Re: the new scheduler

[Matthew Dillon <dillon@xxxxxxxxxxxxxxxxxxxx>]

:Tried burning a DVD with it earlier, everything was fine and it went
:smooth. But now I've tried to play MP3s with mpg123 and XMMS and it's
:_not_ a smooth experience, mpg123 skips every know and then, changing
:mpg123 niceness screws everything up.
:XMMS is slow, takes some while to change song, impossible to use the
:buttons or move the window after a small while, changing nice value
:makes it stop (maybe it stops without that aswell).
:Anyway, it doesn't work good at all.
:...
:
: (and in a followup message)
:niceness -10 seem to be ok for mpg123
:
:using opensound for dragonfly if that helps/matters
:
:CPU: AMD Athlon(tm) XP 1500+ (1312.94-MHz 686-class CPU)
:  Origin = "AuthenticAMD"  Id = 0x662  Stepping = 2
:  Features=0x383fbff<FPU,VME,DE,PSE,TSC,MSR,PAE,MCE,CX8,APIC,SEP,MTRR,PGE,MCA,CMOV,PAT,PSE36,MMX,FXSR,SSE>
:  AMD Features=0xc0400000<AMIE,DSP,3DNow!>

    Hmm.  I have a very similar workstation:

    CPU: AMD Athlon(tm) XP (1243.35-MHz 686-class CPU)
    Origin = "AuthenticAMD"  Id = 0x681  Stepping = 1
    Features=0x383fbff<FPU,VME,DE,PSE,TSC,MSR,PAE,MCE,CX8,APIC,SEP,MTRR,
		PGE,MCA,CMOV,PAT,PSE36,MMX,FXSR,SSE>
    AMD Features=0xc0400000<AMIE,DSP,3DNow!>

    mpg123 seems to work fairly well, but I'm just running it on an mp3 
    file manually.  When I first read your report I thought it might be 
    cpu related but mpg123 seems to actually take very little (only 2% of
    the cpu).  The scheduler is correctly identifying it as an interactive
    program so I suspect that the issue may not be the scheduler at all.

    You could experiment a bit with mpg123's '-b' option.  If a larger buffer
    doesn't solve the problem then it is more then likely that (known) 
    interrupt latency issues are the cause of the skipping you are hearing.

						-Matt