Both functions need to obtain lock on the found PCB, and they can't do
classic inter-lock with the PCB hash lock, due to lock order reversal.
To keep the PCB stable, these functions put a reference on it and after PCB
lock is acquired drop it. If the reference was the last one, this means
we've raced with in_pcbfree() and the PCB is no longer valid.
This approach works okay only if we are acquiring writer-lock on the PCB.
In case of reader-lock, the following scenario can happen:
- 2 threads locate pcb, and do in_pcbref() on it.
- These 2 threads drop the inp hash lock.
- Another thread comes to delete pcb via in_pcbfree(), it obtains hash lock,
does in_pcbremlists(), drops hash lock, and runs in_pcbrele_wlocked(), which
doesn't free the pcb due to two references on it. Then it unlocks the pcb.
- 2 aforementioned threads acquire reader lock on the pcb and run
in_pcbrele_rlocked(). One gets 1 from in_pcbrele_rlocked() and continues,
second gets 0 and considers pcb freed, returns.
- The thread that got 1 continutes working with detached pcb, which later
leads to panic in the underlying protocol level.
To plumb that problem an additional INPCB flag introduced - INP_FREED. We
check for that flag in the in_pcbrele_rlocked() and if it is set, we pretend
that that was the last reference.
Discussed with: rwatson, jhb
Reported by: Vladimir Medvedkin <medved rambler-co.ru>
packets a cmsg of type IP_RECVTOS which contains the TOS byte.
Much like IP_RECVTTL does for TTL. This allows to implement a
protocol on top of UDP and implementing ECN.
MFC after: 3 days
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
The SYSCTL_NODE macro defines a list that stores all child-elements of
that node. If there's no SYSCTL_DECL macro anywhere else, there's no
reason why it shouldn't be static.
inp_socket->so_options dereference when we may not acquire the lock on
the inpcb.
This fixes the crash due to NULL pointer dereference in
in_pcbbind_setup() when inp_socket->so_options in a pcb returned by
in_pcblookup_local() was checked.
Reported by: dave jones <s.dave.jones@gmail.com>, Arnaud Lacombe <lacombar@gmail.com>
Suggested by: rwatson
Glanced by: rwatson
Tested by: dave jones <s.dave.jones@gmail.com>
struct inpcbgroup. pcbgroups, or "connection groups", supplement the
existing inpcbinfo connection hash table, which when pcbgroups are
enabled, might now be thought of more usefully as a per-protocol
4-tuple reservation table.
Connections are assigned to connection groups base on a hash of their
4-tuple; wildcard sockets require special handling, and are members
of all connection groups. During a connection lookup, a
per-connection group lock is employed rather than the global pcbinfo
lock. By aligning connection groups with input path processing,
connection groups take on an effective CPU affinity, especially when
aligned with RSS work placement (see a forthcoming commit for
details). This eliminates cache line migration associated with
global, protocol-layer data structures in steady state TCP and UDP
processing (with the exception of protocol-layer statistics; further
commit to follow).
Elements of this approach were inspired by Willman, Rixner, and Cox's
2006 USENIX paper, "An Evaluation of Network Stack Parallelization
Strategies in Modern Operating Systems". However, there are also
significant differences: we maintain the inpcb lock, rather than using
the connection group lock for per-connection state.
Likewise, the focus of this implementation is alignment with NIC
packet distribution strategies such as RSS, rather than pure software
strategies. Despite that focus, software distribution is supported
through the parallel netisr implementation, and works well in
configurations where the number of hardware threads is greater than
the number of NIC input queues, such as in the RMI XLR threaded MIPS
architecture.
Another important difference is the continued maintenance of existing
hash tables as "reservation tables" -- these are useful both to
distinguish the resource allocation aspect of protocol name management
and the more common-case lookup aspect. In configurations where
connection tables are aligned with hardware hashes, it is desirable to
use the traditional lookup tables for loopback or encapsulated traffic
rather than take the expense of hardware hashes that are hard to
implement efficiently in software (such as RSS Toeplitz).
Connection group support is enabled by compiling "options PCBGROUP"
into your kernel configuration; for the time being, this is an
experimental feature, and hence is not enabled by default.
Subject to the limited MFCability of change dependencies in inpcb,
and its change to the inpcbinfo init function signature, this change
in principle could be merged to FreeBSD 8.x.
Reviewed by: bz
Sponsored by: Juniper Networks, Inc.
hash install, etc. For now, these are arguments are unused, but as we add
RSS support, we will want to use hashes extracted from mbufs, rather than
manually calculated hashes of header fields, due to the expensive of the
software version of Toeplitz (and similar hashes).
