Re: [RFC 4/5] clk: per-user clock accounting for debug

From: Stephen Warren
Date: Fri Jun 27 2014 - 18:44:42 EST


On 06/27/2014 01:57 AM, Tomeu Vizoso wrote:
> From: Rabin Vincent <rabin.vincent@xxxxxxxxxxxxxx>
>
> When a clock has multiple users, the WARNING on imbalance of
> enable/disable may not show the guilty party since although they may
> have commited the error earlier, the warning is emitted later when some
> other user, presumably innocent, disables the clock.
>
> Provide per-user clock enable/disable accounting and disabler tracking
> in order to help debug these problems.
>
> NOTE: with this patch, clk_get_parent() behaves like clk_get(), i.e. it
> needs to be matched with a clk_put(). Otherwise, memory will leak.

> diff --git a/include/linux/clk-private.h b/include/linux/clk-private.h
> index 91659b2..9657fc8 100644
> --- a/include/linux/clk-private.h
> +++ b/include/linux/clk-private.h
> @@ -56,7 +56,11 @@ struct clk_core {
> };
>
> struct clk {
> - struct clk_core clk;
> + struct clk_core *core;
> + unsigned int enable_count;
> + const char *dev_id;
> + const char *con_id;

Why not just store the "struct device *" there instead of pulling the
name out of it, so ...

> diff --git a/drivers/clk/clk.c b/drivers/clk/clk.c

> void clk_disable(struct clk *clk_user)
> {
> - __clk_disable_internal(clk_to_clk_core(clk_user));
> + struct clk_core *clk = clk_to_clk_core(clk_user);
> + unsigned long flags;
> +
> + flags = clk_enable_lock();
> + if (!WARN(clk_user->enable_count == 0,
> + "incorrect disable clk dev %s con %s last disabler %pF\n",
> + clk_user->dev_id, clk_user->con_id, clk_user->last_disable)) {

Here, you could do something like:

if (!clk_user->enable_count) {
dev_err(clk_user->dev, "", ...);
goto out;
}
...
out:
clk_enable_unlock(flags);
}

I suppose that has the disadvantage of not using WARN() so not
generating a full back-trace. Still, you could keep the use of WARN()
and pass as a parameter to the printf parameters dev_name(clk_user->dev)
rather than manually saving the fields separately.
--
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/