linux-kernel - Re: [PATCH v3 1/2] mtd: spi-nand: Add read retry support

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]

Message-ID: <87msessr0u.fsf@bootlin.com>
Date: Tue, 11 Feb 2025 11:40:49 +0100
From: Miquel Raynal <miquel.raynal@...tlin.com>
To: Cheng Ming Lin <linchengming884@...il.com>
Cc: vigneshr@...com,  linux-mtd@...ts.infradead.org,
  linux-kernel@...r.kernel.org,  richard@....at,  alvinzhou@...c.com.tw,
  leoyu@...c.com.tw,  Cheng Ming Lin <chengminglin@...c.com.tw>
Subject: Re: [PATCH v3 1/2] mtd: spi-nand: Add read retry support

On 11/02/2025 at 16:13:32 +08, Cheng Ming Lin <linchengming884@...il.com> wrote:

> Hi Miquel,
>
> Miquel Raynal <miquel.raynal@...tlin.com> 於 2025年2月10日 週一 下午6:07寫道：
>>
>> Hello,
>>
>> >> > +                             ret = spinand->set_read_retry(spinand, retry_mode);
>> >> > +                             if (ret < 0) {
>> >> > +                                     ecc_failed = true;
>> >> > +                                     return ret;
>> >>
>> >> Shall we try to set the read_retry level to 0 upon:
>> >>
>> >>       if (ret < 0 && retry_mode > 1)
>> >>
>> >> ?
>> >
>> > If we set the read_retry level to 0 upon, and set_read_retry fails
>> > when retry_mode equals to 1, it won't return an error. This could
>> > potentially mask an underlying issue.
>>
>> Don't save the return value in this case? But otherwise you would leave
>> the chip in a retry state, no?
>
> However, if we set the read_retry level to 0 upon, the chip would still
> remain in the retry state if set_read_retry fails.

Well, if even this fails, you have bigger troubles than read_retry being
enabled.

>
> I come up with a solution: setting the read_retry level to 0 right before
> the read_retry label. This ensures that subsequent reads always start
> from level 0, and it eliminates the need to reset the read_retry level at
> the end.

I don't see the difference with the previous solution. If it fails, it's
exactly the same.

Please do not add a function call to *every* read regardless of the fact
that we enabled read_retry (which I think is what you are suggesting).

Thanks,
Miquèl