From: Billy McFall <bmcfall@redhat.com>
To: thomas.monjalon@6wind.com, wenzhuo.lu@intel.com, olivier.matz@6wind.com
Cc: dev@dpdk.org, Billy McFall <bmcfall@redhat.com>
Subject: [dpdk-dev] [PATCH v8 2/3] net/e1000: e1000 igb support to free consumed buffers
Date: Fri, 24 Mar 2017 14:55:54 -0400 [thread overview]
Message-ID: <20170324185555.18398-3-bmcfall@redhat.com> (raw)
In-Reply-To: <20170324185555.18398-1-bmcfall@redhat.com>
Add support to the e1000 igb driver for the new API to force free
consumed buffers on Tx ring. This API is independent of the tx_rs_thresh
setting. With this API, buffers should be free even if tx_rs_thresh is
not met.
e1000 igb driver does not implement a tx_rs_thresh to free mbufs, it
frees a slot in the ring as needed. However, it could be implemented at
some future date.
Signed-off-by: Billy McFall <bmcfall@redhat.com>
Acked-by: Wenzhuo Lu <wenzhuo.lu@intel.com>
Acked-by: Keith Wiles <Keith.Wiles@intel.com>
---
doc/guides/nics/features/e1000.ini | 1 +
drivers/net/e1000/e1000_ethdev.h | 2 +
drivers/net/e1000/igb_ethdev.c | 1 +
drivers/net/e1000/igb_rxtx.c | 126 +++++++++++++++++++++++++++++++++++++
4 files changed, 130 insertions(+)
diff --git a/doc/guides/nics/features/e1000.ini b/doc/guides/nics/features/e1000.ini
index 7f6d55c..3aed7d7 100644
--- a/doc/guides/nics/features/e1000.ini
+++ b/doc/guides/nics/features/e1000.ini
@@ -7,6 +7,7 @@
Link status = Y
Link status event = Y
Rx interrupt = Y
+Free Tx mbuf on demand = Y
MTU update = Y
Jumbo frame = Y
Scattered Rx = Y
diff --git a/drivers/net/e1000/e1000_ethdev.h b/drivers/net/e1000/e1000_ethdev.h
index 81a6dbb..39b2f43 100644
--- a/drivers/net/e1000/e1000_ethdev.h
+++ b/drivers/net/e1000/e1000_ethdev.h
@@ -315,6 +315,8 @@ int eth_igb_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);
+int eth_igb_tx_done_cleanup(void *txq, uint32_t free_cnt);
+
int eth_igb_rx_init(struct rte_eth_dev *dev);
void eth_igb_tx_init(struct rte_eth_dev *dev);
diff --git a/drivers/net/e1000/igb_ethdev.c b/drivers/net/e1000/igb_ethdev.c
index a112b38..71d05a9 100644
--- a/drivers/net/e1000/igb_ethdev.c
+++ b/drivers/net/e1000/igb_ethdev.c
@@ -408,6 +408,7 @@ static const struct eth_dev_ops eth_igb_ops = {
.rx_descriptor_done = eth_igb_rx_descriptor_done,
.tx_queue_setup = eth_igb_tx_queue_setup,
.tx_queue_release = eth_igb_tx_queue_release,
+ .tx_done_cleanup = eth_igb_tx_done_cleanup,
.dev_led_on = eth_igb_led_on,
.dev_led_off = eth_igb_led_off,
.flow_ctrl_get = eth_igb_flow_ctrl_get,
diff --git a/drivers/net/e1000/igb_rxtx.c b/drivers/net/e1000/igb_rxtx.c
index 1bb4d85..cba3704 100644
--- a/drivers/net/e1000/igb_rxtx.c
+++ b/drivers/net/e1000/igb_rxtx.c
@@ -1283,6 +1283,132 @@ eth_igb_tx_queue_release(void *txq)
igb_tx_queue_release(txq);
}
+static int
+igb_tx_done_cleanup(struct igb_tx_queue *txq, uint32_t free_cnt)
+{
+ struct igb_tx_entry *sw_ring;
+ volatile union e1000_adv_tx_desc *txr;
+ uint16_t tx_first; /* First segment analyzed. */
+ uint16_t tx_id; /* Current segment being processed. */
+ uint16_t tx_last; /* Last segment in the current packet. */
+ uint16_t tx_next; /* First segment of the next packet. */
+ int count;
+
+ if (txq != NULL) {
+ count = 0;
+ sw_ring = txq->sw_ring;
+ txr = txq->tx_ring;
+
+ /*
+ * tx_tail is the last sent packet on the sw_ring. Goto the end
+ * of that packet (the last segment in the packet chain) and
+ * then the next segment will be the start of the oldest segment
+ * in the sw_ring. This is the first packet that will be
+ * attempted to be freed.
