Messages in this thread | | | Subject | Re: [PATCH 10/15] MTD: nand: make reads using MTD_OOB_RAW affect only ECC validation | From | Maxim Levitsky <> | Date | Tue, 23 Feb 2010 20:32:40 +0200 |
| |
On Mon, 2010-02-22 at 22:29 +0100, Thomas Gleixner wrote: > On Mon, 22 Feb 2010, Maxim Levitsky wrote: > > On Mon, 2010-02-22 at 22:20 +0100, Thomas Gleixner wrote: > > > On Mon, 22 Feb 2010, Maxim Levitsky wrote: > > > > > > > This changes the behavier of MTD_OOB_RAW. It used to read both OOB and data > > > > to the data buffer, however you would still need to specify the dummy oob buffer. > > > > > > > > This is only used in one place, but makes it hard to read data+oob without ECC > > > > test, thus I removed that behavier, and fixed the user. > > > > > > > > Now MTD_OOB_RAW behaves just like MTD_OOB_PLACE, but doesn't do ECC validation > > > > > > Is this tested against existing user space tools like nanddump ? Can I > > > still get the raw data from flash ? > > > > Thats the point. > > Userspace doesn't/can't use that mode. > > It is not exposed through mtdchar. > > Userspace reads the page, and then reads the oob. > > > > It does use MTD_OOB_RAW, but without data buffer, and this path I don't > > change. > > > > The only user of this, is the nand itself, when it reads bad block > > table. > > > > I confess that I didn't run test that I ported this code correctly. > > But I did logically verified many times that the new code works just > > like old one. > > Well, then it's about time to run the tests :) > > Thanks,
Just to be sure, I tested jffs2 and nanddump.
Works just fine.
Best regards, Maxim Levitsky
| |