1
0
mirror of https://git.FreeBSD.org/src.git synced 2024-12-19 10:53:58 +00:00
Commit Graph

1130 Commits

Author SHA1 Message Date
Michael Tuexen
f30ac43257 Pass flowid explicitly through the stack instead of taking it from
the mbuf chain at different places.
While there: Fix several bugs related to VRFs.

MFC after: 3 days
2012-06-14 06:54:48 +00:00
Michael Tuexen
b36dcb9a39 Deliver IPV6_TCLASS, IPV6_HOPLIMIT and IPV6_PKTINFO cmsgs (if
requested) on IPV6 sockets, which have been marked to be not IPV6_V6ONLY,
for each received IPV4 packet.

MFC after: 3 days
2012-06-12 13:57:56 +00:00
Bjoern A. Zeeb
15cc25e9c0 Plug two interface address refcount leaks in early error return cases
in the ioctl path.

Reported by:	rpaulo
 Reviewed by:	emax
MFC after:	3 days
2012-06-05 13:27:37 +00:00
Maksim Yevmenkin
3df0e439b0 Plug reference leak.
Interface routes are refcounted as packets move through the stack,
and there's garbage collection tied to it so that route changes can
safely propagate while traffic is flowing. In our setup, we weren't
changing or deleting any routes, but the refcounting logic in
ip6_input() was wrong and caused a reference leak on every inbound
V6 packet. This eventually caused a 32bit overflow, and the resulting
0 value caused the garbage collection to run on the active route.
That then snowballed into the panic.

Reviewed by:	scottl
MFC after:	3 days
2012-06-03 07:36:59 +00:00
Michael Tuexen
a6cff10f2a Seperate SCTP checksum offloading for IPv4 and IPv6.
While there: remove some trainling whitespaces.

MFC after: 3 days
X-MFC with: 236170
2012-05-30 20:56:07 +00:00
Maksim Yevmenkin
c784f9e5eb When we return deprecated addresses, we need to reference them.
Reviewed by:	bz, scottl
MFC after:	3 days
2012-05-30 20:02:39 +00:00
Bjoern A. Zeeb
356ab07e2d It turns out that too many drivers are not only parsing the L2/3/4
headers for TSO but also for generic checksum offloading.  Ideally we
would only have one common function shared amongst all drivers, and
perhaps when updating them for IPv6 we should introduce that.
Eventually we should provide the meta information along with mbufs to
avoid (re-)parsing entirely.

To not break IPv6 (checksums and offload) and to be able to MFC the
changes without risking to hurt 3rd party drivers, duplicate the v4
framework, as other OSes have done as well.

Introduce interface capability flags for TX/RX checksum offload with
IPv6, to allow independent toggling (where possible).  Add CSUM_*_IPV6
flags for UDP/TCP over IPv6, and reserve further for SCTP, and IPv6
fragmentation.  Define CSUM_DELAY_DATA_IPV6 as we do for legacy IP and
add an alias for CSUM_DATA_VALID_IPV6.

This pretty much brings IPv6 handling in line with IPv4.
TSO is still handled in a different way and not via if_hwassist.

Update ifconfig to allow (un)setting of the new capability flags.
Update loopback to announce the new capabilities and if_hwassist flags.

Individual driver updates will have to follow, as will SCTP.

Reported by:	gallatin, dim, ..
Reviewed by:	gallatin (glanced at?)
MFC after:	3 days
X-MFC with:	r235961,235959,235958
2012-05-28 09:30:13 +00:00
Bjoern A. Zeeb
c69baa7e91 Correctly get the payload length in host byte order. While we
already plan to support >64k payload here, the IPv6 header payload
length obviously is only 16 bit and the calculations need to be right.

Reported by:	dim
Tested by:	dim
MFC after:	1 day
X-MFC:		with r235958
2012-05-26 23:58:51 +00:00
Michael Tuexen
8d9638ab33 Get rid of SCTP specific code to avoid CRC32C computations on loopback.
Just just offloading.
MFC after: 3 days
2012-05-26 09:16:33 +00:00
Bjoern A. Zeeb
39e19560d6 MFp4 bz_ipv6_fast:
Use M_ZERO with malloc rather than calling bzero() ourselves.

  Change if () panic() checks to KASSERT()s as they are only
  catching invariants in code flow but not dependent on network
  input/output.

  Move initial assigments indirecting pointers after the lock
  has been aquired.

  Passing layer boundries, reset M_PROTOFLAGS.

  Remove a NULL assignment before free.

  Sponsored by:	The FreeBSD Foundation
  Sponsored by:	iXsystems

Reviewed by:	gnn (as part of the whole)
MFC After:	3 days
2012-05-25 09:27:16 +00:00
Bjoern A. Zeeb
ae14505058 MFp4 bz_ipv6_fast:
Factor out Hop-By-Hop option processing.  It's still not heavily used,
  it reduces the footprint of ip6_input() and makes ip6_input() more
  readable.

  Sponsored by:	The FreeBSD Foundation
  Sponsored by:	iXsystems

Reviewed by:	gnn (as part of the whole)
MFC After:	3 days
2012-05-25 02:58:21 +00:00
Bjoern A. Zeeb
5aa624a803 MFp4 bz_ipv6_fast:
Defer checksum calulations on UDP6 output and respect the mbuf
  flags set by NICs having done checksum validation for us already,
  thus saving the computing time in the input path as well.

  Sponsored by:	The FreeBSD Foundation
  Sponsored by:	iXsystems

Reviewed by:	gnn (as part of the whole)
MFC After:	3 days
2012-05-25 02:19:17 +00:00
Bjoern A. Zeeb
e7b92e2769 MFp4 bz_ipv6_fast:
Add support for delayed checksum calculations in the IPv6
  output path.  We currently cannot offload to the card if we
  add extension headers (which incl. fragmentation).

  Fix two SCTP offload support copy&paste bugs: calculate
  checksums if fragmenting and no need to flag IPv4 header
  checksums in the IPv6 forwarding path.

  Sponsored by:	The FreeBSD Foundation
  Sponsored by:	iXsystems

Reviewed by:	gnn (as part of the whole)
MFC After:	3 days
2012-05-25 02:17:16 +00:00
Bjoern A. Zeeb
d3443481dc MFp4 bz_ipv6_fast:
Hide the ip6aux functions.  The only one referenced outside ip6_input.c
  is not compiled in yet (__notyet__) in route6.c (r235954).  We do have
  accessor functions that should be used.

  Sponsored by:	The FreeBSD Foundation
  Sponsored by:	iXsystems

Reviewed by:	gnn (as part of the whole)
MFC After:	3 days
X-MFC:		KPI?
2012-05-25 01:48:15 +00:00
Bjoern A. Zeeb
f8315b5fd6 MFp4 bz_ipv6_fast:
Simplify the code removing a return from an earlier else case,
  not differing from the default function return called now.

  Sponsored by:	The FreeBSD Foundation
  Sponsored by:	iXsystems

Reviewed by:	gnn (as part of the whole)
MFC After:	3 days
2012-05-25 01:45:05 +00:00
Bjoern A. Zeeb
1b53a49ad9 MFp4 bz_ipv6_fast:
We currently nowhere set IP6A_SWAP making the entire check useless
  with the current code.  Keep around but do not compile in.

  Sponsored by:	The FreeBSD Foundation
  Sponsored by:	iXsystems

Reviewed by:	gnn (as part of the whole)
MFC After:	3 days
2012-05-25 01:43:52 +00:00
Bjoern A. Zeeb
2cf62998da MFp4 bz_ipv6_fast:
No need to hold the (expensive) rt lock over (expensive) logging.

  Sponsored by:	The FreeBSD Foundation
  Sponsored by:	iXsystems

Reviewed by:	gnn (as part of the whole)
MFC After:	3 days
2012-05-25 01:42:48 +00:00
Bjoern A. Zeeb
ecade87edf MFp4 bz_ipv6_fast:
Introduce a (for now copied stripped down) in6_cksum_pseudo()
  function.  We should be able to use this from in6_cksum() but
  we should also ponder possible MD specific improvements.
  It takes an extra csum argument to allow for easy checks as
  will be done by the upper layer protocol input paths.

  Sponsored by:	The FreeBSD Foundation
  Sponsored by:	iXsystems

Reviewed by:	gnn (as part of the whole)
MFC After:	3 days
2012-05-24 18:25:09 +00:00
Bjoern A. Zeeb
2889eb8bdf MFp4 bz_ipv6_fast:
Optimize in6_cksum(), re-ordering work and limiting variable
  initialization, removing a bzero() for mostly re-initialized
  struct values, making use of the newly introduced in6_getscope(),
  as well as converting an if/panic to a KASSERT().

  Sponsored by:	The FreeBSD Foundation
  Sponsored by:	iXsystems

Reviewed by:	gnn (as part of the whole)
MFC After:	3 days
2012-05-24 18:05:10 +00:00
Bjoern A. Zeeb
0edc703a02 MFp4 bz_ipv6_fast:
Introduce in6_getscope() to allow more effective checksum
  computations without the need to copy the address to clear the
  scope.

