Remove the need for -Wno-pointer-sign. Object diffs checked.

(dholland)

remove unused global variable

(dholland)

sprinkle static in this file

(dholland)

Remove the need for -Wno-pointer-sign. Does not change the output .o file.

(dholland)

use __arraycount, kill N() macro.

(cegger)

Instead of using the net.inet6.ip6.accept_rtadv sysctl for all devices,
make net.inet6.ip6.accept_rtadv the default for individual per-device
settings so people can use the ndp(8) utility to set per-device whether
or not to accept route advertisements.

rtadvd changes to follow.

(Debated on tech-net@ before but almost two weeks passed by without any
comment on the patch.)

(tonnerre)

Bump date for previous.

(wiz)

Ignore comment lines in .editrc from Jess Thrysoee

(christos)

Reword for better HTML output.

(wiz)

use device_xname

(cegger)

add GPIO_DEBUG

(cegger)

make this build w/ GPIO_DEBUG

(cegger)

add new Service Class ID and Protocol UUIDs from the Service Discovery
assigned numbers document at www.bluetooth.com

add detail printing for Message Access Profile v1.0

(plunky)

add in definitions from the latest Service Discovery Assigned Numbers
document from www.bluetooth.com

(plunky)

Change one more struct device * to device_t, noticed by cegger.

(mbalmer)

Mention gpio and kauth changes.

(mbalmer)

Use device_t instead of struct device * per cegger's request.

(mbalmer)

- replace magic number with proper define from mii.h. From OpenBSD.
- fix typo in comment. From OpenBSD.

(cegger)

Allow gpiosim(4) to attach if configured in the kernel configuration.

(mbalmer)

Add new files due recent gpio(4) rework.

(mbalmer)

Document recent gpio(4) changes and introduce a new config file for GPIO.
Integrate with the startup scripts in /etc/rc.d.  Introduce new variable
"gpio" for /etc/rc.conf.

(mbalmer)

Reworked gpioctl(8) command to reflect recent gpio(4) changes.

(mbalmer)

Rework the GPIO framework.  Tie it to the kauth(9) framework to control
access to the GPIO pins.  Device drivers using GPIO pins can now be
attached and detached at runtime.  GPIO pins can be named for easier
reference from userland programs.  Introduce a new gpiosim(4) driver,
which is used for development.

Reviewed by many.

(mbalmer)

Extend the existing security models for upcoming gpio(4) changes.
Reviewed and feedback by Elad Efrat.

(mbalmer)

Call prop_dictionary_set_uint32() before mii_init() so that brgphy
can do prop_dictionary_get_uint32() correctly.

(msaitoh)

When installing via a temporary file, base the name of the temporary
file on the name of the target file, not just the target directory, to
ensure uniqueness when multiple concurrent invocations of install(1)
simultaneously install files in the same directory.  Fixes bin/41512.

(gson)

de-__P() and ANSIfy; no functional change

(dholland)

(1) reject instead of ignore extra arguments;
(2) reject requests for environment variables containing '='
    (PR 41774, but with different patch)
(3) fix capitalization of usage message

(dholland)

There's no such file as arch/i386/i386/core_machdep.c, so don't
list it here.

(dyoung)

new root.cache

(christos)

sync; adds v6 address for L.

(christos)

the movers came.

(christos)

Delay emulation record output, to be processed after the current
syscall; to avoid picking syscall name from the wrong emulation table.

(njoly)

- rework the page pinning API, so that now a function is provided for
each level of indirection encountered during virtual memory translations. Update
pmap accordingly. Pinning looks cleaner that way, and it offers the possibility
to pin lower level pages if necessary (NetBSD does not do it currently).

- some fixes and comments to explain how page validation/invalidation take
place during save/restore/migrate under Xen. L2 shadow entries from PAE are now
handled, so basically, suspend/resume works with PAE.

- fixes an issue reported by Christoph (cegger@) for xencons suspend/resume
in dom0.

TODO:

- PAE save/restore is currently limited to single-user only, multi-user
support requires modifications in PAE pmap that should be discussed first. See
the comments about the L2 shadow pages cached in pmap_pdp_cache in this commit.

- grant table bug is still there; do not use the kernels of this branch
to test suspend/resume, unless you want to experience bad crashes in dom0,
and push the big red button.

Now there is light at the end of the tunnel :)

Note: XEN2 kernels will neither build nor work with this branch.

(jym)

spaces to tab.

(skrll)

G/C DCIAS

(skrll)

Remove trailing whitespace

(skrll)

Wrap some long lines.

Whitespace.

(skrll)

we have xorg 1.6.2 now.

(mrg)

check return code from soreserve() (Sean Boudreau)

(christos)

Sync with HEAD.

