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

Issue 823761 link

Starred by 3 users

Issue metadata

Status: Verified
Owner:
Closed: Apr 2018
Cc:
Components:
EstimatedDays: ----
NextAction: ----
OS: Chrome
Pri: 1
Type: Bug-Regression



Sign in to add a comment

[Kevin, Bob] Loading flash webpage causes kernel crash CPU: 2 PID: 1701 Comm: DrmThread Not tainted)

Project Member Reported by matthewjoseph@chromium.org, Mar 20 2018

Issue description

Google Chrome:   	66.0.3359.43 (Official Build) dev (32-bit)
Revision:	        0
Platform:	        10452.19.0 (Official Build) dev-channel kevin
Firmware Version:	Google_Kevin.8785.220.0
Customization ID:	SAMSUNG-KEVIN1

What steps will reproduce the problem?
(1) Recover kevin device with 10452.19.0
(2) Navigate to https://online-voice-recorder.com - or any other flash based site
    a) If necessary, click the Flash content to enable running Flash  
(3) Observe crash

What is the expected result?
Chrome should render the website

What happens instead?
Screen goes black and then the Kernel crashes

kernel-(WATCHDOG)-*ERROR* Failed to enable vop (-16-359F8C7B) (handling)

*NOTES

I've tried to reproduce on Kevin, Orco, Terra, Eve, and Caroline. So far, I can only reproduce on Kevin.

Kernel Crashes:
- https://crash.corp.google.com/browse?q=reportid=%278038d2bdd01cc879%27
- https://crash.corp.google.com/browse?q=reportid=%27afc18a6c28c7a43e%27
- https://crash.corp.google.com/browse?q=reportid=%272f327371e9e4d4cd%27

Kernel Warnings:
- https://crash.corp.google.com/browse?q=reportid=%2794a9814a7fc28d1d%27
- https://crash.corp.google.com/browse?q=reportid=%27b0bfe6be67da43e7%27


I'm able to reproduce this 100% of the time.


Looking at dmesg shows about 70 entries of the following error:
[  388.099742] rk_iommu ff903f00.iommu: Page fault at 0x0000000014016900 of type read
[  388.099768] rk_iommu ff903f00.iommu: iova = 0x0000000014016900: dte_index: 0x50 pte_index: 0x16 page_offset: 0x900
[  388.099779] rk_iommu ff903f00.iommu: mmu_dte_addr: 0x00000000ede50000 dte@0x00000000ede50140: 0xad7c3001 valid: 1 pte@0x00000000ad7c3058: 0x1e7a6006 valid: 0 page@0x0000000000000000 flags: 0x0
[  388.116386] rk_iommu ff903f00.iommu: Page fault at 0x0000000014017000 of type read
[  388.116408] rk_iommu ff903f00.iommu: iova = 0x0000000014017000: dte_index: 0x50 pte_index: 0x17 page_offset: 0x0
[  388.116418] rk_iommu ff903f00.iommu: mmu_dte_addr: 0x00000000ede50000 dte@0x00000000ede50140: 0xad7c3001 valid: 1 pte@0x00000000ad7c305c: 0x1e7a7006 valid: 0 page@0x0000000000000000 flags: 0x0
 
Cc: vsu...@chromium.org
Labels: ReleaseBlock-Dev
Issue is reproducible on bob as well.  Likely will reproduce on gru boards.
Summary: [Kevin, Bob] Loading flash webpage causes kernel crash (was: [Kevin] Loading flash webpage causes kernel crash )
This doesn't occur when loading the same page in Guest Mode.  Occurs only when logged in with a user account.

Comment 7 by ka...@chromium.org, Mar 20 2018

Components: Internals>Media
Summary: [Kevin, Bob] Loading flash webpage causes kernel crash CPU: 2 PID: 1701 Comm: DrmThread Not tainted) (was: [Kevin, Bob] Loading flash webpage causes kernel crash )
[ 1725.365727] ------------[ cut here ]------------
[ 1725.365753] WARNING: CPU: 2 PID: 1701 at ../../../../../tmp/portage/sys-kernel/chromeos-kernel-4_4-4.4.118-r1396/work/chromeos-kernel-4_4-4.4.118/drivers/iommu/iommu.c:1280 __iommu_detach_group+0x54/0xa0
[ 1725.365762] Modules linked in: snd_usb_audio snd_hwdep snd_usbmidi_lib snd_rawmidi ip6t_REJECT nf_reject_ipv6 veth cmac rfcomm btusb btrtl btbcm btintel bluetooth uinput ip6table_filter uvcvideo mwifiex_pcie mwifiex zram bridge stp llc ipt_MASQUERADE nf_nat_masquerade_ipv4 xt_mark fuse snd_seq_dummy snd_seq snd_seq_device cfg80211 joydev
[ 1725.365913] 
[ 1725.365928] CPU: 2 PID: 1701 Comm: DrmThread Not tainted 4.4.118-13169-g3b0f04b2f8fc #1
[ 1725.365935] Hardware name: Google Kevin (DT)
[ 1725.365944] task: ffffffc0ef188d00 ti: ffffffc0daf84000 task.ti: ffffffc0daf84000
[ 1725.365954] PC is at __iommu_detach_group+0x54/0xa0
[ 1725.365963] LR is at __iommu_detach_group+0x50/0xa0
[ 1725.365971] pc : [<ffffffc000579098>] lr : [<ffffffc000579094>] pstate: 60000145
[ 1725.365978] sp : ffffffc0daf879d0
[ 1725.365985] x29: ffffffc0daf879f0 x28: ffffffc0010f6890 
[ 1725.365998] x27: 0000000000000000 x26: ffffffc000a57d58 
[ 1725.366012] x25: 0000000000000001 x24: 0000000000000028 
[ 1725.366025] x23: ffffffc0edaea028 x22: ffffffc0ede03100 
[ 1725.366038] x21: ffffffc0ee5ea448 x20: ffffffc0ede02880 
[ 1725.366051] x19: ffffffc0ee5ea400 x18: ffffffc00108d000 
[ 1725.366064] x17: ffffffc00108d000 x16: ffffffc0e9ce3ca8 
[ 1725.366077] x15: 0000000000000000 x14: 6330303030783020 
[ 1725.366090] x13: 3a73757461747320 x12: 2c74756f2064656d 
[ 1725.366103] x11: ffffffc0ef188d00 x10: 0000000000000000 
[ 1725.366116] x9 : 0000000000000000 x8 : ffffffc0ef188d00 
[ 1725.366128] x7 : ffffffc0010d1028 x6 : 0000000000000059 
[ 1725.366141] x5 : 0000000000000000 x4 : ffffffc001199329 
[ 1725.366153] x3 : 0000000000000059 x2 : ffffffc0f7b2e4d0 
[ 1725.366166] x1 : 0000000000000140 x0 : 00000000ffffff92 
[ 1725.366180] 
[ 1725.366180] PC: 0xffffffc000579018:
...
[ 1725.371911] 
[ 1725.371921] ---[ end trace 46a780f3efcf4b12 ]---
[ 1725.371929] Call trace:
[ 1725.371938] [<ffffffc000579098>] __iommu_detach_group+0x54/0xa0
[ 1725.371951] [<ffffffc000578f6c>] iommu_detach_device+0x80/0xa0
[ 1725.371968] [<ffffffc0005ab564>] rockchip_drm_dma_detach_device+0x34/0x40
[ 1725.371981] [<ffffffc0005b0894>] vop_crtc_disable+0x1f0/0x234
[ 1725.371993] [<ffffffc0005865f0>] drm_atomic_helper_commit_modeset_disables+0x1a4/0x32c
[ 1725.372007] [<ffffffc0005ac29c>] rockchip_atomic_commit_complete+0x284/0x4a0
[ 1725.372018] [<ffffffc0005aca44>] rockchip_drm_atomic_commit+0x168/0x18c
[ 1725.372033] [<ffffffc00059eb08>] drm_atomic_commit+0x58/0x64
[ 1725.372043] [<ffffffc000587364>] drm_atomic_helper_set_config+0x80/0xb0
[ 1725.372055] [<ffffffc000593f7c>] drm_mode_set_config_internal+0x5c/0xdc
[ 1725.372065] [<ffffffc000594934>] drm_mode_setcrtc+0x1ec/0x48c
[ 1725.372082] [<ffffffc0007a838c>] drm_ioctl+0x1d8/0x3f8
[ 1725.372093] [<ffffffc0005a64a8>] drm_compat_ioctl+0x3c/0x9c
[ 1725.372107] [<ffffffc00039e274>] compat_SyS_ioctl+0x400/0x191c
[ 1725.372121] [<ffffffc000203e90>] __sys_trace_return+0x0/0x4
[ 1725.385854] atmel_mxt_ts 3-004b: Status: 10 Config Checksum: 573e89
[ 1725.393387] atmel_mxt_ts 3-004b: Status: 00 Config Checksum: 573e89
[ 1772.732095] rockchip-vop ff900000.vop: Failed to attach iommu device
[ 1772.732126] rockchip-vop ff900000.vop: failed to attach dma mapping, -16
[ 1772.732235] rockchip-vop ff900000.vop: [drm:vop_crtc_enable] *ERROR* Failed to enable vop (-16)
[ 1779.996778] BUG: spinlock lockup suspected on CPU#5, swapper/5/0
[ 1779.996811]  lock: 0xffffffc0010aec98, .magic: dead4ead, .owner: swapper/2/0, .owner_cpu: 2
[ 1779.996834] CPU: 5 PID: 0 Comm: swapper/5 Tainted: G        W       4.4.118-13169-g3b0f04b2f8fc #1
[ 1779.996845] Hardware name: Google Kevin (DT)
[ 1779.996854] Call trace:
[ 1779.996885] [<ffffffc0002083c4>] dump_backtrace+0x0/0x16c
[ 1779.996901] [<ffffffc0002083b8>] show_stack+0x20/0x2c
[ 1779.996922] [<ffffffc0004d32a8>] __dump_stack+0x20/0x28
[ 1779.996935] [<ffffffc0004d3264>] dump_stack+0x70/0x94
[ 1779.996952] [<ffffffc00025b218>] spin_dump+0x98/0xa8
[ 1779.996968] [<ffffffc0002eff90>] do_raw_spin_lock+0x168/0x1a8
[ 1779.996984] [<ffffffc0009401c8>] _raw_spin_lock+0x30/0x3c
[ 1779.996998] [<ffffffc00027df38>] __tick_broadcast_oneshot_control+0x44/0x234
[ 1779.997011] [<ffffffc00027cf98>] tick_broadcast_oneshot_control+0x3c/0x48
[ 1779.997027] [<ffffffc0007c2a2c>] cpuidle_enter_state+0x100/0x184
[ 1779.997039] [<ffffffc0007c2bf8>] cpuidle_enter+0x30/0x40
[ 1779.997051] [<ffffffc0002eee00>] cpu_startup_entry+0x11c/0x2a8
[ 1779.997064] [<ffffffc00020d2a4>] secondary_start_kernel+0x144/0x150
[ 1779.997074] [<0000000000400c0c>] 0x400c0c
Cc: hsiangc@chromium.org avkodipelli@chromium.org
Owner: ihf@chromium.org
Cc: ihf@chromium.org
Owner: ----
issue also reproduce on https://webcamera.io
Cc: josa...@chromium.org
on https://webcamera.io
issue reproduce if use Flash but no reproduce if use HTML5

Comment 14 by ihf@chromium.org, Mar 20 2018

Components: -OS>Kernel -Internals>Plugins>Flash -Internals>Media OS>Kernel>Graphics
Owner: dbehr@chromium.org
Status: Assigned (was: Untriaged)
Dominik, could you please take a look at the issue? The uprev was M-67 only, right?

Comment 15 by josa...@google.com, Mar 20 2018

was this issue seen on the last dev?
issue can reproduce on last dev 10452.11.0/66.0.3359.31

Comment 17 by josa...@google.com, Mar 20 2018

Labels: -ReleaseBlock-Dev ReleaseBlock-Beta
moving to RBB as per c#16

Comment 18 by josa...@google.com, Mar 27 2018

Any update on this blocking issue?

Comment 19 by dbehr@chromium.org, Mar 27 2018

Cc: dcasta...@chromium.org
Looks like reverting this stuff:
63a8f4b61aab (HEAD) Revert "Revert "CHROMIUM: drm/rockchip: YUV overlays BT.601 color conversion.""
48a8e27fccd0 Revert "CHROMIUM: drm/rockchip: YUV overlays BT.601 color conversion."
190055d71e83 Revert "CHROMIUM: rockchip: Implement drm plane->ctm property."
3ae397b65827 Revert "CHROMIUM: rockchip: Enable 'PLANE_CTM' drm property."

Fixes the problem.



how confident are we on that? 
Also, any side effect?

