Date: Fri, 28 Jun 2002 13:34:07 -0700 (PDT) From: Matthew Dillon <dillon@apollo.backplane.com> To: "Long, Scott" <Scott_Long@adaptec.com> Cc: "'Terry Lambert'" <tlambert2@mindspring.com>, Julian Elischer <julian@elischer.org>, FreeBSD current users <current@freebsd.org>, "'kirk@mckusick.com'" <kirk@mckusick.com> Subject: Re: RE: something funny with soft updates? Message-ID: <200206282034.g5SKY7Un007157@apollo.backplane.com> References: <2C7CBDC6EA58D6119E4A00065B3A24CB0463ED@btcexc01.btc.adaptec.com>
next in thread | previous in thread | raw e-mail | index | archive | help
Before people go off half cocked please do some quantitative tests.
Do at least three buildworld's in a row for each configuration, taking
the timing results of the last two runs. If you do not run at least
three buildworlds your timing results will be seriously skewed.
Major timing differences will result:
* Depending on the state of the /usr/obj tree prior to beginning
the build.
* Depending on the state of the cache in regards to having cached
source files and directories.
* Depending on the amount of debugging turned on in -current.
I'll run a couple of tests with and without softupdates on one of my
(large memory configuration) -stable boxes.
-Matt
Matthew Dillon
<dillon@backplane.com>
:>
:> Maybe the difference in speed is related to the UFS2 commit?
:>
:> -- Terry
:>
:
:I'm seeing similar weirdness in 4.6-STABLE. It seems to be especially
:pronounced on RAID arrays (irregardless of hardware vs software RAID).
:Memory size doesn't seem to make a difference, as an unloaded 2GB
:machine of mine behaves in exactly the same way.
:
:Scott
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?200206282034.g5SKY7Un007157>