Date: Sat, 11 Oct 1997 22:18:41 -0500 (EST) From: "John S. Dyson" <toor@dyson.iquest.net> To: dkelly@hiwaay.net Cc: chat@FreeBSD.ORG Subject: Re: Linux vs freeBSD Message-ID: <199710120318.WAA02179@dyson.iquest.net> In-Reply-To: <199710112008.PAA02585@nospam.hiwaay.net> from "dkelly@hiwaay.net" at "Oct 11, 97 03:08:29 pm"
next in thread | previous in thread | raw e-mail | index | archive | help
dkelly@hiwaay.net said: > > I'm wrong for saying, "I'd guess that its not being carressed..." > but it took several readings of that sentence to see the error as > I saw the emphasis on "...the way a Linux driver would be." and > didn't mean to infer these features didn't or wouldn't exist. This > relates to the stability issue: I have seen 3 Linuxen at a table in > a computer flea market, each holding a different Linux distribution, > reading aloud to each other the features/versions of each, attempting > to decide which was the newest and most featured, and therefore the > best one to buy. In that market failure to mention UDMA33 would be > fatal. And that was the image I had in mind when I wrote that > sentence. > I don't know if it is wise to add to this discussion or not :-). I am one of the people who hack on the wd driver, but certainly not the owner of it. One of the differences between FBSD and Linux is that we on FBSD tend to avoid touting features. I don't know why if it is cultural or fear :-). For example, given my interest in being "responsible", I tend not to tell people about things that are really new. So often, people use FBSD in mission critical applications, and when new, but green features are crowed about, it eventually causes lots of bug reports about something broken in -current, that is shutting down a $10M business. After moving from -current to -stable (usually 6mos to 1yr later), we -current developers aren't that excited about those new features, and our "advertising literature" is missing the feature. Specifically, we have been "supporting" UDMA33 in -current since we (John Hood?) added the DMA code several months ago or so. The place for most of the setup IS in the bios. UDMA33 sometimes requires support for rogue devices (for example, the Promise Ultra/33 controller), but is pretty consistant for PIIX* controllers from Intel. I do understand your point about making sure that it is clear what we support or don't support for pseudo-competitive analysis, but don't know the best way to fix the problem. :-(. -- John dyson@freebsd.org jdyson@nc.com
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199710120318.WAA02179>