Module gpu 

Re-exports

use std::collections::HashMap; 🔒

use std::collections::HashSet; 🔒

use std::path::Path; 🔒

use std::path::PathBuf; 🔒

use anyhow::Result; 🔒

use log::debug; 🔒

use crate::cgroup::CgroupReader; 🔒

Functions

direct_gpu_processes 🔒

Keep only NVML processes whose GPUs resolve to one NUMA node.

expand_gpu_processes 🔒

Expand NVML processes to peer processes in the same exact cgroup.

expand_gpu_processes_with 🔒

fit_gpu_processes 🔒

Fit workload hints into the BPF map without partially expanding a cgroup.