summaryrefslogtreecommitdiff
AgeCommit message (Collapse)AuthorFilesLines
2021-06-30RDMA/core: Always release restrack objectLeon Romanovsky1-1/+1
Change location of rdma_restrack_del() to fix the bug where task_struct was acquired but not released, causing to resource leak. ucma_create_id() { ucma_alloc_ctx(); rdma_create_user_id() { rdma_restrack_new(); rdma_restrack_set_name() { rdma_restrack_attach_task.part.0(); <--- task_struct was gotten } } ucma_destroy_private_ctx() { ucma_put_ctx(); rdma_destroy_id() { _destroy_id() <--- id_priv was freed } } } Fixes: 889d916b6f8a ("RDMA/core: Don't access cm_id after its destruction") Link: https://lore.kernel.org/r/073ec27acb943ca8b6961663c47c5abe78a5c8cc.1624948948.git.leonro@nvidia.com Reported-by: Pavel Skripkin <paskripkin@gmail.com> Signed-off-by: Leon Romanovsky <leonro@nvidia.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-06-30RDMA/mlx5: Don't access NULL-cleared mpi pointerLeon Romanovsky1-1/+2
The "dev->port[i].mp.mpi" is set to NULL during mlx5_ib_unbind_slave_port() execution, however that field is needed to add device to unaffiliated list. Such flow causes to the following kernel panic while unloading mlx5_ib module in multi-port mode, hence the device should be added to the list prior to unbind call. RPC: Unregistered rdma transport module. RPC: Unregistered rdma backchannel transport module. BUG: kernel NULL pointer dereference, address: 0000000000000000 #PF: supervisor write access in kernel mode #PF: error_code(0x0002) - not-present page PGD 0 P4D 0 Oops: 0002 [#1] SMP NOPTI CPU: 4 PID: 1904 Comm: modprobe Not tainted 5.13.0-rc7_for_upstream_min_debug_2021_06_24_12_08 #1 Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS rel-1.13.0-0-gf21b5a4aeb02-prebuilt.qemu.org 04/01/2014 RIP: 0010:mlx5_ib_cleanup_multiport_master+0x18b/0x2d0 [mlx5_ib] Code: 00 04 0f 85 c4 00 00 00 48 89 df e8 ef fa ff ff 48 8b 83 40 0d 00 00 48 8b 15 b9 e8 05 00 4a 8b 44 28 20 48 89 05 ad e8 05 00 <48> c7 00 d0 57 c5 a0 48 89 50 08 48 89 02 39 ab 88 0a 00 00 0f 86 RSP: 0018:ffff888116ee3df8 EFLAGS: 00010296 RAX: 0000000000000000 RBX: ffff8881154f6000 RCX: 0000000000000080 RDX: ffffffffa0c557d0 RSI: ffff88810b69d200 RDI: 000000000002d8a0 RBP: 0000000000000002 R08: ffff888110780408 R09: 0000000000000000 R10: ffff88812452e1c0 R11: fffffffffff7e028 R12: 0000000000000000 R13: 0000000000000080 R14: ffff888102c58000 R15: 0000000000000000 FS: 00007f884393a740(0000) GS:ffff8882f5a00000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 0000000000000000 CR3: 00000001249f6004 CR4: 0000000000370ea0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 Call Trace: mlx5_ib_stage_init_cleanup+0x16/0xd0 [mlx5_ib] __mlx5_ib_remove+0x33/0x90 [mlx5_ib] mlx5r_remove+0x22/0x30 [mlx5_ib] auxiliary_bus_remove+0x18/0x30 __device_release_driver+0x177/0x220 driver_detach+0xc4/0x100 bus_remove_driver+0x58/0xd0 auxiliary_driver_unregister+0x12/0x20 mlx5_ib_cleanup+0x13/0x897 [mlx5_ib] __x64_sys_delete_module+0x154/0x230 ? exit_to_user_mode_prepare+0x104/0x140 do_syscall_64+0x3f/0x80 entry_SYSCALL_64_after_hwframe+0x44/0xae RIP: 0033:0x7f8842e095c7 Code: 73 01 c3 48 8b 0d d9 48 2c 00 f7 d8 64 89 01 48 83 c8 ff c3 66 2e 0f 1f 84 00 00 00 00 00 0f 1f 44 00 00 b8 b0 00 00 00 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 8b 0d a9 48 2c 00 f7 d8 64 89 01 48 RSP: 002b:00007ffc68f6e758 EFLAGS: 00000206 ORIG_RAX: 00000000000000b0 RAX: ffffffffffffffda RBX: 00005638207929c0 RCX: 00007f8842e095c7 RDX: 0000000000000000 RSI: 0000000000000800 RDI: 0000563820792a28 RBP: 00005638207929c0 R08: 00007ffc68f6d701 R09: 0000000000000000 R10: 00007f8842e82880 R11: 0000000000000206 R12: 0000563820792a28 R13: 0000000000000001 R14: 0000563820792a28 R15: 00007ffc68f6fb40 Modules linked in: xt_MASQUERADE nf_conntrack_netlink nfnetlink iptable_nat xt_addrtype xt_conntrack nf_nat nf_conntrack nf_defrag_ipv6 nf_defrag_ipv4 br_netfilter overlay rdma_ucm ib_iser libiscsi scsi_transport_iscsi rdma_cm iw_cm ib_ipoib ib_cm ib_umad mlx5_ib(-) mlx4_ib ib_uverbs ib_core mlx4_en mlx4_core mlx5_core ptp pps_core [last unloaded: rpcrdma] CR2: 0000000000000000 ---[ end trace a0bb7e20804e9e9b ]--- Fixes: 7ce6095e3bff ("RDMA/mlx5: Don't add slave port to unaffiliated list") Link: https://lore.kernel.org/r/899ac1b33a995be5ec0e16a4765c4e43c2b1ba5b.1624956444.git.leonro@nvidia.com Reviewed-by: Itay Aveksis <itayav@nvidia.com> Reviewed-by: Maor Gottlieb <maorg@nvidia.com> Signed-off-by: Leon Romanovsky <leonro@nvidia.