* [PATCH net-next v3 06/10] net: bcmgenet: add main driver file
From: Florian Fainelli @ 2014-02-13 22:53 UTC (permalink / raw)
To: netdev-u79uwXL29TY76Z2rM5mHXA
Cc: davem-fT/PcQaiUtIeIZ0/mPfg9Q, devicetree-u79uwXL29TY76Z2rM5mHXA,
cernekee-Re5JQEeQqe8AvxtiuMwx3w, mark.rutland-5wv7dgnIgG8,
romieu-W8zweXLXuWQS+FvcfC7Uqw, Florian Fainelli
In-Reply-To: <1392332029-24669-1-git-send-email-f.fainelli-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
This patch adds the BCMGENET main driver file which supports the
following:
- GENET hardware from V1 to V4
- support for reading the UniMAC MIB counters statistics
- support for the 5 transmit queues
- support for RX/TX checksum offload and SG
Signed-off-by: Florian Fainelli <f.fainelli-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
---
Changes since v2:
- fixed stylistic issues spotted by Francois Romieu (xmas tree etc...)
- updated bcmgenet_dma_teardown() to also disable the TDMA engine
- removed WIP Wake-on-LAN code
- use of_device_id auxiliary data to get the GENET version
- fixed comment about disabling the GENET main clock
- changed unsigned int types to bool for desc_64b_en, rxhck_en etc...
- use phy_is_internal() where appropriate
- remove bh_lock use
Changes since v1:
- use module_platform_driver boilerplate macro
- renamed bcmgenet_plat_drv to bcmgenet_driver
- renamed bcmgenet_drv_{probe,remove} to bcmgenet_{probe,remove}
- removed priv->phy_type usage and use priv->phy_interface which
contains the exact same value
- removed debug module parameters, unused
- added MODULDE_{AUTHOR,ALIAS,DESCRIPTION} macros
- remove hardcoded queue index check in bcmgenet_xmit
drivers/net/ethernet/broadcom/genet/bcmgenet.c | 2595 ++++++++++++++++++++++++
1 file changed, 2595 insertions(+)
create mode 100644 drivers/net/ethernet/broadcom/genet/bcmgenet.c
diff --git a/drivers/net/ethernet/broadcom/genet/bcmgenet.c b/drivers/net/ethernet/broadcom/genet/bcmgenet.c
new file mode 100644
index 0000000..0ebc297
--- /dev/null
+++ b/drivers/net/ethernet/broadcom/genet/bcmgenet.c
@@ -0,0 +1,2595 @@
+/*
+ * Broadcom GENET (Gigabit Ethernet) controller driver
+ *
+ * Copyright (c) 2014 Broadcom Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
+ */
+
+#define pr_fmt(fmt) "bcmgenet: " fmt
+
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/sched.h>
+#include <linux/types.h>
+#include <linux/fcntl.h>
+#include <linux/interrupt.h>
+#include <linux/string.h>
+#include <linux/if_ether.h>
+#include <linux/init.h>
+#include <linux/errno.h>
+#include <linux/delay.h>
+#include <linux/platform_device.h>
+#include <linux/dma-mapping.h>
+#include <linux/pm.h>
+#include <linux/clk.h>
+#include <linux/version.h>
+#include <linux/of.h>
+#include <linux/of_address.h>
+#include <linux/of_irq.h>
+#include <linux/of_net.h>
+#include <linux/of_platform.h>
+#include <net/arp.h>
+
+#include <linux/mii.h>
+#include <linux/ethtool.h>
+#include <linux/netdevice.h>
+#include <linux/inetdevice.h>
+#include <linux/etherdevice.h>
+#include <linux/skbuff.h>
+#include <linux/in.h>
+#include <linux/ip.h>
+#include <linux/ipv6.h>
+#include <linux/phy.h>
+
+#include <asm/unaligned.h>
+
+#include "bcmgenet.h"
+
+/* Maximum number of hardware queues, downsized if needed */
+#define GENET_MAX_MQ_CNT 4
+
+/* Default highest priority queue for multi queue support */
+#define GENET_Q0_PRIORITY 0
+
+#define GENET_DEFAULT_BD_CNT \
+ (TOTAL_DESC - priv->hw_params->tx_queues * priv->hw_params->bds_cnt)
+
+#define RX_BUF_LENGTH 2048
+#define SKB_ALIGNMENT 32
+
+/* Tx/Rx DMA register offset, skip 256 descriptors */
+#define WORDS_PER_BD(p) (p->hw_params->words_per_bd)
+#define DMA_DESC_SIZE (WORDS_PER_BD(priv) * sizeof(u32))
+
+#define GENET_TDMA_REG_OFF (priv->hw_params->tdma_offset + \
+ TOTAL_DESC * DMA_DESC_SIZE)
+
+#define GENET_RDMA_REG_OFF (priv->hw_params->rdma_offset + \
+ TOTAL_DESC * DMA_DESC_SIZE)
+
+static inline void dmadesc_set_length_status(struct bcmgenet_priv *priv,
+ void __iomem *d, u32 value)
+{
+ __raw_writel(value, d + DMA_DESC_LENGTH_STATUS);
+}
+
+static inline u32 dmadesc_get_length_status(struct bcmgenet_priv *priv,
+ void __iomem *d)
+{
+ return __raw_readl(d + DMA_DESC_LENGTH_STATUS);
+}
+
+static inline void dmadesc_set_addr(struct bcmgenet_priv *priv,
+ void __iomem *d,
+ dma_addr_t addr)
+{
+ __raw_writel(lower_32_bits(addr), d + DMA_DESC_ADDRESS_LO);
+
+ /* Register writes to GISB bus can take couple hundred nanoseconds
+ * and are done for each packet, save these expensive writes unless
+ * the platform is explicitely configured for 64-bits/LPAE.
+ */
+#ifdef CONFIG_PHYS_ADDR_T_64BIT
+ if (priv->hw_params->flags & GENET_HAS_40BITS)
+ __raw_writel(upper_32_bits(addr), d + DMA_DESC_ADDRESS_HI);
+#endif
+}
+
+/* Combined address + length/status setter */
+static inline void dmadesc_set(struct bcmgenet_priv *priv,
+ void __iomem *d, dma_addr_t addr, u32 val)
+{
+ dmadesc_set_length_status(priv, d, val);
+ dmadesc_set_addr(priv, d, addr);
+}
+
+static inline dma_addr_t dmadesc_get_addr(struct bcmgenet_priv *priv,
+ void __iomem *d)
+{
+ dma_addr_t addr;
+
+ addr = __raw_readl(d + DMA_DESC_ADDRESS_LO);
+
+ /* Register writes to GISB bus can take couple hundred nanoseconds
+ * and are done for each packet, save these expensive writes unless
+ * the platform is explicitely configured for 64-bits/LPAE.
+ */
+#ifdef CONFIG_PHYS_ADDR_T_64BIT
+ if (priv->hw_params->flags & GENET_HAS_40BITS)
+ addr |= (u64)__raw_readl(d + DMA_DESC_ADDRESS_HI) << 32;
+#endif
+ return addr;
+}
+
+#define GENET_VER_FMT "%1d.%1d EPHY: 0x%04x"
+
+#define GENET_MSG_DEFAULT (NETIF_MSG_DRV | NETIF_MSG_PROBE | \
+ NETIF_MSG_LINK)
+
+static inline u32 bcmgenet_rbuf_ctrl_get(struct bcmgenet_priv *priv)
+{
+ if (GENET_IS_V1(priv))
+ return bcmgenet_rbuf_readl(priv, RBUF_FLUSH_CTRL_V1);
+ else
+ return bcmgenet_sys_readl(priv, SYS_RBUF_FLUSH_CTRL);
+}
+
+static inline void bcmgenet_rbuf_ctrl_set(struct bcmgenet_priv *priv, u32 val)
+{
+ if (GENET_IS_V1(priv))
+ bcmgenet_rbuf_writel(priv, val, RBUF_FLUSH_CTRL_V1);
+ else
+ bcmgenet_sys_writel(priv, val, SYS_RBUF_FLUSH_CTRL);
+}
+
+/* These macros are defined to deal with register map change
+ * between GENET1.1 and GENET2. Only those currently being used
+ * by driver are defined.
+ */
+static inline u32 bcmgenet_tbuf_ctrl_get(struct bcmgenet_priv *priv)
+{
+ if (GENET_IS_V1(priv))
+ return bcmgenet_rbuf_readl(priv, TBUF_CTRL_V1);
+ else
+ return __raw_readl(priv->base +
+ priv->hw_params->tbuf_offset + TBUF_CTRL);
+}
+
+static inline void bcmgenet_tbuf_ctrl_set(struct bcmgenet_priv *priv, u32 val)
+{
+ if (GENET_IS_V1(priv))
+ bcmgenet_rbuf_writel(priv, val, TBUF_CTRL_V1);
+ else
+ __raw_writel(val, priv->base +
+ priv->hw_params->tbuf_offset + TBUF_CTRL);
+}
+
+static inline u32 bcmgenet_bp_mc_get(struct bcmgenet_priv *priv)
+{
+ if (GENET_IS_V1(priv))
+ return bcmgenet_rbuf_readl(priv, TBUF_BP_MC_V1);
+ else
+ return __raw_readl(priv->base +
+ priv->hw_params->tbuf_offset + TBUF_BP_MC);
+}
+
+static inline void bcmgenet_bp_mc_set(struct bcmgenet_priv *priv, u32 val)
+{
+ if (GENET_IS_V1(priv))
+ bcmgenet_rbuf_writel(priv, val, TBUF_BP_MC_V1);
+ else
+ __raw_writel(val, priv->base +
+ priv->hw_params->tbuf_offset + TBUF_BP_MC);
+}
+
+/* RX/TX DMA register accessors */
+enum dma_reg {
+ DMA_RING_CFG = 0,
+ DMA_CTRL,
+ DMA_STATUS,
+ DMA_SCB_BURST_SIZE,
+ DMA_ARB_CTRL,
+ DMA_PRIORITY,
+ DMA_RING_PRIORITY,
+};
+
+static const u8 bcmgenet_dma_regs_v3plus[] = {
+ [DMA_RING_CFG] = 0x00,
+ [DMA_CTRL] = 0x04,
+ [DMA_STATUS] = 0x08,
+ [DMA_SCB_BURST_SIZE] = 0x0C,
+ [DMA_ARB_CTRL] = 0x2C,
+ [DMA_PRIORITY] = 0x30,
+ [DMA_RING_PRIORITY] = 0x38,
+};
+
+static const u8 bcmgenet_dma_regs_v2[] = {
+ [DMA_RING_CFG] = 0x00,
+ [DMA_CTRL] = 0x04,
+ [DMA_STATUS] = 0x08,
+ [DMA_SCB_BURST_SIZE] = 0x0C,
+ [DMA_ARB_CTRL] = 0x30,
+ [DMA_PRIORITY] = 0x34,
+ [DMA_RING_PRIORITY] = 0x3C,
+};
+
+static const u8 bcmgenet_dma_regs_v1[] = {
+ [DMA_CTRL] = 0x00,
+ [DMA_STATUS] = 0x04,
+ [DMA_SCB_BURST_SIZE] = 0x0C,
+ [DMA_ARB_CTRL] = 0x30,
+ [DMA_PRIORITY] = 0x34,
+ [DMA_RING_PRIORITY] = 0x3C,
+};
+
+/* Set at runtime once bcmgenet version is known */
+static const u8 *bcmgenet_dma_regs;
+
+static inline struct bcmgenet_priv *dev_to_priv(struct device *dev)
+{
+ return netdev_priv(dev_get_drvdata(dev));
+}
+
+static inline u32 bcmgenet_tdma_readl(struct bcmgenet_priv *priv,
+ enum dma_reg r)
+{
+ return __raw_readl(priv->base + GENET_TDMA_REG_OFF +
+ DMA_RINGS_SIZE + bcmgenet_dma_regs[r]);
+}
+
+static inline void bcmgenet_tdma_writel(struct bcmgenet_priv *priv,
+ u32 val, enum dma_reg r)
+{
+ __raw_writel(val, priv->base + GENET_TDMA_REG_OFF +
+ DMA_RINGS_SIZE + bcmgenet_dma_regs[r]);
+}
+
+static inline u32 bcmgenet_rdma_readl(struct bcmgenet_priv *priv,
+ enum dma_reg r)
+{
+ return __raw_readl(priv->base + GENET_RDMA_REG_OFF +
+ DMA_RINGS_SIZE + bcmgenet_dma_regs[r]);
+}
+
+static inline void bcmgenet_rdma_writel(struct bcmgenet_priv *priv,
+ u32 val, enum dma_reg r)
+{
+ __raw_writel(val, priv->base + GENET_RDMA_REG_OFF +
+ DMA_RINGS_SIZE + bcmgenet_dma_regs[r]);
+}
+
+/* RDMA/TDMA ring registers and accessors
+ * we merge the common fields and just prefix with T/D the registers
+ * having different meaning depending on the direction
+ */
+enum dma_ring_reg {
+ TDMA_READ_PTR = 0,
+ RDMA_WRITE_PTR = TDMA_READ_PTR,
+ TDMA_READ_PTR_HI,
+ RDMA_WRITE_PTR_HI = TDMA_READ_PTR_HI,
+ TDMA_CONS_INDEX,
+ RDMA_PROD_INDEX = TDMA_CONS_INDEX,
+ TDMA_PROD_INDEX,
+ RDMA_CONS_INDEX = TDMA_PROD_INDEX,
+ DMA_RING_BUF_SIZE,
+ DMA_START_ADDR,
+ DMA_START_ADDR_HI,
+ DMA_END_ADDR,
+ DMA_END_ADDR_HI,
+ DMA_MBUF_DONE_THRESH,
+ TDMA_FLOW_PERIOD,
+ RDMA_XON_XOFF_THRESH = TDMA_FLOW_PERIOD,
+ TDMA_WRITE_PTR,
+ RDMA_READ_PTR = TDMA_WRITE_PTR,
+ TDMA_WRITE_PTR_HI,
+ RDMA_READ_PTR_HI = TDMA_WRITE_PTR_HI
+};
+
+/* GENET v4 supports 40-bits pointer addressing
+ * for obvious reasons the LO and HI word parts
+ * are contiguous, but this offsets the other
+ * registers.
+ */
+static const u8 genet_dma_ring_regs_v4[] = {
+ [TDMA_READ_PTR] = 0x00,
+ [TDMA_READ_PTR_HI] = 0x04,
+ [TDMA_CONS_INDEX] = 0x08,
+ [TDMA_PROD_INDEX] = 0x0C,
+ [DMA_RING_BUF_SIZE] = 0x10,
+ [DMA_START_ADDR] = 0x14,
+ [DMA_START_ADDR_HI] = 0x18,
+ [DMA_END_ADDR] = 0x1C,
+ [DMA_END_ADDR_HI] = 0x20,
+ [DMA_MBUF_DONE_THRESH] = 0x24,
+ [TDMA_FLOW_PERIOD] = 0x28,
+ [TDMA_WRITE_PTR] = 0x2C,
+ [TDMA_WRITE_PTR_HI] = 0x30,
+};
+
+static const u8 genet_dma_ring_regs_v123[] = {
+ [TDMA_READ_PTR] = 0x00,
+ [TDMA_CONS_INDEX] = 0x04,
+ [TDMA_PROD_INDEX] = 0x08,
+ [DMA_RING_BUF_SIZE] = 0x0C,
+ [DMA_START_ADDR] = 0x10,
+ [DMA_END_ADDR] = 0x14,
+ [DMA_MBUF_DONE_THRESH] = 0x18,
+ [TDMA_FLOW_PERIOD] = 0x1C,
+ [TDMA_WRITE_PTR] = 0x20,
+};
+
+/* Set at runtime once GENET version is known */
+static const u8 *genet_dma_ring_regs;
+
+static inline u32 bcmgenet_tdma_ring_readl(struct bcmgenet_priv *priv,
+ unsigned int ring,
+ enum dma_ring_reg r)
+{
+ return __raw_readl(priv->base + GENET_TDMA_REG_OFF +
+ (DMA_RING_SIZE * ring) +
+ genet_dma_ring_regs[r]);
+}
+
+static inline void bcmgenet_tdma_ring_writel(struct bcmgenet_priv *priv,
+ unsigned int ring,
+ u32 val,
+ enum dma_ring_reg r)
+{
+ __raw_writel(val, priv->base + GENET_TDMA_REG_OFF +
+ (DMA_RING_SIZE * ring) +
+ genet_dma_ring_regs[r]);
+}
+
+static inline u32 bcmgenet_rdma_ring_readl(struct bcmgenet_priv *priv,
+ unsigned int ring,
+ enum dma_ring_reg r)
+{
+ return __raw_readl(priv->base + GENET_RDMA_REG_OFF +
+ (DMA_RING_SIZE * ring) +
+ genet_dma_ring_regs[r]);
+}
+
+static inline void bcmgenet_rdma_ring_writel(struct bcmgenet_priv *priv,
+ unsigned int ring,
+ u32 val,
+ enum dma_ring_reg r)
+{
+ __raw_writel(val, priv->base + GENET_RDMA_REG_OFF +
+ (DMA_RING_SIZE * ring) +
+ genet_dma_ring_regs[r]);
+}
+
+static int bcmgenet_get_settings(struct net_device *dev,
+ struct ethtool_cmd *cmd)
+{
+ struct bcmgenet_priv *priv = netdev_priv(dev);
+
+ if (!netif_running(dev))
+ return -EINVAL;
+
+ if (!priv->phydev)
+ return -ENODEV;
+
+ return phy_ethtool_gset(priv->phydev, cmd);
+}
+
+static int bcmgenet_set_settings(struct net_device *dev,
+ struct ethtool_cmd *cmd)
+{
+ struct bcmgenet_priv *priv = netdev_priv(dev);
+
+ if (!netif_running(dev))
+ return -EINVAL;
+
+ if (!priv->phydev)
+ return -ENODEV;
+
+ return phy_ethtool_sset(priv->phydev, cmd);
+}
+
+static int bcmgenet_set_rx_csum(struct net_device *dev,
+ netdev_features_t wanted)
+{
+ struct bcmgenet_priv *priv = netdev_priv(dev);
+ u32 rbuf_chk_ctrl;
+ bool rx_csum_en;
+
+ rx_csum_en = !!(wanted & NETIF_F_RXCSUM);
+
+ rbuf_chk_ctrl = bcmgenet_rbuf_readl(priv, RBUF_CHK_CTRL);
+
+ /* enable rx checksumming */
+ if (rx_csum_en)
+ rbuf_chk_ctrl |= RBUF_RXCHK_EN;
+ else
+ rbuf_chk_ctrl &= ~RBUF_RXCHK_EN;
+ priv->desc_rxchk_en = rx_csum_en;
+ bcmgenet_rbuf_writel(priv, rbuf_chk_ctrl, RBUF_CHK_CTRL);
+
+ return 0;
+}
+
+static int bcmgenet_set_tx_csum(struct net_device *dev,
+ netdev_features_t wanted)
+{
+ struct bcmgenet_priv *priv = netdev_priv(dev);
+ bool desc_64b_en;
+ u32 tbuf_ctrl, rbuf_ctrl;
+
+ tbuf_ctrl = bcmgenet_tbuf_ctrl_get(priv);
+ rbuf_ctrl = bcmgenet_rbuf_readl(priv, RBUF_CTRL);
+
+ desc_64b_en = !!(wanted & (NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM));
+
+ /* enable 64 bytes descriptor in both directions (RBUF and TBUF) */
+ if (desc_64b_en) {
+ tbuf_ctrl |= RBUF_64B_EN;
+ rbuf_ctrl |= RBUF_64B_EN;
+ } else {
+ tbuf_ctrl &= ~RBUF_64B_EN;
+ rbuf_ctrl &= ~RBUF_64B_EN;
+ }
+ priv->desc_64b_en = desc_64b_en;
+
+ bcmgenet_tbuf_ctrl_set(priv, tbuf_ctrl);
+ bcmgenet_rbuf_writel(priv, rbuf_ctrl, RBUF_CTRL);
+
+ return 0;
+}
+
+static int bcmgenet_set_features(struct net_device *dev,
+ netdev_features_t features)
+{
+ netdev_features_t changed = features ^ dev->features;
+ netdev_features_t wanted = dev->wanted_features;
+ int ret = 0;
+
+ if (changed & (NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM))
+ ret = bcmgenet_set_tx_csum(dev, wanted);
+ if (changed & (NETIF_F_RXCSUM))
+ ret = bcmgenet_set_rx_csum(dev, wanted);
+
+ return ret;
+}
+
+static u32 bcmgenet_get_msglevel(struct net_device *dev)
+{
+ struct bcmgenet_priv *priv = netdev_priv(dev);
+
+ return priv->msg_enable;
+}
+
+static void bcmgenet_set_msglevel(struct net_device *dev, u32 level)
+{
+ struct bcmgenet_priv *priv = netdev_priv(dev);
+
+ priv->msg_enable = level;
+}
+
+/* standard ethtool support functions. */
+enum bcmgenet_stat_type {
+ BCMGENET_STAT_NETDEV = -1,
+ BCMGENET_STAT_MIB_RX,
+ BCMGENET_STAT_MIB_TX,
+ BCMGENET_STAT_RUNT,
+ BCMGENET_STAT_MISC,
+};
+
+struct bcmgenet_stats {
+ char stat_string[ETH_GSTRING_LEN];
+ int stat_sizeof;
+ int stat_offset;
+ enum bcmgenet_stat_type type;
+ /* reg offset from UMAC base for misc counters */
+ u16 reg_offset;
+};
+
+#define STAT_NETDEV(m) { \
+ .stat_string = __stringify(m), \
+ .stat_sizeof = sizeof(((struct net_device_stats *)0)->m), \
+ .stat_offset = offsetof(struct net_device_stats, m), \
+ .type = BCMGENET_STAT_NETDEV, \
+}
+
+#define STAT_GENET_MIB(str, m, _type) { \
+ .stat_string = str, \
+ .stat_sizeof = sizeof(((struct bcmgenet_priv *)0)->m), \
+ .stat_offset = offsetof(struct bcmgenet_priv, m), \
+ .type = _type, \
+}
+
+#define STAT_GENET_MIB_RX(str, m) STAT_GENET_MIB(str, m, BCMGENET_STAT_MIB_RX)
+#define STAT_GENET_MIB_TX(str, m) STAT_GENET_MIB(str, m, BCMGENET_STAT_MIB_TX)
+#define STAT_GENET_RUNT(str, m) STAT_GENET_MIB(str, m, BCMGENET_STAT_RUNT)
+
+#define STAT_GENET_MISC(str, m, offset) { \
+ .stat_string = str, \
+ .stat_sizeof = sizeof(((struct bcmgenet_priv *)0)->m), \
+ .stat_offset = offsetof(struct bcmgenet_priv, m), \
+ .type = BCMGENET_STAT_MISC, \
+ .reg_offset = offset, \
+}
+
+
+/* There is a 0xC gap between the end of RX and beginning of TX stats and then
+ * between the end of TX stats and the beginning of the RX RUNT
+ */
+#define BCMGENET_STAT_OFFSET 0xc
+
+/* Hardware counters must be kept in sync because the order/offset
+ * is important here (order in structure declaration = order in hardware)
+ */
+static const struct bcmgenet_stats bcmgenet_gstrings_stats[] = {
+ /* general stats */
+ STAT_NETDEV(rx_packets),
+ STAT_NETDEV(tx_packets),
+ STAT_NETDEV(rx_bytes),
+ STAT_NETDEV(tx_bytes),
+ STAT_NETDEV(rx_errors),
+ STAT_NETDEV(tx_errors),
+ STAT_NETDEV(rx_dropped),
+ STAT_NETDEV(tx_dropped),
+ STAT_NETDEV(multicast),
+ /* UniMAC RSV counters */
+ STAT_GENET_MIB_RX("rx_64_octets", mib.rx.pkt_cnt.cnt_64),
+ STAT_GENET_MIB_RX("rx_65_127_oct", mib.rx.pkt_cnt.cnt_127),
+ STAT_GENET_MIB_RX("rx_128_255_oct", mib.rx.pkt_cnt.cnt_255),
+ STAT_GENET_MIB_RX("rx_256_511_oct", mib.rx.pkt_cnt.cnt_511),
+ STAT_GENET_MIB_RX("rx_512_1023_oct", mib.rx.pkt_cnt.cnt_1023),
+ STAT_GENET_MIB_RX("rx_1024_1518_oct", mib.rx.pkt_cnt.cnt_1518),
+ STAT_GENET_MIB_RX("rx_vlan_1519_1522_oct", mib.rx.pkt_cnt.cnt_mgv),
+ STAT_GENET_MIB_RX("rx_1522_2047_oct", mib.rx.pkt_cnt.cnt_2047),
+ STAT_GENET_MIB_RX("rx_2048_4095_oct", mib.rx.pkt_cnt.cnt_4095),
+ STAT_GENET_MIB_RX("rx_4096_9216_oct", mib.rx.pkt_cnt.cnt_9216),
+ STAT_GENET_MIB_RX("rx_pkts", mib.rx.pkt),
+ STAT_GENET_MIB_RX("rx_bytes", mib.rx.bytes),
+ STAT_GENET_MIB_RX("rx_multicast", mib.rx.mca),
+ STAT_GENET_MIB_RX("rx_broadcast", mib.rx.bca),
+ STAT_GENET_MIB_RX("rx_fcs", mib.rx.fcs),
+ STAT_GENET_MIB_RX("rx_control", mib.rx.cf),
+ STAT_GENET_MIB_RX("rx_pause", mib.rx.pf),
+ STAT_GENET_MIB_RX("rx_unknown", mib.rx.uo),
+ STAT_GENET_MIB_RX("rx_align", mib.rx.aln),
+ STAT_GENET_MIB_RX("rx_outrange", mib.rx.flr),
+ STAT_GENET_MIB_RX("rx_code", mib.rx.cde),
+ STAT_GENET_MIB_RX("rx_carrier", mib.rx.fcr),
+ STAT_GENET_MIB_RX("rx_oversize", mib.rx.ovr),
+ STAT_GENET_MIB_RX("rx_jabber", mib.rx.jbr),
+ STAT_GENET_MIB_RX("rx_mtu_err", mib.rx.mtue),
+ STAT_GENET_MIB_RX("rx_good_pkts", mib.rx.pok),
+ STAT_GENET_MIB_RX("rx_unicast", mib.rx.uc),
+ STAT_GENET_MIB_RX("rx_ppp", mib.rx.ppp),
+ STAT_GENET_MIB_RX("rx_crc", mib.rx.rcrc),
+ /* UniMAC TSV counters */
+ STAT_GENET_MIB_TX("tx_64_octets", mib.tx.pkt_cnt.cnt_64),
+ STAT_GENET_MIB_TX("tx_65_127_oct", mib.tx.pkt_cnt.cnt_127),
+ STAT_GENET_MIB_TX("tx_128_255_oct", mib.tx.pkt_cnt.cnt_255),
+ STAT_GENET_MIB_TX("tx_256_511_oct", mib.tx.pkt_cnt.cnt_511),
+ STAT_GENET_MIB_TX("tx_512_1023_oct", mib.tx.pkt_cnt.cnt_1023),
+ STAT_GENET_MIB_TX("tx_1024_1518_oct", mib.tx.pkt_cnt.cnt_1518),
+ STAT_GENET_MIB_TX("tx_vlan_1519_1522_oct", mib.tx.pkt_cnt.cnt_mgv),
+ STAT_GENET_MIB_TX("tx_1522_2047_oct", mib.tx.pkt_cnt.cnt_2047),
+ STAT_GENET_MIB_TX("tx_2048_4095_oct", mib.tx.pkt_cnt.cnt_4095),
+ STAT_GENET_MIB_TX("tx_4096_9216_oct", mib.tx.pkt_cnt.cnt_9216),
+ STAT_GENET_MIB_TX("tx_pkts", mib.tx.pkts),
+ STAT_GENET_MIB_TX("tx_multicast", mib.tx.mca),
+ STAT_GENET_MIB_TX("tx_broadcast", mib.tx.bca),
+ STAT_GENET_MIB_TX("tx_pause", mib.tx.pf),
+ STAT_GENET_MIB_TX("tx_control", mib.tx.cf),
+ STAT_GENET_MIB_TX("tx_fcs_err", mib.tx.fcs),
+ STAT_GENET_MIB_TX("tx_oversize", mib.tx.ovr),
+ STAT_GENET_MIB_TX("tx_defer", mib.tx.drf),
+ STAT_GENET_MIB_TX("tx_excess_defer", mib.tx.edf),
+ STAT_GENET_MIB_TX("tx_single_col", mib.tx.scl),
+ STAT_GENET_MIB_TX("tx_multi_col", mib.tx.mcl),
+ STAT_GENET_MIB_TX("tx_late_col", mib.tx.lcl),
+ STAT_GENET_MIB_TX("tx_excess_col", mib.tx.ecl),
+ STAT_GENET_MIB_TX("tx_frags", mib.tx.frg),
+ STAT_GENET_MIB_TX("tx_total_col", mib.tx.ncl),
+ STAT_GENET_MIB_TX("tx_jabber", mib.tx.jbr),
+ STAT_GENET_MIB_TX("tx_bytes", mib.tx.bytes),
+ STAT_GENET_MIB_TX("tx_good_pkts", mib.tx.pok),
+ STAT_GENET_MIB_TX("tx_unicast", mib.tx.uc),
+ /* UniMAC RUNT counters */
+ STAT_GENET_RUNT("rx_runt_pkts", mib.rx_runt_cnt),
+ STAT_GENET_RUNT("rx_runt_valid_fcs", mib.rx_runt_fcs),
+ STAT_GENET_RUNT("rx_runt_inval_fcs_align", mib.rx_runt_fcs_align),
+ STAT_GENET_RUNT("rx_runt_bytes", mib.rx_runt_bytes),
+ /* Misc UniMAC counters */
+ STAT_GENET_MISC("rbuf_ovflow_cnt", mib.rbuf_ovflow_cnt,
+ UMAC_RBUF_OVFL_CNT),
+ STAT_GENET_MISC("rbuf_err_cnt", mib.rbuf_err_cnt, UMAC_RBUF_ERR_CNT),
+ STAT_GENET_MISC("mdf_err_cnt", mib.mdf_err_cnt, UMAC_MDF_ERR_CNT),
+};
+
+#define BCMGENET_STATS_LEN ARRAY_SIZE(bcmgenet_gstrings_stats)
+
+static void bcmgenet_get_drvinfo(struct net_device *dev,
+ struct ethtool_drvinfo *info)
+{
+ strlcpy(info->driver, "bcmgenet", sizeof(info->driver));
+ strlcpy(info->version, "v2.0", sizeof(info->version));
+ info->n_stats = BCMGENET_STATS_LEN;
+
+}
+
+static int bcmgenet_get_sset_count(struct net_device *dev, int string_set)
+{
+ switch (string_set) {
+ case ETH_SS_STATS:
+ return BCMGENET_STATS_LEN;
+ default:
+ return -EOPNOTSUPP;
+ }
+}
+
+static void bcmgenet_get_strings(struct net_device *dev,
+ u32 stringset, u8 *data)
+{
+ int i;
+
+ switch (stringset) {
+ case ETH_SS_STATS:
+ for (i = 0; i < BCMGENET_STATS_LEN; i++) {
+ memcpy(data + i * ETH_GSTRING_LEN,
+ bcmgenet_gstrings_stats[i].stat_string,
+ ETH_GSTRING_LEN);
+ }
+ break;
+ }
+}
+
+static void bcmgenet_update_mib_counters(struct bcmgenet_priv *priv)
+{
+ int i, j = 0;
+
+ for (i = 0; i < BCMGENET_STATS_LEN; i++) {
+ const struct bcmgenet_stats *s;
+ u8 offset = 0;
+ u32 val = 0;
+ char *p;
+
+ s = &bcmgenet_gstrings_stats[i];
+ switch (s->type) {
+ case BCMGENET_STAT_NETDEV:
+ continue;
+ case BCMGENET_STAT_MIB_RX:
+ case BCMGENET_STAT_MIB_TX:
+ case BCMGENET_STAT_RUNT:
+ if (s->type != BCMGENET_STAT_MIB_RX)
+ offset = BCMGENET_STAT_OFFSET;
+ val = bcmgenet_umac_readl(priv, UMAC_MIB_START +
+ j + offset);
+ break;
+ case BCMGENET_STAT_MISC:
+ val = bcmgenet_umac_readl(priv, s->reg_offset);
+ /* clear if overflowed */
+ if (val == ~0)
+ bcmgenet_umac_writel(priv, 0, s->reg_offset);
+ break;
+ }
+
+ j += s->stat_sizeof;
+ p = (char *)priv + s->stat_offset;
+ *(u32 *)p = val;
+ }
+}
+
+static void bcmgenet_get_ethtool_stats(struct net_device *dev,
+ struct ethtool_stats *stats,
+ u64 *data)
+{
+ struct bcmgenet_priv *priv = netdev_priv(dev);
+ int i;
+
+ if (netif_running(dev))
+ bcmgenet_update_mib_counters(priv);
+
+ for (i = 0; i < BCMGENET_STATS_LEN; i++) {
+ const struct bcmgenet_stats *s;
+ char *p;
+
+ s = &bcmgenet_gstrings_stats[i];
+ if (s->type == BCMGENET_STAT_NETDEV)
+ p = (char *)&dev->stats;
+ else
+ p = (char *)priv;
+ p += s->stat_offset;
+ data[i] = *(u32 *)p;
+ }
+}
+
+/* standard ethtool support functions. */
+static struct ethtool_ops bcmgenet_ethtool_ops = {
+ .get_strings = bcmgenet_get_strings,
+ .get_sset_count = bcmgenet_get_sset_count,
+ .get_ethtool_stats = bcmgenet_get_ethtool_stats,
+ .get_settings = bcmgenet_get_settings,
+ .set_settings = bcmgenet_set_settings,
+ .get_drvinfo = bcmgenet_get_drvinfo,
+ .get_link = ethtool_op_get_link,
+ .get_msglevel = bcmgenet_get_msglevel,
+ .set_msglevel = bcmgenet_set_msglevel,
+};
+
+/* Power down the unimac, based on mode. */
+static void bcmgenet_power_down(struct bcmgenet_priv *priv,
+ enum bcmgenet_power_mode mode)
+{
+ u32 reg;
+
+ switch (mode) {
+ case GENET_POWER_CABLE_SENSE:
+ if (priv->phydev)
+ phy_detach(priv->phydev);
+ break;
+
+ case GENET_POWER_PASSIVE:
+ /* Power down LED */
+ bcmgenet_mii_reset(priv->dev);
+ if (priv->hw_params->flags & GENET_HAS_EXT) {
+ reg = bcmgenet_ext_readl(priv, EXT_EXT_PWR_MGMT);
+ reg |= (EXT_PWR_DOWN_PHY |
+ EXT_PWR_DOWN_DLL | EXT_PWR_DOWN_BIAS);
+ bcmgenet_ext_writel(priv, reg, EXT_EXT_PWR_MGMT);
+ }
+ break;
+ default:
+ break;
+ }
+}
+
+static void bcmgenet_power_up(struct bcmgenet_priv *priv,
+ enum bcmgenet_power_mode mode)
+{
+ u32 reg;
+
+ if (!(priv->hw_params->flags & GENET_HAS_EXT))
+ return;
+
+ reg = bcmgenet_ext_readl(priv, EXT_EXT_PWR_MGMT);
+
+ switch (mode) {
+ case GENET_POWER_PASSIVE:
+ reg &= ~(EXT_PWR_DOWN_DLL | EXT_PWR_DOWN_PHY |
+ EXT_PWR_DOWN_BIAS);
+ /* fallthrough */
+ case GENET_POWER_CABLE_SENSE:
+ /* enable APD */
+ reg |= EXT_PWR_DN_EN_LD;
+ break;
+ default:
+ break;
+ }
+
+ bcmgenet_ext_writel(priv, reg, EXT_EXT_PWR_MGMT);
+ bcmgenet_mii_reset(priv->dev);
+}
+
+/* ioctl handle special commands that are not present in ethtool. */
+static int bcmgenet_ioctl(struct net_device *dev, struct ifreq *rq, int cmd)
+{
+ struct bcmgenet_priv *priv = netdev_priv(dev);
+ int val = 0;
+
+ if (!netif_running(dev))
+ return -EINVAL;
+
+ switch (cmd) {
+ case SIOCGMIIPHY:
+ case SIOCGMIIREG:
+ case SIOCSMIIREG:
+ if (!priv->phydev)
+ val = -ENODEV;
+ else
+ val = phy_mii_ioctl(priv->phydev, rq, cmd);
+ break;
+
+ default:
+ val = -EINVAL;
+ break;
+ }
+
+ return val;
+}
+
+static struct enet_cb *bcmgenet_get_txcb(struct bcmgenet_priv *priv,
+ struct bcmgenet_tx_ring *ring)
+{
+ struct enet_cb *tx_cb_ptr;
+
+ tx_cb_ptr = ring->cbs;
+ tx_cb_ptr += ring->write_ptr - ring->cb_ptr;
+ tx_cb_ptr->bd_addr = priv->tx_bds + ring->write_ptr * DMA_DESC_SIZE;
+ /* Advancing local write pointer */
+ if (ring->write_ptr == ring->end_ptr)
+ ring->write_ptr = ring->cb_ptr;
+ else
+ ring->write_ptr++;
+
+ return tx_cb_ptr;
+}
+
+/* Simple helper to free a control block's resources */
+static void bcmgenet_free_cb(struct enet_cb *cb)
+{
+ dev_kfree_skb_any(cb->skb);
+ cb->skb = NULL;
+ dma_unmap_addr_set(cb, dma_addr, 0);
+}
+
+static inline void bcmgenet_tx_ring16_int_disable(struct bcmgenet_priv *priv,
+ struct bcmgenet_tx_ring *ring)
+{
+ bcmgenet_intrl2_0_writel(priv,
+ UMAC_IRQ_TXDMA_BDONE | UMAC_IRQ_TXDMA_PDONE,
+ INTRL2_CPU_MASK_SET);
+}
+
+static inline void bcmgenet_tx_ring16_int_enable(struct bcmgenet_priv *priv,
+ struct bcmgenet_tx_ring *ring)
+{
+ bcmgenet_intrl2_0_writel(priv,
+ UMAC_IRQ_TXDMA_BDONE | UMAC_IRQ_TXDMA_PDONE,
+ INTRL2_CPU_MASK_CLEAR);
+}
+
+static inline void bcmgenet_tx_ring_int_enable(struct bcmgenet_priv *priv,
+ struct bcmgenet_tx_ring *ring)
+{
+ bcmgenet_intrl2_1_writel(priv,
+ (1 << ring->index), INTRL2_CPU_MASK_CLEAR);
+ priv->int1_mask &= ~(1 << ring->index);
+}
+
+static inline void bcmgenet_tx_ring_int_disable(struct bcmgenet_priv *priv,
+ struct bcmgenet_tx_ring *ring)
+{
+ bcmgenet_intrl2_1_writel(priv,
+ (1 << ring->index), INTRL2_CPU_MASK_SET);
+ priv->int1_mask |= (1 << ring->index);
+}
+
+/* Unlocked version of the reclaim routine */
+static void __bcmgenet_tx_reclaim(struct net_device *dev,
+ struct bcmgenet_tx_ring *ring)
+{
+ struct bcmgenet_priv *priv = netdev_priv(dev);
+ int last_tx_cn, last_c_index, num_tx_bds;
+ struct enet_cb *tx_cb_ptr;
+ unsigned int c_index;
+
+ /* Compute how many buffers are transmited since last xmit call */
+ c_index = bcmgenet_tdma_ring_readl(priv, ring->index, TDMA_CONS_INDEX);
+
+ last_c_index = ring->c_index;
+ num_tx_bds = ring->size;
+
+ c_index &= (num_tx_bds - 1);
+
+ if (c_index >= last_c_index)
+ last_tx_cn = c_index - last_c_index;
+ else
+ last_tx_cn = num_tx_bds - last_c_index + c_index;
+
+ netif_dbg(priv, tx_done, dev,
+ "%s ring=%d index=%d last_tx_cn=%d last_index=%d\n",
+ __func__, ring->index,
+ c_index, last_tx_cn, last_c_index);
+
+ /* Reclaim transmitted buffers */
+ while (last_tx_cn-- > 0) {
+ tx_cb_ptr = ring->cbs + last_c_index;
+ if (tx_cb_ptr->skb) {
+ dev->stats.tx_bytes += tx_cb_ptr->skb->len;
+ dma_unmap_single(&dev->dev,
+ dma_unmap_addr(tx_cb_ptr, dma_addr),
+ tx_cb_ptr->skb->len,
+ DMA_TO_DEVICE);
+ bcmgenet_free_cb(tx_cb_ptr);
+ } else if (dma_unmap_addr(tx_cb_ptr, dma_addr)) {
+ dev->stats.tx_bytes +=
+ dma_unmap_len(tx_cb_ptr, dma_len);
+ dma_unmap_page(&dev->dev,
+ dma_unmap_addr(tx_cb_ptr, dma_addr),
+ dma_unmap_len(tx_cb_ptr, dma_len),
+ DMA_TO_DEVICE);
+ dma_unmap_addr_set(tx_cb_ptr, dma_addr, 0);
+ }
+ dev->stats.tx_packets++;
+ ring->free_bds += 1;
+
+ last_c_index++;
+ last_c_index &= (num_tx_bds - 1);
+ }
+
+ if (ring->free_bds > (MAX_SKB_FRAGS + 1))
+ ring->int_disable(priv, ring);
+
+ if (__netif_subqueue_stopped(dev, ring->queue))
+ netif_wake_subqueue(dev, ring->queue);
+
+ ring->c_index = c_index;
+}
+
+static void bcmgenet_tx_reclaim(struct net_device *dev,
+ struct bcmgenet_tx_ring *ring)
+{
+ unsigned long flags;
+
+ spin_lock_irqsave(&ring->lock, flags);
+ __bcmgenet_tx_reclaim(dev, ring);
+ spin_unlock_irqrestore(&ring->lock, flags);
+}
+
+static void bcmgenet_tx_reclaim_all(struct net_device *dev)
+{
+ struct bcmgenet_priv *priv = netdev_priv(dev);
+ int i;
+
+ if (netif_is_multiqueue(dev)) {
+ for (i = 0; i < priv->hw_params->tx_queues; i++)
+ bcmgenet_tx_reclaim(dev, &priv->tx_rings[i]);
+ }
+
+ bcmgenet_tx_reclaim(dev, &priv->tx_rings[DESC_INDEX]);
+}
+
+/* Transmits a single SKB (either head of a fragment or a single SKB)
+ * caller must hold priv->lock
+ */
+static int bcmgenet_xmit_single(struct net_device *dev,
+ struct sk_buff *skb,
+ u16 dma_desc_flags,
+ struct bcmgenet_tx_ring *ring)
+{
+ struct bcmgenet_priv *priv = netdev_priv(dev);
+ struct device *kdev = &priv->pdev->dev;
+ struct enet_cb *tx_cb_ptr;
+ unsigned int skb_len;
+ dma_addr_t mapping;
+ u32 length_status;
+ int ret;
+
+ tx_cb_ptr = bcmgenet_get_txcb(priv, ring);
+
+ if (unlikely(!tx_cb_ptr))
+ BUG();
+
+ tx_cb_ptr->skb = skb;
+
+ skb_len = skb_headlen(skb) < ETH_ZLEN ? ETH_ZLEN : skb_headlen(skb);
+
+ mapping = dma_map_single(kdev, skb->data, skb_len, DMA_TO_DEVICE);
+ ret = dma_mapping_error(kdev, mapping);
+ if (ret) {
+ netif_err(priv, tx_err, dev, "Tx DMA map failed\n");
+ dev_kfree_skb(skb);
+ return ret;
+ }
+
+ dma_unmap_addr_set(tx_cb_ptr, dma_addr, mapping);
+ dma_unmap_len_set(tx_cb_ptr, dma_len, skb->len);
+ length_status = (skb_len << DMA_BUFLENGTH_SHIFT) | dma_desc_flags |
+ (priv->hw_params->qtag_mask << DMA_TX_QTAG_SHIFT) |
+ DMA_TX_APPEND_CRC;
+
+ if (skb->ip_summed == CHECKSUM_PARTIAL)
+ length_status |= DMA_TX_DO_CSUM;
+
+ dmadesc_set(priv, tx_cb_ptr->bd_addr, mapping, length_status);
+
+ /* Decrement total BD count and advance our write pointer */
+ ring->free_bds -= 1;
+ ring->prod_index += 1;
+ ring->prod_index &= DMA_P_INDEX_MASK;
+
+ return 0;
+}
+
+/* Transmit a SKB fragement */
+static int bcmgenet_xmit_frag(struct net_device *dev,
+ skb_frag_t *frag,
+ u16 dma_desc_flags,
+ struct bcmgenet_tx_ring *ring)
+{
+ struct bcmgenet_priv *priv = netdev_priv(dev);
+ struct device *kdev = &priv->pdev->dev;
+ struct enet_cb *tx_cb_ptr;
+ dma_addr_t mapping;
+ int ret;
+
+ tx_cb_ptr = bcmgenet_get_txcb(priv, ring);
+
+ if (unlikely(!tx_cb_ptr))
+ BUG();
+ tx_cb_ptr->skb = NULL;
+
+ mapping = skb_frag_dma_map(kdev, frag, 0,
+ skb_frag_size(frag), DMA_TO_DEVICE);
+ ret = dma_mapping_error(kdev, mapping);
+ if (ret) {
+ netif_err(priv, tx_err, dev, "%s: Tx DMA map failed\n",
+ __func__);
+ return ret;
+ }
+
+ dma_unmap_addr_set(tx_cb_ptr, dma_addr, mapping);
+ dma_unmap_len_set(tx_cb_ptr, dma_len, frag->size);
+
+ dmadesc_set(priv, tx_cb_ptr->bd_addr, mapping,
+ (frag->size << DMA_BUFLENGTH_SHIFT) | dma_desc_flags |
+ (priv->hw_params->qtag_mask << DMA_TX_QTAG_SHIFT));
+
+
+ ring->free_bds -= 1;
+ ring->prod_index += 1;
+ ring->prod_index &= DMA_P_INDEX_MASK;
+
+ return 0;
+}
+
+/* Reallocate the SKB to put enough headroom in front of it and insert
+ * the transmit checksum offsets in the descriptors
+ */
+static int bcmgenet_put_tx_csum(struct net_device *dev, struct sk_buff *skb)
+{
+ struct status_64 *status = NULL;
+ struct sk_buff *new_skb;
+ u16 offset;
+ u8 ip_proto;
+ u16 ip_ver;
+ u32 tx_csum_info;
+
+ if (unlikely(skb_headroom(skb) < sizeof(*status))) {
+ /* If 64 byte status block enabled, must make sure skb has
+ * enough headroom for us to insert 64B status block.
+ */
+ new_skb = skb_realloc_headroom(skb, sizeof(*status));
+ dev_kfree_skb(skb);
+ if (!new_skb) {
+ dev->stats.tx_errors++;
+ dev->stats.tx_dropped++;
+ return -ENOMEM;
+ }
+ skb = new_skb;
+ }
+
+ skb_push(skb, sizeof(*status));
+ status = (struct status_64 *)skb->data;
+
+ if (skb->ip_summed == CHECKSUM_PARTIAL) {
+ ip_ver = htons(skb->protocol);
+ switch (ip_ver) {
+ case ETH_P_IP:
+ ip_proto = ip_hdr(skb)->protocol;
+ break;
+ case ETH_P_IPV6:
+ ip_proto = ipv6_hdr(skb)->nexthdr;
+ break;
+ default:
+ return 0;
+ }
+
+ offset = skb_checksum_start_offset(skb) - sizeof(*status);
+ tx_csum_info = (offset << STATUS_TX_CSUM_START_SHIFT) |
+ (offset + skb->csum_offset);
+
+ /* Set the length valid bit for TCP and UDP and just set
+ * the special UDP flag for IPv4, else just set to 0.
+ */
+ if (ip_proto == IPPROTO_TCP || ip_proto == IPPROTO_UDP) {
+ tx_csum_info |= STATUS_TX_CSUM_LV;
+ if (ip_proto == IPPROTO_UDP && ip_ver == ETH_P_IP)
+ tx_csum_info |= STATUS_TX_CSUM_PROTO_UDP;
+ } else
+ tx_csum_info = 0;
+
+ status->tx_csum_info = tx_csum_info;
+ }
+
+ return 0;
+}
+
+static netdev_tx_t bcmgenet_xmit(struct sk_buff *skb, struct net_device *dev)
+{
+ struct bcmgenet_priv *priv = netdev_priv(dev);
+ struct bcmgenet_tx_ring *ring = NULL;
+ unsigned long flags = 0;
+ int nr_frags, index;
+ u16 dma_desc_flags;
+ int ret;
+ int i;
+
+ index = skb_get_queue_mapping(skb);
+ /* Mapping strategy:
+ * queue_mapping = 0, unclassified, packet xmited through ring16
+ * queue_mapping = 1, goes to ring 0. (highest priority queue
+ * queue_mapping = 2, goes to ring 1.
+ * queue_mapping = 3, goes to ring 2.
+ * queue_mapping = 4, goes to ring 3.
+ */
+ if (index == 0)
+ index = DESC_INDEX;
+ else
+ index -= 1;
+
+ if ((index != DESC_INDEX) && (index > priv->hw_params->tx_queues - 1)) {
+ netdev_err(dev, "%s: queue_mapping %d is invalid\n",
+ __func__, skb_get_queue_mapping(skb));
+ dev->stats.tx_errors++;
+ dev->stats.tx_dropped++;
+ ret = NETDEV_TX_OK;
+ goto out;
+ }
+ nr_frags = skb_shinfo(skb)->nr_frags;
+ ring = &priv->tx_rings[index];
+
+ spin_lock_irqsave(&ring->lock, flags);
+ if (ring->free_bds <= nr_frags + 1) {
+ netif_stop_subqueue(dev, ring->queue);
+ netdev_err(dev, "%s: tx ring %d full when queue %d awake\n",
+ __func__, index, ring->queue);
+ ret = NETDEV_TX_BUSY;
+ goto out;
+ }
+
+ /* reclaim xmited skb every 8 packets. */
+ /*if (ring->free_bds < ring->size - 8)*/
+ /*__bcmgenet_tx_reclaim(dev, ring);*/
+
+ /* set the SKB transmit checksum */
+ if (priv->desc_64b_en) {
+ ret = bcmgenet_put_tx_csum(dev, skb);
+ if (ret) {
+ ret = NETDEV_TX_OK;
+ goto out;
+ }
+ }
+
+ dma_desc_flags = DMA_SOP;
+ if (nr_frags == 0)
+ dma_desc_flags |= DMA_EOP;
+
+ /* Transmit single SKB or head of fragment list */
+ ret = bcmgenet_xmit_single(dev, skb, dma_desc_flags, ring);
+ if (ret) {
+ ret = NETDEV_TX_OK;
+ goto out;
+ }
+
+ /* xmit fragment */
+ for (i = 0; i < nr_frags; i++) {
+ ret = bcmgenet_xmit_frag(dev,
+ &skb_shinfo(skb)->frags[i],
+ (i == nr_frags - 1) ? DMA_EOP : 0, ring);
+ if (ret) {
+ ret = NETDEV_TX_OK;
+ goto out;
+ }
+ }
+
+ /* we kept a software copy of how much we should advance the TDMA
+ * producer index, now write it down to the hardware
+ */
+ bcmgenet_tdma_ring_writel(priv, ring->index,
+ ring->prod_index, TDMA_PROD_INDEX);
+
+ if (ring->free_bds <= (MAX_SKB_FRAGS + 1)) {
+ netif_stop_subqueue(dev, ring->queue);
+ ring->int_enable(priv, ring);
+ }
+
+out:
+ spin_unlock_irqrestore(&ring->lock, flags);
+
+ return ret;
+}
+
+
+static int bcmgenet_rx_refill(struct bcmgenet_priv *priv,
+ struct enet_cb *cb)
+{
+ struct device *kdev = &priv->pdev->dev;
+ struct sk_buff *skb;
+ dma_addr_t mapping;
+ int ret;
+
+ skb = netdev_alloc_skb(priv->dev,
+ priv->rx_buf_len + SKB_ALIGNMENT);
+ if (!skb)
+ return -ENOMEM;
+
+ /* a caller did not release this control block */
+ WARN_ON(cb->skb != NULL);
+ cb->skb = skb;
+ mapping = dma_map_single(kdev, skb->data,
+ priv->rx_buf_len, DMA_FROM_DEVICE);
+ ret = dma_mapping_error(kdev, mapping);
+ if (ret) {
+ bcmgenet_free_cb(cb);
+ netif_err(priv, rx_err, priv->dev,
+ "%s DMA map failed\n", __func__);
+ return ret;
+ }
+
+ dma_unmap_addr_set(cb, dma_addr, mapping);
+ /* assign packet, prepare descriptor, and advance pointer */
+
+ dmadesc_set_addr(priv, priv->rx_bd_assign_ptr, mapping);
+
+ /* turn on the newly assigned BD for DMA to use */
+ priv->rx_bd_assign_index++;
+ priv->rx_bd_assign_index &= (priv->num_rx_bds - 1);
+
+ priv->rx_bd_assign_ptr = priv->rx_bds +
+ (priv->rx_bd_assign_index * DMA_DESC_SIZE);
+
+ return 0;
+}
+
+/* bcmgenet_desc_rx - descriptor based rx process.
