Yang Li [Thu, 29 Apr 2021 10:42:20 +0000 (18:42 +0800)]
IB/qib: Remove redundant assignment to ret
Variable 'ret' is set to zero but this value is never read as it is
overwritten with a new value later on, hence it is a redundant assignment
and can be removed.
Clean up the following clang-analyzer warning:
drivers/infiniband/hw/qib/qib_sd7220.c:690:2: warning: Value stored to
'ret' is never read [clang-analyzer-deadcode.DeadStores]
Gal Pressman [Sun, 18 Apr 2021 12:10:25 +0000 (15:10 +0300)]
RDMA/nldev: Add copy-on-fork attribute to get sys command
The new attribute indicates that the kernel copies DMA pages on fork,
hence libibverbs' fork support through madvise and MADV_DONTFORK is not
needed.
The introduced attribute is always reported as supported since the kernel
has the patch that added the copy-on-fork behavior. This allows the
userspace library to identify older vs newer kernel versions. Extra care
should be taken when backporting this patch as it relies on the fact that
the copy-on-fork patch is merged, hence no check for support is added.
Don't backport this patch unless you also have the following series:
commit 70e806e4e645 ("mm: Do early cow for pinned pages during fork() for
ptes") and commit 4eae4efa2c29 ("hugetlb: do early cow when page pinned on
src mm").
RDMA/bnxt_re: Fix a double free in bnxt_qplib_alloc_res
In bnxt_qplib_alloc_res, it calls bnxt_qplib_alloc_dpi_tbl(). Inside
bnxt_qplib_alloc_dpi_tbl, dpit->dbr_bar_reg_iomem is freed via
pci_iounmap() in unmap_io error branch. After the callee returns err code,
bnxt_qplib_alloc_res calls
bnxt_qplib_free_res()->bnxt_qplib_free_dpi_tbl() in the fail branch. Then
dpit->dbr_bar_reg_iomem is freed in the second time by pci_iounmap().
My patch set dpit->dbr_bar_reg_iomem to NULL after it is freed by
pci_iounmap() in the first time, to avoid the double free.
In siw_alloc_mr(), it calls siw_mr_add_mem(mr,..). In the implementation of
siw_mr_add_mem(), mem is assigned to mr->mem and then mem is freed via
kfree(mem) if xa_alloc_cyclic() failed. Here, mr->mem still point to a
freed object. After, the execution continue up to the err_out branch of
siw_alloc_mr, and the freed mr->mem is used in siw_mr_drop_mem(mr).
My patch moves "mr->mem = mem" behind the if (xa_alloc_cyclic(..)<0) {}
section, to avoid the uaf.
Add QP numbers that are associated with the SRQ to the SRQ information.
The QPs are displayed in a range form.
Sample output:
$ rdma res show srq
dev ibp8s0f0 srqn 0 type BASIC pdn 3 comm [ib_ipoib]
dev ibp8s0f0 srqn 4 type BASIC lqpn 125-128,130-140 pdn 9 pid 3581 comm ibv_srq_pingpon
dev ibp8s0f0 srqn 5 type BASIC lqpn 141-156 pdn 10 pid 3584 comm ibv_srq_pingpon
dev ibp8s0f0 srqn 6 type BASIC lqpn 157-172 pdn 11 pid 3590 comm ibv_srq_pingpon
dev ibp8s0f1 srqn 0 type BASIC pdn 3 comm [ib_ipoib]
dev ibp8s0f1 srqn 1 type BASIC lqpn 329-344 pdn 4 pid 3586 comm ibv_srq_pingpon
$ rdma res show srq lqpn 126-141
dev ibp8s0f0 srqn 4 type BASIC lqpn 126-128,130-140 pdn 9 pid 3581 comm ibv_srq_pingpon
dev ibp8s0f0 srqn 5 type BASIC lqpn 141 pdn 10 pid 3584 comm ibv_srq_pingpon
$ rdma res show srq lqpn 127
dev ibp8s0f0 srqn 4 type BASIC lqpn 127 pdn 9 pid 3581 comm ibv_srq_pingpon
Extend the RDMA nldev return a SRQ information, like SRQ number, SRQ type,
PD number, CQ number and process ID that created that SRQ.
