Wrap potentially blocking operations to drop and retake the kernel
giant lock.

(pooka)

regen from configure.ac 1.20

(lukem)

Reorder the "Configuration results" after AC_OUTPUT, so we can display
an expanded $prefix.

(lukem)

Rewrite kernel lockus maximus handling to be correct.

(pooka)

regen from configure.ac 1.19

(lukem)

Don't bother trying to determine if the result from AC_PROG_YACC
is usuable; AC_PATH_PROG barfs on absolute paths.
(AC_PROG_YACC shouldn't set $YACC to 'yacc' if it's not available ...)

(lukem)

Update config.guess to 2008-09-28.
Update config.sub  to 2008-09-08.

(lukem)

Don't assume that HAVE_STRUCT_PASSWD_PW_CHANGE means you have _PASSWORD_CHGNOW

(lukem)

Use atomic op to get next xid.  Initialize value with arc4random()
at nfs init time instead system time based trickery intermingled
with the runtime code.

le bouef: kills last simple_lock from nfs

(pooka)

#error if WABPL_DEBUG_INODES is defined.  That code has bitrotted
more than casu marzu cheese.

(pooka)

Move functionality for preparing the kernel (including XMS support) into
a new function common_load_kernel.

(joerg)

Move initialisation of boot_argv after the full loading of the kernel,
directly before loading modules.

(joerg)

Setup module_base in module_init().

(joerg)

tickets 1968, 1973

(bouyer)

Pull up following revision(s) (requested by jmcneill in ticket #1973):
libexec/ld.elf_so/symbol.c: revision 1.47
libexec/ld.elf_so/rtld.h: revision 1.79
libexec/ld.elf_so/rtld.c: revision 1.122
lib/39649: dlsym(3) does not follow dependencies
Pull across code from FreeBSD to do a search of the passed object and it's
NEEDED objects (dependencies).
Reviewed by gimpy.

(bouyer)

Pull up following revision(s) (requested by lukem in ticket #1968):
distrib/sets/lists/base/mi: revisions 1.699, 1.721, 1.725, 1.773
share/zoneinfo/Makefile patch
share/zoneinfo/africa patch
share/zoneinfo/antarctica patch
share/zoneinfo/asia patch
share/zoneinfo/australasia patch
share/zoneinfo/backward patch
share/zoneinfo/europe patch
share/zoneinfo/iso3166.tab patch
share/zoneinfo/leapseconds patch
share/zoneinfo/northamerica patch
share/zoneinfo/southamerica patch
share/zoneinfo/zone.tab patch
Update to tzdata2008f.

(bouyer)

Need to include sys/termios.h here too.

(oster)

memset to zero the entire nop_in structure.  Fixes PR#39688 by
Henning Petersen.  Thanks!

(oster)

return ENOTTY instead of EOPNOTSUPP for ioctl(/dev/null, TIOCGETA)
from Andy Shevchenko

(christos)

check return value of rump_init()

(pooka)

Clarifications for PTHREAD_DIAGASSERT.

(ad)

pthread_diagassert is now off by default.

(ad)

Disable diagnostic assertions by default and just return error codes like
other systems. Allows poorly written applications to appear working. If you
are developing pthread apps please turn it on manually by setting the
environment variable.

(ad)

Fix a fairly obvious pasto which has been there since version 1.1 in 2001.

(agc)

Adjust the compat stuff slightly so that the changes are mostly self
contained (Makefile, pthread_compat.c).

(ad)

WARNS=4

(lukem)

use a const struct hash pointer

(lukem)

Since I nuked the old adt7467c driver, I might as well nuke the man page.

(pgoyette)

check rump_init() return value

(pooka)

check rump_init() return value

(pooka)

Embed library build version into the init call and refuse to init
if they don't match.  In other words, this is a lightweight library
major number.

(pooka)

Embed library build version into the init call and refuse to init
if they don't match.  In other words, this is a lightweight library
major number.

(pooka)

check ukfs_init() return value

(pooka)

Increase accuracy in nominal voltage table.

