Message ID | 20220225161855.489923-1-eric.dumazet@gmail.com (mailing list archive) |
---|---|
State | Accepted |
Commit | b3483bc7a1f2673b35331e60668104ba2be46510 |
Delegated to: | Netdev Maintainers |
Headers | show |
Series | [net-next] net/sysctl: avoid two synchronize_rcu() calls | expand |
Hello: This patch was applied to netdev/net-next.git (master) by David S. Miller <davem@davemloft.net>: On Fri, 25 Feb 2022 08:18:55 -0800 you wrote: > From: Eric Dumazet <edumazet@google.com> > > Both rps_sock_flow_sysctl() and flow_limit_cpu_sysctl() > are using synchronize_rcu() right before freeing memory > either by vfree() or kfree() > > They can switch to kvfree_rcu(ptr) and kfree_rcu(ptr) to benefit > from asynchronous mode, instead of blocking the current thread. > > [...] Here is the summary with links: - [net-next] net/sysctl: avoid two synchronize_rcu() calls https://git.kernel.org/netdev/net-next/c/b3483bc7a1f2 You are awesome, thank you!
diff --git a/net/core/sysctl_net_core.c b/net/core/sysctl_net_core.c index dbeb8ecbcd98f0aa6c02c650f925b05faf23ecad..7123fe7feeac634023d4e73247db0a20e3fcc383 100644 --- a/net/core/sysctl_net_core.c +++ b/net/core/sysctl_net_core.c @@ -103,8 +103,7 @@ static int rps_sock_flow_sysctl(struct ctl_table *table, int write, if (orig_sock_table) { static_branch_dec(&rps_needed); static_branch_dec(&rfs_needed); - synchronize_rcu(); - vfree(orig_sock_table); + kvfree_rcu(orig_sock_table); } } } @@ -142,8 +141,7 @@ static int flow_limit_cpu_sysctl(struct ctl_table *table, int write, lockdep_is_held(&flow_limit_update_mutex)); if (cur && !cpumask_test_cpu(i, mask)) { RCU_INIT_POINTER(sd->flow_limit, NULL); - synchronize_rcu(); - kfree(cur); + kfree_rcu(cur); } else if (!cur && cpumask_test_cpu(i, mask)) { cur = kzalloc_node(len, GFP_KERNEL, cpu_to_node(i));