Merge branch 'net/ax88796' of git://git.pengutronix.de/git/mkl/linux-2.6
diff --git a/drivers/net/benet/be.h b/drivers/net/benet/be.h
index 3a800e2..ed709a5 100644
--- a/drivers/net/benet/be.h
+++ b/drivers/net/benet/be.h
@@ -225,6 +225,10 @@
u32 cache_line_barrier[15];
};
+struct be_drv_stats {
+ u8 be_on_die_temperature;
+};
+
struct be_vf_cfg {
unsigned char vf_mac_addr[ETH_ALEN];
u32 vf_if_handle;
@@ -234,6 +238,7 @@
};
#define BE_INVALID_PMAC_ID 0xffffffff
+
struct be_adapter {
struct pci_dev *pdev;
struct net_device *netdev;
@@ -269,6 +274,7 @@
u32 big_page_size; /* Compounded page size shared by rx wrbs */
u8 msix_vec_next_idx;
+ struct be_drv_stats drv_stats;
struct vlan_group *vlan_grp;
u16 vlans_added;
@@ -281,6 +287,7 @@
struct be_dma_mem stats_cmd;
/* Work queue used to perform periodic tasks like getting statistics */
struct delayed_work work;
+ u16 work_counter;
/* Ethtool knobs and info */
bool rx_csum; /* BE card must perform rx-checksumming */
@@ -298,7 +305,7 @@
u32 rx_fc; /* Rx flow control */
u32 tx_fc; /* Tx flow control */
bool ue_detected;
- bool stats_ioctl_sent;
+ bool stats_cmd_sent;
int link_speed;
u8 port_type;
u8 transceiver;
@@ -311,6 +318,8 @@
struct be_vf_cfg vf_cfg[BE_MAX_VF];
u8 is_virtfn;
u32 sli_family;
+ u8 hba_port_num;
+ u16 pvid;
};
#define be_physfn(adapter) (!adapter->is_virtfn)
diff --git a/drivers/net/benet/be_cmds.c b/drivers/net/benet/be_cmds.c
index 619ebc2..1822ecd 100644
--- a/drivers/net/benet/be_cmds.c
+++ b/drivers/net/benet/be_cmds.c
@@ -18,6 +18,9 @@
#include "be.h"
#include "be_cmds.h"
+/* Must be a power of 2 or else MODULO will BUG_ON */
+static int be_get_temp_freq = 32;
+
static void be_mcc_notify(struct be_adapter *adapter)
{
struct be_queue_info *mccq = &adapter->mcc_obj.q;
@@ -81,7 +84,7 @@
be_dws_le_to_cpu(&resp->hw_stats,
sizeof(resp->hw_stats));
netdev_stats_update(adapter);
- adapter->stats_ioctl_sent = false;
+ adapter->stats_cmd_sent = false;
}
} else if ((compl_status != MCC_STATUS_NOT_SUPPORTED) &&
(compl->tag0 != OPCODE_COMMON_NTWK_MAC_QUERY)) {
@@ -124,6 +127,16 @@
}
}
+/*Grp5 PVID evt*/
+static void be_async_grp5_pvid_state_process(struct be_adapter *adapter,
+ struct be_async_event_grp5_pvid_state *evt)
+{
+ if (evt->enabled)
+ adapter->pvid = evt->tag;
+ else
+ adapter->pvid = 0;
+}
+
static void be_async_grp5_evt_process(struct be_adapter *adapter,
u32 trailer, struct be_mcc_compl *evt)
{
@@ -141,6 +154,10 @@
be_async_grp5_qos_speed_process(adapter,
(struct be_async_event_grp5_qos_link_speed *)evt);
break;
+ case ASYNC_EVENT_PVID_STATE:
+ be_async_grp5_pvid_state_process(adapter,
+ (struct be_async_event_grp5_pvid_state *)evt);
+ break;
default:
dev_warn(&adapter->pdev->dev, "Unknown grp5 event!\n");
break;
@@ -1055,6 +1072,9 @@
struct be_sge *sge;
int status = 0;
+ if (MODULO(adapter->work_counter, be_get_temp_freq) == 0)
+ be_cmd_get_die_temperature(adapter);
+
spin_lock_bh(&adapter->mcc_lock);
wrb = wrb_from_mccq(adapter);
@@ -1075,7 +1095,7 @@
sge->len = cpu_to_le32(nonemb_cmd->size);
be_mcc_notify(adapter);
- adapter->stats_ioctl_sent = true;
+ adapter->stats_cmd_sent = true;
err:
spin_unlock_bh(&adapter->mcc_lock);
@@ -1122,6 +1142,44 @@
return status;
}
+/* Uses synchronous mcc */
+int be_cmd_get_die_temperature(struct be_adapter *adapter)
+{
+ struct be_mcc_wrb *wrb;
+ struct be_cmd_req_get_cntl_addnl_attribs *req;
+ int status;
+
+ spin_lock_bh(&adapter->mcc_lock);
+
+ wrb = wrb_from_mccq(adapter);
+ if (!wrb) {
+ status = -EBUSY;
+ goto err;
+ }
+ req = embedded_payload(wrb);
+
+ be_wrb_hdr_prepare(wrb, sizeof(*req), true, 0,
+ OPCODE_COMMON_GET_CNTL_ADDITIONAL_ATTRIBUTES);
+
+ be_cmd_hdr_prepare(&req->hdr, CMD_SUBSYSTEM_COMMON,
+ OPCODE_COMMON_GET_CNTL_ADDITIONAL_ATTRIBUTES, sizeof(*req));
+
+ status = be_mcc_notify_wait(adapter);
+ if (!status) {
+ struct be_cmd_resp_get_cntl_addnl_attribs *resp =
+ embedded_payload(wrb);
+ adapter->drv_stats.be_on_die_temperature =
+ resp->on_die_temperature;
+ }
+ /* If IOCTL fails once, do not bother issuing it again */
+ else
+ be_get_temp_freq = 0;
+
+err:
+ spin_unlock_bh(&adapter->mcc_lock);
+ return status;
+}
+
/* Uses Mbox */
int be_cmd_get_fw_ver(struct be_adapter *adapter, char *fw_ver)
{
@@ -1896,3 +1954,57 @@
spin_unlock_bh(&adapter->mcc_lock);
return status;
}
+
+int be_cmd_get_cntl_attributes(struct be_adapter *adapter)
+{
+ struct be_mcc_wrb *wrb;
+ struct be_cmd_req_cntl_attribs *req;
+ struct be_cmd_resp_cntl_attribs *resp;
+ struct be_sge *sge;
+ int status;
+ int payload_len = max(sizeof(*req), sizeof(*resp));
+ struct mgmt_controller_attrib *attribs;
+ struct be_dma_mem attribs_cmd;
+
+ memset(&attribs_cmd, 0, sizeof(struct be_dma_mem));
+ attribs_cmd.size = sizeof(struct be_cmd_resp_cntl_attribs);
+ attribs_cmd.va = pci_alloc_consistent(adapter->pdev, attribs_cmd.size,
+ &attribs_cmd.dma);
+ if (!attribs_cmd.va) {
+ dev_err(&adapter->pdev->dev,
+ "Memory allocation failure\n");
+ return -ENOMEM;
+ }
+
+ if (mutex_lock_interruptible(&adapter->mbox_lock))
+ return -1;
+
+ wrb = wrb_from_mbox(adapter);
+ if (!wrb) {
+ status = -EBUSY;
+ goto err;
+ }
+ req = attribs_cmd.va;
+ sge = nonembedded_sgl(wrb);
+
+ be_wrb_hdr_prepare(wrb, payload_len, false, 1,
+ OPCODE_COMMON_GET_CNTL_ATTRIBUTES);
+ be_cmd_hdr_prepare(&req->hdr, CMD_SUBSYSTEM_COMMON,
+ OPCODE_COMMON_GET_CNTL_ATTRIBUTES, payload_len);
