Oliver McFadden
9c74291197
r300: Removed the (obsolete) special source register macros.
2008-03-01 06:33:06 +00:00
Oliver McFadden
440a6fbf6e
r300: Cleaned up the special vertex program source register macros.
2008-03-01 06:33:06 +00:00
Oliver McFadden
8c4d811295
r300: Added the vertex program swizzle (aka selection) defines.
2008-03-01 06:33:06 +00:00
Oliver McFadden
b2c02a4a7c
r300: Converted to the new src/dest register defines.
2008-03-01 06:33:06 +00:00
Oliver McFadden
b0e81fd19e
r300: Removed an obsolete comment from the vertex program header file.
2008-03-01 06:33:06 +00:00
Oliver McFadden
41c3ae5778
r300: Converted to the new Math Engine defines.
2008-03-01 06:33:06 +00:00
Oliver McFadden
0ced26099d
r300: Added the Math Engine opcode macro.
2008-03-01 06:33:06 +00:00
Oliver McFadden
7c4add02f8
r300: Renamed the Vector Engine opcode macro.
2008-03-01 06:33:06 +00:00
Oliver McFadden
bd46a482c4
r300: Converted to the new Vector Engine defines.
2008-03-01 06:33:06 +00:00
Oliver McFadden
66952fa5ae
r300: Removed the duplicate dest register defines.
2008-03-01 06:33:05 +00:00
Oliver McFadden
8a646b80ef
r300: Removed the duplicate "easy" vertex program macros.
2008-03-01 06:33:05 +00:00
Oliver McFadden
ea8299040f
r300: Added the vertex program src/dest register defines.
2008-03-01 06:33:05 +00:00
Oliver McFadden
bccb56d5f3
r300: Added the Vector Engine and Math Engine defines from AMD's documentation.
2008-03-01 06:33:05 +00:00
Oliver McFadden
e0a5194a54
r300: Moved the vertex and fragment program macros into the appropriate files.
2008-03-01 06:33:05 +00:00
Roland Scheidegger
642d5ee8c5
state.depth.range alpha value should be 1, not 0 (bug #14733 )
2008-03-01 02:56:08 +01:00
Kristian Høgsberg
3c5a1e4c2d
Update libGL DRI loader to latest DRI interface changes.
2008-02-29 16:42:29 -05:00
Kristian Høgsberg
51cd168da4
glxgears: oops, remove accidental commit of glFinish() hack.
2008-02-29 15:33:40 -05:00
Kristian Høgsberg
6cb3f5c4d8
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.
2008-02-29 15:05:39 -05:00
Kristian Høgsberg
16242a8007
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.
2008-02-29 15:05:39 -05:00
Kristian Høgsberg
5197a31b8e
intel: Set the lock flag early to avoid deadlock.
...
Another regression from the intel_context.c merge.
2008-02-29 15:05:39 -05:00
Xiang, Haihao
de1e9880f8
i965: use _Current pointer instead of Current pointer.
...
fix double free issue(bug#14710). It also corrects glsl/bitmap demo behavior.
2008-02-29 11:16:21 +08:00
Eric Anholt
b39f15b88f
[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.
2008-02-28 16:04:54 -08:00
Eric Anholt
ec5ca6844b
[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).
2008-02-28 16:01:15 -08:00
Christoph Brill
981b7da8a7
r300: R5xx and R3xx use different registers for RS_IP and RS_INST
...
These changes are taken from the xf86-video-ati driver. They update the header
file accordingly and also remove some UNKOWN variables.
2008-02-28 22:37:15 +01:00
Eric Anholt
9c8f27ba13
[965] Bug #9151 : make fragment.position return window coords not screen coords.
2008-02-28 13:18:48 -08:00
Eric Anholt
3f0301d7bc
[965] Fix serious copy'n'paste failure in brw state dumping.
2008-02-28 13:18:48 -08:00
Brian
ddbfa8c462
added git on Windows section
2008-02-28 08:03:34 -07:00
Brian
cef71f58c9
a few words about how the cell driver works.
2008-02-28 08:03:34 -07:00
Xiang, Haihao
3ffd11f71d
mesa: separate shader program object from shader object.
...
Currently a callback delete_shader_cb is used for deleting shader
and shader program objects. Mesa detaches all attached shaders in
_mesa_free_shader_program_data when deleting shader program objects. However
it is likely that these shaders have been freed in _mesa_free_shader,
which will result in unexpected behaviour. This fix uses a single callback for
shader program objects and deletes shader program objects before shader objects.
