* [PATCH net-next v2 8/9] bnxt_en: Check for valid forced speed during ethtool -s.
From: Michael Chan @ 2016-04-05 18:09 UTC (permalink / raw)
To: davem; +Cc: netdev
In-Reply-To: <1459879743-16960-1-git-send-email-michael.chan@broadcom.com>
Check that the forced speed is a valid speed supported by firmware.
If not supported, return -EINVAL.
Signed-off-by: Michael Chan <michael.chan@broadcom.com>
---
drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c | 48 ++++++++++++++++++-----
1 file changed, 38 insertions(+), 10 deletions(-)
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c
index 47e08a8..952b5ba 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c
@@ -739,28 +739,49 @@ static int bnxt_get_settings(struct net_device *dev, struct ethtool_cmd *cmd)
static u32 bnxt_get_fw_speed(struct net_device *dev, u16 ethtool_speed)
{
+ struct bnxt *bp = netdev_priv(dev);
+ struct bnxt_link_info *link_info = &bp->link_info;
+ u16 support_spds = link_info->support_speeds;
+ u32 fw_speed = 0;
+
switch (ethtool_speed) {
case SPEED_100:
- return PORT_PHY_CFG_REQ_AUTO_LINK_SPEED_100MB;
+ if (support_spds & BNXT_LINK_SPEED_MSK_100MB)
+ fw_speed = PORT_PHY_CFG_REQ_AUTO_LINK_SPEED_100MB;
+ break;
case SPEED_1000:
- return PORT_PHY_CFG_REQ_AUTO_LINK_SPEED_1GB;
+ if (support_spds & BNXT_LINK_SPEED_MSK_1GB)
+ fw_speed = PORT_PHY_CFG_REQ_AUTO_LINK_SPEED_1GB;
+ break;
case SPEED_2500:
- return PORT_PHY_CFG_REQ_AUTO_LINK_SPEED_2_5GB;
+ if (support_spds & BNXT_LINK_SPEED_MSK_2_5GB)
+ fw_speed = PORT_PHY_CFG_REQ_AUTO_LINK_SPEED_2_5GB;
+ break;
case SPEED_10000:
- return PORT_PHY_CFG_REQ_AUTO_LINK_SPEED_10GB;
+ if (support_spds & BNXT_LINK_SPEED_MSK_10GB)
+ fw_speed = PORT_PHY_CFG_REQ_AUTO_LINK_SPEED_10GB;
+ break;
case SPEED_20000:
- return PORT_PHY_CFG_REQ_AUTO_LINK_SPEED_20GB;
+ if (support_spds & BNXT_LINK_SPEED_MSK_20GB)
+ fw_speed = PORT_PHY_CFG_REQ_AUTO_LINK_SPEED_20GB;
+ break;
case SPEED_25000:
- return PORT_PHY_CFG_REQ_AUTO_LINK_SPEED_25GB;
+ if (support_spds & BNXT_LINK_SPEED_MSK_25GB)
+ fw_speed = PORT_PHY_CFG_REQ_AUTO_LINK_SPEED_25GB;
+ break;
case SPEED_40000:
- return PORT_PHY_CFG_REQ_AUTO_LINK_SPEED_40GB;
+ if (support_spds & BNXT_LINK_SPEED_MSK_40GB)
+ fw_speed = PORT_PHY_CFG_REQ_AUTO_LINK_SPEED_40GB;
+ break;
case SPEED_50000:
- return PORT_PHY_CFG_REQ_AUTO_LINK_SPEED_50GB;
+ if (support_spds & BNXT_LINK_SPEED_MSK_50GB)
+ fw_speed = PORT_PHY_CFG_REQ_AUTO_LINK_SPEED_50GB;
+ break;
default:
netdev_err(dev, "unsupported speed!\n");
break;
}
- return 0;
+ return fw_speed;
}
u16 bnxt_get_fw_auto_link_speeds(u32 advertising)
@@ -823,6 +844,8 @@ static int bnxt_set_settings(struct net_device *dev, struct ethtool_cmd *cmd)
*/
set_pause = true;
} else {
+ u16 fw_speed;
+
/* TODO: currently don't support half duplex */
if (cmd->duplex == DUPLEX_HALF) {
netdev_err(dev, "HALF DUPLEX is not supported!\n");
@@ -833,7 +856,12 @@ static int bnxt_set_settings(struct net_device *dev, struct ethtool_cmd *cmd)
if (cmd->duplex == DUPLEX_UNKNOWN)
cmd->duplex = DUPLEX_FULL;
speed = ethtool_cmd_speed(cmd);
- link_info->req_link_speed = bnxt_get_fw_speed(dev, speed);
+ fw_speed = bnxt_get_fw_speed(dev, speed);
+ if (!fw_speed) {
+ rc = -EINVAL;
+ goto set_setting_exit;
+ }
+ link_info->req_link_speed = fw_speed;
link_info->req_duplex = BNXT_LINK_DUPLEX_FULL;
link_info->autoneg = 0;
link_info->advertising = 0;
--
1.8.3.1
^ permalink raw reply related
* [PATCH net-next v2 7/9] bnxt_en: Add unsupported SFP+ module warnings.
From: Michael Chan @ 2016-04-05 18:09 UTC (permalink / raw)
To: davem; +Cc: netdev
In-Reply-To: <1459879743-16960-1-git-send-email-michael.chan@broadcom.com>
Add the PORT_CONN_NOT_ALLOWED async event handling logic. The driver
will print an appropriate warning to reflect the SFP+ module enforcement
policy done in the firmware.
Signed-off-by: Michael Chan <michael.chan@broadcom.com>
---
drivers/net/ethernet/broadcom/bnxt/bnxt.c | 55 +++++++++++++++++++++++++++++++
drivers/net/ethernet/broadcom/bnxt/bnxt.h | 3 ++
2 files changed, 58 insertions(+)
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
index dd0b32c..597e472 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
@@ -121,6 +121,7 @@ static const u16 bnxt_vf_req_snif[] = {
static const u16 bnxt_async_events_arr[] = {
HWRM_ASYNC_EVENT_CMPL_EVENT_ID_LINK_STATUS_CHANGE,
HWRM_ASYNC_EVENT_CMPL_EVENT_ID_PF_DRVR_UNLOAD,
+ HWRM_ASYNC_EVENT_CMPL_EVENT_ID_PORT_CONN_NOT_ALLOWED,
};
static bool bnxt_vf_pciid(enum board_idx idx)
@@ -1236,6 +1237,19 @@ next_rx_no_prod:
return rc;
}
+#define BNXT_GET_EVENT_PORT(data) \
+ ((data) & \
+ HWRM_ASYNC_EVENT_CMPL_PORT_CONN_NOT_ALLOWED_EVENT_DATA1_PORT_ID_MASK)
+
+#define BNXT_EVENT_POLICY_MASK \
+ HWRM_ASYNC_EVENT_CMPL_PORT_CONN_NOT_ALLOWED_EVENT_DATA1_ENFORCEMENT_POLICY_MASK
+
+#define BNXT_EVENT_POLICY_SFT \
+ HWRM_ASYNC_EVENT_CMPL_PORT_CONN_NOT_ALLOWED_EVENT_DATA1_ENFORCEMENT_POLICY_SFT
+
+#define BNXT_GET_EVENT_POLICY(data) \
+ (((data) & BNXT_EVENT_POLICY_MASK) >> BNXT_EVENT_POLICY_SFT)
+
static int bnxt_async_event_process(struct bnxt *bp,
struct hwrm_async_event_cmpl *cmpl)
{
@@ -1249,6 +1263,22 @@ static int bnxt_async_event_process(struct bnxt *bp,
case HWRM_ASYNC_EVENT_CMPL_EVENT_ID_PF_DRVR_UNLOAD:
set_bit(BNXT_HWRM_PF_UNLOAD_SP_EVENT, &bp->sp_event);
break;
+ case HWRM_ASYNC_EVENT_CMPL_EVENT_ID_PORT_CONN_NOT_ALLOWED: {
+ u32 data1 = le32_to_cpu(cmpl->event_data1);
+ u16 port_id = BNXT_GET_EVENT_PORT(data1);
+
+ if (BNXT_VF(bp))
+ break;
+
+ if (bp->pf.port_id != port_id)
+ break;
+
+ bp->link_info.last_port_module_event =
+ BNXT_GET_EVENT_POLICY(data1);
+
+ set_bit(BNXT_HWRM_PORT_MODULE_SP_EVENT, &bp->sp_event);
+ break;
+ }
default:
netdev_err(bp->dev, "unhandled ASYNC event (id 0x%x)\n",
event_id);
@@ -5447,6 +5477,28 @@ bnxt_restart_timer:
mod_timer(&bp->timer, jiffies + bp->current_interval);
}
+static void bnxt_port_module_event(struct bnxt *bp)
+{
+ struct bnxt_link_info *link_info = &bp->link_info;
+ struct hwrm_port_phy_qcfg_output *resp = &link_info->phy_qcfg_resp;
+
+ if (bnxt_update_link(bp, true))
+ return;
+
+ if (link_info->last_port_module_event != 0) {
+ netdev_warn(bp->dev, "Unqualified SFP+ module detected on port %d\n",
+ bp->pf.port_id);
+ if (bp->hwrm_spec_code >= 0x10201) {
+ netdev_warn(bp->dev, "Module part number %s\n",
+ resp->phy_vendor_partnumber);
+ }
+ }
+ if (link_info->last_port_module_event == 1)
+ netdev_warn(bp->dev, "TX is disabled\n");
+ if (link_info->last_port_module_event == 3)
+ netdev_warn(bp->dev, "Shutdown SFP+ module\n");
+}
+
static void bnxt_cfg_ntp_filters(struct bnxt *);
static void bnxt_sp_task(struct work_struct *work)
@@ -5494,6 +5546,9 @@ static void bnxt_sp_task(struct work_struct *work)
rtnl_unlock();
}
+ if (test_and_clear_bit(BNXT_HWRM_PORT_MODULE_SP_EVENT, &bp->sp_event))
+ bnxt_port_module_event(bp);
+
if (test_and_clear_bit(BNXT_PERIODIC_STATS_SP_EVENT, &bp->sp_event))
bnxt_hwrm_port_qstats(bp);
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.h b/drivers/net/ethernet/broadcom/bnxt/bnxt.h
index a981e2c..cc8e38a 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.h
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.h
@@ -825,6 +825,8 @@ struct bnxt_link_info {
u16 req_link_speed;
u32 advertising;
bool force_link_chng;
+
+ u8 last_port_module_event;
/* a copy of phy_qcfg output used to report link
* info to VF
*/
@@ -992,6 +994,7 @@ struct bnxt {
#define BNXT_RST_RING_SP_EVENT 7
#define BNXT_HWRM_PF_UNLOAD_SP_EVENT 8
#define BNXT_PERIODIC_STATS_SP_EVENT 9
+#define BNXT_HWRM_PORT_MODULE_SP_EVENT 10
struct bnxt_pf_info pf;
#ifdef CONFIG_BNXT_SRIOV
--
1.8.3.1
^ permalink raw reply related
* [PATCH net-next v2 6/9] bnxt_en: Set async event bits when registering with the firmware.
From: Michael Chan @ 2016-04-05 18:09 UTC (permalink / raw)
To: davem; +Cc: netdev
In-Reply-To: <1459879743-16960-1-git-send-email-michael.chan@broadcom.com>
Currently, the driver only sets bit 0 of the async_event_fwd fields.
To be compatible with the latest spec, we need to set the
appropriate event bits handled by the driver. We should be handling
link change and PF driver unload events, so these 2 bits should be
set.
Signed-off-by: Michael Chan <michael.chan@broadcom.com>
---
drivers/net/ethernet/broadcom/bnxt/bnxt.c | 18 ++++++++++++++----
1 file changed, 14 insertions(+), 4 deletions(-)
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
index 2c3c795..dd0b32c 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
@@ -118,6 +118,11 @@ static const u16 bnxt_vf_req_snif[] = {
HWRM_CFA_L2_FILTER_ALLOC,
};
+static const u16 bnxt_async_events_arr[] = {
+ HWRM_ASYNC_EVENT_CMPL_EVENT_ID_LINK_STATUS_CHANGE,
+ HWRM_ASYNC_EVENT_CMPL_EVENT_ID_PF_DRVR_UNLOAD,
+};
+
static bool bnxt_vf_pciid(enum board_idx idx)
{
return (idx == BCM57304_VF || idx == BCM57404_VF);
@@ -2751,6 +2756,8 @@ static int bnxt_hwrm_func_drv_rgtr(struct bnxt *bp)
{
struct hwrm_func_drv_rgtr_input req = {0};
int i;
+ DECLARE_BITMAP(async_events_bmap, 256);
+ u32 *events = (u32 *)async_events_bmap;
bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_FUNC_DRV_RGTR, -1, -1);
@@ -2759,10 +2766,13 @@ static int bnxt_hwrm_func_drv_rgtr(struct bnxt *bp)
FUNC_DRV_RGTR_REQ_ENABLES_VER |
FUNC_DRV_RGTR_REQ_ENABLES_ASYNC_EVENT_FWD);
- /* TODO: current async event fwd bits are not defined and the firmware
- * only checks if it is non-zero to enable async event forwarding
- */
- req.async_event_fwd[0] |= cpu_to_le32(1);
+ memset(async_events_bmap, 0, sizeof(async_events_bmap));
+ for (i = 0; i < ARRAY_SIZE(bnxt_async_events_arr); i++)
+ __set_bit(bnxt_async_events_arr[i], async_events_bmap);
+
+ for (i = 0; i < 8; i++)
+ req.async_event_fwd[i] |= cpu_to_le32(events[i]);
+
req.os_type = cpu_to_le16(FUNC_DRV_RGTR_REQ_OS_TYPE_LINUX);
req.ver_maj = DRV_VER_MAJ;
req.ver_min = DRV_VER_MIN;
--
1.8.3.1
^ permalink raw reply related
* [PATCH net-next v2 3/9] bnxt_en: Add basic EEE support.
From: Michael Chan @ 2016-04-05 18:08 UTC (permalink / raw)
To: davem; +Cc: netdev
In-Reply-To: <1459879743-16960-1-git-send-email-michael.chan@broadcom.com>
Get EEE capability and the initial EEE settings from firmware.
Add "EEE is active | not active" to link up dmesg.
Signed-off-by: Michael Chan <michael.chan@broadcom.com>
---
drivers/net/ethernet/broadcom/bnxt/bnxt.c | 82 ++++++++++++++++++++++-
drivers/net/ethernet/broadcom/bnxt/bnxt.h | 4 ++
drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c | 2 +-
drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.h | 1 +
4 files changed, 87 insertions(+), 2 deletions(-)
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
index 2b5a541..7442e20 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
@@ -4488,12 +4488,49 @@ static void bnxt_report_link(struct bnxt *bp)
speed = bnxt_fw_to_ethtool_speed(bp->link_info.link_speed);
netdev_info(bp->dev, "NIC Link is Up, %d Mbps %s duplex, Flow control: %s\n",
speed, duplex, flow_ctrl);
+ if (bp->flags & BNXT_FLAG_EEE_CAP)
+ netdev_info(bp->dev, "EEE is %s\n",
+ bp->eee.eee_active ? "active" :
+ "not active");
} else {
netif_carrier_off(bp->dev);
netdev_err(bp->dev, "NIC Link is Down\n");
}
}
+static int bnxt_hwrm_phy_qcaps(struct bnxt *bp)
+{
+ int rc = 0;
+ struct hwrm_port_phy_qcaps_input req = {0};
+ struct hwrm_port_phy_qcaps_output *resp = bp->hwrm_cmd_resp_addr;
+
+ if (bp->hwrm_spec_code < 0x10201)
+ return 0;
+
+ bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_PORT_PHY_QCAPS, -1, -1);
+
+ mutex_lock(&bp->hwrm_cmd_lock);
+ rc = _hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT);
+ if (rc)
+ goto hwrm_phy_qcaps_exit;
+
+ if (resp->eee_supported & PORT_PHY_QCAPS_RESP_EEE_SUPPORTED) {
+ struct ethtool_eee *eee = &bp->eee;
+ u16 fw_speeds = le16_to_cpu(resp->supported_speeds_eee_mode);
+
+ bp->flags |= BNXT_FLAG_EEE_CAP;
+ eee->supported = _bnxt_fw_to_ethtool_adv_spds(fw_speeds, 0);
+ bp->lpi_tmr_lo = le32_to_cpu(resp->tx_lpi_timer_low) &
+ PORT_PHY_QCAPS_RESP_TX_LPI_TIMER_LOW_MASK;
+ bp->lpi_tmr_hi = le32_to_cpu(resp->valid_tx_lpi_timer_high) &
+ PORT_PHY_QCAPS_RESP_TX_LPI_TIMER_HIGH_MASK;
+ }
+
+hwrm_phy_qcaps_exit:
+ mutex_unlock(&bp->hwrm_cmd_lock);
+ return rc;
+}
+
static int bnxt_update_link(struct bnxt *bp, bool chng_link_state)
{
int rc = 0;
@@ -4535,8 +4572,44 @@ static int bnxt_update_link(struct bnxt *bp, bool chng_link_state)
link_info->phy_ver[2] = resp->phy_bld;
link_info->media_type = resp->media_type;
link_info->transceiver = resp->xcvr_pkg_type;
- link_info->phy_addr = resp->eee_config_phy_addr;
+ link_info->phy_addr = resp->eee_config_phy_addr &
+ PORT_PHY_QCFG_RESP_PHY_ADDR_MASK;
+
+ if (bp->flags & BNXT_FLAG_EEE_CAP) {
+ struct ethtool_eee *eee = &bp->eee;
+ u16 fw_speeds;
+
+ eee->eee_active = 0;
+ if (resp->eee_config_phy_addr &
+ PORT_PHY_QCFG_RESP_EEE_CONFIG_EEE_ACTIVE) {
+ eee->eee_active = 1;
+ fw_speeds = le16_to_cpu(
+ resp->link_partner_adv_eee_link_speed_mask);
+ eee->lp_advertised =
+ _bnxt_fw_to_ethtool_adv_spds(fw_speeds, 0);
+ }
+
+ /* Pull initial EEE config */
+ if (!chng_link_state) {
+ if (resp->eee_config_phy_addr &
+ PORT_PHY_QCFG_RESP_EEE_CONFIG_EEE_ENABLED)
+ eee->eee_enabled = 1;
+ fw_speeds = le16_to_cpu(resp->adv_eee_link_speed_mask);
+ eee->advertised =
+ _bnxt_fw_to_ethtool_adv_spds(fw_speeds, 0);
+
+ if (resp->eee_config_phy_addr &
+ PORT_PHY_QCFG_RESP_EEE_CONFIG_EEE_TX_LPI) {
+ __le32 tmr;
+
+ eee->tx_lpi_enabled = 1;
+ tmr = resp->xcvr_identifier_type_tx_lpi_timer;
+ eee->tx_lpi_timer = le32_to_cpu(tmr) &
+ PORT_PHY_QCFG_RESP_TX_LPI_TIMER_MASK;
+ }
+ }
+ }
/* TODO: need to add more logic to report VF link */
if (chng_link_state) {
if (link_info->phy_link_status == BNXT_LINK_LINK)
@@ -5825,6 +5898,13 @@ static int bnxt_probe_phy(struct bnxt *bp)
int rc = 0;
struct bnxt_link_info *link_info = &bp->link_info;
+ rc = bnxt_hwrm_phy_qcaps(bp);
+ if (rc) {
+ netdev_err(bp->dev, "Probe phy can't get phy capabilities (rc: %x)\n",
+ rc);
+ return rc;
+ }
+
rc = bnxt_update_link(bp, false);
if (rc) {
netdev_err(bp->dev, "Probe phy can't update link (rc: %x)\n",
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.h b/drivers/net/ethernet/broadcom/bnxt/bnxt.h
index e98c37a..5e83405 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.h
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.h
@@ -874,6 +874,7 @@ struct bnxt {
#define BNXT_FLAG_RFS 0x100
#define BNXT_FLAG_SHARED_RINGS 0x200
#define BNXT_FLAG_PORT_STATS 0x400
+ #define BNXT_FLAG_EEE_CAP 0x1000
#define BNXT_FLAG_ALL_CONFIG_FEATS (BNXT_FLAG_TPA | \
BNXT_FLAG_RFS | \
@@ -1011,6 +1012,9 @@ struct bnxt {
int ntp_fltr_count;
struct bnxt_link_info link_info;
+ struct ethtool_eee eee;
+ u32 lpi_tmr_lo;
+ u32 lpi_tmr_hi;
};
#ifdef CONFIG_NET_RX_BUSY_POLL
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c
index 99b1740..bdc6220 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c
@@ -597,7 +597,7 @@ static void bnxt_get_drvinfo(struct net_device *dev,
kfree(pkglog);
}
-static u32 _bnxt_fw_to_ethtool_adv_spds(u16 fw_speeds, u8 fw_pause)
+u32 _bnxt_fw_to_ethtool_adv_spds(u16 fw_speeds, u8 fw_pause)
{
u32 speed_mask = 0;
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.h b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.h
index b2d8bd3..e061f8f 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.h
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.h
@@ -12,6 +12,7 @@
extern const struct ethtool_ops bnxt_ethtool_ops;
+u32 _bnxt_fw_to_ethtool_adv_spds(u16, u8);
u32 bnxt_fw_to_ethtool_speed(u16);
#endif
--
1.8.3.1
^ permalink raw reply related
* [PATCH net-next v2 1/9] bnxt_en: Update to Firmware 1.2.2 spec.
From: Michael Chan @ 2016-04-05 18:08 UTC (permalink / raw)
To: davem; +Cc: netdev
In-Reply-To: <1459879743-16960-1-git-send-email-michael.chan@broadcom.com>
Use new field names in API structs and stop using deprecated fields
auto_link_speed and auto_duplex in phy_cfg/phy_qcfg structs.
Update copyright year to 2016.
Signed-off-by: Michael Chan <michael.chan@broadcom.com>
---
drivers/net/ethernet/broadcom/bnxt/bnxt.c | 20 +-
drivers/net/ethernet/broadcom/bnxt/bnxt.h | 8 +-
drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c | 4 +-
drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.h | 2 +-
drivers/net/ethernet/broadcom/bnxt/bnxt_fw_hdr.h | 2 +-
drivers/net/ethernet/broadcom/bnxt/bnxt_hsi.h | 433 +++++++++++++++++++--
drivers/net/ethernet/broadcom/bnxt/bnxt_nvm_defs.h | 2 +-
drivers/net/ethernet/broadcom/bnxt/bnxt_sriov.c | 14 +-
drivers/net/ethernet/broadcom/bnxt/bnxt_sriov.h | 2 +-
9 files changed, 431 insertions(+), 56 deletions(-)
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
index 12a009d..bfe98cb 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
@@ -1,6 +1,6 @@
/* Broadcom NetXtreme-C/E network driver.
*
- * Copyright (c) 2014-2015 Broadcom Corporation
+ * Copyright (c) 2014-2016 Broadcom Corporation
*
* This program is free software; you can redistribute it and/or modify
* it under the terms of the GNU General Public License as published by
@@ -2763,7 +2763,7 @@ static int bnxt_hwrm_func_drv_rgtr(struct bnxt *bp)
* only checks if it is non-zero to enable async event forwarding
*/
req.async_event_fwd[0] |= cpu_to_le32(1);
- req.os_type = cpu_to_le16(1);
+ req.os_type = cpu_to_le16(FUNC_DRV_RGTR_REQ_OS_TYPE_LINUX);
req.ver_maj = DRV_VER_MAJ;
req.ver_min = DRV_VER_MIN;
req.ver_upd = DRV_VER_UPD;
@@ -3726,7 +3726,7 @@ int bnxt_hwrm_func_qcaps(struct bnxt *bp)
pf->fw_fid = le16_to_cpu(resp->fid);
pf->port_id = le16_to_cpu(resp->port_id);
- memcpy(pf->mac_addr, resp->perm_mac_address, ETH_ALEN);
+ memcpy(pf->mac_addr, resp->mac_address, ETH_ALEN);
memcpy(bp->dev->dev_addr, pf->mac_addr, ETH_ALEN);
pf->max_rsscos_ctxs = le16_to_cpu(resp->max_rsscos_ctx);
pf->max_cp_rings = le16_to_cpu(resp->max_cmpl_rings);
@@ -3751,7 +3751,7 @@ int bnxt_hwrm_func_qcaps(struct bnxt *bp)
struct bnxt_vf_info *vf = &bp->vf;
vf->fw_fid = le16_to_cpu(resp->fid);
- memcpy(vf->mac_addr, resp->perm_mac_address, ETH_ALEN);
+ memcpy(vf->mac_addr, resp->mac_address, ETH_ALEN);
if (is_valid_ether_addr(vf->mac_addr))
/* overwrite netdev dev_adr with admin VF MAC */
memcpy(bp->dev->dev_addr, vf->mac_addr, ETH_ALEN);
@@ -3842,6 +3842,8 @@ static int bnxt_hwrm_ver_get(struct bnxt *bp)
memcpy(&bp->ver_resp, resp, sizeof(struct hwrm_ver_get_output));
+ bp->hwrm_spec_code = resp->hwrm_intf_maj << 16 |
+ resp->hwrm_intf_min << 8 | resp->hwrm_intf_upd;
if (resp->hwrm_intf_maj < 1) {
netdev_warn(bp->dev, "HWRM interface %d.%d.%d is older than 1.0.0.\n",
resp->hwrm_intf_maj, resp->hwrm_intf_min,
@@ -4523,7 +4525,6 @@ static int bnxt_update_link(struct bnxt *bp, bool chng_link_state)
else
link_info->link_speed = 0;
link_info->force_link_speed = le16_to_cpu(resp->force_link_speed);
- link_info->auto_link_speed = le16_to_cpu(resp->auto_link_speed);
link_info->support_speeds = le16_to_cpu(resp->support_speeds);
link_info->auto_link_speeds = le16_to_cpu(resp->auto_link_speed_mask);
link_info->lp_auto_link_speeds =
@@ -4533,8 +4534,8 @@ static int bnxt_update_link(struct bnxt *bp, bool chng_link_state)
link_info->phy_ver[1] = resp->phy_min;
link_info->phy_ver[2] = resp->phy_bld;
link_info->media_type = resp->media_type;
- link_info->transceiver = resp->transceiver_type;
- link_info->phy_addr = resp->phy_addr;
+ link_info->transceiver = resp->xcvr_pkg_type;
+ link_info->phy_addr = resp->eee_config_phy_addr;
/* TODO: need to add more logic to report VF link */
if (chng_link_state) {
@@ -4581,7 +4582,7 @@ static void bnxt_hwrm_set_link_common(struct bnxt *bp,
if (autoneg & BNXT_AUTONEG_SPEED) {
req->auto_mode |=
- PORT_PHY_CFG_REQ_AUTO_MODE_MASK;
+ PORT_PHY_CFG_REQ_AUTO_MODE_SPEED_MASK;
req->enables |= cpu_to_le32(
PORT_PHY_CFG_REQ_ENABLES_AUTO_LINK_SPEED_MASK);
@@ -4595,9 +4596,6 @@ static void bnxt_hwrm_set_link_common(struct bnxt *bp,
req->flags |= cpu_to_le32(PORT_PHY_CFG_REQ_FLAGS_FORCE);
}
- /* currently don't support half duplex */
- req->auto_duplex = PORT_PHY_CFG_REQ_AUTO_DUPLEX_FULL;
- req->enables |= cpu_to_le32(PORT_PHY_CFG_REQ_ENABLES_AUTO_DUPLEX);
/* tell chimp that the setting takes effect immediately */
req->flags |= cpu_to_le32(PORT_PHY_CFG_REQ_FLAGS_RESET_PHY);
}
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.h b/drivers/net/ethernet/broadcom/bnxt/bnxt.h
index 709b95b..e98c37a 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.h
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.h
@@ -1,6 +1,6 @@
/* Broadcom NetXtreme-C/E network driver.
*
- * Copyright (c) 2014-2015 Broadcom Corporation
+ * Copyright (c) 2014-2016 Broadcom Corporation
*
* This program is free software; you can redistribute it and/or modify
* it under the terms of the GNU General Public License as published by
@@ -11,7 +11,7 @@
#define BNXT_H
#define DRV_MODULE_NAME "bnxt_en"
-#define DRV_MODULE_VERSION "1.0.0"
+#define DRV_MODULE_VERSION "1.2.0"
#define DRV_VER_MAJ 1
#define DRV_VER_MIN 0
@@ -788,7 +788,7 @@ struct bnxt_link_info {
#define BNXT_LINK_AUTO_ALLSPDS PORT_PHY_QCFG_RESP_AUTO_MODE_ALL_SPEEDS
#define BNXT_LINK_AUTO_ONESPD PORT_PHY_QCFG_RESP_AUTO_MODE_ONE_SPEED
#define BNXT_LINK_AUTO_ONEORBELOW PORT_PHY_QCFG_RESP_AUTO_MODE_ONE_OR_BELOW
-#define BNXT_LINK_AUTO_MSK PORT_PHY_QCFG_RESP_AUTO_MODE_MASK
+#define BNXT_LINK_AUTO_MSK PORT_PHY_QCFG_RESP_AUTO_MODE_SPEED_MASK
#define PHY_VER_LEN 3
u8 phy_ver[PHY_VER_LEN];
u16 link_speed;
@@ -813,7 +813,6 @@ struct bnxt_link_info {
#define BNXT_LINK_SPEED_MSK_40GB PORT_PHY_QCFG_RESP_SUPPORT_SPEEDS_40GB
#define BNXT_LINK_SPEED_MSK_50GB PORT_PHY_QCFG_RESP_SUPPORT_SPEEDS_50GB
u16 lp_auto_link_speeds;
- u16 auto_link_speed;
u16 force_link_speed;
u32 preemphasis;
@@ -940,6 +939,7 @@ struct bnxt {
u32 msg_enable;
+ u32 hwrm_spec_code;
u16 hwrm_cmd_seq;
u32 hwrm_intr_seq_id;
void *hwrm_cmd_resp_addr;
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c
index 2e472f6..f103f9b 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c
@@ -1,6 +1,6 @@
/* Broadcom NetXtreme-C/E network driver.
*
- * Copyright (c) 2014-2015 Broadcom Corporation
+ * Copyright (c) 2014-2016 Broadcom Corporation
*
* This program is free software; you can redistribute it and/or modify
* it under the terms of the GNU General Public License as published by
@@ -728,7 +728,7 @@ static int bnxt_get_settings(struct net_device *dev, struct ethtool_cmd *cmd)
ethtool_speed = bnxt_fw_to_ethtool_speed(link_info->link_speed);
ethtool_cmd_speed_set(cmd, ethtool_speed);
if (link_info->transceiver ==
- PORT_PHY_QCFG_RESP_TRANSCEIVER_TYPE_XCVR_INTERNAL)
+ PORT_PHY_QCFG_RESP_XCVR_PKG_TYPE_XCVR_INTERNAL)
cmd->transceiver = XCVR_INTERNAL;
else
cmd->transceiver = XCVR_EXTERNAL;
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.h b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.h
index 98fa81e..b2d8bd3 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.h
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.h
@@ -1,6 +1,6 @@
/* Broadcom NetXtreme-C/E network driver.
*
- * Copyright (c) 2014-2015 Broadcom Corporation
+ * Copyright (c) 2014-2016 Broadcom Corporation
*
* This program is free software; you can redistribute it and/or modify
* it under the terms of the GNU General Public License as published by
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_fw_hdr.h b/drivers/net/ethernet/broadcom/bnxt/bnxt_fw_hdr.h
index e0aac65..461675c 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt_fw_hdr.h
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_fw_hdr.h
@@ -1,6 +1,6 @@
/* Broadcom NetXtreme-C/E network driver.
*
- * Copyright (c) 2014-2015 Broadcom Corporation
+ * Copyright (c) 2014-2016 Broadcom Corporation
*
* This program is free software; you can redistribute it and/or modify
* it under the terms of the GNU General Public License as published by
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_hsi.h b/drivers/net/ethernet/broadcom/bnxt/bnxt_hsi.h
index 4badbed..80f9556 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt_hsi.h
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_hsi.h
@@ -1,6 +1,6 @@
/* Broadcom NetXtreme-C/E network driver.
*
- * Copyright (c) 2014-2015 Broadcom Corporation
+ * Copyright (c) 2014-2016 Broadcom Corporation
*
* This program is free software; you can redistribute it and/or modify
* it under the terms of the GNU General Public License as published by
@@ -104,6 +104,7 @@ struct hwrm_async_event_cmpl {
#define HWRM_ASYNC_EVENT_CMPL_EVENT_ID_DCB_CONFIG_CHANGE (0x3UL << 0)
#define HWRM_ASYNC_EVENT_CMPL_EVENT_ID_PORT_CONN_NOT_ALLOWED (0x4UL << 0)
#define HWRM_ASYNC_EVENT_CMPL_EVENT_ID_LINK_SPEED_CFG_NOT_ALLOWED (0x5UL << 0)
+ #define HWRM_ASYNC_EVENT_CMPL_EVENT_ID_LINK_SPEED_CFG_CHANGE (0x6UL << 0)
#define HWRM_ASYNC_EVENT_CMPL_EVENT_ID_FUNC_DRVR_UNLOAD (0x10UL << 0)
#define HWRM_ASYNC_EVENT_CMPL_EVENT_ID_FUNC_DRVR_LOAD (0x11UL << 0)
#define HWRM_ASYNC_EVENT_CMPL_EVENT_ID_PF_DRVR_UNLOAD (0x20UL << 0)
@@ -111,6 +112,7 @@ struct hwrm_async_event_cmpl {
#define HWRM_ASYNC_EVENT_CMPL_EVENT_ID_VF_FLR (0x30UL << 0)
#define HWRM_ASYNC_EVENT_CMPL_EVENT_ID_VF_MAC_ADDR_CHANGE (0x31UL << 0)
#define HWRM_ASYNC_EVENT_CMPL_EVENT_ID_PF_VF_COMM_STATUS_CHANGE (0x32UL << 0)
+ #define HWRM_ASYNC_EVENT_CMPL_EVENT_ID_VF_CFG_CHANGE (0x33UL << 0)
#define HWRM_ASYNC_EVENT_CMPL_EVENT_ID_HWRM_ERROR (0xffUL << 0)
__le32 event_data2;
u8 opaque_v;
@@ -141,6 +143,7 @@ struct hwrm_async_event_cmpl_link_status_change {
#define HWRM_ASYNC_EVENT_CMPL_LINK_STATUS_CHANGE_EVENT_DATA1_LINK_CHANGE 0x1UL
#define HWRM_ASYNC_EVENT_CMPL_LINK_STATUS_CHANGE_EVENT_DATA1_LINK_CHANGE_DOWN (0x0UL << 0)
#define HWRM_ASYNC_EVENT_CMPL_LINK_STATUS_CHANGE_EVENT_DATA1_LINK_CHANGE_UP (0x1UL << 0)
+ #define HWRM_ASYNC_EVENT_CMPL_LINK_STATUS_CHANGE_EVENT_DATA1_LINK_CHANGE_LAST HWRM_ASYNC_EVENT_CMPL_LINK_STATUS_CHANGE_EVENT_DATA1_LINK_CHANGE_UP
#define HWRM_ASYNC_EVENT_CMPL_LINK_STATUS_CHANGE_EVENT_DATA1_PORT_MASK 0xeUL
#define HWRM_ASYNC_EVENT_CMPL_LINK_STATUS_CHANGE_EVENT_DATA1_PORT_SFT 1
#define HWRM_ASYNC_EVENT_CMPL_LINK_STATUS_CHANGE_EVENT_DATA1_PORT_ID_MASK 0xffff0UL
@@ -195,6 +198,9 @@ struct hwrm_async_event_cmpl_link_speed_change {
#define HWRM_ASYNC_EVENT_CMPL_LINK_SPEED_CHANGE_EVENT_DATA1_NEW_LINK_SPEED_100MBPS_25GB (0xfaUL << 1)
#define HWRM_ASYNC_EVENT_CMPL_LINK_SPEED_CHANGE_EVENT_DATA1_NEW_LINK_SPEED_100MBPS_40GB (0x190UL << 1)
#define HWRM_ASYNC_EVENT_CMPL_LINK_SPEED_CHANGE_EVENT_DATA1_NEW_LINK_SPEED_100MBPS_50GB (0x1f4UL << 1)
+ #define HWRM_ASYNC_EVENT_CMPL_LINK_SPEED_CHANGE_EVENT_DATA1_NEW_LINK_SPEED_100MBPS_100GB (0x3e8UL << 1)
+ #define HWRM_ASYNC_EVENT_CMPL_LINK_SPEED_CHANGE_EVENT_DATA1_NEW_LINK_SPEED_100MBPS_10MB (0xffffUL << 1)
+ #define HWRM_ASYNC_EVENT_CMPL_LINK_SPEED_CHANGE_EVENT_DATA1_NEW_LINK_SPEED_100MBPS_LAST HWRM_ASYNC_EVENT_CMPL_LINK_SPEED_CHANGE_EVENT_DATA1_NEW_LINK_SPEED_100MBPS_10MB
#define HWRM_ASYNC_EVENT_CMPL_LINK_SPEED_CHANGE_EVENT_DATA1_PORT_ID_MASK 0xffff0000UL
#define HWRM_ASYNC_EVENT_CMPL_LINK_SPEED_CHANGE_EVENT_DATA1_PORT_ID_SFT 16
};
@@ -237,6 +243,55 @@ struct hwrm_async_event_cmpl_port_conn_not_allowed {
__le32 event_data1;
#define HWRM_ASYNC_EVENT_CMPL_PORT_CONN_NOT_ALLOWED_EVENT_DATA1_PORT_ID_MASK 0xffffUL
#define HWRM_ASYNC_EVENT_CMPL_PORT_CONN_NOT_ALLOWED_EVENT_DATA1_PORT_ID_SFT 0
+ #define HWRM_ASYNC_EVENT_CMPL_PORT_CONN_NOT_ALLOWED_EVENT_DATA1_ENFORCEMENT_POLICY_MASK 0xff0000UL
+ #define HWRM_ASYNC_EVENT_CMPL_PORT_CONN_NOT_ALLOWED_EVENT_DATA1_ENFORCEMENT_POLICY_SFT 16
+ #define HWRM_ASYNC_EVENT_CMPL_PORT_CONN_NOT_ALLOWED_EVENT_DATA1_ENFORCEMENT_POLICY_NONE (0x0UL << 16)
+ #define HWRM_ASYNC_EVENT_CMPL_PORT_CONN_NOT_ALLOWED_EVENT_DATA1_ENFORCEMENT_POLICY_DISABLETX (0x1UL << 16)
+ #define HWRM_ASYNC_EVENT_CMPL_PORT_CONN_NOT_ALLOWED_EVENT_DATA1_ENFORCEMENT_POLICY_WARNINGMSG (0x2UL << 16)
+ #define HWRM_ASYNC_EVENT_CMPL_PORT_CONN_NOT_ALLOWED_EVENT_DATA1_ENFORCEMENT_POLICY_PWRDOWN (0x3UL << 16)
+ #define HWRM_ASYNC_EVENT_CMPL_PORT_CONN_NOT_ALLOWED_EVENT_DATA1_ENFORCEMENT_POLICY_LAST HWRM_ASYNC_EVENT_CMPL_PORT_CONN_NOT_ALLOWED_EVENT_DATA1_ENFORCEMENT_POLICY_PWRDOWN
+};
+
+/* HWRM Asynchronous Event Completion Record for link speed config not allowed (16 bytes) */
+struct hwrm_async_event_cmpl_link_speed_cfg_not_allowed {
+ __le16 type;
+ #define HWRM_ASYNC_EVENT_CMPL_LINK_SPEED_CFG_NOT_ALLOWED_TYPE_MASK 0x3fUL
+ #define HWRM_ASYNC_EVENT_CMPL_LINK_SPEED_CFG_NOT_ALLOWED_TYPE_SFT 0
+ #define HWRM_ASYNC_EVENT_CMPL_LINK_SPEED_CFG_NOT_ALLOWED_TYPE_HWRM_ASYNC_EVENT (0x2eUL << 0)
+ __le16 event_id;
+ #define HWRM_ASYNC_EVENT_CMPL_LINK_SPEED_CFG_NOT_ALLOWED_EVENT_ID_LINK_SPEED_CFG_NOT_ALLOWED (0x5UL << 0)
+ __le32 event_data2;
+ u8 opaque_v;
+ #define HWRM_ASYNC_EVENT_CMPL_LINK_SPEED_CFG_NOT_ALLOWED_V 0x1UL
+ #define HWRM_ASYNC_EVENT_CMPL_LINK_SPEED_CFG_NOT_ALLOWED_OPAQUE_MASK 0xfeUL
+ #define HWRM_ASYNC_EVENT_CMPL_LINK_SPEED_CFG_NOT_ALLOWED_OPAQUE_SFT 1
+ u8 timestamp_lo;
+ __le16 timestamp_hi;
+ __le32 event_data1;
+ #define HWRM_ASYNC_EVENT_CMPL_LINK_SPEED_CFG_NOT_ALLOWED_EVENT_DATA1_PORT_ID_MASK 0xffffUL
+ #define HWRM_ASYNC_EVENT_CMPL_LINK_SPEED_CFG_NOT_ALLOWED_EVENT_DATA1_PORT_ID_SFT 0
+};
+
+/* HWRM Asynchronous Event Completion Record for link speed configuration change (16 bytes) */
+struct hwrm_async_event_cmpl_link_speed_cfg_change {
+ __le16 type;
+ #define HWRM_ASYNC_EVENT_CMPL_LINK_SPEED_CFG_CHANGE_TYPE_MASK 0x3fUL
+ #define HWRM_ASYNC_EVENT_CMPL_LINK_SPEED_CFG_CHANGE_TYPE_SFT 0
+ #define HWRM_ASYNC_EVENT_CMPL_LINK_SPEED_CFG_CHANGE_TYPE_HWRM_ASYNC_EVENT (0x2eUL << 0)
+ __le16 event_id;
+ #define HWRM_ASYNC_EVENT_CMPL_LINK_SPEED_CFG_CHANGE_EVENT_ID_LINK_SPEED_CFG_CHANGE (0x6UL << 0)
+ __le32 event_data2;
+ u8 opaque_v;
+ #define HWRM_ASYNC_EVENT_CMPL_LINK_SPEED_CFG_CHANGE_V 0x1UL
+ #define HWRM_ASYNC_EVENT_CMPL_LINK_SPEED_CFG_CHANGE_OPAQUE_MASK 0xfeUL
+ #define HWRM_ASYNC_EVENT_CMPL_LINK_SPEED_CFG_CHANGE_OPAQUE_SFT 1
+ u8 timestamp_lo;
+ __le16 timestamp_hi;
+ __le32 event_data1;
+ #define HWRM_ASYNC_EVENT_CMPL_LINK_SPEED_CFG_CHANGE_EVENT_DATA1_PORT_ID_MASK 0xffffUL
+ #define HWRM_ASYNC_EVENT_CMPL_LINK_SPEED_CFG_CHANGE_EVENT_DATA1_PORT_ID_SFT 0
+ #define HWRM_ASYNC_EVENT_CMPL_LINK_SPEED_CFG_CHANGE_EVENT_DATA1_SUPPORTED_LINK_SPEEDS_CHANGE 0x10000UL
+ #define HWRM_ASYNC_EVENT_CMPL_LINK_SPEED_CFG_CHANGE_EVENT_DATA1_ILLEGAL_LINK_SPEED_CFG 0x20000UL
};
/* HWRM Asynchronous Event Completion Record for Function Driver Unload (16 bytes) */
@@ -363,6 +418,47 @@ struct hwrm_async_event_cmpl_vf_mac_addr_change {
#define HWRM_ASYNC_EVENT_CMPL_VF_MAC_ADDR_CHANGE_EVENT_DATA1_VF_ID_SFT 0
};
+/* HWRM Asynchronous Event Completion Record for PF-VF communication status change (16 bytes) */
+struct hwrm_async_event_cmpl_pf_vf_comm_status_change {
+ __le16 type;
+ #define HWRM_ASYNC_EVENT_CMPL_PF_VF_COMM_STATUS_CHANGE_TYPE_MASK 0x3fUL
+ #define HWRM_ASYNC_EVENT_CMPL_PF_VF_COMM_STATUS_CHANGE_TYPE_SFT 0
+ #define HWRM_ASYNC_EVENT_CMPL_PF_VF_COMM_STATUS_CHANGE_TYPE_HWRM_ASYNC_EVENT (0x2eUL << 0)
+ __le16 event_id;
+ #define HWRM_ASYNC_EVENT_CMPL_PF_VF_COMM_STATUS_CHANGE_EVENT_ID_PF_VF_COMM_STATUS_CHANGE (0x32UL << 0)
+ __le32 event_data2;
+ u8 opaque_v;
+ #define HWRM_ASYNC_EVENT_CMPL_PF_VF_COMM_STATUS_CHANGE_V 0x1UL
+ #define HWRM_ASYNC_EVENT_CMPL_PF_VF_COMM_STATUS_CHANGE_OPAQUE_MASK 0xfeUL
+ #define HWRM_ASYNC_EVENT_CMPL_PF_VF_COMM_STATUS_CHANGE_OPAQUE_SFT 1
+ u8 timestamp_lo;
+ __le16 timestamp_hi;
+ __le32 event_data1;
+ #define HWRM_ASYNC_EVENT_CMPL_PF_VF_COMM_STATUS_CHANGE_EVENT_DATA1_COMM_ESTABLISHED 0x1UL
+};
+
+/* HWRM Asynchronous Event Completion Record for VF configuration change (16 bytes) */
+struct hwrm_async_event_cmpl_vf_cfg_change {
+ __le16 type;
+ #define HWRM_ASYNC_EVENT_CMPL_VF_CFG_CHANGE_TYPE_MASK 0x3fUL
+ #define HWRM_ASYNC_EVENT_CMPL_VF_CFG_CHANGE_TYPE_SFT 0
+ #define HWRM_ASYNC_EVENT_CMPL_VF_CFG_CHANGE_TYPE_HWRM_ASYNC_EVENT (0x2eUL << 0)
+ __le16 event_id;
+ #define HWRM_ASYNC_EVENT_CMPL_VF_CFG_CHANGE_EVENT_ID_VF_CFG_CHANGE (0x33UL << 0)
+ __le32 event_data2;
+ u8 opaque_v;
+ #define HWRM_ASYNC_EVENT_CMPL_VF_CFG_CHANGE_V 0x1UL
+ #define HWRM_ASYNC_EVENT_CMPL_VF_CFG_CHANGE_OPAQUE_MASK 0xfeUL
+ #define HWRM_ASYNC_EVENT_CMPL_VF_CFG_CHANGE_OPAQUE_SFT 1
+ u8 timestamp_lo;
+ __le16 timestamp_hi;
+ __le32 event_data1;
+ #define HWRM_ASYNC_EVENT_CMPL_VF_CFG_CHANGE_EVENT_DATA1_MTU_CHANGE 0x1UL
+ #define HWRM_ASYNC_EVENT_CMPL_VF_CFG_CHANGE_EVENT_DATA1_MRU_CHANGE 0x2UL
+ #define HWRM_ASYNC_EVENT_CMPL_VF_CFG_CHANGE_EVENT_DATA1_DFLT_MAC_ADDR_CHANGE 0x4UL
+ #define HWRM_ASYNC_EVENT_CMPL_VF_CFG_CHANGE_EVENT_DATA1_DFLT_VLAN_CHANGE 0x8UL
+};
+
/* HWRM Asynchronous Event Completion Record for HWRM Error (16 bytes) */
struct hwrm_async_event_cmpl_hwrm_error {
__le16 type;
@@ -377,6 +473,7 @@ struct hwrm_async_event_cmpl_hwrm_error {
#define HWRM_ASYNC_EVENT_CMPL_HWRM_ERROR_EVENT_DATA2_SEVERITY_WARNING (0x0UL << 0)
#define HWRM_ASYNC_EVENT_CMPL_HWRM_ERROR_EVENT_DATA2_SEVERITY_NONFATAL (0x1UL << 0)
#define HWRM_ASYNC_EVENT_CMPL_HWRM_ERROR_EVENT_DATA2_SEVERITY_FATAL (0x2UL << 0)
+ #define HWRM_ASYNC_EVENT_CMPL_HWRM_ERROR_EVENT_DATA2_SEVERITY_LAST HWRM_ASYNC_EVENT_CMPL_HWRM_ERROR_EVENT_DATA2_SEVERITY_FATAL
u8 opaque_v;
#define HWRM_ASYNC_EVENT_CMPL_HWRM_ERROR_V 0x1UL
#define HWRM_ASYNC_EVENT_CMPL_HWRM_ERROR_OPAQUE_MASK 0xfeUL
@@ -387,12 +484,12 @@ struct hwrm_async_event_cmpl_hwrm_error {
#define HWRM_ASYNC_EVENT_CMPL_HWRM_ERROR_EVENT_DATA1_TIMESTAMP 0x1UL
};
-/* HW Resource Manager Specification 1.0.0 */
+/* HW Resource Manager Specification 1.2.2 */
#define HWRM_VERSION_MAJOR 1
-#define HWRM_VERSION_MINOR 0
-#define HWRM_VERSION_UPDATE 0
+#define HWRM_VERSION_MINOR 2
+#define HWRM_VERSION_UPDATE 2
-#define HWRM_VERSION_STR "1.0.0"
+#define HWRM_VERSION_STR "1.2.2"
/*
* Following is the signature for HWRM message field that indicates not
* applicable (All F's). Need to cast it the size of the field if needed.