Add notes that it would be nice to be able to pass mbufs into lookup
routines in pf(4), optimising firewall lookup in the same way, but the
code structure there doesn't facilitate that currently.
(In principle there is no reason this couldn't be MFCed -- the change
extends rather than modifies the KBI. However, it won't be useful without
other previous possibly less MFCable changes.)
Reviewed by: bz
Sponsored by: Juniper Networks, Inc.
- The existing ipi_lock continues to protect the global inpcb list and
inpcb counter. This lock is now relegated to a small number of
allocation and free operations, and occasional operations that walk
all connections (including, awkwardly, certain UDP multicast receive
operations -- something to revisit).
- A new ipi_hash_lock protects the two inpcbinfo hash tables for
looking up connections and bound sockets, manipulated using new
INP_HASH_*() macros. This lock, combined with inpcb locks, protects
the 4-tuple address space.
Unlike the current ipi_lock, ipi_hash_lock follows the individual inpcb
connection locks, so may be acquired while manipulating a connection on
which a lock is already held, avoiding the need to acquire the inpcbinfo
lock preemptively when a binding change might later be required. As a
result, however, lookup operations necessarily go through a reference
acquire while holding the lookup lock, later acquiring an inpcb lock --
if required.
A new function in_pcblookup() looks up connections, and accepts flags
indicating how to return the inpcb. Due to lock order changes, callers
no longer need acquire locks before performing a lookup: the lookup
routine will acquire the ipi_hash_lock as needed. In the future, it will
also be able to use alternative lookup and locking strategies
transparently to callers, such as pcbgroup lookup. New lookup flags are,
supplementing the existing INPLOOKUP_WILDCARD flag:
INPLOOKUP_RLOCKPCB - Acquire a read lock on the returned inpcb
INPLOOKUP_WLOCKPCB - Acquire a write lock on the returned inpcb
Callers must pass exactly one of these flags (for the time being).
Some notes:
- All protocols are updated to work within the new regime; especially,
TCP, UDPv4, and UDPv6. pcbinfo ipi_lock acquisitions are largely
eliminated, and global hash lock hold times are dramatically reduced
compared to previous locking.
- The TCP syncache still relies on the pcbinfo lock, something that we
may want to revisit.
- Support for reverting to the FreeBSD 7.x locking strategy in TCP input
is no longer available -- hash lookup locks are now held only very
briefly during inpcb lookup, rather than for potentially extended
periods. However, the pcbinfo ipi_lock will still be acquired if a
connection state might change such that a connection is added or
removed.
- Raw IP sockets continue to use the pcbinfo ipi_lock for protection,
due to maintaining their own hash tables.
- The interface in6_pcblookup_hash_locked() is maintained, which allows
callers to acquire hash locks and perform one or more lookups atomically
with 4-tuple allocation: this is required only for TCPv6, as there is no
in6_pcbconnect_setup(), which there should be.
- UDPv6 locking remains significantly more conservative than UDPv4
locking, which relates to source address selection. This needs
attention, as it likely significantly reduces parallelism in this code
for multithreaded socket use (such as in BIND).
- In the UDPv4 and UDPv6 multicast cases, we need to revisit locking
somewhat, as they relied on ipi_lock to stablise 4-tuple matches, which
is no longer sufficient. A second check once the inpcb lock is held
should do the trick, keeping the general case from requiring the inpcb
lock for every inpcb visited.
- This work reminds us that we need to revisit locking of the v4/v6 flags,
which may be accessed lock-free both before and after this change.
- Right now, a single lock name is used for the pcbhash lock -- this is
undesirable, and probably another argument is required to take care of
this (or a char array name field in the pcbinfo?).
This is not an MFC candidate for 8.x due to its impact on lookup and
locking semantics. It's possible some of these issues could be worked
around with compatibility wrappers, if necessary.
Reviewed by: bz
Sponsored by: Juniper Networks, Inc.
reworking of inpcbinfo locking:
(1) Convert inpcb reference counting from manually manipulated integers to
the refcount(9) KPI. This allows the refcount to be managed atomically
with an inpcb read lock rather than write lock, or even with no inpcb
lock at all. As a result, in_pcbref() also no longer requires an inpcb
lock, so can be performed solely using the lock used to look up an
inpcb.
(2) Shift more inpcb freeing activity from the in_pcbrele() context (via
in_pcbfree_internal) to the explicit in_pcbfree() context. This means
that the inpcb refcount is increasingly used only to maintain memory
stability, not actually defer the clean up of inpcb protocol parts.
This is desirable as many of those protocol parts required the pcbinfo
lock, which we'd like not to acquire in in_pcbrele() contexts. Document
this in comments better.
