aboutsummaryrefslogtreecommitdiff
path: root/drivers/infiniband/hw
AgeCommit message (Collapse)AuthorFilesLines
2024-04-11RDMA/mana_ib: Prefer struct_size over open coded arithmeticErick Archer1-7/+5
This is an effort to get rid of all multiplications from allocation functions in order to prevent integer overflows [1][2]. As the "req" variable is a pointer to "struct mana_cfg_rx_steer_req_v2" and this structure ends in a flexible array: struct mana_cfg_rx_steer_req_v2 { [...] mana_handle_t indir_tab[] __counted_by(num_indir_entries); }; the preferred way in the kernel is to use the struct_size() helper to do the arithmetic instead of the calculation "size + size * count" in the kzalloc() function. Moreover, use the "offsetof" helper to get the indirect table offset instead of the "sizeof" operator and avoid the open-coded arithmetic in pointers using the new flex member. This new structure member also allow us to remove the "req_indir_tab" variable since it is no longer needed. This way, the code is more readable and safer. This code was detected with the help of Coccinelle, and audited and modified manually. Link: https://www.kernel.org/doc/html/latest/process/deprecated.html#open-coded-arithmetic-in-allocator-arguments [1] Link: https://github.com/KSPP/linux/issues/160 [2] Signed-off-by: Erick Archer <[email protected]> Link: https://lore.kernel.org/r/AS8PR02MB72375EB06EE1A84A67BE722E8B022@AS8PR02MB7237.eurprd02.prod.outlook.com Reviewed-by: Long Li <[email protected]> Reviewed-by: Justin Stitt <[email protected]> Signed-off-by: Leon Romanovsky <[email protected]>
2024-04-09RDMA/hns: Support DSCPJunxian Huang4-29/+108
Add support for DSCP configuration. For DSCP, get dscp-prio mapping via hns3 nic driver api .get_dscp_prio() and fill the SL (in WQE for UD or in QPC for RC) with the priority value. The prio-tc mapping is configured to HW by hns3 nic driver. HW will select a corresponding TC according to SL and the prio-tc mapping. Signed-off-by: Junxian Huang <[email protected]> Link: https://lore.kernel.org/r/[email protected] Signed-off-by: Leon Romanovsky <[email protected]>
2024-04-08ipv4: Set scope explicitly in ip_route_output().Guillaume Nault2-2/+4
Add a "scope" parameter to ip_route_output() so that callers don't have to override the tos parameter with the RTO_ONLINK flag if they want a local scope. This will allow converting flowi4_tos to dscp_t in the future, thus allowing static analysers to flag invalid interactions between "tos" (the DSCP bits) and ECN. Only three users ask for local scope (bonding, arp and atm). The others continue to use RT_SCOPE_UNIVERSE. While there, add a comment to warn users about the limitations of ip_route_output(). Signed-off-by: Guillaume Nault <[email protected]> Acked-by: Leon Romanovsky <[email protected]> # infiniband Signed-off-by: David S. Miller <[email protected]>
2024-04-08RDMA/mlx5: Adding remote atomic access flag to updatable flagsOr Har-Toov1-1/+2
Currently IB_ACCESS_REMOTE_ATOMIC is blocked from being updated via UMR although in some cases it should be possible. These cases are checked in mlx5r_umr_can_reconfig function. Fixes: ef3642c4f54d ("RDMA/mlx5: Fix error unwinds for rereg_mr") Signed-off-by: Or Har-Toov <[email protected]> Link: https://lore.kernel.org/r/24dac73e2fa48cb806f33a932d97f3e402a5ea2c.1712140377.git.leon@kernel.org Signed-off-by: Leon Romanovsky <[email protected]>
2024-04-08RDMA/mlx5: Change check for cacheable mkeysOr Har-Toov2-10/+23
umem can be NULL for user application mkeys in some cases. Therefore umem can't be used for checking if the mkey is cacheable and it is changed for checking a flag that indicates it. Also make sure that all mkeys which are not returned to the cache will be destroyed. Fixes: dd1b913fb0d0 ("RDMA/mlx5: Cache all user cacheable mkeys on dereg MR flow") Signed-off-by: Or Har-Toov <[email protected]> Link: https://lore.kernel.org/r/2690bc5c6896bcb937f89af16a1ff0343a7ab3d0.1712140377.git.leon@kernel.org Signed-off-by: Leon Romanovsky <[email protected]>
2024-04-08RDMA/mlx5: Uncacheable mkey has neither rb_key or cache_entOr Har-Toov1-1/+1
As some mkeys can't be modified with UMR due to some UMR limitations, like the size of translation that can be updated, not all user mkeys can be cached. Fixes: dd1b913fb0d0 ("RDMA/mlx5: Cache all user cacheable mkeys on dereg MR flow") Signed-off-by: Or Har-Toov <[email protected]> Link: https://lore.kernel.org/r/f2742dd934ed73b2d32c66afb8e91b823063880c.1712140377.git.leon@kernel.org Signed-off-by: Leon Romanovsky <[email protected]>
2024-04-08RDMA/mlx5: Fix port number for counter query in multi-port configurationMichael Guralnik1-1/+2
Set the correct port when querying PPCNT in multi-port configuration. Distinguish between cases where switchdev mode was enabled to multi-port configuration and don't overwrite the queried port to 1 in multi-port case. Fixes: 74b30b3ad5ce ("RDMA/mlx5: Set local port to one when accessing counters") Signed-off-by: Michael Guralnik <[email protected]> Link: https://lore.kernel.org/r/9bfcc8ade958b760a51408c3ad654a01b11f7d76.1712134988.git.leon@kernel.org Signed-off-by: Leon Romanovsky <[email protected]>
