Unable to run example

[justinmoon]

I'm on arch linux:

$ cargo run --example gallery
    Finished dev [unoptimized + debuginfo] target(s) in 0.10s
     Running `target/debug/examples/gallery`
MESA-INTEL: warning: Performance support disabled, consider sysctl dev.i915.perf_stream_paranoid=0

[2021-09-23T05:00:46Z ERROR wgpu_core::validation] Unexpected varying type: Array { base: [1], size: Constant([5]), stride: 4 }
[2021-09-23T05:00:46Z ERROR wgpu_core::validation] Unexpected varying type: Array { base: [1], size: Constant([5]), stride: 4 }
[2021-09-23T05:00:46Z ERROR wgpu_core::validation] Unexpected varying type: Array { base: [1], size: Constant([5]), stride: 4 }
[2021-09-23T05:00:46Z ERROR wgpu_core::validation] Unexpected varying type: Array { base: [1], size: Constant([5]), stride: 4 }
[2021-09-23T05:00:46Z ERROR wgpu_core::validation] Unexpected varying type: Array { base: [1], size: Constant([5]), stride: 4 }
[2021-09-23T05:00:46Z ERROR wgpu_core::validation] Unexpected varying type: Array { base: [1], size: Constant([5]), stride: 4 }
[2021-09-23T05:00:46Z ERROR wgpu_core::validation] Unexpected varying type: Array { base: [1], size: Constant([5]), stride: 4 }
[2021-09-23T05:00:46Z ERROR wgpu_core::validation] Unexpected varying type: Array { base: [1], size: Constant([5]), stride: 4 }
[2021-09-23T05:00:46Z ERROR wgpu_core::validation] Unexpected varying type: Array { base: [4], size: Constant([6]), stride: 4 }
[2021-09-23T05:00:46Z ERROR wgpu_core::validation] Unexpected varying type: Array { base: [4], size: Constant([6]), stride: 4 }
[2021-09-23T05:00:46Z ERROR wgpu_core::validation] Unexpected varying type: Array { base: [4], size: Constant([6]), stride: 4 }
[2021-09-23T05:00:46Z ERROR wgpu_core::validation] Unexpected varying type: Array { base: [4], size: Constant([6]), stride: 4 }
MESA-INTEL: error: ../mesa-21.2.1/src/intel/vulkan/anv_batch_chain.c:2053: execbuf2 failed: Invalid argument (VK_ERROR_DEVICE_LOST)
thread 'main' panicked at 'Error in Queue::submit: parent device is lost', /home/justin/.cargo/registry/src/github.com-1ecc6299db9ec823/wgpu-0.10.1/src/backend/direct.rs:186:9
note: run with `RUST_BACKTRACE=1` environment variable to display a backtrace
[2021-09-23T05:00:46Z ERROR wgpu_core::device] failed to wait for the device: Lost

Not sure if I installed dependencies correctly ...

[dhardy]

So you're using Intel graphics. Which CPU generation? Can you try running some wgpu examples?

[justinmoon]

Yes Intel. Skylake. I get the same error when I try to run wgpu's "mipmap" example. But all the other ones I tried worked.

$ cargo run --example mipmap
   Compiling wgpu v0.10.1 (/home/justin/code/clones/wgpu/wgpu)
    Finished dev [unoptimized + debuginfo] target(s) in 11.66s
     Running `/home/justin/code/clones/wgpu/target/debug/examples/mipmap`
MESA-INTEL: warning: Performance support disabled, consider sysctl dev.i915.perf_stream_paranoid=0

Using Intel(R) HD Graphics 620 (KBL GT2) (Vulkan)
Generating mip level 1 took 69.833 μs and called the fragment shader 65536 times
Generating mip level 2 took 41.333 μs and called the fragment shader 16384 times
Generating mip level 3 took 16.083 μs and called the fragment shader 4096 times
Generating mip level 4 took 13.333 μs and called the fragment shader 1024 times
Generating mip level 5 took 13.083 μs and called the fragment shader 256 times
Generating mip level 6 took 13.417 μs and called the fragment shader 64 times
Generating mip level 7 took 14.333 μs and called the fragment shader 16 times
Generating mip level 8 took 14.417 μs and called the fragment shader 4 times
MESA-INTEL: error: ../mesa-21.2.1/src/intel/vulkan/anv_batch_chain.c:2053: execbuf2 failed: Invalid argument (VK_ERROR_DEVICE_LOST)
thread 'main' panicked at 'Error in Queue::submit: parent device is lost', wgpu/src/backend/direct.rs:197:9
note: run with `RUST_BACKTRACE=1` environment variable to display a backtrace
[2021-09-23T16:24:10Z ERROR wgpu_core::device] failed to wait for the device: Lost

[dhardy]

I just tested on my Intel Haswell laptop.. and it doesn't draw (though no obvious error). Will have to investigate further. You might also consider reporting an issue to WGPU. See here if you wish to investigate further yourself.

[dhardy]

I can confirm that the update to WGPU 0.10 is what broke rendering on the Haswell laptop. I see WGPU 0.11 is out now, so testing that is next..

[dhardy]

WGPU v0.11 does not work on the Haswell laptop either, but it introduces an experimental fallback option which does (with some issues). This can be enabled with (in latest master):

KAS_BACKENDS=fallback cargo run --example gallery

@justinmoon it would be worth also testing without the fallback option now that master is updated to WGPU v0.11.

[justinmoon]

$ cargo run --example gallery
...
MESA-INTEL: warning: Performance support disabled, consider sysctl dev.i915.perf_stream_paranoid=0

MESA-INTEL: error: ../mesa-21.2.2/src/intel/vulkan/anv_batch_chain.c:2053: execbuf2 failed: Invalid argument (VK_ERROR_DEVICE_LOST)
thread 'main' panicked at 'Error in Queue::submit: parent device is lost', /home/justin/.cargo/registry/src/github.com-1ecc6299db9ec823/wgpu-0.11.0/src/backend/direct.rs:204:9
note: run with `RUST_BACKTRACE=1` environment variable to display a backtrace
[2021-10-14T17:22:14Z ERROR wgpu_core::device] failed to wait for the device: Lost

$ KAS_BACKENDS=fallback  cargo run --example gallery
    Finished dev [unoptimized + debuginfo] target(s) in 0.12s
     Running `target/debug/examples/gallery`
MESA-INTEL: warning: Performance support disabled, consider sysctl dev.i915.perf_stream_paranoid=0

Error: NoAdapter

[dhardy]

consider sysctl dev.i915.perf_stream_paranoid=0

Did you try this?

Currently KAS does not support pure-CPU rendering (#33). (It may happen eventually, given more motivation/resources.)

[dhardy]

Closing because this largely concerns WGPU and I do not have the capacity to work on this directly.

Instead non-GPU rendering (#287) should solve this (eventually).