Closed
Bug 1202102
Opened 8 years ago
Closed 8 years ago
Android cppunit and xpcshell tests fail to szip libs when --localLib used
Categories
(Testing :: General, defect)
Testing
General
Tracking
(firefox43 fixed)
RESOLVED
FIXED
mozilla43
Tracking | Status | |
---|---|---|
firefox43 | --- | fixed |
People
(Reporter: gbrown, Assigned: gbrown)
References
Details
Attachments
(1 file, 1 obsolete file)
9.68 KB,
patch
|
dminor
:
review+
|
Details | Diff | Splinter Review |
Remote (Android) cppunit and xpcshell tests push libraries to device to support the test executables. Libraries are either extracted from the apk (when --localApk is specified) or found in the specified directory (when --localLib is specified). All is well when --localApk is used, but with --localLib, 'szip -d' is not invoked on the local libraries before pushing. As a result, tests fail with library load failures complaining of "bad elf magic".
![]() |
Assignee | |
Comment 1•8 years ago
|
||
This allows me to run cppunit-tests locally via make. gbrown@mozpad:~/objdirs/droid$ make -C ~/objdirs/droid cppunittests-remote make: Entering directory `/home/gbrown/objdirs/droid' Pushing libmozglue.so.. Pushing libplugin-container.so.. Pushing libplugin-container-pie.so.. Pushing libnssckbi.so.. Pushing libfreebl3.so.. Pushing libomxplugingb235.so.. Pushing libnss3.so.. Pushing libomxpluginkk.so.. Pushing libomxpluginhc.so.. Pushing libxul.so.. Pushing libomxplugingb.so.. Pushing liblgpllibs.so.. Pushing libomxplugin.so.. Pushing libsoftokn3.so.. SUITE-START | Running 1 tests TEST-START | TestFile PROCESS | TestFile | Running nsLocalFile tests... TEST-PASS | Setup TEST-PASS | AppendNative with invalid file name TEST-PASS | GetParent TEST-PASS | Create file TEST-PASS | Remove file TEST-PASS | Create directory TEST-PASS | MoveTo rename file TEST-PASS | CopyTo copy file TEST-PASS | MoveTo move file TEST-PASS | MoveTo move and rename file TEST-PASS | CopyTo copy file across directories TEST-PASS | Normalize with native paths TEST-PASS | Remove directory TEST-PASS | CreateUnique file TEST-PASS | CreateUnique directory TEST-PASS | OpenNSPRFileDesc DELETE_ON_CLOSE Finished running nsLocalFile tests. TEST-PASS | TestFile | took 946ms SUITE-END | took 0s Result summary: cppunittests INFO | Passed: 1 cppunittests INFO | Failed: 0 make: Leaving directory `/home/gbrown/objdirs/droid'
Attachment #8658334 -
Flags: review?(dminor)
Comment 2•8 years ago
|
||
Comment on attachment 8658334 [details] [diff] [review] apply 'szip -d' to local libs Review of attachment 8658334 [details] [diff] [review]: ----------------------------------------------------------------- r+ when comments are addressed. ::: testing/remotecppunittests.py @@ +93,3 @@ > remote_file = posixpath.join(self.remote_bin_dir, file) > + if szip: > + out = subprocess.check_output([szip, '-d', local_file], stderr=subprocess.STDOUT) Please catch CalledProcessError here and log the output if it occurs. @@ +105,3 @@ > remote_file = posixpath.join(self.remote_bin_dir, file) > + if szip: > + out = subprocess.check_output([szip, '-d', local_file], stderr=subprocess.STDOUT) Please catch CalledProcessError here and log the output if it occurs. ::: testing/xpcshell/remotexpcshelltests.py @@ +460,3 @@ > remoteFile = remoteJoin(self.remoteBinDir, file) > + if szip: > + out = subprocess.check_output([szip, '-d', localFile], stderr=subprocess.STDOUT) Please catch CalledProcessError here and log the output if it occurs. @@ +474,3 @@ > remoteFile = remoteJoin(self.remoteBinDir, file) > + if szip: > + out = subprocess.check_output([szip, '-d', localFile], stderr=subprocess.STDOUT) Please catch CalledProcessError here and log the output if it occurs.
Attachment #8658334 -
Flags: review?(dminor) → review+
![]() |
Assignee | |
Comment 3•8 years ago
|
||
https://treeherder.mozilla.org/#/jobs?repo=try&revision=2308d5c7ea3a
https://hg.mozilla.org/mozilla-central/rev/6422d6fcc6ae
Status: NEW → RESOLVED
Closed: 8 years ago
Resolution: --- → FIXED
Target Milestone: --- → mozilla43
Comment 6•7 years ago
|
||
Review commit: https://reviewboard.mozilla.org/r/68448/diff/#index_header See other reviews: https://reviewboard.mozilla.org/r/68448/
Attachment #8776795 -
Flags: review?(dminor)
![]() |
Assignee | |
Comment 7•7 years ago
|
||
If there's a change needed now, please open a new bug.
Updated•7 years ago
|
Attachment #8776795 -
Flags: review?(dminor)
Comment 8•7 years ago
|
||
Comment on attachment 8776795 [details] Bug 1202102 - s/CalledProcessError/subprocess.CalledProcessError/ https://reviewboard.mozilla.org/r/68448/#review65638 This looks good to me, but like Geoff said, should be landed as a separate bug.
Comment 9•7 years ago
|
||
https://reviewboard.mozilla.org/r/68448/#review65638 Is there any benefit of creating a new bug report over directly landing the patch now? This patch merely fixes a small mistake in the first patch for bug 1202102, so I thought that it made sense to link it to that bug (since reviewboard does not support patches without bugs yet).
![]() |
Assignee | |
Comment 10•7 years ago
|
||
My concern is that this bug and its original patch have a tidy history now, with tracking flag showing that the patch landed in firefox 43, the bug status showing that the patch landed, etc. If your patch gets backed out, or needs an uplift to aurora, or whatever, the history and bug status starts to get "interesting" and can cause confusion. I think you should have opened a new bug. That said, this is a small change and if you still want to proceed, I won't stop you.
Updated•7 years ago
|
Attachment #8776795 -
Attachment is obsolete: true
You need to log in
before you can comment on or make changes to this bug.
Description
•