Re: [PATCH v2] memory: tegra: Deduplicate rate request management code

[Krzysztof Kozlowski <krzk@kernel.org>]

On 30/04/2026 08:52, Mikko Perttunen wrote:
> On Tuesday, April 28, 2026 6:12 PM Krzysztof Kozlowski wrote:
>> On 17/04/2026 11:19, Mikko Perttunen wrote:
>>> +
>>> +struct tegra_emc_rate_request {
>>> +	unsigned long min_rate;
>>> +	unsigned long max_rate;
>>> +};
>>> +
>>> +struct tegra_emc_rate_requests {
>>> +	struct tegra_emc_rate_request requested_rate[TEGRA_EMC_RATE_TYPE_MAX];
>>> +	struct mutex rate_lock;
>>
>> You introduced a new checkpatch warning. Wasn't there before. Probably
>> the old comment feels to obvious in this context, so it should be
>> extended to be meaningful, e.g. what data structure, which members,
>> which code paths etc.
> 
> On my system, it's just a 'CHECK', not a 'WARNING', so I thought it
> not necessary to address in this relatively simple circumstance.
> 
> However, I do agree it's best to err on the side of verbosity when
> dealing with locking, so I'll add a description for v3.

It's still a warning both in CI and maintainer commit tools. The meaning
of "CHECK" means only amount of false positives, so you are supposed to
check it. And if you checked it you would see it is valid report.

Best regards,
Krzysztof