cc'ing Intel, in case this turns out to be Intel-specific.

I have tried two Skylake systems - Intel HD Graphics 520 and 530. Both work well. I believe it has to be an Intel Graphics 620 or 630  (Kabylake) to reproduce this issue.

+piman@, sunnyps@ in case it's a sync token issue.

I dont' think there's anything to bisect if this is cause by the OS update

(I very very strongly doubt that it's a SyncToken issue)

chrome://gpu reports Direct Composition and Overlays are both enabled. Might be a DirectComposition bug. Since DirectComposition is closely tied to dwm, it's possible dwm froze because of a bug in the driver/runtime.

Can you try running with --disable-direct-composition? If that fixes the problem, please also try --disable-direct-composition-overlays (without the first flag).

Just tried launching Chrome with --disable-direct-composition : the problem does not happen as frequently as before, but still happens

Correction: it seems to keep happening with --disable-direct-composition-overlays, but not with --disable-direct-composition (after a few hours usage)

Hi

If any help it feels like some sort of deadlocking.  I noticed that on each freeze my wallpaper was always in the middle of fading to the next image, originally I thought it was the wallpaper transition causing the freeze, so set a static wallpaper, the problem was then less frequent but still occurred when using Chrome.  

It always seems to freeze if I'm doing something quick in Chrome, by that I mean launching a new tab and flicking back and forth, so Chrome is doing a few things at once regarding the UI.  Since turning off hardware acceleration in Chrome no freezes.

This is on Intel HD 630 driver version 24.20.100.6025 (the latest as of writing) however I also rollback the driver to the previous version and freezes still happened.

Am UHD 630 owner here. Random hang after win upgrade. Also suspect is due to chrome. 

Perhaps helpful. Same here. 
Thinkpad X1C5, i7-7600U Kaby Lake, HD 620 (so matches the pattern)

crtl-alt-del and most things don't work.
mouse still moves and sound and brightness buttons still respond

holding power to enter sleep and then exiting sleep restores function.

chrome-gpu-after-freeze.html 153 KB View Download

Same issue here. Thinkpad T470s i7-7600U, IGP

This massive problem only started after the April update 1803. But before I always had some minor graphics issues on this device. There were artifacts (mostly green) during video decode in Chrome (Youtube, Netflix, ...). But also inside the Netflix App (which is using Edge).

And also minimizing and maximizing Youtube videos to fullscreen was always very slow on this device. Sometimes after maximizing, the video get's stuck for a few seconds. Or even totally, until closing full screen and trying again.

I suspect the GPU to be damaged. This seems to be a common problem.

Btw, I'm using the newest Lenovo drivers. And I had the feeling things got worse with the newest Intel GPU driver revision from March.

This is Yang Gu from Intel Web Team, and currently I'm working with Google Chrome Team to investigate this issue. Below are some findings at my side today:
1. It seems all the issues happen only on Kabylake (HD 620 and 630, https://en.wikipedia.org/wiki/List_of_Intel_graphics_processing_units). I tried with one Skylake (HD 530) and one Kabylake, and hit no problem with Skylake, but I ran into hang on Kabylake several times.
2. It seemed to me Chrome Stable (mine version is 66.0.3359.139) is prone to the issue, and I didn't run into problem with Chrome Canary (mine version is 68.0.3418.0). Can someone help to try with Canary (https://www.google.com/chrome/browser/canary.html)?
3. Many times when the UI is paused, I just need to wait for a few seconds, and the system can come back to life again. 
4. I didn't repro the issue with --use-angle=gl, though the system sometimes was very slow to respond. Maybe I was just lucky here.


The most problem for me now is I couldn't find a way to easily and steadily repro the issue. Please share if you have some good ideas on this, which might be very helpful for further investigation!

Adding the conops hotlist.

We haven't spotted too many reports on the community side yet, but we're going to try to consolidate them all into this thread as they come in:

- https://productforums.google.com/forum/#!topic/chrome/bzZPHqKXODw

Thanks!

@Yang Gu, yes it would appear to be HD 620/630 graphics are affected.  I had tried leaving the system for some time and it didn't resolve.  I did find turning off and on the monitors would restore a working system as this with Display Port 1.2 and daisy chained monitors this causes Windows to re-enumerate and reset the graphics driver.

It seemed to me having a Windows theme that cycled the wallpaper made it more prone to freezing, as I noticed it always seemed to freeze halfway through a transition of the wallpaper AND when using Chrome.  It still happened though when I set a static image but perhaps not so often.

I will try the Canary channel version of Chrome and report back.

Hi

I've done a quick text compare between the results of chrome://gpu for Canary and stable on my system to see what the differences are.

On stable CheckerImaging: is Disabled, on Canary it is Enabled.

Canary has an additional setting for "Viz Service Display Compositor: Disabled", this setting doesn't exist in the stable version.

Not sure if that helps pinpoint a potential cause.

As per above comments checked the issue on Windows 10 with versions 1703,1709 and version 1803, hp laptop with Intel driver and issue is not reproducible from TE end. Attaching gpu details.

As issue is not reproducible from TE end adding TE-NeedsTriageHelp label to remove this issue from TE triaging bucket. Please change if not the case.

Thanks!

gpu(1).pdf 135 KB Download

@Yang Gu, I've been using Canary version 68.0.3418.0 pretty much continuously since switching to it and haven't had any issues with freezes, which I would have had several times by now with the stable version with GPU Hardware enabled.

@plee5509, thanks for the double check on Canary! At least this gives us more hints about what happened, and you may have a workaround to use Chrome now:) 
If we can find a more efficient way to repro the issue, I can also try to bisect the Chrome builds to see if some commit caused or solved the issue. 
Thanks again!

Thanks so much for testing and investigating on the issue.

To summarize the findings so far
1) It seems --disable-direct-composition works around the issue
2) It seems the issue no longer manifest in Canary

So now we just need a device that we can easily repro the hang, so we can bisect.

plee5509@gmail.com: do you mind also trying Chrome Beta?

https://www.google.com/chrome/browser/beta.html

We still haven't located a device that we can reproduce the hang in house here.

For those who can reproduce this bug, could you please share what machine you have and also share the about:gpu page content from Chrome?

I confirm the issue. I can reproduce it with Slack for Windows running (is it Chromium?) without Chrome running. Happens only after Windows 10 upgrade to 1803.17134.1.
i7-7500U, Mobile, Kaby Lake, Intel HD Graphics 620.
Video Driver v23.20.16.5018 (03/27/2018).

Mouse is moving, mouse buttons not working, keyboard mostly not working (except CapsLock, brightness), Ctrl-Alt-Del not working.

Workaround: close and open lid, or push sleep button and again to resume.

Tested video driver v24.20.100.6025 (25.04.2018), latest on Intel website — same issue.
Third workaround: press Win-Ctrl-Shift-B.

vanav@vanav.org: what is your machine model? I have the same CPU/GPU/OS/driver as yours, but no luck reproducing.

Lenovo Thinkpad X1 Carbon 5th Gen (2017), display non-touch 1920x1080.
Issue happens when I minimize/restore Slack or scroll it. Chrome is not running.
User https://bugs.chromium.org/p/chromium/issues/detail?id=838809#c28 have similar hardware.