(pgoyette)

Remove chip-specific drivers for ADT7463, ADT7467, and ADM1030;  these
have been superseded by the dbcool(4) driver.

(pgoyette)

Finish moving pkg_install to src/external/bsd/pkg_install.

(joerg)

Update to libfetch-2.17. Fix a bug in the line buffering code to not
drop output if the server actually send more than one line.

Status:

Vendor Tag: PKGSRC
Release Tags: libfetch-2-17

(joerg)

Remove the definitions of BUILD_OSTYPE and BUILD_OSVERSION, which have
been unused since the TOOL_JOIN change made in revision 1.9.

(apb)

Add a NOTE saying that invalid usage leads to undefined behaviour.
Inspired by PR 39684.

(apb)

Move all platforms to new dbCool driver in preparation for removal of
chip-specific drivers

(pgoyette)

Explicitly pull in "nbtool_config.h" if available.
Fixes tools/db build, as reported by Matthias Scheler.

(lukem)

Add a constty entry, marked 'off', with otherwise the same settings
as the console. Now when coming up in single user mode the tset in
/.profile can set the terminal type correctly.
Note: I believe constty should be 'on' and console 'off', but thats a
separate discussion

(abs)

Most ports do not create /.profile /.termcap and /sysinst in the target
before extracting sets, so do not try to delete them after install.
Now the '/.profile' from etc.tgz will actually be left after install

Note: bebox, evppc, and sandpoint still have a md_copy_filesystem() which
dumps a copy of the install filesystem into the target before extracting,
so I've left them removing /.profile /.termcap and /sysinst after install.

(abs)

- Replace lwp_t::l_sched_info with union: pointer and timeslice.
- Change minimal time-quantum to ~20 ms.
- Thus remove unneeded pool in M2, and unused sched_lwp_exit().
- Do not increase l_slptime twice for SCHED_4BSD (regression fix).

(rmind)

Mention few more changeable SysV IPC parameters, sync with reality.

(rmind)

- Allow changing of kern.ipc.shmmax, just to avoid confusion.
- Change type of kern.ipc.shmmax to CTLTYPE_QUAD.

(rmind)

Add a note that SysV IPC parameters could be changed dynamically.

(rmind)

Fail with a non-zero exit code if deleting a non-existant key.
Fixes bug introduced in rev 1.15, and reported in PR 39710 from Juan RP.

Don't display various warnings in delete if -q.
Suggested in PR 39710.

Add some sanity enforcing abort()s.

(lukem)

nuke outdated comment

(pooka)

fixed botched update in last

(pgoyette)

Don't use emalloc and friends directly, but call them consistently
bmake_malloc and friends. Implement them via macros for the native case
and provide fallback implementations otherwise. Avoid polluting the
namespace by not defining enomem globally. Don't bother to provide
strdup and strndup, they were only used for the estrdup and estrndup
comapt code.

This addresses the presence of emalloc in system libraries on A/UX and
resulted strange issues as reported by Timothy E. Larson.

(joerg)

OOooops, part of this had already been done.  Remove duplicate descriptions
of -1 and -2 options.  Reorder the synopsis line to put the numeric options
before the alphabetic ones.

(pgoyette)

Document the -1 and -2 options, and include this in amd64 as well as i386

Fixes my PR/38816

(pgoyette)

Temporarily disable sending more than one unblocked upcall at once.
Something's wrong in the copy-out, and programs die as a result.

(wrstuden)

Fix the build if stack protection is enabled.

(tron)

tz{code,data}2008g out; tzdata2008g imported.

(kleink)

Import tzdata2008g.

Status:

Vendor Tag: ADO
Release Tags: TZDATA2008G

(kleink)

Return early if size == 0 in ie_gsc_memcopy{in,out}

(skrll)

Regen.

(njoly)

Cleanup. Reduce diffs against netbsd32 syscalls.master.
- Remove const keyword for pointer types.
- Update some argument types.

(njoly)

Disable "-Wstack-protector" in a slightly nicer way.

(tron)

int -> u_int per source

(pooka)

PR kern/39681 ddb doesn't work with new MODULAR modules

Complete initialization of new symbol table record.

