Merge ^/head r306906 through r307382.

- Revert r300854, r303657 which tried to fix regression from r297225.
- Fix the regression proper way using RO_RTFREE().

Submitted by: ae

Merge ^/head r305687 through r305890.

Remove the 4.3BSD compatible macro m_copy(), use m_copym() instead.

Reviewed by: gnn
Differential Revision: https://reviews.freebsd.org/D7878

Merge ^/head r304700 through r304884.

MFhead @ r304815

Fix L2 caching for UDP over IPv6

ip6_output() was missing cache invalidation code analougous to
ip_output.c. r304545 disabled L2 caching for UDP/IPv6 as a workaround.
This change adds the missing ca

Fix L2 caching for UDP over IPv6

ip6_output() was missing cache invalidation code analougous to
ip_output.c. r304545 disabled L2 caching for UDP/IPv6 as a workaround.
This change adds the missing cache invalidation code and reverts
r304545.

Reviewed by: gnn
Approved by: gnn (mentor)
Tested by: peter@, Mike Andrews
MFC after: 3 weeks
Differential Revision: https://reviews.freebsd.org/D7591

show more ...

Merge ^/head r303250 through r304235.

Fix NULL pointer dereference.
ro pointer can be NULL when IPSec consumes mbuf.

PR: 211486
MFC after: 3 days

Rework IPV6 TCP path MTU discovery to match IPv4

- Re-write tcp_ctlinput6() to closely mimic the IPv4 tcp_ctlinput()

- Now that tcp_ctlinput6() updates t_maxseg, we can allow ip6_output()
to send

Rework IPV6 TCP path MTU discovery to match IPv4

- Re-write tcp_ctlinput6() to closely mimic the IPv4 tcp_ctlinput()

- Now that tcp_ctlinput6() updates t_maxseg, we can allow ip6_output()
to send TCP packets without looking at the tcp host cache for every
single transmit.

- Make the icmp6 code mimic the IPv4 code & avoid returning
PRC_HOSTDEAD because it is so expensive.

Without these changes in place, every TCP6 pmtu discovery or host
unreachable ICMP resulted in a call to in6_pcbnotify() which walks the
tcbinfo table with the write lock held. Because the tcbinfo table is
shared between IPv4 and IPv6, this causes huge scalabilty issues on
servers with lots of (~100K) TCP connections, to the point where even
a small percent of IPv6 traffic had a disproportionate impact on
overall throughput.

Reviewed by: bz, rrs, ae (all earlier versions), lstewart (in Netflix's tree)
Sponsored by: Netflix
Differential Revision: https://reviews.freebsd.org/D7272

show more ...

Fix a page fault in ip6_setpktopt(), occurring when the pflog module is
loaded, and syncthing is started, which uses setsockopt(IPV6_PKGINFO).

This is because pflog interfaces do not normally have a

Fix a page fault in ip6_setpktopt(), occurring when the pflog module is
loaded, and syncthing is started, which uses setsockopt(IPV6_PKGINFO).

This is because pflog interfaces do not normally have an IPv6 address,
causing the ND_IFINFO() macro to dereference a NULL pointer.

Reviewed by: ae
PR: 210943
MFC after: 3 days

show more ...

Cleanup unneded include "opt_ipfw.h".

It was used for conditional build IPFIREWALL_FORWARD support.
But IPFIREWALL_FORWARD option was removed a long time ago.

This change re-adds L2 caching for TCP and UDP, as originally added in D4306
but removed due to other changes in the system. Restore the llentry pointer
to the "struct route", and use it to cache the

This change re-adds L2 caching for TCP and UDP, as originally added in D4306
but removed due to other changes in the system. Restore the llentry pointer
to the "struct route", and use it to cache the L2 lookup (ARP or ND6) as
appropriate.

Submitted by: Mike Karels
Differential Revision: https://reviews.freebsd.org/D6262

show more ...

Plug route reference underleak that happens with FLOWTABLE after r297225.

Submitted by: Mike Karels <mike karels.net>

Remove ip6 adjusting from the place where pointer couldn't be changed.
And add comment after calling PFIL hooks, where it could be changed.

Remove ip6 pointer initialization and strange check from the beginning
of ip6_output(). It isn't used until the first time adjusted.
Remove the comment about adjusting where it is actually initialize

Remove ip6 pointer initialization and strange check from the beginning
of ip6_output(). It isn't used until the first time adjusted.
Remove the comment about adjusting where it is actually initialized.

show more ...

Move IPv6 malloc tag definitions into the IPv6 code.

Since PFIL can change destination address, use its always actual value
from mbuf when calculating path mtu. Remove now unused finaldst variable.
Also constify dst argument in ip6_getpmtu() and ip6_ge

Since PFIL can change destination address, use its always actual value
from mbuf when calculating path mtu. Remove now unused finaldst variable.
Also constify dst argument in ip6_getpmtu() and ip6_getpmtu_ctl().

Reviewed by: melifaro
Obtained from: Yandex LLC
Sponsored by: Yandex LLC

show more ...

Call RO_RTFREE() when we have detected the change of destination
address, otherwise the old route will be used with new destination.

MFC after: 1 week

MFH

Sponsored by: The FreeBSD Foundation

sys/net* : for pointers replace 0 with NULL.

Mostly cosmetical, no functional change.

Found with devel/coccinelle.

MFH

Sponsored by: The FreeBSD Foundation

FreeBSD previously provided route caching for TCP (and UDP). Re-add
route caching for TCP, with some improvements. In particular, invalidate
the route cache if a new route is added, which might be a

FreeBSD previously provided route caching for TCP (and UDP). Re-add
route caching for TCP, with some improvements. In particular, invalidate
the route cache if a new route is added, which might be a better match.
The cache is automatically invalidated if the old route is deleted.

Submitted by: Mike Karels
Reviewed by: gnn
Differential Revision: https://reviews.freebsd.org/D4306

show more ...

Merge ^/head r296007 through r296368.

MFH

Sponsored by: The FreeBSD Foundation