Require for drm control-node

Mon Aug 14 08:02:28 UTC 2017

> Directly letting clients bypass the compositor using the design you're 
> proposing is not something we're supporting in upstream.
Yes, completely agree.

> i'm working on nxp i.mx8 platform
That this works with control nodes actually sounds like a rather nasty 
bug in some of the ARM drivers. What driver does that platform use?

We should either ping the maintainer of that driver or let Dave in his 
role as subsystem maintainer take a look.

Christian.

Am 14.08.2017 um 08:45 schrieb Daniel Vetter:
> Hi Jared,
>
> What you need is something like Wayland's subsurface support, so that 
> part of your apps screen can be passed to the compositor as NV12, and 
> the compositor then uses a plane for that.
>
> Directly letting clients bypass the compositor using the design you're 
> proposing is not something we're supporting in upstream.
> -Daniel
>
> On Mon, Aug 14, 2017 at 3:54 AM, Jared Hu <jared.hu at nxp.com 
> <mailto:jared.hu at nxp.com>> wrote:
>
>     Hi Christian,
>
>     i'm working on nxp i.mx8 platform, as far as i know, kmssink
>     cannot run with window system based on drm, i think kmssink is a
>     good plugin for video playback because of flex display configure
>     and multi screen support. We need this feature to support it.
>
>     best regards,
>     Jared
>
>     获取 Outlook for Android <https://aka.ms/ghei36>
>
>
>
>     发件人: Christian König
>     发送时间: 8月13日星期日 21:27
>     主题: Re: Require for drm control-node
>     收件人: Jared Hu, dri-devel at lists.freedesktop.org
>     <mailto:dri-devel at lists.freedesktop.org>
>     抄送: daniel.vetter at ffwll.ch <mailto:daniel.vetter at ffwll.ch>
>
>
>     Hi Jared,
>
>>     I met a problem that the kmssink will have permission issue when
>>     running with x11 and wayland.
>
>     This is intentional, at least with the current DRM KMS API design
>     only one application is allowed to control a DRM device.
>
>>     So I’m change the code to use control node “controlD64” for
>>     display and file node card0 for memory management. It work really
>>     well.
>
>     That sounds like a rather big bug to me, as far as I know control
>     nodes are completely deprecated and should never be used
>     concurrently with the DRM master on the primary node.
>
>     What hardware are trying this on?
>
>     Regards,
>     Christian.
>
>     Am 10.08.2017 um 10:27 schrieb Jared Hu:
>
>     Hi All,
>
>
>
>     I’m a gstreamer developer and is trying to implement a video
>     display plugin based on DRM/KMS which is named “kmssink”.
>
>
>
>     I met a problem that the kmssink will have permission issue when
>     running with x11 and wayland. Because these two window system will
>     hold DRM-Master in their life time. So I’m change the code to use
>     control node “controlD64” for display and file node card0 for
>     memory management. It work really well. Video can show on multi
>     display with wayland and x11. But I notice that in the Linux
>     kernel mainline, below commit remove the control node:
>
>
>
>     *commit 8a357d10043c75e980e7fcdb60d2b913491564af *
>
>     *Author: Daniel Vetter <**daniel.vetter at ffwll.ch
>     <mailto:daniel.vetter at ffwll.ch>**>*
>
>     *Date: Fri Oct 28 10:10:50 2016 +0200 *
>
>     **
>
>     *drm: Nerf DRM_CONTROL nodes *
>
>     **
>
>     *    Looking at the ioctl permission checks I noticed that it's
>     impossible *
>
>     *    to import gem buffers into a control nodes, and
>     fd2handle/handle2fd *
>
>     *    also don't work, so no joy with dma-bufs. *
>
>     **
>
>     *    The only way to do anything with a control node is by drawing
>     stuff *
>
>     *    into a dumb buffer and displaying that. I suspect control
>     nodes are an *
>
>     *    entirely unused thing, and a cursory check shows that there
>     does not *
>
>     *    seem to be any callers of drmOpenControl nor of the other
>     drmOpen *
>
>     *    functions using DRM_MODE_CONTROL. *
>
>     **
>
>     *    Since I don't like dead uabi, let's remove it. But since this
>     would be *
>
>     *    a really big change I think it's better to start out small by
>     simply *
>
>     *    not registering anything. We can garbage-collect the dead
>     code later *
>
>     *    on, once we're sure it's really not used anywhere. *
>
>     **
>
>     *    Acked-by: Dave Airlie <**airlied at gmail.com
>     <mailto:airlied at gmail.com>**>*
>
>     *Signed-off-by: Daniel Vetter <**daniel.vetter at intel.com
>     <mailto:daniel.vetter at intel.com>**>*
>
>     *Link:
>     **http://patchwork.freedesktop.org/patch/msgid/20161028081050.1042-1-daniel.vetter@ffwll.ch*
>     <http://patchwork.freedesktop.org/patch/msgid/20161028081050.1042-1-daniel.vetter@ffwll.ch>
>
>
>
>     This will have big impace on my case, and I cannot figure out how
>     to fix this issue. I think control node is useful, actually many
>     users are using this node for application development including
>     rockchip and amlogic. details please refer to:
>     http://gstreamer-devel.966125.n4.nabble.com/Can-kmssink-run-with-x11-or-wayland-which-is-based-on-drm-kms-also-td4683783.html#a4683819
>     <http://gstreamer-devel.966125.n4.nabble.com/Can-kmssink-run-with-x11-or-wayland-which-is-based-on-drm-kms-also-td4683783.html#a4683819>
>
>
>
>     I‘m sending this mail for *require restoring control node in file
>     system, or is there any other way to fix my permission issue*?
>
>
>     cid:image002.png at 01D13808.32F8E040
>
>
>
>     Thanks.
>
>
>
>     Best wishes
>
>     *Haihua Hu*
>
>     *MPU Mulitmedia Software*
>
>     NXP Semiconductors
>
>     No.288 ZhuYuan Road
>
>     Suzhou New District 215011, P.R.C
>
>     Phone: +86-512-68
>     cid:image003.png at 01D13808.32F8E040
>
>     0566738
>
>     jared.hu at nxp.com <mailto:jared.hu at nxp.com>
>
>
>
>
>
>
>
>     _______________________________________________ dri-devel mailing
>     list dri-devel at lists.freedesktop.org
>     <mailto:dri-devel at lists.freedesktop.org>
>     https://lists.freedesktop.org/mailman/listinfo/dri-devel
>     <https://lists.freedesktop.org/mailman/listinfo/dri-devel>
>
>
>
>
>
>
>
> -- 
> Daniel Vetter
> Software Engineer, Intel Corporation
> +41 (0) 79 365 57 48 - http://blog.ffwll.ch

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.freedesktop.org/archives/dri-devel/attachments/20170814/e833c4c3/attachment-0001.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: image001.png
Type: image/png
Size: 182 bytes
Desc: not available
URL: <https://lists.freedesktop.org/archives/dri-devel/attachments/20170814/e833c4c3/attachment-0002.png>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: image002.png
Type: image/png
Size: 22041 bytes
Desc: not available
URL: <https://lists.freedesktop.org/archives/dri-devel/attachments/20170814/e833c4c3/attachment-0003.png>