[Intel-gfx] [PATCH CI 1/2] drm/i915/display/skl+: Drop frontbuffer rendering support

[Souza, Jose]

On Wed, 2021-09-15 at 16:48 +0300, Ville Syrjälä wrote:
> On Mon, Sep 13, 2021 at 10:54:14PM +0000, Souza, Jose wrote:
> > On Thu, 2021-09-09 at 23:28 +0300, Ville Syrjälä wrote:
> > > On Thu, Sep 09, 2021 at 08:23:20PM +0000, Souza, Jose wrote:
> > > > On Thu, 2021-09-09 at 23:20 +0300, Ville Syrjälä wrote:
> > > > > On Thu, Sep 09, 2021 at 12:49:16PM -0700, José Roberto de Souza wrote:
> > > > > > By now all the userspace applications should have migrated to atomic
> > > > > > or at least be calling DRM_IOCTL_MODE_DIRTYFB.
> > > > > > 
> > > > > > With that we can kill frontbuffer rendering support in i915 for
> > > > > > modern platforms.
> > > > > > 
> > > > > > So here converting legacy APIs into atomic commits so it can be
> > > > > > properly handled by driver i915.
> > > > > > 
> > > > > > Several IGT tests will fail with this changes, because some tests
> > > > > > were stressing those frontbuffer rendering scenarios that no userspace
> > > > > > should be using by now, fixes to IGT should be sent soon.
> > > > > > 
> > > > > 
> > > > > I just gave this a try here and it's unusable. glxgears went from
> > > > > 9000 to 120 fps (was expecting 60fps tbh, not sure why I get
> > > > > double), everything lags like mad, if I drag a window around
> > > > > glxgears/other windows stop updating entirely, etc. NAK
> > > > 
> > > > Can you share your setup? What GPU? Desktop environment? Mesa version? resolutions of sinks?
> > > > Will try it in my end.
> > > 
> > > Doesn't really matter as long as you don't have a compositor making a
> > > mess of things. This machine is a cfl running mate w/ compositor off,
> > > and some 1920x1200 display.
> > > 
> > 
> > Making drm_atomic_helper_dirtyfb() do a non-blocking atomic commit makes user experience pretty similar to the one with compositing enabled:
> > drm_atomic_commit() + compositor off: https://youtu.be/NBt6smXs99U
> > drm_atomic_nonblocking_commit() + compositor off: https://youtu.be/QiMhkeGX_L8
> > drm_atomic_nonblocking_commit() + compositor on: https://youtu.be/KdpJyJ5k6sQ
> > 
> > 
> > I do not completly agree with the comment in drm_atomic_helper_dirtyfb() about why it uses a blocking implementation.
> > With frontbuffer rendering the registers are programmed but the content could only show up for user a whole frame later.
> > 
> > Perhaps if this solutions is accetable we could have a non-blocking version of drm_atomic_helper_dirtyfb() so the drivers current using it don't have
> > their behavior changed.
> 
> Non-blocking update would make sense to me, whereas a blocking
> update makes no sense given how this is used by actual userspace.
> 
> But this still changes the whole approach to eg. FBC nukes, so it's
> going to require some actual thought and testing to make sure it
> actually works as intended. I think by just essentially page flipping
> to the same buffer we're going to depend on semi-undefined behaviour
> for FBC. So I'm a bit worried that this is going to introduce new bugs.
> I have verified on ancient platforms that FBC does in fact flip nuke
> when flipping to the same buffer (and now we even (ab)use that fact
> in the frontbuffer tracking code), but I've not done that for any new
> platforms.

I believe the current fbc tests in kms_frontbuffer_tracking already stress this.
For PSR2 hardware tracking in SKL and TGL I have validated(PSR2 is not covered by kms_frontbuffer_tracking due pipe CRC not work with it) this and
flipping to the same buffer causes the hardware to properly update the screen.

> 
> The other concern is that introducing additional atomic commits this
> might trip up the page flip->-EBUSY stuff. Whether that is going
> to cause any real grief to userspace is unknown at this time. I guess
> we'd have to come up with a test that actually hits that and make sure
> the ddx doesn't get confused and instead just gracefully falls back to
> the blit path.
>