From: Daniel Vetter <daniel@ffwll.ch>
To: "Ville Syrjälä" <ville.syrjala@linux.intel.com>
Cc: "Gustavo Padovan" <gustavo.padovan@collabora.co.uk>,
"Gustavo Padovan" <gustavo@padovan.org>,
"Daniel Stone" <daniels@collabora.com>,
"Riley Andrews" <riandrews@android.com>,
dri-devel@lists.freedesktop.org, linux-kernel@vger.kernel.org,
"Arve Hjønnevåg" <arve@android.com>,
"John Harrison" <John.C.Harrison@Intel.com>
Subject: Re: [RFC v2 5/8] drm/fence: add in-fences support
Date: Tue, 26 Apr 2016 22:05:05 +0200 [thread overview]
Message-ID: <20160426200505.GD2558@phenom.ffwll.local> (raw)
In-Reply-To: <20160426185506.GH4329@intel.com>
On Tue, Apr 26, 2016 at 09:55:06PM +0300, Ville Syrjälä wrote:
> On Tue, Apr 26, 2016 at 08:23:46PM +0200, Daniel Vetter wrote:
> > On Tue, Apr 26, 2016 at 08:40:45PM +0300, Ville Syrjälä wrote:
> > > On Tue, Apr 26, 2016 at 07:20:49PM +0200, Daniel Vetter wrote:
> > > > On Tue, Apr 26, 2016 at 07:26:21PM +0300, Ville Syrjälä wrote:
> > > > > On Tue, Apr 26, 2016 at 04:36:36PM +0200, Daniel Vetter wrote:
> > > > > > On Tue, Apr 26, 2016 at 11:14:22AM -0300, Gustavo Padovan wrote:
> > > > > > > 2016-04-26 Ville Syrjälä <ville.syrjala@linux.intel.com>:
> > > > > > >
> > > > > > > > On Mon, Apr 25, 2016 at 07:33:25PM -0300, Gustavo Padovan wrote:
> > > > > > > > > From: Gustavo Padovan <gustavo.padovan@collabora.co.uk>
> > > > > > > > >
> > > > > > > > > There is now a new property called FENCE_FD attached to every plane
> > > > > > > > > state that receives the sync_file fd from userspace via the atomic commit
> > > > > > > > > IOCTL.
> > > > > > > >
> > > > > > > > I still don't like this property abuse. Also with atomic, all passed
> > > > > > > > fences must be waited upon before anything is done, so attaching them
> > > > > > > > to planes seems like it might just give people the wrong idea.
> > > > > > >
> > > > > > > I'm actually fine with this as property, but another solutions is use
> > > > > > > an array of {plane, fence_fd} and extend drm_atomic_ioctl args just like
> > > > > > > we have done for out fences. However the FENCE_FD property is easier to
> > > > > > > handle in userspace than the array. Any other idea?
> > > > > >
> > > > > > Imo FENCE_FD is perfectly fine. But what's the concern around giving
> > > > > > people the wrong idea with attaching fences to planes? For nonblocking
> > > > > > commits we need to store them somewhere for the worker, drm_plane_state
> > > > > > seems like an as good place as any other.
> > > > >
> > > > > It gives the impression that each plane might flip as soon as its fence
> > > > > signals.
> > > >
> > > > That wouldn't be atomic. Not sure how someone could come up with that
> > > > idea.
> > >
> > > What else would it mean? It's attached to a specific plane, so why would
> > > it affect other planes?
> > >
> > > > I mean we could move FENCE_FD to the crtc (fence fds can be merged),
> > > > but that's just a needless difference to what hwc expects. I think
> > > > aligning with the only real-world users in this case here makes sense.
> > >
> > > Well it doesn't belong on the crtc either. I would just stick in the
> > > ioctl as a separate thing, then it's clear it's related to the whole
> > > operation rather than any kms object.
> >
> > We want it per-crtc I'd say, so that you could flip each crtc
> > individually.
>
> Then you could just issue multiple ioctls. For eg. those nasty 4k MST
> display (or just otherwise neatly synced displayes) you want to wait for
> all the fences upfront and the flip everything at once, otherwise you'll
> get nasty tears at the seams.
>
> > But really the reason for per-plane is hw composer from
> > Android. I don't see any point in designing an api that's needlessly
> > different from what the main user expects (even if it may be silly).
>
> What are they doing that can't stuff the fences into an array
> instead of props?
