OSDN Git Service
Brian Paul [Tue, 19 Jul 2005 15:27:03 +0000 (15:27 +0000)]
overlay planes test
Brian Paul [Mon, 18 Jul 2005 20:07:19 +0000 (20:07 +0000)]
s/clippping/clipping/
Philippe Houdoin [Mon, 18 Jul 2005 19:27:49 +0000 (19:27 +0000)]
Fixed warnings by using GLUTProc type instead of anonymous (void *) for
GLUT callbacks.
Alan Hourihane [Mon, 18 Jul 2005 13:53:03 +0000 (13:53 +0000)]
re-enable the ffb driver
Alan Hourihane [Mon, 18 Jul 2005 13:52:34 +0000 (13:52 +0000)]
Fix the FFB driver for the renderbuffer changes
Support NEW INTERFACE as well
Alan Hourihane [Mon, 18 Jul 2005 13:50:06 +0000 (13:50 +0000)]
pass in the renderbuffer
Ian Romanick [Mon, 18 Jul 2005 12:33:03 +0000 (12:33 +0000)]
Missed this on the previous commit.
Ian Romanick [Mon, 18 Jul 2005 12:31:24 +0000 (12:31 +0000)]
Wrap every place that accesses a dispatch table with a macro. A new script-
generated file, called src/mesa/glapi/dispatch.h, is added. This file
contains three macros for each API function. It contains a GET, a SET, and
a CALL. Each of the macros take a pointer to the context and a pointer to
the dispatch table.
In several threads on mesa3d-dev we discussed replacing _glapi_add_entrypoint
with a new function called _glapi_add_dispatch. For this discussion, the
important difference between the two is that the caller of _glapi_add_dispatch
does *not* know what the dispatch offset will be at compile time. Because of
this callers need to track the dispatch offset returned by
_glapi_add_dispatch.
http://marc.theaimsgroup.com/?t=
111947074700001&r=1&w=2
The downside is that driver code then has to access the dispatch table two
different ways. It accesses it using structure tags (e.g., exec->Begin) for
functions with fixed offsets and via a remap table (e.g., exec[
remap->NewExtensionFunction ]) for functions without fixed offsets. Yuck!
Using the macros allows both types of functions to be accessed
identically. If a driver needs to set a pointer for Begin, it does
'SET_Begin(ctx, exec, my_begin_function)'. If it needs to set a pointer
for NewExtensionFunction, it does 'SET_NewExtensionFunction(ctx, exec,
my_NewExtensionFunction_function)'. Furthermore, if at some point in
the future a static offset is assigned for NewExtensionFunction, only
the macros need to change (instead of every single place that accesses a
table for that function).
This code differs slightly from the originally posted patches in that the
CALL, GET, and SET marcos no longer take a context pointer as a parameter.
Brian Paul had suggested that the remap table could be stored as a global
since it would be set at CreateScreen time and would be constant for all
contexts. This change reflects that feedback.
http://marc.theaimsgroup.com/?t=
112087194700001&r=1&w=2
Adam Jackson [Sat, 16 Jul 2005 04:02:05 +0000 (04:02 +0000)]
XMesaResizeBuffers needs separate paths for client and server builds
Adam Jackson [Sat, 16 Jul 2005 00:56:20 +0000 (00:56 +0000)]
strip out vestigial #ifdef HAVE_CONFIG_H stanzas, they're confusing the
modular X build
Brian Paul [Fri, 15 Jul 2005 23:13:46 +0000 (23:13 +0000)]
disable call to via_sse_memcpy() until it's fixed
Brian Paul [Fri, 15 Jul 2005 23:08:06 +0000 (23:08 +0000)]
need to flush vertices for GL_OCCLUSION_TEST_RESULT_HP query
Brian Paul [Fri, 15 Jul 2005 22:48:09 +0000 (22:48 +0000)]
Check for null program->Parameters pointer in validate_vertex_program().
GL_NV_vertex_program programs can't directly reference GL state so this
pointer is always NULL.
Brian Paul [Fri, 15 Jul 2005 22:47:00 +0000 (22:47 +0000)]
added comments
Brian Paul [Fri, 15 Jul 2005 20:17:02 +0000 (20:17 +0000)]
remove duplicate "User disable" string
Brian Paul [Fri, 15 Jul 2005 14:31:31 +0000 (14:31 +0000)]
redo previous check-in, just set ARCH_FLAGS = -m32
Brian Paul [Fri, 15 Jul 2005 13:58:58 +0000 (13:58 +0000)]
define CFLAGS and CXXflags with -m32 (Egbert Eich)
Brian Paul [Fri, 15 Jul 2005 13:55:20 +0000 (13:55 +0000)]
replace gcc with $(CC) (Egbert Eich)
Brian Paul [Thu, 14 Jul 2005 23:16:41 +0000 (23:16 +0000)]
eglgears demo (Dane Rushton)
Brian Paul [Thu, 14 Jul 2005 23:10:21 +0000 (23:10 +0000)]
added a few more fallbackStrings (Andreas Stenglein)
Keith Whitwell [Wed, 13 Jul 2005 17:49:22 +0000 (17:49 +0000)]
Use the LDEXPF wrapper macro.
Keith Whitwell [Wed, 13 Jul 2005 17:47:40 +0000 (17:47 +0000)]
Add a LDEXPF wrapper macro.
Keith Whitwell [Wed, 13 Jul 2005 13:50:31 +0000 (13:50 +0000)]
Wrap RESTORE_FPU/RND_NEG_FPU declarations with check for USE_SSE_ASM.
Keith Whitwell [Mon, 11 Jul 2005 10:10:38 +0000 (10:10 +0000)]
Cache texenv programs to avoid repeated compilation (Ben Skeggs)
Keith Whitwell [Sun, 10 Jul 2005 11:23:10 +0000 (11:23 +0000)]
Add runtime compiled x87 implementations for most of the remaining
opcodes. When enabled via environment vars, gears runs and almost
looks right but other apps are still quite buggy.
Keith Whitwell [Sun, 10 Jul 2005 11:14:00 +0000 (11:14 +0000)]
Add lots of x87 fpu instructions.
Ian Romanick [Fri, 8 Jul 2005 21:55:56 +0000 (21:55 +0000)]
Restore linux-x86-32 config for Brian.
Ian Romanick [Thu, 7 Jul 2005 20:43:01 +0000 (20:43 +0000)]
Put back the glPopMatrix. This test was derived from blendminmax.c.
When extra code was cut from blendminmax.c's Display function, the
glPopMatrix call was also accidentally cut.
Ian Romanick [Thu, 7 Jul 2005 19:35:30 +0000 (19:35 +0000)]
Correct the comments in the file headers. Oops.
Ian Romanick [Thu, 7 Jul 2005 19:34:06 +0000 (19:34 +0000)]
Add regression test for bug #3101.
Roland Scheidegger [Wed, 6 Jul 2005 12:49:46 +0000 (12:49 +0000)]
Re-enable r200InitSpanFuncs, which was ifdefed out (probably by mistake?),
so r200SetBuffer, r200SpanRenderStart and r200SpanRenderFinish get called again.
Bugzilla #3705, #3706.
Brian Paul [Tue, 5 Jul 2005 14:13:42 +0000 (14:13 +0000)]
The old MESA_PBUFFER_ALLOC() function allocated memory on 512-byte boundaries.
Restore that behavior with new _mesa_alloc_texmemory() function.
Should fix via_sse_memcpy() problem in found with flightgear.
Jon Smirl [Sat, 2 Jul 2005 21:28:01 +0000 (21:28 +0000)]
Put default back to the normal DRM path
Ian Romanick [Sat, 2 Jul 2005 08:29:57 +0000 (08:29 +0000)]
Add support for assembly (static) dispatch functions on x86-64. This
is basically patch #2939 from X.org bugzilla #3379. This does *not*
fix the bug as it does not dynamically generate stubs at run-time. It
just gets things one step closer.
Ian Romanick [Sat, 2 Jul 2005 07:48:41 +0000 (07:48 +0000)]
linux-x86-32 is unnecessary. The same thing can be acomplished by
'make linux-x86 ARCH_FLAGS=-m32'.
Ian Romanick [Sat, 2 Jul 2005 07:38:40 +0000 (07:38 +0000)]
Massive refactor of (most of) the Linux config files. This reduces
the length of each config and brings a lot of much needed uniformity
to them.
Ian Romanick [Sat, 2 Jul 2005 07:12:44 +0000 (07:12 +0000)]
Add $(CFLAGS) to the link commands. This makes 'make linux-x86
ARCH_FLAGS=-m32' work correctly on x86-64 systems.
Ian Romanick [Sat, 2 Jul 2005 07:11:54 +0000 (07:11 +0000)]
Make linux-x86 more like the other linux configs.
Vladimir Dergachev [Sat, 2 Jul 2005 01:09:10 +0000 (01:09 +0000)]
Expose the fact that we do not do either selection of feedback buffers - either with software or hardware rendering.
Jerome Glisse [Fri, 1 Jul 2005 21:26:43 +0000 (21:26 +0000)]
Adding RV350_NJ ID.
Jerome Glisse [Fri, 1 Jul 2005 21:22:05 +0000 (21:22 +0000)]
Adapt extension string to new GL dispatch schema proposed
by Idr. This patch is based on Idr patch to radeon driver.
Change #if 1 to #if 0 (r300_context.c:l69) for old dispatch
tab.
Karl Schultz [Fri, 1 Jul 2005 21:08:00 +0000 (21:08 +0000)]
Update notes with VC7 info.
Karl Schultz [Fri, 1 Jul 2005 20:31:41 +0000 (20:31 +0000)]
Add windows VC7 project files to filelist for tar/zip files.
Karl Schultz [Fri, 1 Jul 2005 20:25:57 +0000 (20:25 +0000)]
VC7 project file for building glut lib
Karl Schultz [Fri, 1 Jul 2005 20:25:23 +0000 (20:25 +0000)]
VC7 project file for gears demo
Karl Schultz [Fri, 1 Jul 2005 20:24:49 +0000 (20:24 +0000)]
VC7 Solution file for building glut and demos
Karl Schultz [Fri, 1 Jul 2005 20:23:52 +0000 (20:23 +0000)]
VC7 project file for osmesa
Karl Schultz [Fri, 1 Jul 2005 20:23:23 +0000 (20:23 +0000)]
VC7 project file for core mesa lib
Karl Schultz [Fri, 1 Jul 2005 20:22:41 +0000 (20:22 +0000)]
VC7 project file for glu
Karl Schultz [Fri, 1 Jul 2005 20:21:45 +0000 (20:21 +0000)]
VC7 project file for gdi driver
Karl Schultz [Fri, 1 Jul 2005 20:19:47 +0000 (20:19 +0000)]
Visual Studio 7 solution file for mesa lib
Karl Schultz [Fri, 1 Jul 2005 20:11:46 +0000 (20:11 +0000)]
Don't #define __FUNCTION__ on a Microsoft C compiler new enough to have it.
Keith Whitwell [Fri, 1 Jul 2005 17:12:29 +0000 (17:12 +0000)]
Set fp->Opcode for OP_TEX_KILL case, fix from Ben Skeggs.
Karl Schultz [Fri, 1 Jul 2005 15:59:33 +0000 (15:59 +0000)]
Add new files to project.
Karl Schultz [Fri, 1 Jul 2005 15:58:02 +0000 (15:58 +0000)]
Use frexp on Windows; frexpf is not available.
Karl Schultz [Fri, 1 Jul 2005 15:56:14 +0000 (15:56 +0000)]
Fix a prototype
Roland Scheidegger [Fri, 1 Jul 2005 11:28:14 +0000 (11:28 +0000)]
convert sis and radeon driver to use spantmp2 instead of spantmp template. Remove old (pre-renderbuffer) span code instead of converting that too. Remove this old code from mach64 (the dead code was not fully converted to spantmp2 previously) too.
Brian Paul [Fri, 1 Jul 2005 02:03:33 +0000 (02:03 +0000)]
comment-out some unused code to silence warnings
Brian Paul [Fri, 1 Jul 2005 02:03:00 +0000 (02:03 +0000)]
replace some free() calls with _mesa_free()
Brian Paul [Fri, 1 Jul 2005 01:55:30 +0000 (01:55 +0000)]
updates to file list for tarballs
Brian Paul [Fri, 1 Jul 2005 01:54:38 +0000 (01:54 +0000)]
fix dependencies
Brian Paul [Fri, 1 Jul 2005 01:34:29 +0000 (01:34 +0000)]
added some glGetRenderbufferParameterivEXT calls, printfs, and assertions
Brian Paul [Fri, 1 Jul 2005 01:30:03 +0000 (01:30 +0000)]
fix typo in assertions
Brian Paul [Fri, 1 Jul 2005 01:27:00 +0000 (01:27 +0000)]
Move the PYTHON2 and PYTHON_FLAGS assignments to configs/current, and
include $(TOP)/configs/current in glapi/Makefile so those vars can be
easily overridden by any system config, if needed.
Brian Paul [Fri, 1 Jul 2005 01:22:25 +0000 (01:22 +0000)]
Remove NEW_RENDERBUFFER stuff.
Now, OLD_RENDERBUFFER marks code that needs to eventually be removed when
all the drivers are updated to no longer need the SetBuffer() function.
Brian Paul [Fri, 1 Jul 2005 01:10:36 +0000 (01:10 +0000)]
remove OLD_RENDERBUFFER code
Brian Paul [Fri, 1 Jul 2005 01:08:29 +0000 (01:08 +0000)]
remove OLD_RENDERBUFFER stuff
Brian Paul [Fri, 1 Jul 2005 01:04:31 +0000 (01:04 +0000)]
minor fixes
Roland Scheidegger [Fri, 1 Jul 2005 00:08:46 +0000 (00:08 +0000)]
remove common macros used in the span functions of most drivers from the individual drivers and put them in common code. It is still possible for a driver to define its own macros if it has special needs. This affects CLIPPIXEL, CLIPSPAN, HW_CLIPLOOP, HW_ENDCLIPLOOP, and for drivers using the spantmp2 template also GET_SRC_PTR and GET_DST_PTR.
Ian Romanick [Thu, 30 Jun 2005 16:00:48 +0000 (16:00 +0000)]
Replace add_newer_entrypoints (src/mesa/main/context.c) with
device-specific code. A new Python script
(src/mesa/glapi/extension_helper.py) generates a list of all
entry-points for all known extensions. Each driver the selects only
the extensions that it needs and enables the via either
driInitExtensions or driInitSingleExtension.
This code has been compile-tested on a drivers, but has only been
run-tested on mga and i915 (on i830 hardware).
These changes were discussed at length on the mesa3d-dev mailing list.
http://marc.theaimsgroup.com/?t=
111947074700001&r=1&w=2
Brian Paul [Thu, 30 Jun 2005 14:24:44 +0000 (14:24 +0000)]
remove \n from warning message string
Brian Paul [Thu, 30 Jun 2005 14:23:40 +0000 (14:23 +0000)]
added newline at end of file
Brian Paul [Thu, 30 Jun 2005 14:22:23 +0000 (14:22 +0000)]
Add a set of predicate functions for testing matrices instead of directly
testing the flags field.
Move definition of all the MAT_FLAGs into the m_matrix.c file since they're
now private.
Brian Paul [Thu, 30 Jun 2005 13:42:10 +0000 (13:42 +0000)]
Redo all the GL_LINEAR interpolation code in terms of LERP macros/functions.
This cleans up and simplifies the arithmetic quite a bit.
Roland Scheidegger [Thu, 30 Jun 2005 11:42:56 +0000 (11:42 +0000)]
Fix crashes with pixel readback when using the optimized assembly functions.
Pixel count can be negative (this could be fixed elsewhere), so adapt the
functions to work with such inputs correctly (same behaviour as non-optimized
functions).
Bugzilla #2317
Submitted by idr
Aapo Tahkola [Thu, 30 Jun 2005 04:37:19 +0000 (04:37 +0000)]
-"fix" page flipping
-use depth tiling if tiling is enabled
Brian Paul [Thu, 30 Jun 2005 02:34:07 +0000 (02:34 +0000)]
fix a few glitches
Brian Paul [Thu, 30 Jun 2005 02:33:25 +0000 (02:33 +0000)]
fix Darwin dynamic lib naming
Brian Paul [Tue, 28 Jun 2005 03:48:08 +0000 (03:48 +0000)]
just code clean-ups and new comments
Vladimir Dergachev [Mon, 27 Jun 2005 18:46:31 +0000 (18:46 +0000)]
Uncomment a couple of helper functions.
Fix slipup from CVS update that was commented out and did not show up during compilation.
Vladimir Dergachev [Mon, 27 Jun 2005 15:56:14 +0000 (15:56 +0000)]
Rename unchecked state to match DRM driver.
Emit wait idle and pacify r300 before emitting state - this seems to improve stability.
Brian Paul [Mon, 27 Jun 2005 00:45:36 +0000 (00:45 +0000)]
include t_vp_build.h
Brian Paul [Mon, 27 Jun 2005 00:39:33 +0000 (00:39 +0000)]
remove unused MEMSET16 macro
Brian Paul [Mon, 27 Jun 2005 00:34:17 +0000 (00:34 +0000)]
Get rid of the MESA_PBUFFER_ALLOC/FREE() macros.
If that stuff is still needed, lots of other updates are needed anyway.
Also, some misc MALLOC/FREE -> _mesa_malloc/free() changes.
Brian Paul [Mon, 27 Jun 2005 00:30:56 +0000 (00:30 +0000)]
added _tnl_ProgramCacheDestroy() prototype
Brian Paul [Mon, 27 Jun 2005 00:30:37 +0000 (00:30 +0000)]
make create_new_program() static
Eric Anholt [Sun, 26 Jun 2005 21:57:23 +0000 (21:57 +0000)]
Fix crashes during rasterization fallback by avoiding _tnl_need_projected_coords
during fallbacks. In one case, _swsetup_Wakeup had just been called, covering
the need there, and in the other case, we can simply exit the entire
radeonChooseVertexState function, knowing that it will be called again once we
leave the fallback.
Bugzilla #: 2516
Submitted by: sroland
Eric Anholt [Sun, 26 Jun 2005 21:54:43 +0000 (21:54 +0000)]
Remove the #if 0-ed out old span code.
Eric Anholt [Sun, 26 Jun 2005 21:53:11 +0000 (21:53 +0000)]
Explain the reason for TCL fallback when GL_SPHERE_MAP happens with
RADEON_DEBUG=fall.
Eric Anholt [Sun, 26 Jun 2005 21:02:58 +0000 (21:02 +0000)]
Zero out the ENGPACKET allocated on the stack, to avoid sending uninitialized
values to reserved fields on the card, resulting in all-black output and
sometimes hangs.
Submitted by: Thomas Winischhofer
Brian Paul [Sun, 26 Jun 2005 20:08:23 +0000 (20:08 +0000)]
fix for-loop in _mesa_GetDouble to avoid out of bounds memory read
Brian Paul [Sun, 26 Jun 2005 19:57:17 +0000 (19:57 +0000)]
check for null oldBuffer in _mesa_realloc(), cleaned up some comments
Brian Paul [Sun, 26 Jun 2005 19:53:52 +0000 (19:53 +0000)]
need parenthesis in realloc call
Ian Romanick [Fri, 24 Jun 2005 18:54:57 +0000 (18:54 +0000)]
Ignore files generated as part of the build process.
