headers
jg zhao | 1 Apr 2002 03:29
Picon
Favicon

About 28f320c3

Hi !    
  I have some questions about the TE28F320C3TA110. I
configed it as a intel CFI chip. The size of it is
4M,and the address is from 1000000 to 1400000. 1100000
to 1200000 is my kernel code and I want to put my jffs
to the 1300000-1380000. My kernel burned and running
is OK. But when I add the MTD into it ,after this
message:

  physmap flash device: 80000 at 1300000  

my board was dead. The  kernel is running from the
flash . The code and jffs is on the same chip.Did it
work?Thanks for giving me some suggestion!

_________________________________________________________
Do You Yahoo!? 
到世界杯主题公园玩一玩，赢取世界杯门票乐一乐。
http://cn.worldcup.yahoo.com/
Permalink | Reply | 1 comment |
headers
Jörn Engel | 1 Apr 2002 10:31
Picon

Re: About 28f320c3

Hi!

>   I have some questions about the TE28F320C3TA110. I
> configed it as a intel CFI chip. The size of it is
> 4M,and the address is from 1000000 to 1400000. 1100000
> to 1200000 is my kernel code and I want to put my jffs
> to the 1300000-1380000. My kernel burned and running
> is OK. But when I add the MTD into it ,after this
> message:
>        
>   physmap flash device: 80000 at 1300000  
> 
> my board was dead. The  kernel is running from the
> flash . The code and jffs is on the same chip.Did it
> work?Thanks for giving me some suggestion!

(Doing this from memory, better check it yourself)

The message you quoted it printed out before the actual probing takes
place. There will be another message, if either the probing fails or
it detects more than one chip in the provided range. It will not
however print another message, if all goes well.
You should edit drivers/mtd/maps/physmap.c and add an unconditional
printk after the ioremap and after the actual probe, to see, which one
fails. My bet is the ioremap.

Jörn

--

-- 
Release early. Release often. And listen to your customers.
(Continue reading)

Permalink | Reply | 0 comments |
headers
Derek Ross | 1 Apr 2002 17:41
Picon

Can JFFS be placed in middle of flash?

Hello,

Can a JFFS file system be placed in the middle
area of flash? (We will be using the beginning and
end areas for other things).

What I'm concerned about is that many flash
chips have a group of 8 small blocks at the
beginning of the flash space in the chip.
If a JFFS is placed in the middle, is there
a danger that the software will still expect
those small blocks?

Derek Ross.
Permalink | Reply | 1 comment |
headers
Derek Ross | 1 Apr 2002 19:27
Picon

Can cfi_cmdset.c functions be called by anybody?

Hello,

Part of our flash application will be a real-time
logger.  The real time component will be limited
to certain periods of operation, but when it's running,
it will require about 5 bytes/second to be written
to a fixed circular buffer.

This would be simplified if I were able to directly
use the functions from cfi_cmdset_0001.c.

Are the functions available for any application to
use, or are they only available for device drivers?

Thanks,
Derek Ross.
Permalink | Reply | 2 comments |
headers
Jeremy Jackson | 1 Apr 2002 21:20

Award BIOS flash programming

HI,

I'm having trouble writing an MTD driver for some
common flash chips used in motherboard BIOS.
I'd like to merge /dev/bios project's support into MTD.

I'm having trouble getting writes to propagate to the 
bios flash device on two motherboards, 430TX & 430FX
based.  The AWDFLASH utility work ok on both. The
only bit in all the northbridge, southbridge, keyboard
controller, superio, flash chip that refers to write protect
is in the PIIX/PIIX4 southbridge.  Setting this bit to
either value fails to allow writes (logic probe on flash
chip WE#)

Under Linux, I've tried from kernel and user space.
Under dos, I've tried debug and uniflash.

Any perls of wisdom welcome.

Jeremy
Permalink | Reply | 0 comments |
headers
David Woodhouse | 1 Apr 2002 23:34
Favicon

Re: Can JFFS be placed in middle of flash?

dross <at> iders.ca said:
> Can a JFFS file system be placed in the middle area of flash? (We will
> be using the beginning and end areas for other things). 

Yes, we do it all the time. Look at the partitioning stuff in most of the 
'map' drivers.

--
dwmw2
Permalink | Reply | 0 comments |
headers
David Woodhouse | 1 Apr 2002 23:35
Favicon

Re: Can cfi_cmdset.c functions be called by anybody?

dross <at> iders.ca said:
> This would be simplified if I were able to directly use the functions
> from cfi_cmdset_0001.c.

> Are the functions available for any application to use, or are they
> only available for device drivers? 

The read/write/erase functionality of the chip driver is presented to 
userspace via the /dev/mtd[0123...] devices. Read and write work as 
expected, and there's an ioctl to perform an erase.

--
dwmw2
Permalink | Reply | 1 comment |
headers
Derek Ross | 2 Apr 2002 00:44
Picon

Are erase suspends possible?

Hello,

I was looking through the code in 
cfi_cmdset_0001.c, and it appears that
erase suspends are not implemented. Is
this correct?

Suspends are needed by my logging routine, 
in which I'll be writing several bytes per 
second to flash in a circular buffer of blocks.
I will have to erase the next block in
the sequence while writing to the current 
block, to keep the logging as continuous
as possible.

Thanks,
Derek Ross.
Permalink | Reply | 5 comments |
headers
David Woodhouse | 2 Apr 2002 02:14
Favicon

Re: Are erase suspends possible?

dross <at> iders.ca said:
> I was looking through the code in  cfi_cmdset_0001.c, and it appears
> that erase suspends are not implemented. Is this correct? 

No. Erase suspend has been working for ages. What version of 
cfi_cmdset_0001.c are you looking at?

--
dwmw2
Permalink | Reply | 4 comments |
headers
Andrew Dixon | 2 Apr 2002 16:28

PPCBoot can't read my jffs2 image

Hi Everyone,

Sorry for the cross post but this a problem with the way the linux-mtd
layer and PPCBoot are interacting.

So here's the situation, I've got PPCBoot+Linux running on a custom
board.  I've formatted the flash to be one small raw partition for
PPCBoot and the rest is a jffs2 filesystem.  PPCBoot is reading Linux
and an initial ramdisk out of the jffs2.  In my original testing I was
booting the board via tftp and nfs and then copying copying the
filesystem onto the flash via /dev/mtdblock1.  This works great.  I
reboot and PPCBoot sees everything in the jffs2.  Now I'm trying to
automate this a little more so enhanced my build to give me a jffs2
image "wombat.jffs2".  I create this image with mkfs.jffs2 and put it
onto the flash by dd'ing it to /dev/mtd1:

	dd if=/wombat.jffs2 of=/dev/mtd1

now when I mount /dev/mtdblock1 I can see the filesystem that I've put
over there but when I reboot PPCBoot doesn't see any of the files that
are there.  But if I boot back up Linux can still see all of those files
OK.

Any ideas?  

Thanks,
Andy 
--

-- 
Andrew Dixon
Software Engineer
(Continue reading)

Permalink | Reply | 0 comments |

Gmane