nntp: lazily allocate and stash rbuf

Allocating a per-client buffer up front is unnecessary and
wastes a hash slot.  For the majority of (non-malicious)
clients, we won't need to store rbuf in a long-lived object
associated with a client socket at all.

This saves around 10M on 64-bit with 20K connected-but-idle
clients.

1 files changed, 7 insertions, 3 deletions

diff --git a/lib/PublicInbox/NNTP.pm b/lib/PublicInbox/NNTP.pm
index 6acfcc1b..10a2e158 100644
--- a/lib/PublicInbox/NNTP.pm
+++ b/lib/PublicInbox/NNTP.pm
@@ -107,7 +107,6 @@ sub new ($$$) {
         $self->{nntpd} = $nntpd;
         push @$wbuf, \&greet;
         $self->{wbuf} = $wbuf;
-        $self->{rbuf} = '';
         update_idle_time($self);
         $expt ||= PublicInbox::EvCleanup::later(*expire_old);
         $self;
@@ -964,7 +963,7 @@ sub event_step {
         # otherwise we can be buffering infinitely w/o backpressure
 
         use constant LINE_MAX => 512; # RFC 977 section 2.3
-        my $rbuf = \($self->{rbuf});
+        my $rbuf = $self->{rbuf} // (\(my $x = ''));
         my $r = 1;
 
         if (index($$rbuf, "\n") < 0) {
@@ -984,6 +983,11 @@ sub event_step {
         return $self->close if $r < 0;
         my $len = bytes::length($$rbuf);
         return $self->close if ($len >= LINE_MAX);
+        if ($len) {
+                $self->{rbuf} = $rbuf;
+        } else {
+                delete $self->{rbuf};
+        }
         update_idle_time($self);
 
         # maybe there's more pipelined data, or we'll have
@@ -1002,7 +1006,7 @@ sub not_idle_long ($$) {
 # for graceful shutdown in PublicInbox::Daemon:
 sub busy {
         my ($self, $now) = @_;
-        ($self->{rbuf} ne '' || $self->{wbuf} || not_idle_long($self, $now));
+        ($self->{rbuf} || $self->{wbuf} || not_idle_long($self, $now));
 }
 
 1;