mv88e6xxx: Add MAB offload support
Subsystems (4)
v2 cover mv88e6xxx: Add MAB offload support
This patchset adds MAB [1] offload support in mv88e6xxx.
Patch #1: Correct default return value for mv88e6xxx_port_bridge_flags.
Patch #2: Change chip lock handling in ATU interrupt handler.
Patch #3: The MAB implementation for mv88e6xxx.
LOG:
V2: -FID reading patch already applied, so dropped here. [1]
-Patch #2 here as separate patch instead of part of MAB
implementation patch.
-Check if fid is MV88E6XXX_FID_STANDALONE, and not if
fid is zero, as that is the correct check. Do not report an error.
[1] https://git.kernel.org/netdev/net-next/c/4bf24ad09bc0
Hans J. Schultz (3):
net: dsa: mv88e6xxx: change default return of
mv88e6xxx_port_bridge_flags
net: dsa: mv88e6xxx: disable hold of chip lock for handling
net: dsa: mv88e6xxx: mac-auth/MAB implementation
drivers/net/dsa/mv88e6xxx/Makefile | 1 + drivers/net/dsa/mv88e6xxx/chip.c | 20 +++--- drivers/net/dsa/mv88e6xxx/chip.h | 15 +++++ drivers/net/dsa/mv88e6xxx/global1_atu.c | 22 +++++-- drivers/net/dsa/mv88e6xxx/switchdev.c | 83 +++++++++++++++++++++++++ drivers/net/dsa/mv88e6xxx/switchdev.h | 19 ++++++ 6 files changed, 147 insertions(+), 13 deletions(-) create mode 100644 drivers/net/dsa/mv88e6xxx/switchdev.c create mode 100644 drivers/net/dsa/mv88e6xxx/switchdev.h
Signature · 1 line
2.34.1
PATCH v2 1/3 net: dsa: mv88e6xxx: change default return of mv88e6xxx_port_bridge_flags
The default return value -EOPNOTSUPP of mv88e6xxx_port_bridge_flags() came from the return value of the DSA method port_egress_floods() in commit 4f85901f0063 ("net: dsa: mv88e6xxx: add support for bridge flags"), but the DSA API was changed in commit a8b659e7ff75 ("net: dsa: act as passthrough for bridge port flags"), resulting in the return value -EOPNOTSUPP not being valid anymore, and sections for new flags will not need to set the return value to zero on success, as with the new mab flag added in a following patch.
--- drivers/net/dsa/mv88e6xxx/chip.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/net/dsa/mv88e6xxx/chip.c +1 -1
@@ -6546,7 +6546,7 @@ static int mv88e6xxx_port_bridge_flags(struct dsa_switch *ds, int port, struct netlink_ext_ack *extack) {struct mv88e6xxx_chip *chip = ds->priv; -int err = -EOPNOTSUPP ; +int err =0 ;mv88e6xxx_reg_lock (chip);
Signature · 1 line
2.34.1
Re: [PATCH v2 net-next 1/3] net: dsa: mv88e6xxx: change default return of mv88e6xxx_port_bridge_flags
Quoted 11 lines · Hans J. Schultz
The default return value
-EOPNOTSUPPofmv88e6xxx_port_bridge_flags()came from the return value of the DSA methodport_egress_floods()in commit4f85901f0063("net: dsa: mv88e6xxx: add support for bridge flags"), but the DSA API was changed in commita8b659e7ff75("net: dsa: act as passthrough for bridge port flags"), resulting in the return value-EOPNOTSUPPnot being valid anymore, and sections for new flags will not need to set the return value to zero on success, as with the new mab flag added in a following patch.Signed-off-by Hans J. Schultz <[email protected]>---
PATCH v2 2/3 net: dsa: mv88e6xxx: disable hold of chip lock for handling
In reply to Hans J. SchultzAs functions called under the interrupt handler will need to take the netlink lock, we need to release the chip lock before calling those functions as otherwise double lock deadlocks will occur as userspace calls towards the driver often take the netlink lock and then the chip lock.
The deadlock would look like:
Interrupt handler: chip lock taken, but cannot take netlink lock as
userspace config call has netlink lock.
Userspace config: netlink lock taken, but cannot take chip lock as
the interrupt handler has the chip lock.
--- drivers/net/dsa/mv88e6xxx/global1_atu.c | 14 ++++++++------ 1 file changed, 8 insertions(+), 6 deletions(-)
drivers/net/dsa/mv88e6xxx/global1_atu.c +8 -6
@@ -409,11 +409,11 @@ static irqreturn_t mv88e6xxx_g1_atu_prob_irq_thread_fn(int irq, void *dev_id) err =mv88e6xxx_g1_read_atu_violation (chip);if (err) - goto out; + goto out_unlock; err =mv88e6xxx_g1_read (chip,MV88E6XXX_G1_ATU_OP , &val);if (err) - goto out; + goto out_unlock; err =mv88e6xxx_g1_atu_fid_read (chip, &fid);if (err)
@@ -421,11 +421,13 @@ static irqreturn_t mv88e6xxx_g1_atu_prob_irq_thread_fn(int irq, void *dev_id) err =mv88e6xxx_g1_atu_data_read (chip, &entry);if (err) - goto out; + goto out_unlock; err =mv88e6xxx_g1_atu_mac_read (chip, &entry);if (err) - goto out; + goto out_unlock; + +mv88e6xxx_reg_unlock (chip); spid = entry.state;
@@ -449,13 +451,13 @@ static irqreturn_t mv88e6xxx_g1_atu_prob_irq_thread_fn(int irq, void *dev_id) fid); chip->ports[spid].atu_full_violation++; } -mv88e6xxx_reg_unlock (chip);return IRQ_HANDLED ; -out: +out_unlock:mv88e6xxx_reg_unlock (chip); +out:dev_err (chip->dev,"ATU problem: error %d while handling interrupt\n" , err);return IRQ_HANDLED ;
Signature · 1 line
2.34.1
Vladimir Oltean
maintainerRe: [PATCH v2 net-next 2/3] net: dsa: mv88e6xxx: disable hold of chip lock for handling
Quoted 6 lines · Hans J. Schultz
As functions called under the interrupt handler will need to take the netlink lock, we need to release the chip lock before calling those functions as otherwise double lock deadlocks will occur as userspace calls towards the driver often take the netlink lock and then the chip lock.
Quoted 6 lines
The deadlock would look like:
Interrupt handler: chip lock taken, but cannot take netlink lock as
userspace config call has netlink lock.
Userspace config: netlink lock taken, but cannot take chip lock as
the interrupt handler has the chip lock.
Ultimately, none of this explanation is really relevant, and it requires too much prior reviewer knowledge. I would phrase the commit title as "shorten the locked section in mv88e6xxx_g1_atu_prob_irq_thread_fn()" and say, as an explanation, that only the hardware access functions (up until the call to mv88e6xxx_g1_atu_mac_read()) require the register lock. The follow-up code, which processes the ATU violation data, can run unlocked, and in a future patch will even run from a context which is incompatible with the register lock being held. If you wish, you can mention here as a small note that the incompatible context comes from an AB/BA ordering inversion with rtnl_lock().
Quoted 13 lines
Signed-off-by Hans J. Schultz <[email protected]>--- drivers/net/dsa/mv88e6xxx/global1_atu.c | 14 ++++++++------ 1 file changed, 8 insertions(+), 6 deletions(-)drivers/net/dsa/mv88e6xxx/global1_atu.c +0
-0@@ -409,11 +409,11 @@ static irqreturn_t mv88e6xxx_g1_atu_prob_irq_thread_fn(int irq, void *dev_id) err =mv88e6xxx_g1_read_atu_violation (chip);
drivers/net/dsa/mv88e6xxx/global1_atu.c static irqreturn_t mv88e6xxx_g1_atu_prob_irq_thread_fn(int irq, void *dev_id) · line 418
Quoted 10 lines
drivers/net/dsa/mv88e6xxx/global1_atu.c +2
-2if (err) - goto out; + goto out_unlock; err =mv88e6xxx_g1_read (chip,MV88E6XXX_G1_ATU_OP , &val);if (err) - goto out; + goto out_unlock; err =mv88e6xxx_g1_atu_fid_read (chip, &fid);
If mv88e6xxx_g1_atu_fid_read() fails, it will goto out, which will exit the IRQ handler with the mv88e6xxx_reg_lock() still held.
Probably not a good idea, since the driver will access the registers again in the future (errors in IRQ handlers aren't propagated anywhere), and the user might need a computer which is not deadlocked.
