669043 - "Assertion failure: script->code <= target && target < script->code + script->length," with trap

Status: RESOLVED FIXED

(Core :: JavaScript Engine, defect)

Description

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

function f() {
    print(
        Proxy.create((
            function() {
                return {
                    get: Namespace,
                }
            })
        ())
    )
}
dis(f)
trap(f, 0, '')
f()

asserts js debug shell on TM changeset f59568ec0513 with -d at Assertion failure: script->code <= target && target < script->code + script->length,

js> dis(f)
flags: NULL_CLOSURE
loc     op
-----   --
main:
00000:  callgname "print"     <-- trap goes here
00003:  getgname "Proxy"
00006:  callprop "create"
00009:  lambda (function () {return {get: Namespace};})
00012:  nullblockchain
00013:  push
00014:  call 0
00017:  call 1
00020:  call 1
00023:  pop
00024:  stop

Source notes:
 ofs  line    pc  delta desc     args
---- ---- ----- ------ -------- ------
  0:    1     0 [   0] newline 
  1:    2     3 [   3] newline 
  2:    3     6 [   3] pcbase   offset 3
  4:    3     9 [   3] newline 
  5:    4    14 [   5] pcbase   offset 5
  7:    4    17 [   3] pcbase   offset 14
  9:    4    20 [   3] pcbase   offset 20
 11:    4    24 [   4] setline  lineno 10

js> trap(f, 0, '')

autoBisect shows this is probably related to the following changeset:

The first bad revision is:
changeset:   70404:bb9e5496b0ac
user:        Luke Wagner
date:        Fri May 13 08:56:26 2011 -0700
summary:     Bug 656462, part 4 - Simplify stack code, keep track of native calls, create new iterator over native/scripted callstack, make JS_SaveFrameChain fallible (r=waldo,mrbkap)

Comment 1

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

Attachment: stack

stack is hideously long...

Comment 2

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

Still occurs on TM changeset d8e967b8afc8.

Comment 3

[Luke Wagner [:luke]]

Attachment: fix

More DEBUG-only ridiculousness resulting from new strong assertions plus AutoScriptUntrapper.

Comment 4

[Jeff Walden [:Waldo]]

Comment on attachment 544072 [details] [diff] [review]
fix

Review of attachment 544072 [details] [diff] [review]:
-----------------------------------------------------------------

I'm not sure this is the perfect, absolute most cleanest way to do this.  But it works, so meh.

Comment 5

[Luke Wagner [:luke]]

http://hg.mozilla.org/tracemonkey/rev/60b1a6a58531

Comment 6

[Joe Drew (not getting mail)]

This (along with most things committed on Friday afternoon) was backed out of mozilla-inbound in order to clear up orange.

Comment 7

[Luke Wagner [:luke]]

http://hg.mozilla.org/integration/mozilla-inbound/rev/16b9dfded119

Comment 8

[Marco Bonardo [:mak]]

http://hg.mozilla.org/mozilla-central/rev/16b9dfded119