Prevent amd64 compiler flag -mcmodel=kernel for rump kernel code.
This allows to compile rump libraries as shared libraries also on mad64.

Problem spotted by Greg Oster.  I guess his build broke or something ...

(pooka)

Make sure to call dlerror() to clear any error messages after the relocations
are done and there are no errors.  This will prevent spurious undefined weak
symbol errors being returned by dlerror().

Fixes PR bin/39240.

(matt)

rw_vector_exit(): When doing direct hand-off to a writer, only set
RW_WRITE_WANTED if there are additional writers waiting.

(thorpej)

Use command line option parsing code from sbin/mount_msdos to
gain equal usage and behaviour with the "real" counterpart.

(pooka)

Split option parsing into its own routine so that it can be used
by other consumers as well.

(pooka)

Add recently added subdirs under /usr/X11R6.

(he)

document workaround for PR#39192

(lukem)

Disable assembler warnings on mipseb and mipsel for mcount.c
Workaround for PR#39192

(lukem)

* remove unnecessary .PATH
* wrap malloc

(pooka)

NOMAN

(pooka)

Accidentally left behind: these live in src/lib now.

(pooka)

install/39193, fix signedness problem of converting char -> wchar_t.

(tnozaki)

Install rump libraries and utilities to the base system and remove the
private non-installed build infrastructure from sys/rump.

breakdown of commit:
  * install relevant headers into /usr/include/rump
  * build sys/rump/librump/rumpuser and sys/rump/librump/rumpkern
    from src/lib and install as librumpuser and librump, respectively
    + this retains the ability to test a librump build with just the
      kernel sources at hand
  * move sys/rump/fs/lib/libukfs and sys/rump/fs/lib/libp2k to src/lib
    for general consumption, they are not kernel-space dwellers anyway
  * build and install sys/rump/fs/lib/lib$fs as librumpfs_$fs
  * add chapter 3 manual pages for rump, rumpuser, ukfs and p2k
  * build and install userspace kernel file system daemons if MKPUFFS=yes
    is spexified
  * retire fsconsole for now, it will make a comeback with an actually
    implemented version shortly

(pooka)

Add missing file.

(rjs)

fix typo.

(christos)

Remove comments on `need to read in'; they are read in.

(reinoud)

Solve the fstat-wants-to-look-at-kernel-data-structures in a nicer
way: don't export the fs internals to innocent userspace programs
which just want to mount the file system.

(pooka)

postfix-2.5.3 out.

(wiz)

initial import of fontcacheproto-0.1.2

Status:

Vendor Tag: xorg
Release Tags: fontcacheproto-0-1-2

(mrg)

initial import of fixesproto-4.0

Status:

Vendor Tag: xorg
Release Tags: fixesproto-4-0

(mrg)

Sync with head.

(simonb)

initial import of expat-2.0.1

Status:

Vendor Tag: xorg
Release Tags: expat-2-0-1

(mrg)

initial import of evieext-1.0.2

Status:

Vendor Tag: xorg
Release Tags: evieext-1-0-2

(mrg)

initial import of encodings-1.0.2

Status:

Vendor Tag: xorg
Release Tags: encodings-1-0-2

(mrg)

initial import of editres-1.0.3

Status:

Vendor Tag: xorg
Release Tags: editres-1-0-3

(mrg)

initial import of damageproto-1.1.0

Status:

Vendor Tag: xorg
Release Tags: damageproto-1-1-0

(mrg)

initial import of compositeproto-0.4

Status:

Vendor Tag: xorg
Release Tags: compositeproto-0-4

(mrg)

initial import of bitmap-1.0.3

Status:

Vendor Tag: xorg
Release Tags: bitmap-1-0-3

(mrg)

initial import of beforelight-1.0.3

Status:

Vendor Tag: xorg
Release Tags: beforelight-1-0-3

(mrg)

initial import of bigreqsproto-1.0.2

Status:

Vendor Tag: xorg
Release Tags: bigreqsproto-1-0-2

(mrg)

initial import of beforelight-1.0.3

Status:

Vendor Tag: xorg
Release Tags: beforelight-1-0-3

(mrg)

initial import of appres-1.0.1

Status:

Vendor Tag: xorg
Release Tags: appres-1-0-1

(mrg)

initial import of appres-1.0.1

Status:

Vendor Tag: xorg
Release Tags: bdftopcf-1-0-1

(mrg)

initial import of appres-1.0.1

Status:

Vendor Tag: xorg
Release Tags: appres-1-0-3

(mrg)

initial import of MesaLib-7.0.3

Status:

Vendor Tag: xorg
Release Tags: MesaLib-7-0-3

(mrg)

initial import of MesaDemos-7.0.3

Status:

Vendor Tag: xorg
Release Tags: MesaDemos-7-0-3

(mrg)

initial import of the crossbuild framework for modular Xorg.  this
is heavily based upon the work rtr@ did on src/x11, but hacked to
work for modern modular Xorg source split.

the sources will live in xsrc/external/mit/<package>/dist.

documentation on using this will come when the above directories
are fully populated.  there are 192 of them so it may take some time
before that is completed.

