* Weird status information on JFFS2 File System
@ 2002-02-14 19:29 Navin Boppuri
0 siblings, 0 replies; 8+ messages in thread
From: Navin Boppuri @ 2002-02-14 19:29 UTC (permalink / raw)
To: linux-mtd
Hello all,
I have a JFFS2 file system successfully installed on my flash device. Here is the output of the drivers during the Linux kernel boot. I have 1MB of flash partition installed as my flash device( I have two 16bit AMD flash devices connected on a 32 bit bus as one device).
sp flash device: 1000000 at 40000000
Amd/Fujitsu Extended Query Table v1.1 at 0x0040
number of CFI chips: 1
Creating 1 MTD partitions on "Service Processor flash device":
0x00f00000-0x01000000 : "Flash file system"
mtd: Giving out device 0 to Flash file system
After I mount my file system using the following command,
mount -t jffs2 /dev/mtdblock0 /mnt/jffs2
I confirm that the file system is mounted by looking at /proc/mounts.
root@sp_06:/mnt/jffs2# cat /proc/mounts
/dev/root / nfs rw,v2,rsize=4096,wsize=4096,hard,udp,nolock,addr=10.10.30.39 0 0
proc /proc proc rw 0 0
devpts /dev/pts devpts rw 0 0
/dev/mtdblock0 /mnt/jffs2 jffs2 rw 0 0
I now look at the disk usage using the df command
root@sp_06:/mnt/jffs2# df
Filesystem 1k-blocks Used Available Use% Mounted on
/dev/root 30237928 12751536 15950380 45% /
/dev/mtdblock0 1024 644 380 63% /mnt/jffs2
root@sp_06:/mnt/jffs2# ls -al
total 4
drwxr-xr-x 1 root root 0 Jan 1 00:00 .
drwxr-xr-x 3 1342 root 4096 Jan 25 2002 ..
I am surprised to see 63% usage without anything in the file system.
I now create the following three files doing this:
dd if=/dev/zero of=tmp.out bs=1M count=4 conv=sync
dd if=/dev/zero of=tmp2.out bs=1M count=4 conv=sync
dd if=/dev/zero of=tmp3.out bs=1M count=4 conv=sync
I run out of space while trying to create the tmp3.out command. But I am able to create almost 10MB of data in my 1MB file system.
root@sp_06:~# ls -al /mnt/jffs2
total 10152
drwxr-xr-x 1 root root 0 Jan 1 00:00 .
drwxr-xr-x 3 1342 root 4096 Jan 25 2002 ..
-rw-r--r-- 1 root root 4194304 Jan 1 03:01 tmp.out
-rw-r--r-- 1 root root 4194304 Jan 1 03:01 tmp2.out
-rw-r--r-- 1 root root 2002944 Jan 1 03:01 tmp3.out
root@sp_06:~# df
Filesystem 1k-blocks Used Available Use% Mounted on
/dev/root 30237928 12751536 15950380 45% /
/dev/mtdblock0 1024 932 92 92% /mnt/jffs2
I doing something really wrong here. What's going on? The kernel messages are all correct. Then how am I able to access 10MB of flash in this file system? I am only mapping the last 1MB of flash which is 16MB in size. And I am confused about the status of my file system each time I create files.
Can someone give me some pointers on this? Thank you.
Navin.
^ permalink raw reply [flat|nested] 8+ messages in thread
* RE: Weird status information on JFFS2 File System
@ 2002-02-14 20:57 Navin Boppuri
0 siblings, 0 replies; 8+ messages in thread
From: Navin Boppuri @ 2002-02-14 20:57 UTC (permalink / raw)
To: linux-mtd
Well, I guess I think I know what the problem is. I should not use /dev/zero to create files to test the file system size. I decided to use /dev/core instead and that actually creates the correct sized file.
But I still dont understand why the status of my file system is so screwed up. Even with an empty file sytem, it shows 64% used. And I am only able to write around 375K of data into this partition, which is actually 1MB.
Navin.
