profile/ivi/mesa.git
16 years ago [i965] multiple rendering target fix
Zou Nan hai [Fri, 21 Mar 2008 10:16:11 +0000 (18:16 +0800)]
 [i965] multiple rendering target fix

16 years ago[i915] GL_DEPTH_TEXTURE_MODE fix
Zou Nan hai [Thu, 20 Mar 2008 09:47:02 +0000 (17:47 +0800)]
[i915] GL_DEPTH_TEXTURE_MODE fix

16 years agomesa: The span array is only populated in single-pixel
Xiang, Haihao [Thu, 20 Mar 2008 03:22:46 +0000 (11:22 +0800)]
mesa: The span array is only populated in single-pixel
point drawing function.

16 years ago[965] Initialize region surface key structure padding.
Eric Anholt [Thu, 20 Mar 2008 00:53:21 +0000 (17:53 -0700)]
[965] Initialize region surface key structure padding.

Fixes valgrind warnings, and potential performance loss from cache misses.

16 years ago[intel] Fix an uninitialized variable access in PRESUMED_OFFSET clearing.
Eric Anholt [Wed, 19 Mar 2008 23:14:48 +0000 (16:14 -0700)]
[intel] Fix an uninitialized variable access in PRESUMED_OFFSET clearing.

It was harmless, as the only time we need to clear PRESUMED_OFFSET, the
variable had been initialized already.

16 years agofix IEEE_ONE definition for ICC compiler (bug 15134)
Brian [Wed, 19 Mar 2008 23:24:20 +0000 (17:24 -0600)]
fix IEEE_ONE definition for ICC compiler (bug 15134)

16 years agoRadeon 9500 (0x4144) only has one pipe
Alex Deucher [Wed, 19 Mar 2008 22:10:47 +0000 (18:10 -0400)]
Radeon 9500 (0x4144) only has one pipe

confirmed by Reid Linnemann <lreid@cs.okstate.edu>

16 years agox86_init_func_size() returns int to indicate success/fail (bug 15119)
Brian [Wed, 19 Mar 2008 15:52:08 +0000 (09:52 -0600)]
x86_init_func_size() returns int to indicate success/fail (bug 15119)

16 years agos/x86_init_func/x86_init_func_size (fixes bug 15119)
Brian [Wed, 19 Mar 2008 14:38:28 +0000 (08:38 -0600)]
s/x86_init_func/x86_init_func_size (fixes bug 15119)

16 years ago [i915] arb point sprite only support in i965
Zou Nan hai [Wed, 19 Mar 2008 08:32:29 +0000 (16:32 +0800)]
 [i915] arb point sprite only support in i965

16 years ago[i915] fix fragment.position
Zou Nan hai [Wed, 19 Mar 2008 08:29:47 +0000 (16:29 +0800)]
[i915] fix fragment.position

16 years ago[i915] Bug #13634: Fix bugs in 945 cube mipmap layout.
Eric Anholt [Wed, 19 Mar 2008 02:45:30 +0000 (19:45 -0700)]
[i915] Bug #13634: Fix bugs in 945 cube mipmap layout.

The most egregious, and the one the bug report and failure in the cubemap
demo were about was introduced with intel_mipmap_pitch_align(), where a
"* 2" for the pitch calculation was lost.  The base size < 32 case also
failed to align, which may have caused problems with render to texture.
Another bug would have broken 2x2/1x1 base mipmap levels by placing the
data where the hardware wouldn't look for it.

Other bugs remain with the layout of the small mipmap faces (hardware looks
for them in X,Y,Z,-X,-Y,-Z order along the bottom row, but we lay them out
X,-X,Y,-Y,Z,-Z).

16 years ago[i915] Add comments about how cube texture layout works.
Eric Anholt [Wed, 19 Mar 2008 00:49:31 +0000 (17:49 -0700)]
[i915] Add comments about how cube texture layout works.

