Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.36 as permitted sender) client-ip=23.128.96.36;
MIME-Version: 1.0
References: <20231020-mtd-otp-byteswap-v4-1-0d132c06aa9d@linaro.org> <20231023102552.4b053b17@xps-13>
In-Reply-To: <20231023102552.4b053b17@xps-13>
From:   Linus Walleij <linus.walleij@linaro.org>
Date:   Mon, 23 Oct 2023 10:35:24 +0200
Message-ID: <CACRpkdYaq=RxgLJhRanfgw1+280N1K-aRvpy_BTM3SYY9Xtokw@mail.gmail.com>
Subject: Re: [PATCH v4] mtd: cfi_cmdset_0001: Byte swap OTP info
To:     Miquel Raynal <miquel.raynal@bootlin.com>
Cc:     Richard Weinberger <richard@nod.at>,
        Vignesh Raghavendra <vigneshr@ti.com>,
        linux-mtd@lists.infradead.org, linux-kernel@vger.kernel.org,
        stable@vger.kernel.org, Nicolas Pitre <nico@fluxnic.net>
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
Precedence: bulk

On Mon, Oct 23, 2023 at 10:25=E2=80=AFAM Miquel Raynal
<miquel.raynal@bootlin.com> wrote:

> linus.walleij@linaro.org wrote on Fri, 20 Oct 2023 22:30:29 +0200:
>
> > Currently the offset into the device when looking for OTP
> > bits can go outside of the address of the MTD NOR devices,
> > and if that memory isn't readable, bad things happen
> > on the IXP4xx (added prints that illustrate the problem before
> > the crash):
> >
> > cfi_intelext_otp_walk walk OTP on chip 0 start at reg_prot_offset 0x000=
00100
> > ixp4xx_copy_from copy from 0x00000100 to 0xc880dd78
> > cfi_intelext_otp_walk walk OTP on chip 0 start at reg_prot_offset 0x120=
00000
> > ixp4xx_copy_from copy from 0x12000000 to 0xc880dd78
> > 8<--- cut here ---
> > Unable to handle kernel paging request at virtual address db000000
> > [db000000] *pgd=3D00000000
> > (...)
> >
> > This happens in this case because the IXP4xx is big endian and
> > the 32- and 16-bit fields in the struct cfi_intelext_otpinfo are not
> > properly byteswapped. Compare to how the code in read_pri_intelext()
> > byteswaps the fields in struct cfi_pri_intelext.
> >
> > Adding a small byte swapping loop for the OTP in read_pri_intelext()
> > and the crash goes away.
> >
> > The problem went unnoticed for many years until I enabled
> > CONFIG_MTD_OTP on the IXP4xx as well, triggering the bug.
> >
> > Cc: stable@vger.kernel.org
>
> Would you like to add a Fixes tag as well? Or is this skipped on
> purpose?

Yes the actual commit predates existing git history... :/

Git blame looks like that:

^1da177e4c3f4 (Linus Torvalds         2005-04-16 15:20:36 -0700  423)
^1da177e4c3f4 (Linus Torvalds         2005-04-16 15:20:36 -0700  424)
         /* Protection Register info */
b359ed5184aeb (Jean-Philippe Brucker  2020-04-17 16:23:26 +0200  425)
         if (extp->NumProtectionFields)
b359ed5184aeb (Jean-Philippe Brucker  2020-04-17 16:23:26 +0200  426)
                 extra_size +=3D (extp->NumProtectionFields - 1) *
b359ed5184aeb (Jean-Philippe Brucker  2020-04-17 16:23:26 +0200  427)
                               sizeof(struct cfi_intelext_otpinfo);

Jean-Philippe's patch is not the root cause AFAICT, but something
predating it, and predating git history.

The fix easily applies to all maintained stable kernels.

> > Reviewed-by: Nicolas Pitre <nico@fluxnic.net>
> > Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
> > ---
> > ChangeLog v3->v4:
> > - Collected Nico's ACK.
> > - Stalled since june! Has this been missed?
>
> Our current organization relies on Vignesh to pick-up (or tell me to
> pick-up) cfi patches. But he is slightly less active these days, so if I
> don't get any feedback from him soon I will take it for the next merge
> window.
>
> Sorry for the delay anyway.

No worries it's not like the patch is urgent.

Yours,
Linus Walleij