- Linux-stable-mirror - lists.linaro.org

FAILED: patch "[PATCH] net/mlx5: Devcom, serialize devcom registration" failed to apply to 5.10-stable tree

by gregkh＠linuxfoundation.org

The patch below does not apply to the 5.10-stable tree. If someone wants it applied there, or to any other stable or longterm tree, then please email the backport, including the original git commit id to <stable(a)vger.kernel.org>. To reproduce the conflict and resubmit, you may use the following commands: git fetch https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/ linux-5.10.y git checkout FETCH_HEAD git cherry-pick -x 1f893f57a3bf9fe1f4bcb25b55aea7f7f9712fe7 # <resolve conflicts, build, test, etc.> git commit -s git send-email --to '<stable(a)vger.kernel.org>' --in-reply-to '2023052820-childcare-gumminess-835c@gregkh' --subject-prefix 'PATCH 5.10.y' HEAD^.. Possible dependencies: 1f893f57a3bf ("net/mlx5: Devcom, serialize devcom registration") af87194352ca ("net/mlx5: Devcom, fix error flow in mlx5_devcom_register_device") 8a6e75e5f57e ("net/mlx5: devcom only supports 2 ports") thanks, greg k-h ------------------ original commit in Linus's tree ------------------ From 1f893f57a3bf9fe1f4bcb25b55aea7f7f9712fe7 Mon Sep 17 00:00:00 2001 From: Shay Drory <shayd(a)nvidia.com> Date: Tue, 2 May 2023 13:36:42 +0300 Subject: [PATCH] net/mlx5: Devcom, serialize devcom registration From one hand, mlx5 driver is allowing to probe PFs in parallel. From the other hand, devcom, which is a share resource between PFs, is registered without any lock. This might resulted in memory problems. Hence, use the global mlx5_dev_list_lock in order to serialize devcom registration. Fixes: fadd59fc50d0 ("net/mlx5: Introduce inter-device communication mechanism") Signed-off-by: Shay Drory <shayd(a)nvidia.com> Reviewed-by: Mark Bloch <mbloch(a)nvidia.com> Signed-off-by: Saeed Mahameed <saeedm(a)nvidia.com> diff --git a/drivers/net/ethernet/mellanox/mlx5/core/lib/devcom.c b/drivers/net/ethernet/mellanox/mlx5/core/lib/devcom.c index 8f978491dd32..b7d779d08d83 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/lib/devcom.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/lib/devcom.c @@ -3,6 +3,7 @@ #include <linux/mlx5/vport.h> #include "lib/devcom.h" +#include "mlx5_core.h" static LIST_HEAD(devcom_list); @@ -77,6 +78,7 @@ struct mlx5_devcom *mlx5_devcom_register_device(struct mlx5_core_dev *dev) if (MLX5_CAP_GEN(dev, num_lag_ports) != MLX5_DEVCOM_PORTS_SUPPORTED) return NULL; + mlx5_dev_list_lock(); sguid0 = mlx5_query_nic_system_image_guid(dev); list_for_each_entry(iter, &devcom_list, list) { struct mlx5_core_dev *tmp_dev = NULL; @@ -102,8 +104,10 @@ struct mlx5_devcom *mlx5_devcom_register_device(struct mlx5_core_dev *dev) if (!priv) { priv = mlx5_devcom_list_alloc(); - if (!priv) - return ERR_PTR(-ENOMEM); + if (!priv) { + devcom = ERR_PTR(-ENOMEM); + goto out; + } idx = 0; new_priv = true; @@ -114,12 +118,14 @@ struct mlx5_devcom *mlx5_devcom_register_device(struct mlx5_core_dev *dev) if (!devcom) { if (new_priv) kfree(priv); - return ERR_PTR(-ENOMEM); + devcom = ERR_PTR(-ENOMEM); + goto out; } if (new_priv) list_add(&priv->list, &devcom_list); - +out: + mlx5_dev_list_unlock(); return devcom; } @@ -132,6 +138,7 @@ void mlx5_devcom_unregister_device(struct mlx5_devcom *devcom) if (IS_ERR_OR_NULL(devcom)) return; + mlx5_dev_list_lock(); priv = devcom->priv; priv->devs[devcom->idx] = NULL; @@ -142,10 +149,12 @@ void mlx5_devcom_unregister_device(struct mlx5_devcom *devcom) break; if (i != MLX5_DEVCOM_PORTS_SUPPORTED) - return; + goto out; list_del(&priv->list); kfree(priv); +out: + mlx5_dev_list_unlock(); } void mlx5_devcom_register_component(struct mlx5_devcom *devcom,

2 years, 7 months

1
0
0 0

FAILED: patch "[PATCH] net/mlx5: Devcom, serialize devcom registration" failed to apply to 5.15-stable tree

by gregkh＠linuxfoundation.org

The patch below does not apply to the 5.15-stable tree. If someone wants it applied there, or to any other stable or longterm tree, then please email the backport, including the original git commit id to <stable(a)vger.kernel.org>. To reproduce the conflict and resubmit, you may use the following commands: git fetch https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/ linux-5.15.y git checkout FETCH_HEAD git cherry-pick -x 1f893f57a3bf9fe1f4bcb25b55aea7f7f9712fe7 # <resolve conflicts, build, test, etc.> git commit -s git send-email --to '<stable(a)vger.kernel.org>' --in-reply-to '2023052819-emit-savior-ef9a@gregkh' --subject-prefix 'PATCH 5.15.y' HEAD^.. Possible dependencies: 1f893f57a3bf ("net/mlx5: Devcom, serialize devcom registration") af87194352ca ("net/mlx5: Devcom, fix error flow in mlx5_devcom_register_device") 8a6e75e5f57e ("net/mlx5: devcom only supports 2 ports") thanks, greg k-h ------------------ original commit in Linus's tree ------------------ From 1f893f57a3bf9fe1f4bcb25b55aea7f7f9712fe7 Mon Sep 17 00:00:00 2001 From: Shay Drory <shayd(a)nvidia.com> Date: Tue, 2 May 2023 13:36:42 +0300 Subject: [PATCH] net/mlx5: Devcom, serialize devcom registration From one hand, mlx5 driver is allowing to probe PFs in parallel. From the other hand, devcom, which is a share resource between PFs, is registered without any lock. This might resulted in memory problems. Hence, use the global mlx5_dev_list_lock in order to serialize devcom registration. Fixes: fadd59fc50d0 ("net/mlx5: Introduce inter-device communication mechanism") Signed-off-by: Shay Drory <shayd(a)nvidia.com> Reviewed-by: Mark Bloch <mbloch(a)nvidia.com> Signed-off-by: Saeed Mahameed <saeedm(a)nvidia.com> diff --git a/drivers/net/ethernet/mellanox/mlx5/core/lib/devcom.c b/drivers/net/ethernet/mellanox/mlx5/core/lib/devcom.c index 8f978491dd32..b7d779d08d83 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/lib/devcom.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/lib/devcom.c @@ -3,6 +3,7 @@ #include <linux/mlx5/vport.h> #include "lib/devcom.h" +#include "mlx5_core.h" static LIST_HEAD(devcom_list); @@ -77,6 +78,7 @@ struct mlx5_devcom *mlx5_devcom_register_device(struct mlx5_core_dev *dev) if (MLX5_CAP_GEN(dev, num_lag_ports) != MLX5_DEVCOM_PORTS_SUPPORTED) return NULL; + mlx5_dev_list_lock(); sguid0 = mlx5_query_nic_system_image_guid(dev); list_for_each_entry(iter, &devcom_list, list) { struct mlx5_core_dev *tmp_dev = NULL; @@ -102,8 +104,10 @@ struct mlx5_devcom *mlx5_devcom_register_device(struct mlx5_core_dev *dev) if (!priv) { priv = mlx5_devcom_list_alloc(); - if (!priv) - return ERR_PTR(-ENOMEM); + if (!priv) { + devcom = ERR_PTR(-ENOMEM); + goto out; + } idx = 0; new_priv = true; @@ -114,12 +118,14 @@ struct mlx5_devcom *mlx5_devcom_register_device(struct mlx5_core_dev *dev) if (!devcom) { if (new_priv) kfree(priv); - return ERR_PTR(-ENOMEM); + devcom = ERR_PTR(-ENOMEM); + goto out; } if (new_priv) list_add(&priv->list, &devcom_list); - +out: + mlx5_dev_list_unlock(); return devcom; } @@ -132,6 +138,7 @@ void mlx5_devcom_unregister_device(struct mlx5_devcom *devcom) if (IS_ERR_OR_NULL(devcom)) return; + mlx5_dev_list_lock(); priv = devcom->priv; priv->devs[devcom->idx] = NULL; @@ -142,10 +149,12 @@ void mlx5_devcom_unregister_device(struct mlx5_devcom *devcom) break; if (i != MLX5_DEVCOM_PORTS_SUPPORTED) - return; + goto out; list_del(&priv->list); kfree(priv); +out: + mlx5_dev_list_unlock(); } void mlx5_devcom_register_component(struct mlx5_devcom *devcom,

