- 08 Jul, 2010 3 commits
- 02 Jul, 2010 1 commit
-
-
Friedemann Kleint authored
CDB: Add more information and state to Thread. Refactor thread information in CdbCore classes to prepare for more information.
-
- 22 Jun, 2010 2 commits
-
-
Friedemann Kleint authored
-
hjk authored
This replaces the (de facto) singleton engines and data handlers by classes that are instantiated per run. The DebuggerRunControl will now create an object of (a class derived from) DebuggerEngine that contains all the relevant "dynamic" data. DebuggerManager is no more. The "singleton" bits are merged into DebuggerPlugin, whereas the data bits went to DebuggerEngine. There is no formal notion of a "current" DebuggerEngine. However, as there's only one DebuggerEngine at a time that has its data models connected to the view, there's still some "de facto" notion of a "current" engine. Calling SomeModel::setData(int role, QVariant data) with custom role is used as the primary dispatch mechanism from the views to the "current" data models (and the engine, as all data models know their engine).
-
- 15 Jun, 2010 5 commits
-
-
Friedemann Kleint authored
-
hjk authored
-
hjk authored
-
hjk authored
-
Friedemann Kleint authored
...to fix multithread crash. Fix exit crash by temporarily removing output.
-
- 14 Jun, 2010 7 commits
-
-
Robert Loehning authored
Wrote down what hjk told me.
-
hjk authored
-
hjk authored
-
hjk authored
Pass output through the RunControl instead of the DebuggerManager.
-
hjk authored
Plan is to identify make identify an IDebuggerEngine incarnation with a RunControl.
-
Friedemann Kleint authored
-
hjk authored
This replaces most uses of DebuggerStartParameters by DebuggerRunControl which is a simple RunControl with a DebuggerStartParameters member. Plan is to move all global state to the run controls, and possibly introduce specialized ones for core debugging etc.
-
- 20 May, 2010 1 commit
-
-
Friedemann Kleint authored
And use it from CDB's module load/unload callbacks instead of re-setting all modules. Some cleanup./Fixed conflicts in 2.0 Acked-by: hjk
-
- 19 May, 2010 2 commits
-
-
Friedemann Kleint authored
And use it from CDB's module load/unload callbacks instead of re-setting all modules. Some cleanup. Acked-by: hjk
-
Friedemann Kleint authored
-
- 14 May, 2010 1 commit
-
-
Friedemann Kleint authored
Prompt to set up the public symbol server unless it is already configured or the environment variable exists. Change the dialog to be based on the PathChooser to be able to suggest a non-existent directory. Acked-by:
Alessandro Portale <alessandro.portale@nokia.com>
-
- 12 May, 2010 2 commits
-
-
Friedemann Kleint authored
-
Robert Loehning authored
-
- 10 May, 2010 2 commits
-
-
Friedemann Kleint authored
along with some output formatting to find out defaults.
-
Friedemann Kleint authored
-
- 05 May, 2010 1 commit
-
-
ck authored
Reviewed-by: hjk
-
- 04 May, 2010 2 commits
-
-
Robert Loehning authored
-
hjk authored
-
- 27 Apr, 2010 1 commit
-
-
hjk authored
-
- 20 Apr, 2010 1 commit
-
-
Friedemann Kleint authored
-
- 19 Apr, 2010 1 commit
-
-
Erik Verbruggen authored
So now the "Applciation Output" can distinguish between these four, and handle them appropriately.
-
- 15 Apr, 2010 2 commits
-
-
Friedemann Kleint authored
Including reason, etc, and show in status bar. Switch to assembler in activateFrame() in case someone clicks on an unusable frame.
-
Friedemann Kleint authored
Task-number: QTCREATORBUG-1139
-
- 09 Apr, 2010 1 commit
-
-
Friedemann Kleint authored
- Do not reformat boolean values (true,false) as hex/oct. - char/unsigned char types (cdb/gdb): Chop off the quoted printable character output by the debugger, pass around integer only and append the character ourselves in the watchhandler, making reformatting as hex/octal work. - cdb: Check format correctly (fix values output as plain integer). - cdb: Kill needless warning produced by reformatting output of out-of-scope variables. - cdb: Display exceptions in the status bar as non-intrusive crash notification. Reviewed-by: hjk
-
- 18 Mar, 2010 1 commit
-
-
Friedemann Kleint authored
-
- 11 Mar, 2010 1 commit
-
-
Friedemann Kleint authored
The timer for event-delivery time-out checking when attaching to a crashed process in the case of not being spawned by the handler (such as via single-application remote command) triggered too early. Add a 'moduleLoaded()' signal to CoreEngine and trigger from there (due to lack of a proper "engine up/attached" notification. Emit moduleLoaded() from the first timed-out debug event handler when modules no longer change. Reviewed-by:
Robert Loehning <robert.loehning@nokia.com>
-
- 10 Mar, 2010 1 commit
-
-
hjk authored
-
- 05 Mar, 2010 1 commit
-
-
hjk authored
-
- 03 Mar, 2010 1 commit
-
-
Friedemann Kleint authored
Set mode variable correctly, suppress unwanted debug output. Reviewed-by:
hjk <qtc-committer@nokia.com>
-