this is managed in mtd drivers (?).

bye
Estelle



> --- Estelle Hammache wrote:
> 
> > Hi,
> > JFFS2 write pages sequentially from the start of an
> > erase
> > block to the end of the block. The current
> > write block is in c->nextblock. The current write
> > offset
> > in this block is c->nextblock->offset.
> > There is a write buffer (see file wbuf.c) to ensure
> > that
> > only full pages are written, even though data nodes
> > may
> > overlap a page boundary.
> > Data nodes may not overlap an erase block boundary:
> > if there
> > is no more room in the current block, a new (empty)
> > block
> > is selected (and the page buffer is flushed in the
> > old block
> > previously to writing to the new block). The new
> > block
> > selection and the write address selection happen in
> > nodemgmt.c (jffs2_do_reserve_space).
> > Does this answer your question ?
> > Estelle
> >
> > Rudresh NB wrote:
> > >
> > > Hi All,
> > >        Iam using samsung Nand flash (k9f2g08u0m).
> > In
> > > that there is restriction of random page program.
> > I
> > > heard that jffs2 takes care. but iam unable to
> > find
> > > where exactly (In which file ) it is taken care.
> > > Brief explanation is appreciated.
> > >
> > > Thanx in advance
> > >
> > > Rgds
> > > NBR

  reply	other threads:[~2004-12-22  9:26 UTC|newest]

Thread overview: 6+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2004-12-21 12:08 Query Rudresh NB
2004-12-21 12:19 ` Query Artem B. Bityuckiy
2004-12-21 12:42 ` Query Estelle HAMMACHE
2004-12-22  9:05   ` Query Rudresh NB
2004-12-22  9:26     ` Estelle HAMMACHE [this message]
  -- strict thread matches above, loose matches on Subject: below --
2004-06-25 13:13 query Manish RATHI

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=41C93DB6.16162C19@st.com \
    --to=estelle.hammache@st.com \
    --cc=getrudri@yahoo.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