Running a desktop i5-7600k on an Asus Z270G motherboard, April update, 24.20.100.6025, 66.0.3359.139 (Official Build) (64-bit), static desktop background image.
The freezes are not easily reproducible however, have to use chrome for about an hour before it happens. Can't find any one action that consistently triggers it.

gpu.htm 125 KB View Download

also, there are some others posting their configs here: https://www.reddit.com/r/Windows10/comments/8gf68w/chrome_freezes_windows_after_april_update/

I can confirm and reproduce this issue using Chrome. It also started after installing the latest Windows 10 April 2018 feature update to 1803.
My machine is a late-2016 Razer Blade Stealth.
i7-7500U, Kaby-Lake, Integrated Intel HD Graphics 620
Video Driver: 23.20.16.4973

Mouse moves, but mouse buttons do not work. Only option is to turn off power to laptop (close lid) and power back on, which will fix the issue most of the time (not always). Using Canary instead seems to work fine. I've attached a text file of my about:gpu settings.

gpu.txt 11.4 KB View Download

Update: Win-Ctrl-Shift-B workaround sometimes works, but sometimes completely halts the OS, need to do hard reboot with data loose.

Intel folks, can you please get us stack traces for dwm.exe and chrome gpu process when this bug happens?

Easiest way would be to enable full kernel dumps on a keyboard shortcut as described here: https://docs.microsoft.com/en-us/windows-hardware/drivers/debugger/forcing-a-system-crash-from-the-keyboard

But that didn't work for me during testing, so here are instructions for setting up a remote debugger on the failing hardware and grabbing stack traces from a host machine:

First, install the version of Remote Debugging Tools that matches your Visual Studio version on the remote device: https://docs.microsoft.com/en-us/visualstudio/debugger/remote-debugging

Configure Visual Studio on host machine with MS and Chrome symbol servers:
https://msdl.microsoft.com/download/symbols
https://chromium-browser-symsrv.commondatastorage.googleapis.com

Run the remote debugger with authentication or not depending on your setup. Make a note of the port number (default is 4022).

Connect the remote device to ethernet e.g. by using a USB to ethernet adapter, and note its IP address or hostname.

When the bug happens, use Visual Studio on host machine to attach to dwm.exe and Chrome gpu process on the remote machine. The menu entry is "Debug > Attach to Process". Select "Remote (no authentication)" as "Connection type", and enter the IP address or hostname and port in "Connection target" e.g. 172.10.16.24:4022. Then you'll see a list of processes. The processes we're interested in are dwm.exe, and chrome.exe GPU process. You can get PID for GPU process from Chrome task manager (More tools > Task manager), or from Windows Task Manager / procexp (look for --type=gpu-process in command line args).

If you attach to these processes break execution and copy the stack traces here.

Hi, apparently Microsoft are aware of the problem (also affecting Cortana) and have an aim of issuing a patch on Tuesday 8 May.

So the problem appears to be with Windows.

https://answers.microsoft.com/en-us/windows/forum/windows_10-win_cortana/some-devices-may-hang-or-freeze-when-using-certain/612a341b-340a-4ac0-8866-df5346327a52?tm=1525304726596&auth=1

Some of updates as of today:
1. I couldn't get any stack traces. 
I chose the way of remote debugger, and I think I set everything correctly. 
For dwm, I could attach to the process, and downloaded all the symbols from both Microsoft and Chromium servers, but I could see nothing in the stack. BTW I noticed the column "Title" of dwm is "Page Unresponsive (Not Responding)", and the log are as below:
'dwm.exe' (Win32): Loaded 'C:\Windows\System32\dwm.exe'. Symbols loaded.
'dwm.exe' (Win32): Loaded 'C:\Windows\System32\ntdll.dll'. Symbols loaded.
......
'dwm.exe' (Win32): Loaded 'C:\Windows\System32\iertutil.dll'. Symbols loaded.
'dwm.exe' (Win32): Loaded 'C:\Windows\System32\wuceffects.dll'. Symbols loaded.
'dwm.exe' (Win32): Loaded 'C:\Windows\System32\d3d10warp.dll'. Symbols loaded.
The thread 0x1210 has exited with code 0 (0x0).

For GPU process, I couldn't attach it and got the error message "Unable to attach to the process. A device attached to the system is not functioning."

2. I found an effective way to repro the issue, and most of time, the hang could be triggered within a few seconds. 
I used 5 websites as below, and they are all very heavy GPU workloads.
https://webglsamples.org/aquarium/aquarium.html?numFish=30000 (I used a local version instead, but I think the upstream one is the same)
https://webvr.info/samples/test-slow-render.html?heavyGpu=1&cubeScale=0.5&workTime=4&standardSize=1&renderScale=1.0&submitInterval=3
https://webvr.info/samples/test-slow-render.html?heavyGpu=1&cubeScale=0.5&workTime=12&standardSize=1&renderScale=1.0&halfOnly=1
https://webvr.info/samples/test-slow-render.html?heavyGpu=1&cubeScale=0.3&workTime=12&standardSize=1&renderScale=1.0
https://webvr.info/samples/test-slow-render.html?heavyGpu=1&cubeScale=0.3&standardSize=1&renderScale=1.0

I opened them in both Edge and Chrome I wanted to test. Then I used alt+tab to switch between browsers, and held ctrl+tab to switch among different tabs. Most of time, the hang would be triggered within a few seconds (< 10 seconds). Further experiments are all based on this method.

3. Not only Kabylake (HD630), but also Coffeelake (UHD 630) encountered the hang issue. We have 2 Coffeelake and they can both repro the issues, but not as easy as Kabylake. So my main focus was on Kabylake.

4. Both Chrome Canary (68.0.3418.0) and Chrome Beta (67.0.3396.30) are good (never encountered a hang). I know the status of Beta is more important as it might impact the upcoming release. I baked the Beta many times (interleaved with other tests) for a long time (more than 1 hour in total), but didn't encounter any problem. However, Stable (66.0.3359.139) encountered the hang very quickly (< 10 seconds). BTW, I checked passthrough decoder, and they are all false for Canary, Beta and Stable. 

5. I began to bisect the builds to try to understand the diff between Stable and Beta. I haven't finished it yet, but the current situation is a bit complicated. 
I intended to find the commit from bad (hang is quickly encountered) and good (hang is never encountered). However, I met a situation in the middle. For 540442, it took me much longer time (several minutes) to hit the hang, and I did encounter twice. Below are detailed revisions and their corresponding status, and I will continue the bisect. 
555651 (latest Canary) good
540442 (1st M67) much longer time to trigger hang, and I encountered 2 times
540396 bad
540374 bad
540355 bad
540294 bad
540130 bad
539819 bad
539194 bad
537925 bad
535407 bad
530421 (1st M66) bad

6. Win-Ctrl-Shift-B didn't work for me. 

7. I never pressed power button so many times within a day. Hope it can survive;)

Could anyone please double check Chrome Beta (https://www.google.com/chrome/browser/beta.html)? If it's verified to be OK, the Chrome release may not be blocked, and we may even wait for response from Microsoft side. 

Dell 3268 desktop with Intel HD 630 and had the random freezing problem whilst on driver 23.20.16.4849  But since i went into device manager and forced a driver update to 23.20.16.4973 i have not experienced any freezing of chrome.  

Hi

I've tried the links for the high GPU websites and flicked between them for several minutes on Chrome stable and could not reproduce the issue.  I've already this morning had a freeze when using Chrome stable which accidentally launched from Visual Studio to run a website, the Win-Ctrl-Shift-B didn't work either, and locked my PC even worse and I had hard reset.

Without reproducing it reliably in Stable I can not of course confirm if BETA is fixed, although haven't experienced the issue so far while using the BETA version.

It's a frustrating bug, a shame Microsoft hasn't come into here to advise what the issue is.

@plee5509, thanks for continuously working with us! We really appreciate your help! 
According to my experience, we may have to open these websites with Edge too at the same time, and Chrome alone wasn't enough. That is, we need to run these 5 websites within 2 browsers simultaneously. 

Hi, I've got all the sites open in Chrome Stable and in Edge, been flicking about between the two, pulling them out into their own windows, moving tabs, working in Visual Studio, loading other websites and so far no lock up using Chrome Stable and I have confirmed I've re-enabled Hardware acceleration.  This has been running now around 10 minutes.

Task manager shows the GPU is at 99/100%, CPU at 31%, Memory at 28%, my system is responding slower to typing here in Chrome, but no freeze replicated.  It's an odd one this bug.

Another workaround: After the system hangs, we can still connect it with remote desktop. With remote connection, sometimes I saw Chrome was totally killed, sometimes I saw Chrome's GPU process crashed, and sometimes Chrome was still there with many tabs open (Never see the hang within remote connection). In situation the Chrome was still there with many tabs open, I could log on Windows again, and system would again hang there:)
Maybe this is helpful for debugging. 

 Issue 839256  has been merged into this issue.

yang.gu@: if you saw  GPU process crash sometimes, can you enable crash reporting and let us know the crash IDs? We can take a look at our crash database and maybe grab a stack there.

Same problem Facing with Yoga 920 Vibes - i7,8550
Every time have to do ctrl+win+shft+B

Random failures over here as well. 
If screensaver fires up or laptop goes to sleep then only a hard reset will work.
CRT+ALT+DEL or Win+Ctrl+Shift+B won't do anything at all. 
CTRL+SHIFT+ESC won’t work as well.

This bug seems to be triggered when computers enters on screen saver mode or when it goes to sleep. Once awake, fan spins at maximum speed and CRT+ALT+DEL or Win+Ctrl+Shift+B won't do anything at all. CTRL+SHIFT+ESC won’t fire up taskmgr and after some time it shuts down. 

Looks like it gets stuck in an infinite loop forever. 
Resembles the old Fork Bomb prank.

In Event Viewer just a power failure gets recorded. Maybe CPU’s thermal shutdown gets triggered somewhere or a triple fault is generated in the processor? Can’t really tell what’s going on. 
Win 10 1709 x64 works like a charm, so it must be a bug in 1803 release.

Buggy windows Version:
Name: Win10_1803_English_x64.iso
Size: 4692365312 bytes (4474 MB)
SHA1: 08FBB24627FA768F869C09F44C5D6C1E53A57A6F

	2018-05-04 17_07_43.png 61.0 KB View Download

gpu.pdf 138 KB Download

gpu.txt 14.6 KB View Download

Assigining to sunnyps@ for tracking progress.

Removing from sheriff queue since there are active owners.

Managed to grab a stack trace from dwm.exe (note: remote debugger needs to be run as admin):

Not Flagged   1360  0 Worker Thread DWM Compositor Thread win32u.dll!NtGdiDdDDIPresentMultiPlaneOverlay3
            win32u.dll!NtGdiDdDDIPresentMultiPlaneOverlay3()
            d3d11.dll!NDXGI::CDevice::PresentMultiplaneOverlay1CBImpl<struct DXGIDDI_MULTIPLANE_OVERLAY_PLANE_INFO>(void *,struct DXGIDDI_MULTIPLANE_OVERLAY_PLANE_INFO * *,unsigned int,void *)
            d3d11.dll!NDXGI::CDevice::PresentMultiplaneOverlayCB(void *,struct DXGIDDICB_PRESENT_MULTIPLANE_OVERLAY const *)
            igd10iumd64.dll!00007ffd37b30376()
            d3d11.dll!NDXGI::CDevice::PresentMultiplaneOverlay(struct IDXGIResource *,unsigned int,struct _DXGI_PRESENT_MULTIPLANE_OVERLAY const *,void *)
            dxgi.dll!CDXGISwapChain::PresentMultiplaneOverlayInternal(struct SPresentArgsCore<struct CDXGISwapChainWrapper> const *,enum DXGI_HDR_METADATA_TYPE,void const *,unsigned int,struct _DXGI_PRESENT_MULTIPLANE_OVERLAY const *)
            dxgi.dll!CDXGISwapChain::PresentMultiplaneOverlay(unsigned int,unsigned int,enum DXGI_HDR_METADATA_TYPE,void const *,unsigned int,struct _DXGI_PRESENT_MULTIPLANE_OVERLAY const *)
            d2d1.dll!DrawingContext::PresentMultiplaneOverlay(struct IDXGISwapChainDWM1 *,unsigned int,unsigned int,enum DXGI_HDR_METADATA_TYPE,void const *,struct _DXGI_PRESENT_MULTIPLANE_OVERLAY const *,unsigned int)
            d2d1.dll!D2DDeviceContextBase<struct ID2D1BitmapRenderTarget,struct ID2D1BitmapRenderTarget,struct ID2D1DeviceContext6>::PresentMultiplaneOverlay(struct IDXGISwapChainDWM1 *,unsigned int,unsigned int,enum DXGI_HDR_METADATA_TYPE,void const *,struct _DXGI_PRESENT_MULTIPLANE_OVERLAY const *,unsigned int)
            dwmcore.dll!CD2DContext::D2DPresentMultiplaneOverlay(struct IDXGISwapChainDWM1 *,unsigned int,unsigned int,enum DXGI_HDR_METADATA_TYPE,void const *,unsigned int,struct _DXGI_PRESENT_MULTIPLANE_OVERLAY const *)
            dwmcore.dll!CDWMSwapChain::PresentInternal()
            dwmcore.dll!CHwFullScreenRenderTarget::Present()
            dwmcore.dll!CHwndRenderTarget::InternalPresent()
            dwmcore.dll!CHwndRenderTarget::Present()
            dwmcore.dll!CRenderTargetManager::Present()
            dwmcore.dll!CComposition::Present()
            dwmcore.dll!CPartitionVerticalBlankScheduler::ProcessFrame()
            dwmcore.dll!CPartitionVerticalBlankScheduler::ScheduleAndProcessFrame()
            dwmcore.dll!CConnection::RunCompositionThread()
            kernel32.dll!BaseThreadInitThunk()
            ntdll.dll!RtlUserThreadStart()

Not Flagged > 1192  0 Main Thread Main Thread win32u.dll!NtUserGetMessage
            win32u.dll!NtUserGetMessage()
            user32.dll!GetMessageW()
            dwm.exe!CDwmAppHost::Run()
            dwm.exe!wWinMain()
            dwm.exe!__scrt_common_main_seh()
            kernel32.dll!BaseThreadInitThunk()
            ntdll.dll!RtlUserThreadStart()

Not Flagged   1316  0 Worker Thread dwmredir.dll!CPortBase::PortThread() dwmcore.dll!CChannel::WaitForNextMessage
            ntdll.dll!NtWaitForMultipleObjects()
            KernelBase.dll!WaitForMultipleObjectsEx()
            KernelBase.dll!WaitForMultipleObjects()
            dwmcore.dll!CChannel::WaitForNextMessage()
            dwmredir.dll!CPortBase::PortThreadInternal(void)
            dwmredir.dll!CPortBase::PortThread(void *)
            kernel32.dll!BaseThreadInitThunk()
            ntdll.dll!RtlUserThreadStart()

Not Flagged   1424  0 Worker Thread ntdll.dll!TppWorkerThread ntdll.dll!NtWaitForWorkViaWorkerFactory
            ntdll.dll!NtWaitForWorkViaWorkerFactory()
            ntdll.dll!TppWorkerThread()
            kernel32.dll!BaseThreadInitThunk()
            ntdll.dll!RtlUserThreadStart()

