#!/usr/bin/perl

# https://perlmonks.org/?node_id=1228906

use strict;
use warnings;

@ARGV or @ARGV = qw( tst1 tst2 ); # FIXME for testing only
@ARGV == 2 or die "usage: $0 filename1 filename2\n";

my %lines;

for my $file ( @ARGV )
  {
  my $segment;
  open my $fh, '<', $file or die "$! opening $file";
  while( <$fh> )
    {
    if( /^\w/ )
      {
      $segment = $_;
      }
    elsif( $segment && /\w/ )
      {
      $lines{"$segment$_" } .= "\t$file";
      }
    }
  close $fh;
  }
my $segment;
for( sort grep $lines{$_} =~ tr/\t// == 1, keys %lines )
  {
  if( $segment && /^\Q$segment\E(.+)/ )
    {
    print "$1$lines{$_}\n";
    }
  elsif( /^(.*\n)/ )
    {
    $segment = $1;
    print s/(?=\n$)/$lines{$_}/r;
    }
  }
[download]

Outputs:

lodv
     EJHFG      tst1
sdmd
     IJFH       tst2
     LAKJSK     tst2
[download]