| Commit message (Collapse) | Author | Age | Files | Lines |
|
|
|
|
|
|
|
| |
The areamap contains precomputed data on different aliasing types.
It is necessary for good performance.
Signed-off-by: Lauri Kasanen <[email protected]>
Signed-off-by: Brian Paul <[email protected]>
|
|
|
|
|
| |
Signed-off-by: Lauri Kasanen <[email protected]>
Signed-off-by: Brian Paul <[email protected]>
|
|
|
|
|
| |
Signed-off-by: Lauri Kasanen <[email protected]>
Signed-off-by: Brian Paul <[email protected]>
|
|
|
|
|
| |
Signed-off-by: Lauri Kasanen <[email protected]>
Signed-off-by: Brian Paul <[email protected]>
|
|
|
|
|
| |
Signed-off-by: Lauri Kasanen <[email protected]>
Signed-off-by: Brian Paul <[email protected]>
|
|
|
|
|
|
| |
I also needed to make some changes in u_vbuf_mgr in order to override
the caps from the driver and enable the fallback even though the driver
claims the format is supported.
|
| |
|
| |
|
|\
| |
| |
| |
| |
| | |
Conflicts:
src/mesa/state_tracker/st_atom_pixeltransfer.c
src/mesa/state_tracker/st_program.c
|
| | |
|
| | |
|
| |
| |
| |
| | |
Reported by Gustaw Smolarczyk.
|
| |
| |
| |
| | |
Some of those have been in drivers already.
|
| |
| |
| |
| | |
NOTE: This is a candidate for the 7.10 and 7.11 branches.
|
| |
| |
| |
| | |
Fixes: https://bugs.freedesktop.org/show_bug.cgi?id=39276
|
| |
| |
| |
| |
| | |
The position of the quad vertex is calculated in calc_position,
so we don't need the output here any more.
|
| | |
|
| |
| |
| |
| |
| |
| |
| | |
Not sure how I computed these, but they were wrong (which explains why
bumping the polynomial order before never improved precision).
This allows to pass the EXP test cases of PSPrecision/VSPrecision DCTs.
|
| |
| |
| |
| |
| |
| |
| | |
Add an iteration step, which makes rqsqrt precision go from 12bits to
24, and fixes RSQ/NRM test case of PSPrecision/VSPrevision DCTs.
There are no uses of this function outside shader translation.
|
| | |
|
| |
| |
| |
| |
| | |
Never used so far -- we only used the base 2 variants -- which is why
it went unnoticed so far.
|
| | |
|
| |
| |
| |
| |
| |
| | |
This reverts commit b56daf71d2f63d044d4c53ab49c6f87e02991a28.
The bug is actually in softpipe's blend and writemask interaction.
|
|\ \ |
|
| | | |
|
| | | |
|
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
u_vbuf_upload_buffers modifies the buffer offsets. If they are not
restored, and any of the vertex formats is not supported natively, the
next u_vbuf_mgr_draw_begin call will translate the vertex buffers with
incorrect buffer offsets.
|
| | |
| | |
| | |
| | | |
See also comments in the code.
|
| | |
| | |
| | |
| | | |
Fixes xvmc-softpipe MC entrypoint, amongst others.
|
| | |
| | |
| | |
| | | |
Fixes VDPAU CSC-only mode.
|
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
LLVM 3.0svn introduced a new type system. It defines a new way to create
named structs and removes the (now not needed) LLVMInvalidateStructLayout
function. See revision 134829 of LLVM.
Signed-off-by: Tobias Droste <[email protected]>
Signed-off-by: Brian Paul <[email protected]>
|
| | |
| | |
| | |
| | |
| | | |
llvm-3.0svn r135219 renamed createAsmInfo to createMCAsmInfo in
include/llvm/Target/TargetRegistry.h.
|
| | |
| | |
| | |
| | | |
Move that also inside the state tracker where needed.
|
| | |
| | |
| | |
| | | |
Otherwise xine won't scale correctly.
|
| | |
| | |
| | |
| | | |
Using pipe_mpeg12_picture_desc was unintentional here.
|
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
The forward references to video enum types in p_context.h causes
a massive number of compiler warnings (ISO C forbids forward references
to ‘enum’ types).
By putting the new video enums in a separate header that can be included
by p_context.h and p_screen.h we can avoid this.
Acked-by Christian König <[email protected]>
|
| | |
| | |
| | |
| | | |
The array initializer must be a constant expression in MSVC.
|
| | |
| | |
| | |
| | | |
MSVC does not support designated initializers.
|
| | |
| | |
| | |
| | | |
The inline keyword is not available in MSVC C.
|
| | |
| | |
| | |
| | |
| | | |
Signed-off-by: Emil Velikov <[email protected]>
Signed-off-by: Brian Paul <[email protected]>
|
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
There seems to be a bug in r600g when uploading more than one layer of a
3D resource at once with a hardware blit.
So just do them one at a time to workaround this.
|
|\ \ \ |
|
| | | |
| | | |
| | | |
| | | | |
It's about 20% slower, but should at least work with every hardware.
|
| | | |
| | | |
| | | |
| | | | |
So we always know to which idct object a buffer belongs
|
| | | | |
|
| | | | |
|
| | | | |
|
| |\ \ \
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
Conflicts:
src/gallium/drivers/r600/r600_pipe.c
src/gallium/drivers/r600/r600_state_inlines.h
|
| | | | | |
|
| | | | | |
|