Quoted 36 lines
drivers/net/dsa/mv88e6xxx/global1_atu.c +6
-4if (err)@@ -421,11 +421,13 @@ static irqreturn_t mv88e6xxx_g1_atu_prob_irq_thread_fn(int irq, void *dev_id) err =mv88e6xxx_g1_atu_data_read (chip, &entry);if (err) - goto out; + goto out_unlock; err =mv88e6xxx_g1_atu_mac_read (chip, &entry);if (err) - goto out; + goto out_unlock; + +mv88e6xxx_reg_unlock (chip); spid = entry.state;@@ -449,13 +451,13 @@ static irqreturn_t mv88e6xxx_g1_atu_prob_irq_thread_fn(int irq, void *dev_id) fid); chip->ports[spid].atu_full_violation++; } -mv88e6xxx_reg_unlock (chip);return IRQ_HANDLED ; -out: +out_unlock:mv88e6xxx_reg_unlock (chip); +out:dev_err (chip->dev,"ATU problem: error %d while handling interrupt\n" , err);return IRQ_HANDLED ;Signature · 1 line
2.34.1
PATCH v2 3/3 net: dsa: mv88e6xxx: mac-auth/MAB implementation
In reply to Hans J. SchultzThis implementation for the Marvell mv88e6xxx chip series, is based on handling ATU miss violations occurring when packets ingress on a port that is locked with learning on. This will trigger a SWITCHDEV_FDB_ADD_TO_BRIDGE event, which will result in the bridge module adding a locked FDB entry. This bridge FDB entry will not age out as it has the extern_learn flag set.
Userspace daemons can listen to these events and either accept or deny access for the host, by either replacing the locked FDB entry with a simple entry or leave the locked entry.
If the host MAC address is already present on another port, a ATU member violation will occur, but to no real effect, and the packet will be dropped in hardware. Statistics on these violations can be shown with the command and example output of interest:
ethtool -S ethX
NIC statistics:
...
atu_member_violation: 5
atu_miss_violation: 23
...
Where ethX is the interface of the MAB enabled port.
Furthermore, as added vlan interfaces where the vid is not added to the VTU will cause ATU miss violations reporting the FID as MV88E6XXX_FID_STANDALONE, we need to check and skip the miss violations handling in this case.
--- drivers/net/dsa/mv88e6xxx/Makefile | 1 + drivers/net/dsa/mv88e6xxx/chip.c | 18 ++++-- drivers/net/dsa/mv88e6xxx/chip.h | 15 +++++ drivers/net/dsa/mv88e6xxx/global1_atu.c | 8 +++ drivers/net/dsa/mv88e6xxx/switchdev.c | 83 +++++++++++++++++++++++++ drivers/net/dsa/mv88e6xxx/switchdev.h | 19 ++++++ 6 files changed, 138 insertions(+), 6 deletions(-) create mode 100644 drivers/net/dsa/mv88e6xxx/switchdev.c create mode 100644 drivers/net/dsa/mv88e6xxx/switchdev.h
drivers/net/dsa/mv88e6xxx/Makefile +1 -0
@@ -15,6 +15,7 @@ mv88e6xxx-objs += port_hidden.o mv88e6xxx-$(CONFIG_NET_DSA_MV88E6XXX_PTP) += ptp.o mv88e6xxx-objs += serdes.o mv88e6xxx-objs += smi.o +mv88e6xxx-objs += switchdev.o mv88e6xxx-objs += trace.o # for tracing framework to find trace.h
drivers/net/dsa/mv88e6xxx/chip.c +12 -6
@@ -1729,11 +1729,11 @@ static int mv88e6xxx_vtu_get(struct mv88e6xxx_chip *chip, u16 vid,return err; } -static int mv88e6xxx_vtu_walk (struct mv88e6xxx_chip *chip, -int (*cb)(struct mv88e6xxx_chip *chip, -const struct mv88e6xxx_vtu_entry *entry, -void *priv), -void *priv) +int mv88e6xxx_vtu_walk (struct mv88e6xxx_chip *chip, +int (*cb)(struct mv88e6xxx_chip *chip, +const struct mv88e6xxx_vtu_entry *entry, +void *priv), +void *priv) {struct mv88e6xxx_vtu_entry entry = { .vid =mv88e6xxx_max_vid (chip),
@@ -6527,7 +6527,7 @@ static int mv88e6xxx_port_pre_bridge_flags(struct dsa_switch *ds, int port,const struct mv88e6xxx_ops *ops;if (flags.mask & ~(BR_LEARNING |BR_FLOOD |BR_MCAST_FLOOD | -BR_BCAST_FLOOD |BR_PORT_LOCKED )) +BR_BCAST_FLOOD |BR_PORT_LOCKED |BR_PORT_MAB ))return -EINVAL ; ops = chip->info->ops;
@@ -6585,6 +6585,12 @@ static int mv88e6xxx_port_bridge_flags(struct dsa_switch *ds, int port, goto out; } +if (flags.mask &BR_PORT_MAB ) { +bool mab = !!(flags.val &BR_PORT_MAB ); + +mv88e6xxx_port_set_mab (chip, port, mab); + } +if (flags.mask &BR_PORT_LOCKED ) {bool locked = !!(flags.val &BR_PORT_LOCKED );
