6.6-stable review patch. If anyone has any objections, please let me know.
------------------
From: Moshe Shemesh moshe@nvidia.com
[ Upstream commit 902a8bc23a24882200f57cadc270e15a2cfaf2bb ]
Fix lockdep assertion triggered during sync reset unload event. When the sync reset flow is initiated using the devlink reload fw_activate option, the PF already holds the devlink lock while handling unload event. In this case, delegate sync reset unload event handling back to the devlink callback process to avoid double-locking and resolve the lockdep warning.
Kernel log: WARNING: CPU: 9 PID: 1578 at devl_assert_locked+0x31/0x40 [...] Call Trace: <TASK> mlx5_unload_one_devl_locked+0x2c/0xc0 [mlx5_core] mlx5_sync_reset_unload_event+0xaf/0x2f0 [mlx5_core] process_one_work+0x222/0x640 worker_thread+0x199/0x350 kthread+0x10b/0x230 ? __pfx_worker_thread+0x10/0x10 ? __pfx_kthread+0x10/0x10 ret_from_fork+0x8e/0x100 ? __pfx_kthread+0x10/0x10 ret_from_fork_asm+0x1a/0x30 </TASK>
Fixes: 7a9770f1bfea ("net/mlx5: Handle sync reset unload event") Signed-off-by: Moshe Shemesh moshe@nvidia.com Signed-off-by: Mark Bloch mbloch@nvidia.com Link: https://patch.msgid.link/20250825143435.598584-7-mbloch@nvidia.com Signed-off-by: Jakub Kicinski kuba@kernel.org Signed-off-by: Sasha Levin sashal@kernel.org --- .../net/ethernet/mellanox/mlx5/core/devlink.c | 2 +- .../ethernet/mellanox/mlx5/core/fw_reset.c | 108 ++++++++++-------- .../ethernet/mellanox/mlx5/core/fw_reset.h | 1 + 3 files changed, 63 insertions(+), 48 deletions(-)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/devlink.c b/drivers/net/ethernet/mellanox/mlx5/core/devlink.c index f2d1f7cad7e72..8489b5087d9c6 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/devlink.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/devlink.c @@ -107,7 +107,7 @@ static int mlx5_devlink_reload_fw_activate(struct devlink *devlink, struct netli if (err) return err;
- mlx5_unload_one_devl_locked(dev, false); + mlx5_sync_reset_unload_flow(dev, true); err = mlx5_health_wait_pci_up(dev); if (err) NL_SET_ERR_MSG_MOD(extack, "FW activate aborted, PCI reads fail after reset"); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fw_reset.c b/drivers/net/ethernet/mellanox/mlx5/core/fw_reset.c index a1c2dd7f5c2c0..b9986a2656083 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/fw_reset.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/fw_reset.c @@ -12,7 +12,8 @@ enum { MLX5_FW_RESET_FLAGS_NACK_RESET_REQUEST, MLX5_FW_RESET_FLAGS_PENDING_COMP, MLX5_FW_RESET_FLAGS_DROP_NEW_REQUESTS, - MLX5_FW_RESET_FLAGS_RELOAD_REQUIRED + MLX5_FW_RESET_FLAGS_RELOAD_REQUIRED, + MLX5_FW_RESET_FLAGS_UNLOAD_EVENT, };
struct mlx5_fw_reset { @@ -217,7 +218,7 @@ int mlx5_fw_reset_set_live_patch(struct mlx5_core_dev *dev) return mlx5_reg_mfrl_set(dev, MLX5_MFRL_REG_RESET_LEVEL0, 0, 0, false); }
-static void mlx5_fw_reset_complete_reload(struct mlx5_core_dev *dev, bool unloaded) +static void mlx5_fw_reset_complete_reload(struct mlx5_core_dev *dev) { struct mlx5_fw_reset *fw_reset = dev->priv.fw_reset; struct devlink *devlink = priv_to_devlink(dev); @@ -226,8 +227,7 @@ static void mlx5_fw_reset_complete_reload(struct mlx5_core_dev *dev, bool unload if (test_bit(MLX5_FW_RESET_FLAGS_PENDING_COMP, &fw_reset->reset_flags)) { complete(&fw_reset->done); } else { - if (!unloaded) - mlx5_unload_one(dev, false); + mlx5_sync_reset_unload_flow(dev, false); if (mlx5_health_wait_pci_up(dev)) mlx5_core_err(dev, "reset reload flow aborted, PCI reads still not working\n"); else @@ -270,7 +270,7 @@ static void mlx5_sync_reset_reload_work(struct work_struct *work)
mlx5_sync_reset_clear_reset_requested(dev, false); mlx5_enter_error_state(dev, true); - mlx5_fw_reset_complete_reload(dev, false); + mlx5_fw_reset_complete_reload(dev); }
#define MLX5_RESET_POLL_INTERVAL (HZ / 10) @@ -553,6 +553,59 @@ static int mlx5_sync_pci_reset(struct mlx5_core_dev *dev, u8 reset_method) return err; }
