Experimental options

[Eric Anholt]

On Sat, 2004-12-11 at 03:52 -0500, Vladimir Dergachev wrote:
> Hi all,
> 
>      I would like to get some feedback on what people think
> about introducing "experimental" options in the video drivers
> and having a standardized way of marking them as such.
> 
>      In particular, I thought about classifying as "experimental"
> any options that turn on behaviour that can result in lockup or
> breakage of standard protocols.
> 
>      I would propose that all such options be prefixed with "X_"
> and default of "false" (for obvious reasons). It'd be nice to have
> a webpage (and/or README) that explains that all options starting
> with "X_" are experimental and users turn them on at their own risk.
> 
>      For example, I would like to add an option to radeon driver
> to enable 3d for R300 cards with non-existent "r300_dri.so" mesa driver.
> The reason this is useful is that such trick greatly accelerates 3d on
                                                                   ^ 2? :)
> large resolution screens as 2d driver can also use DRM driver and this 
> works perfectly well. This also facilitates work on 3d driver.
> 
>      The 2d speedup is quite considerable, however one would expect all GL
> apps to break because of this, as even software rendering won't work.

libGL should fall back to GLX when the *_dri.so isn't around.  Is it
not?  Even so, it would be nice to make some sort of change to make
libdri not advertise DRI support to clients in this case, but use the
rest of the facilities within the server.  Seems like it should be
pretty possible.

-- 
Eric Anholt                                eta at lclark.edu          
http://people.freebsd.org/~anholt/         anholt at FreeBSD.org