16 years ago[i915] Move miptree layout code into separate functions per target.
Eric Anholt [Tue, 18 Mar 2008 23:06:41 +0000 (16:06 -0700)]
[i915] Move miptree layout code into separate functions per target.

Also clean up some other miscellaneous formatting nits while I'm at it.

16 years ago[intel] Clarify miptree layout by using byte offsets to images.
Eric Anholt [Tue, 18 Mar 2008 23:20:44 +0000 (16:20 -0700)]
[intel] Clarify miptree layout by using byte offsets to images.

16 years ago[945] Remove conditional in 945 3D mipmap layout checking for cube layout.
Eric Anholt [Tue, 18 Mar 2008 22:29:11 +0000 (15:29 -0700)]
[945] Remove conditional in 945 3D mipmap layout checking for cube layout.

16 years agoset outputs_safe to 0 as it's possible for the code generation
Alan Hourihane [Wed, 19 Mar 2008 01:03:48 +0000 (01:03 +0000)]
set outputs_safe to 0 as it's possible for the code generation
to slip over the allocated memory for the vb.

pull in sse fixes from gallium-0.1

16 years agoRevert "[i965] make stipple pattern continue across GL_LINE_LOOP and GL_LINE_STRIP"
Zou Nan hai [Tue, 18 Mar 2008 09:37:25 +0000 (17:37 +0800)]
Revert "[i965] make stipple pattern continue across GL_LINE_LOOP and GL_LINE_STRIP"
There is no information in GS to determinate when to reset line stipple count, still fallback to software
This reverts commit 5a0314b431ab147c6156c3011f4cb54161ba4b25.

16 years agoglx: Add isDirect back to __GLXcontextRec. It is needed
Xiang, Haihao [Tue, 18 Mar 2008 07:02:57 +0000 (15:02 +0800)]
glx: Add isDirect back to __GLXcontextRec. It is needed
to check whether oldGC is used for direct rendering in
function MakeContextCurrent. However it is possible
oldGC->driContext is already freed. fix bug #14926.

16 years ago[i965] make stipple pattern continue across GL_LINE_LOOP and GL_LINE_STRIP
Zou Nan hai [Tue, 18 Mar 2008 02:27:20 +0000 (10:27 +0800)]
[i965] make stipple pattern continue across GL_LINE_LOOP and GL_LINE_STRIP

16 years agor300: add new rs690 pci id
Dave Airlie [Tue, 18 Mar 2008 00:29:16 +0000 (10:29 +1000)]
r300: add new rs690 pci id

16 years ago[965] Fix fp temp reg release code to not usually release all temps.
Andrzej Trznadel [Mon, 17 Mar 2008 22:54:24 +0000 (15:54 -0700)]
[965] Fix fp temp reg release code to not usually release all temps.

Also, use wrapped ffs() instead of native.

16 years agoFix compat implementation of ffs() to return 1-based bit numbers.
Andrzej Trznadel [Mon, 17 Mar 2008 22:52:08 +0000 (15:52 -0700)]
Fix compat implementation of ffs() to return 1-based bit numbers.

16 years agoonly set InputsRead bit if input is really used
Markus Amsler [Mon, 17 Mar 2008 14:35:27 +0000 (08:35 -0600)]
only set InputsRead bit if input is really used

16 years agomesa: Add vertex.attrib 0-15 to arb_input_attrib_string.
Markus Amsler [Mon, 17 Mar 2008 11:11:11 +0000 (12:11 +0100)]
mesa: Add vertex.attrib 0-15 to arb_input_attrib_string.

16 years agor300: Simplify r300VAPInputRoute1.
Markus Amsler [Mon, 17 Mar 2008 11:03:54 +0000 (12:03 +0100)]
r300: Simplify r300VAPInputRoute1.

16 years agor300: Simplify r300VAPInputRoute0, check for valid input.
Markus Amsler [Mon, 17 Mar 2008 11:02:56 +0000 (12:02 +0100)]
r300: Simplify r300VAPInputRoute0, check for valid input.

16 years ago[i965] round pointsize to nearest int according to spec
Zou Nan hai [Mon, 17 Mar 2008 09:52:26 +0000 (17:52 +0800)]
[i965] round pointsize to nearest int according to spec

16 years agointel: fix the error in commit 7ed1fd5d8438e55fe24091844cdfccb0881306bc
Xiang, Haihao [Mon, 17 Mar 2008 09:02:42 +0000 (17:02 +0800)]
intel: fix the error in commit 7ed1fd5d8438e55fe24091844cdfccb0881306bc

16 years agointel: It is needed to allocating texture memory to accommodate
Xiang, Haihao [Mon, 17 Mar 2008 08:52:07 +0000 (16:52 +0800)]
intel: It is needed to allocating texture memory to accommodate
a texture when calling TexImage with pixels set to NULL pointer.

16 years agointel: Remove an assertion from intel_miptree_create. TexImage
Xiang, Haihao [Mon, 17 Mar 2008 08:50:20 +0000 (16:50 +0800)]
intel: Remove an assertion from intel_miptree_create. TexImage
call with zero width/height/depth matches GL spec.

16 years agomesa: avoid to unlock an unlocked array in _mesa_PopClientAttrib
Xiang, Haihao [Mon, 17 Mar 2008 08:48:36 +0000 (16:48 +0800)]
mesa: avoid to unlock an unlocked array in _mesa_PopClientAttrib

16 years agomesa: Follow GL spec to draw DEPTH_COMPONENT pixels when
Xiang, Haihao [Mon, 17 Mar 2008 08:47:42 +0000 (16:47 +0800)]
mesa: Follow GL spec to draw DEPTH_COMPONENT pixels when
there's no depth buffer. Fix bug #11580

16 years ago [i965] fix wpos height 1 pixel higher
Zou Nan hai [Mon, 17 Mar 2008 08:39:10 +0000 (16:39 +0800)]
 [i965] fix wpos height 1 pixel higher

16 years agoTest createNewScreen for NULL, not createNewScreenName in driCreateScreen().
Kristian Høgsberg [Sat, 15 Mar 2008 18:06:42 +0000 (14:06 -0400)]
Test createNewScreen for NULL, not createNewScreenName in driCreateScreen().

16 years agoinit tmpNode to zeros
Brian [Fri, 14 Mar 2008 20:15:55 +0000 (14:15 -0600)]
init tmpNode to zeros

16 years agomesa: fix emit_clamp() so that we don't use an output register as temporary
Brian [Fri, 14 Mar 2008 19:59:32 +0000 (13:59 -0600)]
mesa: fix emit_clamp() so that we don't use an output register as temporary

IR_CLAMP is decomposed into OPCODE_MIN+OPCODE_MAX.  Allocate a temporary
register for the intermediate value so we don't inadvertantly use an output
register (which are write-only on some GPUs).

16 years agoglx: fix rotation regression. bug #14963
Xiang, Haihao [Fri, 14 Mar 2008 07:27:15 +0000 (15:27 +0800)]
glx: fix rotation regression. bug #14963

16 years agointel: fix abort issue with shadowtex demo when use
Xiang, Haihao [Fri, 14 Mar 2008 03:39:08 +0000 (11:39 +0800)]
intel: fix abort issue with shadowtex demo when use
DEPTH_STENCIL texture. (bug#14952).

16 years agofix state.lightprod ambient/specular w value (bug #14983)
Roland Scheidegger [Thu, 13 Mar 2008 11:43:04 +0000 (12:43 +0100)]
fix state.lightprod ambient/specular w value (bug #14983)

16 years ago [i965] multiple rendering target support
Zou Nan hai [Thu, 13 Mar 2008 06:46:38 +0000 (14:46 +0800)]
 [i965] multiple rendering target support

16 years agolibGL: Pull the drawable hash back out in the generic code.
Kristian Høgsberg [Wed, 12 Mar 2008 06:18:12 +0000 (02:18 -0400)]
libGL: Pull the drawable hash back out in the generic code.