  Sponsored by:	The FreeBSD Foundation
  Sponsored by:	iXsystems

Reviewed by:	gnn (as part of the whole)
MFC After:	3 days
2012-05-24 16:30:13 +00:00
Michael Tuexen
807aad636f Use consistent text at the begining of the files.
MFC after: 3 days
2012-05-23 11:26:28 +00:00
Marius Strobl
7195094c3c Rewrite nd6_sysctl_{d,p}rlist() to avoid misaligned accesses to char arrays
casted to structs by getting rid of these buffers entirely. In r169832, it
was tried to paper over this issue by 32-bit aligning the buffers. Depending
on compiler optimizations that still was insufficient for 64-bit architectures
with strong alignment requirements though.
While at it, add comments regarding the total lack of locking in this area.

Tested by:	bz
Reviewed by:	bz (slightly earlier version), yongari (earlier version)
MFC after:	1 week
2012-05-20 05:12:31 +00:00
Michael Tuexen
1a5b79010a Missed to commit this in r235414.
MFC after: 3 days
2012-05-13 19:25:21 +00:00
Michael Tuexen
410a3b1ef0 Use ECONNABORTED in cases where the ABORT was sent to the peer.
MFC after: 3 days
2012-05-13 16:56:16 +00:00
Michael Tuexen
a2b42326b5 Provide in the association change notification the received ABORT chunk
if case of SCTP_COMM_LOST or SCTP_CANT_STR_ASSOC as required by RFC 6458.

MFC after: 3 days
2012-05-12 20:11:35 +00:00
Gleb Smirnoff
d0e6c546a2 in6_pcblookup_local() still can return a pcb with NULL
inp_socket. To avoid panic, do not dereference inp_socket,
but obtain reuse port option from inp_flags2, like this
is done after next call to in_pcblookup_local() a few lines
down below.

Submitted by:	rwatson
2012-03-21 08:43:38 +00:00
Michael Tuexen
dea47f3999 Clean up, no functional change.
MFC after: 3 days.
2012-03-15 14:22:05 +00:00
Bjoern A. Zeeb
abbe8356ea In nd6_options() ignore the RFC 6106 options completely rather than printing
them if nd6_debug is enabled as unknown.  Leave a comment about the RFC4191
option as I am undecided so far.

Discussed with:	hrs
MFC after:	3 days
2012-03-04 18:51:45 +00:00
Hiroki Sato
dead19563c Allow to configure net.inet6.ip6.{accept_rtadv,no_radr} by the loader tunables
as well because they have to be configured before interface initialization for
AF_INET6.
2012-03-02 07:23:28 +00:00
Hiroki Sato
a1875676ca Remove a redundant check. 2012-03-02 07:22:04 +00:00
Bjoern A. Zeeb
5aa7e8edc5 In selectroute() add a missing fibnum argument to an in6_rtalloc()
call in an #if 0 section.

In in6_selecthlim() optimize a case where in6p cannot be NULL due to an
earlier check.

More consistently use u_int instead of int for fibnum function arguments.

Sponsored by:	Cisco Systems, Inc.
MFC after:	3 days
2012-02-24 20:06:04 +00:00
Kip Macy
a93cda789a When using flowtable llentrys can outlive the interface with which they're associated
at which the lle_tbl pointer points to freed memory and the llt_free pointer is no longer
valid.

Move the free pointer in to the llentry itself and update the initalization sites.

MFC after:	2 weeks
2012-02-23 18:21:37 +00:00
Michael Tuexen
310a027788 Remove two clang warnings.
MFC after: 1 month.
2012-02-18 16:06:15 +00:00
Bjoern A. Zeeb
1b46c7f832 Allow to provide a hint to in6_selectsrc() for the interface using the
return ifnet double pointer.
Pass that hint down to in6_selectif() to be used when i) the default FIB
is queried and ii) route lookup fails because the network is not present
(i.e. someone deleted the connected subnet).
This hint should not be generally used from anywhere outside the neighbor
discovery code.  We just make use of it from nd6_ns_output().

Extend the nd6_na_output() interface by a nd6_na_output_fib() version
and pass the FIB number from the NS mbuf on to NA to allow the new mbuf
to inherit the FIB tag and a later lookup from ip6_output() to succeed
in the aformentioned example case.
Provide a wrapper function for the old public interface also used from
CARP but mark it with BURN_BRIDGES to cleanup in HEAD after MFC.

Sponsored by:	Cisco Systems, Inc.
2012-02-14 11:51:32 +00:00
Bjoern A. Zeeb
81d5d46b3c Add multi-FIB IPv6 support to the core network stack supplementing
the original IPv4 implementation from r178888:

- Use RT_DEFAULT_FIB in the IPv4 implementation where noticed.
- Use rt*fib() KPI with explicit RT_DEFAULT_FIB where applicable in
  the NFS code.
- Use the new in6_rt* KPI in TCP, gif(4), and the IPv6 network stack
  where applicable.
- Split in6_rtqtimo() and in6_mtutimo() as done in IPv4 and equally
  prevent multiple initializations of callouts in in6_inithead().
- Use wrapper functions where needed to preserve the current KPI to
  ease MFCs.  Use BURN_BRIDGES to indicate expected future cleanup.
- Fix (related) comments (both technical or style).
- Convert to rtinit() where applicable and only use custom loops where
  currently not possible otherwise.
- Multicast group, most neighbor discovery address actions and faith(4)
  are locked to the default FIB.  Individual IPv6 addresses will only
  appear in the default FIB, however redirect information and prefixes
  of connected subnets are automatically propagated to all FIBs by
  default (mimicking IPv4 behavior as closely as possible).

Sponsored by:	Cisco Systems, Inc.
2012-02-03 13:08:44 +00:00
Bjoern A. Zeeb
ee799639e8 Add SO_SETFIB option support on PF_INET6 sockets and allow inheriting the
FIB number from the process, as set by setfib(2), on socket creation.

Sponsored by:	Cisco Systems, Inc.
2012-02-03 11:00:53 +00:00
Bjoern A. Zeeb
db566a23b6 Provide the IPv6 counterpart to the extended IPv4 rtalloc(9) KPI.
Sponsored by:	Cisco Systems, Inc.
2012-02-03 09:33:58 +00:00
Bjoern A. Zeeb
5490110cf5 In preparation for multi-FIB IPv6 support, factor the code for joining
and leaving multicast groups out from in6_update_ifa() and in6_purgeaddr().

Sponsored by:	Cisco Systems, Inc.
2012-02-03 08:50:19 +00:00
Gleb Smirnoff
9c2ae3b1c8 Remove casts from inet6 address testing macros, thus preserving
qualifier from original argument.

Obtained from:	NetBSD, r. 1.67
Submitted by:	maxim
2012-01-26 12:04:19 +00:00
Sergey Kandaurov
8e4609a4a3 Remove unused variable.
The actual ia6->ia6_lifetime access is hidden in
IFA6_IS_INVALID/IFA6_IS_DEPRECATED macros since a long time ago
(see netinet6/nd6.c, r1.104 of KAME for the reference).

MFC after:	3 days
2012-01-25 08:53:42 +00:00
Bjoern A. Zeeb
4aa7588c8f Plug a possible ifa_ref leak in case of premature return from in6_purgeaddr().
Reviewed by:	rwatson
MFC after:	3 days
2012-01-24 13:57:30 +00:00
Sergey Kandaurov
373de5d88b Remove the stale XXX rt_newaddrmsg comment.
A routing socket message is generated since r192282.

Reviewed by:	bz
MFC after:	3 days
2012-01-24 09:51:42 +00:00
Bjoern A. Zeeb
e0f1891c48 Remove unnecessary line break.
MFC after:	3 days
2012-01-24 06:21:38 +00:00
Bjoern A. Zeeb
83e521ec73 Clean up some #endif comments removing from short sections. Add #endif
comments to longer, also refining strange ones.

Properly use #ifdef rather than #if defined() where possible.  Four
#if defined(PCBGROUP) occurances (netinet and netinet6) were ignored to
avoid conflicts with eventually upcoming changes for RSS.

Reported by:	bde (most)
Reviewed by:	bde
MFC after:	3 days
2012-01-22 02:13:19 +00:00
Michael Tuexen
f7f2907a7e Small cleanup, no functional change. 2012-01-15 14:03:05 +00:00
Michael Tuexen
c58e60be43 Add an SCTP sysctl "blackhole", similar to the one for TCP.
If set to 1, no ABORT is sent back in response to an incoming
INIT. If set to 2, no ABORT is sent back in response to
an out of the blue packet. If set to 0 (the default), ABORTs
are sent.
Discussed with rrs@.

MFC after: 1 month.
2012-01-08 09:56:24 +00:00
John Baldwin
137f91e80f Convert all users of IF_ADDR_LOCK to use new locking macros that specify
either a read lock or write lock.

Reviewed by:	bz
MFC after:	2 weeks
2012-01-05 19:00:36 +00:00
Bjoern A. Zeeb
9d9b92f299 Mark a couple of file local functions static and stop exporting them.
MFC after:	1 week
2012-01-05 01:14:35 +00:00
Bjoern A. Zeeb
f67e13d66d Convert an #ifdef DIAGNOSTIC if/panic to a KASSERT.
MFC after:	1 week
2012-01-05 01:13:25 +00:00
John Baldwin
19b0c9b246 Use the mli_relinmhead list normally used to defer calls to
in6m_release_locked() to defer calls to mld_v1_transmit_report() until
after the IF_ADDR_LOCK is dropped.  This removes a race where the lock
is dropped and reacquired while attempting to walk an interface's
address list.

Reviewed by:	bz
MFC after:	1 week
2012-01-04 13:35:20 +00:00