From: Jun Yang <jun.yang@nxp.com>
To: "Dumitrescu, Cristian" <cristian.dumitrescu@intel.com>,
Hemant Agrawal <hemant.agrawal@nxp.com>,
"dev@dpdk.org" <dev@dpdk.org>
Cc: "Stable@dpdk.org" <Stable@dpdk.org>
Subject: Re: [dpdk-dev] [EXT] RE: [PATCH 2/2] examples/ip_pipeline: fix to use sock id instead of cpuid
Date: Fri, 14 Feb 2020 15:18:39 +0000 [thread overview]
Message-ID: <DB6PR04MB312830BB960A5EAE127BBABD9D150@DB6PR04MB3128.eurprd04.prod.outlook.com> (raw)
In-Reply-To: <SN6PR11MB27170D00627029F8B685753CEB150@SN6PR11MB2717.namprd11.prod.outlook.com>
Hi Cristian,
Per my debug log, the CPU ID is actually DPDK current core ID on NXP ARM64 platform whose socket ID is always 0.
-----Original Message-----
From: Dumitrescu, Cristian [mailto:cristian.dumitrescu@intel.com]
Sent: Friday, February 14, 2020 7:27 PM
To: Hemant Agrawal <hemant.agrawal@nxp.com>; dev@dpdk.org
Cc: Stable@dpdk.org; Jun Yang <jun.yang@nxp.com>
Subject: [EXT] RE: [dpdk-dev] [PATCH 2/2] examples/ip_pipeline: fix to use sock id instead of cpuid
Caution: EXT Email
> -----Original Message-----
> From: dev <dev-bounces@dpdk.org> On Behalf Of Hemant Agrawal
> Sent: Wednesday, December 11, 2019 5:26 AM
> To: dev@dpdk.org
> Cc: Dumitrescu, Cristian <cristian.dumitrescu@intel.com>;
> Stable@dpdk.org; Jun Yang <jun.yang@nxp.com>
> Subject: [dpdk-dev] [PATCH 2/2] examples/ip_pipeline: fix to use sock
> id instead of cpuid
>
> From: Jun Yang <jun.yang@nxp.com>
>
> rte_lcore_to_socket_id should be used to convert cpu ID to socket ID.
>
> Fixes: 6bfe74f8c93e ("examples/ip_pipeline: add mempool object")
> Fixes: d75c371e9b46 ("examples/ip_pipeline: add pipeline object")
> Cc: Stable@dpdk.org
>
> Signed-off-by: Jun Yang <jun.yang@nxp.com>
> ---
> examples/ip_pipeline/mempool.c | 2 +-
> examples/ip_pipeline/pipeline.c | 10 +++++-----
> 2 files changed, 6 insertions(+), 6 deletions(-)
>
> diff --git a/examples/ip_pipeline/mempool.c
> b/examples/ip_pipeline/mempool.c index f5d2a7d108..d7eea85f8f 100644
> --- a/examples/ip_pipeline/mempool.c
> +++ b/examples/ip_pipeline/mempool.c
> @@ -58,7 +58,7 @@ mempool_create(const char *name, struct
> mempool_params *params)
> params->cache_size,
> 0,
> params->buffer_size - sizeof(struct rte_mbuf),
> - params->cpu_id);
> + rte_lcore_to_socket_id(params->cpu_id));
>
> if (m == NULL)
> return NULL;
> diff --git a/examples/ip_pipeline/pipeline.c
> b/examples/ip_pipeline/pipeline.c index b627310a0c..d8df62d785 100644
> --- a/examples/ip_pipeline/pipeline.c
> +++ b/examples/ip_pipeline/pipeline.c
> @@ -94,7 +94,7 @@ pipeline_create(const char *name, struct
> pipeline_params *params)
>
> msgq_req = rte_ring_create(msgq_name,
> PIPELINE_MSGQ_SIZE,
> - params->cpu_id,
> + rte_lcore_to_socket_id(params->cpu_id),
> RING_F_SP_ENQ | RING_F_SC_DEQ);
> if (msgq_req == NULL)
> return NULL;
> @@ -103,7 +103,7 @@ pipeline_create(const char *name, struct
> pipeline_params *params)
>
> msgq_rsp = rte_ring_create(msgq_name,
> PIPELINE_MSGQ_SIZE,
> - params->cpu_id,
> + rte_lcore_to_socket_id(params->cpu_id),
> RING_F_SP_ENQ | RING_F_SC_DEQ);
> if (msgq_rsp == NULL) {
> rte_ring_free(msgq_req); @@ -111,7 +111,7 @@
> pipeline_create(const char *name, struct pipeline_params *params)
> }
>
> pp.name = name;
> - pp.socket_id = (int) params->cpu_id;
> + pp.socket_id = (int) rte_lcore_to_socket_id(params->cpu_id);
> pp.offset_port_id = params->offset_port_id;
>
> p = rte_pipeline_create(&pp);
> @@ -332,7 +332,7 @@ pipeline_port_in_create(const char *pipeline_name,
>
> if (ap) {
> action = rte_port_in_action_create(ap->ap,
> - pipeline->cpu_id);
> + rte_lcore_to_socket_id(pipeline->cpu_id));
> if (action == NULL)
> return -1;
>
> @@ -1002,7 +1002,7 @@ pipeline_table_create(const char *pipeline_name,
>
> if (ap) {
> action = rte_table_action_create(ap->ap,
> - pipeline->cpu_id);
> + rte_lcore_to_socket_id(pipeline->cpu_id));
> if (action == NULL)
> return -1;
>
> --
> 2.17.1
NACK.
This is incorrect, probably a misunderstanding: the params->cpu_id is already the CPU socket ID (passed as parameter), and not the core ID. For core ID, we typically use variable names as core_id or lcore_id, and never cpu_id.
next prev parent reply other threads:[~2020-02-14 18:14 UTC|newest]
Thread overview: 8+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-12-11 5:25 [dpdk-dev] [PATCH 1/2] examples/ip_pipeline: fix crypto queue config Hemant Agrawal
2019-12-11 5:25 ` [dpdk-dev] [PATCH 2/2] examples/ip_pipeline: fix to use sock id instead of cpuid Hemant Agrawal
2020-02-14 11:27 ` Dumitrescu, Cristian
2020-02-14 15:18 ` Jun Yang [this message]
2020-02-18 10:47 ` [dpdk-dev] [EXT] " Dumitrescu, Cristian
2020-02-19 14:17 ` Jun Yang
2020-02-14 10:52 ` [dpdk-dev] [PATCH 1/2] examples/ip_pipeline: fix crypto queue config Zhang, Roy Fan
2020-02-14 11:22 ` Dumitrescu, Cristian
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=DB6PR04MB312830BB960A5EAE127BBABD9D150@DB6PR04MB3128.eurprd04.prod.outlook.com \
--to=jun.yang@nxp.com \
--cc=Stable@dpdk.org \
--cc=cristian.dumitrescu@intel.com \
--cc=dev@dpdk.org \
--cc=hemant.agrawal@nxp.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 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).