Add netlink command that enables/disables privileged QKEY by default.
It is disabled by default, since according to IB spec only privileged
users are allowed to use privileged QKEY.
According to the IB specification rel-1.6, section 3.5.3:
"QKEYs with the most significant bit set are considered controlled
QKEYs, and a HCA does not allow a consumer to arbitrarily specify a
controlled QKEY."
Using rdma tool,
$rdma system set privileged-qkey on
When enabled non-privileged users would be able to use
controlled QKEYs which are considered privileged.
Using rdma tool,
$rdma system set privileged-qkey off
When disabled only privileged users would be able to use
controlled QKEYs.
You can also use the command below to check the parameter state:
$rdma system show
netns shared privileged-qkey off copy-on-fork on
Signed-off-by: Patrisious Haddad <phaddad@nvidia.com>
Link: https://lore.kernel.org/r/90398be70a9d23d2aa9d0f9fd11d2c264c1be534.1696848201.git.leon@kernel.org
Signed-off-by: Leon Romanovsky <leon@kernel.org>
The QP usecnts were incremented through QP attributes structure while
decreased through QP itself. Rely on the ib_creat_qp_user() code that
initialized all QP parameters prior returning to the user and increment
exactly like destroy does.
Link: https://lore.kernel.org/r/25d256a3bb1fc480b77d7fe439817b993de48610.1628014762.git.leonro@nvidia.com
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
The low-level create QP function grew to be larger than any sensible
inline function should be. The inline attribute is not really needed for
that function and can be implemented as exported symbol.
Link: https://lore.kernel.org/r/2c08709d86f876c3dfb77684357b2a939e570ca4.1628014762.git.leonro@nvidia.com
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
The ib_create_named_qp() is kernel verb and no kernel users exist that use
XRC_INI QP. Hence such QP path is not reachable. In addition, delete
duplicated assignments of QP attributes from the initialization structure.
Link: https://lore.kernel.org/r/1b4c0d1def5f8f6d26839e14d19da950cc4a0b05.1628014762.git.leonro@nvidia.com
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Convert QP object to follow IB/core general allocation scheme. That
change allows us to make sure that restrack properly kref the memory.
Link: https://lore.kernel.org/r/48e767124758aeecc433360ddd85eaa6325b34d9.1627040189.git.leonro@nvidia.com
Reviewed-by: Gal Pressman <galpress@amazon.com> #efa
Tested-by: Gal Pressman <galpress@amazon.com>
Reviewed-by: Dennis Dalessandro <dennis.dalessandro@cornelisnetworks.com> #rdma and core
Tested-by: Dennis Dalessandro <dennis.dalessandro@cornelisnetworks.com>
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
Tested-by: Tatyana Nikolova <tatyana.e.nikolova@intel.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Removed port validity check from ib_get_cached_subnet_prefix() as this
check is not needed because "port_num" is valid.
Link: https://lore.kernel.org/r/20210616154509.1047-2-anand.a.khoje@oracle.com
Suggested-by: Leon Romanovsky <leonro@nvidia.com>
Signed-off-by: Anand Khoje <anand.a.khoje@oracle.com>
Signed-off-by: Haakon Bugge <haakon.bugge@oracle.com>
Reviewed-by: Leon Romanovsky <leonro@nvidia.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
This code is trying to attach a list of counters grouped into 4 groups to
the ib_port sysfs. Instead of creating a bunch of kobjects simply express
everything naturally as an ib_port_attribute and add a single
attribute_groups list.
Remove all the naked kobject manipulations.
Link: https://lore.kernel.org/r/0d5a7241ee0fe66622de04fcbaafaf6a791d5c7c.1623427137.git.leonro@nvidia.com
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Instead of calling device_add_groups() add the group to the existing
groups array which is managed through device_add().
This requires setting up the hw_counters before device_add(), so it gets
split up from the already split port sysfs flow.
Move all the memory freeing to the release function.
Link: https://lore.kernel.org/r/666250d937b64f6fdf45da9e2dc0b6e5e4f7abd8.1623427137.git.leonro@nvidia.com
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
It is much saner to store a pointer to the kobject structure that contains
the cannonical stats pointer than to copy the stats pointers into a public
structure.
Future patches will require the sysfs pointer for other purposes.
Link: https://lore.kernel.org/r/f90551dfd296cde1cb507bbef27cca9891d19871.1623427137.git.leonro@nvidia.com
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Current code uses many different types when dealing with a port of a RDMA
device: u8, unsigned int and u32. Switch to u32 to clean up the logic.
This allows us to make (at least) the core view consistent and use the
same type. Unfortunately not all places can be converted. Many uverbs
functions expect port to be u8 so keep those places in order not to break
UAPIs. HW/Spec defined values must also not be changed.
With the switch to u32 we now can support devices with more than 255
ports. U32_MAX is reserved to make control logic a bit easier to deal
with. As a device with U32_MAX ports probably isn't going to happen any
time soon this seems like a non issue.
