Messages in this thread | | | Subject | Re: [PATCH 4.18 000/123] 4.18.6-stable review | From | Guenter Roeck <> | Date | Wed, 5 Sep 2018 08:34:23 -0700 |
| |
On 09/05/2018 02:01 AM, Greg Kroah-Hartman wrote: > On Tue, Sep 04, 2018 at 09:24:34AM -0700, Guenter Roeck wrote: >> On Mon, Sep 03, 2018 at 06:55:44PM +0200, Greg Kroah-Hartman wrote: >>> This is the start of the stable review cycle for the 4.18.6 release. >>> There are 123 patches in this series, all will be posted as a response >>> to this one. If anyone has any issues with these being applied, please >>> let me know. >>> >>> Responses should be made by Wed Sep 5 16:56:53 UTC 2018. >>> Anything received after that time might be too late. >>> >> >> Not directly related to v4.18.6-rc1. I have seen the following hang >> several times with v4.18.5. It happens on a quite regular basis after >> a suspend-resume cycle. CPU is Ryzen 1700X. >> >> Guenter >> >> --- >> [ 9990.754641] watchdog: BUG: soft lockup - CPU#5 stuck for 22s! [kworker/5:1:155] >> [ 9990.762549] Modules linked in: ipt_REJECT nf_reject_ipv4 xt_multiport sp5100_tco squashfs iptable_filter snd_hda_codec_hdmi binfmt_misc edac_mce_amd kvm snd_hda_codec_realtek irqbypass snd_hda_codec_generic snd_seq_midi snd_seq_midi_event crct10dif_pclmul ghash_clmulni_intel snd_rawmidi aesni_intel snd_hda_intel aes_x86_64 crypto_simd cryptd glue_helper snd_hda_codec snd_hda_core wmi_bmof snd_hwdep snd_seq snd_pcm k10temp snd_seq_device snd_timer snd soundcore sch_fq_codel parport_pc sunrpc ppdev lp parport ip_tables x_tables autofs4 hid_generic nouveau mxm_wmi video ttm drm_kms_helper usbhid syscopyarea sysfillrect hid sysimgblt igb fb_sys_fops dca drm i2c_algo_bit i2c_piix4 i2c_core r8169 ahci mii libahci wmi >> [ 9990.762589] CPU: 5 PID: 155 Comm: kworker/5:1 Tainted: G L 4.18.5+ #1 >> [ 9990.762591] Hardware name: Gigabyte Technology Co., Ltd. AB350M-Gaming 3/AB350M-Gaming 3-CF, BIOS F23 08/08/2018 >> [ 9990.762596] Workqueue: events free_work >> [ 9990.762601] RIP: 0010:smp_call_function_many+0x208/0x270 >> [ 9990.762601] Code: e8 0d d1 77 00 3b 05 cb f0 24 01 0f 83 86 fe ff ff 48 63 d0 49 8b 0c 24 48 03 0c d5 00 f7 11 a7 8b 51 18 83 e2 01 74 0a f3 90 <8b> 51 18 83 e2 01 75 f6 eb c7 0f b6 4d d0 4c 89 f2 4c 89 ee 44 89 >> [ 9990.762626] RSP: 0018:ffff95ebc3effd20 EFLAGS: 00000202 ORIG_RAX: ffffffffffffff13 >> [ 9990.762628] RAX: 000000000000000c RBX: ffff94eeded63cc8 RCX: ffff94eedef27bc0 >> [ 9990.762629] RDX: 0000000000000001 RSI: 0000000000000100 RDI: ffff94eeded63cc8 >> [ 9990.762630] RBP: ffff95ebc3effd60 R08: 00000000fffffff0 R09: 00000000000000ff >> [ 9990.762631] R10: ffff94eeded63ce8 R11: ffff94eeded63cc8 R12: ffff94eeded63cc0 >> [ 9990.762632] R13: ffffffffa6076150 R14: 0000000000000000 R15: 0000000000000100 >> [ 9990.762633] FS: 0000000000000000(0000) GS:ffff94eeded40000(0000) knlGS:0000000000000000 >> [ 9990.762635] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 >> [ 9990.762636] CR2: 0000000000a67000 CR3: 00000006f120c000 CR4: 00000000003406e0 >> [ 9990.762637] Call Trace: >> [ 9990.762642] ? load_new_mm_cr3+0xe0/0xe0 >> [ 9990.762644] on_each_cpu+0x2d/0x60 >> [ 9990.762647] flush_tlb_kernel_range+0x4b/0x80 >> [ 9990.762648] ? vunmap_page_range+0x1fe/0x310 >> [ 9990.762650] __purge_vmap_area_lazy+0x50/0xb0 >> [ 9990.762652] free_vmap_area_noflush+0x7d/0x90 >> [ 9990.762654] remove_vm_area+0x74/0x80 >> [ 9990.762656] __vunmap+0x3b/0xc0 >> [ 9990.762657] free_work+0x25/0x40 >> [ 9990.762660] process_one_work+0x15e/0x3f0 >> [ 9990.762662] worker_thread+0x4a/0x440 >> [ 9990.762664] kthread+0x105/0x140 >> [ 9990.762666] ? process_one_work+0x3f0/0x3f0 >> [ 9990.762668] ? kthread_destroy_worker+0x50/0x50 >> [ 9990.762670] ret_from_fork+0x22/0x40 > > Odd. Do you see this on Linus's tree? >
Not tested, but I see it in v4.17.19 and in v4.18.6-rc2. Turns out it is related to heavy load, not to suspend/resume. At this point I suspect that it may be an AMD/Ryzen specific problem - it looks like it disappears if I add "kernel.randomize_va_space = 0" to /etc/sysctl.conf. No idea if it is a CPU bug or some AMD specific code problem. I'll try to analyze it further.
Either case, it is not a concern for the current release since it affects other kernel versions.
Guenter
| |