Commit Graph

142596 Commits

Author SHA1 Message Date
scottl
6de3633d94 Move the CAM passthrough code into a true module so that it doesn't have to be
compiled into the main AMR driver.  It's code that is nice to have but not
required for normal operation, and it is reported to cause problems for some
people.
2008-11-03 00:53:54 +00:00
rwatson
bd5c63c83e Catch up with netsmb locking: explicit thread arguments no longer required. 2008-11-02 23:20:27 +00:00
rwatson
4de0ed8a13 Catch up internal locking routines in netsmb with lockmgr changes --
explicit thread arguments are no longer required in many places.
2008-11-02 23:15:32 +00:00
ivoras
d819bb20f8 Increase the initial sbuf size for CPU topology dump to something more
usable for newer CPUs. The new value allows 2 x quad core configuration
dumps to fit within the initial buffer without reallocations.

Approved by:	gnn (mentor) (older version)
Pointed out by:	rdivacky
2008-11-02 23:11:20 +00:00
rwatson
8287eede82 smb_vc_put() requires that the passed vcp be locked, so lock it before
dropping the connection when the requested service isn't available, or
we may try to release a lock that isn't locked.

This prevents an assertion failure when trying to mount a non-present
share using smbfs with INVARIANTS; a lock order reversal warning that
immediately follows is not yet fixed.

Reported by:	attilio
MFC after:	3 days
2008-11-02 20:22:24 +00:00
rwatson
b563afb164 When encoding an smb name, truncate one byte earlier in order than we did
previously in order to ensure it fit properly in the bufer when encoded.
This prevents a debugging printf from firing if a source or destination
host name for an smb mount exceeds 15 characters.

MFC after:	3 days
Obtained from:	Apple, Inc.
2008-11-02 19:48:15 +00:00
kib
42fb388946 Revert r184516. Option RL_TWISTER_ENABLE is no more after it became
loader tunable.

Pointy hat to:	me
2008-11-02 19:40:24 +00:00
ed
8df5c9ee67 Make the touch pad on my PowerBook G4 12" a little more usable.
For an unknown reason the touch pad of my PowerBook generates button 5
events when you operate it. This causes the adb_mouse code to convert
them to button 2 events, which is not what we want.

Add a new flag, AMS_TOUCHPAD, which is used to distinguish the touch
pad. When set, don't convert button events of unknown buttons to the
last button.

There are still three problems left with respect to user input:

- The mouse button events are not properly processed when the touch pad
  isn't touched.

- The arrow keys on the keyboard don't work inside X11.

- The power button isn't handled by the kernel, similar to the ACPI
  power button on i386/amd64.

Approved by:	nwhitehorn
2008-11-02 19:08:10 +00:00
imp
e4a289ae9d MFp4:
Make the ISA bus keep track of more PNP details.  Plus a minor style
fix while I'm here.  More could be done here, but except for some SBCs
that don't have ACPI, there's limited value to anybody in doing so.
2008-11-02 18:48:54 +00:00
imp
e0025a35ca Make the no driver stuff an ifdef. 2008-11-02 17:35:15 +00:00
imp
fbd442bda5 Move mn over. One of the last stragglers in sys/pci. There's no
module built for this hardware, so no changes needed.
2008-11-02 17:04:54 +00:00
trhodes
3c9c77e154 Document a few sysctls in the NFS client and server code.
Minor style(9) where applicable.

Approved by:	alfred (slightly older version)
2008-11-02 17:00:23 +00:00
imp
1bdd099a16 Document dev.rl.%unit.twister_enable sysctl/tunable.
Submitted by:	Kostik Belousov
2008-11-02 16:51:57 +00:00
imp
0beb2d81e4 Make RL_TWISTER_ENABLE a tunable/sysctl. Eliminate it as an option.
Fix module build.

Submitted by:	Kostik Belousov
2008-11-02 16:50:57 +00:00
mav
ba374a2c72 As soon as we have several threads per process now, it is not correct to
use process ID as ACPI thread ID. Concurrent requests with equal thread
IDs broke ACPI mutexes operation causing unpredictable errors including
AE_AML_MUTEX_NOT_ACQUIRED that I have seen.

