From: "Steve Wise" <swise@opengridcomputing.com>
To: 'Leon Romanovsky' <leon@kernel.org>, 'David Ahern' <dsahern@gmail.com>
Cc: 'Leon Romanovsky' <leonro@mellanox.com>,
'netdev' <netdev@vger.kernel.org>,
'RDMA mailing list' <linux-rdma@vger.kernel.org>,
'Stephen Hemminger' <stephen@networkplumber.org>
Subject: RE: [PATCH iproute2-next v1 04/19] rdma: Provide parent context index for all objects except CM_ID
Date: Thu, 21 Feb 2019 14:12:19 -0600 [thread overview]
Message-ID: <01d401d4ca21$bfad6390$3f082ab0$@opengridcomputing.com> (raw)
In-Reply-To: <20190220072136.3840-5-leon@kernel.org>
> -----Original Message-----
> From: linux-rdma-owner@vger.kernel.org <linux-rdma-
> owner@vger.kernel.org> On Behalf Of Leon Romanovsky
> Sent: Wednesday, February 20, 2019 1:21 AM
> To: David Ahern <dsahern@gmail.com>
> Cc: Leon Romanovsky <leonro@mellanox.com>; netdev
> <netdev@vger.kernel.org>; RDMA mailing list <linux-
> rdma@vger.kernel.org>; Stephen Hemminger
> <stephen@networkplumber.org>
> Subject: [PATCH iproute2-next v1 04/19] rdma: Provide parent context index
> for all objects except CM_ID
>
> From: Leon Romanovsky <leonro@mellanox.com>
>
> Allow users to correlate allocated object with relevant parent
>
> [leonro@server ~]$ rdma res show pd
> dev mlx5_0 users 5 pid 0 comm [ib_core] pdn 1
> dev mlx5_0 users 7 pid 0 comm [ib_ipoib] pdn 2
> dev mlx5_0 users 0 pid 0 comm [mlx5_ib] pdn 3
> dev mlx5_0 users 2 pid 548 comm ibv_rc_pingpong ctxn 0 pdn 4
>
> [leonro@server ~]$ rdma res show cq cqn 0-100
> dev mlx5_0 cqe 2047 users 6 poll-ctx UNBOUND_WORKQUEUE pid 0 comm
> [ib_core] cqn 2
> dev mlx5_0 cqe 255 users 2 poll-ctx SOFTIRQ pid 0 comm [mlx5_ib] cqn 3
> dev mlx5_0 cqe 511 users 1 poll-ctx DIRECT pid 0 comm [ib_ipoib] cqn 4
> dev mlx5_0 cqe 255 users 1 poll-ctx DIRECT pid 0 comm [ib_ipoib] cqn 5
> dev mlx5_0 cqe 255 users 0 poll-ctx SOFTIRQ pid 0 comm [mlx5_ib] cqn 6
> dev mlx5_0 cqe 511 users 2 pid 548 comm ibv_rc_pingpong cqn 7 ctxn 0
>
> [leonro@server ~]$ rdma res show mr
> dev mlx5_0 mrlen 4096 pid 548 comm ibv_rc_pingpong mrn 4 pdn 0
>
> [leonro@nps-server-14-015 ~]$ /images/leonro/src/iproute2/rdma/rdma res
> show qp
> link mlx5_0/1 lqpn 0 type SMI state RTS sq-psn 0 pid 0 comm [ib_core]
> link mlx5_0/1 lqpn 1 type GSI state RTS sq-psn 0 pid 0 comm [ib_core]
> link mlx5_0/1 lqpn 7 type UD state RTS sq-psn 0 pid 0 comm [ib_core]
> link mlx5_0/1 lqpn 8 type UD state RTS sq-psn 0 pid 0 comm [ib_ipoib]
> link mlx5_0/1 lqpn 9 pdn 4 rqpn 0 type RC state INIT rq-psn 0 sq-psn 0
path-
> mig-state MIGRATED pid 548 comm ibv_rc_pingpong
>
> Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
> ---
> rdma/res.c | 110 +++++++++++++++++++++++++++++++++++----------------
> --
> 1 file changed, 72 insertions(+), 38 deletions(-)
>
> diff --git a/rdma/res.c b/rdma/res.c
> index f941de5c..308f66c0 100644
> --- a/rdma/res.c
> +++ b/rdma/res.c
> @@ -320,6 +320,22 @@ static char *get_task_name(uint32_t pid)
> return comm;
> }
>
> +static void print_key(struct rd *rd, const char *name, uint64_t val)
> +{
> + if (rd->json_output)
> + jsonw_xint_field(rd->jw, name, val);
> + else
> + pr_out("%s 0x%" PRIx64 " ", name, val);
> +}
> +
> +static void res_print_uint(struct rd *rd, const char *name, uint64_t val)
> +{
> + if (rd->json_output)
> + jsonw_uint_field(rd->jw, name, val);
> + else
> + pr_out("%s %" PRIu64 " ", name, val);
> +}
> +
> static int res_qp_parse_cb(const struct nlmsghdr *nlh, void *data)
> {
> struct nlattr *tb[RDMA_NLDEV_ATTR_MAX] = {};
> @@ -343,6 +359,7 @@ static int res_qp_parse_cb(const struct nlmsghdr
> *nlh, void *data)
> uint32_t lqpn, rqpn = 0, rq_psn = 0, sq_psn;
> uint8_t type, state, path_mig_state = 0;
> uint32_t port = 0, pid = 0;
> + uint32_t pdn = 0;
> char *comm = NULL;
> int err;
>
> @@ -369,6 +386,11 @@ static int res_qp_parse_cb(const struct nlmsghdr
> *nlh, void *data)
> if (rd_check_is_filtered(rd, "lqpn", lqpn))
> continue;
>
> + if (nla_line[RDMA_NLDEV_ATTR_RES_PDN])
> + pdn =
> mnl_attr_get_u32(nla_line[RDMA_NLDEV_ATTR_RES_PDN]);
> + if (rd_check_is_filtered(rd, "pdn", pdn))
> + continue;
> +
> if (nla_line[RDMA_NLDEV_ATTR_RES_RQPN]) {
> rqpn =
> mnl_attr_get_u32(nla_line[RDMA_NLDEV_ATTR_RES_RQPN]);
> if (rd_check_is_filtered(rd, "rqpn", rqpn))
> @@ -428,6 +450,8 @@ static int res_qp_parse_cb(const struct nlmsghdr
> *nlh, void *data)
> print_link(rd, idx, name, port, nla_line);
>
> print_lqpn(rd, lqpn);
> + if (nla_line[RDMA_NLDEV_ATTR_RES_PDN])
> + res_print_uint(rd, "pdn", pdn);
> print_rqpn(rd, rqpn, nla_line);
>
> print_type(rd, type);
> @@ -549,22 +573,6 @@ static int ss_ntop(struct nlattr *nla_line, char
> *addr_str, uint16_t *port)
> return 0;
> }
>
> -static void print_key(struct rd *rd, const char *name, uint64_t val)
> -{
> - if (rd->json_output)
> - jsonw_xint_field(rd->jw, name, val);
> - else
> - pr_out("%s 0x%" PRIx64 " ", name, val);
> -}
> -
> -static void res_print_uint(struct rd *rd, const char *name, uint64_t val)
> -{
> - if (rd->json_output)
> - jsonw_uint_field(rd->jw, name, val);
> - else
> - pr_out("%s %" PRIu64 " ", name, val);
> -}
> -
> static int res_cm_id_parse_cb(const struct nlmsghdr *nlh, void *data)
> {
