| Message ID | 20200207135950.6655-7-ville.syrjala@linux.intel.com (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
| Series | drm: Try to fix encoder possible_clones/crtc | expand |
On Fri, Feb 07, 2020 at 03:59:50PM +0200, Ville Syrjala wrote: > From: Ville Syrjälä <ville.syrjala@linux.intel.com> > > WARN if the encoder possible_crtcs is effectively empty or contains > bits for non-existing crtcs. > > TODO: Or should we perhapst just filter out any bit for a > non-exisiting crtc? > > Cc: Thomas Zimmermann <tzimmermann@suse.de> > Cc: Daniel Vetter <daniel@ffwll.ch> > Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com> From a quick grep it looks like at least most drivers seem to get this right. Worth a shot to find the hold-outs. Two things: - Imo also best to move into the drm_mode_config_validate I suggested. - Please update the kerneldoc for drm_encoder.possible_crtcs to mention that this will WARN if you get it wrong (and maybe remove the line that most drivers screw this up). Check itself lgtm. -Daniel > --- > drivers/gpu/drm/drm_encoder.c | 18 ++++++++++++++++++ > 1 file changed, 18 insertions(+) > > diff --git a/drivers/gpu/drm/drm_encoder.c b/drivers/gpu/drm/drm_encoder.c > index bc2246f27e0d..f16b2a2518d7 100644 > --- a/drivers/gpu/drm/drm_encoder.c > +++ b/drivers/gpu/drm/drm_encoder.c > @@ -107,6 +107,23 @@ static void validate_possible_clones(struct drm_encoder *encoder) > encoder->possible_clones, encoder_mask); > } > > +static void validate_possible_crtcs(struct drm_encoder *encoder) > +{ > + struct drm_device *dev = encoder->dev; > + struct drm_crtc *crtc; > + u32 crtc_mask = 0; > + > + drm_for_each_crtc(crtc, dev) > + crtc_mask |= drm_crtc_mask(crtc); > + > + WARN((encoder->possible_crtcs & crtc_mask) == 0 || > + (encoder->possible_crtcs & ~crtc_mask) != 0, > + "Bogus possible_crtcs: " > + "[ENCODER:%d:%s] possible_crtcs=0x%x (full crtc mask=0x%x)\n", > + encoder->base.id, encoder->name, > + encoder->possible_crtcs, crtc_mask); > +} > + > int drm_encoder_register_all(struct drm_device *dev) > { > struct drm_encoder *encoder; > @@ -115,6 +132,7 @@ int drm_encoder_register_all(struct drm_device *dev) > fixup_possible_clones(dev); > > drm_for_each_encoder(encoder, dev) { > + validate_possible_crtcs(encoder); > validate_possible_clones(encoder); > > if (encoder->funcs->late_register) > -- > 2.24.1 >
On Fri, Feb 07, 2020 at 05:39:26PM +0100, Daniel Vetter wrote: > On Fri, Feb 07, 2020 at 03:59:50PM +0200, Ville Syrjala wrote: > > From: Ville Syrjälä <ville.syrjala@linux.intel.com> > > > > WARN if the encoder possible_crtcs is effectively empty or contains > > bits for non-existing crtcs. > > > > TODO: Or should we perhapst just filter out any bit for a > > non-exisiting crtc? > > > > Cc: Thomas Zimmermann <tzimmermann@suse.de> > > Cc: Daniel Vetter <daniel@ffwll.ch> > > Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com> > > >From a quick grep it looks like at least most drivers seem to get this > right. Worth a shot to find the hold-outs. > > Two things: > - Imo also best to move into the drm_mode_config_validate I suggested. > - Please update the kerneldoc for drm_encoder.possible_crtcs to mention > that this will WARN if you get it wrong (and maybe remove the line that > most drivers screw this up). ack > > Check itself lgtm. > -Daniel > > > --- > > drivers/gpu/drm/drm_encoder.c | 18 ++++++++++++++++++ > > 1 file changed, 18 insertions(+) > > > > diff --git a/drivers/gpu/drm/drm_encoder.c b/drivers/gpu/drm/drm_encoder.c > > index bc2246f27e0d..f16b2a2518d7 100644 > > --- a/drivers/gpu/drm/drm_encoder.c > > +++ b/drivers/gpu/drm/drm_encoder.c > > @@ -107,6 +107,23 @@ static void validate_possible_clones(struct drm_encoder *encoder) > > encoder->possible_clones, encoder_mask); > > } > > > > +static void validate_possible_crtcs(struct drm_encoder *encoder) > > +{ > > + struct drm_device *dev = encoder->dev; > > + struct drm_crtc *crtc; > > + u32 crtc_mask = 0; > > + > > + drm_for_each_crtc(crtc, dev) > > + crtc_mask |= drm_crtc_mask(crtc); > > + > > + WARN((encoder->possible_crtcs & crtc_mask) == 0 || > > + (encoder->possible_crtcs & ~crtc_mask) != 0, > > + "Bogus possible_crtcs: " > > + "[ENCODER:%d:%s] possible_crtcs=0x%x (full crtc mask=0x%x)\n", > > + encoder->base.id, encoder->name, > > + encoder->possible_crtcs, crtc_mask); > > +} > > + > > int drm_encoder_register_all(struct drm_device *dev) > > { > > struct drm_encoder *encoder; > > @@ -115,6 +132,7 @@ int drm_encoder_register_all(struct drm_device *dev) > > fixup_possible_clones(dev); > > > > drm_for_each_encoder(encoder, dev) { > > + validate_possible_crtcs(encoder); > > validate_possible_clones(encoder); > > > > if (encoder->funcs->late_register) > > -- > > 2.24.1 > > > > -- > Daniel Vetter > Software Engineer, Intel Corporation > http://blog.ffwll.ch
diff --git a/drivers/gpu/drm/drm_encoder.c b/drivers/gpu/drm/drm_encoder.c index bc2246f27e0d..f16b2a2518d7 100644 --- a/drivers/gpu/drm/drm_encoder.c +++ b/drivers/gpu/drm/drm_encoder.c @@ -107,6 +107,23 @@ static void validate_possible_clones(struct drm_encoder *encoder) encoder->possible_clones, encoder_mask); } +static void validate_possible_crtcs(struct drm_encoder *encoder) +{ + struct drm_device *dev = encoder->dev; + struct drm_crtc *crtc; + u32 crtc_mask = 0; + + drm_for_each_crtc(crtc, dev) + crtc_mask |= drm_crtc_mask(crtc); + + WARN((encoder->possible_crtcs & crtc_mask) == 0 || + (encoder->possible_crtcs & ~crtc_mask) != 0, + "Bogus possible_crtcs: " + "[ENCODER:%d:%s] possible_crtcs=0x%x (full crtc mask=0x%x)\n", + encoder->base.id, encoder->name, + encoder->possible_crtcs, crtc_mask); +} + int drm_encoder_register_all(struct drm_device *dev) { struct drm_encoder *encoder; @@ -115,6 +132,7 @@ int drm_encoder_register_all(struct drm_device *dev) fixup_possible_clones(dev); drm_for_each_encoder(encoder, dev) { + validate_possible_crtcs(encoder); validate_possible_clones(encoder); if (encoder->funcs->late_register)