There are two variants of the MT7531 switch IC which got different
features (and pins) regarding port 5:
* MT7531AE: SGMII/1000Base-X/2500Base-X SerDes
* MT7531BE: RGMII
Moving the creation of the SerDes PCS from mt753x_setup to mt7530_probe
with commit 6de285229773 ("net: dsa: mt7530: move SGMII PCS creation to
mt7530_probe function") works fine for MT7531AE which got two instances
of mtk-pcs-lynxi, however, MT7531BE requires mt7531_pll_setup to setup
clocks before the single PCS on port 6 (usually used as CPU port)
starts to work and hence the PCS creation failed on MT7531BE.
Fix this by introducing a pointer to mt7531_create_sgmii function in
struct mt7530_priv and call it again at the end of mt753x_setup like it
was before commit 6de285229773 ("net: dsa: mt7530: move SGMII PCS
creation to mt7530_probe function").
Fixes: 6de285229773 ("net: dsa: mt7530: move SGMII PCS creation to mt7530_probe function")
Signed-off-by: Daniel Golle <[email protected]>
---
drivers/net/dsa/mt7530-mdio.c | 14 +++++++-------
drivers/net/dsa/mt7530.c | 6 ++++++
drivers/net/dsa/mt7530.h | 4 ++++
3 files changed, 17 insertions(+), 7 deletions(-)
diff --git a/drivers/net/dsa/mt7530-mdio.c b/drivers/net/dsa/mt7530-mdio.c
index 34a547b88e497..f17eab67d15fa 100644
--- a/drivers/net/dsa/mt7530-mdio.c
+++ b/drivers/net/dsa/mt7530-mdio.c
@@ -81,14 +81,17 @@ static const struct regmap_bus mt7530_regmap_bus = {
};
static int
-mt7531_create_sgmii(struct mt7530_priv *priv)
+mt7531_create_sgmii(struct mt7530_priv *priv, bool dual_sgmii)
{
struct regmap_config *mt7531_pcs_config[2];
struct phylink_pcs *pcs;
struct regmap *regmap;
int i, ret = 0;
- for (i = 0; i < 2; i++) {
+ /* MT7531AE has two SGMII units for port 5 and port 6
+ * MT7531BE has only one SGMII unit for port 6
+ */
+ for (i = dual_sgmii ? 0 : 1; i < 2; i++) {
mt7531_pcs_config[i] = devm_kzalloc(priv->dev,
sizeof(struct regmap_config),
GFP_KERNEL);
@@ -208,11 +211,8 @@ mt7530_probe(struct mdio_device *mdiodev)
if (IS_ERR(priv->regmap))
return PTR_ERR(priv->regmap);
- if (priv->id == ID_MT7531) {
- ret = mt7531_create_sgmii(priv);
- if (ret)
- return ret;
- }
+ if (priv->id == ID_MT7531)
+ priv->create_sgmii = mt7531_create_sgmii;
return dsa_register_switch(priv->ds);
}
diff --git a/drivers/net/dsa/mt7530.c b/drivers/net/dsa/mt7530.c
index e4bb5037d3525..c680873819b01 100644
--- a/drivers/net/dsa/mt7530.c
+++ b/drivers/net/dsa/mt7530.c
@@ -3018,6 +3018,12 @@ mt753x_setup(struct dsa_switch *ds)
if (ret && priv->irq)
mt7530_free_irq_common(priv);
+ if (priv->create_sgmii) {
+ ret = priv->create_sgmii(priv, mt7531_dual_sgmii_supported(priv));
+ if (ret && priv->irq)
+ mt7530_free_irq(priv);
+ }
+
return ret;
}
diff --git a/drivers/net/dsa/mt7530.h b/drivers/net/dsa/mt7530.h
index 01db5c9724fa8..6e89578b4de02 100644
--- a/drivers/net/dsa/mt7530.h
+++ b/drivers/net/dsa/mt7530.h
@@ -752,6 +752,8 @@ struct mt753x_info {
* @irq: IRQ number of the switch
* @irq_domain: IRQ domain of the switch irq_chip
* @irq_enable: IRQ enable bits, synced to SYS_INT_EN
+ *
+ * @create_sgmii: Pointer to function creating SGMII PCS instance(s)
*/
struct mt7530_priv {
struct device *dev;
@@ -778,6 +780,8 @@ struct mt7530_priv {
int irq;
struct irq_domain *irq_domain;
u32 irq_enable;
+
+ int (*create_sgmii)(struct mt7530_priv *priv, bool dual_sgmii);
};
struct mt7530_hw_vlan_entry {
--
2.40.0
On 4/10/2023 5:11 PM, Daniel Golle wrote:
> There are two variants of the MT7531 switch IC which got different
> features (and pins) regarding port 5:
> * MT7531AE: SGMII/1000Base-X/2500Base-X SerDes
> * MT7531BE: RGMII
>
> Moving the creation of the SerDes PCS from mt753x_setup to mt7530_probe
> with commit 6de285229773 ("net: dsa: mt7530: move SGMII PCS creation to
> mt7530_probe function") works fine for MT7531AE which got two instances
> of mtk-pcs-lynxi, however, MT7531BE requires mt7531_pll_setup to setup
> clocks before the single PCS on port 6 (usually used as CPU port)
> starts to work and hence the PCS creation failed on MT7531BE.
