Features:

Small memory usage

bustd seems to use less memory than other lean process killers, such as earlyoom:

$ ps -F -C bustd
UID          PID    PPID  C    SZ   RSS PSR STIME TTY          TIME CMD
vrmiguel  353609  187407  5   151     8   2 01:20 pts/2    00:00:00 target/x86_64-unknown-linux-musl/release/bustd -V -n

$ ps -F -C earlyoom
UID          PID    PPID  C    SZ   RSS PSR STIME TTY          TIME CMD
vrmiguel  350497    9498  0   597   688   6 01:12 pts/1    00:00:00 ./earlyoom/

Check the README.md for more info on this comparison!

Small CPU usage

bustd uses adaptive sync times between each of its memory polls and at every memory poll, checks the system memory status through the sysinfo syscall.

Other process killers usually read and parse from /proc/meminfo, which is believed to be at least an order of magnitude slower than the sysinfo syscall (source).

As sysinfo doesn't have sufficiently accurate free system memory metrics, we use pressure stall information to know precisely when the system is out of memory and prone to thrashing. More on this later.

Able to lock its memory pages

bustd uses mlockall to lock all pages mapped into its address space.

This allows bustd to not have its memory sent to swap so as to remain responsive during scenarios of stressed memory. In order to fully commit to this concept, bustd constantly reuses two buffers that amount to 150 bytes throughout its entire main loop, avoiding any new heap allocations whatsoever.

Checks for Pressure Stall Information (PSI)

The Linux kernel, since version 4.20 (and built with CONFIG_PSI=y), presents canonical new pressure metrics for memory, CPU, and IO. In the words of Facebook Incubator:

PSI stats are like barometers that provide fair warning of impending resource 
shortages, enabling you to take more proactive, granular, and nuanced steps 
when resources start becoming scarce.

More specifically, bustd checks for how long, in microseconds, processes have stalled in the last 10 seconds. By default, bustd will kill a process when processes have stalled for 25 microseconds in the last ten seconds, but this is adjustable through command-line options.

It's worth noting that, for performance reasons, the PSI figures are only checked when the sysinfo reading of free system memory are sufficiently low (15% memory left, currently).

Objective

bustd doesn't particularly aim to replace or be better than any other OOM killer out there. I'm making this project as a hobby to delve deeper into how OOM killers work and how Linux deals with this sort of thing in general.

Suggestions?

As this is quite a recent work in progress, I'm eager to know if anyone has suggestions for any features it should implement!

Repo: https://github.com/vrmiguel/bustd/