Commit Graph

1712 Commits

Author SHA1 Message Date
Linus Torvalds 19fd08b85b Merge candidates for 4.17 merge window
- Fix RDMA uapi headers to actually compile in userspace and be more
   complete
 
 - Three shared with netdev pull requests from Mellanox:
 
    * 7 patches, mostly to net with 1 IB related one at the back). This
      series addresses an IRQ performance issue (patch 1), cleanups related to
      the fix for the IRQ performance problem (patches 2-6), and then extends
      the fragmented completion queue support that already exists in the net
      side of the driver to the ib side of the driver (patch 7).
 
    * Mostly IB, with 5 patches to net that are needed to support the remaining
      10 patches to the IB subsystem. This series extends the current
      'representor' framework when the mlx5 driver is in switchdev mode from
      being a netdev only construct to being a netdev/IB dev construct. The IB
      dev is limited to raw Eth queue pairs only, but by having an IB dev of
      this type attached to the representor for a switchdev port, it enables
      DPDK to work on the switchdev device.
 
    * All net related, but needed as infrastructure for the rdma driver
 
 - Updates for the hns, i40iw, bnxt_re, cxgb3, cxgb4, hns drivers
 
 - SRP performance updates
 
 - IB uverbs write path cleanup patch series from Leon
 
 - Add RDMA_CM support to ib_srpt. This is disabled by default.  Users need to
   set the port for ib_srpt to listen on in configfs in order for it to be
   enabled (/sys/kernel/config/target/srpt/discovery_auth/rdma_cm_port)
 
 - TSO and Scatter FCS support in mlx4
 
 - Refactor of modify_qp routine to resolve problems seen while working on new
   code that is forthcoming
 
 - More refactoring and updates of RDMA CM for containers support from Parav
 
 - mlx5 'fine grained packet pacing', 'ipsec offload' and 'device memory'
   user API features
 
 - Infrastructure updates for the new IOCTL interface, based on increased usage
 
 - ABI compatibility bug fixes to fully support 32 bit userspace on 64 bit
   kernel as was originally intended. See the commit messages for
   extensive details
 
 - Syzkaller bugs and code cleanups motivated by them
 -----BEGIN PGP SIGNATURE-----
 Version: GnuPG v2
 
 iQIcBAABCgAGBQJax5Z0AAoJEDht9xV+IJsacCwQAJBIgmLCvVp5fBu2kJcXMMVI
 y3l2YNzAUJvDDKv1r5yTC9ugBXEkDtgzi/W/C2/5es2yUG/QeT/zzQ3YPrtsnN68
 5FkiXQ35Tt7+PBHMr0cacGRmF4M3Td3MeW0X5aJaBKhqlNKwA+aF18pjGWBmpVYx
 URYCwLb5BZBKVh4+1Leebsk4i0/7jSauAqE5M+9notuAUfBCoY1/Eve3DipEIBBp
 EyrEnMDIdujYRsg4KHlxFKKJ1EFGItknLQbNL1+SEa0Oe0SnEl5Bd53Yxfz7ekNP
 oOWQe5csTcs3Yr4Ob0TC+69CzI71zKbz6qPDILTwXmsPFZJ9ipJs4S8D6F7ra8tb
 D5aT1EdRzh/vAORPC9T3DQ3VsHdvhwpUMG7knnKrVT9X/g7E+gSji1BqaQaTr/xs
 i40GepHT7lM/TWEuee/6LRpqdhuOhud7vfaRFwn2JGRX9suqTcvwhkBkPUDGV5XX
 5RkHcWOb/7KvmpG7S1gaRGK5kO208LgmAZi7REaJFoZB74FqSneMR6NHIH07ha41
 Zou7rnxV68CT2bgu27m+72EsprgmBkVDeEzXgKxVI/+PZ1oadUFpgcZ3pRLOPWVx
 rEqjHu65rlA/YPog4iXQaMfSwt/oRD3cVJS/n8EdJKXi4Qt2RDDGdyOmt74w4prM
 QuLEdvJIFmwrND1KDoqn
 =Ku8g
 -----END PGP SIGNATURE-----

Merge tag 'for-linus-unmerged' of git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma

Pull rdma updates from Jason Gunthorpe:
 "Doug and I are at a conference next week so if another PR is sent I
  expect it to only be bug fixes. Parav noted yesterday that there are
  some fringe case behavior changes in his work that he would like to
  fix, and I see that Intel has a number of rc looking patches for HFI1
  they posted yesterday.

  Parav is again the biggest contributor by patch count with his ongoing
  work to enable container support in the RDMA stack, followed by Leon
  doing syzkaller inspired cleanups, though most of the actual fixing
  went to RC.

  There is one uncomfortable series here fixing the user ABI to actually
  work as intended in 32 bit mode. There are lots of notes in the commit
  messages, but the basic summary is we don't think there is an actual
  32 bit kernel user of drivers/infiniband for several good reasons.

  However we are seeing people want to use a 32 bit user space with 64
  bit kernel, which didn't completely work today. So in fixing it we
  required a 32 bit rxe user to upgrade their userspace. rxe users are
  still already quite rare and we think a 32 bit one is non-existing.

   - Fix RDMA uapi headers to actually compile in userspace and be more
     complete

   - Three shared with netdev pull requests from Mellanox:

      * 7 patches, mostly to net with 1 IB related one at the back).
        This series addresses an IRQ performance issue (patch 1),
        cleanups related to the fix for the IRQ performance problem
        (patches 2-6), and then extends the fragmented completion queue
        support that already exists in the net side of the driver to the
        ib side of the driver (patch 7).

      * Mostly IB, with 5 patches to net that are needed to support the
        remaining 10 patches to the IB subsystem. This series extends
        the current 'representor' framework when the mlx5 driver is in
        switchdev mode from being a netdev only construct to being a
        netdev/IB dev construct. The IB dev is limited to raw Eth queue
        pairs only, but by having an IB dev of this type attached to the
        representor for a switchdev port, it enables DPDK to work on the
        switchdev device.

      * All net related, but needed as infrastructure for the rdma
        driver

   - Updates for the hns, i40iw, bnxt_re, cxgb3, cxgb4, hns drivers

   - SRP performance updates

   - IB uverbs write path cleanup patch series from Leon

   - Add RDMA_CM support to ib_srpt. This is disabled by default. Users
     need to set the port for ib_srpt to listen on in configfs in order
     for it to be enabled
     (/sys/kernel/config/target/srpt/discovery_auth/rdma_cm_port)

   - TSO and Scatter FCS support in mlx4

   - Refactor of modify_qp routine to resolve problems seen while
     working on new code that is forthcoming

   - More refactoring and updates of RDMA CM for containers support from
     Parav

   - mlx5 'fine grained packet pacing', 'ipsec offload' and 'device
     memory' user API features

   - Infrastructure updates for the new IOCTL interface, based on
     increased usage

   - ABI compatibility bug fixes to fully support 32 bit userspace on 64
     bit kernel as was originally intended. See the commit messages for
     extensive details

   - Syzkaller bugs and code cleanups motivated by them"

* tag 'for-linus-unmerged' of git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma: (199 commits)
  IB/rxe: Fix for oops in rxe_register_device on ppc64le arch
  IB/mlx5: Device memory mr registration support
  net/mlx5: Mkey creation command adjustments
  IB/mlx5: Device memory support in mlx5_ib
  net/mlx5: Query device memory capabilities
  IB/uverbs: Add device memory registration ioctl support
  IB/uverbs: Add alloc/free dm uverbs ioctl support
  IB/uverbs: Add device memory capabilities reporting
  IB/uverbs: Expose device memory capabilities to user
  RDMA/qedr: Fix wmb usage in qedr
  IB/rxe: Removed GID add/del dummy routines
  RDMA/qedr: Zero stack memory before copying to user space
  IB/mlx5: Add ability to hash by IPSEC_SPI when creating a TIR
  IB/mlx5: Add information for querying IPsec capabilities
  IB/mlx5: Add IPsec support for egress and ingress
  {net,IB}/mlx5: Add ipsec helper
  IB/mlx5: Add modify_flow_action_esp verb
  IB/mlx5: Add implementation for create and destroy action_xfrm
  IB/uverbs: Introduce ESP steering match filter
  IB/uverbs: Add modify ESP flow_action
  ...
2018-04-06 17:35:43 -07:00
Ariel Levkovich be934cca9e IB/uverbs: Add device memory registration ioctl support
Adding new ioctl method for the MR object - REG_DM_MR.

This command can be used by users to register an allocated
device memory buffer as an MR and receive lkey and rkey
to be used within work requests.

It is added as a new method under the MR object and using a new
ib_device callback - reg_dm_mr.
The command creates a standard ib_mr object which represents the
registered memory.

Signed-off-by: Ariel Levkovich <lariel@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
2018-04-05 11:16:39 -06:00
Ariel Levkovich bee76d7ab5 IB/uverbs: Add alloc/free dm uverbs ioctl support
This change adds uverbs support for allocation/freeing
of device memory commands.

A new uverbs object is defined of type idr to represent
and track the new resource type allocation per context.

The API requires provider driver to implement 2 new ib_device
callbacks - one for allocation and one for deallocation which
return and accept (respectively) the ib_dm object which represents
the allocated memory on the device.

The support is added via the ioctl command infrastructure
only.

Signed-off-by: Ariel Levkovich <lariel@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
2018-04-05 11:16:39 -06:00
Ariel Levkovich 1d8eeb9f6a IB/uverbs: Add device memory capabilities reporting
This change allows vendors to report device memory capability
max_dm_size - to user via uverbs command.

Signed-off-by: Ariel Levkovich <lariel@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
2018-04-05 11:16:39 -06:00
Matan Barak 56ab0b38b8 IB/uverbs: Introduce ESP steering match filter
Adding a new ESP steering match filter that could match against
spi and seq used in IPSec protocol.

Reviewed-by: Yishai Hadas <yishaih@mellanox.com>
Signed-off-by: Matan Barak <matanb@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
2018-04-04 12:06:26 -06:00
Matan Barak 7d12f8d5a1 IB/uverbs: Add modify ESP flow_action
flow_actions of ESP type could be modified during runtime. This could be
common for example when ESN should be changed. Adding a new
UVERBS_FLOW_ACTION_ESP_MODIFY method for changing ESP parameters of an
existing ESP flow_action.
The new method uses the UVERBS_FLOW_ACTION_ESP_CREATE attributes, but
adds a new IB_FLOW_ACTION_ESP_FLAGS_MOD_ESP_ATTRS which means ESP_ATTRS
should be changed.
In addition, we add a new FLOW_ACTION_ESP_REPLAY_NONE replay type that
could be used when one wants to disable a replay protection over a
specific flow_action.

Reviewed-by: Yishai Hadas <yishaih@mellanox.com>
Signed-off-by: Matan Barak <matanb@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
2018-04-04 12:06:26 -06:00
Matan Barak 9b82844197 IB/uverbs: Add action_handle flow steering specification
Binding a flow_action to flow steering rule requires using a new
specification. Therefore, adding such an IB_FLOW_SPEC_ACTION_HANDLE flow
specification.

Flow steering rules could use flow_action(s) and as of that we need to
avoid deleting flow_action(s) as long as they're being used.
Moreover, when the attached rules are deleted, action_handle reference
count should be decremented. Introducing a new mechanism of flow
resources to keep track on the attached action_handle(s). Later on, this
mechanism should be extended to other attached flow steering resources
like flow counters.

Reviewed-by: Yishai Hadas <yishaih@mellanox.com>
Signed-off-by: Matan Barak <matanb@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
2018-04-04 12:06:25 -06:00
Matan Barak 2eb9beaee5 IB/uverbs: Add flow_action create and destroy verbs
A verbs application may receive and transmits packets using a data
path pipeline. Sometimes, the first stage in the receive pipeline or
the last stage in the transmit pipeline involves transforming a
packet, either in order to make it easier for later stages to process
it or to prepare it for transmission over the wire. Such transformation
could be stripping/encapsulating the packet (i.e. vxlan),
decrypting/encrypting it (i.e. ipsec), altering headers, doing some
complex FPGA changes, etc.

Some hardware could do such transformations without software data path
intervention at all. The flow steering API supports steering a
packet (either to a QP or dropping it) and some simple packet
immutable actions (i.e. tagging a packet). Complex actions, that may
change the packet, could bloat the flow steering API extensively.
Sometimes the same action should be applied to several flows.
In this case, it's easier to bind several flows to the same action and
modify it than change all matching flows.

Introducing a new flow_action object that abstracts any packet
transformation (out of a standard and well defined set of actions).
This flow_action object could be tied to a flow steering rule via a
new specification.

Currently, we support esp flow_action, which encrypts or decrypts a
packet according to the given parameters. However, we present a
flexible schema that could be used to other transformation actions tied
to flow rules.

Reviewed-by: Yishai Hadas <yishaih@mellanox.com>
Signed-off-by: Matan Barak <matanb@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
2018-04-04 12:06:25 -06:00
Matan Barak 766d8551ad IB/uverbs: Refactor kern_spec_to_ib_spec_filter
The current implementation of kern_spec_to_ib_spec_filter, which takes
a uAPI based flow steering specification and creates the respective kernel
API flow steering structure, gets a ib_uverbs_flow_spec structure.
The new flow_action uAPI gets a match mask and filter from user-space
which aren't encoded in the flow steering's ib_uverbs_flow_spec structure.
Exporting the logic out of kern_spec_to_ib_spec_filter to get user-space
blobs rather than ib_uverbs_flow_spec structure.

Reviewed-by: Yishai Hadas <yishaih@mellanox.com>
Signed-off-by: Matan Barak <matanb@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
2018-04-04 12:06:24 -06:00
Matan Barak 494c5580aa IB/uverbs: Add enum attribute type to ioctl() interface
Methods sometimes need to get one attribute out of a group of
pre-defined attributes. This is an enum-like behavior. Since
this is a common requirement, we add a new ENUM attribute to the
generic uverbs ioctl() layer. This attribute is embedded in methods,
like any other attributes we currently have. ENUM attributes point to
an array of standard UVERBS_ATTR_PTR_IN. The user-space encodes the
enum's attribute id in the id field and the internal PTR_IN attr id in
the enum_data.elem_id field. This ENUM attribute could be shared by
several attributes and it can get UVERBS_ATTR_SPEC_F_MANDATORY flag,
stating this attribute must be supported by the kernel, like any other
attribute.

