jffs2

jffs2

[Kári Davíðsson]

Hi,

Are there any known issues of running programs from a jffs2 partition?

I am attempting to run a program from a jffs2 partition (not root
partition which is mounted on /dev/ram)
and the program seem to be running extremely slow and what is worse it
seem to block the kernel at the 
same time, i.e. I can not ping the board nor telnet to it.

If I wait long enough I get output from the program on the console but
everything is very very slow.....

Ideas?

K.D.

Re: jffs2

[Vipin Malik]

Kári Davíðsson wrote:

> Are there any known issues of running programs from a jffs2 partition?

Not that I know of.


> I am attempting to run a program from a jffs2 partition (not root
> partition which is mounted on /dev/ram)

Interesting choice of mount dir names (why "ram"?) :)


> and the program seem to be running extremely slow and what is worse it
> seem to block the kernel at the
> same time, i.e. I can not ping the board nor telnet to it.

Not being able to ping it is bad. Not being able to telnet to it may
depend on how much CPU resources are available
to run the telnet daemon.

>
>
> If I wait long enough I get output from the program on the console but
> everything is very very slow.....
>

What processor are you using? How big is your JFFS2 partition and what is
this program that you are running doing?

Remember that JFFS2 needs to go GC when the fs gets full (or near full if
the background GC task is running). If your program
is reading/writing from/to the JFFS2 fs it will get blocked. Are you
generating a lot of data to the JFFS2 fs?

Vipin

JFFS2

[Herman Oosthuysen]

Hi guys,

Something is not clear to me:
In order to have JFFS2 support, do I have to compile in the JFFS stuff as well?

I modified the following lines in fs/config.in:
## Herman - Add JFFS2 Support
tristate 'Journaling FLASH filesystem (JFFS) support' CONFIG_JFFS_FS
tristate 'Journalling Flash filesystem (JFFS2) support' CONFIG_JFFS2_FS
int 'Journalling Flash filesystem (JFFS2) Debug Verbosity' CONFIG_JFFS2_FS_DEBUG

Cheers,
-- 
Herman Oosthuysen
Phone: 403+569-5687
E-mail: Herman@WirelessNetworksInc.com

Re: JFFS2

[Vipin Malik]

No, JFFS2 is completely separate from JFFS(1). You do of course need MTD support
for both.

Vipin


Herman Oosthuysen wrote:

> Hi guys,
>
> Something is not clear to me:
> In order to have JFFS2 support, do I have to compile in the JFFS stuff as well?
>
> I modified the following lines in fs/config.in:
> ## Herman - Add JFFS2 Support
> tristate 'Journaling FLASH filesystem (JFFS) support' CONFIG_JFFS_FS
> tristate 'Journalling Flash filesystem (JFFS2) support' CONFIG_JFFS2_FS
> int 'Journalling Flash filesystem (JFFS2) Debug Verbosity' CONFIG_JFFS2_FS_DEBUG
>
> Cheers,
> --
> Herman Oosthuysen
> Phone: 403+569-5687
> E-mail: Herman@WirelessNetworksInc.com
>
> ______________________________________________________
> Linux MTD discussion mailing list
> http://lists.infradead.org/mailman/listinfo/linux-mtd/

RE: jffs2

[Kári Davíðsson]

Hi,

> -----Original Message-----
> From: Vipin Malik [mailto:vipin.malik@daniel.com]
> Sent: 1. júní 2001 14:56
> To: Kári Davíðsson
> Cc: linux-mtd@lists.infradead.org
> Subject: Re: jffs2
> 
> 
> Kári Davíðsson wrote:
> 
> > Are there any known issues of running programs from a jffs2 
> partition?
> 
> Not that I know of.
> 
> 
> > I am attempting to run a program from a jffs2 partition (not root
> > partition which is mounted on /dev/ram)
> 
> Interesting choice of mount dir names (why "ram"?) :)

Well that is the initrd. I am limited in flash and I found that
compressing
a e2fs and mount it through the initrd mechanism gives me more space in
the
flash for a jffs2 partition that keeps the stuff I need to write to
flash 
during operation of board (logs, frequently changing configuration
files,
upgradable programs and drivers, etc).

> 
> 
> > and the program seem to be running extremely slow and what 
> is worse it
> > seem to block the kernel at the
> > same time, i.e. I can not ping the board nor telnet to it.
> 
> Not being able to ping it is bad. Not being able to telnet to it may
> depend on how much CPU resources are available
> to run the telnet daemon.

Yes, actually I found out it was a combination of two "bugs", one was
that I was running a very big (unstriped C++ program) in the range of
3.5 MB from the jffs2 partition, so I guess just uncompressing that kind
of file
takes its time. The second bug was another driver that disables
interrupts, was hanging.....

> 
> >
> >
> > If I wait long enough I get output from the program on the 
> console but
> > everything is very very slow.....
> >
> 
> What processor are you using? How big is your JFFS2 partition 

The processor is mpc823 running at 50Mhz, no powerhorse but a resonable
CPU for the things we are dooing.

> and what is

The jffs2 partition is 1.75 MB.

> this program that you are running doing?

Data acquizition.

>
> Remember that JFFS2 needs to go GC when the fs gets full (or 
> near full if
> the background GC task is running). If your program

How full is "near full" for the jffs2?

> is reading/writing from/to the JFFS2 fs it will get blocked. Are you
> generating a lot of data to the JFFS2 fs?

After I striped the executable, it is around 350kb everything is running
much better.

> 
> Vipin

Thanks,

K.D.

jffs2

[Kornel Masłowski]

I'm not sure how jffs2 works.
I have jffs2 on  8bit NAND flash (page 512 B).
I want to append record 32B to the file on flash which current size is 10kB.
Flash device is almost empty.
How many pages will be written by jffs2? 

Re: jffs2

[Artem B. Bityuckiy]

Kornel Masłowski wrote:
> I'm not sure how jffs2 works.
> I have jffs2 on  8bit NAND flash (page 512 B).
> I want to append record 32B to the file on flash which current size is 
> 10kB.
> Flash device is almost empty.
> How many pages will be written by jffs2?
> 
> 
It is hard to calculate because of compression. Moreover, if your file 
was created by
seek to "nowhere" it will cosist of only one "hole" node and fit into 
one NAND page only.

If there is no compression, the exact number of pages also depends on 
how your file was created. Also, there is some platform dependency - 
does you platform has 4K RAM pages? - but this is minor since most 
platforms have 4K RAM pages.

If your 10K file was created by one go and there is no compression, the 
file will consist of 3 JFFS2 nodes (4K + 4K + 2K) and fit into 9 + 9 + 5 
= 23 NAND pages (4K data = 4 NAND pages and one is needed for node 
header; the rest of page won't be used since the next node isn't fit to it).

When you append 32 bytes, one more page will be used (24 total). But if 
your append just after 10K file creation (i.e, the JFFS2 write buffer 
hasn't yet been written to the NAND flash media), the 32B appendix node 
will be added to the 23-th page and there will be only 23 used pages.

If the compression is on, there ought to be fewer used NAND pages.

Ok, this is theory (I didn't check) and my own understanding of how 
JFFS2 works. I might make mistakes :-)

You may check this by creating/appending your file and then calculating 
the numbur of used pages. The nanddump utility may be used.

-- 
Best Regards,
Artem B. Bityuckiy,
St.-Petersburg, Russia.

Re: jffs2

[Artem B. Bityuckiy]

Also, the direntry node and probably the root directory inode node will 
also add 1-2 NAND pages. Thus, 24-27 pages if compression is disabled...

Artem B. Bityuckiy wrote:
> Kornel Masłowski wrote:
> 
>> I'm not sure how jffs2 works.
>> I have jffs2 on  8bit NAND flash (page 512 B).
>> I want to append record 32B to the file on flash which current size is 
>> 10kB.
>> Flash device is almost empty.
>> How many pages will be written by jffs2?
>>
>>
> It is hard to calculate because of compression. Moreover, if your file 
> was created by
> seek to "nowhere" it will cosist of only one "hole" node and fit into 
> one NAND page only.
> 
> If there is no compression, the exact number of pages also depends on 
> how your file was created. Also, there is some platform dependency - 
> does you platform has 4K RAM pages? - but this is minor since most 
> platforms have 4K RAM pages.
> 
> If your 10K file was created by one go and there is no compression, the 
> file will consist of 3 JFFS2 nodes (4K + 4K + 2K) and fit into 9 + 9 + 5 
> = 23 NAND pages (4K data = 4 NAND pages and one is needed for node 
> header; the rest of page won't be used since the next node isn't fit to 
> it).
> 
> When you append 32 bytes, one more page will be used (24 total). But if 
> your append just after 10K file creation (i.e, the JFFS2 write buffer 
> hasn't yet been written to the NAND flash media), the 32B appendix node 
> will be added to the 23-th page and there will be only 23 used pages.
> 
> If the compression is on, there ought to be fewer used NAND pages.
> 
> Ok, this is theory (I didn't check) and my own understanding of how 
> JFFS2 works. I might make mistakes :-)
> 
> You may check this by creating/appending your file and then calculating 
> the numbur of used pages. The nanddump utility may be used.
> 

-- 
Best Regards,
Artem B. Bityuckiy,
St.-Petersburg, Russia.

jffs2

[trimarchi]

Hi all,
I need to know the order of magnitude of time spent to mount a NAND of 128MB
using a jffs2 filesystem.

Regards
Michael


----------------------------------------------------------------
This message was sent using IMP, the Internet Messaging Program.

Re: jffs2

[Josh Boyer]

On 3/7/06, trimarchi@gandalf.sssup.it <trimarchi@gandalf.sssup.it> wrote:
> Hi all,
> I need to know the order of magnitude of time spent to mount a NAND of 128MB
> using a jffs2 filesystem.

