Mupen64 0.5 source code

[beniro]

I worked to build this on Arch Linux. The PKGBUILD is at the link below...but why I'm really posting this is that there's a patch there that may help some people resolve some issues. The patch isn't written by me.

http://bbs.archlinux.org/viewtopic.php?t=15016&start=15

[Hacktarux]

yup, mupen64 will crash if it's configured for one user and you do a "make install"...

i didn't remember that the mupen64 sound plugin was using a config file and forgot to handle the multiuser case.... i'll have to change that

[loserMcloser]

Concerning the segfault... of course you have done a "make install" as root after compilation, you have checked that you have some plugins in PREFIX/share/mupen64/plugins/ ?

Sorry to nitpick, but it should only take a few extra lines to output "No plugins found" and exit cleanly instead of segfaulting...?

Also, note that line 297 of main.c

if(strcmp(entry->d_name + strlen(entry->d_name) - 3, ".so"))

implicitly assumes that strlen(entry->d_name) is at least 3 -- you're going to be out of bounds if it's only 1 or 2.

[smcd]

Sorry to nitpick, but it should only take a few extra lines to output "No plugins found" and exit cleanly instead of segfaulting...?

Also, note that line 297 of main.c

if(strcmp(entry->d_name + strlen(entry->d_name) - 3, ".so"))

implicitly assumes that strlen(entry->d_name) is at least 3 -- you're going to be out of bounds if it's only 1 or 2.

You could use strrchr perhaps to find the first occurence of the '.' from the end of the C-string to compare against?

[Hacktarux]

don't worry sethmcdoogle, it's easy to fix, it's just that i've never looked at that and didn't see there was a potential problem there....

anyway, i'll fix it next time i'll work on mupen64

[kulminaator]

Is there any chance that mupen's source will move to CVS or SVN ?

Patching the original source from the forum definitely is fun
but quite time consuming. Also the subversion/cvs would allow
to make checkouts from historic revisions which arent always
even packaged as a source release.

btw. great job with the emulator, runs just fine under linux
and there almost no bugs with the rice video plugin.

[tpower01]

Okay...so I'm new to these forums, and linux, and emulators. How do I get help?

[GarulfoLinux]

Impossible compilation under Ubuntu Edgy .

[braddabug]

I got it to compile on Debian. I was getting an error inside SDL_ttf.c. It looks like that's an old version of the SDL_ttf files. I tried to replace them with a newer version, but the SDL site is down today. Intead I just replaced line 281 (or somewhere in there. it's just above the line that says "stream->read = RWread") with this:

stream->memory = NULL;

I got it to build and now it's running Ocarina of Time about as well as I could ask an emulator to.

BTW, I second the Subversion repo idea.

[CyZyco]

Hi,

I was studding your code and I have anuy doubts.

In n64info, rdram's range is

0x0000 0000 to 0x001F FFFF RDRAM range 0
0x0020 0000 to 0x003F FFFF RDRAM range 1

so, its has 0x400000 bytes.

In memory.c is
unsigned long rdram[0x800000/4];

has it been superdimensioned?