>I wonder if Robert Love's kernel pre-emption patches would make >this better. Has anybody tried them extensively? From what I've >heard on the kernel mailing list, people are _very_ happy with >them. (Happy == xmms skips once in a blue moon with uptime loads >of >10, including hitting the VM and disk hard.) I'm using it >here, but I'm not doing anything extensive. (Yet :).
>From what I've read, they don't do quite as good a job as Andrew Morton's low latency patches in eliminating all kernel stalls. They work better in dealing with some common stalls, however. >Addendum to Paul's reply: I think ASIO on Windows is able to get >such low latencies because it requires specific ASIO drivers for >sound cards, and these drivers bypass the normal kernel mixers >etc. Thats not it, though this is certainly true. The scheduling latency in Windows is simply better than in mainstream Linux. Linux, like most Unix-related OS's, was designed around high throughput, not low scheduling latency. Thankfully, given the source, its been possible to change that :) --p