Sample output:
$ rdma res show srq
dev ibp8s0f0 srqn 0 type BASIC pdn 3 comm [ib_ipoib]
dev ibp8s0f0 srqn 4 type BASIC pdn 9 pid 3581 comm ibv_srq_pingpon
dev ibp8s0f0 srqn 5 type BASIC pdn 10 pid 3584 comm ibv_srq_pingpon
dev ibp8s0f0 srqn 6 type BASIC pdn 11 pid 3590 comm ibv_srq_pingpon
dev ibp8s0f1 srqn 0 type BASIC pdn 3 comm [ib_ipoib]
dev ibp8s0f1 srqn 1 type BASIC pdn 4 pid 3586 comm ibv_srq_pingpon
Extend the RDMA nldev return a context information, like ctx number and
process ID that created that context. This functionality is helpful to
find orphan contexts that are not closed for some reason.
Sample output:
$ rdma res show ctx
dev ibp8s0f0 ctxn 0 pid 980 comm ibv_rc_pingpong
dev ibp8s0f0 ctxn 1 pid 981 comm ibv_rc_pingpong
dev ibp8s0f0 ctxn 2 pid 992 comm ibv_rc_pingpong
dev ibp8s0f1 ctxn 0 pid 984 comm ibv_rc_pingpong
dev ibp8s0f1 ctxn 1 pid 987 comm ibv_rc_pingpong
$ rdma res show ctx dev ibp8s0f1
dev ibp8s0f1 ctxn 0 pid 984 comm ibv_rc_pingpong
dev ibp8s0f1 ctxn 1 pid 987 comm ibv_rc_pingpong
RDMA/core: Add CM to restrack after successful attachment to a device
The device attach triggers addition of CM_ID to the restrack DB.
However, when error occurs, we releasing this device, but defer CM_ID
release. This causes to the situation where restrack sees CM_ID that
is not valid anymore.
As a solution, add the CM_ID to the resource tracking DB only after the
attachment is finished.
Found by syzcaller:
infiniband syz0: added syz_tun
rdma_rxe: ignoring netdev event = 10 for syz_tun
infiniband syz0: set down
infiniband syz0: ib_query_port failed (-19)
restrack: ------------[ cut here ]------------
infiniband syz0: BUG: RESTRACK detected leak of resources
restrack: User CM_ID object allocated by syz-executor716 is not freed
restrack: ------------[ cut here ]------------
A switchdev RDMA device do not support IB CM. When such device is added to
the RDMA CM's device list, when application invokes rdma_listen(), cma
attempts to listen to such device, however it has IB CM attribute
disabled.
Due to this, rdma_listen() call fails to listen for other non switchdev
devices as well.
Bob Pearson [Wed, 21 Apr 2021 03:59:53 +0000 (22:59 -0500)]
RDMA/rxe: Fix a bug in rxe_fill_ip_info()
Fix a bug in rxe_fill_ip_info() which was attempting to convert from
RDMA_NETWORK_XXX to RXE_NETWORK_XXX. .._IPV6 should have mapped to .._IPV6
not .._IPV4.
Mark Bloch [Sun, 18 Apr 2021 10:59:21 +0000 (13:59 +0300)]
RDMA/mlx5: Expose private query port
Expose a non standard query port via IOCTL that will be used to expose
port attributes that are specific to mlx5 devices.
The new interface receives a port number to query and returns a structure
that contains the available attributes for that port. This will be used
to fill the gap between pure DEVX use cases and use cases where a kernel
needs to inform userspace about various kernel driver configurations that
userspace must use in order to work correctly.
Flags is used to indicate which fields are valid on return.
MLX5_IB_UAPI_QUERY_PORT_VPORT:
The vport number of the queered port.
MLX5_IB_UAPI_QUERY_PORT_VPORT_VHCA_ID:
The VHCA ID of the vport of the queered port.
MLX5_IB_UAPI_QUERY_PORT_VPORT_STEERING_ICM_RX:
The vport's RX ICM address used for sw steering.
MLX5_IB_UAPI_QUERY_PORT_VPORT_STEERING_ICM_TX:
The vport's TX ICM address used for sw steering.