(ad)

kobj_read: plug a leak.

(ad)

Conform previous commit to KNF, pointed out by ad@.

(nakayama)

make sure libdixmods (linked into the Xorg server) is aware of MITSHM and
other server extensions.

(rtr)

ehci_free_itd: use LIST_INSERT_HEAD rather than
LIST_INSERT_AFTER(LIST_FIRST(..), ..) when returning itds to the freelist
to prevent a crash when the freelist is empty. Reviewed by Jeremy Morse.

(jmcneill)

Various clean-ups:
  1. Trange sysctl's belong to the temp sensors, not to the fan controllers
  2. Trange really describes a slope on many chips, so modify the description
  3. Most of the sysctl's are read/write even if the chip's config is
    locked;  reflect that in the CTLFLAGs
  4. Apply correct 'nominal' values for voltage sensors/limits, specify
    them in microVolts, and calculate once rather than each time needed
  5. Be more consistent in register names - for example, use VCC instead of
    SUPPLY_VOLTAGE, to match VCC_LOWLIM & VCC_HIGHLIM
  6. Type of dbcool_islocked() should be bool, not int
  7. Reduce some unnecessary code indentation
  8. Define Vtt and Imon, and add ADT7490 support (excluding PECI sensors)
  9. Split the huge (250+ lines) dbcool_setup() function into a few smaller
    routines for better readability
10. Update sensor tables for ADT7476 and ADT7468 - these chips have five
    voltage sensors, not two
11. Adjust flags for ADT7463 and ADM1027 - these chips can monitor CPU
    VID data bits
12. Update man page

(pgoyette)

Make libnfs compile.

(pooka)

Locally descend into new directories (build does not descend here yet).

(pooka)

Implement a virtual userspace networking interface.  This is required
for any sensible operation when running the entire TCP/IP stack in
userspace (as opposed to libsockin which uses host kernel networking
to provide PF_INET).

While this basically works (although it is quite barebones), it
depends on some bit of cleanup in librump and is not built by
default yet.

(pooka)

Provide the TCP/IP as a rump library.

(pooka)

Provide essentially sys/net as a rump library.

(pooka)

move the code that sets ld_vers and other variables so that it is
always executed, regardless of cached values.  this lets the comdat
version test for gas work properly.

(mrg)

fix a 2 year old fix that broke genmultilib.  ${CONFIG_SHELL}
wasn't expanded inside cat >tmpmultilib<<\EOF due to the \.
use two cat's to generate these files now, one with EOF and
one with \EOF.

(mrg)

hardclock() itself doesn't schedule callout softintrs any more,
so update comment - less is more.

(pooka)

Fix issue where multi-transaction isoc endpoints were forced to single
transactions, from Jeremy Morse.

(jmcneill)

Fix a file name who lost its extension.  Noted by mrg@.

(cube)

Remove the definition of HOST_SED.  HOST_SED was previously used only in
tools/atf-compile, which has now been adjusted to use TOOL_SED.

(apb)

Document TOOL_SED, which has been defined in bsd.sys.mk for more than
two years.

(apb)

Use TOOL_SED, not HOST_SED.  Also use && to catch errors in
building atf-host-compile.

(apb)

Add a .WAIT to ensure that sed is installed early, so that TOOL_SED can
be used during the build of other tools.

(apb)

Sync with HEAD.

(mjf)

I am pretty sure we install all available manual pages for bin and lib now.
A few might have slipped under my radar, of course...

(cube)

Install API manual pages for fontcnfig as well as fonts-conf(5).

(cube)

Note installation of fc-cat(1) and fc-match(1) from fontconfig.

(cube)

Descend into fc-cat and fc-match.

(cube)

Add
gnu/usr.bin/groff/tmac/mdoc.local              patch
sys/sys/param.h                                patch

for 4.0.1 RELEASE

(bouyer)

Add CHANGES files for NetBSD 4.0.2

(bouyer)

Properly depend on libexpat.

(cube)

Note installation of xmlwf(1).

(cube)

Install xmlwf(1) from expat.

