On Tue, 27 Jul 2004, Robert Olsson wrote:

 > Yeah, when the ksoftirqd is taking all the cpu it will be like that, but 
 > when the kernel is behaving normally the starving diff is between 0->1sec.
 Well ksoftirqd makes your kernel load just visible which is good and 
 ksofirqd gets accounted for this when softirq's get deferred to it.
 It may look like goes from 0 to 100% but thats probably not the case.
 The problem is we can starve userland at high loads. As said we were
 trying some way to cure this I may have some old patch if you like to try.

Ok, as I said before I'm willing to test your patches. 

It would be nice that one could use the full capacity of his/her computer.
This is not a big problem for everyday use for a workstation but prevents 
2.6-series to be used in production-enviroments in the servers.
But hey.. we need to do some work and maybe we will resolve this. =)

--
Pasi Sjöholm