Date: Wed, 7 May 2003 05:05:41 -0700 From: "Cliff L. Biffle" <cbiffle@safety.net> To: "W. Josephson" <cvs-D20030429@morphisms.net>, freebsd-current@freebsd.org Subject: Re: recent performance problems? Message-ID: <200305070505.41849.cbiffle@safety.net> In-Reply-To: <20030507054602.GA67627@mero.morphisms.net> References: <20030507022653.GA66145@mero.morphisms.net> <20030506222117.I5620@znfgre.qbhto.arg> <20030507054602.GA67627@mero.morphisms.net>
next in thread | previous in thread | raw e-mail | index | archive | help
On Tuesday 06 May 2003 10:46 pm, W. Josephson wrote: > I wouldn't notice a difference of 10-25% given my current > usage pattern on the laptop -- we're talking a factor of three or so > in the time to build world (~6 hrs on a 900MHz PIII with 5400RPM > disk); opening a new xterm in X11 now takes a second or so. I'm in > the process of building a new kernel and am going to try locking the > CPU into high performance mode. That hasn't been necessary in the > past with either -STABLE or any version of -CURRENT, but the low clock > rate reported when the kernel boots is suspicious. The numbers you report for buildworld time are roughly on par for my 233mhz box here, on a good day. It's a kick-off-the-build, leave-for-work, come-back-after-dinner sort of box. :-) But my point is, that reported speed might not be so far off, for whatever reason. I couldn't tell from your dmesg if you're using ACPI. If not, does enabling it (on a recent -current) help? I've been running into strange situations with CPU throttling on my machines here, though a throttled CPU still reports the right speed for me.... -Cliff L. Biffle
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?200305070505.41849.cbiffle>