[dpdk-dev] [PATCH 11/22] event/dlb2: add port setup

Eads, Gage gage.eads at intel.com
Wed Oct 7 22:34:56 CEST 2020



> -----Original Message-----
> From: McDaniel, Timothy <timothy.mcdaniel at intel.com>
> Sent: Friday, September 11, 2020 3:26 PM
> Cc: dev at dpdk.org; Carrillo, Erik G <erik.g.carrillo at intel.com>; Eads, Gage
> <gage.eads at intel.com>; Van Haaren, Harry <harry.van.haaren at intel.com>;
> jerinj at marvell.com
> Subject: [PATCH 11/22] event/dlb2: add port setup
> 
> Configure the load balanded (ldb) or directed (dir) port.
> The consumer queue (CQ) and producer port (PP) are also
> set up here.
> 
> Signed-off-by: Timothy McDaniel <timothy.mcdaniel at intel.com>
> ---
>  drivers/event/dlb2/dlb2.c                  | 527 +++++++++++++++++
>  drivers/event/dlb2/dlb2_iface.c            |   9 +
>  drivers/event/dlb2/dlb2_iface.h            |   8 +
>  drivers/event/dlb2/pf/base/dlb2_resource.c | 921
> +++++++++++++++++++++++++++++
>  drivers/event/dlb2/pf/dlb2_main.c          |  28 +
>  drivers/event/dlb2/pf/dlb2_pf.c            | 179 ++++++
>  6 files changed, 1672 insertions(+)
> 
> diff --git a/drivers/event/dlb2/dlb2.c b/drivers/event/dlb2/dlb2.c
> index 366e194..a4c8833 100644
> --- a/drivers/event/dlb2/dlb2.c
> +++ b/drivers/event/dlb2/dlb2.c
> @@ -1043,6 +1043,532 @@ dlb2_eventdev_queue_setup(struct rte_eventdev
> *dev,
>  	return ret;
>  }
> 
> +static int
> +dlb2_init_consume_qe(struct dlb2_port *qm_port, char *mz_name)
> +{
> +	struct dlb2_cq_pop_qe *qe;
> +
> +	qe = rte_malloc(mz_name,
> +			DLB2_NUM_QES_PER_CACHE_LINE *
> +				sizeof(struct dlb2_cq_pop_qe),
> +			RTE_CACHE_LINE_SIZE);
> +
> +	if (qe == NULL)	{
> +		DLB2_LOG_ERR("dlb2: no memory for consume_qe\n");
> +		return -ENOMEM;
> +	}
> +	qm_port->consume_qe = qe;
> +
> +	memset(qe, 0, DLB2_NUM_QES_PER_CACHE_LINE *
> +	       sizeof(struct dlb2_cq_pop_qe));

You can use rte_zmalloc() instead (applies to other init_*_qe functions below too), and
no need to zero-init the fields again after the memset.

> +
> +	qe->qe_valid = 0;
> +	qe->qe_frag = 0;
> +	qe->qe_comp = 0;
> +	qe->cq_token = 1;
> +	/* Tokens value is 0-based; i.e. '0' returns 1 token, '1' returns 2,
> +	 * and so on.
> +	 */
> +	qe->tokens = 0;	/* set at run time */
> +	qe->meas_lat = 0;
> +	qe->no_dec = 0;
> +	/* Completion IDs are disabled */
> +	qe->cmp_id = 0;
> +
> +	return 0;
> +}
> +
> +static int
> +dlb2_init_int_arm_qe(struct dlb2_port *qm_port, char *mz_name)
> +{
> +	struct dlb2_enqueue_qe *qe;
> +
> +	qe = rte_malloc(mz_name,
> +			DLB2_NUM_QES_PER_CACHE_LINE *
> +				sizeof(struct dlb2_enqueue_qe),
> +			RTE_CACHE_LINE_SIZE);
> +
> +	if (qe == NULL) {
> +		DLB2_LOG_ERR("dlb2: no memory for complete_qe\n");
> +		return -ENOMEM;
> +	}
> +	qm_port->int_arm_qe = qe;
> +
> +	memset(qe, 0, DLB2_NUM_QES_PER_CACHE_LINE *
> +	       sizeof(struct dlb2_enqueue_qe));
> +
> +	/* V2 - INT ARM is CQ_TOKEN + FRAG */
> +	qe->qe_valid = 0;
> +	qe->qe_frag = 1;
> +	qe->qe_comp = 0;
> +	qe->cq_token = 1;
> +	qe->meas_lat = 0;
> +	qe->no_dec = 0;
> +	/* Completion IDs are disabled */
> +	qe->cmp_id = 0;
> +
> +	return 0;
> +}
> +
> +static int
> +dlb2_init_qe_mem(struct dlb2_port *qm_port, char *mz_name)
> +{
> +	int ret, sz;
> +
> +	sz = DLB2_NUM_QES_PER_CACHE_LINE * sizeof(struct
> dlb2_enqueue_qe);
> +
> +	qm_port->qe4 = rte_malloc(mz_name, sz, RTE_CACHE_LINE_SIZE);
> +
> +	if (qm_port->qe4 == NULL) {
> +		DLB2_LOG_ERR("dlb2: no qe4 memory\n");
> +		ret = -ENOMEM;
> +		goto error_exit;
> +	}
> +
> +	memset(qm_port->qe4, 0, sz);
> +
> +	ret = dlb2_init_int_arm_qe(qm_port, mz_name);
> +	if (ret < 0) {
> +		DLB2_LOG_ERR("dlb2: dlb2_init_int_arm_qe ret=%d\n",
> +			     ret);

