Here's what PERF shows (there's a bunch more lines, I just grabbed the top.
+ 48.50% 0.08% lt-openconnect [kernel.vmlinux] [k] system_call_fastpath
+ 19.57% 0.10% lt-openconnect [kernel.vmlinux] [k] sys_write
+ 19.37% 0.15% lt-openconnect [kernel.vmlinux] [k] vfs_write
+ 18.22% 0.00% lt-openconnect libpthread-2.17.so [.] __write_nocancel
+ 17.74% 0.08% lt-openconnect [kernel.vmlinux] [k] do_sync_write
+ 17.67% 0.08% lt-openconnect [tun] [k] tun_chr_aio_write
+ 17.52% 0.45% lt-openconnect [tun] [k] tun_get_user
+ 17.07% 0.13% lt-openconnect [kernel.vmlinux] [k] ip_local_out_sk
+ 16.92% 0.13% lt-openconnect [kernel.vmlinux] [k] ip_output
+ 16.72% 0.60% lt-openconnect [kernel.vmlinux] [k] ip_finish_output
+ 15.89% 0.08% lt-openconnect libpthread-2.17.so [.] __libc_send
+ 15.87% 0.03% lt-openconnect [kernel.vmlinux] [k] dev_queue_xmit
+ 15.82% 0.58% lt-openconnect [kernel.vmlinux] [k] __dev_queue_xmit
+ 15.69% 0.03% lt-openconnect [kernel.vmlinux] [k] sys_sendto
+ 15.59% 0.03% lt-openconnect [kernel.vmlinux] [k] SYSC_sendto
+ 15.47% 0.28% lt-openconnect [kernel.vmlinux] [k] sock_sendmsg
+ 15.24% 0.08% lt-openconnect [kernel.vmlinux] [k] netif_receive_skb
+ 15.17% 0.00% lt-openconnect [kernel.vmlinux] [k] netif_receive_skb_internal
+ 15.09% 0.08% lt-openconnect [kernel.vmlinux] [k] inet_sendmsg
+ 15.04% 0.38% lt-openconnect [kernel.vmlinux] [k] udp_sendmsg
+ 15.02% 0.20% lt-openconnect [kernel.vmlinux] [k] __netif_receive_skb_core
+ 15.02% 0.03% lt-openconnect [kernel.vmlinux] [k] __netif_receive_skb
+ 14.74% 0.10% lt-openconnect [kernel.vmlinux] [k] sch_direct_xmit
+ 14.71% 0.08% lt-openconnect [kernel.vmlinux] [k] ip_rcv
+ 14.49% 0.20% lt-openconnect [kernel.vmlinux] [k] ip_rcv_finish
+ 13.76% 0.05% lt-openconnect [kernel.vmlinux] [k] ip_local_deliver
+ 13.66% 0.03% lt-openconnect [kernel.vmlinux] [k] ip_local_deliver_finish
+ 13.49% 0.23% lt-openconnect [kernel.vmlinux] [k] tcp_v4_rcv
+ 13.46% 0.00% lt-openconnect [unknown] [k] 0000000000000000
+ 12.76% 0.10% lt-openconnect [kernel.vmlinux] [k] tcp_v4_do_rcv
+ 12.56% 0.25% lt-openconnect [kernel.vmlinux] [k] tcp_rcv_established
+ 11.76% 0.20% lt-openconnect [kernel.vmlinux] [k] udp_send_skb
+ 11.51% 0.03% lt-openconnect [kernel.vmlinux] [k] ip_send_skb
+ 10.49% 0.10% lt-openconnect [kernel.vmlinux] [k] dev_hard_start_xmit
+ 10.06% 9.06% lt-openconnect [vmxnet3] [k] vmxnet3_xmit_frame
+ 7.23% 0.03% lt-openconnect [kernel.vmlinux] [k] __tcp_push_pending_frames
+ 7.03% 0.28% lt-openconnect [kernel.vmlinux] [k] tcp_write_xmit
+ 6.46% 0.23% lt-openconnect [kernel.vmlinux] [k] tcp_transmit_skb
+ 6.36% 0.35% lt-openconnect libc-2.17.so [.] __select
+ 5.88% 0.28% lt-openconnect [kernel.vmlinux] [k] ip_queue_xmit
+ 5.86% 0.23% lt-openconnect [kernel.vmlinux] [k] sys_select
+ 5.46% 0.28% lt-openconnect [kernel.vmlinux] [k] core_sys_select
+ 5.36% 0.95% lt-openconnect [kernel.vmlinux] [k] do_select
+ 4.90% 0.05% lt-openconnect [kernel.vmlinux] [k] neigh_direct_output
4.63% 4.63% lt-openconnect libgnutls.so.28.43.0 [.] 0x00000000000c971d
+ 4.13% 0.05% lt-openconnect [kernel.vmlinux] [k] validate_xmit_skb_list
+ 4.05% 0.10% lt-openconnect [kernel.vmlinux] [k] validate_xmit_skb.isra.102.part.103
+ 3.85% 1.13% lt-openconnect [kernel.vmlinux] [k] tcp_ack
+ 3.53% 0.03% lt-openconnect [kernel.vmlinux] [k] neigh_resolve_output
+ 3.45% 0.20% lt-openconnect [kernel.vmlinux] [k] sys_read
+ 3.18% 0.03% lt-openconnect [kernel.vmlinux] [k] __skb_gso_segment
+ 3.13% 0.00% lt-openconnect [kernel.vmlinux] [k] skb_mac_gso_segment
+ 3.13% 0.08% lt-openconnect libpthread-2.17.so [.] __read_nocancel
+ 3.13% 0.20% lt-openconnect [kernel.vmlinux] [k] inet_gso_segment
+ 3.03% 0.20% lt-openconnect [kernel.vmlinux] [k] vfs_read
+ 2.93% 0.03% lt-openconnect [kernel.vmlinux] [k] tcp4_gso_segment
+ 2.90% 0.08% lt-openconnect [kernel.vmlinux] [k] tcp_gso_segment
+ 2.83% 0.03% lt-openconnect [kernel.vmlinux] [k] poll_schedule_timeout
+ 2.80% 0.05% lt-openconnect [kernel.vmlinux] [k] schedule_hrtimeout_range
+ 2.80% 2.80% lt-openconnect [kernel.vmlinux] [k] _raw_spin_unlock_irqrestore
+ 2.78% 0.38% lt-openconnect libpthread-2.17.so [.] __libc_recv
+ 2.75% 2.45% lt-openconnect libc-2.17.so [.] _int_malloc
+ 2.73% 2.73% lt-openconnect libnettle.so.4.7 [.] _nettle_sha1_compress
-----Original Message-----
From: Nikos Mavrogiannopoulos [mailto:n.mavrogiannopoulos@xxxxxxxxx]
Sent: Sunday, March 10, 2019 12:12 AM
To: Phillips, Tony; David Woodhouse
Cc: openconnect-devel@xxxxxxxxxxxxxxxxxxx
Subject: Re: [EXTERNAL] Re: What throughput is reasonable?
On Fri, 2019-03-08 at 20:26 +0000, Phillips, Tony wrote:
> Resending because the listserver bounced it due to it being HTML
> email.
>
> > > Tony, what's the output of gnutls-cli --benchmark-tls-ciphers on
> > > that platform?
>
> Testing throughput in cipher/MAC combinations (payload: 1400 bytes)
> RSA_ARCFOUR_128_SHA1 94.63 MB/sec
> RSA_ARCFOUR_128_MD5 96.69 MB/sec
> RSA_AES_128_GCM_SHA256 0.66 GB/sec
> RSA_AES_128_CBC_SHA1 146.48 MB/sec
> RSA_CAMELLIA_128_CBC_SHA1 48.80 MB/sec
>
> Testing throughput in cipher/MAC combinations (payload: 15360 bytes)
> RSA_ARCFOUR_128_SHA1 121.17 MB/sec
> RSA_ARCFOUR_128_MD5 106.72 MB/sec
> RSA_AES_128_GCM_SHA256 1.05 GB/sec
> RSA_AES_128_CBC_SHA1 196.41 MB/sec
> RSA_CAMELLIA_128_CBC_SHA1 53.76 MB/sec
The throughput is still much larger than what you see over the VPN.
Using perf is the best option at this point to see whether that's an
issue in openconnect client. You can start it as "perf record -g
openconnect"
regards,
Nikos
_______________________________________________
openconnect-devel mailing list
openconnect-devel@xxxxxxxxxxxxxxxxxxx
http://lists.infradead.org/mailman/listinfo/openconnect-devel
