Pull up following revision(s) (requested by khorben in ticket #671):
external/bsd/ipf/dist/lib/interror.c: revision 1.4
sys/external/bsd/ipf/netinet/fil.c: revision 1.16
Fix for PR kern/48109 (and its duplicate kern/49807)
As provided by Takahiro HAYASHI in PR kern/48109. Additional error
registration in ipf(8) by myself. Changes tested with GENERIC and
XEN3_DOM0. Thanks!

(snj)

Pull up following revision(s) (requested by martin in ticket #670):
lib/libc/softfloat/Makefile.inc: revision 1.20
Restrict the arm compiler optimization hack to gcc 4.5

(snj)

Pull up following revision(s) (requested by skrll in ticket #669):
sys/arch/arm/arm32/pmap.c: revision 1.318
Fix two bugs.  pmap_is_cached fix for MULTIPROCESSOR (not just ASID on
local cpu -> any valid ASID on any cpu).
pmap_deactivate: update curcpu()->ci_pmap_cur_asid to KERNEL_PID too.

(snj)

Pull up following revision(s) (requested by bouyer in ticket #668):
external/gpl3/gcc/dist/gcc/config/host-netbsd.c: revision 1.2
Use the same TRY_EMPTY_VM_SPACE as mips and vax for arm, which also
has VM_MAXUSER_ADDRESS set to 2GB. Fixes "fatal error: had to relocate PCH"
from g++ building wxGTK28 or wxGTK30.
For details see https://gcc.gnu.org/bugzilla/show_bug.cgi?id=58379
thanks to Martin Husemann for the hint.

(snj)

Disable mmap on zfs for now.

The implementation is not coherent with read/write, and apparently it
is is so broken that write(2) on a region mapped from a zfs file will
often fail with EINVAL because something passes an offset that is too
large to getpages.

(riastradh)

KASSERT about page-alignment on initialization too.

(riastradh)

Use KASSERTMSG to show bad base/offset.

(riastradh)

Whitespace.

(skrll)

Need the page's uvm object lock to do pmap_page_protect.

(riastradh)

Pull VOP_BMAP/VOP_STRATEGY loop from getpages into its own function.

No functional change.

In preparation for a gop_read like the existing gop_write.

(riastradh)

Fix a double free. "Suggested" by Brainy.

ok rjs@ riastradh@

(maxv)

add PARALLELLA and ZEDBOARD

(hkenken)

include "GENERIC.common"

(hkenken)

Fix casts which are inconsistent with *printf() formats and remove the
use of a non-standard type in the process.

(tron)

It's 2015.

(wiz)

Remove trailing whitespace.

(wiz)

Typo in previous.

(uebayasi)

More about attribute -> bus, and child configuration.

(uebayasi)

almost add radeondrmkms suspend/resume support.  it unfortunately doesn't work.

(mrg)

Add #define __HAVE_PIC_PENDING_INTRS for the non-GIC PICs.

(matt)

Pass NetBSD VA_UTIMES_NULL through to ZFS as Solaris ATTR_UTIME.

Makes t_unpriv:zfs_times fail the wrong way instead of silently
allowing the change.

(riastradh)

Apparently I fixed t_unpriv:zfs_owner and didn't notice.

(riastradh)

Advisory locking for zfs.

(riastradh)

Remove some vestigial #if 0 sections.

(riastradh)

Forgot to tweak date

(sjg)

jobs mode handles '-' very differently to compat mode, but this is not
reflected in make.1
Also fix handling to match the behavior described in the comment
of JobPrintCommand (only matters if shell sets hasErrCtl).

A better long term solution is needed since the current behavior
is sub-optimal wrt '-' and different from all other makes.

Reviewed by: christos

(sjg)

add new test

(christos)

Split inet_addr(3) tests from t_inet_network into stand-alone t_inet_addr,
so that t_inet_network only contains tests for inet_network(3).

(ginsbach)

Rename beagleboard.img to armv7.img. The new image includes the same
kernels as beagleboard.img plus support for Raspberry Pi 2, ODROID-C1,
Cubieboard2, Cubietruck, Hummingbird A31, and Banana Pi.

