Eric Anholt [Fri, 8 Oct 2004 23:03:38 +0000 (23:03 +0000)]
Add fallback debugging (R128_DEBUG=fall) output, and set DO_DEBUG=1 by default
so that we can use the env var to get output. Add a no_rast driconf option to
force software fallbacks.
Eric Anholt [Fri, 8 Oct 2004 22:21:09 +0000 (22:21 +0000)]
Don't compile WriteMonoDepthSpan if HAVE_HW_DEPTH_SPANS is defined -- it won't
be used in that case, and it wanting WRITE_DEPTH was making r128 die on
the undefined symbol.
Eric Anholt [Fri, 8 Oct 2004 21:37:08 +0000 (21:37 +0000)]
Add support for NV_blend_square, and print errors if an unsupported blend
function is used.
Ian Romanick [Fri, 8 Oct 2004 01:15:52 +0000 (01:15 +0000)]
Massively cleaned up the code that calculates front/back/depth buffer
pitch and size. Cut out a bunch of dead code.
This fixes bugzilla #1555.
Ian Romanick [Fri, 8 Oct 2004 01:03:10 +0000 (01:03 +0000)]
Reject unsupported texture formats passed to glCompressedTexImage?D. This
fixes Mesa bug #1028405.
Ian Romanick [Fri, 8 Oct 2004 00:56:08 +0000 (00:56 +0000)]
Add simple test program for Mesa bug #1028405.
Alex Deucher [Fri, 8 Oct 2004 00:31:51 +0000 (00:31 +0000)]
Apply Eric's stencil wrap patch (Eric Anholt)
Eric Anholt [Thu, 7 Oct 2004 23:30:29 +0000 (23:30 +0000)]
Add Roland Scheidegger's S3TC patch. This patch does not implement the
(patented) S3TC/DXTC algorithms, but adds an option to dlopen a library module
providing functions to do so. Because it uses dlopen, it is only enabled if
USE_EXTERNAL_DXTN_LIB=1 is defined (which is only in linux-dri config, so far).
It adds support for S3TC to several DRI drivers, and adds a DRI config option to
force enabling S3TC even if the software compression/decompression is
unavailable. This may allow people to use apps that require S3TC even though
they don't have a license to implement the patented material themselves, if
those apps use precompressed textures.
Ideally we would get permission from the current holder of the patents to
implement the algorithm in Mesa, at which point the dlopen mess could go away.
Until then, this allows some to run applications they couldn't otherwise, and
hopefully will provide us with more push to get the final step of getting that
permission done.
Ian Romanick [Thu, 7 Oct 2004 16:39:20 +0000 (16:39 +0000)]
Prevent Y-offset from exceeding valid range in texture upload code. This
fixes bugzilla #960.
Felix Kuehling [Thu, 7 Oct 2004 10:07:35 +0000 (10:07 +0000)]
Fix emitting fog without secondary color and vice-versa.
Karl Schultz [Wed, 6 Oct 2004 21:45:48 +0000 (21:45 +0000)]
*** empty log message ***
Brian Paul [Wed, 6 Oct 2004 15:56:38 +0000 (15:56 +0000)]
more GL_ARB_draw_buffers updates from Michale Krol
Brian Paul [Wed, 6 Oct 2004 15:52:43 +0000 (15:52 +0000)]
fix GetVertexAttrib problem
Ian Romanick [Wed, 6 Oct 2004 02:02:53 +0000 (02:02 +0000)]
Convert tdfxDDInitExtensions to use driInitExtensions.
Daniel Borca [Tue, 5 Oct 2004 08:43:22 +0000 (08:43 +0000)]
put Rush in the list of supported HW
Daniel Borca [Tue, 5 Oct 2004 08:42:10 +0000 (08:42 +0000)]
stencil wrap works, either HW or SW (Ian Romanick)
Daniel Borca [Tue, 5 Oct 2004 08:39:32 +0000 (08:39 +0000)]
enable 2_0 extensions
Ian Romanick [Mon, 4 Oct 2004 22:58:39 +0000 (22:58 +0000)]
Use the driFillInModes utility function.
