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.
initialized socket with no qlimit was being passed in. In order
to handle this case properly, we must not use >= when comparing
queue sizes to qlimit. As a result of this improper handling,
a panic could result in certain cases.
PR: 38325
MFC after: 3 days
o Add a mutex (sb_mtx) to struct sockbuf. This protects the data in a
socket buffer. The mutex in the receive buffer also protects the data
in struct socket.
o Determine the lock strategy for each members in struct socket.
o Lock down the following members:
- so_count
- so_options
- so_linger
- so_state
o Remove *_locked() socket APIs. Make the following socket APIs
touching the members above now require a locked socket:
- sodisconnect()
- soisconnected()
- soisconnecting()
- soisdisconnected()
- soisdisconnecting()
- sofree()
- soref()
- sorele()
- sorwakeup()
- sotryfree()
- sowakeup()
- sowwakeup()
Reviewed by: alfred
Turn the sigio sx into a mutex.
Sigio lock is really only needed to protect interrupts from dereferencing
the sigio pointer in an object when the sigio itself is being destroyed.
In order to do this in the most unintrusive manner change pgsigio's
sigio * argument into a **, that way we can lock internally to the
function.
Requested by: bde
Since locking sigio_lock is usually followed by calling pgsigio(),
move the declaration of sigio_lock and the definitions of SIGIO_*() to
sys/signalvar.h.
While I am here, sort include files alphabetically, where possible.
of a socket. This avoids lock order reversal caused by locking a
process in pgsigio().
sowakeup() and the callers of it (sowwakeup, soisconnected, etc.) now
require sigio_lock to be locked. Provide sowwakeup_locked(),
soisconnected_locked(), and so on in case where we have to modify a
socket and wake up a process atomically.
LRU fashion when the listen queue fills up. Previously, there was
no mechanism to kick out old sockets, leading to an easy DoS of
daemons using accept filtering.
Reviewed by: alfred
MFC after: 3 days
Remove the explicit call to aio_proc_rundown() from exit1(), instead AIO
will use at_exit(9).
Add functions at_exec(9), rm_at_exec(9) which function nearly the
same as at_exec(9) and rm_at_exec(9), these functions are called
on behalf of modules at the time of execve(2) after the image
activator has run.
Use a modified version of tegge's suggestion via at_exec(9) to close
an exploitable race in AIO.
Fix SYSCALL_MODULE_HELPER such that it's archetecuterally neutral,
the problem was that one had to pass it a paramater indicating the
number of arguments which were actually the number of "int". Fix
it by using an inline version of the AS macro against the syscall
arguments. (AS should be available globally but we'll get to that
later.)
Add a primative system for dynamically adding kqueue ops, it's really
not as sophisticated as it should be, but I'll discuss with jlemon when
he's around.
code only passed up the connection to the tcp stack when it was complete,
so it went directly into the so_comp (complete) queue. However, with
accept filters, there is an additional phase before calling it "complete".
Reviewed by: jlemon
always deriving the credential for a newly accepted connection from
the listen socket. Previously, the selection of the credential
depended on the protocol: UNIX domain sockets would use the
connecting process's credential, and protocols supporting a creation
of the socket before the receiving end called accept() would use
the listening socket. After this change, it is always the listening
credential.
Reviewed by: green
vnodes. This will hopefully serve as a base from which we can
expand the MP code. We currently do not attempt to obtain any
mutex or SX locks, but the door is open to add them when we nail
down exactly how that part of it is going to work.
- crhold() returns a reference to the ucred whose refcount it bumps.
- crcopy() now simply copies the credentials from one credential to
another and has no return value.
- a new crshared() primitive is added which returns true if a ucred's
refcount is > 1 and false (0) otherwise.
Note ALL MODULES MUST BE RECOMPILED
make the kernel aware that there are smaller units of scheduling than the
process. (but only allow one thread per process at this time).
This is functionally equivalent to teh previousl -current except
that there is a thread associated with each process.
Sorry john! (your next MFC will be a doosie!)
Reviewed by: peter@freebsd.org, dillon@freebsd.org
X-MFC after: ha ha ha ha
around, use a common function for looking up and extracting the tunables
from the kernel environment. This saves duplicating the same function
over and over again. This way typically has an overhead of 8 bytes + the
path string, versus about 26 bytes + the path string.
other "system" header files.
Also help the deprecation of lockmgr.h by making it a sub-include of
sys/lock.h and removing sys/lockmgr.h form kernel .c files.
Sort sys/*.h includes where possible in affected files.
OK'ed by: bde (with reservations)
could not compress into clusters. This could result in lots of
wasted clusters while recieving small packets from an interface
that uses clusters for all it's packets.
Patch is partially from BSDi (limiting the size of the copy) and
based on a patch for 4.1 by Ian Dowse <iedowse@maths.tcd.ie> and
myself.
Reviewed by: bmilekic
Obtained From: BSDi
Submitted by: iedowse
chgsbsize(), which are called rather frequently and may be called from an
interrupt context in the case of chgsbsize(). Instead, do the hash table
lookup and maintenance when credentials are changed, which is a lot less
frequent. Add pointers to the uidinfo structures to the ucred and pcred
structures for fast access. Pass a pointer to the credential to chgproccnt()
and chgsbsize() instead of passing the uid. Add a reference count to the
uidinfo structure and use it to decide when to free the structure rather
than freeing the structure when the resource consumption drops to zero.
Move the resource tracking code from kern_proc.c to kern_resource.c. Move
some duplicate code sequences in kern_prot.c to separate helper functions.
Change KASSERTs in this code to unconditional tests and calls to panic().
the chgsbsize() call to use a "subject" pointer (&sb.sb_hiwat) and
a u_long target to set it to. The whole thing is splnet().
This fixes a problem that jdp has been able to provoke.
1) while allocating a uidinfo struct malloc is called with M_WAITOK,
it's possible that while asleep another process by the same user
could have woken up earlier and inserted an entry into the uid
hash table. Having redundant entries causes inconsistancies that
we can't handle.
fix: do a non-waiting malloc, and if that fails then do a blocking
malloc, after waking up check that no one else has inserted an entry
for us already.
2) Because many checks for sbsize were done as "test then set" in a non
atomic manner it was possible to exceed the limits put up via races.
fix: instead of querying the count then setting, we just attempt to
set the count and leave it up to the function to return success or
failure.
3) The uidinfo code was inlining and repeating, lookups and insertions
and deletions needed to be in their own functions for clarity.
Reviewed by: green
accept filters are now loadable as well as able to be compiled into
the kernel.
two accept filters are provided, one that returns sockets when data
arrives the other when an http request is completed (doesn't work
with 0.9 requests)
Reviewed by: jmg
until the incoming connection has either data waiting or what looks like a
HTTP request header already in the socketbuffer. This ought to reduce
the context switch time and overhead for processing requests.
The initial idea and code for HTTPACCEPT came from Yahoo engineers and has
been cleaned up and a more lightweight DELAYACCEPT for non-http servers
has been added
Reviewed by: silence on hackers.
Now this check is necessary because IPv6 source routing might use
control data bigger than MLEN. (e.g. 16bytes IPv6 addr x 23 hops)
Actually mbuf cluster should be used in uipc_socket.c:sbcreatecontrol()
and uipc_syscalls.c:sockargs() when data size is bigger then MLEN,
and such patches were already in KAME environment and have been
confirmed to work well. I just forgot to merge them into 4.0, sorry.
For safety, I'll postpone such patches until after 4.0 release.
The effect of postponement is followings.
-Ping6 source routing hops are limitted to around 6 or so.
-If some apps do setsockopt IPV6_RTHDR and try to receive
incoming IPv6 source routing info, it can't receive more
than 6 hops source routing info.
(But currently, no apps seems to be doing it.)
Approved by: jkh
Make gratuitous style(9) fixes (me, not the submitter) to make the aio
code more readable.
PR: kern/12053
Submitted by: Chris Sedore <cmsedore@maxwell.syr.edu>
an empty mbuf to stay in the queue, then causing a needless panic
because sb_cc == 0 and sb_mbcnt != 0.
But we still need to panic rather than endlessly looping if, for
some reason, sb_cc == 0 and there are non-empty mbufs in the queue.
PR: kern/11988
Reviewed by: fenner
Make a sonewconn3() which takes an extra argument (proc) so new sockets created
with sonewconn() from a user's system call get the correct credentials, not
just the parent's credentials.
into uipc_mbuf.c. This reduces three sets of identical tunable code to
one set, and puts the initialisation with the mbuf code proper.
Make NMBUFs tunable as well.
Move the nmbclusters sysctl here as well.
Move the initialisation of maxsockets from param.c to uipc_socket2.c,
next to its corresponding sysctl.
Use the new tunable macros for the kern.vm.kmem.size tunable (this should have
been in a separate commit, whoops).
This is the change to struct sockets that gets rid of so_uid and replaces
it with a much more useful struct pcred *so_cred. This is here to be able
to do socket-level credential checks (i.e. IPFW uid/gid support, to be added
to HEAD soon). Along with this comes an update to pidentd which greatly
simplifies the code necessary to get a uid from a socket. Soon to come:
a sysctl() interface to finding individual sockets' credentials.
where select(2) can return that a listening socket has a connected socket
queued, the connection is broken, and the user calls accept(2), which then
blocks because there are no connections queued.
Reviewed by: wollman
Obtained from: NetBSD
(ftp://ftp.NetBSD.ORG/pub/NetBSD/misc/security/patches/19990120-accept)
from an interrupt context and fsetown() wants to peek at curproc, call
malloc(..., M_WAITOK), and fiddle with various unprotected data structures.
The fix is to move the code that duplicates the F_SETOWN/FIOSETOWN state
of the original socket to the new socket from sonewconn() to accept1(),
since accept1() runs in the correct context. Deferring this until the
process calls accept() is harmless since the process can't do anything
useful with SIGIO on the new socket until it has the descriptor for that
socket.
One could make the case for not bothering to duplicate the
F_SETOWN/FIOSETOWN state and requiring the process to explicitly make the
fcntl() or ioctl() call on the new socket, but this would be incompatible
with the previous implementation and might break programs which rely on
the old semantics.
This bug was discovered by Andrew Gallatin <gallatin@cs.duke.edu>.
by bde, a few other tweaks to get the patch to apply cleanly again and
some improvements to the comments.
This change closes some fairly minor security holes associated with
F_SETOWN, fixes a few bugs, and removes some limitations that F_SETOWN
had on tty devices. For more details, see the description on the PR.
Because this patch increases the size of the proc and pgrp structures,
it is necessary to re-install the includes and recompile libkvm,
the vinum lkm, fstat, gcore, gdb, ipfilter, ps, top, and w.
PR: kern/7899
Reviewed by: bde, elvind
Also fix data types and printf formats while I'm here.
PR: misc/8494
Panic instead of looping forever in sbflush(). If sb_mbcnt counts
more mbufs than sb_cc counts bytes, the original code can turn into an
infinite loop of removing 0 bytes from the socket buffer until it's empty.
FreeBSD/alpha. The most significant item is to change the command
argument to ioctl functions from int to u_long. This change brings us
inline with various other BSD versions. Driver writers may like to
use (__FreeBSD_version == 300003) to detect this change.
The prototype FreeBSD/alpha machdep will follow in a couple of days
time.
Define a parameter which indicates the maximum number of sockets in a
system, and use this to size the zone allocators used for sockets and
for certain PCBs.
Convert PF_LOCAL PCB structures to be type-stable and add a version number.
Define an external format for infomation about socket structures and use
it in several places.
Define a mechanism to get all PF_LOCAL and PF_INET PCB lists through
sysctl(3) without blocking network interrupts for an unreasonable
length of time. This probably still has some bugs and/or race
conditions, but it seems to work well enough on my machines.
It is now possible for `netstat' to get almost all of its information
via the sysctl(3) interface rather than reading kmem (changes to follow).
* Figure out UTC relative to boottime. Four new functions provide
time relative to boottime.
* move "runtime" into struct proc. This helps fix the calcru()
problem in SMP.
* kill mono_time.
* add timespec{add|sub|cmp} macros to time.h. (XXX: These may change!)
* nanosleep, select & poll takes long sleeps one day at a time
Reviewed by: bde
Tested by: ache and others
declaration macros so that a semicolon can be added when the macros
are invoked without giving a (pedantic) syntax error. Invocations
need to be followed by a semicolon so that programs like indent and
gtags don't get confused.
Fixed the one invocation that wasn't followed by a trailing semicolon.
since that might cause in_pcballoc to call MALLOC with M_WAITOK during
a software interrupt.
Reviewed by: Garrett Wollman <wollman@khavrinen.lcs.mit.edu>
socket addresses in mbufs. (Socket buffers are the one exception.) A number
of kernel APIs needed to get fixed in order to make this happen. Also,
fix three protocol families which kept PCBs in mbufs to not malloc them
instead. Delete some old compatibility cruft while we're at it, and add
some new routines in the in_cksum family.
switch. I needed 'LINT' to compile for other reasons so I kinda got the
blood on my hands. Note: I don't know how to test this, I don't know if
it works correctly.
This commit includes the following changes:
1) Old-style (pr_usrreq()) protocols are no longer supported, the compatibility
glue for them is deleted, and the kernel will panic on boot if any are compiled
in.
2) Certain protocol entry points are modified to take a process structure,
so they they can easily tell whether or not it is possible to sleep, and
also to access credentials.
3) SS_PRIV is no more, and with it goes the SO_PRIVSTATE setsockopt()
call. Protocols should use the process pointer they are now passed.
4) The PF_LOCAL and PF_ROUTE families have been updated to use the new
style, as has the `raw' skeleton family.
5) PF_LOCAL sockets now obey the process's umask when creating a socket
in the filesystem.
As a result, LINT is now broken. I'm hoping that some enterprising hacker
with a bit more time will either make the broken bits work (should be
easy for netipx) or dike them out.
to removing the connection from the queue. The problem here is that
falloc() may block and this would allow another process to accept the
connection instead. If this happens to leave the queue empty, then the
system will panic with an "accept: nothing queued".
Also changed a wakeup() to a wakeup_one() to avoid the "thundering herd"
problem on new connections in Apache (or any other application that has
multiple processes blocked in accept() for the same socket).
all of the configurables and instrumentation related to
inter-process communication mechanisms. Some variables,
like mbuf statistics, are instrumented here for the first
time.
For mbuf statistics: also keep track of m_copym() and
m_pullup() failures, and provide for the user's inspection
the compiled-in values of MSIZE, MHLEN, MCLBYTES, and MINCLSIZE.
sb_max * MCLBYTES / (MSIZE + MCLBYTES)
used in sbreserve() to overflow, causing all socket creation attempts
to fail. Force the calculation to use u_quad_t's, which makes overflow
less likely.
This will make a number of things easier in the future, as well as (finally!)
avoiding the Id-smashing problem which has plagued developers for so long.
Boy, I'm glad we're not using sup anymore. This update would have been
insane otherwise.
using a sockaddr_dl.
Fix the other packet-information socket options (SO_TIMESTAMP, IP_RECVDSTADDR)
to work for multicast UDP and raw sockets as well. (They previously only
worked for unicast UDP).
(yes I had tested the hell out of this).
I've also temporarily disabled the code so that it behaves as it previously
did (tail drop's the syns) pending discussion with fenner about some socket
state flags that I don't fully understand.
Submitted by: fenner
half way through the range rather than possibly colliding with
fixed elements. Increase the size of the arrays to take this into account..
remember that each element in the array is now only 1 ponter so this
isn't that much..
also note a possible bug in debugging code in uipc_socket2.c (add XXX)
pr_usrreq mechanism which was poorly designed and error-prone. This
commit renames pr_usrreq to pr_ousrreq so that old code which depended on it
would break in an obvious manner. This commit also implements the new
interface for TCP, although the old function is left as an example
(#ifdef'ed out). This commit ALSO fixes a longstanding bug in the
TCP timer processing (introduced by davidg on 1995/04/12) which caused
timer processing on a TCB to always stop after a single timer had
expired (because it misinterpreted the return value from tcp_usrreq()
to indicate that the TCB had been deleted). Finally, some code
related to polling has been deleted from if.c because it is not
relevant t -current and doesn't look at all like my current code.
the high kernel calls into a protocol stack to perform requests on the
user's behalf. We replace the pr_usrreq() entry in struct protosw with a
pointer to a structure containing pointers to functions which implement
the various reuqests; each function is declared with the correct type and
number of arguments. (This is unlike the current scheme in which a quarter
of the requests take arguments of type other than (struct mbuf *) and the
difference is papered over with casts.) There are a few benefits to this
new scheme:
1) Arguments are passed with their correct types, and null-pointer dummies
are no longer necessary.
2) There should be slightly better caching effects from eliminating
the prximity to extraneous code and th switch in pr_usrreq().
3) It becomes much easier to change the types of the arguments to something
other than `struct mbuf *' (e.g.,pushing the work of sosend() into
the protocol as advocated by Van Jacobson).
There is one principal drawback: existing protocol stacks need to
be modified. This is alleviated by compatibility code in
uipc_socket2.c and uipc_domain.c which emulates the new interface
in terms of the old and vice versa.
This idea is not original to me. I read about what Jacobson did
in one of his papers and have tried to implement the first steps
towards something like that here. Much work remains to be done.
the obsolete soqinsque and soqremque functions as well as collapsing
so_q0len and so_qlen into a single queue length of unaccepted connections.
Now the queue of unaccepted & complete connections is checked directly
for queued sockets. The new code should be functionally equivilent to
the old while being substantially faster - especially in cases where
large numbers of connections are often queued for accept (e.g. http).
the range [210:260] by sweeping the problem under the rug. This change
has the following effects:
1) A new MIB variable in the kern branch is defined to allow modification
of the socket buffer layer's ``wastage factor'' (which determines how
much unused-but-allocated space in mbufs and mbuf clusters is allowed
in a socket buffer).
2) The default value of the wastage factor is changed from 2 to 8.
The original value was chosen when MINCLSIZE was 7*MLEN (!), and is not
appropriate for an environment where MINCLSIZE is much less.
The real solution to this problem is to scrap both mbufs and sockbufs
and completely redesign the buffering mechanism used at both levels.
(maximum size of a socket buffer) tunable.
Permit callers of listen(2) to specify a negative backlog, which
is translated into somaxconn. Previously, a negative backlog was
silently translated into 0.
