<div dir="ltr"><div class="gmail_quote"><div dir="ltr" class="gmail_attr">On Wed, Mar 13, 2019 at 3:56 PM Nanley Chery <<a href="mailto:nanleychery@gmail.com">nanleychery@gmail.com</a>> wrote:<br></div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">On Tue, Mar 12, 2019 at 10:56:27PM -0500, Jason Ekstrand wrote:<br> > Cc: <a href="mailto:mesa-stable@lists.freedesktop.org" target="_blank">mesa-stable@lists.freedesktop.org</a><br> > Cc: Nanley Chery <<a href="mailto:nanley.g.chery@intel.com" target="_blank">nanley.g.chery@intel.com</a>><br> > ---<br> >Â src/intel/vulkan/anv_pass.c | 18 +++++++++++++++++-<br> >Â 1 file changed, 17 insertions(+), 1 deletion(-)<br> > <br> > diff --git a/src/intel/vulkan/anv_pass.c b/src/intel/vulkan/anv_pass.c<br> > index 5fac5bbb31c..ec217abfda0 100644<br> > --- a/src/intel/vulkan/anv_pass.c<br> > +++ b/src/intel/vulkan/anv_pass.c<br> > @@ -178,12 +178,28 @@ anv_render_pass_compile(struct anv_render_pass *pass)<br> >Â Â Â * subpasses and checking to see if any of them don't have an external<br> >Â Â Â * dependency.Â Or, we could just be lazy and add a couple extra flushes.<br> >Â Â Â * We choose to be lazy.<br> > +Â Â *<br> > +Â Â * From the documentation for vkCmdNextSubpass:<br> > +Â Â *<br> > +Â Â *Â Â "Moving to the next subpass automatically performs any multisample<br> > +Â Â *Â Â resolve operations in the subpass being ended. End-of-subpass<br> > +Â Â *Â Â multisample resolves are treated as color attachment writes for the<br> > +Â Â *Â Â purposes of synchronization. This applies to resolve operations for<br> > +Â Â *Â Â both color and depth/stencil attachments. That is, they are<br> > +Â Â *Â Â considered to execute in the<br> > +Â Â *Â Â VK_PIPELINE_STAGE_COLOR_ATTACHMENT_OUTPUT_BIT pipeline stage and<br> > +Â Â *Â Â their writes are synchronized with<br> > +Â Â *Â Â VK_ACCESS_COLOR_ATTACHMENT_WRITE_BIT."<br> > +Â Â *<br> > +Â Â * Therefore, the above flags concerning color attachments also apply to<br> > +Â Â * color and depth/stencil resolve attachments.<br> >Â Â Â */<br> >Â Â Â if (all_usage & VK_IMAGE_USAGE_INPUT_ATTACHMENT_BIT) {<br> >Â Â Â Â pass->subpass_flushes[0] |=<br> >Â Â Â Â Â Â ANV_PIPE_TEXTURE_CACHE_INVALIDATE_BIT;<br> >Â Â Â }<br> > -Â Â if (all_usage & VK_IMAGE_USAGE_COLOR_ATTACHMENT_BIT) {<br> > +Â Â if (all_usage & (VK_IMAGE_USAGE_COLOR_ATTACHMENT_BIT |<br> > +Â Â Â Â Â Â Â Â Â Â VK_IMAGE_USAGE_TRANSFER_DST_BIT)) {<br> >Â Â Â Â pass->subpass_flushes[pass->subpass_count] |=<br> >Â Â Â Â Â Â ANV_PIPE_RENDER_TARGET_CACHE_FLUSH_BIT;<br> <br> I'm assuming you meant to s/depth/color/ in the title of the patch?<br></blockquote><div><br></div><div>Right.Â I'll fix that.<br></div><div>Â </div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"> If so and with that change, this patch is<br></blockquote><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"> Reviewed-by: Nanley Chery <<a href="mailto:nanley.g.chery@intel.com" target="_blank">nanley.g.chery@intel.com</a>><br></blockquote><div><br></div><div>Thanks!<br></div><div>Â </div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"> <br> >Â Â Â }<br> > -- <br> > 2.20.1<br> > <br> > _______________________________________________<br> > mesa-dev mailing list<br> > <a href="mailto:mesa-dev@lists.freedesktop.org" target="_blank">mesa-dev@lists.freedesktop.org</a><br> > <a href="https://lists.freedesktop.org/mailman/listinfo/mesa-dev" rel="noreferrer" target="_blank">https://lists.freedesktop.org/mailman/listinfo/mesa-dev</a><br> </blockquote></div></div>