+ */
+
+ /* Get last segment in most recently added packet. */
+ tx_first = sw_ring[txq->tx_tail].last_id;
+
+ /* Get the next segment, which is the oldest segment in ring. */
+ tx_first = sw_ring[tx_first].next_id;
+
+ /* Set the current index to the first. */
+ tx_id = tx_first;
+
+ /*
+ * Loop through each packet. For each packet, verify that an
+ * mbuf exists and that the last segment is free. If so, free
+ * it and move on.
+ */
+ while (1) {
+ tx_last = sw_ring[tx_id].last_id;
+
+ if (sw_ring[tx_last].mbuf) {
+ if (txr[tx_last].wb.status &
+ E1000_TXD_STAT_DD) {
+ /*
+ * Increment the number of packets
+ * freed.
+ */
+ count++;
+
+ /* Get the start of the next packet. */
+ tx_next = sw_ring[tx_last].next_id;
+
+ /*
+ * Loop through all segments in a
+ * packet.
+ */
+ do {
+ rte_pktmbuf_free_seg(sw_ring[tx_id].mbuf);
+ sw_ring[tx_id].mbuf = NULL;
+ sw_ring[tx_id].last_id = tx_id;
+
+ /* Move to next segemnt. */
+ tx_id = sw_ring[tx_id].next_id;
+
+ } while (tx_id != tx_next);
+
+ if (unlikely(count == (int)free_cnt))
+ break;
+ } else
+ /*
+ * mbuf still in use, nothing left to
+ * free.
+ */
+ break;
+ } else {
+ /*
+ * There are multiple reasons to be here:
+ * 1) All the packets on the ring have been
+ * freed - tx_id is equal to tx_first
+ * and some packets have been freed.
+ * - Done, exit
+ * 2) Interfaces has not sent a rings worth of
+ * packets yet, so the segment after tail is
+ * still empty. Or a previous call to this
+ * function freed some of the segments but
+ * not all so there is a hole in the list.
+ * Hopefully this is a rare case.
+ * - Walk the list and find the next mbuf. If
+ * there isn't one, then done.
+ */
+ if (likely((tx_id == tx_first) && (count != 0)))
+ break;
+
+ /*
+ * Walk the list and find the next mbuf, if any.
+ */
+ do {
+ /* Move to next segemnt. */
+ tx_id = sw_ring[tx_id].next_id;
+
+ if (sw_ring[tx_id].mbuf)
+ break;
+
+ } while (tx_id != tx_first);
+
+ /*
+ * Determine why previous loop bailed. If there
+ * is not an mbuf, done.
+ */
+ if (sw_ring[tx_id].mbuf == NULL)
+ break;
+ }
+ }
+ } else
+ count = -ENODEV;
+
+ return count;
+}
+
+int
+eth_igb_tx_done_cleanup(void *txq, uint32_t free_cnt)
+{
+ return igb_tx_done_cleanup(txq, free_cnt);
+}
+
static void
igb_reset_tx_queue_stat(struct igb_tx_queue *txq)
{
--
2.9.3
next prev parent reply other threads:[~2017-03-24 18:56 UTC|newest]
Thread overview: 54+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-01-20 16:01 [dpdk-dev] [PATCH v3 0/3] new API to free consumed buffers in Tx ring Billy McFall
2017-01-20 16:01 ` [dpdk-dev] [PATCH v3 1/3] ethdev: " Billy McFall
2017-01-20 16:01 ` [dpdk-dev] [PATCH v3 2/3] net/e1000: e1000 igb support to free consumed buffers Billy McFall
2017-01-22 3:47 ` Lu, Wenzhuo
2017-01-23 13:49 ` Billy McFall
2017-01-24 0:42 ` Lu, Wenzhuo
2017-01-20 16:01 ` [dpdk-dev] [PATCH v3 3/3] net/vhost: vHost " Billy McFall
2017-01-23 15:25 ` [dpdk-dev] [PATCH v3 0/3] new API to free consumed buffers in Tx ring Thomas Monjalon
