dragonfly.git
5 hours ago<sys/cdefs.h> Also add __malloclike to lint case. master
zrj [Thu, 29 Sep 2016 05:27:24 +0000 (08:27 +0300)]
<sys/cdefs.h> Also add __malloclike to lint case.

Fixes: $ echo "#include <stdlib.h>" | gcc -Wall -Wextra -Dlint -c -xc -

38 hours agokernel - Remove mplock from KTRACE paths
Matthew Dillon [Tue, 27 Sep 2016 21:39:03 +0000 (14:39 -0700)]
kernel - Remove mplock from KTRACE paths

* The mplock is no longer needed for KTRACE, ktrace writes are serialized
  by the vnode lock and everything else is MPSAFE.  Note that this change
  means that even fast system calls may interleave in the ktrace output on
  a multi-threaded program.

* Fix ktrace bug related to vkernels.  The syscall2() code assumes that
  no tokens are held on entry (since we are coming from usermode), but
  a system call made from the vkernel may actually be nested inside
  another syscall2().  The mplock KTRACE held caused this to assert in
  the nested syscall2().  The removal of the mplock from the ktrace path
  also fixes this bug.

* Minor comment adjustment in vm_vmspace.c.

Reported-by: tuxillo
2 days agocpuctl(4): Require write rights for CPUCTL_MSRSBIT and CPUCTL_MSRCBIT.
Imre Vadász [Mon, 26 Sep 2016 23:44:43 +0000 (01:44 +0200)]
cpuctl(4): Require write rights for CPUCTL_MSRSBIT and CPUCTL_MSRCBIT.

* Both CPUCTL_MSRSBIT and CPUCTL_MSRCBIT write MSR registers, so they
  should require write rights like CPUCTL_WRMSR.

2 days agokernel/vmx - Add a missing lwkt_reltoken()
Antonio Huete Jimenez [Mon, 26 Sep 2016 22:45:50 +0000 (00:45 +0200)]
kernel/vmx - Add a missing lwkt_reltoken()

2 days agoif_iwm - Get rid of some more superfluous bus_dmamap_sync calls in rx path.
Imre Vadász [Mon, 26 Sep 2016 20:54:02 +0000 (22:54 +0200)]
if_iwm - Get rid of some more superfluous bus_dmamap_sync calls in rx path.

2 days agokernel - Change IPI recovery watchdog
Matthew Dillon [Mon, 26 Sep 2016 18:09:49 +0000 (11:09 -0700)]
kernel - Change IPI recovery watchdog

* Change the default recovery watchdog timeout for IPIs from 1/16 second
  to 2 seconds and the repeated timeout from 1/2 second to 2 seconds.

* Add missing initialization to pmap_inval_smp_cmpset(), without it an
  improper watchdog timeout/retry could occur.

* This still may not fix issues with VMs when core threads cause extreme
  latencies on the host.

Reported-by: zach
3 days agolibc_r: Tiny adjustment, git add.
zrj [Mon, 26 Sep 2016 10:34:38 +0000 (13:34 +0300)]
libc_r: Tiny adjustment, git add.

Got lost for fdc2e97852d952eedc66a271c9ae1c12f246934b

3 days agolibpthread: Some more adjustments for pthread_attr_setschedpolicy().
Sascha Wildner [Mon, 26 Sep 2016 09:46:01 +0000 (11:46 +0200)]
libpthread: Some more adjustments for pthread_attr_setschedpolicy().

* Adjust libc_r too.

* Remove unneeded braces.

* Adjust manual page.

3 days agoif_iwm - Add support for Firmware paging, needed for newer 8000C firmware.
Imre Vadász [Sun, 25 Sep 2016 00:32:05 +0000 (02:32 +0200)]
if_iwm - Add support for Firmware paging, needed for newer 8000C firmware.

* Uses the IWM_FW_PAGING_BLOCK_CMD firmware command to tell the firmware
  what memory ranges to use for paging.

Taken-From: Linux iwlwifi

3 days agoif_iwm - Move iwm_dma_contig_alloc/_free functions to if_iwm_util.c.
Imre Vadász [Sun, 25 Sep 2016 00:25:37 +0000 (02:25 +0200)]
if_iwm - Move iwm_dma_contig_alloc/_free functions to if_iwm_util.c.

3 days agopthread: Split pthread_exit() to smaller chunks.
zrj [Mon, 19 Sep 2016 10:20:47 +0000 (13:20 +0300)]
pthread: Split pthread_exit() to smaller chunks.

To see what is needed for thread unwind support.

3 days agopthread: Split mutex_init().
zrj [Mon, 19 Sep 2016 09:33:40 +0000 (12:33 +0300)]
pthread: Split mutex_init().

Makes it easier to see what is needed for PTHREAD_PROCESS_SHARED
support, that it is plenty..

3 days agopthread: Prefer explicit return on error.
zrj [Mon, 19 Sep 2016 08:41:17 +0000 (11:41 +0300)]
pthread: Prefer explicit return on error.

Easier to follow the logic.

3 days agopthread: Set default contention scope to system.
zrj [Mon, 19 Sep 2016 06:13:28 +0000 (09:13 +0300)]
pthread: Set default contention scope to system.

It is a bit unclear how scope was handled previously.
Make it explicit to see what falls down.

3 days agopthread: Use spin hint for cpu on x86_64 too.
zrj [Mon, 19 Sep 2016 05:45:39 +0000 (08:45 +0300)]
pthread: Use spin hint for cpu on x86_64 too.

Looks like missed case in initial port to x864_64.

Pointed-by: swildner
3 days agopthread: Move few attr private structs to thr_init.c
zrj [Mon, 19 Sep 2016 05:14:59 +0000 (08:14 +0300)]
pthread: Move few attr private structs to thr_init.c

These symbols should only be available if _thread_init() gets linked in.
Helps with -static troubleshooting.

3 days agopthread: Add vprintf variant of _thread_printf, reformat PANIC()
zrj [Fri, 16 Sep 2016 13:47:17 +0000 (16:47 +0300)]
pthread: Add vprintf variant of _thread_printf, reformat PANIC()

Cleans up intermediate writes.

Taken-from: FreeBSD

3 days agopthread: Fix some warnings, do not raise WARNS to 6.
zrj [Fri, 16 Sep 2016 13:09:34 +0000 (16:09 +0300)]
pthread: Fix some warnings, do not raise WARNS to 6.

Using -Os will create lots of -Winline, thus leave at 5 by default.
Sorry for __is_threaded.

3 days agopthread: Collect separated mutex functions.
zrj [Fri, 16 Sep 2016 11:49:05 +0000 (14:49 +0300)]
pthread: Collect separated mutex functions.

Split and integrate them into thr_mutex.c and thr_mutexattr.c
just as it is currently in FreeBSD to reduce diffs.

3 days agopthread: Change single weak reference to strong one.
zrj [Fri, 16 Sep 2016 10:25:28 +0000 (13:25 +0300)]
pthread: Change single weak reference to strong one.

Mainly to match libc_r and have only strong references in libthread_xu.

3 days agopthread: Move out pthread_md.c from x86_64 dir to common code.
zrj [Fri, 16 Sep 2016 09:58:32 +0000 (12:58 +0300)]
pthread: Move out pthread_md.c from x86_64 dir to common code.

This leaves only assembly code in arch specific dir.
Similary as it was done on FreeBSD.

3 days agopthread: Fix return value for pthread_attr_setschedpolicy()
zrj [Fri, 16 Sep 2016 09:26:04 +0000 (12:26 +0300)]
pthread: Fix return value for pthread_attr_setschedpolicy()

There are to possible interpretations of ENOTSUP in POSIX specs.
After some research and trial and error testing for now stick to:
on unsupported [known] attribute ENOTSUP should be returned.
Check sched parameters explictly.

While there, rename too generic local variable 'local' and
move up the <sys/syscall.h> include.

3 days agopthread: General pre-cleanup (style, typos etc)
zrj [Fri, 16 Sep 2016 09:18:50 +0000 (12:18 +0300)]
pthread: General pre-cleanup (style, typos etc)

No functional change.

3 days agopthread: Sync copyright changes with FreeBSD.
zrj [Fri, 16 Sep 2016 09:12:34 +0000 (12:12 +0300)]
pthread: Sync copyright changes with FreeBSD.

Taken-from: FreeBSD

3 days agowpa_supplicant: Avoid cpp warning during buildworld.
zrj [Sun, 25 Sep 2016 16:02:15 +0000 (19:02 +0300)]
wpa_supplicant: Avoid cpp warning during buildworld.

Used MIN() macro is essentially the same to one in sys/param.h but
for the sake of argument it is better to avoid warning by simply
using system header variant.

Note: there is issue with WPA_OUI_TYPE definition so leave WANRS at 0.

3 days agoOpenSSH: Add missing prototypes and bump WARNS to 3.
zrj [Sun, 25 Sep 2016 15:23:42 +0000 (18:23 +0300)]
OpenSSH: Add missing prototypes and bump WARNS to 3.

Just include sshlogin.h header and make two functions static (used locally).

3 days agoOpenSSH: Suppress last warning in session.c
zrj [Sun, 25 Sep 2016 15:10:27 +0000 (18:10 +0300)]
OpenSSH: Suppress last warning in session.c

Use __DECONST() trick, that is safe here because login_getcapstr() either
returns copy on success or error string that in this case is passed as
def_nl and checked if return pointer is != def_nl.

3 days agoOpenSSH: Suppress warning on login_getcapstr().
zrj [Sun, 25 Sep 2016 15:04:35 +0000 (18:04 +0300)]
OpenSSH: Suppress warning on login_getcapstr().

Taken-from: FreeBSD (r255767 hidden in merge commit)

3 days agogcc47: Suppress last codegen warning.
zrj [Sun, 25 Sep 2016 14:19:48 +0000 (17:19 +0300)]
gcc47: Suppress last codegen warning.

