initial rev

9 files changed, 1393 insertions, 0 deletions

diff --git a/docs/README.AMIWIN b/docs/README.AMIWIN
new file mode 100644
index 00000000000..47cf696cc1e
--- /dev/null
+++ b/docs/README.AMIWIN
@@ -0,0 +1,181 @@
+AMIGA AMIWIN PORT of MESA: THE OPENGL SOFTWARE EMULATION
+========================================================
+Port by Victor Ng-Thow-Hing ([email protected]) 
+Original Author (Brian Paul ([email protected])
+
+Dec.1 , 1995: Port of release Mesa 1.2.5
+ - Modifications made to minimize changes to Mesa distribution.
+
+Nov.25, 1995: Port of release Mesa 1.2.4
+
+
+HISTORY
+=======
+As a 3D graphics progammer, I was increasingly frustrated to see OpenGL 
+appearing on so many platforms EXCEPT the Amiga. Up to now, the task
+of porting OpenGL directly from native Amiga drawing routines seemed like
+a daunting task. However, two important events made this port possible.
+
+First of all, Brian Paul wrote Mesa, the OpenGL software emulator that 
+can be found on many platforms - except the Amiga and Atari (who cares 
+about the latter!). This was pretty ironic considering that Mesa was 
+originally prototyped on an Amiga! The second great event was when 
+Holger Kruse developed AmiWin, the X11R6 server for the Amiga (definitely 
+register for this great piece of software) and released a development kit
+so one could compile X programs with SAS/C.
+
+Since Mesa had X routines as its primitive drawing operations, this made
+a marriage of Mesa and Amiwin feasible. I copied over the sources from
+an ftp site, played with the code, wrote some Smakefiles, and voila, 
+I had OpenGL programs displaying on my Amiga.
+
+Although the speed is nothing to be impressed about, this port can be
+potentially useful to those who want to quickly test their code in
+wireframe or perhaps learn more about programming with the OpenGL API.
+
+I hope Amiga developers will continue to write excellent software for
+their machine, especially more X clients for Amiwin. If you have any 
+solutions so some of my problems in the porting notes, please send me
+some email!
+
+See you around,
+Vic.
+
+HOW TO CREATE THE LIBRARIES AND SAMPLE CODE
+===========================================
+
+Just run the shell script mklib.amiwin in the mesa directory. This will
+make all the libraries and copy them into the mesa/lib directory. If you
+don't want to compile everything, just go to the desired directory and
+type smake in that directory.
+
+Change any of the variables in the smakefiles as necessary. You will REQUIRE
+the Amiwin development kit to compile these libraries since you need X11.LIB
+and the shareable X libraries. Some examples require the AmiTCP4.0
+net.lib static link library and related header files for unix related
+header files and functions like sleep().
+
+HOW TO USE THE MESA LIBRARIES
+=============================
+
+Study the Smakefiles in the demos, samples and book directories for the
+proper SAS/C options and linkable libraries to use. Basically aux calls
+require Mesaaux.LIB, gl calls require MesaGL.LIB, glu calls MesaGLU.LIB,
+tk calls Mesatk.LIB. There is a preliminary port of MesaGLUT.LIB toolkit
+available in the lib directory with the other Mesa libraries. However, 
+it seems to cause crashes on some of the sample code. Someone else may want
+to attempt a more stable port.
+
+PORTING NOTES TO AMIWIN
+=======================
+
+My strategy of porting was to leave as much of the code untouched as
+possible. I surrounded any amiga specific changes with 
+#ifdef AMIWIN ... #endif or #ifndef AMIWIN ... #endif preprocessor
+symbols. The code  was ported on an Amiga 2000, with Fusion 40 accelerator
+and a Picasso II graphics card. The SAS/C 6.56 compiler was used, with
+the AmiWin 2.16 X development kit.
+
+All compilations were done for a 68040 CPU with 68882 math coprocessor for
+maximum  speed. Please edit the smakefile for other compilers.
+I wrote smakefiles for the directories I ported. I omitted the Windows
+and Widgets directories. The former is for MS Windows and the latter 
+requires Motif, which is not easily available for the Amiga.
+
+Here are the changes I did per directory:
+
+* mesa
+Nov. 25, 1995 v 1.2.4
+  - added a mklib.amiwin shell script that will make all the libraries and
+    sample code for Mesa
+  - created this readme file: readme.AMIGA
+
+* mesa/include
+Dec. 1, 1995 v 1.2.5
+  - added the following to GL/xmesa.h 
+     #ifdef AMIWIN
+     #include <pragmas/xlib_pragmas.h>
+     extern struct Library *XLibBase;
+     #endif
+NET CHANGE: xmesa.h
+
+* mesa/src 
+Nov. 25, 1995 v 1.2.4
+  - added the necessary pragma calls for X functions to the following:
+    xmesa1.c, xmesa2.c, xmesa3.c, xfonts.c, glx.c 
+    This prevents undefined symbols errors during the linking phase for 
+    X library calls
+  - created smakefile
+Dec.  1, 1995 v 1.2.5
+  - removed AMIWIN includes from xmesa1.c, xmesa2.c, xmesa3.c, xfonts.c, 
+    glx.c since they are now defined in include/GL/xmesa.h
+NET CHANGE: smakefile
+   
+* mesa/src-tk
+Nov. 25, 1995 v 1.2.4
+  - added the necessary pragma calls for X functions to the following:
+    private.h
+  - created smakefile
+Dec.  1, 1995 v 1.2.5
+  - removed AMIWIN includes from private.h since it is now defined in
+    include/GL/xmesa.h
+NET CHANGE: smakefile
+
+* mesa/src-glu
+Nov. 25, 1995 v 1.2.4
+  - created smakefile
+NET CHANGE: smakefile
+
+* mesa/src-aux
+Nov. 25, 1995 v 1.2.4
+  - added the necessary pragma calls for X functions to the following:
+    glaux.c
+  - created smakefile
+NET CHANGE: glaux.c, smakefile
+
+* mesa/demos
+Nov. 25, 1995 v 1.2.4
+  - added the necessary pragma calls for X functions to the following:
+    xdemo.c, glxdemo.c, offset.c
+  - created smakefile
+  - put #ifndef AMIWIN ... #endif around sleep() calls in xdemo.c since 
+    they are not part of AmigaDOS.
+Dec.  1, 1995 v 1.2.5
+  - removed AMIWIN defines from xdemo.c, glxdemo.c, offset.c since
+    already defined in include/GL/xmesa.h
+  - modified Smakefile to include header and includes from the AmiTCP4.0
+    net.lib linkable library to provide unix-compatible sys/time.h and
+    the sleep() function
+    - removed AMIWIN defines in xdemo.c since sleep() now defined
+NET CHANGE: smakefile
+
+* mesa/samples
+Nov. 25, 1995 v 1.2.4
+  - added the necessary pragma calls for X functions to the following:
+    oglinfo.c
+  - created smakefile
+  - put #ifndef AMIWIN ... #endif around sleep() in blendxor.c
+  - removed olympic from smakefile targets since <sys/time.h> not defined
+Dec.  1, 1995 v 1.2.5
+  - removed AMIWIN defines from oglinfo.c, since already defined in 
+    include/GL/xmesa.h
+  - modified Smakefile to include header and includes from the AmiTCP4.0
+    net.lib linkable library to provide unix-compatible sys/time.h and
+    the sleep() function
+    - removed AMIWIN defines in blendxor.c for sleep()
+    - added AMIWIN defines around _MACHTEN_ in olympic.c since xrandom()
+      functions are not defined in any libraries
+    - added olympic back into the Smakefile targets
+NET CHANGE: smakefile, olympic.c
+
+* mesa/book
+Nov. 25, 1995 v 1.2.4
+- created smakefile
+- removed accpersp and dof from smakefile targets since the SAS/C compile seems to
+  confuse the near,far variables with near/far memory models.
+NET CHANGE: smakefile
+
+* mesa/windows
+Dec.  1, 1995 v 1.2.5
+- Removed directory to save space since this is only needed for Windows based 
+  machines.
diff --git a/docs/README.GGI b/docs/README.GGI
new file mode 100644
index 00000000000..e28d047732c
--- /dev/null
+++ b/docs/README.GGI
@@ -0,0 +1,172 @@
+                       LibGGI driver for Mesa-3.0
+		  by Uwe Maurer ([email protected])	
+
+
+Introduction
+============
+[from libggi.txt by Steve Cheng and Hartmut Niemann]
+
+ "LibGGI, the dynamic GGI (General Graphics Interface) library is a
+  flexible drawing library.
+
+  It provides an opaque interface to the display's acceleration
+  functions. It was originally intended to allow user programs to
+  interface with KGI, the kernel side of the GGI code, but other display
+  types can be easily used by loading the appropriate "display target"
+  (e.g. X, memory).
+
+  LibGGI consists of a main library (libggi.so) and a multitude of
+  dynamic drivers. The library then loads the necessary "drivers" for
+  the requested mode, taking hints from the graphics device if
+  necessary. LibGGI can also load extension libraries, e.g. to provide
+  enhanced 2D and 3D functions.
+
+  It has been designed after having a look at several existing
+  libraries, and so far we have found porting to be quite simple from
+  and to most of them."
+
+----------------------------------------------------------------------------
+
+	More information about the GGI project and LibGGI can be 
+	obtained from the GGI website:
+
+	www.ggi-project.org
+
+----------------------------------------------------------------------------
+
+Installation
+============
+
+	- Install LibGGI
+	
+	- Unpack the Mesa archives
+	
+	- In the Mesa directory type:
+		
+		make linux-ggi
+		su
+		make linux-ggi-install
+		exit
+	
+	- Now you can try some demos.
+	  If they don't work, you can set the GGIMESA_DEBUG
+	  variable to 255 and you will see some information from the
+	  LibGGI-driver.
+		 
+	   export GGIMESA_DEBUG=255
+	
+
+GLUT
+====	
+
+	You can change these default values in ggi/ggiglut.c:
+	#define WIDTH   640
+	#define HEIGHT  400
+	#define GRAPHTYPE_RGB   GT_16BIT
+	#define GRAPHTYPE_INDEX GT_8BIT
+
+	Options:
+		-bpp x		Set graphic mode with x bits per pixel	
+		-size x y	Screen (or window) is x*y pixels 
+
+	Example:
+		demos/gears -size 320 200 -bpp 24
+
+Updates
+=======
+	
+	You can find the latest LibGGI-driver and ggiglut on my
+	homepage:
+	
+	http://home.t-online.de/home/uwe_maurer/ggimesa.htm
+	
+		
+
+Uwe Maurer  -  [email protected]
+
+                       LibGGI driver for Mesa-3.0
+		  by Uwe Maurer ([email protected])	
+
+
+Introduction
+============
+[from libggi.txt by Steve Cheng and Hartmut Niemann]
+
+ "LibGGI, the dynamic GGI (General Graphics Interface) library is a
+  flexible drawing library.
+
+  It provides an opaque interface to the display's acceleration
+  functions. It was originally intended to allow user programs to
+  interface with KGI, the kernel side of the GGI code, but other display
+  types can be easily used by loading the appropriate "display target"
+  (e.g. X, memory).
+
+  LibGGI consists of a main library (libggi.so) and a multitude of
+  dynamic drivers. The library then loads the necessary "drivers" for
+  the requested mode, taking hints from the graphics device if
+  necessary. LibGGI can also load extension libraries, e.g. to provide
+  enhanced 2D and 3D functions.
+
+  It has been designed after having a look at several existing
+  libraries, and so far we have found porting to be quite simple from
+  and to most of them."
+
+----------------------------------------------------------------------------
+
+	More information about the GGI project and LibGGI can be 
+	obtained from the GGI website:
+
+	www.ggi-project.org
+
+----------------------------------------------------------------------------
+
+Installation
+============
+
+	- Install LibGGI
+	
+	- Unpack the Mesa archives
+	
+	- In the Mesa directory type:
+		
+		make linux-ggi
+		su
+		make linux-ggi-install
+		exit
+	
+	- Now you can try some demos.
+	  If they don't work, you can set the GGIMESA_DEBUG
+	  variable to 255 and you will see some information from the
+	  LibGGI-driver.
+		 
+	   export GGIMESA_DEBUG=255
+	
+
+GLUT
+====	
+
+	You can change these default values in ggi/ggiglut.c:
+	#define WIDTH   640
+	#define HEIGHT  400
+	#define GRAPHTYPE_RGB   GT_16BIT
+	#define GRAPHTYPE_INDEX GT_8BIT
+
+	Options:
+		-bpp x		Set graphic mode with x bits per pixel	
+		-size x y	Screen (or window) is x*y pixels 
+
+	Example:
+		demos/gears -size 320 200 -bpp 24
+
+Updates
+=======
+	
+	You can find the latest LibGGI-driver and ggiglut on my
+	homepage:
+	
+	http://home.t-online.de/home/uwe_maurer/ggimesa.htm
+	
+		
+
+Uwe Maurer  -  [email protected]
+
diff --git a/docs/README.MINGW32 b/docs/README.MINGW32
new file mode 100644
index 00000000000..75709ed43d8
--- /dev/null
+++ b/docs/README.MINGW32
@@ -0,0 +1,123 @@
+August 30, 1998 -- Paul Garceau ([email protected])

+

+DISCLAIMER:  I make this extension to the Mesa 3-D Graphics Library as a service

+to the general public.  I can, in no way support or make any guarantee that the

+EGCS-Mingw32 build or any Gnu-Win32 build will work for your system. The

+associated packages and batch files I have included as part of the EGCS-Mingw32

+extension are provided "As-is" with out any guarantee of support or functionality

+from the author of this EGCS-Mingw32 native windows port of the Mesa 3-D Graphics

+Library.

+

+	Feel free to modify or change things as you see fit, just remember that

+I can't support any modifications you might want to make to the files which I

+have included OR the lgpl protected Mesa 3-D Graphics Library.

+

+

+	EGCS-Mingw32 Beta 3.08 Archive Manifest:

+		mingw32.bat

+		src/makefile.nt4

+		src/wmesa.c

+		src-glu/makefile.nt4

+

+###############

+

+Greetings,

+

+	In order to build the Mingw32 set of Mesa 3-D Graphics Library for Beta3.08

+it will be necessary for you to use the Dos or Command Prompt that is available

+on most of the i86 based MS Windows machines.  Also, I believe that this build

+will run on Win95, Win98, WinNT4 and WinNT5.

+

+	I haven't tested Win95/98 or WinNT5.  This build was generated under

+WinNT4 with SP3 installed.

+

+	This has not been tested under any systems outside of

+a WinNT4 Workstation with EGCS-Mingw32 toolchain, v.1.0.2 installed.

+

+	EGCS-Mingw32 uses a variation of gcc to handle its build.  The Mesa 3-D

+Graphics Library build that I have generated is based, in small part, on the

+Cygwin32 build and associated makefiles that Stephane Rehel ([email protected])

+defined back in 1997.  The EGCS-Mingw32 toolchain is capable of generating

+native windows code and, as of the date of this readme, can be obtained from:

+

+	http://www.xraylith.wisc.edu/~khan/software/gnu-win32/egcs-mingw32-102.html

+

+	Much thanks to the combined efforts of Mumit Khan, Jan-Jaap Vanderhagen

+and Colin Peters for making it possible for the EGCS-Mingw32 toolchain to exist.

+

+Installing EGCS-Mingw32 Build Revisions:

+

+	To install the makefile and source revisions incorporated with this build

+of the Mesa 3-D Graphics Library, you'll have to use a version of winzip.  I am

+in the process of finding a suitable Win32 compatible tar executable so that if

+you don't have winzip, you can still decompress the files into their respective

+folders/directories.

+

+	a) Move the mingw32.zip file to the top level of the hard drive on your

+system.

+

+	b) Copy all of the Beta 3.08 src/windows files to the src/ directory.

+

+	b) Open the Winzip file

+

+	c) Verify that the files will be properly extracted.

+

+	d) Extract the files with the Winzip "Overwrite" and "Use Folder Names"

+	   options enabled.

+

+	The zip file directory structure extraction defaults to the top level of

+the hard drive where the mingw32.zip file exists unless otherwise instructed by

+you.

+

+	The version of wmesa.c included with the mingw32 archive needs to replace

+the current version of the Beta 3.08 wmesa.c file in order for the egcs-mingw32

+build to work.  This is because the original Win32 stuff assumes that the glut

+utilities are to be installed.  The Glut utilities are not part of the

+egcs-mingw32 build for Beta 3.08.

+

+

+Build Considerations:

+

+	In order to get the build to work, I needed to create a special makefile

+for each library which the Mesa 3-D Graphics Library requires since there is no

+comparable make-config/config on a native windows platform.

+

+	Since I was only creating a few of the possible libraries for 

+Mesa (gl, glu), I only created the new make files in their respective libraries

+src, src-glu).  For libMesaaux.a. you will find a makefile for it in the

+src-aux directory.  libMesatk.a and libglut.a were not ported.

+

+	The build itself is a .bat based build and uses Gnu Make,Version 3.76.1 to

+process the makefiles noted above.  The build must be run from the directory

+where the mingw32.bat file is.  You can get the binary version of Make 3.76.1

+from Jan-Jaap van der Heijden's site in Germany:

+

+	http://agnes.dida.physik.uni-essen.de/~janjaap/mingw32/download.html

+

+	It was necessary to modify some source code, specifically the source code

+in the src-glu directory.  I needed to modify nurbs.c, quadric.c and tess.c in

+order to get them to work using the EGCS-Mingw32 toolchain.

+

+	The original EGCS-Mingw32 Toolchain, is available from:

+

+	http://www.xraylith.wisc.edu/~khan/software/gnu-win32/egcs-mingw32-102.html	

+

+Running the Build:

+

+	Ok, now that we've got the basics out of the way, follows is all you need

+to do in order to build the EGCS-Mingw32 version of libMesaGL.a and libMesaGLU.a:

+

+	Open your Command Prompt/Dos prompt.

+	Go to your Mesa-3.0 beta 'root' directory.

+		This is the same directory that the Mesa mingw32.zip file was

+		originally stored in if you've installed the Mesa-3.0 beta 3-D

+		Graphics Library source as outlined in the "readme" file included

+		with the Mesa-3.0 beta distribution.

+	At the command line type: mingw32

+		mingw32 is the .bat file that actually does the build.

+

+	Enjoy!

+

+	Peace,

+

+		Paul G. ([email protected])
\ No newline at end of file
diff --git a/docs/README.MITS b/docs/README.MITS
new file mode 100644
index 00000000000..a89176a62e6
--- /dev/null
+++ b/docs/README.MITS
@@ -0,0 +1,102 @@
+
+			Mesa 3.0 MITS Information
+
+
+This software is distributed under the terms of the GNU Library
+General Public License, see the LICENSE file for details.
+
+
+This document is a preliminary introduction to help you get
+started. For more detaile information consult the web page.
+
+http://10-dencies.zkm.de/~mesa/
+
+
+
+Version 0.1 (Yes it's very alpha code so be warned!)
+Contributors: 
+  Emil Briggs    	([email protected])
+  David Bucciarelli 	([email protected])
+  Andreas Schiffler 	([email protected])
+
+
+
+1. Requirements:
+     Mesa 3.0.
+     An SMP capable machine running Linux 2.x
+     libpthread installed on your machine.
+
+
+2. What does MITS stand for?
+     MITS stands for Mesa Internal Threading System. By adding
+     internal threading to Mesa it should be possible to improve
+     performance of OpenGL applications on SMP machines.
+
+
+3. Do applications have to be recoded to take advantage of MITS?
+     No. The threading is internal to Mesa and transparent to
+     applications.
+
+
+4. Will all applications benefit from the current implementation of MITS?
+     No. This implementation splits the processing of the vertex buffer
+     over two threads. There is a certain amount of overhead involved
+     with the thread synchronization and if there is not enough work
+     to be done the extra overhead outweighs any speedup from using
+     dual processors. You will not for example see any speedup when
+     running Quake because it uses GL_POLYGON and there is only one
+     polygon for each vertex buffer processed. Test results on a
+     dual 200 Mhz. Pentium Pro system show that one needs around
+     100-200 vertices in the vertex buffer before any there is any
+     appreciable benefit from the threading.
+
+
+5. Are there any parameters that I can tune to try to improve performance.
+     Yes. You can try to vary the size of the vertex buffer which is
+     define in VB_MAX located in the file src/vb.h from your top level
+     Mesa distribution. The number needs to be a multiple of 12 and
+     the optimum value will probably depend on the capabilities of
+     your machine and the particular application you are running.
+
+
+6. Are there any ways I can modify the application to improve its
+   performance with the MITS?
+     Yes. Try to use as many vertices between each Begin/End pair
+     as possbile. This will reduce the thread synchronization
+     overhead.
+
+
+7. What sort of speedups can I expect?
+     On some benchmarks performance gains of up to 30% have been
+     observerd. Others may see no gain at all and in a few rare
+     cases even some degradation.
+
+
+8. What still needs to be done?
+     Lots of testing and benchmarking.
+     A portable implementation that works within the Mesa thread API.
+     Threading of additional areas of Mesa to improve performance
+     even more.
+
+
+
+Installation:
+
+   1. This assumes that you already have a working Mesa 3.0 installation
+      from source.
+   2. Place the tarball MITS.tar.gz in your top level Mesa directory.
+   3. Unzip it and untar it. It will replace the following files in
+      your Mesa source tree so back them up if you want to save them.
+
+
+	 README.MITS
+         Make-config
+	 Makefile
+	 mklib.glide
+         src/vbxform.c
+	 src/vb.h
+
+   4. Rebuild Mesa using the command
+
+          make linux-386-glide-mits
+
diff --git a/docs/README.NeXT b/docs/README.NeXT
new file mode 100644
index 00000000000..52a62a32650
--- /dev/null
+++ b/docs/README.NeXT
@@ -0,0 +1,6 @@
+The NeXT support has now been incorproated into the OpenStep support.
+You can build NeXT libraries simply by typing "make next", though before
+linking they will need to be ranlib'd by hand. For more information see
+the README.OpenStep file, together with the README files in OpenStep/Old_Demos.
+
+-Pete French. ([email protected]) 28/5/98
diff --git a/docs/README.OS2 b/docs/README.OS2
new file mode 100644
index 00000000000..35b3e7a3294
--- /dev/null
+++ b/docs/README.OS2
@@ -0,0 +1,27 @@
+                README for port of Mesa to XFree86 on OS/2
+                           (as of 19980802)
+
+
+Instructions to build Mesa for XFree86/OS2:
+
+You need a recent version of XFree86 (3.3x or above) installed including
+the supplied programming libraries and tools as well as EMX 0.9c (and above).
+
+Beginning after beta 7 there's again support for creating DLLs.
+The details are handled in "mklib-emx.cmd" a small REXX script.
+By now it does ensure compatiblity by using the function names as
+entry points instead of ordinals. This will cost performance and
+might be fixed in a future patch.
+
+We switched to the usual build method
+(based on Makefile and make-config) beginning with Mesa 3.0 beta 5.
+To use most of the standard files (including shell scripts) you should
+have a un*x shell (sh) in path.
+
+To actually build the (static) libraries and demos type
+  make os2
+
+
+Alexander Mai
[email protected]
[email protected]
diff --git a/docs/README.OpenStep b/docs/README.OpenStep
new file mode 100644
index 00000000000..c4f504c22a2
--- /dev/null
+++ b/docs/README.OpenStep
@@ -0,0 +1,28 @@
+This is a port of Mesa-3.0 to OpenStep and Rhapsody/YellowBox. Only
+the GL and GLU libraries have been ported. As OpenStep has it's own
+window handling code we simply use the offscreen rendering capability
+of Mesa to generate a bitmap which can then be drawn into a View. An
+example application using Mesa can be found in OpenStep/MesaView.
+Currently only static libraries are built. The code has been tested on the
+Intel hardware version of the following systems:
+
+	OpenStep for Mach 4.2
+	Rhapsody (DR1)
+	YellowBox for NT4 (DR1)
+
+It should, however, work on all other variants of OpenStep for other
+processors without modification. Feedback on this would be appreciated.
+
+To build on UNIX based systems simply type "make openstep".
+
+To build on Win95/WinNT based systems run the "win32-openstep.sh" script from
+the Bourne shell provided with the development environment.
+
+Thiss build the libraries, places them in the "lib" directory and also builds
+the "MesaView" example application. Older examples may be found in the
+OpenStep/Old_Demos directory. These only work on UNIX based systems. The CC
+variable is passed around by the Makefiles so fat libraries may be created
+by alreting this on the command line, e.g. for m68k and i486 support you
+can use the command "make CC='cc -arch m68k -arch i386' openstep".
+
+-Pete French. ([email protected]) 28/5/98
diff --git a/docs/README.VMS b/docs/README.VMS
new file mode 100644
index 00000000000..1763f700e3c
--- /dev/null
+++ b/docs/README.VMS
@@ -0,0 +1,15 @@
+
+VMS support contributed by Jouk Jansen ([email protected])
+
+
+I compiled the libs on a VMSAlpha7.0 system using DECC5.3.
+But earlier versions may work as well.  The make files were tested
+use the DIGITAL make utility called MMS.  There is also a public domain
+clone available (MMK) and I  think, but it is not tested, that this
+utility will give no problem.
+
+To make everything just type MMS (or MMK) in the main directory of
+mesagl.  For MMS the deafult makefile is called descrip.mms, and
+that is what I have called it.  I included alse some config files,
+all having mms somewhere in the name which all the makefiles need
+(just as your unix makefiles).
diff --git a/docs/README.WIN32 b/docs/README.WIN32
new file mode 100644
index 00000000000..6382624f568
--- /dev/null
+++ b/docs/README.WIN32
@@ -0,0 +1,739 @@
+
+    Mesa/Readme.win32
+
+    Last Updated: Sun, Dec 06 1998, 08:49 EST - [email protected]
+
+    Note: While this build set supports generation of a 3Dfx specific
+          DLL using Mesa, David Bucciarelli's original build files
+          are the "supported" method. -Ted
+
+*** What's New
+
+- Build environment change: The Glide SDK is no longer assumed to be in
+  the global INCLUDE/LIB environment vars, it is required that you set the
+  value 'GLIDE2X' as either an environment variable pointing to your Glide
+  SDK install directory or that you configure that as a build option to
+  nmake.exe when building fxmesagl32.  Examples:
+
+    nmake /f nmake.mak GLIDE2X=g:\sdk\glide2x fxmesagl32
+
+          <or>
+
+    nmake /f nmake.mak GLIDE2X=g:\sdk\glide2x allfx
+
+          <or>
+
+    nmake /f nmake.mak GLIDE2X=g:\sdk\glide2x progs.3dfx.demos
+
+  The DevStudio workspace files for 3Dfx OpenGL require the definition of
+  GLIDE2SDK as an environment variable pointing to where your copy of the
+  Glide SDK has been installed. Adding this to your AUTOEXEC.BAT would do
+  so (change the directories to match):
+
+       SET GLIDE2SDK=G:\SDK\GLIDE2X
+
+- Static build/link of all non-3Dfx targets works now, cool. Required some
+  nontrivial chicanery in the GLUT headers though to allow it to use the
+  mesa 'wgl' functions statically linked instead of always importing the
+  sytem wgl functions from GDI32. Mostly this was an extension of an
+  existing method used in the headers, just taking into account more
+  functions.
+
+- Handled an undocumented change in how Microsoft NMAKE for DevStudio 6
+  defines _NMAKE_VER so that the makefiles should work properly.
+
+- Microsoft Compiler/Win32 specific build tuneups in the header files
+  providing for dramatically faster compilation times of mesa itself and of
+  code using Mesa. Primarily via the removal of any call sequence which
+  would invoke WINDOWS.H (which triggers ~6-10K lines of code inclusion).
+
+  Supporting this, a macro name change was made in the gl.h/glu.h/glut.h
+  header files which should be client-code transparent. The change was to
+  specify function prototypes with the (newly created) macro GLAPI instead
+  of WINGDIAPI to ease conflicts with multiple definitions of WINGDIAPI
+  (needs to be one thing to access Win32 API calls, another to define
+  export funcs when building a Mesa/GLU/GLUT DLL which themselves need to
+  import functions from Win32, etc.).
+
+  Similarly, call sequence definition usage of APIENTRY, CALLBACK, and
+  WINAPI were renamed to prevent collision/need for windows.h.
+
+  Lastly, these changes were incorporated into the GLUT source to
+  facilitate same compile-time performance increases.
+
+  All of these changes should be transparent to code USING
+  mesa/mesaglu/glut, and in most cases did not require changes in their
+  source files within the Mesa/GLUT code base (but there were a few).
+
+- 'wgl' function prototypes and supporting data types declared in gl/gl.h
+  to provide for the usage of these from client Windows code without
+  requiring WINDOWS.H to be included. If you're using non-trivial functions
+  (such as any that pass structures around) you will need to include
+  WINDOWS.H *before* gl/gl.h, however in that case you needed them anyway.
+
+  One benefit of this is that glut-based code on Windows may now use
+  wglGetProcAddress to get the function pointers for all supported OpenGL
+  extensions, preventing the need for directly importing them into
+  code. The 3Dfx/Demos/glbpaltx.c has been modified to use this mechanism
+  by way of example.
+
+  This code has also been copied into gl/glut.h to support glut programs on
+  Windows that do not use Mesa.
+
+- new command line build target: all.debug
+
+  Builds all DLL files and test programs in release, then debug builds,
+  emplacing the requried DLL files in the EXE directories for ready test
+  runs.
+
+- Expanded MESA_WGL_FX options. When setting it to instruct fxMesa to
+  render windowed you may provide it a clue to your system's display, or
+  instruct it to try to detect the pixel format. This information is used
+  in constructing the windowed-rendering hack to try for optimal
+  performance, such as it is. You specify this by adding one of the
+  following tags in the envvar prefixed by a colon.
+
+    noconv    - does not do any pixel format conversion
+                This is synonymous to '16bgr'.
+    16bgr     - uses a 16-bit RGB sequenced dib section and does no pixel
+                format conversion as this is the internal format of the
+                3Dfx hardware when used by fxMesa.
+                This is synonymous to 'noconv'.
+    16rgb     - uses a 16-bit RGB sequenced dib section and does pixel
+                format conversion each frame from the 3Dfx internal format
+                to 5:6:5 RGB.
+    15rgb     - uses a 16-bit RGB sequenced dib section and does pixel
+                format conversion each frame from the 3Dfx internal format
+                to 5:5:5 RGB.
+    15bgr     - uses a 16-bit RGB sequenced dib section and does pixel
+                format conversion each frame from the 3Dfx internal format
+                to 5:5:5 RGB.
+    24rgb     - uses a 24-bit RGB sequenced dib section and does pixel
+                format conversion each frame from the 3Dfx internal format
+                to 8:8:8 RGB. Also suitable for a 32-bpp display.
+    24bgr     - uses a 24-bit BGR sequenced dib section and does pixel
+                format conversion each frame from the 3Dfx internal format
+                to 8:8:8 RGB. Also suitable for a 32-bpp display.
+    query     - Queries Windows for the displays pixel format and attempts
+                to adjust accordingly. This query is done by using
+                DirectDraw to allocate a "Primary" surface and then
+                checking what that surface's pixel format is (as the Win32
+                API for getting the pixel format from a window device or
+                screen device context is useless as far as I have been able
+                to determine).  This does *NOT* mean that fxMesa is now
+                requiring DDraw, however it will use it if found. This
+                prevents system incompatabilities with WinNT 4SP2 and
+                earlier systems.
+
+    Example, put this in your AUTOEXEC.BAT to have fxMesa detect a best
+    case match on context creation:
+
+         set MESA_WGL_FX=windowed:query
+
+    Additionally, if you wish to find out *what* it detected, you may add a
+    ":logged" to this which will cause it to create/append to a file named
+    FXMESAGL32.LOG whenever a context is created, e.g.:
+
+         set MESA_WGL_FX=windowed:query:logged
+
+    *** WARNING: Some display drivers/windows version combinations do not
+        support all of the possible DIB section formats that can be
+        requested, and thanks to Microsoft there is no way of detecting
+        (that I have yet discovered) that the requested DIB section will
+        not work and when it is utilized the application crashes.
+
+  Eventually I hope to incorporate support for DirectDraw Overlay surfaces
+  to facilitate less data crunch overhead when dealing with the in-window
+  hack. For example: if a primary display support overlays of BGR565
+  sequence then the data could be read from the 3Dfx framebuffer into a
+  DDraw "System Memory" or "Non-Local Video Memory" surface - both of which
+  actually reside in on-board DRAM (the latter in AGP space) which could
+  then be blitted/flipped in a much faster fashion than the current DIB
+  section handling.
+
+  The DIB section handling will always be the default, as that's much more
+  compatible since it utilizes GDI features sets that have more longevity
+  to them.
+
+*** Legalese
+
+These build files are provided as-is and are submitted to be included with
+the "Mesa 3-D Graphics Library" package as (currently) maintained by Brian
+Paul. These project build files are free software; you can redistribute it
+and/or modify it under the terms of the GNU Library General Public License
+as published by the Free Software Foundation; either version 2 of the
+License, or (at your option) any later version.
+
+These project files are distributed in the hope that they will be useful,
+but WITHOUT ANY WARRANTY; without even the implied warranty of
+MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU Library
+General Public License for more details.
+
+You should have received a copy of the GNU Library General Public License
+along with this library; if not, write to the Free Software Foundation,
+Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
+
+*** Maintenance Responsiblity and Technical Support
+
+While these files are now part of the Mesa core distribution please do NOT
+contact Mr. Paul for help with them if you encounter problems as he can't
+help you (currently).  I will, however, attempt my straightforward best in
+assisting anyone with using these files on their system.  I can NOT
+guarantee instant responses owing to other responsiblities, but I do try
+dang hard to answer any mail w/in 24 hours.  I may be contacted at the
+above email address for the forseeable future.
+
+-Ted
+mailto://[email protected]
+http://www.i21.com/~tjump
+
+*** General Information
+
+These build files facilitate convenient building of many variants of Mesa,
+both as static link libraries (including mesaglu) and as dynamic link
+libraries that in some cases may be used as "drop-in" replacements for
+OpenGL32.DLL on both Windows95 and Windows NT.
+
+The construction of the Win32 command-line build files and projects has
+been something of a pet project of mine, and is based upon my own
+"standard" Win32 build environment as supplied by the "nmake.mif" file.
+They have been tested under Windows95 OSR2, Windows NT 4.0SP3, and Windows
+NT 5.0 beta 1.  The libraries that they generated have been tested (via the
+demo programs) in a *limited* fashion on the above three systems, including
+the 3Dfx versions.
+
+The reason I went with command-line build environment instead of the more
+convenient IDE-based project files is for two reasons: 1. These appear to
+have some amount of portability between versions (the nmake syntax hasn't
+changed much since Microsoft C 7.0) while the IDE project files seem to
+change drastically each version. and 2. These are readable with any ascii
+editor and such are better self-documentation of the file relationships for
+more people such that it will facilitate supporting other Win32 compilers.
+
+While these files only deal with building for x86 targeted code it *should*
+be possible to add the necessary logic to them to build for the other MSVC
+supported CPU targets, I simply have no hardware to test them on nor the
+alternative compilers to build with.
+
+*** Prerequisites for use
+
+1. You must have a 32-bit Microsoft compiler installed. I have tested
+this with Visual C 5.0 (SP3) and Visual C 4.2, but with minor
+(possibly no) modification to the nmake.mak and nmake.mif files this
+sequence should work on Visual C 2.0 also. The workspace files
+(mesalib.dsw and mesademos-*.dsw) and their included project files
+(*.dsp) are specific to the DevStudio IDE - I have made no attempt at
+building a VC4 IDE project set as I do not use that any more.  Note
+that the VC workspace files NO LONGER use NORE are dependant upon the
+nmake.mak and nmake.mif files for construction of definition (*.DEF)
+and resource (*.RC) files.
+
+*** Visual C 4.x Users Warning ****
+
+Note that early editions of VC4 do NOT have header files current enough
+for use building this code base. If you are using VC4 you will either need
+to get an update to version 4.2 *or* you may download the Platform SDK
+directly from Microsoft's web site (www.microsoft.com) and update your
+build environment that way.
+
+*** Visual C 4.x Users Warning ****
+
+2. You must have the PATH, INCLUDE, and LIB environment variables set
+properly. With VC5 you can easily get this by executing the VCVARS32.BAT
+file that was created for you upon installation. It is found in the
+DevStudio\VC\BIN directory, wherever you installed DevStudio. VC4 provides
+a similar batch file in it's BIN directory also.
+
+3. (optional) If you're going to build for 3Dfx/Voodoo you will need to
+have previously installed the Glide SDK version 2.3 or later, if I
+recall. This may be retrieved from www.3dfx.com for no money and some
+download time. ;-) These build files assume that you have the Glide SDK
+added to the respective environment variables (LIB and INCLUDE).
+
+4. (optional) If you're going to build for S3/Virge you will need the S3
+Developers Toolkit which may be downloaded from www.s3.com for the price of
+registering on-line and some time. NOTE: I can build the s3mesa.dll file to
+completion, however the compilation of s3mesa.c currently generates a large
+amount of compiler warnings and between that and the fact that I can not at
+all test it I can make no claims to it's ability to execute.  Again, like
+the 3Dfx version before this, these build files assume you have the S3Dtk H
+and LIB files in the path of their respective environment variables.
+Note 2: As of Mesa3.0beta6 I have build files, both command-line and IDE,
+which should be able to build the s3mesa code base if it weren't for updates
+being required in the S3 DD code support (Mesa-3.0/src/s3 directory).
+
+I advise putting any include and lib files for secondary toolkits (Glide,
+S3Tk, whatever) in their respective environment variables *before* the
+Microsoft-assigned default values.
+
+*** FAQ: Frequenty Asked Questions and Other Important Information ***
+
+- When running the 3Dfx demos under Windows NT, they crash on exit, what's
+  up?
+
+  This is apparently a problem in Glide itself. The workaround is to go to
+  your C:\WINNT\SYSTEM32 directory and rename the file FXOEM2X.DLL to
+  FXOEM2X.DL_ to prevent Glide from loading and initializing it upon
+  startup.  This is known to be an issue with cards that do not have "TV
+  out" and is known to cause crashes on Diamond Monster II 8M and 3Dfx
+  Reference boards, all using 3Dfx Reference Drivers version 2.53. Other
+  hardware/driver combinations will also likely exhibit this behavior.
+
+- I'm having a problem building Mesa for static library linking.
+
+  This was caused by some incomplete testing on my part, and a fix is now
+  available in the form of an add-on to the base Mesa 3.0 release.  The
+  file to get is:
+
+       via FTP download from: iris.ssec.wisc.edu
+         you want to go here: /pub/Mesa/patches_to_3.0/
+        you want to get file: Mesa-3.0-w32-static-fixes.tar.gz
+
+  This required a minor addition to INCLUDE/GL for a clean solution, the
+  file "include/gl/mesa_wgl.h" is automatically included by
+  "include/gl/gl.h" when a Win32 non-DLL build is in progress to provide
+  prototypes for the various wgl functions.
+
+  The only remaining hitch in this setup is that the 3Dfx build is not yet
+  running as a static build, because of problems with conflicts in
+  existance of the various GDI functions like ChoosePixelFormat,
+  etc. *sigh*
+
+  Anyway, the "allstatic" target now works as expected and builds all
+  book/sample/demos programs to boot. ;^)
+
+- How do I get fxMesa to render in a window on the desktop instead of only
+  full-screen?
+
+  Use the Microsoft Windows fxMesa-in-a-window hack!
+
+  Seriously, if you want fxMesaGL to render using the 3Dfx Voodoo1 or
+  Voodoo2 hardware into a window on the desktop then all you need to do is
+  set the MESA_WGL_FX environment variable to anything other than
+  "fullscreen" and it will render into a window.  If you wish to go
+  fullscreen then you only need to NOT have the environment variable, or
+  have it set to "fullscreen".  You may also switch at runtime between
+  fullscreen-mode and windowed by pressing ALT-ENTER on the keyboard
+  (unless the application using Mesa does something with those keystrokes,
+  of course).
+
+  As of 8/13/98 this should be running a LOT better for more people as a
+  low-compatability item was cleaned up which prevented it from working on
+  many (most?) display drivers under Windows 9x.
+
+- I have my 3Dfx card hooked to it's own monitor and I want the output to
+  stay on even if I switch to another program, is this possible?
+
+  If the Glide environment variable SST_DUALHEAD is set to '1' then fxMesa
+  will never disable the Voodoo output on a Voodoo1 or Voodoo2 display
+  regardless of whether the fxMesa application is "current" or not. This
+  works regardless of whether it's rendering using the window hack
+  mentioned above or not.
+
+- I want to run the Mesa demos on my Intel740 card using it's own OpenGL
+  acceleration, how do I do this?
+
+  Build GLUT standalone for use with system OpenGL and GLU drivers!
+
+  The Command-line project supports building all test/demo programs against
+  these drivers also! This allows you full use of GLUT on Windows using
+  hardware accelerated OpenGL. Wheee! This includes the "3dfx/demos"
+  directory of which only two programs will not run on "standard"
+  opengl. Note that there are a few of the sample programs which will NOT
+  work without Mesa as they directly call into Mesa instead of using the
+  extension mechanism.
+
+*** Included programs that exhibit unfortunate or bad behavior
+
+- demos/bounce - doesn't run on high-colors screens?  It's requesting an
+  INDEX display from GLUT and that fails on my true-color desktop. Changing
+  this to _RGB let's the program work, but it doesn't display
+  properly. This is probably just an idiosyncracy of my machine though, as
+  if I test the program using GLUT for System OpenGL on my Intel740 OpenGL
+  accelerated machine it's just hunky-dory.
+
+- demos/glutfx - runs, but crashes on exit (but not on my Intel740 machine)
+
+- demos/texobj - runs, but crashes on exit if ESC is pressed. Exits cleanly
+  if the Close box on the window frame is pressed with the mouse. Go figure.
+
+- book/aaindex - doesn't run, can't get pixel format, because it wants an
+  INDEX display maybe (but is okay on my Intel740 machine)?
+
+- most of the book/* demos don't respond to ESC being pressed.
+
+- 3dfx/demos/* - all demos run, however they all crash on exit. I've traced
+  this so far as to determine the call it's happening with. The crash comes
+  from within Glide during the processing of the grGlideShutdown() call, as
+  in invalid memory reference exception. I'm wondering if this is because
+  of some state or processing not being completed before the call. Dunno,
+  but putting grSstIdle() in just before grGlideShutdown() does NOT fix the
+  problem.
+
+- 3dfx/demos/tunnel2 - does not run on my system even with SLI mode
+  disabled. Hmmmm, maybe I need to disconnect my Voodoo2 cards?
+
+*** Important Notes and Changing Default values
+
+- The optimizer settings have been manually reworked in both command line
+  and DevStudio IDE files to hopefully prevent possible irrational code on
+  the part of the code generator.  Formerly, it was configured for "/Ox",
+  now it is configured for safer handling at a slight potential performance
+  cost. This may not be required for Visual Studio 6 but I can't test that
+  (yet).
+
+- These files build with the code targeted for Pentium processors and
+  8-byte structure padding.
+
+- The IDE-built programs seem to be "happier" in that the command line
+  build of the 3Dfx demo "fire" will grenade on exit (?). Otherwise pretty
+  much everything may be built with either interface.
+
+- The currently configured Mesa version is 3.1, and MesaDemos version is
+  the same. To change this permanently you will need to edit NMAKE.MAK and
+  change the lines that look like this (they start o/a line 116):
+
+    # Currently, Mesa is at rev 3.1 ...
+    #
+    !IF "$(MESAVER)" == ""
+    MESAVER=3.1
+    !ENDIF
+
+    # used in building all of the resource files for the Mesa DLLs
+    #
+    !IF "$(MESAFILEVER)" == ""
+    MESAFILEVER=3,1,0,0
+    !ENDIF
+
+- Currently the build files are configured to be used from a Win32
+  directory that is included inside the main Mesa-3.1 heirarchy.
+
+- The build files are smart enough to find the files for the core lib, glu,
+  glut, and the various demo programs if they are unpacked in the current
+  Mesa-3.1 heirarchy, like this:
+
+    \Mesa-3.1
+    \Mesa-3.1\src
+    \Mesa-3.1\src-glu
+    \Mesa-3.1\src-glut
+    \Mesa-3.1\Win32
+    \Mesa-3.1\samples
+    \Mesa-3.1\demos
+    \Mesa-3.1\book
+    \Mesa-3.1\3Dfx\demos
+
+    ... should work.  This arose because my initial build tests for the
+    demo files were done before MesaDemos 2.6 had been released.
+
+- To enable use of MMX instructions by the VC5 compiler you may add the
+  "USE_MMX=1" option to the NMAKE command line, or edit the default in the
+  NMAKE.MAK file.  This does appear to have some affect on the performance
+  on the library and does not seem to harm it in any way *but* I have done
+  *no* verification of this. I have an MMX processor so I figured what the
+  heck. This option is only available with VC5 when building from the
+  command line.
+
+  This is probably required if you are going to modify the code to include
+  inline MMX instructions though.
+
+- With the exception of the static link libraries generated by this file
+  set (mesagl.lib, mesaglu.lib, mesaglut.lib) all DLLs and executables are
+  built against the "Multithreaded DLL" runtime - this means that they
+  require MSVCRT.DLL or MSVCRTD.DLL in the path to execute.
+
+  ** CHANGED 8/11/98 ***
+
+  Note also that the demos are all built aginst the "OpenGL32, GLU32, and
+  GLUT32" and as such they are fairly agnostic wrt: building against Mesa
+  for CPU-rendering, Mesa-for-3Dfx, Mesa-for-S3, or System OpenGL.
+
+  If you want to build them for use on your system and your display card
+  provides full OpenGL acceleration (Permedia, Intel740, Intergraph,
+  whatever) then you only need to build GLUT prior to building any of the
+  demo programs. For convenience, the GLUT project is included in each of
+  the demo projects Workspace files for the DevStudio IDE builds BUT it is
+  not automatically built - you still need to build it first manually.
+
+  Note that if you have GLUT already installed on your system (gl/glut.h in
+  yoru INCLUDE path, glut32.lib/glut32d.lib in your LIB path, and the DLL
+  in your PATH) then you do NOT need to build GLUT prior to the test
+  programs.
+
+- The 3Dfx build of Mesa has primarily been tested with Quake 2 and it runs
+  (mostly) fine on my PC (take that for what you want it)...
+
+  ** CHANGED  8/11/98 ***
+
+  There is still something going on that causes Glide to crash on shutdown,
+  when I run fxMesa under Windows NT, however it does not appear to occur
+  under Windows 9x on either Voodoo1 or Voodoo2 cards. *sigh*
+
+- I can not test the S3 build as I have no machines available with Virge
+  based display cards.
+
+- The multithreaded test code is *not* built as it requires pthreads and I
+  have as of yet spent not time trying to get that running. The latest word
+  that I saw WRT threading support on win32 was that they are intending to
+  support it natively within Win32 - so I'm waiting it out until they get
+  it done.
+
+- Similarly, the 'xdemos' are not currently built because I haven't gotten
+  around to building the client libs for native win32 and getting it all
+  setup for use.
+
+*** Output Files
+
+All final output files (DLL/LIB) are placed in the Mesa-3.1/lib directory,
+with the exception of the fxMesaGL32 build which is placed in
+Mesa-3./lib/FX and the executable images which are placed in their source
+directories.
+
+To be able to execute the various test programs, you will need to copy the
+requisite DLL files into the same directory as the EXE files. Note that
+most of the 3Dfx/demos/* programs WILL run with the non-FX build of Mesa -
+just very slowly. The two programs which are hard-linked with the FX build
+and will not run without it are "glbpaltx" which uses "gl3DfxSetPaletteEXT"
+directly instead of via the extensions mechanism and "tunnel2" which uses
+"fxMesaSelectCurrentBoard" API for selecting between multiple 3Dfx cards
+installed in one system. Likewise, "paltex" directly uses the
+"glColorTableEXT" extension and thus may not run on anything except
+Mesa. If these applications used the proper extension mechanism they could
+then be used on more than "just" fxMesa to good effect (for example, the
+rest of the "3Dfx/demos" run just peachy on the Intel740 card in my test
+machine) under WinNT.
+
+Because I'm anal about my computer and it's organization, and I like to
+prevent collision between builds, each of the subprojects has their own
+intermediate file directory inside .\win32\release (for example, when
+building mesagl.lib all of it's intermediate files will be found in
+.\win32\release\lib.mesagl).  This makes it very easy to cleanup as you
+only need to remove .\win32\release.
+
+*** Okay, Enough, how do I build with this stuff already Ted!
+
+Okay, no major calamity here. The basic way to use the project file is to
+call it via NMAKE from the command line. The format is:
+
+    nmake[.exe] /f nmake.mak [options] [target]
+
+The most likely [options] values you will use may be any combination of the
+following:
+
+    DEBUG=1 or DEBUG=0
+    USE_MMX=1 or USE_MMX=0
+    USE_CRTDLL=1 or USE_CRTDLL=0
+
+    Note that all three of these options are OFF by default.
+
+The [target] includes but is not limited to the following (for full details
+please peruse the NMAKE.MAK and NMAKE.MIF files - but be warned that
+NMAKE.MIF is rather large and sometimes hard to follow):
+
+    --- convenience targets ---
+
+    all                 - builds everything
+    libfiles            - builds all linking library files
+    progs               - builds all executable images
+
+    --- library files, static and dynamic ---
+
+    mesagl              - static lib build of Mesa core.
+    mesaglu             - static lib build of MesaGLU core.
+    mesaglut            - static lib build of Mesa GLUT core.
+
+    mesagl32            - dynamic lib build of Mesa core.
+
+    mesaglu32           - dynamic lib build of GLU core, generates
+                          GLU32.DLL and/or GLU32d.DLL.
+
+    mesaglut32          - dynamic lib build of GLUT core, generates
+                          GLUT32.DLL and/or GLUT32d.dll.
+
+    --- hardware accelerated mesa builds ---
+
+    fxmesagl32          - builds Mesa for use on top of the 3Dfx
+                          Glide runtime libs
+
+    s3mesagl32          - builds mesa for use on top of the S3
+                          'S3Tk' runtime libs.
+
+    --- executable images ---
+
+    progs.book          - builds all programs in \book directory
+    progs.demos         - builds all programs in \demos directory
+    progs.samples       - builds all programs in \samples directory
+
+        These targets generate all of the programs in their respective
+        directories and link the executables against OpenGL32.DLL,
+        GLU32.DLL, and GLUT32.DLL (or their debug equivalents).
+
+    progs.3dfx.demos    - builds all programs in \3dfx\demos directory
+
+        This target generates the 3Dfx/Demo executables, linking them
+        against GLUT32.DLL, GLU32.DLL, OPENGL32.DLL and are thus NOT
+        hard-bound to using Mesa per-se as you can simply NOT build the
+        Mesa core and GLU libraries.
+
+   --- Microsoft/SGI OpenGL-based GLUT and Demo program builds ----
+
+   *** IMPORTANT SAFETY TIP: If you're going to build these variants of
+       GLUT then DO NOT build any other target libraries in this package
+       first, OR from the command line run the "nmake /f nmake.mak clean"
+       command first!  This is because generation of the GLUT for SGI
+       OpenGL target libraries conflicts in naming with the static build
+       libraries of Mesa and it's supporting GLUT build.
+
+   Currently, you may build GLUT as either GLUT32.DLL or GLUT.DLL for
+   use running against either Microsoft or SGI OpenGL for Window,
+   respectively.  This allows for the general use of GLUT 3.7 on Windows
+   systems with fully compliant OpenGL.
+
+   You can build the GLUT DLL files either with the command line by
+   issuing either of these commands:
+
+        nmake /f nmake.mak glut.sysgl
+
+        <or>
+
+        nmake /f nmake.mak glut.sgigl
+
+   OR by using the DevStudio MesaLib Worksapce build the GLUT_SGIGL or
+   GLUT_SYSGL projects within the DevStudio IDE.
+
+   Unfortunately, the only way to build the test programs against this
+   build of GLUT is via the command line, and I will NOT be making
+   duplicate demo program projects for the IDE as it's just not worth it,
+   sorry.
+
+   To build the test programs against either MS or SGI OpenGL, you do so
+   via either of these two commands:
+
+        nmake /f nmake.mak progs.sysgl
+
+        <or>
+
+        nmake /f nmake.mak progs.sgigl
+
+   To use the GLUT-for-system-OpenGL in your own programs, you need to do
+   three things by way of preparation, after building GLUT of course:
+
+         1. Copy include\gl\glut.h to somewhere in your %INCLUDE% path, one
+            likely candidate location would be in your
+            "DevStudio\VC\INCLUDE\GL" directory.
+
+         2. Copy the linking libraries to somewhere in your %LIB% path, one
+            likely candidate location would be in your "DevStudio\VC\LIB"
+            directory. The linking libraries you need to copy are as
+            follows:
+
+                .\Release\GLUT32.LIB
+                .\Release\GLUT.LIB
+                .\Debug\GLUT32.LIB
+                .\Debug\GLUT.LIB
+
+        3. Copy the runtime libraries to somewhere in your %PATH%, one
+           likely candidate location would be in WINDOWS\SYSTEM. the files
+           that you should copy are as follows:
+
+                .\Release\GLUT32.DLL
+                .\Release\GLUT32.PDB
+                .\Release\GLUT.DLL
+                .\Release\GLUT.PDB
+                .\Debug\GLUT32d.DLL
+                .\Debug\GLUT32d.PDB
+                .\Debug\GLUTd.DLL
+                .\Debug\GLUTd.PDB
+
+Some examples are in order ...
+
+    ... build all dynamic-link libs using MSVCRT.DLL for C runtime:
+
+        nmake /f nmake.mak USE_CRTDLL=1 alldynamic
+
+    ... To build all library variants and all test and demonstration
+        programs with the default settings you do this:
+
+        nmake /f nmake.mak all
+
+    ... to build all static link libs and nothing else you do this:
+
+        nmake /f nmake.mak allstatic
+
+    ... to build all non-accelerated dynamic link libs you do this:
+
+        nmake /f nmake.mak alldynamic
+
+    ... to build all 3Dfx targeted dynamic link libs you do this:
+
+        nmake /f nmake.mak allaccel
+
+    ... to build all S3 Virge targetd dynamic link libs you do this:
+
+        nmake /f nmake.mak alls3
+
+    ... to build all libraries, static and dynamic, in all versions
+        you do this:
+
+        nmake /f nmake.mak libfiles
+
+    ... to subsequently build all demo and test programs you do this:
+
+        nmake /f nmake.mak progs
+
+    ... to cleanup all intermediate files you do this:
+
+        nmake /f clean
+
+You get the picture. (I hope) ;^)  You may also specify specify
+single targets in a convenient fashion. The rule is simple, any of the
+above named lib files, static or dynamic, may be built by providing it's
+name on the command line as the target. Examples:
+
+    ... to build only Mesa as OpenGL32.DLL ...
+
+        nmake /f nmake.mak opengl32
+
+    ... to build only Mesa on top of the 3Dfx Glide API ...
+
+        nmake /f nmake.mak fxMesaGL32
+              <or>
+        nmake /f nmake.mak fxMesaGL
+
+    ... to build only Mesa on top of the S3 Toolkit ...
+
+        nmake /f nmake.mak s3MesaGL32
+              <or>
+        nmake /f nmake.mak s3mesaGL
+
+*** Revision history for ./win32 project files
+
+1/18/98 - initial cut submitted and included with core mesa
+2/5/98  - fixed internal dependency within nmake.mif upon there being
+          a $(DEVDIR) variable to make some temporary batch files
+          dependant upon (thanks to Keven T. McDonnell for finding
+          that there was this particular bug). I also updated the
+          build files for 2.6beta6.
+2/8/98  - added DevStudio workspace and project files for all lib
+          files and some test programs. Updated readme.win32.
+6/25/98 - initial revision for Mesa 3.0, does not include IDE files,
+          not everything is running. *sigh*
+7/20/98 - Mesa 3.0beta6 rev of all build files, all libs built and
+          minimally tested, all demo programs built and minimally
+          tested to within limits of my PC. ;^) Eveything looks
+          MUCH better now ...
+7/30/98 - Minor updates/edits based upon feedback from
+          Eero Pajarre <[email protected]>. These updates include a fix
+          to the Mesa-on-3Dfx build such that Quake-II now runs almost
+          properly on my system. It runs, just *very* slowly and with *no*
+          textures. Hmmm. Doesn't make any difference whether Quake is set
+          to use 8-bit textures or not.
+8/13/98 - Lots of build cleanups, minor bug fixes in fxwgl.c, and
+          compatability fix in fxapi.c for in-window rendering using 3Dfx
+          hardware.
+8/26/98 - Final revisions for Mesa 3 release checked
+9/22/98 - Fixed static builds for all but fxMesaGL32 and s3MesaGL32 targets
+9/29/98 - Reorganized FAQ information and added Added faq entry about Glide
+          bug under NT (crash on exit) and a workaround.
+11/21/98 - Updated files for Mesa 3.1 beta 1
+           Updated fxMesa window-hack code
+           Updated fxMesa resolution support to handle 1600x1200 & 1280x1024