Skip to content
This repository has been archived by the owner on Apr 3, 2020. It is now read-only.

[Android] Fix getUserMedia low FPS problem if constraint out of range #193

Open
wants to merge 1,917 commits into
base: master
Choose a base branch
from
Open

[Android] Fix getUserMedia low FPS problem if constraint out of range #193

wants to merge 1,917 commits into from

Conversation

js0701
Copy link

@js0701 js0701 commented Sep 29, 2014

@xingnan @hmin
eg. Constraints requires minFramerate at 31 FPS, while the camera only support FPS range 10 to 30.
In this case, VideoCapture Android will return a min Framerate 10(or 11), and it is used to calurate
the expectedNextFrameTime when one frame of data is captured from camera: in video_capture_android.cc,
on FrameAvailable function, if it judge the frame arrive too early, it will discard the frame.
The problem is that Android Camera device can only accept setting of a FPS range, when set the range of
FPS from 11 to 30, the FPS can be 11 or 30. If the actual FPS is 30 and the returned FPS to calculate the
expectedNextFrameTime is 11, many frames will be thought to arrive too early, but it is not fact. To avoid
this problem, use the default setting to be max FPS of the range. If the range is 11 to 30, use 30 instead
avoid the frames to be discard.
The strategy here is avoid discard frame as much as possible sine the frame has been captured from Camera.
BUG=XWALK-2643

Per and others added 30 commits September 2, 2014 12:28
Change VideoTrackAdapter to not drop frames if the source frame rate …
9ec3990 
…is known. Also change remote video and pepper plugin MediaStreamVideoSource implementations to not set a guessed frame rate.

BUG= 394315

Review URL: https://codereview.chromium.org/517973002

Cr-Commit-Position: refs/heads/master@{#292800}
(cherry picked from commit b18d81f)

Review URL: https://codereview.chromium.org/526173003

Cr-Commit-Position: refs/branch-heads/2125@{#171}
Cr-Branched-From: b68026d-refs/heads/master@{#290040}
[Most Visited] Check for Sync state when using SuggestionsService.
76082ea 
BUG=386454
[email protected], [email protected]
TBR=erikwright
TEST=SuggestionsServiceTest

Review URL: https://codereview.chromium.org/473123002

Cr-Commit-Position: refs/heads/master@{#291116}
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@291116 0039d316-1c4b-4281-b951-d872f2087c98
(cherry picked from commit 1d1ab15)

Review URL: https://codereview.chromium.org/527943002

Cr-Commit-Position: refs/branch-heads/2125@{#172}
Cr-Branched-From: b68026d-refs/heads/master@{#290040}
[Sync] Fix a crash when sign into sync.
59533ea 
The crash was caused by a dangling pointer to a deleted observer
registered by MostVisitedSites.

BUG=406996

Review URL: https://codereview.chromium.org/501973004

Cr-Commit-Position: refs/heads/master@{#292034}
(cherry picked from commit 94a4210)

Review URL: https://codereview.chromium.org/533663002

Cr-Commit-Position: refs/branch-heads/2125@{#173}
Cr-Branched-From: b68026d-refs/heads/master@{#290040}
@mfomitchev
Fixing GestureNav on HDPI displays.
e772e6b 
The GestireNav screenshot dimensions are density-independent, because when we capture the screenshot we scale it to RVH's bounds: https://code.google.com/p/chromium/codesearch#chromium/src/content/browser/frame_host/navigation_entry_screenshot_manager.cc&q=navigation_entry_sc&sq=package:chromium&type=cs&l=136

Thus we don't need to scale it again when we paint during the overscroll gesture.

BUG=402280
TBR=sky

Review URL: https://codereview.chromium.org/507083002

Cr-Commit-Position: refs/heads/master@{#292426}
(cherry picked from commit 7aa72c8)

Review URL: https://codereview.chromium.org/529023002

Cr-Commit-Position: refs/branch-heads/2125@{#174}
Cr-Branched-From: b68026d-refs/heads/master@{#290040}
[Android] Speculative fix for crash during overscroll effect creation
7b87ff7 
Crash reports indicate that the WindowAndroid's Compositor may be NULL
during |RenderWidgetHostViewAndroid::SetContentViewCore()|. While this
is a somewhat surprising result, it is in theory possible and should
probably not be fatal. Avoid creating the overscroll effect in this
case, instead creating it after the compositor is (re-)attached.

BUG=407208
[email protected],[email protected]

Review URL: https://codereview.chromium.org/503813003

Cr-Commit-Position: refs/heads/master@{#291835}
(cherry picked from commit 7936233)

Review URL: https://codereview.chromium.org/532733002

Cr-Commit-Position: refs/branch-heads/2125@{#175}
Cr-Branched-From: b68026d-refs/heads/master@{#290040}
[Android] Clear backing SkBitmap before rendering the overscroll effect
61a2edc 
Skia makes no guarantees about the contents of a newly allocated
SkBitmap. Consequently, it's possible that the bitmap allocated for the
overscroll effect isn't zeroed before the effect itself is painted.
Prevent related issues by always clearing the allocated SkBitmap's
contents (to fully transparent) before painting the effect.

BUG=408225
[email protected]

Review URL: https://codereview.chromium.org/484373004

Cr-Commit-Position: refs/heads/master@{#292280}
(cherry picked from commit 46c0ec3)

Review URL: https://codereview.chromium.org/527363002

Cr-Commit-Position: refs/branch-heads/2125@{#176}
Cr-Branched-From: b68026d-refs/heads/master@{#290040}
aw: Ensure fallback tick unsets |block_invalidates_|
abd5de0 
Currently, if |compositor_needs_continuous_invalidate_| is unset before
PostFallbackTick or FallbackTickFired, then the fallback ticks are
cancelled, but we never unset |block_invalidates_|, which will block
invalidates indefinitely.

Fix is simply always unset |block_invalidates_| even if we skip
the fallback tick.

BUG=408779

Review URL: https://codereview.chromium.org/512403002

Cr-Commit-Position: refs/heads/master@{#292511}
(cherry picked from commit d3ba720)

[email protected]

Review URL: https://codereview.chromium.org/528953003

Cr-Commit-Position: refs/branch-heads/2125@{#177}
Cr-Branched-From: b68026d-refs/heads/master@{#290040}
Prevent the screen from sleeping when the desktop is being captured.
a1cc1f3 
BUG=405784

Review URL: https://codereview.chromium.org/502103003

Cr-Commit-Position: refs/heads/master@{#291908}
(cherry picked from commit aea4a57)

Review URL: https://codereview.chromium.org/531843002

Cr-Commit-Position: refs/branch-heads/2125@{#178}
Cr-Branched-From: b68026d-refs/heads/master@{#290040}
@SergeyUlanov
Fix DesktopMediaPickerViews to handle background parent_web_contents
6debe58 
After r283702 DesktopMediaPickerViews is getting parent_web_contents even
when it corresponds to a background extension page.
DesktopMediaPickerViews was always trying to create a modal dialog when
web_contents is specified, and that doesn't work for background pages.
Now it explicitly checks if the web_contents is a background page and
in that case shows the picker in a separate window.

Also updated example for desktop capture API to make it easier to test
this scenario.

BUG=402579

Review URL: https://codereview.chromium.org/501713002

Cr-Commit-Position: refs/heads/master@{#292055}
(cherry picked from commit 71b32fc)

[email protected]

Review URL: https://codereview.chromium.org/522483007

Cr-Commit-Position: refs/branch-heads/2125@{#179}
Cr-Branched-From: b68026d-refs/heads/master@{#290040}
[Android] Fix overflow menu enter tablet animation
b1838a4 
- On tablets the new anchor point was 6dp off due to the new shadow assets.  To fix
this add a 6dp translate to the enter animation that will move the asset into the
correct location as the scale animation runs.  This effectively mimics the old
animation but allows the offset to be included.

