re: CVS commit: src/sys (sched_pstats() + uvm_meter()/uvm_loadav())

[matthew green <mrg%eterna.com.au@localhost>]

   On Friday 2010-04-16 03:21 +0000, Mindaugas Rasiukevicius output:
   :Module Name:        src
   :Committed By:       rmind
   :Date:               Fri Apr 16 03:21:49 UTC 2010
   :Modified Files:
   :    src/sys/kern: kern_synch.c
   :    src/sys/sys: sched.h
   :    src/sys/uvm: uvm_extern.h uvm_glue.c uvm_meter.c
   :
   :Log Message:
   :- Merge sched_pstats() and uvm_meter()/uvm_loadav().  Avoids double loop
   :  through all LWPs and duplicate locking overhead.
   :
   :- Move sched_pstats() from soft-interrupt context to process 0 main loop.
   :  Avoids blocking effect on real-time threads.  Mostly fixes PR/38792.
   :
   :Note: it might be worth to move the loop above PRI_PGDAEMON.  Also,
   :sched_pstats() might be cleaned-up slightly.
   :
   :To generate a diff of this commit:
   :cvs rdiff -u -r1.280 -r1.281 src/sys/kern/kern_synch.c
   
   Something is funny with this new merged "load average" calculation.
   
        % uptime
         2:06PM  up 33 mins, 1 user, load averages: 11.04, 11.03, 9.96
        %
   
   It should be showing about a tenth of that.
   
   This is i386 with SMP (Core2Duo).

ah, this may be what i'm seeing on a sparc64 box... "build.sh -j4" is
showing load's in the 12-15 range.
   

.mrg.