Date: Wed, 8 Jan 1997 23:35:51 -0500 (EST) From: Charles Henrich <henrich@crh.cl.msu.edu> To: freebsd-hackers@freebsd.org Subject: is time() valid in kernel? Message-ID: <199701090435.XAA06374@crh.cl.msu.edu>
next in thread | raw e-mail | index | archive | help
I want to have the kernel time how long sync() takes in the vfs_update()
routine, is it legit to throw in a standard call to gettimeofday() or is there
a cooler (e.g. macro) that will give me the time so I can find out how long
this sucker is taking?
Why: I want to experiment changing how frequently update() runs on a busy
newserver, and see what the time differences are. I have a gut-feeling that
changing it from 30 seconds to 300 or more seconds wont significantly increase
the sync() time, but as I'm now doing it 1/10th as often I get a large
performance win.. [This system is attached to a UPS that notifies FreeBSD when
the power drops which starts doing sync's and after a few minutes shuts down]
-Crh
Charles Henrich Michigan State University henrich@msu.edu
http://pilot.msu.edu/~henrich
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199701090435.XAA06374>