Brian Paul [Wed, 15 Apr 2009 01:44:16 +0000 (19:44 -0600)]
mesa: use standard offsetof() macro
Brian Paul [Tue, 14 Apr 2009 22:05:28 +0000 (16:05 -0600)]
i965: fix VS constant buffer reads
This mostly came down to finding the right MRF incantation in the
brw_dp_READ_4_vs() function.
Note: this feature is still disabled (but getting close to done).
Adam Jackson [Tue, 14 Apr 2009 21:56:29 +0000 (17:56 -0400)]
glx: Make glXGetScreenDriver() work for DRI2
José Fonseca [Tue, 14 Apr 2009 20:39:54 +0000 (21:39 +0100)]
scons: Support winddk 6001.18002.
José Fonseca [Tue, 14 Apr 2009 20:31:34 +0000 (21:31 +0100)]
scons: Recent Windows DDK do not include LIB.EXE.
Have to use LINK /LIB instead. The biggest problem is when the command
line is very long and all the options are included in a argument file --
link doesn't like if /LIB is included in the argument file.
José Fonseca [Tue, 14 Apr 2009 20:27:58 +0000 (21:27 +0100)]
mesa: Fix gcc assembly enable logic.
The i386 symbol is defined on WINDDK.
Ian Romanick [Sat, 11 Apr 2009 21:02:33 +0000 (14:02 -0700)]
DRI2: Don't fault on NULL DrawBuffer
It is possible for ctx->DrawBuffer to be NULL, so don't fault when
that happens. This change is not being committed to master because it
doesn't appear to be necessary there.
Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
Cherry picked from mesa_7_4_branch, commit
49e0c74ddd91900fc4effb6d305d56e0563b456d
Brian Paul [Tue, 14 Apr 2009 17:08:42 +0000 (11:08 -0600)]
i965: checkpoint commit: VS constant buffers
Hook up a constant buffer, binding table, etc for the VS unit.
This will allow using large constant buffers with vertex shaders.
The new code is disabled at this time (use_const_buffer=FALSE).
Younes Manton [Tue, 14 Apr 2009 01:24:44 +0000 (21:24 -0400)]
dri glx: Swap before checking for cliprects.
We don't update drawables anymore unless they are completely
uninitialized, so we need to swap even if we don't have
cliprects yet, otherwise we never end up calling the driver's
SwapBuffers(). The driver should update the drawable in its
SwapBuffers() anyway.
See
8e753d04045a82062ac34d3b2622eb9dba8af374,
"dri glx: Fix dri_util::driBindContext" for the change that
exposed it.
Ben Skeggs [Mon, 13 Apr 2009 04:41:06 +0000 (14:41 +1000)]
nv50: fix mrt clear
Corbin Simpson [Sat, 11 Apr 2009 01:34:55 +0000 (18:34 -0700)]
r300-gallium: Split up vertex format tallying for HW and SW TCL.
This makes things draw (again) with HW TCL. Yay?
Corbin Simpson [Sat, 11 Apr 2009 01:15:46 +0000 (18:15 -0700)]
r300-gallium: Finish up cleanup of vertex format state.
This makes texcoords route properly, and also fixes a few asserts.
Corbin Simpson [Fri, 10 Apr 2009 23:14:12 +0000 (16:14 -0700)]
r300-gallium: Clean up vertex format setup.
Brian Paul [Fri, 10 Apr 2009 19:32:04 +0000 (13:32 -0600)]
i965: remove unused var
Brian Paul [Fri, 10 Apr 2009 19:21:27 +0000 (13:21 -0600)]
i965: clean-up in prepare_wm_surfaces()
Brian Paul [Fri, 10 Apr 2009 17:17:35 +0000 (11:17 -0600)]
mesa: reduce makefile output
José Fonseca [Fri, 10 Apr 2009 18:48:59 +0000 (19:48 +0100)]
wgl: Protect the framebuffer with a lock.
Unfortunately this doesn't catch all the cases, as the mesa state tracker
can still use the framebuffer without giving the wgl state tracker
the chance to lock it.