This can fit on one line

> +		goto error_exit;
> +	}
> +
> +	ret = dlb2_init_consume_qe(qm_port, mz_name);
> +	if (ret < 0) {
> +		DLB2_LOG_ERR("dlb2: dlb2_init_consume_qe ret=%d\n",
> +			     ret);

This can fit on one line

> +		goto error_exit;
> +	}
> +
> +	return 0;
> +
> +error_exit:
> +
> +	dlb2_free_qe_mem(qm_port);
> +
> +	return ret;
> +}
> +
> +static int
> +dlb2_hw_create_ldb_port(struct dlb2_eventdev *dlb2,
> +			struct dlb2_eventdev_port *ev_port,
> +			uint32_t dequeue_depth,
> +			uint32_t enqueue_depth)
> +{
> +	struct dlb2_hw_dev *handle = &dlb2->qm_instance;
> +	struct dlb2_create_ldb_port_args cfg = {0};
> +	int ret;
> +	struct dlb2_port *qm_port = NULL;
> +	char mz_name[RTE_MEMZONE_NAMESIZE];
> +	uint32_t qm_port_id;
> +	uint16_t ldb_credit_high_watermark;
> +	uint16_t dir_credit_high_watermark;
> +
> +	if (handle == NULL)
> +		return -EINVAL;
> +
> +	if (dequeue_depth < DLB2_MIN_CQ_DEPTH ||
> +	    dequeue_depth > DLB2_MAX_INPUT_QUEUE_DEPTH) {
> +		DLB2_LOG_ERR("dlb2: invalid dequeue_depth, must be %d-
> %d\n",
> +			     DLB2_MIN_CQ_DEPTH,
> DLB2_MAX_INPUT_QUEUE_DEPTH);
> +		return -EINVAL;
> +	}
> +
> +	if (enqueue_depth < DLB2_MIN_ENQUEUE_DEPTH) {
> +		DLB2_LOG_ERR("dlb2: invalid enqueue_depth, must be at least
> %d\n",
> +			     DLB2_MIN_ENQUEUE_DEPTH);
> +		return -EINVAL;
> +	}

The dequeue and enqueue depth checks are suspiciously inconsistent -- only the dequeue
depth is compared against an upper bound. I suspect the enqueue upper bound check is missing
because it's already checked in both dlb2_eventdev_port_setup() and
rte_event_port_setup()...if that's the case, can the dequeue depth max check be dropped as well?

> +
> +	rte_spinlock_lock(&handle->resource_lock);
> +
> +	/* TODO - additional parameter validation */

Leftover TODO

> +	/* We round up to the next power of 2 if necessary */
> +	cfg.cq_depth = rte_align32pow2(dequeue_depth);
> +	cfg.cq_depth_threshold = 1;
> +
> +	cfg.cq_history_list_size =
> DLB2_NUM_HIST_LIST_ENTRIES_PER_LDB_PORT;
> +
> +	if (handle->cos_id == DLB2_COS_DEFAULT)
> +		cfg.cos_id = 0;
> +	else
> +		cfg.cos_id = handle->cos_id;
> +
> +	cfg.cos_strict = 0;
> +
> +	/* User controls the LDB high watermark via enqueue depth. The DIR
> high
> +	 * watermark is equal, unless the directed credit pool is too small.
> +	 */
> +	ldb_credit_high_watermark = enqueue_depth;
> +
> +	/* If there are no directed ports, the kernel driver will ignore this
> +	 * port's directed credit settings. Don't use enqueue_depth if it would
> +	 * require more directed credits than are available.
> +	 */
> +	dir_credit_high_watermark =
> +		RTE_MIN(enqueue_depth,
> +			handle->cfg.num_dir_credits / dlb2->num_ports);
> +
> +	/* Per QM values */
> +
> +	/* DEBUG
> +	 * DLB2_LOG_ERR("create ldb port - grp=%d, devId=%d\n",
> +	 * handle->cfg.domain_id, handle->device_id);
> +	 */

Leftover debug code

> +
> +	ret = dlb2_iface_ldb_port_create(handle, &cfg,  dlb2->poll_mode);
> +	if (ret < 0) {
> +		DLB2_LOG_ERR("dlb2: dlb2_ldb_port_create error, ret=%d
> (driver status: %s)\n",
> +			     ret, dlb2_error_strings[cfg.response.status]);
> +		goto error_exit;
> +	}
> +
> +	qm_port_id = cfg.response.id;
> +
> +	DLB2_LOG_DBG("dlb2: ev_port %d uses qm LB port %d <<<<<\n",
> +		     ev_port->id, qm_port_id);
> +
> +	qm_port = &ev_port->qm_port;
> +	qm_port->ev_port = ev_port; /* back ptr */
> +	qm_port->dlb2 = dlb2; /* back ptr */
> +	/*
> +	 * Allocate and init local qe struct(s).
> +	 * Note: MOVDIR64 requires the enqueue QE (qe4) to be aligned.
> +	 */
> +
> +	snprintf(mz_name, sizeof(mz_name), "%s_ldb_port%d",
> +		 handle->device_name,
> +		 ev_port->id);
> +

I see handle->device_name getting read here, in dlb2_hw_create_dir_port(),
and also in dlb2_eventdev_dump(), but I don't see it written anywhere?

> +	ret = dlb2_init_qe_mem(qm_port, mz_name);
> +	if (ret < 0) {
> +		DLB2_LOG_ERR("dlb2: init_qe_mem failed, ret=%d\n", ret);
> +		goto error_exit;
> +	}
> +
> +	qm_port->id = qm_port_id;
> +
> +	qm_port->cached_ldb_credits = 0;
> +	qm_port->cached_dir_credits = 0;
> +	/* CQs with depth < 8 use an 8-entry queue, but withhold credits so
> +	 * the effective depth is smaller.
> +	 */
> +	qm_port->cq_depth = cfg.cq_depth <= 8 ? 8 : cfg.cq_depth;
> +	qm_port->cq_idx = 0;
> +	qm_port->cq_idx_unmasked = 0;
> +
> +	if (dlb2->poll_mode == DLB2_CQ_POLL_MODE_SPARSE)
> +		qm_port->cq_depth_mask = (qm_port->cq_depth * 4) - 1;
> +	else
> +		qm_port->cq_depth_mask = qm_port->cq_depth - 1;
> +
> +	qm_port->gen_bit_shift = __builtin_popcount(qm_port-
> >cq_depth_mask);
> +	/* starting value of gen bit - it toggles at wrap time */
> +	qm_port->gen_bit = 1;
> +
> +	qm_port->int_armed = false;
> +
> +	/* Save off for later use in info and lookup APIs. */
> +	qm_port->qid_mappings = &dlb2->qm_ldb_to_ev_queue_id[0];
> +
> +	qm_port->dequeue_depth = dequeue_depth;
> +	qm_port->token_pop_thresh = dequeue_depth;
> +
> +	qm_port->owed_tokens = 0;
> +	qm_port->issued_releases = 0;
> +
> +	/* Save config message too. */
> +	rte_memcpy(&qm_port->cfg.ldb, &cfg, sizeof(cfg));

I know qm_port->cfg.ldb and cfg are the same type, but just for safety in case that
ever changes in the future...probably better to use sizeof() on the destination rather
than the source.