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-06-25RDMA/irdma: Fix potential overflow expression in irdma_prm_get_pblesShiraz Saleem2-3/+3
Coverity reports a signed 32-bit overflow on "1 << pprm->pble_shift" when used expression to compute bits_needed that expects 64bit, unsigned. Fix this by using the 1ULL in the left shift operator and convert mem_size to u64. Link: https://lore.kernel.org/r/20210625162329.1654-3-tatyana.e.nikolova@intel.com Reported-by: coverity-bot <keescook+coverity-bot@chromium.org> Addresses-Coverity-ID: 1505157 ("Integer handling issues") Fixes: 915cc7ac0f8e ("RDMA/irdma: Add miscellaneous utility definitions") Signed-off-by: Shiraz Saleem <shiraz.saleem@intel.com> Signed-off-by: Tatyana Nikolova <tatyana.e.nikolova@intel.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-06-25RDMA/irdma: Check contents of user-space irdma_mem_reg_req objectShiraz Saleem1-6/+20
The contents of user-space req object is used in array indexing in irdma_handle_q_mem without checking for valid values. Guard against bad input on each of these req object pages by limiting them to number of pages that make up the region. Link: https://lore.kernel.org/r/20210625162329.1654-2-tatyana.e.nikolova@intel.com Reported-by: coverity-bot <keescook+coverity-bot@chromium.org> Addresses-Coverity-ID: 1505160 ("TAINTED_SCALAR") Fixes: b48c24c2d710 ("RDMA/irdma: Implement device supported verb APIs") Signed-off-by: Shiraz Saleem <shiraz.saleem@intel.com> Signed-off-by: Tatyana Nikolova <tatyana.e.nikolova@intel.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-06-25RDMA/rxe: Missing unlock on error in get_srq_wqe()Dan Carpenter1-0/+1
This error path needs to unlock before returning. Fixes: ec0fa2445c18 ("RDMA/rxe: Fix over copying in get_srq_wqe") Link: https://lore.kernel.org/r/YNXUCmnPsSkPyhkm@mwanda Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com> Reviewed-by: Majd Dibbiny <majd@nvidia.com> Reviewed-by: Bob Pearson <rpearsonhpe@gmail.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-06-25RDMA/cma: Fix rdma_resolve_route() memory leakGerd Rausch1-1/+2
Fix a memory leak when "mda_resolve_route() is called more than once on the same "rdma_cm_id". This is possible if cma_query_handler() triggers the RDMA_CM_EVENT_ROUTE_ERROR flow which puts the state machine back and allows rdma_resolve_route() to be called again. Link: https://lore.kernel.org/r/f6662b7b-bdb7-2706-1e12-47c61d3474b6@oracle.com Signed-off-by: Gerd Rausch <gerd.rausch@oracle.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-06-25RDMA/core/sa_query: Remove unused argumentHåkon Bugge1-4/+2
Fixes:4c33bd1926cc ("IB/SA: Add support to query OPA path records") Link: https://lore.kernel.org/r/1624624257-3677-1-git-send-email-haakon.bugge@oracle.com Signed-off-by: Håkon Bugge <haakon.bugge@oracle.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-06-25RDMA/cma: Fix incorrect Packet Lifetime calculationHåkon Bugge1-2/+4
An approximation for the PacketLifeTime is half the local ACK timeout. The encoding for both timers are logarithmic. If the local ACK timeout is set, but zero, it means the timer is disabled. In this case, we choose the CMA_IBOE_PACKET_LIFETIME value, since 50% of infinite makes no sense. Before this commit, the PacketLifeTime became 255 if local ACK timeout was zero (not running). Fixed by explicitly testing for timeout being zero. Fixes: e1ee1e62bec4 ("RDMA/cma: Use ACK timeout for RoCE packetLifeTime") Link: https://lore.kernel.org/r/1624371207-26710-1-git-send-email-haakon.bugge@oracle.com Signed-off-by: Håkon Bugge <haakon.bugge@oracle.com> Reviewed-by: Leon Romanovsky <leonro@nvidia.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-06-24RDMA/cma: Protect RMW with qp_mutexHåkon Bugge1-1/+19
