* [PATCH net 1/4] ice: Fix SRIOV LAG disable on non-compliant aggregate
2023-11-07 0:48 [PATCH net 0/4][pull request] Intel Wired LAN Driver Updates 2023-11-06 (ice) Tony Nguyen
@ 2023-11-07 0:48 ` Tony Nguyen
2023-11-07 0:48 ` [PATCH net 2/4] ice: lag: in RCU, use atomic allocation Tony Nguyen
` (3 subsequent siblings)
4 siblings, 0 replies; 6+ messages in thread
From: Tony Nguyen @ 2023-11-07 0:48 UTC (permalink / raw)
To: davem, kuba, pabeni, edumazet, netdev
Cc: Dave Ertman, anthony.l.nguyen, daniel.machon, Wojciech Drewek,
Simon Horman, Sujai Buvaneswaran
From: Dave Ertman <david.m.ertman@intel•com>
If an attribute of an aggregate interface disqualifies it from supporting
SRIOV, the driver will unwind the SRIOV support. Currently the driver is
clearing the feature bit for all interfaces in the aggregate, but this is
not allowing the other interfaces to unwind successfully on driver unload.
Only clear the feature bit for the interface that is currently unwinding.
Fixes: bf65da2eb279 ("ice: enforce interface eligibility and add messaging for SRIOV LAG")
Signed-off-by: Dave Ertman <david.m.ertman@intel•com>
Reviewed-by: Wojciech Drewek <wojciech.drewek@intel•com>
Reviewed-by: Simon Horman <horms@kernel•org>
Tested-by: Sujai Buvaneswaran <sujai.buvaneswaran@intel•com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel•com>
---
drivers/net/ethernet/intel/ice/ice_lag.c | 12 +++---------
1 file changed, 3 insertions(+), 9 deletions(-)
diff --git a/drivers/net/ethernet/intel/ice/ice_lag.c b/drivers/net/ethernet/intel/ice/ice_lag.c
index b980f89dc892..95e46bde54fe 100644
--- a/drivers/net/ethernet/intel/ice/ice_lag.c
+++ b/drivers/net/ethernet/intel/ice/ice_lag.c
@@ -1555,18 +1555,12 @@ static void ice_lag_chk_disabled_bond(struct ice_lag *lag, void *ptr)
*/
static void ice_lag_disable_sriov_bond(struct ice_lag *lag)
{
- struct ice_lag_netdev_list *entry;
struct ice_netdev_priv *np;
- struct net_device *netdev;
struct ice_pf *pf;
- list_for_each_entry(entry, lag->netdev_head, node) {
- netdev = entry->netdev;
- np = netdev_priv(netdev);
- pf = np->vsi->back;
-
- ice_clear_feature_support(pf, ICE_F_SRIOV_LAG);
- }
+ np = netdev_priv(lag->netdev);
+ pf = np->vsi->back;
+ ice_clear_feature_support(pf, ICE_F_SRIOV_LAG);
}
/**
--
2.41.0
^ permalink raw reply related [flat|nested] 6+ messages in thread* [PATCH net 2/4] ice: lag: in RCU, use atomic allocation
2023-11-07 0:48 [PATCH net 0/4][pull request] Intel Wired LAN Driver Updates 2023-11-06 (ice) Tony Nguyen
2023-11-07 0:48 ` [PATCH net 1/4] ice: Fix SRIOV LAG disable on non-compliant aggregate Tony Nguyen
@ 2023-11-07 0:48 ` Tony Nguyen
2023-11-07 0:48 ` [PATCH net 3/4] ice: Fix VF-VF filter rules in switchdev mode Tony Nguyen
` (2 subsequent siblings)
4 siblings, 0 replies; 6+ messages in thread
From: Tony Nguyen @ 2023-11-07 0:48 UTC (permalink / raw)
To: davem, kuba, pabeni, edumazet, netdev
Cc: Michal Schmidt, anthony.l.nguyen, daniel.machon, Wojciech Drewek,
Pucha Himasekhar Reddy, Simon Horman
From: Michal Schmidt <mschmidt@redhat•com>
Sleeping is not allowed in RCU read-side critical sections.
Use atomic allocations under rcu_read_lock.
