Skip site navigation (1)Skip section navigation (2)
Date:      Sat, 14 Sep 2002 02:24:25 -0700 (PDT)
From:      Julian Elischer <julian@elischer.org>
To:        Nate Lawson <nate@root.org>
Cc:        FreeBSD current users <current@FreeBSD.ORG>
Subject:   Re: A riddle in -current
Message-ID:  <Pine.BSF.4.21.0209140209470.82711-100000@InterJet.elischer.org>
In-Reply-To: <Pine.BSF.4.21.0209140030490.28545-100000@root.org>

next in thread | previous in thread | raw e-mail | index | archive | help


On Sat, 14 Sep 2002, Nate Lawson wrote:

> On Fri, 13 Sep 2002, Julian Elischer wrote:
> > 2 identical machines.
> > the same kernel binary
> > /etc (basically) identical (/etc/passwd slightly different)
> > the same system sources compiled and installed
> > 
> > yet:
> > 
> > one is 10% faster than the other CONSISTANTLY in buildworld
> > 
> > after a reboot and 5 minutes of the EXACT same work (buildworld up to 
> > the following message:)
> > --------------------------------------------------------------
> > >>> stage 2: cleaning up the object tree
> > --------------------------------------------------------------
> > the slower machine shows in top:
> > 
> > last pid:  3079;  load averages:  0.01,  0.19,  0.14    up 0+00:13:02
> > 06:13:55
> > 24 processes:  1 running, 14 sleeping, 9 stopped
> > CPU states:  0.4% user,  0.0% nice,  0.0% system,  0.4% interrupt, 99.2%
> > idle
> > Mem: 18M Active, 106M Inact, 27M Wired, 124K Cache, 35M Buf, 94M Free
> > Swap: 513M Total, 513M Free
> > 
> > 
> > while the faster machine shows:
> > last pid:  3079;  load averages:  0.00,  0.15,  0.13    up 0+00:13:25
> > 06:15:55
> > 24 processes:  1 running, 14 sleeping, 9 stopped
> > CPU states:  0.0% user,  0.0% nice,  0.4% system,  0.0% interrupt, 99.6%
> > idle
> > Mem: 11M Active, 36M Inact, 24M Wired, 144K Cache, 35M Buf, 174M Free
> > Swap: 513M Total, 513M Free
> > 
> > Notice in particular the differences in Free and Inactive memory...
> 
> Do they have the same kernel modules loaded?  Same BIOS settings?

As one note.. I can imagine a BIOS setting maj=king it slower,
but I have  alot of trouble seeing howsuch a setting can 
make one machine use 2 x the memory of the other to do the same work.
               Active       Inactive      Wired     Cache    Free
Machine 1       18M           106M         27M       124K    94M
Machine 2       11M           36M          24M       144K    174M

these numbers are RADICALLY different, yet the machines were booted
at the same time with the same hardware on the identically same kernel
and did the same work up to this point.
both however add up to 245MB (the machines have 256M.. I guess something
is not getting counted (kernel text and data?)



> 
> -Nate
> 
> 


To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-current" in the body of the message




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?Pine.BSF.4.21.0209140209470.82711-100000>