public inbox for linux-mtd@lists.infradead.org
 help / color / mirror / Atom feed
* Increasing read request size in JFFS2
@ 2007-09-14 11:24 payagond
  2007-09-14 11:38 ` Artem Bityutskiy
  0 siblings, 1 reply; 5+ messages in thread
From: payagond @ 2007-09-14 11:24 UTC (permalink / raw)
  To: linux-mtd

Hi,
I want to know, whether, we can exploit the fast sequential read 
(multiple pages at one go) feature of a nand flash device by existing 
JFFS2 code.
read_pages in readahead decides to call either 
mapping->a_ops->readpages or mapping->a_ops->readpage
can I implement jffs2_readpages to do so?
The problem I feel is (if I have understood correctly)
To read more pages at one shot, we need to anyway read the 
jffs2_raw_inode header first and then calculate the lenth to read, will 
it give me more performance than what is existing?
regards
payagond


________________________________________________________________________
Email and AIM finally together. You've gotta check out free AOL Mail! - 
http://mail.aol.com

^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: Increasing read request size in JFFS2
  2007-09-14 11:24 Increasing read request size in JFFS2 payagond
@ 2007-09-14 11:38 ` Artem Bityutskiy
  2007-09-14 12:50   ` Josh Boyer
  0 siblings, 1 reply; 5+ messages in thread
From: Artem Bityutskiy @ 2007-09-14 11:38 UTC (permalink / raw)
  To: payagond; +Cc: linux-mtd

On Fri, 2007-09-14 at 07:24 -0400, payagond@aol.com wrote:
> Hi,
> I want to know, whether, we can exploit the fast sequential read 
> (multiple pages at one go) feature of a nand flash device by existing 
> JFFS2 code.
> read_pages in readahead decides to call either 
> mapping->a_ops->readpages or mapping->a_ops->readpage
> can I implement jffs2_readpages to do so?
> The problem I feel is (if I have understood correctly)
> To read more pages at one shot, we need to anyway read the 
> jffs2_raw_inode header first and then calculate the lenth to read, will 
> it give me more performance than what is existing?
> regards
> payagond

JFFS2 reads maximum 3 2K NAND pages at a time, although very rarely.
Mostly it reads one page if compression is enabled.

-- 
Best regards,
Artem Bityutskiy (Битюцкий Артём)

^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: Increasing read request size in JFFS2
  2007-09-14 11:38 ` Artem Bityutskiy
@ 2007-09-14 12:50   ` Josh Boyer
  2007-09-17 10:26     ` need patch for SST39VF301flash chips Bruno
  0 siblings, 1 reply; 5+ messages in thread
From: Josh Boyer @ 2007-09-14 12:50 UTC (permalink / raw)
  To: dedekind; +Cc: linux-mtd, payagond

On Fri, 14 Sep 2007 14:38:27 +0300
Artem Bityutskiy <dedekind@infradead.org> wrote:

> On Fri, 2007-09-14 at 07:24 -0400, payagond@aol.com wrote:
> > Hi,
> > I want to know, whether, we can exploit the fast sequential read 
> > (multiple pages at one go) feature of a nand flash device by existing 
> > JFFS2 code.
> > read_pages in readahead decides to call either 
> > mapping->a_ops->readpages or mapping->a_ops->readpage
> > can I implement jffs2_readpages to do so?
> > The problem I feel is (if I have understood correctly)
> > To read more pages at one shot, we need to anyway read the 
> > jffs2_raw_inode header first and then calculate the lenth to read, will 
> > it give me more performance than what is existing?
> > regards
> > payagond
> 
> JFFS2 reads maximum 3 2K NAND pages at a time, although very rarely.
> Mostly it reads one page if compression is enabled.

And isn't a performance gain only seen if the data is actually
in sequential pages on flash?  For fresh filesystems, that may be true.
But over time, the data for each individual file can scatter everywhere.

josh

^ permalink raw reply	[flat|nested] 5+ messages in thread

* need patch for SST39VF301flash chips
  2007-09-14 12:50   ` Josh Boyer
@ 2007-09-17 10:26     ` Bruno
  2007-09-17 11:06       ` Josh Boyer
  0 siblings, 1 reply; 5+ messages in thread
From: Bruno @ 2007-09-17 10:26 UTC (permalink / raw)
  To: linux-mtd



  

Hi...

Has anyone released MTD drivers patch for SST39VF3201 flash chips for linux
kernel 2.4.27?

Need it urgently.

Regards,
Bruno

^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: need patch for SST39VF301flash chips
  2007-09-17 10:26     ` need patch for SST39VF301flash chips Bruno
@ 2007-09-17 11:06       ` Josh Boyer
  0 siblings, 0 replies; 5+ messages in thread
From: Josh Boyer @ 2007-09-17 11:06 UTC (permalink / raw)
  To: bruno; +Cc: linux-mtd

On Mon, 2007-09-17 at 15:56 +0530, Bruno wrote:
> 
>   
> 
> Hi...
> 
> Has anyone released MTD drivers patch for SST39VF3201 flash chips for linux
> kernel 2.4.27?
> 
> Need it urgently.

http://linux-mtd.infradead.org/source.html#kernelversions

josh

^ permalink raw reply	[flat|nested] 5+ messages in thread

end of thread, other threads:[~2007-09-17 11:13 UTC | newest]

Thread overview: 5+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2007-09-14 11:24 Increasing read request size in JFFS2 payagond
2007-09-14 11:38 ` Artem Bityutskiy
2007-09-14 12:50   ` Josh Boyer
2007-09-17 10:26     ` need patch for SST39VF301flash chips Bruno
2007-09-17 11:06       ` Josh Boyer

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox