Bug 185043 – Expose "$262.agent.monotonicNow()" for use in testing Atomic operation timeouts

• Status: RESOLVED
• Resolution: FIXED
• Component: JavaScriptCore

[Rick Waldron]

During a discussion (re: Atomics testing in Test262) with Filip Pizlo this afternoon, we determined that $262.agent should expose a new function that returns a monotonically increasing time value (to avoid relying on Date.now()). I believe this can be implemented by exposing a wrapper for MonotonicTime::now()

(Patch to follow)

[Rick Waldron]

Created attachment 338943 [details]
$262.agent.monotonicNow()

[Filip Pizlo]

Comment on attachment 338943 [details]
$262.agent.monotonicNow()

Looks good. All it needs is a change log and I can r+ it.

[Rick Waldron]

Created attachment 338988 [details]
$262.agent.monotonicNow()

[Rick Waldron]

Created attachment 338989 [details]
$262.agent.monotonicNow()

[Filip Pizlo]

Comment on attachment 338989 [details]
$262.agent.monotonicNow()

R=me but the changelog looks pretty goofy. Why does it claim that you changed all of those functions?

[Rick Waldron]

I'm not sure, but I wonder if the changelog program sees all of the uncommitted whitespace changes that exist in the file. I'd like to "discard changes" for those whitespace bits and regenerate the changelog entry.

[Rick Waldron]

Created attachment 338991 [details]
$262.agent.monotonicNow()

Filip, sure enough—the whitespace changes that existed in jsc.cpp, but not committed, were treated as meaningful changes to report. I'll be more mindful of that in the future.

[Rick Waldron]

Created attachment 339108 [details]
$262.agent.monotonicNow()

[EWS Watchlist]

Comment on attachment 339108 [details]
$262.agent.monotonicNow()

Attachment 339108 [details] did not pass win-ews (win):
Output: http://webkit-queues.webkit.org/results/7507664

New failing tests:
http/tests/security/contentSecurityPolicy/userAgentShadowDOM/allow-audio.html

[EWS Watchlist]

Created attachment 339112 [details]
Archive of layout-test-results from ews200 for win-future

The attached test failures were seen while running run-webkit-tests on the win-ews.
Bot: ews200  Port: win-future  Platform: CYGWIN_NT-6.1-2.9.0-0.318-5-3-x86_64-64bit

[Filip Pizlo]

(In reply to rwaldron from comment #8)
> Created attachment 339108 [details]
> $262.agent.monotonicNow()

Did you mean to commit this version?  Looks like ews is having issues on Windows but it looks like an unrelated flaky failure. If you’d like, I can call+ it again or land manually.

[Rick Waldron]

(In reply to Filip Pizlo from comment #11)
> (In reply to rwaldron from comment #8)
> > Created attachment 339108 [details]
> > $262.agent.monotonicNow()
> 
> Did you mean to commit this version?  

Yes, this new patch was intentional. Because I had to do interactive per-line staging to avoid committing all of the whitespace changes, I accidentally missed this line: 

  static EncodedJSValue JSC_HOST_CALL functionDollarAgentMonotonicNow(ExecState*);

So I had to replace the existing patch with a revised patch. 



> Looks like ews is having issues on
> Windows but it looks like an unrelated flaky failure. 


I have no idea why that windows test is failing, that's definitely odd. 

> If you’d like, I can call+ it again or land manually.


Whichever is easiest for you?

[Filip Pizlo]

Comment on attachment 339108 [details]
$262.agent.monotonicNow()

Let’s try that again.

[WebKit Commit Bot]

Comment on attachment 339108 [details]
$262.agent.monotonicNow()

Clearing flags on attachment: 339108

Committed r231238: <https://trac.webkit.org/changeset/231238>

[WebKit Commit Bot]

All reviewed patches have been landed.  Closing bug.

[Radar WebKit Bug Importer]

<rdar://problem/39908404>