[CUWiN-Dev] nodeinfo build error

[dan blah]

on cuw with rev 4085:

#    create  arithmetic/.depend
rm -f .depend
CC=/u4/blah/cuwin/scratch/T/bin/i386--netbsdelf-gcc
/u4/blah/cuwin/scratch/T/bin                                 /nbmkdep
-d -f .depend -s .o\ .po\ .so\ .ln powi32.d
depend ===> clearsilver
depend ===> clearsilver/cgi
nbmake: don't know how to make cgi.c. Stop

nbmake: stopped in /u4/blah/cuwin/trunk/src/clearsilver/cgi

*** Failed target:  depend-cgi
*** Failed command: _makedirtarget() { dir="$1"; shift; target="$1";
shift; case                                  "${dir}" in /*)
this="${dir}/"; real="${dir}" ;; .) this="clearsilver/"; real="
                         /u4/blah/cuwin/trunk/src/clearsilver" ;; *)
this="clearsilver/${dir}/"; real="/u
4/blah/cuwin/trunk/src/clearsilver/${dir}" ;; esac; show=${this:-.};
echo "${tar                                 get} ===> ${show%/}${1:+
(with: $@)}"; cd "${real}" && /u4/blah/cuwin/scratch/T/
                 bin/nbmake _THISDIR_="${this}" "$@" ${target}; };
_makedirtarget cgi depend
*** Error code 2

Stop.
nbmake: stopped in /u4/blah/cuwin/trunk/src/clearsilver

*** Failed target:  depend-clearsilver
*** Failed command: _makedirtarget() { dir="$1"; shift; target="$1";
shift; case                                  "${dir}" in /*)
this="${dir}/"; real="${dir}" ;; .) this=""; real="/u4/blah/cuw
                         in/trunk/src" ;; *) this="${dir}/";
real="/u4/blah/cuwin/trunk/src/${dir}" ;; es
      ac; show=${this:-.}; echo "${target} ===> ${show%/}${1:+ (with:
$@)}"; cd "${rea                                 l}" &&
/u4/blah/cuwin/scratch/T/bin/nbmake _THISDIR_="${this}" "$@"
${target}; }                                 ; _makedirtarget
clearsilver depend
*** Error code 1

Stop.
nbmake: stopped in /u4/blah/cuwin/trunk/src


On 7/15/06, David Young <dyoung at pobox.com> wrote:
> On Fri, Jul 14, 2006 at 05:58:24PM -0700, Chase Phillips wrote:
> > Bryan, Jeff, or Dave will need to confirm, but there appears to be
> > misuse of the adler pointer in nc_update_adler.  Is there only one
> > call to nc_update_adler -- only from nodeconfig_read_cuw_conf?
> >
> > It appears in nodeconfig.c's current state, fingerprint should just be
> > set to NULL at declaration, and someone should shed light on what
> > circumstances the newly created fingerprint uint32_t would take on a
> > value other than what the compiler gives it.  (E.g. does Linux zero
> > out variable memory at declaration?)
>
> I have fixed the compile errors.  I haven't pulled them up to the 0.7.0
> release branch.
>
> Bryan, occasionally test-compile with GCC4, which is in the new NetBSD
> snapshot I mentioned here a while ago.  GCC4 is more picky than GCC3
> ever was.
>
> Dave
>
> --
> David Young             OJC Technologies
> dyoung at ojctech.com      Urbana, IL * (217) 278-3933
> _______________________________________________
> CU-Wireless-Dev mailing list
> CU-Wireless-Dev at lists.cuwireless.net
> http://lists.chambana.net/cgi-bin/listinfo/cu-wireless-dev
>


-- 
Daniel