Date: Wed, 1 May 1996 12:01:58 -0600 From: Nate Williams <nate@sri.MT.net> To: dwhite@resnet.uoregon.edu Cc: Nate Williams <nate@sri.MT.net>, questions@FreeBSD.org Subject: Re: Hello Message-ID: <199605011801.MAA07508@rocky.sri.MT.net> In-Reply-To: <Pine.BSF.3.91.960501103220.10061E-100000@riley-net170-164.uoregon.edu> References: <199605010034.SAA05536@rocky.sri.MT.net> <Pine.BSF.3.91.960501103220.10061E-100000@riley-net170-164.uoregon.edu>
next in thread | previous in thread | raw e-mail | index | archive | help
> > In order to do this, I need *testers* who can tell me how things work, > > and if folks only use the Nomad code this isn't going to help me at all. > > Telling folks that 'You _want_ to apply the Nomad PCMCIA patches' > > implies that there is no other solution, or that the other solutions are > > somehow 'bad'. Please avoid making my job harder, as I need some help. > > Then, I'm sorry I stepped on your toes. I was going on what (little) > information I had and my personal experience. I wan't aware of what the > current status was regarding your work with it. One of my co-workers is > on -mobile, but he hasn't forwarded any information over, so I'm in the > dark. I started the mobile mailing list, and have yet to post to it. :( This is MY problem, as it's much easier to sort of let it slide rather than taking the time to actually put together a 'status' document. I've got one started, but about the time I'm happy with it things change enough that I have to modify it so I don't send it out. I start modifying it, and then just about the time I get around to sending it out, it needs to be modified once again. That's the problem with having 3 development branches I guess. > My remaining question is, although it may be moot in a couple of months > with 2.1.x, what are 2.1-R people supposed to do for PCMCIA support? That's a fair question, and here's the answer which you aren't going to like. It depends. What are your desires? Do you want to help, or just get a 'working' system? If you just want a working system, then get the Nomad stuff and apply it to 2.1R, until I make my patch-set against -stable. Note, Hosokawa already told me that the next release will contain some stuff that is very destabilizing w/regards to combining APM and PC-CARD support. But, that won't matter since the Nomad's are no longer support 2.1, so you won't have patches anyway. Hopefully the old release will be kept around, but if not then there is *NOTHING* for a 2.1R user to do except upgrade to -current. OR, if you want to help out (please!), you can do one of a couple of things, but all involve getting the Nomad patches. Either upgrade to -current and test out the code (w/out the Nomad patches), and then add the Nomad patches if necessary. This will help me out more than you know in that it tests the kernel functionality of both the APM and PC-CARD code, which is pretty much stable as of now. If your card's not supported, you should be able to add in just the driver (and entry into /etc/pccard.conf) from the Nomad's patchset and see if that's enough. Or, if you fancy yourself a programmer you can either use -current or -stable and try to fit in the Nomad patches into there *WITHOUT* de-stabilizing either one. Adding PC-CARD patches to a driver is fairly easy, but many (most?) of the drivers aren't written to be 'hot' swappable. This means that the drivers has some assumptions about the state of the system which aren't valid w/PC-CARD drivers. For example, the 3C3X9 driver (if_ep.c) assumes that all of the bus architectures exist at boot time, so the probe portions of the driver need to be re-written to allow for how-swapping. This kind of work is also good in that it makes the driver more likely to become an LKM, since that's kind of ability is necessary there as well. (At some point, I suspect the PC-CARD and LKM driver code will merge.) The above tasks aren't that difficult (really!), but it requires someone with the necessary hardware to test and the time to do the job. Any kernel guru wannabe (like me) should be able to handle this job w/no problems. I have 4 laptops available, so I can test *some* things pretty well, but I can't test my PC-CARD mods to the if_ep driver to see if it breaks non-PCMCIA cards. If someone on the list had a 3C589 card *AND* a 3C509 card they would be perfect for this job. Please folks, I can't do this alone. I'm going to start making some changes to the drivers in -current on a branch to avoid hosing up the normal users, but that's only to make the code more available for general consumption. If folks don't actually *test* the code, we're not going to get anywhere. This may sound a bit nasty, but if you aren't interested in helping me test and get the FreeBSD laptop support working better, you may as well run Linux where it already works pretty well. Nate
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199605011801.MAA07508>