From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail-ig0-f174.google.com (mail-ig0-f174.google.com [209.85.213.174]) by dpdk.org (Postfix) with ESMTP id D371E9215 for ; Fri, 4 Dec 2015 19:36:28 +0100 (CET) Received: by igcph11 with SMTP id ph11so38647579igc.1 for ; Fri, 04 Dec 2015 10:36:28 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=semihalf-com.20150623.gappssmtp.com; s=20150623; h=mime-version:in-reply-to:references:from:date:message-id:subject:to :cc:content-type; bh=GmLv9+KGAYJ7toW5tau/u9R4B4kv33BfU+tMenoeu0Q=; b=oPGUjM+voIM0Zl8S5UodLdK4vlJAE9AWrjr506dtaUqy1tzzc5lDrO/LVBsDXw1Yi6 IktVBl2bskxV0g7W3i9yr0rIob4i456no0EpWpgBze8AgdTeBwCOngF3CFhlYdS21vBf J2ShRDX/JS8GKkcFRq+0HSFW6CaSlAXxJw45L77pb18l7hmAdIEts95EvsKDGn6Oic0O d8Tu7Zh2dNu91AHQ8R5WhNif1D1anr0t18omYV31NjucPvUeDR2e08BpCxXCQaWwFBMz JfMud6SdLc2Mpdmc5Y+LimqoRI6zq/2F+k5eR8DGU90LDZygo8iuQ27PcjwdYpIGffmk oh9Q== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=x-gm-message-state:mime-version:in-reply-to:references:from:date :message-id:subject:to:cc:content-type; bh=GmLv9+KGAYJ7toW5tau/u9R4B4kv33BfU+tMenoeu0Q=; b=hmPk2uhOJLf8rId/gEydV3cmDAl0AOKDiEB1kLQpVWio1Ll4s1023tDi3ZikfoNMQa mY8M8iXLnk8W0sirTwq5lBiL5X8w6ZYRmXzarqEKtCcgb5gRqIN/Bre83WpwqZlsIddn rezbSKtqWb5Gn4KgzD/2wVnmj5/Tot05yIu1NEKkLPzKJO6rVPZqgNaQH7gBCHfJdoEt 7DJem7Ck/xHx6e+sE/ofj205kaBuxKh1GtqPeka1d+p5jMJ6zL9uozi4HxHNUfVFJiE0 h8ULXcCd6LR/9a+fRXZfgJA3soEPKvFO0FOUB0b1TP4CMCoeYfDJ+PVoeb8XeguI+Rp0 Dhvw== X-Gm-Message-State: ALoCoQlzk4/Umlh3busDPkikwwNgvohlZXMmHgc7JLM7qdM/PwAlHgO1dVd8qy647I5A9vR/QkDlkk5Xrq0w3e8UgBaMBmNRhQ== X-Received: by 10.50.83.38 with SMTP id n6mr5769551igy.92.1449254188405; Fri, 04 Dec 2015 10:36:28 -0800 (PST) MIME-Version: 1.0 Received: by 10.64.11.225 with HTTP; Fri, 4 Dec 2015 10:36:09 -0800 (PST) In-Reply-To: <1449249260-15165-7-git-send-email-stephen@networkplumber.org> References: <1449249260-15165-1-git-send-email-stephen@networkplumber.org> <1449249260-15165-7-git-send-email-stephen@networkplumber.org> From: Andriy Berestovskyy Date: Fri, 4 Dec 2015 19:36:09 +0100 Message-ID: To: Stephen Hemminger Content-Type: text/plain; charset=UTF-8 Cc: dev@dpdk.org, Eric Kinzie Subject: Re: [dpdk-dev] [PATCH 6/8] bond: handle slaves with fewer queues than bonding device X-BeenThere: dev@dpdk.org X-Mailman-Version: 2.1.15 Precedence: list List-Id: patches and discussions about DPDK List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 04 Dec 2015 18:36:29 -0000 Hi guys, I'm not quite sure if we can support less TX queues on a slave that easy: > queue_id = bond_slave_txqid(internals, i, bd_tx_q->queue_id); > num_tx_slave = rte_eth_tx_burst(slaves[i], queue_id, > slave_bufs[i], slave_nb_pkts[i]); It seems that two different lcores might end up writing to the same slave queue at the same time, isn't it? Regards, Andriy On Fri, Dec 4, 2015 at 6:14 PM, Stephen Hemminger wrote: > From: Eric Kinzie > > In the event that the bonding device has a greater number of tx and/or rx > queues than the slave being added, track the queue limits of the slave. > On receive, ignore queue identifiers beyond what the slave interface > can support. During transmit, pick a different queue id to use if the > intended queue is not available on the slave. > > Signed-off-by: Eric Kinzie > Signed-off-by: Stephen