# Functions sourced via testing framework. getlimits_() { eval $(getlimits) test "$INT_MAX" || error_ "Error running getlimits" } require_acl_() { getfacl --version < /dev/null > /dev/null 2>&1 \ && setfacl --version < /dev/null > /dev/null 2>&1 \ || skip_ "This test requires getfacl and setfacl." id -u bin > /dev/null 2>&1 \ || skip_ "This test requires a local user named bin." } require_hfs_() { mkfs.hfs 2>&1 | grep '^usage:' \ || skip_ "mkfs.hfs: command not found" } require_fat_() { mkfs.vfat 2>&1 | grep '^Usage:' \ || skip_ "mkfs.vfat: command not found" } # Skip this test if we're not in SELinux "enforcing" mode. require_selinux_enforcing_() { test "$(getenforce)" = Enforcing \ || skip_ "This test is useful only with SELinux in Enforcing mode." } require_openat_support_() { # Skip this test if your system has neither the openat-style functions # nor /proc/self/fd support with which to emulate them. test -z "$CONFIG_HEADER" \ && skip_ 'internal error: CONFIG_HEADER not defined' _skip=yes grep '^#define HAVE_OPENAT' "$CONFIG_HEADER" > /dev/null && _skip=no test -d /proc/self/fd && _skip=no if test $_skip = yes; then skip_ 'this system lacks openat support' fi } require_ulimit_() { ulimit_works=yes # Expect to be able to exec a program in 10MB of virtual memory, # but not in 20KB. I chose "date". It must not be a shell built-in # function, so you can't use echo, printf, true, etc. # Of course, in coreutils, I could use $top_builddir/src/true, # but this should be able to work for other projects, too. ( ulimit -v 10000; date ) > /dev/null 2>&1 || ulimit_works=no ( ulimit -v 20; date ) > /dev/null 2>&1 && ulimit_works=no test $ulimit_works = no \ && skip_ "this shell lacks ulimit support" } require_readable_root_() { test -r / || skip_ "/ is not readable" } # Skip the current test if strace is not available or doesn't work # with the named syscall. Usage: require_strace_ unlink require_strace_() { test $# = 1 || framework_failure strace -V < /dev/null > /dev/null 2>&1 || skip_ 'no strace program' strace -qe "$1" echo > /dev/null 2>&1 || skip_ 'strace -qe "'"$1"'" does not work' } # Require a controlling input `terminal'. require_controlling_input_terminal_() { tty -s || have_input_tty=no test -t 0 || have_input_tty=no if test "$have_input_tty" = no; then skip_ 'requires controlling input terminal This test must have a controlling input "terminal", so it may not be run via "batch", "at", or "ssh". On some systems, it may not even be run in the background.' fi } uid_is_privileged_() { # Make sure id -u succeeds. my_uid=$(id -u) \ || { echo "$0: cannot run \`id -u'" 1>&2; return 1; } # Make sure it gives valid output. case $my_uid in 0) ;; *[!0-9]*) echo "$0: invalid output (\`$my_uid') from \`id -u'" 1>&2 return 1 ;; *) return 1 ;; esac } get_process_status_() { sed -n '/^State:[ ]*\([[:alpha:]]\).*/s//\1/p' /proc/$1/status } # Convert an ls-style permission string, like drwxr----x and -rw-r-x-wx # to the equivalent chmod --mode (-m) argument, (=,u=rwx,g=r,o=x and # =,u=rw,g=rx,o=wx). Ignore ACLs. rwx_to_mode_() { case $# in 1) rwx=$1;; *) echo "$0: wrong number of arguments" 1>&2 echo "Usage: $0 ls-style-mode-string" 1>&2 return;; esac case $rwx in [ld-][rwx-][rwx-][rwxsS-][rwx-][rwx-][rwxsS-][rwx-][rwx-][rwxtT-]) ;; [ld-][rwx-][rwx-][rwxsS-][rwx-][rwx-][rwxsS-][rwx-][rwx-][rwxtT-][+.]) ;; *) echo "$0: invalid mode string: $rwx" 1>&2; return;; esac # Perform these conversions: # S s # s xs # T t # t xt # The `T' and `t' ones are only valid for `other'. s='s/S/@/;s/s/x@/;s/@/s/' t='s/T/@/;s/t/x@/;s/@/t/' u=`echo $rwx|sed 's/^.