That's a pretty vague question, so I'll give you a vague answer:

It will take less than 3 years to mount a NAND of 126MiB using a JFFS2
filesystem.

If you specify more information, perhaps you'll get a better answer. 
Things like the type of processor, the type of NAND flash, how full
the filesystem is, whether or not the Summary option is on, etc. all
have a direct impact to your question.

josh

Re: jffs2

[Michael Trimarchi]

s3c2440
128MB nand
full file system

jffs2 and jffs2+Erase Block Summary (EBS)

regards
Michael


Josh Boyer wrote:

>On 3/7/06, trimarchi@gandalf.sssup.it <trimarchi@gandalf.sssup.it> wrote:
>  
>
>>Hi all,
>>I need to know the order of magnitude of time spent to mount a NAND of 128MB
>>using a jffs2 filesystem.
>>    
>>
>
>That's a pretty vague question, so I'll give you a vague answer:
>
>It will take less than 3 years to mount a NAND of 126MiB using a JFFS2
>filesystem.
>
>If you specify more information, perhaps you'll get a better answer. 
>Things like the type of processor, the type of NAND flash, how full
>the filesystem is, whether or not the Summary option is on, etc. all
>have a direct impact to your question.
>
>josh
>
>______________________________________________________
>Linux MTD discussion mailing list
>http://lists.infradead.org/mailman/listinfo/linux-mtd/
>
>  
>

Re: jffs2

[Ferenc Havasi]

Michael Trimarchi wrote:

> s3c2440
> 128MB nand
> full file system
>
> jffs2 and jffs2+Erase Block Summary (EBS)

Do you use sumtool utility after mkfs.jffs2?

Ferenc

jffs2

[Katey Yu]

Hello,

What's wrong?


Creating 5 MTD partitions on " Intel flash memory":
0x00000000-0x00010000 : "bootloader"
0x00010000-0x000b0000 : "kernel"
0x000b0000-0x00200000 : "filesystem"
0x00004000-0x00008000 : "config"
0x00000000-0x00200000 : "complete flash"

# mount -t jffs2  /dev/mtdblock3 /var/tmp
jffs2_scan_eraseblock(): Magic bitmask 0x1985 not found at 0x00000004:
0x000d instead
jffs2_scan_eraseblock(): Magic bitmask 0x1985 not found at 0x00000008:
0x300c instead
jffs2_scan_eraseblock(): Magic bitmask 0x1985 not found at 0x0000000c:
0x0008 instead
jffs2_scan_eraseblock(): Magic bitmask 0x1985 not found at 0x00000010:
0x000a instead
jffs2_scan_eraseblock(): Magic bitmask 0x1985 not found at 0x00000014:
0x300d instead
jffs2_scan_eraseblock(): Magic bitmask 0x1985 not found at 0x00000018:
0x0006 instead
jffs2_scan_eraseblock(): Magic bitmask 0x1985 not found at 0x0000001c:
0x0007 instead
jffs2_scan_eraseblock(): Magic bitmask 0x1985 not found at 0x00000020:
0x0004 instead
jffs2_scan_eraseblock(): Magic bitmask 0x1985 not found at 0x00000024:
0x1006 instead
jffs2_scan_eraseblock(): Magic bitmask 0x1985 not found at 0x00000028:
0x2004 instead
Further such events for this erase block will not be printed
JFFS2: Erase block at 0x00000000 is not formatted. It will be erased
jffs2_scan_eraseblock(): Magic bitmask 0x1985 not found at 0x00002000:
0xc00d instead
jffs2_scan_eraseblock(): Magic bitmask 0x1985 not found at 0x00002004:
0xd870 instead
jffs2_scan_eraseblock(): Magic bitmask 0x1985 not found at 0x00002008:
0x6000 instead
jffs2_scan_eraseblock(): Magic bitmask 0x1985 not found at 0x0000200c:
0x300c instead
jffs2_scan_eraseblock(): Magic bitmask 0x1985 not found at 0x00002010:
0xb004 instead
jffs2_scan_eraseblock(): Magic bitmask 0x1985 not found at 0x00002014:
0x5003 instead
jffs2_scan_eraseblock(): Magic bitmask 0x1985 not found at 0x00002018:
0x0002 instead
jffs2_scan_eraseblock(): Magic bitmask 0x1985 not found at 0x0000201c:
0x001c instead
jffs2_scan_eraseblock(): Magic bitmask 0x1985 not found at 0x00002020:
0x3038 instead
jffs2_scan_eraseblock(): Magic bitmask 0x1985 not found at 0x00002024:
0x20b4 instead
Further such events for this erase block will not be printed
JFFS2: Erase block at 0x00002000 is not formatted. It will be erased
Cowardly refusing to erase blocks on filesystem with no valid JFFS2
nodes
mount: wrong fs type, bad option, bad superblock on /dev/mtdblock3,
       or too many mounted file systems