> +
> +	/* update state */
> +	qm_port->state = PORT_STARTED; /* enabled at create time */
> +	qm_port->config_state = DLB2_CONFIGURED;
> +
> +	qm_port->dir_credits = dir_credit_high_watermark;
> +	qm_port->ldb_credits = ldb_credit_high_watermark;
> +	qm_port->credit_pool[DLB2_DIR_QUEUE] = &dlb2->dir_credit_pool;
> +	qm_port->credit_pool[DLB2_LDB_QUEUE] = &dlb2->ldb_credit_pool;
> +
> +	DLB2_LOG_DBG("dlb2: created ldb port %d, depth = %d, ldb credits=%d,
> dir credits=%d\n",
> +		     qm_port_id,
> +		     dequeue_depth,
> +		     qm_port->ldb_credits,
> +		     qm_port->dir_credits);
> +
> +	rte_spinlock_unlock(&handle->resource_lock);
> +
> +	return 0;
> +
> +error_exit:
> +
> +	if (qm_port)
> +		dlb2_free_qe_mem(qm_port);
> +
> +	rte_spinlock_unlock(&handle->resource_lock);
> +
> +	DLB2_LOG_ERR("dlb2: create ldb port failed!\n");
> +
> +	return ret;
> +}
> +
> +static void
> +dlb2_port_link_teardown(struct dlb2_eventdev *dlb2,
> +			struct dlb2_eventdev_port *ev_port)
> +{
> +	struct dlb2_eventdev_queue *ev_queue;
> +	int i;
> +
> +	for (i = 0; i < DLB2_MAX_NUM_QIDS_PER_LDB_CQ; i++) {
> +		if (!ev_port->link[i].valid)
> +			continue;
> +
> +		ev_queue = &dlb2->ev_queues[ev_port->link[i].queue_id];
> +
> +		ev_port->link[i].valid = false;
> +		ev_port->num_links--;
> +		ev_queue->num_links--;
> +	}
> +}
> +
> +static int
> +dlb2_hw_create_dir_port(struct dlb2_eventdev *dlb2,
> +			struct dlb2_eventdev_port *ev_port,
> +			uint32_t dequeue_depth,
> +			uint32_t enqueue_depth)
> +{
> +	struct dlb2_hw_dev *handle = &dlb2->qm_instance;
> +	struct dlb2_create_dir_port_args cfg = {0};
> +	int ret;
> +	struct dlb2_port *qm_port = NULL;
> +	char mz_name[RTE_MEMZONE_NAMESIZE];
> +	uint32_t qm_port_id;
> +	uint16_t ldb_credit_high_watermark;
> +	uint16_t dir_credit_high_watermark;
> +
> +	if (dlb2 == NULL || handle == NULL)
> +		return -EINVAL;
> +
> +	if (dequeue_depth < DLB2_MIN_CQ_DEPTH ||
> +	    dequeue_depth > DLB2_MAX_INPUT_QUEUE_DEPTH) {
> +		DLB2_LOG_ERR("dlb2: invalid dequeue_depth, must be %d-
> %d\n",
> +			     DLB2_MIN_CQ_DEPTH,
> DLB2_MAX_INPUT_QUEUE_DEPTH);
> +		return -EINVAL;
> +	}

Enqueue depth check needed?

> +
> +	rte_spinlock_lock(&handle->resource_lock);
> +
> +	/* Directed queues are configured at link time. */
> +	cfg.queue_id = -1;
> +
> +	/* We round up to the next power of 2 if necessary */
> +	cfg.cq_depth = rte_align32pow2(dequeue_depth);
> +	cfg.cq_depth_threshold = 1;
> +
> +	/* User controls the LDB high watermark via enqueue depth. The DIR
> high
> +	 * watermark is equal, unless the directed credit pool is too small.
> +	 */
> +	ldb_credit_high_watermark = enqueue_depth;
> +
> +	/* Don't use enqueue_depth if it would require more directed credits
> +	 * than are available.
> +	 */
> +	dir_credit_high_watermark =
> +		RTE_MIN(enqueue_depth,
> +			handle->cfg.num_dir_credits / dlb2->num_ports);
> +
> +	/* Per QM values */
> +
> +	ret = dlb2_iface_dir_port_create(handle, &cfg,  dlb2->poll_mode);
> +	if (ret < 0) {
> +		DLB2_LOG_ERR("dlb2: dlb2_dir_port_create error, ret=%d (driver
> status: %s)\n",
> +			     ret, dlb2_error_strings[cfg.response.status]);
> +		goto error_exit;
> +	}
> +
> +	qm_port_id = cfg.response.id;
> +
> +	DLB2_LOG_DBG("dlb2: ev_port %d uses qm DIR port %d <<<<<\n",
> +		     ev_port->id, qm_port_id);
> +
> +	qm_port = &ev_port->qm_port;
> +	qm_port->ev_port = ev_port; /* back ptr */
> +	qm_port->dlb2 = dlb2;  /* back ptr */
> +
> +	/*
> +	 * Init local qe struct(s).
> +	 * Note: MOVDIR64 requires the enqueue QE to be aligned
> +	 */
> +
> +	snprintf(mz_name, sizeof(mz_name), "%s_dir_port%d",
> +		 handle->device_name,

