Awesome

WebGPU.jl

Julia bindings and native wrapper for the gfx webgpu implementation

Mapping the WGPU API

• adapter
  • create_device deviceid (adapterid, descriptor)
• bind_group
  • destroy (groupid)
• buffer
  • destroy (bufferid)
  • map_read_async (bufferid, start, size, callback, data)
  • map_write_async (bufferid, start, size, callback, data)
  • set_sub_data (bufferid, start, count, data)
  • unmap (bufferid)
• command_buffer
  • copy_buffer_to_buffer
  • copy_buffer_to_texture
  • copy_texture_to_buffer
  • copy_texture_to_texture
• command_encoder
  • begin_compute_pass (encoderid)
  • begin_render_pass (encoderid, renderpassdesc)
  • finish (encoderid)
• compute_pass
  • dispatch (passid, x, y, z)
  • end_pass (passid)
  • set_bind_group (passid, index, groupid)
  • set_pipeline (passid, pipelineid)
• create_instance instanceid ()
• device
  • create_bind_group (deviceid, desc)
  • create_bind_group_layout (deviceid, layourdesc)
  • create_buffer (deviceid, bufferdesc)
  • create_buffer_mapped (deviceid, bufferdesc, map)
  • create_command_encoder (deviceid, desc)
  • create_compute_pipeline (deviceid, desc)
  • create_pipeline_layout (deviceid, desc)
  • create_render_pipeline (deviceid, desc)
  • create_sampler (deviceid, desc)
  • create_shader ()
  • create_swap_chain
  • create_texture
  • destroy
  • get_queue
• instance
  • create_surface_from_macos_layer
  • create_surface_from_windows_hwnd
  • create_surface_from_xlib
  • get_adaptor
• queue
  • submit
• render_pass
  • draw
  • draw_indexed
  • end_pass
  • set_bind_group
  • set_index_buffer
  • set_pipeline
  • set_vertex_buffers
• swap_chain
  • get_next_texture
  • present
• texture
  • create_default_view
  • create_view
  • destroy
  • view_destroy