summaryrefslogtreecommitdiff
path: root/drivers/infiniband
AgeCommit message (Collapse)AuthorFilesLines
2021-12-17RDMA/irdma: Don't arm the CQ more than two times if no CE for this CQTatyana Nikolova5-5/+38
[ Upstream commit 10467ce09fefa2e74359f5b2ab1efb8909402f19 ] Completion events (CEs) are lost if the application is allowed to arm the CQ more than two times when no new CE for this CQ has been generated by the HW. Check if arming has been done for the CQ and if not, arm the CQ for any event otherwise promote to arm the CQ for any event only when the last arm event was solicited. Fixes: b48c24c2d710 ("RDMA/irdma: Implement device supported verb APIs") Link: https://lore.kernel.org/r/20211201231509.1930-2-shiraz.saleem@intel.com 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@nvidia.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
2021-12-17RDMA/irdma: Report correct WC errorsShiraz Saleem1-1/+4
[ Upstream commit 25b5d6fd6d13b2de3780a0ae247befc43c4576fe ] Return IBV_WC_REM_OP_ERR for responder QP errors instead of IBV_WC_REM_ACCESS_ERR. Return IBV_WC_LOC_QP_OP_ERR for errors detected on the SQ with bad opcodes Fixes: 44d9e52977a1 ("RDMA/irdma: Implement device initialization definitions") Link: https://lore.kernel.org/r/20211201231509.1930-1-shiraz.saleem@intel.com Signed-off-by: Shiraz Saleem <shiraz.saleem@intel.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
2021-12-17RDMA/irdma: Fix a potential memory allocation issue in ↵Christophe JAILLET3-12/+4
'irdma_prm_add_pble_mem()' [ Upstream commit 117697cc935b0ab04ec66274d8e64ccfebd7d0d2 ] 'pchunk->bitmapbuf' is a bitmap. Its size (in number of bits) is stored in 'pchunk->sizeofbitmap'. When it is allocated, the size (in bytes) is computed by: size_in_bits >> 3 There are 2 issues (numbers bellow assume that longs are 64 bits): - there is no guarantee here that 'pchunk->bitmapmem.size' is modulo BITS_PER_LONG but bitmaps are stored as longs (sizeofbitmap=8 bits will only allocate 1 byte, instead of 8 (1 long)) - the number of bytes is computed with a shift, not a round up, so we may allocate less memory than needed (sizeofbitmap=65 bits will only allocate 8 bytes (i.e. 1 long), when 2 longs are needed = 16 bytes) Fix both issues by using 'bitmap_zalloc()' and remove the useless 'bitmapmem' from 'struct irdma_chunk'. While at it, remove some useless NULL test before calling kfree/bitmap_free. Fixes: 915cc7ac0f8e ("RDMA/irdma: Add miscellaneous utility definitions") Link: https://lore.kernel.org/r/5e670b640508e14b1869c3e8e4fb970d78cbe997.1638692171.git.christophe.jaillet@wanadoo.fr Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr> Reviewed-by: Shiraz Saleem <shiraz.saleem@intel.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
2021-12-17RDMA/irdma: Fix a user-after-free in add_pble_prmShiraz Saleem1-1/+1
[ Upstream commit 1e11a39a82e95ce86f849f40dda0d9c0498cebd9 ] When irdma_hmc_sd_one fails, 'chunk' is freed while its still on the PBLE info list. Add the chunk entry to the PBLE info list only after successful setting of the SD in irdma_hmc_sd_one. Fixes: e8c4dbc2fcac ("RDMA/irdma: Add PBLE resource manager") Link: https://lore.kernel.org/r/20211207152135.2192-1-shiraz.saleem@intel.com Reported-by: Dan Carpenter <dan.carpenter@oracle.com> Signed-off-by: Shiraz Saleem <shiraz.saleem@intel.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
2021-12-17RDMA/mlx5: Fix releasing unallocated memory in dereg MR flowAlaa Hleihel2-17/+15
[ Upstream commit f0ae4afe3d35e67db042c58a52909e06262b740f ] For the case of IB_MR_TYPE_DM the mr does doesn't have a umem, even though it is a user MR. This causes function mlx5_free_priv_descs() to think that it is a kernel MR, leading to wrongly accessing mr->descs that will get wrong values in the union which leads to attempt to release resources that were not allocated in the first place. For example: DMA-API: mlx5_core 0000:08:00.1: device driver tries to free DMA memory it has not allocated [device address=0x0000000000000000] [size=0 bytes] WARNING: CPU: 8 PID: 1021 at kernel/dma/debug.c:961 check_unmap+0x54f/0x8b0 RIP: 0010:check_unmap+0x54f/0x8b0 Call Trace: debug_dma_unmap_page+0x57/0x60 mlx5_free_priv_descs+0x57/0x70 [mlx5_ib] mlx5_ib_dereg_mr+0x1fb/0x3d0 [mlx5_ib] ib_dereg_mr_user+0x60/0x140 [ib_core] uverbs_destroy_uobject+0x59/0x210 [ib_uverbs] uobj_destroy+0x3f/0x80 [ib_uverbs] ib_uverbs_cmd_verbs+0x435/0xd10 [ib_uverbs] ? uverbs_finalize_object+0x50/0x50 [ib_uverbs] ? lock_acquire+0xc4/0x2e0 ? lock_acquired+0x12/0x380 ? lock_acquire+0xc4/0x2e0 ? lock_acquire+0xc4/0x2e0 ? ib_uverbs_ioctl+0x7c/0x140 [ib_uverbs] ? lock_release+0x28a/0x400 ib_uverbs_ioctl+0xc0/0x140 [ib_uverbs] ? ib_uverbs_ioctl+0x7c/0x140 [ib_uverbs] __x64_sys_ioctl+0x7f/0xb0 do_syscall_64+0x38/0x90 Fix it by reorganizing the dereg flow and mlx5_ib_mr structure: - Move the ib_umem field into the user MRs structure in the union as it's applicable only there. - Function mlx5_ib_dereg_mr() will now call mlx5_free_priv_descs() only in case there isn't udata, which indicates that this isn't a user MR. Fixes: f18ec4223117 ("RDMA/mlx5: Use a union inside mlx5_ib_mr") Link: https://lore.kernel.org/r/66bb1dd253c1fd7ceaa9fc411061eefa457b86fb.1637581144.git.leonro@nvidia.com Signed-off-by: Alaa Hleihel <alaa@nvidia.com> Signed-off-by: Leon Romanovsky <leonro@nvidia.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
2021-12-17RDMA: Fix use-after-free in rxe_queue_cleanupPavel Skripkin1-0/+1
[ Upstream commit 84b01721e8042cdd1e8ffeb648844a09cd4213e0 ] On error handling path in rxe_qp_from_init() qp->sq.queue is freed and then rxe_create_qp() will drop last reference to this object. qp clean up function will try to free this queue one time and it causes UAF bug. Fix it by zeroing queue pointer after freeing queue in rxe_qp_from_init(). Fixes: 514aee660df4 ("RDMA: Globally allocate and release QP memory") Link: https://lore.kernel.org/r/20211121202239.3129-1-paskripkin@gmail.com Reported-by: syzbot+aab53008a5adf26abe91@syzkaller.appspotmail.com Signed-off-by: Pavel Skripkin <paskripkin@gmail.com> Reviewed-by: Zhu Yanjun <zyjzyj2000@gmail.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
2021-12-14RDMA/hns: Do not destroy QP resources in the hw resetting phaseYangyang Li1-1/+11