Hemminger > --- > drivers/net/bonding/rte_eth_bond_api.c | 6 +- > drivers/net/bonding/rte_eth_bond_pmd.c | 141 +++++++++++++++++++++++++---- > drivers/net/bonding/rte_eth_bond_private.h | 5 +- > 3 files changed, 129 insertions(+), 23 deletions(-) > > diff --git a/drivers/net/bonding/rte_eth_bond_api.c b/drivers/net/bonding/rte_eth_bond_api.c > index 630a461..64058ff 100644 > --- a/drivers/net/bonding/rte_eth_bond_api.c > +++ b/drivers/net/bonding/rte_eth_bond_api.c > @@ -340,11 +340,11 @@ __eth_bond_slave_add_lock_free(uint8_t bonded_port_id, uint8_t slave_port_id) > > slave_eth_dev = &rte_eth_devices[slave_port_id]; > > - /* Add slave details to bonded device */ > - slave_add(internals, slave_eth_dev); > - > rte_eth_dev_info_get(slave_port_id, &dev_info); > > + /* Add slave details to bonded device */ > + slave_add(internals, slave_eth_dev, &dev_info); > + > /* We need to store slaves reta_size to be able to synchronize RETA for all > * slave devices even if its sizes are different. > */ > diff --git a/drivers/net/bonding/rte_eth_bond_pmd.c b/drivers/net/bonding/rte_eth_bond_pmd.c > index 77582dd..868e66b 100644 > --- a/drivers/net/bonding/rte_eth_bond_pmd.c > +++ b/drivers/net/bonding/rte_eth_bond_pmd.c > @@ -76,6 +76,47 @@ get_vlan_offset(struct ether_hdr *eth_hdr, uint16_t *proto) > return vlan_offset; > } > > +static uint8_t > +bond_active_slaves_by_rxqid(struct bond_dev_private *internals, int queue_id, > + uint8_t slaves[]) > +{ > + struct bond_slave_details *slave_details; > + uint8_t num_of_slaves; > + uint8_t i = 0; > + > + num_of_slaves = internals->active_slave_count; > + memcpy(slaves, internals->active_slaves, > + sizeof(internals->active_slaves[0]) * num_of_slaves); > + > + if (num_of_slaves < 1 || internals->kvlist) > + return num_of_slaves; > + > + /* remove slaves that don't have a queue numbered "queue_id" */ > + while (i < num_of_slaves) { > + slave_details = &internals->slaves[i]; > + if (unlikely(queue_id >= slave_details->nb_rx_queues)) { > + slaves[i] = slaves[num_of_slaves-1]; > + num_of_slaves--; > + } else > + i++; > + } > + > + return num_of_slaves; > +} > + > +static int > +bond_slave_txqid(struct bond_dev_private *internals, uint8_t slave_id, > + int queue_id) > +{ > + struct bond_slave_details *slave_details; > + > + if (internals->kvlist) > + return queue_id; > + > + slave_details = &internals->slaves[slave_id]; > + return queue_id % slave_details->nb_tx_queues; > +} > + > static uint16_t > bond_ethdev_rx_burst(void *queue, struct rte_mbuf **bufs, uint16_t nb_pkts) > { > @@ -83,6 +124,8 @@ bond_ethdev_rx_burst(void *queue, struct rte_mbuf **bufs, uint16_t nb_pkts) > > uint16_t num_rx_slave = 0; > uint16_t num_rx_total = 0; > + uint8_t slaves[RTE_MAX_ETHPORTS]; > + uint8_t num_of_slaves; > > int i; > > @@ -91,11 +134,13 @@ bond_ethdev_rx_burst(void *queue, struct rte_mbuf **bufs, uint16_t nb_pkts) > > internals = bd_rx_q->dev_private; > > + num_of_slaves = bond_active_slaves_by_rxqid(internals, bd_rx_q->queue_id, > + slaves); > > - for (i = 0; i < internals->active_slave_count && nb_pkts; i++) { > + for (i = 0; i < num_of_slaves && nb_pkts; i++) { > /* Offset of pointer to *bufs increases as packets are received > * from other slaves */ > - num_rx_slave = rte_eth_rx_burst(internals->active_slaves[i], > + num_rx_slave = rte_eth_rx_burst(slaves[i], > bd_rx_q->queue_id, bufs + num_rx_total, nb_pkts); > if (num_rx_slave) { > num_rx_total += num_rx_slave; > @@ -117,8 +162,13 @@ bond_ethdev_rx_burst_active_backup(void *queue, struct rte_mbuf **bufs, > > internals = bd_rx_q->dev_private; > > - return rte_eth_rx_burst(internals->current_primary_port, > - bd_rx_q->queue_id, bufs, nb_pkts); > + uint8_t active_slave = internals->current_primary_port; > + struct rte_eth_dev *dev = &rte_eth_devices[active_slave]; > + > + if (bd_rx_q->queue_id >= dev->data->nb_rx_queues) > + return 0; > + > + return rte_eth_rx_burst(active_slave, bd_rx_q->queue_id, bufs, nb_pkts); > } > > static uint16_t > @@ -144,9 +194,9 @@ bond_ethdev_rx_burst_8023ad(void *queue, struct rte_mbuf **bufs, > rte_eth_macaddr_get(internals->port_id, &bond_mac); > /* Copy slave list to protect against slave up/down changes during tx > * bursting */ > - slave_count = internals->active_slave_count; > - memcpy(slaves, internals->active_slaves, > - sizeof(internals->active_slaves[0]) * slave_count); > + > + slave_count = bond_active_slaves_by_rxqid(internals, bd_rx_q->queue_id, > + slaves); > > for (i = 0; i < slave_count && num_rx_total < nb_pkts; i++) { > j = num_rx_total; > @@ -401,6 +451,7 @@ bond_ethdev_tx_burst_round_robin(void *queue, struct rte_mbuf **bufs, > > static int slave_idx = 0; > int i, cslave_idx = 0, tx_fail_total = 0; > + int queue_id; > > bd_tx_q = (struct bond_tx_queue *)queue; > internals = bd_tx_q->dev_private; > @@ -427,7 +478,9 @@ bond_ethdev_tx_burst_round_robin(void *queue, struct rte_mbuf **bufs, > /* Send packet burst on each slave device */ > for (i = 0; i < num_of_slaves; i++) { > if (slave_nb_pkts[i] > 0) { > - num_tx_slave = rte_eth_tx_burst(slaves[i], bd_tx_q->queue_id, > + queue_id = bond_slave_txqid(internals, i, > + bd_tx_q->queue_id); > + num_tx_slave = rte_eth_tx_burst(slaves[i], queue_id, > slave_bufs[i], slave_nb_pkts[i]); > > /* if tx burst fails move packets to end of bufs */ > @@ -453,14 +506,27 @@ bond_ethdev_tx_burst_active_backup(void *queue, > { > struct bond_dev_private *internals; > struct bond_tx_queue *bd_tx_q; > + int queue_id; > + int i; > + uint8_t num_of_slaves; > + uint8_t slaves[RTE_MAX_ETHPORTS]; > > bd_tx_q = (struct bond_tx_queue *)queue; > internals = bd_tx_q->dev_private; > > - if (internals->active_slave_count < 1) > + num_of_slaves = internals->active_slave_count; > + memcpy(slaves, internals->active_slaves, > + sizeof(internals->active_slaves[0]) * num_of_slaves); > + > + if (num_of_slaves < 1) > return 0; > > - return rte_eth_tx_burst(internals->current_primary_port, bd_tx_q->queue_id, > + for (i = 0; i < num_of_slaves; i++) > + if (slaves[i] == internals->current_primary_port) > + break; > + > + queue_id = bond_slave_txqid(internals, i, bd_tx_q->queue_id); > + return rte_eth_tx_burst(internals->current_primary_port, queue_id, > bufs, nb_pkts); > } > > @@ -696,6 +762,7 @@ bond_ethdev_tx_burst_tlb(void *queue, struct rte_mbuf **bufs, uint16_t nb_pkts) > struct ether_hdr *ether_hdr; > struct ether_addr primary_slave_addr; > struct ether_addr active_slave_addr; > + int queue_id; > > if (num_of_slaves < 1) > return num_tx_total; > @@ -725,7 +792,8 @@ bond_ethdev_tx_burst_tlb(void *queue, struct rte_mbuf **bufs, uint16_t nb_pkts) > #endif > } > > - num_tx_total += rte_eth_tx_burst(slaves[i], bd_tx_q->queue_id, > + queue_id = bond_slave_txqid(internals, i, bd_tx_q->queue_id); > + num_tx_total += rte_eth_tx_burst(slaves[i], queue_id, > bufs + num_tx_total, nb_pkts - num_tx_total); > > if (num_tx_total == nb_pkts) > @@ -903,6 +971,7 @@ bond_ethdev_tx_burst_balance(void *queue, struct rte_mbuf **bufs, > uint16_t num_tx_total = 0, num_tx_slave = 0, tx_fail_total = 0; > > int i, op_slave_id; > + int queue_id; > > struct rte_mbuf *slave_bufs[RTE_MAX_ETHPORTS][nb_pkts]; > uint16_t slave_nb_pkts[RTE_MAX_ETHPORTS] = { 0 }; > @@ -931,7 +1000,9 @@ bond_ethdev_tx_burst_balance(void *queue, struct rte_mbuf **bufs, > /* Send packet burst on each slave device */ > for (i = 0; i < num_of_slaves; i++) { > if (slave_nb_pkts[i] > 0) { > - num_tx_slave = rte_eth_tx_burst(slaves[i], bd_tx_q->queue_id, > + queue_id = bond_slave_txqid(internals, i, > + bd_tx_q->queue_id); > + num_tx_slave = rte_eth_tx_burst(slaves[i], queue_id, > slave_bufs[i], slave_nb_pkts[i]); > > /* if tx burst fails move packets to end of bufs */ > @@ -977,6 +1048,8 @@ bond_ethdev_tx_burst_8023ad(void *queue, struct rte_mbuf **bufs, > /* Slow packets placed in each slave */ > uint8_t slave_slow_nb_pkts[RTE_MAX_ETHPORTS] = { 0 }; > > + int queue_id; > + > bd_tx_q = (struct bond_tx_queue *)queue; > internals = bd_tx_q->dev_private; > > @@ -1022,7 +1095,8 @@ bond_ethdev_tx_burst_8023ad(void *queue, struct rte_mbuf **bufs, > if (slave_nb_pkts[i] == 0) > continue; > > - num_tx_slave = rte_eth_tx_burst(slaves[i], bd_tx_q->queue_id, > + queue_id = bond_slave_txqid(internals, i, bd_tx_q->queue_id); > + num_tx_slave = rte_eth_tx_burst(slaves[i], queue_id, > slave_bufs[i], slave_nb_pkts[i]); > > /* If tx burst fails drop slow packets */ > @@ -1057,6 +1131,7 @@ bond_ethdev_tx_burst_broadcast(void *queue, struct rte_mbuf **bufs, > > int slave_tx_total[RTE_MAX_ETHPORTS]; > int i, most_successful_tx_slave = -1; > + int queue_id; > > bd_tx_q = (struct bond_tx_queue *)queue; > internals = bd_tx_q->dev_private; > @@ -1076,7 +1151,8 @@ bond_ethdev_tx_burst_broadcast(void *queue, struct rte_mbuf **bufs, > > /* Transmit burst on each active slave */ > for (i = 0; i < num_of_slaves; i++) { > - slave_tx_total[i] = rte_eth_tx_burst(slaves[i], bd_tx_q->queue_id, > + queue_id = bond_slave_txqid(internals, i, bd_tx_q->queue_id); > + slave_tx_total[i] = rte_eth_tx_burst(slaves[i], queue_id, > bufs, nb_pkts); > > if (unlikely(slave_tx_total[i] < nb_pkts)) > @@ -1298,9 +1374,22 @@ int > slave_configure(struct rte_eth_dev *bonded_eth_dev, > struct rte_eth_dev *slave_eth_dev) > { > + struct bond_dev_private *internals; > struct bond_rx_queue *bd_rx_q; > struct bond_tx_queue *bd_tx_q; > + int slave_id; > + > + internals = bonded_eth_dev->data->dev_private; > > + for (slave_id = 0; slave_id < internals->slave_count; slave_id++) > + if (internals->slaves[slave_id].port_id == > + slave_eth_dev->data->port_id) > + break; > + > + RTE_VERIFY(slave_id != internals->slave_count); > + > + uint16_t nb_rx_queues = internals->slaves[slave_id].nb_rx_queues; > + uint16_t nb_tx_queues = internals->slaves[slave_id].nb_tx_queues; > int errval; > uint16_t q_id; > > @@ -1331,8 +1420,7 @@ slave_configure(struct rte_eth_dev *bonded_eth_dev, > > /* Configure device */ > errval = rte_eth_dev_configure(slave_eth_dev->data->port_id, > - bonded_eth_dev->data->nb_rx_queues, > - bonded_eth_dev->data->nb_tx_queues, > + nb_rx_queues, nb_tx_queues, > &(slave_eth_dev->data->dev_conf)); > if (errval != 0) { > RTE_BOND_LOG(ERR, "Cannot configure slave device: port %u , err (%d)", > @@ -1343,7 +1431,7 @@ slave_configure(struct rte_eth_dev *bonded_eth_dev, > /* Setup Rx Queues */ > /* Use existing queues, if any */ > for (q_id = slave_eth_dev->data->nb_rx_queues; > - q_id < bonded_eth_dev->data->nb_rx_queues; q_id++) { > + q_id < nb_rx_queues ; q_id++) { > bd_rx_q = (struct bond_rx_queue *)bonded_eth_dev->data->rx_queues[q_id]; > > errval = rte_eth_rx_queue_setup(slave_eth_dev->data->port_id, q_id, > @@ -1361,7 +1449,7 @@ slave_configure(struct rte_eth_dev *bonded_eth_dev, > /* Setup Tx Queues */ > /* Use existing queues, if any */ > for (q_id = slave_eth_dev->data->nb_tx_queues; > - q_id < bonded_eth_dev->data->nb_tx_queues; q_id++) { > + q_id < nb_tx_queues ; q_id++) { > bd_tx_q = (struct bond_tx_queue *)bonded_eth_dev->data->tx_queues[q_id]; > > errval = rte_eth_tx_queue_setup(slave_eth_dev->data->port_id, q_id, > @@ -1440,7 +1528,8 @@ bond_ethdev_slave_link_status_change_monitor(void *cb_arg); > > void > slave_add(struct bond_dev_private *internals, > - struct rte_eth_dev *slave_eth_dev) > + struct rte_eth_dev *slave_eth_dev, > + const struct rte_eth_dev_info *slave_dev_info) > { > struct bond_slave_details *slave_details = > &internals->slaves[internals->slave_count]; > @@ -1448,6 +1537,20 @@ slave_add(struct bond_dev_private *internals, > slave_details->port_id = slave_eth_dev->data->port_id; > slave_details->last_link_status = 0; > > + uint16_t bond_nb_rx_queues = > + rte_eth_devices[internals->port_id].data->nb_rx_queues; > + uint16_t bond_nb_tx_queues = > + rte_eth_devices[internals->port_id].data->nb_tx_queues; > + > + slave_details->nb_rx_queues = > + bond_nb_rx_queues > slave_dev_info->max_rx_queues > + ? slave_dev_info->max_rx_queues > + : bond_nb_rx_queues; > + slave_details->nb_tx_queues = > + bond_nb_tx_queues > slave_dev_info->max_tx_queues > + ? slave_dev_info->max_tx_queues > + : bond_nb_tx_queues; > + > /* If slave device doesn't support interrupts then we need to enabled > * polling to monitor link status */ > if (!(slave_eth_dev->data->dev_flags & RTE_PCI_DRV_INTR_LSC)) { > diff --git a/drivers/net/bonding/rte_eth_bond_private.h b/drivers/net/bonding/rte_eth_bond_private.h > index 6c47a29..02f6de1 100644 > --- a/drivers/net/bonding/rte_eth_bond_private.h > +++ b/drivers/net/bonding/rte_eth_bond_private.h > @@ -101,6 +101,8 @@ struct bond_slave_details { > uint8_t link_status_poll_enabled; > uint8_t link_status_wait_to_complete; > uint8_t last_link_status; > + uint16_t nb_rx_queues; > + uint16_t nb_tx_queues; > /**< Port Id of slave eth_dev */ > struct ether_addr persisted_mac_addr; > > @@ -240,7 +242,8 @@ slave_remove(struct bond_dev_private *internals, > > void > slave_add(struct bond_dev_private *internals, > - struct rte_eth_dev *slave_eth_dev); > + struct rte_eth_dev *slave_eth_dev, > + const struct rte_eth_dev_info *slave_dev_info); > > uint16_t > xmit_l2_hash(const struct rte_mbuf *buf, uint8_t slave_count); > -- > 2.1.4 > -- Andriy Berestovskyy