The Decode call may result in synchronously creating the surface, but we only check again if the surface is there for FLAG_SYNC_DECODE, not FLAG_SYNC_DECODE_IF_FAST.
All of the decoding we do during painting is of the type FLAG_SYNC_DECODE_IF_FAST, which means it would be useless to do that decoding synchronously during painting because the paint doesn't benefit from the result of that decoding.
Looking at the history of this code it looks like https://hg.mozilla.org/mozilla-central/rev/435df926eb10 (part 6 of bug 1119774) was where this bug was introduced. Before that changeset we always did another LookupFrameInternal call after the Decode (called WantDecodedFrames back then). But that changeset changed it to only be done for standard sync decodes, not "sync decode if fast".
In bug 1308472 we are seeing 'make -k check' fail intermittently with
the only apparent error message something like:
make: *** [check] Error 245
This debug should make it more clear which test (if any) is responsible
for setting the return code in 'mach python-test', and whether or not
'mach python-test' is actually reaching the end of the function.
MozReview-Commit-ID: 6IQrZQqs8ij
--HG--
extra : rebase_source : 4d2d4b8b139d7f16bda2b22ce79ab2c84e8fe4c7
Leak checking in mochitest was failing to turn the jobs orange for a period of time. Several leaks slipped
through in the interim. In order to land the fix ASAP to avoid further bustage, these leaks are being
whitelisted. See the patch or the dependency graph of bug 1325148 for the list of leaks/bugs that are being
disabled.
While these bugs won't show up in treeherder, they are a priority to fix, as they could represent leaks that
are already landed in the code.
MozReview-Commit-ID: JY7Gpvswp4r
--HG--
extra : rebase_source : 0d75d80eccddca31a7d2d1c997a85efdb5a4858c
This is being disabled because of the leak in bug 1325215. It slipped in because leak
checking in reftest was temporarily failing to turn the job orange. We are disabling
leak checking so we can land the fix ASAP and prevent further bustage from slipping
through in the mean time. Once bug 1325215 has been fixed, this should be backed out.
MozReview-Commit-ID: 74VGhHvZtIG
--HG--
extra : rebase_source : c965aa1e229a15948d2d4b55dc16c5904c304ad9
This fixes a regression when switching to the StructuredOutputParser in mozharness.
Previously, mozleak was relying on the string "TEST-UNEXPECTED-FAIL" to turn the jobs
orange, but it was doing so at the "warning" level. The StructuredOutputParser requires
the "error" level to set the appropriate tbpl_status.
MozReview-Commit-ID: 9u9mwqrkA6E
--HG--
extra : rebase_source : d31ece08232a87713702e713076071fa10fb8324
On windows, python generally returns windows-style paths,
with drive letters and backslash for a separator. However,
when we offer advice for updating the PATH variable, we're
talking about the msys environment which uses unix-style
paths. Convert to avoid confusion.
This is intended to turn c:\Users\mozilla\ into /c/Users/mozilla/.
MozReview-Commit-ID: FdB8FvjeCV1
--HG--
extra : rebase_source : 6d9e87b460417254bbe2eb5af3813e22f2126fb1
We want to call Request{Audio,Video}Data() instead of Ensure{Audio,Video}DecodeTaskQueued
which checks mState and breaks the encapsulation of the state objects.
MozReview-Commit-ID: 5oydItSvnMF
--HG--
extra : rebase_source : 5da4ad2f0fbf010a78c6b4e7cbab70378a002758
extra : intermediate-source : 0a882fb53fd51a71f9a3914074e9020d4eb87c4f
extra : source : 9927e37c9383e9204cbaf1b0dbc6fbfdff48df82
We want to call Request{Audio,Video}Data() instead of Ensure{Audio,Video}DecodeTaskQueued
which checks mState and breaks the encapsulation of the state objects.
MozReview-Commit-ID: 87CwSdtTwi4
--HG--
extra : rebase_source : a16404e208ae5cd9b008728249cea444a7b229e7
extra : intermediate-source : 3fbfa4532f6a54ea9e1aa7b0e9880539ca57c811
extra : source : 22a75175dd957605e70096b029a677b4d2ac2d42
Commenting out test methods is not the way how we should mark tests
as being skipped. The correct skip methods have to be used instead so
that the final results also show the correct skip count.
MozReview-Commit-ID: LKL4YQCyFko
--HG--
extra : rebase_source : 4c50596a6477e2afa0926b5dd787466d2b6ce89a
Marionette's skip decorators are currently not conform with the ones from the
Python's unittest module, which require a reason as parameter. As such Marionette
should behave the same and should also require a reason for more detailed skip
messages. This is done by wrapping the actual decorator with another enclosing
method.
With the changes we also ensure that the wrapper has the same attributes as the
wrapped function by using functools.wraps(). This hasn't used so far and makes
debugging harder.
Further a couple of skip methods and classes were copied from the unittest module,
which should be better imported instead to reduce code duplication.
MozReview-Commit-ID: 6XT6M6cbCFW
--HG--
extra : rebase_source : 2fb8bce0f17eade182eb5f61479548d0f6ef8ecc
Taking screenshots for skipped tests is useless and should be avoided to
reduce the size of the test logs.
MozReview-Commit-ID: 9HPH7pSXTj9
--HG--
extra : rebase_source : 41ea609a285e037160b1be535dac35de643f31c1
Registered callback handlers for tests should receive the correct
test status when the test has been finished, and not always "Error".
This change allows those callbacks to run specific code for individual
test results, eg. only do screenshots for failures.
MozReview-Commit-ID: FfbCRR0Jvjb
--HG--
extra : rebase_source : c73253acbb666ca62b23f806145c20f0a70c934c
Masatoshi Kimura
Grisha Kruglov
Mark Banner
Sebastian Hengst
Tobias Schneider
Timothy Nikkel
Boris Zbarsky
Mike Shal
Andrew Halberstadt
Ralph Giles
Al Tsai
JW Wang
Cameron McCormack
Henrik Skupin
aceman
Jonathan Watt
Karl Tomlinson
Phil Ringnalda
cku
Ehsan Akhgari