Ian Romanick [Fri, 24 Jun 2005 18:35:31 +0000 (18:35 +0000)]
Fix non-x86 specific builds. The changes from glTexImage3DEXT to
glTexImage3D that caused me so many problems during the re-development
of the API scripts reared its ugly head again. This has been fixed by
tracking the parameter string for each entry-point individually.
This has the annoying side-effect that the names of the parameters in
all aliases of a function must be the same or gl_apitemp.py will
generate bad code. :( The changes in
src/mesa/glapi/{gl_API.xml,glapitable.h} and src/glx/x11/* are caused
by fixing the parameter names in various function aliases that didn't
match.
Reported by: Eric Anholt, Jacob Jansen
Jeremy Kolb [Fri, 24 Jun 2005 02:28:02 +0000 (02:28 +0000)]
Fixed the XCB implementation when always_array is true, thanks to Ian Romanick for pointing it out. Please review.
Brian Paul [Thu, 23 Jun 2005 15:09:32 +0000 (15:09 +0000)]
Remove the xmesa_set_buffer() function. No longer needed since the
span functions pass in a gl_renderbuffer to indicate which color
buffer should be drawn into. Optimized line/triangle routines are
smart enough to know which buffer to draw into as well.
The swrast->SetBuffer() routine should eventually be removed from
all drivers.
Ian Romanick [Wed, 22 Jun 2005 17:11:20 +0000 (17:11 +0000)]
Mark GenQueriesARB at 'always_array="true"'. This eliminates the need
to special-case the handling of that function in glX_proto_send.py.
Jeremy Kolb [Wed, 22 Jun 2005 02:36:23 +0000 (02:36 +0000)]
Fixed a bug where GenQueriesARB was generating the wrong code for XCB and wouldn't compile. Not sure how that slipped through before but it should work now.
Ian Romanick [Tue, 21 Jun 2005 23:42:43 +0000 (23:42 +0000)]
Mammoth update to the Python code generator scripts that live in
src/mesa/glapi. Basically, the scripts that did simple things (like
gl_offsets.py) were simple, and the scripts that did more complicated things
(like glX_proto_send.py) were getting progressively more and more out of
control. So, I re-write the foundation classes on which everything is based.
One problem with the existing code is that the division between the GL API
database representation and the way the output code is generated was either
blury or nonexistant. The new code somewhat follows the
Model-View-Controller pattern, minus the Controller. There is a distinct
set of classes that model the API data, and there is a distinct set of
classes that generate code from that data.
One big change is in the class that represents GL functions (was glFunction,
is now gl_function). There used to be an instance of this calls for each
function and for each alias to that function. For example, there was an
instance for PointParameterivSGIS, PointParameterivEXT, PointParameterivARB,
and PointParameteriv. In the new code, there is one instance. Each
instance has a list of entrypoint names for the function. In the next
revision, this will allow a couple useful things. The script will be able
to verify that the parameters, return type, and GLX protocol for a function
and all it's aliases match.
It will also allow aliases to be represented in the XML more compactly.
Instead of repeating all the information, an alias can be listed as:
<function name="PointParameterivARB" alias="PointParameterivEXT"/>
Because the data representation was changed, the order that the alias
functions are processed by the scripts also changed. This accounts for at
least 2,700 of the ~3,600 lines of diffs in the generated code.
Most of the remaining ~900 lines of diffs are the result of bugs *fixed* by
the new scripts. The old scripts also generated code with some bugs in it.
These bugs were discovered while the new code was being written.
These changes were discussed on the mesa3d-dev mailing list back at the end
of May:
http://marc.theaimsgroup.com/?t=
111714569000004&r=1&w=2
Xorg bug: 3197, 3208