BUG=405352

Review URL: https://codereview.chromium.org/488863002

Cr-Commit-Position: refs/heads/master@{#290773}
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@290773 0039d316-1c4b-4281-b951-d872f2087c98
(cherry picked from commit 6f2a9c8)

Review URL: https://codereview.chromium.org/533813002

Cr-Commit-Position: refs/branch-heads/2125@{#180}
Cr-Branched-From: b68026d-refs/heads/master@{#290040}
@notwaldorf
[Merge] [Win, Linux] Always display the avatar button text at the sam…
5ea6216 
…e size.

The problem is that some locales (Hindi) or platforms (Linux) have a
bigger base font size for the default font (crbug.com/405994). If this
is the case, we should decrease the font size to the one we expect,
so that the font fits in the button correctly.

Just adjusting the baseline is not always enough, as the new avatar
button has a fixed size, and the text might not fit correctly
regardless of the baseline adjustments.

Screenshots: https://drive.google.com/folderview?id=0B1B1Up4p2NRMRHc5WTBxZWNrRm8&usp=sharing

BUG=403466
[email protected]
TEST=Start Chrome with --enable-new-avatar-menu on Linux, or with --enable-new-avatar-menu and --lang-hi on Windows. The avatar button text should fit nicely in the button.

Review URL: https://codereview.chromium.org/470053004

Cr-Commit-Position: refs/heads/master@{#292715}
(cherry picked from commit 899ae44)

Review URL: https://codereview.chromium.org/534713002

Cr-Commit-Position: refs/branch-heads/2125@{#181}
Cr-Branched-From: b68026d-refs/heads/master@{#290040}
@rch-chromium-org
Merge f9f103c
603c048 
Fix bug in QuicHttpStream::SendRequest. If the session is closed before SendRequest is called, a CHECK would fire in SendRequest().

BUG=409101
TBR=amineer

Review URL: https://codereview.chromium.org/523463007

Cr-Commit-Position: refs/heads/master@{#292775}
(cherry picked from commit f9f103c)

Review URL: https://codereview.chromium.org/531903002

Cr-Commit-Position: refs/branch-heads/2125@{#182}
Cr-Branched-From: b68026d-refs/heads/master@{#290040}
@notwaldorf
[Merge] [Mac] Add tab and keyboard navigation to the new avatar bubble
0881617 
Tabbing should navigate through all the visible controls.

I've changed the EditablePhotoButton to actually be a button and
not just an image, so that tabbing and pressing enter actually
works.

BUG=405056
[email protected]

Review URL: https://codereview.chromium.org/489143002

Cr-Commit-Position: refs/heads/master@{#292455}
(cherry picked from commit aaacc30)

Review URL: https://codereview.chromium.org/533833003

Cr-Commit-Position: refs/branch-heads/2125@{#183}
Cr-Branched-From: b68026d-refs/heads/master@{#290040}
@inexorabletash
IndexedDB: Decode correct int type for database name->id map
37897e4 
Database enumeration was decoding id as VarInt, but it's a plain Int.

[email protected]
BUG=407015

Review URL: https://codereview.chromium.org/503913002

Cr-Commit-Position: refs/heads/master@{#291792}
(cherry picked from commit 3f53c98)

Review URL: https://codereview.chromium.org/536463002

Cr-Commit-Position: refs/branch-heads/2125@{#184}
Cr-Branched-From: b68026d-refs/heads/master@{#290040}
Incrementing VERSION to 38.0.2125.43
945db46 
Cr-Commit-Position: refs/branch-heads/2125@{#185}
Cr-Branched-From: b68026d-refs/heads/master@{#290040}
@dzhioev
Fix stale keyboard bounds on login screen
0621a5e 
TBR=nkostylev
BUG=401667

Review URL: https://codereview.chromium.org/495923002

Cr-Commit-Position: refs/heads/master@{#291281}
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@291281 0039d316-1c4b-4281-b951-d872f2087c98
(cherry picked from commit bd7e97a)