(cube)

Switch netbsd-4-0 to 4.0.1_PATCH

(bouyer)

Sync with HEAD.

(mjf)

Use date of netbsd-4-0-1-RELEASE tagging, requested by tsutsui in ticket #1206.

(bouyer)

Welcome to NetBSD 4.0.1

(bouyer)

fix bus_spacification botch
from Marco Trillo in PR39696

(macallan)

Note installation of vmmouse(4).

(cube)

I have all respect for users of NetBSD/i368, but I still think vmmouse only
has a chance of being useful on i386 (and arguably on amd64).

(cube)

Install manual pages for libXScrnSaver.

(cube)

Install manual pages for libXau.

(cube)

Install manual pages for libXcomposite.

(cube)

pool_cache: use BLOCK_SIZE as alignment.
pmap_pv_cache: use large size groups (PR_LARGECACHE) like x86.

(nakayama)

Install manual pages for libXevie.

(cube)

- in flushing cache, use 32 as increment/decrement value to match the
  cache line size.
- remove useless restore of %asi.
- remove nop in delay slot if we can.

(nakayama)

struct cpu_info: reorder member to avoid a padding.

(nakayama)

Install manual pages for libXext.

(cube)

Install manual page for libXfixes.

(cube)

Note installation of xft-config.

(cube)

Install manual pages for libXi.

(cube)

Install manual pages for libXinerama.

(cube)

Install manual pages for libXp.

(cube)

Install manual pages for libXres.

(cube)

Avoid referencing something that isn't defined.  Bring in the definition
from another manual page.

(cube)

Add missing splx(). Funny that it didn't cause issues ...

(bouyer)

Install manual pages (the 267 of them) of libXt.

(cube)

Install manual pages for libXtst.

(cube)

Install manual pages for libXv.

(cube)

Install manual pages for libXxf86misc.

(cube)

in cpu_hatch(), set PIR when the current value is not what we need
rather than only when it's zero.

(chs)

Add entries for the 3 recent pullups, fixing security issues.
Also add a note about ftpd PAM fix/reorganisation, from tsutsui@
Fix typo from ticket #1206, from tsutsui@.
reviewed by tsutsui@ and adrianp@

(bouyer)

lib/39649: dlsym(3) does not follow dependencies

Pull across code from FreeBSD to do a search of the passed object and it's
NEEDED objects (dependencies).

Reviewed by gimpy.

(skrll)

In both postinstall and etcupdate, in modes where the -s argument
refers to tgz files or to a directory in which tgz files have already
been extracted, make it an error for the files that should have come
from etc.tgz to be missing.  This is intended to prevent users from
accidentally deleting necessary files when they run "postinstall -s
xetc.tgz fix".

Use the absence of .../etc/mtree/set.etc in the extracted directory
as a test for the error case.

(apb)

Add previously not installed headers.

(cube)

- install missing headers (reported by hubertf@ through proxy jmcneill@)
- clean up creation/installation of freetype-config

(cube)

POOL_INIT -> pool_init

(pooka)

Compat linux/linux32 nice(2) fix. The syscall argument is an increment
to be added to the current nice value, not an absolute value to be
set.

(njoly)

Fix pointer size typo - affects only debug output.

Henning Petersen, PR lib/39689

(pooka)

PR/39687 - Henning Petersen -- comment typo

(jnemeth)

Hallo, pool_init().  Auf wiedersehen & byebye, link set POOL_INIT().

(pooka)

fix example: ONCE_DECL(), not DECL_ONCE()

(pooka)

Replace more intptr_t to uintptr_t when casting VA.

(rmind)

Make sure the adapter is ready to accept FISs before bringing the PHY up.
Wait for the BSY bit to clear in the status register.
Makes the cdrom drive probe properly on a ATI SB600 controller in AHCI
mode.

(bouyer)

Ticket 1967.

(jdc)

Ticket 1967.

(jdc)

Ticket 1967.

(jdc)

Pull up revisions:
  src/sys/netinet6/in6.c 1.141 via patch
  src/sys/netinet6/in6_var.h 1.59 via patch
  src/sys/netinet6/nd6_nbr.c 1.89-1.90 via patch
(requested by adrianp in ticket #1967).

If a neighbor solictation isn't from the unspecified address, make sure
that the source address matches one of the interfaces address prefixes.

Generalize previous fix so that both NS and NA packets are checked.

(jdc)

Pull up revisions:
  src/sys/netinet6/in6.c 1.141 via patch
  src/sys/netinet6/in6_var.h 1.59 via patch
  src/sys/netinet6/nd6_nbr.c 1.89-1.90 via patch
(requested by adrianp in ticket #1967).

If a neighbor solictation isn't from the unspecified address, make sure
that the source address matches one of the interfaces address prefixes.

Generalize previous fix so that both NS and NA packets are checked.

(jdc)

Pull up revisions:
  src/sys/netinet6/in6.c 1.141 via patch
  src/sys/netinet6/in6_var.h 1.59 via patch
  src/sys/netinet6/nd6_nbr.c 1.89-1.90 via patch
(requested by adrianp in ticket #1967).

If a neighbor solictation isn't from the unspecified address, make sure
that the source address matches one of the interfaces address prefixes.

Generalize previous fix so that both NS and NA packets are checked.

(jdc)

Correct the ticket 1210 entry.

(jdc)

Correct the ticket 1210 entry.

(jdc)

Ticket 1210.

(jdc)

Ticket 1210.

(jdc)

Pull up revisions:
  src/sys/netinet6/in6.c 1.141 via patch
  src/sys/netinet6/in6_var.h 1.59 via patch
  src/sys/netinet6/nd6_nbr.c 1.89-1.90 via patch
(requested by adrianp in ticket #1210).

If a neighbor solictation isn't from the unspecified address, make sure
that the source address matches one of the interfaces address prefixes.

Generalize previous fix so that both NS and NA packets are checked.

(jdc)

Pull up revisions:
  src/sys/netinet6/in6.c 1.141 via patch
  src/sys/netinet6/in6_var.h 1.59 via patch
  src/sys/netinet6/nd6_nbr.c 1.89-1.90 via patch
(requested by adrianp in ticket #1210).

If a neighbor solictation isn't from the unspecified address, make sure
that the source address matches one of the interfaces address prefixes.

Generalize previous fix so that both NS and NA packets are checked.

(jdc)

Ticket 1966.

(jdc)

Ticket 1966.

(jdc)

Ticket 1966.

(jdc)

Pull up revision 1.150 (requested by adrianp in ticket #1966).

Fix for CVE-2008-3530 from matt@
Implement improved checking for MTU values on ICMP 'Packet Too Big Messages'

(jdc)

Pull up revision 1.150 (requested by adrianp in ticket #1966).

Fix for CVE-2008-3530 from matt@
Implement improved checking for MTU values on ICMP 'Packet Too Big Messages'

(jdc)

Pull up revision 1.150 (requested by adrianp in ticket #1966).

Fix for CVE-2008-3530 from matt@
Implement improved checking for MTU values on ICMP 'Packet Too Big Messages'

(jdc)

Ticket 1209.

(jdc)

Tickets 1208 and 1209.

(jdc)

Pull up revisions:
  src/sys/dev/ic/aac.c 1.41 via patch
  src/sys/dev/ic/aacreg.h 1.13 via patch
  src/sys/dev/ic/aacvar.h 1.13 via patch
  src/sys/dev/ic/ld_aac.c 1.22 via patch
(requested by sborrill in ticket #1208).

(jdc)

Pull up revision 1.150 (requested by adrianp in ticket #1209).

Fix for CVE-2008-3530 from matt@
Implement improved checking for MTU values on ICMP 'Packet Too Big Messages'

(jdc)

Pull up revision 1.150 (requested by adrianp in ticket #1209).

Fix for CVE-2008-3530 from matt@
Implement improved checking for MTU values on ICMP 'Packet Too Big Messages'

(jdc)

Fix for CVE-2008-3530 from matt@
Implement improved checking for MTU values on ICMP 'Packet Too Big Messages'

(adrianp)