On Thursday 06 November 2008, Kyungmin Park wrote:
ret = mtd->read_oob(mtd, ofs, &ops);if (ret) {printk("Read failed 0x%x, %d", (unsigned int) ofs,ret); + mtd->block_markbad(mtd, ofs);
You are marking all blocks as bad whenever a failure occurs. I'm not sure if we really should do it this way. Failures could have other reasons as well. I'm inclined to remove this marking in my patch version.
Any comments?
Right, current implementation make a bad on all errors, but actually it need to try again in case read/write error. It will patch it later.
OK, I'll remove this bad block marking for now.
<snip>
I'm currently working on a version of this "bad block aware" OneNAND command support which resembles the NAND command style even more. I hope to have something ready till tomorrow that I can send to the list for review. I would really like to see some comments from you on this since you are much more experienced with the OneNAND stuff.
Do you already have some comments to the patch version I posted yesterday?
Looks good to me. No problem to commit your version.
Acked-by: Kyungmin Park kyungmin.park@samsung.com
Thanks. I'll send an updated version today or tomorrow.
Best regards, Stefan
===================================================================== DENX Software Engineering GmbH, MD: Wolfgang Denk & Detlev Zundel HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany Phone: +49-8142-66989-0 Fax: +49-8142-66989-80 Email: office@denx.de =====================================================================