[PATCH AUTOSEL 4.19 008/192] net/mlx5: Avoid panic when setting vport rate

[PATCH AUTOSEL 4.19 008/192] net/mlx5: Avoid panic when setting vport rate

[Sasha Levin]

From: Tonghao Zhang <xiangxia.m.yue@gmail.com>

[ Upstream commit 24319258660a84dd77f4be026a55b10a12524919 ]

If we try to set VFs rate on a VF (not PF) net device, the kernel
will be crash. The commands are show as below:

$ echo 2 > /sys/class/net/$MLX_PF0/device/sriov_numvfs
$ ip link set $MLX_VF0 vf 0 max_tx_rate 2 min_tx_rate 1

If not applied the first patch ("net/mlx5: Avoid panic when setting
vport mac, getting vport config"), the command:

$ ip link set $MLX_VF0 vf 0 rate 100

can also crash the kernel.

[ 1650.006388] RIP: 0010:mlx5_eswitch_set_vport_rate+0x1f/0x260 [mlx5_core]
[ 1650.007092]  do_setlink+0x982/0xd20
[ 1650.007129]  __rtnl_newlink+0x528/0x7d0
[ 1650.007374]  rtnl_newlink+0x43/0x60
[ 1650.007407]  rtnetlink_rcv_msg+0x2a2/0x320
[ 1650.007484]  netlink_rcv_skb+0xcb/0x100
[ 1650.007519]  netlink_unicast+0x17f/0x230
[ 1650.007554]  netlink_sendmsg+0x2d2/0x3d0
[ 1650.007592]  sock_sendmsg+0x36/0x50
[ 1650.007625]  ___sys_sendmsg+0x280/0x2a0
[ 1650.007963]  __sys_sendmsg+0x58/0xa0
[ 1650.007998]  do_syscall_64+0x5b/0x180
[ 1650.009438]  entry_SYSCALL_64_after_hwframe+0x44/0xa9

Fixes: c9497c98901c ("net/mlx5: Add support for setting VF min rate")
Cc: Mohamad Haj Yahia <mohamad@mellanox.com>
Signed-off-by: Tonghao Zhang <xiangxia.m.yue@gmail.com>
Reviewed-by: Roi Dayan <roid@mellanox.com>
Acked-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 drivers/net/ethernet/mellanox/mlx5/core/eswitch.c | 13 +++++++++----
 1 file changed, 9 insertions(+), 4 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c
index d6706475a3ba..886a4a77c47f 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c
@@ -2044,19 +2044,24 @@ static int normalize_vports_min_rate(struct mlx5_eswitch *esw, u32 divider)
 int mlx5_eswitch_set_vport_rate(struct mlx5_eswitch *esw, int vport,
 				u32 max_rate, u32 min_rate)
 {
-	u32 fw_max_bw_share = MLX5_CAP_QOS(esw->dev, max_tsar_bw_share);
-	bool min_rate_supported = MLX5_CAP_QOS(esw->dev, esw_bw_share) &&
-					fw_max_bw_share >= MLX5_MIN_BW_SHARE;
-	bool max_rate_supported = MLX5_CAP_QOS(esw->dev, esw_rate_limit);
 	struct mlx5_vport *evport;
+	u32 fw_max_bw_share;
 	u32 previous_min_rate;
 	u32 divider;
+	bool min_rate_supported;
+	bool max_rate_supported;
 	int err = 0;
 
 	if (!ESW_ALLOWED(esw))
 		return -EPERM;
 	if (!LEGAL_VPORT(esw, vport))
 		return -EINVAL;
+
+	fw_max_bw_share = MLX5_CAP_QOS(esw->dev, max_tsar_bw_share);
+	min_rate_supported = MLX5_CAP_QOS(esw->dev, esw_bw_share) &&
+				fw_max_bw_share >= MLX5_MIN_BW_SHARE;
+	max_rate_supported = MLX5_CAP_QOS(esw->dev, esw_rate_limit);
+
 	if ((min_rate && !min_rate_supported) || (max_rate && !max_rate_supported))
 		return -EOPNOTSUPP;
 
-- 
2.19.1

[PATCH AUTOSEL 4.19 009/192] net/mlx5: Avoid panic when setting vport mac, getting vport config

[Sasha Levin]

From: Tonghao Zhang <xiangxia.m.yue@gmail.com>

[ Upstream commit 6e77c413e8e73d0f36b5358b601389d75ec4451c ]

If we try to set VFs mac address on a VF (not PF) net device,
the kernel will be crash. The commands are show as below:

$ echo 2 > /sys/class/net/$MLX_PF0/device/sriov_numvfs
$ ip link set $MLX_VF0 vf 0 mac 00:11:22:33:44:00

[exception RIP: mlx5_eswitch_set_vport_mac+41]
[ffffb8b7079e3688] do_setlink at ffffffff8f67f85b
[ffffb8b7079e37a8] __rtnl_newlink at ffffffff8f683778
[ffffb8b7079e3b68] rtnl_newlink at ffffffff8f683a63
[ffffb8b7079e3b90] rtnetlink_rcv_msg at ffffffff8f67d812
[ffffb8b7079e3c10] netlink_rcv_skb at ffffffff8f6b88ab
[ffffb8b7079e3c60] netlink_unicast at ffffffff8f6b808f
[ffffb8b7079e3ca0] netlink_sendmsg at ffffffff8f6b8412
[ffffb8b7079e3d18] sock_sendmsg at ffffffff8f6452f6
[ffffb8b7079e3d30] ___sys_sendmsg at ffffffff8f645860
[ffffb8b7079e3eb0] __sys_sendmsg at ffffffff8f647a38
[ffffb8b7079e3f38] do_syscall_64 at ffffffff8f00401b
[ffffb8b7079e3f50] entry_SYSCALL_64_after_hwframe at ffffffff8f80008c

and

[exception RIP: mlx5_eswitch_get_vport_config+12]
[ffffa70607e57678] mlx5e_get_vf_config at ffffffffc03c7f8f [mlx5_core]
[ffffa70607e57688] do_setlink at ffffffffbc67fa59
[ffffa70607e577a8] __rtnl_newlink at ffffffffbc683778
[ffffa70607e57b68] rtnl_newlink at ffffffffbc683a63
[ffffa70607e57b90] rtnetlink_rcv_msg at ffffffffbc67d812
[ffffa70607e57c10] netlink_rcv_skb at ffffffffbc6b88ab
[ffffa70607e57c60] netlink_unicast at ffffffffbc6b808f
[ffffa70607e57ca0] netlink_sendmsg at ffffffffbc6b8412
[ffffa70607e57d18] sock_sendmsg at ffffffffbc6452f6
[ffffa70607e57d30] ___sys_sendmsg at ffffffffbc645860
[ffffa70607e57eb0] __sys_sendmsg at ffffffffbc647a38
[ffffa70607e57f38] do_syscall_64 at ffffffffbc00401b
[ffffa70607e57f50] entry_SYSCALL_64_after_hwframe at ffffffffbc80008c

Fixes: a8d70a054a718 ("net/mlx5: E-Switch, Disallow vlan/spoofcheck setup if not being esw manager")
Cc: Eli Cohen <eli@mellanox.com>
Signed-off-by: Tonghao Zhang <xiangxia.m.yue@gmail.com>
Reviewed-by: Roi Dayan <roid@mellanox.com>
Acked-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 drivers/net/ethernet/mellanox/mlx5/core/eswitch.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c
index 886a4a77c47f..26c9f9421901 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c
@@ -1797,7 +1797,7 @@ int mlx5_eswitch_set_vport_mac(struct mlx5_eswitch *esw,
 	u64 node_guid;
 	int err = 0;
 
-	if (!MLX5_CAP_GEN(esw->dev, vport_group_manager))
+	if (!esw || !MLX5_CAP_GEN(esw->dev, vport_group_manager))
 		return -EPERM;
 	if (!LEGAL_VPORT(esw, vport) || is_multicast_ether_addr(mac))
 		return -EINVAL;
@@ -1871,7 +1871,7 @@ int mlx5_eswitch_get_vport_config(struct mlx5_eswitch *esw,
 {
 	struct mlx5_vport *evport;
 
-	if (!MLX5_CAP_GEN(esw->dev, vport_group_manager))
+	if (!esw || !MLX5_CAP_GEN(esw->dev, vport_group_manager))
 		return -EPERM;
 	if (!LEGAL_VPORT(esw, vport))
 		return -EINVAL;
-- 
2.19.1

[PATCH AUTOSEL 4.19 069/192] IB/mlx4: Increase the timeout for CM cache

[Sasha Levin]

From: Håkon Bugge <haakon.bugge@oracle.com>

[ Upstream commit 2612d723aadcf8281f9bf8305657129bd9f3cd57 ]

Using CX-3 virtual functions, either from a bare-metal machine or
pass-through from a VM, MAD packets are proxied through the PF driver.

Since the VF drivers have separate name spaces for MAD Transaction Ids
(TIDs), the PF driver has to re-map the TIDs and keep the book keeping
in a cache.

Following the RDMA Connection Manager (CM) protocol, it is clear when
an entry has to evicted form the cache. But life is not perfect,
remote peers may die or be rebooted. Hence, it's a timeout to wipe out
a cache entry, when the PF driver assumes the remote peer has gone.

During workloads where a high number of QPs are destroyed concurrently,
excessive amount of CM DREQ retries has been observed

The problem can be demonstrated in a bare-metal environment, where two
nodes have instantiated 8 VFs each. This using dual ported HCAs, so we
have 16 vPorts per physical server.

64 processes are associated with each vPort and creates and destroys
one QP for each of the remote 64 processes. That is, 1024 QPs per
vPort, all in all 16K QPs. The QPs are created/destroyed using the
CM.

