Merge -r1.66 of dir.c, -r1.171 of job.c, -r1.186 of parse.c from HEAD
(committed by Christos):

Keep track of the location where a dependency is defined, so we can report
about it.

(dholland)

Merge -r1.170 of job.c and -r1.209 of main.c from HEAD (committed by
Christos):

PR/45042: Thomas Cort: HAVE_foo for setrlimit(2) and setpgid(2)

(dholland)

Use real types.

(dholland)

Fix inlining again, more forcibly this time.

(dholland)

Buhbye.

(dholland)

Drop the list library's header.

(dholland)

Switch off the list library.

(dholland)

Remove remaining references to the linked list library.

(dholland)

Use real types.

(dholland)

G/C Dir_CopyDir(), which is no longer used.

(dholland)

Rip the linked lists out of here and make it build again.

(dholland)

Update some #if 0'd code to still compile.

(dholland)

Improve the toBeMade queue code so it doesn't do bulk array shifts
every time you remove something from the queue head. If this turns out
not to be good enough it can be tuned and/or made into a circular
queue later.

(Also, in the long run, the non-tail insertions may go away; they seem
to chiefly be associated with the .ORDER implementation and may not be
needed after the primary graph structure gets some strengthening.)

(dholland)

Improve the toBeMade logic: collect the non-tail insertions in a
separate array and stuff them into the main queue all at once.

(dholland)

Switch the toBeMade logic to an array instead of a linked list.

(dholland)

Simplify the toBeMade logic based on the fact that only one group of
non-tail insertions is done at a time, so the insertion point can be
tracked by the queue.

(dholland)

Wrap the toBeMade list code in a queue abstraction. Use the same list
code on the inside for now.

(dholland)

Change the order of register settings a little to match both Broadcom's
document and Linux tg3.

(msaitoh)

Remove another linked list.

(dholland)

5718 and 57785 document say we should wait 100us.

(msaitoh)

Remove another linked list.

(dholland)

Update comment.

(dholland)

Remove the last linked list.

(dholland)

More arrays, fewer lists.

(dholland)

Oops, left CLEANUP on for testing; turn it off again.

(dholland)

Update another comment.

(dholland)

Update a comment.

(dholland)

Update some more comments.

(dholland)

Update comment for previous

(dholland)

Use real types.

(dholland)

Use more const.

(dholland)

Reorg some of the comments.

(dholland)

Remove non-en translations. en is enough for poor Tier II ports.

(tsutsui)

Rewrite SuffSuffIsSuffix to avoid potential undefined behavior.

(dholland)

Simplify the interface to SuffSuffIsSuffix().

(dholland)

More arrays, fewer lists.

(dholland)

Make the DEBUG_SRC code compile again.

(dholland)

Remove unused back-pointers.

(dholland)

More arrays, fewer lists.

(dholland)

Remove rarely maintained and required non-en translations.

(tsutsui)

Update comment.

(dholland)

More arrays, fewer lists.

(dholland)

one less, 6 left.

(christos)

remove extra space

(christos)

upgrade file

(christos)

bump libmagic

(christos)

merge new file

(christos)

typo.

(wiz)

Sync non-English messages with the changes of msg.mi.en 1.179 to the
best of my ability.

(gson)

import file-5.14
changes are "name" + "use" keyword features, bug fixes

Status:

Vendor Tag: CHRISTOS
Release Tags: FILE5_14

(christos)

correct pdisk information

(christos)

pdisk has moved

(christos)

add Makefile

(christos)

move pdisk

(christos)

Apply our fixes

(christos)

Fix some nits in the English messages for the recently added debug
sets: rename the "Source sets" group to "Source and debug sets" to
better reflect its current content, describe the indivdual sets as
containing as "debug symbols" instead of "debug sets", and capitalize
them to be consistent with the messages for other sets.

(gson)

from:
http://cantaforda.com/cfcl/eryk/linux/pdisk/dist/pdisk.20000516.src.tar

Status:

Vendor Tag: ERYK
Release Tags: PDISK20000516

(christos)

try to fix sun2

(christos)

Use the #error directive.