2024-04-02RDMA/mana_ib: Use struct mana_ib_queue for RAW QPsKonstantin Taranov2-46/+18
Use struct mana_ib_queue and its helpers for RAW QPs Signed-off-by: Konstantin Taranov <[email protected]> Link: https://lore.kernel.org/r/[email protected] Reviewed-by: Long Li <[email protected]> Signed-off-by: Leon Romanovsky <[email protected]>
2024-04-02RDMA/mana_ib: Use struct mana_ib_queue for WQsKonstantin Taranov3-35/+10
Use struct mana_ib_queue and its helpers for WQs Signed-off-by: Konstantin Taranov <[email protected]> Link: https://lore.kernel.org/r/[email protected] Reviewed-by: Long Li <[email protected]> Signed-off-by: Leon Romanovsky <[email protected]>
2024-04-02RDMA/mana_ib: Use struct mana_ib_queue for CQsKonstantin Taranov3-58/+24
Use struct mana_ib_queue and its helpers for CQs Signed-off-by: Konstantin Taranov <[email protected]> Link: https://lore.kernel.org/r/[email protected] Reviewed-by: Long Li <[email protected]> Signed-off-by: Leon Romanovsky <[email protected]>
2024-04-02RDMA/mana_ib: Introduce helpers to create and destroy mana queuesKonstantin Taranov2-0/+53
Intoduce helpers to work with mana ib queues (struct mana_ib_queue). A queue always consists of umem, gdma_region, and id. A queue can become a WQ or a CQ. Signed-off-by: Konstantin Taranov <[email protected]> Link: https://lore.kernel.org/r/[email protected] Reviewed-by: Long Li <[email protected]> Signed-off-by: Leon Romanovsky <[email protected]>
2024-04-01RDMA/erdma: Remove unnecessary __GFP_ZERO flagBoshi Yu2-8/+4
The dma_alloc_coherent() interface automatically zero the memory returned. Thus, we do not need to specify the __GFP_ZERO flag explicitly when we call dma_alloc_coherent(). Reviewed-by: Cheng Xu <[email protected]> Signed-off-by: Boshi Yu <[email protected]> Link: https://lore.kernel.org/r/[email protected] Signed-off-by: Leon Romanovsky <[email protected]>
2024-04-01RDMA/erdma: Unify the names related to doorbell recordsBoshi Yu8-93/+79
There exist two different names for the doorbell records: db_info and db_record. We use dbrec for cpu address of the doorbell record and dbrec_dma for dma address of the doorbell recordi uniformly. Reviewed-by: Cheng Xu <[email protected]> Signed-off-by: Boshi Yu <[email protected]> Link: https://lore.kernel.org/r/[email protected] Signed-off-by: Leon Romanovsky <[email protected]>
2024-04-01RDMA/erdma: Allocate doorbell records from dma poolBoshi Yu6-101/+167
Currently, the 8 byte doorbell record is allocated along with the queue buffer, which may result in waste of dma space when the queue buffer is page aligned. To address this issue, we introduce a dma pool named db_pool and allocate doorbell record from it. Reviewed-by: Cheng Xu <[email protected]> Signed-off-by: Boshi Yu <[email protected]> Link: https://lore.kernel.org/r/[email protected] Signed-off-by: Leon Romanovsky <[email protected]>
2024-03-18Merge tag 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdmaLinus Torvalds23-469/+636
Pull rdma updates from Jason Gunthorpe: "Very small update this cycle: - Minor code improvements in fi, rxe, ipoib, mana, cxgb4, mlx5, irdma, rxe, rtrs, mana - Simplify the hns hem mechanism - Fix EFA's MSI-X allocation in resource constrained configurations - Fix a KASN splat in srpt - Narrow hns's congestion control selection to QPs granularity and allow userspace to select it - Solve a parallel module loading race between the CM module and a driver module - Flexible array cleanup - Dump hns's SCC Conext to 'rdma res' for debugging - Make mana build page lists for HW objects that require a 0 offset correctly - Stuck CM ID debugging" * tag 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma: (29 commits) RDMA/cm: add timeout to cm_destroy_id wait RDMA/mana_ib: Use virtual address in dma regions for MRs RDMA/mana_ib: Fix bug in creation of dma regions RDMA/hns: Append SCC context to the raw dump of QPC RDMA/uverbs: Avoid -Wflex-array-member-not-at-end warnings RDMA/hns: Support userspace configuring congestion control algorithm with QP granularity RDMA/rtrs-clt: Check strnlen return len in sysfs mpath_policy_store() RDMA/uverbs: Remove flexible arrays from struct *_filter RDMA/device: Fix a race between mad_client and cm_client init RDMA/hns: Fix mis-modifying default congestion control algorithm RDMA/rxe: Remove unused 'iova' parameter from rxe_mr_init_user RDMA/srpt: Do not register event handler until srpt device is fully setup RDMA/irdma: Remove duplicate assignment RDMA/efa: Limit EQs to available MSI-X vectors RDMA/mlx5: Delete unused mlx5_ib_copy_pas prototype RDMA/cxgb4: Delete unused c4iw_ep_redirect prototype RDMA/mana_ib: Introduce mana_ib_install_cq_cb helper function RDMA/mana_ib: Introduce mana_ib_get_netdev helper function RDMA/mana_ib: Introduce mdev_to_gc helper function RDMA/hns: Simplify 'struct hns_roce_hem' allocation ...
2024-03-07RDMA/mana_ib: Use virtual address in dma regions for MRsKonstantin Taranov6-20/+45
Introduce mana_ib_create_dma_region() to create dma regions with iova for MRs. It allows creating MRs with any page offset. Previously, only page-aligned addresses worked. For dma regions that must have a zero dma offset (e.g., for queues), mana_ib_create_zero_offset_dma_region() is added. To get the zero offset, ib_umem_find_best_pgoff() is used with zero pgoff_bitmask. Fixes: 0266a177631d ("RDMA/mana_ib: Add a driver for Microsoft Azure Network Adapter") Signed-off-by: Konstantin Taranov <[email protected]> Link: https://lore.kernel.org/r/[email protected] Reviewed-by: Zhu Yanjun <[email protected]> Signed-off-by: Leon Romanovsky <[email protected]>
2024-03-07RDMA/mana_ib: Fix bug in creation of dma regionsKonstantin Taranov1-1/+1
Use ib_umem_dma_offset() helper to calculate correct dma offset. Fixes: 0266a177631d ("RDMA/mana_ib: Add a driver for Microsoft Azure Network Adapter") Signed-off-by: Konstantin Taranov <[email protected]> Link: https://lore.kernel.org/r/[email protected] Signed-off-by: Leon Romanovsky <[email protected]>
2024-03-07RDMA/hns: Append SCC context to the raw dump of QPCwenglianfa5-3/+55
SCCC (SCC Context) is a context with QP granularity that contains information about congestion control. Dump SCCC and QPC together to improve troubleshooting. When dumping raw QPC with rdmatool, there will be a total of 576 bytes data output, where the first 512 bytes is QPC and the last 64 bytes is SCCC. When congestion control is disabled, the 64 byte SCCC will be all 0. Example: $rdma res show qp -jpr [ { "ifindex": 0, "ifname": "hns_0", "data": [ 67,0,0,0... 512bytes 4,0,2... 64bytes] },... } ] Signed-off-by: wenglianfa <[email protected]> Signed-off-by: Junxian Huang <[email protected]> Link: https://lore.kernel.org/r/[email protected] Signed-off-by: Leon Romanovsky <[email protected]>
2024-03-03RDMA/hns: Support userspace configuring congestion control algorithm with QP ↵Junxian Huang5-12/+69
granularity Currently, congestion control algorithm is statically configured in FW, and all QPs use the same algorithm(except UD which has a fixed configuration of DCQCN). This is not flexible enough. Support userspace configuring congestion control algorithm with QP granularity while creating QPs. If the algorithm is not specified in userspace, use the default one. Signed-off-by: Junxian Huang <[email protected]> Link: https://lore.kernel.org/r/[email protected] Signed-off-by: Leon Romanovsky <[email protected]>
2024-02-25qibfs: fix dentry leakAl Viro1-0/+1
simple_recursive_removal() drops the pinning references to all positives in subtree. For the cases when its argument has been kept alive by the pinning alone that's exactly the right thing to do, but here the argument comes from dcache lookup, that needs to be balanced by explicit dput(). Fixes: e41d237818598 "qib_fs: switch to simple_recursive_removal()" Fucked-up-by: Al Viro <[email protected]> Signed-off-by: Al Viro <[email protected]>
2024-02-19RDMA/hns: Fix mis-modifying default congestion control algorithmLuoyouming2-14/+19
Commit 27c5fd271d8b ("RDMA/hns: The UD mode can only be configured with DCQCN") adds a check of congest control alorithm for UD. But that patch causes a problem: hr_dev->caps.congest_type is global, used by all QPs, so modifying this field to DCQCN for UD QPs causes other QPs unable to use any other algorithm except DCQCN. Revert the modification in commit 27c5fd271d8b ("RDMA/hns: The UD mode can only be configured with DCQCN"). Add a new field cong_type to struct hns_roce_qp and configure DCQCN for UD QPs. Fixes: 27c5fd271d8b ("RDMA/hns: The UD mode can only be configured with DCQCN") Fixes: f91696f2f053 ("RDMA/hns: Support congestion control type selection according to the FW") Signed-off-by: Luoyouming <[email protected]> Signed-off-by: Junxian Huang <[email protected]> Link: https://lore.kernel.org/r/[email protected] Signed-off-by: Leon Romanovsky <[email protected]>
2024-02-12RDMA/qedr: Fix qedr_create_user_qp error flowKamal Heib1-1/+10
Avoid the following warning by making sure to free the allocated resources in case that qedr_init_user_queue() fail. -----------[ cut here ]----------- WARNING: CPU: 0 PID: 143192 at drivers/infiniband/core/rdma_core.c:874 uverbs_destroy_ufile_hw+0xcf/0xf0 [ib_uverbs] Modules linked in: tls target_core_user uio target_core_pscsi target_core_file target_core_iblock ib_srpt ib_srp scsi_transport_srp nfsd nfs_acl rpcsec_gss_krb5 auth_rpcgss nfsv4 dns_resolver nfs lockd grace fscache netfs 8021q garp mrp stp llc ext4 mbcache jbd2 opa_vnic ib_umad ib_ipoib sunrpc rdma_ucm ib_isert iscsi_target_mod target_core_mod ib_iser libiscsi scsi_transport_iscsi rdma_cm iw_cm ib_cm hfi1 intel_rapl_msr intel_rapl_common mgag200 qedr sb_edac drm_shmem_helper rdmavt x86_pkg_temp_thermal drm_kms_helper intel_powerclamp ib_uverbs coretemp i2c_algo_bit kvm_intel dell_wmi_descriptor ipmi_ssif sparse_keymap kvm ib_core rfkill syscopyarea sysfillrect video sysimgblt irqbypass ipmi_si ipmi_devintf