| View previous topic :: View next topic |
| Author |
Message |
Skinjob2707
n00b
Joined: 07 Aug 2013
Posts: 57
|
 Posted: Thu Jun 22, 2017 6:13 pm Post subject: Root processes fail to end |
 |
|
I am having a problem where I'm not sure the right place to being troubleshooting. After running for awhile (at least 45 minutes to overnight) root processes fail to close. If I run emerge --newuse --update --ask @world, the first item in the list will be installed, but emerge will stall at the end and never move unto to the next package in the list. If I run gvim as root, click the x in the upper right hand corner to close it, it will sit a little while before the process will not end dialogue comes up. If I then click terminate, the gvim window has everything grayed out, but will not go away without a reboot.
The output from ps aux | grep gvim reads:
| Code: | | root 9967 0.0 0.1 172616 24908 ? Ds 13:57 0:00 gvim nvidia-cudnn-bin-8.0-r6.ebuild |
.
The only way to get it to go away is to reboot. On restarting, the shutdown process often hangs because there is an open file on either the /boot or the root partition.
When I look in journalctl --pager-end I don't see any error messages that look relevant.
My current Kernel config is at:
https://pastebin.com/VEMNFAvv
Thanks in advance for your assistance! |
|
| Back to top |
|
 |
Skinjob2707
n00b
Joined: 07 Aug 2013
Posts: 57
|
| Posted: Thu Jun 22, 2017 6:23 pm Post subject: |
|
|
Looking through the logs, I see the following, but it isn't clear whether it is the cause of the problem:
| Code: | Jun 22 13:14:51 bluemeanie kernel: ------------[ cut here ]------------
Jun 22 13:14:51 bluemeanie kernel: kernel BUG at fs/f2fs/gc.c:899!
Jun 22 13:14:51 bluemeanie kernel: invalid opcode: 0000 [#1] PREEMPT SMP
Jun 22 13:14:51 bluemeanie kernel: Modules linked in: nvidia_drm(PO) arc4 ath9k ath9k_common ath9k_hw mac80211 input_le
Jun 22 13:14:51 bluemeanie kernel: CPU: 0 PID: 1044 Comm: f2fs_gc-8:3 Tainted: P O 4.11.6-gentoo #1
Jun 22 13:14:51 bluemeanie kernel: Hardware name: Micro-Star International Co., Ltd MS-7A34/B350 PC MATE(MS-7A34), BIOS
Jun 22 13:14:51 bluemeanie kernel: task: ffff88040de8cf40 task.stack: ffffc90000468000
Jun 22 13:14:51 bluemeanie kernel: RIP: 0010:do_garbage_collect+0x9e1/0xb00
Jun 22 13:14:51 bluemeanie kernel: RSP: 0018:ffffc9000046bcb0 EFLAGS: 00010297
Jun 22 13:14:51 bluemeanie kernel: RAX: ffff8801944aa000 RBX: 0000000000000000 RCX: 0000000000000000
Jun 22 13:14:51 bluemeanie kernel: RDX: ffff880000000000 RSI: 0000000000000003 RDI: ffffea0005870530
Jun 22 13:14:51 bluemeanie kernel: RBP: ffffc9000046bdb0 R08: ffff880207505b90 R09: ffffea000587054c
Jun 22 13:14:51 bluemeanie kernel: R10: ffffc9000046bc38 R11: 0000000000000040 R12: 0000000000000006
