* [ovs-dev] [PATCH net v2] openvswitch: meter: fix race when getting now_ms.
@ 2021-05-13 13:08 Tao Liu
2021-05-13 14:35 ` Ilya Maximets
2021-05-13 23:00 ` patchwork-bot+netdevbpf
0 siblings, 2 replies; 3+ messages in thread
From: Tao Liu @ 2021-05-13 13:08 UTC (permalink / raw)
To: pshelar
Cc: dev, netdev, linux-kernel, i.maximets, jean.tourrilhes, kuba,
davem, thomas.liu, wenxu
We have observed meters working unexpected if traffic is 3+Gbit/s
with multiple connections.
now_ms is not pretected by meter->lock, we may get a negative
long_delta_ms when another cpu updated meter->used, then:
delta_ms = (u32)long_delta_ms;
which will be a large value.
band->bucket += delta_ms * band->rate;
then we get a wrong band->bucket.
OpenVswitch userspace datapath has fixed the same issue[1] some
time ago, and we port the implementation to kernel datapath.
[1] https://patchwork.ozlabs.org/project/openvswitch/patch/20191025114436.9746-1-i.maximets@ovn.org/
Fixes: 96fbc13d7e77 ("openvswitch: Add meter infrastructure")
Signed-off-by: Tao Liu <thomas.liu@ucloud.cn>
Suggested-by: Ilya Maximets <i.maximets@ovn.org>
---
Changelog:
v2: just set negative long_delta_ms to zero in case of race for meter lock.
v1: make now_ms protected by meter lock.
---
net/openvswitch/meter.c | 8 ++++++++
1 file changed, 8 insertions(+)
diff --git a/net/openvswitch/meter.c b/net/openvswitch/meter.c
index 96b524c..896b8f5 100644
--- a/net/openvswitch/meter.c
+++ b/net/openvswitch/meter.c
@@ -611,6 +611,14 @@ bool ovs_meter_execute(struct datapath *dp, struct sk_buff *skb,
spin_lock(&meter->lock);
long_delta_ms = (now_ms - meter->used); /* ms */
+ if (long_delta_ms < 0) {
+ /* This condition means that we have several threads fighting
+ * for a meter lock, and the one who received the packets a
+ * bit later wins. Assuming that all racing threads received
+ * packets at the same time to avoid overflow.
+ */
+ long_delta_ms = 0;
+ }
/* Make sure delta_ms will not be too large, so that bucket will not
* wrap around below.
--
1.8.3.1
^ permalink raw reply related [flat|nested] 3+ messages in thread
* Re: [ovs-dev] [PATCH net v2] openvswitch: meter: fix race when getting now_ms.
2021-05-13 13:08 [ovs-dev] [PATCH net v2] openvswitch: meter: fix race when getting now_ms Tao Liu
@ 2021-05-13 14:35 ` Ilya Maximets
2021-05-13 23:00 ` patchwork-bot+netdevbpf
1 sibling, 0 replies; 3+ messages in thread
From: Ilya Maximets @ 2021-05-13 14:35 UTC (permalink / raw)
To: Tao Liu, pshelar
Cc: dev, netdev, linux-kernel, i.maximets, jean.tourrilhes, kuba,
davem, wenxu
On 5/13/21 3:08 PM, Tao Liu wrote:
> We have observed meters working unexpected if traffic is 3+Gbit/s
> with multiple connections.
>
> now_ms is not pretected by meter->lock, we may get a negative
> long_delta_ms when another cpu updated meter->used, then:
> delta_ms = (u32)long_delta_ms;
> which will be a large value.
>
> band->bucket += delta_ms * band->rate;
> then we get a wrong band->bucket.
>
> OpenVswitch userspace datapath has fixed the same issue[1] some
> time ago, and we port the implementation to kernel datapath.
>
> [1] https://patchwork.ozlabs.org/project/openvswitch/patch/20191025114436.9746-1-i.maximets@ovn.org/
>
> Fixes: 96fbc13d7e77 ("openvswitch: Add meter infrastructure")
> Signed-off-by: Tao Liu <thomas.liu@ucloud.cn>
> Suggested-by: Ilya Maximets <i.maximets@ovn.org>
> ---
> Changelog:
> v2: just set negative long_delta_ms to zero in case of race for meter lock.
> v1: make now_ms protected by meter lock.
> ---
Thanks!
I didn't test it, but the change looks good to me.
Reviewed-by: Ilya Maximets <i.maximets@ovn.org>
> net/openvswitch/meter.c | 8 ++++++++
> 1 file changed, 8 insertions(+)
>
> diff --git a/net/openvswitch/meter.c b/net/openvswitch/meter.c
> index 96b524c..896b8f5 100644
> --- a/net/openvswitch/meter.c
> +++ b/net/openvswitch/meter.c
> @@ -611,6 +611,14 @@ bool ovs_meter_execute(struct datapath *dp, struct sk_buff *skb,
> spin_lock(&meter->lock);
>
> long_delta_ms = (now_ms - meter->used); /* ms */
> + if (long_delta_ms < 0) {
> + /* This condition means that we have several threads fighting
> + * for a meter lock, and the one who received the packets a
> + * bit later wins. Assuming that all racing threads received
> + * packets at the same time to avoid overflow.
> + */
> + long_delta_ms = 0;
> + }
>
> /* Make sure delta_ms will not be too large, so that bucket will not
> * wrap around below.
>
^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: [ovs-dev] [PATCH net v2] openvswitch: meter: fix race when getting now_ms.
2021-05-13 13:08 [ovs-dev] [PATCH net v2] openvswitch: meter: fix race when getting now_ms Tao Liu
2021-05-13 14:35 ` Ilya Maximets
@ 2021-05-13 23:00 ` patchwork-bot+netdevbpf
1 sibling, 0 replies; 3+ messages in thread
From: patchwork-bot+netdevbpf @ 2021-05-13 23:00 UTC (permalink / raw)
To: Tao Liu
Cc: pshelar, dev, netdev, linux-kernel, i.maximets, jean.tourrilhes,
kuba, davem, wenxu
Hello:
This patch was applied to netdev/net.git (refs/heads/master):
On Thu, 13 May 2021 21:08:00 +0800 you wrote:
> We have observed meters working unexpected if traffic is 3+Gbit/s
> with multiple connections.
>
> now_ms is not pretected by meter->lock, we may get a negative
> long_delta_ms when another cpu updated meter->used, then:
> delta_ms = (u32)long_delta_ms;
> which will be a large value.
>
> [...]
Here is the summary with links:
- [ovs-dev,net,v2] openvswitch: meter: fix race when getting now_ms.
https://git.kernel.org/netdev/net/c/e4df1b0c2435
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] 3+ messages in thread
end of thread, other threads:[~2021-05-13 23:00 UTC | newest]
Thread overview: 3+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2021-05-13 13:08 [ovs-dev] [PATCH net v2] openvswitch: meter: fix race when getting now_ms Tao Liu
2021-05-13 14:35 ` Ilya Maximets
2021-05-13 23:00 ` 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).