Commit graph

38766 commits

Author SHA1 Message Date
Chia-I Wu
ea9e5dbbc2 egl: Simplify _eglBindContext.
Remove the hard-to-get-right _eglBindContextToSurfaces.  As well as fix
an assertion failure from b90a3e7d8b when
such call sequence is hit

  eglMakeCurrent(dpy, surf1, surf1, ctx1);
  eglMakeCurrent(dpy, surf2, surf2, ctx2);
  eglMakeCurrent(dpy, surf1, surf1, ctx1);
2010-09-10 18:41:14 +08:00
Andre Maasikas
67270e1fd6 r600: dont bswap rgba FLOAT formats
fixes at least some readback tests in piglit
2010-09-10 12:14:26 +03:00
Chia-I Wu
67660ccee9 targets/egl: Fix crashes from loading invalid modules.
Be defensive.
2010-09-10 16:03:04 +08:00
Chia-I Wu
5ea092117f gallium: Remove ST_API_OPENGL_ES1 and ST_API_OPENGL_ES2.
They are no longer used.
2010-09-10 15:44:19 +08:00
Chia-I Wu
948e3fa27c st/egl: Use profiles to create OpenGL ES contexts.
Replace all uses of ST_API_OPENGL_ES{1,2} by profiles.  Having 3
st_api's to provide OpenGL, OpenGL ES 1.1, and OpenGL ES 2.0 is not a
sane abstraction, since all of them share glapi for current
context/dispatch management.
2010-09-10 15:44:11 +08:00
Chia-I Wu
0cd480f076 st/dri: Use profiles to create OpenGL ES contexts.
Having 3 st_api's to provide OpenGL, OpenGL ES 1.1, and OpenGL ES 2.0 is
not a sane abstraction, since all of them share glapi for current
context/dispatch management.
2010-09-10 15:43:34 +08:00
Chia-I Wu
4531356817 gallium: Add context profile support to st_api.
Add struct st_context_attribs to describe context profiles and
attributes.  Modify st_api::create_context to take the new struct
instead of an st_visual.

st_context_attribs can be used to support GLX_ARB_create_context_profile
and GLX_EXT_create_context_es2_profile in the future.  But the
motivation for doing it now is to be able to replace ST_API_OPENGL_ES1
and ST_API_OPENGL_ES2 by profiles.

Having 3 st_api's to provide OpenGL, OpenGL ES 1.1, and OpenGL ES 2.0 is
not a sane abstraction, since all of them share glapi for current
context/dispatch management.
2010-09-10 15:37:43 +08:00
Dave Airlie
fcae8ca575 r600g: fixup state calculations for picking states.
for evergreen I ended up using a non-contig array of states, but
this code needs a bit of fixing up to deal with that.
2010-09-10 22:41:00 +10:00
Dave Airlie
f61b241eba r600g: fixup CB state numbering header 2010-09-10 21:59:06 +10:00
Dave Airlie
cc1b3b1013 r600g: fix warning in r600 pipe driver 2010-09-10 14:07:49 +10:00
Dave Airlie
ad5ada4372 r600g: evergreen CBs are more sane to support with a single state 2010-09-10 13:52:02 +10:00
Dave Airlie
7d564fdddd r600g: add multi-buffer flush support properly. 2010-09-10 13:52:02 +10:00
Dave Airlie
ab686d340a r600g: fix regression in multi-buffer tests since CB flush merge 2010-09-10 13:52:02 +10:00
Eric Anholt
6ef5f21234 i965: Add switch cases for ir_unop_noise, which should have been lowered.
Fixes compiler warnings.
2010-09-09 19:49:58 -07:00
Dave Airlie
50526e094f r600g: add initial evergreen support
adds shader opcodes + assembler support (except ARL)
uses constant buffers
add interp instructions in fragment shader
adds all evergreen hw states
adds evergreen pm4 support.

this runs gears for me on my evergreen
2010-09-10 19:41:18 +10:00
Dave Airlie
42da027433 r600g: align flushing of cb/db with DDX/r600c.
the DDX and r600c both flush cb/db after the draw is emitted,
as long as they do that, r600g can't be different, as it races.

