I have 6 devices affected by this issue and we weren't able to recover using version 56 on a USB drive

The customer I was assisting was able to create a Chrome OS recovery image version 54.8743.85.0 and install it on the devices, he was able to recover back the devices functionality, this is a great workaround meanwhile a fixed is found.

After receiving 54.8743.85.0 image from gonzalezjo@google.com, I was able to recover all my chromebooks. Only issue I see in future is will chromebook go from v54 to v58 once 58 is out for Blaze. 

This user lost his V54 recovery image. I didn't tell him to come here and ask for one because I don't know if that's OK. Is there another way to get this image?

Recovery images of HP Chromebook 14 G3 Not work !!

https://productforums.google.com/forum/#!topic/chromebook-central/mjrnIPsWeEU;context-place=forum/chromebook-central

My Device is Chromebook HP 14 G3 Energy Star and this is the Hardware has this issue:
sudo lshw -numeric
alarm                       
    description: ARMv7 Processor rev 3 (v7l)
    product: Google Nyan Blaze Rev 0+
    width: 4294967295 bits
    capabilities: smp
  *-core
       description: Motherboard
       physical id: 0
     *-cpu:0
          description: CPU
          product: cpu
          physical id: 0
          bus info: cpu@0
          size: 2014MHz
          capacity: 2116MHz
          capabilities: swp half thumb fastmult vfp edsp thumbee neon vfpv3 tls vfpv4 idiva idivt vfpd32 lpae cpufreq
     *-cpu:1
          description: CPU
          product: cpu
          physical id: 1
          bus info: cpu@1
          size: 2014MHz
          capacity: 2116MHz
          capabilities: swp half thumb fastmult vfp edsp thumbee neon vfpv3 tls vfpv4 idiva idivt vfpd32 lpae cpufreq
     *-cpu:2
          description: CPU
          product: cpu
          physical id: 2
          bus info: cpu@2
          size: 2014MHz
          capacity: 2116MHz
          capabilities: swp half thumb fastmult vfp edsp thumbee neon vfpv3 tls vfpv4 idiva idivt vfpd32 lpae cpufreq
     *-cpu:3
          description: CPU
          product: cpu
          physical id: 3
          bus info: cpu@3
          size: 2014MHz
          capacity: 2116MHz
          capabilities: swp half thumb fastmult vfp edsp thumbee neon vfpv3 tls vfpv4 idiva idivt vfpd32 lpae cpufreq
     *-memory
          description: System memory
          physical id: 4
          size: 4006MiB
  *-usbhost:0
       product: xHCI Host Controller [1D6B:3]
       vendor: Linux 3.10.18-19-ARCH xhci-hcd [1D6B]
       physical id: 1
       bus info: usb@2
       logical name: usb2
       version: 3.10
       capabilities: usb-3.00
       configuration: driver=hub slots=2 speed=5000Mbit/s
  *-usbhost:1
       product: xHCI Host Controller [1D6B:2]
       vendor: Linux 3.10.18-19-ARCH xhci-hcd [1D6B]
       physical id: 2
       bus info: usb@1
       logical name: usb1
       version: 3.10
       capabilities: usb-2.00
       configuration: driver=hub slots=6 speed=480Mbit/s
     *-usb
          description: USB hub
          product: USB2.0 Hub [5E3:608]
          vendor: Genesys Logic, Inc. [5E3]
          physical id: 2
          bus info: usb@1:2
          version: 85.37
          capabilities: usb-2.00
          configuration: driver=hub maxpower=100mA slots=3 speed=480Mbit/s
        *-usb
             description: Video
             product: HP TrueVision HD Webcam [408:5030]
             vendor: Quanta Computer, Inc. [408]
             physical id: 1
             bus info: usb@1:2.1
             version: 0.03
             capabilities: usb-2.00
             configuration: driver=uvcvideo maxpower=500mA speed=480Mbit/s
  *-network
       description: Ethernet interface
       physical id: 3
       logical name: mlan0
       serial: 28:c2:dd:d5:9f:05
       capabilities: ethernet physical
       configuration: broadcast=yes 

Follow the Information screen OS Damege insert OS USB recovery  with Version > 54 installed in Chromebook HP 14 G3 Energy Start

 active firmaware :Google_nyan_blaze_57.71.63.0
Read only firmware id: Google_nyan_blaze_57.71.63.0
RootKey: c788cee8c79866
gdb.flags 0x 00000000
keyver=0x00020001
TBM fwver = 0x00140002 

 I had trouble ticket opened on store-google about this bug and take i look how the treatment the give for 

Re: Your Google Support Inquiry: Case ID [3-9414000017025]

FROM: store-support@google.com <store-support@google.com>
SEND: quinta-feira, 13 de abril de 2017 17:07
to: Alexandre N. Perfeito
SUBJECT: Re: Your Google Support Inquiry: Case ID [3-9414000017025]
 
Google	 
Hi Alexander,
Thank you for replying! I hope you are doing great!
I appreciate all your time and efforts in troubleshooting the issue. Since the issue persists with another account as well, it confirms it to be an issue with the hardware of the Chromebook. I'd recommend you to contact HP support as they would be the best bet to help us with the warranty and repair options available for your Chromebook.
You may contact HP support at 1-800-474-6836. 
I'm really glad you took the time to get in touch with us, in future we encourage you to do the same, we'll be more than happy to assist you!
It’s been a real pleasure assisting you through the entire conversation. If you have any more questions or concerns related to your device, please let me know by replying to this email. I'm more than happy to help! 
Have a great day!
Thanks!
Henry
The Google Support Team

Has anybody connected this issue, black/blue screen, to the screen itself?  We have been able to fix, at least initially, our G3's with new screens?

Hey I manage to find more update information to you debug the problem take a look: ATTENTION ! the version 55.0 compiled in January 2017 WORKS FINE

I retrieved some stories Me - Google Support "e-mails" before the disaster that we are going through regarding the update of the version of Chrome OS and I was able to take  the information of the last WORK version in which the model Chromebook HP 14 G3 Energy Start was before the Bomb now theses days we going through today update of a look at the email below:

De: Alexandre N. Perfeito 
Enviado: sexta-feira, 14 de abril de 2017 12:57
Para: 
Assunto: Re: Your Google Support Inquiry: Case ID [3-9414000017025]
 
HI Henry,

    Its a not hardware issue because before the update for neweast release  I running with version 55.0 compiled in January 2017 the version more update was running right now 54.0.