2 years, 7 months

1
0
0 0

FAILED: patch "[PATCH] net/mlx5: Handle pairing of E-switch via uplink un/load APIs" failed to apply to 5.4-stable tree

by gregkh＠linuxfoundation.org

The patch below does not apply to the 5.4-stable tree. If someone wants it applied there, or to any other stable or longterm tree, then please email the backport, including the original git commit id to <stable(a)vger.kernel.org>. To reproduce the conflict and resubmit, you may use the following commands: git fetch https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/ linux-5.4.y git checkout FETCH_HEAD git cherry-pick -x 2be5bd42a5bba1a05daedc86cf0e248210009669 # <resolve conflicts, build, test, etc.> git commit -s git send-email --to '<stable(a)vger.kernel.org>' --in-reply-to '2023052830-jeep-cod-05b2@gregkh' --subject-prefix 'PATCH 5.4.y' HEAD^.. Possible dependencies: 2be5bd42a5bb ("net/mlx5: Handle pairing of E-switch via uplink un/load APIs") c9668f0b1d28 ("net/mlx5e: Fix cleanup null-ptr deref on encap lock") d13674b1d14c ("net/mlx5e: TC, map tc action cookie to a hw counter") 0e414518d6d8 ("net/mlx5e: Add hairpin debugfs files") 1a8034720f38 ("net/mlx5e: Add hairpin params structure") 8facc02f22f1 ("net/mlx5e: TC, reuse flow attribute post parser processing") f2bb566f5c97 ("Merge git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net") thanks, greg k-h ------------------ original commit in Linus's tree ------------------ From 2be5bd42a5bba1a05daedc86cf0e248210009669 Mon Sep 17 00:00:00 2001 From: Shay Drory <shayd(a)nvidia.com> Date: Mon, 20 Mar 2023 13:07:53 +0200 Subject: [PATCH] net/mlx5: Handle pairing of E-switch via uplink un/load APIs In case user switch a device from switchdev mode to legacy mode, mlx5 first unpair the E-switch and afterwards unload the uplink vport. From the other hand, in case user remove or reload a device, mlx5 first unload the uplink vport and afterwards unpair the E-switch. The latter is causing a bug[1], hence, handle pairing of E-switch as part of uplink un/load APIs. [1] In case VF_LAG is used, every tc fdb flow is duplicated to the peer esw. However, the original esw keeps a pointer to this duplicated flow, not the peer esw. e.g.: if user create tc fdb flow over esw0, the flow is duplicated over esw1, in FW/HW, but in SW, esw0 keeps a pointer to the duplicated flow. During module unload while a peer tc fdb flow is still offloaded, in case the first device to be removed is the peer device (esw1 in the example above), the peer net-dev is destroyed, and so the mlx5e_priv is memset to 0. Afterwards, the peer device is trying to unpair himself from the original device (esw0 in the example above). Unpair API invoke the original device to clear peer flow from its eswitch (esw0), but the peer flow, which is stored over the original eswitch (esw0), is trying to use the peer mlx5e_priv, which is memset to 0 and result in bellow kernel-oops. [ 157.964081 ] BUG: unable to handle page fault for address: 000000000002ce60 [ 157.964662 ] #PF: supervisor read access in kernel mode [ 157.965123 ] #PF: error_code(0x0000) - not-present page [ 157.965582 ] PGD 0 P4D 0 [ 157.965866 ] Oops: 0000 [#1] SMP [ 157.967670 ] RIP: 0010:mlx5e_tc_del_fdb_flow+0x48/0x460 [mlx5_core] [ 157.976164 ] Call Trace: [ 157.976437 ] <TASK> [ 157.976690 ] __mlx5e_tc_del_fdb_peer_flow+0xe6/0x100 [mlx5_core] [ 157.977230 ] mlx5e_tc_clean_fdb_peer_flows+0x67/0x90 [mlx5_core] [ 157.977767 ] mlx5_esw_offloads_unpair+0x2d/0x1e0 [mlx5_core] [ 157.984653 ] mlx5_esw_offloads_devcom_event+0xbf/0x130 [mlx5_core] [ 157.985212 ] mlx5_devcom_send_event+0xa3/0xb0 [mlx5_core] [ 157.985714 ] esw_offloads_disable+0x5a/0x110 [mlx5_core] [ 157.986209 ] mlx5_eswitch_disable_locked+0x152/0x170 [mlx5_core] [ 157.986757 ] mlx5_eswitch_disable+0x51/0x80 [mlx5_core] [ 157.987248 ] mlx5_unload+0x2a/0xb0 [mlx5_core] [ 157.987678 ] mlx5_uninit_one+0x5f/0xd0 [mlx5_core] [ 157.988127 ] remove_one+0x64/0xe0 [mlx5_core] [ 157.988549 ] pci_device_remove+0x31/0xa0 [ 157.988933 ] device_release_driver_internal+0x18f/0x1f0 [ 157.989402 ] driver_detach+0x3f/0x80 [ 157.989754 ] bus_remove_driver+0x70/0xf0 [ 157.990129 ] pci_unregister_driver+0x34/0x90 [ 157.990537 ] mlx5_cleanup+0xc/0x1c [mlx5_core] [ 157.990972 ] __x64_sys_delete_module+0x15a/0x250 [ 157.991398 ] ? exit_to_user_mode_prepare+0xea/0x110 [ 157.991840 ] do_syscall_64+0x3d/0x90 [ 157.992198 ] entry_SYSCALL_64_after_hwframe+0x46/0xb0 Fixes: 04de7dda7394 ("net/mlx5e: Infrastructure for duplicated offloading of TC flows") Fixes: 1418ddd96afd ("net/mlx5e: Duplicate offloaded TC eswitch rules under uplink LAG") Signed-off-by: Shay Drory <shayd(a)nvidia.com> Reviewed-by: Roi Dayan <roid(a)nvidia.com> Signed-off-by: Saeed Mahameed <saeedm(a)nvidia.com> diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c index 728b82ce4031..65fe40f55d84 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c @@ -5301,6 +5301,8 @@ int mlx5e_tc_esw_init(struct mlx5_rep_uplink_priv *uplink_priv) goto err_action_counter; } + mlx5_esw_offloads_devcom_init(esw); + return 0; err_action_counter: @@ -5329,7 +5331,7 @@ void mlx5e_tc_esw_cleanup(struct mlx5_rep_uplink_priv *uplink_priv) priv = netdev_priv(rpriv->netdev); esw = priv->mdev->priv.eswitch; - mlx5e_tc_clean_fdb_peer_flows(esw); + mlx5_esw_offloads_devcom_cleanup(esw); mlx5e_tc_tun_cleanup(uplink_priv->encap); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h index 1a042c981713..9f007c5438ee 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h @@ -369,6 +369,8 @@ int mlx5_eswitch_enable(struct mlx5_eswitch *esw, int num_vfs); void mlx5_eswitch_disable_sriov(struct mlx5_eswitch *esw, bool clear_vf); void mlx5_eswitch_disable_locked(struct mlx5_eswitch *esw); void mlx5_eswitch_disable(struct mlx5_eswitch *esw); +void mlx5_esw_offloads_devcom_init(struct mlx5_eswitch *esw); +void mlx5_esw_offloads_devcom_cleanup(struct mlx5_eswitch *esw); int mlx5_eswitch_set_vport_mac(struct mlx5_eswitch *esw, u16 vport, const u8 *mac); int mlx5_eswitch_set_vport_state(struct mlx5_eswitch *esw, @@ -767,6 +769,8 @@ static inline void mlx5_eswitch_cleanup(struct mlx5_eswitch *esw) {} static inline int mlx5_eswitch_enable(struct mlx5_eswitch *esw, int num_vfs) { return 0; } static inline void mlx5_eswitch_disable_sriov(struct mlx5_eswitch *esw, bool clear_vf) {} static inline void mlx5_eswitch_disable(struct mlx5_eswitch *esw) {} +static inline void mlx5_esw_offloads_devcom_init(struct mlx5_eswitch *esw) {} +static inline void mlx5_esw_offloads_devcom_cleanup(struct mlx5_eswitch *esw) {} static inline bool mlx5_eswitch_is_funcs_handler(struct mlx5_core_dev *dev) { return false; } static inline int mlx5_eswitch_set_vport_state(struct mlx5_eswitch *esw, u16 vport, int link_state) { return 0; } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c index 69215ffb9999..7c34c7cf506f 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c @@ -2779,7 +2779,7 @@ static int mlx5_esw_offloads_devcom_event(int event, return err; } -static void esw_offloads_devcom_init(struct mlx5_eswitch *esw) +void mlx5_esw_offloads_devcom_init(struct mlx5_eswitch *esw) { struct mlx5_devcom *devcom = esw->dev->priv.devcom; @@ -2802,7 +2802,7 @@ static void esw_offloads_devcom_init(struct mlx5_eswitch *esw) ESW_OFFLOADS_DEVCOM_PAIR, esw); } -static void esw_offloads_devcom_cleanup(struct mlx5_eswitch *esw) +void mlx5_esw_offloads_devcom_cleanup(struct mlx5_eswitch *esw) { struct mlx5_devcom *devcom = esw->dev->priv.devcom; @@ -3250,8 +3250,6 @@ int esw_offloads_enable(struct mlx5_eswitch *esw) if (err) goto err_vports; - esw_offloads_devcom_init(esw); - return 0; err_vports: @@ -3292,7 +3290,6 @@ static int esw_offloads_stop(struct mlx5_eswitch *esw, void esw_offloads_disable(struct mlx5_eswitch *esw) { - esw_offloads_devcom_cleanup(esw); mlx5_eswitch_disable_pf_vf_vports(esw); esw_offloads_unload_rep(esw, MLX5_VPORT_UPLINK); esw_set_passing_vport_metadata(esw, false);

2 years, 7 months

1
0
0 0

FAILED: patch "[PATCH] net/mlx5: Handle pairing of E-switch via uplink un/load APIs" failed to apply to 5.10-stable tree

