mesa: Makefile: 7.2 release candidate 1

mesa: documentation updates for 7.2 release
updated Cell driver info
2008-09-13 16:06:12 -06:00 · 2008-09-13 14:43:54 -06:00 · 2008-09-13 14:39:39 -06:00 · 2008-09-13 14:36:35 -06:00 · 2008-09-13 14:36:17 -06:00 · 2008-09-13 14:34:51 -06:00
73 changed files with 1620 additions and 1661 deletions
--- a/8
+++ b/8
@@ -174,10 +174,10 @@ ultrix-gcc:
 # Rules for making release tarballs
-DIRECTORY = Mesa-7.1-rc4
+DIRECTORY = Mesa-7.2-rc1
-LIB_NAME = MesaLib-7.1-rc4
+LIB_NAME = MesaLib-7.2-rc1
-DEMO_NAME = MesaDemos-7.1-rc4
+DEMO_NAME = MesaDemos-7.2-rc1
-GLUT_NAME = MesaGLUT-7.1-rc4
+GLUT_NAME = MesaGLUT-7.2-rc1
 MAIN_FILES = \
 	$(DIRECTORY)/Makefile*						\
--- a/configs/autoconf.in
+++ b/configs/autoconf.in
@@ -86,7 +86,6 @@ APP_LIB_DEPS = $(EXTRA_LIB_PATH) @APP_LIB_DEPS@
 DRI_LIB_DEPS = $(EXTRA_LIB_PATH) @DRI_LIB_DEPS@
 LIBDRM_CFLAGS = @LIBDRM_CFLAGS@
 LIBDRM_LIB = @LIBDRM_LIBS@
 DRI2PROTO_CFLAGS = @DRI2PROTO_CFLAGS@
 EXPAT_INCLUDES = @EXPAT_INCLUDES@
 # Autoconf directories
--- a/configs/default
+++ b/configs/default
@@ -9,7 +9,7 @@ CONFIG_NAME = default
 # Version info
 MESA_MAJOR=7
-MESA_MINOR=1
+MESA_MINOR=2
 MESA_TINY=0
 MESA_VERSION = $(MESA_MAJOR).$(MESA_MINOR).$(MESA_TINY)
--- a/configure.ac
+++ b/configure.ac
@@ -23,7 +23,6 @@ AC_CANONICAL_HOST
 dnl Versions for external dependencies
 LIBDRM_REQUIRED=2.3.1
 DRI2PROTO_REQUIRED=1.1
 dnl Check for progs
 AC_PROG_CPP
@@ -531,7 +530,6 @@ dri)
    # Check for libdrm
    PKG_CHECK_MODULES([LIBDRM], [libdrm >= $LIBDRM_REQUIRED])
    PKG_CHECK_MODULES([DRI2PROTO], [dri2proto >= $DRI2PROTO_REQUIRED])
    # find the DRI deps for libGL
    if test "$x11_pkgconfig" = yes; then
--- a/docs/cell.html
+++ b/docs/cell.html
@@ -6,7 +6,7 @@
 <BODY>
-<H1>Mesa Cell Driver</H1>
+<H1>Mesa/Gallium Cell Driver</H1>
 <p>
 The Mesa
@@ -23,18 +23,19 @@ Two phases are planned.
 First, to implement the framework for parallel rasterization using the Cell
 SPEs, including texture mapping.
 Second, to implement a full-featured OpenGL driver with support for GLSL, etc.
 The second phase is now underway.
 </p>
 <H2>Source Code</H2>
 <p>
-The Cell driver source code is on the <code>gallium-0.1</code> branch of the
+The latest Cell driver source code is on the <code>gallium-0.2</code> branch
-git repository.
+of the Mesa git repository.
 After you've cloned the repository, check out the branch with:
 </p>
 <pre>
-   git-checkout -b gallium-0.1 origin/gallium-0.1
+   git-checkout -b gallium-0.2 origin/gallium-0.2
 </pre>
 <p>
 To build the driver you'll need the IBM Cell SDK (version 2.1 or 3.0).
@@ -43,12 +44,13 @@ or the Cell Simulator (untested, though).
 </p>
 <p>
-If using Cell SDK 3.0, first edit configs/linux-cell and add
+If using Cell SDK 2.1, see the configs/linux-cell file for some
-<code>-DSPU_MAIN_PARAM_LONG_LONG</code> to the SPU_CFLAGS.
+special changes.
 </p>
 <p>
 To compile the code, run <code>make linux-cell</code>.
 To build in debug mode, run <code>make linux-cell-debug</code>.
 </p>
 <p>
@@ -60,7 +62,7 @@ directory that contains <code>libGL.so</code>.
 Verify that the Cell driver is being used by running <code>glxinfo</code>
 and looking for:
 <pre>
-  OpenGL renderer string: Gallium 0.1, Cell on Xlib
+  OpenGL renderer string: Gallium 0.2, Cell on Xlib
 </pre>
@@ -77,21 +79,49 @@ SPU local store as needed.
 Similarly, textures are tiled and brought into local store as needed.
 </p>
 <p>
 More recently, vertex transformation has been parallelized across the SPUs
 as well.
 </p>
 <H2>Status</H2>
 <p>
-As of February 2008 the driver supports smooth/flat shaded triangle rendering
+As of September 2008, the driver supports smooth/flat shaded triangle rendering
 with Z testing and simple texture mapping.
 Simple demos like gears run successfully.
 To test texture mapping, try progs/demos/texcyl (press right mouse button for
 rendering options).
 </p>
 <p>
 Runtime/dynamic code generation is being done for per-fragment
 operations (Z test, blend, etc) and for fragment programs (though only a
 few opcodes are implemented now).
 </p>
 <p>
 In general, however, the driver is rather slow because all vertex
 transformation is being done by an interpreter running on the PPU.
 Programs with many vertices or complex vertex shaders will run especially
 slow.
 This will be addressed in the future.
 </p>
 <H2>Debug Options</H2>
 <p>
 The CELL_DEBUG env var can be set to a comma-separated list of one or
 more of the following debug options:
 </p>
 <ul>
 <li><b>checker</b> - use a different background clear color for each SPU.
   This lets you see which SPU is rendering which screen tiles.
 <li><b>sync</b> - wait/synchronize after each DMA transfer
 </ul>
 <p>
 If the GALLIUM_NOCELL env var is set, the softpipe driver will be used
 intead of the Cell driver.
 This is useful for comparison/validation.
 </p>
 <H2>Contributing</H2>
--- a/docs/contents.html
+++ b/docs/contents.html
@@ -48,6 +48,7 @@ a:visited {
 <li><a href="lists.html" target="MainFrame">Mailing Lists</a>
 <li><a href="bugs.html" target="MainFrame">Bug Database</a>
 <li><a href="webmaster.html" target="MainFrame">Webmaster</a>
 <li><a href="http://dri.freedesktop.org/" target="MainFrame">Wiki</a>
 </ul>
 <b>User Topics</b>
--- a/docs/download.html
+++ b/docs/download.html
@@ -9,7 +9,9 @@
 <H1>Downloading</H1>
 <p>
-Last stable release: <b>7.0.4</b>
+Current stable release: <b>7.2</b>
 <br>
 Current development: Mesa git/master
 </p>
 <p>
--- a/docs/intro.html
+++ b/docs/intro.html
@@ -22,9 +22,10 @@ for modern GPUs.
 <p>
 Mesa ties into several other open-source projects: the 
-<a href="http://dri.sf.net/" target="_parent">Direct Rendering Infrastructure</a>
+<a href="http://dri.freedesktop.org/" target="_parent">Direct Rendering 
-and <a href="http://x.org" target="_parent">X.org</a> to provide OpenGL
+Infrastructure</a> and <a href="http://x.org" target="_parent">X.org</a> to 
-support to users of X on Linux, FreeBSD and other operating systems.
+provide OpenGL support to users of X on Linux, FreeBSD and other operating 
 systems.
 </p>
--- a/docs/lists.html
+++ b/docs/lists.html
@@ -9,7 +9,7 @@
 <H1>Mailing Lists</H1>
-<p>There are four Mesa mailing lists:
+<p>There are four Mesa 3D / DRI mailing lists:
 </p>
 <ul>
@@ -19,14 +19,14 @@ versions are sent to this list.
 </li>
 <br>
 <li><a href="https://lists.sourceforge.net/lists/listinfo/mesa3d-users"
-target="_parent">mesa3d-users</a> - intended for users of the Mesa library.
+target="_parent">mesa3d-users</a> - intended for users of the Mesa and DRI.
 Newbie questions are appropriate, but please try the general OpenGL
-resources and Mesa documentation first.
+resources and Mesa/DRI documentation first.
 </li>
 <br>
 <li><a href="https://lists.sourceforge.net/lists/listinfo/mesa3d-dev"
-target="_parent">mesa3d-dev</a> - for discussion of Mesa development.
+target="_parent">mesa3d-dev</a> - for discussion of Mesa and Direct Rendering 
-Not for beginners.
+Infrastructure development. Not for beginners.
 </li>
 <br>
 <li><a href="http://lists.freedesktop.org/mailman/listinfo/mesa-commit"
@@ -37,6 +37,9 @@ Note: the old mesa3d-cvs list is no longer in use.
 </li>
 </ul>
 <p>For mailing lists about Direct Rendering Modules (drm) in Linux/BSD 
 kernels, see <a href="http://dri.freedesktop.org/wiki/MailingLists">wiki</a>.
 <p>
 <b>Notice</b>: non-member posts to any of these lists will be automatically
 rejected.
--- a/docs/news.html
+++ b/docs/news.html
@@ -11,6 +11,23 @@
 <H1>News</H1>
 <h2>September 15, 2008</h2>
 <p>
 <a href="relnotes-7.2.html">Mesa 7.2</a> is released.
 This is stable, bug-fix release.
 </p>
 <h2>August 26, 2008</h2>
 <p>
 <a href="relnotes-7.1.html">Mesa 7.1</a> is released.
 This is a new development release.
 It should be relatively stable, but those especially concerned about
 stability should wait for the 7.2 release or use Mesa 7.0.4 (the
 previous stable release).
 </p>
 <h2>August 16, 2008</h2>
 <p>
 <a href="relnotes-7.0.4.html">Mesa 7.0.4</a> is released.
--- a/docs/relnotes-7.1.html
+++ b/docs/relnotes-7.1.html
@@ -8,28 +8,46 @@
 <body bgcolor="#eeeeee">
-<H1>Mesa 7.1 Release Notes / May XX, 2008</H1>
+<H1>Mesa 7.1 Release Notes / August 26, 2008</H1>
 <p>
 Mesa 7.1 is a new development release.
 There have been many internal code changes since Mesa 7.0.x.
 It should be relatively stable, but those who are especially concerned about
-stability should wait for Mesa 7.2.
+stability should wait for Mesa 7.2 or use Mesa 7.0.4 (the previous stable
 release).
 </p>
 </>
 Note that this version of Mesa does not use the GEM memory manager.
 The master branch of git uses GEM.
 </p>
 <p>
 DRM version 2.3.1 should be used with Mesa 7.1
 </p>
 <h2>MD5 checksums</h2>
 <pre>
-TBD
+971c2fe6e6949dc5ba200a6f97a6dc81  MesaLib-7.1.tar.gz
 6bff7f532d16f90f944a400c8bd7074d  MesaLib-7.1.tar.bz2
 d48224bf9d54c3da6776adb4869ba024  MesaLib-7.1.zip
 3de268420efca43e9a19ab506cdfc993  MesaDemos-7.1.tar.gz
 abfc9775e1462363af8ec160d1feb01f  MesaDemos-7.1.tar.bz2
 f7b3623387c4036e9895cd9ac0dfad99  MesaDemos-7.1.zip
 fdf348f78cd09304b6ff801ef8acc8eb  MesaGLUT-7.1.tar.gz
 f6d88a4eeb02e98c7e92f1c895d3c76b  MesaGLUT-7.1.tar.bz2
 4dc102a5ca51e1c41dde87d3f8c7b22a  MesaGLUT-7.1.zip
 </pre>
 <h2>New features</h2>
 <ul>
 <li>autoconf-based configuration (and clean-up of Makefiles)
 <li>Assorted DRI driver enhancements
 <li>Reduced dependencies between X server and Mesa
 <li>GL_EXT_texture_from_pixmap extension for Xlib driver
 <li>Support for the GL shading language with i965 driver (implemented by Intel)
 <li>ATI R500 series support (Radeon X1300&#8211;X1950) in r300 DRI driver
 </ul>
@@ -37,6 +55,7 @@ TBD
 <ul>
 <li>Numerous GLSL fixes
 <li>Fixed some error code/detection bugs in the GLSL-related API functions
 <li>Lots of DRI driver fixes.
 </ul>
--- a/docs/relnotes-7.2.html
+++ b/docs/relnotes-7.2.html
@@ -0,0 +1,86 @@
 <HTML>
 <TITLE>Mesa Release Notes</TITLE>
 <head><link rel="stylesheet" type="text/css" href="mesa.css"></head>
 <BODY>
 <body bgcolor="#eeeeee">
 <H1>Mesa 7.2 Release Notes / (date tbd)</H1>
 <p>
 Mesa 7.2 is a stable release fixing bugs found in 7.1, which was a
 new development release.
 </p>
 <p>
 Mesa 7.2 implements the OpenGL 2.1 API, but the version reported by
 glGetString(GL_VERSION) depends on the particular driver being used.
 Some drivers don't support all the features required in OpenGL 2.1.
 </p>
 <p>
 Note that this version of Mesa does not use the GEM memory manager.
 The master branch of git uses GEM.
 The prototype DRI2 code that was in 7.1 has also been removed.
 </p>
 <p>
 DRM version 2.3.1 should be used with Mesa 7.2
 </p>
 <h2>MD5 checksums</h2>
 <pre>
 tbd
 </pre>
 <h2>New features</h2>
 <ul>
 <li>i965 driver: added support for G41 chipset (Intel)
 </ul>
 <h2>Bug fixes</h2>
 <ul>
 <li>Assorted bug fixes for Ming build
 <li>Fixed some vertex/pixel buffer object reference counting bugs
 <li>Fixed depth/stencil bug in i915/945 driver
 <li>Fixed some shader flow control bugs in i965 driver
 <li>Fixed a few tdfx driver bugs which prevented driver from working
 </ul>
 <h2>To Do (someday) items</h2>
 <ul>
 <li>Remove the MEMCPY() and _mesa_memcpy() wrappers and just use memcpy().
 Probably do the same for malloc, calloc, etc.
 The wrappers were useful in the past for memory debugging but now we
 have valgrind.  Not worried about SunOS 4 support anymore either...
 <li>Switch to freeglut
 <li>Fix linux-glide target/driver.
 <li>Improved lambda and derivative calculation for frag progs.
 </ul>
 <h2>Driver Status</h2>
 <pre>
 Driver			Status
 ----------------------	----------------------
 DRI drivers		varies with the driver
 XMesa/GLX (on Xlib)	implements OpenGL 2.1
 OSMesa (off-screen)	implements OpenGL 2.1
 Windows/Win32		implements OpenGL 2.1
 Glide (3dfx Voodoo1/2)	implements OpenGL 1.3
 SVGA			unsupported
 Wind River UGL		unsupported
 DJGPP			unsupported
 GGI			unsupported
 BeOS			unsupported
 Allegro			unsupported
 D3D			unsupported
 </pre>
 </body>
 </html>
--- a/docs/relnotes.html
+++ b/docs/relnotes.html
@@ -20,6 +20,7 @@ The release notes summarize what's new or changed in each Mesa release.
 </p>
 <UL>
 <LI><A HREF="relnotes-7.2.html">7.2 release notes</A>
 <LI><A HREF="relnotes-7.1.html">7.1 release notes</A>
 <LI><A HREF="relnotes-7.0.4.html">7.0.4 release notes</A>
 <LI><A HREF="relnotes-7.0.3.html">7.0.3 release notes</A>
--- a/progs/demos/rain.cxx
+++ b/progs/demos/rain.cxx
@@ -114,7 +114,7 @@ static void calcposobs(void)
  rainParticle::setRainingArea(obs[0]-7.0f,-0.2f,obs[2]-7.0f,obs[0]+7.0f,8.0f,obs[2]+7.0f);
 }
-static void printstring(void *font, char *string)
+static void printstring(void *font, const char *string)
 {
  int len,i;
--- a/progs/xdemos/glxinfo.c
+++ b/progs/xdemos/glxinfo.c
@@ -447,7 +447,6 @@ print_screen_info(Display *dpy, int scrnum, Bool allowDirect, GLboolean limits)
   if (!ctx) {
      fprintf(stderr, "Error: glXCreateContext failed\n");
      XFree(visinfo);
      XDestroyWindow(dpy, win);
      return;
   }
--- a/src/glu/sgi/libnurbs/interface/mystdio.h
+++ b/src/glu/sgi/libnurbs/interface/mystdio.h
@@ -43,7 +43,7 @@
 #define __glumystdio_h_
 #ifdef STANDALONE
-inline void _glu_dprintf( char *, ... ) { }
+inline void _glu_dprintf( const char *, ... ) { }
 #endif
 #ifdef LIBRARYBUILD
@@ -51,12 +51,12 @@ inline void _glu_dprintf( char *, ... ) { }
 #include <stdio.h>
 #define _glu_dprintf printf
 #else
-inline void _glu_dprintf( char *, ... ) { }
+inline void _glu_dprintf( const char *, ... ) { }
 #endif
 #endif
 #ifdef GLBUILD
-inline void _glu_dprintf( char *, ... ) { }
+inline void _glu_dprintf( const char *, ... ) { }
 #endif
 #ifndef NULL
--- a/src/glu/sgi/libnurbs/internals/backend.cc
+++ b/src/glu/sgi/libnurbs/internals/backend.cc
@@ -246,7 +246,7 @@ Backend::evalVStrip(int n_left, REAL u_left, REAL* left_val,
 *-------------------------------------------------------------------------
 */
 void
-Backend::bgntmesh( char * )		
+Backend::bgntmesh( const char * )		
 {
 #ifndef NOWIREFRAME
--- a/src/glu/sgi/libnurbs/internals/backend.h
+++ b/src/glu/sgi/libnurbs/internals/backend.h
@@ -65,7 +65,7 @@ public:
    void		surfbbox( long, REAL *, REAL * );
    void		surfgrid( REAL, REAL, long, REAL, REAL, long ); 
    void		surfmesh( long, long, long, long ); 
-    void		bgntmesh( char * );
+    void		bgntmesh( const char * );
    void		endtmesh( void );
    void		swaptmesh( void );
    void		tmeshvert( GridTrimVertex * );
--- a/src/glu/sgi/libnurbs/internals/bufpool.cc
+++ b/src/glu/sgi/libnurbs/internals/bufpool.cc
@@ -48,7 +48,7 @@
 * Pool - allocate a new pool of buffers
 *-----------------------------------------------------------------------------
 */
-Pool::Pool( int _buffersize, int initpoolsize, char *n )
+Pool::Pool( int _buffersize, int initpoolsize, const char *n )
 {
    if((unsigned)_buffersize < sizeof(Buffer))
        buffersize = sizeof(Buffer);
--- a/src/glu/sgi/libnurbs/internals/bufpool.h
+++ b/src/glu/sgi/libnurbs/internals/bufpool.h
@@ -55,7 +55,7 @@ class Buffer {
 class Pool {
 public:
-			Pool( int, int, char * );
+			Pool( int, int, const char * );
 			~Pool( void );
    inline void*	new_buffer( void );
    inline void		free_buffer( void * );
@@ -74,7 +74,7 @@ protected:
    int			nextfree;		/* byte offset past next free buffer */
    int			initsize;
    enum Magic { is_allocated = 0xf3a1, is_free = 0xf1a2 };
-    char		*name;			/* name of the pool */
+    const char		*name;			/* name of the pool */
    Magic		magic;			/* marker for valid pool */
 };
--- a/src/glu/sgi/libnurbs/internals/knotvector.cc
+++ b/src/glu/sgi/libnurbs/internals/knotvector.cc
@@ -126,7 +126,7 @@ int Knotvector::validate( void )
    return 0;
 }
-void Knotvector::show( char *msg )
+void Knotvector::show( const char *msg )
 {
 #ifndef NDEBUG
    _glu_dprintf( "%s\n", msg );
--- a/src/glu/sgi/libnurbs/internals/knotvector.h
+++ b/src/glu/sgi/libnurbs/internals/knotvector.h
@@ -49,7 +49,7 @@ struct Knotvector { /* a knot vector */
 			~Knotvector( void );
    void		init( long, long, long, INREAL * );
    int			validate( void );
-    void 		show( char * );
+    void 		show( const char * );
    long		order;		/* order of spline  */
    long		knotcount;	/* number of knots  */
--- a/src/glu/sgi/libnurbs/internals/nurbstess.cc
+++ b/src/glu/sgi/libnurbs/internals/nurbstess.cc
@@ -677,7 +677,7 @@ NurbsTessellator::do_nurbserror( int msg )
 }
 int 
-NurbsTessellator::do_check_knots( Knotvector *knots, char *msg )
+NurbsTessellator::do_check_knots( Knotvector *knots, const char *msg )
 {
    int status = knots->validate();
    if( status ) {
--- a/src/glu/sgi/libnurbs/internals/nurbstess.h
+++ b/src/glu/sgi/libnurbs/internals/nurbstess.h
@@ -112,7 +112,7 @@ protected:
 private:
    void		resetObjects( void );
-    int			do_check_knots( Knotvector *, char * );
+    int			do_check_knots( Knotvector *, const char * );
    void		do_nurbserror( int );
    void		do_bgncurve( O_curve * );
    void		do_endcurve( void );
--- a/src/glx/x11/Makefile
+++ b/src/glx/x11/Makefile
@@ -35,9 +35,7 @@ SOURCES = \
 	  dri_common.c \
 	  dri_glx.c \
 	  XF86dri.c \
-	  glxhash.c \
+	  glxhash.c
 	  dri2_glx.c \
 	  dri2.c
 include $(TOP)/src/mesa/sources
@@ -53,7 +51,6 @@ INCLUDES = -I. \
 	-I$(TOP)/src/mesa/main \
 	-I$(TOP)/src/mesa/glapi \
 	$(LIBDRM_CFLAGS) \
 	$(DRI2PROTO_CFLAGS) \
 	$(X11_INCLUDES)
--- a/src/glx/x11/dri2.c
+++ b/src/glx/x11/dri2.c
@@ -1,252 +0,0 @@
 /*
 * Copyright © 2008 Red Hat, Inc.
 *
 * Permission is hereby granted, free of charge, to any person obtaining a
 * copy of this software and associated documentation files (the "Soft-
 * ware"), to deal in the Software without restriction, including without
 * limitation the rights to use, copy, modify, merge, publish, distribute,
 * and/or sell copies of the Software, and to permit persons to whom the
 * Software is furnished to do so, provided that the above copyright
 * notice(s) and this permission notice appear in all copies of the Soft-
 * ware and that both the above copyright notice(s) and this permission
 * notice appear in supporting documentation.
 *
 * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS
 * OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABIL-
 * ITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT OF THIRD PARTY
 * RIGHTS. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR HOLDERS INCLUDED IN
 * THIS NOTICE BE LIABLE FOR ANY CLAIM, OR ANY SPECIAL INDIRECT OR CONSE-
 * QUENTIAL DAMAGES, OR ANY DAMAGES WHATSOEVER RESULTING FROM LOSS OF USE,
 * DATA OR PROFITS, WHETHER IN AN ACTION OF CONTRACT, NEGLIGENCE OR OTHER
 * TORTIOUS ACTION, ARISING OUT OF OR IN CONNECTION WITH THE USE OR PERFOR-
 * MANCE OF THIS SOFTWARE.
 *
 * Except as contained in this notice, the name of a copyright holder shall
 * not be used in advertising or otherwise to promote the sale, use or
 * other dealings in this Software without prior written authorization of
 * the copyright holder.
 *
 * Authors:
 *   Kristian Høgsberg (krh@redhat.com)
 */
 #define NEED_REPLIES
 #include <X11/Xlibint.h>
 #include <X11/extensions/Xext.h>
 #include <X11/extensions/extutil.h>
 #include <X11/extensions/dri2proto.h>
 #include "glheader.h"
 #include "xf86drm.h"
 #include "dri2.h"
 static char dri2ExtensionName[] = DRI2_NAME;
 static XExtensionInfo *dri2Info;
 static XEXT_GENERATE_CLOSE_DISPLAY (DRI2CloseDisplay, dri2Info)
 static /* const */ XExtensionHooks dri2ExtensionHooks = {
    NULL,				/* create_gc */
    NULL,				/* copy_gc */
    NULL,				/* flush_gc */
    NULL,				/* free_gc */
    NULL,				/* create_font */
    NULL,				/* free_font */
    DRI2CloseDisplay,			/* close_display */
    NULL,				/* wire_to_event */
    NULL,				/* event_to_wire */
    NULL,				/* error */
    NULL,				/* error_string */
 };
 static XEXT_GENERATE_FIND_DISPLAY (DRI2FindDisplay, dri2Info, 
 				   dri2ExtensionName, 
 				   &dri2ExtensionHooks, 
 				   0, NULL)
 Bool DRI2QueryExtension(Display *dpy, int *eventBase, int *errorBase)
 {
    XExtDisplayInfo *info = DRI2FindDisplay(dpy);
    if (XextHasExtension(info)) {
 	*eventBase = info->codes->first_event;
 	*errorBase = info->codes->first_error;
 	return True;
    }
    return False;
 }
 Bool DRI2QueryVersion(Display *dpy, int *major, int *minor)
 {
    XExtDisplayInfo *info = DRI2FindDisplay (dpy);
    xDRI2QueryVersionReply rep;
    xDRI2QueryVersionReq *req;
    XextCheckExtension (dpy, info, dri2ExtensionName, False);
    LockDisplay(dpy);
    GetReq(DRI2QueryVersion, req);
    req->reqType = info->codes->major_opcode;
    req->dri2ReqType = X_DRI2QueryVersion;
    req->majorVersion = DRI2_MAJOR;
    req->minorVersion = DRI2_MINOR;
    if (!_XReply(dpy, (xReply *)&rep, 0, xFalse)) {
 	UnlockDisplay(dpy);
 	SyncHandle();
 	return False;
    }
    *major = rep.majorVersion;
    *minor = rep.minorVersion;
    UnlockDisplay(dpy);
    SyncHandle();
    return True;
 }
 Bool DRI2Connect(Display *dpy, int screen,
 		 char **driverName, char **busId, unsigned int *sareaHandle)
 {
    XExtDisplayInfo *info = DRI2FindDisplay(dpy);
    xDRI2ConnectReply rep;
    xDRI2ConnectReq *req;
    XextCheckExtension (dpy, info, dri2ExtensionName, False);
    LockDisplay(dpy);
    GetReq(DRI2Connect, req);
    req->reqType = info->codes->major_opcode;
    req->dri2ReqType = X_DRI2Connect;
    req->screen = screen;
    if (!_XReply(dpy, (xReply *)&rep, 0, xFalse)) {
 	UnlockDisplay(dpy);
 	SyncHandle();
 	return False;
    }
    *sareaHandle = rep.sareaHandle;
    *driverName = Xmalloc(rep.driverNameLength + 1);
    if (*driverName == NULL) {
 	_XEatData(dpy, 
 		  ((rep.driverNameLength + 3) & ~3) +
 		  ((rep.busIdLength + 3) & ~3));
 	UnlockDisplay(dpy);
 	SyncHandle();
 	return False;
    }
    _XReadPad(dpy, *driverName, rep.driverNameLength);
    (*driverName)[rep.driverNameLength] = '\0';
    *busId = Xmalloc(rep.busIdLength + 1);
    if (*busId == NULL) {
 	Xfree(*driverName);
 	_XEatData(dpy, ((rep.busIdLength + 3) & ~3));
 	UnlockDisplay(dpy);
 	SyncHandle();
 	return False;
    }
    _XReadPad(dpy, *busId, rep.busIdLength);
    (*busId)[rep.busIdLength] = '\0';
    UnlockDisplay(dpy);
    SyncHandle();
    return rep.sareaHandle != 0;
 }
 Bool DRI2AuthConnection(Display *dpy, int screen, drm_magic_t magic)
 {
    XExtDisplayInfo *info = DRI2FindDisplay(dpy);
    xDRI2AuthConnectionReq *req;
    xDRI2AuthConnectionReply rep;
    XextCheckExtension (dpy, info, dri2ExtensionName, False);
    LockDisplay(dpy);
    GetReq(DRI2AuthConnection, req);
    req->reqType = info->codes->major_opcode;
    req->dri2ReqType = X_DRI2AuthConnection;
    req->screen = screen;
    req->magic = magic;
    rep.authenticated = 0;
    if (!_XReply(dpy, (xReply *)&rep, 0, xFalse)) {
 	UnlockDisplay(dpy);
 	SyncHandle();
 	return False;
    }
    UnlockDisplay(dpy);
    SyncHandle();
    return rep.authenticated;
 }
 Bool DRI2CreateDrawable(Display *dpy, XID drawable,
 			unsigned int *handle, unsigned int *head)
 {
    XExtDisplayInfo *info = DRI2FindDisplay(dpy);
    xDRI2CreateDrawableReply rep;
    xDRI2CreateDrawableReq *req;
    XextCheckExtension (dpy, info, dri2ExtensionName, False);
    LockDisplay(dpy);
    GetReq(DRI2CreateDrawable, req);
    req->reqType = info->codes->major_opcode;
    req->dri2ReqType = X_DRI2CreateDrawable;
    req->drawable = drawable;
    if (!_XReply(dpy, (xReply *)&rep, 0, xFalse)) {
 	UnlockDisplay(dpy);
 	SyncHandle();
 	return False;
    }
    UnlockDisplay(dpy);
    SyncHandle();
    *handle = rep.handle;
    *head = rep.head;
    return True;
 }
 void DRI2DestroyDrawable(Display *dpy, XID drawable)
 {
    XExtDisplayInfo *info = DRI2FindDisplay(dpy);
    xDRI2DestroyDrawableReq *req;
    XextSimpleCheckExtension (dpy, info, dri2ExtensionName);
    XSync(dpy, GL_FALSE);
    LockDisplay(dpy);
    GetReq(DRI2DestroyDrawable, req);
    req->reqType = info->codes->major_opcode;
    req->dri2ReqType = X_DRI2DestroyDrawable;
    req->drawable = drawable;
    UnlockDisplay(dpy);
    SyncHandle();
 }
 Bool DRI2ReemitDrawableInfo(Display *dpy, XID drawable, unsigned int *head)
 {
    XExtDisplayInfo *info = DRI2FindDisplay(dpy);
    xDRI2ReemitDrawableInfoReply rep;
    xDRI2ReemitDrawableInfoReq *req;
    XextCheckExtension (dpy, info, dri2ExtensionName, False);
    LockDisplay(dpy);
    GetReq(DRI2ReemitDrawableInfo, req);
    req->reqType = info->codes->major_opcode;
    req->dri2ReqType = X_DRI2ReemitDrawableInfo;
    req->drawable = drawable;
    if (!_XReply(dpy, (xReply *)&rep, 0, xFalse)) {
 	UnlockDisplay(dpy);
 	SyncHandle();
 	return False;
    }
    UnlockDisplay(dpy);
    SyncHandle();
    *head = rep.head;
    return True;
 }
--- a/src/glx/x11/dri2.h
+++ b/src/glx/x11/dri2.h
@@ -1,53 +0,0 @@
 /*
 * Copyright © 2007,2008 Red Hat, Inc.
 *
 * Permission is hereby granted, free of charge, to any person obtaining a
 * copy of this software and associated documentation files (the "Soft-
 * ware"), to deal in the Software without restriction, including without
 * limitation the rights to use, copy, modify, merge, publish, distribute,
 * and/or sell copies of the Software, and to permit persons to whom the
 * Software is furnished to do so, provided that the above copyright
 * notice(s) and this permission notice appear in all copies of the Soft-
 * ware and that both the above copyright notice(s) and this permission
 * notice appear in supporting documentation.
 *
 * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS
 * OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABIL-
 * ITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT OF THIRD PARTY
 * RIGHTS. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR HOLDERS INCLUDED IN
 * THIS NOTICE BE LIABLE FOR ANY CLAIM, OR ANY SPECIAL INDIRECT OR CONSE-
 * QUENTIAL DAMAGES, OR ANY DAMAGES WHATSOEVER RESULTING FROM LOSS OF USE,
 * DATA OR PROFITS, WHETHER IN AN ACTION OF CONTRACT, NEGLIGENCE OR OTHER
 * TORTIOUS ACTION, ARISING OUT OF OR IN CONNECTION WITH THE USE OR PERFOR-
 * MANCE OF THIS SOFTWARE.
 *
 * Except as contained in this notice, the name of a copyright holder shall
 * not be used in advertising or otherwise to promote the sale, use or
 * other dealings in this Software without prior written authorization of
 * the copyright holder.
 *
 * Authors:
 *   Kristian Høgsberg (krh@redhat.com)
 */
 #ifndef _DRI2_H_
 #define _DRI2_H_
 extern Bool
 DRI2QueryExtension(Display *display, int *eventBase, int *errorBase);
 extern Bool
 DRI2QueryVersion(Display *display, int *major, int *minor);
 extern Bool
 DRI2Connect(Display *display, int screen,
 	    char **driverName, char **busId, unsigned int *sareaHandle);
 extern Bool
 DRI2AuthConnection(Display *display, int screen, drm_magic_t magic);
 extern Bool
 DRI2CreateDrawable(Display *display, XID drawable,
 		   unsigned int *handle, unsigned int *head);
 extern void
 DRI2DestroyDrawable(Display *display, XID handle);
 extern Bool
 DRI2ReemitDrawableInfo(Display *dpy, XID handle, unsigned int *head);
 #endif
--- a/src/glx/x11/dri2_glx.c
+++ b/src/glx/x11/dri2_glx.c
@@ -1,371 +0,0 @@
 /*
 * Copyright © 2008 Red Hat, Inc.
 *
 * Permission is hereby granted, free of charge, to any person obtaining a
 * copy of this software and associated documentation files (the "Soft-
 * ware"), to deal in the Software without restriction, including without
 * limitation the rights to use, copy, modify, merge, publish, distribute,
 * and/or sell copies of the Software, and to permit persons to whom the
 * Software is furnished to do so, provided that the above copyright
 * notice(s) and this permission notice appear in all copies of the Soft-
 * ware and that both the above copyright notice(s) and this permission
 * notice appear in supporting documentation.
 *
 * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS
 * OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABIL-
 * ITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT OF THIRD PARTY
 * RIGHTS. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR HOLDERS INCLUDED IN
 * THIS NOTICE BE LIABLE FOR ANY CLAIM, OR ANY SPECIAL INDIRECT OR CONSE-
 * QUENTIAL DAMAGES, OR ANY DAMAGES WHATSOEVER RESULTING FROM LOSS OF USE,
 * DATA OR PROFITS, WHETHER IN AN ACTION OF CONTRACT, NEGLIGENCE OR OTHER
 * TORTIOUS ACTION, ARISING OUT OF OR IN CONNECTION WITH THE USE OR PERFOR-
 * MANCE OF THIS SOFTWARE.
 *
 * Except as contained in this notice, the name of a copyright holder shall
 * not be used in advertising or otherwise to promote the sale, use or
 * other dealings in this Software without prior written authorization of
 * the copyright holder.
 *
 * Authors:
 *   Kristian Høgsberg (krh@redhat.com)
 */
 #ifdef GLX_DIRECT_RENDERING
 #include <X11/Xlib.h>
 #include <X11/extensions/Xfixes.h>
 #include <X11/extensions/Xdamage.h>
 #include "glheader.h"
 #include "glxclient.h"
 #include "glcontextmodes.h"
 #include "xf86dri.h"
 #include "sarea.h"
 #include <dlfcn.h>
 #include <sys/types.h>
 #include <sys/mman.h>
 #include "xf86drm.h"
 #include "dri2.h"
 #include "dri_common.h"
 typedef struct __GLXDRIdisplayPrivateRec __GLXDRIdisplayPrivate;
 typedef struct __GLXDRIcontextPrivateRec __GLXDRIcontextPrivate;
 struct __GLXDRIdisplayPrivateRec {
    __GLXDRIdisplay base;
    /*
    ** XFree86-DRI version information
    */
    int driMajor;
    int driMinor;
    int driPatch;
 };
 struct __GLXDRIcontextPrivateRec {
    __GLXDRIcontext base;
    __DRIcontext *driContext;
    __GLXscreenConfigs *psc;
 };
 static void dri2DestroyContext(__GLXDRIcontext *context,
 			      __GLXscreenConfigs *psc, Display *dpy)
 {
    __GLXDRIcontextPrivate *pcp = (__GLXDRIcontextPrivate *) context;
    const __DRIcoreExtension *core = pcp->psc->core;
    (*core->destroyContext)(pcp->driContext);
    Xfree(pcp);
 }
 static Bool dri2BindContext(__GLXDRIcontext *context,
 			   __GLXDRIdrawable *draw, __GLXDRIdrawable *read)
 {
    __GLXDRIcontextPrivate *pcp = (__GLXDRIcontextPrivate *) context;
    const __DRIcoreExtension *core = pcp->psc->core;
    return (*core->bindContext)(pcp->driContext,
 				draw->driDrawable,
 				read->driDrawable);
 }
 static void dri2UnbindContext(__GLXDRIcontext *context)
 {
    __GLXDRIcontextPrivate *pcp = (__GLXDRIcontextPrivate *) context;
    const __DRIcoreExtension *core = pcp->psc->core;
    (*core->unbindContext)(pcp->driContext);
 }
 static __GLXDRIcontext *dri2CreateContext(__GLXscreenConfigs *psc,
 					 const __GLcontextModes *mode,
 					 GLXContext gc,
 					 GLXContext shareList, int renderType)
 {
    __GLXDRIcontextPrivate *pcp, *pcp_shared;
    __GLXDRIconfigPrivate *config = (__GLXDRIconfigPrivate *) mode;
    const __DRIcoreExtension *core = psc->core;
    __DRIcontext *shared = NULL;
    if (shareList) {
 	pcp_shared = (__GLXDRIcontextPrivate *) shareList->driContext;
 	shared = pcp_shared->driContext;
    }
    pcp = Xmalloc(sizeof *pcp);
    if (pcp == NULL)
 	return NULL;
    pcp->psc = psc;
    pcp->driContext = 
 	(*core->createNewContext)(psc->__driScreen,
 				  config->driConfig, shared, pcp);
    gc->__driContext = pcp->driContext;
    if (pcp->driContext == NULL) {
 	Xfree(pcp);
 	return NULL;
    }
    pcp->base.destroyContext = dri2DestroyContext;
    pcp->base.bindContext = dri2BindContext;
    pcp->base.unbindContext = dri2UnbindContext;
    return &pcp->base;
 }
 static void dri2DestroyDrawable(__GLXDRIdrawable *pdraw)
 {
    const __DRIcoreExtension *core = pdraw->psc->core;
    (*core->destroyDrawable)(pdraw->driDrawable);
    DRI2DestroyDrawable(pdraw->psc->dpy, pdraw->drawable);
    Xfree(pdraw);
 }
 static __GLXDRIdrawable *dri2CreateDrawable(__GLXscreenConfigs *psc,
 					    XID xDrawable,
 					    GLXDrawable drawable,
 					    const __GLcontextModes *modes)
 {
    __GLXDRIdrawable *pdraw;
    __GLXDRIconfigPrivate *config = (__GLXDRIconfigPrivate *) modes;
    unsigned int handle, head;
    const __DRIcoreExtension *core = psc->core;
    pdraw = Xmalloc(sizeof(*pdraw));
    if (!pdraw)
 	return NULL;
    pdraw->destroyDrawable = dri2DestroyDrawable;
    pdraw->xDrawable = xDrawable;
    pdraw->drawable = drawable;
    pdraw->psc = psc;
    fprintf(stderr, "calling DRI2CreateDrawable, XID 0x%lx, GLX ID 0x%lx\n",
 	    xDrawable, drawable);
    if (!DRI2CreateDrawable(psc->dpy, xDrawable, &handle, &head)) {
 	Xfree(pdraw);
 	return NULL;
    }
    fprintf(stderr, "success, head 0x%x, handle 0x%x\n", head, handle);
    /* Create a new drawable */
    pdraw->driDrawable =
 	(*core->createNewDrawable)(psc->__driScreen,
 				   config->driConfig,
 				   handle,
 				   head,
 				   pdraw);
    if (!pdraw->driDrawable) {
 	DRI2DestroyDrawable(psc->dpy, drawable);
 	Xfree(pdraw);
 	return NULL;
    }
    return pdraw;
 }
 static void dri2DestroyScreen(__GLXscreenConfigs *psc)
 {
    /* Free the direct rendering per screen data */
    (*psc->core->destroyScreen)(psc->__driScreen);
    drmClose(psc->fd);
    psc->__driScreen = NULL;
 }
 static void dri2ReemitDrawableInfo(__DRIdrawable *draw, unsigned int *tail,
 				    void *loaderPrivate)
 {
    __GLXDRIdrawable *pdraw = loaderPrivate;
    DRI2ReemitDrawableInfo(pdraw->psc->dpy, pdraw->drawable, tail);
 }
 static void dri2PostDamage(__DRIdrawable *draw,
 			   struct drm_clip_rect *rects,
 			   int numRects, void *loaderPrivate)
 { 
    XRectangle *xrects;
    XserverRegion region;
    __GLXDRIdrawable *glxDraw = loaderPrivate;
    __GLXscreenConfigs *psc = glxDraw->psc;
    Display *dpy = psc->dpy;
    int i;
    xrects = malloc(sizeof(XRectangle) * numRects);
    if (xrects == NULL)
 	return;
    for (i = 0; i < numRects; i++) {
 	xrects[i].x = rects[i].x1;
 	xrects[i].y = rects[i].y1;
 	xrects[i].width = rects[i].x2 - rects[i].x1;
 	xrects[i].height = rects[i].y2 - rects[i].y1;
    }
    region = XFixesCreateRegion(dpy, xrects, numRects);
    free(xrects);
    XDamageAdd(dpy, glxDraw->xDrawable, region);
    XFixesDestroyRegion(dpy, region);
 }
 static const __DRIloaderExtension dri2LoaderExtension = {
    { __DRI_LOADER, __DRI_LOADER_VERSION },
    dri2ReemitDrawableInfo,
    dri2PostDamage
 };
 static const __DRIextension *loader_extensions[] = {
    &dri2LoaderExtension.base,
    &systemTimeExtension.base,
    NULL
 };
 static __GLXDRIscreen *dri2CreateScreen(__GLXscreenConfigs *psc, int screen,
 					__GLXdisplayPrivate *priv)
 {
    const __DRIconfig **driver_configs;
    const __DRIextension **extensions;
    __GLXDRIscreen *psp;
    unsigned int sareaHandle;
    char *driverName, *busID;
    drm_magic_t magic;
    int i;
    psp = Xmalloc(sizeof *psp);
    if (psp == NULL)
 	return NULL;
    /* Initialize per screen dynamic client GLX extensions */
    psc->ext_list_first_time = GL_TRUE;
    if (!DRI2Connect(psc->dpy, screen, &driverName, &busID, &sareaHandle))
 	return NULL;
    psc->driver = driOpenDriver(driverName);
    if (psc->driver == NULL)
 	goto handle_error;
    extensions = dlsym(psc->driver, __DRI_DRIVER_EXTENSIONS);
    if (extensions == NULL) {
 	ErrorMessageF("driver exports no extensions (%s)\n", dlerror());
 	goto handle_error;
    }
    for (i = 0; extensions[i]; i++) {
 	if (strcmp(extensions[i]->name, __DRI_CORE) == 0)
 	    psc->core = (__DRIcoreExtension *) extensions[i];
    }
    if (psc->core == NULL) {
 	ErrorMessageF("core dri extension not found\n");
 	goto handle_error;
    }
    psc->fd = drmOpen(NULL, busID);
    if (psc->fd < 0) {
 	ErrorMessageF("failed to open drm device: %s\n", strerror(errno));
 	return NULL;
    }
    if (drmGetMagic(psc->fd, &magic))
 	return NULL;
    if (!DRI2AuthConnection(psc->dpy, screen, magic)) {
 	ErrorMessageF("failed to authenticate drm access\n");
 	return NULL;
    }
    psc->__driScreen = 
 	psc->core->createNewScreen(screen, psc->fd, sareaHandle,
 				   loader_extensions, &driver_configs, psc);
    if (psc->__driScreen == NULL) {
 	ErrorMessageF("failed to create dri screen\n");
 	return NULL;
    }
    driBindExtensions(psc, 1);
    psc->configs = driConvertConfigs(psc->core, psc->configs, driver_configs);
    psc->visuals = driConvertConfigs(psc->core, psc->visuals, driver_configs);
    psp->destroyScreen = dri2DestroyScreen;
    psp->createContext = dri2CreateContext;
    psp->createDrawable = dri2CreateDrawable;
    Xfree(driverName);
    Xfree(busID);
    return psp;
 handle_error:
    Xfree(driverName);
    Xfree(busID);
    /* FIXME: clean up here */
    return NULL;
 }
 /* Called from __glXFreeDisplayPrivate.
 */
 static void dri2DestroyDisplay(__GLXDRIdisplay *dpy)
 {
    Xfree(dpy);
 }
 /*
 * Allocate, initialize and return a __DRIdisplayPrivate object.
 * This is called from __glXInitialize() when we are given a new
 * display pointer.
 */
 _X_HIDDEN __GLXDRIdisplay *dri2CreateDisplay(Display *dpy)
 {
    __GLXDRIdisplayPrivate *pdp;
    int eventBase, errorBase;
    if (!DRI2QueryExtension(dpy, &eventBase, &errorBase))
 	return NULL;
    pdp = Xmalloc(sizeof *pdp);
    if (pdp == NULL)
 	return NULL;
    if (!DRI2QueryVersion(dpy, &pdp->driMajor, &pdp->driMinor)) {
 	Xfree(pdp);
 	return NULL;
    }
    pdp->driPatch = 0;
    pdp->base.destroyDisplay = dri2DestroyDisplay;
    pdp->base.createScreen = dri2CreateScreen;
    return &pdp->base;
 }
 #endif /* GLX_DIRECT_RENDERING */
--- a/src/glx/x11/glxclient.h
+++ b/src/glx/x11/glxclient.h
@@ -151,7 +151,6 @@ struct __GLXDRIdrawableRec {
 */
 extern __GLXDRIdisplay *driswCreateDisplay(Display *dpy);
 extern __GLXDRIdisplay *driCreateDisplay(Display *dpy);
 extern __GLXDRIdisplay *dri2CreateDisplay(Display *dpy);
 extern void DRI_glXUseXFont( Font font, int first, int count, int listbase );
@@ -570,7 +569,6 @@ struct __GLXdisplayPrivateRec {
     */
    __GLXDRIdisplay *driswDisplay;
    __GLXDRIdisplay *driDisplay;
    __GLXDRIdisplay *dri2Display;
 #endif
 };
--- a/src/glx/x11/glxext.c
+++ b/src/glx/x11/glxext.c
@@ -207,10 +207,6 @@ static int __glXFreeDisplayPrivate(XExtData *extension)
    if (priv->driDisplay)
 	(*priv->driDisplay->destroyDisplay)(priv->driDisplay);
    priv->driDisplay = NULL;
    if (priv->dri2Display)
 	(*priv->dri2Display->destroyDisplay)(priv->dri2Display);
    priv->dri2Display = NULL;
 #endif
    Xfree((char*) priv);
@@ -608,9 +604,6 @@ static Bool AllocAndFetchScreenConfigs(Display *dpy, __GLXdisplayPrivate *priv)
 	if (psc->drawHash == NULL)
 	    continue;
 	if (priv->dri2Display)
 	    psc->driScreen = (*priv->dri2Display->createScreen)(psc, i, priv);
 	if (psc->driScreen == NULL && priv->driDisplay)
 	    psc->driScreen = (*priv->driDisplay->createScreen)(psc, i, priv);
@@ -716,7 +709,6 @@ _X_HIDDEN __GLXdisplayPrivate *__glXInitialize(Display* dpy)
    ** (e.g., those called in AllocAndFetchScreenConfigs).
    */
    if (glx_direct && glx_accel) {
 	dpyPriv->dri2Display = dri2CreateDisplay(dpy);
 	dpyPriv->driDisplay = driCreateDisplay(dpy);
    }
    if (glx_direct)
--- a/src/glx/x11/indirect_vertex_program.c
+++ b/src/glx/x11/indirect_vertex_program.c
@@ -193,7 +193,13 @@ void __indirect_glGetVertexAttribivARB( GLuint index, GLenum pname,
    get_vertex_attrib( gc, 1303, index, pname, (xReply *) & reply );
    if ( reply.size != 0 ) {
-	if ( ! get_attrib_array_data( state, index, pname, params ) ) {
+	GLintptr data;
 	if ( get_attrib_array_data( state, index, pname, & data ) ) {
 	    *params = (GLint) data;
 	}
 	else {
 	    if (reply.size == 1) {
 		*params = (GLint) reply.pad3;
 	    } 
--- a/src/mesa/Makefile.mgw
+++ b/src/mesa/Makefile.mgw
@@ -154,7 +154,7 @@ DRIVER_SOURCES = \
 endif
 endif
-SOURCES = $(CORE_SOURCES) $(X86_SOURCES) $(COMMON_DRIVER_SOURCES) $(DRIVER_SOURCES)
+SOURCES = $(MESA_SOURCES) $(GLAPI_SOURCES) $(X86_SOURCES) $(DRIVER_SOURCES)
 OBJECTS = $(addsuffix .o,$(basename $(SOURCES)))
--- a/src/mesa/drivers/dri/i915/i915_debug.c
+++ b/src/mesa/drivers/dri/i915/i915_debug.c
@@ -177,7 +177,7 @@ static GLboolean debug_variable_length_prim( struct debug_stream *stream )
 #define BITS( dw, hi, lo, ... )				\
 do {							\
-   unsigned himask = ~0UL >> (31 - (hi));		\
+   unsigned himask = 0xffffffffU >> (31 - (hi));		\
   PRINTF("\t\t ");				\
   PRINTF(__VA_ARGS__);			\
   PRINTF(": 0x%x\n", ((dw) & himask) >> (lo));	\
--- a/src/mesa/drivers/dri/i915/i915_texstate.c
+++ b/src/mesa/drivers/dri/i915/i915_texstate.c
@@ -78,7 +78,7 @@ translate_texture_format(GLuint mesa_format, GLenum DepthMode)
      return (MAPSURF_COMPRESSED | MT_COMPRESS_DXT2_3);
   case MESA_FORMAT_RGBA_DXT5:
      return (MAPSURF_COMPRESSED | MT_COMPRESS_DXT4_5);
-   case MESA_FORMAT_Z24_S8:
+   case MESA_FORMAT_S8_Z24:
      return (MAPSURF_32BIT | MT_32BIT_xI824);
   default:
      fprintf(stderr, "%s: bad image format %x\n", __FUNCTION__, mesa_format);
--- a/src/mesa/drivers/dri/i965/brw_eu_emit.c
+++ b/src/mesa/drivers/dri/i965/brw_eu_emit.c
@@ -513,6 +513,8 @@ struct brw_instruction *brw_IF(struct brw_compile *p, GLuint execute_size)
   insn->header.compression_control = BRW_COMPRESSION_NONE;
   insn->header.predicate_control = BRW_PREDICATE_NORMAL;
   insn->header.mask_control = BRW_MASK_ENABLE;
   if (!p->single_program_flow)
       insn->header.thread_control = BRW_THREAD_SWITCH;
   p->current->header.predicate_control = BRW_PREDICATE_NONE;
@@ -538,6 +540,8 @@ struct brw_instruction *brw_ELSE(struct brw_compile *p,
   insn->header.compression_control = BRW_COMPRESSION_NONE;
   insn->header.execution_size = if_insn->header.execution_size;
   insn->header.mask_control = BRW_MASK_ENABLE;
   if (!p->single_program_flow)
       insn->header.thread_control = BRW_THREAD_SWITCH;
   /* Patch the if instruction to point at this instruction.
    */
@@ -579,6 +583,7 @@ void brw_ENDIF(struct brw_compile *p,
      insn->header.compression_control = BRW_COMPRESSION_NONE;
      insn->header.execution_size = patch_insn->header.execution_size;
      insn->header.mask_control = BRW_MASK_ENABLE;
      insn->header.thread_control = BRW_THREAD_SWITCH;
      assert(patch_insn->bits3.if_else.jump_count == 0);
@@ -617,7 +622,7 @@ struct brw_instruction *brw_BREAK(struct brw_compile *p)
   brw_set_src1(insn, brw_imm_d(0x0));
   insn->header.compression_control = BRW_COMPRESSION_NONE;
   insn->header.execution_size = BRW_EXECUTE_8;
-   insn->header.mask_control = BRW_MASK_DISABLE;
+   /* insn->header.mask_control = BRW_MASK_DISABLE; */
   insn->bits3.if_else.pad0 = 0;
   return insn;
 }
@@ -631,7 +636,7 @@ struct brw_instruction *brw_CONT(struct brw_compile *p)
   brw_set_src1(insn, brw_imm_d(0x0));
   insn->header.compression_control = BRW_COMPRESSION_NONE;
   insn->header.execution_size = BRW_EXECUTE_8;
-   insn->header.mask_control = BRW_MASK_DISABLE;
+   /* insn->header.mask_control = BRW_MASK_DISABLE; */
   insn->bits3.if_else.pad0 = 0;
   return insn;
 }
@@ -655,7 +660,7 @@ struct brw_instruction *brw_DO(struct brw_compile *p, GLuint execute_size)
      insn->header.execution_size = execute_size;
      insn->header.predicate_control = BRW_PREDICATE_NONE;
      /* insn->header.mask_control = BRW_MASK_ENABLE; */
-      insn->header.mask_control = BRW_MASK_DISABLE;
+      /* insn->header.mask_control = BRW_MASK_DISABLE; */
      return insn;
   }
@@ -694,7 +699,7 @@ struct brw_instruction *brw_WHILE(struct brw_compile *p,
 /*    insn->header.mask_control = BRW_MASK_ENABLE; */
-   insn->header.mask_control = BRW_MASK_DISABLE;
+   /* insn->header.mask_control = BRW_MASK_DISABLE; */
   p->current->header.predicate_control = BRW_PREDICATE_NONE;   
   return insn;
 }
--- a/src/mesa/drivers/dri/i965/brw_wm_glsl.c
+++ b/src/mesa/drivers/dri/i965/brw_wm_glsl.c
@@ -850,20 +850,20 @@ static void emit_sop(struct brw_wm_compile *c,
    struct brw_reg dst, src0, src1;
    int i;
    brw_push_insn_state(p);
    for (i = 0; i < 4; i++) {
 	if (mask & (1<<i)) {
 	    dst = get_dst_reg(c, inst, i, 1);
 	    src0 = get_src_reg(c, &inst->SrcReg[0], i, 1);
 	    src1 = get_src_reg(c, &inst->SrcReg[1], i, 1);
 	    brw_push_insn_state(p);
 	    brw_CMP(p, brw_null_reg(), cond, src0, src1);
 	    brw_set_predicate_control(p, BRW_PREDICATE_NONE);
 	    brw_MOV(p, dst, brw_imm_f(0.0));
 	    brw_set_predicate_control(p, BRW_PREDICATE_NORMAL);
 	    brw_MOV(p, dst, brw_imm_f(1.0));
 	    brw_pop_insn_state(p);
 	}
    }
    brw_pop_insn_state(p);
 }
 static void emit_slt(struct brw_wm_compile *c,
--- a/src/mesa/drivers/dri/i965/brw_wm_surface_state.c
+++ b/src/mesa/drivers/dri/i965/brw_wm_surface_state.c
@@ -138,7 +138,7 @@ static GLuint translate_tex_format( GLuint mesa_format, GLenum depth_mode )
   case MESA_FORMAT_SRGB_DXT1:
      return BRW_SURFACEFORMAT_BC1_UNORM_SRGB;
-   case MESA_FORMAT_Z24_S8:
+   case MESA_FORMAT_S8_Z24:
      return BRW_SURFACEFORMAT_I24X8_UNORM;
   default:
--- a/src/mesa/drivers/dri/intel/intel_chipset.h
+++ b/src/mesa/drivers/dri/intel/intel_chipset.h
@@ -58,6 +58,7 @@
 #define PCI_CHIP_IGD_E_G                0x2E02
 #define PCI_CHIP_Q45_G                  0x2E12
 #define PCI_CHIP_G45_G                  0x2E22
 #define PCI_CHIP_G41_G                  0x2E32
 #define IS_MOBILE(devid)	(devid == PCI_CHIP_I855_GM || \
 				 devid == PCI_CHIP_I915_GM || \
@@ -70,7 +71,8 @@
 #define IS_GM45_GM(devid)       (devid == PCI_CHIP_GM45_GM)
 #define IS_G4X(devid)           (devid == PCI_CHIP_IGD_E_G || \
                                 devid == PCI_CHIP_Q45_G || \
-                                 devid == PCI_CHIP_G45_G)
+                                 devid == PCI_CHIP_G45_G || \
                                 devid == PCI_CHIP_G41_G)
 #define IS_915(devid)		(devid == PCI_CHIP_I915_G || \
 				 devid == PCI_CHIP_E7221_G || \
--- a/src/mesa/drivers/dri/intel/intel_context.c
+++ b/src/mesa/drivers/dri/intel/intel_context.c
@@ -178,6 +178,9 @@ intelGetString(GLcontext * ctx, GLenum name)
      case PCI_CHIP_Q45_G:
         chipset = "Intel(R) Q45/Q43";
         break;
      case PCI_CHIP_G41_G:
         chipset = "Intel(R) G41";
         break;
      default:
         chipset = "Unknown Intel Chipset";
         break;
--- a/src/mesa/drivers/dri/intel/intel_tex_format.c
+++ b/src/mesa/drivers/dri/intel/intel_tex_format.c
@@ -138,7 +138,7 @@ intelChooseTextureFormat(GLcontext * ctx, GLint internalFormat,
   case GL_DEPTH_STENCIL_EXT:
   case GL_DEPTH24_STENCIL8_EXT:
-      return &_mesa_texformat_z24_s8;
+      return &_mesa_texformat_s8_z24;
 #ifndef I915
   case GL_SRGB_EXT:
--- a/src/mesa/drivers/dri/tdfx/tdfx_screen.c
+++ b/src/mesa/drivers/dri/tdfx/tdfx_screen.c
@@ -65,6 +65,7 @@ DRI_CONF_END;
 static const __DRIextension *tdfxExtensions[] = {
    &driReadDrawableExtension,
    NULL
 };
 static const GLuint __driNConfigOptions = 1;
@@ -350,12 +351,7 @@ tdfxFillInModes(__DRIscreenPrivate *psp,
 		unsigned stencil_bits,
 		GLboolean have_back_buffer)
 {
 	__DRIconfig **configs, **c;
 	__GLcontextModes *m;
 	unsigned num_modes;
 	unsigned vis[2] = { GLX_TRUE_COLOR, GLX_DIRECT_COLOR };
 	unsigned deep = (depth_bits > 17);
 	unsigned i, db, depth, accum, stencil;
 	/* Right now GLX_SWAP_COPY_OML isn't supported, but it would be easy
 	 * enough to add support.  Basically, if a context is created with an
@@ -363,55 +359,32 @@ tdfxFillInModes(__DRIscreenPrivate *psp,
 	 * will never be used.
 	 */
-	num_modes = (depth_bits == 16) ? 32 : 16;
+	static const GLenum db_modes[2] = { GLX_NONE, GLX_SWAP_UNDEFINED_OML };
-
+	uint8_t depth_bits_array[4];
-	configs = _mesa_malloc(num_modes * sizeof *configs);
+	uint8_t stencil_bits_array[4];
-	c = configs;
+	if(deep) {
-
+		depth_bits_array[0] = 0;
-	for (i = 0; i <= 1; i++) {
+		depth_bits_array[1] = 24;
-	    for (db = 0; db <= 1; db++) {
+		stencil_bits_array[0] = 0;
-		for (depth = 0; depth <= 1; depth++) {
+		stencil_bits_array[1] = 8;
-		    for (accum = 0; accum <= 1; accum++) {
+	} else {
-			for (stencil = 0; stencil <= !deep; stencil++) {
+		depth_bits_array[0] = depth_bits;
-			    *c = _mesa_malloc(sizeof **c);
+		depth_bits_array[1] = 0;
-			    m = &(*c++)->modes;
+		depth_bits_array[2] = depth_bits;
-			    if (deep) stencil = depth;
+		depth_bits_array[3] = 0;
-			    m->redBits		= deep ? 8 : 5;
+		stencil_bits_array[0] = 0;
-			    m->greenBits	= deep ? 8 : 6;
+		stencil_bits_array[1] = 0;
-			    m->blueBits		= deep ? 8 : 5;
+		stencil_bits_array[2] = 8;
-			    m->alphaBits	= deep ? 8 : 0;
+		stencil_bits_array[3] = 8;
 			    m->redMask		= deep ?0xFF000000 :0x0000F800;
 			    m->greenMask	= deep ?0x00FF0000 :0x000007E0;
 			    m->blueMask		= deep ?0x0000FF00 :0x0000001F;
 			    m->alphaMask	= deep ? 0x000000FF : 0;
 			    m->rgbBits		= m->redBits + m->greenBits +
 			    			  m->blueBits + m->alphaBits;
 			    m->accumRedBits	= accum ? 16 : 0;
 			    m->accumGreenBits	= accum ? 16 : 0;
 			    m->accumBlueBits	= accum ? 16 : 0;
 			    m->accumAlphaBits	= (accum && deep) ? 16 : 0;
 			    m->stencilBits	= stencil ? 8 : 0;
 			    m->depthBits	= deep
 			    			  ? (depth ? 24 : 0)
 			    			  : (depth ? 0 : depth_bits);
 			    m->visualType	= vis[i];
 			    m->renderType	= GLX_RGBA_BIT;
 			    m->drawableType	= GLX_WINDOW_BIT;
 			    m->rgbMode		= GL_TRUE;
 			    m->doubleBufferMode = db ? GL_TRUE : GL_FALSE;
 			    if (db)
 			    	m->swapMethod = GLX_SWAP_UNDEFINED_OML;
 			    m->visualRating	= ((stencil && !deep) || accum)
 			    			  ? GLX_SLOW_CONFIG
 						  : GLX_NONE;
 			    if (deep) stencil = 0;
 			}
 		    }
 		}
 	    }
 	}
-	return (const __DRIconfig **) configs;
+	return driCreateConfigs(
 		deep ? GL_RGBA : GL_RGB,
 		deep ? GL_UNSIGNED_INT_8_8_8_8 : GL_UNSIGNED_SHORT_5_6_5,
 		depth_bits_array,
 		stencil_bits_array,
 		deep ? 2 : 4,
 		db_modes, 2);
 }
 /**
--- a/src/mesa/drivers/windows/gdi/wmesa.c
+++ b/src/mesa/drivers/windows/gdi/wmesa.c
@@ -118,6 +118,7 @@ static void wmSetPixelFormat(WMesaFramebuffer pwfb, HDC hDC)
    /* Only 16 and 32 bit targets are supported now */
    assert(pwfb->cColorBits == 0 ||
 	   pwfb->cColorBits == 16 || 
 	   pwfb->cColorBits == 24 || 
 	   pwfb->cColorBits == 32);
    switch(pwfb->cColorBits){
@@ -127,6 +128,7 @@ static void wmSetPixelFormat(WMesaFramebuffer pwfb, HDC hDC)
    case 16:
 	pwfb->pixelformat = PF_5R6G5B;
 	break;
    case 24:
    case 32:
 	pwfb->pixelformat = PF_8R8G8B;
 	break;
@@ -854,6 +856,195 @@ static void read_rgba_pixels_32(const GLcontext *ctx,
 }
 /*********************************************************************/
 /* DOUBLE BUFFER 24-bit */
 #define WMSETPIXEL24(pwc, y, x, r, g, b) { \
 LPBYTE lpb = ((LPBYTE)((pwc)->pbPixels + (pwc)->ScanWidth * (y)) + (3 * x)); \
 lpb[0] = (b); \
 lpb[1] = (g); \
 lpb[2] = (r); }
 /* Write a horizontal span of RGBA color pixels with a boolean mask. */
 static void write_rgba_span_24(const GLcontext *ctx, 
 			       struct gl_renderbuffer *rb, 
 			       GLuint n, GLint x, GLint y,
 			       const GLubyte rgba[][4], 
 			       const GLubyte mask[] )
 {
    WMesaContext pwc = wmesa_context(ctx);
    WMesaFramebuffer pwfb = wmesa_framebuffer(ctx->DrawBuffer);
    GLuint i;
    LPBYTE lpb;
    (void) ctx;
    y=FLIP(y);
    lpb = ((LPBYTE)(pwfb->pbPixels + pwfb->ScanWidth * y)) + (3 * x);
    if (mask) {
 	for (i=0; i<n; i++)
 	    if (mask[i]) {
                lpb[3*i] = rgba[i][BCOMP];
                lpb[3*i+1] = rgba[i][GCOMP];
                lpb[3*i+2] = rgba[i][RCOMP];
 	    }
    }
    else {
 	    for (i=0; i<n; i++) {
            *lpb++ = rgba[i][BCOMP];
            *lpb++ = rgba[i][GCOMP];
            *lpb++ = rgba[i][RCOMP];
 	    }
    }
 }
 /* Write a horizontal span of RGB color pixels with a boolean mask. */
 static void write_rgb_span_24(const GLcontext *ctx, 
 			      struct gl_renderbuffer *rb, 
 			      GLuint n, GLint x, GLint y,
 			      const GLubyte rgb[][3], 
 			      const GLubyte mask[] )
 {
    WMesaContext pwc = wmesa_context(ctx);
    WMesaFramebuffer pwfb = wmesa_framebuffer(ctx->DrawBuffer);
    GLuint i;
    LPBYTE lpb;
    (void) ctx;
    y=FLIP(y);
    lpb = ((LPBYTE)(pwfb->pbPixels + pwfb->ScanWidth * y)) + (3 * x);
    if (mask) {
 	for (i=0; i<n; i++)
 	    if (mask[i]) {
            lpb[3*i] = rgb[i][BCOMP];
            lpb[3*i+1] = rgb[i][GCOMP];
            lpb[3*i+2] = rgb[i][RCOMP];
 	    }
    }
    else {
    	for (i=0; i<n; i++) {
    		*lpb++ = rgb[i][BCOMP];
    		*lpb++ = rgb[i][GCOMP];
    		*lpb++ = rgb[i][RCOMP];
    	}
    }
 }
 /*
 * Write a horizontal span of pixels with a boolean mask.  The current color
 * is used for all pixels.
 */
 static void write_mono_rgba_span_24(const GLcontext *ctx, 
 				    struct gl_renderbuffer *rb,
 				    GLuint n, GLint x, GLint y,
 				    const GLchan color[4], 
 				    const GLubyte mask[])
 {
    LPBYTE lpb;
    GLuint i;
    WMesaContext pwc = wmesa_context(ctx);
    WMesaFramebuffer pwfb = wmesa_framebuffer(ctx->DrawBuffer);
    lpb = ((LPBYTE)(pwfb->pbPixels + pwfb->ScanWidth * y)) + (3 * x);
    y=FLIP(y);
    if (mask) {
 	for (i=0; i<n; i++)
 	    if (mask[i]) {
 	    	lpb[3*i] = color[BCOMP];
 	    	lpb[3*i+1] = color[GCOMP];
 	    	lpb[3*i+2] = color[RCOMP];
 	    }
    }
    else
 	for (i=0; i<n; i++) {
 		*lpb++ = color[BCOMP];
 		*lpb++ = color[GCOMP];
 		*lpb++ = color[RCOMP];		
 	}
 }
 /* Write an array of RGBA pixels with a boolean mask. */
 static void write_rgba_pixels_24(const GLcontext *ctx, 
 				 struct gl_renderbuffer *rb,
 				 GLuint n, const GLint x[], const GLint y[],
 				 const GLubyte rgba[][4], 
 				 const GLubyte mask[])
 {
    GLuint i;
    WMesaContext pwc = wmesa_context(ctx);
    WMesaFramebuffer pwfb = wmesa_framebuffer(ctx->DrawBuffer);
    for (i=0; i<n; i++)
 	if (mask[i])
 	    WMSETPIXEL24(pwfb, FLIP(y[i]), x[i],
 			 rgba[i][RCOMP], rgba[i][GCOMP], rgba[i][BCOMP]);
 }
 /*
 * Write an array of pixels with a boolean mask.  The current color
 * is used for all pixels.
 */
 static void write_mono_rgba_pixels_24(const GLcontext *ctx, 
 				      struct gl_renderbuffer *rb,
 				      GLuint n,
 				      const GLint x[], const GLint y[],
 				      const GLchan color[4],
 				      const GLubyte mask[])
 {
    GLuint i;
    WMesaContext pwc = wmesa_context(ctx);
    WMesaFramebuffer pwfb = wmesa_framebuffer(ctx->DrawBuffer);
    for (i=0; i<n; i++)
 	if (mask[i])
 	    WMSETPIXEL24(pwfb, FLIP(y[i]),x[i],color[RCOMP],
 			 color[GCOMP], color[BCOMP]);
 }
 /* Read a horizontal span of color pixels. */
 static void read_rgba_span_24(const GLcontext *ctx, 
 			      struct gl_renderbuffer *rb,
 			      GLuint n, GLint x, GLint y,
 			      GLubyte rgba[][4] )
 {
    GLuint i;
    LPBYTE lpb;
    WMesaContext pwc = wmesa_context(ctx);
    WMesaFramebuffer pwfb = wmesa_framebuffer(ctx->DrawBuffer);
    y = FLIP(y);
    lpb = ((LPBYTE)(pwfb->pbPixels + pwfb->ScanWidth * y)) + (3 * x);
    for (i=0; i<n; i++) {
 	rgba[i][RCOMP] = lpb[3*i+2];
 	rgba[i][GCOMP] = lpb[3*i+1];
 	rgba[i][BCOMP] = lpb[3*i];
 	rgba[i][ACOMP] = 255;
    }
 }
 /* Read an array of color pixels. */
 static void read_rgba_pixels_24(const GLcontext *ctx, 
 				struct gl_renderbuffer *rb,
 				GLuint n, const GLint x[], const GLint y[],
 				GLubyte rgba[][4])
 {
    GLuint i;
    LPBYTE lpb;
    WMesaContext pwc = wmesa_context(ctx);
    WMesaFramebuffer pwfb = wmesa_framebuffer(ctx->DrawBuffer);
    for (i=0; i<n; i++) {
 	GLint y2 = FLIP(y[i]);
 	lpb = ((LPBYTE)(pwfb->pbPixels + pwfb->ScanWidth * y2)) + (3 * x[i]);
 	rgba[i][RCOMP] = lpb[3*i+2];
 	rgba[i][GCOMP] = lpb[3*i+1];
 	rgba[i][BCOMP] = lpb[3*i];
 	rgba[i][ACOMP] = 255;
  }
 }
 /*********************************************************************/
 /* DOUBLE BUFFER 16-bit */
@@ -1077,7 +1268,7 @@ wmesa_renderbuffer_storage(GLcontext *ctx,
 * on if we're drawing to the front or back color buffer.
 */
 void wmesa_set_renderbuffer_funcs(struct gl_renderbuffer *rb, int pixelformat,
-                                  int double_buffer)
+                                  BYTE cColorBits, int double_buffer)
 {
    if (double_buffer) {
        /* back buffer */
@@ -1097,16 +1288,32 @@ void wmesa_set_renderbuffer_funcs(struct gl_renderbuffer *rb, int pixelformat,
            rb->BlueBits = 5;
 	    break;
 	case PF_8R8G8B:
-	    rb->PutRow = write_rgba_span_32;
+		if (cColorBits == 24)
-	    rb->PutRowRGB = write_rgb_span_32;
+		{
-	    rb->PutMonoRow = write_mono_rgba_span_32;
+		    rb->PutRow = write_rgba_span_24;
-	    rb->PutValues = write_rgba_pixels_32;
+		    rb->PutRowRGB = write_rgb_span_24;
-	    rb->PutMonoValues = write_mono_rgba_pixels_32;
+		    rb->PutMonoRow = write_mono_rgba_span_24;
-	    rb->GetRow = read_rgba_span_32;
+		    rb->PutValues = write_rgba_pixels_24;
-	    rb->GetValues = read_rgba_pixels_32;
+		    rb->PutMonoValues = write_mono_rgba_pixels_24;
 		    rb->GetRow = read_rgba_span_24;
 		    rb->GetValues = read_rgba_pixels_24;
 	        rb->RedBits = 8;
 	        rb->GreenBits = 8;
 	        rb->BlueBits = 8;		
 		}
 		else
 		{
 	        rb->PutRow = write_rgba_span_32;
 	        rb->PutRowRGB = write_rgb_span_32;
 	        rb->PutMonoRow = write_mono_rgba_span_32;
 	        rb->PutValues = write_rgba_pixels_32;
 	        rb->PutMonoValues = write_mono_rgba_pixels_32;
 	        rb->GetRow = read_rgba_span_32;
 	        rb->GetValues = read_rgba_pixels_32;
            rb->RedBits = 8;
            rb->GreenBits = 8;
            rb->BlueBits = 8;
 		}
 	    break;
 	default:
 	    break;
@@ -1413,11 +1620,11 @@ void WMesaMakeCurrent(WMesaContext c, HDC hdc)
        if (visual->doubleBufferMode == 1) {
            rb = wmesa_new_renderbuffer();
            _mesa_add_renderbuffer(&pwfb->Base, BUFFER_BACK_LEFT, rb);
-            wmesa_set_renderbuffer_funcs(rb, pwfb->pixelformat, 1);
+            wmesa_set_renderbuffer_funcs(rb, pwfb->pixelformat, pwfb->cColorBits, 1);
 	}
        rb = wmesa_new_renderbuffer();
        _mesa_add_renderbuffer(&pwfb->Base, BUFFER_FRONT_LEFT, rb);
-        wmesa_set_renderbuffer_funcs(rb, pwfb->pixelformat, 0);
+        wmesa_set_renderbuffer_funcs(rb, pwfb->pixelformat, pwfb->cColorBits, 0);
 	/* Let Mesa own the Depth, Stencil, and Accum buffers */
        _mesa_add_soft_renderbuffers(&pwfb->Base,
--- a/src/mesa/glapi/glthread.c
+++ b/src/mesa/glapi/glthread.c
@@ -34,6 +34,7 @@
 #endif
 #include <stdlib.h>
 #include <stdio.h>
 #include "glthread.h"
--- a/src/mesa/main/arrayobj.c
+++ b/src/mesa/main/arrayobj.c
@@ -1,8 +1,8 @@
 /*
 * Mesa 3-D graphics library
- * Version:  6.5
+ * Version:  7.2
 *
- * Copyright (C) 1999-2004  Brian Paul   All Rights Reserved.
+ * Copyright (C) 1999-2008  Brian Paul   All Rights Reserved.
 * (C) Copyright IBM Corporation 2006
 *
 * Permission is hereby granted, free of charge, to any person obtaining a
@@ -77,7 +77,7 @@ lookup_arrayobj(GLcontext *ctx, GLuint id)
 struct gl_array_object *
 _mesa_new_array_object( GLcontext *ctx, GLuint name )
 {
-   struct gl_array_object *obj = MALLOC_STRUCT(gl_array_object);
+   struct gl_array_object *obj = CALLOC_STRUCT(gl_array_object);
   if (obj)
      _mesa_initialize_array_object(ctx, obj, name);
   return obj;
@@ -210,6 +210,15 @@ _mesa_remove_array_object( GLcontext *ctx, struct gl_array_object *obj )
 }
 static void
 unbind_buffer_object( GLcontext *ctx, struct gl_buffer_object *bufObj )
 {
   if (bufObj != ctx->Array.NullBufferObj) {
      _mesa_reference_buffer_object(ctx, &bufObj, NULL);
   }
 }
 /**********************************************************************/
 /* API Functions                                                      */
 /**********************************************************************/
@@ -311,18 +320,18 @@ _mesa_DeleteVertexArraysAPPLE(GLsizei n, const GLuint *ids)
 	 /* Unbind any buffer objects that might be bound to arrays in
 	  * this array object.
 	  */
-	 _mesa_unbind_buffer_object( ctx, obj->Vertex.BufferObj );
+	 unbind_buffer_object( ctx, obj->Vertex.BufferObj );
-	 _mesa_unbind_buffer_object( ctx, obj->Normal.BufferObj );
+	 unbind_buffer_object( ctx, obj->Normal.BufferObj );
-	 _mesa_unbind_buffer_object( ctx, obj->Color.BufferObj );
+	 unbind_buffer_object( ctx, obj->Color.BufferObj );
-	 _mesa_unbind_buffer_object( ctx, obj->SecondaryColor.BufferObj );
+	 unbind_buffer_object( ctx, obj->SecondaryColor.BufferObj );
-	 _mesa_unbind_buffer_object( ctx, obj->FogCoord.BufferObj );
+	 unbind_buffer_object( ctx, obj->FogCoord.BufferObj );
-	 _mesa_unbind_buffer_object( ctx, obj->Index.BufferObj );
+	 unbind_buffer_object( ctx, obj->Index.BufferObj );
 	 for (i = 0; i < MAX_TEXTURE_UNITS; i++) {
-	    _mesa_unbind_buffer_object( ctx, obj->TexCoord[i].BufferObj );
+	    unbind_buffer_object( ctx, obj->TexCoord[i].BufferObj );
 	 }
-	 _mesa_unbind_buffer_object( ctx, obj->EdgeFlag.BufferObj );
+	 unbind_buffer_object( ctx, obj->EdgeFlag.BufferObj );
 	 for (i = 0; i < VERT_ATTRIB_MAX; i++) {
-	    _mesa_unbind_buffer_object( ctx, obj->VertexAttrib[i].BufferObj );
+	    unbind_buffer_object( ctx, obj->VertexAttrib[i].BufferObj );
 	 }
 #endif
--- a/src/mesa/main/attrib.c
+++ b/src/mesa/main/attrib.c
@@ -1,8 +1,8 @@
 /*
 * Mesa 3-D graphics library
- * Version:  7.1
+ * Version:  7.2
 *
- * Copyright (C) 1999-2007  Brian Paul   All Rights Reserved.
+ * Copyright (C) 1999-2008  Brian Paul   All Rights Reserved.
 *
 * Permission is hereby granted, free of charge, to any person obtaining a
 * copy of this software and associated documentation files (the "Software"),
@@ -1274,6 +1274,29 @@ adjust_buffer_object_ref_counts(struct gl_array_attrib *array, GLint step)
 }
 /**
 * Copy gl_pixelstore_attrib from src to dst, updating buffer
 * object refcounts.
 */
 static void
 copy_pixelstore(GLcontext *ctx,
                struct gl_pixelstore_attrib *dst,
                const struct gl_pixelstore_attrib *src)
 {
   dst->Alignment = src->Alignment;
   dst->RowLength = src->RowLength;
   dst->SkipPixels = src->SkipPixels;
   dst->SkipRows = src->SkipRows;
   dst->ImageHeight = src->ImageHeight;
   dst->SkipImages = src->SkipImages;
   dst->SwapBytes = src->SwapBytes;
   dst->LsbFirst = src->LsbFirst;
   dst->ClientStorage = src->ClientStorage;
   dst->Invert = src->Invert;
   _mesa_reference_buffer_object(ctx, &dst->BufferObj, src->BufferObj);
 }
 #define GL_CLIENT_PACK_BIT (1<<20)
 #define GL_CLIENT_UNPACK_BIT (1<<21)
@@ -1292,31 +1315,29 @@ _mesa_PushClientAttrib(GLbitfield mask)
      return;
   }
-   /* Build linked list of attribute nodes which save all attribute */
+   /* Build linked list of attribute nodes which save all attribute
-   /* groups specified by the mask. */
+    * groups specified by the mask.
    */
   head = NULL;
   if (mask & GL_CLIENT_PIXEL_STORE_BIT) {
      struct gl_pixelstore_attrib *attr;
 #if FEATURE_EXT_pixel_buffer_object
      ctx->Pack.BufferObj->RefCount++;
      ctx->Unpack.BufferObj->RefCount++;
 #endif
      /* packing attribs */
-      attr = MALLOC_STRUCT( gl_pixelstore_attrib );
+      attr = CALLOC_STRUCT( gl_pixelstore_attrib );
-      MEMCPY( attr, &ctx->Pack, sizeof(struct gl_pixelstore_attrib) );
+      copy_pixelstore(ctx, attr, &ctx->Pack);
      newnode = new_attrib_node( GL_CLIENT_PACK_BIT );
      newnode->data = attr;
      newnode->next = head;
      head = newnode;
      /* unpacking attribs */
-      attr = MALLOC_STRUCT( gl_pixelstore_attrib );
+      attr = CALLOC_STRUCT( gl_pixelstore_attrib );
-      MEMCPY( attr, &ctx->Unpack, sizeof(struct gl_pixelstore_attrib) );
+      copy_pixelstore(ctx, attr, &ctx->Unpack);
      newnode = new_attrib_node( GL_CLIENT_UNPACK_BIT );
      newnode->data = attr;
      newnode->next = head;
      head = newnode;
   }
   if (mask & GL_CLIENT_VERTEX_ARRAY_BIT) {
      struct gl_array_attrib *attr;
      struct gl_array_object *obj;
@@ -1353,7 +1374,7 @@ _mesa_PushClientAttrib(GLbitfield mask)
 void GLAPIENTRY
 _mesa_PopClientAttrib(void)
 {
-   struct gl_attrib_node *attr, *next;
+   struct gl_attrib_node *node, *next;
   GET_CURRENT_CONTEXT(ctx);
   ASSERT_OUTSIDE_BEGIN_END_AND_FLUSH(ctx);
@@ -1364,37 +1385,31 @@ _mesa_PopClientAttrib(void)
   }
   ctx->ClientAttribStackDepth--;
-   attr = ctx->ClientAttribStack[ctx->ClientAttribStackDepth];
+   node = ctx->ClientAttribStack[ctx->ClientAttribStackDepth];
-   while (attr) {
+   while (node) {
-      switch (attr->kind) {
+      switch (node->kind) {
         case GL_CLIENT_PACK_BIT:
-#if FEATURE_EXT_pixel_buffer_object
+            {
-            ctx->Pack.BufferObj->RefCount--;
+               struct gl_pixelstore_attrib *store =
-            if (ctx->Pack.BufferObj->RefCount <= 0) {
+                  (struct gl_pixelstore_attrib *) node->data;
-               _mesa_remove_buffer_object( ctx, ctx->Pack.BufferObj );
+               copy_pixelstore(ctx, &ctx->Pack, store);
-               (*ctx->Driver.DeleteBuffer)( ctx, ctx->Pack.BufferObj );
+               _mesa_reference_buffer_object(ctx, &store->BufferObj, NULL);
            }
 #endif
            MEMCPY( &ctx->Pack, attr->data,
                    sizeof(struct gl_pixelstore_attrib) );
 	    ctx->NewState |= _NEW_PACKUNPACK;
            break;
         case GL_CLIENT_UNPACK_BIT:
-#if FEATURE_EXT_pixel_buffer_object
+            {
-            ctx->Unpack.BufferObj->RefCount--;
+               struct gl_pixelstore_attrib *store =
-            if (ctx->Unpack.BufferObj->RefCount <= 0) {
+                  (struct gl_pixelstore_attrib *) node->data;
-               _mesa_remove_buffer_object( ctx, ctx->Unpack.BufferObj );
+               copy_pixelstore(ctx, &ctx->Unpack, store);
-               (*ctx->Driver.DeleteBuffer)( ctx, ctx->Unpack.BufferObj );
+               _mesa_reference_buffer_object(ctx, &store->BufferObj, NULL);
            }
 #endif
            MEMCPY( &ctx->Unpack, attr->data,
                    sizeof(struct gl_pixelstore_attrib) );
 	    ctx->NewState |= _NEW_PACKUNPACK;
            break;
         case GL_CLIENT_VERTEX_ARRAY_BIT: {
 	    struct gl_array_attrib * data =
-	      (struct gl_array_attrib *) attr->data;
+	      (struct gl_array_attrib *) node->data;
            adjust_buffer_object_ref_counts(&ctx->Array, -1);
@@ -1431,10 +1446,10 @@ _mesa_PopClientAttrib(void)
            break;
      }
-      next = attr->next;
+      next = node->next;
-      FREE( attr->data );
+      FREE( node->data );
-      FREE( attr );
+      FREE( node );
-      attr = next;
+      node = next;
   }
 }
--- a/src/mesa/main/bufferobj.c
+++ b/src/mesa/main/bufferobj.c
@@ -1,6 +1,6 @@
 /*
 * Mesa 3-D graphics library
- * Version:  7.1
+ * Version:  7.2
 *
 * Copyright (C) 1999-2008  Brian Paul   All Rights Reserved.
 *
@@ -166,22 +166,75 @@ _mesa_delete_buffer_object( GLcontext *ctx, struct gl_buffer_object *bufObj )
   if (bufObj->Data)
      _mesa_free(bufObj->Data);
   /* assign strange values here to help w/ debugging */
   bufObj->RefCount = -1000;
   bufObj->Name = ~0;
   _mesa_free(bufObj);
 }
 /**
 * Set ptr to bufObj w/ reference counting.
 */
 void
-_mesa_unbind_buffer_object( GLcontext *ctx, struct gl_buffer_object *bufObj )
+_mesa_reference_buffer_object(GLcontext *ctx,
                              struct gl_buffer_object **ptr,
                              struct gl_buffer_object *bufObj)
 {
-   if (bufObj != ctx->Array.NullBufferObj) {
+   if (*ptr == bufObj)
-      bufObj->RefCount--;
+      return;
-      if (bufObj->RefCount <= 0) {
+
   if (*ptr) {
      /* Unreference the old texture */
      GLboolean deleteFlag = GL_FALSE;
      struct gl_buffer_object *oldObj = *ptr;
      /*_glthread_LOCK_MUTEX(oldObj->Mutex);*/
      ASSERT(oldObj->RefCount > 0);
      oldObj->RefCount--;
 #if 0
      printf("BufferObj %p %d DECR to %d\n",
             (void *) oldObj, oldObj->Name, oldObj->RefCount);
 #endif
      deleteFlag = (oldObj->RefCount == 0);
      /*_glthread_UNLOCK_MUTEX(oldObj->Mutex);*/
      if (deleteFlag) {
         /* some sanity checking: don't delete a buffer still in use */
 	 ASSERT(ctx->Array.ArrayBufferObj != bufObj);
 	 ASSERT(ctx->Array.ElementArrayBufferObj != bufObj);
 	 ASSERT(ctx->Array.ArrayObj->Vertex.BufferObj != bufObj);
 	 ASSERT(ctx->Driver.DeleteBuffer);
-	 ctx->Driver.DeleteBuffer(ctx, bufObj);
+
         ctx->Driver.DeleteBuffer(ctx, oldObj);
      }
      *ptr = NULL;
   }
   ASSERT(!*ptr);
   if (bufObj) {
      /* reference new texture */
      /*_glthread_LOCK_MUTEX(tex->Mutex);*/
      if (bufObj->RefCount == 0) {
         /* this buffer's being deleted (look just above) */
         /* Not sure this can every really happen.  Warn if it does. */
         _mesa_problem(NULL, "referencing deleted buffer object");
         *ptr = NULL;
      }
      else {
         bufObj->RefCount++;
 #if 0
         printf("BufferObj %p %d INCR to %d\n",
                (void *) bufObj, bufObj->Name, bufObj->RefCount);
 #endif
         *ptr = bufObj;
      }
      /*_glthread_UNLOCK_MUTEX(tex->Mutex);*/
   }
 }
@@ -203,33 +256,6 @@ _mesa_initialize_buffer_object( struct gl_buffer_object *obj,
 }
 /**
 * Add the given buffer object to the buffer object pool.
 */
 void
 _mesa_save_buffer_object( GLcontext *ctx, struct gl_buffer_object *obj )
 {
   if (obj->Name > 0) {
      /* insert into hash table */
      _mesa_HashInsert(ctx->Shared->BufferObjects, obj->Name, obj);
   }
 }
 /**
 * Remove the given buffer object from the buffer object pool.
 * Do not deallocate the buffer object though.
 */
 void
 _mesa_remove_buffer_object( GLcontext *ctx, struct gl_buffer_object *bufObj )
 {
   if (bufObj->Name > 0) {
      /* remove from hash table */
      _mesa_HashRemove(ctx->Shared->BufferObjects, bufObj->Name);
   }
 }
 /**
 * Allocate space for and store data in a buffer object.  Any data that was
 * previously stored in the buffer object is lost.  If \c data is \c NULL,
@@ -400,6 +426,7 @@ _mesa_init_buffer_objects( GLcontext *ctx )
 {
   /* Allocate the default buffer object and set refcount so high that
    * it never gets deleted.
    * XXX with recent/improved refcounting this may not longer be needed.
    */
   ctx->Array.NullBufferObj = _mesa_new_buffer_object(ctx, 0, 0);
   if (ctx->Array.NullBufferObj)
@@ -462,28 +489,16 @@ bind_buffer_object(GLcontext *ctx, GLenum target, GLuint buffer)
            _mesa_error(ctx, GL_OUT_OF_MEMORY, "glBindBufferARB");
            return;
         }
-         _mesa_save_buffer_object(ctx, newBufObj);
+         _mesa_HashInsert(ctx->Shared->BufferObjects, buffer, newBufObj);
      }
   }
-   /* Make new binding */
+   /* bind new buffer */
-   *bindTarget = newBufObj;
+   _mesa_reference_buffer_object(ctx, bindTarget, newBufObj);
   newBufObj->RefCount++;
   /* Pass BindBuffer call to device driver */
   if (ctx->Driver.BindBuffer && newBufObj)
      ctx->Driver.BindBuffer( ctx, target, newBufObj );
   /* decr ref count on old buffer obj, delete if needed */
   if (oldBufObj) {
      oldBufObj->RefCount--;
      assert(oldBufObj->RefCount >= 0);
      if (oldBufObj->RefCount == 0) {
         assert(oldBufObj->Name != 0);
         ASSERT(ctx->Driver.DeleteBuffer);
         ctx->Driver.DeleteBuffer( ctx, oldBufObj );
      }
   }
 }
@@ -716,6 +731,23 @@ _mesa_lookup_bufferobj(GLcontext *ctx, GLuint buffer)
 }
 /**
 * If *ptr points to obj, set ptr = the Null/default buffer object.
 * This is a helper for buffer object deletion.
 * The GL spec says that deleting a buffer object causes it to get
 * unbound from all arrays in the current context.
 */
 static void
 unbind(GLcontext *ctx,
       struct gl_buffer_object **ptr,
       struct gl_buffer_object *obj)
 {
   if (*ptr == obj) {
      _mesa_reference_buffer_object(ctx, ptr, ctx->Array.NullBufferObj);
   }
 }
 /**********************************************************************/
 /* API Functions                                                      */
@@ -759,54 +791,18 @@ _mesa_DeleteBuffersARB(GLsizei n, const GLuint *ids)
         ASSERT(bufObj->Name == ids[i]);
-         if (ctx->Array.ArrayObj->Vertex.BufferObj == bufObj) {
+         unbind(ctx, &ctx->Array.ArrayObj->Vertex.BufferObj, bufObj);
-            bufObj->RefCount--;
+         unbind(ctx, &ctx->Array.ArrayObj->Normal.BufferObj, bufObj);
-            ctx->Array.ArrayObj->Vertex.BufferObj = ctx->Array.NullBufferObj;
+         unbind(ctx, &ctx->Array.ArrayObj->Color.BufferObj, bufObj);
-            ctx->Array.NullBufferObj->RefCount++;
+         unbind(ctx, &ctx->Array.ArrayObj->SecondaryColor.BufferObj, bufObj);
-         }
+         unbind(ctx, &ctx->Array.ArrayObj->FogCoord.BufferObj, bufObj);
-         if (ctx->Array.ArrayObj->Normal.BufferObj == bufObj) {
+         unbind(ctx, &ctx->Array.ArrayObj->Index.BufferObj, bufObj);
-            bufObj->RefCount--;
+         unbind(ctx, &ctx->Array.ArrayObj->EdgeFlag.BufferObj, bufObj);
            ctx->Array.ArrayObj->Normal.BufferObj = ctx->Array.NullBufferObj;
            ctx->Array.NullBufferObj->RefCount++;
         }
         if (ctx->Array.ArrayObj->Color.BufferObj == bufObj) {
            bufObj->RefCount--;
            ctx->Array.ArrayObj->Color.BufferObj = ctx->Array.NullBufferObj;
            ctx->Array.NullBufferObj->RefCount++;
         }
         if (ctx->Array.ArrayObj->SecondaryColor.BufferObj == bufObj) {
            bufObj->RefCount--;
            ctx->Array.ArrayObj->SecondaryColor.BufferObj = ctx->Array.NullBufferObj;
            ctx->Array.NullBufferObj->RefCount++;
         }
         if (ctx->Array.ArrayObj->FogCoord.BufferObj == bufObj) {
            bufObj->RefCount--;
            ctx->Array.ArrayObj->FogCoord.BufferObj = ctx->Array.NullBufferObj;
            ctx->Array.NullBufferObj->RefCount++;
         }
         if (ctx->Array.ArrayObj->Index.BufferObj == bufObj) {
            bufObj->RefCount--;
            ctx->Array.ArrayObj->Index.BufferObj = ctx->Array.NullBufferObj;
            ctx->Array.NullBufferObj->RefCount++;
         }
         if (ctx->Array.ArrayObj->EdgeFlag.BufferObj == bufObj) {
            bufObj->RefCount--;
            ctx->Array.ArrayObj->EdgeFlag.BufferObj = ctx->Array.NullBufferObj;
            ctx->Array.NullBufferObj->RefCount++;
         }
         for (j = 0; j < MAX_TEXTURE_UNITS; j++) {
-            if (ctx->Array.ArrayObj->TexCoord[j].BufferObj == bufObj) {
+            unbind(ctx, &ctx->Array.ArrayObj->TexCoord[j].BufferObj, bufObj);
               bufObj->RefCount--;
               ctx->Array.ArrayObj->TexCoord[j].BufferObj = ctx->Array.NullBufferObj;
               ctx->Array.NullBufferObj->RefCount++;
            }
         }
         for (j = 0; j < VERT_ATTRIB_MAX; j++) {
-            if (ctx->Array.ArrayObj->VertexAttrib[j].BufferObj == bufObj) {
+            unbind(ctx, &ctx->Array.ArrayObj->VertexAttrib[j].BufferObj, bufObj);
               bufObj->RefCount--;
               ctx->Array.ArrayObj->VertexAttrib[j].BufferObj = ctx->Array.NullBufferObj;
               ctx->Array.NullBufferObj->RefCount++;
            }
         }
         if (ctx->Array.ArrayBufferObj == bufObj) {
@@ -824,8 +820,8 @@ _mesa_DeleteBuffersARB(GLsizei n, const GLuint *ids)
         }
         /* The ID is immediately freed for re-use */
-         _mesa_remove_buffer_object(ctx, bufObj);
+         _mesa_HashRemove(ctx->Shared->BufferObjects, bufObj->Name);
-         _mesa_unbind_buffer_object(ctx, bufObj);
+         _mesa_reference_buffer_object(ctx, &bufObj, NULL);
      }
   }
@@ -874,7 +870,7 @@ _mesa_GenBuffersARB(GLsizei n, GLuint *buffer)
         _mesa_error(ctx, GL_OUT_OF_MEMORY, "glGenBuffersARB");
         return;
      }
-      _mesa_save_buffer_object(ctx, bufObj);
+      _mesa_HashInsert(ctx->Shared->BufferObjects, first + i, bufObj);
      buffer[i] = first + i;
   }
--- a/src/mesa/main/bufferobj.h
+++ b/src/mesa/main/bufferobj.h
@@ -1,6 +1,6 @@
 /*
 * Mesa 3-D graphics library
- * Version:  7.1
+ * Version:  7.2
 *
 * Copyright (C) 1999-2008  Brian Paul   All Rights Reserved.
 *
@@ -55,10 +55,9 @@ _mesa_initialize_buffer_object( struct gl_buffer_object *obj,
 				GLuint name, GLenum target );
 extern void
-_mesa_save_buffer_object( GLcontext *ctx, struct gl_buffer_object *obj );
+_mesa_reference_buffer_object(GLcontext *ctx,
-
+                              struct gl_buffer_object **ptr,
-extern void
+                              struct gl_buffer_object *bufObj);
 _mesa_remove_buffer_object( GLcontext *ctx, struct gl_buffer_object *bufObj );
 extern void
 _mesa_buffer_data( GLcontext *ctx, GLenum target, GLsizeiptrARB size,
@@ -118,8 +117,6 @@ _mesa_unmap_readpix_pbo(GLcontext *ctx,
                        const struct gl_pixelstore_attrib *pack);
 extern void
 _mesa_unbind_buffer_object( GLcontext *ctx, struct gl_buffer_object *bufObj );
 /*
 * API functions
--- a/src/mesa/main/extensions.c
+++ b/src/mesa/main/extensions.c
@@ -1,8 +1,8 @@
 /*
 * Mesa 3-D graphics library
- * Version:  6.5.1
+ * Version:  7.1
 *
- * Copyright (C) 1999-2006  Brian Paul   All Rights Reserved.
+ * Copyright (C) 1999-2008  Brian Paul   All Rights Reserved.
 *
 * Permission is hereby granted, free of charge, to any person obtaining a
 * copy of this software and associated documentation files (the "Software"),
--- a/src/mesa/main/framebuffer.c
+++ b/src/mesa/main/framebuffer.c
@@ -1,8 +1,8 @@
 /*
 * Mesa 3-D graphics library
- * Version:  6.5.1
+ * Version:  7.2
 *
- * Copyright (C) 1999-2006  Brian Paul   All Rights Reserved.
+ * Copyright (C) 1999-2008  Brian Paul   All Rights Reserved.
 *
 * Permission is hereby granted, free of charge, to any person obtaining a
 * copy of this software and associated documentation files (the "Software"),
@@ -491,13 +491,13 @@ _mesa_update_framebuffer_visual(struct gl_framebuffer *fb)
   if (fb->Attachment[BUFFER_ACCUM].Renderbuffer) {
      fb->Visual.haveAccumBuffer = GL_TRUE;
      fb->Visual.accumRedBits
-         = fb->Attachment[BUFFER_DEPTH].Renderbuffer->RedBits;
+         = fb->Attachment[BUFFER_ACCUM].Renderbuffer->RedBits;
      fb->Visual.accumGreenBits
-         = fb->Attachment[BUFFER_DEPTH].Renderbuffer->GreenBits;
+         = fb->Attachment[BUFFER_ACCUM].Renderbuffer->GreenBits;
      fb->Visual.accumBlueBits
-         = fb->Attachment[BUFFER_DEPTH].Renderbuffer->BlueBits;
+         = fb->Attachment[BUFFER_ACCUM].Renderbuffer->BlueBits;
      fb->Visual.accumAlphaBits
-         = fb->Attachment[BUFFER_DEPTH].Renderbuffer->AlphaBits;
+         = fb->Attachment[BUFFER_ACCUM].Renderbuffer->AlphaBits;
   }
   compute_depth_max(fb);
--- a/src/mesa/main/getstring.c
+++ b/src/mesa/main/getstring.c
@@ -85,7 +85,7 @@ compute_version(const GLcontext *ctx)
                              ctx->Extensions.ARB_texture_non_power_of_two &&
                              ctx->Extensions.EXT_blend_equation_separate);
   const GLboolean ver_2_1 = (ver_2_0 &&
-                              ctx->Extensions.ARB_shading_language_120 &&
+                              /*ctx->Extensions.ARB_shading_language_120 &&*/
                              ctx->Extensions.EXT_pixel_buffer_object &&
                              ctx->Extensions.EXT_texture_sRGB);
   if (ver_2_1)
@@ -121,12 +121,6 @@ _mesa_GetString( GLenum name )
   static const char *vendor = "Brian Paul";
   static const char *renderer = "Mesa";
 #if FEATURE_ARB_shading_language_120_foo /* support not complete! */
   static const char *sl_version = "1.20";
 #elif FEATURE_ARB_shading_language_100
   static const char *sl_version = "1.10";
 #endif
   if (!ctx)
      return NULL;
@@ -154,8 +148,10 @@ _mesa_GetString( GLenum name )
         return (const GLubyte *) ctx->Extensions.String;
 #if FEATURE_ARB_shading_language_100
      case GL_SHADING_LANGUAGE_VERSION_ARB:
-         if (ctx->Extensions.ARB_shading_language_100)
+         if (ctx->Extensions.ARB_shading_language_120)
-            return (const GLubyte *) sl_version;
+            return (const GLubyte *) "1.20";
         else if (ctx->Extensions.ARB_shading_language_100)
            return (const GLubyte *) "1.10";
         goto error;
 #endif
 #if FEATURE_NV_fragment_program || FEATURE_ARB_fragment_program || \
--- a/src/mesa/main/imports.c
+++ b/src/mesa/main/imports.c
@@ -559,7 +559,7 @@ _mesa_pow(double x, double y)
 int
 _mesa_ffs(int i)
 {
-#if (defined(_WIN32) && !defined(__MINGW32__) ) || defined(__IBMC__) || defined(__IBMCPP__)
+#if (defined(_WIN32) ) || defined(__IBMC__) || defined(__IBMCPP__)
   register int bit = 0;
   if (i != 0) {
      if ((i & 0xffff) == 0) {
--- a/src/mesa/main/light.c
+++ b/src/mesa/main/light.c
@@ -184,6 +184,7 @@ _mesa_Lightfv( GLenum light, GLenum pname, const GLfloat *params )
   GET_CURRENT_CONTEXT(ctx);
   GLint i = (GLint) (light - GL_LIGHT0);
   GLfloat temp[4];
   ASSERT_OUTSIDE_BEGIN_END(ctx);
   if (i < 0 || i >= (GLint) ctx->Const.MaxLights) {
      _mesa_error( ctx, GL_INVALID_ENUM, "glLight(light=0x%x)", light );
--- a/src/mesa/main/texformat.c
+++ b/src/mesa/main/texformat.c
@@ -1207,6 +1207,30 @@ const struct gl_texture_format _mesa_texformat_z24_s8 = {
   store_texel_z24_s8			/* StoreTexel */
 };
 const struct gl_texture_format _mesa_texformat_s8_z24 = {
   MESA_FORMAT_S8_Z24,			/* MesaFormat */
   GL_DEPTH_STENCIL_EXT,		/* BaseFormat */
   GL_UNSIGNED_NORMALIZED_ARB,		/* DataType */
   0,					/* RedBits */
   0,					/* GreenBits */
   0,					/* BlueBits */
   0,					/* AlphaBits */
   0,					/* LuminanceBits */
   0,					/* IntensityBits */
   0,					/* IndexBits */
   24,					/* DepthBits */
   8,					/* StencilBits */
   4,					/* TexelBytes */
   _mesa_texstore_s8_z24,		/* StoreTexImageFunc */
   NULL,				/* FetchTexel1D */
   NULL,				/* FetchTexel2D */
   NULL,				/* FetchTexel3D */
   fetch_texel_1d_f_s8_z24,		/* FetchTexel1Df */
   fetch_texel_2d_f_s8_z24,		/* FetchTexel2Df */
   fetch_texel_3d_f_s8_z24,		/* FetchTexel3Df */
   store_texel_s8_z24			/* StoreTexel */
 };
 const struct gl_texture_format _mesa_texformat_z16 = {
   MESA_FORMAT_Z16,			/* MesaFormat */
   GL_DEPTH_COMPONENT,			/* BaseFormat */
--- a/src/mesa/main/texformat.h
+++ b/src/mesa/main/texformat.h
@@ -84,6 +84,7 @@ enum _format {
   MESA_FORMAT_YCBCR,		/*                     YYYY YYYY UorV UorV */
   MESA_FORMAT_YCBCR_REV,	/*                     UorV UorV YYYY YYYY */
   MESA_FORMAT_Z24_S8,          /* ZZZZ ZZZZ ZZZZ ZZZZ ZZZZ ZZZZ SSSS SSSS */
   MESA_FORMAT_S8_Z24,          /* SSSS SSSS ZZZZ ZZZZ ZZZZ ZZZZ ZZZZ ZZZZ */
   MESA_FORMAT_Z16,             /*                     ZZZZ ZZZZ ZZZZ ZZZZ */
   MESA_FORMAT_Z32,             /* ZZZZ ZZZZ ZZZZ ZZZZ ZZZZ ZZZZ ZZZZ ZZZZ */
   /*@}*/
@@ -211,6 +212,7 @@ extern const struct gl_texture_format _mesa_texformat_l8;
 extern const struct gl_texture_format _mesa_texformat_i8;
 extern const struct gl_texture_format _mesa_texformat_ci8;
 extern const struct gl_texture_format _mesa_texformat_z24_s8;
 extern const struct gl_texture_format _mesa_texformat_s8_z24;
 extern const struct gl_texture_format _mesa_texformat_z16;
 extern const struct gl_texture_format _mesa_texformat_z32;
 /*@}*/
--- a/src/mesa/main/texformat_tmp.h
+++ b/src/mesa/main/texformat_tmp.h
@@ -1117,6 +1117,7 @@ static void FETCH(ci8)( const struct gl_texture_image *texImage,
         break;;
      default:
         _mesa_problem(ctx, "Bad palette format in fetch_texel_ci8");
         return;
      }
 #if CHAN_TYPE == GL_UNSIGNED_BYTE
      COPY_4UBV(texel, texelUB);
@@ -1363,6 +1364,32 @@ static void store_texel_z24_s8(struct gl_texture_image *texImage,
 #endif
 /* MESA_TEXFORMAT_S8_Z24 ***************************************************/
 static void FETCH(f_s8_z24)( const struct gl_texture_image *texImage,
                             GLint i, GLint j, GLint k, GLfloat *texel )
 {
   /* only return Z, not stencil data */
   const GLuint *src = TEXEL_ADDR(GLuint, texImage, i, j, k, 1);
   const GLfloat scale = 1.0F / (GLfloat) 0xffffff;
   texel[0] = ((*src) & 0x00ffffff) * scale;
   ASSERT(texImage->TexFormat->MesaFormat == MESA_FORMAT_S8_Z24);
   ASSERT(texel[0] >= 0.0F);
   ASSERT(texel[0] <= 1.0F);
 }
 #if DIM == 3
 static void store_texel_s8_z24(struct gl_texture_image *texImage,
                               GLint i, GLint j, GLint k, const void *texel)
 {
   /* only store Z, not stencil */
   GLuint *dst = TEXEL_ADDR(GLuint, texImage, i, j, k, 1);
   GLfloat depth = *((GLfloat *) texel);
   GLuint zi = (GLuint) (depth * 0xffffff);
   *dst = zi | (*dst & 0xff000000);
 }
 #endif
 #undef TEXEL_ADDR
 #undef DIM
--- a/src/mesa/main/texstore.c
+++ b/src/mesa/main/texstore.c
@@ -2445,6 +2445,87 @@ _mesa_texstore_z24_s8(TEXSTORE_PARAMS)
 }
 /**
 * Store a combined depth/stencil texture image.
 */
 GLboolean
 _mesa_texstore_s8_z24(TEXSTORE_PARAMS)
 {
   const GLuint depthScale = 0xffffff;
   const GLint srcRowStride
      = _mesa_image_row_stride(srcPacking, srcWidth, srcFormat, srcType)
      / sizeof(GLuint);
   GLint img, row;
   ASSERT(dstFormat == &_mesa_texformat_s8_z24);
   ASSERT(srcFormat == GL_DEPTH_STENCIL_EXT || srcFormat == GL_DEPTH_COMPONENT);
   ASSERT(srcFormat != GL_DEPTH_STENCIL_EXT || srcType == GL_UNSIGNED_INT_24_8_EXT);
   /* Incase we only upload depth we need to preserve the stencil */
   if (srcFormat == GL_DEPTH_COMPONENT) {
      for (img = 0; img < srcDepth; img++) {
         GLuint *dstRow = (GLuint *) dstAddr
            + dstImageOffsets[dstZoffset + img]
            + dstYoffset * dstRowStride / sizeof(GLuint)
            + dstXoffset;
         const GLuint *src
            = (const GLuint *) _mesa_image_address(dims, srcPacking, srcAddr,
                  srcWidth, srcHeight,
                  srcFormat, srcType,
                  img, 0, 0);
         for (row = 0; row < srcHeight; row++) {
            GLuint depth[MAX_WIDTH];
            GLint i;
            _mesa_unpack_depth_span(ctx, srcWidth,
                                    GL_UNSIGNED_INT, /* dst type */
                                    depth, /* dst addr */
                                    depthScale,
                                    srcType, src, srcPacking);
            for (i = 0; i < srcWidth; i++)
               dstRow[i] = depth[i] | (dstRow[i] & 0xFF000000);
            src += srcRowStride;
            dstRow += dstRowStride / sizeof(GLuint);
         }
      }
   } else {
      for (img = 0; img < srcDepth; img++) {
         GLuint *dstRow = (GLuint *) dstAddr
            + dstImageOffsets[dstZoffset + img]
            + dstYoffset * dstRowStride / sizeof(GLuint)
            + dstXoffset;
         const GLuint *src
            = (const GLuint *) _mesa_image_address(dims, srcPacking, srcAddr,
                  srcWidth, srcHeight,
                  srcFormat, srcType,
                  img, 0, 0);
         for (row = 0; row < srcHeight; row++) {
            GLubyte stencil[MAX_WIDTH];
            GLint i;
            /* the 24 depth bits will be in the low position: */
            _mesa_unpack_depth_span(ctx, srcWidth,
                                    GL_UNSIGNED_INT, /* dst type */
                                    dstRow, /* dst addr */
                                    depthScale,
                                    srcType, src, srcPacking);
            /* get the 8-bit stencil values */
            _mesa_unpack_stencil_span(ctx, srcWidth,
                                      GL_UNSIGNED_BYTE, /* dst type */
                                      stencil, /* dst addr */
                                      srcType, src, srcPacking,
                                      ctx->_ImageTransferState);
            /* merge stencil values into depth values */
            for (i = 0; i < srcWidth; i++)
               dstRow[i] |= stencil[i] << 24;
            src += srcRowStride;
            dstRow += dstRowStride / sizeof(GLuint);
         }
      }
   }
   return GL_TRUE;
 }
 /**
 * Store an image in any of the formats:
--- a/src/mesa/main/texstore.h
+++ b/src/mesa/main/texstore.h
@@ -57,6 +57,7 @@ extern GLboolean _mesa_texstore_a8(TEXSTORE_PARAMS);
 extern GLboolean _mesa_texstore_ci8(TEXSTORE_PARAMS);
 extern GLboolean _mesa_texstore_ycbcr(TEXSTORE_PARAMS);
 extern GLboolean _mesa_texstore_z24_s8(TEXSTORE_PARAMS);
 extern GLboolean _mesa_texstore_s8_z24(TEXSTORE_PARAMS);
 extern GLboolean _mesa_texstore_z16(TEXSTORE_PARAMS);
 extern GLboolean _mesa_texstore_z32(TEXSTORE_PARAMS);
 extern GLboolean _mesa_texstore_rgba_float32(TEXSTORE_PARAMS);
--- a/src/mesa/main/varray.c
+++ b/src/mesa/main/varray.c
@@ -1,8 +1,8 @@
 /*
 * Mesa 3-D graphics library
- * Version:  6.5.1
+ * Version:  7.2
 *
- * Copyright (C) 1999-2006  Brian Paul   All Rights Reserved.
+ * Copyright (C) 1999-2008  Brian Paul   All Rights Reserved.
 *
 * Permission is hereby granted, free of charge, to any person obtaining a
 * copy of this software and associated documentation files (the "Software"),
@@ -62,14 +62,9 @@ update_array(GLcontext *ctx, struct gl_client_array *array,
   array->Normalized = normalized;
   array->Ptr = (const GLubyte *) ptr;
 #if FEATURE_ARB_vertex_buffer_object
-   array->BufferObj->RefCount--;
+   _mesa_reference_buffer_object(ctx, &array->BufferObj,
-   if (array->BufferObj->RefCount <= 0) {
+                                 ctx->Array.ArrayBufferObj);
-      ASSERT(array->BufferObj->Name);
+
      _mesa_remove_buffer_object( ctx, array->BufferObj );
      (*ctx->Driver.DeleteBuffer)( ctx, array->BufferObj );
   }
   array->BufferObj = ctx->Array.ArrayBufferObj;
   array->BufferObj->RefCount++;
   /* Compute the index of the last array element that's inside the buffer.
    * Later in glDrawArrays we'll check if start + count > _MaxElement to
    * be sure we won't go out of bounds.
--- a/src/mesa/main/version.h
+++ b/src/mesa/main/version.h
@@ -1,8 +1,8 @@
 /*
 * Mesa 3-D graphics library
- * Version:  7.1
+ * Version:  7.2
 *
- * Copyright (C) 1999-2007  Brian Paul   All Rights Reserved.
+ * Copyright (C) 1999-2008  Brian Paul   All Rights Reserved.
 *
 * Permission is hereby granted, free of charge, to any person obtaining a
 * copy of this software and associated documentation files (the "Software"),
@@ -29,9 +29,9 @@
 /* Mesa version */
 #define MESA_MAJOR 7
-#define MESA_MINOR 1
+#define MESA_MINOR 2
 #define MESA_PATCH 0
-#define MESA_VERSION_STRING "7.1 rc4"
+#define MESA_VERSION_STRING "7.2"
 /* To make version comparison easy */
 #define MESA_VERSION(a,b,c) (((a) << 16) + ((b) << 8) + (c))
--- a/src/mesa/shader/prog_parameter.c
+++ b/src/mesa/shader/prog_parameter.c
@@ -29,9 +29,9 @@
 */
-#include "glheader.h"
+#include "main/glheader.h"
-#include "imports.h"
+#include "main/imports.h"
-#include "macros.h"
+#include "main/macros.h"
 #include "prog_instruction.h"
 #include "prog_parameter.h"
 #include "prog_statevars.h"
@@ -277,6 +277,26 @@ _mesa_add_uniform(struct gl_program_parameter_list *paramList,
 }
 /**
 * Mark the named uniform as 'used'.
 */
 void
 _mesa_use_uniform(struct gl_program_parameter_list *paramList,
                  const char *name)
 {
   GLuint i;
   for (i = 0; i < paramList->NumParameters; i++) {
      struct gl_program_parameter *p = paramList->Parameters + i;
      if (p->Type == PROGRAM_UNIFORM && _mesa_strcmp(p->Name, name) == 0) {
         p->Used = GL_TRUE;
         /* Note that large uniforms may occupy several slots so we're
          * not done searching yet.
          */
      }
   }
 }
 /**
 * Add a sampler to the parameter list.
 * \param name  uniform's name
@@ -591,21 +611,24 @@ _mesa_clone_parameter_list(const struct gl_program_parameter_list *list)
   /** Not too efficient, but correct */
   for (i = 0; i < list->NumParameters; i++) {
      struct gl_program_parameter *p = list->Parameters + i;
      struct gl_program_parameter *pCopy;
      GLuint size = MIN2(p->Size, 4);
      GLint j = _mesa_add_parameter(clone, p->Type, p->Name, size, p->DataType,
                                    list->ParameterValues[i], NULL);
      ASSERT(j >= 0);
      pCopy = clone->Parameters + j;
      pCopy->Used = p->Used;
      /* copy state indexes */
      if (p->Type == PROGRAM_STATE_VAR) {
         GLint k;
         struct gl_program_parameter *q = clone->Parameters + j;
         for (k = 0; k < STATE_LENGTH; k++) {
-            q->StateIndexes[k] = p->StateIndexes[k];
+            pCopy->StateIndexes[k] = p->StateIndexes[k];
         }
      }
      else {
         clone->Parameters[j].Size = p->Size;
      }
   }
   clone->StateFlags = list->StateFlags;
--- a/src/mesa/shader/prog_parameter.h
+++ b/src/mesa/shader/prog_parameter.h
@@ -31,7 +31,7 @@
 #ifndef PROG_PARAMETER_H
 #define PROG_PARAMETER_H
-#include "mtypes.h"
+#include "main/mtypes.h"
 #include "prog_statevars.h"
@@ -48,6 +48,7 @@ struct gl_program_parameter
   enum register_file Type; /**< PROGRAM_NAMED_PARAM, CONSTANT or STATE_VAR */
   GLenum DataType;         /**< GL_FLOAT, GL_FLOAT_VEC2, etc */
   GLuint Size;             /**< Number of components (1..4) */
   GLboolean Used;          /**< Helper flag for GLSL uniform tracking */
   /**
    * A sequence of STATE_* tokens and integers to identify GL state.
    */
@@ -112,6 +113,10 @@ extern GLint
 _mesa_add_uniform(struct gl_program_parameter_list *paramList,
                  const char *name, GLuint size, GLenum datatype);
 extern void
 _mesa_use_uniform(struct gl_program_parameter_list *paramList,
                  const char *name);
 extern GLint
 _mesa_add_sampler(struct gl_program_parameter_list *paramList,
                  const char *name, GLenum datatype);
--- a/src/mesa/shader/slang/library/slang_common_builtin.gc
+++ b/src/mesa/shader/slang/library/slang_common_builtin.gc
@@ -473,32 +473,33 @@ vec4 pow(const vec4 a, const vec4 b)
 float exp(const float a)
 {
-   const float e = 2.71828;
+   // NOTE: log2(e) = 1.44269502
-   __asm float_power __retVal, e, a;
+   float t = a * 1.44269502;
   __asm float_exp2 __retVal, t;
 }
 vec2 exp(const vec2 a)
 {
-   const float e = 2.71828;
+   vec2 t = a * 1.44269502;
-   __asm float_power __retVal.x, e, a.x;
+   __asm float_exp2 __retVal.x, t.x;
-   __asm float_power __retVal.y, e, a.y;
+   __asm float_exp2 __retVal.y, t.y;
 }
 vec3 exp(const vec3 a)
 {
-   const float e = 2.71828;
+   vec3 t = a * 1.44269502;
-   __asm float_power __retVal.x, e, a.x;
+   __asm float_exp2 __retVal.x, t.x;
-   __asm float_power __retVal.y, e, a.y;
+   __asm float_exp2 __retVal.y, t.y;
-   __asm float_power __retVal.z, e, a.z;
+   __asm float_exp2 __retVal.z, t.z;
 }
 vec4 exp(const vec4 a)
 {
-   const float e = 2.71828;
+   vec4 t = a * 1.44269502;
-   __asm float_power __retVal.x, e, a.x;
+   __asm float_exp2 __retVal.x, t.x;
-   __asm float_power __retVal.y, e, a.y;
+   __asm float_exp2 __retVal.y, t.y;
-   __asm float_power __retVal.z, e, a.z;
+   __asm float_exp2 __retVal.z, t.z;
-   __asm float_power __retVal.w, e, a.w;
+   __asm float_exp2 __retVal.w, t.w;
 }
--- a/src/mesa/shader/slang/library/slang_common_builtin_gc.h
+++ b/src/mesa/shader/slang/library/slang_common_builtin_gc.h
--- a/src/mesa/shader/slang/slang_codegen.c
+++ b/src/mesa/shader/slang/slang_codegen.c
@@ -1481,7 +1481,7 @@ _slang_simple_writemask(GLuint writemask, GLuint swizzle)
 * \return GL_FALSE for simple writemasks, GL_TRUE for non-simple
 */
 static GLboolean
-swizzle_to_writemask(GLuint swizzle,
+swizzle_to_writemask(slang_assemble_ctx *A, GLuint swizzle,
                     GLuint *writemaskOut, GLuint *swizzleOut)
 {
   GLuint mask = 0x0, newSwizzle[4];
@@ -1495,6 +1495,18 @@ swizzle_to_writemask(GLuint swizzle,
         break;
      }
      assert(swz >= 0 && swz <= 3);
      if (swizzle != SWIZZLE_XXXX &&
          swizzle != SWIZZLE_YYYY &&
          swizzle != SWIZZLE_ZZZZ &&
          swizzle != SWIZZLE_WWWW &&
          (mask & (1 << swz))) {
         /* a channel can't be specified twice (ex: ".xyyz") */
         slang_info_log_error(A->log, "Invalid writemask '%s'",
                              _mesa_swizzle_string(swizzle, 0, 0));
         return GL_FALSE;
      }
      mask |= (1 << swz);
   }
   assert(mask <= 0xf);
@@ -1590,11 +1602,11 @@ resolve_swizzle(const slang_operation *oper)
 * As above, but produce a writemask.
 */
 static GLuint
-resolve_writemask(const slang_operation *oper)
+resolve_writemask(slang_assemble_ctx *A, const slang_operation *oper)
 {
   GLuint swizzle = resolve_swizzle(oper);
   GLuint writemask, swizzleOut;
-   swizzle_to_writemask(swizzle, &writemask, &swizzleOut);
+   swizzle_to_writemask(A, swizzle, &writemask, &swizzleOut);
   return writemask;
 }
@@ -1668,7 +1680,7 @@ _slang_gen_asm(slang_assemble_ctx *A, slang_operation *oper,
      dest_oper = &oper->children[0];
-      writemask = resolve_writemask(dest_oper);
+      writemask = resolve_writemask(A, dest_oper);
      n0 = _slang_gen_operation(A, dest_oper);
      if (!n0)
@@ -1980,6 +1992,12 @@ _slang_gen_function_call_name(slang_assemble_ctx *A, const char *name,
      fun = _slang_locate_struct_constructor(A, name);
   }
   /*
    * At this point, some heuristics are used to try to find a function
    * that matches the calling signature by means of casting or "unrolling"
    * of constructors.
    */
   if (!fun && _slang_is_vec_mat_type(name)) {
      /* Next, if this call looks like a vec() or mat() constructor call,
       * try "unwinding" the args to satisfy a constructor.
@@ -1995,7 +2013,7 @@ _slang_gen_function_call_name(slang_assemble_ctx *A, const char *name,
      }
   }
-   if (!fun) {
+   if (!fun && _slang_is_vec_mat_type(name)) {
      /* Next, try casting args to the types of the formal parameters */
      int numArgs = oper->num_children;
      fun = _slang_find_function_by_argc(A->space.funcs, name, numArgs);
@@ -2008,6 +2026,13 @@ _slang_gen_function_call_name(slang_assemble_ctx *A, const char *name,
      assert(fun);
   }
   if (!fun) {
      slang_info_log_error(A->log,
                           "Function '%s' not found (check argument types)",
                           name);
      return NULL;
   }
   n = _slang_gen_function_call(A, fun, oper, dest);
   if (n && !n->Store && !dest
@@ -2716,7 +2741,11 @@ _slang_assignment_compatible(slang_assemble_ctx *A,
   slang_typeinfo t0, t1;
   GLuint sz0, sz1;
-   
+   if (op0->type == SLANG_OPER_POSTINCREMENT ||
       op0->type == SLANG_OPER_POSTDECREMENT) {
      return GL_FALSE;
   }
   slang_typeinfo_construct(&t0);
   _slang_typeof_operation(A, op0, &t0);
@@ -3020,7 +3049,7 @@ _slang_gen_assignment(slang_assemble_ctx * A, slang_operation *oper)
      if (lhs && rhs) {
         /* convert lhs swizzle into writemask */
         GLuint writemask, newSwizzle;
-         if (!swizzle_to_writemask(lhs->Store->Swizzle,
+         if (!swizzle_to_writemask(A, lhs->Store->Swizzle,
                                   &writemask, &newSwizzle)) {
            /* Non-simple writemask, need to swizzle right hand side in
             * order to put components into the right place.
--- a/src/mesa/shader/slang/slang_emit.c
+++ b/src/mesa/shader/slang/slang_emit.c
@@ -108,7 +108,9 @@ writemask_to_swizzle(GLuint writemask)
 /**
- * Swizzle a swizzle.  That is, return swz2(swz1)
+ * Swizzle a swizzle (function composition).
 * That is, return swz2(swz1), or said another way: swz1.szw2
 * Example: swizzle_swizzle(".zwxx", ".xxyw") yields ".zzwx"
 */
 GLuint
 _slang_swizzle_swizzle(GLuint swz1, GLuint swz2)
@@ -279,7 +281,7 @@ storage_to_src_reg(struct prog_src_register *src, const slang_ir_storage *st)
   while (st->Parent) {
      st = st->Parent;
      index += st->Index;
-      swizzle = _slang_swizzle_swizzle(st->Swizzle, swizzle);
+      swizzle = _slang_swizzle_swizzle(fix_swizzle(st->Swizzle), swizzle);
   }
   assert(st->File >= 0);
@@ -1636,6 +1638,88 @@ emit_struct_field(slang_emit_info *emitInfo, slang_ir_node *n)
 }
 /**
 * Emit code for a variable declaration.
 * This usually doesn't result in any code generation, but just
 * memory allocation.
 */
 static struct prog_instruction *
 emit_var_decl(slang_emit_info *emitInfo, slang_ir_node *n)
 {
   struct prog_instruction *inst;
   assert(n->Store);
   assert(n->Store->File != PROGRAM_UNDEFINED);
   assert(n->Store->Size > 0);
   /*assert(n->Store->Index < 0);*/
   if (!n->Var || n->Var->isTemp) {
      /* a nameless/temporary variable, will be freed after first use */
      /*NEW*/
      if (n->Store->Index < 0 && !_slang_alloc_temp(emitInfo->vt, n->Store)) {
         slang_info_log_error(emitInfo->log,
                              "Ran out of registers, too many temporaries");
         return NULL;
      }
   }
   else {
      /* a regular variable */
      _slang_add_variable(emitInfo->vt, n->Var);
      if (!_slang_alloc_var(emitInfo->vt, n->Store)) {
         slang_info_log_error(emitInfo->log,
                              "Ran out of registers, too many variables");
         return NULL;
      }
      /*
        printf("IR_VAR_DECL %s %d store %p\n",
        (char*) n->Var->a_name, n->Store->Index, (void*) n->Store);
      */
      assert(n->Var->aux == n->Store);
   }
   if (emitInfo->EmitComments) {
      /* emit NOP with comment describing the variable's storage location */
      char s[1000];
      sprintf(s, "TEMP[%d]%s = variable %s (size %d)",
              n->Store->Index,
              _mesa_swizzle_string(n->Store->Swizzle, 0, GL_FALSE), 
              (n->Var ? (char *) n->Var->a_name : "anonymous"),
              n->Store->Size);
      inst = emit_comment(emitInfo, s);
      return inst;
   }
   return NULL;
 }
 /**
 * Emit code for a reference to a variable.
 * Actually, no code is generated but we may do some memory alloation.
 * In particular, state vars (uniforms) are allocated on an as-needed basis.
 */
 static struct prog_instruction *
 emit_var_ref(slang_emit_info *emitInfo, slang_ir_node *n)
 {
   assert(n->Store);
   assert(n->Store->File != PROGRAM_UNDEFINED);
   if (n->Store->File == PROGRAM_STATE_VAR && n->Store->Index < 0) {
      n->Store->Index = _slang_alloc_statevar(n, emitInfo->prog->Parameters);
   }
   else if (n->Store->File == PROGRAM_UNIFORM) {
      /* mark var as used */
      _mesa_use_uniform(emitInfo->prog->Parameters, (char *) n->Var->a_name);
   }
   if (n->Store->Index < 0) {
      /* probably ran out of registers */
      return NULL;
   }
   assert(n->Store->Size > 0);
   return NULL;
 }
 static struct prog_instruction *
 emit(slang_emit_info *emitInfo, slang_ir_node *n)
 {
@@ -1671,64 +1755,14 @@ emit(slang_emit_info *emitInfo, slang_ir_node *n)
   case IR_VAR_DECL:
      /* Variable declaration - allocate a register for it */
-      assert(n->Store);
+      inst = emit_var_decl(emitInfo, n);
-      assert(n->Store->File != PROGRAM_UNDEFINED);
+      return inst;
      assert(n->Store->Size > 0);
      /*assert(n->Store->Index < 0);*/
      if (!n->Var || n->Var->isTemp) {
         /* a nameless/temporary variable, will be freed after first use */
         /*NEW*/
         if (n->Store->Index < 0 && !_slang_alloc_temp(emitInfo->vt, n->Store)) {
            slang_info_log_error(emitInfo->log,
                                 "Ran out of registers, too many temporaries");
            return NULL;
         }
      }
      else {
         /* a regular variable */
         _slang_add_variable(emitInfo->vt, n->Var);
         if (!_slang_alloc_var(emitInfo->vt, n->Store)) {
            slang_info_log_error(emitInfo->log,
                                 "Ran out of registers, too many variables");
            return NULL;
         }
         /*
         printf("IR_VAR_DECL %s %d store %p\n",
                (char*) n->Var->a_name, n->Store->Index, (void*) n->Store);
         */
         assert(n->Var->aux == n->Store);
      }
      if (emitInfo->EmitComments) {
         /* emit NOP with comment describing the variable's storage location */
         char s[1000];
         sprintf(s, "TEMP[%d]%s = variable %s (size %d)",
                 n->Store->Index,
                 _mesa_swizzle_string(n->Store->Swizzle, 0, GL_FALSE), 
                 (n->Var ? (char *) n->Var->a_name : "anonymous"),
                 n->Store->Size);
         inst = emit_comment(emitInfo, s);
         return inst;
      }
      return NULL;
   case IR_VAR:
      /* Reference to a variable
       * Storage should have already been resolved/allocated.
       */
-      assert(n->Store);
+      return emit_var_ref(emitInfo, n);
      assert(n->Store->File != PROGRAM_UNDEFINED);
      if (n->Store->File == PROGRAM_STATE_VAR &&
          n->Store->Index < 0) {
         n->Store->Index = _slang_alloc_statevar(n, emitInfo->prog->Parameters);
      }
      if (n->Store->Index < 0) {
         /* probably ran out of registers */
         return NULL;
      }
      assert(n->Store->Size > 0);
      break;
   case IR_ELEMENT:
      return emit_array_element(emitInfo, n);
@@ -1939,6 +1973,7 @@ _slang_emit_code(slang_ir_node *n, slang_var_table *vt,
   GET_CURRENT_CONTEXT(ctx);
   GLboolean success;
   slang_emit_info emitInfo;
   GLuint maxUniforms;
   emitInfo.log = log;
   emitInfo.vt = vt;
@@ -1955,6 +1990,19 @@ _slang_emit_code(slang_ir_node *n, slang_var_table *vt,
      emitInfo.EmitHighLevelInstructions = GL_TRUE;
   }      
   /* Check uniform/constant limits */
   if (prog->Target == GL_FRAGMENT_PROGRAM_ARB) {
      maxUniforms = ctx->Const.FragmentProgram.MaxUniformComponents / 4;
   }
   else {
      assert(prog->Target == GL_VERTEX_PROGRAM_ARB);
      maxUniforms = ctx->Const.VertexProgram.MaxUniformComponents / 4;
   }
   if (prog->Parameters->NumParameters > maxUniforms) {
      slang_info_log_error(log, "Constant/uniform register limit exceeded");
      return GL_FALSE;
   }
   (void) emit(&emitInfo, n);
   /* finish up by adding the END opcode to program */
--- a/src/mesa/shader/slang/slang_link.c
+++ b/src/mesa/shader/slang/slang_link.c
@@ -173,7 +173,7 @@ link_uniform_vars(struct gl_shader_program *shProg,
       * Furthermore, we'll need to fix the state-var's size/datatype info.
       */
-      if (p->Type == PROGRAM_UNIFORM ||
+      if ((p->Type == PROGRAM_UNIFORM && p->Used) ||
          p->Type == PROGRAM_SAMPLER) {
         _mesa_append_uniform(shProg->Uniforms, p->Name, prog->Target, i);
      }
--- a/src/mesa/swrast/s_texfilter.c
+++ b/src/mesa/swrast/s_texfilter.c
@@ -342,6 +342,7 @@ lerp_rgba_3d(GLchan result[4], GLfloat a, GLfloat b, GLfloat c,
      break;								\
   default:								\
      _mesa_problem(ctx, "Bad wrap mode");				\
      return;								\
   }									\
 }
@@ -462,6 +463,7 @@ lerp_rgba_3d(GLchan result[4], GLfloat a, GLfloat b, GLfloat c,
      break;								\
   default:								\
      _mesa_problem(ctx, "Bad wrap mode");				\
      return;								\
   }									\
 }
--- a/src/mesa/tnl/t_vp_build.c
+++ b/src/mesa/tnl/t_vp_build.c
@@ -235,7 +235,7 @@ static struct state_key *make_state_key( GLcontext *ctx )
 */
 #define PREFER_DP4 0
-#define MAX_INSN 256
+#define MAX_INSN 350
 /* Use uregs to represent registers internally, translate to Mesa's
 * expected formats on emit.  
--- a/src/mesa/vbo/vbo_exec_api.c
+++ b/src/mesa/vbo/vbo_exec_api.c
@@ -1,6 +1,6 @@
 /**************************************************************************
-Copyright 2002 Tungsten Graphics Inc., Cedar Park, Texas.
+Copyright 2002-2008 Tungsten Graphics Inc., Cedar Park, Texas.
 All Rights Reserved.
@@ -31,6 +31,7 @@ USE OR OTHER DEALINGS IN THE SOFTWARE.
 */
 #include "main/glheader.h"
 #include "main/bufferobj.h"
 #include "main/context.h"
 #include "main/macros.h"
 #include "main/vtxfmt.h"
@@ -655,7 +656,10 @@ void vbo_exec_vtx_init( struct vbo_exec_context *exec )
   /* Allocate a buffer object.  Will just reuse this object
    * continuously.
    */
-   exec->vtx.bufferobj = ctx->Array.NullBufferObj;
+   _mesa_reference_buffer_object(ctx,
                                 &exec->vtx.bufferobj,
                                 ctx->Array.NullBufferObj);
   exec->vtx.buffer_map = ALIGN_MALLOC(VBO_VERT_BUFFER_SIZE * sizeof(GLfloat), 64);
   vbo_exec_vtxfmt_init( exec );
--- a/src/mesa/vbo/vbo_exec_draw.c
+++ b/src/mesa/vbo/vbo_exec_draw.c
@@ -1,8 +1,8 @@
 /*
 * Mesa 3-D graphics library
- * Version:  5.1
+ * Version:  7.2
 *
- * Copyright (C) 1999-2003  Brian Paul   All Rights Reserved.
+ * Copyright (C) 1999-2008  Brian Paul   All Rights Reserved.
 *
 * Permission is hereby granted, free of charge, to any person obtaining a
 * copy of this software and associated documentation files (the "Software"),
@@ -26,6 +26,7 @@
 */
 #include "main/glheader.h"
 #include "main/bufferobj.h"
 #include "main/context.h"
 #include "main/enums.h"
 #include "main/state.h"
@@ -155,8 +156,12 @@ static void vbo_exec_bind_arrays( GLcontext *ctx )
    */
   switch (get_program_mode(exec->ctx)) {
   case VP_NONE:
-      memcpy(arrays,      vbo->legacy_currval, 16 * sizeof(arrays[0]));
+      for (attr = 0; attr < 16; attr++) {
-      memcpy(arrays + 16, vbo->mat_currval,    MAT_ATTRIB_MAX * sizeof(arrays[0]));
+         exec->vtx.inputs[attr] = &vbo->legacy_currval[attr];
      }
      for (attr = 0; attr < MAT_ATTRIB_MAX; attr++) {
         exec->vtx.inputs[attr + 16] = &vbo->mat_currval[attr];
      }
      map = vbo->map_vp_none;
      break;
   case VP_NV:
@@ -165,8 +170,10 @@ static void vbo_exec_bind_arrays( GLcontext *ctx )
       * occurred.  NV vertex programs cannot access material values,
       * nor attributes greater than VERT_ATTRIB_TEX7.  
       */
-      memcpy(arrays,      vbo->legacy_currval,  16 * sizeof(arrays[0]));
+      for (attr = 0; attr < 16; attr++) {
-      memcpy(arrays + 16, vbo->generic_currval, 16 * sizeof(arrays[0]));
+         exec->vtx.inputs[attr] = &vbo->legacy_currval[attr];
         exec->vtx.inputs[attr + 16] = &vbo->generic_currval[attr];
      }
      map = vbo->map_vp_arb;
      break;
   }
@@ -178,13 +185,18 @@ static void vbo_exec_bind_arrays( GLcontext *ctx )
      GLuint src = map[attr];
      if (exec->vtx.attrsz[src]) {
         /* override the default array set above */
         exec->vtx.inputs[attr] = &arrays[attr];
 	 arrays[attr].Ptr = (void *)data;
 	 arrays[attr].Size = exec->vtx.attrsz[src];
 	 arrays[attr].StrideB = exec->vtx.vertex_size * sizeof(GLfloat);
 	 arrays[attr].Stride = exec->vtx.vertex_size * sizeof(GLfloat);
 	 arrays[attr].Type = GL_FLOAT;
 	 arrays[attr].Enabled = 1;
-	 arrays[attr].BufferObj = exec->vtx.bufferobj; /* NullBufferObj */
+         _mesa_reference_buffer_object(ctx,
                                       &arrays[attr].BufferObj,
                                       exec->vtx.bufferobj);
 	 arrays[attr]._MaxElement = count; /* ??? */
 	 data += exec->vtx.attrsz[src] * sizeof(GLfloat);
--- a/src/mesa/vbo/vbo_save.c
+++ b/src/mesa/vbo/vbo_save.c
@@ -1,8 +1,8 @@
 /*
 * Mesa 3-D graphics library
- * Version:  6.3
+ * Version:  7.2
 *
- * Copyright (C) 1999-2005  Brian Paul   All Rights Reserved.
+ * Copyright (C) 1999-2008  Brian Paul   All Rights Reserved.
 *
 * Permission is hereby granted, free of charge, to any person obtaining a
 * copy of this software and associated documentation files (the "Software"),
@@ -27,6 +27,7 @@
 #include "main/mtypes.h"
 #include "main/bufferobj.h"
 #include "main/dlist.h"
 #include "main/vtxfmt.h"
 #include "main/imports.h"
@@ -71,19 +72,24 @@ void vbo_save_destroy( GLcontext *ctx )
 {
   struct vbo_context *vbo = vbo_context(ctx);
   struct vbo_save_context *save = &vbo->save;
   GLuint i;
   if (save->prim_store) {
      if ( --save->prim_store->refcount == 0 ) {
         FREE( save->prim_store );
         save->prim_store = NULL;
      }
      if ( --save->vertex_store->refcount == 0 ) {
-         if (save->vertex_store->bufferobj)
+         _mesa_reference_buffer_object(ctx,
-            ctx->Driver.DeleteBuffer( ctx, save->vertex_store->bufferobj );
+                                       &save->vertex_store->bufferobj, NULL);
         FREE( save->vertex_store );
         save->vertex_store = NULL;
      }
   }
   for (i = 0; i < VBO_ATTRIB_MAX; i++) {
      _mesa_reference_buffer_object(ctx, &save->arrays[i].BufferObj, NULL);
   }
 }
--- a/src/mesa/vbo/vbo_save_api.c
+++ b/src/mesa/vbo/vbo_save_api.c
@@ -1,6 +1,6 @@
 /**************************************************************************
-Copyright 2002 Tungsten Graphics Inc., Cedar Park, Texas.
+Copyright 2002-2008 Tungsten Graphics Inc., Cedar Park, Texas.
 All Rights Reserved.
@@ -68,6 +68,7 @@ USE OR OTHER DEALINGS IN THE SOFTWARE.
 #include "main/glheader.h"
 #include "main/bufferobj.h"
 #include "main/context.h"
 #include "main/dlist.h"
 #include "main/enums.h"
@@ -85,6 +86,10 @@ USE OR OTHER DEALINGS IN THE SOFTWARE.
 #endif
 /* An interesting VBO number/name to help with debugging */
 #define VBO_BUF_ID  12345
 /*
 * NOTE: Old 'parity' issue is gone, but copying can still be
 * wrong-footed on replay.
@@ -170,7 +175,9 @@ static struct vbo_save_vertex_store *alloc_vertex_store( GLcontext *ctx )
    * user.  Perhaps there could be a special number for internal
    * buffers:
    */
-   vertex_store->bufferobj = ctx->Driver.NewBufferObject(ctx, 1, GL_ARRAY_BUFFER_ARB);
+   vertex_store->bufferobj = ctx->Driver.NewBufferObject(ctx,
                                                         VBO_BUF_ID,
                                                         GL_ARRAY_BUFFER_ARB);
   ctx->Driver.BufferData( ctx, 
 			   GL_ARRAY_BUFFER_ARB, 
@@ -190,8 +197,9 @@ static void free_vertex_store( GLcontext *ctx, struct vbo_save_vertex_store *ver
 {
   assert(!vertex_store->buffer);
-   if (vertex_store->bufferobj)
+   if (vertex_store->bufferobj) {
-      ctx->Driver.DeleteBuffer( ctx, vertex_store->bufferobj );
+      _mesa_reference_buffer_object(ctx, &vertex_store->bufferobj, NULL);
   }
   FREE( vertex_store );
 }
@@ -1139,6 +1147,7 @@ void vbo_save_api_init( struct vbo_save_context *save )
   _save_vtxfmt_init( ctx );
   _save_current_init( ctx );
   /* These will actually get set again when binding/drawing */
   for (i = 0; i < VBO_ATTRIB_MAX; i++)
      save->inputs[i] = &save->arrays[i];
--- a/src/mesa/vbo/vbo_save_draw.c
+++ b/src/mesa/vbo/vbo_save_draw.c
@@ -1,8 +1,8 @@
 /*
 * Mesa 3-D graphics library
- * Version:  6.1
+ * Version:  7.2
 *
- * Copyright (C) 1999-2004  Brian Paul   All Rights Reserved.
+ * Copyright (C) 1999-2008  Brian Paul   All Rights Reserved.
 *
 * Permission is hereby granted, free of charge, to any person obtaining a
 * copy of this software and associated documentation files (the "Software"),
@@ -27,6 +27,7 @@
 */
 #include "main/glheader.h"
 #include "main/bufferobj.h"
 #include "main/context.h"
 #include "main/imports.h"
 #include "main/mtypes.h"
@@ -115,8 +116,12 @@ static void vbo_bind_vertex_list( GLcontext *ctx,
    */
   switch (get_program_mode(ctx)) {
   case VP_NONE:
-      memcpy(arrays,      vbo->legacy_currval, 16 * sizeof(arrays[0]));
+      for (attr = 0; attr < 16; attr++) {
-      memcpy(arrays + 16, vbo->mat_currval,    MAT_ATTRIB_MAX * sizeof(arrays[0]));
+         save->inputs[attr] = &vbo->legacy_currval[attr];
      }
      for (attr = 0; attr < MAT_ATTRIB_MAX; attr++) {
         save->inputs[attr + 16] = &vbo->mat_currval[attr];
      }
      map = vbo->map_vp_none;
      break;
   case VP_NV:
@@ -125,8 +130,10 @@ static void vbo_bind_vertex_list( GLcontext *ctx,
       * occurred.  NV vertex programs cannot access material values,
       * nor attributes greater than VERT_ATTRIB_TEX7.  
       */
-      memcpy(arrays,      vbo->legacy_currval,  16 * sizeof(arrays[0]));
+      for (attr = 0; attr < 16; attr++) {
-      memcpy(arrays + 16, vbo->generic_currval, 16 * sizeof(arrays[0]));
+         save->inputs[attr] = &vbo->legacy_currval[attr];
         save->inputs[attr + 16] = &vbo->generic_currval[attr];
      }
      map = vbo->map_vp_arb;
      break;
   }
@@ -135,13 +142,18 @@ static void vbo_bind_vertex_list( GLcontext *ctx,
      GLuint src = map[attr];
      if (node->attrsz[src]) {
-	 arrays[attr].Ptr = (const GLubyte *)data;
+         /* override the default array set above */
         save->inputs[attr] = &arrays[attr];
 	 arrays[attr].Ptr = (const GLubyte *) data;
 	 arrays[attr].Size = node->attrsz[src];
 	 arrays[attr].StrideB = node->vertex_size * sizeof(GLfloat);
 	 arrays[attr].Stride = node->vertex_size * sizeof(GLfloat);
 	 arrays[attr].Type = GL_FLOAT;
 	 arrays[attr].Enabled = 1;
-	 arrays[attr].BufferObj = node->vertex_store->bufferobj;
+         _mesa_reference_buffer_object(ctx,
                                       &arrays[attr].BufferObj,
                                       node->vertex_store->bufferobj);
 	 arrays[attr]._MaxElement = node->count; /* ??? */
 	 assert(arrays[attr].BufferObj->Name);
Author	SHA1	Message	Date
Brian Paul	419353dc2f	mesa: Makefile: 7.2 release candidate 1	2008-09-13 16:06:12 -06:00
Brian Paul	38d3b2262b	mesa: documentation updates for 7.2 release	2008-09-13 14:43:54 -06:00
Brian Paul	4db2a7f36c	updated Cell driver info	2008-09-13 14:39:39 -06:00
Brian Paul	1145042fd2	Mesa: version bumps for 7.2 release	2008-09-13 14:36:35 -06:00
Brian Paul	772131280a	document 7.2 bug fixes, etc	2008-09-13 14:36:17 -06:00
Jakub Bogusz	ee913c0fcc	tdfx: fix crash and lack of visuals bug	2008-09-13 14:34:51 -06:00
Guillaume Melquiond	cd1804c43a	i915: fix himask constant init for 64-bit build	2008-09-13 14:26:35 -06:00
Guillaume Melquiond	253784d45e	glx: fix 64-bit datatype issue	2008-09-13 14:26:25 -06:00
Guillaume Melquiond	463d19d477	mesa: return after _mesa_problem() calls	2008-09-13 14:26:18 -06:00
Guillaume Melquiond	6e4fd90ca5	mesa: #include <stdio.h>	2008-09-13 14:26:13 -06:00
Guillaume Melquiond	06188c0c5c	remove invalid XDestroyWindow() call	2008-09-13 14:26:06 -06:00
Guillaume Melquiond	dec9097647	silence warning	2008-09-13 14:25:59 -06:00
Guillaume Melquiond	e234339dba	GLU: fix asst warnings	2008-09-13 14:25:53 -06:00
Ian Romanick	bb09d23e92	Remove DRI2 support Remove support for the original incarnation of DRI2. This version has since been abandoned and will never ship. This eliminates a build dependency on a component that only exists in Fedora 9 and in now-dead up-stream development trees. I also recommend that we spin a Mesa 7.1.1 release with this patch. Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>	2008-09-12 13:48:14 -07:00
Xiang, Haihao	1ffbcddb6f	i965: Add support for G41 chipset which is another 4 series. (cherry picked from commit `b2e083eba2`)	2008-09-12 09:29:00 +08:00
Shane Blackett	33cf8f5c06	added 24bpp support	2008-09-11 08:35:37 -06:00
Shane Blackett	7b8d878bed	Fixes for Mingw	2008-09-10 08:23:19 -06:00
Brian Paul	86ae34fbb4	mesa: replace MALLOC w/ CALLOC to fix memory error in glPushClientAttrib()	2008-09-05 08:07:07 -06:00
Brian Paul	16f3461fd6	mesa: improved gl_buffer_object reference counting Use new _mesa_reference_buffer_object() function wherever possible. Fixes buffer object/display list crash reported in ParaView.	2008-09-04 15:26:07 -06:00
Xiang, Haihao	0ac8886b51	intel: Fix depth_stencil texture. (cherry picked from commit `0adfd10210`)	2008-09-04 11:54:31 +08:00
Xiang, Haihao	1729d37223	mesa: merge stencil values into depth values for MESA_FORMAT_S8_Z24 (cherry picked from commit `218df7f9c5`)	2008-09-04 11:54:21 +08:00
Jakob Bornecrantz	81edef6f9f	mesa: Support for MESA_FORMAT_S8_Z24 texture cherry-picked from gallium-0.1 (cherry picked from commit `dc44bb8e92`)	2008-09-04 11:54:08 +08:00
Brian Paul	e040e18c7b	fix BUFFER_DEPTH/BUFFER_ACCUM mix-up	2008-09-02 18:13:53 -06:00
Roland Scheidegger	14d186644b	fix no error generated when calling glLight{if}[v] inside begin/end (bug 17408)	2008-09-02 15:49:35 +02:00
Xiang, Haihao	e4535e4d5e	i965: force thread switch after IF/ELSE/ENDIF. partial fix for #16882 . A thread switch is implicitly invoked after the issuance of an IF/ELSE/ENDIF instruction if necessary. Unfortunately it seems sometimes a forced thread switch is needed.	2008-08-29 10:30:12 +08:00
Xiang, Haihao	a008813890	i965: mask control for BREAK/CONT/DO/WHILE. partial fix fox #16882	2008-08-29 10:30:03 +08:00
Xiang, Haihao	da50dc7bb3	i965: Push/pop instruction state. partial fix for #16882	2008-08-29 10:29:57 +08:00
Brian Paul	7f628d9cbc	mesa: bump MAX_INSN to 350	2008-08-28 15:16:33 -06:00
Brian Paul	b2ffd0be03	placeholder for upcoming 7.2 release notes	2008-08-26 16:24:37 -06:00
Brian Paul	912b16c689	added 7.1 MD5 sums	2008-08-26 16:11:56 -06:00
Brian Paul	6d9b25e522	mesa: don't check for GLSL 1.2 to advertise GL 2.1 The GLSL 1.2 features are minor...	2008-08-26 15:50:11 -06:00
Corbin Simpson	31bad23f0f	note ATI R500 series support	2008-08-26 12:39:27 -06:00
Timo Jyrinki	29b9b25e72	asst. html doc updates	2008-08-26 12:36:23 -06:00
Brian Paul	e794f496c6	mesa: assorted updates/prep for 7.1 release	2008-08-25 14:09:14 -06:00
Brian Paul	68067c66d3	mesa: glsl: regenerated file	2008-08-25 09:16:21 -06:00
Brian Paul	ad5474e995	mesa: glsl: grab latest fixes from gallium-0.1 branch Includes: 1. Fixes failed asserting about bad swizzles in src reg emit. 2. Tracks uniform var usage. 3. Emit exp() in terms of EXP2 instruction.	2008-08-25 09:15:16 -06:00