For the complete documentation index, see llms.txt. Markdown versions of all pages are available by appending .md to any URL (e.g. /get-started.md).

Mojo module

device_query

Provides device query utilities for communication primitives.

`comptime` values

comptime GB = 1073741824

comptime KB = 1024

comptime MB = 1048576

DefaultCommTuningConfig: Parameters: ngpus: Number of GPUs for running allreduce. num_bytes: Total number of input bytes supported by the config. sm_version: SM version (as string). num_blocks: Number of thread blocks for running allreduce.

CommTuningConfig: Tuning-table entry for a multi-GPU communication collective.

dispatch_select_comm_config: This function searches for tuning configs with matching sm_version and ngpus. If such configs are found, then the search continues for finding the config x where num_bytes <= x.num_bytes.
get_sm_version: Returns the SM version string for the current compile target.