that protects socket and receive socket buffer state, and a second
mutex to protect send socket buffer state. In some places, the
mutex shared between the socket and receive socket buffer will be
acquired twice, once by each layer, resulting in some
inconsistency, but providing the abstraction benefit of being able
to more easily separate the two mutexes in the future if desired.
When transitioning a socket to the SS_ISDISCONNECTING or
SS_ISDISCONNECTED states, grab the socket/receive socket buffer lock
once rather than grabbing it as the socket lock, modifying socket
state, then grabbing a second time as the receive lock in order to
modify the socket buffer state to indicate no further data can be
read. This change is believed to close a race between the change in
socket state and the change in socket buffer state, which for a
remotely initiated close on a UNIX domain socket, resulted in
soreceive() returning ENOTCONN rather than an EOF condition.
A similar race still exists in the case of send, however, and is
harder to fix as the socket and send socket buffer mutexes are not
the same, and we would like to avoid holding combinations of socket
mutexes over sb_upcall until we've finished clarifying the locking
protocol for upcalls.
This change has the side affect of reducing the number of mutex
operations to initiate disconnect or perform disconnect on a
socket by two.
PR: 78824
Rerported by: Marc Olzheim <marcolz@stack.nl>
MFC after: 2 weeks
so that the socket lock is held over the test-and-set removal of the
accept filter option during connect, and the two socket mutex regions
(transition to connected, perform accept filter) are combined.
connection status before inserting the new socket into the listen
socket's accept queue, or there might be a race in which another thread
wakes up when the accept lock is released, and sees the socket before its
state is set correctly. The wakeup still occurs after the accept lock is
released. There have been no diagnoses of this bug in real-world systems
(as yet).
MFC after: 3 days
families.
The protosw[] array of any particular protocol family ("domain") is of fixed size
defined at compile time. This made it impossible to dynamically add or remove any
protocols to or from it. We work around this by introducing so called SPACER's
which are embedded into the protosw[] array at compile time. The SPACER's have
a special protocol number (32767) to indicate the fact that they are SPACER's but
are otherwise NULL. Only as many protocols can be dynamically loaded as SPACER's
are provided in the protosw[] structure.
The pr_usrreqs structure is treated more special and contains pointers to dummy
functions only returning EOPNOTSUPP. This is needed because the use of those
functions pointers is usually not checked within the kernel because until now it
was assumed to be a valid function pointer. Instead of fixing all potential
callers we just return a proper error code.
Two new functions provide a clean API to register and unregister a protocol. The
register function expects a pointer to a valid and complete struct protosw including
a pointer to struct pru_usrreqs provided by the caller. Upon successful registration
the pr_init() function will be called to finish initialization of the protocol. The
unregister function restores the SPACER in place of the protocol again. It is the
responseability of the caller to ensure proper closing of all sockets and freeing
of memory allocation by the unloading protocol.
sys/protosw.h
o Define generic PROTO_SPACER to be 32767
o Prototypes for all pru_*_notsupp() functions
o Prototypes for pf_proto_[un]register() functions
kern/uipc_domain.c
o Global struct pr_usrreqs nousrreqs containing valid pointers to the
pru_*_notsupp() functions
o New functions pf_proto_[un]register()
kern/uipc_socket2.c
o New functions bodies for all pru_*_notsupp() functions
a more complete subsystem, and removes the knowlege of how things are
implemented from the drivers. Include locking around filter ops, so a
module like aio will know when not to be unloaded if there are outstanding
knotes using it's filter ops.
Currently, it uses the MTX_DUPOK even though it is not always safe to
aquire duplicate locks. Witness currently doesn't support the ability
to discover if a dup lock is ok (in some cases).
Reviewed by: green, rwatson (both earlier versions)
associated with performing a wakeup on the socket buffer:
- When performing an sbappend*() followed by a so[rw]wakeup(), explicitly
acquire the socket buffer lock and use the _locked() variants of both
calls. Note that the _locked() sowakeup() versions unlock the mutex on
return. This is done in uipc_send(), divert_packet(), mroute
socket_send(), raw_append(), tcp_reass(), tcp_input(), and udp_append().
- When the socket buffer lock is dropped before a sowakeup(), remove the
explicit unlock and use the _locked() sowakeup() variant. This is done
in soisdisconnecting(), soisdisconnected() when setting the can't send/
receive flags and dropping data, and in uipc_rcvd() which adjusting
back-pressure on the sockets.
For UNIX domain sockets running mpsafe with a contention-intensive SMP
mysql benchmark, this results in a 1.6% query rate improvement due to
reduce mutex costs.
the socket buffer having its limits adjusted. sbreserve() now acquires
the lock before calling sbreserve_locked(). In soreserve(), acquire
socket buffer locks across read-modify-writes of socket buffer fields,
and calls into sbreserve/sbrelease; make sure to acquire in keeping
with the socket buffer lock order. In tcp_mss(), acquire the socket
buffer lock in the calling context so that we have atomic read-modify
-write on buffer sizes.
- sowakeup() now asserts the socket buffer lock on entry. Move
the call to KNOTE higher in sowakeup() so that it is made with
the socket buffer lock held for consistency with other calls.
Release the socket buffer lock prior to calling into pgsigio(),
so_upcall(), or aio_swake(). Locking for this event management
will need revisiting in the future, but this model avoids lock
order reversals when upcalls into other subsystems result in
socket/socket buffer operations. Assert that the socket buffer
lock is not held at the end of the function.
- Wrapper macros for sowakeup(), sorwakeup() and sowwakeup(), now
have _locked versions which assert the socket buffer lock on
entry. If a wakeup is required by sb_notify(), invoke
sowakeup(); otherwise, unconditionally release the socket buffer
lock. This results in the socket buffer lock being released
whether a wakeup is required or not.
- Break out socantsendmore() into socantsendmore_locked() that
asserts the socket buffer lock. socantsendmore()
unconditionally locks the socket buffer before calling
socantsendmore_locked(). Note that both functions return with
the socket buffer unlocked as socantsendmore_locked() calls
sowwakeup_locked() which has the same properties. Assert that
the socket buffer is unlocked on return.
- Break out socantrcvmore() into socantrcvmore_locked() that
asserts the socket buffer lock. socantrcvmore() unconditionally
locks the socket buffer before calling socantrcvmore_locked().
Note that both functions return with the socket buffer unlocked
as socantrcvmore_locked() calls sorwakeup_locked() which has
similar properties. Assert that the socket buffer is unlocked
on return.
- Break out sbrelease() into a sbrelease_locked() that asserts the
socket buffer lock. sbrelease() unconditionally locks the
socket buffer before calling sbrelease_locked().
sbrelease_locked() now invokes sbflush_locked() instead of
sbflush().
- Assert the socket buffer lock in socket buffer sanity check
functions sblastrecordchk(), sblastmbufchk().
- Assert the socket buffer lock in SBLINKRECORD().
- Break out various sbappend() functions into sbappend_locked()
(and variations on that name) that assert the socket buffer
lock. The !_locked() variations unconditionally lock the socket
buffer before calling their _locked counterparts. Internally,
make sure to call _locked() support routines, etc, if already
holding the socket buffer lock.
- Break out sbinsertoob() into sbinsertoob_locked() that asserts
the socket buffer lock. sbinsertoob() unconditionally locks the
socket buffer before calling sbinsertoob_locked().
- Break out sbflush() into sbflush_locked() that asserts the
socket buffer lock. sbflush() unconditionally locks the socket
buffer before calling sbflush_locked(). Update panic strings
for new function names.
- Break out sbdrop() into sbdrop_locked() that asserts the socket
buffer lock. sbdrop() unconditionally locks the socket buffer
before calling sbdrop_locked().
- Break out sbdroprecord() into sbdroprecord_locked() that asserts
the socket buffer lock. sbdroprecord() unconditionally locks
the socket buffer before calling sbdroprecord_locked().
- sofree() now calls socantsendmore_locked() and re-acquires the
socket buffer lock on return. It also now calls
sbrelease_locked().
- sorflush() now calls socantrcvmore_locked() and re-acquires the
socket buffer lock on return. Clean up/mess up other behavior
in sorflush() relating to the temporary stack copy of the socket
buffer used with dom_dispose by more properly initializing the
temporary copy, and selectively bzeroing/copying more carefully
to prevent WITNESS from getting confused by improperly
initialized mutexes. Annotate why that's necessary, or at
least, needed.
- soisconnected() now calls sbdrop_locked() before unlocking the
socket buffer to avoid locking overhead.
Some parts of this change were:
Submitted by: sam
Sponsored by: FreeBSD Foundation
Obtained from: BSD/OS
lock state. Convert tsleep() into msleep() with socket buffer mutex
as argument. Hold socket buffer lock over sbunlock() to protect sleep
lock state.
Assert socket buffer lock in sbwait() to protect the socket buffer
wait state. Convert tsleep() into msleep() with socket buffer mutex
as argument.
Modify sofree(), sosend(), and soreceive() to acquire SOCKBUF_LOCK()
in order to call into these functions with the lock, as well as to
start protecting other socket buffer use in their implementation. Drop
the socket buffer mutexes around calls into the protocol layer, around
potentially blocking operations, for copying to/from user space, and
VM operations relating to zero-copy. Assert the socket buffer mutex
strategically after code sections or at the beginning of loops. In
some cases, modify return code to ensure locks are properly dropped.
Convert the potentially blocking allocation of storage for the remote
address in soreceive() into a non-blocking allocation; we may wish to
move the allocation earlier so that it can block prior to acquisition
of the socket buffer lock.
Drop some spl use.
NOTE: Some races exist in the current structuring of sosend() and
soreceive(). This commit only merges basic socket locking in this
code; follow-up commits will close additional races. As merged,
these changes are not sufficient to run without Giant safely.
Reviewed by: juli, tjr
- Lock down low hanging fruit use of sb_flags with socket buffer
lock.
- Lock down low hanging fruit use of so_state with socket lock.
- Lock down low hanging fruit use of so_options.
- Lock down low-hanging fruit use of sb_lowwat and sb_hiwat with
socket buffer lock.
- Annotate situations in which we unlock the socket lock and then
grab the receive socket buffer lock, which are currently actually
the same lock. Depending on how we want to play our cards, we
may want to coallesce these lock uses to reduce overhead.
- Convert a if()->panic() into a KASSERT relating to so_state in
soaccept().
- Remove a number of splnet()/splx() references.
More complex merging of socket and socket buffer locking to
follow.
flags relating to several aspects of socket functionality. This change
breaks out several bits relating to send and receive operation into a
new per-socket buffer field, sb_state, in order to facilitate locking.
This is required because, in order to provide more granular locking of
sockets, different state fields have different locking properties. The
following fields are moved to sb_state:
SS_CANTRCVMORE (so_state)
SS_CANTSENDMORE (so_state)
SS_RCVATMARK (so_state)
Rename respectively to:
SBS_CANTRCVMORE (so_rcv.sb_state)
SBS_CANTSENDMORE (so_snd.sb_state)
SBS_RCVATMARK (so_rcv.sb_state)
This facilitates locking by isolating fields to be located with other
identically locked fields, and permits greater granularity in socket
locking by avoiding storing fields with different locking semantics in
the same short (avoiding locking conflicts). In the future, we may
wish to coallesce sb_state and sb_flags; for the time being I leave
them separate and there is no additional memory overhead due to the
packing/alignment of shorts in the socket buffer structure.
SOCK_LOCK(so):
- Hold socket lock over calls to MAC entry points reading or
manipulating socket labels.
- Assert socket lock in MAC entry point implementations.
- When externalizing the socket label, first make a thread-local
copy while holding the socket lock, then release the socket lock
to externalize to userspace.
global mutex, accept_mtx, which serializes access to the following
fields across all sockets:
so_qlen so_incqlen so_qstate
so_comp so_incomp so_list
so_head
While providing only coarse granularity, this approach avoids lock
order issues between sockets by avoiding ownership of the fields
by a specific socket and its per-socket mutexes.
While here, rewrite soclose(), sofree(), soaccept(), and
sonewconn() to add assertions, close additional races and address
lock order concerns. In particular:
- Reorganize the optimistic concurrency behavior in accept1() to
always allocate a file descriptor with falloc() so that if we do
find a socket, we don't have to encounter the "Oh, there wasn't
a socket" race that can occur if falloc() sleeps in the current
code, which broke inbound accept() ordering, not to mention
requiring backing out socket state changes in a way that raced
with the protocol level. We may want to add a lockless read of
the queue state if polling of empty queues proves to be important
to optimize.
- In accept1(), soref() the socket while holding the accept lock
so that the socket cannot be free'd in a race with the protocol
layer. Likewise in netgraph equivilents of the accept1() code.
- In sonewconn(), loop waiting for the queue to be small enough to
insert our new socket once we've committed to inserting it, or
races can occur that cause the incomplete socket queue to
overfill. In the previously implementation, it was sufficient
to simply tested once since calling soabort() didn't release
synchronization permitting another thread to insert a socket as
we discard a previous one.
- In soclose()/sofree()/et al, it is the responsibility of the
caller to remove a socket from the incomplete connection queue
before calling soabort(), which prevents soabort() from having
to walk into the accept socket to release the socket from its
queue, and avoids races when releasing the accept mutex to enter
soabort(), permitting soabort() to avoid lock ordering issues
with the caller.
- Generally cluster accept queue related operations together
throughout these functions in order to facilitate locking.
Annotate new locking in socketvar.h.
the socket is on an accept queue of a listen socket. This change
renames the flags to SQ_COMP and SQ_INCOMP, and moves them to a new
state field on the socket, so_qstate, as the locking for these flags
is substantially different for the locking on the remainder of the
flags in so_state.
mbuma is an Mbuf & Cluster allocator built on top of a number of
extensions to the UMA framework, all included herein.
Extensions to UMA worth noting:
- Better layering between slab <-> zone caches; introduce
Keg structure which splits off slab cache away from the
zone structure and allows multiple zones to be stacked
on top of a single Keg (single type of slab cache);
perhaps we should look into defining a subset API on
top of the Keg for special use by malloc(9),
for example.
- UMA_ZONE_REFCNT zones can now be added, and reference
counters automagically allocated for them within the end
of the associated slab structures. uma_find_refcnt()
does a kextract to fetch the slab struct reference from
the underlying page, and lookup the corresponding refcnt.
mbuma things worth noting:
- integrates mbuf & cluster allocations with extended UMA
and provides caches for commonly-allocated items; defines
several zones (two primary, one secondary) and two kegs.
- change up certain code paths that always used to do:
m_get() + m_clget() to instead just use m_getcl() and
try to take advantage of the newly defined secondary
Packet zone.
- netstat(1) and systat(1) quickly hacked up to do basic
stat reporting but additional stats work needs to be
done once some other details within UMA have been taken
care of and it becomes clearer to how stats will work
within the modified framework.
From the user perspective, one implication is that the
NMBCLUSTERS compile-time option is no longer used. The
maximum number of clusters is still capped off according
to maxusers, but it can be made unlimited by setting
the kern.ipc.nmbclusters boot-time tunable to zero.
Work should be done to write an appropriate sysctl
handler allowing dynamic tuning of kern.ipc.nmbclusters
at runtime.
Additional things worth noting/known issues (READ):
- One report of 'ips' (ServeRAID) driver acting really
slow in conjunction with mbuma. Need more data.
Latest report is that ips is equally sucking with
and without mbuma.
- Giant leak in NFS code sometimes occurs, can't
reproduce but currently analyzing; brueffer is
able to reproduce but THIS IS NOT an mbuma-specific
problem and currently occurs even WITHOUT mbuma.
- Issues in network locking: there is at least one
code path in the rip code where one or more locks
are acquired and we end up in m_prepend() with
M_WAITOK, which causes WITNESS to whine from within
UMA. Current temporary solution: force all UMA
allocations to be M_NOWAIT from within UMA for now
to avoid deadlocks unless WITNESS is defined and we
can determine with certainty that we're not holding
any locks when we're M_WAITOK.
- I've seen at least one weird socketbuffer empty-but-
mbuf-still-attached panic. I don't believe this
to be related to mbuma but please keep your eyes
open, turn on debugging, and capture crash dumps.
This change removes more code than it adds.
A paper is available detailing the change and considering
various performance issues, it was presented at BSDCan2004:
http://www.unixdaemons.com/~bmilekic/netbuf_bmilekic.pdf
Please read the paper for Future Work and implementation
details, as well as credits.
Testing and Debugging:
rwatson,
brueffer,
Ketrien I. Saihr-Kesenchedra,
...
Reviewed by: Lots of people (for different parts)
functions in kern_socket.c.
Rename the "canwait" field to "mflags" and pass M_WAITOK and M_NOWAIT
in from the caller context rather than "1" or "0".
Correct mflags pass into mac_init_socket() from previous commit to not
include M_ZERO.
Submitted by: sam
than a "waitok" argument. Callers now passing M_WAITOK or M_NOWAIT
rather than 0 or 1. This simplifies the soalloc() logic, and also
makes the waiting behavior of soalloc() more clear in the calling
context.
Submitted by: sam
- struct plimit includes a mutex to protect a reference count. The plimit
structure is treated similarly to struct ucred in that is is always copy
on write, so having a reference to a structure is sufficient to read from
it without needing a further lock.
- The proc lock protects the p_limit pointer and must be held while reading
limits from a process to keep the limit structure from changing out from
under you while reading from it.
- Various global limits that are ints are not protected by a lock since
int writes are atomic on all the archs we support and thus a lock
wouldn't buy us anything.
- All accesses to individual resource limits from a process are abstracted
behind a simple lim_rlimit(), lim_max(), and lim_cur() API that return
either an rlimit, or the current or max individual limit of the specified
resource from a process.
- dosetrlimit() was renamed to kern_setrlimit() to match existing style of
other similar syscall helper functions.
- The alpha OSF/1 compat layer no longer calls getrlimit() and setrlimit()
(it didn't used the stackgap when it should have) but uses lim_rlimit()
and kern_setrlimit() instead.
- The svr4 compat no longer uses the stackgap for resource limits calls,
but uses lim_rlimit() and kern_setrlimit() instead.
- The ibcs2 compat no longer uses the stackgap for resource limits. It
also no longer uses the stackgap for accessing sysctl's for the
ibcs2_sysconf() syscall but uses kernel_sysctl() instead. As a result,
ibcs2_sysconf() no longer needs Giant.
- The p_rlimit macro no longer exists.
Submitted by: mtm (mostly, I only did a few cleanups and catchups)
Tested on: i386
Compiled on: alpha, amd64
the MAC label referenced from 'struct socket' in the IPv4 and
IPv6-based protocols. This permits MAC labels to be checked during
network delivery operations without dereferencing inp->inp_socket
to get to so->so_label, which will eventually avoid our having to
grab the socket lock during delivery at the network layer.
This change introduces 'struct inpcb' as a labeled object to the
MAC Framework, along with the normal circus of entry points:
initialization, creation from socket, destruction, as well as a
delivery access control check.
For most policies, the inpcb label will simply be a cache of the
socket label, so a new protocol switch method is introduced,
pr_sosetlabel() to notify protocols that the socket layer label
has been updated so that the cache can be updated while holding
appropriate locks. Most protocols implement this using
pru_sosetlabel_null(), but IPv4/IPv6 protocols using inpcbs use
the the worker function in_pcbsosetlabel(), which calls into the
MAC Framework to perform a cache update.
Biba, LOMAC, and MLS implement these entry points, as do the stub
policy, and test policy.
Reviewed by: sam, bms
Obtained from: TrustedBSD Project
Sponsored by: DARPA, Network Associates Laboratories
thread being waken up. The thread waken up can run at a priority as
high as after tsleep().
- Replace selwakeup()s with selwakeuppri()s and pass appropriate
priorities.
- Add cv_broadcastpri() which raises the priority of the broadcast
threads. Used by selwakeuppri() if collision occurs.
Not objected in: -arch, -current
manipulated directly (rather than using sballoc()/sbfree()); update them
to tweak the new sb_ctl field too.
Sponsored by: NTT Multimedia Communications Labs
of the adjusted sb_max into a sysctl handler for sb_max and assigning it to
a variable that is used instead. This eliminates the 32bit multiply and
divide from the fast path that was being done previously.
expensive (!) 64bit multiply, divide, and comparison aren't necessary
(this came in originally from rev 1.19 to fix an overflow with large
sb_max or MCLBYTES).
The 64bit math in this function was measured in some kernel profiles as
being as much as 5-8% of the total overhead of the TCP/IP stack and
is eliminated with this commit. There is a harmless rounding error (of
about .4% with the standard values) introduced with this change,
however this is in the conservative direction (downward toward a
slightly smaller maximum socket buffer size).
MFC after: 3 days
kernel access control.
Invoke the necessary MAC entry points to maintain labels on sockets.
In particular, invoke entry points during socket allocation and
destruction, as well as creation by a process or during an
accept-scenario (sonewconn). For UNIX domain sockets, also assign
a peer label. As the socket code isn't locked down yet, locking
interactions are not yet clear. Various protocol stack socket
operations (such as peer label assignment for IPv4) will follow.
Obtained from: TrustedBSD Project
Sponsored by: DARPA, NAI Labs
not responding) then drop any data on the outgoing queue in
soisdisconnected because there is no way to get it to its destination
any longer.
The only objection to this patch I got on -net was from Terry, who
wasn't sure that the condition in question could arise, so I provided
some example code.
