Thanks Daniel,
I will await your response on the issue then, regarding using the lock or atomic integers. I presume the atomic integer usage must be faster than the locking approach.
FWIW, a similar locking is being used in 'dp_init_relative_weights()' and 'ds_update_latency' APIs in dispatch.c also.

—
You are receiving this because you are subscribed to this thread.
Reply to this email directly, view it on GitHub, or unsubscribe.