Mail Archives: cygwin/2002/06/25/15:51:34
On Tue, Jun 25, 2002 at 09:44:33PM +0200, Gerrit P. Haase wrote:
>>>(gdb) bt
>>>#0 0x77f34ac4 in _libkernel32_a_iname ()
>>>#1 0xffffffff in ?? ()
>>>#2 0x6107120a in _libkernel32_a_iname ()
>>>#3 0x0049cfcf in Perl_pp_link (my_perl=0xa013cf0) at pp_sys.c:3556
>> ^^^^^^^^^^^ ^^^^^^^^^^^^^
>
>> What function is this calling?
>
>PP(pp_link)
>{
>#ifdef HAS_LINK
> dSP; dTARGET;
> STRLEN n_a;
> char *tmps2 = POPpx;
> char *tmps = SvPV(TOPs, n_a);
> TAINT_PROPER("link");
> SETi( PerlLIO_link(tmps, tmps2) >= 0 ); <----3556
> RETURN;
>#else
> DIE(aTHX_ PL_no_func, "link");
>#endif
>}
>
>iperlsys.h:
>#define PerlLIO_link(oldname, newname) \
> (*PL_LIO->pLink)(PL_LIO, (oldname), (newname))
>
>pp.h:
>#define SETi(i) STMT_START { sv_setiv(TARG, (IV)(i)); SETTARG; } STMT_END
So, that confirms that it is dying in the link() call even though the
location from your stack dump doesn't make sense (not your fault).
Do you have a debugging DLL handy or can I send you one? If you set a
breakpoint in _link in the cygwin DLL and single step along until it
dies, it should be instructive.
Also does the latest snapshot behave any differently?
cgf
--
Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Bug reporting: http://cygwin.com/bugs.html
Documentation: http://cygwin.com/docs.html
FAQ: http://cygwin.com/faq/
- Raw text -