* [PATCH v5 net-next 4/6] net: ethernet: ti: cpsw: add CBS Qdisc offload
From: Ivan Khoronzhuk @ 2018-07-23 21:26 UTC (permalink / raw)
To: davem, grygorii.strashko
Cc: corbet, akpm, netdev, linux-doc, linux-kernel, linux-omap,
vinicius.gomes, henrik, jesus.sanchez-palencia, ilias.apalodimas,
p-varis, spatton, francois.ozog, yogeshs, nsekhar, andrew,
Ivan Khoronzhuk
In-Reply-To: <20180723212634.3219-1-ivan.khoronzhuk@linaro.org>
The cpsw has up to 4 FIFOs per port and upper 3 FIFOs can feed rate
limited queue with shaping. In order to set and enable shaping for
those 3 FIFOs queues the network device with CBS qdisc attached is
needed. The CBS configuration is added for dual-emac/single port mode
only, but potentially can be used in switch mode also, based on
switchdev for instance.
Despite the FIFO shapers can work w/o cpdma level shapers the base
usage must be in combine with cpdma level shapers as described in TRM,
that are set as maximum rates for interface queues with sysfs.
One of the possible configuration with txq shapers and CBS shapers:
Configured with echo RATE >
/sys/class/net/eth0/queues/tx-0/tx_maxrate
/---------------------------------------------------
/
/ cpdma level shapers
+----+ +----+ +----+ +----+ +----+ +----+ +----+ +----+
| c7 | | c6 | | c5 | | c4 | | c3 | | c2 | | c1 | | c0 |
\ / \ / \ / \ / \ / \ / \ / \ /
\ / \ / \ / \ / \ / \ / \ / \ /
\/ \/ \/ \/ \/ \/ \/ \/
+---------|------|------|------|-------------------------------------+
| +----+ | | +---+ |
| | +----+ | | |
| v v v v |
| +----+ +----+ +----+ +----+ p p+----+ +----+ +----+ +----+ |
| | | | | | | | | o o| | | | | | | | |
| | f3 | | f2 | | f1 | | f0 | r CPSW r| f3 | | f2 | | f1 | | f0 | |
| | | | | | | | | t t| | | | | | | | |
| \ / \ / \ / \ / 0 1\ / \ / \ / \ / |
| \ X \ / \ / \ / \ / \ / \ / \ / |
| \/ \ \/ \/ \/ \/ \/ \/ \/ |
+-------\------------------------------------------------------------+
\
\ FIFO shaper, set with CBS offload added in this patch,
\ FIFO0 cannot be rate limited
------------------------------------------------------
CBS shaper configuration is supposed to be used with root MQPRIO Qdisc
offload allowing to add sk_prio->tc->txq maps that direct traffic to
appropriate tx queue and maps L2 priority to FIFO shaper.
The CBS shaper is intended to be used for AVB where L2 priority
(pcp field) is used to differentiate class of traffic. So additionally
vlan needs to be created with appropriate egress sk_prio->l2 prio map.
If CBS has several tx queues assigned to it, the sum of their
bandwidth has not overlap bandwidth set for CBS. It's recomended the
CBS bandwidth to be a little bit more.
The CBS shaper is configured with CBS qdisc offload interface using tc
tool from iproute2 packet.
For instance:
$ tc qdisc replace dev eth0 handle 100: parent root mqprio num_tc 3 \
map 2 2 1 0 2 2 2 2 2 2 2 2 2 2 2 2 queues 1@0 1@1 2@2 hw 1
$ tc -g class show dev eth0
+---(100:ffe2) mqprio
| +---(100:3) mqprio
| +---(100:4) mqprio
|
+---(100:ffe1) mqprio
| +---(100:2) mqprio
|
+---(100:ffe0) mqprio
+---(100:1) mqprio
$ tc qdisc add dev eth0 parent 100:1 cbs locredit -1440 \
hicredit 60 sendslope -960000 idleslope 40000 offload 1
$ tc qdisc add dev eth0 parent 100:2 cbs locredit -1470 \
hicredit 62 sendslope -980000 idleslope 20000 offload 1
The above code set CBS shapers for tc0 and tc1, for that txq0 and
txq1 is used. Pay attention, the real set bandwidth can differ a bit
due to discreteness of configuration parameters.
Here parameters like locredit, hicredit and sendslope are ignored
internally and are supposed to be set with assumption that maximum
frame size for frame - 1500.
It's supposed that interface speed is not changed while reconnection,
not always is true, so inform user in case speed of interface was
changed, as it can impact on dependent shapers configuration.
For more examples see Documentation.
Reviewed-by: Ilias Apalodimas <ilias.apalodimas@linaro.org>
Reviewed-by: Grygorii Strashko <grygorii.strashko@ti.com>
Signed-off-by: Ivan Khoronzhuk <ivan.khoronzhuk@linaro.org>
---
drivers/net/ethernet/ti/cpsw.c | 221 +++++++++++++++++++++++++++++++++
1 file changed, 221 insertions(+)
diff --git a/drivers/net/ethernet/ti/cpsw.c b/drivers/net/ethernet/ti/cpsw.c
index f099e0ed138d..449dc7f1e5f8 100644
--- a/drivers/net/ethernet/ti/cpsw.c
+++ b/drivers/net/ethernet/ti/cpsw.c
@@ -46,6 +46,8 @@
#include "cpts.h"
#include "davinci_cpdma.h"
+#include <net/pkt_sched.h>
+
#define CPSW_DEBUG (NETIF_MSG_HW | NETIF_MSG_WOL | \
NETIF_MSG_DRV | NETIF_MSG_LINK | \
NETIF_MSG_IFUP | NETIF_MSG_INTR | \
@@ -154,8 +156,12 @@ do { \
#define IRQ_NUM 2
#define CPSW_MAX_QUEUES 8
#define CPSW_CPDMA_DESCS_POOL_SIZE_DEFAULT 256
+#define CPSW_FIFO_QUEUE_TYPE_SHIFT 16
+#define CPSW_FIFO_SHAPE_EN_SHIFT 16
+#define CPSW_FIFO_RATE_EN_SHIFT 20
#define CPSW_TC_NUM 4
#define CPSW_FIFO_SHAPERS_NUM (CPSW_TC_NUM - 1)
+#define CPSW_PCT_MASK 0x7f
#define CPSW_RX_VLAN_ENCAP_HDR_PRIO_SHIFT 29
#define CPSW_RX_VLAN_ENCAP_HDR_PRIO_MSK GENMASK(2, 0)
@@ -458,6 +464,8 @@ struct cpsw_priv {
bool rx_pause;
bool tx_pause;
bool mqprio_hw;
+ int fifo_bw[CPSW_TC_NUM];
+ int shp_cfg_speed;
u32 emac_port;
struct cpsw_common *cpsw;
};
@@ -1082,6 +1090,38 @@ static void cpsw_set_slave_mac(struct cpsw_slave *slave,
slave_write(slave, mac_lo(priv->mac_addr), SA_LO);
}
+static bool cpsw_shp_is_off(struct cpsw_priv *priv)
+{
+ struct cpsw_common *cpsw = priv->cpsw;
+ struct cpsw_slave *slave;
+ u32 shift, mask, val;
+
+ val = readl_relaxed(&cpsw->regs->ptype);
+
+ slave = &cpsw->slaves[cpsw_slave_index(cpsw, priv)];
+ shift = CPSW_FIFO_SHAPE_EN_SHIFT + 3 * slave->slave_num;
+ mask = 7 << shift;
+ val = val & mask;
+
+ return !val;
+}
+
+static void cpsw_fifo_shp_on(struct cpsw_priv *priv, int fifo, int on)
+{
+ struct cpsw_common *cpsw = priv->cpsw;
+ struct cpsw_slave *slave;
+ u32 shift, mask, val;
+
+ val = readl_relaxed(&cpsw->regs->ptype);
+
+ slave = &cpsw->slaves[cpsw_slave_index(cpsw, priv)];
+ shift = CPSW_FIFO_SHAPE_EN_SHIFT + 3 * slave->slave_num;
+ mask = (1 << --fifo) << shift;
+ val = on ? val | mask : val & ~mask;
+
+ writel_relaxed(val, &cpsw->regs->ptype);
+}
+
static void _cpsw_adjust_link(struct cpsw_slave *slave,
struct cpsw_priv *priv, bool *link)
{
@@ -1121,6 +1161,12 @@ static void _cpsw_adjust_link(struct cpsw_slave *slave,
mac_control |= BIT(4);
*link = true;
+
+ if (priv->shp_cfg_speed &&
+ priv->shp_cfg_speed != slave->phy->speed &&
+ !cpsw_shp_is_off(priv))
+ dev_warn(priv->dev,
+ "Speed was changed, CBS shaper speeds are changed!");
} else {
mac_control = 0;
/* disable forwarding */
@@ -1590,6 +1636,178 @@ static int cpsw_tc_to_fifo(int tc, int num_tc)
return CPSW_FIFO_SHAPERS_NUM - tc;
}
+static int cpsw_set_fifo_bw(struct cpsw_priv *priv, int fifo, int bw)
+{
+ struct cpsw_common *cpsw = priv->cpsw;
+ u32 val = 0, send_pct, shift;
+ struct cpsw_slave *slave;
+ int pct = 0, i;
+
+ if (bw > priv->shp_cfg_speed * 1000)
+ goto err;
+
+ /* shaping has to stay enabled for highest fifos linearly
+ * and fifo bw no more then interface can allow
+ */
+ slave = &cpsw->slaves[cpsw_slave_index(cpsw, priv)];
+ send_pct = slave_read(slave, SEND_PERCENT);
+ for (i = CPSW_FIFO_SHAPERS_NUM; i > 0; i--) {
+ if (!bw) {
+ if (i >= fifo || !priv->fifo_bw[i])
+ continue;
+
+ dev_warn(priv->dev, "Prev FIFO%d is shaped", i);
+ continue;
+ }
+
+ if (!priv->fifo_bw[i] && i > fifo) {
+ dev_err(priv->dev, "Upper FIFO%d is not shaped", i);
+ return -EINVAL;
+ }
+
+ shift = (i - 1) * 8;
+ if (i == fifo) {
+ send_pct &= ~(CPSW_PCT_MASK << shift);
+ val = DIV_ROUND_UP(bw, priv->shp_cfg_speed * 10);
+ if (!val)
+ val = 1;
+
+ send_pct |= val << shift;
+ pct += val;
+ continue;
+ }
+
+ if (priv->fifo_bw[i])
+ pct += (send_pct >> shift) & CPSW_PCT_MASK;
+ }
+
+ if (pct >= 100)
+ goto err;
+
+ slave_write(slave, send_pct, SEND_PERCENT);
+ priv->fifo_bw[fifo] = bw;
+
+ dev_warn(priv->dev, "set FIFO%d bw = %d\n", fifo,
+ DIV_ROUND_CLOSEST(val * priv->shp_cfg_speed, 100));
+
+ return 0;
+err:
+ dev_err(priv->dev, "Bandwidth doesn't fit in tc configuration");
+ return -EINVAL;
+}
+
+static int cpsw_set_fifo_rlimit(struct cpsw_priv *priv, int fifo, int bw)
+{
+ struct cpsw_common *cpsw = priv->cpsw;
+ struct cpsw_slave *slave;
+ u32 tx_in_ctl_rg, val;
+ int ret;
+
+ ret = cpsw_set_fifo_bw(priv, fifo, bw);
+ if (ret)
+ return ret;
+
+ slave = &cpsw->slaves[cpsw_slave_index(cpsw, priv)];
+ tx_in_ctl_rg = cpsw->version == CPSW_VERSION_1 ?
+ CPSW1_TX_IN_CTL : CPSW2_TX_IN_CTL;
+
+ if (!bw)
+ cpsw_fifo_shp_on(priv, fifo, bw);
+
+ val = slave_read(slave, tx_in_ctl_rg);
+ if (cpsw_shp_is_off(priv)) {
+ /* disable FIFOs rate limited queues */
+ val &= ~(0xf << CPSW_FIFO_RATE_EN_SHIFT);
+
+ /* set type of FIFO queues to normal priority mode */
+ val &= ~(3 << CPSW_FIFO_QUEUE_TYPE_SHIFT);
+
+ /* set type of FIFO queues to be rate limited */
+ if (bw)
+ val |= 2 << CPSW_FIFO_QUEUE_TYPE_SHIFT;
+ else
+ priv->shp_cfg_speed = 0;
+ }
+
+ /* toggle a FIFO rate limited queue */
+ if (bw)
+ val |= BIT(fifo + CPSW_FIFO_RATE_EN_SHIFT);
+ else
+ val &= ~BIT(fifo + CPSW_FIFO_RATE_EN_SHIFT);
+ slave_write(slave, val, tx_in_ctl_rg);
+
+ /* FIFO transmit shape enable */
+ cpsw_fifo_shp_on(priv, fifo, bw);
+ return 0;
+}
+
+/* Defaults:
+ * class A - prio 3
+ * class B - prio 2
+ * shaping for class A should be set first
+ */
+static int cpsw_set_cbs(struct net_device *ndev,
+ struct tc_cbs_qopt_offload *qopt)
+{
+ struct cpsw_priv *priv = netdev_priv(ndev);
+ struct cpsw_common *cpsw = priv->cpsw;
+ struct cpsw_slave *slave;
+ int prev_speed = 0;
+ int tc, ret, fifo;
+ u32 bw = 0;
+
+ tc = netdev_txq_to_tc(priv->ndev, qopt->queue);
+
+ /* enable channels in backward order, as highest FIFOs must be rate
+ * limited first and for compliance with CPDMA rate limited channels
+ * that also used in bacward order. FIFO0 cannot be rate limited.
+ */
+ fifo = cpsw_tc_to_fifo(tc, ndev->num_tc);
+ if (!fifo) {
+ dev_err(priv->dev, "Last tc%d can't be rate limited", tc);
+ return -EINVAL;
+ }
+
+ /* do nothing, it's disabled anyway */
+ if (!qopt->enable && !priv->fifo_bw[fifo])
+ return 0;
+
+ /* shapers can be set if link speed is known */
+ slave = &cpsw->slaves[cpsw_slave_index(cpsw, priv)];
+ if (slave->phy && slave->phy->link) {
+ if (priv->shp_cfg_speed &&
+ priv->shp_cfg_speed != slave->phy->speed)
+ prev_speed = priv->shp_cfg_speed;
+
+ priv->shp_cfg_speed = slave->phy->speed;
+ }
+
+ if (!priv->shp_cfg_speed) {
+ dev_err(priv->dev, "Link speed is not known");
+ return -1;
+ }
+
+ ret = pm_runtime_get_sync(cpsw->dev);
+ if (ret < 0) {
+ pm_runtime_put_noidle(cpsw->dev);
+ return ret;
+ }
+
+ bw = qopt->enable ? qopt->idleslope : 0;
+ ret = cpsw_set_fifo_rlimit(priv, fifo, bw);
+ if (ret) {
+ priv->shp_cfg_speed = prev_speed;
+ prev_speed = 0;
+ }
+
+ if (bw && prev_speed)
+ dev_warn(priv->dev,
+ "Speed was changed, CBS shaper speeds are changed!");
+
+ pm_runtime_put_sync(cpsw->dev);
+ return ret;
+}
+
static int cpsw_ndo_open(struct net_device *ndev)
{
struct cpsw_priv *priv = netdev_priv(ndev);
@@ -2264,6 +2482,9 @@ static int cpsw_ndo_setup_tc(struct net_device *ndev, enum tc_setup_type type,
void *type_data)
{
switch (type) {
+ case TC_SETUP_QDISC_CBS:
+ return cpsw_set_cbs(ndev, type_data);
+
case TC_SETUP_QDISC_MQPRIO:
return cpsw_set_mqprio(ndev, type_data);
--
2.17.1
^ permalink raw reply related
* [PATCH v5 net-next 3/6] net: ethernet: ti: cpsw: add MQPRIO Qdisc offload
From: Ivan Khoronzhuk @ 2018-07-23 21:26 UTC (permalink / raw)
To: davem, grygorii.strashko
Cc: corbet, akpm, netdev, linux-doc, linux-kernel, linux-omap,
vinicius.gomes, henrik, jesus.sanchez-palencia, ilias.apalodimas,
p-varis, spatton, francois.ozog, yogeshs, nsekhar, andrew,
Ivan Khoronzhuk
In-Reply-To: <20180723212634.3219-1-ivan.khoronzhuk@linaro.org>
That's possible to offload vlan to tc priority mapping with
assumption sk_prio == L2 prio.
Example:
$ ethtool -L eth0 rx 1 tx 4
$ qdisc replace dev eth0 handle 100: parent root mqprio num_tc 3 \
map 2 2 1 0 2 2 2 2 2 2 2 2 2 2 2 2 queues 1@0 1@1 2@2 hw 1
$ tc -g class show dev eth0
+---(100:ffe2) mqprio
| +---(100:3) mqprio
| +---(100:4) mqprio
|
+---(100:ffe1) mqprio
| +---(100:2) mqprio
|
+---(100:ffe0) mqprio
+---(100:1) mqprio
Here, 100:1 is txq0, 100:2 is txq1, 100:3 is txq2, 100:4 is txq3
txq0 belongs to tc0, txq1 to tc1, txq2 and txq3 to tc2
The offload part only maps L2 prio to classes of traffic, but not
to transmit queues, so to direct traffic to traffic class vlan has
to be created with appropriate egress map.
Reviewed-by: Ilias Apalodimas <ilias.apalodimas@linaro.org>
Reviewed-by: Grygorii Strashko <grygorii.strashko@ti.com>
Signed-off-by: Ivan Khoronzhuk <ivan.khoronzhuk@linaro.org>
---
drivers/net/ethernet/ti/cpsw.c | 82 ++++++++++++++++++++++++++++++++++
1 file changed, 82 insertions(+)
diff --git a/drivers/net/ethernet/ti/cpsw.c b/drivers/net/ethernet/ti/cpsw.c
index 4425b537b9dd..f099e0ed138d 100644
--- a/drivers/net/ethernet/ti/cpsw.c
+++ b/drivers/net/ethernet/ti/cpsw.c
@@ -39,6 +39,7 @@
#include <linux/sys_soc.h>
#include <linux/pinctrl/consumer.h>
+#include <net/pkt_cls.h>
#include "cpsw.h"
#include "cpsw_ale.h"
@@ -153,6 +154,8 @@ do { \
#define IRQ_NUM 2
#define CPSW_MAX_QUEUES 8
#define CPSW_CPDMA_DESCS_POOL_SIZE_DEFAULT 256
+#define CPSW_TC_NUM 4
+#define CPSW_FIFO_SHAPERS_NUM (CPSW_TC_NUM - 1)
#define CPSW_RX_VLAN_ENCAP_HDR_PRIO_SHIFT 29
#define CPSW_RX_VLAN_ENCAP_HDR_PRIO_MSK GENMASK(2, 0)
@@ -454,6 +457,7 @@ struct cpsw_priv {
u8 mac_addr[ETH_ALEN];
bool rx_pause;
bool tx_pause;
+ bool mqprio_hw;
u32 emac_port;
struct cpsw_common *cpsw;
};
@@ -1578,6 +1582,14 @@ static void cpsw_slave_stop(struct cpsw_slave *slave, struct cpsw_common *cpsw)
soft_reset_slave(slave);
}
+static int cpsw_tc_to_fifo(int tc, int num_tc)
+{
+ if (tc == num_tc - 1)
+ return 0;
+
+ return CPSW_FIFO_SHAPERS_NUM - tc;
+}
+
static int cpsw_ndo_open(struct net_device *ndev)
{
struct cpsw_priv *priv = netdev_priv(ndev);
@@ -2191,6 +2203,75 @@ static int cpsw_ndo_set_tx_maxrate(struct net_device *ndev, int queue, u32 rate)
return ret;
}
+static int cpsw_set_mqprio(struct net_device *ndev, void *type_data)
+{
+ struct tc_mqprio_qopt_offload *mqprio = type_data;
+ struct cpsw_priv *priv = netdev_priv(ndev);
+ struct cpsw_common *cpsw = priv->cpsw;
+ int fifo, num_tc, count, offset;
+ struct cpsw_slave *slave;
+ u32 tx_prio_map = 0;
+ int i, tc, ret;
+
+ num_tc = mqprio->qopt.num_tc;
+ if (num_tc > CPSW_TC_NUM)
+ return -EINVAL;
+
+ if (mqprio->mode != TC_MQPRIO_MODE_DCB)
+ return -EINVAL;
+
+ ret = pm_runtime_get_sync(cpsw->dev);
+ if (ret < 0) {
+ pm_runtime_put_noidle(cpsw->dev);
+ return ret;
+ }
+
+ if (num_tc) {
+ for (i = 0; i < 8; i++) {
+ tc = mqprio->qopt.prio_tc_map[i];
+ fifo = cpsw_tc_to_fifo(tc, num_tc);
+ tx_prio_map |= fifo << (4 * i);
+ }
+
+ netdev_set_num_tc(ndev, num_tc);
+ for (i = 0; i < num_tc; i++) {
+ count = mqprio->qopt.count[i];
+ offset = mqprio->qopt.offset[i];
+ netdev_set_tc_queue(ndev, i, count, offset);
+ }
+ }
+
+ if (!mqprio->qopt.hw) {
+ /* restore default configuration */
+ netdev_reset_tc(ndev);
+ tx_prio_map = TX_PRIORITY_MAPPING;
+ }
+
+ priv->mqprio_hw = mqprio->qopt.hw;
+
+ offset = cpsw->version == CPSW_VERSION_1 ?
+ CPSW1_TX_PRI_MAP : CPSW2_TX_PRI_MAP;
+
+ slave = &cpsw->slaves[cpsw_slave_index(cpsw, priv)];
+ slave_write(slave, tx_prio_map, offset);
+
+ pm_runtime_put_sync(cpsw->dev);
+
+ return 0;
+}
+
+static int cpsw_ndo_setup_tc(struct net_device *ndev, enum tc_setup_type type,
+ void *type_data)
+{
+ switch (type) {
+ case TC_SETUP_QDISC_MQPRIO:
+ return cpsw_set_mqprio(ndev, type_data);
+
+ default:
+ return -EOPNOTSUPP;
+ }
+}
+
static const struct net_device_ops cpsw_netdev_ops = {
.ndo_open = cpsw_ndo_open,
.ndo_stop = cpsw_ndo_stop,
@@ -2206,6 +2287,7 @@ static const struct net_device_ops cpsw_netdev_ops = {
#endif
.ndo_vlan_rx_add_vid = cpsw_ndo_vlan_rx_add_vid,
.ndo_vlan_rx_kill_vid = cpsw_ndo_vlan_rx_kill_vid,
+ .ndo_setup_tc = cpsw_ndo_setup_tc,
};
static int cpsw_get_regs_len(struct net_device *ndev)
--
2.17.1
^ permalink raw reply related
* [PATCH v5 net-next 5/6] net: ethernet: ti: cpsw: restore shaper configuration while down/up
From: Ivan Khoronzhuk @ 2018-07-23 21:26 UTC (permalink / raw)
To: davem, grygorii.strashko
Cc: corbet, akpm, netdev, linux-doc, linux-kernel, linux-omap,
vinicius.gomes, henrik, jesus.sanchez-palencia, ilias.apalodimas,
p-varis, spatton, francois.ozog, yogeshs, nsekhar, andrew,
Ivan Khoronzhuk
In-Reply-To: <20180723212634.3219-1-ivan.khoronzhuk@linaro.org>
Need to restore shapers configuration after interface was down/up.
This is needed as appropriate configuration is still replicated in
kernel settings. This only shapers context restore, so vlan
configuration should be restored by user if needed, especially for
devices with one port where vlan frames are sent via ALE.
Reviewed-by: Ilias Apalodimas <ilias.apalodimas@linaro.org>
Reviewed-by: Grygorii Strashko <grygorii.strashko@ti.com>
Signed-off-by: Ivan Khoronzhuk <ivan.khoronzhuk@linaro.org>
---
drivers/net/ethernet/ti/cpsw.c | 47 ++++++++++++++++++++++++++++++++++
1 file changed, 47 insertions(+)
diff --git a/drivers/net/ethernet/ti/cpsw.c b/drivers/net/ethernet/ti/cpsw.c
index 449dc7f1e5f8..171abcfb6184 100644
--- a/drivers/net/ethernet/ti/cpsw.c
+++ b/drivers/net/ethernet/ti/cpsw.c
@@ -1808,6 +1808,51 @@ static int cpsw_set_cbs(struct net_device *ndev,
return ret;
}
+static void cpsw_cbs_resume(struct cpsw_slave *slave, struct cpsw_priv *priv)
+{
+ int fifo, bw;
+
+ for (fifo = CPSW_FIFO_SHAPERS_NUM; fifo > 0; fifo--) {
+ bw = priv->fifo_bw[fifo];
+ if (!bw)
+ continue;
+
+ cpsw_set_fifo_rlimit(priv, fifo, bw);
+ }
+}
+
+static void cpsw_mqprio_resume(struct cpsw_slave *slave, struct cpsw_priv *priv)
+{
+ struct cpsw_common *cpsw = priv->cpsw;
+ u32 tx_prio_map = 0;
+ int i, tc, fifo;
+ u32 tx_prio_rg;
+
+ if (!priv->mqprio_hw)
+ return;
+
+ for (i = 0; i < 8; i++) {
+ tc = netdev_get_prio_tc_map(priv->ndev, i);
+ fifo = CPSW_FIFO_SHAPERS_NUM - tc;
+ tx_prio_map |= fifo << (4 * i);
+ }
+
+ tx_prio_rg = cpsw->version == CPSW_VERSION_1 ?
+ CPSW1_TX_PRI_MAP : CPSW2_TX_PRI_MAP;
+
+ slave_write(slave, tx_prio_map, tx_prio_rg);
+}
+
+/* restore resources after port reset */
+static void cpsw_restore(struct cpsw_priv *priv)
+{
+ /* restore MQPRIO offload */
+ for_each_slave(priv, cpsw_mqprio_resume, priv);
+
+ /* restore CBS offload */
+ for_each_slave(priv, cpsw_cbs_resume, priv);
+}
+
static int cpsw_ndo_open(struct net_device *ndev)
{
struct cpsw_priv *priv = netdev_priv(ndev);
@@ -1887,6 +1932,8 @@ static int cpsw_ndo_open(struct net_device *ndev)
}
+ cpsw_restore(priv);
+
/* Enable Interrupt pacing if configured */
if (cpsw->coal_intvl != 0) {
struct ethtool_coalesce coal;
--
2.17.1
^ permalink raw reply related
* [PATCH v5 net-next 6/6] Documentation: networking: cpsw: add MQPRIO & CBS offload examples
From: Ivan Khoronzhuk @ 2018-07-23 21:26 UTC (permalink / raw)
To: davem, grygorii.strashko
Cc: corbet, akpm, netdev, linux-doc, linux-kernel, linux-omap,
vinicius.gomes, henrik, jesus.sanchez-palencia, ilias.apalodimas,
p-varis, spatton, francois.ozog, yogeshs, nsekhar, andrew,
Ivan Khoronzhuk
In-Reply-To: <20180723212634.3219-1-ivan.khoronzhuk@linaro.org>
This document describes MQPRIO and CBS Qdisc offload configuration
for cpsw driver based on examples. It potentially can be used in
audio video bridging (AVB) and time sensitive networking (TSN).
Reviewed-by: Ilias Apalodimas <ilias.apalodimas@linaro.org>
Reviewed-by: Grygorii Strashko <grygorii.strashko@ti.com>
Signed-off-by: Ivan Khoronzhuk <ivan.khoronzhuk@linaro.org>
---
Documentation/networking/ti-cpsw.txt | 540 +++++++++++++++++++++++++++
1 file changed, 540 insertions(+)
create mode 100644 Documentation/networking/ti-cpsw.txt
diff --git a/Documentation/networking/ti-cpsw.txt b/Documentation/networking/ti-cpsw.txt
new file mode 100644
index 000000000000..67039205bd69
--- /dev/null
+++ b/Documentation/networking/ti-cpsw.txt
@@ -0,0 +1,540 @@
+* Texas Instruments CPSW ethernet driver
+
+Multiqueue & CBS & MQPRIO
+=====================================================================
+=====================================================================
+
+The cpsw has 3 CBS shapers for each external ports. This document
+describes MQPRIO and CBS Qdisc offload configuration for cpsw driver
+based on examples. It potentially can be used in audio video bridging
+(AVB) and time sensitive networking (TSN).
+
+The following examples were tested on AM572x EVM and BBB boards.
+
+Test setup
+==========
+
+Under consideration two examples with AM572x EVM running cpsw driver
+in dual_emac mode.
+
+Several prerequisites:
+- TX queues must be rated starting from txq0 that has highest priority
+- Traffic classes are used starting from 0, that has highest priority
+- CBS shapers should be used with rated queues
+- The bandwidth for CBS shapers has to be set a little bit more then
+ potential incoming rate, thus, rate of all incoming tx queues has
+ to be a little less
+- Real rates can differ, due to discreetness
+- Map skb-priority to txq is not enough, also skb-priority to l2 prio
+ map has to be created with ip or vconfig tool
+- Any l2/socket prio (0 - 7) for classes can be used, but for
+ simplicity default values are used: 3 and 2
+- only 2 classes tested: A and B, but checked and can work with more,
+ maximum allowed 4, but only for 3 rate can be set.
+
+Test setup for examples
+=======================
+ +-------------------------------+
+ |--+ |
+ | | Workstation0 |
+ |E | MAC 18:03:73:66:87:42 |
++-----------------------------+ +--|t | |
+| | 1 | E | | |h |./tsn_listener -d \ |
+| Target board: | 0 | t |--+ |0 | 18:03:73:66:87:42 -i eth0 \|
+| AM572x EVM | 0 | h | | | -s 1500 |
+| | 0 | 0 | |--+ |
+| Only 2 classes: |Mb +---| +-------------------------------+
+| class A, class B | |
+| | +---| +-------------------------------+
+| | 1 | E | |--+ |
+| | 0 | t | | | Workstation1 |
+| | 0 | h |--+ |E | MAC 20:cf:30:85:7d:fd |
+| |Mb | 1 | +--|t | |
++-----------------------------+ |h |./tsn_listener -d \ |
+ |0 | 20:cf:30:85:7d:fd -i eth0 \|
+ | | -s 1500 |
+ |--+ |
+ +-------------------------------+
+
+*********************************************************************
+*********************************************************************
+*********************************************************************
+Example 1: One port tx AVB configuration scheme for target board
+----------------------------------------------------------------------
+(prints and scheme for AM572x evm, applicable for single port boards)
+
+tc - traffic class
+txq - transmit queue
+p - priority
+f - fifo (cpsw fifo)
+S - shaper configured
+
++------------------------------------------------------------------+ u
+| +---------------+ +---------------+ +------+ +------+ | s
+| | | | | | | | | | e
+| | App 1 | | App 2 | | Apps | | Apps | | r
+| | Class A | | Class B | | Rest | | Rest | |
+| | Eth0 | | Eth0 | | Eth0 | | Eth1 | | s
+| | VLAN100 | | VLAN100 | | | | | | | | p
+| | 40 Mb/s | | 20 Mb/s | | | | | | | | a
+| | SO_PRIORITY=3 | | SO_PRIORITY=2 | | | | | | | | c
+| | | | | | | | | | | | | | e
+| +---|-----------+ +---|-----------+ +---|--+ +---|--+ |
++-----|------------------|------------------|--------|-------------+
+ +-+ +------------+ | |
+ | | +-----------------+ +--+
+ | | | |
++---|-------|-------------|-----------------------|----------------+
+| +----+ +----+ +----+ +----+ +----+ |
+| | p3 | | p2 | | p1 | | p0 | | p0 | | k
+| \ / \ / \ / \ / \ / | e
+| \ / \ / \ / \ / \ / | r
+| \/ \/ \/ \/ \/ | n
+| | | | | | e
+| | | +-----+ | | l
+| | | | | |
+| +----+ +----+ +----+ +----+ | s
+| |tc0 | |tc1 | |tc2 | |tc0 | | p
+| \ / \ / \ / \ / | a
+| \ / \ / \ / \ / | c
+| \/ \/ \/ \/ | e
+| | | +-----+ | |
+| | | | | | |
+| | | | | | |
+| | | | | | |
+| +----+ +----+ +----+ +----+ +----+ |
+| |txq0| |txq1| |txq2| |txq3| |txq4| |
+| \ / \ / \ / \ / \ / |
+| \ / \ / \ / \ / \ / |
+| \/ \/ \/ \/ \/ |
+| +-|------|------|------|--+ +--|--------------+ |
+| | | | | | | Eth0.100 | | Eth1 | |
++---|------|------|------|------------------------|----------------+
+ | | | | |
+ p p p p |
+ 3 2 0-1, 4-7 <- L2 priority |
+ | | | | |
+ | | | | |
++---|------|------|------|------------------------|----------------+
+| | | | | |----------+ |
+| +----+ +----+ +----+ +----+ +----+ |
+| |dma7| |dma6| |dma5| |dma4| |dma3| |
+| \ / \ / \ / \ / \ / | c
+| \S / \S / \ / \ / \ / | p
+| \/ \/ \/ \/ \/ | s
+| | | | +----- | | w
+| | | | | | |
+| | | | | | | d
+| +----+ +----+ +----+p p+----+ | r
+| | | | | | |o o| | | i
+| | f3 | | f2 | | f0 |r r| f0 | | v
+| |tc0 | |tc1 | |tc2 |t t|tc0 | | e
+| \CBS / \CBS / \CBS /1 2\CBS / | r
+| \S / \S / \ / \ / |
+| \/ \/ \/ \/ |
++------------------------------------------------------------------+
+========================================Eth==========================>
+
+1)
+// Add 4 tx queues, for interface Eth0, and 1 tx queue for Eth1
+$ ethtool -L eth0 rx 1 tx 5
+rx unmodified, ignoring
+
+2)
+// Check if num of queues is set correctly:
+$ ethtool -l eth0
+Channel parameters for eth0:
+Pre-set maximums:
+RX: 8
+TX: 8
+Other: 0
+Combined: 0
+Current hardware settings:
+RX: 1
+TX: 5
+Other: 0
+Combined: 0
+
+3)
+// TX queues must be rated starting from 0, so set bws for tx0 and tx1
+// Set rates 40 and 20 Mb/s appropriately.
+// Pay attention, real speed can differ a bit due to discreetness.
+// Leave last 2 tx queues not rated.
+$ echo 40 > /sys/class/net/eth0/queues/tx-0/tx_maxrate
+$ echo 20 > /sys/class/net/eth0/queues/tx-1/tx_maxrate
+
+4)
+// Check maximum rate of tx (cpdma) queues:
+$ cat /sys/class/net/eth0/queues/tx-*/tx_maxrate
+40
+20
+0
+0
+0
+
+5)
+// Map skb->priority to traffic class:
+// 3pri -> tc0, 2pri -> tc1, (0,1,4-7)pri -> tc2
+// Map traffic class to transmit queue:
+// tc0 -> txq0, tc1 -> txq1, tc2 -> (txq2, txq3)
+$ tc qdisc replace dev eth0 handle 100: parent root mqprio num_tc 3 \
+map 2 2 1 0 2 2 2 2 2 2 2 2 2 2 2 2 queues 1@0 1@1 2@2 hw 1
+
+5a)
+// As two interface sharing same set of tx queues, assign all traffic
+// coming to interface Eth1 to separate queue in order to not mix it
+// with traffic from interface Eth0, so use separate txq to send
+// packets to Eth1, so all prio -> tc0 and tc0 -> txq4
+// Here hw 0, so here still default configuration for eth1 in hw
+$ tc qdisc replace dev eth1 handle 100: parent root mqprio num_tc 1 \
+map 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 queues 1@4 hw 0
+
+6)
+// Check classes settings
+$ tc -g class show dev eth0
++---(100:ffe2) mqprio
+| +---(100:3) mqprio
+| +---(100:4) mqprio
+|
++---(100:ffe1) mqprio
+| +---(100:2) mqprio
+|
++---(100:ffe0) mqprio
+ +---(100:1) mqprio
+
+$ tc -g class show dev eth1
++---(100:ffe0) mqprio
+ +---(100:5) mqprio
+
+7)
+// Set rate for class A - 41 Mbit (tc0, txq0) using CBS Qdisc
+// Set it +1 Mb for reserve (important!)
+// here only idle slope is important, others arg are ignored
+// Pay attention, real speed can differ a bit due to discreetness
+$ tc qdisc add dev eth0 parent 100:1 cbs locredit -1438 \
+hicredit 62 sendslope -959000 idleslope 41000 offload 1
+net eth0: set FIFO3 bw = 50
+
+8)
+// Set rate for class B - 21 Mbit (tc1, txq1) using CBS Qdisc:
+// Set it +1 Mb for reserve (important!)
+$ tc qdisc add dev eth0 parent 100:2 cbs locredit -1468 \
+hicredit 65 sendslope -979000 idleslope 21000 offload 1
+net eth0: set FIFO2 bw = 30
+
+9)
+// Create vlan 100 to map sk->priority to vlan qos
+$ ip link add link eth0 name eth0.100 type vlan id 100
+8021q: 802.1Q VLAN Support v1.8
+8021q: adding VLAN 0 to HW filter on device eth0
+8021q: adding VLAN 0 to HW filter on device eth1
+net eth0: Adding vlanid 100 to vlan filter
+
+10)
+// Map skb->priority to L2 prio, 1 to 1
+$ ip link set eth0.100 type vlan \
+egress 0:0 1:1 2:2 3:3 4:4 5:5 6:6 7:7
+
+11)
+// Check egress map for vlan 100
+$ cat /proc/net/vlan/eth0.100
+[...]
+INGRESS priority mappings: 0:0 1:0 2:0 3:0 4:0 5:0 6:0 7:0
+EGRESS priority mappings: 0:0 1:1 2:2 3:3 4:4 5:5 6:6 7:7
+
+12)
+// Run your appropriate tools with socket option "SO_PRIORITY"
+// to 3 for class A and/or to 2 for class B
+// (I took at https://www.spinics.net/lists/netdev/msg460869.html)
+./tsn_talker -d 18:03:73:66:87:42 -i eth0.100 -p3 -s 1500&
+./tsn_talker -d 18:03:73:66:87:42 -i eth0.100 -p2 -s 1500&
+
+13)
+// run your listener on workstation (should be in same vlan)
+// (I took at https://www.spinics.net/lists/netdev/msg460869.html)
+./tsn_listener -d 18:03:73:66:87:42 -i enp5s0 -s 1500
+Receiving data rate: 39012 kbps
+Receiving data rate: 39012 kbps
+Receiving data rate: 39012 kbps
+Receiving data rate: 39012 kbps
+Receiving data rate: 39012 kbps
+Receiving data rate: 39012 kbps
+Receiving data rate: 39012 kbps
+Receiving data rate: 39012 kbps
+Receiving data rate: 39012 kbps
+Receiving data rate: 39012 kbps
+Receiving data rate: 39012 kbps
+Receiving data rate: 39012 kbps
+Receiving data rate: 39000 kbps
+
+14)
+// Restore default configuration if needed
+$ ip link del eth0.100
+$ tc qdisc del dev eth1 root
+$ tc qdisc del dev eth0 root
+net eth0: Prev FIFO2 is shaped
+net eth0: set FIFO3 bw = 0
+net eth0: set FIFO2 bw = 0
+$ ethtool -L eth0 rx 1 tx 1
+
+*********************************************************************
+*********************************************************************
+*********************************************************************
+Example 2: Two port tx AVB configuration scheme for target board
+----------------------------------------------------------------------
+(prints and scheme for AM572x evm, for dual emac boards only)
+
++------------------------------------------------------------------+ u
+| +----------+ +----------+ +------+ +----------+ +----------+ | s
+| | | | | | | | | | | | e
+| | App 1 | | App 2 | | Apps | | App 3 | | App 4 | | r
+| | Class A | | Class B | | Rest | | Class B | | Class A | |
+| | Eth0 | | Eth0 | | | | | Eth1 | | Eth1 | | s
+| | VLAN100 | | VLAN100 | | | | | VLAN100 | | VLAN100 | | p
+| | 40 Mb/s | | 20 Mb/s | | | | | 10 Mb/s | | 30 Mb/s | | a
+| | SO_PRI=3 | | SO_PRI=2 | | | | | SO_PRI=3 | | SO_PRI=2 | | c
+| | | | | | | | | | | | | | | | | e
+| +---|------+ +---|------+ +---|--+ +---|------+ +---|------+ |
++-----|-------------|-------------|---------|-------------|--------+
+ +-+ +-------+ | +----------+ +----+
+ | | +-------+------+ | |
+ | | | | | |
++---|-------|-------------|--------------|-------------|-------|---+
+| +----+ +----+ +----+ +----+ +----+ +----+ +----+ +----+ |
+| | p3 | | p2 | | p1 | | p0 | | p0 | | p1 | | p2 | | p3 | | k
+| \ / \ / \ / \ / \ / \ / \ / \ / | e
+| \ / \ / \ / \ / \ / \ / \ / \ / | r
+| \/ \/ \/ \/ \/ \/ \/ \/ | n
+| | | | | | | | e
+| | | +----+ +----+ | | | l
+| | | | | | | |
+| +----+ +----+ +----+ +----+ +----+ +----+ | s
+| |tc0 | |tc1 | |tc2 | |tc2 | |tc1 | |tc0 | | p
+| \ / \ / \ / \ / \ / \ / | a
+| \ / \ / \ / \ / \ / \ / | c
+| \/ \/ \/ \/ \/ \/ | e
+| | | +-----+ +-----+ | | |
+| | | | | | | | | |
+| | | | | | | | | |
+| | | | | E E | | | | |
+| +----+ +----+ +----+ +----+ t t +----+ +----+ +----+ +----+ |
+| |txq0| |txq1| |txq4| |txq5| h h |txq6| |txq7| |txq3| |txq2| |
+| \ / \ / \ / \ / 0 1 \ / \ / \ / \ / |
+| \ / \ / \ / \ / . . \ / \ / \ / \ / |
+| \/ \/ \/ \/ 1 1 \/ \/ \/ \/ |
+| +-|------|------|------|--+ 0 0 +-|------|------|------|--+ |
+| | | | | | | 0 0 | | | | | | |
++---|------|------|------|---------------|------|------|------|----+
+ | | | | | | | |
+ p p p p p p p p
+ 3 2 0-1, 4-7 <-L2 pri-> 0-1, 4-7 2 3
+ | | | | | | | |
+ | | | | | | | |
++---|------|------|------|---------------|------|------|------|----+
+| | | | | | | | | |
+| +----+ +----+ +----+ +----+ +----+ +----+ +----+ +----+ |
+| |dma7| |dma6| |dma3| |dma2| |dma1| |dma0| |dma4| |dma5| |
+| \ / \ / \ / \ / \ / \ / \ / \ / | c
+| \S / \S / \ / \ / \ / \ / \S / \S / | p
+| \/ \/ \/ \/ \/ \/ \/ \/ | s
+| | | | +----- | | | | | w
+| | | | | +----+ | | | |
+| | | | | | | | | | d
+| +----+ +----+ +----+p p+----+ +----+ +----+ | r
+| | | | | | |o o| | | | | | | i
+| | f3 | | f2 | | f0 |r CPSW r| f3 | | f2 | | f0 | | v
+| |tc0 | |tc1 | |tc2 |t t|tc0 | |tc1 | |tc2 | | e
+| \CBS / \CBS / \CBS /1 2\CBS / \CBS / \CBS / | r
+| \S / \S / \ / \S / \S / \ / |
+| \/ \/ \/ \/ \/ \/ |
++------------------------------------------------------------------+
+========================================Eth==========================>
+
+1)
+// Add 8 tx queues, for interface Eth0, but they are common, so are accessed
+// by two interfaces Eth0 and Eth1.
+$ ethtool -L eth1 rx 1 tx 8
+rx unmodified, ignoring
+
+2)
+// Check if num of queues is set correctly:
+$ ethtool -l eth0
+Channel parameters for eth0:
+Pre-set maximums:
+RX: 8
+TX: 8
+Other: 0
+Combined: 0
+Current hardware settings:
+RX: 1
+TX: 8
+Other: 0
+Combined: 0
+
+3)
+// TX queues must be rated starting from 0, so set bws for tx0 and tx1 for Eth0
+// and for tx2 and tx3 for Eth1. That is, rates 40 and 20 Mb/s appropriately
+// for Eth0 and 30 and 10 Mb/s for Eth1.
+// Real speed can differ a bit due to discreetness
+// Leave last 4 tx queues as not rated
+$ echo 40 > /sys/class/net/eth0/queues/tx-0/tx_maxrate
+$ echo 20 > /sys/class/net/eth0/queues/tx-1/tx_maxrate
+$ echo 30 > /sys/class/net/eth1/queues/tx-2/tx_maxrate
+$ echo 10 > /sys/class/net/eth1/queues/tx-3/tx_maxrate
+
+4)
+// Check maximum rate of tx (cpdma) queues:
+$ cat /sys/class/net/eth0/queues/tx-*/tx_maxrate
+40
+20
+30
+10
+0
+0
+0
+0
+
+5)
+// Map skb->priority to traffic class for Eth0:
+// 3pri -> tc0, 2pri -> tc1, (0,1,4-7)pri -> tc2
+// Map traffic class to transmit queue:
+// tc0 -> txq0, tc1 -> txq1, tc2 -> (txq4, txq5)
+$ tc qdisc replace dev eth0 handle 100: parent root mqprio num_tc 3 \
+map 2 2 1 0 2 2 2 2 2 2 2 2 2 2 2 2 queues 1@0 1@1 2@4 hw 1
+
+6)
+// Check classes settings
+$ tc -g class show dev eth0
++---(100:ffe2) mqprio
+| +---(100:5) mqprio
+| +---(100:6) mqprio
+|
++---(100:ffe1) mqprio
+| +---(100:2) mqprio
+|
++---(100:ffe0) mqprio
+ +---(100:1) mqprio
+
+7)
+// Set rate for class A - 41 Mbit (tc0, txq0) using CBS Qdisc for Eth0
+// here only idle slope is important, others ignored
+// Real speed can differ a bit due to discreetness
+$ tc qdisc add dev eth0 parent 100:1 cbs locredit -1470 \
+hicredit 62 sendslope -959000 idleslope 41000 offload 1
+net eth0: set FIFO3 bw = 50
+
+8)
+// Set rate for class B - 21 Mbit (tc1, txq1) using CBS Qdisc for Eth0
+$ tc qdisc add dev eth0 parent 100:2 cbs locredit -1470 \
+hicredit 65 sendslope -979000 idleslope 21000 offload 1
+net eth0: set FIFO2 bw = 30
+
+9)
+// Create vlan 100 to map sk->priority to vlan qos for Eth0
+$ ip link add link eth0 name eth0.100 type vlan id 100
+net eth0: Adding vlanid 100 to vlan filter
+
+10)
+// Map skb->priority to L2 prio for Eth0.100, one to one
+$ ip link set eth0.100 type vlan \
+egress 0:0 1:1 2:2 3:3 4:4 5:5 6:6 7:7
+
+11)
+// Check egress map for vlan 100
+$ cat /proc/net/vlan/eth0.100
+[...]
+INGRESS priority mappings: 0:0 1:0 2:0 3:0 4:0 5:0 6:0 7:0
+EGRESS priority mappings: 0:0 1:1 2:2 3:3 4:4 5:5 6:6 7:7
+
+12)
+// Map skb->priority to traffic class for Eth1:
+// 3pri -> tc0, 2pri -> tc1, (0,1,4-7)pri -> tc2
+// Map traffic class to transmit queue:
+// tc0 -> txq2, tc1 -> txq3, tc2 -> (txq6, txq7)
+$ tc qdisc replace dev eth1 handle 100: parent root mqprio num_tc 3 \
+map 2 2 1 0 2 2 2 2 2 2 2 2 2 2 2 2 queues 1@2 1@3 2@6 hw 1
+
+13)
+// Check classes settings
+$ tc -g class show dev eth1
++---(100:ffe2) mqprio
+| +---(100:7) mqprio
+| +---(100:8) mqprio
+|
++---(100:ffe1) mqprio
+| +---(100:4) mqprio
+|
++---(100:ffe0) mqprio
+ +---(100:3) mqprio
+
+14)
+// Set rate for class A - 31 Mbit (tc0, txq2) using CBS Qdisc for Eth1
+// here only idle slope is important, others ignored
+// Set it +1 Mb for reserve (important!)
+$ tc qdisc add dev eth1 parent 100:3 cbs locredit -1453 \
+hicredit 47 sendslope -969000 idleslope 31000 offload 1
+net eth1: set FIFO3 bw = 31
+
+15)
+// Set rate for class B - 11 Mbit (tc1, txq3) using CBS Qdisc for Eth1
+// Set it +1 Mb for reserve (important!)
+$ tc qdisc add dev eth1 parent 100:4 cbs locredit -1483 \
+hicredit 34 sendslope -989000 idleslope 11000 offload 1
+net eth1: set FIFO2 bw = 11
+
+16)
+// Create vlan 100 to map sk->priority to vlan qos for Eth1
+$ ip link add link eth1 name eth1.100 type vlan id 100
+net eth1: Adding vlanid 100 to vlan filter
+
+17)
+// Map skb->priority to L2 prio for Eth1.100, one to one
+$ ip link set eth1.100 type vlan \
+egress 0:0 1:1 2:2 3:3 4:4 5:5 6:6 7:7
+
+18)
+// Check egress map for vlan 100
+$ cat /proc/net/vlan/eth1.100
+[...]
+INGRESS priority mappings: 0:0 1:0 2:0 3:0 4:0 5:0 6:0 7:0
+EGRESS priority mappings: 0:0 1:1 2:2 3:3 4:4 5:5 6:6 7:7
+
+19)
+// Run appropriate tools with socket option "SO_PRIORITY" to 3
+// for class A and to 2 for class B. For both interfaces
+./tsn_talker -d 18:03:73:66:87:42 -i eth0.100 -p2 -s 1500&
+./tsn_talker -d 18:03:73:66:87:42 -i eth0.100 -p3 -s 1500&
+./tsn_talker -d 20:cf:30:85:7d:fd -i eth1.100 -p2 -s 1500&
+./tsn_talker -d 20:cf:30:85:7d:fd -i eth1.100 -p3 -s 1500&
+
+20)
+// run your listener on workstation (should be in same vlan)
+// (I took at https://www.spinics.net/lists/netdev/msg460869.html)
+./tsn_listener -d 18:03:73:66:87:42 -i enp5s0 -s 1500
+Receiving data rate: 39012 kbps
+Receiving data rate: 39012 kbps
+Receiving data rate: 39012 kbps
+Receiving data rate: 39012 kbps
+Receiving data rate: 39012 kbps
+Receiving data rate: 39012 kbps
+Receiving data rate: 39012 kbps
+Receiving data rate: 39012 kbps
+Receiving data rate: 39012 kbps
+Receiving data rate: 39012 kbps
+Receiving data rate: 39012 kbps
+Receiving data rate: 39012 kbps
+Receiving data rate: 39000 kbps
+
+21)
+// Restore default configuration if needed
+$ ip link del eth1.100
+$ ip link del eth0.100
+$ tc qdisc del dev eth1 root
+net eth1: Prev FIFO2 is shaped
+net eth1: set FIFO3 bw = 0
+net eth1: set FIFO2 bw = 0
+$ tc qdisc del dev eth0 root
+net eth0: Prev FIFO2 is shaped
+net eth0: set FIFO3 bw = 0
+net eth0: set FIFO2 bw = 0
+$ ethtool -L eth0 rx 1 tx 1
--
2.17.1
^ permalink raw reply related
* Re: [pull request][net 0/8] Mellanox, mlx5 fixes 2018-07-18
From: Saeed Mahameed @ 2018-07-23 21:27 UTC (permalink / raw)
To: davem@davemloft.net; +Cc: netdev@vger.kernel.org
In-Reply-To: <20180721.102050.722565861824892185.davem@davemloft.net>
On Sat, 2018-07-21 at 10:20 -0700, David Miller wrote:
> From: Saeed Mahameed <saeedm@mellanox.com>
> Date: Wed, 18 Jul 2018 18:26:04 -0700
>
> > The following series provides fixes to mlx5 core and net device
> > driver.
> >
> > Please pull and let me know if there's any problem.
>
> Pulled, thanks Saeed.
>
> Based upon the thread with Or, it would be useful to do some auditing
> and make sure all tunnels set skb->encapsulation.
>
Thanks Dave, Eran's patch relies on our driver setting skb-
>encapsulation, since the rps rule is injected directly from the device
rx path once netif_receive_skb_internal is called via the gro stack way
before forwarding the skb to the tunnel netdev.
> > For -stable v4.7
> > net/mlx5e: Don't allow aRFS for encapsulated packets
> > net/mlx5e: Fix quota counting in aRFS expire flow
> >
> > For -stable v4.15
> > net/mlx5e: Only allow offloading decap egress (egdev) flows
> > net/mlx5e: Refine ets validation function
> > net/mlx5: Adjust clock overflow work period
> >
> > For -stable v4.17
> > net/mlx5: E-Switch, UBSAN fix undefined behavior in
> > mlx5_eswitch_mode
>
> Queued up.
>
> Thanks.
^ permalink raw reply
* Re: [PATCH] bpf: Add Python 3 support to selftests scripts for bpf
From: Daniel Borkmann @ 2018-07-23 22:36 UTC (permalink / raw)
To: Dan Rue, Jeremy Cline
Cc: Alexei Starovoitov, Shuah Khan, netdev, linux-kernel,
linux-kselftest, Lawrence Brakmo, jakub.kicinski
In-Reply-To: <20180723173313.ri2uz7srtdtno2hl@linode.therub.org>
On 07/23/2018 07:33 PM, Dan Rue wrote:
> On Mon, Jul 23, 2018 at 10:08:57AM -0400, Jeremy Cline wrote:
>> On 07/20/2018 04:45 PM, Daniel Borkmann wrote:
>>> On 07/18/2018 11:36 PM, Jeremy Cline wrote:
>>>> Adjust tcp_client.py and tcp_server.py to work with Python 3 by using
>>>> the print function, marking string literals as bytes, and using the
>>>> newer exception syntax. This should be functionally equivalent and
>>>> support Python 2.6 through Python 3.7.
>>>>
>>>> Signed-off-by: Jeremy Cline <jcline@redhat.com>
>>>
>>> Thanks for the patch, Jeremy! Given we also have test_offload.py in BPF
>>> kselftests and it is written for python 3 only, it would probably make
>>> sense to adapt the tcp_{client,server}.py towards python 3 as well, so
>>> we wouldn't need to keep extra compat for 2 and have a consistent version
>>> dependency. Lawrence / Jeremy, any objections?
>>
>> I certainly don't object to Python 3 only and I'm happy to drop the
>> Python 2 compatibility from this patch if that's okay.
Sounds good, lets do it, please respin with that.
> This (well, along with introducing python in the first place, which took
> me by surprise), sounds like a policy decision that should be made clear
> in the kselftest documentation (Documentation/dev-tools/kselftest.rst).
> Currently, that file does not mention any python requirement.
Right now each selftest subdir has a config file which lists dependencies,
perhaps it makes sense to have another standardized file there (e.g. 'deps')
which lists user space dependencies, so it's immediately visible what is
needed to run all tests from there. Thoughts?
> That said, I agree that python2 support is no longer necessary.
>
> My use-case (which may be unusual?): We try to run all of kselftest
> against a variety of kernels and architectures for every push to next,
> mainline, and stable/lts branches. It seems that this is not a common
> usecase, but shouldn't it be?
As far as I'm aware the intel lkp-tests bot seems also to regularly run
the kselftests for x86, if also done from arm side e.g. on latest mainline,
even better.
Thanks,
Daniel
^ permalink raw reply
* Re: [pull request][net-next 00/16] Mellanox, mlx5e updates 2018-07-18
From: Saeed Mahameed @ 2018-07-23 21:35 UTC (permalink / raw)
To: davem@davemloft.net; +Cc: netdev@vger.kernel.org
In-Reply-To: <20180719010107.22363-1-saeedm@mellanox.com>
On Wed, 2018-07-18 at 18:00 -0700, Saeed Mahameed wrote:
> Hi dave,
>
> This series includes updates for mlx5e net device driver, with a
> couple
> of major features and some misc updates.
>
> Please notice the mlx5-next merge patch at the beginning:
> "Merge branch 'mlx5-next' of
> git://git.kernel.org/pub/scm/linux/kernel/git/mellanox/linux"
>
> For more information please see tag log below.
>
> Please pull and let me know if there's any problem.
>
I will re-post v2 without the "Support PCIe buffer congestion handling
via Devlink" patches until Eran sorts out the review comments.
Thanks,
Saeed.
> Thanks,
> Saeed.
>
> ---
>
> The following changes since commit
> 681d5d071c8bd5533a14244c0d55d1c0e30aa989:
>
> Merge branch 'mlx5-next' of
> git://git.kernel.org/pub/scm/linux/kernel/git/mellanox/linux (2018-
> 07-18 15:53:31 -0700)
>
> are available in the Git repository at:
>
> git://git.kernel.org/pub/scm/linux/kernel/git/saeed/linux.git
> tags/mlx5e-updates-2018-07-18
>
> for you to fetch changes up to
> a0ba57c09676689eb35f13d48990c9674c9baad4:
>
> net/mlx5e: Use PARTIAL_GSO for UDP segmentation (2018-07-18
> 17:26:28 -0700)
>
> ----------------------------------------------------------------
> mlx5e-updates-2018-07-18
>
> This series includes update for mlx5e net device driver.
>
> 1) From Feras Daoud, Added the support for firmware log tracing,
> first by introducing the firmware API needed for the task and then
> For each PF do the following:
> 1- Allocate memory for the tracer strings database and read it
> from the FW to the SW.
> 2- Allocate and dma map tracer buffers.
>
> Traces that will be written into the buffer will be parsed as a
> group
> of one or more traces, referred to as trace message. The trace
> message
> represents a C-like printf string.
> Once a new trace is available FW will generate an event indicates
> new trace/s are
> available and the driver will parse them and dump them using
> tracepoints
> event tracing
>
> Enable mlx5 fw tracing by:
> echo 1 > /sys/kernel/debug/tracing/events/mlx5/mlx5_fw/enable
>
> Read traces by:
> cat /sys/kernel/debug/tracing/trace
>
> 2) From Eran Ben Elisha, Support PCIe buffer congestion handling
> via Devlink, using the new devlink device parameters API, added the
> new
> parameters:
> - Congestion action
> HW mechanism in the PCIe buffer which monitors the amount
> of
> consumed PCIe buffer per host. This mechanism supports
> the
> following actions in case of threshold overflow:
> - Disabled - NOP (Default)
> - Drop
> - Mark - Mark CE bit in the CQE of received packet
> - Congestion mode
> - Aggressive - Aggressive static trigger threshold
> (Default)
> - Dynamic - Dynamically change the trigger threshold
>
> 3) From Natali, Set ECN for received packets using CQE indication.
> Using Eran's congestion settings a user can enable ECN marking, on
> such case
> driver must update ECN CE IP fields when requested by firmware
> (congestion is sensed).
>
> 4) From Roi Dayan, Remove redundant WARN when we cannot find neigh
> entry
>
> 5) From Jianbo Liu, TC double vlan support
> - Support offloading tc double vlan headers match
> - Support offloading double vlan push/pop tc actions
>
> 6) From Boris, re-visit UDP GSO, remove the splitting of UDP_GSO_L4
> packets
> in the driver, and exposes UDP_GSO_L4 as a PARTIAL_GSO feature.
>
> ----------------------------------------------------------------
> Boris Pismenny (1):
> net/mlx5e: Use PARTIAL_GSO for UDP segmentation
>
> Eran Ben Elisha (3):
> net/mlx5: Move all devlink related functions calls to devlink.c
> net/mlx5: Add MPEGC register configuration functionality
> net/mlx5: Support PCIe buffer congestion handling via Devlink
>
> Feras Daoud (5):
> net/mlx5: FW tracer, implement tracer logic
> net/mlx5: FW tracer, create trace buffer and copy strings
> database
> net/mlx5: FW tracer, events handling
> net/mlx5: FW tracer, parse traces and kernel tracing support
> net/mlx5: FW tracer, Enable tracing
>
> Jianbo Liu (3):
> net/mlx5e: Support offloading tc double vlan headers match
> net/mlx5e: Refactor tc vlan push/pop actions offloading
> net/mlx5e: Support offloading double vlan push/pop tc actions
>
> Natali Shechtman (1):
> net/mlx5e: Set ECN for received packets using CQE indication
>
> Roi Dayan (1):
> net/mlx5e: Remove redundant WARN when we cannot find neigh
> entry
>
> Saeed Mahameed (2):
> net/mlx5: FW tracer, register log buffer memory key
> net/mlx5: FW tracer, Add debug prints
>
> drivers/net/ethernet/mellanox/mlx5/core/Makefile | 6 +-
> drivers/net/ethernet/mellanox/mlx5/core/devlink.c | 267 ++++++
> drivers/net/ethernet/mellanox/mlx5/core/devlink.h | 41 +
> .../ethernet/mellanox/mlx5/core/diag/fw_tracer.c | 947
> +++++++++++++++++++++
> .../ethernet/mellanox/mlx5/core/diag/fw_tracer.h | 175 ++++
> .../mellanox/mlx5/core/diag/fw_tracer_tracepoint.h | 78 ++
> .../mellanox/mlx5/core/en_accel/en_accel.h | 27 +-
> .../ethernet/mellanox/mlx5/core/en_accel/rxtx.c | 109 ---
> .../ethernet/mellanox/mlx5/core/en_accel/rxtx.h | 14 -
> drivers/net/ethernet/mellanox/mlx5/core/en_main.c | 9 +-
> drivers/net/ethernet/mellanox/mlx5/core/en_rx.c | 35 +-
> drivers/net/ethernet/mellanox/mlx5/core/en_stats.c | 3 +
> drivers/net/ethernet/mellanox/mlx5/core/en_stats.h | 2 +
> drivers/net/ethernet/mellanox/mlx5/core/en_tc.c | 134 ++-
> drivers/net/ethernet/mellanox/mlx5/core/eq.c | 11 +
> drivers/net/ethernet/mellanox/mlx5/core/eswitch.h | 21 +-
> .../ethernet/mellanox/mlx5/core/eswitch_offloads.c | 23 +-
> drivers/net/ethernet/mellanox/mlx5/core/main.c | 23 +-
> include/linux/mlx5/device.h | 7 +
> include/linux/mlx5/driver.h | 3 +
> 20 files changed, 1745 insertions(+), 190 deletions(-)
> create mode 100644 drivers/net/ethernet/mellanox/mlx5/core/devlink.c
> create mode 100644 drivers/net/ethernet/mellanox/mlx5/core/devlink.h
> create mode 100644
> drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.c
> create mode 100644
> drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.h
> create mode 100644
> drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer_tracepoint.h
> delete mode 100644
> drivers/net/ethernet/mellanox/mlx5/core/en_accel/rxtx.c
> delete mode 100644
> drivers/net/ethernet/mellanox/mlx5/core/en_accel/rxtx.h
^ permalink raw reply
* Re: [PATCH] net/9p/trans_fd.c: fix race by holding the lock
From: Dominique Martinet @ 2018-07-23 22:54 UTC (permalink / raw)
To: Tomas Bortoli
Cc: ericvh, rminnich, lucho, davem, v9fs-developer, netdev,
linux-kernel, syzkaller
In-Reply-To: <20180723184253.6682-1-tomasbortoli@gmail.com>
Tomas Bortoli wrote on Mon, Jul 23, 2018:
> It may be possible to run p9_fd_cancel() with a deleted req->req_list
> and incur in a double del. To fix hold the client->lock while changing
> the status, so the other threads will be synchronized.
>
> Signed-off-by: Tomas Bortoli <tomasbortoli@gmail.com>
> Reported-by: syzbot+735d926e9d1317c3310c@syzkaller.appspotmail.com
OK, taking this one.
> ---
> net/9p/trans_fd.c | 10 +++++-----
> 1 file changed, 5 insertions(+), 5 deletions(-)
>
> diff --git a/net/9p/trans_fd.c b/net/9p/trans_fd.c
> index 588bf88c3305..f9f96d50d96d 100644
> --- a/net/9p/trans_fd.c
> +++ b/net/9p/trans_fd.c
> @@ -197,15 +197,14 @@ static void p9_mux_poll_stop(struct p9_conn *m)
> static void p9_conn_cancel(struct p9_conn *m, int err)
> {
> struct p9_req_t *req, *rtmp;
> - unsigned long flags;
> LIST_HEAD(cancel_list);
>
> p9_debug(P9_DEBUG_ERROR, "mux %p err %d\n", m, err);
>
> - spin_lock_irqsave(&m->client->lock, flags);
> + spin_lock(&m->client->lock);
>
> if (m->err) {
> - spin_unlock_irqrestore(&m->client->lock, flags);
> + spin_unlock(&m->client->lock);
> return;
> }
>
> @@ -217,7 +216,6 @@ static void p9_conn_cancel(struct p9_conn *m, int err)
> list_for_each_entry_safe(req, rtmp, &m->unsent_req_list, req_list) {
> list_move(&req->req_list, &cancel_list);
> }
> - spin_unlock_irqrestore(&m->client->lock, flags);
>
> list_for_each_entry_safe(req, rtmp, &cancel_list, req_list) {
> p9_debug(P9_DEBUG_ERROR, "call back req %p\n", req);
> @@ -226,6 +224,7 @@ static void p9_conn_cancel(struct p9_conn *m, int err)
> req->t_err = err;
> p9_client_cb(m->client, req, REQ_STATUS_ERROR);
> }
> + spin_unlock(&m->client->lock);
> }
>
> static __poll_t
> @@ -373,8 +372,9 @@ static void p9_read_work(struct work_struct *work)
> if (m->req->status != REQ_STATUS_ERROR)
> status = REQ_STATUS_RCVD;
> list_del(&m->req->req_list);
> - spin_unlock(&m->client->lock);
> + /* update req->status while holding client->lock */
> p9_client_cb(m->client, m->req, status);
> + spin_unlock(&m->client->lock);
> m->rc.sdata = NULL;
> m->rc.offset = 0;
> m->rc.capacity = 0;
^ permalink raw reply
* [PATCH] hinic: Link the logical network device to the pci device in sysfs
From: dann frazier @ 2018-07-23 22:55 UTC (permalink / raw)
To: Aviad Krawczyk, David S. Miller; +Cc: Zhao Chen, netdev, linux-kernel
Otherwise interfaces get exposed under /sys/devices/virtual, which
doesn't give udev the context it needs for PCI-based predictable
interface names.
Signed-off-by: dann frazier <dann.frazier@canonical.com>
---
drivers/net/ethernet/huawei/hinic/hinic_main.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/drivers/net/ethernet/huawei/hinic/hinic_main.c b/drivers/net/ethernet/huawei/hinic/hinic_main.c
index 5b122728dcb47..09e9da10b7865 100644
--- a/drivers/net/ethernet/huawei/hinic/hinic_main.c
+++ b/drivers/net/ethernet/huawei/hinic/hinic_main.c
@@ -983,6 +983,7 @@ static int nic_dev_init(struct pci_dev *pdev)
hinic_hwdev_cb_register(nic_dev->hwdev, HINIC_MGMT_MSG_CMD_LINK_STATUS,
nic_dev, link_status_event_handler);
+ SET_NETDEV_DEV(netdev, &pdev->dev);
err = register_netdev(netdev);
if (err) {
dev_err(&pdev->dev, "Failed to register netdev\n");
--
2.18.0
^ permalink raw reply related
* Re: [**EXTERNAL**] Re: VRF with enslaved L3 enabled bridge
From: David Ahern @ 2018-07-23 22:00 UTC (permalink / raw)
To: D'Souza, Nelson, netdev@vger.kernel.org
In-Reply-To: <F8810C2E-7791-43B8-AB64-6FC843431FEA@ciena.com>
On 7/20/18 1:03 PM, D'Souza, Nelson wrote:
> Setup is as follows:
>
> ethUSB(ingress port) -> mgmtbr0 (bridge) -> mgmtvrf (vrf)
| netns foo
[ test-vrf ] |
| |
[ br0 ] 172.16.1.1 |
| |
[ veth1 ] ============|======= [ veth2 ] lo
| 172.16.1.2 172.16.2.2
|
Copy and paste the following into your environment:
ip netns add foo
ip li add veth1 type veth peer name veth2
ip li set veth2 netns foo
ip -netns foo li set lo up
ip -netns foo li set veth2 up
ip -netns foo addr add 172.16.1.2/24 dev veth2
ip li add test-vrf type vrf table 123
ip li set test-vrf up
ip ro add vrf test-vrf unreachable default
ip li add br0 type bridge
ip li set veth1 master br0
ip li set veth1 up
ip li set br0 up
ip addr add dev br0 172.16.1.1/24
ip li set br0 master test-vrf
ip -netns foo addr add 172.16.2.2/32 dev lo
ip ro add vrf test-vrf 172.16.2.2/32 via 172.16.1.2
Does ping work?
# ping -I test-vrf 172.16.2.2
ping: Warning: source address might be selected on device other than
test-vrf.
PING 172.16.2.2 (172.16.2.2) from 172.16.1.1 test-vrf: 56(84) bytes of data.
64 bytes from 172.16.2.2: icmp_seq=1 ttl=64 time=0.228 ms
64 bytes from 172.16.2.2: icmp_seq=2 ttl=64 time=0.263 ms
and:
# ping -I br0 172.16.2.2
PING 172.16.2.2 (172.16.2.2) from 172.16.1.1 br0: 56(84) bytes of data.
64 bytes from 172.16.2.2: icmp_seq=1 ttl=64 time=0.227 ms
64 bytes from 172.16.2.2: icmp_seq=2 ttl=64 time=0.223 ms
^C
--- 172.16.2.2 ping statistics ---
2 packets transmitted, 2 received, 0% packet loss, time 1001ms
rtt min/avg/max/mdev = 0.223/0.225/0.227/0.002 ms
^ permalink raw reply
* Re: [PATCH nf] netfilter: nf_tables: move dumper state alloation into ->start
From: Pablo Neira Ayuso @ 2018-07-23 22:07 UTC (permalink / raw)
To: Florian Westphal; +Cc: netfilter-devel, netdev, cscnull
In-Reply-To: <20180723104714.24008-1-fw@strlen.de>
On Mon, Jul 23, 2018 at 12:47:14PM +0200, Florian Westphal wrote:
> Shaochun Chen points out we leak dumper filter state allocations
> stored in dump_control->data in case there is an error before netlink sets
> cb_running (after which ->done will be called at some point).
>
> In order to fix this, add .start functions and do the allocations
> there.
>
> ->done is going to clean up, and in case error occurs before
> ->start invocation no cleanups need to be done anymore.
Applied, thanks Florian.
^ permalink raw reply
* [pull request][net-next V2 00/12] Mellanox, mlx5e updates 2018-07-18
From: Saeed Mahameed @ 2018-07-23 22:11 UTC (permalink / raw)
To: David S. Miller; +Cc: netdev, Saeed Mahameed
Hi Dave,
This series includes updates for mlx5e net device driver, with a couple
of major features and some misc updates.
Please notice the mlx5-next merge patch at the beginning:
"Merge branch 'mlx5-next' of git://git.kernel.org/pub/scm/linux/kernel/git/mellanox/linux"
For more information please see tag log below.
Please pull and let me know if there's any problem.
v1->v2:
- Dropped "Support PCIe buffer congestion handling via Devlink" patches until the
comments are addressed.
Thanks,
Saeed.
---
The following changes since commit 7854ac44fe86548f8a6c6001938a1a2593b255e4:
Merge branch 'mlx5-next' of git://git.kernel.org/pub/scm/linux/kernel/git/mellanox/linux (2018-07-23 14:58:46 -0700)
are available in the Git repository at:
git://git.kernel.org/pub/scm/linux/kernel/git/saeed/linux.git tags/mlx5e-updates-2018-07-18-v2
for you to fetch changes up to 3f44899ef2ce0c9da49feb0d6f08098a08cb96ae:
net/mlx5e: Use PARTIAL_GSO for UDP segmentation (2018-07-23 15:01:11 -0700)
----------------------------------------------------------------
mlx5e-updates-2018-07-18
This series includes update for mlx5e net device driver.
1) From Feras Daoud, Added the support for firmware log tracing,
first by introducing the firmware API needed for the task and then
For each PF do the following:
1- Allocate memory for the tracer strings database and read it from the FW to the SW.
2- Allocate and dma map tracer buffers.
Traces that will be written into the buffer will be parsed as a group
of one or more traces, referred to as trace message. The trace message
represents a C-like printf string.
Once a new trace is available FW will generate an event indicates new trace/s are
available and the driver will parse them and dump them using tracepoints
event tracing
Enable mlx5 fw tracing by:
echo 1 > /sys/kernel/debug/tracing/events/mlx5/mlx5_fw/enable
Read traces by:
cat /sys/kernel/debug/tracing/trace
2) From Roi Dayan, Remove redundant WARN when we cannot find neigh entry
3) From Jianbo Liu, TC double vlan support
- Support offloading tc double vlan headers match
- Support offloading double vlan push/pop tc actions
4) From Boris, re-visit UDP GSO, remove the splitting of UDP_GSO_L4 packets
in the driver, and exposes UDP_GSO_L4 as a PARTIAL_GSO feature.
----------------------------------------------------------------
Boris Pismenny (1):
net/mlx5e: Use PARTIAL_GSO for UDP segmentation
Feras Daoud (5):
net/mlx5: FW tracer, implement tracer logic
net/mlx5: FW tracer, create trace buffer and copy strings database
net/mlx5: FW tracer, events handling
net/mlx5: FW tracer, parse traces and kernel tracing support
net/mlx5: FW tracer, Enable tracing
Jianbo Liu (3):
net/mlx5e: Support offloading tc double vlan headers match
net/mlx5e: Refactor tc vlan push/pop actions offloading
net/mlx5e: Support offloading double vlan push/pop tc actions
Roi Dayan (1):
net/mlx5e: Remove redundant WARN when we cannot find neigh entry
Saeed Mahameed (2):
net/mlx5: FW tracer, register log buffer memory key
net/mlx5: FW tracer, Add debug prints
drivers/net/ethernet/mellanox/mlx5/core/Makefile | 6 +-
.../ethernet/mellanox/mlx5/core/diag/fw_tracer.c | 947 +++++++++++++++++++++
.../ethernet/mellanox/mlx5/core/diag/fw_tracer.h | 175 ++++
.../mellanox/mlx5/core/diag/fw_tracer_tracepoint.h | 78 ++
.../mellanox/mlx5/core/en_accel/en_accel.h | 27 +-
.../ethernet/mellanox/mlx5/core/en_accel/rxtx.c | 109 ---
.../ethernet/mellanox/mlx5/core/en_accel/rxtx.h | 14 -
drivers/net/ethernet/mellanox/mlx5/core/en_main.c | 9 +-
drivers/net/ethernet/mellanox/mlx5/core/en_tc.c | 134 ++-
drivers/net/ethernet/mellanox/mlx5/core/eq.c | 11 +
drivers/net/ethernet/mellanox/mlx5/core/eswitch.h | 21 +-
.../ethernet/mellanox/mlx5/core/eswitch_offloads.c | 23 +-
drivers/net/ethernet/mellanox/mlx5/core/main.c | 18 +-
include/linux/mlx5/device.h | 7 +
include/linux/mlx5/driver.h | 3 +
15 files changed, 1399 insertions(+), 183 deletions(-)
create mode 100644 drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.c
create mode 100644 drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.h
create mode 100644 drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer_tracepoint.h
delete mode 100644 drivers/net/ethernet/mellanox/mlx5/core/en_accel/rxtx.c
delete mode 100644 drivers/net/ethernet/mellanox/mlx5/core/en_accel/rxtx.h
^ permalink raw reply
* [net-next V2 01/12] net/mlx5: FW tracer, implement tracer logic
From: Saeed Mahameed @ 2018-07-23 22:11 UTC (permalink / raw)
To: David S. Miller; +Cc: netdev, Feras Daoud, Saeed Mahameed
In-Reply-To: <20180723221129.21625-1-saeedm@mellanox.com>
From: Feras Daoud <ferasda@mellanox.com>
Implement FW tracer logic and registers access, initialization and
cleanup flows.
Initializing the tracer will be part of load one flow, as multiple
PFs will try to acquire ownership but only one will succeed and will
be the tracer owner.
Signed-off-by: Feras Daoud <ferasda@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
.../mellanox/mlx5/core/diag/fw_tracer.c | 196 ++++++++++++++++++
.../mellanox/mlx5/core/diag/fw_tracer.h | 66 ++++++
include/linux/mlx5/driver.h | 3 +
3 files changed, 265 insertions(+)
create mode 100644 drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.c
create mode 100644 drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.h
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.c b/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.c
new file mode 100644
index 000000000000..3ecbf06b4d71
--- /dev/null
+++ b/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.c
@@ -0,0 +1,196 @@
+/*
+ * Copyright (c) 2018, Mellanox Technologies. All rights reserved.
+ *
+ * This software is available to you under a choice of one of two
+ * licenses. You may choose to be licensed under the terms of the GNU
+ * General Public License (GPL) Version 2, available from the file
+ * COPYING in the main directory of this source tree, or the
+ * OpenIB.org BSD license below:
+ *
+ * Redistribution and use in source and binary forms, with or
+ * without modification, are permitted provided that the following
+ * conditions are met:
+ *
+ * - Redistributions of source code must retain the above
+ * copyright notice, this list of conditions and the following
+ * disclaimer.
+ *
+ * - Redistributions in binary form must reproduce the above
+ * copyright notice, this list of conditions and the following
+ * disclaimer in the documentation and/or other materials
+ * provided with the distribution.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
+ * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
+ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ */
+
+#include "fw_tracer.h"
+
+static int mlx5_query_mtrc_caps(struct mlx5_fw_tracer *tracer)
+{
+ u32 *string_db_base_address_out = tracer->str_db.base_address_out;
+ u32 *string_db_size_out = tracer->str_db.size_out;
+ struct mlx5_core_dev *dev = tracer->dev;
+ u32 out[MLX5_ST_SZ_DW(mtrc_cap)] = {0};
+ u32 in[MLX5_ST_SZ_DW(mtrc_cap)] = {0};
+ void *mtrc_cap_sp;
+ int err, i;
+
+ err = mlx5_core_access_reg(dev, in, sizeof(in), out, sizeof(out),
+ MLX5_REG_MTRC_CAP, 0, 0);
+ if (err) {
+ mlx5_core_warn(dev, "FWTracer: Error reading tracer caps %d\n",
+ err);
+ return err;
+ }
+
+ if (!MLX5_GET(mtrc_cap, out, trace_to_memory)) {
+ mlx5_core_dbg(dev, "FWTracer: Device does not support logging traces to memory\n");
+ return -ENOTSUPP;
+ }
+
+ tracer->trc_ver = MLX5_GET(mtrc_cap, out, trc_ver);
+ tracer->str_db.first_string_trace =
+ MLX5_GET(mtrc_cap, out, first_string_trace);
+ tracer->str_db.num_string_trace =
+ MLX5_GET(mtrc_cap, out, num_string_trace);
+ tracer->str_db.num_string_db = MLX5_GET(mtrc_cap, out, num_string_db);
+ tracer->owner = !!MLX5_GET(mtrc_cap, out, trace_owner);
+
+ for (i = 0; i < tracer->str_db.num_string_db; i++) {
+ mtrc_cap_sp = MLX5_ADDR_OF(mtrc_cap, out, string_db_param[i]);
+ string_db_base_address_out[i] = MLX5_GET(mtrc_string_db_param,
+ mtrc_cap_sp,
+ string_db_base_address);
+ string_db_size_out[i] = MLX5_GET(mtrc_string_db_param,
+ mtrc_cap_sp, string_db_size);
+ }
+
+ return err;
+}
+
+static int mlx5_set_mtrc_caps_trace_owner(struct mlx5_fw_tracer *tracer,
+ u32 *out, u32 out_size,
+ u8 trace_owner)
+{
+ struct mlx5_core_dev *dev = tracer->dev;
+ u32 in[MLX5_ST_SZ_DW(mtrc_cap)] = {0};
+
+ MLX5_SET(mtrc_cap, in, trace_owner, trace_owner);
+
+ return mlx5_core_access_reg(dev, in, sizeof(in), out, out_size,
+ MLX5_REG_MTRC_CAP, 0, 1);
+}
+
+static int mlx5_fw_tracer_ownership_acquire(struct mlx5_fw_tracer *tracer)
+{
+ struct mlx5_core_dev *dev = tracer->dev;
+ u32 out[MLX5_ST_SZ_DW(mtrc_cap)] = {0};
+ int err;
+
+ err = mlx5_set_mtrc_caps_trace_owner(tracer, out, sizeof(out),
+ MLX5_FW_TRACER_ACQUIRE_OWNERSHIP);
+ if (err) {
+ mlx5_core_warn(dev, "FWTracer: Acquire tracer ownership failed %d\n",
+ err);
+ return err;
+ }
+
+ tracer->owner = !!MLX5_GET(mtrc_cap, out, trace_owner);
+
+ if (!tracer->owner)
+ return -EBUSY;
+
+ return 0;
+}
+
+static void mlx5_fw_tracer_ownership_release(struct mlx5_fw_tracer *tracer)
+{
+ u32 out[MLX5_ST_SZ_DW(mtrc_cap)] = {0};
+
+ mlx5_set_mtrc_caps_trace_owner(tracer, out, sizeof(out),
+ MLX5_FW_TRACER_RELEASE_OWNERSHIP);
+ tracer->owner = false;
+}
+
+static void mlx5_fw_tracer_ownership_change(struct work_struct *work)
+{
+ struct mlx5_fw_tracer *tracer = container_of(work, struct mlx5_fw_tracer,
+ ownership_change_work);
+ struct mlx5_core_dev *dev = tracer->dev;
+ int err;
+
+ if (tracer->owner) {
+ mlx5_fw_tracer_ownership_release(tracer);
+ return;
+ }
+
+ err = mlx5_fw_tracer_ownership_acquire(tracer);
+ if (err) {
+ mlx5_core_dbg(dev, "FWTracer: Ownership was not granted %d\n", err);
+ return;
+ }
+}
+
+struct mlx5_fw_tracer *mlx5_fw_tracer_create(struct mlx5_core_dev *dev)
+{
+ struct mlx5_fw_tracer *tracer = NULL;
+ int err;
+
+ if (!MLX5_CAP_MCAM_REG(dev, tracer_registers)) {
+ mlx5_core_dbg(dev, "FWTracer: Tracer capability not present\n");
+ return NULL;
+ }
+
+ tracer = kzalloc(sizeof(*tracer), GFP_KERNEL);
+ if (!tracer)
+ return ERR_PTR(-ENOMEM);
+
+ tracer->work_queue = create_singlethread_workqueue("mlx5_fw_tracer");
+ if (!tracer->work_queue) {
+ err = -ENOMEM;
+ goto free_tracer;
+ }
+
+ tracer->dev = dev;
+
+ INIT_WORK(&tracer->ownership_change_work, mlx5_fw_tracer_ownership_change);
+
+ err = mlx5_query_mtrc_caps(tracer);
+ if (err) {
+ mlx5_core_dbg(dev, "FWTracer: Failed to query capabilities %d\n", err);
+ goto destroy_workqueue;
+ }
+
+ mlx5_fw_tracer_ownership_change(&tracer->ownership_change_work);
+
+ return tracer;
+
+destroy_workqueue:
+ tracer->dev = NULL;
+ destroy_workqueue(tracer->work_queue);
+free_tracer:
+ kfree(tracer);
+ return ERR_PTR(err);
+}
+
+void mlx5_fw_tracer_destroy(struct mlx5_fw_tracer *tracer)
+{
+ if (!tracer)
+ return;
+
+ cancel_work_sync(&tracer->ownership_change_work);
+
+ if (tracer->owner)
+ mlx5_fw_tracer_ownership_release(tracer);
+
+ flush_workqueue(tracer->work_queue);
+ destroy_workqueue(tracer->work_queue);
+ kfree(tracer);
+}
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.h b/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.h
new file mode 100644
index 000000000000..721c41a5e827
--- /dev/null
+++ b/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.h
@@ -0,0 +1,66 @@
+/*
+ * Copyright (c) 2018, Mellanox Technologies. All rights reserved.
+ *
+ * This software is available to you under a choice of one of two
+ * licenses. You may choose to be licensed under the terms of the GNU
+ * General Public License (GPL) Version 2, available from the file
+ * COPYING in the main directory of this source tree, or the
+ * OpenIB.org BSD license below:
+ *
+ * Redistribution and use in source and binary forms, with or
+ * without modification, are permitted provided that the following
+ * conditions are met:
+ *
+ * - Redistributions of source code must retain the above
+ * copyright notice, this list of conditions and the following
+ * disclaimer.
+ *
+ * - Redistributions in binary form must reproduce the above
+ * copyright notice, this list of conditions and the following
+ * disclaimer in the documentation and/or other materials
+ * provided with the distribution.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
+ * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
+ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ */
+
+#ifndef __LIB_TRACER_H__
+#define __LIB_TRACER_H__
+
+#include <linux/mlx5/driver.h>
+#include "mlx5_core.h"
+
+#define STRINGS_DB_SECTIONS_NUM 8
+
+struct mlx5_fw_tracer {
+ struct mlx5_core_dev *dev;
+ bool owner;
+ u8 trc_ver;
+ struct workqueue_struct *work_queue;
+ struct work_struct ownership_change_work;
+
+ /* Strings DB */
+ struct {
+ u8 first_string_trace;
+ u8 num_string_trace;
+ u32 num_string_db;
+ u32 base_address_out[STRINGS_DB_SECTIONS_NUM];
+ u32 size_out[STRINGS_DB_SECTIONS_NUM];
+ } str_db;
+};
+
+enum mlx5_fw_tracer_ownership_state {
+ MLX5_FW_TRACER_RELEASE_OWNERSHIP,
+ MLX5_FW_TRACER_ACQUIRE_OWNERSHIP,
+};
+
+struct mlx5_fw_tracer *mlx5_fw_tracer_create(struct mlx5_core_dev *dev);
+void mlx5_fw_tracer_destroy(struct mlx5_fw_tracer *tracer);
+
+#endif
diff --git a/include/linux/mlx5/driver.h b/include/linux/mlx5/driver.h
index 957199c20a0f..86cb0ebf92fa 100644
--- a/include/linux/mlx5/driver.h
+++ b/include/linux/mlx5/driver.h
@@ -816,6 +816,8 @@ struct mlx5_clock {
struct mlx5_pps pps_info;
};
+struct mlx5_fw_tracer;
+
struct mlx5_core_dev {
struct pci_dev *pdev;
/* sync pci state */
@@ -860,6 +862,7 @@ struct mlx5_core_dev {
struct mlx5_clock clock;
struct mlx5_ib_clock_info *clock_info;
struct page *clock_info_page;
+ struct mlx5_fw_tracer *tracer;
};
struct mlx5_db {
--
2.17.0
^ permalink raw reply related
* [net-next V2 03/12] net/mlx5: FW tracer, register log buffer memory key
From: Saeed Mahameed @ 2018-07-23 22:11 UTC (permalink / raw)
To: David S. Miller; +Cc: netdev, Saeed Mahameed
In-Reply-To: <20180723221129.21625-1-saeedm@mellanox.com>
Create a memory key and protection domain for the tracer log buffer.
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
.../mellanox/mlx5/core/diag/fw_tracer.c | 64 ++++++++++++++++++-
1 file changed, 61 insertions(+), 3 deletions(-)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.c b/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.c
index 35107b8f76df..d6cc27b0ff34 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.c
@@ -169,6 +169,48 @@ static void mlx5_fw_tracer_destroy_log_buf(struct mlx5_fw_tracer *tracer)
free_pages((unsigned long)tracer->buff.log_buf, get_order(tracer->buff.size));
}
+static int mlx5_fw_tracer_create_mkey(struct mlx5_fw_tracer *tracer)
+{
+ struct mlx5_core_dev *dev = tracer->dev;
+ int err, inlen, i;
+ __be64 *mtt;
+ void *mkc;
+ u32 *in;
+
+ inlen = MLX5_ST_SZ_BYTES(create_mkey_in) +
+ sizeof(*mtt) * round_up(TRACER_BUFFER_PAGE_NUM, 2);
+
+ in = kvzalloc(inlen, GFP_KERNEL);
+ if (!in)
+ return -ENOMEM;
+
+ MLX5_SET(create_mkey_in, in, translations_octword_actual_size,
+ DIV_ROUND_UP(TRACER_BUFFER_PAGE_NUM, 2));
+ mtt = (u64 *)MLX5_ADDR_OF(create_mkey_in, in, klm_pas_mtt);
+ for (i = 0 ; i < TRACER_BUFFER_PAGE_NUM ; i++)
+ mtt[i] = cpu_to_be64(tracer->buff.dma + i * PAGE_SIZE);
+
+ mkc = MLX5_ADDR_OF(create_mkey_in, in, memory_key_mkey_entry);
+ MLX5_SET(mkc, mkc, access_mode_1_0, MLX5_MKC_ACCESS_MODE_MTT);
+ MLX5_SET(mkc, mkc, lr, 1);
+ MLX5_SET(mkc, mkc, lw, 1);
+ MLX5_SET(mkc, mkc, pd, tracer->buff.pdn);
+ MLX5_SET(mkc, mkc, bsf_octword_size, 0);
+ MLX5_SET(mkc, mkc, qpn, 0xffffff);
+ MLX5_SET(mkc, mkc, log_page_size, PAGE_SHIFT);
+ MLX5_SET(mkc, mkc, translations_octword_size,
+ DIV_ROUND_UP(TRACER_BUFFER_PAGE_NUM, 2));
+ MLX5_SET64(mkc, mkc, start_addr, tracer->buff.dma);
+ MLX5_SET64(mkc, mkc, len, tracer->buff.size);
+ err = mlx5_core_create_mkey(dev, &tracer->buff.mkey, in, inlen);
+ if (err)
+ mlx5_core_warn(dev, "FWTracer: Failed to create mkey, %d\n", err);
+
+ kvfree(in);
+
+ return err;
+}
+
static void mlx5_fw_tracer_free_strings_db(struct mlx5_fw_tracer *tracer)
{
u32 num_string_db = tracer->str_db.num_string_db;
@@ -363,13 +405,26 @@ int mlx5_fw_tracer_init(struct mlx5_fw_tracer *tracer)
if (!tracer->str_db.loaded)
queue_work(tracer->work_queue, &tracer->read_fw_strings_work);
- err = mlx5_fw_tracer_ownership_acquire(tracer);
+ err = mlx5_core_alloc_pd(dev, &tracer->buff.pdn);
if (err) {
- mlx5_core_dbg(dev, "FWTracer: Ownership was not granted %d\n", err);
- return 0; /* return 0 since ownership can be acquired on a later FW event */
+ mlx5_core_warn(dev, "FWTracer: Failed to allocate PD %d\n", err);
+ return err;
}
+ err = mlx5_fw_tracer_create_mkey(tracer);
+ if (err) {
+ mlx5_core_warn(dev, "FWTracer: Failed to create mkey %d\n", err);
+ goto err_dealloc_pd;
+ }
+
+ err = mlx5_fw_tracer_ownership_acquire(tracer);
+ if (err) /* Don't fail since ownership can be acquired on a later FW event */
+ mlx5_core_dbg(dev, "FWTracer: Ownership was not granted %d\n", err);
+
return 0;
+err_dealloc_pd:
+ mlx5_core_dealloc_pd(dev, tracer->buff.pdn);
+ return err;
}
void mlx5_fw_tracer_cleanup(struct mlx5_fw_tracer *tracer)
@@ -381,6 +436,9 @@ void mlx5_fw_tracer_cleanup(struct mlx5_fw_tracer *tracer)
if (tracer->owner)
mlx5_fw_tracer_ownership_release(tracer);
+
+ mlx5_core_destroy_mkey(tracer->dev, &tracer->buff.mkey);
+ mlx5_core_dealloc_pd(tracer->dev, tracer->buff.pdn);
}
void mlx5_fw_tracer_destroy(struct mlx5_fw_tracer *tracer)
--
2.17.0
^ permalink raw reply related
* [net-next V2 02/12] net/mlx5: FW tracer, create trace buffer and copy strings database
From: Saeed Mahameed @ 2018-07-23 22:11 UTC (permalink / raw)
To: David S. Miller; +Cc: netdev, Feras Daoud, Saeed Mahameed
In-Reply-To: <20180723221129.21625-1-saeedm@mellanox.com>
From: Feras Daoud <ferasda@mellanox.com>
For each PF do the following:
1- Allocate memory for the tracer strings database and read the
strings from the FW to the SW. These strings will be used later for
parsing traces.
2- Allocate and dma map tracer buffers.
Traces that will be written into the buffer will be parsed as a group
of one or more traces, referred to as trace message. The trace message
represents a C-like printf string.
First trace of a message holds the pointer to the correct string in
strings database. The following traces holds the variables of the
message.
