Bug 173344 - Flaky Test: media/media-source/media-source-unnecessary-seek-seeked.html
Summary: Flaky Test: media/media-source/media-source-unnecessary-seek-seeked.html
Alias: None
Product: WebKit
Classification: Unclassified
Component: Media (show other bugs)
Version: WebKit Nightly Build
Hardware: Unspecified Unspecified
: P2 Normal
Assignee: WebKit Commit Bot
: 173331 (view as bug list)
Depends on:
Blocks: 50856
  Show dependency treegraph
Reported: 2017-06-13 17:18 PDT by WebKit Commit Bot
Modified: 2017-06-14 10:32 PDT (History)
4 users (show)

See Also:

Archive of layout-test-results from webkit-cq-01 (997.77 KB, application/zip)
2017-06-13 17:18 PDT, WebKit Commit Bot
no flags Details
Patch (4.06 KB, patch)
2017-06-13 22:51 PDT, Jer Noble
no flags Details | Formatted Diff | Diff

Note You need to log in before you can comment on or make changes to this bug.
Description WebKit Commit Bot 2017-06-13 17:18:49 PDT
This is an automatically generated bug from the commit-queue.
media/media-source/media-source-unnecessary-seek-seeked.html has been flaky on the commit-queue.

media/media-source/media-source-unnecessary-seek-seeked.html was authored by jer.noble@apple.com.

The commit-queue just saw media/media-source/media-source-unnecessary-seek-seeked.html flake (text diff) while processing attachment 312824 [details] on bug 173330.
Bot: webkit-cq-01  Port: <class 'webkitpy.common.config.ports.MacPort'>  Platform: Mac OS X 10.11.6

The bots will update this with information from each new failure.

If you believe this bug to be fixed or invalid, feel free to close.  The bots will re-open if the flake re-occurs.

If you would like to track this test fix with another bug, please close this bug as a duplicate.  The bots will follow the duplicate chain when making future comments.
Comment 1 WebKit Commit Bot 2017-06-13 17:18:51 PDT
Created attachment 312834 [details]
Archive of layout-test-results from webkit-cq-01
Comment 2 Jer Noble 2017-06-13 21:59:15 PDT
So the underlying cause of this failure is not new; but the recent change to delay firing the 'seeker' event until the frame at the seek destination is actually decoded makes this happen much more frequently.  It goes like this:

1. MSE client enqueues a bunch of data to be parsed and enqueued
2. In the background, the media engine begins decoding the first visible frame
3. MSE client calls seek(), which enqueues a task to do the seek
4. The first frame is decoded, which triggers a readyState change.
5. The readyState change sees that the seek flag is set, and per the spec, finishes seeking
6. The seek task finally fires.

Previous to this change, the ready state change in step 4 happens between step 2 and 3. After this change, it's much more likely that the ready state change in step 4 happens after the seek request in step 3.
Comment 3 Jer Noble 2017-06-13 22:51:12 PDT
Created attachment 312859 [details]
Comment 4 Matt Lewis 2017-06-14 09:51:13 PDT
*** Bug 173331 has been marked as a duplicate of this bug. ***
Comment 5 WebKit Commit Bot 2017-06-14 10:32:52 PDT
Comment on attachment 312859 [details]

Clearing flags on attachment: 312859

Committed r218268: <http://trac.webkit.org/changeset/218268>
Comment 6 WebKit Commit Bot 2017-06-14 10:32:53 PDT
All reviewed patches have been landed.  Closing bug.