lkml.org 
[lkml]   [2019]   [Dec]   [5]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
From
SubjectRe: [PATCH v2 01/28] drm: Introduce drm_bridge_init()
Date
Hi Laurent,

On Thursday, 5 December 2019 12:40:22 GMT Laurent Pinchart wrote:
> Hi Mihail,
>
> Thank you for the patch.

Thanks for the quick reviews :).

>
> On Wed, Dec 04, 2019 at 11:48:02AM +0000, Mihail Atanassov wrote:
> > A simple convenience function to initialize the struct drm_bridge. The
> > goal is to standardize initialization for any bridge registered with
> > drm_bridge_add() so that we can later add device links for consumers of
> > those bridges.
> >
> > v2:
> > - s/WARN_ON(!funcs)/WARN_ON(!funcs || !dev)/ as suggested by Daniel
> > - expand on some kerneldoc comments as suggested by Daniel
> > - update commit message as suggested by Daniel
> >
> > Signed-off-by: Mihail Atanassov <mihail.atanassov@arm.com>
> > ---
> > drivers/gpu/drm/drm_bridge.c | 34 +++++++++++++++++++++++++++++++++-
> > include/drm/drm_bridge.h | 15 ++++++++++++++-
> > 2 files changed, 47 insertions(+), 2 deletions(-)
> >
> > diff --git a/drivers/gpu/drm/drm_bridge.c b/drivers/gpu/drm/drm_bridge.c
> > index cba537c99e43..50e1c1b46e20 100644
> > --- a/drivers/gpu/drm/drm_bridge.c
> > +++ b/drivers/gpu/drm/drm_bridge.c
> > @@ -64,7 +64,10 @@ static DEFINE_MUTEX(bridge_lock);
> > static LIST_HEAD(bridge_list);
> >
> > /**
> > - * drm_bridge_add - add the given bridge to the global bridge list
> > + * drm_bridge_add - add the given bridge to the global bridge list.
>
> You add a final period here and in the documentation of struct
> drm_bridge, but the new function you're adding doesn't have one :-) I'd
> drop the period here and for drm_bridge to be consistent with the rest
> of the code.
>
> > + *
> > + * Drivers should call drm_bridge_init() prior adding it to the list.
>
> s/should/shall/
> s/prior adding it/prior to adding the bridge/
>
> > + * Drivers should call drm_bridge_remove() to clean up the bridge list.
>
> I'd replace this sentence with "Before deleting a bridge (usually when
> the driver is unbound from the device), drivers shall call
> drm_bridge_remove() to remove it from the global list."
>

> > *
> > * @bridge: bridge control structure
> > */
> > @@ -89,6 +92,35 @@ void drm_bridge_remove(struct drm_bridge *bridge)
> > }
> > EXPORT_SYMBOL(drm_bridge_remove);
> >
> > +/**
> > + * drm_bridge_init - initialise a drm_bridge structure
>
> initialise or initialize ? :-)

I have absolutely no clue :). Judging by the question I'm guessing the
correct answer for the kernel is US spelling.

>
> > + *
> > + * @bridge: bridge control structure
> > + * @funcs: control functions
> > + * @dev: device associated with this drm_bridge
>
> dev goes before funcs
>
> > + * @timings: timing specification for the bridge; optional (may be NULL)
> > + * @driver_private: pointer to the bridge driver internal context (may be NULL)
>
> I'm not too sure about the last two parameters. Having NULL, NULL in
> most callers is confusing, and having a void * as one of the parameters
> means that the compiler won't catch errors if the two parameters are
> reversed. I think this is quite error prone.
>
> There are very few drivers using driver_private (I count 6 of them, with
> 2 additional drivers that set driver_private but never use it). How
> about embedding the bridge structure in those 6 drivers and getting rid
> of the field altogether ? This could be part of a separate series, but
> in any case I'd drop driver_private from drm_bridge_init().

Ok, I'll do that first before refreshing this series.

>
> > + */
> > +void drm_bridge_init(struct drm_bridge *bridge, struct device *dev,
> > + const struct drm_bridge_funcs *funcs,
> > + const struct drm_bridge_timings *timings,
> > + void *driver_private)
> > +{
> > + WARN_ON(!funcs || !dev);
> > +
> > + bridge->dev = NULL;
>
> NULL ? Shouldn't this be dev ?

Hehe, Daniel got caught on that one, too :). This is the drm_device pointer
for the bound consumer, not the struct device that the bridge's lifetime
is tied to. I was planning to rename them with my (eventual) device_links
addition (some discussion around it here:
https://patchwork.freedesktop.org/patch/342472/?series=70039&rev=1).

I guess if I do the drm_device part of the rename first, this patch will
look less confusing, so I'll do that too.

>
> > + bridge->encoder = NULL;
> > + bridge->next = NULL;
> > +
> > +#ifdef CONFIG_OF
> > + bridge->of_node = dev->of_node;
> > +#endif
> > + bridge->timings = timings;
> > + bridge->funcs = funcs;
> > + bridge->driver_private = driver_private;
> > +}
> > +EXPORT_SYMBOL(drm_bridge_init);
> > +
> > /**
> > * drm_bridge_attach - attach the bridge to an encoder's chain
> > *
> > diff --git a/include/drm/drm_bridge.h b/include/drm/drm_bridge.h
> > index c0a2286a81e9..949e4f401a53 100644
> > --- a/include/drm/drm_bridge.h
> > +++ b/include/drm/drm_bridge.h
> > @@ -373,7 +373,16 @@ struct drm_bridge_timings {
> > };
> >
> > /**
> > - * struct drm_bridge - central DRM bridge control structure
> > + * struct drm_bridge - central DRM bridge control structure.
> > + *
> > + * Bridge drivers should call drm_bridge_init() to initialize a bridge
> > + * driver, and then register it with drm_bridge_add().
>
> s/bridge driver/bridge structure/ (or drm_bridge structure)
>
> > + *
> > + * Users of bridges link a bridge driver into their overall display output
> > + * pipeline by calling drm_bridge_attach().
> > + *
> > + * Modular drivers in OF systems can query the list of registered bridges
> > + * with of_drm_find_bridge().
> > */
> > struct drm_bridge {
> > /** @dev: DRM device this bridge belongs to */
> > @@ -402,6 +411,10 @@ struct drm_bridge {
> >
> > void drm_bridge_add(struct drm_bridge *bridge);
> > void drm_bridge_remove(struct drm_bridge *bridge);
> > +void drm_bridge_init(struct drm_bridge *bridge, struct device *dev,
> > + const struct drm_bridge_funcs *funcs,
> > + const struct drm_bridge_timings *timings,
> > + void *driver_private);
> > struct drm_bridge *of_drm_find_bridge(struct device_node *np);
> > int drm_bridge_attach(struct drm_encoder *encoder, struct drm_bridge *bridge,
> > struct drm_bridge *previous);
>
>


--
Mihail



\
 
 \ /
  Last update: 2019-12-05 15:25    [W:0.069 / U:4.080 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site