FYI, we noticed the following commit (built with gcc-7): commit: 8dab1f5c1e6a0b2cb17237035efaf019553d79ea ("[RFC PATCH 1/2] mm/vmalloc: keep track of free blocks for allocation") url: https://github.com/0day-ci/linux/commits/Uladzislau-Rezki-Sony/improve-vmalloc-allocation/20181021-001707 base: https://github.com/thesofproject/linux master in testcase: ftq with following parameters: nr_task: 100% samples: 100000ss test: add freq: 1000 ucode: 0xb00002e cpufreq_governor: performance test-description: The FTQ benchmarks measure hardware and software interference or 'noise' on a node from the applications perspective. test-url: https://github.com/rminnich/ftq on test machine: 192 threads Intel(R) Xeon(R) CPU E7-8890 v4 @ 2.20GHz with 512G memory caused below changes (please refer to attached dmesg/kmsg for entire log/backtrace): +-----------------------------------------------------------------+------------+------------+ | | dd52cb8790 | 8dab1f5c1e | +-----------------------------------------------------------------+------------+------------+ | boot_successes | 36 | 0 | | boot_failures | 11 | 12 | | WARNING:stack_recursion | 11 | | | WARNING:at#for_ip_swapgs_restore_regs_and_return_to_usermode/0x | 10 | | | WARNING:stack_going_in_the_wrong_direction?ip=_cond_resched/0x | 1 | | | WARNING:stack_going_in_the_wrong_direction?ip=__schedule/0x | 1 | | | BUG:soft_lockup-CPU##stuck_for#s | 0 | 11 | | RIP:pcpu_get_vm_areas | 0 | 12 | | Kernel_panic-not_syncing:softlockup:hung_tasks | 0 | 11 | | kernel_BUG_at_mm/vmalloc.c | 0 | 1 | | invalid_opcode:#[##] | 0 | 1 | | Kernel_panic-not_syncing:Fatal_exception | 0 | 1 | +-----------------------------------------------------------------+------------+------------+ [ 37.496001] watchdog: BUG: soft lockup - CPU#97 stuck for 22s! [kworker/97:1:825] [ 37.496001] Modules linked in: [ 37.496001] CPU: 97 PID: 825 Comm: kworker/97:1 Not tainted 4.17.0-rc7-00044-g8dab1f5 #1 [ 37.496001] Hardware name: Intel Corporation BRICKLAND/BRICKLAND, BIOS BRBDXSD1.86B.0338.R00.1603162127 03/16/2016 [ 37.496001] Workqueue: events pcpu_balance_workfn [ 37.496001] RIP: 0010:pcpu_get_vm_areas+0x464/0xa90 [ 37.496001] RSP: 0000:ffffc9000e57bd50 EFLAGS: 00000296 ORIG_RAX: ffffffffffffff13 [ 37.496001] RAX: ffff885bf9eaa078 RBX: ffff885bf9eaa048 RCX: ffff885bf9eaa078 [ 37.496001] RDX: ffff885bf9eaa088 RSI: ffffe8c00d000000 RDI: 0000000000000000 [ 37.496001] RBP: ffffe8e00d000000 R08: ffffffff82516cd0 R09: ffff883bf9b19260 [ 37.496001] R10: ffff881c0ec03038 R11: ffff881c0ec03028 R12: ffff885bf9a8b1c0 [ 37.496001] R13: ffff885bf9b9fea0 R14: 0000000000000010 R15: ffff885bf9a8b1a0 [ 37.496001] FS: 0000000000000000(0000) GS:ffff881c0f840000(0000) knlGS:0000000000000000 [ 37.496001] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 37.496001] CR2: 0000000000000000 CR3: 000000000240a001 CR4: 00000000003606e0 [ 37.496001] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 [ 37.496001] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 [ 37.496001] Call Trace: [ 37.496001] ? __kmalloc+0x1ee/0x230 [ 37.496001] pcpu_create_chunk+0x10a/0x1b0 [ 37.496001] pcpu_balance_workfn+0x27a/0x620 [ 37.496001] ? pick_next_task_fair+0x134/0x600 [ 37.496001] ? __switch_to+0x92/0x4d0 [ 37.496001] ? put_prev_entity+0x20/0x100 [ 37.496001] process_one_work+0x182/0x3a0 [ 37.496001] worker_thread+0x3a/0x3b0 [ 37.496001] ? process_one_work+0x3a0/0x3a0 [ 37.496001] kthread+0x11e/0x140 [ 37.496001] ? kthread_associate_blkcg+0xb0/0xb0 [ 37.496001] ret_from_fork+0x35/0x40 [ 37.496001] Code: ea 01 48 21 d0 48 89 43 08 48 8b 05 7f e0 c7 01 48 85 c0 0f 84 bd 03 00 00 48 8b 33 eb 03 48 89 c8 48 3b 70 e8 48 8d 50 10 72 12 <48> 8b 78 e0 48 39 7b 08 0f 86 91 05 00 00 48 8d 50 08 48 8b 0a [ 37.496001] Kernel panic - not syncing: softlockup: hung tasks [ 37.496001] CPU: 97 PID: 825 Comm: kworker/97:1 Tainted: G L 4.17.0-rc7-00044-g8dab1f5 #1 [ 37.496001] Hardware name: Intel Corporation BRICKLAND/BRICKLAND, BIOS BRBDXSD1.86B.0338.R00.1603162127 03/16/2016 [ 37.496001] Workqueue: events pcpu_balance_workfn [ 37.496001] Call Trace: [ 37.496001] [ 37.496001] dump_stack+0x5c/0x7b [ 37.496001] panic+0xd5/0x242 [ 37.496001] watchdog_timer_fn+0x20f/0x220 [ 37.496001] ? watchdog+0x30/0x30 [ 37.496001] __hrtimer_run_queues+0xfa/0x290 [ 37.496001] hrtimer_interrupt+0xe5/0x240 [ 37.496001] smp_apic_timer_interrupt+0x62/0x140 [ 37.496001] apic_timer_interrupt+0xf/0x20 [ 37.496001] [ 37.496001] RIP: 0010:pcpu_get_vm_areas+0x464/0xa90 [ 37.496001] RSP: 0000:ffffc9000e57bd50 EFLAGS: 00000296 ORIG_RAX: ffffffffffffff13 [ 37.496001] RAX: ffff885bf9eaa078 RBX: ffff885bf9eaa048 RCX: ffff885bf9eaa078 [ 37.496001] RDX: ffff885bf9eaa088 RSI: ffffe8c00d000000 RDI: 0000000000000000 [ 37.496001] RBP: ffffe8e00d000000 R08: ffffffff82516cd0 R09: ffff883bf9b19260 [ 37.496001] R10: ffff881c0ec03038 R11: ffff881c0ec03028 R12: ffff885bf9a8b1c0 [ 37.496001] R13: ffff885bf9b9fea0 R14: 0000000000000010 R15: ffff885bf9a8b1a0 [ 37.496001] ? __kmalloc+0x1ee/0x230 [ 37.496001] pcpu_create_chunk+0x10a/0x1b0 [ 37.496001] pcpu_balance_workfn+0x27a/0x620 [ 37.496001] ? pick_next_task_fair+0x134/0x600 [ 37.496001] ? __switch_to+0x92/0x4d0 [ 37.496001] ? put_prev_entity+0x20/0x100 [ 37.496001] process_one_work+0x182/0x3a0 [ 37.496001] worker_thread+0x3a/0x3b0 [ 37.496001] ? process_one_work+0x3a0/0x3a0 [ 37.496001] kthread+0x11e/0x140 [ 37.496001] ? kthread_associate_blkcg+0xb0/0xb0 [ 37.496001] ret_from_fork+0x35/0x40 ACPI MEMORY or I/O RESET_REG. To reproduce: git clone https://github.com/intel/lkp-tests.git cd lkp-tests bin/lkp install job.yaml # job file is attached in this email bin/lkp run job.yaml Thanks, Rong Chen