mtd/fs/jffs2 erase.c,1.84,1.85 roll back

[Artem B. Bityutskiy]

On Tue, 2005-09-20 at 21:49 +0200, Anders Grafstrom wrote:
> So if we return -EIO, then we go to jffs2_erase_failed(). For NAND, if 
> the MTD driver reported erase failure (bad_offset != 0xffffffff), a 
> retry is made before giving up on the block and marking it bad. Also the 
> read operation may have failed, erase may have failed and MTD missed it, 
> write of clean marker may have failed. As it is these cases will put the 
> block on the bad_list. In the case of NOR it will also be put on the 
> bad_list at once.
> But how about retrying for all of these cases?

Sorry, no more today. What does your patch do? Adds the second try? What
for?

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