When host is powered off, firmware does not send vhca_state event
for every probed host SF on the DPU because it may have deployed
thousands of SFs to the host. Instead it sends a function changed
event. Currently, only VFs handle this event. This commit extends
support to SFs.
When DPU user deactivates[1] SFs, mlx5 expects vhca_state event
and leaves the SF in dangling state[2].
When DPU user deletes[3] SFs, mlx5 also expects vhca_state event
and destroys the SF resources[4].
Fix it by changing SF to the right state and freeing SF resources
when the function changed event is received.
When this event is received, driver checks all SF states.
- If state is in_use, change it to active.
- If state is teardown_request, change it to allocated.
And SF hardware table entry is freed if it is pending for delete.
[1]
# devlink port function set en3f0c1pf0sf0 state inactive
[2]
# devlink port function set en3f0c1pf0sf0 state active
Error: mlx5_core: SF is inactivated but it is still attached.
kernel answers: Device or resource busy
[3]
# devlink port show
pci/0000:03:00.0/229376: type eth netdev en3f0c1pf0sf0 \
flavour pcisf controller 1 pfnum 0 sfnum 0 splittable false
function:
hw_addr 00:00:00:00:00:00 state active opstate attached \
roce enable trust off max_uc_macs 4096 max_io_eqs 8
# devlink port del en3f0c1pf0sf0
[4]
# devlink port add pci/0000:03:00.0 flavour pcisf pfnum 0 sfnum 0 \
controller 1
Error: mlx5_core: SF already exist. Choose different sfnum.
kernel answers: File exists
Fixes: 6a3273217469 ("net/mlx5: SF, Port function state change support")
Signed-off-by: Chris Mi <cmi@nvidia.com>
Reviewed-by: Shay Drori <shayd@nvidia.com>
Signed-off-by: Tariq Toukan <tariqt@nvidia.com>
Link: https://patch.msgid.link/20260729071622.2423270-1-tariqt@nvidia.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
mutex_destroy(&esw->fdb_table.offloads.vports.lock);
}
-static void esw_vfs_changed_event_handler(struct mlx5_eswitch *esw)
+static void esw_changed_event_handler(struct mlx5_eswitch *esw)
{
struct mlx5_esw_pf_info host_pf_info;
u16 new_num_vfs;
host_pf_info = mlx5_esw_get_host_pf_info(esw->dev, out);
new_num_vfs = host_pf_info.num_of_vfs;
+ if (host_pf_info.pf_disabled) {
+ mlx5_sf_table_esw_changed_event_handler(esw->dev);
+ mlx5_sf_hw_table_esw_changed_event_handler(esw->dev);
+ }
+
if (new_num_vfs == esw->esw_funcs.num_vfs || host_pf_info.pf_disabled)
goto free;
esw_funcs = mlx5_nb_cof(nb, struct mlx5_esw_functions, nb);
esw = container_of(esw_funcs, struct mlx5_eswitch, esw_funcs);
- ret = mlx5_esw_add_work(esw, esw_vfs_changed_event_handler,
- GFP_ATOMIC);
+ ret = mlx5_esw_add_work(esw, esw_changed_event_handler, GFP_ATOMIC);
if (ret)
return NOTIFY_DONE;
return xa_empty(&table->function_ids);
}
+
+void mlx5_sf_table_esw_changed_event_handler(struct mlx5_core_dev *dev)
+{
+ struct mlx5_sf_table *table = dev->priv.sf_table;
+ unsigned long index;
+ struct mlx5_sf *sf;
+
+ trace_mlx5_sf_host_pf_disabled(dev);
+
+ if (!table)
+ return;
+
+ mutex_lock(&table->sf_state_lock);
+ xa_for_each(&table->function_ids, index, sf) {
+ if (!sf->controller)
+ continue;
+
+ if (sf->hw_state == MLX5_VHCA_STATE_IN_USE)
+ sf->hw_state = MLX5_VHCA_STATE_ACTIVE;
+ else if (sf->hw_state == MLX5_VHCA_STATE_TEARDOWN_REQUEST)
+ sf->hw_state = MLX5_VHCA_STATE_ALLOCATED;
+ else
+ continue;
+ trace_mlx5_sf_update_state(table->dev, sf->port_index,
+ sf->controller, sf->hw_fn_id,
+ sf->hw_state);
+ }
+ mutex_unlock(&table->sf_state_lock);
+}
#include <linux/mlx5/driver.h>
#include "sf/vhca_event.h"
+TRACE_EVENT(mlx5_sf_host_pf_disabled,
+ TP_PROTO(const struct mlx5_core_dev *dev),
+ TP_ARGS(dev),
+ TP_STRUCT__entry(__string(devname, dev_name(dev->device))),
+ TP_fast_assign(__assign_str(devname);),
+ TP_printk("(%s)\n", __get_str(devname))
+);
+
TRACE_EVENT(mlx5_sf_add,
TP_PROTO(const struct mlx5_core_dev *dev,
unsigned int port_index,
{
return !!dev->priv.sf_hw_table;
}
+
+void mlx5_sf_hw_table_esw_changed_event_handler(struct mlx5_core_dev *dev)
+{
+ struct mlx5_sf_hw_table *table;
+ struct mlx5_sf_hwc_table *hwc;
+ int i;
+
+ table = dev->priv.sf_hw_table;
+ if (!table)
+ return;
+
+ mutex_lock(&table->table_lock);
+ hwc = &table->hwc[MLX5_SF_HWC_EXT_HOST];
+ for (i = 0; i < hwc->max_fn; i++) {
+ struct mlx5_sf_hw *sf_hw;
+
+ sf_hw = &hwc->sfs[i];
+ if (sf_hw->allocated && sf_hw->pending_delete)
+ mlx5_sf_hw_table_hwc_sf_free(dev, hwc, i);
+ }
+ mutex_unlock(&table->table_lock);
+}
int mlx5_sf_hw_notifier_init(struct mlx5_core_dev *dev);
void mlx5_sf_hw_notifier_cleanup(struct mlx5_core_dev *dev);
void mlx5_sf_hw_table_destroy(struct mlx5_core_dev *dev);
+void mlx5_sf_hw_table_esw_changed_event_handler(struct mlx5_core_dev *dev);
int mlx5_sf_notifiers_init(struct mlx5_core_dev *dev);
int mlx5_sf_table_init(struct mlx5_core_dev *dev);
void mlx5_sf_notifiers_cleanup(struct mlx5_core_dev *dev);
void mlx5_sf_table_cleanup(struct mlx5_core_dev *dev);
bool mlx5_sf_table_empty(const struct mlx5_core_dev *dev);
+void mlx5_sf_table_esw_changed_event_handler(struct mlx5_core_dev *dev);
int mlx5_devlink_sf_port_new(struct devlink *devlink,
const struct devlink_port_new_attrs *add_attr,
{
}
+static inline void
+mlx5_sf_hw_table_esw_changed_event_handler(struct mlx5_core_dev *dev)
+{
+}
+
static inline int mlx5_sf_notifiers_init(struct mlx5_core_dev *dev)
{
return 0;
return true;
}
+static inline void
+mlx5_sf_table_esw_changed_event_handler(struct mlx5_core_dev *dev)
+{
+}
+
#endif
#endif
projects / thirdparty / kernel / linux.git / commitdiff