Fixes: 1e0f9881ef79 ("ice: Flesh out implementation of support for SRIOV on bonded interface")
Fixes: 41ccedf5ca8f ("ice: implement lag netdev event handler")
Fixes: 3579aa86fb40 ("ice: update reset path for SRIOV LAG support")
Signed-off-by: Michal Schmidt <mschmidt@redhat•com>
Reviewed-by: Wojciech Drewek <wojciech.drewek@intel•com>
Tested-by: Pucha Himasekhar Reddy <himasekharx.reddy.pucha@intel•com> (A Contingent worker at Intel)
Reviewed-by: Simon Horman <horms@kernel•org>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel•com>
---
drivers/net/ethernet/intel/ice/ice_lag.c | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git a/drivers/net/ethernet/intel/ice/ice_lag.c b/drivers/net/ethernet/intel/ice/ice_lag.c
index 95e46bde54fe..cd065ec48c87 100644
--- a/drivers/net/ethernet/intel/ice/ice_lag.c
+++ b/drivers/net/ethernet/intel/ice/ice_lag.c
@@ -628,7 +628,7 @@ void ice_lag_move_new_vf_nodes(struct ice_vf *vf)
INIT_LIST_HEAD(&ndlist.node);
rcu_read_lock();
for_each_netdev_in_bond_rcu(lag->upper_netdev, tmp_nd) {
- nl = kzalloc(sizeof(*nl), GFP_KERNEL);
+ nl = kzalloc(sizeof(*nl), GFP_ATOMIC);
if (!nl)
break;
@@ -1692,7 +1692,7 @@ ice_lag_event_handler(struct notifier_block *notif_blk, unsigned long event,
rcu_read_lock();
for_each_netdev_in_bond_rcu(upper_netdev, tmp_nd) {
- nd_list = kzalloc(sizeof(*nd_list), GFP_KERNEL);
+ nd_list = kzalloc(sizeof(*nd_list), GFP_ATOMIC);
if (!nd_list)
break;
@@ -2069,7 +2069,7 @@ void ice_lag_rebuild(struct ice_pf *pf)
INIT_LIST_HEAD(&ndlist.node);
rcu_read_lock();
for_each_netdev_in_bond_rcu(lag->upper_netdev, tmp_nd) {
- nl = kzalloc(sizeof(*nl), GFP_KERNEL);
+ nl = kzalloc(sizeof(*nl), GFP_ATOMIC);
if (!nl)
break;
--
2.41.0
^ permalink raw reply related [flat|nested] 6+ messages in thread* [PATCH net 3/4] ice: Fix VF-VF filter rules in switchdev mode
2023-11-07 0:48 [PATCH net 0/4][pull request] Intel Wired LAN Driver Updates 2023-11-06 (ice) Tony Nguyen
2023-11-07 0:48 ` [PATCH net 1/4] ice: Fix SRIOV LAG disable on non-compliant aggregate Tony Nguyen
2023-11-07 0:48 ` [PATCH net 2/4] ice: lag: in RCU, use atomic allocation Tony Nguyen
@ 2023-11-07 0:48 ` Tony Nguyen
2023-11-07 0:48 ` [PATCH net 4/4] ice: Fix VF-VF direction matching in drop rule in switchdev Tony Nguyen
2023-11-09 2:50 ` [PATCH net 0/4][pull request] Intel Wired LAN Driver Updates 2023-11-06 (ice) patchwork-bot+netdevbpf
4 siblings, 0 replies; 6+ messages in thread
From: Tony Nguyen @ 2023-11-07 0:48 UTC (permalink / raw)
To: davem, kuba, pabeni, edumazet, netdev
Cc: Aniruddha Paul, anthony.l.nguyen, Przemek Kitszel,
Wojciech Drewek, Sujai Buvaneswaran
From: Aniruddha Paul <aniruddha.paul@intel•com>
Any packet leaving VSI i.e VF's VSI is considered as
egress traffic by HW, thus failing to match the added
rule.