fb_sys_fops rapl iTCO_wdt mxm_wmi iTCO_vendor_support intel_cstate pcspkr dcdbas intel_uncore ipmi_msghandler lpc_ich acpi_power_meter mei_me mei fuse drm xfs libcrc32c qede sd_mod ahci libahci t10_pi sg crct10dif_pclmul crc32_pclmul crc32c_intel qed libata tg3 ghash_clmulni_intel megaraid_sas crc8 wmi [last unloaded: ib_srpt] CPU: 0 PID: 143192 Comm: fi_rdm_tagged_p Kdump: loaded Not tainted 5.14.0-408.el9.x86_64 #1 Hardware name: Dell Inc. PowerEdge R430/03XKDV, BIOS 2.14.0 01/25/2022 RIP: 0010:uverbs_destroy_ufile_hw+0xcf/0xf0 [ib_uverbs] Code: 5d 41 5c 41 5d 41 5e e9 0f 26 1b dd 48 89 df e8 67 6a ff ff 49 8b 86 10 01 00 00 48 85 c0 74 9c 4c 89 e7 e8 83 c0 cb dd eb 92 <0f> 0b eb be 0f 0b be 04 00 00 00 48 89 df e8 8e f5 ff ff e9 6d ff RSP: 0018:ffffb7c6cadfbc60 EFLAGS: 00010286 RAX: ffff8f0889ee3f60 RBX: ffff8f088c1a5200 RCX: 00000000802a0016 RDX: 00000000802a0017 RSI: 0000000000000001 RDI: ffff8f0880042600 RBP: 0000000000000001 R08: 0000000000000001 R09: 0000000000000000 R10: ffff8f11fffd5000 R11: 0000000000039000 R12: ffff8f0d5b36cd80 R13: ffff8f088c1a5250 R14: ffff8f1206d91000 R15: 0000000000000000 FS: 0000000000000000(0000) GS:ffff8f11d7c00000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 0000147069200e20 CR3: 00000001c7210002 CR4: 00000000001706f0 Call Trace: <TASK> ? show_trace_log_lvl+0x1c4/0x2df ? show_trace_log_lvl+0x1c4/0x2df ? ib_uverbs_close+0x1f/0xb0 [ib_uverbs] ? uverbs_destroy_ufile_hw+0xcf/0xf0 [ib_uverbs] ? __warn+0x81/0x110 ? uverbs_destroy_ufile_hw+0xcf/0xf0 [ib_uverbs] ? report_bug+0x10a/0x140 ? handle_bug+0x3c/0x70 ? exc_invalid_op+0x14/0x70 ? asm_exc_invalid_op+0x16/0x20 ? uverbs_destroy_ufile_hw+0xcf/0xf0 [ib_uverbs] ib_uverbs_close+0x1f/0xb0 [ib_uverbs] __fput+0x94/0x250 task_work_run+0x5c/0x90 do_exit+0x270/0x4a0 do_group_exit+0x2d/0x90 get_signal+0x87c/0x8c0 arch_do_signal_or_restart+0x25/0x100 ? ib_uverbs_ioctl+0xc2/0x110 [ib_uverbs] exit_to_user_mode_loop+0x9c/0x130 exit_to_user_mode_prepare+0xb6/0x100 syscall_exit_to_user_mode+0x12/0x40 do_syscall_64+0x69/0x90 ? syscall_exit_work+0x103/0x130 ? syscall_exit_to_user_mode+0x22/0x40 ? do_syscall_64+0x69/0x90 ? syscall_exit_work+0x103/0x130 ? syscall_exit_to_user_mode+0x22/0x40 ? do_syscall_64+0x69/0x90 ? do_syscall_64+0x69/0x90 ? common_interrupt+0x43/0xa0 entry_SYSCALL_64_after_hwframe+0x72/0xdc RIP: 0033:0x1470abe3ec6b Code: Unable to access opcode bytes at RIP 0x1470abe3ec41. RSP: 002b:00007fff13ce9108 EFLAGS: 00000246 ORIG_RAX: 0000000000000010 RAX: fffffffffffffffc RBX: 00007fff13ce9218 RCX: 00001470abe3ec6b RDX: 00007fff13ce9200 RSI: 00000000c0181b01 RDI: 0000000000000004 RBP: 00007fff13ce91e0 R08: 0000558d9655da10 R09: 0000558d9655dd00 R10: 00007fff13ce95c0 R11: 0000000000000246 R12: 00007fff13ce9358 R13: 0000000000000013 R14: 0000558d9655db50 R15: 00007fff13ce9470 </TASK> --[ end trace 888a9b92e04c5c97 ]-- Fixes: df15856132bc ("RDMA/qedr: restructure functions that create/destroy QPs") Signed-off-by: Kamal Heib <[email protected]> Link: https://lore.kernel.org/r/[email protected] Signed-off-by: Leon Romanovsky <[email protected]>
2024-02-04IB/hfi1: Fix sdma.h tx->num_descs off-by-one errorDaniel Vacek1-1/+1
Unfortunately the commit `fd8958efe877` introduced another error causing the `descs` array to overflow. This reults in further crashes easily reproducible by `sendmsg` system call. [ 1080.836473] general protection fault, probably for non-canonical address 0x400300015528b00a: 0000 [#1] PREEMPT SMP PTI [ 1080.869326] RIP: 0010:hfi1_ipoib_build_ib_tx_headers.constprop.0+0xe1/0x2b0 [hfi1] -- [ 1080.974535] Call Trace: [ 1080.976990] <TASK> [ 1081.021929] hfi1_ipoib_send_dma_common+0x7a/0x2e0 [hfi1] [ 1081.027364] hfi1_ipoib_send_dma_list+0x62/0x270 [hfi1] [ 1081.032633] hfi1_ipoib_send+0x112/0x300 [hfi1] [ 1081.042001] ipoib_start_xmit+0x2a9/0x2d0 [ib_ipoib] [ 1081.046978] dev_hard_start_xmit+0xc4/0x210 -- [ 1081.148347] __sys_sendmsg+0x59/0xa0 crash> ipoib_txreq 0xffff9cfeba229f00 struct ipoib_txreq { txreq = { list = { next = 0xffff9cfeba229f00, prev = 0xffff9cfeba229f00 }, descp = 0xffff9cfeba229f40, coalesce_buf = 0x0, wait = 0xffff9cfea4e69a48, complete = 0xffffffffc0fe0760 <hfi1_ipoib_sdma_complete>, packet_len = 0x46d, tlen = 0x0, num_desc = 0x0, desc_limit = 0x6, next_descq_idx = 0x45c, coalesce_idx = 0x0, flags = 0x0, descs = {{ qw = {0x8024000120dffb00, 0x4} # SDMA_DESC0_FIRST_DESC_FLAG (bit 63) }, { qw = { 0x3800014231b108, 0x4} }, { qw = { 0x310000e4ee0fcf0, 0x8} }, { qw = { 0x3000012e9f8000, 0x8} }, { qw = { 0x59000dfb9d0000, 0x8} }, { qw = { 0x78000e02e40000, 0x8} }} }, sdma_hdr = 0x400300015528b000, <<< invalid pointer in the tx request structure sdma_status = 0x0, SDMA_DESC0_LAST_DESC_FLAG (bit 62) complete = 0x0, priv = 0x0, txq = 0xffff9cfea4e69880, skb = 0xffff9d099809f400 } If an SDMA send consists of exactly 6 descriptors and requires dword padding (in the 7th descriptor), the sdma_txreq descriptor array is not properly expanded and the packet will overflow into the container structure. This results in a panic when the send completion runs. The exact panic varies depending on what elements of the container structure get corrupted. The fix is to use the correct expression in _pad_sdma_tx_descs() to test the need to expand the descriptor array. With this patch the crashes are no longer reproducible and the machine is stable. Fixes: fd8958efe877 ("IB/hfi1: Fix sdma.h tx->num_descs off-by-one errors") Cc: [email protected] Reported-by: Mats Kronberg <[email protected]> Tested-by: Mats Kronberg <[email protected]> Signed-off-by: Daniel Vacek <[email protected]> Link: https://lore.kernel.org/r/[email protected] Signed-off-by: Leon Romanovsky <[email protected]>
2024-02-04RDMA/irdma: Remove duplicate assignmentMustafa Ismail1-2/+1
Remove the unneeded assignment of the qp_num which is already set in irdma_create_qp(). Fixes: b48c24c2d710 ("RDMA/irdma: Implement device supported verb APIs") Signed-off-by: Mustafa Ismail <[email protected]> Signed-off-by: Shiraz Saleem <[email protected]> Signed-off-by: Sindhu Devale <[email protected]> Link: https://lore.kernel.org/r/[email protected] Signed-off-by: Leon Romanovsky <[email protected]>
2024-02-04RDMA/irdma: Add AE for too many RNRSMustafa Ismail2-0/+2
Add IRDMA_AE_LLP_TOO_MANY_RNRS to the list of AE's processed as an abnormal asyncronous event. Fixes: b48c24c2d710 ("RDMA/irdma: Implement device supported verb APIs") Signed-off-by: Mustafa Ismail <[email protected]> Signed-off-by: Shiraz Saleem <[email protected]> Signed-off-by: Sindhu Devale <[email protected]> Link: https://lore.kernel.org/r/[email protected] Signed-off-by: Leon Romanovsky <[email protected]>
2024-02-04RDMA/irdma: Set the CQ read threshold for GEN 1Mustafa Ismail1-3/+2
The CQ shadow read threshold is currently not set for GEN 2. This could cause an invalid CQ overflow condition, so remove the GEN check that exclused GEN 1. Fixes: b48c24c2d710 ("RDMA/irdma: Implement device supported verb APIs") Signed-off-by: Mustafa Ismail <[email protected]> Signed-off-by: Shiraz Saleem <[email protected]> Signed-off-by: Sindhu Devale <[email protected]> Link: https://lore.kernel.org/r/[email protected] Signed-off-by: Leon Romanovsky <[email protected]>
2024-02-04RDMA/irdma: Validate max_send_wr and max_recv_wrShiraz Saleem1-1/+3
Validate that max_send_wr and max_recv_wr is within the supported range. Fixes: b48c24c2d710 ("RDMA/irdma: Implement device supported verb APIs") Change-Id: I2fc8b10292b641fddd20b36986a9dae90a93f4be Signed-off-by: Shiraz Saleem <[email protected]> Signed-off-by: Sindhu Devale <[email protected]> Link: https://lore.kernel.org/r/[email protected] Signed-off-by: Leon Romanovsky <[email protected]>
2024-02-04RDMA/irdma: Fix KASAN issue with taskletMike Marciniszyn1-0/+7
KASAN testing revealed the following issue assocated with freeing an IRQ. [50006.466686] Call Trace: [50006.466691] <IRQ> [50006.489538] dump_stack+0x5c/0x80 [50006.493475] print_address_description.constprop.6+0x1a/0x150 [50006.499872] ? irdma_sc_process_ceq+0x483/0x790 [irdma] [50006.505742] ? irdma_sc_process_ceq+0x483/0x790 [irdma] [50006.511644] kasan_report.cold.11+0x7f/0x118 [50006.516572] ? irdma_sc_process_ceq+0x483/0x790 [irdma] [50006.522473] irdma_sc_process_ceq+0x483/0x790 [irdma] [50006.528232] irdma_process_ceq+0xb2/0x400 [irdma] [50006.533601] ? irdma_hw_flush_wqes_callback+0x370/0x370 [irdma] [50006.540298] irdma_ceq_dpc+0x44/0x100 [irdma] [50006.545306] tasklet_action_common.isra.14+0x148/0x2c0 [50006.551096] __do_softirq+0x1d0/0xaf8 [50006.555396] irq_exit_rcu+0x219/0x260 [50006.559670] irq_exit+0xa/0x20 [50006.563320] smp_apic_timer_interrupt+0x1bf/0x690 [50006.568645] apic_timer_interrupt+0xf/0x20 [50006.573341] </IRQ> The issue is that a tasklet could be pending on another core racing the delete of the irq. Fix by insuring any scheduled tasklet is killed after deleting the irq. Fixes: 44d9e52977a1 ("RDMA/irdma: Implement device initialization definitions") Signed-off-by: Mike Marciniszyn <[email protected]> Signed-off-by: Shiraz Saleem <[email protected]> Signed-off-by: Sindhu Devale <[email protected]> Link: https://lore.kernel.org/r/[email protected] Signed-off-by: Leon Romanovsky <[email protected]>
2024-01-31RDMA/efa: Limit EQs to available MSI-X vectorsYonatan Nachum2-17/+16
When creating EQs we take into consideration the max number of EQs the device reported it can support and the number of available CPUs. There are situations where the number of EQs the device reported it can support and the PCI configuration of MSI-X is different, take it in account as well when creating EQs. Also request at least 1 MSI-X vector for the management queue and allow the kernel to return a number of vectors in a range between 1 and the max supported MSI-X vectors according to the PCI config. Reviewed-by: Michael Margolin <[email protected]> Reviewed-by: Yonatan Goldhirsh <[email protected]> Signed-off-by: Yonatan Nachum <[email protected]> Link: https://lore.kernel.org/r/[email protected] Signed-off-by: Leon Romanovsky <[email protected]>
2024-01-31RDMA/mlx5: Relax DEVX access upon modify commandsYishai Hadas1-1/+1
