Bug 107248 – [EFL] Enable indexed database

• Component: WebKit EFL
• Status: RESOLVED FIXED

[Jussi Kukkonen (jku)]

We should set ENABLE_INDEXED_DATABASE by default at some point.

[Jussi Kukkonen (jku)]

with ThirdParty leveldb (bug 106443) the results are:
27 failures or flakes out of 205 in storage/indexeddb/ and http/tests/inspector/indexeddb/:

storage/indexeddb/cursor-advance-workers.html [ Crash ]
storage/indexeddb/objectstore-basics-workers.html [ Failure Crash ]
storage/indexeddb/objectstore-basics.html [ Failure Crash ]
storage/indexeddb/pending-version-change-stuck-works-with-terminate.html [ Failure Crash ]
storage/indexeddb/transaction-abort.html [ Failure Crash ]
storage/indexeddb/database-quota.html [ Failure ]
storage/indexeddb/dont-commit-on-blocked.html [ Failure ]
storage/indexeddb/intversion-close-in-upgradeneeded.html [ Failure ]
storage/indexeddb/pending-activity-workers.html [ Failure ]
storage/indexeddb/pending-version-change-on-exit.html [ Failure ]
storage/indexeddb/pending-version-change-stuck.html [ Failure ]
storage/indexeddb/transaction-complete-workers.html [ Failure ]
storage/indexeddb/transaction-error.html [ Failure ]
storage/indexeddb/open-twice-workers.html [ Failure Pass ]
storage/indexeddb/index-multientry.html [ Failure Pass ]
storage/indexeddb/intversion-close-in-oncomplete.html [ Failure Pass ]
storage/indexeddb/index-unique.html [ Failure Pass ]
storage/indexeddb/lazy-index-population.html [ Failure Pass ]
storage/indexeddb/intversion-abort-in-initial-upgradeneeded.html [ Failure Pass ]
storage/indexeddb/key-generator.html [ Failure Pass ]
storage/indexeddb/factory-basics-workers.html [ Failure Pass ]
storage/indexeddb/index-basics-workers.html [ Failure Pass ]
storage/indexeddb/deletedatabase-delayed-by-open-and-versionchange-workers.html [ Failure Pass ]

http/tests/inspector/indexeddb/database-data.html [ Crash ]
http/tests/inspector/indexeddb/database-names.html [ Failure ]
http/tests/inspector/indexeddb/database-structure.html [ Failure ]
http/tests/inspector/indexeddb/resources-panel.html [ Failure ]

[Jussi Kukkonen (jku)]

Created attachment 188324 [details]
Patch

WIP patch for thiago

[Jussi Kukkonen (jku)]

(In reply to comment #2)
> Created an attachment (id=188324) [details]
> Patch
> 
> WIP patch for thiago

Ah, this is missing the result for:
 http/tests/inspector/indexeddb/database-data.html [ Crash ]

But otherwise these TestExpectations cover both Debug and Release over quite a few test runs.

[Riju]

Created attachment 193773 [details]
Enable IndexedDB in EFL

[Jussi Kukkonen (jku)]

Comment on attachment 193773 [details]
Enable IndexedDB in EFL

View in context: https://bugs.webkit.org/attachment.cgi?id=193773&action=review

By the way Thiago, do you still see different results for these IndexedDB tests?

> ChangeLog:1
> [phone]  Jussi Kukkonen  <[e-mail]>, Rijubrata Bhaumik  <[e-mail]>

Thanks for taking care of this: You can just drop my name, there's no reason to keep it (same for the other Changelogs).

> LayoutTests/platform/efl/TestExpectations:840
> +# Crashing IndexedDB tests, all with same backtrace
> +webkit.org/b/111171 storage/indexeddb/cursor-advance-workers.html [ Crash Pass ]
> +webkit.org/b/107194 storage/indexeddb/index-unique.html [ Failure Crash Pass ]
> +webkit.org/b/111171 storage/indexeddb/objectstore-basics-workers.html [ Failure Crash Pass ]
> +webkit.org/b/106732 storage/indexeddb/pending-activity-workers.html [ Failure Crash ]
> +webkit.org/b/107194 storage/indexeddb/pending-version-change-on-exit.html [ Failure Crash Pass ]
> +webkit.org/b/107377 storage/indexeddb/keypath-edges.html [ Crash Pass ]
> +webkit.org/b/107194 storage/indexeddb/pending-activity.html [ Crash Pass ]
> +webkit.org/b/112578 storage/indexeddb/transaction-coordination-across-databases.html [ Crash ]

The best thing would be to group these by issue (although I know it's a bit difficult when you have both a crash and failure), but at least make the comment clearer: I assume the crashes don't all have the same backtrace when they point to different bugs?

