All of lore.kernel.org
 help / color / mirror / Atom feed
From: David Marchand <david.marchand@redhat.com>
To: dev@dpdk.org
Cc: Ferruh Yigit <ferruh.yigit@amd.com>,
	Aman Singh <aman.deep.singh@intel.com>,
	Yuying Zhang <yuying.zhang@intel.com>,
	Robin Jarry <rjarry@redhat.com>,
	stable@dpdk.org, Tomasz Kulasek <tomaszx.kulasek@intel.com>,
	Konstantin Ananyev <konstantin.v.ananyev@yandex.ru>
Subject: [PATCH v2 1/9] app/testpmd: fix Tx preparation in checksum engine
Date: Mon, 20 Feb 2023 17:40:55 +0100	[thread overview]
Message-ID: <20230220164103.3041538-2-david.marchand@redhat.com> (raw)
In-Reply-To: <20230220164103.3041538-1-david.marchand@redhat.com>

"unprepared" packets could get to the wire in the retry loop.

Split packets freeing in two stages: one for preparation failure, and
one for transmission failure.
Adjust dropped counter update accordingly.

Fixes: 6b520d54ebfe ("app/testpmd: use Tx preparation in checksum engine")
Cc: stable@dpdk.org

Signed-off-by: David Marchand <david.marchand@redhat.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@amd.com>
---
 app/test-pmd/csumonly.c | 17 ++++++++++-------
 1 file changed, 10 insertions(+), 7 deletions(-)

diff --git a/app/test-pmd/csumonly.c b/app/test-pmd/csumonly.c
index 1c24598515..90a59e0aa5 100644
--- a/app/test-pmd/csumonly.c
+++ b/app/test-pmd/csumonly.c
@@ -1168,10 +1168,13 @@ pkt_burst_checksum_forward(struct fwd_stream *fs)
 
 	nb_prep = rte_eth_tx_prepare(fs->tx_port, fs->tx_queue,
 			tx_pkts_burst, nb_rx);
-	if (nb_prep != nb_rx)
+	if (nb_prep != nb_rx) {
 		fprintf(stderr,
 			"Preparing packet burst to transmit failed: %s\n",
 			rte_strerror(rte_errno));
+		fs->fwd_dropped += (nb_rx - nb_prep);
+		rte_pktmbuf_free_bulk(&tx_pkts_burst[nb_prep], nb_rx - nb_prep);
+	}
 
 	nb_tx = rte_eth_tx_burst(fs->tx_port, fs->tx_queue, tx_pkts_burst,
 			nb_prep);
@@ -1179,12 +1182,12 @@ pkt_burst_checksum_forward(struct fwd_stream *fs)
 	/*
 	 * Retry if necessary
 	 */
-	if (unlikely(nb_tx < nb_rx) && fs->retry_enabled) {
+	if (unlikely(nb_tx < nb_prep) && fs->retry_enabled) {
 		retry = 0;
-		while (nb_tx < nb_rx && retry++ < burst_tx_retry_num) {
+		while (nb_tx < nb_prep && retry++ < burst_tx_retry_num) {
 			rte_delay_us(burst_tx_delay_time);
 			nb_tx += rte_eth_tx_burst(fs->tx_port, fs->tx_queue,
-					&tx_pkts_burst[nb_tx], nb_rx - nb_tx);
+					&tx_pkts_burst[nb_tx], nb_prep - nb_tx);
 		}
 	}
 	fs->tx_packets += nb_tx;
@@ -1194,11 +1197,11 @@ pkt_burst_checksum_forward(struct fwd_stream *fs)
 	fs->rx_bad_outer_ip_csum += rx_bad_outer_ip_csum;
 
 	inc_tx_burst_stats(fs, nb_tx);
-	if (unlikely(nb_tx < nb_rx)) {
-		fs->fwd_dropped += (nb_rx - nb_tx);
+	if (unlikely(nb_tx < nb_prep)) {
+		fs->fwd_dropped += (nb_prep - nb_tx);
 		do {
 			rte_pktmbuf_free(tx_pkts_burst[nb_tx]);
-		} while (++nb_tx < nb_rx);
+		} while (++nb_tx < nb_prep);
 	}
 
 	get_end_cycles(fs, start_tsc);
-- 
2.39.2


  reply	other threads:[~2023-02-20 16:41 UTC|newest]

Thread overview: 48+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-01-24 10:47 [PATCH 0/6] Testpmd code cleanup David Marchand
2023-01-24 10:47 ` [PATCH 1/6] app/testpmd: factorize core cycles record David Marchand
2023-02-14 18:14   ` Ferruh Yigit
2023-01-24 10:47 ` [PATCH 2/6] app/testpmd: don't send unprepared packets David Marchand
2023-02-14 18:14   ` Ferruh Yigit
2023-01-24 10:47 ` [PATCH 3/6] app/testpmd: bulk free mbufs David Marchand
2023-02-14 18:14   ` Ferruh Yigit
2023-01-24 10:47 ` [PATCH 4/6] app/testpmd: factorize fwd engine init David Marchand
2023-02-14 18:14   ` Ferruh Yigit
2023-01-24 10:47 ` [PATCH 5/6] app/testpmd: factorize fwd engine Rx David Marchand
2023-02-14 18:15   ` Ferruh Yigit
2023-01-24 10:47 ` [PATCH 6/6] app/testpmd: factorize fwd engine Tx David Marchand
2023-02-14 11:03   ` Singh, Aman Deep
2023-02-14 18:17     ` Ferruh Yigit
2023-02-16  8:01       ` Singh, Aman Deep
2023-02-16 10:07         ` Ferruh Yigit
2023-02-14 18:16   ` Ferruh Yigit
2023-02-20 16:33     ` David Marchand
2023-01-25 13:50 ` [PATCH 0/6] Testpmd code cleanup Robin Jarry
2023-02-14 18:22   ` Ferruh Yigit
2023-02-20 15:02     ` David Marchand
2023-02-20 16:40 ` [PATCH v2 0/9] " David Marchand
2023-02-20 16:40   ` David Marchand [this message]
2023-02-20 16:40   ` [PATCH v2 2/9] app/testpmd: fix packet count in ieee15888 engine David Marchand
2023-02-20 16:40   ` [PATCH v2 3/9] app/testpmd: rework ieee1588 engine fwd configuration David Marchand
2023-02-24  9:11     ` Singh, Aman Deep
2023-02-20 16:40   ` [PATCH v2 4/9] app/testpmd: fix packet transmission in noisy VNF engine David Marchand
2023-02-20 16:40   ` [PATCH v2 5/9] app/testpmd: bulk free mbufs David Marchand
2023-02-20 16:41   ` [PATCH v2 6/9] app/testpmd: factorize core cycles record David Marchand
2023-02-20 16:41   ` [PATCH v2 7/9] app/testpmd: factorize fwd engine init David Marchand
2023-02-20 16:41   ` [PATCH v2 8/9] app/testpmd: factorize fwd engine Rx David Marchand
2023-02-20 16:41   ` [PATCH v2 9/9] app/testpmd: factorize fwd engine Tx David Marchand
2023-02-20 18:34 ` [PATCH v3 0/9] Testpmd code cleanup David Marchand
2023-02-20 18:34   ` [PATCH v3 1/9] app/testpmd: fix Tx preparation in checksum engine David Marchand
2023-02-20 18:34   ` [PATCH v3 2/9] app/testpmd: fix packet count in ieee15888 engine David Marchand
2023-02-24  8:24     ` Singh, Aman Deep
2023-02-20 18:34   ` [PATCH v3 3/9] app/testpmd: rework ieee1588 engine fwd configuration David Marchand
2023-02-28 18:50     ` Ferruh Yigit
2023-02-20 18:34   ` [PATCH v3 4/9] app/testpmd: fix packet transmission in noisy VNF engine David Marchand
2023-02-28 18:51     ` Ferruh Yigit
2023-02-20 18:34   ` [PATCH v3 5/9] app/testpmd: bulk free mbufs David Marchand
2023-02-20 18:45     ` Stephen Hemminger
2023-02-20 18:34   ` [PATCH v3 6/9] app/testpmd: factorize core cycles record David Marchand
2023-02-20 18:35   ` [PATCH v3 7/9] app/testpmd: factorize fwd engine init David Marchand
2023-02-20 18:35   ` [PATCH v3 8/9] app/testpmd: factorize fwd engine Rx David Marchand
2023-02-20 18:35   ` [PATCH v3 9/9] app/testpmd: factorize fwd engine Tx David Marchand
2023-02-28 18:35     ` Ferruh Yigit
2023-02-28 18:54   ` [PATCH v3 0/9] Testpmd code cleanup 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=20230220164103.3041538-2-david.marchand@redhat.com \
    --to=david.marchand@redhat.com \
    --cc=aman.deep.singh@intel.com \
    --cc=dev@dpdk.org \
    --cc=ferruh.yigit@amd.com \
    --cc=konstantin.v.ananyev@yandex.ru \
    --cc=rjarry@redhat.com \
    --cc=stable@dpdk.org \
    --cc=tomaszx.kulasek@intel.com \
    --cc=yuying.zhang@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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.