m68k-hp problem

[Hans Cappelle]

>>>>> Jeffrey A Law writes:

>   In message < 199811091433.PAA24734@yeti.imec.be >you write:
>> Hi,
>> 
>> I compiled a m68k-hp cross compiler with the egcs-19981101
>> snapshot. To my surprise it fails in dwarf2out.c, even for
>> "trivial" code examples and if no debugging info is requested.
>> 
> I'd be very interested in knowing how/why that configuration is
> using dwarf2 -- I don't think it should be using any of the dwarf2
> code unless your target is using elf or possibly coff.

The INCOMING_RETURN_ADDR_RTX macro in m68k.h triggers
that DWARF2_UNWIND_INFO is defined in defaults.h.
The latter is included in toplev.c.

This causes the implicit use of dwarf2 code for m68k targets.

best regards,

--
Hans Cappelle,

Embedded software engineer.

IMEC / VSDM division, Kapeldreef 75, B-3001 Leuven, Belgium
phone: +32/16/281.601      e-mail: cappelle@imec.be
fax  : +32/16/281.515      www   : http://www.imec.be


PS :

-------------------------------------------------
m68k.h:1804
/* Before the prologue, RA is at 0(%sp).  */
#define INCOMING_RETURN_ADDR_RTX \
  gen_rtx_MEM (VOIDmode, gen_rtx_REG (VOIDmode, STACK_POINTER_REGNUM))

-------------------------------------------------
defaults.h:136
/* If we have a definition of INCOMING_RETURN_ADDR_RTX, assume that
   the rest of the DWARF 2 frame unwind support is also provided.  */
#if !defined (DWARF2_UNWIND_INFO) && defined (INCOMING_RETURN_ADDR_RTX)
#define DWARF2_UNWIND_INFO 1
#endif


-------------------------------------------------
toplev.c includes defaults.h


-------------------------------------------------
backtrace from exit :
(gdb) bt
#0  0x7af59484 in ___exit ()
#1  0x1d35d8 in output_call_frame_info (for_eh=1) at dwarf2out.c:1669
#2  0x1d4d84 in dwarf2out_frame_finish () at dwarf2out.c:1949
#3  0x11c354 in compile_file (name=0x7b03a034 "/imec/vsdm/cappelle/tmp/foo.ii") at toplev.c:3003
#4  0x120a2c in main (argc=9, argv=0x7b03a754, envp=0x7b03a77c) at toplev.c:4838