[MPlayer-dev-eng] shared liba52 and mm_accel macros

[Rich Felker]

On Mon, Apr 02, 2007 at 12:27:30PM +0200, Reimar Döffinger wrote:
> Hello,
> On Mon, Apr 02, 2007 at 11:49:58AM +0200, Guillaume Poirier wrote:
> > on Linux, it's much more hackish (trying to execute an altivec
> > instruction and trap SIGILL if it raises an exception.
> 
> No, for a library that is not an option, and it does not check if the

Indeed, due to the extremely stupid way posix specified the
interaction of signals and threads..

> kernel actually has support for it (don't know if there is any special
> support needed).
> There is a correct way to do it, and it involves a third parameter to
> the main() function. I forgot the details though.

This is not correct, and the third parameter to main (which is
undefined and invalid C anyway!) is just the pointer to the
environment, which is also accessible as extern char **environ;

Rich