aboutsummaryrefslogtreecommitdiff
path: root/drivers/infiniband
AgeCommit message (Collapse)AuthorFilesLines
2018-09-12RDMA/core: Rename rdma_copy_addr to rdma_copy_src_l2_addrParav Pandit3-11/+19
Now that rdma_copy_addr() only copies the source addresses and all callers are interested in copying only source addresses, simplify it to drop the destination address argument. Given that it only copies source layer2 addresses, rename it to rdma_copy_src_l2_addr for better code readability. Signed-off-by: Parav Pandit <[email protected]> Reviewed-by: Daniel Jurgens <[email protected]> Signed-off-by: Leon Romanovsky <[email protected]> Signed-off-by: Jason Gunthorpe <[email protected]>
2018-09-12RDMA/core: Introduce and use rdma_set_src_addr() between IPv4 and IPv6Parav Pandit1-42/+23
rdma_translate_ip() is done while resolving address for the loopback addresses. The current flow is convoluted with resolve neighbor being optional. This patch simplifies the code in following ways. (a) Use common code between IPv4 and IPv6 for address translation, loopback checks and acquiring netdevice. (b) During neigh resolve in addr_resolve_neigh(), only copy destination address. (c) Always resolve the source address before the destination address, because it doesn't depend on resolving neigh being requested or not. This helps to reduce 3 calls of rdma_copy_addr and rdma_translate_ip to one and makes it easier to follow the code flow. Now that ib_nl_fetch_ha() doesn't depend on dst, drop dst argument from ib_nl_fetch_ha(). Signed-off-by: Parav Pandit <[email protected]> Reviewed-by: Daniel Jurgens <[email protected]> Signed-off-by: Leon Romanovsky <[email protected]> Signed-off-by: Jason Gunthorpe <[email protected]>
2018-09-12RDMA/core: Let protocol specific function typecast sockaddr structureParav Pandit1-16/+15
Current code typecasts destination address using extra variable but uses source address as is. Even though the compiler optimizes such code well, just let each protocol specific function typecast for src and dest both and have symmetric code. Signed-off-by: Parav Pandit <[email protected]> Reviewed-by: Daniel Jurgens <[email protected]> Signed-off-by: Leon Romanovsky <[email protected]> Signed-off-by: Jason Gunthorpe <[email protected]>
2018-09-12RDMA/core: Avoid unnecessary sa_family overwriteParav Pandit1-4/+1
addr4_resolve() and addr6_resolve() are called by checking the value of sa_family. Both above functions overwrite the value after typecasting, this is not necessary. Signed-off-by: Parav Pandit <[email protected]> Reviewed-by: Daniel Jurgens <[email protected]> Signed-off-by: Leon Romanovsky <[email protected]> Signed-off-by: Jason Gunthorpe <[email protected]>
2018-09-12RDMA/core Introduce and use rdma_find_ndev_for_src_ip_rcuParav Pandit1-26/+35
This fixes two issues: 1. When address family is other than IPv4 or v6, rdma_translate_ip() returns success which is incorrect. 2. When address familty is AF_INET6, and if the source address is not found, it returns success, which is also incorrect. Therefore, introduce and use rdma_find_ndev_for_src_ip_rcu() helper function which returns correct success or error status and is also useful for future code refactor in addr_resolve(). Signed-off-by: Parav Pandit <[email protected]> Reviewed-by: Daniel Jurgens <[email protected]> Signed-off-by: Leon Romanovsky <[email protected]> Signed-off-by: Jason Gunthorpe <[email protected]>
2018-09-12RDMA/uverbs: Atomically flush and mark closed the comp event queueSteve Wise1-0/+1