When tearing down these 16K QPs, excessive CM DREQ retries (and
duplicates) are observed. With some cat/paste/awk wizardry on the
infiniband_cm sysfs, we observe as sum of the 16 vPorts on one of the
nodes:

cm_rx_duplicates:
      dreq  2102
cm_rx_msgs:
      drep  1989
      dreq  6195
       rep  3968
       req  4224
       rtu  4224
cm_tx_msgs:
      drep  4093
      dreq 27568
       rep  4224
       req  3968
       rtu  3968
cm_tx_retries:
      dreq 23469

Note that the active/passive side is equally distributed between the
two nodes.

Enabling pr_debug in cm.c gives tons of:

[171778.814239] <mlx4_ib> mlx4_ib_multiplex_cm_handler: id{slave:
1,sl_cm_id: 0xd393089f} is NULL!

By increasing the CM_CLEANUP_CACHE_TIMEOUT from 5 to 30 seconds, the
tear-down phase of the application is reduced from approximately 90 to
50 seconds. Retries/duplicates are also significantly reduced:

cm_rx_duplicates:
      dreq  2460
[]
cm_tx_retries:
      dreq  3010
       req    47

Increasing the timeout further didn't help, as these duplicates and
retries stems from a too short CMA timeout, which was 20 (~4 seconds)
on the systems. By increasing the CMA timeout to 22 (~17 seconds), the
numbers fell down to about 10 for both of them.

Adjustment of the CMA timeout is not part of this commit.

Signed-off-by: Håkon Bugge <haakon.bugge@oracle.com>
Acked-by: Jack Morgenstein <jackm@dev.mellanox.co.il>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 drivers/infiniband/hw/mlx4/cm.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/infiniband/hw/mlx4/cm.c b/drivers/infiniband/hw/mlx4/cm.c
index fedaf8260105..8c79a480f2b7 100644
--- a/drivers/infiniband/hw/mlx4/cm.c
+++ b/drivers/infiniband/hw/mlx4/cm.c
@@ -39,7 +39,7 @@
 
 #include "mlx4_ib.h"
 
-#define CM_CLEANUP_CACHE_TIMEOUT  (5 * HZ)
+#define CM_CLEANUP_CACHE_TIMEOUT  (30 * HZ)
 
 struct id_map_entry {
 	struct rb_node node;
-- 
2.19.1

[PATCH AUTOSEL 4.19 119/192] iw_cxgb4: fix srqidx leak during connection abort

[Sasha Levin]

From: Raju Rangoju <rajur@chelsio.com>

[ Upstream commit f368ff188ae4b3ef6f740a15999ea0373261b619 ]

When an application aborts the connection by moving QP from RTS to ERROR,
then iw_cxgb4's modify_rc_qp() RTS->ERROR logic sets the
*srqidxp to 0 via t4_set_wq_in_error(&qhp->wq, 0), and aborts the
connection by calling c4iw_ep_disconnect().

c4iw_ep_disconnect() does the following:
 1. sends up a close_complete_upcall(ep, -ECONNRESET) to libcxgb4.
 2. sends abort request CPL to hw.

But, since the close_complete_upcall() is sent before sending the
ABORT_REQ to hw, libcxgb4 would fail to release the srqidx if the
connection holds one. Because, the srqidx is passed up to libcxgb4 only
after corresponding ABORT_RPL is processed by kernel in abort_rpl().

This patch handle the corner-case by moving the call to
close_complete_upcall() from c4iw_ep_disconnect() to abort_rpl().  So that
libcxgb4 is notified about the -ECONNRESET only after abort_rpl(), and
libcxgb4 can relinquish the srqidx properly.

Signed-off-by: Raju Rangoju <rajur@chelsio.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 drivers/infiniband/hw/cxgb4/cm.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/drivers/infiniband/hw/cxgb4/cm.c b/drivers/infiniband/hw/cxgb4/cm.c
index 0f83cbec33f3..a68569ec86bf 100644
--- a/drivers/infiniband/hw/cxgb4/cm.c
+++ b/drivers/infiniband/hw/cxgb4/cm.c
@@ -1904,8 +1904,10 @@ static int abort_rpl(struct c4iw_dev *dev, struct sk_buff *skb)
 	}
 	mutex_unlock(&ep->com.mutex);
 
-	if (release)
+	if (release) {
+		close_complete_upcall(ep, -ECONNRESET);
 		release_ep_resources(ep);
+	}
 	c4iw_put_ep(&ep->com);
 	return 0;
 }
@@ -3608,7 +3610,6 @@ int c4iw_ep_disconnect(struct c4iw_ep *ep, int abrupt, gfp_t gfp)
 	if (close) {
 		if (abrupt) {
 			set_bit(EP_DISC_ABORT, &ep->com.history);
-			close_complete_upcall(ep, -ECONNRESET);
 			ret = send_abort(ep);
 		} else {
 			set_bit(EP_DISC_CLOSE, &ep->com.history);
-- 
2.19.1