Remove extra .Pp before wizd(8) beats me.

(jruoho)

remove unnecessary step in example code
(sdp_service_attribute() did that already)
while here, start new paragraph for next example

(plunky)

Mention attribute(3) in passing.

(jruoho)

Start to document the GNU C attributes. For now, only __packed, __aligned,
and __section are documented, but I will keep extending this.

(jruoho)

ioctl: KNF switch-case

(cegger)

add PCI VGA drivers

(cegger)

Brdige -> Bridge

(cegger)

ioctl: KNF switch-case, remove useless brackets

(cegger)

ioctl: KNF switch-case

(cegger)

Don't pass unused argument. From Doug Barton.

(dholland)

Use strlcpy, not strncpy, when the desired semantics are strlcpy's
rather than strncpy's.

(dholland)

uebayasi-xip: Add a few files.

(uebayasi)

finish device_t-ification, now it works again

(macallan)

ubc_init needs to run while we're still cold or uvmhist breaks.

(eeh)

Slighty tweak the interrupt debug printfs.

(matt)

Fix typo (damn :q! in vi allow the commit to proceed).

(matt)

Add a workaround (hopefully temporary) for corrupted usb fragments on powerpc.
With this in effect, USB is now usable on some powerpc platforms again.
Basically this prevents the DMA portion of the fragment ever overlapping the
usb_frag_hdr which is enough to solve the problem.  To enable this, add

options USB_FRAG_DMA_WORKAROUND

to your config file.
Switch from malloc to kmem.

(matt)

Invalidate meta_fname when we finish a job, so that in compat
mode, we do not report a previous .meta file as being involved
in an error.

(sjg)

Whitespace police.

(haad)

Only KM_SLEEP | KM_NOSLEEP can be passed down to kmem layer, clean up
tqflags from internal taskq flags (TQ_FRONT) before calling kmem_alloc.

(haad)

fix post-move testsdir.  duh

(pooka)

goop for initial raidframe test

(pooka)

"A few seconds later" I realized the kernel directory is dev/raidframe,
not dev/raid, so reflect this in the test dir.  Affects location
for test of PR kern/44239.

(pooka)

Add first raidframe tests.  This is the test for the kernel fault
described in PR kern/44239

(pooka)

Use proper cleanup.

XXX: the atf sh "compiler" should check for errors.

(pooka)

Guard against supplying the same parameter multiple times in one -d argument.

(pooka)

language fix.  sniped by wizd

(pooka)

+rump.raidctl

(pooka)

Use RUMPPRG.

ok Greg Oster

(pooka)

Merge single-letter options in SYNOPSIS.

(wiz)

Extract searching of description tables into a single function, rather
than duplicating the code every time.  Minor reduction in code size
(about 1200 bytes on amd64), no change in functionality.

(pgoyette)

Add an option to not actually run the power management scripts, and
when in debug mode send all error messages to stderr rather than to
syslog(8).

(pgoyette)

beautify printing of SVM features across multiple lines

(cegger)

document the current syscall bug with rump_server and -l

(pooka)

fix XP bit and U bit definitions of CP15 control register.
These constants are not used in our source tree for now,
so this won't change any kernel bianries.

(bsh)

RLIMIT_NOFILE stabilized in NetBSD 17 years ago, so it's safe now
to remove the #ifdef's from around code which uses it.

(pooka)

Don't need prog_ioctl here, it's a leftover of the previous approach.

pointed out by Paul Goyette

(pooka)

add privcmd ioctl that got introduced with Xen 4

(cegger)

add gnttab ioctl definitions to implement

(cegger)

Remove duplicate asserts from when uvm_fault_lower1() was merged
into uvm_fault_lower() (the duplicates were there already before,
just in different functions).

reported by Alexander Nasonov on tech-kern

(pooka)

Nuke conditionally compiled MULOG stuff.  Itojun wanted to do it
already 8 years ago when he fixed the compilation last time, and
again it hasn't been buildable in years.

(pooka)

take two on module support
XXX this wont work as a builtin module

(ahoka)

Deal with crunch the standard way.

(pooka)

Add usr.sbin/Attfile to test list.

(haad)

Hook traceroute test case to build. it was added to lists so unbreak build now.

(haad)

Sort options in usage same as in man page.
While here, replace \xa0 with ' '. Where do you get these from?

(wiz)

Make this build with CRUNCHEDPROG defined, and default to the sysctl()
method of fetching information.  Apparently we can't simply not define
the prog_ops struct in this program.

(he)

Make this build as part of the dist/miniroot build, where it's still
referenced.  Adjust for the move from dist/ to external/.

(he)

Enable device-mapper driver in MONOLITHIC kernels.

(haad)

Fix typo.

(enami)

put real openssl version.

