On 06/10/2021 15:29, Sam Protsenko wrote:
> On Wed, 6 Oct 2021 at 15:38, Krzysztof Kozlowski
> <krzysztof.kozlowski@xxxxxxxxxxxxx> wrote:
>>
>> On 06/10/2021 12:46, Sam Protsenko wrote:
>>> On Wed, 15 Sept 2021 at 11:21, Krzysztof Kozlowski
>>> <krzysztof.kozlowski@xxxxxxxxxxxxx> wrote:
>>>>
>>>> On 14/09/2021 17:56, Sam Protsenko wrote:
>>>>> By default if bus clock has no users its "enable count" value is 0. It
>>>>> might be actually running if it's already enabled in bootloader, but
>>>>> then in some cases it can be disabled by mistake. For example, such case
>>>>> was observed when dw_mci_probe() enabled bus clock, then failed to do
>>>>> something and disabled that bus clock on error path. After that even
>>>>> attempt to read the 'clk_summary' file in DebugFS freezed forever, as
>>>>> CMU bus clock ended up being disabled and it wasn't possible to access
>>>>> CMU registers anymore.
>>>>>
>>>>> To avoid such cases, CMU driver must increment the ref count for that
>>>>> bus clock by running clk_prepare_enable(). There is already existing
>>>>> '.clk_name' field in struct samsung_cmu_info, exactly for that reason.
>>>>> It was added in commit 523d3de41f02 ("clk: samsung: exynos5433: Add
>>>>> support for runtime PM"). But the clock is actually enabled only in
>>>>> Exynos5433 clock driver. Let's mimic what is done there in generic
>>>>> samsung_cmu_register_one() function, so other drivers can benefit from
>>>>> that `.clk_name' field. As was described above, it might be helpful not
>>>>> only for PM reasons, but also to prevent possible erroneous clock gating
>>>>> on error paths.
>>>>>
>>>>> Another way to workaround that issue would be to use CLOCK_IS_CRITICAL
>>>>> flag for corresponding gate clocks. But that might be not very good
>>>>> design decision, as we might still want to disable that bus clock, e.g.
>>>>> on PM suspend.
>>>>>
>>>>> Signed-off-by: Sam Protsenko <semen.protsenko@xxxxxxxxxx>
>>>>> ---
>>>>> drivers/clk/samsung/clk.c | 13 +++++++++++++
>>>>> 1 file changed, 13 insertions(+)
>>>>>
>>>>> diff --git a/drivers/clk/samsung/clk.c b/drivers/clk/samsung/clk.c
>>>>> index 1949ae7851b2..da65149fa502 100644
>>>>> --- a/drivers/clk/samsung/clk.c
>>>>> +++ b/drivers/clk/samsung/clk.c
>>>>> @@ -357,6 +357,19 @@ struct samsung_clk_provider * __init samsung_cmu_register_one(
>>>>>
>>>>> ctx = samsung_clk_init(np, reg_base, cmu->nr_clk_ids);
>>>>>
>>>>> + /* Keep bus clock running, so it's possible to access CMU registers */
>>>>> + if (cmu->clk_name) {
>>>>> + struct clk *bus_clk;
>>>>> +
>>>>> + bus_clk = __clk_lookup(cmu->clk_name);
>>>>> + if (bus_clk) {
>>>>> + clk_prepare_enable(bus_clk);
>>>>> + } else {
>>>>> + pr_err("%s: could not find bus clock %s\n", __func__,
>>>>> + cmu->clk_name);
>>>>> + }
>>>>> + }
>>>>> +
>>>>
>>>> Solving this problem in generic way makes sense but your solution is
>>>> insufficient. You skipped suspend/resume paths and in such case you
>>>> should remove the Exynos5433-specific code.
>>>>
>>>
>>> Keeping core bus clocks always running seems like a separate
>>> independent feature to me (not related to suspend/resume). It's
>>> mentioned in commit 523d3de41f02 ("clk: samsung: exynos5433: Add
>>> support for runtime PM") this way:
>>>
>>> "Also for each CMU there is one special parent clock, which has to
>>> be enabled all the time when any access to CMU registers is being
>>> done."
>>>
>>> Why do you think suspend/resume paths have to be implemented along
>>> with it? Btw, I didn't add PM ops in clk-exynos850, as PM is not
>>> implemented on my board yet and I can't test it.
>>
>> You can skip the runtime PM, so keep your patch almost like it is now
>> (in respect to Sylwester's comment about __clk_lookup). However now the
>> Exynos5433 will enable the clk_name twice: here and in
>> exynos5433_cmu_probe().
>>
>> If you keep this approach, you need to remove duplicated part in
>> exynos5433_cmu_probe()...
>>
>
> My patch is only touching samsung_cmu_register_one(), and
> exynos5433_cmu_probe() doesn't call samsung_cmu_register_one(). So I
> don't think there can be a problem there. Or I'm missing something?
>
> samsung_cmu_register_one() is actually called from 5433 clk driver,
> but only from CMUs registered with CLK_OF_DECLARE(), and those are not
> setting .clk_name field, so my code is not affecting those either.
You are right.
>
> Real problem I can see is that I can't avoid using __clk_lookup() if I
> implement that code in samsung_cmu_register_one(). Tried to do use
> clk_get(NULL, ...) instead, but it doesn't work with 1st param (dev)
> being NULL, because samsung_clk_register_*() functions don't register
> clkdev (only samsung_clk_register_fixed_rate() does), hence
> LIST_HEAD(clocks) is empty in clkdev.c, and clk_get() fails, when not
> provided with actual 'dev' param, which in turn is not present in
> samsung_cmu_register_one()...
>
> About using platform_driver: as I can see from clk-exynos5433.c, only
> CMUs which belong to Power Domains are registered as platform_driver.
> Rest of CMUs are registered using CLK_OF_DECLARE(), thus they don't
> get platform_device param. That makes it harder to avoid using
> __clk_lookup() inside samsung_cmu_register_one().
>
> All that said, I feel like correct way to implement this patch would be:
> 1. Register all PD-capable CMUs as platform_driver in clk-exynos850
> (all CMUs except CMU_TOP)
> 2. Move bus clock enablement code from samsung_cmu_register_one() to
> corresponding clk-exynos850 probe function
>
> This way I would be able to use clk_get(dev, ...) instead of
> __clk_lookup(), and that won't affect any existing code for sure. Code
> will be more unified w.r.t. how it's done in clk-exynos5433, and
> platform_device will be a foundation for implementing PM ops later.
> Taking into account how much design decisions should be done for using
> that in common code -- I'd say let's do that later, as a separate
> refactoring activity.
>
> Do you think that makes sense?
Yes, makes sense. Thank you!
Best regards,
Krzysztof
