LinuxPPC-Dev Archive on lore.kernel.org
 help / color / mirror / Atom feed
* Re: [PATCH net v3] net: dpaa: fix mode setting
       [not found] <20260717132401.2653252-1-mwalle@kernel.org>
@ 2026-07-17 21:10 ` Sean Anderson
  2026-07-18  7:35   ` Christian Zigotzky
  2026-07-18 16:31   ` Christian Zigotzky
  0 siblings, 2 replies; 4+ messages in thread
From: Sean Anderson @ 2026-07-17 21:10 UTC (permalink / raw)
  To: Michael Walle, Madalin Bucur, Andrew Lunn, David S . Miller,
	Eric Dumazet, Jakub Kicinski, Paolo Abeni, Christian Zigotzky
  Cc: netdev, linux-kernel, linuxppc-dev

On 7/17/26 09:20, Michael Walle wrote:
> Before converting to the phylink interface, the init function would have
> set a non-reserved I/F mode in the maccfg2 register. After converting to
> phylink, 0 is written as mode, which is a reserved value (although it's
> the hardware default). Without a valid mode, a SGMII link is never
> established between the MAC and the PHY and thus .link_up() is never
> called which could set the correct mode according to the actual speed.
> 
> Fix it by setting the maximum speed of the phy_interface_t in use in
> .mac_config() - just like the driver did before the phylink conversion.
> 
> Fixes: 5d93cfcf7360 ("net: dpaa: Convert to phylink")
> Suggested-by: Sean Anderson <sean.anderson@linux.dev>
> Signed-off-by: Michael Walle <mwalle@kernel.org>
> ---
> I didn't grab Sean's Rb tag as this is somewhat different.
> 
> Changes in v3:
>   - keep the mode setting also in .adjust_link().
>   - reword the commit message, to be (hopefully) more precise
>   - Link to v2: https://lore.kernel.org/r/20260710143430.2276141-1-mwalle@kernel.org/
> 
> Changes in v2:
>   - the setting is/was based on the maximum speed, not the current
>     speed. thus, move the setting into mac_config().
>   - Link to v1: https://lore.kernel.org/r/20260706121011.1948906-1-mwalle@kernel.org/
> 
>   .../net/ethernet/freescale/fman/fman_dtsec.c    | 17 ++++++++++++-----
>   1 file changed, 12 insertions(+), 5 deletions(-)
> 
> diff --git a/drivers/net/ethernet/freescale/fman/fman_dtsec.c b/drivers/net/ethernet/freescale/fman/fman_dtsec.c
> index fe35703c509e..b8d70c0ecb6c 100644
> --- a/drivers/net/ethernet/freescale/fman/fman_dtsec.c
> +++ b/drivers/net/ethernet/freescale/fman/fman_dtsec.c
> @@ -900,22 +900,28 @@ static void dtsec_mac_config(struct phylink_config *config, unsigned int mode,
>   {
>   	struct mac_device *mac_dev = fman_config_to_mac(config);
>   	struct dtsec_regs __iomem *regs = mac_dev->fman_mac->regs;
> -	u32 tmp;
> +	u32 ecntrl, maccfg2;
> +
> +	maccfg2 = ioread32be(&regs->maccfg2);
> +	maccfg2 &= ~(MACCFG2_NIBBLE_MODE | MACCFG2_BYTE_MODE);
>   
>   	switch (state->interface) {
>   	case PHY_INTERFACE_MODE_RMII:
> -		tmp = DTSEC_ECNTRL_RMM;
> +		ecntrl = DTSEC_ECNTRL_RMM;
> +		maccfg2 |= MACCFG2_NIBBLE_MODE;
>   		break;
>   	case PHY_INTERFACE_MODE_RGMII:
>   	case PHY_INTERFACE_MODE_RGMII_ID:
>   	case PHY_INTERFACE_MODE_RGMII_RXID:
>   	case PHY_INTERFACE_MODE_RGMII_TXID:
> -		tmp = DTSEC_ECNTRL_GMIIM | DTSEC_ECNTRL_RPM;
> +		ecntrl = DTSEC_ECNTRL_GMIIM | DTSEC_ECNTRL_RPM;
> +		maccfg2 |= MACCFG2_BYTE_MODE;
>   		break;
>   	case PHY_INTERFACE_MODE_SGMII:
>   	case PHY_INTERFACE_MODE_1000BASEX:
>   	case PHY_INTERFACE_MODE_2500BASEX:
> -		tmp = DTSEC_ECNTRL_TBIM | DTSEC_ECNTRL_SGMIIM;
> +		ecntrl = DTSEC_ECNTRL_TBIM | DTSEC_ECNTRL_SGMIIM;
> +		maccfg2 |= MACCFG2_BYTE_MODE;
>   		break;
>   	default:
>   		dev_warn(mac_dev->dev, "cannot configure dTSEC for %s\n",
> @@ -923,7 +929,8 @@ static void dtsec_mac_config(struct phylink_config *config, unsigned int mode,
>   		return;
>   	}
>   
> -	iowrite32be(tmp, &regs->ecntrl);
> +	iowrite32be(ecntrl, &regs->ecntrl);
> +	iowrite32be(maccfg2, &regs->maccfg2);
>   }
>   
>   static void dtsec_link_up(struct phylink_config *config, struct phy_device *phy,

Reviewed-by: Sean Anderson <sean.anderson@linux.dev>

Christian, can you test this patch with ethernet at 100/1G speed if you still have
access to those P5020/P5040 boards?

https://lore.kernel.org/all/0bfc8f3d-cb62-25f4-2590-ff424adbe48a@xenosoft.de/


^ permalink raw reply	[flat|nested] 4+ messages in thread
* Re: [PATCH net v3] net: dpaa: fix mode setting
  2026-07-17 21:10 ` [PATCH net v3] net: dpaa: fix mode setting Sean Anderson
@ 2026-07-18  7:35   ` Christian Zigotzky
  2026-07-18 16:31   ` Christian Zigotzky
  1 sibling, 0 replies; 4+ messages in thread
From: Christian Zigotzky @ 2026-07-18  7:35 UTC (permalink / raw)
  To: Sean Anderson
  Cc: Michael Walle, Madalin Bucur, Andrew Lunn, David S Miller,
	Eric Dumazet, Jakub Kicinski, Abeni Paolo, Christian Zigotzky,
	netdev, linux-kernel, linuxppc-dev, Trevor Dickinson,
	mad skateman, hypexed

On  17 July 2026 at 11:11 am, Sean Anderson <sean.anderson@linux.dev> wrote:

On 7/17/26 09:20, Michael Walle wrote:

---
I didn't grab Sean's Rb tag as this is somewhat different.
Changes in v3:
 - keep the mode setting also in .adjust_link().
 - reword the commit message, to be (hopefully) more precise
 - Link to v2: https://lore.kernel.org/r/20260710143430.2276141-1-mwalle@kernel.org/
Changes in v2:
 - the setting is/was based on the maximum speed, not the current
   speed. thus, move the setting into mac_config().
 - Link to v1: https://lore.kernel.org/r/20260706121011.1948906-1-mwalle@kernel.org/
 .../net/ethernet/freescale/fman/fman_dtsec.c    | 17 ++++++++++++-----
 1 file changed, 12 insertions(+), 5 deletions(-)
diff --git a/drivers/net/ethernet/freescale/fman/fman_dtsec.c b/drivers/net/ethernet/freescale/fman/fman_dtsec.c
index fe35703c509e..b8d70c0ecb6c 100644
--- a/drivers/net/ethernet/freescale/fman/fman_dtsec.c
+++ b/drivers/net/ethernet/freescale/fman/fman_dtsec.c
@@ -900,22 +900,28 @@ static void dtsec_mac_config(struct phylink_config *config, unsigned int mode,
 {
     struct mac_device *mac_dev = fman_config_to_mac(config);
     struct dtsec_regs __iomem *regs = mac_dev->fman_mac->regs;
-    u32 tmp;
+    u32 ecntrl, maccfg2;
+
+    maccfg2 = ioread32be(&regs->maccfg2);
+    maccfg2 &= ~(MACCFG2_NIBBLE_MODE | MACCFG2_BYTE_MODE);
       switch (state->interface) {
     case PHY_INTERFACE_MODE_RMII:
-        tmp = DTSEC_ECNTRL_RMM;
+        ecntrl = DTSEC_ECNTRL_RMM;
+        maccfg2 |= MACCFG2_NIBBLE_MODE;
         break;
     case PHY_INTERFACE_MODE_RGMII:
     case PHY_INTERFACE_MODE_RGMII_ID:
     case PHY_INTERFACE_MODE_RGMII_RXID:
     case PHY_INTERFACE_MODE_RGMII_TXID:
-        tmp = DTSEC_ECNTRL_GMIIM | DTSEC_ECNTRL_RPM;
+        ecntrl = DTSEC_ECNTRL_GMIIM | DTSEC_ECNTRL_RPM;
+        maccfg2 |= MACCFG2_BYTE_MODE;
         break;
     case PHY_INTERFACE_MODE_SGMII:
     case PHY_INTERFACE_MODE_1000BASEX:
     case PHY_INTERFACE_MODE_2500BASEX:
-        tmp = DTSEC_ECNTRL_TBIM | DTSEC_ECNTRL_SGMIIM;
+        ecntrl = DTSEC_ECNTRL_TBIM | DTSEC_ECNTRL_SGMIIM;
+        maccfg2 |= MACCFG2_BYTE_MODE;
         break;
     default:
         dev_warn(mac_dev->dev, "cannot configure dTSEC for %s\n",
@@ -923,7 +929,8 @@ static void dtsec_mac_config(struct phylink_config *config, unsigned int mode,
         return;
     }
 -    iowrite32be(tmp, &regs->ecntrl);
+    iowrite32be(ecntrl, &regs->ecntrl);
+    iowrite32be(maccfg2, &regs->maccfg2);
 }
   static void dtsec_link_up(struct phylink_config *config, struct phy_device *phy,

Reviewed-by: Sean Anderson <sean.anderson@linux.dev>

Christian, can you test this patch with ethernet at 100/1G speed if you still have
access to those P5020/P5040 boards?

- - -

Yes, I will test it as soon as possible.

- Christian


^ permalink raw reply related	[flat|nested] 4+ messages in thread
* Re: [PATCH net v3] net: dpaa: fix mode setting
  2026-07-17 21:10 ` [PATCH net v3] net: dpaa: fix mode setting Sean Anderson
  2026-07-18  7:35   ` Christian Zigotzky
@ 2026-07-18 16:31   ` Christian Zigotzky
  2026-07-19 11:58     ` Christian Zigotzky
  1 sibling, 1 reply; 4+ messages in thread
From: Christian Zigotzky @ 2026-07-18 16:31 UTC (permalink / raw)
  To: Sean Anderson, Michael Walle, Madalin Bucur, Andrew Lunn,
	David S . Miller, Eric Dumazet, Jakub Kicinski, Paolo Abeni,
	Christian Zigotzky
  Cc: netdev, linux-kernel, linuxppc-dev, R.T.Dickinson, mad skateman,
	Damien Stewart

On 17/07/26 23:10, Sean Anderson wrote:
> On 7/17/26 09:20, Michael Walle wrote:
>> Before converting to the phylink interface, the init function would have
>> set a non-reserved I/F mode in the maccfg2 register. After converting to
>> phylink, 0 is written as mode, which is a reserved value (although it's
>> the hardware default). Without a valid mode, a SGMII link is never
>> established between the MAC and the PHY and thus .link_up() is never
>> called which could set the correct mode according to the actual speed.
>>
>> Fix it by setting the maximum speed of the phy_interface_t in use in
>> .mac_config() - just like the driver did before the phylink conversion.
>>
>> Fixes: 5d93cfcf7360 ("net: dpaa: Convert to phylink")
>> Suggested-by: Sean Anderson <sean.anderson@linux.dev>
>> Signed-off-by: Michael Walle <mwalle@kernel.org>
>> ---
>> I didn't grab Sean's Rb tag as this is somewhat different.
>>
>> Changes in v3:
>>   - keep the mode setting also in .adjust_link().
>>   - reword the commit message, to be (hopefully) more precise
>>   - Link to v2: 
>> https://lore.kernel.org/r/20260710143430.2276141-1-mwalle@kernel.org/
>>
>> Changes in v2:
>>   - the setting is/was based on the maximum speed, not the current
>>     speed. thus, move the setting into mac_config().
>>   - Link to v1: 
>> https://lore.kernel.org/r/20260706121011.1948906-1-mwalle@kernel.org/
>>
>>   .../net/ethernet/freescale/fman/fman_dtsec.c    | 17 ++++++++++++-----
>>   1 file changed, 12 insertions(+), 5 deletions(-)
>>
>> diff --git a/drivers/net/ethernet/freescale/fman/fman_dtsec.c 
>> b/drivers/net/ethernet/freescale/fman/fman_dtsec.c
>> index fe35703c509e..b8d70c0ecb6c 100644
>> --- a/drivers/net/ethernet/freescale/fman/fman_dtsec.c
>> +++ b/drivers/net/ethernet/freescale/fman/fman_dtsec.c
>> @@ -900,22 +900,28 @@ static void dtsec_mac_config(struct 
>> phylink_config *config, unsigned int mode,
>>   {
>>       struct mac_device *mac_dev = fman_config_to_mac(config);
>>       struct dtsec_regs __iomem *regs = mac_dev->fman_mac->regs;
>> -    u32 tmp;
>> +    u32 ecntrl, maccfg2;
>> +
>> +    maccfg2 = ioread32be(&regs->maccfg2);
>> +    maccfg2 &= ~(MACCFG2_NIBBLE_MODE | MACCFG2_BYTE_MODE);
>>         switch (state->interface) {
>>       case PHY_INTERFACE_MODE_RMII:
>> -        tmp = DTSEC_ECNTRL_RMM;
>> +        ecntrl = DTSEC_ECNTRL_RMM;
>> +        maccfg2 |= MACCFG2_NIBBLE_MODE;
>>           break;
>>       case PHY_INTERFACE_MODE_RGMII:
>>       case PHY_INTERFACE_MODE_RGMII_ID:
>>       case PHY_INTERFACE_MODE_RGMII_RXID:
>>       case PHY_INTERFACE_MODE_RGMII_TXID:
>> -        tmp = DTSEC_ECNTRL_GMIIM | DTSEC_ECNTRL_RPM;
>> +        ecntrl = DTSEC_ECNTRL_GMIIM | DTSEC_ECNTRL_RPM;
>> +        maccfg2 |= MACCFG2_BYTE_MODE;
>>           break;
>>       case PHY_INTERFACE_MODE_SGMII:
>>       case PHY_INTERFACE_MODE_1000BASEX:
>>       case PHY_INTERFACE_MODE_2500BASEX:
>> -        tmp = DTSEC_ECNTRL_TBIM | DTSEC_ECNTRL_SGMIIM;
>> +        ecntrl = DTSEC_ECNTRL_TBIM | DTSEC_ECNTRL_SGMIIM;
>> +        maccfg2 |= MACCFG2_BYTE_MODE;
>>           break;
>>       default:
>>           dev_warn(mac_dev->dev, "cannot configure dTSEC for %s\n",
>> @@ -923,7 +929,8 @@ static void dtsec_mac_config(struct 
>> phylink_config *config, unsigned int mode,
>>           return;
>>       }
>>   -    iowrite32be(tmp, &regs->ecntrl);
>> +    iowrite32be(ecntrl, &regs->ecntrl);
>> +    iowrite32be(maccfg2, &regs->maccfg2);
>>   }
>>     static void dtsec_link_up(struct phylink_config *config, struct 
>> phy_device *phy,
>
> Reviewed-by: Sean Anderson <sean.anderson@linux.dev>
>
> Christian, can you test this patch with ethernet at 100/1G speed if 
> you still have
> access to those P5020/P5040 boards?
>
> https://lore.kernel.org/all/0bfc8f3d-cb62-25f4-2590-ff424adbe48a@xenosoft.de/ 
>
I tested the patch today. I don't see any differences.

Further information: 
https://github.com/chzigotzky/kernels/releases/tag/v7.2.0-rc3-fman-dtsec-patch

Christian

-- 
Sent with BrassMonkey 34.2.2 (https://github.com/chzigotzky/Web-Browsers-and-Suites-for-Linux-PPC/releases/tag/BrassMonkey_34.2.2)



^ permalink raw reply	[flat|nested] 4+ messages in thread
* [PATCH net v3] net: dpaa: fix mode setting
  2026-07-18 16:31   ` Christian Zigotzky
@ 2026-07-19 11:58     ` Christian Zigotzky
  0 siblings, 0 replies; 4+ messages in thread
From: Christian Zigotzky @ 2026-07-19 11:58 UTC (permalink / raw)
  To: Sean Anderson, Michael Walle, Madalin Bucur, Andrew Lunn,
	David S . Miller, Eric Dumazet, Jakub Kicinski, Paolo Abeni,
	Christian Zigotzky
  Cc: netdev, linux-kernel, linuxppc-dev, R.T.Dickinson, mad skateman,
	Damien Stewart

On 18/07/26 18:31, Christian Zigotzky wrote:
> On 17/07/26 23:10, Sean Anderson wrote:
>> On 7/17/26 09:20, Michael Walle wrote:
>>> Before converting to the phylink interface, the init function would 
>>> have
>>> set a non-reserved I/F mode in the maccfg2 register. After 
>>> converting to
>>> phylink, 0 is written as mode, which is a reserved value (although it's
>>> the hardware default). Without a valid mode, a SGMII link is never
>>> established between the MAC and the PHY and thus .link_up() is never
>>> called which could set the correct mode according to the actual speed.
>>>
>>> Fix it by setting the maximum speed of the phy_interface_t in use in
>>> .mac_config() - just like the driver did before the phylink conversion.
>>>
>>> Fixes: 5d93cfcf7360 ("net: dpaa: Convert to phylink")
>>> Suggested-by: Sean Anderson <sean.anderson@linux.dev>
>>> Signed-off-by: Michael Walle <mwalle@kernel.org>
>>> ---
>>> I didn't grab Sean's Rb tag as this is somewhat different.
>>>
>>> Changes in v3:
>>>   - keep the mode setting also in .adjust_link().
>>>   - reword the commit message, to be (hopefully) more precise
>>>   - Link to v2: 
>>> https://lore.kernel.org/r/20260710143430.2276141-1-mwalle@kernel.org/
>>>
>>> Changes in v2:
>>>   - the setting is/was based on the maximum speed, not the current
>>>     speed. thus, move the setting into mac_config().
>>>   - Link to v1: 
>>> https://lore.kernel.org/r/20260706121011.1948906-1-mwalle@kernel.org/
>>>
>>>   .../net/ethernet/freescale/fman/fman_dtsec.c    | 17 
>>> ++++++++++++-----
>>>   1 file changed, 12 insertions(+), 5 deletions(-)
>>>
>>> diff --git a/drivers/net/ethernet/freescale/fman/fman_dtsec.c 
>>> b/drivers/net/ethernet/freescale/fman/fman_dtsec.c