I can try reverting in M66 branch 
Cc: hoegsberg@chromium.org
Owner: dcasta...@chromium.org
Those patches are needed for efficient night mode, we are targeting 67 for that, so if that is causing issues in 66, please go ahead and revert, I'll try to investigate and find a fix for 67.
dbehr, have you tried reverting only the last two patches (plane ctm)? Is it necessary to revert the first two patches (the BT.601) too to fix the issue?
Cc: kbleicher@google.com allendam@chromium.org
 Issue 826022  has been merged into this issue.
Project Member

Comment 24 by bugdroid1@chromium.org, Apr 2 2018

Labels: merge-merged-chromeos-4.4
The following revision refers to this bug:
  https://chromium.googlesource.com/chromiumos/third_party/kernel/+/e72a9fe83d2ab82514032beb5f3b6a1fad5563ca

commit e72a9fe83d2ab82514032beb5f3b6a1fad5563ca
Author: Daniele Castagna <dcastagna@chromium.org>
Date: Mon Apr 02 23:26:43 2018

CHROMIUM: drm/rockchip: Fix a gone-wrong revert

crrev.com/c/922542 was supposed to revert a patch that added BT.601
YUV conversion.
Something went wrong and y-mirror support partially went away with it,
causing iommu issues since we scanout y-flipped buffers starting from
the last line.

BUG= chromium:823761 
TEST=on kevin following the instruction in chromium:823761

Change-Id: I49f75b192acf7906521d4d2e20d0818c1dee3e93
Reviewed-on: https://chromium-review.googlesource.com/988495
Commit-Ready: Daniele Castagna <dcastagna@chromium.org>
Commit-Ready: Kristian H. Kristensen <hoegsberg@chromium.org>
Tested-by: Daniele Castagna <dcastagna@chromium.org>
Reviewed-by: Kristian H. Kristensen <hoegsberg@chromium.org>

[modify] https://crrev.com/e72a9fe83d2ab82514032beb5f3b6a1fad5563ca/drivers/gpu/drm/rockchip/rockchip_drm_vop.c

Labels: Merge-Request-66
Labels: -Merge-Request-66 Merge-Approved-66
Project Member

Comment 27 by bugdroid1@chromium.org, Apr 2 2018

Labels: merge-merged-release-R66-10452.B-chromeos-4.4
The following revision refers to this bug:
  https://chromium.googlesource.com/chromiumos/third_party/kernel/+/21c1bbdb387ef1f58dfda5ac13adddfbdc47eb8a

commit 21c1bbdb387ef1f58dfda5ac13adddfbdc47eb8a
Author: Daniele Castagna <dcastagna@chromium.org>
Date: Mon Apr 02 23:41:42 2018

CHROMIUM: drm/rockchip: Fix a gone-wrong revert

crrev.com/c/922542 was supposed to revert a patch that added BT.601
YUV conversion.
Something went wrong and y-mirror support partially went away with it,
causing iommu issues since we scanout y-flipped buffers starting from
the last line.

BUG= chromium:823761 
TEST=on kevin following the instruction in chromium:823761

Change-Id: I49f75b192acf7906521d4d2e20d0818c1dee3e93
Reviewed-on: https://chromium-review.googlesource.com/988495
Commit-Ready: Daniele Castagna <dcastagna@chromium.org>
Commit-Ready: Kristian H. Kristensen <hoegsberg@chromium.org>
Tested-by: Daniele Castagna <dcastagna@chromium.org>
Reviewed-by: Kristian H. Kristensen <hoegsberg@chromium.org>
(cherry picked from commit e72a9fe83d2ab82514032beb5f3b6a1fad5563ca)
Reviewed-on: https://chromium-review.googlesource.com/990833
Reviewed-by: Grace Kihumba <gkihumba@chromium.org>
Commit-Queue: Grace Kihumba <gkihumba@chromium.org>
Tested-by: Grace Kihumba <gkihumba@chromium.org>

[modify] https://crrev.com/21c1bbdb387ef1f58dfda5ac13adddfbdc47eb8a/drivers/gpu/drm/rockchip/rockchip_drm_vop.c

Status: Fixed (was: Assigned)
Status: Verified (was: Fixed)
verified on 10452.42.0/66.0.3359.79
Project Member

Comment 30 by sheriffbot@chromium.org, Apr 6 2018

Cc: gkihumba@google.com
This issue has been approved for a merge. Please merge the fix to any appropriate branches as soon as possible!

If all merges have been completed, please remove any remaining Merge-Approved labels from this issue.

Thanks for your time! To disable nags, add the Disable-Nags label.

For more details visit https://www.chromium.org/issue-tracking/autotriage - Your friendly Sheriffbot
Labels: -Merge-Approved-66

Sign in to add a comment