The struct rdma_id_private contains three bit-fields, tos_set, timeout_set, and min_rnr_timer_set. These are set by accessor functions without any synchronization. If two or all accessor functions are invoked in close proximity in time, there will be Read-Modify-Write from several contexts to the same variable, and the result will be intermittent. Fixed by protecting the bit-fields by the qp_mutex in the accessor functions. The consumer of timeout_set and min_rnr_timer_set is in rdma_init_qp_attr(), which is called with qp_mutex held for connected QPs. Explicit locking is added for the consumers of tos and tos_set. This commit depends on ("RDMA/cma: Remove unnecessary INIT->INIT transition"), since the call to rdma_init_qp_attr() from cma_init_conn_qp() does not hold the qp_mutex. Fixes: 2c1619edef61 ("IB/cma: Define option to set ack timeout and pack tos_set") Fixes: 3aeffc46afde ("IB/cma: Introduce rdma_set_min_rnr_timer()") Link: https://lore.kernel.org/r/1624369197-24578-3-git-send-email-haakon.bugge@oracle.com Signed-off-by: Håkon Bugge <haakon.bugge@oracle.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-06-24RDMA/cma: Remove unnecessary INIT->INIT transitionHåkon Bugge1-16/+1
In rdma_create_qp(), a connected QP will be transitioned to the INIT state. Afterwards, the QP will be transitioned to the RTR state by the cma_modify_qp_rtr() function. But this function starts by performing an ib_modify_qp() to the INIT state again, before another ib_modify_qp() is performed to transition the QP to the RTR state. Hence, there is no need to transition the QP to the INIT state in rdma_create_qp(). Link: https://lore.kernel.org/r/1624369197-24578-2-git-send-email-haakon.bugge@oracle.com Signed-off-by: Håkon Bugge <haakon.bugge@oracle.com> Reviewed-by: Leon Romanovsky <leonro@nvidia.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-06-24RDMA/hns: Add window selection field of congestion controlYixing Liu2-0/+14
The window selection field is necessary for congestion control of HIP09, it is got from firmware and then filled into QPC. Some algorithms need it to decide whether to limit the number of windows. Fixes: f91696f2f053 ("RDMA/hns: Support congestion control type selection according to the FW") Link: https://lore.kernel.org/r/1624364163-44185-1-git-send-email-liweihang@huawei.com Signed-off-by: Yixing Liu <liuyixing1@huawei.com> Signed-off-by: Weihang Li <liweihang@huawei.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-06-24RDMA/hfi1: Remove use of kmap()Ira Weiny1-2/+2
kmap() is being deprecated and will break uses of device dax after PKS protection is introduced.[1] The kmap() used in sdma does not need to be global. Use the new kmap_local_page() which works with PKS and may provide better performance for this thread local mapping anyway. [1] https://lore.kernel.org/lkml/20201009195033.3208459-59-ira.weiny@intel.com/ Link: https://lore.kernel.org/r/20210622061422.2633501-2-ira.weiny@intel.com Signed-off-by: Ira Weiny <ira.weiny@intel.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-06-24RDMA/irdma: Remove use of kmap()Ira Weiny1-4/+4
kmap() is being deprecated and will break uses of device dax after PKS protection is introduced.[1] The kmap() used in the irdma CM driver is thread local. Therefore kmap_local_page() is sufficient to use and may provide performance benefits as well. kmap_local_page() will work with device dax and pgmap protected pages. Use kmap_local_page() instead of kmap(). [1] https://lore.kernel.org/lkml/20201009195033.3208459-59-ira.weiny@intel.com/ Link: https://lore.kernel.org/r/20210622165622.2638628-1-ira.weiny@intel.com Signed-off-by: Ira Weiny <ira.weiny@intel.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-06-24RDMA/bnxt_re: Fix uninitialized struct bit field rsvd1Colin Ian King1-4/+1
