Skip to contentSkip to conversation

mlx5 fixes 2024-01-24

GIT PULL Applied Saeed Mahameed 2 participants 16 messages Jan 24, 2024 – Jan 25, 2024

Subsystems (6)
lore
Share & tools
lore
#0 lore raw

From: Saeed Mahameed <[email protected]>

This series provides bug fixes to mlx5 driver. Please pull and let me know if there is any problem.

Thanks,
Saeed.

The following changes since commit 32f2a0afa95fae0d1ceec2ff06e0e816939964b8:

net/sched: flower: Fix chain template offload (2024-01-24 01:33:59 +0000)

are available in the Git repository at:

git://git.kernel.org/pub/scm/linux/kernel/git/saeed/linux.git tags/mlx5-fixes-2024-01-24

for you to fetch changes up to aef855df7e1bbd5aa4484851561211500b22707e:

net/mlx5e: fix a potential double-free in fs_any_create_groups (2024-01-24 00:15:38 -0800)

----------------------------------------------------------------
mlx5-fixes-2024-01-24

----------------------------------------------------------------
Dinghao Liu (1):
net/mlx5e: fix a potential double-free in fs_any_create_groups

Leon Romanovsky (2):
net/mlx5e: Allow software parsing when IPsec crypto is enabled net/mlx5e: Ignore IPsec replay window values on sender side

Moshe Shemesh (1):
net/mlx5: Bridge, fix multicast packets sent to uplink

Rahul Rameshbabu (2):
net/mlx5e: Fix operation precedence bug in port timestamping napi_poll context net/mlx5: Use mlx5 device constant for selecting CQ period mode for ASO

Saeed Mahameed (1):
net/mlx5e: Use the correct lag ports number when creating TISes

Tariq Toukan (2):
net/mlx5: Fix query of sd_group field net/mlx5e: Fix inconsistent hairpin RQT sizes

Vlad Buslov (1):
net/mlx5e: Fix peer flow lists handling

Yevgeny Kliteynik (2):
net/mlx5: DR, Use the right GVMI number for drop action net/mlx5: DR, Can't go to uplink vport on RX rule

Yishai Hadas (1):
net/mlx5: Fix a WARN upon a callback command failure

Zhipeng Lu (1):
net/mlx5e: fix a double-free in arfs_create_groups

 drivers/net/ethernet/mellanox/mlx5/core/cmd.c      |  5 +++--
 drivers/net/ethernet/mellanox/mlx5/core/en.h       |  2 +-
 .../mellanox/mlx5/core/en/fs_tt_redirect.c         |  1 +
 .../net/ethernet/mellanox/mlx5/core/en/params.c    |  4 ++--
 drivers/net/ethernet/mellanox/mlx5/core/en/ptp.c   |  2 +-
 .../ethernet/mellanox/mlx5/core/en_accel/ipsec.c   | 10 +++++++--
 drivers/net/ethernet/mellanox/mlx5/core/en_arfs.c  | 26 +++++++++++++---------
 .../net/ethernet/mellanox/mlx5/core/en_common.c    | 21 ++++++++++-------
 drivers/net/ethernet/mellanox/mlx5/core/en_main.c  |  2 +-
 drivers/net/ethernet/mellanox/mlx5/core/en_tc.c    |  5 +++--
 .../ethernet/mellanox/mlx5/core/esw/bridge_mcast.c |  3 +++
 drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.c   |  2 ++
 .../net/ethernet/mellanox/mlx5/core/ipoib/ipoib.c  |  2 +-
 drivers/net/ethernet/mellanox/mlx5/core/lib/aso.c  |  2 +-
 .../mellanox/mlx5/core/steering/dr_action.c        | 17 +++++++++-----
 drivers/net/ethernet/mellanox/mlx5/core/vport.c    | 21 +++++++++++++++++
 include/linux/mlx5/driver.h                        |  1 +
 include/linux/mlx5/fs.h                            |  1 +
 include/linux/mlx5/mlx5_ifc.h                      | 12 ++++++----
 include/linux/mlx5/vport.h                         |  1 +
 20 files changed, 99 insertions(+), 41 deletions(-)
#1 lore raw

PATCH 1/14 net/mlx5e: Use the correct lag ports number when creating TISes

From: Saeed Mahameed <[email protected]>

The cited commit moved the code of mlx5e_create_tises() and changed the loop to create TISes over MLX5_MAX_PORTS constant value, instead of getting the correct lag ports supported by the device, which can cause FW errors on devices with less than MLX5_MAX_PORTS ports.

Change that back to mlx5e_get_num_lag_ports(mdev).

Also IPoIB interfaces create there own TISes, they don't use the eth TISes, pass a flag to indicate that.

This fixes the following errors that might appear in kernel log:
mlx5_cmd_out_err:808:(pid 650): CREATE_TIS(0x912) op_mod(0x0) failed, status bad parameter(0x3), syndrome (0x595b5d), err(-22) mlx5e_create_mdev_resources:174:(pid 650): alloc tises failed, -22

Fixes b25bd37c859f ("net/mlx5: Move TISes from priv to mdev HW resources")
---
 drivers/net/ethernet/mellanox/mlx5/core/en.h  |  2 +-
 .../ethernet/mellanox/mlx5/core/en_common.c   | 21 ++++++++++++-------
 .../net/ethernet/mellanox/mlx5/core/en_main.c |  2 +-
 .../ethernet/mellanox/mlx5/core/ipoib/ipoib.c |  2 +-
 include/linux/mlx5/driver.h                   |  1 +
 5 files changed, 17 insertions(+), 11 deletions(-)
drivers/net/ethernet/mellanox/mlx5/core/en.h +1 -1
index 0bfe1ca8a364..55c6ace0acd5 100644
@@ -1124,7 +1124,7 @@ static inline bool mlx5_tx_swp_supported(struct mlx5_core_dev *mdev)
 extern const struct ethtool_ops mlx5e_ethtool_ops;
 
 int mlx5e_create_mkey(struct mlx5_core_dev *mdev, u32 pdn, u32 *mkey);
-int mlx5e_create_mdev_resources(struct mlx5_core_dev *mdev);
+int mlx5e_create_mdev_resources(struct mlx5_core_dev *mdev, bool create_tises);
 void mlx5e_destroy_mdev_resources(struct mlx5_core_dev *mdev);
 int mlx5e_refresh_tirs(struct mlx5e_priv *priv, bool enable_uc_lb,
 		       bool enable_mc_lb);
drivers/net/ethernet/mellanox/mlx5/core/en_common.c +13 -8
index 67f546683e85..6ed3a32b7e22 100644
@@ -95,7 +95,7 @@ static void mlx5e_destroy_tises(struct mlx5_core_dev *mdev, u32 tisn[MLX5_MAX_PO
 {
 	int tc, i;
 
-	for (i = 0; i < MLX5_MAX_PORTS; i++)
+	for (i = 0; i < mlx5e_get_num_lag_ports(mdev); i++)
 		for (tc = 0; tc < MLX5_MAX_NUM_TC; tc++)
 			mlx5e_destroy_tis(mdev, tisn[i][tc]);
 }
@@ -110,7 +110,7 @@ static int mlx5e_create_tises(struct mlx5_core_dev *mdev, u32 tisn[MLX5_MAX_PORT
 	int tc, i;
 	int err;
 
-	for (i = 0; i < MLX5_MAX_PORTS; i++) {
+	for (i = 0; i < mlx5e_get_num_lag_ports(mdev); i++) {
 		for (tc = 0; tc < MLX5_MAX_NUM_TC; tc++) {
 			u32 in[MLX5_ST_SZ_DW(create_tis_in)] = {};
 			void *tisc;
@@ -140,7 +140,7 @@ static int mlx5e_create_tises(struct mlx5_core_dev *mdev, u32 tisn[MLX5_MAX_PORT
 	return err;
 }
 
-int mlx5e_create_mdev_resources(struct mlx5_core_dev *mdev)
+int mlx5e_create_mdev_resources(struct mlx5_core_dev *mdev, bool create_tises)
 {
 	struct mlx5e_hw_objs *res = &mdev->mlx5e_res.hw_objs;
 	int err;
@@ -169,11 +169,15 @@ int mlx5e_create_mdev_resources(struct mlx5_core_dev *mdev)
 		goto err_destroy_mkey;
 	}
 
-	err = mlx5e_create_tises(mdev, res->tisn);
-	if (err) {
-		mlx5_core_err(mdev, "alloc tises failed, %d\n", err);
-		goto err_destroy_bfreg;
+	if (create_tises) {
+		err = mlx5e_create_tises(mdev, res->tisn);
+		if (err) {
+			mlx5_core_err(mdev, "alloc tises failed, %d\n", err);
+			goto err_destroy_bfreg;
+		}
+		res->tisn_valid = true;
 	}
+
 	INIT_LIST_HEAD(&res->td.tirs_list);
 	mutex_init(&res->td.list_lock);
 
@@ -203,7 +207,8 @@ void mlx5e_destroy_mdev_resources(struct mlx5_core_dev *mdev)
 
 	mlx5_crypto_dek_cleanup(mdev->mlx5e_res.dek_priv);
 	mdev->mlx5e_res.dek_priv = NULL;
-	mlx5e_destroy_tises(mdev, res->tisn);
+	if (res->tisn_valid)
+		mlx5e_destroy_tises(mdev, res->tisn);
 	mlx5_free_bfreg(mdev, &res->bfreg);
 	mlx5_core_destroy_mkey(mdev, res->mkey);
 	mlx5_core_dealloc_transport_domain(mdev, res->td.tdn);
drivers/net/ethernet/mellanox/mlx5/core/en_main.c +1 -1
index b5f1c4ca38ba..c8e8f512803e 100644
@@ -5992,7 +5992,7 @@ static int mlx5e_resume(struct auxiliary_device *adev)
 	if (netif_device_present(netdev))
 		return 0;
 
-	err = mlx5e_create_mdev_resources(mdev);
+	err = mlx5e_create_mdev_resources(mdev, true);
 	if (err)
 		return err;
 
drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib.c +1 -1
index 58845121954c..d77be1b4dd9c 100644
@@ -783,7 +783,7 @@ static int mlx5_rdma_setup_rn(struct ib_device *ibdev, u32 port_num,
 		}
 
 		/* This should only be called once per mdev */
-		err = mlx5e_create_mdev_resources(mdev);
+		err = mlx5e_create_mdev_resources(mdev, false);
 		if (err)
 			goto destroy_ht;
 	}
include/linux/mlx5/driver.h +1 -0
index 8c55ff351e5f..41f03b352401 100644
@@ -681,6 +681,7 @@ struct mlx5e_resources {
 		struct mlx5_sq_bfreg       bfreg;
 #define MLX5_MAX_NUM_TC 8
 		u32                        tisn[MLX5_MAX_PORTS][MLX5_MAX_NUM_TC];
+		bool			   tisn_valid;
 	} hw_objs;
 	struct net_device *uplink_netdev;
 	struct mutex uplink_netdev_lock;
Signature · 1 line
2.43.0
#2 lore raw

Re: [net 01/14] net/mlx5e: Use the correct lag ports number when creating TISes

This series was applied to netdev/net.git (main) by Saeed Mahameed <[email protected]>:

Hello:

This series was applied to netdev/net.git (main)
by Saeed Mahameed <[email protected]>:

Quoted 8 lines · On Wed, 24 Jan 2024 00:18:42 -0800 you wrote:

From: Saeed Mahameed <[email protected]>

The cited commit moved the code of mlx5e_create_tises() and changed the loop to create TISes over MLX5_MAX_PORTS constant value, instead of getting the correct lag ports supported by the device, which can cause FW errors on devices with less than MLX5_MAX_PORTS ports.

[...]

Here is the summary with links:
- [net,01/14] net/mlx5e: Use the correct lag ports number when creating TISes

    https://git.kernel.org/netdev/net/c/25461ce8b3d2
  - [net,02/14] net/mlx5: Fix query of sd_group field
    https://git.kernel.org/netdev/net/c/cfbc3608a8c6
  - [net,03/14] net/mlx5e: Fix operation precedence bug in port timestamping napi_poll context
    https://git.kernel.org/netdev/net/c/3876638b2c7e
  - [net,04/14] net/mlx5e: Fix inconsistent hairpin RQT sizes
    https://git.kernel.org/netdev/net/c/c20767fd45e8
  - [net,05/14] net/mlx5e: Fix peer flow lists handling
    https://git.kernel.org/netdev/net/c/d76fdd31f953
  - [net,06/14] net/mlx5: Fix a WARN upon a callback command failure
    https://git.kernel.org/netdev/net/c/cc8091587779
  - [net,07/14] net/mlx5: Bridge, fix multicast packets sent to uplink
    https://git.kernel.org/netdev/net/c/ec7cc38ef9f8
  - [net,08/14] net/mlx5: DR, Use the right GVMI number for drop action
    https://git.kernel.org/netdev/net/c/5665954293f1
  - [net,09/14] net/mlx5: DR, Can't go to uplink vport on RX rule
    https://git.kernel.org/netdev/net/c/5b2a2523eeea
  - [net,10/14] net/mlx5: Use mlx5 device constant for selecting CQ period mode for ASO
    https://git.kernel.org/netdev/net/c/20cbf8cbb827
  - [net,11/14] net/mlx5e: Allow software parsing when IPsec crypto is enabled
    https://git.kernel.org/netdev/net/c/20f5468a7988
  - [net,12/14] net/mlx5e: Ignore IPsec replay window values on sender side
    https://git.kernel.org/netdev/net/c/315a597f9bcf
  - [net,13/14] net/mlx5e: fix a double-free in arfs_create_groups
    https://git.kernel.org/netdev/net/c/3c6d5189246f
  - [net,14/14] net/mlx5e: fix a potential double-free in fs_any_create_groups
    https://git.kernel.org/netdev/net/c/aef855df7e1b

You are awesome, thank you!

