[GIT PULL] omapdrm changes for 5.3

[Daniel Vetter]

On Tue, Jun 11, 2019 at 06:19:21PM +0200, Sebastian Reichel wrote:
> Hi Daniel,
> 
> On Tue, Jun 11, 2019 at 01:30:51PM +0200, Daniel Vetter wrote:
> > On Tue, Jun 11, 2019 at 11:05:40AM +0300, Tomi Valkeinen wrote:
> > > Hi Dave,
> > > 
> > > Please pull omapdrm changes for 5.3.
> > > 
> > >  Tomi
> > > 
> > > The following changes since commit 396f9acaffd87e44b8b02bc60f06dc949b832b7e:
> > > 
> > >   Merge branch 'drm-next-5.3' of git://people.freedesktop.org/~agd5f/linux into drm-next (2019-06-06 14:28:26 +1000)
> > > 
> > > are available in the Git repository at:
> > > 
> > >   git://git.kernel.org/pub/scm/linux/kernel/git/tomba/linux.git tags/omapdrm-5.3
> > > 
> > > for you to fetch changes up to 1bb418bffdd6f95a41e8e59be63933bc84fde13c:
> > > 
> > >   drm/omap: add support for manually updated displays (2019-06-10 17:04:15 +0300)
> > > 
> > > ----------------------------------------------------------------
> > > omapdrm changes for 5.3
> > > 
> > > - Add support for DSI command mode displays
> > 
> > Thanks, pulled.
> 
> Finally :)

Hm why? Pull is less than a day old, and I didn't see an older one that
Dave or me missed ...

> > > ----------------------------------------------------------------
> > > Emil Velikov (1):
> > >       drm/omap: remove open-coded drm_invalid_op()
> > > 
> > > Kefeng Wang (1):
> > >       drm/omap: Use dev_get_drvdata()
> > > 
> > > Sebastian Reichel (4):
> > >       drm/omap: use DRM_DEBUG_DRIVER instead of CORE
> > >       drm/omap: don't check dispc timings for DSI
> > >       drm/omap: add framedone interrupt support
> > >       drm/omap: add support for manually updated displays
> > 
> > This seems to reinvent drm_atomic_helper_dirty(), I've cc'ed you on a doc
> > patch. Also the fbdev dirty stuff seems to be missing here, you might want
> > to look at Noralf's generic fbdev stuff to take care of all of this for
> > you.
> 
> Thanks, I already noticed the patch series adding this, but did not
> yet find enough time to check the details. The omapdrm patches
> predate this work. The plan was to get the above work merged first,
> so that it's easier to catch regressions and unblock community work
> on Droid 4.
> 
> My plan was to first convert the DSI part from omapdrm to use
> mipi_dsi_device and drm_panel API and then look at drm_atomic_helper_dirtyfb
> afterwards.

Using drm_atomic_helper_dirty should essentially be a "remove a pile of
lines, replace with one" kind of patch. For an example, see the one Gerd
just posted for virtio. It's pretty low-hanging fruit :-)

I think the bigger fish to fry is switching over to the generic fbdev
emulation.
-Daniel
-- 
Daniel Vetter
Software Engineer, Intel Corporation
http://blog.ffwll.ch