>>> index fe35703c509e..b8d70c0ecb6c 100644
>>> --- a/drivers/net/ethernet/freescale/fman/fman_dtsec.c
>>> +++ b/drivers/net/ethernet/freescale/fman/fman_dtsec.c
>>> @@ -900,22 +900,28 @@ static void dtsec_mac_config(struct 
>>> phylink_config *config, unsigned int mode,
>>>   {
>>>       struct mac_device *mac_dev = fman_config_to_mac(config);
>>>       struct dtsec_regs __iomem *regs = mac_dev->fman_mac->regs;
>>> -    u32 tmp;
>>> +    u32 ecntrl, maccfg2;
>>> +
>>> +    maccfg2 = ioread32be(&regs->maccfg2);
>>> +    maccfg2 &= ~(MACCFG2_NIBBLE_MODE | MACCFG2_BYTE_MODE);
>>>         switch (state->interface) {
>>>       case PHY_INTERFACE_MODE_RMII:
>>> -        tmp = DTSEC_ECNTRL_RMM;
>>> +        ecntrl = DTSEC_ECNTRL_RMM;
>>> +        maccfg2 |= MACCFG2_NIBBLE_MODE;
>>>           break;
>>>       case PHY_INTERFACE_MODE_RGMII:
>>>       case PHY_INTERFACE_MODE_RGMII_ID:
>>>       case PHY_INTERFACE_MODE_RGMII_RXID:
>>>       case PHY_INTERFACE_MODE_RGMII_TXID:
>>> -        tmp = DTSEC_ECNTRL_GMIIM | DTSEC_ECNTRL_RPM;
>>> +        ecntrl = DTSEC_ECNTRL_GMIIM | DTSEC_ECNTRL_RPM;
>>> +        maccfg2 |= MACCFG2_BYTE_MODE;
>>>           break;
>>>       case PHY_INTERFACE_MODE_SGMII:
>>>       case PHY_INTERFACE_MODE_1000BASEX:
>>>       case PHY_INTERFACE_MODE_2500BASEX:
>>> -        tmp = DTSEC_ECNTRL_TBIM | DTSEC_ECNTRL_SGMIIM;
>>> +        ecntrl = DTSEC_ECNTRL_TBIM | DTSEC_ECNTRL_SGMIIM;
>>> +        maccfg2 |= MACCFG2_BYTE_MODE;
>>>           break;
>>>       default:
>>>           dev_warn(mac_dev->dev, "cannot configure dTSEC for %s\n",
>>> @@ -923,7 +929,8 @@ static void dtsec_mac_config(struct 
>>> phylink_config *config, unsigned int mode,
>>>           return;
>>>       }
>>>   -    iowrite32be(tmp, &regs->ecntrl);
>>> +    iowrite32be(ecntrl, &regs->ecntrl);
>>> +    iowrite32be(maccfg2, &regs->maccfg2);
>>>   }
>>>     static void dtsec_link_up(struct phylink_config *config, struct 
>>> phy_device *phy,
>>
>> Reviewed-by: Sean Anderson <sean.anderson@linux.dev>
>>
>> Christian, can you test this patch with ethernet at 100/1G speed if 
>> you still have
>> access to those P5020/P5040 boards?
>>
>> https://lore.kernel.org/all/0bfc8f3d-cb62-25f4-2590-ff424adbe48a@xenosoft.de/ 
>>
> I tested the patch today. I don't see any differences.
>
> Further information: 
> https://github.com/chzigotzky/kernels/releases/tag/v7.2.0-rc3-fman-dtsec-patch
>
> Christian
>
I tested further the new patch today and switching between 100Mbit/s and 
1Gbit/s works without any problems.