+void mlx5_sync_reset_unload_flow(struct mlx5_core_dev *dev, bool locked) +{ + struct mlx5_fw_reset *fw_reset = dev->priv.fw_reset; + unsigned long timeout; + bool reset_action; + u8 rst_state; + int err; + + if (locked) + mlx5_unload_one_devl_locked(dev, false); + else + mlx5_unload_one(dev, false); + + if (!test_bit(MLX5_FW_RESET_FLAGS_UNLOAD_EVENT, &fw_reset->reset_flags)) + return; + + mlx5_set_fw_rst_ack(dev); + mlx5_core_warn(dev, "Sync Reset Unload done, device reset expected\n"); + + reset_action = false; + timeout = jiffies + msecs_to_jiffies(mlx5_tout_ms(dev, RESET_UNLOAD)); + do { + rst_state = mlx5_get_fw_rst_state(dev); + if (rst_state == MLX5_FW_RST_STATE_TOGGLE_REQ || + rst_state == MLX5_FW_RST_STATE_IDLE) { + reset_action = true; + break; + } + msleep(20); + } while (!time_after(jiffies, timeout)); + + if (!reset_action) { + mlx5_core_err(dev, "Got timeout waiting for sync reset action, state = %u\n", + rst_state); + fw_reset->ret = -ETIMEDOUT; + goto done; + } + + mlx5_core_warn(dev, "Sync Reset, got reset action. rst_state = %u\n", + rst_state); + if (rst_state == MLX5_FW_RST_STATE_TOGGLE_REQ) { + err = mlx5_sync_pci_reset(dev, fw_reset->reset_method); + if (err) { + mlx5_core_warn(dev, "mlx5_sync_pci_reset failed, err %d\n", + err); + fw_reset->ret = err; + } + } + +done: + clear_bit(MLX5_FW_RESET_FLAGS_UNLOAD_EVENT, &fw_reset->reset_flags); +} + static void mlx5_sync_reset_now_event(struct work_struct *work) { struct mlx5_fw_reset *fw_reset = container_of(work, struct mlx5_fw_reset, @@ -580,16 +633,13 @@ static void mlx5_sync_reset_now_event(struct work_struct *work) mlx5_enter_error_state(dev, true); done: fw_reset->ret = err; - mlx5_fw_reset_complete_reload(dev, false); + mlx5_fw_reset_complete_reload(dev); }
static void mlx5_sync_reset_unload_event(struct work_struct *work) { struct mlx5_fw_reset *fw_reset; struct mlx5_core_dev *dev; - unsigned long timeout; - bool reset_action; - u8 rst_state; int err;
fw_reset = container_of(work, struct mlx5_fw_reset, reset_unload_work); @@ -598,6 +648,7 @@ static void mlx5_sync_reset_unload_event(struct work_struct *work) if (mlx5_sync_reset_clear_reset_requested(dev, false)) return;
+ set_bit(MLX5_FW_RESET_FLAGS_UNLOAD_EVENT, &fw_reset->reset_flags); mlx5_core_warn(dev, "Sync Reset Unload. Function is forced down.\n");
err = mlx5_cmd_fast_teardown_hca(dev); @@ -606,44 +657,7 @@ static void mlx5_sync_reset_unload_event(struct work_struct *work) else mlx5_enter_error_state(dev, true);
- if (test_bit(MLX5_FW_RESET_FLAGS_PENDING_COMP, &fw_reset->reset_flags)) - mlx5_unload_one_devl_locked(dev, false); - else - mlx5_unload_one(dev, false); - - mlx5_set_fw_rst_ack(dev); - mlx5_core_warn(dev, "Sync Reset Unload done, device reset expected\n"); - - reset_action = false; - timeout = jiffies + msecs_to_jiffies(mlx5_tout_ms(dev, RESET_UNLOAD)); - do { - rst_state = mlx5_get_fw_rst_state(dev); - if (rst_state == MLX5_FW_RST_STATE_TOGGLE_REQ || - rst_state == MLX5_FW_RST_STATE_IDLE) { - reset_action = true; - break; - } - msleep(20); - } while (!time_after(jiffies, timeout)); - - if (!reset_action) { - mlx5_core_err(dev, "Got timeout waiting for sync reset action, state = %u\n", - rst_state); - fw_reset->ret = -ETIMEDOUT; - goto done; - } - - mlx5_core_warn(dev, "Sync Reset, got reset action. rst_state = %u\n", rst_state); - if (rst_state == MLX5_FW_RST_STATE_TOGGLE_REQ) { - err = mlx5_sync_pci_reset(dev, fw_reset->reset_method); - if (err) { - mlx5_core_warn(dev, "mlx5_sync_pci_reset failed, err %d\n", err); - fw_reset->ret = err; - } - } - -done: - mlx5_fw_reset_complete_reload(dev, true); + mlx5_fw_reset_complete_reload(dev); }
static void mlx5_sync_reset_abort_event(struct work_struct *work) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fw_reset.h b/drivers/net/ethernet/mellanox/mlx5/core/fw_reset.h index ea527d06a85f0..d5b28525c960d 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/fw_reset.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/fw_reset.h @@ -12,6 +12,7 @@ int mlx5_fw_reset_set_reset_sync(struct mlx5_core_dev *dev, u8 reset_type_sel, int mlx5_fw_reset_set_live_patch(struct mlx5_core_dev *dev);
int mlx5_fw_reset_wait_reset_done(struct mlx5_core_dev *dev); +void mlx5_sync_reset_unload_flow(struct mlx5_core_dev *dev, bool locked); int mlx5_fw_reset_verify_fw_complete(struct mlx5_core_dev *dev, struct netlink_ext_ack *extack); void mlx5_fw_reset_events_start(struct mlx5_core_dev *dev);