by gregkh＠linuxfoundation.org

The patch below does not apply to the 5.10-stable tree. If someone wants it applied there, or to any other stable or longterm tree, then please email the backport, including the original git commit id to <stable(a)vger.kernel.org>. To reproduce the conflict and resubmit, you may use the following commands: git fetch https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/ linux-5.10.y git checkout FETCH_HEAD git cherry-pick -x 2be5bd42a5bba1a05daedc86cf0e248210009669 # <resolve conflicts, build, test, etc.> git commit -s git send-email --to '<stable(a)vger.kernel.org>' --in-reply-to '2023052828-hatchet-drier-8e0e@gregkh' --subject-prefix 'PATCH 5.10.y' HEAD^.. Possible dependencies: 2be5bd42a5bb ("net/mlx5: Handle pairing of E-switch via uplink un/load APIs") c9668f0b1d28 ("net/mlx5e: Fix cleanup null-ptr deref on encap lock") d13674b1d14c ("net/mlx5e: TC, map tc action cookie to a hw counter") 0e414518d6d8 ("net/mlx5e: Add hairpin debugfs files") 1a8034720f38 ("net/mlx5e: Add hairpin params structure") 8facc02f22f1 ("net/mlx5e: TC, reuse flow attribute post parser processing") f2bb566f5c97 ("Merge git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net") thanks, greg k-h ------------------ original commit in Linus's tree ------------------ From 2be5bd42a5bba1a05daedc86cf0e248210009669 Mon Sep 17 00:00:00 2001 From: Shay Drory <shayd(a)nvidia.com> Date: Mon, 20 Mar 2023 13:07:53 +0200 Subject: [PATCH] net/mlx5: Handle pairing of E-switch via uplink un/load APIs In case user switch a device from switchdev mode to legacy mode, mlx5 first unpair the E-switch and afterwards unload the uplink vport. From the other hand, in case user remove or reload a device, mlx5 first unload the uplink vport and afterwards unpair the E-switch. The latter is causing a bug[1], hence, handle pairing of E-switch as part of uplink un/load APIs. [1] In case VF_LAG is used, every tc fdb flow is duplicated to the peer esw. However, the original esw keeps a pointer to this duplicated flow, not the peer esw. e.g.: if user create tc fdb flow over esw0, the flow is duplicated over esw1, in FW/HW, but in SW, esw0 keeps a pointer to the duplicated flow. During module unload while a peer tc fdb flow is still offloaded, in case the first device to be removed is the peer device (esw1 in the example above), the peer net-dev is destroyed, and so the mlx5e_priv is memset to 0. Afterwards, the peer device is trying to unpair himself from the original device (esw0 in the example above). Unpair API invoke the original device to clear peer flow from its eswitch (esw0), but the peer flow, which is stored over the original eswitch (esw0), is trying to use the peer mlx5e_priv, which is memset to 0 and result in bellow kernel-oops. [ 157.964081 ] BUG: unable to handle page fault for address: 000000000002ce60 [ 157.964662 ] #PF: supervisor read access in kernel mode [ 157.965123 ] #PF: error_code(0x0000) - not-present page [ 157.965582 ] PGD 0 P4D 0 [ 157.965866 ] Oops: 0000 [#1] SMP [ 157.967670 ] RIP: 0010:mlx5e_tc_del_fdb_flow+0x48/0x460 [mlx5_core] [ 157.976164 ] Call Trace: [ 157.976437 ] <TASK> [ 157.976690 ] __mlx5e_tc_del_fdb_peer_flow+0xe6/0x100 [mlx5_core] [ 157.977230 ] mlx5e_tc_clean_fdb_peer_flows+0x67/0x90 [mlx5_core] [ 157.977767 ] mlx5_esw_offloads_unpair+0x2d/0x1e0 [mlx5_core] [ 157.984653 ] mlx5_esw_offloads_devcom_event+0xbf/0x130 [mlx5_core] [ 157.985212 ] mlx5_devcom_send_event+0xa3/0xb0 [mlx5_core] [ 157.985714 ] esw_offloads_disable+0x5a/0x110 [mlx5_core] [ 157.986209 ] mlx5_eswitch_disable_locked+0x152/0x170 [mlx5_core] [ 157.986757 ] mlx5_eswitch_disable+0x51/0x80 [mlx5_core] [ 157.987248 ] mlx5_unload+0x2a/0xb0 [mlx5_core] [ 157.987678 ] mlx5_uninit_one+0x5f/0xd0 [mlx5_core] [ 157.988127 ] remove_one+0x64/0xe0 [mlx5_core] [ 157.988549 ] pci_device_remove+0x31/0xa0 [ 157.988933 ] device_release_driver_internal+0x18f/0x1f0 [ 157.989402 ] driver_detach+0x3f/0x80 [ 157.989754 ] bus_remove_driver+0x70/0xf0 [ 157.990129 ] pci_unregister_driver+0x34/0x90 [ 157.990537 ] mlx5_cleanup+0xc/0x1c [mlx5_core] [ 157.990972 ] __x64_sys_delete_module+0x15a/0x250 [ 157.991398 ] ? exit_to_user_mode_prepare+0xea/0x110 [ 157.991840 ] do_syscall_64+0x3d/0x90 [ 157.992198 ] entry_SYSCALL_64_after_hwframe+0x46/0xb0 Fixes: 04de7dda7394 ("net/mlx5e: Infrastructure for duplicated offloading of TC flows") Fixes: 1418ddd96afd ("net/mlx5e: Duplicate offloaded TC eswitch rules under uplink LAG") Signed-off-by: Shay Drory <shayd(a)nvidia.com> Reviewed-by: Roi Dayan <roid(a)nvidia.com> Signed-off-by: Saeed Mahameed <saeedm(a)nvidia.com> diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c index 728b82ce4031..65fe40f55d84 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c @@ -5301,6 +5301,8 @@ int mlx5e_tc_esw_init(struct mlx5_rep_uplink_priv *uplink_priv) goto err_action_counter; } + mlx5_esw_offloads_devcom_init(esw); + return 0; err_action_counter: @@ -5329,7 +5331,7 @@ void mlx5e_tc_esw_cleanup(struct mlx5_rep_uplink_priv *uplink_priv) priv = netdev_priv(rpriv->netdev); esw = priv->mdev->priv.eswitch; - mlx5e_tc_clean_fdb_peer_flows(esw); + mlx5_esw_offloads_devcom_cleanup(esw); mlx5e_tc_tun_cleanup(uplink_priv->encap); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h index 1a042c981713..9f007c5438ee 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h @@ -369,6 +369,8 @@ int mlx5_eswitch_enable(struct mlx5_eswitch *esw, int num_vfs); void mlx5_eswitch_disable_sriov(struct mlx5_eswitch *esw, bool clear_vf); void mlx5_eswitch_disable_locked(struct mlx5_eswitch *esw); void mlx5_eswitch_disable(struct mlx5_eswitch *esw); +void mlx5_esw_offloads_devcom_init(struct mlx5_eswitch *esw); +void mlx5_esw_offloads_devcom_cleanup(struct mlx5_eswitch *esw); int mlx5_eswitch_set_vport_mac(struct mlx5_eswitch *esw, u16 vport, const u8 *mac); int mlx5_eswitch_set_vport_state(struct mlx5_eswitch *esw, @@ -767,6 +769,8 @@ static inline void mlx5_eswitch_cleanup(struct mlx5_eswitch *esw) {} static inline int mlx5_eswitch_enable(struct mlx5_eswitch *esw, int num_vfs) { return 0; } static inline void mlx5_eswitch_disable_sriov(struct mlx5_eswitch *esw, bool clear_vf) {} static inline void mlx5_eswitch_disable(struct mlx5_eswitch *esw) {} +static inline void mlx5_esw_offloads_devcom_init(struct mlx5_eswitch *esw) {} +static inline void mlx5_esw_offloads_devcom_cleanup(struct mlx5_eswitch *esw) {} static inline bool mlx5_eswitch_is_funcs_handler(struct mlx5_core_dev *dev) { return false; } static inline int mlx5_eswitch_set_vport_state(struct mlx5_eswitch *esw, u16 vport, int link_state) { return 0; } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c index 69215ffb9999..7c34c7cf506f 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c @@ -2779,7 +2779,7 @@ static int mlx5_esw_offloads_devcom_event(int event, return err; } -static void esw_offloads_devcom_init(struct mlx5_eswitch *esw) +void mlx5_esw_offloads_devcom_init(struct mlx5_eswitch *esw) { struct mlx5_devcom *devcom = esw->dev->priv.devcom; @@ -2802,7 +2802,7 @@ static void esw_offloads_devcom_init(struct mlx5_eswitch *esw) ESW_OFFLOADS_DEVCOM_PAIR, esw); } -static void esw_offloads_devcom_cleanup(struct mlx5_eswitch *esw) +void mlx5_esw_offloads_devcom_cleanup(struct mlx5_eswitch *esw) { struct mlx5_devcom *devcom = esw->dev->priv.devcom; @@ -3250,8 +3250,6 @@ int esw_offloads_enable(struct mlx5_eswitch *esw) if (err) goto err_vports; - esw_offloads_devcom_init(esw); - return 0; err_vports: @@ -3292,7 +3290,6 @@ static int esw_offloads_stop(struct mlx5_eswitch *esw, void esw_offloads_disable(struct mlx5_eswitch *esw) { - esw_offloads_devcom_cleanup(esw); mlx5_eswitch_disable_pf_vf_vports(esw); esw_offloads_unload_rep(esw, MLX5_VPORT_UPLINK); esw_set_passing_vport_metadata(esw, false);

2 years, 7 months

1
0
0 0

FAILED: patch "[PATCH] net/mlx5: Handle pairing of E-switch via uplink un/load APIs" failed to apply to 5.15-stable tree

