Re: [PATCHv4 net 1/3] bonding: move IPsec deletion to bond_ipsec_free_sa

[Cosmin Ratiu <cratiu@nvidia.com>]

On Wed, 2025-03-05 at 14:13 +0000, Hangbin Liu wrote:
> On Wed, Mar 05, 2025 at 10:38:36AM +0200, Nikolay Aleksandrov wrote:
> > > @@ -617,8 +614,18 @@ static void bond_ipsec_del_sa_all(struct
> > > bonding *bond)
> > >  
> > >  	mutex_lock(&bond->ipsec_lock);
> > >  	list_for_each_entry(ipsec, &bond->ipsec_list, list) {
> > 
> > Second time - you should use list_for_each_entry_safe if you're
> > walking and deleting
> > elements from the list.
> 
> Sorry, I missed this comment. I will update in next version.
> 
> > 
> > > +		spin_lock_bh(&ipsec->xs->lock);
> > >  		if (!ipsec->xs->xso.real_dev)
> > > -			continue;
> > > +			goto next;
> > > +
> > > +		if (ipsec->xs->km.state == XFRM_STATE_DEAD) {
> > > +			/* already dead no need to delete again
> > > */
> > > +			if (real_dev->xfrmdev_ops-
> > > >xdo_dev_state_free)
> > > +				real_dev->xfrmdev_ops-
> > > >xdo_dev_state_free(ipsec->xs);
> > 
> > Have you checked if .xdo_dev_state_free can sleep?
> > I see at least one that can: mlx5e_xfrm_free_state().
> 
> Hmm, This brings us back to the initial problem. We tried to avoid
> calling
> a spin lock in a sleep context (bond_ipsec_del_sa), but now the new
> code
> encounters this issue again.

The reason the mutex was added (instead of the spinlock used before)
was exactly because the add and free offload operations could sleep.

> With your reply, I also checked the xdo_dev_state_add() in
> bond_ipsec_add_sa_all(), which may also sleep, e.g.
> mlx5e_xfrm_add_state(),
> 
> If we unlock the spin lock, then the race came back again.
> 
> Any idea about this?

The race is between bond_ipsec_del_sa_all and bond_ipsec_del_sa (plus
bond_ipsec_free_sa). The issue is that when bond_ipsec_del_sa_all
releases x->lock, bond_ipsec_del_sa can immediately be called, followed
by bond_ipsec_free_sa.
Maybe dropping x->lock after setting real_dev to NULL? I checked,
real_dev is not used anywhere on the free calls, I think. I have
another series refactoring things around real_dev, I hope to be able to
send it soon.

Here's a sketch of this idea:

--- a/drivers/net/bonding/bond_main.c
+++ b/drivers/net/bonding/bond_main.c
@@ -613,8 +613,11 @@ static void bond_ipsec_del_sa_all(struct bonding
*bond)
 
        mutex_lock(&bond->ipsec_lock);
        list_for_each_entry(ipsec, &bond->ipsec_list, list) {
-               if (!ipsec->xs->xso.real_dev)
+               spin_lock(&ipsec->x->lock);
+               if (!ipsec->xs->xso.real_dev) {
+                       spin_unlock(&ipsec->x->lock);
                        continue;
+               }
 
                if (!real_dev->xfrmdev_ops ||
                    !real_dev->xfrmdev_ops->xdo_dev_state_delete ||
@@ -622,12 +625,16 @@ static void bond_ipsec_del_sa_all(struct bonding
*bond)
                        slave_warn(bond_dev, real_dev,
                                   "%s: no slave
xdo_dev_state_delete\n",
                                   __func__);
-               } else {
-                       real_dev->xfrmdev_ops-
>xdo_dev_state_delete(real_dev, ipsec->xs);
-                       if (real_dev->xfrmdev_ops->xdo_dev_state_free)
-                               real_dev->xfrmdev_ops-
>xdo_dev_state_free(ipsec->xs);
-                       ipsec->xs->xso.real_dev = NULL;
+                       spin_unlock(&ipsec->x->lock);
+                       continue;
                }
+
+               real_dev->xfrmdev_ops->xdo_dev_state_delete(real_dev,
ipsec->xs);
+               ipsec->xs->xso.real_dev = NULL;
+               /* Unlock before freeing device state, it could sleep.
*/
+               spin_unlock(&ipsec->x->lock);
+               if (real_dev->xfrmdev_ops->xdo_dev_state_free)
+                       real_dev->xfrmdev_ops-
>xdo_dev_state_free(ipsec->xs);

Cosmin.