Robin Kertels

Thanks for the heads up. I'll take a look once the PRs that block this land. EDIT: I checked and it looks like I already got that right.

I've hooked it up to the "Backend multithreading option". Before, that option used to control whether or not we submit our Vulkan command buffers on a separate thread. There's no...

I rebased this but it's gonna need a lot more testing. Unfortunately it's not even beneficial across the board as it can increase the submission latency of EFB work which...

D3D12 also has something very similar to VK_KHR_present_wait as far as I know. So it would make sense to design something in VideoCommon.

Which backend fails to compile that shader? I'm pretty sure your fix is incorrect. The GLSL spec says: (emphasis mine) > Any expression whose type evaluates to a Boolean can...

For a future progress report: Old 2020 Qualcomm proprietary driver:  Turnip 23.1: 

@JMC47 OpenGL is not impacted at all. It only replaces the Vulkan driver.

Go home FIFOCI, you're drunk.

@AlexS2412 There's nothing preventing this from working on Mali GPUs aside from the fact that nobody has written an open source Vulkan driver for those GPUs. Besides that, Mali GPUs...

This was useful to figure out that Adreno 615 & 682 hit DEVICE_LOST when encountering the `discard` we emit for the alpha test.