Last Comment Bug 773703 - on B2G xpcshell always returns 1
: on B2G xpcshell always returns 1
Status: NEW
:
Product: Core
Classification: Components
Component: XPConnect (show other bugs)
: unspecified
: ARM Other
: -- normal (vote)
: ---
Assigned To: Nobody; OK to take it and work on it
:
: Andrew Overholt [:overholt]
Mentors:
Depends on:
Blocks:
  Show dependency treegraph
 
Reported: 2012-07-13 09:57 PDT by Mihnea Dobrescu-Balaur (:mihneadb)
Modified: 2012-07-13 15:43 PDT (History)
3 users (show)
See Also:
Crash Signature:
(edit)
QA Whiteboard:
Iteration: ---
Points: ---
Has Regression Range: ---
Has STR: ---


Attachments

Description Mihnea Dobrescu-Balaur (:mihneadb) 2012-07-13 09:57:20 PDT
Running the XPCShell tests on B2G, every test suite shows that "test failed; xpcshell exited with return code 1; see following log:" and the log that follows only contains "test-pass" and "test-info". So for some reason it returns 1 even when everything is OK.
Comment 1 Ted Mielczarek [:ted.mielczarek] 2012-07-13 10:19:17 PDT
If this is actually a bug in the xpcshell binary, then the right component is Core:XPConnect.
Comment 2 Jonathan Griffin (:jgriffin) 2012-07-13 10:25:07 PDT
We've verified that xpcshell always exits with a return code of 1 on B2G, even after successfully running tests.
Comment 3 Ted Mielczarek [:ted.mielczarek] 2012-07-13 10:29:15 PDT
I mostly meant that you should try running xpcshell itself without the test harness and see if the behavior reproduces. If so then it's certainly a bug in the xpcshell binary. If not, then it might just be a harness bug.
Comment 4 Geoff Brown [:gbrown] 2012-07-13 14:17:27 PDT
We see the same thing when running on Android devices: see bug 768491.
Comment 5 Mihnea Dobrescu-Balaur (:mihneadb) 2012-07-13 15:43:36 PDT
(In reply to Ted Mielczarek [:ted] from comment #3)
> I mostly meant that you should try running xpcshell itself without the test
> harness and see if the behavior reproduces. If so then it's certainly a bug
> in the xpcshell binary. If not, then it might just be a harness bug.

http://pastebin.mozilla.org/1702951

Weird return codes everytime. Also, it might be that the get_return_code logic (which basically runs echo $? and parses that) truncates it to only the first digit and get the 1 from thre?

Note You need to log in before you can comment on or make changes to this bug.