Commit Graph

187 Commits

Author SHA1 Message Date
Robert Watson
b37ffd3189 Move some functions and definitions from uipc_socket2.c to uipc_socket.c:
- Move sonewconn(), which creates new sockets for incoming connections on
  listen sockets, so that all socket allocate code is together in
  uipc_socket.c.

- Move 'maxsockets' and associated sysctls to uipc_socket.c with the
  socket allocation code.

- Move kern.ipc sysctl node to uipc_socket.c, add a SYSCTL_DECL() for it
  to sysctl.h and remove lots of scattered implementations in various
  IPC modules.

- Sort sodealloc() after soalloc() in uipc_socket.c for dependency order
  reasons.  Statisticize soalloc() and sodealloc() as they are now
  required only in uipc_socket.c, and are internal to the socket
  implementation.

After this change, socket allocation and deallocation is entirely
centralized in one file, and uipc_socket2.c consists entirely of socket
buffer manipulation and default protocol switch functions.

MFC after:	1 month
2006-06-10 14:34:07 +00:00
Gleb Smirnoff
61fb9bd80c - In pipe() return the error returned by pipe_create(), rather then
hardcoded ENFILES, which is incorrect. pipe_create() can fail due
  to ENOMEM.
- Update manual page, describing ENOMEM return code.

Reviewed by:	arch
2006-01-30 08:25:04 +00:00
Xin LI
6ba9ec2d09 In pipe_write(): when uiomove() fails, do not spin on it forever.
Submitted by:	Kostik Belousov <kostikbel at gmail.com> on -current@
Message-ID:	<20051216151016.GE84442@deviant.zoral.local>
MFC After:	3 weeks
2005-12-16 18:32:39 +00:00
Suleiman Souhlal
571dcd15e2 Fix the recent panics/LORs/hangs created by my kqueue commit by:
- Introducing the possibility of using locks different than mutexes
for the knlist locking. In order to do this, we add three arguments to
knlist_init() to specify the functions to use to lock, unlock and
check if the lock is owned. If these arguments are NULL, we assume
mtx_lock, mtx_unlock and mtx_owned, respectively.

- Using the vnode lock for the knlist locking, when doing kqueue operations
on a vnode. This way, we don't have to lock the vnode while holding a
mutex, in filt_vfsread.

Reviewed by:	jmg
Approved by:	re (scottl), scottl (mentor override)
Pointyhat to:	ssouhlal
Will be happy:	everyone
2005-07-01 16:28:32 +00:00
Mike Silbersack
6792415119 Rearrange the kninit calls for both directions of a pipe so that
they both happen before pipe backing allocation occurs.  Previously,
a pipe memory shortage would cause a panic due to a KNOTE call
on an uninitialized si_note.

Reported by:	Peter Holm
MFC after:	1 week
2005-01-17 07:56:28 +00:00
Warner Losh
9454b2d864 /* -> /*- for copyright notices, minor format tweaks as necessary 2005-01-06 23:35:40 +00:00
Robert Watson
436cac68e6 Correct a bug introduced in sys_pipe.c:1.179: in pipe_ioctl(),
release the pipe mutex before calling fsetown(), as fsetown()
may block.  The sigio code protects the pipe sigio data using
its own mutex, and the pipe reference count held by the caller
will prevent the pipe from being prematurely garbage-collected.

Discovered by:	imp
2004-11-23 22:15:08 +00:00
Poul-Henning Kamp
7f21497282 Add missing break. 2004-11-16 06:57:52 +00:00
Poul-Henning Kamp
f661e9a0bc Straighten the ioctl function out to have only one exit point. 2004-11-15 21:51:28 +00:00
Poul-Henning Kamp
ef11fbd7c4 Introduce fdclose() which will clean an entry in a filedesc.
Replace homerolled versions with call to fdclose().

Make fdunused() static to kern_descrip.c
2004-11-07 22:16:07 +00:00
Mike Silbersack
5173e8f567 Major enhancements to pipe memory usage:
- pipespace is now able to resize non-empty pipes; this allows
  for many more resizing opportunities

- Backing is no longer pre-allocated for the reverse direction
  of pipes.  This direction is rarely (if ever) used, so this cuts the
  amount of map space allocated to a pipe in half.

- Pipe growth is now much more dynamic; a pipe will now grow when
  the total amount of data it contains and the size of the write are
  larger than the size of pipe.  Previously, only individual writes greater
  than the size of the pipe would cause growth.

- In low memory situations, pipes will now shrink during both read
  and write operations, where possible.  Once the memory shortage
  ends, the growth code will cause these pipes to grow back to an appropriate
  size.

- If the full PIPE_SIZE allocation fails when a new pipe is created, the
  allocation will be retried with SMALL_PIPE_SIZE.  This helps to deal
  with the situation of a fragmented map after a low memory period has
  ended.

- Minor documentation + code changes to support the above.

In total, these changes increase the total number of pipes that
can be allocated simultaneously, drastically reducing the chances that
pipe allocation will fail.

Performance appears unchanged due to dynamic resizing.
2004-08-16 01:27:24 +00:00
John-Mark Gurney
ad3b9257c2 Add locking to the kqueue subsystem. This also makes the kqueue subsystem
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)
2004-08-15 06:24:42 +00:00
Mike Silbersack
e10ecdea88 Standardize pipe locking, ensuring that everything is locked via
pipelock(), not via a mixture of mutexes and pipelock().  Additionally,
add a few KASSERTS, and change some statements that should have been
KASSERTS into KASSERTS.

As a result of these cleanups, some segments of code have become
significantly shorter and/or easier to read.
2004-08-03 02:59:15 +00:00
Brian Feldman
b23f72e98a * Add a "how" argument to uma_zone constructors and initialization functions
so that they know whether the allocation is supposed to be able to sleep
  or not.
* Allow uma_zone constructors and initialation functions to return either
  success or error.  Almost all of the ones in the tree currently return
  success unconditionally, but mbuf is a notable exception: the packet
  zone constructor wants to be able to fail if it cannot suballocate an
  mbuf cluster, and the mbuf allocators want to be able to fail in general
  in a MAC kernel if the MAC mbuf initializer fails.  This fixes the
  panics people are seeing when they run out of memory for mbuf clusters.
* Allow debug.nosleepwithlocks on WITNESS to be disabled, without changing
  the default.

Both bmilekic and jeff have reviewed the changes made to make failable
zone allocations work.
2004-08-02 00:18:36 +00:00
Robert Watson
46b25cb5f6 Don't perform pipe endpoint locking during pipe_create(), as the pipe
can't yet be referenced by other threads.

In microbenchmarks, this appears to reduce the cost of
pipe();close();close() on UP by 10%, and SMP by 7%.  The vast majority
of the cost of allocating a pipe remains VM magic.

Suggested by:	silby
2004-07-23 14:11:04 +00:00
Mike Silbersack
eb3d2c61b4 Fix a minor error in pipe_stat - st_size was always reported as 0
when direct writes kicked in.  Whether this affected any applications
is unknown.
2004-07-20 07:06:43 +00:00
Alan Cox
e3b19536fb Revise the direct or optimized case to use uiomove_fromphys() by the reader
instead of ephemeral mappings using pmap_qenter() by the writer.  The
writer is still, however, responsible for wiring the pages, just not
mapping them.  Consequently, the allocation of KVA for the direct case is
unnecessary.  Remove it and the sysctls limiting it, i.e.,
kern.ipc.maxpipekvawired and kern.ipc.amountpipekvawired.  The number
of temporarily wired pages is still, however, limited by
kern.ipc.maxpipekva.

Note: On platforms lacking a direct virtual-to-physical mapping,
uiomove_fromphys() uses sf_bufs to cache ephemeral mappings.  Thus,
the number of available sf_bufs can influence the performance of pipes
on platforms such i386.  Surprisingly, I saw the greatest gain from this
change on such a machine: lmbench's pipe bandwidth result increased from
~1050MB/s to ~1850MB/s on my 2.4GHz, 400MHz FSB P4 Xeon.
2004-03-27 19:50:23 +00:00
Robert Watson
049ffe98a8 Assert pipe mutex in pipeselwakeup(), as we manipulate pipe_state
in a non-atomic manner.  It appears to always be called with the
mutex (good).
2004-02-26 00:18:22 +00:00
Robert Watson
094bdd260c Update comment regarding MAC labels: we no longer pass endpoints
into the MAC Framework, just the pipe pair.

GC 'hadpeer' used in pipedestroy(), which is no longer needed as
we check pipe_present flags on the pair.
2004-02-25 23:30:56 +00:00
Brian Feldman
240160d48b Correct some major SMP-harmful problems in the pipe implementation. First
of all, PIPE_EOF is not checked pervasively after everything that can drop
the pipe mutex and msleep(), so fix.  Additionally, though it might not
harm anything, pipelock() and pipeunlock() are not used consistently.
Third, the kqueue support functions do not use the pipe mutex correctly.
Last, but absolutely not least, is a race: if pipe_busy is not set on
the closing side of the pipe, the other side that is trying to write to
that will crash BECAUSE PIPE_EOF IS NOT SET!  Unconditionally set
PIPE_EOF, and get rid of all the lockups/crashes I have seen trying
to build ports.
2004-02-22 23:00:14 +00:00
Robert Watson
4f638130c3 Don't dec/inc the amountpipes counter every time we resize a pipe --
instead, just dec/inc in the ctor/dtor.  For now, increment/decrement
in two's, since we're now performing the operation once per pair,
not once per pipe.  Not really any measurable performance change
in my micro-benchmarks, but doing less work is good, especially when
it comes to atomic operations.

Suggested by:	alc
2004-02-03 04:55:24 +00:00
Robert Watson
9a830ddc54 Catch instances of (pipe == NULL) that were obsoleted with recent
changes to jointly allocated pipe pairs.  Replace these checks
with pipe_present checks.  This avoids a NULL pointer dereference
when a pipe is half-closed.

Submitted by:	Peter Edwards <peter.edwards@openet-telecom.com>
2004-02-03 02:50:51 +00:00
Robert Watson
4795b82c13 Coalesce pipe allocations and frees. Previously, the pipe code
would allocate two 'struct pipe's from the pipe zone, and malloc a
mutex.

- Create a new "struct pipepair" object holding the two 'struct
  pipe' instances, struct mutex, and struct label reference.  Pipe
  structures now have a back-pointer to the pipe pair, and a
  'pipe_present' flag to indicate whether the half has been
  closed.

- Perform mutex init/destroy in zone init/destroy, avoiding
  reallocating the mutex for each pipe.  Perform most pipe structure
  setup in zone constructor.

- VM memory mappings for pageable buffers are still done outside of
  the UMA zone.

- Change MAC API to speak 'struct pipepair' instead of 'struct pipe',
  update many policies.  MAC labels are also handled outside of the
  UMA zone for now.  Label-only policy modules don't have to be
  recompiled, but if a module is recompiled, its pipe entry points
  will need to be updated.  If a module actually reached into the
  pipe structures (unlikely), that would also need to be modified.

These changes substantially simplify failure handling in the pipe
code as there are many fewer possible failure modes.

On half-close, pipes no longer free the 'struct pipe' for the closed
half until a full-close takes place.  However, VM mapped buffers
are still released on half-close.

Some code refactoring is now possible to clean up some of the back
references, etc; this patch attempts not to change the structure
of most of the pipe implementation, only allocation/free code
paths, so as to avoid introducing bugs (hopefully).

This cuts about 8%-9% off the cost of sequential pipe allocation
and free in system call tests on UP and SMP in my micro-benchmarks.
May or may not make a difference in macro-benchmarks, but doing
less work is good.

Reviewed by:	juli, tjr
Testing help:	dwhite, fenestro, scottl, et al
2004-02-01 05:56:51 +00:00
Robert Watson
26518e8d8c Fix an error in a KASSERT string: it's pipe_free_kmem(), not
pipespace(), that contains this KASSERT.
2004-01-31 23:03:22 +00:00
Dag-Erling Smørgrav
a2fe44e8cf New file descriptor allocation code, derived from similar code introduced
in OpenBSD by Niels Provos.  The patch introduces a bitmap of allocated
file descriptors which is used to locate available descriptors when a new
one is needed.  It also moves the task of growing the file descriptor table
out of fdalloc(), reducing complexity in both fdalloc() and do_dup().

Debts of gratitude are owed to tjr@ (who provided the original patch on
which this work is based), grog@ (for the gdb(4) man page) and rwatson@
(for assistance with pxeboot(8)).
2004-01-15 10:15:04 +00:00
Dag-Erling Smørgrav
ac34dc4e79 Back out 1.160, which was committed by mistake. 2004-01-11 20:08:57 +00:00
Dag-Erling Smørgrav
0e5dfade00 Mechanical whitespace cleanup. 2004-01-11 19:54:45 +00:00
Dag-Erling Smørgrav
012b5531f4 Mechanical whitespace cleanup + minor style nits. 2004-01-11 19:43:14 +00:00
Mike Silbersack
69fba1650a Fix the maxpipekva warning message so that it points to the correct
sysctl, and shorten the message.

Noticed by:	bde
2003-12-28 01:19:58 +00:00
Seigo Tanimura
512824f8f7 - Implement selwakeuppri() which allows raising the priority of a
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
2003-11-09 09:17:26 +00:00
Alan Cox
3b2c54e7bc - Delay the allocation of memory for the pipe mutex until we need it.
This avoids the need to free said memory in various error cases along
   the way.
2003-11-06 05:58:26 +00:00
Alan Cox
fc17df5264 - Simplify pipespace() by eliminating the explicit creation of vm objects.
Instead, let the vm objects be lazily instantiated at fault time.  This
   results in the allocation of fewer vm objects and vm map entries due to
   aggregation in the vm system.
2003-11-06 05:08:12 +00:00
Robert Watson
730ecf8254 Unlock pipe mutex when failing MAC pipe ioctl access control check.
Obtained from:	TrustedBSD Project
Sponsored by:	DARPA, Network Associates Laboratories
2003-11-03 17:58:23 +00:00
Mike Silbersack
184dcdc7c8 Change all SYSCTLS which are readonly and have a related TUNABLE
from CTLFLAG_RD to CTLFLAG_RDTUN so that sysctl(8) can provide
more useful error messages.
2003-10-21 18:28:36 +00:00
David Malone
e1419c08e2 falloc allocates a file structure and adds it to the file descriptor
table, acquiring the necessary locks as it works. It usually returns
two references to the new descriptor: one in the descriptor table
and one via a pointer argument.

As falloc releases the FILEDESC lock before returning, there is a
potential for a process to close the reference in the file descriptor
table before falloc's caller gets to use the file. I don't think this
can happen in practice at the moment, because Giant indirectly protects
closes.

To stop the file being completly closed in this situation, this change
makes falloc set the refcount to two when both references are returned.
This makes life easier for several of falloc's callers, because the
first thing they previously did was grab an extra reference on the
file.

Reviewed by:	iedowse
Idea run past:	jhb
2003-10-19 20:41:07 +00:00
John-Mark Gurney
9e5de980c6 fix a problem referencing free'd memory. This is only a problem for
kqueue write events on a socket and you regularly create tons of pipes
which overwrites the structure causing a panic when removing the knote
from the list.  If the peer has gone away (and it's a write knote), then
don't bother trying to remove the knote from the list.

Submitted by:	Brian Buchanan and myself
Obtained from:	nCircle
2003-10-12 07:06:02 +00:00
Alan Cox
27d203eab3 pipe_build_write_buffer() only requires read access of the page that it
obtains from pmap_extract_and_hold().
2003-09-12 07:13:15 +00:00
Alan Cox
03be99d20c Use pmap_extract_and_hold() in pipe_build_write_buffer(). Consequently,
pipe_build_write_buffer() no longer requires Giant on entry.

Reviewed by:	tegge
2003-09-08 04:58:32 +00:00
Alan Cox
603d3d4a44 Giant is no longer required by pipe_destroy_write_buffer(). Reduce
unnecessary white space from pipe_destroy_write_buffer().
2003-09-06 21:02:10 +00:00
John-Mark Gurney
fc8684cd46 if we got this far, we definately don't have an EBADF. Return a more
sane result of EPIPE.

Reported by:	nCircle dev team
MFC after:	3 day
2003-08-15 04:31:01 +00:00
Alan Cox
77685ea594 - The vm_object pointer in pipe_buffer is unused. Remove it.
- Check for successful initialization of pipe_zone in pipeinit()
   rather than every call to pipe(2).
2003-08-13 20:01:38 +00:00
Alan Cox
ad8204e3f5 Pipespace() no longer requires Giant. 2003-08-11 22:23:25 +00:00
Mike Silbersack
cebde06978 More pipe changes:
From alc:
Move pageable pipe memory to a seperate kernel submap to avoid awkward
vm map interlocking issues.  (Bad explanation provided by me.)

From me:
Rework pipespace accounting code to handle this new layout, and adjust
our default values to account for the fact that we now have a solid
limit on allocations.

Also, remove the "maxpipes" limit, as it no longer has a purpose.
(The limit on kva usage solves the problem of having two many pipes.)
2003-08-11 05:51:51 +00:00
Alan Cox
f9999c67be Use vm_page_hold() instead of vm_page_wire(). Otherwise, a multithreaded
application could cause a wired page to be freed.  In general,
vm_page_hold() should be preferred for ephemeral kernel mappings of pages
borrowed from a user-level address space.  (vm_page_wire() should really be
reserved for indefinite duration pinning by the "owner" of the page.)

Discussed with:	silby
Submitted by:	tegge
2003-08-11 00:17:44 +00:00
Alan Cox
9c62fce085 - Remove GIANT_REQUIRED from pipespace().
- Remove a duplicate initialization from pipe_create().
2003-08-08 22:38:15 +00:00
Alan Cox
f9b1de367e - Remove GIANT_REQUIRED from pipe_free_kmem().
- Remove the acquisition and release of Giant around pipe_kmem_free() and
   uma_zfree() in pipeclose().
2003-08-07 04:32:40 +00:00
Pierre Beyssac
ae9fcf4c66 Remove test in pipe_write() which causes write(2) to return EAGAIN
on a non-blocking pipe in cases where select(2) returns the file
descriptor as ready for write. This in turns causes libc_r, for
one, to busy wait in such cases.

Note: it is a quick performance fix, a more complex fix might be
required in case this turns out to have unexpected side effects.

Reviewed by:	silby
MFC after:	3 days
2003-07-30 22:50:37 +00:00
Alan Cox
93b4c5b707 The introduction of vm object locking has caused witness to reveal
a long-standing mistake in the way a portion of a pipe's KVA is
allocated.  Specifically, kmem_alloc_pageable() is inappropriate
for use in the "direct" case because it allows a preceding vm map entry
and vm object to be extended to support the new KVA allocation.
However, the direct case KVA allocation should not have a backing
vm object.  This is corrected by using kmem_alloc_nofault().

Submitted by:	tegge (with the above explanation by me)
2003-07-30 18:55:04 +00:00
Mike Silbersack
ff56f15e26 A few minor changes:
- Use atomic ops to update the bigpipe count
- Make the bigpipe count sysctl readable
- Remove a duplicate comparison in an if statement
- Comment two SYSCTLs.
2003-07-09 21:59:48 +00:00
Mike Silbersack
289016f2d1 Put some concrete limits on pipe memory consumption:
- Limit the total number of pipes so that we do not
  exhaust all vm objects in the kernel map.  When
  this limit is reached, a ratelimited message will
  be printed to the console.

- Put a soft limit on the amount of memory consumable
  by pipes.  Once the limit has been reached, all new
  pipes will be limited to 4K in size, rather than the
  default of 16K.

- Put a limit on the number of pages that may be used
  for high speed page flipping in order to reduce the
  amount of wired memory.  Pipe writes that occur
  while this limit is exceeded will fall back to
  non-page flipping mode.

The above values are auto-tuned in subr_param.c and
are scaled to take into account both the size of
physical memory and the size of the kernel map.

These limits help to reduce the "kernel resources exhausted"
panics that could be caused by opening a large
number of pipes.  (Pipes alone are no longer able
to exhaust all resources, but other kernel memory hogs
in league with pipes may still be able to do so.)

PR:			53627
Ideas / comments from:	hsu, tjr, dillon@apollo.backplane.com
MFC after:		1 week
2003-07-08 04:02:31 +00:00