should remove that file and just have people regenerate it. Also update
the Makefile.in's to put bins right in targets rather than subdirs under
targets. Removed stale uquake/Makefile.extra.
places in the various Makefile.in's. Having them all in the module root
like that may make it easy to make a change that affects everything all
at once but it's real bitch to try and maintain those files. They're all
gone now and we'll just have to maintain the seperate makefiles.
were tested successfully. There are a few hardwired LDFLAGS to clean up
(-lm, -ldl, -lvga, and -lGL) in various targets still. In order to build
with Mesa versions prior to 3.1 you'll need to change -lGL to -lMesaGL. I'll teach autoconf how to do this tonight. It's safe to come out from behind your pre_autoconf branch now guys. ;>
the -glide targets (now named -3dfx..) Thid doesn't mean that I have
actually written the makefile stuff for these targets, but rather that I
have written the foundation needed to create them.
support for the return of a glide target. Things are still a bit broken
however as I need to write glide build rules. I probably have more to do
in configure.in, I want to make libGL detection work more like Glide's
does now. The Glide detection rule was lifted almost wholesale from Mesa
because it works and I couldn't write anything that would work better.
it's not alpha. qw_server should build on most archs now, qw_client and
uquake apparently have endian issues still. (Not to mention the lack of
software renderers for anything but i386...)
- MesaGL is no more, use a newer mesa that provides libGL.so.1
- The suffix for software X11 renderer is now -x11 for consistancy with
other software that build -x11 and -svga versions
- Unified build is here! Everything goes into $(BUILD_DIR) now
- $(BUILD_DIR) is used for both debug and release, override it for now
if you need to build seperate release and debug objects. The option to
build them in different places should probably go into the autoconf
stuff so Mercury doesn't have to kill me.
Deek told me to remove S_Init() references from host.c to fix a sound bug,
I did so. The difference should be limited to whitespace since I changed
nothing else.
