Bug 191012 – Add support for GetStack FlushedDouble

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

[Tadeu Zagallo]

LowerDFGToB3::compileGetStack assumed that we would not emit GetStack for doubles, but it turns out it may arise from the PutStack sinking phase: if we sink a PutStack into a successor block, other predecessors will emit a GetStack followed by a Upsilon.

[Tadeu Zagallo]

<rdar://problem/45265141>

[Tadeu Zagallo]

Created attachment 353255 [details]
Patch

[EWS Watchlist]

Comment on attachment 353255 [details]
Patch

Attachment 353255 [details] did not pass mac-wk2-ews (mac-wk2):
Output: https://webkit-queues.webkit.org/results/9758572

New failing tests:
http/wpt/beacon/cors/cors-redirect-failure.html

[EWS Watchlist]

Created attachment 353257 [details]
Archive of layout-test-results from ews106 for mac-sierra-wk2

The attached test failures were seen while running run-webkit-tests on the mac-wk2-ews.
Bot: ews106  Port: mac-sierra-wk2  Platform: Mac OS X 10.12.6

[Saam Barati]

Comment on attachment 353255 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=353255&action=review

r=me

> JSTests/stress/get-stack-double.js:1
> +function bar(n)

I think you need to run this with "--maximumFunctionForCallInlineCandidateInstructionCount=1000" based on the Radar.

To do this, I think you can add this comment to the top of this file:

// @runDefault("--maximumFunctionForCallInlineCandidateInstructionCount=1000")

[Tadeu Zagallo]

Comment on attachment 353255 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=353255&action=review

>> JSTests/stress/get-stack-double.js:1
>> +function bar(n)
> 
> I think you need to run this with "--maximumFunctionForCallInlineCandidateInstructionCount=1000" based on the Radar.
> 
> To do this, I think you can add this comment to the top of this file:
> 
> // @runDefault("--maximumFunctionForCallInlineCandidateInstructionCount=1000")

It does crash without the flag, should I add it anyway to mirror the radar?

[Saam Barati]

(In reply to Tadeu Zagallo from comment #6)
> Comment on attachment 353255 [details]
> Patch
> 
> View in context:
> https://bugs.webkit.org/attachment.cgi?id=353255&action=review
> 
> >> JSTests/stress/get-stack-double.js:1
> >> +function bar(n)
> > 
> > I think you need to run this with "--maximumFunctionForCallInlineCandidateInstructionCount=1000" based on the Radar.
> > 
> > To do this, I think you can add this comment to the top of this file:
> > 
> > // @runDefault("--maximumFunctionForCallInlineCandidateInstructionCount=1000")
> 
> It does crash without the flag, should I add it anyway to mirror the radar?

Nope. Ignore me. I thought it required the flag.

[Tadeu Zagallo]

Created attachment 353283 [details]
Patch for landing

[WebKit Commit Bot]

Comment on attachment 353283 [details]
Patch for landing

Rejecting attachment 353283 [details] from commit-queue.

New failing tests:
http/wpt/css/css-animations/start-animation-001.html
Full output: https://webkit-queues.webkit.org/results/9767152

[WebKit Commit Bot]

Created attachment 353285 [details]
Archive of layout-test-results from webkit-cq-01 for mac-sierra

The attached test failures were seen while running run-webkit-tests on the commit-queue.
Bot: webkit-cq-01  Port: mac-sierra  Platform: Mac OS X 10.12.6

[WebKit Commit Bot]

Comment on attachment 353283 [details]
Patch for landing

Clearing flags on attachment: 353283

Committed r237553: <https://trac.webkit.org/changeset/237553>

[WebKit Commit Bot]

All reviewed patches have been landed.  Closing bug.