[PATCH v2 3/4] drm/mgag200: Use simple encoder

Daniel Vetter daniel at ffwll.ch
Fri Feb 21 19:03:07 UTC 2020


On Fri, Feb 21, 2020 at 08:00:57PM +0100, Sam Ravnborg wrote:
> Hi Thomas.
> 
> On Fri, Feb 21, 2020 at 08:48:48AM +0100, Thomas Zimmermann wrote:
> > Hi Sam
> > 
> > thanks for reviewing the patch set.
> > 
> > Am 20.02.20 um 19:56 schrieb Sam Ravnborg:
> > > Hi Thomas.
> > > 
> > > On Tue, Feb 18, 2020 at 09:48:14AM +0100, Thomas Zimmermann wrote:
> > >> The mgag200 driver uses an empty implementation for its encoder. Replace
> > >> the code with the generic simple encoder.
> > >>
> > >> v2:
> > >> 	* rebase onto new simple-encoder interface
> > >>
> > >> Signed-off-by: Thomas Zimmermann <tzimmermann at suse.de>
> > >> ---
> > >>  drivers/gpu/drm/mgag200/mgag200_drv.h  |  7 ---
> > >>  drivers/gpu/drm/mgag200/mgag200_mode.c | 61 ++------------------------
> > >>  2 files changed, 3 insertions(+), 65 deletions(-)
> > >>
> > >> diff --git a/drivers/gpu/drm/mgag200/mgag200_drv.h b/drivers/gpu/drm/mgag200/mgag200_drv.h
> > >> index aa32aad222c2..9bb9e8e14539 100644
> > >> --- a/drivers/gpu/drm/mgag200/mgag200_drv.h
> > >> +++ b/drivers/gpu/drm/mgag200/mgag200_drv.h
> > >> @@ -95,7 +95,6 @@
> > >>  #define MATROX_DPMS_CLEARED (-1)
> > >>  
> > >>  #define to_mga_crtc(x) container_of(x, struct mga_crtc, base)
> > >> -#define to_mga_encoder(x) container_of(x, struct mga_encoder, base)
> > >>  #define to_mga_connector(x) container_of(x, struct mga_connector, base)
> > >>  
> > >>  struct mga_crtc {
> > >> @@ -110,12 +109,6 @@ struct mga_mode_info {
> > >>  	struct mga_crtc *crtc;
> > >>  };
> > >>  
> > >> -struct mga_encoder {
> > >> -	struct drm_encoder base;
> > >> -	int last_dpms;
> > >> -};
> > >> -
> > >> -
> > >>  struct mga_i2c_chan {
> > >>  	struct i2c_adapter adapter;
> > >>  	struct drm_device *dev;
> > > 
> > > Any particular reason why the drm_encoder is not embedded in struct
> > > mga_device?
> > > 
> > > I found it more elegant - like you did it for ast in the previous patch.
> > 
> > I think I wanted something that uses drm_simple_encoder_create(). But I
> > can change that. The embedded variant is indeed better.
> 
> You should consider to drop drm_simple_encoder_create() until there
> is a driver that really needs it.

Yeah +1 on only the _init version. The create version really should use
drmm_kzalloc I think, but we're not quite there yet :-)
-Daniel
-- 
Daniel Vetter
Software Engineer, Intel Corporation
http://blog.ffwll.ch


More information about the dri-devel mailing list