Use kernel thread ID instead of process ID for ACPI thread.
2008-11-02 12:50:16 +00:00
trasz
495f796829 Remove the call to getinoquota() from ntfs_access. How did it get there?!
Approved by:	rwatson (mentor)
2008-11-02 11:49:19 +00:00
attilio
56b77d9915 Reflect changes for the vfs_busy() prototype. 2008-11-02 10:33:57 +00:00
attilio
2cebb9622a Bump __FreeBSD_version in order to reflect vfs_busy() prototype changing
and its flags (MBF_NOWAIT, MBF_MNTLSTLOCK) introduction.
2008-11-02 10:20:18 +00:00
attilio
e1f493235e Improve VFS locking:
- Implement real draining for vfs consumers by not relying on the
  mnt_lock and using instead a refcount in order to keep track of lock
  requesters.
- Due to the change above, remove the mnt_lock lockmgr because it is now
  useless.
- Due to the change above, vfs_busy() is no more linked to a lockmgr.
  Change so its KPI by removing the interlock argument and defining 2 new
  flags for it: MBF_NOWAIT which basically replaces the LK_NOWAIT of the
  old version (which was unlinked from the lockmgr alredy) and
  MBF_MNTLSTLOCK which provides the ability to drop the mountlist_mtx
  once the mnt interlock is held (ability still desired by most consumers).
- The stub used into vfs_mount_destroy(), that allows to override the
  mnt_ref if running for more than 3 seconds, make it totally useless.
  Remove it as it was thought to work into older versions.
  If a problem of "refcount held never going away" should appear, we will
  need to fix properly instead than trust on such hackish solution.
- Fix a bug where returning (with an error) from dounmount() was still
  leaving the MNTK_MWAIT flag on even if it the waiters were actually
  woken up. Just a place in vfs_mount_destroy() is left because it is
  going to recycle the structure in any case, so it doesn't matter.
- Remove the markercnt refcount as it is useless.

This patch modifies VFS ABI and breaks KPI for vfs_busy() so manpages and
__FreeBSD_version will be modified accordingly.

Discussed with:	kib
Tested by:	pho
2008-11-02 10:15:42 +00:00
imp
92bf4795aa Add support for reading Tivo Series 1 partitioning. This likely needs
a little refinement, but is good enough to commit as is.

# Should look to see if I should move swab(3) into the kernel or just
# provide the unoptimized routine here.

Reviewed by:	marcel@
2008-11-02 03:02:56 +00:00
imp
9534b3469f Merge some minor deltas from p4 newcard tree:
(1) Belkin F5D7050_V4000 was also sold as 'Ativa 802.11g wireless card'
(document)
(2) Add HP Office Jet 4215
2008-11-02 03:00:36 +00:00
imp
1c6d3e0cf3 Remove cardbus attachment. It likely was a cut-n-paste left over from
whatever template was used to create this driver.  It is not
necessary, and wouldn't work anyway since (a) this device will never
be in a cardbus tin-can and (b) the driver isn't even PCI, but instead
a built-in NIC on the IDT RC32434 on its internal bus.
2008-11-02 02:58:24 +00:00
peter
3375b8347b Add missing END() macros, as per rev 184547 for amd64. The lack of these
is mostly harmless, but it does upset some of valgrind's functionality.
2008-11-02 01:28:47 +00:00
peter
aa63008f13 We've been lax about matching END() macros in asm code for some time. This
is used to set the ELF size attribute for functions.  It isn't normally
critical but some things can make use of it (gdb for stack traces).
Valgrind needs it so I'm adding it in.  The problem is present on all
branches and on both i386 and amd64.
2008-11-02 01:10:54 +00:00
keramida
4b957256d0 Various comment nits, and typos. 2008-11-02 00:41:26 +00:00
rwatson
20831b1f86 Update introductory comment for audit pipes.
MFC after:	2 months
Sponsored by:	Apple, Inc.
2008-11-02 00:25:48 +00:00
rwatson
368cc5044a Remove stale comment about filtering in audit pipe ioctl routine: we do
support filtering now, although we may want to make it more interesting
in the future.

