summaryrefslogtreecommitdiffstats
path: root/drivers/infiniband/hw/i40iw
Commit message (Collapse)AuthorAgeFilesLines
* Merge tag 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdmaLinus Torvalds2018-12-282-73/+56
|\ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Pull rdma updates from Jason Gunthorpe: "This has been a fairly typical cycle, with the usual sorts of driver updates. Several series continue to come through which improve and modernize various parts of the core code, and we finally are starting to get the uAPI command interface cleaned up. - Various driver fixes for bnxt_re, cxgb3/4, hfi1, hns, i40iw, mlx4, mlx5, qib, rxe, usnic - Rework the entire syscall flow for uverbs to be able to run over ioctl(). Finally getting past the historic bad choice to use write() for command execution - More functional coverage with the mlx5 'devx' user API - Start of the HFI1 series for 'TID RDMA' - SRQ support in the hns driver - Support for new IBTA defined 2x lane widths - A big series to consolidate all the driver function pointers into a big struct and have drivers provide a 'static const' version of the struct instead of open coding initialization - New 'advise_mr' uAPI to control device caching/loading of page tables - Support for inline data in SRPT - Modernize how umad uses the driver core and creates cdev's and sysfs files - First steps toward removing 'uobject' from the view of the drivers" * tag 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma: (193 commits) RDMA/srpt: Use kmem_cache_free() instead of kfree() RDMA/mlx5: Signedness bug in UVERBS_HANDLER() IB/uverbs: Signedness bug in UVERBS_HANDLER() IB/mlx5: Allocate the per-port Q counter shared when DEVX is supported IB/umad: Start using dev_groups of class IB/umad: Use class_groups and let core create class file IB/umad: Refactor code to use cdev_device_add() IB/umad: Avoid destroying device while it is accessed IB/umad: Simplify and avoid dynamic allocation of class IB/mlx5: Fix wrong error unwind IB/mlx4: Remove set but not used variable 'pd' RDMA/iwcm: Don't copy past the end of dev_name() string IB/mlx5: Fix long EEH recover time with NVMe offloads IB/mlx5: Simplify netdev unbinding IB/core: Move query port to ioctl RDMA/nldev: Expose port_cap_flags2 IB/core: uverbs copy to struct or zero helper IB/rxe: Reuse code which sets port state IB/rxe: Make counters thread safe IB/mlx5: Use the correct commands for UMEM and UCTX allocation ...
| * RDMA: Cleanup undesired pd->uobject usageShamir Rabinovitch2018-12-181-23/+22
| | | | | | | | | | | | | | | | | | | | | | | | Drivers should be using udata to determine if a method is invoked from user space or kernel space. A pd does not necessarily say a different objects is kernel or user. Transforming the tests to use udata eliminates a large number of uobject references from the drivers. Signed-off-by: Shamir Rabinovitch <shamir.rabinovitch@oracle.com> Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
| * RDMA/i40iw: Make sure to initialize ib_device_opsKamal Heib2018-12-181-0/+1
| | | | | | | | | | | | | | | | | | The initialization of the ib_device_ops was dropped by mistake when rebasing the ib_device_ops series, this patch fixes that. Fixes: 15644f57cb66 ("RDMA/i40iw: Initialize ib_device_ops struct") Signed-off-by: Kamal Heib <kamalheib1@gmail.com> Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
| * RDMA: Start use ib_device_opsKamal Heib2018-12-121-1/+1
| | | | | | | | | | | | | | Make all the required change to start use the ib_device_ops structure. Signed-off-by: Kamal Heib <kamalheib1@gmail.com> Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
| * RDMA/i40iw: Initialize ib_device_ops structKamal Heib2018-12-111-30/+33
| | | | | | | | | | | | | | | | Initialize ib_device_ops with the supported operations using ib_set_device_ops(). Signed-off-by: Kamal Heib <kamalheib1@gmail.com> Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
| * i40iw: remove support for ib_get_vector_affinitySagi Grimberg2018-11-081-20/+0
