[PATCH net] net: dsa: b53: prevent standalone from trying to forward to other ports

[PATCH net] net: dsa: b53: prevent standalone from trying to forward to other ports

[Jonas Gorski]

When bridged ports and standalone ports share a VLAN, e.g. via VLAN
uppers, or untagged traffic with a vlan unaware bridge, the ASIC will
still try to forward traffic to known FDB entries on standalone ports.
But since the port VLAN masks prevent forwarding to bridged ports, this
traffic will be dropped.

This e.g. can be observed in the bridge_vlan_unaware ping tests, where
this breaks pinging with learning on.

Work around this by enabling the simplified EAP mode on switches
supporting it for standalone ports, which causes the ASIC to redirect
traffic of unknown source MAC addresses to the CPU port.

Since standalone ports do not learn, there are no known source MAC
addresses, so effectively this redirects all incoming traffic to the CPU
port.

Fixes: ff39c2d68679 ("net: dsa: b53: Add bridge support")
Signed-off-by: Jonas Gorski <jonas.gorski@gmail.com>
---
 drivers/net/dsa/b53/b53_common.c | 33 ++++++++++++++++++++++++++++++++
 drivers/net/dsa/b53/b53_regs.h   | 14 ++++++++++++++
 2 files changed, 47 insertions(+)

diff --git a/drivers/net/dsa/b53/b53_common.c b/drivers/net/dsa/b53/b53_common.c
index 9eb39cfa5fb2..7216eb8f9493 100644
--- a/drivers/net/dsa/b53/b53_common.c
+++ b/drivers/net/dsa/b53/b53_common.c
@@ -326,6 +326,26 @@ static void b53_get_vlan_entry(struct b53_device *dev, u16 vid,
 	}
 }
 
+static void b53_set_eap_mode(struct b53_device *dev, int port, int mode)
+{
+	u64 eap_conf;
+
+	if (is5325(dev) || is5365(dev) || dev->chip_id == BCM5389_DEVICE_ID)
+		return;
+
+	b53_read64(dev, B53_EAP_PAGE, B53_PORT_EAP_CONF(port), &eap_conf);
+
+	if (is63xx(dev)) {
+		eap_conf &= ~EAP_MODE_MASK_63XX;
+		eap_conf |= (u64)mode << EAP_MODE_SHIFT_63XX;
+	} else {
+		eap_conf &= ~EAP_MODE_MASK;
+		eap_conf |= (u64)mode << EAP_MODE_SHIFT;
+	}
+
+	b53_write64(dev, B53_EAP_PAGE, B53_PORT_EAP_CONF(port), eap_conf);
+}
+
 static void b53_set_forwarding(struct b53_device *dev, int enable)
 {
 	u8 mgmt;
@@ -586,6 +606,13 @@ int b53_setup_port(struct dsa_switch *ds, int port)
 	b53_port_set_mcast_flood(dev, port, true);
 	b53_port_set_learning(dev, port, false);
 
+	/* Force all traffic to go to the CPU port to prevent the ASIC from
+	 * trying to forward to bridged ports on matching FDB entries, then
+	 * dropping frames because it isn't allowed to forward there.
+	 */
+	if (dsa_is_user_port(ds, port))
+		b53_set_eap_mode(dev, port, EAP_MODE_SIMPLIFIED);
+
 	return 0;
 }
 EXPORT_SYMBOL(b53_setup_port);
@@ -2042,6 +2069,9 @@ int b53_br_join(struct dsa_switch *ds, int port, struct dsa_bridge bridge,
 		pvlan |= BIT(i);
 	}
 
+	/* Disable redirection of unknown SA to the CPU port */
+	b53_set_eap_mode(dev, port, EAP_MODE_BASIC);
+
 	/* Configure the local port VLAN control membership to include
 	 * remote ports and update the local port bitmask
 	 */
@@ -2077,6 +2107,9 @@ void b53_br_leave(struct dsa_switch *ds, int port, struct dsa_bridge bridge)
 			pvlan &= ~BIT(i);
 	}
 
+	/* Enable redirection of unknown SA to the CPU port */
+	b53_set_eap_mode(dev, port, EAP_MODE_SIMPLIFIED);
+
 	b53_write16(dev, B53_PVLAN_PAGE, B53_PVLAN_PORT_MASK(port), pvlan);
 	dev->ports[port].vlan_ctl_mask = pvlan;
 
diff --git a/drivers/net/dsa/b53/b53_regs.h b/drivers/net/dsa/b53/b53_regs.h
index bfbcb66bef66..5f7a0e5c5709 100644
--- a/drivers/net/dsa/b53/b53_regs.h
+++ b/drivers/net/dsa/b53/b53_regs.h
@@ -50,6 +50,9 @@
 /* Jumbo Frame Registers */
 #define B53_JUMBO_PAGE			0x40
 
+/* EAP Registers */
+#define B53_EAP_PAGE			0x42
+
 /* EEE Control Registers Page */
 #define B53_EEE_PAGE			0x92
 
@@ -480,6 +483,17 @@
 #define   JMS_MIN_SIZE			1518
 #define   JMS_MAX_SIZE			9724
 
+/*************************************************************************
+ * EAP Page Registers
+ *************************************************************************/
+#define B53_PORT_EAP_CONF(i)		(0x20 + 8 * (i))
+#define  EAP_MODE_SHIFT			51
+#define  EAP_MODE_SHIFT_63XX		50
+#define  EAP_MODE_MASK			(0x3ull << EAP_MODE_SHIFT)
+#define  EAP_MODE_MASK_63XX		(0x3ull << EAP_MODE_SHIFT_63XX)
+#define  EAP_MODE_BASIC			0
+#define  EAP_MODE_SIMPLIFIED		3
+
 /*************************************************************************
  * EEE Configuration Page Registers
  *************************************************************************/
-- 
2.43.0

Re: [PATCH net] net: dsa: b53: prevent standalone from trying to forward to other ports

[Paolo Abeni]

On 5/8/25 11:14 AM, Jonas Gorski wrote:
> When bridged ports and standalone ports share a VLAN, e.g. via VLAN
> uppers, or untagged traffic with a vlan unaware bridge, the ASIC will
> still try to forward traffic to known FDB entries on standalone ports.
> But since the port VLAN masks prevent forwarding to bridged ports, this
> traffic will be dropped.
> 
> This e.g. can be observed in the bridge_vlan_unaware ping tests, where
> this breaks pinging with learning on.
> 
> Work around this by enabling the simplified EAP mode on switches
> supporting it for standalone ports, which causes the ASIC to redirect
> traffic of unknown source MAC addresses to the CPU port.
> 
> Since standalone ports do not learn, there are no known source MAC
> addresses, so effectively this redirects all incoming traffic to the CPU
> port.
> 
> Fixes: ff39c2d68679 ("net: dsa: b53: Add bridge support")
> Signed-off-by: Jonas Gorski <jonas.gorski@gmail.com>

@Florian, could you please have a look at this one, too?

Thanks,

Paolo

Re: [PATCH net] net: dsa: b53: prevent standalone from trying to forward to other ports

[Florian Fainelli]

On 5/8/2025 11:14 AM, Jonas Gorski wrote:
> When bridged ports and standalone ports share a VLAN, e.g. via VLAN
> uppers, or untagged traffic with a vlan unaware bridge, the ASIC will
> still try to forward traffic to known FDB entries on standalone ports.
> But since the port VLAN masks prevent forwarding to bridged ports, this
> traffic will be dropped.
> 
> This e.g. can be observed in the bridge_vlan_unaware ping tests, where
> this breaks pinging with learning on.
> 
> Work around this by enabling the simplified EAP mode on switches
> supporting it for standalone ports, which causes the ASIC to redirect
> traffic of unknown source MAC addresses to the CPU port.
> 
> Since standalone ports do not learn, there are no known source MAC
> addresses, so effectively this redirects all incoming traffic to the CPU
> port.
> 
> Fixes: ff39c2d68679 ("net: dsa: b53: Add bridge support")
> Signed-off-by: Jonas Gorski <jonas.gorski@gmail.com>

