linux-kernel - Re: [PATCH 1/5] misc: eeprom: at24: Initialise AT24 NVMEM ID field

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]

Date:   Thu, 10 Sep 2020 17:35:39 +0200
From:   Bartosz Golaszewski <bgolaszewski@...libre.com>
To:     Jon Hunter <jonathanh@...dia.com>
Cc:     Rob Herring <robh+dt@...nel.org>,
        Thierry Reding <thierry.reding@...il.com>,
        linux-i2c <linux-i2c@...r.kernel.org>,
        LKML <linux-kernel@...r.kernel.org>,
        linux-devicetree <devicetree@...r.kernel.org>,
        linux-tegra@...r.kernel.org
Subject: Re: [PATCH 1/5] misc: eeprom: at24: Initialise AT24 NVMEM ID field

On Thu, Sep 10, 2020 at 3:43 PM Jon Hunter <jonathanh@...dia.com> wrote:
>
> The AT24 EEPROM driver does not initialise the 'id' field of the
> nvmem_config structure and because the entire structure is not
> initialised, it ends up with a random value. This causes the NVMEM
> driver to append the device 'devid' value to name of the NVMEM
> device. Although this is not a problem per-se, for I2C devices such as
> the AT24, that already have a device unique name, there does not seem
> much value in appending an additional 0 to the I2C name. For example,
> appending a 0 to an I2C device name such as 1-0050 does not seem
> necessary and maybe even a bit confusing. Therefore, fix this by
> setting the NVMEM config.id to NVMEM_DEVID_NONE for AT24 EEPROMs.
>
> Signed-off-by: Jon Hunter <jonathanh@...dia.com>
> ---
>  drivers/misc/eeprom/at24.c | 1 +
>  1 file changed, 1 insertion(+)
>
> diff --git a/drivers/misc/eeprom/at24.c b/drivers/misc/eeprom/at24.c
> index e9df1ca251df..3f7a3bb6a36c 100644
> --- a/drivers/misc/eeprom/at24.c
> +++ b/drivers/misc/eeprom/at24.c
> @@ -715,6 +715,7 @@ static int at24_probe(struct i2c_client *client)
>
>         nvmem_config.name = dev_name(dev);
>         nvmem_config.dev = dev;
> +       nvmem_config.id = NVMEM_DEVID_NONE;
>         nvmem_config.read_only = !writable;
>         nvmem_config.root_only = !(flags & AT24_FLAG_IRUGO);
>         nvmem_config.owner = THIS_MODULE;
> --
> 2.25.1
>

This patch is correct and thanks for catching it. I vaguely recall
wondering at some point why the appended 0 in the nvmem name for at24.
Unfortunately this change would affect how the device is visible in
user-space in /sys/bus/nvmem/devices/ and this could break existing
users. Also: there are many in-kernel users that would need to be
updated. I'm afraid we'll need some sort of backward compatibility.

Bartosz