[ 1692.006428] fsl_dpaa_mac ffe4e8000.ethernet eth0: PHY 
[mdio@ffe4e1120:03] driver [Micrel KSZ9021 Gigabit PHY] (irq=POLL)
[ 1692.006448] fsl_dpaa_mac ffe4e8000.ethernet eth0: configuring for 
phy/rgmii link mode
[ 1692.021436] fsl_dpaa_mac ffe5e8000.ethernet eth2: PHY 
[mdio@ffe4e1120:07] driver [Micrel KSZ9021 Gigabit PHY] (irq=POLL)
[ 1692.021456] fsl_dpaa_mac ffe5e8000.ethernet eth2: configuring for 
phy/rgmii link mode
[ 1695.057699] fsl_dpaa_mac ffe4e8000.ethernet eth0: Link is Up - 
1Gbps/Full - flow control rx/tx
[ 2148.681140] fsl_dpaa_mac ffe4e8000.ethernet eth0: Link is Down
[ 2149.704534] fsl_dpaa_mac ffe4e8000.ethernet eth0: Link is Up - 
100Mbps/Full - flow control rx/tx
[ 2199.875179] fsl_dpaa_mac ffe4e8000.ethernet eth0: Link is Down
[ 2201.923103] fsl_dpaa_mac ffe4e8000.ethernet eth0: Link is Up - 
1Gbps/Full - flow control rx/tx

- Christian

-- 
Sent with BrassMonkey 34.2.2 (https://github.com/chzigotzky/Web-Browsers-and-Suites-for-Linux-PPC/releases/tag/BrassMonkey_34.2.2)



^ permalink raw reply	[flat|nested] 4+ messages in thread
end of thread, other threads:[~2026-07-19 11:59 UTC | newest]

Thread overview: 4+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
     [not found] <20260717132401.2653252-1-mwalle@kernel.org>
2026-07-17 21:10 ` [PATCH net v3] net: dpaa: fix mode setting Sean Anderson
2026-07-18  7:35   ` Christian Zigotzky
2026-07-18 16:31   ` Christian Zigotzky
2026-07-19 11:58     ` Christian Zigotzky

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox