public inbox for insight@sourceware.org
 help / color / mirror / Atom feed
* stack window
@ 2000-04-10 12:13 Tom Tromey
  2000-04-10 12:55 ` James Ingham
  0 siblings, 1 reply; 4+ messages in thread
From: Tom Tromey @ 2000-04-10 12:13 UTC (permalink / raw)
  To: Insight List

Today I tried to use the stack window.

First, it doesn't display nearly enough information.

Second, clicking on a line does not do anything useful.  In fact, it
hoses my session.  The cursor changes to an hourglass over the source
window, and the source window just goes blank.  Nothing I do changes
this.

Tom

^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: stack window
  2000-04-10 12:13 stack window Tom Tromey
@ 2000-04-10 12:55 ` James Ingham
  2000-04-10 13:04   ` Tom Tromey
  0 siblings, 1 reply; 4+ messages in thread
From: James Ingham @ 2000-04-10 12:55 UTC (permalink / raw)
  To: tromey; +Cc: Insight List

Tom,

I need more details.  In ordinary C code, the stack window works
fine...  I bet you have hit some error that didn't get caught.  Maybe
a source file not found error similar to the "pane5 doesn't exist"
error you saw earlier...

Some general diagnostic tricks: if you run with the console window
open, and it is hovering in the hourglass, go to the console, and
type:

tk gdbtk_idle

If this gives you back control, it means that you did get some error
that jumped over the point where the gui meant to call idle...  If you 
can reproduce the error, then while it is still spinning, attach to it 
with something like tkCon, and do:

set ::errorInfo

and see if that gives you any useful info...

On the lack of details.  What we found was that if you are doing
embedded development, then having all the arguments in the stack
window (I presume this is mainly what you want?) caused Insight to be
really slow at stepping, etc.  So we removed these.  The ultimate goal 
is to redo the window so that each level of the stack has a little
exposure triangle that reveals the arguments.  That way, you could
progressively reveal those you were interested in.  At this point, it
would also be easy to add an option to open all the turndowns by
default, which you may or may not want for native development.

But, as you may be coming to expect now, we didn't have the time to
complete the advanced design, so you are left with what is there now.

Jim

 > Today I tried to use the stack window.
 > 
 > First, it doesn't display nearly enough information.
 > 
 > Second, clicking on a line does not do anything useful.  In fact, it
 > hoses my session.  The cursor changes to an hourglass over the source
 > window, and the source window just goes blank.  Nothing I do changes
 > this.
 > 
 > Tom
 > 

^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: stack window
  2000-04-10 12:55 ` James Ingham
@ 2000-04-10 13:04   ` Tom Tromey
  2000-04-10 13:21     ` James Ingham
  0 siblings, 1 reply; 4+ messages in thread
From: Tom Tromey @ 2000-04-10 13:04 UTC (permalink / raw)
  To: James Ingham; +Cc: tromey, Insight List

Jim> I need more details.  In ordinary C code, the stack window works
Jim> fine...  I bet you have hit some error that didn't get caught.  Maybe
Jim> a source file not found error similar to the "pane5 doesn't exist"
Jim> error you saw earlier...

I found out how to reproduce it: if I click on a frame, and gdb can't
find the file for that frame, then I get the bad behavior.

Jim> But, as you may be coming to expect now, we didn't have the time
Jim> to complete the advanced design, so you are left with what is
Jim> there now.

Thanks for the info.

T

^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: stack window
  2000-04-10 13:04   ` Tom Tromey
@ 2000-04-10 13:21     ` James Ingham
  0 siblings, 0 replies; 4+ messages in thread
From: James Ingham @ 2000-04-10 13:21 UTC (permalink / raw)
  To: Tom Tromey; +Cc: James Ingham, Insight List

Tom,

Yeah, this is the same error as the one you reported before where you
got a Tcl stack traceback.  The reason is the same: not finding a
source file causes the Source window's cache to get confused, and you
are pretty much lost from there on out.  This has to be fixed, but the 
real solution is to rip out the code and rewrite it, which is a couple 
two or three of weeks of work.  I really don't want to try bandaging
this particular bit of code any more, as it is pretty bad, and it
implements a very important part of the UI.

Jim

 > Jim> I need more details.  In ordinary C code, the stack window works
 > Jim> fine...  I bet you have hit some error that didn't get caught.  Maybe
 > Jim> a source file not found error similar to the "pane5 doesn't exist"
 > Jim> error you saw earlier...
 > 
 > I found out how to reproduce it: if I click on a frame, and gdb can't
 > find the file for that frame, then I get the bad behavior.
 > 
 > Jim> But, as you may be coming to expect now, we didn't have the time
 > Jim> to complete the advanced design, so you are left with what is
 > Jim> there now.
 > 
 > Thanks for the info.
 > 
 > T
 > 

^ permalink raw reply	[flat|nested] 4+ messages in thread

end of thread, other threads:[~2000-04-10 13:21 UTC | newest]

Thread overview: 4+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2000-04-10 12:13 stack window Tom Tromey
2000-04-10 12:55 ` James Ingham
2000-04-10 13:04   ` Tom Tromey
2000-04-10 13:21     ` James Ingham

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).