From patchwork Mon Nov 21 15:01:08 2016 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Andrew Rybchenko X-Patchwork-Id: 17160 X-Patchwork-Delegate: ferruh.yigit@amd.com Return-Path: X-Original-To: patchwork@dpdk.org Delivered-To: patchwork@dpdk.org Received: from [92.243.14.124] (localhost [IPv6:::1]) by dpdk.org (Postfix) with ESMTP id 49957D6BA; Mon, 21 Nov 2016 16:03:58 +0100 (CET) Received: from nbfkord-smmo02.seg.att.com (nbfkord-smmo02.seg.att.com [209.65.160.78]) by dpdk.org (Postfix) with ESMTP id 471ED5582 for ; Mon, 21 Nov 2016 16:01:57 +0100 (CET) Received: from unknown [12.187.104.26] (EHLO nbfkord-smmo02.seg.att.com) by nbfkord-smmo02.seg.att.com(mxl_mta-7.2.4-7) with ESMTP id 56c03385.2b927801f940.1541332.00-2495.3424383.nbfkord-smmo02.seg.att.com (envelope-from ); Mon, 21 Nov 2016 15:01:57 +0000 (UTC) X-MXL-Hash: 58330c657c2fef50-5999b13b5e58dcc4be2854c23791fb1b24e767fe Received: from unknown [12.187.104.26] by nbfkord-smmo02.seg.att.com(mxl_mta-7.2.4-7) with SMTP id 55c03385.0.1541296.00-2347.3424250.nbfkord-smmo02.seg.att.com (envelope-from ); Mon, 21 Nov 2016 15:01:43 +0000 (UTC) X-MXL-Hash: 58330c5752100c9f-bd49d2a1af7a60dbe81aa6eb75bafbe83055463b Received: from ocex03.SolarFlarecom.com (10.20.40.36) by ocex03.SolarFlarecom.com (10.20.40.36) with Microsoft SMTP Server (TLS) id 15.0.1044.25; Mon, 21 Nov 2016 07:01:23 -0800 Received: from opal.uk.solarflarecom.com (10.17.10.1) by ocex03.SolarFlarecom.com (10.20.40.36) with Microsoft SMTP Server (TLS) id 15.0.1044.25 via Frontend Transport; Mon, 21 Nov 2016 07:01:22 -0800 Received: from uklogin.uk.solarflarecom.com (uklogin.uk.solarflarecom.com [10.17.10.10]) by opal.uk.solarflarecom.com (8.13.8/8.13.8) with ESMTP id uALF1LXe007260; Mon, 21 Nov 2016 15:01:21 GMT Received: from uklogin.uk.solarflarecom.com (localhost.localdomain [127.0.0.1]) by uklogin.uk.solarflarecom.com (8.13.8/8.13.8) with ESMTP id uALF1J3j006765; Mon, 21 Nov 2016 15:01:21 GMT From: Andrew Rybchenko To: CC: Ivan Malov Date: Mon, 21 Nov 2016 15:01:08 +0000 Message-ID: <1479740470-6723-55-git-send-email-arybchenko@solarflare.com> X-Mailer: git-send-email 1.8.2.3 In-Reply-To: <1479740470-6723-1-git-send-email-arybchenko@solarflare.com> References: <1479740470-6723-1-git-send-email-arybchenko@solarflare.com> MIME-Version: 1.0 X-AnalysisOut: [v=2.1 cv=UI/baXry c=1 sm=1 tr=0 a=8BlWFWvVlq5taO8ncb8nKg==] X-AnalysisOut: [:17 a=L24OOQBejmoA:10 a=pK7X0mNQAAAA:8 a=zRKbQ67AAAAA:8 a=] X-AnalysisOut: [qEWDoAS3dpP2Ns0Aji0A:9 a=5HA-qpC1VU4iIGLgRoNS:22 a=PA03WX8] X-AnalysisOut: [tBzeizutn5_OT:22] X-Spam: [F=0.3471414136; CM=0.500; S=0.347(2015072901)] X-MAIL-FROM: X-SOURCE-IP: [12.187.104.26] Subject: [dpdk-dev] [PATCH 54/56] net/sfc: add callbacks to set up and release Tx queues 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: , Errors-To: dev-bounces@dpdk.org Sender: "dev" From: Ivan Malov Reviewed-by: Andy Moreton Signed-off-by: Ivan Malov Signed-off-by: Andrew Rybchenko --- drivers/net/sfc/efx/sfc_ethdev.c | 58 ++++++++++++++ drivers/net/sfc/efx/sfc_ev.h | 2 + drivers/net/sfc/efx/sfc_tx.c | 167 +++++++++++++++++++++++++++++++++++++++ drivers/net/sfc/efx/sfc_tx.h | 36 +++++++++ 4 files changed, 263 insertions(+) diff --git a/drivers/net/sfc/efx/sfc_ethdev.c b/drivers/net/sfc/efx/sfc_ethdev.c index 4afd30c..98e5d83 100644 --- a/drivers/net/sfc/efx/sfc_ethdev.c +++ b/drivers/net/sfc/efx/sfc_ethdev.c @@ -39,6 +39,7 @@ #include "sfc_kvargs.h" #include "sfc_ev.h" #include "sfc_rx.h" +#include "sfc_tx.h" static void @@ -266,6 +267,61 @@ sfc_rx_queue_release(void *queue) sfc_adapter_unlock(sa); } +static int +sfc_tx_queue_setup(struct rte_eth_dev *dev, uint16_t tx_queue_id, + uint16_t nb_tx_desc, unsigned int socket_id, + const struct rte_eth_txconf *tx_conf) +{ + struct sfc_adapter *sa = dev->data->dev_private; + int rc; + + sfc_log_init(sa, "TxQ = %u, nb_tx_desc = %u, socket_id = %u", + tx_queue_id, nb_tx_desc, socket_id); + + sfc_adapter_lock(sa); + + rc = sfc_tx_qinit(sa, tx_queue_id, nb_tx_desc, socket_id, tx_conf); + if (rc != 0) + goto fail_tx_qinit; + + dev->data->tx_queues[tx_queue_id] = sa->txq_info[tx_queue_id].txq; + + sfc_adapter_unlock(sa); + return 0; + +fail_tx_qinit: + sfc_adapter_unlock(sa); + SFC_ASSERT(rc > 0); + return -rc; +} + +static void +sfc_tx_queue_release(void *queue) +{ + struct sfc_txq *txq = queue; + unsigned int sw_index; + struct sfc_adapter *sa; + + if (txq == NULL) + return; + + sw_index = sfc_txq_sw_index(txq); + + SFC_ASSERT(txq->evq != NULL); + sa = txq->evq->sa; + + sfc_log_init(sa, "TxQ = %u", sw_index); + + sfc_adapter_lock(sa); + + SFC_ASSERT(sw_index < sa->eth_dev->data->nb_tx_queues); + sa->eth_dev->data->tx_queues[sw_index] = NULL; + + sfc_tx_qfini(sa, sw_index); + + sfc_adapter_unlock(sa); +} + static const struct eth_dev_ops sfc_eth_dev_ops = { .dev_configure = sfc_dev_configure, .dev_start = sfc_dev_start, @@ -275,6 +331,8 @@ static const struct eth_dev_ops sfc_eth_dev_ops = { .dev_infos_get = sfc_dev_infos_get, .rx_queue_setup = sfc_rx_queue_setup, .rx_queue_release = sfc_rx_queue_release, + .tx_queue_setup = sfc_tx_queue_setup, + .tx_queue_release = sfc_tx_queue_release, }; static int diff --git a/drivers/net/sfc/efx/sfc_ev.h b/drivers/net/sfc/efx/sfc_ev.h index d053182..71e51a5 100644 --- a/drivers/net/sfc/efx/sfc_ev.h +++ b/drivers/net/sfc/efx/sfc_ev.h @@ -41,6 +41,7 @@ extern "C" { struct sfc_adapter; struct sfc_rxq; +struct sfc_txq; enum sfc_evq_state { SFC_EVQ_UNINITIALIZED = 0, @@ -58,6 +59,7 @@ struct sfc_evq { boolean_t exception; efsys_mem_t mem; struct sfc_rxq *rxq; + struct sfc_txq *txq; /* Not used on datapath */ struct sfc_adapter *sa; diff --git a/drivers/net/sfc/efx/sfc_tx.c b/drivers/net/sfc/efx/sfc_tx.c index fecd058..a4ffe9c 100644 --- a/drivers/net/sfc/efx/sfc_tx.c +++ b/drivers/net/sfc/efx/sfc_tx.c @@ -33,6 +33,165 @@ #include "sfc_tx.h" static int +sfc_tx_qcheck_conf(struct sfc_adapter *sa, + const struct rte_eth_txconf *tx_conf) +{ + unsigned int flags = tx_conf->txq_flags; + int rc = 0; + + if (tx_conf->tx_rs_thresh != 0) { + sfc_err(sa, "RS bit in transmit descriptor is not supported"); + rc = EINVAL; + } + + if (tx_conf->tx_free_thresh != 0) { + sfc_err(sa, + "setting explicit TX free threshold is not supported"); + rc = EINVAL; + } + + if (tx_conf->tx_deferred_start != 0) { + sfc_err(sa, "TX queue deferred start is not supported (yet)"); + rc = EINVAL; + } + + if (tx_conf->tx_thresh.pthresh != 0 || + tx_conf->tx_thresh.hthresh != 0 || + tx_conf->tx_thresh.wthresh != 0) { + sfc_err(sa, + "prefetch/host/writeback thresholds are not supported"); + rc = EINVAL; + } + + if ((flags & ETH_TXQ_FLAGS_NOVLANOFFL) == 0) { + sfc_err(sa, "VLAN offload is not supported"); + rc = EINVAL; + } + + if ((flags & ETH_TXQ_FLAGS_NOXSUMSCTP) == 0) { + sfc_err(sa, "SCTP offload is not supported"); + rc = EINVAL; + } + + /* We either perform both TCP and UDP offload, or no offload at all */ + if (((flags & ETH_TXQ_FLAGS_NOXSUMTCP) == 0) != + ((flags & ETH_TXQ_FLAGS_NOXSUMUDP) == 0)) { + sfc_err(sa, "TCP and UDP offloads can't be set independently"); + rc = EINVAL; + } + + return rc; +} + +int +sfc_tx_qinit(struct sfc_adapter *sa, unsigned int sw_index, + uint16_t nb_tx_desc, unsigned int socket_id, + const struct rte_eth_txconf *tx_conf) +{ + struct sfc_txq_info *txq_info; + struct sfc_evq *evq; + struct sfc_txq *txq; + unsigned int evq_index = sfc_evq_index_by_txq_sw_index(sa, sw_index); + int rc = 0; + + sfc_log_init(sa, "TxQ = %u", sw_index); + + rc = sfc_tx_qcheck_conf(sa, tx_conf); + if (rc != 0) + goto fail_bad_conf; + + SFC_ASSERT(sw_index < sa->txq_count); + txq_info = &sa->txq_info[sw_index]; + + SFC_ASSERT(nb_tx_desc <= sa->txq_max_entries); + txq_info->entries = nb_tx_desc; + + rc = sfc_ev_qinit(sa, evq_index, txq_info->entries, socket_id); + if (rc != 0) + goto fail_ev_qinit; + + evq = sa->evq_info[evq_index].evq; + + rc = ENOMEM; + txq = rte_zmalloc_socket("sfc-txq", sizeof(*txq), 0, socket_id); + if (txq == NULL) + goto fail_txq_alloc; + + rc = sfc_dma_alloc(sa, "txq", sw_index, EFX_TXQ_SIZE(txq_info->entries), + socket_id, &txq->mem); + if (rc != 0) + goto fail_dma_alloc; + + rc = ENOMEM; + txq->pend_desc = rte_calloc_socket("sfc-txq-pend-desc", + EFX_TXQ_LIMIT(txq_info->entries), + sizeof(efx_desc_t), 0, socket_id); + if (txq->pend_desc == NULL) + goto fail_pend_desc_alloc; + + rc = ENOMEM; + txq->sw_ring = rte_calloc_socket("sfc-txq-desc", txq_info->entries, + sizeof(*txq->sw_ring), 0, socket_id); + if (txq->sw_ring == NULL) + goto fail_desc_alloc; + + txq->state = SFC_TXQ_INITIALIZED; + txq->ptr_mask = txq_info->entries - 1; + txq->hw_index = sw_index; + txq->flags = tx_conf->txq_flags; + txq->evq = evq; + + evq->txq = txq; + + txq_info->txq = txq; + + return 0; + +fail_desc_alloc: + rte_free(txq->pend_desc); + +fail_pend_desc_alloc: + sfc_dma_free(sa, &txq->mem); + +fail_dma_alloc: + rte_free(txq); + +fail_txq_alloc: + sfc_ev_qfini(sa, evq_index); + +fail_ev_qinit: + txq_info->entries = 0; + +fail_bad_conf: + sfc_log_init(sa, "failed (TxQ = %u, rc = %d)", sw_index, rc); + return rc; +} + +void +sfc_tx_qfini(struct sfc_adapter *sa, unsigned int sw_index) +{ + struct sfc_txq_info *txq_info; + struct sfc_txq *txq; + + sfc_log_init(sa, "TxQ = %u", sw_index); + + SFC_ASSERT(sw_index < sa->txq_count); + txq_info = &sa->txq_info[sw_index]; + + txq = txq_info->txq; + SFC_ASSERT(txq != NULL); + SFC_ASSERT(txq->state == SFC_TXQ_INITIALIZED); + + txq_info->txq = NULL; + txq_info->entries = 0; + + rte_free(txq->sw_ring); + rte_free(txq->pend_desc); + sfc_dma_free(sa, &txq->mem); + rte_free(txq); +} + +static int sfc_tx_qinit_info(struct sfc_adapter *sa, unsigned int sw_index) { struct sfc_txq_info *txq_info = &sa->txq_info[sw_index]; @@ -121,6 +280,14 @@ sfc_tx_init(struct sfc_adapter *sa) void sfc_tx_fini(struct sfc_adapter *sa) { + int sw_index; + + sw_index = sa->txq_count; + while (--sw_index >= 0) { + if (sa->txq_info[sw_index].txq != NULL) + sfc_tx_qfini(sa, sw_index); + } + rte_free(sa->txq_info); sa->txq_info = NULL; sa->txq_count = 0; diff --git a/drivers/net/sfc/efx/sfc_tx.h b/drivers/net/sfc/efx/sfc_tx.h index b2d4875..3278797 100644 --- a/drivers/net/sfc/efx/sfc_tx.h +++ b/drivers/net/sfc/efx/sfc_tx.h @@ -40,13 +40,49 @@ extern "C" { #endif struct sfc_adapter; +struct sfc_evq; + +struct sfc_tx_sw_desc { + struct rte_mbuf *mbuf; +}; + +enum sfc_txq_state_bit { + SFC_TXQ_INITIALIZED_BIT = 0, +#define SFC_TXQ_INITIALIZED (1 << SFC_TXQ_INITIALIZED_BIT) +}; + +struct sfc_txq { + struct sfc_evq *evq; + struct sfc_tx_sw_desc *sw_ring; + unsigned int state; + unsigned int ptr_mask; + efx_desc_t *pend_desc; + efx_txq_t *common; + efsys_mem_t mem; + + unsigned int hw_index; + unsigned int flags; +}; + +static inline unsigned int +sfc_txq_sw_index(const struct sfc_txq *txq) +{ + return txq->hw_index; +} struct sfc_txq_info { + unsigned int entries; + struct sfc_txq *txq; }; int sfc_tx_init(struct sfc_adapter *sa); void sfc_tx_fini(struct sfc_adapter *sa); +int sfc_tx_qinit(struct sfc_adapter *sa, unsigned int sw_index, + uint16_t nb_tx_desc, unsigned int socket_id, + const struct rte_eth_txconf *tx_conf); +void sfc_tx_qfini(struct sfc_adapter *sa, unsigned int sw_index); + #ifdef __cplusplus } #endif