* [PATCH 16/17] mlx4_en: Using blue flame support
From: Yevgeny Petrilin @ 2011-03-07 10:18 UTC (permalink / raw)
To: davem; +Cc: netdev, yevgenyp
Doorbell is used according to usage of BlueFlame.
For Blue Flame to work in Ethernet mode QP number should have 0
at bits 6,7.
Allocating range of QPs accordingly.
Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il>
---
drivers/net/mlx4/en_netdev.c | 10 +++++-
drivers/net/mlx4/en_tx.c | 72 ++++++++++++++++++++++++++++++-----------
drivers/net/mlx4/mlx4_en.h | 4 ++-
3 files changed, 64 insertions(+), 22 deletions(-)
diff --git a/drivers/net/mlx4/en_netdev.c b/drivers/net/mlx4/en_netdev.c
index 4a30047..78717aa 100644
--- a/drivers/net/mlx4/en_netdev.c
+++ b/drivers/net/mlx4/en_netdev.c
@@ -931,6 +931,13 @@ int mlx4_en_alloc_resources(struct mlx4_en_priv *priv)
{
struct mlx4_en_port_profile *prof = priv->prof;
int i;
+ int base_tx_qpn, err;
+
+ err = mlx4_qp_reserve_range(priv->mdev->dev, priv->tx_ring_num, 256, &base_tx_qpn);
+ if (err) {
+ en_err(priv, "failed reserving range for TX rings\n");
+ return err;
+ }
/* Create tx Rings */
for (i = 0; i < priv->tx_ring_num; i++) {
@@ -938,7 +945,7 @@ int mlx4_en_alloc_resources(struct mlx4_en_priv *priv)
prof->tx_ring_size, i, TX))
goto err;
- if (mlx4_en_create_tx_ring(priv, &priv->tx_ring[i],
+ if (mlx4_en_create_tx_ring(priv, &priv->tx_ring[i], base_tx_qpn + i,
prof->tx_ring_size, TXBB_SIZE))
goto err;
}
@@ -958,6 +965,7 @@ int mlx4_en_alloc_resources(struct mlx4_en_priv *priv)
err:
en_err(priv, "Failed to allocate NIC resources\n");
+ mlx4_qp_release_range(priv->mdev->dev, base_tx_qpn, priv->tx_ring_num);
return -ENOMEM;
}
diff --git a/drivers/net/mlx4/en_tx.c b/drivers/net/mlx4/en_tx.c
index a680cd4..01feb8f 100644
--- a/drivers/net/mlx4/en_tx.c
+++ b/drivers/net/mlx4/en_tx.c
@@ -44,6 +44,7 @@
enum {
MAX_INLINE = 104, /* 128 - 16 - 4 - 4 */
+ MAX_BF = 256,
};
static int inline_thold __read_mostly = MAX_INLINE;
@@ -52,7 +53,7 @@ module_param_named(inline_thold, inline_thold, int, 0444);
MODULE_PARM_DESC(inline_thold, "threshold for using inline data");
int mlx4_en_create_tx_ring(struct mlx4_en_priv *priv,
- struct mlx4_en_tx_ring *ring, u32 size,
+ struct mlx4_en_tx_ring *ring, int qpn, u32 size,
u16 stride)
{
struct mlx4_en_dev *mdev = priv->mdev;
@@ -103,23 +104,25 @@ int mlx4_en_create_tx_ring(struct mlx4_en_priv *priv,
"buf_size:%d dma:%llx\n", ring, ring->buf, ring->size,
ring->buf_size, (unsigned long long) ring->wqres.buf.direct.map);
- err = mlx4_qp_reserve_range(mdev->dev, 1, 1, &ring->qpn);
- if (err) {
- en_err(priv, "Failed reserving qp for tx ring.\n");
- goto err_map;
- }
-
+ ring->qpn = qpn;
err = mlx4_qp_alloc(mdev->dev, ring->qpn, &ring->qp);
if (err) {
en_err(priv, "Failed allocating qp %d\n", ring->qpn);
- goto err_reserve;
+ goto err_map;
}
ring->qp.event = mlx4_en_sqp_event;
+ err = mlx4_bf_alloc(mdev->dev, &ring->bf);
+ if (err) {
+ en_dbg(DRV, priv, "working without blueflame (%d)", err);
+ ring->bf.uar = &mdev->priv_uar;
+ ring->bf.uar->map = mdev->uar_map;
+ ring->bf_enabled = false;
+ } else
+ ring->bf_enabled = true;
+
return 0;
-err_reserve:
- mlx4_qp_release_range(mdev->dev, ring->qpn, 1);
err_map:
mlx4_en_unmap_buffer(&ring->wqres.buf);
err_hwq_res:
@@ -139,6 +142,8 @@ void mlx4_en_destroy_tx_ring(struct mlx4_en_priv *priv,
struct mlx4_en_dev *mdev = priv->mdev;
en_dbg(DRV, priv, "Destroying tx ring, qpn: %d\n", ring->qpn);
+ if (ring->bf_enabled)
+ mlx4_bf_free(mdev->dev, &ring->bf);
mlx4_qp_remove(mdev->dev, &ring->qp);
mlx4_qp_free(mdev->dev, &ring->qp);
mlx4_qp_release_range(mdev->dev, ring->qpn, 1);
@@ -171,6 +176,8 @@ int mlx4_en_activate_tx_ring(struct mlx4_en_priv *priv,
mlx4_en_fill_qp_context(priv, ring->size, ring->stride, 1, 0, ring->qpn,
ring->cqn, &ring->context);
+ if (ring->bf_enabled)
+ ring->context.usr_page = cpu_to_be32(ring->bf.uar->index);
err = mlx4_qp_to_ready(mdev->dev, &ring->wqres.mtt, &ring->context,
&ring->qp, &ring->qp_state);
@@ -591,6 +598,11 @@ u16 mlx4_en_select_queue(struct net_device *dev, struct sk_buff *skb)
return skb_tx_hash(dev, skb);
}
+static void mlx4_bf_copy(unsigned long *dst, unsigned long *src, unsigned bytecnt)
+{
+ __iowrite64_copy(dst, src, bytecnt / 8);
+}
+
netdev_tx_t mlx4_en_xmit(struct sk_buff *skb, struct net_device *dev)
{
struct mlx4_en_priv *priv = netdev_priv(dev);
@@ -609,12 +621,13 @@ netdev_tx_t mlx4_en_xmit(struct sk_buff *skb, struct net_device *dev)
int desc_size;
int real_size;
dma_addr_t dma;
- u32 index;
+ u32 index, bf_index;
__be32 op_own;
u16 vlan_tag = 0;
int i;
int lso_header_size;
void *fragptr;
+ bool bounce = false;
if (!priv->port_up)
goto tx_drop;
@@ -657,13 +670,16 @@ netdev_tx_t mlx4_en_xmit(struct sk_buff *skb, struct net_device *dev)
/* Packet is good - grab an index and transmit it */
index = ring->prod & ring->size_mask;
+ bf_index = ring->prod;
/* See if we have enough space for whole descriptor TXBB for setting
* SW ownership on next descriptor; if not, use a bounce buffer. */
if (likely(index + nr_txbb <= ring->size))
tx_desc = ring->buf + index * TXBB_SIZE;
- else
+ else {
tx_desc = (struct mlx4_en_tx_desc *) ring->bounce_buf;
+ bounce = true;
+ }
/* Save skb in tx_info ring */
tx_info = &ring->tx_info[index];
@@ -768,21 +784,37 @@ netdev_tx_t mlx4_en_xmit(struct sk_buff *skb, struct net_device *dev)
ring->prod += nr_txbb;
/* If we used a bounce buffer then copy descriptor back into place */
- if (tx_desc == (struct mlx4_en_tx_desc *) ring->bounce_buf)
+ if (bounce)
tx_desc = mlx4_en_bounce_to_desc(priv, ring, index, desc_size);
/* Run destructor before passing skb to HW */
if (likely(!skb_shared(skb)))
skb_orphan(skb);
- /* Ensure new descirptor hits memory
- * before setting ownership of this descriptor to HW */
- wmb();
- tx_desc->ctrl.owner_opcode = op_own;
+ if (ring->bf_enabled && desc_size <= MAX_BF && !bounce && !vlan_tag) {
+ *(u32 *) (&tx_desc->ctrl.vlan_tag) |= ring->doorbell_qpn;
+ op_own |= htonl((bf_index & 0xffff) << 8);
+ /* Ensure new descirptor hits memory
+ * before setting ownership of this descriptor to HW */
+ wmb();
+ tx_desc->ctrl.owner_opcode = op_own;
- /* Ring doorbell! */
- wmb();
- writel(ring->doorbell_qpn, mdev->uar_map + MLX4_SEND_DOORBELL);
+ wmb();
+
+ mlx4_bf_copy(ring->bf.reg + ring->bf.offset, (unsigned long *) &tx_desc->ctrl,
+ desc_size);
+
+ wmb();
+
+ ring->bf.offset ^= ring->bf.buf_size;
+ } else {
+ /* Ensure new descirptor hits memory
+ * before setting ownership of this descriptor to HW */
+ wmb();
+ tx_desc->ctrl.owner_opcode = op_own;
+ wmb();
+ writel(ring->doorbell_qpn, ring->bf.uar->map + MLX4_SEND_DOORBELL);
+ }
/* Poll CQ here */
mlx4_en_xmit_poll(priv, tx_ind);
diff --git a/drivers/net/mlx4/mlx4_en.h b/drivers/net/mlx4/mlx4_en.h
index 5a2c560..edcb535 100644
--- a/drivers/net/mlx4/mlx4_en.h
+++ b/drivers/net/mlx4/mlx4_en.h
@@ -248,6 +248,8 @@ struct mlx4_en_tx_ring {
unsigned long bytes;
unsigned long packets;
spinlock_t comp_lock;
+ struct mlx4_bf bf;
+ bool bf_enabled;
};
struct mlx4_en_rx_desc {
@@ -518,7 +520,7 @@ u16 mlx4_en_select_queue(struct net_device *dev, struct sk_buff *skb);
netdev_tx_t mlx4_en_xmit(struct sk_buff *skb, struct net_device *dev);
int mlx4_en_create_tx_ring(struct mlx4_en_priv *priv, struct mlx4_en_tx_ring *ring,
- u32 size, u16 stride);
+ int qpn, u32 size, u16 stride);
void mlx4_en_destroy_tx_ring(struct mlx4_en_priv *priv, struct mlx4_en_tx_ring *ring);
int mlx4_en_activate_tx_ring(struct mlx4_en_priv *priv,
struct mlx4_en_tx_ring *ring,
--
1.6.0.2
^ permalink raw reply related
* [PATCH 13/17] mlx4: Add blue flame support for kernel consumers
From: Yevgeny Petrilin @ 2011-03-07 10:17 UTC (permalink / raw)
To: davem; +Cc: netdev, yevgenyp, eli
From: Eli Cohen <eli@mellanox.co.il>
Using blue flame can improve latency by allowing the HW to more efficiently
access the WQE. This patch presents two functions that are used to allocate or
release HW resources for using blue flame; the caller need to supply a struct
mlx4_bf object when allocating resources. Consumers that make use of this API
should post doorbells to the UAR object pointed by the initialized struct
mlx4_bf;
Signed-off-by: Eli Cohen <eli@mellanox.co.il>
Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il>
---
drivers/net/mlx4/main.c | 31 ++++++++++++++
drivers/net/mlx4/mlx4.h | 3 +
drivers/net/mlx4/pd.c | 94 +++++++++++++++++++++++++++++++++++++++++++
include/linux/mlx4/device.h | 13 ++++++
include/linux/mlx4/qp.h | 1 +
5 files changed, 142 insertions(+), 0 deletions(-)
diff --git a/drivers/net/mlx4/main.c b/drivers/net/mlx4/main.c
index c8e2761..34581d2 100644
--- a/drivers/net/mlx4/main.c
+++ b/drivers/net/mlx4/main.c
@@ -39,6 +39,7 @@
#include <linux/pci.h>
#include <linux/dma-mapping.h>
#include <linux/slab.h>
+#include <linux/io-mapping.h>
#include <linux/mlx4/device.h>
#include <linux/mlx4/doorbell.h>
@@ -721,8 +722,31 @@ static void mlx4_free_icms(struct mlx4_dev *dev)
mlx4_free_icm(dev, priv->fw.aux_icm, 0);
}
+static int map_bf_area(struct mlx4_dev *dev)
+{
+ struct mlx4_priv *priv = mlx4_priv(dev);
+ resource_size_t bf_start;
+ resource_size_t bf_len;
+ int err = 0;
+
+ bf_start = pci_resource_start(dev->pdev, 2) + (dev->caps.num_uars << PAGE_SHIFT);