2017-01-23 21:13 ` [dpdk-dev] [PATCH v4 " Billy McFall
2017-01-23 21:13 ` [dpdk-dev] [PATCH v4 1/3] ethdev: " Billy McFall
2017-01-23 21:13 ` [dpdk-dev] [PATCH v4 2/3] net/e1000: e1000 igb support to free consumed buffers Billy McFall
2017-01-23 21:13 ` [dpdk-dev] [PATCH v4 3/3] net/vhost: vHost " Billy McFall
2017-01-27 18:37 ` [dpdk-dev] [PATCH v5 0/3] new API to free consumed buffers in Tx ring Billy McFall
2017-01-27 18:37 ` [dpdk-dev] [PATCH v5 1/3] ethdev: " Billy McFall
2017-02-27 13:48 ` Thomas Monjalon
2017-03-07 14:29 ` Billy McFall
2017-03-07 16:03 ` Thomas Monjalon
2017-03-09 15:49 ` Billy McFall
2017-03-09 17:11 ` Thomas Monjalon
2017-03-07 16:35 ` Mcnamara, John
2017-03-07 16:42 ` Mcnamara, John
2017-01-27 18:37 ` [dpdk-dev] [PATCH v5 2/3] net/e1000: e1000 igb support to free consumed buffers Billy McFall
2017-02-27 13:49 ` Thomas Monjalon
2017-02-28 1:07 ` Lu, Wenzhuo
2017-01-27 18:38 ` [dpdk-dev] [PATCH v5 3/3] net/vhost: vHost " Billy McFall
2017-02-27 13:50 ` Thomas Monjalon
2017-02-28 6:41 ` Yuanhan Liu
2017-03-01 10:15 ` Maxime Coquelin
2017-03-07 21:59 ` [dpdk-dev] [PATCH v5 0/3] new API to free consumed buffers in Tx ring Thomas Monjalon
2017-03-09 20:51 ` [dpdk-dev] [PATCH v6 " Billy McFall
2017-03-09 20:51 ` [dpdk-dev] [PATCH v6 1/3] ethdev: " Billy McFall
2017-03-15 10:29 ` Olivier Matz
2017-03-15 15:01 ` Billy McFall
2017-03-15 10:30 ` Thomas Monjalon
2017-03-09 20:51 ` [dpdk-dev] [PATCH v6 2/3] net/e1000: e1000 igb support to free consumed buffers Billy McFall
2017-03-13 3:17 ` Lu, Wenzhuo
2017-03-09 20:51 ` [dpdk-dev] [PATCH v6 3/3] net/vhost: vHost " Billy McFall
2017-03-15 10:27 ` Thomas Monjalon
2017-03-15 18:02 ` [dpdk-dev] [PATCH v7 0/3] new API to free consumed buffers in Tx ring Billy McFall
2017-03-15 18:02 ` [dpdk-dev] [PATCH v7 1/3] ethdev: " Billy McFall
2017-03-23 10:37 ` Olivier MATZ
2017-03-23 13:32 ` Billy McFall
2017-03-24 12:46 ` Olivier Matz
2017-03-24 13:18 ` Billy McFall
2017-03-24 13:30 ` Olivier Matz
2017-03-15 18:02 ` [dpdk-dev] [PATCH v7 2/3] net/e1000: e1000 igb support to free consumed buffers Billy McFall
2017-03-15 18:02 ` [dpdk-dev] [PATCH v7 3/3] net/vhost: vHost " Billy McFall
2017-03-15 20:25 ` [dpdk-dev] [PATCH v7 0/3] new API to free consumed buffers in Tx ring Wiles, Keith
2017-03-24 18:55 ` [dpdk-dev] [PATCH v8 " Billy McFall
2017-03-24 18:55 ` [dpdk-dev] [PATCH v8 1/3] ethdev: " Billy McFall
2017-03-24 18:55 ` Billy McFall [this message]
2017-03-24 18:55 ` [dpdk-dev] [PATCH v8 3/3] net/vhost: vHost support to free consumed buffers Billy McFall
2017-03-27 15:20 ` [dpdk-dev] [PATCH v8 0/3] new API to free consumed buffers in Tx ring Thomas Monjalon
2017-04-19 16:25 ` Ferruh Yigit
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=20170324185555.18398-3-bmcfall@redhat.com \
--to=bmcfall@redhat.com \
--cc=dev@dpdk.org \
--cc=olivier.matz@6wind.com \
--cc=thomas.monjalon@6wind.com \
--cc=wenzhuo.lu@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).