DPDK patches and discussions
 help / color / mirror / Atom feed
* [dpdk-dev] [PATCH v2] net/af_xdp: optimisations to improve packet loss
@ 2020-06-23 14:29 Ciara Loftus
  2020-06-30 14:14 ` Ferruh Yigit
  0 siblings, 1 reply; 2+ messages in thread
From: Ciara Loftus @ 2020-06-23 14:29 UTC (permalink / raw)
  To: dev; +Cc: stephen, magnus.karlsson, qi.z.zhang, Ciara Loftus

This commit makes some changes to the AF_XDP PMD in an effort to improve
its packet loss characteristics.

1. In the case of failed transmission due to inability to reserve a tx
descriptor, the PMD now pulls from the completion ring, issues a syscall
in which the kernel attempts to complete outstanding tx operations, then
tries to reserve the tx descriptor again. Prior to this we dropped the
packet after the syscall and didn't try to re-reserve.

2. During completion ring cleanup, always pull as many entries as possible
from the ring as opposed to the batch size or just how many packets
we're going to attempt to send. Keeping the completion ring emptier should
reduce failed transmissions in the kernel, as the kernel requires space in
the completion ring to successfully tx.

3. Size the fill ring as twice the receive ring size which may help reduce
allocation failures in the driver.

4. Emulate a tx_free_thresh - when the number of available entries in the
completion ring rises above this, we pull from it. The threshold is set to
1k entries.

With these changes, a benchmark which measured the packet rate at which
0.01% packet loss could be reached improved from ~0.1G to ~3Gbps.

Signed-off-by: Ciara Loftus <ciara.loftus@intel.com>
Acked-by: Xiaolong Ye <xiaolong.ye@intel.com>
---

v1->v2:
 - added emulated tx_free_thresh as suggested by Stephen Hemminger

 drivers/net/af_xdp/rte_eth_af_xdp.c | 20 ++++++++++++--------
 1 file changed, 12 insertions(+), 8 deletions(-)

diff --git a/drivers/net/af_xdp/rte_eth_af_xdp.c b/drivers/net/af_xdp/rte_eth_af_xdp.c
index 06124ba789..2d69221c1b 100644
--- a/drivers/net/af_xdp/rte_eth_af_xdp.c
+++ b/drivers/net/af_xdp/rte_eth_af_xdp.c
@@ -396,6 +396,8 @@ kick_tx(struct pkt_tx_queue *txq)
 {
 	struct xsk_umem_info *umem = txq->umem;
 
+	pull_umem_cq(umem, XSK_RING_CONS__DEFAULT_NUM_DESCS);
+
 #if defined(XDP_USE_NEED_WAKEUP)
 	if (xsk_ring_prod__needs_wakeup(&txq->tx))
 #endif
@@ -407,11 +409,9 @@ kick_tx(struct pkt_tx_queue *txq)
 
 			/* 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,
+					     XSK_RING_CONS__DEFAULT_NUM_DESCS);
 		}
-#ifndef XDP_UMEM_UNALIGNED_CHUNK_FLAG
-	pull_umem_cq(umem, ETH_AF_XDP_TX_BATCH_SIZE);
-#endif
 }
 
 #if defined(XDP_UMEM_UNALIGNED_CHUNK_FLAG)
@@ -427,8 +427,10 @@ af_xdp_tx_zc(void *queue, struct rte_mbuf **bufs, uint16_t nb_pkts)
 	uint16_t count = 0;
 	struct xdp_desc *desc;
 	uint64_t addr, offset;
+	uint32_t free_thresh = umem->cq.size >> 1;
 
-	pull_umem_cq(umem, nb_pkts);
+	if (xsk_cons_nb_avail(&umem->cq, free_thresh) >= free_thresh)
+		pull_umem_cq(umem, XSK_RING_CONS__DEFAULT_NUM_DESCS);
 
