| Date | Mon, 22 Jun 2009 13:54:00 +0200 | From | Ingo Molnar <> | Subject | Re: I.7 - Group validity checking |
| |
> 7/ Group validity checking > > At the user level, an application is only concerned with events > and grouping of those events. The assignment logic is performed by > the kernel. > > For a group to be scheduled, all its events must be compatible > with each other, otherwise the group will never be scheduled. It > is not clear to me when that sanity check will be performed if I > create the group such that it is stopped.
The hardware implementation should verify that the group fits on the PMU for every new group member.
Like said before, we think this and I.2/ are in conflict.
> If the group goes all the way to scheduling, it will never be > scheduled. Counts will be zero and the users will have no idea > why. If the group is put in error state, read will not be > possible. But again, how will the user know why?
It should not be possible to create such a situation. Can you see it being possible?
|