* [PATCH net] inet: ping: use hlist_nulls rcu iterator during lookup
@ 2022-11-29 14:06 Florian Westphal
2022-12-01 11:50 ` patchwork-bot+netdevbpf
0 siblings, 1 reply; 2+ messages in thread
From: Florian Westphal @ 2022-11-29 14:06 UTC (permalink / raw)
To: netdev
Cc: David S. Miller, Jakub Kicinski, Paolo Abeni, Florian Westphal,
Eric Dumazet
ping_lookup() does not acquire the table spinlock, so iteration should
use hlist_nulls_for_each_entry_rcu().
Spotted during code review.
Fixes: dbca1596bbb0 ("ping: convert to RCU lookups, get rid of rwlock")
Cc: Eric Dumazet <edumazet@google.com>
Signed-off-by: Florian Westphal <fw@strlen.de>
---
.clang-format | 1 +
include/net/ping.h | 3 ---
net/ipv4/ping.c | 7 ++++++-
3 files changed, 7 insertions(+), 4 deletions(-)
diff --git a/.clang-format b/.clang-format
index 1247d54f9e49..8d01225bfcb7 100644
--- a/.clang-format
+++ b/.clang-format
@@ -535,6 +535,7 @@ ForEachMacros:
- 'perf_hpp_list__for_each_sort_list_safe'
- 'perf_pmu__for_each_hybrid_pmu'
- 'ping_portaddr_for_each_entry'
+ - 'ping_portaddr_for_each_entry_rcu'
- 'plist_for_each'
- 'plist_for_each_continue'
- 'plist_for_each_entry'
diff --git a/include/net/ping.h b/include/net/ping.h
index e4ff3911cbf5..9233ad3de0ad 100644
--- a/include/net/ping.h
+++ b/include/net/ping.h
@@ -16,9 +16,6 @@
#define PING_HTABLE_SIZE 64
#define PING_HTABLE_MASK (PING_HTABLE_SIZE-1)
-#define ping_portaddr_for_each_entry(__sk, node, list) \
- hlist_nulls_for_each_entry(__sk, node, list, sk_nulls_node)
-
/*
* gid_t is either uint or ushort. We want to pass it to
* proc_dointvec_minmax(), so it must not be larger than MAX_INT
diff --git a/net/ipv4/ping.c b/net/ipv4/ping.c
index bde333b24837..04b4ec07bb06 100644
--- a/net/ipv4/ping.c
+++ b/net/ipv4/ping.c
@@ -49,6 +49,11 @@
#include <net/transp_v6.h>
#endif
+#define ping_portaddr_for_each_entry(__sk, node, list) \
+ hlist_nulls_for_each_entry(__sk, node, list, sk_nulls_node)
+#define ping_portaddr_for_each_entry_rcu(__sk, node, list) \
+ hlist_nulls_for_each_entry_rcu(__sk, node, list, sk_nulls_node)
+
struct ping_table {
struct hlist_nulls_head hash[PING_HTABLE_SIZE];
spinlock_t lock;
@@ -192,7 +197,7 @@ static struct sock *ping_lookup(struct net *net, struct sk_buff *skb, u16 ident)
return NULL;
}
- ping_portaddr_for_each_entry(sk, hnode, hslot) {
+ ping_portaddr_for_each_entry_rcu(sk, hnode, hslot) {
isk = inet_sk(sk);
pr_debug("iterate\n");
--
2.37.4
^ permalink raw reply related [flat|nested] 2+ messages in thread
* Re: [PATCH net] inet: ping: use hlist_nulls rcu iterator during lookup
2022-11-29 14:06 [PATCH net] inet: ping: use hlist_nulls rcu iterator during lookup Florian Westphal
@ 2022-12-01 11:50 ` patchwork-bot+netdevbpf
0 siblings, 0 replies; 2+ messages in thread
From: patchwork-bot+netdevbpf @ 2022-12-01 11:50 UTC (permalink / raw)
To: Florian Westphal; +Cc: netdev, davem, kuba, pabeni, edumazet
Hello:
This patch was applied to netdev/net.git (master)
by Paolo Abeni <pabeni@redhat.com>:
On Tue, 29 Nov 2022 15:06:44 +0100 you wrote:
> ping_lookup() does not acquire the table spinlock, so iteration should
> use hlist_nulls_for_each_entry_rcu().
>
> Spotted during code review.
>
> Fixes: dbca1596bbb0 ("ping: convert to RCU lookups, get rid of rwlock")
> Cc: Eric Dumazet <edumazet@google.com>
> Signed-off-by: Florian Westphal <fw@strlen.de>
>
> [...]
Here is the summary with links:
- [net] inet: ping: use hlist_nulls rcu iterator during lookup
https://git.kernel.org/netdev/net/c/c25b7a7a565e
You are awesome, thank you!
--
Deet-doot-dot, I am a bot.
https://korg.docs.kernel.org/patchwork/pwbot.html
^ permalink raw reply [flat|nested] 2+ messages in thread
end of thread, other threads:[~2022-12-01 11:50 UTC | newest]
Thread overview: 2+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2022-11-29 14:06 [PATCH net] inet: ping: use hlist_nulls rcu iterator during lookup Florian Westphal
2022-12-01 11:50 ` patchwork-bot+netdevbpf
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).