On Tue, May 06, 2014 at 03:32:06PM +0200, Jérémie Courrèges-Anglas wrote: > Denis Fondras <[email protected]> writes: > > >> By the OS, which cleans up after the process exits. If it wasn't that > >> way, we'd all have a much shorter uptime... > >> > > > > Thank you Jérémie :) > > I had not considered it as I can see > > > > ... > > free(ibuf_rde); > > ... > > free(ibuf_main); > > ... > > > > at the end of session_main() in session.c. > > Maybe you should consider this as "best effort". It's a good thing to > clean up at the end of a function in an executable, even in error cases > where the end result is exit(1) / err() / fatal(), because code gets > copied and used in different contexts (where cleaning up could be > necessary).
Bad advice, you should not cleanup before err() or fatal() ever. It could very well be that you hit a bug that trashed your heap and calling free() will not solve anything. Doing cleanup before exit is only used to check for memleaks with simple tools. > But from discussions on tech@ a few months ago, you'll see that there is > no real interest in amending those non-problems if no true audit is > done to track down all occurrences in a particular piece of software. > > That said, thid diff works fine on a simple test setup. > > Index: session.c > =================================================================== > RCS file: /cvs/src/usr.sbin/bgpd/session.c,v > retrieving revision 1.334 > diff -u -p -p -u -r1.334 session.c > --- session.c 22 Jan 2014 04:08:08 -0000 1.334 > +++ session.c 6 May 2014 13:21:26 -0000 > @@ -599,6 +599,7 @@ session_main(int pipe_m2s[2], int pipe_s > free(la); > } > free(conf->listen_addrs); > + free(conf); > free(peer_l); > free(mrt_l); > free(pfd); > > -- > jca | PGP : 0x1524E7EE / 5135 92C1 AD36 5293 2BDF DDCC 0DFA 74AE 1524 E7EE > -- :wq Claudio
