2023-02-08 16:16:15

by Clément Léger

[permalink] [raw]
Subject: [PATCH net-next v3 0/3] net: dsa: rzn1-a5psw: add support for vlan and .port_bridge_flags

While adding support for VLAN, bridge_vlan_unaware.sh and
bridge_vlan_aware.sh were executed and requires .port_bridge_flags
to disable flooding on some specific port. Thus, this series adds
both vlan support and .port_bridge_flags.

----
V3:
- Target net-next tree and correct version...

V2:
- Fixed a few formatting errors
- Add .port_bridge_flags implementation

Clément Léger (3):
net: dsa: rzn1-a5psw: use a5psw_reg_rmw() to modify flooding
resolution
net: dsa: rzn1-a5psw: add support for .port_bridge_flags
net: dsa: rzn1-a5psw: add vlan support

drivers/net/dsa/rzn1_a5psw.c | 224 ++++++++++++++++++++++++++++++++++-
drivers/net/dsa/rzn1_a5psw.h | 8 +-
2 files changed, 223 insertions(+), 9 deletions(-)

--
2.39.0



2023-02-08 16:16:18

by Clément Léger

[permalink] [raw]
Subject: [PATCH net-next v3 1/3] net: dsa: rzn1-a5psw: use a5psw_reg_rmw() to modify flooding resolution

.port_bridge_flags will be added and allows to modify the flood mask
independently for each port. Keeping the existing bridged_ports write
in a5psw_flooding_set_resolution() would potentially messed up this.
Use a read-modify-write to set that value and move bridged_ports
handling in bridge_port_join/leave.

Signed-off-by: Clément Léger <[email protected]>
---
drivers/net/dsa/rzn1_a5psw.c | 12 ++++++------
1 file changed, 6 insertions(+), 6 deletions(-)

diff --git a/drivers/net/dsa/rzn1_a5psw.c b/drivers/net/dsa/rzn1_a5psw.c
index 919027cf2012..8b7d4a371f8b 100644
--- a/drivers/net/dsa/rzn1_a5psw.c
+++ b/drivers/net/dsa/rzn1_a5psw.c
@@ -299,13 +299,9 @@ static void a5psw_flooding_set_resolution(struct a5psw *a5psw, int port,
A5PSW_MCAST_DEF_MASK};
int i;

- if (set)
- a5psw->bridged_ports |= BIT(port);
- else
- a5psw->bridged_ports &= ~BIT(port);
-
for (i = 0; i < ARRAY_SIZE(offsets); i++)
- a5psw_reg_writel(a5psw, offsets[i], a5psw->bridged_ports);
+ a5psw_reg_rmw(a5psw, offsets[i], BIT(port),
+ set ? BIT(port) : 0);
}

static int a5psw_port_bridge_join(struct dsa_switch *ds, int port,
@@ -326,6 +322,8 @@ static int a5psw_port_bridge_join(struct dsa_switch *ds, int port,
a5psw_flooding_set_resolution(a5psw, port, true);
a5psw_port_mgmtfwd_set(a5psw, port, false);

+ a5psw->bridged_ports |= BIT(port);
+
return 0;
}

@@ -334,6 +332,8 @@ static void a5psw_port_bridge_leave(struct dsa_switch *ds, int port,
{
struct a5psw *a5psw = ds->priv;

+ a5psw->bridged_ports &= ~BIT(port);
+
a5psw_flooding_set_resolution(a5psw, port, false);
a5psw_port_mgmtfwd_set(a5psw, port, true);

--
2.39.0


2023-02-08 16:16:26

by Clément Léger

[permalink] [raw]
Subject: [PATCH net-next v3 2/3] net: dsa: rzn1-a5psw: add support for .port_bridge_flags

When running vlan test (bridge_vlan_aware/unaware.sh), there were some
failure due to the lack .port_bridge_flag function to disable port
flooding. Implement this operation for BR_LEARNING, BR_FLOOD,
BR_MCAST_FLOOD and BR_BCAST_FLOOD.

Signed-off-by: Clément Léger <[email protected]>
---
drivers/net/dsa/rzn1_a5psw.c | 45 ++++++++++++++++++++++++++++++++++++
1 file changed, 45 insertions(+)

diff --git a/drivers/net/dsa/rzn1_a5psw.c b/drivers/net/dsa/rzn1_a5psw.c
index 8b7d4a371f8b..0ce3948952db 100644
--- a/drivers/net/dsa/rzn1_a5psw.c
+++ b/drivers/net/dsa/rzn1_a5psw.c
@@ -342,6 +342,49 @@ static void a5psw_port_bridge_leave(struct dsa_switch *ds, int port,
a5psw->br_dev = NULL;
}

+static int a5psw_port_pre_bridge_flags(struct dsa_switch *ds, int port,
+ struct switchdev_brport_flags flags,
+ struct netlink_ext_ack *extack)
+{
+ if (flags.mask & ~(BR_LEARNING | BR_FLOOD | BR_MCAST_FLOOD |
+ BR_BCAST_FLOOD))
+ return -EINVAL;
+
+ return 0;
+}
+
+static int
+a5psw_port_bridge_flags(struct dsa_switch *ds, int port,
+ struct switchdev_brport_flags flags,
+ struct netlink_ext_ack *extack)
+{
+ struct a5psw *a5psw = ds->priv;
+ u32 val;
+
+ if (flags.mask & BR_LEARNING) {
+ val = flags.val & BR_LEARNING ? 0 : A5PSW_INPUT_LEARN_DIS(port);
+ a5psw_reg_rmw(a5psw, A5PSW_INPUT_LEARN,
+ A5PSW_INPUT_LEARN_DIS(port), val);
+ }
+
+ if (flags.mask & BR_FLOOD) {
+ val = flags.val & BR_FLOOD ? BIT(port) : 0;
+ a5psw_reg_rmw(a5psw, A5PSW_UCAST_DEF_MASK, BIT(port), val);
+ }
+
+ if (flags.mask & BR_MCAST_FLOOD) {
+ val = flags.val & BR_MCAST_FLOOD ? BIT(port) : 0;
+ a5psw_reg_rmw(a5psw, A5PSW_MCAST_DEF_MASK, BIT(port), val);
+ }
+
+ if (flags.mask & BR_BCAST_FLOOD) {
+ val = flags.val & BR_BCAST_FLOOD ? BIT(port) : 0;
+ a5psw_reg_rmw(a5psw, A5PSW_BCAST_DEF_MASK, BIT(port), val);
+ }
+
+ return 0;
+}
+
static void a5psw_port_stp_state_set(struct dsa_switch *ds, int port, u8 state)
{
u32 mask = A5PSW_INPUT_LEARN_DIS(port) | A5PSW_INPUT_LEARN_BLOCK(port);
@@ -754,6 +797,8 @@ static const struct dsa_switch_ops a5psw_switch_ops = {
.set_ageing_time = a5psw_set_ageing_time,
.port_bridge_join = a5psw_port_bridge_join,
.port_bridge_leave = a5psw_port_bridge_leave,
+ .port_pre_bridge_flags = a5psw_port_pre_bridge_flags,
+ .port_bridge_flags = a5psw_port_bridge_flags,
.port_stp_state_set = a5psw_port_stp_state_set,
.port_fast_age = a5psw_port_fast_age,
.port_fdb_add = a5psw_port_fdb_add,
--
2.39.0


2023-02-08 16:16:37

by Clément Léger

[permalink] [raw]
Subject: [PATCH net-next v3 3/3] net: dsa: rzn1-a5psw: add vlan support

Add support for vlan operation (add, del, filtering) on the RZN1
driver. The a5psw switch supports up to 32 VLAN IDs with filtering,
tagged/untagged VLANs and PVID for each ports.