José Fonseca [Fri, 10 Apr 2009 17:43:51 +0000 (18:43 +0100)]
wgl: Note down the gallium pixel formats, instead of re-guessing them.
José Fonseca [Fri, 10 Apr 2009 17:14:22 +0000 (18:14 +0100)]
wgl: Query the screen for supported formats.
José Fonseca [Fri, 10 Apr 2009 16:22:21 +0000 (17:22 +0100)]
wgl: Walk through the pixel formats combinations.
José Fonseca [Fri, 10 Apr 2009 15:37:15 +0000 (16:37 +0100)]
wgl: Keep copies of the PIXELFORMATDESCRIPTOR structures.
Robert Noland [Fri, 10 Apr 2009 16:41:27 +0000 (11:41 -0500)]
Build dri drivers for powerpc and sparc on FreeBSD.
Brian Paul [Fri, 10 Apr 2009 14:56:03 +0000 (08:56 -0600)]
mesa: asst. progs/test/Makefile files
Brian Paul [Fri, 10 Apr 2009 14:36:04 +0000 (08:36 -0600)]
i965: added null const_buffer pointer check in update_constant_buffer()
Brian Paul [Fri, 10 Apr 2009 14:35:21 +0000 (08:35 -0600)]
intel: added screen->dri2.loader null pointer check in intel_flush()
Fixes segfaults when rendering to front buffer.
Brian Paul [Fri, 10 Apr 2009 14:05:10 +0000 (08:05 -0600)]
gallium: remove pipe_texture::compressed field
The format field encodes compressed vs. uncompressed already. We can easily
check if a texture is compressed with pf_is_compressed(texture->format).
José Fonseca [Fri, 10 Apr 2009 12:18:14 +0000 (13:18 +0100)]
wgl: Slightly simplify pixelformat creation.
José Fonseca [Fri, 10 Apr 2009 12:08:57 +0000 (13:08 +0100)]
progs: Port glxinfo to wgl.
José Fonseca [Fri, 10 Apr 2009 11:58:05 +0000 (12:58 +0100)]
mesa: Don't define WGL_ARB_extensions_string in mesa_wgl.h.
It breaks wglext.h as it doesn't define PFNWGLGETEXTENSIONSSTRINGARBPROC.
José Fonseca [Fri, 10 Apr 2009 09:30:14 +0000 (10:30 +0100)]
wgl: Move pixel format information into device structure.
José Fonseca [Fri, 10 Apr 2009 09:14:13 +0000 (10:14 +0100)]
wgl: Combine alpha pixelformat info in color.
José Fonseca [Fri, 10 Apr 2009 09:04:31 +0000 (10:04 +0100)]
stw: Reorder the pixelformats, so that no alpha pixelformats get chosen before alpha.
Some applications might not handle a alpha visual when when they did not
request for one.
José Fonseca [Fri, 10 Apr 2009 09:02:58 +0000 (10:02 +0100)]
stw: Don't make assumptions about integer size.
José Fonseca [Fri, 10 Apr 2009 09:02:29 +0000 (10:02 +0100)]
stw: Free the mesa framebuffer.
Michal Krol [Fri, 10 Apr 2009 09:58:49 +0000 (11:58 +0200)]
tgsi/sse2: Cleanup NRM/NRM4 implementation.
Fix comments.
Make sure .w is set to 1.0 for NRM.
Optimise for non-.xyzw writemasks.
Michal Krol [Fri, 10 Apr 2009 07:00:57 +0000 (09:00 +0200)]
rtasm: Silence `static function not used' warning.
Michal Krol [Fri, 10 Apr 2009 06:59:20 +0000 (08:59 +0200)]
draw: Silence pointer-to-integer cast warning.
Michal Krol [Fri, 10 Apr 2009 06:54:47 +0000 (08:54 +0200)]
draw: Fix vertex_header initialisation.
Brian Paul [Thu, 9 Apr 2009 21:31:48 +0000 (15:31 -0600)]
swrast: remove some unneeded CHAN_TYPE!=GL_FLOAT code
Brian Paul [Fri, 10 Apr 2009 00:40:24 +0000 (18:40 -0600)]
gallium: comments for surface_copy(), surface_fill() in p_context.h
Brian Paul [Fri, 10 Apr 2009 00:37:03 +0000 (18:37 -0600)]
i965: re-org of some of the new constant buffer code
Plus, begin the new code for vertex shader const buffers.
Brian Paul [Thu, 9 Apr 2009 23:04:58 +0000 (17:04 -0600)]
demos: fix aspect ratio in Reshape()
Brian Paul [Thu, 9 Apr 2009 23:04:09 +0000 (17:04 -0600)]
i965: new SURF_INDEX_ macros
Used to map drawables, textures and constant buffers to surface binding
table indexes.
Brian Paul [Thu, 9 Apr 2009 22:10:43 +0000 (16:10 -0600)]
gallium: change // comment to /* */ style
Brian Paul [Thu, 9 Apr 2009 22:02:41 +0000 (16:02 -0600)]
gallium: updated comments in p_screen.h
Brian Paul [Thu, 9 Apr 2009 21:09:42 +0000 (15:09 -0600)]
i965simple: remove pipe_texture::compressed reference
Michal Krol [Thu, 9 Apr 2009 21:24:57 +0000 (23:24 +0200)]
tgsi/sse2: Fix build.
Michal Krol [Thu, 9 Apr 2009 21:17:02 +0000 (23:17 +0200)]
draw: Fix LIT instruction.
Ian Romanick [Tue, 7 Apr 2009 18:23:10 +0000 (11:23 -0700)]
intel / DRI2: Accept fake front-buffer from loader
Handle the loader returning a fake front-buffer. Since the driver
never specifically requests a fake front-buffer, the driver assumes
that it will never receive both a fake and a real front-buffer.
Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
Reviewed-by: Kristian Høgsberg <krh@redhat.com>
Ian Romanick [Mon, 6 Apr 2009 22:35:37 +0000 (15:35 -0700)]
DRI2: Assume that there is always a front buffer
Assume that the front-buffer exists even if the server didn't tell the
client that it exists.
Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
Reviewed-by: Kristian Høgsberg <krh@redhat.com>
Ian Romanick [Mon, 6 Apr 2009 20:15:54 +0000 (13:15 -0700)]
intel / DRI2: Track and flush front-buffer rendering
Track two flags: whether or not front-buffer rendering is currently
enabled and whether or not front-buffer rendering has been enabled
since the last glFlush. If the second flag is set, the front-buffer
is flushed via a loader call back. If the first flag is cleared, the
second flag is cleared at this time.
Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
Reviewed-by: Kristian Høgsberg <krh@redhat.com>
Ian Romanick [Mon, 6 Apr 2009 19:44:18 +0000 (12:44 -0700)]
DRI2: Provide an interface for drivers to flush front-buffer rendering
Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
Reviewed-by: Kristian Høgsberg <krh@redhat.com>
José Fonseca [Thu, 9 Apr 2009 21:10:45 +0000 (22:10 +0100)]
python: Fix bad pointer.
Brian Paul [Thu, 9 Apr 2009 21:00:54 +0000 (15:00 -0600)]
st: remove another unneeded 'is compressed' comparison
Brian Paul [Thu, 9 Apr 2009 20:58:40 +0000 (14:58 -0600)]
st: remove unneeded "is compressed" check
The format indicates compressed vs. uncompressed.
Brian Paul [Thu, 9 Apr 2009 20:57:41 +0000 (14:57 -0600)]
st: rearrange some code to be a little more clear
Brian Paul [Thu, 9 Apr 2009 20:55:32 +0000 (14:55 -0600)]
gallium: remove unneeded compressed=0 assignment
Brian Paul [Thu, 9 Apr 2009 20:55:19 +0000 (14:55 -0600)]
gallium: remove unneeded compressed=0 assignment
Brian Paul [Thu, 9 Apr 2009 20:48:21 +0000 (14:48 -0600)]
st: remove comp_byte parameter to st_texture_create()
We can determine if the texture is compressed by checking the format.
Brian Paul [Thu, 9 Apr 2009 20:41:49 +0000 (14:41 -0600)]
st: consolidate format->usage computation
Brian Paul [Thu, 9 Apr 2009 20:35:36 +0000 (14:35 -0600)]
st: add const qualifiers, use GL types
Brian Paul [Thu, 9 Apr 2009 20:32:42 +0000 (14:32 -0600)]
st: make loops over 3D texture slices a litte more intuitive
Brian Paul [Thu, 9 Apr 2009 20:27:36 +0000 (14:27 -0600)]
st: reformatting and clean-ups in texture code
Michal Krol [Thu, 9 Apr 2009 20:47:06 +0000 (22:47 +0200)]
tgsi/sse2: Fix ARL instruction.
Michal Krol [Thu, 9 Apr 2009 20:38:07 +0000 (22:38 +0200)]
tgsi/sse2: Fix LIT instruction.
José Fonseca [Thu, 9 Apr 2009 20:28:31 +0000 (21:28 +0100)]
trace: Dump the fence, not its pointer.
José Fonseca [Thu, 9 Apr 2009 20:27:43 +0000 (21:27 +0100)]
wgl: Remove unused member.
José Fonseca [Thu, 9 Apr 2009 20:27:03 +0000 (21:27 +0100)]
trace: Use 'flags' name consistently.
José Fonseca [Thu, 9 Apr 2009 20:26:15 +0000 (21:26 +0100)]
gallium: Use "flags" consistently.
José Fonseca [Thu, 9 Apr 2009 19:42:09 +0000 (20:42 +0100)]
gdi: Use stw_* names.
José Fonseca [Thu, 9 Apr 2009 19:41:59 +0000 (20:41 +0100)]
wgl: Use more stw_* names.
José Fonseca [Thu, 9 Apr 2009 19:35:55 +0000 (20:35 +0100)]
wgl: Use hooks instead of subclassing the window.
Subclassing the window is invasive: we might call an old window proc even
after it was removed. Glut and another bug just in the wrong time was
provoking this. Hooks don't have this problem.
Brian Paul [Thu, 9 Apr 2009 16:53:01 +0000 (10:53 -0600)]
mesa: fix potential recursive locking deadlock in _mesa_HashWalk()
If the walk callback called _mesa_HashRemove() we'd deadlock.
Brian Paul [Thu, 9 Apr 2009 16:33:08 +0000 (10:33 -0600)]
i965: free shader's constant buffer in brwDeleteProgram()
Fixes mem leak observed with texcombine test.
José Fonseca [Thu, 9 Apr 2009 14:15:12 +0000 (15:15 +0100)]
wgl: Catch the attempt of releasing a context which is not current.
José Fonseca [Thu, 9 Apr 2009 13:58:17 +0000 (14:58 +0100)]
wgl: Move the framebuffer list to the device. Avoid recursive locking.
José Fonseca [Thu, 9 Apr 2009 10:44:33 +0000 (11:44 +0100)]
wgl: More symbol naming normalization.
José Fonseca [Thu, 9 Apr 2009 10:28:23 +0000 (11:28 +0100)]
stw: Use wglext.h's defines and prototypes.
José Fonseca [Thu, 9 Apr 2009 10:09:42 +0000 (11:09 +0100)]
stw: Normalize symbols prefix.
José Fonseca [Thu, 9 Apr 2009 09:54:00 +0000 (10:54 +0100)]
wgl: Add const qualifier to global constant data.
Brian Paul [Thu, 9 Apr 2009 01:35:24 +0000 (19:35 -0600)]
i965: set BRW_MASK_DISABLE flag in "send" instruction in brw_dp_READ_4()
This fixes the random results that were seen when fetching a constant
inside an IF/ELSE clause. Disabling the execution mask ensures that all
the components of the register are written.
Brian Paul [Thu, 9 Apr 2009 01:31:49 +0000 (19:31 -0600)]
i965: clean-ups, debug code in brw_wm_glsl.c
Brian Paul [Thu, 9 Apr 2009 01:29:37 +0000 (19:29 -0600)]
i965: init current_const[i].index = -1
Brian Paul [Wed, 8 Apr 2009 19:48:34 +0000 (13:48 -0600)]
i965: move the fetch_constants() call before setting conditional mod state
Before, the instruction's CondUpdate field was mistakenly effecting the
constant-fetch operation.
Fixes progs/glsl/bump.c demo. But there are some other issues related
to condition flags and IF/ELSE that need investigation...
Brian Paul [Tue, 7 Apr 2009 17:15:27 +0000 (11:15 -0600)]
mesa: minor datatype changes in optimization code
Ben Skeggs [Thu, 9 Apr 2009 01:36:28 +0000 (11:36 +1000)]
nv50: adapt for new clear interface
this is so much nicer :)
Corbin Simpson [Wed, 8 Apr 2009 21:54:17 +0000 (14:54 -0700)]
r300-gallium: Properly emit indexbufs.
This fixes hardlocks with anything using elts.
Corbin Simpson [Tue, 7 Apr 2009 09:04:07 +0000 (02:04 -0700)]
r300-gallium: Add vertex shader constant emit.
José Fonseca [Wed, 8 Apr 2009 20:13:57 +0000 (21:13 +0100)]
python/retrace: Try to cope with failures creating textures.
José Fonseca [Wed, 8 Apr 2009 20:11:51 +0000 (21:11 +0100)]
python: Don't try to finish a null fence.
José Fonseca [Wed, 8 Apr 2009 20:11:24 +0000 (21:11 +0100)]
python/retrace: Use colors on windows console.
Robert Ellison [Wed, 8 Apr 2009 16:58:33 +0000 (10:58 -0600)]
Mesa: allow suppression of debug messages in a debug build
For testing, it's very useful to be able to test on a debug build,
while suppressing the debug messages (messages that are by default
suppressed in a release build), in order to see the same behavior
that users of release builds will see.
For example, the "piglit" test suite will flag an error on
programs that produce unexpected output, which means that a
debug build will always fail due to the extra debug messages.
This change introduces a new value to the MESA_DEBUG
environment variable. In a debug build, explicitly setting MESA_DEBUG
to "0" will suppress all debug messages (both from _mesa_debug() and
from _mesa_warning()). (The former behavior was that debug
messages were never suppressed in debug builds.)
Behavior of non-debug builds has not changed. In such a build,
_mesa_debug() messages are always suppressed, and _mesa_warning()
messages will be suppressed unless MESA_DEBUG is set *to any value*.
Alan Hourihane [Wed, 8 Apr 2009 15:39:17 +0000 (16:39 +0100)]
gallium: when using gl_PointCoord ensure we use the correct attribute.
José Fonseca [Wed, 8 Apr 2009 15:25:29 +0000 (16:25 +0100)]
python/retrace: Allow to specify the range of calls of interest.
José Fonseca [Wed, 8 Apr 2009 15:24:31 +0000 (16:24 +0100)]
trace: Make calls dumping threadsafe.
José Fonseca [Wed, 8 Apr 2009 14:41:08 +0000 (15:41 +0100)]
trace: Make call no global.
José Fonseca [Wed, 8 Apr 2009 14:36:17 +0000 (15:36 +0100)]
python/retrace: Handle recent traces.
Try to keep backwards compatability with old traces as much as possible.
José Fonseca [Wed, 8 Apr 2009 14:32:08 +0000 (15:32 +0100)]
python: Fix method renaming.
José Fonseca [Wed, 8 Apr 2009 14:30:53 +0000 (15:30 +0100)]
python: Cope with null surfaces.
José Fonseca [Wed, 8 Apr 2009 14:30:31 +0000 (15:30 +0100)]
wgl: Integrate the trace driver on debug builds.