Currently a uverbs completion event queue is flushed of events in ib_uverbs_comp_event_close() with the queue spinlock held and then released. Yet setting ev_queue->is_closed is not set until later in uverbs_hot_unplug_completion_event_file(). In between the time ib_uverbs_comp_event_close() releases the lock and uverbs_hot_unplug_completion_event_file() acquires the lock, a completion event can arrive and be inserted into the event queue by ib_uverbs_comp_handler(). This can cause a "double add" list_add warning or crash depending on the kernel configuration, or a memory leak because the event is never dequeued since the queue is already closed down. So add setting ev_queue->is_closed = 1 to ib_uverbs_comp_event_close(). Cc: [email protected] Fixes: 1e7710f3f656 ("IB/core: Change completion channel to use the reworked objects schema") Signed-off-by: Steve Wise <[email protected]> Signed-off-by: Jason Gunthorpe <[email protected]>
2018-09-12IB/mlx5: Allow transition of DCI QP to resetMoni Shoua1-1/+3
The transition is allowed from any state and the atrribute mask must be IB_QP_STATE. Fixes: c32a4f296e1d ("IB/mlx5: Add support for DC Initiator QP") Signed-off-by: Moni Shoua <[email protected]> Reviewed-by: Artemy Kovalyov <[email protected]> Signed-off-by: Leon Romanovsky <[email protected]> Signed-off-by: Jason Gunthorpe <[email protected]>
2018-09-11IB/hfi1,PCI: Allow bus reset while probingDennis Dalessandro1-7/+4
Calling into the new API to reset the secondary bus results in a deadlock. This occurs because the device/bus is already locked at probe time. Reverting back to the old behavior while the API is improved. Link: https://bugzilla.kernel.org/show_bug.cgi?id=200985 Fixes: c6a44ba950d1 ("PCI: Rename pci_try_reset_bus() to pci_reset_bus()") Fixes: 409888e0966e ("IB/hfi1: Use pci_try_reset_bus() for initiating PCI Secondary Bus Reset") Signed-off-by: Dennis Dalessandro <[email protected]> Signed-off-by: Bjorn Helgaas <[email protected]> Reviewed-by: Michael J. Ruhl <[email protected]> Cc: Sinan Kaya <[email protected]>
2018-09-11IB/hfi1: set_intr_bits uses incorrect source for register modificationMichael J. Ruhl1-1/+1
HFI IRQ enable bits are not being set correctly. Send context error and DC IRQs are not being enabled correctly. In addition, send context error IRQs are not being delivered. Because of this, send context errors are not being handled correctly when they occur. When setting the IRQ bits, if an IRQ range is used, and the last bit is on a register boundary (bit 63), the calculated index for the final register modification is incorrect (index + 1 vs. index). The incorrect index calculation causes incorrect IRQ bits to be set. In this case the send context error IRQ is NOT enabled. Fix by using the 'last' value rather than the counted 'src' value to determine the final index to use. This satisfies all cases. Fixes: a2f7bbdc2dba ("IB/hfi1: Rework the IRQ API to be more flexible") Reviewed-by: Mike Marciniszyn <[email protected]> Reviewed-by: Dennis Dalessandro <[email protected]> Signed-off-by: Michael J. Ruhl <[email protected]> Signed-off-by: Dennis Dalessandro <[email protected]> Signed-off-by: Jason Gunthorpe <[email protected]>
2018-09-11IB/hfi1: Missing return value in error path for user sdmaMichael J. Ruhl1-1/+3
If the set_txreq_header_agh() function returns an error, the exit path is chosen. In this path, the code fails to set the return value. This will cause the caller to not realize an error has occurred. Set the return value correctly in the error path. Signed-off-by: Michael J. Ruhl <[email protected]> Signed-off-by: Dennis Dalessandro <[email protected]> Signed-off-by: Jason Gunthorpe <[email protected]>
2018-09-11IB/hfi1: Right size user_sdma sequence numbers and related variablesMichael J. Ruhl4-11/+11
Hardware limits the maximum number of packets to u16 packets. Match that size for all relevant sequence numbers in the user_sdma engine. Reviewed-by: Mike Marciniszyn <[email protected]> Signed-off-by: Michael J. Ruhl <[email protected]> Signed-off-by: Dennis Dalessandro <[email protected]> Signed-off-by: Jason Gunthorpe <[email protected]>
2018-09-11IB/hfi1: Remove race conditions in user_sdma send pathMichael J. Ruhl2-18/+15
Packet queue state is over used to determine SDMA descriptor availablitity and packet queue request state. cpu 0 ret = user_sdma_send_pkts(req, pcount); cpu 0 if (atomic_read(&pq->n_reqs)) cpu 1 IRQ user_sdma_txreq_cb calls pq_update() (state to _INACTIVE) cpu 0 xchg(&pq->state, SDMA_PKT_Q_ACTIVE); At this point pq->n_reqs == 0 and pq->state is incorrectly SDMA_PKT_Q_ACTIVE. The close path will hang waiting for the state to return to _INACTIVE. This can also change the state from _DEFERRED to _ACTIVE. However, this is a mostly benign race. Remove the racy code path. Use n_reqs to determine if a packet queue is active or not. Reviewed-by: Mitko Haralanov <[email protected]> Reviewed-by: Mike Marciniszyn <[email protected]> Signed-off-by: Michael J. Ruhl <[email protected]> Signed-off-by: Dennis Dalessandro <[email protected]> Signed-off-by: Jason Gunthorpe <[email protected]>
2018-09-11IB/hfi1: Eliminate races in the SDMA send error pathMichael J. Ruhl2-51/+39
pq_update() can only be called in two places: from the completion function when the complete (npkts) sequence of packets has been submitted and processed, or from setup function if a subset of the packets were submitted (i.e. the error path). Currently both paths can call pq_update() if an error occurrs. This race will cause the n_req value to go negative, hanging file_close(), or cause a crash by freeing the txlist more than once. Several variables are used to determine SDMA send state. Most of these are unnecessary, and have code inspectible races between the setup function and the completion function, in both the send path and the error path. The request 'status' value can be set by the setup or by the completion function. This is code inspectibly racy. Since the status is not needed in the completion code or by the caller it has been removed. The request 'done' value races between usage by the setup and the completion function. The completion function does not need this. When the number of processed packets matches npkts, it is done. The 'has_error' value races between usage of the setup and the completion function. This can cause incorrect error handling and leave the n_req in an incorrect value (i.e. negative). Simplify the code by removing all of the unneeded state checks and variables. Clean up iovs node when it is freed. Eliminate race conditions in the error path: If all packets are submitted, the completion handler will set the completion status correctly (ok or aborted). If all packets are not submitted, the caller must wait until the submitted packets have completed, and then set the completion status. These two change eliminate the race condition in the error path. Reviewed-by: Mitko Haralanov <[email protected]> Reviewed-by: Mike Marciniszyn <[email protected]> Signed-off-by: Michael J. Ruhl <[email protected]> Signed-off-by: Dennis Dalessandro <[email protected]> Signed-off-by: Jason Gunthorpe <[email protected]>
2018-09-11RDMA/hns: Fix an error code in hns_roce_v2_init_eq_table()Dan Carpenter1-0/+1
The error code isn't set on this path. Signed-off-by: Dan Carpenter <[email protected]> Signed-off-by: Jason Gunthorpe <[email protected]>
2018-09-11IB/{hfi1, qib, rdmavt}: Schedule multi RC/UC packets instead of postingMichael J. Ruhl5-25/+24
