Pull up following revision(s) (requested by dholland in ticket #194):
usr.bin/uudecode/uudecode.c: revision 1.24
Sync usage message with man page. PR 40063.

(snj)

Fix handling of ppp compressor modules, from Andrew Doran's input.
- ref count each compressor
- allow {un,}registration of several modules at once
- une RUN_ONCE to make sure the mutex is initialised, because
  unfortunately built-in (and bootloader-loaded) modules init functions
  are run before pseudo-devices attach (reported by Nick Hudson).

(cube)

Pull up following revision(s) (requested by dholland in ticket #149):
usr.bin/uuencode/uuencode.c: revision 1.14
usr.bin/uuencode/uuencode.1: revision 1.17
Clarify man page with respect to input and output files, and sync usage
message. Prompted by PR 40063 about the usage message.

(snj)

Pull up following revision(s) (requested by mrg in ticket #148):
external/mit/xorg/server/xorg-server/hw/xfree86/utils/scanpci/Makefile: revision 1.4
add -Wl,-rpath,${X11USRLIBDIR}/modules to LDADD, so that scanpci can
find libscanpci.so.0.  fixes PR#40050.

(snj)

Pull up following revision(s) (requested by mrg in ticket #147):
sys/ufs/ext2fs/ext2fs_alloc.c: revision 1.37
sys/ufs/ext2fs/ext2fs_bswap.c: revision 1.14
sys/ufs/ext2fs/ext2fs_dinode.h: revision 1.17
sys/ufs/ext2fs/ext2fs_lookup.c: revision 1.56
sys/ufs/ext2fs/ext2fs_vnops.c: revision 1.83
sys/ufs/ext2fs/ext2fs_vfsops.c: revision 1.140
sys/ufs/ufs/inode.h: revision 1.55
add support for 32 bit uid/gid fields in ext2, but only do so for
when the revision is > REV0.

(snj)

Modernize and expand example. Bump date.

(dholland)

Sync usage message with man page. PR 40063.

(dholland)

Clarify man page with respect to input and output files, and sync usage
message. Prompted by PR 40063 about the usage message.

(dholland)

Tickets 137-146

(bouyer)

Pull up following revision(s) (requested by cegger in ticket #146):
lib/libkvm/kvm_proc.c: revision 1.79
check error case for kvm_getlwps or SIGSEGV occurs due to dereferecing a
NULL pointer.
return NULL if there are no threads or proceed further through all loops.
Fixes PR 40004. Ok rmind@

(bouyer)

Pull up following revision(s) (requested by reed in ticket #145):
usr.bin/rev/rev.1: revision 1.7
May have multiple files on command line.
Fixed PR #40058.

(bouyer)

Pull up following revision(s) (requested by jmcneill in ticket #144):
sys/dev/usb/uvideo.c: revision 1.23
Don't select an endpoint with a larger packet size than the maximum for
the current video stream.

(bouyer)

Pull up following revision(s) (requested by joerg in ticket #143):
sys/dev/pci/if_vge.c: revision 1.43
- ifconfig down support
- ALTQ support

(bouyer)

Pull up following revision(s) (requested by martin in ticket #142):
external/mit/libX11/dist/src/xlibi18n/XlcDL.c: revision 1.2
Do not special case locale names for sparc _LP64 machines - this probably
is a Solaris'm not needed on NetBSD. Fixes PR xsrc/40046.

(bouyer)

Pull up following revision(s) (requested by sborrill in ticket #141):
distrib/utils/libhack/getpwent.c: revision 1.10
Add cut-down getpwent_r() to join existing getpwent() and friends. This
fixes PR 40003

(bouyer)

Pull up following revision(s) (requested by elad in ticket #140):
sys/kern/init_sysctl.c: revision 1.151
PR/40002: Daniel Horecki: sockstat doesn't work for user with sysctl
    security.curtain=1
If the kauth call failed, we'd silently continue the loop, but the error
code would remain and eventually "leak" to userspace. Reset the error to
zero when continuing.
Tested by snj@ and myself. Okay snj@.

(bouyer)

Pull up following revision(s) (requested by jmorse in ticket #139):
sys/dev/usb/uvideo.c: revision 1.22
Modified uvideo_stream_init_desc to correctly calculate max packet size
for high speed endpoints.

