ipsec6_common_input_cb() uses partial copy of ip6_input() to parse
headers. But this isn't correct, when we use tunnel mode IPSec.
When we stripped outer IPv6 header from the decrypted packet, it
can become IPv4 packet and should be handled by ip_input. Also when
we use tunnel mode IPSec with IPv6 traffic, we should pass decrypted
packet with inner IPv6 header to ip6_input, it will correctly handle
it and also can decide to forward it.
The "skip" variable points to offset where payload starts. In tunnel
mode we reset it to zero after stripping the outer header. So, when
it is zero, we should requeue mbuf via netisr.
Differential Revision: https://reviews.freebsd.org/D2306
Reviewed by: adrian, gnn
Sponsored by: Yandex LLC
* in ipsec_encap() embed scope zone ids into link-local addresses
in the new IPv6 header, this helps ip6_output() disambiguate the
scope;
* teach key_ismyaddr6() use in6_localip(). in6_localip() is less
strict than key_sockaddrcmp(). It doesn't compare all fileds of
struct sockaddr_in6, but it is faster and it should be safe,
because all SA's data was checked for correctness. Also, since
IPv6 link-local addresses in the &V_in6_ifaddrhead are stored in
kernel-internal form, we need to embed scope zone id from SA into
the address before calling in6_localip.
* in ipsec_common_input() take scope zone id embedded in the address
and use it to initialize sin6_scope_id, then use this sockaddr
structure to lookup SA, because we keep addresses in the SADB without
embedded scope zone id.
Differential Revision: https://reviews.freebsd.org/D2304
Reviewed by: gnn
Sponsored by: Yandex LLC
The only thing is used from this code is ipip_output() function, that does
IPIP encapsulation. Other parts of XF_IP4 code were removed in r275133.
Also it isn't possible to configure the use of XF_IP4, nor from userland
via setkey(8), nor from the kernel.
Simplify the ipip_output() function and rename it to ipsec_encap().
* move IP_DF handling from ipsec4_process_packet() into ipsec_encap();
* since ipsec_encap() called from ipsec[64]_process_packet(), it
is safe to assume that mbuf is contiguous at least to IP header
for used IP version. Remove all unneeded m_pullup(), m_copydata
and related checks.
* use V_ip_defttl and V_ip6_defhlim for outer headers;
* use V_ip4_ipsec_ecn and V_ip6_ipsec_ecn for outer headers;
* move all diagnostic messages to the ipsec_encap() callers;
* simplify handling of ipsec_encap() results: if it returns non zero
value, print diagnostic message and free mbuf.
* some style(9) fixes.
Differential Revision: https://reviews.freebsd.org/D2303
Reviewed by: glebius
Sponsored by: Yandex LLC
where we want to create a new IP datagram.
o Add support for RFC6864, which allows to set IP ID for atomic IP
datagrams to any value, to improve performance. The behaviour is
controlled by net.inet.ip.rfc6864 sysctl knob, which is enabled by
default.
o In case if we generate IP ID, use counter(9) to improve performance.
o Gather all code related to IP ID into ip_id.c.
Differential Revision: https://reviews.freebsd.org/D2177
Reviewed by: adrian, cy, rpaulo
Tested by: Emeric POUPON <emeric.poupon stormshield.eu>
Sponsored by: Netflix
Sponsored by: Nginx, Inc.
Relnotes: yes
code.
Resurrect the state field in the struct secpolicy, it has
IPSEC_SPSTATE_ALIVE value when security policy linked in the chain,
and IPSEC_SPSTATE_DEAD value in all other cases. This field protects
from trying to unlink one security policy several times from the different
threads.
Take additional reference in the key_flush_spd() to be sure that policy
won't be freed from the different thread while we are sending SPDEXPIRE message.
Add KEY_FREESP() call to the key_unlink() to release additional reference
that we take when use key_getsp*() functions.
