Commit Graph

168031 Commits

Author SHA1 Message Date
Alexander Motin
4f24090354 Realtek CODECs declare support for 32bit samples on S/PDIF input/output
widgets. I am not sure if S/PDIF supports 32bit samples, but my Marantz
SR4001 doesn't, producing only single clicks on playback start/stop.
Because HDA controller requires 32bit alignment for all samples above 16bit,
we can't handle this situation in regular way and have to set 32bit format
in sound(4) for anything above 16bit. To workaround the problem, prefer
to setup hardware to use 24/20bit samples when 32bit format requested. Add
dev.pcm.X.play.32bit and dev.pcm.X.rec.32bit sysctls to control what format
really use for 32bit samples.

MFC after:	2 months
Sponsored by:	iXsystems, Inc.
2012-01-23 17:05:11 +00:00
Gleb Smirnoff
48a47609bc Provide a findhook method for ng_socket(4). The node stores a
hash with names of its hooks. It starts with size of 16, and
grows when number of hooks reaches twice the current size. A
failure to grow (memory is allocated with M_NOWAIT) isn't
fatal, however.

I used standard hash(9) function for the hash. With 25000
hooks named in the mpd (ports/net/mpd5) manner of "b%u", the
distributions is the following: 72.1% entries consist of one
element, 22.1% consist of two, 5.2% consist of three and
0.6% of four.

Speedup in a synthetic test that creates 25000 hooks and then
runs through a long cyclce dereferencing them in a random order
is over 25 times.
2012-01-23 16:43:13 +00:00
Gleb Smirnoff
93a1b4c4cf Convert panic()s to KASSERT()s. This is an optimisation for
hashdestroy() since in absence of INVARIANTS a compiler
will drop the entire for() cycle.
2012-01-23 16:31:46 +00:00
Nathan Whitehorn
3fa1e8a27c Part of r230482 didn't actually work. Revert it for now. This means PR
161047 isn't actually fixed.

PR:		bin/161047
2012-01-23 16:17:54 +00:00
Nathan Whitehorn
a630c48bfa Do a test in a better way. Editing files after testing them is never wise. 2012-01-23 15:50:16 +00:00
Nathan Whitehorn
dc8fd8f2b7 Per popular demand, if installing from a graphics terminal, run the
installer on a VTY with no kernel messages (VTY 2), show the installer
log in real time on VTY 3, and spawn a shell on VTY 4.

PR:		bin/161047, bin/161048
MFC after:	2 weeks
2012-01-23 15:44:52 +00:00
Gleb Smirnoff
4b2b8a370c In ng_socket(4) expose less kernel internals to userland. This commit
breaks ABI, but makes probability of ABI breakage in future less.
2012-01-23 15:39:45 +00:00
Gleb Smirnoff
c4282b741b Convert locks that protect name hash, ID hash and typelist from
mutex(9) to rwlock(9) based locks.

While here remove dropping lock when processing NGM_LISTNODES,
and NGM_LISTTYPES generic commands. We don't need to drop it
since memory allocation is done with M_NOWAIT.
2012-01-23 15:17:14 +00:00
Alexander Leidinger
07c1584e76 Mechanically add a config for all missing drivers.
No cross-referencing was added to the configs, so no
automatic linking to the documentation of other subsystems.

Drivers which already contain doxygen markup:
agp ath bktr bxe cxgb cxgbe dpt drm e1000 iir
ixgbe mwl nxge ofw pccard siba wpi xen
2012-01-23 11:37:40 +00:00
Dag-Erling Smørgrav
e6e3bcd376 Fix two nits in previous commit pointed out by pjd@.
MFC after:	3 weeks
2012-01-23 09:23:07 +00:00
David Schultz
2ee7b1d4ae Add C11 macros describing subnormal numbers to float.h.
Reviewed by:	bde
2012-01-23 06:36:41 +00:00
Mikolaj Golub
c3c314f9f7 Make procstat -l to work with the new version of kern.proc.rlimit.
Submitted by:	Andrey Zonov <andrey at zonov.org>
MFC after:	2 weeks
2012-01-22 20:26:46 +00:00
Mikolaj Golub
8854fe3915 Change kern.proc.rlimit sysctl to:
- retrive only one, specified limit for a process, not the whole
  array, as it was previously (the sysctl has been added recently and
  has not been backported to stable yet, so this change is ok);

- allow to set a resource limit for another process.

