syscount - Summarize syscall counts and latencies. syscount [-h] [-p PID] [-t TID] [-c PPID] [-i INTERVAL] [-d DURATION] [-T TOP] [-x] [-e ERRNO] [-L] [-m] [-P] [-l] [--syscall SYSCALL] This tool traces syscall entry and exit tracepoints and summarizes either the number of syscalls of each type, or the number of syscalls per process. It can also collect latency (invocation time) for each syscall or each process. Since this uses BPF, only the root user can use this tool. CONFIG_BPF and bcc. Linux 4.7+ is required to attach a BPF program to the raw_syscalls:sys_{enter,exit} tracepoints, used by this tool.

-h Print usage message.

-p PID Trace only this process.

-t TID Trace only this thread.

-c PPID Trace only child of this pid.

-i INTERVAL Print the summary at the specified interval (in seconds).

-d DURATION Total duration of trace (in seconds).

-T TOP Print only this many entries. Default: 10.

-x Trace only failed syscalls (i.e., the return value from the syscall was < 0).

-e ERRNO Trace only syscalls that failed with that error (e.g. -e EPERM or -e 1).

-m Display times in milliseconds. Default: microseconds.

-P Summarize by process and not by syscall.

-l List the syscalls recognized by the tool (hard-coded list). Syscalls beyond this list will still be displayed, as "[unknown: nnn]" where nnn is the syscall number.

--syscall SYSCALL Trace this syscall only (use option -l to get all recognized syscalls).

Summarize all syscalls by syscall: # syscount

Summarize all syscalls by process: # syscount -P

Summarize only failed syscalls: # syscount -x

Summarize only syscalls that failed with EPERM: # syscount -e EPERM

Trace PID 181 only: # syscount -p 181

Summarize syscalls counts and latencies: # syscount -L

PID Process ID

COMM Process name

SYSCALL Syscall name, or "[unknown: nnn]" for syscalls that aren't recognized

COUNT The number of events

TIME The total elapsed time (in us or ms)

For most applications, the overhead should be manageable if they perform 1000's or even 10,000's of syscalls per second. For higher rates, the overhead may become considerable. For example, tracing a loop of 4 million calls to geteuid(), slows it down by 1.85x when tracing only syscall counts, and slows it down by more than 5x when tracing syscall counts and latencies. However, this represents a rate of >3.5 million syscalls per second, which should not be typical. This is from bcc. https://github.com/iovisor/bcc

Also look in the bcc distribution for a companion _examples.txt file containing example usage, output, and commentary for this tool.

Linux Unstable - in development. Sasha Goldshtein, Rocky Xing funccount(8), ucalls(8), argdist(8), trace(8), funclatency(8)