Okay. Now we're on to something. After applying that patch (and the minor edit) It's now noticeably faster. I average now 18-20 MBytes/sec (as opposed to 1.9 MB/sec) I ran 10 of the dd tests back to back, and the syslog is littered with about 40,000 of the Requeueing failed ESP send: Resource temporarily unavailable There are now even more UDP send buffer errors and IP Packets Dropped (about 7-10x more per test??), but the TCP Segments Retransmitted dropped to 32 (compared to 1490 before) per test. -----Original Message----- From: David Woodhouse [mailto:dwmw2@xxxxxxxxxxxxx] Sent: Monday, March 25, 2019 11:33 AM To: Phillips, Tony; Daniel Lenski Cc: Nikos Mavrogiannopoulos; openconnect-devel@xxxxxxxxxxxxxxxxxxx Subject: Re: [EXTERNAL] Re: What throughput is reasonable? On Mon, 2019-03-25 at 16:25 +0000, David Woodhouse wrote: > On Mon, 2019-03-25 at 16:05 +0000, David Woodhouse wrote: > > /* XXX: Keep the packet somewhere? */ > > > Try this. Er, no, this. > diff --git a/esp.c b/esp.c > index 9b79c03a..e1a52112 100644 > --- a/esp.c > +++ b/esp.c [snip] this->len = len; [snip] _______________________________________________ openconnect-devel mailing list openconnect-devel@xxxxxxxxxxxxxxxxxxx http://lists.infradead.org/mailman/listinfo/openconnect-devel