(jmcneill)

Only if we __HAVE_PIC_PENDING_INTRS do we need the variables to track them.

(matt)

Fix vattr_init_mask: mode is mode_t, not u_short.

Fixes fs/vfs/t_vnops:zfs_attrs test.

(riastradh)

But rename(..., "x/..") is still supposed to yield EINVAL.  Go figure.

(riastradh)

Tests claim rename(..., "x/.") yields EISDIR, so do that.  Fixes zfs.

(riastradh)

Fix return code of rmdir(nonempty) in zfs: ENOTEMPTY, not EEXIST.

(riastradh)

All SGIs are MPSAFE so establish them as such.

(matt)

Add __HAVE_PIC_HAVE_PENDING_INTRS and define it if __HAVE_PIC_SET_PRIORITY
is undefined (also define in mvsoc_intr.h since their use of the latter is
peculiar).  This new define controls whether the pending interrupt logic is
compiled.  The GIC doesn't use pending interrupts since it uses the priority
level on the GIC to control delivery of interrupts, thus there can never
be a pending interrupt.  The kernel shrinks about 4KB with the removal of
the pending interupt support,

(matt)

Make sure TPIDRPRw_IS_CURLWP is being used when __HAVE_PREEEMPTION is defined.

(matt)

Add basic support for __HAVE_PREEMPTION.
Use atomic ops for ci_astpending if __HAVE_PREEMPTION is defined.

(matt)

Don't clear CI_ASTPENDING in exception return, do it in ast() instead.
Add basic support for __HAVE_PREEMPTION.
Use atomic ops for ci_astpending if __HAVE_PREEMPTION is defined.
Use kpreempt_disable/kpreempt_enable

(matt)

CI_ASTPENDING is now a bitfield and low bit is for user ASTs.

(matt)

Small tweaks for preemption.

(matt)

ssh_packet_write_wait() returns number of bytes now; check for negative
for error instead of 0. Fixes ssh command restrictions, reported by
Tobias Nygren.

(christos)

Fix a bit of indentation.

(riastradh)

Remove unnecessary, and actively wrong and harmful, locking.

Obviously this `enable callback' never actually worked, because it
would deadlock against itself as soon as you try it.

(riastradh)

Don't do broken lazy initialization.  Just do it up front.

(riastradh)

Gather (some) rnd_skew state into a cacheline-aligned static struct.

(riastradh)

Don't forget to bus_dmamap_unload.

(riastradh)

Omit needless state.

(riastradh)

regen after syscall update

(justin)

Add __posix_rename as a rump syscall

(justin)

Don't read past the end of the sample buffer...  (Seriously???)

(riastradh)

Restrict the arm compiler optimization hack to gcc 4.5

(martin)

Added pmf hook.

(nonaka)

Add pmf hook.

fix PR/49783 USB WiFi ural0 rejects hw.acpi.sleep.state=3.

(nonaka)

Make TPIDRPRW_IS_CURLWP work for MULTIPROCESSOR.
get curcpu() from new lwp.
don't set lwp l_cpu (already done).
Remove support for __HAVE_UNNESTED_INTRS
don't set curlwp until after we are done saving the oldlwp.
disable interrupts when setting curlwp/kernel stack pointer.
Overall, these changes simplify cpu_switchto even more.

(matt)

Include sys/lwp.h instead of arm/cpu.h

sys/lwp.h is preferred because arm/cpu.h has a dependency on sys/lwp.h.

Suggested by skrll@ and matt@

(ozaki-r)

Add an agnostic version of curlwp and curcpu() for modules since they won't
know if the kernel was built with TPIDRPRW_IS_CUR{LWP,CPU} or not.

(matt)

remove support to look for GCC 4.5 in gcc.old.

(mrg)

add prototype declarations

(knakahara)

regen

(wiz)

Update Xorg defaults description. From Joachim Henke on netbsd-docs.

(wiz)

Update some defaults, from Joachim Henke on netbsd-docs.