(christos)

Use the source dir atomic.h

(christos)

Remove unneeded <uvm/uvm.h> include

(matt)

remove unneeded <uvm/uvm_extern.h> includes

(matt)

rework <uvm/uvm_extern.h> includes.

(matt)

include <uvm/uvm_extern.h> here instead

(matt)

Remove unneeded includes of <uvm/uvm_extern.h>

(matt)

Revert last commit it's not needed anymore.

(haad)

usr.sbin/traceroute

(pooka)

don't create h_img2cgd dirs anymore since they'll just be harvested by obsolete

(pooka)

+traceroute tp

(pooka)

Add simple tests for traceroute.  The test bootstraps a three-node
network with one router in the middle and checks traceroute from
one endpoint to another with both UDP and ICMP.

(i guess these test networking at least as much as traceroute
itself, but then again that's not a bad thing)

(pooka)

rump.traceroute

(pooka)

Use RUMPPRG.  I think it's safe to say there's not going to be a
new upstream version of traceroute to import.

AS# lookup is still done using host networking.  Rationale: the
relevance to where that data comes from with respect to network
tracing is zero (be it socket, local file, db, whatever).

(pooka)

Need to include <sys/device.h> since vax's <sys/cpu.h> doesn't.

(matt)

make this compile again.

(matt)

Make these compile again.

(matt)

Make this compile again.

(matt)

Make this compilee again.

(matt)

Cleanups includes.  Change <machine/XXX.h> to <sys/XXX.h>
Remove unneeded includes.

(matt)

Wrong rfc mentioned

(christos)

Do not built 64b atomic ops to libzpool they are in libc again.

(haad)

Cleanup includes.  Change <machine/XXX.h> to <sys/XXX.h> for bus, cpu, intr.
Remove unneeded includes.

(matt)

Cleanup includes.  Switch from <machine/XXX.h> to <sys/XXX.h> from
bus, cpu, and intr.  Remove unneeded includes.

(matt)

Include <sys/evcnt.h> since <sys/cpu.h> no includes <sys/device.h>

(matt)

Since CLKF_BASEPRI isn't used any more, merge its test into CLKF_INTR which
results in a simplier comparision.

(matt)

Disable errors reported by unknown pragmas in dtrace module.

(haad)

Switch to use <sys/device_if.h>

(matt)

handle lack of <sys/atomic.h>

(christos)

silence gcc-3 uninitialized variable warning.

(christos)

Handle NetBSD-5 and 4 lack of atomics by disabling threads.

(christos)

handle openssl versions less < 1.

(christos)

filesystem -> file system.

(wiz)

fix uninitialized variable by elimination.

(christos)

Be explicit about byteswapped and ffsv2 file systems being currently
unsupported, while catching up to some changes in my local tree which
will hopefully support them at some time in the future.

Also, change "device" variable to "special", to reflect the fact
that resize_ffs will work on a plain file.

(riz)

Use a larger array to avoid overflow.

from Piotr Meyer

(pooka)

Add an example for a server listening on a tcp port.

also, bump date to avoid problems with wizd

(pooka)

use ping -n, since technically we don't have dns

(pooka)

remove some leftover development garbage

(pooka)

Make compile on non-NetBSD.

(pooka)

Per private email (to myself and core@), cgd has agreed

I'm happy to assign all copyright to this file, ...

So replace original 4-clause cgd with a 2-clause NetBSD.

(pgoyette)

From Roman Hoog Antink <rha@open.ch>: Fix possible null derefence.

(tteras)

add new, remove obsolete

(pooka)

Retire the old C helper in h_img2cgd since we can now write everything
with a shellscript in terms of rump.cgdconfig and dd.

(pooka)

Convert from the oldstyle cgd_kops.op_open to the newstyle prog_open
and define RUMPPRG.  Get rid of the Makefile.inc stuff, since it was
useful only for the oldstyle regime.

(pooka)

Remove functions/macros that are not in <tgmath.h> in NetBSD.
Use .Xr instead of .Fn for those functions/macros that have a manual page.

(jruoho)

msdosfs_sync: check for dead vnode (denode == NULL) before testing denode flags.

(hannken)

Add a manual page for the makedev(), major(), and minor() macros.

Taken from FreeBSD. Written by Ed Schouten.

(jruoho)

Make the definition of CRUNCHOPS independent of SMALLPROG_INET6 / USE_INET6.

Distrib/sparc64/instfs now builds again.

(hannken)

Initialize mutex and cv after sanity checks

(cegger)

Add -d which can be used to register host-backed block devices in
the rump kernel.

It would be cool if this, and the other options, could be specified
in a config file (plist or other).

(pooka)

Remove Xref to format(8) as it is available only on VAX and HP300.

(jruoho)

back out rev. 1.285. The problem I try to hunt down
in PR 42455 is not in the network stack as shown by PR 44206.

