Your message dated Tue, 13 Dec 2016 01:20:41 +0100
with message-id <[email protected]>
and subject line Re: Bug#784086: [[email protected]: Re: gdb-multiarch can't
debug 64-bit programs on 32-bit archs]
has caused the Debian Bug report #784086,
regarding gdb-multiarch can't debug 64-bit programs on 32-bit archs
to be marked as done.
This means that you claim that the problem has been dealt with.
If this is not the case it is now your responsibility to reopen the
Bug report if necessary, and/or fix the problem forthwith.
(NB: If you are a system administrator and have no idea what this
message is talking about, this may indicate a serious mail system
misconfiguration somewhere. Please contact [email protected]
immediately.)
--
784086: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=784086
Debian Bug Tracking System
Contact [email protected] with problems
--- Begin Message ---
Package: gdb-multiarch
Version: 7.7.1+dfsg-5
Severity: normal
Dear GDB Maintainers,
It's quite common to have a 64-bit kernel and 32-bit userspace in
Debian, eg powerpc and sparc, where 64-bit userspace is not fully or
officially supported.
It appears that gdb-multiarch:powerpc fails to debug any ppc64
programs. Similarly gdb-multiarch:i386 cannot debug amd64 programs.
Perhaps it's more useful to have gdb-multiarch:powerpc as a meta
package that depends on gdb-multiarch:ppc64. (also for other 32/64
arch pairs.)
Regards,
Yixuan
A test case, with ppc64 kernel, powerpc userspace, and a ppc64 chroot.
(powerpc) $ cat > hello.c
#include <stdio.h>
int main() {
printf("hello\n");
return 0;
}
(ppc64) $ gcc -static -ggdb -m64 hello.c -o hello
(powerpc) $ gdb-multiarch ./hello
Reading symbols from ./hello...done.
(gdb) start
Temporary breakpoint 1 at 0x100013a4: file hello.c, line 4.
Starting program: .../hello
warning: Selected architecture powerpc:common64 is not compatible with reported
target architecture powerpc:common
warning: Architecture rejected target-supplied description
Program received signal SIGTRAP, Trace/breakpoint trap.
0x100013a40002d032 in ?? ()
(gdb) c
Continuing.
Program received signal SIGTRAP, Trace/breakpoint trap.
0x100013a40002d032 in ?? ()
(gdb) n
Cannot find bounds of current function
(gdb)
--- End Message ---
--- Begin Message ---
On Sat, May 02, 2015 at 08:48:18PM -0400, Sergio Durigan Junior wrote:
> On Saturday, May 02 2015, GUO Yixuan wrote:
>
> > With gdb:powerpc and gdb64:powerpc available, the gdb-multiarch:powerpc
> > package seems rather useless, so it's still make sense to turn it to
> > a meta package to pull in gdb-multiarch:ppc64, if biarch is going to be
> > deprecated(?) in favour of multiarch.
>
> I may be wrong here (still new to Debian), but it seems to me that
> gdb-multiarch is meant to be used to debug not only 32- vs. 64-bit
> binaries, but also binaries from different architectures via gdbserver.
> For example, this section:
>
> <https://anonscm.debian.org/cgit/pkg-gdb/gdb.git/tree/debian/rules#n127>
>
> is responsible for choosing the targets that will be passed to
> --enable-targets when compiling the gdb-multiarch package. Therefore,
> it also seems to me that gdb-multiarch:powerpc should be able to debug a
> 64-bit binary on a 32-bit PowerPC system. Or am I missing something
> here?
Yes, that is the idea, that is correct. That way we can also avoid gdb-cross
package.
I am closing this report.
Regards
--
Hector Oron
signature.asc
Description: PGP signature
--- End Message ---