Differential Revision: https://reviews.freebsd.org/D1914
Tested by: Emeric POUPON <emeric.poupon at stormshield dot eu>
Reviewed by: hrs
Sponsored by: Yandex LLC
socket-buffer implementations, introduce a return value for MCLGET()
(and m_cljget() that underlies it) to allow the caller to avoid testing
M_EXT itself. Update all callers to use the return value.
With this change, very few network device drivers remain aware of
M_EXT; the primary exceptions lie in mbuf-chain pretty printers for
debugging, and in a few cases, custom mbuf and cluster allocation
implementations.
NB: This is a difficult-to-test change as it touches many drivers for
which I don't have physical devices. Instead we've gone for intensive
review, but further post-commit review would definitely be appreciated
to spot errors where changes could not easily be made mechanically,
but were largely mechanical in nature.
Differential Revision: https://reviews.freebsd.org/D1440
Reviewed by: adrian, bz, gnn
Sponsored by: EMC / Isilon Storage Division
IPv6. Initialize it only once in def_policy_init(). Remove its
initialization from key_init() and make it static.
Remove several fields from struct secpolicy:
* lock - it isn't so useful having mutex in the structure, but the only
thing we do with it is initialization and destroying.
* state - it has only two values - DEAD and ALIVE. Instead of take a lock
and change the state to DEAD, then take lock again in GC function and
delete policy from the chain - keep in the chain only ALIVE policies.
* scangen - it was used in GC function to protect from sending several
SADB_SPDEXPIRE messages for one SPD entry. Now we don't keep DEAD entries
in the chain and there is no need to have scangen variable.
Use TAILQ to implement SPD entries chain. Use rmlock to protect access
to SPD entries chain. Protect all SP lookup with RLOCK, and use WLOCK
when we are inserting (or removing) SP entry in the chain.
Instead of using pattern "LOCK(); refcnt++; UNLOCK();", use refcount(9)
API to implement refcounting in SPD. Merge code from key_delsp() and
_key_delsp() into _key_freesp(). And use KEY_FREESP() macro in all cases
when we want to release reference or just delete SP entry.
Obtained from: Yandex LLC
Sponsored by: Yandex LLC
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
and make its prototype similar to ipsec6_process_packet.
The flags argument isn't used here, tunalready is always zero.
Obtained from: Yandex LLC
Sponsored by: Yandex LLC
xform_ipip was used as fallback with low priority for IPIP
encapsulated packets that were decrypted. In some cases
it can decapsulate packets, that it shouldn't. This leads to situations,
when wrong configurations are magically working. Also it can propagate
wrong ingress interface and this can break security.
Now we redesigned the IPSEC code and IPIP encapsulation is called directly
from ipsec_output, and decapsulation is done in the ipsec_input with m_striphdr.
Differential Revision: https://reviews.freebsd.org/D1220
MFC after: 1 month
Sponsored by: Yandex LLC
- Wrong integer type was specified.
- Wrong or missing "access" specifier. The "access" specifier
sometimes included the SYSCTL type, which it should not, except for
procedural SYSCTL nodes.
- Logical OR where binary OR was expected.
- Properly assert the "access" argument passed to all SYSCTL macros,
using the CTASSERT macro. This applies to both static- and dynamically
created SYSCTLs.
- Properly assert the the data type for both static and dynamic
SYSCTLs. In the case of static SYSCTLs we only assert that the data
pointed to by the SYSCTL data pointer has the correct size, hence
there is no easy way to assert types in the C language outside a
C-function.
- Rewrote some code which doesn't pass a constant "access" specifier
when creating dynamic SYSCTL nodes, which is now a requirement.
- Updated "EXAMPLES" section in SYSCTL manual page.
MFC after: 3 days
Sponsored by: Mellanox Technologies
Instead requeue mbuf back to IPv4 protocol handler. If there is one extra IP-IP
encapsulation, it will be handled with tunneling interface. And thus proper
interface will be exposed into mbuf's rcvif. Also, tcpdump that listens on tunneling
interface will see packets in both directions.
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
were primarily used to size the sysctl name list macros that were removed
in r254295. A few other constants either did not have an associated
sysctl node, or the associated node used OID_AUTO instead.
PR: ports/184525 (exp-run)
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.
structure is used, but they already have equal fields in the struct
newipsecstat, that was introduced with FAST_IPSEC and then was merged
together with old ipsecstat structure.
This fixes kernel stack overflow on some architectures after migration
ipsecstat to PCPU counters.
Reported by: Taku YAMAMOTO, Maciej Milewski
examination shows, that although key_alloc_mbuf() could return chains,
the callers never use chains, so m_get2() should suffice.
Sponsored by: Nginx, Inc.
before passing a packet to protocol input routines.
For several protocols this mean that now protocol needs to
do subtraction itself, and for another half this means that
we do not need to add header length back to the packet.
Make ip_stripoptions() to adjust ip_len, since now we enter
this function with a packet header whose ip_len does represent
length of entire packet, not payload only.
in network byte order. Any host byte order processing is
done in local variables and host byte order values are
never[1] written to a packet.
After this change a packet processed by the stack isn't
modified at all[2] except for TTL.
After this change a network stack hacker doesn't need to
scratch his head trying to figure out what is the byte order
at the given place in the stack.
[1] One exception still remains. The raw sockets convert host
byte order before pass a packet to an application. Probably
this would remain for ages for compatibility.
[2] The ip_input() still subtructs header len from ip->ip_len,
but this is planned to be fixed soon.
Reviewed by: luigi, Maxim Dounin <mdounin mdounin.ru>
Tested by: ray, Olivier Cochard-Labbe <olivier cochard.me>
SAs. For now allow same address family bundles. While discovered with
ESP and AH, which does not make a lot of sense, IPcomp could be a possible
problematic candidate.
PR: kern/164400
MFC after: 3 days
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.
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
Unfold the IPSEC_COMMON_INPUT_CB() macro in xform_{ah,esp,ipcomp}.c
to not need three different versions depending on INET, INET6 or both.
Mark two places preparing for not yet supported functionality with IPv6.
Reviewed by: gnn
Sponsored by: The FreeBSD Foundation
Sponsored by: iXsystems
MFC after: 4 days
- Remove contention on ISR during the crypto operation by using rwlock(9).
- Remove a second lookup of the SA in the callback.
Gain on 6 cores CPU with SHA1/AES128 can be up to 30%.
Reviewed by: vanhu
MFC after: 1 month
- AH does not release the SA like in ESP/IPCOMP when handling EAGAIN
- ipsec_process_done incorrectly release the SA.
Reviewed by: vanhu
MFC after: 1 week
DPCPU_DEFINE and VNET_DEFINE macros, as these cause problems for various
people working on the affected files. A better long-term solution is
still being considered. This reversal may give some modules empty
set_pcpu or set_vnet sections, but these are harmless.
Changes reverted:
------------------------------------------------------------------------
r215318 | dim | 2010-11-14 21:40:55 +0100 (Sun, 14 Nov 2010) | 4 lines
Instead of unconditionally emitting .globl's for the __start_set_xxx and
__stop_set_xxx symbols, only emit them when the set_vnet or set_pcpu
sections are actually defined.
------------------------------------------------------------------------
r215317 | dim | 2010-11-14 21:38:11 +0100 (Sun, 14 Nov 2010) | 3 lines
Apply the STATIC_VNET_DEFINE and STATIC_DPCPU_DEFINE macros throughout
the tree.
------------------------------------------------------------------------
r215316 | dim | 2010-11-14 21:23:02 +0100 (Sun, 14 Nov 2010) | 2 lines
Add macros to define static instances of VNET_DEFINE and DPCPU_DEFINE.
feature_present(3) checks.
This will help to run-time detect and conditionally handle specific
optionas of either feature in user space (i.e. in libipsec).
Descriptions read by: rwatson
MFC after: 2 weeks
legacy and IPv6 route destination address.
Previously in case of IPv6, there was a memory overwrite due to not enough
space for the IPv6 address.
PR: kern/122565
MFC After: 2 weeks
Improve IPsec flow distribution for better netisr parallelism.
Instead of using the pointer that would have the last bits masked in a %
statement in netisr_select_cpuid() to select the queue, use the SPI.
