* dwarf2-frame.c and sym_private
@ 2003-06-22 16:52 Daniel Jacobowitz
2003-06-22 21:25 ` Mark Kettenis
0 siblings, 1 reply; 4+ messages in thread
From: Daniel Jacobowitz @ 2003-06-22 16:52 UTC (permalink / raw)
To: gdb, kettenis
The dwarf2-frame code blithely assumes that objfile->sym_private is
always an FDE description. But nothing stops the file from having
non-DWARF-2 debugging information, or no debugging information at
all... would you mind if I switched it to use a new dwarf2-specific
member of the objfile structure instead?
This would also let me avoid a crash in David's new namespace code; he
creates an objfile which does not have its section offsets initialized,
and dwarf2_frame_find_fde aborts.
--
Daniel Jacobowitz
MontaVista Software Debian GNU/Linux Developer
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: dwarf2-frame.c and sym_private
2003-06-22 16:52 dwarf2-frame.c and sym_private Daniel Jacobowitz
@ 2003-06-22 21:25 ` Mark Kettenis
2003-06-22 21:38 ` Andrew Cagney
2003-06-22 21:38 ` Daniel Jacobowitz
0 siblings, 2 replies; 4+ messages in thread
From: Mark Kettenis @ 2003-06-22 21:25 UTC (permalink / raw)
To: drow; +Cc: gdb
Date: Sun, 22 Jun 2003 12:52:35 -0400
From: Daniel Jacobowitz <drow@mvista.com>
The dwarf2-frame code blithely assumes that objfile->sym_private is
always an FDE description. But nothing stops the file from having
non-DWARF-2 debugging information, or no debugging information at
all... would you mind if I switched it to use a new dwarf2-specific
member of the objfile structure instead?
I wouldn't mind, although I'm not sure whether adding debug
format-specific members to the objfile structure is the right
approach. Assuming that an object file won't contain unwind info in
different formats, would adding a member for unwind info to the
objfile structure be an option.
Mark
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: dwarf2-frame.c and sym_private
2003-06-22 21:25 ` Mark Kettenis
@ 2003-06-22 21:38 ` Andrew Cagney
2003-06-22 21:38 ` Daniel Jacobowitz
1 sibling, 0 replies; 4+ messages in thread
From: Andrew Cagney @ 2003-06-22 21:38 UTC (permalink / raw)
To: Mark Kettenis; +Cc: drow, gdb
> Date: Sun, 22 Jun 2003 12:52:35 -0400
> From: Daniel Jacobowitz <drow@mvista.com>
>
> The dwarf2-frame code blithely assumes that objfile->sym_private is
> always an FDE description. But nothing stops the file from having
> non-DWARF-2 debugging information, or no debugging information at
> all... would you mind if I switched it to use a new dwarf2-specific
> member of the objfile structure instead?
>
> I wouldn't mind, although I'm not sure whether adding debug
> format-specific members to the objfile structure is the right
> approach. Assuming that an object file won't contain unwind info in
> different formats, would adding a member for unwind info to the
> objfile structure be an option.
See: The symbol auxilary.
http://sources.redhat.com/ml/gdb/2003-04/msg00017.html
Andrew
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: dwarf2-frame.c and sym_private
2003-06-22 21:25 ` Mark Kettenis
2003-06-22 21:38 ` Andrew Cagney
@ 2003-06-22 21:38 ` Daniel Jacobowitz
1 sibling, 0 replies; 4+ messages in thread
From: Daniel Jacobowitz @ 2003-06-22 21:38 UTC (permalink / raw)
To: Mark Kettenis; +Cc: gdb
On Sun, Jun 22, 2003 at 11:23:59PM +0200, Mark Kettenis wrote:
> Date: Sun, 22 Jun 2003 12:52:35 -0400
> From: Daniel Jacobowitz <drow@mvista.com>
>
> The dwarf2-frame code blithely assumes that objfile->sym_private is
> always an FDE description. But nothing stops the file from having
> non-DWARF-2 debugging information, or no debugging information at
> all... would you mind if I switched it to use a new dwarf2-specific
> member of the objfile structure instead?
>
> I wouldn't mind, although I'm not sure whether adding debug
> format-specific members to the objfile structure is the right
> approach. Assuming that an object file won't contain unwind info in
> different formats, would adding a member for unwind info to the
> objfile structure be an option.
TBH, I don't think I like the assumption. It's pretty academic - right
now dwarf2 is the one we support - but we need to be flexible in the
face of really strange input files.
--
Daniel Jacobowitz
MontaVista Software Debian GNU/Linux Developer
^ permalink raw reply [flat|nested] 4+ messages in thread
end of thread, other threads:[~2003-06-22 21:38 UTC | newest]
Thread overview: 4+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2003-06-22 16:52 dwarf2-frame.c and sym_private Daniel Jacobowitz
2003-06-22 21:25 ` Mark Kettenis
2003-06-22 21:38 ` Andrew Cagney
2003-06-22 21:38 ` Daniel Jacobowitz
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).