(See device_name comment above)

> +		 ev_port->id);
> +
> +	ret = dlb2_init_qe_mem(qm_port, mz_name);
> +
> +	if (ret < 0) {
> +		DLB2_LOG_ERR("dlb2: init_qe_mem failed, ret=%d\n", ret);
> +		goto error_exit;
> +	}
> +
> +	qm_port->id = qm_port_id;
> +
> +	qm_port->cached_ldb_credits = 0;
> +	qm_port->cached_dir_credits = 0;
> +	/* CQs with depth < 8 use an 8-entry queue, but withhold credits so
> +	 * the effective depth is smaller.
> +	 */
> +	qm_port->cq_depth = cfg.cq_depth <= 8 ? 8 : cfg.cq_depth;
> +	qm_port->cq_idx = 0;
> +	qm_port->cq_idx_unmasked = 0;
> +
> +	if (dlb2->poll_mode == DLB2_CQ_POLL_MODE_SPARSE)
> +		qm_port->cq_depth_mask = (cfg.cq_depth * 4) - 1;
> +	else
> +		qm_port->cq_depth_mask = cfg.cq_depth - 1;
> +
> +	qm_port->gen_bit_shift = __builtin_popcount(qm_port-
> >cq_depth_mask);
> +	/* starting value of gen bit - it toggles at wrap time */
> +	qm_port->gen_bit = 1;
> +
> +	qm_port->int_armed = false;
> +
> +	/* Save off for later use in info and lookup APIs. */
> +	qm_port->qid_mappings = &dlb2->qm_dir_to_ev_queue_id[0];
> +
> +	qm_port->dequeue_depth = dequeue_depth;
> +
> +	/* Directed ports are auto-pop, by default. */
> +	qm_port->token_pop_mode = AUTO_POP;
> +	qm_port->owed_tokens = 0;
> +	qm_port->issued_releases = 0;
> +
> +	/* Save config message too. */
> +	rte_memcpy(&qm_port->cfg.dir, &cfg, sizeof(cfg));

(See sizeof() comment above)

> +
> +	/* update state */
> +	qm_port->state = PORT_STARTED; /* enabled at create time */
> +	qm_port->config_state = DLB2_CONFIGURED;
> +
> +	qm_port->dir_credits = dir_credit_high_watermark;
> +	qm_port->ldb_credits = ldb_credit_high_watermark;
> +	qm_port->credit_pool[DLB2_DIR_QUEUE] = &dlb2->dir_credit_pool;
> +	qm_port->credit_pool[DLB2_LDB_QUEUE] = &dlb2->ldb_credit_pool;
> +
> +	DLB2_LOG_DBG("dlb2: created dir port %d, depth = %d cr=%d,%d\n",
> +		     qm_port_id,
> +		     dequeue_depth,
> +		     dir_credit_high_watermark,
> +		     ldb_credit_high_watermark);
> +
> +	rte_spinlock_unlock(&handle->resource_lock);
> +
> +	return 0;
> +
> +error_exit:
> +
> +	if (qm_port)
> +		dlb2_free_qe_mem(qm_port);
> +
> +	rte_spinlock_unlock(&handle->resource_lock);
> +
> +	DLB2_LOG_ERR("dlb2: create dir port failed!\n");
> +
> +	return ret;
> +}
> +
> +static int
> +dlb2_eventdev_port_setup(struct rte_eventdev *dev,
> +			 uint8_t ev_port_id,
> +			 const struct rte_event_port_conf *port_conf)
> +{
> +	struct dlb2_eventdev *dlb2;
> +	struct dlb2_eventdev_port *ev_port;
> +	int ret;
> +
> +	if (dev == NULL || port_conf == NULL) {
> +		DLB2_LOG_ERR("Null parameter\n");
> +		return -EINVAL;
> +	}
> +
> +	dlb2 = dlb2_pmd_priv(dev);
> +
> +	if (ev_port_id >= DLB2_MAX_NUM_PORTS)
> +		return -EINVAL;
> +
> +	if (port_conf->dequeue_depth >
> +		evdev_dlb2_default_info.max_event_port_dequeue_depth ||
> +	    port_conf->enqueue_depth >
> +		evdev_dlb2_default_info.max_event_port_enqueue_depth)
> +		return -EINVAL;
> +
> +	ev_port = &dlb2->ev_ports[ev_port_id];
> +	/* configured? */
> +	if (ev_port->setup_done) {
> +		DLB2_LOG_ERR("evport %d is already configured\n",
> ev_port_id);
> +		return -EINVAL;
> +	}
> +
> +	/* The reserved token interrupt arming scheme requires that one or
> more
> +	 * CQ tokens be reserved by the PMD. This limits the amount of CQ space
> +	 * usable by the DLB, so in order to give an *effective* CQ depth equal
> +	 * to the user-requested value, we double CQ depth and reserve half of
> +	 * its tokens. If the user requests the max CQ depth (256) then we
> +	 * cannot double it, so we reserve one token and give an effective
> +	 * depth of 255 entries.
> +	 */

I don't think this comment applies to the 2.0 device.

> +
> +	ev_port->qm_port.is_directed = port_conf->event_port_cfg &
> +		RTE_EVENT_PORT_CFG_SINGLE_LINK;
> +
> +	if (!ev_port->qm_port.is_directed) {
> +		ret = dlb2_hw_create_ldb_port(dlb2,
> +					      ev_port,
> +					      port_conf->dequeue_depth,
> +					      port_conf->enqueue_depth);
> +		if (ret < 0) {
> +			DLB2_LOG_ERR("Failed to create the lB port ve
> portId=%d\n",
> +				     ev_port_id);
> +
> +			return ret;
> +		}
> +	} else {
> +		ret = dlb2_hw_create_dir_port(dlb2,
> +					      ev_port,
> +					      port_conf->dequeue_depth,
> +					      port_conf->enqueue_depth);
> +		if (ret < 0) {
> +			DLB2_LOG_ERR("Failed to create the DIR port\n");
> +			return ret;
> +		}
> +	}
> +
> +	/* Save off port config for reconfig */
> +	dlb2->ev_ports[ev_port_id].conf = *port_conf;

Nit: 'ev_port' is assigned to &dlb2->ev_ports[ev_port_id] above, use it here and below?

> +
> +	dlb2->ev_ports[ev_port_id].id = ev_port_id;
> +	dlb2->ev_ports[ev_port_id].enq_configured = true;
> +	dlb2->ev_ports[ev_port_id].setup_done = true;
> +	dlb2->ev_ports[ev_port_id].inflight_max =
> +		port_conf->new_event_threshold;
> +	dlb2->ev_ports[ev_port_id].implicit_release =
> +		!(port_conf->event_port_cfg &
> +		  RTE_EVENT_PORT_CFG_DISABLE_IMPL_REL);
> +	dlb2->ev_ports[ev_port_id].outstanding_releases = 0;
> +	dlb2->ev_ports[ev_port_id].inflight_credits = 0;
> +	dlb2->ev_ports[ev_port_id].credit_update_quanta =
> +		RTE_LIBRTE_PMD_DLB2_SW_CREDIT_QUANTA;
> +	dlb2->ev_ports[ev_port_id].dlb2 = dlb2; /* reverse link */
> +
> +	/* Tear down pre-existing port->queue links */
> +	if (dlb2->run_state == DLB2_RUN_STATE_STOPPED)
> +		dlb2_port_link_teardown(dlb2, &dlb2->ev_ports[ev_port_id]);
> +
> +	dev->data->ports[ev_port_id] = &dlb2->ev_ports[ev_port_id];
> +
> +	return 0;
> +}

[...]

> diff --git a/drivers/event/dlb2/pf/dlb2_pf.c b/drivers/event/dlb2/pf/dlb2_pf.c
> index dea70e6..a6824b1 100644
> --- a/drivers/event/dlb2/pf/dlb2_pf.c
> +++ b/drivers/event/dlb2/pf/dlb2_pf.c
> @@ -234,6 +234,183 @@ dlb2_pf_set_sn_allocation(struct dlb2_hw_dev
> *handle,
>  	return ret;
>  }
> 
> +static void *
> +dlb2_alloc_coherent_aligned(uintptr_t *phys, size_t size, int align)
> +{
> +	const struct rte_memzone *mz;
> +	char mz_name[RTE_MEMZONE_NAMESIZE];
> +	uint32_t core_id = rte_lcore_id();
> +	unsigned int socket_id;
> +
> +	snprintf(mz_name, sizeof(mz_name) - 1, "%lx",
> +		 (unsigned long)rte_get_timer_cycles());

For debug purposes, it would be better if this name can trace the mz back to
this driver. How about something like event_dlb2_pf_name + ldb/dir_port + port ID?

> +	if (core_id == (unsigned int)LCORE_ID_ANY)
> +		core_id = rte_get_master_lcore();
> +	socket_id = rte_lcore_to_socket_id(core_id);

Should this use the socket ID devarg (and perhaps fall back to the core's socket if unspecified)?

> +	mz = rte_memzone_reserve_aligned(mz_name, size, socket_id,
> +					 RTE_MEMZONE_IOVA_CONTIG, align);
> +	if (!mz) {
> +		DLB2_LOG_DBG("Unable to allocate DMA memory of size %zu
> bytes - %s\n",
> +			     size, rte_strerror(rte_errno));
> +		*phys = 0;
> +		return NULL;
> +	}
> +	*phys = mz->iova;
> +	return mz->addr;
> +}
> +
> +static int
> +dlb2_pf_ldb_port_create(struct dlb2_hw_dev *handle,
> +			struct dlb2_create_ldb_port_args *cfg,
> +			enum dlb2_cq_poll_modes poll_mode)
> +{
> +	struct dlb2_dev *dlb2_dev = (struct dlb2_dev *)handle->pf_dev;
> +	struct dlb2_cmd_response response = {0};
> +	struct dlb2_port_memory port_memory;
> +	int ret, cq_alloc_depth;
> +	uint8_t *port_base;
> +	int alloc_sz, qe_sz;
> +	phys_addr_t cq_base;
> +	phys_addr_t pp_base;
> +	int is_dir = false;
> +
> +	DLB2_INFO(dev->dlb2_device, "Entering %s()\n", __func__);
> +
> +	if (poll_mode == DLB2_CQ_POLL_MODE_STD)
> +		qe_sz = sizeof(struct dlb2_dequeue_qe);
> +	else
> +		qe_sz = RTE_CACHE_LINE_SIZE;
> +
> +	/* Calculate the port memory required, and round up to the nearest
> +	 * cache line.
> +	 */
> +	cq_alloc_depth = RTE_MAX(cfg->cq_depth,
> DLB2_MIN_HARDWARE_CQ_DEPTH);
> +	alloc_sz = cq_alloc_depth * qe_sz;
> +	alloc_sz = RTE_CACHE_LINE_ROUNDUP(alloc_sz);
> +
> +	port_base = dlb2_alloc_coherent_aligned(&cq_base,
> +						alloc_sz,
> +						PAGE_SIZE);

This can fit on one line

> +	if (port_base == NULL)
> +		return -ENOMEM;
> +
> +	/* Lock the page in memory */
> +	ret = rte_mem_lock_page(port_base);
> +	if (ret < 0) {
> +		DLB2_LOG_ERR("dlb2 pf pmd could not lock page for device
> i/o\n");
> +		goto create_port_err;
> +	}
> +
> +
> +	memset(port_base, 0, alloc_sz);
> +
> +	ret = dlb2_pf_create_ldb_port(&dlb2_dev->hw,
> +				      handle->domain_id,
> +				      cfg,
> +				      cq_base,
> +				      &response);
> +	if (ret)
> +		goto create_port_err;
> +
> +	pp_base = (uintptr_t)dlb2_dev->hw.func_kva + PP_BASE(is_dir);
> +	dlb2_port[response.id][DLB2_LDB_PORT].pp_addr =
> +		(void *)(uintptr_t)(pp_base + (PAGE_SIZE * response.id));

If pp_base is defined as a uintptr_t, I think you can avoid some of the explicit
casts.

> +
> +	dlb2_port[response.id][DLB2_LDB_PORT].cq_base =
> +		(void *)(uintptr_t)(port_base);

Since port_base is a uint8_t*, the uintptr_t cast shouldn't be necessary

> +	memset(&port_memory, 0, sizeof(port_memory));
> +	dlb2_list_init_head(&port_memory.list);
> +
> +	/* Fill out the per-port memory tracking structure */
> +	dlb2_dev->ldb_port_pages[response.id].valid = true;
> +	dlb2_list_splice(&port_memory.list,
> +			 &dlb2_dev->ldb_port_pages[response.id].list);

Does the list serve any purpose? Looks like port_memory is zero-initialized, then it becomes
the sole entry on a per-port list.

> +
> +	cfg->response = response;
> +
> +	DLB2_INFO(dev->dlb2_device, "Exiting %s() with ret=%d\n",
> +		  __func__, ret);
> +
> +create_port_err:

Need to free the memzone in this case.