    At least provide me the version 55.0 compiled in January 2017 with was before the upgrade problem 

   Wait for your requests

     Happy good fryday. 

Alexandre

In my work environment we have HP Chromebook 14 G3's with 1366x768 resolution screens and with 1920x1080 (FHD) screens. We've only noticed this problem on the FHD Chromebooks.

In fact, installing a non-FHD screen into one of the affected Chromebooks renders it workable again. 

Unfortunately the FHD Chromebooks are not managed, so we cannot hold them to v54 or v55. 

I've also noticed that my personal 13" touchscreen Acer Chromebook (I can't recall the model) started to have artifacting (black lines going through the screen, horizontally) on the Chrome boot logo. This only happened AFTER it updated to v57.

These are all NVidia Tegra chips.

I just got the device last night. Digging in this morning.

Any further updates on this? I have about 30 machines being effected. Any way I can get an older .bin file to at least get mine up and running?

brownca@, file a support ticket and point them at:

https://bugs.chromium.org/p/chromium/issues/detail?id=706535#c64

and Google Support should be able to share the 54 recovery file with you.

Quick update:

Looks like DP fast link training is failing on the panel. When I boot an R55 image, I also see fast link training failure messages. The difference, presumably, is that the fbdev has full link training fallback, which succeeds. It doesn't look like full link training is supported in 3.10 drm sor driver, so I'll try to work around the fast link fail. The panel datasheet might also shed some light on training peculiarities.

i wanted to add that we had a similar issue on this same platform previously. NVIDIA got involved to get a resolution. 

https://code.google.com/p/chrome-os-partner/issues/detail?id=51141

i am waiting on some captures from a customer. 

I haven't been able to tickle fast link training on ToT the right way to get it working.

Skimming the diff between 3.18 and 3.10, it's not trivial to backport full training support (tons of unrelated, but conflicting changes). The drm_dp_helper has also undergone major changes between the 2 versions. IMO the fastest way to do this is to just re-implement the feature in 3.10 using fbdev and 3.18 as a guide.

Update time!

I've uploaded 5 patches which implement proper link training in 3.10, and this resolves the issue on the problem panel. I've also tested on my other blaze (with lower res, single lane panel), and it works there.

Starts with https://chromium-review.googlesource.com/#/c/512782/

The bad news is that I noticed a fair number of what seem like render artifacts while using the devices. I'm syncing up to see if I caught a bad build. @marcheu, is this a known issue on ToT?

Circling back on the render artifacts I mentioned above. I loaded my kernel on an R59 image and verified that everything works fine with the problem panel, display comes up without artifacts.

So something in R60 has regressed rendering on nyan (haven't tried any other boards). At any rate, that seems like it's outside the scope of this bug, I just wanted to doubly confirm that my patches weren't causing the badness.

We experienced this yesterday with two devices - one upgrading from 54 to 58, and one upgrading from 55 to 58. I have restored both back to 54 with the .bin file from Google Support, and disabled auto-updates for our domain, so it is limited to two units, but the issue is still happening with 58.

Thanks for getting the patches together!

Do you think we could get these landed today or tomorrow so we can make the 60 branch (Thursday)?

Stephane is out today so is there an alternate reviewer?

We probably want to get these back into 59 if we can validate them in ToT, we are too late for 58 at this point, but if we can do 59 we can get this into stable by early June.

@bhthompson

I think we can get these landed by the branch point. Perhaps dbehr@ can review in marcheu's absence?

I'm also officially on vacation, but I'll do my best to stay on top of this.

FYI I put one of my affected Chromebooks on the dev channel and it still
had the black screen after it updated

Pri-0 bugs are critical regressions or serious emergencies, and this bug has not been updated in three days. Could you please provide an update, or adjust the priority to a more appropriate level if applicable?

If a fix is in active development, please set the status to Started.

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

marcheu@ is there someone else who can pick this up while seanpaul@ is on vacation?

abodenha@ - the issue is getting the review done, not the actual implementation.   That seems to be in place atm.  dbehr@, could you please review chain at https://chromium-review.googlesource.com/c/512786/ ?  

If possible we should try to get this landed this week and merged back into 60 so we can validate it for 59. 

The changes are pretty big, but appear to be pulling code from other kernels, is landing this week feasible?

Update: While trying to land this, Kristian experienced black screen while booting his blaze. The fixes work on my 2 blazes, so I'll see if he can ship me his for testing. 

I have a unit in failing state if google needs a unit for testing. 

Here's a smaller CL that may fix the nyan without backporting the full link training.

https://chromium-review.googlesource.com/c/530192/

It works (doesn't break) my nyan. We haven't yet tried on one of the affected nyans, but we should know on monday. In the meantime, if anybody with a broken nyan can give this a try, that would be great.

The following revision refers to this bug:
  https://chromium.googlesource.com/chromiumos/third_party/kernel/+/9cd52c8c0412bde64b1c24e115ce767a2d2a505f

commit 9cd52c8c0412bde64b1c24e115ce767a2d2a505f
Author: Thierry Reding <treding@nvidia.com>
Date: Tue Jun 13 03:27:44 2017

CHROMIUM: drm/dp: Add link training helper

Add a helper that will perform link training as described in the
DisplayPort specification.

Signed-off-by: Thierry Reding <treding@nvidia.com>
[from https://github.com/thierryreding/linux.git]
(cherry picked from commit 352ae461f2aa8527805616feea3715c4c1dda75d)
Signed-off-by: Sean Paul <seanpaul@chromium.org>

BUG= chromium:706535 
TEST=Tested on 2 blaze devices (lowres and fhd) including flakey panel, display
	comes up on both

Reviewed-on: https://chromium-review.googlesource.com/308867
Commit-Ready: Sean Paul <seanpaul@chromium.org>
Tested-by: Sean Paul <seanpaul@chromium.org>
[seanpaul squashed post cursor functions in]
Reviewed-by: Sean Paul <seanpaul@chromium.org>
Change-Id: I7b5ecdf7a1bb3dd30b09e2c73408473693b03bfd
Reviewed-on: https://chromium-review.googlesource.com/512782
Commit-Ready: Kristian H. Kristensen <hoegsberg@chromium.org>
Tested-by: Kristian H. Kristensen <hoegsberg@chromium.org>
Reviewed-by: Kristian H. Kristensen <hoegsberg@chromium.org>

[modify] https://crrev.com/9cd52c8c0412bde64b1c24e115ce767a2d2a505f/drivers/gpu/drm/drm_dp_helper.c
[modify] https://crrev.com/9cd52c8c0412bde64b1c24e115ce767a2d2a505f/include/drm/drm_dp_helper.h

The following revision refers to this bug:
  https://chromium.googlesource.com/chromiumos/third_party/kernel/+/048d9631514c1f83a0c5d86bff869e35cab6dd6f

commit 048d9631514c1f83a0c5d86bff869e35cab6dd6f
Author: Thierry Reding <treding@nvidia.com>
Date: Tue Jun 13 03:27:46 2017

CHROMIUM: drm/dp: Add drm_dp_link_choose() helper

This helper chooses an appropriate configuration, according to the
bitrate requirements of the video mode and the capabilities of the
DisplayPort sink.

