From d265929930e2ffafc744c0ae05fb70acd53be1ee Mon Sep 17 00:00:00 2001 From: William Tu Date: Wed, 4 May 2022 08:35:59 -0700 Subject: netfilter: nf_conncount: reduce unnecessary GC Currently nf_conncount can trigger garbage collection (GC) at multiple places. Each GC process takes a spin_lock_bh to traverse the nf_conncount_list. We found that when testing port scanning use two parallel nmap, because the number of connection increase fast, the nf_conncount_count and its subsequent call to __nf_conncount_add take too much time, causing several CPU lockup. This happens when user set the conntrack limit to +20,000, because the larger the limit, the longer the list that GC has to traverse. The patch mitigate the performance issue by avoiding unnecessary GC with a timestamp. Whenever nf_conncount has done a GC, a timestamp is updated, and beforce the next time GC is triggered, we make sure it's more than a jiffies. By doin this we can greatly reduce the CPU cycles and avoid the softirq lockup. To reproduce it in OVS, $ ovs-appctl dpctl/ct-set-limits zone=1,limit=20000 $ ovs-appctl dpctl/ct-get-limits At another machine, runs two nmap $ nmap -p1- $ nmap -p1- Signed-off-by: William Tu Co-authored-by: Yifeng Sun Reported-by: Greg Rose Suggested-by: Florian Westphal Signed-off-by: Pablo Neira Ayuso --- include/net/netfilter/nf_conntrack_count.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/net') diff --git a/include/net/netfilter/nf_conntrack_count.h b/include/net/netfilter/nf_conntrack_count.h index 9645b47fa7e4..e227d997fc71 100644 --- a/include/net/netfilter/nf_conntrack_count.h +++ b/include/net/netfilter/nf_conntrack_count.h @@ -10,6 +10,7 @@ struct nf_conncount_data; struct nf_conncount_list { spinlock_t list_lock; + u32 last_gc; /* jiffies at most recent gc */ struct list_head head; /* connections with the same filtering key */ unsigned int count; /* length of list */ }; -- cgit v1.2.3