svn commit: r357004 - in head/sys: kern sys

[Gleb Smirnoff]

On Fri, Jan 24, 2020 at 12:23:33PM -0800, Cy Schubert wrote:
C> > Let not the network epoch become the new Giant of EPOCH's. There might 
C> > be realtime constraints for EPOCH's aswell.
C> 
C> I also had that concern yesterday.
C> 
C> Obtaining an EPOCH higher up the call stack or lower down depends on the 
C> workload. Are there any measurements that moving the EPOCHs higher in the 
C> call stack improves one workload (and configuration) or another? Could one 
C> type of workload or configuration benefit from this at the expense of 
C> another workload or configuration?

Can you imagine a workload where calling epoch_enter/exit more often but
for a shorter period of execution would be beneficial that calling it
once for a packet?

-- 
Gleb Smirnoff