DPDK patches and discussions
 help / color / mirror / Atom feed
From: Junfeng Guo <junfeng.guo@intel.com>
To: qi.z.zhang@intel.com, jingjing.wu@intel.com,
	ferruh.yigit@amd.com, beilei.xing@intel.com
Cc: dev@dpdk.org, xiaoyun.li@intel.com, helin.zhang@intel.com,
	Junfeng Guo <junfeng.guo@intel.com>,
	Rushil Gupta <rushilg@google.com>,
	Jordan Kimbrough <jrkim@google.com>,
	Jeroen de Borst <jeroendb@google.com>
Subject: [RFC v2 3/9] net/gve: support device start and close for DQO
Date: Mon, 30 Jan 2023 14:26:36 +0800	[thread overview]
Message-ID: <20230130062642.3337239-4-junfeng.guo@intel.com> (raw)
In-Reply-To: <20230130062642.3337239-1-junfeng.guo@intel.com>

Add device start and close support for DQO.

Signed-off-by: Junfeng Guo <junfeng.guo@intel.com>
Signed-off-by: Rushil Gupta <rushilg@google.com>
Signed-off-by: Jordan Kimbrough <jrkim@google.com>
Signed-off-by: Jeroen de Borst <jeroendb@google.com>
---
 drivers/net/gve/base/gve_adminq.c | 10 +++----
 drivers/net/gve/gve_ethdev.c      | 43 ++++++++++++++++++++++++++++++-
 2 files changed, 47 insertions(+), 6 deletions(-)

diff --git a/drivers/net/gve/base/gve_adminq.c b/drivers/net/gve/base/gve_adminq.c
index e745b709b2..e963f910a0 100644
--- a/drivers/net/gve/base/gve_adminq.c
+++ b/drivers/net/gve/base/gve_adminq.c
@@ -497,11 +497,11 @@ static int gve_adminq_create_tx_queue(struct gve_priv *priv, u32 queue_index)
 		cmd.create_tx_queue.queue_page_list_id = cpu_to_be32(qpl_id);
 	} else {
 		cmd.create_tx_queue.tx_ring_size =
-			cpu_to_be16(txq->nb_tx_desc);
+			cpu_to_be16(priv->tx_desc_cnt);
 		cmd.create_tx_queue.tx_comp_ring_addr =
-			cpu_to_be64(txq->complq->tx_ring_phys_addr);
+			cpu_to_be64(txq->compl_ring_phys_addr);
 		cmd.create_tx_queue.tx_comp_ring_size =
-			cpu_to_be16(priv->tx_compq_size);
+			cpu_to_be16(priv->tx_compq_size * DQO_TX_MULTIPLIER);
 	}
 
 	return gve_adminq_issue_cmd(priv, &cmd);
@@ -549,9 +549,9 @@ static int gve_adminq_create_rx_queue(struct gve_priv *priv, u32 queue_index)
 		cmd.create_rx_queue.rx_ring_size =
 			cpu_to_be16(priv->rx_desc_cnt);
 		cmd.create_rx_queue.rx_desc_ring_addr =
-			cpu_to_be64(rxq->rx_ring_phys_addr);
+			cpu_to_be64(rxq->compl_ring_phys_addr);
 		cmd.create_rx_queue.rx_data_ring_addr =
-			cpu_to_be64(rxq->bufq->rx_ring_phys_addr);
+			cpu_to_be64(rxq->rx_ring_phys_addr);
 		cmd.create_rx_queue.packet_buffer_size =
 			cpu_to_be16(rxq->rx_buf_len);
 		cmd.create_rx_queue.rx_buff_ring_size =
diff --git a/drivers/net/gve/gve_ethdev.c b/drivers/net/gve/gve_ethdev.c
index 26182b0422..3543378978 100644
--- a/drivers/net/gve/gve_ethdev.c
+++ b/drivers/net/gve/gve_ethdev.c
@@ -78,6 +78,9 @@ gve_free_qpls(struct gve_priv *priv)
 	uint16_t nb_rxqs = priv->max_nb_rxq;
 	uint32_t i;
 
+	if (priv->queue_format != GVE_GQI_QPL_FORMAT)
+		return;
+
 	for (i = 0; i < nb_txqs + nb_rxqs; i++) {
 		if (priv->qpl[i].mz != NULL)
 			rte_memzone_free(priv->qpl[i].mz);
@@ -138,6 +141,41 @@ gve_refill_pages(struct gve_rx_queue *rxq)
 	return 0;
 }
 
+static int
+gve_refill_dqo(struct gve_rx_queue *rxq)
+{
+	struct rte_mbuf *nmb;
+	uint16_t i;
+	int diag;
+
+	diag = rte_pktmbuf_alloc_bulk(rxq->mpool, &rxq->sw_ring[0], rxq->nb_rx_desc);
+	if (diag < 0) {
+		for (i = 0; i < rxq->nb_rx_desc - 1; i++) {
+			nmb = rte_pktmbuf_alloc(rxq->mpool);
+			if (!nmb)
+				break;
+			rxq->sw_ring[i] = nmb;
+		}
+		if (i < rxq->nb_rx_desc - 1)
+			return -ENOMEM;
+	}
+
+	for (i = 0; i < rxq->nb_rx_desc; i++) {
+		if (i == rxq->nb_rx_desc - 1)
+			break;
+		nmb = rxq->sw_ring[i];
+		rxq->rx_ring[i].buf_addr = rte_cpu_to_le_64(rte_mbuf_data_iova_default(nmb));
+		rxq->rx_ring[i].buf_id = rte_cpu_to_le_16(i);
+	}
+
+	rxq->nb_rx_hold = 0;
+	rxq->bufq_tail = rxq->nb_rx_desc - 1;
+
+	rte_write32(rxq->bufq_tail, rxq->qrx_tail);
+
+	return 0;
+}
+
 static int
 gve_link_update(struct rte_eth_dev *dev, __rte_unused int wait_to_complete)
 {
@@ -206,7 +244,10 @@ gve_dev_start(struct rte_eth_dev *dev)
 
 		rte_write32(rte_cpu_to_be_32(GVE_IRQ_MASK), rxq->ntfy_addr);
 
-		err = gve_refill_pages(rxq);
+		if (gve_is_gqi(priv))
+			err = gve_refill_pages(rxq);
+		else
+			err = gve_refill_dqo(rxq);
 		if (err) {
 			PMD_DRV_LOG(ERR, "Failed to refill for RX");
 			goto err_rx;
-- 
2.34.1


  parent reply	other threads:[~2023-01-30  6:32 UTC|newest]

Thread overview: 36+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-01-18  2:53 [RFC 0/8] gve PMD enhancement Junfeng Guo
2023-01-18  2:53 ` [RFC 1/8] net/gve: add Rx queue setup for DQO Junfeng Guo
2023-01-18  2:53 ` [RFC 2/8] net/gve: support device start and close " Junfeng Guo
2023-01-18  2:53 ` [RFC 3/8] net/gve: support queue release and stop " Junfeng Guo
2023-01-18  2:53 ` [RFC 4/8] net/gve: support basic Tx data path " Junfeng Guo
2023-01-18  2:53 ` [RFC 5/8] net/gve: support basic Rx " Junfeng Guo
2023-01-18  2:53 ` [RFC 6/8] net/gve: support basic stats " Junfeng Guo
2023-01-18  2:53 ` [RFC 7/8] net/gve: support jumbo frame for GQI Junfeng Guo
2023-01-18  2:53 ` [RFC 8/8] net/gve: add AdminQ command to verify driver compatibility Junfeng Guo
2023-01-25 13:37 ` [RFC 0/8] gve PMD enhancement Li, Xiaoyun
2023-01-30  6:26 ` [RFC v2 0/9] " Junfeng Guo
2023-01-30  6:26   ` [RFC v2 1/9] net/gve: add Tx queue setup for DQO Junfeng Guo
2023-01-30  6:26   ` [RFC v2 2/9] net/gve: add Rx " Junfeng Guo
2023-01-30  6:26   ` Junfeng Guo [this message]
2023-01-30  6:26   ` [RFC v2 4/9] net/gve: support queue release and stop " Junfeng Guo
2023-01-30  6:26   ` [RFC v2 5/9] net/gve: support basic Tx data path " Junfeng Guo
2023-01-30  6:26   ` [RFC v2 6/9] net/gve: support basic Rx " Junfeng Guo
2023-01-30 18:32     ` Honnappa Nagarahalli
2023-01-30  6:26   ` [RFC v2 7/9] net/gve: support basic stats " Junfeng Guo
2023-01-30 18:27     ` Honnappa Nagarahalli
2023-01-30  6:26   ` [RFC v2 8/9] net/gve: support jumbo frame for GQI Junfeng Guo
2023-01-30  6:26   ` [RFC v2 9/9] net/gve: add AdminQ command to verify driver compatibility Junfeng Guo
2023-02-17  7:32   ` [RFC v3 00/10] gve PMD enhancement Junfeng Guo
2023-02-17  7:32     ` [RFC v3 01/10] net/gve: add Tx queue setup for DQO Junfeng Guo
2023-02-17  7:32     ` [RFC v3 02/10] net/gve: add Rx " Junfeng Guo
2023-02-17  7:32     ` [RFC v3 03/10] net/gve: support device start and close " Junfeng Guo
2023-02-17  7:32     ` [RFC v3 04/10] net/gve: support queue release and stop " Junfeng Guo
2023-02-17  7:32     ` [RFC v3 05/10] net/gve: support basic Tx data path " Junfeng Guo
2023-02-17  7:32     ` [RFC v3 06/10] net/gve: support basic Rx " Junfeng Guo
2023-02-17 15:17       ` Honnappa Nagarahalli
2023-02-23  5:32         ` Guo, Junfeng
2023-02-17  7:32     ` [RFC v3 07/10] net/gve: support basic stats " Junfeng Guo
2023-02-17 15:28       ` Honnappa Nagarahalli
2023-02-17  7:32     ` [RFC v3 08/10] net/gve: enable Tx checksum offload " Junfeng Guo
2023-02-17  7:32     ` [RFC v3 09/10] net/gve: support jumbo frame for GQI Junfeng Guo
2023-02-17  7:32     ` [RFC v3 10/10] net/gve: add AdminQ command to verify driver compatibility Junfeng Guo

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20230130062642.3337239-4-junfeng.guo@intel.com \
    --to=junfeng.guo@intel.com \
    --cc=beilei.xing@intel.com \
    --cc=dev@dpdk.org \
    --cc=ferruh.yigit@amd.com \
    --cc=helin.zhang@intel.com \
    --cc=jeroendb@google.com \
    --cc=jingjing.wu@intel.com \
    --cc=jrkim@google.com \
    --cc=qi.z.zhang@intel.com \
    --cc=rushilg@google.com \
    --cc=xiaoyun.li@intel.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).