Hi, Guus Sliepen schrieb: > This is necessary because some architectures impose alignment restrictions > that > would cause tinc to misbehave or segfault if you cast a random char * pointer > to a mac_t *.
I wouldn't have ever imagined, thanks :)
>
>> Running gprof it reports no time spent but only function calls, which hints
>> that all time is lost in the kernel.
>
> Which function call is the biggest user on your alix?
gprof returns:
Flat profile:
Each sample counts as 0.01 seconds.
no time accumulated
self total
calls Ts/call Ts/call name
1810148 0.00 0.00 sockaddrcmp
1809670 0.00 0.00 node_udp_compare
1675175 0.00 0.00 lookup_subnet_mac
1326891 0.00 0.00 send_packet
1271887 0.00 0.00 route
1198798 0.00 0.00 clamp_mss
1174688 0.00 0.00 get_expired_event
891582 0.00 0.00 avl_search_closest_node
884669 0.00 0.00 avl_search_node
884122 0.00 0.00 avl_search
868745 0.00 0.00 sockaddrunmap
868744 0.00 0.00 handle_incoming_vpn_data
868744 0.00 0.00 lookup_node_udp
868599 0.00 0.00 receive_packet
868599 0.00 0.00 write_packet
458516 0.00 0.00 send_udppacket
403288 0.00 0.00 read_packet
113930 0.00 0.00 send_meta
90005 0.00 0.00 past_request_compare
73089 0.00 0.00 broadcast_packet
58748 0.00 0.00 subnet_compare
55905 0.00 0.00 flush_meta
55835 0.00 0.00 send_request
54992 0.00 0.00 send_tcppacket
14620 0.00 0.00 node_compare
11997 0.00 0.00 avl_rebalance
...
Thanks,
Michael Braun
signature.asc
Description: OpenPGP digital signature
_______________________________________________ tinc mailing list [email protected] http://www.tinc-vpn.org/cgi-bin/mailman/listinfo/tinc
