Hi,

fist of all, thank you for your blog and your perl tool. It's great to have it.

I'm also having several systems with large amount of sysv shmem, where classical "free" fails. What me puzzels is, that in free.pl you calculate:

my $minus_caches_data = {
        'used' => $mem_row_data->{'used'} - $mem_row_data->{'caches'},
        'free' => $mem_row_data->{'free'} + $mem_row_data->{'caches'},
};

of course your $mem_row_data->{'caches'} is far diffrent from classical "free", I wonder why you do not use 'free' => $avail from Linux::MemAvailable::calculate().

comparing free.pl's free+caches with Linux::MemAvailable::calculate()'s $avail ... there seems a (minor) difference (about 100MB sometimes).

In the meantime the procps tools (free, top, ...) have adopted. I've attached a screen shoot of them. Maybe your free.pl can match the new output (for older $enterprice distros that do not adopt so fast).

On Ubuntu 16.04 (kernel 4.0.0.36) the "avail" value differs from system's by about 90M.

~/svn/psutil$ ./free.pl  | grep avail
  -/+ avail           7135660    9189988
~/svn/psutil$ cat /proc/meminfo |grep Avail
MemAvailable:    9091500 kB

At first I thought it was the Perl interpreter overhead but I tried running "cat /proc/meminfo |grep Avail" from the shell script as a subprocess and I get a similar result.
Any idea?