Submitted by:	Andrey Zonov <andrey at zonov.org>
Discussed with:	kib
Reviewed by:	kib
MFC after:	2 weeks
2012-01-22 20:25:00 +00:00
Alexander Motin
661268263e Increase snd_hda(4) default maximal buffer size from 16K to 64K and
maximal from 64K to 256K.

We usually don't need 750 sound interrupts per second (1.3ms latency)
when playing 192K/24/8 stream. 187 should be better. On usual 48K/16/2
it is just enough for hw.snd.latency=9 at hw.snd.latency_profile=1 with
23 and 6 interrupts per second.

MFC after:	2 weeks
Sponsored by:	iXsystems, Inc.
2012-01-22 15:44:20 +00:00
Stefan Farfeleder
a6fdf2db3c Document the values for hw.snd.default_auto. 2012-01-22 14:04:42 +00:00
Jilles Tjoelker
6eff4a75d5 sh: Fix $? in the first command of a 'for'.
In the first command of a 'for', $? should be the exit status of the last
pipeline (command substitution in the word list or command before 'for'),
not always 0.
2012-01-22 14:00:33 +00:00
Christian Brueffer
24835875f2 Connect VirtIO-related manpages to the build. 2012-01-22 13:51:20 +00:00
Konstantin Belousov
7ff824218a Clarify the implementation-defined behaviour in case of close(2)
returning error.

MFC after:	1 week
2012-01-22 11:58:17 +00:00
Konstantin Belousov
62f9324b50 Fix typo.
Submitted by:	John Marino <draco marino st>
MFC after:	3 days
2012-01-22 11:35:50 +00:00
Hiroki Sato
9549f74201 - Replace "wtmp" with "utx.log" because last(1) no longer reads wtmp.
- A real filename is now shown in an output report when "-f file" is specified.
- Add Xr lastlogin into last(1) manual page.

Reviewed by:	ed
MFC after:	1 week
2012-01-22 11:34:24 +00:00
Pawel Jakub Dawidek
4ed472a1cc Free memory that won't be used in child.
MFC after:	1 week
2012-01-22 11:20:42 +00:00
Pawel Jakub Dawidek
e0c980d95d The sys/uio.h header is needed only for readv(2), preadv(2), writev(2) and
pwritev(2). Document it more precisely.

Reviewed by:	jilles
MFC after:	3 days
2012-01-22 11:15:48 +00:00
Pawel Jakub Dawidek
9b9a01792d TDF_* flags should be used with td_flags field and TDP_* flags should be used
with td_pflags field. Correct two places where it was not the case.

Discussed with:	kib
MFC after:	1 week
2012-01-22 11:01:36 +00:00
Pawel Jakub Dawidek
241b3b8122 Use provided name when allocating ksid domain. It isn't really used on FreeBSD,
but should fix a panic when pool is imported from another OS that is using this.

MFC after:	1 week
2012-01-22 10:58:17 +00:00
Hiroki Sato
f6e5ba31d2 Fix several glitches in IPv6-related knobs:
- ipv6_enable + ipv6_gateway_enable should unset ACCEPT_RTADV by default for
  backward compatibility.

- Configurations in ipv6_prefix_IF should be recognized even if there is no
  ifconfig_IF_ipv6.

- DAD wait should be performed at once, not on a per-interface basis, if
  possible.  This fixes an issue that a system with a lot of IPv6-capable
  interfaces takes too long for booting.

