Ugh, I'm to blame on the title -- the graph from which I drew the conclusion
has logarithmic x-axis and linear y-axis, so the increase looks exponential,
whereas it is in fact linear. However, the measurements are still accurate -
after a certain threshold (~ 2048 threads), the scheduling is no longer
constant-time, but becomes linear.
This message posted from opensolaris.org
_______________________________________________
opensolaris-code mailing list
[email protected]
http://mail.opensolaris.org/mailman/listinfo/opensolaris-code
