On wgpu 0.3.0 + winit 0.20.0-alpha3 i'm seeing what appear to be new gpu errors.
The application loads successfully, a single frame is rendered, and then I see:
2019-09-09 22:46:02.894432-0700 test[13776:6361225] Execution of the command buffer was aborted due to an error during execution. Caused GPU Timeout Error (IOAF code 2)
2019-09-09 22:46:02.894697-0700 test[13776:6361225] Execution of the command buffer was aborted due to an error during execution. Discarded (victim of GPU error/recovery) (IOAF code 5)
It will sit in this state until I touch the screen, then winit panics with a dpi assertion:
thread '<unnamed>' panicked at 'assertion failed: validate_hidpi_factor(dpi_factor)', /Users/ataraxis/.cargo/registry/src/github.com-1ecc6299db9ec823/winit-0.20.0-alpha3/src/dpi.rs:113:9
I think this is a red herring; I've printed the dpi I get from winit (3.0), and it seems ok. I suspect the window has already died and dpi returns some random value, which triggers the assert? Or maybe not, I'm not totally sure.
On wgpu
0.3.0+ winit0.20.0-alpha3i'm seeing what appear to be new gpu errors.The application loads successfully, a single frame is rendered, and then I see:
It will sit in this state until I touch the screen, then winit panics with a dpi assertion:
I think this is a red herring; I've printed the dpi I get from winit (
3.0), and it seems ok. I suspect the window has already died and dpi returns some random value, which triggers the assert? Or maybe not, I'm not totally sure.