From: Daniel Vetter <daniel@ffwll.ch>
To: Rob Clark <robdclark@gmail.com>
Cc: dri-devel@lists.freedesktop.org
Subject: Re: [PATCH 10/17] drm: allow FB's in drm_mode_object_find
Date: Mon, 26 May 2014 10:39:32 +0200 [thread overview]
Message-ID: <20140526083932.GA14357@phenom.ffwll.local> (raw)
In-Reply-To: <1400956226-28053-11-git-send-email-robdclark@gmail.com>
On Sat, May 24, 2014 at 02:30:19PM -0400, Rob Clark wrote:
> We do actually want to permit FB's in atomic case, since FB will be
> looked up like any other object property value in a few code paths
> (like property value validation).
>
> So split out into an internal function without the WARN_ON() which
> we can use in those special cases.
>
> Signed-off-by: Rob Clark <robdclark@gmail.com>
See my other mail, but Nack. Imo you want a drm_mode_object_exists
instead, which just returns a bool. Gives you want you want and is safe.
-Daniel
> ---
> drivers/gpu/drm/drm_crtc.c | 33 ++++++++++++++++++++++-----------
> include/drm/drm_crtc.h | 6 ------
> 2 files changed, 22 insertions(+), 17 deletions(-)
>
> diff --git a/drivers/gpu/drm/drm_crtc.c b/drivers/gpu/drm/drm_crtc.c
> index 9b95601..48555724 100644
> --- a/drivers/gpu/drm/drm_crtc.c
> +++ b/drivers/gpu/drm/drm_crtc.c
> @@ -443,6 +443,21 @@ void drm_mode_object_put(struct drm_device *dev,
> mutex_unlock(&dev->mode_config.idr_mutex);
> }
>
> +static struct drm_mode_object *_object_find(struct drm_device *dev,
> + uint32_t id, uint32_t type)
> +{
> + struct drm_mode_object *obj = NULL;
> +
> + mutex_lock(&dev->mode_config.idr_mutex);
> + obj = idr_find(&dev->mode_config.crtc_idr, id);
> + if (!obj || (type != DRM_MODE_OBJECT_ANY && obj->type != type) ||
> + (obj->id != id))
> + obj = NULL;
> + mutex_unlock(&dev->mode_config.idr_mutex);
> +
> + return obj;
> +}
> +
> /**
> * drm_mode_object_find - look up a drm object with static lifetime
> * @dev: drm device
> @@ -455,23 +470,19 @@ void drm_mode_object_put(struct drm_device *dev,
> struct drm_mode_object *drm_mode_object_find(struct drm_device *dev,
> uint32_t id, uint32_t type)
> {
> - struct drm_mode_object *obj = NULL;
> -
> /* Framebuffers are reference counted and need their own lookup
> * function.*/
> WARN_ON(type == DRM_MODE_OBJECT_FB);
> -
> - mutex_lock(&dev->mode_config.idr_mutex);
> - obj = idr_find(&dev->mode_config.crtc_idr, id);
> - if (!obj || (type != DRM_MODE_OBJECT_ANY && obj->type != type) ||
> - (obj->id != id))
> - obj = NULL;
> - mutex_unlock(&dev->mode_config.idr_mutex);
> -
> - return obj;
> + return _object_find(dev, id, type);
> }
> EXPORT_SYMBOL(drm_mode_object_find);
>
> +static struct drm_mode_object *
> +drm_property_get_obj(struct drm_property *property, uint64_t value)
> +{
> + return _object_find(property->dev, value, property->values[0]);
> +}
> +
> /**
> * drm_framebuffer_init - initialize a framebuffer
> * @dev: DRM device
> diff --git a/include/drm/drm_crtc.h b/include/drm/drm_crtc.h
> index 31ed7c6..547b75a 100644
> --- a/include/drm/drm_crtc.h
> +++ b/include/drm/drm_crtc.h
> @@ -1033,12 +1033,6 @@ extern int drm_mode_crtc_set_gamma_size(struct drm_crtc *crtc,
> extern struct drm_mode_object *drm_mode_object_find(struct drm_device *dev,
> uint32_t id, uint32_t type);
>
> -static inline struct drm_mode_object *
> -drm_property_get_obj(struct drm_property *property, uint64_t value)
> -{
> - return drm_mode_object_find(property->dev, value, property->values[0]);
> -}
> -
> /* IOCTLs */
> extern int drm_mode_getresources(struct drm_device *dev,
> void *data, struct drm_file *file_priv);
> --
> 1.9.0
>
--
