Intel NOR flash(P30) query

Intel NOR flash(P30) query

[Manjula D]

Hi All,
Iam a newbie to mtd. We have a MPC8272 custom board with INTEL
TE28F640P30B85 (P30-CFI compilant)NOR boot Flash and NAND flash
with JFFS2 filesystem running linux-2.4.20-8.
The NOR Flash need not have any partition/filesystem.
My requirement is to have an application to copy an image from
the NAND Flash to the NOR Flash.
My current understanding is 
Under RAM/ROM/Flash chip drivers on selecting, 
Detect flash chips by Common Flash Interface (CFI) probe :
  cfi_probe_init() will register with the mtd_chip_driver.
Support for Intel/Sharp flash chips:
  cfi_cmdset_0001.c contains the list of commands supported by
  Intel.
  
1) I need to understand how the application views the NOR flash
    once it is registered as a mtd device.
2) Once i register as mtd_chip_driver how can i call the
    read/write commands of the flash from the application?
3) What is do_map_probe for and how is it different from
    cfi_probe?

  -Thanks
  Manju

________________________________________________
Get your own "800" number
Voicemail, fax, email, and a lot more
http://www.ureach.com/reg/tag

Re: Intel NOR flash(P30) query

[Vitaly Bordug]

I'm not an MTD guru, but...
Manjula D wrote:
> Hi All,
> Iam a newbie to mtd. We have a MPC8272 custom board with INTEL
> TE28F640P30B85 (P30-CFI compilant)NOR boot Flash and NAND flash
> with JFFS2 filesystem running linux-2.4.20-8.
> The NOR Flash need not have any partition/filesystem.
> My requirement is to have an application to copy an image from
> the NAND Flash to the NOR Flash.
> My current understanding is 
> Under RAM/ROM/Flash chip drivers on selecting, 
> Detect flash chips by Common Flash Interface (CFI) probe :
>   cfi_probe_init() will register with the mtd_chip_driver.
> Support for Intel/Sharp flash chips:
>   cfi_cmdset_0001.c contains the list of commands supported by
>   Intel.
>   
> 1) I need to understand how the application views the NOR flash
>     once it is registered as a mtd device.
The answer is trivial - as mtd device. You can enable support for mtd as 
character or block devices - thus you will have say /dev/mtdchar1 to 
work from your application.
> 2) Once i register as mtd_chip_driver how can i call the
>     read/write commands of the flash from the application?
see upper. It can be treated as an ordinary character device.
> 3) What is do_map_probe for and how is it different from
>     cfi_probe?
If the flash is used for firmware storage as well, you need to 
"partition" it with independent accesss to each part, keeping some of 
them write-protected. do_map_probe will apply the map, and will probably 
call cfi_probe for certain flash region - refer to google for details.
> 
>   -Thanks
>   Manju
> 
> ________________________________________________
> Get your own "800" number
> Voicemail, fax, email, and a lot more
> http://www.ureach.com/reg/tag
> 
> ______________________________________________________
> Linux MTD discussion mailing list
> http://lists.infradead.org/mailman/listinfo/linux-mtd/
> 
> 


-- 
Sincerely,
Vitaly

Re: Intel NOR flash(P30) query

[Jörn Engel]

On Tue, 16 August 2005 19:11:46 +0400, Vitaly Bordug wrote:
> 
> I'm not an MTD guru, but...

No shame in that.

> Manjula D wrote:
> >Hi All,
> >Iam a newbie to mtd. We have a MPC8272 custom board with INTEL
> >TE28F640P30B85 (P30-CFI compilant)NOR boot Flash and NAND flash
> >with JFFS2 filesystem running linux-2.4.20-8.
> >The NOR Flash need not have any partition/filesystem.
> >My requirement is to have an application to copy an image from
> >the NAND Flash to the NOR Flash.
> >My current understanding is 
> >Under RAM/ROM/Flash chip drivers on selecting, 
> >Detect flash chips by Common Flash Interface (CFI) probe :
> >  cfi_probe_init() will register with the mtd_chip_driver.
> >Support for Intel/Sharp flash chips:
> >  cfi_cmdset_0001.c contains the list of commands supported by
> >  Intel.
> >  
> >1) I need to understand how the application views the NOR flash
> >    once it is registered as a mtd device.
> The answer is trivial - as mtd device. You can enable support for mtd as 
> character or block devices - thus you will have say /dev/mtdchar1 to 
> work from your application.

Almost correct.  See:
http://www.linux-mtd.infradead.org/faq/general.html

Jörn

-- 
Data dominates. If you've chosen the right data structures and organized
things well, the algorithms will almost always be self-evident. Data
structures, not algorithms, are central to programming.
-- Rob Pike

Re: Intel NOR flash(P30) query

[Todd Poynor]

By the way, if anybody has mtd working on the new P30 chips it would be 
good to know; I haven't seen a system with these chips yet.  They are 
CFI 1.4, and some changes for 1.4 were added by Nicolas recently to 
support M18 "Sibley" chips, but not sure if those changes are required 
for P-series chips.  Thankis,

-- 
Todd

Re: Intel NOR flash(P30) query

[Ralph Siemsen]

Todd Poynor wrote:
> By the way, if anybody has mtd working on the new P30 chips it would be 
> good to know; I haven't seen a system with these chips yet.  They are 
> CFI 1.4, and some changes for 1.4 were added by Nicolas recently to 
> support M18 "Sibley" chips, but not sure if those changes are required 
> for P-series chips.  Thankis,

Are these changes merged into mainline kernel, or only in MTD CVS?  I 
may be in a position to test this soon-ish, as we are looking for a 
replacement for the now obsoleted J/K strataflash parts and seem to be 
headed towards the P30's...

-Ralph

Re: Intel NOR flash(P30) query

[Todd Poynor]

Ralph Siemsen wrote:
> Todd Poynor wrote:
> 
>> By the way, if anybody has mtd working on the new P30 chips it would 
>> be good to know; I haven't seen a system with these chips yet.  They 
>> are CFI 1.4, and some changes for 1.4 were added by Nicolas recently 
>> to support M18 "Sibley" chips, but not sure if those changes are 
>> required for P-series chips.  Thankis,
> 
> 
> Are these changes merged into mainline kernel, or only in MTD CVS?  I 
> may be in a position to test this soon-ish, as we are looking for a 
> replacement for the now obsoleted J/K strataflash parts and seem to be 
> headed towards the P30's...

Pretty sure linux-mtd-only, these came in during the settle down period 
for 2.6.13 rc's.  The CFI 1.3-based code probably works after a quick 
adjustment of a revision level check (to ignore new fields at end of 
vendor-specific extended query table), if you want to try kernel.org, 
but testing linux-mtd is always better for us here.  Thanks,

-- 
Todd

Re: Intel NOR flash(P30) query

[Josh Boyer]

On Tue, 2005-08-16 at 14:14 -0700, Todd Poynor wrote:
> Ralph Siemsen wrote:
> > Todd Poynor wrote:
> > 
> >> By the way, if anybody has mtd working on the new P30 chips it would 
> >> be good to know; I haven't seen a system with these chips yet.  They 
> >> are CFI 1.4, and some changes for 1.4 were added by Nicolas recently 
> >> to support M18 "Sibley" chips, but not sure if those changes are 
> >> required for P-series chips.  Thankis,
> > 
> > 
> > Are these changes merged into mainline kernel, or only in MTD CVS?  I 
> > may be in a position to test this soon-ish, as we are looking for a 
> > replacement for the now obsoleted J/K strataflash parts and seem to be 
> > headed towards the P30's...
> 
> Pretty sure linux-mtd-only, these came in during the settle down period 
> for 2.6.13 rc's.  The CFI 1.3-based code probably works after a quick 
> adjustment of a revision level check (to ignore new fields at end of 
> vendor-specific extended query table), if you want to try kernel.org, 
> but testing linux-mtd is always better for us here.  Thanks,

For what it's worth, I just changed the checks for version 3 of the
query table to allow a version 4 in the cfi_cmdset_0001.c driver and it
seemed to be working.  I haven't had a chance to test out Nicolas'
changes, but it should work just fine.

josh