The dump isn't supported for the statistics bind/unbind commands
because they operate on specific QP counters. This is different
from query commands that can operate on many objects at the same
time.
Let's check the user input and ensure that arguments are valid.
Fixes: a6d0773ebe ("rdma: Add stat manual mode support")
Signed-off-by: Ido Kalir <idok@nvidia.com>
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
The value of a number of booleans is shown as "on" and "off" in the plain
output, and as an actual boolean in JSON mode. Add a function that does
that.
RDMA tool already uses a function named print_on_off(). This function
always shows "on" and "off", even in JSON mode. Since there are probably
very few if any consumers of this interface at this point, migrate it to
the new central print_on_off() as well.
Signed-off-by: Petr Machata <me@pmachata.org>
Reviewed-by: Leon Romanovsky <leonro@nvidia.com>
Signed-off-by: David Ahern <dsahern@gmail.com>
Add 'raw' argument to get the resource in raw format.
When RDMA_NLDEV_ATTR_RES_RAW is set in the netlink message,
then the resource fields are in raw format, print it as byte array.
Example:
$rdma res show qp link rocep0s12f0/1 lqpn 1137 -j -r
[{"ifindex":7,"ifname":"mlx5_1","port":1,
"data":[0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,...]}]
Signed-off-by: Maor Gottlieb <maorg@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: David Ahern <dsahern@kernel.org>
Instead of doing open-coded solution to generate JSON and prints, let's
reuse existing infrastructure and APIs to do the same as ip/*.
Before this change:
if (rd->json_output)
jsonw_uint_field(rd->jw, "sm_lid", sm_lid);
else
pr_out("sm_lid %u ", sm_lid);
After this change:
print_uint(PRINT_ANY, "sm_lid", "sm_lid %u ", sm_lid);
All the print functions are converted to support color but for now the
type of color is COLOR_NONE. This is done as a preparation to addition
of color enable option. Such change will require rewrite of command line
arguments parser which is out-of-scope for this patch.
Signed-off-by: Ido Kalir <idok@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: David Ahern <dsahern@gmail.com>
In order to set adaptive-moderation for an ib device the command is:
rdma dev set [DEV] adaptive-moderation [on|off]
rdma dev show -d
0: mlx5_0: node_type ca fw 16.25.0319 node_guid 248a:0703:00a5:29d0
sys_image_guid 248a:0703:00a5:29d0 adaptive-moderation on
caps: <BAD_PKEY_CNTR, BAD_QKEY_CNTR, AUTO_PATH_MIG, CHANGE_PHY_PORT,
PORT_ACTIVE_EVENT, SYS_IMAGE_GUID, RC_RNR_NAK_GEN, MEM_WINDOW, XRC,
MEM_MGT_EXTENSIONS, BLOCK_MULTICAST_LOOPBACK, MEM_WINDOW_TYPE_2B,
RAW_IP_CSUM, CROSS_CHANNEL, MANAGED_FLOW_STEERING, SIGNATURE_HANDOVER,
ON_DEMAND_PAGING, SG_GAPS_REG, RAW_SCATTER_FCS, PCI_WRITE_END_PADDING>
rdma resource show cq
dev mlx5_0 cqn 0 cqe 1023 users 4 poll-ctx UNBOUND_WORKQUEUE
adaptive-moderation off comm [ib_core]
Signed-off-by: Yamin Friedman <yaminf@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
When strict_port is set, make get_port_from_argv() returns failure if
no valid port is specified.
Signed-off-by: Mark Zhang <markz@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
With per-QP statistic counter support, a user is allowed to monitor
specific QPs categories, which are bound to/unbound from counters
dynamically allocated/deallocated.
In per-port "auto" mode, QPs are bound to counters automatically
according to common criteria. For example a per "type"(qp type)
scheme, where in each process all QPs have same qp type are bind
automatically to a single counter.
Currently only "type" (qp type) is supported. Examples:
$ rdma statistic qp set link mlx5_2/1 auto type on
$ rdma statistic qp set link mlx5_2/1 auto off
Signed-off-by: Mark Zhang <markz@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Add an interface to show which mode is active. Two modes are supported:
- "auto": In this mode all QPs belong to one category are bind automatically
to a single counter set. Currently only "qp type" is supported;
- "manual": In this mode QPs are bound to a counter manually.
Examples:
$ rdma statistic qp mode
0/1: mlx5_0/1: qp auto off
1/1: mlx5_1/1: qp auto off
2/1: mlx5_2/1: qp auto type on
3/1: mlx5_3/1: qp auto off
$ rdma statistic qp mode link mlx5_0
0/1: mlx5_0/1: qp auto off
Signed-off-by: Mark Zhang <markz@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Enrich rdmatool with an option to query rdma subsystem parameter
whether rdma devices are shared among multiple network namespaces
or exclusive to single network namespace.
rdma tool command examples and output.
$ rdma system show
netns shared
$ rdma system set netns exclusive
$ rdma system show
netns exclusive
Reviewed-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Parav Pandit <parav@mellanox.com>
Signed-off-by: David Ahern <dsahern@gmail.com>
This function sends the constructed netlink message and then
receives the response.
Change rd_recv_msg() to display any error messages.
Change 'rdma dev set' to use rd_sendrecv_msg().
Signed-off-by: Steve Wise <larrystevenwise@gmail.com>
Reviewed-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: David Ahern <dsahern@gmail.com>
There is a need to distinguish between driver vs. general exposed
attributes. The most common use case is to expose some internal
garbage under extremely common and sexy name, e.g. pi, ci e.t.c
In order to achieve that, we will add "drv_" prefix to all strings
which were received through RDMA_NLDEV_ATTR_DRIVER_* attributes.
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>a
Tested-by: Lijun Ou <oulijun@huawei.com>
Signed-off-by: David Ahern <dsahern@gmail.com>
Globally replace all filter function in safer variants of those
is_filtered functions, which take into account the availability/lack
of netlink attributes.
Such conversion allowed to fix a number of places in the code, where
the previous implementation didn't honor filter requests if netlink
attribute wasn't present.
Reviewed-by: Steve Wise <swise@opengridcomputing.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: David Ahern <dsahern@gmail.com>
If user provides specific index, we can speedup query
by using .doit callback and save full dump and filtering
after that.
Reviewed-by: Steve Wise <swise@opengridcomputing.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: David Ahern <dsahern@gmail.com>
RDMA subsystem is dual-licensed with "GPL-2.0 OR Linux-OpenIB" proper
license and Mellanox submission are supposed to have this type of license.
Reviewed-by: Steve Wise <swise@opengridcomputing.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: David Ahern <dsahern@gmail.com>
mnl_attr_type_valid() receives maximum attribute type, which means that
we were supposed to supply the latest valid netlink attribute and not
the number of attributes. Such coding mistake caused to failures while
NLA attributes were extended.
Fixes: 74bd75c2b6 ("rdma: Add basic infrastructure for RDMA tool")
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
This reverts commit e99c4443ae.
Patch added to iproute2-master breaks builds of -next because of a
more recent patch in -next that relies on the exports. Revert the
offending patch. Unfortunately this leaves a window where builds
break.
Signed-off-by: David Ahern <dsahern@gmail.com>
In contradiction to various show commands, the set command explicitly
requires to use device name as an argument. Provide new command
execution helper which enforces it.
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Reviewed-by: Steve Wise <swise@opengridcomputing.com>
Signed-off-by: David Ahern <dsahern@gmail.com>
This enhancement allows printing rdma device-specific state, if provided
by the kernel. This is done in a generic manner, so rdma tool doesn't
need to know about the details of every type of rdma device.
Driver attributes for a rdma resource are in the form of <key,
[print_type], value> tuples, where the key is a string and the value can
be any supported driver attribute. The print_type attribute, if present,
provides a print format to use vs the standard print format for the type.
