Keith Packard wrote:
>On Sat, 2007-12-22 at 07:59 +0100, Thomas Hellström wrote:
>
>
>
>>The way I would do this with a small extension to the current
>>implementation is to
>>implement a DRM_I915_FENCE_TYPE_READ fence type and a
>>DRM_I915_FENCE_TYPE_WRITE fence type, which are triggered by t
On Sat, 2007-12-22 at 07:59 +0100, Thomas Hellström wrote:
> The way I would do this with a small extension to the current
> implementation is to
> implement a DRM_I915_FENCE_TYPE_READ fence type and a
> DRM_I915_FENCE_TYPE_WRITE fence type, which are triggered by the
> corresponding buffer obj
On Sat, 2007-12-22 at 09:09 +0100, Thomas Hellström wrote:
> This should be easy. We can have drm_ttm_get_page() come in two
> varieties, one which tests
> for NULL only and one that tests for dummy_page or NULL and allocates a
> new page if needed. drm_ttm_get_page() is called from the nopfn()
On Sat, 2007-12-22 at 10:19 +0100, Thomas Hellström wrote:
> There is indeed, although you and Eric have improved the situation
> considerably.
> I hope there will be some time available to document things more
> properly soon.
I think it works fairly well for me to try and document things as
http://bugs.freedesktop.org/show_bug.cgi?id=13527
[EMAIL PROTECTED] changed:
What|Removed |Added
Status|NEW |RESOLVED
Resolution|
http://bugs.freedesktop.org/show_bug.cgi?id=12164
--- Comment #28 from [EMAIL PROTECTED] 2007-12-22 04:31 PST ---
According to Dietrich Bollmann, the crash is fixed with latest mesa 7.0.x
branch (he's using Debian Mesa 7.0.2-3 which contains the branch up to commit
0107acde).
--
Co
Keith Packard wrote:
>On Fri, 2007-12-21 at 13:21 +0100, Thomas Hellström wrote:
>
>
>
>>Keith,
>>This is purely an optimization.
>>Buffers that are evicted due to a memory aperture space shortage ends up
>>in a
>>state where they are unbound from the aperture, but retain the uncached
>>state
Keith Packard wrote:
>On Fri, 2007-12-21 at 14:01 +0100, Thomas Hellström wrote:
>
>
>
>>This looks like a good thing, but is there a way we can avoid doing this
>>in the ttm backend?
>>Some backends (Poulsbo, and I think Dave's work on Radeon) implement a
>>null populate() function and use th