(bouyer)

Pull up following revision(s) (requested by jmorse in ticket #138):
sys/dev/usb/ehcireg.h: revision 1.29
sys/dev/usb/ehci.c: revision 1.155
Fixed high-speed isoc USB transfers, 64 bit capable host controllers
have a different transfer descriptor to non-capable controllers.
Added dma-sync of data buffer before scheduling transfers

(bouyer)

Pull up following revision(s) (requested by joerg in ticket #137):
etc/MAKEDEV.tmpl: revision 1.118
Change the formula for tmpfs size. The per-inode memory usage is
higher than old estimate on 64bit platforms and the resulting tmpfs
wouldn't be able to hold the required number of devices.

(bouyer)

- don't build compat_linux on alpha until sys/modules/compat_linux/Makefile
  handles required MD files under sys/compat/linux/arch/alpha
- build exec_elf64 for all 64 bit arch
- use MACHINE_ARCH!="alpha" instead of MACHINE!="alpha" for exec_elf32 check

XXX we have many histrical inconsistent use of amd64 vs x86_64.

(tsutsui)

check error case for kvm_getlwps or SIGSEGV occurs due to dereferecing a NULL pointer.
return NULL if there are no threads or proceed further through all loops.
Fixes PR 40004. Ok rmind@

(cegger)

Reorganize lists for MD modules:
- move MD module files from md.* into separate module.* lists
- make those module lists per MACHINE_CPU or MACHINE_ARCH,
  rather than per MACHINE

Tested on several ports.

XXX maybe we should also move usr/tests/modules/k_helper/k_helper.kmod
XXX from lists/tests/mi to lists/tests/module.mi and disable it on evbppc.

(tsutsui)

Split For_Eval() into two functions.
For_Eval() is now only called for the first line of a .for.
For_Accum() is called for the subsequent lines.
Stops any problems with forLevel being left invalid after an error.
Use a return value of -1 from For_Eval() to mean 'skip input line' to stop
a .for line with a syntax error being reparsed by make.

(dsl)

Add an entry for workaround in src/sbin/ifconfig/Makefile
for gcc -O1 loop optimize bug.

(tsutsui)

Specify -fno-loop-optimize on m68000 as workaround for gcc -O1 codegen bug
mentioned in PR bin/40036 and PR toolchain/40066.

(tsutsui)

Add RCS Id.  XXX why separate Makefile.inc?

(tsutsui)

Before deciding to use the default function in a .if, check whether the
token that follows the argument might be '==' or '!='.
If so then treat as a string comparison instead.
Fixes bin/15233 and bin/30967 provided some whitespace is present.
".if A==A" remains a check for defined(A==A) since make places no
restrictions on the names of variables!

(dsl)

Common up parsing .if defined() make() exists() target() commands() and empty()
Add a check for the '(' following the function name, if absent then treat
as if the function name is unknown - usually leading to a syntax error.
No other functional changes intended.

(dsl)

Hand over copyright to TNF

(hubertf)

Rename variables to avoid confusing.  forgotten in previous commit.

(isaki)

Split device_t/softc.  Use aprint*() and device_xname().
Not tested. I cannot test now...

(isaki)

PR/40060 - Bug Hunting -- grammar fix

(jnemeth)

Split device_t/softc.  Use aprint*() and device_xname().
Not tested. I don't have its hardware and no response in port-x68k.

(isaki)

PR/40059 - Bug Hunting -- add Xref to wscons.conf(5)

(jnemeth)

Note that re(4), ath(4), and puc(4) are known to work on sparc64.

(snj)

Don't select an endpoint with a larger packet size than the maximum for
the current video stream.

(jmcneill)

Do not special case locale names for sparc _LP64 machines - this probably
is a Solaris'm not needed on NetBSD. Fixes PR xsrc/40046.

(martin)

Remove empty line. Fix typo. New sentence, new line.

(wiz)

Sort errors.

(wiz)

Fix typo, remove .Pp before .Sh (unneeded).

(wiz)

add -Wl,-rpath,${X11USRLIBDIR}/modules to LDADD, so that scanpci can
find libscanpci.so.0.  fixes PR#40050.

(mrg)

Build compat_linux and exec_elf64 for alpha.

(he)

Ticket 1977

(bouyer)

pullup the following revisions (requested by lukem in ticket #1977):
share/zoneinfo/africa TZDATA2008I
share/zoneinfo/asia TZDATA2008I
share/zoneinfo/southamerica TZDATA2008I
share/zoneinfo/zone.tab TZDATA2008I
distrib/sets/lists/mi 1.778

update zoneinfo to tzdata2008i

(bouyer)

Ticket 1977

(bouyer)

pullup the following revisions (requested by lukem in ticket #1977):
share/zoneinfo/africa TZDATA2008I
share/zoneinfo/asia TZDATA2008I
share/zoneinfo/southamerica TZDATA2008I
share/zoneinfo/zone.tab TZDATA2008I
distrib/sets/lists/mi 1.778

update zoneinfo to tzdata2008i

(bouyer)

handle old IFDATAREQ; this makes ifconfig work and now my machine
comes up multi-user with new libc/new kernel old binaries.

(christos)

May have multiple files on command line.
Fixed PR #40058.

(reed)

Add cut-down getpwent_r() to join existing getpwent() and friends. This
fixes PR 40003

(sborrill)

PR/40002: Daniel Horecki: sockstat doesn't work for user with sysctl
    security.curtain=1

If the kauth call failed, we'd silently continue the loop, but the error
code would remain and eventually "leak" to userspace. Reset the error to
zero when continuing.

Tested by snj@ and myself. Okay snj@.

(elad)

#undef ACTIVE (not ENTER) since that is the variable we defined just above.
Noted on port-i386 by Gao Ya'nan.

(dsl)

Modified uvideo_stream_init_desc to correctly calculate max packet size for high speed endpoints.

(jmorse)

Fixed high-speed isoc USB transfers, 64 bit capable host controllers have a different transfer descriptor to non-capable controllers.

Added dma-sync of data buffer before scheduling transfers

(jmorse)

Change the formula for tmpfs size. The per-inode memory usage is
higher than old estimate on 64bit platforms and the resulting tmpfs
wouldn't be able to hold the required number of devices.

(joerg)

Allow for the exclusive selection of `direct' for rdonly mounts; its disabled
for now since its a lot slower than `rmw' access.

For archs that have trouble with `rmw' for whatever reason can so use it as a
scapegoat to allways mount savely rdonly though slower.

(reinoud)

alpha doesn't have modules for compat_linux and exec_elf64 (yet?).

(tsutsui)

Remove old comment and remove unused calculated variable

(reinoud)

- use Em for HZ
- add xrefs to hz(9) and time_second(9)

(tsutsui)

- reorder description per meanings
- note header files in SYNOPSIS
- remove obsolete internal description about hardclock(9)
- note that microtime(9) and microuptime(9) family functions are prefered
  for atomic operation (from FreeBSD's time(9))
- add more related Xrefs

(tsutsui)

Tyop: v -> vp.  Spotted by Tom's ARM builds.

(pooka)

Sync with microuptime(9):
- improve punctuation
- add explicit xref to hz(9)
- use Em for HZ

(tsutsui)

exec_add, exec_remove: allow zero entries in case a module provides nothing.

(ad)

- use Em for HZ
- also add hz(9) to SEE ALSO

(tsutsui)

fix include filename

(martin)

make the synopsis slightly less overpromising:
"attempt to recover a deleted file" => "remove whiteout"

(pooka)

Copy the sigdebug and sigpid declarations and associated defines
for the DEBUG define from machdep.c, so that this builds again.

(he)

We also need a 32-bit librumpvfs when building on 64-bit arches.
Fix by pooka.

(dogcow)

We now also need globals.c from the common libsa.

(he)

Add wscons/vcons support for cg3.

(jdc)

Use kmem instead of malloc to avoid hassle with dynamically attaching
a malloc type.  Makes nfsserver-as-a-module work.

reported on current-users & tested by Geoff Wing

(pooka)

Clarify that the "amaster" argument of forkpty() is only defined upon
return in the parent process, but the tty name is available to both
the parent and child. Bump date.

(dholland)

g/c unused malloc types

(pooka)

Ticket 136.

(snj)

Pull up following revision(s) (requested by christos in ticket #136):
dist/nawk/b.c: revision 1.17
dist/nawk/proto.h: revision 1.7
don't use signed chars for no reason.