Jun 22 13:14:51 bluemeanie kernel: R13: ffff88040d033708 R14: ffff88040dc0e800 R15: ffffea0005870530
Jun 22 13:14:51 bluemeanie kernel: FS: 0000000000000000(0000) GS:ffff88041ec00000(0000) knlGS:0000000000000000
Jun 22 13:14:51 bluemeanie kernel: CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
Jun 22 13:14:51 bluemeanie kernel: CR2: 00007ffac2da9010 CR3: 00000001a9e2c000 CR4: 00000000003406f0
Jun 22 13:14:51 bluemeanie kernel: Call Trace:
Jun 22 13:14:51 bluemeanie kernel: ? _raw_spin_lock+0x12/0x40
Jun 22 13:14:51 bluemeanie kernel: ? pick_next_task_fair+0x585/0x9c0
Jun 22 13:14:51 bluemeanie kernel: ? find_next_bit+0xb/0x10
Jun 22 13:14:51 bluemeanie kernel: f2fs_gc+0x19f/0x470
Jun 22 13:14:51 bluemeanie kernel: ? f2fs_gc+0x19f/0x470
Jun 22 13:14:51 bluemeanie kernel: ? del_timer_sync+0x30/0x50
Jun 22 13:14:51 bluemeanie kernel: ? preempt_count_add+0xa3/0xc0
Jun 22 13:14:51 bluemeanie kernel: gc_thread_func+0x2eb/0x340
Jun 22 13:14:51 bluemeanie kernel: ? gc_thread_func+0x2eb/0x340
Jun 22 13:14:51 bluemeanie kernel: ? wake_atomic_t_function+0x50/0x50
Jun 22 13:14:51 bluemeanie kernel: kthread+0xff/0x140
Jun 22 13:14:51 bluemeanie kernel: ? f2fs_gc+0x470/0x470
Jun 22 13:14:51 bluemeanie kernel: ? kthread_create_on_node+0x40/0x40
Jun 22 13:14:51 bluemeanie kernel: ? umh_complete+0x40/0x40
Jun 22 13:14:51 bluemeanie kernel: ? call_usermodehelper_exec_async+0x137/0x140
Jun 22 13:14:51 bluemeanie kernel: ret_from_fork+0x29/0x40
Jun 22 13:14:51 bluemeanie kernel: Code: ff e9 d3 fd ff ff 8b 55 8c 44 89 f1 4c 89 ef e8 b6 ee ff ff e9 78 fe ff ff 8b
Jun 22 13:14:51 bluemeanie kernel: RIP: do_garbage_collect+0x9e1/0xb00 RSP: ffffc9000046bcb0
Jun 22 13:14:51 bluemeanie kernel: ---[ end trace 71b180caf0c5dabb ]---
|
|
|
| Back to top |
|
|
Hu
Administrator
Joined: 06 Mar 2007
Posts: 23062
|
| Posted: Fri Jun 23, 2017 1:34 am Post subject: |
|
|
This is: | bdac9683a6553b04e7456ed959162fe6bc696dde:fs/f2fs/gc.c: | 860 static int do_garbage_collect(struct f2fs_sb_info *sbi,
865 struct f2fs_summary_block *sum;
870 unsigned char type = IS_DATASEG(get_seg_entry(sbi, segno)->type) ?
871 SUM_TYPE_DATA : SUM_TYPE_NODE;
898 sum = page_address(sum_page);
899 f2fs_bug_on(sbi, type != GET_SUM_TYPE((&sum->footer))); |
Normally, I would suggest that you try to reproduce the problem with an untainted kernel. I doubt that will matter here, but it's still good practice, if for no other reason than that you are unlikely to get much support upstream for a tainted kernel. |
|
| Back to top |
|
|
Skinjob2707
n00b
Joined: 07 Aug 2013
Posts: 57
|
| Posted: Sat Jun 24, 2017 7:15 pm Post subject: |
|
|
You make a good point about the tainted kernel. Here is the same problem happening with an untainted kernel:
| Code: | [ 494.543223] kernel BUG at fs/f2fs/gc.c:899!
