Brian Paul [Tue, 20 Sep 2005 15:51:44 +0000 (15:51 +0000)]
initial check-in
Brian Paul [Tue, 20 Sep 2005 04:47:34 +0000 (04:47 +0000)]
remove 31-bit Z buffer work-around
Brian Paul [Tue, 20 Sep 2005 04:47:04 +0000 (04:47 +0000)]
change mask[] test in put_mono_row_uint()
Brian Paul [Tue, 20 Sep 2005 04:45:26 +0000 (04:45 +0000)]
compute z[i] as GLuint
Brian Paul [Tue, 20 Sep 2005 04:36:50 +0000 (04:36 +0000)]
some improvements to _swrast_clear_depth_buffer()
Brian Paul [Tue, 20 Sep 2005 04:35:40 +0000 (04:35 +0000)]
remove unneeded sizeof(GLdepth) test
Brian Paul [Tue, 20 Sep 2005 04:34:29 +0000 (04:34 +0000)]
Change zLeft and zval to GLuint. Seems to fix 32-bit Z computation
Brian Paul [Tue, 20 Sep 2005 04:31:24 +0000 (04:31 +0000)]
replace STENCIL_BITS with stencilMax value
Brian Paul [Tue, 20 Sep 2005 04:25:03 +0000 (04:25 +0000)]
remove STENCIL_MAX
Brian Paul [Tue, 20 Sep 2005 02:44:40 +0000 (02:44 +0000)]
change location of ffs() code for windows
Brian Paul [Mon, 19 Sep 2005 21:16:29 +0000 (21:16 +0000)]
init a=0
Brian Paul [Mon, 19 Sep 2005 20:32:21 +0000 (20:32 +0000)]
s/GLchan/GLubyte/, fix error string, add comment
Brian Paul [Mon, 19 Sep 2005 20:29:34 +0000 (20:29 +0000)]
updated comment
Brian Paul [Mon, 19 Sep 2005 20:12:32 +0000 (20:12 +0000)]
additional wrapper updates, bug 4468
Brian Paul [Mon, 19 Sep 2005 17:17:58 +0000 (17:17 +0000)]
s/unsigned long/drm_handle_t/ to get compilation of fb driver to work
Brian Paul [Mon, 19 Sep 2005 17:17:18 +0000 (17:17 +0000)]
compilation fixes, don't know if runtime works
Brian Paul [Fri, 16 Sep 2005 21:15:27 +0000 (21:15 +0000)]
Fix potential segfault when trying to read pixels outside renderbuffer bounds.
Use _swrast_get_values() which does clipping.
Brian Paul [Fri, 16 Sep 2005 21:10:27 +0000 (21:10 +0000)]
remove unused _swrast_read_depth_span()
Brian Paul [Fri, 16 Sep 2005 21:07:48 +0000 (21:07 +0000)]
lots of little clean-ups, improvements
Brian Paul [Fri, 16 Sep 2005 21:06:57 +0000 (21:06 +0000)]
change some comments
Brian Paul [Fri, 16 Sep 2005 21:06:33 +0000 (21:06 +0000)]
pass mask=NULL to write all pixels in clear_ci_buffer_with_masking()
Brian Paul [Fri, 16 Sep 2005 18:18:47 +0000 (18:18 +0000)]
remove unused ctx var
Brian Paul [Fri, 16 Sep 2005 18:14:24 +0000 (18:14 +0000)]
use mesa import wrappers, bug 4468
Brian Paul [Fri, 16 Sep 2005 04:18:09 +0000 (04:18 +0000)]
alphabetize
Brian Paul [Fri, 16 Sep 2005 04:16:48 +0000 (04:16 +0000)]
Rework the texture filtering functions a bit.
No need to pass the texture unit number as an argument.
Brian Paul [Fri, 16 Sep 2005 01:42:07 +0000 (01:42 +0000)]
replace a couple macros with inlined functions
Brian Paul [Fri, 16 Sep 2005 01:41:03 +0000 (01:41 +0000)]
more GLuint->GLbitfield changes
Brian Paul [Thu, 15 Sep 2005 05:00:45 +0000 (05:00 +0000)]
Replace GLuint with GLbitfield where appropriate. Also replace GLuint
with GLboolean in a few places.
Brian Paul [Thu, 15 Sep 2005 03:58:52 +0000 (03:58 +0000)]
Code clean-ups, reorg.
Brian Paul [Thu, 15 Sep 2005 01:58:39 +0000 (01:58 +0000)]
silence warning
Brian Paul [Thu, 15 Sep 2005 01:55:40 +0000 (01:55 +0000)]
Remove a bunch texel fetch functions that can be handled by the new
adaptor functions.
Brian Paul [Thu, 15 Sep 2005 01:08:24 +0000 (01:08 +0000)]
updated comments
Brian Paul [Thu, 15 Sep 2005 01:06:19 +0000 (01:06 +0000)]
New set_fetch_functions() to set a gl_texture_image's FetchTexel functions
from the gl_texture_format's functions.
Added "adaptor" functions to allow sampling a float-valued texture with
GLchans and vice versa. This will allow trimming down the number of
texture fetch routines.
Brian Paul [Thu, 15 Sep 2005 01:00:20 +0000 (01:00 +0000)]
replace s_texture.[ch] with s_texcombine.[ch] and s_texfilter.[ch]
Brian Paul [Thu, 15 Sep 2005 00:58:03 +0000 (00:58 +0000)]
Update includes post-splitting of s_texture.c
Brian Paul [Thu, 15 Sep 2005 00:57:00 +0000 (00:57 +0000)]
Split the s_texture.c file into two new files:
s_texcombine.c - for texture combining/application
s_texfilter.c - for texture sampling/filtering
Brian Paul [Wed, 14 Sep 2005 18:21:05 +0000 (18:21 +0000)]
remove unused vars
Brian Paul [Wed, 14 Sep 2005 18:09:15 +0000 (18:09 +0000)]
remove _tnl_MakeCurrent call
Brian Paul [Wed, 14 Sep 2005 14:34:54 +0000 (14:34 +0000)]
initial 6.5 release notes
Brian Paul [Wed, 14 Sep 2005 14:32:20 +0000 (14:32 +0000)]
Remove _tnl_MakeCurrent() and the unused ctx->Driver.MakeCurrent() callback.
Dave Airlie [Wed, 14 Sep 2005 12:08:32 +0000 (12:08 +0000)]
fix breakage from Brians changes earlier
Brian Paul [Wed, 14 Sep 2005 03:11:36 +0000 (03:11 +0000)]
clean-up some _mesa_make_current() code
Brian Paul [Wed, 14 Sep 2005 02:18:43 +0000 (02:18 +0000)]
remove rm -f line
Brian Paul [Wed, 14 Sep 2005 02:18:09 +0000 (02:18 +0000)]
remove .a file before creating new one
Brian Paul [Wed, 14 Sep 2005 00:49:09 +0000 (00:49 +0000)]
pageflip change from radeon driver
Brian Paul [Wed, 14 Sep 2005 00:42:33 +0000 (00:42 +0000)]
Instead of calling _mesa_ResizeBuffersMESA() in the Viewport function,
call driUpdateFramebufferSize() when window size/position changes.
Brian Paul [Wed, 14 Sep 2005 00:37:10 +0000 (00:37 +0000)]
make some funcs static
Brian Paul [Wed, 14 Sep 2005 00:36:27 +0000 (00:36 +0000)]
Instead of calling _mesa_ResizeBuffersMESA() in the Viewport function,
call driUpdateFramebufferSize() when window size/position changes.
Brian Paul [Wed, 14 Sep 2005 00:33:03 +0000 (00:33 +0000)]
use mklib script to make drivers instead of gcc -shared
Brian Paul [Wed, 14 Sep 2005 00:11:30 +0000 (00:11 +0000)]
updated comment
Brian Paul [Wed, 14 Sep 2005 00:10:56 +0000 (00:10 +0000)]
make some functions static
Brian Paul [Wed, 14 Sep 2005 00:10:27 +0000 (00:10 +0000)]
Instead of calling _mesa_ResizeBuffersMESA() in the Viewport function,
call driUpdateFramebufferSize() when we detect changes in window size/
position in radeonGetLock().
Brian Paul [Wed, 14 Sep 2005 00:09:09 +0000 (00:09 +0000)]
added driUpdateFramebufferSize()
Brian Paul [Tue, 13 Sep 2005 23:37:50 +0000 (23:37 +0000)]
Move _mesa_add_soft_renderbuffers() from framebuffer.c to renderbuffer.c
Update some comments, code, etc.
Brian Paul [Tue, 13 Sep 2005 18:18:47 +0000 (18:18 +0000)]
update like for MD2 viewer
Brian Paul [Tue, 13 Sep 2005 15:04:30 +0000 (15:04 +0000)]
replace semicolon with &&
Brian Paul [Tue, 13 Sep 2005 14:48:28 +0000 (14:48 +0000)]
print version with debug info
Brian Paul [Tue, 13 Sep 2005 13:36:47 +0000 (13:36 +0000)]
fix FreeBSD problem (bug 4435)
Eric Anholt [Tue, 13 Sep 2005 05:19:02 +0000 (05:19 +0000)]
Fix libGL build on FreeBSD by respecting ASM_FLAGS.
Submitted by: Diego Pettenò <flameeyes@gentoo.org>
Brian Paul [Tue, 13 Sep 2005 04:42:09 +0000 (04:42 +0000)]
Replace ctx->Driver.StencilOp/Func/Mask() functions with
ctx->Driver.Stencil*Separate() functions.
Brian Paul [Tue, 13 Sep 2005 02:59:53 +0000 (02:59 +0000)]
OpenGL 2.0's two-sided stencil feature wasn't implemented correctly.
See comment near top of stencil.c for info about OpenGL 2.0 vs.
GL_EXT_stencil_two_side.
Brian Paul [Tue, 13 Sep 2005 02:57:07 +0000 (02:57 +0000)]
just some reformatting
Brian Paul [Tue, 13 Sep 2005 02:03:06 +0000 (02:03 +0000)]
Clear either 8 or 16-bit stencil buffers.
Brian Paul [Tue, 13 Sep 2005 02:02:03 +0000 (02:02 +0000)]
Use GLuint/GLint intead of GLstencil for stencil state.
Brian Paul [Tue, 13 Sep 2005 01:19:29 +0000 (01:19 +0000)]
Move the ctx->FirstTimeCurrent initialization
Brian Paul [Tue, 13 Sep 2005 01:17:01 +0000 (01:17 +0000)]
init StencilOpSeparate to NULL
Roland Scheidegger [Mon, 12 Sep 2005 23:45:05 +0000 (23:45 +0000)]
add GL_ATI_fragment_shader support for r200
Brian Paul [Mon, 12 Sep 2005 23:18:21 +0000 (23:18 +0000)]
added a return upon out-of-memory, remove a needless assignment
Brian Paul [Mon, 12 Sep 2005 21:41:45 +0000 (21:41 +0000)]
restore brace lost in previous check-in
Roland Scheidegger [Mon, 12 Sep 2005 21:20:10 +0000 (21:20 +0000)]
add complete support for ATI_fragment_shader for the r200. Most of the new code is in the new file r200_fragshader.c. Reverse engeneered by Dave Airlie and me
Roland Scheidegger [Mon, 12 Sep 2005 17:21:40 +0000 (17:21 +0000)]
remove unnecessary Swizzle parameter from struct (forgotten when the code was refactored)
Brian Paul [Mon, 12 Sep 2005 15:19:19 +0000 (15:19 +0000)]
use ASM_API variable, bug 4415
Brian Paul [Mon, 12 Sep 2005 15:05:57 +0000 (15:05 +0000)]
make lookup_query_object() static
Brian Paul [Mon, 12 Sep 2005 13:59:29 +0000 (13:59 +0000)]
remove rendundant r,g,b,a vars (bug 4331)
Dave Airlie [Sun, 11 Sep 2005 04:02:41 +0000 (04:02 +0000)]
add 1002:5460 M22 X300 card
Dave Airlie [Sun, 11 Sep 2005 03:58:29 +0000 (03:58 +0000)]
fix dri drivers build
Brian Paul [Sun, 11 Sep 2005 01:45:04 +0000 (01:45 +0000)]
Silence some warnings.
Brian Paul [Sat, 10 Sep 2005 16:57:22 +0000 (16:57 +0000)]
bunch of updates
Brian Paul [Sat, 10 Sep 2005 16:54:05 +0000 (16:54 +0000)]
put GLUT sources into a separate tarball
Brian Paul [Sat, 10 Sep 2005 16:41:17 +0000 (16:41 +0000)]
Clean-up of sources/Makefile, bug 4111
Brian Paul [Sat, 10 Sep 2005 16:27:51 +0000 (16:27 +0000)]
fix a problem w/ color index mode when CHAN_BITS=16
Brian Paul [Sat, 10 Sep 2005 16:20:47 +0000 (16:20 +0000)]
make some vars static
Brian Paul [Sat, 10 Sep 2005 16:19:36 +0000 (16:19 +0000)]
fix GLubyte/GLchan inconsistencies (bug 4331)
Roland Scheidegger [Sat, 10 Sep 2005 01:02:25 +0000 (01:02 +0000)]
new demo to test ATI_fragment_shader, not very creative but easy to know if the output is correct. Tests single-pass as well as multi-pass shader (only pseudo-dependant texture read though, and no tex coord swizzling), src and dst modifiers, src repetition, dst masks, constants, and some ops. Both shaders run succesfully with swrast as well as r200.
Roland Scheidegger [Sat, 10 Sep 2005 00:31:11 +0000 (00:31 +0000)]
add the new drm packets needed for ATI_fragment_shader to the sanity code of the r200 and radeon driver
Roland Scheidegger [Sat, 10 Sep 2005 00:11:27 +0000 (00:11 +0000)]
add reverse-engeneered register bits needed for ATI_fragment_shader (Dave Airlie, me)
Brian Paul [Sat, 10 Sep 2005 00:10:15 +0000 (00:10 +0000)]
Add offset and pitch fields to intelRegion. Remove frontOffset, frontPitch,
backOffset, backPitch, depthOffset, depthPitch, textureOffset fields.
Roland Scheidegger [Fri, 9 Sep 2005 23:31:31 +0000 (23:31 +0000)]
add _tnl_arb_vertex_program_stage to all drivers which already have _tnl_vertex_program_stage. Fixes crashes when using vertex programs. Tested on r200 only (fix by Keith Whitwell)
Brian Paul [Fri, 9 Sep 2005 22:57:05 +0000 (22:57 +0000)]
remove the redundant textureSize field, use tex.size instead, it's always the same value
Brian Paul [Fri, 9 Sep 2005 16:40:47 +0000 (16:40 +0000)]
don't generate selection hits for glBitmap, glDraw/CopyPixels. Misc clean-ups
Roland Scheidegger [Fri, 9 Sep 2005 15:43:11 +0000 (15:43 +0000)]
fix some bugs with handling of GL_DOT4_ATI, GL_DOT3_ATI and GL_DOT2_ADD_ATI ATI_fs instructions.
Roland Scheidegger [Fri, 9 Sep 2005 12:51:38 +0000 (12:51 +0000)]
add missing fallback string for ARB_vertex_program
Roland Scheidegger [Fri, 9 Sep 2005 01:38:21 +0000 (01:38 +0000)]
Fix wrong fallback bit being set by the tcl fallback caused by ARB_vertex_program. Disable ARB_vertex_program by default - it is incredibly slow, and just plain doesn't work anyway (arbvptorus fails an assertion, arbvpmesh has bogus output)
Keith Whitwell [Thu, 8 Sep 2005 21:40:10 +0000 (21:40 +0000)]
use an explicit reserved field, rather than magic number in bound
Brian Paul [Thu, 8 Sep 2005 18:47:36 +0000 (18:47 +0000)]
Fix the comment for the 'bound' field: it's a bitmask, not an index.
In driTexturesGone(), set bound = 0, not 99.
Keith Whitwell [Thu, 8 Sep 2005 18:45:03 +0000 (18:45 +0000)]
Keep NumTemporaries uptodate.
Make sure oC alpha is populated when separate_specular enabled.
Keith Whitwell [Thu, 8 Sep 2005 18:36:39 +0000 (18:36 +0000)]
Treat NegateBase as a bitfield, not a boolean.
Keith Whitwell [Thu, 8 Sep 2005 18:35:48 +0000 (18:35 +0000)]
be consistent in populating NegateBase field
Brian Paul [Thu, 8 Sep 2005 15:28:45 +0000 (15:28 +0000)]
check for either GL_ARB_depth_texture or GL_SGIX_depth_texture in a few places
Brian Paul [Thu, 8 Sep 2005 13:55:05 +0000 (13:55 +0000)]
fix precision problem in depth clearing (bug 4395)