[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: X Strike Force X.Org X11 SVN commit: r2912 - in branches/7.1/xserver/xorg-server/debian: . patches



On Mon, Aug 28, 2006 at 12:13:09PM +0200, Michel Dänzer wrote:
> On Sun, 2006-08-27 at 21:14 +0300, Daniel Stone wrote:
> > > > I didn't realize I needed to ask for permission for such things. If the
> > > > -ignoreABI option didn't already exist, I wouldn't have done so. I didn't
> > > > really add anything new, I just provided a new way to get at it.
> 
> It might have been prudent to assume that there could have been reasons
> for it being command line only, and to ask about them before rushing in
> the option.

I figured that it, like so many other things before it, was simply
forgotten. My previous patch to document enabling or disabling aiglx in the
xorg.conf manpage must have been a simple omission rather than a conscious
decision[0]. I figured that this was much the same.
 
> Ignoring an incompatible ABI doesn't cause problems with proprietary
> drivers only. Keep in mind that bad xorg.conf settings tend to spread
> like viruses, e.g. we still have to regularly educate people about the
> braindeadness of enabling backing store or disabling certain extensions
> in a SubSection "extmod", because these happened to be in some example
> xorg.conf sometime at the end of last millennium.

That's true, but any sort of modification to things will spread like this.
The same would have been true if people modified their startx, and it would
have been far more difficult to catch. If we provide the option, we need to
be able to deal with it. I wouldn't mind ripping out -ignoreABI all
together myself.

> > I personally have extreme disdain for proprietary drivers, but I don't
> > think it's right to jump on David for this.  -ignoreABI was already
> > there, 
> 
> My point is that there is a difference between that and an xorg.conf
> option, and that this doesn't seem to have been considered
> appropriately.

Sorry, I did try to consider it. But I've become more and more impressed
with the ingenuity and fervor with which people will come up with solutions
to run the proprietary drivers. The ignoreABI option came from a web search
where someone figured out that you can enable that and disable render
acceleration in the nvidia driver to get it to work. That must have taken
some serious work for the average user to figure out. With or without the
option, I figured people would find a way like they always do. It sucks,
but at least this way we could guarantee that people were able to use the
nvidia-glx package and guarantee that the drivers behaved more or less
correctly with respect to diverting the GL module and so forth.

> > and the release team apparently explicitly requested IgnoreABI as a 
> > prerequisite for supporting 7.1, which I think we can all agree is
> > a very worthwhile end.  If you want to take the release team to task
> > for implicitly supporting binary drivers, I don't think anyone would
> > begrudge you.
> 
> I am indeed quite irritated that making it *convenient* to use
> proprietary drivers (people seem to have managed in the last couple of
> months since the 7.1 release after all) seems to be considered so
> important as to risk a long term negative impact for free software.

I didn't like it either (see Steve's mail) but we have to face the reality
that people will use such drivers right now, and if we prevent them from
doing so actively then they'll go elsewhere.

> > 'Can't we all just get along?',
> 
> I'm trying, and I certainly don't mean to pick on David personally, but
> I'm trying to explain why I don't like the way this was handled.

Thanks, I appreciate that. I'll try to be more careful in the future.

 - David Nusinow

[0] If not then please tell me.



Reply to: