- 07 May, 2019 6 commits
-
-
Dennis Dalessandro authored
When IPoIB receives an SM LID change event, it reacts by flushing its path record cache and rejoining multicast groups. This is the same behavior it performs when it receives a reregistration event. This behavior is unnecessary as an SM may have database backup or synchronization mechanisms which permit the SM location or LID to change without loss of multicast membership and without impact to path records. Both opensm and the OPA FM issue reregistration events if a new SM is started (or restarted with a new config) or an SM event occurs which results in loss of multicast membership records by the SM (such as opensm failover) or the SM encounters new nodes with Active ports (such as after joining 2 fabrics by connecting switches via ISLs). Hence this event can be depended on as the trigger for IPoIB cache and multicast flushing. It appears that some drivers, such as qib, and hfi1 issue the IB_EVENT_SM_CHANGE but other drivers such as mlx4 and mlx5 do not. Empirical testing on Mellanox EDR using ibv_asyncwatch has confirmed that Mellanox EDR HCAs do not generate SM change events and that opensm does generate reregistration. An SM LID change event is generated by the mentioned drivers to reflect that sm_lid and/or sm_sl in the local port info has changed. The intent of this event is to permit applications and ULPs which have a local copy of this information (or an address handle using it) to update their information. The intent is that the reregistration event (caused by the SM via a bit in Set(PortInfo)) be used to inform nodes that they need to rejoin multicast groups, resubscribe for notices and potentially update path records. When an SM migrates or fails over, a SM LID change event can occur. In response IPoIB discards path records and multicast membership and loses connectivity until these records are restored via SA requests. In very large fabrics, it may take minutes for the SM to be ready and for the SA responses to be supplied. This can result in undesirable and unnecessary IPoIB connectivity impacts. It also can result in an unnecessary storm of SA queries from all nodes in a cluster potentially followed by yet another storm if the SM issues the reregistration request. The fact the Mellanox HCAs do not even generate this event, is further evidence that on modern IB fabrics there will be no ill side effects from the proposed changes below to reduce the reaction by 3 kernel components to this event. So these changes should be benign for Mellanox IB fabrics and will benefit OPA fabrics while also making ib_core and ULP behavor "correct" as intended by the IBTA spec and kernel RDMA event APIs. Address these issues by removing IB_EVENT_SM_CHANGE handling from ipoib. IPoIB does not locally store sm_lid nor sm_sl, so it does not need to do anything on SM LID change. IPoIB makes use of other ib_core components to issue SA requests for it and those components correctly track SM LID and SM LID changes. Also in ib_core multicast handling, remove the test for IB_EVENT_SM_CHANGE. This code is moving all multicast groups to the error state, which will trigger rejoins. This code is used by IPoIB as well as the connection manager and other clients of multicast groups. This kernel module centralizes group membership status and joins since a node can only join a given group once but multiple ULPs or applications may want to join the same group. It makes use of the sa_query.c component in ib_core, which correctly trackes SM LID and SL. This component does not track SM LID nor SL itself and hence need not react to their changes. Similarly in the ib_core cache code remove the handling for the IB_EVENT_SM_CHANGE. In this function. The ib_cache_update function which is ultimately called is updating local copies of the pkey table, gid table and lmc. It does not update nor retain sm_lid nor sm_sl. As such it does not need to be called on an SM LID change. It technically also does not need to be called on