| Age | Commit message (Collapse) | Author |
|---|
|
|
|
|
|
|
|
|
|
The struct is still there though until all winsys layers are updated
|
|
|
|
|
|
|
|
updates fail
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Not convenient and almost not used at all. Better replacements in p_debug.h
|
|
|
|
|
|
This works since we always use XImages for color buffers. It wouldn't always
work for Windows/Pixmaps.
|
|
|
|
The cell driver still uses an internal CELL_FLUSH_WAIT flag, in the long run
proper fencing should be implemented for it.
|
|
Gives about a 3% performance increase in gears on x86-64 (non-tiled)
and about 10% performance increase in gears on Cell (tiled). I
actually expected more of a boost. :(
|
|
|
|
A recent commit removed a bug which essentially caused a sync after each
use of a user buffer. In fixing it, the scratch area can now become
fragmented under some circumstances leading to nouveau_bo_tmp() waiting
forever for a large enough block of free space.
This fixes ppracer, gl-117 and probably a heap of other apps.
|
|
|
|
It's possible for the unsignalled list head to change during fence_flush.
Fixes valgrind complaint exposed by a previous commit.
|
|
Rather than spinning waiting for the GPU to finish with the buffer,
add a callback on the buffer's fence object instead. Gives a sizable
performance increase in OA.
|
|
Fixes some cases where we end up with a list of many unused chunks that are
too small to be useful.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
