ULE/SCHED_SMP diff for 7.0

[Mike Tancsa]

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.

"Works for me" although it seems a bit slower than the old ULE on the 
-j2 buildworld.... But -j4 is as fast.  If useful, I could reload the 
old kernel and run a -j4 to see what times it does.

Regular ULE
[em-test]# time make -j2 buildworld > /var/log/build.out ; time make 
-j2 buildkernel > /var/log/build.out.k
3021.917u 342.449s 32:40.21 171.6%      5900+1085k 29135+4887io 2343pf+0w
298.402u 22.080s 3:00.91 177.1% 6247+1037k 1750+1962io 4pf+0w



[em-test]# time make -j2 buildworld > /var/log/build.out ; time make 
-j2 buildkernel > /var/log/build.out.k ; mv /usr/obj/usr 
/usr/obj/usr.o ; sleep 20 ; time make -j4 buildworld > 
/var/log/build.out ; time make -j4 buildkernel > /var/log/build.out.k
3022.101u 351.076s 45:16.79 124.1%      5871+1080k 25578+4898io 2264pf+0w
299.681u 22.354s 3:56.01 136.4% 6234+1034k 1600+1853io 0pf+0w
3057.276u 357.860s 31:57.84 178.0%      5889+1081k 2310+5057io 534pf+0w
302.924u 23.387s 2:58.57 182.7% 6231+1034k 127+1846io 0pf+0w
[em-test]#


CPU: AMD Athlon(tm) 64 X2 Dual Core Processor 3800+ (2015.01-MHz 686-class CPU)
   Origin = "AuthenticAMD"  Id = 0x20fb1  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=0x1<SSE3>
   AMD Features=0xe2500800<SYSCALL,NX,MMX+,FFXSR,LM,3DNow!+,3DNow!>
   AMD Features2=0x3<LAHF,CMP>
   Cores per package: 2
real memory  = 2147418112 (2047 MB)
avail memory = 2096111616 (1999 MB)