dragonfly.git
5 years agoNFS - Fix refactor bugs.
Matthew Dillon [Fri, 17 Jul 2009 00:03:55 +0000 (17:03 -0700)]
NFS - Fix refactor bugs.

* The new ERROROUT macro unconditionally overwrites error.  Don't use
  it for cases where we wish to retain a prior error.

5 years agoNFS - Marathon - completely refactor the nfsm_* macros.
Matthew Dillon [Thu, 16 Jul 2009 23:21:36 +0000 (16:21 -0700)]
NFS - Marathon - completely refactor the nfsm_* macros.

Completely refactor the nfsm_*() macros in nfsm_subs.h, converting
all of them to procedures in nfsm_subs.c.

Add a nfsm_info structure to track the persistent state that the code
depends on, and remove all related local variables from the other bits
of the NFS code.

5 years agoNFS - create nfsm_subs.c, clean up externs
Matthew Dillon [Thu, 16 Jul 2009 15:19:14 +0000 (08:19 -0700)]
NFS - create nfsm_subs.c, clean up externs

* Move nfsm_*() procedures from nfs_subs.c to nfsm_subs.c

* Clean up externs improperly embedded in .c files.

5 years agoMerge branch 'master' of ssh://crater.dragonflybsd.org/repository/git/dragonfly
Matthew Dillon [Thu, 16 Jul 2009 06:47:44 +0000 (23:47 -0700)]
Merge branch 'master' of ssh://crater.dragonflybsd.org/repository/git/dragonfly

5 years agoNFS - install nfsmount on nfs_mountq, for timeout handling.
Matthew Dillon [Thu, 16 Jul 2009 06:46:07 +0000 (23:46 -0700)]
NFS - install nfsmount on nfs_mountq, for timeout handling.

* This way the timeout callout can find all the requests.

5 years agoNFS - Remove old nfsiod, start adding new kernel thread infrastructure
Matthew Dillon [Thu, 16 Jul 2009 06:21:34 +0000 (23:21 -0700)]
NFS - Remove old nfsiod, start adding new kernel thread infrastructure

* Remove all the nfsiod junk.

* Add two per-mount threads, one for reading from the socket, one for writing
  to the socket, in a new file nfs_iod.c

* Implement a quick and dirty synchronous, single threaded nfs_doio()
  loop in the writer thread to test basic mechanics.

5 years agoNFS - Minor compile fix
Matthew Dillon [Thu, 16 Jul 2009 05:37:37 +0000 (22:37 -0700)]
NFS - Minor compile fix

5 years agotsleep_interlock - Fix bug with TDF_TSLEEPQ
Matthew Dillon [Thu, 16 Jul 2009 05:25:40 +0000 (22:25 -0700)]
tsleep_interlock - Fix bug with TDF_TSLEEPQ

* LWKT previously disallowed the scheduling of a thread flagged
  TDF_TSLEEPQ because the tsleep queue was the same as the threadq.
  Now that they are different, it is possible for the thread to
  still be on a tsleepq when descheduled and later rescheduled.
  Remove the check, allowing the thread to be rescheduled by LWKT.

  This fixes issues with processes just locking up in "D"isk wait.

5 years agoNFS - move nfs_reqq from global to per-nfsmount.
Matthew Dillon [Thu, 16 Jul 2009 04:14:04 +0000 (21:14 -0700)]
NFS - move nfs_reqq from global to per-nfsmount.

5 years agoMerge branch 'master' of ssh://crater.dragonflybsd.org/repository/git/dragonfly
Matthew Dillon [Thu, 16 Jul 2009 02:48:23 +0000 (19:48 -0700)]
Merge branch 'master' of ssh://crater.dragonflybsd.org/repository/git/dragonfly

5 years agoNFS - Use mutex API, begin refactoring the state machine.
Matthew Dillon [Thu, 16 Jul 2009 02:41:34 +0000 (19:41 -0700)]
NFS - Use mutex API, begin refactoring the state machine.

* Use the mtx_*() API instead of roll-your-own locks for the send and
  receive locks.

* Refactor nfs_request().  Break the procedure up into multiple pieces
  for upcoming nfsiod/nfsd work (as in: getting rid of them).

  The main thing here is to split off the 'setup', 'send', 'receive',
  and 'process reply' parts.  This will make it easier to construct
  a kernel thread to i.e. just do the 'setup/send' part, and another
  to do the 'receive/reply' part.

5 years agoMPSAFE - mutex - better exclusive lock sequencer, bug fixes, abort
Matthew Dillon [Thu, 16 Jul 2009 02:38:39 +0000 (19:38 -0700)]
MPSAFE - mutex - better exclusive lock sequencer, bug fixes, abort

* Redo the exclusive lock chaining algorithm.  Use an explicit
  link structure and directly pass ownership to the next thread
  waiting on an exclusive lock.

* Exclusive locks can be aborted via mtx_lock_ex_link() and
  mtx_abort_ex_link().

* Lots of misc bug fixes.

5 years agoFix LINT build.
Sascha Wildner [Wed, 15 Jul 2009 20:23:43 +0000 (22:23 +0200)]
Fix LINT build.

5 years agoMPSAFE - mutexes
Matthew Dillon [Wed, 15 Jul 2009 19:13:13 +0000 (12:13 -0700)]
MPSAFE - mutexes

* Add additional inline functions to test the lock state.

5 years agoMPSAFE - mutex enhancements.
Matthew Dillon [Wed, 15 Jul 2009 19:04:28 +0000 (12:04 -0700)]
MPSAFE - mutex enhancements.

* Add timeout to mtx_lock_{sh,ex}

* Add quick version, mtx_lock_{sh,ex}_quick() which does not take a
  flags or timeout parameter.

5 years agoMerge branch 'master' of ssh://crater.dragonflybsd.org/repository/git/dragonfly
Matthew Dillon [Wed, 15 Jul 2009 18:38:50 +0000 (11:38 -0700)]
Merge branch 'master' of ssh://crater.dragonflybsd.org/repository/git/dragonfly

5 years agoMPSAFE - Add a set of general blocking/spinnable mutex functions.
Matthew Dillon [Wed, 15 Jul 2009 18:37:45 +0000 (11:37 -0700)]
MPSAFE - Add a set of general blocking/spinnable mutex functions.

* (serial wakeup code wasn't included in last commit)

5 years agoMPSAFE - Add a set of general blocking/spinnable mutex functions.
Matthew Dillon [Wed, 15 Jul 2009 18:29:43 +0000 (11:29 -0700)]
MPSAFE - Add a set of general blocking/spinnable mutex functions.

These locks are intended to eventually replace lockmgr locks for most
use cases.

* Optimized based on in-line atomic_cmpset_int() calls with fallback call.

* Recursive shared and exclusive locks.  Downgrading, and non-blocking
  upgrading.

* Interlocked wakeup flags.

* Serial wakeup for exclusive waiters (i.e. optimal in the face of a large
  number of waiting threads).  Mass-wakeup for shared waiters.

* Additional entry points for spinning.

* Ref-count support, separate from lock count.

5 years agoAdjust atomic_cmpset_int/long - Faster version, fix amd64 issue.
Matthew Dillon [Wed, 15 Jul 2009 16:44:22 +0000 (09:44 -0700)]
Adjust atomic_cmpset_int/long - Faster version, fix amd64 issue.

* Instead of using the condition code just compare %eax (or %rax)
  against the old value.  This is considerably faster then using
  sete/movzbl and GCC will also optimize the caller's test for
  both 0 or non-zero.

* AMD64.  long is 64 bits, use cmpxchgq (it was previously using cmpxchgl).

5 years agohammer expand: Also increase number of total bigblocks
Michael Neumann [Wed, 15 Jul 2009 16:34:58 +0000 (18:34 +0200)]
hammer expand: Also increase number of total bigblocks

5 years agohammer expand: Layer 1 formatting (step 2/2)
Michael Neumann [Wed, 15 Jul 2009 15:37:44 +0000 (17:37 +0200)]
hammer expand: Layer 1 formatting (step 2/2)

With this commit online HAMMER filesystem expansion is functional, but
highly experimental and might destroy your filesystem.

5 years agoMinor adjustment which does not change behaviour.
Michael Neumann [Wed, 15 Jul 2009 12:36:58 +0000 (14:36 +0200)]
Minor adjustment which does not change behaviour.

This has no effect because of the bzero(). Nevertheless it's
more logical to generate the crc after setting blocks_free to 0.

5 years agohammer expand: Layer 2 formatting (step 1.5/2)
Michael Neumann [Wed, 15 Jul 2009 11:58:39 +0000 (13:58 +0200)]
hammer expand: Layer 2 formatting (step 1.5/2)

5 years agoisdnd(8): Fix some NULL dereferences.
Sascha Wildner [Wed, 15 Jul 2009 10:56:15 +0000 (12:56 +0200)]
isdnd(8): Fix some NULL dereferences.