by gregkh＠linuxfoundation.org

The patch below does not apply to the 5.15-stable tree. If someone wants it applied there, or to any other stable or longterm tree, then please email the backport, including the original git commit id to <stable(a)vger.kernel.org>. To reproduce the conflict and resubmit, you may use the following commands: git fetch https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/ linux-5.15.y git checkout FETCH_HEAD git cherry-pick -x 2be5bd42a5bba1a05daedc86cf0e248210009669 # <resolve conflicts, build, test, etc.> git commit -s git send-email --to '<stable(a)vger.kernel.org>' --in-reply-to '2023052826-curve-data-272d@gregkh' --subject-prefix 'PATCH 5.15.y' HEAD^.. Possible dependencies: 2be5bd42a5bb ("net/mlx5: Handle pairing of E-switch via uplink un/load APIs") c9668f0b1d28 ("net/mlx5e: Fix cleanup null-ptr deref on encap lock") d13674b1d14c ("net/mlx5e: TC, map tc action cookie to a hw counter") 0e414518d6d8 ("net/mlx5e: Add hairpin debugfs files") 1a8034720f38 ("net/mlx5e: Add hairpin params structure") 8facc02f22f1 ("net/mlx5e: TC, reuse flow attribute post parser processing") f2bb566f5c97 ("Merge git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net") thanks, greg k-h ------------------ original commit in Linus's tree ------------------ From 2be5bd42a5bba1a05daedc86cf0e248210009669 Mon Sep 17 00:00:00 2001 From: Shay Drory <shayd(a)nvidia.com> Date: Mon, 20 Mar 2023 13:07:53 +0200 Subject: [PATCH] net/mlx5: Handle pairing of E-switch via uplink un/load APIs In case user switch a device from switchdev mode to legacy mode, mlx5 first unpair the E-switch and afterwards unload the uplink vport. From the other hand, in case user remove or reload a device, mlx5 first unload the uplink vport and afterwards unpair the E-switch. The latter is causing a bug[1], hence, handle pairing of E-switch as part of uplink un/load APIs. [1] In case VF_LAG is used, every tc fdb flow is duplicated to the peer esw. However, the original esw keeps a pointer to this duplicated flow, not the peer esw. e.g.: if user create tc fdb flow over esw0, the flow is duplicated over esw1, in FW/HW, but in SW, esw0 keeps a pointer to the duplicated flow. During module unload while a peer tc fdb flow is still offloaded, in case the first device to be removed is the peer device (esw1 in the example above), the peer net-dev is destroyed, and so the mlx5e_priv is memset to 0. Afterwards, the peer device is trying to unpair himself from the original device (esw0 in the example above). Unpair API invoke the original device to clear peer flow from its eswitch (esw0), but the peer flow, which is stored over the original eswitch (esw0), is trying to use the peer mlx5e_priv, which is memset to 0 and result in bellow kernel-oops. [ 157.964081 ] BUG: unable to handle page fault for address: 000000000002ce60 [ 157.964662 ] #PF: supervisor read access in kernel mode [ 157.965123 ] #PF: error_code(0x0000) - not-present page [ 157.965582 ] PGD 0 P4D 0 [ 157.965866 ] Oops: 0000 [#1] SMP [ 157.967670 ] RIP: 0010:mlx5e_tc_del_fdb_flow+0x48/0x460 [mlx5_core] [ 157.976164 ] Call Trace: [ 157.976437 ] <TASK> [ 157.976690 ] __mlx5e_tc_del_fdb_peer_flow+0xe6/0x100 [mlx5_core] [ 157.977230 ] mlx5e_tc_clean_fdb_peer_flows+0x67/0x90 [mlx5_core] [ 157.977767 ] mlx5_esw_offloads_unpair+0x2d/0x1e0 [mlx5_core] [ 157.984653 ] mlx5_esw_offloads_devcom_event+0xbf/0x130 [mlx5_core] [ 157.985212 ] mlx5_devcom_send_event+0xa3/0xb0 [mlx5_core] [ 157.985714 ] esw_offloads_disable+0x5a/0x110 [mlx5_core] [ 157.986209 ] mlx5_eswitch_disable_locked+0x152/0x170 [mlx5_core] [ 157.986757 ] mlx5_eswitch_disable+0x51/0x80 [mlx5_core] [ 157.987248 ] mlx5_unload+0x2a/0xb0 [mlx5_core] [ 157.987678 ] mlx5_uninit_one+0x5f/0xd0 [mlx5_core] [ 157.988127 ] remove_one+0x64/0xe0 [mlx5_core] [ 157.988549 ] pci_device_remove+0x31/0xa0 [ 157.988933 ] device_release_driver_internal+0x18f/0x1f0 [ 157.989402 ] driver_detach+0x3f/0x80 [ 157.989754 ] bus_remove_driver+0x70/0xf0 [ 157.990129 ] pci_unregister_driver+0x34/0x90 [ 157.990537 ] mlx5_cleanup+0xc/0x1c [mlx5_core] [ 157.990972 ] __x64_sys_delete_module+0x15a/0x250 [ 157.991398 ] ? exit_to_user_mode_prepare+0xea/0x110 [ 157.991840 ] do_syscall_64+0x3d/0x90 [ 157.992198 ] entry_SYSCALL_64_after_hwframe+0x46/0xb0 Fixes: 04de7dda7394 ("net/mlx5e: Infrastructure for duplicated offloading of TC flows") Fixes: 1418ddd96afd ("net/mlx5e: Duplicate offloaded TC eswitch rules under uplink LAG") Signed-off-by: Shay Drory <shayd(a)nvidia.com> Reviewed-by: Roi Dayan <roid(a)nvidia.com> Signed-off-by: Saeed Mahameed <saeedm(a)nvidia.com> diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c index 728b82ce4031..65fe40f55d84 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c @@ -5301,6 +5301,8 @@ int mlx5e_tc_esw_init(struct mlx5_rep_uplink_priv *uplink_priv) goto err_action_counter; } + mlx5_esw_offloads_devcom_init(esw); + return 0; err_action_counter: @@ -5329,7 +5331,7 @@ void mlx5e_tc_esw_cleanup(struct mlx5_rep_uplink_priv *uplink_priv) priv = netdev_priv(rpriv->netdev); esw = priv->mdev->priv.eswitch; - mlx5e_tc_clean_fdb_peer_flows(esw); + mlx5_esw_offloads_devcom_cleanup(esw); mlx5e_tc_tun_cleanup(uplink_priv->encap); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h index 1a042c981713..9f007c5438ee 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h @@ -369,6 +369,8 @@ int mlx5_eswitch_enable(struct mlx5_eswitch *esw, int num_vfs); void mlx5_eswitch_disable_sriov(struct mlx5_eswitch *esw, bool clear_vf); void mlx5_eswitch_disable_locked(struct mlx5_eswitch *esw); void mlx5_eswitch_disable(struct mlx5_eswitch *esw); +void mlx5_esw_offloads_devcom_init(struct mlx5_eswitch *esw); +void mlx5_esw_offloads_devcom_cleanup(struct mlx5_eswitch *esw); int mlx5_eswitch_set_vport_mac(struct mlx5_eswitch *esw, u16 vport, const u8 *mac); int mlx5_eswitch_set_vport_state(struct mlx5_eswitch *esw, @@ -767,6 +769,8 @@ static inline void mlx5_eswitch_cleanup(struct mlx5_eswitch *esw) {} static inline int mlx5_eswitch_enable(struct mlx5_eswitch *esw, int num_vfs) { return 0; } static inline void mlx5_eswitch_disable_sriov(struct mlx5_eswitch *esw, bool clear_vf) {} static inline void mlx5_eswitch_disable(struct mlx5_eswitch *esw) {} +static inline void mlx5_esw_offloads_devcom_init(struct mlx5_eswitch *esw) {} +static inline void mlx5_esw_offloads_devcom_cleanup(struct mlx5_eswitch *esw) {} static inline bool mlx5_eswitch_is_funcs_handler(struct mlx5_core_dev *dev) { return false; } static inline int mlx5_eswitch_set_vport_state(struct mlx5_eswitch *esw, u16 vport, int link_state) { return 0; } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c index 69215ffb9999..7c34c7cf506f 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c @@ -2779,7 +2779,7 @@ static int mlx5_esw_offloads_devcom_event(int event, return err; } -static void esw_offloads_devcom_init(struct mlx5_eswitch *esw) +void mlx5_esw_offloads_devcom_init(struct mlx5_eswitch *esw) { struct mlx5_devcom *devcom = esw->dev->priv.devcom; @@ -2802,7 +2802,7 @@ static void esw_offloads_devcom_init(struct mlx5_eswitch *esw) ESW_OFFLOADS_DEVCOM_PAIR, esw); } -static void esw_offloads_devcom_cleanup(struct mlx5_eswitch *esw) +void mlx5_esw_offloads_devcom_cleanup(struct mlx5_eswitch *esw) { struct mlx5_devcom *devcom = esw->dev->priv.devcom; @@ -3250,8 +3250,6 @@ int esw_offloads_enable(struct mlx5_eswitch *esw) if (err) goto err_vports; - esw_offloads_devcom_init(esw); - return 0; err_vports: @@ -3292,7 +3290,6 @@ static int esw_offloads_stop(struct mlx5_eswitch *esw, void esw_offloads_disable(struct mlx5_eswitch *esw) { - esw_offloads_devcom_cleanup(esw); mlx5_eswitch_disable_pf_vf_vports(esw); esw_offloads_unload_rep(esw, MLX5_VPORT_UPLINK); esw_set_passing_vport_metadata(esw, false);

2 years, 7 months

1
0
0 0

FAILED: patch "[PATCH] net/mlx5e: do as little as possible in napi poll when budget" failed to apply to 4.19-stable tree

by gregkh＠linuxfoundation.org

The patch below does not apply to the 4.19-stable tree. If someone wants it applied there, or to any other stable or longterm tree, then please email the backport, including the original git commit id to <stable(a)vger.kernel.org>. To reproduce the conflict and resubmit, you may use the following commands: git fetch https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/ linux-4.19.y git checkout FETCH_HEAD git cherry-pick -x afbed3f74830163f9559579dee382cac3cff82da # <resolve conflicts, build, test, etc.> git commit -s git send-email --to '<stable(a)vger.kernel.org>' --in-reply-to '2023052845-probably-overpass-092e@gregkh' --subject-prefix 'PATCH 4.19.y' HEAD^.. Possible dependencies: afbed3f74830 ("net/mlx5e: do as little as possible in napi poll when budget is 0") 214baf22870c ("net/mlx5e: Support HTB offload") 1880bc4e4a96 ("net/mlx5e: Add TX port timestamp support") 145e5637d941 ("net/mlx5e: Add TX PTP port object support") 1a7f51240dfb ("net/mlx5e: Split SW group counters update function") 0b676aaecc25 ("net/mlx5e: Change skb fifo push/pop API to be used without SQ") 579524c6eace ("net/mlx5e: Validate stop_room size upon user input") 3180472f582b ("net/mlx5: Add functions to set/query MFRL register") 573a8095f68c ("Merge tag 'mlx5-updates-2020-09-21' of git://git.kernel.org/pub/scm/linux/kernel/git/saeed/linux") thanks, greg k-h ------------------ original commit in Linus's tree ------------------ From afbed3f74830163f9559579dee382cac3cff82da Mon Sep 17 00:00:00 2001 From: Jakub Kicinski <kuba(a)kernel.org> Date: Tue, 16 May 2023 18:59:35 -0700 Subject: [PATCH] net/mlx5e: do as little as possible in napi poll when budget is 0 NAPI gets called with budget of 0 from netpoll, which has interrupts disabled. We should try to free some space on Tx rings and nothing else. Specifically do not try to handle XDP TX or try to refill Rx buffers - we can't use the page pool from IRQ context. Don't check if IRQs moved, either, that makes no sense in netpoll. Netpoll calls _all_ the rings from whatever CPU it happens to be invoked on. In general do as little as possible, the work quickly adds up when there's tens of rings to poll. The immediate stack trace I was seeing is: __do_softirq+0xd1/0x2c0 __local_bh_enable_ip+0xc7/0x120 </IRQ> <TASK> page_pool_put_defragged_page+0x267/0x320 mlx5e_free_xdpsq_desc+0x99/0xd0 mlx5e_poll_xdpsq_cq+0x138/0x3b0 mlx5e_napi_poll+0xc3/0x8b0 netpoll_poll_dev+0xce/0x150 AFAIU page pool takes a BH lock, releases it and since BH is now enabled tries to run softirqs. Reviewed-by: Tariq Toukan <tariqt(a)nvidia.com> Fixes: 60bbf7eeef10 ("mlx5: use page_pool for xdp_return_frame call") Signed-off-by: Jakub Kicinski <kuba(a)kernel.org> Reviewed-by: Simon Horman <simon.horman(a)corigine.com> Signed-off-by: David S. Miller <davem(a)davemloft.net> diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_txrx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_txrx.c index a50bfda18e96..fbb2d963fb7e 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_txrx.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_txrx.c @@ -161,20 +161,22 @@ int mlx5e_napi_poll(struct napi_struct *napi, int budget) } } + /* budget=0 means we may be in IRQ context, do as little as possible */ + if (unlikely(!budget)) + goto out; + busy |= mlx5e_poll_xdpsq_cq(&c->xdpsq.cq); if (c->xdp) busy |= mlx5e_poll_xdpsq_cq(&c->rq_xdpsq.cq); - if (likely(budget)) { /* budget=0 means: don't poll rx rings */ - if (xsk_open) - work_done = mlx5e_poll_rx_cq(&xskrq->cq, budget); + if (xsk_open) + work_done = mlx5e_poll_rx_cq(&xskrq->cq, budget); - if (likely(budget - work_done)) - work_done += mlx5e_poll_rx_cq(&rq->cq, budget - work_done); + if (likely(budget - work_done)) + work_done += mlx5e_poll_rx_cq(&rq->cq, budget - work_done); - busy |= work_done == budget; - } + busy |= work_done == budget; mlx5e_poll_ico_cq(&c->icosq.cq); if (mlx5e_poll_ico_cq(&c->async_icosq.cq))

