[PATCH blktests 01/11] check: factor out _run_test()

[Shinichiro Kawasaki]

On Apr 11, 2024 / 19:04, Nitesh Shetty wrote:
> On 11/04/24 08:12PM, Shin'ichiro Kawasaki wrote:
> > The function _run_test() is rather complex and has deep nests. Before
> > modifying it for repeated test case runs, simplify it. Factor out some
> > part of the function to the new functions _check_and_call_test() and
> > _check_and_call_test_device().
> > 
> > Signed-off-by: Shin'ichiro Kawasaki <shinichiro.kawasaki at wdc.com>
> > ---
> > check | 90 +++++++++++++++++++++++++++++++++++------------------------
> > 1 file changed, 53 insertions(+), 37 deletions(-)
> > 
> > diff --git a/check b/check
> > index 55871b0..b1f5212 100755
> > --- a/check
> > +++ b/check
> > @@ -463,6 +463,56 @@ _unload_modules() {
> > 	unset MODULES_TO_UNLOAD
> > }
> > 
> > +_check_and_call_test() {
> 
> ret should be declared ret as local ?

Yes, will do so in v2.

> 
> > +	if declare -fF requires >/dev/null; then
> > +		requires
> > +	fi
> > +
> > +	RESULTS_DIR="$OUTPUT/nodev"
> > +	_call_test test
> > +	ret=$?
> > +	if (( RUN_ZONED_TESTS && CAN_BE_ZONED )); then
> > +		RESULTS_DIR="$OUTPUT/nodev_zoned"
> > +		RUN_FOR_ZONED=1
> > +		_call_test test
> > +		ret=$(( ret || $? ))
> > +	fi
> > +
> > +	return $ret
> > +}
> > +
> > +_check_and_call_test_device() {
> > +	local unset_skip_reason
> 
> Same here, ret should declared be local ?

Yes, and thanks for the review :)

> 
> Reviewed-by: Nitesh Shetty <nj.shetty at samsung.com>