When a device with more than 255 ports is created uverbs will report the
RDMA device as having 255 ports as this is the max currently supported.
The verbs interface is not changed yet because the IBTA spec limits the
port size in too many places to be u8 and all applications that relies in
verbs won't be able to cope with this change. At this stage, we are
extending the interfaces that are using vendor channel solely
Once the limitation is lifted mlx5 in switchdev mode will be able to have
thousands of SFs created by the device. As the only instance of an RDMA
device that reports more than 255 ports will be a representor device and
it exposes itself as a RAW Ethernet only device CM/MAD/IPoIB and other
ULPs aren't effected by this change and their sysfs/interfaces that are
exposes to userspace can remain unchanged.
While here cleanup some alignment issues and remove unneeded sanity
checks (mainly in rdmavt),
Link: https://lore.kernel.org/r/20210301070420.439400-1-leon@kernel.org
Signed-off-by: Mark Bloch <mbloch@nvidia.com>
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
The CQ pool mechanism had two problems:
1. The CQ pool lists were uninitialized in the device registration error
flow. As a result, all the list pointers remained NULL. This caused
the kernel to crash (in procedure ib_cq_pool_destroy) when that error
flow was taken (and unregister called). The stack trace snippet:
BUG: kernel NULL pointer dereference, address: 0000000000000000
#PF: supervisor read access in kernel mode
#PF: error_code(0×0000) ? not-present page
PGD 0 P4D 0
Oops: 0000 [#1] SMP PTI
. . .
RIP: 0010:ib_cq_pool_destroy+0x1b/0×70 [ib_core]
. . .
Call Trace:
disable_device+0x9f/0×130 [ib_core]
__ib_unregister_device+0x35/0×90 [ib_core]
ib_register_device+0x529/0×610 [ib_core]
__mlx5_ib_add+0x3a/0×70 [mlx5_ib]
mlx5_add_device+0x87/0×1c0 [mlx5_core]
mlx5_register_interface+0x74/0xc0 [mlx5_core]
do_one_initcall+0x4b/0×1f4
do_init_module+0x5a/0×223
load_module+0x1938/0×1d40
2. At device unregister, when cleaning up the cq pool, the cq's in the
pool lists were freed, but the cq entries were left in the list.
The fix for the first issue is to initialize the cq pool lists when the
ib_device structure is allocated for a new device (in procedure
_ib_alloc_device).
The fix for the second problem is to delete cq entries from the pool lists
when cleaning up the cq pool.
In addition, procedure ib_cq_pool_destroy() is renamed to the more
appropriate name ib_cq_pool_cleanup().
Fixes: 4aa1615268 ("RDMA/core: Fix ordering of CQ pool destruction")
Link: https://lore.kernel.org/r/20201208073545.9723-2-leon@kernel.org
Suggested-by: Jason Gunthorpe <jgg@nvidia.com>
Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il>
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
The latest changes in restrack name handling allowed to simplify the QP
creation code to support all types of QPs.
For example XRC QP are presented with rdmatool.
$ ibv_xsrq_pingpong &
$ rdma res show qp
link ibp0s9/1 lqpn 0 type SMI state RTS sq-psn 0 comm [ib_core]
link ibp0s9/1 lqpn 1 type GSI state RTS sq-psn 0 comm [ib_core]
link ibp0s9/1 lqpn 7 type UD state RTS sq-psn 0 comm [mlx5_ib]
link ibp0s9/1 lqpn 42 type XRC_TGT state INIT sq-psn 0 path-mig-state MIGRATED comm [ib_uverbs]
link ibp0s9/1 lqpn 43 type XRC_INI state INIT sq-psn 0 path-mig-state MIGRATED pdn 197 pid 419 comm ibv_xsrq_pingpong
Link: https://lore.kernel.org/r/20201117070148.1974114-4-leon@kernel.org
Reviewed-by: Mark Zhang <markz@nvidia.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Special QPs (SMI and GSI) have different rules in regards of their QP
numbers. While all other QP numbers are unique per-device, the QP0 and QP1
are created per-port as requested by IBTA.
In multiple port devices, the number of SMI and GSI QPs with be equal to
the number ports.
$ rdma dev
0: ibp0s9: node_type ca fw 4.4.9999 node_guid 5254:00c0:fe12:3455 sys_image_guid 5254:00c0:fe12:3455
$ rdma link
0/1: ibp0s9/1: subnet_prefix fe80:0000:0000:0000 lid 13397 sm_lid 49151 lmc 0 state ACTIVE physical_state LINK_UP
0/2: ibp0s9/2: subnet_prefix fe80:0000:0000:0000 lid 13397 sm_lid 49151 lmc 0 state UNKNOWN physical_state UNKNOWN
Before:
$ rdma res show qp type SMI,GSI
link ibp0s9/1 lqpn 0 type SMI state RTS sq-psn 0 comm [ib_core]
link ibp0s9/1 lqpn 1 type GSI state RTS sq-psn 0 comm [ib_core]
After:
$ rdma res show qp type SMI,GSI
link ibp0s9/1 lqpn 0 type SMI state RTS sq-psn 0 comm [ib_core]
link ibp0s9/1 lqpn 1 type GSI state RTS sq-psn 0 comm [ib_core]
link ibp0s9/2 lqpn 0 type SMI state RTS sq-psn 0 comm [ib_core]
link ibp0s9/2 lqpn 1 type GSI state RTS sq-psn 0 comm [ib_core]
Link: https://lore.kernel.org/r/20201104144008.3808124-4-leon@kernel.org
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Use rdma_restrack_set_name() and rdma_restrack_parent_name() instead of
tricky uses of rdma_restrack_attach_task()/rdma_restrack_uadd().
This uniformly makes all restracks add'd using rdma_restrack_add().
Link: https://lore.kernel.org/r/20200922091106.2152715-6-leon@kernel.org
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Have a single rdma_restrack_add() that adds an entry, there is no reason
to split the user/kernel here, the rdma_restrack_set_task() is responsible
for this difference.
This patch prepares the code to the future requirement of making restrack
is mandatory for managing ib objects.
Link: https://lore.kernel.org/r/20200922091106.2152715-5-leon@kernel.org
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Refactor the restrack code to make sure the kref inside the restrack entry
properly kref's the object in which it is embedded. This slight change is
needed for future conversions of MR and QP which are refcounted before the
release and kfree.
The ideal flow from ib_core perspective as follows:
* Allocate ib_* structure with rdma_zalloc_*.
* Set everything that is known to ib_core to that newly created object.
* Initialize kref with restrack help
* Call to driver specific allocation functions.
* Insert into restrack DB
....
* Return and release restrack with restrack_put.
Largely this means a rdma_restrack_new() should be called near allocating
the containing structure.
Link: https://lore.kernel.org/r/20200922091106.2152715-4-leon@kernel.org
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Allow a ULP to ask the core to provide a completion queue based on a
least-used search on a per-device CQ pools. The device CQ pools grow in a
lazy fashion when more CQs are requested.
This feature reduces the amount of interrupts when using many QPs. Using
shared CQs allows for more effcient completion handling. It also reduces
the amount of overhead needed for CQ contexts.
Test setup:
Intel(R) Xeon(R) Platinum 8176M CPU @ 2.10GHz servers.
Running NVMeoF 4KB read IOs over ConnectX-5EX across Spectrum switch.
TX-depth = 32. The patch was applied in the nvme driver on both the target
and initiator. Four controllers are accessed from each core. In the
current test case we have exposed sixteen NVMe namespaces using four
different subsystems (four namespaces per subsystem) from one NVM port.
Each controller allocated X queues (RDMA QPs) and attached to Y CQs.
Before this series we had X == Y, i.e for four controllers we've created
total of 4X QPs and 4X CQs. In the shared case, we've created 4X QPs and
only X CQs which means that we have four controllers that share a
completion queue per core. Until fourteen cores there is no significant
change in performance and the number of interrupts per second is less than
a million in the current case.
==================================================
|Cores|Current KIOPs |Shared KIOPs |improvement|
|-----|---------------|--------------|-----------|
|14 |2332 |2723 |16.7% |
|-----|---------------|--------------|-----------|
|20 |2086 |2712 |30% |
|-----|---------------|--------------|-----------|
|28 |1971 |2669 |35.4% |
|=================================================
|Cores|Current avg lat|Shared avg lat|improvement|
|-----|---------------|--------------|-----------|
|14 |767us |657us |14.3% |
|-----|---------------|--------------|-----------|
|20 |1225us |943us |23% |
|-----|---------------|--------------|-----------|
|28 |1816us |1341us |26.1% |
========================================================
|Cores|Current interrupts|Shared interrupts|improvement|
|-----|------------------|-----------------|-----------|
|14 |1.6M/sec |0.4M/sec |72% |
|-----|------------------|-----------------|-----------|
|20 |2.8M/sec |0.6M/sec |72.4% |
|-----|------------------|-----------------|-----------|
|28 |2.9M/sec |0.8M/sec |63.4% |
====================================================================
|Cores|Current 99.99th PCTL lat|Shared 99.99th PCTL lat|improvement|
|-----|------------------------|-----------------------|-----------|
|14 |67ms |6ms |90.9% |
|-----|------------------------|-----------------------|-----------|
|20 |5ms |6ms |-10% |
|-----|------------------------|-----------------------|-----------|
|28 |8.7ms |6ms |25.9% |
|===================================================================
Performance improvement with sixteen disks (sixteen CQs per core) is
comparable.
