Closed Bug 818838 Opened 12 years ago Closed 10 years ago

154 jit-test failures when compiling SpiderMonkey from source

Categories

(Core :: JavaScript Engine, defect)

x86_64
Linux
defect
Not set
normal

Tracking

()

RESOLVED INVALID

People

(Reporter: bruant.d, Unassigned)

Details

Attachments

(1 file)

Here is a report: http://pastebin.mozilla.org/1979291 I'm on Linux (Ubuntu 12.10), 64-bit. I don't know what to make of these failures, but I'd be happy to help investigate.
I've never seen any of these tests failure, is your working directory dirty? Can you detail you compilation process and dump your environment that you used during the compilation?
And I forgot to ask, are you on top of inbound or mozilla-central? inbound is never guarantee to be working, while m-c should.
pastes to pbmo can go away, so I'm attaching this to the bug.
also, debugging tips: pass -f into jit-tests, it will give you a full command line that you can just paste into your terminal. Once you have that, I'd recommend that you try running a couple of those and see how they are failing.
(In reply to Nicolas B. Pierron [:pierron] [:nbp] from comment #1) > I've never seen any of these tests failure, is your working directory dirty? No. I followed https://developer.mozilla.org/en-US/docs/SpiderMonkey/Build_Documentation and didn't do anything more. > Can you detail you compilation process and dump your environment that you > used during the compilation? I'm not too familiar with compilation processes. How do I "dump my environment"? I've been thinking about it and maybe it's a compiler issue? I'm using the one that comes with "sudo apt-get install build-essential" which I think is G++ (which has the version "g++ (Ubuntu/Linaro 4.7.2-2ubuntu1) 4.7.2") (In reply to Nicolas B. Pierron [:pierron] [:nbp] from comment #2) > And I forgot to ask, are you on top of inbound or mozilla-central? inbound > is never guarantee to be working, while m-c should. mozilla-central. (In reply to Marty Rosenberg [:mjrosenb] from comment #3) > pastes to pbmo can go away, so I'm attaching this to the bug. Ok. Thanks :-) (In reply to Marty Rosenberg [:mjrosenb] from comment #4) > also, debugging tips: pass -f into jit-tests, it will give you a full > command line that you can just paste into your terminal. Once you have > that, I'd recommend that you try running a couple of those and see how they > are failing. will do and come back as soon as I've done that.
(In reply to Marty Rosenberg [:mjrosenb] from comment #4) > also, debugging tips: pass -f into jit-tests, it will give you a full > command line that you can just paste into your terminal. Once you have > that, I'd recommend that you try running a couple of those and see how they > are failing. So I'm trying to copy/paste the thing but get the same error each time (I haven't tried all of them, but assume it'll be the same) $> js -e "const platform='linux2'; const libdir='/home/david/hgRepo/spidermonkey/js/src/jit-test/lib/'; const scriptdir='/home/david/hgRepo/spidermonkey/js/src/jit-test/tests/basic/'" -f /home/david/hgRepo/spidermonkey/js/src/jit-test/lib/prolog.js -f /home/david/hgRepo/spidermonkey/js/src/jit-test/tests/basic/bug596351-1.js Error: unrecognized flag -f Try --help for options I see that the -f option is passed twice every time. Is it a normal thing?
Assignee: general → nobody
I am guessing this is no longer the case, whatever the specific problem here was. Reopen if not.
Status: NEW → RESOLVED
Closed: 10 years ago
Resolution: --- → INVALID
You need to log in before you can comment on or make changes to this bug.

Attachment

General

Created:
Updated:
Size: