<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>
>  src/intel/vulkan/anv_pass.c | 18 +++++++++++++++++-<br>
>  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>
>      * subpasses and checking to see if any of them don't have an external<br>
>      * dependency.  Or, we could just be lazy and add a couple extra flushes.<br>
>      * We choose to be lazy.<br>
> +    *<br>
> +    * From the documentation for vkCmdNextSubpass:<br>
> +    *<br>
> +    *    "Moving to the next subpass automatically performs any multisample<br>
> +    *    resolve operations in the subpass being ended. End-of-subpass<br>
> +    *    multisample resolves are treated as color attachment writes for the<br>
> +    *    purposes of synchronization. This applies to resolve operations for<br>
> +    *    both color and depth/stencil attachments. That is, they are<br>
> +    *    considered to execute in the<br>
> +    *    VK_PIPELINE_STAGE_COLOR_ATTACHMENT_OUTPUT_BIT pipeline stage and<br>
> +    *    their writes are synchronized with<br>
> +    *    VK_ACCESS_COLOR_ATTACHMENT_WRITE_BIT."<br>
> +    *<br>
> +    * Therefore, the above flags concerning color attachments also apply to<br>
> +    * color and depth/stencil resolve attachments.<br>
>      */<br>
>     if (all_usage & VK_IMAGE_USAGE_INPUT_ATTACHMENT_BIT) {<br>
>        pass->subpass_flushes[0] |=<br>
>           ANV_PIPE_TEXTURE_CACHE_INVALIDATE_BIT;<br>
>     }<br>
> -   if (all_usage & VK_IMAGE_USAGE_COLOR_ATTACHMENT_BIT) {<br>
> +   if (all_usage & (VK_IMAGE_USAGE_COLOR_ATTACHMENT_BIT |<br>
> +                    VK_IMAGE_USAGE_TRANSFER_DST_BIT)) {<br>
>        pass->subpass_flushes[pass->subpass_count] |=<br>
>           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.  I'll fix that.<br></div><div> </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> </div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">
<br>
>     }<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>