Reviewed-by: Yishai Hadas <yishaih@mellanox.com>
Signed-off-by: Matan Barak <matanb@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
2018-04-04 12:06:24 -06:00
Parav Pandit 414448d249 RDMA: Use ib_gid_attr during GID modification
Now that ib_gid_attr contains device, port and index, simplify the
provider APIs add_gid() and del_gid() to use device, port and index
fields from the ib_gid_attr attributes structure.

Signed-off-by: Parav Pandit <parav@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
2018-04-03 21:34:16 -06:00
Parav Pandit 598ff6bae6 IB/core: Refactor GID modify code for RoCE
Code is refactored to prepare separate functions for RoCE which can do more
complex operations related to reference counting, while still
maintainining code readability. This includes
(a) Simplification to not perform netdevice checks and modifications
for IB link layer.
(b) Do not add RoCE GID entry which has NULL netdevice; instead return
an error.
(c) If GID addition fails at provider level add_gid(), do not add the
entry in the cache and keep the entry marked as INVALID.
(d) Simplify and reuse the ib_cache_gid_add()/del() routines so that they
can be used even for modifying default GIDs. This avoid some code
duplication in modifying default GIDs.
(e) find_gid() routine refers to the data entry flags to qualify a GID
as valid or invalid GID rather than depending on attributes and zeroness
of the GID content.
(f) gid_table_reserve_default() sets the GID default attribute at
beginning while setting up the GID table. There is no need to use
default_gid flag in low level functions such as write_gid(), add_gid(),
del_gid(), as they never need to update the DEFAULT property of the GID
entry while during GID table update.

As as result of this refactor, reserved GID 0:0:0:0:0:0:0:0 is no longer
searchable as described below.

A unicast GID entry of 0:0:0:0:0:0:0:0 is Reserved GID as per the IB
spec version 1.3 section 4.1.1, point (6) whose snippet is below.

"The unicast GID address 0:0:0:0:0:0:0:0 is reserved - referred to as
the Reserved GID. It shall never be assigned to any endport. It shall
not be used as a destination address or in a global routing header
(GRH)."

GID table cache now only stores valid GID entries. Before this patch,
Reserved GID 0:0:0:0:0:0:0:0 was searchable in the GID table using
ib_find_cached_gid_by_port() and other similar find routines.

Zero GID is no longer searchable as it shall not to be present in GRH or
path recored entry as described in IB spec version 1.3 section 4.1.1,
point (6), section 12.7.10 and section 12.7.20.

ib_cache_update() is simplified to check link layer once, use unified
locking scheme for all link layers, removed temporary gid table
allocation/free logic.

Additionally,
(a) Expand ib_gid_attr to store port and index so that GID query
routines can get port and index information from the attribute structure.
(b) Expand ib_gid_attr to store device as well so that in future code when
GID reference counting is done, device is used to reach back to the GID
table entry.

Signed-off-by: Parav Pandit <parav@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
2018-04-03 21:33:50 -06:00
Parav Pandit f35faa4ba9 IB/core: Simplify ib_query_gid to always refer to cache
Currently following inconsistencies exist.
1. ib_query_gid() returns GID from the software cache for a RoCE port
and returns GID from the HCA for an IB port.
This is incorrect because software GID cache is maintained regardless
of HCA port type.

2. GID is queries from the HCA via ib_query_gid and updated in the
software cache for IB link layer. Both of them might not be in sync.

ULPs such as SRP initiator, SRP target, IPoIB driver have historically
used ib_query_gid() API to query the GID. However CM used cached version
during CM processing, When software cache was introduced, this
inconsitency remained.

In order to simplify, improve readability and avoid link layer
specific above inconsistencies, this patch brings following changes.

1. ib_query_gid() always refers to the cache layer regardless of link
layer.

2. cache module who reads the GID entry from HCA and builds the cache,
directly invokes the HCA provider verb's query_gid() callback function.

3. ib_query_port() is being called in early stage where GID cache is not
yet build while reading port immutable property. Therefore it needs to
read the default GID from the HCA for IB link layer to publish the
subnet prefix.

Signed-off-by: Parav Pandit <parav@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
2018-04-03 21:33:50 -06:00
Parav Pandit 0e1f9b9244 RDMA/providers: Simplify query_gid callback of RoCE providers
ib_query_gid() fetches the GID from the software cache maintained in
ib_core for RoCE ports.

Therefore, simplify the provider drivers for RoCE to treat query_gid()
callback as never called for RoCE, and only require non-RoCE devices to
implement it.

Signed-off-by: Parav Pandit <parav@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
2018-04-03 21:33:47 -06:00
Roland Dreier 8435168d50 RDMA/ucma: Don't allow setting RDMA_OPTION_IB_PATH without an RDMA device
Check to make sure that ctx->cm_id->device is set before we use it.
Otherwise userspace can trigger a NULL dereference by doing
RDMA_USER_CM_CMD_SET_OPTION on an ID that is not bound to a device.

Cc: <stable@vger.kernel.org>
Reported-by: <syzbot+a67bc93e14682d92fc2f@syzkaller.appspotmail.com>
Signed-off-by: Roland Dreier <roland@purestorage.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
2018-04-03 21:06:14 -06:00
David S. Miller c0b458a946 Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net
Minor conflicts in drivers/net/ethernet/mellanox/mlx5/core/en_rep.c,
we had some overlapping changes:

1) In 'net' MLX5E_PARAMS_LOG_{SQ,RQ}_SIZE -->
   MLX5E_REP_PARAMS_LOG_{SQ,RQ}_SIZE

2) In 'net-next' params->log_rq_size is renamed to be
   params->log_rq_mtu_frames.

3) In 'net-next' params->hard_mtu is added.

Signed-off-by: David S. Miller <davem@davemloft.net>
2018-04-01 19:49:34 -04:00
Parav Pandit 218b9e3eb8 RDMA/cma: Move rdma_cm_state to cma_priv.h
rdma_cm_state enum is internal to rdma_cm kernel module.
It is not required to expose state enums to ULP modules.
So lets keep its scope limited to rdma_cm module in cma_priv.h file.

Signed-off-by: Parav Pandit <parav@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
2018-03-29 13:54:21 -06:00
Parav Pandit fd59015d68 IB/addr: Constify dst_entry pointer
Make dst_entry pointer as const struct dst_entry* to improve code
readablity to make sure that dst structure fields are not modified by
various functions which are using it.

Signed-off-by: Parav Pandit <parav@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
2018-03-29 13:54:20 -06:00
Jason Gunthorpe 6f57c933a4 RDMA: Use u64_to_user_ptr everywhere
This is already used in many places, get the rest of them too, only
to make the code a bit clearer & simpler.

Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
2018-03-29 13:42:29 -06:00
Leon Romanovsky 5b2cc79de8 RDMA/nldev: Provide netdevice name and index
Export the net device name and index to easily find connection
between IB devices and relevant net devices.

We also updated the comment regarding the devices without FW.

Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Reviewed-by: Steve Wise <swise@opengridcomputing.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
2018-03-29 13:32:40 -06:00
Kirill Tkhai f0b07bb151 net: Introduce net_rwsem to protect net_namespace_list
rtnl_lock() is used everywhere, and contention is very high.
When someone wants to iterate over alive net namespaces,
he/she has no a possibility to do that without exclusive lock.
But the exclusive rtnl_lock() in such places is overkill,
and it just increases the contention. Yes, there is already
for_each_net_rcu() in kernel, but it requires rcu_read_lock(),
and this can't be sleepable. Also, sometimes it may be need
really prevent net_namespace_list growth, so for_each_net_rcu()
is not fit there.

This patch introduces new rw_semaphore, which will be used
instead of rtnl_mutex to protect net_namespace_list. It is
sleepable and allows not-exclusive iterations over net
namespaces list. It allows to stop using rtnl_lock()
in several places (what is made in next patches) and makes
less the time, we keep rtnl_mutex. Here we just add new lock,
while the explanation of we can remove rtnl_lock() there are
in next patches.

Fine grained locks generally are better, then one big lock,
so let's do that with net_namespace_list, while the situation
allows that.

Signed-off-by: Kirill Tkhai <ktkhai@virtuozzo.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2018-03-29 13:47:53 -04:00
Steve Wise 2253fc0caa RDMA/CMA: Add rdma_port_space to UAPI
Since the rdma_port_space enum is being passed between user and kernel for
user cm_id setup, we need it in a UAPI header.  So add it to
rdma_user_cm.h.

This also fixes the cm_id restrack changes which pass up the port space
value via the RDMA_NLDEV_ATTR_RES_PS attribute.

Fixes: 00313983cd ("RDMA/nldev: provide detailed CM_ID information")
Signed-off-by: Steve Wise <swise@opengridcomputing.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
2018-03-28 20:50:45 -06:00
Roland Dreier 84652aefb3 RDMA/ucma: Introduce safer rdma_addr_size() variants
There are several places in the ucma ABI where userspace can pass in a
sockaddr but set the address family to AF_IB.  When that happens,
rdma_addr_size() will return a size bigger than sizeof struct sockaddr_in6,
and the ucma kernel code might end up copying past the end of a buffer
not sized for a struct sockaddr_ib.

Fix this by introducing new variants

    int rdma_addr_size_in6(struct sockaddr_in6 *addr);
    int rdma_addr_size_kss(struct __kernel_sockaddr_storage *addr);

that are type-safe for the types used in the ucma ABI and return 0 if the
size computed is bigger than the size of the type passed in.  We can use
these new variants to check what size userspace has passed in before
copying any addresses.

Reported-by: <syzbot+6800425d54ed3ed8135d@syzkaller.appspotmail.com>
Signed-off-by: Roland Dreier <roland@purestorage.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
2018-03-28 16:13:36 -06:00
Parav Pandit 190fb9c4d1 IB/core: Refer to RoCE port property to decide building cache
IB core maintains the GID cache entries for the GID table.
This cache table has to be maintained regardless of HCA's
support of GID table.
For IB and iWarp ports, cache is created by querying the HCA.
For RoCE cache is created based on netdev events.

Therefore just refer to the RoCE port property of the {device, port} to
decide whether to build cache by querying HCA or from netdev events.
There is no need to check if HCA support GID table or not.

ib_cache_update() referred to RoCE attribute before validating
port. Though in all current callers port is valid, it is incorrect
to query RoCE port property before validating the port. Therefore,
rdma_protocol_roce() check is done after rdma_is_port_valid() verifies
that port is valid.

Fixes: 115b68aa6e ("IB/ocrdma: Removed GID add/del null routines")
Reviewed-by: Daniel Jurgens <danielj@mellanox.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>
2018-03-27 16:22:12 -06:00
Parav Pandit 22d24f75a1 IB/core: Search GID only for IB link layer
Even though API is only used by IPoIB driver, its incorrect to refer
RoCE GID table property to search for GID.

Look for only IB link layer to search for the GID.

Fixes: dbb12562f7 ("IB/{core, ipoib}: Simplify ib_find_gid to search only for IB link layer")
Signed-off-by: Parav Pandit <parav@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
2018-03-27 16:22:12 -06:00
Parav Pandit 4ab7cb4bf3 IB/core: Refer to RoCE port property instead of GID table property
ib_find_gid_by_filter() searches GID with filter only for RoCE link
layer regardless of HCA's support for GID table.
Therefore, right way to lookup is compare RoCE port property and not
the GID table property.

Fixes: 99b27e3b5d ("IB/cache: Add ib_find_gid_by_filter cache API")
Signed-off-by: Parav Pandit <parav@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
2018-03-27 16:22:12 -06:00
Parav Pandit 3401857ea3 IB/core: Generate GID change event regardless of RoCE GID table property
Due to following reasons, GID table event is generated regardless of GID
table property.

1. GID table cache is maintained at ib core layer regardless of link layer.
2. GID change event has no relation with IB link layer.
3. GID change event also doesn't depend on whether HCA supports GID table
or not.

Fixes: f3906bd360 ("IB/core: Refactor GID cache's ib_dispatch_event")
Signed-off-by: Parav Pandit <parav@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
2018-03-27 16:22:12 -06:00
Parav Pandit 97c45c2c28 IB/cm: Block processing alternate path handling RoCE Rx cm messages
Due to below reasons, it is better to not support alternate path receive
messages for RoCE in near term.

1. Alternate path for RoCE is not supported at rdmacm layer.
2. It is not supported in uverbs/core layer for RoCE.
3. Alternate path for IPv6 for link local address cannot resolve route
determinstically without a valid incoming interface id whose usecase
make sense only with dual port mode.
4. init_av_from_path while processing LAP messages for IB and RoCE can
lead to adding duplicate entry of AV into the port list, leads to list
corruption.
5. rdma-core userspace a well known userspace implementation has removed
support of libucm which use ucm.ko module, which is the only module that
can trigger alternate path related messages.
6. ucm kernel module is requested to be removed from the IB core in
patch [1].

[1] https://patchwork.kernel.org/patch/10268503/

Signed-off-by: Parav Pandit <parav@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
2018-03-27 16:22:12 -06:00
Mark Bloch e945130b52 IB/core: Protect against concurrent access to hardware stats
Currently access to hardware stats buffer isn't protected, this can
result in multiple writes and reads at the same time to the same
memory location. This can lead to providing an incorrect value to
the user. Add a mutex to protect against it.

Fixes: b40f4757da ("IB/core: Make device counter infrastructure dynamic")
Signed-off-by: Mark Bloch <markb@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
2018-03-27 15:07:21 -06:00
Jason Gunthorpe 611cb92b08 RDMA/ucma: Fix uABI structure layouts for 32/64 compat
The rdma_ucm_event_resp is a different length on 32 and 64 bit compiles.

The kernel requires it to be the expected length or longer so 32 bit
builds running on a 64 bit kernel will not work.

Retain full compat by having all kernels accept a struct with or without
the trailing reserved field.

Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
2018-03-27 14:25:08 -06:00
Leon Romanovsky c8d3bcbfc5 RDMA/ucma: Check that device exists prior to accessing it
Ensure that device exists prior to accessing its properties.