Turns out gcc(compilation as C) differentiates when initializing global variables that
are declared 'extern' on the same line (removal of extern attribute is ok, includes
c-common.h header that properly marks c_family_lang_mask global variable as extern).

Note: do not apply same change to gcc50 since it is compiled as C++ that is not affected.

3 days agogcc47: Fix misplaced NULL usage in strings.
zrj [Sun, 25 Sep 2016 14:17:32 +0000 (17:17 +0300)]
gcc47: Fix misplaced NULL usage in strings.

Specific to c++, just do not do it.

3 days agogcc47: Backport objc fix from gcc trunk.
zrj [Sun, 25 Sep 2016 13:37:22 +0000 (16:37 +0300)]
gcc47: Backport objc fix from gcc trunk.

This prevents spurious warnings when compiling with gcc50.
Fix is needed for objc ABI inline semantics.
Also why we even provide objc language support in base?

Taken-from: gcc trunk (trunk@220175)

3 days agogcc47: Prevent -Wswitch-bool noise.
zrj [Sun, 25 Sep 2016 13:00:29 +0000 (16:00 +0300)]
gcc47: Prevent -Wswitch-bool noise.

In gcc v5 and later -Wswitch-bool is default and creates a lot of
warnings when building gcc v4.7 with gcc v5 compiler.
Since backporting patches to generate "compatible" gcc private headers
would be very error-prune, just take lkml.org/2015/5/27/941 approach.
At this point we only need to catch real issues for backup compiler.

3 days agocrunchgen(1): Fix few spurious warnings.
zrj [Sun, 25 Sep 2016 12:48:16 +0000 (15:48 +0300)]
crunchgen(1): Fix few spurious warnings.

Just add prototypes to help compiler detecting real bugs and not
pollute build logs with false positives.

Note: made crunch_help() static, seems to be fine.

3 days agolibncurses: Set WANRS at 5.
zrj [Sun, 25 Sep 2016 11:13:23 +0000 (14:13 +0300)]
libncurses: Set WANRS at 5.

Turns out private libncurses was completely ignoring WARNS and running loose,
so after previous few changes connect to global lib/Makefile.inc and set
WARNS at 5 (6 is an overkill, also does not survive FORMAT_AUDIT).

3 days agolibncurses: Avoid warning when calling cgetent().
zrj [Sun, 25 Sep 2016 11:08:29 +0000 (14:08 +0300)]
libncurses: Avoid warning when calling cgetent().

Since both read_termcap.c static version of cgetent() and cgetent(3) has second
argument as char** just use __DECONST() trick so to not have big changes.
It should be safe to do that, base libncurses is a private library now.

3 days agolibncurses: Avoid -Woverflow in baud rates.
zrj [Sun, 25 Sep 2016 10:40:48 +0000 (13:40 +0300)]
libncurses: Avoid -Woverflow in baud rates.

NCURSES_OSPEED has type - short so anything >= B38400 overflows.
Since changing NCURSES_OSPEED to use int is an ABI change and given
that overflown values are still unique and only used as identifiers
there is nothing wrong by explictly casting it to used type.

3 days agobootstrap: Expand BOOTSTRAPPING case to avoid warning.
zrj [Sun, 25 Sep 2016 09:18:28 +0000 (12:18 +0300)]
bootstrap: Expand BOOTSTRAPPING case to avoid warning.

Addition to 6d3bd8e3942560036cf2aec0d98a65819c5f3b4c that allows to prevent
"warning: 'encode_b64' defined but not used [-Wunused-function]"

3 days agoath_hal: Add FCC6_FCCA regulatory domain (0x0014)
John Marino [Mon, 26 Sep 2016 00:35:57 +0000 (19:35 -0500)]
ath_hal: Add FCC6_FCCA regulatory domain (0x0014)

Tested on TP-Link N900, AR9380, regdomain 0x0014 (FCC6_FCCA)

Taken From FreeBSD: SVN rev 306323 (25 SEP 2016)

Thanks Adrian!

3 days agokernel/psm: Support newer Synaptics touchpads
Daniel Bilik [Mon, 26 Sep 2016 00:47:17 +0000 (02:47 +0200)]
kernel/psm: Support newer Synaptics touchpads

Taken-from: FreeBSD

3 days agokernel/clock_gettime: Various fixes.
Sascha Wildner [Sun, 25 Sep 2016 22:36:03 +0000 (00:36 +0200)]
kernel/clock_gettime: Various fixes.

* Fix CLOCK_PROF and CLOCK_VIRTUAL to behave like FreeBSD's. They are
  not meant to return the values of ITIMER_PROF and ITIMER_VIRTUAL
  (which are decreasing, see getitimer(2)), like they were implemented
  by 91810a6f0686477493e9915e98cfc5adcbe91363. This also fixes
  CLOCK_PROCESS_CPUTIME_ID.

