Rua
Rua
Yes, pretty much. In the image, track 0 contains sectors whose track number is 1. The GW then writes those to physical track 1, and not to physical track 0...
Ah, I was using Greaseweazle tools 1.13 still. I'll see if I can still reproduce it with 1.16.3.
If `UPDATE_AFTER_BIND` is enabled, then it's possible that the resources in the descriptor set, at the time the command is recorded, are not the same as those that will be...
If I'm wrong, then `GetAdjustedDestImageExtent` may need changing as it currently doesn't account for blocked-to-blocked copies if the block sizes are different.
Also may be important: https://github.com/KhronosGroup/Vulkan-Docs/issues/1807
I'm running Plasma 6.3 though.
What extensions and features are you enabling, and what validation error do you get?
The Vulkan error is not very useful to the user, though. It says that something isn't present, but not what. Vulkano's error gives more information.
If you comply with the Vulkan specification, there shouldn't be any validation errors to begin with.
After thinking about this, I wonder if it makes more sense to re-export all of `ash::vk`?