Using the per-revision bisect providing the bisect results,
Good build: 55.0.2874.0 (Revision: 421409).
Bad build: 55.0.2875.0 (Revision: 421703).

You are probably looking for a change made after 421527 (known good), but no later than 421528 (first known bad).
CHANGELOG URL:
 https://chromium.googlesource.com/chromium/src/+log/34a7f279d80becf1b33570de4606c58dbe51dd9e..5248d2a4cc2ac706f0404b6c6edf196f5b491539

Adding RB Label as this is a recent Regression.

@dtapuska -- Could you please look into the issue, pardon me if it has nothing to do with your changes and if possible please assign it to concern owner.
Thank You.

Original changed reverted. It revealed a pre-existing problem with the code that is being fixed separately before re-landing my original change.

dtapuska@, Thank you for the revert.

Above issue is not reproducible on latest chrome version i.e 55.0.2880.0 and seems to be working as intended.

OS tested :Windows(7,8,8.1,10), Mac(10.11.6)
Chrome version :55.0.2880.0 (Official Build) a20982650ce3e0f7ba76278fdac132a66e1b6a8f-refs/heads/master@{#422654}(32/64-bit)

Since the build 55.0.2880.0 for Linux has failed, could not test the fix on Linux

The above revert fixed this issue on 'Linux' as well for chrome#55.0.2880.0.

Thank you!

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/5bf106420af2df80286f4f404c44dbf818469db1

commit 5bf106420af2df80286f4f404c44dbf818469db1
Author: dtapuska <dtapuska@chromium.org>
Date: Wed Oct 05 16:30:47 2016

RenderWidgetHostViewChildFrame's called a virtual in its ctor.

It appears that calling SetView in the ctor can lead to a path where
SetNeedsBeginFrame is invoked on the newly constructed object.
Since RenderWidgetHostViewGuest is a subclass of RenderWidgetHostViewChildFrame
its virtual function would not get invoked since the vtable had not
be constructed yet.

Change SetView to be called after the class is fully
instantiated.

Also fix RenderWidgetHostViewAura since calling this correctly
now leads to an assert that the observer is only added once.
So track whether the observer has been added so that when the
begin frame source is added an assertion doesn't occur.

BUG= 652212 
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:linux_site_isolation

Review-Url: https://codereview.chromium.org/2385333002
Cr-Commit-Position: refs/heads/master@{#423182}

[modify] https://crrev.com/5bf106420af2df80286f4f404c44dbf818469db1/content/browser/frame_host/render_frame_host_impl.cc
[modify] https://crrev.com/5bf106420af2df80286f4f404c44dbf818469db1/content/browser/frame_host/render_widget_host_view_child_frame.cc
[modify] https://crrev.com/5bf106420af2df80286f4f404c44dbf818469db1/content/browser/frame_host/render_widget_host_view_child_frame.h
[modify] https://crrev.com/5bf106420af2df80286f4f404c44dbf818469db1/content/browser/frame_host/render_widget_host_view_child_frame_unittest.cc
[modify] https://crrev.com/5bf106420af2df80286f4f404c44dbf818469db1/content/browser/frame_host/render_widget_host_view_guest.cc
[modify] https://crrev.com/5bf106420af2df80286f4f404c44dbf818469db1/content/browser/frame_host/render_widget_host_view_guest.h
[modify] https://crrev.com/5bf106420af2df80286f4f404c44dbf818469db1/content/browser/frame_host/render_widget_host_view_guest_unittest.cc
[modify] https://crrev.com/5bf106420af2df80286f4f404c44dbf818469db1/content/browser/renderer_host/render_widget_host_view_aura.cc
[modify] https://crrev.com/5bf106420af2df80286f4f404c44dbf818469db1/content/browser/renderer_host/render_widget_host_view_aura.h
[modify] https://crrev.com/5bf106420af2df80286f4f404c44dbf818469db1/content/browser/renderer_host/render_widget_host_view_aura_unittest.cc
[modify] https://crrev.com/5bf106420af2df80286f4f404c44dbf818469db1/content/browser/renderer_host/render_widget_host_view_mac_unittest.mm
[modify] https://crrev.com/5bf106420af2df80286f4f404c44dbf818469db1/content/browser/web_contents/web_contents_view_child_frame.cc
[modify] https://crrev.com/5bf106420af2df80286f4f404c44dbf818469db1/content/browser/web_contents/web_contents_view_guest.cc

dtapuska@: Please close the issue if there is no further work to be done here.

