[PATCH net-next 0/8] mlxsw: Various updates

[PATCH net-next 0/8] mlxsw: Various updates

[Ido Schimmel]

From: Ido Schimmel <idosch@nvidia.com>

This patchset contains various updates for mlxsw. The most significant
change is the long overdue removal of the abort mechanism in the first
two patches.

Patches #1-#2 remove the route abort mechanism. This change is long
overdue and explained in detail in the commit message.

Patch #3 sets ports down in a few selftests that forgot to do so.
Discovered using a BPF tool (WIP) that monitors ASIC resources.

Patch #4 fixes an issue introduced by commit 557c4d2f780c ("selftests:
devlink_lib: add check for devlink device existence").

Patches #5-#8 modify the driver to read transceiver module's temperature
thresholds using MTMP register (when supported) instead of directly from
the module's EEPROM using MCIA register. This is both more reliable and
more efficient as now the module's temperature and thresholds are read
using one transaction instead of three.

Amit Cohen (3):
  mlxsw: spectrum_router: Remove abort mechanism
  selftests: router_scale: Do not count failed routes
  selftests: Clean forgotten resources as part of cleanup()

Mykola Kostenok (4):
  mlxsw: reg: Extend MTMP register with new threshold field
  mlxsw: core_env: Read module temperature thresholds using MTMP
    register
  mlxsw: thermal: Add function for reading module temperature and
    thresholds
  mlxsw: thermal: Read module temperature thresholds using MTMP register

Petr Machata (1):
  selftests: devlink_lib: Fix bouncing of netdevsim DEVLINK_DEV

 .../net/ethernet/mellanox/mlxsw/core_env.c    |  13 +-
 .../net/ethernet/mellanox/mlxsw/core_hwmon.c  |   6 +-
 .../ethernet/mellanox/mlxsw/core_thermal.c    |  97 ++++++++-----
 drivers/net/ethernet/mellanox/mlxsw/reg.h     |  20 ++-
 .../ethernet/mellanox/mlxsw/spectrum_router.c | 129 +-----------------
 .../ethernet/mellanox/mlxsw/spectrum_router.h |   1 -
 .../net/mlxsw/devlink_trap_l3_drops.sh        |   3 +
 .../net/mlxsw/devlink_trap_l3_exceptions.sh   |   3 +
 .../drivers/net/mlxsw/qos_dscp_bridge.sh      |   2 +
 .../drivers/net/mlxsw/router_scale.sh         |   2 +-
 .../drivers/net/netdevsim/devlink_trap.sh     |   4 +-
 .../selftests/drivers/net/netdevsim/fib.sh    |   6 +-
 .../drivers/net/netdevsim/nexthop.sh          |   4 +-
 .../drivers/net/netdevsim/psample.sh          |   4 +-
 .../selftests/net/forwarding/devlink_lib.sh   |   2 +-
 .../selftests/net/forwarding/pedit_dsfield.sh |   2 +
 .../selftests/net/forwarding/pedit_l4port.sh  |   2 +
 .../net/forwarding/skbedit_priority.sh        |   2 +
 18 files changed, 132 insertions(+), 170 deletions(-)

-- 
2.31.1

[PATCH net-next 1/8] mlxsw: spectrum_router: Remove abort mechanism

[Ido Schimmel]

From: Amit Cohen <amcohen@nvidia.com>

The abort mechanism was introduced in commit 8e05fd7166c6 ("fib: hook
IPv4 fib for hardware offload") with the purpose of falling back to
software-based routing in case of a route programming error in hardware.
The process is irreversible and requires users to reload the offloading
driver or reboot the machine.

While this approach might make sense in theory, it makes very little
sense in practice. In the case of high speed ASICs such as the Spectrum
ASIC, the abort mechanism effectively kills the machine upon a non-fatal
error such as a route programming error.

Such an extreme policy does not belong in the kernel, especially when
user space can simply try to reprogram the route following the
RTM_NEWROUTE failure notification.

Therefore, remove the abort mechanism.

Signed-off-by: Amit Cohen <amcohen@nvidia.com>
Reviewed-by: Petr Machata <petrm@nvidia.com>
Signed-off-by: Ido Schimmel <idosch@nvidia.com>
---
 .../ethernet/mellanox/mlxsw/spectrum_router.c | 129 +-----------------
 .../ethernet/mellanox/mlxsw/spectrum_router.h |   1 -
 2 files changed, 5 insertions(+), 125 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c
index 6decc5a43f98..bc47ed766878 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c
@@ -4312,9 +4312,6 @@ static void mlxsw_sp_nexthop4_event(struct mlxsw_sp *mlxsw_sp,
 	struct mlxsw_sp_nexthop_key key;
 	struct mlxsw_sp_nexthop *nh;
 
-	if (mlxsw_sp->router->aborted)
-		return;
-
 	key.fib_nh = fib_nh;
 	nh = mlxsw_sp_nexthop_lookup(mlxsw_sp, key);
 	if (!nh)
@@ -6422,9 +6419,6 @@ mlxsw_sp_router_fib4_replace(struct mlxsw_sp *mlxsw_sp,
 	struct mlxsw_sp_fib_node *fib_node;
 	int err;
 
-	if (mlxsw_sp->router->aborted)
-		return 0;
-
 	if (fen_info->fi->nh &&
 	    !mlxsw_sp_nexthop_obj_group_lookup(mlxsw_sp, fen_info->fi->nh->id))
 		return 0;
@@ -6485,9 +6479,6 @@ static int mlxsw_sp_router_fib4_del(struct mlxsw_sp *mlxsw_sp,
 	struct mlxsw_sp_fib_node *fib_node;
 	int err;
 
-	if (mlxsw_sp->router->aborted)
-		return 0;
-
 	fib4_entry = mlxsw_sp_fib4_entry_lookup(mlxsw_sp, fen_info);
 	if (!fib4_entry)
 		return 0;
@@ -7070,9 +7061,6 @@ static int mlxsw_sp_router_fib6_replace(struct mlxsw_sp *mlxsw_sp,
 	struct fib6_info *rt = rt_arr[0];
 	int err;
 
-	if (mlxsw_sp->router->aborted)
-		return 0;
-
 	if (rt->fib6_src.plen)
 		return -EINVAL;
 
@@ -7136,9 +7124,6 @@ static int mlxsw_sp_router_fib6_append(struct mlxsw_sp *mlxsw_sp,
 	struct fib6_info *rt = rt_arr[0];
 	int err;
 
-	if (mlxsw_sp->router->aborted)
-		return 0;
-
 	if (rt->fib6_src.plen)
 		return -EINVAL;
 
@@ -7180,9 +7165,6 @@ static int mlxsw_sp_router_fib6_del(struct mlxsw_sp *mlxsw_sp,
 	struct fib6_info *rt = rt_arr[0];
 	int err;
 
-	if (mlxsw_sp->router->aborted)
-		return 0;
-
 	if (mlxsw_sp_fib6_rt_should_ignore(rt))
 		return 0;
 
@@ -7211,55 +7193,6 @@ static int mlxsw_sp_router_fib6_del(struct mlxsw_sp *mlxsw_sp,
 	return err;
 }
 
-static int __mlxsw_sp_router_set_abort_trap(struct mlxsw_sp *mlxsw_sp,
-					    enum mlxsw_sp_l3proto proto,
-					    u8 tree_id)
-{
-	const struct mlxsw_sp_router_ll_ops *ll_ops = mlxsw_sp->router->proto_ll_ops[proto];
-	enum mlxsw_reg_ralxx_protocol ralxx_proto =
-				(enum mlxsw_reg_ralxx_protocol) proto;
-	struct mlxsw_sp_fib_entry_priv *priv;
-	char xralta_pl[MLXSW_REG_XRALTA_LEN];
-	char xralst_pl[MLXSW_REG_XRALST_LEN];
-	int i, err;
-
-	mlxsw_reg_xralta_pack(xralta_pl, true, ralxx_proto, tree_id);
-	err = ll_ops->ralta_write(mlxsw_sp, xralta_pl);
-	if (err)
-		return err;
-
-	mlxsw_reg_xralst_pack(xralst_pl, 0xff, tree_id);
-	err = ll_ops->ralst_write(mlxsw_sp, xralst_pl);
-	if (err)
-		return err;
-
-	for (i = 0; i < MLXSW_CORE_RES_GET(mlxsw_sp->core, MAX_VRS); i++) {
-		struct mlxsw_sp_fib_entry_op_ctx *op_ctx = mlxsw_sp->router->ll_op_ctx;
-		struct mlxsw_sp_vr *vr = &mlxsw_sp->router->vrs[i];
-		char xraltb_pl[MLXSW_REG_XRALTB_LEN];
-
-		mlxsw_sp_fib_entry_op_ctx_clear(op_ctx);
-		mlxsw_reg_xraltb_pack(xraltb_pl, vr->id, ralxx_proto, tree_id);
-		err = ll_ops->raltb_write(mlxsw_sp, xraltb_pl);
-		if (err)
-			return err;
-
-		priv = mlxsw_sp_fib_entry_priv_create(ll_ops);
-		if (IS_ERR(priv))
-			return PTR_ERR(priv);
-
-		ll_ops->fib_entry_pack(op_ctx, proto, MLXSW_SP_FIB_ENTRY_OP_WRITE,
-				       vr->id, 0, NULL, priv);
-		ll_ops->fib_entry_act_ip2me_pack(op_ctx);
-		err = ll_ops->fib_entry_commit(mlxsw_sp, op_ctx, NULL);
-		mlxsw_sp_fib_entry_priv_put(priv);
-		if (err)
-			return err;
-	}
-
-	return 0;
-}
-
 static struct mlxsw_sp_mr_table *
 mlxsw_sp_router_fibmr_family_to_table(struct mlxsw_sp_vr *vr, int family)
 {
@@ -7276,9 +7209,6 @@ static int mlxsw_sp_router_fibmr_add(struct mlxsw_sp *mlxsw_sp,
 	struct mlxsw_sp_mr_table *mrt;
 	struct mlxsw_sp_vr *vr;
 
-	if (mlxsw_sp->router->aborted)
-		return 0;
-
 	vr = mlxsw_sp_vr_get(mlxsw_sp, men_info->tb_id, NULL);
 	if (IS_ERR(vr))
 		return PTR_ERR(vr);
@@ -7293,9 +7223,6 @@ static void mlxsw_sp_router_fibmr_del(struct mlxsw_sp *mlxsw_sp,
 	struct mlxsw_sp_mr_table *mrt;
 	struct mlxsw_sp_vr *vr;
 
-	if (mlxsw_sp->router->aborted)
-		return;
-
 	vr = mlxsw_sp_vr_find(mlxsw_sp, men_info->tb_id);
 	if (WARN_ON(!vr))
 		return;
@@ -7313,9 +7240,6 @@ mlxsw_sp_router_fibmr_vif_add(struct mlxsw_sp *mlxsw_sp,
 	struct mlxsw_sp_rif *rif;
 	struct mlxsw_sp_vr *vr;
 
-	if (mlxsw_sp->router->aborted)
-		return 0;
-
 	vr = mlxsw_sp_vr_get(mlxsw_sp, ven_info->tb_id, NULL);
 	if (IS_ERR(vr))
 		return PTR_ERR(vr);
@@ -7334,9 +7258,6 @@ mlxsw_sp_router_fibmr_vif_del(struct mlxsw_sp *mlxsw_sp,
 	struct mlxsw_sp_mr_table *mrt;
 	struct mlxsw_sp_vr *vr;
 
-	if (mlxsw_sp->router->aborted)
-		return;
-
 	vr = mlxsw_sp_vr_find(mlxsw_sp, ven_info->tb_id);
 	if (WARN_ON(!vr))
 		return;
@@ -7346,25 +7267,6 @@ mlxsw_sp_router_fibmr_vif_del(struct mlxsw_sp *mlxsw_sp,
 	mlxsw_sp_vr_put(mlxsw_sp, vr);
 }
 
-static int mlxsw_sp_router_set_abort_trap(struct mlxsw_sp *mlxsw_sp)
-{
-	enum mlxsw_sp_l3proto proto = MLXSW_SP_L3_PROTO_IPV4;
-	int err;
-
-	err = __mlxsw_sp_router_set_abort_trap(mlxsw_sp, proto,
-					       MLXSW_SP_LPM_TREE_MIN);
-	if (err)
-		return err;
-
-	/* The multicast router code does not need an abort trap as by default,
-	 * packets that don't match any routes are trapped to the CPU.
-	 */
-
-	proto = MLXSW_SP_L3_PROTO_IPV6;
-	return __mlxsw_sp_router_set_abort_trap(mlxsw_sp, proto,
-						MLXSW_SP_LPM_TREE_MIN + 1);
-}
-
 static void mlxsw_sp_fib4_node_flush(struct mlxsw_sp *mlxsw_sp,
 				     struct mlxsw_sp_fib_node *fib_node)
 {
@@ -7451,20 +7353,6 @@ static void mlxsw_sp_router_fib_flush(struct mlxsw_sp *mlxsw_sp)
 	mlxsw_sp->router->adj_discard_index_valid = false;
 }
 
-static void mlxsw_sp_router_fib_abort(struct mlxsw_sp *mlxsw_sp)
-{
-	int err;
-
-	if (mlxsw_sp->router->aborted)
-		return;
-	dev_warn(mlxsw_sp->bus_info->dev, "FIB abort triggered. Note that FIB entries are no longer being offloaded to this device.\n");
-	mlxsw_sp_router_fib_flush(mlxsw_sp);
-	mlxsw_sp->router->aborted = true;
-	err = mlxsw_sp_router_set_abort_trap(mlxsw_sp);
-	if (err)
-		dev_warn(mlxsw_sp->bus_info->dev, "Failed to set abort trap.\n");
-}
-
 struct mlxsw_sp_fib6_event {
 	struct fib6_info **rt_arr;
 	unsigned int nrt6;
@@ -7546,7 +7434,7 @@ static void mlxsw_sp_router_fib4_event_process(struct mlxsw_sp *mlxsw_sp,
 		err = mlxsw_sp_router_fib4_replace(mlxsw_sp, op_ctx, &fib_event->fen_info);
 		if (err) {
 			mlxsw_sp_fib_entry_op_ctx_priv_put_all(op_ctx);
-			mlxsw_sp_router_fib_abort(mlxsw_sp);
+			dev_warn(mlxsw_sp->bus_info->dev, "FIB replace failed.\n");
 			mlxsw_sp_fib4_offload_failed_flag_set(mlxsw_sp,
 							      &fib_event->fen_info);
 		}
@@ -7581,7 +7469,7 @@ static void mlxsw_sp_router_fib6_event_process(struct mlxsw_sp *mlxsw_sp,
 						   fib_event->fib6_event.nrt6);
 		if (err) {
 			mlxsw_sp_fib_entry_op_ctx_priv_put_all(op_ctx);
-			mlxsw_sp_router_fib_abort(mlxsw_sp);
+			dev_warn(mlxsw_sp->bus_info->dev, "FIB replace failed.\n");
 			mlxsw_sp_fib6_offload_failed_flag_set(mlxsw_sp,
 							      fib6_event->rt_arr,
 							      fib6_event->nrt6);
@@ -7593,7 +7481,7 @@ static void mlxsw_sp_router_fib6_event_process(struct mlxsw_sp *mlxsw_sp,
 						  fib_event->fib6_event.nrt6);
 		if (err) {
 			mlxsw_sp_fib_entry_op_ctx_priv_put_all(op_ctx);
-			mlxsw_sp_router_fib_abort(mlxsw_sp);
+			dev_warn(mlxsw_sp->bus_info->dev, "FIB append failed.\n");
 			mlxsw_sp_fib6_offload_failed_flag_set(mlxsw_sp,
 							      fib6_event->rt_arr,
 							      fib6_event->nrt6);
@@ -7625,7 +7513,7 @@ static void mlxsw_sp_router_fibmr_event_process(struct mlxsw_sp *mlxsw_sp,
 