2008-02-28 21:32:36 +08:00
Zou Nan hai
0da03c0aa9
[intel] fix compressed image height
2008-02-28 16:49:38 +08:00
José Fonseca
60159c1b09
Convert crlf->lf line endings.
...
Windows/DOS users should enable core.autocrlf from now on:
git config --global core.autocrlf true
2008-02-28 16:34:32 +09:00
Jose Fonseca
90a036b337
Disable autocrlf for Visual Studio project files.
2008-02-28 12:34:01 +09:00
Brian
cabcee9ea9
document cell -DSPU_MAIN_PARAM_LONG_LONG flag
2008-02-27 18:02:23 -07:00
Christoph Brill
00bc91ac64
r300: properly handle GL_POINT in glPolygonMode
...
Until now the polygon mode was completely turned of when you used
GL_POINT. For me it looked buggy to completely disable the polygon
mode for FrontMode and BackMode if any of these was GL_POINT.
2008-02-27 22:06:38 +01:00
Kristian Høgsberg
60c0f09abb
intel: Always use intelInitExtensions() for initializing extensions.
2008-02-27 11:24:29 -05:00
Markus Amsler
c34b024cf4
r300: add artificial output to match fragment program input
2008-02-27 02:06:17 +01:00
Markus Amsler
12a4a74e94
mesa: set input read only on success
2008-02-27 02:04:06 +01:00
Kristian Høgsberg
20b8bff49c
i965: Setup framebuffer texture in meta_draw_region.
...
With DRI2 we there is no screen region until a drawable is bound to
the context. Set up the framebuffer texture in meta_draw_region instead
which should also handle the case where the draw region changes as a
result of resizing a redirected window or resizing the screen.
2008-02-26 18:01:13 -05:00
Eric Anholt
451bd1941e
[965] Don't segfault on INTEL_DEBUG=batch when a surface buffer is NULL.
2008-02-26 13:58:01 -08:00
Eric Anholt
09aa4b3690
[965] remove dead brw_parameter_list_state_flags.
...
This was replaced by Mesa core code.
2008-02-26 13:57:49 -08:00
Christoph Brill
6e3dde8a54
[r300] fix wow-pinkness regression. Fixed by Markus Amsler
2008-02-26 22:10:54 +01:00
Christoph Brill
bfb7cd653a
[r300] Document based on chapter 10.4 and 10.5
...
This commit adds most of the graphics backend registers and of the rasterizer
registers. Again, some minor bugs were found and marked with TODO or even
fixed.
2008-02-26 21:45:58 +01:00
Zou Nan hai
9bd2cb7f90
[i915] fix texture size exceed limit case when running celestia
2008-02-26 17:20:12 +08:00
Thomas Hellstrom
32879e1461
[intel] Handle -EAGAINs correctly in execbuffer.
...
Dont stop on fence creation errors.
2008-02-26 09:19:27 +01:00
Kristian Høgsberg
f1c82bf9a4
A few more fixes to the dlsym() fix.
...
Argh, seriously, I did test that, I just forgot to amend the commit
before pushing.
2008-02-25 22:24:09 -05:00
Dave Airlie
0a64ffb308
dri2: dlsym requires some quotes..
2008-02-26 12:28:45 +10:00
Kristian Høgsberg
129b9ad16f
Use dlsym() to avoid referencing __dri2DriverInitScreen.
...
Sort-of a stop-gap workaround. There are a couple of nicer ways to
do this that doesn't require dlfcn.h, but they're more invasive.
2008-02-25 20:02:25 -05:00
Kristian Høgsberg
bcecea6ea6
autoconf: Add gcc search patch to MKDEP_OPTIONS to eliminate warnings.
...
Also, use -include to avoid error message when make initially fails to
include the non-existent depend file.
2008-02-25 19:02:07 -05:00
Kristian Høgsberg
7381ccab44
intel: Add missing include file to silence last couple of warnings.
2008-02-25 19:02:07 -05:00
Kristian Høgsberg
54bd506c0f
intel: Initialize color and spec arrays in LOCAL_VARS macro.
...
Avoids spamming compilation output with tons of warnings about
use of possibly uninitialized variables.
2008-02-25 19:02:06 -05:00