Doesn't the pmc MAP_IN event report incomplete information?

[Ryan Stone]

The pmc MAP_IN event returns data in the following structure:

struct pmclog_ev_map_in {
pid_t pl_pid;
uintfptr_t pl_start;
char pl_pathname[PATH_MAX];
};

It seems to me that this information is incomplete. Don't we need the
offset into the file at which the mapping started, along with the size
of the mapping?

I looked at how pmcstat handles this event, and it looks like it
assumes that the entire .text section is being mapped. I'm sure that
this is correct in 99.99% of all cases, but wouldn't it be easier to
have the event contain the size and offset?