Reviewed-by: Florian Fainelli <florian.fainelli@broadcom.com>
-- 
Florian

Re: [PATCH net] net: dsa: b53: prevent standalone from trying to forward to other ports

[Vladimir Oltean]

On Thu, May 08, 2025 at 11:14:24AM +0200, Jonas Gorski wrote:
> When bridged ports and standalone ports share a VLAN, e.g. via VLAN
> uppers, or untagged traffic with a vlan unaware bridge, the ASIC will
> still try to forward traffic to known FDB entries on standalone ports.
> But since the port VLAN masks prevent forwarding to bridged ports, this
> traffic will be dropped.
> 
> This e.g. can be observed in the bridge_vlan_unaware ping tests, where
> this breaks pinging with learning on.
> 
> Work around this by enabling the simplified EAP mode on switches
> supporting it for standalone ports, which causes the ASIC to redirect
> traffic of unknown source MAC addresses to the CPU port.
> 
> Since standalone ports do not learn, there are no known source MAC
> addresses, so effectively this redirects all incoming traffic to the CPU
> port.
> 
> Fixes: ff39c2d68679 ("net: dsa: b53: Add bridge support")
> Signed-off-by: Jonas Gorski <jonas.gorski@gmail.com>
> ---

Reviewed-by: Vladimir Oltean <olteanv@gmail.com>

Re: [PATCH net] net: dsa: b53: prevent standalone from trying to forward to other ports

[patchwork-bot+netdevbpf]

Hello:

This patch was applied to netdev/net.git (main)
by Paolo Abeni <pabeni@redhat.com>:

On Thu,  8 May 2025 11:14:24 +0200 you wrote:
> When bridged ports and standalone ports share a VLAN, e.g. via VLAN
> uppers, or untagged traffic with a vlan unaware bridge, the ASIC will
> still try to forward traffic to known FDB entries on standalone ports.
> But since the port VLAN masks prevent forwarding to bridged ports, this
> traffic will be dropped.
> 
> This e.g. can be observed in the bridge_vlan_unaware ping tests, where
> this breaks pinging with learning on.
> 
> [...]

Here is the summary with links:
  - [net] net: dsa: b53: prevent standalone from trying to forward to other ports
    https://git.kernel.org/netdev/net/c/4227ea91e265

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

Re: [PATCH net] net: dsa: b53: prevent standalone from trying to forward to other ports

[Rafał Miłecki]

W dniu 8.05.2025 o 11:14, Jonas Gorski pisze:
> When bridged ports and standalone ports share a VLAN, e.g. via VLAN
> uppers, or untagged traffic with a vlan unaware bridge, the ASIC will
> still try to forward traffic to known FDB entries on standalone ports.
> But since the port VLAN masks prevent forwarding to bridged ports, this
> traffic will be dropped.
> 
> This e.g. can be observed in the bridge_vlan_unaware ping tests, where
> this breaks pinging with learning on.
> 
> Work around this by enabling the simplified EAP mode on switches
> supporting it for standalone ports, which causes the ASIC to redirect
> traffic of unknown source MAC addresses to the CPU port.
> 
> Since standalone ports do not learn, there are no known source MAC
> addresses, so effectively this redirects all incoming traffic to the CPU
> port.

This change broke standalone ports (those not being part of bridge) on Northstar (BCM5301X) devices. We got reports from users about WAN ports on routers not working anymore:
https://github.com/openwrt/openwrt/issues/21187
https://github.com/openwrt/openwrt/issues/21349

Can you take another look at those changes, see if something me be wrong / missing?


