Should you ever need to read back from a tiled surface and for whatever
reason do not have a CPU bo to accelerate the operation, maybe we could
use the manual tiling instead (as it is useful elsewhere).
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Be kinder to smaller machines by lowering the threshold at which treat
an object as huge and worthy of avoiding duplication.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
As we strive to only keep one copy when working with very large objects,
so try operating inplace on a mapping for CPU operations with a large
GPU bo.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
This is lower latency than the double copy incurred for first moving the
bo to the CPU and then copying it back - but due to the less efficient
tiled memcpy, it has lower throughput. So x11perf -shmget500 suffers
(by about 30%) but real world applications improve by about 2x.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
To provide symmetry with the ability to write into an X-tiled mapping of
a bo, we add the memcpy_from_tiled to be able to read back from the same
bo.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
There are a few more circumstances where the temporary mapping is
beneficial, such as pixmaps that are only shadow copies or operations
that require reads on the destination.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Use the common function so that we correctly check for pinned GPU bo,
and adjust the hints afterwards.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Use the default mechanism for picking a fallback output and mode if
nothing is connected at X startup.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
After checking for user overrides (preferred output modes, rotation and
placement), copy the current CRTC configuration from the kernel into the
desired modes. This should enable X to start without introducing any extra
flicker.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Complete logic fail for finding the bounding box of the boxes to be
copied.
Reported-by: Clemens Eisserer <linuxhippy@gmail.com>
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=66168
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
If the caller requires an exactly constructed bo, abandon the attempt if
we cannot set the tiling as specified.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
As we reacquire master when initialisation the next gen of the server,
to keep the reference counting consistent we need to release our master
reference on CloseScreen.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
As we set the flags for the migration extents, but only operate upon the
boxes in the region, we need to be careful to be sure to migrate data
outside of the boxes overwritten.
Fixes regression from
commit 94cbe7bf7b [2.21.10]
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date: Thu Jun 20 19:40:44 2013 +0100
sna: Mark overwriting CopyArea as not needing the dst damage
If we start passing around regions rather than extents, we coud do
finergrained migration decision.
Reported-by: Clemens Eisserer <linuxhippy@gmail.com>
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=55244
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
So that the tarball matches the git checkout.
Reported-by: Maarten Lankhorst <maarten.lankhorst@canonical.com>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
The miWideDash may also call PolyPoint rather than emit spans, so make
sure that we correctly prepare sna_poly_point before calling mi*.
Reported-by: Tobias Brodel <toby.jamesjoy@gmail.com>
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=66104
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
The combination is just about valid and used along some odd paths, so
replace the assertion with a regular checks.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Fixes regression from
commit 53c113c3cc [2.21.10]
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date: Fri Jun 21 19:27:24 2013 +0100
sna: Allow PutImage to write inplace using manual tiling
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=66059
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Presume that the non-pipelined drawrect is sufficient to serialise
operations to the render cache.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Otherwise it appears that the hardware will readback from memory
bypassing its render cache after a change of modes. There is probably a
lot more subtly to it than this, but this appears to be a good first
approximation.
Reported-by: Clemens Eisserer <linuxhippy@gmail.com>
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=51422
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Chris Wilson
Roy.Li