Not Flagged   1564  0 Worker Thread dwmcore.dll!CSurfaceManager::s_TokenThreadMain() win32u.dll!NtTokenManagerThread
Not Flagged   1568  0 Worker Thread dwmcore.dll!CMit::RunInputThreadStatic() CoreMessaging.dll!Microsoft::CoreUI::Dispatch::WaitArray::Callback_WaitAny
            ntdll.dll!NtWaitForMultipleObjects()
            KernelBase.dll!WaitForMultipleObjectsEx()
            CoreMessaging.dll!Microsoft::CoreUI::Dispatch::WaitArray::Callback_WaitAny()
            CoreMessaging.dll!Microsoft::CoreUI::Dispatch::WaitArray::Callback_DoWait()
            CoreMessaging.dll!Microsoft::CoreUI::Dispatch::WaitController::Callback_DoGeneralWait()
            CoreMessaging.dll!Microsoft::CoreUI::Dispatch::WaitController::Callback_OnDispatch()
            CoreMessaging.dll!Microsoft::CoreUI::Dispatch::EventLoop::Callback_RunCoreLoop()
            CoreMessaging.dll!Microsoft::CoreUI::Dispatch::EventLoop::Callback_Run()
            CoreMessaging.dll!Microsoft::CoreUI::Messaging::MessageSession$R::Microsoft__CoreUI__IExportMessageSession_Impl::Run(class System::Object *)
            CoreMessaging.dll!Microsoft::CoreUI::IExportMessageSession$X__ExportAdapter::Run(void)
            dwmcore.dll!CMit::Run()
            kernel32.dll!BaseThreadInitThunk()
            ntdll.dll!RtlUserThreadStart()

Not Flagged   1572  0 Worker Thread dwmcore.dll!CManipulationManager::s_ThreadMain() CoreMessaging.dll!Microsoft::CoreUI::Dispatch::WaitArray::Callback_WaitAny
            ntdll.dll!NtWaitForMultipleObjects()
            KernelBase.dll!WaitForMultipleObjectsEx()
            CoreMessaging.dll!Microsoft::CoreUI::Dispatch::WaitArray::Callback_WaitAny()
            CoreMessaging.dll!Microsoft::CoreUI::Dispatch::WaitArray::Callback_DoWait()
            CoreMessaging.dll!Microsoft::CoreUI::Dispatch::WaitController::Callback_DoGeneralWait()
            CoreMessaging.dll!Microsoft::CoreUI::Dispatch::WaitController::Callback_OnDispatch()
            CoreMessaging.dll!Microsoft::CoreUI::Dispatch::EventLoop::Callback_RunCoreLoop()
            CoreMessaging.dll!Microsoft::CoreUI::Dispatch::EventLoop::Callback_Run()
            CoreMessaging.dll!Microsoft::CoreUI::Messaging::MessageSession$R::Microsoft__CoreUI__IExportMessageSession_Impl::Run(class System::Object *)
            CoreMessaging.dll!Microsoft::CoreUI::IExportMessageSession$X__ExportAdapter::Run(void)
            dwmcore.dll!CManipulationManager::ManipulationThreadMain()
            dwmcore.dll!CManipulationManager::s_ThreadMain(void *)
            kernel32.dll!BaseThreadInitThunk()
            ntdll.dll!RtlUserThreadStart()

Not Flagged   1652  0 Worker Thread uDWM.dll!CDesktopManager::DwmEventThreadProc  win32u.dll!NtUserMsgWaitForMultipleObjectsEx
            win32u.dll!NtUserMsgWaitForMultipleObjectsEx()
            user32.dll!RealMsgWaitForMultipleObjectsEx()
            uDWM.dll!CDesktopManager::DwmEventThreadProc()
            kernel32.dll!BaseThreadInitThunk()
            ntdll.dll!RtlUserThreadStart()

Not Flagged   1728  0 Worker Thread Windows.Gaming.Input.dll!ProviderManagerWorker::WorkerThreadProcThunk()  Windows.Gaming.Input.dll!ProviderManagerWorker::WorkerThreadProc
            ntdll.dll!NtDelayExecution()
            KernelBase.dll!SleepEx()
            Windows.Gaming.Input.dll!ProviderManagerWorker::WorkerThreadProc(void)
            Windows.Gaming.Input.dll!ProviderManagerWorker::WorkerThreadProcThunk(void *)
            kernel32.dll!BaseThreadInitThunk()
            ntdll.dll!RtlUserThreadStart()

I can reproduce this prob. on my Intel 7th gen NUC with Version 66.0.3359.139 Chrome and even Intel Day0 Spring update display driver.  Disable/enable hardware acceleration does not solve the problem.

Mouse is moving, mouse buttons not working, keyboard mostly not working (except CapsLock, brightness), Ctrl-Alt-Del not working.

Someone mentioned to remote login.  I tried teamviewer but I just get a black screen.

There appear to be 2 failure modes after the 1803 update.  One involves GPU functionality and is being addressed here.  The other involves CryptoServices and is being addressed in bug 838707.

The workarounds for the GPU problem
 --disable-direct-composition-overlays
 disable hardware accelleration
do not fix the crypto problem.

Stop/restart cryptoSvcs does (temporarily?) fix the crypto problem,
as well as a global permissions change to the registry.
Net-export logs are available in 838707.
The crypto problem is reproducible (on user PCs).
I believe it persists on Canary and Beta.
Running Chrome from another Windows account (that didn't install 1803) also works.

There is a request on 838707 to retain that CR for crypto issues.

More later..

Larry I still have the problem after both hardware acceleration and direct-composition are disabled.