Ian Romanick [Mon, 4 Oct 2004 22:23:29 +0000 (22:23 +0000)]
Also export fbconfigs with 0/0 depth/stencil modes. This fixes "driver
claims not to support visual 0xXX" warnings in X.org 6.8.1.
Ian Romanick [Mon, 4 Oct 2004 17:17:49 +0000 (17:17 +0000)]
Add some Savage3D stencil documentation based on discussions during the
20-Sep-2004 #dri-devel meeting.
Brian Paul [Mon, 4 Oct 2004 14:47:48 +0000 (14:47 +0000)]
make clean should remove depend.bak too (Sérgio Moneiro Basto)
Brian Paul [Mon, 4 Oct 2004 14:40:05 +0000 (14:40 +0000)]
ARB_fp support for GL_ARB_draw_buffers (Karl Rasche)
Ian Romanick [Mon, 4 Oct 2004 03:19:01 +0000 (03:19 +0000)]
Set MaxTextureImageUnits and MaxTextureCoordUnits to match MaxTextureUnits.
This fixes bugzilla #1511.
Ian Romanick [Mon, 4 Oct 2004 02:54:49 +0000 (02:54 +0000)]
The alpha post-scale and the RGB post-scale were mistakenly reversed.
Eric Anholt [Sun, 3 Oct 2004 18:09:57 +0000 (18:09 +0000)]
Remove empty r128_vb.h.
Brian Paul [Sat, 2 Oct 2004 22:48:59 +0000 (22:48 +0000)]
put glapi_x86.S into the x86/ directory
Brian Paul [Sat, 2 Oct 2004 22:47:48 +0000 (22:47 +0000)]
remove unused413 stuff, glDrawBuffersARB uses that slot now
Brian Paul [Sat, 2 Oct 2004 22:43:44 +0000 (22:43 +0000)]
GL_ARB_draw_buffers
Brian Paul [Sat, 2 Oct 2004 22:38:43 +0000 (22:38 +0000)]
fix _DrawDestMask[0][0] typos
Brian Paul [Sat, 2 Oct 2004 16:39:09 +0000 (16:39 +0000)]
added support for GL_ARB_draw_buffers
Brian Paul [Sat, 2 Oct 2004 15:56:50 +0000 (15:56 +0000)]
fix indentation
Brian Paul [Sat, 2 Oct 2004 15:56:33 +0000 (15:56 +0000)]
fix a comment
Brian Paul [Sat, 2 Oct 2004 15:43:26 +0000 (15:43 +0000)]
bump version to 6.3
Brian Paul [Sat, 2 Oct 2004 15:43:14 +0000 (15:43 +0000)]
initial stuff for 6.3
Brian Paul [Sat, 2 Oct 2004 15:26:25 +0000 (15:26 +0000)]
a few more updates for the 6.2 release
Brian Paul [Sat, 2 Oct 2004 15:16:59 +0000 (15:16 +0000)]
fix LoadProgramNV regression when I had fixed the RefCount bug
Brian Paul [Sat, 2 Oct 2004 14:59:29 +0000 (14:59 +0000)]
added PBO known issues
Brian Paul [Sat, 2 Oct 2004 14:56:17 +0000 (14:56 +0000)]
set 6.2 release date
Eric Anholt [Sat, 2 Oct 2004 05:22:19 +0000 (05:22 +0000)]
If an application cleared before any state had been emitted, that clear would
happen before any state had been set, causing a hang later on. Fix this by
calling r200Flush instead of FIREVERTICES (which checks if any state has been
emitted but not flushed, before calling Flush) in r200Clear. While here, add
some more debugging info which was useful, and remove an unnecessary
save/restore in BackUpAndEmit.
Eric Anholt [Sat, 2 Oct 2004 01:33:46 +0000 (01:33 +0000)]
Convert Rage 128 over to t_vertex.c. While it's slightly slower (10% in ipers)
it's not that big of a deal in more normal apps, and axes a good bit of code.
And I assume that t_vertex will only get faster. Removes ~43k from compiled
binary.
Tested with: quake3, ut, ipers, texcyl, chromium, tuxracer, neverball (kinda)
Eric Anholt [Sat, 2 Oct 2004 01:28:38 +0000 (01:28 +0000)]
Add an option for vertices emitted to be swapped CPU_TO_LE32, to be used by the
r128 code.