Status:

Vendor Tag: mrg-and-rtr
Release Tags: oinkoink

(mrg)

add "mit" entry

(mrg)

add several new subdirs to /usr/X11R6 to cope with native Xorg.

XXX: should native Xorg use /usr/X11R6?  input required.

(mrg)

if MKXORG != no, use external/mit/xorg subdir instead of x11 subdir for
the mk-x11 target.

(mrg)

begin to add support for cross-built Xorg:

- add MKXORG
- default to DRI-on for Xorg
- add paths for 192 Xorg packages that will exist in
  xsrc/external/mit/<package>/dist
- add some more X11FLAGS.* variables if MKXORG.

this is somewhat based on the xorg work rtr@ did a couple of years ago.

XXX: seriously need to fix the XORG_RELEASE value, it is still set to
XXX: 6.8.1 from when rtr was working on this.

(mrg)

Make uvm_map.? use <sys/rb.h> instead of <sys/tree.h>.  Change the
ambiguous members ownspace/space to gap/maxgap.  Add some evcnt for
evaluation of lookups using tree/list.  Drop threshold of using
tree for lookups from > 30 to > 15.

Bump kernel version to 4.99.71

(matt)

Repository : haad@cvs.netbsd.org:/cvsroot
Module    : src/sys/dev/dm
Working dir: /usr/src/sys/dev/dm/
Tag        : haad-dm

In directory .:
              Modified                device-mapper.c
              Modified                dm.h
              Modified                dm_ioctl.c

--------------------- Empty ---------------------
-- last cmd: cvs -f -n update -d -P --

(haad)

Add status function to targets. This function is called from dm_table_status_ioctl,
when DM_STATUS_TABLE_FLAG is specified and it's purpose is construct params string.
Which can be sent back to libdevmapper.

Add adding removing of the DM_SUSPENDED_FLAG flag in dm_dev_suspend/resume ioctls.

(haad)

add also the dir /usr/include/fs/tmpfs

pointed out by Kurt Schreiner

(pooka)

Streamline allocation to prepare it for metadata partition. Also add
preliminary Metadata partition write support but its disabled still since
its not finished yet and not functioning correctly. All other formats are
checked and should work fine.

(reinoud)

Reacquire softnet_lock after calling soabort which returns with the socket
unlocked.

(matt)

Backup some manual page sketches lest I delete the wrong source tree.

(pooka)

rump now builds on all ports

(pooka)

Give rump a faux vmparam.h to work around problematic ports (hooray
for evbppc).  Obligatory inlined rant: it would be really nice if
we didn't have weird ports like that which sport totally incompatible
code for different kernels.

(pooka)

include correct header

(pooka)

shuffle around some more defs for fstat

noticed by Kurt Schreiner on current-users

(pooka)

- trailing commas in enums
- arithmetic on enums needs cast
from Anon Ymous

(christos)

fix signed comparison warning.

(christos)

just whitespace fixes.

(christos)

fix type punned warnings.

(christos)

Fix remaining register offset shifting left over from proper bus_space-ification
done a while ago. From Marco Trillo.
This should fix PR 39176.

(macallan)

Sync with head.

(simonb)

Not everyone is building from my yet uncommitted trees, so catchup
here also.

tip-of-the-hat to simonb

(pooka)

Fix a printf(3) format.

Noticed by martin@.

(simonb)

lint wins again :-) From Anon Ymous:
- fix a mis-placed parenthesis.
- fix assignment to the wrong variable.

(christos)

-in usbd_probe_and_attach(), split out the code for per-device and
per-interface attachment into individual functions, to ease
maintainance and allow easier plugin of new attachment functions
-keep a counter of USB interfaces in use on a device, and try to
keep track of interfaces claimed by drivers behind the framework's
back

(drochner)

Sync with head.

(simonb)

Sync with head.

(simonb)

add missing device_private(), fixes crash in interrupt handler

(drochner)

Avoid NULL pointer dereference on power handler deregistration if
it was not registered before. I assume that a lot of drivers do not
proper bookkeeping in the case the attach() exits early due to
errors. This is hard to fix and to test, so just be generous here.

(drochner)

Don't pmf_device_deregister() if it was not registered, which happens
if attach() encountered an error before.
In contrast, such a protection should not be needed for the register()
call, so remove it there.

(drochner)

add a comment stating why evbppc doesn't do lkm or modules

(pooka)

blkset.S is needed on vax by some macros, so include it in the build.

(pooka)

A wapbl(8) should be a wapbl(4).

Found by wizd.

(simonb)

Tidy up simonb-wapbl issues section.

(simonb)

Install mount argument structure header just like every other file system.

(pooka)

Add support for creating a WAPBL log in the filesystem.  Will
create an in-filesystem log on first "mount -o log" if one doesn't
exist, and will then continue to use same log in the future.  See
(soon to be added) wapbl(4) for more info.

Adds a new B_CONTIG low-level allocation flag that uses hints in
"struct ffs_inode_ext" to lay out an ffs file's data contiguously.

Thanks to Greg Oster for helping with the design of this and to
Antti Kantee for code review and suggestions.

(simonb)

Clarify some printfs in error cases so we at least know what subsystem
reported them.

(jmcneill)

simonb-wapbl branch now covers sbin/tunefs and usr.sbin/dumpfs.

(simonb)

Mention turning on WAPBL_DEBUG_SERIALIZE to work around performance issue
in simonb-wapbl.

(oster)

Turn on WAPBL_DEBUG_SERIALIZE in order to use RW_WRITER locks instead
of RW_READER locks in wapbl_begin().  Include the following comment as
well:

XXX: The original code calls for the use of a RW_READER lock
here, but it turns out there are performance issues with high
metadata-rate workloads (e.g. multiple simultaneous tar
extractions).  For now, we force the lock to be RW_WRITER,
since that currently has the best performance characteristics
(even for a single tar-file extraction).

Approved by: simonb

(oster)

Change how 'while (0)' is commented out to avoid /* within a comment.

(dsl)

add missing quote

(mlelstv)

Annotate dhcpcd-4.0.0rc3 import.

(joerg)

Update build glue for dhcpcd 4.0.0rc3 and move it to
src/external/dhcpcd.

Make dhcpcd quiet by default when run from rc.

(joerg)

Import dhpcd-4.0.0rc3. Major changes since beta4:
- support for link state notification (e.g. renew lease after carrier
  returned)
- support to start go into background immediately
- don't cleanup interface state. dhcpcd will leave the interface in the
  same state as before if the lease uses separate addresses. -p still
  stops it from removing the configured address on exit
- various smaller bugfixes, optimisations and cleanups

Status:

Vendor Tag: DHCPCD
Release Tags: dhcpcd-4-0-0-rc3

(joerg)

UKFS_UIOINIT() functionality has been provided for ages by rump
function interfaces, so remove the macro.

(pooka)

Adjust comment location to make it make sense.

no functional change

(pooka)

Unbreak twa after the last two commits.

(joerg)

Delay physical partition spacebitmaps from volume-descriptor-checking until
the other supporting structures are read in.

(reinoud)

Unify naming scheme and source code beautify for better readablility.

(reinoud)

Fix typo in cd9660 functionname cd9660_setup_boot_volume_descritpor ->
cd9660_setup_boot_volume_descriptor.

Interestingly the same typo was used in 3 files!

(reinoud)

Remove the '${.CURDIR}/' part in CLEANFILES, so that the files will be
deleted in the case of an objdir too.
Proposed to current-users on 20-Jul-2008 with no reply..

XXX The '*~' part should be removed too, but I leave that to someone else
    since it is used in a lot of other Makefiles too and there maybe a
    valid reason for it.

(veego)

Don't convert cpuid_t through int while testing it for being in range.
(Might truncate it, for one thing. Also, if the shift result is < 0 the
demons have already flown out of our nose.)

Fixes some -Wsign-compare warnings.

(dholland)

Add myself to the list of developers.

(pgoyette)

Forgot to rename those two references to write_space_bitmap too.

(reinoud)

Clarify the current spacemap read/write code is reading/writing physical
partition space maps. This in preperation of Metadata spacetable
reading/writeout work.

(reinoud)

Clarify not implemented allocation schemes; mostly UDF 2.50
HD-DVD/BluRay-RE and UDF 2.60 for BluRay-R

(reinoud)

Add UDF 2.50 metadata partition creation to newfs_udf(8). It allows a
percentage to the volume space to be pre-allocated for metadata on format.
Implementations that can't extent the metadata partitions can thus use the
discs too.

Note that the kernel UDF code can't write to UDF 2.50 metadata partitions
yet; this is a work-in-progress.

Also note that the default formatting version choice is still version 2.01
i.e. without the metadata partition.

(reinoud)

remove comment that is no longer true

(plunky)

sys__lwp_create() and sys__lwp_suspend(): Only test to see if a process is
an SA process ifdef KERN_SA.

sys__sched_setaffinity(): Don't allow changing (setting) the affinity
of an SA process (or a thread in an SA process). To really set the
affinity of a thread in an SA process, we need to set the affility
for all LWPs on which that user thread will run. This really means
setting the affinity on all present and future threads on a VP
and also having the user thread always run on an lwp on that VP. The
latter needs libpthread's intervention.

(wrstuden)

2020447 IPFilter's NAT can undo name server random port selection
        (fix output port range, was a random number in [0,max-min]
(byteswapped on litle endian), instead of [min,max])

(darrenr)

Moved comments for some _POSIX_* constants in front of their definition,
like all the rest of them.

(heinz)

make sure we write little endian. From Anon Ymous

(christos)

Remove all the pending connections in soclose().
'continue' in 'do .. while (0)' doesn't do what ad@ intended.

(dsl)

Change the gross #defines HVIS and SVIS into local functions.

(dsl)