On 1/9/24 02:44, Shin'ichiro Kawasaki wrote:
Introduce a helper function _have_null_blk_feature which checks /sys/kernel/config/features. It allows test cases to adapt to null_blk feature support status. Signed-off-by: Shin'ichiro Kawasaki <shinichiro.kawasaki@xxxxxxx> --- common/null_blk | 15 +++++++++++++++ 1 file changed, 15 insertions(+) diff --git a/common/null_blk b/common/null_blk index 91b78d4..164125d 100644 --- a/common/null_blk +++ b/common/null_blk @@ -10,6 +10,21 @@ _have_null_blk() { _have_driver null_blk }+_have_null_blk_feature() {+ # Ensure that null_blk driver is built-in or loaded + if ! [[ -d /sys/module/null_blk ]]; then + if ! modprobe -q null_blk; then + return 1 + fi + if [[ ! "${MODULES_TO_UNLOAD[*]}" =~ null_blk ]]; then + MODULES_TO_UNLOAD+=(null_blk) + fi + fi + + # Check that null_blk has the specified feature + grep -qe "$1" /sys/kernel/config/nullb/features +} + _remove_null_blk_devices() { if [[ -d /sys/kernel/config/nullb ]]; then find /sys/kernel/config/nullb -mindepth 1 -maxdepth 1 \
Shouldn't _have_null_blk_feature() unload the null_blk kernel module if it loads that kernel module? That will allow to simplify the next patch in this series. Thanks, Bart.