@@ -444,7 +541,7 @@ struct cmd_nums {
#define HWRM_FUNC_BUF_RGTR (0x1fUL)
#define HWRM_PORT_PHY_CFG (0x20UL)
#define HWRM_PORT_MAC_CFG (0x21UL)
- #define RESERVED2 (0x22UL)
+ #define HWRM_PORT_TS_QUERY (0x22UL)
#define HWRM_PORT_QSTATS (0x23UL)
#define HWRM_PORT_LPBK_QSTATS (0x24UL)
#define HWRM_PORT_CLR_STATS (0x25UL)
@@ -452,6 +549,9 @@ struct cmd_nums {
#define HWRM_PORT_PHY_QCFG (0x27UL)
#define HWRM_PORT_MAC_QCFG (0x28UL)
#define HWRM_PORT_BLINK_LED (0x29UL)
+ #define HWRM_PORT_PHY_QCAPS (0x2aUL)
+ #define HWRM_PORT_PHY_I2C_WRITE (0x2bUL)
+ #define HWRM_PORT_PHY_I2C_READ (0x2cUL)
#define HWRM_QUEUE_QPORTCFG (0x30UL)
#define HWRM_QUEUE_QCFG (0x31UL)
#define HWRM_QUEUE_CFG (0x32UL)
@@ -531,6 +631,7 @@ struct cmd_nums {
__le16 unused_0[3];
};
+/* Return Codes (8 bytes) */
struct ret_codes {
__le16 error_code;
#define HWRM_ERR_CODE_SUCCESS (0x0UL)
@@ -875,10 +976,11 @@ struct hwrm_func_vf_cfg_input {
#define FUNC_VF_CFG_REQ_ENABLES_MTU 0x1UL
#define FUNC_VF_CFG_REQ_ENABLES_GUEST_VLAN 0x2UL
#define FUNC_VF_CFG_REQ_ENABLES_ASYNC_EVENT_CR 0x4UL
+ #define FUNC_VF_CFG_REQ_ENABLES_DFLT_MAC_ADDR 0x8UL
__le16 mtu;
__le16 guest_vlan;
__le16 async_event_cr;
- __le16 unused_0[3];
+ u8 dflt_mac_addr[6];
};
/* Output (16 bytes) */
@@ -917,7 +1019,8 @@ struct hwrm_func_qcaps_output {
__le32 flags;
#define FUNC_QCAPS_RESP_FLAGS_PUSH_MODE_SUPPORTED 0x1UL
#define FUNC_QCAPS_RESP_FLAGS_GLOBAL_MSIX_AUTOMASKING 0x2UL
- u8 perm_mac_address[6];
+ #define FUNC_QCAPS_RESP_FLAGS_PTP_SUPPORTED 0x4UL
+ u8 mac_address[6];
__le16 max_rsscos_ctx;
__le16 max_cmpl_rings;
__le16 max_tx_rings;
@@ -942,6 +1045,67 @@ struct hwrm_func_qcaps_output {
u8 valid;
};
+/* hwrm_func_qcfg */
+/* Input (24 bytes) */
+struct hwrm_func_qcfg_input {
+ __le16 req_type;
+ __le16 cmpl_ring;
+ __le16 seq_id;
+ __le16 target_id;
+ __le64 resp_addr;
+ __le16 fid;
+ __le16 unused_0[3];
+};
+
+/* Output (72 bytes) */
+struct hwrm_func_qcfg_output {
+ __le16 error_code;
+ __le16 req_type;
+ __le16 seq_id;
+ __le16 resp_len;
+ __le16 fid;
+ __le16 port_id;
+ __le16 vlan;
+ u8 unused_0;
+ u8 unused_1;
+ u8 mac_address[6];
+ __le16 pci_id;
+ __le16 alloc_rsscos_ctx;
+ __le16 alloc_cmpl_rings;
+ __le16 alloc_tx_rings;
+ __le16 alloc_rx_rings;
+ __le16 alloc_l2_ctx;
+ __le16 alloc_vnics;
+ __le16 mtu;
+ __le16 mru;
+ __le16 stat_ctx_id;
+ u8 port_partition_type;
+ #define FUNC_QCFG_RESP_PORT_PARTITION_TYPE_SPF (0x0UL << 0)
+ #define FUNC_QCFG_RESP_PORT_PARTITION_TYPE_MPFS (0x1UL << 0)
+ #define FUNC_QCFG_RESP_PORT_PARTITION_TYPE_NPAR1_0 (0x2UL << 0)
+ #define FUNC_QCFG_RESP_PORT_PARTITION_TYPE_NPAR1_5 (0x3UL << 0)
+ #define FUNC_QCFG_RESP_PORT_PARTITION_TYPE_NPAR2_0 (0x4UL << 0)
+ #define FUNC_QCFG_RESP_PORT_PARTITION_TYPE_UNKNOWN (0xffUL << 0)
+ u8 unused_2;
+ __le16 dflt_vnic_id;
+ u8 unused_3;
+ u8 unused_4;
+ __le32 min_bw;
+ __le32 max_bw;
+ u8 evb_mode;
+ #define FUNC_QCFG_RESP_EVB_MODE_NO_EVB (0x0UL << 0)
+ #define FUNC_QCFG_RESP_EVB_MODE_VEB (0x1UL << 0)
+ #define FUNC_QCFG_RESP_EVB_MODE_VEPA (0x2UL << 0)
+ u8 unused_5;
+ __le16 unused_6;
+ __le32 alloc_mcast_filters;
+ __le32 alloc_hw_ring_grps;
+ u8 unused_7;
+ u8 unused_8;
+ u8 unused_9;
+ u8 valid;
+};
+
/* hwrm_func_cfg */
/* Input (88 bytes) */
struct hwrm_func_cfg_input {
@@ -1171,6 +1335,7 @@ struct hwrm_func_drv_rgtr_input {
#define FUNC_DRV_RGTR_REQ_OS_TYPE_UNKNOWN (0x0UL << 0)
#define FUNC_DRV_RGTR_REQ_OS_TYPE_OTHER (0x1UL << 0)
#define FUNC_DRV_RGTR_REQ_OS_TYPE_MSDOS (0xeUL << 0)
+ #define FUNC_DRV_RGTR_REQ_OS_TYPE_WINDOWS (0x12UL << 0)
#define FUNC_DRV_RGTR_REQ_OS_TYPE_SOLARIS (0x1dUL << 0)
#define FUNC_DRV_RGTR_REQ_OS_TYPE_LINUX (0x24UL << 0)
#define FUNC_DRV_RGTR_REQ_OS_TYPE_FREEBSD (0x2aUL << 0)
@@ -1302,6 +1467,7 @@ struct hwrm_func_drv_qver_output {
#define FUNC_DRV_QVER_RESP_OS_TYPE_UNKNOWN (0x0UL << 0)
#define FUNC_DRV_QVER_RESP_OS_TYPE_OTHER (0x1UL << 0)
#define FUNC_DRV_QVER_RESP_OS_TYPE_MSDOS (0xeUL << 0)
+ #define FUNC_DRV_QVER_RESP_OS_TYPE_WINDOWS (0x12UL << 0)
#define FUNC_DRV_QVER_RESP_OS_TYPE_SOLARIS (0x1dUL << 0)
#define FUNC_DRV_QVER_RESP_OS_TYPE_LINUX (0x24UL << 0)
#define FUNC_DRV_QVER_RESP_OS_TYPE_FREEBSD (0x2aUL << 0)
@@ -1317,7 +1483,7 @@ struct hwrm_func_drv_qver_output {
};
/* hwrm_port_phy_cfg */
-/* Input (48 bytes) */
+/* Input (56 bytes) */
struct hwrm_port_phy_cfg_input {
__le16 req_type;
__le16 cmpl_ring;
@@ -1329,6 +1495,10 @@ struct hwrm_port_phy_cfg_input {
#define PORT_PHY_CFG_REQ_FLAGS_FORCE_LINK_DOWN 0x2UL
#define PORT_PHY_CFG_REQ_FLAGS_FORCE 0x4UL
#define PORT_PHY_CFG_REQ_FLAGS_RESTART_AUTONEG 0x8UL
+ #define PORT_PHY_CFG_REQ_FLAGS_EEE_ENABLE 0x10UL
+ #define PORT_PHY_CFG_REQ_FLAGS_EEE_DISABLE 0x20UL
+ #define PORT_PHY_CFG_REQ_FLAGS_EEE_TX_LPI_ENABLE 0x40UL
+ #define PORT_PHY_CFG_REQ_FLAGS_EEE_TX_LPI_DISABLE 0x80UL
__le32 enables;
#define PORT_PHY_CFG_REQ_ENABLES_AUTO_MODE 0x1UL
#define PORT_PHY_CFG_REQ_ENABLES_AUTO_DUPLEX 0x2UL
@@ -1339,6 +1509,8 @@ struct hwrm_port_phy_cfg_input {
#define PORT_PHY_CFG_REQ_ENABLES_LPBK 0x40UL
#define PORT_PHY_CFG_REQ_ENABLES_PREEMPHASIS 0x80UL
#define PORT_PHY_CFG_REQ_ENABLES_FORCE_PAUSE 0x100UL
+ #define PORT_PHY_CFG_REQ_ENABLES_EEE_LINK_SPEED_MASK 0x200UL
+ #define PORT_PHY_CFG_REQ_ENABLES_TX_LPI_TIMER 0x400UL
__le16 port_id;
__le16 force_link_speed;
#define PORT_PHY_CFG_REQ_FORCE_LINK_SPEED_100MB (0x1UL << 0)
@@ -1350,12 +1522,14 @@ struct hwrm_port_phy_cfg_input {
#define PORT_PHY_CFG_REQ_FORCE_LINK_SPEED_25GB (0xfaUL << 0)
#define PORT_PHY_CFG_REQ_FORCE_LINK_SPEED_40GB (0x190UL << 0)
#define PORT_PHY_CFG_REQ_FORCE_LINK_SPEED_50GB (0x1f4UL << 0)
+ #define PORT_PHY_CFG_REQ_FORCE_LINK_SPEED_100GB (0x3e8UL << 0)
+ #define PORT_PHY_CFG_REQ_FORCE_LINK_SPEED_10MB (0xffffUL << 0)
u8 auto_mode;
#define PORT_PHY_CFG_REQ_AUTO_MODE_NONE (0x0UL << 0)
#define PORT_PHY_CFG_REQ_AUTO_MODE_ALL_SPEEDS (0x1UL << 0)
#define PORT_PHY_CFG_REQ_AUTO_MODE_ONE_SPEED (0x2UL << 0)
#define PORT_PHY_CFG_REQ_AUTO_MODE_ONE_OR_BELOW (0x3UL << 0)
- #define PORT_PHY_CFG_REQ_AUTO_MODE_MASK (0x4UL << 0)
+ #define PORT_PHY_CFG_REQ_AUTO_MODE_SPEED_MASK (0x4UL << 0)
u8 auto_duplex;
#define PORT_PHY_CFG_REQ_AUTO_DUPLEX_HALF (0x0UL << 0)
#define PORT_PHY_CFG_REQ_AUTO_DUPLEX_FULL (0x1UL << 0)
@@ -1363,6 +1537,7 @@ struct hwrm_port_phy_cfg_input {
u8 auto_pause;
#define PORT_PHY_CFG_REQ_AUTO_PAUSE_TX 0x1UL
#define PORT_PHY_CFG_REQ_AUTO_PAUSE_RX 0x2UL
+ #define PORT_PHY_CFG_REQ_AUTO_PAUSE_AUTONEG_PAUSE 0x4UL
u8 unused_0;
__le16 auto_link_speed;
#define PORT_PHY_CFG_REQ_AUTO_LINK_SPEED_100MB (0x1UL << 0)
@@ -1374,6 +1549,8 @@ struct hwrm_port_phy_cfg_input {
#define PORT_PHY_CFG_REQ_AUTO_LINK_SPEED_25GB (0xfaUL << 0)
#define PORT_PHY_CFG_REQ_AUTO_LINK_SPEED_40GB (0x190UL << 0)
#define PORT_PHY_CFG_REQ_AUTO_LINK_SPEED_50GB (0x1f4UL << 0)
+ #define PORT_PHY_CFG_REQ_AUTO_LINK_SPEED_100GB (0x3e8UL << 0)
+ #define PORT_PHY_CFG_REQ_AUTO_LINK_SPEED_10MB (0xffffUL << 0)
__le16 auto_link_speed_mask;
#define PORT_PHY_CFG_REQ_AUTO_LINK_SPEED_MASK_100MBHD 0x1UL
#define PORT_PHY_CFG_REQ_AUTO_LINK_SPEED_MASK_100MB 0x2UL
@@ -1386,6 +1563,9 @@ struct hwrm_port_phy_cfg_input {
#define PORT_PHY_CFG_REQ_AUTO_LINK_SPEED_MASK_25GB 0x100UL
#define PORT_PHY_CFG_REQ_AUTO_LINK_SPEED_MASK_40GB 0x200UL
#define PORT_PHY_CFG_REQ_AUTO_LINK_SPEED_MASK_50GB 0x400UL
+ #define PORT_PHY_CFG_REQ_AUTO_LINK_SPEED_MASK_100GB 0x800UL
+ #define PORT_PHY_CFG_REQ_AUTO_LINK_SPEED_MASK_10MBHD 0x1000UL
+ #define PORT_PHY_CFG_REQ_AUTO_LINK_SPEED_MASK_10MB 0x2000UL
u8 wirespeed;
#define PORT_PHY_CFG_REQ_WIRESPEED_OFF (0x0UL << 0)
#define PORT_PHY_CFG_REQ_WIRESPEED_ON (0x1UL << 0)
@@ -1398,7 +1578,20 @@ struct hwrm_port_phy_cfg_input {
#define PORT_PHY_CFG_REQ_FORCE_PAUSE_RX 0x2UL
u8 unused_1;
__le32 preemphasis;
- __le32 unused_2;
+ __le16 eee_link_speed_mask;
+ #define PORT_PHY_CFG_REQ_EEE_LINK_SPEED_MASK_RSVD1 0x1UL
+ #define PORT_PHY_CFG_REQ_EEE_LINK_SPEED_MASK_100MB 0x2UL
+ #define PORT_PHY_CFG_REQ_EEE_LINK_SPEED_MASK_RSVD2 0x4UL
+ #define PORT_PHY_CFG_REQ_EEE_LINK_SPEED_MASK_1GB 0x8UL
+ #define PORT_PHY_CFG_REQ_EEE_LINK_SPEED_MASK_RSVD3 0x10UL
+ #define PORT_PHY_CFG_REQ_EEE_LINK_SPEED_MASK_RSVD4 0x20UL
+ #define PORT_PHY_CFG_REQ_EEE_LINK_SPEED_MASK_10GB 0x40UL
+ u8 unused_2;
+ u8 unused_3;
+ __le32 tx_lpi_timer;
+ __le32 unused_4;
+ #define PORT_PHY_CFG_REQ_TX_LPI_TIMER_MASK 0xffffffUL
+ #define PORT_PHY_CFG_REQ_TX_LPI_TIMER_SFT 0
};
/* Output (16 bytes) */
@@ -1426,7 +1619,7 @@ struct hwrm_port_phy_qcfg_input {
__le16 unused_0[3];
};
-/* Output (48 bytes) */
+/* Output (96 bytes) */
struct hwrm_port_phy_qcfg_output {
__le16 error_code;
__le16 req_type;
@@ -1447,6 +1640,8 @@ struct hwrm_port_phy_qcfg_output {
#define PORT_PHY_QCFG_RESP_LINK_SPEED_25GB (0xfaUL << 0)
#define PORT_PHY_QCFG_RESP_LINK_SPEED_40GB (0x190UL << 0)
#define PORT_PHY_QCFG_RESP_LINK_SPEED_50GB (0x1f4UL << 0)
+ #define PORT_PHY_QCFG_RESP_LINK_SPEED_100GB (0x3e8UL << 0)
+ #define PORT_PHY_QCFG_RESP_LINK_SPEED_10MB (0xffffUL << 0)
u8 duplex;
#define PORT_PHY_QCFG_RESP_DUPLEX_HALF (0x0UL << 0)
#define PORT_PHY_QCFG_RESP_DUPLEX_FULL (0x1UL << 0)
@@ -1465,6 +1660,9 @@ struct hwrm_port_phy_qcfg_output {
#define PORT_PHY_QCFG_RESP_SUPPORT_SPEEDS_25GB 0x100UL
#define PORT_PHY_QCFG_RESP_SUPPORT_SPEEDS_40GB 0x200UL
#define PORT_PHY_QCFG_RESP_SUPPORT_SPEEDS_50GB 0x400UL
+ #define PORT_PHY_QCFG_RESP_SUPPORT_SPEEDS_100GB 0x800UL
+ #define PORT_PHY_QCFG_RESP_SUPPORT_SPEEDS_10MBHD 0x1000UL
+ #define PORT_PHY_QCFG_RESP_SUPPORT_SPEEDS_10MB 0x2000UL
__le16 force_link_speed;
#define PORT_PHY_QCFG_RESP_FORCE_LINK_SPEED_100MB (0x1UL << 0)
#define PORT_PHY_QCFG_RESP_FORCE_LINK_SPEED_1GB (0xaUL << 0)
@@ -1475,15 +1673,18 @@ struct hwrm_port_phy_qcfg_output {
#define PORT_PHY_QCFG_RESP_FORCE_LINK_SPEED_25GB (0xfaUL << 0)
#define PORT_PHY_QCFG_RESP_FORCE_LINK_SPEED_40GB (0x190UL << 0)
#define PORT_PHY_QCFG_RESP_FORCE_LINK_SPEED_50GB (0x1f4UL << 0)
+ #define PORT_PHY_QCFG_RESP_FORCE_LINK_SPEED_100GB (0x3e8UL << 0)
+ #define PORT_PHY_QCFG_RESP_FORCE_LINK_SPEED_10MB (0xffffUL << 0)
u8 auto_mode;
#define PORT_PHY_QCFG_RESP_AUTO_MODE_NONE (0x0UL << 0)
#define PORT_PHY_QCFG_RESP_AUTO_MODE_ALL_SPEEDS (0x1UL << 0)
#define PORT_PHY_QCFG_RESP_AUTO_MODE_ONE_SPEED (0x2UL << 0)
#define PORT_PHY_QCFG_RESP_AUTO_MODE_ONE_OR_BELOW (0x3UL << 0)
- #define PORT_PHY_QCFG_RESP_AUTO_MODE_MASK (0x4UL << 0)
+ #define PORT_PHY_QCFG_RESP_AUTO_MODE_SPEED_MASK (0x4UL << 0)
u8 auto_pause;
#define PORT_PHY_QCFG_RESP_AUTO_PAUSE_TX 0x1UL
#define PORT_PHY_QCFG_RESP_AUTO_PAUSE_RX 0x2UL
+ #define PORT_PHY_QCFG_RESP_AUTO_PAUSE_AUTONEG_PAUSE 0x4UL
__le16 auto_link_speed;
#define PORT_PHY_QCFG_RESP_AUTO_LINK_SPEED_100MB (0x1UL << 0)
#define PORT_PHY_QCFG_RESP_AUTO_LINK_SPEED_1GB (0xaUL << 0)
@@ -1494,6 +1695,8 @@ struct hwrm_port_phy_qcfg_output {
#define PORT_PHY_QCFG_RESP_AUTO_LINK_SPEED_25GB (0xfaUL << 0)
#define PORT_PHY_QCFG_RESP_AUTO_LINK_SPEED_40GB (0x190UL << 0)
#define PORT_PHY_QCFG_RESP_AUTO_LINK_SPEED_50GB (0x1f4UL << 0)
+ #define PORT_PHY_QCFG_RESP_AUTO_LINK_SPEED_100GB (0x3e8UL << 0)
+ #define PORT_PHY_QCFG_RESP_AUTO_LINK_SPEED_10MB (0xffffUL << 0)
__le16 auto_link_speed_mask;
#define PORT_PHY_QCFG_RESP_AUTO_LINK_SPEED_MASK_100MBHD 0x1UL
#define PORT_PHY_QCFG_RESP_AUTO_LINK_SPEED_MASK_100MB 0x2UL
@@ -1506,6 +1709,9 @@ struct hwrm_port_phy_qcfg_output {
#define PORT_PHY_QCFG_RESP_AUTO_LINK_SPEED_MASK_25GB 0x100UL
#define PORT_PHY_QCFG_RESP_AUTO_LINK_SPEED_MASK_40GB 0x200UL
#define PORT_PHY_QCFG_RESP_AUTO_LINK_SPEED_MASK_50GB 0x400UL
+ #define PORT_PHY_QCFG_RESP_AUTO_LINK_SPEED_MASK_100GB 0x800UL
+ #define PORT_PHY_QCFG_RESP_AUTO_LINK_SPEED_MASK_10MBHD 0x1000UL
+ #define PORT_PHY_QCFG_RESP_AUTO_LINK_SPEED_MASK_10MB 0x2000UL
u8 wirespeed;
#define PORT_PHY_QCFG_RESP_WIRESPEED_OFF (0x0UL << 0)
#define PORT_PHY_QCFG_RESP_WIRESPEED_ON (0x1UL << 0)
@@ -1516,31 +1722,49 @@ struct hwrm_port_phy_qcfg_output {
u8 force_pause;
#define PORT_PHY_QCFG_RESP_FORCE_PAUSE_TX 0x1UL
#define PORT_PHY_QCFG_RESP_FORCE_PAUSE_RX 0x2UL
- u8 reserved1;
+ u8 module_status;
+ #define PORT_PHY_QCFG_RESP_MODULE_STATUS_NONE (0x0UL << 0)
+ #define PORT_PHY_QCFG_RESP_MODULE_STATUS_DISABLETX (0x1UL << 0)
+ #define PORT_PHY_QCFG_RESP_MODULE_STATUS_WARNINGMSG (0x2UL << 0)
+ #define PORT_PHY_QCFG_RESP_MODULE_STATUS_PWRDOWN (0x3UL << 0)
+ #define PORT_PHY_QCFG_RESP_MODULE_STATUS_NOTINSERTED (0x4UL << 0)
+ #define PORT_PHY_QCFG_RESP_MODULE_STATUS_NOTAPPLICABLE (0xffUL << 0)
__le32 preemphasis;
u8 phy_maj;
u8 phy_min;
u8 phy_bld;
u8 phy_type;
- #define PORT_PHY_QCFG_RESP_PHY_TYPE_BASECR4 (0x1UL << 0)
+ #define PORT_PHY_QCFG_RESP_PHY_TYPE_UNKNOWN (0x0UL << 0)
+ #define PORT_PHY_QCFG_RESP_PHY_TYPE_BASECR (0x1UL << 0)
#define PORT_PHY_QCFG_RESP_PHY_TYPE_BASEKR4 (0x2UL << 0)
- #define PORT_PHY_QCFG_RESP_PHY_TYPE_BASELR4 (0x3UL << 0)
- #define PORT_PHY_QCFG_RESP_PHY_TYPE_BASESR4 (0x4UL << 0)
+ #define PORT_PHY_QCFG_RESP_PHY_TYPE_BASELR (0x3UL << 0)
+ #define PORT_PHY_QCFG_RESP_PHY_TYPE_BASESR (0x4UL << 0)
#define PORT_PHY_QCFG_RESP_PHY_TYPE_BASEKR2 (0x5UL << 0)
- #define PORT_PHY_QCFG_RESP_PHY_TYPE_BASEKX4 (0x6UL << 0)
+ #define PORT_PHY_QCFG_RESP_PHY_TYPE_BASEKX (0x6UL << 0)
#define PORT_PHY_QCFG_RESP_PHY_TYPE_BASEKR (0x7UL << 0)
#define PORT_PHY_QCFG_RESP_PHY_TYPE_BASET (0x8UL << 0)
+ #define PORT_PHY_QCFG_RESP_PHY_TYPE_BASETE (0x9UL << 0)
+ #define PORT_PHY_QCFG_RESP_PHY_TYPE_SGMIIEXTPHY (0xaUL << 0)
u8 media_type;
+ #define PORT_PHY_QCFG_RESP_MEDIA_TYPE_UNKNOWN (0x0UL << 0)
#define PORT_PHY_QCFG_RESP_MEDIA_TYPE_TP (0x1UL << 0)
#define PORT_PHY_QCFG_RESP_MEDIA_TYPE_DAC (0x2UL << 0)
#define PORT_PHY_QCFG_RESP_MEDIA_TYPE_FIBRE (0x3UL << 0)
- u8 transceiver_type;
- #define PORT_PHY_QCFG_RESP_TRANSCEIVER_TYPE_XCVR_INTERNAL (0x1UL << 0)
- #define PORT_PHY_QCFG_RESP_TRANSCEIVER_TYPE_XCVR_EXTERNAL (0x2UL << 0)
- u8 phy_addr;
+ u8 xcvr_pkg_type;
+ #define PORT_PHY_QCFG_RESP_XCVR_PKG_TYPE_XCVR_INTERNAL (0x1UL << 0)
+ #define PORT_PHY_QCFG_RESP_XCVR_PKG_TYPE_XCVR_EXTERNAL (0x2UL << 0)
+ u8 eee_config_phy_addr;
#define PORT_PHY_QCFG_RESP_PHY_ADDR_MASK 0x1fUL
#define PORT_PHY_QCFG_RESP_PHY_ADDR_SFT 0
- u8 unused_2;
+ #define PORT_PHY_QCFG_RESP_EEE_CONFIG_EEE_ENABLED 0x20UL
+ #define PORT_PHY_QCFG_RESP_EEE_CONFIG_EEE_ACTIVE 0x40UL
+ #define PORT_PHY_QCFG_RESP_EEE_CONFIG_EEE_TX_LPI 0x80UL
+ #define PORT_PHY_QCFG_RESP_EEE_CONFIG_MASK 0xe0UL
+ #define PORT_PHY_QCFG_RESP_EEE_CONFIG_SFT 5
+ u8 parallel_detect;
+ #define PORT_PHY_QCFG_RESP_PARALLEL_DETECT 0x1UL
+ #define PORT_PHY_QCFG_RESP_RESERVED_MASK 0xfeUL
+ #define PORT_PHY_QCFG_RESP_RESERVED_SFT 1
__le16 link_partner_adv_speeds;
#define PORT_PHY_QCFG_RESP_LINK_PARTNER_ADV_SPEEDS_100MBHD 0x1UL
#define PORT_PHY_QCFG_RESP_LINK_PARTNER_ADV_SPEEDS_100MB 0x2UL
@@ -1553,15 +1777,48 @@ struct hwrm_port_phy_qcfg_output {
#define PORT_PHY_QCFG_RESP_LINK_PARTNER_ADV_SPEEDS_25GB 0x100UL
#define PORT_PHY_QCFG_RESP_LINK_PARTNER_ADV_SPEEDS_40GB 0x200UL
#define PORT_PHY_QCFG_RESP_LINK_PARTNER_ADV_SPEEDS_50GB 0x400UL
+ #define PORT_PHY_QCFG_RESP_LINK_PARTNER_ADV_SPEEDS_100GB 0x800UL
+ #define PORT_PHY_QCFG_RESP_LINK_PARTNER_ADV_SPEEDS_10MBHD 0x1000UL
+ #define PORT_PHY_QCFG_RESP_LINK_PARTNER_ADV_SPEEDS_10MB 0x2000UL
u8 link_partner_adv_auto_mode;
#define PORT_PHY_QCFG_RESP_LINK_PARTNER_ADV_AUTO_MODE_NONE (0x0UL << 0)
#define PORT_PHY_QCFG_RESP_LINK_PARTNER_ADV_AUTO_MODE_ALL_SPEEDS (0x1UL << 0)
#define PORT_PHY_QCFG_RESP_LINK_PARTNER_ADV_AUTO_MODE_ONE_SPEED (0x2UL << 0)
#define PORT_PHY_QCFG_RESP_LINK_PARTNER_ADV_AUTO_MODE_ONE_OR_BELOW (0x3UL << 0)
- #define PORT_PHY_QCFG_RESP_LINK_PARTNER_ADV_AUTO_MODE_MASK (0x4UL << 0)
+ #define PORT_PHY_QCFG_RESP_LINK_PARTNER_ADV_AUTO_MODE_SPEED_MASK (0x4UL << 0)
u8 link_partner_adv_pause;
#define PORT_PHY_QCFG_RESP_LINK_PARTNER_ADV_PAUSE_TX 0x1UL
#define PORT_PHY_QCFG_RESP_LINK_PARTNER_ADV_PAUSE_RX 0x2UL
+ __le16 adv_eee_link_speed_mask;
+ #define PORT_PHY_QCFG_RESP_ADV_EEE_LINK_SPEED_MASK_RSVD1 0x1UL
+ #define PORT_PHY_QCFG_RESP_ADV_EEE_LINK_SPEED_MASK_100MB 0x2UL
+ #define PORT_PHY_QCFG_RESP_ADV_EEE_LINK_SPEED_MASK_RSVD2 0x4UL
+ #define PORT_PHY_QCFG_RESP_ADV_EEE_LINK_SPEED_MASK_1GB 0x8UL
+ #define PORT_PHY_QCFG_RESP_ADV_EEE_LINK_SPEED_MASK_RSVD3 0x10UL
+ #define PORT_PHY_QCFG_RESP_ADV_EEE_LINK_SPEED_MASK_RSVD4 0x20UL
+ #define PORT_PHY_QCFG_RESP_ADV_EEE_LINK_SPEED_MASK_10GB 0x40UL
+ __le16 link_partner_adv_eee_link_speed_mask;
+ #define PORT_PHY_QCFG_RESP_LINK_PARTNER_ADV_EEE_LINK_SPEED_MASK_RSVD1 0x1UL
+ #define PORT_PHY_QCFG_RESP_LINK_PARTNER_ADV_EEE_LINK_SPEED_MASK_100MB 0x2UL
+ #define PORT_PHY_QCFG_RESP_LINK_PARTNER_ADV_EEE_LINK_SPEED_MASK_RSVD2 0x4UL
+ #define PORT_PHY_QCFG_RESP_LINK_PARTNER_ADV_EEE_LINK_SPEED_MASK_1GB 0x8UL
+ #define PORT_PHY_QCFG_RESP_LINK_PARTNER_ADV_EEE_LINK_SPEED_MASK_RSVD3 0x10UL
+ #define PORT_PHY_QCFG_RESP_LINK_PARTNER_ADV_EEE_LINK_SPEED_MASK_RSVD4 0x20UL
+ #define PORT_PHY_QCFG_RESP_LINK_PARTNER_ADV_EEE_LINK_SPEED_MASK_10GB 0x40UL
+ __le32 xcvr_identifier_type_tx_lpi_timer;
+ #define PORT_PHY_QCFG_RESP_TX_LPI_TIMER_MASK 0xffffffUL
+ #define PORT_PHY_QCFG_RESP_TX_LPI_TIMER_SFT 0
+ #define PORT_PHY_QCFG_RESP_XCVR_IDENTIFIER_TYPE_MASK 0xff000000UL
+ #define PORT_PHY_QCFG_RESP_XCVR_IDENTIFIER_TYPE_SFT 24
+ #define PORT_PHY_QCFG_RESP_XCVR_IDENTIFIER_TYPE_UNKNOWN (0x0UL << 24)
+ #define PORT_PHY_QCFG_RESP_XCVR_IDENTIFIER_TYPE_SFP (0x3UL << 24)
+ #define PORT_PHY_QCFG_RESP_XCVR_IDENTIFIER_TYPE_QSFP (0xcUL << 24)
+ #define PORT_PHY_QCFG_RESP_XCVR_IDENTIFIER_TYPE_QSFPPLUS (0xdUL << 24)
+ #define PORT_PHY_QCFG_RESP_XCVR_IDENTIFIER_TYPE_QSFP28 (0x11UL << 24)
+ __le32 unused_1;
+ char phy_vendor_name[16];
+ char phy_vendor_partnumber[16];
+ __le32 unused_2;
u8 unused_3;
u8 unused_4;
u8 unused_5;
@@ -1569,7 +1826,7 @@ struct hwrm_port_phy_qcfg_output {
};
/* hwrm_port_mac_cfg */
-/* Input (32 bytes) */
+/* Input (40 bytes) */
struct hwrm_port_mac_cfg_input {
__le16 req_type;
__le16 cmpl_ring;
@@ -1581,6 +1838,10 @@ struct hwrm_port_mac_cfg_input {
#define PORT_MAC_CFG_REQ_FLAGS_COS_ASSIGNMENT_ENABLE 0x2UL
#define PORT_MAC_CFG_REQ_FLAGS_TUNNEL_PRI2COS_ENABLE 0x4UL
#define PORT_MAC_CFG_REQ_FLAGS_IP_DSCP2COS_ENABLE 0x8UL
+ #define PORT_MAC_CFG_REQ_FLAGS_PTP_RX_TS_CAPTURE_ENABLE 0x10UL
+ #define PORT_MAC_CFG_REQ_FLAGS_PTP_RX_TS_CAPTURE_DISABLE 0x20UL
+ #define PORT_MAC_CFG_REQ_FLAGS_PTP_TX_TS_CAPTURE_ENABLE 0x40UL
+ #define PORT_MAC_CFG_REQ_FLAGS_PTP_TX_TS_CAPTURE_DISABLE 0x80UL
__le32 enables;
#define PORT_MAC_CFG_REQ_ENABLES_IPG 0x1UL
#define PORT_MAC_CFG_REQ_ENABLES_LPBK 0x2UL
@@ -1588,6 +1849,8 @@ struct hwrm_port_mac_cfg_input {
#define PORT_MAC_CFG_REQ_ENABLES_LCOS_MAP_PRI 0x8UL
#define PORT_MAC_CFG_REQ_ENABLES_TUNNEL_PRI2COS_MAP_PRI 0x10UL
#define PORT_MAC_CFG_REQ_ENABLES_DSCP2COS_MAP_PRI 0x20UL
+ #define PORT_MAC_CFG_REQ_ENABLES_RX_TS_CAPTURE_PTP_MSG_TYPE 0x40UL
+ #define PORT_MAC_CFG_REQ_ENABLES_TX_TS_CAPTURE_PTP_MSG_TYPE 0x80UL
__le16 port_id;
u8 ipg;
u8 lpbk;
@@ -1598,6 +1861,9 @@ struct hwrm_port_mac_cfg_input {
u8 lcos_map_pri;
u8 tunnel_pri2cos_map_pri;
u8 dscp2pri_map_pri;
+ __le16 rx_ts_capture_ptp_msg_type;
+ __le16 tx_ts_capture_ptp_msg_type;
+ __le32 unused_0;
};
/* Output (16 bytes) */
@@ -1754,7 +2020,79 @@ struct hwrm_port_blink_led_output {
u8 valid;
};
-/* hwrm_queue_qportcfg */
+/* hwrm_port_phy_qcaps */
+/* Input (24 bytes) */
+struct hwrm_port_phy_qcaps_input {
+ __le16 req_type;
+ __le16 cmpl_ring;
+ __le16 seq_id;
+ __le16 target_id;
+ __le64 resp_addr;
+ __le16 port_id;
+ __le16 unused_0[3];
+};
+
+/* Output (24 bytes) */
+struct hwrm_port_phy_qcaps_output {
+ __le16 error_code;
+ __le16 req_type;
+ __le16 seq_id;
+ __le16 resp_len;
+ u8 eee_supported;
+ #define PORT_PHY_QCAPS_RESP_EEE_SUPPORTED 0x1UL
+ #define PORT_PHY_QCAPS_RESP_RSVD1_MASK 0xfeUL
+ #define PORT_PHY_QCAPS_RESP_RSVD1_SFT 1
+ u8 unused_0;
+ __le16 supported_speeds_force_mode;
+ #define PORT_PHY_QCAPS_RESP_SUPPORTED_SPEEDS_FORCE_MODE_100MBHD 0x1UL
+ #define PORT_PHY_QCAPS_RESP_SUPPORTED_SPEEDS_FORCE_MODE_100MB 0x2UL
+ #define PORT_PHY_QCAPS_RESP_SUPPORTED_SPEEDS_FORCE_MODE_1GBHD 0x4UL
+ #define PORT_PHY_QCAPS_RESP_SUPPORTED_SPEEDS_FORCE_MODE_1GB 0x8UL
+ #define PORT_PHY_QCAPS_RESP_SUPPORTED_SPEEDS_FORCE_MODE_2GB 0x10UL
+ #define PORT_PHY_QCAPS_RESP_SUPPORTED_SPEEDS_FORCE_MODE_2_5GB 0x20UL
+ #define PORT_PHY_QCAPS_RESP_SUPPORTED_SPEEDS_FORCE_MODE_10GB 0x40UL
+ #define PORT_PHY_QCAPS_RESP_SUPPORTED_SPEEDS_FORCE_MODE_20GB 0x80UL
+ #define PORT_PHY_QCAPS_RESP_SUPPORTED_SPEEDS_FORCE_MODE_25GB 0x100UL
+ #define PORT_PHY_QCAPS_RESP_SUPPORTED_SPEEDS_FORCE_MODE_40GB 0x200UL
+ #define PORT_PHY_QCAPS_RESP_SUPPORTED_SPEEDS_FORCE_MODE_50GB 0x400UL
+ #define PORT_PHY_QCAPS_RESP_SUPPORTED_SPEEDS_FORCE_MODE_100GB 0x800UL
+ #define PORT_PHY_QCAPS_RESP_SUPPORTED_SPEEDS_FORCE_MODE_10MBHD 0x1000UL
+ #define PORT_PHY_QCAPS_RESP_SUPPORTED_SPEEDS_FORCE_MODE_10MB 0x2000UL
+ __le16 supported_speeds_auto_mode;
+ #define PORT_PHY_QCAPS_RESP_SUPPORTED_SPEEDS_AUTO_MODE_100MBHD 0x1UL
+ #define PORT_PHY_QCAPS_RESP_SUPPORTED_SPEEDS_AUTO_MODE_100MB 0x2UL
+ #define PORT_PHY_QCAPS_RESP_SUPPORTED_SPEEDS_AUTO_MODE_1GBHD 0x4UL
+ #define PORT_PHY_QCAPS_RESP_SUPPORTED_SPEEDS_AUTO_MODE_1GB 0x8UL
+ #define PORT_PHY_QCAPS_RESP_SUPPORTED_SPEEDS_AUTO_MODE_2GB 0x10UL
+ #define PORT_PHY_QCAPS_RESP_SUPPORTED_SPEEDS_AUTO_MODE_2_5GB 0x20UL
+ #define PORT_PHY_QCAPS_RESP_SUPPORTED_SPEEDS_AUTO_MODE_10GB 0x40UL
+ #define PORT_PHY_QCAPS_RESP_SUPPORTED_SPEEDS_AUTO_MODE_20GB 0x80UL
+ #define PORT_PHY_QCAPS_RESP_SUPPORTED_SPEEDS_AUTO_MODE_25GB 0x100UL
+ #define PORT_PHY_QCAPS_RESP_SUPPORTED_SPEEDS_AUTO_MODE_40GB 0x200UL
+ #define PORT_PHY_QCAPS_RESP_SUPPORTED_SPEEDS_AUTO_MODE_50GB 0x400UL
+ #define PORT_PHY_QCAPS_RESP_SUPPORTED_SPEEDS_AUTO_MODE_100GB 0x800UL
+ #define PORT_PHY_QCAPS_RESP_SUPPORTED_SPEEDS_AUTO_MODE_10MBHD 0x1000UL
+ #define PORT_PHY_QCAPS_RESP_SUPPORTED_SPEEDS_AUTO_MODE_10MB 0x2000UL
+ __le16 supported_speeds_eee_mode;
+ #define PORT_PHY_QCAPS_RESP_SUPPORTED_SPEEDS_EEE_MODE_RSVD1 0x1UL
+ #define PORT_PHY_QCAPS_RESP_SUPPORTED_SPEEDS_EEE_MODE_100MB 0x2UL
+ #define PORT_PHY_QCAPS_RESP_SUPPORTED_SPEEDS_EEE_MODE_RSVD2 0x4UL
+ #define PORT_PHY_QCAPS_RESP_SUPPORTED_SPEEDS_EEE_MODE_1GB 0x8UL
+ #define PORT_PHY_QCAPS_RESP_SUPPORTED_SPEEDS_EEE_MODE_RSVD3 0x10UL
+ #define PORT_PHY_QCAPS_RESP_SUPPORTED_SPEEDS_EEE_MODE_RSVD4 0x20UL
+ #define PORT_PHY_QCAPS_RESP_SUPPORTED_SPEEDS_EEE_MODE_10GB 0x40UL
+ __le32 tx_lpi_timer_low;
+ #define PORT_PHY_QCAPS_RESP_TX_LPI_TIMER_LOW_MASK 0xffffffUL
+ #define PORT_PHY_QCAPS_RESP_TX_LPI_TIMER_LOW_SFT 0
+ #define PORT_PHY_QCAPS_RESP_RSVD2_MASK 0xff000000UL
+ #define PORT_PHY_QCAPS_RESP_RSVD2_SFT 24
+ __le32 valid_tx_lpi_timer_high;
+ #define PORT_PHY_QCAPS_RESP_TX_LPI_TIMER_HIGH_MASK 0xffffffUL
+ #define PORT_PHY_QCAPS_RESP_TX_LPI_TIMER_HIGH_SFT 0
+ #define PORT_PHY_QCAPS_RESP_VALID_MASK 0xff000000UL
+ #define PORT_PHY_QCAPS_RESP_VALID_SFT 24
+};
+
/* Input (24 bytes) */
struct hwrm_queue_qportcfg_input {
__le16 req_type;
@@ -1766,6 +2104,7 @@ struct hwrm_queue_qportcfg_input {
#define QUEUE_QPORTCFG_REQ_FLAGS_PATH 0x1UL
#define QUEUE_QPORTCFG_REQ_FLAGS_PATH_TX (0x0UL << 0)
#define QUEUE_QPORTCFG_REQ_FLAGS_PATH_RX (0x1UL << 0)
+ #define QUEUE_QPORTCFG_REQ_FLAGS_PATH_LAST QUEUE_QPORTCFG_REQ_FLAGS_PATH_RX
__le16 port_id;
__le16 unused_0;
};
@@ -1838,6 +2177,7 @@ struct hwrm_queue_cfg_input {
#define QUEUE_CFG_REQ_FLAGS_PATH 0x1UL
#define QUEUE_CFG_REQ_FLAGS_PATH_TX (0x0UL << 0)
#define QUEUE_CFG_REQ_FLAGS_PATH_RX (0x1UL << 0)
+ #define QUEUE_CFG_REQ_FLAGS_PATH_LAST QUEUE_CFG_REQ_FLAGS_PATH_RX
__le32 enables;
#define QUEUE_CFG_REQ_ENABLES_DFLT_LEN 0x1UL
#define QUEUE_CFG_REQ_ENABLES_SERVICE_PROFILE 0x2UL
@@ -1875,6 +2215,7 @@ struct hwrm_queue_buffers_cfg_input {
#define QUEUE_BUFFERS_CFG_REQ_FLAGS_PATH 0x1UL
#define QUEUE_BUFFERS_CFG_REQ_FLAGS_PATH_TX (0x0UL << 0)
#define QUEUE_BUFFERS_CFG_REQ_FLAGS_PATH_RX (0x1UL << 0)
+ #define QUEUE_BUFFERS_CFG_REQ_FLAGS_PATH_LAST QUEUE_BUFFERS_CFG_REQ_FLAGS_PATH_RX
__le32 enables;
#define QUEUE_BUFFERS_CFG_REQ_ENABLES_RESERVED 0x1UL
#define QUEUE_BUFFERS_CFG_REQ_ENABLES_SHARED 0x2UL
@@ -1952,6 +2293,7 @@ struct hwrm_queue_pri2cos_cfg_input {
#define QUEUE_PRI2COS_CFG_REQ_FLAGS_PATH 0x1UL
#define QUEUE_PRI2COS_CFG_REQ_FLAGS_PATH_TX (0x0UL << 0)
#define QUEUE_PRI2COS_CFG_REQ_FLAGS_PATH_RX (0x1UL << 0)
+ #define QUEUE_PRI2COS_CFG_REQ_FLAGS_PATH_LAST QUEUE_PRI2COS_CFG_REQ_FLAGS_PATH_RX
#define QUEUE_PRI2COS_CFG_REQ_FLAGS_IVLAN 0x2UL
__le32 enables;
u8 port_id;
@@ -2158,6 +2500,8 @@ struct hwrm_vnic_cfg_input {
#define VNIC_CFG_REQ_FLAGS_DEFAULT 0x1UL