+ * this could be called from bottom half, or from NAPI polling method.
+ */
+static unsigned int bcmgenet_desc_rx(struct bcmgenet_priv *priv,
+ unsigned int budget)
+{
+ struct net_device *dev = priv->dev;
+ struct enet_cb *cb;
+ struct sk_buff *skb;
+ u32 dma_length_status;
+ unsigned long dma_flag;
+ int len, err;
+ unsigned int rxpktprocessed = 0, rxpkttoprocess;
+ unsigned int p_index;
+ unsigned int chksum_ok = 0;
+
+ p_index = bcmgenet_rdma_ring_readl(priv,
+ DESC_INDEX, RDMA_PROD_INDEX);
+ p_index &= DMA_P_INDEX_MASK;
+
+ if (p_index < priv->rx_c_index)
+ rxpkttoprocess = (DMA_C_INDEX_MASK + 1) -
+ priv->rx_c_index + p_index;
+ else
+ rxpkttoprocess = p_index - priv->rx_c_index;
+
+ netif_dbg(priv, rx_status, dev,
+ "RDMA: rxpkttoprocess=%d\n", rxpkttoprocess);
+
+ while ((rxpktprocessed < rxpkttoprocess) &&
+ (rxpktprocessed < budget)) {
+
+ /* Unmap the packet contents such that we can use the
+ * RSV from the 64 bytes descriptor when enabled and save
+ * a 32-bits register read
+ */
+ cb = &priv->rx_cbs[priv->rx_read_ptr];
+ skb = cb->skb;
+ dma_unmap_single(&dev->dev, dma_unmap_addr(cb, dma_addr),
+ priv->rx_buf_len, DMA_FROM_DEVICE);
+
+ if (!priv->desc_64b_en) {
+ dma_length_status = dmadesc_get_length_status(priv,
+ priv->rx_bds +
+ (priv->rx_read_ptr *
+ DMA_DESC_SIZE));
+ } else {
+ struct status_64 *status;
+ status = (struct status_64 *)skb->data;
+ dma_length_status = status->length_status;
+ }
+
+ /* DMA flags and length are still valid no matter how
+ * we got the Receive Status Vector (64B RSB or register)
+ */
+ dma_flag = dma_length_status & 0xffff;
+ len = dma_length_status >> DMA_BUFLENGTH_SHIFT;
+
+ netif_dbg(priv, rx_status, dev,
+ "%s: p_ind=%d c_ind=%d read_ptr=%d len_stat=0x%08x\n",
+ __func__, p_index, priv->rx_c_index, priv->rx_read_ptr,
+ dma_length_status);
+
+ rxpktprocessed++;
+
+ priv->rx_read_ptr++;
+ priv->rx_read_ptr &= (priv->num_rx_bds - 1);
+
+ /* out of memory, just drop packets at the hardware level */
+ if (unlikely(!skb)) {
+ dev->stats.rx_dropped++;
+ dev->stats.rx_errors++;
+ goto refill;
+ }
+
+ if (unlikely(!(dma_flag & DMA_EOP) || !(dma_flag & DMA_SOP))) {
+ netif_err(priv, rx_status, dev,
+ "Droping fragmented packet!\n");
+ dev->stats.rx_dropped++;
+ dev->stats.rx_errors++;
+ dev_kfree_skb_any(cb->skb);
+ cb->skb = NULL;
+ goto refill;
+ }
+ /* report errors */
+ if (unlikely(dma_flag & (DMA_RX_CRC_ERROR |
+ DMA_RX_OV |
+ DMA_RX_NO |
+ DMA_RX_LG |
+ DMA_RX_RXER))) {
+ netif_err(priv, rx_status, dev, "dma_flag=0x%x\n",
+ (unsigned int)dma_flag);
+ if (dma_flag & DMA_RX_CRC_ERROR)
+ dev->stats.rx_crc_errors++;
+ if (dma_flag & DMA_RX_OV)
+ dev->stats.rx_over_errors++;
+ if (dma_flag & DMA_RX_NO)
+ dev->stats.rx_frame_errors++;
+ if (dma_flag & DMA_RX_LG)
+ dev->stats.rx_length_errors++;
+ dev->stats.rx_dropped++;
+ dev->stats.rx_errors++;
+
+ /* discard the packet and advance consumer index.*/
+ dev_kfree_skb_any(cb->skb);
+ cb->skb = NULL;
+ goto refill;
+ } /* error packet */
+
+ chksum_ok = (dma_flag & priv->dma_rx_chk_bit) &&
+ priv->desc_rxchk_en;
+
+ skb_put(skb, len);
+ if (priv->desc_64b_en) {
+ skb_pull(skb, 64);
+ len -= 64;
+ }
+
+ if (likely(chksum_ok))
+ skb->ip_summed = CHECKSUM_UNNECESSARY;
+
+ /* remove hardware 2bytes added for IP alignment */
+ skb_pull(skb, 2);
+ len -= 2;
+
+ if (priv->crc_fwd_en) {
+ skb_trim(skb, len - ETH_FCS_LEN);
+ len -= ETH_FCS_LEN;
+ }
+
+ /*Finish setting up the received SKB and send it to the kernel*/
+ skb->protocol = eth_type_trans(skb, priv->dev);
+ dev->stats.rx_packets++;
+ dev->stats.rx_bytes += len;
+ if (dma_flag & DMA_RX_MULT)
+ dev->stats.multicast++;
+
+ /* Notify kernel */
+ napi_gro_receive(&priv->napi, skb);
+ cb->skb = NULL;
+ netif_dbg(priv, rx_status, dev, "pushed up to kernel\n");
+
+ /* refill RX path on the current control block */
+refill:
+ err = bcmgenet_rx_refill(priv, cb);
+ if (err)
+ netif_err(priv, rx_err, dev, "Rx refill failed\n");
+ }
+
+ return rxpktprocessed;
+}
+
+/* Assign skb to RX DMA descriptor. */
+static int bcmgenet_alloc_rx_buffers(struct bcmgenet_priv *priv)
+{
+ struct enet_cb *cb;
+ int ret = 0;
+ int i;
+
+ netif_dbg(priv, hw, priv->dev, "%s:\n", __func__);
+
+ /* loop here for each buffer needing assign */
+ for (i = 0; i < priv->num_rx_bds; i++) {
+ cb = &priv->rx_cbs[priv->rx_bd_assign_index];
+ if (cb->skb)
+ continue;
+
+ /* set the DMA descriptor length once and for all
+ * it will only change if we support dynamically sizing
+ * priv->rx_buf_len, but we do not
+ */
+ dmadesc_set_length_status(priv, priv->rx_bd_assign_ptr,
+ priv->rx_buf_len << DMA_BUFLENGTH_SHIFT);
+
+ ret = bcmgenet_rx_refill(priv, cb);
+ if (ret)
+ break;
+
+ }
+
+ return ret;
+}
+
+static void bcmgenet_free_rx_buffers(struct bcmgenet_priv *priv)
+{
+ struct enet_cb *cb;
+ int i;
+
+ for (i = 0; i < priv->num_rx_bds; i++) {
+ cb = &priv->rx_cbs[i];
+
+ if (dma_unmap_addr(cb, dma_addr)) {
+ dma_unmap_single(&priv->dev->dev,
+ dma_unmap_addr(cb, dma_addr),
+ priv->rx_buf_len, DMA_FROM_DEVICE);
+ dma_unmap_addr_set(cb, dma_addr, 0);
+ }
+
+ if (cb->skb)
+ bcmgenet_free_cb(cb);
+ }
+}
+
+static int reset_umac(struct bcmgenet_priv *priv)
+{
+ struct device *kdev = &priv->pdev->dev;
+ unsigned int timeout = 0;
+ u32 reg;
+
+ /* 7358a0/7552a0: bad default in RBUF_FLUSH_CTRL.umac_sw_rst */
+ bcmgenet_rbuf_ctrl_set(priv, 0);
+ udelay(10);
+
+ /* disable MAC while updating its registers */
+ bcmgenet_umac_writel(priv, 0, UMAC_CMD);
+
+ /* issue soft reset, wait for it to complete */
+ bcmgenet_umac_writel(priv, CMD_SW_RESET, UMAC_CMD);
+ while (timeout++ < 1000) {
+ reg = bcmgenet_umac_readl(priv, UMAC_CMD);
+ if (!(reg & CMD_SW_RESET))
+ return 0;
+
+ udelay(1);
+ }
+
+ if (timeout == 1000) {
+ dev_err(kdev,
+ "timeout waiting for MAC to come out of resetn\n");
+ return -ETIMEDOUT;
+ }
+
+ return 0;
+}
+
+static int init_umac(struct bcmgenet_priv *priv)
+{
+ struct device *kdev = &priv->pdev->dev;
+ int ret;
+ u32 reg, cpu_mask_clear;
+
+ dev_dbg(&priv->pdev->dev, "bcmgenet: init_umac\n");
+
+ ret = reset_umac(priv);
+ if (ret)
+ return ret;
+
+ bcmgenet_umac_writel(priv, 0, UMAC_CMD);
+ /* clear tx/rx counter */
+ bcmgenet_umac_writel(priv,
+ MIB_RESET_RX | MIB_RESET_TX | MIB_RESET_RUNT, UMAC_MIB_CTRL);
+ bcmgenet_umac_writel(priv, 0, UMAC_MIB_CTRL);
+
+ bcmgenet_umac_writel(priv, ENET_MAX_MTU_SIZE, UMAC_MAX_FRAME_LEN);
+
+ /* init rx registers, enable ip header optimization */
+ reg = bcmgenet_rbuf_readl(priv, RBUF_CTRL);
+ reg |= RBUF_ALIGN_2B;
+ bcmgenet_rbuf_writel(priv, reg, RBUF_CTRL);
+
+ if (!GENET_IS_V1(priv) && !GENET_IS_V2(priv))
+ bcmgenet_rbuf_writel(priv, 1, RBUF_TBUF_SIZE_CTRL);
+
+ /* Mask all interrupts.*/
+ bcmgenet_intrl2_0_writel(priv, 0xFFFFFFFF, INTRL2_CPU_MASK_SET);
+ bcmgenet_intrl2_0_writel(priv, 0xFFFFFFFF, INTRL2_CPU_CLEAR);
+ bcmgenet_intrl2_0_writel(priv, 0, INTRL2_CPU_MASK_CLEAR);
+
+ cpu_mask_clear = UMAC_IRQ_RXDMA_BDONE;
+
+ dev_dbg(kdev, "%s:Enabling RXDMA_BDONE interrupt\n", __func__);
+
+ /* Monitor cable plug/unpluged event for internal PHY */
+ if (phy_is_internal(priv->phydev))
+ cpu_mask_clear |= (UMAC_IRQ_LINK_DOWN | UMAC_IRQ_LINK_UP);
+ else if (priv->ext_phy)
+ cpu_mask_clear |= (UMAC_IRQ_LINK_DOWN | UMAC_IRQ_LINK_UP);
+ else if (priv->phy_interface == PHY_INTERFACE_MODE_MOCA) {
+ reg = bcmgenet_bp_mc_get(priv);
+ reg |= BIT(priv->hw_params->bp_in_en_shift);
+
+ /* bp_mask: back pressure mask */
+ if (netif_is_multiqueue(priv->dev))
+ reg |= priv->hw_params->bp_in_mask;
+ else
+ reg &= ~priv->hw_params->bp_in_mask;
+ bcmgenet_bp_mc_set(priv, reg);
+ }
+
+ /* Enable MDIO interrupts on GENET v3+ */
+ if (priv->hw_params->flags & GENET_HAS_MDIO_INTR)
+ cpu_mask_clear |= UMAC_IRQ_MDIO_DONE | UMAC_IRQ_MDIO_ERROR;
+
+ bcmgenet_intrl2_0_writel(priv, cpu_mask_clear,
+ INTRL2_CPU_MASK_CLEAR);
+
+ /* Enable rx/tx engine.*/
+ dev_dbg(kdev, "done init umac\n");
+
+ return 0;
+}
+
+/* Initialize all house-keeping variables for a TX ring, along
+ * with corresponding hardware registers
+ */
+static void bcmgenet_init_tx_ring(struct bcmgenet_priv *priv,
+ unsigned int index, unsigned int size,
+ unsigned int write_ptr, unsigned int end_ptr)
+{
+ struct bcmgenet_tx_ring *ring = &priv->tx_rings[index];
+ u32 words_per_bd = WORDS_PER_BD(priv);
+ u32 flow_period_val = 0;
+ unsigned int first_bd;
+
+ spin_lock_init(&ring->lock);
+ ring->index = index;
+ if (index == DESC_INDEX) {
+ ring->queue = 0;
+ ring->int_enable = bcmgenet_tx_ring16_int_enable;
+ ring->int_disable = bcmgenet_tx_ring16_int_disable;
+ } else {
+ ring->queue = index + 1;
+ ring->int_enable = bcmgenet_tx_ring_int_enable;
+ ring->int_disable = bcmgenet_tx_ring_int_disable;
+ }
+ ring->cbs = priv->tx_cbs + write_ptr;
+ ring->size = size;
+ ring->c_index = 0;
+ ring->free_bds = size;
+ ring->write_ptr = write_ptr;
+ ring->cb_ptr = write_ptr;
+ ring->end_ptr = end_ptr - 1;
+ ring->prod_index = 0;
+
+ /* Set flow period for ring != 16 */
+ if (index != DESC_INDEX)
+ flow_period_val = ENET_MAX_MTU_SIZE << 16;
+
+ bcmgenet_tdma_ring_writel(priv, index, 0, TDMA_PROD_INDEX);
+ bcmgenet_tdma_ring_writel(priv, index, 0, TDMA_CONS_INDEX);
+ bcmgenet_tdma_ring_writel(priv, index, 1, DMA_MBUF_DONE_THRESH);
+ /* Disable rate control for now */
+ bcmgenet_tdma_ring_writel(priv, index, flow_period_val,
+ TDMA_FLOW_PERIOD);
+ /* Unclassified traffic goes to ring 16 */
+ bcmgenet_tdma_ring_writel(priv, index,
+ ((size << DMA_RING_SIZE_SHIFT) | RX_BUF_LENGTH),
+ DMA_RING_BUF_SIZE);
+
+ first_bd = write_ptr;
+
+ /* Set start and end address, read and write pointers */
+ bcmgenet_tdma_ring_writel(priv, index, first_bd * words_per_bd,
+ DMA_START_ADDR);
+ bcmgenet_tdma_ring_writel(priv, index, first_bd * words_per_bd,
+ TDMA_READ_PTR);
+ bcmgenet_tdma_ring_writel(priv, index, first_bd,
+ TDMA_WRITE_PTR);
+ bcmgenet_tdma_ring_writel(priv, index, end_ptr * words_per_bd - 1,
+ DMA_END_ADDR);
+}
+
+/* Initialize a RDMA ring */
+static int bcmgenet_init_rx_ring(struct bcmgenet_priv *priv,
+ unsigned int index, unsigned int size)
+{
+ u32 words_per_bd = WORDS_PER_BD(priv);
+ int ret;
+
+ priv->num_rx_bds = TOTAL_DESC;
+ priv->rx_bds = priv->base + priv->hw_params->rdma_offset;
+ priv->rx_bd_assign_ptr = priv->rx_bds;
+ priv->rx_bd_assign_index = 0;
+ priv->rx_c_index = 0;
+ priv->rx_read_ptr = 0;
+ priv->rx_cbs = kzalloc(priv->num_rx_bds * sizeof(struct enet_cb),
+ GFP_KERNEL);
+ if (!priv->rx_cbs)
+ return -ENOMEM;
+
+ ret = bcmgenet_alloc_rx_buffers(priv);
+ if (ret) {
+ kfree(priv->rx_cbs);
+ return ret;
+ }
+
+ bcmgenet_rdma_ring_writel(priv, index, 0, RDMA_WRITE_PTR);
+ bcmgenet_rdma_ring_writel(priv, index, 0, RDMA_PROD_INDEX);
+ bcmgenet_rdma_ring_writel(priv, index, 0, RDMA_CONS_INDEX);
+ bcmgenet_rdma_ring_writel(priv, index,
+ ((size << DMA_RING_SIZE_SHIFT) | RX_BUF_LENGTH),
+ DMA_RING_BUF_SIZE);
+ bcmgenet_rdma_ring_writel(priv, index, 0, DMA_START_ADDR);
+ bcmgenet_rdma_ring_writel(priv, index,
+ words_per_bd * size - 1, DMA_END_ADDR);
+ bcmgenet_rdma_ring_writel(priv, index,
+ (DMA_FC_THRESH_LO << DMA_XOFF_THRESHOLD_SHIFT) |
+ DMA_FC_THRESH_HI, RDMA_XON_XOFF_THRESH);
+ bcmgenet_rdma_ring_writel(priv, index, 0, RDMA_READ_PTR);
+
+ return ret;
+}
+
+/* init multi xmit queues, only available for GENET2+
+ * the queue is partitioned as follows:
+ *
+ * queue 0 - 3 is priority based, each one has 32 descriptors,
+ * with queue 0 being the highest priority queue.
+ *
+ * queue 16 is the default tx queue with GENET_DEFAULT_BD_CNT
+ * descriptors: 256 - (number of tx queues * bds per queues) = 128
+ * descriptors.
+ *
+ * The transmit control block pool is then partitioned as following:
+ * - tx_cbs[0...127] are for queue 16
+ * - tx_ring_cbs[0] points to tx_cbs[128..159]
+ * - tx_ring_cbs[1] points to tx_cbs[160..191]
+ * - tx_ring_cbs[2] points to tx_cbs[192..223]
+ * - tx_ring_cbs[3] points to tx_cbs[224..255]
+ */
+static void bcmgenet_init_multiq(struct net_device *dev)
+{
+ struct bcmgenet_priv *priv = netdev_priv(dev);
+ unsigned int i, dma_enable;
+ u32 reg, dma_ctrl, ring_cfg = 0, dma_priority = 0;
+
+ if (!netif_is_multiqueue(dev)) {
+ netdev_warn(dev, "called with non multi queue aware HW\n");
+ return;
+ }
+
+ dma_ctrl = bcmgenet_tdma_readl(priv, DMA_CTRL);
+ dma_enable = dma_ctrl & DMA_EN;
+ dma_ctrl &= ~DMA_EN;
+ bcmgenet_tdma_writel(priv, dma_ctrl, DMA_CTRL);
+
+ /* Enable strict priority arbiter mode */
+ bcmgenet_tdma_writel(priv, DMA_ARBITER_SP, DMA_ARB_CTRL);
+
+ for (i = 0; i < priv->hw_params->tx_queues; i++) {
+ /* first 64 tx_cbs are reserved for default tx queue
+ * (ring 16)
+ */
+ bcmgenet_init_tx_ring(priv, i, priv->hw_params->bds_cnt,
+ i * priv->hw_params->bds_cnt,
+ (i + 1) * priv->hw_params->bds_cnt);
+
+ /* Configure ring as decriptor ring and setup priority */
+ ring_cfg |= 1 << i;
+ dma_priority |= ((GENET_Q0_PRIORITY + i) <<
+ (GENET_MAX_MQ_CNT + 1) * i);
+ dma_ctrl |= 1 << (i + DMA_RING_BUF_EN_SHIFT);
+ }
+
+ /* Enable rings */
+ reg = bcmgenet_tdma_readl(priv, DMA_RING_CFG);
+ reg |= ring_cfg;
+ bcmgenet_tdma_writel(priv, reg, DMA_RING_CFG);
+
+ /* Use configured rings priority and set ring #16 priority */
+ reg = bcmgenet_tdma_readl(priv, DMA_RING_PRIORITY);
+ reg |= ((GENET_Q0_PRIORITY + priv->hw_params->tx_queues) << 20);
+ reg |= dma_priority;
+ bcmgenet_tdma_writel(priv, reg, DMA_PRIORITY);
+
+ /* Configure ring as descriptor ring and re-enable DMA if enabled */
+ reg = bcmgenet_tdma_readl(priv, DMA_CTRL);
+ reg |= dma_ctrl;
+ if (dma_enable)
+ reg |= DMA_EN;
+ bcmgenet_tdma_writel(priv, reg, DMA_CTRL);
+}
+
+static void bcmgenet_fini_dma(struct bcmgenet_priv *priv)
+{
+ int i;
+
+ /* disable DMA */
+ bcmgenet_rdma_writel(priv, 0, DMA_CTRL);
+ bcmgenet_tdma_writel(priv, 0, DMA_CTRL);
+
+ for (i = 0; i < priv->num_tx_bds; i++) {
+ if (priv->tx_cbs[i].skb != NULL) {
+ dev_kfree_skb(priv->tx_cbs[i].skb);
+ priv->tx_cbs[i].skb = NULL;
+ }
+ }
+
+ bcmgenet_free_rx_buffers(priv);
+ kfree(priv->rx_cbs);
+ kfree(priv->tx_cbs);
+}
+
+/* init_edma: Initialize DMA control register */
+static int bcmgenet_init_dma(struct bcmgenet_priv *priv)
+{
+ int ret;
+
+ netif_dbg(priv, hw, priv->dev, "bcmgenet: init_edma\n");
+
+ /* by default, enable ring 16 (descriptor based) */
+ ret = bcmgenet_init_rx_ring(priv, DESC_INDEX, TOTAL_DESC);
+ if (ret) {
+ netdev_err(priv->dev, "failed to initialize RX ring\n");
+ return ret;
+ }
+
+ /* init rDma */
+ bcmgenet_rdma_writel(priv, DMA_MAX_BURST_LENGTH, DMA_SCB_BURST_SIZE);
+
+ /* Init tDma */
+ bcmgenet_tdma_writel(priv, DMA_MAX_BURST_LENGTH, DMA_SCB_BURST_SIZE);
+
+ /* Initialize commont TX ring structures */
+ priv->tx_bds = priv->base + priv->hw_params->tdma_offset;
+ priv->num_tx_bds = TOTAL_DESC;
+ priv->tx_cbs = kzalloc(priv->num_tx_bds * sizeof(struct enet_cb),
+ GFP_KERNEL);
+ if (!priv->tx_cbs) {
+ bcmgenet_fini_dma(priv);
+ return -ENOMEM;
+ }
+
+ /* initialize multi xmit queue */
+ bcmgenet_init_multiq(priv->dev);
+
+ /* initialize special ring 16 */
+ bcmgenet_init_tx_ring(priv, DESC_INDEX, GENET_DEFAULT_BD_CNT,
+ priv->hw_params->tx_queues * priv->hw_params->bds_cnt,
+ TOTAL_DESC);
+
+ return 0;
+}
+
+/* NAPI polling method*/
+static int bcmgenet_poll(struct napi_struct *napi, int budget)
+{
+ struct bcmgenet_priv *priv = container_of(napi,
+ struct bcmgenet_priv, napi);
+ unsigned int work_done;
+
+ /* tx reclaim */
+ bcmgenet_tx_reclaim(priv->dev, &priv->tx_rings[DESC_INDEX]);
+
+ work_done = bcmgenet_desc_rx(priv, budget);
+
+ /* Advancing our consumer index*/
+ priv->rx_c_index += work_done;
+ priv->rx_c_index &= DMA_C_INDEX_MASK;
+ bcmgenet_rdma_ring_writel(priv, DESC_INDEX,
+ priv->rx_c_index, RDMA_CONS_INDEX);
+ if (work_done < budget) {
+ napi_complete(napi);
+ bcmgenet_intrl2_0_writel(priv,
+ UMAC_IRQ_RXDMA_BDONE, INTRL2_CPU_MASK_CLEAR);
+ }
+
+ return work_done;
+}
+
+/* Interrupt bottom half */
+static void bcmgenet_irq_task(struct work_struct *work)
+{
+ struct bcmgenet_priv *priv = container_of(
+ work, struct bcmgenet_priv, bcmgenet_irq_work);
+
+ netif_dbg(priv, intr, priv->dev, "%s\n", __func__);
+
+ /* Link UP/DOWN event */
+ if ((priv->hw_params->flags & GENET_HAS_MDIO_INTR) &&
+ (priv->irq0_stat & (UMAC_IRQ_LINK_UP|UMAC_IRQ_LINK_DOWN))) {
+ if (priv->phydev)
+ phy_mac_interrupt(priv->phydev,
+ (priv->irq0_stat & UMAC_IRQ_LINK_UP));
+ priv->irq0_stat &= ~(UMAC_IRQ_LINK_UP|UMAC_IRQ_LINK_DOWN);
+ }
+}
+
+/* bcmgenet_isr1: interrupt handler for ring buffer. */
+static irqreturn_t bcmgenet_isr1(int irq, void *dev_id)
+{
+ struct bcmgenet_priv *priv = dev_id;
+ unsigned int index;
+
+ /* Save irq status for bottom-half processing. */
+ priv->irq1_stat =
+ bcmgenet_intrl2_1_readl(priv, INTRL2_CPU_STAT) &
+ ~priv->int1_mask;
+ /* clear inerrupts*/
+ bcmgenet_intrl2_1_writel(priv, priv->irq1_stat, INTRL2_CPU_CLEAR);
+
+ netif_dbg(priv, intr, priv->dev,
+ "%s: IRQ=0x%x\n", __func__, priv->irq1_stat);
+ /* Check the MBDONE interrupts.
+ * packet is done, reclaim descriptors
+ */
+ if (priv->irq1_stat & 0x0000ffff) {
+ index = 0;
+ for (index = 0; index < 16; index++) {
+ if (priv->irq1_stat & (1 << index))
+ bcmgenet_tx_reclaim(priv->dev,
+ &priv->tx_rings[index]);
+ }
+ }
+ return IRQ_HANDLED;
+}
+
+/* bcmgenet_isr0: Handle various interrupts. */
+static irqreturn_t bcmgenet_isr0(int irq, void *dev_id)
+{
+ struct bcmgenet_priv *priv = dev_id;
+
+ /* Save irq status for bottom-half processing. */
+ priv->irq0_stat =
+ bcmgenet_intrl2_0_readl(priv, INTRL2_CPU_STAT) &
+ ~bcmgenet_intrl2_0_readl(priv, INTRL2_CPU_MASK_STATUS);
+ /* clear inerrupts*/
+ bcmgenet_intrl2_0_writel(priv, priv->irq0_stat, INTRL2_CPU_CLEAR);
+
+ netif_dbg(priv, intr, priv->dev,
+ "IRQ=0x%x\n", priv->irq0_stat);
+
+ if (priv->irq0_stat & (UMAC_IRQ_RXDMA_BDONE | UMAC_IRQ_RXDMA_PDONE)) {
+ /* We use NAPI(software interrupt throttling, if
+ * Rx Descriptor throttling is not used.
