The branch main has been updated by kbowling: URL: https://cgit.FreeBSD.org/src/commit/?id=17d90d5b9350239a87e66a3612cb9b084b2d75e9
commit 17d90d5b9350239a87e66a3612cb9b084b2d75e9 Author: Kevin Bowling <[email protected]> AuthorDate: 2026-08-11 16:18:06 +0000 Commit: Kevin Bowling <[email protected]> CommitDate: 2026-08-11 20:39:54 +0000 e1000: Reconfigure modern PCH K1 clock synchronization Meteor Lake and newer PCH generations can lose packets while the MAC and PHY clocks synchronize. Move K1 power-down to P1 and extend the PHY K1 exit timeout before PHY access and after reset. Use the longer 1 Gb/s PLL clock-gate timeout added by Linux so K1 can remain enabled without the power penalty of disabling it. Apply the workaround through the newer PTP and NVP generations. This follows DPDK commits ba54bdc79d94 and d88ef2356ecc, with the longer exit time observed in Linux 578294b8b60d. MFC after: 2 weeks --- sys/dev/e1000/e1000_ich8lan.c | 84 ++++++++++++++++++++++++++++++++++++++++--- sys/dev/e1000/e1000_ich8lan.h | 5 +++ 2 files changed, 85 insertions(+), 4 deletions(-) diff --git a/sys/dev/e1000/e1000_ich8lan.c b/sys/dev/e1000/e1000_ich8lan.c index 6090b8995718..29f5b2ff69bd 100644 --- a/sys/dev/e1000/e1000_ich8lan.c +++ b/sys/dev/e1000/e1000_ich8lan.c @@ -315,6 +315,46 @@ static void e1000_toggle_lanphypc_pch_lpt(struct e1000_hw *hw) } } +/** + * e1000_reconfigure_k1_exit_timeout - Reconfigure K1 parameters + * @hw: pointer to the HW structure + * + * Reconfigure the PHY power-down state and K1 exit timeout to avoid a + * MAC/PHY clock synchronization problem on Meteor Lake and newer PCH. + * The caller must hold the PHY semaphore. + **/ +static s32 +e1000_reconfigure_k1_exit_timeout(struct e1000_hw *hw) +{ + u32 fextnvm12; + u16 phy_timeout; + s32 ret_val; + + DEBUGFUNC("e1000_reconfigure_k1_exit_timeout"); + + if (hw->mac.type < e1000_pch_mtp || + hw->mac.type >= e1000_82575) + return E1000_SUCCESS; + + /* Change the K1 power-down state from P0s to P1. */ + fextnvm12 = E1000_READ_REG(hw, E1000_FEXTNVM12); + fextnvm12 &= ~E1000_FEXTNVM12_PHYPD_CTRL_MASK; + fextnvm12 |= E1000_FEXTNVM12_PHYPD_CTRL_P1; + E1000_WRITE_REG(hw, E1000_FEXTNVM12, fextnvm12); + + msec_delay_irq(1); + + ret_val = hw->phy.ops.read_reg_locked(hw, E1000_PHY_TIMEOUTS_REG, + &phy_timeout); + if (ret_val) + return ret_val; + + phy_timeout &= ~E1000_PHY_TIMEOUTS_K1_EXIT_TO_MASK; + phy_timeout |= 0xF00; + return hw->phy.ops.write_reg_locked(hw, E1000_PHY_TIMEOUTS_REG, + phy_timeout); +} + /** * e1000_init_phy_workarounds_pchlan - PHY initialization workarounds * @hw: pointer to the HW structure @@ -357,14 +397,20 @@ static s32 e1000_init_phy_workarounds_pchlan(struct e1000_hw *hw) * LANPHYPC Value bit to force the interconnect to PCIe mode. */ switch (hw->mac.type) { + case e1000_pch_mtp: + case e1000_pch_ptp: + case e1000_pch_nvp: + /* The PHY might be inaccessible here, so do not propagate a + * failure from this preliminary programming attempt. + */ + if (e1000_reconfigure_k1_exit_timeout(hw)) + DEBUGOUT("Failed to reconfigure K1 exit timeout\n"); + /* FALLTHROUGH */ case e1000_pch_lpt: case e1000_pch_spt: case e1000_pch_cnp: case e1000_pch_tgp: case e1000_pch_adp: - case e1000_pch_mtp: - case e1000_pch_ptp: - case e1000_pch_nvp: if (e1000_phy_is_accessible_pchlan(hw)) break; @@ -447,8 +493,21 @@ static s32 e1000_init_phy_workarounds_pchlan(struct e1000_hw *hw) * the PHY is in. */ ret_val = hw->phy.ops.check_reset_block(hw); - if (ret_val) + if (ret_val) { ERROR_REPORT("ME blocked access to PHY after reset\n"); + goto out; + } + + if (hw->mac.type >= e1000_pch_mtp && + hw->mac.type < e1000_82575) { + ret_val = hw->phy.ops.acquire(hw); + if (ret_val) { + DEBUGOUT("Failed to acquire PHY for K1 setup\n"); + goto out; + } + ret_val = e1000_reconfigure_k1_exit_timeout(hw); + hw->phy.ops.release(hw); + } } out: @@ -1662,6 +1721,9 @@ static s32 e1000_check_for_copper_link_ich8lan(struct e1000_hw *hw) phy_reg &= ~I217_PLL_CLOCK_GATE_MASK; if (speed == SPEED_100 || speed == SPEED_10) phy_reg |= 0x3E8; + else if (hw->mac.type >= e1000_pch_mtp && + hw->mac.type < e1000_82575) + phy_reg |= 0x1D5; else phy_reg |= 0xFA; hw->phy.ops.write_reg_locked(hw, @@ -5168,6 +5230,20 @@ static s32 e1000_init_hw_ich8lan(struct e1000_hw *hw) e1000_initialize_hw_bits_ich8lan(hw); + if (hw->mac.type >= e1000_pch_mtp && + hw->mac.type < e1000_82575) { + ret_val = hw->phy.ops.acquire(hw); + if (ret_val) + return ret_val; + + ret_val = e1000_reconfigure_k1_exit_timeout(hw); + hw->phy.ops.release(hw); + if (ret_val) { + DEBUGOUT("Failed to reconfigure K1 exit timeout\n"); + return ret_val; + } + } + /* Initialize identification LED */ ret_val = mac->ops.id_led_init(hw); /* An error is not fatal and we should not stop init due to this */ diff --git a/sys/dev/e1000/e1000_ich8lan.h b/sys/dev/e1000/e1000_ich8lan.h index c6af34cade54..f3e87f24c9f6 100644 --- a/sys/dev/e1000/e1000_ich8lan.h +++ b/sys/dev/e1000/e1000_ich8lan.h @@ -118,6 +118,8 @@ #define E1000_FEXTNVM11_DISABLE_PB_READ 0x00000200 #define E1000_FEXTNVM11_DISABLE_MULR_FIX 0x00002000 #define E1000_FEXTNVM12_DONT_WAK_DPG_CLKREQ 0x00001000 +#define E1000_FEXTNVM12_PHYPD_CTRL_MASK 0x00C00000 +#define E1000_FEXTNVM12_PHYPD_CTRL_P1 0x00800000 /* bit24: RXDCTL thresholds granularity: 0 - cache lines, 1 - descriptors */ #define E1000_RXDCTL_THRESH_UNIT_DESC 0x01000000 @@ -244,6 +246,9 @@ #define HV_PM_CTRL_K1_CLK_REQ 0x200 #define HV_PM_CTRL_K1_ENABLE 0x4000 +#define E1000_PHY_TIMEOUTS_REG PHY_REG(770, 21) +#define E1000_PHY_TIMEOUTS_K1_EXIT_TO_MASK 0x0FC0 + #define I217_PLL_CLOCK_GATE_REG PHY_REG(772, 28) #define I217_PLL_CLOCK_GATE_MASK 0x07FF
