New issue
Advanced search Search tips
Note: Color blocks (like or ) mean that a user may not be available. Tooltip shows the reason.

Issue 597131 link

Starred by 5 users

Issue metadata

Status: Verified
Owner:
Closed: Apr 2016
Cc:
Components:
EstimatedDays: ----
NextAction: ----
OS: Chrome
Pri: 1
Type: Bug



Sign in to add a comment

[Haswell]: kernel-Fatal exception-381E7AFD occurs after suspend/resume

Project Member Reported by hsiangc@chromium.org, Mar 23 2016

Issue description

Google Chrome	51.0.2686.0 (Official Build) dev  (64-bit) 
Platform	8097.0.0 (Official Build) dev-channel peppy
JavaScript	V8 5.1.176
Flash	21.0.0.197 -r1

What steps will reproduce the problem?
(1)boot up system and login with account
(2)close the lid and open it
(3)issue occur, it will show login screen again

What is the expected output?
user should not need to login again

What do you see instead?
it will show login screen again

feedback:will add when it is available

Crash ID 2401666400000000 (ChromeOS)

 
Cc: abod...@chromium.org
Cc: helenzhang@chromium.org
adding feedback:
Report ID: 7692662359
Cc: -abod...@chromium.org ka...@chromium.org

Comment 5 by ka...@chromium.org, Mar 23 2016

Labels: Stability-Crash
It looks like kernel warning from the crash report at https://crash.corp.google.com/browse?stbtiq=2401666400000000

Do we have a reboot with chrome logo splash screen, OR do you see the login screen immediately?

How consistent is this observation?


<4>[    0.098164] ------------[ cut here ]------------
<4>[    0.098171] WARNING: at /mnt/host/source/src/third_party/kernel/v3.8/arch/x86/mm/ioremap.c:171 __ioremap_caller+0x2cf/0x314()
<4>[    0.098173] Hardware name: Peppy
<6>[    0.098174] Info: mapping multiple BARs. Your kernel is fine.
<5>[    0.098176] Modules linked in:
<5>[    0.098181] Pid: 1, comm: swapper/0 Not tainted 3.8.11 #1
<5>[    0.098183] Call Trace:
<5>[    0.098190]  [<ffffffffbe633173>] warn_slowpath_fmt+0x65/0x90
<5>[    0.098194]  [<ffffffffbe629cf0>] __ioremap_caller+0x2cf/0x314
<5>[    0.098198]  [<ffffffffbe629d4c>] ioremap_nocache+0x17/0x19
<5>[    0.098203]  [<ffffffffbe616707>] snb_uncore_imc_init_box+0x7c/0xb1
<5>[    0.098207]  [<ffffffffbe615055>] uncore_box_init+0x2c/0x2e
<5>[    0.098210]  [<ffffffffbe615227>] uncore_pci_probe+0x100/0x156
<5>[    0.098216]  [<ffffffffbe80480f>] pci_device_probe+0x6e/0xb0
<5>[    0.098221]  [<ffffffffbe8ef0f3>] driver_probe_device+0xa2/0x1e3
<5>[    0.098225]  [<ffffffffbe8ef2ce>] __driver_attach+0x5d/0x7f
<5>[    0.098228]  [<ffffffffbe8ef271>] ? __device_attach+0x3d/0x3d
<5>[    0.098232]  [<ffffffffbe8ee2b8>] bus_for_each_dev+0x89/0xac
<5>[    0.098236]  [<ffffffffbe8eec41>] driver_attach+0x1e/0x20
<5>[    0.098240]  [<ffffffffbe8ee889>] bus_add_driver+0x114/0x211
<5>[    0.098246]  [<ffffffffbeedaee6>] ? uncore_cpu_setup+0x13/0x13
<5>[    0.098261]  [<ffffffffbe8ef7ea>] driver_register+0x8c/0xfb
<5>[    0.098266]  [<ffffffffbeedaee6>] ? uncore_cpu_setup+0x13/0x13
<5>[    0.098271]  [<ffffffffbe803f33>] __pci_register_driver+0x60/0x63
<5>[    0.098275]  [<ffffffffbeedafa5>] intel_uncore_init+0xbf/0x305
<5>[    0.098280]  [<ffffffffbeedaee6>] ? uncore_cpu_setup+0x13/0x13
<5>[    0.098284]  [<ffffffffbe60023c>] do_one_initcall+0x7f/0x137
<5>[    0.098288]  [<ffffffffbeed3c49>] kernel_init_freeable+0x112/0x190
<5>[    0.098293]  [<ffffffffbeab545b>] ? rest_init+0x6f/0x6f
<5>[    0.098296]  [<ffffffffbeab5469>] kernel_init+0xe/0xd6
<5>[    0.098302]  [<ffffffffbeac645c>] ret_from_fork+0x7c/0xb0
<5>[    0.098306]  [<ffffffffbeab545b>] ? rest_init+0x6f/0x6f
<4>[    0.098313] ---[ end trace a49a8625d80b5931 ]---
I see the login screen after open the lid. it is 100% reproducible on peppy

