On Thu, May 20, 2010 at 12:35:56PM +0200, Sebastian Reitenbach wrote: > Hi, > > since trying to build dependencies for my planned coccinella update, I found > that jack is not building on sparc with the system gcc.
jack is most likely useless on the sparc. it doesn't support ulaw and that's all audioamd(4) does. afaik, audiocs(4) is busted. it does work on my SGI O2, but it's sorta pointless. anyway, wtf is pulling in jack? jack is not designed to be used as a "system" sound server. it is designed for audio production work. we have a real "system" sound server in base, so I think it's high time people quit trying to use jack for anything other than music production. it's just going to cause headaches otherwise. > appended patch just shuffles the initialization of variables a bit to allow > to > build with old gcc on sparc. Afterwards tested to build on mips64 too. > Note, I don't have a soundcard in both boxes, and no 4.7 or -current box > running with a soundcard to actually test whether I broke sth. > > see attached diff, please test and OK? I've got a WIP update to a newer version of jack. I'm kinda stuck because I'm seeing some odd/rare timing issues (new versions of jack use realtime scheduling my default ... on sysems that have that). I'll see about integrating these changes. netjack has been mostly rewritten ... *sigh*. -- jake...@sdf.lonestar.org SDF Public Access UNIX System - http://sdf.lonestar.org