Relax DEVX access upon modify commands to be UVERBS_ACCESS_READ. The kernel doesn't need to protect what firmware protects, or what causes no damage to anyone but the user. As firmware needs to protect itself from parallel access to the same object, don't block parallel modify/query commands on the same object in the kernel side. This change will allow user space application to run parallel updates to different entries in the same bulk object. Tested-by: Tamar Mashiah <[email protected]> Signed-off-by: Yishai Hadas <[email protected]> Reviewed-by: Michael Guralnik <[email protected]> Link: https://lore.kernel.org/r/7407d5ed35dc427c1097699e12b49c01e1073406.1706433934.git.leon@kernel.org Signed-off-by: Leon Romanovsky <[email protected]>
2024-01-31IB/mlx5: Don't expose debugfs entries for RRoCE general parameters if not ↵Mark Zhang1-0/+6
supported debugfs entries for RRoCE general CC parameters must be exposed only when they are supported, otherwise when accessing them there may be a syndrome error in kernel log, for example: $ cat /sys/kernel/debug/mlx5/0000:08:00.1/cc_params/rtt_resp_dscp cat: '/sys/kernel/debug/mlx5/0000:08:00.1/cc_params/rtt_resp_dscp': Invalid argument $ dmesg mlx5_core 0000:08:00.1: mlx5_cmd_out_err:805:(pid 1253): QUERY_CONG_PARAMS(0x824) op_mod(0x0) failed, status bad parameter(0x3), syndrome (0x325a82), err(-22) Fixes: 66fb1d5df6ac ("IB/mlx5: Extend debug control for CC parameters") Reviewed-by: Edward Srouji <[email protected]> Signed-off-by: Mark Zhang <[email protected]> Link: https://lore.kernel.org/r/e7ade70bad52b7468bdb1de4d41d5fad70c8b71c.1706433934.git.leon@kernel.org Signed-off-by: Leon Romanovsky <[email protected]>
2024-01-31RDMA/mlx5: Fix fortify source warning while accessing Eth segmentLeon Romanovsky1-1/+1
------------[ cut here ]------------ memcpy: detected field-spanning write (size 56) of single field "eseg->inline_hdr.start" at /var/lib/dkms/mlnx-ofed-kernel/5.8/build/drivers/infiniband/hw/mlx5/wr.c:131 (size 2) WARNING: CPU: 0 PID: 293779 at /var/lib/dkms/mlnx-ofed-kernel/5.8/build/drivers/infiniband/hw/mlx5/wr.c:131 mlx5_ib_post_send+0x191b/0x1a60 [mlx5_ib] Modules linked in: 8021q garp mrp stp llc rdma_ucm(OE) rdma_cm(OE) iw_cm(OE) ib_ipoib(OE) ib_cm(OE) ib_umad(OE) mlx5_ib(OE) ib_uverbs(OE) ib_core(OE) mlx5_core(OE) pci_hyperv_intf mlxdevm(OE) mlx_compat(OE) tls mlxfw(OE) psample nft_fib_inet nft_fib_ipv4 nft_fib_ipv6 nft_fib nft_reject_inet nf_reject_ipv4 nf_reject_ipv6 nft_reject nft_ct nft_chain_nat nf_nat nf_conntrack nf_defrag_ipv6 nf_defrag_ipv4 ip_set nf_tables libcrc32c nfnetlink mst_pciconf(OE) knem(OE) vfio_pci vfio_pci_core vfio_iommu_type1 vfio iommufd irqbypass cuse nfsv3 nfs fscache netfs xfrm_user xfrm_algo ipmi_devintf ipmi_msghandler binfmt_misc crct10dif_pclmul crc32_pclmul polyval_clmulni polyval_generic ghash_clmulni_intel sha512_ssse3 snd_pcsp aesni_intel crypto_simd cryptd snd_pcm snd_timer joydev snd soundcore input_leds serio_raw evbug nfsd auth_rpcgss nfs_acl lockd grace sch_fq_codel sunrpc drm efi_pstore ip_tables x_tables autofs4 psmouse virtio_net net_failover failover floppy [last unloaded: mlx_compat(OE)] CPU: 0 PID: 293779 Comm: ssh Tainted: G OE 6.2.0-32-generic #32~22.04.1-Ubuntu Hardware name: Red Hat KVM, BIOS 0.5.1 01/01/2011 RIP: 0010:mlx5_ib_post_send+0x191b/0x1a60 [mlx5_ib] Code: 0c 01 00 a8 01 75 25 48 8b 75 a0 b9 02 00 00 00 48 c7 c2 10 5b fd c0 48 c7 c7 80 5b fd c0 c6 05 57 0c 03 00 01 e8 95 4d 93 da <0f> 0b 44 8b 4d b0 4c 8b 45 c8 48 8b 4d c0 e9 49 fb ff ff 41 0f b7 RSP: 0018:ffffb5b48478b570 EFLAGS: 00010046 RAX: 0000000000000000 RBX: 0000000000000001 RCX: 0000000000000000 RDX: 0000000000000000 RSI: 0000000000000000 RDI: 0000000000000000 RBP: ffffb5b48478b628 R08: 0000000000000000 R09: 0000000000000000 R10: 0000000000000000 R11: 0000000000000000 R12: ffffb5b48478b5e8 R13: ffff963a3c609b5e R14: ffff9639c3fbd800 R15: ffffb5b480475a80 FS: 00007fc03b444c80(0000) GS:ffff963a3dc00000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 0000556f46bdf000 CR3: 0000000006ac6003 CR4: 00000000003706f0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 Call Trace: <TASK> ? show_regs+0x72/0x90 ? mlx5_ib_post_send+0x191b/0x1a60 [mlx5_ib] ? __warn+0x8d/0x160 ? mlx5_ib_post_send+0x191b/0x1a60 [mlx5_ib] ? report_bug+0x1bb/0x1d0 ? handle_bug+0x46/0x90 ? exc_invalid_op+0x19/0x80 ? asm_exc_invalid_op+0x1b/0x20 ? mlx5_ib_post_send+0x191b/0x1a60 [mlx5_ib] mlx5_ib_post_send_nodrain+0xb/0x20 [mlx5_ib] ipoib_send+0x2ec/0x770 [ib_ipoib] ipoib_start_xmit+0x5a0/0x770 [ib_ipoib] dev_hard_start_xmit+0x8e/0x1e0 ? validate_xmit_skb_list+0x4d/0x80 sch_direct_xmit+0x116/0x3a0 __dev_xmit_skb+0x1fd/0x580 __dev_queue_xmit+0x284/0x6b0 ? _raw_spin_unlock_irq+0xe/0x50 ? __flush_work.isra.0+0x20d/0x370 ? push_pseudo_header+0x17/0x40 [ib_ipoib] neigh_connected_output+0xcd/0x110 ip_finish_output2+0x179/0x480 ? __smp_call_single_queue+0x61/0xa0 __ip_finish_output+0xc3/0x190 ip_finish_output+0x2e/0xf0 ip_output+0x78/0x110 ? __pfx_ip_finish_output+0x10/0x10 ip_local_out+0x64/0x70 __ip_queue_xmit+0x18a/0x460 ip_queue_xmit+0x15/0x30 __tcp_transmit_skb+0x914/0x9c0 tcp_write_xmit+0x334/0x8d0 tcp_push_one+0x3c/0x60 tcp_sendmsg_locked+0x2e1/0xac0 tcp_sendmsg+0x2d/0x50 inet_sendmsg+0x43/0x90 sock_sendmsg+0x68/0x80 sock_write_iter+0x93/0x100 vfs_write+0x326/0x3c0 ksys_write+0xbd/0xf0 ? do_syscall_64+0x69/0x90 __x64_sys_write+0x19/0x30 do_syscall_64+0x59/0x90 ? do_user_addr_fault+0x1d0/0x640 ? exit_to_user_mode_prepare+0x3b/0xd0 ? irqentry_exit_to_user_mode+0x9/0x20 ? irqentry_exit+0x43/0x50 ? exc_page_fault+0x92/0x1b0 entry_SYSCALL_64_after_hwframe+0x72/0xdc RIP: 0033:0x7fc03ad14a37 Code: 10 00 f7 d8 64 89 02 48 c7 c0 ff ff ff ff eb b7 0f 1f 00 f3 0f 1e fa 64 8b 04 25 18 00 00 00 85 c0 75 10 b8 01 00 00 00 0f 05 <48> 3d 00 f0 ff ff 77 51 c3 48 83 ec 28 48 89 54 24 18 48 89 74 24 RSP: 002b:00007ffdf8697fe8 EFLAGS: 00000246 ORIG_RAX: 0000000000000001 RAX: ffffffffffffffda RBX: 0000000000008024 RCX: 00007fc03ad14a37 RDX: 0000000000008024 RSI: 0000556f46bd8270 RDI: 0000000000000003 RBP: 0000556f46bb1800 R08: 0000000000007fe3 R09: 0000000000000000 R10: 0000000000000000 R11: 0000000000000246 R12: 0000000000000002 R13: 0000556f46bc66b0 R14: 000000000000000a R15: 0000556f46bb2f50 </TASK> ---[ end trace 0000000000000000 ]--- Link: https://lore.kernel.org/r/8228ad34bd1a25047586270f7b1fb4ddcd046282.1706433934.git.leon@kernel.org Signed-off-by: Leon Romanovsky <[email protected]>
2024-01-25RDMA/mlx5: Delete unused mlx5_ib_copy_pas prototypeAlexey Dobriyan1-1/+0
mlx5_ib_copy_pas() doesn't exist anymore. Signed-off-by: Alexey Dobriyan <[email protected]> Link: https://lore.kernel.org/r/a2cb861e-d11e-4567-8a73-73763d1dc199@p183 Signed-off-by: Leon Romanovsky <[email protected]>
2024-01-25RDMA/cxgb4: Delete unused c4iw_ep_redirect prototypeAlexey Dobriyan1-2/+0
c4iw_ep_redirect() doesn't exist. Signed-off-by: Alexey Dobriyan <[email protected]> Link: https://lore.kernel.org/r/a67881e7-469b-43d9-9973-ad7579eb2c4b@p183 Signed-off-by: Leon Romanovsky <[email protected]>
2024-01-25RDMA/bnxt_re: Add a missing check in bnxt_qplib_query_srqKalesh AP1-1/+2
Before populating the response, driver has to check the status of HWRM command. Fixes: 37cb11acf1f7 ("RDMA/bnxt_re: Add SRQ support for Broadcom adapters") Signed-off-by: Kalesh AP <[email protected]> Signed-off-by: Selvin Xavier <[email protected]> Link: https://lore.kernel.org/r/[email protected] Signed-off-by: Leon Romanovsky <[email protected]>
2024-01-25RDMA/bnxt_re: Return error for SRQ resizeKalesh AP1-3/+2
SRQ resize is not supported in the driver. But driver is not returning error from bnxt_re_modify_srq() for SRQ resize. Fixes: 37cb11acf1f7 ("RDMA/bnxt_re: Add SRQ support for Broadcom adapters") Signed-off-by: Kalesh AP <[email protected]> Signed-off-by: Selvin Xavier <[email protected]> Link: https://lore.kernel.org/r/[email protected] Signed-off-by: Leon Romanovsky <[email protected]>
2024-01-25RDMA/bnxt_re: Fix unconditional fence for newer adaptersKalesh AP1-12/+16
Older adapters required an unconditional fence for non-wire memory operations. Newer adapters doesn't require this and therefore, disabling the unconditional fence. Fixes: 1801d87b3598 ("RDMA/bnxt_re: Support new 5760X P7 devices") Signed-off-by: Kashyap Desai <[email protected]> Signed-off-by: Kalesh AP <[email protected]> Signed-off-by: Selvin Xavier <[email protected]> Link: https://lore.kernel.org/r/[email protected] Signed-off-by: Leon Romanovsky <[email protected]>
2024-01-25RDMA/bnxt_re: Remove a redundant check inside bnxt_re_vf_res_configKalesh AP1-3/+0
After the cited commit, there is no possibility that this check can return true. Remove it. Fixes: a43c26fa2e6c ("RDMA/bnxt_re: Remove the sriov config callback") Signed-off-by: Kalesh AP <[email protected]> Signed-off-by: Selvin Xavier <[email protected]> Link: https://lore.kernel.org/r/[email protected] Signed-off-by: Leon Romanovsky <[email protected]>
2024-01-25RDMA/bnxt_re: Avoid creating fence MR for newer adaptersKalesh AP1-0/+10
Limit the usage of fence MR to adapters older than Gen P5 products. Fixes: 1801d87b3598 ("RDMA/bnxt_re: Support new 5760X P7 devices") Signed-off-by: Kashyap Desai <[email protected]> Signed-off-by: Bhargava Chenna Marreddy <[email protected]> Signed-off-by: Kalesh AP <[email protected]> Signed-off-by: Selvin Xavier <[email protected]> Link: https://lore.kernel.org/r/[email protected] Signed-off-by: Leon Romanovsky <[email protected]>
2024-01-25RDMA/mana_ib: Introduce mana_ib_install_cq_cb helper functionKonstantin Taranov3-26/+27
Use a helper function to install callbacks to CQs. This patch removes code repetition. Signed-off-by: Konstantin Taranov <[email protected]> Link: https://lore.kernel.org/r/[email protected] Signed-off-by: Leon Romanovsky <[email protected]>
2024-01-25RDMA/mana_ib: Introduce mana_ib_get_netdev helper functionKonstantin Taranov3-41/+32
Use a helper function to access netdevs using a port number. This patch removes code repetitions as well as removes the need to explicitly use gdma_dev, which was error-prone. Signed-off-by: Konstantin Taranov <[email protected]> Link: https://lore.kernel.org/r/[email protected] Signed-off-by: Leon Romanovsky <[email protected]>
