[RFC PATCH v2 net-next 00/12] net: dsa: mt7530: fix port 5 phylink, port 6, and MT7988

[arinc9.unal at gmail.com]

Hello!

This patch series is mainly focused on improving the support for port 5,
setting up port 6, and refactoring the MT7530 DSA subdriver. There're also
fixes for the switch on the MT7988 SoC.

I'm asking for your comments on patch 4 and 9.

For patch 4:

If you think priv->p5_interface should not be set when port 5 is used for
PHY muxing, let me know.

For patch 9:

Do I need to protect the register from being accessed by processes while
this operation is being done? I don't see this on mt7530_setup() but it's
being done on mt7530_setup_port5().

There's an oddity here. The XTAL mask is defined on the MT7530_HWTRAP
register, but it's being read from MT7530_MHWTRAP instead which is at a
different address.

HWTRAP_XTAL_MASK and reading HWTRAP_XTAL_MASK from MT7530_MHWTRAP was both
added with:

https://git.kernel.org/pub/scm/linux/kernel/git/netdev/net-next.git/commit?id=7ef6f6f8d237fa6724108b57d9706cb5069688e4

I did this to test it:

diff --git a/drivers/net/dsa/mt7530.c b/drivers/net/dsa/mt7530.c
index 46749aee3c49..7aa3b5828ac3 100644
--- a/drivers/net/dsa/mt7530.c
+++ b/drivers/net/dsa/mt7530.c
@@ -404,7 +404,7 @@ static int
 mt7530_setup_port6(struct dsa_switch *ds, phy_interface_t interface)
 {
 	struct mt7530_priv *priv = ds->priv;
-	u32 ncpo1, ssc_delta, trgint, xtal, val;
+	u32 ncpo1, ssc_delta, trgint, xtal, xtal2, val;
 
 	/* Enable port 6 */
 	val = mt7530_read(priv, MT7530_MHWTRAP);
@@ -413,6 +413,21 @@ mt7530_setup_port6(struct dsa_switch *ds, phy_interface_t interface)
 	mt7530_write(priv, MT7530_MHWTRAP, val);
 
 	xtal = val & HWTRAP_XTAL_MASK;
+	xtal2 = mt7530_read(priv, MT7530_HWTRAP) & HWTRAP_XTAL_MASK;
+
+	if (xtal == HWTRAP_XTAL_20MHZ)
+		dev_info(priv->dev, "xtal 20 Mhz\n");
+	if (xtal == HWTRAP_XTAL_25MHZ)
+		dev_info(priv->dev, "xtal 25 Mhz\n");
+	if (xtal == HWTRAP_XTAL_40MHZ)
+		dev_info(priv->dev, "xtal 40 Mhz\n");
+
+	if (xtal2 == HWTRAP_XTAL_20MHZ)
+		dev_info(priv->dev, "actual xtal 20 Mhz\n");
+	if (xtal2 == HWTRAP_XTAL_25MHZ)
+		dev_info(priv->dev, "actual xtal 25 Mhz\n");
+	if (xtal2 == HWTRAP_XTAL_40MHZ)
+		dev_info(priv->dev, "actual xtal 40 Mhz\n");
 
 	if (xtal == HWTRAP_XTAL_20MHZ) {
 		dev_err(priv->dev,

Both ended up reporting 40 Mhz so I'm not sure if this is a bug or intended
to be done this way.

Please advise.

The only missing piece to properly support port 5 as a CPU port is the
fixes [0] [1] [2] from Richard.

I have very thoroughly tested the patch series with every possible mode to
use. I'll let the name of the dtb files speak for themselves.

MT7621 Unielec:

only-gmac0-mt7621-unielec-u7621-06-16m.dtb
rgmii-only-gmac0-mt7621-unielec-u7621-06-16m.dtb
only-gmac1-mt7621-unielec-u7621-06-16m.dtb
gmac0-and-gmac1-mt7621-unielec-u7621-06-16m.dtb
phy0-muxing-mt7621-unielec-u7621-06-16m.dtb
phy4-muxing-mt7621-unielec-u7621-06-16m.dtb
port5-as-user-mt7621-unielec-u7621-06-16m.dtb

tftpboot 0x80008000 mips-uzImage.bin; tftpboot 0x83000000 mips-rootfs.cpio.uboot; tftpboot 0x83f00000 $dtb; bootm 0x80008000 0x83000000 0x83f00000

MT7623 Bananapi:

only-gmac0-mt7623n-bananapi-bpi-r2.dtb
rgmii-only-gmac0-mt7623n-bananapi-bpi-r2.dtb
only-gmac1-mt7623n-bananapi-bpi-r2.dtb
gmac0-and-gmac1-mt7623n-bananapi-bpi-r2.dtb
phy0-muxing-mt7623n-bananapi-bpi-r2.dtb
phy4-muxing-mt7623n-bananapi-bpi-r2.dtb
port5-as-user-mt7623n-bananapi-bpi-r2.dtb

tftpboot 0x80008000 arm-uImage; tftpboot 0x83000000 arm-rootfs.cpio.uboot; tftpboot 0x83f00000 $dtb; bootm 0x80008000 0x83000000 0x83f00000

Current CPU ports setup of MT7530:

mt7530_setup()
-> mt7530_setup_port5()

mt753x_phylink_mac_config()
-> mt753x_mac_config()
   -> mt7530_mac_config()
      -> mt7530_setup_port5()
-> mt753x_pad_setup()
   -> mt7530_pad_clk_setup() sets up port 6, rename to mt7530_setup_port6()

How it will be with the patch series:

mt7530_setup()
-> mt7530_setup_port5() runs if the port is not used as a CPU, DSA, or user port

mt753x_phylink_mac_config()
-> mt753x_mac_config()
   -> mt7530_mac_config()
      -> mt7530_setup_port5()
      -> mt7530_setup_port6()

CPU ports setup of MT7531 for reference:

mt7531_setup()
-> mt753x_cpu_port_enable()
   -> mt7531_cpu_port_config()
      -> mt7531_mac_config()
         -> mt7531_rgmii_setup()
         -> mt7531_sgmii_setup_mode_an()
         -> etc.

mt753x_phylink_mac_config()
-> mt753x_mac_config()
   -> mt7531_mac_config()
      -> mt7531_rgmii_setup()
      -> mt7531_sgmii_setup_mode_an()
      -> etc.

[0] https://lore.kernel.org/netdev/20230212213949.672443-1-richard@routerhints.com/
[1] https://lore.kernel.org/netdev/20230212215152.673221-1-richard@routerhints.com/
[2] https://lore.kernel.org/netdev/20230212214027.672501-1-richard@routerhints.com/

Arınç

RFC v2: Add two patches related to MT7988 to the end.