[PATCH 1/2] eeprom: at24: added quirk for 24c08 devices

[Joel Stanley]

Hi Brendan,

On Sun, Aug 21, 2016 at 11:04 AM, Brendan Higgins
<brendanhiggins at google.com> wrote:
> Some 24c08 devices exhibits a quirk where they will not NACK durring an
> acknowledge poll. Added a STOP after the dummy write to address problem.

I'm not an i2c nor a eeprom expert. Are we confident that this is this
generic enough that it should be applied to all at24c08 devices?

This needs a signed-off-by. You might want to briefly describe the
hardware you've tested this on.

Once you've added that, please rebase on Linus' tree and send it
directly upstream to the maintainer. You can find the maintainer's
email address and mailing list using scripts/get_maintainer.pl on your
patch.

You can cc me if you would like.

Cheers,

Joel

> ---
>  drivers/misc/eeprom/at24.c         | 12 +++++++++++-
>  include/linux/platform_data/at24.h |  1 +
>  2 files changed, 12 insertions(+), 1 deletion(-)
>
> diff --git a/drivers/misc/eeprom/at24.c b/drivers/misc/eeprom/at24.c
> index 9ceb63b..c647efb 100644
> --- a/drivers/misc/eeprom/at24.c
> +++ b/drivers/misc/eeprom/at24.c
> @@ -120,7 +120,11 @@ static const struct i2c_device_id at24_ids[] = {
>                 AT24_FLAG_READONLY | AT24_FLAG_IRUGO) },
>         { "24c04", AT24_DEVICE_MAGIC(4096 / 8, 0) },
>         /* 24rf08 quirk is handled at i2c-core */
> -       { "24c08", AT24_DEVICE_MAGIC(8192 / 8, 0) },
> +       /* 24c08 exhibits a quirk where it will not NACK durring an
> +        * acknowledge poll so we have to send a stop after the dummy write in
> +        * a read.
> +        */
> +       { "24c08", AT24_DEVICE_MAGIC(8192 / 8, AT24_FLAG_STOPPOLL) },
>         { "24c16", AT24_DEVICE_MAGIC(16384 / 8, 0) },
>         { "24c32", AT24_DEVICE_MAGIC(32768 / 8, AT24_FLAG_ADDR16) },
>         { "24c64", AT24_DEVICE_MAGIC(65536 / 8, AT24_FLAG_ADDR16) },
> @@ -212,6 +216,12 @@ static ssize_t at24_eeprom_read(struct at24_data *at24, char *buf,
>                 msgbuf[i++] = offset;
>
>                 msg[0].addr = client->addr;
> +               /*
> +                * Some EEPROMs do not perform the acknowledge poll correctly
> +                * and have to send a STOP after the dummy write.
> +                */
> +               if (at24->chip.flags & AT24_FLAG_STOPPOLL)
> +                       msg[0].flags = I2C_M_STOP;
>                 msg[0].buf = msgbuf;
>                 msg[0].len = i;
>
> diff --git a/include/linux/platform_data/at24.h b/include/linux/platform_data/at24.h
> index be830b1..1773822 100644
> --- a/include/linux/platform_data/at24.h
> +++ b/include/linux/platform_data/at24.h
> @@ -47,6 +47,7 @@ struct at24_platform_data {
>  #define AT24_FLAG_READONLY     0x40    /* sysfs-entry will be read-only */
>  #define AT24_FLAG_IRUGO                0x20    /* sysfs-entry will be world-readable */
>  #define AT24_FLAG_TAKE8ADDR    0x10    /* take always 8 addresses (24c00) */
> +#define AT24_FLAG_STOPPOLL     0x08    /* Send stop before ACK poll */
>
>         void            (*setup)(struct nvmem_device *nvmem, void *context);
>         void            *context;
> --
> 2.8.0.rc3.226.g39d4020
>
> _______________________________________________
> openbmc mailing list
> openbmc at lists.ozlabs.org
> https://lists.ozlabs.org/listinfo/openbmc