+ sge->pa_hi = cpu_to_le32(upper_32_bits(attribs_cmd.dma));
+ sge->pa_lo = cpu_to_le32(attribs_cmd.dma & 0xFFFFFFFF);
+ sge->len = cpu_to_le32(attribs_cmd.size);
+
+ status = be_mbox_notify_wait(adapter);
+ if (!status) {
+ attribs = (struct mgmt_controller_attrib *)( attribs_cmd.va +
+ sizeof(struct be_cmd_resp_hdr));
+ adapter->hba_port_num = attribs->hba_attribs.phy_port;
+ }
+
+err:
+ mutex_unlock(&adapter->mbox_lock);
+ pci_free_consistent(adapter->pdev, attribs_cmd.size, attribs_cmd.va,
+ attribs_cmd.dma);
+ return status;
+}
diff --git a/drivers/net/benet/be_cmds.h b/drivers/net/benet/be_cmds.h
index 91c5d2b..93e5768 100644
--- a/drivers/net/benet/be_cmds.h
+++ b/drivers/net/benet/be_cmds.h
@@ -88,6 +88,7 @@
#define ASYNC_EVENT_CODE_GRP_5 0x5
#define ASYNC_EVENT_QOS_SPEED 0x1
#define ASYNC_EVENT_COS_PRIORITY 0x2
+#define ASYNC_EVENT_PVID_STATE 0x3
struct be_async_event_trailer {
u32 code;
};
@@ -134,6 +135,18 @@
struct be_async_event_trailer trailer;
} __packed;
+/* When the event code of an async trailer is GRP5 and event type is
+ * PVID state, the mcc_compl must be interpreted as follows
+ */
+struct be_async_event_grp5_pvid_state {
+ u8 enabled;
+ u8 rsvd0;
+ u16 tag;
+ u32 event_tag;
+ u32 rsvd1;
+ struct be_async_event_trailer trailer;
+} __packed;
+
struct be_mcc_mailbox {
struct be_mcc_wrb wrb;
struct be_mcc_compl compl;
@@ -156,6 +169,7 @@
#define OPCODE_COMMON_SET_QOS 28
#define OPCODE_COMMON_MCC_CREATE_EXT 90
#define OPCODE_COMMON_SEEPROM_READ 30
+#define OPCODE_COMMON_GET_CNTL_ATTRIBUTES 32
#define OPCODE_COMMON_NTWK_RX_FILTER 34
#define OPCODE_COMMON_GET_FW_VERSION 35
#define OPCODE_COMMON_SET_FLOW_CONTROL 36
@@ -176,6 +190,7 @@
#define OPCODE_COMMON_GET_BEACON_STATE 70
#define OPCODE_COMMON_READ_TRANSRECV_DATA 73
#define OPCODE_COMMON_GET_PHY_DETAILS 102
+#define OPCODE_COMMON_GET_CNTL_ADDITIONAL_ATTRIBUTES 121
#define OPCODE_ETH_RSS_CONFIG 1
#define OPCODE_ETH_ACPI_CONFIG 2
@@ -619,7 +634,10 @@
u32 rx_drops_invalid_ring; /* dword 145*/
u32 forwarded_packets; /* dword 146*/
u32 rx_drops_mtu; /* dword 147*/
- u32 rsvd0[15];
+ u32 rsvd0[7];
+ u32 port0_jabber_events;
+ u32 port1_jabber_events;
+ u32 rsvd1[6];
};
struct be_erx_stats {
@@ -630,11 +648,16 @@
u32 debug_pmem_pbuf_dealloc; /* dword 47*/
};
+struct be_pmem_stats {
+ u32 eth_red_drops;
+ u32 rsvd[4];
+};
+
struct be_hw_stats {
struct be_rxf_stats rxf;
u32 rsvd[48];
struct be_erx_stats erx;
- u32 rsvd1[6];
+ struct be_pmem_stats pmem;
};
struct be_cmd_req_get_stats {
@@ -647,6 +670,20 @@
struct be_hw_stats hw_stats;
};
+struct be_cmd_req_get_cntl_addnl_attribs {
+ struct be_cmd_req_hdr hdr;
+ u8 rsvd[8];
+};
+
+struct be_cmd_resp_get_cntl_addnl_attribs {
+ struct be_cmd_resp_hdr hdr;
+ u16 ipl_file_number;
+ u8 ipl_file_version;
+ u8 rsvd0;
+ u8 on_die_temperature; /* in degrees centigrade*/
+ u8 rsvd1[3];
+};
+
struct be_cmd_req_vlan_config {
struct be_cmd_req_hdr hdr;
u8 interface_id;
@@ -994,6 +1031,16 @@
u32 rsvd;
};
+/*********************** Controller Attributes ***********************/
+struct be_cmd_req_cntl_attribs {
+ struct be_cmd_req_hdr hdr;
+};
+
+struct be_cmd_resp_cntl_attribs {
+ struct be_cmd_resp_hdr hdr;
+ struct mgmt_controller_attrib attribs;
+};
+
extern int be_pci_fnum_get(struct be_adapter *adapter);
extern int be_cmd_POST(struct be_adapter *adapter);
extern int be_cmd_mac_addr_query(struct be_adapter *adapter, u8 *mac_addr,
@@ -1078,4 +1125,6 @@
struct be_dma_mem *cmd);
extern int be_cmd_set_qos(struct be_adapter *adapter, u32 bps, u32 domain);
extern void be_detect_dump_ue(struct be_adapter *adapter);
+extern int be_cmd_get_die_temperature(struct be_adapter *adapter);
+extern int be_cmd_get_cntl_attributes(struct be_adapter *adapter);
diff --git a/drivers/net/benet/be_ethtool.c b/drivers/net/benet/be_ethtool.c
index 07b4ab9..6e5e433 100644
--- a/drivers/net/benet/be_ethtool.c
+++ b/drivers/net/benet/be_ethtool.c
@@ -26,7 +26,8 @@
int offset;
};
-enum {NETSTAT, PORTSTAT, MISCSTAT, DRVSTAT_TX, DRVSTAT_RX, ERXSTAT};
+enum {NETSTAT, PORTSTAT, MISCSTAT, DRVSTAT_TX, DRVSTAT_RX, ERXSTAT,
+ PMEMSTAT, DRVSTAT};
#define FIELDINFO(_struct, field) FIELD_SIZEOF(_struct, field), \
offsetof(_struct, field)
#define NETSTAT_INFO(field) #field, NETSTAT,\
@@ -43,6 +44,11 @@
field)
#define ERXSTAT_INFO(field) #field, ERXSTAT,\
FIELDINFO(struct be_erx_stats, field)
+#define PMEMSTAT_INFO(field) #field, PMEMSTAT,\
+ FIELDINFO(struct be_pmem_stats, field)
+#define DRVSTAT_INFO(field) #field, DRVSTAT,\
+ FIELDINFO(struct be_drv_stats, \
+ field)
static const struct be_ethtool_stat et_stats[] = {
{NETSTAT_INFO(rx_packets)},
@@ -99,7 +105,11 @@
{MISCSTAT_INFO(rx_drops_too_many_frags)},
{MISCSTAT_INFO(rx_drops_invalid_ring)},
{MISCSTAT_INFO(forwarded_packets)},
- {MISCSTAT_INFO(rx_drops_mtu)}
+ {MISCSTAT_INFO(rx_drops_mtu)},
+ {MISCSTAT_INFO(port0_jabber_events)},
+ {MISCSTAT_INFO(port1_jabber_events)},
+ {PMEMSTAT_INFO(eth_red_drops)},
+ {DRVSTAT_INFO(be_on_die_temperature)}
};
#define ETHTOOL_STATS_NUM ARRAY_SIZE(et_stats)
@@ -121,7 +131,7 @@
"MAC Loopback test",
"PHY Loopback test",
"External Loopback test",
- "DDR DMA test"
+ "DDR DMA test",
"Link test"
};
@@ -276,6 +286,12 @@
case MISCSTAT:
p = &hw_stats->rxf;
break;
+ case PMEMSTAT:
+ p = &hw_stats->pmem;
+ break;
+ case DRVSTAT:
+ p = &adapter->drv_stats;
+ break;
}
p = (u8 *)p + et_stats[i].offset;
@@ -497,7 +513,7 @@
