780405 - "Assertion failure: strictModeState != StrictMode::UNKNOWN,"

Status: RESOLVED FIXED

(Core :: JavaScript Engine, defect)

Description

[Gary Kwong [:gkw] [:nth10sd] (NOT official MoCo now)]

Attachment: stack

function n(b = ((function() {}), function() {
    (a for (r in [function() {}]))
})) {}

asserts js debug shell on m-c changeset 0f66a7ecf8de without any CLI arguments at Assertion failure: strictModeState != StrictMode::UNKNOWN,

autoBisect shows this is probably related to the following changeset:

The first bad revision is:
changeset:   100567:418ad69cfe52
user:        Benjamin Peterson
date:        Mon Jul 09 23:43:06 2012 -0700
summary:     Bug 769072: retroactively apply strict mode to defaults. r=njn

Comment 1

[:Benjamin Peterson]

Attachment: set funbox->kids in LeaveFunction

Jason, I'm giving this to you since njn is gone.

Comment 2

[Jason Orendorff [:jorendorff]]

Comment on attachment 649041 [details] [diff] [review]
set funbox->kids in LeaveFunction

Review of attachment 649041 [details] [diff] [review]:
-----------------------------------------------------------------

Oh, I see. It's because generatorExpr() wasn't setting ->kids?

The comment on LeaveFunction() says it isn't called for functions created using the Function constructor. Test that too?

This is an improvement, so r=me, but feel free to revise it anyway and request review again.

Comment 3

[:Benjamin Peterson]

(In reply to Jason Orendorff [:jorendorff] from comment #2)
> Comment on attachment 649041 [details] [diff] [review]
> set funbox->kids in LeaveFunction
> 
> Review of attachment 649041 [details] [diff] [review]:
> -----------------------------------------------------------------
> 
> Oh, I see. It's because generatorExpr() wasn't setting ->kids?

Yes, so when we went to set strict mode to STRICT on it, its children were unreachable.

> 
> The comment on LeaveFunction() says it isn't called for functions created
> using the Function constructor. Test that too?

You can't have Function() lexically nested in defaults the way you can genexps.

> 
> This is an improvement, so r=me, but feel free to revise it anyway and
> request review again.

What's the revision you're looking for?

Comment 4

[Jason Orendorff [:jorendorff]]

(In reply to Benjamin Peterson from comment #3)
> > The comment on LeaveFunction() says it isn't called for functions created
> > using the Function constructor. Test that too?
> 
> You can't have Function() lexically nested in defaults the way you can
> genexps.

OK.

> What's the revision you're looking for?

None, carry on.

Comment 5

[:Benjamin Peterson]

https://hg.mozilla.org/integration/mozilla-inbound/rev/84739a192aa9

Comment 6

[Ed Morley [:emorley]]

https://hg.mozilla.org/mozilla-central/rev/84739a192aa9

Comment 7

[Christian Holler (:decoder)]

Automatically extracted testcase for this bug was committed:

https://hg.mozilla.org/mozilla-central/rev/efaf8960a929