[Intel-gfx] [PATCH 4/4] drm/i915: Move scanline_offset and timestamping constant setup to intel_modeset_readout_hw_state()

[Maarten Lankhorst]

Hey,

Op 14-09-15 om 16:23 schreef Ville Syrjälä:
> On Mon, Sep 14, 2015 at 01:57:45PM +0200, Maarten Lankhorst wrote:
>> Op 10-09-15 om 17:59 schreef ville.syrjala at linux.intel.com:
>>> From: Ville Syrjälä <ville.syrjala at linux.intel.com>
>>>
>>> intel_modeset_readout_hw_state() seems like the more appropriate place
>>> for populating the scanline_offset and timestamping constants than
>>> intel_sanitize_crtc() since they are basically part of the state we
>>> read out.
>>>
>>> Cc: Maarten Lankhorst <maarten.lankhorst at linux.intel.com>
>>> Cc: Patrik Jakobsson <patrik.jakobsson at linux.intel.com>
>>> Signed-off-by: Ville Syrjälä <ville.syrjala at linux.intel.com>
>>> ---
>>>  drivers/gpu/drm/i915/intel_display.c | 5 +++--
>>>  1 file changed, 3 insertions(+), 2 deletions(-)
>>>
>>> diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
>>> index 5fed120..88d9466 100644
>>> --- a/drivers/gpu/drm/i915/intel_display.c
>>> +++ b/drivers/gpu/drm/i915/intel_display.c
>>> @@ -14918,8 +14918,6 @@ static void intel_sanitize_crtc(struct intel_crtc *crtc)
>>>  	if (crtc->active) {
>>>  		struct intel_plane *plane;
>>>  
>>> -		drm_calc_timestamping_constants(&crtc->base, &crtc->base.hwmode);
>>> -		update_scanline_offset(crtc);
>>>  		drm_crtc_vblank_on(&crtc->base);
>>>  
>>>  		/* Disable everything but the primary plane */
>>> @@ -15216,6 +15214,9 @@ static void intel_modeset_readout_hw_state(struct drm_device *dev)
>>>  			 * recalculation.
>>>  			 */
>>>  			crtc->base.state->mode.private_flags = I915_MODE_FLAG_INHERITED;
>>> +
>>> +			drm_calc_timestamping_constants(&crtc->base, &crtc->base.hwmode);
>>> +			update_scanline_offset(crtc);
>>>  		}
>>>  	}
>>>  }
>> Can you move drm_vblank_reset and drm_vblank_on too?
> I'm not sure I really want to move those. They can actually modify
> the hardware state, so I don't think they really belong in
> intel_modeset_readout_hw_state(). intel_sanitize_crtc() feels like
> a better fit.
Well the changes are all useful and they're blocking other work. So I think it's better these get in as is.

Reviewed-by: Maarten Lankhorst <maarten.lankhorst at linux.intel.com>