DPDK patches and discussions
 help / color / mirror / Atom feed
From: Marat Khalili <marat.khalili@huawei.com>
To: Stephen Hemminger <stephen@networkplumber.org>,
	"dev@dpdk.org" <dev@dpdk.org>
Subject: RE: [PATCH 08/12] net/pcap: optimize calculation of receive timestamp
Date: Wed, 7 Jan 2026 10:58:53 +0000	[thread overview]
Message-ID: <7a43c70576054749920c16e8c2ae6320@huawei.com> (raw)
In-Reply-To: <20260106182823.192350-9-stephen@networkplumber.org>

> -----Original Message-----
> From: Stephen Hemminger <stephen@networkplumber.org>
> Sent: Tuesday 6 January 2026 18:27
> To: dev@dpdk.org
> Cc: Stephen Hemminger <stephen@networkplumber.org>
> Subject: [PATCH 08/12] net/pcap: optimize calculation of receive timestamp
> 
> Avoid doing slow instructions in receive path when calculating
> timestamp. Give all packets in the same rx burst the same timestamp.
> 
> Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
> ---
>  drivers/net/pcap/pcap_ethdev.c | 23 +++++++++++++----------
>  1 file changed, 13 insertions(+), 10 deletions(-)
> 
> diff --git a/drivers/net/pcap/pcap_ethdev.c b/drivers/net/pcap/pcap_ethdev.c
> index 175d6998f9..e283fb3787 100644
> --- a/drivers/net/pcap/pcap_ethdev.c
> +++ b/drivers/net/pcap/pcap_ethdev.c
> @@ -20,6 +20,7 @@
>  #include <bus_vdev_driver.h>
>  #include <rte_os_shim.h>
>  #include <rte_time.h>
> +#include <rte_reciprocal.h>
> 
>  #include "pcap_osdep.h"
> 
> @@ -41,7 +42,7 @@
> 
>  static struct timespec start_time;
>  static uint64_t start_cycles;
> -static uint64_t hz;
> +static struct rte_reciprocal_u64 hz_inv;
> 
>  static uint64_t timestamp_rx_dynflag;
>  static int timestamp_dynfield_offset = -1;
> @@ -362,8 +363,6 @@ eth_null_rx(void *queue __rte_unused,
>  	return 0;
>  }
> 
> -#define NSEC_PER_SEC	1000000000L
> -
>  /*
>   * This function stores nanoseconds in `tv_usec` field of `struct timeval`,
>   * because `ts` goes directly to nanosecond-precision dump.
> @@ -374,8 +373,10 @@ calculate_timestamp(struct timeval *ts) {
>  	struct timespec cur_time;
> 
>  	cycles = rte_get_timer_cycles() - start_cycles;
> -	cur_time.tv_sec = cycles / hz;
> -	cur_time.tv_nsec = (cycles % hz) * NSEC_PER_SEC / hz;
> +	cur_time.tv_sec = rte_reciprocal_divide_u64(cycles, &hz_inv);
> +	/* compute remainder */
> +	cycles -= cur_time.tv_sec * rte_get_timer_hz();

Can be made faster and safer by caching rte_get_timer_hz() result in current translation unit.

> +	cur_time.tv_nsec = rte_reciprocal_divide_u64(cycles * NS_PER_S, &hz_inv);
> 
>  	ts->tv_sec = start_time.tv_sec + cur_time.tv_sec;
>  	ts->tv_usec = start_time.tv_nsec + cur_time.tv_nsec;
> @@ -394,6 +395,7 @@ eth_pcap_tx_dumper(void *queue, struct rte_mbuf **bufs, uint16_t nb_pkts)
>  	unsigned int i;
>  	struct pmd_process_private *pp;
>  	struct pcap_tx_queue *dumper_q = queue;
> +	struct pcap_pkthdr header;
>  	uint16_t num_tx = 0;
>  	uint32_t tx_bytes = 0;
>  	pcap_dumper_t *dumper;
> @@ -406,13 +408,14 @@ eth_pcap_tx_dumper(void *queue, struct rte_mbuf **bufs, uint16_t nb_pkts)
>  	if (unlikely(dumper == NULL || nb_pkts == 0))
>  		return 0;
> 
> -	/* writes the nb_pkts packets to the previously opened pcap file
> -	 * dumper */
> +	/* all packets in burst have same timestamp */
> +	calculate_timestamp(&header.ts);
> +
> +	/* writes the nb_pkts packets to the previously opened pcap file dumper */
>  	for (i = 0; i < nb_pkts; i++) {
>  		struct rte_mbuf *mbuf = bufs[i];
>  		size_t len = rte_pktmbuf_pkt_len(mbuf);
>  		uint8_t temp_data[RTE_ETH_PCAP_SNAPLEN];
> -		struct pcap_pkthdr header;
> 
>  		if (unlikely(len > mtu))
>  			continue;
> @@ -420,7 +423,6 @@ eth_pcap_tx_dumper(void *queue, struct rte_mbuf **bufs, uint16_t nb_pkts)
>  		if ((mbuf->ol_flags & RTE_MBUF_F_TX_VLAN) && rte_vlan_insert(&mbuf))
>  			continue;
> 
> -		calculate_timestamp(&header.ts);
>  		header.len = len;
>  		header.caplen = len;
> 
> @@ -1530,7 +1532,8 @@ pmd_pcap_probe(struct rte_vdev_device *dev)
> 
>  	timespec_get(&start_time, TIME_UTC);
>  	start_cycles = rte_get_timer_cycles();
> -	hz = rte_get_timer_hz();
> +
> +	hz_inv = rte_reciprocal_value_u64(rte_get_timer_hz());
> 
>  	if (rte_eal_process_type() == RTE_PROC_SECONDARY) {
>  		eth_dev = rte_eth_dev_attach_secondary(name);
> --
> 2.51.0

