Skip to contentSkip to conversation

Remove legacy phylink behaviour

PATCH Applied Russell King (Oracle) 3 participants 6 messages Jul 22, 2023 – Jul 25, 2023

Subsystems (6)
lore
Share & tools
lore

Russell King (Oracle)

author maintainer
#0 lore raw

v2 cover Remove legacy phylink behaviour

Hi,

This series removes the - as far as I can tell - unreachable code in mtk_eth_soc that relies upon legacy phylink behaviour, and then removes the support in phylink for this legacy behaviour.

Patch 1 removes the clocking configuration from mtk_eth_soc for non- TRGMII, non-serdes based interface modes, and disables those interface modes prior to phylink configuration.

Patch 2 removes the mac_pcs_get_state() method from mtk_eth_soc which I believe is also not used - mtk_eth_soc appears not to be used with SFPs (which would use a kind of in-band mode) nor does any DT appear to specify in-band mode for any non-serdes based interface mode.

With both of those dealt with, the kernel is now free of any driver relying on the phylink legacy mode. Therefore, patch 3 removes support for this.

Finally, with the advent of a new driver being submitted today that makes use of state->speed in the mac_config() path, patch 4 ensures that any phylink_link_state member that should not be used in mac_config is either cleared or set to an invalid value.

v2: rebase on net-next, add attributations for testing etc

 drivers/net/ethernet/mediatek/mtk_eth_soc.c | 94 +++++------------------------
 drivers/net/ethernet/mediatek/mtk_eth_soc.h |  1 +
 drivers/net/phy/phylink.c                   | 48 ++++++---------
 include/linux/phylink.h                     | 45 ++------------
 4 files changed, 42 insertions(+), 146 deletions(-)
Signature · 2 lines
RMK's Patch system: https://www.armlinux.org.uk/developer/patches/
FTTP is here! 80Mbps down 10Mbps up. Decent connectivity at last!
#1 lore raw

PATCH v2 1/4 net: ethernet: mtk_eth_soc: remove incorrect PLL configuration

MT7623 GMAC0 attempts to configure the system clocking according to the required speed in the .mac_config callback for non-SGMII, non-baseX and non-TRGMII modes.

state->speed setting has never been reliable in the .mac_config callback - there are cases where this is not the link speed, particularly via ethtool paths, so this has always been unreliable (as detailed in phylink's documentation.)

There is the additional issue that mtk_gmac0_rgmii_adjust() will only be called if state->interface changes, which means it only configures the system clocking on the very first .mac_config call, which will be made when the network device is first brought up before any link is established.

Essentially, this code is incredibly buggy, and probably never worked.

Moreover, checking the in-kernel DT files, it seems no platform makes use of this code path.

Therefore, let's remove it, and disable interface modes for port 0 that are not SGMII, 1000base-X, 2500base-X or TRGMII on the MT7623.

---
 drivers/net/ethernet/mediatek/mtk_eth_soc.c | 59 ++++++---------------
 drivers/net/ethernet/mediatek/mtk_eth_soc.h |  1 +
 2 files changed, 17 insertions(+), 43 deletions(-)
drivers/net/ethernet/mediatek/mtk_eth_soc.c +16 -43
index 18a8aca7944d..fe8b7e38decc 100644
@@ -385,10 +385,8 @@ static int mt7621_gmac0_rgmii_adjust(struct mtk_eth *eth,
 }
 
 static void mtk_gmac0_rgmii_adjust(struct mtk_eth *eth,
-				   phy_interface_t interface, int speed)
+				   phy_interface_t interface)
 {
-	unsigned long rate;
-	u32 tck, rck, intf;
 	int ret;
 
 	if (interface == PHY_INTERFACE_MODE_TRGMII) {
@@ -399,30 +397,7 @@ static void mtk_gmac0_rgmii_adjust(struct mtk_eth *eth,
 		return;
 	}
 
-	if (speed == SPEED_1000) {
-		intf = INTF_MODE_RGMII_1000;
-		rate = 250000000;
-		rck = RCK_CTRL_RGMII_1000;
-		tck = TCK_CTRL_RGMII_1000;
-	} else {
-		intf = INTF_MODE_RGMII_10_100;
-		rate = 500000000;
-		rck = RCK_CTRL_RGMII_10_100;
-		tck = TCK_CTRL_RGMII_10_100;
-	}
-
-	mtk_w32(eth, intf, INTF_MODE);
-
-	regmap_update_bits(eth->ethsys, ETHSYS_CLKCFG0,
-			   ETHSYS_TRGMII_CLK_SEL362_5,
-			   ETHSYS_TRGMII_CLK_SEL362_5);
-
-	ret = clk_set_rate(eth->clks[MTK_CLK_TRGPLL], rate);
-	if (ret)
-		dev_err(eth->dev, "Failed to set trgmii pll: %d\n", ret);
-
-	mtk_w32(eth, rck, TRGMII_RCK_CTRL);
-	mtk_w32(eth, tck, TRGMII_TCK_CTRL);
+	dev_err(eth->dev, "Missing PLL configuration, ethernet may not work\n");
 }
 
 static struct phylink_pcs *mtk_mac_select_pcs(struct phylink_config *config,
@@ -498,17 +473,8 @@ static void mtk_mac_config(struct phylink_config *config, unsigned int mode,
 							      state->interface))
 					goto err_phy;
 			} else {
-				/* FIXME: this is incorrect. Not only does it
-				 * use state->speed (which is not guaranteed
-				 * to be correct) but it also makes use of it
-				 * in a code path that will only be reachable
-				 * when the PHY interface mode changes, not
-				 * when the speed changes. Consequently, RGMII
-				 * is probably broken.
-				 */
 				mtk_gmac0_rgmii_adjust(mac->hw,
-						       state->interface,
-						       state->speed);
+						       state->interface);
 
 				/* mt7623_pad_clk_setup */
 				for (i = 0 ; i < NUM_TRGMII_CTRL; i++)
@@ -4366,13 +4332,19 @@ static int mtk_add_mac(struct mtk_eth *eth, struct device_node *np)
 	mac->phylink_config.mac_capabilities = MAC_ASYM_PAUSE | MAC_SYM_PAUSE |
 		MAC_10 | MAC_100 | MAC_1000 | MAC_2500FD;
 
-	__set_bit(PHY_INTERFACE_MODE_MII,
-		  mac->phylink_config.supported_interfaces);
-	__set_bit(PHY_INTERFACE_MODE_GMII,
-		  mac->phylink_config.supported_interfaces);
+	/* MT7623 gmac0 is now missing its speed-specific PLL configuration
+	 * in its .mac_config method (since state->speed is not valid there.
+	 * Disable support for MII, GMII and RGMII.
+	 */
+	if (!mac->hw->soc->disable_pll_modes || mac->id != 0) {
+		__set_bit(PHY_INTERFACE_MODE_MII,
+			  mac->phylink_config.supported_interfaces);
+		__set_bit(PHY_INTERFACE_MODE_GMII,
+			  mac->phylink_config.supported_interfaces);
 
-	if (MTK_HAS_CAPS(mac->hw->soc->caps, MTK_RGMII))
-		phy_interface_set_rgmii(mac->phylink_config.supported_interfaces);
+		if (MTK_HAS_CAPS(mac->hw->soc->caps, MTK_RGMII))
+			phy_interface_set_rgmii(mac->phylink_config.supported_interfaces);
+	}
 
 	if (MTK_HAS_CAPS(mac->hw->soc->caps, MTK_TRGMII) && !mac->id)
 		__set_bit(PHY_INTERFACE_MODE_TRGMII,
@@ -4845,6 +4817,7 @@ static const struct mtk_soc_data mt7623_data = {
 	.offload_version = 1,
 	.hash_offset = 2,
 	.foe_entry_size = MTK_FOE_ENTRY_V1_SIZE,
+	.disable_pll_modes = true,
 	.txrx = {
 		.txd_size = sizeof(struct mtk_tx_dma),
 		.rxd_size = sizeof(struct mtk_rx_dma),
drivers/net/ethernet/mediatek/mtk_eth_soc.h +1 -0
index 707445f6bcb1..28adda0c90c0 100644
@@ -1030,6 +1030,7 @@ struct mtk_soc_data {
 	u16		foe_entry_size;
 	netdev_features_t hw_features;
 	bool		has_accounting;
+	bool		disable_pll_modes;
 	struct {
 		u32	txd_size;
 		u32	rxd_size;
Signature · 1 line
2.30.2
#2 lore raw

PATCH v2 2/4 net: ethernet: mtk_eth_soc: remove mac_pcs_get_state and modernise

In reply to Russell King (Oracle)

Remove the .mac_pcs_get_state function, since as far as I can tell is never called - no DT appears to specify an in-band-status management nor SFP support for this driver.

Removal of this, along with the previous patch to remove the incorrect clocking configuration, means that the driver becomes non-legacy, so we can remove the "legacy_pre_march2020" status from this driver.

---
 drivers/net/ethernet/mediatek/mtk_eth_soc.c | 35 ---------------------
 1 file changed, 35 deletions(-)
drivers/net/ethernet/mediatek/mtk_eth_soc.c +0 -35
index fe8b7e38decc..7490d48000c2 100644
@@ -568,38 +568,6 @@ static int mtk_mac_finish(struct phylink_config *config, unsigned int mode,
 	return 0;
 }
 
-static void mtk_mac_pcs_get_state(struct phylink_config *config,
-				  struct phylink_link_state *state)
-{
-	struct mtk_mac *mac = container_of(config, struct mtk_mac,
-					   phylink_config);
-	u32 pmsr = mtk_r32(mac->hw, MTK_MAC_MSR(mac->id));
-
-	state->link = (pmsr & MAC_MSR_LINK);
-	state->duplex = (pmsr & MAC_MSR_DPX) >> 1;
-
-	switch (pmsr & (MAC_MSR_SPEED_1000 | MAC_MSR_SPEED_100)) {
-	case 0:
-		state->speed = SPEED_10;
-		break;
-	case MAC_MSR_SPEED_100:
-		state->speed = SPEED_100;
-		break;
-	case MAC_MSR_SPEED_1000:
-		state->speed = SPEED_1000;
-		break;
-	default:
-		state->speed = SPEED_UNKNOWN;
-		break;
-	}
-
-	state->pause &= (MLO_PAUSE_RX | MLO_PAUSE_TX);
-	if (pmsr & MAC_MSR_RX_FC)
-		state->pause |= MLO_PAUSE_RX;
-	if (pmsr & MAC_MSR_TX_FC)
-		state->pause |= MLO_PAUSE_TX;
-}
-
 static void mtk_mac_link_down(struct phylink_config *config, unsigned int mode,
 			      phy_interface_t interface)
 {
@@ -722,7 +690,6 @@ static void mtk_mac_link_up(struct phylink_config *config,
 
 static const struct phylink_mac_ops mtk_phylink_ops = {
 	.mac_select_pcs = mtk_mac_select_pcs,
-	.mac_pcs_get_state = mtk_mac_pcs_get_state,
 	.mac_config = mtk_mac_config,
 	.mac_finish = mtk_mac_finish,
 	.mac_link_down = mtk_mac_link_down,
@@ -4327,8 +4294,6 @@ static int mtk_add_mac(struct mtk_eth *eth, struct device_node *np)
 
 	mac->phylink_config.dev = &eth->netdev[id]->dev;
 	mac->phylink_config.type = PHYLINK_NETDEV;
-	/* This driver makes use of state->speed in mac_config */
-	mac->phylink_config.legacy_pre_march2020 = true;
 	mac->phylink_config.mac_capabilities = MAC_ASYM_PAUSE | MAC_SYM_PAUSE |
 		MAC_10 | MAC_100 | MAC_1000 | MAC_2500FD;
 
Signature · 1 line
2.30.2
#3 lore raw

PATCH v2 3/4 net: phylink: strip out pre-March 2020 legacy code

In reply to Russell King (Oracle)

Strip out all the pre-March 2020 legacy code from phylink now that the last user of it is gone.

---
 drivers/net/phy/phylink.c | 25 ++++------------------
 include/linux/phylink.h   | 45 ++++++---------------------------------
 2 files changed, 10 insertions(+), 60 deletions(-)
drivers/net/phy/phylink.c +4 -21
index f07e496319b4..df413fb15088 100644
@@ -1198,13 +1198,6 @@ static int phylink_change_inband_advert(struct phylink *pl)
 	if (test_bit(PHYLINK_DISABLE_STOPPED, &pl->phylink_disable_state))
 		return 0;
 
-	if (!pl->pcs && pl->config->legacy_pre_march2020) {
-		/* Legacy method */
-		phylink_mac_config(pl, &pl->link_config);
-		phylink_pcs_an_restart(pl);
-		return 0;
-	}
-
 	phylink_dbg(pl, "%s: mode=%s/%s adv=%*pb pause=%02x\n", __func__,
 		    phylink_an_mode_str(pl->cur_link_an_mode),
 		    phy_modes(pl->link_config.interface),
@@ -1257,9 +1250,6 @@ static void phylink_mac_pcs_get_state(struct phylink *pl,
 
 	if (pl->pcs)
 		pl->pcs->ops->pcs_get_state(pl->pcs, state);
-	else if (pl->mac_ops->mac_pcs_get_state &&
-		 pl->config->legacy_pre_march2020)
-		pl->mac_ops->mac_pcs_get_state(pl->config, state);
 	else
 		state->link = 0;
 }
@@ -1492,13 +1482,6 @@ static void phylink_resolve(struct work_struct *w)
 			}
 			phylink_major_config(pl, false, &link_state);
 			pl->link_config.interface = link_state.interface;
-		} else if (!pl->pcs && pl->config->legacy_pre_march2020) {
-			/* The interface remains unchanged, only the speed,
-			 * duplex or pause settings have changed. Call the
-			 * old mac_config() method to configure the MAC/PCS
-			 * only if we do not have a legacy MAC driver.
-			 */
-			phylink_mac_config(pl, &link_state);
 		}
 	}
 
@@ -3513,7 +3496,7 @@ static void phylink_decode_usgmii_word(struct phylink_link_state *state,
  *
  * Parse the Clause 37 or Cisco SGMII link partner negotiation word into
  * the phylink @state structure. This is suitable to be used for implementing
- * the mac_pcs_get_state() member of the struct phylink_mac_ops structure if
+ * the pcs_get_state() member of the struct phylink_pcs_ops structure if
  * accessing @bmsr and @lpa cannot be done with MDIO directly.
  */
 void phylink_mii_c22_pcs_decode_state(struct phylink_link_state *state,
@@ -3563,7 +3546,7 @@ EXPORT_SYMBOL_GPL(phylink_mii_c22_pcs_decode_state);
  * Read the MAC PCS state from the MII device configured in @config and
  * parse the Clause 37 or Cisco SGMII link partner negotiation word into
  * the phylink @state structure. This is suitable to be directly plugged
- * into the mac_pcs_get_state() member of the struct phylink_mac_ops
+ * into the pcs_get_state() member of the struct phylink_pcs_ops
  * structure.
  */
 void phylink_mii_c22_pcs_get_state(struct mdio_device *pcs,
@@ -3674,8 +3657,8 @@ EXPORT_SYMBOL_GPL(phylink_mii_c22_pcs_config);
  * clause 37 negotiation.
  *
  * Restart the clause 37 negotiation with the link partner. This is
- * suitable to be directly plugged into the mac_pcs_get_state() member
- * of the struct phylink_mac_ops structure.
+ * suitable to be directly plugged into the pcs_get_state() member
+ * of the struct phylink_pcs_ops structure.
  */
 void phylink_mii_c22_pcs_an_restart(struct mdio_device *pcs)
 {
include/linux/phylink.h +6 -39

93 lines not on this page · raw message

Signature · 1 line
2.30.2
#4 lore raw

PATCH v2 4/4 net: phylink: explicitly invalidate link_state members in mac_config

In reply to Russell King (Oracle)

Explicitly invalidate the phylink_link_state structure members in mac_config that do not contain reliable information for this function, thereby preventing their future incorrect use.

---
 drivers/net/phy/phylink.c | 25 ++++++++++++++++---------
 1 file changed, 16 insertions(+), 9 deletions(-)
drivers/net/phy/phylink.c +16 -9
index df413fb15088..4f1c8bb199e9 100644
@@ -1066,17 +1066,24 @@ static void phylink_pcs_poll_start(struct phylink *pl)
 static void phylink_mac_config(struct phylink *pl,
 			       const struct phylink_link_state *state)
 {
+	struct phylink_link_state st = *state;
+
+	/* Stop drivers incorrectly using these */
+	linkmode_zero(st.lp_advertising);
+	st.speed = SPEED_UNKNOWN;
+	st.duplex = DUPLEX_UNKNOWN;
+	st.an_complete = false;
+	st.link = false;
+
 	phylink_dbg(pl,
-		    "%s: mode=%s/%s/%s/%s/%s adv=%*pb pause=%02x link=%u\n",
+		    "%s: mode=%s/%s/%s adv=%*pb pause=%02x\n",
 		    __func__, phylink_an_mode_str(pl->cur_link_an_mode),
-		    phy_modes(state->interface),
-		    phy_speed_to_str(state->speed),
-		    phy_duplex_to_str(state->duplex),
-		    phy_rate_matching_to_str(state->rate_matching),
-		    __ETHTOOL_LINK_MODE_MASK_NBITS, state->advertising,
-		    state->pause, state->link);
-
-	pl->mac_ops->mac_config(pl->config, pl->cur_link_an_mode, state);
+		    phy_modes(st.interface),
+		    phy_rate_matching_to_str(st.rate_matching),
+		    __ETHTOOL_LINK_MODE_MASK_NBITS, st.advertising,
+		    st.pause);
+
+	pl->mac_ops->mac_config(pl->config, pl->cur_link_an_mode, &st);
 }
 
 static void phylink_pcs_an_restart(struct phylink *pl)
Signature · 1 line
2.30.2
#5 lore raw
In reply to Russell King (Oracle)
This series was applied to netdev/net-next.git (main) by Paolo Abeni <[email protected]>:

Hello:

This series was applied to netdev/net-next.git (main) by Paolo Abeni <[email protected]>:

Quoted 11 lines · On Sat, 22 Jul 2023 21:29:02 +0100 you wrote:

Hi,

This series removes the - as far as I can tell - unreachable code in mtk_eth_soc that relies upon legacy phylink behaviour, and then removes the support in phylink for this legacy behaviour.

Patch 1 removes the clocking configuration from mtk_eth_soc for non- TRGMII, non-serdes based interface modes, and disables those interface modes prior to phylink configuration.

[...]

Here is the summary with links:
- [net-next,v2,1/4] net: ethernet: mtk_eth_soc: remove incorrect PLL configuration

    https://git.kernel.org/netdev/net-next/c/76a4cb755cf9
  - [net-next,v2,2/4] net: ethernet: mtk_eth_soc: remove mac_pcs_get_state and modernise
    https://git.kernel.org/netdev/net-next/c/28e1737d2544
  - [net-next,v2,3/4] net: phylink: strip out pre-March 2020 legacy code
    https://git.kernel.org/netdev/net-next/c/4d72c3bb60dd
  - [net-next,v2,4/4] net: phylink: explicitly invalidate link_state members in mac_config
    https://git.kernel.org/netdev/net-next/c/c5714f68a76b

You are awesome, thank you!

Signature · 2 lines
Deet-doot-dot, I am a bot.
https://korg.docs.kernel.org/patchwork/pwbot.html