Skip to contentSkip to conversation

ipv6: use call_rcu_hurry() in fib6_info_release()

PATCH Applied Eric Dumazet 4 participants 4 messages Apr 26, 2024 – Apr 29, 2024

Subsystems (3)
lore
Share & tools
lore
#0 lore raw

PATCH 1/1 ipv6: use call_rcu_hurry() in fib6_info_release()

This is a followup of commit c4e86b4363ac ("net: add two more call_rcu_hurry()")

fib6_info_destroy_rcu() is calling nexthop_put() or fib6_nh_release()

We must not delay it too much or risk unregister_netdevice/ref_tracker traces because references to netdev are not released in time.

This should speedup device/netns dismantles when CONFIG_RCU_LAZY=y

---
 include/net/ip6_fib.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
include/net/ip6_fib.h +1 -1
index 323c94f1845b9e3eed52a2a19a4871cf8174d9c2..7834f7f29d3c9323e753be9e46499784a619de1b 100644
@@ -338,7 +338,7 @@ static inline void fib6_info_release(struct fib6_info *f6i)
 {
 	if (f6i && refcount_dec_and_test(&f6i->fib6_ref)) {
 		DEBUG_NET_WARN_ON_ONCE(!hlist_unhashed(&f6i->gc_link));
-		call_rcu(&f6i->rcu, fib6_info_destroy_rcu);
+		call_rcu_hurry(&f6i->rcu, fib6_info_destroy_rcu);
 	}
 }
 
Signature · 1 line
2.44.0.769.g3c40516874-goog
Reviewed-by on v1 1/1
#1 lore raw
Quoted 11 lines · Fri, Apr 26, 2024 at 12:47:22PM CEST, [email protected] wrote:

This is a followup of commit c4e86b4363ac ("net: add two more call_rcu_hurry()")

fib6_info_destroy_rcu() is calling nexthop_put() or fib6_nh_release()

We must not delay it too much or risk unregister_netdevice/ref_tracker traces because references to netdev are not released in time.

This should speedup device/netns dismantles when CONFIG_RCU_LAZY=y

maintainer Reviewed-by on v1 1/1
#2 lore raw
In reply to Eric Dumazet
Quoted 8 lines · Eric Dumazet

This is a followup of commit c4e86b4363ac ("net: add two more call_rcu_hurry()")

fib6_info_destroy_rcu() is calling nexthop_put() or fib6_nh_release()

We must not delay it too much or risk unregister_netdevice/ref_tracker traces because references to netdev are not released in time.

Quoted 7 lines

This should speedup device/netns dismantles when CONFIG_RCU_LAZY=y

---
 include/net/ip6_fib.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
#3 lore raw
In reply to Eric Dumazet
This patch was applied to netdev/net-next.git (main) by David S. Miller <[email protected]>:

Hello:

This patch was applied to netdev/net-next.git (main) by David S. Miller <[email protected]>:

Quoted 9 lines · On Fri, 26 Apr 2024 10:47:22 +0000 you wrote:

This is a followup of commit c4e86b4363ac ("net: add two more call_rcu_hurry()")

fib6_info_destroy_rcu() is calling nexthop_put() or fib6_nh_release()

We must not delay it too much or risk unregister_netdevice/ref_tracker traces because references to netdev are not released in time.

[...]

Here is the summary with links:
- [net-next] ipv6: use call_rcu_hurry() in fib6_info_release()

    https://git.kernel.org/netdev/net-next/c/b5327b9a300e

You are awesome, thank you!

Signature · 2 lines
Deet-doot-dot, I am a bot.
https://korg.docs.kernel.org/patchwork/pwbot.html