Reviewed by: rwatson
MFC after: 4 weeks
"Whitspace" churn after the VIMAGE/VNET whirls.
Remove the need for some "init" functions within the network
stack, like pim6_init(), icmp_init() or significantly shorten
others like ip6_init() and nd6_init(), using static initialization
again where possible and formerly missed.
Move (most) variables back to the place they used to be before the
container structs and VIMAGE_GLOABLS (before r185088) and try to
reduce the diff to stable/7 and earlier as good as possible,
to help out-of-tree consumers to update from 6.x or 7.x to 8 or 9.
This also removes some header file pollution for putatively
static global variables.
Revert VIMAGE specific changes in ipfilter::ip_auth.c, that are
no longer needed.
Reviewed by: jhb
Discussed with: rwatson
Sponsored by: The FreeBSD Foundation
Sponsored by: CK Software GmbH
MFC after: 6 days
payload size. Before we had always added the header, no matter if we
actually send out compressed data or not.
With this, after the opencrypto/deflate changes, IPcomp starts to work
apart from edge cases. Leave it disabled by default until those are
fixed as well.
PR: kern/123587
MFC after: 5 days
may come from outside without being discarded before.
Submitted by: aurelien.ansel@netasq.com
Reviewed by: bz (secteam)
Obtained from: NETASQ
MFC after: 1m
compare them together instead of comparing each one with respective
tunnel endpoint.
PR: kern/138439
Submitted by: aurelien.ansel@netasq.com
Obtained from: NETASQ
MFC after: 1 m
several critical bugs, including race conditions and lock order issues:
Replace the single rwlock, ifnet_lock, with two locks, an rwlock and an
sxlock. Either can be held to stablize the lists and indexes, but both
are required to write. This allows the list to be held stable in both
network interrupt contexts and sleepable user threads across sleeping
memory allocations or device driver interactions. As before, writes to
the interface list must occur from sleepable contexts.
Reviewed by: bz, julian
MFC after: 3 days
vnet.h, we now use jails (rather than vimages) as the abstraction
for virtualization management, and what remained was specific to
virtual network stacks. Minor cleanups are done in the process,
and comments updated to reflect these changes.
Reviewed by: bz
Approved by: re (vimage blanket)
network stacks, VNET_SYSINIT:
- Add VNET_SYSINIT and VNET_SYSUNINIT macros to declare events that will
occur each time a network stack is instantiated and destroyed. In the
!VIMAGE case, these are simply mapped into regular SYSINIT/SYSUNINIT.
For the VIMAGE case, we instead use SYSINIT's to track their order and
properties on registration, using them for each vnet when created/
destroyed, or immediately on module load for already-started vnets.
- Remove vnet_modinfo mechanism that existed to serve this purpose
previously, as well as its dependency scheme: we now just use the
SYSINIT ordering scheme.
- Implement VNET_DOMAIN_SET() to allow protocol domains to declare that
they want init functions to be called for each virtual network stack
rather than just once at boot, compiling down to DOMAIN_SET() in the
non-VIMAGE case.
- Walk all virtualized kernel subsystems and make use of these instead
of modinfo or DOMAIN_SET() for init/uninit events. In some cases,
convert modular components from using modevent to using sysinit (where
appropriate). In some cases, do minor rejuggling of SYSINIT ordering
to make room for or better manage events.
Portions submitted by: jhb (VNET_SYSINIT), bz (cleanup)
Discussed with: jhb, bz, julian, zec
Reviewed by: bz
Approved by: re (VIMAGE blanket)
nor destructors, as there's no actual work to do.
In most cases, the constructors weren't needed because of the existing
protocol initialization functions run by net_init_domain() as part of
VNET_MOD_NET, or they were eliminated when support for static
initialization of virtualized globals was added.
Garbage collect dependency references to modules without constructors or
destructors, notably VNET_MOD_INET and VNET_MOD_INET6.
Reviewed by: bz
Approved by: re (vimage blanket)
unused custom mutex/condvar-based sleep locks with two locks: an
rwlock (for non-sleeping use) and sxlock (for sleeping use). Either
acquired for read is sufficient to stabilize the vnet list, but both
must be acquired for write to modify the list.
Replace previous no-op read locking macros, used in various places
in the stack, with actual locking to prevent race conditions. Callers
must declare when they may perform unbounded sleeps or not when
selecting how to lock.
Refactor vnet sysinits so that the vnet list and locks are initialized
before kernel modules are linked, as the kernel linker will use them
for modules loaded by the boot loader.
Update various consumers of these KPIs based on whether they may sleep
or not.
Reviewed by: bz
Approved by: re (kib)
(DPCPU), as suggested by Peter Wemm, and implement a new per-virtual
network stack memory allocator. Modify vnet to use the allocator
instead of monolithic global container structures (vinet, ...). This
change solves many binary compatibility problems associated with
VIMAGE, and restores ELF symbols for virtualized global variables.
Each virtualized global variable exists as a "reference copy", and also
once per virtual network stack. Virtualized global variables are
tagged at compile-time, placing the in a special linker set, which is
loaded into a contiguous region of kernel memory. Virtualized global
variables in the base kernel are linked as normal, but those in modules
are copied and relocated to a reserved portion of the kernel's vnet
region with the help of a the kernel linker.
Virtualized global variables exist in per-vnet memory set up when the
network stack instance is created, and are initialized statically from
the reference copy. Run-time access occurs via an accessor macro, which
converts from the current vnet and requested symbol to a per-vnet
address. When "options VIMAGE" is not compiled into the kernel, normal
global ELF symbols will be used instead and indirection is avoided.
This change restores static initialization for network stack global
variables, restores support for non-global symbols and types, eliminates
the need for many subsystem constructors, eliminates large per-subsystem
structures that caused many binary compatibility issues both for
monitoring applications (netstat) and kernel modules, removes the
per-function INIT_VNET_*() macros throughout the stack, eliminates the
need for vnet_symmap ksym(2) munging, and eliminates duplicate
definitions of virtualized globals under VIMAGE_GLOBALS.
Bump __FreeBSD_version and update UPDATING.
Portions submitted by: bz
Reviewed by: bz, zec
Discussed with: gnn, jamie, jeff, jhb, julian, sam
Suggested by: peter
Approved by: re (kensmith)
for in_ifaddrhead, we stick with an rwlock for the time being, which
we will revisit in the future with a possible move to rmlocks.
Some pieces of code require significant further reworking to be
safe from all classes of writer-writer races.
Reviewed by: bz
MFC after: 6 weeks
in_ifaddrhead and INADDR_HASH address lists.
Previously, these lists were used unsynchronized as they were effectively
never changed in steady state, but we've seen increasing reports of
writer-writer races on very busy VPN servers as core count has gone up
(and similar configurations where address lists change frequently and
concurrently).
For the time being, use rwlocks rather than rmlocks in order to take
advantage of their better lock debugging support. As a result, we don't
enable ip_input()'s read-locking of INADDR_HASH until an rmlock conversion
is complete and a performance analysis has been done. This means that one
class of reader-writer races still exists.
MFC after: 6 weeks
Reviewed by: bz
for the global IPv6 address list (in6_ifaddr -> in6_ifaddrhead). Adopt
the code styles and conventions present in netinet where possible.
Reviewed by: gnn, bz
MFC after: 6 weeks (possibly not MFCable?)
below key_setsaval().
Without that, the lookup for the SA had failed as we were looking for
a SA with the new, updated port numbers instead of the old ones and
were comparing the ports in key_cmpsaidx().
This makes updating the remote -> local SA on the initiator work again.
Problem introduced with: p4 changeset 152114
missing it.
Remove the "hidden" kernel only include of vimage.h from ip_var.h added
with the very first Vimage commit r181803 to avoid further kernel poisoning.
Thanks to (no special order) Emmanuel Dreyfus (manu@netbsd.org), Larry
Baird (lab@gta.com), gnn, bz, and other FreeBSD devs, Julien Vanherzeele
(julien.vanherzeele@netasq.com, for years of bug reporting), the PFSense
team, and all people who used / tried the NAT-T patch for years and
reported bugs, patches, etc...
X-MFC: never
Reviewed by: bz
Approved by: gnn(mentor)
Obtained from: NETASQ
the ROUTETABLES kernel option thus there is no need to include opt_route.h
anymore in all consumers of vnet.h and no longer depend on it for module
builds.
Remove the hidden include in flowtable.h as well and leave the two
explicit #includes in ip_input.c and ip_output.c.
Vnet modules and protocol domains may now register destructor
functions to clean up and release per-module state. The destructor
mechanisms can be triggered by invoking "vimage -d", or a future
equivalent command which will be provided via the new jail framework.
While this patch introduces numerous placeholder destructor functions,
many of those are currently incomplete, thus leaking memory or (even
worse) failing to stop all running timers. Many of such issues are
already known and will be incrementaly fixed over the next weeks in
smaller incremental commits.
Apart from introducing new fields in structs ifnet, domain, protosw
and vnet_net, which requires the kernel and modules to be rebuilt, this
change should have no impact on nooptions VIMAGE builds, since vnet
destructors can only be called in VIMAGE kernels. Moreover,
destructor functions should be in general compiled in only in
options VIMAGE builds, except for kernel modules which can be safely
kldunloaded at run time.
Bump __FreeBSD_version to 800097.
Reviewed by: bz, julian
Approved by: rwatson, kib (re), julian (mentor)
threads:
- Support up to one netisr thread per CPU, each processings its own
workstream, or set of per-protocol queues. Threads may be bound
to specific CPUs, or allowed to migrate, based on a global policy.
In the future it would be desirable to support topology-centric
policies, such as "one netisr per package".
- Allow each protocol to advertise an ordering policy, which can
currently be one of:
NETISR_POLICY_SOURCE: packets must maintain ordering with respect to
an implicit or explicit source (such as an interface or socket).
NETISR_POLICY_FLOW: make use of mbuf flow identifiers to place work,
as well as allowing protocols to provide a flow generation function
for mbufs without flow identifers (m2flow). Falls back on
NETISR_POLICY_SOURCE if now flow ID is available.
NETISR_POLICY_CPU: allow protocols to inspect and assign a CPU for
each packet handled by netisr (m2cpuid).
- Provide utility functions for querying the number of workstreams
being used, as well as a mapping function from workstream to CPU ID,
which protocols may use in work placement decisions.
- Add explicit interfaces to get and set per-protocol queue limits, and
get and clear drop counters, which query data or apply changes across
all workstreams.
- Add a more extensible netisr registration interface, in which
protocols declare 'struct netisr_handler' structures for each
registered NETISR_ type. These include name, handler function,
optional mbuf to flow ID function, optional mbuf to CPU ID function,
queue limit, and ordering policy. Padding is present to allow these
to be expanded in the future. If no queue limit is declared, then
a default is used.
- Queue limits are now per-workstream, and raised from the previous
IFQ_MAXLEN default of 50 to 256.
- All protocols are updated to use the new registration interface, and
with the exception of netnatm, default queue limits. Most protocols
register as NETISR_POLICY_SOURCE, except IPv4 and IPv6, which use
NETISR_POLICY_FLOW, and will therefore take advantage of driver-
generated flow IDs if present.
- Formalize a non-packet based interface between interface polling and
the netisr, rather than having polling pretend to be two protocols.
Provide two explicit hooks in the netisr worker for start and end
events for runs: netisr_poll() and netisr_pollmore(), as well as a
function, netisr_sched_poll(), to allow the polling code to schedule
netisr execution. DEVICE_POLLING still embeds single-netisr
assumptions in its implementation, so for now if it is compiled into
the kernel, a single and un-bound netisr thread is enforced
regardless of tunable configuration.
In the default configuration, the new netisr implementation maintains
the same basic assumptions as the previous implementation: a single,
un-bound worker thread processes all deferred work, and direct dispatch
is enabled by default wherever possible.
Performance measurement shows a marginal performance improvement over
the old implementation due to the use of batched dequeue.
An rmlock is used to synchronize use and registration/unregistration
using the framework; currently, synchronized use is disabled
(replicating current netisr policy) due to a measurable 3%-6% hit in
ping-pong micro-benchmarking. It will be enabled once further rmlock
optimization has taken place. However, in practice, netisrs are
rarely registered or unregistered at runtime.
A new man page for netisr will follow, but since one doesn't currently
exist, it hasn't been updated.
This change is not appropriate for MFC, although the polling shutdown
handler should be merged to 7-STABLE.
Bump __FreeBSD_version.
Reviewed by: bz
kernel option.
This also permits tuning of the option per virtual network stack, as
well as separately per inet, inet6.
The kernel option is left for a transition period, marked deprecated,
and will be removed soon.
Initially requested by: phk (1 year 1 day ago)
MFC after: 4 weeks
previously always pointing to the default vnet context, to a
dynamically changing thread-local one. The currvnet context
should be set on entry to networking code via CURVNET_SET() macros,
and reverted to previous state via CURVNET_RESTORE(). Recursions
on curvnet are permitted, though strongly discuouraged.
This change should have no functional impact on nooptions VIMAGE
kernel builds, where CURVNET_* macros expand to whitespace.
The curthread->td_vnet (aka curvnet) variable's purpose is to be an
indicator of the vnet context in which the current network-related
operation takes place, in case we cannot deduce the current vnet
context from any other source, such as by looking at mbuf's
m->m_pkthdr.rcvif->if_vnet, sockets's so->so_vnet etc. Moreover, so
far curvnet has turned out to be an invaluable consistency checking
aid: it helps to catch cases when sockets, ifnets or any other
vnet-aware structures may have leaked from one vnet to another.
The exact placement of the CURVNET_SET() / CURVNET_RESTORE() macros
was a result of an empirical iterative process, whith an aim to
reduce recursions on CURVNET_SET() to a minimum, while still reducing
the scope of CURVNET_SET() to networking only operations - the
alternative would be calling CURVNET_SET() on each system call entry.
In general, curvnet has to be set in three typicall cases: when
processing socket-related requests from userspace or from within the
kernel; when processing inbound traffic flowing from device drivers
to upper layers of the networking stack, and when executing
timer-driven networking functions.
This change also introduces a DDB subcommand to show the list of all
vnet instances.
Approved by: julian (mentor)
active network stack instance. Turning on options VIMAGE at compile
time yields the following changes relative to default kernel build:
1) V_ accessor macros for virtualized variables resolve to structure
fields via base pointers, instead of being resolved as fields in global
structs or plain global variables. As an example, V_ifnet becomes:
options VIMAGE: ((struct vnet_net *) vnet_net)->_ifnet
default build: vnet_net_0._ifnet
options VIMAGE_GLOBALS: ifnet
2) INIT_VNET_* macros will declare and set up base pointers to be used
by V_ accessor macros, instead of resolving to whitespace:
INIT_VNET_NET(ifp->if_vnet); becomes
struct vnet_net *vnet_net = (ifp->if_vnet)->mod_data[VNET_MOD_NET];
3) Memory for vnet modules registered via vnet_mod_register() is now
allocated at run time in sys/kern/kern_vimage.c, instead of per vnet
module structs being declared as globals. If required, vnet modules
can now request the framework to provide them with allocated bzeroed
memory by filling in the vmi_size field in their vmi_modinfo structures.
4) structs socket, ifnet, inpcbinfo, tcpcb and syncache_head are
extended to hold a pointer to the parent vnet. options VIMAGE builds
will fill in those fields as required.
5) curvnet is introduced as a new global variable in options VIMAGE
builds, always pointing to the default and only struct vnet.
6) struct sysctl_oid has been extended with additional two fields to
store major and minor virtualization module identifiers, oid_v_subs and
oid_v_mod. SYSCTL_V_* family of macros will fill in those fields
accordingly, and store the offset in the appropriate vnet container
struct in oid_arg1.
In sysctl handlers dealing with virtualized sysctls, the
SYSCTL_RESOLVE_V_ARG1() macro will compute the address of the target
variable and make it available in arg1 variable for further processing.
Unused fields in structs vnet_inet, vnet_inet6 and vnet_ipfw have
been deleted.
Reviewed by: bz, rwatson
Approved by: julian (mentor)
This has the effect that IPv6 multicast traffic won't trigger
an SPI allocation when IPSEC is in use, however, this obviously
needs to stomp on locks, and IN6_LOOKUP_MULTI() is about to go away.
This definitely needs to be revisited before 8.x is branched as
a release branch.
KAME had explicit checks at one point using it, so just hide it behind
#if 0 for now until we are sure if we can completely dump it or not.
MFC after: 1 month
dependency tracking and ordering enforcement.
With this change, per-vnet initialization functions introduced with
r190787 are no longer directly called from traditional initialization
functions (which cc in most cases inlined to pre-r190787 code), but are
instead registered via the vnet framework first, and are invoked only
after all prerequisite modules have been initialized. In the long run,
this framework should allow us to both initialize and dismantle
multiple vnet instances in a correct order.
The problem this change aims to solve is how to replay the
initialization sequence of various network stack components, which
have been traditionally triggered via different mechanisms (SYSINIT,
protosw). Note that this initialization sequence was and still can be
subtly different depending on whether certain pieces of code have been
statically compiled into the kernel, loaded as modules by boot
loader, or kldloaded at run time.
The approach is simple - we record the initialization sequence
established by the traditional mechanisms whenever vnet_mod_register()
is called for a particular vnet module. The vnet_mod_register_multi()
variant allows a single initializer function to be registered multiple
times but with different arguments - currently this is only used in
kern/uipc_domain.c by net_add_domain() with different struct domain *
as arguments, which allows for protosw-registered initialization
routines to be invoked in a correct order by the new vnet
initialization framework.
For the purpose of identifying vnet modules, each vnet module has to
have a unique ID, which is statically assigned in sys/vimage.h.
Dynamic assignment of vnet module IDs is not supported yet.
A vnet module may specify a single prerequisite module at registration
time by filling in the vmi_dependson field of its vnet_modinfo struct
with the ID of the module it depends on. Unless specified otherwise,
all vnet modules depend on VNET_MOD_NET (container for ifnet list head,
rt_tables etc.), which thus has to and will always be initialized
first. The framework will panic if it detects any unresolved
dependencies before completing system initialization. Detection of
unresolved dependencies for vnet modules registered after boot
(kldloaded modules) is not provided.
Note that the fact that each module can specify only a single
prerequisite may become problematic in the long run. In particular,
INET6 depends on INET being already instantiated, due to TCP / UDP
structures residing in INET container. IPSEC also depends on INET,
which will in turn additionally complicate making INET6-only kernel
configs a reality.
The entire registration framework can be compiled out by turning on the
VIMAGE_GLOBALS kernel config option.
Reviewed by: bz
Approved by: julian (mentor)
from existing functions for initializing global state.
At this stage, the new per-vnet initializer functions are
directly called from the existing global initialization code,
which should in most cases result in compiler inlining those
new functions, hence yielding a near-zero functional change.
Modify the existing initializer functions which are invoked via
protosw, like ip_init() et. al., to allow them to be invoked
multiple times, i.e. per each vnet. Global state, if any,
is initialized only if such functions are called within the
context of vnet0, which will be determined via the
IS_DEFAULT_VNET(curvnet) check (currently always true).
While here, V_irtualize a few remaining global UMA zones
used by net/netinet/netipsec networking code. While it is
not yet clear to me or anybody else whether this is the right
thing to do, at this stage this makes the code more readable,
and makes it easier to track uncollected UMA-zone-backed
objects on vnet removal. In the long run, it's quite possible
that some form of shared use of UMA zone pools among multiple
vnets should be considered.
Bump __FreeBSD_version due to changes in layout of structs
vnet_ipfw, vnet_inet and vnet_net.
Approved by: julian (mentor)
