**** BEGIN LOGGING AT Fri May 22 02:59:58 2015
May 22 09:57:43 <ant_work> lumag__: well, should I consider the read operation may sleep?
May 22 09:58:12 <lumag__> ant_work, nand read?
May 22 09:58:38 <ant_work> http://www.opensourceforu.com/2012/01/working-with-mtd-devices/
May 22 09:58:49 <ant_work> seems easier to work at mtd level
May 22 09:59:26 <lumag__> Judging from afs partition parser, parsing happens in a non-atomic context
May 22 10:01:23 <ant_work> I hope these sectors are guaranteed good
May 22 10:01:55 <ant_work> then have to check about ECC
May 22 10:02:32 <ant_work> I have an indice: long ago pwgen had a defetcive sector and had to install u-boot
May 22 10:03:03 <ant_work> that could mean, Angell bootloader is looking at fixed addresses
May 22 10:03:25 <ant_work> and not finding relocated badblocks
May 22 10:14:20 <ant_work> lumag__: if I understand it right, it is possible to fill the partitions in mtd_info
May 22 10:14:31 <ant_work> after having read them from it
May 22 10:20:04 <lumag__> I don't remember if partitioning info is reread in case of ECC errors
May 22 10:21:28 <ant_work> we read the two copies, thi sshould be enough
**** ENDING LOGGING AT Sat May 23 02:59:58 2015