FYI, we noticed the below changes on
git://internal_merge_and_test_tree revert-18c5809b7e1183fc5ca4d104b99d53dc77723ef9-18c5809b7e1183fc5ca4d104b99d53dc77723ef9
commit 18c5809b7e1183fc5ca4d104b99d53dc77723ef9 ("fs: clear file privilege bits when mmap writing")
+------------------------------------------------------------------+----------+------------+
| | v4.4-rc3 | 18c5809b7e |
+------------------------------------------------------------------+----------+------------+
| boot_successes | 0 | 0 |
| boot_failures | 39 | 6 |
| BUG:kernel_test_oversize | 4 | |
| BUG:kernel_boot_hang | 34 | 1 |
| invoked_oom-killer:gfp_mask=0x | 1 | 1 |
| Mem-Info | 1 | 1 |
| Kernel_panic-not_syncing:Out_of_memory_and_no_killable_processes | 1 | 1 |
| backtrace:power_supply_changed_work | 1 | |
| INFO:possible_circular_locking_dependency_detected | 0 | 4 |
| backtrace:iterate_dir | 0 | 4 |
| backtrace:SyS_getdents | 0 | 4 |
| backtrace:vm_mmap_pgoff | 0 | 4 |
| backtrace:SyS_mmap_pgoff | 0 | 4 |
| backtrace:SyS_mmap | 0 | 4 |
| backtrace:kmem_cache_alloc | 0 | 1 |
| backtrace:device_private_init | 0 | 1 |
| backtrace:serio_handle_event | 0 | 1 |
+------------------------------------------------------------------+----------+------------+
[ 54.241020] systemd-sysv-generator[171]: Ignoring K01watchdog symlink in rc6.d, not generating watchdog.service.
[ 54.918926]
[ 54.919580] ======================================================
[ 54.921207] [ INFO: possible circular locking dependency detected ]
[ 54.922873] 4.4.0-rc3-00001-g18c5809b7 #1 Not tainted
[ 54.925715] -------------------------------------------------------
[ 54.929512] systemd-journal/185 is trying to acquire lock:
[ 54.929544] (&sb->s_type->i_mutex_key#7){+.+.+.}, at: [<ffffffff81253416>] do_mmap+0x3b6/0x670
[ 54.929545]
[ 54.929545] but task is already holding lock:
[ 54.929551] (&mm->mmap_sem){++++++}, at: [<ffffffff8122e6dc>] vm_mmap_pgoff+0x4c/0xd0
[ 54.929552]
[ 54.929552] which lock already depends on the new lock.
[ 54.929552]
[ 54.929553]
[ 54.929553] the existing dependency chain (in reverse order) is:
[ 54.929556]
[ 54.929556] -> #1 (&mm->mmap_sem){++++++}:
[ 54.929566] [<ffffffff81155f61>] lock_acquire+0xa1/0xd0
[ 54.929570] [<ffffffff8124953d>] __might_fault+0x9d/0xe0
[ 54.929586] [<ffffffff812bfa2c>] filldir+0xdc/0x1b0
[ 54.929594] [<ffffffff812e2089>] dcache_readdir+0x249/0x370
[ 54.929597] [<ffffffff812bfbb5>] iterate_dir+0xb5/0x1a0
[ 54.929600] [<ffffffff812bfe36>] SyS_getdents+0xb6/0x180
[ 54.929617] [<ffffffff81d844f6>] entry_SYSCALL_64_fastpath+0x16/0x7a
[ 54.929622]
[ 54.929622] -> #0 (&sb->s_type->i_mutex_key#7){+.+.+.}:
[ 54.929625] [<ffffffff81155401>] __lock_acquire+0x1e71/0x2150
[ 54.929628] [<ffffffff81155f61>] lock_acquire+0xa1/0xd0
[ 54.929637] [<ffffffff81d7ca89>] mutex_lock_nested+0x99/0x620
[ 54.929641] [<ffffffff81253416>] do_mmap+0x3b6/0x670
[ 54.929644] [<ffffffff8122e70d>] vm_mmap_pgoff+0x7d/0xd0
[ 54.929647] [<ffffffff81250144>] SyS_mmap_pgoff+0x214/0x2e0
[ 54.929662] [<ffffffff8100dd06>] SyS_mmap+0x26/0x40
[ 54.929666] [<ffffffff81d844f6>] entry_SYSCALL_64_fastpath+0x16/0x7a
[ 54.929667]
[ 54.929667] other info that might help us debug this:
[ 54.929667]
[ 54.929668] Possible unsafe locking scenario:
[ 54.929668]
[ 54.929668] CPU0 CPU1
[ 54.929669] ---- ----
[ 54.929671] lock(&mm->mmap_sem);
[ 54.929673] lock(&sb->s_type->i_mutex_key#7);
[ 54.929675] lock(&mm->mmap_sem);
[ 54.929677] lock(&sb->s_type->i_mutex_key#7);
[ 54.929678]
[ 54.929678] *** DEADLOCK ***
[ 54.929678]
[ 54.929679] 1 lock held by systemd-journal/185:
[ 54.929685] #0: (&mm->mmap_sem){++++++}, at: [<ffffffff8122e6dc>] vm_mmap_pgoff+0x4c/0xd0
[ 54.929686]
[ 54.929686] stack backtrace:
[ 54.929700] CPU: 0 PID: 185 Comm: systemd-journal Not tainted 4.4.0-rc3-00001-g18c5809b7 #1
[ 54.929701] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS Debian-1.8.2-1 04/01/2014
[ 54.929714] ffffffff8357a530 ffff88001426fc70 ffffffff816bd2d8 ffffffff8357a530
[ 54.929717] ffff88001426fcb0 ffffffff8114e758 ffff88001426fd00 ffff880000b74bc0
[ 54.929790] ffff880000b74b88 0000000000000001 0000000000000001 ffff880000b74040
[ 54.929793] Call Trace:
[ 54.929814] [<ffffffff816bd2d8>] dump_stack+0x7c/0xb4
[ 54.929817] [<ffffffff8114e758>] print_circular_bug+0x2c8/0x370
[ 54.929820] [<ffffffff81155401>] __lock_acquire+0x1e71/0x2150
[ 54.929831] [<ffffffff8112fb7c>] ? local_clock+0x4c/0x60
[ 54.929834] [<ffffffff81155f61>] lock_acquire+0xa1/0xd0
[ 54.929837] [<ffffffff81253416>] ? do_mmap+0x3b6/0x670
[ 54.929841] [<ffffffff81d7ca89>] mutex_lock_nested+0x99/0x620
[ 54.929843] [<ffffffff81253416>] ? do_mmap+0x3b6/0x670
[ 54.929846] [<ffffffff81253416>] ? do_mmap+0x3b6/0x670
[ 54.929849] [<ffffffff81253416>] do_mmap+0x3b6/0x670
[ 54.929851] [<ffffffff8122e70d>] vm_mmap_pgoff+0x7d/0xd0
[ 54.929854] [<ffffffff81250144>] SyS_mmap_pgoff+0x214/0x2e0
[ 54.929857] [<ffffffff8100dd06>] SyS_mmap+0x26/0x40
[ 54.929860] [<ffffffff81d844f6>] entry_SYSCALL_64_fastpath+0x16/0x7a
To reproduce:
git clone git://git.kernel.org/pub/scm/linux/kernel/git/wfg/lkp-tests.git
cd lkp-tests
Bin/lkp install job.yaml # job file is attached in this email
bin/lkp run job.yaml
Thanks,
Ying Huang