2024-01-25RDMA/mana_ib: Introduce mdev_to_gc helper functionKonstantin Taranov5-37/+27
Use a helper function to access gdma_context from mana_ib_dev. This patch removes code repetitions as well as removes the need to explicitly use gdma_dev, which was error-prone. Signed-off-by: Konstantin Taranov <[email protected]> Link: https://lore.kernel.org/r/[email protected] Signed-off-by: Leon Romanovsky <[email protected]>
2024-01-25IB/hfi1: Fix a memleak in init_credit_returnZhipeng Lu1-1/+5
When dma_alloc_coherent fails to allocate dd->cr_base[i].va, init_credit_return should deallocate dd->cr_base and dd->cr_base[i] that allocated before. Or those resources would be never freed and a memleak is triggered. Fixes: 7724105686e7 ("IB/hfi1: add driver files") Signed-off-by: Zhipeng Lu <[email protected]> Link: https://lore.kernel.org/r/[email protected] Acked-by: Dennis Dalessandro <[email protected]> Signed-off-by: Leon Romanovsky <[email protected]>
2024-01-25RDMA/hns: Simplify 'struct hns_roce_hem' allocationYunsheng Lin3-136/+24
'struct hns_roce_hem' is used to refer to the last level of dma buffer managed by the hw, pointed by a single BA(base address) in the previous level of BT(base table), so the dma buffer in 'struct hns_roce_hem' must be contiguous. Right now the size of dma buffer in 'struct hns_roce_hem' is decided by mhop->buf_chunk_size in get_hem_table_config(), which ensure the mhop->buf_chunk_size is power of two of PAGE_SIZE, so there will be only one contiguous dma buffer allocated in hns_roce_alloc_hem(), which means hem->chunk_list and chunk->mem for linking multi dma buffers is unnecessary. This patch removes the hem->chunk_list and chunk->mem and other related macro and function accordingly. Signed-off-by: Yunsheng Lin <[email protected]> Signed-off-by: Junxian Huang <[email protected]> Link: https://lore.kernel.org/r/[email protected] Signed-off-by: Leon Romanovsky <[email protected]>
2024-01-25RDMA/hns: Support adaptive PBL hopnumChengchang Tang1-3/+54
In the current implementation, a fixed addressing level is used for PBL. But in fact, the necessary addressing level is related to page size and the size of MR. This patch calculates the addressing level according to page size and the size of MR, and uses the addressing level to configure the PBL. Signed-off-by: Chengchang Tang <[email protected]> Signed-off-by: Junxian Huang <[email protected]> Link: https://lore.kernel.org/r/[email protected] Signed-off-by: Leon Romanovsky <[email protected]>
2024-01-25RDMA/hns: Support flexible umem page sizeChengchang Tang2-38/+92
In the current implementation, a fixed page size is used to configure the umem PBL, which is not flexible enough and is not conducive to the performance of the HW. Find a best page size to get better performance. Signed-off-by: Chengchang Tang <[email protected]> Signed-off-by: Junxian Huang <[email protected]> Link: https://lore.kernel.org/r/[email protected] Signed-off-by: Leon Romanovsky <[email protected]>
2024-01-25RDMA/hns: Alloc MTR memory before alloc_mtt()Chengchang Tang1-20/+27
MTR memory allocation do not depend on allocation of mtt. This patch moves the allocation of mtr before mtt in preparation for the following optimization. Signed-off-by: Chengchang Tang <[email protected]> Signed-off-by: Junxian Huang <[email protected]> Link: https://lore.kernel.org/r/[email protected] Signed-off-by: Leon Romanovsky <[email protected]>
2024-01-25RDMA/hns: Refactor mtr_init_buf_cfg()Chengchang Tang1-31/+45
page_shift and page_cnt is only used in mtr_map_bufs(). And these parameter could be calculated indepedently. Strip the computation of page_shift and page_cnt from mtr_init_buf_cfg(), reducing the number of parameters of it. This helps reducing coupling between mtr_init_buf_cfg() and mtr_map_bufs(). Signed-off-by: Chengchang Tang <[email protected]> Signed-off-by: Junxian Huang <[email protected]> Link: https://lore.kernel.org/r/[email protected] Signed-off-by: Leon Romanovsky <[email protected]>
2024-01-25RDMA/hns: Refactor mtr findChengchang Tang4-85/+121
hns_roce_mtr_find() is a collection of multiple functions, and the return value is also difficult to understand, which is not conducive to modification and maintenance. Separate the function of obtaining MTR root BA from this function. And some adjustments has been made to improve readability. Signed-off-by: Chengchang Tang <[email protected]> Signed-off-by: Junxian Huang <[email protected]> Link: https://lore.kernel.org/r/[email protected] Signed-off-by: Leon Romanovsky <[email protected]>
2024-01-25IB/hfi1: fix spellos and kernel-docRandy Dunlap1-12/+13
Fix spelling mistakes as reported by codespell. Fix kernel-doc warnings. Signed-off-by: Randy Dunlap <[email protected]> Cc: Dennis Dalessandro <[email protected]> Cc: Jason Gunthorpe <[email protected]> Cc: Leon Romanovsky <[email protected]> Cc: [email protected] Link: https://lore.kernel.org/r/[email protected] Signed-off-by: Leon Romanovsky <[email protected]>
Powered by cgit, Gruvboxed by Blaster4385.