> struct nlattr *tb[RDMA_NLDEV_ATTR_MAX] = {};
> @@ -768,6 +776,7 @@ static int res_cq_parse_cb(const struct nlmsghdr *nlh,
> void *data)
> char *comm = NULL;
> uint32_t pid = 0;
> uint8_t poll_ctx = 0;
> + uint32_t ctxn = 0;
> uint32_t cqn = 0;
> uint64_t users;
> uint32_t cqe;
> @@ -815,6 +824,12 @@ static int res_cq_parse_cb(const struct nlmsghdr
> *nlh, void *data)
> if (rd_check_is_filtered(rd, "cqn", cqn))
> continue;
>
> + if (nla_line[RDMA_NLDEV_ATTR_RES_CTXN])
> + ctxn = mnl_attr_get_u32(
> + nla_line[RDMA_NLDEV_ATTR_RES_CTXN]);
> + if (rd_check_is_filtered(rd, "ctxn", ctxn))
> + continue;
> +
> if (nla_line[RDMA_NLDEV_ATTR_RES_KERN_NAME])
> /* discard const from mnl_attr_get_str */
> comm = (char *)mnl_attr_get_str(
> @@ -833,6 +848,8 @@ static int res_cq_parse_cb(const struct nlmsghdr *nlh,
> void *data)
>
> if (nla_line[RDMA_NLDEV_ATTR_RES_CQN])
> res_print_uint(rd, "cqn", cqn);
> + if (nla_line[RDMA_NLDEV_ATTR_RES_CTXN])
> + res_print_uint(rd, "ctxn", ctxn);
>
> if (nla_line[RDMA_NLDEV_ATTR_RES_PID])
> free(comm);
> @@ -866,6 +883,7 @@ static int res_mr_parse_cb(const struct nlmsghdr
> *nlh, void *data)
> uint32_t rkey = 0, lkey = 0;
> uint64_t iova = 0, mrlen;
> char *comm = NULL;
> + uint32_t pdn = 0;
> uint32_t mrn = 0;
> uint32_t pid = 0;
> int err;
> @@ -911,6 +929,12 @@ static int res_mr_parse_cb(const struct nlmsghdr
> *nlh, void *data)
> if (rd_check_is_filtered(rd, "mrn", mrn))
> continue;
>
> + if (nla_line[RDMA_NLDEV_ATTR_RES_PDN])
> + pdn = mnl_attr_get_u32(
> + nla_line[RDMA_NLDEV_ATTR_RES_PDN]);
> + if (rd_check_is_filtered(rd, "pdn", pdn))
> + continue;
> +
> if (nla_line[RDMA_NLDEV_ATTR_RES_KERN_NAME])
> /* discard const from mnl_attr_get_str */
> comm = (char *)mnl_attr_get_str(
> @@ -933,6 +957,9 @@ static int res_mr_parse_cb(const struct nlmsghdr
> *nlh, void *data)
> if (nla_line[RDMA_NLDEV_ATTR_RES_MRN])
> res_print_uint(rd, "mrn", mrn);
>
> + if (nla_line[RDMA_NLDEV_ATTR_RES_PDN])
> + res_print_uint(rd, "pdn", pdn);
> +
> if (nla_line[RDMA_NLDEV_ATTR_RES_PID])
> free(comm);
>
> @@ -964,6 +991,7 @@ static int res_pd_parse_cb(const struct nlmsghdr
> *nlh, void *data)
> uint32_t local_dma_lkey = 0, unsafe_global_rkey = 0;
> struct nlattr *nla_line[RDMA_NLDEV_ATTR_MAX] = {};
> char *comm = NULL;
> + uint32_t ctxn = 0;
> uint32_t pid = 0;
> uint32_t pdn = 0;
> uint64_t users;
> @@ -997,7 +1025,13 @@ static int res_pd_parse_cb(const struct nlmsghdr
> *nlh, void *data)
> comm = get_task_name(pid);
> }
>
> - if (rd_check_is_filtered(rd, "pid", pid))
> + if (rd_check_is_filtered(rd, "pid", pid))
Did this add some whitespace problem?