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Devices that does not use managed affinity can not export a vector affinity as the consumer relies on having a static mapping it can map to upper layer affinity (e.g. sw queues). If the driver allows the user to set the device irq affinity, then the affinitization of a long term existing entites is not relevant. For example, nvme-rdma controllers queue-irq affinitization is determined at init time so if the irq affinity changes over time, we are no longer aligned. Signed-off-by: Sagi Grimberg <sagi@grimberg.me> Acked-by: Shiraz Saleem <shiraz.saleem@intel.com> Signed-off-by: Doug Ledford <dledford@redhat.com> Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
* | i40iw: remove use of VLAN_TAG_PRESENTMichał Mirosław2018-11-081-4/+4
|/ | | | | Signed-off-by: Michał Mirosław <mirq-linux@rere.qmqm.pl> Signed-off-by: David S. Miller <davem@davemloft.net>
* RDMA/drivers: Use core provided API for registering device attributesParav Pandit2018-10-171-47/+23
| | | | | | | | | Use rdma_set_device_sysfs_group() to register device attributes and simplify the driver. Signed-off-by: Parav Pandit <parav@mellanox.com> Signed-off-by: Leon Romanovsky <leonro@mellanox.com> Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
* RDMA: Fully setup the device name in ib_register_deviceJason Gunthorpe2018-09-261-2/+1
| | | | | | | | | | | | | | | | | | | | | | | The current code has two copies of the device name, ibdev->dev and dev_name(&ibdev->dev), and they are setup at different times, which is very confusing. Set them both up at the same time and make dev_name() the lead name, which is the proper use of the driver core APIs. To make it very clear that the name is not valid until registration pass it in to the ib_register_device() call rather than messing with ibdev->name directly. Also the reorganization now checks that dev_name is unique even if it does not contain a %. Signed-off-by: Jason Gunthorpe <jgg@mellanox.com> Acked-by: Adit Ranadive <aditr@vmware.com> Reviewed-by: Steve Wise <swise@opengridcomputing.com> Acked-by: Devesh Sharma <devesh.sharma@broadcom.com> Reviewed-by: Shiraz Saleem <shiraz.saleem@intel.com> Reviewed-by: Leon Romanovsky <leonro@mellanox.com> Reviewed-by: Dennis Dalessandro <dennis.dalessandro@intel.com> Reviewed-by: Michael J. Ruhl <michael.j.ruhl@intel.com>
* RDMA/i40iw: Fix incorrect iterator typeHåkon Bugge2018-09-191-1/+1
| | | | | | | | | | Commit f27b4746f378 ("i40iw: add connection management code") uses an incorrect rcu iterator, whilst holding the rtnl_lock. Since the critical region invokes i40iw_manage_qhash(), which is a sleeping function, the rcu locking and traversal cannot be used. Signed-off-by: Håkon Bugge <haakon.bugge@oracle.com> Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
* RDMA/providers: Remove pointless functionsKamal Heib2018-07-301-40/+0
| | | | | | | | | The rdma core is taking care of return the right error code when the rdma device callbacks aren't supported. Signed-off-by: Kamal Heib <kamalheib1@gmail.com> Acked-by: Shiraz Saleem <shiraz.saleem@intel.com> Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
* RDMA, core and ULPs: Declare ib_post_send() and ib_post_recv() arguments constBart Van Assche2018-07-301-5/+4
| | | | | | | | | | | | | | | | | | | | | | Since neither ib_post_send() nor ib_post_recv() modify the data structure their second argument points at, declare that argument const. This change makes it necessary to declare the 'bad_wr' argument const too and also to modify all ULPs that call ib_post_send(), ib_post_recv() or ib_post_srq_recv(). This patch does not change any functionality but makes it possible for the compiler to verify whether the ib_post_(send|recv|srq_recv) really do not modify the posted work request. To make this possible, only one cast had to be introduce that casts away constness, namely in rpcrdma_post_recvs(). The only way I can think of to avoid that cast is to introduce an additional loop in that function or to change the data type of bad_wr from struct ib_recv_wr ** into int (an index that refers to an element in the work request list). However, both approaches would require even more extensive changes than this patch. Signed-off-by: Bart Van Assche <bart.vanassche@wdc.com> Reviewed-by: Chuck Lever <chuck.lever@oracle.com> Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
* infiniband: i40iw, nes: don't use wall time for TCP sequence numbersArnd Bergmann2018-07-112-5/+22
| | | | | | | | | | | | | | | | | | | | | | | | | | | | The nes infiniband driver uses current_kernel_time() to get a nanosecond granunarity timestamp to initialize its tcp sequence counters. This is one of only a few remaining users of that deprecated function, so we should try to get rid of it. Aside from using a deprecated API, there are several problems I see here: - Using a CLOCK_REALTIME based time source makes it predictable in case the time base is synchronized. - Using a coarse timestamp means it only gets updated once per jiffie, making it even more predictable in order to avoid having to access the hardware clock source - The upper 2 bits are always zero because the nanoseconds are at most 999999999. For the Linux TCP implementation, we use secure_tcp_seq(), which appears to be appropriate here as well, and solves all the above problems. i40iw uses a variant of the same code, so I do that same thing there for ipv4. Unlike nes, i40e also supports ipv6, which needs to call secure_tcpv6_seq instead. Acked-by: Shiraz Saleem <shiraz.saleem@intel.com> Signed-off-by: Arnd Bergmann <arnd@arndb.de> Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
* RDMA/i40w: Hold read semaphore while looking after VMALeon Romanovsky2018-07-041-0/+2
| | | | | | | | VMA lookup is supposed to be performed while mmap_sem is held. Fixes: f26c7c83395b ("i40iw: Add 2MB page support") Signed-off-by: Leon Romanovsky <leonro@mellanox.com> Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
* IB/core: add max_send_sge and max_recv_sge attributesSteve Wise2018-06-181-1/+2
| | | | | | | | | | | | | | | | | | | This patch replaces the ib_device_attr.max_sge with max_send_sge and max_recv_sge. It allows ulps to take advantage of devices that have very different send and recv sge depths. For example cxgb4 has a max_recv_sge of 4, yet a max_send_sge of 16. Splitting out these attributes allows much more efficient use of the SQ for cxgb4 with ulps that use the RDMA_RW API. Consider a large RDMA WRITE that has 16 scattergather entries. With max_sge of 4, the ulp would send 4 WRITE WRs, but with max_sge of 16, it can be done with 1 WRITE WR. Acked-by: Sagi Grimberg <sagi@grimberg.me> Acked-by: Christoph Hellwig <hch@lst.de> Acked-by: Selvin Xavier <selvin.xavier@broadcom.com> Acked-by: Shiraz Saleem <shiraz.saleem@intel.com> Acked-by: Dennis Dalessandro <dennis.dalessandro@intel.com> Signed-off-by: Steve Wise <swise@opengridcomputing.com> Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
* i40iw: Reorganize acquire/release of locks in i40iw_manage_apbvtShiraz Saleem2018-06-181-32/+51
| | | | | | | | | | | | | | | | | | Commit f43c00c04bbf ("i40iw: Extend port reuse support for listeners") introduces a sparse warning: include/linux/spinlock.h:365:9: sparse: context imbalance in 'i40iw_manage_apbvt' - unexpected unlock Fix this by reorganizing the acquire/release of locks in i40iw_manage_apbvt and add a new function i40iw_cqp_manage_abvpt_cmd to perform the CQP command. Also, use __clear_bit and __test_and_set_bit as we do not need atomic versions. Fixes: f43c00c04bbf ("i40iw: Extend port reuse support for listeners") Suggested-by: Jason Gunthorpe <jgg@mellanox.com> Signed-off-by: Shiraz Saleem <shiraz.saleem@intel.com> Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
* Merge branch 'mr_fix' into ↵Jason Gunthorpe2018-05-286-10/+18
|\ | | | | | | | | | | | | | | | | | | | | | | | | git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma for-next Update mlx4 to support user MR creation against read-only memory, previously it required the memory to be writable. Based on rdma for-rc due to dependencies. * mr_fix: (2 commits) IB/mlx4: Mark user MR as writable if actual virtual memory is writable IB/core: Make testing MR flags for writability a static inline function
| * RDMA/i40iw: Avoid panic when reading back the IRQ affinity hintAndrew Boyer2018-05-092-4/+4
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The current code sets an affinity hint with a cpumask_t stored on the stack. This value can then be accessed through /proc/irq/*/affinity_hint/, causing a segfault or returning corrupt data. Move the cpumask_t into struct i40iw_msix_vector so it is available later. Backtrace: BUG: unable to handle kernel paging request at ffffb16e600e7c90 IP: irq_affinity_hint_proc_show+0x60/0xf0 PGD 17c0c6d067 PUD 17c0c6e067 PMD 15d4a0e067 PTE 0 Oops: 0000 [#1] SMP Modules linked in: ... CPU: 3 PID: 172543 Comm: grep Tainted: G OE ... #1 Hardware name: ... task: ffff9a5caee08000 task.stack: ffffb16e659d8000 RIP: 0010:irq_affinity_hint_proc_show+0x60/0xf0 RSP: 0018:ffffb16e659dbd20 EFLAGS: 00010086 RAX: 0000000000000246 RBX: ffffb16e659dbd20 RCX: 0000000000000000 RDX: ffffb16e600e7c90 RSI: 0000000000000003 RDI: 0000000000000046 RBP: ffffb16e659dbd88 R08: 0000000000000038 R09: 0000000000000001 R10: 0000000070803079 R11: 0000000000000000 R12: ffff9a59d1d97a00 R13: ffff9a5da47a6cd8 R14: ffff9a5da47a6c00 R15: ffff9a59d1d97a00 FS: 00007f946c31d740(0000) GS:ffff9a5dc1800000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: ffffb16e600e7c90 CR3: 00000016a4339000 CR4: 00000000007406e0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 PKRU: 55555554 Call Trace: seq_read+0x12d/0x430 ? sched_clock_cpu+0x11/0xb0 proc_reg_read+0x48/0x70 __vfs_read+0x37/0x140 ? security_file_permission+0xa0/0xc0 vfs_read+0x96/0x140 SyS_read+0x58/0xc0 do_syscall_64+0x5a/0x190 entry_SYSCALL64_slow_path+0x25/0x25 RIP: 0033:0x7f946bbc97e0 RSP: 002b:00007ffdd0c4ae08 EFLAGS: 00000246 ORIG_RAX: 0000000000000000 RAX: ffffffffffffffda RBX: 000000000096b000 RCX: 00007f946bbc97e0 RDX: 000000000096b000 RSI: 00007f946a2f0000 RDI: 0000000000000004 RBP: 0000000000001000 R08: 00007f946a2ef011 R09: 000000000000000a R10: 0000000000001000 R11: 0000000000000246 R12: 00007f946a2f0000 R13: 0000000000000004 R14: 0000000000000000 R15: 00007f946a2f0000 Code: b9 08 00 00 00 49 89 c6 48 89 df 31 c0 4d 8d ae d8 00 00 00 f3 48 ab 4c 89 ef e8 6c 9a 56 00 49 8b 96 30 01 00 00 48 85 d2 74 3f <48> 8b 0a 48 89 4d 98 48 8b 4a 08 48 89 4d a0 48 8b 4a 10 48 89 RIP: irq_affinity_hint_proc_show+0x60/0xf0 RSP: ffffb16e659dbd20 CR2: ffffb16e600e7c90 Fixes: 8e06af711bf2 ("i40iw: add main, hdr, status") Signed-off-by: Andrew Boyer <andrew.boyer@dell.com> Reviewed-by: Shiraz Saleem <shiraz.saleem@intel.com> Signed-off-by: Doug Ledford <dledford@redhat.com>
| * RDMA/i40iw: Avoid reference leaks when processing the AEQAndrew Boyer2018-05-091-2/+2
| | | | | | | | | | | | | | | | | | | | | | | | In this switch there is a reference held on the QP. 'continue' will grab the next event without releasing the reference, causing a leak. Change it to 'break' to drop the reference before grabbing the next event. Fixes: 4e9042e647ff ("i40iw: add hw and utils files") Signed-off-by: Andrew Boyer <andrew.boyer@dell.com> Reviewed-by: Shiraz Saleem <shiraz.saleem@intel.com> Signed-off-by: Doug Ledford <dledford@redhat.com>
| * RDMA/i40iw: Avoid panic when objects are being created and destroyedAndrew Boyer2018-05-092-2/+10
| | | | | | | | | | | | | | | | | | | | | | | | | | | | A panic occurs when there is a newly-registered element on the QP/CQ MR list waiting to be attached, but a different MR is deregistered. The current code only checks for whether the list is empty, not whether the element being deregistered is actually on the list. Fix the panic by adding a boolean to track if the object is on the list. Fixes: d37498417947 ("i40iw: add files for iwarp interface") Signed-off-by: Andrew Boyer <andrew.boyer@dell.com> Reviewed-by: Shiraz Saleem <shiraz.saleem@intel.com> Signed-off-by: Doug Ledford <dledford@redhat.com>
| * i40iw: Use correct address in dst_neigh_lookup for IPv6Mustafa Ismail2018-05-091-1/+1
| | | | | | | | | | | | | | | | | | | | | | Use of incorrect structure address for IPv6 neighbor lookup causes connections to IPv6 addresses to fail. Fix this by using correct address in call to dst_neigh_lookup. Fixes: f27b4746f378 ("i40iw: add connection management code") Signed-off-by: Mustafa Ismail <mustafa.ismail@intel.com> Signed-off-by: Shiraz Saleem <shiraz.saleem@intel.com> Signed-off-by: Doug Ledford <dledford@redhat.com>
| * i40iw: Fix memory leak in error path of create QPMustafa Ismail2018-05-091-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | If i40iw_allocate_dma_mem fails when creating a QP, the memory allocated for the QP structure using kzalloc is not freed because iwqp->allocated_buffer is used to free the memory and it is not setup until later. Fix this by setting iwqp->allocated_buffer before allocating the dma memory. Fixes: d37498417947 ("i40iw: add files for iwarp interface") Signed-off-by: Mustafa Ismail <mustafa.ismail@intel.com> Signed-off-by: Shiraz Saleem <shiraz.saleem@intel.com> Signed-off-by: Doug Ledford <dledford@redhat.com>
