01:25.50 |
starseeker |
sighs - well, guess I know
what to work on Monday then |
01:39.21 |
*** join/#brlcad juanman
(~quassel@unaffiliated/juanman) |
03:54.25 |
starseeker |
O.o |
03:54.27 |
starseeker |
./.libs/libbu.so: undefined reference to
`fnblank' |
03:54.32 |
starseeker |
when building with clang |
03:54.42 |
starseeker |
and several others |
03:55.04 |
starseeker |
error comes when trying to link
htester |
06:45.17 |
brlcad |
fnblank is in fnmatch.c so it might be
dropping the references since they are declared inline |
06:46.15 |
brlcad |
removing the inline declaration should make
the undefined references go away, though I'd be curious to know
exactly why it's failing |
06:57.52 |
*** join/#brlcad Ralith
(~ralith@S010600221561996a.vc.shawcable.net) |
14:40.53 |
starseeker |
yeah, removing inline did it - weird |
14:41.53 |
starseeker |
maybe something about the way it's handing the
charclasses array? |
14:57.13 |
*** join/#brlcad juanman
(~quassel@unaffiliated/juanman) |
15:14.04 |
``Erik |
probably assuming that inline and static are
synonymous |
15:14.25 |
``Erik |
or that inline insinuates static |
15:15.34 |
``Erik |
(which is wrong, inline is a hint, static is a
statement) |
16:23.04 |
*** join/#brlcad R0b0t1
(~Enigma@64-136-219-55.dyn.everestkc.net) |
16:23.04 |
*** join/#brlcad R0b0t1
(~Enigma@unaffiliated/r0b0t1) |
19:47.14 |
*** join/#brlcad Ralith
(~ralith@S010600221561996a.vc.shawcable.net) |
21:28.39 |
brlcad |
starseeker: can you dig in and figure out that
cause, whether it's a compiler bug or linker issue? maybe ask
their devs if the docs don't say |
21:29.07 |
brlcad |
if it's a code problem, then we should edit
for portability -- if it's a bug, they can fix it |
23:03.47 |
starseeker |
brlcad: sure, I'll take a wack at it - I've
been holding off bugging them til things stablized, but now that
they're saying they've pretty much completed the current C++
standard features I guess it's time |
23:15.57 |
starseeker |
actually... |
23:17.00 |
starseeker |
when I replace /usr/bin/ld with llvm-ld it
works... |
23:17.09 |
starseeker |
tries a clean build to make
sure that's for real |
23:30.03 |
starseeker |
yep |
23:30.28 |
starseeker |
so it's probably some sort of incompatibility
between clang and ld |
23:32.03 |
starseeker |
dingnabbit, it still doesn't like the
bu_byteoffset stuff |
23:35.46 |
starseeker |
brlcad: I'm not sure what exactly to file as a
bug report about that byteoffset issue, but it's been there for a
looong time now so presumably it will stay there unless we file
something |
23:37.45 |
starseeker |
tries the cfe-dev
list |