Review URL: https://codereview.chromium.org/528043002

Cr-Commit-Position: refs/branch-heads/2125@{#186}
Cr-Branched-From: b68026d-refs/heads/master@{#290040}
Merge: Update easy unlock 'unlocked' icon resource
6efd10b 
The icon is changed so it's more distinguishable from the 'locked' icon

BUG=405770

Review URL: https://codereview.chromium.org/508703002

Cr-Commit-Position: refs/heads/master@{#291963}
(cherry picked from commit 86b7dd0)

[email protected]

Review URL: https://codereview.chromium.org/531253002

Cr-Commit-Position: refs/branch-heads/2125@{#187}
Cr-Branched-From: b68026d-refs/heads/master@{#290040}
@stevenjb
Fix GSM in Setings
725b744 
This fixes a bug introduced in https://codereview.chromium.org/427903004

BUG=406114
[email protected]

Review URL: https://codereview.chromium.org/492373002

Cr-Commit-Position: refs/heads/master@{#291415}
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@291415 0039d316-1c4b-4281-b951-d872f2087c98
(cherry picked from commit 8e0e343)

Review URL: https://codereview.chromium.org/533933004

Cr-Commit-Position: refs/branch-heads/2125@{#188}
Cr-Branched-From: b68026d-refs/heads/master@{#290040}
Fix OEM menu issue
36e3f28 
- Some OEMs don't let us change the background, so the padding is incorrect.
- Also including sizing the menu too large causes clipping.

BUG=398837

Review URL: https://codereview.chromium.org/469673006

Cr-Commit-Position: refs/heads/master@{#291193}
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@291193 0039d316-1c4b-4281-b951-d872f2087c98
(cherry picked from commit 2b66232)

Review URL: https://codereview.chromium.org/531263002

Cr-Commit-Position: refs/branch-heads/2125@{#189}
Cr-Branched-From: b68026d-refs/heads/master@{#290040}
Reverting since merge is breaking build.
7a0797a 
Revert "Fix DesktopMediaPickerViews to handle background parent_web_contents"

This reverts commit 6debe58.

BUG=None

Review URL: https://codereview.chromium.org/534033002

Cr-Commit-Position: refs/branch-heads/2125@{#190}
Cr-Branched-From: b68026d-refs/heads/master@{#290040}
@quisquous
cc: Don't infinitely throttle large raster tasks
ed16153 
Previously, if a raster task had a resource size that was larger than
the throttling limit for the raster worker, that task would never get
scheduled.  This leads to starvation for that task, preventing tree
activation, and causing freezes.

The fix is to always allow such large tasks if it is the first task.

[email protected]
BUG=403446

Review URL: https://codereview.chromium.org/489293002

Cr-Commit-Position: refs/heads/master@{#291484}
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@291484 0039d316-1c4b-4281-b951-d872f2087c98
(cherry picked from commit 604e202)

Review URL: https://codereview.chromium.org/534823002

Cr-Commit-Position: refs/branch-heads/2125@{#191}
Cr-Branched-From: b68026d-refs/heads/master@{#290040}
@jrummell-chromium
Merge 291602 "Changing the behavior of ReleaseSession() from Close() …
baa7b33 
…to Remove()"

With the recent EME spec changes, ReleaseSession() got replaced by
CloseSession() and RemoveSession(). Until all the changes to support
both methods are in place, calls to ReleaseSession() should call
RemoveSession() in order for existing prefixed EME applications
to continue to work.

Unprefixed Close() calls now call RemoveSession(), and thus don't do
the correct thing. This will be fixed when both CloseSession() and
RemoveSession() are passed through Pepper.