Signed-off-by: Clément Léger <[email protected]>
---
drivers/net/dsa/rzn1_a5psw.c | 167 +++++++++++++++++++++++++++++++++++
drivers/net/dsa/rzn1_a5psw.h | 8 +-
2 files changed, 172 insertions(+), 3 deletions(-)

diff --git a/drivers/net/dsa/rzn1_a5psw.c b/drivers/net/dsa/rzn1_a5psw.c
index 0ce3948952db..de6b18ec647d 100644
--- a/drivers/net/dsa/rzn1_a5psw.c
+++ b/drivers/net/dsa/rzn1_a5psw.c
@@ -583,6 +583,147 @@ static int a5psw_port_fdb_dump(struct dsa_switch *ds, int port,
return ret;
}

+static int a5psw_port_vlan_filtering(struct dsa_switch *ds, int port,
+ bool vlan_filtering,
+ struct netlink_ext_ack *extack)
+{
+ u32 mask = BIT(port + A5PSW_VLAN_VERI_SHIFT) |
+ BIT(port + A5PSW_VLAN_DISC_SHIFT);
+ struct a5psw *a5psw = ds->priv;
+ u32 val = 0;
+
+ if (vlan_filtering)
+ val = BIT(port + A5PSW_VLAN_VERI_SHIFT) |
+ BIT(port + A5PSW_VLAN_DISC_SHIFT);
+
+ a5psw_reg_rmw(a5psw, A5PSW_VLAN_VERIFY, mask, val);
+
+ return 0;
+}
+
+static int a5psw_find_vlan_entry(struct a5psw *a5psw, u16 vid)
+{
+ u32 vlan_res;
+ int i;
+
+ /* Find vlan for this port */
+ for (i = 0; i < A5PSW_VLAN_COUNT; i++) {
+ vlan_res = a5psw_reg_readl(a5psw, A5PSW_VLAN_RES(i));
+ if (FIELD_GET(A5PSW_VLAN_RES_VLANID, vlan_res) == vid)
+ return i;
+ }
+
+ return -1;
+}
+
+static int a5psw_get_vlan_res_entry(struct a5psw *a5psw, u16 newvid)
+{
+ u32 vlan_res;
+ int i;
+
+ /* Find a free VLAN entry */
+ for (i = 0; i < A5PSW_VLAN_COUNT; i++) {
+ vlan_res = a5psw_reg_readl(a5psw, A5PSW_VLAN_RES(i));
+ if (!(FIELD_GET(A5PSW_VLAN_RES_PORTMASK, vlan_res))) {
+ vlan_res = FIELD_PREP(A5PSW_VLAN_RES_VLANID, newvid);
+ a5psw_reg_writel(a5psw, A5PSW_VLAN_RES(i), vlan_res);
+ return i;
+ }
+ }
+
+ return -1;
+}
+
+static void a5psw_port_vlan_tagged_cfg(struct a5psw *a5psw, int vlan_res_id,
+ int port, bool set)
+{
+ u32 mask = A5PSW_VLAN_RES_WR_PORTMASK | A5PSW_VLAN_RES_RD_TAGMASK |
+ BIT(port);
+ u32 vlan_res_off = A5PSW_VLAN_RES(vlan_res_id);
+ u32 val = A5PSW_VLAN_RES_WR_TAGMASK, reg;
+
+ if (set)
+ val |= BIT(port);
+
+ /* Toggle tag mask read */
+ a5psw_reg_writel(a5psw, vlan_res_off, A5PSW_VLAN_RES_RD_TAGMASK);
+ reg = a5psw_reg_readl(a5psw, vlan_res_off);
+ a5psw_reg_writel(a5psw, vlan_res_off, A5PSW_VLAN_RES_RD_TAGMASK);
+
+ reg &= ~mask;
+ reg |= val;
+ a5psw_reg_writel(a5psw, vlan_res_off, reg);
+}
+
+static void a5psw_port_vlan_cfg(struct a5psw *a5psw, int vlan_res_id, int port,
+ bool set)
+{
+ u32 mask = A5PSW_VLAN_RES_WR_TAGMASK | BIT(port);
+ u32 reg = A5PSW_VLAN_RES_WR_PORTMASK;
+
+ if (set)
+ reg |= BIT(port);
+
+ a5psw_reg_rmw(a5psw, A5PSW_VLAN_RES(vlan_res_id), mask, reg);
+}
+
+static int a5psw_port_vlan_add(struct dsa_switch *ds, int port,
+ const struct switchdev_obj_port_vlan *vlan,
+ struct netlink_ext_ack *extack)
+{
+ bool tagged = !(vlan->flags & BRIDGE_VLAN_INFO_UNTAGGED);
+ bool pvid = vlan->flags & BRIDGE_VLAN_INFO_PVID;
+ struct a5psw *a5psw = ds->priv;
+ u16 vid = vlan->vid;
+ int vlan_res_id;
+
+ dev_dbg(a5psw->dev, "Add VLAN %d on port %d, %s, %s\n",
+ vid, port, tagged ? "tagged" : "untagged",
+ pvid ? "PVID" : "no PVID");
+
+ vlan_res_id = a5psw_find_vlan_entry(a5psw, vid);
+ if (vlan_res_id < 0) {
+ vlan_res_id = a5psw_get_vlan_res_entry(a5psw, vid);
+ if (vlan_res_id < 0)
+ return -EINVAL;
+ }
+
+ a5psw_port_vlan_cfg(a5psw, vlan_res_id, port, true);
+ if (tagged)
+ a5psw_port_vlan_tagged_cfg(a5psw, vlan_res_id, port, true);
+
+ if (pvid) {
+ a5psw_reg_rmw(a5psw, A5PSW_VLAN_IN_MODE_ENA, BIT(port),
+ BIT(port));
+ a5psw_reg_writel(a5psw, A5PSW_SYSTEM_TAGINFO(port), vid);
+ }
+
+ return 0;
+}
+
+static int a5psw_port_vlan_del(struct dsa_switch *ds, int port,
+ const struct switchdev_obj_port_vlan *vlan)
+{
+ struct a5psw *a5psw = ds->priv;
+ u16 vid = vlan->vid;
+ int vlan_res_id;
+
+ dev_dbg(a5psw->dev, "Removing VLAN %d on port %d\n", vid, port);
+
+ vlan_res_id = a5psw_find_vlan_entry(a5psw, vid);
+ if (vlan_res_id < 0)
+ return -EINVAL;
+
+ a5psw_port_vlan_cfg(a5psw, vlan_res_id, port, false);
+ a5psw_port_vlan_tagged_cfg(a5psw, vlan_res_id, port, false);
+
+ /* Disable PVID if the vid is matching the port one */
+ if (vid == a5psw_reg_readl(a5psw, A5PSW_SYSTEM_TAGINFO(port)))
+ a5psw_reg_rmw(a5psw, A5PSW_VLAN_IN_MODE_ENA, BIT(port), 0);
+
+ return 0;
+}
+
static u64 a5psw_read_stat(struct a5psw *a5psw, u32 offset, int port)
{
u32 reg_lo, reg_hi;
@@ -700,6 +841,27 @@ static void a5psw_get_eth_ctrl_stats(struct dsa_switch *ds, int port,
ctrl_stats->MACControlFramesReceived = stat;
}

+static void a5psw_vlan_setup(struct a5psw *a5psw, int port)
+{
+ u32 reg;
+
+ /* Enable TAG always mode for the port, this is actually controlled
+ * by VLAN_IN_MODE_ENA field which will be used for PVID insertion
+ */
+ reg = A5PSW_VLAN_IN_MODE_TAG_ALWAYS;
+ reg <<= A5PSW_VLAN_IN_MODE_PORT_SHIFT(port);
+ a5psw_reg_rmw(a5psw, A5PSW_VLAN_IN_MODE, A5PSW_VLAN_IN_MODE_PORT(port),
+ reg);
+
+ /* Set transparent mode for output frame manipulation, this will depend
+ * on the VLAN_RES configuration mode
+ */
+ reg = A5PSW_VLAN_OUT_MODE_TRANSPARENT;
+ reg <<= A5PSW_VLAN_OUT_MODE_PORT_SHIFT(port);
+ a5psw_reg_rmw(a5psw, A5PSW_VLAN_OUT_MODE,
+ A5PSW_VLAN_OUT_MODE_PORT(port), reg);
+}
+
static int a5psw_setup(struct dsa_switch *ds)
{
struct a5psw *a5psw = ds->priv;
@@ -772,6 +934,8 @@ static int a5psw_setup(struct dsa_switch *ds)
/* Enable management forward only for user ports */
if (dsa_port_is_user(dp))
a5psw_port_mgmtfwd_set(a5psw, port, true);
+
+ a5psw_vlan_setup(a5psw, port);
}

return 0;
@@ -801,6 +965,9 @@ static const struct dsa_switch_ops a5psw_switch_ops = {
.port_bridge_flags = a5psw_port_bridge_flags,
.port_stp_state_set = a5psw_port_stp_state_set,
.port_fast_age = a5psw_port_fast_age,
+ .port_vlan_filtering = a5psw_port_vlan_filtering,
+ .port_vlan_add = a5psw_port_vlan_add,
+ .port_vlan_del = a5psw_port_vlan_del,
.port_fdb_add = a5psw_port_fdb_add,
.port_fdb_del = a5psw_port_fdb_del,
.port_fdb_dump = a5psw_port_fdb_dump,
diff --git a/drivers/net/dsa/rzn1_a5psw.h b/drivers/net/dsa/rzn1_a5psw.h
index c67abd49c013..2bad2e3edc2a 100644
--- a/drivers/net/dsa/rzn1_a5psw.h
+++ b/drivers/net/dsa/rzn1_a5psw.h
@@ -50,7 +50,9 @@
#define A5PSW_VLAN_IN_MODE_TAG_ALWAYS 0x2

#define A5PSW_VLAN_OUT_MODE 0x2C
-#define A5PSW_VLAN_OUT_MODE_PORT(port) (GENMASK(1, 0) << ((port) * 2))
+#define A5PSW_VLAN_OUT_MODE_PORT_SHIFT(port) ((port) * 2)
+#define A5PSW_VLAN_OUT_MODE_PORT(port) (GENMASK(1, 0) << \
+ A5PSW_VLAN_OUT_MODE_PORT_SHIFT(port))
#define A5PSW_VLAN_OUT_MODE_DIS 0x0
#define A5PSW_VLAN_OUT_MODE_STRIP 0x1
#define A5PSW_VLAN_OUT_MODE_TAG_THROUGH 0x2
@@ -59,7 +61,7 @@
#define A5PSW_VLAN_IN_MODE_ENA 0x30
#define A5PSW_VLAN_TAG_ID 0x34

-#define A5PSW_SYSTEM_TAGINFO(port) (0x200 + A5PSW_PORT_OFFSET(port))
+#define A5PSW_SYSTEM_TAGINFO(port) (0x200 + 4 * (port))

#define A5PSW_AUTH_PORT(port) (0x240 + 4 * (port))
#define A5PSW_AUTH_PORT_AUTHORIZED BIT(0)
@@ -68,7 +70,7 @@
#define A5PSW_VLAN_RES_WR_PORTMASK BIT(30)
#define A5PSW_VLAN_RES_WR_TAGMASK BIT(29)
#define A5PSW_VLAN_RES_RD_TAGMASK BIT(28)
-#define A5PSW_VLAN_RES_ID GENMASK(16, 5)
+#define A5PSW_VLAN_RES_VLANID GENMASK(16, 5)
#define A5PSW_VLAN_RES_PORTMASK GENMASK(4, 0)

#define A5PSW_RXMATCH_CONFIG(port) (0x3e80 + 4 * (port))
--
2.39.0


2023-02-08 17:27:27

by Florian Fainelli

[permalink] [raw]
Subject: Re: [PATCH net-next v3 1/3] net: dsa: rzn1-a5psw: use a5psw_reg_rmw() to modify flooding resolution

On 2/8/23 08:17, Clément Léger wrote:
> .port_bridge_flags will be added and allows to modify the flood mask
> independently for each port. Keeping the existing bridged_ports write
> in a5psw_flooding_set_resolution() would potentially messed up this.
> Use a read-modify-write to set that value and move bridged_ports
> handling in bridge_port_join/leave.
>
> Signed-off-by: Clément Léger <[email protected]>

Reviewed-by: Florian Fainelli <[email protected]>
--
Florian


2023-02-08 17:28:32

by Florian Fainelli

[permalink] [raw]
Subject: Re: [PATCH net-next v3 2/3] net: dsa: rzn1-a5psw: add support for .port_bridge_flags

On 2/8/23 08:17, Clément Léger wrote:
> When running vlan test (bridge_vlan_aware/unaware.sh), there were some
> failure due to the lack .port_bridge_flag function to disable port
> flooding. Implement this operation for BR_LEARNING, BR_FLOOD,
> BR_MCAST_FLOOD and BR_BCAST_FLOOD.
>
> Signed-off-by: Clément Léger <[email protected]>

Reviewed-by: Florian Fainelli <[email protected]>
--
Florian


2023-02-08 17:42:20

by Florian Fainelli

[permalink] [raw]
Subject: Re: [PATCH net-next v3 3/3] net: dsa: rzn1-a5psw: add vlan support

On 2/8/23 08:17, Clément Léger wrote:
> Add support for vlan operation (add, del, filtering) on the RZN1
> driver. The a5psw switch supports up to 32 VLAN IDs with filtering,
> tagged/untagged VLANs and PVID for each ports.
>
> Signed-off-by: Clément Léger <[email protected]>

This looks good to me, just a few nits/suggestions below.

