> -----Original Message----- > From: Morten Brørup > Sent: Sunday, September 15, 2019 11:07 AM
[...] > [...] (Except that the use of likely/unlikely > in the mbuf library heavily favors single-segment mbufs, so such an > optimization would go against this favoritism.) > Ignore that statement. I got two things mixed up... Single-reference mbufs are favored, not single-segment mbufs. Med venlig hilsen / kind regards - Morten Brørup
