Re: BUG: sleeping function called from invalid context at kernel/locking/mutex.c:908

[Jeff Kirsher <jeffrey.t.kirsher@intel.com>]

[-- Attachment #1: Type: text/plain, Size: 4002 bytes --]

On Fri, 2019-01-18 at 18:31 +0900, Tetsuo Handa wrote:
> This is not a MM bug. ;-)
> And this change is made by below commit.
> 
> commit 59361316afcb08569af21e1af83e89c7051c055a
> Author: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
> Date:   Thu Aug 2 10:13:10 2018 -0700
> 
>     igb: reduce CPU0 latency when updating statistics
> 

This was already identified as a problem and I already sent a patch to
revert this commit to David Miller's net tree.

commit 5642e27bf6a935c1d0d9149e83d625dfba286463
Author: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Date:   Wed Jan 9 09:10:53 2019 -0800

    Revert "igb: reduce CPU0 latency when updating statistics"
    
    This reverts commit 59361316afcb08569af21e1af83e89c7051c055a.
    
    Due to problems found in additional testing, this causes an illegal
    context switch in the RCU read-side critical section.
    
    CC: Dave Jones <davej@codemonkey.org.uk>
    CC: Cong Wang <xiyou.wangcong@gmail.com>
    CC: Jan Jablonsky <jan.jablonsky@thalesgroup.com>
    Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
    Signed-off-by: David S. Miller <davem@davemloft.net>

> 
> On 2019/01/18 18:21, Mikhail Gavrilov wrote:
> > Hi folks.
> > 
> > Starting from 5.0 rc1 every time after launch atop I see subject
> > error message.
> > 
> > And this issue afle negatively impact on system performance.
> > 
> > Unclear on which subsystem it related but I hope here in mm list
> > somebody help clarify what is happened.
> > 
> > Thanks.
> > 
> > Full dmesg also attached.
> > 
> > [ 2164.694709] BUG: sleeping function called from invalid context
> > at
> > kernel/locking/mutex.c:908
> > [ 2164.694724] in_atomic(): 1, irqs_disabled(): 0, pid: 10498,
> > name: atop
> > [ 2164.694729] 2 locks held by atop/10498:
> > [ 2164.694732]  #0: 0000000061e564e5 (&p->lock){+.+.}, at:
> > seq_read+0x41/0x430
> > [ 2164.694753]  #1: 0000000081e34167 (rcu_read_lock){....}, at:
> > dev_seq_start+0x5/0x120
> > [ 2164.694762] CPU: 10 PID: 10498 Comm: atop Tainted: G         C
> >   5.0.0-0.rc2.git1.2.fc30.x86_64 #1
> > [ 2164.694765] Hardware name: System manufacturer System Product
> > Name/ROG STRIX X470-I GAMING, BIOS 1103 11/16/2018
> > [ 2164.694768] Call Trace:
> > [ 2164.694773]  dump_stack+0x85/0xc0
> > [ 2164.694778]  ___might_sleep.cold.73+0xac/0xbc
> > [ 2164.694783]  __mutex_lock+0x55/0x9a0
> > [ 2164.694788]  ? seq_vprintf+0x30/0x50
> > [ 2164.694792]  ? seq_printf+0x53/0x70
> > [ 2164.694806]  ? igb_get_stats64+0x29/0x80 [igb]
> > [ 2164.694813]  igb_get_stats64+0x29/0x80 [igb]
> > [ 2164.694818]  dev_get_stats+0x5b/0xc0
> > [ 2164.694822]  dev_seq_printf_stats+0x32/0xe0
> > [ 2164.694836]  dev_seq_show+0x10/0x30
> > [ 2164.694840]  seq_read+0x2fd/0x430
> > [ 2164.694847]  proc_reg_read+0x39/0x60
> > [ 2164.694852]  __vfs_read+0x36/0x1a0
> > [ 2164.694861]  vfs_read+0x9f/0x160
> > [ 2164.694866]  ksys_read+0x52/0xc0
> > [ 2164.694872]  do_syscall_64+0x60/0x1f0
> > [ 2164.694876]  entry_SYSCALL_64_after_hwframe+0x49/0xbe
> > [ 2164.694880] RIP: 0033:0x7fa0f51bd255
> > [ 2164.694883] Code: fe ff ff 50 48 8d 3d 2a 03 0a 00 e8 35 01 02
> > 00
> > 0f 1f 44 00 00 f3 0f 1e fa 48 8d 05 c5 94 0d 00 8b 00 85 c0 75 0f
> > 31
> > c0 0f 05 <48> 3d 00 f0 ff ff 77 53 c3 66 90 41 54 49 89 d4 55 48 89
> > f5
> > 53 89
> > [ 2164.694886] RSP: 002b:00007ffe79288db8 EFLAGS: 00000246
> > ORIG_RAX:
> > 0000000000000000
> > [ 2164.694890] RAX: ffffffffffffffda RBX: 0000000000a55260 RCX:
> > 00007fa0f51bd255
> > [ 2164.694892] RDX: 0000000000000400 RSI: 0000000000a55490 RDI:
> > 0000000000000004
> > [ 2164.694895] RBP: 0000000000000d68 R08: 0000000000000001 R09:
> > 0000000000000000
> > [ 2164.694898] R10: 00007fa0f50a3740 R11: 0000000000000246 R12:
> > 00007fa0f528d740
> > [ 2164.694900] R13: 00007fa0f528e340 R14: 00000000000007ff R15:
> > 0000000000a55260
> > 
> > 
> > 
> > --
> > Best Regards,
> > Mike Gavrilov.
> > 


[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 833 bytes --]