> ---
> drivers/net/dsa/rzn1_a5psw.c | 167 +++++++++++++++++++++++++++++++++++
> drivers/net/dsa/rzn1_a5psw.h | 8 +-
> 2 files changed, 172 insertions(+), 3 deletions(-)
>
> diff --git a/drivers/net/dsa/rzn1_a5psw.c b/drivers/net/dsa/rzn1_a5psw.c
> index 0ce3948952db..de6b18ec647d 100644
> --- a/drivers/net/dsa/rzn1_a5psw.c
> +++ b/drivers/net/dsa/rzn1_a5psw.c
> @@ -583,6 +583,147 @@ static int a5psw_port_fdb_dump(struct dsa_switch *ds, int port,
> return ret;
> }
>
> +static int a5psw_port_vlan_filtering(struct dsa_switch *ds, int port,
> + bool vlan_filtering,
> + struct netlink_ext_ack *extack)
> +{
> + u32 mask = BIT(port + A5PSW_VLAN_VERI_SHIFT) |
> + BIT(port + A5PSW_VLAN_DISC_SHIFT);
> + struct a5psw *a5psw = ds->priv;
> + u32 val = 0;
> +
> + if (vlan_filtering)
> + val = BIT(port + A5PSW_VLAN_VERI_SHIFT) |
> + BIT(port + A5PSW_VLAN_DISC_SHIFT);
> +
> + a5psw_reg_rmw(a5psw, A5PSW_VLAN_VERIFY, mask, val);
> +
> + return 0;
> +}
> +
> +static int a5psw_find_vlan_entry(struct a5psw *a5psw, u16 vid)
> +{
> + u32 vlan_res;
> + int i;
> +
> + /* Find vlan for this port */
> + for (i = 0; i < A5PSW_VLAN_COUNT; i++) {
> + vlan_res = a5psw_reg_readl(a5psw, A5PSW_VLAN_RES(i));
> + if (FIELD_GET(A5PSW_VLAN_RES_VLANID, vlan_res) == vid)
> + return i;
> + }
> +
> + return -1;
> +}
> +
> +static int a5psw_get_vlan_res_entry(struct a5psw *a5psw, u16 newvid)

The name seems a bit misleading with respect to what it does, maybe
a5psw_new_vlan_res_entry()?

> +{
> + u32 vlan_res;
> + int i;
> +
> + /* Find a free VLAN entry */
> + for (i = 0; i < A5PSW_VLAN_COUNT; i++) {
> + vlan_res = a5psw_reg_readl(a5psw, A5PSW_VLAN_RES(i));
> + if (!(FIELD_GET(A5PSW_VLAN_RES_PORTMASK, vlan_res))) {
> + vlan_res = FIELD_PREP(A5PSW_VLAN_RES_VLANID, newvid);
> + a5psw_reg_writel(a5psw, A5PSW_VLAN_RES(i), vlan_res);
> + return i;
> + }
> + }
> +
> + return -1;
> +}
> +
> +static void a5psw_port_vlan_tagged_cfg(struct a5psw *a5psw, int vlan_res_id,

unsigned int vlan_res_id

> + int port, bool set)
> +{
> + u32 mask = A5PSW_VLAN_RES_WR_PORTMASK | A5PSW_VLAN_RES_RD_TAGMASK |
> + BIT(port);
> + u32 vlan_res_off = A5PSW_VLAN_RES(vlan_res_id);
> + u32 val = A5PSW_VLAN_RES_WR_TAGMASK, reg;
> +
> + if (set)
> + val |= BIT(port);
> +
> + /* Toggle tag mask read */
> + a5psw_reg_writel(a5psw, vlan_res_off, A5PSW_VLAN_RES_RD_TAGMASK);
> + reg = a5psw_reg_readl(a5psw, vlan_res_off);
> + a5psw_reg_writel(a5psw, vlan_res_off, A5PSW_VLAN_RES_RD_TAGMASK);
> +
> + reg &= ~mask;
> + reg |= val;
> + a5psw_reg_writel(a5psw, vlan_res_off, reg);
> +}
> +
> +static void a5psw_port_vlan_cfg(struct a5psw *a5psw, int vlan_res_id, int port,

Likewise

> + bool set)
> +{
> + u32 mask = A5PSW_VLAN_RES_WR_TAGMASK | BIT(port);
> + u32 reg = A5PSW_VLAN_RES_WR_PORTMASK;
> +
> + if (set)
> + reg |= BIT(port);
> +
> + a5psw_reg_rmw(a5psw, A5PSW_VLAN_RES(vlan_res_id), mask, reg);
> +}
> +
> +static int a5psw_port_vlan_add(struct dsa_switch *ds, int port,
> + const struct switchdev_obj_port_vlan *vlan,
> + struct netlink_ext_ack *extack)
> +{
> + bool tagged = !(vlan->flags & BRIDGE_VLAN_INFO_UNTAGGED);
> + bool pvid = vlan->flags & BRIDGE_VLAN_INFO_PVID;
> + struct a5psw *a5psw = ds->priv;
> + u16 vid = vlan->vid;
> + int vlan_res_id;
> +
> + dev_dbg(a5psw->dev, "Add VLAN %d on port %d, %s, %s\n",
> + vid, port, tagged ? "tagged" : "untagged",
> + pvid ? "PVID" : "no PVID");
> +
> + vlan_res_id = a5psw_find_vlan_entry(a5psw, vid);
> + if (vlan_res_id < 0) {
> + vlan_res_id = a5psw_get_vlan_res_entry(a5psw, vid);
> + if (vlan_res_id < 0)
> + return -EINVAL;

return -ENOSPC?

> + }
> +
> + a5psw_port_vlan_cfg(a5psw, vlan_res_id, port, true);
> + if (tagged)
> + a5psw_port_vlan_tagged_cfg(a5psw, vlan_res_id, port, true);
> +
> + if (pvid) {
> + a5psw_reg_rmw(a5psw, A5PSW_VLAN_IN_MODE_ENA, BIT(port),
> + BIT(port));
> + a5psw_reg_writel(a5psw, A5PSW_SYSTEM_TAGINFO(port), vid);
> + }
> +
> + return 0;
> +}
> +
> +static int a5psw_port_vlan_del(struct dsa_switch *ds, int port,
> + const struct switchdev_obj_port_vlan *vlan)
> +{
> + struct a5psw *a5psw = ds->priv;
> + u16 vid = vlan->vid;
> + int vlan_res_id;
> +
> + dev_dbg(a5psw->dev, "Removing VLAN %d on port %d\n", vid, port);
> +
> + vlan_res_id = a5psw_find_vlan_entry(a5psw, vid);
> + if (vlan_res_id < 0)
> + return -EINVAL;

-EINVAL looks legit here

> +
> + a5psw_port_vlan_cfg(a5psw, vlan_res_id, port, false);
> + a5psw_port_vlan_tagged_cfg(a5psw, vlan_res_id, port, false);
> +
> + /* Disable PVID if the vid is matching the port one */
> + if (vid == a5psw_reg_readl(a5psw, A5PSW_SYSTEM_TAGINFO(port)))
> + a5psw_reg_rmw(a5psw, A5PSW_VLAN_IN_MODE_ENA, BIT(port), 0);
> +
> + return 0;
> +}
> +
> static u64 a5psw_read_stat(struct a5psw *a5psw, u32 offset, int port)
> {
> u32 reg_lo, reg_hi;
> @@ -700,6 +841,27 @@ static void a5psw_get_eth_ctrl_stats(struct dsa_switch *ds, int port,
> ctrl_stats->MACControlFramesReceived = stat;
> }
>
> +static void a5psw_vlan_setup(struct a5psw *a5psw, int port)
> +{
> + u32 reg;
> +
> + /* Enable TAG always mode for the port, this is actually controlled
> + * by VLAN_IN_MODE_ENA field which will be used for PVID insertion
> + */
> + reg = A5PSW_VLAN_IN_MODE_TAG_ALWAYS;
> + reg <<= A5PSW_VLAN_IN_MODE_PORT_SHIFT(port);
> + a5psw_reg_rmw(a5psw, A5PSW_VLAN_IN_MODE, A5PSW_VLAN_IN_MODE_PORT(port),
> + reg);

If we always enable VLAN mode, which VLAN ID do switch ports not part of
a VLAN aware bridge get classified into?

> +
> + /* Set transparent mode for output frame manipulation, this will depend
> + * on the VLAN_RES configuration mode
> + */
> + reg = A5PSW_VLAN_OUT_MODE_TRANSPARENT;
> + reg <<= A5PSW_VLAN_OUT_MODE_PORT_SHIFT(port);
> + a5psw_reg_rmw(a5psw, A5PSW_VLAN_OUT_MODE,
> + A5PSW_VLAN_OUT_MODE_PORT(port), reg);

Sort of a follow-on to the previous question, what does transparent
mean? Does that mean the frames ingressing with a certain VLAN tag will
egress with the same VLAN tag in the absence of a VLAN configuration
rewriting the tag?

> +}
> +
> static int a5psw_setup(struct dsa_switch *ds)
> {
> struct a5psw *a5psw = ds->priv;
> @@ -772,6 +934,8 @@ static int a5psw_setup(struct dsa_switch *ds)
> /* Enable management forward only for user ports */
> if (dsa_port_is_user(dp))
> a5psw_port_mgmtfwd_set(a5psw, port, true);
> +
> + a5psw_vlan_setup(a5psw, port);
> }
>
> return 0;
> @@ -801,6 +965,9 @@ static const struct dsa_switch_ops a5psw_switch_ops = {
> .port_bridge_flags = a5psw_port_bridge_flags,
> .port_stp_state_set = a5psw_port_stp_state_set,
> .port_fast_age = a5psw_port_fast_age,
> + .port_vlan_filtering = a5psw_port_vlan_filtering,
> + .port_vlan_add = a5psw_port_vlan_add,
> + .port_vlan_del = a5psw_port_vlan_del,
> .port_fdb_add = a5psw_port_fdb_add,
> .port_fdb_del = a5psw_port_fdb_del,
> .port_fdb_dump = a5psw_port_fdb_dump,
> diff --git a/drivers/net/dsa/rzn1_a5psw.h b/drivers/net/dsa/rzn1_a5psw.h
> index c67abd49c013..2bad2e3edc2a 100644
> --- a/drivers/net/dsa/rzn1_a5psw.h
> +++ b/drivers/net/dsa/rzn1_a5psw.h
> @@ -50,7 +50,9 @@
> #define A5PSW_VLAN_IN_MODE_TAG_ALWAYS 0x2
>
> #define A5PSW_VLAN_OUT_MODE 0x2C
> -#define A5PSW_VLAN_OUT_MODE_PORT(port) (GENMASK(1, 0) << ((port) * 2))
> +#define A5PSW_VLAN_OUT_MODE_PORT_SHIFT(port) ((port) * 2)
> +#define A5PSW_VLAN_OUT_MODE_PORT(port) (GENMASK(1, 0) << \
> + A5PSW_VLAN_OUT_MODE_PORT_SHIFT(port))
> #define A5PSW_VLAN_OUT_MODE_DIS 0x0
> #define A5PSW_VLAN_OUT_MODE_STRIP 0x1
> #define A5PSW_VLAN_OUT_MODE_TAG_THROUGH 0x2
> @@ -59,7 +61,7 @@
> #define A5PSW_VLAN_IN_MODE_ENA 0x30
> #define A5PSW_VLAN_TAG_ID 0x34
>
> -#define A5PSW_SYSTEM_TAGINFO(port) (0x200 + A5PSW_PORT_OFFSET(port))
> +#define A5PSW_SYSTEM_TAGINFO(port) (0x200 + 4 * (port))
>
> #define A5PSW_AUTH_PORT(port) (0x240 + 4 * (port))
> #define A5PSW_AUTH_PORT_AUTHORIZED BIT(0)
> @@ -68,7 +70,7 @@
> #define A5PSW_VLAN_RES_WR_PORTMASK BIT(30)
> #define A5PSW_VLAN_RES_WR_TAGMASK BIT(29)
> #define A5PSW_VLAN_RES_RD_TAGMASK BIT(28)
> -#define A5PSW_VLAN_RES_ID GENMASK(16, 5)
> +#define A5PSW_VLAN_RES_VLANID GENMASK(16, 5)
> #define A5PSW_VLAN_RES_PORTMASK GENMASK(4, 0)
>
> #define A5PSW_RXMATCH_CONFIG(port) (0x3e80 + 4 * (port))

--
Florian


2023-02-08 21:38:08

by Vladimir Oltean

[permalink] [raw]
Subject: Re: [PATCH net-next v3 1/3] net: dsa: rzn1-a5psw: use a5psw_reg_rmw() to modify flooding resolution