> LayoutTests/platform/efl/TestExpectations:854
> +Bug(EFL) storage/indexeddb/index-multientry.html [ Failure Pass ]
> +Bug(EFL) storage/indexeddb/intversion-abort-in-initial-upgradeneeded.html [ Failure Pass ]
> +Bug(EFL) storage/indexeddb/intversion-close-in-oncomplete.html [ Failure Pass ]
> +Bug(EFL) storage/indexeddb/intversion-close-in-upgradeneeded.html [ Failure Pass ]
> +Bug(EFL) storage/indexeddb/key-generator.html [ Failure Pass ]
> +Bug(EFL) storage/indexeddb/lazy-index-population.html [ Failure Pass ]
> +Bug(EFL) storage/indexeddb/objectstore-basics.html [ Failure Pass ]
> +Bug(EFL) storage/indexeddb/transaction-abort.html [ Failure Pass ]
> +Bug(EFL) storage/indexeddb/transaction-error.html [ Failure Pass ]

would be great if there were bug(s) filed for these as well.

[Riju]

Created attachment 194490 [details]
patch

[Riju]

Created attachment 194497 [details]
patch

updated ChangeLog

[Thiago Marcos P. Santos]

Comment on attachment 194497 [details]
patch

LGTM.

[Sudarsana Nagineni (babu)]

Comment on attachment 194497 [details]
patch

LGTM too.

[Laszlo Gombos]

Comment on attachment 194497 [details]
patch

r=me.

[WebKit Review Bot]

Comment on attachment 194497 [details]
patch

Rejecting attachment 194497 [details] from commit-queue.

Failed to run "['/mnt/git/webkit-commit-queue/Tools/Scripts/webkit-patch', '--status-host=webkit-commit-queue.appspot.com', '--bot-id=gce-cq-04', 'apply-attachment', '--no-update', '--non-interactive', 194497, '--port=chromium-xvfb']" exit_code: 2 cwd: /mnt/git/webkit-commit-queue

Last 500 characters of output:
tations
Hunk #2 FAILED at 1107.
Hunk #3 succeeded at 1819 (offset 2 lines).
1 out of 3 hunks FAILED -- saving rejects to file LayoutTests/platform/efl/TestExpectations.rej
patching file Source/cmake/OptionsEfl.cmake
patching file Tools/ChangeLog
Hunk #1 succeeded at 1 with fuzz 3.
patching file Tools/Scripts/webkitperl/FeatureList.pm

Failed to run "[u'/mnt/git/webkit-commit-queue/Tools/Scripts/svn-apply', '--force', '--reviewer', 'Laszlo Gombos']" exit_code: 1 cwd: /mnt/git/webkit-commit-queue

Full output: http://webkit-commit-queue.appspot.com/results/17246411

[Riju]

Created attachment 194864 [details]
patch

patch needed a rebase

[Riju]

Created attachment 195043 [details]
patch

[WebKit Review Bot]

Comment on attachment 195043 [details]
patch

Clearing flags on attachment: 195043

Committed r146864: <http://trac.webkit.org/changeset/146864>

[WebKit Review Bot]

All reviewed patches have been landed.  Closing bug.

[Gyuyoung Kim]

It looks EFL layout test bot has broken since this patch. Could you check it ?

http://build.webkit.org/builders/EFL%20Linux%2064-bit%20Release%20WK2/builds/6516

[WebKit Review Bot]

Re-opened since this is blocked by bug 113391

[Dominik Röttsches (drott)]

Riju, the stack traces look like the reason could be a mismatched alloc/free due to the different allocation backends used in DEBUG vs. RELEASE (fastMalloc -> alloc in DEBUG, vs. custom allocator in RELEASE).

[Riju]

Created attachment 196650 [details]
patch

Disabling GLOBAL_FAST_MALLOC_NEW by default webkit.org/b/114031 solves the mismatch caused earlier by different memory allocators specially in the Release build.

[Chris Dumez]

Comment on attachment 196650 [details]
patch

This patch does not disable GLOBAL_FAST_MALLOC_NEW as advertised?

[Riju]

(In reply to comment #20)
> (From update of attachment 196650 [details])
> This patch does not disable GLOBAL_FAST_MALLOC_NEW as advertised?

Chris, this patch -> webkit.org/b/114031 , by rakuco does it.

[Chris Dumez]

(In reply to comment #21)
> (In reply to comment #20)
> > (From update of attachment 196650 [details] [details])
> > This patch does not disable GLOBAL_FAST_MALLOC_NEW as advertised?
> 
> Chris, this patch -> webkit.org/b/114031 , by rakuco does it.

ok :)

[WebKit Commit Bot]

Comment on attachment 196650 [details]
patch

Clearing flags on attachment: 196650

Committed r147768: <http://trac.webkit.org/changeset/147768>

[WebKit Commit Bot]

All reviewed patches have been landed.  Closing bug.