To: sdl@libsdl.org
Date: Sun, 11 Dec 2005 22:57:37 -0500
Subject: [SDL] exec stack in libsdl update
i posted back in September a patch to remove executable stacks:
http://www.devolution.com/pipermail/sdl/2005-September/070626.html
later in November, a similar patch was merged it seems:
http://www.libsdl.org/cgi/cvsweb.cgi/SDL12/src/hermes/mmx_main.asm
however, this lacks the additional output format checks that i posted in my
patch ... this isnt a problem if the hermes asm code is only ever used to
produce ELF objects, but if this is not true, then the additional checks in
my original patch will need to be merged
-mike
--HG--
extra : convert_revision : svn%3Ac70aab31-4412-0410-b14c-859654838e24/trunk%401202
From: Jonathan Atkins <jcatki@jcatki.no-ip.org>
Subject: SDL patch: DGA key events
I kept seeing DGA die on me whenever I press a key.
I tracked it down and it seems that the new indirect X via pointers
system messed it up. It needed to have the pX* pointers well defined to
call, in particular, pXNextEvent. Also, the X11_TranslateKey function
prototype was changed, but not updated in the dga driver. perhaps other
files are also affected.
Not sure what releases may suffer from this too.
thanks!
-Jon Atkins
--HG--
extra : convert_revision : svn%3Ac70aab31-4412-0410-b14c-859654838e24/trunk%401197
From: Christian Walther <cwalther@gmx.ch>
Date: Fri, 18 Nov 2005 23:39:02 +0100
Subject: [SDL] Mouse position bugs on Mac OS X
The attached patch fixes a few bugs in SDL related to the mouse position
in windowed mode on Mac OS X, reproduced using the attached minimal test
program - at least here on 10.3.9, with SDL CVS from today. Could anyone
test whether the bugs exist and are fixed by the patch on 10.2 and 10.4?
1. When using OpenGL, the vertical mouse positions obtained through
events or SDL_GetMouseState() are off by one.
2. When using OpenGL, SDL_WarpMouse() inverts the y coordinate.
3. Clicks on the topmost pixel row of the window are not recognized.
1 and 2 do not occur in non-OpenGL mode, while 3 does. All three only
occur in windowed mode, not in fullscreen.
The cause for 1 and 3 is that in Cocoa, "the location of the mouse"
seems to be defined as "the location of the top left corner of the mouse
pointer's hot pixel" (this is not documented, it's just what I found out
here), which together with the fact that Cocoa's usual y coordinates
start at the bottom and increase upwards means that the y coordinate of
the mouse runs from 1 to h, not from 0 to h-1, in a window of height h.
If it does work on 10.2 and 10.4 (I'll try to test it as soon as I can,
but at the moment all I have at hand is 10.3.9), can this be applied to
the CVS?
-Christian
To: sdl@libsdl.org
From: Christian Walther <cwalther@gmx.ch>
Date: Mon, 28 Nov 2005 10:41:51 +0100
Subject: [SDL] Re: Mouse position bugs on Mac OS X
I wrote:
> I'll try to test it as soon as I can, but at the moment all I have at hand is 10.3.9
So, here are the results of my tests (with patched and unpatched
frameworks compiled with Xcode 1.5 (gcc 3.3) on 10.3.9):
On 10.1.5, my test program doesn't run because of "Undefined symbols:
SDL undefined reference to _CGMainDisplayID expected to be defined in
Carbon". I guess not supporting 10.1 was a deliberate decision then and
that's OK with me.
On 10.2.8, 10.3.9, and 10.4.0, the bugs exist as described in my
original post and are fixed by my patch. That is, there is no difference
between pre/post 10.3 and the patched version works correctly in all
combinations of GL/non-GL and windowed/fullscreen.
I therefore recommend the patch for inclusion.
-Christian
--HG--
extra : convert_revision : svn%3Ac70aab31-4412-0410-b14c-859654838e24/trunk%401195
a more official way to do OpenGL, explicitly check for glX on Mac OS X, and
use SDL_LoadObject for platforms that have glX but don't have dlopen().
--HG--
extra : convert_revision : svn%3Ac70aab31-4412-0410-b14c-859654838e24/trunk%401194
This was mostly, if not entirely, written by "Doodle" and "Caetano":
doodle@scenergy.dfmk.hudaniel@caetano.eng.br
--ryan.
--HG--
extra : convert_revision : svn%3Ac70aab31-4412-0410-b14c-859654838e24/trunk%401193
From: Staffan Ulfberg <staffan@ulfberg.se>
Date: 19 Nov 2005 01:00:48 +0100
Subject: [SDL] New driver for OpenBSD/wscons
Hello,
I've written an SDL driver for OpenBSD/wscons (console mode, somewhat
resembling the functionality of the svga driver for Linux). I use it
for playing MAME on my Sharp Zaurus. The alternative is to play under
X, which is slower.
I asked how to submit the driver a few days ago, and posted a link to
the patch in a follow-up, so maybe it was missed?
Anyway, the patch is on the web at:
http://multivac.fatburen.org/SDL-wscons.patch
Comments?
Staffan
--HG--
extra : convert_revision : svn%3Ac70aab31-4412-0410-b14c-859654838e24/trunk%401190
only exists at /usr/sys/arch/i386/include/joystick.h, for example.
--HG--
extra : convert_revision : svn%3Ac70aab31-4412-0410-b14c-859654838e24/trunk%401188
From: Mattias Karlsson <betasoft@acc.umu.se>
To: sdl@libsdl.org
Subject: [SDL] [PATCH] Fix (tiny) build problems on solaris (fwd)
This small patch fixes building audio on solaris 2.9 with gcc 4.0
--HG--
extra : convert_revision : svn%3Ac70aab31-4412-0410-b14c-859654838e24/trunk%401187
Driver can now open whatever library is specified in SDL_GL_LoadLibrary()
call (previously, it ignored this parameter), and uses the default system
library when NULL is specified.
Also, library is loaded once in SDL_GL_LoadLibrary() and not every call to
SDL_GL_GetProcAddress().
--HG--
extra : convert_revision : svn%3Ac70aab31-4412-0410-b14c-859654838e24/trunk%401184
To: sdl@libsdl.org
Date: Fri, 30 Sep 2005 02:24:50 -0600
Subject: [SDL] WinCE timers, continued
Here's a strange timer for Windows CE that doesn't ignore time across
suspends. It uses GetSystemTime to keep the time continuous, and GetTicks to
get finer-grained readings than 1 second. It detects the difference between
the GetTicks time and GetSystemTime time on power-on to keep the error within
one second max.
It's not a patch on the current win32 timer code -- took one look at that and
figured it had more than enough #ifdefs already. It's windows-ce specific.
Another thing I've noticed is that the Windows CE 4.0 and newer API has
functions warn processes about suspends. This is something SDL REALLY needs
for audio in particular, because turning it off while it's playing causes
anything that uses audio to hardlock the system on power-on. Unfortunately I
don't have 4.0 to play with. :(
--HG--
extra : convert_revision : svn%3Ac70aab31-4412-0410-b14c-859654838e24/trunk%401183
Quartz target...fixed this to match the documented behaviour.
--HG--
extra : convert_revision : svn%3Ac70aab31-4412-0410-b14c-859654838e24/trunk%401180
From: Mark Schreiber <mark7@alumni.cmu.edu>
To: ryan@clutteredmind.org
Subject: [PATCH]SDL mprotect() crash fix
(I'm going to throw this patch your way at the suggestion of #SDL --
for some reason, I had some difficulty sending it to the main list
last time, and I go bonkers subscribing to send each email or
patch...)
Currently, when I run SDL applications as non-root using
SDL_VIDEODRIVER=dga, the fbdev fallback mprotect()s read/write the
proper size of mmapped /dev/fb0 (7.5MB), but on framebuffer release
mprotect()s read-only the range by the entire size of my video memory
(128MB), which causes a segfault:
#0 0x002a9a27 in ?? () from /lib/libc.so.6
#1 0x04a63eb6 in SDL_XDGAUnmapFramebuffer (screen=3D0) at XF86DGA2.c:978
#2 0x04a63efc in SDL_XDGACloseFramebuffer (dpy=3D0x9d3f008, screen=3D0)
at XF86DGA2.c:268
#3 0x04a68b57 in DGA_Available () at SDL_dgavideo.c:98
#4 0x04a53677 in SDL_VideoInit (driver_name=3D0xbfb0bfc7 "dga", flags=3D0)
at SDL_video.c:180
#5 0x04a2613f in SDL_InitSubSystem (flags=3D32) at SDL.c:74
#6 0x04a2617c in SDL_Init (flags=3D32) at SDL.c:166
#7 0x08049722 in main (argc=3D1, argv=3D0x0) at testwin.c:32
This is SDL 1.2.8 on Fedora Core 4, radeon driver for a Radeon 9250,
xorg-x11-6.8.2-37.
I've attached a one-line patch against SDL CVS that updates the size
of the framebuffer at framebuffer map time so that the mprotect() on
unmap will be the same size. I'm not sure if this is the best
approach (i.e. one might want to retain the original value), but it
does make my SDL applications work without segfaulting.
-- Best of luck, Mark Schreiber
--HG--
extra : convert_revision : svn%3Ac70aab31-4412-0410-b14c-859654838e24/trunk%401179
instead replace it with the string "(null)", like glibc's printf() would do.
--HG--
extra : convert_revision : svn%3Ac70aab31-4412-0410-b14c-859654838e24/trunk%401175
From: Hayashi Naoyuki <titan@culzean.org>
To: "A list for developers using the SDL library. \(includes SDL-announce\)" <sdl@libsdl.org>
Subject: Re: [SDL] Dynamic X11...
1. Compilation produce the following error on Tru64 UNIX.
cc: Severe: SDL_x11dyn.h, line 31: Cannot find file
<X11/extensions/extutil.h> specified in #include directive. (noinclfilef)
#include <X11/extensions/extutil.h>
Because Tru64 UNIX doesn't have extutil.h, this error is caused.
2. Compilation with --enable-x11-shared=no produce the following error.
cc: Error: SDL_x11sym.h, line 115: In this statement,
"Xutf8TextListToTextProperty" is not declared. (undeclared)
SDL_X11_SYM(int,Xutf8TextListToTextProperty,(Display*,char**,int,XICCEncodingStyle,XTextProperty*))
Though it doesn't have Xutf8TextListToTextProperty,
"pXutf8TextListToTextProperty = Xutf8TextListToTextProperty;"
in SDL_x11dyn.c,117-119
#define SDL_X11_SYM(r,fn,arg) p##fn = fn;
#include "SDL_x11sym.h"
#undef SDL_X11_SYM
-- Hayashi Naoyuki Key fingerprint = 60A0 D5D3 F58B 3633 2E52 0147 D17F 5578 3FDF F5B6
--HG--
extra : convert_revision : svn%3Ac70aab31-4412-0410-b14c-859654838e24/trunk%401172
This allows you to run an SDL program on a system without Xlib, since it'll
just report the x11 target unavailable at runtime.
--HG--
extra : convert_revision : svn%3Ac70aab31-4412-0410-b14c-859654838e24/trunk%401171
From: Hayashi Naoyuki
Subject: SDL_numjoysticks: multiply defined
The following linker error is produced when I try to build SDL CVS on Tru64.
ld:
.libs/libSDL.lax/libjoystick.a/SDL_sysjoystick.o: SDL_numjoysticks:
multiply defined
SDL_numjoysticks is defined in both src/joystick/SDL_joystick.c:48: and
SDL12/src/joystick/dummy/SDL_sysjoystick.c:37:.
As a result, compilation produce "multiply defined" errors, when link.
--HG--
extra : convert_revision : svn%3Ac70aab31-4412-0410-b14c-859654838e24/trunk%401170
From: Dirk Mueller
Subject: [PATCH] build SDL with nonexecutable stack
libSDL is by default marked with an executable stack, which it doesn't
actually need. the reason for this is that there are assembler files in the
source tree not properly annotated with the "noexec stack" section. As such
the linker does a safe-fallback and marks the whole lib as "requires
executable stack".
the patch below removes this by adding annotations. As far as I can see it
shouldn't break anything.
--HG--
extra : convert_revision : svn%3Ac70aab31-4412-0410-b14c-859654838e24/trunk%401169
From: Mike Frysinger <vapier@gentoo.org>
To: sdl@libsdl.org
Subject: Re: [SDL] libsdl needs some tweaks for DirectFB 0.9.23
On Fri, Oct 28, 2005 at 01:23:57AM +0000, Mike Frysinger wrote:
> the new release of DirectFB breaks the libsdl DirectRB video module
>
> specifically, this change:
> http://www.directfb.org/index.php/viewcvs.cgi/DirectFB/include/directfb.h.diff?r1=1.266&r2=1.267
>
> but (unless i missed something), it should be trivial to fix (just annoying)
> ... ive done so in Gentoo (also attached):
> http://viewcvs.gentoo.org/media-libs/libsdl/files/libsdl-1.2.9-DirectFB-updates.patch
hmm, i did miss something ... need to include directfb_version.h before trying
to test version defines :)
updated patch attached as well as previous URL
-mike
--HG--
extra : convert_revision : svn%3Ac70aab31-4412-0410-b14c-859654838e24/trunk%401168
From: Thomas Omilian <paccy@arcor.de>
To: sdl@libsdl.org
Subject: [SDL] SDL_SoftStretch() fixed!
Ok, after debugging the night I found the error:
in video/SDL_stretch.c in Line 81
/* See if we need to regenerate the copy buffer */
if ( (src_w == last.src_w) &&
(dst_w == last.src_w) && (bpp == last.bpp) ) {
return(0);
}
the second comparison should be (dst_w == last dst_w).
Perhaps someone could apply it to the source...
Regards,
Thomas Omilian
--HG--
extra : convert_revision : svn%3Ac70aab31-4412-0410-b14c-859654838e24/trunk%401167
Casts to (char *) will disable strict aliasing when the compiler sees it.
--HG--
extra : convert_revision : svn%3Ac70aab31-4412-0410-b14c-859654838e24/trunk%401164
just be flat out missing from glibc, so force use of previous gettimeofday()
behaviour for now.
--HG--
extra : convert_revision : svn%3Ac70aab31-4412-0410-b14c-859654838e24/trunk%401161
From: Tommi Kyntola <tommi.kyntola@ray.fi>
To: sdl@libsdl.org
Subject: [SDL] [RFC] get_ticks broken on linux
It uses gettimeofday to calculate the timedifference.
Gettimeofday returns current time which is seldom monotonous.
This breaks SDL timer subsystem. (time callbacks and all that
get borked when the time difference ms is suddenly ~ 2^32)
I posted a message about this earlier but got no response.
Some thoughts on this matter would be appreciated.
(Or even an explanation for the lack of interest.)
A patch below would use the posix timers that have been around
since posix 93 and do provide a good source of monotonous time
on linux boxes (and on few others too).
The following patch is also availabe at:
http://www.hut.fi/u/tkyntola/SDL-1.2.9-clockfix.patch
It's against 1.2.9, but I can easily rediffit against
the cvs branch is needed.
cheers,
Tommi Kyntola tommi.kyntola@ray.fi
--HG--
extra : convert_revision : svn%3Ac70aab31-4412-0410-b14c-859654838e24/trunk%401160
From: =?ISO-8859-1?Q?Martin_Storsj=F6?= <martin@martin.st>
To: sdl@libsdl.org
Subject: [SDL] [PATCH] Use nanosleep on OS X
Hi,
The current version of SDL_Delay on OS X seems to always sleep at least 10
msec. OS X has nanosleep(), which performs shorter sleeps well. The
attached patch makes it use that one instead of select() as currently.
// Martin
--HG--
extra : convert_revision : svn%3Ac70aab31-4412-0410-b14c-859654838e24/trunk%401159
if SDL is built with a non-cdecl calling convention, and it's just generally
bad practice anyhow.
Now programs explicitly call SDL_Quit() where appropriate, wrap SDL_Quit() in
a cdecl function where it can't be avoided, and rely on the parachute where
a crash might have hit the atexit() before (these ARE test programs, after
all!).
--HG--
extra : convert_revision : svn%3Ac70aab31-4412-0410-b14c-859654838e24/trunk%401154
