[openwrt/openwrt] generic: add pending patches for RealTek Ethernet PHYs
LEDE Commits
lede-commits at lists.infradead.org
Fri Jan 17 04:22:12 PST 2025
noltari pushed a commit to openwrt/openwrt.git, branch main:
https://git.openwrt.org/41c164d32bedb6b27ad1fd022dec6d25338db6d6
commit 41c164d32bedb6b27ad1fd022dec6d25338db6d6
Author: Daniel Golle <daniel at makrotopia.org>
AuthorDate: Thu Jan 16 20:11:43 2025 +0000
generic: add pending patches for RealTek Ethernet PHYs
Import patches to fix several issues with status reporting of RealTek
2.5G PHYs.
Signed-off-by: Daniel Golle <daniel at makrotopia.org>
---
...ltek-clear-1000Base-T-lpa-bits-if-link-is.patch | 78 ++++++++++++++++++++++
...ltek-clear-master_slave_state-if-link-is-.patch | 61 +++++++++++++++++
...ltek-always-clear-10gbase-t-link-partner-.patch | 67 +++++++++++++++++++
3 files changed, 206 insertions(+)
diff --git a/target/linux/generic/pending-6.6/721-net-phy-realtek-clear-1000Base-T-lpa-bits-if-link-is.patch b/target/linux/generic/pending-6.6/721-net-phy-realtek-clear-1000Base-T-lpa-bits-if-link-is.patch
new file mode 100644
index 0000000000..5bc77ed3b8
--- /dev/null
+++ b/target/linux/generic/pending-6.6/721-net-phy-realtek-clear-1000Base-T-lpa-bits-if-link-is.patch
@@ -0,0 +1,78 @@
+From patchwork Wed Jan 15 14:43:35 2025
+Content-Type: text/plain; charset="utf-8"
+MIME-Version: 1.0
+Content-Transfer-Encoding: 7bit
+X-Patchwork-Submitter: Daniel Golle <daniel at makrotopia.org>
+X-Patchwork-Id: 13940510
+X-Patchwork-Delegate: kuba at kernel.org
+Date: Wed, 15 Jan 2025 14:43:35 +0000
+From: Daniel Golle <daniel at makrotopia.org>
+To: Andrew Lunn <andrew at lunn.ch>, Heiner Kallweit <hkallweit1 at gmail.com>,
+ Russell King <linux at armlinux.org.uk>,
+ "David S. Miller" <davem at davemloft.net>,
+ Eric Dumazet <edumazet at google.com>,
+ Jakub Kicinski <kuba at kernel.org>, Paolo Abeni <pabeni at redhat.com>,
+ Daniel Golle <daniel at makrotopia.org>,
+ "Russell King (Oracle)" <linux at armlinux.org.uk>,
+ netdev at vger.kernel.org, linux-kernel at vger.kernel.org
+Subject: [PATCH net 1/3] net: phy: realtek: clear 1000Base-T lpa if link is
+ down
+Message-ID:
+ <67e38eee4c46b921938fb33f5bc86c8979b9aa33.1736951652.git.daniel at makrotopia.org>
+References: <cover.1736951652.git.daniel at makrotopia.org>
+Precedence: bulk
+X-Mailing-List: netdev at vger.kernel.org
+List-Id: <netdev.vger.kernel.org>
+List-Subscribe: <mailto:netdev+subscribe at vger.kernel.org>
+List-Unsubscribe: <mailto:netdev+unsubscribe at vger.kernel.org>
+MIME-Version: 1.0
+Content-Disposition: inline
+In-Reply-To: <cover.1736951652.git.daniel at makrotopia.org>
+X-Patchwork-Delegate: kuba at kernel.org
+
+Only read 1000Base-T link partner advertisement if autonegotiation has
+completed and otherwise 1000Base-T link partner advertisement bits.
+
+This fixes bogus 1000Base-T link partner advertisement after link goes
+down (eg. by disconnecting the wire).
+Fixes: 5cb409b3960e ("net: phy: realtek: clear 1000Base-T link partner advertisement")
+Signed-off-by: Daniel Golle <daniel at makrotopia.org>
+Reviewed-by: Michal Swiatkowski <michal.swiatkowski at linux.intel.com>
+---
+ drivers/net/phy/realtek/realtek_main.c | 19 ++++++++-----------
+ 1 file changed, 8 insertions(+), 11 deletions(-)
+
+--- a/drivers/net/phy/realtek/realtek_main.c
++++ b/drivers/net/phy/realtek/realtek_main.c
+@@ -1068,23 +1068,20 @@ static int rtl822x_c45_read_status(struc
+ {
+ int ret, val;
+
+- ret = genphy_c45_read_status(phydev);
+- if (ret < 0)
+- return ret;
+-
+- if (phydev->autoneg == AUTONEG_DISABLE ||
+- !genphy_c45_aneg_done(phydev))
+- mii_stat1000_mod_linkmode_lpa_t(phydev->lp_advertising, 0);
+-
+ /* Vendor register as C45 has no standardized support for 1000BaseT */
+- if (phydev->autoneg == AUTONEG_ENABLE) {
++ if (phydev->autoneg == AUTONEG_ENABLE && genphy_c45_aneg_done(phydev)) {
+ val = phy_read_mmd(phydev, MDIO_MMD_VEND2,
+ RTL822X_VND2_GANLPAR);
+ if (val < 0)
+ return val;
+-
+- mii_stat1000_mod_linkmode_lpa_t(phydev->lp_advertising, val);
++ } else {
++ val = 0;
+ }
++ mii_stat1000_mod_linkmode_lpa_t(phydev->lp_advertising, val);
++
++ ret = genphy_c45_read_status(phydev);
++ if (ret < 0)
++ return ret;
+
+ if (!phydev->link)
+ return 0;
diff --git a/target/linux/generic/pending-6.6/722-net-phy-realtek-clear-master_slave_state-if-link-is-.patch b/target/linux/generic/pending-6.6/722-net-phy-realtek-clear-master_slave_state-if-link-is-.patch
new file mode 100644
index 0000000000..b7e5f8b58f
--- /dev/null
+++ b/target/linux/generic/pending-6.6/722-net-phy-realtek-clear-master_slave_state-if-link-is-.patch
@@ -0,0 +1,61 @@
+From patchwork Wed Jan 15 14:43:43 2025
+Content-Type: text/plain; charset="utf-8"
+MIME-Version: 1.0
+Content-Transfer-Encoding: 7bit
+X-Patchwork-Submitter: Daniel Golle <daniel at makrotopia.org>
+X-Patchwork-Id: 13940511
+X-Patchwork-Delegate: kuba at kernel.org
+Date: Wed, 15 Jan 2025 14:43:43 +0000
+From: Daniel Golle <daniel at makrotopia.org>
+To: Andrew Lunn <andrew at lunn.ch>, Heiner Kallweit <hkallweit1 at gmail.com>,
+ Russell King <linux at armlinux.org.uk>,
+ "David S. Miller" <davem at davemloft.net>,
+ Eric Dumazet <edumazet at google.com>,
+ Jakub Kicinski <kuba at kernel.org>, Paolo Abeni <pabeni at redhat.com>,
+ Daniel Golle <daniel at makrotopia.org>,
+ "Russell King (Oracle)" <linux at armlinux.org.uk>,
+ netdev at vger.kernel.org, linux-kernel at vger.kernel.org
+Subject: [PATCH net 2/3] net: phy: realtek: clear master_slave_state if link
+ is down
+Message-ID:
+ <2155887c3a665e4132a034df1e9cfdeec0ae48c9.1736951652.git.daniel at makrotopia.org>
+References: <cover.1736951652.git.daniel at makrotopia.org>
+Precedence: bulk
+X-Mailing-List: netdev at vger.kernel.org
+List-Id: <netdev.vger.kernel.org>
+List-Subscribe: <mailto:netdev+subscribe at vger.kernel.org>
+List-Unsubscribe: <mailto:netdev+unsubscribe at vger.kernel.org>
+MIME-Version: 1.0
+Content-Disposition: inline
+In-Reply-To: <cover.1736951652.git.daniel at makrotopia.org>
+X-Patchwork-Delegate: kuba at kernel.org
+
+rtlgen_decode_physr() which sets master_slave_state isn't called in case
+the link is down and other than rtlgen_read_status(),
+rtl822x_c45_read_status() doesn't implicitely clear master_slave_state.
+
+Avoid stale master_slave_state by always setting it to
+MASTER_SLAVE_STATE_UNKNOWN in rtl822x_c45_read_status() in case the link
+is down.
+
+Fixes: 081c9c0265c9 ("net: phy: realtek: read duplex and gbit master from PHYSR register")
+Signed-off-by: Daniel Golle <daniel at makrotopia.org>
+Reviewed-by: Michal Swiatkowski <michal.swiatkowski at linux.intel.com>
+---
+ drivers/net/phy/realtek/realtek_main.c | 4 +++-
+ 1 file changed, 3 insertions(+), 1 deletion(-)
+
+--- a/drivers/net/phy/realtek/realtek_main.c
++++ b/drivers/net/phy/realtek/realtek_main.c
+@@ -1083,8 +1083,10 @@ static int rtl822x_c45_read_status(struc
+ if (ret < 0)
+ return ret;
+
+- if (!phydev->link)
++ if (!phydev->link) {
++ phydev->master_slave_state = MASTER_SLAVE_STATE_UNKNOWN;
+ return 0;
++ }
+
+ /* Read actual speed from vendor register. */
+ val = phy_read_mmd(phydev, MDIO_MMD_VEND2, RTL_VND2_PHYSR);
diff --git a/target/linux/generic/pending-6.6/723-net-phy-realtek-always-clear-10gbase-t-link-partner-.patch b/target/linux/generic/pending-6.6/723-net-phy-realtek-always-clear-10gbase-t-link-partner-.patch
new file mode 100644
index 0000000000..23947c2200
--- /dev/null
+++ b/target/linux/generic/pending-6.6/723-net-phy-realtek-always-clear-10gbase-t-link-partner-.patch
@@ -0,0 +1,67 @@
+From patchwork Wed Jan 15 14:45:00 2025
+Content-Type: text/plain; charset="utf-8"
+MIME-Version: 1.0
+Content-Transfer-Encoding: 7bit
+X-Patchwork-Submitter: Daniel Golle <daniel at makrotopia.org>
+X-Patchwork-Id: 13940519
+X-Patchwork-Delegate: kuba at kernel.org
+Date: Wed, 15 Jan 2025 14:45:00 +0000
+From: Daniel Golle <daniel at makrotopia.org>
+To: Andrew Lunn <andrew at lunn.ch>, Heiner Kallweit <hkallweit1 at gmail.com>,
+ Russell King <linux at armlinux.org.uk>,
+ "David S. Miller" <davem at davemloft.net>,
+ Eric Dumazet <edumazet at google.com>,
+ Jakub Kicinski <kuba at kernel.org>, Paolo Abeni <pabeni at redhat.com>,
+ Daniel Golle <daniel at makrotopia.org>,
+ "Russell King (Oracle)" <linux at armlinux.org.uk>,
+ netdev at vger.kernel.org, linux-kernel at vger.kernel.org
+Subject: [PATCH net 3/3] net: phy: realtek: always clear NBase-T lpa
+Message-ID:
+ <566d4771d68a1e18d2d48860e25891e468e26551.1736951652.git.daniel at makrotopia.org>
+References: <cover.1736951652.git.daniel at makrotopia.org>
+Precedence: bulk
+X-Mailing-List: netdev at vger.kernel.org
+List-Id: <netdev.vger.kernel.org>
+List-Subscribe: <mailto:netdev+subscribe at vger.kernel.org>
+List-Unsubscribe: <mailto:netdev+unsubscribe at vger.kernel.org>
+MIME-Version: 1.0
+Content-Disposition: inline
+In-Reply-To: <cover.1736951652.git.daniel at makrotopia.org>
+X-Patchwork-Delegate: kuba at kernel.org
+
+Clear NBase-T link partner advertisement before calling
+rtlgen_read_status() to avoid phy_resolve_aneg_linkmode() wrongly
+setting speed and duplex.
+
+This fixes bogus 2.5G/5G/10G link partner advertisement and thus
+speed and duplex being set by phy_resolve_aneg_linkmode() due to stale
+NBase-T lpa.
+
+Fixes: 68d5cd09e891 ("net: phy: realtek: change order of calls in C22 read_status()")
+Signed-off-by: Daniel Golle <daniel at makrotopia.org>
+Reviewed-by: Michal Swiatkowski <michal.swiatkowski at linux.intel.com>
+---
+ drivers/net/phy/realtek/realtek_main.c | 6 +++---
+ 1 file changed, 3 insertions(+), 3 deletions(-)
+
+--- a/drivers/net/phy/realtek/realtek_main.c
++++ b/drivers/net/phy/realtek/realtek_main.c
+@@ -997,15 +997,15 @@ static int rtl822x_read_status(struct ph
+ {
+ int lpadv, ret;
+
++ mii_10gbt_stat_mod_linkmode_lpa_t(phydev->lp_advertising, 0);
++
+ ret = rtlgen_read_status(phydev);
+ if (ret < 0)
+ return ret;
+
+ if (phydev->autoneg == AUTONEG_DISABLE ||
+- !phydev->autoneg_complete) {
+- mii_10gbt_stat_mod_linkmode_lpa_t(phydev->lp_advertising, 0);
++ !phydev->autoneg_complete)
+ return 0;
+- }
+
+ lpadv = phy_read_paged(phydev, 0xa5d, 0x13);
+ if (lpadv < 0)
More information about the lede-commits
mailing list