Signature · 2 lines
Deet-doot-dot, I am a bot.
https://korg.docs.kernel.org/patchwork/pwbot.html
#3 lore raw

PATCH 2/14 net/mlx5: Fix query of sd_group field

In reply to Saeed Mahameed

From: Tariq Toukan <[email protected]>

The sd_group field moved in the HW spec from the MPIR register to the vport context.
Align the query accordingly.

Fixes f5e956329960 ("net/mlx5: Expose Management PCIe Index Register (MPIR)")
---
 .../net/ethernet/mellanox/mlx5/core/vport.c   | 21 +++++++++++++++++++
 include/linux/mlx5/mlx5_ifc.h                 | 10 ++++++---
 include/linux/mlx5/vport.h                    |  1 +
 3 files changed, 29 insertions(+), 3 deletions(-)
drivers/net/ethernet/mellanox/mlx5/core/vport.c +21 -0
index 21753f327868..1005bb6935b6 100644
@@ -440,6 +440,27 @@ int mlx5_query_nic_vport_system_image_guid(struct mlx5_core_dev *mdev,
 }
 EXPORT_SYMBOL_GPL(mlx5_query_nic_vport_system_image_guid);
 
+int mlx5_query_nic_vport_sd_group(struct mlx5_core_dev *mdev, u8 *sd_group)
+{
+	int outlen = MLX5_ST_SZ_BYTES(query_nic_vport_context_out);
+	u32 *out;
+	int err;
+
+	out = kvzalloc(outlen, GFP_KERNEL);
+	if (!out)
+		return -ENOMEM;
+
+	err = mlx5_query_nic_vport_context(mdev, 0, out);
+	if (err)
+		goto out;
+
+	*sd_group = MLX5_GET(query_nic_vport_context_out, out,
+			     nic_vport_context.sd_group);
+out:
+	kvfree(out);
+	return err;
+}
+
 int mlx5_query_nic_vport_node_guid(struct mlx5_core_dev *mdev, u64 *node_guid)
 {
 	u32 *out;
include/linux/mlx5/mlx5_ifc.h +7 -3
index bf5320b28b8b..37230253f9f1 100644
@@ -4036,8 +4036,13 @@ struct mlx5_ifc_nic_vport_context_bits {
 	u8	   affiliation_criteria[0x4];
 	u8	   affiliated_vhca_id[0x10];
 
-	u8	   reserved_at_60[0xd0];
+	u8	   reserved_at_60[0xa0];
 
+	u8	   reserved_at_100[0x1];
+	u8         sd_group[0x3];
+	u8	   reserved_at_104[0x1c];
+
+	u8	   reserved_at_120[0x10];
 	u8         mtu[0x10];
 
 	u8         system_image_guid[0x40];
@@ -10122,8 +10127,7 @@ struct mlx5_ifc_mpir_reg_bits {
 	u8         reserved_at_20[0x20];
 
 	u8         local_port[0x8];
-	u8         reserved_at_28[0x15];
-	u8         sd_group[0x3];
+	u8         reserved_at_28[0x18];
 
 	u8         reserved_at_60[0x20];
 };
include/linux/mlx5/vport.h +1 -0
index fbb9bf447889..c36cc6d82926 100644
@@ -72,6 +72,7 @@ int mlx5_query_nic_vport_mtu(struct mlx5_core_dev *mdev, u16 *mtu);
 int mlx5_modify_nic_vport_mtu(struct mlx5_core_dev *mdev, u16 mtu);
 int mlx5_query_nic_vport_system_image_guid(struct mlx5_core_dev *mdev,
 					   u64 *system_image_guid);
+int mlx5_query_nic_vport_sd_group(struct mlx5_core_dev *mdev, u8 *sd_group);
 int mlx5_query_nic_vport_node_guid(struct mlx5_core_dev *mdev, u64 *node_guid);
 int mlx5_modify_nic_vport_node_guid(struct mlx5_core_dev *mdev,
 				    u16 vport, u64 node_guid);
Signature · 1 line
2.43.0
#4 lore raw

PATCH 3/14 net/mlx5e: Fix operation precedence bug in port timestamping napi_poll context

In reply to Saeed Mahameed

From: Rahul Rameshbabu <[email protected]>

Indirection (*) is of lower precedence than postfix increment (++). Logic in napi_poll context would cause an out-of-bound read by first increment the pointer address by byte address space and then dereference the value. Rather, the intended logic was to dereference first and then increment the underlying value.

Fixes 92214be5979c ("net/mlx5e: Update doorbell for port timestamping CQ before the software counter")
---
 drivers/net/ethernet/mellanox/mlx5/core/en/ptp.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/net/ethernet/mellanox/mlx5/core/en/ptp.c +1 -1
index c206cc0a8483..078f56a3cbb2 100644
@@ -213,7 +213,7 @@ static void mlx5e_ptp_handle_ts_cqe(struct mlx5e_ptpsq *ptpsq,
 	mlx5e_ptpsq_mark_ts_cqes_undelivered(ptpsq, hwtstamp);
 out:
 	napi_consume_skb(skb, budget);
-	md_buff[*md_buff_sz++] = metadata_id;
+	md_buff[(*md_buff_sz)++] = metadata_id;
 	if (unlikely(mlx5e_ptp_metadata_map_unhealthy(&ptpsq->metadata_map)) &&
 	    !test_and_set_bit(MLX5E_SQ_STATE_RECOVERING, &sq->state))
 		queue_work(ptpsq->txqsq.priv->wq, &ptpsq->report_unhealthy_work);
Signature · 1 line
2.43.0
#5 lore raw

PATCH 4/14 net/mlx5e: Fix inconsistent hairpin RQT sizes

In reply to Saeed Mahameed

From: Tariq Toukan <[email protected]>

The processing of traffic in hairpin queues occurs in HW/FW and does not involve the cpus, hence the upper bound on max num channels does not apply to them. Using this bound for the hairpin RQT max_table_size is wrong. It could be too small, and cause the error below [1]. As the RQT size provided on init does not get modified later, use the same value for both actual and max table sizes.

[1]
mlx5_core 0000:08:00.1: mlx5_cmd_out_err:805:(pid 1200): CREATE_RQT(0x916) op_mod(0x0) failed, status bad parameter(0x3), syndrome (0x538faf), err(-22)

Fixes 74a8dadac17e ("net/mlx5e: Preparations for supporting larger number of channels")
---
 drivers/net/ethernet/mellanox/mlx5/core/en_tc.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/net/ethernet/mellanox/mlx5/core/en_tc.c +1 -1
index 30932c9c9a8f..047b465fc6a5 100644
@@ -761,7 +761,7 @@ static int mlx5e_hairpin_create_indirect_rqt(struct mlx5e_hairpin *hp)
 
 	err = mlx5e_rss_params_indir_init(&indir, mdev,
 					  mlx5e_rqt_size(mdev, hp->num_channels),
-					  mlx5e_rqt_size(mdev, priv->max_nch));
+					  mlx5e_rqt_size(mdev, hp->num_channels));
 	if (err)
 		return err;
 
Signature · 1 line
2.43.0
#6 lore raw

PATCH 5/14 net/mlx5e: Fix peer flow lists handling

In reply to Saeed Mahameed

From: Vlad Buslov <[email protected]>

The cited change refactored mlx5e_tc_del_fdb_peer_flow() to only clear DUP flag when list of peer flows has become empty. However, if any concurrent user holds a reference to a peer flow (for example, the neighbor update workqueue task is updating peer flow's parent encap entry concurrently), then the flow will not be removed from the peer list and, consecutively, DUP flag will remain set. Since mlx5e_tc_del_fdb_peers_flow() calls mlx5e_tc_del_fdb_peer_flow() for every possible peer index the algorithm will try to remove the flow from eswitch instances that it has never peered with causing either NULL pointer dereference when trying to remove the flow peer list head of peer_index that was never initialized or a warning if the list debug config is enabled[0].

Fix the issue by always removing the peer flow from the list even when not releasing the last reference to it.

[0]:

[ 3102.985806] ------------[ cut here ]------------
[ 3102.986223] list_del corruption, ffff888139110698->next is NULL
[ 3102.986757] WARNING: CPU: 2 PID: 22109 at lib/list_debug.c:53 __list_del_entry_valid_or_report+0x4f/0xc0
[ 3102.987561] Modules linked in: act_ct nf_flow_table bonding act_tunnel_key act_mirred act_skbedit vxlan cls_matchall nfnetlink_cttimeout act_gact cls_flower sch_ingress mlx5_vdpa vringh vhost_iotlb vdpa openvswitch nsh xt_MASQUERADE nf_conntrack_netlink nfnetlink iptable_nat xt_addrtype xt_conntrack nf_nat br_netfilter rpcsec_gss_krb5 auth_rpcg
ss oid_registry overlay rpcrdma rdma_ucm ib_iser libiscsi scsi_transport_iscsi ib_umad rdma_cm ib_ipoib iw_cm ib_cm mlx5_ib ib_uverbs ib_core mlx5_core [last unloaded: bonding]
[ 3102.991113] CPU: 2 PID: 22109 Comm: revalidator28 Not tainted 6.6.0-rc6+ #3
[ 3102.991695] Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS rel-1.13.0-0-gf21b5a4aeb02-prebuilt.qemu.org 04/01/2014
[ 3102.992605] RIP: 0010:__list_del_entry_valid_or_report+0x4f/0xc0
[ 3102.993122] Code: 39 c2 74 56 48 8b 32 48 39 fe 75 62 48 8b 51 08 48 39 f2 75 73 b8 01 00 00 00 c3 48 89 fe 48 c7 c7 48 fd 0a 82 e8 41 0b ad ff <0f> 0b 31 c0 c3 48 89 fe 48 c7 c7 70 fd 0a 82 e8 2d 0b ad ff 0f 0b
[ 3102.994615] RSP: 0018:ffff8881383e7710 EFLAGS: 00010286
[ 3102.995078] RAX: 0000000000000000 RBX: 0000000000000002 RCX: 0000000000000000
[ 3102.995670] RDX: 0000000000000001 RSI: ffff88885f89b640 RDI: ffff88885f89b640
[ 3102.997188] DEL flow 00000000be367878 on port 0
[ 3102.998594] RBP: dead000000000122 R08: 0000000000000000 R09: c0000000ffffdfff
[ 3102.999604] R10: 0000000000000008 R11: ffff8881383e7598 R12: dead000000000100
[ 3103.000198] R13: 0000000000000002 R14: ffff888139110000 R15: ffff888101901240
[ 3103.000790] FS:  00007f424cde4700(0000) GS:ffff88885f880000(0000) knlGS:0000000000000000
[ 3103.001486] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 3103.001986] CR2: 00007fd42e8dcb70 CR3: 000000011e68a003 CR4: 0000000000370ea0
[ 3103.002596] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[ 3103.003190] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[ 3103.003787] Call Trace:
[ 3103.004055]  <TASK>
[ 3103.004297]  ? __warn+0x7d/0x130
[ 3103.004623]  ? __list_del_entry_valid_or_report+0x4f/0xc0
[ 3103.005094]  ? report_bug+0xf1/0x1c0
[ 3103.005439]  ? console_unlock+0x4a/0xd0
[ 3103.005806]  ? handle_bug+0x3f/0x70
[ 3103.006149]  ? exc_invalid_op+0x13/0x60
[ 3103.006531]  ? asm_exc_invalid_op+0x16/0x20
[ 3103.007430]  ? __list_del_entry_valid_or_report+0x4f/0xc0
[ 3103.007910]  mlx5e_tc_del_fdb_peers_flow+0xcf/0x240 [mlx5_core]
[ 3103.008463]  mlx5e_tc_del_flow+0x46/0x270 [mlx5_core]
[ 3103.008944]  mlx5e_flow_put+0x26/0x50 [mlx5_core]
[ 3103.009401]  mlx5e_delete_flower+0x25f/0x380 [mlx5_core]
[ 3103.009901]  tc_setup_cb_destroy+0xab/0x180
[ 3103.010292]  fl_hw_destroy_filter+0x99/0xc0 [cls_flower]
[ 3103.010779]  __fl_delete+0x2d4/0x2f0 [cls_flower]
[ 3103.011207]  fl_delete+0x36/0x80 [cls_flower]
[ 3103.011614]  tc_del_tfilter+0x56f/0x750
[ 3103.011982]  rtnetlink_rcv_msg+0xff/0x3a0
[ 3103.012362]  ? netlink_ack+0x1c7/0x4e0
[ 3103.012719]  ? rtnl_calcit.isra.44+0x130/0x130
[ 3103.013134]  netlink_rcv_skb+0x54/0x100
[ 3103.013533]  netlink_unicast+0x1ca/0x2b0
[ 3103.013902]  netlink_sendmsg+0x361/0x4d0
[ 3103.014269]  __sock_sendmsg+0x38/0x60
[ 3103.014643]  ____sys_sendmsg+0x1f2/0x200
[ 3103.015018]  ? copy_msghdr_from_user+0x72/0xa0
[ 3103.015265]  ___sys_sendmsg+0x87/0xd0
[ 3103.016608]  ? copy_msghdr_from_user+0x72/0xa0
[ 3103.017014]  ? ___sys_recvmsg+0x9b/0xd0
[ 3103.017381]  ? ttwu_do_activate.isra.137+0x58/0x180
[ 3103.017821]  ? wake_up_q+0x49/0x90
[ 3103.018157]  ? futex_wake+0x137/0x160
[ 3103.018521]  ? __sys_sendmsg+0x51/0x90
[ 3103.018882]  __sys_sendmsg+0x51/0x90
[ 3103.019230]  ? exit_to_user_mode_prepare+0x56/0x130
[ 3103.019670]  do_syscall_64+0x3c/0x80
[ 3103.020017]  entry_SYSCALL_64_after_hwframe+0x46/0xb0
[ 3103.020469] RIP: 0033:0x7f4254811ef4
[ 3103.020816] Code: 89 f3 48 83 ec 10 48 89 7c 24 08 48 89 14 24 e8 42 eb ff ff 48 8b 14 24 41 89 c0 48 89 de 48 8b 7c 24 08 b8 2e 00 00 00 0f 05 <48> 3d 00 f0 ff ff 77 30 44 89 c7 48 89 04 24 e8 78 eb ff ff 48 8b
[ 3103.022290] RSP: 002b:00007f424cdd9480 EFLAGS: 00000293 ORIG_RAX: 000000000000002e
[ 3103.022970] RAX: ffffffffffffffda RBX: 00007f424cdd9510 RCX: 00007f4254811ef4
[ 3103.023564] RDX: 0000000000000000 RSI: 00007f424cdd9510 RDI: 0000000000000012
[ 3103.024158] RBP: 00007f424cdda238 R08: 0000000000000000 R09: 00007f41d801a4b0
[ 3103.024748] R10: 0000000000000000 R11: 0000000000000293 R12: 0000000000000001
[ 3103.025341] R13: 00007f424cdd9510 R14: 00007f424cdda240 R15: 00007f424cdd99a0
[ 3103.025931]  </TASK>
[ 3103.026182] ---[ end trace 0000000000000000 ]---
[ 3103.027033] ------------[ cut here ]------------
Fixes 9be6c21fdcf8 ("net/mlx5e: Handle offloads flows per peer")
---
 drivers/net/ethernet/mellanox/mlx5/core/en_tc.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
drivers/net/ethernet/mellanox/mlx5/core/en_tc.c +2 -1
index 047b465fc6a5..9fb2c057bd78 100644
@@ -2014,9 +2014,10 @@ static void mlx5e_tc_del_fdb_peer_flow(struct mlx5e_tc_flow *flow,
 	list_for_each_entry_safe(peer_flow, tmp, &flow->peer_flows, peer_flows) {
 		if (peer_index != mlx5_get_dev_index(peer_flow->priv->mdev))
 			continue;
+
+		list_del(&peer_flow->peer_flows);
 		if (refcount_dec_and_test(&peer_flow->refcnt)) {
 			mlx5e_tc_del_fdb_flow(peer_flow->priv, peer_flow);
-			list_del(&peer_flow->peer_flows);
 			kfree(peer_flow);
 		}
 	}
Signature · 1 line
2.43.0
#7 lore raw

