Vecvec

> Ah, you mean the D3DMetal.framework but the source code for that is not public, binary distribution only. I'd assumed that apple might make some way of showing what each...

@Lichtso are you able to use a debugger on the tests? (It looks to be possible at least under `cargo test`) If so, could you see what acceleration structure sizes...

@SpeedCrash100, wgpu claims it can't find the vulkan validation layers, do you have them installed? It might catch something if we're doing it wrong.

@SpeedCrash100, in the shader from `prevent_invalid_ray_query_calls` there are multiple blocks, could you try removing them (maybe in multiple orders?) and see if some configurations work? It could also be useful...

@SpeedCrash100, no problem. I don't expect people to immediately respond (and I was away during that time anyway). It's useful to have found the cause. Does it still segfault with...

> It crashes same way, however, if rayQueryProceed line removed validation error will appear I'm not sure if that's expected, but its a wgpu validation error (and so internal). >...

I have opened the bug report as https://gitlab.freedesktop.org/mesa/mesa/-/issues/14421

That's not completely surprising, this is in an optimization pass. I'd assumed this was a RADV specific optimization, but guess not.

Could it be one of the old segfaults from llvmpipe?

Windows Intel UHD 770 + Nvidia RTX 3050 commit [db907d90b1c0397098a849d92babb5e9c4a81af7](https://github.com/gfx-rs/wgpu/commit/db907d90b1c0397098a849d92babb5e9c4a81af7) [wgpu-info.txt](https://github.com/user-attachments/files/23912909/wgpu-info.txt) [tests.txt](https://github.com/user-attachments/files/23912910/tests.txt) ````cmd FAIL [ 0.818s] wgpu-examples [Executed] [Vulkan/NVIDIA GeForce RTX 3050/0] bunnymark FAIL [ 0.992s] wgpu-test::wgpu-gpu [Executed] [Dx12/Intel(R) UHD...