Re: ULE/SCHED_SMP diff for 7.0

At 03:26 PM 7/17/2007, Mike Tancsa wrote:
>At 02:35 AM 7/17/2007, Jeff Roberson wrote:
>
>>Even "works for me!" type responses are welcome so I know roughly 
>>how many people have tested before I commit this close to release.

Had some newer boxes I was testing on the bench before putting them 
in the perf cluster.  Same hardware, CURRENT from yesterday



time make -j2 buildworld > /var/log/build.out
time make -j4 buildworld > /var/log/build.out
time make -j8 buildworld > /var/log/build.out
time make -j10 buildworld > /var/log/build.out
time make -j14 buildworld > /var/log/build.out
time make -j16 buildworld > /var/log/build.out
time make -j18 buildworld > /var/log/build.out
time make -j24 buildworld > /var/log/build.out


0[hydra1]# sysctl -a kern.sched
kern.sched.runq_fuzz: 1
kern.sched.preemption: 1
kern.sched.ipiwakeup.htt2: 0
kern.sched.ipiwakeup.onecpu: 0
kern.sched.ipiwakeup.useloop: 0
kern.sched.ipiwakeup.usemask: 1
kern.sched.ipiwakeup.delivered: 36221638
kern.sched.ipiwakeup.requested: 36221461
kern.sched.ipiwakeup.enabled: 1
kern.sched.quantum: 100000
kern.sched.name: 4BSD
0[hydra1]#

2787.488u 451.970s 29:55.01 180.4%      5765+1063k 20950+181419io 1935pf+0w
2829.800u 593.261s 17:26.51 327.0%      5703+1055k 1747+181416io 536pf+0w
2914.180u 831.262s 12:14.97 509.6%      5650+1048k 1547+181416io 534pf+0w
2943.049u 864.558s 12:29.44 508.0%      -5630+1048k 1549+181449io 534pf+0w
2951.000u 908.972s 12:42.39 506.2%      -5481+1048k 1544+181439io 534pf+0w
2952.665u 931.201s 12:39.08 511.6%      -5413+1047k 1541+181397io 534pf+0w
2961.467u 937.847s 12:49.09 507.0%      -5371+1047k 1543+181423io 534pf+0w
2961.957u 983.947s 12:58.20 507.0%      -5238+1047k 1542+181435io 534pf+0w

vs
ULE
0[hydra2]# sysctl kern.sched
kern.sched.preemption: 1
kern.sched.topology: 0
kern.sched.steal_thresh: 2
kern.sched.steal_idle: 1
kern.sched.steal_htt: 0
kern.sched.balance_secs: 1
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
0[hydra2]#
2780.017u 425.475s 28:40.12 186.3%      5806+1068k 21162+4934io 1929pf+0w
2820.171u 521.349s 16:18.90 341.3%      5772+1064k 1693+4968io 534pf+0w
2866.390u 676.119s 12:19.83 478.8%      5714+1057k 1648+4970io 534pf+0w
2883.521u 709.969s 12:29.52 479.4%      5701+1055k 1647+5006io 534pf+0w
2911.001u 801.767s 12:55.74 478.6%      5680+1053k 1650+4974io 534pf+0w
2926.022u 863.136s 13:17.31 475.2%      -5656+1052k 1643+5013io 534pf+0w
2937.156u 923.333s 13:28.78 477.3%      -5448+1051k 1640+5048io 534pf+0w
2969.080u 1096.393s 13:50.15 489.7%     -4885+1051k 1642+5126io 534pf+0w


CPU: Intel(R) Xeon(R) CPU           E5320  _at_ 1.86GHz (1866.74-MHz 
686-class CPU)
   Origin = "GenuineIntel"  Id = 0x6f7  Stepping = 7
   Features=0xbfebfbff<FPU,VME,DE,PSE,TSC,MSR,PAE,MCE,CX8,APIC,SEP,MTRR,PGE,MCA,CMOV,PAT,PSE36,CLFLUSH,DTS,ACPI,MMX,FXSR,SSE,SSE2,SS,HTT,TM,PBE>
   Features2=0x4e3bd<SSE3,RSVD2,MON,DS_CPL,VMX,EST,TM2,SSSE3,CX16,xTPR,PDCM,DCA>
   AMD Features=0x20100000<NX,LM>
   AMD Features2=0x1<LAHF>
   Cores per package: 4
real memory  = 3489005568 (3327 MB)
avail memory = 3409797120 (3251 MB)
ACPI APIC Table: <PTLTD          APIC  >
FreeBSD/SMP: Multiprocessor System Detected: 8 CPUs
  cpu0 (BSP): APIC ID:  0
  cpu1 (AP): APIC ID:  1
  cpu2 (AP): APIC ID:  2
  cpu3 (AP): APIC ID:  3
  cpu4 (AP): APIC ID:  4
  cpu5 (AP): APIC ID:  5
  cpu6 (AP): APIC ID:  6
  cpu7 (AP): APIC ID:  7
ioapic0 <Version 2.0> irqs 0-23 on motherboard
ioapic1 <Version 2.0> irqs 24-47 on motherboard


         ---Mike