Reported-by: <syzbot+71655d44855ac3e76366@syzkaller.appspotmail.com>
Fixes: 7521663857 ("RDMA/cma: Export rdma cm interface to userspace")
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
2018-03-27 14:10:45 -06:00
Leon Romanovsky 4b658d1bbc RDMA/ucma: Check that device is connected prior to access it
Add missing check that device is connected prior to access it.

[   55.358652] BUG: KASAN: null-ptr-deref in rdma_init_qp_attr+0x4a/0x2c0
[   55.359389] Read of size 8 at addr 00000000000000b0 by task qp/618
[   55.360255]
[   55.360432] CPU: 1 PID: 618 Comm: qp Not tainted 4.16.0-rc1-00071-gcaf61b1b8b88 #91
[   55.361693] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.11.0-0-g63451fca13-prebuilt.qemu-project.org 04/01/2014
[   55.363264] Call Trace:
[   55.363833]  dump_stack+0x5c/0x77
[   55.364215]  kasan_report+0x163/0x380
[   55.364610]  ? rdma_init_qp_attr+0x4a/0x2c0
[   55.365238]  rdma_init_qp_attr+0x4a/0x2c0
[   55.366410]  ucma_init_qp_attr+0x111/0x200
[   55.366846]  ? ucma_notify+0xf0/0xf0
[   55.367405]  ? _get_random_bytes+0xea/0x1b0
[   55.367846]  ? urandom_read+0x2f0/0x2f0
[   55.368436]  ? kmem_cache_alloc_trace+0xd2/0x1e0
[   55.369104]  ? refcount_inc_not_zero+0x9/0x60
[   55.369583]  ? refcount_inc+0x5/0x30
[   55.370155]  ? rdma_create_id+0x215/0x240
[   55.370937]  ? _copy_to_user+0x4f/0x60
[   55.371620]  ? mem_cgroup_commit_charge+0x1f5/0x290
[   55.372127]  ? _copy_from_user+0x5e/0x90
[   55.372720]  ucma_write+0x174/0x1f0
[   55.373090]  ? ucma_close_id+0x40/0x40
[   55.373805]  ? __lru_cache_add+0xa8/0xd0
[   55.374403]  __vfs_write+0xc4/0x350
[   55.374774]  ? kernel_read+0xa0/0xa0
[   55.375173]  ? fsnotify+0x899/0x8f0
[   55.375544]  ? fsnotify_unmount_inodes+0x170/0x170
[   55.376689]  ? __fsnotify_update_child_dentry_flags+0x30/0x30
[   55.377522]  ? handle_mm_fault+0x174/0x320
[   55.378169]  vfs_write+0xf7/0x280
[   55.378864]  SyS_write+0xa1/0x120
[   55.379270]  ? SyS_read+0x120/0x120
[   55.379643]  ? mm_fault_error+0x180/0x180
[   55.380071]  ? task_work_run+0x7d/0xd0
[   55.380910]  ? __task_pid_nr_ns+0x120/0x140
[   55.381366]  ? SyS_read+0x120/0x120
[   55.381739]  do_syscall_64+0xeb/0x250
[   55.382143]  entry_SYSCALL_64_after_hwframe+0x21/0x86
[   55.382841] RIP: 0033:0x7fc2ef803e99
[   55.383227] RSP: 002b:00007fffcc5f3be8 EFLAGS: 00000217 ORIG_RAX: 0000000000000001
[   55.384173] RAX: ffffffffffffffda RBX: 0000000000000000 RCX: 00007fc2ef803e99
[   55.386145] RDX: 0000000000000057 RSI: 0000000020000080 RDI: 0000000000000003
[   55.388418] RBP: 00007fffcc5f3c00 R08: 0000000000000000 R09: 0000000000000000
[   55.390542] R10: 0000000000000000 R11: 0000000000000217 R12: 0000000000400480
[   55.392916] R13: 00007fffcc5f3cf0 R14: 0000000000000000 R15: 0000000000000000
[   55.521088] Code: e5 4d 1e ff 48 89 df 44 0f b6 b3 b8 01 00 00 e8 65 50 1e ff 4c 8b 2b 49
8d bd b0 00 00 00 e8 56 50 1e ff 41 0f b6 c6 48 c1 e0 04 <49> 03 85 b0 00 00 00 48 8d 78 08
48 89 04 24 e8 3a 4f 1e ff 48
[   55.525980] RIP: rdma_init_qp_attr+0x52/0x2c0 RSP: ffff8801e2c2f9d8
[   55.532648] CR2: 00000000000000b0
[   55.534396] ---[ end trace 70cee64090251c0b ]---

Fixes: 7521663857 ("RDMA/cma: Export rdma cm interface to userspace")
Fixes: d541e45500 ("IB/core: Convert ah_attr from OPA to IB when copying to user")
Reported-by: <syzbot+7b62c837c2516f8f38c8@syzkaller.appspotmail.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
2018-03-27 14:10:45 -06:00
Jason Gunthorpe 9137108cc3 RDMA/rdma_cm: Fix use after free race with process_one_req
process_one_req() can race with rdma_addr_cancel():

           CPU0                                 CPU1
           ====                                 ====
 process_one_work()
  debug_work_deactivate(work);
  process_one_req()
                                        rdma_addr_cancel()
	                                  mutex_lock(&lock);
 			    	           set_timeout(&req->work,..);
                                              __queue_work()
				   	       debug_work_activate(work);
	                                  mutex_unlock(&lock);

   mutex_lock(&lock);
[..]
	list_del(&req->list);
   mutex_unlock(&lock);
[..]

   // ODEBUG explodes since the work is still queued.
   kfree(req);

Causing ODEBUG to detect the use after free:

ODEBUG: free active (active state 0) object type: work_struct hint: process_one_req+0x0/0x6c0 include/net/dst.h:165
WARNING: CPU: 0 PID: 79 at lib/debugobjects.c:291 debug_print_object+0x166/0x220 lib/debugobjects.c:288
kvm: emulating exchange as write
Kernel panic - not syncing: panic_on_warn set ...

CPU: 0 PID: 79 Comm: kworker/u4:3 Not tainted 4.16.0-rc6+ #361
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011
Workqueue: ib_addr process_one_req
Call Trace:
 __dump_stack lib/dump_stack.c:17 [inline]
 dump_stack+0x194/0x24d lib/dump_stack.c:53
 panic+0x1e4/0x41c kernel/panic.c:183
 __warn+0x1dc/0x200 kernel/panic.c:547
 report_bug+0x1f4/0x2b0 lib/bug.c:186
 fixup_bug.part.11+0x37/0x80 arch/x86/kernel/traps.c:178
 fixup_bug arch/x86/kernel/traps.c:247 [inline]
 do_error_trap+0x2d7/0x3e0 arch/x86/kernel/traps.c:296
 do_invalid_op+0x1b/0x20 arch/x86/kernel/traps.c:315
 invalid_op+0x1b/0x40 arch/x86/entry/entry_64.S:986
RIP: 0010:debug_print_object+0x166/0x220 lib/debugobjects.c:288
RSP: 0000:ffff8801d966f210 EFLAGS: 00010086
RAX: dffffc0000000008 RBX: 0000000000000003 RCX: ffffffff815acd6e
RDX: 0000000000000000 RSI: 1ffff1003b2cddf2 RDI: 0000000000000000
RBP: ffff8801d966f250 R08: 0000000000000000 R09: 1ffff1003b2cddc8
R10: ffffed003b2cde71 R11: ffffffff86f39a98 R12: 0000000000000001
R13: ffffffff86f15540 R14: ffffffff86408700 R15: ffffffff8147c0a0
 __debug_check_no_obj_freed lib/debugobjects.c:745 [inline]
 debug_check_no_obj_freed+0x662/0xf1f lib/debugobjects.c:774
 kfree+0xc7/0x260 mm/slab.c:3799
 process_one_req+0x2e7/0x6c0 drivers/infiniband/core/addr.c:592
 process_one_work+0xc47/0x1bb0 kernel/workqueue.c:2113
 worker_thread+0x223/0x1990 kernel/workqueue.c:2247
 kthread+0x33c/0x400 kernel/kthread.c:238
 ret_from_fork+0x3a/0x50 arch/x86/entry/entry_64.S:406

