From mboxrd@z Thu Jan 1 00:00:00 1970 From: Boris Brezillon Subject: Re: [PATCH v3 1/4] drm/uapi: add documentation for atomic flags Date: Fri, 12 Apr 2019 15:31:17 +0200 Message-ID: <20190412153117.1333f3d1@collabora.com> References: <20190412125827.5877-1-helen.koike@collabora.com> <20190412125827.5877-2-helen.koike@collabora.com> Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <20190412125827.5877-2-helen.koike@collabora.com> Sender: linux-kernel-owner@vger.kernel.org To: Helen Koike Cc: dri-devel@lists.freedesktop.org, David Airlie , dnicoara@chromium.org, daniels@collabora.com, alexandros.frantzis@collabora.com, daniel.vetter@ffwll.ch, linux-kernel@vger.kernel.org, tomasz Figa , tina.zhang@intel.com, Sean Paul , kernel@collabora.com, nicholas.kazlauskas@amd.com, =?UTF-8?B?U3TDqXBoYW5l?= Marchesin , Gustavo Padovan , Sean Paul , Maarten Lankhorst , Maxime Ripard , Daniel Vetter List-Id: dri-devel@lists.freedesktop.org Hello Helen, On Fri, 12 Apr 2019 09:58:24 -0300 Helen Koike wrote: > add a brief description of the flags used in an atomic commit > > Signed-off-by: Helen Koike > --- > > Changes in v3: None > Changes in v2: None > Changes in v1: None > > include/uapi/drm/drm_mode.h | 18 +++++++++++++++++- > 1 file changed, 17 insertions(+), 1 deletion(-) > > diff --git a/include/uapi/drm/drm_mode.h b/include/uapi/drm/drm_mode.h > index 83cd1636b9be..88ef2cf04d13 100644 > --- a/include/uapi/drm/drm_mode.h > +++ b/include/uapi/drm/drm_mode.h > @@ -729,7 +729,23 @@ struct drm_mode_destroy_dumb { > __u32 handle; > }; > > -/* page-flip flags are valid, plus: */ > +/* You might want to use a standard kernel/sphynx doc header so that it can be parsed by sphynx and included in the DRM doc. > + * drm atomic flags > + * > + * page-flip flags are valid, plus: > + * > + * DRM_MODE_ATOMIC_TEST_ONLY > + * Used with fences to check if the Sync File is a valid one. I think it can be used for any kind of tests, like when you want to know if a combination of property updates is supported. It's fine having the fence example, but I think you should be clear that it's just one use case. > + * > + * DRM_MODE_ATOMIC_NONBLOCK > + * Perform a normal atomic update but do not block the ioctl until the request > + * is finished, return the ioctl call immediately. > + * > + * DRM_MODE_ATOMIC_ALLOW_MODESET > + * Indicates whether a full modeset is acceptable or not. > + */ > + > +/* */ > #define DRM_MODE_ATOMIC_TEST_ONLY 0x0100 > #define DRM_MODE_ATOMIC_NONBLOCK 0x0200 > #define DRM_MODE_ATOMIC_ALLOW_MODESET 0x0400 Regards, Boris