+ bf_len = pci_resource_len(dev->pdev, 2) - (dev->caps.num_uars << PAGE_SHIFT);
+ priv->bf_mapping = io_mapping_create_wc(bf_start, bf_len);
+ if (!priv->bf_mapping)
+ err = -ENOMEM;
+
+ return err;
+}
+
+static void unmap_bf_area(struct mlx4_dev *dev)
+{
+ if (mlx4_priv(dev)->bf_mapping)
+ io_mapping_free(mlx4_priv(dev)->bf_mapping);
+}
+
static void mlx4_close_hca(struct mlx4_dev *dev)
{
+ unmap_bf_area(dev);
mlx4_CLOSE_HCA(dev, 0);
mlx4_free_icms(dev);
mlx4_UNMAP_FA(dev);
@@ -775,6 +799,9 @@ static int mlx4_init_hca(struct mlx4_dev *dev)
goto err_stop_fw;
}
+ if (map_bf_area(dev))
+ mlx4_dbg(dev, "Kernel support for blue flame is not available for kernels < 2.6.28\n");
+
init_hca.log_uar_sz = ilog2(dev->caps.num_uars);
err = mlx4_init_icm(dev, &dev_cap, &init_hca, icm_size);
@@ -805,6 +832,7 @@ err_free_icm:
mlx4_free_icms(dev);
err_stop_fw:
+ unmap_bf_area(dev);
mlx4_UNMAP_FA(dev);
mlx4_free_icm(dev, priv->fw.fw_icm, 0);
@@ -1196,6 +1224,9 @@ static int __mlx4_init_one(struct pci_dev *pdev, const struct pci_device_id *id)
pci_read_config_byte(pdev, PCI_REVISION_ID, &dev->rev_id);
+ INIT_LIST_HEAD(&priv->bf_list);
+ mutex_init(&priv->bf_mutex);
+
/*
* Now reset the HCA before we touch the PCI capabilities or
* attempt a firmware command, since a boot ROM may have left
diff --git a/drivers/net/mlx4/mlx4.h b/drivers/net/mlx4/mlx4.h
index a50923a..008bf95 100644
--- a/drivers/net/mlx4/mlx4.h
+++ b/drivers/net/mlx4/mlx4.h
@@ -353,6 +353,9 @@ struct mlx4_priv {
struct mutex port_mutex;
struct mlx4_msix_ctl msix_ctl;
struct mlx4_steer *steer;
+ struct list_head bf_list;
+ struct mutex bf_mutex;
+ struct io_mapping *bf_mapping;
};
static inline struct mlx4_priv *mlx4_priv(struct mlx4_dev *dev)
diff --git a/drivers/net/mlx4/pd.c b/drivers/net/mlx4/pd.c
index c4988d6..5210a0f 100644
--- a/drivers/net/mlx4/pd.c
+++ b/drivers/net/mlx4/pd.c
@@ -32,6 +32,7 @@
*/
#include <linux/errno.h>
+#include <linux/io-mapping.h>
#include <asm/page.h>
@@ -77,6 +78,7 @@ int mlx4_uar_alloc(struct mlx4_dev *dev, struct mlx4_uar *uar)
return -ENOMEM;
uar->pfn = (pci_resource_start(dev->pdev, 2) >> PAGE_SHIFT) + uar->index;
+ uar->map = NULL;
return 0;
}
@@ -88,6 +90,98 @@ void mlx4_uar_free(struct mlx4_dev *dev, struct mlx4_uar *uar)
}
EXPORT_SYMBOL_GPL(mlx4_uar_free);
+int mlx4_bf_alloc(struct mlx4_dev *dev, struct mlx4_bf *bf)
+{
+ struct mlx4_priv *priv = mlx4_priv(dev);
+ struct mlx4_uar *uar;
+ int err = 0;
+ int idx;
+
+ if (!priv->bf_mapping)
+ return -ENOMEM;
+
+ mutex_lock(&priv->bf_mutex);
+ if (!list_empty(&priv->bf_list))
+ uar = list_entry(priv->bf_list.next, struct mlx4_uar, bf_list);
+ else {
+ uar = kmalloc(sizeof *uar, GFP_KERNEL);
+ if (!uar) {
+ err = -ENOMEM;
+ goto out;
+ }
+ err = mlx4_uar_alloc(dev, uar);
+ if (err)
+ goto free_kmalloc;
+
+ uar->map = ioremap(uar->pfn << PAGE_SHIFT, PAGE_SIZE);
+ if (!uar->map) {
+ err = -ENOMEM;
+ goto free_uar;
+ }
+
+ uar->bf_map = io_mapping_map_wc(priv->bf_mapping, uar->index << PAGE_SHIFT);
+ if (!uar->bf_map) {
+ err = -ENOMEM;
+ goto unamp_uar;
+ }
+ uar->free_bf_bmap = 0;
+ list_add(&uar->bf_list, &priv->bf_list);
+ }
+
+ bf->uar = uar;
+ idx = ffz(uar->free_bf_bmap);
+ uar->free_bf_bmap |= 1 << idx;
+ bf->uar = uar;
+ bf->offset = 0;
+ bf->buf_size = dev->caps.bf_reg_size / 2;
+ bf->reg = uar->bf_map + idx * dev->caps.bf_reg_size;
+ if (uar->free_bf_bmap == (1 << dev->caps.bf_regs_per_page) - 1)
+ list_del_init(&uar->bf_list);
+
+ goto out;
+
+unamp_uar:
+ bf->uar = NULL;
+ iounmap(uar->map);
+
+free_uar:
+ mlx4_uar_free(dev, uar);
+
+free_kmalloc:
+ kfree(uar);
+
+out:
+ mutex_unlock(&priv->bf_mutex);
+ return err;
+}
+EXPORT_SYMBOL_GPL(mlx4_bf_alloc);
+
+void mlx4_bf_free(struct mlx4_dev *dev, struct mlx4_bf *bf)
+{
+ struct mlx4_priv *priv = mlx4_priv(dev);
+ int idx;
+
+ if (!bf->uar || !bf->uar->bf_map)
+ return;
+
+ mutex_lock(&priv->bf_mutex);
+ idx = (bf->reg - bf->uar->bf_map) / dev->caps.bf_reg_size;
+ bf->uar->free_bf_bmap &= ~(1 << idx);
+ if (!bf->uar->free_bf_bmap) {
+ if (!list_empty(&bf->uar->bf_list))
+ list_del(&bf->uar->bf_list);
+
+ io_mapping_unmap(bf->uar->bf_map);
+ iounmap(bf->uar->map);
+ mlx4_uar_free(dev, bf->uar);
+ kfree(bf->uar);
+ } else if (list_empty(&bf->uar->bf_list))
+ list_add(&bf->uar->bf_list, &priv->bf_list);
+
+ mutex_unlock(&priv->bf_mutex);
+}
+EXPORT_SYMBOL_GPL(mlx4_bf_free);
+
int mlx4_init_uar_table(struct mlx4_dev *dev)
{
if (dev->caps.num_uars <= 128) {
diff --git a/include/linux/mlx4/device.h b/include/linux/mlx4/device.h
index f1cb31b..229cd65 100644
--- a/include/linux/mlx4/device.h
+++ b/include/linux/mlx4/device.h
@@ -351,6 +351,17 @@ struct mlx4_fmr {
struct mlx4_uar {
unsigned long pfn;
int index;
+ struct list_head bf_list;
+ unsigned free_bf_bmap;
+ void __iomem *map;
+ void __iomem *bf_map;
+};
+
+struct mlx4_bf {
+ unsigned long offset;
+ int buf_size;
+ struct mlx4_uar *uar;
+ void __iomem *reg;
};
struct mlx4_cq {
@@ -478,6 +489,8 @@ void mlx4_pd_free(struct mlx4_dev *dev, u32 pdn);
int mlx4_uar_alloc(struct mlx4_dev *dev, struct mlx4_uar *uar);
void mlx4_uar_free(struct mlx4_dev *dev, struct mlx4_uar *uar);
+int mlx4_bf_alloc(struct mlx4_dev *dev, struct mlx4_bf *bf);
+void mlx4_bf_free(struct mlx4_dev *dev, struct mlx4_bf *bf);
int mlx4_mtt_init(struct mlx4_dev *dev, int npages, int page_shift,
struct mlx4_mtt *mtt);
diff --git a/include/linux/mlx4/qp.h b/include/linux/mlx4/qp.h
index 0eeb2a1..9e9eb21 100644
--- a/include/linux/mlx4/qp.h
+++ b/include/linux/mlx4/qp.h
@@ -303,6 +303,7 @@ struct mlx4_wqe_data_seg {
enum {
MLX4_INLINE_ALIGN = 64,
+ MLX4_INLINE_SEG = 1 << 31,
};
struct mlx4_wqe_inline_seg {
--
1.6.0.2
^ permalink raw reply related
* [PATCH 08/17] mlx4_en: Reporting HW revision in ethtool -i
From: Yevgeny Petrilin @ 2011-03-07 10:17 UTC (permalink / raw)
To: davem; +Cc: netdev, yevgenyp, eugenia
HW revision is derived from device ID and rev id.
Signed-off-by: Eugenia Emantayev <eugenia@mellanox.co.il>
Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il>
---
drivers/infiniband/hw/mlx4/main.c | 1 -
drivers/net/mlx4/en_ethtool.c | 15 ++++++++++++++-
drivers/net/mlx4/main.c | 2 ++
drivers/net/mlx4/mlx4_en.h | 3 +++
include/linux/mlx4/device.h | 2 +-
5 files changed, 20 insertions(+), 3 deletions(-)
diff --git a/drivers/infiniband/hw/mlx4/main.c b/drivers/infiniband/hw/mlx4/main.c
index c7a6213..66e3eec 100644
--- a/drivers/infiniband/hw/mlx4/main.c
+++ b/drivers/infiniband/hw/mlx4/main.c
@@ -721,7 +721,6 @@ static int init_node_data(struct mlx4_ib_dev *dev)
if (err)
goto out;
- dev->dev->rev_id = be32_to_cpup((__be32 *) (out_mad->data + 32));
memcpy(&dev->ib_dev.node_guid, out_mad->data + 12, 8);
out:
diff --git a/drivers/net/mlx4/en_ethtool.c b/drivers/net/mlx4/en_ethtool.c
index c1f351f..62ace6c 100644
--- a/drivers/net/mlx4/en_ethtool.c
+++ b/drivers/net/mlx4/en_ethtool.c
@@ -45,7 +45,20 @@ mlx4_en_get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *drvinfo)
struct mlx4_en_priv *priv = netdev_priv(dev);
struct mlx4_en_dev *mdev = priv->mdev;
- sprintf(drvinfo->driver, DRV_NAME " (%s)", mdev->dev->board_id);
+ switch (mdev->dev->rev_id) {
+ case 0xa0:
+ if (dev->dev_id >= MLX4_EN_CX3_LOW_ID && dev->dev_id <= MLX4_EN_CX3_HIGH_ID)
+ sprintf(drvinfo->driver, DRV_NAME " (%s_CX-3)", mdev->dev->board_id);
+ else
+ sprintf(drvinfo->driver, DRV_NAME " (%s_CX)", mdev->dev->board_id);
+ break;
+ case 0xb0:
+ sprintf(drvinfo->driver, DRV_NAME " (%s_CX-2)", mdev->dev->board_id);
+ break;
+ default:
+ sprintf(drvinfo->driver, DRV_NAME " (%s)", mdev->dev->board_id);
+ break;
+ }
strncpy(drvinfo->version, DRV_VERSION " (" DRV_RELDATE ")", 32);
sprintf(drvinfo->fw_version, "%d.%d.%d",
(u16) (mdev->dev->caps.fw_ver >> 32),
diff --git a/drivers/net/mlx4/main.c b/drivers/net/mlx4/main.c
index 42d4fb4..5bebb88 100644
--- a/drivers/net/mlx4/main.c
+++ b/drivers/net/mlx4/main.c
@@ -1139,6 +1139,8 @@ static int __mlx4_init_one(struct pci_dev *pdev, const struct pci_device_id *id)
INIT_LIST_HEAD(&priv->pgdir_list);
mutex_init(&priv->pgdir_mutex);
+ pci_read_config_byte(pdev, PCI_REVISION_ID, &dev->rev_id);
+
/*
* Now reset the HCA before we touch the PCI capabilities or
* attempt a firmware command, since a boot ROM may have left
diff --git a/drivers/net/mlx4/mlx4_en.h b/drivers/net/mlx4/mlx4_en.h
index 07aea8d..ad4df66 100644
--- a/drivers/net/mlx4/mlx4_en.h
+++ b/drivers/net/mlx4/mlx4_en.h
@@ -216,6 +216,9 @@ struct mlx4_en_tx_desc {
#define MLX4_EN_USE_SRQ 0x01000000
+#define MLX4_EN_CX3_LOW_ID 0x1000
+#define MLX4_EN_CX3_HIGH_ID 0x1005
+
struct mlx4_en_rx_alloc {
struct page *page;
u16 offset;
diff --git a/include/linux/mlx4/device.h b/include/linux/mlx4/device.h
index 2460356..fe2a3a3 100644
--- a/include/linux/mlx4/device.h
+++ b/include/linux/mlx4/device.h
@@ -422,7 +422,7 @@ struct mlx4_dev {
unsigned long flags;
struct mlx4_caps caps;
struct radix_tree_root qp_table_tree;
- u32 rev_id;
+ u8 rev_id;
char board_id[MLX4_BOARD_ID_LEN];
};
--
1.6.0.2
^ permalink raw reply related
* [PATCH 07/17] mlx4: Wake on LAN support
From: Yevgeny Petrilin @ 2011-03-07 10:17 UTC (permalink / raw)
To: davem; +Cc: netdev, yevgenyp, igory
The driver queries the FW for WOL support.
Ethtool get/set_wol is implemented accordingly.
Only magic packets are supported at the time.
Signed-off-by: Igor Yarovinsky <igory@mellanox.co.il>
Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il>
---
drivers/net/mlx4/en_ethtool.c | 62 +++++++++++++++++++++++++++++++++++++++-
drivers/net/mlx4/fw.c | 20 +++++++++++++
drivers/net/mlx4/fw.h | 1 +
drivers/net/mlx4/main.c | 1 +
drivers/net/mlx4/mlx4_en.h | 7 ++++
include/linux/mlx4/device.h | 4 ++
6 files changed, 93 insertions(+), 2 deletions(-)
diff --git a/drivers/net/mlx4/en_ethtool.c b/drivers/net/mlx4/en_ethtool.c
index 8cfe858..c1f351f 100644
--- a/drivers/net/mlx4/en_ethtool.c
+++ b/drivers/net/mlx4/en_ethtool.c
@@ -131,8 +131,65 @@ static void mlx4_en_set_msglevel(struct net_device *dev, u32 val)
static void mlx4_en_get_wol(struct net_device *netdev,
struct ethtool_wolinfo *wol)
{
- wol->supported = 0;
- wol->wolopts = 0;
+ struct mlx4_en_priv *priv = netdev_priv(netdev);
+ int err = 0;
+ u64 config = 0;
+
+ if (!priv->mdev->dev->caps.wol) {
+ wol->supported = 0;
+ wol->wolopts = 0;
+ return;
+ }
+
+ err = mlx4_wol_read(priv->mdev->dev, &config, priv->port);
+ if (err) {
+ en_err(priv, "Failed to get WoL information\n");
+ return;
+ }
+
+ if (config & MLX4_EN_WOL_MAGIC)
+ wol->supported = WAKE_MAGIC;
+ else
+ wol->supported = 0;
+
+ if (config & MLX4_EN_WOL_ENABLED)
+ wol->wolopts = WAKE_MAGIC;
+ else
+ wol->wolopts = 0;
+}
+
+static int mlx4_en_set_wol(struct net_device *netdev,
+ struct ethtool_wolinfo *wol)
+{
+ struct mlx4_en_priv *priv = netdev_priv(netdev);
+ u64 config = 0;
+ int err = 0;
+
+ if (!priv->mdev->dev->caps.wol)
+ return -EOPNOTSUPP;
+
+ if (wol->supported & ~WAKE_MAGIC)
+ return -EINVAL;
+
+ err = mlx4_wol_read(priv->mdev->dev, &config, priv->port);
+ if (err) {
+ en_err(priv, "Failed to get WoL info, unable to modify\n");
+ return err;
+ }
+
+ if (wol->wolopts & WAKE_MAGIC) {
+ config |= MLX4_EN_WOL_DO_MODIFY | MLX4_EN_WOL_ENABLED |
+ MLX4_EN_WOL_MAGIC;
+ } else {
+ config &= ~(MLX4_EN_WOL_ENABLED | MLX4_EN_WOL_MAGIC);
+ config |= MLX4_EN_WOL_DO_MODIFY;
+ }
+
+ err = mlx4_wol_write(priv->mdev->dev, config, priv->port);
+ if (err)
+ en_err(priv, "Failed to set WoL information\n");
+
+ return err;
}
static int mlx4_en_get_sset_count(struct net_device *dev, int sset)
@@ -442,6 +499,7 @@ const struct ethtool_ops mlx4_en_ethtool_ops = {
.get_ethtool_stats = mlx4_en_get_ethtool_stats,
.self_test = mlx4_en_self_test,
.get_wol = mlx4_en_get_wol,
+ .set_wol = mlx4_en_set_wol,
.get_msglevel = mlx4_en_get_msglevel,
.set_msglevel = mlx4_en_set_msglevel,
.get_coalesce = mlx4_en_get_coalesce,
diff --git a/drivers/net/mlx4/fw.c b/drivers/net/mlx4/fw.c
index 5de1db8..fd1c51b 100644
--- a/drivers/net/mlx4/fw.c
+++ b/drivers/net/mlx4/fw.c
@@ -276,6 +276,7 @@ int mlx4_QUERY_DEV_CAP(struct mlx4_dev *dev, struct mlx4_dev_cap *dev_cap)
dev_cap->udp_rss = field & 0x1;
MLX4_GET(field, outbox, QUERY_DEV_CAP_ETH_UC_LOOPBACK_OFFSET);
dev_cap->loopback_support = field & 0x1;
+ dev_cap->wol = field & 0x40;
MLX4_GET(dev_cap->flags, outbox, QUERY_DEV_CAP_FLAGS_OFFSET);
MLX4_GET(field, outbox, QUERY_DEV_CAP_RSVD_UAR_OFFSET);
dev_cap->reserved_uars = field >> 4;
@@ -908,3 +909,22 @@ int mlx4_NOP(struct mlx4_dev *dev)
/* Input modifier of 0x1f means "finish as soon as possible." */
return mlx4_cmd(dev, 0, 0x1f, 0, MLX4_CMD_NOP, 100);
}
+
+#define MLX4_WOL_SETUP_MODE (5 << 28)
+int mlx4_wol_read(struct mlx4_dev *dev, u64 *config, int port)
+{
+ u32 in_mod = MLX4_WOL_SETUP_MODE | port << 8;
+
+ return mlx4_cmd_imm(dev, 0, config, in_mod, 0x3,
+ MLX4_CMD_MOD_STAT_CFG, MLX4_CMD_TIME_CLASS_A);
+}
+EXPORT_SYMBOL_GPL(mlx4_wol_read);
+
+int mlx4_wol_write(struct mlx4_dev *dev, u64 config, int port)
+{
+ u32 in_mod = MLX4_WOL_SETUP_MODE | port << 8;
+
+ return mlx4_cmd(dev, config, in_mod, 0x1, MLX4_CMD_MOD_STAT_CFG,
+ MLX4_CMD_TIME_CLASS_A);
+}
+EXPORT_SYMBOL_GPL(mlx4_wol_write);
diff --git a/drivers/net/mlx4/fw.h b/drivers/net/mlx4/fw.h
index 65cc72e..f7b9cc2 100644
--- a/drivers/net/mlx4/fw.h
+++ b/drivers/net/mlx4/fw.h
@@ -80,6 +80,7 @@ struct mlx4_dev_cap {
u16 stat_rate_support;
int udp_rss;
int loopback_support;
+ int wol;
u32 flags;
int reserved_uars;
int uar_size;
diff --git a/drivers/net/mlx4/main.c b/drivers/net/mlx4/main.c
index 517ca34..42d4fb4 100644
--- a/drivers/net/mlx4/main.c
+++ b/drivers/net/mlx4/main.c
@@ -227,6 +227,7 @@ static int mlx4_dev_cap(struct mlx4_dev *dev, struct mlx4_dev_cap *dev_cap)
dev->caps.stat_rate_support = dev_cap->stat_rate_support;
dev->caps.udp_rss = dev_cap->udp_rss;
dev->caps.loopback_support = dev_cap->loopback_support;
+ dev->caps.wol = dev_cap->wol;
dev->caps.max_gso_sz = dev_cap->max_gso_sz;
dev->caps.log_num_macs = log_num_mac;
diff --git a/drivers/net/mlx4/mlx4_en.h b/drivers/net/mlx4/mlx4_en.h
index 2db245f..07aea8d 100644
--- a/drivers/net/mlx4/mlx4_en.h
+++ b/drivers/net/mlx4/mlx4_en.h
@@ -479,6 +479,13 @@ struct mlx4_en_priv {
int mc_addrs_cnt;
struct mlx4_en_stat_out_mbox hw_stats;
int vids[128];
+ bool wol;
+};
+
+enum mlx4_en_wol {
+ MLX4_EN_WOL_MAGIC = (1ULL << 61),
+ MLX4_EN_WOL_ENABLED = (1ULL << 62),
+ MLX4_EN_WOL_DO_MODIFY = (1ULL << 63),
};
diff --git a/include/linux/mlx4/device.h b/include/linux/mlx4/device.h
index 7838082..2460356 100644
--- a/include/linux/mlx4/device.h
+++ b/include/linux/mlx4/device.h
@@ -251,6 +251,7 @@ struct mlx4_caps {
u16 stat_rate_support;
int udp_rss;
int loopback_support;
+ int wol;
u8 port_width_cap[MLX4_MAX_PORTS + 1];
int max_gso_sz;
int reserved_qps_cnt[MLX4_NUM_QP_REGION];
@@ -535,4 +536,7 @@ int mlx4_test_interrupts(struct mlx4_dev *dev);
int mlx4_assign_eq(struct mlx4_dev *dev, char* name , int* vector);
void mlx4_release_eq(struct mlx4_dev *dev, int vec);
+int mlx4_wol_read(struct mlx4_dev *dev, u64 *config, int port);
+int mlx4_wol_write(struct mlx4_dev *dev, u64 config, int port);
+
#endif /* MLX4_DEVICE_H */
--
1.6.0.2
^ permalink raw reply related
* [PATCH 06/17] mlx4_en: using new mlx4 interrupt scheme
From: Yevgeny Petrilin @ 2011-03-07 10:17 UTC (permalink / raw)
To: davem; +Cc: netdev, yevgenyp, markuze
Each RX ring will have its own interrupt vector, and TX rings will share one
(we mostly use polling for TX completions).
The vectors are assigned first time device is opened, and its name includes
the interface name and ring number.
Signed-off-by: Markuze Alex <markuze@mellanox.co.il>
Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il>
---
drivers/net/mlx4/en_cq.c | 38 +++++++++++++++++++++++++++++++-------
drivers/net/mlx4/en_ethtool.c | 2 +-
drivers/net/mlx4/en_main.c | 20 +++++++++++---------
drivers/net/mlx4/en_netdev.c | 18 ++++++++++++++----
drivers/net/mlx4/mlx4_en.h | 7 +++++--
5 files changed, 62 insertions(+), 23 deletions(-)
diff --git a/drivers/net/mlx4/en_cq.c b/drivers/net/mlx4/en_cq.c
index 21786ad..ec4b6d0 100644
--- a/drivers/net/mlx4/en_cq.c
+++ b/drivers/net/mlx4/en_cq.c
@@ -51,13 +51,10 @@ int mlx4_en_create_cq(struct mlx4_en_priv *priv,
int err;
cq->size = entries;
- if (mode == RX) {
+ if (mode == RX)
cq->buf_size = cq->size * sizeof(struct mlx4_cqe);
- cq->vector = ring % mdev->dev->caps.num_comp_vectors;
- } else {
+ else
cq->buf_size = sizeof(struct mlx4_cqe);
- cq->vector = 0;
- }
cq->ring = ring;
cq->is_tx = mode;
@@ -80,7 +77,8 @@ int mlx4_en_create_cq(struct mlx4_en_priv *priv,
int mlx4_en_activate_cq(struct mlx4_en_priv *priv, struct mlx4_en_cq *cq)
{
struct mlx4_en_dev *mdev = priv->mdev;
- int err;
+ int err = 0;
+ char name[25];
cq->dev = mdev->pndev[priv->port];
cq->mcq.set_ci_db = cq->wqres.db.db;
@@ -89,6 +87,29 @@ int mlx4_en_activate_cq(struct mlx4_en_priv *priv, struct mlx4_en_cq *cq)
*cq->mcq.arm_db = 0;
memset(cq->buf, 0, cq->buf_size);
+ if (cq->is_tx == RX) {
+ if (mdev->dev->caps.comp_pool) {
+ if (!cq->vector) {
+ sprintf(name , "%s-rx-%d", priv->dev->name, cq->ring);
+ if (mlx4_assign_eq(mdev->dev, name, &cq->vector)) {
+ cq->vector = (cq->ring + 1 + priv->port) %
+ mdev->dev->caps.num_comp_vectors;
+ mlx4_warn(mdev, "Failed Assigning an EQ to "
+ "%s_rx-%d ,Falling back to legacy EQ's\n",
+ priv->dev->name, cq->ring);
+ }
+ }
+ } else {
+ cq->vector = (cq->ring + 1 + priv->port) %
+ mdev->dev->caps.num_comp_vectors;
+ }
+ } else {
+ if (!cq->vector || !mdev->dev->caps.comp_pool) {
+ /*Fallback to legacy pool in case of error*/
+ cq->vector = 0;
+ }
+ }
+
if (!cq->is_tx)
cq->size = priv->rx_ring[cq->ring].actual_size;
@@ -112,12 +133,15 @@ int mlx4_en_activate_cq(struct mlx4_en_priv *priv, struct mlx4_en_cq *cq)
return 0;
}
-void mlx4_en_destroy_cq(struct mlx4_en_priv *priv, struct mlx4_en_cq *cq)
+void mlx4_en_destroy_cq(struct mlx4_en_priv *priv, struct mlx4_en_cq *cq,
+ bool reserve_vectors)
{
struct mlx4_en_dev *mdev = priv->mdev;
mlx4_en_unmap_buffer(&cq->wqres.buf);
mlx4_free_hwq_res(mdev->dev, &cq->wqres, cq->buf_size);
+ if (priv->mdev->dev->caps.comp_pool && cq->vector && !reserve_vectors)
+ mlx4_release_eq(priv->mdev->dev, cq->vector);
cq->buf_size = 0;
cq->buf = NULL;
}
diff --git a/drivers/net/mlx4/en_ethtool.c b/drivers/net/mlx4/en_ethtool.c
index 056152b..8cfe858 100644
--- a/drivers/net/mlx4/en_ethtool.c
+++ b/drivers/net/mlx4/en_ethtool.c
@@ -388,7 +388,7 @@ static int mlx4_en_set_ringparam(struct net_device *dev,
mlx4_en_stop_port(dev);
}
- mlx4_en_free_resources(priv);
+ mlx4_en_free_resources(priv, true);
priv->prof->tx_ring_size = tx_size;
priv->prof->rx_ring_size = rx_size;
diff --git a/drivers/net/mlx4/en_main.c b/drivers/net/mlx4/en_main.c
index 1ff6ca6..29aaa43 100644
--- a/drivers/net/mlx4/en_main.c
+++ b/drivers/net/mlx4/en_main.c
@@ -241,16 +241,18 @@ static void *mlx4_en_add(struct mlx4_dev *dev)
mlx4_foreach_port(i, dev, MLX4_PORT_TYPE_ETH)
mdev->port_cnt++;
- /* If we did not receive an explicit number of Rx rings, default to
- * the number of completion vectors populated by the mlx4_core */
mlx4_foreach_port(i, dev, MLX4_PORT_TYPE_ETH) {
- mlx4_info(mdev, "Using %d tx rings for port:%d\n",
- mdev->profile.prof[i].tx_ring_num, i);
- mdev->profile.prof[i].rx_ring_num = min_t(int,
- roundup_pow_of_two(dev->caps.num_comp_vectors),
- MAX_RX_RINGS);
- mlx4_info(mdev, "Defaulting to %d rx rings for port:%d\n",
- mdev->profile.prof[i].rx_ring_num, i);
+ if (!dev->caps.comp_pool) {
+ mdev->profile.prof[i].rx_ring_num =
+ rounddown_pow_of_two(max_t(int, MIN_RX_RINGS,
+ min_t(int,
+ dev->caps.num_comp_vectors,
+ MAX_RX_RINGS)));
+ } else {
+ mdev->profile.prof[i].rx_ring_num = rounddown_pow_of_two(
+ min_t(int, dev->caps.comp_pool/
+ dev->caps.num_ports - 1 , MAX_MSIX_P_PORT - 1));
+ }
}
/* Create our own workqueue for reset/multicast tasks
diff --git a/drivers/net/mlx4/en_netdev.c b/drivers/net/mlx4/en_netdev.c
index 5727bf5..f0d06a0 100644
--- a/drivers/net/mlx4/en_netdev.c
+++ b/drivers/net/mlx4/en_netdev.c
@@ -557,6 +557,7 @@ int mlx4_en_start_port(struct net_device *dev)
int err = 0;
int i;
int j;
+ char name[32];
if (priv->port_up) {
en_dbg(DRV, priv, "start port called while port already up\n");
@@ -601,10 +602,19 @@ int mlx4_en_start_port(struct net_device *dev)
goto cq_err;
}
+ if (mdev->dev->caps.comp_pool && !priv->tx_vector) {
+ sprintf(name , "%s_tx", priv->dev->name);
+ if (mlx4_assign_eq(mdev->dev , name, &priv->tx_vector)) {
+ mlx4_warn(mdev, "Failed Assigning an EQ to "
+ "%s_tx ,Falling back to legacy "
+ "EQ's\n", priv->dev->name);
+ }
+ }
/* Configure tx cq's and rings */
for (i = 0; i < priv->tx_ring_num; i++) {
/* Configure cq */
cq = &priv->tx_cq[i];
+ cq->vector = priv->tx_vector;
err = mlx4_en_activate_cq(priv, cq);
if (err) {
en_err(priv, "Failed allocating Tx CQ\n");
@@ -819,7 +829,7 @@ static int mlx4_en_close(struct net_device *dev)
return 0;
}
-void mlx4_en_free_resources(struct mlx4_en_priv *priv)
+void mlx4_en_free_resources(struct mlx4_en_priv *priv, bool reserve_vectors)
{
int i;
@@ -827,14 +837,14 @@ void mlx4_en_free_resources(struct mlx4_en_priv *priv)
if (priv->tx_ring[i].tx_info)
mlx4_en_destroy_tx_ring(priv, &priv->tx_ring[i]);
if (priv->tx_cq[i].buf)
- mlx4_en_destroy_cq(priv, &priv->tx_cq[i]);
+ mlx4_en_destroy_cq(priv, &priv->tx_cq[i], reserve_vectors);
}
for (i = 0; i < priv->rx_ring_num; i++) {
if (priv->rx_ring[i].rx_info)
mlx4_en_destroy_rx_ring(priv, &priv->rx_ring[i]);
if (priv->rx_cq[i].buf)
- mlx4_en_destroy_cq(priv, &priv->rx_cq[i]);
+ mlx4_en_destroy_cq(priv, &priv->rx_cq[i], reserve_vectors);
}
}
@@ -896,7 +906,7 @@ void mlx4_en_destroy_netdev(struct net_device *dev)
mdev->pndev[priv->port] = NULL;
mutex_unlock(&mdev->state_lock);
- mlx4_en_free_resources(priv);
+ mlx4_en_free_resources(priv, false);
free_netdev(dev);
}
diff --git a/drivers/net/mlx4/mlx4_en.h b/drivers/net/mlx4/mlx4_en.h
index 5387c4e..2db245f 100644
--- a/drivers/net/mlx4/mlx4_en.h
+++ b/drivers/net/mlx4/mlx4_en.h
@@ -62,6 +62,7 @@
#define MLX4_EN_PAGE_SHIFT 12
#define MLX4_EN_PAGE_SIZE (1 << MLX4_EN_PAGE_SHIFT)
#define MAX_RX_RINGS 16
+#define MIN_RX_RINGS 4
#define TXBB_SIZE 64
#define HEADROOM (2048 / TXBB_SIZE + 1)
#define STAMP_STRIDE 64
@@ -462,6 +463,7 @@ struct mlx4_en_priv {
u16 log_rx_info;
struct mlx4_en_tx_ring tx_ring[MAX_TX_RINGS];
+ int tx_vector;
struct mlx4_en_rx_ring rx_ring[MAX_RX_RINGS];
struct mlx4_en_cq tx_cq[MAX_TX_RINGS];
struct mlx4_en_cq rx_cq[MAX_RX_RINGS];
@@ -487,12 +489,13 @@ int mlx4_en_init_netdev(struct mlx4_en_dev *mdev, int port,
int mlx4_en_start_port(struct net_device *dev);
void mlx4_en_stop_port(struct net_device *dev);
-void mlx4_en_free_resources(struct mlx4_en_priv *priv);
+void mlx4_en_free_resources(struct mlx4_en_priv *priv, bool reserve_vectors);
int mlx4_en_alloc_resources(struct mlx4_en_priv *priv);
int mlx4_en_create_cq(struct mlx4_en_priv *priv, struct mlx4_en_cq *cq,
int entries, int ring, enum cq_type mode);
-void mlx4_en_destroy_cq(struct mlx4_en_priv *priv, struct mlx4_en_cq *cq);
+void mlx4_en_destroy_cq(struct mlx4_en_priv *priv, struct mlx4_en_cq *cq,
+ bool reserve_vectors);
int mlx4_en_activate_cq(struct mlx4_en_priv *priv, struct mlx4_en_cq *cq);
void mlx4_en_deactivate_cq(struct mlx4_en_priv *priv, struct mlx4_en_cq *cq);
int mlx4_en_set_cq_moder(struct mlx4_en_priv *priv, struct mlx4_en_cq *cq);
--
1.6.0.2
^ permalink raw reply related
* [PATCH 04/17] mlx4_en: bringing link up when registering net device
From: Yevgeny Petrilin @ 2011-03-07 10:17 UTC (permalink / raw)
To: davem; +Cc: netdev, yevgenyp, eugenia
Signed-off-by: Eugenia Emantayev <eugenia@mellanox.co.il>
Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il>
---
drivers/net/mlx4/en_netdev.c | 17 +++++++++++++++++
1 files changed, 17 insertions(+), 0 deletions(-)
diff --git a/drivers/net/mlx4/en_netdev.c b/drivers/net/mlx4/en_netdev.c
index b1180dd..5727bf5 100644
--- a/drivers/net/mlx4/en_netdev.c
+++ b/drivers/net/mlx4/en_netdev.c
@@ -1069,6 +1069,23 @@ int mlx4_en_init_netdev(struct mlx4_en_dev *mdev, int port,
en_warn(priv, "Using %d TX rings\n", prof->tx_ring_num);
en_warn(priv, "Using %d RX rings\n", prof->rx_ring_num);
+ /* Configure port */
+ err = mlx4_SET_PORT_general(mdev->dev, priv->port,
+ MLX4_EN_MIN_MTU,
+ 0, 0, 0, 0);
+ if (err) {
+ en_err(priv, "Failed setting port general configurations "
+ "for port %d, with error %d\n", priv->port, err);
+ goto out;
+ }
+
+ /* Init port */
+ en_warn(priv, "Initializing port\n");
+ err = mlx4_INIT_PORT(mdev->dev, priv->port);
+ if (err) {
+ en_err(priv, "Failed Initializing port\n");
+ goto out;
+ }
priv->registered = 1;
mlx4_en_set_default_moderation(priv);
queue_delayed_work(mdev->workqueue, &priv->stats_task, STATS_DELAY);
--
1.6.0.2
^ permalink raw reply related
* [PATCH 02/17] mlx4_en: moderation parameters are not reseted.
From: Yevgeny Petrilin @ 2011-03-07 10:17 UTC (permalink / raw)
To: davem; +Cc: netdev, yevgenyp
Instead of reseting the module parameters each ifup or mtu change,
they are being set once at device initialization
Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il>
---
drivers/net/mlx4/en_netdev.c | 3 +--
1 files changed, 1 insertions(+), 2 deletions(-)
diff --git a/drivers/net/mlx4/en_netdev.c b/drivers/net/mlx4/en_netdev.c
index 897f576..56627c2 100644
--- a/drivers/net/mlx4/en_netdev.c
+++ b/drivers/net/mlx4/en_netdev.c
@@ -801,7 +801,6 @@ static int mlx4_en_open(struct net_device *dev)
priv->rx_ring[i].packets = 0;
}
- mlx4_en_set_default_moderation(priv);
err = mlx4_en_start_port(dev);
if (err)
en_err(priv, "Failed starting port:%d\n", priv->port);
@@ -932,7 +931,6 @@ static int mlx4_en_change_mtu(struct net_device *dev, int new_mtu)
en_dbg(DRV, priv, "Change MTU called with card down!?\n");
} else {
mlx4_en_stop_port(dev);
- mlx4_en_set_default_moderation(priv);
err = mlx4_en_start_port(dev);
if (err) {
en_err(priv, "Failed restarting port:%d\n",
@@ -1080,6 +1078,7 @@ int mlx4_en_init_netdev(struct mlx4_en_dev *mdev, int port,
en_warn(priv, "Using %d RX rings\n", prof->rx_ring_num);
priv->registered = 1;
+ mlx4_en_set_default_moderation(priv);
queue_delayed_work(mdev->workqueue, &priv->stats_task, STATS_DELAY);
return 0;
--
1.6.0.2
^ permalink raw reply related
* [PATCH 00/17] mlx4_en: driver updates
From: Yevgeny Petrilin @ 2011-03-07 10:16 UTC (permalink / raw)
To: davem; +Cc: netdev, yevgenyp
Hello David,
This patch set presents a set of fixes, performance improvements and new features for the mlx4_core and mlx4_en modules.
main changes are:
- MSI-X per ring
- Changing RX packet steering model for supported devices
- BlueFlame support for better TX latency.
- Adaptive moderation improvements.
Patches were generated against net-next-2.6.git
Thanks,
Yevgeny
^ permalink raw reply
* [PATCH 2/2 net-2.6] r6040: bump to version 0.27 and date 23Feb2011
From: Florian Fainelli @ 2011-03-07 10:09 UTC (permalink / raw)
To: netdev@vger.kernel.org; +Cc: David Miller
From: Florian Fainelli <florian@openwrt.org>
Signed-off-by: Florian Fainelli <florian@openwrt.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
diff --git a/drivers/net/r6040.c b/drivers/net/r6040.c
index 7965ae4..e3ebd90 100644
--- a/drivers/net/r6040.c
+++ b/drivers/net/r6040.c
@@ -49,8 +49,8 @@
#include <asm/processor.h>
#define DRV_NAME "r6040"
-#define DRV_VERSION "0.26"
-#define DRV_RELDATE "30May2010"
+#define DRV_VERSION "0.27"
+#define DRV_RELDATE "23Feb2011"
/* PHY CHIP Address */
#define PHY1_ADDR 1 /* For MAC1 */
--
1.7.1
^ permalink raw reply related
* [PATCH 1/2 net-2.6] r6040: fix multicast operations
From: Florian Fainelli @ 2011-03-07 10:09 UTC (permalink / raw)
To: netdev@vger.kernel.org; +Cc: David Miller
From: Shawn Lin <shawn@dmp.com.tw>
The original code does not work well when the number of mulitcast
address to handle is greater than MCAST_MAX. It only enable promiscous
mode instead of multicast hash table mode, so the hash table function
will not be activated and all multicast frames will be recieved in this
condition.
This patch fixes the following issues with the r6040 NIC operating in
multicast:
1) When the IFF_ALLMULTI flag is set, we should write 0xffff to the NIC
hash table registers to make it process multicast traffic.
2) When the number of multicast address to handle is smaller than
MCAST_MAX, we should use the NIC multicast registers MID1_{L,M,H}.
3) The hashing of the address was not correct, due to an invalid
substraction (15 - (crc & 0x0f)) instead of (crc & 0x0f) and an
incorrect crc algorithm (ether_crc_le) instead of (ether_crc).
4) If necessary, we should set HASH_EN flag in MCR0 to enable multicast
hash table function.
Reported-by: Marc Leclerc <marc-leclerc@signaturealpha.com>
Tested-by: Marc Leclerc <marc-leclerc@signaturealpha.com>
Signed-off-by: Shawn Lin <shawn@dmp.com.tw>
Signed-off-by: Albert Chen <albert.chen@rdc.com.tw>
Signed-off-by: Florian Fainelli <florian@openwrt.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
diff --git a/drivers/net/r6040.c b/drivers/net/r6040.c
index 27e6f6d..7965ae4 100644
--- a/drivers/net/r6040.c
+++ b/drivers/net/r6040.c
@@ -69,6 +69,8 @@
/* MAC registers */
#define MCR0 0x00 /* Control register 0 */
+#define MCR0_PROMISC 0x0020 /* Promiscuous mode */
+#define MCR0_HASH_EN 0x0100 /* Enable multicast hash table function */
#define MCR1 0x04 /* Control register 1 */
#define MAC_RST 0x0001 /* Reset the MAC */
#define MBCR 0x08 /* Bus control */
@@ -851,77 +853,92 @@ static void r6040_multicast_list(struct net_device *dev)
{
struct r6040_private *lp = netdev_priv(dev);
void __iomem *ioaddr = lp->base;
- u16 *adrp;
- u16 reg;
unsigned long flags;
struct netdev_hw_addr *ha;
int i;
+ u16 *adrp;
+ u16 hash_table[4] = { 0 };
+
+ spin_lock_irqsave(&lp->lock, flags);
- /* MAC Address */
+ /* Keep our MAC Address */
adrp = (u16 *)dev->dev_addr;
iowrite16(adrp[0], ioaddr + MID_0L);
iowrite16(adrp[1], ioaddr + MID_0M);
iowrite16(adrp[2], ioaddr + MID_0H);
- /* Promiscous Mode */
- spin_lock_irqsave(&lp->lock, flags);
-
/* Clear AMCP & PROM bits */
- reg = ioread16(ioaddr) & ~0x0120;
- if (dev->flags & IFF_PROMISC) {
- reg |= 0x0020;
- lp->mcr0 |= 0x0020;
- }
- /* Too many multicast addresses
- * accept all traffic */
- else if ((netdev_mc_count(dev) > MCAST_MAX) ||
- (dev->flags & IFF_ALLMULTI))
- reg |= 0x0020;
+ lp->mcr0 = ioread16(ioaddr + MCR0) & ~(MCR0_PROMISC | MCR0_HASH_EN);
- iowrite16(reg, ioaddr);
- spin_unlock_irqrestore(&lp->lock, flags);
+ /* Promiscuous mode */
+ if (dev->flags & IFF_PROMISC)
+ lp->mcr0 |= MCR0_PROMISC;
- /* Build the hash table */
- if (netdev_mc_count(dev) > MCAST_MAX) {
- u16 hash_table[4];
- u32 crc;
+ /* Enable multicast hash table function to
+ * receive all multicast packets. */
+ else if (dev->flags & IFF_ALLMULTI) {
+ lp->mcr0 |= MCR0_HASH_EN;
- for (i = 0; i < 4; i++)
- hash_table[i] = 0;
+ for (i = 0; i < MCAST_MAX ; i++) {
+ iowrite16(0, ioaddr + MID_1L + 8 * i);
+ iowrite16(0, ioaddr + MID_1M + 8 * i);
+ iowrite16(0, ioaddr + MID_1H + 8 * i);
+ }
+ for (i = 0; i < 4; i++)
+ hash_table[i] = 0xffff;
+ }
+ /* Use internal multicast address registers if the number of
+ * multicast addresses is not greater than MCAST_MAX. */
+ else if (netdev_mc_count(dev) <= MCAST_MAX) {
+ i = 0;
netdev_for_each_mc_addr(ha, dev) {
- char *addrs = ha->addr;
+ u16 *adrp = (u16 *) ha->addr;
+ iowrite16(adrp[0], ioaddr + MID_1L + 8 * i);
+ iowrite16(adrp[1], ioaddr + MID_1M + 8 * i);
+ iowrite16(adrp[2], ioaddr + MID_1H + 8 * i);
+ i++;
+ }
+ while (i < MCAST_MAX) {
+ iowrite16(0, ioaddr + MID_1L + 8 * i);
+ iowrite16(0, ioaddr + MID_1M + 8 * i);
+ iowrite16(0, ioaddr + MID_1H + 8 * i);
+ i++;
+ }
+ }
+ /* Otherwise, Enable multicast hash table function. */
+ else {
+ u32 crc;
- if (!(*addrs & 1))
- continue;
+ lp->mcr0 |= MCR0_HASH_EN;
+
+ for (i = 0; i < MCAST_MAX ; i++) {
+ iowrite16(0, ioaddr + MID_1L + 8 * i);
+ iowrite16(0, ioaddr + MID_1M + 8 * i);
+ iowrite16(0, ioaddr + MID_1H + 8 * i);
+ }
- crc = ether_crc_le(6, addrs);
+ /* Build multicast hash table */
+ netdev_for_each_mc_addr(ha, dev) {
+ u8 *addrs = ha->addr;
+
+ crc = ether_crc(ETH_ALEN, addrs);
crc >>= 26;
- hash_table[crc >> 4] |= 1 << (15 - (crc & 0xf));
+ hash_table[crc >> 4] |= 1 << (crc & 0xf);
}
- /* Fill the MAC hash tables with their values */
+ }
+
+ iowrite16(lp->mcr0, ioaddr + MCR0);
+
+ /* Fill the MAC hash tables with their values */
+ if (lp->mcr0 && MCR0_HASH_EN) {
iowrite16(hash_table[0], ioaddr + MAR0);
iowrite16(hash_table[1], ioaddr + MAR1);
iowrite16(hash_table[2], ioaddr + MAR2);
iowrite16(hash_table[3], ioaddr + MAR3);
}
- /* Multicast Address 1~4 case */
- i = 0;
- netdev_for_each_mc_addr(ha, dev) {
- if (i >= MCAST_MAX)
- break;
- adrp = (u16 *) ha->addr;
- iowrite16(adrp[0], ioaddr + MID_1L + 8 * i);
- iowrite16(adrp[1], ioaddr + MID_1M + 8 * i);
- iowrite16(adrp[2], ioaddr + MID_1H + 8 * i);
- i++;
- }
- while (i < MCAST_MAX) {
- iowrite16(0xffff, ioaddr + MID_1L + 8 * i);
- iowrite16(0xffff, ioaddr + MID_1M + 8 * i);
- iowrite16(0xffff, ioaddr + MID_1H + 8 * i);
- i++;
- }
+
+ spin_unlock_irqrestore(&lp->lock, flags);
}
static void netdev_get_drvinfo(struct net_device *dev,
--
1.7.1
^ permalink raw reply related
* Re: [PATCH net-next-2.6] net: harmonize the call to ptype_all and ptype_base handlers.
From: Jiri Pirko @ 2011-03-07 10:03 UTC (permalink / raw)
To: Nicolas de Pesloüan
Cc: netdev, davem, shemminger, eric.dumazet, kaber, fubar, andy
In-Reply-To: <1299417916-14198-1-git-send-email-nicolas.2p.debian@free.fr>
Sun, Mar 06, 2011 at 02:25:16PM CET, nicolas.2p.debian@free.fr wrote:
>Until now, ptype_all and ptype_base delivery in __netif_receive_skb() is
>inconsistent.
>
>- For ptype_all, we deliver to every device crossed while walking the
>rx_handler path (inside the another_round loop), and there is no way to stop
>wildcard delivery (no exact match logic).
>- For ptype_base, we deliver to the lowest device (orig_dev) and to the highest
>(skb->dev) and we can ask for exact match delivery.
>
>This patch try and fix this, by:
>
>1/ Doing exact match delivery for both ptype_all and ptype_base, while walking
> the rx_handler path.
>2/ Doing wildcard match delivery at the end of __netif_receive_skb(), if not
> asked to do exact match delivery only.
>
>Signed-off-by: Nicolas de Pesloüan <nicolas.2p.debian@free.fr>
>---
>
>This apply on top of the last batch of patch from Jiri Pirko.
>---
> net/core/dev.c | 32 ++++++++++++++++++++++++--------
> 1 files changed, 24 insertions(+), 8 deletions(-)
>
I tend to like this patch. However I'm not sure if extra 2 loops don't
introduce noticable overhead :/
^ permalink raw reply
* Re: [patch net-next-2.6 4/8] bonding: wrap slave state work
From: Jiri Pirko @ 2011-03-07 9:58 UTC (permalink / raw)
To: Nicolas de Pesloüan
Cc: netdev, davem, shemminger, kaber, fubar, eric.dumazet, andy
In-Reply-To: <4D7254E9.6090605@gmail.com>
>>
>>+static inline void bond_set_active_slave(struct slave *slave)
>>+{
>>+ slave->backup = 0;
>
>In the comment above, you said that the possible value for backup
>corresponds with BOND_STATE_ACTIVE and BOND_STATE_BACKUP.
>
>So, should be:
>
>slave->backup = BOND_STATE_ACTIVE;
>
>>+}
>>+
>>+static inline void bond_set_backup_slave(struct slave *slave)
>>+{
>>+ slave->backup = 1;
>
>slave->backup = BOND_STATE_BACKUP;
>
Well, I think it's weird and misleading to assign some define to :1
bitfield. Should be 0 or 1, nothing else.
^ permalink raw reply
* Re: [PATCH] drivers/net/macvtap: fix error check
From: Arnd Bergmann @ 2011-03-07 9:59 UTC (permalink / raw)
To: Nicolas Kaiser; +Cc: David S. Miller, Eric Dumazet, netdev, linux-kernel
In-Reply-To: <20110305004941.7b195e9d@absol.kitzblitz>
On Saturday 05 March 2011, Nicolas Kaiser wrote:
> 'len' is unsigned of type size_t and can't be negative.
>
> Signed-off-by: Nicolas Kaiser <nikai@nikai.net>
Acked-by: Arnd Bergmann <arnd@arndb.de>
I think it's harmless: the worst thing that can happen is
macvtap_alloc_skb() failing with ENOMEM when it gets a
large argument, but we could have it in -stable just to
be sure.
Arnd
^ permalink raw reply
* Re: [PATCH net-next-2.6] net: enhance the documentation for rx_handler.
From: Jiri Pirko @ 2011-03-07 9:54 UTC (permalink / raw)
To: Nicolas de Pesloüan
Cc: netdev, davem, shemminger, eric.dumazet, kaber, fubar, andy
In-Reply-To: <1299441608-25482-1-git-send-email-nicolas.2p.debian@free.fr>
Sun, Mar 06, 2011 at 09:00:08PM CET, nicolas.2p.debian@free.fr wrote:
>Signed-off-by: Nicolas de Pesloüan <nicolas.2p.debian@free.fr>
>---
>This apply on top of Jiri's last patch serie, including the last one that
>commented the RX_HANDLER_* values.
>
> include/linux/netdevice.h | 53 ++++++++++++++++++++++++++++++++++++++------
> net/core/dev.c | 2 +
> 2 files changed, 47 insertions(+), 8 deletions(-)
>
>diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
>index 26e03f9..0c9dc93 100644
>--- a/include/linux/netdevice.h
>+++ b/include/linux/netdevice.h
>@@ -390,15 +390,52 @@ enum gro_result {
> };
> typedef enum gro_result gro_result_t;
>
>+/**
>+ * enum rx_handler_result - Possible return values for rx_handlers.
>+ * @RX_HANDLER_CONSUMED: skb was consumed by rx_handler, do not process it
>+ * further.
>+ * @RX_HANDLER_ANOTHER: Do another round in receive path. This is indicated in
>+ * case skb->dev was changed by rx_handler.
>+ * @RX_HANDLER_EXACT: Force exact delivery, no wildcard.
>+ * @RX_HANDLER_PASS: Do nothing, passe the skb as if no rx_handler was called.
>+ *
>+ * rx_handlers are functions called from inside __netif_receive_skb(), to do
>+ * special processing of the skb, prior to delivery to protocol handlers.
>+ *
>+ * Currently, a net_device can only have a single rx_handler registered. Trying
>+ * to register a second rx_handler will return -EBUSY.
>+ *
>+ * To register a rx_handler on a net_device, use netdev_rx_handler_register().
>+ * To unregister a rx_handler on a net_device, use
>+ * netdev_rx_handler_unregister().
>+ *
>+ * Upon return, rx_handler is expected to tell __netif_receive_skb() what to
>+ * do with the skb.
>+ *
>+ * If the rx_handler consumed to skb in some way, it should return
>+ * RX_HANDLER_CONSUMED. This is appropriate when the rx_handler arranged for
>+ * the skb to be delivered in some other ways.
>+ *
>+ * If the rx_handler changed skb->dev, to divert the skb to another
>+ * net_device, it should return RX_HANDLER_ANOTHER. The rx_handler for the
>+ * new device will be called if it exists.
>+ *
>+ * If the rx_handler consider the skb should be ignored, it should return
>+ * RX_HANDLER_EXACT. The skb will only be delivered to protocol handlers that
>+ * are registred on exact device (ptype->dev == skb->dev).
>+ *
>+ * If the rx_handler didn't changed skb->dev, but want the skb to be normally
>+ * delivered, it should return RX_HANDLER_PASS.
>+ *
>+ * A device without a registered rx_handler will behave as if rx_handler
>+ * returned RX_HANDLER_PASS.
>+ */
>+
> enum rx_handler_result {
>- RX_HANDLER_CONSUMED, /* skb was consumed by rx_handler,
>- do not process it further. */
>- RX_HANDLER_ANOTHER, /* Do another round in receive path.
>- This is indicated in case skb->dev
>- was changed by rx_handler */
>- RX_HANDLER_EXACT, /* Force exact delivery, no wildcard */
>- RX_HANDLER_PASS, /* Do nothing, pass the skb as if
>- no rx_handler was called */
>+ RX_HANDLER_CONSUMED,
>+ RX_HANDLER_ANOTHER,
>+ RX_HANDLER_EXACT,
>+ RX_HANDLER_PASS,
> };
> typedef enum rx_handler_result rx_handler_result_t;
> typedef rx_handler_result_t rx_handler_func_t(struct sk_buff **pskb);
>diff --git a/net/core/dev.c b/net/core/dev.c
>index a368223..3630722 100644
>--- a/net/core/dev.c
>+++ b/net/core/dev.c
>@@ -3060,6 +3060,8 @@ out:
> * on a failure.
> *
> * The caller must hold the rtnl_mutex.
>+ *
>+ * For a general description of rx_handler, see enum rx_handler_result.
> */
> int netdev_rx_handler_register(struct net_device *dev,
> rx_handler_func_t *rx_handler,
>--
>1.7.2.3
>
>--
>To unsubscribe from this list: send the line "unsubscribe netdev" in
>the body of a message to majordomo@vger.kernel.org
>More majordomo info at http://vger.kernel.org/majordomo-info.html
Reviewed-by: Jiri Pirko <jpirko@redhat.com>
^ permalink raw reply
* Re: [PATCH] pktgen: fix errata in show results
From: Daniel Turull @ 2011-03-07 9:44 UTC (permalink / raw)
To: David Miller; +Cc: netdev, ljw, robert, Daniel Turull
In-Reply-To: <20110307.011942.115928916.davem@davemloft.net>
The units in show_results in pktgen were not correct.
The results are in usec but it was displayed nsec.
Reported-by: Jong-won Lee <ljw@handong.edu>
Signed-off-by: Daniel Turull <daniel.turull@gmail.com>
---
Here it is the new version.
Sorry for the previous mail.
Daniel
---
diff --git a/net/core/pktgen.c b/net/core/pktgen.c
index d73b77a..f0aec6c 100644
--- a/net/core/pktgen.c
+++ b/net/core/pktgen.c
@@ -3271,7 +3271,7 @@ static void show_results(struct pktgen_dev *pkt_dev, int nr_frags)
pkt_dev->started_at);
ktime_t idle = ns_to_ktime(pkt_dev->idle_acc);
- p += sprintf(p, "OK: %llu(c%llu+d%llu) nsec, %llu (%dbyte,%dfrags)\n",
+ p += sprintf(p, "OK: %llu(c%llu+d%llu) usec, %llu (%dbyte,%dfrags)\n",
(unsigned long long)ktime_to_us(elapsed),
(unsigned long long)ktime_to_us(ktime_sub(elapsed, idle)),
(unsigned long long)ktime_to_us(idle),
^ permalink raw reply related
* Re: mii_bus->read return checking in phy_device.c
From: Florian Fainelli @ 2011-03-07 9:43 UTC (permalink / raw)
To: Fleming Andy-AFLEMING; +Cc: netdev@vger.kernel.org, David Miller
In-Reply-To: <BB72AC9C-654E-43CE-9361-255767DCFC31@freescale.com>
On Friday 04 March 2011 19:19:10 Fleming Andy-AFLEMING wrote:
> On Mar 4, 2011, at 12:10, "Florian Fainelli" <florian@openwrt.org> wrote:
> > On Friday 04 March 2011 19:06:20 Fleming Andy-AFLEMING wrote:
> >> On Mar 4, 2011, at 11:24, "Florian Fainelli" <florian@openwrt.org> wrote:
> >>> Hello Andy,
> >>>
> >>> While debugging a PHY probing issue with the au1000_eth, I stumbled
> >>> upon this
> >>>
> >>> in drivers/net/phy/phy_device.c:
> >>> phy_reg = bus->read(bus, addr, MII_PHYSID1);
> >>>
> >>> if (phy_reg < 0)
> >>>
> >>> return -EIO;
> >>>
> >>> most drivers implement phylib's mdio_read callback by simply returning
> >>> the contents of their MDIO register after a readl, ioread ... which is
> >>> unsigned. Would not it rather make sense to check for phy_reg <= 0
> >>> instead?
> >>
> >> That isn't a check for a non-existent PHY. PHY registers are unsigned
> >> 16-bit quantities. The negative 32-bit return value would be the result
> >> of something going wrong in the bus transaction.
> >
> > Ok, but 0 is not an acceptable value either for both ID1 and ID2.
>
> I don't remember the exact details, but i recall we had a discussion about
> this several years ago, and decided that 0 should not be interpreted as a
> non-existent PHY. I know I have a part that has an internal PHY which
> doesnt have anything in the ID registers. If your driver is aware that it
> did not get a response from the PHY, it should return 0xffff. Otherwise,
> you can return 0, and just be aware that the PHY subsystem will believe
> there's a PHY there.
Allright, thanks for the clarification, I have sorted this in platform code
registering the particular ethernet driver.
--
Florian
^ permalink raw reply
* Re: [PATCH] pktgen: fix errata in show results
From: David Miller @ 2011-03-07 9:19 UTC (permalink / raw)
To: daniel.turull; +Cc: netdev, ljw, robert
In-Reply-To: <4D74A0D7.3020506@gmail.com>
From: Daniel Turull <daniel.turull@gmail.com>
Date: Mon, 07 Mar 2011 10:09:43 +0100
> The units in show_results in pktgen were not correct.
> The results are in usec but it was displayed nsec.
>
> Reported-by: Jong-won Lee <ljw@handong.edu>
> Signed-off-by: Daniel Turull <daniel.turull@gmail.com>
> ---
Your email client corrupted the patch.
^ permalink raw reply
* Re: pull request: batman-adv 2011-03-05
From: David Miller @ 2011-03-07 9:19 UTC (permalink / raw)
To: sven; +Cc: netdev, b.a.t.m.a.n
In-Reply-To: <201103071001.39111.sven@narfation.org>
From: Sven Eckelmann <sven@narfation.org>
Date: Mon, 7 Mar 2011 10:01:36 +0100
> On Monday 07 March 2011 03:14:32 David Miller wrote:
>> Can you like sync with me when you have less than 20+ patches queued
>> up?
>>
>> It's too much at once to reasonably review, and makes regressions take
>> longer to bisect when people hit them.
>
> I'll try. The problem is that most of it is one work package (the removal of
> orig_hash) which I got in this amount of patches.
>
> What do you think about another pull request which only has some of the
> cleanup/fixes and leave the rest for the time after 2.6.39-rc1 - so it may get
> into 2.6.40?
I already pulled this stuff into net-next-2.6
^ permalink raw reply
* [PATCH] pktgen: fix errata in show results
From: Daniel Turull @ 2011-03-07 9:09 UTC (permalink / raw)
To: netdev; +Cc: 이종원, Robert Olsson
The units in show_results in pktgen were not correct.
The results are in usec but it was displayed nsec.
Reported-by: Jong-won Lee <ljw@handong.edu>
Signed-off-by: Daniel Turull <daniel.turull@gmail.com>
---
diff --git a/net/core/pktgen.c b/net/core/pktgen.c
index d73b77a..f0aec6c 100644
--- a/net/core/pktgen.c
+++ b/net/core/pktgen.c
@@ -3271,7 +3271,7 @@ static void show_results(struct pktgen_dev
*pkt_dev, int nr_frags)
pkt_dev->started_at);
ktime_t idle = ns_to_ktime(pkt_dev->idle_acc);
- p += sprintf(p, "OK: %llu(c%llu+d%llu) nsec, %llu (%dbyte,%dfrags)\n",
+ p += sprintf(p, "OK: %llu(c%llu+d%llu) usec, %llu (%dbyte,%dfrags)\n",
(unsigned long long)ktime_to_us(elapsed),
(unsigned long long)ktime_to_us(ktime_sub(elapsed, idle)),
(unsigned long long)ktime_to_us(idle),
^ permalink raw reply related
* Re: pull request: batman-adv 2011-03-05
From: Sven Eckelmann @ 2011-03-07 9:01 UTC (permalink / raw)
To: David Miller; +Cc: netdev, b.a.t.m.a.n
In-Reply-To: <20110306.181432.104062354.davem@davemloft.net>
[-- Attachment #1: Type: Text/Plain, Size: 580 bytes --]
On Monday 07 March 2011 03:14:32 David Miller wrote:
> Can you like sync with me when you have less than 20+ patches queued
> up?
>
> It's too much at once to reasonably review, and makes regressions take
> longer to bisect when people hit them.
I'll try. The problem is that most of it is one work package (the removal of
orig_hash) which I got in this amount of patches.
What do you think about another pull request which only has some of the
cleanup/fixes and leave the rest for the time after 2.6.39-rc1 - so it may get
into 2.6.40?
Best regards,
Sven
[-- Attachment #2: This is a digitally signed message part. --]
[-- Type: application/pgp-signature, Size: 836 bytes --]
^ permalink raw reply
* Re: [V2 Patch] bonding: move procfs code into bond_procfs.c
From: Jiri Pirko @ 2011-03-07 7:49 UTC (permalink / raw)
To: Amerigo Wang; +Cc: linux-kernel, davem, Jay Vosburgh, netdev
In-Reply-To: <1299484726-29905-1-git-send-email-amwang@redhat.com>
Mon, Mar 07, 2011 at 08:58:46AM CET, amwang@redhat.com wrote:
>V2: Move #ifdef CONFIG_PROC_FS into bonding.h, as suggested by David.
>
>bond_main.c is bloating, separate the procfs code out,
>move them to bond_procfs.c
>
>Signed-off-by: WANG Cong <amwang@redhat.com>
>
>---
> drivers/net/bonding/Makefile | 3 +
> drivers/net/bonding/bond_main.c | 302 +------------------------------------
> drivers/net/bonding/bond_procfs.c | 276 +++++++++++++++++++++++++++++++++
> drivers/net/bonding/bonding.h | 26 +++
> 4 files changed, 307 insertions(+), 300 deletions(-)
>
Looking good to me
Reviewed-by: Jiri Pirko <jpirko@redhat.com>
^ permalink raw reply
* Re: [PATCH 1/2] Issue NETDEV_CHANGE notification when bridge changes state
From: Nicolas de Pesloüan @ 2011-03-07 7:44 UTC (permalink / raw)
To: Stephen Hemminger
Cc: Adam Majer, David S. Miller, Alexey Kuznetsov,
Pekka Savola (ipv6), James Morris, Hideaki YOSHIFUJI,
Patrick McHardy, bridge, netdev, Andy Gospodarek, Jay Vosburgh
In-Reply-To: <1649722795.14144.1299480074110.JavaMail.root@tahiti.vyatta.com>
Le 07/03/2011 07:41, Stephen Hemminger a écrit :
>
>> On Sun, Mar 06, 2011 at 09:45:41AM -0800, Stephen Hemminger wrote:
>>> Since this a generic problem, it needs a better solution.
>>> Sending NETDEV_CHANGE impacts lots of other pieces, and even
>>> user space has similar problems.
>>
>> It does seem a little broad notification type. I've checked over
>> all the currently defined NETDEV notifiers, and it seems that
>> NETDEV_NOTIFY_PEERS may be a better option to use when bridge
>> has a potential topology change.
>>
>> Currently it is only used in ipv4/devinet.c: where it is used to issue
>> a gratuitous ARP.
>
> I was thinking of fixing bridge to not actually bring the link
> up until in forwarding mode. Other applications (DHCP, etc)
> see the link up and really don't like being in half duplex
> during that period.
I think it is the right way to manage this situation. And bonding should behave the same, if not
already true.
Nicolas.
^ permalink raw reply
* [V2 Patch] bonding: move procfs code into bond_procfs.c
From: Amerigo Wang @ 2011-03-07 7:58 UTC (permalink / raw)
To: linux-kernel; +Cc: davem, WANG Cong, Jay Vosburgh, netdev
V2: Move #ifdef CONFIG_PROC_FS into bonding.h, as suggested by David.
bond_main.c is bloating, separate the procfs code out,
move them to bond_procfs.c
Signed-off-by: WANG Cong <amwang@redhat.com>
---
drivers/net/bonding/Makefile | 3 +
drivers/net/bonding/bond_main.c | 302 +------------------------------------
drivers/net/bonding/bond_procfs.c | 276 +++++++++++++++++++++++++++++++++
drivers/net/bonding/bonding.h | 26 +++
4 files changed, 307 insertions(+), 300 deletions(-)
diff --git a/drivers/net/bonding/Makefile b/drivers/net/bonding/Makefile
index 0e2737e..3c5c014 100644
--- a/drivers/net/bonding/Makefile
+++ b/drivers/net/bonding/Makefile
@@ -6,6 +6,9 @@ obj-$(CONFIG_BONDING) += bonding.o
bonding-objs := bond_main.o bond_3ad.o bond_alb.o bond_sysfs.o bond_debugfs.o
+proc-$(CONFIG_PROC_FS) += bond_procfs.o
+bonding-objs += $(proc-y)
+
ipv6-$(subst m,y,$(CONFIG_IPV6)) += bond_ipv6.o
bonding-objs += $(ipv6-y)
diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c
index 584f97b..7abed73 100644
--- a/drivers/net/bonding/bond_main.c
+++ b/drivers/net/bonding/bond_main.c
@@ -65,8 +65,6 @@
#include <linux/skbuff.h>
#include <net/sock.h>
#include <linux/rtnetlink.h>
-#include <linux/proc_fs.h>
-#include <linux/seq_file.h>
#include <linux/smp.h>
#include <linux/if_ether.h>
#include <net/arp.h>
@@ -173,9 +171,6 @@ MODULE_PARM_DESC(resend_igmp, "Number of IGMP membership reports to send on link
atomic_t netpoll_block_tx = ATOMIC_INIT(0);
#endif
-static const char * const version =
- DRV_DESCRIPTION ": v" DRV_VERSION " (" DRV_RELDATE ")\n";
-
int bond_net_id __read_mostly;
static __be32 arp_target[BOND_MAX_ARP_TARGETS];
@@ -245,7 +240,7 @@ static void bond_uninit(struct net_device *bond_dev);
/*---------------------------- General routines -----------------------------*/
-static const char *bond_mode_name(int mode)
+const char *bond_mode_name(int mode)
{
static const char *names[] = {
[BOND_MODE_ROUNDROBIN] = "load balancing (round-robin)",
@@ -3292,299 +3287,6 @@ out:
read_unlock(&bond->lock);
}
-/*------------------------------ proc/seq_file-------------------------------*/
-
-#ifdef CONFIG_PROC_FS
-
-static void *bond_info_seq_start(struct seq_file *seq, loff_t *pos)
- __acquires(RCU)
- __acquires(&bond->lock)
-{
- struct bonding *bond = seq->private;
- loff_t off = 0;
- struct slave *slave;
- int i;
-
- /* make sure the bond won't be taken away */
- rcu_read_lock();
- read_lock(&bond->lock);
-
- if (*pos == 0)
- return SEQ_START_TOKEN;
-
- bond_for_each_slave(bond, slave, i) {
- if (++off == *pos)
- return slave;
- }
-
- return NULL;
-}
-
-static void *bond_info_seq_next(struct seq_file *seq, void *v, loff_t *pos)
-{
- struct bonding *bond = seq->private;
- struct slave *slave = v;
-
- ++*pos;
- if (v == SEQ_START_TOKEN)
- return bond->first_slave;
-
- slave = slave->next;
-
- return (slave == bond->first_slave) ? NULL : slave;
-}
-
-static void bond_info_seq_stop(struct seq_file *seq, void *v)
- __releases(&bond->lock)
- __releases(RCU)
-{
- struct bonding *bond = seq->private;
-
- read_unlock(&bond->lock);
- rcu_read_unlock();
-}
-
-static void bond_info_show_master(struct seq_file *seq)
-{
- struct bonding *bond = seq->private;
- struct slave *curr;
- int i;
-
- read_lock(&bond->curr_slave_lock);
- curr = bond->curr_active_slave;
- read_unlock(&bond->curr_slave_lock);
-
- seq_printf(seq, "Bonding Mode: %s",
- bond_mode_name(bond->params.mode));
-
- if (bond->params.mode == BOND_MODE_ACTIVEBACKUP &&
- bond->params.fail_over_mac)
- seq_printf(seq, " (fail_over_mac %s)",
- fail_over_mac_tbl[bond->params.fail_over_mac].modename);
-
- seq_printf(seq, "\n");
-
- if (bond->params.mode == BOND_MODE_XOR ||
- bond->params.mode == BOND_MODE_8023AD) {
- seq_printf(seq, "Transmit Hash Policy: %s (%d)\n",
- xmit_hashtype_tbl[bond->params.xmit_policy].modename,
- bond->params.xmit_policy);
- }
-
- if (USES_PRIMARY(bond->params.mode)) {
- seq_printf(seq, "Primary Slave: %s",
- (bond->primary_slave) ?
- bond->primary_slave->dev->name : "None");
- if (bond->primary_slave)
- seq_printf(seq, " (primary_reselect %s)",
- pri_reselect_tbl[bond->params.primary_reselect].modename);
-
- seq_printf(seq, "\nCurrently Active Slave: %s\n",
- (curr) ? curr->dev->name : "None");
- }
-
- seq_printf(seq, "MII Status: %s\n", netif_carrier_ok(bond->dev) ?
- "up" : "down");
- seq_printf(seq, "MII Polling Interval (ms): %d\n", bond->params.miimon);
- seq_printf(seq, "Up Delay (ms): %d\n",
- bond->params.updelay * bond->params.miimon);
- seq_printf(seq, "Down Delay (ms): %d\n",
- bond->params.downdelay * bond->params.miimon);
-
-
- /* ARP information */
- if (bond->params.arp_interval > 0) {
- int printed = 0;
- seq_printf(seq, "ARP Polling Interval (ms): %d\n",
- bond->params.arp_interval);
-
- seq_printf(seq, "ARP IP target/s (n.n.n.n form):");
-
- for (i = 0; (i < BOND_MAX_ARP_TARGETS); i++) {
- if (!bond->params.arp_targets[i])
- break;
- if (printed)
- seq_printf(seq, ",");
- seq_printf(seq, " %pI4", &bond->params.arp_targets[i]);
- printed = 1;
- }
- seq_printf(seq, "\n");
- }
-
- if (bond->params.mode == BOND_MODE_8023AD) {
- struct ad_info ad_info;
-
- seq_puts(seq, "\n802.3ad info\n");
- seq_printf(seq, "LACP rate: %s\n",
- (bond->params.lacp_fast) ? "fast" : "slow");
- seq_printf(seq, "Aggregator selection policy (ad_select): %s\n",
- ad_select_tbl[bond->params.ad_select].modename);
-
- if (bond_3ad_get_active_agg_info(bond, &ad_info)) {
- seq_printf(seq, "bond %s has no active aggregator\n",
- bond->dev->name);
- } else {
- seq_printf(seq, "Active Aggregator Info:\n");
-
- seq_printf(seq, "\tAggregator ID: %d\n",
- ad_info.aggregator_id);
- seq_printf(seq, "\tNumber of ports: %d\n",
- ad_info.ports);
- seq_printf(seq, "\tActor Key: %d\n",
- ad_info.actor_key);
- seq_printf(seq, "\tPartner Key: %d\n",
- ad_info.partner_key);
- seq_printf(seq, "\tPartner Mac Address: %pM\n",
- ad_info.partner_system);
- }
- }
-}
-
-static void bond_info_show_slave(struct seq_file *seq,
- const struct slave *slave)
-{
- struct bonding *bond = seq->private;
-
- seq_printf(seq, "\nSlave Interface: %s\n", slave->dev->name);
- seq_printf(seq, "MII Status: %s\n",
- (slave->link == BOND_LINK_UP) ? "up" : "down");
- seq_printf(seq, "Speed: %d Mbps\n", slave->speed);
- seq_printf(seq, "Duplex: %s\n", slave->duplex ? "full" : "half");
- seq_printf(seq, "Link Failure Count: %u\n",
- slave->link_failure_count);
-
- seq_printf(seq, "Permanent HW addr: %pM\n", slave->perm_hwaddr);
-
- if (bond->params.mode == BOND_MODE_8023AD) {
- const struct aggregator *agg
- = SLAVE_AD_INFO(slave).port.aggregator;
-
- if (agg)
- seq_printf(seq, "Aggregator ID: %d\n",
- agg->aggregator_identifier);
- else
- seq_puts(seq, "Aggregator ID: N/A\n");
- }
- seq_printf(seq, "Slave queue ID: %d\n", slave->queue_id);
-}
-
-static int bond_info_seq_show(struct seq_file *seq, void *v)
-{
- if (v == SEQ_START_TOKEN) {
- seq_printf(seq, "%s\n", version);
- bond_info_show_master(seq);
- } else
- bond_info_show_slave(seq, v);
-
- return 0;
-}
-
-static const struct seq_operations bond_info_seq_ops = {
- .start = bond_info_seq_start,
- .next = bond_info_seq_next,
- .stop = bond_info_seq_stop,
- .show = bond_info_seq_show,
-};
-
-static int bond_info_open(struct inode *inode, struct file *file)
-{
- struct seq_file *seq;
- struct proc_dir_entry *proc;
- int res;
-
- res = seq_open(file, &bond_info_seq_ops);
- if (!res) {
- /* recover the pointer buried in proc_dir_entry data */
- seq = file->private_data;
- proc = PDE(inode);
- seq->private = proc->data;
- }
-
- return res;
-}
-
-static const struct file_operations bond_info_fops = {
- .owner = THIS_MODULE,
- .open = bond_info_open,
- .read = seq_read,
- .llseek = seq_lseek,
- .release = seq_release,
-};
-
-static void bond_create_proc_entry(struct bonding *bond)
-{
- struct net_device *bond_dev = bond->dev;
- struct bond_net *bn = net_generic(dev_net(bond_dev), bond_net_id);
-
- if (bn->proc_dir) {
- bond->proc_entry = proc_create_data(bond_dev->name,
- S_IRUGO, bn->proc_dir,
- &bond_info_fops, bond);
- if (bond->proc_entry == NULL)
- pr_warning("Warning: Cannot create /proc/net/%s/%s\n",
- DRV_NAME, bond_dev->name);
- else
- memcpy(bond->proc_file_name, bond_dev->name, IFNAMSIZ);
- }
-}
-
-static void bond_remove_proc_entry(struct bonding *bond)
-{
- struct net_device *bond_dev = bond->dev;
- struct bond_net *bn = net_generic(dev_net(bond_dev), bond_net_id);
-
- if (bn->proc_dir && bond->proc_entry) {
- remove_proc_entry(bond->proc_file_name, bn->proc_dir);
- memset(bond->proc_file_name, 0, IFNAMSIZ);
- bond->proc_entry = NULL;
- }
-}
-
-/* Create the bonding directory under /proc/net, if doesn't exist yet.
- * Caller must hold rtnl_lock.
- */
-static void __net_init bond_create_proc_dir(struct bond_net *bn)
-{
- if (!bn->proc_dir) {
- bn->proc_dir = proc_mkdir(DRV_NAME, bn->net->proc_net);
- if (!bn->proc_dir)
- pr_warning("Warning: cannot create /proc/net/%s\n",
- DRV_NAME);
- }
-}
-
-/* Destroy the bonding directory under /proc/net, if empty.
- * Caller must hold rtnl_lock.
- */
-static void __net_exit bond_destroy_proc_dir(struct bond_net *bn)
-{
- if (bn->proc_dir) {
- remove_proc_entry(DRV_NAME, bn->net->proc_net);
- bn->proc_dir = NULL;
- }
-}
-
-#else /* !CONFIG_PROC_FS */
-
-static void bond_create_proc_entry(struct bonding *bond)
-{
-}
-
-static void bond_remove_proc_entry(struct bonding *bond)
-{
-}
-
-static inline void bond_create_proc_dir(struct bond_net *bn)
-{
-}
-
-static inline void bond_destroy_proc_dir(struct bond_net *bn)
-{
-}
-
-#endif /* CONFIG_PROC_FS */
-
-
/*-------------------------- netdev event handling --------------------------*/
/*
@@ -5388,7 +5090,7 @@ static int __init bonding_init(void)
int i;
int res;
- pr_info("%s", version);
+ pr_info("%s", bond_version);
res = bond_check_params(&bonding_defaults);
if (res)
diff --git a/drivers/net/bonding/bond_procfs.c b/drivers/net/bonding/bond_procfs.c
new file mode 100644
index 0000000..c18a7ce
--- /dev/null
+++ b/drivers/net/bonding/bond_procfs.c
@@ -0,0 +1,276 @@
+#include <linux/proc_fs.h>
+#include <net/net_namespace.h>
+#include <net/netns/generic.h>
+#include "bonding.h"
+
+
+extern const char *bond_mode_name(int mode);
+
+static void *bond_info_seq_start(struct seq_file *seq, loff_t *pos)
+ __acquires(RCU)
+ __acquires(&bond->lock)
+{
+ struct bonding *bond = seq->private;
+ loff_t off = 0;
+ struct slave *slave;
+ int i;
+
+ /* make sure the bond won't be taken away */
+ rcu_read_lock();
+ read_lock(&bond->lock);
+
+ if (*pos == 0)
+ return SEQ_START_TOKEN;
+
+ bond_for_each_slave(bond, slave, i) {
+ if (++off == *pos)
+ return slave;
+ }
+
+ return NULL;
+}
+
+static void *bond_info_seq_next(struct seq_file *seq, void *v, loff_t *pos)
+{
+ struct bonding *bond = seq->private;
+ struct slave *slave = v;
+
+ ++*pos;
+ if (v == SEQ_START_TOKEN)
+ return bond->first_slave;
+
+ slave = slave->next;
+
+ return (slave == bond->first_slave) ? NULL : slave;
+}
+
+static void bond_info_seq_stop(struct seq_file *seq, void *v)
+ __releases(&bond->lock)
+ __releases(RCU)
+{
+ struct bonding *bond = seq->private;
+
+ read_unlock(&bond->lock);
+ rcu_read_unlock();
+}
+
+static void bond_info_show_master(struct seq_file *seq)
+{
+ struct bonding *bond = seq->private;
+ struct slave *curr;
+ int i;
+
+ read_lock(&bond->curr_slave_lock);
+ curr = bond->curr_active_slave;
+ read_unlock(&bond->curr_slave_lock);
+
+ seq_printf(seq, "Bonding Mode: %s",
+ bond_mode_name(bond->params.mode));
+
+ if (bond->params.mode == BOND_MODE_ACTIVEBACKUP &&
+ bond->params.fail_over_mac)
+ seq_printf(seq, " (fail_over_mac %s)",
+ fail_over_mac_tbl[bond->params.fail_over_mac].modename);
+
+ seq_printf(seq, "\n");
+
+ if (bond->params.mode == BOND_MODE_XOR ||
+ bond->params.mode == BOND_MODE_8023AD) {
+ seq_printf(seq, "Transmit Hash Policy: %s (%d)\n",
+ xmit_hashtype_tbl[bond->params.xmit_policy].modename,
+ bond->params.xmit_policy);
+ }
+
+ if (USES_PRIMARY(bond->params.mode)) {
+ seq_printf(seq, "Primary Slave: %s",
+ (bond->primary_slave) ?
+ bond->primary_slave->dev->name : "None");
+ if (bond->primary_slave)
+ seq_printf(seq, " (primary_reselect %s)",
+ pri_reselect_tbl[bond->params.primary_reselect].modename);
+
+ seq_printf(seq, "\nCurrently Active Slave: %s\n",
+ (curr) ? curr->dev->name : "None");
+ }
+
+ seq_printf(seq, "MII Status: %s\n", netif_carrier_ok(bond->dev) ?
+ "up" : "down");
+ seq_printf(seq, "MII Polling Interval (ms): %d\n", bond->params.miimon);
+ seq_printf(seq, "Up Delay (ms): %d\n",
+ bond->params.updelay * bond->params.miimon);
+ seq_printf(seq, "Down Delay (ms): %d\n",
+ bond->params.downdelay * bond->params.miimon);
+
+
+ /* ARP information */
+ if (bond->params.arp_interval > 0) {
+ int printed = 0;
+ seq_printf(seq, "ARP Polling Interval (ms): %d\n",
+ bond->params.arp_interval);
+
+ seq_printf(seq, "ARP IP target/s (n.n.n.n form):");
+
+ for (i = 0; (i < BOND_MAX_ARP_TARGETS); i++) {
+ if (!bond->params.arp_targets[i])
+ break;
+ if (printed)
+ seq_printf(seq, ",");
+ seq_printf(seq, " %pI4", &bond->params.arp_targets[i]);
+ printed = 1;
+ }
+ seq_printf(seq, "\n");
+ }
+
+ if (bond->params.mode == BOND_MODE_8023AD) {
+ struct ad_info ad_info;
+
+ seq_puts(seq, "\n802.3ad info\n");
+ seq_printf(seq, "LACP rate: %s\n",
+ (bond->params.lacp_fast) ? "fast" : "slow");
+ seq_printf(seq, "Aggregator selection policy (ad_select): %s\n",
+ ad_select_tbl[bond->params.ad_select].modename);
+
+ if (bond_3ad_get_active_agg_info(bond, &ad_info)) {
+ seq_printf(seq, "bond %s has no active aggregator\n",
+ bond->dev->name);
+ } else {
+ seq_printf(seq, "Active Aggregator Info:\n");
+
+ seq_printf(seq, "\tAggregator ID: %d\n",
+ ad_info.aggregator_id);
+ seq_printf(seq, "\tNumber of ports: %d\n",
+ ad_info.ports);
+ seq_printf(seq, "\tActor Key: %d\n",
+ ad_info.actor_key);
+ seq_printf(seq, "\tPartner Key: %d\n",
+ ad_info.partner_key);
+ seq_printf(seq, "\tPartner Mac Address: %pM\n",
+ ad_info.partner_system);
+ }
+ }
+}
+
+static void bond_info_show_slave(struct seq_file *seq,
+ const struct slave *slave)
+{
+ struct bonding *bond = seq->private;
+
+ seq_printf(seq, "\nSlave Interface: %s\n", slave->dev->name);
+ seq_printf(seq, "MII Status: %s\n",
+ (slave->link == BOND_LINK_UP) ? "up" : "down");
+ seq_printf(seq, "Speed: %d Mbps\n", slave->speed);
+ seq_printf(seq, "Duplex: %s\n", slave->duplex ? "full" : "half");
+ seq_printf(seq, "Link Failure Count: %u\n",
+ slave->link_failure_count);
+
+ seq_printf(seq, "Permanent HW addr: %pM\n", slave->perm_hwaddr);
+
+ if (bond->params.mode == BOND_MODE_8023AD) {
+ const struct aggregator *agg
+ = SLAVE_AD_INFO(slave).port.aggregator;
+
+ if (agg)
+ seq_printf(seq, "Aggregator ID: %d\n",
+ agg->aggregator_identifier);
+ else
+ seq_puts(seq, "Aggregator ID: N/A\n");
+ }
+ seq_printf(seq, "Slave queue ID: %d\n", slave->queue_id);
+}
+
+static int bond_info_seq_show(struct seq_file *seq, void *v)
+{
+ if (v == SEQ_START_TOKEN) {
+ seq_printf(seq, "%s\n", bond_version);
+ bond_info_show_master(seq);
+ } else
+ bond_info_show_slave(seq, v);
+
+ return 0;
+}
+
+static const struct seq_operations bond_info_seq_ops = {
+ .start = bond_info_seq_start,
+ .next = bond_info_seq_next,
+ .stop = bond_info_seq_stop,
+ .show = bond_info_seq_show,
+};
+
+static int bond_info_open(struct inode *inode, struct file *file)
+{
+ struct seq_file *seq;
+ struct proc_dir_entry *proc;
+ int res;
+
+ res = seq_open(file, &bond_info_seq_ops);
+ if (!res) {
+ /* recover the pointer buried in proc_dir_entry data */
+ seq = file->private_data;
+ proc = PDE(inode);
+ seq->private = proc->data;
+ }
+
+ return res;
+}
+
+static const struct file_operations bond_info_fops = {
+ .owner = THIS_MODULE,
+ .open = bond_info_open,
+ .read = seq_read,
+ .llseek = seq_lseek,
+ .release = seq_release,
+};
+
+void bond_create_proc_entry(struct bonding *bond)
+{
+ struct net_device *bond_dev = bond->dev;
+ struct bond_net *bn = net_generic(dev_net(bond_dev), bond_net_id);
+
+ if (bn->proc_dir) {
+ bond->proc_entry = proc_create_data(bond_dev->name,
+ S_IRUGO, bn->proc_dir,
+ &bond_info_fops, bond);
+ if (bond->proc_entry == NULL)
+ pr_warning("Warning: Cannot create /proc/net/%s/%s\n",
+ DRV_NAME, bond_dev->name);
+ else
+ memcpy(bond->proc_file_name, bond_dev->name, IFNAMSIZ);
+ }
+}
+
+void bond_remove_proc_entry(struct bonding *bond)
+{
+ struct net_device *bond_dev = bond->dev;
+ struct bond_net *bn = net_generic(dev_net(bond_dev), bond_net_id);
+
+ if (bn->proc_dir && bond->proc_entry) {
+ remove_proc_entry(bond->proc_file_name, bn->proc_dir);
+ memset(bond->proc_file_name, 0, IFNAMSIZ);
+ bond->proc_entry = NULL;
+ }
+}
+
+/* Create the bonding directory under /proc/net, if doesn't exist yet.
+ * Caller must hold rtnl_lock.
+ */
+void __net_init bond_create_proc_dir(struct bond_net *bn)
+{
+ if (!bn->proc_dir) {
+ bn->proc_dir = proc_mkdir(DRV_NAME, bn->net->proc_net);
+ if (!bn->proc_dir)
+ pr_warning("Warning: cannot create /proc/net/%s\n",
+ DRV_NAME);
+ }
+}
+
+/* Destroy the bonding directory under /proc/net, if empty.
+ * Caller must hold rtnl_lock.
+ */
+void __net_exit bond_destroy_proc_dir(struct bond_net *bn)
+{
+ if (bn->proc_dir) {
+ remove_proc_entry(DRV_NAME, bn->net->proc_net);
+ bn->proc_dir = NULL;
+ }
+}
+
diff --git a/drivers/net/bonding/bonding.h b/drivers/net/bonding/bonding.h
index a401b8d..8e8691a 100644
--- a/drivers/net/bonding/bonding.h
+++ b/drivers/net/bonding/bonding.h
@@ -29,6 +29,8 @@
#define DRV_NAME "bonding"
#define DRV_DESCRIPTION "Ethernet Channel Bonding Driver"
+#define bond_version DRV_DESCRIPTION ": v" DRV_VERSION " (" DRV_RELDATE ")\n"
+
#define BOND_MAX_ARP_TARGETS 16
#define IS_UP(dev) \
@@ -413,6 +415,30 @@ struct bond_net {
#endif
};
+#ifdef CONFIG_PROC_FS
+void bond_create_proc_entry(struct bonding *bond);
+void bond_remove_proc_entry(struct bonding *bond);
+void bond_create_proc_dir(struct bond_net *bn);
+void bond_destroy_proc_dir(struct bond_net *bn);
+#else
+static inline void bond_create_proc_entry(struct bonding *bond)
+{
+}
+
+static inline void bond_remove_proc_entry(struct bonding *bond)
+{
+}
+
+static inline void bond_create_proc_dir(struct bond_net *bn)
+{
+}
+
+static inline void bond_destroy_proc_dir(struct bond_net *bn)
+{
+}
+#endif
+
+
/* exported from bond_main.c */
extern int bond_net_id;
extern const struct bond_parm_tbl bond_lacp_tbl[];
^ permalink raw reply related
* Re: [PATCH] sctp: do not mark chunk abandoned if peer has no PRSCTP capable
From: Wei Yongjun @ 2011-03-07 7:08 UTC (permalink / raw)
To: Vlad Yasevich; +Cc: netdev@vger.kernel.org, lksctp, David Miller
In-Reply-To: <1299244083.2581.4.camel@oscar>
> On Fri, 2011-03-04 at 13:10 +0800, Wei Yongjun wrote:
>>> On 03/02/2011 11:20 PM, Wei Yongjun wrote:
>>>> Chunk is marked abandoned if the chunk is expires, and it not be
>>>> retransmited even if the peer has no PRSCTP capable, but the peer
>>>> will still wait for retransmit it to update CTSN.
>>>> This patch disable mark chunk abandoned if peer has no PRSCTP
>>>> capable.
>>>>
>>>> Signed-off-by: Wei Yongjun <yjwei@cn.fujitsu.com>
>>>> ---
>>>> net/sctp/chunk.c | 3 +++
>>>> 1 files changed, 3 insertions(+), 0 deletions(-)
>>>>
>>>> diff --git a/net/sctp/chunk.c b/net/sctp/chunk.c
>>>> index 6c85564..0d4832d 100644
>>>> --- a/net/sctp/chunk.c
>>>> +++ b/net/sctp/chunk.c
>>>> @@ -347,6 +347,9 @@ int sctp_chunk_abandoned(struct sctp_chunk *chunk)
>>>> {
>>>> struct sctp_datamsg *msg = chunk->msg;
>>>>
>>>> + if (!chunk->asoc->peer.prsctp_capable)
>>>> + return 0;
>>>> +
>>>> if (!msg->can_abandon)
>>>> return 0;
>>>>
>>> The trouble is that timetolive can be set on a message independent of Partial Reliability.
>>> The difference in behavior is that when PR can't be used, a chunk can only be abandoned
>>> if it has not yet been transmitted. With PR enabled, the chunk can be abandoned at any time.
>>>
>>> So, you can't blindly disallow abandonment.
>> But, how can we do PR if peer has no PRSCTP capable?
> That's already taken care of. We report the message as unsent. When
> PRSCTP is disabled, you may only abandon messages/chunks that have not
> been transmitted (assigned a TSN).
I got it, thanks
>> Return error to application in sendmsg()? If so, how we check this if there is no asoc?
> How would you have data without an association?
>
> -vlad
>
>>
>>> -vlad
>>> --
>>> To unsubscribe from this list: send the line "unsubscribe netdev" in
>>> the body of a message to majordomo@vger.kernel.org
>>> More majordomo info at http://vger.kernel.org/majordomo-info.html
>>>
>
>
> --
> To unsubscribe from this list: send the line "unsubscribe netdev" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at http://vger.kernel.org/majordomo-info.html
>
^ permalink raw reply
* Re: [PATCH 1/2] Issue NETDEV_CHANGE notification when bridge changes state
From: Stephen Hemminger @ 2011-03-07 6:41 UTC (permalink / raw)
To: Adam Majer
Cc: David S. Miller, Alexey Kuznetsov, Pekka Savola (ipv6),
James Morris, Hideaki YOSHIFUJI, Patrick McHardy, bridge, netdev
In-Reply-To: <20110307002543.GA4242@mira.lan.galacticasoftware.com>
> On Sun, Mar 06, 2011 at 09:45:41AM -0800, Stephen Hemminger wrote:
> > Since this a generic problem, it needs a better solution.
> > Sending NETDEV_CHANGE impacts lots of other pieces, and even
> > user space has similar problems.
>
> It does seem a little broad notification type. I've checked over
> all the currently defined NETDEV notifiers, and it seems that
> NETDEV_NOTIFY_PEERS may be a better option to use when bridge
> has a potential topology change.
>
> Currently it is only used in ipv4/devinet.c: where it is used to issue
> a gratuitous ARP.
I was thinking of fixing bridge to not actually bring the link
up until in forwarding mode. Other applications (DHCP, etc)
see the link up and really don't like being in half duplex
during that period.
^ permalink raw reply
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