Mark the direction for redirect rules as below:
1. VF-VF - Egress
2. Uplink-VF - Ingress
3. VF-Uplink - Egress
4. Link_Partner-Uplink - Ingress
5. Link_Partner-VF - Ingress
Fixes: 0960a27bd479 ("ice: Add direction metadata")
Reviewed-by: Przemek Kitszel <przemyslaw.kitszel@intel•com>
Reviewed-by: Wojciech Drewek <wojciech.drewek@intel•com>
Signed-off-by: Aniruddha Paul <aniruddha.paul@intel•com>
Tested-by: Sujai Buvaneswaran <sujai.buvaneswaran@intel•com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel•com>
---
drivers/net/ethernet/intel/ice/ice_tc_lib.c | 90 ++++++++++++++-------
1 file changed, 62 insertions(+), 28 deletions(-)
diff --git a/drivers/net/ethernet/intel/ice/ice_tc_lib.c b/drivers/net/ethernet/intel/ice/ice_tc_lib.c
index 37b54db91df2..0e75fc6b3c06 100644
--- a/drivers/net/ethernet/intel/ice/ice_tc_lib.c
+++ b/drivers/net/ethernet/intel/ice/ice_tc_lib.c
@@ -630,32 +630,61 @@ bool ice_is_tunnel_supported(struct net_device *dev)
return ice_tc_tun_get_type(dev) != TNL_LAST;
}
-static int
-ice_eswitch_tc_parse_action(struct ice_tc_flower_fltr *fltr,
- struct flow_action_entry *act)
+static bool ice_tc_is_dev_uplink(struct net_device *dev)
+{
+ return netif_is_ice(dev) || ice_is_tunnel_supported(dev);
+}
+
+static int ice_tc_setup_redirect_action(struct net_device *filter_dev,
+ struct ice_tc_flower_fltr *fltr,
+ struct net_device *target_dev)
{
struct ice_repr *repr;
+ fltr->action.fltr_act = ICE_FWD_TO_VSI;
+
+ if (ice_is_port_repr_netdev(filter_dev) &&
+ ice_is_port_repr_netdev(target_dev)) {
+ repr = ice_netdev_to_repr(target_dev);
+
+ fltr->dest_vsi = repr->src_vsi;
+ fltr->direction = ICE_ESWITCH_FLTR_EGRESS;
+ } else if (ice_is_port_repr_netdev(filter_dev) &&
+ ice_tc_is_dev_uplink(target_dev)) {
+ repr = ice_netdev_to_repr(filter_dev);
+
+ fltr->dest_vsi = repr->src_vsi->back->switchdev.uplink_vsi;
+ fltr->direction = ICE_ESWITCH_FLTR_EGRESS;
+ } else if (ice_tc_is_dev_uplink(filter_dev) &&
+ ice_is_port_repr_netdev(target_dev)) {
+ repr = ice_netdev_to_repr(target_dev);
+
+ fltr->dest_vsi = repr->src_vsi;
+ fltr->direction = ICE_ESWITCH_FLTR_INGRESS;
+ } else {
+ NL_SET_ERR_MSG_MOD(fltr->extack,
+ "Unsupported netdevice in switchdev mode");
+ return -EINVAL;
+ }
+
+ return 0;
+}
+
+static int ice_eswitch_tc_parse_action(struct net_device *filter_dev,
+ struct ice_tc_flower_fltr *fltr,
+ struct flow_action_entry *act)
+{
+ int err;
+
switch (act->id) {
case FLOW_ACTION_DROP:
fltr->action.fltr_act = ICE_DROP_PACKET;
break;
case FLOW_ACTION_REDIRECT:
- fltr->action.fltr_act = ICE_FWD_TO_VSI;
-
- if (ice_is_port_repr_netdev(act->dev)) {
- repr = ice_netdev_to_repr(act->dev);
-
- fltr->dest_vsi = repr->src_vsi;
- fltr->direction = ICE_ESWITCH_FLTR_INGRESS;
- } else if (netif_is_ice(act->dev) ||
- ice_is_tunnel_supported(act->dev)) {
- fltr->direction = ICE_ESWITCH_FLTR_EGRESS;
- } else {
- NL_SET_ERR_MSG_MOD(fltr->extack, "Unsupported netdevice in switchdev mode");
- return -EINVAL;
- }
+ err = ice_tc_setup_redirect_action(filter_dev, fltr, act->dev);
+ if (err)
+ return err;
break;
@@ -696,10 +725,6 @@ ice_eswitch_add_tc_fltr(struct ice_vsi *vsi, struct ice_tc_flower_fltr *fltr)
goto exit;
}
- /* egress traffic is always redirect to uplink */
- if (fltr->direction == ICE_ESWITCH_FLTR_EGRESS)
- fltr->dest_vsi = vsi->back->switchdev.uplink_vsi;
-
rule_info.sw_act.fltr_act = fltr->action.fltr_act;
if (fltr->action.fltr_act != ICE_DROP_PACKET)
