rcu: INFO: rcu_sched self-detected stall on CPU

rcu: INFO: rcu_sched self-detected stall on CPU

[Arthur Marsh]

Hi, still getting kswapd going to 100 percent CPU load on current Linus git head when loading Firefox with enough tabs to exhaust physical memory.

Tried bisecting but results were inconclusive, last reproducable good kernel was after:

https://github.com/torvalds/linux/commit/3644e2d2dda78e21edd8f5415b6d7ab03f5f54f3 mm/filemap: fix infinite loop in generic_file_buffered_read()

With current Linus git head 22nd December 2020 1:35 UTC:

202012221117-dmesg.log:[ 1765.483421] rcu: INFO: rcu_sched self-detected stall on CPU

202012221117-dmesglog:[ 1765.483425] rcu: 0-....: (5249 ticks this GP) idle=6fa/1/0x4000000000000000 softirq=43076/43076 fqs=2590 

202012221117-dmesg.log:[ 1765.483431] (t=5250 jiffies g=79277 q=13726)

202012221117-dmesg.log:[ 1765.483434] NMI backtrace for cpu 0

202012221117-dmesg.log:[ 1765.483436] CPU: 0 PID: 69 Comm: kswapd0 Tainted: G        W         5.10.0+ #4215

202012221117-dmesg.log:[ 1765.483438] Hardware name: System manufacturer System Product Name/M3A78 PRO, BIOS 1701    01/27/2011

202012221117-dmesg.log:[ 1765.483440] Call Trace:

202012221117-dmesg.log:[ 1765.483442]  <IRQ>

202012221117-dmesg.log:[ 1765.483443]  dump_stack+0x57/0x6a

202012221117-dmesg.log:[ 1765.483447]  nmi_cpu_backtrace.cold+0x32/0x69

202012221117-dmesg.log:[ 1765.483450]  ? lapic_can_unplug_cpu+0xa0/0xa0

202012221117-dmesg.log:[ 1765.483459]  nmi_trigger_cpumask_backtrace+0x80/0xa0

202012221117-dmesg.log:[ 1765.483463]  rcu_dump_cpu_stacks+0x9a/0xc8

202012221117-dmesg.log:[ 1765.483466]  rcu_sched_clock_irq.cold+0x202/0x3bb

202012221117-dmesg.log:[ 1765.483469]  update_process_times+0x8c/0xc0

202012221117-dmesg.log:[ 1765.483474]  tick_sched_handle+0x21/0x70

202012221117-dmesg.log:[ 1765.483479]  tick_sched_timer+0x94/0xb0

202012221117-dmesg.log:[ 1765.483482]  ? tick_nohz_handler+0xe0/0xe0

202012221117-dmesg.log:[ 1765.483485]  __hrtimer_run_queues+0x115/0x250

202012221117-dmesg.log:[ 1765.483488]  hrtimer_interrupt+0x10d/0x290

202012221117-dmesg.log:[ 1765.483492]  __sysvec_apic_timer_interrupt+0x5f/0xd0

202012221117-dmesg.log:[ 1765.483495]  asm_call_irq_on_stack+0x12/0x20

202012221117-dmesg.log:[ 1765.483500]  </IRQ>

202012221117-dmesg.log:[ 1765.483501]  sysvec_apic_timer_interrupt+0x69/0x80

202012221117-dmesg.log:[ 1765.483504]  asm_sysvec_apic_timer_interrupt+0x12/0x20

202012221117-dmesg.log:[ 1765.483507] RIP: 0010:ttm_pool_type_take+0xe/0xb0 [ttm]

202012221117-dmesg.log:[ 1765.483515] Code: 0f 1f 40 00 5b 31 c0 5d 41 5c 41 5d c3 66 66 2e 0f 1f 84 00 00 00 00 00 0f 1f 40 00 0f 1f 44 00 00 48 83 ec 20 4c 89 64 24 10 <4c> 8d 67 20 48 89 1c 24 48 89 6c 24 08 48 89 fb 4c 89 6c 24 18 4c

202012221117-dmesg.log:[ 1765.483517] RSP: 0018:ffffb12840197bf8 EFLAGS: 00000282

202012221117-dmesg.log:[ 1765.483520] RAX: 0000000000000000 RBX: ffffffffc0bf8e00 RCX: 0000000000000027

202012221117-dmesg.log:[ 1765.483522] RDX: 0000000000000001 RSI: ffffffffc0bf9080 RDI: ffffffffc0bf8df0

202012221117-dmesg.log:[ 1765.483524] RBP: 0000000000000080 R08: ffffffffc0bf8ba0 R09: ffffffffc0bf8ba0

202012221117-dmesg.log:[ 1765.483525] R10: ffffffff99ab6280 R11: ffffffff99a4b440 R12: 0000000000000000

202012221117-dmesg.log:[ 1765.483527] R13: 0000000000000000 R14: 0000000000000084 R15: ffffffffc0bf8b60

202012221117-dmesg.log:[ 1765.483531]  ? ttm_pool_type_take+0x2b/0xb0 [ttm]

202012221117-dmesg.log:[ 1765.483537]  ttm_pool_shrink+0x28/0xb0 [ttm]

202012221117-dmesg.log:[ 1765.483544]  ttm_pool_shrinker_scan+0xa/0x20 [ttm]

202012221117-dmesg.log:[ 1765.483550]  do_shrink_slab+0x144/0x240

202012221117-dmesglog:[ 1765.483554]  shrink_slab+0x9c/0x270

202012221117-dmesg.log:[ 1765.483557]  shrink_node+0x2d8/0x6e0

202012221117-dmesg.log:[ 1765.483560]  balance_pgdat+0x2d8/0x5d0

202012221117-dmesg.log:[ 1765.483563]  kswapd+0x1c5/0x340

202012221117-dmesg.log:[ 1765.483565]  ? finish_wait+0xb0/0xb0

202012221117-dmesg.log:[ 1765.483568]  ? balance_pgdat+0x5d0/0x5d0

202012221117-dmesg.log:[ 1765.483570]  kthread+0x126/0x150

202012221117-dmesg.log:[ 1765.483572]  ? kthread_associate_blkcg+0xc0/0xc0

202012221117-dmesg.log:[ 1765.483575]  ret_from_fork+0x22/0x30

End. 

AMD Athlon II X4 640, gcc 11, can supply config and build information if anyone wants it. 

Arthur. 



-- 
Sent from my Android device with K-9 Mail. Please excuse my brevity.