Daniel Vetter
Software Engineer, Intel Corporation
+41 (0) 79 365 57 48 - http://blog.ffwll.ch
next prev parent reply other threads:[~2014-05-26 8:39 UTC|newest]
Thread overview: 69+ messages / expand[flat|nested] mbox.gz Atom feed top
2014-05-24 18:30 [PATCH 00/17] prepare for atomic/nuclear modeset/pageflip Rob Clark
2014-05-24 18:30 ` [PATCH 01/17] drm: fix typo Rob Clark
2014-05-24 18:30 ` [PATCH 02/17] drm: add atomic fxns Rob Clark
2014-05-24 18:30 ` [PATCH 03/17] drm: convert crtc and mode_config to ww_mutex Rob Clark
2014-05-25 22:10 ` Daniel Vetter
2014-05-25 23:16 ` Rob Clark
2014-05-26 8:23 ` Daniel Vetter
2014-05-26 11:56 ` Rob Clark
2014-05-26 14:35 ` Daniel Vetter
2014-05-26 14:36 ` Daniel Vetter
2014-05-26 15:04 ` Rob Clark
2014-05-26 15:07 ` Daniel Vetter
2014-05-26 15:20 ` Rob Clark
2014-05-26 15:35 ` Daniel Vetter
2014-05-26 15:49 ` Rob Clark
2014-05-26 16:09 ` Daniel Vetter
2014-05-24 18:30 ` [PATCH 04/17] drm: add object property type Rob Clark
2014-05-26 8:29 ` Daniel Vetter
2014-05-26 8:33 ` Daniel Vetter
2014-05-26 11:06 ` Rob Clark
2014-05-24 18:30 ` [PATCH 05/17] drm: add signed-range " Rob Clark
2014-05-24 18:30 ` [PATCH 06/17] drm: helpers to find mode objects Rob Clark
2014-05-26 8:37 ` Daniel Vetter
2014-05-26 8:55 ` Daniel Vetter
2014-05-26 11:12 ` Rob Clark
2014-05-24 18:30 ` [PATCH 07/17] drm: split propvals out and blob property support Rob Clark
2014-05-24 18:30 ` [PATCH 08/17] drm: Allow drm_mode_object_find() to look up an object of any type Rob Clark
2014-05-24 18:30 ` [PATCH 09/17] drm: Refactor object property check code Rob Clark
2014-05-24 18:30 ` [PATCH 10/17] drm: allow FB's in drm_mode_object_find Rob Clark
2014-05-26 8:39 ` Daniel Vetter [this message]
2014-05-24 18:30 ` [PATCH 11/17] drm: convert plane to properties/state Rob Clark
2014-05-26 9:12 ` Daniel Vetter
2014-05-26 11:32 ` Rob Clark
2014-05-26 14:52 ` Daniel Vetter
2014-05-24 18:30 ` [PATCH 12/17] drm: convert crtc " Rob Clark
2014-05-26 9:31 ` Daniel Vetter
2014-05-26 11:35 ` Rob Clark
2014-05-26 14:56 ` Daniel Vetter
2014-05-26 15:15 ` Rob Clark
2014-05-26 15:23 ` Ville Syrjälä
2014-05-26 15:37 ` Daniel Vetter
2014-05-26 15:42 ` Rob Clark
2014-05-26 15:46 ` Ville Syrjälä
2014-05-26 16:12 ` Daniel Vetter
2014-05-24 18:30 ` [PATCH 13/17] drm: push locking down into restore_fbdev_mode Rob Clark
2014-05-26 9:34 ` Daniel Vetter
2014-05-24 18:30 ` [PATCH 14/17] drm/msm: add atomic support Rob Clark
2014-05-26 17:54 ` Daniel Vetter
2014-05-27 15:58 ` Rob Clark
2014-05-27 17:50 ` Daniel Vetter
2014-05-27 18:48 ` Rob Clark
2014-05-27 19:26 ` Daniel Vetter
2014-05-27 20:06 ` Rob Clark
2014-05-27 22:09 ` Daniel Vetter
2014-05-27 23:32 ` Rob Clark
2014-05-28 13:21 ` Daniel Vetter
2014-05-28 14:14 ` Ville Syrjälä
2014-05-28 14:50 ` Daniel Vetter
2014-05-28 15:19 ` Rob Clark
2014-05-27 23:47 ` Rob Clark
2014-05-28 13:32 ` Daniel Vetter
2014-05-24 18:30 ` [PATCH 15/17] drm: spiff out FB refcnting traces Rob Clark
2014-05-24 18:30 ` [PATCH 16/17] drm: more conservative locking Rob Clark
2014-05-24 18:30 ` [PATCH 17/17] drm: Fix up the atomic legacy paths so they work Rob Clark
2014-05-26 10:40 ` [PATCH 00/17] prepare for atomic/nuclear modeset/pageflip Daniel Vetter
2014-05-26 12:48 ` Rob Clark
2014-05-26 15:24 ` Daniel Vetter
2014-05-26 16:12 ` Rob Clark
2014-05-26 17:36 ` Daniel Vetter
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20140526083932.GA14357@phenom.ffwll.local \
--to=daniel@ffwll.ch \
--cc=dri-devel@lists.freedesktop.org \
--cc=robdclark@gmail.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox