ipv4: reduce percpu needs for icmpmsg mibs

Reading /proc/net/snmp on a machine with a lot of cpus is very expensive
(can be ~88000 us).

This is because ICMPMSG MIB uses 4096 bytes per cpu, and folding values
for all possible cpus can read 16 Mbytes of memory.

ICMP messages are not considered as fast path on a typical server, and
eventually few cpus handle them anyway. We can afford an atomic
operation instead of using percpu data.

This saves 4096 bytes per cpu and per network namespace.

Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This commit is contained in:
Eric Dumazet 2011-11-08 13:04:43 +00:00 committed by David S. Miller
parent e56c57d0d3
commit acb32ba3de
5 changed files with 12 additions and 13 deletions

View file

@ -31,8 +31,8 @@ struct icmp_err {
extern const struct icmp_err icmp_err_convert[];
#define ICMP_INC_STATS(net, field) SNMP_INC_STATS((net)->mib.icmp_statistics, field)
#define ICMP_INC_STATS_BH(net, field) SNMP_INC_STATS_BH((net)->mib.icmp_statistics, field)
#define ICMPMSGOUT_INC_STATS(net, field) SNMP_INC_STATS((net)->mib.icmpmsg_statistics, field+256)
#define ICMPMSGIN_INC_STATS_BH(net, field) SNMP_INC_STATS_BH((net)->mib.icmpmsg_statistics, field)
#define ICMPMSGOUT_INC_STATS(net, field) SNMP_INC_STATS_ATOMIC_LONG((net)->mib.icmpmsg_statistics, field+256)
#define ICMPMSGIN_INC_STATS_BH(net, field) SNMP_INC_STATS_ATOMIC_LONG((net)->mib.icmpmsg_statistics, field)
struct dst_entry;
struct net_proto_family;