2 years, 7 months

1
0
0 0

FAILED: patch "[PATCH] net/mlx5e: Fix deadlock in tc route query code" failed to apply to 5.15-stable tree

by gregkh＠linuxfoundation.org

The patch below does not apply to the 5.15-stable tree. If someone wants it applied there, or to any other stable or longterm tree, then please email the backport, including the original git commit id to <stable(a)vger.kernel.org>. To reproduce the conflict and resubmit, you may use the following commands: git fetch https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/ linux-5.15.y git checkout FETCH_HEAD git cherry-pick -x 691c041bf20899fc13c793f92ba61ab660fa3a30 # <resolve conflicts, build, test, etc.> git commit -s git send-email --to '<stable(a)vger.kernel.org>' --in-reply-to '2023052817-obligate-dramatic-13ed@gregkh' --subject-prefix 'PATCH 5.15.y' HEAD^.. Possible dependencies: 691c041bf208 ("net/mlx5e: Fix deadlock in tc route query code") 8a6e75e5f57e ("net/mlx5: devcom only supports 2 ports") thanks, greg k-h ------------------ original commit in Linus's tree ------------------ From 691c041bf20899fc13c793f92ba61ab660fa3a30 Mon Sep 17 00:00:00 2001 From: Vlad Buslov <vladbu(a)nvidia.com> Date: Fri, 31 Mar 2023 14:20:51 +0200 Subject: [PATCH] net/mlx5e: Fix deadlock in tc route query code Cited commit causes ABBA deadlock[0] when peer flows are created while holding the devcom rw semaphore. Due to peer flows offload implementation the lock is taken much higher up the call chain and there is no obvious way to easily fix the deadlock. Instead, since tc route query code needs the peer eswitch structure only to perform a lookup in xarray and doesn't perform any sleeping operations with it, refactor the code for lockless execution in following ways: - RCUify the devcom 'data' pointer. When resetting the pointer synchronously wait for RCU grace period before returning. This is fine since devcom is currently only used for synchronization of pairing/unpairing of eswitches which is rare and already expensive as-is. - Wrap all usages of 'paired' boolean in {READ|WRITE}_ONCE(). The flag has already been used in some unlocked contexts without proper annotations (e.g. users of mlx5_devcom_is_paired() function), but it wasn't an issue since all relevant code paths checked it again after obtaining the devcom semaphore. Now it is also used by mlx5_devcom_get_peer_data_rcu() as "best effort" check to return NULL when devcom is being unpaired. Note that while RCU read lock doesn't prevent the unpaired flag from being changed concurrently it still guarantees that reader can continue to use 'data'. - Refactor mlx5e_tc_query_route_vport() function to use new mlx5_devcom_get_peer_data_rcu() API which fixes the deadlock. [0]: [ 164.599612] ====================================================== [ 164.600142] WARNING: possible circular locking dependency detected [ 164.600667] 6.3.0-rc3+ #1 Not tainted [ 164.601021] ------------------------------------------------------ [ 164.601557] handler1/3456 is trying to acquire lock: [ 164.601998] ffff88811f1714b0 (&esw->offloads.encap_tbl_lock){+.+.}-{3:3}, at: mlx5e_attach_encap+0xd8/0x8b0 [mlx5_core] [ 164.603078] but task is already holding lock: [ 164.603617] ffff88810137fc98 (&comp->sem){++++}-{3:3}, at: mlx5_devcom_get_peer_data+0x37/0x80 [mlx5_core] [ 164.604459] which lock already depends on the new lock. [ 164.605190] the existing dependency chain (in reverse order) is: [ 164.605848] -> #1 (&comp->sem){++++}-{3:3}: [ 164.606380] down_read+0x39/0x50 [ 164.606772] mlx5_devcom_get_peer_data+0x37/0x80 [mlx5_core] [ 164.607336] mlx5e_tc_query_route_vport+0x86/0xc0 [mlx5_core] [ 164.607914] mlx5e_tc_tun_route_lookup+0x1a4/0x1d0 [mlx5_core] [ 164.608495] mlx5e_attach_decap_route+0xc6/0x1e0 [mlx5_core] [ 164.609063] mlx5e_tc_add_fdb_flow+0x1ea/0x360 [mlx5_core] [ 164.609627] __mlx5e_add_fdb_flow+0x2d2/0x430 [mlx5_core] [ 164.610175] mlx5e_configure_flower+0x952/0x1a20 [mlx5_core] [ 164.610741] tc_setup_cb_add+0xd4/0x200 [ 164.611146] fl_hw_replace_filter+0x14c/0x1f0 [cls_flower] [ 164.611661] fl_change+0xc95/0x18a0 [cls_flower] [ 164.612116] tc_new_tfilter+0x3fc/0xd20 [ 164.612516] rtnetlink_rcv_msg+0x418/0x5b0 [ 164.612936] netlink_rcv_skb+0x54/0x100 [ 164.613339] netlink_unicast+0x190/0x250 [ 164.613746] netlink_sendmsg+0x245/0x4a0 [ 164.614150] sock_sendmsg+0x38/0x60 [ 164.614522] ____sys_sendmsg+0x1d0/0x1e0 [ 164.614934] ___sys_sendmsg+0x80/0xc0 [ 164.615320] __sys_sendmsg+0x51/0x90 [ 164.615701] do_syscall_64+0x3d/0x90 [ 164.616083] entry_SYSCALL_64_after_hwframe+0x46/0xb0 [ 164.616568] -> #0 (&esw->offloads.encap_tbl_lock){+.+.}-{3:3}: [ 164.617210] __lock_acquire+0x159e/0x26e0 [ 164.617638] lock_acquire+0xc2/0x2a0 [ 164.618018] __mutex_lock+0x92/0xcd0 [ 164.618401] mlx5e_attach_encap+0xd8/0x8b0 [mlx5_core] [ 164.618943] post_process_attr+0x153/0x2d0 [mlx5_core] [ 164.619471] mlx5e_tc_add_fdb_flow+0x164/0x360 [mlx5_core] [ 164.620021] __mlx5e_add_fdb_flow+0x2d2/0x430 [mlx5_core] [ 164.620564] mlx5e_configure_flower+0xe33/0x1a20 [mlx5_core] [ 164.621125] tc_setup_cb_add+0xd4/0x200 [ 164.621531] fl_hw_replace_filter+0x14c/0x1f0 [cls_flower] [ 164.622047] fl_change+0xc95/0x18a0 [cls_flower] [ 164.622500] tc_new_tfilter+0x3fc/0xd20 [ 164.622906] rtnetlink_rcv_msg+0x418/0x5b0 [ 164.623324] netlink_rcv_skb+0x54/0x100 [ 164.623727] netlink_unicast+0x190/0x250 [ 164.624138] netlink_sendmsg+0x245/0x4a0 [ 164.624544] sock_sendmsg+0x38/0x60 [ 164.624919] ____sys_sendmsg+0x1d0/0x1e0 [ 164.625340] ___sys_sendmsg+0x80/0xc0 [ 164.625731] __sys_sendmsg+0x51/0x90 [ 164.626117] do_syscall_64+0x3d/0x90 [ 164.626502] entry_SYSCALL_64_after_hwframe+0x46/0xb0 [ 164.626995] other info that might help us debug this: [ 164.627725] Possible unsafe locking scenario: [ 164.628268] CPU0 CPU1 [ 164.628683] ---- ---- [ 164.629098] lock(&comp->sem); [ 164.629421] lock(&esw->offloads.encap_tbl_lock); [ 164.630066] lock(&comp->sem); [ 164.630555] lock(&esw->offloads.encap_tbl_lock); [ 164.630993] *** DEADLOCK *** [ 164.631575] 3 locks held by handler1/3456: [ 164.631962] #0: ffff888124b75130 (&block->cb_lock){++++}-{3:3}, at: tc_setup_cb_add+0x5b/0x200 [ 164.632703] #1: ffff888116e512b8 (&esw->mode_lock){++++}-{3:3}, at: mlx5_esw_hold+0x39/0x50 [mlx5_core] [ 164.633552] #2: ffff88810137fc98 (&comp->sem){++++}-{3:3}, at: mlx5_devcom_get_peer_data+0x37/0x80 [mlx5_core] [ 164.634435] stack backtrace: [ 164.634883] CPU: 17 PID: 3456 Comm: handler1 Not tainted 6.3.0-rc3+ #1 [ 164.635431] Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS rel-1.13.0-0-gf21b5a4aeb02-prebuilt.qemu.org 04/01/2014 [ 164.636340] Call Trace: [ 164.636616] <TASK> [ 164.636863] dump_stack_lvl+0x47/0x70 [ 164.637217] check_noncircular+0xfe/0x110 [ 164.637601] __lock_acquire+0x159e/0x26e0 [ 164.637977] ? mlx5_cmd_set_fte+0x5b0/0x830 [mlx5_core] [ 164.638472] lock_acquire+0xc2/0x2a0 [ 164.638828] ? mlx5e_attach_encap+0xd8/0x8b0 [mlx5_core] [ 164.639339] ? lock_is_held_type+0x98/0x110 [ 164.639728] __mutex_lock+0x92/0xcd0 [ 164.640074] ? mlx5e_attach_encap+0xd8/0x8b0 [mlx5_core] [ 164.640576] ? __lock_acquire+0x382/0x26e0 [ 164.640958] ? mlx5e_attach_encap+0xd8/0x8b0 [mlx5_core] [ 164.641468] ? mlx5e_attach_encap+0xd8/0x8b0 [mlx5_core] [ 164.641965] mlx5e_attach_encap+0xd8/0x8b0 [mlx5_core] [ 164.642454] ? lock_release+0xbf/0x240 [ 164.642819] post_process_attr+0x153/0x2d0 [mlx5_core] [ 164.643318] mlx5e_tc_add_fdb_flow+0x164/0x360 [mlx5_core] [ 164.643835] __mlx5e_add_fdb_flow+0x2d2/0x430 [mlx5_core] [ 164.644340] mlx5e_configure_flower+0xe33/0x1a20 [mlx5_core] [ 164.644862] ? lock_acquire+0xc2/0x2a0 [ 164.645219] tc_setup_cb_add+0xd4/0x200 [ 164.645588] fl_hw_replace_filter+0x14c/0x1f0 [cls_flower] [ 164.646067] fl_change+0xc95/0x18a0 [cls_flower] [ 164.646488] tc_new_tfilter+0x3fc/0xd20 [ 164.646861] ? tc_del_tfilter+0x810/0x810 [ 164.647236] rtnetlink_rcv_msg+0x418/0x5b0 [ 164.647621] ? rtnl_setlink+0x160/0x160 [ 164.647982] netlink_rcv_skb+0x54/0x100 [ 164.648348] netlink_unicast+0x190/0x250 [ 164.648722] netlink_sendmsg+0x245/0x4a0 [ 164.649090] sock_sendmsg+0x38/0x60 [ 164.649434] ____sys_sendmsg+0x1d0/0x1e0 [ 164.649804] ? copy_msghdr_from_user+0x6d/0xa0 [ 164.650213] ___sys_sendmsg+0x80/0xc0 [ 164.650563] ? lock_acquire+0xc2/0x2a0 [ 164.650926] ? lock_acquire+0xc2/0x2a0 [ 164.651286] ? __fget_files+0x5/0x190 [ 164.651644] ? find_held_lock+0x2b/0x80 [ 164.652006] ? __fget_files+0xb9/0x190 [ 164.652365] ? lock_release+0xbf/0x240 [ 164.652723] ? __fget_files+0xd3/0x190 [ 164.653079] __sys_sendmsg+0x51/0x90 [ 164.653435] do_syscall_64+0x3d/0x90 [ 164.653784] entry_SYSCALL_64_after_hwframe+0x46/0xb0 [ 164.654229] RIP: 0033:0x7f378054f8bd [ 164.654577] Code: 28 89 54 24 1c 48 89 74 24 10 89 7c 24 08 e8 6a c3 f4 ff 8b 54 24 1c 48 8b 74 24 10 41 89 c0 8b 7c 24 08 b8 2e 00 00 00 0f 05 <48> 3d 00 f0 ff ff 77 33 44 89 c7 48 89 44 24 08 e8 be c3 f4 ff 48 [ 164.656041] RSP: 002b:00007f377fa114b0 EFLAGS: 00000293 ORIG_RAX: 000000000000002e [ 164.656701] RAX: ffffffffffffffda RBX: 0000000000000001 RCX: 00007f378054f8bd [ 164.657297] RDX: 0000000000000000 RSI: 00007f377fa11540 RDI: 0000000000000014 [ 164.657885] RBP: 00007f377fa12278 R08: 0000000000000000 R09: 000000000000015c [ 164.658472] R10: 00007f377fa123d0 R11: 0000000000000293 R12: 0000560962d99bd0 [ 164.665317] R13: 0000000000000000 R14: 0000560962d99bd0 R15: 00007f377fa11540 Fixes: f9d196bd632b ("net/mlx5e: Use correct eswitch for stack devices with lag") Signed-off-by: Vlad Buslov <vladbu(a)nvidia.com> Reviewed-by: Roi Dayan <roid(a)nvidia.com> Reviewed-by: Shay Drory <shayd(a)nvidia.com> Reviewed-by: Tariq Toukan <tariqt(a)nvidia.com> Signed-off-by: Saeed Mahameed <saeedm(a)nvidia.com> diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c index 65fe40f55d84..416ab6b6da97 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c @@ -1665,11 +1665,9 @@ bool mlx5e_tc_is_vf_tunnel(struct net_device *out_dev, struct net_device *route_ int mlx5e_tc_query_route_vport(struct net_device *out_dev, struct net_device *route_dev, u16 *vport) { struct mlx5e_priv *out_priv, *route_priv; - struct mlx5_devcom *devcom = NULL; struct mlx5_core_dev *route_mdev; struct mlx5_eswitch *esw; u16 vhca_id; - int err; out_priv = netdev_priv(out_dev); esw = out_priv->mdev->priv.eswitch; @@ -1678,6 +1676,9 @@ int mlx5e_tc_query_route_vport(struct net_device *out_dev, struct net_device *ro vhca_id = MLX5_CAP_GEN(route_mdev, vhca_id); if (mlx5_lag_is_active(out_priv->mdev)) { + struct mlx5_devcom *devcom; + int err; + /* In lag case we may get devices from different eswitch instances. * If we failed to get vport num, it means, mostly, that we on the wrong * eswitch. @@ -1686,16 +1687,16 @@ int mlx5e_tc_query_route_vport(struct net_device *out_dev, struct net_device *ro if (err != -ENOENT) return err; + rcu_read_lock(); devcom = out_priv->mdev->priv.devcom; - esw = mlx5_devcom_get_peer_data(devcom, MLX5_DEVCOM_ESW_OFFLOADS); - if (!esw) - return -ENODEV; + esw = mlx5_devcom_get_peer_data_rcu(devcom, MLX5_DEVCOM_ESW_OFFLOADS); + err = esw ? mlx5_eswitch_vhca_id_to_vport(esw, vhca_id, vport) : -ENODEV; + rcu_read_unlock(); + + return err; } - err = mlx5_eswitch_vhca_id_to_vport(esw, vhca_id, vport); - if (devcom) - mlx5_devcom_release_peer_data(devcom, MLX5_DEVCOM_ESW_OFFLOADS); - return err; + return mlx5_eswitch_vhca_id_to_vport(esw, vhca_id, vport); } static int diff --git a/drivers/net/ethernet/mellanox/mlx5/core/lib/devcom.c b/drivers/net/ethernet/mellanox/mlx5/core/lib/devcom.c index adefde3ea941..070d55f13419 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/lib/devcom.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/lib/devcom.c @@ -13,7 +13,7 @@ static LIST_HEAD(devcom_list); struct mlx5_devcom_component { struct { - void *data; + void __rcu *data; } device[MLX5_DEVCOM_PORTS_SUPPORTED]; mlx5_devcom_event_handler_t handler; @@ -162,7 +162,7 @@ void mlx5_devcom_register_component(struct mlx5_devcom *devcom, comp = &devcom->priv->components[id]; down_write(&comp->sem); comp->handler = handler; - comp->device[devcom->idx].data = data; + rcu_assign_pointer(comp->device[devcom->idx].data, data); up_write(&comp->sem); } @@ -176,8 +176,9 @@ void mlx5_devcom_unregister_component(struct mlx5_devcom *devcom, comp = &devcom->priv->components[id]; down_write(&comp->sem); - comp->device[devcom->idx].data = NULL; + RCU_INIT_POINTER(comp->device[devcom->idx].data, NULL); up_write(&comp->sem); + synchronize_rcu(); } int mlx5_devcom_send_event(struct mlx5_devcom *devcom, @@ -193,12 +194,15 @@ int mlx5_devcom_send_event(struct mlx5_devcom *devcom, comp = &devcom->priv->components[id]; down_write(&comp->sem); - for (i = 0; i < MLX5_DEVCOM_PORTS_SUPPORTED; i++) - if (i != devcom->idx && comp->device[i].data) { - err = comp->handler(event, comp->device[i].data, - event_data); + for (i = 0; i < MLX5_DEVCOM_PORTS_SUPPORTED; i++) { + void *data = rcu_dereference_protected(comp->device[i].data, + lockdep_is_held(&comp->sem)); + + if (i != devcom->idx && data) { + err = comp->handler(event, data, event_data); break; } + } up_write(&comp->sem); return err; @@ -213,7 +217,7 @@ void mlx5_devcom_set_paired(struct mlx5_devcom *devcom, comp = &devcom->priv->components[id]; WARN_ON(!rwsem_is_locked(&comp->sem)); - comp->paired = paired; + WRITE_ONCE(comp->paired, paired); } bool mlx5_devcom_is_paired(struct mlx5_devcom *devcom, @@ -222,7 +226,7 @@ bool mlx5_devcom_is_paired(struct mlx5_devcom *devcom, if (IS_ERR_OR_NULL(devcom)) return false; - return devcom->priv->components[id].paired; + return READ_ONCE(devcom->priv->components[id].paired); } void *mlx5_devcom_get_peer_data(struct mlx5_devcom *devcom, @@ -236,7 +240,7 @@ void *mlx5_devcom_get_peer_data(struct mlx5_devcom *devcom, comp = &devcom->priv->components[id]; down_read(&comp->sem); - if (!comp->paired) { + if (!READ_ONCE(comp->paired)) { up_read(&comp->sem); return NULL; } @@ -245,7 +249,29 @@ void *mlx5_devcom_get_peer_data(struct mlx5_devcom *devcom, if (i != devcom->idx) break; - return comp->device[i].data; + return rcu_dereference_protected(comp->device[i].data, lockdep_is_held(&comp->sem)); +} + +void *mlx5_devcom_get_peer_data_rcu(struct mlx5_devcom *devcom, enum mlx5_devcom_components id) +{ + struct mlx5_devcom_component *comp; + int i; + + if (IS_ERR_OR_NULL(devcom)) + return NULL; + + for (i = 0; i < MLX5_DEVCOM_PORTS_SUPPORTED; i++) + if (i != devcom->idx) + break; + + comp = &devcom->priv->components[id]; + /* This can change concurrently, however 'data' pointer will remain + * valid for the duration of RCU read section. + */ + if (!READ_ONCE(comp->paired)) + return NULL; + + return rcu_dereference(comp->device[i].data); } void mlx5_devcom_release_peer_data(struct mlx5_devcom *devcom, diff --git a/drivers/net/ethernet/mellanox/mlx5/core/lib/devcom.h b/drivers/net/ethernet/mellanox/mlx5/core/lib/devcom.h index 94313c18bb64..9a496f4722da 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/lib/devcom.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/lib/devcom.h @@ -41,6 +41,7 @@ bool mlx5_devcom_is_paired(struct mlx5_devcom *devcom, void *mlx5_devcom_get_peer_data(struct mlx5_devcom *devcom, enum mlx5_devcom_components id); +void *mlx5_devcom_get_peer_data_rcu(struct mlx5_devcom *devcom, enum mlx5_devcom_components id); void mlx5_devcom_release_peer_data(struct mlx5_devcom *devcom, enum mlx5_devcom_components id);