drivers/net/dsa/mv88e6xxx/chip.h +15 -0
@@ -280,6 +280,9 @@ struct mv88e6xxx_port {unsigned int serdes_irq;char serdes_irq_name[64 ];struct devlink_region *region; + +/* MacAuth Bypass control flag */ +bool mab; };enum mv88e6xxx_region_id {
@@ -784,6 +787,12 @@ static inline bool mv88e6xxx_is_invalid_port(struct mv88e6xxx_chip *chip, int poreturn (chip->info->invalid_port_mask &BIT (port)) !=0 ; } +static inline void mv88e6xxx_port_set_mab (struct mv88e6xxx_chip *chip, +int port,bool mab) +{ + chip->ports[port].mab = mab; +} +int mv88e6xxx_read (struct mv88e6xxx_chip *chip,int addr,int reg,u16 *val);int mv88e6xxx_write (struct mv88e6xxx_chip *chip,int addr,int reg,u16 val); int mv88e6xxx_wait_mask(struct mv88e6xxx_chip *chip, int addr, int reg,
@@ -802,6 +811,12 @@ static inline void mv88e6xxx_reg_unlock(struct mv88e6xxx_chip *chip) mutex_unlock(&chip->reg_lock); } +int mv88e6xxx_vtu_walk (struct mv88e6xxx_chip *chip, +int (*cb)(struct mv88e6xxx_chip *chip, +const struct mv88e6xxx_vtu_entry *entry, +void *priv), +void *priv); +int mv88e6xxx_fid_map (struct mv88e6xxx_chip *chip,unsigned long *bitmap);#endif /* _MV88E6XXX_CHIP_H */
drivers/net/dsa/mv88e6xxx/global1_atu.c +8 -0
@@ -12,6 +12,7 @@#include "chip.h" #include "global1.h" +#include "switchdev.h" #include "trace.h" /* Offset 0x01: ATU FID Register */
@@ -443,6 +444,13 @@ static irqreturn_t mv88e6xxx_g1_atu_prob_irq_thread_fn(int irq, void *dev_id) entry.portvec, entry.mac, fid); chip->ports[spid].atu_miss_violation++; + +if (fid !=MV88E6XXX_FID_STANDALONE && chip->ports[spid].mab) { + err =mv88e6xxx_handle_miss_violation (chip, spid, + &entry, fid); +if (err) + goto out; + } }if (val &MV88E6XXX_G1_ATU_OP_FULL_VIOLATION ) {
drivers/net/dsa/mv88e6xxx/switchdev.c new file +83 -0
83 lines not on this page · raw message
drivers/net/dsa/mv88e6xxx/switchdev.h new file +19 -0
@@ -0,0 +1,19 @@ +/* SPDX-License-Identifier: GPL-2.0-or-later +* +* switchdev.h +* +* Authors: +* Hans J. Schultz <[email protected]> +* +*/ + +#ifndef _MV88E6XXX_SWITCHDEV_H_ +#define _MV88E6XXX_SWITCHDEV_H_ + +#include "chip.h" + +int mv88e6xxx_handle_miss_violation (struct mv88e6xxx_chip *chip,int port, +struct mv88e6xxx_atu_entry *entry, +u16 fid); + +#endif /* _MV88E6XXX_SWITCHDEV_H_ */
Signature · 1 line
2.34.1
Re: [PATCH v2 net-next 3/3] net: dsa: mv88e6xxx: mac-auth/MAB implementation
Quoted 1 line · Hans J. Schultz
This implementation for the Marvell mv88e6xxx chip series, is based on
Comma between subject and predicate?
Quoted 24 lines
handling ATU miss violations occurring when packets ingress on a port that is locked with learning on. This will trigger a
SWITCHDEV_FDB_ADD_TO_BRIDGEevent, which will result in the bridge module adding a locked FDB entry. This bridge FDB entry will not age out as it has theextern_learnflag set.Userspace daemons can listen to these events and either accept or deny access for the host, by either replacing the locked FDB entry with a simple entry or leave the locked entry.
If the host MAC address is already present on another port, a ATU member violation will occur, but to no real effect, and the packet will be dropped in hardware. Statistics on these violations can be shown with the command and example output of interest:
ethtool -S ethX
NIC statistics:
...
atu_member_violation: 5
atu_miss_violation: 23
...Where ethX is the interface of the MAB enabled port.
Quoted 7 lines
Furthermore, as added vlan interfaces where the vid is not added to the VTU will cause ATU miss violations reporting the FID as
MV88E6XXX_FID_STANDALONE, we need to check and skip the miss violations handling in this case.Signed-off-by Hans J. Schultz <[email protected]>---