We end up with r600g flush, set CB, DDX set CB, flush. This
was causing misrendering on my evergreen, where sometimes the drawing
would go to an old CB.
2010-09-10 11:29:23 +10:00
Dave Airlie
e795ca8f31 r600g: don't need 3 bos here.
the code should reloc correctly a single BO 3 times.
2010-09-10 11:29:23 +10:00
Brian Paul
48fda8c446 graw: added test for finding shader mem leaks 2010-09-09 19:19:28 -06:00
Brian Paul
451dfe5413 draw: minor reformatting 2010-09-09 19:19:28 -06:00
Brian Paul
821e390018 graw: emit warnings when context/surface creation failes 2010-09-09 19:19:28 -06:00
Brian Paul
5c3f678790 winsys: emit warning in null_sw_displaytarget_create() 2010-09-09 19:19:28 -06:00
Brian Paul
d851dae919 graw: fix array size, indentation, 2010-09-09 19:19:28 -06:00
Jakob Bornecrantz
379e2e77ba glsl2: Fix scons build for all platforms 2010-09-10 01:17:56 +02:00
Kristian Høgsberg
042a333028 Revert "glapi: Implement optional dispatch logging"
This reverts commit b9abc6139a and the
follow on fixes (7aae704 and 6fe1b47).  It's changing the glapi/driver
ABI and causes a number of problems for debug/non-debug builds.
2010-09-09 19:02:55 -04:00
Ian Romanick
1f3c7d968c glsl2: Implement noise[1234] built-in functions using ir_unop_noise 2010-09-09 15:39:52 -07:00
Ian Romanick
2b70dbfe09 glsl2: Add EmitNoNoise flag, use it to remove noise opcodes 2010-09-09 15:39:52 -07:00
Ian Romanick
547131ac87 glsl2: Add lowering pass to remove noise opcodes 2010-09-09 15:39:51 -07:00
Ian Romanick
3a5ce85cfa glsl2: Add ir_unop_noise 2010-09-09 15:39:51 -07:00
Kenneth Graunke
6dcca5a308 glsl/builtins: normalize of a negative scalar should be -1.0. 2010-09-09 15:23:12 -07:00
Kristian Høgsberg
6fe1b479ca mesa: Only reference logging symbols in debug builds 2010-09-09 18:00:44 -04:00
Kristian Høgsberg
144356f992 mesa: Don't reuse DummyFramebuffer as the incomplete framebuffer
Binding framebuffer 0 on a context that doesn't have a winsys drawable
will try to bind the incomplete framebuffer.  That fails when that's
also the dummy framebuffer.
2010-09-09 17:08:12 -04:00
Jakob Bornecrantz
7aae70406b glapi: Fix non-debug builds 2010-09-09 22:47:42 +02:00
ben
1d61793824 st/dri: support EGL_MESA_image_drm: queryImage
Signed-off-by: Jakob Bornecrantz <wallbraker@gmail.com>
2010-09-09 22:47:38 +02:00
ben
a572e3198d st/dri: support EGL_MESA_image_drm: createImage
Signed-off-by: Jakob Bornecrantz <wallbraker@gmail.com>
2010-09-09 22:47:31 +02:00
Kristian Høgsberg
6eda3f311b eglglx: Convert glx visuals/fbconfigs straight to EGL configs
In other words, skip the __GLcontextModes middle man.
2010-09-09 15:00:18 -04:00
Kristian Høgsberg
01a7eebc4c egl: Remove old egldri driver
This driver doesn't work with any of the DRI drivers in the source tree.
2010-09-09 14:05:30 -04:00
Kristian Høgsberg
b9abc6139a glapi: Implement optional dispatch logging
There's a useful feature buried in glapi to log all API calls to stderr.
Unfortunately it requires editing the code and then it's enabled
unconditionally for that build.  This patch builds in API logging for
debug builds and makes it run-time switchable by setting MESA_DEBUG=dispatch.
2010-09-09 13:21:15 -04:00
Kristian Høgsberg
94118fe2d4 glx: Optimize out no-op make current calls
This make a lot more sense now that we might have to recreate the
glx drawables for legacy code paths.
2010-09-09 13:18:43 -04:00
Chia-I Wu
916c8ed2c8 egl: Use _EGL_CHECK_DISPLAY in eglCreateContext.
_EGL_CHECK_DISPLAY checks the display and returns from eglCreateContext
on error.
2010-09-10 00:18:23 +08:00
Chia-I Wu
08a482e7a9 egl: Display may be NULL in _eglLookup*.
This fixes several NULL dereferences.
2010-09-10 00:18:14 +08:00
Brian Paul
3ec296390c llvmpipe: remove redundant tgsi_dup_tokens() call
We were calling this twice so the first allocation was orphaned/leaked.
2010-09-09 10:04:39 -06:00
Francisco Jerez
cdd5f21eac dri/nouveau: Expose EXT_texture_env_combine. 2010-09-09 14:19:36 +02:00
Francisco Jerez
3bbad7f108 dri/nv10-nv20: Add support for NV_texture_env_combine4. 2010-09-09 14:19:35 +02:00
Francisco Jerez
699749cfee dri/nv04: Add support for NV_texture_env_combine4. 2010-09-09 14:19:35 +02:00
Francisco Jerez
a76f6dc849 dri/nouveau: Minor cleanup. 2010-09-09 14:19:35 +02:00
Kristian Høgsberg
80e48ddcf6 glx: Fix another use-after-free problem 2010-09-09 08:06:51 -04:00
Andre Maasikas
777f352e60 r600: add TXL instruction and note about TXB 2010-09-09 11:08:11 +03:00
Andre Maasikas
1f12945d2f r600: remove depth exports move, just set to output x <- z in the export intruction 2010-09-09 11:08:11 +03:00
Andre Maasikas
7753416c5b r600: protect cleanup instructions from double free
We might get the cleanup when we have not translated the shader yet
e.g 2 programstringnotifys in a row
2010-09-09 11:08:11 +03:00
Andre Maasikas
5697bf1660 r600: remove mask from output intructions
in case of relative addressing we never get to know which comps
were really written.
2010-09-09 11:08:11 +03:00