mm: system crashed
A message that is not in the archive (sent off-list or not yet received): [email protected]
Hi All,
I hit the below crashed when doing memory related tests[1] on s390x:
--------------- snip ---------------------
� 15929.351639¨ � <000000000021c0a6>¨ shrink_inactive_list+0x1c6/0x56c � 15929.351647¨ � <000000000021c69e>¨ shrink_lruvec+0x252/0x56c � 15929.351654¨ � <000000000021ca44>¨ shrink_zone+0x8c/0x1bc � 15929.351662¨ � <000000000021d080>¨ balance_pgdat+0x50c/0x658 � 15929.351671¨ � <000000000021d318>¨ kswapd+0x14c/0x470 � 15929.351680¨ � <0000000000158292>¨ kthread+0xda/0xe4 � 15929.351690¨ � <000000000062a5de>¨ kernel_thread_starter+0x6/0xc � 15929.351700¨ � <000000000062a5d8>¨ kernel_thread_starter+0x0/0xc
� 16109.346061¨ INFO: rcu_sched self-detected stall on CPU { 0} (t=24006 jiffies
g=89766 c=89765 q=10544)
� 16109.346101¨ CPU: 0 Tainted: G D 3.9.0-rc6+ #1
� 16109.346106¨ Process kswapd0 (pid: 28, task: 000000003b2a0000, ksp: 000000003b 2ab8c0)
� 16109.346110¨ 000000000001bb60 000000000001bb70 0000000000000002 0000000
000000000
000000000001bc00 000000000001bb78 000000000001bb78 00000000001009ca
0000000000000000 0000000000002930 000000000000000a 000000000000000a
000000000001bbc0 000000000001bb60 0000000000000000 0000000000000000
000000000063bb18 00000000001009ca 000000000001bb60 000000000001bbb0
� 16109.346170¨ Call Trace:
� 16109.346179¨ (� <0000000000100920>¨ show_trace+0x128/0x12c)
� 16109.346195¨ � <00000000001cd320>¨ rcu_check_callbacks+0x458/0xccc
� 16109.346209¨ � <0000000000140f2e>¨ update_process_times+0x4a/0x74
� 16109.346222¨ � <0000000000199452>¨ tick_sched_handle.isra.12+0x5e/0x70
� 16109.346235¨ � <00000000001995aa>¨ tick_sched_timer+0x6a/0x98
� 16109.346247¨ � <000000000015c1ea>¨ __run_hrtimer+0x8e/0x200
� 16109.346381¨ � <000000000015d1b2>¨ hrtimer_interrupt+0x212/0x2b0
� 16109.346385¨ � <00000000001040f6>¨ clock_comparator_work+0x4a/0x54
� 16109.346390¨ � <000000000010d658>¨ do_extint+0x158/0x15c
� 16109.346396¨ � <000000000062aa24>¨ ext_skip+0x38/0x3c
� 16109.346404¨ � <00000000001153c8>¨ smp_yield_cpu+0x44/0x48
� 16109.346412¨ (� <000003d10051aec0>¨ 0x3d10051aec0)
� 16109.346457¨ � <000000000024206a>¨ __page_check_address+0x16a/0x170
� 16109.346466¨ � <00000000002423a2>¨ page_referenced_one+0x3e/0xa0
� 16109.346501¨ � <000000000024427c>¨ page_referenced+0x32c/0x41c
� 16109.346510¨ � <000000000021b1dc>¨ shrink_page_list+0x380/0xb9c
� 16109.346521¨ � <000000000021c0a6>¨ shrink_inactive_list+0x1c6/0x56c
� 16109.346532¨ � <000000000021c69e>¨ shrink_lruvec+0x252/0x56c
� 16109.346542¨ � <000000000021ca44>¨ shrink_zone+0x8c/0x1bc
� 16109.346553¨ � <000000000021d080>¨ balance_pgdat+0x50c/0x658
� 16109.346564¨ � <000000000021d318>¨ kswapd+0x14c/0x470
� 16109.346576¨ � <0000000000158292>¨ kthread+0xda/0xe4
� 16109.346656¨ � <000000000062a5de>¨ kernel_thread_starter+0x6/0xc
� 16109.346682¨ � <000000000062a5d8>¨ kernel_thread_starter+0x0/0xc [-- MARK -- Fri Apr 12 06:15:00 2013]
� 16289.386061¨ INFO: rcu_sched self-detected stall on CPU { 0} (t=42010 jiffies
g=89766 c=89765 q=10627)
-------------- snip ----------------------
The testing system has 1Gb RAM, kernel is new latest mainline. please let me know if you need any more info.
[1] reproducer is come from LTP: https://github.com/linux-test-project/ltp/blob/master/testcases/kernel/mem/mtest06/mmap2.c
and execute it using this command: ./mmap2 -x 0.002 -a -p
Signature · 8 lines
Thanks, Zhouping -- To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to [email protected]. For more info on Linux MM, see: http://www.linux-mm.org/ . Don't email: <a href=mailto:"[email protected]"> [email protected] </a>
Quoted 49 lines · Zhouping Liu
Hi All,
I hit the below crashed when doing memory related tests[1] on s390x:
--------------- snip ---------------------
i? 1/2 15929.351639A? i? 1/2 <000000000021c0a6>A? shrink_inactive_list+0x1c6/0x56c i? 1/2 15929.351647A? i? 1/2 <000000000021c69e>A? shrink_lruvec+0x252/0x56c i? 1/2 15929.351654A? i? 1/2 <000000000021ca44>A? shrink_zone+0x8c/0x1bc i? 1/2 15929.351662A? i? 1/2 <000000000021d080>A? balance_pgdat+0x50c/0x658 i? 1/2 15929.351671A? i? 1/2 <000000000021d318>A? kswapd+0x14c/0x470 i? 1/2 15929.351680A? i? 1/2 <0000000000158292>A? kthread+0xda/0xe4 i? 1/2 15929.351690A? i? 1/2 <000000000062a5de>A? kernel_thread_starter+0x6/0xc i? 1/2 15929.351700A? i? 1/2 <000000000062a5d8>A? kernel_thread_starter+0x0/0xc i? 1/2 16109.346061A? INFO: rcu_sched self-detected stall on CPU { 0} (t=24006 jiffiesg=89766 c=89765 q=10544)
i? 1/2 16109.346101A? CPU: 0 Tainted: G D 3.9.0-rc6+ #1
i? 1/2 16109.346106A? Process kswapd0 (pid: 28, task:000000003b2a0000, ksp:000000003b2ab8c0)
i? 1/2 16109.346110A?000000000001bb60000000000001bb700000000000000002 0000000
000000000000000000001bc00 000000000001bb78 000000000001bb78 00000000001009ca 0000000000000000 0000000000002930 000000000000000a 000000000000000a 000000000001bbc0 000000000001bb60 0000000000000000 0000000000000000 000000000063bb18 00000000001009ca 000000000001bb60 000000000001bbb0 i? 1/2 16109.346170A? Call Trace: i? 1/2 16109.346179A? (i? 1/2 <0000000000100920>A? show_trace+0x128/0x12c) i? 1/2 16109.346195A? i? 1/2 <00000000001cd320>A? rcu_check_callbacks+0x458/0xccc i? 1/2 16109.346209A? i? 1/2 <0000000000140f2e>A? update_process_times+0x4a/0x74 i? 1/2 16109.346222A? i? 1/2 <0000000000199452>A? tick_sched_handle.isra.12+0x5e/0x70 i? 1/2 16109.346235A? i? 1/2 <00000000001995aa>A? tick_sched_timer+0x6a/0x98 i? 1/2 16109.346247A? i? 1/2 <000000000015c1ea>A? __run_hrtimer+0x8e/0x200 i? 1/2 16109.346381A? i? 1/2 <000000000015d1b2>A? hrtimer_interrupt+0x212/0x2b0 i? 1/2 16109.346385A? i? 1/2 <00000000001040f6>A? clock_comparator_work+0x4a/0x54 i? 1/2 16109.346390A? i? 1/2 <000000000010d658>A? do_extint+0x158/0x15c i? 1/2 16109.346396A? i? 1/2 <000000000062aa24>A? ext_skip+0x38/0x3c i? 1/2 16109.346404A? i? 1/2 <00000000001153c8>A? smp_yield_cpu+0x44/0x48 i? 1/2 16109.346412A? (i? 1/2 <000003d10051aec0>A? 0x3d10051aec0) i? 1/2 16109.346457A? i? 1/2 <000000000024206a>A? __page_check_address+0x16a/0x170 i? 1/2 16109.346466A? i? 1/2 <00000000002423a2>A? page_referenced_one+0x3e/0xa0 i? 1/2 16109.346501A? i? 1/2 <000000000024427c>A? page_referenced+0x32c/0x41c i? 1/2 16109.346510A? i? 1/2 <000000000021b1dc>A? shrink_page_list+0x380/0xb9c i? 1/2 16109.346521A? i? 1/2 <000000000021c0a6>A? shrink_inactive_list+0x1c6/0x56c i? 1/2 16109.346532A? i? 1/2 <000000000021c69e>A? shrink_lruvec+0x252/0x56c i? 1/2 16109.346542A? i? 1/2 <000000000021ca44>A? shrink_zone+0x8c/0x1bc i? 1/2 16109.346553A? i? 1/2 <000000000021d080>A? balance_pgdat+0x50c/0x658 i? 1/2 16109.346564A? i? 1/2 <000000000021d318>A? kswapd+0x14c/0x470 i? 1/2 16109.346576A? i? 1/2 <0000000000158292>A? kthread+0xda/0xe4 i? 1/2 16109.346656A? i? 1/2 <000000000062a5de>A? kernel_thread_starter+0x6/0xc i? 1/2 16109.346682A? i? 1/2 <000000000062a5d8>A? kernel_thread_starter+0x0/0xc
Quoted 3 lines
[-- MARK -- Fri Apr 12 06:15:00 2013]
i? 1/2 16289.386061A? INFO:rcu_schedself-detected stall on CPU { 0} (t=42010 jiffies
g=89766 c=89765 q=10627)
Did the system really crash or did you just see the rcu related warning(s)?
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to [email protected]. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"[email protected]"> [email protected] </a>
Quoted 53 lines · Heiko Carstens
Quoted 49 lines · Zhouping Liu
Hi All,
I hit the below crashed when doing memory related tests[1] on s390x:
--------------- snip ---------------------
i? 1/2 15929.351639A? i? 1/2 <000000000021c0a6>A? shrink_inactive_list+0x1c6/0x56c i? 1/2 15929.351647A? i? 1/2 <000000000021c69e>A? shrink_lruvec+0x252/0x56c i? 1/2 15929.351654A? i? 1/2 <000000000021ca44>A? shrink_zone+0x8c/0x1bc i? 1/2 15929.351662A? i? 1/2 <000000000021d080>A? balance_pgdat+0x50c/0x658 i? 1/2 15929.351671A? i? 1/2 <000000000021d318>A? kswapd+0x14c/0x470 i? 1/2 15929.351680A? i? 1/2 <0000000000158292>A? kthread+0xda/0xe4 i? 1/2 15929.351690A? i? 1/2 <000000000062a5de>A? kernel_thread_starter+0x6/0xc i? 1/2 15929.351700A? i? 1/2 <000000000062a5d8>A? kernel_thread_starter+0x0/0xc i? 1/2 16109.346061A? INFO: rcu_sched self-detected stall on CPU { 0} (t=24006 jiffiesg=89766 c=89765 q=10544)
i? 1/2 16109.346101A? CPU: 0 Tainted: G D 3.9.0-rc6+ #1
i? 1/2 16109.346106A? Process kswapd0 (pid: 28, task:000000003b2a0000, ksp:000000003b2ab8c0)
i? 1/2 16109.346110A?000000000001bb60000000000001bb700000000000000002 0000000
000000000000000000001bc00 000000000001bb78 000000000001bb78 00000000001009ca 0000000000000000 0000000000002930 000000000000000a 000000000000000a 000000000001bbc0 000000000001bb60 0000000000000000 0000000000000000 000000000063bb18 00000000001009ca 000000000001bb60 000000000001bbb0 i? 1/2 16109.346170A? Call Trace: i? 1/2 16109.346179A? (i? 1/2 <0000000000100920>A? show_trace+0x128/0x12c) i? 1/2 16109.346195A? i? 1/2 <00000000001cd320>A? rcu_check_callbacks+0x458/0xccc i? 1/2 16109.346209A? i? 1/2 <0000000000140f2e>A? update_process_times+0x4a/0x74 i? 1/2 16109.346222A? i? 1/2 <0000000000199452>A? tick_sched_handle.isra.12+0x5e/0x70 i? 1/2 16109.346235A? i? 1/2 <00000000001995aa>A? tick_sched_timer+0x6a/0x98 i? 1/2 16109.346247A? i? 1/2 <000000000015c1ea>A? __run_hrtimer+0x8e/0x200 i? 1/2 16109.346381A? i? 1/2 <000000000015d1b2>A? hrtimer_interrupt+0x212/0x2b0 i? 1/2 16109.346385A? i? 1/2 <00000000001040f6>A? clock_comparator_work+0x4a/0x54 i? 1/2 16109.346390A? i? 1/2 <000000000010d658>A? do_extint+0x158/0x15c i? 1/2 16109.346396A? i? 1/2 <000000000062aa24>A? ext_skip+0x38/0x3c i? 1/2 16109.346404A? i? 1/2 <00000000001153c8>A? smp_yield_cpu+0x44/0x48 i? 1/2 16109.346412A? (i? 1/2 <000003d10051aec0>A? 0x3d10051aec0) i? 1/2 16109.346457A? i? 1/2 <000000000024206a>A? __page_check_address+0x16a/0x170 i? 1/2 16109.346466A? i? 1/2 <00000000002423a2>A? page_referenced_one+0x3e/0xa0 i? 1/2 16109.346501A? i? 1/2 <000000000024427c>A? page_referenced+0x32c/0x41c i? 1/2 16109.346510A? i? 1/2 <000000000021b1dc>A? shrink_page_list+0x380/0xb9c i? 1/2 16109.346521A? i? 1/2 <000000000021c0a6>A? shrink_inactive_list+0x1c6/0x56c i? 1/2 16109.346532A? i? 1/2 <000000000021c69e>A? shrink_lruvec+0x252/0x56c i? 1/2 16109.346542A? i? 1/2 <000000000021ca44>A? shrink_zone+0x8c/0x1bc i? 1/2 16109.346553A? i? 1/2 <000000000021d080>A? balance_pgdat+0x50c/0x658 i? 1/2 16109.346564A? i? 1/2 <000000000021d318>A? kswapd+0x14c/0x470 i? 1/2 16109.346576A? i? 1/2 <0000000000158292>A? kthread+0xda/0xe4 i? 1/2 16109.346656A? i? 1/2 <000000000062a5de>A? kernel_thread_starter+0x6/0xc i? 1/2 16109.346682A? i? 1/2 <000000000062a5d8>A? kernel_thread_starter+0x0/0xcQuoted 3 lines
[-- MARK -- Fri Apr 12 06:15:00 2013]
i? 1/2 16289.386061A? INFO:rcu_schedself-detected stall on CPU { 0} (t=42010 jiffies
g=89766 c=89765 q=10627)
Quoted 1 line
Did the system really crash or did you just see the rcu related warning(s)?
I just check it again, actually at first the system didn't really crash, but the system is very slow in response. and the reproducer process can't be killed, after I did some common actions such as 'ls' 'vim' etc, the system seemed to be really crashed, no any response.
also in the previous testing, I can remember that the system would be no any response for a long time, just only repeatedly print out the such above 'Call Trace' into console.
Thanks,
Zhouping
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to [email protected]. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"[email protected]"> [email protected] </a>
Quoted 38 lines · Zhouping Liu
Quoted 29 lines · Heiko Carstens
Quoted 25 lines · Zhouping Liu
i? 1/2 16109.346170A? Call Trace: i? 1/2 16109.346179A? (i? 1/2 <0000000000100920>A? show_trace+0x128/0x12c) i? 1/2 16109.346195A? i? 1/2 <00000000001cd320>A? rcu_check_callbacks+0x458/0xccc i? 1/2 16109.346209A? i? 1/2 <0000000000140f2e>A? update_process_times+0x4a/0x74 i? 1/2 16109.346222A? i? 1/2 <0000000000199452>A? tick_sched_handle.isra.12+0x5e/0x70 i? 1/2 16109.346235A? i? 1/2 <00000000001995aa>A? tick_sched_timer+0x6a/0x98 i? 1/2 16109.346247A? i? 1/2 <000000000015c1ea>A? __run_hrtimer+0x8e/0x200 i? 1/2 16109.346381A? i? 1/2 <000000000015d1b2>A? hrtimer_interrupt+0x212/0x2b0 i? 1/2 16109.346385A? i? 1/2 <00000000001040f6>A? clock_comparator_work+0x4a/0x54 i? 1/2 16109.346390A? i? 1/2 <000000000010d658>A? do_extint+0x158/0x15c i? 1/2 16109.346396A? i? 1/2 <000000000062aa24>A? ext_skip+0x38/0x3c i? 1/2 16109.346404A? i? 1/2 <00000000001153c8>A? smp_yield_cpu+0x44/0x48 i? 1/2 16109.346412A? (i? 1/2 <000003d10051aec0>A? 0x3d10051aec0) i? 1/2 16109.346457A? i? 1/2 <000000000024206a>A? __page_check_address+0x16a/0x170 i? 1/2 16109.346466A? i? 1/2 <00000000002423a2>A? page_referenced_one+0x3e/0xa0 i? 1/2 16109.346501A? i? 1/2 <000000000024427c>A? page_referenced+0x32c/0x41c i? 1/2 16109.346510A? i? 1/2 <000000000021b1dc>A? shrink_page_list+0x380/0xb9c i? 1/2 16109.346521A? i? 1/2 <000000000021c0a6>A? shrink_inactive_list+0x1c6/0x56c i? 1/2 16109.346532A? i? 1/2 <000000000021c69e>A? shrink_lruvec+0x252/0x56c i? 1/2 16109.346542A? i? 1/2 <000000000021ca44>A? shrink_zone+0x8c/0x1bc i? 1/2 16109.346553A? i? 1/2 <000000000021d080>A? balance_pgdat+0x50c/0x658 i? 1/2 16109.346564A? i? 1/2 <000000000021d318>A? kswapd+0x14c/0x470 i? 1/2 16109.346576A? i? 1/2 <0000000000158292>A? kthread+0xda/0xe4 i? 1/2 16109.346656A? i? 1/2 <000000000062a5de>A? kernel_thread_starter+0x6/0xc i? 1/2 16109.346682A? i? 1/2 <000000000062a5d8>A? kernel_thread_starter+0x0/0xcQuoted 3 lines
[-- MARK -- Fri Apr 12 06:15:00 2013]
i? 1/2 16289.386061A? INFO:rcu_schedself-detected stall on CPU { 0} (t=42010 jiffies
g=89766 c=89765 q=10627)Quoted 1 line
Did the system really crash or did you just see the rcu related warning(s)?
I just check it again, actually at first the system didn't really crash, but the system is very slow in response.
and the reproducer process can't be killed, after I did some common actions such as 'ls' 'vim' etc, the system
seemed to be really crashed, no any response.
Quoted 3 lines
also in the previous testing, I can remember that the system would be no any response for a long time, just only repeatedly print out the such above 'Call Trace' into console.
Ok, thanks.
Just a couple of more questions: did you see this also on other archs, or just s390 (if you tried other platforms at all).
If you have some time, could you please repeat your test with the kernel command line option " user_mode=home "?
As far as I can tell there was only one s390 patch merged that was mmap related: 486c0a0bc80d370471b21662bf03f04fbb37cdc6 "s390/mm: Fix crst upgrade of mmap with MAP_FIXED".
Even though I don't think it explains the bug you've seen it might be worth to try to revert it.
And at last, can you share your kernel config?
Thanks,
Heiko
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to [email protected]. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"[email protected]"> [email protected] </a>
Hi Heiko,
Quoted 41 lines · Heiko Carstens
Quoted 37 lines · Zhouping Liu
Quoted 29 lines · Heiko Carstens
Quoted 25 lines · Zhouping Liu
i? 1/2 16109.346170A? Call Trace: i? 1/2 16109.346179A? (i? 1/2 <0000000000100920>A? show_trace+0x128/0x12c) i? 1/2 16109.346195A? i? 1/2 <00000000001cd320>A? rcu_check_callbacks+0x458/0xccc i? 1/2 16109.346209A? i? 1/2 <0000000000140f2e>A? update_process_times+0x4a/0x74 i? 1/2 16109.346222A? i? 1/2 <0000000000199452>A? tick_sched_handle.isra.12+0x5e/0x70 i? 1/2 16109.346235A? i? 1/2 <00000000001995aa>A? tick_sched_timer+0x6a/0x98 i? 1/2 16109.346247A? i? 1/2 <000000000015c1ea>A? __run_hrtimer+0x8e/0x200 i? 1/2 16109.346381A? i? 1/2 <000000000015d1b2>A? hrtimer_interrupt+0x212/0x2b0 i? 1/2 16109.346385A? i? 1/2 <00000000001040f6>A? clock_comparator_work+0x4a/0x54 i? 1/2 16109.346390A? i? 1/2 <000000000010d658>A? do_extint+0x158/0x15c i? 1/2 16109.346396A? i? 1/2 <000000000062aa24>A? ext_skip+0x38/0x3c i? 1/2 16109.346404A? i? 1/2 <00000000001153c8>A? smp_yield_cpu+0x44/0x48 i? 1/2 16109.346412A? (i? 1/2 <000003d10051aec0>A? 0x3d10051aec0) i? 1/2 16109.346457A? i? 1/2 <000000000024206a>A? __page_check_address+0x16a/0x170 i? 1/2 16109.346466A? i? 1/2 <00000000002423a2>A? page_referenced_one+0x3e/0xa0 i? 1/2 16109.346501A? i? 1/2 <000000000024427c>A? page_referenced+0x32c/0x41c i? 1/2 16109.346510A? i? 1/2 <000000000021b1dc>A? shrink_page_list+0x380/0xb9c i? 1/2 16109.346521A? i? 1/2 <000000000021c0a6>A? shrink_inactive_list+0x1c6/0x56c i? 1/2 16109.346532A? i? 1/2 <000000000021c69e>A? shrink_lruvec+0x252/0x56c i? 1/2 16109.346542A? i? 1/2 <000000000021ca44>A? shrink_zone+0x8c/0x1bc i? 1/2 16109.346553A? i? 1/2 <000000000021d080>A? balance_pgdat+0x50c/0x658 i? 1/2 16109.346564A? i? 1/2 <000000000021d318>A? kswapd+0x14c/0x470 i? 1/2 16109.346576A? i? 1/2 <0000000000158292>A? kthread+0xda/0xe4 i? 1/2 16109.346656A? i? 1/2 <000000000062a5de>A? kernel_thread_starter+0x6/0xc i? 1/2 16109.346682A? i? 1/2 <000000000062a5d8>A? kernel_thread_starter+0x0/0xcQuoted 3 lines
[-- MARK -- Fri Apr 12 06:15:00 2013]
i? 1/2 16289.386061A? INFO:rcu_schedself-detected stall on CPU { 0} (t=42010 jiffies
g=89766 c=89765 q=10627)Quoted 1 line
Did the system really crash or did you just see the rcu related warning(s)?
I just check it again, actually at first the system didn't really crash, but the system is very slow in response.
and the reproducer process can't be killed, after I did some common actions such as 'ls' 'vim' etc, the system
seemed to be really crashed, no any response.Quoted 3 lines
also in the previous testing, I can remember that the system would be no any response for a long time, just only repeatedly print out the such above 'Call Trace' into console.
Quoted 6 lines
Ok, thanks.
Just a couple of more questions: did you see this also on other archs, or just s390 (if you tried other platforms at all).If you have some time, could you please repeat your test with the kernel command line option " user_mode=home "?
What's the meaning of this command line? I can't find it in Documentation/kernel-parameters.txt/
Quoted 9 lines
As far as I can tell there was only one s390 patch merged that was mmap related:
486c0a0bc80d370471b21662bf03f04fbb37cdc6"s390/mm: Fix crst upgrade of mmap withMAP_FIXED".
Even though I don't think it explains the bug you've seen it might be worth to try to revert it.And at last, can you share your kernel config?
Quoted 8 lines
Thanks,
Heiko--
To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to [email protected]. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"[email protected]"> [email protected] </a>
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to [email protected]. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"[email protected]"> [email protected] </a>
Quoted 6 lines · Simon Jeons
Hi Heiko,
Quoted 2 lines
If you have some time, could you please repeat your test with the kernel command line option " user_mode=home "?
What's the meaning of this command line? I can't find it in
Documentation/kernel-parameters.txt/
It switches the architectural address space where kernel and user space reside in.
We only recently switched the default address space for user space from home space to primary space, since that's needed for kvm. The user space runs in home space mode will be removed in the future; we keep it currently as fallback, just in case something breaks.
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to [email protected]. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"[email protected]"> [email protected] </a>
Quoted 38 lines · Heiko Carstens
Quoted 37 lines · Zhouping Liu
Quoted 29 lines · Heiko Carstens
Quoted 25 lines · Zhouping Liu
i? 1/2 16109.346170A? Call Trace: i? 1/2 16109.346179A? (i? 1/2 <0000000000100920>A? show_trace+0x128/0x12c) i? 1/2 16109.346195A? i? 1/2 <00000000001cd320>A? rcu_check_callbacks+0x458/0xccc i? 1/2 16109.346209A? i? 1/2 <0000000000140f2e>A? update_process_times+0x4a/0x74 i? 1/2 16109.346222A? i? 1/2 <0000000000199452>A? tick_sched_handle.isra.12+0x5e/0x70 i? 1/2 16109.346235A? i? 1/2 <00000000001995aa>A? tick_sched_timer+0x6a/0x98 i? 1/2 16109.346247A? i? 1/2 <000000000015c1ea>A? __run_hrtimer+0x8e/0x200 i? 1/2 16109.346381A? i? 1/2 <000000000015d1b2>A? hrtimer_interrupt+0x212/0x2b0 i? 1/2 16109.346385A? i? 1/2 <00000000001040f6>A? clock_comparator_work+0x4a/0x54 i? 1/2 16109.346390A? i? 1/2 <000000000010d658>A? do_extint+0x158/0x15c i? 1/2 16109.346396A? i? 1/2 <000000000062aa24>A? ext_skip+0x38/0x3c i? 1/2 16109.346404A? i? 1/2 <00000000001153c8>A? smp_yield_cpu+0x44/0x48 i? 1/2 16109.346412A? (i? 1/2 <000003d10051aec0>A? 0x3d10051aec0) i? 1/2 16109.346457A? i? 1/2 <000000000024206a>A? __page_check_address+0x16a/0x170 i? 1/2 16109.346466A? i? 1/2 <00000000002423a2>A? page_referenced_one+0x3e/0xa0 i? 1/2 16109.346501A? i? 1/2 <000000000024427c>A? page_referenced+0x32c/0x41c i? 1/2 16109.346510A? i? 1/2 <000000000021b1dc>A? shrink_page_list+0x380/0xb9c i? 1/2 16109.346521A? i? 1/2 <000000000021c0a6>A? shrink_inactive_list+0x1c6/0x56c i? 1/2 16109.346532A? i? 1/2 <000000000021c69e>A? shrink_lruvec+0x252/0x56c i? 1/2 16109.346542A? i? 1/2 <000000000021ca44>A? shrink_zone+0x8c/0x1bc i? 1/2 16109.346553A? i? 1/2 <000000000021d080>A? balance_pgdat+0x50c/0x658 i? 1/2 16109.346564A? i? 1/2 <000000000021d318>A? kswapd+0x14c/0x470 i? 1/2 16109.346576A? i? 1/2 <0000000000158292>A? kthread+0xda/0xe4 i? 1/2 16109.346656A? i? 1/2 <000000000062a5de>A? kernel_thread_starter+0x6/0xc i? 1/2 16109.346682A? i? 1/2 <000000000062a5d8>A? kernel_thread_starter+0x0/0xcQuoted 3 lines
[-- MARK -- Fri Apr 12 06:15:00 2013]
i? 1/2 16289.386061A? INFO:rcu_schedself-detected stall on CPU { 0} (t=42010 jiffies
g=89766 c=89765 q=10627)Quoted 1 line
Did the system really crash or did you just see the rcu related warning(s)?
I just check it again, actually at first the system didn't really crash, but the system is very slow in response.
and the reproducer process can't be killed, after I did some common actions such as 'ls' 'vim' etc, the system
seemed to be really crashed, no any response.
Quoted 6 lines
Quoted 3 lines
also in the previous testing, I can remember that the system would be no any response for a long time, just only repeatedly print out the such above 'Call Trace' into console.
Ok, thanks.
Just a couple of more questions: did you see this also on other archs, or just s390 (if you tried other platforms at all).
Didn't meet the issue on other arches, just only on s390x, I will check further on other arches.
Quoted 3 lines
If you have some time, could you please repeat your test with the kernel command line option " user_mode=home "?
Sure, but the testing machine is not in my hand now, I will test it ASAP when I get it, also I'll try it on other s390x machine.
Quoted 6 lines
As far as I can tell there was only one s390 patch merged that was mmap related:
486c0a0bc80d370471b21662bf03f04fbb37cdc6"s390/mm: Fix crst upgrade of mmap withMAP_FIXED".
Even though I don't think it explains the bug you've seen it might be worth to try to revert it.
OK, I'll try it later.
Quoted 2 lines
And at last, can you share your kernel config?
Of course.
Thanks,
Zhouping
Quoted 4 lines
Thanks,
Heiko
#
# Automatically generated file; DO NOT EDIT.
# Linux/s390 3.9.0-rc6 Kernel Configuration
#CONFIG_MMU=yCONFIG_ZONE_DMA=yCONFIG_LOCKDEP_SUPPORT=y CONFIG_STACKTRACE_SUPPORT=y CONFIG_HAVE_LATENCYTOP_SUPPORT=y CONFIG_RWSEM_XCHGADD_ALGORITHM=y
# CONFIG_ARCH_HAS_ILOG2_U32 is not set
# CONFIG_ARCH_HAS_ILOG2_U64 is not set CONFIG_GENERIC_HWEIGHT=yCONFIG_GENERIC_BUG=y CONFIG_GENERIC_BUG_RELATIVE_POINTERS=y CONFIG_ARCH_DMA_ADDR_T_64BIT=y CONFIG_ARCH_SUPPORTS_DEBUG_PAGEALLOC=y CONFIG_KEXEC=yCONFIG_AUDIT_ARCH=yCONFIG_NO_IOPORT=y
# CONFIG_PCI_QUIRKS is not setCONFIG_S390=yCONFIG_SCHED_OMIT_FRAME_POINTER=y CONFIG_DEFCONFIG_LIST="/lib/modules/$UNAME_RELEASE/.config" CONFIG_IRQ_WORK=yCONFIG_BUILDTIME_EXTABLE_SORT=y
#
# General setup
#CONFIG_INIT_ENV_ARG_LIMIT=32CONFIG_CROSS_COMPILE=""CONFIG_LOCALVERSION=""
# CONFIG_LOCALVERSION_AUTO is not setCONFIG_HAVE_KERNEL_GZIP=yCONFIG_HAVE_KERNEL_BZIP2=yCONFIG_HAVE_KERNEL_LZMA=yCONFIG_HAVE_KERNEL_XZ=yCONFIG_HAVE_KERNEL_LZO=yCONFIG_KERNEL_GZIP=y
# CONFIG_KERNEL_BZIP2 is not set
# CONFIG_KERNEL_LZMA is not set
# CONFIG_KERNEL_XZ is not set
# CONFIG_KERNEL_LZO is not setCONFIG_DEFAULT_HOSTNAME="(none)"CONFIG_SWAP=yCONFIG_SYSVIPC=yCONFIG_SYSVIPC_SYSCTL=yCONFIG_POSIX_MQUEUE=yCONFIG_POSIX_MQUEUE_SYSCTL=yCONFIG_FHANDLE=yCONFIG_AUDIT=yCONFIG_AUDITSYSCALL=yCONFIG_AUDIT_WATCH=yCONFIG_AUDIT_TREE=yCONFIG_AUDIT_LOGINUID_IMMUTABLE=y
# CONFIG_ALWAYS_USE_PERSISTENT_CLOCK is not setCONFIG_GENERIC_TIME_VSYSCALL_OLD=yCONFIG_GENERIC_CLOCKEVENTS=yCONFIG_GENERIC_CLOCKEVENTS_BUILD=y
#
# Timers subsystem
#CONFIG_TICK_ONESHOT=yCONFIG_NO_HZ=yCONFIG_HIGH_RES_TIMERS=y
#
# CPU/Task time and stats accounting
#CONFIG_VIRT_CPU_ACCOUNTING=yCONFIG_VIRT_CPU_ACCOUNTING_NATIVE=yCONFIG_BSD_PROCESS_ACCT=yCONFIG_BSD_PROCESS_ACCT_V3=yCONFIG_TASKSTATS=yCONFIG_TASK_DELAY_ACCT=yCONFIG_TASK_XACCT=yCONFIG_TASK_IO_ACCOUNTING=y
#
# RCU Subsystem
#CONFIG_TREE_RCU=y
# CONFIG_PREEMPT_RCU is not setCONFIG_RCU_STALL_COMMON=yCONFIG_RCU_FANOUT=64CONFIG_RCU_FANOUT_LEAF=16
# CONFIG_RCU_FANOUT_EXACT is not set
# CONFIG_RCU_FAST_NO_HZ is not set
# CONFIG_TREE_RCU_TRACE is not set
# CONFIG_RCU_NOCB_CPU is not set
# CONFIG_IKCONFIG is not setCONFIG_LOG_BUF_SHIFT=17CONFIG_CGROUPS=y
# CONFIG_CGROUP_DEBUG is not setCONFIG_CGROUP_FREEZER=yCONFIG_CGROUP_DEVICE=yCONFIG_CPUSETS=yCONFIG_PROC_PID_CPUSET=yCONFIG_CGROUP_CPUACCT=yCONFIG_RESOURCE_COUNTERS=yCONFIG_MEMCG=yCONFIG_MEMCG_SWAP=yCONFIG_MEMCG_SWAP_ENABLED=yCONFIG_MEMCG_KMEM=y
# CONFIG_CGROUP_HUGETLB is not setCONFIG_CGROUP_PERF=yCONFIG_CGROUP_SCHED=yCONFIG_FAIR_GROUP_SCHED=yCONFIG_CFS_BANDWIDTH=yCONFIG_RT_GROUP_SCHED=yCONFIG_BLK_CGROUP=y
# CONFIG_DEBUG_BLK_CGROUP is not set
# CONFIG_CHECKPOINT_RESTORE is not setCONFIG_NAMESPACES=yCONFIG_UTS_NS=yCONFIG_IPC_NS=yCONFIG_PID_NS=yCONFIG_NET_NS=yCONFIG_SCHED_AUTOGROUP=yCONFIG_MM_OWNER=y
# CONFIG_SYSFS_DEPRECATED is not setCONFIG_RELAY=yCONFIG_BLK_DEV_INITRD=yCONFIG_INITRAMFS_SOURCE=""CONFIG_RD_GZIP=yCONFIG_RD_BZIP2=yCONFIG_RD_LZMA=yCONFIG_RD_XZ=yCONFIG_RD_LZO=y
# CONFIG_CC_OPTIMIZE_FOR_SIZE is not setCONFIG_SYSCTL=yCONFIG_ANON_INODES=y
# CONFIG_EXPERT is not set
# CONFIG_SYSCTL_SYSCALL is not setCONFIG_SYSCTL_EXCEPTION_TRACE=yCONFIG_KALLSYMS=yCONFIG_KALLSYMS_ALL=yCONFIG_HOTPLUG=yCONFIG_PRINTK=yCONFIG_BUG=yCONFIG_ELF_CORE=yCONFIG_BASE_FULL=yCONFIG_FUTEX=yCONFIG_EPOLL=yCONFIG_SIGNALFD=yCONFIG_TIMERFD=yCONFIG_EVENTFD=yCONFIG_SHMEM=yCONFIG_AIO=y
# CONFIG_EMBEDDED is not setCONFIG_HAVE_PERF_EVENTS=y
#
# Kernel Performance Events And Counters
#CONFIG_PERF_EVENTS=y
# CONFIG_DEBUG_PERF_USE_VMALLOC is not setCONFIG_VM_EVENT_COUNTERS=yCONFIG_SLUB_DEBUG=y
# CONFIG_COMPAT_BRK is not set
# CONFIG_SLAB is not setCONFIG_SLUB=yCONFIG_PROFILING=yCONFIG_TRACEPOINTS=yCONFIG_OPROFILE=mCONFIG_HAVE_OPROFILE=yCONFIG_KPROBES=yCONFIG_JUMP_LABEL=yCONFIG_HAVE_64BIT_ALIGNED_ACCESS=yCONFIG_HAVE_SYSCALL_WRAPPERS=yCONFIG_KRETPROBES=yCONFIG_HAVE_KPROBES=yCONFIG_HAVE_KRETPROBES=yCONFIG_HAVE_ARCH_TRACEHOOK=y
# CONFIG_HAVE_DMA_ATTRS is not setCONFIG_USE_GENERIC_SMP_HELPERS=yCONFIG_GENERIC_SMP_IDLE_THREAD=yCONFIG_HAVE_REGS_AND_STACK_ACCESS_API=yCONFIG_HAVE_ARCH_JUMP_LABEL=yCONFIG_HAVE_ARCH_MUTEX_CPU_RELAX=yCONFIG_ARCH_HAVE_NMI_SAFE_CMPXCHG=yCONFIG_HAVE_ALIGNED_STRUCT_PAGE=yCONFIG_HAVE_CMPXCHG_LOCAL=yCONFIG_HAVE_CMPXCHG_DOUBLE=yCONFIG_ARCH_WANT_IPC_PARSE_VERSION=yCONFIG_ARCH_WANT_COMPAT_IPC_PARSE_VERSION=yCONFIG_ARCH_WANT_OLD_COMPAT_IPC=yCONFIG_HAVE_ARCH_SECCOMP_FILTER=yCONFIG_SECCOMP_FILTER=yCONFIG_HAVE_VIRT_CPU_ACCOUNTING=yCONFIG_HAVE_ARCH_TRANSPARENT_HUGEPAGE=yCONFIG_HAVE_MOD_ARCH_SPECIFIC=yCONFIG_MODULES_USE_ELF_RELA=yCONFIG_CLONE_BACKWARDS2=yCONFIG_OLD_SIGSUSPEND3=yCONFIG_OLD_SIGACTION=yCONFIG_COMPAT_OLD_SIGACTION=y
#
# GCOV-based kernel profiling
#
# CONFIG_GCOV_KERNEL is not set
# CONFIG_HAVE_GENERIC_DMA_COHERENT is not setCONFIG_SLABINFO=yCONFIG_RT_MUTEXES=yCONFIG_BASE_SMALL=0CONFIG_MODULES=yCONFIG_MODULE_FORCE_LOAD=yCONFIG_MODULE_UNLOAD=y
# CONFIG_MODULE_FORCE_UNLOAD is not setCONFIG_MODVERSIONS=yCONFIG_MODULE_SRCVERSION_ALL=yCONFIG_MODULE_SIG=y
# CONFIG_MODULE_SIG_FORCE is not setCONFIG_MODULE_SIG_ALL=yCONFIG_MODULE_SIG_SHA1=y
# CONFIG_MODULE_SIG_SHA224 is not set
# CONFIG_MODULE_SIG_SHA256 is not set
# CONFIG_MODULE_SIG_SHA384 is not set
# CONFIG_MODULE_SIG_SHA512 is not setCONFIG_MODULE_SIG_HASH="sha1"CONFIG_INIT_ALL_POSSIBLE=yCONFIG_STOP_MACHINE=yCONFIG_BLOCK=yCONFIG_BLK_DEV_BSG=yCONFIG_BLK_DEV_BSGLIB=yCONFIG_BLK_DEV_INTEGRITY=yCONFIG_BLK_DEV_THROTTLING=y
#
# Partition Types
#CONFIG_PARTITION_ADVANCED=y
# CONFIG_ACORN_PARTITION is not set
# CONFIG_OSF_PARTITION is not setCONFIG_AMIGA_PARTITION=y
# CONFIG_ATARI_PARTITION is not setCONFIG_IBM_PARTITION=y
# CONFIG_MAC_PARTITION is not setCONFIG_MSDOS_PARTITION=yCONFIG_BSD_DISKLABEL=yCONFIG_MINIX_SUBPARTITION=yCONFIG_SOLARIS_X86_PARTITION=yCONFIG_UNIXWARE_DISKLABEL=y
# CONFIG_LDM_PARTITION is not set
# CONFIG_SGI_PARTITION is not set
# CONFIG_ULTRIX_PARTITION is not set
# CONFIG_SUN_PARTITION is not setCONFIG_KARMA_PARTITION=yCONFIG_EFI_PARTITION=y
# CONFIG_SYSV68_PARTITION is not setCONFIG_BLOCK_COMPAT=y
#
# IO Schedulers
#CONFIG_IOSCHED_NOOP=yCONFIG_IOSCHED_DEADLINE=yCONFIG_IOSCHED_CFQ=yCONFIG_CFQ_GROUP_IOSCHED=yCONFIG_DEFAULT_DEADLINE=y
# CONFIG_DEFAULT_CFQ is not set
# CONFIG_DEFAULT_NOOP is not setCONFIG_DEFAULT_IOSCHED="deadline"CONFIG_PADATA=yCONFIG_ASN1=yCONFIG_ARCH_INLINE_SPIN_TRYLOCK=yCONFIG_ARCH_INLINE_SPIN_TRYLOCK_BH=yCONFIG_ARCH_INLINE_SPIN_LOCK=yCONFIG_ARCH_INLINE_SPIN_LOCK_BH=yCONFIG_ARCH_INLINE_SPIN_LOCK_IRQ=yCONFIG_ARCH_INLINE_SPIN_LOCK_IRQSAVE=yCONFIG_ARCH_INLINE_SPIN_UNLOCK=yCONFIG_ARCH_INLINE_SPIN_UNLOCK_BH=yCONFIG_ARCH_INLINE_SPIN_UNLOCK_IRQ=yCONFIG_ARCH_INLINE_SPIN_UNLOCK_IRQRESTORE=yCONFIG_ARCH_INLINE_READ_TRYLOCK=yCONFIG_ARCH_INLINE_READ_LOCK=yCONFIG_ARCH_INLINE_READ_LOCK_BH=yCONFIG_ARCH_INLINE_READ_LOCK_IRQ=yCONFIG_ARCH_INLINE_READ_LOCK_IRQSAVE=yCONFIG_ARCH_INLINE_READ_UNLOCK=yCONFIG_ARCH_INLINE_READ_UNLOCK_BH=yCONFIG_ARCH_INLINE_READ_UNLOCK_IRQ=yCONFIG_ARCH_INLINE_READ_UNLOCK_IRQRESTORE=yCONFIG_ARCH_INLINE_WRITE_TRYLOCK=yCONFIG_ARCH_INLINE_WRITE_LOCK=yCONFIG_ARCH_INLINE_WRITE_LOCK_BH=yCONFIG_ARCH_INLINE_WRITE_LOCK_IRQ=yCONFIG_ARCH_INLINE_WRITE_LOCK_IRQSAVE=yCONFIG_ARCH_INLINE_WRITE_UNLOCK=yCONFIG_ARCH_INLINE_WRITE_UNLOCK_BH=yCONFIG_ARCH_INLINE_WRITE_UNLOCK_IRQ=yCONFIG_ARCH_INLINE_WRITE_UNLOCK_IRQRESTORE=yCONFIG_INLINE_SPIN_TRYLOCK=yCONFIG_INLINE_SPIN_TRYLOCK_BH=yCONFIG_INLINE_SPIN_LOCK=yCONFIG_INLINE_SPIN_LOCK_BH=yCONFIG_INLINE_SPIN_LOCK_IRQ=yCONFIG_INLINE_SPIN_LOCK_IRQSAVE=yCONFIG_INLINE_SPIN_UNLOCK_BH=yCONFIG_INLINE_SPIN_UNLOCK_IRQ=yCONFIG_INLINE_SPIN_UNLOCK_IRQRESTORE=yCONFIG_INLINE_READ_TRYLOCK=yCONFIG_INLINE_READ_LOCK=yCONFIG_INLINE_READ_LOCK_BH=yCONFIG_INLINE_READ_LOCK_IRQ=yCONFIG_INLINE_READ_LOCK_IRQSAVE=yCONFIG_INLINE_READ_UNLOCK=yCONFIG_INLINE_READ_UNLOCK_BH=yCONFIG_INLINE_READ_UNLOCK_IRQ=yCONFIG_INLINE_READ_UNLOCK_IRQRESTORE=yCONFIG_INLINE_WRITE_TRYLOCK=yCONFIG_INLINE_WRITE_LOCK=yCONFIG_INLINE_WRITE_LOCK_BH=yCONFIG_INLINE_WRITE_LOCK_IRQ=yCONFIG_INLINE_WRITE_LOCK_IRQSAVE=yCONFIG_INLINE_WRITE_UNLOCK=yCONFIG_INLINE_WRITE_UNLOCK_BH=yCONFIG_INLINE_WRITE_UNLOCK_IRQ=yCONFIG_INLINE_WRITE_UNLOCK_IRQRESTORE=yCONFIG_MUTEX_SPIN_ON_OWNER=yCONFIG_FREEZER=y
#
# Processor type and features
#CONFIG_HAVE_MARCH_Z900_FEATURES=yCONFIG_HAVE_MARCH_Z990_FEATURES=yCONFIG_HAVE_MARCH_Z9_109_FEATURES=y
# CONFIG_HAVE_MARCH_Z10_FEATURES is not set
# CONFIG_HAVE_MARCH_Z196_FEATURES is not set
# CONFIG_HAVE_MARCH_ZEC12_FEATURES is not set
# CONFIG_MARCH_Z900 is not set
# CONFIG_MARCH_Z990 is not setCONFIG_MARCH_Z9_109=y
# CONFIG_MARCH_Z10 is not set
# CONFIG_MARCH_Z196 is not set
# CONFIG_MARCH_ZEC12 is not setCONFIG_64BIT=yCONFIG_COMPAT=yCONFIG_SYSVIPC_COMPAT=yCONFIG_KEYS_COMPAT=yCONFIG_SMP=yCONFIG_NR_CPUS=64CONFIG_HOTPLUG_CPU=yCONFIG_SCHED_MC=yCONFIG_SCHED_BOOK=yCONFIG_PREEMPT_NONE=y
# CONFIG_PREEMPT_VOLUNTARY is not set
# CONFIG_PREEMPT is not setCONFIG_HZ_100=y
# CONFIG_HZ_250 is not set
# CONFIG_HZ_300 is not set
# CONFIG_HZ_1000 is not setCONFIG_HZ=100CONFIG_SCHED_HRTICK=y
#
# Memory setup
#CONFIG_ARCH_SPARSEMEM_ENABLE=yCONFIG_ARCH_SPARSEMEM_DEFAULT=yCONFIG_ARCH_SELECT_MEMORY_MODEL=yCONFIG_ARCH_ENABLE_MEMORY_HOTPLUG=yCONFIG_ARCH_ENABLE_MEMORY_HOTREMOVE=yCONFIG_FORCE_MAX_ZONEORDER=9CONFIG_SELECT_MEMORY_MODEL=yCONFIG_SPARSEMEM_MANUAL=yCONFIG_SPARSEMEM=yCONFIG_HAVE_MEMORY_PRESENT=yCONFIG_SPARSEMEM_EXTREME=yCONFIG_SPARSEMEM_VMEMMAP_ENABLE=yCONFIG_SPARSEMEM_VMEMMAP=yCONFIG_HAVE_MEMBLOCK=yCONFIG_HAVE_MEMBLOCK_NODE_MAP=yCONFIG_ARCH_DISCARD_MEMBLOCK=yCONFIG_MEMORY_ISOLATION=y
# CONFIG_HAVE_BOOTMEM_INFO_NODE is not setCONFIG_MEMORY_HOTPLUG=yCONFIG_MEMORY_HOTPLUG_SPARSE=yCONFIG_MEMORY_HOTREMOVE=yCONFIG_PAGEFLAGS_EXTENDED=yCONFIG_SPLIT_PTLOCK_CPUS=4CONFIG_BALLOON_COMPACTION=yCONFIG_COMPACTION=yCONFIG_MIGRATION=yCONFIG_PHYS_ADDR_T_64BIT=yCONFIG_ZONE_DMA_FLAG=1CONFIG_BOUNCE=yCONFIG_VIRT_TO_BUS=yCONFIG_KSM=yCONFIG_DEFAULT_MMAP_MIN_ADDR=4096
# CONFIG_TRANSPARENT_HUGEPAGE is not setCONFIG_CROSS_MEMORY_ATTACH=yCONFIG_CLEANCACHE=yCONFIG_FRONTSWAP=yCONFIG_PACK_STACK=y
# CONFIG_SMALL_STACK is not setCONFIG_CHECK_STACK=yCONFIG_STACK_GUARD=256
# CONFIG_WARN_DYNAMIC_STACK is not set
#
# I/O subsystem
#CONFIG_QDIO=m
# CONFIG_PCI is not set
# CONFIG_PCI_DOMAINS is not set
# CONFIG_HAS_IOMEM is not set
# CONFIG_IOMMU_HELPER is not set
# CONFIG_HAS_DMA is not set
# CONFIG_NEED_SG_DMA_LENGTH is not set
# CONFIG_NEED_DMA_MAP_STATE is not setCONFIG_CHSC_SCH=mCONFIG_SCM_BUS=yCONFIG_EADM_SCH=m
#
# Dump support
#CONFIG_CRASH_DUMP=y
# CONFIG_ZFCPDUMP is not set
#
# Executable file formats / Emulations
#CONFIG_BINFMT_ELF=yCONFIG_COMPAT_BINFMT_ELF=yCONFIG_CORE_DUMP_DEFAULT_ELF_HEADERS=y
# CONFIG_HAVE_AOUT is not setCONFIG_BINFMT_MISC=mCONFIG_COREDUMP=yCONFIG_SECCOMP=y
#
# Power Management
#CONFIG_ARCH_HIBERNATION_POSSIBLE=yCONFIG_HIBERNATE_CALLBACKS=yCONFIG_HIBERNATION=yCONFIG_ARCH_SAVE_PAGE_KEYS=yCONFIG_PM_STD_PARTITION="/dev/jokes"CONFIG_PM_SLEEP=yCONFIG_PM_SLEEP_SMP=y
# CONFIG_PM_AUTOSLEEP is not set
# CONFIG_PM_WAKELOCKS is not setCONFIG_PM_RUNTIME=yCONFIG_PM=y
# CONFIG_PM_DEBUG is not setCONFIG_NET=y
#
# Networking options
#CONFIG_PACKET=y
# CONFIG_PACKET_DIAG is not setCONFIG_UNIX=yCONFIG_UNIX_DIAG=mCONFIG_XFRM=yCONFIG_XFRM_ALGO=yCONFIG_XFRM_USER=yCONFIG_XFRM_SUB_POLICY=yCONFIG_XFRM_MIGRATE=yCONFIG_XFRM_STATISTICS=yCONFIG_XFRM_IPCOMP=mCONFIG_NET_KEY=mCONFIG_NET_KEY_MIGRATE=yCONFIG_IUCV=yCONFIG_AFIUCV=mCONFIG_INET=yCONFIG_IP_MULTICAST=yCONFIG_IP_ADVANCED_ROUTER=yCONFIG_IP_FIB_TRIE_STATS=yCONFIG_IP_MULTIPLE_TABLES=yCONFIG_IP_ROUTE_MULTIPATH=yCONFIG_IP_ROUTE_VERBOSE=yCONFIG_IP_ROUTE_CLASSID=y
# CONFIG_IP_PNP is not setCONFIG_NET_IPIP=mCONFIG_NET_IPGRE_DEMUX=mCONFIG_NET_IPGRE=mCONFIG_NET_IPGRE_BROADCAST=yCONFIG_IP_MROUTE=yCONFIG_IP_MROUTE_MULTIPLE_TABLES=yCONFIG_IP_PIMSM_V1=yCONFIG_IP_PIMSM_V2=y
# CONFIG_ARPD is not setCONFIG_SYN_COOKIES=yCONFIG_NET_IPVTI=mCONFIG_INET_AH=mCONFIG_INET_ESP=mCONFIG_INET_IPCOMP=mCONFIG_INET_XFRM_TUNNEL=mCONFIG_INET_TUNNEL=mCONFIG_INET_XFRM_MODE_TRANSPORT=mCONFIG_INET_XFRM_MODE_TUNNEL=mCONFIG_INET_XFRM_MODE_BEET=mCONFIG_INET_LRO=yCONFIG_INET_DIAG=mCONFIG_INET_TCP_DIAG=mCONFIG_INET_UDP_DIAG=mCONFIG_TCP_CONG_ADVANCED=yCONFIG_TCP_CONG_BIC=mCONFIG_TCP_CONG_CUBIC=yCONFIG_TCP_CONG_WESTWOOD=mCONFIG_TCP_CONG_HTCP=mCONFIG_TCP_CONG_HSTCP=mCONFIG_TCP_CONG_HYBLA=mCONFIG_TCP_CONG_VEGAS=mCONFIG_TCP_CONG_SCALABLE=mCONFIG_TCP_CONG_LP=mCONFIG_TCP_CONG_VENO=mCONFIG_TCP_CONG_YEAH=mCONFIG_TCP_CONG_ILLINOIS=mCONFIG_DEFAULT_CUBIC=y
# CONFIG_DEFAULT_RENO is not setCONFIG_DEFAULT_TCP_CONG="cubic"CONFIG_TCP_MD5SIG=yCONFIG_IPV6=yCONFIG_IPV6_PRIVACY=yCONFIG_IPV6_ROUTER_PREF=yCONFIG_IPV6_ROUTE_INFO=yCONFIG_IPV6_OPTIMISTIC_DAD=yCONFIG_INET6_AH=mCONFIG_INET6_ESP=mCONFIG_INET6_IPCOMP=mCONFIG_IPV6_MIP6=mCONFIG_INET6_XFRM_TUNNEL=mCONFIG_INET6_TUNNEL=mCONFIG_INET6_XFRM_MODE_TRANSPORT=mCONFIG_INET6_XFRM_MODE_TUNNEL=mCONFIG_INET6_XFRM_MODE_BEET=mCONFIG_INET6_XFRM_MODE_ROUTEOPTIMIZATION=mCONFIG_IPV6_SIT=mCONFIG_IPV6_SIT_6RD=yCONFIG_IPV6_NDISC_NODETYPE=yCONFIG_IPV6_TUNNEL=m
# CONFIG_IPV6_GRE is not setCONFIG_IPV6_MULTIPLE_TABLES=y
# CONFIG_IPV6_SUBTREES is not setCONFIG_IPV6_MROUTE=yCONFIG_IPV6_MROUTE_MULTIPLE_TABLES=yCONFIG_IPV6_PIMSM_V2=yCONFIG_NETLABEL=yCONFIG_NETWORK_SECMARK=yCONFIG_NETWORK_PHY_TIMESTAMPING=yCONFIG_NETFILTER=y
# CONFIG_NETFILTER_DEBUG is not setCONFIG_NETFILTER_ADVANCED=yCONFIG_BRIDGE_NETFILTER=y
#
# Core Netfilter Configuration
#CONFIG_NETFILTER_NETLINK=mCONFIG_NETFILTER_NETLINK_ACCT=mCONFIG_NETFILTER_NETLINK_QUEUE=mCONFIG_NETFILTER_NETLINK_LOG=mCONFIG_NF_CONNTRACK=mCONFIG_NF_CONNTRACK_MARK=yCONFIG_NF_CONNTRACK_SECMARK=yCONFIG_NF_CONNTRACK_ZONES=yCONFIG_NF_CONNTRACK_PROCFS=yCONFIG_NF_CONNTRACK_EVENTS=y
# CONFIG_NF_CONNTRACK_TIMEOUT is not setCONFIG_NF_CONNTRACK_TIMESTAMP=yCONFIG_NF_CT_PROTO_DCCP=mCONFIG_NF_CT_PROTO_GRE=mCONFIG_NF_CT_PROTO_SCTP=mCONFIG_NF_CT_PROTO_UDPLITE=mCONFIG_NF_CONNTRACK_AMANDA=mCONFIG_NF_CONNTRACK_FTP=mCONFIG_NF_CONNTRACK_H323=mCONFIG_NF_CONNTRACK_IRC=mCONFIG_NF_CONNTRACK_BROADCAST=mCONFIG_NF_CONNTRACK_NETBIOS_NS=mCONFIG_NF_CONNTRACK_SNMP=mCONFIG_NF_CONNTRACK_PPTP=mCONFIG_NF_CONNTRACK_SANE=mCONFIG_NF_CONNTRACK_SIP=mCONFIG_NF_CONNTRACK_TFTP=mCONFIG_NF_CT_NETLINK=m
# CONFIG_NF_CT_NETLINK_TIMEOUT is not setCONFIG_NF_CT_NETLINK_HELPER=mCONFIG_NETFILTER_NETLINK_QUEUE_CT=yCONFIG_NF_NAT=mCONFIG_NF_NAT_NEEDED=yCONFIG_NF_NAT_PROTO_DCCP=mCONFIG_NF_NAT_PROTO_UDPLITE=mCONFIG_NF_NAT_PROTO_SCTP=mCONFIG_NF_NAT_AMANDA=mCONFIG_NF_NAT_FTP=mCONFIG_NF_NAT_IRC=mCONFIG_NF_NAT_SIP=mCONFIG_NF_NAT_TFTP=mCONFIG_NETFILTER_TPROXY=mCONFIG_NETFILTER_XTABLES=y
#
# Xtables combined modules
#CONFIG_NETFILTER_XT_MARK=mCONFIG_NETFILTER_XT_CONNMARK=mCONFIG_NETFILTER_XT_SET=m
#
# Xtables targets
#CONFIG_NETFILTER_XT_TARGET_AUDIT=mCONFIG_NETFILTER_XT_TARGET_CHECKSUM=mCONFIG_NETFILTER_XT_TARGET_CLASSIFY=mCONFIG_NETFILTER_XT_TARGET_CONNMARK=mCONFIG_NETFILTER_XT_TARGET_CONNSECMARK=mCONFIG_NETFILTER_XT_TARGET_CT=mCONFIG_NETFILTER_XT_TARGET_DSCP=mCONFIG_NETFILTER_XT_TARGET_HL=mCONFIG_NETFILTER_XT_TARGET_HMARK=mCONFIG_NETFILTER_XT_TARGET_IDLETIMER=mCONFIG_NETFILTER_XT_TARGET_LOG=mCONFIG_NETFILTER_XT_TARGET_MARK=mCONFIG_NETFILTER_XT_TARGET_NETMAP=mCONFIG_NETFILTER_XT_TARGET_NFLOG=mCONFIG_NETFILTER_XT_TARGET_NFQUEUE=mCONFIG_NETFILTER_XT_TARGET_NOTRACK=mCONFIG_NETFILTER_XT_TARGET_RATEEST=mCONFIG_NETFILTER_XT_TARGET_REDIRECT=mCONFIG_NETFILTER_XT_TARGET_TEE=mCONFIG_NETFILTER_XT_TARGET_TPROXY=mCONFIG_NETFILTER_XT_TARGET_TRACE=mCONFIG_NETFILTER_XT_TARGET_SECMARK=mCONFIG_NETFILTER_XT_TARGET_TCPMSS=mCONFIG_NETFILTER_XT_TARGET_TCPOPTSTRIP=m