* Fix CLOCK_PROCESS_THREAD_ID. It was adding the values wrongly in a
  way that could cause tv_nsec to overflow (i.e. become >= 1000000000).

* Fix clock_getres() for CLOCK_{PROCESS,THREAD}_CPUTIME_ID.

* Mention CLOCK_{PROCESS,THREAD}_CPUTIME_ID in clock_gettime()'s manual
  page.

* Bring in some minor manual page fixes from FreeBSD.

Reported-by: zhtw
Dragonfly-bug: <https://bugs.dragonflybsd.org/issues/2951>

3 days agoif_iwm - Switch 7265D and 3165 devices to 7265D firmware.
Imre Vadász [Tue, 20 Sep 2016 22:33:02 +0000 (00:33 +0200)]
if_iwm - Switch 7265D and 3165 devices to 7265D firmware.

3 days agoiwmfw - Import 7265D firmware for AC3165 and newer AC7265 revisions.
Imre Vadász [Tue, 6 Sep 2016 21:20:39 +0000 (23:20 +0200)]
iwmfw - Import 7265D firmware for AC3165 and newer AC7265 revisions.

3 days agoif_iwm - Move struct iwm_cfg chipset configs to if_iwm_7000.c and _8000.c
Imre Vadász [Sun, 25 Sep 2016 19:21:20 +0000 (21:21 +0200)]
if_iwm - Move struct iwm_cfg chipset configs to if_iwm_7000.c and _8000.c

* This is more similar to how code/definitions are distributed in
  Linux's iwlwifi.

* This should make recognizing new chipset variants, and adding additional
  flags from the Linux iwlwifi code easier, without blowing up if_iwm.c

Taken-From: Linux iwlwifi

3 days agoif_iwm - Simplify device id matching code a bit.
Imre Vadász [Sun, 25 Sep 2016 18:59:55 +0000 (20:59 +0200)]
if_iwm - Simplify device id matching code a bit.

* Just add the struct iwm_cfg pointers to the iwm_devices array, to get
  rid of the large switch clause.

3 days agokernel/vmm - Fix build with VMM_DEBUG
Antonio Huete Jimenez [Sun, 25 Sep 2016 10:57:09 +0000 (12:57 +0200)]
kernel/vmm - Fix build with VMM_DEBUG

5 days agopthread: Init libthread_xu also in pthread_key_create()
zrj [Fri, 16 Sep 2016 04:42:37 +0000 (07:42 +0300)]
pthread: Init libthread_xu also in pthread_key_create()

Previously libtreahd_xu was only prepared in pthread_create()
but there are cases where keys need to be initialized before.

Solves runtime of statically linked gfortran programs with -pthread
(libgfortran calls newlocale() when setting up the environment).

5 days ago<pthread.h>: Avoid namespace pollution.
zrj [Thu, 22 Sep 2016 11:32:51 +0000 (14:32 +0300)]
<pthread.h>: Avoid namespace pollution.

Rename MUTEX_TYPE_MAX -> PTHREAD_MUTEX_TYPE_MAX to avoid clashes
with types in user source codes. Adjust all use cases.

While there, remove deprecated unused compat mutex types.

5 days ago<pthread.h>: Mark some __nonnull(x) cases.
zrj [Thu, 22 Sep 2016 11:25:10 +0000 (14:25 +0300)]
<pthread.h>: Mark some __nonnull(x) cases.

5 days ago<pthread.h>: Visibility changes.
zrj [Thu, 22 Sep 2016 07:04:32 +0000 (10:04 +0300)]
<pthread.h>: Visibility changes.

While there, remove prototypes of pthread_attr_{get,set}floatstate()
looks like they never were implemented, so do not advertise them.

5 days ago<pthread.h>: Perform some style(9) cleanup.
zrj [Thu, 22 Sep 2016 06:57:39 +0000 (09:57 +0300)]
<pthread.h>: Perform some style(9) cleanup.

Change indentation to a single tab, since both protype names and arguments
are quite long and later added attribute decorations will only make it worse.

Strictly no functional change, just shifting stuff around for readability and
upcoming changes.

5 days agoUse recently added attributes in few standard headers.
zrj [Thu, 22 Sep 2016 06:19:10 +0000 (09:19 +0300)]
Use recently added attributes in few standard headers.

There attributes are to assits compiler making better analysis
doring compilation time. Like branch prediction and hinting.

While there, for now prefer to use __attribute__((__noreturn__)) to
_Noreturn, since neither OpenBSD or NetBSD has it at all and in our
variant we don't handle it in c++ case ([[noreturn]]) anyway.
Looks to be only a workaround for some old clang++ version.

5 days ago<stdlib.h>: Change visibility of few POSIX extensions to ISO C.
zrj [Thu, 22 Sep 2016 05:44:57 +0000 (08:44 +0300)]
<stdlib.h>: Change visibility of few POSIX extensions to ISO C.

Seems it should be like this.

5 days agostyle(9): Perform some cleanup in few standard headers.
zrj [Thu, 22 Sep 2016 05:32:10 +0000 (08:32 +0300)]
style(9): Perform some cleanup in few standard headers.

For upcomming changes. No functional change.

Kindly-requested-by: swildner@
5 days ago<sys/cdefs.h>: Add support for alloc_align, alloc_size and malloc attributes.
zrj [Tue, 20 Sep 2016 07:27:39 +0000 (10:27 +0300)]
<sys/cdefs.h>: Add support for alloc_align, alloc_size and malloc attributes.

  * __alloc_align tells the compiler that one of the arguments is a
    minimal alignment for pointer to help with pointers analysis.
  * __alloc_size tells the compiler that one of the arguments is a
    size of a pointer returned by function.
  * __malloclike tells the compiler that returned pointer does not
    alias any other pointer to help futher optimizations.

While there, add partial support for clang (as it is in FreeBSD).

5 days ago<sys/cdefs.h>: Be more careful with _Noreturn.
zrj [Tue, 20 Sep 2016 08:48:55 +0000 (11:48 +0300)]
<sys/cdefs.h>: Be more careful with _Noreturn.

With modern gcc started using g++ for bootstraps _Noreturn must
be handled with a bit of extra care.

5 days ago<sys/cdefs.h>: Move __warn_unused_result__ to 3.4
zrj [Tue, 20 Sep 2016 07:14:21 +0000 (10:14 +0300)]
<sys/cdefs.h>: Move __warn_unused_result__ to 3.4

Looks like it is reliably available only at 3.4 so change it.

5 days ago<sys/cdefs.h>: Style(9) fixes.
zrj [Tue, 20 Sep 2016 06:43:53 +0000 (09:43 +0300)]
<sys/cdefs.h>: Style(9) fixes.

Mostly tabs, no functional change except for removal
of extra semicolon in __strong_reference() macro.

5 days agobootstrap: Fix old boostrapping issue once and for all.
zrj [Wed, 21 Sep 2016 10:45:50 +0000 (13:45 +0300)]
bootstrap: Fix old boostrapping issue once and for all.

This was fallout after iconv() removal of 'const' attribute to conform
with POSIX. All the later hacks are starting to accumulate and prevent
easy further base system upgrading. So given that main issue is c++
specific and only needed for bootstraping process deal with it straight
and remove all hacks(include shoehornings) that were caused by it.

To fix bootstrapping problems add WORLD_VERSION variable that indicates
current DragonFly base version and could be used to retrospectively
help bootstrapping from older releases. Makefile changes that depend on
WORLD_VERSION should *only* be used as a once switch and during native
compilations should *not* change default world makefiles logic.

5 days agoif_iwm - Remove OpenBSD-specific comment. Beautify pci cfg space accesses.
Imre Vadász [Sat, 17 Sep 2016 21:06:42 +0000 (23:06 +0200)]
if_iwm - Remove OpenBSD-specific comment. Beautify pci cfg space accesses.

5 days agoif_iwm - Fix typo in commit a7697ea01c11fd493aec52260a02f31df680eb91.
Imre Vadász [Fri, 23 Sep 2016 23:35:04 +0000 (01:35 +0200)]
if_iwm - Fix typo in commit a7697ea01c11fd493aec52260a02f31df680eb91.

* The sc->sc_uc.uc_error_event_table value is now at sc->error_event_table,
  and not sc->umac_error_event_table.

5 days agoif_iwm - Get rid of unneeded IWM_FWDMASEGSZ and IWM_FWMAXSIZE constants
Imre Vadász [Fri, 23 Sep 2016 22:27:49 +0000 (00:27 +0200)]
if_iwm - Get rid of unneeded IWM_FWDMASEGSZ and IWM_FWMAXSIZE constants

* We only need IWM_FH_MEM_TB_MAX_LENGTH bytes for the fw_dma memory.

5 days agoif_iwm - With the previous commits, this DELAY no longer seems to be needed.
Imre Vadász [Fri, 23 Sep 2016 22:31:26 +0000 (00:31 +0200)]
if_iwm - With the previous commits, this DELAY no longer seems to be needed.

5 days agoif_iwm - Sync IWM_MVM_ALIVE waiting and start_fw handling with iwlwifi.
Imre Vadász [Fri, 23 Sep 2016 17:09:13 +0000 (19:09 +0200)]
if_iwm - Sync IWM_MVM_ALIVE waiting and start_fw handling with iwlwifi.

* Use the notification wait API, like it's done in the Linux iwlwifi code,
  to wait for the IWM_MVM_ALIVE notification.

* This also should fix some firmware load interrupt issues, and errors
  in the nic lock using.

Taken-From: Linux iwlwifi

5 days agoif_iwm - Synchronize firmware loading code with Linux iwlwifi.
Imre Vadász [Fri, 9 Sep 2016 22:00:07 +0000 (00:00 +0200)]
if_iwm - Synchronize firmware loading code with Linux iwlwifi.

* While there, rename some functions to match the names and functionality
  of the similarly named functions in Linux iwlwifi.

Taken-From: Linux iwlwifi

6 days agoiwmfw - Fixup revert to version 16 for the 8000C firmware.
Imre Vadász [Thu, 22 Sep 2016 13:24:01 +0000 (15:24 +0200)]
iwmfw - Fixup revert to version 16 for the 8000C firmware.

7 days agoif_iwm - Improve reliability of iwm_release on disassociation a bit.
Imre Vadász [Sat, 17 Sep 2016 20:57:06 +0000 (22:57 +0200)]
if_iwm - Improve reliability of iwm_release on disassociation a bit.

* We need to first call ivp->iv_newstate(), to enqueue the deauth/deassoc
  mgmt frame, then flush the tx queue, before actually calling
  iwm_release().

* cycling a wlan connection via wpa_cli frontend to wpa_supplicant, by
  issuing disconnect and reconnect commandos works quite well.
  (There is still an issue when disconnecting to quickly after completing
   authentication/association).

* Reassociating or roaming via wpa_supplicant is still broken.

7 days agoif_iwm - Remove 1s delay after fw loading. Can't reproduce issues on AC8260.
Imre Vadász [Thu, 22 Sep 2016 09:03:26 +0000 (11:03 +0200)]
if_iwm - Remove 1s delay after fw loading. Can't reproduce issues on AC8260.

The 1s delay was added in the update to version 16 fw, where Family 8000
support was added.

7 days agoif_iwm - SCAN_ABORT_UMAC response doesn't use a wide id
Imre Vadász [Thu, 22 Sep 2016 09:01:51 +0000 (11:01 +0200)]
if_iwm - SCAN_ABORT_UMAC response doesn't use a wide id

7 days agoif_iwm - Revert 8000C firmware (for AC8260) to version 16 for now.
Imre Vadász [Thu, 22 Sep 2016 09:04:39 +0000 (11:04 +0200)]
if_iwm - Revert 8000C firmware (for AC8260) to version 16 for now.

8 days agovkernel - Invalidate pte before setting attributes to the vm_page
Antonio Huete Jimenez [Tue, 20 Sep 2016 23:31:58 +0000 (01:31 +0200)]
vkernel - Invalidate pte before setting attributes to the vm_page

- Fixes a problem at mountroot time where it doesn't find any disk
  even though the disk is detected earlier.

8 days agovkernel - Fix a vkernel lockup on startup
Antonio Huete Jimenez [Tue, 20 Sep 2016 22:03:05 +0000 (00:03 +0200)]
vkernel - Fix a vkernel lockup on startup

- During ap_init() any pending IPIs is processed manually so
  clear gd_npoll as the real kernel does.
- Do not disable interrupts for vkernels during lwkt_send_ipiq3()
  because they don't seem to be re-enabled afterwards as they should.
  I'm not entirely sure this is the right fix, more investigation
  is required.

8 days agoif_iwm - Recognize the IWM_UCODE_TLV_FW_MEM_SEG firmware section type.
Imre Vadász [Tue, 20 Sep 2016 22:35:40 +0000 (00:35 +0200)]
if_iwm - Recognize the IWM_UCODE_TLV_FW_MEM_SEG firmware section type.

* Will be needed for loading version 22 of 7265D firmware.

8 days agoiwmfw - Update to version 17 firmware for all devices supported by iwm.
Imre Vadász [Mon, 19 Sep 2016 19:26:11 +0000 (21:26 +0200)]
iwmfw - Update to version 17 firmware for all devices supported by iwm.

9 days agoif_iwm - Recognize IWM_DTS_MEASUREMENT_NOTIF_WIDE notification.
Imre Vadász [Mon, 19 Sep 2016 19:46:05 +0000 (21:46 +0200)]
if_iwm - Recognize IWM_DTS_MEASUREMENT_NOTIF_WIDE notification.

* Add the command groups enum, and the iwm_phy_ops_subcmd_ids enum
  to if_iwmreg.h definitions.

* The IWM_DTS_MEASUREMENT_NOTIF_WIDE notification will be generated by
  version 17 firmware.

Taken-From: Linux iwlwifi

9 days agoif_iwm - Very basic DTS thermal sensor support (prints temp as debug msg).
Imre Vadász [Thu, 14 Jul 2016 15:39:27 +0000 (17:39 +0200)]
if_iwm - Very basic DTS thermal sensor support (prints temp as debug msg).

* Adds IWM_DEBUG_TEMP debug message type, for printing messages related
  to temperature sensors and thermal/TDP infos.

* The firmware regularly sends us DTS measurement notifications, so just
  print the temperature value as a debugging message.

10 days agoif_iwm - Remove some more fields from struct nvm_data which are irrelevant.
Imre Vadász [Sun, 18 Sep 2016 15:33:01 +0000 (17:33 +0200)]
if_iwm - Remove some more fields from struct nvm_data which are irrelevant.

* These values are only used for dvm devices in Linux's iwlwifi
  (i.e. not relevant for iwm).