(isaki)

Remove about updating fpsr. It was introduced by me but obviously
duplicated with fpu_emul_arith().

(isaki)

Add KUROBOX_PRO as an EVBARM_BOARDS member.

(he)

Add the ability to install the KUROBOX_PRO kernel.
Adds the ability to handle 9 different kernel sets (was 8).

(he)

If built with filemon support, set .MAKE.PATH_FILEMON to _PATH_FILEMON
so that makefiles can test for it.

(sjg)

Declare types before data.

(dholland)

Add unit-test for export-env and gmake export

(sjg)

ParseGmakeExport: terminate variable at '=', or setenv(3) will fail
on some systems.

(sjg)

Fix fpu_ftox().
Update not only exponential but also mantissa when an integer part
becomes 2 by rounding up.  Without this fix, the extended precision
value becomes 0.0 because mantissa (including explicit integer bit)
is all-zero.

(isaki)

Fix typo in function name in DIAGNOSTIC prinf by using __func__

(skrll)

When dumping the done list in ohci_softintr / OHCI_DEBUG use the correct
list next pointer.

PR/33450 and part of PR/35379.

(skrll)

Remove 'return' right after 'goto'

(skrll)

avoid arraycount for toolz

(christos)

dhclient -> dhcpcd

(tsutsui)

move the code to expand variables in one place and use it to expand
variables inside conditionals. still conditionals with variables don't
work.

(christos)

file-5.14 out.

(wiz)

mandoc can now print numbered registers.

(christos)

support expanding numbered registers

(christos)

Ooops, backout local change not intended to be commited with the last change

(martin)

Add syntatic sugar to avoid hurting my eyes

(martin)

Bump minor number for removal of kauth_access_action(), as per
  http://mail-index.netbsd.org/source-changes/2013/03/18/msg042340.html

Welcome to 6.99.18

(plunky)

new test directories

(christos)

Instead of always checking for ANYC, make it a valid channel

(christos)

PR/47678: Dr. Wolfgang Stukenbrock: remove local iscsi entry which conflicted
with the iana assigned one. This port has been assigned to iscsi-target.

(christos)

Don't attempt to dereference ANYC (since it is a pointer to 0x1).
Fixes random crashes in hostap mode (race conditions in the interrupt handler
while the interface comes up or down).

(christos)

new dlopen tests for libpthread.

(christos)

new dlopen tests for libpthread from manu@

(christos)

- Allow libpthread to be dlopened again, by providing libc stubs to libpthread.
- Fail if the dlopened libpthread does pthread_create(). From manu@
- Discussed at length in the mailing lists; approved by core@
- This was chosen as the least intrusive patch that will provide
  the necessary functionality.
XXX: pullup to 6

(christos)

fix ALL_KERNELS handling

(mlelstv)

Fix stat64 by making it match i386.  Other archs might still be broken,
but this is the one I could test.

(pooka)

PR/47677 Aktado: x86_errata() should be avoided if NetBSD runs as a KVM guest.
XXX: pullup to 6

(christos)

Wait for the bootcode to complete initialization for 5717 and newer devices.
See BCM5718 programmer's guide's "step 13, Device reset Procedure, Section 7".
57781's document has the same note.

I verified the loop really looped on those chips.

But, I don't know why tg3 doesn't it.

(msaitoh)

Add some defines from the USB 2.0 specification and suppliments. Also,
add some accessor macros while I'm here.

(skrll)

- Sync with FreeBSD and OpenBSD. Almost the same as OpenBSD rev. 1.325:
- Sync the ring setup code closer to FreeBSD's driver
- Do not touch the jumbo replenish threshold register on chips that do not
  have jumbo support
- Add/sync some of the comments
- Use macro.
- Remove unused code.

(msaitoh)

Add a bunch of names in the ChangeLog to THANKS.
Strip out email addresses; they're often out of date.

(lukem)

Convert from __attribute__(foo) to the __foo equiv in <sys/cdefs.h>

(lukem)

remove unnecessary compat define of __attribute__

(lukem)

Add missing .El.

