freebsd-dev/sys/netipsec
George V. Neville-Neil 26882b4239 Turning on IPSEC used to introduce a slight amount of performance
degradation (7%) for host host TCP connections over 10Gbps links,
even when there were no secuirty policies in place. There is no
change in performance on 1Gbps network links. Testing GENERIC vs.
GENERIC-NOIPSEC vs. GENERIC with this change shows that the new
code removes any overhead introduced by having IPSEC always in the
kernel.

Differential Revision:	D3993
MFC after:	1 month
Sponsored by:	Rubicon Communications (Netgate)
2015-10-27 00:42:15 +00:00
..
ah_var.h
ah.h
esp_var.h
esp.h Summary: Remove spurious, extra, next header comments. 2015-05-15 18:04:49 +00:00
ipcomp_var.h
ipcomp.h
ipsec6.h
ipsec_input.c IPSEC, remove variable argument function its already due. 2015-07-21 21:46:24 +00:00
ipsec_mbuf.c
ipsec_output.c Take extra reference to security policy before calling crypto_dispatch(). 2015-09-30 08:16:33 +00:00
ipsec.c Turning on IPSEC used to introduce a slight amount of performance 2015-10-27 00:42:15 +00:00
ipsec.h IPSEC, remove variable argument function its already due. 2015-07-21 21:46:24 +00:00
key_debug.c Make IPsec work with AES-GCM and AES-ICM (aka CTR) in OCF... IPsec 2015-08-04 17:47:11 +00:00
key_debug.h
key_var.h
key.c Make IPsec work with AES-GCM and AES-ICM (aka CTR) in OCF... IPsec 2015-08-04 17:47:11 +00:00
key.h drop key_sa_stir_iv as it isn't used... 2015-06-11 13:05:37 +00:00
keydb.h Make IPsec work with AES-GCM and AES-ICM (aka CTR) in OCF... IPsec 2015-08-04 17:47:11 +00:00
keysock.c Remove unneded mbuf length adjustment, M_PREPEND() already did that. 2015-05-19 17:14:27 +00:00
keysock.h
xform_ah.c Take extra reference to security policy before calling crypto_dispatch(). 2015-09-30 08:16:33 +00:00
xform_esp.c Take extra reference to security policy before calling crypto_dispatch(). 2015-09-30 08:16:33 +00:00
xform_ipcomp.c Take extra reference to security policy before calling crypto_dispatch(). 2015-09-30 08:16:33 +00:00
xform_tcp.c
xform.h RFC4868 section 2.3 requires that the output be half... This fixes 2015-07-29 07:15:16 +00:00