Hi Philip: Did you mean this is happening only when gn arguments contain "Chrome" branding plus proprietory codecs?

 Issue 684560  has been merged into this issue.

This happens with default chromium build (w/o any branding or codecs).

See the merged bug for stack trace and other details.

Navid, please check if it is a regression in M57. 

No, I just pasted my args.gn in case it doesn't reproduce with other settings.

Hitting the same issue here.

#6: my args.gn

is_component_build = false #true
is_debug = true
use_goma = true
enable_nacl = false

remove_webcore_debug_symbols = true


#use_openh264 = true
#rtc_use_h264 = tru

I removed the DCHECK to continue working. But it might better to move this bug as P0, since Chrome cannot be used at all on macOS.

Bumping it up to P0 to have the DCHECK removed asap as jlebel@ suggested.

Navid, after the ad hoc fix, downgrade the bug to P1 again to keep investigating the real cause. This could be a sign of uninitialized fields in coalesced events through some cracks in plumbing.

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/cb1b5c8361a0758575517519d9914d8a74228997

commit cb1b5c8361a0758575517519d9914d8a74228997
Author: nzolghadr <nzolghadr@chromium.org>
Date: Thu Jan 26 17:40:12 2017

Comment a DCHECK that fails temporarily

BUG= 684292 

Review-Url: https://codereview.chromium.org/2659523004
Cr-Commit-Position: refs/heads/master@{#446367}

[modify] https://crrev.com/cb1b5c8361a0758575517519d9914d8a74228997/third_party/WebKit/Source/core/events/PointerEventFactory.cpp

The problem comes from this line
https://cs.chromium.org/chromium/src/content/browser/renderer_host/input/web_input_event_builders_mac.mm?q=web_input_event_builders_mac.mm&sq=package:chromium&dr&l=325

Since we send mouseenter as mousemove and we don't set the type on Mac correctly due to limitation of the APIs this will hit this DCHECK.
This CL is going to fix this.
https://codereview.chromium.org/2373733002/

What do you suggest?
I have landed another CL up to comment that DCHECK for now. But I'd say let's wait for the above CL to land to correctly fix the problem. Then I add back the DCHECK. Does this sound good?

This looks good, thanks.

We don't need the ad hoc fix in M-57 branch, so labeling this for M-58.

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/6b4cb83ed1e58193e8f3935ff406ce0ddb6310c6

commit 6b4cb83ed1e58193e8f3935ff406ce0ddb6310c6
Author: nzolghadr <nzolghadr@chromium.org>
Date: Mon Feb 13 20:49:16 2017

Re-enable DCHECK for pointertypes of coalesced events

The modified DCHECK was failing on MAC when mouse was
leaving window due to the missing pointer type from
the event Chrome was receiveing from the OS. So we
were setting the pointerType to Unknown.
After the following fix:
https://chromium.googlesource.com/chromium/src/+/e97eb52977dd7ed3936227f0d240cfeed865e520
we have a heuristic in place to set the type to either
pen or mouse. So this DCHECK can be enabled again.

BUG= 684292 

Review-Url: https://codereview.chromium.org/2688793003
Cr-Commit-Position: refs/heads/master@{#450075}

[modify] https://crrev.com/6b4cb83ed1e58193e8f3935ff406ce0ddb6310c6/third_party/WebKit/Source/core/events/PointerEventFactory.cpp

 Issue 693637  has been merged into this issue.

It's still failing. Reopening the bug.

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/47542d73322fa629f5d02975b7a2694089b28a58

commit 47542d73322fa629f5d02975b7a2694089b28a58
Author: alph <alph@chromium.org>
Date: Sat Feb 18 06:25:40 2017

Revert of Re-enable DCHECK for pointertypes of coalesced events (patchset #1 id:1 of https://codereview.chromium.org/2688793003/ )

Reason for revert:
Caused flakiness of WebRtcFromWebAccessibleResourceTest.GetUserMediaInWebAccessibleResourceFail

https://chromium-try-flakes.appspot.com/all_flake_occurrences?key=ahVzfmNocm9taXVtLXRyeS1mbGFrZXNyVgsSBUZsYWtlIktXZWJSdGNGcm9tV2ViQWNjZXNzaWJsZVJlc291cmNlVGVzdC5HZXRVc2VyTWVkaWFJbldlYkFjY2Vzc2libGVSZXNvdXJjZUZhaWwM

Original issue's description:
> Re-enable DCHECK for pointertypes of coalesced events
>
> The modified DCHECK was failing on MAC when mouse was
> leaving window due to the missing pointer type from
> the event Chrome was receiveing from the OS. So we
> were setting the pointerType to Unknown.
> After the following fix:
> https://chromium.googlesource.com/chromium/src/+/e97eb52977dd7ed3936227f0d240cfeed865e520
> we have a heuristic in place to set the type to either
> pen or mouse. So this DCHECK can be enabled again.
>
> BUG= 684292 
>
> Review-Url: https://codereview.chromium.org/2688793003
> Cr-Commit-Position: refs/heads/master@{#450075}
> Committed: https://chromium.googlesource.com/chromium/src/+/6b4cb83ed1e58193e8f3935ff406ce0ddb6310c6

TBR=mustaq@chromium.org,bokan@chromium.org,rbyers@chromium.org,nzolghadr@chromium.org
# Not skipping CQ checks because original CL landed more than 1 days ago.
BUG= 684292 

Review-Url: https://codereview.chromium.org/2706513002
Cr-Commit-Position: refs/heads/master@{#451427}

[modify] https://crrev.com/47542d73322fa629f5d02975b7a2694089b28a58/third_party/WebKit/Source/core/events/PointerEventFactory.cpp

Detected 3 new flakes for test/step "WebRtcFromWebAccessibleResourceTest.GetUserMediaInWebAccessibleResourceFail". To see the actual flakes, please visit https://chromium-try-flakes.appspot.com/all_flake_occurrences?key=ahVzfmNocm9taXVtLXRyeS1mbGFrZXNyVgsSBUZsYWtlIktXZWJSdGNGcm9tV2ViQWNjZXNzaWJsZVJlc291cmNlVGVzdC5HZXRVc2VyTWVkaWFJbldlYkFjY2Vzc2libGVSZXNvdXJjZUZhaWwM. This message was posted automatically by the chromium-try-flakes app. Since flakiness is ongoing, the issue was moved back into Sheriff Bug Queue (unless already there).

 Issue 694472  has been merged into this issue.

 Issue 694403  has been merged into this issue.

 Issue 693832  has been merged into this issue.

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/936e699a3399c5699d5ab30490707f8918e9fc08

commit 936e699a3399c5699d5ab30490707f8918e9fc08
Author: lanwei <lanwei@chromium.org>
Date: Wed Feb 22 15:35:51 2017

Disable the DCHECK on id because of some flaky tests

There are some flaky tests failed on
[1:1:0220/085659.710205:FATAL:PointerEventFactory.cpp(276)] Check failed:
mouseEvent.id == coalescedMouseEvent.id (2147483647 vs. 0)

We are currently disabling this DCHECK. We will inject id from low-level OS events
and add this DCHECK back again.

BUG= 684292 ,  694403 

Review-Url: https://codereview.chromium.org/2709033002
Cr-Commit-Position: refs/heads/master@{#452064}

[modify] https://crrev.com/936e699a3399c5699d5ab30490707f8918e9fc08/third_party/WebKit/Source/core/events/PointerEventFactory.cpp

Detected 6 new flakes for test/step "FindRequestManagerTests/FindRequestManagerTest.NavigateFrame/1". To see the actual flakes, please visit https://chromium-try-flakes.appspot.com/all_flake_occurrences?key=ahVzfmNocm9taXVtLXRyeS1mbGFrZXNySQsSBUZsYWtlIj5GaW5kUmVxdWVzdE1hbmFnZXJUZXN0cy9GaW5kUmVxdWVzdE1hbmFnZXJUZXN0Lk5hdmlnYXRlRnJhbWUvMQw. This message was posted automatically by the chromium-try-flakes app.

Detected 3 new flakes for test/step "WebViewInteractiveTests/WebViewDragDropInteractiveTest.DragDropWithinWebView/1". To see the actual flakes, please visit https://chromium-try-flakes.appspot.com/all_flake_occurrences?key=ahVzfmNocm9taXVtLXRyeS1mbGFrZXNyWQsSBUZsYWtlIk5XZWJWaWV3SW50ZXJhY3RpdmVUZXN0cy9XZWJWaWV3RHJhZ0Ryb3BJbnRlcmFjdGl2ZVRlc3QuRHJhZ0Ryb3BXaXRoaW5XZWJWaWV3LzEM. This message was posted automatically by the chromium-try-flakes app.

Removing sheriff label, since no new flakes seem to have happened today. (It'll get re-added if new flakes occur.)

treib@ the flakes don't seem to be related to this bug at all. They are all related to the  issue 694742 . How does the chromium-try-flakes continue commenting on this bug?

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/ef2a8e07616635ec90e8b30a95fb8a43a7f0ec9a

commit ef2a8e07616635ec90e8b30a95fb8a43a7f0ec9a
Author: nzolghadr <nzolghadr@chromium.org>
Date: Tue Feb 28 15:20:22 2017

Add back DCHECK for pointer type in PointerEventFactory

This DCHECK was originally commented here
https://codereview.chromium.org/2659523004/

and after the fix landed here
https://codereview.chromium.org/2688793003/

but it got mistakenly reverted due to  crbug.com/694742 
which had nothing to do with this DCHECK. So this
re-re-enables this pointertype DCHECK for coalesced
pointer events.

BUG= 684292 

Review-Url: https://codereview.chromium.org/2722643002
Cr-Commit-Position: refs/heads/master@{#453603}

[modify] https://crrev.com/ef2a8e07616635ec90e8b30a95fb8a43a7f0ec9a/third_party/WebKit/Source/core/events/PointerEventFactory.cpp