MLX5_IB_UAPI_QUERY_PORT_VPORT_REG_C0:
The metadata used to tag egress packets of the vport.
MLX5_IB_UAPI_QUERY_PORT_ESW_OWNER_VHCA_ID:
The E-Switch owner vhca id of the vport.
missing qpid increment leads to skipping few qpids while allocating QP.
This eventually leads to adapter running out of qpids after establishing
fewer connections than it actually supports.
Current patch increments the qpid correctly.
Leon Romanovsky [Thu, 1 Apr 2021 06:57:13 +0000 (09:57 +0300)]
RDMA/bnxt_re: Get rid of custom module reference counting
Instead of manually messing with parent driver module reference counting
rely on export symbol mechanism to ensure that proper probe/remove chain
is performed.
Leon Romanovsky [Thu, 1 Apr 2021 06:57:12 +0000 (09:57 +0300)]
RDMA/bnxt_re: Create direct symbol link between bnxt modules
Convert indirect probe call to its direct equivalent to create a symbol
link between RDMA and netdev modules. This will give us an ability to
remove custom module reference counting that doesn't belong to the driver.
Leon Romanovsky [Thu, 1 Apr 2021 06:57:11 +0000 (09:57 +0300)]
RDMA/bnxt_re: Depend on bnxt ethernet driver and not blindly select it
The "select" kconfig keyword provides reverse dependency, however it
doesn't check that selected symbol meets its own dependencies. Usually
"select" is used for non-visible symbols, so instead of trying to keep
dependencies in sync with BNXT ethernet driver, simply "depends on" it,
like Kconfig documentation suggest.
* CONFIG_PCI is already required by BNXT
* CONFIG_NETDEVICES and CONFIG_ETHERNET are needed to chose BNXT
IB/ipoib: Improve latency in ipoib/cm connection formation
Currently IPoIB connected mode queries the device to get the pkey table
entry during connection formation. This will increase the time taken to
form the connection, especially when limited pkeys are in use. This gets
worse when multiple connection attempts are done in parallel.
Since ipoib interfaces are locked to a single pkey, use the pkey index
that was determined at link up time instead of searching for anything.
This improved the latency from 500ms to 1ms on an internal setup.
Introduce the VF support by adding code changes to allow VF PCI device
initialization, assgining the reserved resource of the PF to the active
VFs, setting the default abilities, applying the interruptions, resetting
and reducing the default QP/GID number to aovid exceeding the hardware
limitation.
RDMA/hns: Query the number of functions supported by the PF
Query how many functions are supported by the PF from the FW and store it
in the hns_roce_dev structure which will be used to support the
configuration of virtual functions.
Xi Wang [Tue, 6 Apr 2021 13:25:09 +0000 (21:25 +0800)]
RDMA/hns: Simplify function's resource related command
Use hr_reg_write/read() to simplify codes about configuring function's
resource. And because the design of PF/VF fields is same, they can be
defined only once.
Gioh Kim [Wed, 7 Apr 2021 11:34:41 +0000 (13:34 +0200)]
RDMA/rtrs-clt: Add a minimum latency multipath policy
This patch adds new multipath policy: min-latency. Client checks the
latency of each path when it sends the heart-beat. And it sends IO to the
path with the minimum latency.
Jason Gunthorpe [Tue, 13 Apr 2021 22:37:17 +0000 (19:37 -0300)]
Merge branch 'mlx5_memic_ops' of git://git.kernel.org/pub/scm/linux/kernel/git/mellanox/linux
Maor Gottlieb says:
====================
This series from Maor extends MEMIC to support atomic operations from the
host in addition to already supported regular read/write.
====================
* 'memic_ops':
RDMA/mlx5: Expose UAPI to query DM
RDMA/mlx5: Add support in MEMIC operations
RDMA/mlx5: Add support to MODIFY_MEMIC command
RDMA/mlx5: Re-organize the DM code
RDMA/mlx5: Move all DM logic to separate file
RDMA/uverbs: Make UVERBS_OBJECT_METHODS to consider line number
net/mlx5: Add MEMIC operations related bits
Maor Gottlieb [Sun, 11 Apr 2021 12:29:24 +0000 (15:29 +0300)]
RDMA/mlx5: Expose UAPI to query DM
Expose UAPI to query MEMIC DM, this will let user space application
that didn't allocate the DM but has access to by owning the matching
command FD to retrieve its information.