rule_info.sw_act.vsi_handle = fltr->dest_vsi->idx;
@@ -713,13 +738,21 @@ ice_eswitch_add_tc_fltr(struct ice_vsi *vsi, struct ice_tc_flower_fltr *fltr)
rule_info.flags_info.act_valid = true;
if (fltr->direction == ICE_ESWITCH_FLTR_INGRESS) {
+ /* Uplink to VF */
rule_info.sw_act.flag |= ICE_FLTR_RX;
rule_info.sw_act.src = hw->pf_id;
rule_info.flags_info.act = ICE_SINGLE_ACT_LB_ENABLE;
- } else {
+ } else if (fltr->direction == ICE_ESWITCH_FLTR_EGRESS &&
+ fltr->dest_vsi == vsi->back->switchdev.uplink_vsi) {
+ /* VF to Uplink */
rule_info.sw_act.flag |= ICE_FLTR_TX;
rule_info.sw_act.src = vsi->idx;
rule_info.flags_info.act = ICE_SINGLE_ACT_LAN_ENABLE;
+ } else {
+ /* VF to VF */
+ rule_info.sw_act.flag |= ICE_FLTR_TX;
+ rule_info.sw_act.src = vsi->idx;
+ rule_info.flags_info.act = ICE_SINGLE_ACT_LB_ENABLE;
}
/* specify the cookie as filter_rule_id */
@@ -1745,16 +1778,17 @@ ice_tc_parse_action(struct ice_vsi *vsi, struct ice_tc_flower_fltr *fltr,
/**
* ice_parse_tc_flower_actions - Parse the actions for a TC filter
+ * @filter_dev: Pointer to device on which filter is being added
* @vsi: Pointer to VSI
* @cls_flower: Pointer to TC flower offload structure
* @fltr: Pointer to TC flower filter structure
*
* Parse the actions for a TC filter
*/
-static int
-ice_parse_tc_flower_actions(struct ice_vsi *vsi,
- struct flow_cls_offload *cls_flower,
- struct ice_tc_flower_fltr *fltr)
+static int ice_parse_tc_flower_actions(struct net_device *filter_dev,
+ struct ice_vsi *vsi,
+ struct flow_cls_offload *cls_flower,
+ struct ice_tc_flower_fltr *fltr)
{
struct flow_rule *rule = flow_cls_offload_flow_rule(cls_flower);
struct flow_action *flow_action = &rule->action;
@@ -1769,7 +1803,7 @@ ice_parse_tc_flower_actions(struct ice_vsi *vsi,
flow_action_for_each(i, act, flow_action) {
if (ice_is_eswitch_mode_switchdev(vsi->back))
- err = ice_eswitch_tc_parse_action(fltr, act);
+ err = ice_eswitch_tc_parse_action(filter_dev, fltr, act);
else
err = ice_tc_parse_action(vsi, fltr, act);
if (err)
@@ -1856,7 +1890,7 @@ ice_add_tc_fltr(struct net_device *netdev, struct ice_vsi *vsi,
if (err < 0)
goto err;
- err = ice_parse_tc_flower_actions(vsi, f, fltr);
+ err = ice_parse_tc_flower_actions(netdev, vsi, f, fltr);
if (err < 0)
goto err;
--
2.41.0
^ permalink raw reply related [flat|nested] 6+ messages in thread* [PATCH net 4/4] ice: Fix VF-VF direction matching in drop rule in switchdev
2023-11-07 0:48 [PATCH net 0/4][pull request] Intel Wired LAN Driver Updates 2023-11-06 (ice) Tony Nguyen
` (2 preceding siblings ...)
2023-11-07 0:48 ` [PATCH net 3/4] ice: Fix VF-VF filter rules in switchdev mode Tony Nguyen
@ 2023-11-07 0:48 ` Tony Nguyen
2023-11-09 2:50 ` [PATCH net 0/4][pull request] Intel Wired LAN Driver Updates 2023-11-06 (ice) patchwork-bot+netdevbpf
4 siblings, 0 replies; 6+ messages in thread
From: Tony Nguyen @ 2023-11-07 0:48 UTC (permalink / raw)
To: davem, kuba, pabeni, edumazet, netdev
Cc: Marcin Szycik, anthony.l.nguyen, Michal Swiatkowski,
Sujai Buvaneswaran, Simon Horman
From: Marcin Szycik <marcin.szycik@linux•intel.com>
When adding a drop rule on a VF, rule direction is not being set, which
results in it always being set to ingress (ICE_ESWITCH_FLTR_INGRESS
equals 0). Because of this, drop rules added on port representors don't
match any packets.
To fix it, set rule direction in drop action to egress when netdev is a
port representor, otherwise set it to ingress.
