mesa: Changes to 'refs/tags/mesa_7_1_rc2'

Brian Paul brianp at kemper.freedesktop.org
Tue Jul 8 14:17:44 PDT 2008


The annotated tag, mesa_7_1_rc2 has been created
        Commits at: http://cgit.freedesktop.org/mesa/mesa/log/?h=mesa_7_1_rc2

        at  5ea776d3bdbbc3fdcc52503243c5ff32447ca2f2 (tag)
   tagging  6fe78576291d38108c8815dc33fa44a8ae656f71 (commit)
  replaces  mesa_7_1_rc1
 tagged by  Brian Paul
        on  Tue Jul 8 15:17:16 2008 -0600

- Commits -----------------------------------------------
Mesa 7.1 release candidate 2

Alan Coopersmith (3):
      Fix builds with compilers other than gcc 3.0 & newer
      Make minstall work with Solaris /bin/sh
      Solaris port of Mesa 7.1 with autoconf support

Alan Hourihane (4):
      Silence warning
      Check for NULL pointer
      Check in SwapBuffers for any new pending dri2 events
      glcontextmodes.c is required remove the reference in .gitignore

Alex Deucher (29):
      Merge branch 'r500test' of git://people.freedesktop.org/~airlied/mesa into r345-cleanup
      R5xx:  Add R5xx pci ids
      R5xx: various updates
      Update comment
      R300: clean up VAP_PROG_STREAM_CNTL* register usage
      R300: cleanup VAP_CLIP_CNTL
      R300: fix VAP_OUTPUT_VTX_FMT_1 defines
      R3/4/5: fix TCL on r5xx, cleanup PVS code
      R5xx: Add R5xx pci ids
      R5xx: various updates
      Update comment
      R300: clean up VAP_PROG_STREAM_CNTL* register usage
      R300: cleanup VAP_CLIP_CNTL
      R300: fix VAP_OUTPUT_VTX_FMT_1 defines
      R3/4/5: fix TCL on r5xx, cleanup PVS code
      R300: fix rebase conflicts
      R500: add support for 4k textures
      Merge branch 'r500-support' of git+ssh://agd5f@git.freedesktop.org/git/mesa/mesa into r500-support
      R3xx: more PVS cleanup
      R300: cleanup FS code and fill in missing details
      R500: fixup r300EmitClearState() FP for r5xx
      R300: clean up Fog registers
      R300: clean up CB registers
      R3xx: clean up ZB registers
      R300: clean up GA registers
      r500: write out the correct FP registers
      r300: SC register naming cleanup
      R3/4/5xx: update to use drm get_param for num gb pipes
      R300: 1002:5657 is actually RV410

Brian (1):
      remove realglx.[ch] from build

Brian Paul (93):
      replace make with $(MAKE) (bug 16133)
      updated function list
      Use the GLuint64EXT type defined by GL_EXT_timer_query
      Assorted Windows fixes (Karl Schultz)
      undo prev change to this VC7 project file
      call glutDestroyWindow() before exit
      remove old/unused foomesa.h example header
      Set the attribute as used.
      remove references to incorrect -winwidth/-winheight, replace with -geometry
      code refactoring (draw_frame(0, handle_event())
      press 'a' to toggle animation
      fix event handling issues (such as when resizing when not animating)
      insert mesa/ before include files
      Move construction of libOSMesa.so into src/mesa/drivers/osmesa/Makefile
      minor changes
      prefix some includes with mesa/ or glapi/
      new Makefile for building "stand-alone", Xlib-based libGL
      Move building of stand-alone Mesa into drivers/x11/Makefile
      added dependencies
      remove X11, Glide, OSMesa sources
      reorder some lines, more make clean cmds
      Move fbdev driver construction into src/mesa/drivers/fbdev/Makefile
      move directfb-libgl into drivers/directfb/Makefile
      new drivers/Makefile with simple default/clean/install rules
      remove more old stuff, insert new driver_subdir build rules (but not enabled yet)
      Remove the old, complicated default build rule; use the new driver_subdirs rule
      move a beos-ism down into drivers/beos/Makefile
      remove stray semicolon
      move the install rules
      s/subdirs/asm_subdirs/
      document makefile changes
      Makefile clean-ups
      more old stuff, rename SOLO->MESA, etc
      Makefile clean-ups
      s/SOLO/MESA/
      fix Xlib libGL.so build problem.
      add some missing rule dependencies
      remove unnecessary swrast, tnl includes
      comments
      remove obsoleted doxygen \param comments
      mesa: allocate pixel zoom arrays on heap, not stack
      glu: silence warnings
      mesa: fix inconsistent use of GL_UNSIGNED_INT vs. GL_UNSIGNED_INT_24_8_EXT for Z unpacking
      mesa: tweak mipmap level selection for cube maps
      added pointer to beta/release candidates
      remove old comments
      s/inline/INLINE
      replace __inline and __inline__ with INLINE macro
      replace __inline and __inline__ with INLINE macro
      replace __inline and __inline__ with INLINE macro
      replace __inline and __inline__ with INLINE macro
      replace __inline and __inline__ with INLINE macro
      #undef DEBUG to silence warnings
      replace __inline and __inline__ with INLINE macro
      replace __inline and __inline__ with INLINE macro
      replace __inline and __inline__ with INLINE macro
      replace __inline and __inline__ with INLINE macro
      mesa: added null ptr checks
      s/GL_INVALID_VALUE/GL_INVALID_OPERATION/ in _mesa_get_uniformfv()
      init machine->Samplers (fixes vertex program texture fetches)
      press 'm' to toggle multisample enable/disable
      link to DRM 2.3.1
      mesa: add/fix some IrInfo entries for debugging purposes
      mesa: better function inlining in the presence of 'return' statements
      mesa: make _slang_swizzle_swizzle() non-private
      mesa: move some functions
      mesa: fix a GLSL vector subscript/writemask bug
      set ctx->Const.MaxVertexTextureImageUnits = 0
      generate a link error if the vertex shader references too many textures
      mesa: fix all(bvec2) function typo, add missing bvec2/3/4() constuctors
      mesa: regenerated files
      mesa: added some missing equal() notEqual() intrinsics
      mesa: regenerated
      mesa: fix error codes in _mesa_shader_source(), _mesa_get_shader_source()
      mesa: added some debug code (disabled)
      mesa: when linking a shader program, make sure all the shaders compiled OK
      mesa: fix vertex array validation test for attribute 0 (vert pos)
      mesa: fix incorrect array size, added assertion
      mesa: fix array storage allocation bug
      mesa: additional vec4 constructor
      mesa: regenerated file
      mesa: additional error checking, fix error codes
      mesa: fix problem freeing framebuffer/renderbuffer objects
      mesa: fix some error codes in _mesa_ShaderSourceARB()
      mesa: fix various error codes
      mesa: remove incorrect assertion
      mesa: generate GL_INVALID_OPERATION in _mesa_get_uniform_location() if program isn't linked
      GLSL-related bug fixes
      mesa: add missing VARYING case to storage_string()
      mesa: compute global var size before doing codegen
      mesa: more debug output
      added null texObj ptr check (bug 15567)
      mesa: bump version to rc2

Corbin Simpson (86):
      Add chip id 71D5 (RV530 M66) to radeon_chipset.h
      Initial r5xx fragment program compiler support.
      Added OPCODE_ABS, slightly fixed ADD/SUB
      Start of TEX/TXP support.
      r5xx: Fragprog shader now handles TEX/TXP correctly.
      Merge branch 'r500test' of git://people.freedesktop.org/~airlied/mesa into r500test
      Make radeon stuff build again.
      Merge branch 'r345-cleanup' of git://people.freedesktop.org/~agd5f/mesa into r500test
      Merge branch 'r500-support' into r500test
      Merge branch 'r500test' of git://people.freedesktop.org/~csimpson/mesa into r500-support
      r5xx: Fix dumb shader.
      r5xx: Fix for loops.
      r5xx: Use max_temp_idx.
      r5xx: Fix typo.
      r5xx: We update max_temp_idx now, so no need to hard-code it.
      r5xx: Emit an OUT instruction at the end of execution.
      r5xx: FP: Make MOV/ABS look pretty.
      r5xx: FP: Add OPCODE_TXB.
      r5xx: Index inputs and temps.
      r5xx: Fix false error with DP3/DP4.
      r5xx: Fix FP inputs. (For good?)
      r5xx: Unbreak MAX and MIN.
      r5xx: First swing at OPCODE_COS.
      r5xx: Adding more opcodes.
      r5xx: Add OPCODE_SCS.
      r5xx: Add OPCODE_SWZ.
      r5xx: Fix SCS.
      r5xx: Fix FRC.
      r5xx: Added OPCODE_DPH.
      r5xx: Add OPCODE_KIL.
      r5xx: ALU/OUT fixups.
      r5xx: Fix typo of epic proportions.
      r5xx: Swap sources for CMP.
      r5xx: Fix magic offsets for output fifo write masks.
      r5xx: Fixup emit_tex, add debugging info, enable temp temps.
      r5xx: New fix for COS/SIN/SCS.
      r5xx: Fixup SOP insts.
      r5xx: Count refs so we don't have to guess on temp reg allocation.
      r5xx: Add OPCODE_POW.
      r5xx: Initial (broken) OPCODE_LRP.
      r5xx: Change debug info for readability.
      r5xx: Fixed LRP.
      r5xx: More trig work.
      r5xx: Add OPCODE_DST.
      r5xx: Move dumb_shader.
      r5xx: Clean up some compiler warnings.
      r5xx: Remove some debugging cruft.
      r5xx: Fix SGE/SLT.
      r5xx: Consolidate FP tex insts.
      r5xx: Add emit_mad() for FP.
      r5xx: More emit_alu().
      r5xx: Negation masks for every inst except SWZ.
      r5xx: Massive MAD cleanup.
      r5xx: Unbreak texture swizzling.
      Replace copyright on r500_fragprog.c
      r5xx: First stab at LIT.
      r5xx: Moar LIT.
      r5xx: Enhance emit_mov().
      r5xx: Fix FP temp counting.
      r5xx: Fix emit_mov() regression.
      r5xx: Just a few small LIT fixes.
      r5xx: Add OPCODE_XPD. In working condition, I might add.
      r5xx: Dump shader constants when dumping program assembly.
      r5xx: Cleanup SOP with emit_sop().
      r5xx: More emit_sop(), stage 2.
      r5xx: Last emit_sop() for now.
      r5xx: Add OPCODE_FLR.
      r5xx: Enable depth write emission.
      r5xx: Fix speedy LIT once and for all.
      r5xx: Unbreak anisotropic filtering?
      r300: Allow driconf to set a default anisotropy.
      r3xx/r5xx: Don't force aniso.
      r5xx: FP refactor, take one.
      r5xx: New FP code, take two.
      r5xx: More FP rewriting; fix texrect FP insts.
      r5xx: Detangle FP fallback and translation switches.
      r5xx: Enable fragment.position, partial ARB_shadow.
      r300: Unbreak LOD biasing, a bit.
      Oops, misordered a few instructions.
      r300: Enable LOD bias state emission.
      r300: Forgot to clear old state before writing new state.
      r300: Fix new incarnation of bug 3195.
      r300: Make LOD bias a persistent environment variable.
      Revert accidental edit to progs/demos/lodbias.
      r300: Change LOD bias emission to more closely follow per-tex rules.
      r3xx/r5xx: Enable ARB_point_parameters.

Dan Nicholson (12):
      autoconf: Collect arch/platform settings in one location
      Distribute version.mk in case autoconf needs to be run again
      Update .gitignore for progs/glsl
      No more drivers/xorg
      Create $(TOP)/$(LIB_DIR) for install, too
      Don't make libmesa.a or libglapi.a depend on asm_subdirs
      Ensure all objects are built when installing DRI
      autoconf: Check for posix_memalign
      DRI-specific pkg-config file
      autoconf: Improve the visibility of the swrast DRI driver
      Set library and header installation directories from configuration
      autoconf: Allow commas or spaces to separate DRI drivers

Daniel Zhu (1):
      Rename quad function to quadr to avoid clash with Solaris quad typedef

Dave Airlie (95):
      r500: fragprog
      some basic r500 portage
      more r500 vs r300 kickin
      r500 RS unit setup
      mesa: cleanup state emission and rs for r500
      r300: this code really shouldn't be here.
      r300: add rv530 pci id for the t60p laptop
      r500: setup fragment program constant emission atom
      r500: fixup support for emitting fragment program to hardware.
      r500: enable gb tiling for r5xx
      r500: fixup fake shader to keep gears going
      Merge branch 'master' into r500test
      fixup r500 bits for renaming
      r500: fragprog
      some basic r500 portage
      more r500 vs r300 kickin
      r500 RS unit setup
      mesa: cleanup state emission and rs for r500
      r300: this code really shouldn't be here.
      r300: add rv530 pci id for the t60p laptop
      r500: setup fragment program constant emission atom
      r500: fixup support for emitting fragment program to hardware.
      r500: enable gb tiling for r5xx
      r500: fixup fake shader to keep gears going
      fixup r500 bits for renaming
      r500: fragprog
      some basic r500 portage
      more r500 vs r300 kickin
      r500 RS unit setup
      mesa: cleanup state emission and rs for r500
      r300: this code really shouldn't be here.
      r300: add rv530 pci id for the t60p laptop
      r500: setup fragment program constant emission atom
      r500: fixup support for emitting fragment program to hardware.
      r500: enable gb tiling for r5xx
      r500: fixup fake shader to keep gears going
      fixup r500 bits for renaming
      r300: add R300_NO_TCL to allow testing of non-tcl on tcl cards
      r300: recombine the vap input route 0 code and clean
      r300: update r300 rs unit for swtcl need to fix r500 most likely
      r300: setup vte according to inputs
      r300: remove debugging code
      r300: fix swtcl texrect path properly.
      r500: make sure we emit max temp atom.
      r500: make tri-param work
      r500: mov cleanup macros
      r500: consolidate tex instructions
      r500: for rectangular textures set to unscaled coordinates.
      r500: cleanup r500 RS setup
      r500: we just need to emit a colour for clear drop tex instruction
      r500: some trivial fixups to get tri working.
      r500: add cmp support in theory
      r300: set screen so that context init can find out chip ids
      r500: fixup the program allocations to be the correct sizes
      r3/500: emit RS state before VAP
      r500: split output/pixel masks and emit in the correct places
      r500: remove some debugging
      r500: fixup r500 rs unit texture coordinate counting
      r500: shift tex src properly
      r500: default rsunit swizzle like fglrx
      r500: set fragprog end to correct place
      r500: you can have a single texcoord
      r300: fixup US_OUT_FMT bits
      r500: add fragment program debug dumper
      r500: add mask debugging
      r500: fix swz gets and some returns
      r500: add more input srcs
      r300/r500: fix RS col fmt bits
      r300: some ctrl-m's wierd.
      r300/r500: fixup some of the register write sizes
      r500: finish main texture instruction decoding
      r500: set the RS unit register for R500 not R300 dangnammit..
      r500: print out opcode string
      r500: bump state atom size up for fp and fp constants
      r500: add depth output write
      r500: missed a couple of inst4s.
      r500: initial support for tmu mappings
      r300/r500: fix r500 fragment program texture unit references
      r500: remove debugging
      r500: hopefully fix 4096 texture harder
      r300/r500: emit flush inside vap_cntl state atom
      r500: reset fp/fp_const counts
      r500: need to re-setup inst offset/end for translated programs
      r500: the cs is shared per context - doh so don't use for program upload.
      r500: cleanup warnings and include files
      r500: remove warnings and only start on newer drms.
      Merge branch 'r500-support'
      r300: disable cmdbuf debug again
      r300: disable the lowimpact fallbacks by default.
      r500: add missing brackets around depth testing
      mesa/drm/ttm: allow build against non-TTM aware libdrm
      sparc: add DRI drivers for sparc patch from Fedora
      sparc: add swrast to dri drivers
      mesa: make mm.c use unsigned ints for offsets.
      dri: drop asserts to make build against stable libdrm

Dennis Kasprzyk (2):
      Report correct damage rectangle in CopySubBuffer.
      R1/2/3/4/5xx: fixed calculation of cliprects in CopyBuffer.

Eric Anholt (10):
      i915: Bug #14313: Fix accelerated (PBO) ReadPixels.
      i915: Allow accelerated pixel ops to be disabled with INTEL_NO_BLIT.
      i915: Fix read != draw drawable for glCopyPixels.
      i915: Add support for accelerated glBitmap, shared from 965.
      intel: Note reasons for blit pixel op fallbacks under INTEL_DEBUG=pix.
      intel: Merge check_blit_fragment_ops between i915/i965.
      intel: Avoid glBitmap software fallback for blending when no blending occurs.
      intel: Same pixel function init for everyone now.
      i965: Use the shared intel_pixel_copy.c.
      intel: Fix glCopyPixels when x or y are < 0 in hw coordinates.

George Sapountzis (8):
      dri/swrast: minor cosmetic
      glx: add dri_common.c ala xserver
      glx: use goto's vs. nested if's  ala xserver
      glx: use ErrorMessageF
      dri/swrast: add OpenGL 2.1 functions
      dri/swrast: make unbind a noop
      glx: load swrast_dri.so
      glx: add LIBGL_ALWAYS_SOFTWARE

H. Verbeet (2):
      Don't compute vertex fog when fragment program is active.
      Compute/setup fog coordinates even if vertex program is enabled.

Ian Romanick (4):
      bug #16503: use INSTALL macro instead of hard-coding to bin/minstall
      VBO: add GLX related annotations
      VBO: Add missing functions related to VERTEX_ATTRIB_ARRAY_BUFFER_BINDING_ARB
      VBO: Regenerate files based on recent changes to gl_API.xml

Jose Fonseca (2):
      Fix CRLF line endings.
      Fix CRLF line endings.

Karl Schultz (1):
      added prog_uniform.[ch]

Luc Verhaegen (1):
      Fix segfault on exit in unichrome driver (bug 16150)

Markus Amsler (3):
      r300: Set correct VAP_CNTL per vertex program.
      r300: fix R300_FG_ALPHA_FUNC_VAL_MASK
      r500: Use 8bit fog alpha compare value.

Nicolai Haehnle (45):
      [r300] Fix KIL instruction and swizzling of texture coordinates
      r300: Writing to result.depth in fragment programs (R3xx; only stub for R5xx)
      [texenvprogram] Fix refcounting-related memory leak.
      [t_vp_build] Fix refcounting-related memory leak
      r300: Remove unused variable r300_tex_obj::dirty_state and related defines
      r300: Cleanup TX_MIN_FILTER defines
      r300: Improve texture layout calculations
      r300: Update LOD_BIAS register constants
      r300: Further anisotropic filtering fixes
      r300: Fix stencil buffer clear
      r300: Apply writemask when clearing stencil buffer
      r300: Some cleanups in depth and stencil state handling
      r300: Workaround hardware readcache problem
      r300: Reenable t->filter_1 (though currently not used)
      r300: Fix depth texture GL_LUMINANCE vs. GL_INTENSITY mixup
      r300/r500: Separate fragprog compiler structures
      r300_fragprog: Remove dead code
      r300_fragprog: Refactoring and cleanup
      r300_fragprog: Remove dead declarations, move NOP declarations into source
      r300: Do not include r300_fragprog.h from r300_context.h and other cleanups
      r300: Add radeon_program and trivial refactoring of r300_fragprog to use it
      r300_fragprog: Refactor TEX transformation
      r300: Add radeonCompilerDump for debugging
      texenvprogram: Mark textures using ARB_shadow as ShadowSamplers
      r300: Implement GL_ARB_shadow and GL_EXT_shadow_funcs
      r300_fragprog: Use less complex instructions
      r300: Add radeonTransformALU and fix a bug in r300_fragprog DPH
      r300: Fix wrap mode for 1D textures
      demos/shadowtex: Don't set TEXTURE_WRAP_T for 1D texture
      r300: Cleanup LodBias support
      r300: Fix dumb mistake in LOD bias translation
      Enable TexGen based on InputsRead when a fragment program is active
      r300: Fix depth texture in compare mode
      _mesa_clone_program: Copy ShadowSamplers
      r500: Fix blend color.
      r500: Fix a mixup in fragment program LRP instruction emit
      r500_fragprog: Cleanup some unused variables and code.
      r300: Remove clause stuff for now in favour of a cloned generic gl_program
      r500: Major refactoring of fragment program emit
      r300: Correctly scan for used temporary registers
      r300: Allow adding parameters during fragprog transform, share LIT code
      r300: Translate fragment program LRP in radeon_program_alu.c
      r300: Translate fragment program DST in radeon_program_alu
      r300_fragprog: Emulate trigonometric functions in radeon_program_alu
      r500_fragprog: Fix RSQ with negative parameters

Paulo Cesar Pereira de Andrade (1):
      Bring over commit 8d4d0b47a07a298a20ffae9fefe96c8c7ca9dccc from xserver tree

Roland Scheidegger (4):
      i965: fix OPCODE_TEX when additional ops are needed
      copy msaa visuals capability
      r200: fix typo in r200TryDrawPixels parameter validation (bug 16406)
      mesa: fix issues around multisample enable

Wilfried Holzke (1):
      assorted glide driver fixes

Xiang, Haihao (6):
      i915: Fix GL_DEPTH_TEXTURE_MODE issue. (bug #16221)
      i965: apply commit 6c1a98e97affb2163e776551eb3a9e669ff99bbf to glsl
      i965: add support for Intel 4 series chipsets
      dri: Take the base image size into account when computing
      mesa: fix polygon offset issue (bug #12061)
      i965: official name for GM45 chipset

Zack Rusin (7):
      glsl: implement variable array indexes
      shaders: fix allocation of arrays of parameters
      almost forgot - RelAddr is a boolean so use it here
      glsl: fix array size initialiazers using const variables
      glsl: allow uniforms
      glsl: make sure we replace all output reads with temporaries
      silly bug: it's PROGRAM_VARYING

---------------------------------


More information about the mesa-commit mailing list