(3) Introduce new read-locked and write-locked in_pcbrele() variations,
in_pcbrele_rlocked() and in_pcbrele_wlocked(), which allow the inpcb to
be properly unlocked as needed. in_pcbrele() is a wrapper around the
latter, and should probably go away at some point. This makes it
easier to use this weak reference model when holding only a read lock,
as will happen in the future.
This may well be safe to MFC, but some more KBI analysis is required.
Reviewed by: bz
MFC after: 3 weeks
Sponsored by: Juniper Networks, Inc.
in_pcb_lport(), in_pcblookup_local(), and in_pcblookup_hash(), and similarly
for IPv6 functions. In the future, we would like to support other flags
relating to locking strategy.
This change doesn't appear to modify the KBI in practice, as callers already
passed in INPLOOKUP_WILDCARD rather than a simple boolean.
MFC after: 3 weeks
Reviewed by: bz
Sponsored by: Juniper Networks, Inc.
Move the ipport_tick_callout and related functions from ip_input.c
to in_pcb.c. The random source port allocation code has been merged
and is now local to in_pcb.c only.
Use a SYSINIT to get the callout started and no longer depend on
initialization from the inet code, which would not work in an IPv6
only setup.
Reviewed by: gnn
Sponsored by: The FreeBSD Foundation
Sponsored by: iXsystems
MFC after: 4 days
In some cases as udp6_connect() without an earlier bind(2) to an
address, v4-mapped scokets allowed and a non mapped destination
address, we can end up here with both v4 and v6 indicated:
inp_vflag = (INP_IPV4|INP_IPV6|INP_IPV6PROTO)
In that case however laddrp is NULL as the IPv6 path does not
pass in a copy currently.
Reported by: Pawel Worach (pawel.worach gmail.com)
Tested by: Pawel Worach (pawel.worach gmail.com)
MFC after: 6 days
X-MFC with: r219579
an unbound socket, regardless of any multicast options.
If an address is specified via a multicast option, then
let it override normal the source address selection.
This fixes a bug where source address selection was
not being performed when multicast options were present
but without an interface being specified.
Reviewed by: bz
MFC after: 1 day
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.
even if there is no route out to that mcast address. The code in
in_pcb inadvertantly would error (no route) even though
the user may have specified the address with the
proper socket option (to specify the egress interface).
Thanks bz for reminding me I forgot to commit this ;-)
Reviewed by: bz
MFC after: 1 week
their calling contexts in {IP divert, raw IP sockets, TCP, UDP} and
create new helper functions: in_pcbinfo_init() and in_pcbinfo_destroy()
to do this work in a central spot. As inpcbinfo becomes more complex
due to ongoing work to add connection groups, this will reduce code
duplication.
MFC after: 1 month
Reviewed by: bz
Sponsored by: Juniper Networks
whether to use source address selection (default) or the primary
jail address for unbound outgoing connections.
This is intended to be used by people upgrading from single-IP
jails to multi-IP jails but not having to change firewall rules,
application ACLs, ... but to force their connections (unless
otherwise changed) to the primry jail IP they had been used for
years, as well as for people prefering to implement similar policies.
Note that for IPv6, if configured incorrectly, this might lead to
scope violations, which single-IPv6 jails could as well, as by the
design of jails. [1]
Reviewed by: jamie, hrs (ipv6 part)
Pointed out by: hrs [1]
MFC After: 2 weeks
Asked for by: Jase Thew (bazerka beardz.net)
within the system that owns the interface. Packets destined to
the local end point leak to the wire towards the default gateway
if one exists. This behavior is changed as part of the L2/L3
rewrite efforts. The local end point is now reachable within the
system. The inpcb code needs to consider this fact during the
address selection process.
Reviewed by: bz
MFC after: immediately
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)
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)
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
rather than pointers, requiring callers to properly dispose of those
references. The following routines now return references:
ifaddr_byindex
ifa_ifwithaddr
ifa_ifwithbroadaddr
ifa_ifwithdstaddr
ifa_ifwithnet
ifaof_ifpforaddr
ifa_ifwithroute
ifa_ifwithroute_fib
rt_getifa
rt_getifa_fib
IFP_TO_IA
ip_rtaddr
in6_ifawithifp
in6ifa_ifpforlinklocal
in6ifa_ifpwithaddr
in6_ifadd
carp_iamatch6
ip6_getdstifaddr
Remove unused macro which didn't have required referencing:
IFP_TO_IA6
This closes many small races in which changes to interface
or address lists while an ifaddr was in use could lead to use of freed
memory (etc). In a few cases, add missing if_addr_list locking
required to safely acquire references.
Because of a lack of deep copying support, we accept a race in which
an in6_ifaddr pointed to by mbuf tags and extracted with
ip6_getdstifaddr() doesn't hold a reference while in transmit. Once
we have mbuf tag deep copy support, this can be fixed.
Reviewed by: bz
Obtained from: Apple, Inc. (portions)
MFC after: 6 weeks (portions)
a pointer to an ifaddr matching the passed socket address, returns a
boolean indicating whether one was present. In the (near) future,
ifa_ifwithaddr() will return a referenced ifaddr rather than a raw
ifaddr pointer, and the new wrapper will allow callers that care only
about the boolean condition to avoid having to free that reference.
MFC after: 3 weeks
and used in a large number of files, but also because an increasing number
of incorrect uses of MAC calls were sneaking in due to copy-and-paste of
MAC-aware code without the associated opt_mac.h include.
Discussed with: pjd
modules are loaded by avoiding mbuf label lookups when policies aren't
loaded, pushing further socket locking into MAC policy modules, and
avoiding locking MAC ifnet locks when no policies are loaded:
- Check mac_policies_count before looking for mbuf MAC label m_tags in MAC
Framework entry points. We will still pay label lookup costs if MAC
policies are present but don't require labels (typically a single mbuf
header field read, but perhaps further indirection if IPSEC or other
m_tag consumers are in use).
- Further push socket locking for socket-related access control checks and
events into MAC policies from the MAC Framework, so that sockets are
only locked if a policy specifically requires a lock to protect a label.
This resolves lock order issues during sonewconn() and also in local
domain socket cross-connect where multiple socket locks could not be
held at once for the purposes of propagatig MAC labels across multiple
sockets. Eliminate mac_policy_count check in some entry points where it
no longer avoids locking.
- Add mac_policy_count checking in some entry points relating to network
interfaces that otherwise lock a global MAC ifnet lock used to protect
ifnet labels.
Obtained from: TrustedBSD Project
with OpenBSD (and BSD/OS originally). We can't easly do it SOL_SOCKET option
as there is no more space for more SOL_SOCKET options, but this option also
fits better as an IP socket option, it seems.
- Implement this functionality also for IPv6 and RAW IP sockets.
- Always compile it in (don't use additional kernel options).
- Remove sysctl to turn this functionality on and off.
- Introduce new privilege - PRIV_NETINET_BINDANY, which allows to use this
functionality (currently only unjail root can use it).
Discussed with: julian, adrian, jhb, rwatson, kmacy
by creating a child jail, which is visible to that jail and to any
parent jails. Child jails may be restricted more than their parents,
but never less. Jail names reflect this hierarchy, being MIB-style
dot-separated strings.
Every thread now points to a jail, the default being prison0, which
contains information about the physical system. Prison0's root
directory is the same as rootvnode; its hostname is the same as the
global hostname, and its securelevel replaces the global securelevel.
Note that the variable "securelevel" has actually gone away, which
should not cause any problems for code that properly uses
securelevel_gt() and securelevel_ge().
Some jail-related permissions that were kept in global variables and
set via sysctls are now per-jail settings. The sysctls still exist for
backward compatibility, used only by the now-deprecated jail(2) system
call.
Approved by: bz (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)
certain flags that should have been in inp_flags ended up in inp_vflag,
meaning that they were inconsistently locked, and in one case,
interpreted. Move the following flags from inp_vflag to gaps in the
inp_flags space (and clean up the inp_flags constants to make gaps
more obvious to future takers):
INP_TIMEWAIT
INP_SOCKREF
INP_ONESBCAST
INP_DROPPED
Some aspects of this change have no effect on kernel ABI at all, as these
are UDP/TCP/IP-internal uses; however, netstat and sockstat detect
INP_TIMEWAIT when listing TCP sockets, so any MFC will need to take this
into account.
MFC after: 1 week (or after dependencies are MFC'd)
Reviewed by: bz
or not the inpcb is currenty on various hash lookup lists, rather
than using (lport != 0) to detect this. This means that the full
4-tuple of a connection can be retained after close, which should
lead to more sensible netstat output in the window between TCP
close and socket close.
MFC after: 2 weeks
return zero on success and an error code otherwise. The possible errors
are EADDRNOTAVAIL if an address being checked for doesn't match the
prison, and EAFNOSUPPORT if the prison doesn't have any addresses in
that address family. For most callers of these functions, use the
returned error code instead of e.g. a hard-coded EADDRNOTAVAIL or
EINVAL.
Always include a jailed() check in these functions, where a non-jailed
cred always returns success (and makes no changes). Remove the explicit
jailed() checks that preceded many of the function calls.
Approved by: bz (mentor)