commit b0969f83890bf8b47f5c8bd42539599b2b52fdeb upstream. When hns_roce_v2_destroy_qp() is called, the brief calling process of the driver is as follows: ...... hns_roce_v2_destroy_qp hns_roce_v2_qp_modify hns_roce_cmd_mbox hns_roce_qp_destroy If hns_roce_cmd_mbox() detects that the hardware is being reset during the execution of the hns_roce_cmd_mbox(), the driver will not be able to get the return value from the hardware (the firmware cannot respond to the driver's mailbox during the hardware reset phase). The driver needs to wait for the hardware reset to complete before continuing to execute hns_roce_qp_destroy(), otherwise it may happen that the driver releases the resources but the hardware is still accessing. In order to fix this problem, HNS RoCE needs to add a piece of code to wait for the hardware reset to complete. The original interface get_hw_reset_stat() is the instantaneous state of the hardware reset, which cannot accurately reflect whether the hardware reset is completed, so it needs to be replaced with the ae_dev_reset_cnt interface. The sign that the hardware reset is complete is that the return value of the ae_dev_reset_cnt interface is greater than the original value reset_cnt recorded by the driver. Fixes: 6a04aed6afae ("RDMA/hns: Fix the chip hanging caused by sending mailbox&CMQ during reset") Link: https://lore.kernel.org/r/20211123142402.26936-1-liangwenpeng@huawei.com Signed-off-by: Yangyang Li <liyangyang20@huawei.com> Signed-off-by: Wenpeng Liang <liangwenpeng@huawei.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2021-12-14RDMA/hns: Do not halt commands during reset until laterYangyang Li1-2/+0
commit 52414e27d6b568120b087d1fbafbb4482b0ccaab upstream. is_reset is used to indicate whether the hardware starts to reset. When hns_roce_hw_v2_reset_notify_down() is called, the hardware has not yet started to reset. If is_reset is set at this time, all mailbox operations of resource destroy actions will be intercepted by driver. When the driver cleans up resources, but the hardware is still accessed, the following errors will appear: arm-smmu-v3 arm-smmu-v3.2.auto: event 0x10 received: arm-smmu-v3 arm-smmu-v3.2.auto: 0x0000350100000010 arm-smmu-v3 arm-smmu-v3.2.auto: 0x000002088000003f arm-smmu-v3 arm-smmu-v3.2.auto: 0x00000000a50e0800 arm-smmu-v3 arm-smmu-v3.2.auto: 0x0000000000000000 arm-smmu-v3 arm-smmu-v3.2.auto: event 0x10 received: arm-smmu-v3 arm-smmu-v3.2.auto: 0x0000350100000010 arm-smmu-v3 arm-smmu-v3.2.auto: 0x000002088000043e arm-smmu-v3 arm-smmu-v3.2.auto: 0x00000000a50a0800 arm-smmu-v3 arm-smmu-v3.2.auto: 0x0000000000000000 arm-smmu-v3 arm-smmu-v3.2.auto: event 0x10 received: arm-smmu-v3 arm-smmu-v3.2.auto: 0x0000350100000010 arm-smmu-v3 arm-smmu-v3.2.auto: 0x0000020880000436 arm-smmu-v3 arm-smmu-v3.2.auto: 0x00000000a50a0880 arm-smmu-v3 arm-smmu-v3.2.auto: 0x0000000000000000 arm-smmu-v3 arm-smmu-v3.2.auto: event 0x10 received: arm-smmu-v3 arm-smmu-v3.2.auto: 0x0000350100000010 arm-smmu-v3 arm-smmu-v3.2.auto: 0x000002088000043a arm-smmu-v3 arm-smmu-v3.2.auto: 0x00000000a50e0840 hns3 0000:35:00.0: INT status: CMDQ(0x0) HW errors(0x0) other(0x0) arm-smmu-v3 arm-smmu-v3.2.auto: 0x0000000000000000 hns3 0000:35:00.0: received unknown or unhandled event of vector0 arm-smmu-v3 arm-smmu-v3.2.auto: event 0x10 received: arm-smmu-v3 arm-smmu-v3.2.auto: 0x0000350100000010 {34}[Hardware Error]: Hardware error from APEI Generic Hardware Error Source: 7 is_reset will be set correctly in check_aedev_reset_status(), so the setting in hns_roce_hw_v2_reset_notify_down() should be deleted. Fixes: 726be12f5ca0 ("RDMA/hns: Set reset flag when hw resetting") Link: https://lore.kernel.org/r/20211123084809.37318-1-liangwenpeng@huawei.com Signed-off-by: Yangyang Li <liyangyang20@huawei.com> Signed-off-by: Wenpeng Liang <liangwenpeng@huawei.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2021-12-14IB/hfi1: Correct guard on eager buffer deallocationMike Marciniszyn1-1/+1
commit 9292f8f9a2ac42eb320bced7153aa2e63d8cc13a upstream. The code tests the dma address which legitimately can be 0. The code should test the kernel logical address to avoid leaking eager buffer allocations that happen to map to a dma address of 0. Fixes: 60368186fd85 ("IB/hfi1: Fix user-space buffers mapping with IOMMU enabled") Link: https://lore.kernel.org/r/20211129191952.101968.17137.stgit@awfm-01.cornelisnetworks.com Signed-off-by: Mike Marciniszyn <mike.marciniszyn@cornelisnetworks.com> Signed-off-by: Dennis Dalessandro <dennis.dalessandro@cornelisnetworks.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2021-12-14IB/hfi1: Fix leak of rcvhdrtail_dummy_kvaddrMike Marciniszyn1-19/+14
commit 60a8b5a1611b4a26de4839ab9c1fc2a9cf3e17c1 upstream. This buffer is currently allocated in hfi1_init(): if (reinit) ret = init_after_reset(dd); else ret = loadtime_init(dd); if (ret) goto done; /* allocate dummy tail memory for all receive contexts */ dd->rcvhdrtail_dummy_kvaddr = dma_alloc_coherent(&dd->pcidev->dev, sizeof(u64), &dd->rcvhdrtail_dummy_dma, GFP_KERNEL); if (!dd->rcvhdrtail_dummy_kvaddr) { dd_dev_err(dd, "cannot allocate dummy tail memory\n"); ret = -ENOMEM; goto done; } The reinit triggered path will overwrite the old allocation and leak it. Fix by moving the allocation to hfi1_alloc_devdata() and the deallocation to hfi1_free_devdata(). Link: https://lore.kernel.org/r/20211129192008.101968.91302.stgit@awfm-01.cornelisnetworks.com Cc: stable@vger.kernel.org Fixes: 46b010d3eeb8 ("staging/rdma/hfi1: Workaround to prevent corruption during packet delivery") Signed-off-by: Mike Marciniszyn <mike.marciniszyn@cornelisnetworks.com> Signed-off-by: Dennis Dalessandro <dennis.dalessandro@cornelisnetworks.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2021-12-14IB/hfi1: Fix early init panicMike Marciniszyn3-3/+6
commit f6a3cfec3c01f9983e961c3327cef0db129a3c43 upstream. The following trace can be observed with an init failure such as firmware load failures: BUG: unable to handle kernel NULL pointer dereference at 0000000000000000 PGD 0 P4D 0 Oops: 0010 [#1] SMP PTI CPU: 0 PID: 537 Comm: kworker/0:3 Tainted: G OE --------- - - 4.18.0-240.el8.x86_64 #1 Workqueue: events work_for_cpu_fn RIP: 0010:0x0 Code: Bad RIP value. RSP: 0000:ffffae5f878a3c98 EFLAGS: 00010046 RAX: 0000000000000000 RBX: ffff95e48e025c00 RCX: 0000000000000000 RDX: 0000000000000001 RSI: 0000000000000000 RDI: ffff95e48e025c00 RBP: ffff95e4bf3660a4 R08: 0000000000000000 R09: ffffffff86d5e100 R10: ffff95e49e1de600 R11: 0000000000000001 R12: ffff95e4bf366180 R13: ffff95e48e025c00 R14: ffff95e4bf366028 R15: ffff95e4bf366000 FS: 0000000000000000(0000) GS:ffff95e4df200000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: ffffffffffffffd6 CR3: 0000000f86a0a003 CR4: 00000000001606f0 Call Trace: receive_context_interrupt+0x1f/0x40 [hfi1] __free_irq+0x201/0x300 free_irq+0x2e/0x60 pci_free_irq+0x18/0x30 msix_free_irq.part.2+0x46/0x80 [hfi1] msix_clean_up_interrupts+0x2b/0x70 [hfi1] hfi1_init_dd+0x640/0x1a90 [hfi1] do_init_one.isra.19+0x34d/0x680 [hfi1] local_pci_probe+0x41/0x90 work_for_cpu_fn+0x16/0x20 process_one_work+0x1a7/0x360 worker_thread+0x1cf/0x390 ? create_worker+0x1a0/0x1a0 kthread+0x112/0x130 ? kthread_flush_work_fn+0x10/0x10 ret_from_fork+0x35/0x40 The free_irq() results in a callback to the registered interrupt handler, and rcd->do_interrupt is NULL because the receive context data structures are not fully initialized. Fix by ensuring that the do_interrupt is always assigned and adding a guards in the slow path handler to detect and handle a partially initialized receive context and noop the receive. Link: https://lore.kernel.org/r/20211129192003.101968.33612.stgit@awfm-01.cornelisnetworks.com Cc: stable@vger.kernel.org Fixes: b0ba3c18d6bf ("IB/hfi1: Move normal functions from hfi1_devdata to const array") Signed-off-by: Mike Marciniszyn <mike.marciniszyn@cornelisnetworks.com> Signed-off-by: Dennis Dalessandro <dennis.dalessandro@cornelisnetworks.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2021-12-14IB/hfi1: Insure use of smp_processor_id() is preempt disabledMike Marciniszyn1-1/+1