Graphics Feature Status
Canvas: Software only, hardware acceleration unavailable
CheckerImaging: Disabled
Flash: Software only. Hardware acceleration disabled
Flash Stage3D: Software only. Hardware acceleration disabled
Flash Stage3D Baseline profile: Software only. Hardware acceleration disabled
Compositing: Software only. Hardware acceleration disabled
Multiple Raster Threads: Enabled
Native GpuMemoryBuffers: Software only. Hardware acceleration disabled
Rasterization: Software only. Hardware acceleration disabled
Surface Synchronization: Enabled
Video Decode: Software only. Hardware acceleration disabled
WebGL: Software only, hardware acceleration unavailable
WebGL2: Software only, hardware acceleration unavailable
Problems Detected
Native GpuMemoryBuffers have been disabled, either via about:flags or command line.
Disabled Features: native_gpu_memory_buffers
Checker-imaging has been disabled via finch trial or the command line.
Disabled Features: checker_imaging
Version Information
Data exported	2018-05-05T20:00:44.822Z
Chrome version	Chrome/66.0.3359.139
Operating system	Windows NT 10.0.17134
Software rendering list URL	https://chromium.googlesource.com/chromium/src/+/a020eddf0d85fe84d4a6787b304f50aafb670969/gpu/config/software_rendering_list.json
Driver bug list URL	https://chromium.googlesource.com/chromium/src/+/a020eddf0d85fe84d4a6787b304f50aafb670969/gpu/config/gpu_driver_bug_list.json
ANGLE commit id	22c768fbda54
2D graphics backend	Skia/66 773868fdade5f9f0e7697e6d09c9bd80aaa9b402-
Command Line	"C:\Program Files (x86)\Google\Chrome\Application\chrome.exe" --flag-switches-begin --flag-switches-end
Driver Information
Initialization time	111
In-process GPU	false
Passthrough Command Decoder	false
Direct Composition	false
Supports overlays	false
Sandboxed	true
GPU0	VENDOR = 0x8086, DEVICE= 0x5926 *ACTIVE*
Optimus	false
Optimus	false
AMD switchable	false
Desktop compositing	Aero Glass
Diagonal Monitor Size of \\.\DISPLAY2	26.9"
Diagonal Monitor Size of \\.\DISPLAY1	23.7"
Driver vendor	Intel Corporation
Driver version	24.20.100.6025
Driver date	4-25-2018
Pixel shader version	1.00
Vertex shader version	1.00
Max. MSAA samples	4
Machine model name	
Machine model version	
GL_VENDOR	Google Inc.
GL_RENDERER	Google SwiftShader
GL_VERSION	OpenGL ES 2.0 SwiftShader 4.0.0.3
GL_EXTENSIONS	GL_OES_compressed_ETC1_RGB8_texture GL_OES_depth24 GL_OES_depth32 GL_OES_depth_texture GL_OES_depth_texture_cube_map GL_OES_EGL_image GL_OES_EGL_image_external GL_OES_EGL_sync GL_OES_element_index_uint GL_OES_framebuffer_object GL_OES_packed_depth_stencil GL_OES_rgb8_rgba8 GL_OES_standard_derivatives GL_OES_surfaceless_context GL_OES_texture_float GL_OES_texture_float_linear GL_OES_texture_half_float GL_OES_texture_half_float_linear GL_OES_texture_npot GL_OES_texture_3D GL_OES_vertex_half_float GL_EXT_blend_minmax GL_EXT_color_buffer_half_float GL_EXT_draw_buffers GL_EXT_instanced_arrays GL_EXT_occlusion_query_boolean GL_EXT_read_format_bgra GL_EXT_texture_compression_dxt1 GL_EXT_texture_filter_anisotropic GL_EXT_texture_format_BGRA8888 GL_EXT_texture_rg GL_ARB_texture_rectangle GL_ANGLE_framebuffer_blit GL_ANGLE_framebuffer_multisample GL_ANGLE_instanced_arrays GL_ANGLE_texture_compression_dxt3 GL_ANGLE_texture_compression_dxt5 GL_APPLE_texture_format_BGRA8888 GL_CHROMIUM_color_buffer_float_rgba GL_CHROMIUM_texture_filtering_hint GL_NV_fence GL_NV_framebuffer_blit GL_NV_read_depth GL_EXT_color_buffer_float
Disabled Extensions	
Disabled WebGL Extensions	
Window system binding vendor	Google Inc.
Window system binding version	1.4 SwiftShader 4.0.0.3
Window system binding extensions	EGL_KHR_create_context EGL_KHR_get_all_proc_addresses EGL_KHR_gl_texture_2D_image EGL_KHR_gl_texture_cubemap_image EGL_KHR_gl_renderbuffer_image EGL_KHR_fence_sync EGL_KHR_image_base EGL_KHR_surfaceless_context EGL_ANDROID_framebuffer_target EGL_ANDROID_recordable
Direct rendering	Yes
Reset notification strategy	0x0000
GPU process crash count	0
Compositor Information
Tile Update Mode	One-copy
Partial Raster	Enabled
GpuMemoryBuffers Status
ATC	Software only
ATCIA	Software only
DXT1	Software only
DXT5	Software only
ETC1	Software only
R_8	Software only
R_16	Software only
RG_88	Software only
BGR_565	Software only
RGBA_4444	Software only
RGBX_8888	GPU_READ, SCANOUT
RGBA_8888	GPU_READ, SCANOUT
BGRX_8888	Software only
BGRX_1010102	Software only
RGBX_1010102	Software only
BGRA_8888	Software only
RGBA_F16	Software only
YVU_420	Software only
YUV_420_BIPLANAR	Software only
UYVY_422	Software only
Display(s) Information
Info	Display[2528732444] bounds=0,0 1920x1080, workarea=0,0 1920x1040, scale=1, external
Color space information	{primaries:BT709, transfer:IEC61966_2_1, matrix:RGB, range:FULL}
Bits per color component	8
Bits per pixel	24
Info	Display[2779098405] bounds=1920,0 2560x1440, workarea=1920,0 2560x1400, scale=1, external
Color space information	{primaries:BT709, transfer:IEC61966_2_1, matrix:RGB, range:FULL}
Bits per color component	8
Bits per pixel	24
Video Acceleration Information
Encode h264 baseline	up to 3840x2176 pixels and/or 30.000 fps
Encode h264 main	up to 3840x2176 pixels and/or 30.000 fps
Encode h264 high	up to 3840x2176 pixels and/or 30.000 fps
Diagnostics
0
b3DAccelerationEnabled	true
b3DAccelerationExists	true
bAGPEnabled	true
bAGPExistenceValid	true
bAGPExists	true
bCanRenderWindow	true
bDDAccelerationEnabled	true
bDriverBeta	false
bDriverDebug	false
bDriverSigned	false
bDriverSignedValid	false
bNoHardware	false
dwBpp	32
dwDDIVersion	12
dwHeight	1080
dwRefreshRate	59
dwWHQLLevel	0
dwWidth	1920
iAdapter	0
lDriverSize	2037704
lMiniVddSize	0
szAGPStatusEnglish	Enabled
szAGPStatusLocalized	Enabled
szChipType	Intel(R) Iris(R) Graphics Family
szD3DStatusEnglish	Enabled
szD3DStatusLocalized	Enabled
szDACType	Internal
szDDIVersionEnglish	12
szDDIVersionLocalized	12
szDDStatusEnglish	Enabled
szDDStatusLocalized	Enabled
szDXVAHDEnglish	Supported
szDXVAModes	ModeMPEG2_A ModeMPEG2_C ModeWMV9_C ModeVC1_C
szDescription	Intel(R) Iris(R) Plus Graphics 640
szDeviceId	0x5926
szDeviceIdentifier	{D7B78E66-1A66-11CF-BFF4-0C00BDC2D535}
szDeviceName	\\.\DISPLAY1
szDisplayMemoryEnglish	8264 MB
szDisplayMemoryLocalized	8264 MB
szDisplayModeEnglish	1920 x 1080 (32 bit) (59Hz)
szDisplayModeLocalized	1920 x 1080 (32 bit) (59Hz)
szDriverAssemblyVersion	24.20.100.6025
szDriverAttributes	Final Retail
szDriverDateEnglish	4/25/2018 8:00:00 AM
szDriverDateLocalized	4/25/2018 08:00:00
szDriverLanguageEnglish	English
szDriverLanguageLocalized	English
szDriverModelEnglish	WDDM 2.4
szDriverModelLocalized	WDDM 2.4
szDriverName	C:\Windows\System32\DriverStore\FileRepository\igdlh64.inf_amd64_0707757077710fff\igdumdim64.dll,C:\Windows\System32\DriverStore\FileRepository\igdlh64.inf_amd64_0707757077710fff\igd10iumd64.dll,C:\Windows\System32\DriverStore\FileRepository\igdlh64.inf_amd64_0707757077710fff\igd10iumd64.dll,C:\Windows\System32\DriverStore\FileRepository\igdlh64.inf_amd64_0707757077710fff\igd12umd64.dll
szDriverNodeStrongName	oem3.inf:5f63e534fc3bd056:iKBLD_w10_DS:24.20.100.6025:pci\ven_8086&dev_5926
szDriverSignDate	Unknown
szDriverVersion	24.20.0100.6025
szKeyDeviceID	Enum\PCI\VEN_8086&DEV_5926&SUBSYS_20688086&REV_06
szKeyDeviceKey	\Registry\Machine\System\CurrentControlSet\Control\Video\{D8BE90AF-4EF9-11E8-91E0-A8893D0D92E9}\0000
szManufacturer	Intel Corporation
szMiniVdd	unknown
szMiniVddDateEnglish	Unknown
szMiniVddDateLocalized	unknown
szMonitorMaxRes	Unknown
szMonitorName	Generic PnP Monitor
szNotesEnglish	No problems found.
szNotesLocalized	No problems found.
szOverlayEnglish	Supported
szRankOfInstalledDriver	00D12001
szRegHelpText	Unknown
szRevision	Unknown
szRevisionId	0x0006
szSubSysId	0x20688086
szTestResultD3D7English	Not run
szTestResultD3D7Localized	Not run
szTestResultD3D8English	Not run
szTestResultD3D8Localized	Not run
szTestResultD3D9English	Not run
szTestResultD3D9Localized	Not run
szTestResultDDEnglish	Not run
szTestResultDDLocalized	Not run
szVdd	unknown
szVendorId	0x8086
1
b3DAccelerationEnabled	true
b3DAccelerationExists	true
bAGPEnabled	true
bAGPExistenceValid	true
bAGPExists	true
bCanRenderWindow	true
bDDAccelerationEnabled	true
bDriverBeta	false
bDriverDebug	false
bDriverSigned	false
bDriverSignedValid	false
bNoHardware	false
dwBpp	32
dwDDIVersion	12
dwHeight	1440
dwRefreshRate	59
dwWHQLLevel	0
dwWidth	2560
iAdapter	1
lDriverSize	2037704
lMiniVddSize	0
szAGPStatusEnglish	Enabled
szAGPStatusLocalized	Enabled
szChipType	Intel(R) Iris(R) Graphics Family
szD3DStatusEnglish	Enabled
szD3DStatusLocalized	Enabled
szDACType	Internal
szDDIVersionEnglish	12
szDDIVersionLocalized	12
szDDStatusEnglish	Enabled
szDDStatusLocalized	Enabled
szDXVAHDEnglish	Supported
szDXVAModes	ModeMPEG2_A ModeMPEG2_C ModeWMV9_C ModeVC1_C
szDescription	Intel(R) Iris(R) Plus Graphics 640
szDeviceId	0x5926
szDeviceIdentifier	{D7B78E66-1A66-11CF-BFF4-0C00BDC2D535}
szDeviceName	\\.\DISPLAY2
szDisplayMemoryEnglish	8264 MB
szDisplayMemoryLocalized	8264 MB
szDisplayModeEnglish	2560 x 1440 (32 bit) (59Hz)
szDisplayModeLocalized	2560 x 1440 (32 bit) (59Hz)
szDriverAssemblyVersion	24.20.100.6025
szDriverAttributes	Final Retail
szDriverDateEnglish	4/25/2018 8:00:00 AM
szDriverDateLocalized	4/25/2018 08:00:00
szDriverLanguageEnglish	English
szDriverLanguageLocalized	English
szDriverModelEnglish	WDDM 2.4
szDriverModelLocalized	WDDM 2.4
szDriverName	C:\Windows\System32\DriverStore\FileRepository\igdlh64.inf_amd64_0707757077710fff\igdumdim64.dll,C:\Windows\System32\DriverStore\FileRepository\igdlh64.inf_amd64_0707757077710fff\igd10iumd64.dll,C:\Windows\System32\DriverStore\FileRepository\igdlh64.inf_amd64_0707757077710fff\igd10iumd64.dll,C:\Windows\System32\DriverStore\FileRepository\igdlh64.inf_amd64_0707757077710fff\igd12umd64.dll
szDriverNodeStrongName	oem3.inf:5f63e534fc3bd056:iKBLD_w10_DS:24.20.100.6025:pci\ven_8086&dev_5926
szDriverSignDate	Unknown
szDriverVersion	24.20.0100.6025
szKeyDeviceID	Enum\PCI\VEN_8086&DEV_5926&SUBSYS_20688086&REV_06
szKeyDeviceKey	\Registry\Machine\System\CurrentControlSet\Control\Video\{D8BE90AF-4EF9-11E8-91E0-A8893D0D92E9}\0001
szManufacturer	Intel Corporation
szMiniVdd	unknown
szMiniVddDateEnglish	Unknown
szMiniVddDateLocalized	unknown
szMonitorMaxRes	Unknown
szMonitorName	Generic PnP Monitor
szNotesEnglish	No problems found.
szNotesLocalized	No problems found.
szOverlayEnglish	Supported
szRankOfInstalledDriver	00D12001
szRegHelpText	Unknown
szRevision	Unknown
szRevisionId	0x0006
szSubSysId	0x20688086
szTestResultD3D7English	Not run
szTestResultD3D7Localized	Not run
szTestResultD3D8English	Not run
szTestResultD3D8Localized	Not run
szTestResultD3D9English	Not run
szTestResultD3D9Localized	Not run
szTestResultDDEnglish	Not run
szTestResultDDLocalized	Not run
szVdd	unknown
szVendorId	0x8086
Log Messages
GpuProcessHostUIShim: The GPU process exited normally. Everything is okay.