Ian Romanick [Sat, 2 Oct 2004 01:23:46 +0000 (01:23 +0000)]
Cut out all of the old texture environment code and rewrote it from
scratch. There were just too many cut-and-paste errors in the code,
and it was too hard to follow.
This fixes Mesa bugzilla #1508. #1509 is probably also fixed, but
I'd like to do some more testing before I close that particular bug.
Additionally, this cuts almost 4,600 lines of code from the driver.
Karl Schultz [Fri, 1 Oct 2004 21:22:10 +0000 (21:22 +0000)]
minor updates and improvements
Karl Schultz [Fri, 1 Oct 2004 20:55:23 +0000 (20:55 +0000)]
Add the glut project to the demos workspace.
Karl Schultz [Fri, 1 Oct 2004 20:53:28 +0000 (20:53 +0000)]
Remove glut project from the Mesa workspace. This allows users to build
this entire workspace by only unzipping MesaLib.zip. The project will
be moved to the demos workspace, which is contained in the MesaDemos zip
file.
Karl Schultz [Fri, 1 Oct 2004 20:19:16 +0000 (20:19 +0000)]
Add HP shared lib configs.
Karl Schultz [Fri, 1 Oct 2004 20:18:42 +0000 (20:18 +0000)]
HP shared lib config
Ian Romanick [Fri, 1 Oct 2004 18:32:25 +0000 (18:32 +0000)]
Replace the numerous switch-statements to derive GL_COMBINE post-scale
state with a function. This function sets the bits correctly. Did some
trivial refactoring on some of the GL_COMBINE code.
First pass at replacing classic texture environments with GL_COMBINE.
This is controlled by the EXPERIMENTAL_COMBINE_MODE define.
Karl Schultz [Fri, 1 Oct 2004 16:29:39 +0000 (16:29 +0000)]
fix problems in previous version
Karl Schultz [Fri, 1 Oct 2004 13:33:26 +0000 (13:33 +0000)]
Enable builds on AIX 64-bit
Dave Airlie [Fri, 1 Oct 2004 04:34:01 +0000 (04:34 +0000)]
Fix mesa solo with doublebuffered apps, the drawable index/stamp weren't
being updated properly..
Ian Romanick [Thu, 30 Sep 2004 21:18:23 +0000 (21:18 +0000)]
Factored out code for setting texture coordinate wrap bits. Added support
for GL_ARB_texture_mirrored_repeat. Enabled GL_NV_blend_square. It has
always actually been supported. Removed redundant EXT versions of ARB
extension strings.
Brian Paul [Thu, 30 Sep 2004 15:39:43 +0000 (15:39 +0000)]
set span.y prior to each span write in draw_depth_pixels, like Karl's other fix
Eric Anholt [Thu, 30 Sep 2004 00:40:21 +0000 (00:40 +0000)]
Bugzilla #1058: Fix some potential 64bit pointer issues by storing differences
between pointers in appropriate types.
Submitted by: Ronny V. Vindenes <s864@ii.uib.no>
Eric Anholt [Thu, 30 Sep 2004 00:08:05 +0000 (00:08 +0000)]
OK, one more time. Simplify the state-backup system by just storing the full
state in a ready-to-emit cmdbuf, which avoids the issue Nicolai Haehnle reported
where the check() could return differently during backup-and-emit than it should
have if it were called at the right time. Move the lit emission before most of
the TCL state emission on r200, which fixes neverball issues.
Tested with: r100/r200 with neverball, tuxracer, chromium, quake3, ipers
Karl Schultz [Tue, 28 Sep 2004 18:53:44 +0000 (18:53 +0000)]
Fix bug in glDrawPixels when writing color indexed images on color indexed
devices. In draw_index_pixels(), use the correct variables for the span Y
and span end values.