(wiz)

Oops -- we already hold rndpool_mtx here.

(riastradh)

Simplify logic to disable HWRNGs that fail rngtest.

- Mark them RND_FLAG_NO_COLLECT.
- Skip samples from sources marked RND_FLAG_NO_COLLECT.
- Don't ever call rnd_detach_source in softint context.
- Remove bogus use of cpu_softintr_p.

The call to rnd_detach_source in softint can't possibly have ever
worked: if the owner of the source ever tried to detach it by calling
rnd_detach_source again, it would LIST_REMOVE a removed element,
which is broken.

(riastradh)

Name the simpleq of rnd_sample_t structure.

(riastradh)

Gather sample queue and lock into a static cacheline-aligned struct.

(riastradh)

Let's use the same lock for insertion and deletion, shall we?

(riastradh)

Reduce some indentation.

(riastradh)

Reduce #ifdef RND_VERBOSE clutter.

(riastradh)

Make rnd_extract_data_locked static.

(riastradh)

Don't reserve if atomic -- caller must have pre-pinned the buffer.

(riastradh)

Don't reserve if atomic -- caller must have pre-pinned the buffer.

(riastradh)

Move in6if_do_dad() to if_do_dad() as the routine is not INET6 specific
and could equally be used by INET.

(roy)

delete GCC 4.5.

(mrg)

__HAVE_UNNESTED_INTRS is never used.

(matt)

Fix two bugs.  pmap_is_cached fix for MULTIPROCESSOR (not just ASID on
local cpu -> any valid ASID on any cpu).
pmap_deactivate: update curcpu()->ci_pmap_cur_asid to KERNEL_PID too.

(matt)

fix typo in previous; pass -y unconditionally to resize_ffs, not -p

(jmcneill)

remove ancient dup.

(christos)

From: Joachim Henke

In -current, build.sh fails due to missing zless/zless.1. The attached
patch brings external/zlib/pigz/bin/pigz/Makefile in sync with
usr.bin/gzip/Makefile, which fixes the build.

(christos)

largely apply patch from PR bin/47392 by Abhinav Upadhyay

change some comments to reflect reality, a variable name to enhance
readability, and adds an assert for safety.

(plunky)

new ntp

(christos)

merge conflicts

(christos)

change _whatdb => _mandb as that is the correct keyword here,
and move it to the correct position in the list.

(plunky)

Oops.  Fix previous.

(skrll)

Deal with super speed port powered reporting.  From Takahiro HAYASHI.

(skrll)

Some quirk flags.  From Takahiro HAYASHI.

(skrll)

Comments.  From Takahiro HAYASHI.

(skrll)

USB 3.0 snapshot from Takahiro HAYASHI

(skrll)

Another DPRINTF.  From Takahiro HAYASHI.

(skrll)

KNF

(skrll)

Clean up rtcache_setdst

We can assume sockaddr_copy never return NULL when we pass
non-NULL dst (1st argument).

(ozaki-r)

Add missing #include <arm/cpu.h>

(ozaki-r)

Add missing semicolon

(ozaki-r)

show resize_ffs progress bar when resizing root partition

(jmcneill)

if the -p flag is specified, display a progress bar while growing the file-system

(jmcneill)

add KEYWORD: interactive, honour resize_root_flags and resize_root_postcmd rc vars

(jmcneill)

initialize resize before pulling in board config

(jmcneill)

Document xz decompression support better. From Joachim Henke on netbsd-docs.

(wiz)

Grammar, whitespace; make AUTHORS section from HISTORY since it mostly cites
people.

(wiz)

Add support for auto-growing the root partition. Enable it for rpi.img.

(jmcneill)

Cosmetic/Whitespace changes from Takahiro HAYASHI

(skrll)

666, the number of the pullup ticket (666+1 as well)

(snj)

Pull up following revision(s) (requested by jmcneill in ticket #667):
distrib/utils/embedded/mkimage: revision 1.54
resizing of ffsv2 still doesn't work well, so revert previous, and instead
explicitly set fsize=2048/bsize=16384, which gives a pretty decent boost
to resize_ffs performance on its own

(snj)

Pull up following revision(s) (requested by hannken in ticket #666):
sys/dev/sysmon/sysmon_envsys.c: revision 1.129
sys/dev/sysmon/sysmon_envsys_events.c: revision 1.114
sys/dev/sysmon/sysmonvar.h: revision 1.46
Add a counter of busy events and stop enqueueing more work if a device =
is busy.
Protect this counter with a new short time lock "sme_work_mtx" and
keep "sme_mtx" as long time lock.
Removes a deadlock where an active event holds "sme_mtx", the callout
"sme_events_check" blocks on "sme_mtx" and callout processing stops.

(snj)

(skrll)

resizing of ffsv2 still doesn't work well, so revert previous, and instead
explicitly set fsize=2048/bsize=16384, which gives a pretty decent boost
to resize_ffs performance on its own

(jmcneill)

btmagic(4): add Magic Trackpad support

(bouyer)

Mention magic trackpad support

(bouyer)

Add support for Apple Magic Trackpad.
3 button emulation by detecting in which area of the bottom of
the device the trackpad's button is pressed.
Pointer move support with 1 finger touch, X/Y scroll with 2-finger touch.
TODO:
- detect tap to emulate button press and drag/n/drop.
- Detect and support zoom, if wsmouse allows to report this

(bouyer)

Regen (Add Apple Magic Trackpad)

(bouyer)

Add Apple "Magic Trackpad" device id (bluetooth device)

(bouyer)

Sync with HEAD

(skrll)

Parse config file like ifconfig.if(5).
First word is always the module name, followed by options.
Load modules right after root filesystem is checked and before it
becomes writable.

(mlelstv)

Add UTC as a synonym for GMT (%Z).  [from FreeBSD]

(ginsbach)

Fix important typo, from Joachim Henke on netbsd-docs.
Bump date.

(wiz)

Add CI20 install kernel and build that and the normal image as kernel set.

(joerg)

Allow local overrides in a CI20.local file

(martin)

factor out geometry calculation and use this also in CheckOnly mode.
be a bit more verbose in Verbose mode.

(mlelstv)

Handle case in grow() where last cylinder group is too small for ufs2.
Align with code in shrink().

(mlelstv)

Remove useless break;

(skrll)

ehci_init returns int not usb_status_t

(skrll)

ehci_init return int not usb_status_t

(skrll)

Make DIOCKLABEL work. Set default to keep the disklabel after close to
not change current behaviour.

(mlelstv)

Force gcc to pass -msoft-float to the assembler (to make newer binutils
happy)

(martin)

Classify and sort prototype declarations

No functional change.

(ozaki-r)

Fix membars around rtld internal mutex.

This fixes the most of lockups i observed with Open vSwitch
on NetBSD/amd64.  ("most of" because it still occasionally
locks up because of other problems.  see PR/49816)

(yamt)

Use anonymous unions in various structs and rename the ctl/iso members of
the pipe structs to ctrl/isoc.

No functional change intended.

(skrll)

Change udf to vcache, keyed on udf_node->loc.loc.

(hannken)

Make rt_maskedcopy static

(ozaki-r)

Move the removal of unlinked nodes from udf_inactive() to udf_reclaim().

(hannken)

Make sure to halt (not just stop) the bge_tick callout during detach.

(martin)

Use an anonymous union in struct ehci_pipe and rename the ctl member to
ctrl.

No functional change.

(skrll)

Add hint comments for big ifdef

(ozaki-r)

Include sys/sysctl.h ifdef RTFLUSH_DEBUG

(ozaki-r)

Remove unnecessary inclusions

(ozaki-r)

Missed a zero: it's 0x100000, not 0x10000.

(riastradh)

tickets 655-665

(snj)

Pull up following revision(s) (requested by joerg in ticket #665):
libexec/ld.elf_so/rtld.c: revision 1.176
lib/49813: Release mutex on error.