Signed-off-by: Feras Daoud <ferasda@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
.../mellanox/mlx5/core/diag/fw_tracer.c | 209 +++++++++++++++++-
.../mellanox/mlx5/core/diag/fw_tracer.h | 18 ++
2 files changed, 224 insertions(+), 3 deletions(-)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.c b/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.c
index 3ecbf06b4d71..35107b8f76df 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.c
@@ -119,6 +119,163 @@ static void mlx5_fw_tracer_ownership_release(struct mlx5_fw_tracer *tracer)
tracer->owner = false;
}
+static int mlx5_fw_tracer_create_log_buf(struct mlx5_fw_tracer *tracer)
+{
+ struct mlx5_core_dev *dev = tracer->dev;
+ struct device *ddev = &dev->pdev->dev;
+ dma_addr_t dma;
+ void *buff;
+ gfp_t gfp;
+ int err;
+
+ tracer->buff.size = TRACE_BUFFER_SIZE_BYTE;
+
+ gfp = GFP_KERNEL | __GFP_ZERO;
+ buff = (void *)__get_free_pages(gfp,
+ get_order(tracer->buff.size));
+ if (!buff) {
+ err = -ENOMEM;
+ mlx5_core_warn(dev, "FWTracer: Failed to allocate pages, %d\n", err);
+ return err;
+ }
+ tracer->buff.log_buf = buff;
+
+ dma = dma_map_single(ddev, buff, tracer->buff.size, DMA_FROM_DEVICE);
+ if (dma_mapping_error(ddev, dma)) {
+ mlx5_core_warn(dev, "FWTracer: Unable to map DMA: %d\n",
+ dma_mapping_error(ddev, dma));
+ err = -ENOMEM;
+ goto free_pages;
+ }
+ tracer->buff.dma = dma;
+
+ return 0;
+
+free_pages:
+ free_pages((unsigned long)tracer->buff.log_buf, get_order(tracer->buff.size));
+
+ return err;
+}
+
+static void mlx5_fw_tracer_destroy_log_buf(struct mlx5_fw_tracer *tracer)
+{
+ struct mlx5_core_dev *dev = tracer->dev;
+ struct device *ddev = &dev->pdev->dev;
+
+ if (!tracer->buff.log_buf)
+ return;
+
+ dma_unmap_single(ddev, tracer->buff.dma, tracer->buff.size, DMA_FROM_DEVICE);
+ free_pages((unsigned long)tracer->buff.log_buf, get_order(tracer->buff.size));
+}
+
+static void mlx5_fw_tracer_free_strings_db(struct mlx5_fw_tracer *tracer)
+{
+ u32 num_string_db = tracer->str_db.num_string_db;
+ int i;
+
+ for (i = 0; i < num_string_db; i++) {
+ kfree(tracer->str_db.buffer[i]);
+ tracer->str_db.buffer[i] = NULL;
+ }
+}
+
+static int mlx5_fw_tracer_allocate_strings_db(struct mlx5_fw_tracer *tracer)
+{
+ u32 *string_db_size_out = tracer->str_db.size_out;
+ u32 num_string_db = tracer->str_db.num_string_db;
+ int i;
+
+ for (i = 0; i < num_string_db; i++) {
+ tracer->str_db.buffer[i] = kzalloc(string_db_size_out[i], GFP_KERNEL);
+ if (!tracer->str_db.buffer[i])
+ goto free_strings_db;
+ }
+
+ return 0;
+
+free_strings_db:
+ mlx5_fw_tracer_free_strings_db(tracer);
+ return -ENOMEM;
+}
+
+static void mlx5_tracer_read_strings_db(struct work_struct *work)
+{
+ struct mlx5_fw_tracer *tracer = container_of(work, struct mlx5_fw_tracer,
+ read_fw_strings_work);
+ u32 num_of_reads, num_string_db = tracer->str_db.num_string_db;
+ struct mlx5_core_dev *dev = tracer->dev;
+ u32 in[MLX5_ST_SZ_DW(mtrc_cap)] = {0};
+ u32 leftovers, offset;
+ int err = 0, i, j;
+ u32 *out, outlen;
+ void *out_value;
+
+ outlen = MLX5_ST_SZ_BYTES(mtrc_stdb) + STRINGS_DB_READ_SIZE_BYTES;
+ out = kzalloc(outlen, GFP_KERNEL);
+ if (!out) {
+ err = -ENOMEM;
+ goto out;
+ }
+
+ for (i = 0; i < num_string_db; i++) {
+ offset = 0;
+ MLX5_SET(mtrc_stdb, in, string_db_index, i);
+ num_of_reads = tracer->str_db.size_out[i] /
+ STRINGS_DB_READ_SIZE_BYTES;
+ leftovers = (tracer->str_db.size_out[i] %
+ STRINGS_DB_READ_SIZE_BYTES) /
+ STRINGS_DB_LEFTOVER_SIZE_BYTES;
+
+ MLX5_SET(mtrc_stdb, in, read_size, STRINGS_DB_READ_SIZE_BYTES);
+ for (j = 0; j < num_of_reads; j++) {
+ MLX5_SET(mtrc_stdb, in, start_offset, offset);
+
+ err = mlx5_core_access_reg(dev, in, sizeof(in), out,
+ outlen, MLX5_REG_MTRC_STDB,
+ 0, 1);
+ if (err) {
+ mlx5_core_dbg(dev, "FWTracer: Failed to read strings DB %d\n",
+ err);
+ goto out_free;
+ }
+
+ out_value = MLX5_ADDR_OF(mtrc_stdb, out, string_db_data);
+ memcpy(tracer->str_db.buffer[i] + offset, out_value,
+ STRINGS_DB_READ_SIZE_BYTES);
+ offset += STRINGS_DB_READ_SIZE_BYTES;
+ }
+
+ /* Strings database is aligned to 64, need to read leftovers*/
+ MLX5_SET(mtrc_stdb, in, read_size,
+ STRINGS_DB_LEFTOVER_SIZE_BYTES);
+ for (j = 0; j < leftovers; j++) {
+ MLX5_SET(mtrc_stdb, in, start_offset, offset);
+
+ err = mlx5_core_access_reg(dev, in, sizeof(in), out,
+ outlen, MLX5_REG_MTRC_STDB,
+ 0, 1);
+ if (err) {
+ mlx5_core_dbg(dev, "FWTracer: Failed to read strings DB %d\n",
+ err);
+ goto out_free;
+ }
+
+ out_value = MLX5_ADDR_OF(mtrc_stdb, out, string_db_data);
+ memcpy(tracer->str_db.buffer[i] + offset, out_value,
+ STRINGS_DB_LEFTOVER_SIZE_BYTES);
+ offset += STRINGS_DB_LEFTOVER_SIZE_BYTES;
+ }
+ }
+
+ tracer->str_db.loaded = true;
+
+out_free:
+ kfree(out);
+out:
+ return;
+}
+
static void mlx5_fw_tracer_ownership_change(struct work_struct *work)
{
struct mlx5_fw_tracer *tracer = container_of(work, struct mlx5_fw_tracer,
@@ -161,6 +318,7 @@ struct mlx5_fw_tracer *mlx5_fw_tracer_create(struct mlx5_core_dev *dev)
tracer->dev = dev;
INIT_WORK(&tracer->ownership_change_work, mlx5_fw_tracer_ownership_change);
+ INIT_WORK(&tracer->read_fw_strings_work, mlx5_tracer_read_strings_db);
err = mlx5_query_mtrc_caps(tracer);
if (err) {
@@ -168,10 +326,22 @@ struct mlx5_fw_tracer *mlx5_fw_tracer_create(struct mlx5_core_dev *dev)
goto destroy_workqueue;
}
- mlx5_fw_tracer_ownership_change(&tracer->ownership_change_work);
+ err = mlx5_fw_tracer_create_log_buf(tracer);
+ if (err) {
+ mlx5_core_warn(dev, "FWTracer: Create log buffer failed %d\n", err);
+ goto destroy_workqueue;
+ }
+
+ err = mlx5_fw_tracer_allocate_strings_db(tracer);
+ if (err) {
+ mlx5_core_warn(dev, "FWTracer: Allocate strings database failed %d\n", err);
+ goto free_log_buf;
+ }
return tracer;
+free_log_buf:
+ mlx5_fw_tracer_destroy_log_buf(tracer);
destroy_workqueue:
tracer->dev = NULL;
destroy_workqueue(tracer->work_queue);
@@ -180,17 +350,50 @@ struct mlx5_fw_tracer *mlx5_fw_tracer_create(struct mlx5_core_dev *dev)
return ERR_PTR(err);
}
-void mlx5_fw_tracer_destroy(struct mlx5_fw_tracer *tracer)
+int mlx5_fw_tracer_init(struct mlx5_fw_tracer *tracer)
{
- if (!tracer)
+ struct mlx5_core_dev *dev;
+ int err;
+
+ if (IS_ERR_OR_NULL(tracer))
+ return 0;
+
+ dev = tracer->dev;
+
+ if (!tracer->str_db.loaded)
+ queue_work(tracer->work_queue, &tracer->read_fw_strings_work);
+
+ err = mlx5_fw_tracer_ownership_acquire(tracer);
+ if (err) {
+ mlx5_core_dbg(dev, "FWTracer: Ownership was not granted %d\n", err);
+ return 0; /* return 0 since ownership can be acquired on a later FW event */
+ }
+
+ return 0;
+}
+
+void mlx5_fw_tracer_cleanup(struct mlx5_fw_tracer *tracer)
+{
+ if (IS_ERR_OR_NULL(tracer))
return;
cancel_work_sync(&tracer->ownership_change_work);
if (tracer->owner)
mlx5_fw_tracer_ownership_release(tracer);
+}
+void mlx5_fw_tracer_destroy(struct mlx5_fw_tracer *tracer)
+{
+ if (IS_ERR_OR_NULL(tracer))
+ return;
+
+ cancel_work_sync(&tracer->read_fw_strings_work);
+ mlx5_fw_tracer_free_strings_db(tracer);
+ mlx5_fw_tracer_destroy_log_buf(tracer);
flush_workqueue(tracer->work_queue);
destroy_workqueue(tracer->work_queue);
kfree(tracer);
}
+
+EXPORT_TRACEPOINT_SYMBOL(mlx5_fw);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.h b/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.h
index 721c41a5e827..66cb7e7ada28 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.h
@@ -37,6 +37,11 @@
#include "mlx5_core.h"
#define STRINGS_DB_SECTIONS_NUM 8
+#define STRINGS_DB_READ_SIZE_BYTES 256
+#define STRINGS_DB_LEFTOVER_SIZE_BYTES 64
+#define TRACER_BUFFER_PAGE_NUM 64
+#define TRACER_BUFFER_CHUNK 4096
+#define TRACE_BUFFER_SIZE_BYTE (TRACER_BUFFER_PAGE_NUM * TRACER_BUFFER_CHUNK)
struct mlx5_fw_tracer {
struct mlx5_core_dev *dev;
@@ -44,6 +49,7 @@ struct mlx5_fw_tracer {
u8 trc_ver;
struct workqueue_struct *work_queue;
struct work_struct ownership_change_work;
+ struct work_struct read_fw_strings_work;
/* Strings DB */
struct {
@@ -52,7 +58,19 @@ struct mlx5_fw_tracer {
u32 num_string_db;
u32 base_address_out[STRINGS_DB_SECTIONS_NUM];
u32 size_out[STRINGS_DB_SECTIONS_NUM];
+ void *buffer[STRINGS_DB_SECTIONS_NUM];
+ bool loaded;
} str_db;
+
+ /* Log Buffer */
+ struct {
+ u32 pdn;
+ void *log_buf;
+ dma_addr_t dma;
+ u32 size;
+ struct mlx5_core_mkey mkey;
+
+ } buff;
};
enum mlx5_fw_tracer_ownership_state {
--
2.17.0
^ permalink raw reply related
* [net-next V2 05/12] net/mlx5: FW tracer, parse traces and kernel tracing support
From: Saeed Mahameed @ 2018-07-23 22:11 UTC (permalink / raw)
To: David S. Miller; +Cc: netdev, Feras Daoud, Erez Shitrit, Saeed Mahameed
In-Reply-To: <20180723221129.21625-1-saeedm@mellanox.com>
From: Feras Daoud <ferasda@mellanox.com>
For each message the driver should do the following:
1- Find the message string in the strings database
2- Count the param number of each message
3- Wait for the param events and accumulate them
4- Calculate the event timestamp using the local event timestamp
and the first timestamp event following it.
5- Print message to trace log
Enable the tracing by:
echo 1 > /sys/kernel/debug/tracing/events/mlx5/mlx5_fw/enable
Read traces by:
cat /sys/kernel/debug/tracing/trace
Signed-off-by: Feras Daoud <ferasda@mellanox.com>
Signed-off-by: Erez Shitrit <erezsh@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
.../mellanox/mlx5/core/diag/fw_tracer.c | 235 +++++++++++++++++-
.../mellanox/mlx5/core/diag/fw_tracer.h | 22 ++
.../mlx5/core/diag/fw_tracer_tracepoint.h | 78 ++++++
3 files changed, 333 insertions(+), 2 deletions(-)
create mode 100644 drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer_tracepoint.h
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.c b/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.c
index bd887d1d3396..309842de272c 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.c
@@ -29,8 +29,9 @@
* CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
* SOFTWARE.
*/
-
+#define CREATE_TRACE_POINTS
#include "fw_tracer.h"
+#include "fw_tracer_tracepoint.h"
static int mlx5_query_mtrc_caps(struct mlx5_fw_tracer *tracer)
{
@@ -332,6 +333,109 @@ static void mlx5_fw_tracer_arm(struct mlx5_core_dev *dev)
mlx5_core_warn(dev, "FWTracer: Failed to arm tracer event %d\n", err);
}
+static const char *VAL_PARM = "%llx";
+static const char *REPLACE_64_VAL_PARM = "%x%x";
+static const char *PARAM_CHAR = "%";
+
+static int mlx5_tracer_message_hash(u32 message_id)
+{
+ return jhash_1word(message_id, 0) & (MESSAGE_HASH_SIZE - 1);
+}
+
+static struct tracer_string_format *mlx5_tracer_message_insert(struct mlx5_fw_tracer *tracer,
+ struct tracer_event *tracer_event)
+{
+ struct hlist_head *head =
+ &tracer->hash[mlx5_tracer_message_hash(tracer_event->string_event.tmsn)];
+ struct tracer_string_format *cur_string;
+
+ cur_string = kzalloc(sizeof(*cur_string), GFP_KERNEL);
+ if (!cur_string)
+ return NULL;
+
+ hlist_add_head(&cur_string->hlist, head);
+
+ return cur_string;
+}
+
+static struct tracer_string_format *mlx5_tracer_get_string(struct mlx5_fw_tracer *tracer,
+ struct tracer_event *tracer_event)
+{
+ struct tracer_string_format *cur_string;
+ u32 str_ptr, offset;
+ int i;
+
+ str_ptr = tracer_event->string_event.string_param;
+
+ for (i = 0; i < tracer->str_db.num_string_db; i++) {
+ if (str_ptr > tracer->str_db.base_address_out[i] &&
+ str_ptr < tracer->str_db.base_address_out[i] +
+ tracer->str_db.size_out[i]) {
+ offset = str_ptr - tracer->str_db.base_address_out[i];
+ /* add it to the hash */
+ cur_string = mlx5_tracer_message_insert(tracer, tracer_event);
+ if (!cur_string)
+ return NULL;
+ cur_string->string = (char *)(tracer->str_db.buffer[i] +
+ offset);
+ return cur_string;
+ }
+ }
+
+ return NULL;
+}
+
+static void mlx5_tracer_clean_message(struct tracer_string_format *str_frmt)
+{
+ hlist_del(&str_frmt->hlist);
+ kfree(str_frmt);
+}
+
+static int mlx5_tracer_get_num_of_params(char *str)
+{
+ char *substr, *pstr = str;
+ int num_of_params = 0;
+
+ /* replace %llx with %x%x */
+ substr = strstr(pstr, VAL_PARM);
+ while (substr) {
+ memcpy(substr, REPLACE_64_VAL_PARM, 4);
+ pstr = substr;
+ substr = strstr(pstr, VAL_PARM);
+ }
+
+ /* count all the % characters */
+ substr = strstr(str, PARAM_CHAR);
+ while (substr) {
+ num_of_params += 1;
+ str = substr + 1;
+ substr = strstr(str, PARAM_CHAR);
+ }
+
+ return num_of_params;
+}
+
+static struct tracer_string_format *mlx5_tracer_message_find(struct hlist_head *head,
+ u8 event_id, u32 tmsn)
+{
+ struct tracer_string_format *message;
+
+ hlist_for_each_entry(message, head, hlist)
+ if (message->event_id == event_id && message->tmsn == tmsn)
+ return message;
+
+ return NULL;
+}
+
+static struct tracer_string_format *mlx5_tracer_message_get(struct mlx5_fw_tracer *tracer,
+ struct tracer_event *tracer_event)
+{
+ struct hlist_head *head =
+ &tracer->hash[mlx5_tracer_message_hash(tracer_event->string_event.tmsn)];
+
+ return mlx5_tracer_message_find(head, tracer_event->event_id, tracer_event->string_event.tmsn);
+}
+
static void poll_trace(struct mlx5_fw_tracer *tracer,
struct tracer_event *tracer_event, u64 *trace)
{
@@ -396,6 +500,128 @@ static u64 get_block_timestamp(struct mlx5_fw_tracer *tracer, u64 *ts_event)
return tracer_event.timestamp_event.timestamp;
}
+static void mlx5_fw_tracer_clean_print_hash(struct mlx5_fw_tracer *tracer)
+{
+ struct tracer_string_format *str_frmt;
+ struct hlist_node *n;
+ int i;
+
+ for (i = 0; i < MESSAGE_HASH_SIZE; i++) {
+ hlist_for_each_entry_safe(str_frmt, n, &tracer->hash[i], hlist)
+ mlx5_tracer_clean_message(str_frmt);
+ }
+}
+
+static void mlx5_fw_tracer_clean_ready_list(struct mlx5_fw_tracer *tracer)
+{
+ struct tracer_string_format *str_frmt, *tmp_str;
+
+ list_for_each_entry_safe(str_frmt, tmp_str, &tracer->ready_strings_list,
+ list)
+ list_del(&str_frmt->list);
+}
+
+static void mlx5_tracer_print_trace(struct tracer_string_format *str_frmt,
+ struct mlx5_core_dev *dev,
+ u64 trace_timestamp)
+{
+ char tmp[512];
+
+ snprintf(tmp, sizeof(tmp), str_frmt->string,
+ str_frmt->params[0],
+ str_frmt->params[1],
+ str_frmt->params[2],
+ str_frmt->params[3],
+ str_frmt->params[4],
+ str_frmt->params[5],
+ str_frmt->params[6]);
+
+ trace_mlx5_fw(dev->tracer, trace_timestamp, str_frmt->lost,
+ str_frmt->event_id, tmp);
+
+ /* remove it from hash */
+ mlx5_tracer_clean_message(str_frmt);
+}
+
+static int mlx5_tracer_handle_string_trace(struct mlx5_fw_tracer *tracer,
+ struct tracer_event *tracer_event)
+{
+ struct tracer_string_format *cur_string;
+
+ if (tracer_event->string_event.tdsn == 0) {
+ cur_string = mlx5_tracer_get_string(tracer, tracer_event);
+ if (!cur_string)
+ return -1;
+
+ cur_string->num_of_params = mlx5_tracer_get_num_of_params(cur_string->string);
+ cur_string->last_param_num = 0;
+ cur_string->event_id = tracer_event->event_id;
+ cur_string->tmsn = tracer_event->string_event.tmsn;
+ cur_string->timestamp = tracer_event->string_event.timestamp;
+ cur_string->lost = tracer_event->lost_event;
+ if (cur_string->num_of_params == 0) /* trace with no params */
+ list_add_tail(&cur_string->list, &tracer->ready_strings_list);
+ } else {
+ cur_string = mlx5_tracer_message_get(tracer, tracer_event);
+ if (!cur_string) {
+ pr_debug("%s Got string event for unknown string tdsm: %d\n",
+ __func__, tracer_event->string_event.tmsn);
+ return -1;
+ }
+ cur_string->last_param_num += 1;
+ if (cur_string->last_param_num > TRACER_MAX_PARAMS) {
+ pr_debug("%s Number of params exceeds the max (%d)\n",
+ __func__, TRACER_MAX_PARAMS);
+ list_add_tail(&cur_string->list, &tracer->ready_strings_list);
+ return 0;
+ }
+ /* keep the new parameter */
+ cur_string->params[cur_string->last_param_num - 1] =
+ tracer_event->string_event.string_param;
+ if (cur_string->last_param_num == cur_string->num_of_params)
+ list_add_tail(&cur_string->list, &tracer->ready_strings_list);
+ }
+
+ return 0;
+}
+
+static void mlx5_tracer_handle_timestamp_trace(struct mlx5_fw_tracer *tracer,
+ struct tracer_event *tracer_event)
+{
+ struct tracer_timestamp_event timestamp_event =
+ tracer_event->timestamp_event;
+ struct tracer_string_format *str_frmt, *tmp_str;
+ struct mlx5_core_dev *dev = tracer->dev;
+ u64 trace_timestamp;
+
+ list_for_each_entry_safe(str_frmt, tmp_str, &tracer->ready_strings_list, list) {
+ list_del(&str_frmt->list);
+ if (str_frmt->timestamp < (timestamp_event.timestamp & MASK_6_0))
+ trace_timestamp = (timestamp_event.timestamp & MASK_52_7) |
+ (str_frmt->timestamp & MASK_6_0);
+ else
+ trace_timestamp = ((timestamp_event.timestamp & MASK_52_7) - 1) |
+ (str_frmt->timestamp & MASK_6_0);
+
+ mlx5_tracer_print_trace(str_frmt, dev, trace_timestamp);
+ }
+}
+
+static int mlx5_tracer_handle_trace(struct mlx5_fw_tracer *tracer,
+ struct tracer_event *tracer_event)
+{
+ if (tracer_event->type == TRACER_EVENT_TYPE_STRING) {
+ mlx5_tracer_handle_string_trace(tracer, tracer_event);
+ } else if (tracer_event->type == TRACER_EVENT_TYPE_TIMESTAMP) {
+ if (!tracer_event->timestamp_event.unreliable)
+ mlx5_tracer_handle_timestamp_trace(tracer, tracer_event);
+ } else {
+ pr_debug("%s Got unrecognised type %d for parsing, exiting..\n",
+ __func__, tracer_event->type);
+ }
+ return 0;
+}
+
static void mlx5_fw_tracer_handle_traces(struct work_struct *work)
{
struct mlx5_fw_tracer *tracer =
@@ -452,8 +678,10 @@ static void mlx5_fw_tracer_handle_traces(struct work_struct *work)
}
/* Parse events */
- for (i = 0; i < TRACES_PER_BLOCK ; i++)
+ for (i = 0; i < TRACES_PER_BLOCK ; i++) {
poll_trace(tracer, &tracer_event, &tmp_trace_block[i]);
+ mlx5_tracer_handle_trace(tracer, &tracer_event);
+ }
tracer->buff.consumer_index =
(tracer->buff.consumer_index + 1) & (block_count - 1);
@@ -578,6 +806,7 @@ struct mlx5_fw_tracer *mlx5_fw_tracer_create(struct mlx5_core_dev *dev)
tracer->dev = dev;
+ INIT_LIST_HEAD(&tracer->ready_strings_list);
INIT_WORK(&tracer->ownership_change_work, mlx5_fw_tracer_ownership_change);
INIT_WORK(&tracer->read_fw_strings_work, mlx5_tracer_read_strings_db);
INIT_WORK(&tracer->handle_traces_work, mlx5_fw_tracer_handle_traces);
@@ -675,6 +904,8 @@ void mlx5_fw_tracer_destroy(struct mlx5_fw_tracer *tracer)
return;
cancel_work_sync(&tracer->read_fw_strings_work);
+ mlx5_fw_tracer_clean_ready_list(tracer);
+ mlx5_fw_tracer_clean_print_hash(tracer);
mlx5_fw_tracer_free_strings_db(tracer);
mlx5_fw_tracer_destroy_log_buf(tracer);
flush_workqueue(tracer->work_queue);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.h b/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.h
index 3915e91486b2..8d310e7d6743 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.h
@@ -46,6 +46,13 @@
#define TRACER_BLOCK_SIZE_BYTE 256
#define TRACES_PER_BLOCK 32
+#define TRACER_MAX_PARAMS 7
+#define MESSAGE_HASH_BITS 6
+#define MESSAGE_HASH_SIZE BIT(MESSAGE_HASH_BITS)
+
+#define MASK_52_7 (0x1FFFFFFFFFFF80)
+#define MASK_6_0 (0x7F)
+
struct mlx5_fw_tracer {
struct mlx5_core_dev *dev;
bool owner;
@@ -77,6 +84,21 @@ struct mlx5_fw_tracer {
u64 last_timestamp;
struct work_struct handle_traces_work;
+ struct hlist_head hash[MESSAGE_HASH_SIZE];
+ struct list_head ready_strings_list;
+};
+
+struct tracer_string_format {
+ char *string;
+ int params[TRACER_MAX_PARAMS];
+ int num_of_params;
+ int last_param_num;
+ u8 event_id;
+ u32 tmsn;
+ struct hlist_node hlist;
+ struct list_head list;
+ u32 timestamp;
+ bool lost;
};
enum mlx5_fw_tracer_ownership_state {
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer_tracepoint.h b/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer_tracepoint.h
new file mode 100644
index 000000000000..83f90e9aff45
--- /dev/null
+++ b/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer_tracepoint.h
@@ -0,0 +1,78 @@
+/*
+ * Copyright (c) 2018, Mellanox Technologies. All rights reserved.
