Bill Long writes:
>> My understanding from MFE sec. 19.13.6 and FDIS (10-007r1) sec 8.5.7
>> par 2 is that the correct behaviour of this program is:
>>
>> 1. Image 1 initiates normal termination 2. At "sync all
>> (stat=errstat)" errstat
>> becomes defined with "stat_stopped_image"
>> on all images, but image 1.
>> 3. Hence all images, but image 1, must output "mumu".
> Yes, this is what should happen. At least in the case of the Cray
compiler, there is already a bug submitted about this issue.
There is also a bug submitted about this for Intel - it was reported by John
Reid back in August. We have done some preliminary work on this for our next
major release and the behavior is now a bit different in that most of the
images do output "mumu" but some still hang, probably for the reason Bill
notes as the "implementation challenge". I have passed this information on
to the development team.
Steve Lionel
Intel Developer Support
Merrimack, NH
|