Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Remove transient bit for depth image usage flags #883

Open
wants to merge 1 commit into
base: main
Choose a base branch
from

Conversation

asuessenbach
Copy link
Contributor

Description

With a couple of samples (at least afbc, command_buffer_usage, and [hpp_]pipeline_cache), you get this validation layer warning:
Validation Performance Warning: [ UNASSIGNED-BestPractices-vkCreateFramebuffer-attachment-should-not-be-transient ] | MessageID = 0xc80d9cbd | vkCreateFramebuffer(): Attachment 1 in VkFramebuffer uses loadOp/storeOps which need to access physical memory, but the image backing the image view has VK_IMAGE_USAGE_TRANSIENT_ATTACHMENT_BIT set. Physical memory will need to be backed lazily to this image, potentially causing stalls.
Removing the transient bit from the depth_image resolves it.
But I'm not sure, if that's the correct resolution of this warning. Maybe one should adjust the corresponding loadOp/storeOps instead.

Build tested on Win10 with VS2022. Run tested on Win10 with NVidia GPU.

General Checklist:

Please ensure the following points are checked:

  • My code follows the coding style
  • [ x I have reviewed file licenses
  • I have commented any added functions (in line with Doxygen)
  • I have commented any code that could be hard to understand
  • My changes do not add any new compiler warnings
  • My changes do not add any new validation layer errors or warnings
  • I have used existing framework/helper functions where possible
  • My changes do not add any regressions
  • I have tested every sample to ensure everything runs correctly
  • This PR describes the scope and expected impact of the changes I am making

Note: The Samples CI runs a number of checks including:

  • I have updated the header Copyright to reflect the current year (CI build will fail if Copyright is out of date)
  • My changes build on Windows, Linux, macOS and Android. Otherwise I have documented any exceptions

Sample Checklist

If your PR contains a new or modified sample, these further checks must be carried out in addition to the General Checklist:

  • I have tested the sample on at least one compliant Vulkan implementation
  • If the sample is vendor-specific, I have tagged it appropriately
  • I have stated on what implementation the sample has been tested so that others can test on different implementations and platforms
  • Any dependent assets have been merged and published in downstream modules
  • For new samples, I have added a paragraph with a summary to the appropriate chapter in the samples readme
  • For new samples, I have added a tutorial README.md file to guide users through what they need to know to implement code using this feature. For example, see conditional_rendering
  • For new samples, I have added a link to the Antora navigation so that the sample will be listed at the Vulkan documentation site

@asuessenbach asuessenbach marked this pull request as ready for review January 17, 2024 13:09
@asuessenbach asuessenbach requested a review from a team January 17, 2024 13:09
Copy link
Contributor

@gary-sweet gary-sweet left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

If you remove the transient bit generically like this then you prevent any sample from ever being able to use a transient depth buffer. This will definitely hurt performance on tilers.

I guess you would need to override the bit for particular samples only.

@JoseEmilio-ARM
Copy link
Collaborator

JoseEmilio-ARM commented Apr 8, 2024

HI @asuessenbach , I tried the samples listed above with a similar setup, and cannot reproduce the error messages (get an assertion for command_buffer_usage though...). I'm using SDK 1.3.275.

Update: apologies, had not enabled Best Practice warnings, can see it now and will try to fix it

@JoseEmilio-ARM
Copy link
Collaborator

HI @asuessenbach , what do you think of this solution?

diff --git a/framework/rendering/render_pipeline.cpp b/framework/rendering/render_pipeline.cpp
index 49085b81..3bb6842e 100644
--- a/framework/rendering/render_pipeline.cpp
+++ b/framework/rendering/render_pipeline.cpp
@@ -34,6 +34,14 @@ RenderPipeline::RenderPipeline(std::vector<std::unique_ptr<Subpass>> &&subpasses
 {
        prepare();

+       // Default load/store for swapchain
+       load_store[0].load_op  = VK_ATTACHMENT_LOAD_OP_CLEAR;
+       load_store[0].store_op = VK_ATTACHMENT_STORE_OP_STORE;
+
+       // Default load/store for depth attachment
+       load_store[1].load_op  = VK_ATTACHMENT_LOAD_OP_CLEAR;
+       load_store[1].store_op = VK_ATTACHMENT_STORE_OP_DONT_CARE;
+
        // Default clear value
        clear_value[0].color        = {0.0f, 0.0f, 0.0f, 1.0f};
        clear_value[1].depthStencil = {0.0f, ~0U};

I tested it and it fixes the validation messages on my side, without risking making the depth attachment non-transient by default.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

None yet

3 participants