linux-kernel - Re: [NAND Question] When oob

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Date:   Tue, 28 Feb 2017 17:26:53 +0900
From:   Masahiro Yamada <yamada.masahiro@...ionext.com>
To:     Boris Brezillon <boris.brezillon@...e-electrons.com>
Cc:     linux-mtd@...ts.infradead.org,
        Linux Kernel Mailing List <linux-kernel@...r.kernel.org>
Subject: Re: [NAND Question] When oob_required is used?

Hi Boris,


2017-02-28 17:02 GMT+09:00 Boris Brezillon <boris.brezillon@...e-electrons.com>:
> On Tue, 28 Feb 2017 16:21:50 +0900
> Masahiro Yamada <yamada.masahiro@...ionext.com> wrote:
>
>> Hi.
>>
>>
>> I am seeing the reference implementations for the callback
>> in drivers/mtd/nand/nand_base.c
>>
>>
>> I see only some of them use the argument "oob_required".
>>
>>
>> - oob_required is used
>> nand_read_page_raw()
>> nand_write_page_raw()
>> nand_write_subpage_hwecc()
>>
>>
>> - oob_required is ignored
>> nand_read_page_raw_syndrome()
>> nand_read_page_swecc()
>> nand_read_page_hwecc()
>> nand_read_page_hwecc_oob_first()
>> nand_read_page_syndrome()
>> nand_write_page_raw_syndrome()
>> nand_write_page_swecc()
>> nand_write_page_hwecc()
>> nand_write_page_syndrome()
>>
>>
>>
>> I could not get the logic
>> when "oob_required" is used.
>>
>>
>>
>> When implementing drivers, the "oob_required"
>> should be respected (i.e. oob_poi should be transferred as well),
>> or is it OK to ignore it?
>>
>>
>
> oob_required = true => you must retrieve OOB bytes
> oob_required = false => you can optimize the process and ignore OOB
> bytes, but you can also retrieve them if you don't care about
> optimization. That's usually what's done when you have to retrieve ECC
> bytes (retrieving a few extra OOB bytes is not expensive and usually
> simplifies the logic, especially when free and ECC bytes are
> interleaved in the OOB region).

Ah, I see.

The functions that ignore "oob_required"
always transfer the oob_poi.


Thanks!


-- 
Best Regards
Masahiro Yamada