Closed
Bug 486783
Opened 15 years ago
Closed 15 years ago
Run unit tests on pre-existing nightly builds
Categories
(Release Engineering :: General, defect, P2)
Release Engineering
General
Tracking
(Not tracked)
RESOLVED
FIXED
People
(Reporter: joduinn, Assigned: catlee)
References
Details
Attachments
(4 files)
3.43 KB,
patch
|
bhearsum
:
review+
lsblakk
:
review+
catlee
:
checked-in+
|
Details | Diff | Splinter Review |
11.43 KB,
patch
|
bhearsum
:
review+
catlee
:
checked-in+
|
Details | Diff | Splinter Review |
1.49 KB,
patch
|
nthomas
:
checked-in+
|
Details | Diff | Splinter Review |
1.98 KB,
patch
|
bhearsum
:
review+
catlee
:
checked-in+
|
Details | Diff | Splinter Review |
Once we are able to run unittests on incremental builds, we should have the automation run unittests on nightly and release builds also. Its separate work, and blocked, so filing separate bug in Future to track this.
Reporter | ||
Comment 1•15 years ago
|
||
Tweak summary to distinguish from bug#457753 and bug#372581.
Summary: run unittests on nightly and release builds → Run unit tests on pre-existing nightly builds
Comment 2•15 years ago
|
||
Ted said this will enable unit-testing of PGO builds. Is this the case? We want to enable PGO for a few places where it's disabled (bug 419893), but I'm loathe to do that without knowing that it'll pass the tests.
Comment 3•15 years ago
|
||
bug 457753 is a Q3 goal, I assume this (being very similar) will be fixed at the same time? Should it be un-futured?
Reporter | ||
Comment 4•15 years ago
|
||
(In reply to comment #3) > bug 457753 is a Q3 goal, I assume this (being very similar) will be fixed at > the same time? Should it be un-futured? Yep, not a goal, but happening about the same time, as its similar work. Lukas said she would take this on last week, as is similar to the work she is doing for running unittests on nightly builds and on debug builds, but neither of us updated the bug with that info. Moved from Future and assigned to Lukas, to match reality.
Assignee: nobody → lsblakk
Component: Release Engineering: Future → Release Engineering
Comment 5•15 years ago
|
||
Thanks for the update John, and thanks Lukas for working on this. PGO'd unit testing is blocking the enabling of PGO for several areas of code, which could improve application performance significantly.
Assignee | ||
Updated•15 years ago
|
Assignee: lsblakk → catlee
Assignee | ||
Comment 6•15 years ago
|
||
Attachment #401012 -
Flags: review?(lsblakk)
Attachment #401012 -
Flags: review?(bhearsum)
Updated•15 years ago
|
Attachment #401012 -
Flags: review?(bhearsum) → review+
Updated•15 years ago
|
Attachment #401012 -
Flags: review?(lsblakk) → review+
Assignee | ||
Comment 7•15 years ago
|
||
Attachment #401020 -
Flags: review?(lsblakk)
Attachment #401020 -
Flags: review?(bhearsum)
Comment 8•15 years ago
|
||
Comment on attachment 401020 [details] [diff] [review] buildbot-configs for unittests on optimized builds After our chat on IRC, this is fine.
Attachment #401020 -
Flags: review?(bhearsum) → review+
Assignee | ||
Updated•15 years ago
|
Priority: -- → P2
Assignee | ||
Comment 9•15 years ago
|
||
Comment on attachment 401012 [details] [diff] [review] buildbotcustom changes for running unittests on nightly/depend (optimized) builds changeset: 418:47b1cd1c81dc
Attachment #401012 -
Flags: checked-in+
Assignee | ||
Comment 10•15 years ago
|
||
Comment on attachment 401020 [details] [diff] [review] buildbot-configs for unittests on optimized builds changeset: 1544:83db6bbf7abd
Attachment #401020 -
Flags: review?(lsblakk) → checked-in+
Assignee | ||
Comment 11•15 years ago
|
||
These look like they're running properly on m-c
Comment 12•15 years ago
|
||
Please drop --enable-tests from the mozconfigs. It's the default, so having it there is redundant.
Comment 13•15 years ago
|
||
Today's mac m-c nightly ran out of space packaging the tests. It was asking for 5G and got 5.66G; tests seem to be about 91M unpacked so increasing it to 6G should hopefully be enough. http://hg.mozilla.org/build/buildbot-configs/rev/3e61bd4a8509 p-m & p-m02 reconfig'd.
Attachment #403150 -
Flags: checked-in+
Comment 14•15 years ago
|
||
is this ready to be marked fixed? is this running on all branches w/ tinderboxes?
Comment 15•15 years ago
|
||
(In reply to comment #14) > is this ready to be marked fixed? is this running on all branches w/ > tinderboxes? (if it needs to. maybe it doesn't?)
Assignee | ||
Comment 16•15 years ago
|
||
It's currently only running on mozilla-central. I'd like to see it on the mozilla-1.9.2 branch, and project branches as well at some point. Not sure if it's worth doing on the 1.9.1 branch at this point.
Comment 17•15 years ago
|
||
OS X 10.5.2 mozilla-central test opt everythingelse went green after I landed bug 518641. Yay! OS X 10.5.2 mozilla-central test opt everythingelse is still orange for two reasons: 1) It's failing to run mochitest-a11y. Our nightly builds don't have accessibility enabled on mac, so we'll have to turn these tests off on opt builds currently. 2) One failing xpcshell test: bug 519727
Assignee | ||
Comment 18•15 years ago
|
||
Neither opt nor debug builds are compiled with --enable-accessibility on mac, so running the mochitest-a11y tests will always fail. This patch creates a new UnittestPackagedBuildFactory for mac opt and debug tests that makes sure that they don't try and run mochitest-a11y.
Attachment #404698 -
Flags: review?(bhearsum)
Updated•15 years ago
|
Attachment #404698 -
Flags: review?(bhearsum) → review+
Assignee | ||
Comment 19•15 years ago
|
||
Comment on attachment 404698 [details] [diff] [review] Disable mochitest-a11y on non-unittest mac builds changeset: 445:b6e8ad13700c
Attachment #404698 -
Flags: checked-in+
Assignee | ||
Comment 20•15 years ago
|
||
Looks good. Future work (e.g. running on m-1.9.2, bug 523388) will be tracked in other bugs.
Status: NEW → RESOLVED
Closed: 15 years ago
Resolution: --- → FIXED
Updated•11 years ago
|
Product: mozilla.org → Release Engineering
You need to log in
before you can comment on or make changes to this bug.
Description
•