 		err = mlxsw_sp_router_fibmr_add(mlxsw_sp, &fib_event->men_info, replace);
 		if (err)
-			mlxsw_sp_router_fib_abort(mlxsw_sp);
+			dev_warn(mlxsw_sp->bus_info->dev, "MR entry add failed.\n");
 		mr_cache_put(fib_event->men_info.mfc);
 		break;
 	case FIB_EVENT_ENTRY_DEL:
@@ -7636,7 +7524,7 @@ static void mlxsw_sp_router_fibmr_event_process(struct mlxsw_sp *mlxsw_sp,
 		err = mlxsw_sp_router_fibmr_vif_add(mlxsw_sp,
 						    &fib_event->ven_info);
 		if (err)
-			mlxsw_sp_router_fib_abort(mlxsw_sp);
+			dev_warn(mlxsw_sp->bus_info->dev, "MR VIF add failed.\n");
 		dev_put(fib_event->ven_info.dev);
 		break;
 	case FIB_EVENT_VIF_DEL:
@@ -7800,9 +7688,6 @@ static int mlxsw_sp_router_fib_rule_event(unsigned long event,
 	if (event == FIB_EVENT_RULE_DEL)
 		return 0;
 
-	if (mlxsw_sp->router->aborted)
-		return 0;
-
 	fr_info = container_of(info, struct fib_rule_notifier_info, info);
 	rule = fr_info->rule;
 
@@ -7860,10 +7745,6 @@ static int mlxsw_sp_router_fib_event(struct notifier_block *nb,
 	case FIB_EVENT_ENTRY_ADD:
 	case FIB_EVENT_ENTRY_REPLACE:
 	case FIB_EVENT_ENTRY_APPEND:
-		if (router->aborted) {
-			NL_SET_ERR_MSG_MOD(info->extack, "FIB offload was aborted. Not configuring route");
-			return notifier_from_errno(-EINVAL);
-		}
 		if (info->family == AF_INET) {
 			struct fib_entry_notifier_info *fen_info = ptr;
 
diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.h b/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.h
index be7708a375e1..c5d7007f9173 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.h
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.h
@@ -58,7 +58,6 @@ struct mlxsw_sp_router {
 #define MLXSW_SP_UNRESOLVED_NH_PROBE_INTERVAL 5000 /* ms */
 	struct list_head nexthop_neighs_list;
 	struct list_head ipip_list;
-	bool aborted;
 	struct notifier_block nexthop_nb;
 	struct notifier_block fib_nb;
 	struct notifier_block netevent_nb;
-- 
2.31.1

[PATCH net-next 2/8] selftests: router_scale: Do not count failed routes

[Ido Schimmel]

From: Amit Cohen <amcohen@nvidia.com>

To check how many routes are installed in hardware, the test runs "ip
route" and greps for "offload", which includes routes with state
"offload_failed".

Till now, this wrong check was not found because after one failure in
route insertion, the driver moved to "abort" mode, which means that user
cannot try to add more routes.

The previous patch removed the abort mechanism and now failed routes are
counted as offloaded.

Fix this by not considering routes with "offload_failed" flag as
offloaded.

Signed-off-by: Amit Cohen <amcohen@nvidia.com>
Signed-off-by: Ido Schimmel <idosch@nvidia.com>
---
 tools/testing/selftests/drivers/net/mlxsw/router_scale.sh | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/tools/testing/selftests/drivers/net/mlxsw/router_scale.sh b/tools/testing/selftests/drivers/net/mlxsw/router_scale.sh
index e93878d42596..683759d29199 100644
--- a/tools/testing/selftests/drivers/net/mlxsw/router_scale.sh
+++ b/tools/testing/selftests/drivers/net/mlxsw/router_scale.sh
@@ -68,7 +68,7 @@ wait_for_routes()
 	local t0=$1; shift
 	local route_count=$1; shift
 
-	local t1=$(ip route | grep -o 'offload' | wc -l)
+	local t1=$(ip route | grep 'offload' | grep -v 'offload_failed' | wc -l)
 	local delta=$((t1 - t0))
 	echo $delta
 	[[ $delta -ge $route_count ]]
-- 
2.31.1

[PATCH net-next 3/8] selftests: Clean forgotten resources as part of cleanup()

[Ido Schimmel]

From: Amit Cohen <amcohen@nvidia.com>

Several tests do not set some ports down as part of their cleanup(),
resulting in IPv6 link-local addresses and associated routes not being
deleted.

These leaks were found using a BPF tool that monitors ASIC resources.

Solve this by setting the ports down at the end of the tests.

Signed-off-by: Amit Cohen <amcohen@nvidia.com>
Reviewed-by: Petr Machata <petrm@nvidia.com>
Signed-off-by: Ido Schimmel <idosch@nvidia.com>
---
 .../selftests/drivers/net/mlxsw/devlink_trap_l3_drops.sh       | 3 +++
 .../selftests/drivers/net/mlxsw/devlink_trap_l3_exceptions.sh  | 3 +++
 tools/testing/selftests/drivers/net/mlxsw/qos_dscp_bridge.sh   | 2 ++
 tools/testing/selftests/net/forwarding/pedit_dsfield.sh        | 2 ++
 tools/testing/selftests/net/forwarding/pedit_l4port.sh         | 2 ++
 tools/testing/selftests/net/forwarding/skbedit_priority.sh     | 2 ++
 6 files changed, 14 insertions(+)

diff --git a/tools/testing/selftests/drivers/net/mlxsw/devlink_trap_l3_drops.sh b/tools/testing/selftests/drivers/net/mlxsw/devlink_trap_l3_drops.sh
index 4029833f7e27..160891dcb4bc 100755
--- a/tools/testing/selftests/drivers/net/mlxsw/devlink_trap_l3_drops.sh
+++ b/tools/testing/selftests/drivers/net/mlxsw/devlink_trap_l3_drops.sh
@@ -109,6 +109,9 @@ router_destroy()
 	__addr_add_del $rp1 del 192.0.2.2/24 2001:db8:1::2/64
 
 	tc qdisc del dev $rp2 clsact
+
+	ip link set dev $rp2 down
+	ip link set dev $rp1 down
 }
 
 setup_prepare()
diff --git a/tools/testing/selftests/drivers/net/mlxsw/devlink_trap_l3_exceptions.sh b/tools/testing/selftests/drivers/net/mlxsw/devlink_trap_l3_exceptions.sh
index 42d44e27802c..190c1b6b5365 100755
--- a/tools/testing/selftests/drivers/net/mlxsw/devlink_trap_l3_exceptions.sh
+++ b/tools/testing/selftests/drivers/net/mlxsw/devlink_trap_l3_exceptions.sh
@@ -111,6 +111,9 @@ router_destroy()
 	__addr_add_del $rp1 del 192.0.2.2/24 2001:db8:1::2/64
 
 	tc qdisc del dev $rp2 clsact
+
+	ip link set dev $rp2 down
+	ip link set dev $rp1 down
 }
 
 setup_prepare()
diff --git a/tools/testing/selftests/drivers/net/mlxsw/qos_dscp_bridge.sh b/tools/testing/selftests/drivers/net/mlxsw/qos_dscp_bridge.sh
index 5cbff8038f84..28a570006d4d 100755
--- a/tools/testing/selftests/drivers/net/mlxsw/qos_dscp_bridge.sh
+++ b/tools/testing/selftests/drivers/net/mlxsw/qos_dscp_bridge.sh
@@ -93,7 +93,9 @@ switch_destroy()
 	lldptool -T -i $swp1 -V APP -d $(dscp_map 10) >/dev/null
 	lldpad_app_wait_del
 
