Clone of mesa.
You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.

RELNOTES-5.1 9.7KB

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566676869707172737475767778798081828384858687888990919293949596979899100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130131132133134135136137138139140141142143144145146147148149150151152153154155156157158159160161162163164165166167168169170171172173174175176177178179180181182183184185186187188189190191192193194195196197198199200201202203204205206207208209210211212213214215216217218219220221222223224225226227228229230231232233234235236237238239240241242243244245246247248249250251252253254255256257258259260261262263264265266267268269270271272273274275276277278279
  1. Mesa 5.1 release notes
  2. December 17, 2003
  3. PLEASE READ!!!!
  4. Introduction
  5. ------------
  6. Mesa uses an even/odd version number scheme like the Linux kernel.
  7. Even-numbered versions (such as 5.0) designate stable releases.
  8. Odd-numbered versions (such as 5.1) designate new developmental releases.
  9. Bug fixes
  10. ---------
  11. See the VERSIONS file for a list of bugs fixed in this release.
  12. New Features in Mesa 5.1
  13. ------------------------
  14. GL_ARB_vertex_program / GL_ARB_fragment_program
  15. Michal Krol and Karl Rasche implemented these extensions. Thanks!
  16. Be aware that there may be some rough edges and lurking bugs.
  17. GL_ATI_texture_env_combine3 extension
  18. This adds a few new texture combine modes.
  19. Contributed by Ian Romanick.
  20. GL_SGI_texture_color_table
  21. Adds a color table lookup to the RGBA texture path. There's a separate
  22. color table for each texture unit.
  23. Contributed by Eric Plante.
  24. GL_NV_fragment_program
  25. NVIDIA's fragment-level programming feature.
  26. Possible lurking bugs:
  27. - the DDX and DDY commands aren't fully tested
  28. - there may be bugs in the parser
  29. - the TEX and TXP instructions both do perspective correction
  30. - the pack/unpack instructions may not be correct
  31. GL_EXT_depth_bounds_test
  32. This extension adds a scissor-like test for the Z axis. It's used to
  33. optimize stencil-volume shadow algorithms.
  34. GL_NV_light_max_exponent
  35. Lifts the 128 limit for max light exponent.
  36. GL_EXT_texture_rectangle
  37. Identical to GL_NV_texture_rectangle
  38. GL_ARB_occlusion_query
  39. Useful for visibility-based culling.
  40. GL_ARB_texture_non_power_of_two
  41. Removes the restriction that texture dimensions must be powers of two.
  42. GL_ARB_vertex_buffer_object
  43. Allows server-side vertex arrays, optimized host/card data transfers, etc.
  44. GL_ARB_point_sprite
  45. ARB-approved version of GL_NV_point_sprite. Basically allows textures
  46. to be applied to points.
  47. GL_IBM_multimode_draw_arrays
  48. Allows multiple vertex arrays to be drawn with one call, including arrays
  49. of different types of primitives.
  50. GL_SUN_multi_draw_arrays
  51. An alias for GL_EXT_multi_draw_arrays, standard in OpenGL 1.4.
  52. Faster glDrawPixels / glCopyPixels in X11 driver
  53. If your X screen is 32bpp, glDrawPixels to the front color buffer will
  54. be accelerated (via XPutImage()) if the image format is GL_BGRA and the
  55. type is GL_UNSIGNED_BYTE. No raster operations, such as depth test,
  56. blend, fog, etc. can be enabled.
  57. If your X screen is 16bpp, glDrawPixels to the front color buffer will
  58. be accelerated (via XPutImage()) if the image format is GL_RGB and the
  59. type is GL_UNSIGNED_SHORT_5_6_5. No raster operations, such as depth
  60. test, blend, fog, etc. can be enabled.
  61. glCopyPixels() calls for the front color buffer will be accelerated
  62. (via XCopyArea()) if no raster operations, such as depth test, blend,
  63. fog, pixel zoom, etc. are enabled.
  64. The speed-up over typical software rendering is a factor of 10 for
  65. glDrawPixels and 100 for glCopyPixels.
  66. With the addition of GL_ARB_occlusion_query, GL_ARB_vertex_buffer_object,
  67. GL_ARB_texture_non_power_of_two and GL_EXT_shadow_funcs, Mesa 5.1 supports
  68. all the new features of OpenGL 1.5. Mesa 6.0 (the next stable release)
  69. will advertise GL_VERSION = "1.5".
  70. Vertex/Fragment program debugger
  71. --------------------------------
  72. GL_MESA_program_debug is an experimental extension to support
  73. interactive debugging of vertex and fragment programs. See the
  74. docs/MESA_program_debug.spec file for details.
  75. The bulk of the vertex/fragment program debugger is implemented
  76. outside of Mesa. The GL_MESA_program_debug extension just has minimal
  77. hooks for stopping running programs and inspecting programs.
  78. The progs/tests/debugger.c (only in CVS) program is an example of how
  79. the extension can be used. Presently, the debugger code and demo code
  80. is in the same file. Eventually the debugger code should be moved
  81. into a reusable module.
  82. As it is now, the demo lets you set breakpoings in vertex/fragment
  83. programs, single step, and print intermediate register values. It's
  84. basically just a proof of concept.
  85. Directory tree reorganization
  86. -----------------------------
  87. The directory structure for Mesa has been overhauled to improve its layout.
  88. All source code for Mesa, GLU, GLUT, etc is now under the src/ directory
  89. in appropriate subdirectories.
  90. The Mesa source code and drivers has been reorganized under src/mesa/.
  91. All demonstration programs and tests are now in subdirectories under progs/.
  92. Build System Changes
  93. --------------------
  94. The GNU automake/autoconf support has been removed. As it was, it seldom
  95. worked on anything but Linux. The Mesa developers aren't big fans of
  96. automake/autoconf/libtool and didn't have the time to maintain it.
  97. If someone wants to contribute new automake/autoconf support (and is
  98. willing to maintain it), it may be re-incorporated into Mesa, subject
  99. to some requirements.
  100. The "old style" makefile system has been updated:
  101. 1. Make-config has been trimmed down to fewer, modern configurations.
  102. 2. Most of the bin/mklib.* scripts have been rolled into a new "mklib"
  103. script that works on all sorts of systems. There are probably some
  104. bugs in it, but it's been tested on Linux, SunOS 5.8 and IRIX 6.5.
  105. Improvements/contributes are greatly appreciated.
  106. 3. The Makefile.X11 files have been cleaned up in various ways
  107. Source File Changes
  108. -------------------
  109. The mmath.[ch] files are obsolete. Their contents have been moved
  110. into the imports.[ch] and macros.[ch] files.
  111. The files related to vertex and fragment programming have changed.
  112. Old files:
  113. vpexec.[ch]
  114. vpparse.[ch]
  115. vpstate.[ch]
  116. New files:
  117. program.[ch] - generic ARB/NV program code
  118. arbprogram.[ch] - ARB program API functions
  119. arbfragparse.[ch] - ARB fragment program parsing
  120. arbvertparse.[ch] - ARB vertex program parsing
  121. arbparse.[ch] - ARB vertex/fragment parsing
  122. arbparse_syn.h - vertex/fragment program syntax
  123. nvprogram.[ch] - NV program API functions
  124. nvvertprog.h - NV vertex program definitions
  125. nvfragprog.h - NV fragment program definitions
  126. nvvertparse.[ch] - NV vertex program parser
  127. nvfragparse.[ch] - NV fragment program parser
  128. nvvertexec.[ch] - NV vertex program execution
  129. swrast/s_nvfragprog.[ch] - NV fragment program execution
  130. The files related to per-vertex handling have changed.
  131. Old files:
  132. tnl/t_eval_api.c - old per-vertex code
  133. tnl/t_imm_alloc.c - old per-vertex code
  134. tnl/t_imm_api.c - old per-vertex code
  135. tnl/t_imm_debug.c - old per-vertex code
  136. tnl/t_imm_dlist.c - old per-vertex code
  137. tnl/t_imm_elt.c - old per-vertex code
  138. tnl/t_imm_eval.c - old per-vertex code
  139. tnl/t_imm_exec.c - old per-vertex code
  140. tnl/t_imm_fixup.c - old per-vertex code
  141. tnl/t_vtx_sse.c - old per-vertex code
  142. tnl/t_vtx_x86.c - old per-vertex code
  143. New files:
  144. tnl/t_save_api.c - new per-vertex code
  145. tnl/t_save_loopback.c - new per-vertex code
  146. tnl/t_save_playback.c - new per-vertex code
  147. tnl/t_vtx_eval.c - old per-vertex code
  148. Other new files:
  149. bufferobj.[ch] - GL_ARB_vertex_buffer_object functions
  150. version.h - defines the Mesa version info
  151. Other removed files:
  152. swrast/s_histogram.[ch] - moved into src/histogram.c
  153. Other Changes
  154. -------------
  155. The ctx->Driver.CreateTexture function has been removed - it wasn't used.
  156. New device driver hook functions:
  157. NewTextureObject - used to allocate struct gl_texture_objects
  158. NewTextureImage - used to allocate struct gl_texture_images
  159. New ctx->Texture._EnabledCoordUnits field:
  160. With the addition of GL_NV_fragment_program we may need to interpolate
  161. various sets of texture coordinates even when the corresponding texture
  162. unit is not enabled. That is, glEnable(GL_TEXTURE_xD) may never get
  163. called but we still may have to interpolate texture coordinates across
  164. triangles so that the fragment program will get them.
  165. This new field indicates which sets of texture coordinates are needed.
  166. If a bit is set in the ctx->Texture._EnabledUnits bitmask is set, the
  167. same bit MUST be set in ctx->Texture._EnabledCoordUnits.
  168. The ctx->_TriangleCaps field is deprecated.
  169. Instead of testing the DD_* bits in _TriangleCaps, you should instead
  170. directly test the relevant state variables, or use one of the helper
  171. functions like NEED_SECONDARY_COLOR() at the bottom of context.h
  172. While testing _TriangleCaps bits was fast, it was kludgey, and setting
  173. the bits in the first place could be error prone.
  174. New vertex processing code.
  175. The code behind glBegin, glEnd, glVertex, glNormal, etc. has been
  176. totally rewritten. It's a cleaner implementation now and should use
  177. less memory. (Keith)
  178. To Do
  179. -----
  180. Add screen-awareness to fakeglx.c
  181. Device Driver Status
  182. --------------------
  183. A number of Mesa's software drivers haven't been actively maintained for
  184. some time. We rely on volunteers to maintain many of these drivers.
  185. Here's the current status of all included drivers:
  186. Driver Status
  187. ---------------------- ---------------------
  188. XMesa (Xlib) implements OpenGL 1.4
  189. OSMesa (off-screen) implements OpenGL 1.4
  190. FX (3dfx Voodoo1/2) implements OpenGL 1.3
  191. SVGA implements OpenGL 1.3
  192. Wind River UGL implements OpenGL 1.3
  193. Windows/Win32 implements OpenGL 1.4
  194. DJGPP implements OpenGL 1.4
  195. GGI implements OpenGL 1.3
  196. BeOS implements OpenGL 1.4
  197. Allegro needs updating
  198. D3D needs updating
  199. Note: supporting OpenGL 1.4 (vs. 1.3 or 1.2) usually only requires that the
  200. driver call the _mesa_enable_1_4_extensions() function.
  201. ----------------------------------------------------------------------