kernel/linux.git/drivers/infiniband, branch linux-4.20.y

IB/ipoib: Fix for use-after-free in ipoib_cm_tx_start

2019-03-13T21:04:02+00:00

[ Upstream commit 6ab4aba00f811a5265acc4d3eb1863bb3ca60562 ] The following BUG was reported by kasan: BUG: KASAN: use-after-free in ipoib_cm_tx_start+0x430/0x1390 [ib_ipoib] Read of size 80 at addr ffff88034c30bcd0 by task kworker/u16:1/24020 Workqueue: ipoib_wq ipoib_cm_tx_start [ib_ipoib] Call Trace: dump_stack+0x9a/0xeb print_address_description+0xe3/0x2e0 kasan_report+0x18a/0x2e0 ? ipoib_cm_tx_start+0x430/0x1390 [ib_ipoib] memcpy+0x1f/0x50 ipoib_cm_tx_start+0x430/0x1390 [ib_ipoib] ? kvm_clock_read+0x1f/0x30 ? ipoib_cm_skb_reap+0x610/0x610 [ib_ipoib] ? __lock_is_held+0xc2/0x170 ? process_one_work+0x880/0x1960 ? process_one_work+0x912/0x1960 process_one_work+0x912/0x1960 ? wq_pool_ids_show+0x310/0x310 ? lock_acquire+0x145/0x440 worker_thread+0x87/0xbb0 ? process_one_work+0x1960/0x1960 kthread+0x314/0x3d0 ? kthread_create_worker_on_cpu+0xc0/0xc0 ret_from_fork+0x3a/0x50 Allocated by task 0: kasan_kmalloc+0xa0/0xd0 kmem_cache_alloc_trace+0x168/0x3e0 path_rec_create+0xa2/0x1f0 [ib_ipoib] ipoib_start_xmit+0xa98/0x19e0 [ib_ipoib] dev_hard_start_xmit+0x159/0x8d0 sch_direct_xmit+0x226/0xb40 __dev_queue_xmit+0x1d63/0x2950 neigh_update+0x889/0x1770 arp_process+0xc47/0x21f0 arp_rcv+0x462/0x760 __netif_receive_skb_core+0x1546/0x2da0 netif_receive_skb_internal+0xf2/0x590 napi_gro_receive+0x28e/0x390 ipoib_ib_handle_rx_wc_rss+0x873/0x1b60 [ib_ipoib] ipoib_rx_poll_rss+0x17d/0x320 [ib_ipoib] net_rx_action+0x427/0xe30 __do_softirq+0x28e/0xc42 Freed by task 26680: __kasan_slab_free+0x11d/0x160 kfree+0xf5/0x360 ipoib_flush_paths+0x532/0x9d0 [ib_ipoib] ipoib_set_mode_rss+0x1ad/0x560 [ib_ipoib] set_mode+0xc8/0x150 [ib_ipoib] kernfs_fop_write+0x279/0x440 __vfs_write+0xd8/0x5c0 vfs_write+0x15e/0x470 ksys_write+0xb8/0x180 do_syscall_64+0x9b/0x420 entry_SYSCALL_64_after_hwframe+0x49/0xbe The buggy address belongs to the object at ffff88034c30bcc8 which belongs to the cache kmalloc-512 of size 512 The buggy address is located 8 bytes inside of 512-byte region [ffff88034c30bcc8, ffff88034c30bec8) The buggy address belongs to the page: The following race between change mode and xmit flow is the reason for this use-after-free: Change mode Send packet 1 to GID XX Send packet 2 to GID XX | | | start | | | | | | | | | Create new path for GID XX | | and update neigh path | | | | | | | | | | flush_paths | | | | queue_work(cm.start_task) | | Path for GID XX not found | create new path | | start_task runs with old released path There is no locking to protect the lifetime of the path through the ipoib_cm_tx struct, so delete it entirely and always use the newly looked up path under the priv->lock. Fixes: 546481c2816e ("IB/ipoib: Fix memory corruption in ipoib cm mode connect flow") Signed-off-by: Feras Daoud Reviewed-by: Erez Shitrit Signed-off-by: Leon Romanovsky Signed-off-by: Jason Gunthorpe Signed-off-by: Sasha Levin

RDMA/umem: Add missing initialization of owning_mm

2019-03-13T21:04:01+00:00

[ Upstream commit a2093dd35f8cfb28dd7c878ccbd020c1bb20d0d7 ] When allocating a umem leaf for implicit ODP MR during page fault the field owning_mm was not set. Initialize and take a reference on this field to avoid kernel panic when trying to access this field. BUG: unable to handle kernel NULL pointer dereference at 0000000000000058 PGD 800000022dfed067 P4D 800000022dfed067 PUD 22dfcf067 PMD 0 Oops: 0000 [#1] SMP PTI CPU: 0 PID: 634 Comm: kworker/u33:0 Not tainted 4.20.0-rc6+ #89 Hardware name: Red Hat KVM, BIOS 0.5.1 01/01/2011 Workqueue: mlx5_ib_page_fault mlx5_ib_eqe_pf_action [mlx5_ib] RIP: 0010:ib_umem_odp_map_dma_pages+0xf3/0x710 [ib_core] Code: 45 c0 48 21 f3 48 89 75 b0 31 f6 4a 8d 04 33 48 89 45 a8 49 8b 44 24 60 48 8b 78 10 e8 66 16 a8 c5 49 8b 54 24 08 48 89 45 98 <8b> 42 58 85 c0 0f 84 8e 05 00 00 8d 48 01 48 8d 72 58 f0 0f b1 4a RSP: 0000:ffffb610813a7c20 EFLAGS: 00010202 RAX: ffff95ace6e8ac80 RBX: 0000000000000000 RCX: 000000000000000c RDX: 0000000000000000 RSI: 0000000000000850 RDI: ffff95aceaadae80 RBP: ffffb610813a7ce0 R08: 0000000000000000 R09: 0000000000080c77 R10: ffff95acfffdbd00 R11: 0000000000000000 R12: ffff95aceaa20a00 R13: 0000000000001000 R14: 0000000000001000 R15: 000000000000000c FS: 0000000000000000(0000) GS:ffff95acf7800000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 0000000000000058 CR3: 000000022c834001 CR4: 00000000001606f0 Call Trace: pagefault_single_data_segment+0x1df/0xc60 [mlx5_ib] mlx5_ib_eqe_pf_action+0x7bc/0xa70 [mlx5_ib] ? __switch_to+0xe1/0x470 process_one_work+0x174/0x390 worker_thread+0x4f/0x3e0 kthread+0x102/0x140 ? drain_workqueue+0x130/0x130 ? kthread_stop+0x110/0x110 ret_from_fork+0x1f/0x30 Fixes: f27a0d50a4bc ("RDMA/umem: Use umem->owning_mm inside ODP") Signed-off-by: Artemy Kovalyov Signed-off-by: Moni Shoua Signed-off-by: Leon Romanovsky Signed-off-by: Jason Gunthorpe Signed-off-by: Sasha Levin

IB/{hfi1, qib}: Fix WC.byte_len calculation for UD_SEND_WITH_IMM

2019-03-13T21:04:00+00:00

[ Upstream commit 904bba211acc2112fdf866e5a2bc6cd9ecd0de1b ] The work completion length for a receiving a UD send with immediate is short by 4 bytes causing application using this opcode to fail. The UD receive logic incorrectly subtracts 4 bytes for immediate value. These bytes are already included in header length and are used to calculate header/payload split, so the result is these 4 bytes are subtracted twice, once when the header length subtracted from the overall length and once again in the UD opcode specific path. Remove the extra subtraction when handling the opcode. Fixes: 7724105686e7 ("IB/hfi1: add driver files") Reviewed-by: Michael J. Ruhl Signed-off-by: Brian Welty Signed-off-by: Mike Marciniszyn Signed-off-by: Dennis Dalessandro Signed-off-by: Jason Gunthorpe Signed-off-by: Sasha Levin

RDMA/srp: Rework SCSI device reset handling

2019-02-27T09:09:54+00:00

commit 48396e80fb6526ea5ed267bd84f028bae56d2f9e upstream. Since .scsi_done() must only be called after scsi_queue_rq() has finished, make sure that the SRP initiator driver does not call .scsi_done() while scsi_queue_rq() is in progress. Although invoking sg_reset -d while I/O is in progress works fine with kernel v4.20 and before, that is not the case with kernel v5.0-rc1. This patch avoids that the following crash is triggered with kernel v5.0-rc1: BUG: unable to handle kernel NULL pointer dereference at 0000000000000138 CPU: 0 PID: 360 Comm: kworker/0:1H Tainted: G B 5.0.0-rc1-dbg+ #1 Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.10.2-1 04/01/2014 Workqueue: kblockd blk_mq_run_work_fn RIP: 0010:blk_mq_dispatch_rq_list+0x116/0xb10 Call Trace: blk_mq_sched_dispatch_requests+0x2f7/0x300 __blk_mq_run_hw_queue+0xd6/0x180 blk_mq_run_work_fn+0x27/0x30 process_one_work+0x4f1/0xa20 worker_thread+0x67/0x5b0 kthread+0x1cf/0x1f0 ret_from_fork+0x24/0x30 Cc: Fixes: 94a9174c630c ("IB/srp: reduce lock coverage of command completion") Signed-off-by: Bart Van Assche Signed-off-by: Jason Gunthorpe Signed-off-by: Greg Kroah-Hartman

RDMA/mthca: Clear QP objects during their allocation

2019-02-27T09:09:46+00:00

[ Upstream commit 9d9f59b4204bc41896c866b3e5856e5b416aa199 ] As part of audit process to update drivers to use rdma_restrack_add() ensure that QP objects is cleared before access. Such change fixes the crash observed with uninitialized non zero sgid attr accessed by ib_destroy_qp(). CPU: 3 PID: 74 Comm: kworker/u16:1 Not tainted 4.19.10-300.fc29.x86_64 Workqueue: ipoib_wq ipoib_cm_tx_reap [ib_ipoib] RIP: 0010:rdma_put_gid_attr+0x9/0x30 [ib_core] RSP: 0018:ffffb7ad819dbde8 EFLAGS: 00010202 RAX: 0000000000000000 RBX: ffff8d1bdf5a2e00 RCX: 0000000000002699 RDX: 206c656e72656af8 RSI: ffff8d1bf7ae6160 RDI: 206c656e72656b20 RBP: 0000000000000000 R08: 0000000000026160 R09: ffffffffc06b45bf R10: ffffe849887da000 R11: 0000000000000002 R12: ffff8d1be30cb400 R13: ffff8d1bdf681800 R14: ffff8d1be2272400 R15: ffff8d1be30ca000 FS: 0000000000000000(0000) GS:ffff8d1bf7ac0000(0000) knlGS:0000000000000000 Trace: ib_destroy_qp+0xc9/0x240 [ib_core] ipoib_cm_tx_reap+0x1f9/0x4e0 [ib_ipoib] process_one_work+0x1a1/0x3a0 worker_thread+0x30/0x380 ? pwq_unbound_release_workfn+0xd0/0xd0 kthread+0x112/0x130 ? kthread_create_worker_on_cpu+0x70/0x70 ret_from_fork+0x22/0x40 Reported-by: Alexander Murashkin Tested-by: Alexander Murashkin Fixes: 1a1f460ff151 ("RDMA: Hold the sgid_attr inside the struct ib_ah/qp") Signed-off-by: Parav Pandit Signed-off-by: Leon Romanovsky Signed-off-by: Jason Gunthorpe Signed-off-by: Sasha Levin

IB/hfi1: Unreserve a reserved request when it is completed

2019-02-12T19:02:15+00:00

[ Upstream commit ca95f802ef5139722acc8d30aeaab6fe5bbe939e ] Currently, When a reserved operation is completed, its entry in the send queue will not be unreserved, which leads to the miscalculation of qp->s_avail and thus the triggering of a WARN_ON call trace. This patch fixes the problem by unreserving the reserved operation when it is completed. Fixes: 856cc4c237ad ("IB/hfi1: Add the capability for reserved operations") Reviewed-by: Mike Marciniszyn Signed-off-by: Kaike Wan Signed-off-by: Dennis Dalessandro Signed-off-by: Jason Gunthorpe Signed-off-by: Sasha Levin

RDMA/core: Sync unregistration with netlink commands

2019-02-12T19:02:10+00:00

[ Upstream commit 01b671170d7f82b959dad6d5dbb44d7a915e647d ] When the rdma device is getting removed, get resource info can race with device removal, as below: CPU-0 CPU-1 -------- -------- rdma_nl_rcv_msg() nldev_res_get_cq_dumpit() mutex_lock(device_lock); get device reference mutex_unlock(device_lock); [..] ib_unregister_device() /* Valid reference to * device->dev exists. */ ib_dealloc_device() [..] provider->fill_res_entry(); Even though device object is not freed, fill_res_entry() can get called on device which doesn't have a driver anymore. Kernel core device reference count is not sufficient, as this only keeps the structure valid, and doesn't guarantee the driver is still loaded. Similar race can occur with device renaming and device removal, where device_rename() tries to rename a unregistered device. While this is fine for devices of a class which are not net namespace aware, but it is incorrect for net namespace aware class coming in subsequent series. If a class is net namespace aware, then the below [1] call trace is observed in above situation. Therefore, to avoid the race, keep a reference count and let device unregistration wait until all netlink users drop the reference. [1] Call trace: kernfs: ns required in 'infiniband' for 'mlx5_0' WARNING: CPU: 18 PID: 44270 at fs/kernfs/dir.c:842 kernfs_find_ns+0x104/0x120 libahci i2c_core mlxfw libata dca [last unloaded: devlink] RIP: 0010:kernfs_find_ns+0x104/0x120 Call Trace: kernfs_find_and_get_ns+0x2e/0x50 sysfs_rename_link_ns+0x40/0xb0 device_rename+0xb2/0xf0 ib_device_rename+0xb3/0x100 [ib_core] nldev_set_doit+0x165/0x190 [ib_core] rdma_nl_rcv_msg+0x249/0x250 [ib_core] ? netlink_deliver_tap+0x8f/0x3e0 rdma_nl_rcv+0xd6/0x120 [ib_core] netlink_unicast+0x17c/0x230 netlink_sendmsg+0x2f0/0x3e0 sock_sendmsg+0x30/0x40 __sys_sendto+0xdc/0x160 Fixes: da5c85078215 ("RDMA/nldev: add driver-specific resource tracking") Signed-off-by: Parav Pandit Signed-off-by: Leon Romanovsky Signed-off-by: Jason Gunthorpe Signed-off-by: Sasha Levin

IB/hfi1: Add limit test for RC/UC send via loopback

2019-02-06T16:27:43+00:00

commit 09ce351dff8e7636af0beb72cd4a86c3904a0500 upstream. Fix potential memory corruption and panic in loopback for IB_WR_SEND variants. The code blindly assumes the posted length will fit in the fetched rwqe, which is not a valid assumption. Fix by adding a limit test, and triggering the appropriate send completion and putting the QP in an error state. This mimics the handling for non-loopback QPs. Fixes: 15703461533a ("IB/{hfi1, qib, rdmavt}: Move ruc_loopback to rdmavt") Cc: #v4.20+ Reviewed-by: Michael J. Ruhl Signed-off-by: Mike Marciniszyn Signed-off-by: Dennis Dalessandro Signed-off-by: Jason Gunthorpe Signed-off-by: Greg Kroah-Hartman

IB/hfi1: Remove overly conservative VM_EXEC flag check

2019-02-06T16:27:43+00:00

commit 7709b0dc265f28695487712c45f02bbd1f98415d upstream. Applications that use the stack for execution purposes cause userspace PSM jobs to fail during mmap(). Both Fortran (non-standard format parsing) and C (callback functions located in the stack) applications can be written such that stack execution is required. The linker notes this via the gnu_stack ELF flag. This causes READ_IMPLIES_EXEC to be set which forces all PROT_READ mmaps to have PROT_EXEC for the process. Checking for VM_EXEC bit and failing the request with EPERM is overly conservative and will break any PSM application using executable stacks. Cc: #v4.14+ Fixes: 12220267645c ("IB/hfi: Protect against writable mmap") Reviewed-by: Mike Marciniszyn Reviewed-by: Dennis Dalessandro Reviewed-by: Ira Weiny Signed-off-by: Michael J. Ruhl Signed-off-by: Dennis Dalessandro Signed-off-by: Jason Gunthorpe Signed-off-by: Greg Kroah-Hartman

IB/uverbs: Fix OOPs in uverbs_user_mmap_disassociate

2019-02-06T16:27:43+00:00

commit 7b21b69ab203136fdc153c7707fa6c409e523c2e upstream. The vma->vm_mm can become impossible to get before rdma_umap_close() is called, in this case we must not try to get an mm that is already undergoing process exit. In this case there is no need to wait for anything as the VMA will be destroyed by another thread soon and is already effectively 'unreachable' by userspace. BUG: unable to handle kernel NULL pointer dereference at 0000000000000000 PGD 800000012bc50067 P4D 800000012bc50067 PUD 129db5067 PMD 0 Oops: 0000 [#1] SMP PTI CPU: 1 PID: 2050 Comm: bash Tainted: G W OE 4.20.0-rc6+ #3 Hardware name: Red Hat KVM, BIOS 0.5.1 01/01/2011 RIP: 0010:__rb_erase_color+0xb9/0x280 Code: 84 17 01 00 00 48 3b 68 10 0f 84 15 01 00 00 48 89 58 08 48 89 de 48 89 ef 4c 89 e3 e8 90 84 22 00 e9 60 ff ff ff 48 8b 5d 10 03 01 0f 84 9c 00 00 00 48 8b 43 10 48 85 c0 74 09 f6 00 01 0f RSP: 0018:ffffbecfc090bab8 EFLAGS: 00010246 RAX: ffff97616346cf30 RBX: 0000000000000000 RCX: 0000000000000101 RDX: 0000000000000000 RSI: ffff97623b6ca828 RDI: ffff97621ef10828 RBP: ffff97621ef10828 R08: ffff97621ef10828 R09: 0000000000000000 R10: 0000000000000000 R11: 0000000000000000 R12: ffff97623b6ca838 R13: ffffffffbb3fef50 R14: ffff97623b6ca828 R15: 0000000000000000 FS: 00007f7a5c31d740(0000) GS:ffff97623bb00000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 0000000000000000 CR3: 000000011255a000 CR4: 00000000000006e0 Call Trace: unlink_file_vma+0x3b/0x50 free_pgtables+0xa1/0x110 exit_mmap+0xca/0x1a0 ? mlx5_ib_dealloc_pd+0x28/0x30 [mlx5_ib] mmput+0x54/0x140 uverbs_user_mmap_disassociate+0xcc/0x160 [ib_uverbs] uverbs_destroy_ufile_hw+0xf7/0x120 [ib_uverbs] ib_uverbs_remove_one+0xea/0x240 [ib_uverbs] ib_unregister_device+0xfb/0x200 [ib_core] mlx5_ib_remove+0x51/0xe0 [mlx5_ib] mlx5_remove_device+0xc1/0xd0 [mlx5_core] mlx5_unregister_device+0x3d/0xb0 [mlx5_core] remove_one+0x2a/0x90 [mlx5_core] pci_device_remove+0x3b/0xc0 device_release_driver_internal+0x16d/0x240 unbind_store+0xb2/0x100 kernfs_fop_write+0x102/0x180 __vfs_write+0x36/0x1a0 ? __alloc_fd+0xa9/0x170 ? set_close_on_exec+0x49/0x70 vfs_write+0xad/0x1a0 ksys_write+0x52/0xc0 do_syscall_64+0x5b/0x180 entry_SYSCALL_64_after_hwframe+0x44/0xa9 Cc: # 4.19 Fixes: 5f9794dc94f5 ("RDMA/ucontext: Add a core API for mmaping driver IO memory") Signed-off-by: Yishai Hadas Signed-off-by: Leon Romanovsky Signed-off-by: Jason Gunthorpe Signed-off-by: Greg Kroah-Hartman