The bit field rsvd1 in resp is not being initialized and garbage data is being copied from the stack back to userspace via the ib_copy_to_udata call. Fix this by setting the entire struct resp to zero; this will ensure that further new bit fields in the future will be zero'd too. Link: https://lore.kernel.org/r/20210623182437.163801-1-colin.king@canonical.com Addresses-Coverity: ("Uninitialized scalar variable") Fixes: 879740517dab ("RDMA/bnxt_re: Update ABI to pass wqe-mode to user space") Signed-off-by: Colin Ian King <colin.king@canonical.com> [jgg: remove extra zeroing] Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-06-24IB/isert: Align target max I/O size to initiator sizeMax Gurtovoy2-5/+2
Since the Linux iser initiator default max I/O size set to 512KB and since there is no handshake procedure for this size in iser protocol, set the default max IO size of the target to 512KB as well. For changing the default values, there is a module parameter for both drivers. Link: https://lore.kernel.org/r/20210524085215.29005-1-mgurtovoy@nvidia.com Reviewed-by: Alaa Hleihel <alaa@nvidia.com> Reviewed-by: Israel Rukshin <israelr@nvidia.com> Signed-off-by: Max Gurtovoy <mgurtovoy@nvidia.com> Acked-by: Sagi Grimberg <sagi@grimberg.me> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-06-23RDMA/hns: Fix incorrect vlan enable bit in QPCWeihang Li1-2/+2
The QPC_RQ/SQ_VLAN_EN bit in QPC should be enabled, not the QPC mask. Fixes: f0cb411aad23 ("RDMA/hns: Use new interface to modify QP context") Link: https://lore.kernel.org/r/1624438201-11915-1-git-send-email-liweihang@huawei.com Signed-off-by: Weihang Li <liweihang@huawei.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-06-23MAINTAINERS: Update Broadcom RDMA maintainersSelvin Xavier1-2/+0
Updating the maintainers file as Devesh decidied to leave Broadcom. Link: https://lore.kernel.org/r/1624436089-28263-1-git-send-email-selvin.xavier@broadcom.com Signed-off-by: Selvin Xavier <selvin.xavier@broadcom.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-06-23RDMA/irdma: Use the queried port attributesKamal Heib1-1/+1
Instead of hard code the gid_table_len value, use the value from the ib_query_port() attributes. Fixes: b48c24c2d710 ("RDMA/irdma: Implement device supported verb APIs") Link: https://lore.kernel.org/r/20210620201503.67055-1-kamalheib1@gmail.com Signed-off-by: Kamal Heib <kamalheib1@gmail.com> Acked-by: Tatyana Nikolova <tatyana.e.nikolova@intel.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-06-22RDMA/rxe: Fix redundant skb_put_zeroBob Pearson1-1/+1
rxe_init_packet() in rxe_net.c calls skb_put_zero() to reserve space for the payload and zero it out. All these bytes are then re-written with RoCE headers and payload. Remove this useless extra copy. Fixes: ecb238f6a7f3 ("IB/cxgb4: use skb_put_zero()/__skb_put_zero") Link: https://lore.kernel.org/r/20210618045742.204195-7-rpearsonhpe@gmail.com Signed-off-by: Bob Pearson <rpearsonhpe@gmail.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-06-22RDMA/rxe: Fix extra copy in prepare_ack_packetBob Pearson1-10/+3
Currently prepare_ack_packet writes almost all the fields of the BTH in the ack packet twice. Replace code with the subroutine init_bth(). Fixes: 8700e3e7c485 ("Soft RoCE driver") Link: https://lore.kernel.org/r/20210618045742.204195-6-rpearsonhpe@gmail.com Signed-off-by: Bob Pearson <rpearsonhpe@gmail.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-06-22RDMA/rxe: Fix over copying in get_srq_wqeBob Pearson1-2/+8