(snj)

Pull up following revision(s) (requested by christos in ticket #136):
tools/awk/Makefile: revision 1.2
enable extensions for isblank() and solaris

(snj)

Tickets 124[12].

(snj)

Pull up following revision(s) (requested by dholland in ticket #1242):
usr.bin/who/who.c: revision 1.23
PR/39201: Brian Marcotte: "who am i" problem when using ptyfs, because it
strips pts/

(snj)

Pull up following revision(s) (requested by martin in ticket #1241):
lib/libc/arch/sparc64/sys/__sigaction14_sigtramp.c: revision 1.8
Remove __LIBC12_SOURCE__

(snj)

Add missing fflush() in one of three places that print stuff repeatedly
every so often. PR 32836 from Takuro KUBOTA.

(dholland)

Ticket 135.

(snj)

Pull up following revision(s) (requested by christos in ticket #135):
usr.bin/make/make.1: revision 1.146
I think ':' is called "colon", not "column" :)

(snj)

Pull up following revision(s) (requested by christos in ticket #135):
usr.bin/make/make.1: revision 1.145
PR/40026: Aleksej Saushev: make(1) doesn't document VPATH, document it.

(snj)

More work in progress.

Bring across elroy, astro and uturn from OpenBSD.

Update device scanning to use the OpenBSD method for versions of PDC that
don't report devices correctly.

(skrll)

better error messages, and more messages for DIAGNOSTIC.

(christos)

delete unused variable (and don't use it!)

(christos)

add more 50 compat syscalls.

(christos)

handle ifdatareq

(christos)

PR/40043: Geoff C. Wing: syslogd timestamping and protocol parsing deficiencies
Syslogd does not properly handle:

1) the ADDDATE flag which is set with -T invocation and when messages
  come from the kernel.  Other cases where it is set it is ignored
  as timestamping is always done (e.g. logmsg_async())

2) the variable found_ts in check_timestamp().  It would determine
  whether or not the message had a (possibly valid) timestamp, set
  found_ts to true, then ignore that in most cases.  If we can't find
  a timestamp return.

3) messages without a parsable timestamp should get one when outputting
  the BSD syslog format so that a syslog-protocol timestamp isn't
  injected (chopped off with BSD syslog length) giving something like:
    "2008-11-27T15:0 cisco -: 1790:"
    ^ time might have been 2008-11-27T15:02:35.296497+11:00

4) syslog protocol version checking only checked for a leading numeral
  one (1) then skipped two places (presuming a space).  Messages sent
  from some sources (e.g. my cisco) may be
    "1795: Nov 27 04:12:52: %LINEPROTO-5-..."
  which would be chopped to
      "95: Nov 27 04:12:52: %LINEPROTO-5-..."

(christos)

Tickets 132-134.

(snj)

Pull up following revision(s) (requested by reinoud in ticket #133):
sys/fs/udf/udf_strat_rmw.c: revision 1.12
Specify blobsize as the `pagingsize' of the eccline blob pool. This will relax
the requirements of the pagingsize.

(snj)

Pull up following revision(s) (requested by reinoud in ticket #132):
usr.sbin/mmcformat/mmcformat.8: revision 1.4
Fix typos in the manpage. Thanks for snj for the patch.

(snj)

Pull up following revision(s) (requested by reinoud in ticket #134):
usr.sbin/mmcformat/mmcformat.8: revision 1.3
Sort sections, use more markup, sort options, spelling fixes, improve
formatting.

(snj)

* hang all mountpoints from rootvnode in our virtual mount hierarchy
* set VV_ROOT for the rootvnode to make getcwd realize it has met
  its journey's end

(pooka)

LIBDPLIBS on librumpvfs

(pooka)

Remove various COMPAT_ options for unsupported releases and optional
features; remove parallel printer driver.
Bring back pppoe support.

(hauke)

ModeConfig fixes

(vanhu)

Set up a default value for Mode Config Pool size if pool address specified but pool size not specified

(vanhu)

Fixed pool resizing

(vanhu)

ModeConfig fixes

(vanhu)

Set up a default value for Mode Config Pool size if pool address specified but pool size not specified

(vanhu)

Fixed pool resizing

(vanhu)

Print the trapframe pointer in the trace output.

(skrll)

Add nfs server rump library.
(not descended into on purpose)

(pooka)

Use struct nfs_fattr in struct flrep instead of uint32_t array
acrobatics to get rid of type punning warning.