Link: https://lore.kernel.org/r/1590568495-101621-3-git-send-email-yaminf@mellanox.com
Signed-off-by: Yamin Friedman <yaminf@mellanox.com>
Reviewed-by: Or Gerlitz <ogerlitz@mellanox.com>
Reviewed-by: Max Gurtovoy <maxg@mellanox.com>
Reviewed-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
Currently when the low level driver notifies Pkey, GID, and port change
events they are notified to the registered handlers in the order they are
registered.
IB core and other ULPs such as IPoIB are interested in GID, LID, Pkey
change events.
Since all GID queries done by ULPs are serviced by IB core, and the IB
core deferes cache updates to a work queue, it is possible for other
clients to see stale cache data when they handle their own events.
For example, the below call tree shows how ipoib will call
rdma_query_gid() concurrently with the update to the cache sitting in the
WQ.
mlx5_ib_handle_event()
ib_dispatch_event()
ib_cache_event()
queue_work() -> slow cache update
[..]
ipoib_event()
queue_work()
[..]
work handler
ipoib_ib_dev_flush_light()
__ipoib_ib_dev_flush()
ipoib_dev_addr_changed_valid()
rdma_query_gid() <- Returns old GID, cache not updated.
Move all the event dispatch to a work queue so that the cache update is
always done before any clients are notified.
Fixes: f35faa4ba9 ("IB/core: Simplify ib_query_gid to always refer to cache")
Link: https://lore.kernel.org/r/20191212113024.336702-3-leon@kernel.org
Signed-off-by: Parav Pandit <parav@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Reviewed-by: Jason Gunthorpe <jgg@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
Mainly a collection of smaller of driver updates this cycle.
- Various driver updates and bug fixes for siw, bnxt_re, hns, qedr,
iw_cxgb4, vmw_pvrdma, mlx5
- Improvements in SRPT from working with iWarp
- SRIOV VF support for bnxt_re
- Skeleton kernel-doc files for drivers/infiniband
- User visible counters for events related to ODP
- Common code for tracking of mmap lifetimes so that drivers can link HW
object liftime to a VMA
- ODP bug fixes and rework
- RDMA READ support for efa
- Removal of the very old cxgb3 driver
-----BEGIN PGP SIGNATURE-----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=UV2u
-----END PGP SIGNATURE-----
Merge tag 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma
Pull rdma updates from Jason Gunthorpe:
"Again another fairly quiet cycle with few notable core code changes
and the usual variety of driver bug fixes and small improvements.
- Various driver updates and bug fixes for siw, bnxt_re, hns, qedr,
iw_cxgb4, vmw_pvrdma, mlx5
- Improvements in SRPT from working with iWarp
- SRIOV VF support for bnxt_re
- Skeleton kernel-doc files for drivers/infiniband
- User visible counters for events related to ODP
- Common code for tracking of mmap lifetimes so that drivers can link
HW object liftime to a VMA
- ODP bug fixes and rework
- RDMA READ support for efa
- Removal of the very old cxgb3 driver"
* tag 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma: (168 commits)
RDMA/hns: Delete unnecessary callback functions for cq
RDMA/hns: Rename the functions used inside creating cq
RDMA/hns: Redefine the member of hns_roce_cq struct
RDMA/hns: Redefine interfaces used in creating cq
RDMA/efa: Expose RDMA read related attributes
RDMA/efa: Support remote read access in MR registration
RDMA/efa: Store network attributes in device attributes
IB/hfi1: remove redundant assignment to variable ret
RDMA/bnxt_re: Fix missing le16_to_cpu
RDMA/bnxt_re: Fix stat push into dma buffer on gen p5 devices
RDMA/bnxt_re: Fix chip number validation Broadcom's Gen P5 series
RDMA/bnxt_re: Fix Kconfig indentation
IB/mlx5: Implement callbacks for getting VFs GUID attributes
IB/ipoib: Add ndo operation for getting VFs GUID attributes
IB/core: Add interfaces to get VF node and port GUIDs
net/core: Add support for getting VF GUIDs
RDMA/qedr: Fix null-pointer dereference when calling rdma_user_mmap_get_offset
RDMA/cm: Use refcount_t type for refcount variable
IB/mlx5: Support extended number of strides for Striding RQ
IB/mlx4: Update HW GID table while adding vlan GID
...
The rdma_user_mmap_io interface created a common interface for drivers to
correctly map hw resources and zap them once the ucontext is destroyed
enabling the drivers to safely free the hw resources.
However, this meant the drivers need to delay freeing the resource to the
ucontext destroy phase to ensure they were no longer mapped. The new
mechanism for a common way of handling user/driver address mapping enabled
notifying the driver if all umap_priv mappings were removed, and enabled
freeing the hw resources when they are done with and not delay it until
ucontext destroy.
Since not all drivers use the mechanism, NULL can be sent to the
rdma_user_mmap_io interface to continue working as before. Drivers that
use the mmap_xa interface can pass the entry being mapped to the
rdma_user_mmap_io function to be linked together.
