Skip to contentSkip to conversation

net/af_xdp: support need wakeup feature

PATCH Superseded Xiaolong Ye 4 participants 12 messages Jun 17, 2019 – Jun 28, 2019 Network drivers

lore
Share & tools
lore
#0 lore raw

PATCH 1/1 net/af_xdp: support need wakeup feature

This patch adds a new devarg to support the need_wakeup flag for Tx and fill rings, when this flag is set by the driver, it means that the userspace application has to explicitly wake up the kernel Rx or kernel Tx processing by issuing a syscall. Poll() can wake up both and sendto() or its alternatives will wake up Tx processing only.

This feature is to provide efficient support for case that application and driver are executing on the same core.

---

Original busy poll feature has morphed into need_wakeup flag in kernel side, the main purpose is the same, that is to support both application and driver executing on the same core efficiently.

kernel side patchset can be found at netdev mailing list.
https://lore.kernel.org/netdev/CAJ8uoz2szX=+JXXAMyuVmvSsMXZuDqp6a8rjDQpTioxbZwxFmQ@mail.gmail.com/T/#t

It is targeted for v5.3

 drivers/net/af_xdp/rte_eth_af_xdp.c | 51 ++++++++++++++++++++---------
 1 file changed, 36 insertions(+), 15 deletions(-)
drivers/net/af_xdp/rte_eth_af_xdp.c +36 -15
index c638d9227..198b00147 100644
@@ -5,6 +5,7 @@
 #include <errno.h>
 #include <stdlib.h>
 #include <string.h>
+#include <poll.h>
 #include <netinet/in.h>
 #include <net/if.h>
 #include <sys/socket.h>
@@ -90,6 +91,7 @@ struct pkt_rx_queue {
 	struct rx_stats stats;
 
 	struct pkt_tx_queue *pair;
+	struct pollfd fds[1];
 	int xsk_queue_idx;
 };
 
@@ -117,6 +119,7 @@ struct pmd_internals {
 	int combined_queue_cnt;
 
 	int pmd_zc;
+	int need_wakeup;
 	struct rte_ether_addr eth_addr;
 
 	struct pkt_rx_queue *rx_queues;
@@ -127,12 +130,14 @@ struct pmd_internals {
 #define ETH_AF_XDP_START_QUEUE_ARG		"start_queue"
 #define ETH_AF_XDP_QUEUE_COUNT_ARG		"queue_count"
 #define ETH_AF_XDP_PMD_ZC_ARG			"pmd_zero_copy"
+#define ETH_AF_XDP_NEED_WAKEUP_ARG		"need_wakeup"
 
 static const char * const valid_arguments[] = {
 	ETH_AF_XDP_IFACE_ARG,
 	ETH_AF_XDP_START_QUEUE_ARG,
 	ETH_AF_XDP_QUEUE_COUNT_ARG,
 	ETH_AF_XDP_PMD_ZC_ARG,
+	ETH_AF_XDP_NEED_WAKEUP_ARG,
 	NULL
 };
 
@@ -206,8 +211,12 @@ eth_af_xdp_rx(void *queue, struct rte_mbuf **bufs, uint16_t nb_pkts)
 		return 0;
 
 	rcvd = xsk_ring_cons__peek(rx, nb_pkts, &idx_rx);
-	if (rcvd == 0)
+	if (rcvd == 0) {
+		if (xsk_ring_prod__needs_wakeup(fq))
+			(void)poll(rxq->fds, 1, 1000);
+
 		goto out;
+	}
 
 	if (xsk_prod_nb_free(fq, free_thresh) >= free_thresh)
 		(void)reserve_fill_queue(umem, ETH_AF_XDP_RX_BATCH_SIZE);
@@ -279,16 +288,17 @@ kick_tx(struct pkt_tx_queue *txq)
 {
 	struct xsk_umem_info *umem = txq->pair->umem;
 
-	while (send(xsk_socket__fd(txq->pair->xsk), NULL,
-		      0, MSG_DONTWAIT) < 0) {
-		/* some thing unexpected */
-		if (errno != EBUSY && errno != EAGAIN && errno != EINTR)
-			break;
+	if (xsk_ring_prod__needs_wakeup(&txq->tx))
+		while (send(xsk_socket__fd(txq->pair->xsk), NULL,
+			    0, MSG_DONTWAIT) < 0) {
+			/* some thing unexpected */
+			if (errno != EBUSY && errno != EAGAIN && errno != EINTR)
+				break;
 
-		/* pull from completion queue to leave more space */
-		if (errno == EAGAIN)
-			pull_umem_cq(umem, ETH_AF_XDP_TX_BATCH_SIZE);
-	}
+			/* pull from completion queue to leave more space */
+			if (errno == EAGAIN)
+				pull_umem_cq(umem, ETH_AF_XDP_TX_BATCH_SIZE);
+		}
 	pull_umem_cq(umem, ETH_AF_XDP_TX_BATCH_SIZE);
 }
 
@@ -621,7 +631,7 @@ xsk_configure(struct pmd_internals *internals, struct pkt_rx_queue *rxq,
 	cfg.tx_size = ring_size;
 	cfg.libbpf_flags = 0;
 	cfg.xdp_flags = XDP_FLAGS_UPDATE_IF_NOEXIST;
-	cfg.bind_flags = 0;
+	cfg.bind_flags = internals->need_wakeup ? XDP_USE_NEED_WAKEUP : 0;
 	ret = xsk_socket__create(&rxq->xsk, internals->if_name,
 			rxq->xsk_queue_idx, rxq->umem->umem, &rxq->rx,
 			&txq->tx, &cfg);
@@ -683,6 +693,9 @@ eth_rx_queue_setup(struct rte_eth_dev *dev,
 		goto err;
 	}
 
+	rxq->fds[0].fd = xsk_socket__fd(rxq->xsk);
+	rxq->fds[0].events = POLLIN;
+
 	rxq->umem->pmd_zc = internals->pmd_zc;
 
 	dev->data->rx_queues[rx_queue_id] = rxq;
@@ -856,7 +869,7 @@ xdp_get_channels_info(const char *if_name, int *max_queues,
 
 static int
 parse_parameters(struct rte_kvargs *kvlist, char *if_name, int *start_queue,
-			int *queue_cnt, int *pmd_zc)
+		 int *queue_cnt, int *pmd_zc, int *need_wakeup)
 {
 	int ret;
 
@@ -882,6 +895,9 @@ parse_parameters(struct rte_kvargs *kvlist, char *if_name, int *start_queue,
 	if (ret < 0)
 		goto free_kvlist;
 
+	ret = rte_kvargs_process(kvlist, ETH_AF_XDP_NEED_WAKEUP_ARG,
+				 &parse_integer_arg, need_wakeup);
+
 free_kvlist:
 	rte_kvargs_free(kvlist);
 	return ret;
@@ -919,7 +935,7 @@ get_iface_info(const char *if_name,
 
 static struct rte_eth_dev *
 init_internals(struct rte_vdev_device *dev, const char *if_name,
-			int start_queue_idx, int queue_cnt, int pmd_zc)
+	       int start_queue_idx, int queue_cnt, int pmd_zc, int need_wakeup)
 {
 	const char *name = rte_vdev_device_name(dev);
 	const unsigned int numa_node = dev->device.numa_node;
@@ -935,6 +951,7 @@ init_internals(struct rte_vdev_device *dev, const char *if_name,
 	internals->start_queue_idx = start_queue_idx;
 	internals->queue_cnt = queue_cnt;
 	internals->pmd_zc = pmd_zc;
+	internals->need_wakeup = !!need_wakeup;
 	strlcpy(internals->if_name, if_name, IFNAMSIZ);
 
 	if (xdp_get_channels_info(if_name, &internals->max_queue_cnt,
@@ -993,6 +1010,9 @@ init_internals(struct rte_vdev_device *dev, const char *if_name,
 	if (internals->pmd_zc)
 		AF_XDP_LOG(INFO, "Zero copy between umem and mbuf enabled.\n");
 