Thanks in advance!

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/5bf106420af2df80286f4f404c44dbf818469db1

commit 5bf106420af2df80286f4f404c44dbf818469db1
Author: dtapuska <dtapuska@chromium.org>
Date: Wed Oct 05 16:30:47 2016

RenderWidgetHostViewChildFrame's called a virtual in its ctor.

It appears that calling SetView in the ctor can lead to a path where
SetNeedsBeginFrame is invoked on the newly constructed object.
Since RenderWidgetHostViewGuest is a subclass of RenderWidgetHostViewChildFrame
its virtual function would not get invoked since the vtable had not
be constructed yet.

Change SetView to be called after the class is fully
instantiated.

Also fix RenderWidgetHostViewAura since calling this correctly
now leads to an assert that the observer is only added once.
So track whether the observer has been added so that when the
begin frame source is added an assertion doesn't occur.

BUG= 652212 
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:linux_site_isolation

Review-Url: https://codereview.chromium.org/2385333002
Cr-Commit-Position: refs/heads/master@{#423182}

[modify] https://crrev.com/5bf106420af2df80286f4f404c44dbf818469db1/content/browser/frame_host/render_frame_host_impl.cc
[modify] https://crrev.com/5bf106420af2df80286f4f404c44dbf818469db1/content/browser/frame_host/render_widget_host_view_child_frame.cc
[modify] https://crrev.com/5bf106420af2df80286f4f404c44dbf818469db1/content/browser/frame_host/render_widget_host_view_child_frame.h
[modify] https://crrev.com/5bf106420af2df80286f4f404c44dbf818469db1/content/browser/frame_host/render_widget_host_view_child_frame_unittest.cc
[modify] https://crrev.com/5bf106420af2df80286f4f404c44dbf818469db1/content/browser/frame_host/render_widget_host_view_guest.cc
[modify] https://crrev.com/5bf106420af2df80286f4f404c44dbf818469db1/content/browser/frame_host/render_widget_host_view_guest.h
[modify] https://crrev.com/5bf106420af2df80286f4f404c44dbf818469db1/content/browser/frame_host/render_widget_host_view_guest_unittest.cc
[modify] https://crrev.com/5bf106420af2df80286f4f404c44dbf818469db1/content/browser/renderer_host/render_widget_host_view_aura.cc
[modify] https://crrev.com/5bf106420af2df80286f4f404c44dbf818469db1/content/browser/renderer_host/render_widget_host_view_aura.h
[modify] https://crrev.com/5bf106420af2df80286f4f404c44dbf818469db1/content/browser/renderer_host/render_widget_host_view_aura_unittest.cc
[modify] https://crrev.com/5bf106420af2df80286f4f404c44dbf818469db1/content/browser/renderer_host/render_widget_host_view_mac_unittest.mm
[modify] https://crrev.com/5bf106420af2df80286f4f404c44dbf818469db1/content/browser/web_contents/web_contents_view_child_frame.cc
[modify] https://crrev.com/5bf106420af2df80286f4f404c44dbf818469db1/content/browser/web_contents/web_contents_view_guest.cc

Issue 660107 has been merged into this issue.

[Automated comment] removing mislabelled merge-merged-2840

Removing in-correct label: "merge-merged-2840".

dtapuska@, could you please merge the above fix to 2840 branch? This is still broken on Latest Stable#54.0.2840.87 for Mac OS X (Ref: crbug/660107).

Thank you!

It must be broken on M54 for some other reason. The change that originally caused this didn't get into M54.

It is true that this bug existed in the code before but perhaps you should do a bisect on M54 and find the cause if on Mac OSX.

Note that merging into M54 will require approval and a post mortem.

Sounds good. I will do a bisect on M54 and update the thread.

This seems to be the actual culprit: https://chromium.googlesource.com/chromium/src/+/d020d554c9ffc0476d43f5364a47c522b64e3d9a

Thank you!

And it is fixed in M55 right? Since enne's change was in July it was in M53 as well which means this isn't serious and should just wait for the fix in M56.

Yes, except M54 Stable#54.0.2840.87 other channels (Beta, Dev & Canary) are good.

This regression is introduced in M54#54.0.2840.59 and M53#53.0.2785.143 is working fine w/o any issues.

I still think that this isn't a M54 merge request it should just wait until M55 fixes it.

sure, thank you!