Re: [PATCH v2 4/5] media: iris: Fix power-off ordering to disable power domain after clocks

[sashiko-bot@kernel.org]

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:

Pre-existing issues:
- [High] The patch fixes the clock/power domain disable sequence in `iris_vpu_power_off_hw()` but misses identical sequence bugs in `iris_vpu35_power_off_hw()` and `iris_vpu33_power_off_controller()`, leaving them to operate on unpowered clock controllers.
--

--- Patch [4]: [PATCH v2 4/5] media: iris: Fix power-off ordering to disable power domain after clocks ---
commit f43aeec1ae78e41029af71f7d7db01a62fedc472
Author: Vishnu Reddy <busanna.reddy@oss.qualcomm.com>

media: iris: Fix power-off ordering to disable power domain after clocks

This commit fixes the power-off ordering in iris_vpu_power_off_hw() to
ensure clocks are disabled before the associated power domain is removed,
preventing unpredictable behavior or hardware hangs.

> diff --git a/drivers/media/platform/qcom/iris/iris_vpu_common.c b/drivers/media/platform/qcom/iris/iris_vpu_common.c
> index e4847c1077097..3cd493b06d84c 100644
> --- a/drivers/media/platform/qcom/iris/iris_vpu_common.c
> +++ b/drivers/media/platform/qcom/iris/iris_vpu_common.c
> @@ -223,10 +223,10 @@ int iris_vpu_power_off_controller(struct iris_core *core)
>  void iris_vpu_power_off_hw(struct iris_core *core)
>  {
>  	dev_pm_genpd_set_hwmode(core->pmdomain_tbl->pd_devs[IRIS_HW_POWER_DOMAIN], false);
> -	iris_disable_power_domains(core, core->pmdomain_tbl->pd_devs[IRIS_HW_POWER_DOMAIN]);
>  	iris_disable_unprepare_clock(core, IRIS_BSE_HW_CLK);
>  	iris_disable_unprepare_clock(core, IRIS_HW_AHB_CLK);
>  	iris_disable_unprepare_clock(core, IRIS_HW_CLK);
> +	iris_disable_power_domains(core, core->pmdomain_tbl->pd_devs[IRIS_HW_POWER_DOMAIN]);
>  }

[Severity: High]
This is a pre-existing issue and was not introduced by this patch, but does
this same ordering problem exist in
drivers/media/platform/qcom/iris/iris_vpu3x.c?

Looking at iris_vpu33_power_off_controller(), IRIS_CTRL_POWER_DOMAIN is
removed before disabling IRIS_AXI_CLK:

disable_power:
	iris_disable_power_domains(core, core->pmdomain_tbl->pd_devs[IRIS_CTRL_POWER_DOMAIN]);
	iris_disable_unprepare_clock(core, IRIS_AXI_CLK);

Similarly, in iris_vpu35_power_off_hw(), iris_vpu33_power_off_hardware()
removes the domain before IRIS_HW_FREERUN_CLK and IRIS_AXI_CLK are disabled:

	iris_vpu33_power_off_hardware(core);

	iris_disable_unprepare_clock(core, IRIS_HW_FREERUN_CLK);
	iris_disable_unprepare_clock(core, IRIS_AXI_CLK);

Should these paths also be updated to prevent operating on unpowered clock
controllers and causing system hangs on device teardown?

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20260818-iris-fixes-dma-pseq-fint-v2-0-8dc8c4b3880e@oss.qualcomm.com?part=4