Re: [PATCH] misc: eeprom: at24: Always append device id even if label property is set.

[Bartosz Golaszewski <bgolaszewski@baylibre.com>]

On Thu, Jul 1, 2021 at 5:28 PM <jglisse@redhat.com> wrote:
>
> From: Jérôme Glisse <jglisse@redhat.com>
>
> We need to append device id even if eeprom have a label property set as some
> platform can have multiple eeproms with same label and we can not register
> each of those with same label. Failing to register those eeproms trigger
> cascade failures on such platform (system is no longer working).
>
> This fix regression on such platform introduced with 4e302c3b568e
>
> Signed-off-by: Jérôme Glisse <jglisse@redhat.com>
> Cc: Diego Santa Cruz <Diego.SantaCruz@spinetix.com>
> Cc: Bartosz Golaszewski <bgolaszewski@baylibre.com>
> Cc: Jon Hunter <jonathanh@nvidia.com>
> Cc: stable@vger.kernel.org
> Cc: linux-i2c@vger.kernel.org
> ---
>  drivers/misc/eeprom/at24.c | 17 +++++++----------
>  1 file changed, 7 insertions(+), 10 deletions(-)
>
> diff --git a/drivers/misc/eeprom/at24.c b/drivers/misc/eeprom/at24.c
> index 7a6f01ace78a..305ffad131a2 100644
> --- a/drivers/misc/eeprom/at24.c
> +++ b/drivers/misc/eeprom/at24.c
> @@ -714,23 +714,20 @@ static int at24_probe(struct i2c_client *client)
>         }
>
>         /*
> -        * If the 'label' property is not present for the AT24 EEPROM,
> -        * then nvmem_config.id is initialised to NVMEM_DEVID_AUTO,
> -        * and this will append the 'devid' to the name of the NVMEM
> -        * device. This is purely legacy and the AT24 driver has always
> -        * defaulted to this. However, if the 'label' property is
> -        * present then this means that the name is specified by the
> -        * firmware and this name should be used verbatim and so it is
> -        * not necessary to append the 'devid'.
> +        * We initialize nvmem_config.id to NVMEM_DEVID_AUTO even if the
> +        * label property is set as some platform can have multiple eeproms
> +        * with same label and we can not register each of those with same
> +        * label. Failing to register those eeproms trigger cascade failure
> +        * on such platform.
>          */
> +       nvmem_config.id = NVMEM_DEVID_AUTO;
> +
>         if (device_property_present(dev, "label")) {
> -               nvmem_config.id = NVMEM_DEVID_NONE;
>                 err = device_property_read_string(dev, "label",
>                                                   &nvmem_config.name);
>                 if (err)
>                         return err;
>         } else {
> -               nvmem_config.id = NVMEM_DEVID_AUTO;
>                 nvmem_config.name = dev_name(dev);
>         }
>
> --
> 2.31.1
>

Cc'ing Alexander Fomichev who reported this issue first.

This is the second time someone raises this problem so it seems that
this change really broke many existing systems. I will apply this
patch and send it for stable.

Bart