Link: https://lore.kernel.org/r/20191030094417.16866-4-michal.kalderon@marvell.com
Signed-off-by: Ariel Elior <ariel.elior@marvell.com>
Signed-off-by: Michal Kalderon <michal.kalderon@marvell.com>
Reviewed-by: Jason Gunthorpe <jgg@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
Move functionality that is called by the driver, which is
related to umap, to a new file that will be linked in ib_core.
This is a first step in later enabling ib_uverbs to be optional.
vm_ops is now initialized in ib_uverbs_mmap instead of
priv_init to avoid having to move all the rdma_umap functions
as well.
Link: https://lore.kernel.org/r/20191030094417.16866-2-michal.kalderon@marvell.com
Suggested-by: Jason Gunthorpe <jgg@mellanox.com>
Signed-off-by: Ariel Elior <ariel.elior@marvell.com>
Signed-off-by: Michal Kalderon <michal.kalderon@marvell.com>
Reviewed-by: Jason Gunthorpe <jgg@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
When rdmacm module is not loaded, and when netlink message is received to
get char device info, it results into a deadlock due to recursive locking
of rdma_nl_mutex with the below call sequence.
[..]
rdma_nl_rcv()
mutex_lock()
[..]
rdma_nl_rcv_msg()
ib_get_client_nl_info()
request_module()
iw_cm_init()
rdma_nl_register()
mutex_lock(); <- Deadlock, acquiring mutex again
Due to above call sequence, following call trace and deadlock is observed.
kernel: __mutex_lock+0x35e/0x860
kernel: ? __mutex_lock+0x129/0x860
kernel: ? rdma_nl_register+0x1a/0x90 [ib_core]
kernel: rdma_nl_register+0x1a/0x90 [ib_core]
kernel: ? 0xffffffffc029b000
kernel: iw_cm_init+0x34/0x1000 [iw_cm]
kernel: do_one_initcall+0x67/0x2d4
kernel: ? kmem_cache_alloc_trace+0x1ec/0x2a0
kernel: do_init_module+0x5a/0x223
kernel: load_module+0x1998/0x1e10
kernel: ? __symbol_put+0x60/0x60
kernel: __do_sys_finit_module+0x94/0xe0
kernel: do_syscall_64+0x5a/0x270
kernel: entry_SYSCALL_64_after_hwframe+0x49/0xbe
process stack trace:
[<0>] __request_module+0x1c9/0x460
[<0>] ib_get_client_nl_info+0x5e/0xb0 [ib_core]
[<0>] nldev_get_chardev+0x1ac/0x320 [ib_core]
[<0>] rdma_nl_rcv_msg+0xeb/0x1d0 [ib_core]
[<0>] rdma_nl_rcv+0xcd/0x120 [ib_core]
[<0>] netlink_unicast+0x179/0x220
[<0>] netlink_sendmsg+0x2f6/0x3f0
[<0>] sock_sendmsg+0x30/0x40
[<0>] ___sys_sendmsg+0x27a/0x290
[<0>] __sys_sendmsg+0x58/0xa0
[<0>] do_syscall_64+0x5a/0x270
[<0>] entry_SYSCALL_64_after_hwframe+0x49/0xbe
To overcome this deadlock and to allow multiple netlink messages to
progress in parallel, following scheme is implemented.
1. Split the lock protecting the cb_table into a per-index lock, and make
it a rwlock. This lock is used to ensure no callbacks are running after
unregistration returns. Since a module will not be registered once it
is already running callbacks, this avoids the deadlock.
2. Use smp_store_release() to update the cb_table during registration so
that no lock is required. This avoids lockdep problems with thinking
all the rwsems are the same lock class.
Fixes: 0e2d00eb6f ("RDMA: Add NLDEV_GET_CHARDEV to allow char dev discovery and autoload")
Link: https://lore.kernel.org/r/20191015080733.18625-1-leon@kernel.org
Signed-off-by: Parav Pandit <parav@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
Jason Gunthorpe says:
====================
This is a collection of general cleanups for ODP to clarify some of the
flows around umem creation and use of the interval tree.
====================
The branch is based on v5.3-rc5 due to dependencies
* odp_fixes:
RDMA/mlx5: Use odp instead of mr->umem in pagefault_mr
RDMA/mlx5: Use ib_umem_start instead of umem.address
RDMA/core: Make invalidate_range a device operation
RDMA/odp: Use kvcalloc for the dma_list and page_list
RDMA/odp: Check for overflow when computing the umem_odp end
RDMA/odp: Provide ib_umem_odp_release() to undo the allocs
RDMA/odp: Split creating a umem_odp from ib_umem_get
RDMA/odp: Make the three ways to create a umem_odp clear
RMDA/odp: Consolidate umem_odp initialization
RDMA/odp: Make it clearer when a umem is an implicit ODP umem
RDMA/odp: Iterate over the whole rbtree directly
RDMA/odp: Use the common interval tree library instead of generic
RDMA/mlx5: Fix MR npages calculation for IB_ACCESS_HUGETLB
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
The check for QP type different than XRC has excluded driver QP
types from the resource tracker.
As a result, "rdma resource show" user command would not show opened
driver QPs which does not reflect the real state of the system.
Check QP type explicitly instead of assuming enum values/ordering.
Fixes: 40909f664d ("RDMA/efa: Add EFA verbs implementation")
Signed-off-by: Gal Pressman <galpress@amazon.com>
Reviewed-by: Leon Romanovsky <leonro@mellanox.com>
Link: https://lore.kernel.org/r/20190801104354.11417-1-galpress@amazon.com
Signed-off-by: Doug Ledford <dledford@redhat.com>
Now that IB core supports RDMA device binding with specific net namespace,
enable IB core to accept netlink commands in non init_net namespaces.
This is done by having per net namespace netlink socket.
At present only netlink device handling client RDMA_NL_NLDEV supports
device handling in multiple net namespaces. Hence do not accept netlink
messages for other clients in non init_net net namespaces.
Link: https://lore.kernel.org/r/20190723070205.6247-1-leon@kernel.org
Signed-off-by: Parav Pandit <parav@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Reviewed-by: Jason Gunthorpe <jgg@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
Added parameter in ib_device for enabling dynamic interrupt moderation so
that it can be configured in userspace using rdma tool.
In order to set adaptive-moderation for an ib device the command is:
rdma dev set [DEV] adaptive-moderation [on|off]
Please set on/off.
rdma dev show
0: mlx5_0: node_type ca fw 16.26.0055 node_guid 248a:0703:00a5:29d0
sys_image_guid 248a:0703:00a5:29d0 adaptive-moderation on
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>
Reviewed-by: Sagi Grimberg <sagi@grimberg.me>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
Allow userspace to issue a netlink query against the ib_device for
something like "uverbs" and get back the char dev name, inode major/minor,
and interface ABI information for "uverbs0".
Since we are now in netlink this can also trigger a module autoload to
make the uverbs device come into existence.
Largely this will let us replace searching and reading inside sysfs to
setup devices, and provides an alternative (using driver_id) to device
name based provider binding for things like rxe.
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
Signed-off-by: Doug Ledford <dledford@redhat.com>
The driver interface cannot manipulate the sysfs of the compat device,
only of the full device so we must avoid calling the driver sysfs APIs on
compat devices.
This prevents an oops:
Call Trace:
dump_stack+0x5a/0x73
kobject_init+0x74/0x80
kobject_init_and_add+0x35/0xb0
hfi1_create_port_files+0x6e/0x3c0 [hfi1]
ib_setup_port_attrs+0x43b/0x560 [ib_core]
add_one_compat_dev+0x16a/0x230 [ib_core]
rdma_dev_init_net+0x110/0x160 [ib_core]
ops_init+0x38/0xf0
setup_net+0xcf/0x1e0
copy_net_ns+0xb7/0x130
create_new_namespaces+0x11a/0x1b0
unshare_nsproxy_namespaces+0x55/0xa0
ksys_unshare+0x1a7/0x340
__x64_sys_unshare+0xe/0x20
do_syscall_64+0x5b/0x180
entry_SYSCALL_64_after_hwframe+0x44/0xa9
Fixes: 5417783eab ("RDMA/core: Support core port attributes in non init_net")
Reported-by: Mike Marciniszyn <mike.marciniszyn@intel.com>
Tested-by: Mike Marciniszyn <mike.marciniszyn@intel.com>
Signed-off-by: Parav Pandit <parav@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
real_qp should be initialized before ib_destroy_qp() is called.
ib_destroy_qp() may be called in the error flow if ib_create_qp_security()
failed.
Signed-off-by: Artemy Kovalyov <artemyko@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
Provide an option to change the net namespace of a rdma device through a
netlink command. When multiple rdma devices exists in a system, and when
containers are used, this will limit rdma device visibility to a specified
net namespace.
An example command to change net namespace of mlx5_1 device to the
previously created net namespace 'foo' is:
$ ip netns add foo
$ rdma dev set mlx5_1 netns foo
Signed-off-by: Parav Pandit <parav@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
The sysfs layout is created by CM incorrectly presented RDMA devices with
InfiniBand link layer. Layout of such devices represents device tree of
connections. By moving CM statistics to be under relevant port of IB
device, we will fix the following issues:
* Symlink name - It used device name instead of specific identifier.
* Target location - It was supposed to point to PCI-ID/infiniband_cm/
instead of PCI-ID/infiniband/
* Target name - It created extra device file under already existing
device folder, e.g. mlx5_0/mlx5_0
* Crash during boot with RDMA persistent naming patches.
sysfs: cannot create duplicate filename '/class/infiniband_cm/mlx5_0'
CPU: 29 PID: 433 Comm: modprobe Not tainted 5.0.0-rc5+ #178
Call Trace:
dump_stack+0xcc/0x180
sysfs_warn_dup.cold.3+0x17/0x2d
sysfs_do_create_link_sd.isra.2+0xd0/0xf0
device_add+0x7cb/0x1450
device_create_groups_vargs+0x1ae/0x220
device_create+0x93/0xc0
cm_add_one+0x38f/0xf60 [ib_cm]
add_client_context+0x167/0x210 [ib_core]
enable_device_and_get+0x230/0x3f0 [ib_core]
ib_register_device+0x823/0xbf0 [ib_core]
__mlx5_ib_add+0x45/0x150 [mlx5_ib]
mlx5_ib_add+0x1b3/0x5e0 [mlx5_ib]
mlx5_add_device+0x130/0x3a0 [mlx5_core]
mlx5_register_interface+0x1a9/0x270 [mlx5_core]
do_one_initcall+0x14f/0x5de
do_init_module+0x247/0x7c0
load_module+0x4c2f/0x60d0
entry_SYSCALL_64_after_hwframe+0x49/0xbe
After this change:
[leonro@server ~]$ ls -al /sys/class/infiniband/ibp0s12f0/ports/1/
drwxr-xr-x 2 root root 0 Mar 11 11:17 cm_rx_duplicates
drwxr-xr-x 2 root root 0 Mar 11 11:17 cm_rx_msgs
drwxr-xr-x 2 root root 0 Mar 11 11:17 cm_tx_msgs
drwxr-xr-x 2 root root 0 Mar 11 11:17 cm_tx_retries
Fixes: 110cf374a8 ("infiniband: make cm_device use a struct device and not a kobject.")
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
Add netlink command that enables/disables sharing rdma device among
multiple net namespaces.
Using rdma tool,
$rdma sys set netns shared (default mode)
When rdma subsystem netns mode is set to shared mode, rdma devices
will be accessible in all net namespaces.
Using rdma tool,
$rdma sys set netns exclusive
When rdma subsystem netns mode is set to exclusive mode, devices
will be accessible in only one net namespace at any given
point of time.
If there are any net namespaces other than default init_net exists,
while executing this command, it will fail and mode cannot be changed.
To change this mode, netlink command is used instead of sysctl, because
netlink command allows to auto load a module.
Signed-off-by: Parav Pandit <parav@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
Add an interface via netlink command to query whether rdma devices are
shared among multiple net namespaces or not. When using RDMAtool, it can
be queried as,
$rdma system show netns
netns shared
Signed-off-by: Parav Pandit <parav@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
Extend ib_device_get_by_index() API to check device access for
net namespace for serving netlink commands.
Also enforce net ns check on dumpit commands which iterate over all
registered rdma devices and which don't call ib_device_get_by_index().
Signed-off-by: Parav Pandit <parav@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
Now that sysfs compatibility layer for non init_net exists, add core port
attributes such as pkey and gid table to non init_net ns.
Signed-off-by: Parav Pandit <parav@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
The associated netdev should not actually be very dynamic, so for most
drivers there is no reason for a callback like this. Provide an API to
inform the core code about the net dev affiliation and use a core
maintained data structure instead.
This allows the core code to be more aware of the ndev relationship which
will allow some new APIs based around this.
This also uses locking that makes some kind of sense, many drivers had a
confusing RCU lock, or missing locking which isn't right.
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
Move core device addition and removal from sysfs.c to device.c as device.c
is more appropriate place for device management.
Signed-off-by: Parav Pandit <parav@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
Since this only frees memory it should be done during the release
callback. Otherwise there are possible error flows where it might not get
called if registration aborts.
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
When creating many MAD agents in a short period of time, receive packet
processing can be delayed long enough to cause timeouts while new agents
are being added to the atomic notifier chain with IRQs disabled. Notifier
chain registration and unregstration is an O(n) operation. With large
numbers of MAD agents being created and destroyed simultaneously the CPUs
spend too much time with interrupts disabled.
Instead of each MAD agent registering for it's own LSM notification,
maintain a list of agents internally and register once, this registration
already existed for handling the PKeys. This list is write mostly, so a
normal spin lock is used vs a read/write lock. All MAD agents must be
checked, so a single list is used instead of breaking them down per
device.
Notifier calls are done under rcu_read_lock, so there isn't a risk of
similar packet timeouts while checking the MAD agents security settings
when notified.
Signed-off-by: Daniel Jurgens <danielj@mellanox.com>
Reviewed-by: Parav Pandit <parav@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Acked-by: Paul Moore <paul@paul-moore.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
-----BEGIN PGP SIGNATURE-----
iQFSBAABCAA8FiEEq68RxlopcLEwq+PEeb4+QwBBGIYFAlxXYaEeHHRvcnZhbGRz
QGxpbnV4LWZvdW5kYXRpb24ub3JnAAoJEHm+PkMAQRiGkSQH/2yrfnviNPFYpZOR
QQdc71Bfhkd8m85SmWIsSebkxmi3hKFVj15sGbWXd6+0/VxjEEGvQCZpvVwJceke
LwDxtkKGg/74wAqJvlSAWxFNZ+Had4jDeoSoeQChddsBVXBBCxQx2v6ECg3o2x7W
k8Z8t4+3RijDf8fYXY9ETyO2zW8R/wgT+dnl+DPgUH7u4dxh7FzAUfc4bgZIDg+i
FzBQfbTJuz4BU7uRZ9IJiwhWKv0Iyi2DR3BY8Z1pqEpRaUMJMrCs2WGytHbTgt9e
0EtO1airbVneU4eumU/ZaF9cyEbah9HousEPnP7J09WG4s/Odxc4zE+uK1QqS2im
5Xv88is=
=dVd1
-----END PGP SIGNATURE-----
Merge tag 'v5.0-rc5' into rdma.git for-next
Linux 5.0-rc5
Needed to merge the include/uapi changes so we have an up to date
single-tree for these files. Patches already posted are also expected to
need this for dependencies.
It turns out future patches need this capability quite widely now, not
just for netlink, so provide two global functions to manage the
registration lock refcount.
This also moves the point the lock becomes 1 to within
ib_register_device() so that the semantics of the public API are very sane
and clear. Calling ib_device_try_get() will fail on devices that are only
allocated but not yet registered.
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
Reviewed-by: Steve Wise <swise@opengridcomputing.com>
Reviewed-by: Parav Pandit <parav@mellanox.com>
RDMA cgroup registration routine always returns success, so simplify
function to be void and run clang formatter over whole CONFIG_CGROUP_RDMA
art of core_priv.h.
This reduces unwinding error path for regular registration and future net
namespace change functionality for rdma device.
Signed-off-by: Parav Pandit <parav@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Acked-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
Most provider routines are callback routines which ib core invokes.
_callback suffix doesn't convey information about when such callback is
invoked. Therefore, rename port_callback to init_port.
Additionally, store the init_port function pointer in ib_device_ops, so
that it can be accessed in subsequent patches when binding rdma device to
net namespace.
Signed-off-by: Parav Pandit <parav@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
Having uobject pointer embedded in ib core objects is not aligned with a
future shared ib_x model. The resource tracker only does this to keep
track of user/kernel objects - track this directly instead.
Signed-off-by: Shamir Rabinovitch <shamir.rabinovitch@oracle.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
Make all the required change to start use the ib_device_ops structure.
Signed-off-by: Kamal Heib <kamalheib1@gmail.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
When the rdma device is getting removed, get resource info can race with
device removal, as below:
CPU-0 CPU-1
-------- --------
rdma_nl_rcv_msg()
nldev_res_get_cq_dumpit()
mutex_lock(device_lock);
get device reference
mutex_unlock(device_lock); [..]
ib_unregister_device()
/* Valid reference to
* device->dev exists.
*/
ib_dealloc_device()
[..]
provider->fill_res_entry();
Even though device object is not freed, fill_res_entry() can get called on
device which doesn't have a driver anymore. Kernel core device reference
count is not sufficient, as this only keeps the structure valid, and
doesn't guarantee the driver is still loaded.
Similar race can occur with device renaming and device removal, where
device_rename() tries to rename a unregistered device. While this is fine
for devices of a class which are not net namespace aware, but it is
incorrect for net namespace aware class coming in subsequent series. If a
class is net namespace aware, then the below [1] call trace is observed in
above situation.
Therefore, to avoid the race, keep a reference count and let device
unregistration wait until all netlink users drop the reference.
[1] Call trace:
kernfs: ns required in 'infiniband' for 'mlx5_0'
WARNING: CPU: 18 PID: 44270 at fs/kernfs/dir.c:842 kernfs_find_ns+0x104/0x120
libahci i2c_core mlxfw libata dca [last unloaded: devlink]
RIP: 0010:kernfs_find_ns+0x104/0x120
Call Trace:
kernfs_find_and_get_ns+0x2e/0x50
sysfs_rename_link_ns+0x40/0xb0
device_rename+0xb2/0xf0
ib_device_rename+0xb3/0x100 [ib_core]
nldev_set_doit+0x165/0x190 [ib_core]
rdma_nl_rcv_msg+0x249/0x250 [ib_core]
? netlink_deliver_tap+0x8f/0x3e0
rdma_nl_rcv+0xd6/0x120 [ib_core]
netlink_unicast+0x17c/0x230
netlink_sendmsg+0x2f0/0x3e0
sock_sendmsg+0x30/0x40
__sys_sendto+0xdc/0x160
Fixes: da5c850782 ("RDMA/nldev: add driver-specific resource tracking")
Signed-off-by: Parav Pandit <parav@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
Currently several rdma_cm module specific functions are declared in
core_priv.h file. Now that we have cma_priv.h file specific to rdma_cm
kernel module, move them from core_priv.h to cma_priv.h
Signed-off-by: Parav Pandit <parav@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
