Date: Fri, 18 Dec 1998 21:04:05 -0800 (PST) From: Chris Timmons <skynyrd@opus.cts.cwu.edu> To: "Jordan K. Hubbard" <jkh@zippy.cdrom.com> Cc: current@FreeBSD.ORG Subject: Re: Also seeing IRQ oddness with -current now. Message-ID: <Pine.BSF.3.96.981218205354.13313A-100000@opus.cts.cwu.edu> In-Reply-To: <31115.913981805@zippy.cdrom.com>
next in thread | previous in thread | raw e-mail | index | archive | help
I think the irq messages have been there all along for SMP. The most dramatic performance increase I perceive from SMP is in overall throughput on a machine with diverse tasks running concurrently. Put your 'make -j12 buildworld' in a loop, and put the build of ports/x11/XFree86 in a loop, such that you get a new logfile of stdout for each iteration of the builds. Try this for a while with one CPU, and then add the 2nd CPU. You'll see a pretty pronounced difference in your buildworld time then. And that is the real value of SMP right now - not breathtaking speed for single tasks (even when aided by make's parallelism) but increased throughput for highly loaded machines. I'm going to put cvsup.freebsd.org on a dual ppro machine shortly which will maybe help me decide the question of how ready for prime time 3.0-current is. Right now the uniprocessor pro/512 2.2-stable machine looks like this: 9:00PM up 16 days, 22:14, 2 users, load averages: 9.82, 9.07, 9.40 certainly no wcarchive but it does do it's share :) -Chris On Fri, 18 Dec 1998, Jordan K. Hubbard wrote: [...] > With one CPU, it's 1:05. With two CPUs, it's 1:01. This either > proves that I'm seriously I/O bound or that multiple CPUs aren't > helping the compile times for some weird reason. :-) [...] 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.3.96.981218205354.13313A-100000>