Commit 6f6f0ac6 authored by Jakub Kicinski's avatar Jakub Kicinski
Browse files

Merge tag 'mlx5-fixes-2021-12-22' of git://git.kernel.org/pub/scm/linux/kernel/git/saeed/linux 

Saeed Mahameed says:

====================
mlx5 fixes 2021-12-22

This series provides bug fixes to mlx5 driver.

* tag 'mlx5-fixes-2021-12-22' of git://git.kernel.org/pub/scm/linux/kernel/git/saeed/linux:
  net/mlx5: Fix some error handling paths in 'mlx5e_tc_add_fdb_flow()'
  net/mlx5e: Delete forward rule for ct or sample action
  net/mlx5e: Fix ICOSQ recovery flow for XSK
  net/mlx5e: Fix interoperability between XSK and ICOSQ recovery flow
  net/mlx5e: Fix skb memory leak when TC classifier action offloads are disabled
  net/mlx5e: Wrap the tx reporter dump callback to extract the sq
  net/mlx5: Fix tc max supported prio for nic mode
  net/mlx5: Fix SF health recovery flow
  net/mlx5: Fix error print in case of IRQ request failed
  net/mlx5: Use first online CPU instead of hard coded CPU
  net/mlx5: DR, Fix querying eswitch manager vport for ECPF
  net/mlx5: DR, Fix NULL vs IS_ERR checking in dr_domain_init_resources
====================

Link: https://lore.kernel.org/r/20211223190441.153012-1-saeed@kernel.org


Signed-off-by: default avatarJakub Kicinski <kuba@kernel.org>
parents 95b40115 4390c6ed

drivers/net/ethernet/mellanox/mlx5/core/en.h

+2 −3
Original line number Diff line number Diff line
@@ -783,6 +783,8 @@ struct mlx5e_channel { 
	DECLARE_BITMAP(state, MLX5E_CHANNEL_NUM_STATES);

	int                        ix;

	int                        cpu;

	/* Sync between icosq recovery and XSK enable/disable. */

	struct mutex               icosq_recovery_lock;

};



struct mlx5e_ptp;
@@ -1014,9 +1016,6 @@ int mlx5e_create_rq(struct mlx5e_rq *rq, struct mlx5e_rq_param *param); 
void mlx5e_destroy_rq(struct mlx5e_rq *rq);



struct mlx5e_sq_param;

int mlx5e_open_icosq(struct mlx5e_channel *c, struct mlx5e_params *params,

		     struct mlx5e_sq_param *param, struct mlx5e_icosq *sq);

void mlx5e_close_icosq(struct mlx5e_icosq *sq);

int mlx5e_open_xdpsq(struct mlx5e_channel *c, struct mlx5e_params *params,

		     struct mlx5e_sq_param *param, struct xsk_buff_pool *xsk_pool,

		     struct mlx5e_xdpsq *sq, bool is_redirect);

drivers/net/ethernet/mellanox/mlx5/core/en/health.h

+2 −0
Original line number Diff line number Diff line
@@ -30,6 +30,8 @@ void mlx5e_reporter_rx_destroy(struct mlx5e_priv *priv); 
void mlx5e_reporter_icosq_cqe_err(struct mlx5e_icosq *icosq);

void mlx5e_reporter_rq_cqe_err(struct mlx5e_rq *rq);

void mlx5e_reporter_rx_timeout(struct mlx5e_rq *rq);

void mlx5e_reporter_icosq_suspend_recovery(struct mlx5e_channel *c);

void mlx5e_reporter_icosq_resume_recovery(struct mlx5e_channel *c);



#define MLX5E_REPORTER_PER_Q_MAX_LEN 256

drivers/net/ethernet/mellanox/mlx5/core/en/rep/tc.h

+1 −1
Original line number Diff line number Diff line
@@ -66,7 +66,7 @@ mlx5e_rep_setup_tc(struct net_device *dev, enum tc_setup_type type, 


static inline void

mlx5e_rep_tc_receive(struct mlx5_cqe64 *cqe, struct mlx5e_rq *rq,

		     struct sk_buff *skb) {}

		     struct sk_buff *skb) { napi_gro_receive(rq->cq.napi, skb); }



#endif /* CONFIG_MLX5_CLS_ACT */

drivers/net/ethernet/mellanox/mlx5/core/en/reporter_rx.c

+34 −1
Original line number Diff line number Diff line
@@ -62,6 +62,7 @@ static void mlx5e_reset_icosq_cc_pc(struct mlx5e_icosq *icosq) 


static int mlx5e_rx_reporter_err_icosq_cqe_recover(void *ctx)

