linux-kernel - Re: [GIT PULL 16/20] lightnvm: error handling when whole line is bad

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

Message-ID: <19F83A5D-1449-4F9F-B8E3-0B2B104DA1D2@cnexlabs.com>
Date:   Tue, 29 May 2018 18:29:19 +0000
From:   Javier Gonzalez <javier@...xlabs.com>
To:     "Konopko, Igor J" <igor.j.konopko@...el.com>
CC:     Matias Bjørling <mb@...htnvm.io>,
        Jens Axboe <axboe@...com>,
        "linux-block@...r.kernel.org" <linux-block@...r.kernel.org>,
        LKML <linux-kernel@...r.kernel.org>,
        "Dziegielewski, Marcin" <marcin.dziegielewski@...el.com>
Subject: Re: [GIT PULL 16/20] lightnvm: error handling when whole line is bad

> On 29 May 2018, at 15.15, Konopko, Igor J <igor.j.konopko@...el.com> wrote:
> 
>> From: Javier Gonzalez [mailto:javier@...xlabs.com]
>> This case cannot occur on the only user of the function
>> (pblk_recov_l2p()). On the previous check (pblk_line_was_written()), we
>> verify the state of the line and the position of the first good chunk. In
>> the case of a bad line (less chunks than a given threshold to allow
>> emeta), the recovery will not be carried out in the line.
> You are right. It looks that during my testing there was some
> inconsistency between chunks state table which is verified inside
> pblk_line_was_written() and blk_bitmap which was read from emeta and
> is verified in pblk_line_smeta_start(). I'm living decision to
> maintainers whether we should keep this sanity check or not - it
> really just pass gracefully the result from pblk_line_smeta_start()
> where similar sanity check is present.
> 

Let's avoid an extra check now that there is no users for it in the
current flow. If we have a new use for pblk_line_smeta_start() on a flow
that does cannot offer the same guarantees, we can add it at that point.

Javier

Download attachment "signature.asc" of type "application/pgp-signature" (834 bytes)