For example, the default print type for a PROVIDER_S32 value is "%d ",
but "0x%x " if the print_type of PRINT_TYPE_HEX is included inthe tuple.
Driver resources are only printed when the -dd flag is present.
If -p is present, then the output is formatted to not exceed 80 columns,
otherwise it is printed as a single row to be grep/awk friendly.
Example output:
# rdma resource show qp lqpn 1028 -dd -p
link cxgb4_0/- lqpn 1028 rqpn 0 type RC state RTS rq-psn 0 sq-psn 0 path-mig-state MIGRATED pid 0 comm [nvme_rdma]
sqid 1028 flushed 0 memsize 123968 cidx 85 pidx 85 wq_pidx 106 flush_cidx 85 in_use 0
size 386 flags 0x0 rqid 1029 memsize 16768 cidx 43 pidx 41 wq_pidx 171 msn 44 rqt_hwaddr 0x2a8a5d00
rqt_size 256 in_use 128 size 130
Signed-off-by: Steve Wise <swise@opengridcomputing.com>
Signed-off-by: David Ahern <dsahern@gmail.com>
The RDMA devices are operated in RoCE and iWARP modes have net device
underneath. Present their names in regular output and their net index
in detailed mode.
[root@nps ~]# rdma link show mlx5_3/1
4/1: mlx5_3/1: state ACTIVE physical_state LINK_UP netdev ens7
[root@nps ~]# rdma link show mlx5_3/1 -d
4/1: mlx5_3/1: state ACTIVE physical_state LINK_UP netdev ens7 netdev_index 7
caps: <CM, IP_BASED_GIDS>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Reviewed-by: Steve Wise <swise@opengridcomputing.com>
Signed-off-by: David Ahern <dsahern@gmail.com>
The check if netlink attributes supplied more than maximum supported
is to strict and may lead to backward compatibility issues with old
application with a newer kernel that supports new attribute.
CC: Steve Wise <swise@opengridcomputing.com>
Fixes: 74bd75c2b6 ("rdma: Add basic infrastructure for RDMA tool")
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Reviewed-by: Steve Wise <swise@opengridcomputing.com>
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
This patch adds ss-similar interface to view various resource
tracked objects. At this stage, only QP is presented.
1. Get all QPs for the specific device:
$ rdma res show qp link mlx5_4
link mlx5_4/- lqpn 8 type UD state RESET sq-psn 0 pid 0 comm [ib_ipoib]
link mlx5_4/1 lqpn 7 type UD state RTS sq-psn 0 pid 0 comm [ib_core]
link mlx5_4/1 lqpn 1 type GSI state RTS sq-psn 0 pid 0 comm [ib_core]
link mlx5_4/1 lqpn 0 type SMI state RTS sq-psn 0 pid 0 comm [ib_core]
$ rdma res show qp link mlx5_4/
link mlx5_4/- lqpn 8 type UD state RESET sq-psn 0 pid 0 comm [ib_ipoib]
link mlx5_4/1 lqpn 7 type UD state RTS sq-psn 0 pid 0 comm [ib_core]
link mlx5_4/1 lqpn 1 type GSI state RTS sq-psn 0 pid 0 comm [ib_core]
link mlx5_4/1 lqpn 0 type SMI state RTS sq-psn 0 pid 0 comm [ib_core]
2. Provide illegal port number (0 is illegal):
$ rdma res show qp link mlx5_4/0
Wrong device name
3. Get QPs of specific port:
$ rdma res show qp link mlx5_4/1
link mlx5_4/1 lqpn 7 type UD state RTS sq-psn 0 pid 0 comm [ib_core]
link mlx5_4/1 lqpn 1 type GSI state RTS sq-psn 0 pid 0 comm [ib_core]
link mlx5_4/1 lqpn 0 type SMI state RTS sq-psn 0 pid 0 comm [ib_core]
4. Get QPs which have not assigned port yet:
link mlx5_4/- lqpn 8 type UD state RESET sq-psn 0 pid 0 comm [ib_ipoib]
5. Limit to specific Local QPNs:
$ rdma res show qp link mlx5_4/1 lqpn 1-3,7
link mlx5_4/1 lqpn 7 type UD state RTS sq-psn 0 pid 0 comm [ib_core]
link mlx5_4/1 lqpn 1 type GSI state RTS sq-psn 0 pid 0 comm [ib_core]
link mlx5_4/1 lqpn 0 type SMI state RTS sq-psn 0 pid 0 comm [ib_core]
. Filter types (strings):
$ rdma res show qp link mlx5_4/1 type UD,gSi
link mlx5_4/1 lqpn 7 type UD state RTS sq-psn 0 pid 0 comm [ib_core]
link mlx5_4/1 lqpn 1 type GSI state RTS sq-psn 0 pid 0 comm [ib_core]
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
The global resource summary information. The object names, current utilization
and maximum numbers are received as is from the kernel.
$ rdma res
1: mlx5_0: pd 3 cq 5 qp 4
2: mlx5_1: pd 3 cq 5 qp 4
3: mlx5_2: pd 3 cq 5 qp 4
4: mlx5_3: pd 2 cq 3 qp 2
5: mlx5_4: pd 3 cq 5 qp 4
$ rdma res show mlx5_4
5: mlx5_4: pd 3 cq 5 qp 44
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
The dev and link execution callbacks expects that next
command line argument is device or port name.
Set pointer to device or port name position prior calls to
rd_exec_dev()/rd_exec_link().
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
This patch adds general infrastructure to RDMAtool to handle various
filtering options needed for the downstream resource tracking patches.
The infrastructure is generic and stores filters in list of key<->value
entries. There are three types of filters:
1. Numeric - the values are intended to be digits combined with '-' to
mark range and ',' to mark multiple entries, e.g. pid 1-100,234,400-401
is perfectly legit filter to limit process ids.
2. String - the values are consist from strings and "," as a denominator.
3. Link - special case to allow '/' in string to provide link name, e.g.
link mlx4_1/2.
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
According to the IBTA spec [1], the physical connected port is provided
for the QP in RTR-to-INIT stage performed by modify_qp(). It causes
to do not have port number for newly created QPs.
The following patch adds "-" sign to present absence of port, because
QPs are going to be associated with rdmatool link object, which needs
port number as an index.
[1] InfiniBand Architecture Release 1.3 -
"Table 96 QP State Transition Properties"
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
The dev_map_free() is called once only and it is short,
so it is better to integrate it into the caller's site.
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: David Ahern <dsahern@gmail.com>
Most of the proposed objects are working in the scope "dev"
and will implement the same logic. Move the code to utils.c,
so other objects will be able to reuse the code.
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: David Ahern <dsahern@gmail.com>
Despite the fact that all callers to dev_map_lookup are ensuring that
there is always device name prior to call to that function, it is better
and safer to check that in the dev_map_lookup itself.
Fixes: 40df8263a0 ("rdma: Add dev object")
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: David Ahern <dsahern@gmail.com>
There is no external users of _dev_map_lookup function,
so let's limit its scope to be local.
Fixes: 40df8263a0 ("rdma: Add dev object")
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: David Ahern <dsahern@gmail.com>
Link (port) object represent struct ib_port to the user space.
Link properties:
* Port capabilities
* IB subnet prefix
* LID, SM_LID and LMC
* Port state
* Physical state
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Device (dev) object represents struct ib_device to the user space.
Device properties:
* Device capabilities
* FW version to the device output
* node_guid and sys_image_guid
* node_type
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
RDMA devices are cross-functional devices from one side,
but very tailored for the specific markets from another.
Such diversity caused to spread of RDMA related configuration
across various tools, e.g. devlink, ip, ethtool, ib specific and
vendor specific solutions.
This patch adds ability to fill device and port information
by reading RDMA netlink.
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