+	if (internals->need_wakeup)
+		AF_XDP_LOG(INFO, "need_wakeup feature is explicitly turned on.\n");
+
 	return eth_dev;
 
 err_free_tx:
@@ -1014,6 +1034,7 @@ rte_pmd_af_xdp_probe(struct rte_vdev_device *dev)
 	struct rte_eth_dev *eth_dev = NULL;
 	const char *name;
 	int pmd_zc = 0;
+	int need_wakeup = 0;
 
 	AF_XDP_LOG(INFO, "Initializing pmd_af_xdp for %s\n",
 		rte_vdev_device_name(dev));
@@ -1041,7 +1062,7 @@ rte_pmd_af_xdp_probe(struct rte_vdev_device *dev)
 		dev->device.numa_node = rte_socket_id();
 
 	if (parse_parameters(kvlist, if_name, &xsk_start_queue_idx,
-			     &xsk_queue_cnt, &pmd_zc) < 0) {
+			     &xsk_queue_cnt, &pmd_zc, &need_wakeup) < 0) {
 		AF_XDP_LOG(ERR, "Invalid kvargs value\n");
 		return -EINVAL;
 	}
@@ -1052,7 +1073,7 @@ rte_pmd_af_xdp_probe(struct rte_vdev_device *dev)
 	}
 
 	eth_dev = init_internals(dev, if_name, xsk_start_queue_idx,
-					xsk_queue_cnt, pmd_zc);
+				 xsk_queue_cnt, pmd_zc, need_wakeup);
 	if (eth_dev == NULL) {
 		AF_XDP_LOG(ERR, "Failed to init internals\n");
 		return -1;
Signature · 1 line
2.17.1
#1 lore raw
Quoted 16 lines · Xiaolong Ye

This patch adds a new devarg to support the need_wakeup flag for Tx and fill rings, when this flag is set by the driver, it means that the userspace application has to explicitly wake up the kernel Rx or kernel Tx processing by issuing a syscall. Poll() can wake up both and sendto() or its alternatives will wake up Tx processing only.

This feature is to provide efficient support for case that application and driver are executing on the same core.

---

Original busy poll feature has morphed into need_wakeup flag in kernel side, the main purpose is the same, that is to support both application and driver executing on the same core efficiently.

Quoted 6 lines

kernel side patchset can be found at netdev mailing list.

https://lore.kernel.org/netdev/CAJ8uoz2szX=+JXXAMyuVmvSsMXZuDqp6a8rjDQpTioxbZwxFmQ@mail.gmail.com/T/#t

It is targeted for v5.3

- Is this really optional? Adding too many options is just a nightmare later...

- I suppose this will break compilation with kernels that have af_xdp but are < 5.3.

Signature · 1 line
David Marchand
#2 lore raw
Quoted 20 lines · David Marchand
Quoted 18 lines · Xiaolong Ye

This patch adds a new devarg to support the need_wakeup flag for Tx and fill rings, when this flag is set by the driver, it means that the userspace application has to explicitly wake up the kernel Rx or kernel Tx processing by issuing a syscall. Poll() can wake up both and sendto() or its alternatives will wake up Tx processing only.

This feature is to provide efficient support for case that application and driver are executing on the same core.

---

Original busy poll feature has morphed into need_wakeup flag in kernel side, the main purpose is the same, that is to support both application and driver executing on the same core efficiently.

kernel side patchset can be found at netdev mailing list.

Quoted 7 lines
Quoted 4 lines

https://lore.kernel.org/netdev/CAJ8uoz2szX=+JXXAMyuVmvSsMXZuDqp6a8rjDQpTioxbZwxFmQ@mail.gmail.com/T/#t

It is targeted for v5.3

- Is this really optional? Adding too many options is just a nightmare later...

Hmm, I think we can remove this option and alway turn the need_wakeup flag on since it provides better performance for 1 core case and doesn't downgrage the 2 core case performance.

Quoted 3 lines

- I suppose this will break compilation with kernels that have af_xdp but are < 5.3.

Yes, that is true. It will break the compilation with early kernel, I feel it's sort of common issue, we enable some features in dpdk that's based on kernel features, then kernel side features keep evolving, we need to keep the pace, but it will hurt the compatiblity with the old kernel.

What's dpdk's convention for handling this kind of case? Add some notes in doc to reminder the prerequisite or use the KERNEL_VERSION macro in code?

Thanks,
Xiaolong

Quoted 4 lines
Signature · 2 lines
-- 
David Marchand
#3 lore raw
Quoted 51 lines · Ye Xiaolong
Quoted 1 line · David Marchand

On Mon, Jun 17, 2019 at 9:42 AM Xiaolong Ye <[email protected]>

Quoted 4 lines · wrote:
Quoted 4 lines

This patch adds a new devarg to support the need_wakeup flag for Tx and fill rings, when this flag is set by the driver, it means that the userspace application has to explicitly wake up the kernel Rx or kernel

Tx

Quoted 4 lines
Quoted 4 lines

processing by issuing a syscall. Poll() can wake up both and sendto() or its alternatives will wake up Tx processing only.

This feature is to provide efficient support for case that application

and

Quoted 5 lines
Quoted 5 lines

driver are executing on the same core.

---

Quoted 7 lines
Quoted 7 lines

Original busy poll feature has morphed into need_wakeup flag in kernel side, the main purpose is the same, that is to support both application and driver executing on the same core efficiently.

kernel side patchset can be found at netdev mailing list.

https://lore.kernel.org/netdev/CAJ8uoz2szX=+JXXAMyuVmvSsMXZuDqp6a8rjDQpTioxbZwxFmQ@mail.gmail.com/T/#t

Quoted 6 lines
Quoted 3 lines

It is targeted for v5.3

- Is this really optional? Adding too many options is just a nightmare later...

Hmm, I think we can remove this option and alway turn the need_wakeup flag on
since it provides better performance for 1 core case and doesn't downgrage the
2 core case performance.

Quoted 3 lines

- I suppose this will break compilation with kernels that have af_xdp but are < 5.3.

Yes, that is true. It will break the compilation with early kernel, I feel it's
sort of common issue, we enable some features in dpdk that's based on kernel
features, then kernel side features keep evolving, we need to keep the pace,
but it will hurt the compatiblity with the old kernel.

Quoted 4 lines

What's dpdk's convention for handling this kind of case? Add some notes in doc
to reminder the prerequisite or use the KERNEL_VERSION macro in code?

Rather than a kernel version, you can check that XDP_USE_NEED_WAKEUP is defined (present in the uapi kernel header).

Signature · 1 line
David Marchand
#4 lore raw
Quoted 53 lines · David Marchand
Quoted 51 lines · Ye Xiaolong
Quoted 1 line · David Marchand

On Mon, Jun 17, 2019 at 9:42 AM Xiaolong Ye <[email protected]>

Quoted 4 lines · wrote:
Quoted 4 lines

This patch adds a new devarg to support the need_wakeup flag for Tx and fill rings, when this flag is set by the driver, it means that the userspace application has to explicitly wake up the kernel Rx or kernel

Tx

Quoted 4 lines
Quoted 4 lines

processing by issuing a syscall. Poll() can wake up both and sendto() or its alternatives will wake up Tx processing only.

This feature is to provide efficient support for case that application

and

Quoted 5 lines
Quoted 5 lines

driver are executing on the same core.

---

Quoted 7 lines
Quoted 7 lines

Original busy poll feature has morphed into need_wakeup flag in kernel side, the main purpose is the same, that is to support both application and driver executing on the same core efficiently.

kernel side patchset can be found at netdev mailing list.

https://lore.kernel.org/netdev/CAJ8uoz2szX=+JXXAMyuVmvSsMXZuDqp6a8rjDQpTioxbZwxFmQ@mail.gmail.com/T/#t

Quoted 6 lines
Quoted 3 lines

It is targeted for v5.3

- Is this really optional? Adding too many options is just a nightmare later...

