mirror of
https://github.com/torvalds/linux.git
synced 2026-10-11 21:07:03 +02:00
smb: smbdirect: free completion queues with ib_free_cq()
smbdirect_connection_destroy_qp() creates the send and receive completion
queues with ib_alloc_cq_any(), which for IB_POLL_WORKQUEUE arms an
internal completion handler that runs ib_cq_poll_work() on a workqueue.
Tearing those CQs down with ib_destroy_cq() frees them without first
cancelling that poll work.
If the provider posts a completion late -- for example Soft-RoCE (rxe)
posting an RNR error from rxe_receiver() after rdma_destroy_qp() -- the
handler re-queues ib_cq_poll_work() on the already-freed CQ, and a
follow-on access faults in rxe_req_notify_cq().
Use ib_free_cq(), which cancel_work_sync()es the poll work before freeing
the CQ, so no completion handler can run against a freed queue.
[ 1236.599526] ==================================================================
[ 1236.602142] BUG: KASAN: slab-use-after-free in ib_cq_poll_work+0xd0/0x1a0
[ 1236.605524] Read of size 8 at addr ffff888111865800 by task kworker/4:1H/82
[ 1236.609017]
[ 1236.609270] CPU: 4 UID: 0 PID: 82 Comm: kworker/4:1H Not tainted 7.2.0-rc3-next-20260717-virtme #110 PREEMPT(lazy)
[ 1236.609287] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.17.0-debian-1.17.0-1 04/01/2014
[ 1236.609498] Workqueue: ib-comp-wq ib_cq_poll_work
[ 1236.609525] Call Trace:
[ 1236.609536] <TASK>
[ 1236.609545] __dump_stack+0x21/0x60
[ 1236.609562] dump_stack_lvl+0xc2/0x100
[ 1236.609573] print_address_description+0x77/0x200
[ 1236.609587] ? ib_cq_poll_work+0xd0/0x1a0
[ 1236.609597] print_report+0x58/0x70
[ 1236.609607] kasan_report+0x117/0x150
[ 1236.609623] ? ib_cq_poll_work+0xd0/0x1a0
[ 1236.609636] ? process_scheduled_works+0x954/0x1600
[ 1236.609650] ib_cq_poll_work+0xd0/0x1a0
[ 1236.609662] ? process_scheduled_works+0x954/0x1600
[ 1236.609674] process_scheduled_works+0xc22/0x1600
[ 1236.609698] ? __pfx_process_scheduled_works+0x10/0x10
[ 1236.609713] ? __pfx_assign_work+0x10/0x10
[ 1236.609726] ? lock_is_held_type+0x7b/0x110
[ 1236.609741] worker_thread+0x975/0xee0
[ 1236.609757] ? __pfx_do_raw_spin_lock+0x10/0x10
[ 1236.609775] ? __kthread_parkme+0x21e/0x260
[ 1236.609789] kthread+0x3a6/0x490
[ 1236.609800] ? __pfx_worker_thread+0x10/0x10
[ 1236.609809] ? __pfx_kthread+0x10/0x10
[ 1236.609820] ret_from_fork+0x55a/0xa20
[ 1236.609835] ? __pfx_ret_from_fork+0x10/0x10
[ 1236.609850] ? __pfx_kthread+0x10/0x10
[ 1236.609861] ret_from_fork_asm+0x1a/0x30
[ 1236.609880] </TASK>
[ 1236.609886]
[ 1236.661292] Allocated by task 5076:
[ 1236.662640] kasan_save_track+0x3e/0x80
[ 1236.663842] __kasan_kmalloc+0x72/0x90
[ 1236.664763] __kmalloc_noprof+0x2b0/0x5d0
[ 1236.665356] __ib_alloc_cq+0x284/0x1000
[ 1236.666573] __ib_alloc_cq_any+0x23e/0x340
[ 1236.668654] smbdirect_connection_create_qp+0x6f7/0x1070
[ 1236.669757] smbdirect_accept_connect_request+0x500/0x1ca0
[ 1236.672625] smbdirect_listen_rdma_event_handler+0x1655/0x1c50
[ 1236.673930] cma_listen_handler+0x1bf/0x260
[ 1236.674923] cma_cm_event_handler+0x128/0x380
[ 1236.676926] cma_ib_req_handler+0x2d3d/0x4de0
[ 1236.678368] cm_process_work+0xb0/0x530
[ 1236.680454] cm_queue_work_unlock+0xb1/0x230
[ 1236.681673] cm_work_handler+0x969f/0xdca0
[ 1236.682704] process_scheduled_works+0xc22/0x1600
[ 1236.683447] worker_thread+0x975/0xee0
[ 1236.685901] kthread+0x3a6/0x490
[ 1236.688164] ret_from_fork+0x55a/0xa20
[ 1236.689522] ret_from_fork_asm+0x1a/0x30
[ 1236.690073]
[ 1236.690378] Freed by task 5137:
[ 1236.692242] kasan_save_track+0x3e/0x80
[ 1236.694272] kasan_save_free_info+0x40/0x50
[ 1236.695514] __kasan_slab_free+0x3a/0x60
[ 1236.696773] kfree+0x14e/0x4e0
[ 1236.697216] ib_destroy_cq_user+0x18d/0x250
[ 1236.699817] smbdirect_connection_destroy_qp+0xf2/0x280
[ 1236.702115] smbdirect_socket_destroy_sync+0x1607/0x2720
[ 1236.704062] smbdirect_socket_release+0x140/0x280
[ 1236.705286] smb_direct_free_transport+0x3b/0x90
[ 1236.707241] __ksmbd_conn_release_work+0x99/0xf0
[ 1236.709287] process_scheduled_works+0xc22/0x1600
[ 1236.710763] worker_thread+0x975/0xee0
[ 1236.711262] kthread+0x3a6/0x490
[ 1236.711720] ret_from_fork+0x55a/0xa20
[ 1236.712232] ret_from_fork_asm+0x1a/0x30
[ 1236.712762]
[ 1236.712992] Last potentially related work creation:
[ 1236.715157] kasan_save_stack+0x3e/0x60
[ 1236.716993] kasan_record_aux_stack+0x99/0xb0
[ 1236.718864] insert_work+0xb2/0x4a0
[ 1236.720916] __queue_work+0xebb/0x1260
[ 1236.722397] queue_work_on+0x23b/0x350
[ 1236.723809] ib_cq_completion_workqueue+0xac/0x160
[ 1236.724895] rxe_cq_post+0x433/0x7c0
[ 1236.726273] rxe_receiver+0xa41/0xd0d0
[ 1236.727754] do_work+0x272/0x860
[ 1236.728896] process_scheduled_works+0xc22/0x1600
[ 1236.730026] worker_thread+0x975/0xee0
[ 1236.731499] kthread+0x3a6/0x490
[ 1236.732132] ret_from_fork+0x55a/0xa20
[ 1236.733171] ret_from_fork_asm+0x1a/0x30
[ 1236.734224]
[ 1236.734871] Second to last potentially related work creation:
[ 1236.736001] kasan_save_stack+0x3e/0x60
[ 1236.737161] kasan_record_aux_stack+0x99/0xb0
[ 1236.739074] insert_work+0xb2/0x4a0
[ 1236.740414] __queue_work+0xebb/0x1260
[ 1236.740932] queue_work_on+0x23b/0x350
[ 1236.741849] ib_cq_completion_workqueue+0xac/0x160
[ 1236.744099] rxe_cq_post+0x433/0x7c0
[ 1236.745514] rxe_receiver+0xa41/0xd0d0
[ 1236.746091] do_work+0x272/0x860
[ 1236.747187] process_scheduled_works+0xc22/0x1600
[ 1236.749060] worker_thread+0x975/0xee0
[ 1236.750224] kthread+0x3a6/0x490
[ 1236.751480] ret_from_fork+0x55a/0xa20
[ 1236.751989] ret_from_fork_asm+0x1a/0x30
[ 1236.752974]
[ 1236.753627] The buggy address belongs to the object at ffff888111865800
[ 1236.753627] which belongs to the cache kmalloc-1k of size 1024
[ 1236.757729] The buggy address is located 0 bytes inside of
[ 1236.757729] freed 1024-byte region [ffff888111865800, ffff888111865c00)
[ 1236.760816]
[ 1236.761373] The buggy address belongs to the physical page:
[ 1236.762599] page: refcount:0 mapcount:0 mapping:0000000000000000 index:0x0 pfn:0x111860
[ 1236.764306] head: order:3 mapcount:0 entire_mapcount:0 nr_pages_mapped:0 pincount:0
[ 1236.765281] flags: 0x100000000000040(head|node=0|zone=2)
[ 1236.765979] page_type: f5(slab)
[ 1236.766410] raw: 0100000000000040 ffff8881000430c0 ffffea0004a7e210 ffffea0004586210
[ 1236.770418] raw: 0000000000000000 00000000000a000a 00000000f5000000 0000000000000000
[ 1236.775899] head: 0100000000000040 ffff8881000430c0 ffffea0004a7e210 ffffea0004586210
[ 1236.782823] head: 0000000000000000 00000000000a000a 00000000f5000000 0000000000000000
[ 1236.786239] head: 0100000000000003 fffffffffffffe01 00000000ffffffff 00000000ffffffff
[ 1236.790658] head: 0000000000000000 0000000000000000 00000000ffffffff 0000000000000000
[ 1236.794132] page dumped because: kasan: bad access detected
[ 1236.798301]
[ 1236.799640] Memory state around the buggy address:
[ 1236.802028] ffff888111865700: fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc
[ 1236.806254] ffff888111865780: fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc
[ 1236.809036] >ffff888111865800: fa fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb
[ 1236.813968] ^
[ 1236.816416] ffff888111865880: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb
[ 1236.819454] ffff888111865900: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb
[ 1236.823143] ==================================================================
[ 1236.830365] Disabling lock debugging due to kernel taint
[ 1236.831136] BUG: unable to handle page fault for address: ffffc90006dc8080
[ 1236.838157] #PF: supervisor read access in kernel mode
[ 1236.843686] #PF: error_code(0x0000) - not-present page
[ 1236.849393] PGD 100000067 P4D 100000067 PUD 100366067 PMD 12913e067 PTE 0
[ 1236.854156] Oops: Oops: 0000 [#1] SMP KASAN NOPTI
[ 1236.857893] CPU: 4 UID: 0 PID: 82 Comm: kworker/4:1H Tainted: G B 7.2.0-rc3-next-20260717-virtme #110 PREEMPT(lazy)
[ 1236.860893] ksmbd: smb_direct: smbdirect_connection_recv_io_refill() failed -ECONNRESET
[ 1236.864209] Tainted: [B]=BAD_PAGE
[ 1236.864220] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.17.0-debian-1.17.0-1 04/01/2014
[ 1236.864233] Workqueue: ib-comp-wq ib_cq_poll_work
[ 1236.878314] RIP: 0010:rxe_req_notify_cq+0x13a/0x350
[ 1236.881683] Code: 20 87 fd 4c 89 fe 48 ba 00 00 00 00 00 fc ff df 4c 8b 3e 49 83 ef 80 4c 89 f8 48 c1 e8 03 0f b6 04 10 84 c0 0f 85 9b 01 00 00 <45> 8b 2f 41 80 3c 16 00 74 18 49 89 f6 48 89 f7 e8 71 20 87 fd 4c
[ 1236.886819] ksmbd: smb_direct: smbdirect_connection_recv_io_refill() failed -ECONNRESET
[ 1236.890671] RSP: 0018:ffff88810222f920 EFLAGS: 00010046
[ 1236.890705] RAX: 0000000000000000 RBX: ffff88810222f920 RCX: ffffffff84c92863
[ 1236.901613] RDX: dffffc0000000000 RSI: ffff888120456d48 RDI: ffff888120456d48
[ 1236.903979] RBP: ffff88810222fa20 R08: 0000000000000003 R09: 0000000000000004
[ 1236.908958] R10: dffffc0000000000 R11: ffffed1020445f10 R12: ffff888120456d40
[ 1236.914473] R13: dffffc0000000000 R14: 1ffff1102408ada9 R15: ffffc90006dc8080
[ 1236.918946] FS: 0000000000000000(0000) GS:ffff88842600d000(0000) knlGS:0000000000000000
[ 1236.921600] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 1236.925779] CR2: ffffc90006dc8080 CR3: 000000012b35f003 CR4: 0000000000f72ef0
[ 1236.926809] ksmbd: smb_direct: smbdirect_connection_recv_io_refill() failed -ECONNRESET
[ 1236.928302] PKRU: 55555554
[ 1236.928328] Call Trace:
[ 1236.928338] <TASK>
[ 1236.928352] ? ib_cq_poll_work+0xd0/0x1a0
[ 1236.928374] ? __pfx_rxe_req_notify_cq+0x10/0x10
[ 1236.941601] ? ib_cq_poll_work+0xd0/0x1a0
[ 1236.943306] ib_cq_poll_work+0xfe/0x1a0
[ 1236.943961] ? process_scheduled_works+0x954/0x1600
[ 1236.947036] process_scheduled_works+0xc22/0x1600
[ 1236.951626] ? __pfx_process_scheduled_works+0x10/0x10
[ 1236.954316] ? __pfx_assign_work+0x10/0x10
[ 1236.958110] ? lock_is_held_type+0x7b/0x110
[ 1236.960042] worker_thread+0x975/0xee0
[ 1236.962668] ? __pfx_do_raw_spin_lock+0x10/0x10