+	ip link set dev $swp2 down
 	ip link set dev $swp2 nomaster
+	ip link set dev $swp1 down
 	ip link set dev $swp1 nomaster
 	ip link del dev br1
 }
diff --git a/tools/testing/selftests/net/forwarding/pedit_dsfield.sh b/tools/testing/selftests/net/forwarding/pedit_dsfield.sh
index 55eeacf59241..64fbd211d907 100755
--- a/tools/testing/selftests/net/forwarding/pedit_dsfield.sh
+++ b/tools/testing/selftests/net/forwarding/pedit_dsfield.sh
@@ -75,7 +75,9 @@ switch_destroy()
 	tc qdisc del dev $swp2 clsact
 	tc qdisc del dev $swp1 clsact
 
+	ip link set dev $swp2 down
 	ip link set dev $swp2 nomaster
+	ip link set dev $swp1 down
 	ip link set dev $swp1 nomaster
 	ip link del dev br1
 }
diff --git a/tools/testing/selftests/net/forwarding/pedit_l4port.sh b/tools/testing/selftests/net/forwarding/pedit_l4port.sh
index 5f20d289ee43..10e594c55117 100755
--- a/tools/testing/selftests/net/forwarding/pedit_l4port.sh
+++ b/tools/testing/selftests/net/forwarding/pedit_l4port.sh
@@ -71,7 +71,9 @@ switch_destroy()
 	tc qdisc del dev $swp2 clsact
 	tc qdisc del dev $swp1 clsact
 
+	ip link set dev $swp2 down
 	ip link set dev $swp2 nomaster
+	ip link set dev $swp1 down
 	ip link set dev $swp1 nomaster
 	ip link del dev br1
 }
diff --git a/tools/testing/selftests/net/forwarding/skbedit_priority.sh b/tools/testing/selftests/net/forwarding/skbedit_priority.sh
index e3bd8a6bb8b4..bde11dc27873 100755
--- a/tools/testing/selftests/net/forwarding/skbedit_priority.sh
+++ b/tools/testing/selftests/net/forwarding/skbedit_priority.sh
@@ -72,7 +72,9 @@ switch_destroy()
 	tc qdisc del dev $swp2 clsact
 	tc qdisc del dev $swp1 clsact
 
+	ip link set dev $swp2 down
 	ip link set dev $swp2 nomaster
+	ip link set dev $swp1 down
 	ip link set dev $swp1 nomaster
 	ip link del dev br1
 }
-- 
2.31.1

[PATCH net-next 4/8] selftests: devlink_lib: Fix bouncing of netdevsim DEVLINK_DEV

[Ido Schimmel]

From: Petr Machata <petrm@nvidia.com>

In the commit referenced below, a check was added to devlink_lib that
asserts the existence of a devlink device referenced by $DEVLINK_DEV.
Unfortunately, several netdevsim tests point DEVLINK_DEV at a device that
does not exist at the time that devlink_lib is sourced. Thus these tests
spuriously fail.

Fix this by introducing an override. By setting DEVLINK_DEV to an empty
string, the user declares their intention to handle DEVLINK_DEV management
on their own.

In all netdevsim tests that use devlink_lib and set DEVLINK_DEV, set
instead an empty DEVLINK_DEV just before sourcing devlink_lib, and set it
to the correct value right afterwards.

Fixes: 557c4d2f780c ("selftests: devlink_lib: add check for devlink device existence")
Signed-off-by: Petr Machata <petrm@nvidia.com>
Signed-off-by: Ido Schimmel <idosch@nvidia.com>
---
 .../testing/selftests/drivers/net/netdevsim/devlink_trap.sh | 4 +++-
 tools/testing/selftests/drivers/net/netdevsim/fib.sh        | 6 ++++--
 tools/testing/selftests/drivers/net/netdevsim/nexthop.sh    | 4 +++-
 tools/testing/selftests/drivers/net/netdevsim/psample.sh    | 4 +++-
 tools/testing/selftests/net/forwarding/devlink_lib.sh       | 2 +-
 5 files changed, 14 insertions(+), 6 deletions(-)

diff --git a/tools/testing/selftests/drivers/net/netdevsim/devlink_trap.sh b/tools/testing/selftests/drivers/net/netdevsim/devlink_trap.sh
index da49ad2761b5..6165901a1cf3 100755
--- a/tools/testing/selftests/drivers/net/netdevsim/devlink_trap.sh
+++ b/tools/testing/selftests/drivers/net/netdevsim/devlink_trap.sh
@@ -24,13 +24,15 @@ ALL_TESTS="
 NETDEVSIM_PATH=/sys/bus/netdevsim/
 DEV_ADDR=1337
 DEV=netdevsim${DEV_ADDR}
-DEVLINK_DEV=netdevsim/${DEV}
 DEBUGFS_DIR=/sys/kernel/debug/netdevsim/$DEV/
 SLEEP_TIME=1
 NETDEV=""
 NUM_NETIFS=0
 source $lib_dir/lib.sh
+
+DEVLINK_DEV=
 source $lib_dir/devlink_lib.sh
+DEVLINK_DEV=netdevsim/${DEV}
 
 require_command udevadm
 
diff --git a/tools/testing/selftests/drivers/net/netdevsim/fib.sh b/tools/testing/selftests/drivers/net/netdevsim/fib.sh
index 251f228ce63e..fc794cd30389 100755
--- a/tools/testing/selftests/drivers/net/netdevsim/fib.sh
+++ b/tools/testing/selftests/drivers/net/netdevsim/fib.sh
@@ -33,13 +33,15 @@ ALL_TESTS="
 NETDEVSIM_PATH=/sys/bus/netdevsim/
 DEV_ADDR=1337
 DEV=netdevsim${DEV_ADDR}
-DEVLINK_DEV=netdevsim/${DEV}
 SYSFS_NET_DIR=/sys/bus/netdevsim/devices/$DEV/net/
 NUM_NETIFS=0
 source $lib_dir/lib.sh
-source $lib_dir/devlink_lib.sh
 source $lib_dir/fib_offload_lib.sh
 
