New issue
Advanced search Search tips
Note: Color blocks (like or ) mean that a user may not be available. Tooltip shows the reason.

Issue 667527 link

Starred by 3 users

Issue metadata

Status: Fixed
Owner:
Last visit > 30 days ago
Closed: Apr 2017
Cc:
EstimatedDays: ----
NextAction: ----
OS: ----
Pri: 3
Type: Bug-Regression



Sign in to add a comment

Build broken with enable_webrtc=false w/ MediaDeviceInfo issue

Project Member Reported by mbjorge@google.com, Nov 21 2016

Issue description

<b>Version: <Kenneth, what is the frequency?></b>
<b>OS: <please tell me it's not XP></b>

What steps will reproduce the problem?
(1) Get a recent checkout
(2) Set enable_webrtc=false
(3) ninja content/renderer

What is the expected result?
It builds cleanly

What happens instead?
In file included from ../../content/renderer/pepper/pepper_media_device_manager.cc:5:
In file included from ../../content/renderer/pepper/pepper_media_device_manager.h:15:
In file included from gen/content/common/media/media_devices.mojom.h:15:
In file included from ../../mojo/public/cpp/bindings/associated_interface_ptr.h:23:
In file included from ../../mojo/public/cpp/bindings/lib/associated_interface_ptr_state.h:15:
In file included from ../../base/bind.h:8:
../../base/bind_internal.h:214:31: error: no viable conversion from 'const vector<content::MediaDeviceInfo, allocator<content::MediaDeviceInfo>>' to 'const vector<std::__debug::vector<content::MediaDeviceInfo, std::allocator<content::MediaDeviceInfo> >, allocator<std::__debug::vector<content::MediaDeviceInfo, std::allocator<content::MediaDeviceInfo> >>>'
    return (receiver.*method)(std::forward<RunArgs>(args)...);



 

Comment 1 by mbjorge@google.com, Nov 21 2016

Cc: guidou@chromium.org

Comment 2 by mbjorge@google.com, Nov 21 2016

Cc: brettw@chromium.org

Comment 3 by mbjorge@google.com, Nov 21 2016

Ah, I figured out the issue here. Will put up a CL. (I was having trouble parsing all the compiler template output at first).

Comment 4 by mbjorge@google.com, Nov 21 2016

Owner: mbjorge@chromium.org

Comment 5 by mbjorge@google.com, Nov 21 2016

(Added brettw@ becuase I just saw https://groups.google.com/a/chromium.org/forum/#!topic/chromium-dev/ZiSmKRiQiic and thought it was tangentially related)
Project Member

Comment 6 by bugdroid1@chromium.org, Nov 22 2016

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/53189bba54cc9bb057f9e35ede472088bed73a1d

commit 53189bba54cc9bb057f9e35ede472088bed73a1d
Author: mbjorge <mbjorge@chromium.org>
Date: Tue Nov 22 02:39:22 2016

[Buildfix] Fix enable_webrtc=false build.

BUG= 667527 
TEST=ninja content/renderer

Review-Url: https://codereview.chromium.org/2522663003
Cr-Commit-Position: refs/heads/master@{#433753}

[modify] https://crrev.com/53189bba54cc9bb057f9e35ede472088bed73a1d/content/renderer/pepper/pepper_media_device_manager.cc

Comment 7 by gca...@gmail.com, Jan 30 2017

This patch renders flash unusable if built with enable_webrtc=false (browser tab crashes if flash content is accessed), however flash works as expected with enable_webrtc=true.

os: Arch Linux, kernel: 4.9.6-1-ARCH
version: chromium 56.0.2924.76-1 with inox-patchset (https://github.com/gcarq/inox-patchset)
flash version: pepper-flash 24.0.0.194-1

Relevant issue: https://github.com/gcarq/inox-patchset/issues/57
Hm, it seems unlikely it was because of this patch. This patch just updated some old code that got missed when an interface changed (because none of the chromium bots build with enable_webrtc=false). The code literally didn't even compile with enable_webrtc=false without this patch.

I feel like it's highly more likely that things are just broken in general with enable_webrtc=false since it seems to not get tested much (if at all)

Comment 9 by gca...@gmail.com, Jan 31 2017

Alright, thank you for clarification. Why is there no chromium bot running with enable_webrtc=false?, is there something I can do to push that?

PS: Should I raise a new issue regarding the broken pepper flash?
I do have a bug for make a bot with enable_webrtc=fales (https://bugs.chromium.org/p/chromium/issues/detail?id=667546), but it would be doing a chromecast build and not a normal Chrome build, so hard to say if it would catch the issue you ran into.

I would say you should file a new bug. If you can, include a symbolized callstack of the crash 
Status: Fixed (was: Untriaged)
(There is now a Cast Audio Linux bot that builds with enable_webrtc=true)

Sign in to add a comment