The post_send() path determines if it should post directly or, schedule the post for later. The current logic is: if the swqe ring is empty or (for hfi1) wqe->length <= piothreshold post the send else schedule This can allow large requests to call the send engine directly. Large requests can potentially produce a large number of packets prior to returning to the caller, blocking the caller from posting more requests, and allowing better parallel processing. Allow the driver(s) more say in this logic (pass call_send to the driver, rather than examining a return value). Update hfi1/qib logic to schedule the send engine if an RC or UC message is larger than the QP MTU size. Reviewed-by: Mike Marciniszyn <[email protected]> Reviewed-by: Ira Weiny <[email protected]> Signed-off-by: Michael J. Ruhl <[email protected]> Signed-off-by: Dennis Dalessandro <[email protected]> Signed-off-by: Jason Gunthorpe <[email protected]>
2018-09-11infiniband: remove redundant condition check before debugfs_removezhong jiang1-2/+1
debugfs_remove has taken the IS_ERR_OR_NULL into account. Just remove the unnecessary condition. Signed-off-by: zhong jiang <[email protected]> Signed-off-by: Jason Gunthorpe <[email protected]>
2018-09-11RDMA/mlx5: Allow creating a matcher for a NIC TX flow tableMark Bloch2-6/+33
Currently a matcher can only be created and attached to a NIC RX flow table. Extend it to allow it on NIC TX flow tables as well. In order to achieve that, we: 1) Expose a new attribute: MLX5_IB_ATTR_FLOW_MATCHER_FLOW_FLAGS. enum ib_flow_flags is used as valid flags. Only IB_FLOW_ATTR_FLAGS_EGRESS is supported. 2) Remove the requirement to have a DEVX or QP destination when creating a flow. A flow added to NIC TX flow table will forward the packet outside of the vport (Wire or E-Switch in the SR-iOV case). Signed-off-by: Mark Bloch <[email protected]> Reviewed-by: Yishai Hadas <[email protected]> Signed-off-by: Leon Romanovsky <[email protected]> Signed-off-by: Jason Gunthorpe <[email protected]>
2018-09-11RDMA/mlx5: Add NIC TX namespace when getting a flow tableMark Bloch3-11/+32
Add the ability to get a NIC TX flow table when using _get_flow_table(). This will allow to create a matcher and a flow rule on the NIC TX path. Signed-off-by: Mark Bloch <[email protected]> Reviewed-by: Yishai Hadas <[email protected]> Signed-off-by: Leon Romanovsky <[email protected]> Signed-off-by: Jason Gunthorpe <[email protected]>
2018-09-11RDMA/mlx5: Add flow actions support to raw create flowMark Bloch2-8/+43
Support attaching flow actions to a flow rule via raw create flow. For now only NIC RX path is supported. This change requires to export flow resources management functions so we can maintain proper bookkeeping of flow actions. Signed-off-by: Mark Bloch <[email protected]> Reviewed-by: Yishai Hadas <[email protected]> Signed-off-by: Leon Romanovsky <[email protected]> Signed-off-by: Jason Gunthorpe <[email protected]>
2018-09-11RDMA/mlx5: Refactor raw flow creationMark Bloch3-7/+13
Move struct mlx5_flow_act to be passed from the method entry point, this will allow to add support for flow action for the raw create flow path. Signed-off-by: Mark Bloch <[email protected]> Reviewed-by: Yishai Hadas <[email protected]> Signed-off-by: Leon Romanovsky <[email protected]> Signed-off-by: Jason Gunthorpe <[email protected]>
2018-09-11RDMA/mlx5: Don't overwrite action if already setMark Bloch1-0/+10
We support only a single action type per flow rule, in case the user passes the same type of flow actions fail the flow creation. Signed-off-by: Mark Bloch <[email protected]> Reviewed-by: Yishai Hadas <[email protected]> Signed-off-by: Leon Romanovsky <[email protected]> Signed-off-by: Jason Gunthorpe <[email protected]>
2018-09-11RDMA/mlx5: Refactor flow action parsing to be more genericMark Bloch2-6/+10
Make the parsing of flow actions more generic so it could be used by mlx5 raw create flow. Signed-off-by: Mark Bloch <[email protected]> Reviewed-by: Yishai Hadas <[email protected]> Signed-off-by: Leon Romanovsky <[email protected]> Signed-off-by: Jason Gunthorpe <[email protected]>