+ * Disable interrupt, will be enabled in the poll method.
+ */
+ if (likely(napi_schedule_prep(&priv->napi))) {
+ bcmgenet_intrl2_0_writel(priv,
+ UMAC_IRQ_RXDMA_BDONE, INTRL2_CPU_MASK_SET);
+ __napi_schedule(&priv->napi);
+ }
+ }
+ if (priv->irq0_stat &
+ (UMAC_IRQ_TXDMA_BDONE | UMAC_IRQ_TXDMA_PDONE)) {
+ /* Tx reclaim */
+ bcmgenet_tx_reclaim(priv->dev, &priv->tx_rings[DESC_INDEX]);
+ }
+ if (priv->irq0_stat & (UMAC_IRQ_PHY_DET_R |
+ UMAC_IRQ_PHY_DET_F |
+ UMAC_IRQ_LINK_UP |
+ UMAC_IRQ_LINK_DOWN |
+ UMAC_IRQ_HFB_SM |
+ UMAC_IRQ_HFB_MM |
+ UMAC_IRQ_MPD_R)) {
+ /* all other interested interrupts handled in bottom half */
+ schedule_work(&priv->bcmgenet_irq_work);
+ }
+
+ if ((priv->hw_params->flags & GENET_HAS_MDIO_INTR) &&
+ priv->irq0_stat & (UMAC_IRQ_MDIO_DONE | UMAC_IRQ_MDIO_ERROR)) {
+ priv->irq0_stat &= ~(UMAC_IRQ_MDIO_DONE | UMAC_IRQ_MDIO_ERROR);
+ wake_up(&priv->wq);
+ }
+
+ return IRQ_HANDLED;
+}
+
+static void bcmgenet_umac_reset(struct bcmgenet_priv *priv)
+{
+ u32 reg;
+
+ reg = bcmgenet_rbuf_ctrl_get(priv);
+ reg |= BIT(1);
+ bcmgenet_rbuf_ctrl_set(priv, reg);
+ udelay(10);
+
+ reg &= ~BIT(1);
+ bcmgenet_rbuf_ctrl_set(priv, reg);
+ udelay(10);
+}
+
+static void bcmgenet_set_hw_addr(struct bcmgenet_priv *priv,
+ unsigned char *addr)
+{
+ bcmgenet_umac_writel(priv, (addr[0] << 24) | (addr[1] << 16) |
+ (addr[2] << 8) | addr[3], UMAC_MAC0);
+ bcmgenet_umac_writel(priv, (addr[4] << 8) | addr[5], UMAC_MAC1);
+}
+
+static int bcmgenet_wol_resume(struct bcmgenet_priv *priv)
+{
+ int ret;
+
+ /* From WOL-enabled suspend, switch to regular clock */
+ clk_disable(priv->clk_wol);
+ /* init umac registers to synchronize s/w with h/w */
+ ret = init_umac(priv);
+ if (ret)
+ return ret;
+
+ if (priv->phydev)
+ phy_init_hw(priv->phydev);
+ /* Speed settings must be restored */
+ bcmgenet_mii_config(priv->dev);
+
+ return 0;
+}
+
+/* Returns a reusable dma control register value */
+static u32 bcmgenet_dma_disable(struct bcmgenet_priv *priv)
+{
+ u32 reg;
+ u32 dma_ctrl;
+
+ /* disable DMA */
+ dma_ctrl = 1 << (DESC_INDEX + DMA_RING_BUF_EN_SHIFT) | DMA_EN;
+ reg = bcmgenet_tdma_readl(priv, DMA_CTRL);
+ reg &= ~dma_ctrl;
+ bcmgenet_tdma_writel(priv, reg, DMA_CTRL);
+
+ reg = bcmgenet_rdma_readl(priv, DMA_CTRL);
+ reg &= ~dma_ctrl;
+ bcmgenet_rdma_writel(priv, reg, DMA_CTRL);
+
+ bcmgenet_umac_writel(priv, 1, UMAC_TX_FLUSH);
+ udelay(10);
+ bcmgenet_umac_writel(priv, 0, UMAC_TX_FLUSH);
+
+ return dma_ctrl;
+}
+
+static void bcmgenet_enable_dma(struct bcmgenet_priv *priv, u32 dma_ctrl)
+{
+ u32 reg;
+
+ reg = bcmgenet_rdma_readl(priv, DMA_CTRL);
+ reg |= dma_ctrl;
+ bcmgenet_rdma_writel(priv, reg, DMA_CTRL);
+
+ reg = bcmgenet_tdma_readl(priv, DMA_CTRL);
+ reg |= dma_ctrl;
+ bcmgenet_tdma_writel(priv, reg, DMA_CTRL);
+}
+
+static int bcmgenet_open(struct net_device *dev)
+{
+ struct bcmgenet_priv *priv = netdev_priv(dev);
+ unsigned long dma_ctrl;
+ u32 reg;
+ int ret;
+
+ netif_dbg(priv, ifup, dev, "bcmgenet_open\n");
+
+ /* Turn on the clock */
+ if (!IS_ERR(priv->clk))
+ clk_prepare_enable(priv->clk);
+
+ /* take MAC out of reset */
+ bcmgenet_umac_reset(priv);
+
+ ret = init_umac(priv);
+ if (ret)
+ goto err_clk_disable;
+
+ /* disable ethernet MAC while updating its registers */
+ reg = bcmgenet_umac_readl(priv, UMAC_CMD);
+ reg &= ~(CMD_TX_EN | CMD_RX_EN);
+ bcmgenet_umac_writel(priv, reg, UMAC_CMD);
+
+ bcmgenet_set_hw_addr(priv, dev->dev_addr);
+
+ if (priv->wol_enabled) {
+ ret = bcmgenet_wol_resume(priv);
+ if (ret)
+ return ret;
+ }
+
+ if (phy_is_internal(priv->phydev)) {
+ reg = bcmgenet_ext_readl(priv, EXT_EXT_PWR_MGMT);
+ reg |= EXT_ENERGY_DET_MASK;
+ bcmgenet_ext_writel(priv, reg, EXT_EXT_PWR_MGMT);
+ }
+
+ /* Disable RX/TX DMA and flush TX queues */
+ dma_ctrl = bcmgenet_dma_disable(priv);
+
+ /* Reinitialize TDMA and RDMA and SW housekeeping */
+ ret = bcmgenet_init_dma(priv);
+ if (ret) {
+ netdev_err(dev, "failed to initialize DMA\n");
+ goto err_fini_dma;
+ }
+
+ /* Always enable ring 16 - descriptor ring */
+ bcmgenet_enable_dma(priv, dma_ctrl);
+
+ ret = request_irq(priv->irq0, bcmgenet_isr0, IRQF_SHARED,
+ dev->name, priv);
+ if (ret < 0) {
+ netdev_err(dev, "can't request IRQ %d\n", priv->irq0);
+ goto err_fini_dma;
+ }
+
+ ret = request_irq(priv->irq1, bcmgenet_isr1, IRQF_SHARED,
+ dev->name, priv);
+ if (ret < 0) {
+ netdev_err(dev, "can't request IRQ %d\n", priv->irq1);
+ goto err_irq0;
+ }
+
+ /* Start the network engine */
+ napi_enable(&priv->napi);
+
+ reg = bcmgenet_umac_readl(priv, UMAC_CMD);
+ reg |= (CMD_TX_EN | CMD_RX_EN);
+ bcmgenet_umac_writel(priv, reg, UMAC_CMD);
+
+ /* Make sure we reflect the value of CRC_CMD_FWD */
+ priv->crc_fwd_en = !!(reg & CMD_CRC_FWD);
+
+ device_set_wakeup_capable(&dev->dev, 1);
+
+ if (phy_is_internal(priv->phydev))
+ bcmgenet_power_up(priv, GENET_POWER_PASSIVE);
+
+ netif_tx_start_all_queues(dev);
+
+ if (priv->phydev)
+ phy_start(priv->phydev);
+
+ return 0;
+
+err_irq0:
+ free_irq(priv->irq0, dev);
+err_fini_dma:
+ bcmgenet_fini_dma(priv);
+err_clk_disable:
+ if (!IS_ERR(priv->clk))
+ clk_disable_unprepare(priv->clk);
+ return ret;
+}
+
+static int bcmgenet_dma_teardown(struct bcmgenet_priv *priv)
+{
+ int ret = 0;
+ int timeout = 0;
+ u32 reg;
+
+ /* Disable TDMA to stop add more frames in TX DMA */
+ reg = bcmgenet_tdma_readl(priv, DMA_CTRL);
+ reg &= ~DMA_EN;
+ bcmgenet_tdma_writel(priv, reg, DMA_CTRL);
+
+ /* Check TDMA status register to confirm TDMA is disabled */
+ while (timeout++ < DMA_TIMEOUT_VAL) {
+ reg = bcmgenet_tdma_readl(priv, DMA_STATUS);
+ if (reg & DMA_DISABLED)
+ break;
+
+ udelay(1);
+ }
+
+ if (timeout == DMA_TIMEOUT_VAL) {
+ netdev_warn(priv->dev,
+ "Timed out while disabling TX DMA\n");
+ ret = -ETIMEDOUT;
+ }
+
+ /* Wait 10ms for packet drain in both tx and rx dma */
+ usleep_range(10000, 20000);
+
+ /* Disable RDMA */
+ reg = bcmgenet_rdma_readl(priv, DMA_CTRL);
+ reg &= ~DMA_EN;
+ bcmgenet_rdma_writel(priv, reg, DMA_CTRL);
+
+ timeout = 0;
+ /* Check RDMA status register to confirm RDMA is disabled */
+ while (timeout++ < DMA_TIMEOUT_VAL) {
+ reg = bcmgenet_rdma_readl(priv, DMA_STATUS);
+ if (reg & DMA_DISABLED)
+ break;
+
+ udelay(1);
+ }
+
+ if (timeout == DMA_TIMEOUT_VAL) {
+ netdev_warn(priv->dev,
+ "Timed out while disabling RX DMA\n");
+ ret = -ETIMEDOUT;
+ }
+
+ return ret;
+}
+
+static int bcmgenet_close(struct net_device *dev)
+{
+ struct bcmgenet_priv *priv = netdev_priv(dev);
+ int ret;
+ u32 reg;
+
+ netif_dbg(priv, ifdown, dev, "bcmgenet_close\n");
+
+ if (priv->phydev)
+ phy_stop(priv->phydev);
+
+ /* Disable MAC receive */
+ reg = bcmgenet_umac_readl(priv, UMAC_CMD);
+ reg &= ~CMD_RX_EN;
+ bcmgenet_umac_writel(priv, reg, UMAC_CMD);
+
+ netif_tx_stop_all_queues(dev);
+
+ ret = bcmgenet_dma_teardown(priv);
+ if (ret)
+ return ret;
+
+ /* Disable MAC transmit. TX DMA disabled have to done before this */
+ reg = bcmgenet_umac_readl(priv, UMAC_CMD);
+ reg &= ~CMD_TX_EN;
+ bcmgenet_umac_writel(priv, reg, UMAC_CMD);
+
+ napi_disable(&priv->napi);
+
+ /* tx reclaim */
+ bcmgenet_tx_reclaim_all(dev);
+ bcmgenet_fini_dma(priv);
+
+ free_irq(priv->irq0, priv);
+ free_irq(priv->irq1, priv);
+
+ /* Wait for pending work items to complete - we are stopping
+ * the clock now. Since interrupts are disabled, no new work
+ * will be scheduled.
+ */
+ cancel_work_sync(&priv->bcmgenet_irq_work);
+
+ if (phy_is_internal(priv->phydev))
+ bcmgenet_power_down(priv, GENET_POWER_PASSIVE);
+
+ if (priv->wol_enabled)
+ clk_enable(priv->clk_wol);
+
+ if (!IS_ERR(priv->clk))
+ clk_disable_unprepare(priv->clk);
+
+ return 0;
+}
+
+static void bcmgenet_timeout(struct net_device *dev)
+{
+ struct bcmgenet_priv *priv = netdev_priv(dev);
+
+ netif_dbg(priv, tx_err, dev, "bcmgenet_timeout\n");
+
+ dev->trans_start = jiffies;
+
+ dev->stats.tx_errors++;
+
+ netif_tx_wake_all_queues(dev);
+}
+
+#define MAX_MC_COUNT 16
+
+static inline void bcmgenet_set_mdf_addr(struct bcmgenet_priv *priv,
+ unsigned char *addr,
+ int *i,
+ int *mc)
+{
+ u32 reg;
+
+ bcmgenet_umac_writel(priv,
+ addr[0] << 8 | addr[1], UMAC_MDF_ADDR + (*i * 4));
+ bcmgenet_umac_writel(priv,
+ addr[2] << 24 | addr[3] << 16 |
+ addr[4] << 8 | addr[5],
+ UMAC_MDF_ADDR + ((*i + 1) * 4));
+ reg = bcmgenet_umac_readl(priv, UMAC_MDF_CTRL);
+ reg |= (1 << (MAX_MC_COUNT - *mc));
+ bcmgenet_umac_writel(priv, reg, UMAC_MDF_CTRL);
+ *i += 2;
+ (*mc)++;
+}
+
+static void bcmgenet_set_rx_mode(struct net_device *dev)
+{
+ struct bcmgenet_priv *priv = netdev_priv(dev);
+ struct netdev_hw_addr *ha;
+ int i, mc;
+ u32 reg;
+
+ netif_dbg(priv, hw, dev, "%s: %08X\n", __func__, dev->flags);
+
+ /* Promiscous mode */
+ reg = bcmgenet_umac_readl(priv, UMAC_CMD);
+ if (dev->flags & IFF_PROMISC) {
+ reg |= CMD_PROMISC;
+ bcmgenet_umac_writel(priv, reg, UMAC_CMD);
+ bcmgenet_umac_writel(priv, 0, UMAC_MDF_CTRL);
+ return;
+ } else {
+ reg &= ~CMD_PROMISC;
+ bcmgenet_umac_writel(priv, reg, UMAC_CMD);
+ }
+
+ /* UniMac doesn't support ALLMULTI */
+ if (dev->flags & IFF_ALLMULTI) {
+ netdev_warn(dev, "ALLMULTI is not supported\n");
+ return;
+ }
+
+ /* update MDF filter */
+ i = 0;
+ mc = 0;
+ /* Broadcast */
+ bcmgenet_set_mdf_addr(priv, dev->broadcast, &i, &mc);
+ /* my own address.*/
+ bcmgenet_set_mdf_addr(priv, dev->dev_addr, &i, &mc);
+ /* Unicast list*/
+ if (netdev_uc_count(dev) > (MAX_MC_COUNT - mc))
+ return;
+
+ if (!netdev_uc_empty(dev))
+ netdev_for_each_uc_addr(ha, dev)
+ bcmgenet_set_mdf_addr(priv, ha->addr, &i, &mc);
+ /* Multicast */
+ if (netdev_mc_empty(dev) || netdev_mc_count(dev) >= (MAX_MC_COUNT - mc))
+ return;
+
+ netdev_for_each_mc_addr(ha, dev)
+ bcmgenet_set_mdf_addr(priv, ha->addr, &i, &mc);
+}
+
+/* Set the hardware MAC address. */
+static int bcmgenet_set_mac_addr(struct net_device *dev, void *p)
+{
+ struct sockaddr *addr = p;
+
+ /* Setting the MAC address at the hardware level is not possible
+ * without disabling the UniMAC RX/TX enable bits.
+ */
+ if (netif_running(dev))
+ return -EBUSY;
+
+ ether_addr_copy(dev->dev_addr, addr->sa_data);
+
+ return 0;
+}
+
+static u16 bcmgenet_select_queue(struct net_device *dev,
+ struct sk_buff *skb, void *accel_priv)
+{
+ return netif_is_multiqueue(dev) ? skb->queue_mapping : 0;
+}
+
+static const struct net_device_ops bcmgenet_netdev_ops = {
+ .ndo_open = bcmgenet_open,
+ .ndo_stop = bcmgenet_close,
+ .ndo_start_xmit = bcmgenet_xmit,
+ .ndo_select_queue = bcmgenet_select_queue,
+ .ndo_tx_timeout = bcmgenet_timeout,
+ .ndo_set_rx_mode = bcmgenet_set_rx_mode,
+ .ndo_set_mac_address = bcmgenet_set_mac_addr,
+ .ndo_do_ioctl = bcmgenet_ioctl,
+ .ndo_set_features = bcmgenet_set_features,
+};
+
+/* Array of GENET hardware parameters/characteristics */
+static struct bcmgenet_hw_params bcmgenet_hw_params[] = {
+ [GENET_V1] = {
+ .tx_queues = 0,
+ .rx_queues = 0,
+ .bds_cnt = 0,
+ .bp_in_en_shift = 16,
+ .bp_in_mask = 0xffff,
+ .hfb_filter_cnt = 16,
+ .qtag_mask = 0x1F,
+ .hfb_offset = 0x1000,
+ .rdma_offset = 0x2000,
+ .tdma_offset = 0x3000,
+ .words_per_bd = 2,
+ },
+ [GENET_V2] = {
+ .tx_queues = 4,
+ .rx_queues = 4,
+ .bds_cnt = 32,
+ .bp_in_en_shift = 16,
+ .bp_in_mask = 0xffff,
+ .hfb_filter_cnt = 16,
+ .qtag_mask = 0x1F,
+ .tbuf_offset = 0x0600,
+ .hfb_offset = 0x1000,
+ .hfb_reg_offset = 0x2000,
+ .rdma_offset = 0x3000,
+ .tdma_offset = 0x4000,
+ .words_per_bd = 2,
+ .flags = GENET_HAS_EXT,
+ },
+ [GENET_V3] = {
+ .tx_queues = 4,
+ .rx_queues = 4,
+ .bds_cnt = 32,
+ .bp_in_en_shift = 17,
+ .bp_in_mask = 0x1ffff,
+ .hfb_filter_cnt = 48,
+ .qtag_mask = 0x3F,
+ .tbuf_offset = 0x0600,
+ .hfb_offset = 0x8000,
+ .hfb_reg_offset = 0xfc00,
+ .rdma_offset = 0x10000,
+ .tdma_offset = 0x11000,
+ .words_per_bd = 2,
+ .flags = GENET_HAS_EXT | GENET_HAS_MDIO_INTR,
+ },
+ [GENET_V4] = {
+ .tx_queues = 4,
+ .rx_queues = 4,
+ .bds_cnt = 32,
+ .bp_in_en_shift = 17,
+ .bp_in_mask = 0x1ffff,
+ .hfb_filter_cnt = 48,
+ .qtag_mask = 0x3F,
+ .tbuf_offset = 0x0600,
+ .hfb_offset = 0x8000,
+ .hfb_reg_offset = 0xfc00,
+ .rdma_offset = 0x2000,
+ .tdma_offset = 0x4000,
+ .words_per_bd = 3,
+ .flags = GENET_HAS_40BITS | GENET_HAS_EXT | GENET_HAS_MDIO_INTR,
+ },
+};
+
+/* Infer hardware parameters from the detected GENET version */
+static void bcmgenet_set_hw_params(struct bcmgenet_priv *priv)
+{
+ struct bcmgenet_hw_params *params;
+ u32 reg;
+ u8 major;
+
+ if (GENET_IS_V4(priv)) {
+ bcmgenet_dma_regs = bcmgenet_dma_regs_v3plus;
+ genet_dma_ring_regs = genet_dma_ring_regs_v4;
+ priv->dma_rx_chk_bit = DMA_RX_CHK_V3PLUS;
+ priv->version = GENET_V4;
+ } else if (GENET_IS_V3(priv)) {
+ bcmgenet_dma_regs = bcmgenet_dma_regs_v3plus;
+ genet_dma_ring_regs = genet_dma_ring_regs_v123;
+ priv->dma_rx_chk_bit = DMA_RX_CHK_V3PLUS;
+ priv->version = GENET_V3;
+ } else if (GENET_IS_V2(priv)) {
+ bcmgenet_dma_regs = bcmgenet_dma_regs_v2;
+ genet_dma_ring_regs = genet_dma_ring_regs_v123;
+ priv->dma_rx_chk_bit = DMA_RX_CHK_V12;
+ priv->version = GENET_V2;
+ } else if (GENET_IS_V1(priv)) {
+ bcmgenet_dma_regs = bcmgenet_dma_regs_v1;
+ genet_dma_ring_regs = genet_dma_ring_regs_v123;
+ priv->dma_rx_chk_bit = DMA_RX_CHK_V12;
+ priv->version = GENET_V1;
+ }
+
+ /* enum genet_version starts at 1 */
+ priv->hw_params = &bcmgenet_hw_params[priv->version];
+ params = priv->hw_params;
+
+ /* Read GENET HW version */
+ reg = bcmgenet_sys_readl(priv, SYS_REV_CTRL);
+ major = (reg >> 24 & 0x0f);
+ if (major == 5)
+ major = 4;
+ else if (major == 0)
+ major = 1;
+ if (major != priv->version) {
+ dev_err(&priv->pdev->dev,
+ "GENET version mismatch, got: %d, configured for: %d\n",
+ major, priv->version);
+ }
+
+ /* Print the GENET core version */
+ dev_info(&priv->pdev->dev, "GENET " GENET_VER_FMT,
+ major, (reg >> 16) & 0x0f, reg & 0xffff);
+
+#ifdef CONFIG_PHYS_ADDR_T_64BIT
+ if (!(params->flags & GENET_HAS_40BITS))
+ pr_warn("GENET does not support 40-bits PA\n");
+#endif
+
+ pr_debug("Configuration for version: %d\n"
+ "TXq: %1d, RXq: %1d, BDs: %1d\n"
+ "BP << en: %2d, BP msk: 0x%05x\n"
+ "HFB count: %2d, QTAQ msk: 0x%05x\n"
+ "TBUF: 0x%04x, HFB: 0x%04x, HFBreg: 0x%04x\n"
+ "RDMA: 0x%05x, TDMA: 0x%05x\n"
+ "Words/BD: %d\n",
+ priv->version,
+ params->tx_queues, params->rx_queues, params->bds_cnt,
+ params->bp_in_en_shift, params->bp_in_mask,
+ params->hfb_filter_cnt, params->qtag_mask,
+ params->tbuf_offset, params->hfb_offset,
+ params->hfb_reg_offset,
+ params->rdma_offset, params->tdma_offset,
+ params->words_per_bd);
+}
+
+static const struct of_device_id bcmgenet_match[] = {
+ { .compatible = "brcm,genet-v1", .data = (void *)GENET_V1 },
+ { .compatible = "brcm,genet-v2", .data = (void *)GENET_V2 },
+ { .compatible = "brcm,genet-v3", .data = (void *)GENET_V3 },
+ { .compatible = "brcm,genet-v4", .data = (void *)GENET_V4 },
+ { },
+};
+
+static int bcmgenet_probe(struct platform_device *pdev)
+{
+ struct device_node *dn = pdev->dev.of_node;
+ const struct of_device_id *of_id;
+ struct bcmgenet_priv *priv;
+ struct net_device *dev;
+ const void *macaddr;
+ struct resource *r;
+ int err = -EIO;
+
+ /* Up to GENET_MAX_MQ_CNT + 1 TX queues and a single RX queue */
+ dev = alloc_etherdev_mqs(sizeof(*priv), GENET_MAX_MQ_CNT + 1, 1);
+ if (!dev) {
+ dev_err(&pdev->dev, "can't allocate net device\n");
+ return -ENOMEM;
+ }
+
+ of_id = of_match_node(bcmgenet_match, dn);
+ if (!of_id)
+ return -EINVAL;
+
+ priv = netdev_priv(dev);
+ priv->irq0 = platform_get_irq(pdev, 0);
+ priv->irq1 = platform_get_irq(pdev, 1);
+ if (!priv->irq0 || !priv->irq1) {
+ dev_err(&pdev->dev, "can't find IRQs\n");
+ err = -EINVAL;
+ goto err;
+ }
+
+ macaddr = of_get_mac_address(dn);
+ if (!macaddr) {
+ dev_err(&pdev->dev, "can't find MAC address\n");
+ err = -EINVAL;
+ goto err;
+ }
+
+ r = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+ priv->base = devm_request_and_ioremap(&pdev->dev, r);
+ if (!priv->base) {
+ dev_err(&pdev->dev, "can't ioremap\n");
+ err = -EINVAL;
+ goto err;
+ }
+
+ SET_NETDEV_DEV(dev, &pdev->dev);
+ dev_set_drvdata(&pdev->dev, dev);
+ ether_addr_copy(dev->dev_addr, macaddr);
+ dev->watchdog_timeo = 2 * HZ;
+ SET_ETHTOOL_OPS(dev, &bcmgenet_ethtool_ops);
+ dev->netdev_ops = &bcmgenet_netdev_ops;
+ netif_napi_add(dev, &priv->napi, bcmgenet_poll, 64);
+
+ priv->msg_enable = netif_msg_init(-1, GENET_MSG_DEFAULT);
+
+ /* Set hardware features */
+ dev->hw_features |= NETIF_F_SG | NETIF_F_IP_CSUM |
+ NETIF_F_IPV6_CSUM | NETIF_F_RXCSUM;
+
+ /* Set the needed headroom to account for any possible
+ * features enabling/disabling at runtime
+ */
+ dev->needed_headroom += 64;
+
+ netdev_boot_setup_check(dev);
+
+ priv->dev = dev;
+ priv->pdev = pdev;
+ priv->version = (enum bcmgenet_version)of_id->data;
+
+ bcmgenet_set_hw_params(priv);
+
+ spin_lock_init(&priv->lock);
+ /* Mii wait queue */
+ init_waitqueue_head(&priv->wq);
+ /* Always use RX_BUF_LENGTH (2KB) buffer for all chips */
+ priv->rx_buf_len = RX_BUF_LENGTH;
+ INIT_WORK(&priv->bcmgenet_irq_work, bcmgenet_irq_task);
+
+ priv->clk = devm_clk_get(&priv->pdev->dev, "enet");
+ if (IS_ERR(priv->clk))
+ dev_warn(&priv->pdev->dev, "failed to get enet clock\n");
+
+ priv->clk_wol = devm_clk_get(&priv->pdev->dev, "enet-wol");
+ if (IS_ERR(priv->clk_wol))
+ dev_warn(&priv->pdev->dev, "failed to get enet-wol clock\n");
+
+ if (!IS_ERR(priv->clk))
+ clk_prepare_enable(priv->clk);
+
+ err = reset_umac(priv);
+ if (err)
+ goto err_clk_disable;
+
+ err = bcmgenet_mii_init(dev);
+ if (err)
+ goto err_clk_disable;
+
+ /* setup number of real queues + 1 (GENET_V1 has 0 hardware queues
+ * just the ring 16 descriptor based TX
+ */
+ netif_set_real_num_tx_queues(priv->dev, priv->hw_params->tx_queues + 1);
+ netif_set_real_num_rx_queues(priv->dev, priv->hw_params->rx_queues + 1);
+
+ err = register_netdev(dev);
+ if (err)
+ goto err_clk_disable;
+
+ /* Turn off the main clock, WOL clock is handled separately */
+ if (!IS_ERR(priv->clk))
+ clk_disable_unprepare(priv->clk);
+
+ return err;
+
+err_clk_disable:
+ if (!IS_ERR(priv->clk))
+ clk_disable_unprepare(priv->clk);
+err:
+ free_netdev(dev);
+ return err;
+}
+
+static int bcmgenet_remove(struct platform_device *pdev)
+{
+ struct bcmgenet_priv *priv = dev_to_priv(&pdev->dev);
+
+ dev_set_drvdata(&pdev->dev, NULL);
+ unregister_netdev(priv->dev);
+ bcmgenet_mii_exit(priv->dev);
+ free_netdev(priv->dev);
+
+ return 0;
+}
+
+
+static struct platform_driver bcmgenet_driver = {
+ .probe = bcmgenet_probe,
+ .remove = bcmgenet_remove,
+ .driver = {
+ .name = "bcmgenet",
+ .owner = THIS_MODULE,
+ .of_match_table = bcmgenet_match,
+ },
+};
+module_platform_driver(bcmgenet_driver);
+
+MODULE_AUTHOR("Broadcom Corporation");
+MODULE_DESCRIPTION("Broadcom GENET Ethernet controller driver");
+MODULE_ALIAS("platform:bcmgenet");
+MODULE_LICENSE("GPL");
--
1.8.3.2
--
To unsubscribe from this list: send the line "unsubscribe devicetree" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
^ permalink raw reply related
* [PATCH net-next v3 04/10] net: phy: add Broadcom BCM7xxx internal PHY driver
From: Florian Fainelli @ 2014-02-13 22:53 UTC (permalink / raw)
To: netdev-u79uwXL29TY76Z2rM5mHXA
Cc: davem-fT/PcQaiUtIeIZ0/mPfg9Q, devicetree-u79uwXL29TY76Z2rM5mHXA,
cernekee-Re5JQEeQqe8AvxtiuMwx3w, mark.rutland-5wv7dgnIgG8,
romieu-W8zweXLXuWQS+FvcfC7Uqw, Florian Fainelli
In-Reply-To: <1392332029-24669-1-git-send-email-f.fainelli-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
This patch adds support for the Broadcom BCM7xxx Set Top Box SoCs
internal PHYs. This driver supports the following generation of SoCs:
- BCM7366, BCM7439, BCM7445 (28nm process)
- all 40nm and 65nm (older MIPS-based SoCs)
The PHYs on these SoCs require a bunch of workarounds to operate
correctly, both during configuration time and at suspend/resume time,
the driver handles that for us.
Signed-off-by: Florian Fainelli <f.fainelli-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
---
Changes in v3:
- use register definitions from brcmphy.h where applicable
- add new register definitions for 40nm workaround
- check return code of phy_set_clr_bits
- minor stylistic changes as reported by Francois Romieu
Changes in v2:
- update the 28nm AFE workaround to clear the EEE LPI
timer (value 0xeb17 previously now 0xeb19)
drivers/net/phy/Kconfig | 6 +
drivers/net/phy/Makefile | 1 +
drivers/net/phy/bcm7xxx.c | 343 ++++++++++++++++++++++++++++++++++++++++++++++
include/linux/brcmphy.h | 9 ++
4 files changed, 359 insertions(+)
create mode 100644 drivers/net/phy/bcm7xxx.c
diff --git a/drivers/net/phy/Kconfig b/drivers/net/phy/Kconfig
index 9b5d46c..6a17f92 100644
--- a/drivers/net/phy/Kconfig
+++ b/drivers/net/phy/Kconfig
@@ -71,6 +71,12 @@ config BCM63XX_PHY
---help---
Currently supports the 6348 and 6358 PHYs.
+config BCM7XXX_PHY
+ tristate "Drivers for Broadcom 7xxx SOCs internal PHYs"
+ ---help---
+ Currently supports the BCM7366, BCM7439, BCM7445, and
+ 40nm and 65nm generation of BCM7xxx Set Top Box SoCs.
+
config BCM87XX_PHY
tristate "Driver for Broadcom BCM8706 and BCM8727 PHYs"
help
diff --git a/drivers/net/phy/Makefile b/drivers/net/phy/Makefile
index 9013dfa..07d2402 100644
--- a/drivers/net/phy/Makefile
+++ b/drivers/net/phy/Makefile
@@ -12,6 +12,7 @@ obj-$(CONFIG_SMSC_PHY) += smsc.o
obj-$(CONFIG_VITESSE_PHY) += vitesse.o
obj-$(CONFIG_BROADCOM_PHY) += broadcom.o
obj-$(CONFIG_BCM63XX_PHY) += bcm63xx.o
+obj-$(CONFIG_BCM7XXX_PHY) += bcm7xxx.o
obj-$(CONFIG_BCM87XX_PHY) += bcm87xx.o
obj-$(CONFIG_ICPLUS_PHY) += icplus.o
obj-$(CONFIG_REALTEK_PHY) += realtek.o
diff --git a/drivers/net/phy/bcm7xxx.c b/drivers/net/phy/bcm7xxx.c
new file mode 100644
index 0000000..6973372
--- /dev/null
+++ b/drivers/net/phy/bcm7xxx.c
@@ -0,0 +1,343 @@
+/*
+ * Broadcom BCM7xxx internal transceivers support.
+ *
+ * Copyright (C) 2014, Broadcom Corporation
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; either version
+ * 2 of the License, or (at your option) any later version.
+ */
+
+#include <linux/module.h>
+#include <linux/phy.h>
+#include <linux/delay.h>
+#include <linux/bitops.h>
+#include <linux/brcmphy.h>
+
+/* Broadcom BCM7xxx internal PHY registers */
+#define MII_BCM7XXX_CHANNEL_WIDTH 0x2000
+
+/* 40nm only register definitions */
+#define MII_BCM7XXX_100TX_AUX_CTL 0x10
+#define MII_BCM7XXX_100TX_FALSE_CAR 0x13
+#define MII_BCM7XXX_100TX_DISC 0x14
+#define MII_BCM7XXX_AUX_MODE 0x1d
+#define MII_BCM7XX_64CLK_MDIO BIT(12)
+#define MII_BCM7XXX_CORE_BASE1E 0x1e
+#define MII_BCM7XXX_TEST 0x1f
+#define MII_BCM7XXX_SHD_MODE_2 BIT(2)
+
+static int bcm7445_config_init(struct phy_device *phydev)
+{
+ int ret;
+ const struct bcm7445_regs {
+ int reg;
+ u16 value;
+ } bcm7445_regs_cfg[] = {
+ /* increases ADC latency by 24ns */
+ { MII_BCM54XX_EXP_SEL, 0x0038 },
+ { MII_BCM54XX_EXP_DATA, 0xAB95 },
+ /* increases internal 1V LDO voltage by 5% */
+ { MII_BCM54XX_EXP_SEL, 0x2038 },
+ { MII_BCM54XX_EXP_DATA, 0xBB22 },
+ /* reduce RX low pass filter corner frequency */
+ { MII_BCM54XX_EXP_SEL, 0x6038 },
+ { MII_BCM54XX_EXP_DATA, 0xFFC5 },
+ /* reduce RX high pass filter corner frequency */
+ { MII_BCM54XX_EXP_SEL, 0x003a },
+ { MII_BCM54XX_EXP_DATA, 0x2002 },
+ };
+ unsigned int i;
+
+ for (i = 0; i < ARRAY_SIZE(bcm7445_regs_cfg); i++) {
+ ret = phy_write(phydev,
+ bcm7445_regs_cfg[i].reg,
+ bcm7445_regs_cfg[i].value);
+ if (ret)
+ return ret;
+ }
+
+ return 0;
+}
+
+static void phy_write_exp(struct phy_device *phydev,
+ u16 reg, u16 value)
+{
+ phy_write(phydev, MII_BCM54XX_EXP_SEL, MII_BCM54XX_EXP_SEL_ER | reg);
+ phy_write(phydev, MII_BCM54XX_EXP_DATA, value);
+}
+
+static void phy_write_misc(struct phy_device *phydev,
+ u16 reg, u16 chl, u16 value)
+{
+ int tmp;
+
+ phy_write(phydev, MII_BCM54XX_AUX_CTL, MII_BCM54XX_AUXCTL_SHDWSEL_MISC);
+
+ tmp = phy_read(phydev, MII_BCM54XX_AUX_CTL);
+ tmp |= MII_BCM54XX_AUXCTL_ACTL_SMDSP_ENA;
+ phy_write(phydev, MII_BCM54XX_AUX_CTL, tmp);
+
+ tmp = (chl * MII_BCM7XXX_CHANNEL_WIDTH) | reg;
+ phy_write(phydev, MII_BCM54XX_EXP_SEL, tmp);
+
+ phy_write(phydev, MII_BCM54XX_EXP_DATA, value);
+}
+
+static int bcm7xxx_28nm_afe_config_init(struct phy_device *phydev)
+{
+ /* write AFE_RXCONFIG_0 */
+ phy_write_misc(phydev, 0x38, 0x0000, 0xeb19);
+
+ /* write AFE_RXCONFIG_1 */
+ phy_write_misc(phydev, 0x38, 0x0001, 0x9a3f);
+
+ /* write AFE_RX_LP_COUNTER */
+ phy_write_misc(phydev, 0x38, 0x0003, 0x7fc7);
+
+ /* write AFE_HPF_TRIM_OTHERS */
+ phy_write_misc(phydev, 0x3A, 0x0000, 0x000b);
+
+ /* write AFTE_TX_CONFIG */
+ phy_write_misc(phydev, 0x39, 0x0000, 0x0800);
+
+ /* Increase VCO range to prevent unlocking problem of PLL at low
+ * temp
+ */
+ phy_write_misc(phydev, 0x0032, 0x0001, 0x0048);
+
+ /* Change Ki to 011 */
+ phy_write_misc(phydev, 0x0032, 0x0002, 0x021b);
+
+ /* Disable loading of TVCO buffer to bandgap, set bandgap trim
+ * to 111
+ */
+ phy_write_misc(phydev, 0x0033, 0x0000, 0x0e20);
+
+ /* Adjust bias current trim by -3 */
+ phy_write_misc(phydev, 0x000a, 0x0000, 0x690b);
+
+ /* Switch to CORE_BASE1E */
+ phy_write(phydev, MII_BCM7XXX_CORE_BASE1E, 0xd);
+
+ /* Reset R_CAL/RC_CAL Engine */
+ phy_write_exp(phydev, 0x00b0, 0x0010);
+
+ /* Disable Reset R_CAL/RC_CAL Engine */
+ phy_write_exp(phydev, 0x00b0, 0x0000);
+
+ return 0;
+}
+
+static int bcm7xxx_28nm_config_init(struct phy_device *phydev)
+{
+ int ret;
+
+ ret = bcm7445_config_init(phydev);
+ if (ret)
+ return ret;
+
+ return bcm7xxx_28nm_afe_config_init(phydev);
+}
+
+static int phy_set_clr_bits(struct phy_device *dev, int location,
+ int set_mask, int clr_mask)
+{
+ int v, ret;
+
+ v = phy_read(dev, location);
+ if (v < 0)
+ return v;
+
+ v &= ~clr_mask;
+ v |= set_mask;
+
+ ret = phy_write(dev, location, v);
+ if (ret < 0)
+ return ret;
+
+ return v;
+}
+
+static int bcm7xxx_config_init(struct phy_device *phydev)
+{
+ int ret;
+
+ /* Enable 64 clock MDIO */
+ phy_write(phydev, MII_BCM7XXX_AUX_MODE, MII_BCM7XX_64CLK_MDIO);
+ phy_read(phydev, MII_BCM7XXX_AUX_MODE);
+
+ /* Workaround only required for 100Mbits/sec */
+ if (!(phydev->dev_flags & PHY_BRCM_100MBPS_WAR))
+ return 0;
+
+ /* set shadow mode 2 */
+ ret = phy_set_clr_bits(phydev, MII_BCM7XXX_TEST,
+ MII_BCM7XXX_SHD_MODE_2, MII_BCM7XXX_SHD_MODE_2);
+ if (ret < 0)
+ return ret;
+
+ /* set iddq_clkbias */
+ phy_write(phydev, MII_BCM7XXX_100TX_DISC, 0x0F00);
+ udelay(10);
+
+ /* reset iddq_clkbias */
+ phy_write(phydev, MII_BCM7XXX_100TX_DISC, 0x0C00);
+
+ phy_write(phydev, MII_BCM7XXX_100TX_FALSE_CAR, 0x7555);
+
+ /* reset shadow mode 2 */
+ ret = phy_set_clr_bits(phydev, MII_BCM7XXX_TEST, MII_BCM7XXX_SHD_MODE_2, 0);
+ if (ret < 0)
+ return ret;
+
+ return 0;
+}
+
+/* Workaround for putting the PHY in IDDQ mode, required
+ * for all BCM7XXX PHYs
+ */
+static int bcm7xxx_suspend(struct phy_device *phydev)
+{
+ int ret;
+ const struct bcm7xxx_regs {
+ int reg;
+ u16 value;
+ } bcm7xxx_suspend_cfg[] = {
+ { MII_BCM7XXX_TEST, 0x008b },
+ { MII_BCM7XXX_100TX_AUX_CTL, 0x01c0 },
+ { MII_BCM7XXX_100TX_DISC, 0x7000 },
+ { MII_BCM7XXX_TEST, 0x000f },
+ { MII_BCM7XXX_100TX_AUX_CTL, 0x20d0 },
+ { MII_BCM7XXX_TEST, 0x000b },
+ };
+ unsigned int i;
+
+ for (i = 0; i < ARRAY_SIZE(bcm7xxx_suspend_cfg); i++) {
+ ret = phy_write(phydev,
+ bcm7xxx_suspend_cfg[i].reg,
+ bcm7xxx_suspend_cfg[i].value);
+ if (ret)
+ return ret;
+ }
+
+ return 0;
+}
+
+static int bcm7xxx_dummy_config_init(struct phy_device *phydev)
+{
+ return 0;
+}
+
+static struct phy_driver bcm7xxx_driver[] = {
+{
+ .phy_id = PHY_ID_BCM7366,
+ .phy_id_mask = 0xfffffff0,
+ .name = "Broadcom BCM7366",
+ .features = PHY_GBIT_FEATURES |
+ SUPPORTED_Pause | SUPPORTED_Asym_Pause,
+ .flags = PHY_IS_INTERNAL,
+ .config_init = bcm7xxx_28nm_afe_config_init,
+ .config_aneg = genphy_config_aneg,
+ .read_status = genphy_read_status,
+ .suspend = bcm7xxx_suspend,
+ .resume = bcm7xxx_28nm_afe_config_init,
+ .driver = { .owner = THIS_MODULE },
+}, {
+ .phy_id = PHY_ID_BCM7439,
+ .phy_id_mask = 0xfffffff0,
+ .name = "Broadcom BCM7439",
+ .features = PHY_GBIT_FEATURES |
+ SUPPORTED_Pause | SUPPORTED_Asym_Pause,
+ .flags = PHY_IS_INTERNAL,
+ .config_init = bcm7xxx_28nm_afe_config_init,
+ .config_aneg = genphy_config_aneg,
+ .read_status = genphy_read_status,
+ .suspend = bcm7xxx_suspend,
+ .resume = bcm7xxx_28nm_afe_config_init,
+ .driver = { .owner = THIS_MODULE },
+}, {
+ .phy_id = PHY_ID_BCM7445,
+ .phy_id_mask = 0xfffffff0,
+ .name = "Broadcom BCM7445",
+ .features = PHY_GBIT_FEATURES |
+ SUPPORTED_Pause | SUPPORTED_Asym_Pause,
+ .flags = PHY_IS_INTERNAL,
+ .config_init = bcm7xxx_28nm_config_init,
+ .config_aneg = genphy_config_aneg,
+ .read_status = genphy_read_status,
+ .suspend = bcm7xxx_suspend,
+ .resume = bcm7xxx_28nm_config_init,
+ .driver = { .owner = THIS_MODULE },
+}, {
+ .name = "Broadcom BCM7XXX 28nm",
+ .phy_id = PHY_ID_BCM7XXX_28,
+ .phy_id_mask = PHY_BCM_OUI_MASK,
+ .features = PHY_GBIT_FEATURES |
+ SUPPORTED_Pause | SUPPORTED_Asym_Pause,
+ .flags = PHY_IS_INTERNAL,
+ .config_init = bcm7xxx_28nm_config_init,
+ .config_aneg = genphy_config_aneg,
+ .read_status = genphy_read_status,
+ .suspend = bcm7xxx_suspend,
+ .resume = bcm7xxx_28nm_config_init,
+ .driver = { .owner = THIS_MODULE },
+}, {
+ .phy_id = PHY_BCM_OUI_4,
+ .phy_id_mask = 0xffff0000,
+ .name = "Broadcom BCM7XXX 40nm",
+ .features = PHY_GBIT_FEATURES |
+ SUPPORTED_Pause | SUPPORTED_Asym_Pause,
+ .flags = PHY_IS_INTERNAL,
+ .config_init = bcm7xxx_config_init,
+ .config_aneg = genphy_config_aneg,
+ .read_status = genphy_read_status,
+ .suspend = bcm7xxx_suspend,
+ .resume = bcm7xxx_config_init,
+ .driver = { .owner = THIS_MODULE },
+}, {
+ .phy_id = PHY_BCM_OUI_5,
+ .phy_id_mask = 0xffffff00,
+ .name = "Broadcom BCM7XXX 65nm",
+ .features = PHY_BASIC_FEATURES |
+ SUPPORTED_Pause | SUPPORTED_Asym_Pause,
+ .flags = PHY_IS_INTERNAL,
+ .config_init = bcm7xxx_dummy_config_init,
+ .config_aneg = genphy_config_aneg,
+ .read_status = genphy_read_status,
+ .suspend = bcm7xxx_suspend,
+ .resume = bcm7xxx_config_init,
+ .driver = { .owner = THIS_MODULE },
+} };
+
+static struct mdio_device_id __maybe_unused bcm7xxx_tbl[] = {
+ { PHY_ID_BCM7366, 0xfffffff0, },
+ { PHY_ID_BCM7439, 0xfffffff0, },
+ { PHY_ID_BCM7445, 0xfffffff0, },
+ { PHY_ID_BCM7XXX_28, 0xfffffc00 },
+ { PHY_BCM_OUI_4, 0xffff0000 },
+ { PHY_BCM_OUI_5, 0xffffff00 },
+ { }
+};
+
+static int __init bcm7xxx_phy_init(void)
+{
+ return phy_drivers_register(bcm7xxx_driver,
+ ARRAY_SIZE(bcm7xxx_driver));
+}
+
+static void __exit bcm7xxx_phy_exit(void)
+{
+ phy_drivers_unregister(bcm7xxx_driver,
+ ARRAY_SIZE(bcm7xxx_driver));
+}
+
+module_init(bcm7xxx_phy_init);
+module_exit(bcm7xxx_phy_exit);
+
+MODULE_DEVICE_TABLE(mdio, bcm7xxx_tbl);
+
+MODULE_DESCRIPTION("Broadcom BCM7xxx internal PHY driver");
+MODULE_LICENSE("GPL");
+MODULE_AUTHOR("Broadcom Corporation");
diff --git a/include/linux/brcmphy.h b/include/linux/brcmphy.h
index 104e3ef..6f76277 100644
--- a/include/linux/brcmphy.h
+++ b/include/linux/brcmphy.h
@@ -13,10 +13,17 @@
#define PHY_ID_BCM5461 0x002060c0
#define PHY_ID_BCM57780 0x03625d90
+#define PHY_ID_BCM7366 0x600d8490
+#define PHY_ID_BCM7439 0x600d8480
+#define PHY_ID_BCM7445 0x600d8510
+#define PHY_ID_BCM7XXX_28 0x600d8400
+
#define PHY_BCM_OUI_MASK 0xfffffc00
#define PHY_BCM_OUI_1 0x00206000
#define PHY_BCM_OUI_2 0x0143bc00
#define PHY_BCM_OUI_3 0x03625c00
+#define PHY_BCM_OUI_4 0x600d0000
+#define PHY_BCM_OUI_5 0x03625e00
#define PHY_BCM_FLAGS_MODE_COPPER 0x00000001
@@ -31,6 +38,8 @@
#define PHY_BRCM_EXT_IBND_TX_ENABLE 0x00002000
#define PHY_BRCM_CLEAR_RGMII_MODE 0x00004000
#define PHY_BRCM_DIS_TXCRXC_NOENRGY 0x00008000
+/* Broadcom BCM7xxx specific workarounds */
+#define PHY_BRCM_100MBPS_WAR 0x00010000
#define PHY_BCM_FLAGS_VALID 0x80000000
/* Broadcom BCM54XX register definitions, common to most Broadcom PHYs */
--
1.8.3.2
--
To unsubscribe from this list: send the line "unsubscribe devicetree" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
^ permalink raw reply related
* Re: [PATCH 2/3] net: UDP gro_receive accept csum=0
From: Or Gerlitz @ 2014-02-13 22:50 UTC (permalink / raw)
To: Tom Herbert, Joseph Gasparakis
Cc: Or Gerlitz, David Miller, Linux Netdev List, Jerry Chu,
Eric Dumazet
In-Reply-To: <CA+mtBx_+fc5YdQE+M49QD5ymuD7bfDdedfi2AuCDhU2=CawN6Q@mail.gmail.com>
On Fri, Feb 14, 2014 at 12:27 AM, Tom Herbert <therbert@google.com> wrote:
>> [...] this is according to the conventions set by
>> 0afb166 vxlan: Add capability of Rx checksum offload for inner packet
>> 6a674e9 net: Add support for hardware-offloaded encapsulation
>> B/c after finalizing the GRO work and decapsulating, skb injected up into
>> the TCP stack with ip_summed equals to CHECKSUM_NONE are rejected. If >> this assumption is wrong, maybe we can remove testing the ip_summed field
>> here altogether?
> If I'm interpreting the semantics correctly, when skb->encapsulation
> is set the ip_summed is valid for both the inner and outer header
> (e.g. CHECKSUM_COMPLETE is always assumed okay for both layers). If
> skb->encapsulation is not set then ip_summed is only valid for outer header.
Yep, I think this would be correct interpertation, Joseph, agree?
> So then the patch is broken in the case that encap is not set,
> ip_summed is CHECKSUM_UNNECESSARY, csum == 0, and we need to
> validate the inner checksum.
Just to make sure, by "the patch" you refer to your patch or the current code?
> But even worse, is there a fundamental issue where udp4_csum_init is able
> to change ip_summed to be CHECKSUM_UNNECESSARY (either check == 0
> or ip_summed == CHECKSUM_UNNECESSARY) regardless of
> skb->encapsulation, sending the packet into encap_rcv which could
> ultimately incorrectly apply ip_summed on the inner TCP/UDP packet?
By fundamental you mean performance issue or functionality issue (bug) or both?
^ permalink raw reply
* Re: [PATCH net-next] tipc: explicitly include core.h in addr.h
From: David Miller @ 2014-02-13 22:49 UTC (permalink / raw)
To: andreas.bofjall; +Cc: netdev, tipc-discussion, maloy
In-Reply-To: <1392194058-16271-1-git-send-email-andreas.bofjall@ericsson.com>
From: <andreas.bofjall@ericsson.com>
Date: Wed, 12 Feb 2014 09:34:18 +0100
> From: Andreas Bofjäll <andreas.bofjall@ericsson.com>
>
> The inline functions in addr.h uses tipc_own_addr which is exported by
> core.h, but addr.h never actually includes it. It works because it is
> explicitly included where this is used, but it looks a bit strange.
>
> Include core.h in addr.h explicitly to make the dependency clearer.
>
> Signed-off-by: Andreas Bofjäll <andreas.bofjall@ericsson.com>
> Reviewed-by: Jon Maloy <jon.maloy@ericsson.com>
Applied, thank you.
^ permalink raw reply
* Re: [PATCH net-next 0/2] bonding: remove the redundant judgements for bonding
From: David Miller @ 2014-02-13 22:49 UTC (permalink / raw)
To: dingtianhong; +Cc: fubar, vfalico, andy, netdev
In-Reply-To: <1392188330-17208-1-git-send-email-dingtianhong@huawei.com>
From: Ding Tianhong <dingtianhong@huawei.com>
Date: Wed, 12 Feb 2014 14:58:48 +0800
> Remove the redundant judgements for bond_set_mac_address() and
> bond_option_queue_id_set().
Series applied, thanks.
^ permalink raw reply
* Re: [RFC PATCH v2 tip 0/7] 64-bit BPF insn set and tracing filters
From: H. Peter Anvin @ 2014-02-13 22:47 UTC (permalink / raw)
To: Daniel Borkmann
Cc: Alexei Starovoitov, Ingo Molnar, David S. Miller, Steven Rostedt,
Peter Zijlstra, Thomas Gleixner, Masami Hiramatsu, Tom Zanussi,
Jovi Zhangwei, Eric Dumazet, Linus Torvalds, Andrew Morton,
Frederic Weisbecker, Arnaldo Carvalho de Melo, Pekka Enberg,
Arjan van de Ven, Christoph Hellwig, linux-kernel, netdev
In-Reply-To: <52FD4AC0.7090502@redhat.com>
On 02/13/2014 02:44 PM, Daniel Borkmann wrote:
>
> Well, if that would be the case, then seccomp would have had JIT support
> long ago. ;-) Right now BPF filters with seccomp are not JIT compiled
> for _any_ architecture.
>
Really, I was under the impression there were. They *should be*, that
was an important concept in the development of the seccomp filters.
-hpa
^ permalink raw reply
* Re: [PATCH net-next V2] intel: Remove unnecessary OOM messages
From: David Miller @ 2014-02-13 22:47 UTC (permalink / raw)
To: joe; +Cc: jeffrey.t.kirsher, e1000-devel, netdev
In-Reply-To: <1392168652.1868.1.camel@joe-AO722>
From: Joe Perches <joe@perches.com>
Date: Tue, 11 Feb 2014 17:30:52 -0800
> Don't emit these as there's a generic OOM with a dump_stack()
> on allocation failures.
>
> Signed-off-by: Joe Perches <joe@perches.com>
> ---
>
> With patch attached this time...
I expect the Intel folks to pick this up.
^ permalink raw reply
* Re: [Patch net-next v2] net: allow setting mac address of loopback device
From: David Miller @ 2014-02-13 22:47 UTC (permalink / raw)
To: xiyou.wangcong; +Cc: netdev, hannes, nhorman, stephen, edumazet
In-Reply-To: <1392168086-25607-1-git-send-email-xiyou.wangcong@gmail.com>
From: Cong Wang <xiyou.wangcong@gmail.com>
Date: Tue, 11 Feb 2014 17:21:26 -0800
> We are trying to mirror the local traffic from lo to eth0,
> allowing setting mac address of lo to eth0 would make
> the ether addresses in these packets correct, so that
> we don't have to modify the ether header again.
>
> Since usually no one cares about its mac address (all-zero),
> it is safe to allow those who care to set its mac address.
>
> Cc: Hannes Frederic Sowa <hannes@stressinduktion.org>
> Cc: Neil Horman <nhorman@tuxdriver.com>
> Cc: Stephen Hemminger <stephen@networkplumber.org>
> Cc: Eric Dumazet <edumazet@google.com>
> Cc: David S. Miller <davem@davemloft.net>
> Signed-off-by: Cong Wang <xiyou.wangcong@gmail.com>
Applied, thanks.
^ permalink raw reply
* Re: [PATCH net-next 4/4] sch_netem: replace spin_(un)lock_bh with sch_tree_(un)lock
From: David Miller @ 2014-02-13 22:46 UTC (permalink / raw)
To: yangyingliang; +Cc: netdev, stephen
In-Reply-To: <1392173895-5012-5-git-send-email-yangyingliang@huawei.com>
From: Yang Yingliang <yangyingliang@huawei.com>
Date: Wed, 12 Feb 2014 10:58:15 +0800
> spin_(un)lock_bh(root_lock) is same as sch_tree_(un)lock.
>
> Signed-off-by: Yang Yingliang <yangyingliang@huawei.com>
...
> @@ -684,11 +683,9 @@ static int get_dist_table(struct Qdisc *sch, const struct nlattr *attr)
> for (i = 0; i < n; i++)
> d->table[i] = data[i];
>
> - root_lock = qdisc_root_sleeping_lock(sch);
> -
> - spin_lock_bh(root_lock);
> + sch_tree_lock(sch);
> swap(q->delay_dist, d);
> - spin_unlock_bh(root_lock);
> + sch_tree_unlock(sch);
>
> dist_free(d);
> return 0;
This is more expensive than the existing code.
We will now calculate qdisc_root_sleeping_lock() twice which is at
least two pointer dereferences each.
Without explicitly open-coding this, the compiler cannot cache the
result, because the spin lock operations have memory barriers (if
implemented inline) or are considered to potentially modify all memory
(if implemented as function calls).
^ permalink raw reply
* Re: [RFC PATCH v2 tip 0/7] 64-bit BPF insn set and tracing filters
From: Daniel Borkmann @ 2014-02-13 22:44 UTC (permalink / raw)
To: H. Peter Anvin
Cc: Alexei Starovoitov, Ingo Molnar, David S. Miller, Steven Rostedt,
Peter Zijlstra, Thomas Gleixner, Masami Hiramatsu, Tom Zanussi,
Jovi Zhangwei, Eric Dumazet, Linus Torvalds, Andrew Morton,
Frederic Weisbecker, Arnaldo Carvalho de Melo, Pekka Enberg,
Arjan van de Ven, Christoph Hellwig, linux-kernel, netdev
In-Reply-To: <52FD47F9.2090907@zytor.com>
On 02/13/2014 11:32 PM, H. Peter Anvin wrote:
> On 02/06/2014 05:20 PM, Alexei Starovoitov wrote:
>>
>> I believe that old BPF outlived itself and BPF64 should
>> replace it in all current use cases plus a lot more.
>> It just cannot happen at once.
>> BPF64 can come in. bpf32->bpf64 converter functioning.
>> JIT from bpf64->aarch64 and may be sparc64 needs to be in place.
>> Then old bpf can fade away.
>
> I don't think that is doable any time soon. Right now pretty much all
> mobile devices, for example, are 32 bits and they really want to use
> syscall filtering for security. Performance matters greatly there.
Well, if that would be the case, then seccomp would have had JIT support
long ago. ;-) Right now BPF filters with seccomp are not JIT compiled
for _any_ architecture.
> As such, 32-bit JIT support is going to be very important for a long
> time to come.
True, I think that pretty much depends if we can manage to find a way
to cleanly integrate it into net/core/filter.c while still supporting
the old instructions as I've mentioned earlier.
> -hpa
>
>
^ permalink raw reply
* Re: [PATCH] net: clear iflink when moving to a new netns
From: Cong Wang @ 2014-02-13 22:44 UTC (permalink / raw)
To: Eric W. Biederman
Cc: Nicolas Dichtel, Cong Wang, netdev, David S. Miller, Eric Dumazet,
Hannes Frederic Sowa
In-Reply-To: <87sirn4vqo.fsf@xmission.com>
On Wed, Feb 12, 2014 at 6:00 PM, Eric W. Biederman
<ebiederm@xmission.com> wrote:
>
> There was an effort not long ago to make tunnels safe to pass between
> namespaces. NETIF_F_NETNS_LOCAL was removed from ip6_tunnel in that
> effort. Apparently something was overlooked.
>
> Making iflink a netdevice reference or finding a way to remove it
> entirely seems better that masking the problem.
>
Yeah, looks like the trend is allowing more stacked devices to move to
a new netns.
Sounds good. I am working on a draft patch now.
Thanks.
^ permalink raw reply
* Re: [PATCH v3 0/3] dp83640: Get pin and master/slave configuration from DT
From: David Miller @ 2014-02-13 22:39 UTC (permalink / raw)
To: stefan.sorensen-usnHOLptxrsHrNJx0XZkJA
Cc: richardcochran-Re5JQEeQqe8AvxtiuMwx3w,
grant.likely-QSEj5FYQhm4dnm+yROfE0A,
robh+dt-DgEjT+Ai2ygdnm+yROfE0A, mark.rutland-5wv7dgnIgG8,
devicetree-u79uwXL29TY76Z2rM5mHXA,
linux-kernel-u79uwXL29TY76Z2rM5mHXA,
netdev-u79uwXL29TY76Z2rM5mHXA
In-Reply-To: <1392302129-24947-1-git-send-email-stefan.sorensen-usnHOLptxrsHrNJx0XZkJA@public.gmane.org>
From: Stefan Sørensen <stefan.sorensen-usnHOLptxrsHrNJx0XZkJA@public.gmane.org>
Date: Thu, 13 Feb 2014 15:35:22 +0100
> This patch series add DT configuration to the DP83640 PHY driver and makes
> the configuration of periodic output pins dynamic.
>
> Changes since v2:
> - Add patch to properly configure perout triggers 0+1
> - Keep extts and perout numbers separate
> - Shorten pr_err lines
>
> Changes since v1:
> - Add bindings documentation
> - Keep module parameters
> - Rename gpio->pin
> - Split patch into DT part and dynamic periodic output support
>
> Stefan Sørensen (3):
> net:phy:dp83640: Program pulsewidth2 values of perout triggers 0 and 1
> net:phy:dp83640: Support a configurable number of periodic outputs
> net:phy:dp83640: Get pin and master/slave configuration from DT
I'm not applying this.
This is a great example of why I hate module parameters.
Whoever added these module parameters in the first place took the easy
road and didn't have to think about whether this was a good way or not
to specify these parameters. Do you really think we can, like magic,
just remove this module parameter without having to think about the
consequences of that action at all?
The entities currently specifying this parameter, do they just lose?
Sorry, that's unacceptable.
You'll have to code this in a way such that people who currently specify
this parameter keep working.
If you don't like it, be angry at whoever added this module parameter
in the first place, not me. Module parameters are 99 times out of 100
not the right way to configure something, really.
Thanks.
--
To unsubscribe from this list: send the line "unsubscribe devicetree" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
^ permalink raw reply
* Re: [RFC PATCH v2 tip 0/7] 64-bit BPF insn set and tracing filters
From: H. Peter Anvin @ 2014-02-13 22:32 UTC (permalink / raw)
To: Alexei Starovoitov, Daniel Borkmann
Cc: Ingo Molnar, David S. Miller, Steven Rostedt, Peter Zijlstra,
Thomas Gleixner, Masami Hiramatsu, Tom Zanussi, Jovi Zhangwei,
Eric Dumazet, Linus Torvalds, Andrew Morton, Frederic Weisbecker,
Arnaldo Carvalho de Melo, Pekka Enberg, Arjan van de Ven,
Christoph Hellwig, linux-kernel, netdev
In-Reply-To: <CAMEtUuza3NuM7L0RnVVgU-MTfjE2M==8wb4UY2TAo0Pg6DKZqA@mail.gmail.com>
On 02/06/2014 05:20 PM, Alexei Starovoitov wrote:
>
> I believe that old BPF outlived itself and BPF64 should
> replace it in all current use cases plus a lot more.
> It just cannot happen at once.
> BPF64 can come in. bpf32->bpf64 converter functioning.
> JIT from bpf64->aarch64 and may be sparc64 needs to be in place.
> Then old bpf can fade away.
>
I don't think that is doable any time soon. Right now pretty much all
mobile devices, for example, are 32 bits and they really want to use
syscall filtering for security. Performance matters greatly there.
As such, 32-bit JIT support is going to be very important for a long
time to come.
-hpa
^ permalink raw reply
* Re: [PATCH regression] dma debug: account for cachelines and read-only mappings in overlap tracking
From: Dan Williams @ 2014-02-13 22:33 UTC (permalink / raw)
To: Andrew Morton
Cc: Wei Liu, Eric Dumazet, Konrad Rzeszutek Wilk, Netdev,
linux-kernel@vger.kernel.org, Sander Eikelenboom, Francois Romieu,
Dave Jones
In-Reply-To: <20140213140523.099fd03418d8fe1467db3bd9@linux-foundation.org>
On Thu, Feb 13, 2014 at 2:05 PM, Andrew Morton
<akpm@linux-foundation.org> wrote:
> On Thu, 13 Feb 2014 13:58:00 -0800 Dan Williams <dan.j.williams@intel.com> wrote:
>
>> While debug_dma_assert_idle() checks if a given *page* is actively
>> undergoing dma the valid granularity of a dma mapping is a *cacheline*.
>> Sander's testing shows that the warning message "DMA-API: exceeded 7
>> overlapping mappings of pfn..." is falsely triggering. The test is
>> simply mapping multiple cachelines in a given page.
>>
>> Ultimately we want overlap tracking to be valid as it is a real api
>> violation, so we need to track active mappings by cachelines. Update
>> the active dma tracking to use the page-frame-relative cacheline of the
>> mapping as the key, and update debug_dma_assert_idle() to check for all
>> possible mapped cachelines for a given page.
>>
>> However, the need to track active mappings is only relevant when the
>> dma-mapping is writable by the device. In fact it is fairly standard
>> for read-only mappings to have hundreds or thousands of overlapping
>> mappings at once. Limiting the overlap tracking to writable
>> (!DMA_TO_DEVICE) eliminates this class of false-positive overlap
>> reports.
>>
>> Note, the radix gang lookup is sub-optimal. It would be best if it
>> stopped fetching entries once the search passed a page boundary.
>> Nevertheless, this implementation does not perturb the original net_dma
>> failing case. That is to say the extra overhead does not show up in
>> terms of making the failing case pass due to a timing change.
>>
>> References:
>> http://marc.info/?l=linux-netdev&m=139232263419315&w=2
>> http://marc.info/?l=linux-netdev&m=139217088107122&w=2
>>
>> ...
>>
>> --- a/lib/dma-debug.c
>> +++ b/lib/dma-debug.c
>> @@ -424,111 +424,132 @@ void debug_dma_dump_mappings(struct device *dev)
>> EXPORT_SYMBOL(debug_dma_dump_mappings);
>>
>> /*
>> - * For each page mapped (initial page in the case of
>> - * dma_alloc_coherent/dma_map_{single|page}, or each page in a
>> - * scatterlist) insert into this tree using the pfn as the key. At
>> + * For each mapping (initial cacheline in the case of
>> + * dma_alloc_coherent/dma_map_page, initial cacheline in each page of a
>> + * scatterlist, or the cacheline specified in dma_map_single) insert
>> + * into this tree using the cacheline as the key. At
>> * dma_unmap_{single|sg|page} or dma_free_coherent delete the entry. If
>> - * the pfn already exists at insertion time add a tag as a reference
>> + * the entry already exists at insertion time add a tag as a reference
>> * count for the overlapping mappings. For now, the overlap tracking
>> - * just ensures that 'unmaps' balance 'maps' before marking the pfn
>> - * idle, but we should also be flagging overlaps as an API violation.
>> + * just ensures that 'unmaps' balance 'maps' before marking the
>> + * cacheline idle, but we should also be flagging overlaps as an API
>> + * violation.
>> *
>> * Memory usage is mostly constrained by the maximum number of available
>> * dma-debug entries in that we need a free dma_debug_entry before
>> - * inserting into the tree. In the case of dma_map_{single|page} and
>> - * dma_alloc_coherent there is only one dma_debug_entry and one pfn to
>> - * track per event. dma_map_sg(), on the other hand,
>> - * consumes a single dma_debug_entry, but inserts 'nents' entries into
>> - * the tree.
>> + * inserting into the tree. In the case of dma_map_page and
>> + * dma_alloc_coherent there is only one dma_debug_entry and one
>> + * dma_active_cacheline entry to track per event. dma_map_sg(), on the
>> + * other hand, consumes a single dma_debug_entry, but inserts 'nents'
>> + * entries into the tree.
>> *
>> * At any time debug_dma_assert_idle() can be called to trigger a
>> - * warning if the given page is in the active set.
>> + * warning if any cachelines in the given page are in the active set.
>> */
>> -static RADIX_TREE(dma_active_pfn, GFP_NOWAIT);
>> +static RADIX_TREE(dma_active_cacheline, GFP_NOWAIT);
>> static DEFINE_SPINLOCK(radix_lock);
>> -#define ACTIVE_PFN_MAX_OVERLAP ((1 << RADIX_TREE_MAX_TAGS) - 1)
>> +#define ACTIVE_CLN_MAX_OVERLAP ((1 << RADIX_TREE_MAX_TAGS) - 1)
>> +#define CACHELINE_PER_PAGE_SHIFT (PAGE_SHIFT - L1_CACHE_SHIFT)
>> +#define CACHELINES_PER_PAGE (1 << CACHELINE_PER_PAGE_SHIFT)
>>
>> -static int active_pfn_read_overlap(unsigned long pfn)
>> +unsigned long to_cln(struct dma_debug_entry *entry)
>> +{
>> + return (entry->pfn << CACHELINE_PER_PAGE_SHIFT) +
>> + (entry->offset >> L1_CACHE_SHIFT);
>> +}
>
> "cln" is ugly and isn't a well-known kernel abbreviation. We typically
> spell these things out, so "cacheline". But I think you mean
> "cacheline number", and that is too long to spell out.
>
I do mean cacheline number.
> So I guess "cln" just became a well-known kernel abbreviation.
I can at least make the function names use "cacheline" to give better
context about the local 'cln' variable.
>> ....
>>
>> void debug_dma_assert_idle(struct page *page)
>> {
>> + unsigned long cln = page_to_pfn(page) << CACHELINE_PER_PAGE_SHIFT;
>
> This worries me. Are you sure we cannot overflow the ulong here under
> any circumstances? 32GB PAE with sparsemem or whatever?
You're right, I can't be sure. Certainly page_to_pfn() and max_pfn
are unsigned long, but I don't know how much headroom we have to play
with on all memory-models... so better make a 'cacheline number' be a
phys_addr_t to be safe.
--
Dan
^ permalink raw reply
* [PATCH net-next v2 14/14] tipc: add node_lock protection to link lookup function
From: Jon Maloy @ 2014-02-13 22:29 UTC (permalink / raw)
To: davem
Cc: netdev, Paul Gortmaker, erik.hugne, ying.xue, maloy,
tipc-discussion, Jon Maloy
In-Reply-To: <1392330558-19048-1-git-send-email-jon.maloy@ericsson.com>
In an earlier commit, ("tipc: remove links list from bearer struct")
we described three issues that need to be pre-emptively resolved before
we can remove tipc_net_lock. Here we resolve issue a) described in that
commit:
"a) In access method #2, we access the link before taking the
protecting node_lock. This will not work once net_lock is gone,
so we will have to change the access order. We will deal with
this in a later commit in this series."
Here, we change that access order, by ensuring that the function
link_find_link() returns only a safe reference for finding
the link, i.e., a node pointer and an index into its 'links' array,
not the link pointer itself. We also change all callers of this
function to first take the node lock before they can check if there
still is a valid link pointer at the returned index. Since the
function now returns a node pointer rather than a link pointer,
we rename it to the more appropriate 'tipc_link_find_owner().
Signed-off-by: Jon Maloy <jon.maloy@ericsson.com>
Reviewed-by: Ying Xue <ying.xue@windriver.com>
---
net/tipc/link.c | 110 ++++++++++++++++++++++++++++++++-----------------------
1 file changed, 65 insertions(+), 45 deletions(-)
diff --git a/net/tipc/link.c b/net/tipc/link.c
index 0307516..4fb4ae0 100644
--- a/net/tipc/link.c
+++ b/net/tipc/link.c
@@ -2390,35 +2390,40 @@ void tipc_link_set_queue_limits(struct tipc_link *l_ptr, u32 window)
l_ptr->queue_limit[MSG_FRAGMENTER] = 4000;
}
-/**
- * link_find_link - locate link by name
- * @name: ptr to link name string
- * @node: ptr to area to be filled with ptr to associated node
- *
+/* tipc_link_find_owner - locate owner node of link by link's name
+ * @name: pointer to link name string
+ * @bearer_id: pointer to index in 'node->links' array where the link was found.
* Caller must hold 'tipc_net_lock' to ensure node and bearer are not deleted;
* this also prevents link deletion.
*
- * Returns pointer to link (or 0 if invalid link name).
+ * Returns pointer to node owning the link, or 0 if no matching link is found.
*/
-static struct tipc_link *link_find_link(const char *name,
- struct tipc_node **node)
+static struct tipc_node *tipc_link_find_owner(const char *link_name,
+ unsigned int *bearer_id)
{
struct tipc_link *l_ptr;
struct tipc_node *n_ptr;
+ struct tipc_node *tmp_n_ptr;
+ struct tipc_node *found_node = 0;
+
int i;
- list_for_each_entry(n_ptr, &tipc_node_list, list) {
+ *bearer_id = 0;
+ list_for_each_entry_safe(n_ptr, tmp_n_ptr, &tipc_node_list, list) {
+ spin_lock(&n_ptr->lock);
for (i = 0; i < MAX_BEARERS; i++) {
l_ptr = n_ptr->links[i];
- if (l_ptr && !strcmp(l_ptr->name, name))
- goto found;
+ if (l_ptr && !strcmp(l_ptr->name, link_name)) {
+ *bearer_id = i;
+ found_node = n_ptr;
+ break;
+ }
}
+ spin_unlock(&n_ptr->lock);
+ if (found_node)
+ break;
}
- l_ptr = NULL;
- n_ptr = NULL;
-found:
- *node = n_ptr;
- return l_ptr;
+ return found_node;
}
/**
@@ -2460,32 +2465,33 @@ static int link_cmd_set_value(const char *name, u32 new_value, u16 cmd)
struct tipc_link *l_ptr;
struct tipc_bearer *b_ptr;
struct tipc_media *m_ptr;
+ int bearer_id;
int res = 0;
- l_ptr = link_find_link(name, &node);
- if (l_ptr) {
- /*
- * acquire node lock for tipc_link_send_proto_msg().
- * see "TIPC locking policy" in net.c.
- */
+ node = tipc_link_find_owner(name, &bearer_id);
+ if (node) {
tipc_node_lock(node);
- switch (cmd) {
- case TIPC_CMD_SET_LINK_TOL:
- link_set_supervision_props(l_ptr, new_value);
- tipc_link_send_proto_msg(l_ptr,
- STATE_MSG, 0, 0, new_value, 0, 0);
- break;
- case TIPC_CMD_SET_LINK_PRI:
- l_ptr->priority = new_value;
- tipc_link_send_proto_msg(l_ptr,
- STATE_MSG, 0, 0, 0, new_value, 0);
- break;
- case TIPC_CMD_SET_LINK_WINDOW:
- tipc_link_set_queue_limits(l_ptr, new_value);
- break;
- default:
- res = -EINVAL;
- break;
+ l_ptr = node->links[bearer_id];
+
+ if (l_ptr) {
+ switch (cmd) {
+ case TIPC_CMD_SET_LINK_TOL:
+ link_set_supervision_props(l_ptr, new_value);
+ tipc_link_send_proto_msg(l_ptr, STATE_MSG, 0,
+ 0, new_value, 0, 0);
+ break;
+ case TIPC_CMD_SET_LINK_PRI:
+ l_ptr->priority = new_value;
+ tipc_link_send_proto_msg(l_ptr, STATE_MSG, 0,
+ 0, 0, new_value, 0);
+ break;
+ case TIPC_CMD_SET_LINK_WINDOW:
+ tipc_link_set_queue_limits(l_ptr, new_value);
+ break;
+ default:
+ res = -EINVAL;
+ break;
+ }
}
tipc_node_unlock(node);
return res;
@@ -2580,6 +2586,7 @@ struct sk_buff *tipc_link_cmd_reset_stats(const void *req_tlv_area, int req_tlv_
char *link_name;
struct tipc_link *l_ptr;
struct tipc_node *node;
+ unsigned int bearer_id;
if (!TLV_CHECK(req_tlv_area, req_tlv_space, TIPC_TLV_LINK_NAME))
return tipc_cfg_reply_error_string(TIPC_CFG_TLV_ERROR);
@@ -2590,15 +2597,19 @@ struct sk_buff *tipc_link_cmd_reset_stats(const void *req_tlv_area, int req_tlv_
return tipc_cfg_reply_error_string("link not found");
return tipc_cfg_reply_none();
}
-
read_lock_bh(&tipc_net_lock);
- l_ptr = link_find_link(link_name, &node);
+ node = tipc_link_find_owner(link_name, &bearer_id);
+ if (!node) {
+ read_unlock_bh(&tipc_net_lock);
+ return tipc_cfg_reply_error_string("link not found");
+ }
+ spin_lock(&node->lock);
+ l_ptr = node->links[bearer_id];
if (!l_ptr) {
+ tipc_node_unlock(node);
read_unlock_bh(&tipc_net_lock);
return tipc_cfg_reply_error_string("link not found");
}
-
- tipc_node_lock(node);
link_reset_statistics(l_ptr);
tipc_node_unlock(node);
read_unlock_bh(&tipc_net_lock);
@@ -2628,18 +2639,27 @@ static int tipc_link_stats(const char *name, char *buf, const u32 buf_size)
struct tipc_node *node;
char *status;
u32 profile_total = 0;
+ unsigned int bearer_id;
int ret;
if (!strcmp(name, tipc_bclink_name))
return tipc_bclink_stats(buf, buf_size);
read_lock_bh(&tipc_net_lock);
- l = link_find_link(name, &node);
- if (!l) {
+ node = tipc_link_find_owner(name, &bearer_id);
+ if (!node) {
read_unlock_bh(&tipc_net_lock);
return 0;
}
tipc_node_lock(node);
+
+ l = node->links[bearer_id];
+ if (!l) {
+ tipc_node_unlock(node);
+ read_unlock_bh(&tipc_net_lock);
+ return 0;
+ }
+
s = &l->stats;
if (tipc_link_is_active(l))
--
1.7.9.5
^ permalink raw reply related
* [PATCH net-next v2 12/14] tipc: delay delete of link when failover is needed
From: Jon Maloy @ 2014-02-13 22:29 UTC (permalink / raw)
To: davem
Cc: netdev, Paul Gortmaker, erik.hugne, ying.xue, maloy,
tipc-discussion, Jon Maloy
In-Reply-To: <1392330558-19048-1-git-send-email-jon.maloy@ericsson.com>
When a bearer is disabled, all its attached links are deleted.
Ideally, we should do link failover to redundant links on other bearers,
if there are any, in such cases. This would be consistent with current
behavior when a link is reset, but not deleted. However, due to the
complexity involved, and the (wrongly) perceived low demand for this
feature, it was never implemented until now.
We mark the doomed link for deletion with a new flag, but wait until the
failover process is finished before we actually delete it. With the
improved link tunnelling/failover code introduced earlier in this commit
series, it is now easy to identify a spot in the code where the failover
is finished and it is safe to delete the marked link. Moreover, the test
for the flag and the deletion can be done synchronously, and outside the
most time critical data path.
Signed-off-by: Jon Maloy <jon.maloy@ericsson.com>
Reviewed-by: Ying Xue <ying.xue@windriver.com>
---
net/tipc/bearer.c | 12 ++++++------
net/tipc/link.c | 31 ++++++++++++++++++++++---------
net/tipc/link.h | 2 +-
net/tipc/node.c | 8 +++++++-
4 files changed, 36 insertions(+), 17 deletions(-)
diff --git a/net/tipc/bearer.c b/net/tipc/bearer.c
index 4931eea..60caa45 100644
--- a/net/tipc/bearer.c
+++ b/net/tipc/bearer.c
@@ -51,7 +51,7 @@ static struct tipc_media * const media_info_array[] = {
struct tipc_bearer tipc_bearers[MAX_BEARERS];
-static void bearer_disable(struct tipc_bearer *b_ptr);
+static void bearer_disable(struct tipc_bearer *b_ptr, bool shutting_down);
/**
* tipc_media_find - locates specified media object by name
@@ -331,7 +331,7 @@ restart:
res = tipc_disc_create(b_ptr, &b_ptr->bcast_addr, disc_domain);
if (res) {
- bearer_disable(b_ptr);
+ bearer_disable(b_ptr, false);
pr_warn("Bearer <%s> rejected, discovery object creation failed\n",
name);
goto exit;
@@ -363,14 +363,14 @@ static int tipc_reset_bearer(struct tipc_bearer *b_ptr)
*
* Note: This routine assumes caller holds tipc_net_lock.
*/
-static void bearer_disable(struct tipc_bearer *b_ptr)
+static void bearer_disable(struct tipc_bearer *b_ptr, bool shutting_down)
{
struct tipc_link_req *temp_req;
pr_info("Disabling bearer <%s>\n", b_ptr->name);
spin_lock_bh(&b_ptr->lock);
b_ptr->media->disable_media(b_ptr);
- tipc_link_delete_list(b_ptr->identity);
+ tipc_link_delete_list(b_ptr->identity, shutting_down);
temp_req = b_ptr->link_req;
b_ptr->link_req = NULL;
spin_unlock_bh(&b_ptr->lock);
@@ -392,7 +392,7 @@ int tipc_disable_bearer(const char *name)
pr_warn("Attempt to disable unknown bearer <%s>\n", name);
res = -EINVAL;
} else {
- bearer_disable(b_ptr);
+ bearer_disable(b_ptr, false);
res = 0;
}
write_unlock_bh(&tipc_net_lock);
@@ -612,6 +612,6 @@ void tipc_bearer_stop(void)
for (i = 0; i < MAX_BEARERS; i++) {
if (tipc_bearers[i].active)
- bearer_disable(&tipc_bearers[i]);
+ bearer_disable(&tipc_bearers[i], true);
}
}
diff --git a/net/tipc/link.c b/net/tipc/link.c
index 663623c..0307516 100644
--- a/net/tipc/link.c
+++ b/net/tipc/link.c
@@ -281,7 +281,7 @@ struct tipc_link *tipc_link_create(struct tipc_node *n_ptr,
}
-void tipc_link_delete_list(unsigned int bearer_id)
+void tipc_link_delete_list(unsigned int bearer_id, bool shutting_down)
{
struct tipc_link *l_ptr;
struct tipc_node *n_ptr;
@@ -291,12 +291,20 @@ void tipc_link_delete_list(unsigned int bearer_id)
l_ptr = n_ptr->links[bearer_id];
if (l_ptr) {
tipc_link_reset(l_ptr);
- tipc_node_detach_link(n_ptr, l_ptr);
- spin_unlock_bh(&n_ptr->lock);
-
- /* Nobody else can access this link now: */
- del_timer_sync(&l_ptr->timer);
- kfree(l_ptr);
+ if (shutting_down || !tipc_node_is_up(n_ptr)) {
+ tipc_node_detach_link(l_ptr->owner, l_ptr);
+ tipc_link_reset_fragments(l_ptr);
+ spin_unlock_bh(&n_ptr->lock);
+
+ /* Nobody else can access this link now: */
+ del_timer_sync(&l_ptr->timer);
+ kfree(l_ptr);
+ } else {
+ /* Detach/delete when failover is finished: */
+ l_ptr->flags |= LINK_STOPPED;
+ spin_unlock_bh(&n_ptr->lock);
+ del_timer_sync(&l_ptr->timer);
+ }
continue;
}
spin_unlock_bh(&n_ptr->lock);
@@ -481,6 +489,9 @@ static void link_state_event(struct tipc_link *l_ptr, unsigned int event)
struct tipc_link *other;
u32 cont_intv = l_ptr->continuity_interval;
+ if (l_ptr->flags & LINK_STOPPED)
+ return;
+
if (!(l_ptr->flags & LINK_STARTED) && (event != STARTING_EVT))
return; /* Not yet. */
@@ -2167,8 +2178,11 @@ static struct sk_buff *tipc_link_failover_rcv(struct tipc_link *l_ptr,
&buf);
}
}
-
exit:
+ if ((l_ptr->exp_msg_count == 0) && (l_ptr->flags & LINK_STOPPED)) {
+ tipc_node_detach_link(l_ptr->owner, l_ptr);
+ kfree(l_ptr);
+ }
return buf;
}
@@ -2201,7 +2215,6 @@ static int tipc_link_tunnel_rcv(struct tipc_node *n_ptr,
*buf = tipc_link_failover_rcv(l_ptr, t_buf);
else
pr_warn("%sunknown tunnel pkt received\n", link_co_err);
-
exit:
kfree_skb(t_buf);
return *buf != NULL;
diff --git a/net/tipc/link.h b/net/tipc/link.h
index 3340fc1..45b9cd0 100644
--- a/net/tipc/link.h
+++ b/net/tipc/link.h
@@ -214,7 +214,7 @@ struct tipc_port;
struct tipc_link *tipc_link_create(struct tipc_node *n_ptr,
struct tipc_bearer *b_ptr,
const struct tipc_media_addr *media_addr);
-void tipc_link_delete_list(unsigned int bearer_id);
+void tipc_link_delete_list(unsigned int bearer_id, bool shutting_down);
void tipc_link_failover_send_queue(struct tipc_link *l_ptr);
void tipc_link_dup_send_queue(struct tipc_link *l_ptr,
struct tipc_link *dest);
diff --git a/net/tipc/node.c b/net/tipc/node.c
index efe4d41..833324b 100644
--- a/net/tipc/node.c
+++ b/net/tipc/node.c
@@ -249,7 +249,13 @@ void tipc_node_attach_link(struct tipc_node *n_ptr, struct tipc_link *l_ptr)
void tipc_node_detach_link(struct tipc_node *n_ptr, struct tipc_link *l_ptr)
{
- n_ptr->links[l_ptr->b_ptr->identity] = NULL;
+ int i;
+
+ for (i = 0; i < MAX_BEARERS; i++) {
+ if (l_ptr == n_ptr->links[i])
+ break;
+ }
+ n_ptr->links[i] = NULL;
atomic_dec(&tipc_num_links);
n_ptr->link_cnt--;
}
--
1.7.9.5
^ permalink raw reply related
* [PATCH net-next v2 11/14] tipc: changes to general packet reception algorithm
From: Jon Maloy @ 2014-02-13 22:29 UTC (permalink / raw)
To: davem
Cc: netdev, Paul Gortmaker, erik.hugne, ying.xue, maloy,
tipc-discussion, Jon Maloy
In-Reply-To: <1392330558-19048-1-git-send-email-jon.maloy@ericsson.com>
We change the order of checking for destination users when processing
incoming packets. By placing the checks for users that may potentially
replace the processed buffer, i.e., CHANGEOVER_PROTOCOL and
MSG_FRAGMENTER, in a separate step before we check for the true end
users, we get rid of a label and a 'goto', at the same time making the
code more comprehensible and easy to follow.
This commit does not change any functionality, it is just a cosmetic
code reshuffle.
Signed-off-by: Jon Maloy <jon.maloy@ericsson.com>
Reviewed-by: Ying Xue <ying.xue@windriver.com>
---
net/tipc/link.c | 76 +++++++++++++++++++++++++++++--------------------------
1 file changed, 40 insertions(+), 36 deletions(-)
diff --git a/net/tipc/link.c b/net/tipc/link.c
index b678c2e..663623c 100644
--- a/net/tipc/link.c
+++ b/net/tipc/link.c
@@ -1484,7 +1484,7 @@ void tipc_rcv(struct sk_buff *head, struct tipc_bearer *b_ptr)
if ((n_ptr->block_setup & WAIT_PEER_DOWN) &&
msg_user(msg) == LINK_PROTOCOL &&
(msg_type(msg) == RESET_MSG ||
- msg_type(msg) == ACTIVATE_MSG) &&
+ msg_type(msg) == ACTIVATE_MSG) &&
!msg_redundant_link(msg))
n_ptr->block_setup &= ~WAIT_PEER_DOWN;
@@ -1503,7 +1503,6 @@ void tipc_rcv(struct sk_buff *head, struct tipc_bearer *b_ptr)
while ((crs != l_ptr->next_out) &&
less_eq(buf_seqno(crs), ackd)) {
struct sk_buff *next = crs->next;
-
kfree_skb(crs);
crs = next;
released++;
@@ -1516,14 +1515,17 @@ void tipc_rcv(struct sk_buff *head, struct tipc_bearer *b_ptr)
/* Try sending any messages link endpoint has pending */
if (unlikely(l_ptr->next_out))
tipc_link_push_queue(l_ptr);
+
if (unlikely(!list_empty(&l_ptr->waiting_ports)))
tipc_link_wakeup_ports(l_ptr, 0);
+
if (unlikely(++l_ptr->unacked_window >= TIPC_MIN_LINK_WIN)) {
l_ptr->stats.sent_acks++;
- tipc_link_send_proto_msg(l_ptr, STATE_MSG, 0, 0, 0, 0, 0);
+ tipc_link_send_proto_msg(l_ptr, STATE_MSG,
+ 0, 0, 0, 0, 0);
}
- /* Now (finally!) process the incoming message */
+ /* Process the incoming packet */
if (unlikely(!link_working_working(l_ptr))) {
if (msg_user(msg) == LINK_PROTOCOL) {
link_recv_proto_msg(l_ptr, buf);
@@ -1555,14 +1557,40 @@ void tipc_rcv(struct sk_buff *head, struct tipc_bearer *b_ptr)
l_ptr->next_in_no++;
if (unlikely(l_ptr->oldest_deferred_in))
head = link_insert_deferred_queue(l_ptr, head);
-deliver:
- if (likely(msg_isdata(msg))) {
+
+ /* Deliver packet/message to correct user: */
+ if (unlikely(msg_user(msg) == CHANGEOVER_PROTOCOL)) {
+ if (!tipc_link_tunnel_rcv(n_ptr, &buf)) {
+ tipc_node_unlock(n_ptr);
+ continue;
+ }
+ msg = buf_msg(buf);
+ } else if (msg_user(msg) == MSG_FRAGMENTER) {
+ int rc;
+
+ l_ptr->stats.recv_fragments++;
+ rc = tipc_link_frag_rcv(&l_ptr->reasm_head,
+ &l_ptr->reasm_tail,
+ &buf);
+ if (rc == LINK_REASM_COMPLETE) {
+ l_ptr->stats.recv_fragmented++;
+ msg = buf_msg(buf);
+ } else {
+ if (rc == LINK_REASM_ERROR)
+ tipc_link_reset(l_ptr);
+ tipc_node_unlock(n_ptr);
+ continue;
+ }
+ }
+
+ switch (msg_user(msg)) {
+ case TIPC_LOW_IMPORTANCE:
+ case TIPC_MEDIUM_IMPORTANCE:
+ case TIPC_HIGH_IMPORTANCE:
+ case TIPC_CRITICAL_IMPORTANCE:
tipc_node_unlock(n_ptr);
tipc_port_recv_msg(buf);
continue;
- }
- switch (msg_user(msg)) {
- int ret;
case MSG_BUNDLER:
l_ptr->stats.recv_bundles++;
l_ptr->stats.recv_bundled += msg_msgcnt(msg);
@@ -1574,44 +1602,20 @@ deliver:
tipc_node_unlock(n_ptr);
tipc_named_recv(buf);
continue;
- case BCAST_PROTOCOL:
- tipc_link_recv_sync(n_ptr, buf);
- tipc_node_unlock(n_ptr);
- continue;
case CONN_MANAGER:
tipc_node_unlock(n_ptr);
tipc_port_recv_proto_msg(buf);
continue;
- case MSG_FRAGMENTER:
- l_ptr->stats.recv_fragments++;
- ret = tipc_link_frag_rcv(&l_ptr->reasm_head,
- &l_ptr->reasm_tail,
- &buf);
- if (ret == LINK_REASM_COMPLETE) {
- l_ptr->stats.recv_fragmented++;
- msg = buf_msg(buf);
- goto deliver;
- }
- if (ret == LINK_REASM_ERROR)
- tipc_link_reset(l_ptr);
- tipc_node_unlock(n_ptr);
- continue;
- case CHANGEOVER_PROTOCOL:
- if (!tipc_link_tunnel_rcv(n_ptr, &buf))
- break;
- msg = buf_msg(buf);
- seq_no = msg_seqno(msg);
- goto deliver;
+ case BCAST_PROTOCOL:
+ tipc_link_recv_sync(n_ptr, buf);
+ break;
default:
kfree_skb(buf);
- buf = NULL;
break;
}
tipc_node_unlock(n_ptr);
- tipc_net_route_msg(buf);
continue;
unlock_discard:
-
tipc_node_unlock(n_ptr);
discard:
kfree_skb(buf);
--
1.7.9.5
^ permalink raw reply related
* [PATCH net-next v2 10/14] tipc: rename stack variables in function tipc_link_tunnel_rcv
From: Jon Maloy @ 2014-02-13 22:29 UTC (permalink / raw)
To: davem
Cc: netdev, Paul Gortmaker, erik.hugne, ying.xue, maloy,
tipc-discussion, Jon Maloy
In-Reply-To: <1392330558-19048-1-git-send-email-jon.maloy@ericsson.com>
After the previous redesign of the tunnel reception algorithm and
functions, we finalize it by renaming a couple of stack variables
in tipc_tunnel_rcv(). This makes it more consistent with the naming
scheme elsewhere in this part of the code.
This change is purely cosmetic, with no functional changes.
Signed-off-by: Jon Maloy <jon.maloy@ericsson.com>
---
net/tipc/link.c | 22 +++++++++++-----------
1 file changed, 11 insertions(+), 11 deletions(-)
diff --git a/net/tipc/link.c b/net/tipc/link.c
index 3136788..b678c2e 100644
--- a/net/tipc/link.c
+++ b/net/tipc/link.c
@@ -2177,29 +2177,29 @@ exit:
static int tipc_link_tunnel_rcv(struct tipc_node *n_ptr,
struct sk_buff **buf)
{
- struct sk_buff *tunnel_buf = *buf;
- struct tipc_link *dest_link;
- struct tipc_msg *tunnel_msg = buf_msg(tunnel_buf);
- u32 bearer_id = msg_bearer_id(tunnel_msg);
+ struct sk_buff *t_buf = *buf;
+ struct tipc_link *l_ptr;
+ struct tipc_msg *t_msg = buf_msg(t_buf);
+ u32 bearer_id = msg_bearer_id(t_msg);
*buf = NULL;
if (bearer_id >= MAX_BEARERS)
goto exit;
- dest_link = n_ptr->links[bearer_id];
- if (!dest_link)
+ l_ptr = n_ptr->links[bearer_id];
+ if (!l_ptr)
goto exit;
- if (msg_type(tunnel_msg) == DUPLICATE_MSG)
- tipc_link_dup_rcv(dest_link, tunnel_buf);
- else if (msg_type(tunnel_msg) == ORIGINAL_MSG)
- *buf = tipc_link_failover_rcv(dest_link, tunnel_buf);
+ if (msg_type(t_msg) == DUPLICATE_MSG)
+ tipc_link_dup_rcv(l_ptr, t_buf);
+ else if (msg_type(t_msg) == ORIGINAL_MSG)
+ *buf = tipc_link_failover_rcv(l_ptr, t_buf);
else
pr_warn("%sunknown tunnel pkt received\n", link_co_err);
exit:
- kfree_skb(tunnel_buf);
+ kfree_skb(t_buf);
return *buf != NULL;
}
--
1.7.9.5
^ permalink raw reply related
* [PATCH net-next v2 09/14] tipc: more cleanup of tunnelling reception function
From: Jon Maloy @ 2014-02-13 22:29 UTC (permalink / raw)
To: davem
Cc: netdev, Paul Gortmaker, erik.hugne, ying.xue, maloy,
tipc-discussion, Jon Maloy
In-Reply-To: <1392330558-19048-1-git-send-email-jon.maloy@ericsson.com>
We simplify and slim down the code in function tipc_tunnel_rcv()
No impact on the users of this function.
Signed-off-by: Jon Maloy <jon.maloy@ericsson.com>
---
net/tipc/link.c | 21 +++++++--------------
1 file changed, 7 insertions(+), 14 deletions(-)
diff --git a/net/tipc/link.c b/net/tipc/link.c
index f9f9068..3136788 100644
--- a/net/tipc/link.c
+++ b/net/tipc/link.c
@@ -2180,9 +2180,10 @@ static int tipc_link_tunnel_rcv(struct tipc_node *n_ptr,
struct sk_buff *tunnel_buf = *buf;
struct tipc_link *dest_link;
struct tipc_msg *tunnel_msg = buf_msg(tunnel_buf);
- u32 msg_typ = msg_type(tunnel_msg);
u32 bearer_id = msg_bearer_id(tunnel_msg);
+ *buf = NULL;
+
if (bearer_id >= MAX_BEARERS)
goto exit;
@@ -2190,24 +2191,16 @@ static int tipc_link_tunnel_rcv(struct tipc_node *n_ptr,
if (!dest_link)
goto exit;
- if (msg_typ == DUPLICATE_MSG) {
+ if (msg_type(tunnel_msg) == DUPLICATE_MSG)
tipc_link_dup_rcv(dest_link, tunnel_buf);
- goto exit;
- }
-
- if (msg_type(tunnel_msg) == ORIGINAL_MSG) {
+ else if (msg_type(tunnel_msg) == ORIGINAL_MSG)
*buf = tipc_link_failover_rcv(dest_link, tunnel_buf);
+ else
+ pr_warn("%sunknown tunnel pkt received\n", link_co_err);
- /* Do we have a buffer/buffer chain to return? */
- if (*buf != NULL) {
- kfree_skb(tunnel_buf);
- return 1;
- }
- }
exit:
- *buf = NULL;
kfree_skb(tunnel_buf);
- return 0;
+ return *buf != NULL;
}
/*
--
1.7.9.5
^ permalink raw reply related
* [PATCH net-next v2 07/14] tipc: change reception of tunnelled failover packets
From: Jon Maloy @ 2014-02-13 22:29 UTC (permalink / raw)
To: davem
Cc: netdev, Paul Gortmaker, erik.hugne, ying.xue, maloy,
tipc-discussion, Jon Maloy
In-Reply-To: <1392330558-19048-1-git-send-email-jon.maloy@ericsson.com>
When a link is reset, and there is a redundant link available, all
sender sockets will steer their subsequent traffic through the
remaining link. In order to guarantee preserved packet order and
cardinality during the transition, we tunnel the failing link's send
queue through the remaining link before we allow any sockets to use it.
In this commit, we change the algorithm for receiving failover
("ORIGINAL_MSG") packets in tipc_link_tunnel_rcv(), at the same time
delegating it to a new subfuncton, tipc_link_failover_rcv(). Instead
of directly returning an extracted inner packet to the packet reception
loop in tipc_rcv(), we first check if it is a message fragment, in which
case we append it to the reset link's fragment chain. If the fragment
chain is complete, we return the whole chain instead of the individual
buffer, eliminating any need for the tipc_rcv() loop to do reassembly of
tunneled packets.
This change makes it possible to further simplify tipc_link_tunnel_rcv(),
as well as the calling tipc_rcv() loop. We will do that in later
commits. It also makes it possible to identify a single spot in the code
where we can tell that a failover procedure is finished, something that
is useful when we are deleting links after a failover. This will also
be done in a later commit.
Signed-off-by: Jon Maloy <jon.maloy@ericsson.com>
Reviewed-by: Ying Xue <ying.xue@windriver.com>
---
net/tipc/link.c | 75 ++++++++++++++++++++++++++++++++++---------------------
1 file changed, 47 insertions(+), 28 deletions(-)
diff --git a/net/tipc/link.c b/net/tipc/link.c
index f227a38..26a54f4 100644
--- a/net/tipc/link.c
+++ b/net/tipc/link.c
@@ -2124,6 +2124,50 @@ static void tipc_link_dup_rcv(struct tipc_link *l_ptr,
link_handle_out_of_seq_msg(l_ptr, buf);
}
+/* tipc_link_failover_rcv(): Receive a tunnelled ORIGINAL_MSG packet
+ * Owner node is locked.
+ */
+static struct sk_buff *tipc_link_failover_rcv(struct tipc_link *l_ptr,
+ struct sk_buff *t_buf)
+{
+ struct tipc_msg *t_msg = buf_msg(t_buf);
+ struct sk_buff *buf = NULL;
+ struct tipc_msg *msg;
+
+ if (tipc_link_is_up(l_ptr))
+ tipc_link_reset(l_ptr);
+
+ /* First failover packet? */
+ if (l_ptr->exp_msg_count == START_CHANGEOVER)
+ l_ptr->exp_msg_count = msg_msgcnt(t_msg);
+
+ /* Should there be an inner packet? */
+ if (l_ptr->exp_msg_count) {
+ l_ptr->exp_msg_count--;
+ buf = buf_extract(t_buf, INT_H_SIZE);
+ if (buf == NULL) {
+ pr_warn("%sno inner failover pkt\n", link_co_err);
+ goto exit;
+ }
+ msg = buf_msg(buf);
+
+ if (less(msg_seqno(msg), l_ptr->reset_checkpoint)) {
+ kfree_skb(buf);
+ buf = NULL;
+ goto exit;
+ }
+ if (msg_user(msg) == MSG_FRAGMENTER) {
+ l_ptr->stats.recv_fragments++;
+ tipc_link_frag_rcv(&l_ptr->reasm_head,
+ &l_ptr->reasm_tail,
+ &buf);
+ }
+ }
+
+exit:
+ return buf;
+}
+
/* tipc_link_tunnel_rcv(): Receive a tunnelled packet, sent
* via other link as result of a failover (ORIGINAL_MSG) or
* a new active link (DUPLICATE_MSG). Failover packets are
@@ -2135,10 +2179,8 @@ static int tipc_link_tunnel_rcv(struct tipc_link **l_ptr,
{
struct sk_buff *tunnel_buf = *buf;
struct tipc_link *dest_link;
- struct tipc_msg *msg;
struct tipc_msg *tunnel_msg = buf_msg(tunnel_buf);
u32 msg_typ = msg_type(tunnel_msg);
- u32 msg_count = msg_msgcnt(tunnel_msg);
u32 bearer_id = msg_bearer_id(tunnel_msg);
if (bearer_id >= MAX_BEARERS)
@@ -2153,42 +2195,19 @@ static int tipc_link_tunnel_rcv(struct tipc_link **l_ptr,
goto exit;
}
*l_ptr = dest_link;
- msg = msg_get_wrapped(tunnel_msg);
if (msg_typ == DUPLICATE_MSG) {
tipc_link_dup_rcv(dest_link, tunnel_buf);
goto exit;
}
- /* First original message ?: */
- if (tipc_link_is_up(dest_link)) {
- pr_info("%s<%s>, changeover initiated by peer\n", link_rst_msg,
- dest_link->name);
- tipc_link_reset(dest_link);
- dest_link->exp_msg_count = msg_count;
- if (!msg_count)
- goto exit;
- } else if (dest_link->exp_msg_count == START_CHANGEOVER) {
- dest_link->exp_msg_count = msg_count;
- if (!msg_count)
- goto exit;
- }
+ if (msg_type(tunnel_msg) == ORIGINAL_MSG) {
+ *buf = tipc_link_failover_rcv(dest_link, tunnel_buf);
- /* Receive original message */
- if (dest_link->exp_msg_count == 0) {
- pr_warn("%sgot too many tunnelled messages\n", link_co_err);
- goto exit;
- }
- dest_link->exp_msg_count--;
- if (less(msg_seqno(msg), dest_link->reset_checkpoint)) {
- goto exit;
- } else {
- *buf = buf_extract(tunnel_buf, INT_H_SIZE);
+ /* Do we have a buffer/buffer chain to return? */
if (*buf != NULL) {
kfree_skb(tunnel_buf);
return 1;
- } else {
- pr_warn("%soriginal msg dropped\n", link_co_err);
}
}
exit:
--
1.7.9.5
^ permalink raw reply related
* [PATCH net-next v2 05/14] tipc: remove 'links' list from tipc_bearer struct
From: Jon Maloy @ 2014-02-13 22:29 UTC (permalink / raw)
To: davem
Cc: netdev, Paul Gortmaker, erik.hugne, ying.xue, maloy,
tipc-discussion, Jon Maloy
In-Reply-To: <1392330558-19048-1-git-send-email-jon.maloy@ericsson.com>
From: Ying Xue <ying.xue@windriver.com>
In our ongoing effort to simplify the TIPC locking structure,
we see a need to remove the linked list for tipc_links
in the bearer. This can be explained as follows.
Currently, we have three different ways to access a link,
via three different lists/tables:
1: Via a node hash table:
Used by the time-critical outgoing/incoming data paths.
(e.g. link_send_sections_fast() and tipc_recv_msg() ):
grab net_lock(read)
find node from node hash table
grab node_lock
select link
grab bearer_lock
send_msg()
release bearer_lock
release node lock
release net_lock
2: Via a global linked list for nodes:
Used by configuration commands (link_cmd_set_value())
grab net_lock(read)
find node and link from global node list (using link name)
grab node_lock
update link
release node lock
release net_lock
(Same locking order as above. No problem.)
3: Via the bearer's linked link list:
Used by notifications from interface (e.g. tipc_disable_bearer() )
grab net_lock(write)
grab bearer_lock
get link ptr from bearer's link list
get node from link
grab node_lock
delete link
release node lock
release bearer_lock
release net_lock
(Different order from above, but works because we grab the
outer net_lock in write mode first, excluding all other access.)
The first major goal in our simplification effort is to get rid
of the "big" net_lock, replacing it with rcu-locks when accessing
the node list and node hash array. This will come in a later patch
series.
But to get there we first need to rewrite access methods ##2 and 3,
since removal of net_lock would introduce three major problems:
a) In access method #2, we access the link before taking the
protecting node_lock. This will not work once net_lock is gone,
so we will have to change the access order. We will deal with
this in a later commit in this series, "tipc: add node lock
protection to link found by link_find_link()".
b) When the outer protection from net_lock is gone, taking
bearer_lock and node_lock in opposite order of method 1) and 2)
will become an obvious deadlock hazard. This is fixed in the
commit ("tipc: remove bearer_lock from tipc_bearer struct")
later in this series.
c) Similar to what is described in problem a), access method #3
starts with using a link pointer that is unprotected by node_lock,
in order to via that pointer find the correct node struct and
lock it. Before we remove net_lock, this access order must be
altered. This is what we do with this commit.
We can avoid introducing problem problem c) by even here using the
global node list to find the node, before accessing its links. When
we loop though the node list we use the own bearer identity as search
criteria, thus easily finding the links that are associated to the
resetting/disabling bearer. It should be noted that although this
method is somewhat slower than the current list traversal, it is in
no way time critical. This is only about resetting or deleting links,
something that must be considered relatively infrequent events.
As a bonus, we can get rid of the mutual pointers between links and
bearers. After this commit, pointer dependency go in one direction
only: from the link to the bearer.
This commit pre-empts introduction of problem c) as described above.
Signed-off-by: Ying Xue <ying.xue@windriver.com>
Reviewed-by: Paul Gortmaker <paul.gortmaker@windriver.com>
Signed-off-by: Jon Maloy <jon.maloy@ericsson.com>
---
net/tipc/bearer.c | 5 ++--
net/tipc/bearer.h | 2 --
net/tipc/core.c | 2 +-
net/tipc/link.c | 67 +++++++++++++++++++----------------------------------
net/tipc/link.h | 8 +++----
5 files changed, 30 insertions(+), 54 deletions(-)
diff --git a/net/tipc/bearer.c b/net/tipc/bearer.c
index a5be053..4931eea 100644
--- a/net/tipc/bearer.c
+++ b/net/tipc/bearer.c
@@ -327,7 +327,6 @@ restart:
b_ptr->net_plane = bearer_id + 'A';
b_ptr->active = 1;
b_ptr->priority = priority;
- INIT_LIST_HEAD(&b_ptr->links);
spin_lock_init(&b_ptr->lock);
res = tipc_disc_create(b_ptr, &b_ptr->bcast_addr, disc_domain);
@@ -353,7 +352,7 @@ static int tipc_reset_bearer(struct tipc_bearer *b_ptr)
read_lock_bh(&tipc_net_lock);
pr_info("Resetting bearer <%s>\n", b_ptr->name);
spin_lock_bh(&b_ptr->lock);
- tipc_link_reset_list(b_ptr);
+ tipc_link_reset_list(b_ptr->identity);
spin_unlock_bh(&b_ptr->lock);
read_unlock_bh(&tipc_net_lock);
return 0;
@@ -371,7 +370,7 @@ static void bearer_disable(struct tipc_bearer *b_ptr)
pr_info("Disabling bearer <%s>\n", b_ptr->name);
spin_lock_bh(&b_ptr->lock);
b_ptr->media->disable_media(b_ptr);
- tipc_link_delete_list(b_ptr);
+ tipc_link_delete_list(b_ptr->identity);
temp_req = b_ptr->link_req;
b_ptr->link_req = NULL;
spin_unlock_bh(&b_ptr->lock);
diff --git a/net/tipc/bearer.h b/net/tipc/bearer.h
index 4f5db9a..647cb1d 100644
--- a/net/tipc/bearer.h
+++ b/net/tipc/bearer.h
@@ -120,7 +120,6 @@ struct tipc_media {
* @tolerance: default link tolerance for bearer
* @identity: array index of this bearer within TIPC bearer array
* @link_req: ptr to (optional) structure making periodic link setup requests
- * @links: list of non-congested links associated with bearer
* @active: non-zero if bearer structure is represents a bearer
* @net_plane: network plane ('A' through 'H') currently associated with bearer
* @nodes: indicates which nodes in cluster can be reached through bearer
@@ -142,7 +141,6 @@ struct tipc_bearer {
u32 tolerance;
u32 identity;
struct tipc_link_req *link_req;
- struct list_head links;
int active;
char net_plane;
struct tipc_node_map nodes;
diff --git a/net/tipc/core.c b/net/tipc/core.c
index f9e88d8..3f76b98 100644
--- a/net/tipc/core.c
+++ b/net/tipc/core.c
@@ -1,7 +1,7 @@
/*
* net/tipc/core.c: TIPC module code
*
- * Copyright (c) 2003-2006, Ericsson AB
+ * Copyright (c) 2003-2006, 2013, Ericsson AB
* Copyright (c) 2005-2006, 2010-2013, Wind River Systems
* All rights reserved.
*
diff --git a/net/tipc/link.c b/net/tipc/link.c
index 2070d03..e7e44ab 100644
--- a/net/tipc/link.c
+++ b/net/tipc/link.c
@@ -147,11 +147,6 @@ int tipc_link_is_active(struct tipc_link *l_ptr)
/**
* link_timeout - handle expiration of link timer
* @l_ptr: pointer to link
- *
- * This routine must not grab "tipc_net_lock" to avoid a potential deadlock conflict
- * with tipc_link_delete(). (There is no risk that the node will be deleted by
- * another thread because tipc_link_delete() always cancels the link timer before
- * tipc_node_delete() is called.)
*/
static void link_timeout(struct tipc_link *l_ptr)
{
@@ -213,8 +208,8 @@ static void link_set_timer(struct tipc_link *l_ptr, u32 time)
* Returns pointer to link.
*/
struct tipc_link *tipc_link_create(struct tipc_node *n_ptr,
- struct tipc_bearer *b_ptr,
- const struct tipc_media_addr *media_addr)
+ struct tipc_bearer *b_ptr,
+ const struct tipc_media_addr *media_addr)
{
struct tipc_link *l_ptr;
struct tipc_msg *msg;
@@ -279,47 +274,32 @@ struct tipc_link *tipc_link_create(struct tipc_node *n_ptr,
k_init_timer(&l_ptr->timer, (Handler)link_timeout,
(unsigned long)l_ptr);
- list_add_tail(&l_ptr->link_list, &b_ptr->links);
link_state_event(l_ptr, STARTING_EVT);
return l_ptr;
}
-/**
- * tipc_link_delete - delete a link
- * @l_ptr: pointer to link
- *
- * Note: 'tipc_net_lock' is write_locked, bearer is locked.
- * This routine must not grab the node lock until after link timer cancellation
- * to avoid a potential deadlock situation.
- */
-void tipc_link_delete(struct tipc_link *l_ptr)
-{
- if (!l_ptr) {
- pr_err("Attempt to delete non-existent link\n");
- return;
- }
-
- k_cancel_timer(&l_ptr->timer);
- tipc_node_lock(l_ptr->owner);
- tipc_link_reset(l_ptr);
- tipc_node_detach_link(l_ptr->owner, l_ptr);
- tipc_link_purge_queues(l_ptr);
- list_del_init(&l_ptr->link_list);
- tipc_node_unlock(l_ptr->owner);
- k_term_timer(&l_ptr->timer);
- kfree(l_ptr);
-}
-
-void tipc_link_delete_list(struct tipc_bearer *b_ptr)
+void tipc_link_delete_list(unsigned int bearer_id)
{
struct tipc_link *l_ptr;
- struct tipc_link *temp_l_ptr;
+ struct tipc_node *n_ptr;
- list_for_each_entry_safe(l_ptr, temp_l_ptr, &b_ptr->links, link_list) {
- tipc_link_delete(l_ptr);
+ list_for_each_entry(n_ptr, &tipc_node_list, list) {
+ spin_lock_bh(&n_ptr->lock);
+ l_ptr = n_ptr->links[bearer_id];
+ if (l_ptr) {
+ tipc_link_reset(l_ptr);
+ tipc_node_detach_link(n_ptr, l_ptr);
+ spin_unlock_bh(&n_ptr->lock);
+
+ /* Nobody else can access this link now: */
+ del_timer_sync(&l_ptr->timer);
+ kfree(l_ptr);
+ continue;
+ }
+ spin_unlock_bh(&n_ptr->lock);
}
}
@@ -470,15 +450,16 @@ void tipc_link_reset(struct tipc_link *l_ptr)
link_reset_statistics(l_ptr);
}
-void tipc_link_reset_list(struct tipc_bearer *b_ptr)
+void tipc_link_reset_list(unsigned int bearer_id)
{
struct tipc_link *l_ptr;
+ struct tipc_node *n_ptr;
- list_for_each_entry(l_ptr, &b_ptr->links, link_list) {
- struct tipc_node *n_ptr = l_ptr->owner;
-
+ list_for_each_entry(n_ptr, &tipc_node_list, list) {
spin_lock_bh(&n_ptr->lock);
- tipc_link_reset(l_ptr);
+ l_ptr = n_ptr->links[bearer_id];
+ if (l_ptr)
+ tipc_link_reset(l_ptr);
spin_unlock_bh(&n_ptr->lock);
}
}
diff --git a/net/tipc/link.h b/net/tipc/link.h
index a900e74..3340fc1 100644
--- a/net/tipc/link.h
+++ b/net/tipc/link.h
@@ -59,6 +59,7 @@
/* Link endpoint execution states
*/
#define LINK_STARTED 0x0001
+#define LINK_STOPPED 0x0002
/* Starting value for maximum packet size negotiation on unicast links
* (unless bearer MTU is less)
@@ -102,7 +103,6 @@ struct tipc_stats {
* @media_addr: media address to use when sending messages over link
* @timer: link timer
* @owner: pointer to peer node
- * @link_list: adjacent links in bearer's list of links
* @flags: execution state flags for link endpoint instance
* @checkpoint: reference point for triggering link continuity checking
* @peer_session: link session # being used by peer end of link
@@ -149,7 +149,6 @@ struct tipc_link {
struct tipc_media_addr media_addr;
struct timer_list timer;
struct tipc_node *owner;
- struct list_head link_list;
/* Management and link supervision data */
unsigned int flags;
@@ -215,11 +214,10 @@ struct tipc_port;
struct tipc_link *tipc_link_create(struct tipc_node *n_ptr,
struct tipc_bearer *b_ptr,
const struct tipc_media_addr *media_addr);
-void tipc_link_delete(struct tipc_link *l_ptr);
+void tipc_link_delete_list(unsigned int bearer_id);
void tipc_link_failover_send_queue(struct tipc_link *l_ptr);
void tipc_link_dup_send_queue(struct tipc_link *l_ptr,
struct tipc_link *dest);
-void tipc_link_delete_list(struct tipc_bearer *b_ptr);
void tipc_link_reset_fragments(struct tipc_link *l_ptr);
int tipc_link_is_up(struct tipc_link *l_ptr);
int tipc_link_is_active(struct tipc_link *l_ptr);
@@ -232,7 +230,7 @@ struct sk_buff *tipc_link_cmd_show_stats(const void *req_tlv_area,
struct sk_buff *tipc_link_cmd_reset_stats(const void *req_tlv_area,
int req_tlv_space);
void tipc_link_reset(struct tipc_link *l_ptr);
-void tipc_link_reset_list(struct tipc_bearer *b_ptr);
+void tipc_link_reset_list(unsigned int bearer_id);
int tipc_link_send(struct sk_buff *buf, u32 dest, u32 selector);
void tipc_link_send_names(struct list_head *message_list, u32 dest);
int tipc_link_send_buf(struct tipc_link *l_ptr, struct sk_buff *buf);
--
1.7.9.5
^ permalink raw reply related
* [PATCH net-next v2 02/14] tipc: move code for resetting links from bearer.c to link.c
From: Jon Maloy @ 2014-02-13 22:29 UTC (permalink / raw)
To: davem
Cc: netdev, Paul Gortmaker, erik.hugne, ying.xue, maloy,
tipc-discussion, Jon Maloy
In-Reply-To: <1392330558-19048-1-git-send-email-jon.maloy@ericsson.com>
From: Ying Xue <ying.xue@windriver.com>
We break out the code for resetting attached links in the
function tipc_reset_bearer(), and define a new function named
tipc_link_reset_list() to do this job.
This commit incurs no functional changes, but makes the code
of function tipc_reset_bearer() cleaner. It is also a preparation
for a more important change to the bearer code, in a subsequent
commit in this series.
Signed-off-by: Ying Xue <ying.xue@windriver.com>
Reviewed-by: Paul Gortmaker <paul.gortmaker@windriver.com>
Signed-off-by: Jon Maloy <jon.maloy@ericsson.com>
---
net/tipc/bearer.c | 11 +----------
net/tipc/link.c | 12 ++++++++++++
net/tipc/link.h | 1 +
3 files changed, 14 insertions(+), 10 deletions(-)
diff --git a/net/tipc/bearer.c b/net/tipc/bearer.c
index a38c899..a3bdf5c 100644
--- a/net/tipc/bearer.c
+++ b/net/tipc/bearer.c
@@ -350,19 +350,10 @@ exit:
*/
static int tipc_reset_bearer(struct tipc_bearer *b_ptr)
{
- struct tipc_link *l_ptr;
- struct tipc_link *temp_l_ptr;
-
read_lock_bh(&tipc_net_lock);
pr_info("Resetting bearer <%s>\n", b_ptr->name);
spin_lock_bh(&b_ptr->lock);
- list_for_each_entry_safe(l_ptr, temp_l_ptr, &b_ptr->links, link_list) {
- struct tipc_node *n_ptr = l_ptr->owner;
-
- spin_lock_bh(&n_ptr->lock);
- tipc_link_reset(l_ptr);
- spin_unlock_bh(&n_ptr->lock);
- }
+ tipc_link_reset_list(b_ptr);
spin_unlock_bh(&b_ptr->lock);
read_unlock_bh(&tipc_net_lock);
return 0;
diff --git a/net/tipc/link.c b/net/tipc/link.c
index 17fbd15..3ff34e8 100644
--- a/net/tipc/link.c
+++ b/net/tipc/link.c
@@ -461,6 +461,18 @@ void tipc_link_reset(struct tipc_link *l_ptr)
link_reset_statistics(l_ptr);
}
+void tipc_link_reset_list(struct tipc_bearer *b_ptr)
+{
+ struct tipc_link *l_ptr;
+
+ list_for_each_entry(l_ptr, &b_ptr->links, link_list) {
+ struct tipc_node *n_ptr = l_ptr->owner;
+
+ spin_lock_bh(&n_ptr->lock);
+ tipc_link_reset(l_ptr);
+ spin_unlock_bh(&n_ptr->lock);
+ }
+}
static void link_activate(struct tipc_link *l_ptr)
{
diff --git a/net/tipc/link.h b/net/tipc/link.h
index 8addc5e..73beecb 100644
--- a/net/tipc/link.h
+++ b/net/tipc/link.h
@@ -231,6 +231,7 @@ struct sk_buff *tipc_link_cmd_show_stats(const void *req_tlv_area,
struct sk_buff *tipc_link_cmd_reset_stats(const void *req_tlv_area,
int req_tlv_space);
void tipc_link_reset(struct tipc_link *l_ptr);
+void tipc_link_reset_list(struct tipc_bearer *b_ptr);
int tipc_link_send(struct sk_buff *buf, u32 dest, u32 selector);
void tipc_link_send_names(struct list_head *message_list, u32 dest);
int tipc_link_send_buf(struct tipc_link *l_ptr, struct sk_buff *buf);
--
1.7.9.5
^ permalink raw reply related
* [PATCH net-next v2 01/14] tipc: stricter behavior of message reassembly function
From: Jon Maloy @ 2014-02-13 22:29 UTC (permalink / raw)
To: davem
Cc: netdev, Paul Gortmaker, erik.hugne, ying.xue, maloy,
tipc-discussion, Jon Maloy
In-Reply-To: <1392330558-19048-1-git-send-email-jon.maloy@ericsson.com>
The function tipc_link_recv_fragment(struct sk_buff **buf) currently
leaves the value of the input buffer pointer undefined when it returns,
except when the return code indicates that the reassembly is complete.
This despite the fact that it always consumes the input buffer.
Here, we enforce a stricter behavior by this function, ensuring that
the returned buffer pointer is non-NULL if and only if the reassembly
is complete. This makes it possible to test for the buffer pointer as
criteria for successful reassembly.
We also rename the function to tipc_link_frag_rcv(), which is both
shorter and more in line with common naming practice in the network
subsystem.
Apart from the new name, these changes have no impact on current
users of the function, but makes it more practical for use in some
planned future commits.
Signed-off-by: Jon Maloy <jon.maloy@ericsson.com>
Reviewed-by: Ying Xue <ying.xue@windriver.com>
---
net/tipc/bcast.c | 6 +++---
net/tipc/link.c | 16 +++++++++-------
net/tipc/link.h | 6 +++---
3 files changed, 15 insertions(+), 13 deletions(-)
diff --git a/net/tipc/bcast.c b/net/tipc/bcast.c
index bf860d9..af35f76 100644
--- a/net/tipc/bcast.c
+++ b/net/tipc/bcast.c
@@ -481,9 +481,9 @@ receive:
tipc_link_recv_bundle(buf);
} else if (msg_user(msg) == MSG_FRAGMENTER) {
int ret;
- ret = tipc_link_recv_fragment(&node->bclink.reasm_head,
- &node->bclink.reasm_tail,
- &buf);
+ ret = tipc_link_frag_rcv(&node->bclink.reasm_head,
+ &node->bclink.reasm_tail,
+ &buf);
if (ret == LINK_REASM_ERROR)
goto unlock;
spin_lock_bh(&bc_lock);
diff --git a/net/tipc/link.c b/net/tipc/link.c
index d4b5de4..17fbd15 100644
--- a/net/tipc/link.c
+++ b/net/tipc/link.c
@@ -1584,9 +1584,9 @@ deliver:
continue;
case MSG_FRAGMENTER:
l_ptr->stats.recv_fragments++;
- ret = tipc_link_recv_fragment(&l_ptr->reasm_head,
- &l_ptr->reasm_tail,
- &buf);
+ ret = tipc_link_frag_rcv(&l_ptr->reasm_head,
+ &l_ptr->reasm_tail,
+ &buf);
if (ret == LINK_REASM_COMPLETE) {
l_ptr->stats.recv_fragmented++;
msg = buf_msg(buf);
@@ -2277,12 +2277,11 @@ static int link_send_long_buf(struct tipc_link *l_ptr, struct sk_buff *buf)
return dsz;
}
-/*
- * tipc_link_recv_fragment(): Called with node lock on. Returns
+/* tipc_link_frag_rcv(): Called with node lock on. Returns
* the reassembled buffer if message is complete.
*/
-int tipc_link_recv_fragment(struct sk_buff **head, struct sk_buff **tail,
- struct sk_buff **fbuf)
+int tipc_link_frag_rcv(struct sk_buff **head, struct sk_buff **tail,
+ struct sk_buff **fbuf)
{
struct sk_buff *frag = *fbuf;
struct tipc_msg *msg = buf_msg(frag);
@@ -2296,6 +2295,7 @@ int tipc_link_recv_fragment(struct sk_buff **head, struct sk_buff **tail,
goto out_free;
*head = frag;
skb_frag_list_init(*head);
+ *fbuf = NULL;
return 0;
} else if (*head &&
skb_try_coalesce(*head, frag, &headstolen, &delta)) {
@@ -2315,10 +2315,12 @@ int tipc_link_recv_fragment(struct sk_buff **head, struct sk_buff **tail,
*tail = *head = NULL;
return LINK_REASM_COMPLETE;
}
+ *fbuf = NULL;
return 0;
out_free:
pr_warn_ratelimited("Link unable to reassemble fragmented message\n");
kfree_skb(*fbuf);
+ *fbuf = NULL;
return LINK_REASM_ERROR;
}
diff --git a/net/tipc/link.h b/net/tipc/link.h
index 3b6aa65..8addc5e 100644
--- a/net/tipc/link.h
+++ b/net/tipc/link.h
@@ -239,9 +239,9 @@ int tipc_link_send_sections_fast(struct tipc_port *sender,
struct iovec const *msg_sect,
unsigned int len, u32 destnode);
void tipc_link_recv_bundle(struct sk_buff *buf);
-int tipc_link_recv_fragment(struct sk_buff **reasm_head,
- struct sk_buff **reasm_tail,
- struct sk_buff **fbuf);
+int tipc_link_frag_rcv(struct sk_buff **reasm_head,
+ struct sk_buff **reasm_tail,
+ struct sk_buff **fbuf);
void tipc_link_send_proto_msg(struct tipc_link *l_ptr, u32 msg_typ, int prob,
u32 gap, u32 tolerance, u32 priority,
u32 acked_mtu);
--
1.7.9.5
^ permalink raw reply related
* [PATCH net-next v2 00/14] tipc: clean up media and bearer layer
From: Jon Maloy @ 2014-02-13 22:29 UTC (permalink / raw)
To: davem
Cc: netdev, Paul Gortmaker, erik.hugne, ying.xue, maloy,
tipc-discussion, Jon Maloy
This commit series aims at facilitating future changes to the
locking policy around nodes, links and bearers.
Currently, we have a big read/write lock (net_lock) that is used for
serializing all changes to the node, link and bearer lists, as well
as to their mutual pointers and references.
But, in order to allow for concurrent access to the contents of these
structures, net_lock is only used in read mode by the data path code,
and hence a finer granular locking policy must be applied inside the
scope of net_lock: a spinlock (node_lock) for each node structure,
and another one (bearer_lock) for protection of bearer structures.
This locking policy has proved hard to maintain. We have several
times encountered contention problems between node_lock and
bearer_lock, and with the advent of the RCU locking mechanism we
feel it is anyway obsolete and ripe for improvements.
We now plan to replace net_lock with an RCU lock, as well as
getting rid of bearer_lock altogether. This will both reduce data
path overhead and make the code more manageable, while reducing the
risk of future lock contention problems.
Prior to these changes, we need to do some necessary cleanup and
code consolidation. This is what we do with this commit series,
before we finally remove bearer_lock. In a later series we will
replace net_lock with an RCU lock.
v2:
- Re-inserted a removed kerneldoc entry in commit#5, based on
feedback from D. Miller.
Jon Maloy (9):
tipc: stricter behavior of message reassembly function
tipc: change reception of tunnelled duplicate packets
tipc: change reception of tunnelled failover packets
tipc: change signature of tunnelling reception function
tipc: more cleanup of tunnelling reception function
tipc: rename stack variables in function tipc_link_tunnel_rcv
tipc: changes to general packet reception algorithm
tipc: delay delete of link when failover is needed
tipc: add node_lock protection to link lookup function
Ying Xue (5):
tipc: move code for resetting links from bearer.c to link.c
tipc: move code for deleting links from bearer.c to link.c
tipc: redefine 'started' flag in struct link to bitmap
tipc: remove 'links' list from tipc_bearer struct
tipc: remove bearer_lock from tipc_bearer struct
net/tipc/bcast.c | 7 +-
net/tipc/bearer.c | 42 ++----
net/tipc/bearer.h | 7 +-
net/tipc/core.c | 2 +-
net/tipc/link.c | 432 ++++++++++++++++++++++++++++++-----------------------
net/tipc/link.h | 34 ++---
net/tipc/node.c | 8 +-
7 files changed, 283 insertions(+), 249 deletions(-)
--
1.7.9.5
^ permalink raw reply
* [PATCH net-next v2 13/14] tipc: remove bearer_lock from tipc_bearer struct
From: Jon Maloy @ 2014-02-13 22:29 UTC (permalink / raw)
To: davem; +Cc: Jon Maloy, netdev, tipc-discussion
In-Reply-To: <1392330558-19048-1-git-send-email-jon.maloy@ericsson.com>
From: Ying Xue <ying.xue@windriver.com>
After the earlier commits ("tipc: remove 'links' list from
tipc_bearer struct") and ("tipc: introduce new spinlock to protect
struct link_req"), there is no longer any need to protect struct
link_req or or any link list by use of bearer_lock. Furthermore,
we have eliminated the need for using bearer_lock during downcalls
(send) from the link to the bearer, since we have ensured that
bearers always have a longer life cycle that their associated links,
and always contain valid data.
So, the only need now for a lock protecting bearers is for guaranteeing
consistency of the bearer list itself. For this, it is sufficient, at
least for the time being, to continue applying 'net_lock´ in write mode.
By removing bearer_lock we also pre-empt introduction of issue b) descibed
in the previous commit "tipc: remove 'links' list from tipc_bearer struct":
"b) When the outer protection from net_lock is gone, taking
bearer_lock and node_lock in opposite order of method 1) and 2)
will become an obvious deadlock hazard".
Therefore, we now eliminate the bearer_lock spinlock.
Signed-off-by: Ying Xue <ying.xue@windriver.com>
Reviewed-by: Paul Gortmaker <paul.gortmaker@windriver.com>
Signed-off-by: Jon Maloy <jon.maloy@ericsson.com>
---
net/tipc/bcast.c | 1 -
net/tipc/bearer.c | 16 +++-------------
net/tipc/bearer.h | 5 +----
3 files changed, 4 insertions(+), 18 deletions(-)
diff --git a/net/tipc/bcast.c b/net/tipc/bcast.c
index af35f76..06a639c 100644
--- a/net/tipc/bcast.c
+++ b/net/tipc/bcast.c
@@ -785,7 +785,6 @@ void tipc_bclink_init(void)
bcl->owner = &bclink->node;
bcl->max_pkt = MAX_PKT_DEFAULT_MCAST;
tipc_link_set_queue_limits(bcl, BCLINK_WIN_DEFAULT);
- spin_lock_init(&bcbearer->bearer.lock);
bcl->b_ptr = &bcbearer->bearer;
bcl->state = WORKING_WORKING;
strlcpy(bcl->name, tipc_bclink_name, TIPC_MAX_LINK_NAME);
diff --git a/net/tipc/bearer.c b/net/tipc/bearer.c
index 60caa45..242cddd 100644
--- a/net/tipc/bearer.c
+++ b/net/tipc/bearer.c
@@ -327,7 +327,6 @@ restart:
b_ptr->net_plane = bearer_id + 'A';
b_ptr->active = 1;
b_ptr->priority = priority;
- spin_lock_init(&b_ptr->lock);
res = tipc_disc_create(b_ptr, &b_ptr->bcast_addr, disc_domain);
if (res) {
@@ -351,9 +350,7 @@ static int tipc_reset_bearer(struct tipc_bearer *b_ptr)
{
read_lock_bh(&tipc_net_lock);
pr_info("Resetting bearer <%s>\n", b_ptr->name);
- spin_lock_bh(&b_ptr->lock);
tipc_link_reset_list(b_ptr->identity);
- spin_unlock_bh(&b_ptr->lock);
read_unlock_bh(&tipc_net_lock);
return 0;
}
@@ -365,19 +362,12 @@ static int tipc_reset_bearer(struct tipc_bearer *b_ptr)
*/
static void bearer_disable(struct tipc_bearer *b_ptr, bool shutting_down)
{
- struct tipc_link_req *temp_req;
-
pr_info("Disabling bearer <%s>\n", b_ptr->name);
- spin_lock_bh(&b_ptr->lock);
b_ptr->media->disable_media(b_ptr);
- tipc_link_delete_list(b_ptr->identity, shutting_down);
- temp_req = b_ptr->link_req;
- b_ptr->link_req = NULL;
- spin_unlock_bh(&b_ptr->lock);
-
- if (temp_req)
- tipc_disc_delete(temp_req);
+ tipc_link_delete_list(b_ptr->identity, shutting_down);
+ if (b_ptr->link_req)
+ tipc_disc_delete(b_ptr->link_req);
memset(b_ptr, 0, sizeof(struct tipc_bearer));
}
diff --git a/net/tipc/bearer.h b/net/tipc/bearer.h
index 647cb1d..425dd81 100644
--- a/net/tipc/bearer.h
+++ b/net/tipc/bearer.h
@@ -107,10 +107,8 @@ struct tipc_media {
/**
* struct tipc_bearer - Generic TIPC bearer structure
- * @dev: ptr to associated network device
- * @usr_handle: pointer to additional media-specific information about bearer
+ * @media_ptr: pointer to additional media-specific information about bearer
* @mtu: max packet size bearer can support
- * @lock: spinlock for controlling access to bearer
* @addr: media-specific address associated with bearer
* @name: bearer name (format = media:interface)
* @media: ptr to media structure associated with bearer
@@ -133,7 +131,6 @@ struct tipc_bearer {
u32 mtu; /* initalized by media */
struct tipc_media_addr addr; /* initalized by media */
char name[TIPC_MAX_BEARER_NAME];
- spinlock_t lock;
struct tipc_media *media;
struct tipc_media_addr bcast_addr;
u32 priority;
--
1.7.9.5
------------------------------------------------------------------------------
Android apps run on BlackBerry 10
Introducing the new BlackBerry 10.2.1 Runtime for Android apps.
Now with support for Jelly Bean, Bluetooth, Mapview and more.
Get your Android app in front of a whole new audience. Start now.
http://pubads.g.doubleclick.net/gampad/clk?id=124407151&iu=/4140/ostg.clktrk
_______________________________________________
tipc-discussion mailing list
tipc-discussion@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/tipc-discussion
^ permalink raw reply related
page: next (older) | prev (newer) | latest
- recent:[subjects (threaded)|topics (new)|topics (active)]
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox