[Mesa-dev] [PATCH v2 18/18] anv/image: Allow HiZ on input attachment-capable depth/stencil images

Jason Ekstrand jason at jlekstrand.net
Thu Mar 2 19:03:44 UTC 2017


On Mar 2, 2017 10:57 AM, "Nanley Chery" <nanleychery at gmail.com> wrote:

On Wed, Mar 01, 2017 at 04:46:42PM -0800, Jason Ekstrand wrote:
> I didn't re-read every patch, just the ones I made comments on.  Assuming
> only the commented patches made real changes, the series is
>
> Reviewed-by: Jason Ekstrand <jason at jlekstrand>
>
> I really like the direction things are going.
>

Thanks!

I had to update one patch due to rebasing conflicts that occurred today.
Please see:
[PATCH v3] anv/descriptor_set: Store aux usage of sampled image descriptors


I might have split that into two patches with the refactoring separate from
the aux usage change bit they're fine squashed together.

Also, updating the patch, "anv/image: Add anv_layout_to_aux_usage()",
caused the patches that modify transition_depth_buffer() to change. The


Yup.  I saw that.  Looks fine to me.

end result looks about the same to me. Does your Rb still apply?


Yup.  Land it.


-Nanley

> On Wed, Mar 1, 2017 at 2:30 PM, Nanley Chery <nanleychery at gmail.com>
wrote:
>
> > While an input attachment may only take on one of those two layouts,
> > other depth/stencil attachments that use the same image may have
> > HiZ-enabled layouts. Improves the average frame rate on a release
> > candidate of a proprietary Vulkan benchmark by 9.94% over 3 runs on my
> > SKL GT4.
> >
> > Signed-off-by: Nanley Chery <nanley.g.chery at intel.com>
> >
> > --
> >
> > The tests have been fixed upstream.
> > ---
> >  src/intel/vulkan/anv_image.c | 14 --------------
> >  1 file changed, 14 deletions(-)
> >
> > diff --git a/src/intel/vulkan/anv_image.c b/src/intel/vulkan/anv_image.c
> > index 95ba12c928..59f730c7c6 100644
> > --- a/src/intel/vulkan/anv_image.c
> > +++ b/src/intel/vulkan/anv_image.c
> > @@ -190,20 +190,6 @@ make_surface(const struct anv_device *dev,
> >         */
> >        if (!(image->usage & VK_IMAGE_USAGE_DEPTH_STENCIL_
ATTACHMENT_BIT))
> > {
> >           /* It will never be used as an attachment, HiZ is pointless.
*/
> > -      } else if (image->usage & VK_IMAGE_USAGE_INPUT_ATTACHMENT_BIT) {
> > -         /* From the 1.0.37 spec:
> > -          *
> > -          *    "An attachment used as an input attachment and
> > depth/stencil
> > -          *    attachment must be in either VK_IMAGE_LAYOUT_GENERAL or
> > -          *    VK_IMAGE_LAYOUT_DEPTH_STENCIL_READ_ONLY_OPTIMAL."
> > -          *
> > -          * It will never have a layout of
> > -          * VK_IMAGE_LAYOUT_DEPTH_STENCIL_ATTACHMENT_OPTIMAL, so HiZ is
> > -          * currently pointless. If transfer operations learn to use
the
> > HiZ
> > -          * buffer, we can enable HiZ for VK_IMAGE_LAYOUT_GENERAL and
> > support
> > -          * input attachments.
> > -          */
> > -         anv_finishme("Implement HiZ for input attachments");
> >        } else if (!env_var_as_boolean("INTEL_VK_HIZ", dev->info.gen >=
> > 8)) {
> >           anv_finishme("Implement gen7 HiZ");
> >        } else if (vk_info->mipLevels > 1) {
> > --
> > 2.12.0
> >
> > _______________________________________________
> > mesa-dev mailing list
> > mesa-dev at lists.freedesktop.org
> > https://lists.freedesktop.org/mailman/listinfo/mesa-dev
> >
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.freedesktop.org/archives/mesa-dev/attachments/20170302/c153cd90/attachment.html>


More information about the mesa-dev mailing list