a reregistration. The LID_CHANGE, PKEY_CHANGE, GID_CHANGE and port state change events (PORT_ERR, PORT_ACTICE) should be sufficient triggers. It is worth noting that the alternative of simply having the hfi1 and qib drivers not generate the SM LID change event was explored. While this would duplicate what Mellanox drivers do now, it is not the correct behavior and removes the ability for an SM to migrate without requiring reregistration. Since both opensm and OPA SM have mechanisms to backup or synchronize registration information, it is desirable to let them perform SM migrations (with LID or SL changes) without requiring reregistration when they deem it appropriate. Suggested-by:
Todd Rimmer <todd.rimmer@intel.com> Tested-by:
Michael Brooks <michael.brooks@intel.com> Reviewed-by:
Mike Marciniszyn <mike.marciniszyn@intel.com> Reviewed-by:
Dennis Dalessandro <dennis.dalessandro@intel.com> Signed-off-by:
Jason Gunthorpe <jgg@mellanox.com>
-
Leon Romanovsky authored
When the refcount is 0 the device is invisible to netlink. However in the patch below the refcount = 1 was moved to after the device_add(). This creates a race where userspace can issue a netlink query after the device_add() event and not see the device as visible. Ensure that no uevent is fired before device is fully registered. Fixes: d79af724 ("RDMA/device: Expose ib_device_try_get(()") Signed-off-by:
Leon Romanovsky <leonro@mellanox.com> Signed-off-by:
-
Gal Pressman authored
Commit d901b276 ("lib/scatterlist: Provide a DMA page iterator") added the sg DMA iterator but a leftover remained in the sg_page_iter documentation as you cannot get the page dma address (only the page itself), fix it. Signed-off-by:
Gal Pressman <galpress@amazon.com> Reviewed-by:
Mukesh Ojha <mojha@codeaurora.org> Signed-off-by:
-
Gal Pressman authored
Add EFA Makefile and Kconfig. Signed-off-by:
Steve Wise <swise@opengridcomputing.com> Signed-off-by:
-
Gal Pressman authored
Add the main EFA module file which takes care of device probe/initialization/registration/etc. Signed-off-by:
Shiraz Saleem <shiraz.saleem@intel.com> Reviewed-by:
-
Gal Pressman authored
Add a file that implements the EFA verbs. Signed-off-by:
-
- 06 May, 2019 20 commits
-
-
Gal Pressman authored
Add the EFA common commands implementation. Signed-off-by:
-
Gal Pressman authored
Add admin commands submissions/completions implementation. Signed-off-by:
-
Gal Pressman authored
Add the EFA ABI file exposed to userspace. Signed-off-by:
-
Gal Pressman authored
Header file for the various commands that can be sent through admin queue. This includes queue create/modify/destroy, setting up and remove protection domains, address handlers, and memory registration, etc. Signed-off-by:
-
Gal Pressman authored
A helper header file for EFA admin queue, admin queue completion, asynchronous notification queue, and various hardware configuration data structures and functions. Signed-off-by:
-
Gal Pressman authored
Add EFA driver generic header file defining driver's device independent internal data structures and definitions. Signed-off-by:
-
Gal Pressman authored
EFA PCIe device implements a single Admin Queue (AQ) and Admin Completion Queue (ACQ) pair to initialize and communicate configuration with the device. Through this pair, we run set/get commands for querying and configuring the device, create/modify/destroy queues, and IB specific commands like Address Handler (AH), Memory Registration (MR) and Protection Domains (PD). In addition to admin (AQ/ACQ), we have data path queues that get classified as Queue Pairs (QP) and Completion Queues (CQ). Signed-off-by:
-
Gal Pressman authored
Add EFA driver ID to the IOCTL interface uapi. This patch also adds unspecified node/transport type that will be used by EFA (usnic is left unchanged as it's already part of our ABI). Signed-off-by:
-
Shiraz Saleem authored
The drivers i40iw and bnxt_re no longer dependent on the hugetlb flag. So remove this flag from ib_umem structure. Reviewed-by:
Michael J. Ruhl <michael.j.ruhl@intel.com> Signed-off-by:
-
Shiraz Saleem authored
Call the core helpers to retrieve the HW aligned address to use for the MR, within a supported bnxt_re page size. Remove checking the umem->hugtetlb flag as it is no longer required. The new DMA block iterator will return the 2M aligned address if the MR is backed by 2M huge pages. Acked-by:
Selvin Xavier <selvin.xavier@broadcom.com> Signed-off-by:
-
Shiraz Saleem authored
Call the core helpers to retrieve the HW aligned address to use for the MR, within a supported i40iw page size. Remove code in i40iw to determine when MR is backed by 2M huge pages which involves checking the umem->hugetlb flag and VMA inspection. The new DMA iterator will return the 2M aligned address if the MR is backed by 2M pages. Fixes: f26c7c83 ("i40iw: Add 2MB page support") Reviewed-by:
-
Shiraz Saleem authored
This helper iterates over a DMA-mapped SGL and returns contiguous memory blocks aligned to a HW supported page size. Suggested-by:
Jason Gunthorpe <jgg@ziepe.ca> Signed-off-by:
-
Shiraz Saleem authored
This helper iterates through the SG list to find the best page size to use from a bitmap of HW supported page sizes. Drivers that support multiple page sizes, but not mixed sizes in an MR can use this API. Suggested-by:
-
Mike Marciniszyn authored
The work_item cancels that occur when a QP is destroyed can elicit the following trace: workqueue: WQ_MEM_RECLAIM ipoib_wq:ipoib_cm_tx_reap [ib_ipoib] is flushing !WQ_MEM_RECLAIM hfi0_0:_hfi1_do_send [hfi1] WARNING: CPU: 7 PID: 1403 at kernel/workqueue.c:2486 check_flush_dependency+0xb1/0x100 Call Trace: __flush_work.isra.29+0x8c/0x1a0 ? __switch_to_asm+0x40/0x70 __cancel_work_timer+0x103/0x190 ? schedule+0x32/0x80 iowait_cancel_work+0x15/0x30 [hfi1] rvt_reset_qp+0x1f8/0x3e0 [rdmavt] rvt_destroy_qp+0x65/0x1f0 [rdmavt] ? _cond_resched+0x15/0x30 ib_destroy_qp+0xe9/0x230 [ib_core] ipoib_cm_tx_reap+0x21c/0x560 [ib_ipoib] process_one_work+0x171/0x370 worker_thread+0x49/0x3f0 kthread+0xf8/0x130 ? max_active_store+0x80/0x80 ? kthread_bind+0x10/0x10 ret_from_fork+0x35/0x40 Since QP destruction frees memory, hfi1_wq should have the WQ_MEM_RECLAIM. The hfi1_wq does not allocate memory with GFP_KERNEL or otherwise become entangled with memory reclaim, so this flag is appropriate. Fixes: 0a226edd ("staging/rdma/hfi1: Use parallel workqueue for SDMA engines") Reviewed-by:
-
Leon Romanovsky authored
MAYEXEC flag was mistakenly added in the commit cited in the fixes line. Fixes: 4eb6ab13 ("RDMA: Remove rdma_user_mmap_page") Signed-off-by:
-
Ariel Levkovich authored
For DEVX users who have SYS_RAWIO capability, we set the internal device resources capability when creating the UCTX. This will allow the device to restrict the allocation of internal device resources such as SW ICM memory to privileged DEVX users only. Signed-off-by:
Ariel Levkovich <lariel@mellanox.com> Reviewed-by:
Eli Cohen <eli@mellanox.com> Reviewed-by:
Mark Bloch <markb@mellanox.com> Signed-off-by:
-
Ariel Levkovich authored
This patch adds support for allocating, deallocating and registering a new device memory type, STEERING_SW_ICM. This memory can be allocated and used by a privileged user for direct rule insertion and management of the device's steering tables. The type is provided by the user via the dedicated attribute in the alloc_dm ioctl command. Signed-off-by:
-
Ariel Levkovich authored
Adding a warning on allocated MEMIC buffers that weren't freed prior to driver tear down. Signed-off-by:
-
Ariel Levkovich authored
This patch intoruduces a new mlx5_ib driver attribute to the DM allocation method - the DM type. In order to allow addition of new types in downstream patches this patch also refactors the allocation, deallocation and registration handlers to consider the requested type and perform the necessary actions according to it. Since not all future device memory types will be such that are mapped to user memory, the mandatory page index output attribute is modified to be optional. Signed-off-by:
