Add several "restrict" lines to the default ntp.conf, with comments.

(apb)

Add "cdroms" to the SUBDIR list for src/distrib/ews4800mips.

(apb)

Use Fn for functions and Dv for null.

(wiz)

Fix article.

(wiz)

Use Nx. Remove superfluous Pp. Increase column width to widest entry.

(wiz)

For filesystems mounted without -o use_ino, readdir is not
able to fetch inode number. We perfom an addtional lookup
on each file to get it.

In that case, do not lookup .. from root, as it breaks
out of the filesystem and hits NULL pointers.

(manu)

screen_end can be called multiple times for the same screen.
Work around segfault in TAILQ_REMOVE by abusing the tqe_prev pointer
as a flag that identifies an already removed node.

This should really be solved by calling screen_end only once or by
keeping state explicitely in the screen structure.

(mlelstv)

Make libsa fsmod string pointer const.

(jakllsch)

Make name argument to module_add_common const; use this to allow the libsa
fsmod string to soon become const.

(jakllsch)

sprinkle const

(jakllsch)

Drop duplicate and unused extern global variable.

(jakllsch)

No need to return a empty string literal of the wrong const-ness when we
already have a empty string of the correct const-ness.

(jakllsch)

don't shadow local variable

(jakllsch)

Restore cast hack for GCC's "new qualifiers in middle of multi-level
non-const cast" warning.

(joerg)

Update build system and import script for mdocml 1.12.3.

(joerg)

Merge mdocml 1.12.3.

(joerg)

Sync with interface change in mdocml 1.12.3.

(joerg)

PR/48502: M. Levinson: vi(1) no longer moves the cursor when yanking text
using a mark (remove dangling else)

(christos)

Update LLVM family to 3.5svn r198450. Interesting changes include:
- Improvements in the SPARC64 support
- ARM backend can almost build a full NetBSD/EABI system with the
  integrated assembler
- PowerPC grows disassembler support
- LLD supports copy relocation, beginning of MIPS backend
- New diagnostic for use of bool expression as size_t argument for
  strncmp and friends

(joerg)

- add new clk_sel240x.c

(kardel)

PR/48494: Ryo Onodera: After ACPICA 20131218 update, uhub topology is
misanderstood on macbook air. Ryo by bisection found that the following
commit causes the problem: https://github.com/acpica/acpica/commit/\
637accd073a205eb4acac935f34d0eaee0c301a8. This is tracked in:
https://bugs.acpica.org/show_bug.cgi?id=1016

(christos)

- enable the rest of the PARSE clocks (missed config option when merging)

(kardel)

Skip suffixes like -eabi for the host target, when invoking configure.

(joerg)

Build tools without debug symbols by default.

(joerg)

Use head to remove final #endif, newer versions will contain multiple
conditional blocks for dealing with Apple universal binaries.

(joerg)

- enable PARSE clocks (missed config option when merging)
- remove all conflict markers/finish merge in refclock_parse.c:1.10

(kardel)

We now expect this test to succeed

(martin)

PR bin/46743: mark the "next free cluster" value in the file system
info block as unitialized.

(martin)

Fix previous slightly differently.
No need to have an extra variable and to write to the list head.

(dsl)

Detect boot device on PROM load command and set proper default boot device.

Also remove now unnecessary the old netboot check.

Note all addresses of boot device info set by monitor are checked by
memory dump on several LUNA ROM versions, and these addres might be
different on earlier monitor ROM versions.

Bump version to denote the visible change.

(tsutsui)

Move physmem calculations before nptpage initialization.

Should fix next68k specific part of PR port-m68k/45915
(panic: pmap_enter_ptpage: can't get KPT page), and
this is the last possible m68k MD part of this PR.

Note this change is not tested on the actual machine (yet),
but as noted in comment next68k/pmap_bootstrap.c is based on
the mvme68k one which has been fixed by the similar diff.

Should be pulled up to all netbsd-6 branches.

(tsutsui)

avoid use-after-free in *coredump().
fixes kernel crashes during coredump on sparc64.

(mrg)

When crypto(4) is built-in, crypto_modcmd() doesn't need to handle all
the auto-config stuff.

While here, ensure that we depend on opencrypto.

(pgoyette)

Clean removed files.

(joerg)

Add Energy Performance Bias bit.

(msaitoh)

undo the ld script hack now that the default scripts have been modified to
do this.

(christos)

undo the linker script hack now that the linker scripts do this.

(christos)

gcc-4.8 has begun putting code in .text.unlikely etc. This breaks our
assumption that .text code comes first in kernels about bootblock so
that the first symbol defined becomes the start address. This change
puts .text before other .text like sections.

(christos)

Remove duplicated entry. Modify comments a bit.

(msaitoh)

- Rename x86_print_cacheinfo() to x86_print_cache_and_tlb_info() because
  this function prints TLB info, too.
- Remove an extra printf when verbose flag is set.
- Print the highest extended info level as the basic info level.
- Sort function.

(msaitoh)

__unuse the right function
( note to self - don't commit half asleep )

(macallan)

Merge EINVAL descriptions.

(wiz)

Remove second copy for RCS Id.

(wiz)

New sentence, new line.

(wiz)

Sort sections.

(wiz)

Sort sections.

(wiz)

Sort sections. Remove trailing whitespace.

(wiz)

Sort sections. Punctuation formatting nits.

(wiz)

New sentence, new line. Remove trailing whitespace.

(wiz)

IPv6 UDP uses the IPv4 pcb tables, and therefore the stats, so need
to create the percpu UDPv4 counters even in a v6-only system.

(pooka)

fix CS bit.

(nonaka)

FLOGNP1(-0.0) is -0.0, not +0.0.  Found by XM6i.

(isaki)

Fix a race where thread1 runs VOP_REMOVE() and gets preempted in
tmpfs_reclaim() before the call to tmpfs_free_node().  Thread2
runs VFS_FHTOVP() and gets a new vnode attached to the node thread1
is about to destroy.

Change tmpfs_fhtovp() to check the generation number after
tmpfs_vnode_get() succeeded.

(hannken)

Remove .note.netbsd.mcmodel and .note.netbsd.pax from the chrunched binaries.
We don't know wether the former would make sense (as currently used, it never
makes sense for static binaries) and we can not conclude safe PAX flags from
combined binaries.

(martin)

Fix string compare.

(joerg)

Add Z68 LPC.

(msaitoh)

Add missing volatile for hidden side effects of asm.

(joerg)

regen.

(msaitoh)

Add Intel Z68 LPC

(msaitoh)

Fix memcmp use.

(joerg)

Temporary disable C2000 PCU because the behavior of the wdog is little strange.

(msaitoh)

Explicitly force a CPU for the assembler for the compatibility
cpu_func_* files. LLVM complains about the use of deprecated
instructions when -march=armv7 or equivalent is in use.

(joerg)

Remove __HAVE_PROCESS_XFPREGS and add the extra parameter for the size
  of the fp save area to all the process_read_fpregs() and
  process_write_fpregs() functions.
None of the functions have been modified to use the new parameters.
The size is set for all the writes, but some of the arch-specific reads
  just pass NULL.
The amd64 (and i386) need variable sized fp register save areas in order
  to support AVX and other enhanced register areas.
These functions are rarely called - so the extra argument won't matter.

(dsl)

Add C2000 Platform Controller Unit(PCU).

(msaitoh)

Avoid implicit cast on the return value.

(joerg)

Note import of dhcpcd-6.2.0

(roy)

Build with dhcpcd-6.2.0

(roy)

Sync

(roy)

Fix bug in previous (panic during process core dump).
Change the interface to ELFNAMEEND(coredump_savenote) so that the caller
  doesn't need to know the type of the elf note header.
Simplifies the calling code somewhat.

(dsl)

Add a load script that puts text.unlikely after text so that the start
address stays where we are. This is copied from m68kelfnbsd.xbn

(christos)

There is no need for uvm_coredump_walkmap() to explicity pass the proc_t
  pointer to the calller's function.
If the code needs the process its address can be placed in the caller's
  cookie.

(dsl)

Instead of generating all the 'note' sections twice (and hoping that the
  'fast path' size on the first path matches the actual size on the second)
  save all the notes (mostly the cpu registers for all the LWPs) in
  malloced memory on the first pass.
Sanity check that the number of memory segments matches written matches
  the count obtained earlier. If gcore() is used they could differ.
  (Not sure that returning ENOMEM is ideal, but it is better than a crash.)

(dsl)

Don't bother with the h (short) modifier. The predicated use of bswap16
confuses the type tracking in clang and the use here is not portable
anyway, since it assumes short == int16_t.

(joerg)

fix sun2 build. libfreetype requires -lbz2 -lz

(christos)

Remove the incorrect and unused Elf64_Shalf.
At one time it was used for a 32bit item - but that definition has long
  since been changed to use a reasonable 32bit type.

(dsl)

Remove the borked definition of ELF64_Shalf from some more places.
Remove the _KERNEL parts of some clones of sys/elf_exec.h
One might ask why this file has got cloned so many times.

(dsl)

Remove the _KERNEL part of these two clones of sys/exec_elf.h
Remove the incorrect (and completely unused everywhere) Elf64_Shalf.
Put a better include guard on arch/epoc32/stand/e32boot/include/elf.h
Should be ok because I can't find any references to KERNEL in these
  source trees.

(dsl)

Add hifn(4) and ubsec(4) to the NOTYET section, and note that they can
be successfully built only on architectures that have PCI.

(pgoyette)

Add module glue for hifn(4)

(pgoyette)

Modularize

(pgoyette)

If you're going to build this driver as a module, make sure that it
depends on opencrypto module.

(pgoyette)

do as joerg requested and mark an unused function __unused

(macallan)

Simplify error path and fix typos. From Maxime Villard and me.

(christos)

Minor changes to the process coredump code.
- Add some extra comments.
- Add some XXX comments because the process state might not be stable,
- Add uvm_coredump_count_segs() to simplify the calling code.
- uvm code now only returns non-empty sections/segments.
- Put the 'iocookie' into the 'cookie' block passed to uvm_coredump_walkmap()
  instead of passing it through as an additional parameter.
amd64 can still generate core dumps that gdb can read.

(dsl)

Small bit of tidyup and remove a XXX now that it's possible.

(skrll)

Change wildcard for mvgbe configuration.  Don't attach, if clock was disabled.

(kiyohara)

netstat expected output includes ipv6 sucket, so add netinet6

(pooka)

ldpd wants inet6

(pooka)

missed one inet6 check

(pooka)

Call callout_reset on the xfer timeoute_handle before calling
dwc2_hcd_urb_enqueue

(skrll)

Change a KASSERT to a KASSERTMSG and add a DPRINTFN.

(skrll)

Fix undefined reference to amiga_bus_stride_0x1000.

(rkujawa)

Fix a race where thread1 runs VOP_REMOVE() and gets preempted in
tmpfs_reclaim() before the call to tmpfs_free_node().  Thread2
runs VFS_FHTOVP() and gets a new vnode attached to the node thread1
is about to destroy.

Change tmpfs_alloc_node() to always assign non-zero generation number
and tmpfs_inactive() to set the generation number of unlinked nodes
to zero.

(hannken)

Port from newer gcc:
Add a "mem_noofs_operand" predicate and corresponding "w" constraint.
Use those instead of "memory_reg_operand"/"m" for some atomic instrinsic
patterns: casx (and friends) do not accept an offset from the pointer
register (they are synthetic instructions mapping to e.g. casxa [..]
ASI_P, ... and the opcode space encoding offsets in other instructions
is needed to encode the ASI here).

This fixes the build of gtk+-3.10.6 from pkgsrc on sparc64.

(martin)

Pull more cleanups from OpenBSD/luna88k:

- remove unused declarations
- rename struct scsi_fmt_cdb to scsi_generic_cdb

(tsutsui)

Fix allocator routines. Instead of abusing CIRCLEQ pointers, add an explicit
type field to tag memory nodes that are or are not on the free list.

(mlelstv)

Pull more fixes from OpenBSD/luna88k:

- Bring getline() - a.k.a libsa gets() with a prompt prefix - in par with libsa
  gets(), featurewise; this means support for ^u to clear the input.
- constify

(tsutsui)

Pull more fixes from OpenBSD/luna88k:

- accept empty controller and partition numbers, as well as empty filenames,
  and use defaults (0, 0 and "netbsd") instead of complaining the boot path
  is invalid
- move a macro where actually necessary

Also bump version to denote the user visible change.

(tsutsui)

Disable unnecessary boot device check. From OpenBSD/luna88k.

(tsutsui)

Misc cleanup from OpenBSD/luna88k.

- remove unused code
- use proper prefix for structure members for readability

(tsutsui)

move crypto_openssl.c back into the right room.

(christos)

mention new wpa

(christos)

merge conflicts

(christos)

update for 2.0

(christos)

Remove unused code (from OpenBSD/luna88k). Some KNF.

(tsutsui)

Rework handling of Commodore Gayle chip. Avoid ugly struct casts, introduce
proper bus_space accesses and some abstraction layer. All drivers that utilised
Gayle also had to be refactored.

I tried not to break anything more, but this clearly needs more testing...

(rkujawa)

Treat '~' as a meta char requiring a shell.
Patch from Steve McIntyre 93sam at debian.org

Reviewed by: christos

(sjg)

PR/4891: Wiz: readlink moved from xopen to base at issue 5.

(christos)

Add myself

(tho)

Update warning flags to reflect the fixes. Restrict GCC flags to GCC.

(joerg)

Don't pass in a format string, but flags for using %f / %g and the
desired precision. Allows format string verification to work correctly.

(joerg)

Add format string attribute for better detection.

(joerg)

Avoid pointer operations on the format string.

(joerg)

don't lose the RPC error from CLNT_CALL.

(christos)

Misc cleanup.  Mostly from OpenBSD/luna88k.

- fix isprint() macro
- hight -> height
- remove an unused function
- adjust tabs/spaces
- constify the bitmap font
- consistently use #define<space>
- use #if 0/#endif to disable code block
- u_long -> uint32_t, u_short -> uint16_t
- some KNF

(tsutsui)

Replace broken (by SCCS) ISO-2022-JP comments with UTF-8.

Recovered and decoded by isaki@.

(tsutsui)

disable ctf handling for the shuttle .c -> '' rule since we don't create
an object file to work on.

(christos)

add gnu attributes to the known sections

(christos)

Add more gnu sections

(christos)

Fix const use.

(joerg)

need atomic.h, from uwe

(pooka)

Add bound check of a controller number in scrun().

Noticed by miod@openbsd.

(tsutsui)

Allow kernels compiled with INET+INET6 to be booted as IPv4-only or IPv6-only.

(pooka)

Use separate variables for the format strings to get working format
string checking with both GCC and Clang.

(joerg)

Fix a few "share" flags

. Add missing "share" flag on usr/share/examples/npf files.
. Drop "share" for directories in examples/{libsaslc,npf,tmux}. This is how it
  is done for other directories in usr/share.

This fixes checkflist with MKSHARE=no.

ok christos@.

(tho)

Do not install files in usr/share/ when MKSHARE=no

ok christos@

(tho)

Change the SCSI select timeout from 2ms to 250ms as kernel spc(4) driver does.

Noticed by miod@openbsd.

(tsutsui)

Change the SCSI select timeout from 2ms to 250ms as kernel spc(4) driver does.

Noticed by miod@openbsd.

(tsutsui)

allow arm*{,eb]-*-netbsdelf*

(matt)

Drop superflous "picinstall" flag on "obsolete" items

Reported by apb@, thanks!

(tho)

Protect access of urb->hcpriv by adapting the following change from the
Raspberry PI dwc_otg driver.

https://github.com/raspberrypi/linux/commit/38753ce72d4f10d5d0f1ed27fa691a2ba8910941

dwc_otg: prevent OOPSes during device disconnects

The dwc_otg_urb_enqueue function is thread-unsafe. In particular the
access of urb->hcpriv, usb_hcd_link_urb_to_ep, dwc_otg_urb->qtd and
friends does not occur within a critical section and so if a device
was unplugged during activity there was a high chance that the
usbcore hub_thread would try to disable the endpoint with partially-
formed entries in the URB queue. This would result in BUG() or null
pointer dereferences.

Fix so that access of urb->hcpriv, enqueuing to the hardware and
adding to usbcore endpoint URB lists is contained within a single
critical section.

(skrll)

grep -2.16 out.

(wiz)

Mark all _pic.a archive files with the 'picinstall' flag.

This fixes checkflist with MKPICINSTALL=no.

ok christos@.

(tho)

sort the contributor list, and update the defines list to include lua.

(mrg)

update for 20140102.

(mrg)

update version, date and copyright.

(mrg)

- update CHANGES with recent changes
- update version to 20140102
- update copyrights
- use getcwd() over getwd()
- fix lean build (don't include lua)

(mrg)

rename libtool.m4 m4

(christos)

Remove initialization of unused t_softc in struct tty (to make pullup easier).

(tsutsui)

Define kernel_pmap_store as a pmap followed by ALPHA_MAXPROCS pmap_asn_info.
kernel_pmap_ptr now points to the pmap in kernel_pmap_store.

(matt)

remove commented out -fno-reorder-functions

(christos)

Use a custom ldscript so that we can keep boot_start at 0.

(christos)

Provide a boot ldscript that puts the text.unlikely section after the text
section so that boot_start gets placed first. We also set the entry and the
start address, although we still override them from the command line.
This is a copy of elf_i386.xbn

(christos)

fd == 0 is a valid descriptor!

(christos)

whitespace fixes

(christos)

There was a case where \n did not increase plinno

(christos)

Change the type of the 'cookie' that holds the state of the core dump file
from 'void *' to the actual type 'struct coredump_iostate *'.
In most of the code the contents of the structure are still unknown.
This just stops the wrong type of pointer being passed to the 'void *'
parameter.
I hope I've found everything, amd64 GENERIC and i386 GENERIC & ALL compile.

(dsl)

clarify further.

(christos)

explain the previous fix.

(christos)

Re-sort

(pgoyette)

Fix PMAP_SIZEOF() to deal with that the pmap has 0 pmap_asn_info now.
Addresses PR/48488.

(matt)

Add new modules for opencrypto

(pgoyette)

Modularize the opencrypto components and link to the build

(pgoyette)