2018-09-11RDMA/uverbs: Move flow resources initializationMark Bloch3-24/+3
Use ib_set_flow() when initializing flow related resources. Signed-off-by: Mark Bloch <[email protected]> Reviewed-by: Yishai Hadas <[email protected]> Signed-off-by: Leon Romanovsky <[email protected]> Signed-off-by: Jason Gunthorpe <[email protected]>
2018-09-11IB/uverbs: Add IDRs array attribute type to ioctl() interfaceGuy Levi2-0/+126
Methods sometimes need to get a flexible set of IDRs and not a strict set as can be achieved today by the conventional IDR attribute. Add a new IDRS_ARRAY attribute to the generic uverbs ioctl layer. IDRS_ARRAY points to array of idrs of the same object type and same access rights, only write and read are supported. Signed-off-by: Guy Levi <[email protected]> Signed-off-by: Jason Gunthorpe <[email protected]>`` Signed-off-by: Leon Romanovsky <[email protected]> Signed-off-by: Jason Gunthorpe <[email protected]>
2018-09-11RDMA/mlx5: Enable attaching packet reformat action to steering flowsMark Bloch1-0/+8
Any matching rules will be mutated based on the packet reformat context which is attached to that given flow rule. Signed-off-by: Mark Bloch <[email protected]> Signed-off-by: Leon Romanovsky <[email protected]> Signed-off-by: Jason Gunthorpe <[email protected]>
2018-09-11RDMA/mlx5: Enable reformat on NIC RX if supportedMark Bloch1-0/+4
A L3_TUNNEL_TO_L2 decap flow action requires to enable the encap bit on the flow table, enable it if supported. This will allow to attach those flow actions to NIC RX steering. We don't enable if running on a representor. Signed-off-by: Mark Bloch <[email protected]> Signed-off-by: Leon Romanovsky <[email protected]> Signed-off-by: Jason Gunthorpe <[email protected]>
2018-09-11RDMA/mlx5: Enable attaching DECAP action to steering flowsMark Bloch1-0/+5
Any matching packet will be stripped of it's VXLAN tunnel, only the inner L2 onward is left. The user will receive the decapsulated packet. Signed-off-by: Mark Bloch <[email protected]> Signed-off-by: Leon Romanovsky <[email protected]> Signed-off-by: Jason Gunthorpe <[email protected]>
2018-09-11RDMA/mlx5: Enable decap and packet reformat on flow tablesMark Bloch1-4/+13
If NIC RX flow tables support decap operation, enable it on creation, This allows to perform decapsulation of tunnelled packets by steering rules. If NIC TX flow tables support reformat operation, enable it on creation. We don't enable those capabilities on representors as the E-Switch should handle packet modification (can be configured via TC) and as current hardware can't handle both FDB and NIC flow tables with decap/packet reformat support. Signed-off-by: Mark Bloch <[email protected]> Signed-off-by: Leon Romanovsky <[email protected]> Signed-off-by: Jason Gunthorpe <[email protected]>
2018-09-11RDMA/mlx5: Enable attaching modify header to steering flowsMark Bloch1-0/+8
When creating a flow steering rule, allow the user to attach a modify header action. Signed-off-by: Mark Bloch <[email protected]> Signed-off-by: Leon Romanovsky <[email protected]> Signed-off-by: Jason Gunthorpe <[email protected]>
2018-09-11RDMA/mlx5: Add NIC TX steering supportMark Bloch2-10/+20
Just like ingress steering, allow a user to create steering rules that match egress vport traffic. We expose the same number of priorities as the bypass (NIC RX) steering. Signed-off-by: Mark Bloch <[email protected]> Signed-off-by: Leon Romanovsky <[email protected]> Signed-off-by: Jason Gunthorpe <[email protected]>
2018-09-10infiniband: nes: Use skb_peek_next() and skb_queue_walk().David S. Miller1-5/+3
Instead of direct SKB list accesses. Signed-off-by: David S. Miller <[email protected]>
