Re: [PATCH] drm/i915: vlv: init only needed state during early power well enabling

[Daniel Vetter <daniel@ffwll.ch>]

On Fri, Apr 25, 2014 at 01:19:05PM +0300, Imre Deak wrote:
> During the initial power well enabling on the driver init/resume path
> we can avoid initialzing part of the HW/SW state that will be
> initialized anyway by the subsequent init/resume code. For some steps
> like HPD initialization this redundancy is not only an overhead but an
> actual problem, since they can't be run this early in the overall init
> sequence.
> 
> Add a flag marking the init phase and skip reinitialzing state that is
> not strictly necessary based on that.
> 
> This is also needed by the upcoming HPD init restructuring by Thierry
> and Daniel.
> 
> Signed-off-by: Imre Deak <imre.deak@intel.com>

Not too happy with the layering violation. But we've discussed this
extensively on irc, and there doesn't seem to be anything less ugly
available.

Queued for -next, thanks for the patch.
-Daniel

> ---
>  drivers/gpu/drm/i915/i915_drv.h |  1 +
>  drivers/gpu/drm/i915/intel_pm.c | 14 ++++++++++----
>  2 files changed, 11 insertions(+), 4 deletions(-)
> 
> diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
> index 7d6acb4..b589fc0 100644
> --- a/drivers/gpu/drm/i915/i915_drv.h
> +++ b/drivers/gpu/drm/i915/i915_drv.h
> @@ -929,6 +929,7 @@ struct i915_power_domains {
>  	 * time are on. They are kept on until after the first modeset.
>  	 */
>  	bool init_power_on;
> +	bool initializing;
>  	int power_well_count;
>  
>  	struct mutex lock;
> diff --git a/drivers/gpu/drm/i915/intel_pm.c b/drivers/gpu/drm/i915/intel_pm.c
> index 75c1c76..34b689f 100644
> --- a/drivers/gpu/drm/i915/intel_pm.c
> +++ b/drivers/gpu/drm/i915/intel_pm.c
> @@ -5622,11 +5622,13 @@ static void vlv_display_power_well_enable(struct drm_i915_private *dev_priv,
>  	spin_unlock_irq(&dev_priv->irq_lock);
>  
>  	/*
> -	 * During driver initialization we need to defer enabling hotplug
> -	 * processing until fbdev is set up.
> +	 * During driver initialization/resume we can avoid restoring the
> +	 * part of the HW/SW state that will be inited anyway explicitly.
>  	 */
> -	if (dev_priv->enable_hotplug_processing)
> -		intel_hpd_init(dev_priv->dev);
> +	if (dev_priv->power_domains.initializing)
> +		return;
> +
> +	intel_hpd_init(dev_priv->dev);
>  
>  	i915_redisable_vga_power_on(dev_priv->dev);
>  }
> @@ -5990,9 +5992,13 @@ static void intel_power_domains_resume(struct drm_i915_private *dev_priv)
>  
>  void intel_power_domains_init_hw(struct drm_i915_private *dev_priv)
>  {
> +	struct i915_power_domains *power_domains = &dev_priv->power_domains;
> +
> +	power_domains->initializing = true;
>  	/* For now, we need the power well to be always enabled. */
>  	intel_display_set_init_power(dev_priv, true);
>  	intel_power_domains_resume(dev_priv);
> +	power_domains->initializing = false;
>  }
>  
>  void intel_aux_display_runtime_get(struct drm_i915_private *dev_priv)
> -- 
> 1.8.4
> 
> _______________________________________________
> Intel-gfx mailing list
> Intel-gfx@lists.freedesktop.org
> http://lists.freedesktop.org/mailman/listinfo/intel-gfx

-- 
Daniel Vetter
Software Engineer, Intel Corporation
+41 (0) 79 365 57 48 - http://blog.ffwll.ch