commit b6d57e24ce6cc3df8a8845e1b193e88a65d501b1 upstream. The following BUG has just surfaced with our 5.16 testing: BUG: using smp_processor_id() in preemptible [00000000] code: mpicheck/1581081 caller is sdma_select_user_engine+0x72/0x210 [hfi1] CPU: 0 PID: 1581081 Comm: mpicheck Tainted: G S 5.16.0-rc1+ #1 Hardware name: Intel Corporation S2600WT2R/S2600WT2R, BIOS SE5C610.86B.01.01.0016.033120161139 03/31/2016 Call Trace: <TASK> dump_stack_lvl+0x33/0x42 check_preemption_disabled+0xbf/0xe0 sdma_select_user_engine+0x72/0x210 [hfi1] ? _raw_spin_unlock_irqrestore+0x1f/0x31 ? hfi1_mmu_rb_insert+0x6b/0x200 [hfi1] hfi1_user_sdma_process_request+0xa02/0x1120 [hfi1] ? hfi1_write_iter+0xb8/0x200 [hfi1] hfi1_write_iter+0xb8/0x200 [hfi1] do_iter_readv_writev+0x163/0x1c0 do_iter_write+0x80/0x1c0 vfs_writev+0x88/0x1a0 ? recalibrate_cpu_khz+0x10/0x10 ? ktime_get+0x3e/0xa0 ? __fget_files+0x66/0xa0 do_writev+0x65/0x100 do_syscall_64+0x3a/0x80 Fix this long standing bug by moving the smp_processor_id() to after the rcu_read_lock(). The rcu_read_lock() implicitly disables preemption. Link: https://lore.kernel.org/r/20211129191958.101968.87329.stgit@awfm-01.cornelisnetworks.com Cc: stable@vger.kernel.org Fixes: 0cb2aa690c7e ("IB/hfi1: Add sysfs interface for affinity setup") Signed-off-by: Mike Marciniszyn <mike.marciniszyn@cornelisnetworks.com> Signed-off-by: Dennis Dalessandro <dennis.dalessandro@cornelisnetworks.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2021-11-25RDMA/mlx4: Do not fail the registration on port statsJack Wang1-3/+15
[ Upstream commit 378c67413de18b69fb3bb78d8c4f0f1192cfa973 ] If the FW doesn't support MLX4_DEV_CAP_FLAG2_DIAG_PER_PORT, mlx4 driver will fail the ib_setup_port_attrs, which is called from ib_register_device()/enable_device_and_get(), in the end leads to device not detected[1][2] To fix it, add a new mlx4_ib_hw_stats_ops1, w/o alloc_hw_port_stats if FW does not support MLX4_DEV_CAP_FLAG2_DIAG_PER_PORT. [1] https://bugzilla.redhat.com/show_bug.cgi?id=2014094 [2] https://lore.kernel.org/linux-rdma/CAMGffEn2wvEnmzc0xe=xYiCLqpphiHDBxCxqAELrBofbUAMQxw@mail.gmail.com Fixes: 4b5f4d3fb408 ("RDMA: Split the alloc_hw_stats() ops to port and device variants") Link: https://lore.kernel.org/r/20211115101519.27210-1-jinpu.wang@ionos.com Signed-off-by: Jack Wang <jinpu.wang@ionos.com> Reviewed-by: Leon Romanovsky <leonro@nvidia.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
2021-11-25RDMA/core: Set send and receive CQ before forwarding to the driverLeon Romanovsky1-0/+3
[ Upstream commit 6cd7397d01c4a3e09757840299e4f114f0aa5fa0 ] Preset both receive and send CQ pointers prior to call to the drivers and overwrite it later again till the mlx4 is going to be changed do not overwrite ibqp properties. This change is needed for mlx5, because in case of QP creation failure, it will go to the path of QP destroy which relies on proper CQ pointers. BUG: KASAN: use-after-free in create_qp.cold+0x164/0x16e [mlx5_ib] Write of size 8 at addr ffff8880064c55c0 by task a.out/246 CPU: 0 PID: 246 Comm: a.out Not tainted 5.15.0+ #291 Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS rel-1.13.0-0-gf21b5a4aeb02-prebuilt.qemu.org 04/01/2014 Call Trace: dump_stack_lvl+0x45/0x59 print_address_description.constprop.0+0x1f/0x140 kasan_report.cold+0x83/0xdf create_qp.cold+0x164/0x16e [mlx5_ib] mlx5_ib_create_qp+0x358/0x28a0 [mlx5_ib] create_qp.part.0+0x45b/0x6a0 [ib_core] ib_create_qp_user+0x97/0x150 [ib_core] ib_uverbs_handler_UVERBS_METHOD_QP_CREATE+0x92c/0x1250 [ib_uverbs] ib_uverbs_cmd_verbs+0x1c38/0x3150 [ib_uverbs] ib_uverbs_ioctl+0x169/0x260 [ib_uverbs] __x64_sys_ioctl+0x866/0x14d0 do_syscall_64+0x3d/0x90 entry_SYSCALL_64_after_hwframe+0x44/0xae Allocated by task 246: kasan_save_stack+0x1b/0x40 __kasan_kmalloc+0xa4/0xd0 create_qp.part.0+0x92/0x6a0 [ib_core] ib_create_qp_user+0x97/0x150 [ib_core] ib_uverbs_handler_UVERBS_METHOD_QP_CREATE+0x92c/0x1250 [ib_uverbs] ib_uverbs_cmd_verbs+0x1c38/0x3150 [ib_uverbs] ib_uverbs_ioctl+0x169/0x260 [ib_uverbs] __x64_sys_ioctl+0x866/0x14d0 do_syscall_64+0x3d/0x90 entry_SYSCALL_64_after_hwframe+0x44/0xae Freed by task 246: kasan_save_stack+0x1b/0x40 kasan_set_track+0x1c/0x30 kasan_set_free_info+0x20/0x30 __kasan_slab_free+0x10c/0x150 slab_free_freelist_hook+0xb4/0x1b0 kfree+0xe7/0x2a0 create_qp.part.0+0x52b/0x6a0 [ib_core] ib_create_qp_user+0x97/0x150 [ib_core] ib_uverbs_handler_UVERBS_METHOD_QP_CREATE+0x92c/0x1250 [ib_uverbs] ib_uverbs_cmd_verbs+0x1c38/0x3150 [ib_uverbs] ib_uverbs_ioctl+0x169/0x260 [ib_uverbs] __x64_sys_ioctl+0x866/0x14d0 do_syscall_64+0x3d/0x90 entry_SYSCALL_64_after_hwframe+0x44/0xae Fixes: 514aee660df4 ("RDMA: Globally allocate and release QP memory") Link: https://lore.kernel.org/r/2dbb2e2cbb1efb188a500e5634be1d71956424ce.1636631035.git.leonro@nvidia.com Signed-off-by: Leon Romanovsky <leonro@nvidia.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
2021-11-25RDMA/core: Use kvzalloc when allocating the struct ib_portwangyugui1-2/+2
[ Upstream commit 911a81c9c7092bfd75432ce79b2ef879127ea065 ] The 'struct attribute' flex array contains some struct lock_class_key's which become big when lockdep is turned on. Big enough that some drivers will not load when CONFIG_PROVE_LOCKING=y because they cannot allocate enough memory: WARNING: CPU: 36 PID: 8 at mm/page_alloc.c:5350 __alloc_pages+0x27e/0x3e0 Call Trace: kmalloc_order+0x2a/0xb0 kmalloc_order_trace+0x19/0xf0 __kmalloc+0x231/0x270 ib_setup_port_attrs+0xd8/0x870 [ib_core] ib_register_device+0x419/0x4e0 [ib_core] bnxt_re_task+0x208/0x2d0 [bnxt_re] Link: https://lore.kernel.org/r/20211019002656.17745-1-wangyugui@e16-tech.com Signed-off-by: wangyugui <wangyugui@e16-tech.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