(joerg)

Expand and/or tweak some entries.

(lukem)

Add some entries for people who submitted fixes by PRs.

(lukem)

regen with autoconf 2.69, automake 1.11.1, libtool 2.4.2

(lukem)

Style tweak: use () with argument-less macros.

(lukem)

Consistently use AM_CPPFLAGS

(lukem)

Tweak quotes in a description to avoid syntax highlight fubar

(lukem)

Require autoconf 2.69
Require libtool 2.4
Update to LT_INIT instead of AC_PROG_LIBTOOL
Crank copyrigth
Fix AC_PROG_YACC workaround to only be used if $YACC gets set to "yacc"

(lukem)

glob updated
imported NetBSD ftpd 20130321

(lukem)

Fix and revive test of atan_inf_neg, atan_inf_pos and atan_tan on i386.
PR port-i386/46108.
The machine epsilon 1.0e-40 is too severe and nonsense for double
because DBL_EPSILON is about 2.2e-16 .  I think that 1.0e-15 is
enough good, in this case.
XXX However, test of atan_tan should be replaced for other reasons.

(isaki)

fix groff html generation

(christos)

The man page has numbered registers and conditionals and cannot be handled
by mandoc, so make it use groff. Unfortunately since we don't install cat
pages, this does not work. At least it works for html!

(christos)

handle NOMANDOC in the html case too

(christos)

revert previous, this is a mandoc issue (does not handle .if and .nr)

(christos)

compat workarounds for __attribute__ and __dead

(lukem)

Merge NetBSD-20130321

(lukem)

Import NetBSD-ftpd 20100320:
- Fix warnings and build issues
- Fix memcpy usage
- Fix remote crash (fix was already pulled up manually to tnftpd).  PR/43023
- Reduce priority of syslog message if getpeername returns ENOTCONN.  PR/18934

Status:

Vendor Tag: TNF
Release Tags: NetBSD-20130321

(lukem)

PR/47675: Nick Hudson: top(1) man page missing default number of seconds

(christos)

Update version to 20110904 for the user-visible change I made back then:
  Reduce priority of syslog message if getpeername returns ENOTCONN.
  PR bin/18934 by Greg A. Woods (with supplied fix).

(lukem)

Use "unsigned FOO" instead of less-portable "u_FOO".

(lukem)

If the tripplite SMARTLCD ups connects as a usbhid device, it doesn't
work at all, connect as ugen instead.  Verified working with ups-nut.

(garbled)

Regen

(garbled)

Add my tripplite UPS

(garbled)

apply kludge to make this more or less work on mips64 with n32 userland
with this ldd can handle both 64 and n32 binaries ( previously it would do 64
only ), o32 support is still broken
Someone more familiar with this code needs to fix this properly.

(macallan)

Fix an empty shell function here, per Greg Troxel.  Apparently empty shell
functions are not 100% portable.

(garbled)

Bump MSGBUFSIZE on arm ports to 16KB.

OK matt@

(skrll)

Attempt two for the previous: _ILP32 comes from headers, so don't use
it to decide if we want to muddle with _FILE_OFFSET_BITS.

(pooka)

Use _FILE_OFFSET_BITS=64 on Solaris too (necessary for 32bit builds).

(pooka)

Only set wMaxPacketSize to 64 for HS/FS and set to 8 for LS to be within
the USB spec.

Should help PR/46696.

Requesting 64 bytes via wLength is indeed common, but fallback to 8
would probably help.

(skrll)

regen for TOOL_YACC test

(lukem)

No need to look for <tzfile.h>; we don't include it in "tnftpd.h"

(lukem)

revert
  1.14 cmds.c
  1.14 extern.h
  1.16 ftpcmd.y
  1.32 ftpd.c
because it wasn't necessary in tnftpd as these files have been modified
to use "tnftpd.h" instead of the various NetBSD system headers.

