Hi folks,
Since this commit:
commit edb0872f44ec9976ea6d052cb4b93cd2d23ac2ba
Author: Christoph Hellwig <hch@xxxxxx>
Date: Mon Aug 9 16:17:43 2021 +0200
block: move the bdi from the request_queue to the gendisk
Looping xfstests generic/108 or xfs/279 on mountpoints with fsdax
enabled can lead to panic like this:
[ 373.882656] BUG: kernel NULL pointer dereference, address: 0000000000000300
[ 373.893342] #PF: supervisor read access in kernel mode
[ 373.899082] #PF: error_code(0x0000) - not-present page
[ 373.904820] PGD 0 P4D 0
[ 373.907647] Oops: 0000 [#1] SMP PTI
[ 373.911542] CPU: 5 PID: 0 Comm: swapper/5 Kdump: loaded Tainted: G I 5.14.0-rc4+ #1
[ 373.921544] Hardware name: Lenovo ThinkSystem SR630 -[7X02CTO1WW]-/-[7X02CTO1WW]-, BIOS -[IVE116S-1.20]- 02/08/2018
[ 373.933193] RIP: 0010:wb_timer_fn+0x3a/0x3c0
[ 373.937966] Code: 60 4c 8b 67 50 8b 9d 98 00 00 00 8b 95 b8 00 00 00 8b 85 d8 00 00 00 4c 8b 6d 28 01 d3 01 c3 48 8b 45 60 48 8b 80 90 00 00 00 <48> 8b 80 00 03 00 00 4c 8b b0 98 00 00 00 4d 85 ed 0f 84 ca 00 00
[ 373.958915] RSP: 0018:ffffaa0f465e4eb0 EFLAGS: 00010246
[ 373.964748] RAX: 0000000000000000 RBX: 0000000000000000 RCX: 0000000000000020
[ 373.972712] RDX: 0000000000000000 RSI: 0000000000000000 RDI: ffff937c41649500
[ 373.980676] RBP: ffff937bf0f38400 R08: 0000000000000020 R09: 0000000000000000
[ 373.988641] R10: 0000000000000020 R11: 000000000000003d R12: ffff937d2bbf0d20
[ 373.996606] R13: 0000000000000000 R14: 0000000000000000 R15: ffff937f2f55c0c0
[ 374.004570] FS: 0000000000000000(0000) GS:ffff937f2f540000(0000) knlGS:0000000000000000
[ 374.013602] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 374.020015] CR2: 0000000000000300 CR3: 0000000c14c10003 CR4: 00000000007706e0
[ 374.027981] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[ 374.035944] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[ 374.043907] PKRU: 55555554
[ 374.046927] Call Trace:
[ 374.049657] <IRQ>
[ 374.051902] ? blk_stat_free_callback_rcu+0x30/0x30
[ 374.057350] call_timer_fn+0x26/0xf0
[ 374.061344] __run_timers.part.0+0x1c0/0x220
[ 374.066112] ? __hrtimer_run_queues+0x136/0x270
[ 374.071169] ? recalibrate_cpu_khz+0x10/0x10
[ 374.075940] ? ktime_get+0x38/0x90
[ 374.079737] run_timer_softirq+0x26/0x50
[ 374.084106] __do_softirq+0xca/0x276
[ 374.088102] __irq_exit_rcu+0xc1/0xe0
[ 374.092195] sysvec_apic_timer_interrupt+0x72/0x90
[ 374.097545] </IRQ>
[ 374.099885] asm_sysvec_apic_timer_interrupt+0x12/0x20
[ 374.105613] RIP: 0010:acpi_idle_do_entry+0x4c/0x50
It can be reliably reproduced on xfs or ext4. It was tested on pmem ramdisks.
Thanks,
Murphy