>
> Fix this by introducing a pointer to mt7531_create_sgmii function in
> struct mt7530_priv and call it again at the end of mt753x_setup like it
> was before commit 6de285229773 ("net: dsa: mt7530: move SGMII PCS
> creation to mt7530_probe function").
>
> Fixes: 6de285229773 ("net: dsa: mt7530: move SGMII PCS creation to mt7530_probe function")
> Signed-off-by: Daniel Golle <[email protected]>
> ---
> drivers/net/dsa/mt7530-mdio.c | 14 +++++++-------
> drivers/net/dsa/mt7530.c | 6 ++++++
> drivers/net/dsa/mt7530.h | 4 ++++
> 3 files changed, 17 insertions(+), 7 deletions(-)
>
> diff --git a/drivers/net/dsa/mt7530-mdio.c b/drivers/net/dsa/mt7530-mdio.c
> index 34a547b88e497..f17eab67d15fa 100644
> --- a/drivers/net/dsa/mt7530-mdio.c
> +++ b/drivers/net/dsa/mt7530-mdio.c
> @@ -81,14 +81,17 @@ static const struct regmap_bus mt7530_regmap_bus = {
> };
>
> static int
> -mt7531_create_sgmii(struct mt7530_priv *priv)
> +mt7531_create_sgmii(struct mt7530_priv *priv, bool dual_sgmii)
> {
> struct regmap_config *mt7531_pcs_config[2];
[1] use = {};
> struct phylink_pcs *pcs;
> struct regmap *regmap;
> int i, ret = 0;
>
> - for (i = 0; i < 2; i++) {
> + /* MT7531AE has two SGMII units for port 5 and port 6
> + * MT7531BE has only one SGMII unit for port 6
> + */
> + for (i = dual_sgmii ? 0 : 1; i < 2; i++) {
> mt7531_pcs_config[i] = devm_kzalloc(priv->dev,
hm, I don't like this, you're no longer initializing array[0] if
dual_sgmii is set. This seems like a recipe for disaster.
you could initialize the array to NULL above when allocating this on the
stack? [1]
> sizeof(struct regmap_config),
> GFP_KERNEL);
> @@ -208,11 +211,8 @@ mt7530_probe(struct mdio_device *mdiodev)
> if (IS_ERR(priv->regmap))
> return PTR_ERR(priv->regmap);
>
> - if (priv->id == ID_MT7531) {
> - ret = mt7531_create_sgmii(priv);
> - if (ret)
> - return ret;
> - }
> + if (priv->id == ID_MT7531)
> + priv->create_sgmii = mt7531_create_sgmii;
>
> return dsa_register_switch(priv->ds);
> }
> diff --git a/drivers/net/dsa/mt7530.c b/drivers/net/dsa/mt7530.c
> index e4bb5037d3525..c680873819b01 100644
> --- a/drivers/net/dsa/mt7530.c
> +++ b/drivers/net/dsa/mt7530.c
> @@ -3018,6 +3018,12 @@ mt753x_setup(struct dsa_switch *ds)
> if (ret && priv->irq)
> mt7530_free_irq_common(priv);
>
> + if (priv->create_sgmii) {
> + ret = priv->create_sgmii(priv, mt7531_dual_sgmii_supported(priv));
> + if (ret && priv->irq)
> + mt7530_free_irq(priv);
> + }
> +
> return ret;
> }
>
> diff --git a/drivers/net/dsa/mt7530.h b/drivers/net/dsa/mt7530.h
> index 01db5c9724fa8..6e89578b4de02 100644
> --- a/drivers/net/dsa/mt7530.h
> +++ b/drivers/net/dsa/mt7530.h
> @@ -752,6 +752,8 @@ struct mt753x_info {
> * @irq: IRQ number of the switch
> * @irq_domain: IRQ domain of the switch irq_chip
> * @irq_enable: IRQ enable bits, synced to SYS_INT_EN
> + *
no extra blank line needed here.