> Fixes: ff39c2d68679 ("net: dsa: b53: Add bridge support")
> Signed-off-by: Jonas Gorski <jonas.gorski@gmail.com>
> ---
>   drivers/net/dsa/b53/b53_common.c | 33 ++++++++++++++++++++++++++++++++
>   drivers/net/dsa/b53/b53_regs.h   | 14 ++++++++++++++
>   2 files changed, 47 insertions(+)
> 
> diff --git a/drivers/net/dsa/b53/b53_common.c b/drivers/net/dsa/b53/b53_common.c
> index 9eb39cfa5fb2..7216eb8f9493 100644
> --- a/drivers/net/dsa/b53/b53_common.c
> +++ b/drivers/net/dsa/b53/b53_common.c
> @@ -326,6 +326,26 @@ static void b53_get_vlan_entry(struct b53_device *dev, u16 vid,
>   	}
>   }
>   
> +static void b53_set_eap_mode(struct b53_device *dev, int port, int mode)
> +{
> +	u64 eap_conf;
> +
> +	if (is5325(dev) || is5365(dev) || dev->chip_id == BCM5389_DEVICE_ID)
> +		return;
> +
> +	b53_read64(dev, B53_EAP_PAGE, B53_PORT_EAP_CONF(port), &eap_conf);
> +
> +	if (is63xx(dev)) {
> +		eap_conf &= ~EAP_MODE_MASK_63XX;
> +		eap_conf |= (u64)mode << EAP_MODE_SHIFT_63XX;
> +	} else {
> +		eap_conf &= ~EAP_MODE_MASK;
> +		eap_conf |= (u64)mode << EAP_MODE_SHIFT;
> +	}
> +
> +	b53_write64(dev, B53_EAP_PAGE, B53_PORT_EAP_CONF(port), eap_conf);
> +}
> +
>   static void b53_set_forwarding(struct b53_device *dev, int enable)
>   {
>   	u8 mgmt;
> @@ -586,6 +606,13 @@ int b53_setup_port(struct dsa_switch *ds, int port)
>   	b53_port_set_mcast_flood(dev, port, true);
>   	b53_port_set_learning(dev, port, false);
>   
> +	/* Force all traffic to go to the CPU port to prevent the ASIC from
> +	 * trying to forward to bridged ports on matching FDB entries, then
> +	 * dropping frames because it isn't allowed to forward there.
> +	 */
> +	if (dsa_is_user_port(ds, port))
> +		b53_set_eap_mode(dev, port, EAP_MODE_SIMPLIFIED);
> +
>   	return 0;
>   }
>   EXPORT_SYMBOL(b53_setup_port);
> @@ -2042,6 +2069,9 @@ int b53_br_join(struct dsa_switch *ds, int port, struct dsa_bridge bridge,
>   		pvlan |= BIT(i);
>   	}
>   
> +	/* Disable redirection of unknown SA to the CPU port */
> +	b53_set_eap_mode(dev, port, EAP_MODE_BASIC);
> +
>   	/* Configure the local port VLAN control membership to include
>   	 * remote ports and update the local port bitmask
>   	 */
> @@ -2077,6 +2107,9 @@ void b53_br_leave(struct dsa_switch *ds, int port, struct dsa_bridge bridge)
>   			pvlan &= ~BIT(i);
>   	}
>   
> +	/* Enable redirection of unknown SA to the CPU port */
> +	b53_set_eap_mode(dev, port, EAP_MODE_SIMPLIFIED);
> +
>   	b53_write16(dev, B53_PVLAN_PAGE, B53_PVLAN_PORT_MASK(port), pvlan);
>   	dev->ports[port].vlan_ctl_mask = pvlan;
>   
> diff --git a/drivers/net/dsa/b53/b53_regs.h b/drivers/net/dsa/b53/b53_regs.h
> index bfbcb66bef66..5f7a0e5c5709 100644
> --- a/drivers/net/dsa/b53/b53_regs.h
> +++ b/drivers/net/dsa/b53/b53_regs.h
> @@ -50,6 +50,9 @@
>   /* Jumbo Frame Registers */
>   #define B53_JUMBO_PAGE			0x40
>   
> +/* EAP Registers */
> +#define B53_EAP_PAGE			0x42
> +
>   /* EEE Control Registers Page */
>   #define B53_EEE_PAGE			0x92
>   
> @@ -480,6 +483,17 @@
>   #define   JMS_MIN_SIZE			1518
>   #define   JMS_MAX_SIZE			9724
>   
> +/*************************************************************************
> + * EAP Page Registers
> + *************************************************************************/
> +#define B53_PORT_EAP_CONF(i)		(0x20 + 8 * (i))
> +#define  EAP_MODE_SHIFT			51
> +#define  EAP_MODE_SHIFT_63XX		50
> +#define  EAP_MODE_MASK			(0x3ull << EAP_MODE_SHIFT)
> +#define  EAP_MODE_MASK_63XX		(0x3ull << EAP_MODE_SHIFT_63XX)
> +#define  EAP_MODE_BASIC			0
> +#define  EAP_MODE_SIMPLIFIED		3
> +
>   /*************************************************************************
>    * EEE Configuration Page Registers
>    *************************************************************************/

Re: [PATCH net] net: dsa: b53: prevent standalone from trying to forward to other ports

[Florian Fainelli]

On 2/26/26 12:24, Rafał Miłecki wrote:
> W dniu 8.05.2025 o 11:14, Jonas Gorski pisze:
>> When bridged ports and standalone ports share a VLAN, e.g. via VLAN
>> uppers, or untagged traffic with a vlan unaware bridge, the ASIC will
>> still try to forward traffic to known FDB entries on standalone ports.
>> But since the port VLAN masks prevent forwarding to bridged ports, this
>> traffic will be dropped.
>>
>> This e.g. can be observed in the bridge_vlan_unaware ping tests, where
>> this breaks pinging with learning on.
>>
>> Work around this by enabling the simplified EAP mode on switches
>> supporting it for standalone ports, which causes the ASIC to redirect
>> traffic of unknown source MAC addresses to the CPU port.
>>
>> Since standalone ports do not learn, there are no known source MAC
>> addresses, so effectively this redirects all incoming traffic to the CPU
>> port.
> 
> This change broke standalone ports (those not being part of bridge) on 
> Northstar (BCM5301X) devices. We got reports from users about WAN ports 
> on routers not working anymore:
> https://github.com/openwrt/openwrt/issues/21187
> https://github.com/openwrt/openwrt/issues/21349
> 
> Can you take another look at those changes, see if something me be 
> wrong / missing?

Checking the BCM5301X datasheet, the register definitions look correct.

The first report appears to be for a Netgear R7000 which AFAICT would 
utilize port 8 of the switch as the CPU port, which would be correct.

The other report is for a Netgear R6250 where port 5 would be acting as 
the CPU port, that could be an issue unless the switch is also 
programmed to use port 5 as the IMP port.

Could you try to program GMNGCFG (page 0x02, offset 0) bits 7:6 to have 
the value 0b11 which is dual IMP port (IMPO is port 8 and IMP1 is port 
5) and see if that helps?
-- 
Florian

Re: [PATCH net] net: dsa: b53: prevent standalone from trying to forward to other ports

