Using the workaround on trac addressed my problem with building evolution-data-server, but it's still a hack. Still, if that problem comes up in the future, there are at least two people who have been able to get around it. The next problem is liboil. Particularly, a weird Assembler pseudo-op called ".balign" is getting introduced, and GCC apparently doesn't know how to deal with it. I'm still unable to log into trac, so I can't open a bug on this myself. In the interim, I've sent a 'port -dv install liboil' log to Daniel Ericsson. Has anyone found this .balign problem before? If so, how did it get solved? A (non- -dv) port install output follows here. ===== ---> Building liboil with target all Error: Target com.apple.build returned: shell command "cd "/opt/local/var/db/dports/build/_opt_local_var_db_dports_sources_rsync.rsync.darwinports.org_dpupdate_dports_devel_liboil/work/liboil-0.3.6" && make all" returned error 2 Command output: make all-recursive Making all in liboil Making all in colorspace make[3]: Nothing to be done for `all'. Making all in conv make[3]: Nothing to be done for `all'. Making all in copy make[3]: Nothing to be done for `all'. Making all in dct make[3]: Nothing to be done for `all'. Making all in i386 if /bin/sh ../../libtool --tag=CC --mode=compile gcc -DHAVE_CONFIG_H -I. -I. -I../.. -I/opt/local/include -L/opt/local/lib -Wall -D_BSD_SOURCE -D_GNU_SOURCE -I../.. -O2 -L/opt/local/lib -MT libi386_la-copy8x8_i386.lo -MD -MP -MF ".deps/libi386_la-copy8x8_i386.Tpo" -c -o libi386_la-copy8x8_i386.lo `test -f 'copy8x8_i386.c' || echo './'`copy8x8_i386.c; \ then mv -f ".deps/libi386_la-copy8x8_i386.Tpo" ".deps/libi386_la-copy8x8_i386.Plo"; else rm -f ".deps/libi386_la-copy8x8_i386.Tpo"; exit 1; fi gcc -DHAVE_CONFIG_H -I. -I. -I../.. -I/opt/local/include -L/opt/local/lib -Wall -D_BSD_SOURCE -D_GNU_SOURCE -I../.. -O2 -L/opt/local/lib -MT libi386_la-copy8x8_i386.lo -MD -MP -MF .deps/libi386_la-copy8x8_i386.Tpo -c copy8x8_i386.c -fno-common -DPIC -o .libs/libi386_la-copy8x8_i386.o /var/tmp//ccgwIKbW.s:24:Unknown pseudo-op: .balign /var/tmp//ccgwIKbW.s:24:Rest of line ignored. 1st junk character valued 49 (1). make[3]: *** [libi386_la-copy8x8_i386.lo] Error 1 make[2]: *** [all-recursive] Error 1 make[1]: *** [all-recursive] Error 1 make: *** [all] Error 2
On Dec 6, 2006, at 22:33, Robert J. Hansen wrote:
Using the workaround on trac addressed my problem with building evolution-data-server, but it's still a hack. Still, if that problem comes up in the future, there are at least two people who have been able to get around it.
The next problem is liboil. Particularly, a weird Assembler pseudo-op called ".balign" is getting introduced, and GCC apparently doesn't know how to deal with it.
I'm still unable to log into trac, so I can't open a bug on this myself. In the interim, I've sent a 'port -dv install liboil' log to Daniel Ericsson.
Has anyone found this .balign problem before? If so, how did it get solved?
A (non- -dv) port install output follows here.
[snip]
/var/tmp//ccgwIKbW.s:24:Unknown pseudo-op: .balign /var/tmp//ccgwIKbW.s:24:Rest of line ignored. 1st junk character valued 49 (1). make[3]: *** [libi386_la-copy8x8_i386.lo] Error 1 make[2]: *** [all-recursive] Error 1 make[1]: *** [all-recursive] Error 1 make: *** [all] Error 2
320 results for "Unknown pseudo-op: .balign" on Google. Looks to me like that's a PowerPC thing, and you're on Intel. The liboil port is at version 0.3.6 which seems to be from November 2005. Current version is 0.3.10 from November 2006. Perhaps that fixes it. Port has no maintainer. Any reason we can't update this to 0.3.10? I have no idea what it is, what it's for or what might be using it.
Ryan Schmidt wrote:
320 results for "Unknown pseudo-op: .balign" on Google. Looks to me like that's a PowerPC thing, and you're on Intel.
It's actually not a PowerPC thing. If I remember my gas right, it's just a macro; you should be able to substitute '.align X' for any set of '.balign Y'. If memory serves, X = log2 of Y, or maybe I have that reversed. I haven't needed to know gas since undergrad. It appears to be the case, judging from some traffic on the Fink list, that it's caused by Mactel gas being old and out-of-date compared to the current gas in binutils.
it. Port has no maintainer. Any reason we can't update this to 0.3.10? I have no idea what it is, what it's for or what might be using it.
liboil.freedesktop.org describes it as "a library of simple functions that are optimized for various CPUs". A lot of stuff in GNOME depends upon it.
participants (2)
-
Robert J. Hansen
-
Ryan Schmidt