> + * @create_sgmii: Pointer to function creating SGMII PCS instance(s)
> */
> struct mt7530_priv {
> struct device *dev;
> @@ -778,6 +780,8 @@ struct mt7530_priv {
> int irq;
> struct irq_domain *irq_domain;
> u32 irq_enable;
> +
> + int (*create_sgmii)(struct mt7530_priv *priv, bool dual_sgmii);
> };
>
> struct mt7530_hw_vlan_entry {
On 11.04.2023 03:11, Daniel Golle wrote:
> There are two variants of the MT7531 switch IC which got different
> features (and pins) regarding port 5:
> * MT7531AE: SGMII/1000Base-X/2500Base-X SerDes
> * MT7531BE: RGMII
>
> Moving the creation of the SerDes PCS from mt753x_setup to mt7530_probe
> with commit 6de285229773 ("net: dsa: mt7530: move SGMII PCS creation to
> mt7530_probe function") works fine for MT7531AE which got two instances
> of mtk-pcs-lynxi, however, MT7531BE requires mt7531_pll_setup to setup
> clocks before the single PCS on port 6 (usually used as CPU port)
> starts to work and hence the PCS creation failed on MT7531BE.
>
> Fix this by introducing a pointer to mt7531_create_sgmii function in
> struct mt7530_priv and call it again at the end of mt753x_setup like it
> was before commit 6de285229773 ("net: dsa: mt7530: move SGMII PCS
> creation to mt7530_probe function").
If I understand correctly, this patch does two things.
Run mt7531_create_sgmii() from mt753x_setup(), after mt7531_setup() and
mt7531_setup_common() is run so that PCS on MT7531BE works.
Run the PCS creation code inside the loop only once if
mt7531_dual_sgmii_supported() is false so it doesn't set the nonexistent
port 5 SGMII on MT7531BE.
Regarding the first part:
I was actually in the middle of moving the code until after
mt7530_pll_setup() and mt7531_pll_setup() on mt7530_setup() and
mt7531_setup() to mt7530_probe(). To me it makes more sense to run them
on mt7530_probe() as there's a good amount of duplicate code on
mt7530_setup() and mt7531_setup().
This will resolve the problem here, and make my future work regarding
the PHY muxing feature on the MT7530 switch possible to do.
Regarding the second part:
I'll take your changes to my current RFC patch series while addressing
Jesse's suggestion if this is fine by you.
Arınç
On Tue, Apr 11, 2023 at 10:30:06PM +0300, Arınç ÜNAL wrote:
> On 11.04.2023 03:11, Daniel Golle wrote:
> > There are two variants of the MT7531 switch IC which got different
> > features (and pins) regarding port 5:
> > * MT7531AE: SGMII/1000Base-X/2500Base-X SerDes
> > * MT7531BE: RGMII
> >
> > Moving the creation of the SerDes PCS from mt753x_setup to mt7530_probe
> > with commit 6de285229773 ("net: dsa: mt7530: move SGMII PCS creation to
> > mt7530_probe function") works fine for MT7531AE which got two instances
> > of mtk-pcs-lynxi, however, MT7531BE requires mt7531_pll_setup to setup
> > clocks before the single PCS on port 6 (usually used as CPU port)
> > starts to work and hence the PCS creation failed on MT7531BE.
> >
> > Fix this by introducing a pointer to mt7531_create_sgmii function in
> > struct mt7530_priv and call it again at the end of mt753x_setup like it
> > was before commit 6de285229773 ("net: dsa: mt7530: move SGMII PCS
> > creation to mt7530_probe function").
>
> If I understand correctly, this patch does two things.
