Re: [PATCH] drm/i915: Convert intel_runtime_pm_get_noresume towards raw wakeref

[Imre Deak <imre.deak@intel.com>]

On Thu, Apr 18, 2024 at 06:13:20PM -0400, Rodrigo Vivi wrote:
> In the past, the noresume function was used by the GEM code to ensure
> wakelocks were held and bump its usage. This is no longer the case
> and this function was totally unused until it started to be used again
> by display with commit 77e619a82fc3 ("drm/i915/display: convert inner
> wakeref get towards get_if_in_use")
> 
> However, on the display code, most of the callers are using the
> raw wakeref, rather then the wakelock version. What caused a
> major regression caught by CI.
> 
> Another option to this patch is to go with the original plan and
> use the get_if_in_use variant in the display code, what is enough
> to fulfil our needs. Then, an extra patch to delete the unused
> _noresume variant.
> 
> Cc: Imre Deak <imre.deak@intel.com>
> Cc: Francois Dugast <francois.dugast@intel.com>
> Cc: Ville Syrjälä <ville.syrjala@linux.intel.com>
> Fixes: 77e619a82fc3 ("drm/i915/display: convert inner wakeref get towards get_if_in_use")
> Closes: https://gitlab.freedesktop.org/drm/intel/-/issues/10875
> Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
> ---
>  .../gpu/drm/i915/display/intel_display_power.c    |  6 ------
>  drivers/gpu/drm/i915/intel_runtime_pm.c           | 15 +++++----------
>  2 files changed, 5 insertions(+), 16 deletions(-)
> 
> diff --git a/drivers/gpu/drm/i915/display/intel_display_power.c b/drivers/gpu/drm/i915/display/intel_display_power.c
> index 048943d0a881..03dc7edcc443 100644
> --- a/drivers/gpu/drm/i915/display/intel_display_power.c
> +++ b/drivers/gpu/drm/i915/display/intel_display_power.c
> @@ -640,12 +640,6 @@ release_async_put_domains(struct i915_power_domains *power_domains,
>  	enum intel_display_power_domain domain;
>  	intel_wakeref_t wakeref;
>  
> -	/*
> -	 * The caller must hold already raw wakeref, upgrade that to a proper
> -	 * wakeref to make the state checker happy about the HW access during
> -	 * power well disabling.
> -	 */
> -	assert_rpm_raw_wakeref_held(rpm);
>  	wakeref = intel_runtime_pm_get_noresume(rpm);
>  
>  	for_each_power_domain(domain, mask) {
> diff --git a/drivers/gpu/drm/i915/intel_runtime_pm.c b/drivers/gpu/drm/i915/intel_runtime_pm.c
> index d4e844128826..e27b2ab82da0 100644
> --- a/drivers/gpu/drm/i915/intel_runtime_pm.c
> +++ b/drivers/gpu/drm/i915/intel_runtime_pm.c
> @@ -272,15 +272,11 @@ intel_wakeref_t intel_runtime_pm_get_if_active(struct intel_runtime_pm *rpm)
>   * intel_runtime_pm_get_noresume - grab a runtime pm reference
>   * @rpm: the intel_runtime_pm structure
>   *
> - * This function grabs a device-level runtime pm reference (mostly used for GEM
> - * code to ensure the GTT or GT is on).
> + * This function grabs a runtime pm reference.
>   *
> - * It will _not_ power up the device but instead only check that it's powered
> - * on.  Therefore it is only valid to call this functions from contexts where
> - * the device is known to be powered up and where trying to power it up would
> - * result in hilarity and deadlocks. That pretty much means only the system
> - * suspend/resume code where this is used to grab runtime pm references for
> - * delayed setup down in work items.
> + * It will _not_ resume the device but instead only get an extra wakeref.
> + * Therefore it is only valid to call this functions from contexts where
> + * the device is known to be active and with another wakeref previously hold.
>   *
>   * Any runtime pm reference obtained by this function must have a symmetric
>   * call to intel_runtime_pm_put() to release the reference again.
> @@ -289,10 +285,9 @@ intel_wakeref_t intel_runtime_pm_get_if_active(struct intel_runtime_pm *rpm)
>   */
>  intel_wakeref_t intel_runtime_pm_get_noresume(struct intel_runtime_pm *rpm)
>  {
> -	assert_rpm_wakelock_held(rpm);
>  	pm_runtime_get_noresume(rpm->kdev);
>  
> -	intel_runtime_pm_acquire(rpm, true);
> +	intel_runtime_pm_acquire(rpm, false);

This needs to stay a wakelock, so that the HW access in
release_async_put_domains() will not lead to a wakelock not held assert.
Only the above assert_rpm_wakelock_held() needs to be changed to
assert_rpm_raw_wakeref_held().

>  
>  	return track_intel_runtime_pm_wakeref(rpm);
>  }
> -- 
> 2.44.0
>