On 4/8/2023 5:02 pm, Russell King (Oracle) wrote:
On Fri, Aug 04, 2023 at 04:45:25PM +0800, Choong Yong Liang wrote:
From: "Tan, Tee Min" <tee.min.tan@xxxxxxxxxxxxxxx>
Add cur_link_an_mode into phy_device struct for PHY drivers to
communicate the in-band AN mode setting with phylink framework.
As there is a mechanism in PHY drivers to switch the PHY interface
between SGMII and 2500BaseX according to link speed. In this case,
the in-band AN mode should be switching based on the PHY interface
as well, if the PHY interface has been changed/updated by PHY driver.
For e.g., disable in-band AN in 2500BaseX mode, or enable in-band AN
back for SGMII mode (10/100/1000Mbps).
Signed-off-by: Tan, Tee Min <tee.min.tan@xxxxxxxxxxxxxxx>
Signed-off-by: Choong Yong Liang <yong.liang.choong@xxxxxxxxxxxxxxx>
---
drivers/net/phy/marvell10g.c | 6 ++++++
drivers/net/phy/phylink.c | 4 ++++
include/linux/phy.h | 3 +++
3 files changed, 13 insertions(+)
diff --git a/drivers/net/phy/marvell10g.c b/drivers/net/phy/marvell10g.c
index d4bb90d76881..a9df19278618 100644
--- a/drivers/net/phy/marvell10g.c
+++ b/drivers/net/phy/marvell10g.c
@@ -30,6 +30,7 @@
#include <linux/phy.h>
#include <linux/sfp.h>
#include <linux/netdevice.h>
+#include <linux/phylink.h>
#define MV_PHY_ALASKA_NBT_QUIRK_MASK 0xfffffffe
#define MV_PHY_ALASKA_NBT_QUIRK_REV (MARVELL_PHY_ID_88X3310 | 0xa)
@@ -946,6 +947,9 @@ static void mv3310_update_interface(struct phy_device *phydev)
* xaui / rxaui modes according to the speed.
* Florian suggests setting phydev->interface to communicate this to the
* MAC. Only do this if we are already in one of the above modes.
+ * In-band Auto-negotiation is not supported in 2500BASE-X.
+ * Setting phydev->cur_link_an_mode to communicate this to the
+ * phylink framework.
*/
switch (phydev->speed) {
case SPEED_10000:
@@ -956,11 +960,13 @@ static void mv3310_update_interface(struct phy_device *phydev)
break;
case SPEED_2500:
phydev->interface = PHY_INTERFACE_MODE_2500BASEX;
+ phydev->cur_link_an_mode = MLO_AN_PHY;
break;
case SPEED_1000:
case SPEED_100:
case SPEED_10:
phydev->interface = PHY_INTERFACE_MODE_SGMII;
+ phydev->cur_link_an_mode = MLO_AN_INBAND;
break;
default:
break;
diff --git a/drivers/net/phy/phylink.c b/drivers/net/phy/phylink.c
index 4f1c8bb199e9..f9cbb6d7e134 100644
--- a/drivers/net/phy/phylink.c
+++ b/drivers/net/phy/phylink.c
@@ -1720,6 +1720,8 @@ static void phylink_phy_change(struct phy_device *phydev, bool up)
pl->phy_state.pause |= MLO_PAUSE_RX;
pl->phy_state.interface = phydev->interface;
pl->phy_state.link = up;
+ pl->cur_link_an_mode = phydev->cur_link_an_mode;
+ pl->cfg_link_an_mode = phydev->cur_link_an_mode;
mutex_unlock(&pl->state_mutex);
phylink_run_resolve(pl);
@@ -1824,6 +1826,8 @@ static int phylink_bringup_phy(struct phylink *pl, struct phy_device *phy,
if (pl->config->mac_managed_pm)
phy->mac_managed_pm = true;
+ pl->phydev->cur_link_an_mode = pl->cur_link_an_mode;
I am really not happy with exposing phylink's AN mode into phylib.
Hi Russell,
Thank you for the feedback.
After conducting further analysis on my end, it appears that this line
"pl->phydev->cur_link_an_mode = pl->cur_link_an_mode;" is not necessary.
If we remove this line of code, would the implementation of this patch be
acceptable to you?
