Commit Graph

9268 Commits

Author SHA1 Message Date
Craig Rodrigues
0c89bb0a02 Add "update" mount option to global_opts array,
for use with vfs_filteropt().
2006-05-26 02:38:48 +00:00
Craig Rodrigues
5eb304a91a Remove calls to vfs_export() for exporting a filesystem for NFS mounting
from individual filesystems.  Call it instead in vfs_mount.c,
after we call VFS_MOUNT() for a specific filesystem.
2006-05-26 00:32:21 +00:00
Robert Watson
20bdac8a4f Use getsock() and fput() instead of fgetsock() and fputsock() in
sendfile().  This causes sendfile() to use the file descriptor
reference to the socket instead of bumping the socket reference
count, which avoids an additional refcount operation, as well as a
potential expensive socket refcount drop, which can lead to
contention on the accept mutex.  This change also has the side
effect of further reducing the number of cases where an in-progress
I/O operation can occur on a socket after close, as using the file
descriptor refcount prevents the socket from closing while in use.

MFC after:	3 months
2006-05-25 15:10:13 +00:00
Stephan Uphoff
dcf67e65d2 Do not set B_NOCACHE on buffers when releasing them in flushbuflist().
If B_NOCACHE is set the pages of vm backed buffers will be invalidated.
However clean buffers can be backed by dirty VM pages so invalidating them
can lead to data loss.
Add support for flush dirty page in the data invalidation function
of some network file systems.

This fixes data losses during vnode recycling (and other code paths
using invalbuf(*,V_SAVE,*,*)) for data written using an mmaped file.

Collaborative effort by: jhb@,mohans@,peter@,ps@,ups@
Reviewed by:	tegge@
MFC after:	7 days
2006-05-25 01:00:35 +00:00
Sam Leffler
75b773ae3d When starting up threads in taskqueue_start_threads create them
stopped before adjusting their priority and setting them on the run
q so they cannot race for resources (pointed out by njl).

While here add a console printf on thread create fails; otherwise
noone may notice (e.g. return value is always 0 and caller has no
way to verify).

Reviewed by:	jhb, scottl
MFC after:	2 weeks
2006-05-24 22:11:07 +00:00
David Xu
f705bbe8b1 Don't allow non-root user to set a scheduler policy, otherwise this could
be a local DOS.

Submitted by: Diane Bruce at db at db.net
2006-05-21 00:40:38 +00:00
David Xu
f6c040a2c5 Style fixes.
Submitted by: Diane Bruce < db at db dot net >
2006-05-19 06:37:24 +00:00
David Xu
7b8d821268 Move flag TDF_UMTXQ into structure umtxq, this eliminates the requirement
of scheduler lock in some umtx code.
2006-05-18 08:43:46 +00:00
Poul-Henning Kamp
d595182f0b Make the printfs relating to purging threads from a device less intrusive. 2006-05-17 06:37:14 +00:00
Poul-Henning Kamp
c40da00ca3 Since DELAY() was moved, most <machine/clock.h> #includes have been
unnecessary.
2006-05-16 14:37:58 +00:00
Paul Saab
6befa6ae1b Allow concurrent read(2)/readv(2) access to a file.
Lock file offset against multiple read calls.

Submitted by:	ups
Obtained from:	Yahoo!
MFC after:	2 weeks
2006-05-16 07:50:54 +00:00
Kelly Yancey
c9ad8a67af Restore the ability to mount procfs and fdescfs filesystems via the
mount(2) system call:

  * Add cmount hook to fdescfs and pseudofs (and, by extension, procfs and
    linprocfs).  This (mostly) restores the ability to mount these
    filesystems using the old mount(2) system call (see below for the
    rest of the fix).

