Nick Piggin writes:
And child-runs-first in wake_up_new_task. Please don't.
It does child runs first by design in staircase. You don't need any more.
Also, basic interactivity in X is bad with the interactive sysctl set to 0
Well duh... disable interactivity and interactivity is bad. What's the problem? It's not meant to be used on a desktop in that way.
(is X supposed to be at nice 0?), however fairness is bad when interactive is 1.
I'm not sure if this is an acceptable tradeoff - are you planning to fix it?
Why? A single user desktop is hardly needing extremely accurate cpu distribution... we see that already in 2.6.
It has interactivity problems with "thud". Also the mouse can freeze for .5 to 1
second when moving between windows while there is disk IO going on in the background
(this is with interactive = 1). The test-starve problem is back.
Hmm? a minor mouse freeze with a _test_ starvation program is not starvation; nor is it an interactivity problem. Yours is the first complaint about interactivity during i/o.
Increasing priority (negative nice) doesn't have much impact. -20 CPU hog only gets
about double the CPU of a 0 priority CPU hog and only about 120% the CPU time of a
nice -10 hog.
-20 is 40 rr intervals. 0 is 20 rr intervals. +19 is 1 rr interval. Seems to me the cpu distribution is working our absolutely perfectly as designed.
Why is the only critic of this the person with a competing design? Does anyone else object to these things? I certainly dont feel objective enough to criticise yours.