>
> Run mt7531_create_sgmii() from mt753x_setup(), after mt7531_setup() and
> mt7531_setup_common() is run so that PCS on MT7531BE works.
>
> Run the PCS creation code inside the loop only once if
> mt7531_dual_sgmii_supported() is false so it doesn't set the nonexistent
> port 5 SGMII on MT7531BE.
Yes, both is correct.
>
> Regarding the first part:
> I was actually in the middle of moving the code until after
> mt7530_pll_setup() and mt7531_pll_setup() on mt7530_setup() and
> mt7531_setup() to mt7530_probe(). To me it makes more sense to run them on
> mt7530_probe() as there's a good amount of duplicate code on mt7530_setup()
> and mt7531_setup().
I thought about doing that as well, however, note that you will have to
move all the reset and regulator setup procedure to mt7530_probe() as
well then, as PLL setup currently happens after that, and that's
probably for a reason.
As the reset and regulator setup works differently on MT7530 and
MT7531, and depending on whether it's a standalone IC package or MCM, I
believe changes unifying this will have to be tested on a lot of
boards...
>
> This will resolve the problem here, and make my future work regarding the
> PHY muxing feature on the MT7530 switch possible to do.
>
> Regarding the second part:
> I'll take your changes to my current RFC patch series while addressing
> Jesse's suggestion if this is fine by you.
Yes, I'd appreciate that and I'm ready to test and review once you post
your updated series.
On 11.04.2023 22:43, Daniel Golle wrote:
> On Tue, Apr 11, 2023 at 10:30:06PM +0300, Arınç ÜNAL wrote:
>> On 11.04.2023 03:11, Daniel Golle wrote:
>>> There are two variants of the MT7531 switch IC which got different
>>> features (and pins) regarding port 5:
>>> * MT7531AE: SGMII/1000Base-X/2500Base-X SerDes
>>> * MT7531BE: RGMII
>>>
>>> Moving the creation of the SerDes PCS from mt753x_setup to mt7530_probe
>>> with commit 6de285229773 ("net: dsa: mt7530: move SGMII PCS creation to
>>> mt7530_probe function") works fine for MT7531AE which got two instances
>>> of mtk-pcs-lynxi, however, MT7531BE requires mt7531_pll_setup to setup
>>> clocks before the single PCS on port 6 (usually used as CPU port)
>>> starts to work and hence the PCS creation failed on MT7531BE.
>>>
>>> Fix this by introducing a pointer to mt7531_create_sgmii function in
>>> struct mt7530_priv and call it again at the end of mt753x_setup like it
>>> was before commit 6de285229773 ("net: dsa: mt7530: move SGMII PCS
>>> creation to mt7530_probe function").
>>
>> If I understand correctly, this patch does two things.
>>
>> Run mt7531_create_sgmii() from mt753x_setup(), after mt7531_setup() and
>> mt7531_setup_common() is run so that PCS on MT7531BE works.
>
>>
>> Run the PCS creation code inside the loop only once if
>> mt7531_dual_sgmii_supported() is false so it doesn't set the nonexistent
>> port 5 SGMII on MT7531BE.
>
> Yes, both is correct.
>
>>
>> Regarding the first part:
>> I was actually in the middle of moving the code until after
>> mt7530_pll_setup() and mt7531_pll_setup() on mt7530_setup() and
>> mt7531_setup() to mt7530_probe(). To me it makes more sense to run them on
>> mt7530_probe() as there's a good amount of duplicate code on mt7530_setup()
>> and mt7531_setup().
>
> I thought about doing that as well, however, note that you will have to
> move all the reset and regulator setup procedure to mt7530_probe() as
> well then, as PLL setup currently happens after that, and that's
> probably for a reason.
>
> As the reset and regulator setup works differently on MT7530 and
> MT7531, and depending on whether it's a standalone IC package or MCM, I
> believe changes unifying this will have to be tested on a lot of
> boards...
Not if we don't change the behaviour at all for both switches, which is
what I intend to do.
>
>>
>> This will resolve the problem here, and make my future work regarding the
>> PHY muxing feature on the MT7530 switch possible to do.
>>
>> Regarding the second part:
>> I'll take your changes to my current RFC patch series while addressing
>> Jesse's suggestion if this is fine by you.
>
> Yes, I'd appreciate that and I'm ready to test and review once you post
> your updated series.
Sounds good, cheers.
Arınç