[Intel-gfx] [PATCH 13/14] drm/i915: add some barriers when changing DIPs

Daniel Vetter daniel at ffwll.ch
Wed May 30 23:08:21 CEST 2012


On Mon, May 28, 2012 at 04:43:00PM -0300, Paulo Zanoni wrote:
> From: Paulo Zanoni <paulo.r.zanoni at intel.com>
> 
> On IVB and older, we basically have two registers: the control and the
> data register. We write a few consecutitve times to the control
> register, and we need these writes to arrive exactly in the specified
> order.
> 
> Also, when we're changing the data register, we need to guarantee that
> anything written to the control register already arrived (since
> changing the control register can change where the data register
> points to). Also, we need to make sure all the writes to the data
> register happen exactly in the specified order, and we also *can't*
> read the data register during this process, since reading and/or
> writing it will change the place it points to.
> 
> So invoke the "better safe than sorry" rule and just be careful and
> put barriers everywhere :)
> 
> On HSW we still have a control register that we write many times, but
> we have many data registers.
> 
> Demanded-by: Chris Wilson <chris at chris-wilson.co.uk>
> Signed-off-by: Paulo Zanoni <paulo.r.zanoni at intel.com>

Ok, I've all patches from this series for -next safe for the patches 10,
12 and 14:
- for 14 I've already dumped a bikeshed
- 10 and 12 I like, but I fear we'll get too many merge conflicts against
  -fixes if I merge them this early. I'd still like to include them for
  3.6, so can you please resend these two later in the 3.5 cycle, when
  things have settled a bit for -fixes?

Thanks a lot for digging into this infoframe maze.

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



More information about the Intel-gfx mailing list