Kevin in 9693.0.0,61.0.3142.0. 

Recovery via usb stick is successful. 

However after taking out the usb stick, splash screen shows, then it goes black. Mouse cursor can be seen. VT2 works. 

No workaround
Rebooting the device will not resolve the issue.

I am noticing this on peach_pit as well. After getting to the boot screen with the Chrome logo, it freezes there for a long time. If you go to the terminal and try to go back to the UI, the screen is not redrawn and the cursor appears (over the terminal).

Recovery works, but as mentioned in #2, rebooting doesn't fix the issue.

Try changing the channel to beta from VT2? I was able to avoid further issues by doing that on my kevin.

log:
https://pantheon.corp.google.com/storage/browser/chromiumos-test-logs/bugfiles/cr/737707

Seen black screen after AU to M61-9693.0.0 Kevin.

Only graphix test indication any new issue is graphics_VTSwitch on suites -
- graphics - https://screenshot.googleplex.com/2i7RuCoUhj7
- graphics_bvt - https://screenshot.googleplex.com/2ycXXkAQvCZ

Comparing with other boards I see several others in trouble at 
https://wmatrix.googleplex.com/unfiltered?tests=graphics_VTSwitch&releases=tot&hide_missing=True

Issue starts with R61-9692.0.0	

Looks like all ARM boards affected.

 Issue 737785  has been merged into this issue.

Seen distorted screen/graphics in Blaze just after getting latest M61 build through recovery.
 
M	ChromeOS	Chrome	ARC	Type	Channel
61	9693.0.0	61.0.3142.0	(multiple)	release	dev

Dominik, was this an issue you were looking at (potential side effect of Intel sync patches)?

Is https://chromium-review.googlesource.com/c/551181/ in that image?
If you see "Failing to initialize image" or something similar in the logs that revert will fix it.

Graphics tests on bvt don't test Chrome on purpose. In general graphics was alive, graphics_Sanity ran on kevin and captured fine screenshots
https://pantheon.corp.google.com/storage/browser/chromeos-autotest-results/125603915-chromeos-test/chromeos2-row8-rack9-host11/graphics_Sanity/results/

That means to catch this one would need to capture a screenshot after logging into Chrome (AFAIK no test does it as background images keep changing, but one could cook up something). But that may not be enough. The only way to truly prevent this is to check chameleon tests, which don't run on paladin.

avkodipelli@ per#14 can you please confirm do you see "Failing to initialize image" or something similar in the logs?

Re #15: chameleon tests are e2e, flaky by nature, and have their integration cons. They check mainly external display, and could bring too much noise and maintenance issues.
I think as a first attempt to catch such bugs little earlier, we can review and promote graphics_VTSwitch to bvt-cq. Separate effort should search for better testing solution.

Issue 737945 has been merged into this issue.

The issue is no longer there for Kevin in Chrome OS 9693.1.0, 61.0.3144.0. 

Minnie and nyan blaze are good too in Chrome OS 9693.1.0, 61.0.3144.0. 

Dup? I'm seeing "blank screen" on chell in June 23 builds.
https://bugs.chromium.org/p/chromium/issues/detail?id=738060

ChromeOS stucks at Chrome Logo after cros flash ChromeOS ToT on caroline

/var/ui/ui.LATEST says 
session_manager: error while loading shared libraries: libprotobuf-lite.so.9: cannot open shared object file: No such file or directory

What build was this discovered on? Chrome OS 9693.1.0, 61.0.3144.0. ?

Re#23
It's self build image with ChromeOS ToT checkout source code

Locally marked as 
R61-9697.0.timestamp

 Issue 738140  has been merged into this issue.

I can confirm the issue on Reks, Chell and Minnie.

I'm seeing this on Kevin with 9700.0.0 canary image from today.

seeing this today on big as well. top of tree build.

Per comment #19:

Was somebody expecting this problem to be resolved on Chrome 61.0.3144? I tested this out myself with build 9693.1.0 (Chrome 61.0.3144) and that also seemed to fix the problem for me. But it seems that this was the only build with a recent Chrome. The PFQ has been failing for a while, and we're still stuck at 61.0.3142. (Which would explain people *still* seeing this problem.)

If that's all correct, then it seems like getting the Chrome PFQ in order would be the next step here...

As what Brian described after deploying a new chrome (top of tree) problem is solved.

Ah, wait. 

on nyan-big, top of tree chrome-os + top of tree chrome:

Black screen is no longer the case. But the screen blinks with/without input (mouse, keyboard).

Kristian, maybe you should check with xiaochu@ to see if his nyan device is a repro for your bug :)

This issue seems resolved - graphics_VTSwitch passes since R61-9703.0.0
Not observed anymore on boards with ToT builds.

The following revision refers to this bug:
  https://chromium.googlesource.com/chromiumos/third_party/autotest/+/26a2475c2fa30d931cdce5e1e0a745c15618544b

commit 26a2475c2fa30d931cdce5e1e0a745c15618544b
Author: David Haddock <dhaddock@chromium.org>
Date: Fri Nov 03 06:21:51 2017

[autotest] Test that there is something on screen.

We have had a few builds in the past year that have been passing BVTs
even though there was nothing on the screen. This test aims to find
these builds before the Test team begins their manual testing on them.

8530.86.0 got stuck on the chrome logo at boot.
On samus, daisy and others Intel devies, trying to start chrome failed.
On minnie and arm devices, Chrome started and showed just a black screen (with
cursor).

9693.0.0 showed a black screen (with cursor) on arm devices after Chrome
started. Intel was fine.

Screenshot tests have a problem when the display goes to sleep. We need
to either start Chrome or bring the screen back to life with keyboard
input. I've added the ability to wake the screen to graphic_utils.

BUG=chromium:647436
BUG=chromium:647145
BUG= chromium:737707 
TEST=test_that <ip> graphics_Sanity (screen on and off)
TEST=test_that <ip> graphics_VTSwitch (still works)
CQ-DEPEND=CL:741031
Change-Id: I2c91ed29930a0f5b68d353f6b8c1e4066f1dff53
Reviewed-on: https://chromium-review.googlesource.com/717636
Commit-Ready: David Haddock <dhaddock@chromium.org>
Tested-by: David Haddock <dhaddock@chromium.org>
Reviewed-by: Ilja H. Friedel <ihf@chromium.org>
Reviewed-by: Ruchi Jahagirdar <rjahagir@chromium.org>

[modify] https://crrev.com/26a2475c2fa30d931cdce5e1e0a745c15618544b/client/site_tests/graphics_Sanity/control
[modify] https://crrev.com/26a2475c2fa30d931cdce5e1e0a745c15618544b/client/cros/graphics/graphics_utils.py
[modify] https://crrev.com/26a2475c2fa30d931cdce5e1e0a745c15618544b/client/site_tests/graphics_VTSwitch/graphics_VTSwitch.py
[modify] https://crrev.com/26a2475c2fa30d931cdce5e1e0a745c15618544b/client/site_tests/graphics_Sanity/graphics_Sanity.py