[Intel-gfx] [PATCH 1/4] drm/i915: Track fence setup separately from fenced object lifetime

[Daniel Vetter]

On Sat, Mar 19, 2011 at 10:42:10PM +0000, Chris Wilson wrote:
> It's meant to be an optimisation where we note that even though this might
> be a pipelined request, the object does not have any outstanding GPU
> fenced access and so we can write the fence register immediately.

Some more thinking cleared stuff up here: last_fenced_seqno can retire
much earlier that last_seqno, so we have to ad-hoc track that for
efficiency.

I still like a

	obj->last_fenced_seqno = 0;

in move_off_active, just to ensure that we don't have terribly old seqnos
lying around.

> Worth a comment after cleaning it up (see the later patch).

So scrap these review comments from me for that patch and perhaps add a
small comment instead as to why we have to track this by hand.

> -Chris

Cheers, Daniel
-- 
Daniel Vetter
Mail: daniel at ffwll.ch
Mobile: +41 (0)79 365 57 48