Re: [patch] sched: accurate user accounting

[Gene Heskett <gene.heskett@gmail.com>]

On Sunday 25 March 2007, Con Kolivas wrote:
>On Sunday 25 March 2007 22:32, Gene Heskett wrote:
>> On Sunday 25 March 2007, Con Kolivas wrote:
>> >On Sunday 25 March 2007 21:46, Con Kolivas wrote:
>> >> On Sunday 25 March 2007 21:34, malc wrote:
>> >> > On Sun, 25 Mar 2007, Ingo Molnar wrote:
>> >> > > * Con Kolivas <kernel@kolivas.org> wrote:
>> >> > >> For an rsdl 0.33 patched kernel. Comments? Overhead worth it?
>> >> > >
>> >> > > we want to do this - and we should do this to the vanilla
>> >> > > scheduler first and check the results. I've back-merged the
>> >> > > patch to before RSDL and have tested it - find the patch below.
>> >> > > Vale, could you try this patch against a 2.6.21-rc4-ish kernel
>> >> > > and re-test your testcase?
>> >> >
>> >> > [..snip..]
>> >> >
>> >> > Compilation failed with:
>> >> > kernel/built-in.o(.sched.text+0x564): more undefined references
>> >> > to `__udivdi3' follow
>> >> >
>> >> > $ gcc --version | head -1
>> >> > gcc (GCC) 3.4.6
>> >> >
>> >> > $ cat /proc/cpuinfo | grep cpu
>> >> > cpu             : 7447A, altivec supported
>> >> >
>> >> > Can't say i really understand why 64bit arithmetics suddenly
>> >> > became an issue here.
>> >>
>> >> Probably due to use of:
>> >>
>> >> #define NS_TO_JIFFIES(TIME)	((TIME) / (1000000000 / HZ))
>> >> #define JIFFIES_TO_NS(TIME)	((TIME) * (1000000000 / HZ))
>> >>
>> >> Excuse our 64bit world while we strive to correct our 32bit
>> >> blindness and fix this bug.
>> >
>> >Please try this (akpm please don't include till we confirm it builds
>> > on ppc, sorry). For 2.6.21-rc4
>> >
>> >---
>> >Currently we only do cpu accounting to userspace based on what is
>> >actually happening precisely on each tick. The accuracy of that
>> >accounting gets progressively worse the lower HZ is. As we already
>> > keep accounting of nanosecond resolution we can accurately track
>> > user cpu, nice cpu and idle cpu if we move the accounting to
>> > update_cpu_clock with a nanosecond cpu_usage_stat entry. This
>> > increases overhead slightly but avoids the problem of tick aliasing
>> > errors making accounting unreliable.
>>
>> I'm playing again because the final 2.6.20.4 does NOT break amanda,
>> where 2.6.20.4-rc1 did.
>
>Yes only the original version I posted on this email thread was for an
> RSDL 0.33 patched kernel. That original patch should build fine on i386
> and x86_64 (where I tried it). This version I sent out following Ingo's
> lead has 2.6.21-rc4 in mind (without rsdl).

And since I have an amdump session running in the background using the 
first 2.6.20.4-rdsl-0.33 patch I got from your web page about an hour 
ago, I am very pleased with this, I still have a machine as gzip is being 
restrained to a max of about 83% of the cpu.  And according to my 
grepping of the /tmp/amanda-dbg/client/Daily/sendsize* files, this patch 
does not break amanda.

Also, to Ingo Molnar, I've removed that sched-batch thing from my scripts 
as this feels considerably smoother in comparison.

Thank you very much, Con, and I hope you are improving.

-- 
Cheers, Gene
"There are four boxes to be used in defense of liberty:
 soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author)
Aberdeen was so small that when the family with the car went
on vacation, the gas station and drive-in theatre had to close.