On aarch64, we're experiencing a test failure in tests/misc/printenv.sh on the openSUSE Build Service: the output of 'env' is sorted exactly in reverse order compared to that of 'printenv'.
https://build.opensuse.org/package/live_build_log/Base:System/coreutils-testsuite/openSUSE_Factory_ARM/aarch64 However, if I apply the following patch, then the test would succeed: https://build.opensuse.org/package/live_build_log/home:bernhard-voelker:aarch64/coreutils-testsuite/openSUSE_Factory_ARM/aarch64 --- a/tests/misc/printenv.sh +++ b/tests/misc/printenv.sh @@ -23,7 +23,7 @@ print_ver_ printenv # printenv as a builtin, so we must invoke it via "env". # But beware of $_, set by many shells to the last command run. # Also, filter out LD_PRELOAD, which is set when running under valgrind. -env | grep -Ev '^(_|LD_PRELOAD=)' > exp || framework_failure_ +env -- env | grep -Ev '^(_|LD_PRELOAD=)' > exp || framework_failure_ env -- printenv | grep -Ev '^(_|LD_PRELOAD=)' > out || fail=1 compare exp out || fail=1 I'm not 100% sure what's happening here. According to 'which' and 'type', both 'env' and 'printenv' are called correctly from src/: [ 1718s] + which env [ 1718s] /home/abuild/rpmbuild/BUILD/coreutils-8.23/src/env [ 1718s] + which printenv [ 1718s] /home/abuild/rpmbuild/BUILD/coreutils-8.23/src/printenv [ 1718s] + type env [ 1718s] env is hashed (/home/abuild/rpmbuild/BUILD/coreutils-8.23/./src/env) [ 1718s] + type printenv [ 1718s] printenv is /home/abuild/rpmbuild/BUILD/coreutils-8.23/./src/printenv Any idea what's happening here? I could prepare a proper patch for the above diff, but I'd prefer to understand why this would make a difference. Did I miss the obvious? Thanks & have a nice day, Berny
