[Mesa-dev] [PATCH V2 10/11] genX/cmd_buffer: Enable fast depth clears

Jason Ekstrand jason at jlekstrand.net
Tue Oct 4 01:11:31 UTC 2016


On Tue, Sep 27, 2016 at 3:23 PM, Nanley Chery <nanleychery at gmail.com> wrote:

> On Tue, Sep 27, 2016 at 03:12:17PM -0700, Chad Versace wrote:
> > On Tue 27 Sep 2016, Nanley Chery wrote:
> > > On Tue, Sep 27, 2016 at 11:00:21AM -0700, Chad Versace wrote:
> >
> > > > As a consequence of that reasoning, we should set
> 3DSTATE_CLEAR_PARAMS.DepthClearValueValid = 1
> > > > whenever hiz is enabled, even if we don't care about the actual
> clear value.
>

The logic seems to imply that we can't trust the context to save/restore
our depth clear value so we have to set it every time.  At the very least,
once per batch?  In any case, I doubt there's all that much cost involved
in emitting 3DSTATE_CLEAR_PARAMS so I don't think re-emitting it is that
big of a deal.


> > >
> > > In the V3, I plan to emit that packet once at device initialization
> time
> > > HSW+, and to always emit it (in the expected location) for IVB/BYT.
> Only
> > > the latter platforms have the restriction that it must always be
> > > programmed with the other depth/stencil commands.
> >
> > Is there any benefit to emitting it multiple times on ivb/byt? Does
> > emitting once during initialization, as for hsw, also work for ivb/byt?
> > If so, the code is cleaner if the two gens share the same workaround
> > code.
>
> The benefit for emitting it multiple times on IVB/BYT is that we're
> (possibly) following the oddly-worded programming note for the packet:
>
>    From the IVB PRM Vol2P1, 11.5.5.4 3DSTATE_CLEAR_PARAMS:
>
>       3DSTATE_CLEAR_PARAMS must always be programmed in the along with
>       the other Depth/Stencil state commands(i.e.  3DSTATE_DEPTH_BUFFER,
>       3DSTATE_STENCIL_BUFFER, or 3DSTATE_HIER_DEPTH_BUFFER)
>
> HSW+ doesn't have this restriction, so we're free to only do it once.
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.freedesktop.org/archives/mesa-dev/attachments/20161003/a019d7a3/attachment-0001.html>


More information about the mesa-dev mailing list