  * Remove not-NULL check for the data argument from the mount(2) entry
    point.  Per the mount(2) man page, it is up to the individual
    filesystem being mounted to verify data.  Or, in the case of procfs,
    etc. the filesystem is free to ignore the data parameter if it does
    not use it.  Enforcing data to be not-NULL in the mount(2) system call
    entry point prevented passing NULL to filesystems which ignored the
    data pointer value.  Apparently, passing NULL was common practice
    in such cases, as even our own mount_std(8) used to do it in the
    pre-nmount(2) world.

All userland programs in the tree were converted to nmount(2) long ago,
but I've found at least one external program which broke due to this
(presumably unintentional) mount(2) API change.  One could argue that
external programs should also be converted to nmount(2), but then there
isn't much point in keeping the mount(2) interface for backward
compatibility if it isn't backward compatible.
2006-05-15 19:42:10 +00:00
Benno Rice
77fe443878 The VERBOSE_SYSINIT stuff sees the DDB define a lot better if we include
opt_ddb.h.

Spotted by:	benno
Pointy hat to:	benno
2006-05-14 07:11:28 +00:00
Craig Rodrigues
5250012a1d For nmount(), if "rw" is specified as a mount option,
add "noro" to the list of mount options.  This allows
a read-only mount to be converted to read-write via:
mount -u -o rw

Requested by:	kris
2006-05-14 01:51:38 +00:00
John Baldwin
73dbd3da73 Remove various bits of conditional Alpha code and fixup a few comments. 2006-05-12 05:04:46 +00:00
Benno Rice
26ab616fdc Add a new kernel config option, VERBOSE_SYSINIT.
When porting FreeBSD to a new platform, one of the more useful things to do is
get mi_startup() to let you know which SYSINIT it's up to.  Most people tend to
whack a printf in the SYSINIT loop to print the address of the function it's
about to call.  Going one better, jhb made a version that uses DDB to look up
the name of the function and print that instead.  This version is essentially
his with the addition of some ifdeffery to make it optional and to allow it to
work (although using only the function address, not the symbol) if you forgot
to enable DDB.

All the cool bits by:	jhb
Approved by:		scottl, rink, cognet, imp
2006-05-12 02:01:38 +00:00
Poul-Henning Kamp
99ab8292c7 Remove more straggling CPU_ macro references 2006-05-11 17:53:26 +00:00
David Xu
005efcdb0e Use wakeup_one to avoid thundering herd.
Tested by: kris
2006-05-09 13:00:46 +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
Tor Egge
11991ab418 Call vn_finished_write() before calling the coredump handler which will
indirectly call vn_start_write() as necessary for each write.
2006-05-07 22:50:22 +00:00
Tor Egge
d302786c87 Temporarily unlock vnode for new image being executed to avoid lock order
reversals that can lead to deadlocks.  Normally vn_close(), namei() or vrele()
should not be called while holding vnode locks.
2006-05-05 20:25:05 +00:00
Pawel Jakub Dawidek
643df192de vn_start_write()/vn_finished_write() is not needed here, because
vn_start_write() is always called earlier in the code path and calling
the function recursively may lead to a deadlock.

Confirmed by:	tegge
MFC after:	2 weeks
2006-04-29 21:57:38 +00:00
Kris Kennaway
cef31ff7d9 Lock giant when assigning ni_vp and keep vfslocked state valid.
Committed for:	jeff
2006-04-29 07:13:49 +00:00
Pawel Jakub Dawidek
122410eea2 vn_start_write() is called only when v_type != VCHR, so corresponding
vn_finished_write() should also be called only then.

BTW. I fixed two functions here: vn_rdwr() and vn_write(). The latter seems
to be unused.

MFC after:	3 weeks
2006-04-28 21:54:05 +00:00
Robert Watson
3bf14fd5e9 Also check use_pty in the ptmx clone lookup; this means that when ptmx
support is turned off using the sysctl, we no longer even allow the
ptmx device to be looked up.

Foot provided by:	peter
2006-04-28 21:39:57 +00:00
Marcel Moolenaar
8f405ed335 Remove the puc-specific hacks. The puc(4) driver now properly uses
the rman(9) interface.
2006-04-28 21:23:09 +00:00
Jeff Roberson
6ca9fcc586 - Add a BO_NEEDSGIANT flag to the bufobj. This flag forces all child
buffers to go on the buf daemon's DIRTYGIANT queue.
 - Set BO_NEEDSGIANT on ffs's devvp since the ffs_copyonwrite handler
   runs in the context of the buf daemon and may require Giant.
2006-04-28 01:05:31 +00:00
Jeff Roberson
4b5b86816c - Consistently track ni_dvp and ni_vp with dvfslocked and vfslocked rather
than trying to optimize it into a single lock.  This adds more calls to
   lock giant with non smpsafe filesystems but is the only way to reliably
   hold the correct lock.
 - Remove an invalid assert in the mountedhere case in lookup and fix the
   code to properly deal with the scenario.  We can actually have a lookup
   that returns dp == dvp with mountedhere set with certain unmount races.

Tested by:	kris
Reported by:	kris/mohans
2006-04-28 00:59:48 +00:00
John-Mark Gurney
5c06d111b8 back out for now... revert ccpu to being kern.ccpu... 2006-04-27 17:57:59 +00:00
John-Mark Gurney
c71ce6a445 move remaining sysctl into the kern.sched tree... 2006-04-26 19:42:38 +00:00
John Baldwin
ae110b53d1 Add some new commands to hopefully make it easier to diagnose lock-related
problems in ddb:
- "show threadchain [thread]" will start with the specified thread (or the
  current kdb thread by default) and show it's state.  If it is blocked on
  a lock, it will find the owner of the lock and show its state, etc.
- "show allchains" will find all of the threads that are blocked on a
  lock (but do not have any threads blocked on a lock they hold) and show
  the resulting thread chain.
- "show lockchain <lock>" takes a pointer to a lock_object (such as a
  mutex or rwlock).  If there is a turnstile for that lock, then it will
  display all the threads blocked on the lock.  In addition, for each
  thread blocked on the lock, it will display any contested locks they
  hold, and recurse on those locks to show any threads blocked on those
  locks, etc.
2006-04-25 20:28:17 +00:00
John Baldwin
de833b7c0c Use db_lookup_thread() to lookup the thread for the passed in address
and change 'show locks' to only list the locks for a given thread
rather than for all the threads in the process containing a specified
thread.
2006-04-25 20:24:23 +00:00
Marius Strobl
fa63296aba Remove last vestiges of sab(4). 2006-04-25 19:43:53 +00:00
Robert Watson
102ea03373 Extend getsock() to return the struct file flags read while holding the
file lock, in the style of fgetsock().

Modify accept1() to use getsock() instead of fgetsock(), relying on the
file descriptor reference rather than an acquired socket reference to
prevent the listen socket from being destroyed during accept().  This
avoids additional reference count operations, which should improve
performance, and also avoids accept1() operating on a socket whose file
descriptor has been torn down, which may have resulted in protocol
shutdown starting.

MFC after:	3 months
2006-04-25 11:48:16 +00:00
Maxim Konovalov
481f8fe85f Inherit LOCAL_CREDS option from listen socket for sockets returned
by accept(2).

PR:		kern/90644
Submitted by:	Andrey Simonenko
OK'ed by:	mdodd
Tested by:	NetBSD regress/sys/kern/unfdpass/unfdpass.c
MFC after:	1 month
2006-04-24 19:09:33 +00:00
Marcel Moolenaar
845652dd28 MFp4: Add the ipend() method to the serdev I/F to allow umbrella
drivers to obtain pending interrupt status from subordinate
	drivers.
2006-04-23 22:12:39 +00:00
Robert Watson
0cec9959e8 Assert that sockets passed into soabort() not be SQ_COMP or SQ_INCOMP,
since that removal should have been done a layer up.

