DPDK patches and discussions
 help / color / mirror / Atom feed
From: Stephen Hemminger <stephen@networkplumber.org>
To: dev@dpdk.org
Cc: Stephen Hemminger <stephen@networkplumber.org>
Subject: [dpdk-dev] [PATCH v8 02/15] ethdev: add linkstatus get/set helper functions
Date: Thu, 25 Jan 2018 18:01:38 -0800	[thread overview]
Message-ID: <20180126020151.979-3-stephen@networkplumber.org> (raw)
In-Reply-To: <20180126020151.979-1-stephen@networkplumber.org>

Many drivers are all doing copy/paste of the same code to atomically
update the link status. Reduce duplication, and allow for future
changes by having common function for this.

Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
---
 lib/librte_ether/rte_ethdev.c        | 22 +++----------
 lib/librte_ether/rte_ethdev_driver.h | 63 ++++++++++++++++++++++++++++++++++++
 2 files changed, 67 insertions(+), 18 deletions(-)

diff --git a/lib/librte_ether/rte_ethdev.c b/lib/librte_ether/rte_ethdev.c
index f285ba278967..344397101abd 100644
--- a/lib/librte_ether/rte_ethdev.c
+++ b/lib/librte_ether/rte_ethdev.c
@@ -1578,20 +1578,6 @@ rte_eth_allmulticast_get(uint16_t port_id)
 	return dev->data->all_multicast;
 }
 
-static inline int
-rte_eth_dev_atomic_read_link_status(struct rte_eth_dev *dev,
-				struct rte_eth_link *link)
-{
-	struct rte_eth_link *dst = link;
-	struct rte_eth_link *src = &(dev->data->dev_link);
-
-	if (rte_atomic64_cmpset((uint64_t *)dst, *(uint64_t *)dst,
-					*(uint64_t *)src) == 0)
-		return -1;
-
-	return 0;
-}
-
 void
 rte_eth_link_get(uint16_t port_id, struct rte_eth_link *eth_link)
 {
@@ -1600,8 +1586,8 @@ rte_eth_link_get(uint16_t port_id, struct rte_eth_link *eth_link)
 	RTE_ETH_VALID_PORTID_OR_RET(port_id);
 	dev = &rte_eth_devices[port_id];
 
-	if (dev->data->dev_conf.intr_conf.lsc != 0)
-		rte_eth_dev_atomic_read_link_status(dev, eth_link);
+	if (dev->data->dev_conf.intr_conf.lsc)
+		rte_eth_linkstatus_get(dev, eth_link);
 	else {
 		RTE_FUNC_PTR_OR_RET(*dev->dev_ops->link_update);
 		(*dev->dev_ops->link_update)(dev, 1);
@@ -1617,8 +1603,8 @@ rte_eth_link_get_nowait(uint16_t port_id, struct rte_eth_link *eth_link)
 	RTE_ETH_VALID_PORTID_OR_RET(port_id);
 	dev = &rte_eth_devices[port_id];
 
-	if (dev->data->dev_conf.intr_conf.lsc != 0)
-		rte_eth_dev_atomic_read_link_status(dev, eth_link);
+	if (dev->data->dev_conf.intr_conf.lsc)
+		rte_eth_linkstatus_get(dev, eth_link);
 	else {
 		RTE_FUNC_PTR_OR_RET(*dev->dev_ops->link_update);
 		(*dev->dev_ops->link_update)(dev, 0);
diff --git a/lib/librte_ether/rte_ethdev_driver.h b/lib/librte_ether/rte_ethdev_driver.h
index 45f08c65ec1e..fd3e1c0bd21f 100644
--- a/lib/librte_ether/rte_ethdev_driver.h
+++ b/lib/librte_ether/rte_ethdev_driver.h
@@ -68,6 +68,69 @@ struct rte_eth_dev *rte_eth_dev_attach_secondary(const char *name);
  */
 int rte_eth_dev_release_port(struct rte_eth_dev *eth_dev);
 
+/**
+ * @internal
+ * Atomically set the link status for the specific device.
+ * It is for use by DPDK device driver use only.
+ * User applications should not call it
+ *
+ * @param dev
+ *  Pointer to struct rte_eth_dev.
+ * @param link
+ *  New link status value.
+ * @return
+ *  Same convention as eth_link_update operation.
+ *  0   if link up status has changed
+ *  -1  if link up status was unchanged
+ */
+static inline int
+rte_eth_linkstatus_set(struct rte_eth_dev *dev,
+		       const struct rte_eth_link *new_link)
+{
+	volatile uint64_t *dev_link
+		 = (volatile uint64_t *)&(dev->data->dev_link);
+	union {
+		uint64_t val64;
+		struct rte_eth_link link;
+	} orig;
+
+	RTE_BUILD_BUG_ON(sizeof(*new_link) != sizeof(uint64_t));
+
+	orig.val64 = rte_atomic64_exchange(dev_link,
+					   *(const uint64_t *)new_link);
+
+	return (orig.link.link_status == new_link->link_status) ? -1 : 0;
+}
+
+/**
+ * @internal
+ * Atomically get the link speed and status.
+ *
+ * @param dev
+ *  Pointer to struct rte_eth_dev.
+ * @param link
+ *  link status value.
+ */
+static inline void
+rte_eth_linkstatus_get(const struct rte_eth_dev *dev,
+		       struct rte_eth_link *link)
+{
+	volatile uint64_t *src = (uint64_t *)&(dev->data->dev_link);
+	uint64_t *dst = (uint64_t *)link;
+
+	RTE_BUILD_BUG_ON(sizeof(*link) != sizeof(uint64_t));
+
+#ifdef __LP64__
+	/* if cpu arch has 64 bit unsigned lon then implicitly atomic */
+	*dst = *src;
+#else
+	/* can't use rte_atomic64_read because it returns signed int */
+	do {
+		*dst = *src;
+	} while (!rte_atomic64_cmpset(src, *dst, *dst));
+#endif
+}
+
 /**
  * @internal
  * Release device queues and clear its configuration to force the user
-- 
2.15.1

  parent reply	other threads:[~2018-01-26  2:01 UTC|newest]

Thread overview: 24+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-01-26  2:01 [dpdk-dev] [PATCH v8 00/15] common linkstatus functions Stephen Hemminger
2018-01-26  2:01 ` [dpdk-dev] [PATCH v8 01/15] eal: introduce atomic exchange operation Stephen Hemminger
2018-01-26 17:14   ` Thomas Monjalon
2018-01-26 17:24     ` Bruce Richardson
2018-01-26 21:54       ` Stephen Hemminger
2018-01-29  9:51         ` Bruce Richardson
2018-01-26 17:59   ` Ananyev, Konstantin
2018-01-26  2:01 ` Stephen Hemminger [this message]
2018-01-26  2:01 ` [dpdk-dev] [PATCH v8 03/15] net/virtio: use eth_linkstatus_set Stephen Hemminger
2018-01-26  2:01 ` [dpdk-dev] [PATCH v8 04/15] net/vmxnet3: use rte_eth_linkstatus_set Stephen Hemminger
2018-01-26  2:01 ` [dpdk-dev] [PATCH v8 05/15] net/dpaa2: " Stephen Hemminger
2018-01-26  2:01 ` [dpdk-dev] [PATCH v8 06/15] net/nfp: use rte_eth_linkstatus functions Stephen Hemminger
2018-01-26  2:01 ` [dpdk-dev] [PATCH v8 07/15] net/e1000: use rte_eth_linkstatus helpers Stephen Hemminger
2018-01-26  2:01 ` [dpdk-dev] [PATCH v8 08/15] net/ixgbe: use rte_eth_linkstatus functions Stephen Hemminger
2018-01-26  2:01 ` [dpdk-dev] [PATCH v8 09/15] net/sfc: use new " Stephen Hemminger
2018-01-26  6:31   ` Andrew Rybchenko
2018-01-26  2:01 ` [dpdk-dev] [PATCH v8 10/15] net/i40e: use " Stephen Hemminger
2018-01-26  2:01 ` [dpdk-dev] [PATCH v8 11/15] net/liquidio: use rte_eth_linkstatus_set Stephen Hemminger
2018-01-26  2:01 ` [dpdk-dev] [PATCH v8 12/15] net/thunderx: " Stephen Hemminger
2018-01-26  2:01 ` [dpdk-dev] [PATCH v8 13/15] net/szedata: use _rte_eth_linkstatus_set Stephen Hemminger
2018-01-26  2:01 ` [dpdk-dev] [PATCH v8 14/15] net/octeontx: use rte_eth_linkstatus_set Stephen Hemminger
2018-01-26  2:01 ` [dpdk-dev] [PATCH v8 15/15] net/enic: " Stephen Hemminger
2018-03-15 17:38 ` [dpdk-dev] [PATCH v8 00/15] common linkstatus functions Ferruh Yigit
2018-03-16 10:31   ` 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=20180126020151.979-3-stephen@networkplumber.org \
    --to=stephen@networkplumber.org \
    --cc=dev@dpdk.org \
    /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).