Commit Graph

242 Commits

Author SHA1 Message Date
Konstantin Belousov
d8b0556c6d Adapt vfs kqfilter to the shared vnode lock used by zfs write vop. Use
vnode interlock to protect the knote fields [1]. The locking assumes
that shared vnode lock is held, thus we get exclusive access to knote
either by exclusive vnode lock protection, or by shared vnode lock +
vnode interlock.

Do not use kl_locked() method to assert either lock ownership or the
fact that curthread does not own the lock. For shared locks, ownership
is not recorded, e.g. VOP_ISLOCKED can return LK_SHARED for the shared
lock not owned by curthread, causing false positives in kqueue subsystem
assertions about knlist lock.

Remove kl_locked method from knlist lock vector, and add two separate
assertion methods kl_assert_locked and kl_assert_unlocked, that are
supposed to use proper asserts. Change knlist_init accordingly.

Add convenience function knlist_init_mtx to reduce number of arguments
for typical knlist initialization.

Submitted by:	jhb [1]
Noted by:	jhb [2]
Reviewed by:	jhb
Tested by:	rnoland
2009-06-10 20:59:32 +00:00
John Baldwin
74fb0ba732 Rework socket upcalls to close some races with setup/teardown of upcalls.
- Each socket upcall is now invoked with the appropriate socket buffer
  locked.  It is not permissible to call soisconnected() with this lock
  held; however, so socket upcalls now return an integer value.  The two
  possible values are SU_OK and SU_ISCONNECTED.  If an upcall returns
  SU_ISCONNECTED, then the soisconnected() will be invoked on the
  socket after the socket buffer lock is dropped.
- A new API is provided for setting and clearing socket upcalls.  The
  API consists of soupcall_set() and soupcall_clear().
- To simplify locking, each socket buffer now has a separate upcall.
- When a socket upcall returns SU_ISCONNECTED, the upcall is cleared from
  the receive socket buffer automatically.  Note that a SO_SND upcall
  should never return SU_ISCONNECTED.
- All this means that accept filters should now return SU_ISCONNECTED
  instead of calling soisconnected() directly.  They also no longer need
  to explicitly clear the upcall on the new socket.
- The HTTP accept filter still uses soupcall_set() to manage its internal
  state machine, but other accept filters no longer have any explicit
  knowlege of socket upcall internals aside from their return value.
- The various RPC client upcalls currently drop the socket buffer lock
  while invoking soreceive() as a temporary band-aid.  The plan for
  the future is to add a new flag to allow soreceive() to be called with
  the socket buffer locked.
- The AIO callback for socket I/O is now also invoked with the socket
  buffer locked.  Previously sowakeup() would drop the socket buffer
  lock only to call aio_swake() which immediately re-acquired the socket
  buffer lock for the duration of the function call.

Discussed with:	rwatson, rmacklem
2009-06-01 21:17:03 +00:00
John Baldwin
e588eeb1dc Use the correct type for the timeout parameter to the 32-bit
compat version aio_waitcomplete().

Reminded by:	bz
Submitted by:	jamie
MFC after:	3 days
2009-01-23 13:23:17 +00:00
John Baldwin
3858a1f4f5 - Add 32-bit compat system calls for VFS_AIO. The system calls live in the
aio code and are registered via the recently added SYSCALL32_*() helpers.
- Since the aio code likes to invoke fuword and suword a lot down in the
  "bowels" of system calls, add a structure holding a set of operations for
  things like storing errors, copying in the aiocb structure, storing
  status, etc.  The 32-bit system calls use a separate operations vector to
  handle fuword32 vs fuword, etc.  Also, the oldsigevent handling is now
  done by having seperate operation vectors with different aiocb copyin
  routines.
- Split out kern_foo() functions for the various AIO system calls so the
  32-bit front ends can manage things like copying in and converting
  timespec structures, etc.
- For both the native and 32-bit aio_suspend() and lio_listio() calls,
  just use copyin() to read the array of aiocb pointers instead of using
  a for loop that iterated over fuword/fuword32.  The error handling in
  the old case was incomplete (lio_listio() just ignored any aiocb's that
  it got an EFAULT trying to read rather than reporting an error), and
  possibly slower.

MFC after:	1 month
2008-12-10 20:56:19 +00:00
Oleksandr Tymoshenko
22035f4727 Use minimum of max_aio_procs and target_aio_procs when spawning new
aiod since there should be no more then max_aio_procs processes.
2008-06-21 11:34:34 +00:00
Robert Watson
e603be7ada Use FEATURE() macro to advertise aio availability. 2008-02-01 11:59:14 +00:00
Jean-Sébastien Pédron
a8afa221cc When asked to use kqueue, AIO stores its internal state in the
`kn_sdata' member of the newly registered knote. The problem is that
this member is overwritten by a call to kevent(2) with the EV_ADD flag,
targetted at the same kevent/knote. For instance, a userland application
may set the pointer to NULL, leading to a panic.

A testcase was provided by the submitter.

PR:	kern/118911
Submitted by:	MOROHOSHI Akihiko <moro@remus.dti.ne.jp>
MFC after:	1 day
2008-01-24 17:10:19 +00:00
Attilio Rao
22db15c06f VOP_LOCK1() (and so VOP_LOCK()) and VOP_UNLOCK() are only used in
conjuction with 'thread' argument passing which is always curthread.
Remove the unuseful extra-argument and pass explicitly curthread to lower
layer functions, when necessary.

KPI results broken by this change, which should affect several ports, so
version bumping and manpage update will be further committed.

Tested by: kris, pho, Diego Sardina <siarodx at gmail dot com>
2008-01-13 14:44:15 +00:00
Attilio Rao
cb05b60a89 vn_lock() is currently only used with the 'curthread' passed as argument.
Remove this argument and pass curthread directly to underlying
VOP_LOCK1() VFS method. This modify makes the code cleaner and in
particular remove an annoying dependence helping next lockmgr() cleanup.
KPI results, obviously, changed.

Manpage and FreeBSD_version will be updated through further commits.

As a side note, would be valuable to say that next commits will address
a similar cleanup about VFS methods, in particular vop_lock1 and
vop_unlock.

Tested by:	Diego Sardina <siarodx at gmail dot com>,
		Andrea Di Pasquale <whyx dot it at gmail dot com>
2008-01-10 01:10:58 +00:00
Julian Elischer
3745c395ec Rename the kthread_xxx (e.g. kthread_create()) calls
to kproc_xxx as they actually make whole processes.
Thos makes way for us to add REAL kthread_create() and friends
that actually make theads. it turns out that most of these
calls actually end up being moved back to the thread version
when it's added. but we need to make this cosmetic change first.

I'd LOVE to do this rename in 7.0  so that we can eventually MFC the
new kthread_xxx() calls.
2007-10-20 23:23:23 +00:00
Konstantin Belousov
5114048b63 Destroy the kaio_mtx on the freeing the struct kaioinfo in the
aio_proc_rundown.

Do not allow for zero-length read to be passed to the fo_read file method
by aio.

Reported and tested by:	Peter Holm
Approved by:	re (kensmith)
2007-08-20 11:53:26 +00:00
Matt Jacob
a659386c7e Remove unused variable. 2007-06-10 01:50:05 +00:00
Jeff Roberson
1c4bcd050a - Move rusage from being per-process in struct pstats to per-thread in
td_ru.  This removes the requirement for per-process synchronization in
   statclock() and mi_switch().  This was previously supported by
   sched_lock which is going away.  All modifications to rusage are now
   done in the context of the owning thread.  reads proceed without locks.
 - Aggregate exiting threads rusage in thread_exit() such that the exiting
   thread's rusage is not lost.
 - Provide a new routine, rufetch() to fetch an aggregate of all rusage
   structures from all threads in a process.  This routine must be used
   in any place requiring a rusage from a process prior to it's exit.  The
   exited process's rusage is still available via p_ru.
 - Aggregate tick statistics only on demand via rufetch() or when a thread
   exits.  Tick statistics are kept in the thread and protected by sched_lock
   until it exits.

Initial patch by:	attilio
Reviewed by:		attilio, bde (some objections), arch (mostly silent)
2007-06-01 01:12:45 +00:00
Robert Watson
873fbcd776 Further system call comment cleanup:
- Remove also "MP SAFE" after prior "MPSAFE" pass. (suggested by bde)
- Remove extra blank lines in some cases.
- Add extra blank lines in some cases.
- Remove no-op comments consisting solely of the function name, the word
  "syscall", or the system call name.
- Add punctuation.
- Re-wrap some comments.
2007-03-05 13:10:58 +00:00
Tom Rhodes
6aeb05d7be Merge posix4/* into normal kernel hierarchy.
Reviewed by:	glanced at by jhb
Approved by:	silence on -arch@ and -standards@
2006-11-11 16:26:58 +00:00
Alexander Leidinger
6a1162d4cd MFP4 (with some minor changes):
Implement the linux_io_* syscalls (AIO). They are only enabled if the native
AIO code is available (either compiled in to the kernel or as a module) at
the time the functions are used. If the AIO stuff is not available there
will be a ENOSYS.

From the submitter:
---snip---
DESIGN NOTES:

1. Linux permits a process to own multiple AIO queues (distinguished by
   "context"), but FreeBSD creates only one single AIO queue per process.
   My code maintains a request queue (STAILQ of queue(3)) per "context",
   and throws all AIO requests of all contexts owned by a process into
   the single FreeBSD per-process AIO queue.

   When the process calls io_destroy(2), io_getevents(2), io_submit(2) and
   io_cancel(2), my code can pick out requests owned by the specified context
   from the single FreeBSD per-process AIO queue according to the per-context
   request queues maintained by my code.

2. The request queue maintained by my code stores contrast information between
   Linux IO control blocks (struct linux_iocb) and FreeBSD IO control blocks
   (struct aiocb). FreeBSD IO control block actually exists in userland memory
   space, required by FreeBSD native aio_XXXXXX(2).

3. It is quite troubling that the function io_getevents() of libaio-0.3.105
   needs to use Linux-specific "struct aio_ring", which is a partial mirror
   of context in user space. I would rather take the address of context in
   kernel as the context ID, but the io_getevents() of libaio forces me to
   take the address of the "ring" in user space as the context ID.

   To my surprise, one comment line in the file "io_getevents.c" of
   libaio-0.3.105 reads:

             Ben will hate me for this

REFERENCE:

1. Linux kernel source code:   http://www.kernel.org/pub/linux/kernel/v2.6/
   (include/linux/aio_abi.h, fs/aio.c)

2. Linux manual pages:         http://www.kernel.org/pub/linux/docs/manpages/
   (io_setup(2), io_destroy(2), io_getevents(2), io_submit(2), io_cancel(2))

3. Linux Scalability Effort:   http://lse.sourceforge.net/io/aio.html
   The design notes:           http://lse.sourceforge.net/io/aionotes.txt

4. The package libaio, both source and binary:
       http://rpmfind.net/linux/rpm2html/search.php?query=libaio
   Simple transparent interface to Linux AIO system calls.

5. Libaio-oracle:              http://oss.oracle.com/projects/libaio-oracle/
   POSIX AIO implementation based on Linux AIO system calls (depending on
   libaio).
---snip---

Submitted by:	Li, Xiao <intron@intron.ac>
2006-10-15 14:22:14 +00:00
John-Mark Gurney
4db71d27a1 hide kqueue_register from public view, and replace it w/ kqfd_register...
this eliminates a possible race in aio registering a kevent..
2006-09-24 04:47:47 +00:00
Mark Peek
f6d004d510 Remove call to fdfree() for the AIO daemons to prevent kernel panics
with linprocfs. This call is not needed since file descriptor sharing
was removed in v1.125.

Reviewed by:	alc, davidxu, ambrisko
MFC after:	3 days
2006-09-06 15:11:20 +00:00
Alexander Leidinger
993182e57c - Change process_exec function handlers prototype to include struct
image_params arg.
- Change struct image_params to include struct sysentvec pointer and
  initialize it.
- Change all consumers of process_exit/process_exec eventhandlers to
  new prototypes (includes splitting up into distinct exec/exit functions).
- Add eventhandler to userret.

Sponsored by:		Google SoC 2006
Submitted by:		rdivacky
Parts suggested by:	jhb (on hackers@)
2006-08-15 12:10:57 +00:00
Doug Ambrisko
51e37c7f37 Make lio ident more consistant with aio ident. 2006-06-02 17:45:48 +00:00
David Xu
759ccccadb Use a dedicated mutex to protect aio queues, the movation is to reduce
lock contention with other parts.
2006-05-09 00:10:11 +00:00
David Xu
dbbccfe923 1. Move code for scanning pending I/O from aio_fsync to aio_aqueue,
it has less overhead.
2. Avoid scheduling task if maximum number of I/O threads is reached.
2006-03-24 00:50:06 +00:00
David Xu
99eee864ad Implement aio_fsync() syscall. 2006-03-23 08:46:42 +00:00
David Xu
27b8220d12 1. Remove aio entry from lists earlier in aio_free_entry,
so other threads can not see it if we unlock the proc
   lock (this can happen in knlist_delete).  Don't do wakeup,
   it is not necessary.

2. Decrease kaio_buffer_count in biohelper rather than
   doing it in aio_bio_done_notify.

3. In aio_bio_done_notify, don't send notification if KAIO_RUNDOWN
   was set, because the process is already in single thread mode.

4. Use assignment to initialize aiothreadflags.

5. AIOCBLIST_RUNDOWN is not useful, axe the code using it.

6. use LIO_NOP instead of zero.
2006-02-26 12:56:23 +00:00
David Xu
ad8de0f243 If block size is zero, use normal file operations to do I/O,
this eliminates a divided-by-zero fault.

Recommended by: phk
2006-02-22 00:05:12 +00:00
David Xu
6d53aa6297 Just like dofilewrite(), call bwillwrite before fo_write. 2006-01-27 08:02:25 +00:00
David Xu
03d66b36c7 return final error code in aio_return rather than a hardcoded 0. 2006-01-27 04:14:16 +00:00
David Xu
55a122bf28 in aio_aqueue, store same return code into job->_aiocb_private.error.
in aio_return, unlock proc lock before suword.
2006-01-26 08:37:02 +00:00
David Xu
1aa4c324ee Add locking annotation and comments about socket, pipe, fifo problem.
Temporarily fix a locking problem for socket I/O.
2006-01-24 07:24:24 +00:00
David Xu
e6bdc05ff7 Er, rescure a deleted comment line. 2006-01-24 02:50:42 +00:00
David Xu
bd793be3c6 More cleanup for aio code:
1) unregsiter kqueue filter for EVFILT_LIO.
2) free uma_zones.
3) call setsid directly to enter another session rather than
   implementing by itself.

Submitted by: jhb
2006-01-24 02:46:15 +00:00
David Xu
7f34b521c7 Add bracket. 2006-01-23 23:46:30 +00:00
David Xu
68d7111884 Verify all supported notification types. 2006-01-23 10:27:15 +00:00
David Xu
a9bf5e37ae 1) Merge _aio_aqueue and aio_aqueue, check quota in aio_aqueue,
so that lio_listio won't exceed the quota.
2) Remove lio_ref_count, it is no longer used.
2006-01-23 02:49:34 +00:00
David Xu
8c0d9af5bf Fix a bogus panic. 2006-01-22 09:39:59 +00:00
David Xu
9b84335c84 Decrease kaio_active_count first, because user process may go away
after we notified it.
2006-01-22 09:25:52 +00:00
David Xu
1ce9182407 Make aio code MP safe. 2006-01-22 05:59:27 +00:00
Christian S.J. Peron
8213baf002 Initialize ki to p->p_aioinfo after we know it's going to be referencing
a valid kaioinfo structure. This avoids a potential NULL pointer dereference.

Found with:	Coverity Prevent(tm)
MFC after:	2 weeks
2006-01-15 01:55:45 +00:00
John Baldwin
af56abaab5 Return error from fget_write() rather than hardcoding EBADF now that
fget_write() DTRT.

Requested by:	bde
2006-01-06 16:34:22 +00:00
David Xu
323fe56580 In aio_waitcomplete, do not return EAGAIN if no other threads
have started aio, instead, initialize aio management structure
if it hasn't been done, the reason to adjust this behavior is
to make it a bit friendly for threaded program, consider two
threads, one submits aio_write, and another just calls
aio_waitcomplete to wait any I/O to be completed and recycle the
aio requests, before submitter doing any I/O, the recycler wants
to wait in kernel. This also fixes inconsistency with other aio
syscalls.
2005-11-08 23:48:32 +00:00
John Baldwin
2a522eb9d3 Various and sundry cleanups:
- Use curthread for calls to knlist_delete() and add a big comment
  explaining why as well as appropriate assertions.
- Use TAILQ_FOREACH and TAILQ_FOREACH_SAFE instead of handrolling them.
- Use fget() family of functions to lookup file objects instead of
  grovelling around in file descriptor tables.
- Destroy the aio_freeproc mutex if we are unloaded.

Tested on:	i386
2005-11-08 17:43:05 +00:00
David Xu
8f0371f19d Fix name compatible problem with POSIX standard. the sigval_ptr and
sigval_int really should be sival_ptr and sival_int.
Also sigev_notify_function accepts a union sigval value but not a
pointer.
2005-11-04 09:41:00 +00:00
David Xu
4c0fb2cfff Support sending realtime signal information via signal queue, realtime
signal memory is pre-allocated, so kernel can always notify user code.
2005-11-03 05:25:26 +00:00
John Baldwin
68a17869c1 Push down Giant into fdfree() and remove it from two of the callers.
Other callers such as some rfork() cases weren't locking Giant anyway.

Reviewed by:	csjp
MFC after:	1 week
2005-11-01 17:13:05 +00:00
David Xu
0972628aff Fix sigevent's POSIX incompatible problem by adding member fields
sigev_notify_function and sigev_notify_attributes. AIO syscalls
use sigevent, so they have to be adjusted.

Reviewed by:	alc
2005-10-30 02:12:49 +00:00
Doug Ambrisko
db43cd0417 Fix tinderbox box by removing incomplete/bad spl usage. Proper giant free
locking is required in for aio.

Pointed out by:	imp
2005-10-12 22:33:22 +00:00
Doug Ambrisko
69cd28dacb Add in kqueue support to LIO event notification and fix how it handled
notifications when LIO operations completed.  These were the problems
with LIO event complete notification:
      -	Move all LIO/AIO event notification into one general function
	so we don't have bugs in different data paths.  This unification
	got rid of several notification bugs one of which if kqueue was
	used a SIGILL could get sent to the process.
      -	Change the LIO event accounting to count all AIO request that
	could have been split across the fast path and daemon mode.
	The prior accounting only kept track of AIO op's in that
	mode and not the entire list of operations.  This could cause
	a bogus LIO event complete notification to occur when all of
	the fast path AIO op's completed and not the AIO op's that
	ended up queued for the daemon.

Suggestions from:	alc
2005-10-12 17:51:31 +00:00
Alan Cox
ec9c9e7363 Eliminate inconsistency in the setting of the B_DONE flag. Specifically,
make the b_iodone callback responsible for setting it if it is needed.
Previously, it was set unconditionally by bufdone() without holding
whichever lock is shared by the b_iodone callback and the corresponding
top-half function.  Consequently, in a race, the top-half function could
conclude that operation was done before the b_iodone callback finished.
See, for example, aio_physwakeup() and aio_fphysio().

Note: I don't believe that the other, more widely-used b_iodone callbacks
are affected.

Discussed with: jeff
Reviewed by: phk
MFC after: 2 weeks
2005-07-20 19:06:06 +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
Alan Cox
b490cc72b2 In lio_listio(2) change jobref from an int to a long so that
lio_listio(LIO_WAIT, ...) works correctly on 64-bit architectures.

Reviewed by: tegge
2005-06-07 05:28:21 +00:00