Glucose status/instructions request, (and notes on stale branches)

[Keith Packard]

On Thu, 2007-10-18 at 19:20 +0100, Alan Hourihane wrote:

> Why Keith ? Any technical reason ?

I want to get to one acceleration architecture sooner, rather than
later. Right now, that cannot be glucose as it can only run on one VT at
a time. I expect that will be fixed in a reasonable amount of time
though.

My current plan is to abandon XAA support as soon as we've got ttm-based
EXA working, which should at least resolve the worst of the EXA
performance problems.

That'll give us one acceleration architecture that supports our existing
architecture, and also give us a good basis to compare against glucose
to see where full custom code wins and where the better hardware support
in mesa wins.

> It's an optional component, no one is forced to use it.

Users always find these 'optional' components, and I end up having to
support them. I want to encourage OS distributions to ship fewer
options, rather than more so that we can work on creating a stable and
supportable driver.

-- 
keith.packard at intel.com
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 189 bytes
Desc: This is a digitally signed message part
URL: <http://lists.x.org/archives/xorg/attachments/20071018/bd10fd87/attachment.pgp>