MFC after:	1 week
2012-01-22 10:57:32 +00:00
Bjoern A. Zeeb
08adfbbf62 Make #error messages string-literals and remove punctuation.
Reported by:	bde (for ip_divert)
Reviewed by:	bde
MFC after:	3 days
2012-01-22 10:41:58 +00:00
Alexander Motin
3d741b14a7 Complete rewrite of the snd_hda(4) volume control.
Previous code was relatively dumb. During CODEC probe it was tracing signals
and statically binding amplifier controls to the OSS mixer controls. To set
volume it just set all bound amplifier controls proportionally to mixer
level, not looking on their hierarchy and amplification levels/offsets.

New code is much smarter. It also traces signals during probe, but mostly
to find out possible amplification control rages in dB for each specific
signal. To set volume it retraces each affected signal again and sets
amplifiers controls recursively to reach desired amplification level in dB.
It would be nice to export values in dB to user, but unluckily our OSS mixer
API is too simple for that.

As result of this change:
 - cascaded amplifiers will work together to reach maximal precision.
If some input has 0/+40dB preamplifier with 10dB step and -10/+10dB mixer
with 1dB step after it, new code will use both to provide 0/+40dB control
with 1dB step! We could even get -10/+50dB range there, but that is
intentionally blocked for now.
 - different channels of multichannel associations on non-uniform CODECs
such as VIA VT1708S will have the same volume, not looking that control
ranges are different. It was not good when fronts were 12dB louder.
 - for multiplexed recording, when we can record from only one source at
a time, we can now use recording amplifier controls to set different
volume levels for different inputs if they have no own controls of they
are less precise. If recording source change, amplifiers will be
reconfigured.

To improve out-of-the-box behavior, ignore default volume levels set by
sound(4) and use own, more reasonable: +20dB for mics, -10dB for analog
output volume and 0dB for the rest of controls. sound(4) defaults of 75%
mean absolutely random things for different controls of different CODECs
because of very different control ranges.
Together with further planned automatic recording source selection this
should allow users to get fine playback and recording without touching
mixer first.
Note that existing users should delete /var/db/mixer*-state and reboot
or trigger CODEC reconfiguration to get new default values.

MFC after:	2 months
Sponsored by:	iXsystems, Inc.
2012-01-22 10:24:12 +00:00
Christian Brueffer
28083eaa4d General cleanup.
Prodded by:	grehan
2012-01-22 10:16:24 +00:00
Martin Matuska
3742a70580 Merge illumos revisions 13540, 13562:
illumos rev 13540 [1]:
Removal of pyzfs broke delegation for volumes

illumos rev 13562 [2]:
zfs allow arguments not parsed correctly after pyzfs removal

References:
https://www.illumos.org/issues/1726 [1]
https://www.illumos.org/issues/1977 [2]

Obtained from:	illumos (issues #1726, #1977)
MFC after:	1 week
2012-01-22 08:06:36 +00:00
Adrian Chadd
7b2b15eb98 Mark the taskqueue as the _net80211_ taskqueue.
This makes it much easier to determine whether an event occurs in the
net80211 taskqueue (which was called "ath0 taskq") or the ath driver
taskqueue (which is also called "ath0 taskq".)
2012-01-22 05:30:29 +00:00
Adrian Chadd
11e66657d7 Bring over the visibility control tool for Monthadar's wtap project.
This allows basic control over which wtap nodes can see which other
wtap nodes.
2012-01-22 04:51:00 +00:00
Bjoern A. Zeeb
aa57e971bc Fix ip_divert handling of inet and inet6 and module building some more.
Properly sort the "carp" case in modules/Makefile after it was renamed.

Reported by:	bde (most)
Reviewed by:	bde
MFC after:	3 days
2012-01-22 02:16:31 +00:00
Bjoern A. Zeeb
83e521ec73 Clean up some #endif comments removing from short sections. Add #endif
comments to longer, also refining strange ones.

Properly use #ifdef rather than #if defined() where possible.  Four
#if defined(PCBGROUP) occurances (netinet and netinet6) were ignored to
avoid conflicts with eventually upcoming changes for RSS.

Reported by:	bde (most)
Reviewed by:	bde
MFC after:	3 days
2012-01-22 02:13:19 +00:00
Konstantin Belousov
c2b396f294 Remove the nc_time and nc_ticks elements from struct namecache, and
provide struct namecache_ts which is the old struct namecache. Only
allocate struct namecache_ts if non-null struct timespec *tsp was
passed to cache_enter_time, otherwise use struct namecache.

Change struct namecache allocation and deallocation macros into static
functions, since logic becomes somewhat twisty.  Provide accessor for
the nc_name member of struct namecache to hide difference between
struct namecache and namecache_ts.

The aim of the change is to not waste 20 bytes per small namecache
entry.

Reviewed by:	 jhb
MFC after: 2 weeks
X-MFC-note:  after r230394
2012-01-22 01:11:06 +00:00
Pawel Jakub Dawidek
1698a6aec9 Dramatically optimize listing snapshots when user requests only snapshot
names and wants to sort them by name, ie. when executes:

	# zfs list -t snapshot -o name -s name

Because only name is needed we don't have to read all snapshot properties.

Below you can find how long does it take to list 34509 snapshots from a single
disk pool before and after this change with cold and warm cache:

before:

	# time zfs list -t snapshot -o name -s name > /dev/null
	cold cache: 525s
	warm cache: 218s

after:

	# time zfs list -t snapshot -o name -s name > /dev/null
	cold cache: 1.7s
	warm cache: 1.1s

MFC after:	1 week
2012-01-21 21:12:53 +00:00
Jilles Tjoelker
0105d7224d sh: Remove "kill" example function, which is superseded by the kill builtin
MFC after:	1 week
2012-01-21 20:39:17 +00:00
Pawel Jakub Dawidek
2ce9c023fd Fix minor memory leak.
MFC after:	3 days
2012-01-21 20:13:37 +00:00
Konstantin Belousov
10280ca601 Use getcontextx(3) internal API instead of getcontext(2) to provide
the signal handlers with the context information in the deferrred
case.

Only enable the use of getcontextx(3) in the deferred signal delivery
code on amd64 and i386. Sparc64 seems to have some undetermined issues
with interaction of alloca(3) and signal delivery.

Tested by:	flo (who also provided sparc64 harware access for me), pho
Discussed with:	marius
MFC after:	1 month
2012-01-21 18:06:18 +00:00
Konstantin Belousov
2b1de0afd1 Add API for obtaining extended machine context states that cannot be
fit into existing mcontext_t.

On i386 and amd64 do return the extended FPU states using
getcontextx(3). For other architectures, getcontextx(3) returns the
same information as getcontext(2).

Tested by:  pho
MFC after:  1 month
2012-01-21 18:00:28 +00:00
Konstantin Belousov
f1c8f19055 Adopt to new layout of struct pcb.
MFC after:	1 month
2012-01-21 17:50:14 +00:00
Konstantin Belousov
8c6f8f3d5b Add support for the extended FPU states on amd64, both for native
64bit and 32bit ABIs.  As a side-effect, it enables AVX on capable
CPUs.

In particular:

- Query the CPU support for XSAVE, list of the supported extensions
  and the required size of FPU save area. The hw.use_xsave tunable is
  provided for disabling XSAVE, and hw.xsave_mask may be used to
  select the enabled extensions.

- Remove the FPU save area from PCB and dynamically allocate the
  (run-time sized) user save area on the top of the kernel stack,
  right above the PCB. Reorganize the thread0 PCB initialization to
  postpone it after BSP is queried for save area size.

- The dumppcb, stoppcbs and susppcbs now do not carry the FPU state as
  well. FPU state is only useful for suspend, where it is saved in
  dynamically allocated suspfpusave area.

- Use XSAVE and XRSTOR to save/restore FPU state, if supported and
  enabled.

- Define new mcontext_t flag _MC_HASFPXSTATE, indicating that
  mcontext_t has a valid pointer to out-of-struct extended FPU
  state. Signal handlers are supplied with stack-allocated fpu
  state. The sigreturn(2) and setcontext(2) syscall honour the flag,
  allowing the signal handlers to inspect and manipilate extended
  state in the interrupted context.

