What i965 bottlenecks would still exist if we fixed memcpy and I830WaitSync?
cworth at cworth.org
Fri Jul 13 15:25:51 PDT 2007
Here's a note I've written up on trying to determine what bottlenecks
would still exist in the case of EXA+i965 if a couple of the recent
problems I identified were first fixed:
I'd be quite interested in any details on current work toward fixing
those existing problems. The things I mention are:
* Glyphs pinned in system memory, (Keith says he's working on a fix
* Artificial separation of system and video memory on i965 (memcpy
pain). I'm told that the TTM work will fix this. Where can I find
status on that work or perhaps find how I could help?
* Excessive waiting in the i965 driver. I can hopefully mitigate much
of this with some simple caching, (reducing the waiting to happen
only every N operations instead of on every operation). Eliminating
the waiting altogether would be even better---but I'd need some
advice on how to do that. Anybody have that?
If anyone has any other feedback on these, (or the subsequent problems
that will also need to be fixed as I describe in the post linked to
above), I would greatly appreciate it.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Size: 189 bytes
Desc: not available
More information about the xorg