Re: -var-update @

[Vladimir Prus <vladimir at codesourcery dot com>]

On Thursday 27 March 2008 16:24:40 Marc Khouzam wrote:
> 
> > Incidentally, it seems to be that a really smart frontend might be updating only
> > those variable objects that a visible on screen. To support this case efficiently,
> > we'd better support
> > 
> > 	-var-update var1 var2 var3 ...
> > 
> > syntax. I'm not proposing such a syntax right now -- we'd need to actually play
> > with such a smart frontend.
> 
> DSF only updates varObj that are visible on screen.
> So currently, it always uses -var-update with a single varObj name (never use *).
> 
> If I understand correctly,
> > 	-var-update var1 var2 var3 ...
> would allow the frontend to update multiple variable objects with a single command.
> With the goal of reducing the number of MI commands.  Any other benefits?

Someday, GDB might be able to batch requests to the targets. So, instead of N
roundtrips over slow JTAG connection, you can have single one.

> In the case of DSF, we wouldn't be able to use such a command though.
> The reason is that the views which show the variables are de-coupled from the
> variable object manager; and those views request the value of each variable
> individually, so the variable manager, which sends -var-update only gets
> a single varObj request at a time.

Heh, that's what you get for using nice component design :-)
Of course, batching requests in GDB might prove even more complicated.

> Not to say that
> > 	-var-update var1 var2 var3 ...
> would not be useful to other "really smart frontends" :-)
> 
> And, who knows, it may not be too hard for the DSF views to send batch requests
> containing all the visible variable objects, someday.

Yes.

- Volodya