+ *
+ * This software is available to you under a choice of one of two
+ * licenses. You may choose to be licensed under the terms of the GNU
+ * General Public License (GPL) Version 2, available from the file
+ * COPYING in the main directory of this source tree, or the
+ * OpenIB.org BSD license below:
+ *
+ * Redistribution and use in source and binary forms, with or
+ * without modification, are permitted provided that the following
+ * conditions are met:
+ *
+ * - Redistributions of source code must retain the above
+ * copyright notice, this list of conditions and the following
+ * disclaimer.
+ *
+ * - Redistributions in binary form must reproduce the above
+ * copyright notice, this list of conditions and the following
+ * disclaimer in the documentation and/or other materials
+ * provided with the distribution.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
+ * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
+ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ */
+
+#if !defined(__LIB_TRACER_TRACEPOINT_H__) || defined(TRACE_HEADER_MULTI_READ)
+#define __LIB_TRACER_TRACEPOINT_H__
+
+#include <linux/tracepoint.h>
+#include "fw_tracer.h"
+
+#undef TRACE_SYSTEM
+#define TRACE_SYSTEM mlx5
+
+/* Tracepoint for FWTracer messages: */
+TRACE_EVENT(mlx5_fw,
+ TP_PROTO(const struct mlx5_fw_tracer *tracer, u64 trace_timestamp,
+ bool lost, u8 event_id, const char *msg),
+
+ TP_ARGS(tracer, trace_timestamp, lost, event_id, msg),
+
+ TP_STRUCT__entry(
+ __string(dev_name, dev_name(&tracer->dev->pdev->dev))
+ __field(u64, trace_timestamp)
+ __field(bool, lost)
+ __field(u8, event_id)
+ __string(msg, msg)
+ ),
+
+ TP_fast_assign(
+ __assign_str(dev_name, dev_name(&tracer->dev->pdev->dev));
+ __entry->trace_timestamp = trace_timestamp;
+ __entry->lost = lost;
+ __entry->event_id = event_id;
+ __assign_str(msg, msg);
+ ),
+
+ TP_printk("%s [0x%llx] %d [0x%x] %s",
+ __get_str(dev_name),
+ __entry->trace_timestamp,
+ __entry->lost, __entry->event_id,
+ __get_str(msg))
+);
+
+#endif
+
+#undef TRACE_INCLUDE_PATH
+#undef TRACE_INCLUDE_FILE
+#define TRACE_INCLUDE_PATH ./diag
+#define TRACE_INCLUDE_FILE fw_tracer_tracepoint
+#include <trace/define_trace.h>
--
2.17.0
^ permalink raw reply related
* [net-next V2 04/12] net/mlx5: FW tracer, events handling
From: Saeed Mahameed @ 2018-07-23 22:11 UTC (permalink / raw)
To: David S. Miller; +Cc: netdev, Feras Daoud, Saeed Mahameed
In-Reply-To: <20180723221129.21625-1-saeedm@mellanox.com>
From: Feras Daoud <ferasda@mellanox.com>
The tracer has one event, event 0x26, with two subtypes:
- Subtype 0: Ownership change
- Subtype 1: Traces available
An ownership change occurs in the following cases:
1- Owner releases his ownership, in this case, an event will be
sent to inform others to reattempt acquire ownership.
2- Ownership was taken by a higher priority tool, in this case
the owner should understand that it lost ownership, and go through
tear down flow.
The second subtype indicates that there are traces in the trace buffer,
in this case, the driver polls the tracer buffer for new traces, parse
them and prepares the messages for printing.
The HW starts tracing from the first address in the tracer buffer.
Driver receives an event notifying that new trace block exists.
HW posts a timestamp event at the last 8B of every 256B block.
Comparing the timestamp to the last handled timestamp would indicate
that this is a new trace block. Once the new timestamp is detected,
the entire block is considered valid.
Block validation and parsing, should be done after copying the current
block to a different location, in order to avoid block overwritten
during processing.
Signed-off-by: Feras Daoud <ferasda@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
.../mellanox/mlx5/core/diag/fw_tracer.c | 268 +++++++++++++++++-
.../mellanox/mlx5/core/diag/fw_tracer.h | 71 ++++-
drivers/net/ethernet/mellanox/mlx5/core/eq.c | 11 +
include/linux/mlx5/device.h | 7 +
4 files changed, 347 insertions(+), 10 deletions(-)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.c b/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.c
index d6cc27b0ff34..bd887d1d3396 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.c
@@ -318,25 +318,244 @@ static void mlx5_tracer_read_strings_db(struct work_struct *work)
return;
}
-static void mlx5_fw_tracer_ownership_change(struct work_struct *work)
+static void mlx5_fw_tracer_arm(struct mlx5_core_dev *dev)
{
- struct mlx5_fw_tracer *tracer = container_of(work, struct mlx5_fw_tracer,
- ownership_change_work);
- struct mlx5_core_dev *dev = tracer->dev;
+ u32 out[MLX5_ST_SZ_DW(mtrc_ctrl)] = {0};
+ u32 in[MLX5_ST_SZ_DW(mtrc_ctrl)] = {0};
int err;
- if (tracer->owner) {
- mlx5_fw_tracer_ownership_release(tracer);
+ MLX5_SET(mtrc_ctrl, in, arm_event, 1);
+
+ err = mlx5_core_access_reg(dev, in, sizeof(in), out, sizeof(out),
+ MLX5_REG_MTRC_CTRL, 0, 1);
+ if (err)
+ mlx5_core_warn(dev, "FWTracer: Failed to arm tracer event %d\n", err);
+}
+
+static void poll_trace(struct mlx5_fw_tracer *tracer,
+ struct tracer_event *tracer_event, u64 *trace)
+{
+ u32 timestamp_low, timestamp_mid, timestamp_high, urts;
+
+ tracer_event->event_id = MLX5_GET(tracer_event, trace, event_id);
+ tracer_event->lost_event = MLX5_GET(tracer_event, trace, lost);
+
+ switch (tracer_event->event_id) {
+ case TRACER_EVENT_TYPE_TIMESTAMP:
+ tracer_event->type = TRACER_EVENT_TYPE_TIMESTAMP;
+ urts = MLX5_GET(tracer_timestamp_event, trace, urts);
+ if (tracer->trc_ver == 0)
+ tracer_event->timestamp_event.unreliable = !!(urts >> 2);
+ else
+ tracer_event->timestamp_event.unreliable = !!(urts & 1);
+
+ timestamp_low = MLX5_GET(tracer_timestamp_event,
+ trace, timestamp7_0);
+ timestamp_mid = MLX5_GET(tracer_timestamp_event,
+ trace, timestamp39_8);
+ timestamp_high = MLX5_GET(tracer_timestamp_event,
+ trace, timestamp52_40);
+
+ tracer_event->timestamp_event.timestamp =
+ ((u64)timestamp_high << 40) |
+ ((u64)timestamp_mid << 8) |
+ (u64)timestamp_low;
+ break;
+ default:
+ if (tracer_event->event_id >= tracer->str_db.first_string_trace ||
+ tracer_event->event_id <= tracer->str_db.first_string_trace +
+ tracer->str_db.num_string_trace) {
+ tracer_event->type = TRACER_EVENT_TYPE_STRING;
+ tracer_event->string_event.timestamp =
+ MLX5_GET(tracer_string_event, trace, timestamp);
+ tracer_event->string_event.string_param =
+ MLX5_GET(tracer_string_event, trace, string_param);
+ tracer_event->string_event.tmsn =
+ MLX5_GET(tracer_string_event, trace, tmsn);
+ tracer_event->string_event.tdsn =
+ MLX5_GET(tracer_string_event, trace, tdsn);
+ } else {
+ tracer_event->type = TRACER_EVENT_TYPE_UNRECOGNIZED;
+ }
+ break;
+ }
+}
+
+static u64 get_block_timestamp(struct mlx5_fw_tracer *tracer, u64 *ts_event)
+{
+ struct tracer_event tracer_event;
+ u8 event_id;
+
+ event_id = MLX5_GET(tracer_event, ts_event, event_id);
+
+ if (event_id == TRACER_EVENT_TYPE_TIMESTAMP)
+ poll_trace(tracer, &tracer_event, ts_event);
+ else
+ tracer_event.timestamp_event.timestamp = 0;
+
+ return tracer_event.timestamp_event.timestamp;
+}
+
+static void mlx5_fw_tracer_handle_traces(struct work_struct *work)
+{
+ struct mlx5_fw_tracer *tracer =
+ container_of(work, struct mlx5_fw_tracer, handle_traces_work);
+ u64 block_timestamp, last_block_timestamp, tmp_trace_block[TRACES_PER_BLOCK];
+ u32 block_count, start_offset, prev_start_offset, prev_consumer_index;
+ u32 trace_event_size = MLX5_ST_SZ_BYTES(tracer_event);
+ struct tracer_event tracer_event;
+ struct mlx5_core_dev *dev;
+ int i;
+
+ if (!tracer->owner)
return;
+
+ dev = tracer->dev;
+ block_count = tracer->buff.size / TRACER_BLOCK_SIZE_BYTE;
+ start_offset = tracer->buff.consumer_index * TRACER_BLOCK_SIZE_BYTE;
+
+ /* Copy the block to local buffer to avoid HW override while being processed*/
+ memcpy(tmp_trace_block, tracer->buff.log_buf + start_offset,
+ TRACER_BLOCK_SIZE_BYTE);
+
+ block_timestamp =
+ get_block_timestamp(tracer, &tmp_trace_block[TRACES_PER_BLOCK - 1]);
+
+ while (block_timestamp > tracer->last_timestamp) {
+ /* Check block override if its not the first block */
+ if (!tracer->last_timestamp) {
+ u64 *ts_event;
+ /* To avoid block override be the HW in case of buffer
+ * wraparound, the time stamp of the previous block
+ * should be compared to the last timestamp handled
+ * by the driver.
+ */
+ prev_consumer_index =
+ (tracer->buff.consumer_index - 1) & (block_count - 1);
+ prev_start_offset = prev_consumer_index * TRACER_BLOCK_SIZE_BYTE;
+
+ ts_event = tracer->buff.log_buf + prev_start_offset +
+ (TRACES_PER_BLOCK - 1) * trace_event_size;
+ last_block_timestamp = get_block_timestamp(tracer, ts_event);
+ /* If previous timestamp different from last stored
+ * timestamp then there is a good chance that the
+ * current buffer is overwritten and therefore should
+ * not be parsed.
+ */
+ if (tracer->last_timestamp != last_block_timestamp) {
+ mlx5_core_warn(dev, "FWTracer: Events were lost\n");
+ tracer->last_timestamp = block_timestamp;
+ tracer->buff.consumer_index =
+ (tracer->buff.consumer_index + 1) & (block_count - 1);
+ break;
+ }
+ }
+
+ /* Parse events */
+ for (i = 0; i < TRACES_PER_BLOCK ; i++)
+ poll_trace(tracer, &tracer_event, &tmp_trace_block[i]);
+
+ tracer->buff.consumer_index =
+ (tracer->buff.consumer_index + 1) & (block_count - 1);
+
+ tracer->last_timestamp = block_timestamp;
+ start_offset = tracer->buff.consumer_index * TRACER_BLOCK_SIZE_BYTE;
+ memcpy(tmp_trace_block, tracer->buff.log_buf + start_offset,
+ TRACER_BLOCK_SIZE_BYTE);
+ block_timestamp = get_block_timestamp(tracer,
+ &tmp_trace_block[TRACES_PER_BLOCK - 1]);
}
+ mlx5_fw_tracer_arm(dev);
+}
+
+static int mlx5_fw_tracer_set_mtrc_conf(struct mlx5_fw_tracer *tracer)
+{
+ struct mlx5_core_dev *dev = tracer->dev;
+ u32 out[MLX5_ST_SZ_DW(mtrc_conf)] = {0};
+ u32 in[MLX5_ST_SZ_DW(mtrc_conf)] = {0};
+ int err;
+
+ MLX5_SET(mtrc_conf, in, trace_mode, TRACE_TO_MEMORY);
+ MLX5_SET(mtrc_conf, in, log_trace_buffer_size,
+ ilog2(TRACER_BUFFER_PAGE_NUM));
+ MLX5_SET(mtrc_conf, in, trace_mkey, tracer->buff.mkey.key);
+
+ err = mlx5_core_access_reg(dev, in, sizeof(in), out, sizeof(out),
+ MLX5_REG_MTRC_CONF, 0, 1);
+ if (err)
+ mlx5_core_warn(dev, "FWTracer: Failed to set tracer configurations %d\n", err);
+
+ return err;
+}
+
+static int mlx5_fw_tracer_set_mtrc_ctrl(struct mlx5_fw_tracer *tracer, u8 status, u8 arm)
+{
+ struct mlx5_core_dev *dev = tracer->dev;
+ u32 out[MLX5_ST_SZ_DW(mtrc_ctrl)] = {0};
+ u32 in[MLX5_ST_SZ_DW(mtrc_ctrl)] = {0};
+ int err;
+
+ MLX5_SET(mtrc_ctrl, in, modify_field_select, TRACE_STATUS);
+ MLX5_SET(mtrc_ctrl, in, trace_status, status);
+ MLX5_SET(mtrc_ctrl, in, arm_event, arm);
+
+ err = mlx5_core_access_reg(dev, in, sizeof(in), out, sizeof(out),
+ MLX5_REG_MTRC_CTRL, 0, 1);
+
+ if (!err && status)
+ tracer->last_timestamp = 0;
+
+ return err;
+}
+
+static int mlx5_fw_tracer_start(struct mlx5_fw_tracer *tracer)
+{
+ struct mlx5_core_dev *dev = tracer->dev;
+ int err;
+
err = mlx5_fw_tracer_ownership_acquire(tracer);
if (err) {
mlx5_core_dbg(dev, "FWTracer: Ownership was not granted %d\n", err);
+ /* Don't fail since ownership can be acquired on a later FW event */
+ return 0;
+ }
+
+ err = mlx5_fw_tracer_set_mtrc_conf(tracer);
+ if (err) {
+ mlx5_core_warn(dev, "FWTracer: Failed to set tracer configuration %d\n", err);
+ goto release_ownership;
+ }
+
+ /* enable tracer & trace events */
+ err = mlx5_fw_tracer_set_mtrc_ctrl(tracer, 1, 1);
+ if (err) {
+ mlx5_core_warn(dev, "FWTracer: Failed to enable tracer %d\n", err);
+ goto release_ownership;
+ }
+
+ return 0;
+
+release_ownership:
+ mlx5_fw_tracer_ownership_release(tracer);
+ return err;
+}
+
+static void mlx5_fw_tracer_ownership_change(struct work_struct *work)
+{
+ struct mlx5_fw_tracer *tracer =
+ container_of(work, struct mlx5_fw_tracer, ownership_change_work);
+
+ if (tracer->owner) {
+ tracer->owner = false;
+ tracer->buff.consumer_index = 0;
return;
}
+
+ mlx5_fw_tracer_start(tracer);
}
+/* Create software resources (Buffers, etc ..) */
struct mlx5_fw_tracer *mlx5_fw_tracer_create(struct mlx5_core_dev *dev)
{
struct mlx5_fw_tracer *tracer = NULL;
@@ -361,6 +580,8 @@ struct mlx5_fw_tracer *mlx5_fw_tracer_create(struct mlx5_core_dev *dev)
INIT_WORK(&tracer->ownership_change_work, mlx5_fw_tracer_ownership_change);
INIT_WORK(&tracer->read_fw_strings_work, mlx5_tracer_read_strings_db);
+ INIT_WORK(&tracer->handle_traces_work, mlx5_fw_tracer_handle_traces);
+
err = mlx5_query_mtrc_caps(tracer);
if (err) {
@@ -392,6 +613,9 @@ struct mlx5_fw_tracer *mlx5_fw_tracer_create(struct mlx5_core_dev *dev)
return ERR_PTR(err);
}
+/* Create HW resources + start tracer
+ * must be called before Async EQ is created
+ */
int mlx5_fw_tracer_init(struct mlx5_fw_tracer *tracer)
{
struct mlx5_core_dev *dev;
@@ -417,22 +641,25 @@ int mlx5_fw_tracer_init(struct mlx5_fw_tracer *tracer)
goto err_dealloc_pd;
}
- err = mlx5_fw_tracer_ownership_acquire(tracer);
- if (err) /* Don't fail since ownership can be acquired on a later FW event */
- mlx5_core_dbg(dev, "FWTracer: Ownership was not granted %d\n", err);
+ mlx5_fw_tracer_start(tracer);
return 0;
+
err_dealloc_pd:
mlx5_core_dealloc_pd(dev, tracer->buff.pdn);
return err;
}
+/* Stop tracer + Cleanup HW resources
+ * must be called after Async EQ is destroyed
+ */
void mlx5_fw_tracer_cleanup(struct mlx5_fw_tracer *tracer)
{
if (IS_ERR_OR_NULL(tracer))
return;
cancel_work_sync(&tracer->ownership_change_work);
+ cancel_work_sync(&tracer->handle_traces_work);
if (tracer->owner)
mlx5_fw_tracer_ownership_release(tracer);
@@ -441,6 +668,7 @@ void mlx5_fw_tracer_cleanup(struct mlx5_fw_tracer *tracer)
mlx5_core_dealloc_pd(tracer->dev, tracer->buff.pdn);
}
+/* Free software resources (Buffers, etc ..) */
void mlx5_fw_tracer_destroy(struct mlx5_fw_tracer *tracer)
{
if (IS_ERR_OR_NULL(tracer))
@@ -454,4 +682,26 @@ void mlx5_fw_tracer_destroy(struct mlx5_fw_tracer *tracer)
kfree(tracer);
}
+void mlx5_fw_tracer_event(struct mlx5_core_dev *dev, struct mlx5_eqe *eqe)
+{
+ struct mlx5_fw_tracer *tracer = dev->tracer;
+
+ if (!tracer)
+ return;
+
+ switch (eqe->sub_type) {
+ case MLX5_TRACER_SUBTYPE_OWNERSHIP_CHANGE:
+ if (test_bit(MLX5_INTERFACE_STATE_UP, &dev->intf_state))
+ queue_work(tracer->work_queue, &tracer->ownership_change_work);
+ break;
+ case MLX5_TRACER_SUBTYPE_TRACES_AVAILABLE:
+ if (likely(tracer->str_db.loaded))
+ queue_work(tracer->work_queue, &tracer->handle_traces_work);
+ break;
+ default:
+ mlx5_core_dbg(dev, "FWTracer: Event with unrecognized subtype: sub_type %d\n",
+ eqe->sub_type);
+ }
+}
+
EXPORT_TRACEPOINT_SYMBOL(mlx5_fw);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.h b/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.h
index 66cb7e7ada28..3915e91486b2 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.h
@@ -43,6 +43,9 @@
#define TRACER_BUFFER_CHUNK 4096
#define TRACE_BUFFER_SIZE_BYTE (TRACER_BUFFER_PAGE_NUM * TRACER_BUFFER_CHUNK)
+#define TRACER_BLOCK_SIZE_BYTE 256
+#define TRACES_PER_BLOCK 32
+
struct mlx5_fw_tracer {
struct mlx5_core_dev *dev;
bool owner;
@@ -69,8 +72,11 @@ struct mlx5_fw_tracer {
dma_addr_t dma;
u32 size;
struct mlx5_core_mkey mkey;
-
+ u32 consumer_index;
} buff;
+
+ u64 last_timestamp;
+ struct work_struct handle_traces_work;
};
enum mlx5_fw_tracer_ownership_state {
@@ -78,7 +84,70 @@ enum mlx5_fw_tracer_ownership_state {
MLX5_FW_TRACER_ACQUIRE_OWNERSHIP,
};
+enum tracer_ctrl_fields_select {
+ TRACE_STATUS = 1 << 0,
+};
+
+enum tracer_event_type {
+ TRACER_EVENT_TYPE_STRING,
+ TRACER_EVENT_TYPE_TIMESTAMP = 0xFF,
+ TRACER_EVENT_TYPE_UNRECOGNIZED,
+};
+
+enum tracing_mode {
+ TRACE_TO_MEMORY = 1 << 0,
+};
+
+struct tracer_timestamp_event {
+ u64 timestamp;
+ u8 unreliable;
+};
+
+struct tracer_string_event {
+ u32 timestamp;
+ u32 tmsn;
+ u32 tdsn;
+ u32 string_param;
+};
+
+struct tracer_event {
+ bool lost_event;
+ u32 type;
+ u8 event_id;
+ union {
+ struct tracer_string_event string_event;
+ struct tracer_timestamp_event timestamp_event;
+ };
+};
+
+struct mlx5_ifc_tracer_event_bits {
+ u8 lost[0x1];
+ u8 timestamp[0x7];
+ u8 event_id[0x8];
+ u8 event_data[0x30];
+};
+
+struct mlx5_ifc_tracer_string_event_bits {
+ u8 lost[0x1];
+ u8 timestamp[0x7];
+ u8 event_id[0x8];
+ u8 tmsn[0xd];
+ u8 tdsn[0x3];
+ u8 string_param[0x20];
+};
+
+struct mlx5_ifc_tracer_timestamp_event_bits {
+ u8 timestamp7_0[0x8];
+ u8 event_id[0x8];
+ u8 urts[0x3];
+ u8 timestamp52_40[0xd];
+ u8 timestamp39_8[0x20];
+};
+
struct mlx5_fw_tracer *mlx5_fw_tracer_create(struct mlx5_core_dev *dev);
+int mlx5_fw_tracer_init(struct mlx5_fw_tracer *tracer);
+void mlx5_fw_tracer_cleanup(struct mlx5_fw_tracer *tracer);
void mlx5_fw_tracer_destroy(struct mlx5_fw_tracer *tracer);
+void mlx5_fw_tracer_event(struct mlx5_core_dev *dev, struct mlx5_eqe *eqe) { return; }
#endif
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eq.c b/drivers/net/ethernet/mellanox/mlx5/core/eq.c
index 406c23862f5f..7669b4380779 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eq.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eq.c
@@ -40,6 +40,7 @@
#include "mlx5_core.h"
#include "fpga/core.h"
#include "eswitch.h"
+#include "diag/fw_tracer.h"
enum {
MLX5_EQE_SIZE = sizeof(struct mlx5_eqe),
@@ -168,6 +169,8 @@ static const char *eqe_type_str(u8 type)
return "MLX5_EVENT_TYPE_FPGA_QP_ERROR";
case MLX5_EVENT_TYPE_GENERAL_EVENT:
return "MLX5_EVENT_TYPE_GENERAL_EVENT";
+ case MLX5_EVENT_TYPE_DEVICE_TRACER:
+ return "MLX5_EVENT_TYPE_DEVICE_TRACER";
default:
return "Unrecognized event";
}
@@ -576,6 +579,11 @@ static irqreturn_t mlx5_eq_int(int irq, void *eq_ptr)
case MLX5_EVENT_TYPE_GENERAL_EVENT:
general_event_handler(dev, eqe);
break;
+
+ case MLX5_EVENT_TYPE_DEVICE_TRACER:
+ mlx5_fw_tracer_event(dev, eqe);
+ break;
+
default:
mlx5_core_warn(dev, "Unhandled event 0x%x on EQ 0x%x\n",
eqe->type, eq->eqn);
@@ -853,6 +861,9 @@ int mlx5_start_eqs(struct mlx5_core_dev *dev)
if (MLX5_CAP_GEN(dev, temp_warn_event))
async_event_mask |= (1ull << MLX5_EVENT_TYPE_TEMP_WARN_EVENT);
+ if (MLX5_CAP_MCAM_REG(dev, tracer_registers))
+ async_event_mask |= (1ull << MLX5_EVENT_TYPE_DEVICE_TRACER);
+
err = mlx5_create_map_eq(dev, &table->cmd_eq, MLX5_EQ_VEC_CMD,
MLX5_NUM_CMD_EQE, 1ull << MLX5_EVENT_TYPE_CMD,
"mlx5_cmd_eq", MLX5_EQ_TYPE_ASYNC);
diff --git a/include/linux/mlx5/device.h b/include/linux/mlx5/device.h
index 0566c6a94805..d489494b0a84 100644
--- a/include/linux/mlx5/device.h
+++ b/include/linux/mlx5/device.h
@@ -332,6 +332,13 @@ enum mlx5_event {
MLX5_EVENT_TYPE_FPGA_ERROR = 0x20,
MLX5_EVENT_TYPE_FPGA_QP_ERROR = 0x21,
+
+ MLX5_EVENT_TYPE_DEVICE_TRACER = 0x26,
+};
+
+enum {
+ MLX5_TRACER_SUBTYPE_OWNERSHIP_CHANGE = 0x0,
+ MLX5_TRACER_SUBTYPE_TRACES_AVAILABLE = 0x1,
};
enum {
--
2.17.0
^ permalink raw reply related
* [net-next V2 07/12] net/mlx5: FW tracer, Add debug prints
From: Saeed Mahameed @ 2018-07-23 22:11 UTC (permalink / raw)
To: David S. Miller; +Cc: netdev, Saeed Mahameed
In-Reply-To: <20180723221129.21625-1-saeedm@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
.../ethernet/mellanox/mlx5/core/diag/fw_tracer.c | 13 +++++++++++--
1 file changed, 11 insertions(+), 2 deletions(-)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.c b/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.c
index 309842de272c..d4ec93bde4de 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.c
@@ -629,14 +629,14 @@ static void mlx5_fw_tracer_handle_traces(struct work_struct *work)
u64 block_timestamp, last_block_timestamp, tmp_trace_block[TRACES_PER_BLOCK];
u32 block_count, start_offset, prev_start_offset, prev_consumer_index;
u32 trace_event_size = MLX5_ST_SZ_BYTES(tracer_event);
+ struct mlx5_core_dev *dev = tracer->dev;
struct tracer_event tracer_event;
- struct mlx5_core_dev *dev;
int i;
+ mlx5_core_dbg(dev, "FWTracer: Handle Trace event, owner=(%d)\n", tracer->owner);
if (!tracer->owner)
return;
- dev = tracer->dev;
block_count = tracer->buff.size / TRACER_BLOCK_SIZE_BYTE;
start_offset = tracer->buff.consumer_index * TRACER_BLOCK_SIZE_BYTE;
@@ -762,6 +762,7 @@ static int mlx5_fw_tracer_start(struct mlx5_fw_tracer *tracer)
goto release_ownership;
}
+ mlx5_core_dbg(dev, "FWTracer: Ownership granted and active\n");
return 0;
release_ownership:
@@ -774,6 +775,7 @@ static void mlx5_fw_tracer_ownership_change(struct work_struct *work)
struct mlx5_fw_tracer *tracer =
container_of(work, struct mlx5_fw_tracer, ownership_change_work);
+ mlx5_core_dbg(tracer->dev, "FWTracer: ownership changed, current=(%d)\n", tracer->owner);
if (tracer->owner) {
tracer->owner = false;
tracer->buff.consumer_index = 0;
@@ -830,6 +832,8 @@ struct mlx5_fw_tracer *mlx5_fw_tracer_create(struct mlx5_core_dev *dev)
goto free_log_buf;
}
+ mlx5_core_dbg(dev, "FWTracer: Tracer created\n");
+
return tracer;
free_log_buf:
@@ -887,6 +891,9 @@ void mlx5_fw_tracer_cleanup(struct mlx5_fw_tracer *tracer)
if (IS_ERR_OR_NULL(tracer))
return;
+ mlx5_core_dbg(tracer->dev, "FWTracer: Cleanup, is owner ? (%d)\n",
+ tracer->owner);
+
cancel_work_sync(&tracer->ownership_change_work);
cancel_work_sync(&tracer->handle_traces_work);
@@ -903,6 +910,8 @@ void mlx5_fw_tracer_destroy(struct mlx5_fw_tracer *tracer)
if (IS_ERR_OR_NULL(tracer))
return;
+ mlx5_core_dbg(tracer->dev, "FWTracer: Destroy\n");
+
cancel_work_sync(&tracer->read_fw_strings_work);
mlx5_fw_tracer_clean_ready_list(tracer);
mlx5_fw_tracer_clean_print_hash(tracer);
--
2.17.0
^ permalink raw reply related
* [net-next V2 06/12] net/mlx5: FW tracer, Enable tracing
From: Saeed Mahameed @ 2018-07-23 22:11 UTC (permalink / raw)
To: David S. Miller; +Cc: netdev, Feras Daoud, Saeed Mahameed
In-Reply-To: <20180723221129.21625-1-saeedm@mellanox.com>
From: Feras Daoud <ferasda@mellanox.com>
Add the tracer file to the makefile and add the init
function to the load one flow.
