ULE vs 4BSD in RELENG_7

[Gelsema, P (Patrick) - FreeBSD]

On Mon, November 5, 2007 00:50, Gelsema, P (Patrick) - FreeBSD wrote:
> On Sun, November 4, 2007 22:27, Jeff Roberson wrote:
>> On Sun, 4 Nov 2007, Gelsema, P (Patrick) - FreeBSD wrote:
>>
>>> Hi Jeff,
>>>
>>> I tried your patch. Ran a buildkernel, timed. Recompiled kernel
>>> including
>>> your patch, rebooted and reran. Please find results below.
>>>
>>> w/o patch
>>> hulk# time make -j8 buildkernel
>>> 837.808u 138.167s 10:28.96 155.1%       6349+1349k 2873+7780io 303pf+0w
>>>
>>> w patch
>>> hulk# time make -j8 buildkernel
>>> 838.554u 168.316s 10:52.10 154.4%       6263+1332k 6489+7791io 11pf+0w
>>>
>>> I only understand the 10:52 seconds thing, but it looks like it's
>>> stressing less and still getting things done.
>>>
>>
>> Well this doesn't look very good at all.  System time increased by 30
>> seconds!  Must be too many extra context switches.  The minimum slice
>> value is probably too small and you've got an average of 4 threads
>> per-core.
>>
>> Can you try re-running with kern.sched.slice_min = 4 and
>> kern.sched.slice
>> = 12 ?
>>
>> Can you tell me how 4BSD does on this machine?
>
> ULE w/o patch
> hulk# time make -j8 buildkernel
> 837.808u 138.167s 10:28.96 155.1%       6349+1349k 2873+7780io 303pf+0w
>
> ULE w latest patch
> hulk# time make -j8 buildkernel
> 846.008u 144.377s 10:41.28 154.4%       6364+1354k 7131+7890io 216pf+0w
> kern.sched.interact: 30
> kern.sched.slice_min: 2
> kern.sched.slice: 8
> kern.sched.name: ULE
>
> ULE w latest patch (changed slice)
> hulk# time make -j8 buildkernel
> 840.783u 159.639s 10:37.41 156.9%       6280+1338k 656+8025io 0pf+0w
> kern.sched.slice_min: 4
> kern.sched.slice: 12
> kern.sched.name: ULE
>
> 4BSD
> hulk# time make -j8 buildkernel
> 823.898u 136.524s 9:50.61 162.6%        6263+1334k 7098+7774io 216pf+0w

Hi Jeff,

With your latest patch applied to RELENG_7 (the one that takes packages
into consideration when distributing over cores), I get the following
result.

ULE
hulk# time make -j8 buildkernel
838.774u 147.212s 10:38.59 154.3%       6295+1341k 7139+7880io 214pf+0w

hulk# sysctl kern | grep sched
kern.sched.preemption: 1
kern.sched.topology: 0
kern.sched.steal_thresh: 1
kern.sched.steal_idle: 1
kern.sched.steal_htt: 1
kern.sched.balance_interval: 133
kern.sched.balance: 1
kern.sched.tryself: 1
kern.sched.affinity: 3
kern.sched.pick_pri: 1
kern.sched.preempt_thresh: 64
kern.sched.interact: 30
kern.sched.slice: 13
kern.sched.name: ULE

Rgds,

Patrick

>
> Seems 4BSD wins when it compares to buildkernel. Please let me know if you
> want me to run other tests.
>
> System is:
>
> Timecounter "i8254" frequency 1193182 Hz quality 0
> CPU: AMD Athlon(tm) 64 X2 Dual Core Processor 4800+ (2511.47-MHz K8-class
> CPU)
> Origin = "AuthenticAMD"  Id = 0x60fb1  Stepping = 1
> Features=0x178bfbff<FPU,VME,DE,PSE,TSC,MSR,PAE,MCE,
> CX8,APIC,SEP,MTRR,PGE,MCA,CMOV,PAT,PSE36,CLFLUSH,MMX,FXSR,SSE,SSE2,HTT>
> Features2=0x2001<SSE3,CX16>
> AMD Features=0xea500800<SYSCALL,NX,MMX+,FFXSR,RDTSCP,LM,3DNow!+,3DNow!>
> AMD Features2=0x11f<LAHF,CMP,SVM,ExtAPIC,CR8,Prefetch>
> Cores per package: 2
> usable memory = 3527241728 (3363 MB)
> avail memory  = 3411648512 (3253 MB)
>
> Cheers
>
> Patrick
>
>>
>> Thanks,
>> Jeff
>>
>>
>>> System is AMD Athlon X264 2.2Ghz
>>>
>>> Kernel is GENERIC.
>>> hulk# uname -a
>>> FreeBSD hulk.superhero.nl 7.0-BETA2 FreeBSD 7.0-BETA2 #0: Sun Nov  4
>>> 20:30:23 UTC 2007
>>> root at hulk.superhero.nl:/usr/obj/usr/src/sys/GENERIC
>>> amd64
>>>
>>> I see you have a new patch. Could you send that to me? When I download
>>> it
>>> from the mailman site I get weird characters in the downloaded file :(
>>>
>>> Cheers
>>>
>>> Patrick
>>>
>>
>
>