Re: [PATCH v4 16/18] drm/xe/vsec: Add PMT GUID internal access

[sashiko-bot@kernel.org]

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [Medium] `xe_vsec_get_guid()` returns 0 (success) instead of an error code for unsupported platforms.
--

commit 5ceb34095c8fcb408f81cf6301878d346676cb2c
Author: Michael J. Ruhl <michael.j.ruhl@intel.com>

drm/xe/vsec: Add PMT GUID internal access

This commit adds platform ID usage for the PMT GUID register access. It
provides a new API to allow accessing the GUID based on the platform for
internal usage by Xe.

> diff --git a/drivers/gpu/drm/xe/xe_vsec.c b/drivers/gpu/drm/xe/xe_vsec.c
> --- a/drivers/gpu/drm/xe/xe_vsec.c
> +++ b/drivers/gpu/drm/xe/xe_vsec.c
[ ... ]
> +int xe_vsec_get_guid(struct xe_device *xe, u32 *guid)
> +{
[ ... ]
> +	switch (xe->info.platform) {
> +	case XE_BATTLEMAGE:
> +		*guid = xe_mmio_read32(mmio, BMG_PUNIT_TELEMETRY_GUID);
> +		break;
> +
> +	case XE_CRESCENTISLAND:
> +		xe->soc_remapper.set_telem_region(xe, CRI_IDX_TELEM_DISCOVERY);
> +		*guid = xe_mmio_read32(mmio, CRI_PUNIT_TELEMETRY_GUID);
> +		break;
> +
> +	default:
> +		*guid = 0;
> +		drm_err(&xe->drm, "Unsupported platform: %u\n", xe->info.platform);
> +		break;
> +	}
> +
> +	xe->pmt.punit_guid_cache = *guid;
> +
> +unlock:
> +	mutex_unlock(&xe->pmt.lock);
> +	xe_pm_runtime_put(xe);
> +
> +	return 0;

[Severity: Medium]
Should this return a negative error code (such as -ENODEV) when an unsupported
platform is encountered?

Returning 0 (success) here means callers like read_residency_counter() will
proceed and call xe_pmt_telem_read() with an invalid GUID (0), which will fail
downstream and trigger additional warnings.

Additionally, because *guid is 0, xe->pmt.punit_guid_cache is set to 0. Since
0 evaluates to false in the cache check earlier in the function:

    if (xe->pmt.punit_guid_cache) {
        *guid = xe->pmt.punit_guid_cache;
        goto unlock;
    }

the cache will be bypassed on subsequent calls, causing the kernel to log the
"Unsupported platform" error every time this is called (for example, when
userspace repeatedly reads from debugfs).

> +}

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20260901192736.626777-20-michael.j.ruhl@intel.com?part=16