2 years, 7 months

1
0
0 0

FAILED: patch "[PATCH] net/mlx5e: Fix SQ wake logic in ptp napi_poll context" failed to apply to 5.15-stable tree

by gregkh＠linuxfoundation.org

The patch below does not apply to the 5.15-stable tree. If someone wants it applied there, or to any other stable or longterm tree, then please email the backport, including the original git commit id to <stable(a)vger.kernel.org>. To reproduce the conflict and resubmit, you may use the following commands: git fetch https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/ linux-5.15.y git checkout FETCH_HEAD git cherry-pick -x 7aa50380191635e5897a773f272829cc961a2be5 # <resolve conflicts, build, test, etc.> git commit -s git send-email --to '<stable(a)vger.kernel.org>' --in-reply-to '2023052800-antennae-collected-11b6@gregkh' --subject-prefix 'PATCH 5.15.y' HEAD^.. Possible dependencies: 7aa503801916 ("net/mlx5e: Fix SQ wake logic in ptp napi_poll context") 19b43a432e3e ("net/mlx5e: Extend SKB room check to include PTP-SQ") b8d91145ed7c ("net/mlx5e: Fix wrong calculation of header index in HW_GRO") 92552d3abd32 ("net/mlx5e: HW_GRO cqe handler implementation") 64509b052525 ("net/mlx5e: Add data path for SHAMPO feature") f97d5c2a453e ("net/mlx5e: Add handle SHAMPO cqe support") e5ca8fb08ab2 ("net/mlx5e: Add control path for SHAMPO feature") eaee12f04692 ("net/mlx5e: Rename TIR lro functions to TIR packet merge functions") 50f477fe9933 ("net/mlx5e: Rename lro_timeout to packet_merge_timeout") thanks, greg k-h ------------------ original commit in Linus's tree ------------------ From 7aa50380191635e5897a773f272829cc961a2be5 Mon Sep 17 00:00:00 2001 From: Rahul Rameshbabu <rrameshbabu(a)nvidia.com> Date: Tue, 21 Feb 2023 16:18:48 -0800 Subject: [PATCH] net/mlx5e: Fix SQ wake logic in ptp napi_poll context Check in the mlx5e_ptp_poll_ts_cq context if the ptp tx sq should be woken up. Before change, the ptp tx sq may never wake up if the ptp tx ts skb fifo is full when mlx5e_poll_tx_cq checks if the queue should be woken up. Fixes: 1880bc4e4a96 ("net/mlx5e: Add TX port timestamp support") Signed-off-by: Rahul Rameshbabu <rrameshbabu(a)nvidia.com> Reviewed-by: Tariq Toukan <tariqt(a)nvidia.com> Signed-off-by: Saeed Mahameed <saeedm(a)nvidia.com> diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/ptp.c b/drivers/net/ethernet/mellanox/mlx5/core/en/ptp.c index eb5abd0e55d9..3cbebfba582b 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/ptp.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/ptp.c @@ -175,6 +175,8 @@ static bool mlx5e_ptp_poll_ts_cq(struct mlx5e_cq *cq, int budget) /* ensure cq space is freed before enabling more cqes */ wmb(); + mlx5e_txqsq_wake(&ptpsq->txqsq); + return work_done == budget; } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/txrx.h b/drivers/net/ethernet/mellanox/mlx5/core/en/txrx.h index 47381e949f1f..879d698b6119 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/txrx.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/txrx.h @@ -193,6 +193,8 @@ static inline u16 mlx5e_txqsq_get_next_pi(struct mlx5e_txqsq *sq, u16 size) return pi; } +void mlx5e_txqsq_wake(struct mlx5e_txqsq *sq); + static inline u16 mlx5e_shampo_get_cqe_header_index(struct mlx5e_rq *rq, struct mlx5_cqe64 *cqe) { return be16_to_cpu(cqe->shampo.header_entry_index) & (rq->mpwqe.shampo->hd_per_wq - 1); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c index df5e780e8e6a..c7eb6b238c2b 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c @@ -762,6 +762,17 @@ static void mlx5e_tx_wi_consume_fifo_skbs(struct mlx5e_txqsq *sq, struct mlx5e_t } } +void mlx5e_txqsq_wake(struct mlx5e_txqsq *sq) +{ + if (netif_tx_queue_stopped(sq->txq) && + mlx5e_wqc_has_room_for(&sq->wq, sq->cc, sq->pc, sq->stop_room) && + mlx5e_ptpsq_fifo_has_room(sq) && + !test_bit(MLX5E_SQ_STATE_RECOVERING, &sq->state)) { + netif_tx_wake_queue(sq->txq); + sq->stats->wake++; + } +} + bool mlx5e_poll_tx_cq(struct mlx5e_cq *cq, int napi_budget) { struct mlx5e_sq_stats *stats; @@ -861,13 +872,7 @@ bool mlx5e_poll_tx_cq(struct mlx5e_cq *cq, int napi_budget) netdev_tx_completed_queue(sq->txq, npkts, nbytes); - if (netif_tx_queue_stopped(sq->txq) && - mlx5e_wqc_has_room_for(&sq->wq, sq->cc, sq->pc, sq->stop_room) && - mlx5e_ptpsq_fifo_has_room(sq) && - !test_bit(MLX5E_SQ_STATE_RECOVERING, &sq->state)) { - netif_tx_wake_queue(sq->txq); - stats->wake++; - } + mlx5e_txqsq_wake(sq); return (i == MLX5E_TX_CQ_POLL_BUDGET); }

2 years, 7 months

1
0
0 0

FAILED: patch "[PATCH] cxl: Move cxl_await_media_ready() to before capacity info" failed to apply to 5.15-stable tree

by gregkh＠linuxfoundation.org

The patch below does not apply to the 5.15-stable tree. If someone wants it applied there, or to any other stable or longterm tree, then please email the backport, including the original git commit id to <stable(a)vger.kernel.org>. To reproduce the conflict and resubmit, you may use the following commands: git fetch https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/ linux-5.15.y git checkout FETCH_HEAD git cherry-pick -x e764f12208b99ac7892c4e3f6bf88d71ca71036f # <resolve conflicts, build, test, etc.> git commit -s git send-email --to '<stable(a)vger.kernel.org>' --in-reply-to '2023052827-dancing-proud-0d5b@gregkh' --subject-prefix 'PATCH 5.15.y' HEAD^.. Possible dependencies: e764f12208b9 ("cxl: Move cxl_await_media_ready() to before capacity info retrieval") fd35fdcbf75b ("cxl/test: Add mock test for set_timestamp") f8d22bf50ca5 ("tools/testing/cxl: Mock support for Get Poison List") a5fcd228ca1d ("Merge branch 'for-6.3/cxl-rr-emu' into cxl/next") thanks, greg k-h ------------------ original commit in Linus's tree ------------------ From e764f12208b99ac7892c4e3f6bf88d71ca71036f Mon Sep 17 00:00:00 2001 From: Dave Jiang <dave.jiang(a)intel.com> Date: Thu, 18 May 2023 16:38:20 -0700 Subject: [PATCH] cxl: Move cxl_await_media_ready() to before capacity info retrieval Move cxl_await_media_ready() to cxl_pci probe before driver starts issuing IDENTIFY and retrieving memory device information to ensure that the device is ready to provide the information. Allow cxl_pci_probe() to succeed even if media is not ready. Cache the media failure in cxlds and don't ask the device for any media information. The rationale for proceeding in the !media_ready case is to allow for mailbox operations to interrogate and/or remediate the device. After media is repaired then rebinding the cxl_pci driver is expected to restart the capacity scan. Suggested-by: Dan Williams <dan.j.williams(a)intel.com> Fixes: b39cb1052a5c ("cxl/mem: Register CXL memX devices") Reviewed-by: Ira Weiny <ira.weiny(a)intel.com> Signed-off-by: Dave Jiang <dave.jiang(a)intel.com> Link: https://lore.kernel.org/r/168445310026.3251520.8124296540679268206.stgit@dj… [djbw: fixup cxl_test] Signed-off-by: Dan Williams <dan.j.williams(a)intel.com> diff --git a/drivers/cxl/core/mbox.c b/drivers/cxl/core/mbox.c index 23b9ff920d7e..2c8dc7e2b84d 100644 --- a/drivers/cxl/core/mbox.c +++ b/drivers/cxl/core/mbox.c @@ -1028,7 +1028,7 @@ static int cxl_mem_get_partition_info(struct cxl_dev_state *cxlds) * cxl_dev_state_identify() - Send the IDENTIFY command to the device. * @cxlds: The device data for the operation * - * Return: 0 if identify was executed successfully. + * Return: 0 if identify was executed successfully or media not ready. * * This will dispatch the identify command to the device and on success populate * structures to be exported to sysfs. @@ -1041,6 +1041,9 @@ int cxl_dev_state_identify(struct cxl_dev_state *cxlds) u32 val; int rc; + if (!cxlds->media_ready) + return 0; + mbox_cmd = (struct cxl_mbox_cmd) { .opcode = CXL_MBOX_OP_IDENTIFY, .size_out = sizeof(id), @@ -1115,10 +1118,12 @@ int cxl_mem_create_range_info(struct cxl_dev_state *cxlds) cxlds->persistent_only_bytes, "pmem"); } - rc = cxl_mem_get_partition_info(cxlds); - if (rc) { - dev_err(dev, "Failed to query partition information\n"); - return rc; + if (cxlds->media_ready) { + rc = cxl_mem_get_partition_info(cxlds); + if (rc) { + dev_err(dev, "Failed to query partition information\n"); + return rc; + } } rc = add_dpa_res(dev, &cxlds->dpa_res, &cxlds->ram_res, 0, diff --git a/drivers/cxl/cxlmem.h b/drivers/cxl/cxlmem.h index db12b6313afb..a2845a7a69d8 100644 --- a/drivers/cxl/cxlmem.h +++ b/drivers/cxl/cxlmem.h @@ -266,6 +266,7 @@ struct cxl_poison_state { * @regs: Parsed register blocks * @cxl_dvsec: Offset to the PCIe device DVSEC * @rcd: operating in RCD mode (CXL 3.0 9.11.8 CXL Devices Attached to an RCH) + * @media_ready: Indicate whether the device media is usable * @payload_size: Size of space for payload * (CXL 2.0 8.2.8.4.3 Mailbox Capabilities Register) * @lsa_size: Size of Label Storage Area @@ -303,6 +304,7 @@ struct cxl_dev_state { int cxl_dvsec; bool rcd; + bool media_ready; size_t payload_size; size_t lsa_size; struct mutex mbox_mutex; /* Protects device mailbox and firmware */ diff --git a/drivers/cxl/mem.c b/drivers/cxl/mem.c index 10caf180b3fa..519edd0eb196 100644 --- a/drivers/cxl/mem.c +++ b/drivers/cxl/mem.c @@ -124,6 +124,9 @@ static int cxl_mem_probe(struct device *dev) struct dentry *dentry; int rc; + if (!cxlds->media_ready) + return -EBUSY; + /* * Someone is trying to reattach this device after it lost its port * connection (an endpoint port previously registered by this memdev was diff --git a/drivers/cxl/pci.c b/drivers/cxl/pci.c index f7a5b8e9c102..0872f2233ed0 100644 --- a/drivers/cxl/pci.c +++ b/drivers/cxl/pci.c @@ -708,6 +708,12 @@ static int cxl_pci_probe(struct pci_dev *pdev, const struct pci_device_id *id) if (rc) dev_dbg(&pdev->dev, "Failed to map RAS capability.\n"); + rc = cxl_await_media_ready(cxlds); + if (rc == 0) + cxlds->media_ready = true; + else + dev_warn(&pdev->dev, "Media not active (%d)\n", rc); + rc = cxl_pci_setup_mailbox(cxlds); if (rc) return rc; diff --git a/drivers/cxl/port.c b/drivers/cxl/port.c index 17a95f469c26..c23b6164e1c0 100644 --- a/drivers/cxl/port.c +++ b/drivers/cxl/port.c @@ -117,12 +117,6 @@ static int cxl_endpoint_port_probe(struct cxl_port *port) if (rc) return rc; - rc = cxl_await_media_ready(cxlds); - if (rc) { - dev_err(&port->dev, "Media not active (%d)\n", rc); - return rc; - } - rc = devm_cxl_enumerate_decoders(cxlhdm, &info); if (rc) return rc; diff --git a/tools/testing/cxl/test/mem.c b/tools/testing/cxl/test/mem.c index ba572d03c687..34b48027b3de 100644 --- a/tools/testing/cxl/test/mem.c +++ b/tools/testing/cxl/test/mem.c @@ -1256,6 +1256,7 @@ static int cxl_mock_mem_probe(struct platform_device *pdev) if (rc) return rc; + cxlds->media_ready = true; rc = cxl_dev_state_identify(cxlds); if (rc) return rc;

