--- Dave Airlie <[EMAIL PROTECTED]> wrote:
> >
> > I have not looked into this in detail.
> > But I fail to see how DRM are different than for example lib-sata,
> > that do not play all these tricks.
>
> we don't currently split our drivers into a library and a core part,
> so we
> have to avoid naming collisions when building everything into the
> kernel,
>
> We sort of include the core part in each driver separately but of
> course
> these collide so the DRM() macro puts the driver name DRM(hello) goes
> to
> i810_hello in i810 and i915_hello in i915, to avoid colliding at
> link...
>
> it's messy and I'd like to cut down uses of it to a bare minimum...
Why don't we just address the real problem and split DRM into
core/personalities. Doing this would eliminate the whole intermodule
mess. Intermodule support could even be removed from the kernel since
DRM is the only user. With all of the macro elimination and current
code rework in DRM I don't this splitting a core off is that hard to
do.
As for the original subject of this email, there is a ref count problem
in the stealth mode code. I can reproduce the bug that was reported.
I've been trying to track it down but I haven't found the cause yet. I
sent Dave my debug code, maybe he can spot it.
The bug probably has something to do with the intermodule support. Code
that could be deleted if we changed to a drm_core model.
=====
Jon Smirl
[EMAIL PROTECTED]
__________________________________
Do you Yahoo!?
Yahoo! Mail - 50x more storage than other providers!
http://promotions.yahoo.com/new_mail
-------------------------------------------------------
This SF.Net email is sponsored by BEA Weblogic Workshop
FREE Java Enterprise J2EE developer tools!
Get your free copy of BEA WebLogic Workshop 8.1 today.
http://ads.osdn.com/?ad_id=5047&alloc_id=10808&op=click
--
_______________________________________________
Dri-devel mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/dri-devel
