On Sun, 1 Mar 2015 16:26:20 +0100 Michael Niedermayer <michae...@gmx.at> wrote:
> On Sun, Mar 01, 2015 at 01:11:08PM +0100, wm4 wrote: > > On Sun, 1 Mar 2015 12:09:28 +0100 > > Carl Eugen Hoyos <ceho...@ag.or.at> wrote: > > > > > On Sunday 01 March 2015 11:22:10 am Michael Niedermayer wrote: > > > > On Sun, Mar 01, 2015 at 09:44:30AM +0100, Carl Eugen Hoyos wrote: > > > > > Attached untested patch should fix compilation on broken platforms. > > > > > We don't use round() currently because it appears less portable than > > > > > lrintf(). > > > > > > > yes reducing float code usage should be fine, less potential to > > > > cause problems with regression tests > > > > > > New patch attached. > > > > > > Please review, Carl Eugen > > > > Seriously? You replace kind of magic, but sourced constants with > > completely magic ones? > > one could replace the round by a cast to int and a +0.5 > or figure out why it doesnt work on sunos > > iam fine with whatever color people prefer as long as it works > > [...] I prefer the color of fuck broken platforms nobody uses. _______________________________________________ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