Nicolai Haehnle [Tue, 28 Sep 2004 10:59:23 +0000 (10:59 +0000)]
Initial revision
Daniel Borca [Tue, 28 Sep 2004 06:44:46 +0000 (06:44 +0000)]
use GLuint instead of uint
Brian Paul [Mon, 27 Sep 2004 22:40:31 +0000 (22:40 +0000)]
version bump
Brian Paul [Mon, 27 Sep 2004 22:40:22 +0000 (22:40 +0000)]
change include of glx_mangle.h to match other headers
Brian Paul [Mon, 27 Sep 2004 22:38:47 +0000 (22:38 +0000)]
remove beos/*.h files from tarball list
Brian Paul [Mon, 27 Sep 2004 22:38:25 +0000 (22:38 +0000)]
bump minor version to 2
Brian Paul [Mon, 27 Sep 2004 16:19:17 +0000 (16:19 +0000)]
VBO RefCount fix (David Reveman)
Daniel Borca [Mon, 27 Sep 2004 07:23:20 +0000 (07:23 +0000)]
stencil wrap works, either HW or SW (Ian Romanick)
Eric Anholt [Sat, 25 Sep 2004 21:08:56 +0000 (21:08 +0000)]
Bump driver_date for today's fixes.
Eric Anholt [Sat, 25 Sep 2004 21:08:30 +0000 (21:08 +0000)]
Can't bail in sisDDLogicOpCode when logicop is off, because it's called with
GL_COPY to turn off logicop. Fixes glean's logicop test.
Eric Anholt [Sat, 25 Sep 2004 21:06:58 +0000 (21:06 +0000)]
Triangle stipple is a fallback, since we don't have code for the stippling.
Doesn't seem to help with glean's paths test, but I'm pretty sure it's correct.
Eric Anholt [Sat, 25 Sep 2004 19:43:30 +0000 (19:43 +0000)]
Correct a couple of comments.
Eric Anholt [Sat, 25 Sep 2004 19:01:18 +0000 (19:01 +0000)]
Fix SiS AGP vertex dispatch by not trying to emit 0 vertices, which would hang
the hardware. Re-enable AGP by default.
Eric Anholt [Sat, 25 Sep 2004 16:56:58 +0000 (16:56 +0000)]
Add sis to the list of DRI drivers, since it had already been converted to the
new interface.
Eric Anholt [Sat, 25 Sep 2004 07:11:12 +0000 (07:11 +0000)]
Refactor the linux-dri config files so that linux-dri-x86 inherits from
linux-dri. Turn on -O by default (-O2 provides a slight improvement in
performance, at a large cost to debuggability). Turn off -fPIC on x86 by
default. Turn on -Wall, to catch more stupid mistakes. This could be both done
cleaner, and done for more config files, but it works for me for now.
Eric Anholt [Sat, 25 Sep 2004 07:00:15 +0000 (07:00 +0000)]
The previous code would emit a full set of state during the first EmitState on
a new cmdbuf, to ensure that state wasn't lost across UNLOCK/LOCK pairs (in the
case of context switching). This was rather inefficient. Instead, after
flushing a cmdbuf, mark the state as needing to be saved on unlock. Then, at
the beginning of flushing a cmdbuf, if we actually have lost the context, go
back and emit a new cmdbuf with the full set of state, before continuing with
the cmdbuf flush. Provides a 10-15% improvement in ipers performance in my
tests, along with other apps.
Tested with: ipers, glxgears, quake3
Eric Anholt [Sat, 25 Sep 2004 06:12:40 +0000 (06:12 +0000)]
Fix texturing in quake3. Some code was left over from the dirty/clean list
setup that now removed atoms from the atomlist on texture deletion.
Eric Anholt [Sat, 25 Sep 2004 03:00:27 +0000 (03:00 +0000)]
Bump DRIVER_DATE for yesterday's changes.
Brian Paul [Fri, 24 Sep 2004 23:01:21 +0000 (23:01 +0000)]
moved to demos/ directory
Brian Paul [Fri, 24 Sep 2004 23:00:52 +0000 (23:00 +0000)]
moved in from the tests/ directory to include in distro
Brian Paul [Fri, 24 Sep 2004 20:31:02 +0000 (20:31 +0000)]
updated osmesa info
Brian Paul [Fri, 24 Sep 2004 14:30:13 +0000 (14:30 +0000)]
patches for clearing hw depth buffers from software fallback (Nicolai Haehnle)
Daniel Borca [Fri, 24 Sep 2004 07:03:53 +0000 (07:03 +0000)]
added (back!) option to use non-packedcolor
Eric Anholt [Fri, 24 Sep 2004 04:20:58 +0000 (04:20 +0000)]
-O -Wall warnings cleanups in r200.