int status;
u32 cur;
- be_cmd_get_beacon_state(adapter, adapter->port_num, &cur);
+ be_cmd_get_beacon_state(adapter, adapter->hba_port_num, &cur);
if (cur == BEACON_STATE_ENABLED)
return 0;
@@ -505,12 +521,12 @@
if (data < 2)
data = 2;
- status = be_cmd_set_beacon_state(adapter, adapter->port_num, 0, 0,
+ status = be_cmd_set_beacon_state(adapter, adapter->hba_port_num, 0, 0,
BEACON_STATE_ENABLED);
set_current_state(TASK_INTERRUPTIBLE);
schedule_timeout(data*HZ);
- status = be_cmd_set_beacon_state(adapter, adapter->port_num, 0, 0,
+ status = be_cmd_set_beacon_state(adapter, adapter->hba_port_num, 0, 0,
BEACON_STATE_DISABLED);
return status;
@@ -589,12 +605,12 @@
static u64 be_loopback_test(struct be_adapter *adapter, u8 loopback_type,
u64 *status)
{
- be_cmd_set_loopback(adapter, adapter->port_num,
+ be_cmd_set_loopback(adapter, adapter->hba_port_num,
loopback_type, 1);
- *status = be_cmd_loopback_test(adapter, adapter->port_num,
+ *status = be_cmd_loopback_test(adapter, adapter->hba_port_num,
loopback_type, 1500,
2, 0xabc);
- be_cmd_set_loopback(adapter, adapter->port_num,
+ be_cmd_set_loopback(adapter, adapter->hba_port_num,
BE_NO_LOOPBACK, 1);
return *status;
}
@@ -633,7 +649,8 @@
&qos_link_speed) != 0) {
test->flags |= ETH_TEST_FL_FAILED;
data[4] = -1;
- } else if (mac_speed) {
+ } else if (!mac_speed) {
+ test->flags |= ETH_TEST_FL_FAILED;
data[4] = 1;
}
}
diff --git a/drivers/net/benet/be_hw.h b/drivers/net/benet/be_hw.h
index 4096d97..3f459f7 100644
--- a/drivers/net/benet/be_hw.h
+++ b/drivers/net/benet/be_hw.h
@@ -327,6 +327,53 @@
u32 dw[4];
};
+struct mgmt_hba_attribs {
+ u8 flashrom_version_string[32];
+ u8 manufacturer_name[32];
+ u32 supported_modes;
+ u32 rsvd0[3];
+ u8 ncsi_ver_string[12];
+ u32 default_extended_timeout;
+ u8 controller_model_number[32];
+ u8 controller_description[64];
+ u8 controller_serial_number[32];
+ u8 ip_version_string[32];
+ u8 firmware_version_string[32];
+ u8 bios_version_string[32];
+ u8 redboot_version_string[32];
+ u8 driver_version_string[32];
+ u8 fw_on_flash_version_string[32];
+ u32 functionalities_supported;
+ u16 max_cdblength;
+ u8 asic_revision;
+ u8 generational_guid[16];
+ u8 hba_port_count;
+ u16 default_link_down_timeout;
+ u8 iscsi_ver_min_max;
+ u8 multifunction_device;
+ u8 cache_valid;
+ u8 hba_status;
+ u8 max_domains_supported;
+ u8 phy_port;
+ u32 firmware_post_status;
+ u32 hba_mtu[8];
+ u32 rsvd1[4];
+};
+
+struct mgmt_controller_attrib {
+ struct mgmt_hba_attribs hba_attribs;
+ u16 pci_vendor_id;
+ u16 pci_device_id;
+ u16 pci_sub_vendor_id;
+ u16 pci_sub_system_id;
+ u8 pci_bus_number;
+ u8 pci_device_number;
+ u8 pci_function_number;
+ u8 interface_type;
+ u64 unique_identifier;
+ u32 rsvd0[5];
+};
+
struct controller_id {
u32 vendor;
u32 device;
diff --git a/drivers/net/benet/be_main.c b/drivers/net/benet/be_main.c
index aad7ea3..0bdccb1 100644
--- a/drivers/net/benet/be_main.c
+++ b/drivers/net/benet/be_main.c
@@ -1047,6 +1047,9 @@
if ((adapter->function_mode & 0x400) && !vtm)
vlanf = 0;
+ if ((adapter->pvid == vlanf) && !adapter->vlan_tag[vlanf])
+ vlanf = 0;
+
if (unlikely(vlanf)) {
if (!adapter->vlan_grp || adapter->vlans_added == 0) {
kfree_skb(skb);
@@ -1087,6 +1090,9 @@
if ((adapter->function_mode & 0x400) && !vtm)
vlanf = 0;
+ if ((adapter->pvid == vlanf) && !adapter->vlan_tag[vlanf])
+ vlanf = 0;
+
skb = napi_get_frags(&eq_obj->napi);
if (!skb) {
be_rx_compl_discard(adapter, rxo, rxcp);
@@ -1873,7 +1879,7 @@
goto reschedule;
}
- if (!adapter->stats_ioctl_sent)
+ if (!adapter->stats_cmd_sent)
be_cmd_get_stats(adapter, &adapter->stats_cmd);
be_tx_rate_update(adapter);
@@ -2862,6 +2868,10 @@
else
adapter->max_vlans = BE_NUM_VLANS_SUPPORTED;
+ status = be_cmd_get_cntl_attributes(adapter);
+ if (status)
+ return status;
+
return 0;
}
diff --git a/drivers/net/fec.c b/drivers/net/fec.c
index 74798be..634c0da 100644
--- a/drivers/net/fec.c
+++ b/drivers/net/fec.c
@@ -983,8 +983,6 @@
for (i = 0; i < PHY_MAX_ADDR; i++)
fep->mii_bus->irq[i] = PHY_POLL;
- platform_set_drvdata(ndev, fep->mii_bus);
-
if (mdiobus_register(fep->mii_bus))
goto err_out_free_mdio_irq;
diff --git a/drivers/net/sfc/efx.c b/drivers/net/sfc/efx.c
index d4e0425..35b7bc5 100644
--- a/drivers/net/sfc/efx.c
+++ b/drivers/net/sfc/efx.c
@@ -21,6 +21,7 @@
#include <linux/ethtool.h>
#include <linux/topology.h>
#include <linux/gfp.h>
+#include <linux/cpu_rmap.h>
#include "net_driver.h"
#include "efx.h"
#include "nic.h"
@@ -307,6 +308,8 @@
channel->irq_mod_score = 0;
}
+ efx_filter_rfs_expire(channel);
+
/* There is no race here; although napi_disable() will
* only wait for napi_complete(), this isn't a problem
* since efx_channel_processed() will have no effect if
@@ -1175,10 +1178,32 @@
return count;
}
+static int
+efx_init_rx_cpu_rmap(struct efx_nic *efx, struct msix_entry *xentries)
+{
+#ifdef CONFIG_RFS_ACCEL
+ int i, rc;
+
+ efx->net_dev->rx_cpu_rmap = alloc_irq_cpu_rmap(efx->n_rx_channels);
+ if (!efx->net_dev->rx_cpu_rmap)
+ return -ENOMEM;
+ for (i = 0; i < efx->n_rx_channels; i++) {
+ rc = irq_cpu_rmap_add(efx->net_dev->rx_cpu_rmap,
+ xentries[i].vector);
+ if (rc) {
+ free_irq_cpu_rmap(efx->net_dev->rx_cpu_rmap);
+ efx->net_dev->rx_cpu_rmap = NULL;
+ return rc;
+ }
+ }
+#endif
+ return 0;
+}
+
/* Probe the number and type of interrupts we are able to obtain, and
* the resulting numbers of channels and RX queues.