Hmm, I think we can remove this option and alway turn the need_wakeup flag on
since it provides better performance for 1 core case and doesn't downgrage the
2 core case performance.

Quoted 3 lines

- I suppose this will break compilation with kernels that have af_xdp but are < 5.3.

Yes, that is true. It will break the compilation with early kernel, I feel it's
sort of common issue, we enable some features in dpdk that's based on kernel
features, then kernel side features keep evolving, we need to keep the pace,
but it will hurt the compatiblity with the old kernel.

Quoted 7 lines
Quoted 4 lines

What's dpdk's convention for handling this kind of case? Add some notes in doc
to reminder the prerequisite or use the KERNEL_VERSION macro in code?

Rather than a kernel version, you can check that XDP_USE_NEED_WAKEUP is defined (present in the uapi kernel header).

Sounds better, will try.

Thanks,
Xiaolong

Quoted 4 lines
Signature · 2 lines
-- 
David Marchand
#5 lore raw
In reply to David Marchand
Quoted 53 lines · David Marchand
Quoted 51 lines · Ye Xiaolong
Quoted 1 line · David Marchand

On Mon, Jun 17, 2019 at 9:42 AM Xiaolong Ye <[email protected]>

Quoted 4 lines · wrote:
Quoted 4 lines

This patch adds a new devarg to support the need_wakeup flag for Tx and fill rings, when this flag is set by the driver, it means that the userspace application has to explicitly wake up the kernel Rx or kernel

Tx

Quoted 4 lines
Quoted 4 lines

processing by issuing a syscall. Poll() can wake up both and sendto() or its alternatives will wake up Tx processing only.

This feature is to provide efficient support for case that application

and

Quoted 5 lines
Quoted 5 lines

driver are executing on the same core.

---

Quoted 7 lines
Quoted 7 lines

Original busy poll feature has morphed into need_wakeup flag in kernel side, the main purpose is the same, that is to support both application and driver executing on the same core efficiently.

kernel side patchset can be found at netdev mailing list.

https://lore.kernel.org/netdev/CAJ8uoz2szX=+JXXAMyuVmvSsMXZuDqp6a8rjDQpTioxbZwxFmQ@mail.gmail.com/T/#t

Quoted 6 lines
Quoted 3 lines

It is targeted for v5.3

- Is this really optional? Adding too many options is just a nightmare later...

Hmm, I think we can remove this option and alway turn the need_wakeup flag on
since it provides better performance for 1 core case and doesn't downgrage the
2 core case performance.

Quoted 3 lines

- I suppose this will break compilation with kernels that have af_xdp but are < 5.3.

Yes, that is true. It will break the compilation with early kernel, I feel it's
sort of common issue, we enable some features in dpdk that's based on kernel
features, then kernel side features keep evolving, we need to keep the pace,
but it will hurt the compatiblity with the old kernel.

Quoted 8 lines
Quoted 4 lines

What's dpdk's convention for handling this kind of case? Add some notes in doc
to reminder the prerequisite or use the KERNEL_VERSION macro in code?

Rather than a kernel version, you can check that XDP_USE_NEED_WAKEUP is defined (present in the uapi kernel header).

+1 for this.

Also, since AF_XDP is still fairly new with ongoing development on the kernel side, I think it is reasonable to limit out PMD to only working with sufficiently updated kernels. Hopefully in 6 months or so, the feature set we need should be locked down and we can specify a fixed baseline requirement.

/Bruce

#6 lore raw

PATCH v2 1/1 net/af_xdp: support need wakeup feature

In reply to Xiaolong Ye

This patch enables need_wakeup flag for Tx and fill rings, when this flag is set by the driver, it means that the userspace application has to explicitly wake up the kernel Rx or kernel Tx processing by issuing a syscall. Poll() can wake up both and sendto() or its alternatives will wake up Tx processing only.

This feature is to provide efficient support for case that application and driver are executing on the same core.

---

v2 changes:

1. remove need_wakeup devarg to make need_wakeup feature enabled unconditionally.
2. add conditional compilation directive to avoid breaking build with kernel which doesn't support need_wakeup feature yet.

Note:

Original busy poll feature has morphed into need_wakeup flag in kernel side, the main purpose is the same, that is to support both application and driver executing on the same core efficiently.

kernel side patchset can be found at netdev mailing list.
https://lore.kernel.org/netdev/CAJ8uoz2szX=+JXXAMyuVmvSsMXZuDqp6a8rjDQpTioxbZwxFmQ@mail.gmail.com/T/#t

It is targeted for v5.3

Cc David Marchand <[email protected]>
Cc Bruce Richardson <[email protected]>
 drivers/net/af_xdp/rte_eth_af_xdp.c | 41 +++++++++++++++++++++--------
 1 file changed, 30 insertions(+), 11 deletions(-)
drivers/net/af_xdp/rte_eth_af_xdp.c +30 -11
index c638d9227..5ce90a760 100644
@@ -5,6 +5,7 @@
 #include <errno.h>
 #include <stdlib.h>
 #include <string.h>
+#include <poll.h>
 #include <netinet/in.h>
 #include <net/if.h>
 #include <sys/socket.h>
@@ -90,6 +91,7 @@ struct pkt_rx_queue {
 	struct rx_stats stats;
 
 	struct pkt_tx_queue *pair;
+	struct pollfd fds[1];
 	int xsk_queue_idx;
 };
 
@@ -206,8 +208,14 @@ eth_af_xdp_rx(void *queue, struct rte_mbuf **bufs, uint16_t nb_pkts)
 		return 0;
 
 	rcvd = xsk_ring_cons__peek(rx, nb_pkts, &idx_rx);
-	if (rcvd == 0)
+	if (rcvd == 0) {
+#if defined(XDP_USE_NEED_WAKEUP)
+		if (xsk_ring_prod__needs_wakeup(fq))
+			(void)poll(rxq->fds, 1, 1000);
+#endif
+
 		goto out;
+	}
 
 	if (xsk_prod_nb_free(fq, free_thresh) >= free_thresh)
 		(void)reserve_fill_queue(umem, ETH_AF_XDP_RX_BATCH_SIZE);
@@ -279,16 +287,19 @@ kick_tx(struct pkt_tx_queue *txq)
 {
 	struct xsk_umem_info *umem = txq->pair->umem;
 
-	while (send(xsk_socket__fd(txq->pair->xsk), NULL,
-		      0, MSG_DONTWAIT) < 0) {
-		/* some thing unexpected */
-		if (errno != EBUSY && errno != EAGAIN && errno != EINTR)
-			break;
-
-		/* pull from completion queue to leave more space */
-		if (errno == EAGAIN)
-			pull_umem_cq(umem, ETH_AF_XDP_TX_BATCH_SIZE);
-	}
+#if defined(XDP_USE_NEED_WAKEUP)
+	if (xsk_ring_prod__needs_wakeup(&txq->tx))
+#endif
+		while (send(xsk_socket__fd(txq->pair->xsk), NULL,
+			    0, MSG_DONTWAIT) < 0) {
+			/* some thing unexpected */
+			if (errno != EBUSY && errno != EAGAIN && errno != EINTR)
+				break;
+
+			/* pull from completion queue to leave more space */
+			if (errno == EAGAIN)
+				pull_umem_cq(umem, ETH_AF_XDP_TX_BATCH_SIZE);
+		}
 	pull_umem_cq(umem, ETH_AF_XDP_TX_BATCH_SIZE);
 }
 
@@ -622,6 +633,11 @@ xsk_configure(struct pmd_internals *internals, struct pkt_rx_queue *rxq,
 	cfg.libbpf_flags = 0;
 	cfg.xdp_flags = XDP_FLAGS_UPDATE_IF_NOEXIST;
 	cfg.bind_flags = 0;
+
+#if defined(XDP_USE_NEED_WAKEUP)
+	cfg.bind_flags |= XDP_USE_NEED_WAKEUP;
+#endif
+
 	ret = xsk_socket__create(&rxq->xsk, internals->if_name,
 			rxq->xsk_queue_idx, rxq->umem->umem, &rxq->rx,
 			&txq->tx, &cfg);
@@ -683,6 +699,9 @@ eth_rx_queue_setup(struct rte_eth_dev *dev,
 		goto err;
 	}
 
+	rxq->fds[0].fd = xsk_socket__fd(rxq->xsk);
+	rxq->fds[0].events = POLLIN;
+
 	rxq->umem->pmd_zc = internals->pmd_zc;
 
 	dev->data->rx_queues[rx_queue_id] = rxq;
Signature · 1 line
2.17.1
#7 lore raw

PATCH v3 1/1 net/af_xdp: support need wakeup feature

In reply to Xiaolong Ye

This patch enables need_wakeup flag for Tx and fill rings, when this flag is set by the driver, it means that the userspace application has to explicitly wake up the kernel Rx or kernel Tx processing by issuing a syscall. Poll() can wake up both and sendto() or its alternatives will wake up Tx processing only.

This feature is to provide efficient support for case that application and driver executing on the same core.

---
v3 changes:

1. add introduction in 19.08 release note

v2 changes:

1. remove need_wakeup devarg to make need_wakeup feature enabled unconditionally.
2. add conditional compilation directive to avoid breaking build with kernel which doesn't support need_wakeup feature yet.

Note:

Original busy poll feature has morphed into need_wakeup flag in kernel side, the main purpose is the same, that is to support both application and driver executing on the same core efficiently.

 doc/guides/rel_notes/release_19_08.rst |  2 ++
 drivers/net/af_xdp/rte_eth_af_xdp.c    | 41 +++++++++++++++++++-------
 2 files changed, 32 insertions(+), 11 deletions(-)
doc/guides/rel_notes/release_19_08.rst +2 -0
index 8c3932d06..49656c3cc 100644
@@ -87,6 +87,8 @@ New Features
     high performance
   * Added multi-queue support to allow one af_xdp vdev with multiple netdev
     queues
+  * Enabled need_wakeup feature which can provide efficient support for case
+    that application and driver executing on the same core.
 
 
 Removed Items
drivers/net/af_xdp/rte_eth_af_xdp.c +30 -11

78 lines not on this page · raw message

Signature · 1 line
2.17.1
#8 lore raw
Quoted 22 lines · Xiaolong Ye

This patch enables need_wakeup flag for Tx and fill rings, when this flag is set by the driver, it means that the userspace application has to explicitly wake up the kernel Rx or kernel Tx processing by issuing a syscall. Poll() can wake up both and sendto() or its alternatives will wake up Tx processing only.

This feature is to provide efficient support for case that application and driver executing on the same core.

---
v3 changes:

1. add introduction in 19.08 release note

v2 changes:

1. remove need_wakeup devarg to make need_wakeup feature enabled unconditionally.
2. add conditional compilation directive to avoid breaking build with kernel which doesn't support need_wakeup feature yet.

Quoted 7 lines

Note:

Original busy poll feature has morphed into need_wakeup flag in kernel side, the main purpose is the same, that is to support both application and driver executing on the same core efficiently.

 doc/guides/rel_notes/release_19_08.rst |  2 ++
 drivers/net/af_xdp/rte_eth_af_xdp.c    | 41 +++++++++++++++++++-------

Hi Xiaolong,

Can you please document this feature in af_xdp documentation? And if there specific kernel version requirement for this, can you please add this information too?

#9 lore raw

Hi, Ferruh

Quoted 31 lines · Ferruh Yigit
Quoted 22 lines · Xiaolong Ye

This patch enables need_wakeup flag for Tx and fill rings, when this flag is set by the driver, it means that the userspace application has to explicitly wake up the kernel Rx or kernel Tx processing by issuing a syscall. Poll() can wake up both and sendto() or its alternatives will wake up Tx processing only.

This feature is to provide efficient support for case that application and driver executing on the same core.

---
v3 changes:

1. add introduction in 19.08 release note

v2 changes:

1. remove need_wakeup devarg to make need_wakeup feature enabled unconditionally.
2. add conditional compilation directive to avoid breaking build with kernel which doesn't support need_wakeup feature yet.

Quoted 7 lines

Note:

Original busy poll feature has morphed into need_wakeup flag in kernel side, the main purpose is the same, that is to support both application and driver executing on the same core efficiently.

 doc/guides/rel_notes/release_19_08.rst |  2 ++
 drivers/net/af_xdp/rte_eth_af_xdp.c    | 41 +++++++++++++++++++-------
Quoted 5 lines

Hi Xiaolong,

Can you please document this feature in af_xdp documentation? And if there specific kernel version requirement for this, can you please add this information too?

Got it, will do in next version.

Thanks,
Xiaolong

Quoted 1 line
#10 lore raw

PATCH v4 1/1 net/af_xdp: support need wakeup feature

In reply to Xiaolong Ye

This patch enables need_wakeup flag for Tx and fill rings, when this flag is set by the driver, it means that the userspace application has to explicitly wake up the kernel Rx or kernel Tx processing by issuing a syscall. Poll() can wake up both and sendto() or its alternatives will wake up Tx processing only.

This feature is to provide efficient support for case that application and driver executing on the same core.

---

v4 changes:

1. document need_wakeup feature in the af_xdp.rst

v3 changes:

1. add introduction in 19.08 release note

v2 changes:

1. remove need_wakeup devarg to make need_wakeup feature enabled unconditionally.
2. add conditional compilation directive to avoid breaking build with kernel which doesn't support need_wakeup feature yet.

Note:

Original busy poll feature has morphed into need_wakeup flag in kernel side, the main purpose is the same, that is to support both application and driver executing on the same core efficiently.

 doc/guides/nics/af_xdp.rst             |  7 +++++
 doc/guides/rel_notes/release_19_08.rst |  2 ++
 drivers/net/af_xdp/rte_eth_af_xdp.c    | 41 +++++++++++++++++++-------
 3 files changed, 39 insertions(+), 11 deletions(-)
doc/guides/nics/af_xdp.rst +7 -0
index 18defcda3..c99c4c715 100644
@@ -21,6 +21,12 @@ be added later.
 Note that MTU of AF_XDP PMD is limited due to XDP lacks support for
 fragmentation.
 
+AF_XDP PMD enables need_wakeup flag by default if it is supported. This
+need_wakeup feature is used to support executing application and driver on the
+same core efficiently. This feature not only has a large positive performance
+impact for the one core case, but also does not degrade 2 core performance and
+actually improves it for Tx heavy workloads.
+
 Options
 -------
 
@@ -41,6 +47,7 @@ This is a Linux-specific PMD, thus the following prerequisites apply:
    User can install libbpf via `make install_lib` && `make install_headers` in
    <kernel src tree>/tools/lib/bpf;
 *  A Kernel bound interface to attach to;
+*  For need_wakeup feature, it requires kernel version later than v5.3-rc1;
 
 Set up an af_xdp interface
 -----------------------------
doc/guides/rel_notes/release_19_08.rst +2 -0
index 3da266705..1c9efcbe7 100644
@@ -93,6 +93,8 @@ New Features
     high performance
   * Added multi-queue support to allow one af_xdp vdev with multiple netdev
     queues
+  * Enabled need_wakeup feature which can provide efficient support for case
+    that application and driver executing on the same core.
 
 * **Updated telemetry library for global metrics support.**
 
drivers/net/af_xdp/rte_eth_af_xdp.c +30 -11

78 lines not on this page · raw message

Signature · 1 line
2.17.1
Reviewed-by on v4 1/1 Applied
#11 lore raw
Quoted 10 lines · Xiaolong Ye

This patch enables need_wakeup flag for Tx and fill rings, when this flag is set by the driver, it means that the userspace application has to explicitly wake up the kernel Rx or kernel Tx processing by issuing a syscall. Poll() can wake up both and sendto() or its alternatives will wake up Tx processing only.

This feature is to provide efficient support for case that application and driver executing on the same core.

Applied to dpdk-next-net/master, thanks.