Hi,

The return value of yaffs_MarkBlockBad seems to be ignored always.

Is it safe to do so? What will be the ramifications in YAFFS if this
operation fails? Will the block which failed to get marked as bad, get
used again?

I guess this means that using block 0 for storing bad block table is the
only complete solution. What do most in-production systems do? Do all of
them use block 0 of NAND or another non-volatile storage media to
maintain the bad-block table?

Does any NAND device actually guarantee that marking a block bad will
not fail? Or that OOB area writes are not prone to program-disturb
failure? 

Thanks,
Sekhar Nori.