suggested xorg-compatible video HW for FreeBSD/amd64 ?

[Kostik Belousov]

On Tue, Nov 29, 2011 at 06:03:16PM -0500, Adam K Kirchhoff wrote:
> On 11/29/11 15:17, Jung-uk Kim wrote:
> >On Tuesday 29 November 2011 02:18 pm, Adam K Kirchhoff wrote:
> >>
> >>It is my understanding that to simplify have one unified API for
> >>interacting with the DRM code, the radeon developers (and others)
> >>agreed/decided to use the GEM userspace API, even though the
> >>internals (for radeon DRM) require functionality provided by TTM.
> >Understood.
> >
> >>For what it's worth, the radeon developer I just spoke to even said
> >>in order to remove TTM from the equation, "something" would have to
> >>be recoded to do "partly what ttm does".
> >Yes, *partly*.  That's exactly what I was talking about.  If porting
> >entire TTM layer is harder than recoding to do "partly what TTM
> >does", then it is worth considering, IMHO.
> >
> >Jung-uk Kim
> 
> So we end up with more questions than answers :-)
> 
> Of course, everything would also likely depend on the exact goals of 
> this completely hypothetical Radeon DRM project.
> 
> How much of TTM would need to be re-implemented/ported to simply support 
> 2D acceleration on newer radeon hardware (HD5xxx and higher, and the new 
> APUs)?  How much would be required to support DRI2 and gallium3D?  How 
> much for a full port of KMS?  We can only speculate on what the FreeBSD 
> Foundation would be interested in sponsoring, or what would interest the 
> developer doing the work.
Use of the abbreviations goes in the strange and unexplainable ways.

Would it be easier if I say that whole TTM, execution and KMS bits needs
to be ported ? There are two big DRM infrastructure bits that are missing
right now: TTM and multi-master support. TTM is absolutely critical for
anything non-Intel. Multi-master can be lived without.

TTM as such has no GPU-specific bits, there is a large piece of code
that manages execution for the GPU families.
> 
> Out of curiosity: Can anyone tell me if DRI2 is currently supported on 
> the intel GPUs with Kostik's patches?  Has anyone tried the i915g 
> gallium driver?  It's unofficial, unsupported by Intel, but still has 
> development going on (as compared to i965g, which was dropped from Mesa 
> today).
Do you mean DRI2 protocol ? Yes, it is supported and works, in particular,
vblanks work.

i915g could work, but Gen3 and earlier hardware is supported by my
patchset on the best effort basis - I have no access to such old
machines, and most testers do not either. I had one report of successful
use, see AGP_Testing wiki page. I expect that gallium driver would work,
but I might have to fix a bug or two. Note that it seems that gallium
driver for Gen4 and newer chips was removed several hours ago.

> 
> As a side-note, but still relevant to the discussion:  the r300 and r600 
> classic mesa drivers were dropped from Mesa a few weeks ago.  They 
> are/were the only functioning 3D drivers on FreeBSD for everything from 
> the Radeon 9500 to the HD4950.
> 
> Adam
> 
> 
> 
> _______________________________________________
> freebsd-x11 at freebsd.org mailing list
> http://lists.freebsd.org/mailman/listinfo/freebsd-x11
> To unsubscribe, send any mail to "freebsd-x11-unsubscribe at freebsd.org"
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 196 bytes
Desc: not available
Url : http://lists.freebsd.org/pipermail/freebsd-x11/attachments/20111130/1e6bbe0a/attachment.pgp