[
Date Prev][
Date Next][
Thread Prev][
Thread Next][
Date Index][
Thread Index]
[
List Home]
|
Re: [cdt-patch] Register View Persistence
|
At 04:40 PM 4/4/2003 -0500, Mikhail Khodjaiants wrote:
What if the input is from different target?
I actually tested this case. We sell configurable processors, it's pretty
common to see different register sets on different processors.
This actually means that if you
expand you register tree once you will get it expanded for all targets.
Yes, that's what it does. You will see If you have opened a register set on
one target and switch to a target that has a register set of the same name
then it will be open on the new target. Better or worse than always closing
them? Our folks think better.
Eclipse should (I think) offer debug UI configuration per launch
configuration. If I set up my register view a certain way, or if I have
typed cerrtain values into a memory view then they should be that way when
I launch that config again. That's the "right thing" but involves quite a
bit more work.
And
this preserves only the first level structure of the register tree.
As I read RegisterViewContentProvider, the register view is limited to 2
levels with the final level being nodes with no children.
It would be much easier to add a simple check button in the toolbar.
I don't understand. What does that do and how does it work?
Thanks!
-Chris
Mikhail
----- Original Message -----
From: "Chris Songer" <songer@xxxxxxxxxxxxx>
To: <cdt-patch@xxxxxxxxxxx>
Sent: Friday, April 04, 2003 1:20 PM
Subject: [cdt-patch] Register View Persistence
>
> Hi,
>
> Step through a function return and see the open groups in your register
> view close. That seems to be because the IStackFrame is different, even
> though the underlying register model is the same. (Still, the register
> model could change I suppose.) Our usability testers hated it.
>
> So, please find attached a patch that causes the register view to remember
> what register groups are expanded and restore them to their expanded state
> across stackframe changes. The patch was based on 1.0.1. It looks to me
> like it will cleanly apply to the head, but your mileage may vary.
>
> Thanks!
> -Chris
_______________________________________________
cdt-patch mailing list
cdt-patch@xxxxxxxxxxx
http://dev.eclipse.org/mailman/listinfo/cdt-patch