#define VNIC_CFG_REQ_FLAGS_VLAN_STRIP_MODE 0x2UL
#define VNIC_CFG_REQ_FLAGS_BD_STALL_MODE 0x4UL
+ #define VNIC_CFG_REQ_FLAGS_ROCE_DUAL_VNIC_MODE 0x8UL
+ #define VNIC_CFG_REQ_FLAGS_ROCE_ONLY_VNIC_MODE 0x10UL
__le32 enables;
#define VNIC_CFG_REQ_ENABLES_DFLT_RING_GRP 0x1UL
#define VNIC_CFG_REQ_ENABLES_RSS_RULE 0x2UL
@@ -2622,6 +2966,7 @@ struct hwrm_cfa_l2_filter_alloc_input {
#define CFA_L2_FILTER_ALLOC_REQ_FLAGS_PATH 0x1UL
#define CFA_L2_FILTER_ALLOC_REQ_FLAGS_PATH_TX (0x0UL << 0)
#define CFA_L2_FILTER_ALLOC_REQ_FLAGS_PATH_RX (0x1UL << 0)
+ #define CFA_L2_FILTER_ALLOC_REQ_FLAGS_PATH_LAST CFA_L2_FILTER_ALLOC_REQ_FLAGS_PATH_RX
#define CFA_L2_FILTER_ALLOC_REQ_FLAGS_LOOPBACK 0x2UL
#define CFA_L2_FILTER_ALLOC_REQ_FLAGS_DROP 0x4UL
#define CFA_L2_FILTER_ALLOC_REQ_FLAGS_OUTERMOST 0x8UL
@@ -2747,6 +3092,7 @@ struct hwrm_cfa_l2_filter_cfg_input {
#define CFA_L2_FILTER_CFG_REQ_FLAGS_PATH 0x1UL
#define CFA_L2_FILTER_CFG_REQ_FLAGS_PATH_TX (0x0UL << 0)
#define CFA_L2_FILTER_CFG_REQ_FLAGS_PATH_RX (0x1UL << 0)
+ #define CFA_L2_FILTER_CFG_REQ_FLAGS_PATH_LAST CFA_L2_FILTER_CFG_REQ_FLAGS_PATH_RX
#define CFA_L2_FILTER_CFG_REQ_FLAGS_DROP 0x2UL
__le32 enables;
#define CFA_L2_FILTER_CFG_REQ_ENABLES_DST_ID 0x1UL
@@ -3337,6 +3683,41 @@ struct hwrm_fw_reset_output {
u8 valid;
};
+/* hwrm_fw_qstatus */
+/* Input (24 bytes) */
+struct hwrm_fw_qstatus_input {
+ __le16 req_type;
+ __le16 cmpl_ring;
+ __le16 seq_id;
+ __le16 target_id;
+ __le64 resp_addr;
+ u8 embedded_proc_type;
+ #define FW_QSTATUS_REQ_EMBEDDED_PROC_TYPE_BOOT (0x0UL << 0)
+ #define FW_QSTATUS_REQ_EMBEDDED_PROC_TYPE_MGMT (0x1UL << 0)
+ #define FW_QSTATUS_REQ_EMBEDDED_PROC_TYPE_NETCTRL (0x2UL << 0)
+ #define FW_QSTATUS_REQ_EMBEDDED_PROC_TYPE_ROCE (0x3UL << 0)
+ #define FW_QSTATUS_REQ_EMBEDDED_PROC_TYPE_RSVD (0x4UL << 0)
+ u8 unused_0[7];
+};
+
+/* Output (16 bytes) */
+struct hwrm_fw_qstatus_output {
+ __le16 error_code;
+ __le16 req_type;
+ __le16 seq_id;
+ __le16 resp_len;
+ u8 selfrst_status;
+ #define FW_QSTATUS_RESP_SELFRST_STATUS_SELFRSTNONE (0x0UL << 0)
+ #define FW_QSTATUS_RESP_SELFRST_STATUS_SELFRSTASAP (0x1UL << 0)
+ #define FW_QSTATUS_RESP_SELFRST_STATUS_SELFRSTPCIERST (0x2UL << 0)
+ u8 unused_0;
+ __le16 unused_1;
+ u8 unused_2;
+ u8 unused_3;
+ u8 unused_4;
+ u8 valid;
+};
+
/* hwrm_exec_fwd_resp */
/* Input (128 bytes) */
struct hwrm_exec_fwd_resp_input {
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_nvm_defs.h b/drivers/net/ethernet/broadcom/bnxt/bnxt_nvm_defs.h
index 43ef392..40a7b0e 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt_nvm_defs.h
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_nvm_defs.h
@@ -1,6 +1,6 @@
/* Broadcom NetXtreme-C/E network driver.
*
- * Copyright (c) 2014-2015 Broadcom Corporation
+ * Copyright (c) 2014-2016 Broadcom Corporation
*
* This program is free software; you can redistribute it and/or modify
* it under the terms of the GNU General Public License as published by
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_sriov.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_sriov.c
index 0c5f510..8457850 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt_sriov.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_sriov.c
@@ -1,6 +1,6 @@
/* Broadcom NetXtreme-C/E network driver.
*
- * Copyright (c) 2014-2015 Broadcom Corporation
+ * Copyright (c) 2014-2016 Broadcom Corporation
*
* This program is free software; you can redistribute it and/or modify
* it under the terms of the GNU General Public License as published by
@@ -771,12 +771,8 @@ static int bnxt_vf_set_link(struct bnxt *bp, struct bnxt_vf_info *vf)
PORT_PHY_QCFG_RESP_LINK_NO_LINK) {
phy_qcfg_resp.link =
PORT_PHY_QCFG_RESP_LINK_LINK;
- if (phy_qcfg_resp.auto_link_speed)
- phy_qcfg_resp.link_speed =
- phy_qcfg_resp.auto_link_speed;
- else
- phy_qcfg_resp.link_speed =
- phy_qcfg_resp.force_link_speed;
+ phy_qcfg_resp.link_speed = cpu_to_le16(
+ PORT_PHY_QCFG_RESP_LINK_SPEED_10GB);
phy_qcfg_resp.duplex =
PORT_PHY_QCFG_RESP_DUPLEX_FULL;
phy_qcfg_resp.pause =
@@ -859,8 +855,8 @@ void bnxt_update_vf_mac(struct bnxt *bp)
* default but the stored zero MAC will allow the VF user to change
* the random MAC address using ndo_set_mac_address() if he wants.
*/
- if (!ether_addr_equal(resp->perm_mac_address, bp->vf.mac_addr))
- memcpy(bp->vf.mac_addr, resp->perm_mac_address, ETH_ALEN);
+ if (!ether_addr_equal(resp->mac_address, bp->vf.mac_addr))
+ memcpy(bp->vf.mac_addr, resp->mac_address, ETH_ALEN);
/* overwrite netdev dev_addr with admin VF MAC */
if (is_valid_ether_addr(bp->vf.mac_addr))
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_sriov.h b/drivers/net/ethernet/broadcom/bnxt/bnxt_sriov.h
index c151280..3f08354 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt_sriov.h
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_sriov.h
@@ -1,6 +1,6 @@
/* Broadcom NetXtreme-C/E network driver.
*
- * Copyright (c) 2014-2015 Broadcom Corporation
+ * Copyright (c) 2014-2016 Broadcom Corporation
*
* This program is free software; you can redistribute it and/or modify
* it under the terms of the GNU General Public License as published by
--
1.8.3.1
^ permalink raw reply related
* [PATCH net-next v2 0/9] bnxt_en: Updates for net-next.
From: Michael Chan @ 2016-04-05 18:08 UTC (permalink / raw)
To: davem; +Cc: netdev
Update to latest firmware interface, add EEE feature, unsupported SFP+
module warning, and ethtool -s improvements.
v2: Removed the GEEPROM patch and added more comments to the get_eee patch.
Michael Chan (9):
bnxt_en: Update to Firmware 1.2.2 spec.
bnxt_en: Improve flow control autoneg with Firmware 1.2.1 interface.
bnxt_en: Add basic EEE support.
bnxt_en: Add EEE setup code.
bnxt_en: Add get_eee() and set_eee() ethtool support.
bnxt_en: Set async event bits when registering with the firmware.
bnxt_en: Add unsupported SFP+ module warnings.
bnxt_en: Check for valid forced speed during ethtool -s.
bnxt_en: Improve ethtool .get_settings().
drivers/net/ethernet/broadcom/bnxt/bnxt.c | 259 ++++++++++--
drivers/net/ethernet/broadcom/bnxt/bnxt.h | 17 +-
drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c | 165 ++++++--
drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.h | 4 +-
drivers/net/ethernet/broadcom/bnxt/bnxt_fw_hdr.h | 2 +-
drivers/net/ethernet/broadcom/bnxt/bnxt_hsi.h | 433 +++++++++++++++++++--
drivers/net/ethernet/broadcom/bnxt/bnxt_nvm_defs.h | 2 +-
drivers/net/ethernet/broadcom/bnxt/bnxt_sriov.c | 14 +-
drivers/net/ethernet/broadcom/bnxt/bnxt_sriov.h | 2 +-
9 files changed, 803 insertions(+), 95 deletions(-)
--
1.8.3.1
^ permalink raw reply
* [PATCH net-next v2 4/9] bnxt_en: Add EEE setup code.
From: Michael Chan @ 2016-04-05 18:08 UTC (permalink / raw)
To: davem; +Cc: netdev
In-Reply-To: <1459879743-16960-1-git-send-email-michael.chan@broadcom.com>
1. Add bnxt_hwrm_set_eee() function to setup EEE firmware parameters based
on the bp->eee settings.
2. The new function bnxt_eee_config_ok() will check if EEE parameters need
to be modified due to autoneg changes.
3. bnxt_hwrm_set_link() has added a new parameter to update EEE. If the
parameter is set, it will call bnxt_hwrm_set_eee().
Signed-off-by: Michael Chan <michael.chan@broadcom.com>
---
drivers/net/ethernet/broadcom/bnxt/bnxt.c | 58 ++++++++++++++++++++++-
drivers/net/ethernet/broadcom/bnxt/bnxt.h | 2 +-
drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c | 4 +-
drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.h | 1 +
4 files changed, 60 insertions(+), 5 deletions(-)
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
index 7442e20..2c3c795 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
@@ -4711,7 +4711,30 @@ int bnxt_hwrm_set_pause(struct bnxt *bp)
return rc;
}
-int bnxt_hwrm_set_link_setting(struct bnxt *bp, bool set_pause)
+static void bnxt_hwrm_set_eee(struct bnxt *bp,
+ struct hwrm_port_phy_cfg_input *req)
+{
+ struct ethtool_eee *eee = &bp->eee;
+
+ if (eee->eee_enabled) {
+ u16 eee_speeds;
+ u32 flags = PORT_PHY_CFG_REQ_FLAGS_EEE_ENABLE;
+
+ if (eee->tx_lpi_enabled)
+ flags |= PORT_PHY_CFG_REQ_FLAGS_EEE_TX_LPI_ENABLE;
+ else
+ flags |= PORT_PHY_CFG_REQ_FLAGS_EEE_TX_LPI_DISABLE;
+
+ req->flags |= cpu_to_le32(flags);
+ eee_speeds = bnxt_get_fw_auto_link_speeds(eee->advertised);
+ req->eee_link_speed_mask = cpu_to_le16(eee_speeds);
+ req->tx_lpi_timer = cpu_to_le32(eee->tx_lpi_timer);
+ } else {
+ req->flags |= cpu_to_le32(PORT_PHY_CFG_REQ_FLAGS_EEE_DISABLE);
+ }
+}
+
+int bnxt_hwrm_set_link_setting(struct bnxt *bp, bool set_pause, bool set_eee)
{
struct hwrm_port_phy_cfg_input req = {0};
@@ -4720,14 +4743,42 @@ int bnxt_hwrm_set_link_setting(struct bnxt *bp, bool set_pause)
bnxt_hwrm_set_pause_common(bp, &req);
bnxt_hwrm_set_link_common(bp, &req);
+
+ if (set_eee)
+ bnxt_hwrm_set_eee(bp, &req);
return hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT);
}
+static bool bnxt_eee_config_ok(struct bnxt *bp)
+{
+ struct ethtool_eee *eee = &bp->eee;
+ struct bnxt_link_info *link_info = &bp->link_info;
+
+ if (!(bp->flags & BNXT_FLAG_EEE_CAP))
+ return true;
+
+ if (eee->eee_enabled) {
+ u32 advertising =
+ _bnxt_fw_to_ethtool_adv_spds(link_info->advertising, 0);
+
+ if (!(link_info->autoneg & BNXT_AUTONEG_SPEED)) {
+ eee->eee_enabled = 0;
+ return false;
+ }
+ if (eee->advertised & ~advertising) {
+ eee->advertised = advertising & eee->supported;
+ return false;
+ }
+ }
+ return true;
+}
+
static int bnxt_update_phy_setting(struct bnxt *bp)
{
int rc;
bool update_link = false;
bool update_pause = false;
+ bool update_eee = false;
struct bnxt_link_info *link_info = &bp->link_info;
rc = bnxt_update_link(bp, true);
@@ -4757,8 +4808,11 @@ static int bnxt_update_phy_setting(struct bnxt *bp)
update_link = true;
}
+ if (!bnxt_eee_config_ok(bp))
+ update_eee = true;
+
if (update_link)
- rc = bnxt_hwrm_set_link_setting(bp, update_pause);
+ rc = bnxt_hwrm_set_link_setting(bp, update_pause, update_eee);
else if (update_pause)
rc = bnxt_hwrm_set_pause(bp);
if (rc) {
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.h b/drivers/net/ethernet/broadcom/bnxt/bnxt.h
index 5e83405..a981e2c 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.h
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.h
@@ -1112,7 +1112,7 @@ int hwrm_send_message_silent(struct bnxt *, void *, u32, int);
int bnxt_hwrm_set_coal(struct bnxt *);
int bnxt_hwrm_func_qcaps(struct bnxt *);
int bnxt_hwrm_set_pause(struct bnxt *);
-int bnxt_hwrm_set_link_setting(struct bnxt *, bool);
+int bnxt_hwrm_set_link_setting(struct bnxt *, bool, bool);
int bnxt_open_nic(struct bnxt *, bool, bool);
int bnxt_close_nic(struct bnxt *, bool, bool);
int bnxt_get_max_rings(struct bnxt *, int *, int *, bool);
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c
index bdc6220..14f0520 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c
@@ -763,7 +763,7 @@ static u32 bnxt_get_fw_speed(struct net_device *dev, u16 ethtool_speed)
return 0;
}
-static u16 bnxt_get_fw_auto_link_speeds(u32 advertising)
+u16 bnxt_get_fw_auto_link_speeds(u32 advertising)
{
u16 fw_speed_mask = 0;
@@ -840,7 +840,7 @@ static int bnxt_set_settings(struct net_device *dev, struct ethtool_cmd *cmd)
}
if (netif_running(dev))
- rc = bnxt_hwrm_set_link_setting(bp, set_pause);
+ rc = bnxt_hwrm_set_link_setting(bp, set_pause, false);
set_setting_exit:
return rc;
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.h b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.h
index e061f8f..3abc03b 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.h
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.h
@@ -14,5 +14,6 @@ extern const struct ethtool_ops bnxt_ethtool_ops;
u32 _bnxt_fw_to_ethtool_adv_spds(u16, u8);
u32 bnxt_fw_to_ethtool_speed(u16);
+u16 bnxt_get_fw_auto_link_speeds(u32);
#endif
--
1.8.3.1
^ permalink raw reply related
* [PATCH net-next v2 2/9] bnxt_en: Improve flow control autoneg with Firmware 1.2.1 interface.
From: Michael Chan @ 2016-04-05 18:08 UTC (permalink / raw)
To: davem; +Cc: netdev
In-Reply-To: <1459879743-16960-1-git-send-email-michael.chan@broadcom.com>
Make use of the new AUTONEG_PAUSE bit in the new interface to better
control autoneg flow control settings, independent of RX and TX
advertisement settings.
Signed-off-by: Michael Chan <michael.chan@broadcom.com>
---
drivers/net/ethernet/broadcom/bnxt/bnxt.c | 28 +++++++++++++++++++----
drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c | 10 ++++----
2 files changed, 27 insertions(+), 11 deletions(-)
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
index bfe98cb..2b5a541 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
@@ -4557,6 +4557,9 @@ static void
bnxt_hwrm_set_pause_common(struct bnxt *bp, struct hwrm_port_phy_cfg_input *req)
{
if (bp->link_info.autoneg & BNXT_AUTONEG_FLOW_CTRL) {
+ if (bp->hwrm_spec_code >= 0x10201)
+ req->auto_pause =
+ PORT_PHY_CFG_REQ_AUTO_PAUSE_AUTONEG_PAUSE;
if (bp->link_info.req_flow_ctrl & BNXT_LINK_PAUSE_RX)
req->auto_pause |= PORT_PHY_CFG_REQ_AUTO_PAUSE_RX;
if (bp->link_info.req_flow_ctrl & BNXT_LINK_PAUSE_TX)
@@ -4570,6 +4573,11 @@ bnxt_hwrm_set_pause_common(struct bnxt *bp, struct hwrm_port_phy_cfg_input *req)
req->force_pause |= PORT_PHY_CFG_REQ_FORCE_PAUSE_TX;
req->enables |=
cpu_to_le32(PORT_PHY_CFG_REQ_ENABLES_FORCE_PAUSE);
+ if (bp->hwrm_spec_code >= 0x10201) {
+ req->auto_pause = req->force_pause;
+ req->enables |= cpu_to_le32(
+ PORT_PHY_CFG_REQ_ENABLES_AUTO_PAUSE);
+ }
}
}
@@ -4656,7 +4664,8 @@ static int bnxt_update_phy_setting(struct bnxt *bp)
return rc;
}
if ((link_info->autoneg & BNXT_AUTONEG_FLOW_CTRL) &&
- link_info->auto_pause_setting != link_info->req_flow_ctrl)
+ (link_info->auto_pause_setting & BNXT_LINK_PAUSE_BOTH) !=
+ link_info->req_flow_ctrl)
update_pause = true;
if (!(link_info->autoneg & BNXT_AUTONEG_FLOW_CTRL) &&
link_info->force_pause_setting != link_info->req_flow_ctrl)
@@ -5825,15 +5834,24 @@ static int bnxt_probe_phy(struct bnxt *bp)
/*initialize the ethool setting copy with NVM settings */
if (BNXT_AUTO_MODE(link_info->auto_mode)) {
- link_info->autoneg = BNXT_AUTONEG_SPEED |
- BNXT_AUTONEG_FLOW_CTRL;
+ link_info->autoneg = BNXT_AUTONEG_SPEED;
+ if (bp->hwrm_spec_code >= 0x10201) {
+ if (link_info->auto_pause_setting &
+ PORT_PHY_CFG_REQ_AUTO_PAUSE_AUTONEG_PAUSE)
+ link_info->autoneg |= BNXT_AUTONEG_FLOW_CTRL;
+ } else {
+ link_info->autoneg |= BNXT_AUTONEG_FLOW_CTRL;
+ }
link_info->advertising = link_info->auto_link_speeds;
- link_info->req_flow_ctrl = link_info->auto_pause_setting;
} else {
link_info->req_link_speed = link_info->force_link_speed;
link_info->req_duplex = link_info->duplex_setting;
- link_info->req_flow_ctrl = link_info->force_pause_setting;
}
+ if (link_info->autoneg & BNXT_AUTONEG_FLOW_CTRL)
+ link_info->req_flow_ctrl =
+ link_info->auto_pause_setting & BNXT_LINK_PAUSE_BOTH;
+ else
+ link_info->req_flow_ctrl = link_info->force_pause_setting;
return rc;
}
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c
index f103f9b..99b1740 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c
@@ -874,7 +874,9 @@ static int bnxt_set_pauseparam(struct net_device *dev,
return -EINVAL;
link_info->autoneg |= BNXT_AUTONEG_FLOW_CTRL;
- link_info->req_flow_ctrl |= BNXT_LINK_PAUSE_BOTH;
+ if (bp->hwrm_spec_code >= 0x10201)
+ link_info->req_flow_ctrl =
+ PORT_PHY_CFG_REQ_AUTO_PAUSE_AUTONEG_PAUSE;
} else {
/* when transition from auto pause to force pause,
* force a link change
@@ -882,17 +884,13 @@ static int bnxt_set_pauseparam(struct net_device *dev,
if (link_info->autoneg & BNXT_AUTONEG_FLOW_CTRL)
link_info->force_link_chng = true;
link_info->autoneg &= ~BNXT_AUTONEG_FLOW_CTRL;
- link_info->req_flow_ctrl &= ~BNXT_LINK_PAUSE_BOTH;
+ link_info->req_flow_ctrl = 0;
}
if (epause->rx_pause)
link_info->req_flow_ctrl |= BNXT_LINK_PAUSE_RX;
- else
- link_info->req_flow_ctrl &= ~BNXT_LINK_PAUSE_RX;
if (epause->tx_pause)
link_info->req_flow_ctrl |= BNXT_LINK_PAUSE_TX;
- else
- link_info->req_flow_ctrl &= ~BNXT_LINK_PAUSE_TX;
if (netif_running(dev))
rc = bnxt_hwrm_set_pause(bp);
--
1.8.3.1
^ permalink raw reply related
* Re: [PATCH net-next 1/8] perf: optimize perf_fetch_caller_regs
From: Alexei Starovoitov @ 2016-04-05 17:41 UTC (permalink / raw)
To: Peter Zijlstra
Cc: Steven Rostedt, David S . Miller, Ingo Molnar, Daniel Borkmann,
Arnaldo Carvalho de Melo, Wang Nan, Josef Bacik, Brendan Gregg,
netdev, linux-kernel, kernel-team
In-Reply-To: <20160405120626.GM3448@twins.programming.kicks-ass.net>
On 4/5/16 5:06 AM, Peter Zijlstra wrote:
> On Mon, Apr 04, 2016 at 09:52:47PM -0700, Alexei Starovoitov wrote:
>> avoid memset in perf_fetch_caller_regs, since it's the critical path of all tracepoints.
>> It's called from perf_sw_event_sched, perf_event_task_sched_in and all of perf_trace_##call
>> with this_cpu_ptr(&__perf_regs[..]) which are zero initialized by perpcu_alloc
>
> Its not actually allocated; but because its a static uninitialized
> variable we get .bss like behaviour and the initial value is copied to
> all CPUs when the per-cpu allocator thingy bootstraps SMP IIRC.
yes, it's .bss-like in a special section. I think static percpu still
goes through some fancy boot time init similar to dynamic.
What I tried to emphasize that either static or dynamic percpu areas
are guaranteed to be zero initialized.
>> and
>> subsequent call to perf_arch_fetch_caller_regs initializes the same fields on all archs,
>> so we can safely drop memset from all of the above cases and
>
> Indeed.
>
>> move it into
>> perf_ftrace_function_call that calls it with stack allocated pt_regs.
>
> Hmm, is there a reason that's still on-stack instead of using the
> per-cpu thing, Steve?
>
>> Signed-off-by: Alexei Starovoitov <ast@kernel.org>
>
> In any case,
>
> Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Thanks for the quick review.
^ permalink raw reply
* Re: [net PATCH v2 2/2] ipv4/GRO: Make GRO conform to RFC 6864
From: Tom Herbert @ 2016-04-05 17:38 UTC (permalink / raw)
To: Edward Cree
Cc: Herbert Xu, Alexander Duyck, Alexander Duyck, Jesse Gross,
Eric Dumazet, Netdev, David Miller
In-Reply-To: <5703F0A6.5010201@solarflare.com>
On Tue, Apr 5, 2016 at 2:06 PM, Edward Cree <ecree@solarflare.com> wrote:
> On 05/04/16 16:36, Tom Herbert wrote:
>> I thought about that some. It seems like we would want to do both GRO
>> and retain all the individual packets in the skb so that we could use
>> those for forwarding instead of GSO as I think you're saying.
> I didn't quite mean that, I meant just pass around the skb list, don't
> do GRO at all. The receiving TCP stack ends up just getting called
> several times, in quick succession, without I$ loss from network stack
> traversal in between.
>
>> This
>> would would work great in the plain forwarding case, but one problem
>> is what to do if the host modifies the super packet (for instance when
>> forwarding over a tunnel we might add encapsulation header). This
>> should work in GSO (although we need to address the limitations around
>> 1 encap level), not sure this is easy if we need to add a header to
>> each packet in a batch.
> This is indeed a problem with what I was proposing; perhaps the answer
> is that as you process these SKB lists you also update something like a
> GRO CB, then if you do decide to transform the packets you can coalesce
> them at that point. But doing 'the rest' of GRO might cost as much as
> just transforming all the packets, in which case you only win if you want
> to transform them multiple times.
> And if we assume that we're going to be touching all the headers anyway,
> it probably doesn't cost us much to transform all the packets in the list
> since our code and data are both hot in cache. Well, the code is cold
> for the first packet in the list, but equally it's cold for the
> superpacket in the current approach.
>
> If this is as performant as GRO in the normal (non-forwarding) receive
> case (and that's a *big* if, which can only be resolved by trying it), it
> might make sense to just not have GRO, while TSO only gets used for
> locally-generated sends, and for the case where you're forwarding between
> networks with different MTUs (e.g. from a 64k VM netdevice to the wire).
>
> What do you think? Am I crazy? (More than my usual?)
>
It's not clear to me how important optimizing the GRO to GSO
forwarding case really is. We definitely need GRO for locally received
packets, but how much benefit do we really get when applying GRO to
forwarding and does that really outweigh the latency we're adding in
the forwarding path to to GRO? Also, once we implement high
performance forwarding in XDP there is really is no need to consider
GRO, we'll do some sort of I/O batching for sure but that would be
batching packets to same TX queue not same destination.
Tom
> -Ed
^ permalink raw reply
* Re: [PATCH v4 net-next 01/15] nfp: correct RX buffer length calculation
From: Jakub Kicinski @ 2016-04-05 17:24 UTC (permalink / raw)
To: David Miller; +Cc: netdev
In-Reply-To: <20160405.113917.846241936652070162.davem@davemloft.net>
On Tue, 05 Apr 2016 11:39:17 -0400 (EDT), David Miller wrote:
> From: Jakub Kicinski <jakub.kicinski@netronome.com>
> Date: Fri, 1 Apr 2016 22:06:37 +0100
>
> > When calculating the RX buffer length we need to account for
> > up to 2 VLAN tags and up to 8 MPLS labels. Rounding up to 1k
> > is an relic of a distant past and can be removed. While at
> > it also remove trivial print statement.
> >
> > Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
>
> I disagree with the MPLS aspect of this change.
>
> VLAN is special, in that when the hardware supports VLAN properly, the
> VLAN header doesn't eat into the MTU and is sort of "transparent".
>
> But MPLS doesn't work that way.
>
> MPLS is in the main frame and takes up MTU space.
Makes sense. RFC3032 counts MPLS label stack as Frame Payload.
> Therefore I see no reason to increase the buffer length by 8 * MPLS
> which is just a rediculous amount of wasted space.
>
> I'm not applying this without at least some more explanations about
> why exactly you need to account for these values in the commit message.
It's just what FW guys asked me for. I'll try to find out what their
reasoning was.
I have a patch queued up which gets rid of unconditionally reserving
64B for firmware prepend, I guess that made me too excited to pay
attention to the fact the accounting for MPLS is indeed questionable...
^ permalink raw reply
* Re: [RFC PATCH 6/6] ppp: add rtnetlink device creation support
From: walter harms @ 2016-04-05 17:18 UTC (permalink / raw)
To: Guillaume Nault; +Cc: netdev, linux-ppp, Paul Mackerras, David Miller
In-Reply-To: <dc5fd88775581dba073d0051bbfb4c2c9c5c2d23.1459807527.git.g.nault@alphalink.fr>
Am 05.04.2016 02:56, schrieb Guillaume Nault:
> Define PPP device handlers for use with rtnetlink.
> The only PPP specific attribute is IFLA_PPP_DEV_FD. It is mandatory and
> contains the file descriptor of the associated /dev/ppp instance (the
> file descriptor which would have been used for ioctl(PPPIOCNEWUNIT) in
> the ioctl-based API). The PPP device is removed when this file
> descriptor is released (same behaviour as with ioctl based PPP
> devices).
>
> PPP devices created with the rtnetlink API behave like the ones created
> with ioctl(PPPIOCNEWUNIT). In particular existing ioctls work the same
> way, no matter how the PPP device was created.
>
> However, there are a few differences between rtnl and ioctl based PPP
> devices. Rtnl based PPP devices can be removed with RTM_DELLINK
> messages (e.g. with "ip link del"), while the ones created with
> ioctl(PPPIOCNEWUNIT) can't.
> The interface name is also built differently: the number following the
> "ppp" prefix corresponds to the PPP unit number for ioctl based
> devices, while it is just an unrelated incrementing index for rtnl
> ones.
>
> Signed-off-by: Guillaume Nault <g.nault@alphalink.fr>
> ---
> drivers/net/ppp/ppp_generic.c | 143 +++++++++++++++++++++++++++++++++++++-----
> include/uapi/linux/if_link.h | 8 +++
> 2 files changed, 136 insertions(+), 15 deletions(-)
>
> diff --git a/drivers/net/ppp/ppp_generic.c b/drivers/net/ppp/ppp_generic.c
> index 516f8dc..ae40368 100644
> --- a/drivers/net/ppp/ppp_generic.c
> +++ b/drivers/net/ppp/ppp_generic.c
> @@ -46,6 +46,7 @@
> #include <linux/device.h>
> #include <linux/mutex.h>
> #include <linux/slab.h>
> +#include <linux/file.h>
> #include <asm/unaligned.h>
> #include <net/slhc_vj.h>
> #include <linux/atomic.h>
> @@ -185,7 +186,9 @@ struct channel {
>
> struct ppp_config {
> struct file *file;
> + s32 fd;
> s32 unit;
> + bool ifname_is_set;
> };
>
> /*
> @@ -286,6 +289,7 @@ static int unit_get(struct idr *p, void *ptr);
> static int unit_set(struct idr *p, void *ptr, int n);
> static void unit_put(struct idr *p, int n);
> static void *unit_find(struct idr *p, int n);
> +static void ppp_setup(struct net_device *dev);
>
> static const struct net_device_ops ppp_netdev_ops;
>
> @@ -989,7 +993,7 @@ static struct pernet_operations ppp_net_ops = {
> .size = sizeof(struct ppp_net),
> };
>
> -static int ppp_unit_register(struct ppp *ppp, int unit)
> +static int ppp_unit_register(struct ppp *ppp, int unit, bool ifname_is_set)
> {
> struct ppp_net *pn = ppp_pernet(ppp->ppp_net);
> int ret;
> @@ -1019,7 +1023,8 @@ static int ppp_unit_register(struct ppp *ppp, int unit)
> }
> ppp->file.index = ret;
>
> - snprintf(ppp->dev->name, IFNAMSIZ, "ppp%i", ppp->file.index);
> + if (!ifname_is_set)
> + snprintf(ppp->dev->name, IFNAMSIZ, "ppp%i", ppp->file.index);
>
> ret = register_netdevice(ppp->dev);
> if (ret < 0)
> @@ -1043,12 +1048,39 @@ static int ppp_dev_configure(struct net *src_net, struct net_device *dev,
> const struct ppp_config *conf)
> {
> struct ppp *ppp = netdev_priv(dev);
> + struct file *file;
> int indx;
> + int err;
> +
> + if (conf->fd < 0) {
> + file = conf->file;
> + if (!file) {
> + err = -EBADF;
> + goto out;
why not just return -EBADF;
> + }
> + } else {
> + file = fget(conf->fd);
> + if (!file) {
> + err = -EBADF;
> + goto out;
why not just return -EBADF;
> + }
just my 2 cents,
re,
wh
> +
> + if (file->f_op != &ppp_device_fops) {
> + err = -EBADF;
> + goto out;
> + }
> + }
> +
> + mutex_lock(&ppp_mutex);
> + if (file->private_data) {
> + err = -ENOTTY;
> + goto out_mutex;
> + }
>
> ppp->dev = dev;
> ppp->mru = PPP_MRU;
> ppp->ppp_net = src_net;
> - ppp->owner = conf->file;
> + ppp->owner = file;
>
> init_ppp_file(&ppp->file, INTERFACE);
> ppp->file.hdrlen = PPP_HDRLEN - 2; /* don't count proto bytes */
> @@ -1067,9 +1099,88 @@ static int ppp_dev_configure(struct net *src_net, struct net_device *dev,
> ppp->active_filter = NULL;
> #endif /* CONFIG_PPP_FILTER */
>
> - return ppp_unit_register(ppp, conf->unit);
> + err = ppp_unit_register(ppp, conf->unit, conf->ifname_is_set);
> + if (err < 0)
> + goto out_mutex;
> +
> + file->private_data = &ppp->file;
> +
> +out_mutex:
> + mutex_unlock(&ppp_mutex);
> +out:
> + if (conf->fd >= 0 && file)
> + fput(file);
> +
> + return err;
> }
>
> +static const struct nla_policy ppp_nl_policy[IFLA_PPP_MAX + 1] = {
> + [IFLA_PPP_DEV_FD] = { .type = NLA_S32 },
> +};
> +
> +static int ppp_nl_validate(struct nlattr *tb[], struct nlattr *data[])
> +{
> + if (!data)
> + return -EINVAL;
> +
> + if (!data[IFLA_PPP_DEV_FD])
> + return -EINVAL;
> + if (nla_get_s32(data[IFLA_PPP_DEV_FD]) < 0)
> + return -EBADF;
> +
> + return 0;
> +}
> +
> +static int ppp_nl_newlink(struct net *src_net, struct net_device *dev,
> + struct nlattr *tb[], struct nlattr *data[])
> +{
> + struct ppp_config conf = {
> + .file = NULL,
> + .unit = -1,
> + .ifname_is_set = true,
> + };
> +
> + conf.fd = nla_get_s32(data[IFLA_PPP_DEV_FD]);
> +
> + return ppp_dev_configure(src_net, dev, &conf);
> +}
> +
> +static void ppp_nl_dellink(struct net_device *dev, struct list_head *head)
> +{
> + unregister_netdevice_queue(dev, head);
> +}
> +
> +static size_t ppp_nl_get_size(const struct net_device *dev)
> +{
> + return 0;
> +}
> +
> +static int ppp_nl_fill_info(struct sk_buff *skb, const struct net_device *dev)
> +{
> + return 0;
> +}
> +
> +static struct net *ppp_nl_get_link_net(const struct net_device *dev)
> +{
> + struct ppp *ppp = netdev_priv(dev);
> +
> + return ppp->ppp_net;
> +}
> +
> +static struct rtnl_link_ops ppp_link_ops __read_mostly = {
> + .kind = "ppp",
> + .maxtype = IFLA_PPP_MAX,
> + .policy = ppp_nl_policy,
> + .priv_size = sizeof(struct ppp),
> + .setup = ppp_setup,
> + .validate = ppp_nl_validate,
> + .newlink = ppp_nl_newlink,
> + .dellink = ppp_nl_dellink,
> + .get_size = ppp_nl_get_size,
> + .fill_info = ppp_nl_fill_info,
> + .get_link_net = ppp_nl_get_link_net,
> +};
> +
> #define PPP_MAJOR 108
>
> /* Called at boot time if ppp is compiled into the kernel,
> @@ -1098,11 +1209,19 @@ static int __init ppp_init(void)
> goto out_chrdev;
> }
>
> + err = rtnl_link_register(&ppp_link_ops);
> + if (err) {
> + pr_err("failed to register rtnetlink PPP handler\n");
> + goto out_class;
> + }
> +
> /* not a big deal if we fail here :-) */
> device_create(ppp_class, NULL, MKDEV(PPP_MAJOR, 0), NULL, "ppp");
>
> return 0;
>
> +out_class:
> + class_destroy(ppp_class);
> out_chrdev:
> unregister_chrdev(PPP_MAJOR, "ppp");
> out_net:
> @@ -2846,7 +2965,9 @@ static int ppp_create_interface(struct net *net, struct file *file, int *unit)
> {
> struct ppp_config conf = {
> .file = file,
> + .fd = -1,
> .unit = *unit,
> + .ifname_is_set = false,
> };
> struct net_device *dev;
> struct ppp *ppp;
> @@ -2861,27 +2982,17 @@ static int ppp_create_interface(struct net *net, struct file *file, int *unit)
> dev_net_set(dev, net);
>
> rtnl_lock();
> - mutex_lock(&ppp_mutex);
> - if (file->private_data) {
> - err = -ENOTTY;
> - goto err_dev;
> - }
> -
> err = ppp_dev_configure(net, dev, &conf);
> if (err < 0)
> goto err_dev;
> + rtnl_unlock();
>
> ppp = netdev_priv(dev);
> *unit = ppp->file.index;
> - file->private_data = &ppp->file;
> -
> - mutex_unlock(&ppp_mutex);
> - rtnl_unlock();
>
> return 0;
>
> err_dev:
> - mutex_unlock(&ppp_mutex);
> rtnl_unlock();
> free_netdev(dev);
> err:
> @@ -3074,6 +3185,7 @@ static void __exit ppp_cleanup(void)
> /* should never happen */
> if (atomic_read(&ppp_unit_count) || atomic_read(&channel_count))
> pr_err("PPP: removing module but units remain!\n");
> + rtnl_link_unregister(&ppp_link_ops);
> unregister_chrdev(PPP_MAJOR, "ppp");
> device_destroy(ppp_class, MKDEV(PPP_MAJOR, 0));
> class_destroy(ppp_class);
> @@ -3132,4 +3244,5 @@ EXPORT_SYMBOL(ppp_register_compressor);
> EXPORT_SYMBOL(ppp_unregister_compressor);
> MODULE_LICENSE("GPL");
> MODULE_ALIAS_CHARDEV(PPP_MAJOR, 0);
> +MODULE_ALIAS_RTNL_LINK("ppp");
> MODULE_ALIAS("devname:ppp");
> diff --git a/include/uapi/linux/if_link.h b/include/uapi/linux/if_link.h
> index c488066..f238de9 100644
> --- a/include/uapi/linux/if_link.h
> +++ b/include/uapi/linux/if_link.h
> @@ -515,6 +515,14 @@ enum {
> };
> #define IFLA_GENEVE_MAX (__IFLA_GENEVE_MAX - 1)
>
> +/* PPP section */
> +enum {
> + IFLA_PPP_UNSPEC,
> + IFLA_PPP_DEV_FD,
> + __IFLA_PPP_MAX,
> +};
> +#define IFLA_PPP_MAX (__IFLA_PPP_MAX - 1)
> +
> /* Bonding section */
>
> enum {
^ permalink raw reply
* Re: [net PATCH v2 2/2] ipv4/GRO: Make GRO conform to RFC 6864
From: Edward Cree @ 2016-04-05 17:06 UTC (permalink / raw)
To: Tom Herbert
Cc: Herbert Xu, Alexander Duyck, Alexander Duyck, Jesse Gross,
Eric Dumazet, Netdev, David Miller
In-Reply-To: <CALx6S37p7sBZ_=xDpmW90-Aqs2sgnOjcjdvYJhEGZ7Y2=6+gcg@mail.gmail.com>
On 05/04/16 16:36, Tom Herbert wrote:
> I thought about that some. It seems like we would want to do both GRO
> and retain all the individual packets in the skb so that we could use
> those for forwarding instead of GSO as I think you're saying.
I didn't quite mean that, I meant just pass around the skb list, don't
do GRO at all. The receiving TCP stack ends up just getting called
several times, in quick succession, without I$ loss from network stack
traversal in between.
> This
> would would work great in the plain forwarding case, but one problem
> is what to do if the host modifies the super packet (for instance when
> forwarding over a tunnel we might add encapsulation header). This
> should work in GSO (although we need to address the limitations around
> 1 encap level), not sure this is easy if we need to add a header to
> each packet in a batch.
This is indeed a problem with what I was proposing; perhaps the answer
is that as you process these SKB lists you also update something like a
GRO CB, then if you do decide to transform the packets you can coalesce
them at that point. But doing 'the rest' of GRO might cost as much as
just transforming all the packets, in which case you only win if you want
to transform them multiple times.
And if we assume that we're going to be touching all the headers anyway,
it probably doesn't cost us much to transform all the packets in the list
since our code and data are both hot in cache. Well, the code is cold
for the first packet in the list, but equally it's cold for the
superpacket in the current approach.
If this is as performant as GRO in the normal (non-forwarding) receive
case (and that's a *big* if, which can only be resolved by trying it), it
might make sense to just not have GRO, while TSO only gets used for
locally-generated sends, and for the case where you're forwarding between
networks with different MTUs (e.g. from a 64k VM netdevice to the wire).
What do you think? Am I crazy? (More than my usual?)
-Ed
^ permalink raw reply
* Re: [PATCH net 4/4] lib/test_bpf: Add additional BPF_ADD tests
From: Naveen N. Rao @ 2016-04-05 16:51 UTC (permalink / raw)
To: Alexei Starovoitov
Cc: linux-kernel, linuxppc-dev, netdev, Daniel Borkmann,
David S. Miller, Ananth N Mavinakayanahalli, Michael Ellerman,
Paul Mackerras
In-Reply-To: <5703E7B2.6050706@fb.com>
On 2016/04/05 09:28AM, Alexei Starovoitov wrote:
> On 4/5/16 3:02 AM, Naveen N. Rao wrote:
> >Some of these tests proved useful with the powerpc eBPF JIT port due to
> >sign-extended 16-bit immediate loads. Though some of these aspects get
> >covered in other tests, it is better to have explicit tests so as to
> >quickly tag the precise problem.
> >
> >Cc: Alexei Starovoitov <ast@fb.com>
> >Cc: Daniel Borkmann <daniel@iogearbox.net>
> >Cc: "David S. Miller" <davem@davemloft.net>
> >Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
> >Cc: Michael Ellerman <mpe@ellerman.id.au>
> >Cc: Paul Mackerras <paulus@samba.org>
> >Signed-off-by: Naveen N. Rao <naveen.n.rao@linux.vnet.ibm.com>
>
> Makes sense. Looks like ppc jit will be using quite a bit of
> available ppc instructions. Nice.
>
> I'm assuming all these new tests passed with x64 jit?
Yes, all these tests pass on x86_64.
- Naveen
^ permalink raw reply
* Re: [PATCH net 2/4] lib/test_bpf: Add tests for unsigned BPF_JGT
From: Naveen N. Rao @ 2016-04-05 16:50 UTC (permalink / raw)
To: Alexei Starovoitov
Cc: linux-kernel, linuxppc-dev, netdev, Paul Mackerras,
David S. Miller, Daniel Borkmann
In-Reply-To: <5703E5C5.5060901@fb.com>
On 2016/04/05 09:20AM, Alexei Starovoitov wrote:
> On 4/5/16 3:02 AM, Naveen N. Rao wrote:
> >Unsigned Jump-if-Greater-Than.
> >
> >Cc: Alexei Starovoitov <ast@fb.com>
> >Cc: Daniel Borkmann <daniel@iogearbox.net>
> >Cc: "David S. Miller" <davem@davemloft.net>
> >Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
> >Cc: Michael Ellerman <mpe@ellerman.id.au>
> >Cc: Paul Mackerras <paulus@samba.org>
> >Signed-off-by: Naveen N. Rao <naveen.n.rao@linux.vnet.ibm.com>
>
> I think some of the tests already cover it, but extra tests are
> always great.
> Acked-by: Alexei Starovoitov <ast@kernel.org>
>
> I think the whole set belongs in net-next.
> Next time you submit the patches please say [PATCH net-next] in subject.
> [PATCH net] is for bugfixes only.
Ah, sure. Thanks for the review!
- Naveen
^ permalink raw reply
* Re: [RFC PATCH 4/5] mlx4: add support for fast rx drop bpf program
From: Or Gerlitz @ 2016-04-05 14:15 UTC (permalink / raw)
To: Alexei Starovoitov, Eric Dumazet
Cc: Jesper Dangaard Brouer, Brenden Blanco, davem, netdev, tom,
daniel, john.fastabend, Eran Ben Elisha, Rana Shahout,
Matan Barak
In-Reply-To: <20160404185010.GD68392@ast-mbp.thefacebook.com>
On 4/4/2016 9:50 PM, Alexei Starovoitov wrote:
> On Mon, Apr 04, 2016 at 08:22:03AM -0700, Eric Dumazet wrote:
>> A single flow is able to use 40Gbit on those 40Gbit NIC, so there is not
>> a single 10GB trunk used for a given flow.
>>
>> This 14Mpps thing seems to be a queue limitation on mlx4.
> yeah, could be queueing related. Multiple cpus can send ~30Mpps of the same 64 byte packet,
> but mlx4 can only receive 14.5Mpps. Odd.
>
> Or (and other mellanox guys), what is really going on inside 40G nic?
Hi Alexei,
Not that I know everything that goes inside there, and not that if I
knew it all I could have posted that here (I heard HWs sometimes have
IP)... but, anyway, as for your questions:
ConnectX3 40Gbs NIC can receive > 10Gbs packet-worthy (14.5M) in single
ring and Mellanox
100Gbs NICs can receive > 25Gbs packet-worthy (37.5M) in single ring,
people that use DPDK (...) even see this numbers and AFAIU we now
attempt to see that in the kernel with XDP :)
I realize that we might have some issues in the mlx4 driver reporting on
HW drops. Eran (cc-ed) and Co are looking on that.
In parallel to doing so, I would suggest you to do some experiments that
might shed some more light, if on the TX side you do
$ ./pktgen_sample03_burst_single_flow.sh -i $DEV -d $IP -m $MAC -t 4
On the RX side, skip RSS and force the packets that match that traffic
pattern to go to (say) ring (==action) 0
$ ethtool -U $DEV flow-type ip4 dst-mac $MAC dst-ip $IP action 0 loc 0
to go back to RSS remove the rule
$ ethtool -U $DEV delete action 0
FWIW (not that I see how it helps you now), you can do HW drop on the RX
side with ring -1
$ ethtool -U $DEV flow-type ip4 dst-mac $MAC dst-ip $IP action -1 loc 0
Or.
^ permalink raw reply
* Re: [net PATCH v2 2/2] ipv4/GRO: Make GRO conform to RFC 6864
From: Alexander Duyck @ 2016-04-05 16:45 UTC (permalink / raw)
To: Eric Dumazet
Cc: Herbert Xu, Alexander Duyck, Tom Herbert, Jesse Gross,
Eric Dumazet, Netdev, David Miller
In-Reply-To: <1459873806.6473.358.camel@edumazet-glaptop3.roam.corp.google.com>
On Tue, Apr 5, 2016 at 9:30 AM, Eric Dumazet <eric.dumazet@gmail.com> wrote:
> On Tue, 2016-04-05 at 08:52 -0700, Alexander Duyck wrote:
>
>>
>> I disagree I think it will have to be part of the default
>> configuration. The problem is the IP ID is quickly becoming
>> meaningless. When you consider that a 40Gb/s link can wrap the IP ID
>> value nearly 50 times a second using a 1500 MTU the IP ID field should
>> just be ignored anyway because you cannot guarantee that it will be
>> unique without limiting the Tx window size. That was the whole point
>> of RFC6864. Basically the IP ID field is so small that as we push
>> into the higher speeds you cannot guarantee that the field will have
>> any meaning so for any case where you don't need to use it you
>> shouldn't because it will likely not provide enough useful data.
>
> Just because a few flows reach 40Gbit , we should remind that vast
> majority of the Internet runs with < 50Mbits flows.
>
> I prefer the argument of IPv6 not having ID ;)
Okay, maybe I'll try to use that argument more often then.. :-)
> We should do our best to keep interoperability, this is the selling
> point.
>
> And quite frankly your last patch makes perfect sense to me :
Yes. It was a compromise, though I might still have to go through and
refine it more. It might make sense for the IP header associated with
the TCP flow, but for outer headers it actually is worse because we
end up blocking several different possibilities. What I might need to
do is capture the state of the DF bit as we work a flow up through the
stack and once it is in the list of GRO SKBs use that DF bit as a flag
to indicate if we support a incrementing or fixed pattern for the
values. That way tunnels can optionally ignore the IP ID if the DF
bit is set since their values may not be as clean as that of TCP.
> The aggregation is done only if the TCP headers of consecutive packets
> matches. So who cares of IPv4 ID really ?
> This is a very minor detail. The possible gains outperform the
> theoretical 'problem'
>
> GRO already reorder flows, it never had a guarantee of being 'ínvisible'
> as Herbert claims.
I can see what he is trying to get at. I just think it is a bit too
strict on the interpretation of what values have to be maintained. My
plan going forward is to add a sysctl that will probably allow us some
wiggle room in regards to IP ID for GRO and GSO so that when it is
disabled we will not perform GSO partial nor allow for repeating IP ID
in GRO on devices that cannot get the IP ID right.
- Alex
^ permalink raw reply
* Re: [PATCH net-next 0/3] udp: support SO_PEEK_OF
From: Willem de Bruijn @ 2016-04-05 16:43 UTC (permalink / raw)
To: Network Development
Cc: David Miller, Sam Kumar, Eric Dumazet, Willem de Bruijn
In-Reply-To: <1459874301-92389-1-git-send-email-willemdebruijn.kernel@gmail.com>
> Support peeking at a non-zero offset for UDP sockets. Match the
> existing behavior on Unix datagram sockets.
>
> 1/3 makes the sk_peek_offset functions safe to use outside locks
> 2/3 removes udp headers before enqueue, to simplify offset arithmetic
> 3/3 introduces SO_PEEK_OFFSET support, with Unix socket peek semantics.
Please ignore this cover letter. I sent it to soon. It is superseded
by a v2 patch set.
^ permalink raw reply
* [PATCH net] ipv6: Count in extension headers in skb->network_header
From: Jakub Sitnicki @ 2016-04-05 16:41 UTC (permalink / raw)
To: netdev; +Cc: David S. Miller, Ji Jianwen, Hannes Frederic Sowa
When sending a UDPv6 message longer than MTU, account for the length
of fragmentable IPv6 extension headers in skb->network_header offset.
Same as we do in alloc_new_skb path in __ip6_append_data().
This ensures that later on __ip6_make_skb() will make space in
headroom for fragmentable extension headers:
/* move skb->data to ip header from ext header */
if (skb->data < skb_network_header(skb))
__skb_pull(skb, skb_network_offset(skb));
Prevents a splat due to skb_under_panic:
skbuff: skb_under_panic: text:ffffffff8143397b len:2126 put:14 \
head:ffff880005bacf50 data:ffff880005bacf4a tail:0x48 end:0xc0 dev:lo
------------[ cut here ]------------
kernel BUG at net/core/skbuff.c:104!
invalid opcode: 0000 [#1] KASAN
CPU: 0 PID: 160 Comm: reproducer Not tainted 4.6.0-rc2 #65
[...]
Call Trace:
[<ffffffff813eb7b9>] skb_push+0x79/0x80
[<ffffffff8143397b>] eth_header+0x2b/0x100
[<ffffffff8141e0d0>] neigh_resolve_output+0x210/0x310
[<ffffffff814eab77>] ip6_finish_output2+0x4a7/0x7c0
[<ffffffff814efe3a>] ip6_output+0x16a/0x280
[<ffffffff815440c1>] ip6_local_out+0xb1/0xf0
[<ffffffff814f1115>] ip6_send_skb+0x45/0xd0
[<ffffffff81518836>] udp_v6_send_skb+0x246/0x5d0
[<ffffffff8151985e>] udpv6_sendmsg+0xa6e/0x1090
[...]
Reported-by: Ji Jianwen <jiji@redhat.com>
Signed-off-by: Jakub Sitnicki <jkbs@redhat.com>
Acked-by: Hannes Frederic Sowa <hannes@stressinduktion.org>
---
Can be reproduced by sending a UDPv6 message longer than MTU when
Destination Options are present, as shown below.
Original reproducer has been developed by Ji Jianwen. Cut down
version included.
# ip link set dev lo mtu 1500
# ./reproducer 0 1024 # works
# ./reproducer 8 1024 # works
# ./reproducer 64 1024 # works
# ./reproducer 0 2048 # works
# ./reproducer 8 2048 # crash
# ./reproducer 64 2048 # crash
/* reproducer.c */
#include <sys/types.h>
#include <sys/socket.h>
#include <netdb.h>
#include <assert.h>
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
static void set_dstopts(int s, int len)
{
char *dstopts;
int r;
assert(len % 8 == 0);
dstopts = calloc(len, 1);
dstopts[1] = len / 8 - 1; /* Hdr Ext Len */
dstopts[2] = 1; /* PadN Option */
dstopts[3] = len - 4; /* Opt Data Len */
r = setsockopt(s, IPPROTO_IPV6, IPV6_DSTOPTS, dstopts, len);
if (r < 0) {
perror("setsockopt");
exit(1);
}
free(dstopts);
}
static void do_send(int s, const struct addrinfo *ai, int len)
{
struct msghdr msg;
struct iovec iov[1];
char *data;
int r;
data = malloc(len);
memset(data, 'A', len);
memset(&msg, 0, sizeof(msg));
iov[0].iov_base = data;
iov[0].iov_len = len;
msg.msg_name = ai->ai_addr;
msg.msg_namelen = ai->ai_addrlen;
msg.msg_iov = iov;
msg.msg_iovlen = 1;
msg.msg_control = 0;
msg.msg_controllen = 0;
r = sendmsg(s, &msg, 0);
if (r < 0) {
perror("sendmsg");
exit(1);
}
free(data);
}
int main(int argc, char *argv[])
{
struct addrinfo *ai = NULL;
int dstopts_len, data_len;
int r, s;
if (argc != 3) {
fprintf(stderr, "Usage: %s <dstopts-bytes> <data-bytes>\n", argv[0]);
return 1;
}
dstopts_len = atoi(argv[1]);
data_len = atoi(argv[2]);
r = getaddrinfo("::1", "12345", NULL, &ai);
assert(r == 0);
s = socket(ai->ai_family, SOCK_DGRAM, IPPROTO_UDP);
assert(s != -1);
if (dstopts_len > 0)
set_dstopts(s, dstopts_len);
do_send(s, ai, data_len);
freeaddrinfo(ai);
return 0;
}
net/ipv6/ip6_output.c | 8 ++++----
1 file changed, 4 insertions(+), 4 deletions(-)
diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c
index 9428345..bc972e7 100644
--- a/net/ipv6/ip6_output.c
+++ b/net/ipv6/ip6_output.c
@@ -1090,8 +1090,8 @@ static inline int ip6_ufo_append_data(struct sock *sk,
int getfrag(void *from, char *to, int offset, int len,
int odd, struct sk_buff *skb),
void *from, int length, int hh_len, int fragheaderlen,
- int transhdrlen, int mtu, unsigned int flags,
- const struct flowi6 *fl6)
+ int exthdrlen, int transhdrlen, int mtu,
+ unsigned int flags, const struct flowi6 *fl6)
{
struct sk_buff *skb;
@@ -1116,7 +1116,7 @@ static inline int ip6_ufo_append_data(struct sock *sk,
skb_put(skb, fragheaderlen + transhdrlen);
/* initialize network header pointer */
- skb_reset_network_header(skb);
+ skb_set_network_header(skb, exthdrlen);
/* initialize protocol header pointer */
skb->transport_header = skb->network_header + fragheaderlen;
@@ -1358,7 +1358,7 @@ emsgsize:
(rt->dst.dev->features & NETIF_F_UFO) &&
(sk->sk_type == SOCK_DGRAM) && !udp_get_no_check6_tx(sk)) {
err = ip6_ufo_append_data(sk, queue, getfrag, from, length,
- hh_len, fragheaderlen,
+ hh_len, fragheaderlen, exthdrlen,
transhdrlen, mtu, flags, fl6);
if (err)
goto error;
--
2.5.5
^ permalink raw reply related
* [PATCH net-next v2 3/3] udp: enable MSG_PEEK at non-zero offset
From: Willem de Bruijn @ 2016-04-05 16:41 UTC (permalink / raw)
To: netdev; +Cc: davem, samanthakumar, edumazet, willemb
In-Reply-To: <1459874476-92838-1-git-send-email-willemdebruijn.kernel@gmail.com>
From: samanthakumar <samanthakumar@google.com>
Enable peeking at UDP datagrams at the offset specified with socket
option SOL_SOCKET/SO_PEEK_OFF. Peek at any datagram in the queue, up
to the end of the given datagram.
Implement the SO_PEEK_OFF semantics introduced in commit ef64a54f6e55
("sock: Introduce the SO_PEEK_OFF sock option"). Increase the offset
on peek, decrease it on regular reads.
When peeking, always checksum the packet immediately, to avoid
recomputation on subsequent peeks and final read.
The socket lock is not held for the duration of udp_recvmsg, so
peek and read operations can run concurrently. Only the last store
to sk_peek_off is preserved.
Signed-off-by: Sam Kumar <samanthakumar@google.com>
Signed-off-by: Willem de Bruijn <willemb@google.com>
---
include/linux/skbuff.h | 7 ++++++-
include/net/sock.h | 2 ++
net/core/datagram.c | 9 ++++++---
net/core/sock.c | 9 +++++++++
net/ipv4/af_inet.c | 1 +
net/ipv4/udp.c | 22 +++++++++++-----------
net/ipv6/af_inet6.c | 1 +
net/ipv6/udp.c | 22 +++++++++++-----------
8 files changed, 47 insertions(+), 26 deletions(-)
diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h
index 15d0df9..0073812 100644
--- a/include/linux/skbuff.h
+++ b/include/linux/skbuff.h
@@ -2949,7 +2949,12 @@ int skb_copy_datagram_from_iter(struct sk_buff *skb, int offset,
struct iov_iter *from, int len);
int zerocopy_sg_from_iter(struct sk_buff *skb, struct iov_iter *frm);
void skb_free_datagram(struct sock *sk, struct sk_buff *skb);
-void skb_free_datagram_locked(struct sock *sk, struct sk_buff *skb);
+void __skb_free_datagram_locked(struct sock *sk, struct sk_buff *skb, int len);
+static inline void skb_free_datagram_locked(struct sock *sk,
+ struct sk_buff *skb)
+{
+ __skb_free_datagram_locked(sk, skb, 0);
+}
int skb_kill_datagram(struct sock *sk, struct sk_buff *skb, unsigned int flags);
int skb_copy_bits(const struct sk_buff *skb, int offset, void *to, int len);
int skb_store_bits(struct sk_buff *skb, int offset, const void *from, int len);
diff --git a/include/net/sock.h b/include/net/sock.h
index b759989..1decb7a 100644
--- a/include/net/sock.h
+++ b/include/net/sock.h
@@ -457,6 +457,8 @@ struct sock {
#define SK_CAN_REUSE 1
#define SK_FORCE_REUSE 2
+int sk_set_peek_off(struct sock *sk, int val);
+
static inline int sk_peek_offset(struct sock *sk, int flags)
{
if (unlikely(flags & MSG_PEEK)) {
diff --git a/net/core/datagram.c b/net/core/datagram.c
index fa9dc64..b7de71f 100644
--- a/net/core/datagram.c
+++ b/net/core/datagram.c
@@ -301,16 +301,19 @@ void skb_free_datagram(struct sock *sk, struct sk_buff *skb)
}
EXPORT_SYMBOL(skb_free_datagram);
-void skb_free_datagram_locked(struct sock *sk, struct sk_buff *skb)
+void __skb_free_datagram_locked(struct sock *sk, struct sk_buff *skb, int len)
{
bool slow;
if (likely(atomic_read(&skb->users) == 1))
smp_rmb();
- else if (likely(!atomic_dec_and_test(&skb->users)))
+ else if (likely(!atomic_dec_and_test(&skb->users))) {
+ sk_peek_offset_bwd(sk, len);
return;
+ }
slow = lock_sock_fast(sk);
+ sk_peek_offset_bwd(sk, len);
skb_orphan(skb);
sk_mem_reclaim_partial(sk);
unlock_sock_fast(sk, slow);
@@ -318,7 +321,7 @@ void skb_free_datagram_locked(struct sock *sk, struct sk_buff *skb)
/* skb is now orphaned, can be freed outside of locked section */
__kfree_skb(skb);
}
-EXPORT_SYMBOL(skb_free_datagram_locked);
+EXPORT_SYMBOL(__skb_free_datagram_locked);
/**
* skb_kill_datagram - Free a datagram skbuff forcibly
diff --git a/net/core/sock.c b/net/core/sock.c
index e12197b..2ce76e8 100644
--- a/net/core/sock.c
+++ b/net/core/sock.c
@@ -2187,6 +2187,15 @@ void __sk_mem_reclaim(struct sock *sk, int amount)
}
EXPORT_SYMBOL(__sk_mem_reclaim);
+int sk_set_peek_off(struct sock *sk, int val)
+{
+ if (val < 0)
+ return -EINVAL;
+
+ sk->sk_peek_off = val;
+ return 0;
+}
+EXPORT_SYMBOL_GPL(sk_set_peek_off);
/*
* Set of default routines for initialising struct proto_ops when
diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c
index 9e48199..a38b991 100644
--- a/net/ipv4/af_inet.c
+++ b/net/ipv4/af_inet.c
@@ -948,6 +948,7 @@ const struct proto_ops inet_dgram_ops = {
.recvmsg = inet_recvmsg,
.mmap = sock_no_mmap,
.sendpage = inet_sendpage,
+ .set_peek_off = sk_set_peek_off,
#ifdef CONFIG_COMPAT
.compat_setsockopt = compat_sock_common_setsockopt,
.compat_getsockopt = compat_sock_common_getsockopt,
diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c
index cf747e8..d80312d 100644
--- a/net/ipv4/udp.c
+++ b/net/ipv4/udp.c
@@ -1294,7 +1294,7 @@ int udp_recvmsg(struct sock *sk, struct msghdr *msg, size_t len, int noblock,
DECLARE_SOCKADDR(struct sockaddr_in *, sin, msg->msg_name);
struct sk_buff *skb;
unsigned int ulen, copied;
- int peeked, off = 0;
+ int peeked, peeking, off;
int err;
int is_udplite = IS_UDPLITE(sk);
bool checksum_valid = false;
@@ -1304,15 +1304,16 @@ int udp_recvmsg(struct sock *sk, struct msghdr *msg, size_t len, int noblock,
return ip_recv_error(sk, msg, len, addr_len);
try_again:
+ peeking = off = sk_peek_offset(sk, flags);
skb = __skb_recv_datagram(sk, flags | (noblock ? MSG_DONTWAIT : 0),
&peeked, &off, &err);
if (!skb)
- goto out;
+ return err;
ulen = skb->len;
copied = len;
- if (copied > ulen)
- copied = ulen;
+ if (copied > ulen - off)
+ copied = ulen - off;
else if (copied < ulen)
msg->msg_flags |= MSG_TRUNC;
@@ -1322,16 +1323,16 @@ try_again:
* coverage checksum (UDP-Lite), do it before the copy.
*/
- if (copied < ulen || UDP_SKB_CB(skb)->partial_cov) {
+ if (copied < ulen || UDP_SKB_CB(skb)->partial_cov || peeking) {
checksum_valid = !udp_lib_checksum_complete(skb);
if (!checksum_valid)
goto csum_copy_err;
}
if (checksum_valid || skb_csum_unnecessary(skb))
- err = skb_copy_datagram_msg(skb, 0, msg, copied);
+ err = skb_copy_datagram_msg(skb, off, msg, copied);
else {
- err = skb_copy_and_csum_datagram_msg(skb, 0, msg);
+ err = skb_copy_and_csum_datagram_msg(skb, off, msg);
if (err == -EINVAL)
goto csum_copy_err;
@@ -1344,7 +1345,8 @@ try_again:
UDP_INC_STATS_USER(sock_net(sk),
UDP_MIB_INERRORS, is_udplite);
}
- goto out_free;
+ skb_free_datagram_locked(sk, skb);
+ return err;
}
if (!peeked)
@@ -1368,9 +1370,7 @@ try_again:
if (flags & MSG_TRUNC)
err = ulen;
-out_free:
- skb_free_datagram_locked(sk, skb);
-out:
+ __skb_free_datagram_locked(sk, skb, peeking ? -err : err);
return err;
csum_copy_err:
diff --git a/net/ipv6/af_inet6.c b/net/ipv6/af_inet6.c
index b11c37c..2b78aad 100644
--- a/net/ipv6/af_inet6.c
+++ b/net/ipv6/af_inet6.c
@@ -561,6 +561,7 @@ const struct proto_ops inet6_dgram_ops = {
.recvmsg = inet_recvmsg, /* ok */
.mmap = sock_no_mmap,
.sendpage = sock_no_sendpage,
+ .set_peek_off = sk_set_peek_off,
#ifdef CONFIG_COMPAT
.compat_setsockopt = compat_sock_common_setsockopt,
.compat_getsockopt = compat_sock_common_getsockopt,
diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c
index 84c8d7b..87bd7af 100644
--- a/net/ipv6/udp.c
+++ b/net/ipv6/udp.c
@@ -357,7 +357,7 @@ int udpv6_recvmsg(struct sock *sk, struct msghdr *msg, size_t len,
struct inet_sock *inet = inet_sk(sk);
struct sk_buff *skb;
unsigned int ulen, copied;
- int peeked, off = 0;
+ int peeked, peeking, off;
int err;
int is_udplite = IS_UDPLITE(sk);
bool checksum_valid = false;
@@ -371,15 +371,16 @@ int udpv6_recvmsg(struct sock *sk, struct msghdr *msg, size_t len,
return ipv6_recv_rxpmtu(sk, msg, len, addr_len);
try_again:
+ peeking = off = sk_peek_offset(sk, flags);
skb = __skb_recv_datagram(sk, flags | (noblock ? MSG_DONTWAIT : 0),
&peeked, &off, &err);
if (!skb)
- goto out;
+ return err;
ulen = skb->len;
copied = len;
- if (copied > ulen)
- copied = ulen;
+ if (copied > ulen - off)
+ copied = ulen - off;
else if (copied < ulen)
msg->msg_flags |= MSG_TRUNC;
@@ -391,16 +392,16 @@ try_again:
* coverage checksum (UDP-Lite), do it before the copy.
*/
- if (copied < ulen || UDP_SKB_CB(skb)->partial_cov) {
+ if (copied < ulen || UDP_SKB_CB(skb)->partial_cov || peeking) {
checksum_valid = !udp_lib_checksum_complete(skb);
if (!checksum_valid)
goto csum_copy_err;
}
if (checksum_valid || skb_csum_unnecessary(skb))
- err = skb_copy_datagram_msg(skb, 0, msg, copied);
+ err = skb_copy_datagram_msg(skb, off, msg, copied);
else {
- err = skb_copy_and_csum_datagram_msg(skb, 0, msg);
+ err = skb_copy_and_csum_datagram_msg(skb, off, msg);
if (err == -EINVAL)
goto csum_copy_err;
}
@@ -417,7 +418,8 @@ try_again:
UDP_MIB_INERRORS,
is_udplite);
}
- goto out_free;
+ skb_free_datagram_locked(sk, skb);
+ return err;
}
if (!peeked) {
if (is_udp4)
@@ -465,9 +467,7 @@ try_again:
if (flags & MSG_TRUNC)
err = ulen;
-out_free:
- skb_free_datagram_locked(sk, skb);
-out:
+ __skb_free_datagram_locked(sk, skb, peeking ? -err : err);
return err;
csum_copy_err:
--
2.8.0.rc3.226.g39d4020
^ permalink raw reply related
* [PATCH net-next v2 2/3] udp: remove headers from UDP packets before queueing
From: Willem de Bruijn @ 2016-04-05 16:41 UTC (permalink / raw)
To: netdev; +Cc: davem, samanthakumar, edumazet, willemb
In-Reply-To: <1459874476-92838-1-git-send-email-willemdebruijn.kernel@gmail.com>
From: samanthakumar <samanthakumar@google.com>
Remove UDP transport headers before queueing packets for reception.
This change simplifies a follow-up patch to add MSG_PEEK support.
Signed-off-by: Sam Kumar <samanthakumar@google.com>
Signed-off-by: Willem de Bruijn <willemb@google.com>
---
include/net/sock.h | 1 +
include/net/udp.h | 9 +++++++++
net/core/sock.c | 19 +++++++++++++------
net/ipv4/udp.c | 20 +++++++++++---------
net/ipv6/udp.c | 12 +++++++-----
5 files changed, 41 insertions(+), 20 deletions(-)
diff --git a/include/net/sock.h b/include/net/sock.h
index 09aec75..b759989 100644
--- a/include/net/sock.h
+++ b/include/net/sock.h
@@ -1864,6 +1864,7 @@ void sk_reset_timer(struct sock *sk, struct timer_list *timer,
void sk_stop_timer(struct sock *sk, struct timer_list *timer);
+int __sock_queue_rcv_skb(struct sock *sk, struct sk_buff *skb);
int sock_queue_rcv_skb(struct sock *sk, struct sk_buff *skb);
int sock_queue_err_skb(struct sock *sk, struct sk_buff *skb);
diff --git a/include/net/udp.h b/include/net/udp.h
index d870ec1..a0b0da9 100644
--- a/include/net/udp.h
+++ b/include/net/udp.h
@@ -158,6 +158,15 @@ static inline __sum16 udp_v4_check(int len, __be32 saddr,
void udp_set_csum(bool nocheck, struct sk_buff *skb,
__be32 saddr, __be32 daddr, int len);
+static inline void udp_csum_pull_header(struct sk_buff *skb)
+{
+ if (skb->ip_summed == CHECKSUM_NONE)
+ skb->csum = csum_partial(udp_hdr(skb), sizeof(struct udphdr),
+ skb->csum);
+ skb_pull_rcsum(skb, sizeof(struct udphdr));
+ UDP_SKB_CB(skb)->cscov -= sizeof(struct udphdr);
+}
+
struct sk_buff **udp_gro_receive(struct sk_buff **head, struct sk_buff *skb,
struct udphdr *uh);
int udp_gro_complete(struct sk_buff *skb, int nhoff);
diff --git a/net/core/sock.c b/net/core/sock.c
index 2f517ea..e12197b 100644
--- a/net/core/sock.c
+++ b/net/core/sock.c
@@ -402,9 +402,8 @@ static void sock_disable_timestamp(struct sock *sk, unsigned long flags)
}
-int sock_queue_rcv_skb(struct sock *sk, struct sk_buff *skb)
+int __sock_queue_rcv_skb(struct sock *sk, struct sk_buff *skb)
{
- int err;
unsigned long flags;
struct sk_buff_head *list = &sk->sk_receive_queue;
@@ -414,10 +413,6 @@ int sock_queue_rcv_skb(struct sock *sk, struct sk_buff *skb)
return -ENOMEM;
}
- err = sk_filter(sk, skb);
- if (err)
- return err;
-
if (!sk_rmem_schedule(sk, skb, skb->truesize)) {
atomic_inc(&sk->sk_drops);
return -ENOBUFS;
@@ -440,6 +435,18 @@ int sock_queue_rcv_skb(struct sock *sk, struct sk_buff *skb)
sk->sk_data_ready(sk);
return 0;
}
+EXPORT_SYMBOL(__sock_queue_rcv_skb);
+
+int sock_queue_rcv_skb(struct sock *sk, struct sk_buff *skb)
+{
+ int err;
+
+ err = sk_filter(sk, skb);
+ if (err)
+ return err;
+
+ return __sock_queue_rcv_skb(sk, skb);
+}
EXPORT_SYMBOL(sock_queue_rcv_skb);
int sk_receive_skb(struct sock *sk, struct sk_buff *skb, const int nested)
diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c
index 355bdb2..cf747e8 100644
--- a/net/ipv4/udp.c
+++ b/net/ipv4/udp.c
@@ -1309,7 +1309,7 @@ try_again:
if (!skb)
goto out;
- ulen = skb->len - sizeof(struct udphdr);
+ ulen = skb->len;
copied = len;
if (copied > ulen)
copied = ulen;
@@ -1329,11 +1329,9 @@ try_again:
}
if (checksum_valid || skb_csum_unnecessary(skb))
- err = skb_copy_datagram_msg(skb, sizeof(struct udphdr),
- msg, copied);
+ err = skb_copy_datagram_msg(skb, 0, msg, copied);
else {
- err = skb_copy_and_csum_datagram_msg(skb, sizeof(struct udphdr),
- msg);
+ err = skb_copy_and_csum_datagram_msg(skb, 0, msg);
if (err == -EINVAL)
goto csum_copy_err;
@@ -1500,7 +1498,7 @@ static int __udp_queue_rcv_skb(struct sock *sk, struct sk_buff *skb)
sk_incoming_cpu_update(sk);
}
- rc = sock_queue_rcv_skb(sk, skb);
+ rc = __sock_queue_rcv_skb(sk, skb);
if (rc < 0) {
int is_udplite = IS_UDPLITE(sk);
@@ -1616,10 +1614,14 @@ int udp_queue_rcv_skb(struct sock *sk, struct sk_buff *skb)
}
}
- if (rcu_access_pointer(sk->sk_filter) &&
- udp_lib_checksum_complete(skb))
- goto csum_error;
+ if (rcu_access_pointer(sk->sk_filter)) {
+ if (udp_lib_checksum_complete(skb))
+ goto csum_error;
+ if (sk_filter(sk, skb))
+ goto drop;
+ }
+ udp_csum_pull_header(skb);
if (sk_rcvqueues_full(sk, sk->sk_rcvbuf)) {
UDP_INC_STATS_BH(sock_net(sk), UDP_MIB_RCVBUFERRORS,
is_udplite);
diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c
index 78a7dfd..84c8d7b 100644
--- a/net/ipv6/udp.c
+++ b/net/ipv6/udp.c
@@ -376,7 +376,7 @@ try_again:
if (!skb)
goto out;
- ulen = skb->len - sizeof(struct udphdr);
+ ulen = skb->len;
copied = len;
if (copied > ulen)
copied = ulen;
@@ -398,10 +398,9 @@ try_again:
}
if (checksum_valid || skb_csum_unnecessary(skb))
- err = skb_copy_datagram_msg(skb, sizeof(struct udphdr),
- msg, copied);
+ err = skb_copy_datagram_msg(skb, 0, msg, copied);
else {
- err = skb_copy_and_csum_datagram_msg(skb, sizeof(struct udphdr), msg);
+ err = skb_copy_and_csum_datagram_msg(skb, 0, msg);
if (err == -EINVAL)
goto csum_copy_err;
}
@@ -554,7 +553,7 @@ static int __udpv6_queue_rcv_skb(struct sock *sk, struct sk_buff *skb)
sk_incoming_cpu_update(sk);
}
- rc = sock_queue_rcv_skb(sk, skb);
+ rc = __sock_queue_rcv_skb(sk, skb);
if (rc < 0) {
int is_udplite = IS_UDPLITE(sk);
@@ -648,8 +647,11 @@ int udpv6_queue_rcv_skb(struct sock *sk, struct sk_buff *skb)
if (rcu_access_pointer(sk->sk_filter)) {
if (udp_lib_checksum_complete(skb))
goto csum_error;
+ if (sk_filter(sk, skb))
+ goto drop;
}
+ udp_csum_pull_header(skb);
if (sk_rcvqueues_full(sk, sk->sk_rcvbuf)) {
UDP6_INC_STATS_BH(sock_net(sk),
UDP_MIB_RCVBUFERRORS, is_udplite);
--
2.8.0.rc3.226.g39d4020
^ permalink raw reply related
* [PATCH net-next v2 1/3] sock: convert sk_peek_offset functions to WRITE_ONCE
From: Willem de Bruijn @ 2016-04-05 16:41 UTC (permalink / raw)
To: netdev; +Cc: davem, samanthakumar, edumazet, willemb
In-Reply-To: <1459874476-92838-1-git-send-email-willemdebruijn.kernel@gmail.com>
From: Willem de Bruijn <willemb@google.com>
Make the peek offset interface safe to use in lockless environments.
Use READ_ONCE and WRITE_ONCE to avoid race conditions between testing
and updating the peek offset.
Suggested-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Willem de Bruijn <willemb@google.com>
---
include/net/sock.h | 24 +++++++++++++-----------
1 file changed, 13 insertions(+), 11 deletions(-)
diff --git a/include/net/sock.h b/include/net/sock.h
index 310c436..09aec75 100644
--- a/include/net/sock.h
+++ b/include/net/sock.h
@@ -459,26 +459,28 @@ struct sock {
static inline int sk_peek_offset(struct sock *sk, int flags)
{
- if ((flags & MSG_PEEK) && (sk->sk_peek_off >= 0))
- return sk->sk_peek_off;
- else
- return 0;
+ if (unlikely(flags & MSG_PEEK)) {
+ s32 off = READ_ONCE(sk->sk_peek_off);
+ if (off >= 0)
+ return off;
+ }
+
+ return 0;
}
static inline void sk_peek_offset_bwd(struct sock *sk, int val)
{
- if (sk->sk_peek_off >= 0) {
- if (sk->sk_peek_off >= val)
- sk->sk_peek_off -= val;
- else
- sk->sk_peek_off = 0;
+ s32 off = READ_ONCE(sk->sk_peek_off);
+
+ if (unlikely(off >= 0)) {
+ off = max_t(s32, off - val, 0);
+ WRITE_ONCE(sk->sk_peek_off, off);
}
}
static inline void sk_peek_offset_fwd(struct sock *sk, int val)
{
- if (sk->sk_peek_off >= 0)
- sk->sk_peek_off += val;
+ sk_peek_offset_bwd(sk, -val);
}
/*
--
2.8.0.rc3.226.g39d4020
^ permalink raw reply related
* [PATCH net-next v2 0/3] udp: support SO_PEEK_OFF
From: Willem de Bruijn @ 2016-04-05 16:41 UTC (permalink / raw)
To: netdev; +Cc: davem, samanthakumar, edumazet, willemb
From: Willem de Bruijn <willemb@google.com>
Support peeking at a non-zero offset for UDP sockets. Match the
existing behavior on Unix datagram sockets.
1/3 makes the sk_peek_offset functions safe to use outside locks
2/3 removes udp headers before enqueue, to simplify offset arithmetic
3/3 introduces SO_PEEK_OFFSET support, with Unix socket peek semantics.
Changes
v1->v2
- squash patches 3 and 4
Willem de Bruijn (1):
sock: convert sk_peek_offset functions to WRITE_ONCE
samanthakumar (2):
udp: remove headers from UDP packets before queueing
udp: enable MSG_PEEK at non-zero offset
include/linux/skbuff.h | 7 ++++++-
include/net/sock.h | 27 ++++++++++++++++-----------
include/net/udp.h | 9 +++++++++
net/core/datagram.c | 9 ++++++---
net/core/sock.c | 28 ++++++++++++++++++++++------
net/ipv4/af_inet.c | 1 +
net/ipv4/udp.c | 38 ++++++++++++++++++++------------------
net/ipv6/af_inet6.c | 1 +
net/ipv6/udp.c | 30 ++++++++++++++++--------------
9 files changed, 97 insertions(+), 53 deletions(-)
--
2.8.0.rc3.226.g39d4020
^ permalink raw reply
* [PATCH net-next 0/3] udp: support SO_PEEK_OF
From: Willem de Bruijn @ 2016-04-05 16:38 UTC (permalink / raw)
To: netdev; +Cc: davem, samanthakumar, edumazet, willemb
From: Willem de Bruijn <willemb@google.com>
Support peeking at a non-zero offset for UDP sockets. Match the
existing behavior on Unix datagram sockets.
1/3 makes the sk_peek_offset functions safe to use outside locks
2/3 removes udp headers before enqueue, to simplify offset arithmetic
3/3 introduces SO_PEEK_OFFSET support, with Unix socket peek semantics.
Changes
v1->v2
- squash patches 3 and 4
Willem de Bruijn (1):
sock: convert sk_peek_offset functions to WRITE_ONCE
samanthakumar (2):
udp: remove headers from UDP packets before queueing
udp: enable MSG_PEEK at non-zero offset
include/linux/skbuff.h | 7 ++++++-
include/net/sock.h | 27 ++++++++++++++++-----------
include/net/udp.h | 9 +++++++++
net/core/datagram.c | 9 ++++++---
net/core/sock.c | 28 ++++++++++++++++++++++------
net/ipv4/af_inet.c | 1 +
net/ipv4/udp.c | 38 ++++++++++++++++++++------------------
net/ipv6/af_inet6.c | 1 +
net/ipv6/udp.c | 30 ++++++++++++++++--------------
9 files changed, 97 insertions(+), 53 deletions(-)
--
2.8.0.rc3.226.g39d4020
^ permalink raw reply
* Re: [net PATCH v2 2/2] ipv4/GRO: Make GRO conform to RFC 6864
From: Eric Dumazet @ 2016-04-05 16:30 UTC (permalink / raw)
To: Alexander Duyck
Cc: Herbert Xu, Alexander Duyck, Tom Herbert, Jesse Gross,
Eric Dumazet, Netdev, David Miller
In-Reply-To: <CAKgT0Uc=9MMzSHWgZib5+EwPCQokjSSnC9vfOtWKULvmwUQpdQ@mail.gmail.com>
On Tue, 2016-04-05 at 08:52 -0700, Alexander Duyck wrote:
>
> I disagree I think it will have to be part of the default
> configuration. The problem is the IP ID is quickly becoming
> meaningless. When you consider that a 40Gb/s link can wrap the IP ID
> value nearly 50 times a second using a 1500 MTU the IP ID field should
> just be ignored anyway because you cannot guarantee that it will be
> unique without limiting the Tx window size. That was the whole point
> of RFC6864. Basically the IP ID field is so small that as we push
> into the higher speeds you cannot guarantee that the field will have
> any meaning so for any case where you don't need to use it you
> shouldn't because it will likely not provide enough useful data.
Just because a few flows reach 40Gbit , we should remind that vast
majority of the Internet runs with < 50Mbits flows.
I prefer the argument of IPv6 not having ID ;)
We should do our best to keep interoperability, this is the selling
point.
And quite frankly your last patch makes perfect sense to me :
The aggregation is done only if the TCP headers of consecutive packets
matches. So who cares of IPv4 ID really ?
This is a very minor detail. The possible gains outperform the
theoretical 'problem'
GRO already reorder flows, it never had a guarantee of being 'ínvisible'
as Herbert claims.
^ 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