-----Original Message-----
From: Navin Boppuri
Sent: Thursday, February 14, 2002 1:29 PM
To: linux-mtd@lists.infradead.org
Subject: Weird status information on JFFS2 File System
Hello all,
I have a JFFS2 file system successfully installed on my flash device. Here is the output of the drivers during the Linux kernel boot. I have 1MB of flash partition installed as my flash device( I have two 16bit AMD flash devices connected on a 32 bit bus as one device).
sp flash device: 1000000 at 40000000
Amd/Fujitsu Extended Query Table v1.1 at 0x0040
number of CFI chips: 1
Creating 1 MTD partitions on "Service Processor flash device":
0x00f00000-0x01000000 : "Flash file system"
mtd: Giving out device 0 to Flash file system
After I mount my file system using the following command,
mount -t jffs2 /dev/mtdblock0 /mnt/jffs2
I confirm that the file system is mounted by looking at /proc/mounts.
root@sp_06:/mnt/jffs2# cat /proc/mounts
/dev/root / nfs rw,v2,rsize=4096,wsize=4096,hard,udp,nolock,addr=10.10.30.39 0 0
proc /proc proc rw 0 0
devpts /dev/pts devpts rw 0 0
/dev/mtdblock0 /mnt/jffs2 jffs2 rw 0 0
I now look at the disk usage using the df command
root@sp_06:/mnt/jffs2# df
Filesystem 1k-blocks Used Available Use% Mounted on
/dev/root 30237928 12751536 15950380 45% /
/dev/mtdblock0 1024 644 380 63% /mnt/jffs2
root@sp_06:/mnt/jffs2# ls -al
total 4
drwxr-xr-x 1 root root 0 Jan 1 00:00 .
drwxr-xr-x 3 1342 root 4096 Jan 25 2002 ..
I am surprised to see 63% usage without anything in the file system.
I now create the following three files doing this:
dd if=/dev/zero of=tmp.out bs=1M count=4 conv=sync
dd if=/dev/zero of=tmp2.out bs=1M count=4 conv=sync
dd if=/dev/zero of=tmp3.out bs=1M count=4 conv=sync
I run out of space while trying to create the tmp3.out command. But I am able to create almost 10MB of data in my 1MB file system.
root@sp_06:~# ls -al /mnt/jffs2
total 10152
drwxr-xr-x 1 root root 0 Jan 1 00:00 .
drwxr-xr-x 3 1342 root 4096 Jan 25 2002 ..
-rw-r--r-- 1 root root 4194304 Jan 1 03:01 tmp.out
-rw-r--r-- 1 root root 4194304 Jan 1 03:01 tmp2.out
-rw-r--r-- 1 root root 2002944 Jan 1 03:01 tmp3.out
root@sp_06:~# df
Filesystem 1k-blocks Used Available Use% Mounted on
/dev/root 30237928 12751536 15950380 45% /
/dev/mtdblock0 1024 932 92 92% /mnt/jffs2
I doing something really wrong here. What's going on? The kernel messages are all correct. Then how am I able to access 10MB of flash in this file system? I am only mapping the last 1MB of flash which is 16MB in size. And I am confused about the status of my file system each time I create files.
Can someone give me some pointers on this? Thank you.
Navin.
______________________________________________________
Linux MTD discussion mailing list
http://lists.infradead.org/mailman/listinfo/linux-mtd/
^ permalink raw reply [flat|nested] 8+ messages in thread
* RE: Weird status information on JFFS2 File System
@ 2002-02-14 22:26 Navin Boppuri
0 siblings, 0 replies; 8+ messages in thread
From: Navin Boppuri @ 2002-02-14 22:26 UTC (permalink / raw)
To: linux-mtd
I tried using the JFFS file system instead of JFFS2. Now I see that my file system size is 512Kb instead of 1MB. Also, I am only able to write to around 75% of the file sytem.
What's going on here? Can somone please give me some inputs on this. I am really new to JFFS and this stuff is driving me crazy.
Navin.
-----Original Message-----
From: Navin Boppuri
Sent: Thursday, February 14, 2002 2:57 PM
To: linux-mtd@lists.infradead.org
Subject: RE: Weird status information on JFFS2 File System
Well, I guess I think I know what the problem is. I should not use /dev/zero to create files to test the file system size. I decided to use /dev/core instead and that actually creates the correct sized file.
But I still dont understand why the status of my file system is so screwed up. Even with an empty file sytem, it shows 64% used. And I am only able to write around 375K of data into this partition, which is actually 1MB.
Navin.
-----Original Message-----
From: Navin Boppuri
Sent: Thursday, February 14, 2002 1:29 PM
To: linux-mtd@lists.infradead.org
Subject: Weird status information on JFFS2 File System
Hello all,
I have a JFFS2 file system successfully installed on my flash device. Here is the output of the drivers during the Linux kernel boot. I have 1MB of flash partition installed as my flash device( I have two 16bit AMD flash devices connected on a 32 bit bus as one device).
sp flash device: 1000000 at 40000000
Amd/Fujitsu Extended Query Table v1.1 at 0x0040
number of CFI chips: 1
Creating 1 MTD partitions on "Service Processor flash device":
0x00f00000-0x01000000 : "Flash file system"
mtd: Giving out device 0 to Flash file system
After I mount my file system using the following command,
mount -t jffs2 /dev/mtdblock0 /mnt/jffs2
I confirm that the file system is mounted by looking at /proc/mounts.
root@sp_06:/mnt/jffs2# cat /proc/mounts
/dev/root / nfs rw,v2,rsize=4096,wsize=4096,hard,udp,nolock,addr=10.10.30.39 0 0
proc /proc proc rw 0 0
devpts /dev/pts devpts rw 0 0
/dev/mtdblock0 /mnt/jffs2 jffs2 rw 0 0
I now look at the disk usage using the df command
root@sp_06:/mnt/jffs2# df
Filesystem 1k-blocks Used Available Use% Mounted on
/dev/root 30237928 12751536 15950380 45% /
/dev/mtdblock0 1024 644 380 63% /mnt/jffs2
root@sp_06:/mnt/jffs2# ls -al
total 4
drwxr-xr-x 1 root root 0 Jan 1 00:00 .
drwxr-xr-x 3 1342 root 4096 Jan 25 2002 ..
I am surprised to see 63% usage without anything in the file system.
I now create the following three files doing this:
dd if=/dev/zero of=tmp.out bs=1M count=4 conv=sync
dd if=/dev/zero of=tmp2.out bs=1M count=4 conv=sync
dd if=/dev/zero of=tmp3.out bs=1M count=4 conv=sync
I run out of space while trying to create the tmp3.out command. But I am able to create almost 10MB of data in my 1MB file system.
root@sp_06:~# ls -al /mnt/jffs2
total 10152
drwxr-xr-x 1 root root 0 Jan 1 00:00 .
drwxr-xr-x 3 1342 root 4096 Jan 25 2002 ..
-rw-r--r-- 1 root root 4194304 Jan 1 03:01 tmp.out
-rw-r--r-- 1 root root 4194304 Jan 1 03:01 tmp2.out
-rw-r--r-- 1 root root 2002944 Jan 1 03:01 tmp3.out
root@sp_06:~# df
Filesystem 1k-blocks Used Available Use% Mounted on
/dev/root 30237928 12751536 15950380 45% /
/dev/mtdblock0 1024 932 92 92% /mnt/jffs2
I doing something really wrong here. What's going on? The kernel messages are all correct. Then how am I able to access 10MB of flash in this file system? I am only mapping the last 1MB of flash which is 16MB in size. And I am confused about the status of my file system each time I create files.
Can someone give me some pointers on this? Thank you.
Navin.
______________________________________________________
Linux MTD discussion mailing list
http://lists.infradead.org/mailman/listinfo/linux-mtd/
______________________________________________________
Linux MTD discussion mailing list
http://lists.infradead.org/mailman/listinfo/linux-mtd/
^ permalink raw reply [flat|nested] 8+ messages in thread
* Re: Weird status information on JFFS2 File System
[not found] <D3A72C5007329A4F991C0DD87202259F096A6F@sekhmet.ad.newisys. com>
@ 2002-02-15 5:21 ` Chris Vagnini
0 siblings, 0 replies; 8+ messages in thread
From: Chris Vagnini @ 2002-02-15 5:21 UTC (permalink / raw)
To: Navin Boppuri; +Cc: linux-mtd
Hi Navin,
You're running into two documented behaviors of jffs2
1. jffs2 needs to keep 5 erase blocks of the flash free at all times. This
is so that there's always room for garbage collection. There have been
discussions of why this is necessary and how it might be reduced in the
past; search the email archive for details. These five blocks appear as
used space in the df command. Since you're showing 640K used, your flash
must use sectors of 64K each (x 2 chips).
2. jffs2 compresses the data to be stored the flash. Extremely
compressible data (like the contents of /dev/zero) occupy almost no space
when compressed, so you can fit a *lot* of it. Most data averages
compression more like 2:1, unlike the 30-1 or so you were seeing with
/dev/zero, so you can expect to store about 700KB in your flash which
starts out with 380K free. Of course this compression costs you processing
time, so you might want to turn it off while building jffs2 if speed is
more important than capacity.
--chris vagnini
At 02:29 PM 2/14/2002, Navin Boppuri wrote:
>Hello all,
>
>I have a JFFS2 file system successfully installed on my flash device. Here
>is the output of the drivers during the Linux kernel boot. I have 1MB of
>flash partition installed as my flash device( I have two 16bit AMD flash
>devices connected on a 32 bit bus as one device).
>
>
> sp flash device: 1000000 at 40000000
> Amd/Fujitsu Extended Query Table v1.1 at 0x0040
> number of CFI chips: 1
> Creating 1 MTD partitions on "Service Processor flash device":
> 0x00f00000-0x01000000 : "Flash file system"
> mtd: Giving out device 0 to Flash file system
>
>After I mount my file system using the following command,
> mount -t jffs2 /dev/mtdblock0 /mnt/jffs2
>
>I confirm that the file system is mounted by looking at /proc/mounts.
>
> root@sp_06:/mnt/jffs2# cat /proc/mounts
> /dev/root / nfs
> rw,v2,rsize=4096,wsize=4096,hard,udp,nolock,addr=10.10.30.39 0 0
> proc /proc proc rw 0 0
> devpts /dev/pts devpts rw 0 0
> /dev/mtdblock0 /mnt/jffs2 jffs2 rw 0 0
>
>I now look at the disk usage using the df command
>
> root@sp_06:/mnt/jffs2# df
> Filesystem 1k-blocks Used Available Use% Mounted on
> /dev/root 30237928 12751536 15950380 45% /
> /dev/mtdblock0 1024 644 380 63% /mnt/jffs2
>
> root@sp_06:/mnt/jffs2# ls -al
> total 4
> drwxr-xr-x 1 root root 0 Jan 1 00:00 .
> drwxr-xr-x 3 1342 root 4096 Jan 25 2002 ..
>
>I am surprised to see 63% usage without anything in the file system.
>
>I now create the following three files doing this:
>
> dd if=/dev/zero of=tmp.out bs=1M count=4 conv=sync
> dd if=/dev/zero of=tmp2.out bs=1M count=4 conv=sync
> dd if=/dev/zero of=tmp3.out bs=1M count=4 conv=sync
>
>I run out of space while trying to create the tmp3.out command. But I am
>able to create almost 10MB of data in my 1MB file system.
>
> root@sp_06:~# ls -al /mnt/jffs2
> total 10152
> drwxr-xr-x 1 root root 0 Jan 1 00:00 .
> drwxr-xr-x 3 1342 root 4096 Jan 25 2002 ..
> -rw-r--r-- 1 root root 4194304 Jan 1 03:01 tmp.out
> -rw-r--r-- 1 root root 4194304 Jan 1 03:01 tmp2.out
> -rw-r--r-- 1 root root 2002944 Jan 1 03:01 tmp3.out
>
> root@sp_06:~# df
> Filesystem 1k-blocks Used Available Use% Mounted on
> /dev/root 30237928 12751536 15950380 45% /
> /dev/mtdblock0 1024 932 92 92% /mnt/jffs2
>
>I doing something really wrong here. What's going on? The kernel messages
>are all correct. Then how am I able to access 10MB of flash in this file
>system? I am only mapping the last 1MB of flash which is 16MB in size. And
>I am confused about the status of my file system each time I create files.
>
>Can someone give me some pointers on this? Thank you.
>
>Navin.
>
>
>______________________________________________________
>Linux MTD discussion mailing list
>http://lists.infradead.org/mailman/listinfo/linux-mtd/
^ permalink raw reply [flat|nested] 8+ messages in thread
* RE: Weird status information on JFFS2 File System
@ 2002-02-15 9:49 Ashok M Padmanaban
0 siblings, 0 replies; 8+ messages in thread
From: Ashok M Padmanaban @ 2002-02-15 9:49 UTC (permalink / raw)
To: MTD for Linux, navin.boppuri
Hi Navin
Your Jffs2 file system size is 1MB
Jffs2 reserves around 5 sectors( erase sectors) for garbage collection.
So even when ur FS is empty , command df will show that 5 erase sectors
have been already occupied
Regards
ashok
Navin Boppuri wrote:
> Well, I guess I think I know what the problem is. I should not use
/dev/zero to create files to test the file system size. I decided to use
/dev/core instead and that actually creates the correct sized file.
>
> But I still dont understand why the status of my file system is so
screwed up. Even with an empty file sytem, it shows 64% used. And I am
only able to write around 375K of data into this partition, which is
actually 1MB.
>
> Navin.
>
> -----Original Message-----
> From: Navin Boppuri
> Sent: Thursday, February 14, 2002 1:29 PM
> To: linux-mtd@lists.infradead.org
> Subject: Weird status information on JFFS2 File System
>
> Hello all,
>
> I have a JFFS2 file system successfully installed on my flash device.
Here is the output of the drivers during the Linux kernel boot. I have
1MB of flash partition installed as my flash device( I have two 16bit
AMD flash devices connected on a 32 bit bus as one device).
>
> sp flash device: 1000000 at 40000000
> Amd/Fujitsu Extended Query Table v1.1 at 0x0040
> number of CFI chips: 1
> Creating 1 MTD partitions on "Service Processor flash device":
> 0x00f00000-0x01000000 : "Flash file system"
> mtd: Giving out device 0 to Flash file system
>
> After I mount my file system using the following command,
> mount -t jffs2 /dev/mtdblock0 /mnt/jffs2
>
> I confirm that the file system is mounted by looking at /proc/mounts.
>
> root@sp_06:/mnt/jffs2# cat /proc/mounts
> /dev/root / nfs
rw,v2,rsize=4096,wsize=4096,hard,udp,nolock,addr=10.10.30.39 0 0
> proc /proc proc rw 0 0
> devpts /dev/pts devpts rw 0 0
> /dev/mtdblock0 /mnt/jffs2 jffs2 rw 0 0
>
> I now look at the disk usage using the df command
>
> root@sp_06:/mnt/jffs2# df
> Filesystem 1k-blocks Used Available Use%
Mounted on
> /dev/root 30237928 12751536 15950380 45% /
> /dev/mtdblock0 1024 644 380 63%
/mnt/jffs2
>
> root@sp_06:/mnt/jffs2# ls -al
> total 4
> drwxr-xr-x 1 root root 0 Jan 1 00:00 .
> drwxr-xr-x 3 1342 root 4096 Jan 25 2002 ..
>
> I am surprised to see 63% usage without anything in the file system.
>
> I now create the following three files doing this:
>
> dd if=/dev/zero of=tmp.out bs=1M count=4 conv=sync
> dd if=/dev/zero of=tmp2.out bs=1M count=4 conv=sync
> dd if=/dev/zero of=tmp3.out bs=1M count=4 conv=sync
>
> I run out of space while trying to create the tmp3.out command. But I
am able to create almost 10MB of data in my 1MB file system.
>
> root@sp_06:~# ls -al /mnt/jffs2
> total 10152
> drwxr-xr-x 1 root root 0 Jan 1 00:00 .
> drwxr-xr-x 3 1342 root 4096 Jan 25 2002 ..
> -rw-r--r-- 1 root root 4194304 Jan 1 03:01
tmp.out
> -rw-r--r-- 1 root root 4194304 Jan 1 03:01
tmp2.out
> -rw-r--r-- 1 root root 2002944 Jan 1 03:01
tmp3.out
>
> root@sp_06:~# df
> Filesystem 1k-blocks Used Available Use%
Mounted on
> /dev/root 30237928 12751536 15950380 45% /
> /dev/mtdblock0 1024 932 92 92%
/mnt/jffs2
>
> I doing something really wrong here. What's going on? The kernel
messages are all correct. Then how am I able to access 10MB of flash in
this file system? I am only mapping the last 1MB of flash which is 16MB
in size. And I am confused about the status of my file system each time
I create files.
>
> Can someone give me some pointers on this? Thank you.
>
> Navin.
>
> ______________________________________________________
> Linux MTD discussion mailing list
> http://lists.infradead.org/mailman/listinfo/linux-mtd/
>
> ______________________________________________________
> Linux MTD discussion mailing list
> http://lists.infradead.org/mailman/listinfo/linux-mtd/
^ permalink raw reply [flat|nested] 8+ messages in thread
* RE: Weird status information on JFFS2 File System
@ 2002-02-15 16:06 Navin Boppuri
2002-02-15 16:10 ` David Woodhouse
0 siblings, 1 reply; 8+ messages in thread
From: Navin Boppuri @ 2002-02-15 16:06 UTC (permalink / raw)
To: Chris Vagnini; +Cc: linux-mtd
Well, I should have read more about the JFFS2 file system. But your answers explain everything.
Is there some way I can change the 5 erase blocks default to a smaller size? What are the implications of doing that?
Navin.
-----Original Message-----
From: Chris Vagnini [mailto:cvagnini@bigfoot.com]
Sent: Thursday, February 14, 2002 11:21 PM
To: Navin Boppuri
Cc: linux-mtd@lists.infradead.org
Subject: Re: Weird status information on JFFS2 File System
Hi Navin,
You're running into two documented behaviors of jffs2
1. jffs2 needs to keep 5 erase blocks of the flash free at all times. This
is so that there's always room for garbage collection. There have been
discussions of why this is necessary and how it might be reduced in the
past; search the email archive for details. These five blocks appear as
used space in the df command. Since you're showing 640K used, your flash
must use sectors of 64K each (x 2 chips).
2. jffs2 compresses the data to be stored the flash. Extremely
compressible data (like the contents of /dev/zero) occupy almost no space
when compressed, so you can fit a *lot* of it. Most data averages
compression more like 2:1, unlike the 30-1 or so you were seeing with
/dev/zero, so you can expect to store about 700KB in your flash which
starts out with 380K free. Of course this compression costs you processing
time, so you might want to turn it off while building jffs2 if speed is
more important than capacity.
--chris vagnini
At 02:29 PM 2/14/2002, Navin Boppuri wrote:
>Hello all,
>
>I have a JFFS2 file system successfully installed on my flash device. Here
>is the output of the drivers during the Linux kernel boot. I have 1MB of
>flash partition installed as my flash device( I have two 16bit AMD flash
>devices connected on a 32 bit bus as one device).
>
>
> sp flash device: 1000000 at 40000000
> Amd/Fujitsu Extended Query Table v1.1 at 0x0040
> number of CFI chips: 1
> Creating 1 MTD partitions on "Service Processor flash device":
> 0x00f00000-0x01000000 : "Flash file system"
> mtd: Giving out device 0 to Flash file system
>
>After I mount my file system using the following command,
> mount -t jffs2 /dev/mtdblock0 /mnt/jffs2
>
>I confirm that the file system is mounted by looking at /proc/mounts.
>
> root@sp_06:/mnt/jffs2# cat /proc/mounts
> /dev/root / nfs
> rw,v2,rsize=4096,wsize=4096,hard,udp,nolock,addr=10.10.30.39 0 0
> proc /proc proc rw 0 0
> devpts /dev/pts devpts rw 0 0
> /dev/mtdblock0 /mnt/jffs2 jffs2 rw 0 0
>
>I now look at the disk usage using the df command
>
> root@sp_06:/mnt/jffs2# df
> Filesystem 1k-blocks Used Available Use% Mounted on
> /dev/root 30237928 12751536 15950380 45% /
> /dev/mtdblock0 1024 644 380 63% /mnt/jffs2
>
> root@sp_06:/mnt/jffs2# ls -al
> total 4
> drwxr-xr-x 1 root root 0 Jan 1 00:00 .
> drwxr-xr-x 3 1342 root 4096 Jan 25 2002 ..
>
>I am surprised to see 63% usage without anything in the file system.
>
>I now create the following three files doing this:
>
> dd if=/dev/zero of=tmp.out bs=1M count=4 conv=sync
> dd if=/dev/zero of=tmp2.out bs=1M count=4 conv=sync
> dd if=/dev/zero of=tmp3.out bs=1M count=4 conv=sync
>
>I run out of space while trying to create the tmp3.out command. But I am
>able to create almost 10MB of data in my 1MB file system.
>
> root@sp_06:~# ls -al /mnt/jffs2
> total 10152
> drwxr-xr-x 1 root root 0 Jan 1 00:00 .
> drwxr-xr-x 3 1342 root 4096 Jan 25 2002 ..
> -rw-r--r-- 1 root root 4194304 Jan 1 03:01 tmp.out
> -rw-r--r-- 1 root root 4194304 Jan 1 03:01 tmp2.out
> -rw-r--r-- 1 root root 2002944 Jan 1 03:01 tmp3.out
>
> root@sp_06:~# df
> Filesystem 1k-blocks Used Available Use% Mounted on
> /dev/root 30237928 12751536 15950380 45% /
> /dev/mtdblock0 1024 932 92 92% /mnt/jffs2
>
>I doing something really wrong here. What's going on? The kernel messages
>are all correct. Then how am I able to access 10MB of flash in this file
>system? I am only mapping the last 1MB of flash which is 16MB in size. And
>I am confused about the status of my file system each time I create files.
>
>Can someone give me some pointers on this? Thank you.
>
>Navin.
>
>
>______________________________________________________
>Linux MTD discussion mailing list
>http://lists.infradead.org/mailman/listinfo/linux-mtd/
^ permalink raw reply [flat|nested] 8+ messages in thread
* Re: Weird status information on JFFS2 File System
2002-02-15 16:06 Weird status information on JFFS2 File System Navin Boppuri
@ 2002-02-15 16:10 ` David Woodhouse
0 siblings, 0 replies; 8+ messages in thread
From: David Woodhouse @ 2002-02-15 16:10 UTC (permalink / raw)
To: Navin Boppuri; +Cc: Chris Vagnini, linux-mtd
navin.boppuri@newisys.com said:
> Is there some way I can change the 5 erase blocks default to a
> smaller size? What are the implications of doing that?
grep RESERVED fs/jffs2/nodelist.h
The danger in reducing it too far is that you'll run out of free space
entirely, so that garbage collection cannot continue and your file system
gets entirely stuck. We ought to be able to reduce it to only a block or
two, but the code wants auditing for safety first - we have to ensure that
we never expand while garbage-collecting.
--
dwmw2
^ permalink raw reply [flat|nested] 8+ messages in thread
* RE: Weird status information on JFFS2 File System
@ 2002-02-15 16:45 Navin Boppuri
0 siblings, 0 replies; 8+ messages in thread
From: Navin Boppuri @ 2002-02-15 16:45 UTC (permalink / raw)
To: David Woodhouse; +Cc: linux-mtd
I have two uniform sector (64K each 16 bit) Flash devices connected to a 32 bit bus. So, my sector size is actually 128K. For a 1MB file system, what should be the optimal size of the reserved space for garbage collection? I only have 8 sectors for 1MB out of which 5 sectors are already being taken up by GC. Do I really need 5*128K GC area for 3*128K available space left on the flash?
Thanks a lot.
Navin.
-----Original Message-----
From: David Woodhouse [mailto:dwmw2@infradead.org]
Sent: Friday, February 15, 2002 10:10 AM
To: Navin Boppuri
Cc: Chris Vagnini; linux-mtd@lists.infradead.org
Subject: Re: Weird status information on JFFS2 File System
navin.boppuri@newisys.com said:
> Is there some way I can change the 5 erase blocks default to a
> smaller size? What are the implications of doing that?
grep RESERVED fs/jffs2/nodelist.h
The danger in reducing it too far is that you'll run out of free space
entirely, so that garbage collection cannot continue and your file system
gets entirely stuck. We ought to be able to reduce it to only a block or
two, but the code wants auditing for safety first - we have to ensure that
we never expand while garbage-collecting.
--
dwmw2
^ permalink raw reply [flat|nested] 8+ messages in thread
end of thread, other threads:[~2002-02-15 16:34 UTC | newest]
Thread overview: 8+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2002-02-15 16:06 Weird status information on JFFS2 File System Navin Boppuri
2002-02-15 16:10 ` David Woodhouse
-- strict thread matches above, loose matches on Subject: below --
2002-02-15 16:45 Navin Boppuri
2002-02-15 9:49 Ashok M Padmanaban
[not found] <D3A72C5007329A4F991C0DD87202259F096A6F@sekhmet.ad.newisys. com>
2002-02-15 5:21 ` Chris Vagnini
2002-02-14 22:26 Navin Boppuri
2002-02-14 20:57 Navin Boppuri
2002-02-14 19:29 Navin Boppuri
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox