Ryan C. Gordon <icculus@icculus.org> [Fri, 15 Jun 2007 06:29:52 +0000] rev 3962
Support for Gamma Ramps on Mac OS Classic in both the macrom and macdsp video
targets, thanks to Gregory Smith!
Fixes Bugzilla #431.
Sam Lantinga <slouken@libsdl.org> [Fri, 15 Jun 2007 05:41:22 +0000] rev 3961
Don't install the extra SDL config headers or SDL_copying.h
Fixes Bugzilla #361.
Sam Lantinga <slouken@libsdl.org> [Fri, 15 Jun 2007 05:21:33 +0000] rev 3960
Updated docs with bug fixes in subversion.
Ryan C. Gordon <icculus@icculus.org> [Thu, 14 Jun 2007 14:09:25 +0000] rev 3959
Prevent arts audio target from crashing/hanging SDL if the audio hardware
isn't available.
Fixes Bugzilla #372.
Ryan C. Gordon <icculus@icculus.org> [Wed, 13 Jun 2007 08:00:10 +0000] rev 3958
Actually, this is a better fix...clear the error state once if everything we
need loaded; it's more efficient, and works even if the last unnecessary
xrandr symbol failed to load. Otherwise, leave the original loadso error, so
the end user can find out what symbol failed.
Ryan C. Gordon <icculus@icculus.org> [Wed, 13 Jun 2007 07:54:10 +0000] rev 3957
Date: Mon, 4 Jun 2007 06:17:59 +0000 (UTC)
From: George Gensure <werkt0@gmail.com>
Subject: [SDL] error removal patch for SDL_x11dyn.c
I found that calling SDL_GetError when an error hasn't necessarily occurred is
still reporting problems from loadso regarding dynamic functions in X11. I've
added the following to my local copy to avoid revealing the 'many library
lookup' approach adopted by x11dyn.c:
Ryan C. Gordon <icculus@icculus.org> [Wed, 13 Jun 2007 04:13:21 +0000] rev 3956
Print keysym with key events.
Patrice Mandin <patmandin@gmail.com> [Sat, 09 Jun 2007 19:52:05 +0000] rev 3955
Avoid switch to supervisor mode in SDL_GetTicks, by updating system counter from vbl interrupt
Patrice Mandin <patmandin@gmail.com> [Fri, 08 Jun 2007 21:44:40 +0000] rev 3954
Set supervisor as volatile as it is modified from an interrupt
Ryan C. Gordon <icculus@icculus.org> [Mon, 04 Jun 2007 11:43:47 +0000] rev 3953
C fallback for SDL_revcpy() was off by one. Thanks to Emmanuel Jeandel for
the catch.
Fixes Bugzilla #424.
Ryan C. Gordon <icculus@icculus.org> [Mon, 04 Jun 2007 11:22:23 +0000] rev 3952
Better handling of multiple queued Cocoa mouse events. Thanks to Christian
Walther for the patch.
Fixes Bugzilla #353.
Ryan C. Gordon <icculus@icculus.org> [Mon, 04 Jun 2007 11:15:58 +0000] rev 3951
Patch to allow DirectFB target to report a correct screen size.
Fixes Bugzilla #399.
Ryan C. Gordon <icculus@icculus.org> [Mon, 04 Jun 2007 11:06:48 +0000] rev 3950
Check if yasm supports a nasm-specific syntax we use (it doesn't in 0.5.0,
which is still common in Linux distros, and does in 0.6.0). Disable it if we
don't, attempting to fallback to using nasm. Thanks to Mike Frysinger for the
patch.
Fixes Bugzilla #393.
Ryan C. Gordon <icculus@icculus.org> [Sun, 03 Jun 2007 19:21:07 +0000] rev 3949
Minor macrom comment cleanup.
Ryan C. Gordon <icculus@icculus.org> [Sun, 03 Jun 2007 19:19:16 +0000] rev 3948
Minor tab fix in mac drawsprocket code.
Ryan C. Gordon <icculus@icculus.org> [Sun, 03 Jun 2007 06:52:58 +0000] rev 3947
Patched to compile on Mac OS X (not that you should necessarily be using this
on Mac OS X, but still) ...
Ryan C. Gordon <icculus@icculus.org> [Sun, 03 Jun 2007 06:52:06 +0000] rev 3946
Use system memset/memcpy on Mac OS X, since Apple hand-tunes these for each
processor they ship (and thus, it's likely to beat our code on PowerPC and
Intel and whatever variants or new archs show up later).
Ryan C. Gordon <icculus@icculus.org> [Tue, 29 May 2007 12:01:07 +0000] rev 3945
Mac OS X joystick calibration code should start with device-specified bounds.
Fixes Bugzilla #344.
Patrice Mandin <patmandin@gmail.com> [Sun, 27 May 2007 08:53:48 +0000] rev 3944
Allow creation of window bigger than visible size
Patrice Mandin <patmandin@gmail.com> [Sat, 26 May 2007 20:12:58 +0000] rev 3943
Oops
Patrice Mandin <patmandin@gmail.com> [Sat, 26 May 2007 20:09:56 +0000] rev 3942
Simply resize existing window
Patrice Mandin <patmandin@gmail.com> [Sat, 26 May 2007 19:44:37 +0000] rev 3941
GEM has a window manager, of course
Ryan C. Gordon <icculus@icculus.org> [Tue, 22 May 2007 22:15:54 +0000] rev 3940
Check correct glX extension name for swap_control, but it doesn't look like
we ever actually look up the symbol...that's probably a bug. :/
Ryan C. Gordon <icculus@icculus.org> [Sun, 13 May 2007 23:12:46 +0000] rev 3939
Committed PulseAudio driver. Thanks, Stephan!
Ryan C. Gordon <icculus@icculus.org> [Tue, 17 Apr 2007 09:09:48 +0000] rev 3938
Removing .gdbinit from test directory.
Ryan C. Gordon <icculus@icculus.org> [Tue, 17 Apr 2007 08:28:19 +0000] rev 3937
Let Windows OpenGL users use SDL_GL_SWAP_CONTROL even if WGL_ARB_pixel_format
isn't available...this looks like it got tucked into here with the other
attributes, which are wrapped in a pixel_format availability check.
Ryan C. Gordon <icculus@icculus.org> [Fri, 06 Apr 2007 20:30:41 +0000] rev 3936
Let app set SDL_VIDEO_ALLOW_SCREENSAVER environment variable to override SDL's
attempt to disable screen savers. Works for Quartz (Mac OS X) and X11.
Need a formal API for this in 1.3, still.
Fixes Bugzilla #415.
Ryan C. Gordon <icculus@icculus.org> [Wed, 04 Apr 2007 10:34:42 +0000] rev 3935
There are several subsystems that need IOKit linked in on Mac OS X, so handle
cases where one or more of them are disabled.
Ryan C. Gordon <icculus@icculus.org> [Wed, 04 Apr 2007 10:30:07 +0000] rev 3934
Crap, put this patch in the wrong place.
THIS fixes Bugzilla #403. :/
Ryan C. Gordon <icculus@icculus.org> [Wed, 04 Apr 2007 10:28:14 +0000] rev 3933
Don't compile the joystick code if --disable-joystick is specified at
./configure time, and define SDL_JOYSTICK_DUMMY if the subsystem is wanted
but there aren't any drivers available for the platform.
Fixes Bugzilla #403.