libmcrypt 2.4.19 fails to compile with the following error: rm -fr .libs/loki97.la .libs/loki97.* .libs/loki97.* i586-pc-linux-gnu-gcc -shared loki97.lo /usr/lib/libltdl.so -Wl,-soname -Wl,loki97.so -o .libs/loki97.so ar cru .libs/loki97.a loki97.o ranlib .libs/loki97.a creating loki97.la (cd .libs && rm -f loki97.la && ln -s ../loki97.la loki97.la) /bin/sh ../../libtool --mode=compile i586-pc-linux-gnu-gcc -DHAVE_CONFIG_H -I. -I../.. -I../../lib -mcpu=k6 -march=k6 -O3 -pipe -c gost.c rm -f .libs/gost.lo i586-pc-linux-gnu-gcc -DHAVE_CONFIG_H -I. -I../.. -I../../lib -mcpu=k6 -march=k6 -O3 -pipe -c gost.c -fPIC -DPIC -o .libs/gost.lo gost.c:98: parse error before `1' gost.c: In function `_mcrypt_kboxinit': gost.c:153: subscripted value is neither array nor pointer make[3]: *** [gost.lo] Error 1 make[3]: Leaving directory `/var/tmp/portage/libmcrypt-2.4.19/work/libmcrypt-2.4.19/modules/algorithms' make[2]: *** [all-recursive] Error 1 make[2]: Leaving directory `/var/tmp/portage/libmcrypt-2.4.19/work/libmcrypt-2.4.19/modules' make[1]: *** [all-recursive] Error 1 make[1]: Leaving directory `/var/tmp/portage/libmcrypt-2.4.19/work/libmcrypt-2.4.19' make: *** [all-recursive-am] Error 2 !!! ERROR: The ebuild did not complete successfully. !!! Function src_compile, Line -63, Exitcode 2 !!! (no error message) !!! emerge aborting on /usr/portage/dev-libs/libmcrypt/libmcrypt-2.4.19.ebuild
*** Bug 2277 has been marked as a duplicate of this bug. ***
What gcc, m4, autoconf are you using?
using: gcc 2.95.3 m4 1.4p autoconf 2.52d
Works for me && only difference in my setup is autoconf. I'm using version sys-devel/autoconf-2.13-r2
I have the same setup as comment #3. Works fine here...
Doesn't compile for me either. The problem arises when using -mcpu=k6 -march=k6 optimizations. A quick work around is to use the i586 option instead. versions: (as reported by emerge) gcc 2.95.3 m4 1.4p autoconf 2.52d # autoconf --version Autoconf version 2.13 # m4 --version GNU m4 1.4ppre2 (options: modules) # gcc --version 2.95.3
Works fine for me when I have -march=k6, although this do not look like a automake/autoconf problem, but rather a gcc/"included headers" problem. What version linux-headers do you guys have ?
Hmm, ok, it is the "-mcpu=k6" that breaks things, as modules/algorithms/gosh.c have a "k6" array.
Hi guys, libmcrypt-2.4.19-r1 should fix this.