public inbox for insight@sourceware.org
 help / color / mirror / Atom feed
* symbol with in disassembly window
@ 2002-01-04 13:41 Duane Ellis
  2002-01-04 13:48 ` Keith Seitz
  0 siblings, 1 reply; 4+ messages in thread
From: Duane Ellis @ 2002-01-04 13:41 UTC (permalink / raw)
  To: insight


In the insight source window - when showing dissassembly is there a
prefered way to limit the symbol width? Or a planned way?

The problem [cosmetic, but really bad] occurs when you have a C++
function with many parameters and a long name, there ain't no screen
wide enough.

ie: A GDB set command, or a TK variable instead?

What I did before was an arbitary hack (set/unset a global variable
dynamically while building the screen), I'm looking for a better
suggestion.  Or is something like this in the works?

-Duane.

FYI - I'm updating my custom port of Insight (based on the 19991108
snapshot) for a custom RISC CPU and eBook platform you may have seen
it:

     http://www.franklin.com/ebookman

It's working - partly - and I'm working though a list of problems

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

* Re: symbol with in disassembly window
  2002-01-04 13:41 symbol with in disassembly window Duane Ellis
@ 2002-01-04 13:48 ` Keith Seitz
  2002-01-04 14:03   ` Duane Ellis
  0 siblings, 1 reply; 4+ messages in thread
From: Keith Seitz @ 2002-01-04 13:48 UTC (permalink / raw)
  To: Duane Ellis; +Cc: insight

On Fri, 4 Jan 2002, Duane Ellis wrote:

> In the insight source window - when showing dissassembly is there a
> prefered way to limit the symbol width? Or a planned way?

This is something that gets forced-fed to us. We might be able to hack
something in, but I am hestitant to do so, since this problem also
presumably exists in gdb (right?).

I would recommend asking over on gdb@sources.redhat.com about this and the
"disassemble" command.

Keith


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

* Re: symbol with in disassembly window
  2002-01-04 13:48 ` Keith Seitz
@ 2002-01-04 14:03   ` Duane Ellis
  2002-01-04 14:12     ` Keith Seitz
  0 siblings, 1 reply; 4+ messages in thread
From: Duane Ellis @ 2002-01-04 14:03 UTC (permalink / raw)
  To: insight


keiths> since this problem also presumably exists in gdb (right?).

Most certianly, however there are times when 

A) you want the fully un-ambigious name (Even though it is 100+ chars
   long) and thus I too am hestitant to change perminately.

B) There are times though when a prefix or abeviated version is
   enough, ie:

	someFunc(int,int,int,int,int ....: 

This is *VERY* apparent in the disassembly window when dealing with
managled C++ code, not so much with plain C.

It seems that the gdb core function:

   print_address_symbolic()

nees an extra parameter, "maxwidth" with something like 0 
meaning the sky's the limit.

This would neccesisate changes in three places.

(A) printcmd.c 
    Wrapper or a new function.

(B) in gdbtk-cmds.c, to pass the new parameter.

(C) some place - gdb/gdbtk/generic/???.c or gdb/???.c to set/show the
    parameter.

-Duane.

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

* Re: symbol with in disassembly window
  2002-01-04 14:03   ` Duane Ellis
@ 2002-01-04 14:12     ` Keith Seitz
  0 siblings, 0 replies; 4+ messages in thread
From: Keith Seitz @ 2002-01-04 14:12 UTC (permalink / raw)
  To: Duane Ellis; +Cc: insight

On Fri, 4 Jan 2002, Duane Ellis wrote:

>
> keiths> since this problem also presumably exists in gdb (right?).
>
> Most certianly, however there are times when
>
> A) you want the fully un-ambigious name (Even though it is 100+ chars
>    long) and thus I too am hestitant to change perminately.
>
> B) There are times though when a prefix or abeviated version is
>    enough, ie:
>
> 	someFunc(int,int,int,int,int ....:
>

Yes, but either way, gdb's "disassemble" command suffers the same problems
and would benefit equally by such a change.

> It seems that the gdb core function:
>
>    print_address_symbolic()

That was my thought, as well. This function is not under my control,
though. You'll have to discuss changing it with the gdb folks.

I am not disagreeing that the change would be good. I'm just disagreeing
with changing it in insight alone. I think the proper place to do this (as
you've pointed out) is in "core" gdb.

Keith


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

end of thread, other threads:[~2002-01-04 22:12 UTC | newest]

Thread overview: 4+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2002-01-04 13:41 symbol with in disassembly window Duane Ellis
2002-01-04 13:48 ` Keith Seitz
2002-01-04 14:03   ` Duane Ellis
2002-01-04 14:12     ` Keith Seitz

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).