Please explain bbt_create (problems)

[Thomas Gleixner]

Please fix your mail client to do a line wrap at 76 characters

Read 

On Fri, 2005-07-15 at 16:18 +0200, Andriy Korud wrote:
> Hi, 
> I'm trying to understand how initial bbt_ceate works? 
> I have NAND chip erased for jffs2 (flash_erasall -j) with some data in
> OOB (jffs2 clean masrkers). And bbt_create() calls check_pattern()
> which compare each byte with 0xff, in my case comparision fails (jffs2
> cleanmarker != 0xff) and I have every block marked as bad.
> Should not check_pattern() checks only byte at 0x05 in OOB (according
> to http://www.linux-mtd.infradead.org/tech/mtdnand/x255.html )? 


Maybe you read also:

http://www.linux-mtd.infradead.org/tech/mtdnand/x144.html

"...Therefor we scan the chip(s) when we detect them the first time for
good blocks and store this information in
a bad block table before erasing any of the blocks." 

Emphasis is on the phrase "first time".

I can see that this could have been described more clear, but it
definitely makes no sense to create a flash based bad block table
_after_ using the chip.


> Also is it possible to scan chip for bbt starting from certain block
> (not 0) without dirty changes in mtd code?

no

tglx