Date: Mon, 14 Sep 1998 17:29:30 -0500 (CDT) From: Joel Ray Holveck <joelh@gnu.org> To: Terry Lambert <tlambert@primenet.com> Cc: joelh@gnu.org, tlambert@primenet.com, tom@uniserve.com, gpalmer@FreeBSD.ORG, irc@cooltime.simplenet.com, freebsd-current@FreeBSD.ORG Subject: Re: Download of FreeBSD 3.0-SNAP Message-ID: <199809142229.RAA09230@detlev.UUCP> In-Reply-To: <199809141811.LAA18415@usr05.primenet.com> References: <199809141811.LAA18415@usr05.primenet.com>
next in thread | previous in thread | raw e-mail | index | archive | help
> Note that David points out that FreeBSD *does* do elevator sorting; > it's still not optimal, however, since physical and logical cylinder > boundaries are infrequently the same on modern hardware. I have to > look before I say any more (since I thought the code was removed > circa 2.2.1). David says it's called on all "dumb" drivers (wd, > etc.); I'm not sure the "Ultra" DMA EIDE drivers are still in this > category. I would guess (and this is a guess, feel free to correct) that it's still fairly good, if it simply is performing an elevator sort keyed on block numbers. I would expect that most translations leave the order of the blocks alone. (ie, if block n is closer to the spindle than n+1, then both are closer than n+2, assuming a spiral instead of an actual CHS. Latency is still an issue, but not much.) This would mean that such a sort would still be 100% valid. Best, joelh -- Joel Ray Holveck - joelh@gnu.org - http://www.wp.com/piquan Fourth law of programming: Anything that can go wrong wi sendmail: segmentation violation - core dumped 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?199809142229.RAA09230>