As there may be some latency between the wakeup and the queue, do a
final check if there is a pending vblank event before we decide whether
to chain up or perform an immediate flip.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
As we may recurse into the vblank handler whilst waiting for the
TearFree shadow, we need to be careful that we do not queue the frame
from the vblank handler that we are in the process of constructing.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
As the Window may be resized between the client sending the swap request
and the xchg being processed, we need to validate the swap against the
final target not the buffers supplied originally by the client.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
fbcon doesn't adjust the backlight when it takes over. Therefore if X
performs a VT switch whilst its outputs are off, fbcon wakes up with no
backlight and users are unhappy. Make the assumption that whoever takes
over the VT will set the outputs as it desires, and that the failsafe
value is to then turn the backlight to full.
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=67025
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
The switch was moved after we checked whether pageflips are supported,
but the log message was left in the old location.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
In order to free up memory and improve our bookkeeping when the X server
is switched away from the current VT, disable the CRTC.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
The tearfree code forgot to update the shadow bo, so after the first
flip it would be rendering to the scanout anyway and flip to itself
periodically.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
This was disabled in
commit 9f4f855ba3
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date: Mon May 26 07:06:18 2014 +0100
sna: Implicit release of upload buffers considered bad
as retiring the buffers during the command setup could free one of the
earlier bo used in the command. But discarding the snooped bo could
still be advantageous. So restore the automatic discard of upload
proxies, but make sure we only do between operations.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Matti Hämäläinen <ccr@tnsp.org>
With a shadow, we may never attach a GPU bo to the frontbuffer, so be
careful when checking the scanout not to dereference that NULL pointer.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Fallout from commit 8369166349
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date: Wed Jun 4 08:29:51 2014 +0100
sna/dri2: Enable immediate buffer exchanges
We set the flush hint too early, as we may need to swap the GPU bo for a
more appropriate buffer for DRI.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Changes to the backlights are notified through uevents. Hooking up a
udev monitor to listen out for external changes to the backlight (e.g.
through ACPI function keys, or by the user writing to
/sys/class/backlight directly) is easier than enabling polling on the
backlight sysfs file using X's select() mechanism.
Since we listen to backlight changes, we have to be careful not to
confuse the side-effects of disabling connectors (which may cause either
ourselves or the kernel to turn off the backlight) with the user value.
Many thanks to Alexander Mezin for the suggestion to use udev for
tracking the notifications for external changes to the backlight.
Reported-by: Alexander Mezin <mezin.alexander@gmail.com>
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=79699
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
As we have reorder the application of damage on a TearFree frontbuffer
to before we attach the CRTC, we do not need to then schedule the update
afterwards (until it gets drawn by the client).
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
It is far easily to pass the PixmapPtr into the function and have it
pluck out the width and height than do so in all callers.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
When we discard a GPU bo, we need to make sure that the remaining
content is marked as only accessible via the CPU shadow pointer.
Regression from commit 65301412ec
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date: Sat May 17 20:59:38 2014 +0100
sna: Discard active GPU buffers before uploading into them
Reported-by: Jan Alexander Steffens <jan.steffens@gmail.com>
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=79517
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
The primary benefit of this is avoid the extra blit when using a
compositor and instead propagate the compositor flip on the frontbuffer
to the scanout, or equivalently allows a fullscreen game to flip onto
the scanout without intervention by TearFree.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
XORG_DRIVER_CHECK_EXT only adds the package name to the list of
REQUIRED_MODULES (which is later passed to pkg-config). For our optional
dependencies, we must do a later call to pkg-config to verify the
package is available and then remember to include the includes with the
CFLAGS.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
We only depend upon the original DRI1 infrastructure when building i810
with DRI support. So unless that is true, do not throw an error if the
legacy protocol is absent.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
78dc0c0474 added REQUIRED_MODULES to the
driver link line for... some reason. That pulled in the libs from the
XF86DRI check, which near as I can tell has always been wrong, all of
the other extension checks just look for the protocol module and
xextproto doesn't define dri1 protocol in any case.
Signed-off-by: Adam Jackson <ajax@redhat.com>
In commit 961139f587
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date: Fri May 30 09:45:15 2014 +0100
sna: Use manual detiling for downloads
the code for deciding when to use the GPU was refactored into a new
function that also performed the transfer, but failed to notice the
early return.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
The other half of the commit that accidentally got included with
commit bb49222a51
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date: Mon Jun 2 08:25:52 2014 +0100
sna: Add DBG hints for using inplace CPU mmappings
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Make sure we can compile against the pre-release tarballs at the expense
of an innocuous compile warning if compiled against xserver.git
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
This refactors the drm interrupt handling logic quite a bit, both to
allow for either DRI2 or Present handlers, but also to eliminate
passing pointers through the kernel. Passing pointers left the kernel
holding the only reference to some internal X server data structures.
After a server reset, the X server would end up using stale pointers
stored in those structures. Using simple integers makes it possible to
empty the queue of pending interrupt data and then ignore the stale
kernel data.
Signed-off-by: Keith Packard <keithp@keithp.com>
Reviewed-by: Eric Anholt <eric@anholt.net>
The kernel sometimes reports bogus MSC values, especially when
suspending and resuming the machine. Deal with this by tracking an
offset to ensure that the MSC seen by applications increases
monotonically, and at a reasonable pace.
Also, provide a full 64 bits of MSC value by noticing wrapping and
tracking the high 32-bits of MSC separately.
Signed-off-by: Keith Packard <keithp@keithp.com>
Reviewed-by: Eric Anholt <eric@anholt.net>
Conflicts:
src/uxa/intel_dri.c
Only force the even-tile-row alignment if we have an old GPU with an old
kernel that doesn't perform conservative alignment for us (as required).
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
This is a simple little test to create a pixmap from a local bo, copy it
to a normal pixmap, then read it back by importing it into anther local
bo. It tests the fundamental mechanisms of opening a DRI3 render device,
importing into pixmaps, exporting into /buffers and a read-barrier.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
If a flip fails, attempt to restore the previous working configuration
(using a modeset) then disable further flipping.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
OML_sync_control mandates that MSC must be monotonic, so if the kernel
reports that they go backwards, lie.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Chris Wilson
Adam Jackson
Keith Packard