> +
> +	return ret;
> +}
> +
> +static int
> +dlb2_pf_dir_port_create(struct dlb2_hw_dev *handle,
> +			struct dlb2_create_dir_port_args *cfg,
> +			enum dlb2_cq_poll_modes poll_mode)
> +{
> +	struct dlb2_dev *dlb2_dev = (struct dlb2_dev *)handle->pf_dev;
> +	struct dlb2_cmd_response response = {0};
> +	struct dlb2_port_memory port_memory;
> +	int ret;
> +	uint8_t *port_base;
> +	int alloc_sz, qe_sz;
> +	phys_addr_t cq_base;
> +	phys_addr_t pp_base;
> +	int is_dir = true;
> +
> +	DLB2_INFO(dev->dlb2_device, "Entering %s()\n", __func__);
> +
> +	if (poll_mode == DLB2_CQ_POLL_MODE_STD)
> +		qe_sz = sizeof(struct dlb2_dequeue_qe);
> +	else
> +		qe_sz = RTE_CACHE_LINE_SIZE;
> +
> +	/* Calculate the port memory required, and round up to the nearest
> +	 * cache line.
> +	 */
> +	alloc_sz = cfg->cq_depth * qe_sz;
> +	alloc_sz = RTE_CACHE_LINE_ROUNDUP(alloc_sz);
> +
> +	port_base = dlb2_alloc_coherent_aligned(&cq_base,
> +						alloc_sz,
> +						PAGE_SIZE);

This can fit on one line

> +	if (port_base == NULL)
> +		return -ENOMEM;
> +
> +	/* Lock the page in memory */
> +	ret = rte_mem_lock_page(port_base);
> +	if (ret < 0) {
> +		DLB2_LOG_ERR("dlb2 pf pmd could not lock page for device
> i/o\n");
> +		goto create_port_err;
> +	}
> +
> +	memset(port_base, 0, alloc_sz);
> +
> +	ret = dlb2_pf_create_dir_port(&dlb2_dev->hw,
> +				      handle->domain_id,
> +				      cfg,
> +				      cq_base,
> +				      &response);
> +	if (ret)
> +		goto create_port_err;
> +
> +	pp_base = (uintptr_t)dlb2_dev->hw.func_kva + PP_BASE(is_dir);

(See uintptr_t comment above)

> +	dlb2_port[response.id][DLB2_DIR_PORT].pp_addr =
> +		(void *)(uintptr_t)(pp_base + (PAGE_SIZE * response.id));
> +
> +	dlb2_port[response.id][DLB2_DIR_PORT].cq_base =
> +		(void *)(uintptr_t)(port_base);

(See port_base comment above)

> +	memset(&port_memory, 0, sizeof(port_memory));
> +	dlb2_list_init_head(&port_memory.list);
> +
> +	/* Fill out the per-port memory tracking structure */
> +	dlb2_dev->dir_port_pages[response.id].valid = true;
> +	dlb2_list_splice(&port_memory.list,
> +			 &dlb2_dev->dir_port_pages[response.id].list);
> +

(See list comment above)

> +	cfg->response = response;
> +
> +	DLB2_INFO(dev->dlb2_device, "Exiting %s() with ret=%d\n",
> +		  __func__, ret);
> +
> +create_port_err:

Need to free the memzone in this case

> +
> +	return ret;
> +}
> +
>  static void
>  dlb2_pf_iface_fn_ptrs_init(void)
>  {
> @@ -247,6 +424,8 @@ dlb2_pf_iface_fn_ptrs_init(void)
>  	dlb2_iface_get_cq_poll_mode = dlb2_pf_get_cq_poll_mode;
>  	dlb2_iface_sched_domain_create = dlb2_pf_sched_domain_create;
>  	dlb2_iface_ldb_queue_create = dlb2_pf_ldb_queue_create;
> +	dlb2_iface_ldb_port_create = dlb2_pf_ldb_port_create;
> +	dlb2_iface_dir_port_create = dlb2_pf_dir_port_create;
>  	dlb2_iface_get_sn_allocation = dlb2_pf_get_sn_allocation;
>  	dlb2_iface_set_sn_allocation = dlb2_pf_set_sn_allocation;
>  	dlb2_iface_get_sn_occupancy = dlb2_pf_get_sn_occupancy;
> --
> 2.6.4

I don't see the port memzones getting freed anywhere, e.g. if the event device is reset.
Looks like a possible memory leak.

Thanks,
Gage


More information about the dev mailing list