Fixes: 5fff41e1f8 ("IB/core: Fix race condition in resolving IP to MAC")
Reported-by: <syzbot+3b4acab09b6463472d0a@syzkaller.appspotmail.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
2018-03-27 13:19:01 -06:00
Kirill Tkhai 2f635ceeb2 net: Drop pernet_operations::async
Synchronous pernet_operations are not allowed anymore.
All are asynchronous. So, drop the structure member.

Signed-off-by: Kirill Tkhai <ktkhai@virtuozzo.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2018-03-27 13:18:09 -04:00
Parav Pandit 114cc9c4b1 IB/cma: Resolve route only while receiving CM requests
Currently CM request for RoCE follows following flow.
rdma_create_id()
rdma_resolve_addr()
rdma_resolve_route()
For RC QPs:
rdma_connect()
->cma_connect_ib()
  ->ib_send_cm_req()
    ->cm_init_av_by_path()
      ->ib_init_ah_attr_from_path()
For UD QPs:
rdma_connect()
->cma_resolve_ib_udp()
  ->ib_send_cm_sidr_req()
    ->cm_init_av_by_path()
      ->ib_init_ah_attr_from_path()

In both the flows, route is already resolved before sending CM requests.
Therefore, code is refactored to avoid resolving route second time in
ib_cm layer.
ib_init_ah_attr_from_path() is extended to resolve route when it is not
yet resolved for RoCE link layer. This is achieved by caller setting
route_resolved field in path record whenever it has route already
resolved.

Signed-off-by: Parav Pandit <parav@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
2018-03-23 10:58:05 -06:00
David S. Miller 03fe2debbb Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net
Fun set of conflict resolutions here...

For the mac80211 stuff, these were fortunately just parallel
adds.  Trivially resolved.

In drivers/net/phy/phy.c we had a bug fix in 'net' that moved the
function phy_disable_interrupts() earlier in the file, whilst in
'net-next' the phy_error() call from this function was removed.

In net/ipv4/xfrm4_policy.c, David Ahern's changes to remove the
'rt_table_id' member of rtable collided with a bug fix in 'net' that
added a new struct member "rt_mtu_locked" which needs to be copied
over here.

The mlxsw driver conflict consisted of net-next separating
the span code and definitions into separate files, whilst
a 'net' bug fix made some changes to that moved code.

The mlx5 infiniband conflict resolution was quite non-trivial,
the RDMA tree's merge commit was used as a guide here, and
here are their notes:

====================

    Due to bug fixes found by the syzkaller bot and taken into the for-rc
    branch after development for the 4.17 merge window had already started
    being taken into the for-next branch, there were fairly non-trivial
    merge issues that would need to be resolved between the for-rc branch
    and the for-next branch.  This merge resolves those conflicts and
    provides a unified base upon which ongoing development for 4.17 can
    be based.

    Conflicts:
            drivers/infiniband/hw/mlx5/main.c - Commit 42cea83f95
            (IB/mlx5: Fix cleanup order on unload) added to for-rc and
            commit b5ca15ad7e (IB/mlx5: Add proper representors support)
            add as part of the devel cycle both needed to modify the
            init/de-init functions used by mlx5.  To support the new
            representors, the new functions added by the cleanup patch
            needed to be made non-static, and the init/de-init list
            added by the representors patch needed to be modified to
            match the init/de-init list changes made by the cleanup
            patch.
    Updates:
            drivers/infiniband/hw/mlx5/mlx5_ib.h - Update function
            prototypes added by representors patch to reflect new function
            names as changed by cleanup patch
            drivers/infiniband/hw/mlx5/ib_rep.c - Update init/de-init
            stage list to match new order from cleanup patch
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
2018-03-23 11:31:58 -04:00
Parav Pandit 98f1f4e0ed IB/core: Refer to RoCE port property instead of GID table property
ib_query_gid() in commit [1] refers to RoCE GID table capability of
the HCA using rdma_cap_roce_gid_table().
ib_core maintains the GID table cache regardless of the HCA provider
drivers capability to maintain RoCE GID table.
Therefore, whether to return a GID table entry from the software cache or
from HCA should be done based on whether the port is RoCE or not.

[1] commit 03db3a2d81 ("IB/core: Add RoCE GID table management")

Reviewed-by: Mark Bloch <markb@mellanox.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>
2018-03-22 12:42:49 -06:00
Leon Romanovsky 03286030ac RDMA/restrack: Remove ambiguity in resource track clean logic
The restrack clean routine had simple, but powerful WARN_ON check
to see if all resources are cleared prior to releasing device.

The WARN_ON check performed very well, but lack of information
which device caused to resource leak, the object type and origin
made debug to be fun and challenging at the same time.

The fact that all dumps were the same because restrack_clean() is
called in dealloc() didn't help either.

So let's fix spelling error and convert WARN_ON to be more debug
friendly. The dmesg cut below gives example of how the output
will look output for the case fixed in patch [1]

[  438.421372] restrack: ------------[ cut here ]------------
[  438.423448] restrack: BUG: RESTRACK detected leak of resources on mlx5_2
[  438.425600] restrack: Kernel PD object allocated by mlx5_ib is not freed
[  438.427753] restrack: Kernel CQ object allocated by mlx5_ib is not freed
[  438.429660] restrack: ------------[ cut here ]------------

[1] https://patchwork.kernel.org/patch/10298695/

Cc: Michal Kalderon <Michal.Kalderon@cavium.com>
Cc: Chuck Lever <chuck.lever@oracle.com>
Reviewed-by: Mark Bloch <markb@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
2018-03-22 12:42:48 -06:00
Chien Tin Tung 5f3e3b85cc RDMA/ucma: Correct option size check using optlen
The option size check is using optval instead of optlen
causing the set option call to fail. Use the correct
field, optlen, for size check.

Fixes: 6a21dfc0d0 ("RDMA/ucma: Limit possible option size")
Signed-off-by: Chien Tin Tung <chien.tin.tung@intel.com>
Signed-off-by: Shiraz Saleem <shiraz.saleem@intel.com>
Reviewed-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
2018-03-21 14:22:22 -06:00
Leon Romanovsky 103140eccb RDMA/restrack: Move restrack_clean to be symmetrical to restrack_init
The fact that resource tracking commit 02d8883f52 ("RDMA/restrack: Add
general infrastructure to track RDMA resources") was added immediately
after commit 16c1975f10 ("IB/mlx5: Create profile infrastructure to add
and remove stages") caused us to miss the fact that PD and CQ are created
after ib_register_device, but released after ib_unregister_device() and
not before as it is expected from normal flow.

Fix introduced in commit 42cea83f95 ("IB/mlx5: Fix cleanup order on
unload") revealed this fact, so this patch is needed to avoid from
restrack warnings

It fixes resource tracking warnings during shutdown.

[   43.473906] CPU: 5 PID: 3016 Comm: modprobe Not tainted 4.16.0-rc5-for-linust-perf-2018-03-19_07-01-58-14 #1
[   43.473907] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS Ubuntu-1.8.2-1ubuntu2 04/01/2014
[   43.473919] RIP: 0010:rdma_restrack_clean+0x25/0x30 [ib_core]
[   43.473921] RSP: 0018:ffffc9000267be48 EFLAGS: 00010282
[   43.473924] RAX: 0000000000000000 RBX: ffff88033c690070 RCX: 0000000180080006
[   43.473925] RDX: ffff88035ce922e0 RSI: ffffea000cf1a200 RDI: ffff88033c6907c8
[   43.473926] RBP: ffff88033c690070 R08: ffff88033c689000 R09: 0000000180080006
[   43.473927] R10: 000000003c68a001 R11: ffff88033c689000 R12: ffff88033c690000
[   43.473929] R13: ffff88033c69005c R14: 0000000000000000 R15: 0000000000000000
[   43.473932] FS:  00007f5928359740(0000) GS:ffff88036c540000(0000) knlGS:0000000000000000
[   43.473933] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[   43.473935] CR2: 00007ffffc760cc8 CR3: 000000035620c000 CR4: 00000000000006e0
[   43.473940] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[   43.473941] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[   43.473942] Call Trace:
[   43.473969]  ib_unregister_device+0xf5/0x190 [ib_core]
[   43.474000]  __mlx5_ib_remove+0x2e/0x40 [mlx5_ib]
[   43.474098]  mlx5_remove_device+0xf5/0x120 [mlx5_core]
[   43.474132]  mlx5_unregister_interface+0x37/0x90 [mlx5_core]
[   43.474142]  mlx5_ib_cleanup+0xc/0x16a [mlx5_ib]
[   43.474152]  SyS_delete_module+0x159/0x260
[   43.474159]  do_syscall_64+0x61/0x110
[   43.474165]  entry_SYSCALL_64_after_hwframe+0x3d/0xa2
[   43.474168] RIP: 0033:0x7f59278466b7
[   43.474170] RSP: 002b:00007ffffc763e38 EFLAGS: 00000202 ORIG_RAX: 00000000000000b0
[   43.474172] RAX: ffffffffffffffda RBX: 000000000130d590 RCX: 00007f59278466b7
[   43.474173] RDX: 0000000000000000 RSI: 0000000000000800 RDI: 000000000130d5f8
[   43.474175] RBP: 0000000000000000 R08: 00007f5927b0b060 R09: 00007f59278b6a40
[   43.474176] R10: 00007ffffc763bc0 R11: 0000000000000202 R12: 0000000000000000
[   43.474177] R13: 0000000000000001 R14: 000000000130d5f8 R15: 0000000000000000
[   43.474179] Code: 84 00 00 00 00 00 0f 1f 44 00 00 48 83 c7 28 31 c0
eb 0c 48 83 c0 08 48 3d 00 08 00 00 74 0f 48 8d 14 07 48 8b 12 48 85 d2
74 e8 <0f> 0b c3 f3 c3 66 0f 1f 44 00 00 0f 1f 44 00 00 53 48 8b 47 28
[   43.474221] ---[ end trace e89771e2250ffc23 ]---

Fixes: 42cea83f95 ("IB/mlx5: Fix cleanup order on unload")
Reviewed-by: Mark Bloch <markb@mellanox.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>
2018-03-21 14:22:22 -06:00
Leon Romanovsky e8980d67d6 RDMA/ucma: Ensure that CM_ID exists prior to access it
Prior to access UCMA commands, the context should be initialized
and connected to CM_ID with ucma_create_id(). In case user skips
this step, he can provide non-valid ctx without CM_ID and cause
to multiple NULL dereferences.

Also there are situations where the create_id can be raced with
other user access, ensure that the context is only shared to
other threads once it is fully initialized to avoid the races.

[  109.088108] BUG: unable to handle kernel NULL pointer dereference at 0000000000000020
[  109.090315] IP: ucma_connect+0x138/0x1d0
[  109.092595] PGD 80000001dc02d067 P4D 80000001dc02d067 PUD 1da9ef067 PMD 0
[  109.095384] Oops: 0000 [#1] SMP KASAN PTI
[  109.097834] CPU: 0 PID: 663 Comm: uclose Tainted: G    B 4.16.0-rc1-00062-g2975d5de6428 #45
[  109.100816] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.11.0-0-g63451fca13-prebuilt.qemu-project.org 04/01/2014
[  109.105943] RIP: 0010:ucma_connect+0x138/0x1d0
[  109.108850] RSP: 0018:ffff8801c8567a80 EFLAGS: 00010246
[  109.111484] RAX: 0000000000000000 RBX: 1ffff100390acf50 RCX: ffffffff9d7812e2
[  109.114496] RDX: 1ffffffff3f507a5 RSI: 0000000000000297 RDI: 0000000000000297
[  109.117490] RBP: ffff8801daa15600 R08: 0000000000000000 R09: ffffed00390aceeb
[  109.120429] R10: 0000000000000001 R11: ffffed00390aceea R12: 0000000000000000
[  109.123318] R13: 0000000000000120 R14: ffff8801de6459c0 R15: 0000000000000118
[  109.126221] FS:  00007fabb68d6700(0000) GS:ffff8801e5c00000(0000) knlGS:0000000000000000
[  109.129468] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[  109.132523] CR2: 0000000000000020 CR3: 00000001d45d8003 CR4: 00000000003606b0
[  109.135573] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[  109.138716] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[  109.142057] Call Trace:
[  109.144160]  ? ucma_listen+0x110/0x110
[  109.146386]  ? wake_up_q+0x59/0x90
[  109.148853]  ? futex_wake+0x10b/0x2a0
[  109.151297]  ? save_stack+0x89/0xb0
[  109.153489]  ? _copy_from_user+0x5e/0x90
[  109.155500]  ucma_write+0x174/0x1f0
[  109.157933]  ? ucma_resolve_route+0xf0/0xf0
[  109.160389]  ? __mod_node_page_state+0x1d/0x80
[  109.162706]  __vfs_write+0xc4/0x350
[  109.164911]  ? kernel_read+0xa0/0xa0
[  109.167121]  ? path_openat+0x1b10/0x1b10
[  109.169355]  ? fsnotify+0x899/0x8f0
[  109.171567]  ? fsnotify_unmount_inodes+0x170/0x170
[  109.174145]  ? __fget+0xa8/0xf0
[  109.177110]  vfs_write+0xf7/0x280
[  109.179532]  SyS_write+0xa1/0x120
[  109.181885]  ? SyS_read+0x120/0x120
[  109.184482]  ? compat_start_thread+0x60/0x60
[  109.187124]  ? SyS_read+0x120/0x120
[  109.189548]  do_syscall_64+0xeb/0x250
[  109.192178]  entry_SYSCALL_64_after_hwframe+0x21/0x86
[  109.194725] RIP: 0033:0x7fabb61ebe99
[  109.197040] RSP: 002b:00007fabb68d5e98 EFLAGS: 00000202 ORIG_RAX: 0000000000000001
[  109.200294] RAX: ffffffffffffffda RBX: 0000000000000000 RCX: 00007fabb61ebe99
[  109.203399] RDX: 0000000000000120 RSI: 00000000200001c0 RDI: 0000000000000004
[  109.206548] RBP: 00007fabb68d5ec0 R08: 0000000000000000 R09: 0000000000000000
[  109.209902] R10: 0000000000000000 R11: 0000000000000202 R12: 00007fabb68d5fc0
[  109.213327] R13: 0000000000000000 R14: 00007fff40ab2430 R15: 00007fabb68d69c0
[  109.216613] Code: 88 44 24 2c 0f b6 84 24 6e 01 00 00 88 44 24 2d 0f
b6 84 24 69 01 00 00 88 44 24 2e 8b 44 24 60 89 44 24 30 e8 da f6 06 ff
31 c0 <66> 41 83 7c 24 20 1b 75 04 8b 44 24 64 48 8d 74 24 20 4c 89 e7
[  109.223602] RIP: ucma_connect+0x138/0x1d0 RSP: ffff8801c8567a80
[  109.226256] CR2: 0000000000000020

