Wasn't sure if this should be in *nix or Programming but thought about it and decided that here would be better since you can do this on pretty much any OS you want.
I'm running Ubuntu off of VMWare (VMWare part doesn't really matter though).
I'm attempting to cross compile Xvid from their CVS. Bear in mind that I really have no experience with coding, and I'm just trying to compile something.
What I'm trying to do is cross compile Xvid in Linux (--build=i686-pc-linux-gnu) to Windows (--host=i686-mingw32).
Now I've already successfully compiled it with --host being linux. But when I switch it to mingw32, I get a "unrecognized command line option -mno-cygwin".
And so I'm not able to cross compile it. I'm not really sure what's causing the error. I've tried installing a number of different development things, and I've pretty much been at it for hours, so I finally decided to phone a friend (in this case Programming forum at AT 😛).
Also I have a few other compiler questions.
I've been taking a look at this Safe CFLAGS site, and it says that -mfpmath=sse is sometimes slower then not using it at all. Also read on one of their forums (don't have the link) that Athlon processors are faster with -mfpmath=387 instead of using sse. Anybody know anything about that?
Thanks 🙂
Edit:
BTW, I've been using these as my compile options below. Anything wrong with it (yes, I know it's a quote and not "code" but it works right)? 😛
Edit2:
I'm basically trying to get as much speed as possible out of it. Any suggestions for CFLAGS, LDFLAGS, or CXXFLAGS, to use? Don't care about debugging or potentially breaking the app, all I care about is speed for this.
I'm running Ubuntu off of VMWare (VMWare part doesn't really matter though).
I'm attempting to cross compile Xvid from their CVS. Bear in mind that I really have no experience with coding, and I'm just trying to compile something.
What I'm trying to do is cross compile Xvid in Linux (--build=i686-pc-linux-gnu) to Windows (--host=i686-mingw32).
Now I've already successfully compiled it with --host being linux. But when I switch it to mingw32, I get a "unrecognized command line option -mno-cygwin".
And so I'm not able to cross compile it. I'm not really sure what's causing the error. I've tried installing a number of different development things, and I've pretty much been at it for hours, so I finally decided to phone a friend (in this case Programming forum at AT 😛).
Also I have a few other compiler questions.
I've been taking a look at this Safe CFLAGS site, and it says that -mfpmath=sse is sometimes slower then not using it at all. Also read on one of their forums (don't have the link) that Athlon processors are faster with -mfpmath=387 instead of using sse. Anybody know anything about that?
Thanks 🙂
Edit:
BTW, I've been using these as my compile options below. Anything wrong with it (yes, I know it's a quote and not "code" but it works right)? 😛
Edit2:
I'm basically trying to get as much speed as possible out of it. Any suggestions for CFLAGS, LDFLAGS, or CXXFLAGS, to use? Don't care about debugging or potentially breaking the app, all I care about is speed for this.