* | i40iw: Extend port reuse support for listenersShiraz Saleem2018-05-163-42/+59
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | If two listeners are created with different IP's but same port, the second rdma_listen fails due to a duplicate port entry being added from the CQP add APBVT OP. commit f16dc0aa5ea2 ("i40iw: Add support for port reuse on active side connections") does not account for listener side port reuse. Check for duplicate port before invoking the CQP command to add APBVT entry and delete the entry only if the port is not in use. Additionally, consolidate all port-reuse logic into i40iw_manage_apbvt. Fixes: f16dc0aa5ea2 ("i40iw: Add support for port reuse on active side connections") Signed-off-by: Shiraz Saleem <shiraz.saleem@intel.com> Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
* | infiniband: i40iw: Replace GFP_ATOMIC with GFP_KERNEL in i40iw_l2param_changeJia-Ju Bai2018-04-171-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | i40iw_l2param_change() is never called in atomic context. i40iw_make_listen_node() is only set as ".l2_param_change" in struct i40e_client_ops, and this function pointer is not called in atomic context. Despite never getting called from atomic context, i40iw_l2param_change() calls kzalloc() with GFP_ATOMIC, which does not sleep for allocation. GFP_ATOMIC is not necessary and can be replaced with GFP_KERNEL, which can sleep and improve the possibility of sucessful allocation. This is found by a static analysis tool named DCNS written by myself. And I also manually check it. Signed-off-by: Jia-Ju Bai <baijiaju1990@gmail.com> Acked-by: Shiraz Saleem <shiraz.saleem@intel.com> Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
* | infiniband: i40iw: Replace GFP_ATOMIC with GFP_KERNEL in i40iw_make_listen_nodeJia-Ju Bai2018-04-171-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | i40iw_make_listen_node() is never called in atomic context. i40iw_make_listen_node() is only called by i40iw_create_listen, which is set as ".create_listen" in struct iw_cm_verbs. Despite never getting called from atomic context, i40iw_make_listen_node() calls kzalloc() with GFP_ATOMIC, which does not sleep for allocation. GFP_ATOMIC is not necessary and can be replaced with GFP_KERNEL, which can sleep and improve the possibility of sucessful allocation. This is found by a static analysis tool named DCNS written by myself. And I also manually check it. Signed-off-by: Jia-Ju Bai <baijiaju1990@gmail.com> Acked-by: Shiraz Saleem <shiraz.saleem@intel.com> Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
* | infiniband: i40iw: Replace GFP_ATOMIC with GFP_KERNEL in i40iw_add_mqh_4Jia-Ju Bai2018-04-171-1/+1
|/ | | | | | | | | | | | | | | | | | i40iw_add_mqh_4() is never called in atomic context, because it calls rtnl_lock() that can sleep. Despite never getting called from atomic context, i40iw_add_mqh_4() calls kzalloc() with GFP_ATOMIC, which does not sleep for allocation. GFP_ATOMIC is not necessary and can be replaced with GFP_KERNEL, which can sleep and improve the possibility of sucessful allocation. This is found by a static analysis tool named DCNS written by myself. And I also manually check it. Signed-off-by: Jia-Ju Bai <baijiaju1990@gmail.com> Acked-by: Shiraz Saleem <shiraz.saleem@intel.com> Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
* i40iw: Remove pre-production workaround for resource profile 1Shiraz Saleem2018-04-031-2/+0
| | | | | | | | | Support for resource profile 1 is currenlty deprecated due to a pre-production errata. Remove this workaround as its no longer needed. Signed-off-by: Shiraz Saleem <shiraz.saleem@intel.com> Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
* IB/uverbs: Extend uverbs_ioctl header with driver_idMatan Barak2018-03-191-0/+1
| | | | | | | | | | | | | | 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>
* i40iw: Tear-down connection after CQP Modify QP failureHenry Orosco2018-03-154-10/+30