Fixes: 0960a27bd479 ("ice: Add direction metadata")
Reviewed-by: Michal Swiatkowski <michal.swiatkowski@linux•intel.com>
Signed-off-by: Marcin Szycik <marcin.szycik@linux•intel.com>
Tested-by: Sujai Buvaneswaran <sujai.buvaneswaran@intel•com>
Reviewed-by: Simon Horman <horms@kernel•org>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel•com>
---
drivers/net/ethernet/intel/ice/ice_tc_lib.c | 24 ++++++++++++++++++++-
1 file changed, 23 insertions(+), 1 deletion(-)
diff --git a/drivers/net/ethernet/intel/ice/ice_tc_lib.c b/drivers/net/ethernet/intel/ice/ice_tc_lib.c
index 0e75fc6b3c06..dd03cb69ad26 100644
--- a/drivers/net/ethernet/intel/ice/ice_tc_lib.c
+++ b/drivers/net/ethernet/intel/ice/ice_tc_lib.c
@@ -670,6 +670,25 @@ static int ice_tc_setup_redirect_action(struct net_device *filter_dev,
return 0;
}
+static int
+ice_tc_setup_drop_action(struct net_device *filter_dev,
+ struct ice_tc_flower_fltr *fltr)
+{
+ fltr->action.fltr_act = ICE_DROP_PACKET;
+
+ if (ice_is_port_repr_netdev(filter_dev)) {
+ fltr->direction = ICE_ESWITCH_FLTR_EGRESS;
+ } else if (ice_tc_is_dev_uplink(filter_dev)) {
+ fltr->direction = ICE_ESWITCH_FLTR_INGRESS;
+ } else {
+ NL_SET_ERR_MSG_MOD(fltr->extack,
+ "Unsupported netdevice in switchdev mode");
+ return -EINVAL;
+ }
+
+ return 0;
+}
+
static int ice_eswitch_tc_parse_action(struct net_device *filter_dev,
struct ice_tc_flower_fltr *fltr,
struct flow_action_entry *act)
@@ -678,7 +697,10 @@ static int ice_eswitch_tc_parse_action(struct net_device *filter_dev,
switch (act->id) {
case FLOW_ACTION_DROP:
- fltr->action.fltr_act = ICE_DROP_PACKET;
+ err = ice_tc_setup_drop_action(filter_dev, fltr);
+ if (err)
+ return err;
+
break;
case FLOW_ACTION_REDIRECT:
--
2.41.0
^ permalink raw reply related [flat|nested] 6+ messages in thread* Re: [PATCH net 0/4][pull request] Intel Wired LAN Driver Updates 2023-11-06 (ice)
2023-11-07 0:48 [PATCH net 0/4][pull request] Intel Wired LAN Driver Updates 2023-11-06 (ice) Tony Nguyen
` (3 preceding siblings ...)
2023-11-07 0:48 ` [PATCH net 4/4] ice: Fix VF-VF direction matching in drop rule in switchdev Tony Nguyen
@ 2023-11-09 2:50 ` patchwork-bot+netdevbpf
4 siblings, 0 replies; 6+ messages in thread
From: patchwork-bot+netdevbpf @ 2023-11-09 2:50 UTC (permalink / raw)
To: Tony Nguyen; +Cc: davem, kuba, pabeni, edumazet, netdev
Hello:
This series was applied to netdev/net.git (main)
by Tony Nguyen <anthony.l.nguyen@intel•com>:
On Mon, 6 Nov 2023 16:48:38 -0800 you wrote:
> This series contains updates to ice driver only.
>
> Dave removes SR-IOV LAG attribute for only the interface being disabled
> to allow for proper unwinding of all interfaces.
>
> Michal Schmidt changes some LAG allocations from GFP_KERNEL to GFP_ATOMIC
> due to non-allowed sleeping.
>
> [...]
Here is the summary with links:
- [net,1/4] ice: Fix SRIOV LAG disable on non-compliant aggregate
https://git.kernel.org/netdev/net/c/3e39da4fa16c
- [net,2/4] ice: lag: in RCU, use atomic allocation
https://git.kernel.org/netdev/net/c/e1db8c2a01d7
- [net,3/4] ice: Fix VF-VF filter rules in switchdev mode
https://git.kernel.org/netdev/net/c/8b3c8c55ccbc
- [net,4/4] ice: Fix VF-VF direction matching in drop rule in switchdev
https://git.kernel.org/netdev/net/c/68c51db3a16d
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] 6+ messages in thread