>>>> diff --git a/tests/xfs/group b/tests/xfs/group >>>> index 2319f58..fb74e82 100644 >>>> --- a/tests/xfs/group >>>> +++ b/tests/xfs/group >>>> @@ -445,3 +445,4 @@ >>>> 445 auto quick filestreams >>>> 446 auto quick >>>> 447 auto mount >>>> +448 auto >> >> >> Isn't this test quick? >> >> Thanks, >> Amir. > > > It actually takes about a minute and a half for me, though I guess it would > depend on how fast the host is. Most of the time goes into the loop of 100: > create a hard link, and then scan through all the entries to see if the new > parent pointer is there. This was mostly to make sure the attr tree was > expanding correctly back when I was trying to get the transaction rolling > straightened out. > That is still a bit surprising to me. I suspect the parent pointer parsing code that invokes awk so many times may be the culprit. We soemtimes use little C progs to assist in the test (see in src/) in cases where bash scripting is insufficient or inefficient. Perhaps if you use more bash array tricks instead of awk, parsing code will be faster? assuming that parsing code is the slow down factor? In any case, a "quick" test gets run more frequently by more people so it would be a big benefit IMO if you are able to squeeze that test runtime down to "quick". Thanks, Amir. -- To unsubscribe from this list: send the line "unsubscribe linux-xfs" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
