Re: [PATCH 4/4] watchdog: st_wdt: Add new driver for ST's LPC Watchdog

From: Guenter Roeck
Date: Mon Sep 08 2014 - 09:32:03 EST


On 09/08/2014 05:32 AM, Lee Jones wrote:
On Fri, 05 Sep 2014, Guenter Roeck wrote:
...
+
+static struct st_wdog_syscfg stid127_syscfg = {
+ .type_mask = BIT(2),
+ .enable_mask = BIT(2),
+};
+
+static struct st_wdog_syscfg stih415_syscfg = {
+ .type_mask = BIT(6),
+ .enable_mask = BIT(7),
+};
+
+static struct st_wdog_syscfg stih416_syscfg = {
+ .type_mask = BIT(6),
+ .enable_mask = BIT(7),
+};
+
+static struct st_wdog_syscfg stih407_syscfg = {
+ .enable_mask = BIT(19),
+};
+
...

+ /* Mask/unmask watchdog reset */
+ regmap_update_bits(st_wdog->syscfg->regmap,
+ st_wdog->syscfg->enable_reg,
+ st_wdog->syscfg->enable_mask,
+ !enable);

enable is a bool, but is supposed to provide the value to be put into the
register, masked with enable_mask. Unless I am missing something, the value
is not shifted in regmap_update_bits. So I don't think this can work, but
effectively always writes zero into the mask unless the mask happens to be
at bit position 0 - which never happens.

Same is true for warm_reset above, which also has values 0 or 1.

I know it does not really matter in C (at least when it comes to handling
0 and 1), but I would suggest to avoid mixing booleans with bit masks.

You're right of course, great spot.

How about?

!enable << ffs(st_wdog->syscfg->enable_mask).

Seems to add a lot of complexity (as in 'makes it difficult to understand')
to avoid a conditional, and assumes that enable_mask will never have more
than one bit set. I would go with
enable ? st_wdog->syscfg->enable_mask : 0
to avoid confusion, but your call.

Guenter

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/