linux-kernel - Re: [PATCH] lightnvm: pblk: fix crash in pblk_end_partial

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:   Wed, 10 Apr 2019 14:52:49 +0200
From:   Hans Holmberg <hans@...tronix.com>
To:     Javier González <javier@...igon.com>
Cc:     Matias Bjørling <mb@...htnvm.io>,
        "Konopko, Igor J" <igor.j.konopko@...el.com>,
        Klaus Birkelund Jensen <klaus.jensen@...xlabs.com>,
        linux-block@...r.kernel.org,
        Linux Kernel Mailing List <linux-kernel@...r.kernel.org>,
        Hans Holmberg <hans.holmberg@...xlabs.com>
Subject: Re: [PATCH] lightnvm: pblk: fix crash in pblk_end_partial_read due to
 multipage bvecs

On Wed, Apr 10, 2019 at 2:22 PM Javier González <javier@...igon.com> wrote:
>
> > On 8 Apr 2019, at 09.33, hans@...tronix.com wrote:
> >
> > From: Hans Holmberg <hans.holmberg@...xlabs.com>
> >
> > Ever since '07173c3ec276 ("block: enable multipage bvecs")' we
> > need to handle bios with multipage bvecs in pblk.
> >
> > Currently, a multipage bvec results in a crash[1].
> > Fix this by using bvec iterators in stead of direct bvec indexing.
> >
> > Also add a dcache flush, for the same reasons as in:
> > '6e6e811d747b ("block: Add missing flush_dcache_page() call")'
> >
> > [1] https://github.com/OpenChannelSSD/linux/issues/61
> >
> > Reported-by: Klaus Jensen <klaus.jensen@...xlabs.com>
> > Signed-off-by: Hans Holmberg <hans.holmberg@...xlabs.com>
> > ---
> >
> > It ain't pretty, but let's fix the breakage while waiting for Igor's
> > partial read cleanup to be ready.
> >
>
> I would prefer to have the real fix, but it is good to have this fixed for
> 5.2. Can we agree on a revert if Igor sends the patch within the 5.2
> window?

yeah! please chuck this patch out of the window when we have a nicer fix ready.

Thanks,
Hans