This will be shared between dri and dri2 code.

16 years agoAdd another missing check for uninitialized DRI.
Kristian Høgsberg [Wed, 12 Mar 2008 06:01:40 +0000 (02:01 -0400)]
Add another missing check for uninitialized DRI.

Spotted by Chris Taylor.

16 years agoFix include path for rain demo so glut.h is found
Dan Nicholson [Tue, 11 Mar 2008 13:24:56 +0000 (06:24 -0700)]
Fix include path for rain demo so glut.h is found

16 years agoautoconf: Enable xdemos by default when we build libGL
Dan Nicholson [Mon, 10 Mar 2008 21:05:46 +0000 (14:05 -0700)]
autoconf: Enable xdemos by default when we build libGL

Since the xdemos only link to libGL now, we can enable them for all but
the osmesa driver target.

16 years agoDarwin: Fixed small error in darwin config files
Jeremy Huddleston [Mon, 10 Mar 2008 23:02:55 +0000 (16:02 -0700)]
Darwin: Fixed small error in darwin config files
(cherry picked from commit a21c61ee8bc86a8843024cbf8e9daf4b39a7571a)

16 years agoMove make install logic for libGL back into src/mesa/Makefile.
Kristian Høgsberg [Mon, 10 Mar 2008 22:43:53 +0000 (18:43 -0400)]
Move make install logic for libGL back into src/mesa/Makefile.

This makes make install work again for non-glx libGL implementations.
The make install logic is split into three sub-targets: install-libgl,
install-osmesa, install-drivers.  The install target in src/glx/x11
is then implemented using the src/mesa make install-libgl rule.

Thanks to Dan Nicholson for pointing out the breakage.

16 years agoOnly try to call DRI function if DRI got initialized properly.
Kristian Høgsberg [Mon, 10 Mar 2008 21:59:46 +0000 (17:59 -0400)]
Only try to call DRI function if DRI got initialized properly.

16 years agoDarwin: Config/source fixes to now build glxdemo apps and OSMesa
Jeremy Huddleston [Mon, 10 Mar 2008 22:13:28 +0000 (15:13 -0700)]
Darwin: Config/source fixes to now build glxdemo apps and OSMesa
Also added darwin-fat-32bit darwin-fat-all configs and deleted old darwin-x86ppc config
(cherry picked from commit 7120c0089d663a2b7e7b0c97da38f9bc233fbdd7)

16 years agoautoconf: Add autogen.sh from Xorg for easier setup from git
Dan Nicholson [Fri, 7 Mar 2008 20:04:17 +0000 (12:04 -0800)]
autoconf: Add autogen.sh from Xorg for easier setup from git

The defacto method to rebuild the autotools and run the generated
configure is an autogen.sh script. It is much more discoverable than the
custom `make configure' used here. The Makefile targets are still useful
for creating tarballs, though. This autogen.sh is copied from Xorg.

16 years agoIgnore more demo programs
Dan Nicholson [Mon, 10 Mar 2008 19:36:10 +0000 (12:36 -0700)]
Ignore more demo programs

16 years agoremove extra #include of assert.h (bug 14932)
Brian [Mon, 10 Mar 2008 13:42:58 +0000 (07:42 -0600)]
remove extra #include of assert.h (bug 14932)

16 years agofix parsing of state.texenv.color (bug 14931)
Brian [Mon, 10 Mar 2008 13:41:53 +0000 (07:41 -0600)]
fix parsing of state.texenv.color (bug 14931)

16 years agodri: fix function call to add new parameter.
Dave Airlie [Mon, 10 Mar 2008 09:13:34 +0000 (19:13 +1000)]
dri: fix function call to add new parameter.

This may not be correct but it should get the build going.

16 years agoDRI2: Make setTexBuffer take a __DRIdrawable instead of a BO handle.
Kristian Høgsberg [Mon, 10 Mar 2008 01:28:04 +0000 (21:28 -0400)]
DRI2: Make setTexBuffer take a __DRIdrawable instead of a BO handle.

