All of lore.kernel.org
 help / color / mirror / Atom feed
* [PATCH net] openvswitch: Fix ovs_flow_key_update()
@ 2017-03-30  0:14 Yi-Hung Wei
  2017-03-30 13:22 ` Jiri Benc
  0 siblings, 1 reply; 4+ messages in thread
From: Yi-Hung Wei @ 2017-03-30  0:14 UTC (permalink / raw)
  To: jbenc, joe, netdev; +Cc: Yi-Hung Wei, linux-kernel

ovs_flow_key_update() is called when the flow key is invalid, and it is
used to update and revalidate the flow key. Commit 329f45bc4f19
("openvswitch: add mac_proto field to the flow key") introduces mac_proto
field to flow key and use it to determine whether the flow key is valid.
However, the commit does not update the code path in ovs_flow_key_update()
to revalidate the flow key which may cause BUG_ON() on execute_recirc().
This patch addresses the aforementioned issue.

Fixes: 329f45bc4f19 ("openvswitch: add mac_proto field to the flow key")
Signed-off-by: Yi-Hung Wei <yihung.wei@gmail.com>
---
 net/openvswitch/flow.c | 9 +++++++++
 1 file changed, 9 insertions(+)

diff --git a/net/openvswitch/flow.c b/net/openvswitch/flow.c
index 9d4bb8eb63f2..bb33f5341294 100644
--- a/net/openvswitch/flow.c
+++ b/net/openvswitch/flow.c
@@ -53,6 +53,8 @@
 #include "flow_netlink.h"
 #include "vport.h"
 
+static int key_extract_mac_proto(struct sk_buff *skb);
+
 u64 ovs_flow_used_time(unsigned long flow_jiffies)
 {
 	struct timespec cur_ts;
@@ -745,6 +747,13 @@ static int key_extract(struct sk_buff *skb, struct sw_flow_key *key)
 
 int ovs_flow_key_update(struct sk_buff *skb, struct sw_flow_key *key)
 {
+	int res;
+
+	res = key_extract_mac_proto(skb);
+	if (res < 0)
+		return res;
+	key->mac_proto = res;
+
 	return key_extract(skb, key);
 }
 
-- 
2.7.4

^ permalink raw reply related	[flat|nested] 4+ messages in thread

* Re: [PATCH net] openvswitch: Fix ovs_flow_key_update()
  2017-03-30  0:14 [PATCH net] openvswitch: Fix ovs_flow_key_update() Yi-Hung Wei
@ 2017-03-30 13:22 ` Jiri Benc
  2017-03-30 18:39   ` Yi-Hung Wei
  0 siblings, 1 reply; 4+ messages in thread
From: Jiri Benc @ 2017-03-30 13:22 UTC (permalink / raw)
  To: Yi-Hung Wei; +Cc: joe, netdev, linux-kernel

On Wed, 29 Mar 2017 17:14:10 -0700, Yi-Hung Wei wrote:
> ovs_flow_key_update() is called when the flow key is invalid, and it is
> used to update and revalidate the flow key. Commit 329f45bc4f19
> ("openvswitch: add mac_proto field to the flow key") introduces mac_proto
> field to flow key and use it to determine whether the flow key is valid.
> However, the commit does not update the code path in ovs_flow_key_update()
> to revalidate the flow key which may cause BUG_ON() on execute_recirc().
> This patch addresses the aforementioned issue.

Good catch.

>  int ovs_flow_key_update(struct sk_buff *skb, struct sw_flow_key *key)
>  {
> +	int res;
> +
> +	res = key_extract_mac_proto(skb);
> +	if (res < 0)
> +		return res;
> +	key->mac_proto = res;
> +
>  	return key_extract(skb, key);
>  }

But this should just reset the SW_FLOW_KEY_INVALID flag, there's no
need to recompute mac_proto.

