Received: from smtp.kernel.org (aws-us-west-2-korg-mail-alma10-1.taild15c8.ts.net [100.103.45.18]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id 5ABA02BDC0E; Sat, 12 Sep 2026 11:24:43 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=100.103.45.18 ARC-Seal:i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1789212285; cv=none; b=FG6g5W/Sox1EJCXXbR2+VwVC9cXN7241g5kgZgLAfN9tjOFVD1wxzzXNg+4pxpH63kUokBFuDBrsF3UBN+jySj03MLYbb+kS7yBqk/+KqLl62Pr5seoJlYJC/da+0eJ71JIU1diFt/f+b+g2BR/2FWHKCYxG9RYkN0NA1lbCpog= ARC-Message-Signature:i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1789212285; c=relaxed/simple; bh=YSHbjwEZ0GdWc346ugirOffJn3tyiYKbu5FqDQ3ebeo=; h=From:To:Cc:Subject:Date:Message-ID:In-Reply-To:References: MIME-Version; b=guryBMapDI1VXjHb0JSy2cryGSZaIdEsqN6MWIi9fm2opRx8P6C5VACTaADUpiETU8s4r61IlguCAQReVkikG8Uh7jKpgvy/2ymAxJXo7oihuClQEVId+wfQyeVy9HlQwZ8XWKgG9fre3IqPvlNQQkhbM3/pDXeBLY/yGVhJD1Q= ARC-Authentication-Results:i=1; smtp.subspace.kernel.org; dkim=pass (1024-bit key) header.d=linuxfoundation.org header.i=@linuxfoundation.org header.b=QxUtDMkT; arc=none smtp.client-ip=100.103.45.18 Authentication-Results: smtp.subspace.kernel.org; dkim=pass (1024-bit key) header.d=linuxfoundation.org header.i=@linuxfoundation.org header.b="QxUtDMkT" Received: by smtp.kernel.org (Postfix) with ESMTPSA id C51971F000FF; Sat, 12 Sep 2026 11:24:41 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=linuxfoundation.org; s=korg; t=1789212283; bh=94HcpVXMEAt5J5tOf5+X1xAHwxPSTWKB3c+i2q0csFw=; h=From:To:Cc:Subject:Date:In-Reply-To:References; b=QxUtDMkT1RE7SmWlbhGSJ5px4pGfmo7V7hus90iZKl/jmcNmCcc10hN3pdDaqY8iD 0lNEGhBALGX2aG5eYa4MyHFtpRQDD4+Qx6jOjjeFdnHDI1FqKc6xtrZegfGZKfGUL0 ySkdmWj07R7vM0hmQEzpd3nXSmvJTnVsLWV/thXg= From: Greg Kroah-Hartman To: stable@vger.kernel.org Cc: Greg Kroah-Hartman , patches@lists.linux.dev, Wei Fang , Simon Horman , Jakub Kicinski , Sasha Levin Subject: [PATCH 6.18 1405/1518] net: enetc: restore RX ring congestion mode after ring reconfiguration Date: Sat, 12 Sep 2026 08:59:34 +0200 Message-ID: <20260912065655.330516306@linuxfoundation.org> X-Mailer: git-send-email 2.55.0 In-Reply-To: <20260912065623.398859879@linuxfoundation.org> References: <20260912065623.398859879@linuxfoundation.org> User-Agent: quilt/0.69 X-stable: review X-Patchwork-Hint: ignore Precedence: bulk X-Mailing-List: patches@lists.linux.dev List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Transfer-Encoding: 8bit 6.18-stable review patch. If anyone has any objections, please let me know. ------------------ From: Wei Fang [ Upstream commit 9e94d8e14ae45e3fda8c0919f745db3ed3715433 ] The RX ring congestion mode (CM) is only configured in the phylink .mac_link_up() callback enetc_pl_mac_link_up(), which sets the ENETC_RBMR_CM bit when tx_pause is enabled. This callback runs only when the link status changes. However, enetc_reconfigure() tears down and re-creates the RX BD rings at runtime without any link status change, for example when attaching or detaching an XDP program, or when enabling/disabling PTP RX hardware timestamping. The rings are rebuilt from a cleared RBMR, so the CM bit is lost. Since the link status does not change, enetc_pl_mac_link_up() is not called again and the CM bit is never restored. As a result, the ENETC MAC can no longer generate PAUSE frames on ingress congestion, and flow control stops working after such a reconfiguration. Track the desired CM state in a software flag ENETC_RXBDR_CM. Set or clear this flag in enetc_pl_mac_link_up() according to tx_pause. When the RX BD rings are (re)enabled, enetc_enable_rxbdr() consults this flag and restores the ENETC_RBMR_CM bit accordingly, so flow control survives ring reconfiguration even when the link status does not change. RBMR is now written as a whole word from enetc_enable_rxbdr() rather than by read-modify-write from several call sites. Serialize the remaining RBMR read-modify-write paths, the congestion mode update and the RX VLAN offload update, with the new si->gen_lock so they cannot race each other. Fixes: 5093406c784f ("net: enetc: implement ring reconfiguration procedure for PTP RX timestamping") Signed-off-by: Wei Fang Reviewed-by: Simon Horman Link: https://patch.msgid.link/20260821064140.1315611-2-wei.fang@oss.nxp.com Signed-off-by: Jakub Kicinski Signed-off-by: Sasha Levin --- drivers/net/ethernet/freescale/enetc/enetc.c | 72 +++++++++++++++---- drivers/net/ethernet/freescale/enetc/enetc.h | 9 +++ .../net/ethernet/freescale/enetc/enetc_pf.c | 14 +--- 3 files changed, 68 insertions(+), 27 deletions(-) diff --git a/drivers/net/ethernet/freescale/enetc/enetc.c b/drivers/net/ethernet/freescale/enetc/enetc.c index 6603ed6962967..dec047c8128d5 100644 --- a/drivers/net/ethernet/freescale/enetc/enetc.c +++ b/drivers/net/ethernet/freescale/enetc/enetc.c @@ -72,6 +72,33 @@ void enetc_reset_mac_addr_filter(struct enetc_mac_filter *filter) } EXPORT_SYMBOL_GPL(enetc_reset_mac_addr_filter); +void enetc_set_congestion_mode(struct enetc_ndev_priv *priv, bool enable) +{ + struct enetc_si *si = priv->si; + struct enetc_hw *hw = &si->hw; + + spin_lock(&si->gen_lock); + + if (enable) + set_bit(ENETC_RXBDR_CM, &priv->flags); + else + clear_bit(ENETC_RXBDR_CM, &priv->flags); + + for (int i = 0; i < priv->num_rx_rings; i++) { + u32 old_rbmr = enetc_rxbdr_rd(hw, i, ENETC_RBMR); + u32 rbmr; + + rbmr = u32_replace_bits(old_rbmr, enable, ENETC_RBMR_CM); + if (rbmr == old_rbmr) + continue; + + enetc_rxbdr_wr(hw, i, ENETC_RBMR, rbmr); + } + + spin_unlock(&si->gen_lock); +} +EXPORT_SYMBOL_GPL(enetc_set_congestion_mode); + static int enetc_num_stack_tx_queues(struct enetc_ndev_priv *priv) { int num_tx_rings = priv->num_tx_rings; @@ -2624,7 +2651,6 @@ static void enetc_setup_rxbdr(struct enetc_hw *hw, struct enetc_bdr *rx_ring, bool extended) { int idx = rx_ring->index; - u32 rbmr = 0; enetc_rxbdr_wr(hw, idx, ENETC_RBBAR0, lower_32_bits(rx_ring->bd_dma_base)); @@ -2652,12 +2678,6 @@ static void enetc_setup_rxbdr(struct enetc_hw *hw, struct enetc_bdr *rx_ring, enetc_rxbdr_wr(hw, idx, ENETC_RBICR0, ENETC_RBICR0_ICEN | 0x1); rx_ring->ext_en = extended; - if (rx_ring->ext_en) - rbmr |= ENETC_RBMR_BDS; - - if (rx_ring->ndev->features & NETIF_F_HW_VLAN_CTAG_RX) - rbmr |= ENETC_RBMR_VTE; - rx_ring->rcir = hw->reg + ENETC_BDR(RX, idx, ENETC_RBCIR); rx_ring->idr = hw->reg + ENETC_SIRXIDR; @@ -2668,8 +2688,6 @@ static void enetc_setup_rxbdr(struct enetc_hw *hw, struct enetc_bdr *rx_ring, enetc_lock_mdio(); enetc_refill_rx_ring(rx_ring, enetc_bd_unused(rx_ring)); enetc_unlock_mdio(); - - enetc_rxbdr_wr(hw, idx, ENETC_RBMR, rbmr); } static void enetc_setup_bdrs(struct enetc_ndev_priv *priv, bool extended) @@ -2696,21 +2714,34 @@ static void enetc_enable_txbdr(struct enetc_hw *hw, struct enetc_bdr *tx_ring) static void enetc_enable_rxbdr(struct enetc_hw *hw, struct enetc_bdr *rx_ring) { + struct enetc_ndev_priv *priv = netdev_priv(rx_ring->ndev); int idx = rx_ring->index; - u32 rbmr; + u32 rbmr = ENETC_RBMR_EN; + + if (rx_ring->ext_en) + rbmr |= ENETC_RBMR_BDS; + + if (rx_ring->ndev->features & NETIF_F_HW_VLAN_CTAG_RX) + rbmr |= ENETC_RBMR_VTE; + + if (test_bit(ENETC_RXBDR_CM, &priv->flags)) + rbmr |= ENETC_RBMR_CM; - rbmr = enetc_rxbdr_rd(hw, idx, ENETC_RBMR); - rbmr |= ENETC_RBMR_EN; enetc_rxbdr_wr(hw, idx, ENETC_RBMR, rbmr); } static void enetc_enable_rx_bdrs(struct enetc_ndev_priv *priv) { - struct enetc_hw *hw = &priv->si->hw; + struct enetc_si *si = priv->si; + struct enetc_hw *hw = &si->hw; int i; + spin_lock(&si->gen_lock); + for (i = 0; i < priv->num_rx_rings; i++) enetc_enable_rxbdr(hw, priv->rx_ring[i]); + + spin_unlock(&si->gen_lock); } static void enetc_enable_tx_bdrs(struct enetc_ndev_priv *priv) @@ -2740,11 +2771,16 @@ static void enetc_disable_txbdr(struct enetc_hw *hw, struct enetc_bdr *rx_ring) static void enetc_disable_rx_bdrs(struct enetc_ndev_priv *priv) { - struct enetc_hw *hw = &priv->si->hw; + struct enetc_si *si = priv->si; + struct enetc_hw *hw = &si->hw; int i; + spin_lock(&si->gen_lock); + for (i = 0; i < priv->num_rx_rings; i++) enetc_disable_rxbdr(hw, priv->rx_ring[i]); + + spin_unlock(&si->gen_lock); } static void enetc_disable_tx_bdrs(struct enetc_ndev_priv *priv) @@ -3336,11 +3372,16 @@ EXPORT_SYMBOL_GPL(enetc_get_stats); static void enetc_enable_rxvlan(struct net_device *ndev, bool en) { struct enetc_ndev_priv *priv = netdev_priv(ndev); - struct enetc_hw *hw = &priv->si->hw; + struct enetc_si *si = priv->si; + struct enetc_hw *hw = &si->hw; int i; + spin_lock(&si->gen_lock); + for (i = 0; i < priv->num_rx_rings; i++) enetc_bdr_enable_rxvlan(hw, i, en); + + spin_unlock(&si->gen_lock); } static void enetc_enable_txvlan(struct net_device *ndev, bool en) @@ -3670,6 +3711,7 @@ int enetc_pci_probe(struct pci_dev *pdev, const char *name, int sizeof_priv) si = PTR_ALIGN(p, ENETC_SI_ALIGN); si->pad = (char *)si - (char *)p; + spin_lock_init(&si->gen_lock); pci_set_drvdata(pdev, si); si->pdev = pdev; diff --git a/drivers/net/ethernet/freescale/enetc/enetc.h b/drivers/net/ethernet/freescale/enetc/enetc.h index 60c7205ea9ff5..0e2effa947b2d 100644 --- a/drivers/net/ethernet/freescale/enetc/enetc.h +++ b/drivers/net/ethernet/freescale/enetc/enetc.h @@ -309,6 +309,13 @@ struct enetc_si { struct net_device *ndev; /* back ref. */ + /* General-purpose lock serializing updates that must not race, + * e.g. read-modify-write of shared hardware registers and of + * selected priv->flags bits between the phylink link callbacks + * and the ring (re)configuration path. + */ + spinlock_t gen_lock; + union { struct enetc_cbdr cbd_ring; /* Only ENETC 1.0 */ struct ntmp_user ntmp_user; /* ENETC 4.1 and later */ @@ -413,6 +420,7 @@ enum enetc_active_offloads { enum enetc_flags_bit { ENETC_TX_ONESTEP_TSTAMP_IN_PROGRESS = 0, ENETC_TX_DOWN, + ENETC_RXBDR_CM, }; /* interrupt coalescing modes */ @@ -509,6 +517,7 @@ int enetc_get_driver_data(struct enetc_si *si); void enetc_add_mac_addr_ht_filter(struct enetc_mac_filter *filter, const unsigned char *addr); void enetc_reset_mac_addr_filter(struct enetc_mac_filter *filter); +void enetc_set_congestion_mode(struct enetc_ndev_priv *priv, bool enable); int enetc_open(struct net_device *ndev); int enetc_close(struct net_device *ndev); diff --git a/drivers/net/ethernet/freescale/enetc/enetc_pf.c b/drivers/net/ethernet/freescale/enetc/enetc_pf.c index de0fb272c8474..ba630f0982104 100644 --- a/drivers/net/ethernet/freescale/enetc/enetc_pf.c +++ b/drivers/net/ethernet/freescale/enetc/enetc_pf.c @@ -691,8 +691,7 @@ static void enetc_pl_mac_link_up(struct phylink_config *config, struct enetc_hw *hw = &pf->si->hw; struct enetc_si *si = pf->si; struct enetc_ndev_priv *priv; - u32 rbmr, cmd_cfg; - int idx; + u32 cmd_cfg; priv = netdev_priv(pf->si->ndev); @@ -704,16 +703,7 @@ static void enetc_pl_mac_link_up(struct phylink_config *config, enetc_force_rgmii_mac(si, speed, duplex); /* Flow control */ - for (idx = 0; idx < priv->num_rx_rings; idx++) { - rbmr = enetc_rxbdr_rd(hw, idx, ENETC_RBMR); - - if (tx_pause) - rbmr |= ENETC_RBMR_CM; - else - rbmr &= ~ENETC_RBMR_CM; - - enetc_rxbdr_wr(hw, idx, ENETC_RBMR, rbmr); - } + enetc_set_congestion_mode(priv, tx_pause); if (tx_pause) { /* When the port first enters congestion, send a PAUSE request -- 2.53.0