On Sonntag 06 September 2009, Nikos Chantziaras wrote: > On 09/06/2009 03:43 AM, Volker Armin Hemmann wrote: > > On Sonntag 06 September 2009, Nikos Chantziaras wrote: > >> On 09/06/2009 02:23 AM, Volker Armin Hemmann wrote: > >>> On Sonntag 06 September 2009, Alan McKinnon wrote: > >>>> On Sunday 06 September 2009 00:48:40 Volker Armin Hemmann wrote: > >>>>>> Also, have you considered that you got it all backwards? The kernel > >>>>>> configuration tells you that for lower latencies, you should use > >>>>>> 1000Hz and PREEMPT. It even says "Desktop" right there. Why should > >>>>>> I take your word over that of the kernel devs who actually wrote > >>>>>> that code? > >>>>> > >>>>> low latency means bad throughput and that hurts IO. > >>>> > >>>> The average desktop user on Linux needs low latency. To get that, one > >>>> must sacrifice some throughput efficiency. > >>> > >>> and then complain, that IO hurts? > >> > >> No one complains that IO hurts. Perhaps you aren't even reading. The > >> complaints are about GUI stalls. Not slow IO. > > > > yeah, and most GUI stalls happen with big io. ooops. > > > >>> btw, I think, I am an average user, doing the average stuff, and I am > >>> pretty happy with 'voluntary preemption'. > >> > >> We are not. > > > > 'we'. > > > >>>> have you considered that there is a large population of users whose > >>>> needs and workload are totally different from yours and therefore > >>>> require something completely different to you? > >>> > >>> yes, but I consider my workload average. Surfing the web, watching tv, > >>> watching movies with xine/vlc/mplayer. Listening to music with amarok > >>> and alsaplayer. Typing stuff. Sometimes skype. Burning a dvd once in a > >>> while. vegastrike, ut2004, if I want to play a game. > >>> > >>> See? Average. > >>> > >>> Now 32 sound streams in ardour (or whatever tool you use for that), > >>> that is hardly 'average'. > >> > >> Doesn't matter if it's not average. What does matter is that Linux is > >> not up to the task while Windows and OS X are. > > > > and windows has a completly different, gui centric architecture. Oh, and > > for serious audio stuff you need special low latency drivers. Humpf. > > > > So what was your point again? Windows needs special drivers for serious > > audio stuff - and people are complaining that they have to turn on rt in > > linux for the same tasks? > > I don't even know why I keep talking to you. You don't seem to *want* > to understand. It's not about special drivers. Or specialized audio > processing. It's about my 3D desktop cube getting skippy sometimes > while video is playing. It's about the transparency effect applied > while moving mplayer resulting in some frame skipping. It's about a > slight pause while scrolling a web page if I have an emerge running. > About the mouse freezing for about 0.1 seconds once in a while while > doing the same. > > Can't you understand, or don't you want to understand? >
http://marc.info/?l=linux-kernel&m=125227082723350&w=2 seems like BFS is even crappier than CFS. --------------------------- I saw really bad interactivity in the BFS test here - the system was starved for as long as the test ran. I stopped the tests at 8 loops - the system was unusable and i was getting IO timeouts due to the scheduling lag: sd 0:0:0:0: [sda] Unhandled error code sd 0:0:0:0: [sda] Result: hostbyte=DID_OK driverbyte=DRIVER_TIMEOUT end_request: I/O error, dev sda, sector 81949243 Aborting journal on device sda2. ext3_abort called. EXT3-fs error (device sda2): ext3_journal_start_sb: Detected aborted journal Remounting filesystem read-only oh great. BFS really rocks! --------------- And unfortunately this is where it ends for now, since BFS doesn't boot on the two boxes I tried. It hard hangs right after disk detection. But the latency numbers look pretty appalling for CFQ, so it's a bit of a shame that I did not get to compare. I'll try again later with a newer revision, when available. sounds great. A system not booting is obviously a system not being laggy. And pissing of Jens Axboe? Really smart move. At least you posted your little success story. I hope, you will also be willing to test patches, if Ingo Molnar would ask for it.