Signed-off-by: Feras Daoud <ferasda@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
.../net/ethernet/mellanox/mlx5/core/Makefile | 2 +-
.../mellanox/mlx5/core/diag/fw_tracer.h | 2 +-
drivers/net/ethernet/mellanox/mlx5/core/main.c | 18 ++++++++++++++++--
3 files changed, 18 insertions(+), 4 deletions(-)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/Makefile b/drivers/net/ethernet/mellanox/mlx5/core/Makefile
index d923f2f58608..55d5a5c2e9d8 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/Makefile
+++ b/drivers/net/ethernet/mellanox/mlx5/core/Makefile
@@ -6,7 +6,7 @@ mlx5_core-y := main.o cmd.o debugfs.o fw.o eq.o uar.o pagealloc.o \
health.o mcg.o cq.o srq.o alloc.o qp.o port.o mr.o pd.o \
mad.o transobj.o vport.o sriov.o fs_cmd.o fs_core.o \
fs_counters.o rl.o lag.o dev.o wq.o lib/gid.o lib/clock.o \
- diag/fs_tracepoint.o
+ diag/fs_tracepoint.o diag/fw_tracer.o
mlx5_core-$(CONFIG_MLX5_ACCEL) += accel/ipsec.o accel/tls.o
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.h b/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.h
index 8d310e7d6743..0347f2dd5cee 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.h
@@ -170,6 +170,6 @@ struct mlx5_fw_tracer *mlx5_fw_tracer_create(struct mlx5_core_dev *dev);
int mlx5_fw_tracer_init(struct mlx5_fw_tracer *tracer);
void mlx5_fw_tracer_cleanup(struct mlx5_fw_tracer *tracer);
void mlx5_fw_tracer_destroy(struct mlx5_fw_tracer *tracer);
-void mlx5_fw_tracer_event(struct mlx5_core_dev *dev, struct mlx5_eqe *eqe) { return; }
+void mlx5_fw_tracer_event(struct mlx5_core_dev *dev, struct mlx5_eqe *eqe);
#endif
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/main.c b/drivers/net/ethernet/mellanox/mlx5/core/main.c
index f9b950e1bd85..6ddbb70e95de 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/main.c
@@ -62,6 +62,7 @@
#include "accel/ipsec.h"
#include "accel/tls.h"
#include "lib/clock.h"
+#include "diag/fw_tracer.h"
MODULE_AUTHOR("Eli Cohen <eli@mellanox.com>");
MODULE_DESCRIPTION("Mellanox 5th generation network adapters (ConnectX series) core driver");
@@ -990,6 +991,8 @@ static int mlx5_init_once(struct mlx5_core_dev *dev, struct mlx5_priv *priv)
goto err_sriov_cleanup;
}
+ dev->tracer = mlx5_fw_tracer_create(dev);
+
return 0;
err_sriov_cleanup:
@@ -1015,6 +1018,7 @@ static int mlx5_init_once(struct mlx5_core_dev *dev, struct mlx5_priv *priv)
static void mlx5_cleanup_once(struct mlx5_core_dev *dev)
{
+ mlx5_fw_tracer_destroy(dev->tracer);
mlx5_fpga_cleanup(dev);
mlx5_sriov_cleanup(dev);
mlx5_eswitch_cleanup(dev->priv.eswitch);
@@ -1167,10 +1171,16 @@ static int mlx5_load_one(struct mlx5_core_dev *dev, struct mlx5_priv *priv,
goto err_put_uars;
}
+ err = mlx5_fw_tracer_init(dev->tracer);
+ if (err) {
+ dev_err(&pdev->dev, "Failed to init FW tracer\n");
+ goto err_fw_tracer;
+ }
+
err = alloc_comp_eqs(dev);
if (err) {
dev_err(&pdev->dev, "Failed to alloc completion EQs\n");
- goto err_stop_eqs;
+ goto err_comp_eqs;
}
err = mlx5_irq_set_affinity_hints(dev);
@@ -1252,7 +1262,10 @@ static int mlx5_load_one(struct mlx5_core_dev *dev, struct mlx5_priv *priv,
err_affinity_hints:
free_comp_eqs(dev);
-err_stop_eqs:
+err_comp_eqs:
+ mlx5_fw_tracer_cleanup(dev->tracer);
+
+err_fw_tracer:
mlx5_stop_eqs(dev);
err_put_uars:
@@ -1320,6 +1333,7 @@ static int mlx5_unload_one(struct mlx5_core_dev *dev, struct mlx5_priv *priv,
mlx5_fpga_device_stop(dev);
mlx5_irq_clear_affinity_hints(dev);
free_comp_eqs(dev);
+ mlx5_fw_tracer_cleanup(dev->tracer);
mlx5_stop_eqs(dev);
mlx5_put_uars_page(dev, priv->uar);
mlx5_free_irq_vectors(dev);
--
2.17.0
^ permalink raw reply related
* [net-next V2 08/12] net/mlx5e: Remove redundant WARN when we cannot find neigh entry
From: Saeed Mahameed @ 2018-07-23 22:11 UTC (permalink / raw)
To: David S. Miller; +Cc: netdev, Roi Dayan, Saeed Mahameed
In-Reply-To: <20180723221129.21625-1-saeedm@mellanox.com>
From: Roi Dayan <roid@mellanox.com>
It is possible for neigh entry not to exist if it was cleaned already.
When we bring down an interface the neigh gets deleted but it could be
that our listener for neigh event to clear the encap valid bit didn't
start yet and the neigh update last used work is started first.
In this scenario the encap entry has valid bit set but the neigh entry
doesn't exist.
Signed-off-by: Roi Dayan <roid@mellanox.com>
Reviewed-by: Paul Blakey <paulb@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
drivers/net/ethernet/mellanox/mlx5/core/en_tc.c | 4 +---
1 file changed, 1 insertion(+), 3 deletions(-)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
index 0edf4751a8ba..335a08bc381d 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
@@ -1032,10 +1032,8 @@ void mlx5e_tc_update_neigh_used_value(struct mlx5e_neigh_hash_entry *nhe)
* dst ip pair
*/
n = neigh_lookup(tbl, &m_neigh->dst_ip, m_neigh->dev);
- if (!n) {
- WARN(1, "The neighbour already freed\n");
+ if (!n)
return;
- }
neigh_event_send(n, NULL);
neigh_release(n);
--
2.17.0
^ permalink raw reply related
* [net-next V2 09/12] net/mlx5e: Support offloading tc double vlan headers match
From: Saeed Mahameed @ 2018-07-23 22:11 UTC (permalink / raw)
To: David S. Miller; +Cc: netdev, Jianbo Liu, Saeed Mahameed
In-Reply-To: <20180723221129.21625-1-saeedm@mellanox.com>
From: Jianbo Liu <jianbol@mellanox.com>
We can match on both outer and inner vlan tags, add support for
offloading that.
Signed-off-by: Jianbo Liu <jianbol@mellanox.com>
Reviewed-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
.../net/ethernet/mellanox/mlx5/core/en_tc.c | 55 ++++++++++++++++++-
1 file changed, 52 insertions(+), 3 deletions(-)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
index 335a08bc381d..dcb8c4993811 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
@@ -1235,6 +1235,10 @@ static int __parse_cls_flower(struct mlx5e_priv *priv,
outer_headers);
void *headers_v = MLX5_ADDR_OF(fte_match_param, spec->match_value,
outer_headers);
+ void *misc_c = MLX5_ADDR_OF(fte_match_param, spec->match_criteria,
+ misc_parameters);
+ void *misc_v = MLX5_ADDR_OF(fte_match_param, spec->match_value,
+ misc_parameters);
u16 addr_type = 0;
u8 ip_proto = 0;
@@ -1245,6 +1249,7 @@ static int __parse_cls_flower(struct mlx5e_priv *priv,
BIT(FLOW_DISSECTOR_KEY_BASIC) |
BIT(FLOW_DISSECTOR_KEY_ETH_ADDRS) |
BIT(FLOW_DISSECTOR_KEY_VLAN) |
+ BIT(FLOW_DISSECTOR_KEY_CVLAN) |
BIT(FLOW_DISSECTOR_KEY_IPV4_ADDRS) |
BIT(FLOW_DISSECTOR_KEY_IPV6_ADDRS) |
BIT(FLOW_DISSECTOR_KEY_PORTS) |
@@ -1325,9 +1330,18 @@ static int __parse_cls_flower(struct mlx5e_priv *priv,
skb_flow_dissector_target(f->dissector,
FLOW_DISSECTOR_KEY_VLAN,
f->mask);
- if (mask->vlan_id || mask->vlan_priority) {
- MLX5_SET(fte_match_set_lyr_2_4, headers_c, cvlan_tag, 1);
- MLX5_SET(fte_match_set_lyr_2_4, headers_v, cvlan_tag, 1);
+ if (mask->vlan_id || mask->vlan_priority || mask->vlan_tpid) {
+ if (key->vlan_tpid == htons(ETH_P_8021AD)) {
+ MLX5_SET(fte_match_set_lyr_2_4, headers_c,
+ svlan_tag, 1);
+ MLX5_SET(fte_match_set_lyr_2_4, headers_v,
+ svlan_tag, 1);
+ } else {
+ MLX5_SET(fte_match_set_lyr_2_4, headers_c,
+ cvlan_tag, 1);
+ MLX5_SET(fte_match_set_lyr_2_4, headers_v,
+ cvlan_tag, 1);
+ }
MLX5_SET(fte_match_set_lyr_2_4, headers_c, first_vid, mask->vlan_id);
MLX5_SET(fte_match_set_lyr_2_4, headers_v, first_vid, key->vlan_id);
@@ -1339,6 +1353,41 @@ static int __parse_cls_flower(struct mlx5e_priv *priv,
}
}
+ if (dissector_uses_key(f->dissector, FLOW_DISSECTOR_KEY_CVLAN)) {
+ struct flow_dissector_key_vlan *key =
+ skb_flow_dissector_target(f->dissector,
+ FLOW_DISSECTOR_KEY_CVLAN,
+ f->key);
+ struct flow_dissector_key_vlan *mask =
+ skb_flow_dissector_target(f->dissector,
+ FLOW_DISSECTOR_KEY_CVLAN,
+ f->mask);
+ if (mask->vlan_id || mask->vlan_priority || mask->vlan_tpid) {
+ if (key->vlan_tpid == htons(ETH_P_8021AD)) {
+ MLX5_SET(fte_match_set_misc, misc_c,
+ outer_second_svlan_tag, 1);
+ MLX5_SET(fte_match_set_misc, misc_v,
+ outer_second_svlan_tag, 1);
+ } else {
+ MLX5_SET(fte_match_set_misc, misc_c,
+ outer_second_cvlan_tag, 1);
+ MLX5_SET(fte_match_set_misc, misc_v,
+ outer_second_cvlan_tag, 1);
+ }
+
+ MLX5_SET(fte_match_set_misc, misc_c, outer_second_vid,
+ mask->vlan_id);
+ MLX5_SET(fte_match_set_misc, misc_v, outer_second_vid,
+ key->vlan_id);
+ MLX5_SET(fte_match_set_misc, misc_c, outer_second_prio,
+ mask->vlan_priority);
+ MLX5_SET(fte_match_set_misc, misc_v, outer_second_prio,
+ key->vlan_priority);
+
+ *match_level = MLX5_MATCH_L2;
+ }
+ }
+
if (dissector_uses_key(f->dissector, FLOW_DISSECTOR_KEY_BASIC)) {
struct flow_dissector_key_basic *key =
skb_flow_dissector_target(f->dissector,
--
2.17.0
^ permalink raw reply related
* [net-next V2 10/12] net/mlx5e: Refactor tc vlan push/pop actions offloading
From: Saeed Mahameed @ 2018-07-23 22:11 UTC (permalink / raw)
To: David S. Miller; +Cc: netdev, Jianbo Liu, Saeed Mahameed
In-Reply-To: <20180723221129.21625-1-saeedm@mellanox.com>
From: Jianbo Liu <jianbol@mellanox.com>
Extract actions offloading code to a new function, and also extend data
structures for double vlan actions.
Signed-off-by: Jianbo Liu <jianbol@mellanox.com>
Reviewed-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
.../net/ethernet/mellanox/mlx5/core/en_tc.c | 51 ++++++++++++-------
.../net/ethernet/mellanox/mlx5/core/eswitch.h | 6 +--
.../mellanox/mlx5/core/eswitch_offloads.c | 12 ++---
3 files changed, 41 insertions(+), 28 deletions(-)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
index dcb8c4993811..35b3e135ae1d 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
@@ -2578,6 +2578,32 @@ static int mlx5e_attach_encap(struct mlx5e_priv *priv,
return err;
}
+static int parse_tc_vlan_action(struct mlx5e_priv *priv,
+ const struct tc_action *a,
+ struct mlx5_esw_flow_attr *attr,
+ u32 *action)
+{
+ if (tcf_vlan_action(a) == TCA_VLAN_ACT_POP) {
+ *action |= MLX5_FLOW_CONTEXT_ACTION_VLAN_POP;
+ } else if (tcf_vlan_action(a) == TCA_VLAN_ACT_PUSH) {
+ *action |= MLX5_FLOW_CONTEXT_ACTION_VLAN_PUSH;
+ attr->vlan_vid[0] = tcf_vlan_push_vid(a);
+ if (mlx5_eswitch_vlan_actions_supported(priv->mdev)) {
+ attr->vlan_prio[0] = tcf_vlan_push_prio(a);
+ attr->vlan_proto[0] = tcf_vlan_push_proto(a);
+ if (!attr->vlan_proto[0])
+ attr->vlan_proto[0] = htons(ETH_P_8021Q);
+ } else if (tcf_vlan_push_proto(a) != htons(ETH_P_8021Q) ||
+ tcf_vlan_push_prio(a)) {
+ return -EOPNOTSUPP;
+ }
+ } else { /* action is TCA_VLAN_ACT_MODIFY */
+ return -EOPNOTSUPP;
+ }
+
+ return 0;
+}
+
static int parse_tc_fdb_actions(struct mlx5e_priv *priv, struct tcf_exts *exts,
struct mlx5e_tc_flow_parse_attr *parse_attr,
struct mlx5e_tc_flow *flow)
@@ -2589,6 +2615,7 @@ static int parse_tc_fdb_actions(struct mlx5e_priv *priv, struct tcf_exts *exts,
LIST_HEAD(actions);
bool encap = false;
u32 action = 0;
+ int err;
if (!tcf_exts_has_actions(exts))
return -EINVAL;
@@ -2605,8 +2632,6 @@ static int parse_tc_fdb_actions(struct mlx5e_priv *priv, struct tcf_exts *exts,
}
if (is_tcf_pedit(a)) {
- int err;
-
err = parse_tc_pedit_action(priv, a, MLX5_FLOW_NAMESPACE_FDB,
parse_attr);
if (err)
@@ -2673,23 +2698,11 @@ static int parse_tc_fdb_actions(struct mlx5e_priv *priv, struct tcf_exts *exts,
}
if (is_tcf_vlan(a)) {
- if (tcf_vlan_action(a) == TCA_VLAN_ACT_POP) {
- action |= MLX5_FLOW_CONTEXT_ACTION_VLAN_POP;
- } else if (tcf_vlan_action(a) == TCA_VLAN_ACT_PUSH) {
- action |= MLX5_FLOW_CONTEXT_ACTION_VLAN_PUSH;
- attr->vlan_vid = tcf_vlan_push_vid(a);
- if (mlx5_eswitch_vlan_actions_supported(priv->mdev)) {
- attr->vlan_prio = tcf_vlan_push_prio(a);
- attr->vlan_proto = tcf_vlan_push_proto(a);
- if (!attr->vlan_proto)
- attr->vlan_proto = htons(ETH_P_8021Q);
- } else if (tcf_vlan_push_proto(a) != htons(ETH_P_8021Q) ||
- tcf_vlan_push_prio(a)) {
- return -EOPNOTSUPP;
- }
- } else { /* action is TCA_VLAN_ACT_MODIFY */
- return -EOPNOTSUPP;
- }
+ err = parse_tc_vlan_action(priv, a, attr, &action);
+
+ if (err)
+ return err;
+
attr->mirror_count = attr->out_count;
continue;
}
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h
index b174da2884c5..befa0011efee 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h
@@ -256,9 +256,9 @@ struct mlx5_esw_flow_attr {
int out_count;
int action;
- __be16 vlan_proto;
- u16 vlan_vid;
- u8 vlan_prio;
+ __be16 vlan_proto[1];
+ u16 vlan_vid[1];
+ u8 vlan_prio[1];
bool vlan_handled;
u32 encap_id;
u32 mod_hdr_id;
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
index f32e69170b30..552954d7184e 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
@@ -70,9 +70,9 @@ mlx5_eswitch_add_offloaded_rule(struct mlx5_eswitch *esw,
flow_act.action &= ~(MLX5_FLOW_CONTEXT_ACTION_VLAN_PUSH |
MLX5_FLOW_CONTEXT_ACTION_VLAN_POP);
else if (flow_act.action & MLX5_FLOW_CONTEXT_ACTION_VLAN_PUSH) {
- flow_act.vlan[0].ethtype = ntohs(attr->vlan_proto);
- flow_act.vlan[0].vid = attr->vlan_vid;
- flow_act.vlan[0].prio = attr->vlan_prio;
+ flow_act.vlan[0].ethtype = ntohs(attr->vlan_proto[0]);
+ flow_act.vlan[0].vid = attr->vlan_vid[0];
+ flow_act.vlan[0].prio = attr->vlan_prio[0];
}
if (flow_act.action & MLX5_FLOW_CONTEXT_ACTION_FWD_DEST) {
@@ -266,7 +266,7 @@ static int esw_add_vlan_action_check(struct mlx5_esw_flow_attr *attr,
/* protects against (1) setting rules with different vlans to push and
* (2) setting rules w.o vlans (attr->vlan = 0) && w. vlans to push (!= 0)
*/
- if (push && in_rep->vlan_refcount && (in_rep->vlan != attr->vlan_vid))
+ if (push && in_rep->vlan_refcount && (in_rep->vlan != attr->vlan_vid[0]))
goto out_notsupp;
return 0;
@@ -324,11 +324,11 @@ int mlx5_eswitch_add_vlan_action(struct mlx5_eswitch *esw,
if (vport->vlan_refcount)
goto skip_set_push;
- err = __mlx5_eswitch_set_vport_vlan(esw, vport->vport, attr->vlan_vid, 0,
+ err = __mlx5_eswitch_set_vport_vlan(esw, vport->vport, attr->vlan_vid[0], 0,
SET_VLAN_INSERT | SET_VLAN_STRIP);
if (err)
goto out;
- vport->vlan = attr->vlan_vid;
+ vport->vlan = attr->vlan_vid[0];
skip_set_push:
vport->vlan_refcount++;
}
--
2.17.0
^ permalink raw reply related
* [net-next V2 11/12] net/mlx5e: Support offloading double vlan push/pop tc actions
From: Saeed Mahameed @ 2018-07-23 22:11 UTC (permalink / raw)
To: David S. Miller; +Cc: netdev, Jianbo Liu, Saeed Mahameed
In-Reply-To: <20180723221129.21625-1-saeedm@mellanox.com>
From: Jianbo Liu <jianbol@mellanox.com>
As we can configure two push/pop actions in one flow table entry,
add support to offload those double vlan actions in a rule to HW.
Signed-off-by: Jianbo Liu <jianbol@mellanox.com>
Reviewed-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
.../net/ethernet/mellanox/mlx5/core/en_tc.c | 46 ++++++++++++++-----
.../net/ethernet/mellanox/mlx5/core/eswitch.h | 21 ++++++---
.../mellanox/mlx5/core/eswitch_offloads.c | 11 +++--
3 files changed, 58 insertions(+), 20 deletions(-)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
index 35b3e135ae1d..e9888d6c1f7c 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
@@ -2583,24 +2583,48 @@ static int parse_tc_vlan_action(struct mlx5e_priv *priv,
struct mlx5_esw_flow_attr *attr,
u32 *action)
{
+ u8 vlan_idx = attr->total_vlan;
+
+ if (vlan_idx >= MLX5_FS_VLAN_DEPTH)
+ return -EOPNOTSUPP;
+
if (tcf_vlan_action(a) == TCA_VLAN_ACT_POP) {
- *action |= MLX5_FLOW_CONTEXT_ACTION_VLAN_POP;
+ if (vlan_idx) {
+ if (!mlx5_eswitch_vlan_actions_supported(priv->mdev,
+ MLX5_FS_VLAN_DEPTH))
+ return -EOPNOTSUPP;
+
+ *action |= MLX5_FLOW_CONTEXT_ACTION_VLAN_POP_2;
+ } else {
+ *action |= MLX5_FLOW_CONTEXT_ACTION_VLAN_POP;
+ }
} else if (tcf_vlan_action(a) == TCA_VLAN_ACT_PUSH) {
- *action |= MLX5_FLOW_CONTEXT_ACTION_VLAN_PUSH;
- attr->vlan_vid[0] = tcf_vlan_push_vid(a);
- if (mlx5_eswitch_vlan_actions_supported(priv->mdev)) {
- attr->vlan_prio[0] = tcf_vlan_push_prio(a);
- attr->vlan_proto[0] = tcf_vlan_push_proto(a);
- if (!attr->vlan_proto[0])
- attr->vlan_proto[0] = htons(ETH_P_8021Q);
- } else if (tcf_vlan_push_proto(a) != htons(ETH_P_8021Q) ||
- tcf_vlan_push_prio(a)) {
- return -EOPNOTSUPP;
+ attr->vlan_vid[vlan_idx] = tcf_vlan_push_vid(a);
+ attr->vlan_prio[vlan_idx] = tcf_vlan_push_prio(a);
+ attr->vlan_proto[vlan_idx] = tcf_vlan_push_proto(a);
+ if (!attr->vlan_proto[vlan_idx])
+ attr->vlan_proto[vlan_idx] = htons(ETH_P_8021Q);
+
+ if (vlan_idx) {
+ if (!mlx5_eswitch_vlan_actions_supported(priv->mdev,
+ MLX5_FS_VLAN_DEPTH))
+ return -EOPNOTSUPP;
+
+ *action |= MLX5_FLOW_CONTEXT_ACTION_VLAN_PUSH_2;
+ } else {
+ if (!mlx5_eswitch_vlan_actions_supported(priv->mdev, 1) &&
+ (tcf_vlan_push_proto(a) != htons(ETH_P_8021Q) ||
+ tcf_vlan_push_prio(a)))
+ return -EOPNOTSUPP;
+
+ *action |= MLX5_FLOW_CONTEXT_ACTION_VLAN_PUSH;
}
} else { /* action is TCA_VLAN_ACT_MODIFY */
return -EOPNOTSUPP;
}
+ attr->total_vlan = vlan_idx + 1;
+
return 0;
}
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h
index befa0011efee..c17bfcab517c 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h
@@ -38,6 +38,7 @@
#include <net/devlink.h>
#include <linux/mlx5/device.h>
#include <linux/mlx5/eswitch.h>
+#include <linux/mlx5/fs.h>
#include "lib/mpfs.h"
#ifdef CONFIG_MLX5_ESWITCH
@@ -256,9 +257,10 @@ struct mlx5_esw_flow_attr {
int out_count;
int action;
- __be16 vlan_proto[1];
- u16 vlan_vid[1];
- u8 vlan_prio[1];
+ __be16 vlan_proto[MLX5_FS_VLAN_DEPTH];
+ u16 vlan_vid[MLX5_FS_VLAN_DEPTH];
+ u8 vlan_prio[MLX5_FS_VLAN_DEPTH];
+ u8 total_vlan;
bool vlan_handled;
u32 encap_id;
u32 mod_hdr_id;
@@ -282,10 +284,17 @@ int mlx5_eswitch_del_vlan_action(struct mlx5_eswitch *esw,
int __mlx5_eswitch_set_vport_vlan(struct mlx5_eswitch *esw,
int vport, u16 vlan, u8 qos, u8 set_flags);
-static inline bool mlx5_eswitch_vlan_actions_supported(struct mlx5_core_dev *dev)
+static inline bool mlx5_eswitch_vlan_actions_supported(struct mlx5_core_dev *dev,
+ u8 vlan_depth)
{
- return MLX5_CAP_ESW_FLOWTABLE_FDB(dev, pop_vlan) &&
- MLX5_CAP_ESW_FLOWTABLE_FDB(dev, push_vlan);
+ bool ret = MLX5_CAP_ESW_FLOWTABLE_FDB(dev, pop_vlan) &&
+ MLX5_CAP_ESW_FLOWTABLE_FDB(dev, push_vlan);
+
+ if (vlan_depth == 1)
+ return ret;
+
+ return ret && MLX5_CAP_ESW_FLOWTABLE_FDB(dev, pop_vlan_2) &&
+ MLX5_CAP_ESW_FLOWTABLE_FDB(dev, push_vlan_2);
}
#define MLX5_DEBUG_ESWITCH_MASK BIT(3)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
index 552954d7184e..f72b5c9dcfe9 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
@@ -66,13 +66,18 @@ mlx5_eswitch_add_offloaded_rule(struct mlx5_eswitch *esw,
flow_act.action = attr->action;
/* if per flow vlan pop/push is emulated, don't set that into the firmware */
- if (!mlx5_eswitch_vlan_actions_supported(esw->dev))
+ if (!mlx5_eswitch_vlan_actions_supported(esw->dev, 1))
flow_act.action &= ~(MLX5_FLOW_CONTEXT_ACTION_VLAN_PUSH |
MLX5_FLOW_CONTEXT_ACTION_VLAN_POP);
else if (flow_act.action & MLX5_FLOW_CONTEXT_ACTION_VLAN_PUSH) {
flow_act.vlan[0].ethtype = ntohs(attr->vlan_proto[0]);
flow_act.vlan[0].vid = attr->vlan_vid[0];
flow_act.vlan[0].prio = attr->vlan_prio[0];
+ if (flow_act.action & MLX5_FLOW_CONTEXT_ACTION_VLAN_PUSH_2) {
+ flow_act.vlan[1].ethtype = ntohs(attr->vlan_proto[1]);
+ flow_act.vlan[1].vid = attr->vlan_vid[1];
+ flow_act.vlan[1].prio = attr->vlan_prio[1];
+ }
}
if (flow_act.action & MLX5_FLOW_CONTEXT_ACTION_FWD_DEST) {
@@ -284,7 +289,7 @@ int mlx5_eswitch_add_vlan_action(struct mlx5_eswitch *esw,
int err = 0;
/* nop if we're on the vlan push/pop non emulation mode */
- if (mlx5_eswitch_vlan_actions_supported(esw->dev))
+ if (mlx5_eswitch_vlan_actions_supported(esw->dev, 1))
return 0;
push = !!(attr->action & MLX5_FLOW_CONTEXT_ACTION_VLAN_PUSH);
@@ -347,7 +352,7 @@ int mlx5_eswitch_del_vlan_action(struct mlx5_eswitch *esw,
int err = 0;
/* nop if we're on the vlan push/pop non emulation mode */
- if (mlx5_eswitch_vlan_actions_supported(esw->dev))
+ if (mlx5_eswitch_vlan_actions_supported(esw->dev, 1))
return 0;
if (!attr->vlan_handled)
--
2.17.0
^ permalink raw reply related
* [net-next V2 12/12] net/mlx5e: Use PARTIAL_GSO for UDP segmentation
From: Saeed Mahameed @ 2018-07-23 22:11 UTC (permalink / raw)
To: David S. Miller; +Cc: netdev, Boris Pismenny, Saeed Mahameed
In-Reply-To: <20180723221129.21625-1-saeedm@mellanox.com>
From: Boris Pismenny <borisp@mellanox.com>
This patch removes the splitting of UDP_GSO_L4 packets in the driver,
and exposes UDP_GSO_L4 as a PARTIAL_GSO feature. Thus, the network stack
is not responsible for splitting the packet into two.
Signed-off-by: Boris Pismenny <borisp@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
.../net/ethernet/mellanox/mlx5/core/Makefile | 4 +-
.../mellanox/mlx5/core/en_accel/en_accel.h | 27 +++--
.../mellanox/mlx5/core/en_accel/rxtx.c | 109 ------------------
.../mellanox/mlx5/core/en_accel/rxtx.h | 14 ---
.../net/ethernet/mellanox/mlx5/core/en_main.c | 9 +-
5 files changed, 23 insertions(+), 140 deletions(-)
delete mode 100644 drivers/net/ethernet/mellanox/mlx5/core/en_accel/rxtx.c
delete mode 100644 drivers/net/ethernet/mellanox/mlx5/core/en_accel/rxtx.h
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/Makefile b/drivers/net/ethernet/mellanox/mlx5/core/Makefile
index 55d5a5c2e9d8..fa7fcca5dc78 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/Makefile
+++ b/drivers/net/ethernet/mellanox/mlx5/core/Makefile
@@ -14,8 +14,8 @@ mlx5_core-$(CONFIG_MLX5_FPGA) += fpga/cmd.o fpga/core.o fpga/conn.o fpga/sdk.o \
fpga/ipsec.o fpga/tls.o
mlx5_core-$(CONFIG_MLX5_CORE_EN) += en_main.o en_common.o en_fs.o en_ethtool.o \
- en_tx.o en_rx.o en_dim.o en_txrx.o en_accel/rxtx.o en_stats.o \
- vxlan.o en_arfs.o en_fs_ethtool.o en_selftest.o en/port.o
+ en_tx.o en_rx.o en_dim.o en_txrx.o en_stats.o vxlan.o \
+ en_arfs.o en_fs_ethtool.o en_selftest.o en/port.o
mlx5_core-$(CONFIG_MLX5_MPFS) += lib/mpfs.o
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/en_accel.h b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/en_accel.h
index 39a5d13ba459..1dd225380a66 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/en_accel.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/en_accel.h
@@ -38,14 +38,22 @@
#include <linux/netdevice.h>
#include "en_accel/ipsec_rxtx.h"
#include "en_accel/tls_rxtx.h"
-#include "en_accel/rxtx.h"
#include "en.h"
-static inline struct sk_buff *mlx5e_accel_handle_tx(struct sk_buff *skb,
- struct mlx5e_txqsq *sq,
- struct net_device *dev,
- struct mlx5e_tx_wqe **wqe,
- u16 *pi)
+static inline void
+mlx5e_udp_gso_handle_tx_skb(struct sk_buff *skb)
+{
+ int payload_len = skb_shinfo(skb)->gso_size + sizeof(struct udphdr);
+
+ udp_hdr(skb)->len = htons(payload_len);
+}
+
+static inline struct sk_buff *
+mlx5e_accel_handle_tx(struct sk_buff *skb,
+ struct mlx5e_txqsq *sq,
+ struct net_device *dev,
+ struct mlx5e_tx_wqe **wqe,
+ u16 *pi)
{
#ifdef CONFIG_MLX5_EN_TLS
if (test_bit(MLX5E_SQ_STATE_TLS, &sq->state)) {
@@ -63,11 +71,8 @@ static inline struct sk_buff *mlx5e_accel_handle_tx(struct sk_buff *skb,
}
#endif
- if (skb_shinfo(skb)->gso_type & SKB_GSO_UDP_L4) {
- skb = mlx5e_udp_gso_handle_tx_skb(dev, sq, skb, wqe, pi);
- if (unlikely(!skb))
- return NULL;
- }
+ if (skb_is_gso(skb) && skb_shinfo(skb)->gso_type & SKB_GSO_UDP_L4)
+ mlx5e_udp_gso_handle_tx_skb(skb);
return skb;
}
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/rxtx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/rxtx.c
deleted file mode 100644
index 7b7ec3998e84..000000000000
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/rxtx.c
+++ /dev/null
@@ -1,109 +0,0 @@
-#include "en_accel/rxtx.h"
-
-static void mlx5e_udp_gso_prepare_last_skb(struct sk_buff *skb,
- struct sk_buff *nskb,
- int remaining)
-{
- int bytes_needed = remaining, remaining_headlen, remaining_page_offset;
- int headlen = skb_transport_offset(skb) + sizeof(struct udphdr);
- int payload_len = remaining + sizeof(struct udphdr);
- int k = 0, i, j;
-
- skb_copy_bits(skb, 0, nskb->data, headlen);
- nskb->dev = skb->dev;
- skb_reset_mac_header(nskb);
- skb_set_network_header(nskb, skb_network_offset(skb));
- skb_set_transport_header(nskb, skb_transport_offset(skb));
- skb_set_tail_pointer(nskb, headlen);
-
- /* How many frags do we need? */
- for (i = skb_shinfo(skb)->nr_frags - 1; i >= 0; i--) {
- bytes_needed -= skb_frag_size(&skb_shinfo(skb)->frags[i]);
- k++;
- if (bytes_needed <= 0)
- break;
- }
-
- /* Fill the first frag and split it if necessary */
- j = skb_shinfo(skb)->nr_frags - k;
- remaining_page_offset = -bytes_needed;
- skb_fill_page_desc(nskb, 0,
- skb_shinfo(skb)->frags[j].page.p,
- skb_shinfo(skb)->frags[j].page_offset + remaining_page_offset,
- skb_shinfo(skb)->frags[j].size - remaining_page_offset);
-
- skb_frag_ref(skb, j);
-
- /* Fill the rest of the frags */
- for (i = 1; i < k; i++) {
- j = skb_shinfo(skb)->nr_frags - k + i;
-
- skb_fill_page_desc(nskb, i,
- skb_shinfo(skb)->frags[j].page.p,
- skb_shinfo(skb)->frags[j].page_offset,
- skb_shinfo(skb)->frags[j].size);
- skb_frag_ref(skb, j);
- }
- skb_shinfo(nskb)->nr_frags = k;
-
- remaining_headlen = remaining - skb->data_len;
-
- /* headlen contains remaining data? */
- if (remaining_headlen > 0)
- skb_copy_bits(skb, skb->len - remaining, nskb->data + headlen,
- remaining_headlen);
- nskb->len = remaining + headlen;
- nskb->data_len = payload_len - sizeof(struct udphdr) +
- max_t(int, 0, remaining_headlen);
- nskb->protocol = skb->protocol;
- if (nskb->protocol == htons(ETH_P_IP)) {
- ip_hdr(nskb)->id = htons(ntohs(ip_hdr(nskb)->id) +
- skb_shinfo(skb)->gso_segs);
- ip_hdr(nskb)->tot_len =
- htons(payload_len + sizeof(struct iphdr));
- } else {
- ipv6_hdr(nskb)->payload_len = htons(payload_len);
- }
- udp_hdr(nskb)->len = htons(payload_len);
- skb_shinfo(nskb)->gso_size = 0;
- nskb->ip_summed = skb->ip_summed;
- nskb->csum_start = skb->csum_start;
- nskb->csum_offset = skb->csum_offset;
- nskb->queue_mapping = skb->queue_mapping;
-}
-
-/* might send skbs and update wqe and pi */
-struct sk_buff *mlx5e_udp_gso_handle_tx_skb(struct net_device *netdev,
- struct mlx5e_txqsq *sq,
- struct sk_buff *skb,
- struct mlx5e_tx_wqe **wqe,
- u16 *pi)
-{
- int payload_len = skb_shinfo(skb)->gso_size + sizeof(struct udphdr);
- int headlen = skb_transport_offset(skb) + sizeof(struct udphdr);
- int remaining = (skb->len - headlen) % skb_shinfo(skb)->gso_size;
- struct sk_buff *nskb;
-
- if (skb->protocol == htons(ETH_P_IP))
- ip_hdr(skb)->tot_len = htons(payload_len + sizeof(struct iphdr));
- else
- ipv6_hdr(skb)->payload_len = htons(payload_len);
- udp_hdr(skb)->len = htons(payload_len);
- if (!remaining)
- return skb;
-
- sq->stats->udp_seg_rem++;
- nskb = alloc_skb(max_t(int, headlen, headlen + remaining - skb->data_len), GFP_ATOMIC);
- if (unlikely(!nskb)) {
- sq->stats->dropped++;
- return NULL;
- }
-
- mlx5e_udp_gso_prepare_last_skb(skb, nskb, remaining);
-
- skb_shinfo(skb)->gso_segs--;
- pskb_trim(skb, skb->len - remaining);
- mlx5e_sq_xmit(sq, skb, *wqe, *pi);
- mlx5e_sq_fetch_wqe(sq, wqe, pi);
- return nskb;
-}
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/rxtx.h b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/rxtx.h
deleted file mode 100644
index ed42699a78b3..000000000000
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/rxtx.h
+++ /dev/null
@@ -1,14 +0,0 @@
-
-#ifndef __MLX5E_EN_ACCEL_RX_TX_H__
-#define __MLX5E_EN_ACCEL_RX_TX_H__
-
-#include <linux/skbuff.h>
-#include "en.h"
-
-struct sk_buff *mlx5e_udp_gso_handle_tx_skb(struct net_device *netdev,
- struct mlx5e_txqsq *sq,
- struct sk_buff *skb,
- struct mlx5e_tx_wqe **wqe,
- u16 *pi);
-
-#endif
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
index 712b9766485f..dccde18f6170 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
@@ -4538,7 +4538,6 @@ static void mlx5e_build_nic_netdev(struct net_device *netdev)
netdev->hw_features |= NETIF_F_HW_VLAN_STAG_TX;
if (mlx5e_vxlan_allowed(mdev) || MLX5_CAP_ETH(mdev, tunnel_stateless_gre)) {
- netdev->hw_features |= NETIF_F_GSO_PARTIAL;
netdev->hw_enc_features |= NETIF_F_IP_CSUM;
netdev->hw_enc_features |= NETIF_F_IPV6_CSUM;
netdev->hw_enc_features |= NETIF_F_TSO;
@@ -4563,6 +4562,11 @@ static void mlx5e_build_nic_netdev(struct net_device *netdev)
NETIF_F_GSO_GRE_CSUM;
}
+ netdev->hw_features |= NETIF_F_GSO_PARTIAL;
+ netdev->gso_partial_features |= NETIF_F_GSO_UDP_L4;
+ netdev->hw_features |= NETIF_F_GSO_UDP_L4;
+ netdev->features |= NETIF_F_GSO_UDP_L4;
+
mlx5_query_port_fcs(mdev, &fcs_supported, &fcs_enabled);
if (fcs_supported)
@@ -4595,9 +4599,6 @@ static void mlx5e_build_nic_netdev(struct net_device *netdev)
netdev->features |= NETIF_F_HIGHDMA;
netdev->features |= NETIF_F_HW_VLAN_STAG_FILTER;
- netdev->features |= NETIF_F_GSO_UDP_L4;
- netdev->hw_features |= NETIF_F_GSO_UDP_L4;
-
netdev->priv_flags |= IFF_UNICAST_FLT;
mlx5e_set_netdev_dev_addr(netdev);
--
2.17.0
^ permalink raw reply related
* [PATCH net-next] tls: Fix improper revert in zerocopy_from_iter
From: Doron Roberts-Kedes @ 2018-07-23 22:20 UTC (permalink / raw)
To: David S . Miller
Cc: Dave Watson, Vakul Garg, Matt Mullins, netdev,
Doron Roberts-Kedes
The current code is problematic because the iov_iter is reverted and
never advanced in the non-error case. This patch skips the revert in the
non-error case. This patch also fixes the amount by which the iov_iter
is reverted. Currently, iov_iter is reverted by size, which can be
greater than the amount by which the iter was actually advanced.
Instead, mimic the tx path which reverts by the difference before and
after zerocopy_from_iter.
Fixes: 4718799817c5 ("tls: Fix zerocopy_from_iter iov handling")
Signed-off-by: Doron Roberts-Kedes <doronrk@fb.com>
---
net/tls/tls_sw.c | 13 +++++++------
1 file changed, 7 insertions(+), 6 deletions(-)
diff --git a/net/tls/tls_sw.c b/net/tls/tls_sw.c
index 490f2bcc6313..2ea000baebf8 100644
--- a/net/tls/tls_sw.c
+++ b/net/tls/tls_sw.c
@@ -276,7 +276,7 @@ static int zerocopy_from_iter(struct sock *sk, struct iov_iter *from,
int length, int *pages_used,
unsigned int *size_used,
struct scatterlist *to, int to_max_pages,
- bool charge, bool revert)
+ bool charge)
{
struct page *pages[MAX_SKB_FRAGS];
@@ -327,8 +327,6 @@ static int zerocopy_from_iter(struct sock *sk, struct iov_iter *from,
out:
*size_used = size;
*pages_used = num_elem;
- if (revert)
- iov_iter_revert(from, size);
return rc;
}
@@ -431,7 +429,7 @@ int tls_sw_sendmsg(struct sock *sk, struct msghdr *msg, size_t size)
&ctx->sg_plaintext_size,
ctx->sg_plaintext_data,
ARRAY_SIZE(ctx->sg_plaintext_data),
- true, false);
+ true);
if (ret)
goto fallback_to_reg_send;
@@ -811,6 +809,7 @@ int tls_sw_recvmsg(struct sock *sk,
likely(!(flags & MSG_PEEK))) {
struct scatterlist sgin[MAX_SKB_FRAGS + 1];
int pages = 0;
+ int orig_chunk = chunk;
zc = true;
sg_init_table(sgin, MAX_SKB_FRAGS + 1);
@@ -820,9 +819,11 @@ int tls_sw_recvmsg(struct sock *sk,
err = zerocopy_from_iter(sk, &msg->msg_iter,
to_copy, &pages,
&chunk, &sgin[1],
- MAX_SKB_FRAGS, false, true);
- if (err < 0)
+ MAX_SKB_FRAGS, false);
+ if (err < 0) {
+ iov_iter_revert(&msg->msg_iter, chunk - orig_chunk);
goto fallback_to_reg_recv;
+ }
err = decrypt_skb(sk, skb, sgin);
for (; pages > 0; pages--)
--
2.17.1
^ permalink raw reply related
page: next (older) | prev (newer) | latest
- recent:[subjects (threaded)|topics (new)|topics (active)]
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox