On Thu, 2019-04-18 at 21:16 +0000, Phillips, Tony wrote: > Also of note is that when you do your tests, you’re getting CPU/bound > at 100%. I never paid much attention to overall CPU until I saw your > stats last week. > > We can’t get OC to run over maybe 35%. > > This may boil down to how we have vSphere configured (or VMWare in > general versus AWS’s hypervisor). Hm, it's only single-threaded but it's supposed to spend *all* of its time shovelling packets out the interface, never sleeping when it has work to do. Where's the rest of the CPU time going? Is that 35% only counting userspace? Is there a lot of steal time spent in the hypervisor? Is the hypervisor actually *limiting* the amount of time a given vCPU can have? Seeing another run of (sudo perf report -g `pidof lt-openconnect`) which starts the moment after you start an iperf UDP test, and finished a moment before iperf finishes, might be enlightening.
Attachment:
smime.p7s
Description: S/MIME cryptographic signature
_______________________________________________ openconnect-devel mailing list openconnect-devel@xxxxxxxxxxxxxxxxxxx http://lists.infradead.org/mailman/listinfo/openconnect-devel