Currently get_srq_wqe() in rxe_resp.c copies the maximum possible number of bytes from the wqe into the QPs copy of the SRQ wqe. This is usually extra work and risks reading past the end of the SRQ circular buffer if the SRQ is configured with less than the maximum possible number of SGEs. Check the number of SGEs is not too large. Compute the actual number of bytes in the WR and copy only those. Fixes: 8700e3e7c485 ("Soft RoCE driver") Link: https://lore.kernel.org/r/20210618045742.204195-5-rpearsonhpe@gmail.com Signed-off-by: Bob Pearson <rpearsonhpe@gmail.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-06-22RDMA/rxe: Fix extra copies in build_rdma_network_hdrBob Pearson1-17/+12
build_rdma_network_hdr() in rxe_resp.c does more copying than is needed. Remove this subroutine and eliminate the extra copies for IPV6 and reduce the extra copying for IPV4. Fixes: e404f945a610 ("IB/rxe: improved debug prints & code cleanup") Link: https://lore.kernel.org/r/20210618045742.204195-4-rpearsonhpe@gmail.com Signed-off-by: Bob Pearson <rpearsonhpe@gmail.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-06-22RDMA/rxe: Fix redundant call to ip_send_checkBob Pearson1-2/+0
For IPV4 packets sent on the wire the rxe driver calls ip_local_out() which immediately calls __ip_local_out() which sets iph->tot_len and calls ip_send_check(). This code is duplicated in prepare4(). On the loopback path the IP header checksum and tot_len fields are not used so they do not need to be set. Remove this redundant code. Fixes: 8700e3e7c485 ("Soft RoCE driver") Link: https://lore.kernel.org/r/20210618045742.204195-3-rpearsonhpe@gmail.com Signed-off-by: Bob Pearson <rpearsonhpe@gmail.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-06-22RDMA/rxe: Fix useless copy in send_atomic_ackBob Pearson1-4/+0
In send_atomic_ack() in rxe_resp.c there is code copying ack_pkt into the skb->cb[]. This doesn't do anything useful because the cb[] is not used in the transmit path by the rxe driver. Remove this code. Fixes: 4c93496f18ce ("IB/rxe: do not copy extra stack memory to skb") Link: https://lore.kernel.org/r/20210618045742.204195-2-rpearsonhpe@gmail.com Signed-off-by: Bob Pearson <rpearson@hpe.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-06-22RDMA/hns: Add vendor_err info to error WCLang Cheng1-0/+1
ULP can get more error information of CQ through verbs instead of prints. Link: https://lore.kernel.org/r/1624362836-11631-1-git-send-email-liweihang@huawei.com Signed-off-by: Lang Cheng <chenglang@huawei.com> Signed-off-by: Weihang Li <liweihang@huawei.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-06-22RDMA/irdma: Check return value from ib_umem_find_best_pgszShiraz Saleem1-1/+7
iwmr->page_size stores the return from ib_umem_find_best_pgsz and maybe zero when used in ib_umem_num_dma_blocks thus causing a divide by zero error. Fix this by erroring out of irdma_reg_user when 0 is returned from ib_umem_find_best_pgsz. Link: https://lore.kernel.org/r/20210622175232.439-3-tatyana.e.nikolova@intel.com Reported-by: coverity-bot <keescook+coverity-bot@chromium.org> Addresses-Coverity-ID: 1505149 ("Integer handling issues") Fixes: b48c24c2d710 ("RDMA/irdma: Implement device supported verb APIs") Signed-off-by: Shiraz Saleem <shiraz.saleem@intel.com> Signed-off-by: Tatyana Nikolova <tatyana.e.nikolova@intel.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-06-22RDMA/hns: Fix spelling mistakes of originalLang Cheng1-1/+1
'orignal' should be 'original'. Fixes: 9a4435375cd1 ("IB/hns: Add driver files for hns RoCE driver") Link: https://lore.kernel.org/r/1624011020-16992-11-git-send-email-liweihang@huawei.com Signed-off-by: Lang Cheng <chenglang@huawei.com> Signed-off-by: Weihang Li <liweihang@huawei.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-06-22RDMA/hns: Simplify the judgment in hns_roce_v2_post_send()Yixing Liu1-3/+3
The QP type has been checked in check_send_valid(), if it's not RC, it will process the UD/GSI branch. Link: https://lore.kernel.org/r/1624011020-16992-10-git-send-email-liweihang@huawei.com Signed-off-by: Yixing Liu <liuyixing1@huawei.com> Signed-off-by: Weihang Li <liweihang@huawei.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-06-22RDMA/hns: Encapsulate flushing CQE as a functionWenpeng Liang3-35/+21
The process of flushing CQE can be encapsultated into a function, which can reduce duplicate code. Link: https://lore.kernel.org/r/1624011020-16992-9-git-send-email-liweihang@huawei.com Signed-off-by: Wenpeng Liang <liangwenpeng@huawei.com> Signed-off-by: Weihang Li <liweihang@huawei.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-06-22RDMA/hns: Modify function return value typeYangyang Li3-11/+3
hns_roce_init_qp_table() will only return 0, because this function does not need to return a value, so it is modified to void type. Link: https://lore.kernel.org/r/1624011020-16992-8-git-send-email-liweihang@huawei.com Signed-off-by: Yangyang Li <liyangyang20@huawei.com> Signed-off-by: Weihang Li <liweihang@huawei.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-06-22RDMA/hns: Clean definitions of EQC structureXi Wang1-15/+4
Remove unused members in EQ context structure. Fixes: 782832f25404 ("RDMA/hns: Simplify the function config_eqc()") Link: https://lore.kernel.org/r/1624011020-16992-7-git-send-email-liweihang@huawei.com Signed-off-by: Xi Wang <wangxi11@huawei.com> Signed-off-by: Weihang Li <liweihang@huawei.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-06-22RDMA/hns: Delete unnecessary branch of hns_roce_v2_query_qpYangyang Li1-7/+2
When query_qp is called by userspace, max_send_wr and max_send_sge are set to 0 by the kernel driver. However, the userspace does not use these two return values from the kernel driver, but uses its own calculated values. So there is no need for special treatment. Fixes: 926a01dc000d ("RDMA/hns: Add QP operations support for hip08 SoC") Link: https://lore.kernel.org/r/1624011020-16992-6-git-send-email-liweihang@huawei.com Signed-off-by: Yangyang Li <liyangyang20@huawei.com> Signed-off-by: Weihang Li <liweihang@huawei.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-06-22RDMA/hns: Add member assignments for qp_init_attrYangyang Li1-0/+5
Some kernel ULPs need to use the return value of qp_init_attr, so add member assignments for qp_init_attr. Fixes: 926a01dc000d ("RDMA/hns: Add QP operations support for hip08 SoC") Link: https://lore.kernel.org/r/1624011020-16992-5-git-send-email-liweihang@huawei.com Signed-off-by: Yangyang Li <liyangyang20@huawei.com> Signed-off-by: Weihang Li <liweihang@huawei.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-06-22RDMA/hns: Fix some print issuesYixing Liu2-4/+2
Remove redundant print and fix a character type mismatch. Fixes: 0e0ab04b5bbe ("RDMA/hns: Refactor the MTR creation flow") Link: https://lore.kernel.org/r/1624011020-16992-4-git-send-email-liweihang@huawei.com Signed-off-by: Yixing Liu <liuyixing1@huawei.com> Signed-off-by: Weihang Li <liweihang@huawei.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-06-22RDMA/hns: Fix uninitialized variableYixing Liu1-1/+1
A random value will be returned if the condition below is not met, so it needs to be initialized. Fixes: 9ea9a53ea93b ("RDMA/hns: Add mapped page count checking for MTR") Link: https://lore.kernel.org/r/1624011020-16992-3-git-send-email-liweihang@huawei.com Signed-off-by: Yixing Liu <liuyixing1@huawei.com> Signed-off-by: Weihang Li <liweihang@huawei.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-06-22RDMA/hns: Force rewrite inline flag of WQELang Cheng1-2/+2
When a non-inline WR reuses a WQE that was used for inline last time, the remaining inline flag should be cleared. Fixes: 62490fd5a865 ("RDMA/hns: Avoid unnecessary memset on WQEs in post_send") Link: https://lore.kernel.org/r/1624011020-16992-2-git-send-email-liweihang@huawei.com Signed-off-by: Lang Cheng <chenglang@huawei.com> Signed-off-by: Weihang Li <liweihang@huawei.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-06-22Merge branch 'mlx5_realtime_ts' into rdma.git for-nextJason Gunthorpe9-87/+94
Aharon Landau says: ==================== In case device supports only real-time timestamp, the kernel will fail to create QP despite rdma-core requested such timestamp type. It is because device returns free-running timestamp, and the conversion from free-running to real-time is performed in the user space. This series fixes it, by returning real-time timestamp. ==================== * mlx5_realtime_ts: RDMA/mlx5: Support real-time timestamp directly from the device RDMA/mlx5: Refactor get_ts_format functions to simplify code Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-06-22Merge tag 'v5.13-rc7' into rdma.git for-nextJason Gunthorpe1665-8916/+18578
Linux 5.13-rc7 Needed for dependencies in following patches. Merge conflict in rxe_cmop.c resolved by compining both patches. Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-06-22RDMA/mlx5: Support real-time timestamp directly from the deviceAharon Landau5-6/+45
Currently, if the user asks for a real-time timestamp, the device will return a free-running one, and the timestamp will be translated to real-time in the user-space. When the device supports only real-time timestamp and not free-running, the creation of the QP will fail even though the user needs supported the real-time one. To prevent this, we will return the real-time timestamp directly from the device. Link: https://lore.kernel.org/r/c6cfc8e6f038575c5c2de6505830f7e74e4de80d.1623829775.git.leonro@nvidia.com Signed-off-by: Aharon Landau <aharonl@nvidia.com> Reviewed-by: Maor Gottlieb <maorg@nvidia.com> Signed-off-by: Leon Romanovsky <leonro@nvidia.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-06-22RDMA/core: Use flexible array for mad dataKees Cook1-1/+1
In preparation for FORTIFY_SOURCE performing compile-time and run-time field bounds checking for memcpy(), memmove(), and memset(), avoid intentionally reading across neighboring array fields. Without a flexible array, this looks like an attempt to perform a memcpy() read beyond the end of the packet->mad.data array: drivers/infiniband/core/user_mad.c: memcpy(packet->msg->mad, packet->mad.data, IB_MGMT_MAD_HDR); Switch from [0] to [] to use the appropriately handled type for trailing bytes. Link: https://lore.kernel.org/r/20210616202615.1247242-1-keescook@chromium.org Signed-off-by: Kees Cook <keescook@chromium.org> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-06-22RDMA/mlx5: Refactor get_ts_format functions to simplify codeAharon Landau5-87/+55
QPC, SQC and RQC timestamp formats and capabilities are always equal because they represent general hardware support. So instead of code duplication, let's merge them into general enum and logic. Signed-off-by: Aharon Landau <aharonl@nvidia.com> Reviewed-by: Maor Gottlieb <maorg@nvidia.com> Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
2021-06-22RDMA/rxe: Don't overwrite errno from ib_umem_get()Xiao Yang1-1/+1
rxe_mr_init_user() always returns the fixed -EINVAL when ib_umem_get() fails so it's hard for user to know which actual error happens in ib_umem_get(). For example, ib_umem_get() will return -EOPNOTSUPP when trying to pin pages on a DAX file. Return actual error as mlx4/mlx5 does. Link: https://lore.kernel.org/r/20210621071456.4259-1-ice_yangxiao@163.com Signed-off-by: Xiao Yang <yangx.jy@fujitsu.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-06-22IB/mlx4: Avoid field-overflowing memcpy()Kees Cook1-1/+1
In preparation for FORTIFY_SOURCE performing compile-time and run-time field bounds checking for memcpy(), memmove(), and memset(), avoid intentionally writing across neighboring array fields. Use the ether_addr_copy() helper instead, as already done for smac. Link: https://lore.kernel.org/r/20210616203744.1248551-1-keescook@chromium.org Signed-off-by: Kees Cook <keescook@chromium.org> Reviewed-by: Leon Romanovsky <leonro@nvidia.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-06-22rnbd/rtrs-clt: Query and use max_segments from rtrs-clt.Jack Wang4-17/+13
With fast memory registration on write request, rnbd-clt can do bigger IO without split. rnbd-clt now can query rtrs-clt to get the max_segments, instead of using BMAX_SEGMENTS. BMAX_SEGMENTS is not longer needed, so remove it. Link: https://lore.kernel.org/r/20210621055340.11789-6-jinpu.wang@ionos.com Cc: Jens Axboe <axboe@kernel.dk> Signed-off-by: Jack Wang <jinpu.wang@cloud.ionos.com> Reviewed-by: Md Haris Iqbal <haris.iqbal@cloud.ionos.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-06-22RDMA/rtrs-clt: Raise MAX_SEGMENTSJack Wang1-2/+4
As we can do fast memory registration on write, we can increase the max_segments, default to 512K. Link: https://lore.kernel.org/r/20210621055340.11789-5-jinpu.wang@ionos.com Signed-off-by: Jack Wang <jinpu.wang@cloud.ionos.com> Reviewed-by: Md Haris Iqbal <haris.iqbal@cloud.ionos.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-06-22RDMA/rtrs_clt: Alloc less memory with write path fast memory registrationJack Wang1-3/+2
With write path fast memory registration, we need less memory for each request. With fast memory registration, we can reduce max_send_sge to save memory usage. Also convert the kmalloc_array to kcalloc. Link: https://lore.kernel.org/r/20210621055340.11789-4-jinpu.wang@ionos.com Signed-off-by: Jack Wang <jinpu.wang@cloud.ionos.com> Reviewed-by: Md Haris Iqbal <haris.iqbal@cloud.ionos.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-06-22RDMA/rtrs-clt: Write path fast memory registrationJack Wang2-27/+74
With fast memory registration in write path, we can reduce the memory consumption by using less max_send_sge, support IO bigger than 116 KB (29 segments * 4 KB) without splitting, and it also make the IO path more symmetric. To avoid some times MR reg failed, waiting for the invalidation to finish before the new mr reg. Introduce a refcount, only finish the request when both local invalidation and io reply are there. Link: https://lore.kernel.org/r/20210621055340.11789-3-jinpu.wang@ionos.com Signed-off-by: Jack Wang <jinpu.wang@cloud.ionos.com> Signed-off-by: Md Haris Iqbal <haris.iqbal@ionos.com> Signed-off-by: Dima Stepanov <dmitrii.stepanov@ionos.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-06-22RDMA/rtrs: Introduce head/tail wrJack Wang3-20/+27
Introduce tail wr, we can send as the last wr, we want to send the local invalidate wr after rdma wr in later patch. While at it, also fix coding style issue. Link: https://lore.kernel.org/r/20210621055340.11789-2-jinpu.wang@ionos.com Signed-off-by: Jack Wang <jinpu.wang@cloud.ionos.com> Reviewed-by: Md Haris Iqbal <haris.iqbal@cloud.ionos.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-06-22RDMA/bnxt_re: Update ABI to pass wqe-mode to user spaceDevesh Sharma4-7/+15
Changing ucontext ABI response structure to pass wqe_mode to user library. A flag in comp_mask has been set to indicate presence of wqe_mode. Moved wqe-mode ABI to uapi/rdma/bnxt_re-abi.h Link: https://lore.kernel.org/r/20210616202817.1185276-1-devesh.sharma@broadcom.com Signed-off-by: Devesh Sharma <devesh.sharma@broadcom.com> Reviewed-by: Leon Romanovsky <leonro@nvidia.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-06-22IB/core: Shuffle locks in ib_port_data to save memoryAnand Khoje1-1/+3
pahole shows two 4-byte holes in struct ib_port_data after pkey_list_lock and netdev_lock respectively. Shuffling the netdev_lock to be after pkey_list_lock, this shaves off eight bytes from the struct. Link: https://lore.kernel.org/r/20210616154509.1047-3-anand.a.khoje@oracle.com Suggested-by: Haakon Bugge <haakon.bugge@oracle.com> Signed-off-by: Anand Khoje <anand.a.khoje@oracle.com> Reviewed-by: Leon Romanovsky <leonro@nvidia.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
generated by cgit v1.2.3 (git 2.39.0) at 2025-09-27 11:36:05 +0300