Something like this:

 int ovs_flow_key_update(struct sk_buff *skb, struct sw_flow_key *key)
 {
-	return key_extract(skb, key);
+	int res;
+
+	res = key_extract(skb, key);
+	if (!res)
+		key->mac_proto &= ~SW_FLOW_KEY_INVALID;
+	return res;
 }

Thanks,

 Jiri

^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: [PATCH net] openvswitch: Fix ovs_flow_key_update()
  2017-03-30 13:22 ` Jiri Benc
@ 2017-03-30 18:39   ` Yi-Hung Wei
  2017-03-30 18:47     ` Jiri Benc
  0 siblings, 1 reply; 4+ messages in thread
From: Yi-Hung Wei @ 2017-03-30 18:39 UTC (permalink / raw)
  To: Jiri Benc; +Cc: Joe Stringer, netdev, linux-kernel

On Thu, Mar 30, 2017 at 6:22 AM, Jiri Benc <jbenc@redhat.com> wrote:
> On Wed, 29 Mar 2017 17:14:10 -0700, Yi-Hung Wei wrote:
>> ovs_flow_key_update() is called when the flow key is invalid, and it is
>> used to update and revalidate the flow key. Commit 329f45bc4f19
>> ("openvswitch: add mac_proto field to the flow key") introduces mac_proto
>> field to flow key and use it to determine whether the flow key is valid.
>> However, the commit does not update the code path in ovs_flow_key_update()
>> to revalidate the flow key which may cause BUG_ON() on execute_recirc().
>> This patch addresses the aforementioned issue.
>
> Good catch.
>
>>  int ovs_flow_key_update(struct sk_buff *skb, struct sw_flow_key *key)
>>  {
>> +     int res;
>> +
>> +     res = key_extract_mac_proto(skb);
>> +     if (res < 0)
>> +             return res;
>> +     key->mac_proto = res;
>> +
>>       return key_extract(skb, key);
>>  }
>
> But this should just reset the SW_FLOW_KEY_INVALID flag, there's no
> need to recompute mac_proto.
>
> Something like this:
>
>  int ovs_flow_key_update(struct sk_buff *skb, struct sw_flow_key *key)
>  {
> -       return key_extract(skb, key);
> +       int res;
> +
> +       res = key_extract(skb, key);
> +       if (!res)
> +               key->mac_proto &= ~SW_FLOW_KEY_INVALID;
> +       return res;
>  }
Hi Jiri,

One case that I worry is that key_extract() currently relies on mac_proto to
decide whether to process the link layer.  So if we update key->mac_proto
after key_extract(), wouldn't we run into a problem like the following?

If we invalidate a flow key of a L3 packet, the flow's mac_proto is like this
(MAC_PROTO_NONE | SW_FLOW_KEY_INVALID), then key_extract() will
process the link layer of this L3 packet since mac_proto !=MAC_PROTO_NONE?

In this case, shall we update key_extract() like this
static int key_extract(struct sk_buff *skb, struct sw_flow_key *key)

        /* Link layer. */
        clear_vlan(key);
-       if (key->mac_proto == MAC_PROTO_NONE) {
+       if (key->mac_proto & MAC_PROTO_NONE) {
                if (unlikely(eth_type_vlan(skb->protocol)))
                        return -EINVAL;

Thanks,

-Yi-Hung

>
> Thanks,
>
>  Jiri

^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: [PATCH net] openvswitch: Fix ovs_flow_key_update()
  2017-03-30 18:39   ` Yi-Hung Wei
@ 2017-03-30 18:47     ` Jiri Benc
  0 siblings, 0 replies; 4+ messages in thread
From: Jiri Benc @ 2017-03-30 18:47 UTC (permalink / raw)
  To: Yi-Hung Wei; +Cc: Joe Stringer, netdev, linux-kernel

On Thu, 30 Mar 2017 11:39:35 -0700, Yi-Hung Wei wrote:
> If we invalidate a flow key of a L3 packet, the flow's mac_proto is like this
> (MAC_PROTO_NONE | SW_FLOW_KEY_INVALID), then key_extract() will
> process the link layer of this L3 packet since mac_proto !=MAC_PROTO_NONE?
> 
> In this case, shall we update key_extract() like this
> static int key_extract(struct sk_buff *skb, struct sw_flow_key *key)
> 
>         /* Link layer. */
>         clear_vlan(key);
> -       if (key->mac_proto == MAC_PROTO_NONE) {
> +       if (key->mac_proto & MAC_PROTO_NONE) {

Use ovs_key_mac_proto(key) == MAC_PROTO_NONE.

 Jiri

^ permalink raw reply	[flat|nested] 4+ messages in thread

end of thread, other threads:[~2017-03-30 18:47 UTC | newest]

Thread overview: 4+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2017-03-30  0:14 [PATCH net] openvswitch: Fix ovs_flow_key_update() Yi-Hung Wei
2017-03-30 13:22 ` Jiri Benc
2017-03-30 18:39   ` Yi-Hung Wei
2017-03-30 18:47     ` Jiri Benc

This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.