2021-11-25RDMA/rxe: Separate HW and SW l/rkeysBob Pearson5-51/+81
[ Upstream commit 001345339f4ca85790a1644a74e33ae77ac116be ] Separate software and simulated hardware lkeys and rkeys for MRs and MWs. This makes struct ib_mr and struct ib_mw isolated from hardware changes triggered by executing work requests. This change fixes a bug seen in blktest. Link: https://lore.kernel.org/r/20210914164206.19768-4-rpearsonhpe@gmail.com Signed-off-by: Bob Pearson <rpearsonhpe@gmail.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
2021-11-25RDMA/bnxt_re: Check if the vlan is valid before reportingSelvin Xavier1-3/+9
[ Upstream commit 6bda39149d4b8920fdb8744090653aca3daa792d ] When VF is configured with default vlan, HW strips the vlan from the packet and driver receives it in Rx completion. VLAN needs to be reported for UD work completion only if the vlan is configured on the host. Add a check for valid vlan in the UD receive path. Link: https://lore.kernel.org/r/1631709163-2287-12-git-send-email-selvin.xavier@broadcom.com Signed-off-by: Selvin Xavier <selvin.xavier@broadcom.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
2021-11-18RDMA/core: Require the driver to set the IOVA correctly during rereg_mrAharon Landau1-3/+0
[ Upstream commit f1a090f09f42be5a5542009f0be310fdb3e768fc ] If the driver returns a new MR during rereg it has to fill it with the IOVA from the proper source. If IB_MR_REREG_TRANS is set then the IOVA is cmd.hca_va, otherwise the IOVA comes from the old MR. mlx5 for example has two calls inside rereg_mr: return create_real_mr(new_pd, umem, mr->ibmr.iova, new_access_flags); and return create_real_mr(new_pd, new_umem, iova, new_access_flags); Unconditionally overwriting the iova in the newly allocated MR will corrupt the iova if the first path is used. Remove the redundant initializations from ib_uverbs_rereg_mr(). Fixes: 6e0954b11c05 ("RDMA/uverbs: Allow drivers to create a new HW object during rereg_mr") Link: https://lore.kernel.org/r/4b0a31bbc372842613286a10d7a8cbb0ee6069c7.1635400472.git.leonro@nvidia.com Signed-off-by: Aharon Landau <aharonl@nvidia.com> Signed-off-by: Leon Romanovsky <leonro@nvidia.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
2021-11-18RDMA/hns: Modify the value of MAX_LP_MSG_LEN to meet hardware compatibilityYixing Liu1-2/+2
[ Upstream commit 0e60778efb072d47efc7100c4009b5bd97273b0b ] The upper limit of MAX_LP_MSG_LEN on HIP08 is 64K, and the upper limit on HIP09 is 16K. Regardless of whether it is HIP08 or HIP09, only 16K will be used. In order to ensure compatibility, it is unified to 16K. Setting MAX_LP_MSG_LEN to 16K will not cause performance loss on HIP08. Fixes: fbed9d2be292 ("RDMA/hns: Fix configuration of ack_req_freq in QPC") Link: https://lore.kernel.org/r/20211029100537.27299-1-liangwenpeng@huawei.com Signed-off-by: Yixing Liu <liuyixing1@huawei.com> Signed-off-by: Wenpeng Liang <liangwenpeng@huawei.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
2021-11-18RDMA/hns: Fix initial arm_st of CQHaoyue Xu1-1/+1
[ Upstream commit 571fb4fb78a3bf0fcadbe65eca9ca4ccee885af4 ] We set the init CQ status to ARMED before. As a result, an unexpected CEQE would be reported. Therefore, the init CQ status should be set to no_armed rather than REG_NXT_CEQE. Fixes: a5073d6054f7 ("RDMA/hns: Add eq support of hip08") Link: https://lore.kernel.org/r/20211029095846.26732-1-liangwenpeng@huawei.com Signed-off-by: Haoyue Xu <xuhaoyue1@hisilicon.com> Signed-off-by: Wenpeng Liang <liangwenpeng@huawei.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
2021-11-18RDMA/mlx4: Return missed an error if device doesn't support steeringLeon Romanovsky1-1/+3
[ Upstream commit f4e56ec4452f48b8292dcf0e1c4bdac83506fb8b ] The error flow fixed in this patch is not possible because all kernel users of create QP interface check that device supports steering before set IB_QP_CREATE_NETIF_QP flag. Fixes: c1c98501121e ("IB/mlx4: Add support for steerable IB UD QPs") Link: https://lore.kernel.org/r/91c61f6e60eb0240f8bbc321fda7a1d2986dd03c.1634023677.git.leonro@nvidia.com Reported-by: Dan Carpenter <dan.carpenter@oracle.com> Signed-off-by: Leon Romanovsky <leonro@nvidia.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
2021-11-18RDMA/bnxt_re: Fix query SRQ failureSelvin Xavier1-1/+2
[ Upstream commit 598d16fa1bf93431ad35bbab3ed1affe4fb7b562 ] Fill the missing parameters for the FW command while querying SRQ. Fixes: 37cb11acf1f7 ("RDMA/bnxt_re: Add SRQ support for Broadcom adapters") Link: https://lore.kernel.org/r/1631709163-2287-8-git-send-email-selvin.xavier@broadcom.com Signed-off-by: Selvin Xavier <selvin.xavier@broadcom.com> Reviewed-by: Leon Romanovsky <leonro@nvidia.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
2021-11-18RDMA/rxe: Fix wrong port_cap_flagsJunji Wei1-1/+1
[ Upstream commit dcd3f985b20ffcc375f82ca0ca9f241c7025eb5e ] The port->attr.port_cap_flags should be set to enum ib_port_capability_mask_bits in ib_mad.h, not RDMA_CORE_CAP_PROT_ROCE_UDP_ENCAP. Fixes: 8700e3e7c485 ("Soft RoCE driver") Link: https://lore.kernel.org/r/20210831083223.65797-1-weijunji@bytedance.com Signed-off-by: Junji Wei <weijunji@bytedance.com> Reviewed-by: Leon Romanovsky <leonro@nvidia.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
2021-11-18RDMA/qedr: Fix NULL deref for query_qp on the GSI QPAlok Prasad1-6/+9
commit 4f960393a0ee9a39469ceb7c8077ae8db665cc12 upstream. This patch fixes a crash caused by querying the QP via netlink, and corrects the state of GSI qp. GSI qp's have a NULL qed_qp. The call trace is generated by: $ rdma res show BUG: kernel NULL pointer dereference, address: 0000000000000034 Hardware name: Dell Inc. PowerEdge R720/0M1GCR, BIOS 1.2.6 05/10/2012 RIP: 0010:qed_rdma_query_qp+0x33/0x1a0 [qed] RSP: 0018:ffffba560a08f580 EFLAGS: 00010206 RAX: 0000000200000000 RBX: ffffba560a08f5b8 RCX: 0000000000000000 RDX: ffffba560a08f5b8 RSI: 0000000000000000 RDI: ffff9807ee458090 RBP: ffffba560a08f5a0 R08: 0000000000000000 R09: ffff9807890e7048 R10: ffffba560a08f658 R11: 0000000000000000 R12: 0000000000000000 R13: ffff9807ee458090 R14: ffff9807f0afb000 R15: ffffba560a08f7ec FS: 00007fbbf8bfe740(0000) GS:ffff980aafa00000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 0000000000000034 CR3: 00000001720ba001 CR4: 00000000000606f0 Call Trace: qedr_query_qp+0x82/0x360 [qedr] ib_query_qp+0x34/0x40 [ib_core] ? ib_query_qp+0x34/0x40 [ib_core] fill_res_qp_entry_query.isra.26+0x47/0x1d0 [ib_core] ? __nla_put+0x20/0x30 ? nla_put+0x33/0x40 fill_res_qp_entry+0xe3/0x120 [ib_core] res_get_common_dumpit+0x3f8/0x5d0 [ib_core] ? fill_res_cm_id_entry+0x1f0/0x1f0 [ib_core] nldev_res_get_qp_dumpit+0x1a/0x20 [ib_core] netlink_dump+0x156/0x2f0 __netlink_dump_start+0x1ab/0x260 rdma_nl_rcv+0x1de/0x330 [ib_core] ? nldev_res_get_cm_id_dumpit+0x20/0x20 [ib_core] netlink_unicast+0x1b8/0x270 netlink_sendmsg+0x33e/0x470 sock_sendmsg+0x63/0x70 __sys_sendto+0x13f/0x180 ? setup_sgl.isra.12+0x70/0xc0 __x64_sys_sendto+0x28/0x30 do_syscall_64+0x3a/0xb0 entry_SYSCALL_64_after_hwframe+0x44/0xae Cc: stable@vger.kernel.org Fixes: cecbcddf6461 ("qedr: Add support for QP verbs") Link: https://lore.kernel.org/r/20211027184329.18454-1-palok@marvell.com Signed-off-by: Ariel Elior <aelior@marvell.com> Signed-off-by: Shai Malin <smalin@marvell.com> Signed-off-by: Prabhakar Kushwaha <pkushwaha@marvell.com> Signed-off-by: Alok Prasad <palok@marvell.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2021-10-25RDMA/sa_query: Use strscpy_pad instead of memcpy to copy a stringMark Zhang1-2/+3
When copying the device name, the length of the data memcpy copied exceeds the length of the source buffer, which cause the KASAN issue below. Use strscpy_pad() instead. BUG: KASAN: slab-out-of-bounds in ib_nl_set_path_rec_attrs+0x136/0x320 [ib_core] Read of size 64 at addr ffff88811a10f5e0 by task rping/140263 CPU: 3 PID: 140263 Comm: rping Not tainted 5.15.0-rc1+ #1 Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS rel-1.13.0-0-gf21b5a4aeb02-prebuilt.qemu.org 04/01/2014 Call Trace: dump_stack_lvl+0x57/0x7d print_address_description.constprop.0+0x1d/0xa0 kasan_report+0xcb/0x110 kasan_check_range+0x13d/0x180 memcpy+0x20/0x60 ib_nl_set_path_rec_attrs+0x136/0x320 [ib_core] ib_nl_make_request+0x1c6/0x380 [ib_core] send_mad+0x20a/0x220 [ib_core] ib_sa_path_rec_get+0x3e3/0x800 [ib_core] cma_query_ib_route+0x29b/0x390 [rdma_cm] rdma_resolve_route+0x308/0x3e0 [rdma_cm] ucma_resolve_route+0xe1/0x150 [rdma_ucm] ucma_write+0x17b/0x1f0 [rdma_ucm] vfs_write+0x142/0x4d0 ksys_write+0x133/0x160 do_syscall_64+0x43/0x90 entry_SYSCALL_64_after_hwframe+0x44/0xae RIP: 0033:0x7f26499aa90f Code: 89 54 24 18 48 89 74 24 10 89 7c 24 08 e8 29 fd ff ff 48 8b 54 24 18 48 8b 74 24 10 41 89 c0 8b 7c 24 08 b8 01 00 00 00 0f 05 <48> 3d 00 f0 ff ff 77 31 44 89 c7 48 89 44 24 08 e8 5c fd ff ff 48 RSP: 002b:00007f26495f2dc0 EFLAGS: 00000293 ORIG_RAX: 0000000000000001 RAX: ffffffffffffffda RBX: 00000000000007d0 RCX: 00007f26499aa90f RDX: 0000000000000010 RSI: 00007f26495f2e00 RDI: 0000000000000003 RBP: 00005632a8315440 R08: 0000000000000000 R09: 0000000000000001 R10: 0000000000000000 R11: 0000000000000293 R12: 00007f26495f2e00 R13: 00005632a83154e0 R14: 00005632a8315440 R15: 00005632a830a810 Allocated by task 131419: kasan_save_stack+0x1b/0x40 __kasan_kmalloc+0x7c/0x90 proc_self_get_link+0x8b/0x100 pick_link+0x4f1/0x5c0 step_into+0x2eb/0x3d0 walk_component+0xc8/0x2c0 link_path_walk+0x3b8/0x580 path_openat+0x101/0x230 do_filp_open+0x12e/0x240 do_sys_openat2+0x115/0x280 __x64_sys_openat+0xce/0x140 do_syscall_64+0x43/0x90 entry_SYSCALL_64_after_hwframe+0x44/0xae Fixes: 2ca546b92a02 ("IB/sa: Route SA pathrecord query through netlink") Link: https://lore.kernel.org/r/72ede0f6dab61f7f23df9ac7a70666e07ef314b0.1635055496.git.leonro@nvidia.com Signed-off-by: Mark Zhang <markzhang@nvidia.com> Reviewed-by: Mark Bloch <mbloch@nvidia.com> Signed-off-by: Leon Romanovsky <leonro@nvidia.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-10-20RDMA/irdma: Do not hold qos mutex twice on QP resumeMustafa Ismail1-6/+7
When irdma_ws_add fails, irdma_ws_remove is used to cleanup the leaf node. This lead to holding the qos mutex twice in the QP resume path. Fix this by avoiding the call to irdma_ws_remove and unwinding the error in irdma_ws_add. This skips the call to irdma_tc_in_use function which is not needed in the error unwind cases. Fixes: 3ae331c75128 ("RDMA/irdma: Add QoS definitions") Link: https://lore.kernel.org/r/20211019151654.1943-2-shiraz.saleem@intel.com Signed-off-by: Mustafa Ismail <mustafa.ismail@intel.com> Signed-off-by: Shiraz Saleem <shiraz.saleem@intel.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-10-20RDMA/irdma: Set VLAN in UD work completion correctlyMustafa Ismail1-2/+6
Currently VLAN is reported in UD work completion when VLAN id is zero, i.e. no VLAN case. Report VLAN in UD work completion only when VLAN id is non-zero. Fixes: b48c24c2d710 ("RDMA/irdma: Implement device supported verb APIs") Link: https://lore.kernel.org/r/20211019151654.1943-1-shiraz.saleem@intel.com Signed-off-by: Mustafa Ismail <mustafa.ismail@intel.com> Signed-off-by: Shiraz Saleem <shiraz.saleem@intel.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-10-20RDMA/mlx5: Initialize the ODP xarray when creating an ODP MRAharon Landau1-1/+1
Normally the zero fill would hide the missing initialization, but an errant set to desc_size in reg_create() causes a crash: BUG: unable to handle page fault for address: 0000000800000000 PGD 0 P4D 0 Oops: 0000 [#1] SMP PTI CPU: 5 PID: 890 Comm: ib_write_bw Not tainted 5.15.0-rc4+ #47 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_dereg_mr+0x14/0x3b0 [mlx5_ib] Code: 48 63 cd 4c 89 f7 48 89 0c 24 e8 37 30 03 e1 48 8b 0c 24 eb a0 90 0f 1f 44 00 00 41 56 41 55 41 54 55 53 48 89 fb 48 83 ec 30 <48> 8b 2f 65 48 8b 04 25 28 00 00 00 48 89 44 24 28 31 c0 8b 87 c8 RSP: 0018:ffff88811afa3a60 EFLAGS: 00010286 RAX: 000000000000001c RBX: 0000000800000000 RCX: 0000000000000000 RDX: 0000000000000000 RSI: 0000000000000000 RDI: 0000000800000000 RBP: 0000000800000000 R08: 0000000000000000 R09: c0000000fffff7ff R10: ffff88811afa38f8 R11: ffff88811afa38f0 R12: ffffffffa02c7ac0 R13: 0000000000000000 R14: ffff88811afa3cd8 R15: ffff88810772fa00 FS: 00007f47b9080740(0000) GS:ffff88852cd40000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 0000000800000000 CR3: 000000010761e003 CR4: 0000000000370ea0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 Call Trace: mlx5_ib_free_odp_mr+0x95/0xc0 [mlx5_ib] mlx5_ib_dereg_mr+0x128/0x3b0 [mlx5_ib] ib_dereg_mr_user+0x45/0xb0 [ib_core] ? xas_load+0x8/0x80 destroy_hw_idr_uobject+0x1a/0x50 [ib_uverbs] uverbs_destroy_uobject+0x2f/0x150 [ib_uverbs] uobj_destroy+0x3c/0x70 [ib_uverbs] ib_uverbs_cmd_verbs+0x467/0xb00 [ib_uverbs] ? uverbs_finalize_object+0x60/0x60 [ib_uverbs] ? ttwu_queue_wakelist+0xa9/0xe0 ? pty_write+0x85/0x90 ? file_tty_write.isra.33+0x214/0x330 ? process_echoes+0x60/0x60 ib_uverbs_ioctl+0xa7/0x110 [ib_uverbs] __x64_sys_ioctl+0x10d/0x8e0 ? vfs_write+0x17f/0x260 do_syscall_64+0x3c/0x80 entry_SYSCALL_64_after_hwframe+0x44/0xae Add the missing xarray initialization and remove the desc_size set. Fixes: a639e66703ee ("RDMA/mlx5: Zero out ODP related items in the mlx5_ib_mr") Link: https://lore.kernel.org/r/a4846a11c9de834663e521770da895007f9f0d30.1634642730.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>
2021-10-20rdma/qedr: Fix crash due to redundant release of device's qp memoryPrabhakar Kushwaha3-2/+6
Device's QP memory should only be allocated and released by IB layer. This patch removes the redundant release of the device's qp memory and uses completion APIs to make sure that .destroy_qp() only return, when qp reference becomes 0. Fixes: 514aee660df4 ("RDMA: Globally allocate and release QP memory") Link: https://lore.kernel.org/r/20211019082212.7052-1-pkushwaha@marvell.com Acked-by: Michal Kalderon <michal.kalderon@marvell.com> Signed-off-by: Ariel Elior <aelior@marvell.com> Signed-off-by: Shai Malin <smalin@marvell.com> Signed-off-by: Alok Prasad <palok@marvell.com> Signed-off-by: Prabhakar Kushwaha <pkushwaha@marvell.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-10-13RDMA/rdmavt: Fix error code in rvt_create_qp()Dan Carpenter1-1/+1
Return negative -ENOMEM instead of positive ENOMEM. Returning a postive value will cause an Oops because it becomes an ERR_PTR() in the create_qp() function. Fixes: 514aee660df4 ("RDMA: Globally allocate and release QP memory") Link: https://lore.kernel.org/r/20211013080645.GD6010@kili Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com> Acked-by: Dennis Dalessandro <dennis.dalessandro@cornelisnetworks.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-10-13IB/hfi1: Fix abba locking issue with sc_disable()Mike Marciniszyn1-3/+6
sc_disable() after having disabled the send context wakes up any waiters by calling hfi1_qp_wakeup() while holding the waitlock for the sc. This is contrary to the model for all other calls to hfi1_qp_wakeup() where the waitlock is dropped and a local is used to drive calls to hfi1_qp_wakeup(). Fix by moving the sc->piowait into a local list and driving the wakeup calls from the list. Fixes: 099a884ba4c0 ("IB/hfi1: Handle wakeup of orphaned QPs for pio") Link: https://lore.kernel.org/r/20211013141852.128104.2682.stgit@awfm-01.cornelisnetworks.com Signed-off-by: Mike Marciniszyn <mike.marciniszyn@cornelisnetworks.com> Reported-by: TOTE Robot <oslab@tsinghua.edu.cn> Signed-off-by: Dennis Dalessandro <dennis.dalessandro@cornelisnetworks.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-10-13IB/qib: Protect from buffer overflow in struct qib_user_sdma_pkt fieldsMike Marciniszyn1-10/+23
Overflowing either addrlimit or bytes_togo can allow userspace to trigger a buffer overflow of kernel memory. Check for overflows in all the places doing math on user controlled buffers. Fixes: f931551bafe1 ("IB/qib: Add new qib driver for QLogic PCIe InfiniBand adapters") Link: https://lore.kernel.org/r/20211012175519.7298.77738.stgit@awfm-01.cornelisnetworks.com Reported-by: Ilja Van Sprundel <ivansprundel@ioactive.com> Reviewed-by: Dennis Dalessandro <dennis.dalessandro@cornelisnetworks.com> Signed-off-by: Mike Marciniszyn <mike.marciniszyn@cornelisnetworks.com> Signed-off-by: Dennis Dalessandro <dennis.dalessandro@cornelisnetworks.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-10-06RDMA/mlx5: Set user priority for DCTPatrisious Haddad1-0/+2
Currently, the driver doesn't set the PCP-based priority for DCT, hence DCT response packets are transmitted without user priority. Fix it by setting user provided priority in the eth_prio field in the DCT context, which in turn sets the value in the transmitted packet. Fixes: 776a3906b692 ("IB/mlx5: Add support for DC target QP") Link: https://lore.kernel.org/r/5fd2d94a13f5742d8803c218927322257d53205c.1633512672.git.leonro@nvidia.com Signed-off-by: Patrisious Haddad <phaddad@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>
2021-10-06RDMA/irdma: Process extended CQ entries correctlyShiraz Saleem1-2/+2
The valid bit for extended CQE's written by HW is retrieved from the incorrect quad-word. This leads to missed completions for any UD traffic particularly after a wrap-around. Get the valid bit for extended CQE's from the correct quad-word in the descriptor. Fixes: 551c46edc769 ("RDMA/irdma: Add user/kernel shared libraries") Link: https://lore.kernel.org/r/20211005182302.374-1-shiraz.saleem@intel.com Signed-off-by: Shiraz Saleem <shiraz.saleem@intel.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-09-27RDMA/hns: Add the check of the CQE size of the user spaceWenpeng Liang1-9/+22
If the CQE size of the user space is not the size supported by the hardware, the creation of CQ should be stopped. Fixes: 09a5f210f67e ("RDMA/hns: Add support for CQE in size of 64 Bytes") Link: https://lore.kernel.org/r/20210927125557.15031-3-liangwenpeng@huawei.com Signed-off-by: Wenpeng Liang <liangwenpeng@huawei.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-09-27RDMA/hns: Fix the size setting error when copying CQE in clean_cq()Wenpeng Liang1-1/+1
The size of CQE is different for different versions of hardware, so the driver needs to specify the size of CQE explicitly. Fixes: 09a5f210f67e ("RDMA/hns: Add support for CQE in size of 64 Bytes") Link: https://lore.kernel.org/r/20210927125557.15031-2-liangwenpeng@huawei.com Signed-off-by: Wenpeng Liang <liangwenpeng@huawei.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-09-27RDMA/hfi1: Fix kernel pointer leakGuo Zhi1-4/+4
Pointers should be printed with %p or %px rather than cast to 'unsigned long long' and printed with %llx. Change %llx to %p to print the secured pointer. Fixes: 042a00f93aad ("IB/{ipoib,hfi1}: Add a timeout handler for rdma_netdev") Link: https://lore.kernel.org/r/20210922134857.619602-1-qtxuning1999@sjtu.edu.cn Signed-off-by: Guo Zhi <qtxuning1999@sjtu.edu.cn> Acked-by: Mike Marciniszyn <mike.marciniszyn@cornelisnetworks.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-09-24RDMA/usnic: Lock VF with mutex instead of spinlockLeon Romanovsky3-10/+10
Usnic VF doesn't need lock in atomic context to create QPs, so it is safe to use mutex instead of spinlock. Such change fixes the following smatch error. Smatch static checker warning: lib/kobject.c:289 kobject_set_name_vargs() warn: sleeping in atomic context Fixes: 514aee660df4 ("RDMA: Globally allocate and release QP memory") Link: https://lore.kernel.org/r/2a0e295786c127e518ebee8bb7cafcb819a625f6.1631520231.git.leonro@nvidia.com Reported-by: Dan Carpenter <dan.carpenter@oracle.com> Signed-off-by: Leon Romanovsky <leonro@nvidia.com> Reviewed-by: Håkon Bugge <haakon.bugge@oracle.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-09-24RDMA/hns: Work around broken constant propagation in gcc 8Jason Gunthorpe1-5/+6
gcc 8.3 and 5.4 throw this: In function 'modify_qp_init_to_rtr', ././include/linux/compiler_types.h:322:38: error: call to '__compiletime_assert_1859' declared with attribute error: FIELD_PREP: value too large for the field _compiletime_assert(condition, msg, __compiletime_assert_, __COUNTER__) [..] drivers/infiniband/hw/hns/hns_roce_common.h:91:52: note: in expansion of macro 'FIELD_PREP' *((__le32 *)ptr + (field_h) / 32) |= cpu_to_le32(FIELD_PREP( \ ^~~~~~~~~~ drivers/infiniband/hw/hns/hns_roce_common.h:95:39: note: in expansion of macro '_hr_reg_write' #define hr_reg_write(ptr, field, val) _hr_reg_write(ptr, field, val) ^~~~~~~~~~~~~ drivers/infiniband/hw/hns/hns_roce_hw_v2.c:4412:2: note: in expansion of macro 'hr_reg_write' hr_reg_write(context, QPC_LP_PKTN_INI, lp_pktn_ini); Because gcc has miscalculated the constantness of lp_pktn_ini: mtu = ib_mtu_enum_to_int(ib_mtu); if (WARN_ON(mtu < 0)) [..] lp_pktn_ini = ilog2(MAX_LP_MSG_LEN / mtu); Since mtu is limited to {256,512,1024,2048,4096} lp_pktn_ini is between 4 and 8 which is compatible with the 4 bit field in the FIELD_PREP. Work around this broken compiler by adding a 'can never be true' constraint on lp_pktn_ini's value which clears out the problem. Fixes: f0cb411aad23 ("RDMA/hns: Use new interface to modify QP context") Link: https://lore.kernel.org/r/0-v1-c773ecb137bc+11f-hns_gcc8_jgg@nvidia.com Reported-by: Geert Uytterhoeven <geert@linux-m68k.org> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-09-23RDMA/cma: Ensure rdma_addr_cancel() happens before issuing more requestsJason Gunthorpe2-0/+24
The FSM can run in a circle allowing rdma_resolve_ip() to be called twice on the same id_priv. While this cannot happen without going through the work, it violates the invariant that the same address resolution background request cannot be active twice. CPU 1 CPU 2 rdma_resolve_addr(): RDMA_CM_IDLE -> RDMA_CM_ADDR_QUERY rdma_resolve_ip(addr_handler) #1 process_one_req(): for #1 addr_handler(): RDMA_CM_ADDR_QUERY -> RDMA_CM_ADDR_BOUND mutex_unlock(&id_priv->handler_mutex); [.. handler still running ..] rdma_resolve_addr(): RDMA_CM_ADDR_BOUND -> RDMA_CM_ADDR_QUERY rdma_resolve_ip(addr_handler) !! two requests are now on the req_list rdma_destroy_id(): destroy_id_handler_unlock(): _destroy_id(): cma_cancel_operation(): rdma_addr_cancel() // process_one_req() self removes it spin_lock_bh(&lock); cancel_delayed_work(&req->work); if (!list_empty(&req->list)) == true ! rdma_addr_cancel() returns after process_on_req #1 is done kfree(id_priv) process_one_req(): for #2 addr_handler(): mutex_lock(&id_priv->handler_mutex); !! Use after free on id_priv rdma_addr_cancel() expects there to be one req on the list and only cancels the first one. The self-removal behavior of the work only happens after the handler has returned. This yields a situations where the req_list can have two reqs for the same "handle" but rdma_addr_cancel() only cancels the first one. The second req remains active beyond rdma_destroy_id() and will use-after-free id_priv once it inevitably triggers. Fix this by remembering if the id_priv has called rdma_resolve_ip() and always cancel before calling it again. This ensures the req_list never gets more than one item in it and doesn't cost anything in the normal flow that never uses this strange error path. Link: https://lore.kernel.org/r/0-v1-3bc675b8006d+22-syz_cancel_uaf_jgg@nvidia.com Cc: stable@vger.kernel.org Fixes: e51060f08a61 ("IB: IP address based RDMA connection manager") Reported-by: syzbot+dc3dfba010d7671e05f5@syzkaller.appspotmail.com Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-09-22RDMA/cma: Do not change route.addr.src_addr.ss_familyJason Gunthorpe1-2/+6
If the state is not idle then rdma_bind_addr() will immediately fail and no change to global state should happen. For instance if the state is already RDMA_CM_LISTEN then this will corrupt the src_addr and would cause the test in cma_cancel_operation(): if (cma_any_addr(cma_src_addr(id_priv)) && !id_priv->cma_dev) To view a mangled src_addr, eg with a IPv6 loopback address but an IPv4 family, failing the test. This would manifest as this trace from syzkaller: BUG: KASAN: use-after-free in __list_add_valid+0x93/0xa0 lib/list_debug.c:26 Read of size 8 at addr ffff8881546491e0 by task syz-executor.1/32204 CPU: 1 PID: 32204 Comm: syz-executor.1 Not tainted 5.12.0-rc8-syzkaller #0 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011 Call Trace: __dump_stack lib/dump_stack.c:79 [inline] dump_stack+0x141/0x1d7 lib/dump_stack.c:120 print_address_description.constprop.0.cold+0x5b/0x2f8 mm/kasan/report.c:232 __kasan_report mm/kasan/report.c:399 [inline] kasan_report.cold+0x7c/0xd8 mm/kasan/report.c:416 __list_add_valid+0x93/0xa0 lib/list_debug.c:26 __list_add include/linux/list.h:67 [inline] list_add_tail include/linux/list.h:100 [inline] cma_listen_on_all drivers/infiniband/core/cma.c:2557 [inline] rdma_listen+0x787/0xe00 drivers/infiniband/core/cma.c:3751 ucma_listen+0x16a/0x210 drivers/infiniband/core/ucma.c:1102 ucma_write+0x259/0x350 drivers/infiniband/core/ucma.c:1732 vfs_write+0x28e/0xa30 fs/read_write.c:603 ksys_write+0x1ee/0x250 fs/read_write.c:658 do_syscall_64+0x2d/0x70 arch/x86/entry/common.c:46 entry_SYSCALL_64_after_hwframe+0x44/0xae Which is indicating that an rdma_id_private was destroyed without doing cma_cancel_listens(). Instead of trying to re-use the src_addr memory to indirectly create an any address build one explicitly on the stack and bind to that as any other normal flow would do. Link: https://lore.kernel.org/r/0-v1-9fbb33f5e201+2a-cma_listen_jgg@nvidia.com Cc: stable@vger.kernel.org Fixes: 732d41c545bb ("RDMA/cma: Make the locking for automatic state transition more clear") Reported-by: syzbot+6bb0528b13611047209c@syzkaller.appspotmail.com Tested-by: Hao Sun <sunhao.th@gmail.com> Reviewed-by: Leon Romanovsky <leonro@nvidia.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-09-20RDMA/irdma: Report correct WC error when there are MW bind errorsSindhu Devale3-0/+8
Report the correct WC error when MW bind error related asynchronous events are generated by HW. Fixes: b48c24c2d710 ("RDMA/irdma: Implement device supported verb APIs") Link: https://lore.kernel.org/r/20210916191222.824-5-shiraz.saleem@intel.com Signed-off-by: Sindhu Devale <sindhu.devale@intel.com> Signed-off-by: Shiraz Saleem <shiraz.saleem@intel.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-09-20RDMA/irdma: Report correct WC error when transport retry counter is exceededSindhu Devale3-0/+6
When the retry counter exceeds, as the remote QP didn't send any Ack or Nack an asynchronous event (AE) for too many retries is generated. Add code to handle the AE and set the correct IB WC error code IB_WC_RETRY_EXC_ERR. Fixes: b48c24c2d710 ("RDMA/irdma: Implement device supported verb APIs") Link: https://lore.kernel.org/r/20210916191222.824-4-shiraz.saleem@intel.com Signed-off-by: Sindhu Devale <sindhu.devale@intel.com> Signed-off-by: Shiraz Saleem <shiraz.saleem@intel.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-09-20RDMA/irdma: Validate number of CQ entries on create CQSindhu Devale1-1/+1