Signed-off-by: Thierry Reding <treding@nvidia.com>
[from https://github.com/thierryreding/linux.git]
(cherry picked from commit 2a30ac9d9bcfac2bd076961dc0aefb09d5bcadca)
Signed-off-by: Sean Paul <seanpaul@chromium.org>

BUG= chromium:706535 
TEST=Tested on 2 blaze devices (lowres and fhd) including flakey panel, display
	comes up on both

Reviewed-on: https://chromium-review.googlesource.com/308870
Commit-Ready: Sean Paul <seanpaul@chromium.org>
Tested-by: Sean Paul <seanpaul@chromium.org>
Reviewed-by: Sean Paul <seanpaul@chromium.org>
Change-Id: I4cc973f99847c990fd74972b88f35f37739269a6
Reviewed-on: https://chromium-review.googlesource.com/512783
Commit-Ready: Kristian H. Kristensen <hoegsberg@chromium.org>
Tested-by: Kristian H. Kristensen <hoegsberg@chromium.org>
Reviewed-by: Kristian H. Kristensen <hoegsberg@chromium.org>

[modify] https://crrev.com/048d9631514c1f83a0c5d86bff869e35cab6dd6f/drivers/gpu/drm/drm_dp_helper.c
[modify] https://crrev.com/048d9631514c1f83a0c5d86bff869e35cab6dd6f/include/drm/drm_dp_helper.h

The following revision refers to this bug:
  https://chromium.googlesource.com/chromiumos/third_party/kernel/+/7278e48b04be417f0e3ce16b86d8fe6e8af5f8c9

commit 7278e48b04be417f0e3ce16b86d8fe6e8af5f8c9
Author: Thierry Reding <treding@nvidia.com>
Date: Tue Jun 13 03:27:45 2017

CHROMIUM: drm/dp: link: Track capabilities alongside settings

Store capabilities in max_* fields and add separate fields for the
currently selected settings.

Signed-off-by: Thierry Reding <treding@nvidia.com>
[from https://github.com/thierryreding/linux.git]
(cherry picked from commit 282aadf46077d8801b3b0219b812a9c6d9fbda20)
Signed-off-by: Sean Paul <seanpaul@chromium.org>

BUG= chromium:706535 
TEST=Tested on 2 blaze devices (lowres and fhd) including flakey panel, display
	comes up on both

Reviewed-on: https://chromium-review.googlesource.com/308869
Commit-Ready: Sean Paul <seanpaul@chromium.org>
Tested-by: Sean Paul <seanpaul@chromium.org>
Reviewed-by: Sean Paul <seanpaul@chromium.org>
Change-Id: I279860c3439cbd5b232a2714021dcddf25e0d4f4
Reviewed-on: https://chromium-review.googlesource.com/512784
Commit-Ready: Kristian H. Kristensen <hoegsberg@chromium.org>
Tested-by: Kristian H. Kristensen <hoegsberg@chromium.org>
Reviewed-by: Kristian H. Kristensen <hoegsberg@chromium.org>

[modify] https://crrev.com/7278e48b04be417f0e3ce16b86d8fe6e8af5f8c9/drivers/gpu/drm/drm_dp_helper.c
[modify] https://crrev.com/7278e48b04be417f0e3ce16b86d8fe6e8af5f8c9/include/drm/drm_dp_helper.h
[modify] https://crrev.com/7278e48b04be417f0e3ce16b86d8fe6e8af5f8c9/drivers/gpu/drm/tegra/sor.c

The following revision refers to this bug:
  https://chromium.googlesource.com/chromiumos/third_party/kernel/+/2e46d0bc24c379be0580f30adb208d504769ab26

commit 2e46d0bc24c379be0580f30adb208d504769ab26
Author: Sean Paul <seanpaul@chromium.org>
Date: Tue Jun 13 03:27:47 2017

CHROMIUM: drm/tegra: Shuffle lane power up code

Pull out the lane power up code from sor_enable, and introduce a
configure function for dp. This code is adapted from 3.18.

BUG= chromium:706535 
TEST=Tested on 2 blaze devices (lowres and fhd) including flakey panel, display
	comes up on both

Signed-off-by: Sean Paul <seanpaul@chromium.org>
Change-Id: I0e5d2000a9bfdbb835b1a6d49cc6c9cc069a13ab
Reviewed-on: https://chromium-review.googlesource.com/512785
Commit-Ready: Kristian H. Kristensen <hoegsberg@chromium.org>
Tested-by: Kristian H. Kristensen <hoegsberg@chromium.org>
Reviewed-by: Kristian H. Kristensen <hoegsberg@chromium.org>

[modify] https://crrev.com/2e46d0bc24c379be0580f30adb208d504769ab26/drivers/gpu/drm/tegra/sor.c

The following revision refers to this bug:
  https://chromium.googlesource.com/chromiumos/third_party/kernel/+/17b67fc18ff443e8af099506b0a3db4b859b0fd7

commit 17b67fc18ff443e8af099506b0a3db4b859b0fd7
Author: Sean Paul <seanpaul@chromium.org>
Date: Tue Jun 13 03:27:48 2017

CHROMIUM: drm/tegra: Implement full link training

Pull back and wrangle enough of the training/configuration code
from 3.18 to implement full link training.