#
# Xtables matches
#CONFIG_NETFILTER_XT_MATCH_ADDRTYPE=m
# CONFIG_NETFILTER_XT_MATCH_BPF is not setCONFIG_NETFILTER_XT_MATCH_CLUSTER=mCONFIG_NETFILTER_XT_MATCH_COMMENT=mCONFIG_NETFILTER_XT_MATCH_CONNBYTES=m
# CONFIG_NETFILTER_XT_MATCH_CONNLABEL is not setCONFIG_NETFILTER_XT_MATCH_CONNLIMIT=mCONFIG_NETFILTER_XT_MATCH_CONNMARK=mCONFIG_NETFILTER_XT_MATCH_CONNTRACK=mCONFIG_NETFILTER_XT_MATCH_CPU=mCONFIG_NETFILTER_XT_MATCH_DCCP=mCONFIG_NETFILTER_XT_MATCH_DEVGROUP=mCONFIG_NETFILTER_XT_MATCH_DSCP=mCONFIG_NETFILTER_XT_MATCH_ECN=mCONFIG_NETFILTER_XT_MATCH_ESP=mCONFIG_NETFILTER_XT_MATCH_HASHLIMIT=mCONFIG_NETFILTER_XT_MATCH_HELPER=mCONFIG_NETFILTER_XT_MATCH_HL=mCONFIG_NETFILTER_XT_MATCH_IPRANGE=mCONFIG_NETFILTER_XT_MATCH_IPVS=mCONFIG_NETFILTER_XT_MATCH_LENGTH=mCONFIG_NETFILTER_XT_MATCH_LIMIT=mCONFIG_NETFILTER_XT_MATCH_MAC=mCONFIG_NETFILTER_XT_MATCH_MARK=mCONFIG_NETFILTER_XT_MATCH_MULTIPORT=mCONFIG_NETFILTER_XT_MATCH_NFACCT=mCONFIG_NETFILTER_XT_MATCH_OSF=mCONFIG_NETFILTER_XT_MATCH_OWNER=mCONFIG_NETFILTER_XT_MATCH_POLICY=mCONFIG_NETFILTER_XT_MATCH_PHYSDEV=mCONFIG_NETFILTER_XT_MATCH_PKTTYPE=mCONFIG_NETFILTER_XT_MATCH_QUOTA=mCONFIG_NETFILTER_XT_MATCH_RATEEST=mCONFIG_NETFILTER_XT_MATCH_REALM=mCONFIG_NETFILTER_XT_MATCH_RECENT=mCONFIG_NETFILTER_XT_MATCH_SCTP=mCONFIG_NETFILTER_XT_MATCH_SOCKET=mCONFIG_NETFILTER_XT_MATCH_STATE=mCONFIG_NETFILTER_XT_MATCH_STATISTIC=mCONFIG_NETFILTER_XT_MATCH_STRING=mCONFIG_NETFILTER_XT_MATCH_TCPMSS=mCONFIG_NETFILTER_XT_MATCH_TIME=mCONFIG_NETFILTER_XT_MATCH_U32=mCONFIG_IP_SET=mCONFIG_IP_SET_MAX=256CONFIG_IP_SET_BITMAP_IP=mCONFIG_IP_SET_BITMAP_IPMAC=mCONFIG_IP_SET_BITMAP_PORT=mCONFIG_IP_SET_HASH_IP=mCONFIG_IP_SET_HASH_IPPORT=mCONFIG_IP_SET_HASH_IPPORTIP=mCONFIG_IP_SET_HASH_IPPORTNET=mCONFIG_IP_SET_HASH_NET=mCONFIG_IP_SET_HASH_NETPORT=mCONFIG_IP_SET_HASH_NETIFACE=mCONFIG_IP_SET_LIST_SET=mCONFIG_IP_VS=mCONFIG_IP_VS_IPV6=y
# CONFIG_IP_VS_DEBUG is not setCONFIG_IP_VS_TAB_BITS=12
#
# IPVS transport protocol load balancing support
#CONFIG_IP_VS_PROTO_TCP=yCONFIG_IP_VS_PROTO_UDP=yCONFIG_IP_VS_PROTO_AH_ESP=yCONFIG_IP_VS_PROTO_ESP=yCONFIG_IP_VS_PROTO_AH=yCONFIG_IP_VS_PROTO_SCTP=y
#
# IPVS scheduler
#CONFIG_IP_VS_RR=mCONFIG_IP_VS_WRR=mCONFIG_IP_VS_LC=mCONFIG_IP_VS_WLC=mCONFIG_IP_VS_LBLC=mCONFIG_IP_VS_LBLCR=mCONFIG_IP_VS_DH=mCONFIG_IP_VS_SH=mCONFIG_IP_VS_SED=mCONFIG_IP_VS_NQ=m
#
# IPVS SH scheduler
#CONFIG_IP_VS_SH_TAB_BITS=8
#
# IPVS application helper
#CONFIG_IP_VS_FTP=mCONFIG_IP_VS_NFCT=yCONFIG_IP_VS_PE_SIP=m
#
# IP: Netfilter Configuration
#CONFIG_NF_DEFRAG_IPV4=mCONFIG_NF_CONNTRACK_IPV4=m
# CONFIG_NF_CONNTRACK_PROC_COMPAT is not setCONFIG_IP_NF_IPTABLES=mCONFIG_IP_NF_MATCH_AH=mCONFIG_IP_NF_MATCH_ECN=mCONFIG_IP_NF_MATCH_RPFILTER=mCONFIG_IP_NF_MATCH_TTL=mCONFIG_IP_NF_FILTER=mCONFIG_IP_NF_TARGET_REJECT=mCONFIG_IP_NF_TARGET_ULOG=mCONFIG_NF_NAT_IPV4=mCONFIG_IP_NF_TARGET_MASQUERADE=mCONFIG_IP_NF_TARGET_NETMAP=mCONFIG_IP_NF_TARGET_REDIRECT=mCONFIG_NF_NAT_SNMP_BASIC=mCONFIG_NF_NAT_PROTO_GRE=mCONFIG_NF_NAT_PPTP=mCONFIG_NF_NAT_H323=mCONFIG_IP_NF_MANGLE=mCONFIG_IP_NF_TARGET_CLUSTERIP=mCONFIG_IP_NF_TARGET_ECN=mCONFIG_IP_NF_TARGET_TTL=mCONFIG_IP_NF_RAW=mCONFIG_IP_NF_SECURITY=mCONFIG_IP_NF_ARPTABLES=mCONFIG_IP_NF_ARPFILTER=mCONFIG_IP_NF_ARP_MANGLE=m
#
# IPv6: Netfilter Configuration
#CONFIG_NF_DEFRAG_IPV6=mCONFIG_NF_CONNTRACK_IPV6=mCONFIG_IP6_NF_IPTABLES=mCONFIG_IP6_NF_MATCH_AH=mCONFIG_IP6_NF_MATCH_EUI64=mCONFIG_IP6_NF_MATCH_FRAG=mCONFIG_IP6_NF_MATCH_OPTS=mCONFIG_IP6_NF_MATCH_HL=mCONFIG_IP6_NF_MATCH_IPV6HEADER=mCONFIG_IP6_NF_MATCH_MH=mCONFIG_IP6_NF_MATCH_RPFILTER=mCONFIG_IP6_NF_MATCH_RT=mCONFIG_IP6_NF_TARGET_HL=mCONFIG_IP6_NF_FILTER=mCONFIG_IP6_NF_TARGET_REJECT=mCONFIG_IP6_NF_MANGLE=mCONFIG_IP6_NF_RAW=mCONFIG_IP6_NF_SECURITY=mCONFIG_NF_NAT_IPV6=m
# CONFIG_IP6_NF_TARGET_MASQUERADE is not set
# CONFIG_IP6_NF_TARGET_NPT is not setCONFIG_BRIDGE_NF_EBTABLES=mCONFIG_BRIDGE_EBT_BROUTE=mCONFIG_BRIDGE_EBT_T_FILTER=mCONFIG_BRIDGE_EBT_T_NAT=mCONFIG_BRIDGE_EBT_802_3=mCONFIG_BRIDGE_EBT_AMONG=mCONFIG_BRIDGE_EBT_ARP=mCONFIG_BRIDGE_EBT_IP=mCONFIG_BRIDGE_EBT_IP6=mCONFIG_BRIDGE_EBT_LIMIT=mCONFIG_BRIDGE_EBT_MARK=mCONFIG_BRIDGE_EBT_PKTTYPE=mCONFIG_BRIDGE_EBT_STP=mCONFIG_BRIDGE_EBT_VLAN=mCONFIG_BRIDGE_EBT_ARPREPLY=mCONFIG_BRIDGE_EBT_DNAT=mCONFIG_BRIDGE_EBT_MARK_T=mCONFIG_BRIDGE_EBT_REDIRECT=mCONFIG_BRIDGE_EBT_SNAT=mCONFIG_BRIDGE_EBT_LOG=mCONFIG_BRIDGE_EBT_ULOG=mCONFIG_BRIDGE_EBT_NFLOG=mCONFIG_IP_DCCP=mCONFIG_INET_DCCP_DIAG=m
#
# DCCP CCIDs Configuration
#
# CONFIG_IP_DCCP_CCID2_DEBUG is not setCONFIG_IP_DCCP_CCID3=y
# CONFIG_IP_DCCP_CCID3_DEBUG is not setCONFIG_IP_DCCP_TFRC_LIB=y
#
# DCCP Kernel Hacking
#
# CONFIG_IP_DCCP_DEBUG is not set
# CONFIG_NET_DCCPPROBE is not setCONFIG_IP_SCTP=mCONFIG_NET_SCTPPROBE=m
# CONFIG_SCTP_DBG_MSG is not set
# CONFIG_SCTP_DBG_OBJCNT is not set
# CONFIG_SCTP_DEFAULT_COOKIE_HMAC_MD5 is not setCONFIG_SCTP_DEFAULT_COOKIE_HMAC_SHA1=y
# CONFIG_SCTP_DEFAULT_COOKIE_HMAC_NONE is not setCONFIG_SCTP_COOKIE_HMAC_MD5=yCONFIG_SCTP_COOKIE_HMAC_SHA1=yCONFIG_RDS=mCONFIG_RDS_TCP=m
# CONFIG_RDS_DEBUG is not set
# CONFIG_TIPC is not setCONFIG_ATM=mCONFIG_ATM_CLIP=m
# CONFIG_ATM_CLIP_NO_ICMP is not setCONFIG_ATM_LANE=m
# CONFIG_ATM_MPOA is not setCONFIG_ATM_BR2684=m
# CONFIG_ATM_BR2684_IPFILTER is not setCONFIG_L2TP=mCONFIG_L2TP_DEBUGFS=mCONFIG_L2TP_V3=yCONFIG_L2TP_IP=mCONFIG_L2TP_ETH=mCONFIG_STP=mCONFIG_GARP=mCONFIG_BRIDGE=mCONFIG_BRIDGE_IGMP_SNOOPING=y
# CONFIG_BRIDGE_VLAN_FILTERING is not setCONFIG_VLAN_8021Q=mCONFIG_VLAN_8021Q_GVRP=y
# CONFIG_VLAN_8021Q_MVRP is not set
# CONFIG_DECNET is not setCONFIG_LLC=m
# CONFIG_LLC2 is not set
# CONFIG_IPX is not set
# CONFIG_ATALK is not set
# CONFIG_X25 is not set
# CONFIG_LAPB is not set
# CONFIG_PHONET is not setCONFIG_IEEE802154=mCONFIG_IEEE802154_6LOWPAN=mCONFIG_MAC802154=mCONFIG_NET_SCHED=y
#
# Queueing/Scheduling
#CONFIG_NET_SCH_CBQ=mCONFIG_NET_SCH_HTB=mCONFIG_NET_SCH_HFSC=mCONFIG_NET_SCH_ATM=mCONFIG_NET_SCH_PRIO=mCONFIG_NET_SCH_MULTIQ=mCONFIG_NET_SCH_RED=mCONFIG_NET_SCH_SFB=mCONFIG_NET_SCH_SFQ=mCONFIG_NET_SCH_TEQL=mCONFIG_NET_SCH_TBF=mCONFIG_NET_SCH_GRED=mCONFIG_NET_SCH_DSMARK=mCONFIG_NET_SCH_NETEM=mCONFIG_NET_SCH_DRR=mCONFIG_NET_SCH_MQPRIO=mCONFIG_NET_SCH_CHOKE=mCONFIG_NET_SCH_QFQ=mCONFIG_NET_SCH_CODEL=mCONFIG_NET_SCH_FQ_CODEL=mCONFIG_NET_SCH_INGRESS=mCONFIG_NET_SCH_PLUG=m
#
# Classification
#CONFIG_NET_CLS=yCONFIG_NET_CLS_BASIC=mCONFIG_NET_CLS_TCINDEX=mCONFIG_NET_CLS_ROUTE4=mCONFIG_NET_CLS_FW=mCONFIG_NET_CLS_U32=mCONFIG_CLS_U32_PERF=yCONFIG_CLS_U32_MARK=yCONFIG_NET_CLS_RSVP=mCONFIG_NET_CLS_RSVP6=mCONFIG_NET_CLS_FLOW=mCONFIG_NET_CLS_CGROUP=yCONFIG_NET_EMATCH=yCONFIG_NET_EMATCH_STACK=32CONFIG_NET_EMATCH_CMP=mCONFIG_NET_EMATCH_NBYTE=mCONFIG_NET_EMATCH_U32=mCONFIG_NET_EMATCH_META=mCONFIG_NET_EMATCH_TEXT=mCONFIG_NET_EMATCH_IPSET=mCONFIG_NET_CLS_ACT=yCONFIG_NET_ACT_POLICE=mCONFIG_NET_ACT_GACT=mCONFIG_GACT_PROB=yCONFIG_NET_ACT_MIRRED=mCONFIG_NET_ACT_IPT=mCONFIG_NET_ACT_NAT=mCONFIG_NET_ACT_PEDIT=mCONFIG_NET_ACT_SIMP=mCONFIG_NET_ACT_SKBEDIT=mCONFIG_NET_ACT_CSUM=mCONFIG_NET_CLS_IND=yCONFIG_NET_SCH_FIFO=yCONFIG_DCB=yCONFIG_DNS_RESOLVER=mCONFIG_BATMAN_ADV=mCONFIG_BATMAN_ADV_BLA=yCONFIG_BATMAN_ADV_DAT=y
# CONFIG_BATMAN_ADV_DEBUG is not setCONFIG_OPENVSWITCH=m
# CONFIG_VSOCKETS is not setCONFIG_RPS=yCONFIG_XPS=yCONFIG_NETPRIO_CGROUP=mCONFIG_BQL=yCONFIG_BPF_JIT=y
#
# Network testing
#CONFIG_NET_PKTGEN=m
# CONFIG_NET_TCPPROBE is not setCONFIG_NET_DROP_MONITOR=y
# CONFIG_CAN is not set
# CONFIG_AF_RXRPC is not setCONFIG_FIB_RULES=yCONFIG_WIMAX=mCONFIG_WIMAX_DEBUG_LEVEL=8CONFIG_RFKILL=mCONFIG_RFKILL_INPUT=y
# CONFIG_NET_9P is not set
# CONFIG_CAIF is not set
# CONFIG_CEPH_LIB is not setCONFIG_NFC=mCONFIG_NFC_NCI=mCONFIG_NFC_HCI=mCONFIG_NFC_SHDLC=yCONFIG_NFC_LLCP=y
#
# Near Field Communication (NFC) devices
#
# CONFIG_NFC_PN544 is not set
# CONFIG_NFC_MICROREAD is not setCONFIG_HAVE_BPF_JIT=y
# CONFIG_PCMCIA is not setCONFIG_CCW=y
#
# Device Drivers
#
#
# Generic Driver Options
#CONFIG_UEVENT_HELPER_PATH=""CONFIG_DEVTMPFS=yCONFIG_DEVTMPFS_MOUNT=yCONFIG_STANDALONE=yCONFIG_PREVENT_FIRMWARE_BUILD=yCONFIG_FW_LOADER=y
# CONFIG_FIRMWARE_IN_KERNEL is not setCONFIG_EXTRA_FIRMWARE=""CONFIG_FW_LOADER_USER_HELPER=y
# CONFIG_DEBUG_DRIVER is not set
# CONFIG_DEBUG_DEVRES is not setCONFIG_SYS_HYPERVISOR=y
# CONFIG_GENERIC_CPU_DEVICES is not set
# CONFIG_DMA_SHARED_BUFFER is not set
#
# Bus devices
#CONFIG_CONNECTOR=yCONFIG_PROC_EVENTS=yCONFIG_BLK_DEV=y
# CONFIG_BLK_DEV_COW_COMMON is not setCONFIG_BLK_DEV_LOOP=yCONFIG_BLK_DEV_LOOP_MIN_COUNT=8CONFIG_BLK_DEV_CRYPTOLOOP=m
# CONFIG_BLK_DEV_DRBD is not set
# CONFIG_BLK_DEV_NBD is not setCONFIG_BLK_DEV_OSD=mCONFIG_BLK_DEV_RAM=mCONFIG_BLK_DEV_RAM_COUNT=16CONFIG_BLK_DEV_RAM_SIZE=16384
# CONFIG_BLK_DEV_XIP is not set
# CONFIG_CDROM_PKTCDVD is not set
# CONFIG_ATA_OVER_ETH is not set
#
# S/390 block device drivers
#CONFIG_BLK_DEV_XPRAM=mCONFIG_DCSSBLK=mCONFIG_DASD=mCONFIG_DASD_PROFILE=yCONFIG_DASD_ECKD=mCONFIG_DASD_FBA=mCONFIG_DASD_DIAG=mCONFIG_DASD_EER=yCONFIG_SCM_BLOCK=mCONFIG_SCM_BLOCK_CLUSTER_WRITE=yCONFIG_VIRTIO_BLK=m
# CONFIG_BLK_DEV_RBD is not set
#
# Misc devices
#
# CONFIG_SENSORS_LIS3LV02D is not setCONFIG_ENCLOSURE_SERVICES=m
# CONFIG_C2PORT is not set
#
# EEPROM support
#
# CONFIG_EEPROM_93CX6 is not set
#
# Texas Instruments shared transport line discipline #
#
# Altera FPGA firmware download module
#
#
# SCSI device support
#CONFIG_SCSI_MOD=yCONFIG_RAID_ATTRS=mCONFIG_SCSI=y
# CONFIG_SCSI_DMA is not setCONFIG_SCSI_TGT=mCONFIG_SCSI_NETLINK=yCONFIG_SCSI_PROC_FS=y
#
# SCSI support type (disk, tape, CD-ROM)
#CONFIG_BLK_DEV_SD=mCONFIG_CHR_DEV_ST=mCONFIG_CHR_DEV_OSST=mCONFIG_BLK_DEV_SR=mCONFIG_BLK_DEV_SR_VENDOR=yCONFIG_CHR_DEV_SG=mCONFIG_CHR_DEV_SCH=mCONFIG_SCSI_ENCLOSURE=mCONFIG_SCSI_MULTI_LUN=yCONFIG_SCSI_CONSTANTS=yCONFIG_SCSI_LOGGING=yCONFIG_SCSI_SCAN_ASYNC=y
#
# SCSI Transports
#CONFIG_SCSI_SPI_ATTRS=mCONFIG_SCSI_FC_ATTRS=mCONFIG_SCSI_FC_TGT_ATTRS=yCONFIG_SCSI_ISCSI_ATTRS=mCONFIG_SCSI_SAS_ATTRS=mCONFIG_SCSI_SAS_LIBSAS=mCONFIG_SCSI_SAS_HOST_SMP=yCONFIG_SCSI_SRP_ATTRS=mCONFIG_SCSI_SRP_TGT_ATTRS=yCONFIG_SCSI_LOWLEVEL=yCONFIG_ISCSI_TCP=mCONFIG_ISCSI_BOOT_SYSFS=m
# CONFIG_SCSI_UFSHCD is not set
# CONFIG_LIBFC is not set
# CONFIG_LIBFCOE is not setCONFIG_SCSI_DEBUG=mCONFIG_ZFCP=mCONFIG_SCSI_VIRTIO=mCONFIG_SCSI_DH=yCONFIG_SCSI_DH_RDAC=yCONFIG_SCSI_DH_HP_SW=yCONFIG_SCSI_DH_EMC=yCONFIG_SCSI_DH_ALUA=yCONFIG_SCSI_OSD_INITIATOR=mCONFIG_SCSI_OSD_ULD=mCONFIG_SCSI_OSD_DPRINT_SENSE=1
# CONFIG_SCSI_OSD_DEBUG is not setCONFIG_MD=yCONFIG_BLK_DEV_MD=yCONFIG_MD_AUTODETECT=yCONFIG_MD_LINEAR=mCONFIG_MD_RAID0=mCONFIG_MD_RAID1=mCONFIG_MD_RAID10=mCONFIG_MD_RAID456=m
# CONFIG_MD_MULTIPATH is not setCONFIG_MD_FAULTY=mCONFIG_BLK_DEV_DM=mCONFIG_DM_DEBUG=yCONFIG_DM_BUFIO=mCONFIG_DM_BIO_PRISON=mCONFIG_DM_PERSISTENT_DATA=mCONFIG_DM_CRYPT=mCONFIG_DM_SNAPSHOT=mCONFIG_DM_THIN_PROVISIONING=m
# CONFIG_DM_DEBUG_BLOCK_STACK_TRACING is not set
# CONFIG_DM_CACHE is not setCONFIG_DM_MIRROR=mCONFIG_DM_RAID=mCONFIG_DM_LOG_USERSPACE=mCONFIG_DM_ZERO=mCONFIG_DM_MULTIPATH=mCONFIG_DM_MULTIPATH_QL=mCONFIG_DM_MULTIPATH_ST=mCONFIG_DM_DELAY=mCONFIG_DM_UEVENT=yCONFIG_DM_FLAKEY=mCONFIG_DM_VERITY=mCONFIG_TARGET_CORE=mCONFIG_TCM_IBLOCK=mCONFIG_TCM_FILEIO=mCONFIG_TCM_PSCSI=mCONFIG_LOOPBACK_TARGET=mCONFIG_ISCSI_TARGET=mCONFIG_NETDEVICES=yCONFIG_NET_CORE=yCONFIG_BONDING=mCONFIG_DUMMY=m
# CONFIG_EQUALIZER is not set
# CONFIG_MII is not setCONFIG_IFB=mCONFIG_NET_TEAM=mCONFIG_NET_TEAM_MODE_BROADCAST=mCONFIG_NET_TEAM_MODE_ROUNDROBIN=mCONFIG_NET_TEAM_MODE_ACTIVEBACKUP=mCONFIG_NET_TEAM_MODE_LOADBALANCE=mCONFIG_MACVLAN=mCONFIG_MACVTAP=mCONFIG_VXLAN=mCONFIG_NETCONSOLE=mCONFIG_NETCONSOLE_DYNAMIC=yCONFIG_NETPOLL=yCONFIG_NETPOLL_TRAP=yCONFIG_NET_POLL_CONTROLLER=yCONFIG_TUN=mCONFIG_VETH=mCONFIG_VIRTIO_NET=mCONFIG_ATM_DRIVERS=y
# CONFIG_ATM_DUMMY is not setCONFIG_ATM_TCP=m
#
# CAIF transport drivers
#CONFIG_ETHERNET=yCONFIG_NET_VENDOR_INTEL=y
