Bug 166300 – Web Inspector: Address failures under LayoutTests/inspector/debugger/stepping

• Component: Web Inspector
• Status: RESOLVED FIXED

[Joseph Pecoraro]

Summary:
Address failures under LayoutTests/inspector/debugger/stepping

When running these tests with --force I've seen some consistent failures in a few places.

Steps to Reproduce:
1. $ run-webkit-tests LayoutTests/inspector/debugger/stepping --force --iterations=5
  => Failures

Notes:
- Looks like stepping past the end of a program is causing a pause in later executed code but that should have resumed
- Some numbers need to be updated.

[Joseph Pecoraro]

Created attachment 297578 [details]
[PATCH] Proposed Fix

[Blaze Burg]

Comment on attachment 297578 [details]
[PATCH] Proposed Fix

View in context: https://bugs.webkit.org/attachment.cgi?id=297578&action=review

r=me

> LayoutTests/ChangeLog:8
> +        * inspector/debugger/stepping/stepping-classes-expected.txt:

Should we unskip this test?

> LayoutTests/ChangeLog:9
> +        Rebasline these results to new column values.

Speling

> Source/JavaScriptCore/ChangeLog:10
> +        When continuing clear states that would have had us pause again.

continuing,

> Source/JavaScriptCore/ChangeLog:14
> +        When resuming after becoming idle be sure to clear Debugger state.

idle,

> Source/JavaScriptCore/inspector/agents/InspectorDebuggerAgent.cpp:793
> +        m_scriptDebugServer.continueProgram();
>          m_frontendDispatcher->resumed();

Hmm, in what conditions was this bug user-visible, stepping past the last step into an idle state?

[Joseph Pecoraro]

> > Source/JavaScriptCore/inspector/agents/InspectorDebuggerAgent.cpp:793
> > +        m_scriptDebugServer.continueProgram();
> >          m_frontendDispatcher->resumed();
> 
> Hmm, in what conditions was this bug user-visible, stepping past the last
> step into an idle state?

Correct. Though I wasn't always seeing this locally I certainly was in the tests.

[Joseph Pecoraro]

<https://trac.webkit.org/changeset/210251>