net: proc: speedup /proc/net/netstat
Subsystems (2)
PATCH 1/1 net: proc: speedup /proc/net/netstat
From: Eric Dumazet <[email protected]>
Use cache friendly helpers to better use cpu caches while reading /proc/net/netstat
Tested on a platform with 256 threads (AMD Rome)
Before: 305 usec spent in netstat_seq_show()
After: 130 usec spent in netstat_seq_show()
Signed-off-by Eric Dumazet <[email protected]>
--- net/ipv4/proc.c | 48 +++++++++++++++++++++++++++++++++++------------- 1 file changed, 35 insertions(+), 13 deletions(-)
net/ipv4/proc.c +35 -13
@@ -464,30 +464,52 @@ static int snmp_seq_show(struct seq_file *seq, void *v)*/ static int netstat_seq_show (struct seq_file *seq,void *v) { -int i; +const int ip_cnt =ARRAY_SIZE (snmp4_ipextstats_list) -1 ; +const int tcp_cnt =ARRAY_SIZE (snmp4_net_list) -1 ;struct net *net = seq->private; +unsigned long *buff; +int i;seq_puts (seq,"TcpExt:" ); -for (i =0 ; snmp4_net_list[i].name; i++) +for (i =0 ; i < tcp_cnt; i++)seq_printf (seq," %s" , snmp4_net_list[i].name);seq_puts (seq,"\nTcpExt:" ); -for (i =0 ; snmp4_net_list[i].name; i++) -seq_printf (seq," %lu" , -snmp_fold_field (net->mib.net_statistics, - snmp4_net_list[i].entry)); - + buff =kzalloc (max (tcp_cnt *sizeof (long ), ip_cnt *sizeof (u64)), +GFP_KERNEL ); +if (buff) { +snmp_get_cpu_field_batch (buff, snmp4_net_list, + net->mib.net_statistics); +for (i =0 ; i < tcp_cnt; i++) +seq_printf (seq," %lu" , buff[i]); + }else { +for (i =0 ; i < tcp_cnt; i++) +seq_printf (seq," %lu" , +snmp_fold_field (net->mib.net_statistics, + snmp4_net_list[i].entry)); + }seq_puts (seq,"\nIpExt:" ); -for (i =0 ; snmp4_ipextstats_list[i].name; i++) +for (i =0 ; i < ip_cnt; i++)seq_printf (seq," %s" , snmp4_ipextstats_list[i].name);seq_puts (seq,"\nIpExt:" ); -for (i =0 ; snmp4_ipextstats_list[i].name; i++) -seq_printf (seq," %llu" , -snmp_fold_field64 (net->mib.ip_statistics, - snmp4_ipextstats_list[i].entry, - offsetof(struct ipstats_mib , syncp))); +if (buff) { +u64 *buff64 = (u64 *)buff; +memset (buff64,0 , ip_cnt *sizeof (u64)); +snmp_get_cpu_field64_batch (buff64, snmp4_ipextstats_list, + net->mib.ip_statistics, + offsetof(struct ipstats_mib , syncp)); +for (i =0 ; i < ip_cnt; i++) +seq_printf (seq," %llu" , buff64[i]); + }else { +for (i =0 ; i < ip_cnt; i++) +seq_printf (seq," %llu" , +snmp_fold_field64 (net->mib.ip_statistics, + snmp4_ipextstats_list[i].entry, + offsetof(struct ipstats_mib , syncp))); + } +kfree (buff);seq_putc (seq,'\n' );mptcp_seq_show (seq);return 0 ;
Signature · 1 line
2.30.0.280.ga3ce27912f-goog
Jakub Kicinski
maintainerQuoted 11 lines · On Thu, 28 Jan 2021 08:21:45 -0800 Eric Dumazet wrote:
From: Eric Dumazet <[email protected]>
Use cache friendly helpers to better use cpu caches while reading
/proc/net/netstatTested on a platform with 256 threads (AMD Rome)
Before: 305 usec spent in
netstat_seq_show()
After: 130 usec spent innetstat_seq_show()Signed-off-by Eric Dumazet <[email protected]>
The bot just doesn't like to reply to you :) If it happens again I'll poke Konstantin. Maybe third time is the charm.
Anyway, applied a few hours ago, thank you!