 	for (i = 0; i < nb_pkts; i++) {
 		mbuf = bufs[i];
@@ -436,7 +438,9 @@ af_xdp_tx_zc(void *queue, struct rte_mbuf **bufs, uint16_t nb_pkts)
 		if (mbuf->pool == umem->mb_pool) {
 			if (!xsk_ring_prod__reserve(&txq->tx, 1, &idx_tx)) {
 				kick_tx(txq);
-				goto out;
+				if (!xsk_ring_prod__reserve(&txq->tx, 1,
+							    &idx_tx))
+					goto out;
 			}
 			desc = xsk_ring_prod__tx_desc(&txq->tx, idx_tx);
 			desc->len = mbuf->pkt_len;
@@ -758,7 +762,7 @@ xsk_umem_info *xdp_umem_configure(struct pmd_internals *internals __rte_unused,
 	struct xsk_umem_info *umem;
 	int ret;
 	struct xsk_umem_config usr_config = {
-		.fill_size = ETH_AF_XDP_DFLT_NUM_DESCS,
+		.fill_size = ETH_AF_XDP_DFLT_NUM_DESCS * 2,
 		.comp_size = ETH_AF_XDP_DFLT_NUM_DESCS,
 		.flags = XDP_UMEM_UNALIGNED_CHUNK_FLAG};
 	void *base_addr = NULL;
@@ -867,7 +871,7 @@ xsk_configure(struct pmd_internals *internals, struct pkt_rx_queue *rxq,
 	struct xsk_socket_config cfg;
 	struct pkt_tx_queue *txq = rxq->pair;
 	int ret = 0;
-	int reserve_size = ETH_AF_XDP_DFLT_NUM_DESCS / 2;
+	int reserve_size = ETH_AF_XDP_DFLT_NUM_DESCS;
 	struct rte_mbuf *fq_bufs[reserve_size];
 
 	rxq->umem = xdp_umem_configure(internals, rxq);
-- 
2.17.1


^ permalink raw reply	[flat|nested] 2+ messages in thread

* Re: [dpdk-dev] [PATCH v2] net/af_xdp: optimisations to improve packet loss
  2020-06-23 14:29 [dpdk-dev] [PATCH v2] net/af_xdp: optimisations to improve packet loss Ciara Loftus
@ 2020-06-30 14:14 ` Ferruh Yigit
  0 siblings, 0 replies; 2+ messages in thread
From: Ferruh Yigit @ 2020-06-30 14:14 UTC (permalink / raw)
  To: Ciara Loftus, dev; +Cc: stephen, magnus.karlsson, qi.z.zhang

On 6/23/2020 3:29 PM, Ciara Loftus wrote:
> This commit makes some changes to the AF_XDP PMD in an effort to improve
> its packet loss characteristics.
> 
> 1. In the case of failed transmission due to inability to reserve a tx
> descriptor, the PMD now pulls from the completion ring, issues a syscall
> in which the kernel attempts to complete outstanding tx operations, then
> tries to reserve the tx descriptor again. Prior to this we dropped the
> packet after the syscall and didn't try to re-reserve.
> 
> 2. During completion ring cleanup, always pull as many entries as possible
> from the ring as opposed to the batch size or just how many packets
> we're going to attempt to send. Keeping the completion ring emptier should
> reduce failed transmissions in the kernel, as the kernel requires space in
> the completion ring to successfully tx.
> 
> 3. Size the fill ring as twice the receive ring size which may help reduce
> allocation failures in the driver.
> 
> 4. Emulate a tx_free_thresh - when the number of available entries in the
> completion ring rises above this, we pull from it. The threshold is set to
> 1k entries.
> 
> With these changes, a benchmark which measured the packet rate at which
> 0.01% packet loss could be reached improved from ~0.1G to ~3Gbps.
> 
> Signed-off-by: Ciara Loftus <ciara.loftus@intel.com>
> Acked-by: Xiaolong Ye <xiaolong.ye@intel.com>

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

^ permalink raw reply	[flat|nested] 2+ messages in thread

end of thread, other threads:[~2020-06-30 14:14 UTC | newest]

Thread overview: 2+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2020-06-23 14:29 [dpdk-dev] [PATCH v2] net/af_xdp: optimisations to improve packet loss Ciara Loftus
2020-06-30 14:14 ` Ferruh Yigit

DPDK patches and discussions

This inbox may be cloned and mirrored by anyone:

	git clone --mirror https://inbox.dpdk.org/dev/0 dev/git/0.git

	# If you have public-inbox 1.1+ installed, you may
	# initialize and index your mirror using the following commands:
	public-inbox-init -V2 dev dev/ https://inbox.dpdk.org/dev \
		dev@dpdk.org
	public-inbox-index dev

Example config snippet for mirrors.
Newsgroup available over NNTP:
	nntp://inbox.dpdk.org/inbox.dpdk.dev


AGPL code for this site: git clone https://public-inbox.org/public-inbox.git