Am 28.11.2014 um 13:42 schrieb Stefan Hajnoczi:
> On Thu, Nov 27, 2014 at 11:27:04AM +0100, Peter Lieven wrote:
>> This reverts commit ac2662a913ee5854b1269256adbdc14e57ba480a.
> Justification?
>
> Won't we hit the same problem as the global pool: either pool size is
> too large or too small.  If you exceed the pool size performance drops
> dramatically.

Each thread has its own pool. But you can ignore this series I think.
The idea Paolo came up with seems to be better. With
my last improvement it seems to have equal performance without
the drawback of different threads creating and destroying the coroutines.

Peter

Reply via email to