On Fri, 2003-07-11 at 18:10, Alex Deucher wrote:
> A couple of questions have come up in my mind as I've been looking
> throught the DRM code. In radeon_sarea.h, there is this line:
> int crtc2_base; /* for pageflipping with CloneMode */
>
> I've also noticed references to crtc2 in the
> xc/programs/Xserver/hw/xfree86/os-support/shared/drm/kernel/radeon*
> tree.
>
> The comments regarding this code are a bit sparce and seem to imply
> that the support was added for clone mode. Mergedfb works similarly to
> clone mode, however there are some differences; for instance if you
> have two heads side by side. 3D seems to work pretty well in mergedfb
> mode, but there are a few little issues,
Can you elaborate on these?
> and I'm wondering if they need to be considered in the drm code
> (possibly for pageflipping). this snippet here seems like it might
> need updating (radeon_state.c):
>
> RADEON_WAIT_UNTIL_3D_IDLE();
> OUT_RING_REG( RADEON_CRTC_OFFSET, ( ( sarea->frame.y *
> dev_priv->front_pitch
> + sarea->frame.x
> * ( dev_priv->color_fmt - 2 ) ) & ~7 )
> + offset );
> OUT_RING_REG( RADEON_CRTC2_OFFSET, dev_priv->sarea_priv->crtc2_base
> + offset );
>
> ADVANCE_RING();
>
> Should crtc2 take into account the sarea stuff that crtc1 does?
No, dev_priv->sarea_priv->crtc2_base is supposed to take the CRTC2 frame
position into account. The X server driver must set it correctly in
MergedFB mode, should be easy looking at the clone code.
> The other issue is how to add additional cliprects to make 3D work for
> contexts larger than 2048x2048. Michel made this comment:
>
> --------------------
>
> AFAICS the cliprects are generated in the X server dri module, in
> DRIGetDrawableInfo(). A rough plan would be:
>
> * dri module: add support for maximum cliprect size and position,
> and split the cliprects in DRIGetDrawableInfo() if necessary
> * radeon driver (or any driver simlarly, for that matter):
> * if we're dealing with the new dri module (detected via
> module version or presence/absence of new symbols), set
> the maximum supported cliprect size and position, and
> disable the DRI if the framebuffer size exceeds 4096
> (or
> whatever the correct limit will be) horizontally or
> vertically
> * otherwise, disable the DRI if the framebuffer size
> exceeds 2048 horizontally or vertically.
>
> Of course, I'd first hack in support for splitting the cliprects to
> verify that it actually works as intended.
>
> ----------------------------
>
> what exactly is a cliprect? I've read over the FAQ and the source and
> it seems to define a clipping region if you have overlapping windows.
In X terminology, a cliprect is a rectangular area which is visible.
This translates pretty well into the chip scissor registers. The DRM
basically does in pseudo-code:
foreach ( cliprect ) {
write cliprect coordinates into scissor registers;
flush primitives;
}
> I'm not sure I understand how this fits with the maximum size of the 3D
> context.
The scissor registers seem to be the limiting factor as they only take
12 bits for each dimension of size and position of the rectangle.
> I assume each 3D context has some number of clip rects
> associated with it (at least one?).
Exactly, in fact this is true for every window.
> Does the cliprect define the 3D region? so basically, the code would
> have to check the size of the 3D region requested and if it was wider
> or taller than 2048, it would allocate an additional cliprect?
Exactly.
> I guess I can't quite see how all this is supposed to work together.
I know the feeling. :) You seem to be on the right track though.
> Also what source files should I look in to mess with this?
As I said above, I believe DRIGetDrawableInfo() in
programs/Xserver/GL/dri/dri.c to be the place.
--
Earthling Michel D�nzer \ Debian (powerpc), XFree86 and DRI developer
Software libre enthusiast \ http://svcs.affero.net/rm.php?r=daenzer
-------------------------------------------------------
This SF.Net email sponsored by: Parasoft
Error proof Web apps, automate testing & more.
Download & eval WebKing and get a free book.
www.parasoft.com/bulletproofapps1
_______________________________________________
Dri-devel mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/dri-devel