- The getcontext(2) never returns extended state, since there is no
  place in the fixed-sized mcontext_t to place variable-sized save
  area. And, since mcontext_t is embedded into ucontext_t, makes it
  impossible to fix in a reasonable way.  Instead of extending
  getcontext(2) syscall, provide a sysarch(2) facility to query
  extended FPU state.

- Add ptrace(2) support for getting and setting extended state; while
  there, implement missed PT_I386_{GET,SET}XMMREGS for 32bit binaries.

- Change fpu_kern KPI to not expose struct fpu_kern_ctx layout to
  consumers, making it opaque. Internally, struct fpu_kern_ctx now
  contains a space for the extended state. Convert in-kernel consumers
  of fpu_kern KPI both on i386 and amd64.

First version of the support for AVX was submitted by Tim Bird
<tim.bird am sony com> on behalf of Sony. This version was written
from scratch.

Tested by:	pho (previous version), Yamagi Burmeister <lists yamagi org>
MFC after:	1 month
2012-01-21 17:45:27 +00:00
Hans Petter Selasky
dd03e19c08 Add support for the DesignWare USB 2.0 OTG controller chipset.
Currently the code is not built by any modules. That will
be fixed later. The Atmel ARM bus interface file part of this
commit is just for sake of example. All registers and bits are
declared like macros and not C-structures like in official
Synopsis header files. This driver mostly origins from the
musb_otg.c driver in FreeBSD except that the chip specific
programming has been replaced by the one for DWC 2.0 USB OTG.
Some parts related to system suspend and resume have been left
like empty functions for the future. USB suspend and resume is
fully supported.
2012-01-21 13:31:38 +00:00
Edward Tomasz Napierala
89ef44efa3 Replace the beerware license on mdconfig(8) with standard 2-clause BSD.
Approved by:	phk@
2012-01-21 11:48:48 +00:00
Andreas Tobler
7abf4584c6 Revert r230390. 2012-01-21 11:42:40 +00:00
Adrian Chadd
e148ff4992 Fix the hwmp code to handle nodes in a "line" topology.
For example, this particular topology didn't work correctly from all
nodes:

[A] - [B] - [C] - [D]

Submitted by:	Monthadar Al Jaberi <monthadar@gmail.com>
Reviewed by:	bschmidt, adrian
2012-01-21 00:42:28 +00:00
Adrian Chadd
f7c90cbb77 Change the hwmp debugging to use %6D rather than ether_sprintf().
This allows for multiple MAC addresses to be printed on the same
debugging line.  ether_sprintf() uses a static char buffer and
thus isn't very useful here.

Submitted by:	Monthadar Al Jaberi <monthadar@gmail.com>
2012-01-21 00:38:18 +00:00
Martin Matuska
6dfe0a3dc2 Use separate buffer for global path to avoid overflow of path buffer.
Reviewed by:	jamie@
MFC after:	3 weeks
2012-01-21 00:06:21 +00:00
Oleksandr Tymoshenko
cccc098d9b We use port_index field of struct octusb_qh to reference USB state
of root HUB. Although it is initialized with port index of the
device's parent hub, which is worng. So track the USB tree up to
root HUB  and initialize this filed ptroprly

Rename port_index to root_port_index in order to reflect its
real semantics.
2012-01-20 23:37:04 +00:00
Martin Matuska
08acd455f3 Add one more copyright line accidentially removed in r228103
MFC after:	3 days
2012-01-20 22:56:57 +00:00
Brooks Davis
2dd3bcd48d When creating the jails /dev/log symlink, do it by full path to avoid
creating stray "log" symlinks if the mount fails.  That apparently
happens in some ezjail configs.

PR:		conf/143084
Submitted by:	Dirk Engling <erdgeist at erdgeist.org>
Reviewed by:	simon
MFC after:	2 weeks
2012-01-20 22:55:19 +00:00