497998 - Firebug crashes Firefox with "Assertion failure: cx->fp, .../jsdbgapi.cpp:1246"

Status: VERIFIED FIXED

(Core :: JavaScript Engine, defect)

Description

[Curtis Bartley [:cbartley]]

Attachment: testcase

After Firebug stops at a breakpoint, clicking inside the Firebug UI causes the following assertion failure:

Assertion failure: cx->fp, at /Users/cbartley/Dev/mozilla-c/src/js/src/jsdbgapi.cpp:1246

This is related to bug 497119.

Comment 1

[Blake Kaplan (:mrbkap) (inactive)]

Attachment: Fix

Forgot to remove this when I removed the assumption that fp is active (as opposed to dormant).

Comment 2

[Blake Kaplan (:mrbkap) (inactive)]

http://hg.mozilla.org/tracemonkey/rev/9b0e68c238eb and http://hg.mozilla.org/mozilla-central/rev/ecbd68d83cfa

Comment 3

[Curtis Bartley [:cbartley]]

Verified fixed on trunk 29163:ecbd68d83cfa.

Comment 5

[Blake Kaplan (:mrbkap) (inactive)]

Comment on attachment 383020 [details] [diff] [review]
Fix

Since bug 496824 was checked in on the 1.9.1 branch, we need this there too.

Comment 7

[Mark Banner (:standard8)]

Comment on attachment 383020 [details] [diff] [review]
Fix

Moving request to 1.9.1.1. This fixes debugging using Venkman or Firebug something which at least Thunderbird will need for a while.

Comment 8

[Blake Kaplan (:mrbkap) (inactive)]

http://hg.mozilla.org/releases/mozilla-1.9.1/rev/f3031a3f5fc6

Comment 9

[Aakash Desai [:aakashd]]

I see a hang on OSX using the testcase attached to this bug for build:
Mozilla/5.0 (Macintosh; U; Intel Mac OS X 10.5; en-US; rv:1.9.1.1pre) Gecko/20090715 Shiretoko/3.5.1pre ID:20090715031842 with firebug 1.4.0b7



...but is verified FIXED on 1.9.1 nightly on WinXP: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1.1pre) Gecko/20090715 Shiretoko/3.5.1pre (.NET CLR 3.5.30729) ID:20090715044728 with firebug 1.4.0b7

Comment 10

[timeless]

aakash: can you file a new bug and get a stack trace from a shark enabled build?

Comment 11

[Aakash Desai [:aakashd]]

why would I file a new bug timeless?

Comment 12

[Curtis Bartley [:cbartley]]

Two notes:

This bug only ever appeared in debug builds, so it will be necessary to verify it on a debug build.

Right now it's blocked by bug 497028 which has not yet landed on branch (hopefully it will land in 1.9.1.2; it's too late for 1.9.1.1).

The easiest course of action is to wait for 497028's patch to land on branch before trying to verify this bug on branch.

Comment 13

[Mike Beltzner [:beltzner, not reading bugmail]]

Curtis or Blake: can you verify that this is fixed in latest-mozilla1.9.1 nightly or better yet the 3.5.1 release candidate: ftp://ftp.mozilla.org/pub/mozilla.org/firefox/nightly/3.5.1-candidates/build1/

Comment 14

[Curtis Bartley [:cbartley]]

(In reply to comment #13)
> Curtis or Blake: can you verify that this is fixed in latest-mozilla1.9.1
> nightly or better yet the 3.5.1 release candidate:
> ftp://ftp.mozilla.org/pub/mozilla.org/firefox/nightly/3.5.1-candidates/build1/

I don't think we *can* verify in a nightly or the release candidate.  My understanding is that this bug was an erroneous assertion that only triggered in debug builds.  So the bad news is that we can only verify it in a debug build.  The good news is that it was never in non-debug builds anyway. (Unless we have nightlies with assertions enabled that I don't know about.)

Then there is the whole issue with bug 497028 blocking the reproduction of this bug (on OS X) anyway.  The repro steps are essentially the same, and that bug hangs the browser.

Comment 15

[juan becerra [:juanb]]

Yesterday I created a debug build on OS X, on which I tried the steps to reproduce the problem. I did not see the assertion in the bug summary, and the application did not hang right away. I tried a few more times, and eventually the build hung.

Curtis, can you confirm that this is sufficient to verify the bug, and that the hang is an independent issue from this?

Comment 16

[Curtis Bartley [:cbartley]]

(In reply to comment #15)
> Yesterday I created a debug build on OS X, on which I tried the steps to
> reproduce the problem. I did not see the assertion in the bug summary, and the
> application did not hang right away. I tried a few more times, and eventually
> the build hung.
> 
> Curtis, can you confirm that this is sufficient to verify the bug, and that the
> hang is an independent issue from this?

I think that it does verify the bug, but I can't be 100% certain that it does.  It seems like we should be able to just look at the source code and verify that the assertion statement is no longer there...