PATCH 6/14 net/mlx5: Fix a WARN upon a callback command failure

In reply to Saeed Mahameed

From: Yishai Hadas <[email protected]>

The below WARN [1] is reported once a callback command failed.

As a callback runs under an interrupt context, needs to use the IRQ save/restore variant.

[1]
DEBUG_LOCKS_WARN_ON(lockdep_hardirq_context())

WARNING: CPU: 15 PID: 0 at kernel/locking/lockdep.c:4353
              lockdep_hardirqs_on_prepare+0x11b/0x180
Modules linked in: vhost_net vhost tap mlx5_vfio_pci

vfio_pci vfio_pci_core vfio_iommu_type1 vfio mlx5_vdpa vringh vhost_iotlb vdpa nfnetlink_cttimeout openvswitch nsh ip6table_mangle ip6table_nat ip6table_filter ip6_tables iptable_mangle xt_conntrackxt_MASQUERADE nf_conntrack_netlink nfnetlink xt_addrtype iptable_nat nf_nat br_netfilter rpcsec_gss_krb5 auth_rpcgss oid_registry overlay rpcrdma rdma_ucm ib_iser libiscsi scsi_transport_iscsi rdma_cm iw_cm ib_umad ib_ipoib ib_cm mlx5_ib ib_uverbs ib_core fuse mlx5_core

CPU: 15 PID: 0 Comm: swapper/15 Tainted: G        W 6.7.0-rc4+ #1587
Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS
rel-1.13.0-0-gf21b5a4aeb02-prebuilt.qemu.org 04/01/2014
RIP: 0010:lockdep_hardirqs_on_prepare+0x11b/0x180
Code: 00 5b c3 c3 e8 e6 0d 58 00 85 c0 74 d6 8b 15 f0 c3
      76 01 85 d2 75 cc 48 c7 c6 04 a5 3b 82 48 c7 c7 f1
      e9 39 82 e8 95 12 f9 ff <0f> 0b 5b c3 e8 bc 0d 58 00
      85 c0 74 ac 8b 3d c6 c3 76 01 85 ff 75
RSP: 0018:ffffc900003ecd18 EFLAGS: 00010086
RAX: 0000000000000000 RBX: 0000000000000000 RCX: 0000000000000027
RDX: 0000000000000000 RSI: ffff88885fbdb880 RDI: ffff88885fbdb888
RBP: 00000000ffffff87 R08: 0000000000000000 R09: 0000000000000001
R10: 0000000000000000 R11: 284e4f5f4e524157 R12: 00000000002c9aa1
R13: ffff88810aace980 R14: ffff88810aace9b8 R15: 0000000000000003
FS:  0000000000000000(0000) GS:ffff88885fbc0000(0000)
knlGS:0000000000000000
CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 00007f731436f4c8 CR3: 000000010aae6001 CR4: 0000000000372eb0
DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
Call Trace:
 <IRQ>
? __warn+0x81/0x170
? lockdep_hardirqs_on_prepare+0x11b/0x180
? report_bug+0xf8/0x1c0
? handle_bug+0x3f/0x70
? exc_invalid_op+0x13/0x60
? asm_exc_invalid_op+0x16/0x20
? lockdep_hardirqs_on_prepare+0x11b/0x180
? lockdep_hardirqs_on_prepare+0x11b/0x180
trace_hardirqs_on+0x4a/0xa0
raw_spin_unlock_irq+0x24/0x30
cmd_status_err+0xc0/0x1a0 [mlx5_core]
cmd_status_err+0x1a0/0x1a0 [mlx5_core]
mlx5_cmd_exec_cb_handler+0x24/0x40 [mlx5_core]
mlx5_cmd_comp_handler+0x129/0x4b0 [mlx5_core]
cmd_comp_notifier+0x1a/0x20 [mlx5_core]
notifier_call_chain+0x3e/0xe0
atomic_notifier_call_chain+0x5f/0x130
mlx5_eq_async_int+0xe7/0x200 [mlx5_core]
notifier_call_chain+0x3e/0xe0
atomic_notifier_call_chain+0x5f/0x130
irq_int_handler+0x11/0x20 [mlx5_core]
__handle_irq_event_percpu+0x99/0x220
? tick_irq_enter+0x5d/0x80
handle_irq_event_percpu+0xf/0x40
handle_irq_event+0x3a/0x60
handle_edge_irq+0xa2/0x1c0
__common_interrupt+0x55/0x140
common_interrupt+0x7d/0xa0
</IRQ>
<TASK>
asm_common_interrupt+0x22/0x40
RIP: 0010:default_idle+0x13/0x20
Code: c0 08 00 00 00 4d 29 c8 4c 01 c7 4c 29 c2 e9 72 ff

ff ff cc cc cc cc 8b 05 ea 08 25 01 85 c0 7e 07 0f 00 2d 7f b0 26 00 fb f4 <fa> c3 90 66 2e 0f 1f 84 00 00 00 00 00 65 48 8b 04 25 80 d0 02 00
RSP: 0018:ffffc9000010fec8 EFLAGS: 00000242
RAX: 0000000000000001 RBX: 000000000000000f RCX: 4000000000000000
RDX: 0000000000000000 RSI: 0000000000000000 RDI: ffffffff811c410c
RBP: ffffffff829478c0 R08: 0000000000000001 R09: 0000000000000001
R10: 0000000000000000 R11: 0000000000000000 R12: 0000000000000000
R13: 0000000000000000 R14: 0000000000000000 R15: 0000000000000000

