[Fedora] Re: Nvidia sucks, sucks, sucks !
Jeff Vian
jvian10 at charter.net
Fri Nov 10 00:47:09 UTC 2006
On Thu, 2006-11-09 at 14:33 -0700, Kim Lux wrote:
> On Thu, 2006-11-09 at 14:04 -0700, Ashley M. Kirchner wrote:
> > Kim Lux wrote:
> > > Here is my post on the subject:
> > >
> > > "When I install a new kernel, my
> > > mouse doesn't quit working !"
> > >
> > > I didn't say the kernel provides mouse support in X !
> > By making the statement you made above, you are suggesting that.
>
> No, I don't. I'm saying that when I install a kernel my mouse doesn't
> quit ! That is it. That is all the sentence says.
>
> > Now, if you said, 'when I install a new gmp rpm, my mouse doesn't quit
> > working', that would make a much better argument.
>
> No it wouldn't. The typical user doesn't know anything about a mouse
> driver. Doesn't need to. Shouldn't have to. It is a reasonable
> expectation that nothing quits when a new kernel is installed !
>
> > The kernel has
> > nothing to do with the mouse, don't compare apples and oranges. You
> > will lose that battle.
>
> Users shouldn't have to know if the kernel has anything to do with the
> mouse or not. Just that the mouse shouldn't quit when they change
> kernels ! Get it ?
>
> > The issue here is that nvidia relies (heavily) on the kernel and
> > thus ties into several aspects of it.
>
> So do other pieces of hardware.
>
> > Under STABLE circumstances, this
> > works fine (as many have already told you.)
>
> Ummm... sorry, but it doesn't.
>
> First of all, thanks once again for introducing the red herring concept
> of me running a test kernel. If you keep repeating a red herring,
> sooner or later some people might get confused, but not really.
>
> If you read my original post, you'll note that I had problems building a
> driver for kernel 2798. As far as I know that kernel is considered
> stable. After all, it is running on most fc6 boxes.
>
> And the fact that the build failed is also irrelevant.
>
> THE RELEVANT POINT IN THIS WHOLE DISCUSSION IS THAT USERS HAVE TO
> *BUILD* A DRIVER IN ORDER TO RUN THE NVIDIA HARDWARE, unless they can
> live with what the non proprietary driver does. (Caps for emphasis, not
> shouting !)
>
You also obviously are overlooking (intentionally or otherwise) the
comment about the i586 kernel install with the i386 arch on fc6. This is
a bug that has been beat to death on this list. That simple bug fix
would have prevented your compile/install of the nvidia driver and would
have allowed the yum install from livna to work.
No one I know has to *BUILD* an nvidia driver for use with any
*released* kernel for FC6. A simple yum install takes care of that.
> > But once again, you fail to
> > comprehend that. You want to run test kernels, that's fine. But don't
> > expect everything to work, otherwise it wouldn't be a test kernel.
>
> This has NOTHING to do with being a test kernel. Are you dense ? The
> nvidia driver has to be built for EVERY kernel and for me the failure
> rate on that build is probably 50%.
>
> > The
> > NTFS module also ties in to the kernel, and when there are kernel
> > updates, I have to wait a day or two to actually install the kernel
> > because the NTFS module hasn't been applied to the new one yet.
>
> Funny, I run ntfs on our server and I don't know a darn thing about how
> it ties into the kernel ! I run yum update on the server, it does its
> thing and I'm done. ntfs has never failed to run when I've rebooted.
> That is the way software is supposed to work.
>
> > But
> > that's just it, I WAIT FOR IT because I know it makes absolutely no
> > sense in running the new kernel while not having the updated drivers,
> > and then come bitch and moan that things don't work. Get that into your
> > mind please: everything will, at some point or another, lack behind a
> > kernel update.
>
> Funny, but the only thing that lags behind a kernel update for me is
> nvidia, now that I run bcm43xx instead of ndiswrapper.
>
> > It's YOUR responsibility to make sure you have the
> > correct drivers, modules, and/or patches BEFORE you update your kernel.
>
> Well, funny thing about that, with nvida YOU CAN'T ! At this point I
> would like to know if you have ever built an Nvidia driver ?
>
> Here is the thing... as far as I know, one can only build an Nvidia
> driver for the kernel that they are running. Maybe the wizards know a
> way around this, but, like I said before, I have a life outside of
> Linux ! So what happens is you boot with the Nvidia driver and then you
> build it and that is when you find out if it works or not.
>
> > Whether the nvidia folks decide to take 24 hours or 24 days to release
> > an update, that's THEIR call, and they will have a reason to do that.
>
> And to hell with the Linux folks and what they need. Because here is
> the thing... I don't have to wait 24 hours or 24 days for the ntfs
> driver or the mouse driver or any of that. It just seems to be there
> when the kernel is ready. I have yet to update a kernel and lose use of
> ntfs or my mouse !
>
> > How many times haven't you heard Microsoft delaying a critical update,
> > while other third party companies have already released theirs? They
> > have a reason to, whether it's for more testing or because the guy
> > responsible is taking a long piss.
>
> See, now open source software is different for all the other drivers
> other than those from nvidia. That is my point ! The source is there
> and when the kernel gets built, so does the latest driver with all the
> recent updates in it. Its a beautiful system ! Kudos to Linus and the
> boys on a job very, very well done. I won't say what I think of nvidia
> being the exception to this system !
>
> > GET OVER IT and deal with the fact
> > that YOU, and only YOU made the choice to run test kernels on your
> > hardware. No one told you that you must.
>
> Once again this has NOTHING to do with running test kernels ! If I
> change kernels right now, to one I have run before, guess what ? I have
> to build another Nvidia driver ! Doesn't matter if the kernel is a bit
> torrent from www.kernel.org or 2.4.x ! Livna *might* have a driver
> built for it, but then again, they might not. Funny, I don't have to go
> to my mouse manufacturers and download an install script for my mouse !
>
> > Once again, if you want to run test kernels, DO NOT EXPECT
> > EVERYTHING TO WORK WITH THEM. It's the whole nature of being a test
> > kernel (or any other package for that matter.)
>
> Once again, nothing to do with the fact its a test kernel !
>
> > It's for people to find
> > the problems and report them, and not for you to come bitching and
> > moaning because you lost some precious 60 minutes of your life. You
> > know what, I couldn't care less how much time you lost.
>
> I didn't expect you to care about how much time I lost, I expected you
> to realize that there is some potential to improve Linux in this regard
> and focus on that.
>
> --
> Kim Lux, Diesel Research Inc.
>
>
More information about the fedora-list
mailing list