On Fri, Nov 05, 2021 at 09:18:15AM +0100, Claudio Jeker wrote: > Noticed the other day. The ip addr arrays and as number array are > marshalled element by element which is not very efficent. > All the data is in one big blob of memory so just use the basic io > operations for a memory blob and ship the full array at once. > > This seems to reduce runtime by 5-10% (in my unscientific testing). > Also it makes the code a fair bit simpler.
OK job@
