Date: Sun, 9 Aug 1998 16:44:37 +0930 From: Greg Lehey <grog@lemis.com> To: dg@root.com Cc: John Baldwin <jobaldwi@vt.edu>, FreeBSD Hackers <hackers@FreeBSD.ORG> Subject: Re: AMD-specific kernel code (was: How long a wait?) Message-ID: <19980809164437.C11095@freebie.lemis.com> In-Reply-To: <199808090649.XAA23702@implode.root.com>; from David Greenman on Sat, Aug 08, 1998 at 11:49:51PM -0700 References: <19980809160829.A11214@freebie.lemis.com> <199808090649.XAA23702@implode.root.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On Saturday, 8 August 1998 at 23:49:51 -0700, David Greenman wrote: >> The effect on CPU performance is also noticable: here are "make >> world" times: >> >> Without write allocate: >> >> real 106m58.441s >> user 63m0.960s >> sys 20m25.035s >> >> With write allocate: >> >> real 114m16.402s >> user 69m41.733s >> sys 17m54.862s >> >> These were only a rough test (I had other stuff running at the same >> time), but in practice this normally only makes about 1 minute >> difference. I'd guess that the difference *is* real. Obviously there >> must be a reason for enabling or disabling this behaviour, and I'd >> guess that write allocation is most effective when one process is >> using much CPU time. 'make world' starts thousands of short-lived >> process, and may thus be the worst-case scenario. In this connection, >> it's also interesting to note that system time was down and user time >> was up with write allocation. >> >> I have patches available. Does anybody want to commit them? > > Based on the above performance numbers, I'd say no - they appear to be > a pessimization. If they were always like that, I'd agree. But as I said, I think it depends on the application, and I'm prepared to believe that long-running servers could benefit significantly. Still, that's not proven. I've put the patches on ftp://ftp.lemis.com/pub/AMD-write-allocate-patch. They relate to today's -CURRENT. How about some other people grabbing them and trying them out? Greg -- See complete headers for address and phone numbers finger grog@lemis.com for PGP public key To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hackers" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?19980809164437.C11095>