10 days agoif_iwm - Only for family 7000 power-down busmaster DMA clocks when stopping.
Imre Vadász [Sun, 18 Sep 2016 12:29:30 +0000 (14:29 +0200)]
if_iwm - Only for family 7000 power-down busmaster DMA clocks when stopping.

Taken-From: Linux iwlwifi

10 days agoif_iwm - Implement apmg_wake_up_wa workaround properly for 7000 family.
Imre Vadász [Fri, 16 Sep 2016 20:44:40 +0000 (22:44 +0200)]
if_iwm - Implement apmg_wake_up_wa workaround properly for 7000 family.

* Add iwm_pcie_set_cmd_in_flight() and iwm_pcie_clear_cmd_in_flight()
  helper methods.

* Use ring->queued tracking in the command queue to set/clear the
  cmd_hold_nic_awake bit at the right points.

Taken-From: Linux iwlwifi

11 days agoif_iwm - Fix up RSSI calculations for scan results and normal RX operations.
Imre Vadász [Sun, 18 Sep 2016 08:07:27 +0000 (10:07 +0200)]
if_iwm - Fix up RSSI calculations for scan results and normal RX operations.

* hard code a noise floor of -96 for now. The noise floor update code
  returns some "interesting" values that I can't map to anything useful
  right now.
* Ensure a default noise floor is set - otherwise the initial scan results
  have a noise floor of '0'.
* Fix up the RSSI calculation to be correctly relative to the noise floor.
  The RSSI routines return an absolute value in dBm - so fix this up.
* Cap RSSI values appropriately.
* Ensure we pass in a 1/2 dB unit value in to net80211.

Taken-From: FreeBSD (SVN r305917)

12 days agoBelatedly bump DF version after SSL libraries were privatized
John Marino [Fri, 16 Sep 2016 14:30:32 +0000 (09:30 -0500)]
Belatedly bump DF version after SSL libraries were privatized

13 days agoUpdate the pciconf(8) database.
Sascha Wildner [Thu, 15 Sep 2016 13:28:43 +0000 (15:28 +0200)]
Update the pciconf(8) database.

September 13, 2016 snapshot from http://pciids.sourceforge.net/

2 weeks agoif_iwm - Use iwm_mvm_scan_stop_wait to properly abort scans.
Imre Vadász [Sun, 17 Jul 2016 17:30:00 +0000 (19:30 +0200)]
if_iwm - Use iwm_mvm_scan_stop_wait to properly abort scans.

* Add IWM_FLAG_SCAN_RUNNING to sc->sc_flags to track whether the firmware
  is currently running a scan, in order to decide wheter iwm_scan_end
  needs to abort a running scan.

* In iwm_scan_end, if the scan is still running, we now abort it, in order
  to keep the firmware scanning state in sync.

* Try to make things a bit simpler, by reacting on the
  IWM_SCAN_OFFLOAD_COMPLETE and IWM_SCAN_COMPLETE_UMAC notifications,
  instead of IWM_SCAN_ITERATION_COMPLETE and
  IWM_SCAN_ITERATION_COMPLETE_UMAC. This should be fine since we always
  only tell the firmware to do a single scan iteration anyway.

2 weeks agoif_iwm - Add scan abort functions, to properly cancel a running scan.
Imre Vadász [Mon, 18 Jul 2016 20:03:27 +0000 (22:03 +0200)]
if_iwm - Add scan abort functions, to properly cancel a running scan.

* Uses the notification wait api to wait for the corresponding scan
  complete notification after sending the abort command.

Taken-From: Linux iwlwifi

2 weeks agoif_iwm - Get rid of iwm_disable_rx_dma, just use iwm_pcie_rx_stop directly.
Imre Vadász [Sun, 11 Sep 2016 20:23:50 +0000 (22:23 +0200)]
if_iwm - Get rid of iwm_disable_rx_dma, just use iwm_pcie_rx_stop directly.

* This also fixes one of many small nic lock handling bugs, and matches
  iwlwifi's code.

2 weeks agoif_iwm - Use notification wait API to wait for calibration to complete.
Imre Vadász [Sun, 11 Sep 2016 19:10:43 +0000 (21:10 +0200)]
if_iwm - Use notification wait API to wait for calibration to complete.

Taken-From: Linux iwlwifi

2 weeks agoif_iwm - Add implementation of the notification wait api from iwlwifi.
Imre Vadász [Mon, 18 Jul 2016 19:39:24 +0000 (21:39 +0200)]
if_iwm - Add implementation of the notification wait api from iwlwifi.

Taken-From: Linux iwlwifi

2 weeks agoif_iwm - Change 2nd arg of iwm_phy_db_set_section() to struct iwm_rx_packet.
Imre Vadász [Sun, 11 Sep 2016 16:12:28 +0000 (18:12 +0200)]
if_iwm - Change 2nd arg of iwm_phy_db_set_section() to struct iwm_rx_packet.

* This matches the function declaration in Linux's iwlwifi.

2 weeks agoif_iwm - Store paging_mem_size field in firmware image information struct.
Imre Vadász [Sun, 11 Sep 2016 15:48:46 +0000 (17:48 +0200)]
if_iwm - Store paging_mem_size field in firmware image information struct.

2 weeks agoif_iwm - Get rid of some gratuitous constant renaming wrt. Linux iwlwifi.
Imre Vadász [Tue, 6 Sep 2016 20:01:02 +0000 (22:01 +0200)]
if_iwm - Get rid of some gratuitous constant renaming wrt. Linux iwlwifi.

* IWM_UCODE_SECT_MAX -> IWM_UCODE_SECTION_MAX

* IWM_UCODE_TYPE_* -> IWM_UCODE_* (except for IWM_UCODE_TYPE_MAX which
  stays).

2 weeks agoif_iwm - Get rid of SYNC_RESP_STRUCT and SYNC_RESP_PTR macros.
Imre Vadász [Sun, 11 Sep 2016 15:18:02 +0000 (17:18 +0200)]
if_iwm - Get rid of SYNC_RESP_STRUCT and SYNC_RESP_PTR macros.

* Not needed, because we just do a single bus_dmamap_sync for syncing the
  complete received frame, instead of explicitly bus_dmamap_sync-ing
  subranges of the frame like in the original OpenBSD code.

2 weeks agoif_iwm - Remove unused fw_totlen field in struct iwm_fw_sects.
Imre Vadász [Sun, 11 Sep 2016 15:05:22 +0000 (17:05 +0200)]
if_iwm - Remove unused fw_totlen field in struct iwm_fw_sects.

2 weeks agosys/vfs/hammer: Change unused PFS ioctl flag to reserved
Tomohiro Kusumi [Fri, 9 Sep 2016 17:02:28 +0000 (02:02 +0900)]
sys/vfs/hammer: Change unused PFS ioctl flag to reserved

2 weeks agosbin/hammer: Add clrpfs()
Tomohiro Kusumi [Fri, 9 Sep 2016 16:00:24 +0000 (01:00 +0900)]
sbin/hammer: Add clrpfs()

The one in getpfs() is okay to clear pfs_id with -1,
because scanpfsid() will overwrite pfs_id without using it.

It's also okay to clear the whole &pfs before init_pfsd().
This is actually better than not clearing header part.

2 weeks agopowerd - Detect power state changes
Matthew Dillon [Thu, 8 Sep 2016 23:02:07 +0000 (16:02 -0700)]
powerd - Detect power state changes

* The list of available frequencies changes when the power state changes,
  detect such changes and set parameters for all cpus verses making
  incremental changes.

* Fixes issue with powerd leaving the laptop running at a lower frequency
  when the laptop is unplugged and then plugged back in.

2 weeks agosys/vfs/hammer: Remove wrong comment
Tomohiro Kusumi [Thu, 8 Sep 2016 16:07:09 +0000 (01:07 +0900)]
sys/vfs/hammer: Remove wrong comment

This comment is true for other ioctls, but not for HAMMERIOC_SET_PSEUDOFS.

hammer_mkroot_pseudofs() uses this ip to make sure localization
equals that of PFS#0. In other words, a new PFS can only be created
under PFS#0.

2 weeks agosbin/hammer: Return -1 if readlink(2) failed
Tomohiro Kusumi [Thu, 8 Sep 2016 12:24:10 +0000 (21:24 +0900)]
sbin/hammer: Return -1 if readlink(2) failed

If the path is a link and readlink(2) failed, it shouldn't continue
with empty buf[].

2 weeks agosbin/hammer: Refactoring
Tomohiro Kusumi [Thu, 8 Sep 2016 12:22:44 +0000 (21:22 +0900)]
sbin/hammer: Refactoring

fd is < 0 only when both scanpfs() and open(2) failed.

2 weeks agocorepower: Fix identify method for multi-cpu-package systems
Sepherosa Ziehau [Thu, 8 Sep 2016 13:15:06 +0000 (21:15 +0800)]
corepower: Fix identify method for multi-cpu-package systems

CHIP_LEVEL means physical cpu package, while PACKAGE_LEVEL means
the whole box; admittedly it is kinda confusion to me at least.

3 weeks agoFix build with NO_OPENSSL set
John Marino [Thu, 8 Sep 2016 06:21:59 +0000 (01:21 -0500)]
Fix build with NO_OPENSSL set

In addition to several missing CFLAGS to point to the private libressl
headers, various issues related to libressl were fixed.  These were
exposed by setting NO_OPENSSL in make.conf (and they were obscured
when the compiler found the original openssl headers).

3 weeks agolibc - restir arc4random() on fork()
Matthew Dillon [Thu, 8 Sep 2016 06:09:12 +0000 (23:09 -0700)]
libc - restir arc4random() on fork()

* Fix an issue where the arc4random() function was not being re-stirred
  on a fork.

Reported-by: zrj