On Tue, 2019-04-09 at 09:34 +0300, David Woodhouse wrote: > > It's interesting that Nikos's code does 50% better than mine. I'll > investigate that. It isn't that I'm using malloc() and free() for each packet. If I use a pool, a tiny amount of _int_malloc() disappears from the perf report but the overall performance remains the same. Probably that I need to pay attention to which file descriptors woke me from poll() and service only those which need it, rather than doing everything, every time.
Attachment:
smime.p7s
Description: S/MIME cryptographic signature
_______________________________________________ openconnect-devel mailing list openconnect-devel@xxxxxxxxxxxxxxxxxxx http://lists.infradead.org/mailman/listinfo/openconnect-devel
