| Message ID | 20191115035403.20173-2-gerald.yang@canonical.com |
|---|---|
| State | New |
| Headers | show |
| Series | Fix for LP:#1852663 | expand |
Acked-by: Sultan Alsawaf <sultan.alsawaf@canonical.com> On Thu, Nov 14, 2019, 7:54 PM Gerald Yang <gerald.yang@canonical.com> wrote: > BugLink: https://bugs.launchpad.net/bugs/1852663 > > Fix a general protection in i40e_config_vf_promiscuous_mode > > When deleting VMs with VFs created by i40e, a general protection > fault occurs in i40e_config_vf_promiscuous_mode due to race > condition for vsi->mac_filter_hash > And it also happens when deleteing pod with VFs > > This issue was reported in e1000-devel mailling list > https://sourceforge.net/p/e1000/mailman/message/36766306/ > > Suggested-by: Billy McFall <bmcfall@redhat.com> > Signed-off-by: Gerald Yang <gerald.yang@canonical.com> > --- > drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c | 5 ++++- > 1 file changed, 4 insertions(+), 1 deletion(-) > > diff --git a/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c > b/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c > index 02b09a8ad54c..6f78db626031 100644 > --- a/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c > +++ b/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c > @@ -1121,6 +1121,7 @@ static i40e_status > i40e_config_vf_promiscuous_mode(struct i40e_vf *vf, > struct i40e_pf *pf = vf->pf; > struct i40e_hw *hw = &pf->hw; > struct i40e_mac_filter *f; > + struct hlist_node *h; > i40e_status aq_ret = 0; > struct i40e_vsi *vsi; > int bkt; > @@ -1160,7 +1161,8 @@ static i40e_status > i40e_config_vf_promiscuous_mode(struct i40e_vf *vf, > } > return aq_ret; > } else if (i40e_getnum_vf_vsi_vlan_filters(vsi)) { > - hash_for_each(vsi->mac_filter_hash, bkt, f, hlist) { > + spin_lock_bh(&vsi->mac_filter_hash_lock); > + hash_for_each_safe(vsi->mac_filter_hash, bkt, h, f, hlist) > { > if (f->vlan < 0 || f->vlan > I40E_MAX_VLANID) > continue; > aq_ret = i40e_aq_set_vsi_mc_promisc_on_vlan(hw, > @@ -1193,6 +1195,7 @@ static i40e_status > i40e_config_vf_promiscuous_mode(struct i40e_vf *vf, > i40e_aq_str(&pf->hw, aq_err)); > } > } > + spin_unlock_bh(&vsi->mac_filter_hash_lock); > return aq_ret; > } > aq_ret = i40e_aq_set_vsi_multicast_promiscuous(hw, vsi->seid, > allmulti, > -- > 2.17.1 > > > -- > kernel-team mailing list > kernel-team@lists.ubuntu.com > https://lists.ubuntu.com/mailman/listinfo/kernel-team >
On 11/14/19 7:54 PM, Gerald Yang wrote: > BugLink: https://bugs.launchpad.net/bugs/1852663 > > Fix a general protection in i40e_config_vf_promiscuous_mode > > When deleting VMs with VFs created by i40e, a general protection > fault occurs in i40e_config_vf_promiscuous_mode due to race > condition for vsi->mac_filter_hash > And it also happens when deleteing pod with VFs > > This issue was reported in e1000-devel mailling list > https://sourceforge.net/p/e1000/mailman/message/36766306/ > > Suggested-by: Billy McFall <bmcfall@redhat.com> > Signed-off-by: Gerald Yang <gerald.yang@canonical.com> This seems reasonable to me, it looks like the lock is well-managed. It doesn't look like there's been much discussion in the mailing list it was submitted to regarding acceptance, so it would be good to keep an eye on that. Acked-by: Connor Kuehl <connor.kuehl@canonical.com> > --- > drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c | 5 ++++- > 1 file changed, 4 insertions(+), 1 deletion(-) > > diff --git a/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c b/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c > index 02b09a8ad54c..6f78db626031 100644 > --- a/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c > +++ b/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c > @@ -1121,6 +1121,7 @@ static i40e_status i40e_config_vf_promiscuous_mode(struct i40e_vf *vf, > struct i40e_pf *pf = vf->pf; > struct i40e_hw *hw = &pf->hw; > struct i40e_mac_filter *f; > + struct hlist_node *h; > i40e_status aq_ret = 0; > struct i40e_vsi *vsi; > int bkt; > @@ -1160,7 +1161,8 @@ static i40e_status i40e_config_vf_promiscuous_mode(struct i40e_vf *vf, > } > return aq_ret; > } else if (i40e_getnum_vf_vsi_vlan_filters(vsi)) { > - hash_for_each(vsi->mac_filter_hash, bkt, f, hlist) { > + spin_lock_bh(&vsi->mac_filter_hash_lock); > + hash_for_each_safe(vsi->mac_filter_hash, bkt, h, f, hlist) { > if (f->vlan < 0 || f->vlan > I40E_MAX_VLANID) > continue; > aq_ret = i40e_aq_set_vsi_mc_promisc_on_vlan(hw, > @@ -1193,6 +1195,7 @@ static i40e_status i40e_config_vf_promiscuous_mode(struct i40e_vf *vf, > i40e_aq_str(&pf->hw, aq_err)); > } > } > + spin_unlock_bh(&vsi->mac_filter_hash_lock); > return aq_ret; > } > aq_ret = i40e_aq_set_vsi_multicast_promiscuous(hw, vsi->seid, allmulti, >
diff --git a/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c b/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c index 02b09a8ad54c..6f78db626031 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c +++ b/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c @@ -1121,6 +1121,7 @@ static i40e_status i40e_config_vf_promiscuous_mode(struct i40e_vf *vf, struct i40e_pf *pf = vf->pf; struct i40e_hw *hw = &pf->hw; struct i40e_mac_filter *f; + struct hlist_node *h; i40e_status aq_ret = 0; struct i40e_vsi *vsi; int bkt; @@ -1160,7 +1161,8 @@ static i40e_status i40e_config_vf_promiscuous_mode(struct i40e_vf *vf, } return aq_ret; } else if (i40e_getnum_vf_vsi_vlan_filters(vsi)) { - hash_for_each(vsi->mac_filter_hash, bkt, f, hlist) { + spin_lock_bh(&vsi->mac_filter_hash_lock); + hash_for_each_safe(vsi->mac_filter_hash, bkt, h, f, hlist) { if (f->vlan < 0 || f->vlan > I40E_MAX_VLANID) continue; aq_ret = i40e_aq_set_vsi_mc_promisc_on_vlan(hw, @@ -1193,6 +1195,7 @@ static i40e_status i40e_config_vf_promiscuous_mode(struct i40e_vf *vf, i40e_aq_str(&pf->hw, aq_err)); } } + spin_unlock_bh(&vsi->mac_filter_hash_lock); return aq_ret; } aq_ret = i40e_aq_set_vsi_multicast_promiscuous(hw, vsi->seid, allmulti,
BugLink: https://bugs.launchpad.net/bugs/1852663 Fix a general protection in i40e_config_vf_promiscuous_mode When deleting VMs with VFs created by i40e, a general protection fault occurs in i40e_config_vf_promiscuous_mode due to race condition for vsi->mac_filter_hash And it also happens when deleteing pod with VFs This issue was reported in e1000-devel mailling list https://sourceforge.net/p/e1000/mailman/message/36766306/ Suggested-by: Billy McFall <bmcfall@redhat.com> Signed-off-by: Gerald Yang <gerald.yang@canonical.com> --- drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-)