{

	struct mlx5e_rq *xskrq = NULL;

	struct mlx5_core_dev *mdev;

	struct mlx5e_icosq *icosq;

	struct net_device *dev;
@@ -70,7 +71,13 @@ static int mlx5e_rx_reporter_err_icosq_cqe_recover(void *ctx) 
	int err;



	icosq = ctx;



	mutex_lock(&icosq->channel->icosq_recovery_lock);



	/* mlx5e_close_rq cancels this work before RQ and ICOSQ are killed. */

	rq = &icosq->channel->rq;

	if (test_bit(MLX5E_RQ_STATE_ENABLED, &icosq->channel->xskrq.state))

		xskrq = &icosq->channel->xskrq;

	mdev = icosq->channel->mdev;

	dev = icosq->channel->netdev;

	err = mlx5_core_query_sq_state(mdev, icosq->sqn, &state);
@@ -84,6 +91,9 @@ static int mlx5e_rx_reporter_err_icosq_cqe_recover(void *ctx) 
		goto out;



	mlx5e_deactivate_rq(rq);

	if (xskrq)

		mlx5e_deactivate_rq(xskrq);



	err = mlx5e_wait_for_icosq_flush(icosq);

	if (err)

		goto out;
@@ -97,15 +107,28 @@ static int mlx5e_rx_reporter_err_icosq_cqe_recover(void *ctx) 
		goto out;



	mlx5e_reset_icosq_cc_pc(icosq);



	mlx5e_free_rx_in_progress_descs(rq);

	if (xskrq)

		mlx5e_free_rx_in_progress_descs(xskrq);



	clear_bit(MLX5E_SQ_STATE_RECOVERING, &icosq->state);

	mlx5e_activate_icosq(icosq);

	mlx5e_activate_rq(rq);



	mlx5e_activate_rq(rq);

	rq->stats->recover++;



	if (xskrq) {

		mlx5e_activate_rq(xskrq);

		xskrq->stats->recover++;

	}



	mutex_unlock(&icosq->channel->icosq_recovery_lock);



	return 0;

out:

	clear_bit(MLX5E_SQ_STATE_RECOVERING, &icosq->state);

	mutex_unlock(&icosq->channel->icosq_recovery_lock);

	return err;

}
@@ -706,6 +729,16 @@ void mlx5e_reporter_icosq_cqe_err(struct mlx5e_icosq *icosq) 
	mlx5e_health_report(priv, priv->rx_reporter, err_str, &err_ctx);

}



void mlx5e_reporter_icosq_suspend_recovery(struct mlx5e_channel *c)

{

	mutex_lock(&c->icosq_recovery_lock);

}



void mlx5e_reporter_icosq_resume_recovery(struct mlx5e_channel *c)

{

	mutex_unlock(&c->icosq_recovery_lock);

}



static const struct devlink_health_reporter_ops mlx5_rx_reporter_ops = {

	.name = "rx",

	.recover = mlx5e_rx_reporter_recover,

drivers/net/ethernet/mellanox/mlx5/core/en/reporter_tx.c

+9 −1
Original line number Diff line number Diff line
@@ -466,6 +466,14 @@ static int mlx5e_tx_reporter_dump_sq(struct mlx5e_priv *priv, struct devlink_fms 
	return mlx5e_health_fmsg_named_obj_nest_end(fmsg);

}



static int mlx5e_tx_reporter_timeout_dump(struct mlx5e_priv *priv, struct devlink_fmsg *fmsg,

					  void *ctx)

{

	struct mlx5e_tx_timeout_ctx *to_ctx = ctx;



	return mlx5e_tx_reporter_dump_sq(priv, fmsg, to_ctx->sq);

}



static int mlx5e_tx_reporter_dump_all_sqs(struct mlx5e_priv *priv,

					  struct devlink_fmsg *fmsg)

{
@@ -561,7 +569,7 @@ int mlx5e_reporter_tx_timeout(struct mlx5e_txqsq *sq) 
	to_ctx.sq = sq;

	err_ctx.ctx = &to_ctx;

	err_ctx.recover = mlx5e_tx_reporter_timeout_recover;

	err_ctx.dump = mlx5e_tx_reporter_dump_sq;

	err_ctx.dump = mlx5e_tx_reporter_timeout_dump;

	snprintf(err_str, sizeof(err_str),

		 "TX timeout on queue: %d, SQ: 0x%x, CQ: 0x%x, SQ Cons: 0x%x SQ Prod: 0x%x, usecs since last trans: %u",

		 sq->ch_ix, sq->sqn, sq->cq.mcq.cqn, sq->cc, sq->pc,