BUG= chromium:706535 
TEST=Tested on 2 blaze devices (lowres and fhd) including flakey panel, display
	comes up on both

Signed-off-by: Sean Paul <seanpaul@chromium.org>
Change-Id: I80c019928cd8120e8dbb2564ea1d20a3aaf5a324
Reviewed-on: https://chromium-review.googlesource.com/512786
Commit-Ready: Kristian H. Kristensen <hoegsberg@chromium.org>
Tested-by: Kristian H. Kristensen <hoegsberg@chromium.org>
Reviewed-by: Kristian H. Kristensen <hoegsberg@chromium.org>

[modify] https://crrev.com/17b67fc18ff443e8af099506b0a3db4b859b0fd7/drivers/gpu/drm/tegra/dpaux.c
[modify] https://crrev.com/17b67fc18ff443e8af099506b0a3db4b859b0fd7/include/drm/drm_dp_helper.h
[modify] https://crrev.com/17b67fc18ff443e8af099506b0a3db4b859b0fd7/drivers/gpu/drm/tegra/drm.h
[modify] https://crrev.com/17b67fc18ff443e8af099506b0a3db4b859b0fd7/drivers/gpu/drm/tegra/sor.h
[modify] https://crrev.com/17b67fc18ff443e8af099506b0a3db4b859b0fd7/drivers/gpu/drm/tegra/sor.c

@bhthompson: Please advise on which releases the fix should be merged to

If the patches are ok on ToT lets start with a merge request for 60. 

If they work ok on the next 60 release we can try to get them into a later 59 given the user impact here.

This bug requires manual review: M60 has already been promoted to the beta branch, so this requires manual review
Please contact the milestone owner if you have questions.
Owners: amineer@(Android), cmasso@(iOS), josafat@(ChromeOS), bustamante@(Desktop)

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

The following revision refers to this bug:
  https://chromium.googlesource.com/chromiumos/third_party/kernel/+/b1f7f320804270c00f1b4d353f69a2b546b59cc1

commit b1f7f320804270c00f1b4d353f69a2b546b59cc1
Author: Thierry Reding <treding@nvidia.com>
Date: Thu Jun 15 17:27:08 2017

CHROMIUM: drm/dp: Add link training helper

Add a helper that will perform link training as described in the
DisplayPort specification.

Signed-off-by: Thierry Reding <treding@nvidia.com>
[from https://github.com/thierryreding/linux.git]
(cherry picked from commit 352ae461f2aa8527805616feea3715c4c1dda75d)
Signed-off-by: Sean Paul <seanpaul@chromium.org>

BUG= chromium:706535 
TEST=Tested on 2 blaze devices (lowres and fhd) including flakey panel, display
	comes up on both

Reviewed-on: https://chromium-review.googlesource.com/308867
Commit-Ready: Sean Paul <seanpaul@chromium.org>
Tested-by: Sean Paul <seanpaul@chromium.org>
[seanpaul squashed post cursor functions in]
Reviewed-by: Sean Paul <seanpaul@chromium.org>
Change-Id: I7b5ecdf7a1bb3dd30b09e2c73408473693b03bfd
Reviewed-on: https://chromium-review.googlesource.com/512782
Commit-Ready: Kristian H. Kristensen <hoegsberg@chromium.org>
Tested-by: Kristian H. Kristensen <hoegsberg@chromium.org>
Reviewed-by: Kristian H. Kristensen <hoegsberg@chromium.org>
(cherry picked from commit 9cd52c8c0412bde64b1c24e115ce767a2d2a505f)
Reviewed-on: https://chromium-review.googlesource.com/537194
Commit-Queue: Kristian H. Kristensen <hoegsberg@chromium.org>
Trybot-Ready: Kristian H. Kristensen <hoegsberg@chromium.org>

[modify] https://crrev.com/b1f7f320804270c00f1b4d353f69a2b546b59cc1/drivers/gpu/drm/drm_dp_helper.c
[modify] https://crrev.com/b1f7f320804270c00f1b4d353f69a2b546b59cc1/include/drm/drm_dp_helper.h

The following revision refers to this bug:
  https://chromium.googlesource.com/chromiumos/third_party/kernel/+/84a083cfdc07caefe97597c6445b9be0c006e948

commit 84a083cfdc07caefe97597c6445b9be0c006e948
Author: Thierry Reding <treding@nvidia.com>
Date: Thu Jun 15 17:57:43 2017

CHROMIUM: drm/dp: link: Track capabilities alongside settings

Store capabilities in max_* fields and add separate fields for the
currently selected settings.

Signed-off-by: Thierry Reding <treding@nvidia.com>
[from https://github.com/thierryreding/linux.git]
(cherry picked from commit 282aadf46077d8801b3b0219b812a9c6d9fbda20)
Signed-off-by: Sean Paul <seanpaul@chromium.org>

BUG= chromium:706535 
TEST=Tested on 2 blaze devices (lowres and fhd) including flakey panel, display
	comes up on both

Reviewed-on: https://chromium-review.googlesource.com/308869
Commit-Ready: Sean Paul <seanpaul@chromium.org>
Tested-by: Sean Paul <seanpaul@chromium.org>
Reviewed-by: Sean Paul <seanpaul@chromium.org>
Change-Id: I279860c3439cbd5b232a2714021dcddf25e0d4f4
Reviewed-on: https://chromium-review.googlesource.com/512784
Commit-Ready: Kristian H. Kristensen <hoegsberg@chromium.org>
Tested-by: Kristian H. Kristensen <hoegsberg@chromium.org>
Reviewed-by: Kristian H. Kristensen <hoegsberg@chromium.org>
(cherry picked from commit 7278e48b04be417f0e3ce16b86d8fe6e8af5f8c9)
Reviewed-on: https://chromium-review.googlesource.com/537195
Commit-Queue: Kristian H. Kristensen <hoegsberg@chromium.org>

[modify] https://crrev.com/84a083cfdc07caefe97597c6445b9be0c006e948/drivers/gpu/drm/drm_dp_helper.c
[modify] https://crrev.com/84a083cfdc07caefe97597c6445b9be0c006e948/include/drm/drm_dp_helper.h
[modify] https://crrev.com/84a083cfdc07caefe97597c6445b9be0c006e948/drivers/gpu/drm/tegra/sor.c

The following revision refers to this bug:
  https://chromium.googlesource.com/chromiumos/third_party/kernel/+/4cc948bdf8d7f3f96a9857ab804817cbf42842fa

commit 4cc948bdf8d7f3f96a9857ab804817cbf42842fa
Author: Sean Paul <seanpaul@chromium.org>
Date: Thu Jun 15 17:57:48 2017

CHROMIUM: drm/tegra: Shuffle lane power up code

Pull out the lane power up code from sor_enable, and introduce a
configure function for dp. This code is adapted from 3.18.

BUG= chromium:706535 
TEST=Tested on 2 blaze devices (lowres and fhd) including flakey panel, display
	comes up on both

Signed-off-by: Sean Paul <seanpaul@chromium.org>
Change-Id: I0e5d2000a9bfdbb835b1a6d49cc6c9cc069a13ab
Reviewed-on: https://chromium-review.googlesource.com/512785
Commit-Ready: Kristian H. Kristensen <hoegsberg@chromium.org>
Tested-by: Kristian H. Kristensen <hoegsberg@chromium.org>
Reviewed-by: Kristian H. Kristensen <hoegsberg@chromium.org>
(cherry picked from commit 2e46d0bc24c379be0580f30adb208d504769ab26)
Reviewed-on: https://chromium-review.googlesource.com/537197
Commit-Queue: Kristian H. Kristensen <hoegsberg@chromium.org>

[modify] https://crrev.com/4cc948bdf8d7f3f96a9857ab804817cbf42842fa/drivers/gpu/drm/tegra/sor.c

The following revision refers to this bug:
  https://chromium.googlesource.com/chromiumos/third_party/kernel/+/ab5ed9a3f33448891dc54d6ac4dd6714e5bfbf57

commit ab5ed9a3f33448891dc54d6ac4dd6714e5bfbf57
Author: Thierry Reding <treding@nvidia.com>
Date: Thu Jun 15 17:57:53 2017

CHROMIUM: drm/dp: Add drm_dp_link_choose() helper

This helper chooses an appropriate configuration, according to the
bitrate requirements of the video mode and the capabilities of the
DisplayPort sink.

Signed-off-by: Thierry Reding <treding@nvidia.com>
[from https://github.com/thierryreding/linux.git]
(cherry picked from commit 2a30ac9d9bcfac2bd076961dc0aefb09d5bcadca)
Signed-off-by: Sean Paul <seanpaul@chromium.org>

BUG= chromium:706535 
TEST=Tested on 2 blaze devices (lowres and fhd) including flakey panel, display
	comes up on both

Reviewed-on: https://chromium-review.googlesource.com/308870
Commit-Ready: Sean Paul <seanpaul@chromium.org>
Tested-by: Sean Paul <seanpaul@chromium.org>
Reviewed-by: Sean Paul <seanpaul@chromium.org>
Change-Id: I4cc973f99847c990fd74972b88f35f37739269a6
Reviewed-on: https://chromium-review.googlesource.com/512783
Commit-Ready: Kristian H. Kristensen <hoegsberg@chromium.org>
Tested-by: Kristian H. Kristensen <hoegsberg@chromium.org>
Reviewed-by: Kristian H. Kristensen <hoegsberg@chromium.org>
(cherry picked from commit 048d9631514c1f83a0c5d86bff869e35cab6dd6f)
Reviewed-on: https://chromium-review.googlesource.com/537196
Commit-Queue: Kristian H. Kristensen <hoegsberg@chromium.org>

[modify] https://crrev.com/ab5ed9a3f33448891dc54d6ac4dd6714e5bfbf57/drivers/gpu/drm/drm_dp_helper.c
[modify] https://crrev.com/ab5ed9a3f33448891dc54d6ac4dd6714e5bfbf57/include/drm/drm_dp_helper.h

The following revision refers to this bug:
  https://chromium.googlesource.com/chromiumos/third_party/kernel/+/a9be1466df851fe1749a450504adcb879317ec4f

commit a9be1466df851fe1749a450504adcb879317ec4f
Author: Sean Paul <seanpaul@chromium.org>
Date: Thu Jun 15 18:04:56 2017

CHROMIUM: drm/tegra: Implement full link training

Pull back and wrangle enough of the training/configuration code
from 3.18 to implement full link training.

BUG= chromium:706535 
TEST=Tested on 2 blaze devices (lowres and fhd) including flakey panel, display
	comes up on both

Signed-off-by: Sean Paul <seanpaul@chromium.org>
Change-Id: I80c019928cd8120e8dbb2564ea1d20a3aaf5a324
Reviewed-on: https://chromium-review.googlesource.com/512786
Commit-Ready: Kristian H. Kristensen <hoegsberg@chromium.org>
Tested-by: Kristian H. Kristensen <hoegsberg@chromium.org>
Reviewed-by: Kristian H. Kristensen <hoegsberg@chromium.org>
(cherry picked from commit 17b67fc18ff443e8af099506b0a3db4b859b0fd7)
Reviewed-on: https://chromium-review.googlesource.com/537200
Commit-Queue: Kristian H. Kristensen <hoegsberg@chromium.org>

[modify] https://crrev.com/a9be1466df851fe1749a450504adcb879317ec4f/drivers/gpu/drm/tegra/dpaux.c
[modify] https://crrev.com/a9be1466df851fe1749a450504adcb879317ec4f/include/drm/drm_dp_helper.h
[modify] https://crrev.com/a9be1466df851fe1749a450504adcb879317ec4f/drivers/gpu/drm/tegra/drm.h
[modify] https://crrev.com/a9be1466df851fe1749a450504adcb879317ec4f/drivers/gpu/drm/tegra/sor.h
[modify] https://crrev.com/a9be1466df851fe1749a450504adcb879317ec4f/drivers/gpu/drm/tegra/sor.c

Backported and tested on R60.

Please add merge-request for 59 is needed.

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

We appreciate all the hard work put into this issue. In layman's terms, once the merge for version 59(Currently the Stable version) is complete, will the Chrome Recovery Utility download 59 to a USB so we can apply this fix to our affected Chromebooks?

Requesting merge to 59. While this is a late change for 59, it resolves a critical issue and should be considered.

Impact of the change is limited, as the only systems using 3.10 kernel today are Nyan, we should check a big and a kitty with the 59 build including this to be sure. 

Re comment #115, in short yes, however it will take some time as we need to qualify the new OS build and such before we can safely update the recovery image. 

If you have an account manager at Google you work with they may be able to help get a recovery image earlier.

I tried today in 10:58 PST upgrade my ChromeBook version 54 using setting options about chrome OS hyperlink and button check updates using stable channel  and after download and press button restart my chromebook stil the same above about this bug. Fails hard again  in the black screen.
In screen developed mode I tap the key TAB and follow the information above 

Google_Nyan_Blaze 5771.63.0  --> BOMB  

Back to version 54 again !!!

Do we have the specific 59 version the solution will be implemented inside?

I now have display problems on the HP Chromebook 14 G3, could it be related to
this fix?  I did not see the blank screen problem, but the display is blinking
and displaying random garbage, especially at edges of drawed areas.  I was
running the development version when this occurred on the update to 60.x then
I set my chromebook to beta and now that is update to 60.x and the issue occurs
again!  Both times I reported the problem.

Please get this merged into M59 in the next hour

The following revision refers to this bug:
  https://chromium.googlesource.com/chromiumos/third_party/kernel/+/2a3e5cf234dedd6c1faee4fcae13b74a768e68b2

commit 2a3e5cf234dedd6c1faee4fcae13b74a768e68b2
Author: Thierry Reding <treding@nvidia.com>
Date: Wed Jun 21 21:08:42 2017

CHROMIUM: drm/dp: Add link training helper

Add a helper that will perform link training as described in the
DisplayPort specification.

Signed-off-by: Thierry Reding <treding@nvidia.com>
[from https://github.com/thierryreding/linux.git]
(cherry picked from commit 352ae461f2aa8527805616feea3715c4c1dda75d)
Signed-off-by: Sean Paul <seanpaul@chromium.org>

BUG= chromium:706535 
TEST=Tested on 2 blaze devices (lowres and fhd) including flakey panel, display
	comes up on both

Reviewed-on: https://chromium-review.googlesource.com/308867
Commit-Ready: Sean Paul <seanpaul@chromium.org>
Tested-by: Sean Paul <seanpaul@chromium.org>
[seanpaul squashed post cursor functions in]
Reviewed-by: Sean Paul <seanpaul@chromium.org>
Change-Id: I7b5ecdf7a1bb3dd30b09e2c73408473693b03bfd
Reviewed-on: https://chromium-review.googlesource.com/512782
Commit-Ready: Kristian H. Kristensen <hoegsberg@chromium.org>
Tested-by: Kristian H. Kristensen <hoegsberg@chromium.org>
Reviewed-by: Kristian H. Kristensen <hoegsberg@chromium.org>
Reviewed-on: https://chromium-review.googlesource.com/544055
Commit-Queue: Sean Paul <seanpaul@chromium.org>

[modify] https://crrev.com/2a3e5cf234dedd6c1faee4fcae13b74a768e68b2/drivers/gpu/drm/drm_dp_helper.c
[modify] https://crrev.com/2a3e5cf234dedd6c1faee4fcae13b74a768e68b2/include/drm/drm_dp_helper.h

The following revision refers to this bug:
  https://chromium.googlesource.com/chromiumos/third_party/kernel/+/2a3e5cf234dedd6c1faee4fcae13b74a768e68b2

commit 2a3e5cf234dedd6c1faee4fcae13b74a768e68b2
Author: Thierry Reding <treding@nvidia.com>
Date: Wed Jun 21 21:08:42 2017

CHROMIUM: drm/dp: Add link training helper

Add a helper that will perform link training as described in the
DisplayPort specification.

Signed-off-by: Thierry Reding <treding@nvidia.com>
[from https://github.com/thierryreding/linux.git]
(cherry picked from commit 352ae461f2aa8527805616feea3715c4c1dda75d)
Signed-off-by: Sean Paul <seanpaul@chromium.org>

BUG= chromium:706535 
TEST=Tested on 2 blaze devices (lowres and fhd) including flakey panel, display
	comes up on both

Reviewed-on: https://chromium-review.googlesource.com/308867
Commit-Ready: Sean Paul <seanpaul@chromium.org>
Tested-by: Sean Paul <seanpaul@chromium.org>
[seanpaul squashed post cursor functions in]
Reviewed-by: Sean Paul <seanpaul@chromium.org>
Change-Id: I7b5ecdf7a1bb3dd30b09e2c73408473693b03bfd
Reviewed-on: https://chromium-review.googlesource.com/512782
Commit-Ready: Kristian H. Kristensen <hoegsberg@chromium.org>
Tested-by: Kristian H. Kristensen <hoegsberg@chromium.org>
Reviewed-by: Kristian H. Kristensen <hoegsberg@chromium.org>
Reviewed-on: https://chromium-review.googlesource.com/544055
Commit-Queue: Sean Paul <seanpaul@chromium.org>

[modify] https://crrev.com/2a3e5cf234dedd6c1faee4fcae13b74a768e68b2/drivers/gpu/drm/drm_dp_helper.c
[modify] https://crrev.com/2a3e5cf234dedd6c1faee4fcae13b74a768e68b2/include/drm/drm_dp_helper.h

The following revision refers to this bug:
  https://chromium.googlesource.com/chromiumos/third_party/kernel/+/aea786788347584e716ff5008a258c3fee8072a5

commit aea786788347584e716ff5008a258c3fee8072a5
Author: Thierry Reding <treding@nvidia.com>
Date: Wed Jun 21 21:08:54 2017

CHROMIUM: drm/dp: link: Track capabilities alongside settings

Store capabilities in max_* fields and add separate fields for the
currently selected settings.

Signed-off-by: Thierry Reding <treding@nvidia.com>
[from https://github.com/thierryreding/linux.git]
(cherry picked from commit 282aadf46077d8801b3b0219b812a9c6d9fbda20)
Signed-off-by: Sean Paul <seanpaul@chromium.org>

BUG= chromium:706535 
TEST=Tested on 2 blaze devices (lowres and fhd) including flakey panel, display
	comes up on both

Reviewed-on: https://chromium-review.googlesource.com/308869
Commit-Ready: Sean Paul <seanpaul@chromium.org>
Tested-by: Sean Paul <seanpaul@chromium.org>
Reviewed-by: Sean Paul <seanpaul@chromium.org>
Change-Id: I279860c3439cbd5b232a2714021dcddf25e0d4f4
Reviewed-on: https://chromium-review.googlesource.com/512784
Commit-Ready: Kristian H. Kristensen <hoegsberg@chromium.org>
Tested-by: Kristian H. Kristensen <hoegsberg@chromium.org>
Reviewed-by: Kristian H. Kristensen <hoegsberg@chromium.org>
Reviewed-on: https://chromium-review.googlesource.com/544037
Commit-Queue: Sean Paul <seanpaul@chromium.org>

[modify] https://crrev.com/aea786788347584e716ff5008a258c3fee8072a5/drivers/gpu/drm/drm_dp_helper.c
[modify] https://crrev.com/aea786788347584e716ff5008a258c3fee8072a5/include/drm/drm_dp_helper.h
[modify] https://crrev.com/aea786788347584e716ff5008a258c3fee8072a5/drivers/gpu/drm/tegra/sor.c

The following revision refers to this bug:
  https://chromium.googlesource.com/chromiumos/third_party/kernel/+/bb27529fe62b6a40a97b25edc72826709180caf5

commit bb27529fe62b6a40a97b25edc72826709180caf5
Author: Thierry Reding <treding@nvidia.com>
Date: Wed Jun 21 21:09:05 2017

CHROMIUM: drm/dp: Add drm_dp_link_choose() helper

This helper chooses an appropriate configuration, according to the
bitrate requirements of the video mode and the capabilities of the
DisplayPort sink.

