Re: 2.6.4-mm2

[Nick Piggin <piggin@cyberone.com.au>]

Andrew Morton wrote:

>Nick Piggin <piggin@cyberone.com.au> wrote:
>
>> >>>
>> >>Thanks, so it's the CPU scheduler changes.  Is that machine hyperthreaded? 
>> >>And do you have CONFIG_X86_HT enabled?
>> >>
>> >
>> >Yes and CONFIG_X86_HT is enabled but I have hyperthreading disabled with
>> >'acpi=off noht' (whichever one does it.)  
>> >
>>
>>
>> The oprofile for the 01 kernel says
>> CPU: P4 / Xeon, speed 1497.76 MHz (estimated)
>> while the 02 kernel says
>> CPU: P4 / Xeon with 2 hyper-threads, speed 1497.57 MHz (estimated)
>> What's going on there?
>>
>
>Does the sched-domains patch break `acpi=off' or `noht'?
>
>

Shouldnt.

>> Other than that, nothing in the kernel profile jumps out at me:
>> schedule, __copy_from_user_ll and __copy_to_user_ll are all
>> significantly lower *after* the CPU scheduler changes, which
>> is an indicator that cache behaviour is better.
>>
>
>No, it indicates that the kernel is getting less work done.
>
>

If you are measuring the same period of time, yes. If you
are measuring the same amount of work, no. I assumed the
latter. Maybe I'm wrong.


>> Sar says average context switches/second were 9064 and  6567 before
>> and after.
>>
>> The only thing I can see is the CPU utilisation averages show the
>> scheduler patches have more of a tendancy to load up one CPU more
>> before moving to another. This actually should be good behaviour,
>> generally but I wonder if it is hurting at all. I would be really
>> surprised if it was that significant.
>>
>
>This machine is I/O-bound, the CPUs are mostly idle.  It would appear to be
>some interaction between the I/O system and the CPU scheduler.  Haven't we
>seen that with reaim also?
>
>

I can't remember how much CPU reaim uses, I thought it was
quite a lot (ie. not IO bound).