BUG=406606
TEST=All EME browser_tests pass

Review URL: https://codereview.chromium.org/497153005

Cr-Commit-Position: refs/heads/master@{#291602}
(cherry picked from commit ec6d38e)

[email protected]

Review URL: https://codereview.chromium.org/531313002

Cr-Commit-Position: refs/branch-heads/2125@{#192}
Cr-Branched-From: b68026d-refs/heads/master@{#290040}
Incrementing VERSION to 38.0.2125.44
0ec9fbd 
Cr-Commit-Position: refs/branch-heads/2125@{#193}
Cr-Branched-From: b68026d-refs/heads/master@{#290040}
Incrementing VERSION to 38.0.2125.45
d5ad37b 
Cr-Commit-Position: refs/branch-heads/2125@{#194}
Cr-Branched-From: b68026d-refs/heads/master@{#290040}
@liutikas
[Android] Allow passing in extra intent flags for share menu item.
519cf33 
BUG=409104
b/17294369

[email protected]

Review URL: https://codereview.chromium.org/510023003

Cr-Commit-Position: refs/heads/master@{#293015}

Review URL: https://codereview.chromium.org/534863002

Cr-Commit-Position: refs/branch-heads/2125@{#195}
Cr-Branched-From: b68026d-refs/heads/master@{#290040}
Incrementing VERSION to 38.0.2125.46
771ca28 
Cr-Commit-Position: refs/branch-heads/2125@{#196}
Cr-Branched-From: b68026d-refs/heads/master@{#290040}
Remove Files.app's own profile badge.
51721a5 
Now that the Files.app uses default system header, the own profile badge is redundant.

BUG=408816
TEST=confirmed manually, ran browser_tests
[email protected]

Review URL: https://codereview.chromium.org/520623002

Cr-Commit-Position: refs/heads/master@{#292827}
(cherry picked from commit 14cea1d)

Review URL: https://codereview.chromium.org/535763002

Cr-Commit-Position: refs/branch-heads/2125@{#197}
Cr-Branched-From: b68026d-refs/heads/master@{#290040}
@SergeyUlanov
Fix DesktopMediaPickerViews to handle background parent_web_contents
fa2d622 
After r283702 DesktopMediaPickerViews is getting parent_web_contents even
when it corresponds to a background extension page.
DesktopMediaPickerViews was always trying to create a modal dialog when
web_contents is specified, and that doesn't work for background pages.
Now it explicitly checks if the web_contents is a background page and
in that case shows the picker in a separate window.

Also updated example for desktop capture API to make it easier to test
this scenario.

BUG=402579

Review URL: https://codereview.chromium.org/501713002

Cr-Commit-Position: refs/heads/master@{#292055}

[email protected]

Review URL: https://codereview.chromium.org/529293002

Cr-Commit-Position: refs/branch-heads/2125@{#198}
Cr-Branched-From: b68026d-refs/heads/master@{#290040}
Revert "[Android WebView] Terminate execution of stuck JS code on nav…
8f03226 
…igation requests"

This reverts commit 69538c3.

As described in the bug, the proposed approach is invalid, as Blink's bindings
code isn't ready to deal with terminated scripts properly, resulting in random
crashes. Hence reverting the patch.

BUG=390906

Review URL: https://codereview.chromium.org/532313002

Cr-Commit-Position: refs/branch-heads/2125@{#199}
Cr-Branched-From: b68026d-refs/heads/master@{#290040}
Merge: Show search box on settings page in ChromeOS guest mode.
927fde1 
BUG=408030
[email protected]
TEST=In ChromeOS: Open a guest. Go to settings. Verify search box there.
On desktop: Open a guest. Go to settings. Verify search box not there.
[email protected]

Review URL: https://codereview.chromium.org/517633002

Cr-Commit-Position: refs/heads/master@{#292935}
(cherry picked from commit 5aa9a79)

Review URL: https://codereview.chromium.org/534363003

Cr-Commit-Position: refs/branch-heads/2125@{#200}
Cr-Branched-From: b68026d-refs/heads/master@{#290040}
@rakuco rakuco force-pushed the master branch 2 times, most recently from 5d9bef8 to 18a33ff Compare September 16, 2015 10:57
mrunalk pushed a commit that referenced this pull request Sep 22, 2015
Incrementing VERSION to 46.0.2490.22
77498a9 
Cr-Commit-Position: refs/branch-heads/2490@{#193}
Cr-Branched-From: 7790a35-refs/heads/master@{#344925}
@rakuco rakuco force-pushed the master branch 2 times, most recently from 830d9f6 to 5ad5a18 Compare November 19, 2015 09:27
mrunalk pushed a commit that referenced this pull request Dec 18, 2015
@johnmellor
Remove RuntimeEnabled from NotificationEventInit
cccd9ad 
This is needed to complete the shipping of action buttons.
It was accidentally not included in https://crrev.com/358304
(or https://codereview.chromium.org/1487763003).

See also https://crbug.com/546006.

BUG=513671

Review URL: https://codereview.chromium.org/1489773003

Cr-Commit-Position: refs/heads/master@{#362423}
(cherry picked from commit a193248)

Review URL: https://codereview.chromium.org/1489203002 .

Cr-Commit-Position: refs/branch-heads/2564@{#193}
Cr-Branched-From: 1283eca-refs/heads/master@{#359700}
@rakuco rakuco force-pushed the master branch 2 times, most recently from 1072537 to 512d729 Compare January 21, 2016 15:50
@rakuco rakuco force-pushed the master branch 2 times, most recently from 0f31e18 to 3da217c Compare February 4, 2016 11:32
rakuco pushed a commit that referenced this pull request Mar 3, 2016
@yfriedman
Add flag to disable mmap in all sql connection in work build
bf25518 
Work Chrome runs in environment without a proper mmap support.
Disable it for all work chrome.

BUG=554269
[email protected],[email protected]

Review URL: https://codereview.chromium.org/1637683003

Cr-Commit-Position: refs/heads/master@{#372179}
(cherry picked from commit 7ce37c1)

Review URL: https://codereview.chromium.org/1649023002 .

Cr-Commit-Position: refs/branch-heads/2623@{#193}
Cr-Branched-From: 92d7753-refs/heads/master@{#369907}
huningxin pushed a commit to huningxin/chromium-croswalk that referenced this pull request May 17, 2016
Incrementing VERSION to 50.0.2661.29
206c8e3 
Cr-Commit-Position: refs/branch-heads/2661@{crosswalk-project#193}
Cr-Branched-From: ef6f6ae-refs/heads/master@{#378081}
huningxin pushed a commit to huningxin/chromium-croswalk that referenced this pull request Oct 9, 2016
Incrementing VERSION to 52.0.2743.27
6ad5924 
Cr-Commit-Position: refs/branch-heads/2743@{crosswalk-project#193}
Cr-Branched-From: 2b3ae3b-refs/heads/master@{#394939}
imreotto pushed a commit to tenta-browser/chromium-crosswalk that referenced this pull request Nov 2, 2017
[CrOS Tether] Eliminate token fetch from Initializer.
79da5a0 
This code was unnecessary, since the CryptAuth component is the one that
needs to listen for these tokens. Additionally, the listener in this
class was set up improperly, which made it possible to get "stuck"
without a token.

[email protected]

(cherry picked from commit a88ec4a)

Bug: 763506, 672263
Change-Id: I3e02f10aa201b129591028a8606b6fb4eefbaf11
Reviewed-on: https://chromium-review.googlesource.com/663978
Commit-Queue: Kyle Horimoto <[email protected]>
Reviewed-by: Ryan Hansberry <[email protected]>
Cr-Original-Commit-Position: refs/heads/master@{#501486}
Reviewed-on: https://chromium-review.googlesource.com/663982
Reviewed-by: Kyle Horimoto <[email protected]>
Cr-Commit-Position: refs/branch-heads/3202@{crosswalk-project#193}
Cr-Branched-From: fa6a5d8-refs/heads/master@{#499098}
Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.
Reviewers
No reviews
Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.