? do_idle+0x1ec/0x210
default_idle_call+0x6c/0x90
do_idle+0x1ec/0x210
cpu_startup_entry+0x26/0x30
start_secondary+0x11b/0x150
secondary_startup_64_no_verify+0x165/0x16b
</TASK>
irq event stamp: 833284
hardirqs last  enabled at (833283): [<ffffffff811c410c>]
do_idle+0x1ec/0x210
hardirqs last disabled at (833284): [<ffffffff81daf9ef>]
common_interrupt+0xf/0xa0
softirqs last  enabled at (833224): [<ffffffff81dc199f>]
__do_softirq+0x2bf/0x40e
softirqs last disabled at (833177): [<ffffffff81178ddf>]
irq_exit_rcu+0x7f/0xa0
Fixes 34f46ae0d4b3 ("net/mlx5: Add command failures data to debugfs")
---
 drivers/net/ethernet/mellanox/mlx5/core/cmd.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)
drivers/net/ethernet/mellanox/mlx5/core/cmd.c +3 -2
index a7b1f9686c09..4957412ff1f6 100644
@@ -1923,6 +1923,7 @@ static void cmd_status_log(struct mlx5_core_dev *dev, u16 opcode, u8 status,
 {
 	const char *namep = mlx5_command_str(opcode);
 	struct mlx5_cmd_stats *stats;
+	unsigned long flags;
 
 	if (!err || !(strcmp(namep, "unknown command opcode")))
 		return;
@@ -1930,7 +1931,7 @@ static void cmd_status_log(struct mlx5_core_dev *dev, u16 opcode, u8 status,
 	stats = xa_load(&dev->cmd.stats, opcode);
 	if (!stats)
 		return;
-	spin_lock_irq(&stats->lock);
+	spin_lock_irqsave(&stats->lock, flags);
 	stats->failed++;
 	if (err < 0)
 		stats->last_failed_errno = -err;
@@ -1939,7 +1940,7 @@ static void cmd_status_log(struct mlx5_core_dev *dev, u16 opcode, u8 status,
 		stats->last_failed_mbox_status = status;
 		stats->last_failed_syndrome = syndrome;
 	}
-	spin_unlock_irq(&stats->lock);
+	spin_unlock_irqrestore(&stats->lock, flags);
 }
 
 /* preserve -EREMOTEIO for outbox.status != OK, otherwise return err as is */
Signature · 1 line
2.43.0
#8 lore raw

PATCH 7/14 net/mlx5: Bridge, fix multicast packets sent to uplink

In reply to Saeed Mahameed

From: Moshe Shemesh <[email protected]>

To enable multicast packets which are offloaded in bridge multicast offload mode to be sent also to uplink, FTE bit uplink_hairpin_en should be set. Add this bit to FTE for the bridge multicast offload rules.

Fixes 18c2916cee12 ("net/mlx5: Bridge, snoop igmp/mld packets")
---
 drivers/net/ethernet/mellanox/mlx5/core/esw/bridge_mcast.c | 3 +++
 drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.c           | 2 ++
 include/linux/mlx5/fs.h                                    | 1 +
 include/linux/mlx5/mlx5_ifc.h                              | 2 +-
 4 files changed, 7 insertions(+), 1 deletion(-)
drivers/net/ethernet/mellanox/mlx5/core/esw/bridge_mcast.c +3 -0
index a7ed87e9d842..22dd30cf8033 100644
@@ -83,6 +83,7 @@ mlx5_esw_bridge_mdb_flow_create(u16 esw_owner_vhca_id, struct mlx5_esw_bridge_md
 		i++;
 	}
 
+	rule_spec->flow_context.flags |= FLOW_CONTEXT_UPLINK_HAIRPIN_EN;
 	rule_spec->match_criteria_enable = MLX5_MATCH_OUTER_HEADERS;
 	dmac_v = MLX5_ADDR_OF(fte_match_param, rule_spec->match_value, outer_headers.dmac_47_16);
 	ether_addr_copy(dmac_v, entry->key.addr);
@@ -587,6 +588,7 @@ mlx5_esw_bridge_mcast_vlan_flow_create(u16 vlan_proto, struct mlx5_esw_bridge_po
 	if (!rule_spec)
 		return ERR_PTR(-ENOMEM);
 
+	rule_spec->flow_context.flags |= FLOW_CONTEXT_UPLINK_HAIRPIN_EN;
 	rule_spec->match_criteria_enable = MLX5_MATCH_OUTER_HEADERS;
 
 	flow_act.action |= MLX5_FLOW_CONTEXT_ACTION_PACKET_REFORMAT;
@@ -662,6 +664,7 @@ mlx5_esw_bridge_mcast_fwd_flow_create(struct mlx5_esw_bridge_port *port)
 		dest.vport.flags = MLX5_FLOW_DEST_VPORT_VHCA_ID;
 		dest.vport.vhca_id = port->esw_owner_vhca_id;
 	}
+	rule_spec->flow_context.flags |= FLOW_CONTEXT_UPLINK_HAIRPIN_EN;
 	handle = mlx5_add_flow_rules(port->mcast.ft, rule_spec, &flow_act, &dest, 1);
 
 	kvfree(rule_spec);
drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.c +2 -0
index 1616a6144f7b..9b8599c200e2 100644
@@ -566,6 +566,8 @@ static int mlx5_cmd_set_fte(struct mlx5_core_dev *dev,
 		 fte->flow_context.flow_tag);
 	MLX5_SET(flow_context, in_flow_context, flow_source,
 		 fte->flow_context.flow_source);
+	MLX5_SET(flow_context, in_flow_context, uplink_hairpin_en,
+		 !!(fte->flow_context.flags & FLOW_CONTEXT_UPLINK_HAIRPIN_EN));
 
 	MLX5_SET(flow_context, in_flow_context, extended_destination,
 		 extended_dest);
include/linux/mlx5/fs.h +1 -0
index 6f7725238abc..3fb428ce7d1c 100644
@@ -132,6 +132,7 @@ struct mlx5_flow_handle;
 
 enum {
 	FLOW_CONTEXT_HAS_TAG = BIT(0),
+	FLOW_CONTEXT_UPLINK_HAIRPIN_EN = BIT(1),
 };
 
 struct mlx5_flow_context {
include/linux/mlx5/mlx5_ifc.h +1 -1
index 37230253f9f1..c726f90ab752 100644
@@ -3576,7 +3576,7 @@ struct mlx5_ifc_flow_context_bits {
 	u8         action[0x10];
 
 	u8         extended_destination[0x1];
-	u8         reserved_at_81[0x1];
+	u8         uplink_hairpin_en[0x1];
 	u8         flow_source[0x2];
 	u8         encrypt_decrypt_type[0x4];
 	u8         destination_list_size[0x18];
Signature · 1 line
2.43.0
#9 lore raw

PATCH 8/14 net/mlx5: DR, Use the right GVMI number for drop action

In reply to Saeed Mahameed

From: Yevgeny Kliteynik <[email protected]>

When FW provides ICM addresses for drop RX/TX, the provided capability is 64 bits that contain its GVMI as well as the ICM address itself. In case of TX DROP this GVMI is different from the GVMI that the domain is operating on.

This patch fixes the action to use these GVMI IDs, as provided by FW.

Fixes 9db810ed2d37 ("net/mlx5: DR, Expose steering action functionality")
---
 drivers/net/ethernet/mellanox/mlx5/core/steering/dr_action.c | 1 +
 1 file changed, 1 insertion(+)
drivers/net/ethernet/mellanox/mlx5/core/steering/dr_action.c +1 -0
index 6f9790e97fed..95517c4aca0f 100644
@@ -788,6 +788,7 @@ int mlx5dr_actions_build_ste_arr(struct mlx5dr_matcher *matcher,
 		switch (action_type) {
 		case DR_ACTION_TYP_DROP:
 			attr.final_icm_addr = nic_dmn->drop_icm_addr;
+			attr.hit_gvmi = nic_dmn->drop_icm_addr >> 48;
 			break;
 		case DR_ACTION_TYP_FT:
 			dest_action = action;
Signature · 1 line
2.43.0
#10 lore raw

PATCH 9/14 net/mlx5: DR, Can't go to uplink vport on RX rule

In reply to Saeed Mahameed

From: Yevgeny Kliteynik <[email protected]>

Go-To-Vport action on RX is not allowed when the vport is uplink. In such case, the packet should be dropped.

Fixes 9db810ed2d37 ("net/mlx5: DR, Expose steering action functionality")
---
 .../mellanox/mlx5/core/steering/dr_action.c      | 16 +++++++++++-----
 1 file changed, 11 insertions(+), 5 deletions(-)
drivers/net/ethernet/mellanox/mlx5/core/steering/dr_action.c +11 -5
index 95517c4aca0f..2ebb61ef3ea9 100644
@@ -874,11 +874,17 @@ int mlx5dr_actions_build_ste_arr(struct mlx5dr_matcher *matcher,
 							action->sampler->tx_icm_addr;
 			break;
 		case DR_ACTION_TYP_VPORT:
-			attr.hit_gvmi = action->vport->caps->vhca_gvmi;
-			dest_action = action;
-			attr.final_icm_addr = rx_rule ?
-				action->vport->caps->icm_address_rx :
-				action->vport->caps->icm_address_tx;
+			if (unlikely(rx_rule && action->vport->caps->num == MLX5_VPORT_UPLINK)) {
+				/* can't go to uplink on RX rule - dropping instead */
+				attr.final_icm_addr = nic_dmn->drop_icm_addr;
+				attr.hit_gvmi = nic_dmn->drop_icm_addr >> 48;
+			} else {
+				attr.hit_gvmi = action->vport->caps->vhca_gvmi;
+				dest_action = action;
+				attr.final_icm_addr = rx_rule ?
+						      action->vport->caps->icm_address_rx :
+						      action->vport->caps->icm_address_tx;
+			}
 			break;
 		case DR_ACTION_TYP_POP_VLAN:
 			if (!rx_rule && !(dmn->ste_ctx->actions_caps &
Signature · 1 line
2.43.0
#11 lore raw

PATCH 10/14 net/mlx5: Use mlx5 device constant for selecting CQ period mode for ASO

In reply to Saeed Mahameed

From: Rahul Rameshbabu <[email protected]>

mlx5 devices have specific constants for choosing the CQ period mode. These constants do not have to match the constants used by the kernel software API for DIM period mode selection.

Fixes cdd04f4d4d71 ("net/mlx5: Add support to create SQ and CQ for ASO")
---
 drivers/net/ethernet/mellanox/mlx5/core/lib/aso.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/net/ethernet/mellanox/mlx5/core/lib/aso.c +1 -1
index 40c7be124041..58bd749b5e4d 100644
@@ -98,7 +98,7 @@ static int create_aso_cq(struct mlx5_aso_cq *cq, void *cqc_data)
 	mlx5_fill_page_frag_array(&cq->wq_ctrl.buf,
 				  (__be64 *)MLX5_ADDR_OF(create_cq_in, in, pas));
 
-	MLX5_SET(cqc,   cqc, cq_period_mode, DIM_CQ_PERIOD_MODE_START_FROM_EQE);
+	MLX5_SET(cqc,   cqc, cq_period_mode, MLX5_CQ_PERIOD_MODE_START_FROM_EQE);
 	MLX5_SET(cqc,   cqc, c_eqn_or_apu_element, eqn);
 	MLX5_SET(cqc,   cqc, uar_page,      mdev->priv.uar->index);
 	MLX5_SET(cqc,   cqc, log_page_size, cq->wq_ctrl.buf.page_shift -
Signature · 1 line
2.43.0
#12 lore raw

PATCH 11/14 net/mlx5e: Allow software parsing when IPsec crypto is enabled

In reply to Saeed Mahameed

From: Leon Romanovsky <[email protected]>

All ConnectX devices have software parsing capability enabled, but it is more correct to set allow_swp only if capability exists, which for IPsec means that crypto offload is supported.

Fixes 2451da081a34 ("net/mlx5: Unify device IPsec capabilities check")
---
 drivers/net/ethernet/mellanox/mlx5/core/en/params.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/net/ethernet/mellanox/mlx5/core/en/params.c +2 -2
index 284253b79266..5d213a9886f1 100644
@@ -1064,8 +1064,8 @@ void mlx5e_build_sq_param(struct mlx5_core_dev *mdev,
 	void *wq = MLX5_ADDR_OF(sqc, sqc, wq);
 	bool allow_swp;
 
-	allow_swp =
-		mlx5_geneve_tx_allowed(mdev) || !!mlx5_ipsec_device_caps(mdev);
+	allow_swp = mlx5_geneve_tx_allowed(mdev) ||
+		    (mlx5_ipsec_device_caps(mdev) & MLX5_IPSEC_CAP_CRYPTO);
 	mlx5e_build_sq_param_common(mdev, param);
 	MLX5_SET(wq, wq, log_wq_sz, params->log_sq_size);
 	MLX5_SET(sqc, sqc, allow_swp, allow_swp);
Signature · 1 line
2.43.0
#13 lore raw

PATCH 12/14 net/mlx5e: Ignore IPsec replay window values on sender side

In reply to Saeed Mahameed

From: Leon Romanovsky <[email protected]>

XFRM stack doesn't prevent from users to configure replay window in TX side and strongswan sets replay_window to be 1. It causes to failures in validation logic when trying to offload the SA.

Replay window is not relevant in TX side and should be ignored.

Fixes cded6d80129b ("net/mlx5e: Store replay window in XFRM attributes")
---
 .../net/ethernet/mellanox/mlx5/core/en_accel/ipsec.c   | 10 ++++++++--
 1 file changed, 8 insertions(+), 2 deletions(-)
drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.c +8 -2
index 161c5190c236..05612d9c6080 100644
@@ -336,12 +336,17 @@ void mlx5e_ipsec_build_accel_xfrm_attrs(struct mlx5e_ipsec_sa_entry *sa_entry,
 	/* iv len */
 	aes_gcm->icv_len = x->aead->alg_icv_len;
 
+	attrs->dir = x->xso.dir;
+
 	/* esn */
 	if (x->props.flags & XFRM_STATE_ESN) {
 		attrs->replay_esn.trigger = true;
 		attrs->replay_esn.esn = sa_entry->esn_state.esn;
 		attrs->replay_esn.esn_msb = sa_entry->esn_state.esn_msb;
 		attrs->replay_esn.overlap = sa_entry->esn_state.overlap;
+		if (attrs->dir == XFRM_DEV_OFFLOAD_OUT)
+			goto skip_replay_window;
+
 		switch (x->replay_esn->replay_window) {
 		case 32:
 			attrs->replay_esn.replay_window =
@@ -365,7 +370,7 @@ void mlx5e_ipsec_build_accel_xfrm_attrs(struct mlx5e_ipsec_sa_entry *sa_entry,
 		}
 	}
 
-	attrs->dir = x->xso.dir;
+skip_replay_window:
 	/* spi */
 	attrs->spi = be32_to_cpu(x->id.spi);
 
@@ -501,7 +506,8 @@ static int mlx5e_xfrm_validate_state(struct mlx5_core_dev *mdev,
 			return -EINVAL;
 		}
 
-		if (x->replay_esn && x->replay_esn->replay_window != 32 &&
+		if (x->replay_esn && x->xso.dir == XFRM_DEV_OFFLOAD_IN &&
+		    x->replay_esn->replay_window != 32 &&
 		    x->replay_esn->replay_window != 64 &&
 		    x->replay_esn->replay_window != 128 &&
 		    x->replay_esn->replay_window != 256) {
Signature · 1 line
2.43.0
#14 lore raw

PATCH 13/14 net/mlx5e: fix a double-free in arfs_create_groups

In reply to Saeed Mahameed

From: Zhipeng Lu <[email protected]>

When in allocated by kvzalloc fails, arfs_create_groups will free ft->g and return an error. However, arfs_create_table, the only caller of arfs_create_groups, will hold this error and call to mlx5e_destroy_flow_table, in which the ft->g will be freed again.

Fixes 1cabe6b0965e ("net/mlx5e: Create aRFS flow tables")
---
 .../net/ethernet/mellanox/mlx5/core/en_arfs.c | 26 +++++++++++--------
 1 file changed, 15 insertions(+), 11 deletions(-)
drivers/net/ethernet/mellanox/mlx5/core/en_arfs.c +15 -11
index bb7f86c993e5..e66f486faafe 100644
@@ -254,11 +254,13 @@ static int arfs_create_groups(struct mlx5e_flow_table *ft,
 
 	ft->g = kcalloc(MLX5E_ARFS_NUM_GROUPS,
 			sizeof(*ft->g), GFP_KERNEL);
-	in = kvzalloc(inlen, GFP_KERNEL);
-	if  (!in || !ft->g) {
-		kfree(ft->g);
-		kvfree(in);
+	if (!ft->g)
 		return -ENOMEM;
+
+	in = kvzalloc(inlen, GFP_KERNEL);
+	if (!in) {
+		err = -ENOMEM;
+		goto err_free_g;
 	}
 
 	mc = MLX5_ADDR_OF(create_flow_group_in, in, match_criteria);
@@ -278,7 +280,7 @@ static int arfs_create_groups(struct mlx5e_flow_table *ft,
 		break;
 	default:
 		err = -EINVAL;
-		goto out;
+		goto err_free_in;
 	}
 
 	switch (type) {
@@ -300,7 +302,7 @@ static int arfs_create_groups(struct mlx5e_flow_table *ft,
 		break;
 	default:
 		err = -EINVAL;
-		goto out;
+		goto err_free_in;
 	}
 
 	MLX5_SET_CFG(in, match_criteria_enable, MLX5_MATCH_OUTER_HEADERS);
@@ -309,7 +311,7 @@ static int arfs_create_groups(struct mlx5e_flow_table *ft,
 	MLX5_SET_CFG(in, end_flow_index, ix - 1);
 	ft->g[ft->num_groups] = mlx5_create_flow_group(ft->t, in);
 	if (IS_ERR(ft->g[ft->num_groups]))
-		goto err;
+		goto err_clean_group;
 	ft->num_groups++;
 
 	memset(in, 0, inlen);
@@ -318,18 +320,20 @@ static int arfs_create_groups(struct mlx5e_flow_table *ft,
 	MLX5_SET_CFG(in, end_flow_index, ix - 1);
 	ft->g[ft->num_groups] = mlx5_create_flow_group(ft->t, in);
 	if (IS_ERR(ft->g[ft->num_groups]))
-		goto err;
+		goto err_clean_group;
 	ft->num_groups++;
 
 	kvfree(in);
 	return 0;
 
-err:
+err_clean_group:
 	err = PTR_ERR(ft->g[ft->num_groups]);
 	ft->g[ft->num_groups] = NULL;
-out:
+err_free_in:
 	kvfree(in);
-
+err_free_g:
+	kfree(ft->g);
+	ft->g = NULL;
 	return err;
 }
 
Signature · 1 line
2.43.0
#15 lore raw

PATCH 14/14 net/mlx5e: fix a potential double-free in fs_any_create_groups

In reply to Saeed Mahameed

From: Dinghao Liu <[email protected]>

When kcalloc() for ft->g succeeds but kvzalloc() for in fails, fs_any_create_groups() will free ft->g. However, its caller fs_any_create_table() will free ft->g again through calling mlx5e_destroy_flow_table(), which will lead to a double-free. Fix this by setting ft->g to NULL in fs_any_create_groups().

Fixes 0f575c20bf06 ("net/mlx5e: Introduce Flow Steering ANY API")
---
 drivers/net/ethernet/mellanox/mlx5/core/en/fs_tt_redirect.c | 1 +
 1 file changed, 1 insertion(+)
drivers/net/ethernet/mellanox/mlx5/core/en/fs_tt_redirect.c +1 -0
index e1283531e0b8..671adbad0a40 100644
@@ -436,6 +436,7 @@ static int fs_any_create_groups(struct mlx5e_flow_table *ft)
 	in = kvzalloc(inlen, GFP_KERNEL);
 	if  (!in || !ft->g) {
 		kfree(ft->g);
+		ft->g = NULL;
 		kvfree(in);
 		return -ENOMEM;
 	}
Signature · 1 line
2.43.0