# CONFIG_NET_VENDOR_I825XX is not set
# CONFIG_NET_VENDOR_MARVELL is not setCONFIG_NET_VENDOR_NATSEMI=yCONFIG_NET_VENDOR_8390=y
# CONFIG_PHYLIB is not set
# CONFIG_PPP is not set
# CONFIG_SLIP is not set
#
# S/390 network device drivers
#CONFIG_LCS=mCONFIG_CTCM=m
# CONFIG_NETIUCV is not setCONFIG_SMSGIUCV=mCONFIG_SMSGIUCV_EVENT=m
# CONFIG_CLAW is not setCONFIG_QETH=mCONFIG_QETH_L2=mCONFIG_QETH_L3=mCONFIG_QETH_IPV6=yCONFIG_CCWGROUP=m
#
# WiMAX Wireless Broadband devices
#
#
# Enable USB support to see WiMAX USB drivers
#CONFIG_WAN=yCONFIG_HDLC=mCONFIG_HDLC_RAW=m
# CONFIG_HDLC_RAW_ETH is not setCONFIG_HDLC_CISCO=mCONFIG_HDLC_FR=mCONFIG_HDLC_PPP=m
#
# X.25/LAPB support is disabled
#CONFIG_DLCI=mCONFIG_DLCI_MAX=8CONFIG_IEEE802154_DRIVERS=mCONFIG_IEEE802154_FAKEHARD=mCONFIG_IEEE802154_FAKELB=m
#
# Input device support
#CONFIG_INPUT=y
# CONFIG_INPUT_FF_MEMLESS is not set
# CONFIG_INPUT_POLLDEV is not set
# CONFIG_INPUT_SPARSEKMAP is not set
# CONFIG_INPUT_MATRIXKMAP is not set
#
# Userland interfaces
#CONFIG_INPUT_MOUSEDEV=yCONFIG_INPUT_MOUSEDEV_PSAUX=yCONFIG_INPUT_MOUSEDEV_SCREEN_X=1024CONFIG_INPUT_MOUSEDEV_SCREEN_Y=768
# CONFIG_INPUT_JOYDEV is not set
# CONFIG_INPUT_EVDEV is not set
# CONFIG_INPUT_EVBUG is not set
#
# Input Device Drivers
#CONFIG_INPUT_KEYBOARD=yCONFIG_KEYBOARD_ATKBD=y
# CONFIG_KEYBOARD_LKKBD is not set
# CONFIG_KEYBOARD_NEWTON is not set
# CONFIG_KEYBOARD_OPENCORES is not set
# CONFIG_KEYBOARD_STOWAWAY is not set
# CONFIG_KEYBOARD_SUNKBD is not set
# CONFIG_KEYBOARD_XTKBD is not setCONFIG_INPUT_MOUSE=yCONFIG_MOUSE_PS2=yCONFIG_MOUSE_PS2_ALPS=yCONFIG_MOUSE_PS2_LOGIPS2PP=yCONFIG_MOUSE_PS2_SYNAPTICS=yCONFIG_MOUSE_PS2_CYPRESS=yCONFIG_MOUSE_PS2_TRACKPOINT=y
# CONFIG_MOUSE_PS2_ELANTECH is not set
# CONFIG_MOUSE_PS2_SENTELIC is not set
# CONFIG_MOUSE_PS2_TOUCHKIT is not set
# CONFIG_MOUSE_SERIAL is not set
# CONFIG_MOUSE_VSXXXAA is not set
# CONFIG_INPUT_JOYSTICK is not set
# CONFIG_INPUT_TABLET is not set
# CONFIG_INPUT_TOUCHSCREEN is not set
# CONFIG_INPUT_MISC is not set
#
# Hardware I/O ports
#CONFIG_SERIO=yCONFIG_SERIO_SERPORT=yCONFIG_SERIO_LIBPS2=y
# CONFIG_SERIO_RAW is not set
# CONFIG_SERIO_ALTERA_PS2 is not set
# CONFIG_SERIO_PS2MULT is not set
# CONFIG_GAMEPORT is not set
#
# Character devices
#CONFIG_TTY=yCONFIG_UNIX98_PTYS=yCONFIG_DEVPTS_MULTIPLE_INSTANCES=y
# CONFIG_LEGACY_PTYS is not setCONFIG_N_GSM=m
# CONFIG_TRACE_SINK is not set
# CONFIG_DEVKMEM is not setCONFIG_HVC_DRIVER=yCONFIG_HVC_IUCV=yCONFIG_VIRTIO_CONSOLE=yCONFIG_HW_RANDOM=yCONFIG_HW_RANDOM_VIRTIO=m
# CONFIG_R3964 is not setCONFIG_RAW_DRIVER=yCONFIG_MAX_RAW_DEVS=8192CONFIG_HANGCHECK_TIMER=m
#
# S/390 character device drivers
#CONFIG_TN3270=yCONFIG_TN3270_TTY=yCONFIG_TN3270_FS=mCONFIG_TN3270_CONSOLE=yCONFIG_TN3215=yCONFIG_TN3215_CONSOLE=yCONFIG_CCW_CONSOLE=yCONFIG_SCLP_TTY=yCONFIG_SCLP_CONSOLE=yCONFIG_SCLP_VT220_TTY=yCONFIG_SCLP_VT220_CONSOLE=yCONFIG_SCLP_CPI=mCONFIG_SCLP_ASYNC=mCONFIG_S390_TAPE=m
#
# S/390 tape hardware support
#CONFIG_S390_TAPE_34XX=mCONFIG_S390_TAPE_3590=mCONFIG_VMLOGRDR=mCONFIG_VMCP=yCONFIG_MONREADER=mCONFIG_MONWRITER=mCONFIG_S390_VMUR=m
# CONFIG_I2C is not set
# CONFIG_HSI is not set
#
# PPS support
#CONFIG_PPS=m
# CONFIG_PPS_DEBUG is not set
#
# PPS clients support
#
# CONFIG_PPS_CLIENT_KTIMER is not setCONFIG_PPS_CLIENT_LDISC=m
#
# PPS generators support
#
#
# PTP clock support
#CONFIG_PTP_1588_CLOCK=m
#
# Enable PHYLIB and NETWORK_PHY_TIMESTAMPING to see the additional clocks. #
# CONFIG_PTP_1588_CLOCK_PCH is not setCONFIG_POWER_SUPPLY=y
# CONFIG_POWER_SUPPLY_DEBUG is not set
# CONFIG_TEST_POWER is not set
# CONFIG_BATTERY_BQ27x00 is not setCONFIG_POWER_RESET=y
# CONFIG_POWER_AVS is not set
# CONFIG_THERMAL is not setCONFIG_WATCHDOG=yCONFIG_WATCHDOG_CORE=y
# CONFIG_WATCHDOG_NOWAYOUT is not set
#
# Watchdog Device Drivers
#CONFIG_SOFT_WATCHDOG=mCONFIG_ZVM_WATCHDOG=m
# CONFIG_REGULATOR is not set
#
# HID support
#CONFIG_HID=y
# CONFIG_HID_BATTERY_STRENGTH is not set
# CONFIG_HIDRAW is not set
# CONFIG_UHID is not setCONFIG_HID_GENERIC=y
#
# Special HID drivers
#
# CONFIG_USB_ARCH_HAS_OHCI is not set
# CONFIG_USB_ARCH_HAS_EHCI is not set
# CONFIG_USB_ARCH_HAS_XHCI is not setCONFIG_MEMSTICK=m
# CONFIG_MEMSTICK_DEBUG is not set
#
# MemoryStick drivers
#
# CONFIG_MEMSTICK_UNSAFE_RESUME is not set
# CONFIG_MSPRO_BLOCK is not set
#
# MemoryStick Host Controller Drivers
#
# CONFIG_NEW_LEDS is not set
# CONFIG_ACCESSIBILITY is not setCONFIG_AUXDISPLAY=y
# CONFIG_UIO is not setCONFIG_VIRTIO=y
#
# Virtio drivers
#CONFIG_VIRTIO_BALLOON=m
#
# Microsoft Hyper-V guest support
#CONFIG_STAGING=y
# CONFIG_ECHO is not set
# CONFIG_COMEDI is not set
# CONFIG_ZSMALLOC is not set
# CONFIG_FT1000 is not set
#
# Speakup console speech
#CONFIG_STAGING_MEDIA=y
#
# Android
#
# CONFIG_ANDROID is not set
# CONFIG_WIMAX_GDM72XX is not set
# CONFIG_DGRP is not set
# CONFIG_ZCACHE is not set
#
# Hardware Spinlock drivers
#
# CONFIG_MAILBOX is not setCONFIG_IOMMU_SUPPORT=y
#
# Remoteproc drivers
#
#
# Rpmsg drivers
#
# CONFIG_VIRT_DRIVERS is not set
# CONFIG_PM_DEVFREQ is not set
# CONFIG_EXTCON is not set
# CONFIG_MEMORY is not set
# CONFIG_PWM is not set
#
# File systems
#
# CONFIG_EXT2_FS is not set
# CONFIG_EXT3_FS is not setCONFIG_EXT4_FS=yCONFIG_EXT4_USE_FOR_EXT23=yCONFIG_EXT4_FS_POSIX_ACL=yCONFIG_EXT4_FS_SECURITY=y
# CONFIG_EXT4_DEBUG is not setCONFIG_JBD2=y
# CONFIG_JBD2_DEBUG is not setCONFIG_FS_MBCACHE=y
# CONFIG_REISERFS_FS is not set
# CONFIG_JFS_FS is not setCONFIG_XFS_FS=mCONFIG_XFS_QUOTA=yCONFIG_XFS_POSIX_ACL=y
# CONFIG_XFS_RT is not set
# CONFIG_XFS_DEBUG is not set
# CONFIG_GFS2_FS is not set
# CONFIG_OCFS2_FS is not setCONFIG_BTRFS_FS=mCONFIG_BTRFS_FS_POSIX_ACL=y
# CONFIG_BTRFS_FS_CHECK_INTEGRITY is not set
# CONFIG_NILFS2_FS is not setCONFIG_FS_POSIX_ACL=yCONFIG_EXPORTFS=yCONFIG_FILE_LOCKING=yCONFIG_FSNOTIFY=yCONFIG_DNOTIFY=yCONFIG_INOTIFY_USER=yCONFIG_FANOTIFY=yCONFIG_FANOTIFY_ACCESS_PERMISSIONS=yCONFIG_QUOTA=yCONFIG_QUOTA_NETLINK_INTERFACE=yCONFIG_PRINT_QUOTA_WARNING=y
# CONFIG_QUOTA_DEBUG is not setCONFIG_QUOTA_TREE=y
# CONFIG_QFMT_V1 is not setCONFIG_QFMT_V2=yCONFIG_QUOTACTL=yCONFIG_AUTOFS4_FS=yCONFIG_FUSE_FS=mCONFIG_CUSE=mCONFIG_GENERIC_ACL=y
#
# Caches
#CONFIG_FSCACHE=mCONFIG_FSCACHE_STATS=y
# CONFIG_FSCACHE_HISTOGRAM is not set
# CONFIG_FSCACHE_DEBUG is not setCONFIG_FSCACHE_OBJECT_LIST=yCONFIG_CACHEFILES=m
# CONFIG_CACHEFILES_DEBUG is not set
# CONFIG_CACHEFILES_HISTOGRAM is not set
#
# CD-ROM/DVD Filesystems
#CONFIG_ISO9660_FS=yCONFIG_JOLIET=yCONFIG_ZISOFS=yCONFIG_UDF_FS=mCONFIG_UDF_NLS=y
#
# DOS/FAT/NT Filesystems
#CONFIG_FAT_FS=mCONFIG_MSDOS_FS=mCONFIG_VFAT_FS=mCONFIG_FAT_DEFAULT_CODEPAGE=437CONFIG_FAT_DEFAULT_IOCHARSET="ascii"
# CONFIG_NTFS_FS is not set
#
# Pseudo filesystems
#CONFIG_PROC_FS=yCONFIG_PROC_KCORE=yCONFIG_PROC_VMCORE=yCONFIG_PROC_SYSCTL=y CONFIG_PROC_PAGE_MONITOR=yCONFIG_SYSFS=yCONFIG_TMPFS=yCONFIG_TMPFS_POSIX_ACL=yCONFIG_TMPFS_XATTR=yCONFIG_HUGETLBFS=yCONFIG_HUGETLB_PAGE=yCONFIG_CONFIGFS_FS=y CONFIG_MISC_FILESYSTEMS=y
# CONFIG_ADFS_FS is not set
# CONFIG_AFFS_FS is not set
# CONFIG_ECRYPT_FS is not set
# CONFIG_HFS_FS is not set
# CONFIG_HFSPLUS_FS is not set
# CONFIG_BEFS_FS is not set
# CONFIG_BFS_FS is not set
# CONFIG_EFS_FS is not set
# CONFIG_LOGFS is not setCONFIG_CRAMFS=mCONFIG_SQUASHFS=mCONFIG_SQUASHFS_XATTR=y CONFIG_SQUASHFS_ZLIB=yCONFIG_SQUASHFS_LZO=yCONFIG_SQUASHFS_XZ=y
# CONFIG_SQUASHFS_4K_DEVBLK_SIZE is not set
# CONFIG_SQUASHFS_EMBEDDED is not set CONFIG_SQUASHFS_FRAGMENT_CACHE_SIZE=3
# CONFIG_VXFS_FS is not set
# CONFIG_MINIX_FS is not set
# CONFIG_MINIX_FS_NATIVE_ENDIAN is not set
# CONFIG_OMFS_FS is not set
# CONFIG_HPFS_FS is not set
# CONFIG_QNX4FS_FS is not set
# CONFIG_QNX6FS_FS is not set
# CONFIG_ROMFS_FS is not setCONFIG_PSTORE=y
# CONFIG_PSTORE_CONSOLE is not set
# CONFIG_SYSV_FS is not set
# CONFIG_UFS_FS is not set
# CONFIG_EXOFS_FS is not set
# CONFIG_F2FS_FS is not setCONFIG_ORE=mCONFIG_NETWORK_FILESYSTEMS=yCONFIG_NFS_FS=mCONFIG_NFS_V2=mCONFIG_NFS_V3=mCONFIG_NFS_V3_ACL=yCONFIG_NFS_V4=m
# CONFIG_NFS_SWAP is not set CONFIG_NFS_V4_1=yCONFIG_PNFS_FILE_LAYOUT=mCONFIG_PNFS_BLOCK=mCONFIG_PNFS_OBJLAYOUT=m CONFIG_NFS_V4_1_IMPLEMENTATION_ID_DOMAIN="kernel.org" CONFIG_NFS_FSCACHE=y
# CONFIG_NFS_USE_LEGACY_DNS is not set CONFIG_NFS_USE_KERNEL_DNS=y CONFIG_NFS_DEBUG=yCONFIG_NFSD=mCONFIG_NFSD_V2_ACL=yCONFIG_NFSD_V3=yCONFIG_NFSD_V3_ACL=yCONFIG_NFSD_V4=y
# CONFIG_NFSD_FAULT_INJECTION is not set CONFIG_LOCKD=mCONFIG_LOCKD_V4=yCONFIG_NFS_ACL_SUPPORT=mCONFIG_NFS_COMMON=yCONFIG_SUNRPC=mCONFIG_SUNRPC_GSS=m CONFIG_SUNRPC_BACKCHANNEL=y CONFIG_RPCSEC_GSS_KRB5=m CONFIG_SUNRPC_DEBUG=y
# CONFIG_CEPH_FS is not setCONFIG_CIFS=mCONFIG_CIFS_STATS=y
# CONFIG_CIFS_STATS2 is not set CONFIG_CIFS_WEAK_PW_HASH=y CONFIG_CIFS_UPCALL=yCONFIG_CIFS_XATTR=yCONFIG_CIFS_POSIX=yCONFIG_CIFS_ACL=yCONFIG_CIFS_DEBUG=y
# CONFIG_CIFS_DEBUG2 is not set CONFIG_CIFS_DFS_UPCALL=yCONFIG_CIFS_SMB2=y
# CONFIG_CIFS_FSCACHE is not set
# CONFIG_NCP_FS is not set
# CONFIG_CODA_FS is not set
# CONFIG_AFS_FS is not setCONFIG_NLS=yCONFIG_NLS_DEFAULT="utf8" CONFIG_NLS_CODEPAGE_437=y CONFIG_NLS_CODEPAGE_737=m CONFIG_NLS_CODEPAGE_775=m CONFIG_NLS_CODEPAGE_850=m CONFIG_NLS_CODEPAGE_852=m CONFIG_NLS_CODEPAGE_855=m CONFIG_NLS_CODEPAGE_857=m CONFIG_NLS_CODEPAGE_860=m CONFIG_NLS_CODEPAGE_861=m CONFIG_NLS_CODEPAGE_862=m CONFIG_NLS_CODEPAGE_863=m CONFIG_NLS_CODEPAGE_864=m CONFIG_NLS_CODEPAGE_865=m CONFIG_NLS_CODEPAGE_866=m CONFIG_NLS_CODEPAGE_869=m CONFIG_NLS_CODEPAGE_936=m CONFIG_NLS_CODEPAGE_950=m CONFIG_NLS_CODEPAGE_932=m CONFIG_NLS_CODEPAGE_949=m CONFIG_NLS_CODEPAGE_874=m CONFIG_NLS_ISO8859_8=m CONFIG_NLS_CODEPAGE_1250=m CONFIG_NLS_CODEPAGE_1251=mCONFIG_NLS_ASCII=yCONFIG_NLS_ISO8859_1=mCONFIG_NLS_ISO8859_2=mCONFIG_NLS_ISO8859_3=mCONFIG_NLS_ISO8859_4=mCONFIG_NLS_ISO8859_5=mCONFIG_NLS_ISO8859_6=mCONFIG_NLS_ISO8859_7=mCONFIG_NLS_ISO8859_9=m CONFIG_NLS_ISO8859_13=m CONFIG_NLS_ISO8859_14=m CONFIG_NLS_ISO8859_15=mCONFIG_NLS_KOI8_R=mCONFIG_NLS_KOI8_U=mCONFIG_NLS_MAC_ROMAN=m CONFIG_NLS_MAC_CELTIC=m CONFIG_NLS_MAC_CENTEURO=m CONFIG_NLS_MAC_CROATIAN=m CONFIG_NLS_MAC_CYRILLIC=m CONFIG_NLS_MAC_GAELIC=m CONFIG_NLS_MAC_GREEK=m CONFIG_NLS_MAC_ICELAND=m CONFIG_NLS_MAC_INUIT=m CONFIG_NLS_MAC_ROMANIAN=m CONFIG_NLS_MAC_TURKISH=mCONFIG_NLS_UTF8=mCONFIG_DLM=mCONFIG_DLM_DEBUG=y
#
# Kernel hacking
#CONFIG_TRACE_IRQFLAGS_SUPPORT=yCONFIG_PRINTK_TIME=yCONFIG_DEFAULT_MESSAGE_LOGLEVEL=4
# CONFIG_ENABLE_WARN_DEPRECATED is not setCONFIG_ENABLE_MUST_CHECK=yCONFIG_FRAME_WARN=1024CONFIG_MAGIC_SYSRQ=yCONFIG_STRIP_ASM_SYMS=y
# CONFIG_READABLE_ASM is not set
# CONFIG_UNUSED_SYMBOLS is not setCONFIG_DEBUG_FS=yCONFIG_HEADERS_CHECK=yCONFIG_DEBUG_SECTION_MISMATCH=yCONFIG_DEBUG_KERNEL=y
# CONFIG_PANIC_ON_OOPS is not setCONFIG_PANIC_ON_OOPS_VALUE=0CONFIG_DETECT_HUNG_TASK=yCONFIG_DEFAULT_HUNG_TASK_TIMEOUT=120
# CONFIG_BOOTPARAM_HUNG_TASK_PANIC is not setCONFIG_BOOTPARAM_HUNG_TASK_PANIC_VALUE=0CONFIG_SCHED_DEBUG=yCONFIG_SCHEDSTATS=yCONFIG_TIMER_STATS=y
# CONFIG_DEBUG_OBJECTS is not set
# CONFIG_SLUB_DEBUG_ON is not set
# CONFIG_SLUB_STATS is not setCONFIG_HAVE_DEBUG_KMEMLEAK=y
# CONFIG_DEBUG_KMEMLEAK is not set
# CONFIG_DEBUG_RT_MUTEXES is not set
# CONFIG_RT_MUTEX_TESTER is not set
# CONFIG_DEBUG_SPINLOCK is not set
# CONFIG_DEBUG_MUTEXES is not set
# CONFIG_DEBUG_LOCK_ALLOC is not set
# CONFIG_PROVE_LOCKING is not set
# CONFIG_LOCK_STAT is not set
# CONFIG_DEBUG_ATOMIC_SLEEP is not set
# CONFIG_DEBUG_LOCKING_API_SELFTESTS is not setCONFIG_STACKTRACE=y
# CONFIG_DEBUG_STACK_USAGE is not set
# CONFIG_DEBUG_KOBJECT is not setCONFIG_DEBUG_BUGVERBOSE=yCONFIG_DEBUG_INFO=y
# CONFIG_DEBUG_INFO_REDUCED is not setCONFIG_DEBUG_VM=y
# CONFIG_DEBUG_VM_RB is not set
# CONFIG_DEBUG_WRITECOUNT is not setCONFIG_DEBUG_MEMORY_INIT=yCONFIG_DEBUG_LIST=y
# CONFIG_TEST_LIST_SORT is not set
# CONFIG_DEBUG_SG is not set
# CONFIG_DEBUG_NOTIFIERS is not set
# CONFIG_DEBUG_CREDENTIALS is not set
#
# RCU Debugging
#CONFIG_SPARSE_RCU_POINTER=y
# CONFIG_RCU_TORTURE_TEST is not setCONFIG_RCU_CPU_STALL_TIMEOUT=60
# CONFIG_RCU_CPU_STALL_INFO is not set
# CONFIG_RCU_TRACE is not set
# CONFIG_KPROBES_SANITY_TEST is not set
# CONFIG_BACKTRACE_SELF_TEST is not set
# CONFIG_DEBUG_BLOCK_EXT_DEVT is not set
# CONFIG_DEBUG_FORCE_WEAK_PER_CPU is not set
# CONFIG_DEBUG_PER_CPU_MAPS is not set
# CONFIG_LKDTM is not set
# CONFIG_NOTIFIER_ERROR_INJECTION is not set
# CONFIG_FAULT_INJECTION is not setCONFIG_LATENCYTOP=y
# CONFIG_DEBUG_PAGEALLOC is not setCONFIG_NOP_TRACER=yCONFIG_HAVE_FUNCTION_TRACER=yCONFIG_HAVE_FUNCTION_GRAPH_TRACER=yCONFIG_HAVE_FUNCTION_TRACE_MCOUNT_TEST=yCONFIG_HAVE_DYNAMIC_FTRACE=yCONFIG_HAVE_FTRACE_MCOUNT_RECORD=yCONFIG_HAVE_SYSCALL_TRACEPOINTS=yCONFIG_HAVE_C_RECORDMCOUNT=yCONFIG_TRACER_MAX_TRACE=yCONFIG_TRACE_CLOCK=yCONFIG_RING_BUFFER=yCONFIG_EVENT_TRACING=yCONFIG_CONTEXT_SWITCH_TRACER=yCONFIG_RING_BUFFER_ALLOW_SWAP=yCONFIG_TRACING=yCONFIG_GENERIC_TRACER=yCONFIG_TRACING_SUPPORT=yCONFIG_FTRACE=y
# CONFIG_FUNCTION_TRACER is not set
# CONFIG_IRQSOFF_TRACER is not setCONFIG_SCHED_TRACER=y
# CONFIG_FTRACE_SYSCALLS is not set
# CONFIG_TRACER_SNAPSHOT is not setCONFIG_BRANCH_PROFILE_NONE=y
# CONFIG_PROFILE_ANNOTATED_BRANCHES is not set
# CONFIG_PROFILE_ALL_BRANCHES is not set
# CONFIG_STACK_TRACER is not setCONFIG_BLK_DEV_IO_TRACE=yCONFIG_KPROBE_EVENT=yCONFIG_PROBE_EVENTS=y
# CONFIG_FTRACE_STARTUP_TEST is not setCONFIG_RING_BUFFER_BENCHMARK=m
# CONFIG_RBTREE_TEST is not set
# CONFIG_INTERVAL_TREE_TEST is not setCONFIG_BUILD_DOCSRC=yCONFIG_DYNAMIC_DEBUG=yCONFIG_ATOMIC64_SELFTEST=yCONFIG_ASYNC_RAID6_TEST=m
# CONFIG_SAMPLES is not setCONFIG_TEST_KSTRTOX=yCONFIG_STRICT_DEVMEM=y
# CONFIG_DEBUG_STRICT_USER_COPY_CHECKS is not set
# CONFIG_S390_PTDUMP is not setCONFIG_DEBUG_SET_MODULE_RONX=y
#
# Security options
#CONFIG_KEYS=yCONFIG_ENCRYPTED_KEYS=m CONFIG_KEYS_DEBUG_PROC_KEYS=y
# CONFIG_SECURITY_DMESG_RESTRICT is not set CONFIG_SECURITY=yCONFIG_SECURITYFS=yCONFIG_SECURITY_NETWORK=y CONFIG_SECURITY_NETWORK_XFRM=y
# CONFIG_SECURITY_PATH is not set CONFIG_LSM_MMAP_MIN_ADDR=65535 CONFIG_SECURITY_SELINUX=y CONFIG_SECURITY_SELINUX_BOOTPARAM=y CONFIG_SECURITY_SELINUX_BOOTPARAM_VALUE=1 CONFIG_SECURITY_SELINUX_DISABLE=y CONFIG_SECURITY_SELINUX_DEVELOP=y CONFIG_SECURITY_SELINUX_AVC_STATS=y CONFIG_SECURITY_SELINUX_CHECKREQPROT_VALUE=1
# CONFIG_SECURITY_SELINUX_POLICYDB_VERSION_MAX is not set # CONFIG_SECURITY_SMACK is not set
# CONFIG_SECURITY_TOMOYO is not set
# CONFIG_SECURITY_APPARMOR is not set
# CONFIG_SECURITY_YAMA is not set
# CONFIG_IMA is not set
# CONFIG_EVM is not set CONFIG_DEFAULT_SECURITY_SELINUX=y
# CONFIG_DEFAULT_SECURITY_DAC is not set CONFIG_DEFAULT_SECURITY="selinux" CONFIG_XOR_BLOCKS=mCONFIG_ASYNC_CORE=mCONFIG_ASYNC_MEMCPY=mCONFIG_ASYNC_XOR=mCONFIG_ASYNC_PQ=mCONFIG_ASYNC_RAID6_RECOV=mCONFIG_CRYPTO=y
#
# Crypto core or helper
#CONFIG_CRYPTO_FIPS=yCONFIG_CRYPTO_ALGAPI=yCONFIG_CRYPTO_ALGAPI2=yCONFIG_CRYPTO_AEAD=yCONFIG_CRYPTO_AEAD2=yCONFIG_CRYPTO_BLKCIPHER=yCONFIG_CRYPTO_BLKCIPHER2=yCONFIG_CRYPTO_HASH=yCONFIG_CRYPTO_HASH2=yCONFIG_CRYPTO_RNG=yCONFIG_CRYPTO_RNG2=yCONFIG_CRYPTO_PCOMP=mCONFIG_CRYPTO_PCOMP2=yCONFIG_CRYPTO_MANAGER=yCONFIG_CRYPTO_MANAGER2=yCONFIG_CRYPTO_USER=m
# CONFIG_CRYPTO_MANAGER_DISABLE_TESTS is not setCONFIG_CRYPTO_GF128MUL=mCONFIG_CRYPTO_NULL=mCONFIG_CRYPTO_PCRYPT=mCONFIG_CRYPTO_WORKQUEUE=y
# CONFIG_CRYPTO_CRYPTD is not setCONFIG_CRYPTO_AUTHENC=mCONFIG_CRYPTO_TEST=m
#
# Authenticated Encryption with Associated Data
#CONFIG_CRYPTO_CCM=mCONFIG_CRYPTO_GCM=mCONFIG_CRYPTO_SEQIV=y
#
# Block modes
#CONFIG_CRYPTO_CBC=yCONFIG_CRYPTO_CTR=yCONFIG_CRYPTO_CTS=mCONFIG_CRYPTO_ECB=yCONFIG_CRYPTO_LRW=mCONFIG_CRYPTO_PCBC=mCONFIG_CRYPTO_XTS=m
#
# Hash modes
#CONFIG_CRYPTO_HMAC=yCONFIG_CRYPTO_XCBC=mCONFIG_CRYPTO_VMAC=m
#
# Digest
#CONFIG_CRYPTO_CRC32C=y
# CONFIG_CRYPTO_CRC32 is not setCONFIG_CRYPTO_GHASH=mCONFIG_CRYPTO_MD4=mCONFIG_CRYPTO_MD5=yCONFIG_CRYPTO_MICHAEL_MIC=mCONFIG_CRYPTO_RMD128=mCONFIG_CRYPTO_RMD160=mCONFIG_CRYPTO_RMD256=mCONFIG_CRYPTO_RMD320=mCONFIG_CRYPTO_SHA1=yCONFIG_CRYPTO_SHA256=yCONFIG_CRYPTO_SHA512=mCONFIG_CRYPTO_TGR192=mCONFIG_CRYPTO_WP512=m
#
# Ciphers
#CONFIG_CRYPTO_AES=yCONFIG_CRYPTO_ANUBIS=mCONFIG_CRYPTO_ARC4=mCONFIG_CRYPTO_BLOWFISH=mCONFIG_CRYPTO_BLOWFISH_COMMON=mCONFIG_CRYPTO_CAMELLIA=mCONFIG_CRYPTO_CAST_COMMON=mCONFIG_CRYPTO_CAST5=mCONFIG_CRYPTO_CAST6=mCONFIG_CRYPTO_DES=mCONFIG_CRYPTO_FCRYPT=mCONFIG_CRYPTO_KHAZAD=mCONFIG_CRYPTO_SALSA20=mCONFIG_CRYPTO_SEED=mCONFIG_CRYPTO_SERPENT=mCONFIG_CRYPTO_TEA=mCONFIG_CRYPTO_TWOFISH=mCONFIG_CRYPTO_TWOFISH_COMMON=m
#
# Compression
#CONFIG_CRYPTO_DEFLATE=mCONFIG_CRYPTO_ZLIB=mCONFIG_CRYPTO_LZO=m
#
# Random Number Generation
#CONFIG_CRYPTO_ANSI_CPRNG=mCONFIG_CRYPTO_USER_API=yCONFIG_CRYPTO_USER_API_HASH=yCONFIG_CRYPTO_USER_API_SKCIPHER=yCONFIG_CRYPTO_HW=yCONFIG_ZCRYPT=mCONFIG_CRYPTO_SHA1_S390=mCONFIG_CRYPTO_SHA256_S390=mCONFIG_CRYPTO_SHA512_S390=mCONFIG_CRYPTO_DES_S390=mCONFIG_CRYPTO_AES_S390=mCONFIG_S390_PRNG=mCONFIG_CRYPTO_GHASH_S390=mCONFIG_ASYMMETRIC_KEY_TYPE=yCONFIG_ASYMMETRIC_PUBLIC_KEY_SUBTYPE=yCONFIG_PUBLIC_KEY_ALGO_RSA=yCONFIG_X509_CERTIFICATE_PARSER=yCONFIG_BINARY_PRINTF=y
#
# Library routines
#CONFIG_RAID6_PQ=mCONFIG_BITREVERSE=y
# CONFIG_GENERIC_IO is not setCONFIG_CRC_CCITT=mCONFIG_CRC16=yCONFIG_CRC_T10DIF=mCONFIG_CRC_ITU_T=mCONFIG_CRC32=y
# CONFIG_CRC32_SELFTEST is not setCONFIG_CRC32_SLICEBY8=y
# CONFIG_CRC32_SLICEBY4 is not set
# CONFIG_CRC32_SARWATE is not set
# CONFIG_CRC32_BIT is not set
# CONFIG_CRC7 is not setCONFIG_LIBCRC32C=mCONFIG_CRC8=mCONFIG_ZLIB_INFLATE=yCONFIG_ZLIB_DEFLATE=mCONFIG_LZO_COMPRESS=yCONFIG_LZO_DECOMPRESS=yCONFIG_XZ_DEC=yCONFIG_XZ_DEC_X86=yCONFIG_XZ_DEC_POWERPC=yCONFIG_XZ_DEC_IA64=yCONFIG_XZ_DEC_ARM=yCONFIG_XZ_DEC_ARMTHUMB=yCONFIG_XZ_DEC_SPARC=yCONFIG_XZ_DEC_BCJ=y
# CONFIG_XZ_DEC_TEST is not setCONFIG_DECOMPRESS_GZIP=yCONFIG_DECOMPRESS_BZIP2=yCONFIG_DECOMPRESS_LZMA=yCONFIG_DECOMPRESS_XZ=yCONFIG_DECOMPRESS_LZO=yCONFIG_TEXTSEARCH=yCONFIG_TEXTSEARCH_KMP=mCONFIG_TEXTSEARCH_BM=mCONFIG_TEXTSEARCH_FSM=mCONFIG_DQL=yCONFIG_NLATTR=yCONFIG_ARCH_HAS_ATOMIC64_DEC_IF_POSITIVE=y
# CONFIG_AVERAGE is not setCONFIG_CLZ_TAB=yCONFIG_CORDIC=m
# CONFIG_DDR is not setCONFIG_MPILIB=yCONFIG_OID_REGISTRY=y
#
# Virtualization
#CONFIG_PFAULT=yCONFIG_CMM=mCONFIG_CMM_IUCV=yCONFIG_APPLDATA_BASE=yCONFIG_APPLDATA_MEM=mCONFIG_APPLDATA_OS=mCONFIG_APPLDATA_NET_SUM=mCONFIG_S390_HYPFS_FS=yCONFIG_HAVE_KVM=yCONFIG_VIRTUALIZATION=y
# CONFIG_KVM is not setCONFIG_VHOST_NET=mCONFIG_TCM_VHOST=mCONFIG_S390_GUEST=y
Hello Heiko,
----- Original Message -----
Quoted 53 lines
From: "Heiko Carstens" <[email protected]>
To: "Zhouping Liu" <[email protected]>
Cc: [email protected], "LKML" <[email protected]>, "caiqian" <[email protected]>, "Caspar Zhang" <[email protected]>, "Martin Schwidefsky" <[email protected]>
Sent: Tuesday, April 16, 2013 3:50:47 PM
Subject: Re: [BUG][s390x] mm: system crashedQuoted 41 lines · Zhouping Liu
Quoted 31 lines · Heiko Carstens
Quoted 26 lines · Zhouping Liu
� 16109.346170¨ Call Trace:
� 16109.346179¨ (� <0000000000100920>¨show_trace+0x128/0x12c)� 16109.346195¨ � <00000000001cd320>¨ rcu_check_callbacks+0x458/0xccc � 16109.346209¨ � <0000000000140f2e>¨ update_process_times+0x4a/0x74 � 16109.346222¨ � <0000000000199452>¨ tick_sched_handle.isra.12+0x5e/0x70 � 16109.346235¨ � <00000000001995aa>¨ tick_sched_timer+0x6a/0x98 � 16109.346247¨ � <000000000015c1ea>¨ __run_hrtimer+0x8e/0x200 � 16109.346381¨ � <000000000015d1b2>¨ hrtimer_interrupt+0x212/0x2b0 � 16109.346385¨ � <00000000001040f6>¨ clock_comparator_work+0x4a/0x54 � 16109.346390¨ � <000000000010d658>¨ do_extint+0x158/0x15c � 16109.346396¨ � <000000000062aa24>¨ ext_skip+0x38/0x3c � 16109.346404¨ � <00000000001153c8>¨ smp_yield_cpu+0x44/0x48 � 16109.346412¨ (� <000003d10051aec0>¨ 0x3d10051aec0) � 16109.346457¨ � <000000000024206a>¨ __page_check_address+0x16a/0x170 � 16109.346466¨ � <00000000002423a2>¨ page_referenced_one+0x3e/0xa0 � 16109.346501¨ � <000000000024427c>¨ page_referenced+0x32c/0x41c � 16109.346510¨ � <000000000021b1dc>¨ shrink_page_list+0x380/0xb9c � 16109.346521¨ � <000000000021c0a6>¨ shrink_inactive_list+0x1c6/0x56c � 16109.346532¨ � <000000000021c69e>¨ shrink_lruvec+0x252/0x56c � 16109.346542¨ � <000000000021ca44>¨ shrink_zone+0x8c/0x1bc � 16109.346553¨ � <000000000021d080>¨ balance_pgdat+0x50c/0x658 � 16109.346564¨ � <000000000021d318>¨ kswapd+0x14c/0x470 � 16109.346576¨ � <0000000000158292>¨ kthread+0xda/0xe4 � 16109.346656¨ � <000000000062a5de>¨ kernel_thread_starter+0x6/0xc � 16109.346682¨ � <000000000062a5d8>¨ kernel_thread_starter+0x0/0xcQuoted 4 lines
[-- MARK -- Fri Apr 12 06:15:00 2013]
� 16289.386061¨ INFO:rcu_schedself-detected stall on CPU { 0} (t=42010 jiffies
g=89766 c=89765 q=10627)Quoted 2 lines
Did the system really crash or did you just see the rcu related warning(s)?
I just check it again, actually at first the system didn't really crash, but the system is very slow in response.
and the reproducer process can't be killed, after I did some common actions such as 'ls' 'vim' etc, the system
seemed to be really crashed, no any response.Quoted 3 lines
also in the previous testing, I can remember that the system would be no any response for a long time, just only repeatedly print out the such above 'Call Trace' into console.
Quoted 7 lines
Ok, thanks.
Just a couple of more questions: did you see this also on other archs, or just
s390 (if you tried other platforms at all).If you have some time, could you please repeat your test with the kernel command line option " user_mode=home "?
I tested the system with the kernel parameter, but the issue still appeared, I just to say it takes longer time to reproduce the issue than the before.
Quoted 4 lines
As far as I can tell there was only one s390 patch merged that was mmap related:
486c0a0bc80d370471b21662bf03f04fbb37cdc6"s390/mm: Fix crst upgrade of mmap withMAP_FIXED".
also I tested the revert commit, unluckily, the same issue as the before.
Signature · 8 lines
Thanks, Zhouping -- To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to [email protected]. For more info on Linux MM, see: http://www.linux-mm.org/ . Don't email: <a href=mailto:"[email protected]"> [email protected] </a>
Quoted 8 lines · Zhouping Liu
Hello Heiko,
Quoted 2 lines
If you have some time, could you please repeat your test with the kernel command line option " user_mode=home "?
I tested the system with the kernel parameter, but the issue still appeared, I just to say it takes longer time to reproduce the issue than the before.
Quoted 1 line
Quoted 5 lines
Quoted 3 lines
As far as I can tell there was only one s390 patch merged that was mmap related:
486c0a0bc80d370471b21662bf03f04fbb37cdc6"s390/mm: Fix crst upgrade of mmap withMAP_FIXED".also I tested the revert commit, unluckily, the same issue as the before.
Ok, thanks for verifying! I'll look into it; hopefully I can reproduce it here as well.
Thanks!
[v3.9-rc8]: kernel BUG at mm/memcontrol.c:3994! (was: Re: [BUG][s390x] mm: system crashed)
Quoted 2 lines · Heiko Carstens
Ok, thanks for verifying! I'll look into it; hopefully I can reproduce it here as well.
That seems to be a common code bug. I can easily trigger the VM_BUG_ON() below (when I force the system to swap):
[ 48.347963] ------------[ cut here ]------------
[ 48.347972] kernel BUG at mm/memcontrol.c:3994!
[ 48.348012] illegal operation: 0001 [#1] SMP
[ 48.348015] Modules linked in:
[ 48.348017] CPU: 1 Not tainted 3.9.0-rc8+ #38
[ 48.348020] Process mmap2 (pid: 635, task: 0000000029476100, ksp: 000000002e91b938)
[ 48.348022] Krnl PSW : 0704f00180000000 000000000026552c (__mem_cgroup_uncharge_common+0x2c4/0x33c)
[ 48.348032] R:0 T:1 IO:1 EX:1 Key:0 M:1 W:0 P:0 AS:3 CC:3 PM:0 EA:3
Krnl GPRS: 0000000000000008 0000000000000009 000003d1002a9200 0000000000000000
[ 48.348039] 0000000000000000 00000000006812d8 000003ffdf339000 00000000321a6f98
[ 48.348043] 000003fffce11000 0000000000000000 0000000000000001 000003d1002a9200
[ 48.348046] 0000000000000001 0000000000681b88 000000002e91bc18 000000002e91bbd0
[ 48.348057] Krnl Code: 000000000026551e: c0e5fffaa2a1 brasl %r14,1b9a60
0000000000265524: a7f4ff7d brc 15,26541e
#0000000000265528: a7f40001 brc 15,26552a
>000000000026552c: e3c0b8200124 stg %r12,6176(%r11)
0000000000265532: a7f4ff57 brc 15,2653e0
0000000000265536: e310b8280104 lg %r1,6184(%r11)
000000000026553c: a71b0001 aghi %r1,1
0000000000265540: e310b8280124 stg %r1,6184(%r11)
[ 48.348099] Call Trace:
[ 48.348100] ([<000003d1002a91c0>] 0x3d1002a91c0)
[ 48.348102] [<00000000002404aa>] page_remove_rmap+0xf2/0x16c
[ 48.348106] [<0000000000232dc8>] unmap_single_vma+0x494/0x7d8
[ 48.348107] [<0000000000233ac0>] unmap_vmas+0x50/0x74
[ 48.348109] [<00000000002396ec>] unmap_region+0x9c/0x110
[ 48.348110] [<000000000023bd18>] do_munmap+0x284/0x470
[ 48.348111] [<000000000023bf56>] vm_munmap+0x52/0x70
[ 48.348113] [<000000000023cf32>] SyS_munmap+0x3a/0x4c
[ 48.348114] [<0000000000665e14>] sysc_noemu+0x22/0x28
[ 48.348118] [<000003fffcf187b2>] 0x3fffcf187b2
[ 48.348119] Last Breaking-Event-Address:
[ 48.348120] [<0000000000265528>] __mem_cgroup_uncharge_common+0x2c0/0x33cLooking at the code, the code flow is:
page_remove_rmap() -> mem_cgroup_uncharge_page() -> __mem_cgroup_uncharge_common()
Note that in mem_cgroup_uncharge_page() the page in question passed the check:
[...]
if (PageSwapCache(page))
return;[...]
and just a couple of instructions later the VM_BUG_ON() within __mem_cgroup_uncharge_common() triggers:
[...]
if (mem_cgroup_disabled())
return NULL;
VM_BUG_ON(PageSwapCache(page));[...]
Which means that another cpu changed the pageflags concurrently. In fact, looking at the dump a different cpu is indeed busy with running kswapd.
So.. this seems to be somewhat broken. Anyone familiar with memcontrol?
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to [email protected]. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"[email protected]"> [email protected] </a>
Re: [v3.9-rc8]: kernel BUG at mm/memcontrol.c:3994! (was: Re: [BUG][s390x] mm: system crashed)
Quoted 56 lines · Heiko Carstens
Quoted 2 lines · Heiko Carstens
Ok, thanks for verifying! I'll look into it; hopefully I can reproduce it here as well.
That seems to be a common code bug. I can easily trigger the
VM_BUG_ON()below (when I force the system to swap):[ 48.347963] ------------[ cut here ]------------ [ 48.347972] kernel BUG at mm/memcontrol.c:3994! [ 48.348012] illegal operation: 0001 [#1] SMP [ 48.348015] Modules linked in: [ 48.348017] CPU: 1 Not tainted 3.9.0-rc8+ #38 [ 48.348020] Process mmap2 (pid: 635, task: 0000000029476100, ksp: 000000002e91b938) [ 48.348022] Krnl PSW : 0704f00180000000 000000000026552c (__mem_cgroup_uncharge_common+0x2c4/0x33c) [ 48.348032] R:0 T:1 IO:1 EX:1 Key:0 M:1 W:0 P:0 AS:3 CC:3 PM:0 EA:3 Krnl GPRS: 0000000000000008 0000000000000009 000003d1002a9200 0000000000000000 [ 48.348039] 0000000000000000 00000000006812d8 000003ffdf339000 00000000321a6f98 [ 48.348043] 000003fffce11000 0000000000000000 0000000000000001 000003d1002a9200 [ 48.348046] 0000000000000001 0000000000681b88 000000002e91bc18 000000002e91bbd0 [ 48.348057] Krnl Code: 000000000026551e: c0e5fffaa2a1 brasl %r14,1b9a60 0000000000265524: a7f4ff7d brc 15,26541e #0000000000265528: a7f40001 brc 15,26552a >000000000026552c: e3c0b8200124 stg %r12,6176(%r11) 0000000000265532: a7f4ff57 brc 15,2653e0 0000000000265536: e310b8280104 lg %r1,6184(%r11) 000000000026553c: a71b0001 aghi %r1,1 0000000000265540: e310b8280124 stg %r1,6184(%r11) [ 48.348099] Call Trace: [ 48.348100] ([<000003d1002a91c0>] 0x3d1002a91c0) [ 48.348102] [<00000000002404aa>] page_remove_rmap+0xf2/0x16c [ 48.348106] [<0000000000232dc8>] unmap_single_vma+0x494/0x7d8 [ 48.348107] [<0000000000233ac0>] unmap_vmas+0x50/0x74 [ 48.348109] [<00000000002396ec>] unmap_region+0x9c/0x110 [ 48.348110] [<000000000023bd18>] do_munmap+0x284/0x470 [ 48.348111] [<000000000023bf56>] vm_munmap+0x52/0x70 [ 48.348113] [<000000000023cf32>] SyS_munmap+0x3a/0x4c [ 48.348114] [<0000000000665e14>] sysc_noemu+0x22/0x28 [ 48.348118] [<000003fffcf187b2>] 0x3fffcf187b2 [ 48.348119] Last Breaking-Event-Address: [ 48.348120] [<0000000000265528>] __mem_cgroup_uncharge_common+0x2c0/0x33cLooking at the code, the code flow is:
page_remove_rmap()->mem_cgroup_uncharge_page()->__mem_cgroup_uncharge_common()Note that in
mem_cgroup_uncharge_page()the page in question passed the check:[...]
if (PageSwapCache(page)) return;[...]
and just a couple of instructions later the
VM_BUG_ON()within__mem_cgroup_uncharge_common()triggers:[...]
Quoted 8 lines
if (mem_cgroup_disabled()) return NULL; VM_BUG_ON(PageSwapCache(page));[...]
Which means that another cpu changed the pageflags concurrently. In fact, looking at the dump a different cpu is indeed busy with running kswapd.
Hmm, maybe I am missing something but it really looks like we can race here. Reclaim path takes the page lock while zap_pte takes page table lock so nothing prevents them from racing here:shrink_page_list zap_pte_range
trylock_page pte_offset_map_lock
add_to_swap page_remove_rmap
/* Page can be still mapped */
add_to_swap_cache atomic_add_negative(_mapcount)
__add_to_swap_cache mem_cgroup_uncharge_page
(PageSwapCache(page)) && return
SetPageSwapCache
__mem_cgroup_uncharge_common
VM_BUG_ON(PageSwapCache(page))Maybe not many people run with CONFIG_DEBUG_VM enabled these days so we do not this more often (even me testing configs are not consistent in that regards and only few have it on). The only thing that changed in this area recently is 0c59b89c which made the test VM_BUG_ON rather then simple return in 3.6
And maybe the BUG_ON is too harsh as CgroupUsed should guarantee that the uncharge will eventually go away. What do you think Johannes?
Quoted 1 line
So.. this seems to be somewhat broken. Anyone familiar with memcontrol?
Signature · 8 lines
Michal Hocko SUSE Labs -- To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to [email protected]. For more info on Linux MM, see: http://www.linux-mm.org/ . Don't email: <a href=mailto:"[email protected]"> [email protected] </a>
Re: [v3.9-rc8]: kernel BUG at mm/memcontrol.c:3994! (was: Re: [BUG][s390x] mm: system crashed)
Quoted 80 lines · Michal Hocko
Quoted 56 lines · Heiko Carstens
Quoted 2 lines · Heiko Carstens
Ok, thanks for verifying! I'll look into it; hopefully I can reproduce it here as well.
That seems to be a common code bug. I can easily trigger the
VM_BUG_ON()below (when I force the system to swap):[ 48.347963] ------------[ cut here ]------------ [ 48.347972] kernel BUG at mm/memcontrol.c:3994! [ 48.348012] illegal operation: 0001 [#1] SMP [ 48.348015] Modules linked in: [ 48.348017] CPU: 1 Not tainted 3.9.0-rc8+ #38 [ 48.348020] Process mmap2 (pid: 635, task: 0000000029476100, ksp: 000000002e91b938) [ 48.348022] Krnl PSW : 0704f00180000000 000000000026552c (__mem_cgroup_uncharge_common+0x2c4/0x33c) [ 48.348032] R:0 T:1 IO:1 EX:1 Key:0 M:1 W:0 P:0 AS:3 CC:3 PM:0 EA:3 Krnl GPRS: 0000000000000008 0000000000000009 000003d1002a9200 0000000000000000 [ 48.348039] 0000000000000000 00000000006812d8 000003ffdf339000 00000000321a6f98 [ 48.348043] 000003fffce11000 0000000000000000 0000000000000001 000003d1002a9200 [ 48.348046] 0000000000000001 0000000000681b88 000000002e91bc18 000000002e91bbd0 [ 48.348057] Krnl Code: 000000000026551e: c0e5fffaa2a1 brasl %r14,1b9a60 0000000000265524: a7f4ff7d brc 15,26541e #0000000000265528: a7f40001 brc 15,26552a >000000000026552c: e3c0b8200124 stg %r12,6176(%r11) 0000000000265532: a7f4ff57 brc 15,2653e0 0000000000265536: e310b8280104 lg %r1,6184(%r11) 000000000026553c: a71b0001 aghi %r1,1 0000000000265540: e310b8280124 stg %r1,6184(%r11) [ 48.348099] Call Trace: [ 48.348100] ([<000003d1002a91c0>] 0x3d1002a91c0) [ 48.348102] [<00000000002404aa>] page_remove_rmap+0xf2/0x16c [ 48.348106] [<0000000000232dc8>] unmap_single_vma+0x494/0x7d8 [ 48.348107] [<0000000000233ac0>] unmap_vmas+0x50/0x74 [ 48.348109] [<00000000002396ec>] unmap_region+0x9c/0x110 [ 48.348110] [<000000000023bd18>] do_munmap+0x284/0x470 [ 48.348111] [<000000000023bf56>] vm_munmap+0x52/0x70 [ 48.348113] [<000000000023cf32>] SyS_munmap+0x3a/0x4c [ 48.348114] [<0000000000665e14>] sysc_noemu+0x22/0x28 [ 48.348118] [<000003fffcf187b2>] 0x3fffcf187b2 [ 48.348119] Last Breaking-Event-Address: [ 48.348120] [<0000000000265528>] __mem_cgroup_uncharge_common+0x2c0/0x33cLooking at the code, the code flow is:
page_remove_rmap()->mem_cgroup_uncharge_page()->__mem_cgroup_uncharge_common()Note that in
mem_cgroup_uncharge_page()the page in question passed the check:[...]
if (PageSwapCache(page)) return;[...]
and just a couple of instructions later the
VM_BUG_ON()within__mem_cgroup_uncharge_common()triggers:[...]
Quoted 8 lines
if (mem_cgroup_disabled()) return NULL; VM_BUG_ON(PageSwapCache(page));[...]
Which means that another cpu changed the pageflags concurrently. In fact, looking at the dump a different cpu is indeed busy with running kswapd.
Hmm, maybe I am missing something but it really looks like we can race here. Reclaim path takes the page lock while
zap_ptetakes page table lock so nothing prevents them from racing here:shrink_page_listzap_pte_range
trylock_pagepte_offset_map_lock
add_to_swappage_remove_rmap/* Page can be still mapped */ add_to_swap_cache atomic_add_negative(_mapcount) __add_to_swap_cache mem_cgroup_uncharge_page (PageSwapCache(page)) && return SetPageSwapCache __mem_cgroup_uncharge_common VM_BUG_ON(PageSwapCache(page))
Quoted 7 lines
Maybe not many people run with
CONFIG_DEBUG_VMenabled these days so we do not this more often (even me testing configs are not consistent in that regards and only few have it on). The only thing that changed in this area recently is0c59b89cwhich made the testVM_BUG_ONrather then simple return in 3.6
And maybe theBUG_ONis too harsh as CgroupUsed should guarantee that the uncharge will eventually go away. What do you think Johannes?
Interesting. We need to ensure there is ordering between setting PG_swapcache and installing swap entries because I think we are the only ones looking at PG_swapcache without the page lock held. So we don't have a safe way to check for PG_swapcache but if we get it wrong, we may steal an uncharge that uncharge_swapcache() should be doing instead and that means we mess up the swap statistics accounting.
So how can we, without holding the page lock, either safely back off from a page in swapcache or make sure we do the swap statistics accounting when uncharging a swapcache page from the final unmap?
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to [email protected]. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"[email protected]"> [email protected] </a>
Re: [v3.9-rc8]: kernel BUG at mm/memcontrol.c:3994! (was: Re: [BUG][s390x] mm: system crashed)
Quoted 97 lines · Johannes Weiner
Quoted 80 lines · Michal Hocko
Quoted 56 lines · Heiko Carstens
Quoted 2 lines · Heiko Carstens
Ok, thanks for verifying! I'll look into it; hopefully I can reproduce it here as well.
That seems to be a common code bug. I can easily trigger the
VM_BUG_ON()below (when I force the system to swap):[ 48.347963] ------------[ cut here ]------------ [ 48.347972] kernel BUG at mm/memcontrol.c:3994! [ 48.348012] illegal operation: 0001 [#1] SMP [ 48.348015] Modules linked in: [ 48.348017] CPU: 1 Not tainted 3.9.0-rc8+ #38 [ 48.348020] Process mmap2 (pid: 635, task: 0000000029476100, ksp: 000000002e91b938) [ 48.348022] Krnl PSW : 0704f00180000000 000000000026552c (__mem_cgroup_uncharge_common+0x2c4/0x33c) [ 48.348032] R:0 T:1 IO:1 EX:1 Key:0 M:1 W:0 P:0 AS:3 CC:3 PM:0 EA:3 Krnl GPRS: 0000000000000008 0000000000000009 000003d1002a9200 0000000000000000 [ 48.348039] 0000000000000000 00000000006812d8 000003ffdf339000 00000000321a6f98 [ 48.348043] 000003fffce11000 0000000000000000 0000000000000001 000003d1002a9200 [ 48.348046] 0000000000000001 0000000000681b88 000000002e91bc18 000000002e91bbd0 [ 48.348057] Krnl Code: 000000000026551e: c0e5fffaa2a1 brasl %r14,1b9a60 0000000000265524: a7f4ff7d brc 15,26541e #0000000000265528: a7f40001 brc 15,26552a >000000000026552c: e3c0b8200124 stg %r12,6176(%r11) 0000000000265532: a7f4ff57 brc 15,2653e0 0000000000265536: e310b8280104 lg %r1,6184(%r11) 000000000026553c: a71b0001 aghi %r1,1 0000000000265540: e310b8280124 stg %r1,6184(%r11) [ 48.348099] Call Trace: [ 48.348100] ([<000003d1002a91c0>] 0x3d1002a91c0) [ 48.348102] [<00000000002404aa>] page_remove_rmap+0xf2/0x16c [ 48.348106] [<0000000000232dc8>] unmap_single_vma+0x494/0x7d8 [ 48.348107] [<0000000000233ac0>] unmap_vmas+0x50/0x74 [ 48.348109] [<00000000002396ec>] unmap_region+0x9c/0x110 [ 48.348110] [<000000000023bd18>] do_munmap+0x284/0x470 [ 48.348111] [<000000000023bf56>] vm_munmap+0x52/0x70 [ 48.348113] [<000000000023cf32>] SyS_munmap+0x3a/0x4c [ 48.348114] [<0000000000665e14>] sysc_noemu+0x22/0x28 [ 48.348118] [<000003fffcf187b2>] 0x3fffcf187b2 [ 48.348119] Last Breaking-Event-Address: [ 48.348120] [<0000000000265528>] __mem_cgroup_uncharge_common+0x2c0/0x33cLooking at the code, the code flow is:
page_remove_rmap()->mem_cgroup_uncharge_page()->__mem_cgroup_uncharge_common()Note that in
mem_cgroup_uncharge_page()the page in question passed the check:[...]
if (PageSwapCache(page)) return;[...]
and just a couple of instructions later the
VM_BUG_ON()within__mem_cgroup_uncharge_common()triggers:[...]
Quoted 8 lines
if (mem_cgroup_disabled()) return NULL; VM_BUG_ON(PageSwapCache(page));[...]
Which means that another cpu changed the pageflags concurrently. In fact, looking at the dump a different cpu is indeed busy with running kswapd.
Hmm, maybe I am missing something but it really looks like we can race here. Reclaim path takes the page lock while
zap_ptetakes page table lock so nothing prevents them from racing here:shrink_page_listzap_pte_range
trylock_pagepte_offset_map_lock
add_to_swappage_remove_rmap/* Page can be still mapped */ add_to_swap_cache atomic_add_negative(_mapcount) __add_to_swap_cache mem_cgroup_uncharge_page (PageSwapCache(page)) && return SetPageSwapCache __mem_cgroup_uncharge_common VM_BUG_ON(PageSwapCache(page))Quoted 7 lines
Maybe not many people run with
CONFIG_DEBUG_VMenabled these days so we do not this more often (even me testing configs are not consistent in that regards and only few have it on). The only thing that changed in this area recently is0c59b89cwhich made the testVM_BUG_ONrather then simple return in 3.6
And maybe theBUG_ONis too harsh as CgroupUsed should guarantee that the uncharge will eventually go away. What do you think Johannes?Interesting. We need to ensure there is ordering between setting
PG_swapcacheand installing swap entries because I think we are the only ones looking atPG_swapcachewithout the page lock held. So we don't have a safe way to check forPG_swapcachebut if we get it wrong, we may steal an uncharge thatuncharge_swapcache()should be doing instead and that means we mess up the swap statistics accounting.
Quoted 3 lines
So how can we, without holding the page lock, either safely back off from a page in swapcache or make sure we do the swap statistics accounting when uncharging a swapcache page from the final unmap?
Awkward.
I agree that the actual memcg uncharging should be okay, but the memsw swap stats will go wrong (doesn't matter toooo much), and mem_cgroup_put get missed (leaking a struct mem_cgroup).
I confess to having seen this myself, just once, on x86_64 3.8-rc6-mm1 plus some patches I was testing; but I never got back to look into it.
I wonder if something gone into 3.9 has changed the timing to make it more easily reproduced now; but the underlying problem has been there a long time - long before your VM_BUG_ON was there to expose it.
For now I guess the best is just to remove the VM_BUG_ON, or restore it to a (usually redundant) return NULL, and live with the old leak.
But the real fix... I fear it may involve adding another PageCgroup flag, just for memsw, to ensure that its counts are kept in balance even when SwapCache races cause what's been charged in one way to get uncharged in an unexpected way.
All those PageSwapCache and page_mapped tests are suspect: I expect most will turn out to be okay, one way or another, but they are suspicious and deserve audit.
(But I won't be working on this myself, sorry.)
Hugh
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to [email protected]. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"[email protected]"> [email protected] </a>
Re: [v3.9-rc8]: kernel BUG at mm/memcontrol.c:3994! (was: Re: [BUG][s390x] mm: system crashed)
Quoted 102 lines · Hugh Dickins
Quoted 97 lines · Johannes Weiner
Quoted 80 lines · Michal Hocko
Quoted 56 lines · Heiko Carstens
Quoted 2 lines · Heiko Carstens
Ok, thanks for verifying! I'll look into it; hopefully I can reproduce it here as well.
That seems to be a common code bug. I can easily trigger the
VM_BUG_ON()below (when I force the system to swap):[ 48.347963] ------------[ cut here ]------------ [ 48.347972] kernel BUG at mm/memcontrol.c:3994! [ 48.348012] illegal operation: 0001 [#1] SMP [ 48.348015] Modules linked in: [ 48.348017] CPU: 1 Not tainted 3.9.0-rc8+ #38 [ 48.348020] Process mmap2 (pid: 635, task: 0000000029476100, ksp: 000000002e91b938) [ 48.348022] Krnl PSW : 0704f00180000000 000000000026552c (__mem_cgroup_uncharge_common+0x2c4/0x33c) [ 48.348032] R:0 T:1 IO:1 EX:1 Key:0 M:1 W:0 P:0 AS:3 CC:3 PM:0 EA:3 Krnl GPRS: 0000000000000008 0000000000000009 000003d1002a9200 0000000000000000 [ 48.348039] 0000000000000000 00000000006812d8 000003ffdf339000 00000000321a6f98 [ 48.348043] 000003fffce11000 0000000000000000 0000000000000001 000003d1002a9200 [ 48.348046] 0000000000000001 0000000000681b88 000000002e91bc18 000000002e91bbd0 [ 48.348057] Krnl Code: 000000000026551e: c0e5fffaa2a1 brasl %r14,1b9a60 0000000000265524: a7f4ff7d brc 15,26541e #0000000000265528: a7f40001 brc 15,26552a >000000000026552c: e3c0b8200124 stg %r12,6176(%r11) 0000000000265532: a7f4ff57 brc 15,2653e0 0000000000265536: e310b8280104 lg %r1,6184(%r11) 000000000026553c: a71b0001 aghi %r1,1 0000000000265540: e310b8280124 stg %r1,6184(%r11) [ 48.348099] Call Trace: [ 48.348100] ([<000003d1002a91c0>] 0x3d1002a91c0) [ 48.348102] [<00000000002404aa>] page_remove_rmap+0xf2/0x16c [ 48.348106] [<0000000000232dc8>] unmap_single_vma+0x494/0x7d8 [ 48.348107] [<0000000000233ac0>] unmap_vmas+0x50/0x74 [ 48.348109] [<00000000002396ec>] unmap_region+0x9c/0x110 [ 48.348110] [<000000000023bd18>] do_munmap+0x284/0x470 [ 48.348111] [<000000000023bf56>] vm_munmap+0x52/0x70 [ 48.348113] [<000000000023cf32>] SyS_munmap+0x3a/0x4c [ 48.348114] [<0000000000665e14>] sysc_noemu+0x22/0x28 [ 48.348118] [<000003fffcf187b2>] 0x3fffcf187b2 [ 48.348119] Last Breaking-Event-Address: [ 48.348120] [<0000000000265528>] __mem_cgroup_uncharge_common+0x2c0/0x33cLooking at the code, the code flow is:
page_remove_rmap()->mem_cgroup_uncharge_page()->__mem_cgroup_uncharge_common()Note that in
mem_cgroup_uncharge_page()the page in question passed the check:[...]
if (PageSwapCache(page)) return;[...]
and just a couple of instructions later the
VM_BUG_ON()within__mem_cgroup_uncharge_common()triggers:[...]
Quoted 8 lines
if (mem_cgroup_disabled()) return NULL; VM_BUG_ON(PageSwapCache(page));[...]
Which means that another cpu changed the pageflags concurrently. In fact, looking at the dump a different cpu is indeed busy with running kswapd.
Hmm, maybe I am missing something but it really looks like we can race here. Reclaim path takes the page lock while
zap_ptetakes page table lock so nothing prevents them from racing here:shrink_page_listzap_pte_range
trylock_pagepte_offset_map_lock
add_to_swappage_remove_rmap/* Page can be still mapped */ add_to_swap_cache atomic_add_negative(_mapcount) __add_to_swap_cache mem_cgroup_uncharge_page (PageSwapCache(page)) && return SetPageSwapCache __mem_cgroup_uncharge_common VM_BUG_ON(PageSwapCache(page))Quoted 7 lines
Maybe not many people run with
CONFIG_DEBUG_VMenabled these days so we do not this more often (even me testing configs are not consistent in that regards and only few have it on). The only thing that changed in this area recently is0c59b89cwhich made the testVM_BUG_ONrather then simple return in 3.6
And maybe theBUG_ONis too harsh as CgroupUsed should guarantee that the uncharge will eventually go away. What do you think Johannes?Interesting. We need to ensure there is ordering between setting
PG_swapcacheand installing swap entries because I think we are the only ones looking atPG_swapcachewithout the page lock held. So we don't have a safe way to check forPG_swapcachebut if we get it wrong, we may steal an uncharge thatuncharge_swapcache()should be doing instead and that means we mess up the swap statistics accounting.Quoted 3 lines
So how can we, without holding the page lock, either safely back off from a page in swapcache or make sure we do the swap statistics accounting when uncharging a swapcache page from the final unmap?
Quoted 5 lines
Awkward.
I agree that the actual memcg uncharging should be okay, but the memsw swap stats will go wrong (doesn't matter toooo much), and
mem_cgroup_putget missed (leaking a structmem_cgroup).
Ok, so I just went over this again. For the swapout path the memsw uncharge is deferred, but if we "steal" this uncharge from the swap code, we actually do uncharge memsw in mem_cgroup_do_uncharge(), so we may prematurely unaccount the swap page, but we never leak a charge. Good.
Because of this stealing, we also don't do the following:
if (do_swap_account && ctype == MEM_CGROUP_CHARGE_TYPE_SWAPOUT) {
mem_cgroup_swap_statistics(memcg, true);
mem_cgroup_get(memcg);
}I.e. it does not matter that mem_cgroup_uncharge_swap() doesn't do the put, we are also not doing the get. We should not leak references.
So the only thing that I can see go wrong is that we may have a swapped out page that is not charged to memsw and not accounted as MEM_CGROUP_STAT_SWAP. But I don't know how likely that is, because we check for PG_swapcache in this uncharge path after the last pte is torn down, so even though the page is put on swap cache, it probably won't be swapped. It would require that the PG_swapcache setting would become visible only after the page has been added to the swap cache AND rmap has established at least one swap pte for us to uncharge a page that actually continues to be used. And that's a bit of a stretch, I think.
Did I miss something? If not, I'll just send a patch that removes the VM_BUG_ON() and adds a comment describing the scenarios and a note that we may want to fix this in the future.
Thanks!
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to [email protected]. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"[email protected]"> [email protected] </a>
Re: [v3.9-rc8]: kernel BUG at mm/memcontrol.c:3994! (was: Re: [BUG][s390x] mm: system crashed)
Quoted 37 lines · Johannes Weiner
Quoted 13 lines · Hugh Dickins
Quoted 12 lines · Johannes Weiner
Quoted 11 lines · Michal Hocko
Quoted 10 lines · Heiko Carstens
On Thu, Apr 18, 2013 at 09:13:03AM +0200, Heiko Carstens wrote:
[ 48.347963] ------------[ cut here ]------------ [ 48.347972] kernel BUG at mm/memcontrol.c:3994!
__mem_cgroup_uncharge_common()triggers:[...]
if (mem_cgroup_disabled()) return NULL;Quoted 6 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
VM_BUG_ON(PageSwapCache(page));[...]
I agree that the actual memcg uncharging should be okay, but the memsw swap stats will go wrong (doesn't matter toooo much), and
mem_cgroup_putget missed (leaking a structmem_cgroup).Ok, so I just went over this again. For the swapout path the memsw uncharge is deferred, but if we "steal" this uncharge from the swap code, we actually do uncharge memsw in
mem_cgroup_do_uncharge(), so we may prematurely unaccount the swap page, but we never leak a charge. Good.Because of this stealing, we also don't do the following:
if (do_swap_account && ctype == MEM_CGROUP_CHARGE_TYPE_SWAPOUT) { mem_cgroup_swap_statistics(memcg, true); mem_cgroup_get(memcg); }I.e. it does not matter that
mem_cgroup_uncharge_swap()doesn't do the put, we are also not doing the get. We should not leak references.
Quoted 10 lines
So the only thing that I can see go wrong is that we may have a swapped out page that is not charged to memsw and not accounted as
MEM_CGROUP_STAT_SWAP. But I don't know how likely that is, because we check forPG_swapcachein this uncharge path after the last pte is torn down, so even though the page is put on swap cache, it probably won't be swapped. It would require that thePG_swapcachesetting would become visible only after the page has been added to the swap cache AND rmap has established at least one swap pte for us to uncharge a page that actually continues to be used. And that's a bit of a stretch, I think.
Sorry, our minds seem to work in different ways, I understood very little of what you wrote above :-(
But once I try to disprove you with a counter-example, I seem to arrive at the same conclusion as you have (well, I haven't quite arrived there yet, but cannot give it any more time).
Looking at it from my point of view, I concentrate on the racy
if (PageSwapCache(page)) return; __mem_cgroup_uncharge_common(page, MEM_CGROUP_CHARGE_TYPE_ANON, false);
in mem_cgroup_uncharge_page().
Now, that may or may not catch the case where last reference to page is unmapped at the same time as the page is added to swap: but being a MEM_CGROUP_CHARGE_TYPE_ANON call, it does not interfere with the memsw stats and get/put at all, those remain in balance.
And mem_cgroup_uncharge_swap() has all along been prepared to get a zero id from swap_cgroup_record(), if a SwapCache page should be uncharged when it was never quite charged as such.
Yes, we may occasionally fail to charge a SwapCache page as such if its final unmap from userspace races with its being added to swap; but it's heading towards swap_writepage()'s try_to_free_swap() anyway, so I don't think that's anything to worry about.
(If I had time to stop and read through that, I'd probably find it just as hard to understand as what you wrote!)
Quoted 4 lines
Did I miss something? If not, I'll just send a patch that removes the
VM_BUG_ON()and adds a comment describing the scenarios and a note that we may want to fix this in the future.
I don't think you missed something. Yes, please just send Linus and Andrew a patch to remove the VM_BUG_ON() (with Cc stable tag), I now agree that's all that's really needed - thanks.
Hugh
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to [email protected]. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"[email protected]"> [email protected] </a>
Re: [v3.9-rc8]: kernel BUG at mm/memcontrol.c:3994! (was: Re: [BUG][s390x] mm: system crashed)
Quoted 49 lines · Hugh Dickins
Quoted 37 lines · Johannes Weiner
Quoted 13 lines · Hugh Dickins
Quoted 12 lines · Johannes Weiner
Quoted 11 lines · Michal Hocko
Quoted 10 lines · Heiko Carstens
On Thu, Apr 18, 2013 at 09:13:03AM +0200, Heiko Carstens wrote:
[ 48.347963] ------------[ cut here ]------------ [ 48.347972] kernel BUG at mm/memcontrol.c:3994!
__mem_cgroup_uncharge_common()triggers:[...]
if (mem_cgroup_disabled()) return NULL;Quoted 6 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
VM_BUG_ON(PageSwapCache(page));[...]
I agree that the actual memcg uncharging should be okay, but the memsw swap stats will go wrong (doesn't matter toooo much), and
mem_cgroup_putget missed (leaking a structmem_cgroup).Ok, so I just went over this again. For the swapout path the memsw uncharge is deferred, but if we "steal" this uncharge from the swap code, we actually do uncharge memsw in
mem_cgroup_do_uncharge(), so we may prematurely unaccount the swap page, but we never leak a charge. Good.Because of this stealing, we also don't do the following:
if (do_swap_account && ctype == MEM_CGROUP_CHARGE_TYPE_SWAPOUT) { mem_cgroup_swap_statistics(memcg, true); mem_cgroup_get(memcg); }I.e. it does not matter that
mem_cgroup_uncharge_swap()doesn't do the put, we are also not doing the get. We should not leak references.Quoted 10 lines
So the only thing that I can see go wrong is that we may have a swapped out page that is not charged to memsw and not accounted as
MEM_CGROUP_STAT_SWAP. But I don't know how likely that is, because we check forPG_swapcachein this uncharge path after the last pte is torn down, so even though the page is put on swap cache, it probably won't be swapped. It would require that thePG_swapcachesetting would become visible only after the page has been added to the swap cache AND rmap has established at least one swap pte for us to uncharge a page that actually continues to be used. And that's a bit of a stretch, I think.
Quoted 6 lines
Sorry, our minds seem to work in different ways, I understood very little of what you wrote above :-(
But once I try to disprove you with a counter-example, I seem to arrive at the same conclusion as you have (well, I haven't quite arrived there yet, but cannot give it any more time).
I might be losing my mind. But since you are reaching the same conclusion, and I see the same mental milestones in your thought process described below, it's more likely that I suck at describing my train of thought coherently. Or the third possibility: we're both losing it!
Quoted 10 lines
Looking at it from my point of view, I concentrate on the racy
if (PageSwapCache(page)) return; __mem_cgroup_uncharge_common(page, MEM_CGROUP_CHARGE_TYPE_ANON, false);in
mem_cgroup_uncharge_page().Now, that may or may not catch the case where last reference to page is unmapped at the same time as the page is added to swap: but being a
MEM_CGROUP_CHARGE_TYPE_ANONcall, it does not interfere with the memsw stats and get/put at all, those remain in balance.
Yes, exactly.
Quoted 8 lines
And
mem_cgroup_uncharge_swap()has all along been prepared to get a zero id fromswap_cgroup_record(), if a SwapCache page should be uncharged when it was never quite charged as such.Yes, we may occasionally fail to charge a SwapCache page as such if its final unmap from userspace races with its being added to swap; but it's heading towards swap_writepage()'s
try_to_free_swap()anyway, so I don't think that's anything to worry about.
Agreed as well. If there are no pte references to the swap slot, it will be freed either way. I didn't even think of the try_to_free_swap() in the writeout call, but was looking at the __remove_mapping later on in reclaim that will do a swapcache_free().
The only case I was worried about is the following:
#0 #1page_remove_rmap() shrink_page_list()
if --page->mapcount == 0: add_to_swap()
mem_cgroup_uncharge_page() __add_to_swap_cache()
if PageSwapCache: SetPageSwapCache()
return try_to_unmap()
__mem_cgroup_uncharge_common() for each pte:
install swp_entry_t page->mapcount--
Looking at #1, I don't see anything that would force concurrent threads to observe SetSwapCache ordered against the page->mapcount--. My concern was that if those get reordered, #0 may see page->mapcount == 1 AND !PageSwapcache, and then go ahead and uncharge the page while there is actually a swp_entry_t pointing to it. The page will be a proper long-term swap page without being charged as such.
Quoted 11 lines
(If I had time to stop and read through that, I'd probably find it just as hard to understand as what you wrote!)
Quoted 4 lines
Did I miss something? If not, I'll just send a patch that removes the
VM_BUG_ON()and adds a comment describing the scenarios and a note that we may want to fix this in the future.I don't think you missed something. Yes, please just send Linus and Andrew a patch to remove the
VM_BUG_ON()(with Cc stable tag), I now agree that's all that's really needed - thanks.
Will do, thanks for taking them time to think through it again, even after failing to decipher my ramblings...
Johannes
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to [email protected]. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"[email protected]"> [email protected] </a>
Re: [v3.9-rc8]: kernel BUG at mm/memcontrol.c:3994! (was: Re: [BUG][s390x] mm: system crashed)
Quoted 95 lines · Johannes Weiner
Quoted 11 lines · Hugh Dickins
Quoted 10 lines · Johannes Weiner
Quoted 9 lines · Hugh Dickins
Quoted 8 lines · Johannes Weiner
Quoted 7 lines · Michal Hocko
Quoted 6 lines · Heiko Carstens
On Thu, Apr 18, 2013 at 09:13:03AM +0200, Heiko Carstens wrote:
[ 48.347963] ------------[ cut here ]------------ [ 48.347972] kernel BUG at mm/memcontrol.c:3994!
__mem_cgroup_uncharge_common()triggers:Quoted 8 lines
Quoted 8 lines
Quoted 8 lines
Quoted 6 lines
Quoted 6 lines
Quoted 6 lines
[...]
if (mem_cgroup_disabled()) return NULL; VM_BUG_ON(PageSwapCache(page));[...]
I agree that the actual memcg uncharging should be okay, but the
memsw
Quoted 1 line
Quoted 1 line
Quoted 1 line
swap stats will go wrong (doesn't matter toooo much), and
mem_cgroup_putQuoted 29 lines
Quoted 18 lines
Quoted 1 line
get missed (leaking a struct
mem_cgroup).Ok, so I just went over this again. For the swapout path the memsw uncharge is deferred, but if we "steal" this uncharge from the swap code, we actually do uncharge memsw in
mem_cgroup_do_uncharge(), so we may prematurely unaccount the swap page, but we never leak a charge. Good.Because of this stealing, we also don't do the following:
if (do_swap_account && ctype == MEM_CGROUP_CHARGE_TYPE_SWAPOUT) { mem_cgroup_swap_statistics(memcg, true); mem_cgroup_get(memcg); }I.e. it does not matter that
mem_cgroup_uncharge_swap()doesn't do the put, we are also not doing the get. We should not leak references.Quoted 10 lines
So the only thing that I can see go wrong is that we may have a swapped out page that is not charged to memsw and not accounted as
MEM_CGROUP_STAT_SWAP. But I don't know how likely that is, because we check forPG_swapcachein this uncharge path after the last pte is torn down, so even though the page is put on swap cache, it probably won't be swapped. It would require that thePG_swapcachesetting would become visible only after the page has been added to the swap cache AND rmap has established at least one swap pte for us to uncharge a page that actually continues to be used. And that's a bit of a stretch, I think.Quoted 6 lines
Sorry, our minds seem to work in different ways, I understood very little of what you wrote above :-(
But once I try to disprove you with a counter-example, I seem to arrive at the same conclusion as you have (well, I haven't quite arrived there yet, but cannot give it any more time).
I might be losing my mind. But since you are reaching the same conclusion, and I see the same mental milestones in your thought process described below, it's more likely that I suck at describing my train of thought coherently. Or the third possibility: we're both losing it!
Quoted 4 lines
Looking at it from my point of view, I concentrate on the racy
if (PageSwapCache(page)) return; __mem_cgroup_uncharge_common(page, MEM_CGROUP_CHARGE_TYPE_ANON,false);
Quoted 6 lines
in
mem_cgroup_uncharge_page().Now, that may or may not catch the case where last reference to page is unmapped at the same time as the page is added to swap: but being a
MEM_CGROUP_CHARGE_TYPE_ANONcall, it does not interfere with the memsw stats and get/put at all, those remain in balance.Yes, exactly.
Quoted 8 lines
And
mem_cgroup_uncharge_swap()has all along been prepared to get a zero id fromswap_cgroup_record(), if a SwapCache page should be uncharged when it was never quite charged as such.Yes, we may occasionally fail to charge a SwapCache page as such if its final unmap from userspace races with its being added to swap; but it's heading towards swap_writepage()'s
try_to_free_swap()anyway, so I don't think that's anything to worry about.Agreed as well. If there are no pte references to the swap slot, it will be freed either way. I didn't even think of the
try_to_free_swap()in the writeout call, but was looking at the__remove_mappinglater on in reclaim that will do aswapcache_free().The only case I was worried about is the following:
Quoted 10 lines
#0 #1
page_remove_rmap()shrink_page_list()
if --page->mapcount == 0:add_to_swap()
mem_cgroup_uncharge_page()__add_to_swap_cache()
if PageSwapCache:SetPageSwapCache()
returntry_to_unmap()
__mem_cgroup_uncharge_common()for each pte:
installswp_entry_tpage->mapcount--
Thanks for spelling it out for me in more detail, this time I think I do grasp your concern.
Quoted 8 lines
Looking at #1, I don't see anything that would force concurrent threads to observe SetSwapCache ordered against the page->mapcount--. My concern was that if those get reordered, #0 may see
page->mapcount== 1 AND !PageSwapcache, and then go ahead and uncharge the page while there is actually aswp_entry_tpointing to it. The page will be a proper long-term swap page without being charged as such.
But I don't see any problem with ordering here. #0 is using an atomic operation which returns a result on page->mapcount, so that amounts to (more than) an smp_rmb ensuring it reads mapcount before reading PageSwapCache flag. And in #1, there's at least an unlock of the radix_tree lock (after adding to swap tree) and a lock of the page table lock (before unmapping the page), and that pairing amounts to (more than) an smp_wmb.
Hugh
Quoted 17 lines
Quoted 11 lines
(If I had time to stop and read through that, I'd probably find it just as hard to understand as what you wrote!)
Quoted 4 lines
Did I miss something? If not, I'll just send a patch that removes the
VM_BUG_ON()and adds a comment describing the scenarios and a note that we may want to fix this in the future.I don't think you missed something. Yes, please just send Linus and Andrew a patch to remove the
VM_BUG_ON()(with Cc stable tag), I now agree that's all that's really needed - thanks.Will do, thanks for taking them time to think through it again, even after failing to decipher my ramblings...
Johannes