(lukem)

sync to:
    othersrc/usr.bin/tnftp/libnetbsd/ftpglob.h 1.6
    othersrc/usr.bin/tnftp/libnetbsd/glob.c 1.10
for:
    replace glob with newer copy that does not suffer from DoS
    exhaustion attacks.
    XXX: Needs more testing.

(lukem)

Actually ensure bison or yacc is present after AC_PROG_YACC,
because the latter defaults to "yacc" if it can't find one...

(lukem)

Remove some unportable assumptions that crept in in rev 1.9:
* Just include "tnftp.h" instead of all the system headers
* Restore a fix for AIX 4.1.5
* Use "unsigned char" instead of "u_char" (reinstate part of rev 1.7)

(lukem)

Remove some unportable assumptions that crept in in rev 1.5.
* Consumers of this header will include "tnftp.h" first which should
  include the appropriate files for the types (size_t, struct stat).
* Always define the extensions; this is a private header included
  as a different name.

(lukem)

Use mkubootimage, if obs600.

(kiyohara)

Make ALG's autoloadable by providing in the config file:
alg "algname"

(christos)

Add a new installer image for Raspberry PI to evbarm

Includes the following major changes:

1) Add new configmenu item to sysinst to allow creation of users at
post-install time.
2) Add an sshramdisk type to evbarm, which is a ramdisk with sshd
enabled, allowing users to install on a headless PI by sshing to the
device with the account "sysinst" password "netbsd" to begin the
installation. (Note, neither the account, nor the ssh setup are copied to
the finalized installation image)
3) Change mkimage to build a boot-only image, without the root filesystems.

Much of the sshramdisk code could be reused on other ports easily.

(garbled)

Need assym.h for libkern

(christos)

To quote the BCM2835 datasheet that's publicly available

"The BCM2835 system uses an AMBA AXI-compatible interface
structure. In order to keep the system complexity low and data
throughput high, the BCM2835 AXI system does not always
return read data in-order."

Deal with this by defining DSB for _ARM_ARCH_6.

OK matt@

(skrll)

rearrange things a little bit to avoid showing garbled VRAM content while
switching colour depth

(macallan)

Fix compile error if BGE_DEBUG is set.

(msaitoh)

Fix argument order for "n14"

(martin)

const-ify.

(isaki)

Remove 'register'.

(isaki)

KNF a bit.

(msaitoh)

- Rename PHY related flags for the consistency. It's the same as FreeBSD.
- Remove BGE_10_100_ONLY flag because this was not used.
  For 10/100 devices, when calling mii_attach(), mask BMSR_EXTSTAT flag to
  not to check Gigabit flags. It's the same as FreeBSD.

(msaitoh)

Fix three bugs:
- An workaround for TX data corruption is only for 5719 "A0".
  Fix the wrong evaluation.
- Check BGE_RXBDFLAG_IPV6 flag for 5717_PLUS case.
  Note that {tcp,udp}6csum flag is currently not added in the capability.
- Add delay after clearing BGE_MACMODE_TBI_SEND_CFGS for the link checking.
  FreeBSD has the same delay().

(msaitoh)

Fix Buffer Manager Mbuf Pool settings which was modified in rev. 1.202.

(msaitoh)

can we please stop breaking the build? These get compiled by the kernel
environment when there are no userland header files!

(christos)

fix some lint on i386, noticed by Greg Troxel, thanks!

(agc)

Terminate result of readlink().  Makes writefilewatch_setup() work in
cases when the buffer doesn't accidentally contain suitable zeroes.

Thanks to Juan RP who debugged this with me!

(pooka)

Mark a few more test cases known to only work by chance as failing
and point to PR 47665.
Julio: how do we make such cases not fail when they actually happen to
work (by pure luck)?

(martin)

C99 section 6.7.2.3 (Tags) Note 3 states that:

  A type specifier of the form

enum identifier

  without an enumerator list shall only appear after the type it
  specifies is complete.

which means that we cannot pass an "enum vtype" argument to
kauth_access_action() without fully specifying the type first.
Unfortunately there is a complicated include file loop which
makes that difficult, so convert this minimal function into a
macro (and capitalize it).

(ok elad@)

(plunky)