Fixes: 7521663857 ("RDMA/cma: Export rdma cm interface to userspace")
Reported-by: <syzbot+36712f50b0552615bf59@syzkaller.appspotmail.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
2018-03-20 11:07:21 -06:00
Matan Barak 185899ee8d IB/uverbs: Enable ioctl() uAPI by default for new verbs
Enable the ioctl() uAPI for IB by default if the standard write()
uAPI (INFINIBAND_USER_ACCESS) is enabled. Verbs that are
also available under the old write() uAPI are put inside a new
INFINIBAND_EXP_LEGACY_VERBS_NEW_UAPI Kconfig.

Reviewed-by: Yishai Hadas <yishaih@mellanox.com>
Signed-off-by: Matan Barak <matanb@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
2018-03-19 14:45:17 -06:00
Matan Barak 41b2a71fc8 IB/uverbs: Move ioctl path of create_cq and destroy_cq to a new file
Currently, all objects are declared in uverbs_std_types. This could lead
to a huge file once we implement all objects, methods and handlers.
Moving each object to its own file to keep the files smaller and more
readable. uverbs_std_types.c will only contain the parsing tree
definition and objects without any methods.

Signed-off-by: Matan Barak <matanb@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
2018-03-19 14:45:17 -06:00
Matan Barak dfb1395573 IB/uverbs: Expose parsing tree of all common objects to providers
The ioctl() based uverbs is based on merging feature trees. This teaches
the generic parser how to parse methods according to the provider's
support. In order to support merging with the common objects, exporting
the common-object-tree to the provider drivers.

Reviewed-by: Yishai Hadas <yishaih@mellanox.com>
Signed-off-by: Matan Barak <matanb@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
2018-03-19 14:45:17 -06:00
Matan Barak c66db31113 IB/uverbs: Safely extend existing attributes
Previously, we've used UVERBS_ATTR_SPEC_F_MIN_SZ for extending existing
attributes. The behavior of this flag was the kernel accepts anything
bigger than the minimum size it specified. This is unsafe, since in
order to safely extend an attribute, we need to make sure unknown size
is zeroed. Replacing UVERBS_ATTR_SPEC_F_MIN_SZ with
UVERBS_ATTR_SPEC_F_MIN_SZ_OR_ZERO, which essentially checks that the
unknown size is zero. In addition, attributes are now decorated with
UVERBS_ATTR_TYPE and UVERBS_ATTR_STRUCT, so we can provide the minimum
and known length.

Users of this flag needs to use copy_from_or_zero functions/macros.

Reviewed-by: Yishai Hadas <yishaih@mellanox.com>
Signed-off-by: Matan Barak <matanb@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
2018-03-19 14:45:17 -06:00
Matan Barak 1f07e08fab IB/uverbs: Enable compact representation of uverbs_attr_spec
Downstream patches extend uverbs_attr_spec with new fields.
In order to save space, we move the type and flags fields to
the various attribute flavors contained in the union.

Reviewed-by: Yishai Hadas <yishaih@mellanox.com>
Signed-off-by: Matan Barak <matanb@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
2018-03-19 14:45:17 -06:00
Matan Barak 0ede73bc01 IB/uverbs: Extend uverbs_ioctl header with driver_id
Extending uverbs_ioctl header with driver_id and another reserved
field. driver_id should be used in order to identify the driver.
Since every driver could have its own parsing tree, this is necessary
for strace support.
Downstream patches take off the EXPERIMENTAL flag from the ioctl() IB
support and thus we add some reserved fields for future usage.

Reviewed-by: Yishai Hadas <yishaih@mellanox.com>
Signed-off-by: Matan Barak <matanb@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
2018-03-19 14:45:17 -06:00
Matan Barak 1f7ff9d5d3 IB/uverbs: Move to new headers and make naming consistent
Use macros to make names consistent in ioctl() uAPI:
The ioctl() uAPI works with object-method hierarchy. The method part
also states which handler should be executed when this method is called
from user-space. Therefore, we need to tie method, method's id, method's
handler and the object owning this method together.
Previously, this was done through explicit developer chosen names.
This makes grepping the code harder. Changing the method's name,
method's handler and object's name to be automatically generated based
on the ids.

The headers are split in a way so they be included and used by
user-space. One header strictly contains structures that are used
directly by user-space applications, where another header is used for
internal library (i.e. libibverbs) to form the ioctl() commands.
Other header simply contains the required general command structure.

Reviewed-by: Yishai Hadas <yishaih@mellanox.com>
Signed-off-by: Matan Barak <matanb@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
2018-03-19 14:45:17 -06:00
Leon Romanovsky ed65a4dc22 RDMA/ucma: Fix use-after-free access in ucma_close
The error in ucma_create_id() left ctx in the list of contexts belong
to ucma file descriptor. The attempt to close this file descriptor causes
to use-after-free accesses while iterating over such list.

Fixes: 7521663857 ("RDMA/cma: Export rdma cm interface to userspace")
Reported-by: <syzbot+dcfd344365a56fbebd0f@syzkaller.appspotmail.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Reviewed-by: Sean Hefty <sean.hefty@intel.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
2018-03-19 14:01:35 -06:00
Parav Pandit 6d5b2047fe IB/core: Use rdma_is_port_valid()
Use rdma_is_port_valid() which performs port validity check instead of
open coding the same check.

Signed-off-by: Parav Pandit <parav@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Reviewed-by: Yuval Shaia <yuval.shaia@oracle.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
2018-03-19 11:41:40 -06:00