Opened 2 years ago
Closed 2 years ago
#66273 closed defect (worksforme)
gnutls @3.6.16_4 does not build on PPC Tiger, Mac OS X 10.4.11, because a configure script hangs in an endless loop?
Reported by: | ballapete (Peter "Pete" Dyballa) | Owned by: | |
---|---|---|---|
Priority: | Normal | Milestone: | |
Component: | ports | Version: | 2.8.0 |
Keywords: | tiger ppc | Cc: | mps@… |
Port: | gnutls |
Description
I am building in GNU Emacs
. The configure
reaches/outputs the last line of text:
checking for CMOCKA... no checking for LIBIDN2... yes checking for nettle_get_secp_192r1 in -lhogweed... yes checking for nettle_rsa_sec_decrypt... yes checking for nettle_cfb8_encrypt... yes checking for union nettle_block8... yes checking for nettle_cmac128_update... yes checking for nettle_xts_encrypt_message... yes checking for nettle_gosthash94cp_update... yes checking for nettle_gost28147_set_key... no checking for nettle_streebog512_update... yes checking for nettle_curve448_mul... yes checking for nettle_ed448_shake256_sign... yes checking for nettle_get_gost_gc256b... yes checking for nettle_get_gost_gc512a... yes checking for nettle_gostdsa_sign... yes checking for nettle_chacha_set_counter... yes checking for nettle_siv_cmac_aes128_set_key... yes
After some time my PowerBook G4
starts to activate the fans, which usually does not happen. Then I send a break signal
to GNU Emacs to interrupt compilation. The output then is (excerpt):
checking gmp soname... Command failed: cd "/opt/local/var/macports/build/_opt_local_var_macports_sources_nue.de.rsync.macports.org_macports_release_tarballs_ports_devel_gnutls/gnutls/work/gnutls-3.6.16" && ./configure --prefix=/opt/local --disable-hardware-acceleration --disable-guile --disable-silent-rules --disable-static --enable-local-libopts --enable-openssl-compatibility --with-p11-kit --with-system-priority-file="/opt/local/etc/gnutls/default-priorities" --with-default-trust-store-pkcs11=pkcs11: ac_cv_prog_AWK=/usr/bin/awk --host=powerpc-apple-darwin8 --build=powerpc-apple-darwin8 --with-unbound-root-key-file="/opt/local/var/run/unbound/root.key" Exit code: 1
Attachments (1)
Change History (6)
Changed 2 years ago by ballapete (Peter "Pete" Dyballa)
comment:1 Changed 2 years ago by ballapete (Peter "Pete" Dyballa)
The last tests in config.log
are:
configure:66234: $? = 0 configure:66234: result: yes configure:66258: checking for nettle_siv_cmac_aes128_set_key configure:66258: /opt/local/bin/gcc-mp-7 -o conftest -pipe -Os -I/opt/local/include/LegacySupport -arch ppc -I/opt/local/include -D__WORDSIZE=32 -I/opt/local/include/LegacySupport -L/opt/local/lib -Wl,-headerpad_max_install_names -lMacportsLegacySupport -arch ppc conftest.c -L/opt/local/lib -lnettle >&5 configure:66258: $? = 0 configure:66258: result: yes configure:66280: checking gmp soname configure:66296: /opt/local/bin/gcc-mp-7 -o conftest -pipe -Os -I/opt/local/include/LegacySupport -arch ppc -I/opt/local/include -D__WORDSIZE=32 -I/opt/local/include/LegacySupport -L/opt/local/lib -Wl,-headerpad_max_install_names -lMacportsLegacySupport -arch ppc conftest.c -lgmp >&5 configure:66296: $? = 0
then follows the section of:
## ---------------- ## ## Cache variables. ## ## ---------------- ##
Presumingly these are the corresponding lines in the configure
script:
66298 # Check for SIV-CMAC 66299 save_LIBS=$LIBS 66300 LIBS="$LIBS $NETTLE_LIBS" 66301 for ac_func in nettle_siv_cmac_aes128_set_key 66302 do : 66303 ac_fn_c_check_func "$LINENO" "nettle_siv_cmac_aes128_set_key" "ac_cv_func_nettle_siv_cmac_aes128_set_key" 66304 if test "x$ac_cv_func_nettle_siv_cmac_aes128_set_key" = xyes; then : 66305 cat >>confdefs.h <<_ACEOF 66306 #define HAVE_NETTLE_SIV_CMAC_AES128_SET_KEY 1 66307 _ACEOF 66308 66309 fi 66310 done 66311 66312 LIBS=$save_LIBS 66313 if test "$ac_cv_func_nettle_siv_cmac_aes128_set_key" != "yes"; then 66314 NEED_SIV_TRUE= 66315 NEED_SIV_FALSE='#' 66316 else 66317 NEED_SIV_TRUE='#' 66318 NEED_SIV_FALSE= 66319 fi 66320 66321 66322 # Check sonames of the linked libraries needed for FIPS selftests. 66323 save_LIBS=$LIBS 66324 LIBS="$LIBS $GMP_LIBS" 66325 { $as_echo "$as_me:${as_lineno-$LINENO}: checking gmp soname" >&5 66326 $as_echo_n "checking gmp soname... " >&6; } 66327 cat confdefs.h - <<_ACEOF >conftest.$ac_ext 66328 /* end confdefs.h. */ 66329 66330 #include <gmp.h> 66331 int 66332 main () 66333 { 66334 66335 mpz_t n; 66336 mpz_init(n); 66337 ; 66338 return 0; 66339 } 66340 _ACEOF 66341 if ac_fn_c_try_link "$LINENO"; then : 66342 gmp_so=`(eval "$LDDPROG conftest$EXEEXT $LDDPOSTPROC") | grep '^libgmp\.so'` 66343 else 66344 gmp_so=none 66345 fi 66346 rm -f core conftest.err conftest.$ac_objext \ 66347 conftest$ac_exeext conftest.$ac_ext 66348 if test -z "$gmp_so"; then 66349 gmp_so=none 66350 fi 66351 { $as_echo "$as_me:${as_lineno-$LINENO}: result: $gmp_so" >&5 66352 $as_echo "$gmp_so" >&6; } 66353 66354 cat >>confdefs.h <<_ACEOF 66355 #define GMP_LIBRARY_SONAME "$gmp_so" 66356 _ACEOF
The problem seems to be that configure
checks for .so
files. On Mac OS X they use the extension .dylib
…
comment:2 Changed 2 years ago by ballapete (Peter "Pete" Dyballa)
This morning my PowerBook G4
did not want to present the login window when opening the lid. So it had to be restarted. And now while rev-upgrade
ing from upgrading icu
the configure script
has no problem…
The build process has almost finished.
comment:3 Changed 2 years ago by kencu (Ken)
I am presently trying to see if I can reproduce this on my Tiger system...
comment:4 Changed 2 years ago by ballapete (Peter "Pete" Dyballa)
After a reboot it was gone… Before I could reproduce it a few times on different days.
comment:5 Changed 2 years ago by kencu (Ken)
Resolution: | → worksforme |
---|---|
Status: | new → closed |
Yes, mine installs fine without any trouble:
$ port -v installed gnutls The following ports are currently installed: gnutls @3.6.16_4 (active) requested_variants='' platform='darwin 8' archs='ppc' date='2022-11-30T16:49:43-0800'
Main.log from PPC Tiger, Mac OS X 10.4.11