2009-09-24 17:01:02     NAND flash problems when moving to new disty

Document created by Aaronwu Employee on Aug 19, 2013
Version 1Show Document
  • View in full screen mode

2009-09-24 17:01:02     NAND flash problems when moving to new disty

Jay Ku (UNITED STATES)

Message: 80401   

 

I'm brining my custom BF532 board up on 2009R1-RC6.  I have a 4Gbit (512Mbyte) NAND flash that I had been using with the 2008R1.5 distribution.  I can mount it, read/write, etc - seems to work fine.  When I run bonnie++, I get:

 

root:/mnt> bonnie++ -u root -d /mnt

Using uid:0, gid:0.

Writing a byte at a time...done

Writing intelligently...done

Rewriting...done

Reading a byte at a time...done

Reading intelligently...done

start 'em...done...done...done...done...done...

Create files in sequential order...save exit: isCheckpointed 1

done.

Stat files in sequential order...done.

Delete files in sequential order...==>> yaffs bug: fs/yaffs2/yaffs_guts.c 6832

==>> yaffs bug: fs/yaffs2/yaffs_guts.c 6759

==>> yaffs bug: fs/yaffs2/yaffs_guts.c 6759

save exit: isCheckpointed 1

done.

Create files in random order...save exit: isCheckpointed 1

done.

Stat files in random order...done.

Delete files in random order...==>> yaffs bug: fs/yaffs2/yaffs_guts.c 6832

==>> yaffs bug: fs/yaffs2/yaffs_guts.c 6759

==>> yaffs bug: fs/yaffs2/yaffs_guts.c 6759

save exit: isCheckpointed 1

done.

Version  1.94       ------Sequential Output------ --Sequential Input- --Random-

Concurrency   1     -Per Chr- --Block-- -Rewrite- -Per Chr- --Block-- --Seeks--

Machine        Size K/sec %CP K/sec %CP K/sec %CP K/sec %CP K/sec %CP  /sec %CP

blackfin       300M    45  99  2772  89  1533  99    84  99  3102  86 253.4  99

Latency               220ms     720ms   40000us     128ms     655ms     539ms

Version  1.94       ------Sequential Create------ --------Random Create--------

blackfin            -Create-- --Read--- -Delete-- -Create-- --Read--- -Delete--

              files  /sec %CP  /sec %CP  /sec %CP  /sec %CP  /sec %CP  /sec %CP

                 16    83  99  3022  99   336  99    82  99  6272 100   169  99

Latency             28000us   24000us   27999us   28000us    4000us   19999us

1.93c,1.94,blackfin,1,1167609674,300M,,45,99,2772,89,1533,99,84,99,3102,86,253.4,99,16,,,,,83,99,3022,99,336,99,82,99,6272,100,169,99,220ms,720ms,40000us,128ms,655ms,539ms,28000us,24000us,27999us,28000us,4000us,19999us

root:/mnt>

 

Seems to complete, but any idea what that yaffs message is all about?  The block throughputs seem reasonable, but I'm not sure if those latencies make sense (or what they really mean).

 

Something else I've noticed with the new distribution - when I try to use the new flash_eraseall, I get:

 

root:/> flash_eraseall /dev/mtdblock1

mtdblock_open

ok

flmst_drbslalock_release

k1/dev/mtobko

   unable to get MTD device info

root:/>

 

(I'm running this before I've tried to mount mtdblock1.)

 

Any thoughts?

 

- jay

QuoteReplyEditDelete

 

 

2009-09-24 17:17:36     Re: NAND flash problems when moving to new disty

Mike Frysinger (UNITED STATES)

Message: 80402   

 

using flash commands on the mtdblocks makes no sense.  those are block interfaces which handle the read/write/erase cycles for you.  use the flash commands on the simple mtd node like /dev/mtd1.

QuoteReplyEditDelete

 

 

2009-09-24 18:00:23     Re: NAND flash problems when moving to new disty

Jay Ku (UNITED STATES)

Message: 80403   

 

My mistake... I hadn't enabled 'Direct char device access to MTD devices', so I didn't have /dev/mtdX (and I just used what I saw)... flash_eraseall /dev/mtd1 works fine.

 

I still get the same 'yaffs bug' messages when I run bonnie++, though.

Attachments

    Outcomes