Maor Gottlieb [Sun, 11 Apr 2021 12:29:23 +0000 (15:29 +0300)]
RDMA/mlx5: Add support in MEMIC operations
MEMIC buffer, in addition to regular read and write operations, can
support atomic operations from the host.
Introduce and implement new UAPI to allocate address space for MEMIC
operations such as atomic. This includes:
1. Expose new IOCTL for request mapping of MEMIC operation.
2. Hold the operations address in a list, so same operation to same DM
will be allocated only once.
3. Manage refcount on the mlx5_ib_dm object, so it would be keep valid
until all addresses were unmapped.
Maor Gottlieb [Sun, 11 Apr 2021 12:29:21 +0000 (15:29 +0300)]
RDMA/mlx5: Re-organize the DM code
1. Inline the checks from check_dm_type_support() into their
respective allocation functions.
2. Fix use after free when driver fails to copy the MEMIC address to the
user by moving the allocation code into their respective functions,
hence we avoid the explicit call to free the DM in the error flow.
3. Split mlx5_ib_dm struct to memic and icm proper typesafety
throughout.
The AIP and VNIC implementation turns that model inside out and used a
kfree() to free what appears to be a netdev struct when in reality, it is
a struct that enbodies the rx state as well as the dummy netdev used to
support napi_poll across disparate receive contexts. The relationship is
infered by the odd allocation:
Correct the issue by:
- Correctly naming the alloc and free functions
- Renaming hfi1_netdev_priv to hfi1_netdev_rx
- Replacing dd dummy_netdev with a netdev_rx pointer
- Embedding the net_device in hfi1_netdev_rx
- Moving the init_dummy_netdev to the alloc routine
- Adjusting wrappers to fit the new model
Gioh Kim [Mon, 12 Apr 2021 08:40:02 +0000 (10:40 +0200)]
RDMA/rtrs-clt: destroy sysfs after removing session from active list
A session can be removed dynamically by sysfs interface "remove_path" that
eventually calls rtrs_clt_remove_path_from_sysfs function. The current
rtrs_clt_remove_path_from_sysfs first removes the sysfs interfaces and
frees sess->stats object. Second it removes the session from the active
list.
Therefore some functions could access non-connected session and access the
freed sess->stats object even-if they check the session status before
accessing the session.
For instance rtrs_clt_request and get_next_path_min_inflight check the
session status and try to send IO to the session. The session status
could be changed when they are trying to send IO but they could not catch
the change and update the statistics information in sess->stats object,
and generate use-after-free problem.
(see: "RDMA/rtrs-clt: Check state of the rtrs_clt_sess before reading its
stats")
This patch changes the rtrs_clt_remove_path_from_sysfs to remove the
session from the active session list and then destroy the sysfs
interfaces.
Each function still should check the session status because closing or
error recovery paths can change the status.
Håkon Bugge [Wed, 31 Mar 2021 18:43:14 +0000 (20:43 +0200)]
rds: ib: Remove two ib_modify_qp() calls
For some HCAs, ib_modify_qp() is an expensive operation running
virtualized.
For both the active and passive side, the QP returned by the CM has the
state set to RTS, so no need for this excess RTS -> RTS transition. With
IB Core's ability to set the RNR Retry timer, we use this interface to
shave off another ib_modify_qp().
Håkon Bugge [Wed, 31 Mar 2021 18:43:13 +0000 (20:43 +0200)]
IB/cma: Introduce rdma_set_min_rnr_timer()
Introduce the ability for kernel ULPs to adjust the minimum RNR Retry
timer. The INIT -> RTR transition executed by RDMA CM will be used for
this adjustment. This avoids an additional ib_modify_qp() call.
rdma_set_min_rnr_timer() must be called before the call to rdma_connect()
on the active side and before the call to rdma_accept() on the passive
side.
The default value of RNR Retry timer is zero, which translates to 655
ms. When the receiver is not ready to accept a send messages, it encodes
the RNR Retry timer value in the NAK. The requestor will then wait at
least the specified time value before retrying the send.
The 5-bit value to be supplied to the rdma_set_min_rnr_timer() is
documented in IBTA Table 45: "Encoding for RNR NAK Timer Field".
- Add the necessary braces to maintain context alignment.
- Fix the open '{' that is not on the same line as "switch".
- Remove braces that are not necessary for single statement blocks.
- Fix "else" that doesn't follow close brace '}'.
Jason Gunthorpe [Mon, 12 Apr 2021 16:49:48 +0000 (13:49 -0300)]
Merge branch 'mlx5-next' of git://git.kernel.org/pub/scm/linux/kernel/git/mellanox/linux
Saeed Mahameed says:
====================
This pr contains changes from mlx5-next branch,
already reviewed on netdev and rdma mailing lists, links below.
1) From Leon, Dynamically assign MSI-X vectors count
Already Acked by Bjorn Helgaas.
https://patchwork.kernel.org/project/netdevbpf/cover/20210314124256[email protected]/
From Mark, E-Switch cleanups and refactoring, and the addition
of single FDB mode needed HW bits.
From Mikhael, Remove unused struct field
From Saeed, Cleanup W=1 prototype warning
From Zheng, Esw related cleanup
From Tariq, User order-0 page allocation for EQs
====================
* mlx5-next:
net/mlx5: Implement sriov_get_vf_total_msix/count() callbacks
net/mlx5: Dynamically assign MSI-X vectors count
net/mlx5: Add dynamic MSI-X capabilities bits
PCI/IOV: Add sysfs MSI-X vector assignment interface
net/mlx5: Use order-0 allocations for EQs
net/mlx5: Add IFC bits needed for single FDB mode
net/mlx5: E-Switch, Refactor send to vport to be more generic
RDMA/mlx5: Use representor E-Switch when getting netdev and metadata
net/mlx5: E-Switch, Add eswitch pointer to each representor
net/mlx5: E-Switch, Add match on vhca id to default send rules
net/mlx5: Remove unused mlx5_core_health member recover_work
net/mlx5: simplify the return expression of mlx5_esw_offloads_pair()
net/mlx5: Cleanup prototype warning
Bob Pearson [Fri, 2 Apr 2021 00:10:17 +0000 (19:10 -0500)]
RDMA/rxe: Fix missing acks from responder
All responder errors from request packets that do not consume a receive
WQE fail to generate acks for RC QPs. This patch corrects this behavior
by making the flow follow the same path as request packets that do consume
a WQE after the completion.
Yixian Liu [Tue, 6 Apr 2021 08:46:12 +0000 (16:46 +0800)]
RDMA/core: Make the wc status prompt message clearer
Local invalidate is also a kind of memory management operation, not only
memory bind operation. Furthermore, as invalidate operations include local
and remote, add prefix to the prompt message to make it clearer.
IB/mlx5: Reduce max order of memory allocated for xlt update
To update xlt (during mlx5_ib_reg_user_mr()), the driver can request up to
1 MB (order-8) memory, depending on the size of the MR. This costly
allocation can sometimes take very long to return (a few seconds). This
causes the calling application to hang for a long time, especially when
the system is fragmented. To avoid these long latency spikes, the calls
the higher order allocations need to fail faster in case they are not
available.
In order to acheive this we need __GFP_NORETRY flag in the gfp_mask before
during fetching the free pages. Allow the algorithm to automatically fall
back to smaller page sizes.
Given that there in no init for the mmu_notifier, a kzalloc() should
be used to insure that any newly added fields are given a predictable
initial value of zero.
Mike Marciniszyn [Mon, 29 Mar 2021 13:54:08 +0000 (09:54 -0400)]
IB/{ipoib,hfi1}: Add a timeout handler for rdma_netdev
The current rdma_netdev handling in ipoib hooks the tx_timeout handler,
but prints out a totally useless message that prevents effective debugging
especially when multiple transmit queues are being used.
Add a tx_timeout rdma_netdev hook and implement the callback in the hfi1
to print additional information.
The existing non-helpful message is avoided when the driver has presented
a callback.