Signed-off-by: Thierry Reding <treding@nvidia.com>
[from https://github.com/thierryreding/linux.git]
(cherry picked from commit 2a30ac9d9bcfac2bd076961dc0aefb09d5bcadca)
Signed-off-by: Sean Paul <seanpaul@chromium.org>

BUG= chromium:706535 
TEST=Tested on 2 blaze devices (lowres and fhd) including flakey panel, display
	comes up on both

Reviewed-on: https://chromium-review.googlesource.com/308870
Commit-Ready: Sean Paul <seanpaul@chromium.org>
Tested-by: Sean Paul <seanpaul@chromium.org>
Reviewed-by: Sean Paul <seanpaul@chromium.org>
Change-Id: I4cc973f99847c990fd74972b88f35f37739269a6
Reviewed-on: https://chromium-review.googlesource.com/512783
Commit-Ready: Kristian H. Kristensen <hoegsberg@chromium.org>
Tested-by: Kristian H. Kristensen <hoegsberg@chromium.org>
Reviewed-by: Kristian H. Kristensen <hoegsberg@chromium.org>
Reviewed-on: https://chromium-review.googlesource.com/544038
Commit-Queue: Sean Paul <seanpaul@chromium.org>

[modify] https://crrev.com/bb27529fe62b6a40a97b25edc72826709180caf5/drivers/gpu/drm/drm_dp_helper.c
[modify] https://crrev.com/bb27529fe62b6a40a97b25edc72826709180caf5/include/drm/drm_dp_helper.h

The following revision refers to this bug:
  https://chromium.googlesource.com/chromiumos/third_party/kernel/+/2645ac801b0a7c3e5acc113aa68d95c941d8b81f

commit 2645ac801b0a7c3e5acc113aa68d95c941d8b81f
Author: Sean Paul <seanpaul@chromium.org>
Date: Wed Jun 21 21:09:16 2017

CHROMIUM: drm/tegra: Shuffle lane power up code

Pull out the lane power up code from sor_enable, and introduce a
configure function for dp. This code is adapted from 3.18.

BUG= chromium:706535 
TEST=Tested on 2 blaze devices (lowres and fhd) including flakey panel, display
	comes up on both

Signed-off-by: Sean Paul <seanpaul@chromium.org>
Change-Id: I0e5d2000a9bfdbb835b1a6d49cc6c9cc069a13ab
Reviewed-on: https://chromium-review.googlesource.com/512785
Commit-Ready: Kristian H. Kristensen <hoegsberg@chromium.org>
Tested-by: Kristian H. Kristensen <hoegsberg@chromium.org>
Reviewed-by: Kristian H. Kristensen <hoegsberg@chromium.org>
Reviewed-on: https://chromium-review.googlesource.com/544039

[modify] https://crrev.com/2645ac801b0a7c3e5acc113aa68d95c941d8b81f/drivers/gpu/drm/tegra/sor.c

The following revision refers to this bug:
  https://chromium.googlesource.com/chromiumos/third_party/kernel/+/ae9f659bff391b1650808f194e48f9eac8b0da02

commit ae9f659bff391b1650808f194e48f9eac8b0da02
Author: Sean Paul <seanpaul@chromium.org>
Date: Wed Jun 21 21:09:27 2017

CHROMIUM: drm/tegra: Implement full link training

Pull back and wrangle enough of the training/configuration code
from 3.18 to implement full link training.

BUG= chromium:706535 
TEST=Tested on 2 blaze devices (lowres and fhd) including flakey panel, display
	comes up on both

Signed-off-by: Sean Paul <seanpaul@chromium.org>
Change-Id: I80c019928cd8120e8dbb2564ea1d20a3aaf5a324
Reviewed-on: https://chromium-review.googlesource.com/512786
Commit-Ready: Kristian H. Kristensen <hoegsberg@chromium.org>
Tested-by: Kristian H. Kristensen <hoegsberg@chromium.org>
Reviewed-by: Kristian H. Kristensen <hoegsberg@chromium.org>
Reviewed-on: https://chromium-review.googlesource.com/544040

[modify] https://crrev.com/ae9f659bff391b1650808f194e48f9eac8b0da02/drivers/gpu/drm/tegra/dpaux.c
[modify] https://crrev.com/ae9f659bff391b1650808f194e48f9eac8b0da02/include/drm/drm_dp_helper.h
[modify] https://crrev.com/ae9f659bff391b1650808f194e48f9eac8b0da02/drivers/gpu/drm/tegra/drm.h
[modify] https://crrev.com/ae9f659bff391b1650808f194e48f9eac8b0da02/drivers/gpu/drm/tegra/sor.h
[modify] https://crrev.com/ae9f659bff391b1650808f194e48f9eac8b0da02/drivers/gpu/drm/tegra/sor.c

I tried today in 11:16 PST upgrade my ChromeBook version 54 using theses steps 
 setting --> options ---> about chrome OS hyperlink -->  press button check updates using stable channel. 
 After download and press button restart my chromebook Restart NORMALY !!!! New Logo (ChromeOS) Refreshed Menu Bars: Check it out below : uname -a

Linux localhost 3.10.18 #1 SMP Thu Jun 22 14:28:07 PDT 2017 armv7l ARMv7 Processor rev 3 (v7l) NVIDIA Tegra SoC (Flattened Device Tree) GNU/Linux

My Device is HP Chromebook G3 --> Tegra TK1 

Now running Version 59.0.3071.113 (Official Build) (32-bit) 

Now its OK .. Thanks ... 

@perfeitoan
Thanks for the instruction. It seems the HP G3 Chromebooks will update to Chrome OS 59.0.3071.113 if the OU in the admin console is set to unrestricted on the stable channel, but CURRENTLY if the version is set to 59* on the stable channel, it will update to a non-working version. A version earlier than 59.0.3071.113.

Thank you Guys for the fix and can confirm working. Could you apply 59.0.3071.113 to be pointed to when 59* is selected in the admin console or is that something that happens automatically after some time has passed? 

@ddero...@gmail.com.

This update was made by Graphics Interface as I said before. I not tried in via console admin update. I tested like a normal update tool give by ChormeOS and.
Version 54 --> 59 OK ! directly no bumbs.
stable channel selected

note that this issue is causing black screen on certain panels for Big see  issue 737618  for more details 

@perfeitoan@hotmail.com and @ddero...@gmail.com I changed our admin console setting to allow our devices of this model to update to version 59 and updated my HP Chromebook 14 G3, last night. It updated to Version 59.0.3071.91 and booted to a black screen.

If you pin the version from the Admin Console to 59.* it will update to 59.0.3071.91 and the fix has been released to 59.0.3071.113, I recommend to allow the auto updates and set version restriction to none, then return back the device to version 54 via USB and then allow the update and the device should get the version 59.0.3071.113, after that you will be able to set the restriction policy to version 59.* if you wish, this last action won't reverse the current version on the device because only way to downgrade is via USB recovery image.

I hope this helps.

Thank you @gonzalezjo@google.com, - I completed the USB restore to version 54, set the admin console to have no version restriction (Comment 163), and successfully upgraded to Version 59.0.3071.113 (Official Build) (32-bit) with no black screen issues.

@nathan.papendorf@westby-norse.org

You're welcome, remember you can now set the restriction back to 59.* with no problem. 

I am very happy to hear this is now resolved. We have over 1000 chromebooks for my school that had this problem. 

My wife's chromebook screen stayed black after the 54 update.  How do I get her's updated to 59 and fixed with no screen functionality?

When should this be the image in the Recovery Utility? We have several labs that need to be re-imaged with a valid working image this summer. This would help Comment #140 as well.

Comment #140, 141 - Open a case with Google Support. They will link you the download for version 54. From there, you'll be able to update to 59.0.3071.113.

@bhthompson, can we get Blaze recovery image updated to 9460.67.0 publicly so that users who are in a bad state can self-service recover without needing to request the 54 recovery image?

Passing to Grace for 59 recovery push.

I have google how to open a case with google support and cannot find any information except post in a forum because google support monitors.  Also, I tried to download a new update and install and now it simply says I have no operating system, but nothing happens with the jump drive and the newest update.  Help!

@jwall@centralcss.org

Please check the solution to in comment 136
https://bugs.chromium.org/p/chromium/issues/detail?id=706535#c136

If you need Chrome OS 54 to get back to a functional operating system and follow the instruction in comment 136 please open a case with Google Cloud Support and Chrome team will be able to provide that to you.
https://support.google.com/a/answer/1047213

I've a customer with a different HWID with this issue:
customer HW:
HP Chromebook 14 x000-x999 / HP Chromebook 14 G3
BLAZE F2A-F3F-A3A-I32
the customer tried:
upgrade to 59.0.3071.113 (black screen)
connecting an external monitor (black screen)
enabling developer mode with SSH server (connection refused)
switch to console (ctrl alt right arrow) (black screen)
upgrade to beta  
Google Chrome Version 60.0.3112.50 Platform Version 9592.37.0 (Official Build) beta-channel nyan_blaze
(black screen)
upgrade to developer (black screen)

I am asking how to open a case with google Cloud Support.  Have googled the answer and can find nothing about how to do it.

For your personal device, you can use the contact area here:
https://support.google.com/chromebook/#topic=3399709&contact=1

Thanks for the link. I have been in a chat with support for ten minutes.  The best they can do so far is refer me to the recovery tool which does not work.  I linked them to this thread and asked specifically for a 54 install.  He is speaking to a specialist, but I have my doubts that I will get any resolution.

jkwal...@gmail.com/jw...@centralcss.org - if you're still having issues, email me at papennat@gmail.com and I can share you the steps they gave me.

Hello team, we have a case that affected 11 devices, 4 of them are still affected after attempting to update from 54 to version 59 and beta, the devices are the same model. we tried the recovery tool, but at this moment it is providing the build 9460.60.0.

Do we have an ETA of 9460.67.0 to the recovery tool? 

The 9460.67 nyan_blaze recovery images are making their way onto the recovery tool, they should be live by later this afternoon.

It appears to be there now.

If there are further issues here please let us know, otherwise I am closing this bug out.

We have a case where 4 devices are affected, even if they try the recovery with the new version, the issue persists and the only work around is version 54, if he updates or implements the recovery tool he gets a black screen, if an external monitor is connected it is shown as extended.

Any advise will be appreciated. 

That is unfortunate, this may be another edge case.

There is not much we can do for the R59 image here but there is a chance that a fix that just went in for another Tegra based device might help these blaze systems as well.

If the system is put on to beta channel such that it gets an update to 9592.66.0 does the display work again?

If this is the case the fix should be out on stable within the next two weeks with the R60 release. 

We tried beta channel, but the issue keeps happening on version 60.

Was it on beta as of yesterday evening? The build with the fix for big just went out yesterday, earlier R60 betas did not have any potential fixes for this, it needs to be 9592.66+. 

Thank you, the devices are working fine on the latest beta. 

Anyone have this problem after version 60? A customer updated to 60 just now and lo and behold, screen is dead again. Have not heard that it is widespread yet though. 

@Comment 162 Christia...
Yes! Some of our Chromebooks had pre-downloaded versions before 60 after unrestricting the pinned version of 54.
The latest versions are available via the Chromebook Recovery Utility. Create a recovery USB to get the latest version for these Chromebooks, and use it to update to 60+

Has this issue crept back in? We have a Chromebook running 54.0.2840.101 and get this again when updating to latest OS (62.0.3202.97) 

Thanks, Simon

I've had about 15 come to us with the black screen.  After they die, some of them will come back on for about 5 minutes to run the chrome recovery and then turn off as I start to sign in.  It is frustrating that there is no solution.  Neither going backwards or forwards in version helps.  I've been trying to send the motherboards in repair to AgParts because it is cheaper than a brand new one, but I'm not sure how long it will solve the problem as we just got our first ones back for repair.