The hw composer interface is one in-fence per plane. That's really the
major reason why the kernel interface is built to match. And I really
don't think we should diverge just because we have a slight different
color preference ;-)
As long as you end up with a pile of fences somehow it'll work.
-Daniel
--
Daniel Vetter
Software Engineer, Intel Corporation
http://blog.ffwll.ch
next prev parent reply other threads:[~2016-04-26 20:05 UTC|newest]
Thread overview: 47+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-04-25 22:33 [RFC v2 0/8] drm: explicit fencing support Gustavo Padovan
2016-04-25 22:33 ` [RFC v2 1/8] dma-buf/fence: add fence_collection fences Gustavo Padovan
2016-04-26 14:41 ` Daniel Vetter
2016-04-26 15:02 ` Gustavo Padovan
2016-04-27 6:36 ` Daniel Vetter
2016-04-26 15:09 ` Chris Wilson
2016-04-28 14:47 ` Gustavo Padovan
2016-04-25 22:33 ` [RFC v2 2/8] Documentation: add fence-collection to kernel DocBook Gustavo Padovan
2016-04-25 22:33 ` [RFC v2 3/8] dma-buf/sync_file: add sync_file_fences_get() Gustavo Padovan
2016-04-25 22:33 ` [RFC v2 4/8] drm/fence: allow fence waiting to be interrupted by userspace Gustavo Padovan
2016-04-25 22:33 ` [RFC v2 5/8] drm/fence: add in-fences support Gustavo Padovan
2016-04-26 10:10 ` Ville Syrjälä
2016-04-26 14:14 ` Gustavo Padovan
2016-04-26 14:36 ` Daniel Vetter
2016-04-26 16:26 ` Ville Syrjälä
2016-04-26 17:20 ` Daniel Vetter
2016-04-26 17:40 ` Ville Syrjälä
2016-04-26 18:23 ` Daniel Vetter
2016-04-26 18:55 ` Ville Syrjälä
2016-04-26 20:05 ` Daniel Vetter [this message]
2016-04-26 20:48 ` Greg Hackmann
2016-04-27 6:39 ` Daniel Vetter
2016-04-28 21:28 ` Rob Clark
2016-04-29 7:48 ` Daniel Stone
2016-04-29 22:23 ` Rob Clark
2016-04-29 21:14 ` Greg Hackmann
2016-04-27 6:57 ` Daniel Stone
2016-04-28 14:36 ` Gustavo Padovan
2016-04-28 14:38 ` Daniel Vetter
2016-04-28 16:56 ` Ville Syrjälä
2016-04-28 17:43 ` Daniel Vetter
2016-04-28 17:51 ` Ville Syrjälä
2016-04-28 17:55 ` Gustavo Padovan
2016-04-28 18:02 ` Daniel Vetter
2016-04-28 18:17 ` Ville Syrjälä
2016-04-28 20:40 ` Daniel Vetter
2016-04-26 16:25 ` Ville Syrjälä
2016-04-25 22:33 ` [RFC v2 6/8] drm/fence: add fence to drm_pending_event Gustavo Padovan
2016-04-25 22:33 ` [RFC v2 7/8] drm/fence: add fence timeline to drm_crtc Gustavo Padovan
2016-04-26 10:12 ` Ville Syrjälä
2016-04-26 14:23 ` Gustavo Padovan
2016-04-26 16:34 ` Ville Syrjälä
2016-04-27 8:23 ` Daniel Stone
2016-04-25 22:33 ` [RFC v2 8/8] drm/fence: add out-fences support Gustavo Padovan
2016-04-26 14:53 ` Daniel Vetter
2016-04-28 15:23 ` Gustavo Padovan
[not found] ` <CAHbf0-HVNM=akFaE54U6B=51eegwumFmH=dUv+HHbnGOgGD=nw@mail.gmail.com>
2016-04-26 6:30 ` [RFC v2 0/8] drm: explicit fencing support 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=20160426200505.GD2558@phenom.ffwll.local \
--to=daniel@ffwll.ch \
--cc=John.C.Harrison@Intel.com \
--cc=arve@android.com \
--cc=daniels@collabora.com \
--cc=dri-devel@lists.freedesktop.org \
--cc=gustavo.padovan@collabora.co.uk \
--cc=gustavo@padovan.org \
--cc=linux-kernel@vger.kernel.org \
--cc=riandrews@android.com \
--cc=ville.syrjala@linux.intel.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;
as well as URLs for NNTP newsgroup(s).