From: "Ananyev, Konstantin" <konstantin.ananyev@intel.com>
To: "Iremonger, Bernard" <bernard.iremonger@intel.com>,
"dev@dpdk.org" <dev@dpdk.org>,
"akhil.goyal@nxp.com" <akhil.goyal@nxp.com>
Cc: "stable@dpdk.org" <stable@dpdk.org>
Subject: Re: [dpdk-dev] [PATCH] app/test/ipsec: fix logic around dequeue burst
Date: Fri, 12 Apr 2019 10:31:15 +0000 [thread overview]
Message-ID: <2601191342CEEE43887BDE71AB9772580148A969D8@irsmsx105.ger.corp.intel.com> (raw)
Message-ID: <20190412103115.VAKVtBF86x-4DEHPymHYGAaLq3ZvLqZCz41OPPlRvew@z> (raw)
In-Reply-To: <1554988097-18819-1-git-send-email-bernard.iremonger@intel.com>
Hi Bernard,
/ipsec: fix logic around dequeue burst
>
> Call rte_crypto_dequeue_burst() in a loop with
> a delay to ensure that all the packets are dequeued
> from the crtpto device.
>
> Fixes: 59d7353b0df0 ("test/ipsec: fix test suite setup")
> Cc: stable@dpdk.org
> Signed-off-by: Bernard Iremonger <bernard.iremonger@intel.com>
> ---
> app/test/test_ipsec.c | 42 +++++++++++++++++++++++++++++++-----------
> 1 file changed, 31 insertions(+), 11 deletions(-)
>
> diff --git a/app/test/test_ipsec.c b/app/test/test_ipsec.c
> index 80a2d25..25a7df4 100644
> --- a/app/test/test_ipsec.c
> +++ b/app/test/test_ipsec.c
> @@ -9,7 +9,7 @@
> #include <rte_mbuf.h>
> #include <rte_malloc.h>
> #include <rte_memcpy.h>
> -#include <rte_pause.h>
> +#include <rte_cycles.h>
> #include <rte_bus_vdev.h>
> #include <rte_ip.h>
>
> @@ -42,6 +42,7 @@
> #define OUTBOUND_SPI 17
> #define BURST_SIZE 32
> #define REORDER_PKTS 1
> +#define DEQUEUE_COUNT 1000
>
> struct user_params {
> enum rte_crypto_sym_xform_type auth;
> @@ -758,7 +759,9 @@ crypto_ipsec(uint16_t num_pkts)
> struct ipsec_testsuite_params *ts_params = &testsuite_params;
> struct ipsec_unitest_params *ut_params = &unittest_params;
> uint32_t k, ng;
> + uint32_t pkt_cnt;
> struct rte_ipsec_group grp[1];
> + int i;
>
> /* call crypto prepare */
> k = rte_ipsec_pkt_crypto_prepare(&ut_params->ss[0], ut_params->ibuf,
> @@ -774,9 +777,15 @@ crypto_ipsec(uint16_t num_pkts)
> return TEST_FAILED;
> }
>
> - k = rte_cryptodev_dequeue_burst(ts_params->valid_dev, 0,
> - ut_params->cop, num_pkts);
> - if (k != num_pkts) {
> + for (i = 0, pkt_cnt = 0;
> + i < DEQUEUE_COUNT && pkt_cnt != num_pkts; i++) {
> + k = rte_cryptodev_dequeue_burst(ts_params->valid_dev, 0,
> + &ut_params->cop[pkt_cnt], num_pkts - pkt_cnt);
> + pkt_cnt += k;
> + rte_delay_us(1);
> + }
Looks good to me.
My only suggestion would be to put that loop in a separate function
to avoid duplications.
Apart from that:
Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
> +
> + if (pkt_cnt != num_pkts) {
> RTE_LOG(ERR, USER1, "rte_cryptodev_dequeue_burst fail\n");
> return TEST_FAILED;
> }
> @@ -868,7 +877,7 @@ crypto_ipsec_2sa(void)
> struct ipsec_testsuite_params *ts_params = &testsuite_params;
> struct ipsec_unitest_params *ut_params = &unittest_params;
> struct rte_ipsec_group grp[BURST_SIZE];
> -
> + uint32_t pkt_cnt;
> uint32_t k, ng, i, r;
>
> for (i = 0; i < BURST_SIZE; i++) {
> @@ -890,9 +899,14 @@ crypto_ipsec_2sa(void)
> }
> }
>
> - k = rte_cryptodev_dequeue_burst(ts_params->valid_dev, 0,
> - ut_params->cop, BURST_SIZE);
> - if (k != BURST_SIZE) {
> + for (i = 0, k = 0, pkt_cnt = 0;
> + i < DEQUEUE_COUNT && pkt_cnt != BURST_SIZE; i++) {
> + k = rte_cryptodev_dequeue_burst(ts_params->valid_dev, 0,
> + &ut_params->cop[pkt_cnt], BURST_SIZE - pkt_cnt);
> + pkt_cnt += k;
> + rte_delay_us(1);
> + }
> + if (pkt_cnt != BURST_SIZE) {
> RTE_LOG(ERR, USER1, "rte_cryptodev_dequeue_burst fail\n");
> return TEST_FAILED;
> }
> @@ -1007,6 +1021,7 @@ crypto_ipsec_2sa_4grp(void)
> struct ipsec_unitest_params *ut_params = &unittest_params;
> struct rte_ipsec_group grp[BURST_SIZE];
> uint32_t k, ng, i, j;
> + uint32_t pkt_cnt;
> uint32_t rc = 0;
>
> for (i = 0; i < BURST_SIZE; i++) {
> @@ -1029,9 +1044,14 @@ crypto_ipsec_2sa_4grp(void)
> }
> }
>
> - k = rte_cryptodev_dequeue_burst(ts_params->valid_dev, 0,
> - ut_params->cop, BURST_SIZE);
> - if (k != BURST_SIZE) {
> + for (i = 0, k = 0, pkt_cnt = 0;
> + i < DEQUEUE_COUNT && pkt_cnt != BURST_SIZE; i++) {
> + k = rte_cryptodev_dequeue_burst(ts_params->valid_dev, 0,
> + &ut_params->cop[pkt_cnt], BURST_SIZE - pkt_cnt);
> + pkt_cnt += k;
> + rte_delay_us(1);
> + }
> + if (pkt_cnt != BURST_SIZE) {
> RTE_LOG(ERR, USER1, "rte_cryptodev_dequeue_burst fail\n");
> return TEST_FAILED;
> }
> --
> 2.7.4
next prev parent reply other threads:[~2019-04-12 10:31 UTC|newest]
Thread overview: 10+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-04-11 13:08 Bernard Iremonger
2019-04-11 13:08 ` Bernard Iremonger
2019-04-11 17:14 ` Trahe, Fiona
2019-04-11 17:14 ` Trahe, Fiona
2019-04-12 10:31 ` Ananyev, Konstantin [this message]
2019-04-12 10:31 ` Ananyev, Konstantin
2019-04-12 11:24 ` Iremonger, Bernard
2019-04-12 11:24 ` Iremonger, Bernard
2019-04-12 14:08 ` [dpdk-dev] [PATCH v2] " Bernard Iremonger
2019-04-12 14:08 ` Bernard Iremonger
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=2601191342CEEE43887BDE71AB9772580148A969D8@irsmsx105.ger.corp.intel.com \
--to=konstantin.ananyev@intel.com \
--cc=akhil.goyal@nxp.com \
--cc=bernard.iremonger@intel.com \
--cc=dev@dpdk.org \
--cc=stable@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).