Date: Thu, 6 May 2004 08:37:39 -0700 From: Joe Rhett <jrhett@isite.net> To: =?unknown-8bit?B?U8O4cmVu?= Schmidt <sos@DeepCore.dk> Cc: current@freebsd.org Subject: Re: Updated SATA support patches for Intel ICH & Promise cards Message-ID: <20040506153739.GA21615@isite.net> In-Reply-To: <409A0C7D.7070909@DeepCore.dk> References: <20040504235754.GA8@isite.net> <Pine.LNX.4.44.0405041909590.31369-100000@pancho> <20040505224912.GA9165@isite.net> <409A0C7D.7070909@DeepCore.dk>
next in thread | previous in thread | raw e-mail | index | archive | help
I'm not certain I'll have the time to assist with code, but I do have large amounts of various hardware so I can certainly spend time testing the code and/or getting you a build environment on the hardware if you don't have it already. On Thu, May 06, 2004 at 11:59:25AM +0200, Sren Schmidt wrote: > Joe Rhett wrote: > >Just so you know, you're singing to the choir here. I think everyone > >realizes that its going to take coordination. But Soren didn't say > >"whoah, let's try this a different way" he went directly to "you're wasting > >my time" without even a basic attempt to consolidate the efforts. > > I dont recall using those words, neither does my mail archive. > > So now that we are wasting our time on this lets gets things straight. > > On april 3rd Doug posted a patch to solve some of the ich problems on > -current. I responded with a couble of questions and some of the stuff > in there was fixed (differently or correctly depending on your view) in > -current back when. Then time passed, I got access to Intel HW here, and > then this patch shows up. I tell Doug that it conflicts with the work > I've been doing to get the Intel parts flying, but that the RAID > metadata code he talks about would be welcomed (but thats not in the > patch at hand mind you). Then all hell breaks loose... > > Now, since I'm working on getting *ALL* SATA controllers working > properly (not just Intel) and getting error/status info from them (which > is pretty much ignored currently), I have to take a more general look at > things, and have devised a scheme and implemented most of it already. > This scheme doesn't fit with Doug's work, but parts of it is semilar in > functionality for the Intel parts (surprise). > > Maybe I should spend more time publishing whats going on here about ATA > but until now I've prefered to use whatever time and resources I have to > make things happend instead. And frankly I havn't noticed any interest > in this from anyone, except when things break I get plenty of nagging > emails telling me so. Time is a precious resource these days and having > to deal with situations like this doesn't help motivation either. > > So, for those *really* interested in ATA work and I mean interested in > the sense of spending hours on it every day or at least on a very > regular basis, lets talk about it off the lists. The project needs one > to take on doing ATA maintenance on -stable, which would be a fine > introduction to doing ATA work that could make a difference.... > > That said, I'll still take my position as ATA developer/maintainer up to > revision, it takes *ALOT* of my time, and frankly it hasn't been anyway > near as fun lately as it used to be... > > >On Tue, May 04, 2004 at 07:15:15PM -0500, Mark Linimon wrote: > > > >>>It sounds very much like you're putting your ego far, far in front > >>>of functional code. > >> > >>I think you should go back and read the history of the work that > >>Soren has done before you criticize him. I am not at all surprised > >>that he has a large number of patches in the queue due to the amount > >>of work that he puts in on ATA (mostly unheralded). > >> > >>There is certainly plenty of work to go around in this area, but > >>no doubt needs careful coordination due to the complexity of the > >>work involved -- it has seemed, in the past, that seemingly-trivial > >>changes break old, brain-damaged, ATA hardware. > >> > >>Please, folks, let's try to work together on these really hard > >>problems, and respect that fact that coordinating lots of changes > >>are always going to require extra work -- but it will get FreeBSD > >>a lot further along towards its goals in the long run. > >> > >>mcl > > > > > > > -- > -Søren -- Joe Rhett Chief Geek JRhett@Isite.Net Isite Services, Inc.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20040506153739.GA21615>