-
Leon Romanovsky authored
Prior to commit d3456914 ("RDMA: Handle AH allocations by IB/core"), AH destroy path is rdmavt returned -EBUSY warning to application and caused to potential leakage of kernel memory of AH structure. After that commit, the AH structure is always freed but such early return in driver code can potentially cause to use-after-free error. Add warning to catch such situation to help driver developers to fix AH release path. Signed-off-by:
-
- 03 May, 2019 12 commits
-
-
Parav Pandit authored
When there is active traffic through a GID, a QP/AH holds reference to this GID entry. RoCE GID entry holds reference to its attached netdevice. Due to this when netdevice is deleted by admin user, its refcount is not dropped. Therefore, while deleting RoCE GID, wait for all GID attribute's netdev users to finish accessing netdev in rcu context. Once all users done accessing it, release the netdev refcount. Signed-off-by:
Huy Nguyen <huyn@mellanox.com> Signed-off-by:
Parav Pandit <parav@mellanox.com> Signed-off-by:
-
Parav Pandit authored
Use core provided API to fill the source MAC address and use rdma_read_gid_attr_ndev_rcu() to get stable netdev. This is preparation patch to allow gid attribute to become NULL when associated net device is removed. Signed-off-by:
-
Parav Pandit authored
Use rdma_read_gid_attr_ndev_rcu() to access netdevice attached to GID entry under rcu lock. This ensures that while working on the netdevice of the GID, it doesn't get freed. Signed-off-by:
-
Parav Pandit authored
To access the netdevice of the GID attribute, use an existing API rdma_read_gid_attr_ndev_rcu(). This further reduces dependency on open access to netdevice of GID attribute. Signed-off-by:
-
Parav Pandit authored
Instead of RoCE drivers figuring out vlan, smac fields while working on QP/AH, provide a helper routine to read the L2 fields such as vlan_id and source mac address. This moves logic from mlx5 driver to core for wider usage for RoCE ports. This is a preparation patch to allow detaching netdev in subsequent patch. Signed-off-by:
-
Parav Pandit authored
GID type to path record type conversion can be done directly based on port type and gid attribute type. There is no need to find out using indirect way by its GID attribute's ndev field. Signed-off-by:
-
Parav Pandit authored
Always consider the skb reserve space based on netdevice of the GID attribute, regardless of vlan or non vlan netdevice. Fixes: 43c9fc50 ("rdma_rxe: make rxe work over 802.1q VLAN devices") Signed-off-by:
-
Kamal Heib authored
Integrate iw_cm_verbs data members into ib_device_ops and ib_device structs, this is done to achieve the following: 1) Avoid memory related bugs durring error unwind 2) Make the code more cleaner 3) Reduce code duplication Signed-off-by:
Kamal Heib <kamalheib1@gmail.com> Signed-off-by:
-
Parav Pandit authored
The driver interface cannot manipulate the sysfs of the compat device, only of the full device so we must avoid calling the driver sysfs APIs on compat devices. This prevents an oops: Call Trace: dump_stack+0x5a/0x73 kobject_init+0x74/0x80 kobject_init_and_add+0x35/0xb0 hfi1_create_port_files+0x6e/0x3c0 [hfi1] ib_setup_port_attrs+0x43b/0x560 [ib_core] add_one_compat_dev+0x16a/0x230 [ib_core] rdma_dev_init_net+0x110/0x160 [ib_core] ops_init+0x38/0xf0 setup_net+0xcf/0x1e0 copy_net_ns+0xb7/0x130 create_new_namespaces+0x11a/0x1b0 unshare_nsproxy_namespaces+0x55/0xa0 ksys_unshare+0x1a7/0x340 __x64_sys_unshare+0xe/0x20 do_syscall_64+0x5b/0x180 entry_SYSCALL_64_after_hwframe+0x44/0xa9 Fixes: 5417783e ("RDMA/core: Support core port attributes in non init_net") Reported-by:
-
Artemy Kovalyov authored
real_qp should be initialized before ib_destroy_qp() is called. ib_destroy_qp() may be called in the error flow if ib_create_qp_security() failed. Signed-off-by:
Artemy Kovalyov <artemyko@mellanox.com> Signed-off-by:
-
Jack Morgenstein authored
The QP transition optional parameters for the various transition for XRC QPs are identical to those for RC QPs. Many of the XRC QP transition optional parameter bits are missing from the QP optional mask table. These omissions caused failures when doing XRC QP state transitions. For example, when trying to change the response timer of an XRC receive QP via the RTS2RTS transition, the new timer value was ignored because MLX5_QP_OPTPAR_RNR_TIMEOUT bit was missing from the optional params mask for XRC qps for the RTS2RTS transition. Fix this by adding the missing XRC optional parameters for all QP transitions to the opt_mask table. Fixes: e126ba97 ("mlx5: Add driver for Mellanox Connect-IB adapters") Fixes: a4774e90 ("IB/mlx5: Fix opt param mask according to firmware spec") Signed-off-by:
Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by:
-
Shamir Rabinovitch authored
ib_uverbs_get_context does not have a uobject so it does not call the rdma_lookup_get_uobject which is used to set up the uverbs_attr_bundle ucontext. For ib_uverbs_get_context we need to set up this manually before we send the uverbs_attr_bundle down to the driver layer. This completes the change that was done in commit 70f06b26 ("IB: ucontext should be set properly for all cmd & ioctl paths") Signed-off-by:
Shamir Rabinovitch <shamir.rabinovitch@oracle.com> Signed-off-by:
-
- 02 May, 2019 2 commits
-
-
Gal Pressman authored
Cited commit introduced the udata parameter to different destroy flows but the uapi method definition does not have udata (i.e has_udata flag is not set). As a result, an uninitialized udata struct is being passed down to the driver callbacks. Fix that by clearing the driver udata even in cases where has_udata flag is not set. Fixes: c4367a26 ("IB: Pass uverbs_attr_bundle down ib_x destroy path") Cc: Shamir Rabinovitch <shamir.rabinovitch@oracle.com> Co-developed-by:
-
Shiraz Saleem authored
The flag update_cur_sg tracks whether contiguous pages from a new set of page_list pages can be merged into the SGE passed into ib_umem_add_sg_table(). If this flag is true, but the total segment length exceeds the max_seg_size supported by HW, we avoid combining to this SGE and move to a new SGE (x) and merge 'len' pages to it. However, if i < npages, the next iteration can incorrectly merge 'len' contiguous pages into x instead of into a new SGE since update_cur_sg is still true. Reset update_cur_sg to false always after the check to merge pages into the first SGE passed in to ib_umem_add_sg_table(). Also, prevent a new SGE's segment length from ever exceeding HW max_seg_sz. There is a crash on hfi1 as result of this where-in max_seg_sz is defaulting to 64K. Due to above bug, unfolding SGE's in __ib_umem_release points to a bad page ptr. TEST comp-wfr.perfnative.STL-22166-WDT _ perftest native 2-Write_4097QP_4MB STARTING at 1555387093 BUG: Bad page state in process ib_write_bw pfn:7ebca0 page:ffffcd675faf2800 count:0 mapcount:1 mapping:0000000000000000 index:0x1 flags: 0x17ffffc0000000() raw: 0017ffffc0000000 dead000000000100 dead000000000200 0000000000000000 raw: 0000000000000001 0000000000000000 0000000000000000 0000000000000000 page dumped because: nonzero mapcount CPU: 18 PID: 15853 Comm: ib_write_bw Tainted: G B 5.1.0-rc4 #1 Hardware name: Intel Corporation S2600CWR/S2600CW, BIOS SE5C610.86B.01.01.0014.121820151719 12/18/2015 Call Trace: dump_stack+0x5a/0x73 bad_page+0xf5/0x10f free_pcppages_bulk+0x62c/0x680 free_unref_page+0x54/0x70 __ib_umem_release+0x148/0x1a0 [ib_uverbs] ib_umem_release+0x22/0x80 [ib_uverbs] rvt_dereg_mr+0x67/0xb0 [rdmavt] ib_dereg_mr_user+0x37/0x60 [ib_core] destroy_hw_idr_uobject+0x1c/0x50 [ib_uverbs] uverbs_destroy_uobject+0x2e/0x180 [ib_uverbs] uobj_destroy+0x4d/0x60 [ib_uverbs] __uobj_get_destroy+0x33/0x50 [ib_uverbs] __uobj_perform_destroy+0xa/0x30 [ib_uverbs] ib_uverbs_dereg_mr+0x66/0x90 [ib_uverbs] ib_uverbs_write+0x3e1/0x500 [ib_uverbs] vfs_write+0xad/0x1b0 ksys_write+0x5a/0xd0 do_syscall_64+0x5b/0x180 entry_SYSCALL_64_after_hwframe+0x44/0xa9 Fixes: d10bcf94 ("RDMA/umem: Combine contiguous PAGE_SIZE regions in SGEs") Tested-by:
-
