discover VK_KHR_vulkan_memory_model

[nihui]

No description provided.

[Copilot]

Pull Request Overview

Adds support for the VK_KHR_vulkan_memory_model extension by defining its structures and macros, integrating detection in GpuInfo, enabling it during device creation, and appending its features in SPIR-V compilation.

• Introduce extension definitions in vulkan_header_fix.h
• Expose support/query methods in GpuInfo (gpu.h/gpu.cpp)
• Enable the extension in VulkanDevice and SPIR-V feature appending in gpu.cpp

Reviewed Changes

Copilot reviewed 3 out of 3 changed files in this pull request and generated no comments.

File	Description
src/vulkan_header_fix.h	Added VK_KHR_vulkan_memory_model macro and feature struct
src/gpu.h	Declared support/query methods for the new extension
src/gpu.cpp	Initialized flags, chained feature queries, enabled extension, and appended SPIR-V features

Comments suppressed due to low confidence (3)

src/vulkan_header_fix.h:1640

• Consider adding the standard VK_KHR_VULKAN_MEMORY_MODEL_SPEC_VERSION and VK_KHR_VULKAN_MEMORY_MODEL_EXTENSION_NAME macros to fully match the official Vulkan extension headers.

#define VK_KHR_vulkan_memory_model                                         1

src/gpu.h:342

• [nitpick] You may want to add a brief Doxygen or comment describing what this method checks and what its return value represents to keep the API docs consistent.

    int support_VK_KHR_vulkan_memory_model() const;

src/gpu.cpp:1859

• There aren't any unit tests covering this new support flag. Consider adding tests that simulate devices with and without VK_KHR_vulkan_memory_model to validate the flag and query methods.

int GpuInfo::support_VK_KHR_vulkan_memory_model() const

[codecov-commenter]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 95.79%. Comparing base (6e3052a) to head (c2c23c2).
Report is 1 commits behind head on master.

Additional details and impacted files

@@            Coverage Diff             @@
##           master    #6121      +/-   ##
==========================================
+ Coverage   95.70%   95.79%   +0.09%     
==========================================
  Files         829      829              
  Lines      270340   270769     +429     
==========================================
+ Hits       258724   259392     +668     
+ Misses      11616    11377     -239     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[github-actions]

The binary size change of libncnn.so (bytes)

architecture	base size	pr size	difference
x86_64	16488168	16488320	+152 ⚠️
armhf	7368772	7368864	+92 ⚠️
aarch64	10779040	10779000	-40 😘