marking a block bad

Jeff Hane jhane at
Tue Feb 5 16:39:59 EST 2008


 I am using yaffs for my root filesystem and I am getting some ECC
errors for reads and writes.  What I'm trying to understand is what
determines if a block has gone bad when the block is marked bad. I see
the way to mark a block bad in the mtd code but not clear on when this
will be used.
 Of course this leads to another problem, if a block is marked bad and
the system is rebooted the data that was in that block would become
unaccessible(assume the system can reboot).  In this case, is there
something that can be done, by the filesystem or other methods, to try
to re-locate the data before reboot or rebuilding the bbt?


More information about the linux-mtd mailing list