MFC after:	2 months
Sponsored by:	Apple, Inc.
2008-11-02 00:18:19 +00:00
rwatson
3f0f3e5028 Add comment for per-pipe stats.
MFC after:	2 months
Sponsored by:	Apple, Inc.
2008-11-01 23:05:49 +00:00
rwatson
64f6525f93 We only allow a partial read of the first record in an audit pipe
record queue, so move the offset field from the per-record
audit_pipe_entry structure to the audit_pipe structure.

Now that we support reading more than one record at a time, add a
new summary field to audit_pipe, ap_qbyteslen, which tracks the
total number of bytes present in a pipe, and return that (minus
the current offset) via FIONREAD and kqueue's data variable for
the pending byte count rather than the number of bytes remaining
in only the first record.

Add a number of asserts to confirm that these counts and offsets
following the expected rules.

MFC after:	2 months
Sponsored by:	Apple, Inc.
2008-11-01 21:56:45 +00:00
rwatson
f8873b326d Allow a single read(2) system call on an audit pipe to retrieve data from
more than one audit record at a time in order to improve efficiency.

MFC after:	2 months
Sponsored by:	Apple, Inc.
2008-11-01 21:16:09 +00:00
kientzle
b93042ab7d Correct the documented declaration of the archive_write_callback to
match the code.

PR:		docs/128089
Submitted by:	Mel
MFC after:	3 days
2008-11-01 19:11:21 +00:00
trasz
d773e920c4 Remove obsolete pseudocode from VOP_ACCESS.9, replacing it with something
closer to reality.

Approved by:	rwatson (mentor)
2008-11-01 19:02:05 +00:00
imp
77c74d7544 Fix a few typos/spelling errors in my comments from the last commit,
plus a few others that had lingered in this driver...

Submitted by:	"b." bf2006a att yahoo KIBO com
2008-11-01 17:02:01 +00:00
brueffer
5d99062f35 - Add one more supported adapter (1)
- Fix a couple of typos

Submitted by:	Horvath Andras (1)
2008-11-01 14:19:55 +00:00
ed
57b4089c20 Clamp the values of t_column to 5 digits in pstat -t' and show all ttys'.
We often run into these very high column numbers when we run curses
applications, because they don't print any newlines. This messes up the
table output of `pstat -t'. If these numbers get really high, they
aren't of any use to the reader anyway. Convert them to `99999' when
they run out of bounds.
2008-11-01 13:40:46 +00:00
ed
c2c324d379 Reimplement the /dev/console device node.
One of the pieces of code that I had left alone during the development
of the MPSAFE TTY layer, was tty_cons.c. This file actually has two
different functions:

- It contains low-level console input/output routines (cnputc(), etc).

- It creates /dev/console and wraps all its cdevsw calls to the
  appropriate TTY.

This commit reimplements the second set of functions by moving it
directly into the TTY layer. /dev/console is now a character device node
that's basically a regular TTY, but does a lookup of `si_drv1' each time
you open it. d_write has also been changed to call log_console().
d_close() is not present, because we must make sure we don't revoke the
TTY after writing a log message to it.

Even though I'm not convinced this is in line with the future directions
of our console code, it is a good move for now. It removes recursive
locking from the top half of the TTY layer. The previous implementation
called into the TTY layer with Giant held.

I'm renaming tty_cons.c to kern_cons.c now. The code hardly contains any
TTY related bits, so we'd better give it a less misleading name.

Tested by:	Andrzej Tobola <ato iem pw edu pl>,
		Carlos A.M. dos Santos <unixmania gmail com>,
		Eygene Ryabinkin <rea-fbsd codelabs ru>
2008-11-01 08:35:28 +00:00
ed
ec4658731c Allow a read() on /dev/ams[0-9] to be interrupted.
Right now ams_read() uses cv_wait() to wait for new data to arrive on
the mouse device. This means that when you run `cat /dev/ams0', it
cannot be interrupted directly. After you press ^C, you first need to
move the mouse before cat will quit. Make this function use
cv_wait_sig(), which allows it to be interrupted directly.