2018-09-06RDMA/core: Assign device ifindex before publishing the deviceParav Pandit1-1/+2
Even though device->ifindex is assigned before adding the device in the list which is read by netlink flow, it is better to assign rdma device index before publishing the device in the system to users and clients. Signed-off-by: Parav Pandit <[email protected]> Reviewed-by: Daniel Jurgens <[email protected]> Signed-off-by: Leon Romanovsky <[email protected]> Signed-off-by: Jason Gunthorpe <[email protected]>
2018-09-06RDMA/core: Follow correct unregister order between sysfs and cgroupParav Pandit1-1/+1
During register_device() init sequence is, (a) register with rdma cgroup followed by (b) register with sysfs Therefore, unregister_device() sequence should follow the reverse order. Signed-off-by: Parav Pandit <[email protected]> Reviewed-by: Daniel Jurgens <[email protected]> Signed-off-by: Leon Romanovsky <[email protected]> Signed-off-by: Jason Gunthorpe <[email protected]>
2018-09-06RDMA/umem: Restore lockdep check while downgrading lockLeon Romanovsky1-6/+0
Lockdep engine handles correctly downgrade of locks and it simply incorrect to disable lockdep checks prior to calling mmu_notifier. Remove lockdep_off and ensure locks correctness. Signed-off-by: Leon Romanovsky <[email protected]> Signed-off-by: Jason Gunthorpe <[email protected]>
2018-09-06RDMA/core: Define client_data_lock as rwlock instead of spinlockParav Pandit1-15/+15
Even though device registration/unregistration and client registration/unregistration is not a performance path, define the client_data_lock as rwlock for code clarity. Signed-off-by: Parav Pandit <[email protected]> Reviewed-by: Daniel Jurgens <[email protected]> Signed-off-by: Leon Romanovsky <[email protected]> Signed-off-by: Jason Gunthorpe <[email protected]>
2018-09-06RDMA/core: Use simpler spin lock irq API from blocking contextParav Pandit1-11/+9
add_client_context(), ib_unregister_device() and ib_unregister_client() are designed to call from blocking context. There is no need to save and restore last interrupt state when called from such blocking context. Even though this is not a performance path, using the right spin lock API is desired for code clarity. To avoid checkpatch warning while removing flags, sizeof() is used. Signed-off-by: Parav Pandit <[email protected]> Reviewed-by: Daniel Jurgens <[email protected]> Signed-off-by: Leon Romanovsky <[email protected]> Signed-off-by: Jason Gunthorpe <[email protected]>
2018-09-06RDMA/core: Remove context entries from list while unregistering deviceParav Pandit1-1/+5
While unregistering a device, remove the context elements from the list to not have any stale entries. With that any errors/bugs can be checked when device is freed. Signed-off-by: Parav Pandit <[email protected]> Reviewed-by: Daniel Jurgens <[email protected]> Signed-off-by: Leon Romanovsky <[email protected]> Signed-off-by: Jason Gunthorpe <[email protected]>
2018-09-06RDMA/core: Use simplified list_for_eachParav Pandit1-5/+4
While traversing client_data_list in following conditions, linked list is only read, no elements of the list are removed. Therefore, use list_for_each_entry(), instead of list_for_each_safe(). Signed-off-by: Parav Pandit <[email protected]> Reviewed-by: Daniel Jurgens <[email protected]> Signed-off-by: Leon Romanovsky <[email protected]> Signed-off-by: Jason Gunthorpe <[email protected]>
2018-09-06RDMA/core: No need to protect kfree with spin lock and semaphoreParav Pandit1-1/+1
While unregistering a client, only context removal should be protected with lock. There is no need to protect a freeing of such context which is already removed from the list. Signed-off-by: Parav Pandit <[email protected]> Reviewed-by: Daniel Jurgens <[email protected]> Signed-off-by: Leon Romanovsky <[email protected]> Signed-off-by: Jason Gunthorpe <[email protected]>
2018-09-06RDMA/{cma, core}: Avoid callback on rdma_addr_cancel()Parav Pandit2-9/+7
Currently rdma_addr_cancel() is an async operation, which notifies that cancel is done by executing the callback function given during rdma_resolve_ip(). If resolve_ip request is already completed than callback is not executed. Instead, now rdma_resolve_addr() and rdma_addr_cancel() simplified in following ways. 1. rdma_addr_cancel() now a synchronous method. If request was pending, after it is cancelled, no callback is notified. 2. rdma_resolve_addr() and respective addr_handler() callback doesn't need to hold reference to cm_id. Signed-off-by: Parav Pandit <[email protected]> Signed-off-by: Leon Romanovsky <[email protected]> Signed-off-by: Jason Gunthorpe <[email protected]>
2018-09-06RDMA/core: Rate limit MAD error messagesParav Pandit1-35/+37
While registering a mad agent, a user space can trigger various errors and flood the logs. Therefore, decrease verbosity and rate limit such error messages. While we are at it, use __func__ to print function name. Signed-off-by: Parav Pandit <[email protected]> Signed-off-by: Leon Romanovsky <[email protected]> Reviewed-by: Dennis Dalessandro <[email protected]> Signed-off-by: Jason Gunthorpe <[email protected]>
2018-09-06IB/ipoib: Ensure that MTU isn't less than minimum permittedMuhammad Sammar1-1/+2
It is illegal to change MTU to a value lower than the minimum MTU stated in ethernet spec. In addition to that we need to add 4 bytes for encapsulation header (IPOIB_ENCAP_LEN). Before "ifconfig ib0 mtu 0" command, succeeds while it obviously shouldn't. Signed-off-by: Muhammad Sammar <[email protected]> Reviewed-by: Feras Daoud <[email protected]> Signed-off-by: Leon Romanovsky <[email protected]> Signed-off-by: Jason Gunthorpe <[email protected]>
2018-09-06IB/mlx5: Don't hold spin lock while checking device stateParav Pandit1-14/+12
mdev->state device state is not protected by the QP for which WRs are being processed. Therefore, there is no need to hold spin lock while checking mdev state. Given that device fatal error is unlikely situation, wrap the condition check with unlikely(). Additionally, kernel QP1 is also a kernel ULP for which soft CQEs needs to be generated. Therefore, check for device fatal error before processing QP1 work requests. Fixes: 89ea94a7b6c4 ("IB/mlx5: Reset flow support for IB kernel ULPs") Signed-off-by: Parav Pandit <[email protected]> Reviewed-by: Daniel Jurgens <[email protected]> Reviewed-by: Maor Gottlieb <[email protected]> Signed-off-by: Leon Romanovsky <[email protected]> Signed-off-by: Jason Gunthorpe <[email protected]>
2018-09-06RDMA/core: Fail early if unsupported QP is providedParav Pandit1-0/+4
When requested QP type is not supported for a {device, port}, return the error right away before validating all parameters during mad agent registration time. Signed-off-by: Parav Pandit <[email protected]> Reviewed-by: Daniel Jurgens <[email protected]> Signed-off-by: Leon Romanovsky <[email protected]> Signed-off-by: Jason Gunthorpe <[email protected]>
2018-09-06RDMA/mlx4: Ensure that maximal send/receive SGE less than supported by HWLeon Romanovsky1-3/+5
In calculating the global maximum number of the Scatter/Gather elements supported, the following four maximum parameters must be taken into consideration: max_sg_rq, max_sg_sq, max_desc_sz_rq and max_desc_sz_sq. However instead of bringing this complexity to query_device, which still won't be sufficient anyway (the calculations are dependent on QP type), the safer approach will be to restore old code, which will give us 32 SGEs. Fixes: 33023fb85a42 ("IB/core: add max_send_sge and max_recv_sge attributes") Reported-by: Chuck Lever <[email protected]> Signed-off-by: Leon Romanovsky <[email protected]> Signed-off-by: Jason Gunthorpe <[email protected]>
2018-09-06RDMA/cma: Protect cma dev list with lockParav Pandit1-5/+7
When AF_IB addresses are used during rdma_resolve_addr() a lock is not held. A cma device can get removed while list traversal is in progress which may lead to crash. ie CPU0 CPU1 ==== ==== rdma_resolve_addr() cma_resolve_ib_dev() list_for_each() cma_remove_one() cur_dev->device mutex_lock(&lock) list_del(); mutex_unlock(&lock); cma_process_remove(); Therefore, hold a lock while traversing the list which avoids such situation. Cc: <[email protected]> # 3.10 Fixes: f17df3b0dede ("RDMA/cma: Add support for AF_IB to rdma_resolve_addr()") Signed-off-by: Parav Pandit <[email protected]> Reviewed-by: Daniel Jurgens <[email protected]> Signed-off-by: Leon Romanovsky <[email protected]> Reviewed-by: Dennis Dalessandro <[email protected]> Signed-off-by: Jason Gunthorpe <[email protected]>
2018-09-05Merge branch 'uverbs_dev_cleanups' into rdma.git for-nextJason Gunthorpe9-85/+92
For dependencies, branch based on rdma.git 'for-rc' of https://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma.git/ Pull 'uverbs_dev_cleanups' from Leon Romanovsky: ==================== Reuse the char device code interfaces to simplify ib_uverbs_device creation and destruction. As part of this series, we are sending fix to cleanup path, which was discovered during internal review, The fix definitely can go to -rc, but it means that this series will be dependent on rdma-rc. ==================== * branch 'uverbs_dev_cleanups': RDMA/uverbs: Use device.groups to initialize device attributes RDMA/uverbs: Use cdev_device_add() instead of cdev_add() RDMA/core: Depend on device_add() to add device attributes RDMA/uverbs: Fix error cleanup path of ib_uverbs_add_one() Resolved conflict in ib_device_unregister_sysfs() Signed-off-by: Jason Gunthorpe <[email protected]>
2018-09-05RDMA/uverbs: Use device.groups to initialize device attributesParav Pandit2-13/+19
Instead of explicitly adding device attribute files and handling such error conditions, depend on device core layer to create device attributes files based group pointer NULL terminated array. Signed-off-by: Parav Pandit <[email protected]> Reviewed-by: Daniel Jurgens <[email protected]> Signed-off-by: Leon Romanovsky <[email protected]> Signed-off-by: Jason Gunthorpe <[email protected]>
2018-09-05RDMA/uverbs: Use cdev_device_add() instead of cdev_add()Parav Pandit2-39/+30
Instead of doing two step process to add char device and create underlying device, use cdev_device_add() which does both. Currently a kobject per uverbs_device is created to keep reference to its holding ib_uverbs_device in addition to its underlying device 'dev'. Instead just use uverbs_device->dev to keep a reference to. With this change there is single reference tracker for ib_uverbs_device structure. This allows for subsequent patch to registers group attribute as well using single API cdev_device_add(). Signed-off-by: Parav Pandit <[email protected]> Reviewed-by: Daniel Jurgens <[email protected]> Signed-off-by: Leon Romanovsky <[email protected]> Signed-off-by: Jason Gunthorpe <[email protected]>
2018-09-05RDMA/core: Depend on device_add() to add device attributesParav Pandit1-34/+27
Instead of adding/removing device attribute files, depend on device_add() which considers adding these device files based on NULL terminated attributes group array. Signed-off-by: Parav Pandit <[email protected]> Reviewed-by: Daniel Jurgens <[email protected]> Signed-off-by: Leon Romanovsky <[email protected]> Signed-off-by: Jason Gunthorpe <[email protected]>
Powered by cgit, Gruvboxed by Blaster4385.