Re: [RFC PATCH] reset: Add devm_reset_control_deassert helper

[Philipp Zabel]

Hi,

On Do, 2024-04-18 at 14:26 +0100, Prabhakar wrote:
> From: Lad Prabhakar <prabhakar.mahadev-lad.rj@xxxxxxxxxxxxxx>
> 
> A typical code pattern for reset_control_deassert() call is to call it in
> the _probe function and to call reset_control_assert() both from _probe
> error path and from _remove function.
> 
> Add helper function to replace this bolierplate piece of code. Calling
> devm_reset_control_deassert() removes the need for calling
> reset_control_assert() both in the probe()'s error path and in the
> remove() function.
> 
> Suggested-by: Geert Uytterhoeven <geert+renesas@xxxxxxxxx>
> Signed-off-by: Lad Prabhakar <prabhakar.mahadev-lad.rj@xxxxxxxxxxxxxx>

I'm not sure this aligns well with the intended use of devres for
resource acquisition and release.

Note how there is no devm_clk_prepare_enable, devm_regulator_enable,
devm_gpiod_set_value_cansleep, or devm_pwm_enable either.

I've sent an alternative suggestion that adds
devm_reset_control_get..._deasserted calls, similarly to the existing
devm_clk_get..._enabled calls. Please let me know what you think.

regards
Philipp