Reviewed by:	nwhitehorn
2008-11-01 08:07:02 +00:00
maxim
782d4d3f4b o OpenBSD 4.4 added. 2008-11-01 06:48:28 +00:00
imp
f91e0afe64 Add RL_TWISTER_ENABLE. 2008-11-01 00:28:44 +00:00
imp
6abe76f096 Add RL_TWISTER_ENABLE option. This enables the magic bits to do long
cable tuning.  This has helped in some installations for hardware
deployed by a former employer.  Made optional because the lists aren't
full of complaints about these cards... even when they were wildly
popular.

Reviewed by:	attilio@, jhb@, trhodes@ (all an older version of the patch)
2008-10-31 23:24:13 +00:00
rwatson
7e60f291b2 In example use of err(3) and errx(3), use sysexits(3) constants.
MFC after:	3 days
Submitted by:	Bruce Cran <bruce at cran dot org dot uk>
2008-10-31 15:14:40 +00:00
rwatson
efc5b661a1 Since there is no longer the opportunity for record truncation, just
return 0 if the truncation counter is queried on an audit pipe.

MFC after:	2 months
Sponsored by:	Apple, Inc.
2008-10-31 15:11:01 +00:00
rwatson
06bee05bb5 In style(9) examples of err() and errx(), use sysexits(3) errors rather
than returning 1.

Submitted by:	Bruce Cran <bruce at cran dot org dot uk>
MFC after:	3 days
2008-10-31 14:47:15 +00:00
rwatson
6f79887fc5 Historically, /dev/auditpipe has allows only whole records to be read via
read(2), which meant that records longer than the buffer passed to read(2)
were dropped.  Instead take the approach of allowing partial reads to be
continued across multiple system calls more in the style of streaming
character device.

This means retaining a record on the per-pipe queue in a partially read
state, so maintain a current offset into the record.  Keep the record on
the queue during a read, so add a new lock, ap_sx, to serialize removal
of records from the queue by either read(2) or ioctl(2) requesting a pipe
flush.  Modify the kqueue handler to return bytes left in the current
record rather than simply the size of the current record.

It is now possible to use praudit, which used the standard FILE * buffer
sizes, to track much larger record sizes from /dev/auditpipe, such as
very long command lines to execve(2).

MFC after:	2 months
Sponsored by:	Apple, Inc.
2008-10-31 14:40:21 +00:00
kib
288874a97d The code in linux_proc_exit() contains a race when multiple linux based
processes exits at the same time.  The linux_emuldata structure is freed
but p->p_emuldata is left as a dangling pointer to the just freed memory.

The check for W_EXIT in the loop scanning the child processes isn't safe
since the state of the child process can change right afterwards. Lock
the process and check the W_EXIT before delivering signal.

Submitted by:	tegge
Reviewed by:	davidxu
MFC after:	1 week
2008-10-31 10:38:30 +00:00
kib
e200f7c9b6 The file was inadvertently excluded from r184499. 2008-10-31 10:14:28 +00:00
kib
37e83ace48 Revert r184136. Instead, push the check for crashdumpmap overflow into the
MD i386 and amd64 dump code.

Requested by:	jhb
Retested by:	pho
MFC after:	3 days (+ 176304 + 184136)
2008-10-31 10:11:35 +00:00
peter
1f7fd22cbb Add three extra to the kinfo_proc_vmmap data. kve_offset - the offset
within an object that a mapping refers to.  fileid and fsid are inode/dev
for vnodes.  (Linux procfs has these and valgrind is really unhappy
without them.)  I believe I didn't change the size of the struct.
2008-10-31 05:43:19 +00:00
nwhitehorn
197f3f98a5 Modify our boot block to pick an output device, without which boot1 will fail
on G4 machines. On the assumption that most people using FreeBSD on Apple
hardware are not using serial consoles, set boot1's output to screen. This
should be revisited. While here, reduce verbosity of boot1.
2008-10-31 00:52:31 +00:00
rwatson
81bbfda754 When we drop an audit record going to and audit pipe because the audit
pipe has overflowed, drop the newest, rather than oldest, record.  This
makes overflow drop behavior consistent with memory allocation failure
leading to drop, avoids touching the consumer end of the queue from a
producer, and lowers the CPU overhead of dropping a record by dropping
before memory allocation and copying.

Obtained from:	Apple, Inc.
MFC after:	2 months
2008-10-30 23:09:19 +00:00