Leon Romanovsky [Sun, 14 Mar 2021 12:42:55 +0000 (14:42 +0200)]
net/mlx5: Dynamically assign MSI-X vectors count
The number of MSI-X vectors is a PCI property visible through lspci. The
field is read-only and configured by the device. The mlx5 devices work in
a static or dynamic assignment mode.
Static assignment means that all newly created VFs have a preset number of
MSI-X vectors determined by device configuration parameters. This can
result in some VFs having too many or too few MSI-X vectors. Till now this
has been the only means of fine-tuning the MSI-X vector count and it was
acceptable for small numbers of VFs.
With dynamic assignment the inefficiency of having a fixed number of MSI-X
vectors can be avoided with each VF having exactly the required
vectors. Userspace will provide this information while provisioning the VF
for use, based on the intended use. For instance if being used with a VM,
the MSI-X vector count might be matched to the CPU count of the VM.
For compatibility mlx5 continues to start up with MSI-X vector assignment,
but the kernel can now access a larger dynamic vector pool and assign more
vectors to created VFs.
A typical cloud provider SR-IOV use case is to create many VFs for use by
guest VMs. The VFs may not be assigned to a VM until a customer requests a
VM of a certain size, e.g., number of CPUs. A VF may need MSI-X vectors
proportional to the number of CPUs in the VM, but there is no standard way
to change the number of MSI-X vectors supported by a VF.
Some Mellanox ConnectX devices support dynamic assignment of MSI-X vectors
to SR-IOV VFs. This can be done by the PF driver after VFs are enabled,
and it can be done without affecting VFs that are already in use. The
hardware supports a limited pool of MSI-X vectors that can be assigned to
the PF or to individual VFs. This is device-specific behavior that
requires support in the PF driver.
Add a read-only "sriov_vf_total_msix" sysfs file for the PF and a writable
"sriov_vf_msix_count" file for each VF. Management software may use these
to learn how many MSI-X vectors are available and to dynamically assign
them to VFs before the VFs are passed through to a VM.
If the PF driver implements the ->sriov_get_vf_total_msix() callback,
"sriov_vf_total_msix" contains the total number of MSI-X vectors available
for distribution among VFs.
If no driver is bound to the VF, writing "N" to "sriov_vf_msix_count" uses
the PF driver ->sriov_set_msix_vec_count() callback to assign "N" MSI-X
vectors to the VF. When a VF driver subsequently reads the MSI-X Message
Control register, it will see the new Table Size "N".
Yixian Liu [Sat, 27 Mar 2021 10:25:38 +0000 (18:25 +0800)]
RDMA/hns: Reorganize doorbell update interfaces for all queues
The doorbell update interfaces are very similar for different queues, such
as SQ, RQ, SRQ, CQ and EQ. So reorganize these code and also fix some
inappropriate naming.
Yixian Liu [Sat, 27 Mar 2021 10:25:37 +0000 (18:25 +0800)]
RDMA/hns: Support configuring doorbell mode of RQ and CQ
HIP08 supports both normal and record doorbell mode for RQ and CQ, SQ
record doorbell for userspace is also supported by the software for
flushing CQE process. As now the capability of HIP08 are exposed to the
user and are configurable, the support of normal doorbell should be added
back.
Note that, if switching to normal doorbell, the kernel will report "flush
cqe is unsupported" if modify qp to error status as the flush is based on
record doorbell.
Xi Wang [Sat, 27 Mar 2021 03:21:32 +0000 (11:21 +0800)]
RDMA/hns: Refactor reset state checking flow
The 'HNS_ROCE_OPC_QUERY_MB_ST' command will response the mailbox complete
status and hardware busy flag, and the complete status is only valid when
the busy flag is 0, so it's better to query these two fields at a time
rather than separately.
Weihang Li [Mon, 29 Mar 2021 08:46:24 +0000 (16:46 +0800)]
MAINTAINERS: remove Xavier as maintainer of HISILICON ROCE DRIVER
Wei Hu(Xavier) has left Hisilicon and his email address is invalid now.
I'd be glad to add him back with another address if he wants to continue
maintain this module.
projects / linux.git / log