[ 494.543812] invalid opcode: 0000 [#1] PREEMPT SMP
[ 494.544396] Modules linked in: arc4 ath9k ath9k_common ath9k_hw mac80211 input_leds ath cfg80211 nouveau snd_hda_codec_realtek snd_hda_codec_generic video led_class i2c_algo_bit hwmon drm_kms_helper syscopyarea sysfillrect snd_hda_intel sysimgblt fb_sys_fops snd_hda_codec ttm kvm snd_hwdep snd_hda_core irqbypass snd_pcm drm snd_timer pcspkr i2c_piix4 snd wmi 8250 8250_base serial_core button acpi_cpufreq r8169 mii efivarfs
[ 494.545753] CPU: 0 PID: 1120 Comm: f2fs_gc-8:3 Not tainted 4.11.6-gentoo #1
[ 494.546424] Hardware name: Micro-Star International Co., Ltd MS-7A34/B350 PC MATE(MS-7A34), BIOS A.30 04/19/2017
[ 494.547107] task: ffff88040e48ae80 task.stack: ffffc90000418000
[ 494.547798] RIP: 0010:do_garbage_collect+0x9e1/0xb00
[ 494.548483] RSP: 0018:ffffc9000041bcb0 EFLAGS: 00010297
[ 494.549162] RAX: ffff880255dc8000 RBX: 0000000000000000 RCX: 0000000000000000
[ 494.549845] RDX: ffff880000000000 RSI: 0000000000000003 RDI: ffffea00082c83c0
[ 494.550524] RBP: ffffc9000041bdb0 R08: ffff88040767b4d8 R09: ffffea00082c83dc
[ 494.551201] R10: ffffc9000041bc38 R11: 0000000000000040 R12: 0000000000000007
[ 494.551877] R13: ffff88040e3b04c8 R14: ffff88040d1d9800 R15: ffffea00082c83c0
[ 494.552552] FS: 0000000000000000(0000) GS:ffff88041ec00000(0000) knlGS:0000000000000000
[ 494.553230] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 494.553906] CR2: 00007f41d4caf038 CR3: 0000000407be6000 CR4: 00000000003406f0
[ 494.554592] Call Trace:
[ 494.555275] ? find_next_bit+0xb/0x10
[ 494.555953] f2fs_gc+0x19f/0x470
[ 494.556626] ? f2fs_gc+0x19f/0x470
[ 494.557296] ? del_timer_sync+0x20/0x50
[ 494.557960] ? preempt_count_add+0xa3/0xc0
[ 494.558620] gc_thread_func+0x2eb/0x340
[ 494.559278] ? gc_thread_func+0x2eb/0x340
[ 494.559939] ? wake_atomic_t_function+0x50/0x50
[ 494.560603] kthread+0xff/0x140
[ 494.561258] ? f2fs_gc+0x470/0x470
[ 494.561916] ? kthread_create_on_node+0x40/0x40
[ 494.562569] ret_from_fork+0x29/0x40
[ 494.563216] Code: ff e9 d3 fd ff ff 8b 55 8c 44 89 f1 4c 89 ef e8 b6 ee ff ff e9 78 fe ff ff 8b 75 94 4c 89 ff e8 56 7f 00 00 e9 06 fc ff ff 0f 0b <0f> 0b 44 89 fe 48 89 cf e8 52 8e 00 00 e9 db f8 ff ff 0f b6 b5
[ 494.563935] RIP: do_garbage_collect+0x9e1/0xb00 RSP: ffffc9000041bcb0
[ 494.570544] ---[ end trace bb6a511b13617ddd ]--- |
The Google produced this mailing list entry with at least the same source code line in common: https://www.mail-archive.com/linux-f2fs-devel@lists.sourceforge.net/msg06152.html.
Is the f2fs-devel mail list the right place to inquire about this issue? Or should I try and apply the patch using portage's patch facility?
Thanks for your help![/code] |
|
| Back to top |
|
|
Hu
Administrator
Joined: 06 Mar 2007
Posts: 23062
|
| Posted: Sat Jun 24, 2017 9:02 pm Post subject: |
|
|
| I am not sufficiently familiar with f2fs to know whether that patch is useful. I suggest contacting the f2fs mailing list, explaining your problem, and asking their opinion on the patch. |
|
| Back to top |
|
|
Skinjob2707
n00b
Joined: 07 Aug 2013
Posts: 57
|
| Posted: Tue Jun 27, 2017 4:14 pm Post subject: |
|
|
I upgraded on Sunday morning to kernel 4.11.7 and haven't had the problem since. Unless it recurs, I'm going to consider the problem solved.
Thanks for your help! |
|
| Back to top |
|
|
|
Kernel & Hardware