(pooka)

Link in a networking domain, required by nfs.

(pooka)

From Arnaud Ebalard:
Remove MAXNESTEDSA weirdness. It's probably meant for bundle support which
is not done. When someone actually writes bundle support, the nested SA
stuff would probably be reworked too anyway.

(tteras)

From: Matthew Krenzer
Ability to set pfkey socket buffer size via configuration file directive.
(Indentation and minor fixes by me.)

(tteras)

No need to dig into mips/conf since mips/conf/Makefile has been removed.

(nakayama)

g/c #if 0'd ubc_flush()

(pooka)

Move more vfs-related vm routines from rumpkern to rumpvfs.

(pooka)

Remove the whole magic ubc window thingie.  All file systems use
ubc_uiomove() now, so we can hook ourselves there.

(pooka)

Tickets 128-131.

(snj)

Pull up following revision(s) (requested by nakayama in ticket #131):
sys/arch/sparc64/sparc64/db_disasm.c: revision 1.15
sys/arch/sparc64/sparc64/db_trace.c: revision 1.41
sys/arch/sparc64/sparc64/db_interface.c: revision 1.113
sys/ddb/db_command.c: revision 1.124
sys/arch/sparc64/include/db_machdep.h: revision 1.26
Make DDB shows right register values on 32-bit sparc64 kernels.
- define db_expr_t as int64_t.
- %gsr is 32-bit, so use db_sparc_intop.
- sync arguments with function prototype.
- ansify.

(snj)

Pull up following revision(s) (requested by nakayama in ticket #130):
usr.sbin/postinstall/postinstall: revision 1.78
Add scan obsolete minor shared libraries in /usr/X11R7/lib.
Also scan in /usr/lib/i386 for amd64, /usr/lib/sparc for sparc64.

(snj)

Pull up following revision(s) (requested by bouyer in ticket #129):
sys/arch/i386/isa/npx.c: revision 1.134
npxdna(): even if there's no effective FPU context switch set ci_fpused;
FPU will be used and needs to be switched on next context switch.
Fix "panic: npxintr: wrong process" with Xen.

(snj)

Pull up following revision(s) (requested by bouyer in ticket #128):
sys/arch/i386/isa/npx.c: revision 1.133
npxsave_cpu() set ci_fpused so that the FPU will be disabled on next
context switch on Xen. Fix "npxintr: came from nowhere" kernel messages on
Xen/i386.

(snj)

Another assym.h dependency.

(joerg)

- ifconfig down support
- ALTQ support

(joerg)

Fix typos in the manpage. Thanks for snj for the patch.

(reinoud)

Rototill all remaining file systems to use ubc_uiomove() instead
of the ubc_alloc() - uiomove() - ubc_release() dance.

(pooka)

Dynamically load ext2 support, since the linker can't handle more
that one file system module (ext2fs uses both ffs and itself).

(pooka)

I think ':' is called "colon", not "column" :)

(wiz)

Fix two mdoclint warnings.

(wiz)

Improve grammar.

(wiz)

New sentence, new line. Use more markup. The character is nul with one l.

(wiz)

Fix part of previous, and add link to hz(9) so it will
not happen again.

(wiz)

Casing and punctuation improvements.

(wiz)

WARNS=4

(ginsbach)

Add a man page.

(ginsbach)

- rename set list lkm.mi -> module.mi
- move obsolete lists of usr/lkm files to mi

(tsutsui)

Add a few symbols required by nfsd

(pooka)

new sentence, new line

(ginsbach)

regen

(pooka)

RUMP some syscalls required by nfsd.

(pooka)

puffs_setfhsize() in init

(pooka)

support:
a) readdir cookies and eofflag
b) file handle routines.

  File handle routines are REALLY ugly and are not stable, i.e.
  the file handle will be lost if the file server goes byebye.
  They probably should be redone to not use getfh(), but for nullfs
  I wanted to make them the same as when the file system is accessed
  without the null layer.  Well, turns out it doesn't pass anyway
  since the kernel adds the fsid tags for fhandle_t.  Anyway, it's
  useful for some limited hacking purposes, so I'll commit this
  since it works .. kinda.

(pooka)

Add a section which describes struct bintime.

(tsutsui)

Remove dup entries.

