Bug 122862 - Add ftlShouldBe() function to test FTL jitted code.
Summary: Add ftlShouldBe() function to test FTL jitted code.
Status: UNCONFIRMED
Alias: None
Product: WebKit
Classification: Unclassified
Component: New Bugs (show other bugs)
Version: 528+ (Nightly build)
Hardware: Unspecified Unspecified
: P2 Normal
Assignee: Nobody
URL:
Keywords:
Depends on:
Blocks:
 
Reported: 2013-10-15 14:05 PDT by Nadav Rotem
Modified: 2016-05-24 22:03 PDT (History)
5 users (show)

See Also:


Attachments
Patch (12.99 KB, patch)
2013-10-15 14:08 PDT, Nadav Rotem
beidson: review-
buildbot: commit-queue-
Details | Formatted Diff | Diff
Archive of layout-test-results from webkit-ews-15 for mac-mountainlion-wk2 (452.26 KB, application/zip)
2013-10-15 15:06 PDT, Build Bot
no flags Details
Archive of layout-test-results from webkit-ews-04 for mac-mountainlion (468.27 KB, application/zip)
2013-10-15 15:35 PDT, Build Bot
no flags Details
Archive of layout-test-results from webkit-ews-07 for mac-mountainlion (471.45 KB, application/zip)
2013-10-15 16:48 PDT, Build Bot
no flags Details

Note You need to log in before you can comment on or make changes to this bug.
Description Nadav Rotem 2013-10-15 14:05:32 PDT
Add ftlShouldBe() function to test FTL jitted code.
Comment 1 Nadav Rotem 2013-10-15 14:08:23 PDT
Created attachment 214298 [details]
Patch
Comment 2 Filip Pizlo 2013-10-15 14:13:00 PDT
Comment on attachment 214298 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=214298&action=review

> Source/JavaScriptCore/bytecode/CodeBlock.cpp:2911
> +    return JITCode::isOptimizingJITFTL(replacement()->jitType());

I would just say replacement()->jitType() == JITCode::FTLJIT;

> Source/JavaScriptCore/jit/JITCode.h:149
> -    
> +
> +    static bool isOptimizingJITFTL(JITType jitType)
> +    {
> +        return jitType == FTLJIT;
> +    }
> +

Don't need this.  It would only be useful if we planned to add a fifth tier or something. ;-)

> LayoutTests/resources/standalone-pre.js:265
> +function ftlCompiled(argument)
> +{
> +    if (!("f" in argument))
> +        throw new Error("ftlCompiled called with invalid argument.");
> +    return testRunner.compiledByFTL(argument.f);
> +}

What happens if this runs in run-webkit-tests?  Do you have logic to stub out compiledByFTL()?  I don't see the logic to wire compiledByFTL into the DumpRenderTree or WebKitTestRunner testRunner's.
Comment 3 Build Bot 2013-10-15 15:06:44 PDT
Comment on attachment 214298 [details]
Patch

Attachment 214298 [details] did not pass mac-wk2-ews (mac-wk2):
Output: http://webkit-queues.appspot.com/results/3482093

New failing tests:
js/ftl-not-string.html
Comment 4 Build Bot 2013-10-15 15:06:46 PDT
Created attachment 214306 [details]
Archive of layout-test-results from webkit-ews-15 for mac-mountainlion-wk2

The attached test failures were seen while running run-webkit-tests on the mac-wk2-ews.
Bot: webkit-ews-15  Port: mac-mountainlion-wk2  Platform: Mac OS X 10.8.5
Comment 5 Build Bot 2013-10-15 15:35:49 PDT
Comment on attachment 214298 [details]
Patch

Attachment 214298 [details] did not pass mac-ews (mac):
Output: http://webkit-queues.appspot.com/results/4104039

New failing tests:
js/ftl-not-string.html
Comment 6 Build Bot 2013-10-15 15:35:51 PDT
Created attachment 214309 [details]
Archive of layout-test-results from webkit-ews-04 for mac-mountainlion

The attached test failures were seen while running run-webkit-tests on the mac-ews.
Bot: webkit-ews-04  Port: mac-mountainlion  Platform: Mac OS X 10.8.5
Comment 7 Build Bot 2013-10-15 16:48:27 PDT
Comment on attachment 214298 [details]
Patch

Attachment 214298 [details] did not pass mac-ews (mac):
Output: http://webkit-queues.appspot.com/results/4108070

New failing tests:
js/ftl-not-string.html
Comment 8 Build Bot 2013-10-15 16:48:29 PDT
Created attachment 214320 [details]
Archive of layout-test-results from webkit-ews-07 for mac-mountainlion

The attached test failures were seen while running run-webkit-tests on the mac-ews.
Bot: webkit-ews-07  Port: mac-mountainlion  Platform: Mac OS X 10.8.5
Comment 9 Brady Eidson 2016-05-24 22:03:21 PDT
Comment on attachment 214298 [details]
Patch

Assuming that patches for review since 2013 are stale, r-