*/
-static void efx_probe_interrupts(struct efx_nic *efx)
+static int efx_probe_interrupts(struct efx_nic *efx)
{
int max_channels =
min_t(int, efx->type->phys_addr_channels, EFX_MAX_CHANNELS);
@@ -1220,6 +1245,11 @@
efx->n_tx_channels = efx->n_channels;
efx->n_rx_channels = efx->n_channels;
}
+ rc = efx_init_rx_cpu_rmap(efx, xentries);
+ if (rc) {
+ pci_disable_msix(efx->pci_dev);
+ return rc;
+ }
for (i = 0; i < n_channels; i++)
efx_get_channel(efx, i)->irq =
xentries[i].vector;
@@ -1253,6 +1283,8 @@
efx->n_tx_channels = 1;
efx->legacy_irq = efx->pci_dev->irq;
}
+
+ return 0;
}
static void efx_remove_interrupts(struct efx_nic *efx)
@@ -1289,7 +1321,9 @@
/* Determine the number of channels and queues by trying to hook
* in MSI-X interrupts. */
- efx_probe_interrupts(efx);
+ rc = efx_probe_interrupts(efx);
+ if (rc)
+ goto fail;
if (efx->n_channels > 1)
get_random_bytes(&efx->rx_hash_key, sizeof(efx->rx_hash_key));
@@ -1304,6 +1338,10 @@
efx_init_irq_moderation(efx, tx_irq_mod_usec, rx_irq_mod_usec, true);
return 0;
+
+fail:
+ efx->type->remove(efx);
+ return rc;
}
static void efx_remove_nic(struct efx_nic *efx)
@@ -1837,6 +1875,9 @@
.ndo_poll_controller = efx_netpoll,
#endif
.ndo_setup_tc = efx_setup_tc,
+#ifdef CONFIG_RFS_ACCEL
+ .ndo_rx_flow_steer = efx_filter_rfs,
+#endif
};
static void efx_update_name(struct efx_nic *efx)
@@ -2274,6 +2315,10 @@
*/
static void efx_pci_remove_main(struct efx_nic *efx)
{
+#ifdef CONFIG_RFS_ACCEL
+ free_irq_cpu_rmap(efx->net_dev->rx_cpu_rmap);
+ efx->net_dev->rx_cpu_rmap = NULL;
+#endif
efx_nic_fini_interrupt(efx);
efx_fini_channels(efx);
efx_fini_port(efx);
diff --git a/drivers/net/sfc/efx.h b/drivers/net/sfc/efx.h
index 0cb198a..cbce62b 100644
--- a/drivers/net/sfc/efx.h
+++ b/drivers/net/sfc/efx.h
@@ -76,6 +76,21 @@
struct efx_filter_spec *spec);
extern void efx_filter_clear_rx(struct efx_nic *efx,
enum efx_filter_priority priority);
+#ifdef CONFIG_RFS_ACCEL
+extern int efx_filter_rfs(struct net_device *net_dev, const struct sk_buff *skb,
+ u16 rxq_index, u32 flow_id);
+extern bool __efx_filter_rfs_expire(struct efx_nic *efx, unsigned quota);
+static inline void efx_filter_rfs_expire(struct efx_channel *channel)
+{
+ if (channel->rfs_filters_added >= 60 &&
+ __efx_filter_rfs_expire(channel->efx, 100))
+ channel->rfs_filters_added -= 60;
+}
+#define efx_filter_rfs_enabled() 1
+#else
+static inline void efx_filter_rfs_expire(struct efx_channel *channel) {}
+#define efx_filter_rfs_enabled() 0
+#endif
/* Channels */
extern void efx_process_channel_now(struct efx_channel *channel);
diff --git a/drivers/net/sfc/filter.c b/drivers/net/sfc/filter.c
index d4722c4..95a980f 100644
--- a/drivers/net/sfc/filter.c
+++ b/drivers/net/sfc/filter.c
@@ -8,6 +8,7 @@
*/
#include <linux/in.h>
+#include <net/ip.h>
#include "efx.h"
#include "filter.h"
#include "io.h"
@@ -27,6 +28,10 @@
*/
#define FILTER_CTL_SRCH_MAX 200
+/* Don't try very hard to find space for performance hints, as this is
+ * counter-productive. */
+#define FILTER_CTL_SRCH_HINT_MAX 5
+
enum efx_filter_table_id {
EFX_FILTER_TABLE_RX_IP = 0,
EFX_FILTER_TABLE_RX_MAC,
@@ -47,6 +52,10 @@
struct efx_filter_state {
spinlock_t lock;
struct efx_filter_table table[EFX_FILTER_TABLE_COUNT];
+#ifdef CONFIG_RFS_ACCEL
+ u32 *rps_flow_id;
+ unsigned rps_expire_index;
+#endif
};
/* The filter hash function is LFSR polynomial x^16 + x^3 + 1 of a 32-bit
@@ -325,15 +334,16 @@
struct efx_filter_spec *spec, u32 key,
bool for_insert, int *depth_required)
{
- unsigned hash, incr, filter_idx, depth;
+ unsigned hash, incr, filter_idx, depth, depth_max;
struct efx_filter_spec *cmp;
hash = efx_filter_hash(key);
incr = efx_filter_increment(key);
+ depth_max = (spec->priority <= EFX_FILTER_PRI_HINT ?
+ FILTER_CTL_SRCH_HINT_MAX : FILTER_CTL_SRCH_MAX);
for (depth = 1, filter_idx = hash & (table->size - 1);
- depth <= FILTER_CTL_SRCH_MAX &&
- test_bit(filter_idx, table->used_bitmap);
+ depth <= depth_max && test_bit(filter_idx, table->used_bitmap);
++depth) {
cmp = &table->spec[filter_idx];
if (efx_filter_equal(spec, cmp))
@@ -342,7 +352,7 @@
}
if (!for_insert)
return -ENOENT;
- if (depth > FILTER_CTL_SRCH_MAX)
+ if (depth > depth_max)
return -EBUSY;
found:
*depth_required = depth;
@@ -562,6 +572,13 @@
spin_lock_init(&state->lock);
if (efx_nic_rev(efx) >= EFX_REV_FALCON_B0) {
+#ifdef CONFIG_RFS_ACCEL
+ state->rps_flow_id = kcalloc(FR_BZ_RX_FILTER_TBL0_ROWS,
+ sizeof(*state->rps_flow_id),
+ GFP_KERNEL);
+ if (!state->rps_flow_id)
+ goto fail;
+#endif
table = &state->table[EFX_FILTER_TABLE_RX_IP];
table->id = EFX_FILTER_TABLE_RX_IP;
table->offset = FR_BZ_RX_FILTER_TBL0;
@@ -607,5 +624,97 @@
kfree(state->table[table_id].used_bitmap);
vfree(state->table[table_id].spec);
}
+#ifdef CONFIG_RFS_ACCEL
+ kfree(state->rps_flow_id);
+#endif
kfree(state);
}
+
+#ifdef CONFIG_RFS_ACCEL
+
+int efx_filter_rfs(struct net_device *net_dev, const struct sk_buff *skb,
+ u16 rxq_index, u32 flow_id)
+{
+ struct efx_nic *efx = netdev_priv(net_dev);
+ struct efx_channel *channel;
+ struct efx_filter_state *state = efx->filter_state;
+ struct efx_filter_spec spec;
+ const struct iphdr *ip;
+ const __be16 *ports;
+ int nhoff;
+ int rc;
+
+ nhoff = skb_network_offset(skb);
+
+ if (skb->protocol != htons(ETH_P_IP))
+ return -EPROTONOSUPPORT;
+
+ /* RFS must validate the IP header length before calling us */
+ EFX_BUG_ON_PARANOID(!pskb_may_pull(skb, nhoff + sizeof(*ip)));
+ ip = (const struct iphdr *)(skb->data + nhoff);
+ if (ip->frag_off & htons(IP_MF | IP_OFFSET))
+ return -EPROTONOSUPPORT;
+ EFX_BUG_ON_PARANOID(!pskb_may_pull(skb, nhoff + 4 * ip->ihl + 4));
+ ports = (const __be16 *)(skb->data + nhoff + 4 * ip->ihl);
+
+ efx_filter_init_rx(&spec, EFX_FILTER_PRI_HINT, 0, rxq_index);
+ rc = efx_filter_set_ipv4_full(&spec, ip->protocol,
+ ip->daddr, ports[1], ip->saddr, ports[0]);
+ if (rc)
+ return rc;
+
+ rc = efx_filter_insert_filter(efx, &spec, true);
+ if (rc < 0)
+ return rc;
+
+ /* Remember this so we can check whether to expire the filter later */
+ state->rps_flow_id[rc] = flow_id;
+ channel = efx_get_channel(efx, skb_get_rx_queue(skb));
+ ++channel->rfs_filters_added;
+
+ netif_info(efx, rx_status, efx->net_dev,
+ "steering %s %pI4:%u:%pI4:%u to queue %u [flow %u filter %d]\n",
+ (ip->protocol == IPPROTO_TCP) ? "TCP" : "UDP",
+ &ip->saddr, ntohs(ports[0]), &ip->daddr, ntohs(ports[1]),
+ rxq_index, flow_id, rc);
+
+ return rc;
+}
+
+bool __efx_filter_rfs_expire(struct efx_nic *efx, unsigned quota)
+{
+ struct efx_filter_state *state = efx->filter_state;
+ struct efx_filter_table *table = &state->table[EFX_FILTER_TABLE_RX_IP];
+ unsigned mask = table->size - 1;
+ unsigned index;
+ unsigned stop;
+
+ if (!spin_trylock_bh(&state->lock))
+ return false;
+
+ index = state->rps_expire_index;
+ stop = (index + quota) & mask;
+
+ while (index != stop) {
+ if (test_bit(index, table->used_bitmap) &&
+ table->spec[index].priority == EFX_FILTER_PRI_HINT &&
+ rps_may_expire_flow(efx->net_dev,
+ table->spec[index].dmaq_id,
+ state->rps_flow_id[index], index)) {
+ netif_info(efx, rx_status, efx->net_dev,
+ "expiring filter %d [flow %u]\n",
+ index, state->rps_flow_id[index]);
+ efx_filter_table_clear_entry(efx, table, index);
+ }
+ index = (index + 1) & mask;
+ }
+
+ state->rps_expire_index = stop;
+ if (table->used == 0)
+ efx_filter_table_reset_search_depth(table);
+
+ spin_unlock_bh(&state->lock);
+ return true;
+}
+
+#endif /* CONFIG_RFS_ACCEL */
diff --git a/drivers/net/sfc/net_driver.h b/drivers/net/sfc/net_driver.h
index 96e22ad..15b9068 100644
--- a/drivers/net/sfc/net_driver.h
+++ b/drivers/net/sfc/net_driver.h
@@ -362,6 +362,9 @@
unsigned int irq_count;
unsigned int irq_mod_score;
+#ifdef CONFIG_RFS_ACCEL
+ unsigned int rfs_filters_added;
+#endif
int rx_alloc_level;
int rx_alloc_push_pages;
diff --git a/drivers/net/sh_eth.c b/drivers/net/sh_eth.c
index 819c175..095e525 100644
--- a/drivers/net/sh_eth.c
+++ b/drivers/net/sh_eth.c
@@ -32,10 +32,17 @@
#include <linux/io.h>
#include <linux/pm_runtime.h>
#include <linux/slab.h>
+#include <linux/ethtool.h>
#include <asm/cacheflush.h>
#include "sh_eth.h"
+#define SH_ETH_DEF_MSG_ENABLE \
+ (NETIF_MSG_LINK | \
+ NETIF_MSG_TIMER | \
+ NETIF_MSG_RX_ERR| \
+ NETIF_MSG_TX_ERR)
+
/* There is CPU dependent code */
#if defined(CONFIG_CPU_SUBTYPE_SH7724)
#define SH_ETH_RESET_DEFAULT 1
@@ -817,6 +824,20 @@
return 0;
}
+static void sh_eth_rcv_snd_disable(u32 ioaddr)
+{
+ /* disable tx and rx */
+ writel(readl(ioaddr + ECMR) &
