Re: [PATCH v10 0/7] Rust Abstractions for PWM subsystem with TH1520 PWM driver

[Michal Wilczynski <m.wilczynski@samsung.com>]

On 7/10/25 17:25, Uwe Kleine-König wrote:
> Hello Michal,
> 
> On Thu, Jul 10, 2025 at 03:48:08PM +0200, Michal Wilczynski wrote:
>> On 7/10/25 15:10, Uwe Kleine-König wrote:
>>> On Thu, Jul 10, 2025 at 10:42:07AM +0200, Michal Wilczynski wrote:
>>>> On 7/7/25 11:48, Michal Wilczynski wrote:
>>>>> The series is structured as follows:
>>>>>  - Expose static function pwmchip_release.
>>>
>>> Is this really necessary? I didn't try to understand the requirements
>>> yet, but I wonder about that. If you get the pwmchip from
>>> __pwmchip_add() the right thing to do to release it is to call
>>> pwmchip_remove(). Feels like a layer violation.
>>
>> It's required to prevent a memory leak in a specific, critical failure
>> scenario. The sequence of events is as follows:
>>
>>     pwm::Chip::new() succeeds, allocating both the C struct pwm_chip and
>>     the Rust drvdata.
>>
>>     pwm::Registration::register() (which calls pwmchip_add()) fails for
>>     some reason.
> 
> If you called pwmchip_alloc() but not yet pwmchip_add(), the right
> function to call for cleanup is pwmchip_put().
> 
>>     The ARef<Chip> returned by new() is dropped, its reference count
>>     goes to zero, and our custom release_callback is called.
>>
>> [...]
>>>>> ---
>>>>> base-commit: 47753b5a1696283930a78aae79b29371f96f5bca
>>>
>>> I have problems applying this series and don't have this base commit in
>>> my repo.
>>
>> Sorry for the confusion. Base commit doesn't exist in the mainline
>> kernel or linux-next, cause I've added some dependecies for compilation,
>> like IoMem for the driver (uploaded full branch on github [1]). The
>> bindings however doesn't depend on anything that's not in linux-next.
> 
> The series didn't apply to my pwm/for-next branch.
> 
> Note that the base-commit should always be a publically known commit.
> See the chapter about "Base Tree Information" in git-format-patch(1).

Hello Uwe,

Okay, thank you for the clarification. I understand the requirement for
a public base commit.

My intention was to include the TH1520 driver primarily as a practical
demonstration of the new abstractions. However the driver can't be
merged as is, since it depends on the unmerged IoMem series and won't
compile against a public commit.

I will rebase the series on pwm/for-next and drop the driver and its
associated device tree patches for now. I'll send a new version
containing just the core PWM abstraction patches, which apply cleanly.

I will resubmit the driver patches once their dependencies are available
in a public tree.

> 
> Best regards
> Uwe

Best regards,
-- 
Michal Wilczynski <m.wilczynski@samsung.com>