The issue cannot be seen in the attached build.log:
checking beecrypt/beecrypt.h usability... yes
checking beecrypt/beecrypt.h presence... yes
checking for beecrypt/beecrypt.h... yes
checking for mpfprintln in -lbeecrypt... no
-----------------------------------------^^ at least you see it failed, but that is not much helpful
I should have included config.log. ;( The test for the libbeecrypt.so failed and left unset WITH_BEECRYPT_INCLUDE="" in config.status instead of "-I/usr/include/beecrypt", due to some unresolved gcc 3.4.1? symbols. Please note that I wanted to compile rpm with gcc-4.3.x or 4.4.
After recompiling beecrypt with the current gcc I could subsequently compile also rpm against it.
I wanted to repeat the error but I cannot use i686-pc-linux-gnu-3.3.6 (no 3.4.1 anymore):
The issue cannot be seen in the attached build.log:
checking beecrypt/beecrypt.h usability... yes beecrypt. h... yes ------- ------- ------- ------- ------^ ^ at least you see it failed, but that
is not much helpful
checking beecrypt/beecrypt.h presence... yes
checking for beecrypt/
checking for mpfprintln in -lbeecrypt... no
-------
I should have included config.log. ;( The test for the libbeecrypt.so failed and left unset WITH_BEECRYPT_ INCLUDE= "" in config.status instead of "-I/usr/ include/ beecrypt" , due to some unresolved gcc 3.4.1? symbols. Please note that I wanted to compile rpm with gcc-4.3.x or 4.4.
After recompiling beecrypt with the current gcc I could subsequently compile also rpm against it.
I wanted to repeat the error but I cannot use i686-pc- linux-gnu- 3.3.6 (no 3.4.1 anymore):
i686-pc- linux-gnu- g++ -m32 -DHAVE_CONFIG_H -I. -I../.. -I../../include -D_LARGEFILE_SOURCE -D_FILE_ OFFSET_ BITS=64 -I/usr/include -I/usr/ lib/gcc- lib/i686- pc-linux- gnu/3.3. 6/include -DOPTIMIZE_I386 -O2 -march=pentium4 -mmmx -msse -msse2 -pipe -fno-strict- aliasing -ggdb -DNDEBUG -Wa,--noexecstack -MT PushbackInputSt ream.lo -MD -MP -MF .deps/PushbackI nputStream. Tpo -c PushbackInputSt ream.cxx -o PushbackInputSt ream.o >/dev/null 2>&1 am.Tpo .deps/InputStre am.Plo beecrypt/ c++/io/ Writer. h:42: sorry, unimplemented: adjusting beecrypt/ c++/io/ Writer. h:42: sorry, unimplemented: adjusting beecrypt/ c++/io/ Writer. h:42: sorry, unimplemented: adjusting beecrypt/ c++/io/ Writer. h:42: sorry, unimplemented: adjusting nputStream. Tpo .deps/PushbackI nputStream. Plo am.Tpo .deps/PrintStre am.Plo portage/ dev-libs/ beecrypt- 4.2.1/work/ beecrypt- 4.2.1/c+ +/io'
mv -f .deps/InputStre
../../include/
pointers for covariant returns
../../include/
pointers for covariant returns
../../include/
pointers for covariant returns
../../include/
pointers for covariant returns
distcc[31266] ERROR: compile Writer.cxx on localhost failed
make[3]: *** [Writer.lo] Error 1
make[3]: *** Waiting for unfinished jobs....
mv -f .deps/PushbackI
mv -f .deps/PrintStre
make[3]: Leaving directory `/var/tmp/
make[2]: *** [all-recursive] Error 1
When I tried to compile beecrypt with i686-pc- linux-gnu- 4.2.4 and afterwards tried to compile rpm, it worked as well.
My conclusion is, that the objects produced by different gcc versions were not compatible. I wonder whether this could be detectable (by portage?).
The bug #234422 is I believe a dupe of this one.