+DEVLINK_DEV=
+source $lib_dir/devlink_lib.sh
+DEVLINK_DEV=netdevsim/${DEV}
+
 ipv4_identical_routes()
 {
 	fib_ipv4_identical_routes_test "testns1"
diff --git a/tools/testing/selftests/drivers/net/netdevsim/nexthop.sh b/tools/testing/selftests/drivers/net/netdevsim/nexthop.sh
index ba75c81cda91..e8e0dc088d6a 100755
--- a/tools/testing/selftests/drivers/net/netdevsim/nexthop.sh
+++ b/tools/testing/selftests/drivers/net/netdevsim/nexthop.sh
@@ -44,12 +44,14 @@ ALL_TESTS="
 NETDEVSIM_PATH=/sys/bus/netdevsim/
 DEV_ADDR=1337
 DEV=netdevsim${DEV_ADDR}
-DEVLINK_DEV=netdevsim/${DEV}
 SYSFS_NET_DIR=/sys/bus/netdevsim/devices/$DEV/net/
 DEBUGFS_NET_DIR=/sys/kernel/debug/netdevsim/$DEV/
 NUM_NETIFS=0
 source $lib_dir/lib.sh
+
+DEVLINK_DEV=
 source $lib_dir/devlink_lib.sh
+DEVLINK_DEV=netdevsim/${DEV}
 
 nexthop_check()
 {
diff --git a/tools/testing/selftests/drivers/net/netdevsim/psample.sh b/tools/testing/selftests/drivers/net/netdevsim/psample.sh
index ee10b1a8933c..e689ff7a0b12 100755
--- a/tools/testing/selftests/drivers/net/netdevsim/psample.sh
+++ b/tools/testing/selftests/drivers/net/netdevsim/psample.sh
@@ -14,13 +14,15 @@ ALL_TESTS="
 NETDEVSIM_PATH=/sys/bus/netdevsim/
 DEV_ADDR=1337
 DEV=netdevsim${DEV_ADDR}
-DEVLINK_DEV=netdevsim/${DEV}
 SYSFS_NET_DIR=/sys/bus/netdevsim/devices/$DEV/net/
 PSAMPLE_DIR=/sys/kernel/debug/netdevsim/$DEV/psample/
 CAPTURE_FILE=$(mktemp)
 NUM_NETIFS=0
 source $lib_dir/lib.sh
+
+DEVLINK_DEV=
 source $lib_dir/devlink_lib.sh
+DEVLINK_DEV=netdevsim/${DEV}
 
 # Available at https://github.com/Mellanox/libpsample
 require_command psample
diff --git a/tools/testing/selftests/net/forwarding/devlink_lib.sh b/tools/testing/selftests/net/forwarding/devlink_lib.sh
index c19e001f138b..39fb9b8e7b58 100644
--- a/tools/testing/selftests/net/forwarding/devlink_lib.sh
+++ b/tools/testing/selftests/net/forwarding/devlink_lib.sh
@@ -18,7 +18,7 @@ if [[ ! -v DEVLINK_DEV ]]; then
 
 	DEVLINK_VIDDID=$(lspci -s $(echo $DEVLINK_DEV | cut -d"/" -f2) \
 			 -n | cut -d" " -f3)
-else
+elif [[ ! -z "$DEVLINK_DEV" ]]; then
 	devlink dev show $DEVLINK_DEV &> /dev/null
 	if [ $? -ne 0 ]; then
 		echo "SKIP: devlink device \"$DEVLINK_DEV\" not found"
-- 
2.31.1

[PATCH net-next 5/8] mlxsw: reg: Extend MTMP register with new threshold field

[Ido Schimmel]

From: Mykola Kostenok <c_mykolak@nvidia.com>

Extend Management Temperature (MTMP) register with new field specifying
the maximum temperature threshold.

Extend mlxsw_reg_mtmp_unpack() function with two extra arguments,
providing high and maximum temperature thresholds. For modules, these
thresholds correspond to critical and emergency thresholds that are read
from the module's EEPROM.

Signed-off-by: Mykola Kostenok <c_mykolak@nvidia.com>
Acked-by: Vadim Pasternak <vadimp@nvidia.com>
Signed-off-by: Ido Schimmel <idosch@nvidia.com>
---
 .../net/ethernet/mellanox/mlxsw/core_env.c    |  2 +-
 .../net/ethernet/mellanox/mlxsw/core_hwmon.c  |  6 +++---
 .../ethernet/mellanox/mlxsw/core_thermal.c    |  6 +++---
 drivers/net/ethernet/mellanox/mlxsw/reg.h     | 20 ++++++++++++++++++-
 4 files changed, 26 insertions(+), 8 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlxsw/core_env.c b/drivers/net/ethernet/mellanox/mlxsw/core_env.c
index dd26865bd587..bcad1327d861 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/core_env.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/core_env.c
@@ -142,7 +142,7 @@ int mlxsw_env_module_temp_thresholds_get(struct mlxsw_core *core, int module,
 	err = mlxsw_reg_query(core, MLXSW_REG(mtmp), mtmp_pl);
 	if (err)
 		return err;
-	mlxsw_reg_mtmp_unpack(mtmp_pl, &module_temp, NULL, NULL);
+	mlxsw_reg_mtmp_unpack(mtmp_pl, &module_temp, NULL, NULL, NULL, NULL);
 	if (!module_temp) {
 		*temp = 0;
 		return 0;
diff --git a/drivers/net/ethernet/mellanox/mlxsw/core_hwmon.c b/drivers/net/ethernet/mellanox/mlxsw/core_hwmon.c
index 2196c946698a..d41afdfbd085 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/core_hwmon.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/core_hwmon.c
@@ -72,7 +72,7 @@ static ssize_t mlxsw_hwmon_temp_show(struct device *dev,
 		dev_err(mlxsw_hwmon->bus_info->dev, "Failed to query temp sensor\n");
 		return err;
 	}
-	mlxsw_reg_mtmp_unpack(mtmp_pl, &temp, NULL, NULL);
+	mlxsw_reg_mtmp_unpack(mtmp_pl, &temp, NULL, NULL, NULL, NULL);
 	return sprintf(buf, "%d\n", temp);
 }
 
@@ -95,7 +95,7 @@ static ssize_t mlxsw_hwmon_temp_max_show(struct device *dev,
 		dev_err(mlxsw_hwmon->bus_info->dev, "Failed to query temp sensor\n");
 		return err;
 	}
-	mlxsw_reg_mtmp_unpack(mtmp_pl, NULL, &temp_max, NULL);
+	mlxsw_reg_mtmp_unpack(mtmp_pl, NULL, &temp_max, NULL, NULL, NULL);
 	return sprintf(buf, "%d\n", temp_max);
 }
 
@@ -239,7 +239,7 @@ static int mlxsw_hwmon_module_temp_get(struct device *dev,
 		dev_err(dev, "Failed to query module temperature\n");
 		return err;
 	}
-	mlxsw_reg_mtmp_unpack(mtmp_pl, p_temp, NULL, NULL);
+	mlxsw_reg_mtmp_unpack(mtmp_pl, p_temp, NULL, NULL, NULL, NULL);
 
 	return 0;
 }
diff --git a/drivers/net/ethernet/mellanox/mlxsw/core_thermal.c b/drivers/net/ethernet/mellanox/mlxsw/core_thermal.c
index 85f0ce285146..8d844d3c2e40 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/core_thermal.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/core_thermal.c
@@ -281,7 +281,7 @@ static int mlxsw_thermal_get_temp(struct thermal_zone_device *tzdev,
 		dev_err(dev, "Failed to query temp sensor\n");
 		return err;
 	}
-	mlxsw_reg_mtmp_unpack(mtmp_pl, &temp, NULL, NULL);
+	mlxsw_reg_mtmp_unpack(mtmp_pl, &temp, NULL, NULL, NULL, NULL);
 	if (temp > 0)
 		mlxsw_thermal_tz_score_update(thermal, tzdev, thermal->trips,
 					      temp);
@@ -442,7 +442,7 @@ static int mlxsw_thermal_module_temp_get(struct thermal_zone_device *tzdev,
 		*p_temp = (int) temp;
 		return 0;
 	}
-	mlxsw_reg_mtmp_unpack(mtmp_pl, &temp, NULL, NULL);
+	mlxsw_reg_mtmp_unpack(mtmp_pl, &temp, NULL, NULL, NULL, NULL);
 	*p_temp = temp;
 
 	if (!temp)
@@ -560,7 +560,7 @@ static int mlxsw_thermal_gearbox_temp_get(struct thermal_zone_device *tzdev,
 	if (err)
 		return err;
 
-	mlxsw_reg_mtmp_unpack(mtmp_pl, &temp, NULL, NULL);
+	mlxsw_reg_mtmp_unpack(mtmp_pl, &temp, NULL, NULL, NULL, NULL);
 	if (temp > 0)
 		mlxsw_thermal_tz_score_update(thermal, tzdev, tz->trips, temp);
 
diff --git a/drivers/net/ethernet/mellanox/mlxsw/reg.h b/drivers/net/ethernet/mellanox/mlxsw/reg.h
index 8eff2fc21283..93f1db3927af 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/reg.h
+++ b/drivers/net/ethernet/mellanox/mlxsw/reg.h
@@ -9463,6 +9463,14 @@ MLXSW_ITEM32(reg, mtmp, sensor_index, 0x00, 0, 12);
 					  ((s16)((GENMASK(15, 0) + (v_) + 1) \
 					   * 125)); })
 
+/* reg_mtmp_max_operational_temperature
+ * The highest temperature in the nominal operational range. Reading is in
+ * 0.125 Celsius degrees units.
+ * In case of module this is SFF critical temperature threshold.
+ * Access: RO
+ */
+MLXSW_ITEM32(reg, mtmp, max_operational_temperature, 0x04, 16, 16);
+
 /* reg_mtmp_temperature
  * Temperature reading from the sensor. Reading is in 0.125 Celsius
  * degrees units.
@@ -9541,7 +9549,9 @@ static inline void mlxsw_reg_mtmp_pack(char *payload, u16 sensor_index,
 }
 
 static inline void mlxsw_reg_mtmp_unpack(char *payload, int *p_temp,
-					 int *p_max_temp, char *sensor_name)
+					 int *p_max_temp, int *p_temp_hi,
+					 int *p_max_oper_temp,
+					 char *sensor_name)
 {
 	s16 temp;
 
@@ -9553,6 +9563,14 @@ static inline void mlxsw_reg_mtmp_unpack(char *payload, int *p_temp,
 		temp = mlxsw_reg_mtmp_max_temperature_get(payload);
 		*p_max_temp = MLXSW_REG_MTMP_TEMP_TO_MC(temp);
 	}
+	if (p_temp_hi) {
+		temp = mlxsw_reg_mtmp_temperature_threshold_hi_get(payload);
+		*p_temp_hi = MLXSW_REG_MTMP_TEMP_TO_MC(temp);
+	}
+	if (p_max_oper_temp) {
+		temp = mlxsw_reg_mtmp_max_operational_temperature_get(payload);
+		*p_max_oper_temp = MLXSW_REG_MTMP_TEMP_TO_MC(temp);
+	}
 	if (sensor_name)
 		mlxsw_reg_mtmp_sensor_name_memcpy_from(payload, sensor_name);
 }
-- 
2.31.1

[PATCH net-next 6/8] mlxsw: core_env: Read module temperature thresholds using MTMP register

[Ido Schimmel]

From: Mykola Kostenok <c_mykolak@nvidia.com>

Currently, module temperature thresholds are obtained from Management
Cable Info Access (MCIA) register by specifying the thresholds offsets
within module EEPROM layout. This data does not pass validation and in
some cases can be unreliable. For example, due to some problem with the
module.

Add support for a new feature provided by Management Temperature (MTMP)
register for sanitization of temperature thresholds values.

Extend mlxsw_env_module_temp_thresholds_get() to get temperature
thresholds through MTMP field 'max_operational_temperature' - if it is
not zero, feature is supported. Otherwise fallback to old method and get
the thresholds through MCIA.

Signed-off-by: Mykola Kostenok <c_mykolak@nvidia.com>
Acked-by: Vadim Pasternak <vadimp@nvidia.com>
Signed-off-by: Ido Schimmel <idosch@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlxsw/core_env.c | 13 +++++++++++--
 1 file changed, 11 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlxsw/core_env.c b/drivers/net/ethernet/mellanox/mlxsw/core_env.c
index bcad1327d861..b3ca5bd33a7f 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/core_env.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/core_env.c
@@ -125,6 +125,7 @@ mlxsw_env_query_module_eeprom(struct mlxsw_core *mlxsw_core, int module,
 int mlxsw_env_module_temp_thresholds_get(struct mlxsw_core *core, int module,
 					 int off, int *temp)
 {
+	unsigned int module_temp, module_crit, module_emerg;
 	char eeprom_tmp[MLXSW_REG_MCIA_EEPROM_SIZE];
 	union {
 		u8 buf[MLXSW_REG_MCIA_TH_ITEM_SIZE];
@@ -132,7 +133,6 @@ int mlxsw_env_module_temp_thresholds_get(struct mlxsw_core *core, int module,
 	} temp_thresh;
 	char mcia_pl[MLXSW_REG_MCIA_LEN] = {0};
 	char mtmp_pl[MLXSW_REG_MTMP_LEN];
-	unsigned int module_temp;
 	bool qsfp, cmis;
 	int page;
 	int err;
@@ -142,12 +142,21 @@ int mlxsw_env_module_temp_thresholds_get(struct mlxsw_core *core, int module,
 	err = mlxsw_reg_query(core, MLXSW_REG(mtmp), mtmp_pl);
 	if (err)
 		return err;
-	mlxsw_reg_mtmp_unpack(mtmp_pl, &module_temp, NULL, NULL, NULL, NULL);
+	mlxsw_reg_mtmp_unpack(mtmp_pl, &module_temp, NULL, &module_crit,
+			      &module_emerg, NULL);
 	if (!module_temp) {
 		*temp = 0;
 		return 0;
 	}
 
+	/* Validate if threshold reading is available through MTMP register,
+	 * otherwise fallback to read through MCIA.
+	 */
+	if (module_emerg) {
+		*temp = off == SFP_TEMP_HIGH_WARN ? module_crit : module_emerg;
+		return 0;
+	}
+
 	/* Read Free Side Device Temperature Thresholds from page 03h
 	 * (MSB at lower byte address).
 	 * Bytes:
-- 
2.31.1

[PATCH net-next 7/8] mlxsw: thermal: Add function for reading module temperature and thresholds

[Ido Schimmel]

From: Mykola Kostenok <c_mykolak@nvidia.com>

Provide new function mlxsw_thermal_module_temp_and_thresholds_get() for
reading temperature and temperature thresholds by a single operation.
The motivation is to reduce the number of transactions with the device
which is important when operating over a slow bus such as I2C.

Currently, the sole caller of the function is only using it to read the
module's temperature. The next patch will also use it to query the
module's temperature thresholds.

Signed-off-by: Mykola Kostenok <c_mykolak@nvidia.com>
Acked-by: Vadim Pasternak <vadimp@nvidia.com>
Signed-off-by: Ido Schimmel <idosch@nvidia.com>
---
 .../ethernet/mellanox/mlxsw/core_thermal.c    | 50 +++++++++++++------
 1 file changed, 35 insertions(+), 15 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlxsw/core_thermal.c b/drivers/net/ethernet/mellanox/mlxsw/core_thermal.c
index 8d844d3c2e40..e1ef913c81ef 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/core_thermal.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/core_thermal.c
@@ -420,29 +420,49 @@ static int mlxsw_thermal_module_unbind(struct thermal_zone_device *tzdev,
 	return err;
 }
 
+static void
+mlxsw_thermal_module_temp_and_thresholds_get(struct mlxsw_core *core,
+					     u16 sensor_index, int *p_temp,
+					     int *p_crit_temp,
+					     int *p_emerg_temp)
+{
+	char mtmp_pl[MLXSW_REG_MTMP_LEN];
+	int err;
+
+	/* Read module temperature and thresholds. */
+	mlxsw_reg_mtmp_pack(mtmp_pl, sensor_index, false, false);
+	err = mlxsw_reg_query(core, MLXSW_REG(mtmp), mtmp_pl);
+	if (err) {
+		/* Set temperature and thresholds to zero to avoid passing
+		 * uninitialized data back to the caller.
+		 */
+		*p_temp = 0;
+		*p_crit_temp = 0;
+		*p_emerg_temp = 0;
+
+		return;
+	}
+	mlxsw_reg_mtmp_unpack(mtmp_pl, p_temp, NULL, p_crit_temp, p_emerg_temp,
+			      NULL);
+}
+
 static int mlxsw_thermal_module_temp_get(struct thermal_zone_device *tzdev,
 					 int *p_temp)
 {
 	struct mlxsw_thermal_module *tz = tzdev->devdata;
 	struct mlxsw_thermal *thermal = tz->parent;
-	struct device *dev = thermal->bus_info->dev;
-	char mtmp_pl[MLXSW_REG_MTMP_LEN];
+	struct device *dev;
+	u16 sensor_index;
 	int temp;
 	int err;
 
-	/* Read module temperature. */
-	mlxsw_reg_mtmp_pack(mtmp_pl, MLXSW_REG_MTMP_MODULE_INDEX_MIN +
-			    tz->module, false, false);
-	err = mlxsw_reg_query(thermal->core, MLXSW_REG(mtmp), mtmp_pl);
-	if (err) {
-		/* Do not return error - in case of broken module's sensor
-		 * it will cause error message flooding.
-		 */
-		temp = 0;
-		*p_temp = (int) temp;
-		return 0;
-	}
-	mlxsw_reg_mtmp_unpack(mtmp_pl, &temp, NULL, NULL, NULL, NULL);
+	dev = thermal->bus_info->dev;
+	sensor_index = MLXSW_REG_MTMP_MODULE_INDEX_MIN + tz->module;
+
+	/* Read module temperature and thresholds. */
+	mlxsw_thermal_module_temp_and_thresholds_get(thermal->core,
+						     sensor_index, &temp, NULL,
+						     NULL);
 	*p_temp = temp;
 
 	if (!temp)
-- 
2.31.1

[PATCH net-next 8/8] mlxsw: thermal: Read module temperature thresholds using MTMP register

[Ido Schimmel]

From: Mykola Kostenok <c_mykolak@nvidia.com>

mlxsw_thermal_module_trips_update() is used to update the trip points of
the module's thermal zone. Currently, this is done by querying the
thresholds from the module's EEPROM via MCIA register. This data does
not pass validation and in some cases can be unreliable. For example,
due to some problem with transceiver module.

Previous patch made it possible to read module's temperature and
thresholds via MTMP register. Therefore, extend
mlxsw_thermal_module_trips_update() to use the thresholds queried from
MTMP, if valid.

This is both more reliable and more efficient than current method, as
temperature and thresholds are queried in one transaction instead of
three. This is significant when working over a slow bus such as I2C.

Signed-off-by: Mykola Kostenok <c_mykolak@nvidia.com>
Acked-by: Vadim Pasternak <vadimp@nvidia.com>
Signed-off-by: Ido Schimmel <idosch@nvidia.com>
---
 .../ethernet/mellanox/mlxsw/core_thermal.c    | 47 ++++++++++++-------
 1 file changed, 30 insertions(+), 17 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlxsw/core_thermal.c b/drivers/net/ethernet/mellanox/mlxsw/core_thermal.c
index e1ef913c81ef..d54b67d0bda7 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/core_thermal.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/core_thermal.c
@@ -149,22 +149,27 @@ mlxsw_thermal_module_trips_reset(struct mlxsw_thermal_module *tz)
 
 static int
 mlxsw_thermal_module_trips_update(struct device *dev, struct mlxsw_core *core,
-				  struct mlxsw_thermal_module *tz)
+				  struct mlxsw_thermal_module *tz,
+				  int crit_temp, int emerg_temp)
 {
-	int crit_temp, emerg_temp;
 	int err;
 
-	err = mlxsw_env_module_temp_thresholds_get(core, tz->module,
-						   SFP_TEMP_HIGH_WARN,
-						   &crit_temp);
-	if (err)
-		return err;
+	/* Do not try to query temperature thresholds directly from the module's
+	 * EEPROM if we got valid thresholds from MTMP.
+	 */
+	if (!emerg_temp || !crit_temp) {
+		err = mlxsw_env_module_temp_thresholds_get(core, tz->module,
+							   SFP_TEMP_HIGH_WARN,
+							   &crit_temp);
+		if (err)
+			return err;
 
-	err = mlxsw_env_module_temp_thresholds_get(core, tz->module,
-						   SFP_TEMP_HIGH_ALARM,
-						   &emerg_temp);
-	if (err)
-		return err;
+		err = mlxsw_env_module_temp_thresholds_get(core, tz->module,
+							   SFP_TEMP_HIGH_ALARM,
+							   &emerg_temp);
+		if (err)
+			return err;
+	}
 
 	if (crit_temp > emerg_temp) {
 		dev_warn(dev, "%s : Critical threshold %d is above emergency threshold %d\n",
@@ -451,9 +456,9 @@ static int mlxsw_thermal_module_temp_get(struct thermal_zone_device *tzdev,
 {
 	struct mlxsw_thermal_module *tz = tzdev->devdata;
 	struct mlxsw_thermal *thermal = tz->parent;
+	int temp, crit_temp, emerg_temp;
 	struct device *dev;
 	u16 sensor_index;
-	int temp;
 	int err;
 
 	dev = thermal->bus_info->dev;
@@ -461,15 +466,16 @@ static int mlxsw_thermal_module_temp_get(struct thermal_zone_device *tzdev,
 
 	/* Read module temperature and thresholds. */
 	mlxsw_thermal_module_temp_and_thresholds_get(thermal->core,
-						     sensor_index, &temp, NULL,
-						     NULL);
+						     sensor_index, &temp,
+						     &crit_temp, &emerg_temp);
 	*p_temp = temp;
 
 	if (!temp)
 		return 0;
 
 	/* Update trip points. */
-	err = mlxsw_thermal_module_trips_update(dev, thermal->core, tz);
+	err = mlxsw_thermal_module_trips_update(dev, thermal->core, tz,
+						crit_temp, emerg_temp);
 	if (!err && temp > 0)
 		mlxsw_thermal_tz_score_update(thermal, tzdev, tz->trips, temp);
 
@@ -737,7 +743,10 @@ mlxsw_thermal_module_init(struct device *dev, struct mlxsw_core *core,
 			  struct mlxsw_thermal *thermal, u8 module)
 {
 	struct mlxsw_thermal_module *module_tz;
+	int crit_temp, emerg_temp;
+	u16 sensor_index;
 
+	sensor_index = MLXSW_REG_MTMP_MODULE_INDEX_MIN + module;
 	module_tz = &thermal->tz_module_arr[module];
 	/* Skip if parent is already set (case of port split). */
 	if (module_tz->parent)
@@ -748,8 +757,12 @@ mlxsw_thermal_module_init(struct device *dev, struct mlxsw_core *core,
 	       sizeof(thermal->trips));
 	/* Initialize all trip point. */
 	mlxsw_thermal_module_trips_reset(module_tz);
+	/* Read module temperature and thresholds. */
+	mlxsw_thermal_module_temp_and_thresholds_get(core, sensor_index, NULL,
+						     &crit_temp, &emerg_temp);
 	/* Update trip point according to the module data. */
-	return mlxsw_thermal_module_trips_update(dev, core, module_tz);
+	return mlxsw_thermal_module_trips_update(dev, core, module_tz,
+						 crit_temp, emerg_temp);
 }
 
 static void mlxsw_thermal_module_fini(struct mlxsw_thermal_module *module_tz)
-- 
2.31.1

Re: [PATCH net-next 0/8] mlxsw: Various updates

[patchwork-bot+netdevbpf]

Hello:

This series was applied to netdev/net-next.git (refs/heads/master):

On Tue,  8 Jun 2021 15:44:06 +0300 you wrote:
> From: Ido Schimmel <idosch@nvidia.com>
> 
> This patchset contains various updates for mlxsw. The most significant
> change is the long overdue removal of the abort mechanism in the first
> two patches.
> 
> Patches #1-#2 remove the route abort mechanism. This change is long
> overdue and explained in detail in the commit message.
> 
> [...]

Here is the summary with links:
  - [net-next,1/8] mlxsw: spectrum_router: Remove abort mechanism
    https://git.kernel.org/netdev/net-next/c/a08a61934cfa
  - [net-next,2/8] selftests: router_scale: Do not count failed routes
    https://git.kernel.org/netdev/net-next/c/00190c2b19eb
  - [net-next,3/8] selftests: Clean forgotten resources as part of cleanup()
    https://git.kernel.org/netdev/net-next/c/e67dfb8d15de
  - [net-next,4/8] selftests: devlink_lib: Fix bouncing of netdevsim DEVLINK_DEV
    https://git.kernel.org/netdev/net-next/c/0521a262f043
  - [net-next,5/8] mlxsw: reg: Extend MTMP register with new threshold field
    https://git.kernel.org/netdev/net-next/c/314dbb19f95b
  - [net-next,6/8] mlxsw: core_env: Read module temperature thresholds using MTMP register
    https://git.kernel.org/netdev/net-next/c/befc2048088a
  - [net-next,7/8] mlxsw: thermal: Add function for reading module temperature and thresholds
    https://git.kernel.org/netdev/net-next/c/e57977b34ab5
  - [net-next,8/8] mlxsw: thermal: Read module temperature thresholds using MTMP register
    https://git.kernel.org/netdev/net-next/c/72a64c2fe9d8

You are awesome, thank you!
--
Deet-doot-dot, I am a bot.
https://korg.docs.kernel.org/patchwork/pwbot.html

Re: [PATCH net-next 0/8] mlxsw: Various updates

[patchwork-bot+netdevbpf]

Hello:

This series was applied to netdev/net-next.git (master)
by David S. Miller <davem@davemloft.net>:

On Wed,  4 May 2022 09:29:01 +0300 you wrote:
> Patches #1-#3 add missing topology diagrams in selftests and perform
> small cleanups.
> 
> Patches #4-#5 make small adjustments in QoS configuration. See detailed
> description in the commit messages.
> 
> Patches #6-#8 reduce the number of background EMAD transactions. The
> driver periodically queries the device (via EMAD transactions) about
> updates that cannot happen in certain situations. This can negatively
> impact the latency of time critical transactions, as the device is busy
> processing other transactions.
> 
> [...]

Here is the summary with links:
  - [net-next,1/8] selftests: mlxsw: bail_on_lldpad before installing the cleanup trap
    https://git.kernel.org/netdev/net-next/c/18d2c710e5df
  - [net-next,2/8] selftests: router_vid_1: Add a diagram, fix coding style
    https://git.kernel.org/netdev/net-next/c/5ade50e2df2b
  - [net-next,3/8] selftests: router.sh: Add a diagram
    https://git.kernel.org/netdev/net-next/c/faa7521add89
  - [net-next,4/8] mlxsw: spectrum_dcb: Do not warn about priority changes
    https://git.kernel.org/netdev/net-next/c/b6b584562cbe
  - [net-next,5/8] mlxsw: Treat LLDP packets as control
    https://git.kernel.org/netdev/net-next/c/0106668cd2f9
  - [net-next,6/8] mlxsw: spectrum_acl: Do not report activity for multicast routes
    https://git.kernel.org/netdev/net-next/c/d1314096fbe9
  - [net-next,7/8] mlxsw: spectrum_switchdev: Only query FDB notifications when necessary
    https://git.kernel.org/netdev/net-next/c/b8950003849d
  - [net-next,8/8] mlxsw: spectrum_router: Only query neighbour activity when necessary
    https://git.kernel.org/netdev/net-next/c/cff9437605d5

You are awesome, thank you!
-- 
Deet-doot-dot, I am a bot.
https://korg.docs.kernel.org/patchwork/pwbot.html

[PATCH net-next 0/8] mlxsw: Various updates

[Ido Schimmel]

Patches #1-#3 add missing topology diagrams in selftests and perform
small cleanups.

Patches #4-#5 make small adjustments in QoS configuration. See detailed
description in the commit messages.

Patches #6-#8 reduce the number of background EMAD transactions. The
driver periodically queries the device (via EMAD transactions) about
updates that cannot happen in certain situations. This can negatively
impact the latency of time critical transactions, as the device is busy
processing other transactions.

Before:

 # perf stat -a -e devlink:devlink_hwmsg -- sleep 10

  Performance counter stats for 'system wide':

                452      devlink:devlink_hwmsg

       10.009736160 seconds time elapsed

After:

 # perf stat -a -e devlink:devlink_hwmsg -- sleep 10

  Performance counter stats for 'system wide':

                  0      devlink:devlink_hwmsg

       10.001726333 seconds time elapsed

Ido Schimmel (3):
  mlxsw: spectrum_acl: Do not report activity for multicast routes
  mlxsw: spectrum_switchdev: Only query FDB notifications when necessary
  mlxsw: spectrum_router: Only query neighbour activity when necessary

Petr Machata (5):
  selftests: mlxsw: bail_on_lldpad before installing the cleanup trap
  selftests: router_vid_1: Add a diagram, fix coding style
  selftests: router.sh: Add a diagram
  mlxsw: spectrum_dcb: Do not warn about priority changes
  mlxsw: Treat LLDP packets as control

 .../mellanox/mlxsw/spectrum_acl_tcam.c        |  5 ++-
 .../ethernet/mellanox/mlxsw/spectrum_dcb.c    | 13 --------
 .../ethernet/mellanox/mlxsw/spectrum_router.c |  6 ++++
 .../ethernet/mellanox/mlxsw/spectrum_router.h |  1 +
 .../mellanox/mlxsw/spectrum_switchdev.c       | 31 ++++++++++++-------
 .../ethernet/mellanox/mlxsw/spectrum_trap.c   |  2 +-
 .../drivers/net/mlxsw/qos_headroom.sh         |  4 +--
 .../selftests/drivers/net/mlxsw/qos_pfc.sh    |  4 +--
 .../drivers/net/mlxsw/sch_red_ets.sh          |  5 ++-
 .../drivers/net/mlxsw/sch_red_root.sh         |  5 ++-
 .../selftests/net/forwarding/router.sh        | 18 +++++++++++
 .../selftests/net/forwarding/router_vid_1.sh  | 27 +++++++++++++++-
 12 files changed, 82 insertions(+), 39 deletions(-)

-- 
2.35.1