Re: [PATCH] x86/tracing: introduce enter/exit tracepoint pairs for page faults

[Junxuan Liao <ljx@cs.wisc.edu>]

On 4/14/25 5:20 PM, Steven Rostedt wrote:
> It's useful for me ;-)
> The above shows a histogram in microseconds where the buckets increase in a
> power of two. The biggest bucket is between 2^4 (16) and 2^5 (32) microseconds
> with 108790 hits.

With patch v2, I can do something similar in bpftrace too. :)

tracepoint:exceptions:page_fault_enter
/ args.user_mode == true /
{
    @start[tid] = nsecs;
}

tracepoint:exceptions:page_fault_exit
/ args.user_mode == true /
{
    @lat[tid] = hist(nsecs - @start[tid]);
}

Junxuan