On Wed, Feb 08, 2023 at 05:17:47PM +0100, Cl?ment L?ger wrote:
> .port_bridge_flags will be added and allows to modify the flood mask
> independently for each port. Keeping the existing bridged_ports write
> in a5psw_flooding_set_resolution() would potentially messed up this.
> Use a read-modify-write to set that value and move bridged_ports
> handling in bridge_port_join/leave.
>
> Signed-off-by: Cl?ment L?ger <[email protected]>
> ---
> drivers/net/dsa/rzn1_a5psw.c | 12 ++++++------
> 1 file changed, 6 insertions(+), 6 deletions(-)
>
> diff --git a/drivers/net/dsa/rzn1_a5psw.c b/drivers/net/dsa/rzn1_a5psw.c
> index 919027cf2012..8b7d4a371f8b 100644
> --- a/drivers/net/dsa/rzn1_a5psw.c
> +++ b/drivers/net/dsa/rzn1_a5psw.c
> @@ -299,13 +299,9 @@ static void a5psw_flooding_set_resolution(struct a5psw *a5psw, int port,
> A5PSW_MCAST_DEF_MASK};
> int i;
>
> - if (set)
> - a5psw->bridged_ports |= BIT(port);
> - else
> - a5psw->bridged_ports &= ~BIT(port);
> -
> for (i = 0; i < ARRAY_SIZE(offsets); i++)
> - a5psw_reg_writel(a5psw, offsets[i], a5psw->bridged_ports);
> + a5psw_reg_rmw(a5psw, offsets[i], BIT(port),
> + set ? BIT(port) : 0);
> }
>
> static int a5psw_port_bridge_join(struct dsa_switch *ds, int port,
> @@ -326,6 +322,8 @@ static int a5psw_port_bridge_join(struct dsa_switch *ds, int port,
> a5psw_flooding_set_resolution(a5psw, port, true);
> a5psw_port_mgmtfwd_set(a5psw, port, false);
>
> + a5psw->bridged_ports |= BIT(port);
> +
> return 0;
> }
>
> @@ -334,6 +332,8 @@ static void a5psw_port_bridge_leave(struct dsa_switch *ds, int port,
> {
> struct a5psw *a5psw = ds->priv;
>
> + a5psw->bridged_ports &= ~BIT(port);
> +
> a5psw_flooding_set_resolution(a5psw, port, false);
> a5psw_port_mgmtfwd_set(a5psw, port, true);
>
> --
> 2.39.0
>

What about the a5psw_flooding_set_resolution() call for the CPU port, from a5psw_setup()?
Isn't this an undocumented change? Does this logic in a5psw_port_bridge_leave() still work,
now that bridged_ports will no longer contain BIT(A5PSW_CPU_PORT)?

/* No more ports bridged */
if (a5psw->bridged_ports == BIT(A5PSW_CPU_PORT))
a5psw->br_dev = NULL;

2023-02-08 22:02:29

by Vladimir Oltean

[permalink] [raw]
Subject: Re: [PATCH net-next v3 3/3] net: dsa: rzn1-a5psw: add vlan support

On Wed, Feb 08, 2023 at 09:38:04AM -0800, Florian Fainelli wrote:
> > + /* Enable TAG always mode for the port, this is actually controlled
> > + * by VLAN_IN_MODE_ENA field which will be used for PVID insertion
> > + */
> > + reg = A5PSW_VLAN_IN_MODE_TAG_ALWAYS;
> > + reg <<= A5PSW_VLAN_IN_MODE_PORT_SHIFT(port);
> > + a5psw_reg_rmw(a5psw, A5PSW_VLAN_IN_MODE, A5PSW_VLAN_IN_MODE_PORT(port),
> > + reg);
>
> If we always enable VLAN mode, which VLAN ID do switch ports not part of a
> VLAN aware bridge get classified into?

Good question. I'd guess 0, since otherwise, the VLAN-unaware FDB
entries added with a5psw_port_fdb_add() wouldn't work.

But the driver has to survive the following chain of commands, which, by
looking at the current code structure, it doesn't:

ip link add br0 type bridge vlan_filtering 0
ip link set swp0 master br0 # PVID should remain at a value chosen privately by the driver
bridge vlan add dev swp0 vid 100 pvid untagged # PVID should not change in hardware yet
ip link set br0 type bridge vlan_filtering 1 # PVID should change to 100 now
ip link set br0 type bridge vlan_filtering 0 # PVID should change to the value chosen by the driver

Essentially, what I'm saying is that VLANs added with "bridge vlan add"
should only be active while vlan_filtering=1.

If you search for "commit_pvid" in drivers/net/dsa, you'll find a number
of drivers which have a more elaborate code structure which allows the
commands above to work properly.

2023-02-08 22:03:20

by Vladimir Oltean

[permalink] [raw]
Subject: Re: [PATCH net-next v3 3/3] net: dsa: rzn1-a5psw: add vlan support

On Wed, Feb 08, 2023 at 05:17:49PM +0100, Cl?ment L?ger wrote:
> +static void a5psw_port_vlan_tagged_cfg(struct a5psw *a5psw, int vlan_res_id,
> + int port, bool set)
> +{
> + u32 mask = A5PSW_VLAN_RES_WR_PORTMASK | A5PSW_VLAN_RES_RD_TAGMASK |
> + BIT(port);
> + u32 vlan_res_off = A5PSW_VLAN_RES(vlan_res_id);
> + u32 val = A5PSW_VLAN_RES_WR_TAGMASK, reg;
> +
> + if (set)
> + val |= BIT(port);
> +
> + /* Toggle tag mask read */
> + a5psw_reg_writel(a5psw, vlan_res_off, A5PSW_VLAN_RES_RD_TAGMASK);
> + reg = a5psw_reg_readl(a5psw, vlan_res_off);
> + a5psw_reg_writel(a5psw, vlan_res_off, A5PSW_VLAN_RES_RD_TAGMASK);

Is it intentional that this register is written twice?

> +
> + reg &= ~mask;
> + reg |= val;
> + a5psw_reg_writel(a5psw, vlan_res_off, reg);
> +}

2023-02-09 08:15:56

by Arun Ramadoss

[permalink] [raw]
Subject: Re: [PATCH net-next v3 3/3] net: dsa: rzn1-a5psw: add vlan support

Hi Clement,
On Wed, 2023-02-08 at 17:17 +0100, Clément Léger wrote:
> EXTERNAL EMAIL: Do not click links or open attachments unless you
> know the content is safe
>
> Add support for vlan operation (add, del, filtering) on the RZN1
> driver. The a5psw switch supports up to 32 VLAN IDs with filtering,
> tagged/untagged VLANs and PVID for each ports.
>
> Signed-off-by: Clément Léger <[email protected]>
> ---
> drivers/net/dsa/rzn1_a5psw.c | 167
> +++++++++++++++++++++++++++++++++++
> drivers/net/dsa/rzn1_a5psw.h | 8 +-
> 2 files changed, 172 insertions(+), 3 deletions(-)
>
> diff --git a/drivers/net/dsa/rzn1_a5psw.c
> b/drivers/net/dsa/rzn1_a5psw.c
> index 0ce3948952db..de6b18ec647d 100644
> --- a/drivers/net/dsa/rzn1_a5psw.c
> +++ b/drivers/net/dsa/rzn1_a5psw.c
> @@ -583,6 +583,147 @@ static int a5psw_port_fdb_dump(struct
> dsa_switch *ds, int port,
> return ret;
> }
>
> +static int a5psw_port_vlan_filtering(struct dsa_switch *ds, int
> port,
> + bool vlan_filtering,
> + struct netlink_ext_ack *extack)
> +{
> + u32 mask = BIT(port + A5PSW_VLAN_VERI_SHIFT) |
> + BIT(port + A5PSW_VLAN_DISC_SHIFT);

nit: if initialization of mask is separated from declaration may
increase the readability.

