FYI, we noticed the following commit (built with gcc-7):
commit: e692535ca01126a764ee1c3ac1b352c88c10070b ("hugetlbfs: Convert to fs_context")
https://git.kernel.org/cgit/linux/kernel/git/viro/vfs.git mount-fixes2
in testcase: boot
on test machine: qemu-system-x86_64 -enable-kvm -cpu host -smp 2 -m 2G
caused below changes (please refer to attached dmesg/kmsg for entire log/backtrace):
+-------------------------------------------------------------------------------------------------------------+------------+------------+
| | 4531d23f8f | e692535ca0 |
+-------------------------------------------------------------------------------------------------------------+------------+------------+
| boot_successes | 0 | 0 |
| boot_failures | 20 | 16 |
| kernel_BUG_at_include/linux/fs.h | 20 | |
| invalid_opcode:#[##] | 20 | |
| RIP:__fput | 20 | |
| Kernel_panic-not_syncing:Fatal_exception | 20 | |
| BUG_hugetlbfs_inode_cache(Not_tainted):Objects_remaining_in_hugetlbfs_inode_cache_on__kmem_cache_shutdown() | 0 | 16 |
| INFO:Slab0x(ptrval)objects=#used=#fp=0x(ptrval)flags= | 0 | 16 |
| INFO:Object0x(ptrval)@offset= | 0 | 16 |
+-------------------------------------------------------------------------------------------------------------+------------+------------+
[ 1.997041] hpet0: 3 comparators, 64-bit 100.000000 MHz counter
[ 2.003127] clocksource: Switched to clocksource kvm-clock
[ 2.245168] VFS: Disk quotas dquot_6.6.0
[ 2.252987] VFS: Dquot-cache hash table entries: 512 (order 0, 4096 bytes)
[ 2.264260] =============================================================================
[ 2.265034] BUG hugetlbfs_inode_cache (Not tainted): Objects remaining in hugetlbfs_inode_cache on __kmem_cache_shutdown()
[ 2.265034] -----------------------------------------------------------------------------
[ 2.265034]
[ 2.265034] Disabling lock debugging due to kernel taint
[ 2.265034] INFO: Slab 0x (ptrval) objects=25 used=2 fp=0x (ptrval) flags=0xfffffc0008100
[ 2.265034] CPU: 0 PID: 1 Comm: swapper/0 Tainted: G B 4.17.0-rc5-00190-ge692535 #1
[ 2.265034] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.10.2-1 04/01/2014
[ 2.265034] Call Trace:
[ 2.265034] dump_stack+0x5c/0x7b
[ 2.265034] slab_err+0x9e/0xb0
[ 2.265034] ? on_each_cpu_cond+0xb2/0x150
[ 2.265034] ? __kmalloc+0x1ee/0x230
[ 2.265034] __kmem_cache_shutdown+0x188/0x3a0
[ 2.265034] ? cpumask_next+0x17/0x20
[ 2.265034] shutdown_cache+0x15/0x1b0
[ 2.265034] kmem_cache_destroy+0x1e0/0x1f0
[ 2.265034] init_hugetlbfs_fs+0x85/0x161
[ 2.265034] ? init_ramfs_fs+0x20/0x20
[ 2.265034] ? set_debug_rodata+0x11/0x11
[ 2.265034] do_one_initcall+0x36/0x1bb
[ 2.265034] kernel_init_freeable+0x1ab/0x240
[ 2.265034] ? rest_init+0xd0/0xd0
[ 2.265034] kernel_init+0xa/0x110
[ 2.265034] ret_from_fork+0x35/0x40
[ 2.265034] INFO: Object 0x (ptrval) @offset=0
[ 2.265034] INFO: Object 0x (ptrval) @offset=632
[ 2.518522] kmem_cache_destroy hugetlbfs_inode_cache: Slab cache still has objects
[ 2.530732] CPU: 0 PID: 1 Comm: swapper/0 Tainted: G B 4.17.0-rc5-00190-ge692535 #1
[ 2.531633] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.10.2-1 04/01/2014
[ 2.531633] Call Trace:
[ 2.531633] dump_stack+0x5c/0x7b
[ 2.531633] kmem_cache_destroy+0x1d3/0x1f0
[ 2.531633] init_hugetlbfs_fs+0x85/0x161
[ 2.531633] ? init_ramfs_fs+0x20/0x20
[ 2.531633] ? set_debug_rodata+0x11/0x11
[ 2.531633] do_one_initcall+0x36/0x1bb
[ 2.531633] kernel_init_freeable+0x1ab/0x240
[ 2.531633] ? rest_init+0xd0/0xd0
[ 2.531633] kernel_init+0xa/0x110
[ 2.531633] ret_from_fork+0x35/0x40
[ 2.653396] pnp: PnP ACPI init
[ 2.660462] pnp 00:00: Plug and Play ACPI device, IDs PNP0b00 (active)
[ 2.672477] pnp 00:01: Plug and Play ACPI device, IDs PNP0303 (active)
[ 2.685601] pnp 00:02: Plug and Play ACPI device, IDs PNP0f13 (active)
[ 2.698958] pnp 00:03: [dma 2]
To reproduce:
git clone https://github.com/intel/lkp-tests.git
cd lkp-tests
bin/lkp qemu -k <bzImage> job-script # job-script is attached in this email
Thanks,
Xiaolong