Comment 7 by ka...@chromium.org, Mar 23 2016

Can you check "Require password to wake from sleep" setting under chrome://settings ?
Screenshot - https://screenshot.googleplex.com/ADVKFHzbBO0

Do you see a new crash report under chrome://crashes every time after you open the lid and login again? How many crashes are indicated on this page?
"Require password to wake from sleep" under setting doesn't check and new crash report is created after open lid & login again
Cc: mu...@chromium.org
issue is still reproducible on ToT 8138.0.0/51.0.2695.1
Components: OS>Kernel>Power
Cc: marc...@chromium.org
Components: OS>Kernel>Graphics
If I'm reading the crash report correctly, this is in the i915 driver.

Does this happen every time or was it just a one-off crash?
Yeah the actual kernel crash is 

<1>[  191.299476] BUG: unable to handle kernel NULL pointer dereference at           (null)
<1>[  191.299511] IP: [<          (null)>]           (null)
<5>[  191.299543] PGD 0 
<5>[  191.299570] Oops: 0010 [#1] SMP 
<0>[  191.301973] gsmi: Log Shutdown Reason 0x03
<5>[  191.301988] Modules linked in: i2c_dev uinput rfcomm snd_hda_codec_realtek snd_hda_codec_hdmi memconsole snd_hda_intel snd_hda_codec isl29018(C) snd_hwdep snd_pcm zram(C) industrialio snd_page_alloc zsmalloc(C) fuse nf_conntrack_ipv6 nf_defrag_ipv6 ip6table_filter ip6_tables ath9k_btcoex ath9k_common_btcoex ath9k_hw_btcoex ath mac80211 cfg80211 ath3k btusb btrtl btbcm btintel bluetooth uvcvideo videobuf2_vmalloc videobuf2_memops videobuf2_core joydev snd_seq_midi snd_seq_midi_event snd_rawmidi snd_seq snd_seq_device snd_timer ppp_async ppp_generic slhc tun
<5>[  191.302415] CPU 1 
<5>[  191.302428] Pid: 17416, comm: kworker/u:8 Tainted: G        WC   3.8.11 #1
<5>[  191.302451] RIP: 0010:[<0000000000000000>]  [<          (null)>]           (null)
<5>[  191.302475] RSP: 0000:ffff88015a87db80  EFLAGS: 00010246
<5>[  191.302503] RAX: ffff880177590000 RBX: 0000000000000000 RCX: 0000000000000040
<5>[  191.302521] RDX: 0000000000000000 RSI: ffff88017759a5c0 RDI: ffff880177542400
<5>[  191.302545] RBP: ffff88015a87dc28 R08: 0000000000000000 R09: ffff88015a87dbd8
<5>[  191.302563] R10: 0000000000000020 R11: 0000000000000000 R12: 0000000000000000
<5>[  191.302581] R13: ffff880177590000 R14: ffff88017802f038 R15: ffff880177542400
<5>[  191.302600] FS:  0000000000000000(0000) GS:ffff88017cb00000(0000) knlGS:0000000000000000
<5>[  191.302630] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
<5>[  191.302646] CR2: 0000000000000000 CR3: 000000003ee0c000 CR4: 00000000000407e0
<5>[  191.302669] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
<5>[  191.302687] DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000400
<5>[  191.302707] Process kworker/u:8 (pid: 17416, threadinfo ffff88015a87c000, task ffff88015aa77800)
<5>[  191.302726] Stack:
<5>[  191.302743]  ffffffffbe8e05c9 00000000be8bf938 0000000000000000 0000000000000000
<5>[  191.302793]  ffff88017802f000 0000000000000000 ffff880165b8ec00 0000000000000000
<5>[  191.302830]  ffff880165926cc0 0000000000000000 0000000000000000 0000000000000000
<5>[  191.302879] Call Trace:
<5>[  191.302901]  [<ffffffffbe8e05c9>] ? intel_update_plane+0x572/0x66b
<5>[  191.302925]  [<ffffffffbe8e10dc>] intel_plane_restore+0x57/0x5d
<5>[  191.302945]  [<ffffffffbe8c26d3>] intel_modeset_setup_hw_state+0x35a/0x493
<5>[  191.302970]  [<ffffffffbe8971d1>] __i915_drm_thaw+0x13f/0x1b1
<5>[  191.303001]  [<ffffffffbe8976d4>] i915_resume+0x8c/0xa5
<5>[  191.303021]  [<ffffffffbe897703>] i915_pm_resume+0x16/0x18
<5>[  191.303042]  [<ffffffffbe804bc3>] pci_pm_resume+0xc4/0xeb
<5>[  191.303060]  [<ffffffffbe804aff>] ? pci_pm_prepare+0x40/0x40
<5>[  191.303081]  [<ffffffffbe8f5075>] dpm_run_callback.isra.3+0x2e/0x83
<5>[  191.303101]  [<ffffffffbe8f51d5>] device_resume+0x10b/0x14d
<5>[  191.303124]  [<ffffffffbe8f5234>] async_resume+0x1d/0x43
<5>[  191.303144]  [<ffffffffbe657a64>] async_run_entry_fn+0xc1/0x1a3
<5>[  191.303167]  [<ffffffffbe64b034>] process_one_work+0x18a/0x2af
<5>[  191.303188]  [<ffffffffbe64d283>] worker_thread+0x135/0x1fb
<5>[  191.303208]  [<ffffffffbe64d14e>] ? flush_delayed_work+0x3e/0x3e
<5>[  191.303227]  [<ffffffffbe650e51>] kthread+0xc0/0xc8
<5>[  191.303247]  [<ffffffffbe650d91>] ? __kthread_parkme+0x6b/0x6b
<5>[  191.303270]  [<ffffffffbeac645c>] ret_from_fork+0x7c/0xb0
<5>[  191.303288]  [<ffffffffbe650d91>] ? __kthread_parkme+0x6b/0x6b
<5>[  191.303304] Code:  Bad RIP value.
<1>[  191.303330] RIP  [<          (null)>]           (null)
<5>[  191.303363]  RSP <ffff88015a87db80>
<5>[  191.303375] CR2: 0000000000000000
<4>[  191.303413] ---[ end trace a49a8625d80b5933 ]---
<0>[  191.317774] Kernel panic - not syncing: Fatal exception
<0>[  191.317806] Kernel Offset: 0x3d600000 from 0xffffffff81000000 (relocation range: 0xffffffff80000000-0xffffffffbfffffff)
<0>[  191.317992] gsmi: Log Shutdown Reason 0x02
re #12: According to hsiangc@, it happens everytime. 