Reviewed-by: Steve Wise <swise@opengridcomputing.com>
next prev parent reply other threads:[~2019-02-21 20:12 UTC|newest]
Thread overview: 46+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-02-20 7:21 [PATCH iproute2-next v1 00/19] Export object IDs to users Leon Romanovsky
2019-02-20 7:21 ` Leon Romanovsky
2019-02-20 7:21 ` [PATCH iproute2-next v1 01/19] rdma: update uapi headers Leon Romanovsky
2019-02-21 20:02 ` Steve Wise
2019-02-21 22:11 ` Leon Romanovsky
2019-02-20 7:21 ` [PATCH iproute2-next v1 02/19] rdma: Remove duplicated print code Leon Romanovsky
2019-02-21 15:23 ` David Ahern
2019-02-21 15:39 ` Leon Romanovsky
2019-02-21 20:04 ` Steve Wise
2019-02-20 7:21 ` [PATCH iproute2-next v1 03/19] rdma: Provide unique indexes for all visible objects Leon Romanovsky
2019-02-21 20:08 ` Steve Wise
2019-02-20 7:21 ` [PATCH iproute2-next v1 04/19] rdma: Provide parent context index for all objects except CM_ID Leon Romanovsky
2019-02-21 20:12 ` Steve Wise [this message]
2019-02-23 9:03 ` Leon Romanovsky
2019-02-20 7:21 ` [PATCH iproute2-next v1 05/19] rdma: Move resource PD logic to separate file Leon Romanovsky
2019-02-21 20:15 ` Steve Wise
2019-02-23 9:05 ` Leon Romanovsky
2019-02-20 7:21 ` [PATCH iproute2-next v1 06/19] rdma: Refactor out resource MR " Leon Romanovsky
2019-02-21 20:15 ` Steve Wise
2019-02-20 7:21 ` [PATCH iproute2-next v1 07/19] rdma: Move out resource CQ " Leon Romanovsky
2019-02-21 20:16 ` Steve Wise
2019-02-20 7:21 ` [PATCH iproute2-next v1 08/19] rdma: Move out resource CM-ID " Leon Romanovsky
2019-02-21 20:17 ` Steve Wise
2019-02-20 7:21 ` [PATCH iproute2-next v1 09/19] rdma: Move resource PD " Leon Romanovsky
2019-02-21 20:21 ` Steve Wise
2019-02-20 7:21 ` [PATCH iproute2-next v1 10/19] rdma: Properly mark RDMAtool license Leon Romanovsky
2019-02-21 20:22 ` Steve Wise
2019-02-20 7:21 ` [PATCH iproute2-next v1 11/19] rdma: Simplify code to reuse existing functions Leon Romanovsky
2019-02-21 20:23 ` Steve Wise
2019-02-20 7:21 ` [PATCH iproute2-next v1 12/19] rdma: Simplify CM_ID print code Leon Romanovsky
2019-02-21 20:26 ` Steve Wise
2019-02-20 7:21 ` [PATCH iproute2-next v1 13/19] rdma: Refactor CQ prints Leon Romanovsky
2019-02-21 20:27 ` Steve Wise
2019-02-20 7:21 ` [PATCH iproute2-next v1 14/19] rdma: Separate MR code Leon Romanovsky
2019-02-21 20:28 ` Steve Wise
2019-02-20 7:21 ` [PATCH iproute2-next v1 15/19] rdma: Separate PD code Leon Romanovsky
2019-02-21 20:29 ` Steve Wise
2019-02-20 7:21 ` [PATCH iproute2-next v1 16/19] rdma: Move QP code to separate function Leon Romanovsky
2019-02-21 20:29 ` Steve Wise
2019-02-20 7:21 ` [PATCH iproute2-next v1 17/19] rdma: Unify netlink attribute checks prior to prints Leon Romanovsky
2019-02-21 20:31 ` Steve Wise
2019-02-20 7:21 ` [PATCH iproute2-next v1 18/19] rdma: Perform single .doit call to query specific objects Leon Romanovsky
2019-02-21 20:35 ` Steve Wise
2019-02-20 7:21 ` [PATCH iproute2-next v1 19/19] rdma: Provide and reuse filter functions Leon Romanovsky
2019-02-21 20:39 ` Steve Wise
2019-02-21 15:26 ` [PATCH iproute2-next v1 00/19] Export object IDs to users David Ahern
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='01d401d4ca21$bfad6390$3f082ab0$@opengridcomputing.com' \
--to=swise@opengridcomputing.com \
--cc=dsahern@gmail.com \
--cc=leon@kernel.org \
--cc=leonro@mellanox.com \
--cc=linux-rdma@vger.kernel.org \
--cc=netdev@vger.kernel.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 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.