[Rafał Miłecki]

czw., 26 lut 2026 o 22:26 Florian Fainelli
<florian.fainelli@broadcom.com> napisał(a):
> On 2/26/26 12:24, Rafał Miłecki wrote:
> > W dniu 8.05.2025 o 11:14, Jonas Gorski pisze:
> >> When bridged ports and standalone ports share a VLAN, e.g. via VLAN
> >> uppers, or untagged traffic with a vlan unaware bridge, the ASIC will
> >> still try to forward traffic to known FDB entries on standalone ports.
> >> But since the port VLAN masks prevent forwarding to bridged ports, this
> >> traffic will be dropped.
> >>
> >> This e.g. can be observed in the bridge_vlan_unaware ping tests, where
> >> this breaks pinging with learning on.
> >>
> >> Work around this by enabling the simplified EAP mode on switches
> >> supporting it for standalone ports, which causes the ASIC to redirect
> >> traffic of unknown source MAC addresses to the CPU port.
> >>
> >> Since standalone ports do not learn, there are no known source MAC
> >> addresses, so effectively this redirects all incoming traffic to the CPU
> >> port.
> >
> > This change broke standalone ports (those not being part of bridge) on
> > Northstar (BCM5301X) devices. We got reports from users about WAN ports
> > on routers not working anymore:
> > https://github.com/openwrt/openwrt/issues/21187
> > https://github.com/openwrt/openwrt/issues/21349
> >
> > Can you take another look at those changes, see if something me be
> > wrong / missing?
>
> Checking the BCM5301X datasheet, the register definitions look correct.
>
> The first report appears to be for a Netgear R7000 which AFAICT would
> utilize port 8 of the switch as the CPU port, which would be correct.
>
> The other report is for a Netgear R6250 where port 5 would be acting as
> the CPU port, that could be an issue unless the switch is also
> programmed to use port 5 as the IMP port.
>
> Could you try to program GMNGCFG (page 0x02, offset 0) bits 7:6 to have
> the value 0b11 which is dual IMP port (IMPO is port 8 and IMP1 is port
> 5) and see if that helps?

On my Luxul XWR-3150 (it uses gmac0 and port 5) the value of
B53_MGMT_PAGE (0x02) / B53_GLOBAL_CONFIG (0x00) is 0xc2. Nothing to
set there I suppose?

-- 
Rafał

Re: [PATCH net] net: dsa: b53: prevent standalone from trying to forward to other ports

[Rafał Miłecki]

pt., 27 lut 2026 o 13:23 Rafał Miłecki <zajec5@gmail.com> napisał(a):
> czw., 26 lut 2026 o 22:26 Florian Fainelli
> <florian.fainelli@broadcom.com> napisał(a):
> > Could you try to program GMNGCFG (page 0x02, offset 0) bits 7:6 to have
> > the value 0b11 which is dual IMP port (IMPO is port 8 and IMP1 is port
> > 5) and see if that helps?
>
> On my Luxul XWR-3150 (it uses gmac0 and port 5) the value of
> B53_MGMT_PAGE (0x02) / B53_GLOBAL_CONFIG (0x00) is 0xc2. Nothing to
> set there I suppose?

I tried other possible values:
0x02
0x82
0x42

None affected my switch / wan / lan ports.

-- 
Rafał

Re: [PATCH net] net: dsa: b53: prevent standalone from trying to forward to other ports

[Florian Fainelli]

On 2/27/26 05:32, Rafał Miłecki wrote:
> pt., 27 lut 2026 o 13:23 Rafał Miłecki <zajec5@gmail.com> napisał(a):
>> czw., 26 lut 2026 o 22:26 Florian Fainelli
>> <florian.fainelli@broadcom.com> napisał(a):
>>> Could you try to program GMNGCFG (page 0x02, offset 0) bits 7:6 to have
>>> the value 0b11 which is dual IMP port (IMPO is port 8 and IMP1 is port
>>> 5) and see if that helps?
>>
>> On my Luxul XWR-3150 (it uses gmac0 and port 5) the value of
>> B53_MGMT_PAGE (0x02) / B53_GLOBAL_CONFIG (0x00) is 0xc2. Nothing to
>> set there I suppose?
> 
> I tried other possible values:
> 0x02
> 0x82
> 0x42
> 
> None affected my switch / wan / lan ports.
> 

Thanks for giving it a try, how about the BRCM_HDR_CTRL register 
programming (page 0x02, offset 0x03)? Can you see whether it is set to 
0x3 or 0x1?
-- 
Florian

Re: [PATCH net] net: dsa: b53: prevent standalone from trying to forward to other ports

[Rafał Miłecki]

pt., 27 lut 2026 o 18:58 Florian Fainelli <f.fainelli@gmail.com> napisał(a):
> On 2/27/26 05:32, Rafał Miłecki wrote:
> > pt., 27 lut 2026 o 13:23 Rafał Miłecki <zajec5@gmail.com> napisał(a):
> >> czw., 26 lut 2026 o 22:26 Florian Fainelli
> >> <florian.fainelli@broadcom.com> napisał(a):
> >>> Could you try to program GMNGCFG (page 0x02, offset 0) bits 7:6 to have
> >>> the value 0b11 which is dual IMP port (IMPO is port 8 and IMP1 is port
> >>> 5) and see if that helps?
> >>
> >> On my Luxul XWR-3150 (it uses gmac0 and port 5) the value of
> >> B53_MGMT_PAGE (0x02) / B53_GLOBAL_CONFIG (0x00) is 0xc2. Nothing to
> >> set there I suppose?
> >
> > I tried other possible values:
> > 0x02
> > 0x82
> > 0x42
> >
> > None affected my switch / wan / lan ports.
> >
>
> Thanks for giving it a try, how about the BRCM_HDR_CTRL register
> programming (page 0x02, offset 0x03)? Can you see whether it is set to
> 0x3 or 0x1?

It's 0x02 which seems to match BRCM_HDR_P5_EN. It makes sense as
XWR-3150 uses gmac0 + port 5.

-- 
Rafał

Re: [PATCH net] net: dsa: b53: prevent standalone from trying to forward to other ports

[Jonas Gorski]

Hi,

sorry for being late to the party; currently on vacation (be back
early next week).

On Fri, Feb 27, 2026 at 7:13 PM Rafał Miłecki <zajec5@gmail.com> wrote:
>
> pt., 27 lut 2026 o 18:58 Florian Fainelli <f.fainelli@gmail.com> napisał(a):
> > On 2/27/26 05:32, Rafał Miłecki wrote:
> > > pt., 27 lut 2026 o 13:23 Rafał Miłecki <zajec5@gmail.com> napisał(a):
> > >> czw., 26 lut 2026 o 22:26 Florian Fainelli
> > >> <florian.fainelli@broadcom.com> napisał(a):
> > >>> Could you try to program GMNGCFG (page 0x02, offset 0) bits 7:6 to have
> > >>> the value 0b11 which is dual IMP port (IMPO is port 8 and IMP1 is port
> > >>> 5) and see if that helps?
> > >>
> > >> On my Luxul XWR-3150 (it uses gmac0 and port 5) the value of
> > >> B53_MGMT_PAGE (0x02) / B53_GLOBAL_CONFIG (0x00) is 0xc2. Nothing to
> > >> set there I suppose?
> > >
> > > I tried other possible values:
> > > 0x02
> > > 0x82
> > > 0x42
> > >
> > > None affected my switch / wan / lan ports.
> > >
> >
> > Thanks for giving it a try, how about the BRCM_HDR_CTRL register
> > programming (page 0x02, offset 0x03)? Can you see whether it is set to
> > 0x3 or 0x1?
>
> It's 0x02 which seems to match BRCM_HDR_P5_EN. It makes sense as
> XWR-3150 uses gmac0 + port 5.

Does this affect all ports, or only the wan port?

Have you tried to check the hardware counters if they get sent out a
different (cpu) port? AFAIK they should be available as ethtool
counters via gmacs, though I'm not 100% sure.

Can you check the contents of the EAP Port config registers? Maybe
some extra mode is set there by default.