Taken-from: FreeBSD

5 years agopipes - set kern.pipe.mpsafe default to 1.
Matthew Dillon [Wed, 15 Jul 2009 06:23:55 +0000 (23:23 -0700)]
pipes - set kern.pipe.mpsafe default to 1.

Enable MPSAFE operation on pipes by default.

5 years agoWhitespace cleanup
Matthew Dillon [Wed, 15 Jul 2009 06:09:25 +0000 (23:09 -0700)]
Whitespace cleanup

5 years agoNFS - Make the NFS read-ahead defaults & limits more generous.
Matthew Dillon [Wed, 15 Jul 2009 06:00:52 +0000 (23:00 -0700)]
NFS - Make the NFS read-ahead defaults & limits more generous.

5 years agoundo - Add <sys/param.h> to get MAXPATHLEN
Matthew Dillon [Wed, 15 Jul 2009 03:19:13 +0000 (20:19 -0700)]
undo - Add <sys/param.h> to get MAXPATHLEN

Reported-by: corecode
5 years agoLWKT - Make sure an exiting thread is not sitting on a tsleep queue.
Matthew Dillon [Wed, 15 Jul 2009 03:09:41 +0000 (20:09 -0700)]
LWKT - Make sure an exiting thread is not sitting on a tsleep queue.

5 years agoLWKT - Rename lwkt_yield_quick() to a more appropriate splz_check().
Matthew Dillon [Wed, 15 Jul 2009 03:05:51 +0000 (20:05 -0700)]
LWKT - Rename lwkt_yield_quick() to a more appropriate splz_check().

5 years agoMPSAFE - tsleep_interlock, BUF/BIO, cluster, swap_pager.
Matthew Dillon [Wed, 15 Jul 2009 02:31:18 +0000 (19:31 -0700)]
MPSAFE - tsleep_interlock, BUF/BIO, cluster, swap_pager.

* tsleep_interlock()/tsleep() could miss wakeups during periods of
  heavy cpu activity.  What would happen is code inbetween the two
  calls would try to send an IPI (say, issue a wakeup()), but while
  sending the IPI the kernel would be forced to process incoming IPIs
  synchronous to avoid a deadlock.

  The new tsleep_interlock()/tsleep() code adds another TAILQ_ENTRY to
  the thread structure allowing tsleep_interlock() to formally place
  the thread on the appropriate sleep queue without having to deschedule
  the thread.  Any wakeup which occurs between the interlock and the
  real tsleep() call will remove the thread from the queue and the
  later tsleep() call will recognize this and simply return without sleeping.

  The new tsleep() call requires PINTERLOCKED to be passed to tsleep
  so tsleep() knows that the thread has already been placed on a sleep
  queue.

* Continue making BUF/BIO MPSAFE.  Remove B_ASYNC and B_WANT from buf->b_flag
  and add a new bio->bio_flags field to the bio.  Add BIO_SYNC, BIO_WANT,
  and BIO_DONE.  Use atomic_cmpset_int() (aka cmpxchg) to interlock
  biodone() against biowait().

  vn_strategy() and dev_dstrategy() call semantics now require that
  synchronous BIO's install a bio_done function and set BIO_SYNC in
  the bio.

* Clean up the cluster code a bit.

* Redo the swap_pager code.  Instead of issuing I/O during the collection,
  which depended on critical sections to avoid races in the cluster append,
  we now build the entire collection first and then dispatch the I/O.
  This allows us to use only async completion for the BIOs, instead of
  a hybrid sync-or-async completion.

5 years agoMerge branch 'master' of ssh://crater.dragonflybsd.org/repository/git/dragonfly
Matthew Dillon [Wed, 15 Jul 2009 02:24:43 +0000 (19:24 -0700)]
Merge branch 'master' of ssh://crater.dragonflybsd.org/repository/git/dragonfly

5 years agoMerge branch 'master' of ssh://crater.dragonflybsd.org/repository/git/dragonfly
Matthew Dillon [Wed, 15 Jul 2009 02:23:34 +0000 (19:23 -0700)]
Merge branch 'master' of ssh://crater.dragonflybsd.org/repository/git/dragonfly

5 years agoMinor adjustments of recent commits to fix UP kernel build.
Matthew Dillon [Wed, 15 Jul 2009 02:23:01 +0000 (19:23 -0700)]
Minor adjustments of recent commits to fix UP kernel build.

5 years agoFix some typos in manpages.
Sascha Wildner [Wed, 15 Jul 2009 02:09:19 +0000 (04:09 +0200)]
Fix some typos in manpages.

5 years agoMerge branch 'master' of ssh://crater.dragonflybsd.org/repository/git/dragonfly
Michael Neumann [Tue, 14 Jul 2009 17:30:15 +0000 (19:30 +0200)]
Merge branch 'master' of ssh://crater.dragonflybsd.org/repository/git/dragonfly

5 years agohammer expand: Do not assume sequential vol_no's
Michael Neumann [Tue, 14 Jul 2009 17:29:13 +0000 (19:29 +0200)]
hammer expand: Do not assume sequential vol_no's

5 years agoFix randomly seen problems with linking C++ programs.
Hasso Tepper [Tue, 14 Jul 2009 16:14:15 +0000 (19:14 +0300)]
Fix randomly seen problems with linking C++ programs.

Binutils 2.17 supports COMDAT groups and .nsubspa comdat option. It should
fix randomly seen linking errors like:

`.gnu.linkonce.t.something' referenced in section `.gnu.linkonce.r.something'
of some_object: defined in discarded section `.gnu.linkonce.t.something'
of some_object.

Digged-out-by: corecode@
5 years agoMerge branch 'master' of ssh://crater.dragonflybsd.org/repository/git/dragonfly
Michael Neumann [Tue, 14 Jul 2009 14:01:40 +0000 (16:01 +0200)]
Merge branch 'master' of ssh://crater.dragonflybsd.org/repository/git/dragonfly

5 years agoExpand HAMMER filesystem step 1/2
Michael Neumann [Tue, 14 Jul 2009 13:55:02 +0000 (15:55 +0200)]
Expand HAMMER filesystem step 1/2

Format the volume and increase other volumes' vol_count field.
The space of the new volume is not yet given to the filesystem (that's step 2).

Example:

newfs_hammer -L TEST /dev/da0
mount_hammer /dev/da0 /hammer
hammer expand /hammer /dev/da1
umount /hammer
mount_hammer /dev/da0:/dev/da1 /hammer

# the following fails (wrong number of volumes)
mount_hammer /dev/da0 /hammer

5 years agoFix typos in mse(4) and ppp(8) manpages: compliment -> complement.
Sascha Wildner [Tue, 14 Jul 2009 12:54:36 +0000 (14:54 +0200)]
Fix typos in mse(4) and ppp(8) manpages: compliment -> complement.

5 years agoShuffle IPI handlers to the last two blocks of the IDT.
Sepherosa Ziehau [Mon, 13 Jul 2009 13:44:11 +0000 (21:44 +0800)]
Shuffle IPI handlers to the last two blocks of the IDT.

Relative position of the IPI handlers is kept.  This gives us
more contiguous free IDT vectors.

5 years agoAdd age(4) to GENERIC and LINT
Sepherosa Ziehau [Mon, 13 Jul 2009 06:10:30 +0000 (14:10 +0800)]
Add age(4) to GENERIC and LINT

5 years agoAdd ae(4) manpage
Sepherosa Ziehau [Mon, 13 Jul 2009 06:06:44 +0000 (14:06 +0800)]
Add ae(4) manpage

Submitted-by: Alexander Polakov <polachok@gmail.com>
Obtained-from: FreeBSD

5 years agoAdd ae(4) to LINT and GENERIC
Sepherosa Ziehau [Mon, 13 Jul 2009 06:03:24 +0000 (14:03 +0800)]
Add ae(4) to LINT and GENERIC

5 years agoAdd ae(4) into module building
Sepherosa Ziehau [Mon, 13 Jul 2009 05:51:23 +0000 (13:51 +0800)]
Add ae(4) into module building

5 years agoAdd driver for Attansic/Atheros L2 FastEthernet
Sepherosa Ziehau [Mon, 13 Jul 2009 05:45:25 +0000 (13:45 +0800)]
Add driver for Attansic/Atheros L2 FastEthernet

Submitted-by: Alexander Polakov <polachok@gmail.com> w/ modification
Obtained-from: FreeBSD

5 years agolapic_init() -> lapic_map()
Sepherosa Ziehau [Sun, 12 Jul 2009 08:05:49 +0000 (16:05 +0800)]
lapic_init() -> lapic_map()

5 years agoRemove the distribute, redistribute and distributeworld targets.
Sascha Wildner [Tue, 14 Jul 2009 12:01:34 +0000 (14:01 +0200)]
Remove the distribute, redistribute and distributeworld targets.

We don't split things into different distributions and none of this
was ever used.

Pointed-out-by: pavalos
5 years agoMerge branch 'master' of ssh://crater.dragonflybsd.org/repository/git/dragonfly
Matthew Dillon [Tue, 14 Jul 2009 10:36:06 +0000 (03:36 -0700)]
Merge branch 'master' of ssh://crater.dragonflybsd.org/repository/git/dragonfly

5 years agotsleep() - Use PINTERLOCKED in msleep() and serialized_sleep().
Matthew Dillon [Tue, 14 Jul 2009 10:34:56 +0000 (03:34 -0700)]
tsleep() - Use PINTERLOCKED in msleep() and serialized_sleep().

5 years agotermios.h: Expose tcgetsid(3) function.
Stathis Kamperis [Wed, 1 Jul 2009 20:49:43 +0000 (23:49 +0300)]
termios.h: Expose tcgetsid(3) function.

Dragonfly-bug: <http://bugs.dragonflybsd.org/issue1419>
Reviewed-by: hasso@, corecode@
5 years agotsleep() - Add PINTERLOCKED flag to catch edge case.
Matthew Dillon [Tue, 14 Jul 2009 10:14:33 +0000 (03:14 -0700)]
tsleep() - Add PINTERLOCKED flag to catch edge case.

When the tsleep_interlock() + UNLOCK + tsleep() combination is used it is
possible for an incoming wakeup IPI to be processed even if the combination
is used within a critical section, because operations inbetween the two
may send an IPI.  Under heavy loads sending an IPI can force incoming IPIs
to be processed synchronously to avoid deadlocks.

It is also possible for tsleep itself to create this condition when it
releases the user process schedule prior to descheduling itself.

PINTERLOCKED causes tsleep to check whether the bit set by tsleep_interlock()
is still set.  If it is not set we simply return without sleeping.

5 years agoHAMMER - new hammer_lock code, fix mplock bug in last commit, mpsafe getattr.
Matthew Dillon [Tue, 14 Jul 2009 07:14:12 +0000 (00:14 -0700)]
HAMMER - new hammer_lock code, fix mplock bug in last commit, mpsafe getattr.

* Fix a mplock bug in the last commit that can cause a panic in hammer_read.
  The mplock check code wasn't setting up the hammer transaction properly.

* VMP_GETATTR is now set on all HAMMER vnodes.  If sysctl vfs.getattr_mpsafe=1
  all fstat() operations will run without the MP lock.

* Rewrite HAMMER's lock code, making it MPSAFE.  The new lock code uses
  atomic_cmpset_int().

5 years agoMPSAFE - Regenerate system calls for MPSAFE changes.
Matthew Dillon [Tue, 14 Jul 2009 07:08:11 +0000 (00:08 -0700)]
MPSAFE - Regenerate system calls for MPSAFE changes.

5 years agoMPSAFE - Implement sysctl vfs.getattr_mpsafe, make fstat() MPSAFE
Matthew Dillon [Tue, 14 Jul 2009 06:02:21 +0000 (23:02 -0700)]
MPSAFE - Implement sysctl vfs.getattr_mpsafe, make fstat() MPSAFE

Make fstat() MPSAFE.  Adjust fileops->fo_stat functions to be MPSAFE as
necessary (uipc socket stat mainly).

5 years agoMPSAFE - Document a few procedures as MPSAFE (non-operational)
Matthew Dillon [Tue, 14 Jul 2009 05:58:59 +0000 (22:58 -0700)]
MPSAFE - Document a few procedures as MPSAFE (non-operational)

5 years agoMPSAFE - system calls set_tls_area() and get_tls_area() are MPSAFE
Matthew Dillon [Tue, 14 Jul 2009 05:26:53 +0000 (22:26 -0700)]
MPSAFE - system calls set_tls_area() and get_tls_area() are MPSAFE

5 years agoMerge branch 'master' of ssh://crater.dragonflybsd.org/repository/git/dragonfly
Matthew Dillon [Tue, 14 Jul 2009 03:49:10 +0000 (20:49 -0700)]
Merge branch 'master' of ssh://crater.dragonflybsd.org/repository/git/dragonfly

5 years agoHAMMER - Add MPSAFE suppor for hammer_vop_read().
Matthew Dillon [Tue, 14 Jul 2009 03:46:45 +0000 (20:46 -0700)]
HAMMER - Add MPSAFE suppor for hammer_vop_read().

This is still a bit of a hack.  If vfs.read_mpsafe is set HAMMER will
run MPSAFE for reads from regular files which are cached in the buffer
cache.  Data which is not cached will fall-back to getting the MP lock.

At the moment enabling this operation will cause atime to not get updated
on fully cached files, because the rest of HAMMER is not MPSAFE yet so I
cannot modify the internal hammer_inode state.

5 years agoMPSAFE - Add read() and write() path MPSAFE support, and more.
Matthew Dillon [Tue, 14 Jul 2009 03:41:30 +0000 (20:41 -0700)]
MPSAFE - Add read() and write() path MPSAFE support, and more.

* Add sysctl vfs.read_mpsafe, vfs.write_mpsafe, and vfs.getattr_mpsafe.
  Only vfs.read_mpsafe is currently implemented, and only for HAMMER
  filesystems.

* read() and write() will now call VOP_READ()/VOP_WRITE() without the
  MP lock if the vnode->v_flag says it is ok to do so.

* read() and write() now serialize reads and writes which depend on the
  file offset stored in the fp, when the concurrent operations share the
  same fp.

* Add vnode flags VMP_READ, VMP_WRITE, VMP_GETATTR.

* Add vnode flag VNOTSEEKABLE indicating that fp->f_offset will never be
  used and no serialization is needed (which e.g. can mess up read()'s
  blocking devices such as ttys).

* Correct bug in FIONREAD related to the data field being too small.

5 years agomxge: Clean up Makefile a little more. v2.3.2
Sascha Wildner [Tue, 14 Jul 2009 02:04:49 +0000 (04:04 +0200)]
mxge: Clean up Makefile a little more.

5 years agomxge: Add opt_inet.h to the Makefile.
Sascha Wildner [Tue, 14 Jul 2009 01:42:10 +0000 (03:42 +0200)]
mxge: Add opt_inet.h to the Makefile.

5 years agoCCMS - Correct bug, add #include <sys/spinlock2.h> to kern_ccms.c
Matthew Dillon [Tue, 14 Jul 2009 00:19:09 +0000 (17:19 -0700)]
CCMS - Correct bug, add #include <sys/spinlock2.h> to kern_ccms.c

5 years agoCCMS - bug in last commit, remove #include <sys/spinlock2.h> from ccms.h
Matthew Dillon [Tue, 14 Jul 2009 00:17:21 +0000 (17:17 -0700)]
CCMS - bug in last commit, remove #include <sys/spinlock2.h> from ccms.h

5 years agoNormalize the defines for the vnode flags (non-operational change).
Matthew Dillon [Mon, 13 Jul 2009 23:48:45 +0000 (16:48 -0700)]
Normalize the defines for the vnode flags (non-operational change).

5 years agoCCMS - Make the cache coherency locks MPSAFE.
Matthew Dillon [Mon, 13 Jul 2009 23:44:51 +0000 (16:44 -0700)]
CCMS - Make the cache coherency locks MPSAFE.

5 years agoBUF/BIO - MPSAFE work on core buffer cache routines.
Matthew Dillon [Mon, 13 Jul 2009 22:57:53 +0000 (15:57 -0700)]
BUF/BIO - MPSAFE work on core buffer cache routines.

Do initial MPSAFE work on getblk(), bread(), brelse(), bqrelse(),
biowait(), and friends.  Use get_mplock()/rel_mplock() in areas not
yet converted.

Add a flags argument to findblk() and have it return a locked buffer
cache buffer by default.  Callers may request that the buffer be
locked non-blocking, or that it not be locked at all.

Adjust all callers of findblk() for those cases where an unlocked
buffer is desired to lock and re-check the returned buffer's parameters,
since otherwise it can change out from under the caller whether a critical
section is used or not (for SMP).

5 years agotelnet: Fix a bug (really use the return value of tty_istrapsig()).
Sascha Wildner [Mon, 13 Jul 2009 22:03:33 +0000 (00:03 +0200)]
telnet: Fix a bug (really use the return value of tty_istrapsig()).

5 years agoExpand HAMMER filesystem stubs
Michael Neumann [Mon, 13 Jul 2009 21:40:38 +0000 (23:40 +0200)]
Expand HAMMER filesystem stubs

5 years agopipe - add required critical section around tsleep_interlock() call.
Matthew Dillon [Mon, 13 Jul 2009 21:11:11 +0000 (14:11 -0700)]
pipe - add required critical section around tsleep_interlock() call.

5 years agobuffer cache - protect bufqueues[] with the bufspin spinlock.
Matthew Dillon [Mon, 13 Jul 2009 20:41:12 +0000 (13:41 -0700)]
buffer cache - protect bufqueues[] with the bufspin spinlock.

5 years agomxge.4: Some small changes.
Sascha Wildner [Mon, 13 Jul 2009 19:16:16 +0000 (21:16 +0200)]
mxge.4: Some small changes.

* We use local time line, not FreeBSD's.

* Kick unneeded .Pp.

* First appeared in 2.3.

5 years agoMerge branch 'mxge'
Aggelos Economopoulos [Mon, 13 Jul 2009 18:47:37 +0000 (21:47 +0300)]
Merge branch 'mxge'

5 years agomxge.4: sync with the status of the driver in DragonFly
Aggelos Economopoulos [Mon, 13 Jul 2009 18:46:35 +0000 (21:46 +0300)]
mxge.4: sync with the status of the driver in DragonFly

5 years agomxge: import mxge.4 from FreeBSD
Aggelos Economopoulos [Mon, 13 Jul 2009 18:45:09 +0000 (21:45 +0300)]
mxge: import mxge.4 from FreeBSD

5 years agoDocument pmap_qremove() as being MPSAFE.
Matthew Dillon [Mon, 13 Jul 2009 18:39:03 +0000 (11:39 -0700)]
Document pmap_qremove() as being MPSAFE.

5 years agoMake getpbuf(), trypbuf(), and relpbuf() MPSAFE.
Matthew Dillon [Mon, 13 Jul 2009 18:38:09 +0000 (11:38 -0700)]
Make getpbuf(), trypbuf(), and relpbuf() MPSAFE.

5 years agoMake biowait() MPSAFE.
Matthew Dillon [Mon, 13 Jul 2009 18:37:35 +0000 (11:37 -0700)]
Make biowait() MPSAFE.

5 years agotelnet.1: Remove unnecessary .Pp
Sascha Wildner [Mon, 13 Jul 2009 18:07:54 +0000 (20:07 +0200)]
telnet.1: Remove unnecessary .Pp

5 years agopriv.9: Update for recent changes.
Sascha Wildner [Mon, 13 Jul 2009 17:58:58 +0000 (19:58 +0200)]
priv.9: Update for recent changes.

Approved-by: mneumann
5 years agobuffer cache - Control all access to the buf red-black trees with vp->v_token
Matthew Dillon [Mon, 13 Jul 2009 17:55:24 +0000 (10:55 -0700)]
buffer cache - Control all access to the buf red-black trees with vp->v_token

Access to the buffer cache's RB trees is now controlled via vp->v_token
instead of a critical section.  We still hold the BGL but this is not
quite as simple as it seems because an interrupt calling biodone() on a
B_INVAL buffer may now potentially block, where as it would not have before.
The buffer is locked.

5 years agolwkt_token - Add ASSERT_LWKT_TOKEN_HELD(token) macro.
Matthew Dillon [Mon, 13 Jul 2009 17:53:03 +0000 (10:53 -0700)]
lwkt_token - Add ASSERT_LWKT_TOKEN_HELD(token) macro.

5 years agobio_track - fix bug in last commit (the negative count assertion was wrong)
Matthew Dillon [Mon, 13 Jul 2009 17:42:34 +0000 (10:42 -0700)]
bio_track - fix bug in last commit (the negative count assertion was wrong)

5 years agoMake operations on struct bio_track MPSAFE.
Matthew Dillon [Mon, 13 Jul 2009 16:30:22 +0000 (09:30 -0700)]
Make operations on struct bio_track MPSAFE.

Use atomic ops to make reference counting and waiting on bio_track structures
MPSAFE.

5 years agomxge: use ether_input_chain()
Aggelos Economopoulos [Mon, 13 Jul 2009 15:12:38 +0000 (18:12 +0300)]
mxge: use ether_input_chain()

5 years agomxge: call bus_setup_intr() after the if serializer has been set
Aggelos Economopoulos [Mon, 13 Jul 2009 15:11:10 +0000 (18:11 +0300)]
mxge: call bus_setup_intr() after the if serializer has been set

5 years agomxge: properly initialize m_pkthdr.len as well
Aggelos Economopoulos [Mon, 13 Jul 2009 15:10:01 +0000 (18:10 +0300)]
mxge: properly initialize m_pkthdr.len as well

5 years agomxge: use normal ethernet mtu by default
Aggelos Economopoulos [Mon, 13 Jul 2009 15:05:50 +0000 (18:05 +0300)]
mxge: use normal ethernet mtu by default

We don't properly support jumbo frames yet, so
change the default mtu

5 years agomxge: properly remove the sysctls
Aggelos Economopoulos [Mon, 13 Jul 2009 15:04:19 +0000 (18:04 +0300)]
mxge: properly remove the sysctls

5 years agovnode - move v_pollinfo.vpi_token to v_token
Matthew Dillon [Mon, 13 Jul 2009 14:57:26 +0000 (07:57 -0700)]
vnode - move v_pollinfo.vpi_token to v_token

* Rename the lwkt_token used for the vnode's v_pollinfo for general use by
  the vnode.  This token will soon be used to lock the buffer cache.

5 years agopipe - fix UP build error.
Matthew Dillon [Mon, 13 Jul 2009 14:40:48 +0000 (07:40 -0700)]
pipe - fix UP build error.

Reported-by: Hasso Tepper <hasso@estpak.ee>
5 years agoSYSPERF - update pipe1, pipe2
Matthew Dillon [Mon, 13 Jul 2009 05:44:08 +0000 (22:44 -0700)]
SYSPERF - update pipe1, pipe2

* pipe1 now does a 1-second timing run.

* pipe2 now does a 2-second timing run, just so it runs just a tad longer.

5 years agoMerge branch 'master' of ssh://crater.dragonflybsd.org/repository/git/dragonfly
Matthew Dillon [Mon, 13 Jul 2009 05:24:54 +0000 (22:24 -0700)]
Merge branch 'master' of ssh://crater.dragonflybsd.org/repository/git/dragonfly

5 years agopipe - replace use of DELAY() with a better solution, add features.
Matthew Dillon [Mon, 13 Jul 2009 05:14:43 +0000 (22:14 -0700)]
pipe - replace use of DELAY() with a better solution, add features.

* Use the new tsc functions to poll for new read data in a pipe for a short
  period of time on a SMP box.  This greatly increases the odds of a pipe
  writer on one cpu being able to pipeline data to a reader on another cpu
  without having to use an IPI or tsleep/wakeup.

  For the pipe1 test this brings the synchronous communications path
  over a pipe (Awrite, Bread, Bwrite, Aread) down from 7uS to around 2uS.

  For the pipe2 test this value greatly reduces the number of IPIs
  and improves bandwidth by a few hundred megabytes/sec (the old DELAY
  did the same thing so there is no change for the pipe2 test).

* Add sysctl kern.pipe.delay which defaults to 5000 nanoseconds (5uS).
  This is the maximum a pipe reader will wait for additional data before
  falling back to tsleep/wakeup (and related ipis).  pipe_delay may be
  set to 0 to disable the function.  I value of at least 3000 is
  recomended.  Pipelining large buffers efficiently requires a higher value,
  say up to 8000 or so.

* Allow kern.pipe.mpsafe to be set to 2 which adds a predictive wakeup
  when a writer is found to be stalled.  This currently has no significant
  effect on operations due to token collisions.

* Add statistics: kern.pipe.wblocked and kern.pipe.rblocked, counting
  the number of times a pipe blocks in "pipewr" or "piperd".

* Fix MP races in pipe_ioctl().

5 years agoclocks - Add fine-grained spin delay function via TSC
Matthew Dillon [Mon, 13 Jul 2009 05:12:50 +0000 (22:12 -0700)]
clocks - Add fine-grained spin delay function via TSC

Add int64_t target = tsc_get_target(ns) and tsc_test_target(target).
See routines for details.  These functions are available when the system
supports an extremely fine-grained counter such as a TSC and may be used
to generate finely tuned delays.

5 years agolibtelnet: Fix some mis-usage of NULL and raise WARNS to 2.
Sascha Wildner [Mon, 13 Jul 2009 02:41:45 +0000 (04:41 +0200)]
libtelnet: Fix some mis-usage of NULL and raise WARNS to 2.

5 years agoOops! Actually install /usr/libexec/telnetd.
Peter Avalos [Mon, 13 Jul 2009 02:31:54 +0000 (16:31 -1000)]
Oops!  Actually install /usr/libexec/telnetd.

Noticed-by: swildner
5 years agobootpgw: Use err(3).
Peter Avalos [Mon, 13 Jul 2009 02:23:04 +0000 (16:23 -1000)]
bootpgw:  Use err(3).

Obtained-from: FreeBSD

5 years agopipe - Fix piperd/pipewr deadlock in mpsafe case
Matthew Dillon [Mon, 13 Jul 2009 01:37:46 +0000 (18:37 -0700)]
pipe - Fix piperd/pipewr deadlock in mpsafe case

* Fix a bug where piperd and pipewr could be entered simultaniously
  by the reader and writer and neither gets woken up.

* Redo the hysteresis slightly to reduce IPIs even more.