Re: [PATCH v5 4/4] drm/xe/forcewake: enable configurable delayed forcewake release

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [Low] The function `query_engine_cycles` mixes goto-based error handling with scope-based cleanup macros.
--

--- Patch [4]: [PATCH v5 4/4] drm/xe/forcewake: enable configurable delayed forcewake release ---
commit 7d3e5746e331fb4da88407a96b40b25e06abf763
Author: Zongyao Bai <[email protected]>

drm/xe/forcewake: enable configurable delayed forcewake release

This commit introduces a configurable delayed forcewake release for
engine-cycle queries to avoid repeated wake and ACK cycles during bursts.
It adds a configfs attribute to control the hold interval before releasing
an idle forcewake domain.

> diff --git a/drivers/gpu/drm/xe/xe_query.c b/drivers/gpu/drm/xe/xe_query.c
> --- a/drivers/gpu/drm/xe/xe_query.c
> +++ b/drivers/gpu/drm/xe/xe_query.c
> @@ -168,7 +168,7 @@ query_engine_cycles(struct xe_device *xe,
>  
>  	fw_domain = xe_hw_engine_to_fw_domain(hwe);
>  
> -	xe_with_force_wake(fw_ref, gt_to_fw(gt), fw_domain) {
> +	xe_with_force_wake_delay(fw_ref, gt_to_fw(gt), fw_domain) {
>  		if (!xe_force_wake_ref_has_domain(fw_ref.domains, fw_domain)) {
>  			err = -EIO;
>  			goto out;

[Severity: Low]
Does this code mix goto-based error handling with scope-based cleanup macros?

According to the cleanup subsystem guidelines, the expectation is that usage of
goto and cleanup helpers is never mixed in the same function.

The function query_engine_cycles() uses goto out; for manual drm_dev_exit()
cleanup while simultaneously using the xe_with_force_wake_delay() macro, which
internally uses the scope-based CLASS() cleanup macro. While jumping out of the
loop block correctly invokes the scope destructor here without causing a
double-free or resource leak, it violates the subsystem structural guidelines
against mixing these cleanup styles.

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/[email protected]?part=4