(cegger)

Fix xref; module(4) -> module(7).

(jruoho)

By majority vote, move module(4) to the section 7.

(jruoho)

Revert change made in revision 1.66 by ad@ this is not true and 64bit
atomic ops should be enabled in libc by default.

(haad)

on reflection we can use uintmax_t here instead of trying to decode
the value manually, as only 17 bits are defined and the value is
unlikely to exceed UINTMAX_MAX (would only prevent decoding)

(plunky)

Get DTrace to build again.

(tron)

Fix small format error. Noted by pgoyette@.

(jruoho)

When invoking a content handler specified via -C, set the REDIRECT_STATUS
CGI variable to "200".  This approximates the Apache behavior which is
required by phpcgi in PHP 5.3 (and by some compilations of older PHP
versions).

I have a revolutionary idea.  Perhaps instead of imposing bizarre
requirements on other programs in the name of "security", the PHP
developers should make their own code compile with oh, I don't know,
maybe -Wuninitialized.  That might have an effect on the security of
systems using PHP rather than on the "security" of such systems...

...excuse me, got to go now, I think my airquotes just wore out.

(tls)

Reference strsuftoll(3) and orders(7) with each other.

(jruoho)

Add another version of the simple ping test, this time written as a
shell script and using rump_server, rump.ifconfig and rump.ping.

XXX: uses rump_allserver for now, though, since i noticed a problem
where the rump kernel syscall vector does not get updated for
dlopen()'d libraries (and hence if you dlopen librumpnet.so, socket()
still gives ENOSYS).  Me be fixink it later.

(pooka)

A few weeks ago when i added "linkstr" support I didn't know that
the string parser gives a proplib object which isn't a terminated
string.  Fix that mistake now.

(pooka)

bah set lists. expect this lot back soon(ish).

(skrll)

Sort SECTIONS, fix xref.

(wiz)

Wording.

(wiz)

Fix Xr and use Sx.

(wiz)

Xref number(1) and units(1).

(jruoho)

Use only cas versions of 64 bit atomic ops.

(haad)

Note specifically that kernel modules can be loaded at securelevel 0.
In addition, some markup improvements.

(jruoho)

Use a consistent approach for rump kernel calls made by libs.
Should investigate a dynamic approach in the future.

(pooka)

Remove references to delete tests.

(skrll)

Convert a couple of old regress tests to ATF.

Remove the regress/usr.bin/rtld/dlopen/dlopen.c test as well as it's
covered by existing ATF test(s).

(skrll)

Don't special-case v1 superblocks - especially with an incorrect
calculation.

(riz)

video(4) changes to support analog tv capture devices:
- support interlacing with VIDIOC_G_FMT
- set V4L2_CAP_TUNER if driver implements the set_tuner/get_tuner
  callbacks
- set V4L2_CAP_AUDIO if driver implements the set_audio/get_audio/enum_audio
  callbacks
- add support for the following ioctls: VIDIOC_ENUMSTD, VIDIOC_G_STD,
  VIDIOC_S_STD, VIDIOC_ENUMINPUT, VIDIOC_G_INPUT, VIDIOC_S_INPUT,
  VIDIOC_ENUMAUDIO, VIDIOC_G_AUDIO, VIDIOC_S_AUDIO, VIDIOC_G_TUNER,
  VIDIOC_S_TUNER, VIDIOC_G_FREQUENCY, VIDIOC_S_FREQUENCY
- in video_submit_payload(), fix support for signaling sample complete
  using frame numbers
- new optional callbacks for drivers: enum_standard, get_standard,
  set_standard, enum_input, get_input, set_input, enum_audio, get_audio,
  set_audio, get_tuner, set_tuner, get_frequency, set_frequency

for drivers that don't provide enum_standard, get_standard, set_standard,
enum_input, get_input and set_input, the original stub implementations are
provided

(jmcneill)

moved to external

(christos)

Stop wm(4) from needlessly resetting when you add or delete a vlan(4):

ifconfig vlan0 create vlan 2 vlanif wm0
ifconfig vlan0 destroy

It pays to avoid a reset because after a reset it can take about 30
seconds before wm0 is back on the net.

The patch garnered no objections on tech-net@.

Sponsored by CoyotePoint Systems Inc.

(dyoung)

Prefix ctype bitmask macros with _CTYPE

(joerg)

Add NetBSD version of zfs snapshot management code. IT needs testing it
was only compile tested for now, but I would like to have some code in to be
sure that it will not be lost.

(haad)

Define mutex_owned.

(haad)

Remove now unneeded print_timestamp.

(haad)

Fix amd64 build where sysinfo was not defined, disable libshare for now.

(haad)

Do not include debug.h

(haad)

Add some more solaris vnode defs for zfs_ctldir.

(haad)

Include sys/debug.h to get ASSERT defined.

