On 05/20, Boris Brezillon wrote:
> Clock rates are stored in an unsigned long field, but ->determine_rate()
> (which returns a rounded rate from a requested one) returns a long
> value (errors are reported using negative error codes), which can lead
> to long overflow if the clock rate exceed 2Ghz.
>
> Change ->determine_rate() prototype to return 0 or an error code, and pass
> a pointer to a clk_rate_request structure containing the expected target
> rate and the rate constraints imposed by clk users.
>
> The clk_rate_request structure might be extended in the future to contain
> other kind of constraints like the rounding policy, the maximum clock
> inaccuracy or other things that are not yet supported by the CCF
> (power consumption constraints ?).
>
> Signed-off-by: Boris Brezillon <boris.brezillon@xxxxxxxxxxxxxxxxxx>
>
> CC: Jonathan Corbet <corbet@xxxxxxx>
> CC: Tony Lindgren <tony@xxxxxxxxxxx>
> CC: Ralf Baechle <ralf@xxxxxxxxxxxxxx>
> CC: "Emilio López" <emilio@xxxxxxxxxxxxx>
> CC: Maxime Ripard <maxime.ripard@xxxxxxxxxxxxxxxxxx>
> CC: Tero Kristo <t-kristo@xxxxxx>
> CC: linux-doc@xxxxxxxxxxxxxxx
> CC: linux-kernel@xxxxxxxxxxxxxxx
> CC: linux-arm-kernel@xxxxxxxxxxxxxxxxxxx
> CC: linux-omap@xxxxxxxxxxxxxxx
> CC: linux-mips@xxxxxxxxxxxxxx
> ---
>
> Hi Stephen,
>
> This patch is based on clk-next and contains the changes you suggested
> in your previous review.
>
> It was tested on sama5d4 and compile tested on several ARM platforms
> (those enabled in multi_v7_defconfig).
>
Thanks. I think we should wait until the next -rc1 drops to apply the
patch for the next merge window. That will make it least likely to conflict
with other trees, and we can provide it on a stable branch should there
be clock providers going through other trees somewhere. Please
remind me if I forget.
> @@ -1186,15 +1191,21 @@ EXPORT_SYMBOL_GPL(__clk_determine_rate);
> */
> unsigned long __clk_round_rate(struct clk *clk, unsigned long rate)
> {
> - unsigned long min_rate;
> - unsigned long max_rate;
> +
> + struct clk_rate_request req;
> + int ret;
>
> if (!clk)
> return 0;
>
> - clk_core_get_boundaries(clk->core, &min_rate, &max_rate);
> + clk_core_get_boundaries(clk->core, &req.min_rate, &req.max_rate);
> + req.rate = rate;
> +
> + ret = clk_core_round_rate_nolock(clk->core, &req);
> + if (ret)
> + return ret;
This returns a negative int for unsigned long. Is that intentional?
--
Qualcomm Innovation Center, Inc. is a member of Code Aurora Forum,
a Linux Foundation Collaborative Project