MFC after:	3 months
2006-04-23 18:15:54 +00:00
Robert Watson
28ea180136 Add missing 'not' to SQ_COMP comment.
MFC after:	3 months
2006-04-23 15:37:23 +00:00
Robert Watson
6ca35d4b81 Move handling of SQ_COMP exception case in sofree() to the top of the
function along with the remainder of the reference checking code.  Move
comment from body to header with remainder of comments.  Inclusion of a
socket in a completed connection queue counts as a true reference, and
should not be handled as an under-documented edge case.

MFC after:	3 months
2006-04-23 15:33:38 +00:00
John Baldwin
f9ab2f134f Print td_name instead of p_comm if td_name is non-empty for
'show turnstile' and 'show sleepq'.
2006-04-21 20:40:43 +00:00
Paul Saab
95f16c1e2c Don't try to kill embryonic processes in killpg1(). This prevents
a race condition between fork() and kill(pid,sig) with pid < 0 that
can cause a kernel panic.

Submitted by:	up
MFC after:	3 weeks
2006-04-21 19:26:21 +00:00
Paul Saab
4f590175b7 Allow for nmbclusters and maxsockets to be increased via sysctl.
An eventhandler is used to update all the various zones that depend
on these values.
2006-04-21 09:25:40 +00:00
John-Mark Gurney
be4db476a6 const'ify resource_spec to note that we won't be changing anything while
releasing resources... also, NULL out the resources as we free them...
2006-04-20 01:44:16 +00:00
Warner Losh
0385d64761 r_spare1 and r_spare2 aren't needed. They aren't used. They can't be
accessed from outside of subr_rman.c.  Remove them.

Reviewed by: jmg (in theory)
2006-04-19 21:25:55 +00:00
John Baldwin
fea3efe5bf Implement rw_try_upgrade() and rw_downgrade(). rw_try_upgrade() makes a
single attempt at upgrading a read lock to a write lock, and rw_downgrade()
converts curthread's write lock into a read lock.
2006-04-19 21:06:52 +00:00
Wojciech A. Koszek
5884c1a098 'owner' is not used without SMP. Fix kernel build for such kernel
configurations.

Approved by:	jhb
2006-04-18 20:32:42 +00:00
John Baldwin
efa86db61d Adaptively spin before blocking on the turnstile if an rwlock is write
locked.  In general the adaptive spinning is similar to the same code
for mutexes with some extra trickiness in rw_wunlock_hard().  Specifically,
even though both wait bits might be set and we might have a turnstile with
at least one waiting thread, there might not be any threads blocked on the
queue we are not waking up (they might all be spinning), and we should
only preserve the waiting flag for the queue we aren't waking up if there
are in fact threads blocked on that queue.  Secondly, there might not be
any threads blocked on the queue we have chosen to waken threads from
(there might only be threads blocked on the other queue and the threads
for this queue are all spinning) in which case we disown the turnstile
instead of doing a braodcast and unpend.
2006-04-18 18:27:54 +00:00
John Baldwin
f1a4b852dc - Bring back turnstile_empty() which can check to see if an individual
queue on a turnstile is empty.
- Add a turnstile_disown() function that allows a thread to give up
  ownership of a turnstile w/o waking up any waiters.
2006-04-18 18:16:54 +00:00
Xin LI
4207c279d4 In vfs_hash_get(): mount point should never be changed
so explicitly constify the mp parameter.

Reviewed by:	phk
2006-04-18 08:05:08 +00:00
John Baldwin
38bf165fa1 - Add a rw_wowner() macro that just returns the owner of a write lock and
use it in places that only care about the write owner instead of
  rw_owner() as a baby step towards limited read-lock owner.
- Tidy the code that sets the WAITER flag bits to not duplicate a test
  around the atomic operation and the KTR trace in both of the lock
  functions.
2006-04-17 21:11:01 +00:00