Segmentation Fault in Linux

If you're having problems with Snes9x, or think you've found a bug, this is the place to be.
User avatar
BearOso
Official GTK/Linux Porter/Dev
Posts: 460
Joined: Tue Oct 02, 2007 12:50 am

Post by BearOso »

mangamuscle wrote:Yep, downloaded r295 and no more crashes when trying to activate this option, albeit it would be nice if the Cannot load GLSL shader functions. appeared in a windows instead of in the terminal.

Though I still needed --without-assembler to create the binary; I forgot to mention last time, but when I use autogen.sh I get this error messages:

Code: Select all

Please add the files
  codeset.m4 gettext.m4 glibc21.m4 iconv.m4 isc-posix.m4 lcmessage.m4
  progtest.m4
from the /aclocal directory to your autoconf macro directory
or directly to your aclocal.m4 file.
You will also need config.guess and config.sub, which you can get from
ftp://ftp.gnu.org/pub/gnu/config/.

/usr/share/aclocal/nspr.m4:8: warning: underquoted definition of AM_PATH_NSPR
  run info '(automake)Extending aclocal'
  or see http://sources.redhat.com/automake/automake.html#Extending-aclocal
/usr/share/aclocal/audiofile.m4:12: warning: underquoted definition of AM_PATH_AUDIOFILE
Maybe this is related to the error that prevents me to compile normally the source?
Nah, completely unrelated. It just means the nspr and audiofile libraries on your system (which this program doesn't use), have outdated autoconf syntax. :-)
If I were using an ATI card I would say you are right, but NVIDIA's drivers are accurate AFAIK. I suppose there must be a reason that explains this odd behavior ... is there an SDL or OpenGL guru in the audience?
Considering pixel buffer objects are very seldom used, I don't imagine there's many people that would know what's going wrong with the use here. I'll just have to investigate to see if there's something I'm using out of spec for your card that I'm not checking for.

Could you ensure the latest SVN changes still work for you? (without the pbo option, obviously)
mangamuscle
Snes9x White Belt
Posts: 13
Joined: Tue Feb 16, 2010 3:38 am

Post by mangamuscle »

BearOso wrote:Could you ensure the latest SVN changes still work for you? (without the pbo option, obviously)
Downloaded r296, it still requires --without-assembler, it still works but seems slower than r295. I delayed since I tested somethings, the only interesting one is that if I compile from the unix folder I do not require --without-assembler to create the binary (as I do when compiling from the gtk folder).

I see you already have a r297, lets see if this time I am fast enough to compile it before the next release ^^;
mangamuscle
Snes9x White Belt
Posts: 13
Joined: Tue Feb 16, 2010 3:38 am

r297 is toxic

Post by mangamuscle »

downloaded r297 and this are the results. When running autoconf I get:

Code: Select all

configure.ac:13: error: possibly undefined macro: AM_INIT_AUTOMAKE
      If this token and others are legitimate, please use m4_pattern_allow.
      See the Autoconf documentation.
configure.ac:17: error: possibly undefined macro: AM_PROG_CC_C_O
configure.ac:23: error: possibly undefined macro: AM_GLIB_GNU_GETTEXT
configure.ac:200: error: possibly undefined macro: AC_CHECK_PROG
configure.ac:473: error: possibly undefined macro: AM_CONDITIONAL
and when I run configure it abruptly stops after checking for ranlib:

Code: Select all

configure: error: cannot find install-sh, install.sh, or shtool in "." "./.." "./../.."
As usual I deleted everything in the folder before downloading and compiling the next release.

edit: adding missing configuration files from the 1.52 source I could compile without the need for --without-assembler, albeit I think it still somewhat slower than previous versions.
User avatar
BearOso
Official GTK/Linux Porter/Dev
Posts: 460
Joined: Tue Oct 02, 2007 12:50 am

Re: r297 is toxic

Post by BearOso »

mangamuscle wrote:downloaded r297 and this are the results. When running autoconf I get:

Code: Select all

configure.ac:13: error: possibly undefined macro: AM_INIT_AUTOMAKE
      If this token and others are legitimate, please use m4_pattern_allow.
      See the Autoconf documentation.
configure.ac:17: error: possibly undefined macro: AM_PROG_CC_C_O
configure.ac:23: error: possibly undefined macro: AM_GLIB_GNU_GETTEXT
configure.ac:200: error: possibly undefined macro: AC_CHECK_PROG
configure.ac:473: error: possibly undefined macro: AM_CONDITIONAL
and when I run configure it abruptly stops after checking for ranlib:

Code: Select all

configure: error: cannot find install-sh, install.sh, or shtool in "." "./.." "./../.."
Your autotools setup just looks outdated or broken. The fact that the configure script from the release tarball works proves it.
albeit I think it still somewhat slower than previous versions.
Is it actually slower during emulation or is it just slower in other regards? The actual change was letting GDK handle the expose, which means it's properly syncing the display when other windows move over it or the window resizes. These things are intentionally going to seem slower. If it's actually significantly faster with just a native X window, though, I can certainly go back to that.
mangamuscle
Snes9x White Belt
Posts: 13
Joined: Tue Feb 16, 2010 3:38 am

Re: r297 is toxic

Post by mangamuscle »

BearOso wrote:Your autotools setup just looks outdated or broken. The fact that the configure script from the release tarball works proves it.
Yeah, I should update the OS but since I might have to reinstall al my applications (that continue to work no problem) I will probably wait until I really have to update to a new version.
Is it actually slower during emulation or is it just slower in other regards? The actual change was letting GDK handle the expose, which means it's properly syncing the display when other windows move over it or the window resizes. These things are intentionally going to seem slower. If it's actually significantly faster with just a native X window, though, I can certainly go back to that.
When I say it is slower I refer only to the fps showed on screen (I have that turned on from day one looking to get full speed somehow). Yeah, I know this machine is damn slow, but I thought snes9x was a compromise between bsnes exact (but too slow) emulation and zsnes outdated (but fast) emulation.
User avatar
BearOso
Official GTK/Linux Porter/Dev
Posts: 460
Joined: Tue Oct 02, 2007 12:50 am

Re: r297 is toxic

Post by BearOso »

mangamuscle wrote: When I say it is slower I refer only to the fps showed on screen (I have that turned on from day one looking to get full speed somehow). Yeah, I know this machine is damn slow, but I thought snes9x was a compromise between bsnes exact (but too slow) emulation and zsnes outdated (but fast) emulation.
How much of a difference (between r295 and current) are we talking about here? Can you give some comparative measurements from some relatively still part of a particular game?
mangamuscle
Snes9x White Belt
Posts: 13
Joined: Tue Feb 16, 2010 3:38 am

Re: r297 is toxic

Post by mangamuscle »

BearOso wrote:How much of a difference (between r295 and current) are we talking about here? Can you give some comparative measurements from some relatively still part of a particular game?
Sorry for the delay ^^; It is hard to say since the fps is quite bumpy; all this time I have been ysing the same game (the chaos engine) for testing fps but even though the games only plays some music and shows an static graphic as an intro the fps counter never stays in one number changes at least every second). I would say tha deiference is between 10~15 fps, I wish there was some way to create logs of the performance :S
User avatar
BearOso
Official GTK/Linux Porter/Dev
Posts: 460
Joined: Tue Oct 02, 2007 12:50 am

Re: r297 is toxic

Post by BearOso »

mangamuscle wrote:
BearOso wrote:How much of a difference (between r295 and current) are we talking about here? Can you give some comparative measurements from some relatively still part of a particular game?
Sorry for the delay ^^; It is hard to say since the fps is quite bumpy; all this time I have been ysing the same game (the chaos engine) for testing fps but even though the games only plays some music and shows an static graphic as an intro the fps counter never stays in one number changes at least every second). I would say tha deiference is between 10~15 fps, I wish there was some way to create logs of the performance :S
Ok, it seems you're missing a vsync breakpoint, so that's fairly significant. The GdkWindow might be doing some kind of implicit conversion that's affecting performance. I'll stick to native Xlib.

Do the pixel buffer objects still crash on you, though?
Post Reply