So yesterday I built -current successfully, ran into the
pthread_atfork() problem while testing the result, then before I went to
sleep I did another "git pull" because kre had committed a new patch for
that, and so I fired off another build, only to be greeted by the
following today:
/Volumes/work/woods/g-NetBSD-src/tools/gdb/../../external/gpl3/gdb/dist/gdb/cli/cli-cmds.c:928:14: error: 'vfork' is deprecated: Use posix_spawn or fork [-Werror,-Wdeprecated-declarations]
928 | if ((pid = vfork ()) == 0)
| ^
/Applications/Xcode.app/Contents/Developer/Platforms/MacOSX.platform/Developer/SDKs/MacOSX.sdk/usr/include/unistd.h:604:1: note: 'vfork' has been explicitly marked deprecated here
604 | __deprecated_msg("Use posix_spawn or fork")
| ^
/Applications/Xcode.app/Contents/Developer/Platforms/MacOSX.platform/Developer/SDKs/MacOSX.sdk/usr/include/sys/cdefs.h:218:48: note: expanded from macro '__deprecated_msg'
218 | #define __deprecated_msg(_msg) __attribute__((__deprecated__(_msg)))
| ^
CXX compile/compile-object-load.o
CXX compile/compile-object-run.o
1 error generated.
nbgmake[1]: *** [cli/cli-cmds.o] Error 1
nbgmake[1]: *** Waiting for unfinished jobs....
nbgmake: *** [install-gdb] Error 2
--- .install_done ---
*** Failed target: .install_done
*** In directory: /Volumes/work/woods/g-NetBSD-src/tools/gdb
*** Failed commands:
@(cd ${.OBJDIR}/build && ${BUILD_COMMAND} ${INSTALL_TARGET})
=> @(cd /Users/woods/build/woods/very.local/trunk-x86_64-amd64-obj/Volumes/work/woods/g-NetBSD-src/tools/gdb/build && /usr/bin/env -i AR=ar AWK=/Users/woods/build/woods/very.local/trunk-x86_64-amd64-tools/bin/nbawk CC=cc CFLAGS=-O\ -O2\ -O2\ -no-cpp-precomp CONFIG_SHELL=/bin/dash CPPFLAGS=\ -I/Users/woods/build/woods/very.local/trunk-x86_64-amd64-tools/include CXX=c++ CXXFLAGS=-O\ -O2\ -D__STDC_FORMAT_MACROS\ -D__STDC_LIMIT_MACROS\ -D__STDC_CONSTANT_MACROS INSTALL=/Users/woods/build/woods/very.local/trunk-x86_64-amd64-tools/bin/x86_64--netbsd-install\ -c\ -p\ -r LDFLAGS=-O LEX=/Users/woods/build/woods/very.local/trunk-x86_64-amd64-tools/bin/nblex FLEX=/Users/woods/build/woods/very.local/trunk-x86_64-amd64-tools/bin/nblex M4=/Users/woods/build/woods/very.local/trunk-x86_64-amd64-tools/bin/nbm4 MAKE=/Users/woods/build/woods/very.local/trunk-x86_64-amd64-tools/bin/nbgmake PATH="/Users/woods/build/woods/very.local/trunk-x86_64-amd64-tools/bin:$PATH" RANLIB=ranlib
YACC=/Users/woods/build/woods/very.local/trunk-x86_64-amd64-tools/bin/nbyacc am_cv_func_iconv=no MACHINE= BISON=true DESTDIR= INSTALL=/Users/woods/build/woods/very.local/trunk-x86_64-amd64-tools/bin/x86_64--netbsd-install\ -c\ -p\ -r V=0 /Users/woods/build/woods/very.local/trunk-x86_64-amd64-tools/bin/nbgmake -j 24 -e MACHINE= BISON=true DESTDIR= INSTALL=/Users/woods/build/woods/very.local/trunk-x86_64-amd64-tools/bin/x86_64--netbsd-install\ -c\ -p\ -r -s V=0 install-gdb)
@touch $@
=> @touch .install_done
*** [.install_done] Error code 2
nbmake[6]: stopped making "install" in /Volumes/work/woods/g-NetBSD-src/tools/gdb
I'm cross-building on macOS, for amd64.
I don't see anything relevant in the changes I pulled between builds,
and there shouldn't have been any changes to the macOS toolchain.
I'm not even sure why tools/gdb had to be rebuilt, nor why it is being
rebuilt during "make install-gdb". It's a massively parallel build so
reading the log is difficult. All I see that might be relevant is that
for some reason it wanted to re-run config.status:
install-gcc ===> tools (with: install-gdb)
install ===> tools/gdb
GEN config.status
GEN Makefile
GEN stamp-h
GEN jit-reader.h
GEN stamp-version
GEN init.c
GEN gdb-gdb.py
GEN gdb-gdb.gdb
GEN xml-builtin.c
GEN gcore
Maybe it's just stupidity in the gdb Makefile as when I manually run
"dependall" it does nothing, but "make install" wants to build things:
12:54 [4947] $ cd tools/gdb
12:55 [4948] $ mynbmake dependall
12:55 [4949] $ mynbmake install
CXX cli/cli-cmds.o
One thing I do know is that there might have been some incomplete
intermediate targets in the build tree from earlier build failures in an
unrelated part of the tree. The change Roland made to "make" the other
day didn't actually fix the problem for me (more in a separate email).
I guess I'll clean it out and start with a fresh new whole build again.
--
Greg A. Woods <gwoods%acm.org@localhost>
Kelowna, BC +1 250 762-7675 RoboHack <woods%robohack.ca@localhost>
Planix, Inc. <woods%planix.com@localhost> Avoncote Farms <woods%avoncote.ca@localhost>
Attachment:
pgpAypfACS0eE.pgp
Description: OpenPGP Digital Signature