2 years, 7 months

1
0
0 0

FAILED: patch "[PATCH] cxl: Move cxl_await_media_ready() to before capacity info" failed to apply to 6.1-stable tree

by gregkh＠linuxfoundation.org

The patch below does not apply to the 6.1-stable tree. If someone wants it applied there, or to any other stable or longterm tree, then please email the backport, including the original git commit id to <stable(a)vger.kernel.org>. To reproduce the conflict and resubmit, you may use the following commands: git fetch https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/ linux-6.1.y git checkout FETCH_HEAD git cherry-pick -x e764f12208b99ac7892c4e3f6bf88d71ca71036f # <resolve conflicts, build, test, etc.> git commit -s git send-email --to '<stable(a)vger.kernel.org>' --in-reply-to '2023052826-shallow-paltry-04ed@gregkh' --subject-prefix 'PATCH 6.1.y' HEAD^.. Possible dependencies: e764f12208b9 ("cxl: Move cxl_await_media_ready() to before capacity info retrieval") fd35fdcbf75b ("cxl/test: Add mock test for set_timestamp") f8d22bf50ca5 ("tools/testing/cxl: Mock support for Get Poison List") a5fcd228ca1d ("Merge branch 'for-6.3/cxl-rr-emu' into cxl/next") thanks, greg k-h ------------------ original commit in Linus's tree ------------------ From e764f12208b99ac7892c4e3f6bf88d71ca71036f Mon Sep 17 00:00:00 2001 From: Dave Jiang <dave.jiang(a)intel.com> Date: Thu, 18 May 2023 16:38:20 -0700 Subject: [PATCH] cxl: Move cxl_await_media_ready() to before capacity info retrieval Move cxl_await_media_ready() to cxl_pci probe before driver starts issuing IDENTIFY and retrieving memory device information to ensure that the device is ready to provide the information. Allow cxl_pci_probe() to succeed even if media is not ready. Cache the media failure in cxlds and don't ask the device for any media information. The rationale for proceeding in the !media_ready case is to allow for mailbox operations to interrogate and/or remediate the device. After media is repaired then rebinding the cxl_pci driver is expected to restart the capacity scan. Suggested-by: Dan Williams <dan.j.williams(a)intel.com> Fixes: b39cb1052a5c ("cxl/mem: Register CXL memX devices") Reviewed-by: Ira Weiny <ira.weiny(a)intel.com> Signed-off-by: Dave Jiang <dave.jiang(a)intel.com> Link: https://lore.kernel.org/r/168445310026.3251520.8124296540679268206.stgit@dj… [djbw: fixup cxl_test] Signed-off-by: Dan Williams <dan.j.williams(a)intel.com> diff --git a/drivers/cxl/core/mbox.c b/drivers/cxl/core/mbox.c index 23b9ff920d7e..2c8dc7e2b84d 100644 --- a/drivers/cxl/core/mbox.c +++ b/drivers/cxl/core/mbox.c @@ -1028,7 +1028,7 @@ static int cxl_mem_get_partition_info(struct cxl_dev_state *cxlds) * cxl_dev_state_identify() - Send the IDENTIFY command to the device. * @cxlds: The device data for the operation * - * Return: 0 if identify was executed successfully. + * Return: 0 if identify was executed successfully or media not ready. * * This will dispatch the identify command to the device and on success populate * structures to be exported to sysfs. @@ -1041,6 +1041,9 @@ int cxl_dev_state_identify(struct cxl_dev_state *cxlds) u32 val; int rc; + if (!cxlds->media_ready) + return 0; + mbox_cmd = (struct cxl_mbox_cmd) { .opcode = CXL_MBOX_OP_IDENTIFY, .size_out = sizeof(id), @@ -1115,10 +1118,12 @@ int cxl_mem_create_range_info(struct cxl_dev_state *cxlds) cxlds->persistent_only_bytes, "pmem"); } - rc = cxl_mem_get_partition_info(cxlds); - if (rc) { - dev_err(dev, "Failed to query partition information\n"); - return rc; + if (cxlds->media_ready) { + rc = cxl_mem_get_partition_info(cxlds); + if (rc) { + dev_err(dev, "Failed to query partition information\n"); + return rc; + } } rc = add_dpa_res(dev, &cxlds->dpa_res, &cxlds->ram_res, 0, diff --git a/drivers/cxl/cxlmem.h b/drivers/cxl/cxlmem.h index db12b6313afb..a2845a7a69d8 100644 --- a/drivers/cxl/cxlmem.h +++ b/drivers/cxl/cxlmem.h @@ -266,6 +266,7 @@ struct cxl_poison_state { * @regs: Parsed register blocks * @cxl_dvsec: Offset to the PCIe device DVSEC * @rcd: operating in RCD mode (CXL 3.0 9.11.8 CXL Devices Attached to an RCH) + * @media_ready: Indicate whether the device media is usable * @payload_size: Size of space for payload * (CXL 2.0 8.2.8.4.3 Mailbox Capabilities Register) * @lsa_size: Size of Label Storage Area @@ -303,6 +304,7 @@ struct cxl_dev_state { int cxl_dvsec; bool rcd; + bool media_ready; size_t payload_size; size_t lsa_size; struct mutex mbox_mutex; /* Protects device mailbox and firmware */ diff --git a/drivers/cxl/mem.c b/drivers/cxl/mem.c index 10caf180b3fa..519edd0eb196 100644 --- a/drivers/cxl/mem.c +++ b/drivers/cxl/mem.c @@ -124,6 +124,9 @@ static int cxl_mem_probe(struct device *dev) struct dentry *dentry; int rc; + if (!cxlds->media_ready) + return -EBUSY; + /* * Someone is trying to reattach this device after it lost its port * connection (an endpoint port previously registered by this memdev was diff --git a/drivers/cxl/pci.c b/drivers/cxl/pci.c index f7a5b8e9c102..0872f2233ed0 100644 --- a/drivers/cxl/pci.c +++ b/drivers/cxl/pci.c @@ -708,6 +708,12 @@ static int cxl_pci_probe(struct pci_dev *pdev, const struct pci_device_id *id) if (rc) dev_dbg(&pdev->dev, "Failed to map RAS capability.\n"); + rc = cxl_await_media_ready(cxlds); + if (rc == 0) + cxlds->media_ready = true; + else + dev_warn(&pdev->dev, "Media not active (%d)\n", rc); + rc = cxl_pci_setup_mailbox(cxlds); if (rc) return rc; diff --git a/drivers/cxl/port.c b/drivers/cxl/port.c index 17a95f469c26..c23b6164e1c0 100644 --- a/drivers/cxl/port.c +++ b/drivers/cxl/port.c @@ -117,12 +117,6 @@ static int cxl_endpoint_port_probe(struct cxl_port *port) if (rc) return rc; - rc = cxl_await_media_ready(cxlds); - if (rc) { - dev_err(&port->dev, "Media not active (%d)\n", rc); - return rc; - } - rc = devm_cxl_enumerate_decoders(cxlhdm, &info); if (rc) return rc; diff --git a/tools/testing/cxl/test/mem.c b/tools/testing/cxl/test/mem.c index ba572d03c687..34b48027b3de 100644 --- a/tools/testing/cxl/test/mem.c +++ b/tools/testing/cxl/test/mem.c @@ -1256,6 +1256,7 @@ static int cxl_mock_mem_probe(struct platform_device *pdev) if (rc) return rc; + cxlds->media_ready = true; rc = cxl_dev_state_identify(cxlds); if (rc) return rc;

2 years, 7 months

1
0
0 0

2025

2024

2023

2022

2021

2020

2019

2018

2017

Linux-stable-mirror