+ ~(ECMR_RE | ECMR_TE), ioaddr + ECMR);
+}
+
+static void sh_eth_rcv_snd_enable(u32 ioaddr)
+{
+ /* enable tx and rx */
+ writel(readl(ioaddr + ECMR) |
+ (ECMR_RE | ECMR_TE), ioaddr + ECMR);
+}
+
/* error control function */
static void sh_eth_error(struct net_device *ndev, int intr_status)
{
@@ -843,11 +864,9 @@
if (mdp->ether_link_active_low)
link_stat = ~link_stat;
}
- if (!(link_stat & PHY_ST_LINK)) {
- /* Link Down : disable tx and rx */
- writel(readl(ioaddr + ECMR) &
- ~(ECMR_RE | ECMR_TE), ioaddr + ECMR);
- } else {
+ if (!(link_stat & PHY_ST_LINK))
+ sh_eth_rcv_snd_disable(ioaddr);
+ else {
/* Link Up */
writel(readl(ioaddr + EESIPR) &
~DMAC_M_ECI, ioaddr + EESIPR);
@@ -857,8 +876,7 @@
writel(readl(ioaddr + EESIPR) |
DMAC_M_ECI, ioaddr + EESIPR);
/* enable tx and rx */
- writel(readl(ioaddr + ECMR) |
- (ECMR_RE | ECMR_TE), ioaddr + ECMR);
+ sh_eth_rcv_snd_enable(ioaddr);
}
}
}
@@ -867,6 +885,8 @@
/* Write buck end. unused write back interrupt */
if (intr_status & EESR_TABT) /* Transmit Abort int */
mdp->stats.tx_aborted_errors++;
+ if (netif_msg_tx_err(mdp))
+ dev_err(&ndev->dev, "Transmit Abort\n");
}
if (intr_status & EESR_RABT) {
@@ -874,14 +894,23 @@
if (intr_status & EESR_RFRMER) {
/* Receive Frame Overflow int */
mdp->stats.rx_frame_errors++;
- dev_err(&ndev->dev, "Receive Frame Overflow\n");
+ if (netif_msg_rx_err(mdp))
+ dev_err(&ndev->dev, "Receive Abort\n");
}
}
- if (!mdp->cd->no_ade) {
- if (intr_status & EESR_ADE && intr_status & EESR_TDE &&
- intr_status & EESR_TFE)
- mdp->stats.tx_fifo_errors++;
+ if (intr_status & EESR_TDE) {
+ /* Transmit Descriptor Empty int */
+ mdp->stats.tx_fifo_errors++;
+ if (netif_msg_tx_err(mdp))
+ dev_err(&ndev->dev, "Transmit Descriptor Empty\n");
+ }
+
+ if (intr_status & EESR_TFE) {
+ /* FIFO under flow */
+ mdp->stats.tx_fifo_errors++;
+ if (netif_msg_tx_err(mdp))
+ dev_err(&ndev->dev, "Transmit FIFO Under flow\n");
}
if (intr_status & EESR_RDE) {
@@ -890,12 +919,22 @@
if (readl(ioaddr + EDRRR) ^ EDRRR_R)
writel(EDRRR_R, ioaddr + EDRRR);
- dev_err(&ndev->dev, "Receive Descriptor Empty\n");
+ if (netif_msg_rx_err(mdp))
+ dev_err(&ndev->dev, "Receive Descriptor Empty\n");
}
+
if (intr_status & EESR_RFE) {
/* Receive FIFO Overflow int */
mdp->stats.rx_fifo_errors++;
- dev_err(&ndev->dev, "Receive FIFO Overflow\n");
+ if (netif_msg_rx_err(mdp))
+ dev_err(&ndev->dev, "Receive FIFO Overflow\n");
+ }
+
+ if (!mdp->cd->no_ade && (intr_status & EESR_ADE)) {
+ /* Address Error */
+ mdp->stats.tx_fifo_errors++;
+ if (netif_msg_tx_err(mdp))
+ dev_err(&ndev->dev, "Address Error\n");
}
mask = EESR_TWB | EESR_TABT | EESR_ADE | EESR_TDE | EESR_TFE;
@@ -1012,7 +1051,7 @@
mdp->duplex = -1;
}
- if (new_state)
+ if (new_state && netif_msg_link(mdp))
phy_print_status(phydev);
}
@@ -1063,6 +1102,132 @@
return 0;
}
+static int sh_eth_get_settings(struct net_device *ndev,
+ struct ethtool_cmd *ecmd)
+{
+ struct sh_eth_private *mdp = netdev_priv(ndev);
+ unsigned long flags;
+ int ret;
+
+ spin_lock_irqsave(&mdp->lock, flags);
+ ret = phy_ethtool_gset(mdp->phydev, ecmd);
+ spin_unlock_irqrestore(&mdp->lock, flags);
+
+ return ret;
+}
+
+static int sh_eth_set_settings(struct net_device *ndev,
+ struct ethtool_cmd *ecmd)
+{
+ struct sh_eth_private *mdp = netdev_priv(ndev);
+ unsigned long flags;
+ int ret;
+ u32 ioaddr = ndev->base_addr;
+
+ spin_lock_irqsave(&mdp->lock, flags);
+
+ /* disable tx and rx */
+ sh_eth_rcv_snd_disable(ioaddr);
+
+ ret = phy_ethtool_sset(mdp->phydev, ecmd);
+ if (ret)
+ goto error_exit;
+
+ if (ecmd->duplex == DUPLEX_FULL)
+ mdp->duplex = 1;
+ else
+ mdp->duplex = 0;
+
+ if (mdp->cd->set_duplex)
+ mdp->cd->set_duplex(ndev);
+
+error_exit:
+ mdelay(1);
+
+ /* enable tx and rx */
+ sh_eth_rcv_snd_enable(ioaddr);
+
+ spin_unlock_irqrestore(&mdp->lock, flags);
+
+ return ret;
+}
+
+static int sh_eth_nway_reset(struct net_device *ndev)
+{
+ struct sh_eth_private *mdp = netdev_priv(ndev);
+ unsigned long flags;
+ int ret;
+
+ spin_lock_irqsave(&mdp->lock, flags);
+ ret = phy_start_aneg(mdp->phydev);
+ spin_unlock_irqrestore(&mdp->lock, flags);
+
+ return ret;
+}
+
+static u32 sh_eth_get_msglevel(struct net_device *ndev)
+{
+ struct sh_eth_private *mdp = netdev_priv(ndev);
+ return mdp->msg_enable;
+}
+
+static void sh_eth_set_msglevel(struct net_device *ndev, u32 value)
+{
+ struct sh_eth_private *mdp = netdev_priv(ndev);
+ mdp->msg_enable = value;
+}
+
+static const char sh_eth_gstrings_stats[][ETH_GSTRING_LEN] = {
+ "rx_current", "tx_current",
+ "rx_dirty", "tx_dirty",
+};
+#define SH_ETH_STATS_LEN ARRAY_SIZE(sh_eth_gstrings_stats)
+
+static int sh_eth_get_sset_count(struct net_device *netdev, int sset)
+{
+ switch (sset) {
+ case ETH_SS_STATS:
+ return SH_ETH_STATS_LEN;
+ default:
+ return -EOPNOTSUPP;
+ }
+}
+
+static void sh_eth_get_ethtool_stats(struct net_device *ndev,
+ struct ethtool_stats *stats, u64 *data)
+{
+ struct sh_eth_private *mdp = netdev_priv(ndev);
+ int i = 0;
+
+ /* device-specific stats */
+ data[i++] = mdp->cur_rx;
+ data[i++] = mdp->cur_tx;
+ data[i++] = mdp->dirty_rx;
+ data[i++] = mdp->dirty_tx;
+}
+
+static void sh_eth_get_strings(struct net_device *ndev, u32 stringset, u8 *data)
+{
+ switch (stringset) {
+ case ETH_SS_STATS:
+ memcpy(data, *sh_eth_gstrings_stats,
+ sizeof(sh_eth_gstrings_stats));
+ break;
+ }
+}
+
+static struct ethtool_ops sh_eth_ethtool_ops = {
+ .get_settings = sh_eth_get_settings,
+ .set_settings = sh_eth_set_settings,
+ .nway_reset = sh_eth_nway_reset,
+ .get_msglevel = sh_eth_get_msglevel,
+ .set_msglevel = sh_eth_set_msglevel,
+ .get_link = ethtool_op_get_link,
+ .get_strings = sh_eth_get_strings,
+ .get_ethtool_stats = sh_eth_get_ethtool_stats,
+ .get_sset_count = sh_eth_get_sset_count,
+};
+
/* network device open function */
static int sh_eth_open(struct net_device *ndev)
{
@@ -1073,8 +1238,8 @@
ret = request_irq(ndev->irq, sh_eth_interrupt,
#if defined(CONFIG_CPU_SUBTYPE_SH7763) || \
- defined(CONFIG_CPU_SUBTYPE_SH7764) || \
- defined(CONFIG_CPU_SUBTYPE_SH7757)
+ defined(CONFIG_CPU_SUBTYPE_SH7764) || \
+ defined(CONFIG_CPU_SUBTYPE_SH7757)
IRQF_SHARED,
#else
0,
@@ -1123,8 +1288,8 @@
netif_stop_queue(ndev);
- /* worning message out. */
- printk(KERN_WARNING "%s: transmit timed out, status %8.8x,"
+ if (netif_msg_timer(mdp))
+ dev_err(&ndev->dev, "%s: transmit timed out, status %8.8x,"
" resetting...\n", ndev->name, (int)readl(ioaddr + EESR));
/* tx_errors count up */
@@ -1167,6 +1332,8 @@
spin_lock_irqsave(&mdp->lock, flags);
if ((mdp->cur_tx - mdp->dirty_tx) >= (TX_RING_SIZE - 4)) {
if (!sh_eth_txfree(ndev)) {
+ if (netif_msg_tx_queued(mdp))
+ dev_warn(&ndev->dev, "TxFD exhausted.\n");
netif_stop_queue(ndev);
spin_unlock_irqrestore(&mdp->lock, flags);
return NETDEV_TX_BUSY;
@@ -1497,8 +1664,11 @@
/* set function */
ndev->netdev_ops = &sh_eth_netdev_ops;
+ SET_ETHTOOL_OPS(ndev, &sh_eth_ethtool_ops);
ndev->watchdog_timeo = TX_TIMEOUT;
+ /* debug message level */
+ mdp->msg_enable = SH_ETH_DEF_MSG_ENABLE;
mdp->post_rx = POST_RX >> (devno << 1);
mdp->post_fw = POST_FW >> (devno << 1);
diff --git a/include/linux/net.h b/include/linux/net.h
index 16faa13..94de83c 100644
--- a/include/linux/net.h
+++ b/include/linux/net.h
@@ -118,6 +118,7 @@
};
struct socket_wq {
+ /* Note: wait MUST be first field of socket_wq */
wait_queue_head_t wait;
struct fasync_struct *fasync_list;
struct rcu_head rcu;
@@ -142,7 +143,7 @@
unsigned long flags;
- struct socket_wq *wq;
+ struct socket_wq __rcu *wq;
struct file *file;
struct sock *sk;
diff --git a/include/net/sock.h b/include/net/sock.h
index e3893a2..da0534d 100644
--- a/include/net/sock.h
+++ b/include/net/sock.h
@@ -281,7 +281,7 @@
int sk_rcvbuf;
struct sk_filter __rcu *sk_filter;
- struct socket_wq *sk_wq;
+ struct socket_wq __rcu *sk_wq;
#ifdef CONFIG_NET_DMA
struct sk_buff_head sk_async_wait_queue;
@@ -1266,7 +1266,8 @@
static inline wait_queue_head_t *sk_sleep(struct sock *sk)
{
- return &sk->sk_wq->wait;
+ BUILD_BUG_ON(offsetof(struct socket_wq, wait) != 0);
+ return &rcu_dereference_raw(sk->sk_wq)->wait;
}
/* Detach socket from process context.
* Announce socket dead, detach it from wait queue and inode.
@@ -1287,7 +1288,7 @@
static inline void sock_graft(struct sock *sk, struct socket *parent)
{
write_lock_bh(&sk->sk_callback_lock);
- rcu_assign_pointer(sk->sk_wq, parent->wq);
+ sk->sk_wq = parent->wq;
parent->sk = sk;
sk_set_socket(sk, parent);
security_sock_graft(sk, parent);
diff --git a/net/core/dev.c b/net/core/dev.c
index 9d8bfd9..578415c 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -2610,7 +2610,8 @@
int rc;
/* Should we steer this flow to a different hardware queue? */
- if (!skb_rx_queue_recorded(skb) || !dev->rx_cpu_rmap)
+ if (!skb_rx_queue_recorded(skb) || !dev->rx_cpu_rmap ||
+ !(dev->features & NETIF_F_NTUPLE))
goto out;
rxq_index = cpu_rmap_lookup_index(dev->rx_cpu_rmap, next_cpu);
if (rxq_index == skb_get_rx_queue(skb))
diff --git a/net/sctp/socket.c b/net/sctp/socket.c
index 8e02550..b53b2eb 100644
--- a/net/sctp/socket.c
+++ b/net/sctp/socket.c
@@ -6102,15 +6102,16 @@
wake_up_interruptible(&asoc->wait);
if (sctp_writeable(sk)) {
- if (sk_sleep(sk) && waitqueue_active(sk_sleep(sk)))
- wake_up_interruptible(sk_sleep(sk));
+ wait_queue_head_t *wq = sk_sleep(sk);
+
+ if (wq && waitqueue_active(wq))
+ wake_up_interruptible(wq);
/* Note that we try to include the Async I/O support
* here by modeling from the current TCP/UDP code.
* We have not tested with it yet.
*/
- if (sock->wq->fasync_list &&
- !(sk->sk_shutdown & SEND_SHUTDOWN))
+ if (!(sk->sk_shutdown & SEND_SHUTDOWN))
sock_wake_async(sock,
SOCK_WAKE_SPACE, POLL_OUT);
}
diff --git a/net/socket.c b/net/socket.c
index ac2219f..9fa1e3b 100644
--- a/net/socket.c
+++ b/net/socket.c
@@ -240,17 +240,19 @@
static struct inode *sock_alloc_inode(struct super_block *sb)
{
struct socket_alloc *ei;
+ struct socket_wq *wq;
ei = kmem_cache_alloc(sock_inode_cachep, GFP_KERNEL);
if (!ei)
return NULL;
- ei->socket.wq = kmalloc(sizeof(struct socket_wq), GFP_KERNEL);
- if (!ei->socket.wq) {
+ wq = kmalloc(sizeof(*wq), GFP_KERNEL);
+ if (!wq) {
kmem_cache_free(sock_inode_cachep, ei);
return NULL;
}
- init_waitqueue_head(&ei->socket.wq->wait);
- ei->socket.wq->fasync_list = NULL;
+ init_waitqueue_head(&wq->wait);
+ wq->fasync_list = NULL;
+ RCU_INIT_POINTER(ei->socket.wq, wq);
ei->socket.state = SS_UNCONNECTED;
ei->socket.flags = 0;
@@ -273,9 +275,11 @@
static void sock_destroy_inode(struct inode *inode)
{
struct socket_alloc *ei;
+ struct socket_wq *wq;
ei = container_of(inode, struct socket_alloc, vfs_inode);
- call_rcu(&ei->socket.wq->rcu, wq_free_rcu);
+ wq = rcu_dereference_protected(ei->socket.wq, 1);
+ call_rcu(&wq->rcu, wq_free_rcu);
kmem_cache_free(sock_inode_cachep, ei);
}
@@ -524,7 +528,7 @@
module_put(owner);
}
- if (sock->wq->fasync_list)
+ if (rcu_dereference_protected(sock->wq, 1)->fasync_list)
printk(KERN_ERR "sock_release: fasync list not empty!\n");
percpu_sub(sockets_in_use, 1);
@@ -1108,15 +1112,16 @@
{
struct socket *sock = filp->private_data;
struct sock *sk = sock->sk;
+ struct socket_wq *wq;
if (sk == NULL)
return -EINVAL;
lock_sock(sk);
+ wq = rcu_dereference_protected(sock->wq, sock_owned_by_user(sk));
+ fasync_helper(fd, filp, on, &wq->fasync_list);
- fasync_helper(fd, filp, on, &sock->wq->fasync_list);
-
- if (!sock->wq->fasync_list)
+ if (!wq->fasync_list)
sock_reset_flag(sk, SOCK_FASYNC);
else
sock_set_flag(sk, SOCK_FASYNC);
diff --git a/net/sunrpc/svcsock.c b/net/sunrpc/svcsock.c
index d802e94..b7d435c 100644
--- a/net/sunrpc/svcsock.c
+++ b/net/sunrpc/svcsock.c
@@ -420,6 +420,7 @@
static void svc_udp_data_ready(struct sock *sk, int count)
{
struct svc_sock *svsk = (struct svc_sock *)sk->sk_user_data;
+ wait_queue_head_t *wq = sk_sleep(sk);
if (svsk) {
dprintk("svc: socket %p(inet %p), count=%d, busy=%d\n",
@@ -428,8 +429,8 @@
set_bit(XPT_DATA, &svsk->sk_xprt.xpt_flags);
svc_xprt_enqueue(&svsk->sk_xprt);
}
- if (sk_sleep(sk) && waitqueue_active(sk_sleep(sk)))
- wake_up_interruptible(sk_sleep(sk));
+ if (wq && waitqueue_active(wq))
+ wake_up_interruptible(wq);
}
/*
@@ -438,6 +439,7 @@
static void svc_write_space(struct sock *sk)
{
struct svc_sock *svsk = (struct svc_sock *)(sk->sk_user_data);
+ wait_queue_head_t *wq = sk_sleep(sk);
if (svsk) {
dprintk("svc: socket %p(inet %p), write_space busy=%d\n",
@@ -445,10 +447,10 @@
svc_xprt_enqueue(&svsk->sk_xprt);
}
- if (sk_sleep(sk) && waitqueue_active(sk_sleep(sk))) {
+ if (wq && waitqueue_active(wq)) {
dprintk("RPC svc_write_space: someone sleeping on %p\n",
svsk);
- wake_up_interruptible(sk_sleep(sk));
+ wake_up_interruptible(wq);
}
}
@@ -739,6 +741,7 @@
static void svc_tcp_listen_data_ready(struct sock *sk, int count_unused)
{
struct svc_sock *svsk = (struct svc_sock *)sk->sk_user_data;
+ wait_queue_head_t *wq;
dprintk("svc: socket %p TCP (listen) state change %d\n",
sk, sk->sk_state);
@@ -761,8 +764,9 @@
printk("svc: socket %p: no user data\n", sk);
}
- if (sk_sleep(sk) && waitqueue_active(sk_sleep(sk)))
- wake_up_interruptible_all(sk_sleep(sk));
+ wq = sk_sleep(sk);
+ if (wq && waitqueue_active(wq))
+ wake_up_interruptible_all(wq);
}
/*
@@ -771,6 +775,7 @@
static void svc_tcp_state_change(struct sock *sk)
{
struct svc_sock *svsk = (struct svc_sock *)sk->sk_user_data;
+ wait_queue_head_t *wq = sk_sleep(sk);
dprintk("svc: socket %p TCP (connected) state change %d (svsk %p)\n",
sk, sk->sk_state, sk->sk_user_data);
@@ -781,13 +786,14 @@
set_bit(XPT_CLOSE, &svsk->sk_xprt.xpt_flags);
svc_xprt_enqueue(&svsk->sk_xprt);
}
- if (sk_sleep(sk) && waitqueue_active(sk_sleep(sk)))
- wake_up_interruptible_all(sk_sleep(sk));
+ if (wq && waitqueue_active(wq))
+ wake_up_interruptible_all(wq);
}
static void svc_tcp_data_ready(struct sock *sk, int count)
{
struct svc_sock *svsk = (struct svc_sock *)sk->sk_user_data;
+ wait_queue_head_t *wq = sk_sleep(sk);
dprintk("svc: socket %p TCP data ready (svsk %p)\n",
sk, sk->sk_user_data);
@@ -795,8 +801,8 @@
set_bit(XPT_DATA, &svsk->sk_xprt.xpt_flags);
svc_xprt_enqueue(&svsk->sk_xprt);
}
- if (sk_sleep(sk) && waitqueue_active(sk_sleep(sk)))
- wake_up_interruptible(sk_sleep(sk));
+ if (wq && waitqueue_active(wq))
+ wake_up_interruptible(wq);
}
/*
@@ -1531,6 +1537,7 @@
{
struct svc_sock *svsk = container_of(xprt, struct svc_sock, sk_xprt);
struct sock *sk = svsk->sk_sk;
+ wait_queue_head_t *wq;
dprintk("svc: svc_sock_detach(%p)\n", svsk);
@@ -1539,8 +1546,9 @@
sk->sk_data_ready = svsk->sk_odata;
sk->sk_write_space = svsk->sk_owspace;
- if (sk_sleep(sk) && waitqueue_active(sk_sleep(sk)))
- wake_up_interruptible(sk_sleep(sk));
+ wq = sk_sleep(sk);
+ if (wq && waitqueue_active(wq))
+ wake_up_interruptible(wq);
}
/*
diff --git a/net/unix/af_unix.c b/net/unix/af_unix.c
index d8d98d5..217fb7f 100644
--- a/net/unix/af_unix.c
+++ b/net/unix/af_unix.c
@@ -1171,7 +1171,7 @@
newsk->sk_type = sk->sk_type;
init_peercred(newsk);
newu = unix_sk(newsk);
- newsk->sk_wq = &newu->peer_wq;
+ RCU_INIT_POINTER(newsk->sk_wq, &newu->peer_wq);
otheru = unix_sk(other);
/* copy address information from listening to new sock*/