cheungts@ - Try services restart CryptSvc . If this resolves the problem, at least temporarily, let us know and tag in at bug 838707

Correction: The GPU fix is --disable-direct-composition
  not --diable-direct-composition-overlays (see my C#66)

FYI: At least the CryptSvc bug appears to depend on the Windows account that installed 1803.  For repro, this may be an issue.  See Bug 838707 comments also re: registry permissions.

Hi Larry,

How to -disable-direct-composition ?

Thanks.

cheugnts@ - Did task manager> Services> restart CryptSvc help?

Add --disable-direct-composition in the Target field of a Windows shortcut to launch Chrome.  Note the double '-' and the space after '"'.  Open the shortcut with right-click> properties, to access the Target field. It should look like this  ...chrome.exe" --disable-direct-composition

Use about:version to verify startup options.

Larry,

That fixed my chrome problem but I get another problem which not happened before.

I connect hdmi to my 24" Dell montior which also carries the audio via the analog out on the Dell.  After wake up from sleep DUE TO INACTIVITY, audio is gone (can't find it even in the mixer).

It doesnt do it if I manually make my NUC sleep and wake it up later though.

@cheugnts: Please confirm: 
  --disable-direct-composition got Chrome running again
and
  task manager> Services> restart CryptSvc - did not help

Please followup on the audio problem on the forum main thread here
  https://productforums.google.com/forum/#!topic/chrome/bzZPHqKXODw
or on this one which mentions sleep (after the 1803 upgrade)
  https://productforums.google.com/forum/#!topic/chrome/sD99ucKqA0c

 Issue 839940  has been merged into this issue.

Hi, All,
Here are some new findings today :
1. We've caught the freeze issue(screen freeze and cannot recovery for long time) on chrome 66 stable and use Windows Remote Desktop Connection to get the uploaded crash reports, the ID is 3d592dfa-27ff-4cea-a53e-c6659ebf1d52
The GPU page is attached below.

2. Chrome 67 Beta has the similar freeze issue but could recovery itself for a while. The chrome://gpu page shows GPU process exited successfully.

3. We find playing 8k video and aquarium with 30000 fishes at the same time will increase the hit chance.

gpu.html 125 KB View Download

*** Bulk Edit ***
M67 Stable promotion is coming VERY soon. Your bug is labelled as Stable ReleaseBlock, pls make sure to land the fix and request a merge into the release branch ASAP. 

If fix is already merged to M67 and nothing else is pending, pls mark the bug as fixed. Thank you.

shaobo@ - There are 2 Win update 1803 failure modes, see c#66.  
They can be distinguished by their workarounds and CryptSvc usage.
If you're reporting a CryptSvc problem, please update bug 838707

larry@ and all:
Double check my situation and confirm they are not CryptSvc problem. Because when the screen is frozen, I cannot call task manager out at all.
About --disable-direct-compositor flags:
1. Chrome 66 Stable with this flag hit the same freeze issue but could recovery for a while(about 1 min), no crash reports generated, and chrome://gpu shows GPU process exited and restart.
2. Chrome 67 beta with the flag still has the same behavior. It hit freeze issue but could recovery for a while (about 1 min), no crash reports generated and chrome://gpu shows GPU process exited and restart. This gpu page is attached below

More infos pls see c#76 

gpu.html 125 KB View Download

shaobo@ - Please check the spelling of --disable-direct-composition . The last character is an 'n'.

Oops, thx larry@ for the correction but the result is the same when I use "--disable-direct-composition" and it seems not a CryptSvc. I've done some more tests on coffeelake(UHD630) and kabylake(HD630) today.

To summarize all of the findings today and yesterday:

1. We've caught the freeze issue(screen freeze and cannot recovery for long time) on chrome 66 stable on both kabylake(HD630) and coffeelake(UHD630) and use Windows Remote Desktop Connection to get the uploaded crash reports, the ID is 3d592dfa-27ff-4cea-a53e-c6659ebf1d52(coffeelake) and fc0ff588-7586-4703-af5f-76588a1143a6(kabylake) 
The GPU page is attached as "gpu-stable-kabylake.html" and "gpu-stable-coffeelake.html".

2. Chrome 67 Beta has the similar freeze issue but could recovery itself for a while. The chrome://gpu page shows GPU process exited successfully.

3. By adding "--disable-direct-composition", Chrome 66 stable hit the same freeze issue but could recovery for a while(about 1 min), no crash report generated. Chrome version file is attached as "About Version-stable-with-flag.html", About GPU page is attached as "gpu-stable-with-flag.html".

4. adding "--disable-direct-composition" flag has not effect with Chrome 67 beta. It hit freeze issue but could recovery for a while(about 1 min), no crash report generated. Chrome version file is attached as "About Version-beta-with-flag", About GPU page is attached as "gpu-beta-with-flag.html".

5. We find playing 8k video and aquarium with 30000 fishes at the same time will increase the hit chance.

gpu-stable-kabylake.html 124 KB View Download

gpu-stable-coffeelake.html 125 KB View Download

About Version-stable-with-flag.html 4.5 KB View Download

gpu-stable-with-flag.html 122 KB View Download

About Version-beta-with-flag.html 4.4 KB View Download

gpu-beta-with-flag.html 125 KB View Download

shaobo.yan@intel.com: the crash IDs you provided are local crash IDs. When they get uploaded to the server, you will get another set of hash IDs. Can you share them?

For example, something like

Uploaded Crash Report ID 9feb38954594cb0d (Local Crash ID: e03b9c86-552c-49c8-9173-61ee88b22332)

and we need 9feb38954594cb0d

Wondering if the May patch/update fixed the chrome related problems.
The changelog did not specifically mentioned about the graphic bugs.

Thx.

Seems this bug is fixed in latest public Windows Update (1803 build 17134.48).

Thanks, #84! If other users could confirm, we will close this bug.

The 2018-05 Cumulative Update KB4103721, updated my Win10 Home 17134.1 to 17134.48, i.e. .48 is the May monthly update.  As of yesterday 5/7 the 1803 upgrade was still 17134.1.  Looks like 5/8 is again an update Tuesday.

Hi Larry,

Seems resolved my chrome bug and audio disappeared after sleep bug(I think it reset the audio device because I see it initally still has the red cross and then 2 sec later it comes back).

Will report back as only testing in the past 2hr or so.

Btw, wondering will there be a new version of chrome coming soon?

Thx.

zmo@, my crash file pending on the state "upload requested by user, not yet uploaded " so I attach the dmp file in attachments.

sunnyps@, I've updated to 17134.48, and I do the tests about 1hr on both coffeelake and kabylake with chrome stable. No freeze issue hitted. I think the update has fixed this bug.

3d592dfa-27ff-4cea-a53e-c6659ebf1d52.dmp 628 KB Download

fc0ff588-7586-4703-af5f-76588a1143a6.dmp 611 KB Download

17134.48 update has fixed this freezing issue on my 7700HQ Win10 Home Aero15. I can use Chromium apps again!

Looking at 3d592dfa-27ff-4cea-a53e-c6659ebf1d52.dmp, crashing thread is gpu
watchdog thread.

The main thread has this stack frame in D3D:
d3d11!NDXGI::CDevice::WaitForSynchronizationObjectFromCpuCB+0x72

Then we have stack frames in Intel driver, but it doesn't seem to have
frame pointers. Reverse stack walk in WinDBG wasn't very useful. I see a
bunch of Chrome symbols but it's not clear which are part of the stack.

Intel folks: you can probably symbolize with debug symbols for the driver.
Also, do you know if you have frame pointers enabled? If not, can you
please enable them for future driver releases?

There's a video decoder thread too with this stack trace:
00 ntdll!NtWaitForAlertByThreadId
01 ntdll!RtlpWaitOnAddressWithTimeout
02 ntdll!RtlpWaitOnCriticalSection
03 ntdll!RtlpEnterCriticalSectionContended
04 ntdll!RtlEnterCriticalSection
05 d3d11!CDeviceChild<ID3D11DeviceContext4>::CondObjectLock::CondObjectLock
06 d3d11!CContext::ReleaseDecoderBuffer
07 MSVP9DEC
08 MSVP9DEC
09 MSVP9DEC
0a MSVP9DEC!DllGetActivationFactory
0b MSVP9DEC!DllGetActivationFactory
0c MSVP9DEC!DllGetActivationFactory
0d MSVP9DEC!DllGetActivationFactory
0e MSVP9DEC!DllGetActivationFactory
0f MSVP9DEC!DllGetActivationFactory
10 MSVP9DEC
11 MSVP9DEC
12 MSVP9DEC
13 chrome_child!media::DXVAVideoDecodeAccelerator::DoDecode
14 chrome_child!media::DXVAVideoDecodeAccelerator::DecodeInternal
15 chrome_child!media::DXVAVideoDecodeAccelerator::DecodePendingInputBuffers
16 chrome_child!base::OnceCallback<void ()>::Run
17 chrome_child!base::debug::TaskAnnotator::RunTask
18 chrome_child!base::MessageLoop::RunTask
19 chrome_child!base::MessageLoop::DeferOrRunPendingTask
1a chrome_child!base::MessageLoop::DoWork
1b chrome_child!base::MessagePumpDefault::Run
1c chrome_child!base::RunLoop::Run
1d chrome_child!base::Thread::ThreadMain
1e chrome_child!base::`anonymous namespace'::ThreadFunc
1f KERNEL32!BaseThreadInitThunk
20 ntdll!RtlUserThreadStart

fc0ff588-7586-4703-af5f-76588a1143a6.dmp is similar in that the watchdog
triggers. However, the main thread has this stack trace:

00 ntdll!NtWaitForAlertByThreadId
01 ntdll!RtlpWaitOnAddressWithTimeout
02 ntdll!RtlpWaitOnCriticalSection
03 ntdll!RtlpEnterCriticalSectionContended
04 ntdll!RtlEnterCriticalSection
05
d3d11!CLayeredObjectRoot<ID3D11LayeredDevice>::CondObjectLock::CondObjectLock
06 d3d11!CDevice::CondObjectLock::CondObjectLock
07 d3d11!CContext::TID3D11DeviceContext_ClearView_<2>
08 libglesv2!rx::Clear11::clearFramebuffer
09 libglesv2!gl::Error::isError
0a libglesv2!rx::Framebuffer11::clearImpl
0b libglesv2!rx::FramebufferD3D::clear
0c libglesv2!gl::Framebuffer::clear
0d libglesv2!gl::Context::clear
0e chrome_child!gpu::gles2::GLES2DecoderImpl::DoClear
0f chrome_child!gpu::gles2::GLES2DecoderImpl::HandleClear
10 chrome_child!gpu::gles2::GLES2DecoderImpl::DoCommandsImpl<0>
11 chrome_child!gpu::gles2::GLES2DecoderImpl::DoCommands
12 chrome_child!gpu::CommandBufferService::Flush
13 chrome_child!gpu::CommandBufferStub::OnAsyncFlush
14 chrome_child!base::DispatchToMethodImpl
15 chrome_child!base::DispatchToMethod
16 chrome_child!IPC::DispatchToMethod
17
chrome_child!IPC::MessageT<GpuCommandBufferMsg_AsyncFlush_Meta,std::tuple<int,unsigned
int,bool>,void>::Dispatch<gpu::CommandBufferStub,gpu::CommandBufferStub,void,void
(gpu::CommandBufferStub::*)(int, unsigned int, bool)>
18 chrome_child!gpu::CommandBufferStub::OnMessageReceived
19 chrome_child!IPC::MessageRouter::RouteMessage
1a chrome_child!gpu::GpuChannel::HandleMessageHelper
1b chrome_child!gpu::GpuChannel::HandleMessage
1c chrome_child!base::OnceCallback<void ()>::Run
1d chrome_child!gpu::Scheduler::RunNextTask
1e chrome_child!base::OnceCallback<void ()>::Run
1f chrome_child!base::debug::TaskAnnotator::RunTask
20 chrome_child!base::MessageLoop::RunTask
21 chrome_child!base::MessageLoop::DeferOrRunPendingTask
22 chrome_child!base::MessageLoop::DoWork
23 chrome_child!base::MessagePumpDefault::Run
24 chrome_child!base::RunLoop::Run
25 chrome_child!content::GpuMain
26 chrome_child!content::RunNamedProcessTypeMain
27 chrome_child!content::ContentMainRunnerImpl::Run
28 chrome_child!service_manager::Main
29 chrome_child!content::ContentMain
2a chrome_child!ChromeMain
2b chrome!MainDllLoader::Launch
2c chrome!wWinMain
2d chrome!invoke_main
2e chrome!__scrt_common_main_seh
2f KERNEL32!BaseThreadInitThunk
30 ntdll!RtlUserThreadStart

And there's a thread with the same DXGI signature:

00 win32u!NtGdiDdDDIWaitForSynchronizationObjectFromCpu
01 d3d11!NDXGI::CDevice::WaitForSynchronizationObjectFromCpuCB
02 igd11dxva64
03 0x0
04 d3d11!SDeviceWriteTracking::SDeviceWriteTracking

Reverse stack walk is not helpful here too.

Hi

Just to add the 17134.48 update has fixed it for me as well, no crashed in Chrome stable for the last 9 hours.

Thanks for confirming!

Hi, I´m installed 17134.48 update but not fixed it for me, chrome version 66.0.3359.139 (build oficial)

I have problem to sign in with my e-mail account.


See the pictures attached

Thnkxs!

	picture 2.JPG 23.7 KB View Download

picture 1 

sorry 

	picture1.JPG 31.6 KB View Download

#94/95: That is a separate issue. This issue is for when all of Windows desktop hangs when using Chrome. Please contact the website owner if you're seeing issues with a particular website. If you think the bug is in Chrome, please file a new bug.

martin #94/95: Please submit/look on the Chrome Help Forum for related failures. The blank box is the Signin dialog.
Forum: 
https://productforums.google.com/forum/#!categories/chrome
o mira aqui
https://productforums.google.com/forum/#!forum/chrome-es

have positive result with 17134.48, using Version 66.0.3359.139 (Official Build) (64-bit)

im on version 1803(os build 17134.48), and my chrome still hangs and sometimes my whole system will hang aswell, i have to restart to use my computer back. Anyone know how to fix this? i thought windows update KB4103721 would fix this hangs. Even updated my chrome version to the latest [(Version 66.0.3359.181 (Official Build) (64-bit)]

Black@ There are other 1803 failure modes involving CryptSvc running hot or blocking.  See bug 838707 for details and followup there.

@larrylac is the a fix yet? i contacted windows live support and they asked me to install latest nvidia drivers so far its fine.(no hangs on chrome/or desktop freeze)

HP Spectre 13 af002tu
WIN10 17134.48
Windows freezes even I don't use chrome.
!!!
Microsoft is disappointing!!

Hi,

I can confirm to still have these problems on my laptop
after installing the April update.  Before April update the laptop ran 100% butter smooth, no issues, no crashes, 10 hours a day straight for months!

I have updated the following:

Google Chrome is up-to-date
Versie 68.0.3440.84 (Officiële build) (64-bits)

Install latest nvida Game ready drivers 398.82

Using Intel HD 24.20.100.6194 drivers (note it also happens using the sotck drivers from manufacturer)

Chrome was running with Youtube playing music.

Today it froze the mouse and keyboard again. Music continued playing for ~15 seconds and thereafter it totally frooze.

Had to close lid of the laptop to be able to continue work again:

Operating System
	Windows 10 Pro 64-bit
CPU
	Intel Core i7 7700HQ @ 2.80GHz	54 °C
	Kaby Lake 14nm Technology
RAM
	8,00GB Single-Channel Unknown @ 1197MHz (16-16-16-39)
Motherboard
	Notebook N85_N87HJHJ1HK1 (U3E1)
Graphics
	Generic PnP Monitor (1920x1080@59Hz)
	Acer S231HL (1920x1080@59Hz)
	Intel HD Graphics 630 (CLEVO/KAPOK Computer)
	4095MB NVIDIA GeForce GTX 1050 Ti (CLEVO/KAPOK Computer)	40 °C
	ForceWare version: 398.82
	SLI Disabled
Storage
	489GB Crucial_CT525MX300SSD1 (SSD)	34 °C


Is there some automatic logging in chrome?
Maybe I can log thins for you guys and send the report here.

I really think its Chrome because yesterday it happened as well but 3x times short in a row (after I openend chrome again)  after the first time happened that day.

Today immediately after it happened first time and closed/ openend lid, I ended chrome via taskbar, made sure it was not running and it did NOT directly happen again. 


Hope this helps.

Here is my chrome://gpu  result attached

gpu.html 180 KB View Download

ppvd: This CR was resolved with Windows update to 17134.48 back in May.  Your comments will likely be ignored.  

You may open a new issue here
https://crbug.com/new

Hi, Thanks for the reply

Installing 17134.48 did not help to solve the problem!
I installed the April update in Juli so it had the 17134.48 included.
I cannot install it as seperate update cause I think I aready got it.

Issue 872224 has been merged into this issue.

Please check issue https://bugs.chromium.org/p/chromium/issues/detail?id=872224#c4

its not the SAME issue!!

Please undup 872224. 

I am having this issue too. 
DELL G3 3579
Intel core i7-8750H@2.20GHz
Intel HD Graphics 630 (Driver: 24.20.100.6286)
4095MB NVIDIA GeForce GTX 1050 Ti (Driver:399.24)
Windows 10 Home Version 1803  Build 17134.285

I have noticed this problem when visiting pages with videos and doing a scroll (eg nfl.com)

@conta

Sorry, but comments on 'Wont fix' pages are generally ignored, particularly this one which was for a graphics driver problem fixed with the Windows 17134.48 update long ago.

Turning off hardware acceleration may prevent the freeze.

Pls post elsewhere and include your Chrome version.

Sorry..

Is this issue fixed in chrome 69?

CR's target a specific cause for a general symptom.  The root cause for 'this' specific graphics problem was resolved long ago.

If you are still experiencing a similar symptom, please look for another (not Wont Fix) CR or open a new one:
  https://crbug.com/new

See also bug 838707 also related to a freeze in Establishing Secure Connection, introduced with the April Windows 1803 feature update.

838707 is caused by a CryptSvc problem with the HKCU ProtectedRoots key on isolated machines.  A registry repair is available here
https://productforums.google.com/forum/#!msg/chrome/s5S1uPI0kMc/b2Mud3SKCgAJ