[Yaffs-archive] Re: Performance

Top Page
Attachments:
Message as email
+ (text/plain)
Delete this message
Reply to this message
Author: Luc Van Oostenryck
Date:  
To: YAFFS list
Subject: [Yaffs-archive] Re: Performance
Christian Gan wrote:
> By the way, what kind of performance is everybody getting from YAFFS/NAND
> MTD on Linux?
>
> I found that the NAND MTD was really slow and doing things that wasted time
> (ie. resets for every command, using for loops for ram copies to the data
> buf, etc). So I had to rewrite portions that appeared the slowest on the
> logic analyzer.
>
> So I'm curious to see what performance others are getting (can you also
> include your proc and bus speeds?).
>
> Thanks!
>
> Christian Gan
>
>


Christian,

Here is an extract of one of my previous mail:
>
> For the moment what annoy me is the MTD layer that doesn't offer a good
> interface for NAND and page access. Next week, I will probably write a
> small MTD-like block device that fit better to NAND access (only page
> access with HW ECC).
> In a second time, I will investigate the possiblity to use
> generic_file_read() and maybe generic_file_write() so we can profit from
> the disk cache.
>
> Best regards
>
> Luc Van Oosternyck
>


IOW I have exactly the problems concerning the NAND MTD.

For performance result (hey show yours, I will show mine :-)
      Read:    185Kb/s
      Write:      85Kb/s
      Write:      175Kb/s        when removing the yaffs code that read back
                each chunk to verify if the write is OK.
      Board is a prototype ARM720 running at 24MHz with a 16bit 12MHz bus, with the SW ECC.
      Definitive version will be faster and will also use HW ECC.


Cheers






---------------------------------------------------------------------------------------
This mailing list is hosted by Toby Churchill open software (www.toby-churchill.org).
If mailing list membership is no longer wanted you can remove yourself from the list by
sending an email to with the text "unsubscribe"
(without the quotes) as the subject.