This fixes a problem where texturing from the same Pixmap more than
once per batchbuffer would hang the DRI driver.  We just use the region
associated with the front left renderbuffer of the __DRIdrawable for
texturing, which avoids creating different regions for the same BO.

This change also make GLX_EXT_texture_from_pixmap work for direct
rendering, since tracking the __DRIdrawable -> BO handle now uses
the standard DRI2 event buffer.  Of course, DRI2 direct rendering
doesn't exist yet.

Finally, this commit bumps the DRI interface version again, accounting
for the change in the DRI_TEX_BUFFER extension and the change in
commit 0bba0e5be7a4a7275dad1edc34bdcc134ea1f424 to pass in the
event buffer head index on drawable creation.

16 years agoDRI2: Drop DriverAPI.UpdateBuffer.
Kristian Høgsberg [Mon, 10 Mar 2008 01:22:49 +0000 (21:22 -0400)]
DRI2: Drop DriverAPI.UpdateBuffer.

__dri2ParseEvents() would determine the kind of event, but then call
UpdateBuffer() in either case, and UpdateBuffer() would then have to
figure that out again to dispatch to HandleBufferAttach() or
HandleDrawableConfig().  Pretty pointless.

16 years agoDRI2: Pass the context instead of the screen to __dri2ParseEvents().
Kristian Høgsberg [Mon, 10 Mar 2008 01:19:17 +0000 (21:19 -0400)]
DRI2: Pass the context instead of the screen to __dri2ParseEvents().

Makes a lot more sense, since the screen is always implicit in the
DRI drawable, but it may not be possible to track down a context from
just a drawable.

16 years agoDRI2: Add event buffer head as an argument to driCreateNewDrawable().
Kristian Høgsberg [Mon, 10 Mar 2008 01:16:30 +0000 (21:16 -0400)]
DRI2: Add event buffer head as an argument to driCreateNewDrawable().

The DRI driver needs to know where in the buffer to start reading.

16 years agofix botched test for clearing color buffers (should fix depth peeling regression)
Brian [Mon, 10 Mar 2008 00:30:07 +0000 (18:30 -0600)]
fix botched test for clearing color buffers (should fix depth peeling regression)

16 years agoSet normalized flag for GLubyte arrays in _mesa_VertexAttribPointerNV()
Markus Amsler [Sun, 9 Mar 2008 23:53:22 +0000 (17:53 -0600)]
Set normalized flag for GLubyte arrays in _mesa_VertexAttribPointerNV()

16 years agoinit vertex weight attrib to (1,0,0,0)
Markus Amsler [Sun, 9 Mar 2008 23:51:11 +0000 (17:51 -0600)]
init vertex weight attrib to (1,0,0,0)

16 years agofix __builtin_expect() definition test for IBM XLC (sf bug 1909832)
Brian [Sun, 9 Mar 2008 16:55:01 +0000 (10:55 -0600)]
fix __builtin_expect() definition test for IBM XLC (sf bug 1909832)

16 years agoProperly revert the libGL loader path breakage.
Kristian Høgsberg [Sun, 9 Mar 2008 23:02:56 +0000 (19:02 -0400)]
Properly revert the libGL loader path breakage.

16 years agolibGL: Fall back to DEFAULT_DRIVER_DIR properly.
Kristian Høgsberg [Sun, 9 Mar 2008 20:21:29 +0000 (16:21 -0400)]
libGL: Fall back to DEFAULT_DRIVER_DIR properly.

After commit 6fd82f6fbd208dc7b1839ea408a5fb28589ee622, we would
overwrite the libPath default value with NULL if libGL was running
non-setuid and none of the env vars were set.

Thanks to Magnus Kessler <Magnus.Kessler@gmx.net> for spotting it.

16 years agoreplace // comment with /* */ (bug 14916)
Brian [Sun, 9 Mar 2008 16:41:50 +0000 (10:41 -0600)]
replace // comment with /* */ (bug 14916)

16 years agoadded info about checking out gallium-0.1 for cell code
Brian [Thu, 28 Feb 2008 16:48:26 +0000 (09:48 -0700)]
added info about checking out gallium-0.1 for cell code

16 years agoUse _X_HIDDEN to hide a bunch of leaked symbols.
Kristian Høgsberg [Sun, 9 Mar 2008 03:28:01 +0000 (22:28 -0500)]
Use _X_HIDDEN to hide a bunch of leaked symbols.

16 years agoMove DRI context functions into dri_glx.c.
Kristian Høgsberg [Sun, 9 Mar 2008 02:57:29 +0000 (21:57 -0500)]
Move DRI context functions into dri_glx.c.

Also drop isDirect flag; if gc->driContext is non-NULL, it's direct.

16 years agoMove DRI drawable creation into dri_glx.c.
Kristian Høgsberg [Sun, 9 Mar 2008 02:02:23 +0000 (21:02 -0500)]
Move DRI drawable creation into dri_glx.c.

16 years agoIntroduce __GLXDRIscreen so we can start moving function pointers in there.
Kristian Høgsberg [Sun, 9 Mar 2008 01:34:24 +0000 (20:34 -0500)]
Introduce __GLXDRIscreen so we can start moving function pointers in there.

Temporarily rename the __DRIscreen member to __driScreen.  Eventually,
we'll move that into __GLXDRIscreen and only access it in dri_glx.c.

16 years agoMove DRI specific parts of CreateContext into dri_glx.c.
Kristian Høgsberg [Sun, 9 Mar 2008 01:02:22 +0000 (20:02 -0500)]
Move DRI specific parts of CreateContext into dri_glx.c.

16 years agoConvert driCreateScreen and driDestroyScreen to function pointers.
Kristian Høgsberg [Sun, 9 Mar 2008 00:15:50 +0000 (19:15 -0500)]
Convert driCreateScreen and driDestroyScreen to function pointers.

We avoid leaking the symbols and will be able to replace them with
DRI2 implementation later on.

16 years agoAbstract __DRIdisplayPrivateRec away in dri_glx.c.
Kristian Høgsberg [Sun, 9 Mar 2008 00:02:10 +0000 (19:02 -0500)]
Abstract __DRIdisplayPrivateRec away in dri_glx.c.

This patch moves __DRIdisplayPrivateRec definition into dri_glx.c and
let's dri_glx.c allocate the __DRIdisplay struct pointer to from
__GLXdisplayPrivate.

A small step towards moving more of the dri functionality into dri_glx.c.

16 years agoMake __glXCloseDisplay static instead of prototyping it to avoid warning.
Kristian Høgsberg [Sat, 8 Mar 2008 22:38:02 +0000 (17:38 -0500)]
Make __glXCloseDisplay static instead of prototyping it to avoid warning.

16 years agoMove libGL install target to the libGL Makefile.
Kristian Høgsberg [Sat, 8 Mar 2008 22:19:39 +0000 (17:19 -0500)]
Move libGL install target to the libGL Makefile.

16 years agoSimplify LIBGL_DRIVERS_PATH walking.
Kristian Høgsberg [Fri, 7 Mar 2008 06:49:33 +0000 (01:49 -0500)]
Simplify LIBGL_DRIVERS_PATH walking.

16 years agoSimplify dri loading code by eliminating dlopen "cache".
Kristian Høgsberg [Fri, 7 Mar 2008 06:37:08 +0000 (01:37 -0500)]
Simplify dri loading code by eliminating dlopen "cache".

No need to jump through hoops to track __DRIdrivers and avoid dlopening the
same .so more than twice, dlopen() does this internally.  Besides, we
were already bypassing this and dlopening drivers for each screen,
whether or not they were already dlopened.

16 years agolibGL: Consolidate DRI initialization in dri_glx.c
Kristian Høgsberg [Fri, 7 Mar 2008 05:45:54 +0000 (00:45 -0500)]
libGL: Consolidate DRI initialization in dri_glx.c

Move a lot of code over from glx_ext.c.