| | | | | | | | | | | | | | | | | There is no explicit tear-down sequence initiated on connections if the Control QP OP, Modify QP to close, fails. Fix this by triggering a driver generated Asynchronous Event (AE) on Modify QP failures and tear-down the connection on receipt of the AE. This fix can be generalized to other Modify QP failures (i.e. RTS->TERM, IDLE->RTS, etc) as any modify failure will require a connection tear-down. Fixes: d37498417947 ("i40iw: add files for iwarp interface") Signed-off-by: Henry Orosco <henry.orosco@intel.com> Signed-off-by: Shiraz Saleem <shiraz.saleem@intel.com> Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
* i40iw: Refactor of driver generated AEsHenry Orosco2018-03-156-10/+108
| | | | | | | | | | | | The flush CQP OP can be used to optionally generate Asynchronous Events (AEs) in addition to QP flush. Consolidate all HW AE generation code under a new function i40iw_gen_ae which use the flush CQP OP to only generate AEs. Signed-off-by: Henry Orosco <henry.orosco@intel.com> Signed-off-by: Shiraz Saleem <shiraz.saleem@intel.com> Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
* RDMA/i40iw: Move uapi header to include/uapiJason Gunthorpe2018-03-152-108/+1
| | | | | | | | All of these defines are part of the uABI for the driver, this header duplicates providers/i40iw/i40iw-abi.h in rdma-core. Acked-by: Shiraz Saleem <shiraz.saleem@intel.com> Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
* RDMA/i40iw: include linux/irq.hArnd Bergmann2018-03-141-0/+1
| | | | | | | | | | | | | | | | We get a build failure on ARM unless the header is included explicitly: drivers/infiniband/hw/i40iw/i40iw_verbs.c: In function 'i40iw_get_vector_affinity': drivers/infiniband/hw/i40iw/i40iw_verbs.c:2747:9: error: implicit declaration of function 'irq_get_affinity_mask'; did you mean 'irq_create_affinity_masks'? [-Werror=implicit-function-declaration] return irq_get_affinity_mask(msix_vec->irq); ^~~~~~~~~~~~~~~~~~~~~ irq_create_affinity_masks drivers/infiniband/hw/i40iw/i40iw_verbs.c:2747:9: error: returning 'int' from a function with return type 'const struct cpumask *' makes pointer from integer without a cast [-Werror=int-conversion] return irq_get_affinity_mask(msix_vec->irq); Fixes: 7e952b19eb63 ("i40iw: Implement get_vector_affinity API") Signed-off-by: Arnd Bergmann <arnd@arndb.de> Signed-off-by: Doug Ledford <dledford@redhat.com>
* i40iw: Implement get_vector_affinity APIShiraz Saleem2018-03-061-0/+20
| | | | | | | | | | | | | | Storage ULPs (like NVMEoF) benefit from exposing affinity mapping per completion vector to find the optimal multi-queue affinity assignments. The ULPs call the verbs API ib_get_vector_affinity introduced in commit c66cd353bbe ("RDMA/core: expose affinity mappings per completion vector") to get the underlying devices affinity mappings. Add support in driver to expose the affinity masks per MSI-X completion vector. Signed-off-by: Shiraz Saleem <shiraz.saleem@intel.com> Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
* i40iw: Improve CM node lookup time on connection setupShiraz Saleem2018-03-064-37/+100
| | | | | | | | | | | | | | | | | | | | | Currently all CM nodes involved in a connection are maintained in a connected_node list per dev. During connection setup, we need to search this every time we receive a packet on the iWARP LAN Queue (ILQ) and this can be pretty inefficient for large number of connections. Fix this by organizing the CM nodes in two lists - accelerated list and non-accelerated list. The search on ILQ receive would be limited to only non accelerated nodes. When a node moves to RTS, it is added to the accelerated list. Benchmarking ucmatose 16k connections shows a 20% improvement in test completion time. Signed-off-by: Shiraz Saleem <shiraz.saleem@intel.com> Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
* i40iw: Refactor handling of txpend listMustafa Ismail2018-03-061-6/+2
| | | | | | | | | | Currently the TX pending lists for IEQ and ILQ are handled separately. The handling of both can be consolidated in i40iw_poll_completion. Signed-off-by: Mustafa Ismail <mustafa.ismail@intel.com> Signed-off-by: Shiraz Saleem <shiraz.saleem@intel.com> Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
* i40iw: Free IEQ resourcesMustafa Ismail2018-01-163-2/+3
| | | | | | | | | | The iWARP Exception Queue (IEQ) resources are not freed when a QP is destroyed. Fix this by freeing IEQ resources when freeing QP resources. Fixes: d37498417947 ("i40iw: add files for iwarp interface") Signed-off-by: Mustafa Ismail <mustafa.ismail@intel.com> Signed-off-by: Shiraz Saleem <shiraz.saleem@intel.com> Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
* i40iw: Remove setting of rem_addr.lenMustafa Ismail2018-01-161-3/+0
| | | | | | | | | | Remove setting of rem_addr.len before calling iw_rdma_write, iw_inline_rdma_write and rdma_read. rem_addr.len is not used in those functions. Signed-off-by: Mustafa Ismail <mustafa.ismail@intel.com> Signed-off-by: Shiraz Saleem <shiraz.saleem@intel.com> Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
* i40iw: Remove limit on re-posting AEQ entries to HWSindhu Devale2018-01-162-3/+0
| | | | | | | | | | | | | | | Currently, if the number of processed Asynchronous Event Queue (AEQ) entries exceeds 255, they are not returned to HW for re-use. During scale-up, the unreturned AEQ entries can grow to the max AEQ size and cause the HW to report an AEQ overflow. Remove the check which limits the number of processed AEQ entries returned to HW. Fixes: 86dbcd0f12e9 ("RDMA/i40iw: add file to handle cqp calls") Signed-off-by: Sindhu Devale <sindhu.devale@intel.com> Signed-off-by: Shiraz Saleem <shiraz.saleem@intel.com> Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
* i40iw: Zero-out consumer key on allocate stag for FMRShiraz Saleem2018-01-161-0/+1
| | | | | | | | | | | | | If the application invalidates the MR before the FMR WR, HW parses the consumer key portion of the stag and returns an invalid stag key Asynchronous Event (AE) that tears down the QP. Fix this by zeroing-out the consumer key portion of the allocated stag returned to application for FMR. Fixes: ee855d3b93f3 ("RDMA/i40iw: Add base memory management extensions") Signed-off-by: Shiraz Saleem <shiraz.saleem@intel.com> Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
* i40iw: Remove extra call to i40iw_est_sd()Shiraz Saleem2018-01-161-2/+0
| | | | | | | | Remove redundant estimate SD function call. sd_needed should already be updated at the end of the do while resource reduction loop. Signed-off-by: Shiraz Saleem <shiraz.saleem@intel.com> Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
* i40iw: Replace mdelay with msleep in i40iw_wait_pe_readyJia-Ju Bai2018-01-051-1/+1
| | | | | | | | | | | | i40iw_wait_pe_ready is not called in an interrupt handler nor holding a spinlock. The function mdelay in it can be replaced with msleep, to reduce busy wait. Signed-off-by: Jia-Ju Bai <baijiaju1990@gmail.com> Reviewed-by: Leon Romanovsky <leonro@mellanox.com> Acked-by: Shiraz Saleem <shiraz.saleem@intel.com> Signed-off-by: Doug Ledford <dledford@redhat.com>
* Merge branch 'from-rc' of ↵Jason Gunthorpe2017-12-274-28/+56
|\ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma.git Patches for 4.16 that are dependent on patches sent to 4.15-rc. These are small clean ups for the vmw_pvrdma and i40iw drivers. * 'from-rc' of git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma.git: RDMA/vmw_pvrdma: Remove usage of BIT() from UAPI header RDMA/vmw_pvrdma: Use refcount_t instead of atomic_t RDMA/vmw_pvrdma: Use more specific sizeof in kcalloc RDMA/vmw_pvrdma: Clarify QP and CQ is_kernel logic RDMA/vmw_pvrdma: Add UAR SRQ macros in ABI header file i40iw: Change accelerated flag to bool
| * i40iw: Change accelerated flag to boolHenry Orosco2017-12-272-3/+3
| | | | | | | | | | | | | | | | | | | | The accelerated flag only utilizes two values: 0 and 1. Modify accelerated flag in struct i40iw_cm_node to bool. Signed-off-by: Henry Orosco <henry.orosco@intel.com> Signed-off-by: Shiraz Saleem <shiraz.saleem@intel.com> Reviewed-by: Yuval Shaia <yuval.shaia@oracle.com> Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
| * i40iw: Notify user of established connection after QP in RTSHenry Orosco2017-12-011-6/+11
| | | | | | | | | | | | | | | | | | | | | | | | Established CM event is sent prior to modifying QP to RTS state. This can result in application closing the connection before the QP is actually in RTS state. Move sending of established CM event to after modify QP to RTS. Fixes: f27b4746f378 ("i40iw: add connection management code") Signed-off-by: Henry Orosco <henry.orosco@intel.com> Signed-off-by: Shiraz Saleem <shiraz.saleem@intel.com> Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
| * i40iw: Move MPA request event for loopback after connectTatyana Nikolova2017-12-011-2/+6
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | For loopback, a MPA request event is generated when cm_node is initialized, which allows applications to act on the connect request before i40iw_connect() has completed. In some cases, the reject flow executes in parallel with the connect flow and doesn't delete an APBVT entry, because the apbvt_set variable is still not set by the connect flow. Move the MPA request event to the end of i40iw_connect() to notify application for a connect request, after connect has completed. Fixes: f27b4746f378 ("i40iw: add connection management code") Signed-off-by: Tatyana Nikolova <tatyana.e.nikolova@intel.com> Signed-off-by: Henry Orosco <henry.orosco@intel.com> Signed-off-by: Shiraz Saleem <shiraz.saleem@intel.com> Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
| * i40iw: Correct ARP index maskMustafa Ismail2017-12-011-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | The ARP table entry indexes are aliased to 12bits instead of the intended 16bits when uploaded to the QP Context. This will present an issue when the number of connections exceeds 4096 as ARP entries are reused. Fix this by adjusting the mask to account for the full 16bits. Fixes: 4e9042e647ff ("i40iw: add hw and utils files") Signed-off-by: Mustafa Ismail <mustafa.ismail@intel.com> Signed-off-by: Shiraz Saleem <shiraz.saleem@intel.com> Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
| * i40iw: Do not free sqbuf when event is I40IW_TIMER_TYPE_CLOSEMustafa Ismail2017-12-011-3/+3
| | | | | | | | | | | | | | | | | | | | When the event type is I40IW_TIMER_TYPE_CLOSE, there is no sqbuf and it should not be freed as one in i40iw_schedule_cm_timer(). Fixes: f27b4746f378 ("i40iw: add connection management code") Signed-off-by: Mustafa Ismail <mustafa.ismail@intel.com> Signed-off-by: Shiraz Saleem <shiraz.saleem@intel.com> Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
| * i40iw: Allocate a sdbuf per CQP WQEChien Tin Tung2017-12-012-14/+33
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Currently there is only one sdbuf per Control QP (CQP) for programming Segment Descriptor (SD). If multiple SD work requests are posted simultaneously, the sdbuf is reused by all WQEs and new WQEs can corrupt previous WQEs sdbuf leading to incorrect SD programming. Fix this by allocating one sdbuf per CQP SQ WQE. When an SD command is posted, it will use the corresponding sdbuf for the WQE. Fixes: 86dbcd0f12e9 ("i40iw: add file to handle cqp calls") Signed-off-by: Chien Tin Tung <chien.tin.tung@intel.com> Signed-off-by: Shiraz Saleem <shiraz.saleem@intel.com> Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
* | i40iw: Fix the connection ORD value for loopbackTatyana Nikolova2017-12-221-12/+14
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The accepting QP ORD value should be adjusted not to exceed the peer QP IRD value (RFC 6581). This is skipped for loopback. After the ORD is validated by i40iw_record_ird_ord(), adjust the ORD value of the loopback accepting QP to prevent overrunning the IRD space of the peer QP. Also move the ORD accounting for 0-byte RDMA read to i40iw_record_ird_ord(). Fixes: f27b4746f378 ("i40iw: add connection management code") Signed-off-by: Tatyana Nikolova <tatyana.e.nikolova@intel.com> Signed-off-by: Shiraz Saleem <shiraz.saleem@intel.com> Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
* | i40iw: Validate correct IRD/ORD connection parametersTatyana Nikolova2017-12-221-2/+3
| | | | | | | | | | | | | | | | | | | | | | | | | | Casting to u16 before validating IRD/ORD connection parameters could cause recording wrong IRD/ORD values in the cm_node. Validate the IRD/ORD parameters as they are passed by the application before recording them. Fixes: f27b4746f378 ("i40iw: add connection management code") Signed-off-by: Tatyana Nikolova <tatyana.e.nikolova@intel.com> Signed-off-by: Shiraz Saleem <shiraz.saleem@intel.com> Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
OpenPOWER on IntegriCloud