In the Linux kernel, the following vulnerability has been resolved:<\/p>\n
nvme: fix cdev lifetime<\/p>\n
Sashiko bot reported a potential problem for the cdev lifetime in [0]
\n– the code there is heavily based on the NVMe code.<\/p>\n
Currently the NS head .open and .release file_operations methods take and
\nput a reference to the nvme_ns_head to ensure that this structure does not
\ndisappear while we open fds for that cdev.<\/p>\n
In multipath mode, when we teardown the NS head, we call nvme_cdev_del() ->
\ncdev_device_del() -> cdev_del(). However after cdev_del() returns, cdevs
\nalready open will remain and their fops will still be callable. As such,
\nwe can still reference the cdev after the nvme_ns_head reference count
\ndrops to 0 (and is freed).<\/p>\n
This can be shown with an application which delays between opening the cdev
\nand issuing the ioctl while the NS head is being torn down:<\/p>\n
# .\/ioctl_file \/dev\/ng1n1 &
\n# waiting 10 seconds ….<\/p>\n
# .\/ini_nvme_teardown.sh
\n[ 21.221718] nvme nvme1: Removing ctrl: NQN “nvme-test-target”
\n[ 21.274609] nvme nvme2: Removing ctrl: NQN “nvme-test-target”
\n# now going to issue ioctl ….
\n[ 26.549285] ==================================================================
\n[ 26.550841] BUG: KASAN: slab-use-after-free in cdev_put.part.0+0x3d\/0x40
\n[ 26.552352] Read of size 8 at addr ffff88811e7fa170 by task ioctl_file\/237
\n[ 26.553805]
\n[ 26.554227] CPU: 3 UID: 0 PID: 237 Comm: ioctl_file Not tainted 7.2.0-rc1-00004-g6852a10e32d4 #921 PREEMPT(lazy)
\n[ 26.554236] Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS 1.16.3-debian-1.16.3-2 04\/01\/2014
\n[ 26.554241] Call Trace:
\n[ 26.554245]
\n[ 26.554248] dump_stack_lvl+0x68\/0xa0
\n[ 26.554266] print_report+0x10d\/0x5d0
\n[ 26.554276] ? __virt_addr_valid+0x21d\/0x3f0
\n[ 26.554287] ? cdev_put.part.0+0x3d\/0x40
\n[ 26.554292] kasan_report+0x96\/0xd0
\n[ 26.554300] ? cdev_put.part.0+0x3d\/0x40
\n[ 26.554307] cdev_put.part.0+0x3d\/0x40
\n[ 26.554313] __fput+0x7bc\/0xa70
\n[ 26.554322] fput_close_sync+0xd8\/0x190
\n[ 26.554328] ? __pfx_fput_close_sync+0x10\/0x10
\n[ 26.554337] __x64_sys_close+0x79\/0xd0
\n[ 26.554344] do_syscall_64+0x117\/0x6b0
\n[ 26.554351] entry_SYSCALL_64_after_hwframe+0x77\/0x7f
\n[ 26.554358] RIP: 0033:0x7f938c067727
\n[ 26.554364] Code: 48 89 fa 4c 89 df e8 28 ad 00 00 8b 93 08 03 00 00 59 5e 48 83 f8 fc 74 1a 5b c3 0f 1f 84 00 00 00 00 00 48 8b 44 24 10 0f 05 <5bf
\n[ 26.554369] RSP: 002b:00007fff49f05980 EFLAGS: 00000202 ORIG_RAX: 0000000000000003
\n[ 26.554376] RAX: ffffffffffffffda RBX: 00007f938bfd7780 RCX: 00007f938c067727
\n[ 26.554380] RDX: 0000000000000000 RSI: 0000000000000000 RDI: 0000000000000003
\n[ 26.554383] RBP: 00007fff49f05a10 R08: 0000000000000000 R09: 0000000000000000
\n[ 26.554386] R10: 0000000000000000 R11: 0000000000000202 R12: 0000000000000000
\n[ 26.554389] R13: 00007fff49f05b40 R14: 00007f938c207000 R15: 000055c148471d78
\n[ 26.554397]
\n[ 26.554399]
\n[ 26.575612] Allocated by task 100:
\n[ 26.575871] kasan_save_stack+0x24\/0x50
\n[ 26.576157] kasan_save_track+0x14\/0x30
\n[ 26.576418] __kasan_kmalloc+0x7f\/0x90
\n[ 26.576668] __kmalloc_noprof+0x281\/0x6c0
\n[ 26.576938] nvme_alloc_ns+0x7f7\/0x3170
\n[ 26.577206] nvme_scan_ns+0x508\/0x880
\n[ 26.577449] async_run_entry_fn+0x8c\/0x350
\n[ 26.577723] process_scheduled_works+0xb6f\/0x1a00
\n[ 26.578034] worker_thread+0x4ad\/0xb40
\n[ 26.578283] kthread+0x34f\/0x450
\n[ 26.578501] ret_from_fork+0x563\/0x800
\n[ 26.578752] ret_from_fork_asm+0x1a\/0x30
\n[ 26.579012]
\n[ 26.579124] Freed by task 237:
\n[ 26.579335] kasan_save_stack+0x24\/0x50
\n[ 26.579596] kasan_save_track+0x14\/0x30
\n[ 26.579855] kasan_save_free_info+0x3a\/0x60
\n[ 26.580131] __kasan_slab_free+0x43\/0x70
\n[ 26.580388] kfree+0x321\/0x500
\n[ 26.580591] nvme_ns_head_chr_release+0x39\/0x50
\n[ 26.580883] __fput+0x352\/0xa70
\n[ 26.581095] fput_close_sync+0xd8\/0x190
\n[ 26.581350] __x64_sys_close+0x79\/0xd0
\n[ 26.581595] do_syscall_64+0x117\/0x6b0
\n[ 26.581842] entry_SYSCAL
\n—truncated—<\/p>\n","protected":false},"excerpt":{"rendered":"
In the Linux kernel, the following vulnerability has been resolved:<\/p>\n
nvme: fix cdev lifetime<\/p>\n
Sashiko bot reported a potential problem for the cdev lifetime in [0]
\n– the code there is heavily based on<\/p>\n","protected":false},"template":"","meta":[],"categories":[],"tags":[],"class_list":["post-41287","vuln","type-vuln","status-publish","hentry"],"_links":{"self":[{"href":"https:\/\/www.54iter.cn\/index.php?rest_route=\/wp\/v2\/vuln\/41287","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.54iter.cn\/index.php?rest_route=\/wp\/v2\/vuln"}],"about":[{"href":"https:\/\/www.54iter.cn\/index.php?rest_route=\/wp\/v2\/types\/vuln"}],"wp:attachment":[{"href":"https:\/\/www.54iter.cn\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=41287"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.54iter.cn\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=41287"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.54iter.cn\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=41287"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}