Unload debugger when message handler is cleared

When message handler is set to NULL and there is no debugger listener the debugger is unloaded immediately unless it's entered, in which case it will be unloaded when last instance of EnterDebugger is destroyed.

Without the change the debugger may crash as Debugger::EventActive(v8::Break) called from OnDebugBreak may clear current debugger context. 

Also when compilation cache was enabled debugger could fail on second attach for the same reason(see AfterCompileMessageWhenMessageHandlerIsReset).

BUG=12404
Committed: http://code.google.com/p/v8/source/detail?r=2035

[apavlov, 2009-05-22 22:24:13]

Minor comment-related comments :) The solution seems reasonable but I'll leave
the logics check to Soeren.

http://codereview.chromium.org/115709/diff/1003/1005
File src/debug.h (right):

http://codereview.chromium.org/115709/diff/1003/1005#newcode658
Line 658: static bool debugger_unload_pending_;  // Was message handler cleared?
should the field comment be changed a bit?

http://codereview.chromium.org/115709/diff/1003/1004
File test/cctest/test-debug.cc (right):

http://codereview.chromium.org/115709/diff/1003/1004#newcode4918
Line 4918: // Tests that after compile event is sent as many times as there
scripts
as there scripts compiled -> as there are scripts compiled

http://codereview.chromium.org/115709/diff/1003/1004#newcode4934
Line 4934: // Setting listener to should cause debugger unload.
..to NULL..

http://codereview.chromium.org/115709/diff/1003/1004#newcode4958
Line 4958: // Call f while enabeling and disabling the script break point.
enabling

http://codereview.chromium.org/115709/diff/1003/1004#newcode4993
Line 4993: // Call f while enabeling and disabling the script break point.
ditto

[Søren Thygesen Gjesse, 2009-05-25 07:18:29]

LGTM

http://codereview.chromium.org/115709/diff/1003/1004
File test/cctest/test-debug.cc (right):

http://codereview.chromium.org/115709/diff/1003/1004#newcode4958
Line 4958: // Call f while enabeling and disabling the script break point.
I don't understand the comment what script break point is enabled/disabled?

http://codereview.chromium.org/115709/diff/1003/1004#newcode4993
Line 4993: // Call f while enabeling and disabling the script break point.
ditto

[yurys, 2009-05-25 07:50:18]

http://codereview.chromium.org/115709/diff/1003/1004
File test/cctest/test-debug.cc (right):

http://codereview.chromium.org/115709/diff/1003/1004#newcode4918
Line 4918: // Tests that after compile event is sent as many times as there
scripts
On 2009/05/22 22:24:13, apavlov wrote:
> as there scripts compiled -> as there are scripts compiled

Done.

http://codereview.chromium.org/115709/diff/1003/1004#newcode4934
Line 4934: // Setting listener to should cause debugger unload.
On 2009/05/22 22:24:13, apavlov wrote:
> ..to NULL..

Done.

http://codereview.chromium.org/115709/diff/1003/1004#newcode4958
Line 4958: // Call f while enabeling and disabling the script break point.
On 2009/05/25 07:18:29, Søren Gjesse wrote:
> I don't understand the comment what script break point is enabled/disabled?
Done. It's a result of copy/paste.

http://codereview.chromium.org/115709/diff/1003/1004#newcode4993
Line 4993: // Call f while enabeling and disabling the script break point.
On 2009/05/25 07:18:29, Søren Gjesse wrote:
> ditto

Done. Removed the confusing comment.

http://codereview.chromium.org/115709/diff/1003/1004#newcode4993
Line 4993: // Call f while enabeling and disabling the script break point.
On 2009/05/22 22:24:13, apavlov wrote:
> ditto

Done.