rightkit-gpu 0.1.0

Shared wgpu device context for Right Suite apps: per-OS backend policy, a crash sentinel around device creation, adapter pick and limits, an exit gate for in-flight GPU work, and a device-memory budget that adapts to out-of-memory.
Documentation

rightkit-gpu

The shared wgpu device context for Right Suite apps: per-OS backend policy, a crash sentinel around device creation, adapter pick and limits, an exit gate for in-flight GPU work, and a device-memory budget that halves on out-of-memory. Native only (wgpu 30).

Use it for the device and its lifecycle in Ember and CutRight. Do not use it for kernels, buffer pools or render paths: those stay in the app.

use rightkit_gpu::{backend, exit, ContextOptions, GpuContext};

backend::set_init_marker(Some(marker_path));
if let Some(backends) = backend::compute_backends("EMBER_GPU_BACKEND") {
    let options = ContextOptions::new(backends, "ember");
    let gpu = GpuContext::create(&options, |reason| { /* stop using the GPU */ })?;
    // build kernels under backend::with_marker("kernel compilation", || ...)
}
// at exit:
exit::begin_shutdown();
exit::wait_idle(std::time::Duration::from_secs(5));

Every entry into the device goes through exit::enter(). See CHANGELOG.md for the wgpu pin decision and NOTICE for the upstream attribution.