I don't have NS(+) device, so I can only guess. Though I always
wondered if the setup which CPU port (5/8/both) to use is just a
software thing or actually a hardware thing ... .

Best regards,
Jonas

Re: [PATCH net] net: dsa: b53: prevent standalone from trying to forward to other ports

[Rafał Miłecki]

sob., 28 lut 2026 o 14:30 Jonas Gorski <jonas.gorski@gmail.com> napisał(a):
> Does this affect all ports, or only the wan port?

Problem affects every standalone port (one that isn't part of a bridge).


> Have you tried to check the hardware counters if they get sent out a
> different (cpu) port? AFAIK they should be available as ethtool
> counters via gmacs, though I'm not 100% sure.

With your patch:
1. "rx_packets" stays 0
2. "rx_bytes" stays 0
3. "RxDiscarded" goes higher than 0


> Can you check the contents of the EAP Port config registers? Maybe
> some extra mode is set there by default.

All my B53_EAP_PAGE / B53_PORT_EAP_CONF(port) registers are set to 0 by default.


> I don't have NS(+) device, so I can only guess. Though I always
> wondered if the setup which CPU port (5/8/both) to use is just a
> software thing or actually a hardware thing ... .

Without your patch, 10 s after connecting cable to "wan" port:
NIC statistics:
     tx_packets: 13
     tx_bytes: 1998
     rx_packets: 2
     rx_bytes: 1180
     TxOctets: 2052
     TxDropPkts: 0
     TxBroadcastPkts: 2
     TxMulticastPkts: 11
     TxUnicastPkts: 0
     TxCollisions: 0
     TxSingleCollision: 0
     TxMultipleCollision: 0
     TxDeferredTransmit: 0
     TxLateCollision: 0
     TxExcessiveCollision: 0
     TxPausePkts: 0
     RxOctets: 1188
     RxUndersizePkts: 0
     RxPausePkts: 0
     Pkts64Octets: 0
     Pkts65to127Octets: 0
     Pkts128to255Octets: 0
     Pkts256to511Octets: 0
     Pkts512to1023Octets: 2
     Pkts1024to1522Octets: 0
     RxOversizePkts: 0
     RxJabbers: 0
     RxAlignmentErrors: 0
     RxFCSErrors: 0
     RxGoodOctets: 1188
     RxDropPkts: 0
     RxUnicastPkts: 2
     RxMulticastPkts: 0
     RxBroadcastPkts: 0
     RxSAChanges: 1
     RxFragments: 0
     RxJumboPkts: 0
     RxSymbolErrors: 0
     RxDiscarded: 0

With your patch, 10 s after connecting cable to "wan" port:
NIC statistics:
     tx_packets: 14
     tx_bytes: 2334
     rx_packets: 0
     rx_bytes: 0
     TxOctets: 2392
     TxDropPkts: 0
     TxBroadcastPkts: 3
     TxMulticastPkts: 11
     TxUnicastPkts: 0
     TxCollisions: 0
     TxSingleCollision: 0
     TxMultipleCollision: 0
     TxDeferredTransmit: 0
     TxLateCollision: 0
     TxExcessiveCollision: 0
     TxPausePkts: 0
     RxOctets: 1782
     RxUndersizePkts: 0
     RxPausePkts: 0
     Pkts64Octets: 0
     Pkts65to127Octets: 0
     Pkts128to255Octets: 0
     Pkts256to511Octets: 0
     Pkts512to1023Octets: 3
     Pkts1024to1522Octets: 0
     RxOversizePkts: 0
     RxJabbers: 0
     RxAlignmentErrors: 0
     RxFCSErrors: 0
     RxGoodOctets: 1782
     RxDropPkts: 0
     RxUnicastPkts: 3
     RxMulticastPkts: 0
     RxBroadcastPkts: 0
     RxSAChanges: 1
     RxFragments: 0
     RxJumboPkts: 0
     RxSymbolErrors: 0
     RxDiscarded: 3

-- 
Rafał