Date: Mon, 27 Dec 1999 00:35:25 -0500 (EST) From: Donn Miller <dmmiller@cvzoom.net> To: current@freebsd.org Subject: ESS 1868, mpg123, heavy CPU usage Message-ID: <Pine.BSF.4.20.9912270026001.287-100000@lca111.cvzoom.net>
next in thread | raw e-mail | index | archive | help
Something funny's going on with mpg123 and the pcm sound driver with the ESS 1868. Whenever a clip is playing with mpg123, the load average shoots up to around 2 (when running X). When the clip is "finished", my machine then feels sluggish as if the sound driver is somehow tying up the cpu. Then, the load average drops down to around 1 and stays there. When I start mpg123 again to play a clip, the load average shoots up to 3. Then, the cpu usage becomes unbearable. Of course, rebooting solves the problem. RealPlayer 5.0 and G2 at least don't exhibit this problem. RealPlayer has it's problems, but it doesn't tie up the cpu after RealPlayer has exited. I'm guessing the codec from mpg123 is still active, although the actual program has exited. Of course, sometimes mpg123 doesn't exit after the clip is finished, and I have to kill it manually. It would be nice if we had a special option to ps that showed the resource usage on drivers, because now ps only shows the resource usage on daemons and userland programs. I think newpcm should be a loadable kernel module, as it would make it easy to kill an "out of control" sound driver. Actually, on Linux, I had to unload and reload the sound driver for the ESS 1868 on one occasion. - Donn To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-current" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?Pine.BSF.4.20.9912270026001.287-100000>