Add lower bound check for CQ entries at creation time. Fixes: b48c24c2d710 ("RDMA/irdma: Implement device supported verb APIs") Link: https://lore.kernel.org/r/20210916191222.824-3-shiraz.saleem@intel.com Signed-off-by: Sindhu Devale <sindhu.devale@intel.com> Signed-off-by: Shiraz Saleem <shiraz.saleem@intel.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-09-20RDMA/irdma: Skip CQP ring during a resetSindhu Devale6-10/+8
Due to duplicate reset flags, CQP commands are processed during reset. This leads CQP failures such as below: irdma0: [Delete Local MAC Entry Cmd Error][op_code=49] status=-27 waiting=1 completion_err=0 maj=0x0 min=0x0 Remove the redundant flag and set the correct reset flag so CPQ is paused during reset Fixes: 8498a30e1b94 ("RDMA/irdma: Register auxiliary driver and implement private channel OPs") Link: https://lore.kernel.org/r/20210916191222.824-2-shiraz.saleem@intel.com Reported-by: LiLiang <liali@redhat.com> Signed-off-by: Sindhu Devale <sindhu.devale@intel.com> Signed-off-by: Shiraz Saleem <shiraz.saleem@intel.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-09-15RDMA/cma: Fix listener leak in rdma_cma_listen_on_all() failureTao Liu1-3/+10
If cma_listen_on_all() fails it leaves the per-device ID still on the listen_list but the state is not set to RDMA_CM_ADDR_BOUND. When the cmid is eventually destroyed cma_cancel_listens() is not called due to the wrong state, however the per-device IDs are still holding the refcount preventing the ID from being destroyed, thus deadlocking: task:rping state:D stack: 0 pid:19605 ppid: 47036 flags:0x00000084 Call Trace: __schedule+0x29a/0x780 ? free_unref_page_commit+0x9b/0x110 schedule+0x3c/0xa0 schedule_timeout+0x215/0x2b0 ? __flush_work+0x19e/0x1e0 wait_for_completion+0x8d/0xf0 _destroy_id+0x144/0x210 [rdma_cm] ucma_close_id+0x2b/0x40 [rdma_ucm] __destroy_id+0x93/0x2c0 [rdma_ucm] ? __xa_erase+0x4a/0xa0 ucma_destroy_id+0x9a/0x120 [rdma_ucm] ucma_write+0xb8/0x130 [rdma_ucm] vfs_write+0xb4/0x250 ksys_write+0xb5/0xd0 ? syscall_trace_enter.isra.19+0x123/0x190 do_syscall_64+0x33/0x40 entry_SYSCALL_64_after_hwframe+0x44/0xa9 Ensure that cma_listen_on_all() atomically unwinds its action under the lock during error. Fixes: c80a0c52d85c ("RDMA/cma: Add missing error handling of listen_id") Link: https://lore.kernel.org/r/20210913093344.17230-1-thomas.liu@ucloud.cn Signed-off-by: Tao Liu <thomas.liu@ucloud.cn> Reviewed-by: Leon Romanovsky <leonro@nvidia.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-09-14IB/cma: Do not send IGMP leaves for sendonly Multicast groupsChristoph Lameter1-1/+6
ROCE uses IGMP for Multicast instead of the native Infiniband system where joins are required in order to post messages on the Multicast group. On Ethernet one can send Multicast messages to arbitrary addresses without the need to subscribe to a group. So ROCE correctly does not send IGMP joins during rdma_join_multicast(). F.e. in cma_iboe_join_multicast() we see: if (addr->sa_family == AF_INET) { if (gid_type == IB_GID_TYPE_ROCE_UDP_ENCAP) { ib.rec.hop_limit = IPV6_DEFAULT_HOPLIMIT; if (!send_only) { err = cma_igmp_send(ndev, &ib.rec.mgid, true); } } } else { So the IGMP join is suppressed as it is unnecessary. However no such check is done in destroy_mc(). And therefore leaving a sendonly multicast group will send an IGMP leave. This means that the following scenario can lead to a multicast receiver unexpectedly being unsubscribed from a MC group: 1. Sender thread does a sendonly join on MC group X. No IGMP join is sent. 2. Receiver thread does a regular join on the same MC Group x. IGMP join is sent and the receiver begins to get messages. 3. Sender thread terminates and destroys MC group X. IGMP leave is sent and the receiver no longer receives data. This patch adds the same logic for sendonly joins to destroy_mc() that is also used in cma_iboe_join_multicast(). Fixes: ab15c95a17b3 ("IB/core: Support for CMA multicast join flags") Link: https://lore.kernel.org/r/alpine.DEB.2.22.394.2109081340540.668072@gentwo.de Signed-off-by: Christoph Lameter <cl@linux.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-09-13IB/qib: Fix clang confusion of NULL pointer comparisonJason Gunthorpe1-1/+1
clang becomes confused due to the comparison to NULL in a integer constant expression context: >> drivers/infiniband/hw/qib/qib_sysfs.c:413:1: error: static_assert expression is not an integral constant expression QIB_DIAGC_ATTR(rc_resends); ^~~~~~~~~~~~~~~~~~~~~~~~~~ drivers/infiniband/hw/qib/qib_sysfs.c:406:16: note: expanded from macro 'QIB_DIAGC_ATTR' static_assert(&((struct qib_ibport *)0)->rvp.n_##N != (u64 *)NULL); \ Nathan found __same_type that solves this problem nicely, so use it instead. Reported-by: kernel test robot <lkp@intel.com> Suggested-by: Nathan Chancellor <nathan@kernel.org> Reviewed-by: Nick Desaulniers <ndesaulniers@google.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
2021-09-09Merge tag 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdmaLinus Torvalds4-6/+8
Pull rdma fixes from Jason Gunthorpe: "I don't usually send a second PR in the merge window, but the fix to mlx5 is significant enough that it should start going through the process ASAP. Along with it comes some of the usual -rc stuff that would normally wait for a -rc2 or so. Summary: Important error case regression fixes in mlx5: - Wrong size used when computing the error path smaller allocation request leads to corruption - Confusing but ultimately harmless alignment mis-calculation Static checker warning fixes: - NULL pointer subtraction in qib - kcalloc in bnxt_re - Missing static on global variable in hfi1" * tag 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma: IB/hfi1: make hist static RDMA/bnxt_re: Prefer kcalloc over open coded arithmetic IB/qib: Fix null pointer subtraction compiler warning RDMA/mlx5: Fix xlt_chunk_align calculation RDMA/mlx5: Fix number of allocated XLT entries
2021-09-08IB/hfi1: make hist staticchongjiapeng1-1/+1
This symbol is not used outside of trace.c, so marks it static. Fix the following sparse warning: drivers/infiniband/hw/hfi1/trace.c:491:23: warning: symbol 'hist' was not declared. Should it be static? Link: https://lore.kernel.org/r/1630921723-21545-1-git-send-email-jiapeng.chong@linux.alibaba.com Reported-by: Abaci Robot <abaci@linux.alibaba.com> Signed-off-by: chongjiapeng <jiapeng.chong@linux.alibaba.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
generated by cgit v1.2.3 (git 2.39.0) at 2025-09-04 16:44:20 +0300