> + struct a5psw *a5psw = ds->priv;
> + u32 val = 0;
> +
> + if (vlan_filtering)
> + val = BIT(port + A5PSW_VLAN_VERI_SHIFT) |
> + BIT(port + A5PSW_VLAN_DISC_SHIFT);
> +
> + a5psw_reg_rmw(a5psw, A5PSW_VLAN_VERIFY, mask, val);
> +
> + return 0;
> +}
> +
> +
> +static void a5psw_port_vlan_tagged_cfg(struct a5psw *a5psw, int
> vlan_res_id,
> + int port, bool set)
> +{
> + u32 mask = A5PSW_VLAN_RES_WR_PORTMASK |
> A5PSW_VLAN_RES_RD_TAGMASK |
> + BIT(port);

same here.

> + u32 vlan_res_off = A5PSW_VLAN_RES(vlan_res_id);
> + u32 val = A5PSW_VLAN_RES_WR_TAGMASK, reg;
> +
> + if (set)
> + val |= BIT(port);
> +
> + /* Toggle tag mask read */
> + a5psw_reg_writel(a5psw, vlan_res_off,
> A5PSW_VLAN_RES_RD_TAGMASK);
> + reg = a5psw_reg_readl(a5psw, vlan_res_off);
> + a5psw_reg_writel(a5psw, vlan_res_off,
> A5PSW_VLAN_RES_RD_TAGMASK+static int a5psw_port_vlan_add(struct
> dsa_switch *ds, int port,
> > + const struct switchdev_obj_port_vlan
> > *vlan,
> > + struct netlink_ext_ack *extack)
> > +{
> > + bool tagged = !(vlan->flags & BRIDGE_VLAN_INFO_UNTAGGED);
> > + bool pvid = vlan->flags & BRIDGE_VLAN_INFO_PVID;
> > + struct a5psw *a5psw = ds->priv;
> > + u16 vid = vlan->vid;
> > + int vlan_res_id;
> > +
> > + dev_dbg(a5psw->dev, "Add VLAN %d on port %d, %s, %s\n",
> > + vid, port, tagged ? "tagged" : "untagged",
> > + pvid ? "PVID" : "no PVID");
> > +
> > + vlan_res_id = a5psw_find_vlan_entry(a5psw, vid);
> > + if (vlan_res_id < 0) {
> > + vlan_res_id = a5psw_get_vlan_res_entry(a5psw, vid);
> > + if (vlan_res_id < 0)
> > + return -EINVAL;
> > + }
> > +
> > + a5psw_port_vlan_cfg(a5psw, vlan_res_id, port, true);
> > + if (tagged)
> > + a5psw_port_vlan_tagged_cfg(a5psw, vlan_res_id,
> port,
> > true);
> > +
> > + if (pvid) {
> > + a5psw_reg_rmw(a5psw, A5PSW_VLAN_IN_MODE_ENA,
> > BIT(port),
> > + BIT(port));
> > + a5psw_reg_writel(a5psw, A5PSW_SYSTEM_TAGINFO(port),
> > vid);
> > + }
> > +
> > + return 0;
> > +}
> > +
>

2023-02-09 08:38:08

by Clément Léger

[permalink] [raw]
Subject: Re: [PATCH net-next v3 1/3] net: dsa: rzn1-a5psw: use a5psw_reg_rmw() to modify flooding resolution

Le Wed, 8 Feb 2023 23:37:57 +0200,
Vladimir Oltean <[email protected]> a écrit :

> On Wed, Feb 08, 2023 at 05:17:47PM +0100, Clément Léger wrote:
> > .port_bridge_flags will be added and allows to modify the flood mask
> > independently for each port. Keeping the existing bridged_ports write
> > in a5psw_flooding_set_resolution() would potentially messed up this.
> > Use a read-modify-write to set that value and move bridged_ports
> > handling in bridge_port_join/leave.
> >
> > Signed-off-by: Clément Léger <[email protected]>
> > ---
> > drivers/net/dsa/rzn1_a5psw.c | 12 ++++++------
> > 1 file changed, 6 insertions(+), 6 deletions(-)
> >
> > diff --git a/drivers/net/dsa/rzn1_a5psw.c b/drivers/net/dsa/rzn1_a5psw.c
> > index 919027cf2012..8b7d4a371f8b 100644
> > --- a/drivers/net/dsa/rzn1_a5psw.c
> > +++ b/drivers/net/dsa/rzn1_a5psw.c
> > @@ -299,13 +299,9 @@ static void a5psw_flooding_set_resolution(struct a5psw *a5psw, int port,
> > A5PSW_MCAST_DEF_MASK};
> > int i;
> >
> > - if (set)
> > - a5psw->bridged_ports |= BIT(port);
> > - else
> > - a5psw->bridged_ports &= ~BIT(port);
> > -
> > for (i = 0; i < ARRAY_SIZE(offsets); i++)
> > - a5psw_reg_writel(a5psw, offsets[i], a5psw->bridged_ports);
> > + a5psw_reg_rmw(a5psw, offsets[i], BIT(port),
> > + set ? BIT(port) : 0);
> > }
> >
> > static int a5psw_port_bridge_join(struct dsa_switch *ds, int port,
> > @@ -326,6 +322,8 @@ static int a5psw_port_bridge_join(struct dsa_switch *ds, int port,
> > a5psw_flooding_set_resolution(a5psw, port, true);
> > a5psw_port_mgmtfwd_set(a5psw, port, false);
> >
> > + a5psw->bridged_ports |= BIT(port);
> > +
> > return 0;
> > }
> >
> > @@ -334,6 +332,8 @@ static void a5psw_port_bridge_leave(struct dsa_switch *ds, int port,
> > {
> > struct a5psw *a5psw = ds->priv;
> >
> > + a5psw->bridged_ports &= ~BIT(port);
> > +
> > a5psw_flooding_set_resolution(a5psw, port, false);
> > a5psw_port_mgmtfwd_set(a5psw, port, true);
> >
> > --
> > 2.39.0
> >
>
> What about the a5psw_flooding_set_resolution() call for the CPU port, from a5psw_setup()?
> Isn't this an undocumented change? Does this logic in a5psw_port_bridge_leave() still work,
> now that bridged_ports will no longer contain BIT(A5PSW_CPU_PORT)?
>
> /* No more ports bridged */
> if (a5psw->bridged_ports == BIT(A5PSW_CPU_PORT))
> a5psw->br_dev = NULL;

You are right, this actually disallow to create a bridge multiple
times. I'll fix that.

--
Clément Léger,
Embedded Linux and Kernel engineer at Bootlin
https://bootlin.com

2023-02-09 08:42:12

by Clément Léger

[permalink] [raw]
Subject: Re: [PATCH net-next v3 3/3] net: dsa: rzn1-a5psw: add vlan support

Le Thu, 9 Feb 2023 00:02:19 +0200,
Vladimir Oltean <[email protected]> a écrit :

> On Wed, Feb 08, 2023 at 09:38:04AM -0800, Florian Fainelli wrote:
> > > + /* Enable TAG always mode for the port, this is actually controlled
> > > + * by VLAN_IN_MODE_ENA field which will be used for PVID insertion
> > > + */
> > > + reg = A5PSW_VLAN_IN_MODE_TAG_ALWAYS;
> > > + reg <<= A5PSW_VLAN_IN_MODE_PORT_SHIFT(port);
> > > + a5psw_reg_rmw(a5psw, A5PSW_VLAN_IN_MODE, A5PSW_VLAN_IN_MODE_PORT(port),
> > > + reg);
> >
> > If we always enable VLAN mode, which VLAN ID do switch ports not part of a
> > VLAN aware bridge get classified into?
>
> Good question. I'd guess 0, since otherwise, the VLAN-unaware FDB
> entries added with a5psw_port_fdb_add() wouldn't work.

The name of the mode is probably missleading. When setting VLAN_IN_MODE
with A5PSW_VLAN_IN_MODE_TAG_ALWAYS, the input packet will be tagged
_only_ if VLAN_IN_MODE_ENA port bit is set. If this bit is not set,
packet will passthrough transparently. This bit is actually enabled in
a5psw_port_vlan_add() when a PVID is set and unset when the PVID is
removed. Maybe the comment above these lines was not clear enough.

>
> But the driver has to survive the following chain of commands, which, by
> looking at the current code structure, it doesn't:
>
> ip link add br0 type bridge vlan_filtering 0
> ip link set swp0 master br0 # PVID should remain at a value chosen privately by the driver
> bridge vlan add dev swp0 vid 100 pvid untagged # PVID should not change in hardware yet
> ip link set br0 type bridge vlan_filtering 1 # PVID should change to 100 now
> ip link set br0 type bridge vlan_filtering 0 # PVID should change to the value chosen by the driver
>
> Essentially, what I'm saying is that VLANs added with "bridge vlan add"
> should only be active while vlan_filtering=1.
>
> If you search for "commit_pvid" in drivers/net/dsa, you'll find a number
> of drivers which have a more elaborate code structure which allows the
> commands above to work properly.



--
Clément Léger,
Embedded Linux and Kernel engineer at Bootlin
https://bootlin.com

2023-02-09 08:45:13

by Clément Léger

[permalink] [raw]
Subject: Re: [PATCH net-next v3 3/3] net: dsa: rzn1-a5psw: add vlan support

Le Thu, 9 Feb 2023 00:03:09 +0200,
Vladimir Oltean <[email protected]> a écrit :

> On Wed, Feb 08, 2023 at 05:17:49PM +0100, Clément Léger wrote:
> > +static void a5psw_port_vlan_tagged_cfg(struct a5psw *a5psw, int vlan_res_id,
> > + int port, bool set)
> > +{
> > + u32 mask = A5PSW_VLAN_RES_WR_PORTMASK | A5PSW_VLAN_RES_RD_TAGMASK |
> > + BIT(port);
> > + u32 vlan_res_off = A5PSW_VLAN_RES(vlan_res_id);
> > + u32 val = A5PSW_VLAN_RES_WR_TAGMASK, reg;
> > +
> > + if (set)
> > + val |= BIT(port);
> > +
> > + /* Toggle tag mask read */
> > + a5psw_reg_writel(a5psw, vlan_res_off, A5PSW_VLAN_RES_RD_TAGMASK);
> > + reg = a5psw_reg_readl(a5psw, vlan_res_off);
> > + a5psw_reg_writel(a5psw, vlan_res_off, A5PSW_VLAN_RES_RD_TAGMASK);
>
> Is it intentional that this register is written twice?

Yes, the A5PSW_VLAN_RES_RD_TAGMASK bit is a toggle-bit (toggled
by writing a 1 in it) and it allows to read the tagmask (for
vlan tagging) instead of the portmask (for vlan membership):

"""
b28 read_tagmask:

Select contents of mask bits (4:0) when reading the
register. If this bit is set during a write into the register, all
other bits of the write are ignored (i.e. 30,29,16:0) and the bit 28 of
the register toggles (1-> 0; 0-> 1). This is used only to allow
changing the bit 28 without changing any table contents.
"""

>
> > +
> > + reg &= ~mask;
> > + reg |= val;
> > + a5psw_reg_writel(a5psw, vlan_res_off, reg);
> > +}



--
Clément Léger,
Embedded Linux and Kernel engineer at Bootlin
https://bootlin.com

2023-02-09 13:47:54

by Clément Léger

[permalink] [raw]
Subject: Re: [PATCH net-next v3 3/3] net: dsa: rzn1-a5psw: add vlan support

Le Wed, 8 Feb 2023 09:38:04 -0800,
Florian Fainelli <[email protected]> a écrit :

> > +static void a5psw_vlan_setup(struct a5psw *a5psw, int port)
> > +{
> > + u32 reg;
> > +
> > + /* Enable TAG always mode for the port, this is actually controlled
> > + * by VLAN_IN_MODE_ENA field which will be used for PVID insertion
> > + */
> > + reg = A5PSW_VLAN_IN_MODE_TAG_ALWAYS;
> > + reg <<= A5PSW_VLAN_IN_MODE_PORT_SHIFT(port);
> > + a5psw_reg_rmw(a5psw, A5PSW_VLAN_IN_MODE, A5PSW_VLAN_IN_MODE_PORT(port),
> > + reg);
>
> If we always enable VLAN mode, which VLAN ID do switch ports not part of
> a VLAN aware bridge get classified into?

As answered on Vladimir question, it is VLAN_IN_MODE_ENAnot always VLAN enabled as
stated by the comment above but only if VLAN_IN_MODE_ENA is set (which
is done when setting a PVID only).

>
> > +
> > + /* Set transparent mode for output frame manipulation, this will depend
> > + * on the VLAN_RES configuration mode
> > + */
> > + reg = A5PSW_VLAN_OUT_MODE_TRANSPARENT;
> > + reg <<= A5PSW_VLAN_OUT_MODE_PORT_SHIFT(port);
> > + a5psw_reg_rmw(a5psw, A5PSW_VLAN_OUT_MODE,
> > + A5PSW_VLAN_OUT_MODE_PORT(port), reg);
>
> Sort of a follow-on to the previous question, what does transparent
> mean? Does that mean the frames ingressing with a certain VLAN tag will
> egress with the same VLAN tag in the absence of a VLAN configuration
> rewriting the tag?

Yes, here is an excerpt of the documentation which should clarified your
question (VLAN Table is actually stored in VLAN_RES registers):

- If frame’s VLAN id is found in the VLAN table (see Section
4.5.3.9(3)(b), VLAN Domain Resolution / VLAN Table) and the port is
defined as tagged for the VLAN, the frame is not modified.

- If frame’s VLAN id is found in the VLAN table and the port is defined
as untagged for the VLAN, the first VLAN tag is removed from the frame.

- If frame’s VLAN id is not found in the VLAN table, the frame is not
modified.

Thanks,

--
Clément Léger,
Embedded Linux and Kernel engineer at Bootlin
https://bootlin.com