(haad)

Define print_timestamp during build, to shut up gcc warning.

(haad)

Simple routine to convert long to string.

(haad)

NetBSD doesn't have /etc/mnttab replace it with /dev/null.

(haad)

Don't __weak_alias non-existent symbols.  Apparently the alpha
compiler doesn't like it.

(pooka)

use crunchops for crunchables

(pooka)

use CRUNCHOPS for a crunched prog

(pooka)

befriend RUMPPRG with CRUNCHEDPRG

(pooka)

Add netstat rump client.  For now, it always sets -X, i.e. will
use only sysctl and no kvm (implementing /dev/mem for a rump kernel
would probably not be hard, but still a non-zero effort).

Note: since there is absolutely no network activity in a fresh rump
kernel, rump.netstat usually displays exactly nothing when invoked
without parameters.  Arguments like -r, -bi, -p icmp etc. produce
more stuff.

(pooka)

add attributes from the Basic Printing Profile v1.2r00

(plunky)

last of the RUMP_ACTION syscall swappers

(pooka)

+rump.mod*

(pooka)

Add rump client support for module utilities.  It should be noted
that while modstat and modunload and builtin modules work exactly
the same as in the host case, modload loads file system kernel
modules from the rump kernel namespace.  By default, archs which
have rump support for the kernel kernel ABI have the host module
directory mapped into the rump kernel namespace at the same location
(/stand/...).  Therefore, if the *host* module directory is populated,
"rump.modload foo" will work as expected.  Otherwise, RUMP_MODULEBASE
can be used to point to the module directory.

(pooka)

there's a manpage now ... just not here

(pooka)

Add PT_GNU_EH_FRAME definition

(joerg)

manpage

(pooka)

Prompted by dholland@; add the C99 TCORs.

(jruoho)

Clarifications. Improvements to wording and markup. Also follow the common
conventions and split the NOTES into CAVEATS (all these really highlight the
work-in-progress nature of the modules). Split the note about security into
SECURITY CONSIDERATIONS. Add an EXAMPLE on how to manually update modules.

(jruoho)

be a happy crunch build

(pooka)

op_read return value is a ssize_t, not an int.

(njoly)

+rump.envstat

(pooka)

RUMP_ACTION -> RUMPPRG

(pooka)

add fresh rump clients

(pooka)

RUMP_ACTION -> RUMPPRG

(pooka)

RUMP_ACTION -> RUMPPRG

(pooka)

Convert from RUMP_ACTION to RUMPPRG.

(pooka)

Define PROG here since Makefile.inc no longer defines it.

(pooka)

Convert from the .ifdef RUMP_ACTION stuff to RUMPPRG.

(pooka)

Define RUMPPRG, which is just like PROG, except it additionally
builds a rump client with the name rump.${PROG}.  The rump client
is not installed suid/sgid by default even if BINMODE is defined.

(pooka)

PR/44229: Henning Petersen: Remove dup check for whitespace.

(christos)

- no casts for malloc and free
- malloc(type) -> malloc(*value)
- malloc+strcpy -> strdup

(christos)

Change definition of pc_bus_maxdevs to get a void * instead a pci_chipset_tag_t
to match its use in pci_bus_maxdevs.  This makes it the same as pci_conf_read
and pci_conf_write.
Change all struct device * to device_t.
[Obviously nothing ever used the first argument to pc_bus_maxdevs since it
could have never worked because it wasn't being passed a pci_chipset_tag_t.]

(matt)

Also support ld syntax in -l.

was: -l librumpvfs.so
now: -lrumpvfs

(pooka)

Back out rev that shouldn't have been committed.

(matt)

Mirror sysctlgetmibinfo RUMP_ACTION change.

(pooka)

Add -m which can be used to load modules (which is a completely
different code path than using dlopen() before rump_init(), since
the former uses the in-kernel linker and the latter links the object
in rtld).

So:
golem> ./rump_server -l librumpvfs.so -m /sys/modules/tmpfs/tmpfs.kmod unix:///tmp/commsuck
==>
golem> env RUMP_SERVER=unix:///tmp/commsuck rump.modstat
NAME            CLASS      SOURCE    REFS  SIZE    REQUIRES
suser            secmodel  builtin    0    -        -
tmpfs            vfs        filesys    0    16713    -
wapbl            vfs        builtin    0    -        -

Source is filesys instead of builtin, as expected.

Notably, for -m you *must* use -l librumpvfs.so.  This is because
you need VFS in your kernel to be able to load modules from the
file system.  In a regular kernel "librumpvfs.so" is linked at
kernel build time and loaded by the bootloader.  Here we use dlopen()
for both effects (the other choices would have been to link
rump_server with -lrumpvfs, but that would limit the flexibility,
or link tmpfs.kmod directly into the binary, but that would limit
the flexibility even more).

(pooka)