* multi-arch debug doesn't dump definitions as requested
@ 2002-02-06 9:34 Richard Earnshaw
2002-02-07 18:38 ` Andrew Cagney
0 siblings, 1 reply; 4+ messages in thread
From: Richard Earnshaw @ 2002-02-06 9:34 UTC (permalink / raw)
To: gdb; +Cc: Richard.Earnshaw
According to
http://sources.redhat.com/gdb/papers/multi-arch/howto.html
Rebuild and then run GDB vis:
gdb/gdb
(gdb)
(gdb) set archdebug 1
(gdb) file x-y-z
....
If all goes well, gdb will report the current value of all the
architecture dependent macro's. At this stage must are still
being taken from hardwired definitions in your
tm-XXX.h file. Failing that, an internal error message
reporting an uninitialised architecture vector member will be
reported.
Except that that doesn't happen. All I get is
(gdb) set archdebug 1
Warning: command 'set archdebug' is deprecated.
Use 'set debug arch'.
(gdb) file gdb
gdbarch_update: info.bfd_arch_info arm
gdbarch_update: info.byte_order 1 (little)
gdbarch_update: info.abfd 0x1d2e00
gdbarch_update: info.tdep_info 0x0
gdbarch_update: Architecture 0x001d3000 (arm) unchanged
Reading symbols from gdb...done.
Note, "set debug arch 1" doesn't do anything different either.
How about a "maint" (or maint info) command?
R.
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: multi-arch debug doesn't dump definitions as requested
2002-02-06 9:34 multi-arch debug doesn't dump definitions as requested Richard Earnshaw
@ 2002-02-07 18:38 ` Andrew Cagney
2002-02-08 2:22 ` Richard Earnshaw
0 siblings, 1 reply; 4+ messages in thread
From: Andrew Cagney @ 2002-02-07 18:38 UTC (permalink / raw)
To: Richard.Earnshaw; +Cc: gdb
> According to
> http://sources.redhat.com/gdb/papers/multi-arch/howto.html
>
> Rebuild and then run GDB vis:
>
> gdb/gdb
> (gdb)
> (gdb) set archdebug 1
> (gdb) file x-y-z
> ....
>
> If all goes well, gdb will report the current value of all the
> architecture dependent macro's. At this stage must are still
> being taken from hardwired definitions in your
> tm-XXX.h file. Failing that, an internal error message
> reporting an uninitialised architecture vector member will be
> reported.
>
> Except that that doesn't happen. All I get is
>
> (gdb) set archdebug 1
> Warning: command 'set archdebug' is deprecated.
> Use 'set debug arch'.
>
> (gdb) file gdb
> gdbarch_update: info.bfd_arch_info arm
> gdbarch_update: info.byte_order 1 (little)
> gdbarch_update: info.abfd 0x1d2e00
> gdbarch_update: info.tdep_info 0x0
> gdbarch_update: Architecture 0x001d3000 (arm) unchanged
> Reading symbols from gdb...done.
>
> Note, "set debug arch 1" doesn't do anything different either.
>
> How about a "maint" (or maint info) command?
Try:
(gdb) maint print architecture [ <savefile> ]
and look at sizeof.exp.
The lack of output comes, at least in part, because it isn't possible to
print the macros/values safely.
Andrew
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: multi-arch debug doesn't dump definitions as requested
2002-02-07 18:38 ` Andrew Cagney
@ 2002-02-08 2:22 ` Richard Earnshaw
2002-02-08 2:41 ` Pierre Muller
0 siblings, 1 reply; 4+ messages in thread
From: Richard Earnshaw @ 2002-02-08 2:22 UTC (permalink / raw)
To: Andrew Cagney; +Cc: Richard.Earnshaw, gdb
>
> Try:
>
> (gdb) maint print architecture [ <savefile> ]
>
> and look at sizeof.exp.
>
> The lack of output comes, at least in part, because it isn't possible to
> print the macros/values safely.
>
> Andrew
Thanks. Shouldn't http://sources.redhat.com/gdb/papers/multi-arch/howto.ht
ml be updated to show this. At present it says:
Add:
#ifndef GDB_MULTI_ARCH
#define GDB_MULTI_ARCH 1
#endif
to the top of your tm-XXX.h file. Rebuild and then run GDB vis:
gdb/gdb
(gdb)
(gdb) set archdebug 1
(gdb) file x-y-z
....
If all goes well, gdb will report the current value of all the
architecture dependent macro's. ...
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: multi-arch debug doesn't dump definitions as requested
2002-02-08 2:22 ` Richard Earnshaw
@ 2002-02-08 2:41 ` Pierre Muller
0 siblings, 0 replies; 4+ messages in thread
From: Pierre Muller @ 2002-02-08 2:41 UTC (permalink / raw)
To: Richard.Earnshaw, Andrew Cagney; +Cc: gdb
At 11:21 08/02/2002 , Richard Earnshaw a écrit:
> >
> > Try:
> >
> > (gdb) maint print architecture [ <savefile> ]
> >
> > and look at sizeof.exp.
> >
> > The lack of output comes, at least in part, because it isn't possible to
> > print the macros/values safely.
> >
> > Andrew
>
>Thanks. Shouldn't http://sources.redhat.com/gdb/papers/multi-arch/howto.ht
>ml be updated to show this. At present it says:
>
>Add:
>
>#ifndef GDB_MULTI_ARCH
>#define GDB_MULTI_ARCH 1
>#endif
>
>to the top of your tm-XXX.h file. Rebuild and then run GDB vis:
>
>gdb/gdb
>(gdb)
>(gdb) set archdebug 1
>(gdb) file x-y-z
>....
>
>If all goes well, gdb will report the current value of all the
>architecture dependent macro's. ...
I made a simple test on a recent cygwin native gdb built from CVS tree.
>(gdb) set archdebug 1
(gdb) set archdebug 1
Warning: command 'set archdebug' is deprecated.
Use 'set debug arch'.
Maybe this should also be changed on html, unless there are still some hidden differences...
afterr setting debug arch to 1
(gdb) file ./gdb
gdbarch_update: info.bfd_arch_info i386
gdbarch_update: info.byte_order 1 (little)
gdbarch_update: info.abfd 0xa035030
gdbarch_update: info.tdep_info 0x0
gdbarch_update: Architecture 0x0a0281f0 (i386) unchanged
Reading symbols from ./gdb...done.
Is this all that is defined in i386-tdep.c ?
I see more set_gdbarch_****
calls in i386-tdep.c source, but they are not reflected here....
If I try to set debug arch to higher values I get lots of
output, but not much understandable...
Pierre Muller
Institut Charles Sadron
6,rue Boussingault
F 67083 STRASBOURG CEDEX (France)
mailto:muller@ics.u-strasbg.fr
Phone : (33)-3-88-41-40-07 Fax : (33)-3-88-41-40-99
^ permalink raw reply [flat|nested] 4+ messages in thread
end of thread, other threads:[~2002-02-08 10:41 UTC | newest]
Thread overview: 4+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2002-02-06 9:34 multi-arch debug doesn't dump definitions as requested Richard Earnshaw
2002-02-07 18:38 ` Andrew Cagney
2002-02-08 2:22 ` Richard Earnshaw
2002-02-08 2:41 ` Pierre Muller
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox