1530423 - Debugger does not pause at breakpoints except for the 1st window (round 2)

Status: RESOLVED FIXED

(DevTools :: Debugger, defect, P2)

Description

[Oriol Brufau [:Oriol]]

Bug 1528654 wasn't completely fixed:

1. Start with a clean profile
2. Press Ctrl+Shift+I to open devtools
3. Press F1 to open devtools preferences
4. Check "Enable browser chrome and add-on debugging toolboxes"
5. Check "Enable remote debugging"
6. Press Ctrl+Alt+Shift+I to open Browser Toolbox
7. In the Browser Toolbox, switch to the debugger panel
8. Press Ctrl+P to find file /content/browser.js, open it
9. Press Ctrl+F to search gBrowser.stop(), add a breakpoint in that line.
10. In the Firefox window you should have two tabs (first run and privacy notice)
11. Right-click one of them, choose "Move Tab" and "Move to New Window".
12. The debugger pauses as expected, resume it.
13. The tab will move into a new window.
14. Open a 2nd tab in this new window.
15. Right-click it and move it into a new window.

Expected: the debugger pauses again.
Actual: the debugger doesn't pause.

Comment 1

[Logan Smyth [:loganfsmyth]]

I tracked this down and see what's going on, but it actually looks like this is probably an existing issue that was not caused by https://bugzilla.mozilla.org/show_bug.cgi?id=1518661.

It's the same root cause as the other bug you filed. We're making an incorrect assumption about the relationship between scripts and sources, but it is in code that I didn't change as part of the breakpoint changes.

Comment 2

[Oriol Brufau [:Oriol]]

You are right, this was caused by bug 1513095 instead. I should have rechecked.
With mozregression I get https://hg.mozilla.org/integration/autoland/pushloghtml?fromchange=daf05866f9ddcd5a362b4d436477f3e01abd4234&tochange=957cc2dce89c9c93265bb24445792eb46fc24d4e

Comment 3

[Logan Smyth [:loganfsmyth]]

Interesting, I'll revisit when I get a chance. If that did introduce it, it's likely entirely by chance, since I don't think there are any explicit issues with that patch on its own, probably just the way it interacts with other existing code.

Comment 4

[Anthony Xie [:ravefalcon92]]

Attachment: Bug 1530423 - Ensure breakpoints are added to newly-cloned JSScripts for existing sources. r=loganfsmyth

Comment 5

[Pulsebot]

Pushed by [e-mail]:
https://hg.mozilla.org/integration/autoland/rev/173e4de60b4a
Ensure breakpoints are added to newly-cloned JSScripts for existing sources. r=loganfsmyth

Comment 6

[Andreea Pavel [:apavel]]

https://hg.mozilla.org/mozilla-central/rev/173e4de60b4a

Comment 7

[Oriol Brufau [:Oriol]]

Is this safe to uplift?

Comment 8

[Logan Smyth [:loganfsmyth]]

Attachment: bug1530423.patch

Beta/Release Uplift Approval Request

• Feature/Bug causing the regression: Bug 1513095
• User impact if declined: Users of the browser toolbox JS debugger may have certain breakpoints ignored entirely, making debugging more difficult.
• Is this code covered by automated tests?: Yes
• Has the fix been verified in Nightly?: Yes
• Needs manual test from QE?: No
• If yes, steps to reproduce:
• List of other uplifts needed: None
• Risk to taking this patch: Low
• Why is the change risky/not risky? (and alternatives if risky): This change is not risky because this just removes a special-case for previously-seen objects and makes things consistent across more cases.
• String changes made/needed:

Comment 9

[Logan Smyth [:loganfsmyth]]

This uplift patch also includes additional changes not in the original patch here, but they are only applicable changes for unit tests, so no concern there.

Comment 10

[Liz Henry (:lizzard) (relman/hg->git project)]

Comment on attachment 9049667 [details] [diff] [review]
bug1530423.patch

Fix for regression in 66, adds new tests, let's take it for the RC.

Comment 11

[Bogdan Tara[:bogdan_tara | bogdant]]

https://hg.mozilla.org/releases/mozilla-beta/rev/fb07e38d917f