From: Raed Salem <raeds@nvidia.com>
Date: Thu, 27 Oct 2022 15:47:17 +0300
Subject: net/mlx5e: MACsec, fix RX data path 16 RX security channel limit
Patch-mainline: v6.1-rc8
Git-commit: 813115c4669d74cdf6fc253c5a1768d1202073c3
References: jsc#PED-1549

Currently the data path metadata flow id mask wrongly limits the
number of different RX security channels (SC) to 16, whereas in
adding RX SC the limit is "2^16 - 1" this cause an overlap in
metadata flow id once more than 16 RX SCs is added, this corrupts
MACsec RX offloaded flow handling.

Fix by using the correct mask, while at it improve code to use this
mask when adding the Rx rule and improve visibility of such errors
by adding debug massage.

Fixes: b7c9400cbc48 ("net/mlx5e: Implement MACsec Rx data path using MACsec skb_metadata_dst")
Signed-off-by: Raed Salem <raeds@nvidia.com>
Reviewed-by: Emeel Hakim <ehakim@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Acked-by: Thomas Bogendoerfer <tbogendoerfer@suse.de>
---
 drivers/net/ethernet/mellanox/mlx5/core/en_accel/macsec.c    |   11 ++++++++---
 drivers/net/ethernet/mellanox/mlx5/core/en_accel/macsec.h    |    6 ++++--
 drivers/net/ethernet/mellanox/mlx5/core/en_accel/macsec_fs.c |    4 ++--
 3 files changed, 14 insertions(+), 7 deletions(-)

--- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/macsec.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/macsec.c
@@ -736,9 +736,14 @@ static int mlx5e_macsec_add_rxsc(struct
 
 	sc_xarray_element->rx_sc = rx_sc;
 	err = xa_alloc(&macsec->sc_xarray, &sc_xarray_element->fs_id, sc_xarray_element,
-		       XA_LIMIT(1, USHRT_MAX), GFP_KERNEL);
-	if (err)
+		       XA_LIMIT(1, MLX5_MACEC_RX_FS_ID_MAX), GFP_KERNEL);
+	if (err) {
+		if (err == -EBUSY)
+			netdev_err(ctx->netdev,
+				   "MACsec offload: unable to create entry for RX SC (%d Rx SCs already allocated)\n",
+				   MLX5_MACEC_RX_FS_ID_MAX);
 		goto destroy_sc_xarray_elemenet;
+	}
 
 	rx_sc->md_dst = metadata_dst_alloc(0, METADATA_MACSEC, GFP_KERNEL);
 	if (!rx_sc->md_dst) {
@@ -1748,7 +1753,7 @@ void mlx5e_macsec_offload_handle_rx_skb(
 	if (!macsec)
 		return;
 
-	fs_id = MLX5_MACSEC_METADATA_HANDLE(macsec_meta_data);
+	fs_id = MLX5_MACSEC_RX_METADAT_HANDLE(macsec_meta_data);
 
 	rcu_read_lock();
 	sc_xarray_element = xa_load(&macsec->sc_xarray, fs_id);
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/macsec.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/macsec.h
@@ -10,9 +10,11 @@
 #include <net/macsec.h>
 #include <net/dst_metadata.h>
 
-/* Bit31 - 30: MACsec marker, Bit3-0: MACsec id */
+/* Bit31 - 30: MACsec marker, Bit15-0: MACsec id */
+#define MLX5_MACEC_RX_FS_ID_MAX USHRT_MAX /* Must be power of two */
+#define MLX5_MACSEC_RX_FS_ID_MASK MLX5_MACEC_RX_FS_ID_MAX
 #define MLX5_MACSEC_METADATA_MARKER(metadata)  ((((metadata) >> 30) & 0x3)  == 0x1)
-#define MLX5_MACSEC_METADATA_HANDLE(metadata)  ((metadata) & GENMASK(3, 0))
+#define MLX5_MACSEC_RX_METADAT_HANDLE(metadata)  ((metadata) & MLX5_MACSEC_RX_FS_ID_MASK)
 
 struct mlx5e_priv;
 struct mlx5e_macsec;
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/macsec_fs.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/macsec_fs.c
@@ -1146,10 +1146,10 @@ macsec_fs_rx_add_rule(struct mlx5e_macse
 	ft_crypto = &rx_tables->ft_crypto;
 
 	/* Set bit[31 - 30] macsec marker - 0x01 */
-	/* Set bit[3-0] fs id */
+	/* Set bit[15-0] fs id */
 	MLX5_SET(set_action_in, action, action_type, MLX5_ACTION_TYPE_SET);
 	MLX5_SET(set_action_in, action, field, MLX5_ACTION_IN_FIELD_METADATA_REG_B);
-	MLX5_SET(set_action_in, action, data, fs_id | BIT(30));
+	MLX5_SET(set_action_in, action, data, MLX5_MACSEC_RX_METADAT_HANDLE(fs_id) | BIT(30));
 	MLX5_SET(set_action_in, action, offset, 0);
 	MLX5_SET(set_action_in, action, length, 32);