Experimental Linux kernel fast-path patches for SRAM-based AI inference servers, targeting io_uring submission, registered buffers, CQ polling, wakeup attribution, and completion latency.
benchmarking latency linux-kernel sram low-latency performance-analysis ebpf operating-systems ftrace systems-performance bpftrace io-uring async-io kernel-development kernel-tracing ai-inference ai-infrastructure inference-systems kernel-fastpath
-
Updated
May 3, 2026 - C