- Add RATELIMIT kernel configuration keyword which must be set to
enable the new functionality.
- Add support for hardware driven, Receive Side Scaling, RSS aware, rate
limited sendqueues and expose the functionality through the already
established SO_MAX_PACING_RATE setsockopt(). The API support rates in
the range from 1 to 4Gbytes/s which are suitable for regular TCP and
UDP streams. The setsockopt(2) manual page has been updated.
- Add rate limit function callback API to "struct ifnet" which supports
the following operations: if_snd_tag_alloc(), if_snd_tag_modify(),
if_snd_tag_query() and if_snd_tag_free().
- Add support to ifconfig to view, set and clear the IFCAP_TXRTLMT
flag, which tells if a network driver supports rate limiting or not.
- This patch also adds support for rate limiting through VLAN and LAGG
intermediate network devices.
- How rate limiting works:
1) The userspace application calls setsockopt() after accepting or
making a new connection to set the rate which is then stored in the
socket structure in the kernel. Later on when packets are transmitted
a check is made in the transmit path for rate changes. A rate change
implies a non-blocking ifp->if_snd_tag_alloc() call will be made to the
destination network interface, which then sets up a custom sendqueue
with the given rate limitation parameter. A "struct m_snd_tag" pointer is
returned which serves as a "snd_tag" hint in the m_pkthdr for the
subsequently transmitted mbufs.
2) When the network driver sees the "m->m_pkthdr.snd_tag" different
from NULL, it will move the packets into a designated rate limited sendqueue
given by the snd_tag pointer. It is up to the individual drivers how the rate
limited traffic will be rate limited.
3) Route changes are detected by the NIC drivers in the ifp->if_transmit()
routine when the ifnet pointer in the incoming snd_tag mismatches the
one of the network interface. The network adapter frees the mbuf and
returns EAGAIN which causes the ip_output() to release and clear the send
tag. Upon next ip_output() a new "snd_tag" will be tried allocated.
4) When the PCB is detached the custom sendqueue will be released by a
non-blocking ifp->if_snd_tag_free() call to the currently bound network
interface.
Reviewed by: wblock (manpages), adrian, gallatin, scottl (network)
Differential Revision: https://reviews.freebsd.org/D3687
Sponsored by: Mellanox Technologies
MFC after: 3 months
- 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
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
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
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
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
The m_ext.ext_cnt pointer becomes a union. It can now hold the refcount
value itself. To tell that m_ext.ext_flags flag EXT_FLAG_EMBREF is used.
The first mbuf to attach a cluster stores the refcount. The further mbufs
to reference the cluster point at refcount in the first mbuf. The first
mbuf is freed only when the last reference is freed.
The benefit over refcounts stored in separate slabs is that now refcounts
of different, unrelated mbufs do not share a cache line.
For EXT_EXTREF mbufs the zone_ext_refcnt is no longer needed, and m_extadd()
becomes void, making widely used M_EXTADD macro safe.
For EXT_SFBUF mbufs the sf_ext_ref() is removed, which was an optimization
exactly against the cache aliasing problem with regular refcounting.
Discussed with: rrs, rwatson, gnn, hiren, sbruno, np
Reviewed by: rrs
Differential Revision: https://reviews.freebsd.org/D5396
Sponsored by: Netflix
Add ro_mtu field to 'struct route' to be able to pass lookup MTU back to
the caller.
Currently, ip6_getpmtu() has 2 totally different use cases:
1) control plane (IPV6_PATHMTU req), where we just need to calculate MTU
and return it, w/o any reusability.
2) Actual ip6_output() data path where we (nearly) always use the provided
route lookup data. If this data is not 'valid' we need to perform another
lookup and save the result (which cannot be re-used by ip6_output()).
Given that, handle 1) by calling separate function doing rte lookup itself.
Resulting MTU is calculated by (newly-added) ip6_calcmtu() used by both
ip6_getpmtu_ctl() and ip6_getpmtu().
For 2) instead of storing ref'ed rte, store mtu (the only needed data
from the lookup result) inside newly-added ro_mtu field.
'struct route' was shrinked by 8(or 4 bytes) in r292978. Grow it again
by 4 bytes. New ro_mtu field will be used in other places like
ip/tcp_output (EMSGSIZE handling from output routines).
Reviewed by: ae
Problem description:
How do we currently perform layer 2 resolution and header imposition:
For IPv4 we have the following chain:
ip_output() -> (ether|atm|whatever)_output() -> arpresolve()
Lookup is done in proper place (link-layer output routine) and it is possible
to provide cached lle data.
For IPv6 situation is more complex:
ip6_output() -> nd6_output() -> nd6_output_ifp() -> (whatever)_output() ->
nd6_storelladdr()
We have ip6_ouput() which calls nd6_output() instead of link output routine.
nd6_output() does the following:
* checks if lle exists, creates it if needed (similar to arpresolve())
* performes lle state transitions (similar to arpresolve())
* calls nd6_output_ifp() which pushes packets to link output routine along
with running SeND/MAC hooks regardless of lle state
(e.g. works as run-hooks placeholder).
After that, iface output routine like ether_output() calls nd6_storelladdr()
which performs lle lookup once again.
As a result, we perform lookup twice for each outgoing packet for most types
of interfaces. We also need to maintain runtime-checked table of 'nd6-free'
interfaces (see nd6_need_cache()).
Fix this behavior by eliminating first ND lookup. To be more specific:
* make all nd6_output() consumers use nd6_output_ifp() instead
* rename nd6_output[_slow]() to nd6_resolve_[slow]()
* convert nd6_resolve() and nd6_resolve_slow() to arpresolve() semantics,
e.g. copy L2 address to buffer instead of pushing packet towards lower
layers
* Make all nd6_storelladdr() users use nd6_resolve()
* eliminate nd6_storelladdr()
The resulting callchain is the following:
ip6_output() -> nd6_output_ifp() -> (whatever)_output() -> nd6_resolve()
Error handling:
Currently sending packet to non-existing la results in ip6_<output|forward>
-> nd6_output() -> nd6_output _lle() which returns 0.
In new scenario packet is propagated to <ether|whatever>_output() ->
nd6_resolve() which will return EWOULDBLOCK, and that result
will be converted to 0.
(And EWOULDBLOCK is actually used by IB/TOE code).
Sponsored by: Yandex LLC
Differential Revision: https://reviews.freebsd.org/D1469
Do not pass 'dst' sockaddr to ip[6]_mloopback:
- We have explicit check for AF_INET in ip_output()
- We assume ip header inside passed mbuf in ip_mloopback
- We assume ip6 header inside passed mbuf in ip6_mloopback
When forwarding fragmented IPv6 packets and filtering with PF we
reassemble and refragment. That means we generate new fragment headers
and a new fragment ID.
We already save the fragment IDs so we can do the reassembly so it's
straightforward to apply the incoming fragment ID on the refragmented
packets.
Differential Revision: https://reviews.freebsd.org/D2188
Approved by: gnn (mentor)
When several threads are trying to send datagram to the same destination,
but fragmentation is disabled and datagram size exceeds link MTU,
ip6_output() calls pfctlinput2(PRC_MSGSIZE). It does notify all
sockets wanted to know MTU to this destination. And since all threads
hold PCB lock while sending, taking the lock for each PCB in the
in6_pcbnotify() leads to deadlock.
RFC 3542 p.11.3 suggests notify all application wanted to receive
IPV6_PATHMTU ancillary data for each ICMPv6 packet too big message.
But it doesn't require this, when we don't receive ICMPv6 message.
Change ip6_notify_pmtu() function to be able use it directly from
ip6_output() to notify only one socket, and to notify all sockets
when ICMPv6 packet too big message received.
PR: 197059
Differential Revision: https://reviews.freebsd.org/D1949
Reviewed by: no objection from #network
Obtained from: Yandex LLC
MFC after: 1 week
Sponsored by: Yandex LLC
bits.
The motivation here is to eventually teach netisr and potentially
other networking subsystems a bit more about how RSS work queues / buckets
are configured so things have a hope of auto-configuring in the future.
* net/rss_config.[ch] takes care of the generic bits for doing
configuration, hash function selection, etc;
* topelitz.[ch] is now in net/ rather than netinet/;
* (and would be in libkern if it didn't directly include RSS_KEYSIZE;
that's a later thing to fix up.)
* netinet/in_rss.[ch] now just contains the IPv4 specific methods;
* and netinet/in6_rss.[ch] now just contains the IPv6 specific methods.
This should have no functional impact on anyone currently using
the RSS support.
Differential Revision: D1383
Reviewed by: gnn, jfv (intel driver bits)
use ifqueue at all. Second, there is no point in this lockless check.
Either positive or negative result of the check could be incorrect after
a tick.
Sponsored by: Nginx, Inc.
the knowledge of mbuf layout, and in particular constants such as M_EXT,
MLEN, MHLEN, and so on, in mbuf consumers by unifying various alignment
utility functions (M_ALIGN(), MH_ALIGN(), MEXT_ALIGN() in a single
M_ALIGN() macro, implemented by a now-inlined m_align() function:
- Move m_align() from uipc_mbuf.c to mbuf.h; mark as __inline.
- Reimplement M_ALIGN(), MH_ALIGN(), and MEXT_ALIGN() using m_align().
- Update consumers around the tree to simply use M_ALIGN().
This change eliminates a number of cases where mbuf consumers must be aware
of whether or not mbufs returned by the allocator use external storage, but
also assumptions about the size of the returned mbuf. This will make it
easier to introduce changes in how we use external storage, as well as
features such as variable-size mbufs.
Differential Revision: https://reviews.freebsd.org/D1436
Reviewed by: glebius, trasz, gnn, bz
Sponsored by: EMC / Isilon Storage Division
ipsec_getpolicybyaddr()
ipsec4_checkpolicy()
ip_ipsec_output()
ip6_ipsec_output()
The only flag used here was IP_FORWARDING.
Obtained from: Yandex LLC
Sponsored by: Yandex LLC
from the FreeBSD network code. The flag is still kept around in the
"sys/mbuf.h" header file, but does no longer have any users. Instead
the "m_pkthdr.rsstype" field in the mbuf structure is now used to
decide the meaning of the "m_pkthdr.flowid" field. To modify the
"m_pkthdr.rsstype" field please use the existing "M_HASHTYPE_XXX"
macros as defined in the "sys/mbuf.h" header file.
This patch introduces new behaviour in the transmit direction.
Previously network drivers checked if "M_FLOWID" was set in "m_flags"
before using the "m_pkthdr.flowid" field. This check has now now been
replaced by checking if "M_HASHTYPE_GET(m)" is different from
"M_HASHTYPE_NONE". In the future more hashtypes will be added, for
example hashtypes for hardware dedicated flows.
"M_HASHTYPE_OPAQUE" indicates that the "m_pkthdr.flowid" value is
valid and has no particular type. This change removes the need for an
"if" statement in TCP transmit code checking for the presence of a
valid flowid value. The "if" statement mentioned above is now a direct
variable assignment which is then later checked by the respective
network drivers like before.
Additional notes:
- The SCTP code changes will be committed as a separate patch.
- Removal of the "M_FLOWID" flag will also be done separately.
- The FreeBSD version has been bumped.
MFC after: 1 month
Sponsored by: Mellanox Technologies
Update route MTU in case of ifnet MTU change.
Add new RTF_FIXEDMTU to track explicitly specified MTU.
Old behavior:
ifconfig em0 mtu 1500->9000 -> all routes traversing em0 do not change MTU.
User has to manually update all routes.
ifconfig em0 mtu 9000->1500 -> all routes traversing em0 do not change MTU.
However, if ip[6]_output finds route with rt_mtu > interface mtu, rt_mtu
gets updated.
New behavior:
ifconfig em0 mtu 1500->9000 -> all interface routes in all fibs gets updated
with new MTU unless RTF_FIXEDMTU flag set on them.
ifconfig em0 mtu 9000->1500 -> all routes in all fibs gets updated with new
MTU unless RTF_FIXEDMTU flag set on them AND rt_mtu is less than ifp mtu.
route add ... -mtu XXX automatically sets RTF_FIXEDMTU flag.
route change .. -mtu 0 automatically removes RTF_FIXEDMTU flag.
PR: 194238
MFC after: 1 month
CR: D1125
have chosen different (and more traditional) stateless/statuful
NAT64 as translation mechanism. Last non-trivial commits to both
faith(4) and faithd(8) happened more than 12 years ago, so I assume
it is time to drop RFC3142 in FreeBSD.
No objections from: net@
data in an mbuf, use M_WRITABLE() instead of a direct test of M_EXT;
the latter both unnecessarily exposes mbuf-allocator internals in the
protocol stack and is also insufficient to catch all cases of
non-writability.
(NB: m_pullup() does not actually guarantee that a writable mbuf is
returned, so further refinement of all of these code paths continues to
be required.)
Reviewed by: bz
MFC after: 3 days
Sponsored by: EMC / Isilon Storage Division
Differential Revision: https://reviews.freebsd.org/D900
* Return ENETDOWN when interface specified by ipi6_ifindex is not
enabled for IPv6 use.
* Return EADDRNOTAVAIL when ipi6_ifindex specifies an interface, but the
address ipi6_addr is not available for use on that interface.
* Return EINVAL when ipi6_addr is multicast address.
Obtained from: Yandex LLC
Sponsored by: Yandex LLC
For IPv6-in-IPv4, you may need to do the following command
on the tunnel interface if it is configured as IPv4 only:
ifconfig <interface> inet6 -ifdisabled
Code logic inspired from NetBSD.
PR: kern/169438
Submitted by: emeric.poupon@netasq.com
Reviewed by: fabient, ae
Obtained from: NETASQ
- Use counter(9) for rt_pksent (former rt_rmx.rmx_pksent). This
removes another cache trashing ++ from packet forwarding path.
- Create zini/fini methods for the rtentry UMA zone. Via initialize
mutex and counter in them.
- Fix reporting of rmx_pksent to routing socket.
- Fix netstat(1) to report "Use" both in kvm(3) and sysctl(3) mode.
The change is mostly targeted for stable/10 merge. For head,
rt_pksent is expected to just disappear.
Discussed with: melifaro
Sponsored by: Netflix
Sponsored by: Nginx, Inc.
tested and is unfinished. However, I've tested my version,
it works okay. As before it is unfinished: timeout aren't
driven by TCP session state. To enable the HASH_ALL mode,
one needs in kernel config:
options FLOWTABLE_HASH_ALL
o Reduce the alignment on flentry to 64 bytes. Without
the FLOWTABLE_HASH_ALL option, twice less memory would
be consumed by flows.
o API to ip_output()/ip6_output() got even more thin: 1 liner.
o Remove unused unions. Simply use fle->f_key[].
o Merge all IPv4 code into flowtable_lookup_ipv4(), and do same
flowtable_lookup_ipv6(). Stop copying data to on stack
sockaddr structures, simply use key[] on stack.
o Move code from flowtable_lookup_common() that actually works
on insertion into flowtable_insert().
Sponsored by: Netflix
Sponsored by: Nginx, Inc.
- ip_output() and ip_output6() simply call flowtable_lookup(),
passing mbuf and address family. That's the only code under
#ifdef FLOWTABLE in the protocols code now.
o Revamp statistics gathering and export.
- Remove hand made pcpu stats, and utilize counter(9).
- Snapshot of statistics is available via 'netstat -rs'.
- All sysctls are moved into net.flowtable namespace, since
spreading them over net.inet isn't correct.
o Properly separate at compile time INET and INET6 parts.
o General cleanup.
- Remove chain of multiple flowtables. We simply have one for
IPv4 and one for IPv6.
- Flowtables are allocated in flowtable.c, symbols are static.
- With proper argument to SYSINIT() we no longer need flowtable_ready.
- Hash salt doesn't need to be per-VNET.
- Removed rudimentary debugging, which use quite useless in dtrace era.
The runtime behavior of flowtable shouldn't be changed by this commit.
Sponsored by: Netflix
Sponsored by: Nginx, Inc.
to this event, adding if_var.h to files that do need it. Also, include
all includes that now are included due to implicit pollution via if_var.h
Sponsored by: Netflix
Sponsored by: Nginx, Inc.
