Re: New MI maintainer

[Vladimir Prus <ghost@cs.msu.su>]

[-- Attachment #1: Type: text/plain, Size: 458 bytes --]

Bob Rossi wrote:

> On Tue, Feb 19, 2008 at 02:12:22PM -0500, Daniel Jacobowitz wrote:
>> I'm glad to announce that GDB finally has an official MI maintainer
>> again: Vladimir Prus.
> 
> Congratulations Vladmir.

Thanks.

> Could you tell us if you have any goals as the official MI maintainer?

Attached is my current list of MI problems that I find most important
to resolve. The item most close to completion is custom display of
varobjs.

- Volodya





[-- Attachment #2: MI Development(2).txt --]
[-- Type: text/plain, Size: 2283 bytes --]

 

MI Development

The GDB/MI interface as it is today has the following important problems.

Variable objects and RTTI
For C++, variable objects are not able to look at the real type of the object.
Only the static type is shown. We should be able to implement display of
real type, using Apple's branch as reference.


CLI commands bypass MI
When a CLI command is issued, we don't see "^running". As result, frontend can easily think 
that gdb is waiting for commands while inferior is running.

Variable objects and scopes

Variable objects don't care much about C++ scopes. For example, it's not possible to
create a variable object for a given expression in particular scope, which makes it impossible
to accurately implement variable tooltips. Also, it's not possible to list all local variables in the
entire function, which requires extraordinary effort to display all local variables as the enter
scope and leave scope.


Change notification style
Generally, all commands that a frontend might want to issue at each step -- list of breakpoints,
list of threads, list frames, list of local variables -- should have a notification to match, which is
emitted whenever gdb thinks the result of those command can possibly change. Wit this,
the frontend won't have to automatically issue any commands after step -- if anything changes,
gdb will report that itself.


Static fields

The way -var-list-children displays always displays static field is a bit annoying, it can also
result in nice never ending trees, like in https://bugs.eclipse.org/bugs/show_bug.cgi?id=136627 
Need to make it customizable.


Custom display
For many kinds of objects there's some more reasonable representation than pure varobj tree.
For example -- for QString we need the real string. For vector we need content of the vector
to be the children.  This requires some scripting level, that can customize creation of children
of variable object and display as string.  In order for -var-update to work, that customization
level should actually be able to construct arbitrary value and use that value for a variable object.
For -var-update together with vector, we should be able to check if the vector size/storage
has changed. I'm not sure how we'd create new children if a vector is resized.