rumblehhh

Hi Marek, Please see the following json stats. m_Id is 0 as it seems we only use a single block. `{ "General": { "API": "Vulkan", "apiVersion": "1.3.0", "GPU": "SwiftShader Device...

No problem! VMA_HEAVY_ASSERT didn't trigger during the application run - I checked that VmaIsPow2 was called inside VmaAlignUp so VMA_HEAVY_ASSERT was definitely enabled. The application itself is multithreaded but all...

No problem. >Maybe you are using some other options that interfere with debug margin that we are not aware of We're not explicitly enabling any other features but we will...

We have the following pattern: 1. Lots of allocations 2. Lots of frees 3. One allocation 4. Lots of frees I'm attaching: - a log with the allocations/frees - a...

Thanks for the proposed fix! Unfortunately I still see the same assert :(. If I additionally enable VMA_HEAVY_ASSERT by setting it to VMA_ASSERT I then hit the following assert virtual...

Sorry for the delay - I've been OoO for the last week. No problem glad I can help. Please see attached the attached files as requested: [before_freeing_memory_2.txt](https://github.com/GPUOpen-LibrariesAndSDKs/VulkanMemoryAllocator/files/9240901/before_freeing_memory_2.txt) [before_freeing_allocator.txt](https://github.com/GPUOpen-LibrariesAndSDKs/VulkanMemoryAllocator/files/9240900/before_freeing_allocator.txt)

Jumping on this to add that we were recently bitten by this. We use vulkaninfo's json option for checking implementation features/properties to avoid running unsupported tests/content. After upgrading to a...

@charles-lunarg thanks for the changes! I've checked out the branch and I believe this should work fine for us. Should this, or another option if the name changes, actually show...

Has there been anymore recent investigation into any of the limitations of the Mock ICD that @null77 pointed out? >A second example is mapping memory ranges. After a first map...