Binutils version for GCC 4.8

Binutils version for GCC 4.8

[Sebastian Huber]

Hello,

do I need a specific Binutils version (I use version 2.22) for GCC 4.8 (trunk)? 
  I have problems with the debug information:

powerpc-rtems4.11-objdump --dwarf some.exe  > /dev/null
powerpc-rtems4.11-objdump: Warning: There is a hole [0xf2dc7 - 0xf2def] in 
.debug_loc section.
powerpc-rtems4.11-objdump: Warning: There is a hole [0xf53ea - 0xf5474] in 
.debug_loc section.
powerpc-rtems4.11-objdump: Warning: There is an overlap [0x30 - 0x18] in 
.debug_ranges section.
powerpc-rtems4.11-objdump: Warning: There is an overlap [0x160 - 0x148] in 
.debug_ranges section.
powerpc-rtems4.11-objdump: Warning: There is an overlap [0x560 - 0x548] in 
.debug_ranges section.
powerpc-rtems4.11-objdump: Warning: There is an overlap [0x8e8 - 0x8c0] in 
.debug_ranges section.
[...]
powerpc-rtems4.11-objdump: Warning: There is an overlap [0x32a30 - 0x32a10] in 
.debug_ranges section.
powerpc-rtems4.11-objdump: Warning: There is an overlap [0x32b28 - 0x32ae0] in 
.debug_ranges section.
powerpc-rtems4.11-objdump: Warning: There is an overlap [0x32be0 - 0x32bc8] in 
.debug_ranges section.

GDB 7.4.1 says that:

(gdb) s
Single stepping until exit from function _Workspace_Handler_initialization,
which has no line number information.
0x00000410 in bsp_work_area_initialize ()

Other debugger refuse to load the ELF file.  Has someone a hint how I can check 
that the debug information is ok?

-- 
Sebastian Huber, embedded brains GmbH

Address : Obere Lagerstr. 30, D-82178 Puchheim, Germany
Phone   : +49 89 18 90 80 79-6
Fax     : +49 89 18 90 80 79-9
E-Mail  : sebastian.huber@embedded-brains.de
PGP     : Public key available on request.

Diese Nachricht ist keine geschäftliche Mitteilung im Sinne des EHUG.

Re: Binutils version for GCC 4.8

[Sebastian Huber]

On 11/09/2012 04:54 PM, Sebastian Huber wrote:
> Hello,
>
> do I need a specific Binutils version (I use version 2.22) for GCC 4.8 (trunk)?
>   I have problems with the debug information:
>
> powerpc-rtems4.11-objdump --dwarf some.exe  > /dev/null
> powerpc-rtems4.11-objdump: Warning: There is a hole [0xf2dc7 - 0xf2def] in
> .debug_loc section.
> powerpc-rtems4.11-objdump: Warning: There is a hole [0xf53ea - 0xf5474] in
> .debug_loc section.
> powerpc-rtems4.11-objdump: Warning: There is an overlap [0x30 - 0x18] in
> .debug_ranges section.
> powerpc-rtems4.11-objdump: Warning: There is an overlap [0x160 - 0x148] in
> .debug_ranges section.
> powerpc-rtems4.11-objdump: Warning: There is an overlap [0x560 - 0x548] in
> .debug_ranges section.
> powerpc-rtems4.11-objdump: Warning: There is an overlap [0x8e8 - 0x8c0] in
> .debug_ranges section.
> [...]
> powerpc-rtems4.11-objdump: Warning: There is an overlap [0x32a30 - 0x32a10] in
> .debug_ranges section.
> powerpc-rtems4.11-objdump: Warning: There is an overlap [0x32b28 - 0x32ae0] in
> .debug_ranges section.
> powerpc-rtems4.11-objdump: Warning: There is an overlap [0x32be0 - 0x32bc8] in
> .debug_ranges section.

Ok, the bug with the overlaps disappeared with a more recent trunk version.

>
> GDB 7.4.1 says that:
>
> (gdb) s
> Single stepping until exit from function _Workspace_Handler_initialization,
> which has no line number information.
> 0x00000410 in bsp_work_area_initialize ()
>
> Other debugger refuse to load the ELF file.  Has someone a hint how I can check
> that the debug information is ok?
>

This was due to the new DWARF 4 debug information emitted by GCC 4.8.  GDB 7.5 
can cope with it.

-- 
Sebastian Huber, embedded brains GmbH

Address : Obere Lagerstr. 30, D-82178 Puchheim, Germany
Phone   : +49 89 18 90 80 79-6
Fax     : +49 89 18 90 80 79-9
E-Mail  : sebastian.huber@embedded-brains.de
PGP     : Public key available on request.

Diese Nachricht ist keine geschäftliche Mitteilung im Sinne des EHUG.