16 years agoautoconf: Print GCC include path directly and be more robust using it
Dan Nicholson [Fri, 7 Mar 2008 19:48:09 +0000 (11:48 -0800)]
autoconf: Print GCC include path directly and be more robust using it

Rather than constructing the GCC include path from `-print-search-dirs',
we can get the path directly from `-print-file-name=include'. This is
used in the Linux kernel build, for example. If no output is returned
from the command, then we don't append a -I path the the makedepend
options.

16 years ago[intel] Only enable GL_EXT_texture_sRGB on i965.
Kristian Høgsberg [Fri, 7 Mar 2008 18:45:09 +0000 (13:45 -0500)]
[intel] Only enable GL_EXT_texture_sRGB on i965.

Fixes #14799.

16 years ago [i965] fix fd.o bug #11471 and #11478
Zou Nan hai [Fri, 7 Mar 2008 07:11:28 +0000 (15:11 +0800)]
  [i965] fix fd.o bug #11471 and #11478
1. Follow EXT_texture_rectangle with YCbCr texture
2. swap UV component for MESA_FORMAT_YCBCR

16 years agoi965: use RGB565 to render a bitmap if Depth is 16
Xiang, Haihao [Fri, 7 Mar 2008 01:48:11 +0000 (09:48 +0800)]
i965: use RGB565 to render a bitmap if Depth is 16

16 years agoi965: Fix double free issue to pass glean/maskedClear test
Xiang, Haihao [Thu, 6 Mar 2008 05:20:57 +0000 (13:20 +0800)]
i965: Fix double free issue to pass glean/maskedClear test

16 years agoi965:fix segfault issue when clearing the window which
Xiang, Haihao [Thu, 6 Mar 2008 01:41:41 +0000 (09:41 +0800)]
i965:fix segfault issue when clearing the window which
is created with mode GLUT_SINGLE|GLUT_RGB|GLUT_DEPTH.
This issue is introduced by 20b8bff49cba3e8246e29004c5ff38f231d589ff

16 years ago[intel] Add a driconf option to cache freed buffer objects for reuse.
Eric Anholt [Wed, 5 Mar 2008 22:14:54 +0000 (14:14 -0800)]
[intel] Add a driconf option to cache freed buffer objects for reuse.

This is defaulted off as it has potentially large memory costs for a modest
performance gain.  Ideally we will improve DRM performance to the point where
this optimization is not worth the memory cost in any case, or find some
middle ground in caching only limited numbers of certain buffers.  For now,
this provides a modest 4% improvement in openarena on GM965 and 10% in openarena
on GM945.

16 years agor300: replace some hard coded mask by define in stencil area
Christoph Brill [Tue, 4 Mar 2008 20:52:49 +0000 (21:52 +0100)]
r300: replace some hard coded mask by define in stencil area

16 years agor300: Fix some issues with masks in stencil buffer area
Christoph Brill [Tue, 4 Mar 2008 20:37:56 +0000 (21:37 +0100)]
r300: Fix some issues with masks in stencil buffer area

16 years ago[dri2] Add tail pointer to reemitDrawableInfo callback.
Kristian Høgsberg [Tue, 4 Mar 2008 00:08:26 +0000 (19:08 -0500)]
[dri2] Add tail pointer to reemitDrawableInfo callback.

When the DRI doesn't parse the event buffer for a while, the X server
may overwrite data that the driver didn't get a chance to look at.  The
reemitDrawableInfo callback requests that the X server reemit all info
for the specified drawable.  To make use of this, the drive needs to know
the new tail pointer so it know where to start reading from.

16 years ago[dri2] Optimize event parsing to skip obsolete events.
Kristian Høgsberg [Tue, 4 Mar 2008 00:04:22 +0000 (19:04 -0500)]
[dri2] Optimize event parsing to skip obsolete events.

This also fixes the problem where the X server does multiple resizes before
the DRI driver gets the events.  The obsolete buffer attach events then
reference already destroyed buffer objects.

16 years ago[intel] Silence unused variable warning when compiling for i965.
Kristian Høgsberg [Tue, 4 Mar 2008 00:07:49 +0000 (19:07 -0500)]
[intel] Silence unused variable warning when compiling for i965.

16 years agonouveau: compilation fixes
Patrice Mandin [Sun, 2 Mar 2008 11:16:23 +0000 (12:16 +0100)]
nouveau: compilation fixes

16 years agostate.depth.range alpha value should be 1, not 0 (bug #14733)
Roland Scheidegger [Sat, 1 Mar 2008 01:56:08 +0000 (02:56 +0100)]
state.depth.range alpha value should be 1, not 0 (bug #14733)

16 years agoUpdate libGL DRI loader to latest DRI interface changes.
Kristian Høgsberg [Fri, 29 Feb 2008 21:42:29 +0000 (16:42 -0500)]
Update libGL DRI loader to latest DRI interface changes.

16 years agoglxgears: oops, remove accidental commit of glFinish() hack.
Kristian Høgsberg [Fri, 29 Feb 2008 20:33:40 +0000 (15:33 -0500)]
glxgears: oops, remove accidental commit of glFinish() hack.

16 years agoUse __DRIextension mechanism providing loader functionality to the driver.
Kristian Høgsberg [Thu, 28 Feb 2008 15:32:28 +0000 (10:32 -0500)]
Use __DRIextension mechanism providing loader functionality to the driver.

Instead of passing in a fixed struct, the loader now passes in a list
of __DRIextension structs, to advertise the functionality it can provide
to the driver.  Each extension is individually versioned and can be
extended or phased out as the interface develops.

16 years agoReduce the versioning madness required to create a DRI2 screen.
Kristian Høgsberg [Tue, 26 Feb 2008 04:37:23 +0000 (23:37 -0500)]
Reduce the versioning madness required to create a DRI2 screen.

Right now the DRI2 screen constructor takes 3 different versions:
DRI, DDX and DRM.  This is mostly useless, though:

  DRI: The DRI driver doesn't actually care about the DRI protocol,
  it only talks to the loader, which in turn speaks DRI protocol.  Thus,
  the DRI protocol version is of not interest to the DRI driver, but it
  needs to know what functionality the loader provides.  At this point
  that's reflected in the __DRIinterfaceMethods struct and the
  internal_version integer.

  DDX: The DDX version number is essentially used to track extensions
  to the SAREA.  With DRI2 the SAREA consists of a number of versioned,
  self-describing blocks, so the DDX version is no longer interesting.

  DRM: We have the fd, lets just ask the kernel ourselves.

16 years agointel: Set the lock flag early to avoid deadlock.
Kristian Høgsberg [Fri, 29 Feb 2008 20:04:26 +0000 (15:04 -0500)]
intel: Set the lock flag early to avoid deadlock.

Another regression from the intel_context.c merge.

16 years agoi965: use _Current pointer instead of Current pointer.
Xiang, Haihao [Fri, 29 Feb 2008 03:15:02 +0000 (11:15 +0800)]
i965: use _Current pointer instead of Current pointer.
fix double free issue(bug#14710). It also corrects glsl/bitmap demo behavior.

16 years ago[965] Fix conditional sequencing to allow the frame_buffer_texobj case.
Eric Anholt [Fri, 29 Feb 2008 00:04:54 +0000 (16:04 -0800)]
[965] Fix conditional sequencing to allow the frame_buffer_texobj case.

Previously the frame_buffer_texobj case would have been bound as a normal
texture, at best resulting in no surface data being associated with it.

16 years ago[intel] Bug #14575: Unmap buffers when deleting/dataing as required.
Eric Anholt [Thu, 28 Feb 2008 23:45:46 +0000 (15:45 -0800)]
[intel] Bug #14575: Unmap buffers when deleting/dataing as required.

Otherwise, we would assertion error when doing the final unreference of the
buffer with an outstanding map (catching the memory leak).