Xavier de Gaye added the comment: With configure run as './configure CC="$(CC)" ...' and when CC is set to: CC = clang --sysroot=$(SYSROOT) -target $(TARGET) -gcc-toolchain $(GCC_TOOLCHAIN)
and building for armv7 on android-21, then configure runs $CPP as: checking for CPP... clang --sysroot=/opt/android-ndk/platforms/android-21/arch-arm -target armv7-none-linux-androideabi -gcc-toolchain /opt/android-ndk/toolchains/arm-linux-androideabi-4.9/prebuilt/linux-x86_64 -E as can be seen by inserting in configure.ac: AC_MSG_CHECKING([for CPP]) AC_MSG_RESULT([$CPP]) and the shared libraries are named 'module_name.cpython-36m-arm-linux-gnueabi.so'. OTOH, the 'configure' script runs its own preprocessings with the following idiom: if (eval "$ac_cpp conftest.$ac_ext") 2>&5 | $EGREP "memchr" >/dev/null 2>&1; then : where 'ac_cpp' is set to ac_cpp='$CPP $CPPFLAGS'. So the proposed patch LGTM. PLATFORM_TRIPLETs are specific to multiarch systems such as debian and should not be relied upon on Android, but they should not be entirely wrong of course. ---------- nosy: +doko _______________________________________ Python tracker <rep...@bugs.python.org> <http://bugs.python.org/issue27453> _______________________________________ _______________________________________________ Python-bugs-list mailing list Unsubscribe: https://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com