It only happens on Peppy, other members of the Haswell family such as Wolf and Falco are fine.
Please try below steps:
1. go to settings UI page.
2. under People uncheck Require password to wake from sleep and try. 
#CBC-RS/TC-watchlist

Experiencing this on Wolf version 51.0.2694.1. Uploaded crashes and submitted feedback. 
Crash ID e143008400000000 - With "Require password to wake from sleep" CHECKED
Crash ID f08bc08400000000 - With "Require password to wake from sleep" UNCHECKED

Summary: [peppy,wolf]: kernel-Fatal exception-381E7AFD occurs after suspend/resume (was: [peppy]: kernel-Fatal exception-381E7AFD occurs after suspend/resume)
issue also reproduce with 8150.0.0/51.0.2699.0 on wolf
Summary: [Haswell]: kernel-Fatal exception-381E7AFD occurs after suspend/resume (was: [peppy,wolf]: kernel-Fatal exception-381E7AFD occurs after suspend/resume)
issue also reproduce on falco with 8150.0.0/51.0.2699.0
issue only reproduce on normal mode and no reproduce on dev mode
Re.#21: I'm seeing this in Dev mode
Issue 596926 was raised from failing suspend/resume in external display tests with chameleon. The kernel crash pretty much matches the one here.

https://wmatrix.googleplex.com/unfiltered?tests=display_HotPlugAtSuspend*,&platforms=falco&days_back=30&releases=51

The last good state was at R51-8075.0.0 and the crashes started at R51-8077.0.0
Thanks Kalin that's super useful. For reference here is the diff:

https://crosland.corp.google.com/log/8075.0.0..8077.0.0


Owner: dbasehore@chromium.org
Derek, that's very likely to come from your changes. Let me know what you want to do about it.
Status: Assigned (was: Untriaged)
Issue 596926 has been merged into this issue.
Status: Started (was: Assigned)
If we merge https://codereview.chromium.org/1861593002/ we don't need my kernel patches, so they could be reverted.

I'll try to see what's going on first, though.
Project Member

Comment 30 by bugdroid1@chromium.org, Apr 7 2016

Labels: merge-merged-chromeos-3.8
The following revision refers to this bug:
  https://chromium.googlesource.com/chromiumos/third_party/kernel/+/71edefddb401f3caf1b1d37dc8995f0f1bd7b462

commit 71edefddb401f3caf1b1d37dc8995f0f1bd7b462
Author: Derek Basehore <dbasehore@chromium.org>
Date: Wed Apr 06 22:20:43 2016

Revert "UPSTREAM: PM: make VT switching to the suspend console optional v3"

This reverts commit 46c333e66a70b08235cb39792524d0845550eafe.

We're seeing crashes on resume in some cases due to this patch series.

BUG= 597131 
TEST=suspend/resume on peppy

Change-Id: Ie22230ef6884d70c8136e378ee38aa13369eb202
Signed-off-by: Derek Basehore <dbasehore@chromium.org>
Reviewed-on: https://chromium-review.googlesource.com/337590
Reviewed-by: Stéphane Marchesin <marcheu@chromium.org>

[modify] https://crrev.com/71edefddb401f3caf1b1d37dc8995f0f1bd7b462/kernel/power/console.c
[modify] https://crrev.com/71edefddb401f3caf1b1d37dc8995f0f1bd7b462/include/linux/pm.h

Project Member

Comment 31 by bugdroid1@chromium.org, Apr 7 2016

The following revision refers to this bug:
  https://chromium.googlesource.com/chromiumos/third_party/kernel/+/733ec4c0438c391b35df96d84550faf2d1d13661

commit 733ec4c0438c391b35df96d84550faf2d1d13661
Author: Derek Basehore <dbasehore@chromium.org>
Date: Wed Apr 06 22:20:19 2016

Revert "UPSTREAM: fb: add support for drivers not needing VT switch at suspend/resume time"

This reverts commit fc62fdd54572142855c9a9888f347a7e0839a616.

We're seeing crashes on resume in some cases due to this patch series.

BUG= 597131 
TEST=suspend/resume on peppy

Change-Id: I77418235eed3f9a6bc276578f191033c2784c29a
Signed-off-by: Derek Basehore <dbasehore@chromium.org>
Reviewed-on: https://chromium-review.googlesource.com/337409
Reviewed-by: Stéphane Marchesin <marcheu@chromium.org>

[modify] https://crrev.com/733ec4c0438c391b35df96d84550faf2d1d13661/include/linux/fb.h
[modify] https://crrev.com/733ec4c0438c391b35df96d84550faf2d1d13661/drivers/video/fbmem.c

Project Member

Comment 32 by bugdroid1@chromium.org, Apr 7 2016

Labels: merge-merged-chromeos-3.8
The following revision refers to this bug:
  https://chromium.googlesource.com/chromiumos/third_party/kernel/+/309b66209026de92195c79ac42b44571ff943504

commit 309b66209026de92195c79ac42b44571ff943504
Author: Derek Basehore <dbasehore@chromium.org>
Date: Wed Apr 06 22:19:51 2016

Revert "BACKPORT: drm/i915: restore cursor and sprite state when forcing a config restore v2"

This reverts commit 21e91aa6954d8c0ace44556c6560b25170432e7c.

We're seeing crashes on resume in some cases due to this patch series.

BUG= 597131 
TEST=suspend/resume on peppy

Change-Id: I7568de1f09293368ca90f9f97af6444bfc056256
Signed-off-by: Derek Basehore <dbasehore@chromium.org>
Reviewed-on: https://chromium-review.googlesource.com/337408
Reviewed-by: Stéphane Marchesin <marcheu@chromium.org>

[modify] https://crrev.com/309b66209026de92195c79ac42b44571ff943504/drivers/gpu/drm/i915/intel_display.c

Project Member

Comment 33 by bugdroid1@chromium.org, Apr 7 2016

The following revision refers to this bug:
  https://chromium.googlesource.com/chromiumos/third_party/kernel/+/f658e63acdfd7d206570c241055623f388b1f0fd

commit f658e63acdfd7d206570c241055623f388b1f0fd
Author: Derek Basehore <dbasehore@chromium.org>
Date: Wed Apr 06 22:19:13 2016

Revert "BACKPORT: drm/i915: enable VT switchless resume v3"

This reverts commit d1ea88541b8d25ff7f09a276aea517a229f30fa3.

We're seeing crashes on resume in some cases due to this patch series.

BUG= 597131 
TEST=suspend/resume on peppy

Change-Id: I731867587df23351c5e2cb0fd29990afd54d13e0
Signed-off-by: Derek Basehore <dbasehore@chromium.org>
Reviewed-on: https://chromium-review.googlesource.com/337407
Reviewed-by: Stéphane Marchesin <marcheu@chromium.org>

[modify] https://crrev.com/f658e63acdfd7d206570c241055623f388b1f0fd/drivers/gpu/drm/i915/i915_drv.c
[modify] https://crrev.com/f658e63acdfd7d206570c241055623f388b1f0fd/drivers/gpu/drm/i915/intel_fbdev.c

Project Member

Comment 34 by bugdroid1@chromium.org, Apr 7 2016

The following revision refers to this bug:
  https://chromium.googlesource.com/chromiumos/third_party/kernel/+/1e054645bbcc500ff44fb745672e79a5c97cb648

commit 1e054645bbcc500ff44fb745672e79a5c97cb648
Author: Derek Basehore <dbasehore@chromium.org>
Date: Wed Apr 06 22:17:56 2016

Revert "CHROMIUM: drm/i915: take mode config lock around crtc disable at suspend"

This reverts commit ab96a12cf5d5ea499b3c67ac80421020bee60652.

We're seeing crashes on resume in some cases due to this patch series.

BUG= 597131 
TEST=suspend/resume on peppy

Change-Id: I5cdda5ab02fc33f238fd45963bea093142559acd
Signed-off-by: Derek Basehore <dbasehore@chromium.org>
Reviewed-on: https://chromium-review.googlesource.com/337406
Reviewed-by: Stéphane Marchesin <marcheu@chromium.org>

[modify] https://crrev.com/1e054645bbcc500ff44fb745672e79a5c97cb648/drivers/gpu/drm/i915/i915_drv.c

Status: Fixed (was: Started)
 Issue 600017  has been merged into this issue.
Issue 600141 has been merged into this issue.
Status: Verified (was: Fixed)
Verified on 8712.6.0/51.0.2704.15

Comment 39 by sco...@scosol.org, Apr 20 2016

Verified as fixed.

Sign in to add a comment