* [PATCH net] net: openvswitch: silence suspicious RCU usage warning
@ 2020-10-29 14:53 Eelco Chaudron
2020-10-30 21:28 ` Jakub Kicinski
0 siblings, 1 reply; 5+ messages in thread
From: Eelco Chaudron @ 2020-10-29 14:53 UTC (permalink / raw)
To: netdev; +Cc: davem, dev, kuba
Silence suspicious RCU usage warning in ovs_flow_tbl_masks_cache_resize()
by replacing rcu_dereference() with rcu_dereference_ovsl().
In addition, when creating a new datapath, make sure it's configured under
the ovs_lock.
Fixes: 9bf24f594c6a ("net: openvswitch: make masks cache size configurable")
Reported-by: syzbot+9a8f8bfcc56e8578016c@syzkaller.appspotmail.com
Signed-off-by: Eelco Chaudron <echaudro@redhat.com>
---
net/openvswitch/datapath.c | 8 ++++----
net/openvswitch/flow_table.c | 2 +-
2 files changed, 5 insertions(+), 5 deletions(-)
diff --git a/net/openvswitch/datapath.c b/net/openvswitch/datapath.c
index 832f898edb6a..020f8539fede 100644
--- a/net/openvswitch/datapath.c
+++ b/net/openvswitch/datapath.c
@@ -1695,6 +1695,9 @@ static int ovs_dp_cmd_new(struct sk_buff *skb, struct genl_info *info)
if (err)
goto err_destroy_ports;
+ /* So far only local changes have been made, now need the lock. */
+ ovs_lock();
+
/* Set up our datapath device. */
parms.name = nla_data(a[OVS_DP_ATTR_NAME]);
parms.type = OVS_VPORT_TYPE_INTERNAL;
@@ -1707,9 +1710,6 @@ static int ovs_dp_cmd_new(struct sk_buff *skb, struct genl_info *info)
if (err)
goto err_destroy_meters;
- /* So far only local changes have been made, now need the lock. */
- ovs_lock();
-
vport = new_vport(&parms);
if (IS_ERR(vport)) {
err = PTR_ERR(vport);
@@ -1725,7 +1725,6 @@ static int ovs_dp_cmd_new(struct sk_buff *skb, struct genl_info *info)
ovs_dp_reset_user_features(skb, info);
}
- ovs_unlock();
goto err_destroy_meters;
}
@@ -1742,6 +1741,7 @@ static int ovs_dp_cmd_new(struct sk_buff *skb, struct genl_info *info)
return 0;
err_destroy_meters:
+ ovs_unlock();
ovs_meters_exit(dp);
err_destroy_ports:
kfree(dp->ports);
diff --git a/net/openvswitch/flow_table.c b/net/openvswitch/flow_table.c
index f3486a37361a..c89c8da99f1a 100644
--- a/net/openvswitch/flow_table.c
+++ b/net/openvswitch/flow_table.c
@@ -390,7 +390,7 @@ static struct mask_cache *tbl_mask_cache_alloc(u32 size)
}
int ovs_flow_tbl_masks_cache_resize(struct flow_table *table, u32 size)
{
- struct mask_cache *mc = rcu_dereference(table->mask_cache);
+ struct mask_cache *mc = rcu_dereference_ovsl(table->mask_cache);
struct mask_cache *new;
if (size == mc->cache_size)
^ permalink raw reply related [flat|nested] 5+ messages in thread
* Re: [PATCH net] net: openvswitch: silence suspicious RCU usage warning
2020-10-29 14:53 [PATCH net] net: openvswitch: silence suspicious RCU usage warning Eelco Chaudron
@ 2020-10-30 21:28 ` Jakub Kicinski
2020-11-02 8:52 ` Eelco Chaudron
0 siblings, 1 reply; 5+ messages in thread
From: Jakub Kicinski @ 2020-10-30 21:28 UTC (permalink / raw)
To: Eelco Chaudron; +Cc: netdev, davem, dev
On Thu, 29 Oct 2020 15:53:21 +0100 Eelco Chaudron wrote:
> Silence suspicious RCU usage warning in ovs_flow_tbl_masks_cache_resize()
> by replacing rcu_dereference() with rcu_dereference_ovsl().
>
> In addition, when creating a new datapath, make sure it's configured under
> the ovs_lock.
>
> Fixes: 9bf24f594c6a ("net: openvswitch: make masks cache size configurable")
> Reported-by: syzbot+9a8f8bfcc56e8578016c@syzkaller.appspotmail.com
> Signed-off-by: Eelco Chaudron <echaudro@redhat.com>
> ---
> net/openvswitch/datapath.c | 8 ++++----
> net/openvswitch/flow_table.c | 2 +-
> 2 files changed, 5 insertions(+), 5 deletions(-)
>
> diff --git a/net/openvswitch/datapath.c b/net/openvswitch/datapath.c
> index 832f898edb6a..020f8539fede 100644
> --- a/net/openvswitch/datapath.c
> +++ b/net/openvswitch/datapath.c
> @@ -1695,6 +1695,9 @@ static int ovs_dp_cmd_new(struct sk_buff *skb, struct genl_info *info)
> if (err)
> goto err_destroy_ports;
>
> + /* So far only local changes have been made, now need the lock. */
> + ovs_lock();
Should we move the lock below assignments to param?
Looks a little strange to protect stack variables with a global lock.
> /* Set up our datapath device. */
> parms.name = nla_data(a[OVS_DP_ATTR_NAME]);
> parms.type = OVS_VPORT_TYPE_INTERNAL;
> @@ -1707,9 +1710,6 @@ static int ovs_dp_cmd_new(struct sk_buff *skb, struct genl_info *info)
> if (err)
> goto err_destroy_meters;
>
> - /* So far only local changes have been made, now need the lock. */
> - ovs_lock();
> -
> vport = new_vport(&parms);
> if (IS_ERR(vport)) {
> err = PTR_ERR(vport);
> @@ -1725,7 +1725,6 @@ static int ovs_dp_cmd_new(struct sk_buff *skb, struct genl_info *info)
> ovs_dp_reset_user_features(skb, info);
> }
>
> - ovs_unlock();
> goto err_destroy_meters;
> }
>
> @@ -1742,6 +1741,7 @@ static int ovs_dp_cmd_new(struct sk_buff *skb, struct genl_info *info)
> return 0;
>
> err_destroy_meters:
Let's update the name of the label.
> + ovs_unlock();
> ovs_meters_exit(dp);
> err_destroy_ports:
> kfree(dp->ports);
> diff --git a/net/openvswitch/flow_table.c b/net/openvswitch/flow_table.c
> index f3486a37361a..c89c8da99f1a 100644
> --- a/net/openvswitch/flow_table.c
> +++ b/net/openvswitch/flow_table.c
> @@ -390,7 +390,7 @@ static struct mask_cache *tbl_mask_cache_alloc(u32 size)
> }
> int ovs_flow_tbl_masks_cache_resize(struct flow_table *table, u32 size)
> {
> - struct mask_cache *mc = rcu_dereference(table->mask_cache);
> + struct mask_cache *mc = rcu_dereference_ovsl(table->mask_cache);
> struct mask_cache *new;
>
> if (size == mc->cache_size)
>
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [PATCH net] net: openvswitch: silence suspicious RCU usage warning
2020-10-30 21:28 ` Jakub Kicinski
@ 2020-11-02 8:52 ` Eelco Chaudron
2020-11-02 19:51 ` Jakub Kicinski
0 siblings, 1 reply; 5+ messages in thread
From: Eelco Chaudron @ 2020-11-02 8:52 UTC (permalink / raw)
To: Jakub Kicinski; +Cc: netdev, davem, dev
On 30 Oct 2020, at 22:28, Jakub Kicinski wrote:
> On Thu, 29 Oct 2020 15:53:21 +0100 Eelco Chaudron wrote:
>> Silence suspicious RCU usage warning in
>> ovs_flow_tbl_masks_cache_resize()
>> by replacing rcu_dereference() with rcu_dereference_ovsl().
>>
>> In addition, when creating a new datapath, make sure it's configured
>> under
>> the ovs_lock.
>>
>> Fixes: 9bf24f594c6a ("net: openvswitch: make masks cache size
>> configurable")
>> Reported-by: syzbot+9a8f8bfcc56e8578016c@syzkaller.appspotmail.com
>> Signed-off-by: Eelco Chaudron <echaudro@redhat.com>
>> ---
>> net/openvswitch/datapath.c | 8 ++++----
>> net/openvswitch/flow_table.c | 2 +-
>> 2 files changed, 5 insertions(+), 5 deletions(-)
>>
>> diff --git a/net/openvswitch/datapath.c b/net/openvswitch/datapath.c
>> index 832f898edb6a..020f8539fede 100644
>> --- a/net/openvswitch/datapath.c
>> +++ b/net/openvswitch/datapath.c
>> @@ -1695,6 +1695,9 @@ static int ovs_dp_cmd_new(struct sk_buff *skb,
>> struct genl_info *info)
>> if (err)
>> goto err_destroy_ports;
>>
>> + /* So far only local changes have been made, now need the lock. */
>> + ovs_lock();
>
> Should we move the lock below assignments to param?
>
> Looks a little strange to protect stack variables with a global lock.
You are right, I should have moved it down after the assignment. I will
send out a v2.
>> /* Set up our datapath device. */
>> parms.name = nla_data(a[OVS_DP_ATTR_NAME]);
>> parms.type = OVS_VPORT_TYPE_INTERNAL;
>> @@ -1707,9 +1710,6 @@ static int ovs_dp_cmd_new(struct sk_buff *skb,
>> struct genl_info *info)
>> if (err)
>> goto err_destroy_meters;
>>
>> - /* So far only local changes have been made, now need the lock. */
>> - ovs_lock();
>> -
>> vport = new_vport(&parms);
>> if (IS_ERR(vport)) {
>> err = PTR_ERR(vport);
>> @@ -1725,7 +1725,6 @@ static int ovs_dp_cmd_new(struct sk_buff *skb,
>> struct genl_info *info)
>> ovs_dp_reset_user_features(skb, info);
>> }
>>
>> - ovs_unlock();
>> goto err_destroy_meters;
>> }
>>
>> @@ -1742,6 +1741,7 @@ static int ovs_dp_cmd_new(struct sk_buff *skb,
>> struct genl_info *info)
>> return 0;
>>
>> err_destroy_meters:
>
> Let's update the name of the label.
Guess now it is, unlock and destroy meters, so what label are you
looking for?
err_unlock_and_destroy_meters: which looks a bit long, or just
err_unlock:
>> + ovs_unlock();
>> ovs_meters_exit(dp);
>> err_destroy_ports:
>> kfree(dp->ports);
>> diff --git a/net/openvswitch/flow_table.c
>> b/net/openvswitch/flow_table.c
>> index f3486a37361a..c89c8da99f1a 100644
>> --- a/net/openvswitch/flow_table.c
>> +++ b/net/openvswitch/flow_table.c
>> @@ -390,7 +390,7 @@ static struct mask_cache
>> *tbl_mask_cache_alloc(u32 size)
>> }
>> int ovs_flow_tbl_masks_cache_resize(struct flow_table *table, u32
>> size)
>> {
>> - struct mask_cache *mc = rcu_dereference(table->mask_cache);
>> + struct mask_cache *mc = rcu_dereference_ovsl(table->mask_cache);
>> struct mask_cache *new;
>>
>> if (size == mc->cache_size)
>>
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [PATCH net] net: openvswitch: silence suspicious RCU usage warning
2020-11-02 8:52 ` Eelco Chaudron
@ 2020-11-02 19:51 ` Jakub Kicinski
2020-11-03 8:26 ` Eelco Chaudron
0 siblings, 1 reply; 5+ messages in thread
From: Jakub Kicinski @ 2020-11-02 19:51 UTC (permalink / raw)
To: Eelco Chaudron; +Cc: netdev, davem, dev
On Mon, 02 Nov 2020 09:52:19 +0100 Eelco Chaudron wrote:
> On 30 Oct 2020, at 22:28, Jakub Kicinski wrote:
> >> @@ -1695,6 +1695,9 @@ static int ovs_dp_cmd_new(struct sk_buff *skb,
> >> struct genl_info *info)
> >> if (err)
> >> goto err_destroy_ports;
> >>
> >> + /* So far only local changes have been made, now need the lock. */
> >> + ovs_lock();
> >
> > Should we move the lock below assignments to param?
> >
> > Looks a little strange to protect stack variables with a global lock.
>
> You are right, I should have moved it down after the assignment. I will
> send out a v2.
>
> > Let's update the name of the label.
>
> Guess now it is, unlock and destroy meters, so what label are you
> looking for?
>
> err_unlock_and_destroy_meters: which looks a bit long, or just
> err_unlock:
I feel like I saw some names like err_unlock_and_destroy_meters in OvS
code, but can't find them in this file right now.
I'd personally go for kist err_unlock, or maybe err_unlock_ovs as is
used in other functions in this file.
But as long as it starts with err_unlock it's fine by me :)
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [PATCH net] net: openvswitch: silence suspicious RCU usage warning
2020-11-02 19:51 ` Jakub Kicinski
@ 2020-11-03 8:26 ` Eelco Chaudron
0 siblings, 0 replies; 5+ messages in thread
From: Eelco Chaudron @ 2020-11-03 8:26 UTC (permalink / raw)
To: Jakub Kicinski; +Cc: netdev, davem, dev
On 2 Nov 2020, at 20:51, Jakub Kicinski wrote:
> On Mon, 02 Nov 2020 09:52:19 +0100 Eelco Chaudron wrote:
>> On 30 Oct 2020, at 22:28, Jakub Kicinski wrote:
>>>> @@ -1695,6 +1695,9 @@ static int ovs_dp_cmd_new(struct sk_buff *skb,
>>>> struct genl_info *info)
>>>> if (err)
>>>> goto err_destroy_ports;
>>>>
>>>> + /* So far only local changes have been made, now need the lock. */
>>>> + ovs_lock();
>>>
>>> Should we move the lock below assignments to param?
>>>
>>> Looks a little strange to protect stack variables with a global lock.
>>
>> You are right, I should have moved it down after the assignment. I will
>> send out a v2.
>>
>>> Let's update the name of the label.
>>
>> Guess now it is, unlock and destroy meters, so what label are you
>> looking for?
>>
>> err_unlock_and_destroy_meters: which looks a bit long, or just
>> err_unlock:
>
> I feel like I saw some names like err_unlock_and_destroy_meters in OvS
> code, but can't find them in this file right now.
>
> I'd personally go for kist err_unlock, or maybe err_unlock_ovs as is
> used in other functions in this file.
>
> But as long as it starts with err_unlock it's fine by me :)
Ack, sent out a v2.
^ permalink raw reply [flat|nested] 5+ messages in thread
end of thread, other threads:[~2020-11-03 8:27 UTC | newest]
Thread overview: 5+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2020-10-29 14:53 [PATCH net] net: openvswitch: silence suspicious RCU usage warning Eelco Chaudron
2020-10-30 21:28 ` Jakub Kicinski
2020-11-02 8:52 ` Eelco Chaudron
2020-11-02 19:51 ` Jakub Kicinski
2020-11-03 8:26 ` Eelco Chaudron
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.