\(...\).*/,u=\1/;s/-//g;s/^,u=$//;'$s` g=`echo $rwx|sed 's/^....\(...\).*/,g=\1/;s/-//g;s/^,g=$//;'$s` o=`echo $rwx|sed 's/^.......\(...\).*/,o=\1/;s/-//g;s/^,o=$//;'$s';'$t` echo "=$u$g$o" } require_selinux_() { case `ls -Zd .` in '? .'|'unlabeled .') skip_ "this system (or maybe just" \ "the current file system) lacks SELinux support" ;; esac } very_expensive_() { if test "$RUN_VERY_EXPENSIVE_TESTS" != yes; then skip_ 'very expensive: disabled by default This test is very expensive, so it is disabled by default. To run it anyway, rerun make check with the RUN_VERY_EXPENSIVE_TESTS environment variable set to yes. E.g., env RUN_VERY_EXPENSIVE_TESTS=yes make check ' fi } expensive_() { if test "$RUN_EXPENSIVE_TESTS" != yes; then skip_ 'expensive: disabled by default This test is relatively expensive, so it is disabled by default. To run it anyway, rerun make check with the RUN_EXPENSIVE_TESTS environment variable set to yes. E.g., env RUN_EXPENSIVE_TESTS=yes make check ' fi } require_root_() { uid_is_privileged_ || skip_ "must be run as root" NON_ROOT_USERNAME=${NON_ROOT_USERNAME=nobody} NON_ROOT_GROUP=${NON_ROOT_GROUP=$(id -g $NON_ROOT_USERNAME)} } skip_if_root_() { uid_is_privileged_ && skip_ "must be run as non-root"; } error_() { echo "$0: $@" 1>&2; Exit 1; } framework_failure() { error_ 'failure in testing framework'; } # Set `groups' to a space-separated list of at least two groups # of which the user is a member. require_membership_in_two_groups_() { test $# = 0 || framework_failure groups=${COREUTILS_GROUPS-`(id -G || /usr/xpg4/bin/id -G) 2>/dev/null`} case "$groups" in *' '*) ;; *) skip_ 'requires membership in two groups this test requires that you be a member of more than one group, but running `id -G'\'' either failed or found just one. If you really are a member of at least two groups, then rerun this test with COREUTILS_GROUPS set in your environment to the space-separated list of group names or numbers. E.g., env COREUTILS_GROUPS='users cdrom' make check ' ;; esac } # Is /proc/$PID/status supported? require_proc_pid_status_() { sleep 2 & local pid=$! sleep .5 grep '^State:[ ]*[S]' /proc/$pid/status > /dev/null 2>&1 || skip_ "/proc/$pid/status: missing or 'different'" kill $pid } # Does the current (working-dir) file system support sparse files? require_sparse_support_() { test $# = 0 || framework_failure # Test whether we can create a sparse file. # For example, on Darwin6.5 with a file system of type hfs, it's not possible. # NTFS requires 128K before a hole appears in a sparse file. t=sparse.$$ dd bs=1 seek=128K of=$t < /dev/null 2> /dev/null set x `du -sk $t` kb_size=$2 rm -f $t if test $kb_size -ge 128; then skip_ 'this file system does not support sparse files' fi } mkfifo_or_skip_() { test $# = 1 || framework_failure if ! mkfifo "$1"; then # Make an exception of this case -- usually we interpret framework-creation # failure as a test failure. However, in this case, when running on a SunOS # system using a disk NFS mounted from OpenBSD, the above fails like this: # mkfifo: cannot make fifo `fifo-10558': Not owner skip_ 'NOTICE: unable to create test prerequisites' fi } # Disable the current test if the working directory seems to have # the setgid bit set. skip_if_setgid_() { setgid_tmpdir=setgid-$$ (umask 77; mkdir $setgid_tmpdir) perms=$(stat --printf %A $setgid_tmpdir) rmdir $setgid_tmpdir case $perms in drwx------);; drwxr-xr-x);; # Windows98 + DJGPP 2.03 *) skip_ 'this directory has the setgid bit set';; esac } skip_if_mcstransd_is_running_() { test $# = 0 || framework_failure # When mcstransd is running, you'll see only the 3-component # version of file-system context strings. Detect that, # and if it's running, skip this test. __ctx=$(stat --printf='%C\n' .) || framework_failure case $__ctx in *:*:*:*) ;; # four components is ok *) # anything else probably means mcstransd is running skip_ "unexpected context '$__ctx'; turn off mcstransd" ;; esac } # Skip the current test if umask doesn't work as usual. # This test should be run in the temporary directory that ends # up being removed via the trap commands. working_umask_or_skip_() { umask 022 touch file1 file2 chmod 644 file2 perms=`ls -l file1 file2 | sed 's/ .*//' | uniq` rm -f file1 file2 case $perms in *' '*) skip_ 'your build directory has unusual umask semantics' esac } emit_superuser_warning() { uid=`id -u` || uid=1 test "$uid" != 0 && echo 'WARNING: You are not superuser. Watch out for permissions.' || : } require_mdadm_() { mdadm --version || skip_ "find mdadm executable" } # Will look for an md number that is not in use and create a md device with # that number. If the system has more than 9 md devices, it will fail. mdadm_create_linear_device_() { lo_dev=$1 mdd=$G_dev_/md0 for i in 0 1 2 3 4 5 6 7 8 9 ; do mdd=$G_dev_/md$i mdadm --create --force $mdd --level=linear --raid-devices=1 $lo_dev \ > /dev/null 2>&1 \ && break if [ $i -eq 9 ]; then echo $mdd ; return 1 ; fi done echo $mdd return 0 } # Often, when parted cannot use the specified size or start/endpoints # of a partition, it outputs a warning or error like this: # # Error: You requested a partition from 512B to 50.7kB (...). # The closest location we can manage is 17.4kB to 33.8kB (...). # # But those numbers depend on sector size, so # replace the specific values with place-holders, # so tests do not depend on sector size. normalize_part_diag_() { local file=$1 sed 's/ [0-9.k]*B to [0-9.k]*B (sectors .*$/ X to Y./' $file > $file.t \ && mv $file.t $file && return 0 return 1 } require_xfs_() { mkfs.xfs -V || skip_ "this test requires XFS support" } require_dvhtool_() { dvhtool --help \ || skip_ 'dvhtool is required for this test' } # Helper function: wait 2s (via .1s increments) for FILE to appear. # Usage: wait_for_dev_to_appear_ /dev/sdg # Return 0 upon success, 1 upon failure. wait_for_dev_to_appear_() { local file=$1 local i=0 local incr=1 while :; do udevadm settle ls "$file" > /dev/null 2>&1 && return 0 sleep .1 2>/dev/null || { sleep 1; incr=10; } i=$(expr $i + $incr); test $i = 20 && break done return 1 } # Like the above, but don't hard-code the max timeout. wait_for_dev_to_disappear_() { local file=$1 local n_sec=$2 local i=0 local incr=1 while :; do udevadm settle ls "$file" > /dev/null 2>&1 || return 0 sleep .1 2>/dev/null || { sleep 1; incr=10; } i=$(expr $i + $incr); test $i -ge $(expr $n_sec \* 10) && break done return 1 } device_mapper_required_() { . "$abs_top_srcdir/tests/t-lvm.sh" lvm_init_root_dir_ || fail_ "device mapper setup failed" } # Require a 64bit system require_64bit_() { case $(uname -m) in aarch64|mips64|ppc64|x86_64) return 0;; *) skip_ "This test requires a 64 bit system" ;; esac } # Check if the specified filesystem is either built into the kernel, or can be loaded # as a module # Usage: has_filesystem vfat # Ruturns 0 if the filesystem is available, otherwise skips the test require_filesystem_() { grep $1 /proc/filesystems >/dev/null && return 0 modprobe --quiet --dry-run $1 || skip_ "this test requires kernel support for $1" }