[ 1236.965334] ? __kthread_parkme+0x21e/0x260
[ 1236.966058] kthread+0x3a6/0x490
[ 1236.968115] ? __pfx_worker_thread+0x10/0x10
[ 1236.971020] ? __pfx_kthread+0x10/0x10
[ 1236.974488] ret_from_fork+0x55a/0xa20
[ 1236.977419] ? __pfx_ret_from_fork+0x10/0x10
[ 1236.979846] ? __pfx_kthread+0x10/0x10
[ 1236.981238] ret_from_fork_asm+0x1a/0x30
[ 1236.984086] </TASK>
[ 1236.986181] Modules linked in:
[ 1236.989048] CR2: ffffc90006dc8080
[ 1236.990412] ---[ end trace 0000000000000000 ]---
[ 1236.994119] RIP: 0010:rxe_req_notify_cq+0x13a/0x350
[ 1236.998482] Code: 20 87 fd 4c 89 fe 48 ba 00 00 00 00 00 fc ff df 4c 8b 3e 49 83 ef 80 4c 89 f8 48 c1 e8 03 0f b6 04 10 84 c0 0f 85 9b 01 00 00 <45> 8b 2f 41 80 3c 16 00 74 18 49 89 f6 48 89 f7 e8 71 20 87 fd 4c
[ 1237.006635] RSP: 0018:ffff88810222f920 EFLAGS: 00010046
[ 1237.008513] RAX: 0000000000000000 RBX: ffff88810222f920 RCX: ffffffff84c92863
[ 1237.014100] RDX: dffffc0000000000 RSI: ffff888120456d48 RDI: ffff888120456d48
[ 1237.021058] RBP: ffff88810222fa20 R08: 0000000000000003 R09: 0000000000000004
[ 1237.024939] R10: dffffc0000000000 R11: ffffed1020445f10 R12: ffff888120456d40
[ 1237.030203] R13: dffffc0000000000 R14: 1ffff1102408ada9 R15: ffffc90006dc8080
[ 1237.034299] FS: 0000000000000000(0000) GS:ffff88842600d000(0000) knlGS:0000000000000000
[ 1237.037311] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 1237.042838] CR2: ffffc90006dc8080 CR3: 000000012b35f003 CR4: 0000000000f72ef0
Found with ksmbdzzer [2], a KSMBD fuzzer that drives libFuzzer with a
kcov-dataflow [1] coverage vector: it folds each instrumented
comparison/argument's runtime operand value together with its PC (the
default arm mixes them as pc⊕val) so that a new operand value at a known
site counts as new coverage.
[1] https://lwn.net/Articles/1077606/
[2] https://github.com/yskzalloc/kcov-dataflow
Fixes: 6073eb3e31 ("smb: smbdirect: introduce smbdirect_connection_{create,destroy}_qp()")
Signed-off-by: Yunseong Kim <yunseong.kim@est.tech>
Acked-by: Stefan Metzmacher <metze@samba.org>
Signed-off-by: Namjae Jeon <linkinjeon@kernel.org>
This commit is contained in:
Yunseong Kim
Namjae Jeon
parent
7405d0ba29
commit
fe2c0cacbc
|
|
@ -403,12 +403,21 @@ void smbdirect_connection_destroy_qp(struct smbdirect_socket *sc)
|
|||
sc->ib.qp = NULL;
|
||||
rdma_destroy_qp(sc->rdma.cm_id);
|
||||
}
|
||||
/*
|
||||
* These CQs were created with ib_alloc_cq_any(), which arms an internal
|
||||
* completion handler (ib_cq_poll_work for IB_POLL_WORKQUEUE). They MUST be
|
||||
* torn down with ib_free_cq(), which cancel_work_sync()es that poll work
|
||||
* before freeing the CQ. ib_destroy_cq() skips that step, so a completion
|
||||
* posted late by the (software) provider — e.g. rxe posting an RNR error
|
||||
* from rxe_receiver after rdma_destroy_qp() — re-queues ib_cq_poll_work on
|
||||
* an already-freed CQ (KASAN slab-use-after-free in ib_cq_poll_work).
|
||||
*/
|
||||
if (sc->ib.recv_cq) {
|
||||
ib_destroy_cq(sc->ib.recv_cq);
|
||||
ib_free_cq(sc->ib.recv_cq);
|
||||
sc->ib.recv_cq = NULL;
|
||||
}
|
||||
if (sc->ib.send_cq) {
|
||||
ib_destroy_cq(sc->ib.send_cq);
|
||||
ib_free_cq(sc->ib.send_cq);
|
||||
sc->ib.send_cq = NULL;
|
||||
}
|
||||
if (sc->ib.pd) {
|
||||
|
|
|
|||
Loading…
Reference in New Issue
Block a user