linux-kernel - Re: [RFC] Driver for Real Time Clock chip ST M41T65

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]
Message-ID: <87a5b0800808260130i383ec8d0y45690f85366c5059@mail.gmail.com>
Date:	Tue, 26 Aug 2008 09:30:22 +0100
From:	"Will Newton" <will.newton@...il.com>
To:	"Steven A. Falco" <sfalco@...ris.com>
Cc:	linux-kernel@...r.kernel.org
Subject: Re: [RFC] Driver for Real Time Clock chip ST M41T65

On Mon, Aug 25, 2008 at 6:27 PM, Steven A. Falco <sfalco@...ris.com> wrote:
> I will be using a Real Time Clock chip, ST M41T65, on a new embedded system.
> The chip is quite similar to the M41T8x family, which already has driver
> rtc-m41t80.c.
>
> Would it be more acceptible to generalize rtc-m41t80.c (perhaps renaming it to
> rtc-m41txx.c) or would it be better to create a new rtc-m41t6x.c?
>
> The main differences I see between the M41T65 and M41T80 are that:
>
> 1) The M41T65 watchdog timer has three bits controlling resolution (versus two
>   for the M41T80).
>
> 2) There is no register 0x13 for controlling square-wave output.
>
> Here is the patch I am currently using (on some prototype hardware), for
> reference.  I've introduced two new feature bits to cover the above
> differences.  I have not yet done any renaming, nor have I modified any of the
> defconfigs to account for a name change.
>
>        Steve Falco
>        sfalco@...ris.com
>
> diff --git a/drivers/rtc/rtc-m41t80.c b/drivers/rtc/rtc-m41t80.c
> index a3e0880..4d9b1a9 100644
> --- a/drivers/rtc/rtc-m41t80.c
> +++ b/drivers/rtc/rtc-m41t80.c
> @@ -58,18 +58,21 @@
>
>  #define M41T80_FEATURE_HT      (1 << 0)
>  #define M41T80_FEATURE_BL      (1 << 1)
> +#define M41T80_FEATURE_SQ      (1 << 2)
> +#define M41T80_FEATURE_WD      (1 << 3)

Perhaps these feature flags should be documented with a comment?

>  #define DRV_VERSION "0.05"
>
>  static const struct i2c_device_id m41t80_id[] = {
> -       { "m41t80", 0 },
> -       { "m41t81", M41T80_FEATURE_HT },
> -       { "m41t81s", M41T80_FEATURE_HT | M41T80_FEATURE_BL },
> -       { "m41t82", M41T80_FEATURE_HT | M41T80_FEATURE_BL },
> -       { "m41t83", M41T80_FEATURE_HT | M41T80_FEATURE_BL },
> -       { "m41st84", M41T80_FEATURE_HT | M41T80_FEATURE_BL },
> -       { "m41st85", M41T80_FEATURE_HT | M41T80_FEATURE_BL },
> -       { "m41st87", M41T80_FEATURE_HT | M41T80_FEATURE_BL },
> +       { "m41t65", M41T80_FEATURE_HT | M41T80_FEATURE_WD },
> +       { "m41t80", M41T80_FEATURE_SQ },
> +       { "m41t81", M41T80_FEATURE_HT | M41T80_FEATURE_SQ},
> +       { "m41t81s", M41T80_FEATURE_HT | M41T80_FEATURE_BL | M41T80_FEATURE_SQ },
> +       { "m41t82", M41T80_FEATURE_HT | M41T80_FEATURE_BL | M41T80_FEATURE_SQ },
> +       { "m41t83", M41T80_FEATURE_HT | M41T80_FEATURE_BL | M41T80_FEATURE_SQ },
> +       { "m41st84", M41T80_FEATURE_HT | M41T80_FEATURE_BL | M41T80_FEATURE_SQ },
> +       { "m41st85", M41T80_FEATURE_HT | M41T80_FEATURE_BL | M41T80_FEATURE_SQ },
> +       { "m41st87", M41T80_FEATURE_HT | M41T80_FEATURE_BL | M41T80_FEATURE_SQ },
>        { }
>  };
>  MODULE_DEVICE_TABLE(i2c, m41t80_id);
> @@ -385,8 +388,12 @@ static ssize_t m41t80_sysfs_show_sqwfreq(struct device *dev,
>                                struct device_attribute *attr, char *buf)
>  {
>        struct i2c_client *client = to_i2c_client(dev);
> +       struct m41t80_data *clientdata = i2c_get_clientdata(client);
>        int val;
>
> +       if (!(clientdata->features & M41T80_FEATURE_SQ))
> +               return -EINVAL;
> +
>        val = i2c_smbus_read_byte_data(client, M41T80_REG_SQW);
>        if (val < 0)
>                return -EIO;
> @@ -407,9 +414,13 @@ static ssize_t m41t80_sysfs_set_sqwfreq(struct device *dev,
>                                const char *buf, size_t count)
>  {
>        struct i2c_client *client = to_i2c_client(dev);
> +       struct m41t80_data *clientdata = i2c_get_clientdata(client);
>        int almon, sqw;
>        int val = simple_strtoul(buf, NULL, 0);
>
> +       if (!(clientdata->features & M41T80_FEATURE_SQ))
> +               return -EINVAL;
> +
>        if (val) {
>                if (!is_power_of_2(val))
>                        return -EINVAL;
> @@ -498,6 +509,8 @@ static void wdt_ping(void)
>                        .buf    = i2c_data,
>                },
>        };
> +       struct m41t80_data *clientdata = i2c_get_clientdata(save_client);
> +
>        i2c_data[0] = 0x09;             /* watchdog register */
>
>        if (wdt_margin > 31)
> @@ -508,6 +521,12 @@ static void wdt_ping(void)
>                 */
>                i2c_data[1] = wdt_margin<<2 | 0x82;
>
> +       /* M41T65 has three bits for resolution.  Don't set bit 7, as that would
> +        * be an invalid resolution.
> +        */

I believe the kernel comment style would be:

/*
 * M41T65 has three bits for resolution.  Don't set bit 7, as that would
 * be an invalid resolution.
 */

> +       if (clientdata->features & M41T80_FEATURE_WD)
> +               i2c_data[1] &= ~0x80;

Could this magic constant be moved to a #define?

> +
>        i2c_transfer(save_client->adapter, msgs1, 1);
>  }
>
> --
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to majordomo@...r.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at  http://www.tux.org/lkml/
>
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/