(snj)

Pull up following revision(s) (requested by jmcneill in ticket #664):
distrib/utils/embedded/mkimage: revision 1.53
Use FFSv2 instead of FFSv1 for the root file-system. This should fix the
"resize_ffs takes forever" issues with evbarm images -- growing a v2
file-system is much faster than v1.

(snj)

Pull up following revision(s) (requested by jmcneill in ticket #663):
sys/arch/arm/arm32/cpu.c: revision 1.105
Increase the `freqbuf' buffer that holds the result of humanize_number(). Now
it prints 1600 Mhz instead of 1 Ghz.
Pitty enough humanize_number(9) doesn't do 1.6 Ghz type output.

(snj)

Pull up following revision(s) (requested by skrll in ticket #662):
sys/arch/arm/cortex/gtmr.c: revision 1.10-1.14
port-arm/49737: armgtmr0 timer broken
Use physical timer instead of virtual timer for timecounter. For platforms
that implement virtualization extensions, the CNTVOFF register defines a
virtual offset between the physical count and virtual count. Unfortunately,
the CNTVOFF register is only accessible in secure mode and the value is
per-CPU, so we may end up in a scenario where virtual count reads from
CPU A -> B -> A are not monotonic. No offset applied to physical timer,
so physical count reads are guaranteed to be monotonic.
--
Sprinkle some arm_isb() to force execution of the mcrr instructions to get
more accurate values.  (Given that the generic timer is usually running at
a 64th or less of the cpu clock, this only settle to a tick or so.)
--
Move isb's to before reading count.  Change interrupt to level.
--
Trailing whitespace.
--
Whitespace

(snj)

On swapcontext(3) preserve r12 too.  Properly fixes PR port-sh3/49597.

(uwe)

Pull up following revision(s) (requested by jmcneill in ticket #661):
sys/arch/arm/amlogic/amlogic_genfb.c: revision 1.5
sys/arch/arm/amlogic/amlogic_vpureg.h: revision 1.3
sys/arch/evbarm/amlogic/amlogic_machdep.c: revision 1.20
Allow for choosing between 16 and 24bpp framebuffers. The default is now
16 instead of 24. Set 'fb.depth=16' or 'fb.depth=24' on kernel command-line
to explicitly select a colour depth.

(snj)

Pull up following revision(s) (requested by jmcneill in ticket #660):
sys/arch/evbarm/conf/ODROID-C1: revision 1.15
comment out DDB_COMMANDONENTER

(snj)

Pull up following revision(s) (requested by jmcneill in ticket #659):
sys/arch/arm/amlogic/amlogic_com.c: revision 1.5
sys/arch/arm/amlogic/amlogic_comreg.h: revision 1.4
Fix cnmagic handling, now I can enter ddb from serial console.

(snj)

Pull up following revision(s) (requested by jmcneill in ticket #658):
sys/arch/arm/amlogic/amlogic_board.c: revision 1.11
sys/arch/arm/amlogic/amlogic_crureg.h: revision 1.8
sys/arch/arm/amlogic/amlogic_rng.c: revision 1.2
Use a callback (with ugly lock dance from bcm2835_rng) instead of callout.
Make sure the ring oscillator is enabled as well.

(snj)

Pull up following revision(s) (requested by mbalmer in ticket #657):
usr.sbin/sysinst/msg.mi.en: revision 1.7
usr.sbin/sysinst/msg.mi.pl: revision 1.7
IPv4 number -> IPv4 address

(snj)

Pull up following revision(s) (requested by hikaru in ticket #656):
sys/kern/subr_tftproot.c: revision 1.14
sys/nfs/krpc_subr.c: revision 1.39
sys/nfs/nfs_boot.c: revision 1.82
sys/nfs/nfs_bootdhcp.c: revision 1.53
sys/nfs/nfsdiskless.h: revision 1.31
m_pullup() is called in rcvproc callback functions,
so nfs_boot_sendrecv() should keep track of the head of mbuf chain.
fixes kern/48746

(snj)

Pull up following revision(s) (requested by martin in ticket #655):
sys/netinet6/in6.c: revision 1.182 via patch
sys/netinet6/in6_ifattach.c: revision 1.95 via patch
sys/netinet6/nd6.c: revision 1.158 via patch
sys/netinet6/nd6.h: revision 1.62 via patch
sys/netinet6/nd6_nbr.c: revision 1.104 via patch
sys/netinet6/nd6_rtr.c: revision 1.96 via patch
Rearange interface detachement slightly: before we free the INET6 specific
per-interface data, make sure to call nd6_purge() with it to remove
routing entries pointing to the going interface.
When we should happen to call this function again later, with the data
already gone, just return.
Fixes PR kern/49682, ok: christos.

(snj)

Don't let anyone else allocate the video BIOS either.

(riastradh)

fix typo bindinging -> binding

(rtr)

change return from EINVAL to EAFNOSUPPORT when the domain of the socket
does not match the family of the address to be bound.

fixes atf test lib/libc/sys/t_bind bind_foreign_family

(rtr)

add another test program for connect(2) that checks that connect fails
with EAFNOSUPPORT (similar to the bind test) if the domain of the socket
does not match the address family of the supplied address.

test currently fails as it should

(rtr)

-      ATF_REQUIRE(EINVAL == errno);
+      ATF_REQUIRE(EAFNOSUPPORT == errno);

as per http://pubs.opengroup.org/onlinepubs/009695399/functions/bind.html

while here remove some unnecessary initialization of port and addr members
and replace with a comment explaining why.

(rtr)

don't forget to include config.h otherwise it thinks we don't have vnsnprintf
and fortify compiles fail.

(christos)

Sort errors. Bump date for previous.

(wiz)

sun4v: Ensure that the TTEs CP and CV bits are the same (both set) for the primary cpu and secondary cpus INTSTACK mappings

(palle)

Note that bind(2) can fail with EAFNOSUPPORT.  The description of why is
taken from the connect(2) manpage.

(rtr)

sun4v: Ensure that the W (Write) bit is not present in the TTEs unless for the data mapping. This way the permanent mappings for the primary cpu done by ofwboot are the same as the permanent mappings for the secondary cpus done by the kernel. This should fix an issue when booting the kernel on a T2 based system (T5120).

(palle)

Use FFSv2 instead of FFSv1 for the root file-system. This should fix the
"resize_ffs takes forever" issues with evbarm images -- growing a v2
file-system is much faster than v1.

(jmcneill)

Don't use the video BIOS memory as an i915 flush page!

(riastradh)

More shell quoting.  /etc/rc.d/* file names with embedded spaces are
not expected to work, but let's quote them anyway.

(apb)

PR/49814: USB Quirk for Texas Instrument MSP430 FET programmer

(skrll)

Regen

(skrll)

More Texas Instruments (not Texas Intel) products.

Part of PR/49814

(skrll)

More debug.

(skrll)

Regen

(skrll)

Conexant USB Modem from Ryo ONODERA on current-users.

(skrll)

Remove dangling cross-ref to localtime(3).  There's another cross-ref in
the next sentence that actually makes sense.

(pgoyette)

Fix inverted logic.  Spotted by Takahiro HAYASHI.  Thanks.

(skrll)

Add new bind(2) test program

(martin)

New test program for bind(2), with a single (currently eroneously failing)
test case from Tyler Retzlaff.

(martin)

make bind() fail with EINVAL if the address family of the provided
socket does not match the address family of the sockaddr received.

(rtr)

build ops.h first, because trying to build dependencies and ops.h at the
same time leads to spurious failures.

(christos)

Ticket #653

(martin)

Apply commit fb6e5377f392555b8c725f66b8b701f0061a3695:
-Fix/add ND_TCHECK2 tests,
-Fix a buffer overflow,
-Remove a debug printf

(christos)

lib/49813: Release mutex on error.

(joerg)

In DRM_SPIN_WAIT_ON, don't stop after waiting only one tick.

Continue the loop to recheck the condition and count the whole
duration.

(riastradh)

fix leak, found by Brainy.

(christos)

fix leak, found by Brainy.

(christos)

another missing free dict.

(christos)