Problems with debug info in m68k-coff objdump/objcopy
Steve deRosier
sderosier@vari-lite.com
Fri Jan 19 09:34:00 GMT 2001
I am converting a project from an MRI tool chain to the GNU tool chain for
various m68k chips. The main problem that I am having has to do with not
being able to display C/C++ source in the emulator window. We are using an
HP emulator (HPI64749) which requires an ieee format file. I have the
project running on the target hardware+emulator just fine, but still having
problems with the debugging info.
I thought perhaps that when I use objcopy to create my ieee file from my
m68k-coff format file (can I create a m68k-ieee cross tool chain? I looked
and experimented, but I don't think this is a supported target; if I'm wrong
please let me know) I should use the --debugging switch and that might do a
conversion on the debug info and solve my problems.
Unfortunately, objcopy didn't like the debug info in the file. I've tried
using both coff and stabs format debug info switches on my compiler, but it
doesn't make a difference (not surprising, since the file referenced in the
info is not directly part of my project).
objdump has the same problem reading the file, so I have included the output
from it below.
All in all I find this a little confusing, because I would expect that gcc
would create debugging info that binutils would understand. So--my problem,
or is it a bug? If my problem, any ideas on how to fix it? If not, does
someone have a patch, or can direct me on how to fix the problem?
Thanks,
- Steve
(sorry about the cross-post, did it incase anyone at crossgcc had ideas)
-------
beam 27: mobjdump -g cs.coff.x
cs.coff.x: file format coff-m68k
Bad stab: warning: _kill is not implemented and will always fail
Last stabs entries before error:
n_type n_desc n_value string
LSYM 0 00000000
__sbuf:T(7,14)=s8_base:(7,15)=*(0,11),0,32;_size:(0,1),32,32;;
LSYM 74 00000000 _fpos_t:t(7,16)=(0,3)
LSYM 0 00000000
__sFILE:T(7,5)=s88_p:(7,15),0,32;_r:(0,1),32,32;_w:(0,1),64,32;_flags:(0,8),
96,16;_file:(0,8),112,16;_bf:(7,14),128,64;_lbfsize:(0,1),192,32;_cookie:(7,
17)=*(0,19),224,32;read:(7,18)=*(7,19)=f(0,1),256,32;write:(7,20)=*(7,21)=f(
0,1),288,32;_seek:(7,22)=*(7,23)=f(7,16),320,32;close:(7,24)=*(7,25)=f(0,1),
352,32;_ub:(7,14),384,64;_up:(7,15),448,32;_ur:(0,1),480,32;_ubuf:(7,26)=ar(
0,1);0;2;(0,11),512,24;_nbuf:(7,27)=ar(0,1);0;0;(0,11),536,8;_lb:(7,14),544,
64;_blksize:(0,1),608,32;_offset:(0,1),640,32;_data:(7,28)=*(7,29)=xs_reent:
,672,32;;
LSYM 0 00000000
_reent:T(7,29)=s746_errno:(0,1),0,32;_stdin:(7,4),32,32;_stdout:(7,4),64,32;
_stderr:(7,4),96,32;_inc:(0,1),128,32;_emergency:(7,30)=ar(0,1);0;24;(0,2),1
60,200;_current_category:(0,1),368,32;_current_locale:(7,31)=*(0,2),400,32;_
_sdidinit:(0,1),432,32;__cleanup:(7,32)=*(7,33)=f(0,19),464,32;_result:(7,7)
,496,32;_result_k:(0,1),528,32;_p5s:(7,7),560,32;_freelist:(7,34)=*(7,7),592
,32;_cvtlen:(0,1),624,32;_cvtbuf:(7,35)=*(0,2),656,32;_new:(7,36)=u240_reent
:(7,37)=s74_rand_next:(0,4),0,32;_strtok_last:(7,35),32,32;_asctime_buf:(7,3
8)=ar(0,1);0;25;(0,2),64,208;_localtime_buf:(8,3),272,288;_gamma_signgam:(0,
1),560,32;;,0,592;_unused:(7,39)=s240_nextf:(7,40)=ar(0,1);0;29;(7,15),0,960
;_nmalloc:(7,41)=ar(0,1);0;29;(0,4),960,960;;,0,1920;;,688,1920;_atexit:(7,1
0),2608,32;_atexit0:(7,9),2640,1088;_sig_func:(7,42)=*(7,43)=*(7,44)=f(0,19)
,3728,32;__sglue:(7,2),3760,96;__sf:(7,45)=ar(0,1);0;2;(7,5),3856,2112;;
EINCL 0 00000000
BINCL 0 00000000
../../../../../../source/gcc-2.95.2/libgloss/libnosys/warning.h
EINCL 0 00000000
FUN 17 0081870c kill:F(0,1)
PSYM 15 00000008 pid:p(0,1)
PSYM 16 0000000c sig:p(0,1)
SLINE 17 00000000
SLINE 18 00000004
SLINE 19 0000000c
SLINE 20 0000000e
FUN 0 00000012
WARNING 0 00000000 warning: _kill is not implemented and will always
fail
------
--
Steve deRosier
Embedded Software Engineer
Vari-Lite International, Inc.
------
Want more information? See the CrossGCC FAQ, http://www.objsw.com/CrossGCC/
Want to unsubscribe? Send a note to crossgcc-unsubscribe@sourceware.cygnus.com
More information about the crossgcc
mailing list