Gary Sweet

This all seems to work ok on my platform now, but I notice there is a clang format check failure so I can't approve it yet.

> Would you mind to resolve this issue like it's done in #833? Done. That's a much nicer solution, thanks.

This is an Android-only change, so I can't test it I'm afraid.

I'm surprised this even ran for me since we don't support some of the features being used here. I suspect you may be missing some checks. See https://vulkan.gpuinfo.org/displayreport.php?id=27052 for the...

> @gary-sweet About testing this plugin, I think easiest/quickest way would be to use patch that I attached to this comment, It is basically revert of commits that removed plugin...

I'm seeing a couple of build problems. The first is a redefinition of LOGE: ``` [100%] Building CXX object app/CMakeFiles/vulkan_samples.dir/main.cpp.o In file included from Vulkan-Samples/components/filesystem/include/filesystem/filesystem.hpp:26, from Vulkan-Samples/app/main.cpp:23: Vulkan-Samples/components/core/include/core/util/logging.hpp:28: warning: "LOGE"...

The second issue is that `std::filesystem` isn't available before gcc8 unless you explicitly include `stdc++fs` as a link library. We've had to work around that a number of times using...

> @gary-sweet I created temporary PR like you requested - #953 Also there is one fail and canceled checks here that looks like infrastructure issue, is there a way to...

I don't have a driver that supports swapchain_maintenance1 yet, so all I can do is confirm that this doesn't break the existing functionality for me.

> Unless the IHVs are targeting some kind of bespoke CPU / OS, I think it's unlikely to be a problem. We have our own compiler toolchains for cross-compiling for...