Weird performances: -CURRENT vs 5.2.1

Mike Tancsa mike at
Wed Aug 18 14:01:31 PDT 2004

At 04:50 PM 18/08/2004, Kris Kennaway wrote:
>On Wed, Aug 18, 2004 at 10:05:30PM +0200, Marc Fonvieille wrote:
> > Hello,
> >
> > Since I installed a recent -CURRENT (from 16/08) on my laptop, I noted
> > a big difference during "make clean" in a port directory.
> > With 5.2.1, the make clean operation is transparent, I mean I don't
> > notice any difference on system load or user interactivity.  With
> > -CURRENT, I see some little freezes, xmms skips etc. and a large swap is
> > used...
>Did you read the big note at the head of UPDATING?

         As someone who is also going to start testing the waters with 
RELENG_5 on some non critical production servers (e.g. one of n spam and av 
scanning machines), I take it you mean,

         FreeBSD 5.x has many debugging features turned on, in
         both the kernel and userland.  These features attempt to detect
         incorrect use of system primitives, and encourage loud failure
         through extra sanity checking and fail stop semantics.  They
         also substantially impact system performance.  If you want to
         do performance measurement, benchmarking, and optimization,
         you'll want to turn them off.  This includes various WITNESS-
         related kernel options, INVARIANTS, malloc debugging flags
         in userland, and various verbose features in the kernel.  Many
         developers choose to disable these features on build machines
         to maximize performance.

Are there more details somewhere as to what exactly to turn off and tweak ?


More information about the freebsd-current mailing list