Subject: Re: Font handling in the mplayer package and enable-runtime-cpudetection (was: CVS commit: pkgsrc/graphics/mplayer-share)
To: None <email@example.com>
From: Bernd Ernesti <firstname.lastname@example.org>
Date: 12/23/2002 00:35:08
On Mon, Dec 23, 2002 at 12:19:59AM +0100, Thomas Klausner wrote:
> On Sun, Dec 22, 2002 at 07:33:15PM +0100, Bernd Ernesti wrote:
> > Why not install
> > a default link which can be changed by making the link in ~/.mplayer/font?
> > The $HOME link will overwrite the one in $PREFIX.
> Good idea. Send me a patch, or commit it yourself.
I'm still catching up with my mail and then going to update my system, but don't
expect that this will happen this year or soon next year.
So I can't test anything.
> > And the turning on the --enable-runtime-cpudetection without any way to
> > disable it, better only enable it if a flag is set, is also a performance
> > thing which should not be done.
> This allows us to make binary packages available.
> An optional flag to turn it off is a good idea, though, I'll add one.
> > Oh, and i forgot one thing about that. It should be converted to use sysctl.
> > I made a patch to the current cvs version of mplayer, but didn't had the time to test
> > it.
> > If someone could do that today, then it could be integrated in the next release
> > candidate of mplayer.
> Sorry, my computer doesn't have SSE2.
Neither do mine, only SSE, and I'm not sure that testing for SSE2 is a good idea.
I'm not familiar with sse/sse2, so i don't know if mplayer has a problem if sse2 is
disabled and as I said: This patch is not tested.