Eric Anholt [Fri, 24 Sep 2004 03:09:49 +0000 (03:09 +0000)]
Clean up some warnings in the R100 driver with -O -Wall.
Eric Anholt [Fri, 24 Sep 2004 01:33:15 +0000 (01:33 +0000)]
Remove an unnecessary calculation of the dest pointer.
Brian Paul [Thu, 23 Sep 2004 17:34:52 +0000 (17:34 +0000)]
don't test for NULL pixels pointer here, do that in the 'store' routines after validating PBO address
Brian Paul [Thu, 23 Sep 2004 17:32:12 +0000 (17:32 +0000)]
removed hacked GL tokens, fixed a printf
Brian Paul [Thu, 23 Sep 2004 17:31:21 +0000 (17:31 +0000)]
added a comment in BindTexture
Brian Paul [Thu, 23 Sep 2004 17:30:54 +0000 (17:30 +0000)]
just some comments and formatting changes
Dave Airlie [Thu, 23 Sep 2004 08:24:03 +0000 (08:24 +0000)]
fix for Erics new emit state code, rework out the dest pointer
after we check the buffer
Eric Anholt [Wed, 22 Sep 2004 06:27:02 +0000 (06:27 +0000)]
The previous code would emit a full set of state during the first EmitState on
a new cmdbuf, to ensure that state wasn't lost across UNLOCK/LOCK pairs (in the
case of context switching). This was rather inefficient. Instead, after
flushing a cmdbuf, mark the state as needing to be saved on UNLOCK. Then, at
the beginning of flushing a cmdbuf, if we actually have lost the context, go
back and emit a new cmdbuf with the full set of state, before continuing with
the cmdbuf flush. Also, remove the dirty/clean atom lists, since atoms are
emitted in a fixed order these days, and go with a simpler single list.
Provides a 14% improvement in ipers performance in my tests, along with other
apps.
Dave Airlie [Tue, 21 Sep 2004 00:39:59 +0000 (00:39 +0000)]
Add xmlconfig to the i810 makefile
Daniel Borca [Mon, 20 Sep 2004 08:33:38 +0000 (08:33 +0000)]
clamp size of wide points
Ian Romanick [Sun, 19 Sep 2004 08:03:46 +0000 (08:03 +0000)]
Add GL_ARB_texture_cube_map support for i830. Most of the code was
lifted from the i915 side. i830 will now report version 1.3! Hurrah!
With the exception of GL_EXT_texture_compression_s3tc, the i830 driver
now supports all the extensions that its Windows counterpart supports.
Ian Romanick [Sun, 19 Sep 2004 07:15:38 +0000 (07:15 +0000)]
Added GL_ARB_point_parameters support for i830.
Ian Romanick [Sun, 19 Sep 2004 07:01:26 +0000 (07:01 +0000)]
Merge in all the i830 functional differences from the old i830 driver.
Karl Schultz [Fri, 17 Sep 2004 22:02:05 +0000 (22:02 +0000)]
rearrange a couple of lines of code to avoid compilation error in VC 6,
probably due to a compiler bug.
Eric Anholt [Fri, 17 Sep 2004 05:10:33 +0000 (05:10 +0000)]
Symbol names are prepended with an underscore on CYGWIN as well.
X.Org Bugzilla: 1079
Submitted by: Alexander Gottwald <ago@freedesktop.org>
Brian Paul [Thu, 16 Sep 2004 19:40:22 +0000 (19:40 +0000)]
silence warnings
Brian Paul [Thu, 16 Sep 2004 19:39:04 +0000 (19:39 +0000)]
silence warnings, re-indent code
Brian Paul [Thu, 16 Sep 2004 19:33:44 +0000 (19:33 +0000)]
VBO refcount and spriteblast fixes
Brian Paul [Thu, 16 Sep 2004 19:32:49 +0000 (19:32 +0000)]
clean-up the hokey transformation code so that window resizes actually work
Brian Paul [Wed, 15 Sep 2004 14:49:33 +0000 (14:49 +0000)]
replace glNormal3f(v) with GL_CALL(Normal3f)(v), etc (Andreas Stenglein