Apart from one minor optimization suggestion above,

Acked-by: Marat Khalili <marat.khalili@huawei.com>

  reply	other threads:[~2026-01-07 10:58 UTC|newest]

Thread overview: 28+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-01-06 18:26 [PATCH 00/12] net/pcap: cleanups and test Stephen Hemminger
2026-01-06 18:26 ` [PATCH 01/12] net/pcap: avoid using rte_malloc and rte_memcpy Stephen Hemminger
2026-01-06 18:26 ` [PATCH 02/12] net/pcap: support MTU set Stephen Hemminger
2026-01-06 18:26 ` [PATCH 03/12] net/pcap: use bool for flags Stephen Hemminger
2026-01-07 10:28   ` Marat Khalili
2026-01-09  0:23     ` Stephen Hemminger
2026-01-06 18:26 ` [PATCH 04/12] net/pcap: support Tx offloads Stephen Hemminger
2026-01-06 18:26 ` [PATCH 05/12] net/pcap: support nanosecond timestamp precision Stephen Hemminger
2026-01-06 18:26 ` [PATCH 06/12] net/pcap: remove global variables Stephen Hemminger
2026-01-07  9:48   ` Marat Khalili
2026-01-06 18:26 ` [PATCH 07/12] net/pcap: avoid use of volatile Stephen Hemminger
2026-01-07 10:31   ` Marat Khalili
2026-01-06 18:26 ` [PATCH 08/12] net/pcap: optimize calculation of receive timestamp Stephen Hemminger
2026-01-07 10:58   ` Marat Khalili [this message]
2026-01-06 18:26 ` [PATCH 09/12] net/pcap: report receive clock Stephen Hemminger
2026-01-06 18:26 ` [PATCH 10/12] net/pcap: cleanup MAC address handling Stephen Hemminger
2026-01-06 18:26 ` [PATCH 11/12] net/pcap: support MAC address set Stephen Hemminger
2026-01-06 18:26 ` [PATCH 12/12] test: add test for pcap PMD Stephen Hemminger
2026-01-09  1:16 ` [PATCH v2 0/9] pcap: cleanup pcap PMD and add test Stephen Hemminger
2026-01-09  1:16   ` [PATCH v2 1/9] net/pcap: avoid using rte_malloc and rte_memcpy Stephen Hemminger
2026-01-09  1:16   ` [PATCH v2 2/9] net/pcap: support MTU set Stephen Hemminger
2026-01-09  1:16   ` [PATCH v2 3/9] net/pcap: use bool for flags Stephen Hemminger
2026-01-09  1:16   ` [PATCH v2 4/9] net/pcap: support Tx offloads Stephen Hemminger
2026-01-09  1:16   ` [PATCH v2 5/9] net/pcap: support nanosecond timestamp precision Stephen Hemminger
2026-01-09  1:16   ` [PATCH v2 6/9] net/pcap: remove global variables Stephen Hemminger
2026-01-09  1:16   ` [PATCH v2 7/9] net/pcap: avoid use of volatile Stephen Hemminger
2026-01-09  1:16   ` [PATCH v2 8/9] net/pcap: support MAC address set Stephen Hemminger
2026-01-09  1:16   ` [PATCH v2 9/9] test: add test for pcap PMD Stephen Hemminger

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=7a43c70576054749920c16e8c2ae6320@huawei.com \
    --to=marat.khalili@huawei.com \
    --cc=dev@dpdk.org \
    --cc=stephen@networkplumber.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).