(tsutsui)

Adapt to the removal of exec_elf32 from the alpha port, necessitating
the module be added to all the other ports' md.* files.
Also add the compat_linux module for alpha.

(he)

Add the recently added filecorefs, msdosfs, nullfs, and umapfs modules.

(he)

Try to honor rump_threads by not creating iothreads if threads are off.

(pooka)

Add dummy pr_ctloutput.  I'm sure I had one at some point, since
it's required for rump_nfs to work, but I guess I lost it somewhere.
(yes, it should actually be implemented also)

(pooka)

gpn, not gnp.

(matt)

Add children of geminiipi

(matt)

Use binuptime(9) and emulate 1GiHz (2^30) counter rather than
nanouptime(9) and 1GHz counter to avoid extra rescaling.
Suggested and reviewed by kardel@.

(tsutsui)

Document my changes from 2008-11-08 and today.

(bad)

Avoid using MSG_NOSIGNAL as it is not available everywhere.
Ignore SIGPIPE instead.

(bad)

u_int32_t -> uint32_t

(matt)

Ignore unspecified and looback addresses.  Ignoring unspecified addresses
prevents racoon from trying to bind to the wildcard address and specific
addresses simultaneously after e.g. dhclient has changed an interface's
address to 0.0.0.0.

(bad)

RTM_DELETE and RTM_IFINFO don't carry info for added or deleted addresses.
Ignore them silently.

(bad)

npxdna(): even if there's no effective FPU context switch set ci_fpused;
FPU will be used and needs to be switched on next context switch.
Fix "panic: npxintr: wrong process" with Xen.

(bouyer)

Ignoring an unsuitable address is not an error.  Therefore log it as
informational.
Make it clear from the log message that a route message is not interesting.

(bad)

Use insmyaddr() instead of open coding it.

(bad)

Do not return erroneously from isakmp_open() when setting IPV6_USE_MIN_MTU
fails.

(bad)

Keep myaddr.sock at -1 when no socket is opened.

(bad)

Add dev/crmfbreg.h.

(he)

Make sockin radix-capable.

(pooka)

Support PRU_BIND / PRU_LISTEN / PRU_ACCEPT in sockin.

(pooka)

Take reference for proc0 rootdir vnode.

(pooka)

PR/40026: Aleksej Saushev: make(1) doesn't document VPATH, document it.

(christos)

"Manually" call rn_init() to make sure it's done.  The real kernel
hooks it up with domain constructors in a fascinating way.

(pooka)

enable extensions for isblank() and solaris

(christos)

don't use signed chars for no reason.

(christos)

fix comment (no functional change)

(pooka)

Heave-ho radix.c from librumpnet_net to librumpnet.

(pooka)

npxsave_cpu() set ci_fpused so that the FPU will be disabled on next context
switch on Xen. Fix "npxintr: came from nowhere" kernel messages on Xen/i386.

(bouyer)

Make dom_maxrtkey of inet/inet6domain the size of the ip_encap pack
structures.  This is far from optimal, but gets rid of iffy
#ifdef INET in radix.c.  The radix bonsai still needs lots of love
before loading domains dynamically is possible...

(pooka)

Tickets 121-127.

(snj)

Pull up following revision(s) (requested by ad in ticket #127):
sys/arch/amd64/amd64/fpu.c: revision 1.30
fpudna: enable interrupts in an error path.

(snj)

Pull up following revision(s) (requested by ad in ticket #126):
sys/arch/amd64/amd64/vector.S: revision 1.29
trap07: do NOT enable interrupts before calling fpudna or we could be
preempted.

(snj)

Pull up following revision(s) (requested by joerg in ticket #125):
sys/arch/x86/acpi/acpi_wakeup.c: revision 1.11
sys/dev/acpi/acpi_pci_link.c: revision 1.14
sys/dev/acpi/acpivar.h: revision 1.34
On resum-from-RAM explicitly restore PCI link device state before
reenabling interrupts. At least one BIOS doesn't do this automatically
as reported by Christoph Egger.

(snj)

Pull up following revision(s) (requested by martin in ticket #124):
lib/libc/arch/sparc64/sys/__sigaction14_sigtramp.c: revision 1.8
Remove __LIBC12_SOURCE__

(snj)

Regen for ticket 123.

(snj)