(jym)

Move the RAID shutdown to the raid(4) detachment routine, and use
config_detach(9) to shutdown a RAID.

Detach raid(4) units at shutdown.

Ok by oster@.

(dyoung)

Extract ldlastclose() and use it in ldclose().  At the top of
ldbegindetach(), call disk_begindetach(..., ldlastclose, ...).

Compiles.  Not tested.

(dyoung)

Use kpause(9), cv_timedwait(9), and cv_signal(9) instead of tsleep(9)
and wakeup(9).

Use mstohz(9).

XXX Protection against spurious wakeups is still needed, but this patch
XXX makes the code no worse than before in this regard.

(dyoung)

Avoid a kernel assertion failure upstream by using FSTATE_NOTFOUND
rather than FSTATE_FOUND when setting the unit number directly.

config_attach_pseudo() will convert it to FSTATE_FOUND just after the
assertion.

(plunky)

Fix typos in comments and __PRINTKs.

(jym)

Bump date for previous.

(wiz)

Bump date for previous.

(wiz)

Provide bus_space_mmap. Still needs implementing.

(skrll)

update

(yamt)

quick hack to allow brightness control via PMF

(macallan)

Reduce the difference against FreeBSD and OpenBSD

  Define BGE_IS_JUMBO_CAPABLE and use it. No functional change.

  Cleanup the code for BGE_PCI_DMA_RW_CTL regster setting. No functional
  change.

  Fix a printf message.

(msaitoh)

Fix from Bert Kiers, also noted by jak - be consistent in naming the
key for the target hostname variable - it's "target hostname", not
"target host".

(agc)

Clarify the description of the errno behavior, based on the recent
thread in tech-userlevel.

(dholland)

Mention, along with the behavior of succeeding system calls, the
normal behavior of succeeding library functions regarding errno.
(That is, they may leave arbitrary nonzero values in errno.)

Ok kleink@; related to PR lib/41567 as well as a recent thread on
tech-userlevel.

(dholland)

Restore error behaviour bulldozed in rev 1.246.

might fix PR kern/41769

(pooka)

return proper errno from mount

(pooka)

Bump date.

(alc)

Constify the fourth argument of sysctlbtname(3) (ie. `newp', the pointer to
the new value).

This change sync sysctl(3) and sysctlbtname(3) prototypes.

No objection on <tech-userlevel>

(alc)

Ticket 1338.

(snj)

Apply patch (requested by joerg in ticket #1338):
- Add support for SHA224 to libc
- Fix Big Endian support in some cases
- Use libc support in libcrypto

(snj)

Ticket 1338.

(snj)

Apply patch (requested by joerg in ticket #1338):
- Add support for SHA224 to libc
- Fix Big Endian support in some cases
- Use libc support in libcrypto

(snj)

Don't set debug for the directory.

(joerg)

Since rumpfs can now be used from publically exported routines,
convert unsupported checking KASSERTs to return EOPNOTSUPP.

(pooka)

delint

(pooka)

* catch up with the ukfs changes
* leave mounted fs reference counting totally up to le coq grande
  and simply force an unmount if the kernel has decided the mountpoint
  is not busy

(pooka)

Since rump rootfs now supports creating directories, mount ukfs
mounts at the requested mountpoint instead of as the rump kernel
rootfs.  While doing so, generally convert everything to use system
calls instead of handcrafter rump interfaces (e.g. rump_sys_mount()
instead of rump_mnt_mount()).  This avoids a lot of unnecessary
fuss.

Change UKFS_DEFAULTMP from / to /ukfs and bump ukfs version.

(pooka)

Declare strsignal(3) in <string.h>, per POSIX-2008, and adjust
documentation accordingly.

Fixes PR standards/41325 from Perry E. Metzger.

(kleink)

Sort supported hardware by name.

(wiz)

Remove unnecessary .Pp.

(wiz)

Only need path if we have issetugid... From Anon Ymous

(christos)

Bump date

(alc)

Update supported chips list.

(alc)

HAL redistribution is no longer constrained by specific redistribution rules.

(alc)

The HAL is no longer a kernel blob, reflect this.

(alc)

Make compat netbsd32 sendmsg follow the native version by allowing
empty messages. This let unfdpass regression test work as expected.

(njoly)

Prepare curses for the possibility of changing from termcap to terminfo.
term.h #defines lines, pad_char and no_color_video macros which conflict
with existing curses code. We change lines to alines and nlines depending
on use, pad_char to padchar and no_color_video becomes no_color_attributes
but with a strong alias from no_color_video.

(roy)

Don't depend on side effects inside an assert
From Michael Cook mcook at bbn dot com

(christos)

Fix memory leaks in error paths.
From Michael Cook mcook at bbn dot com

(christos)

Always initialize nread since it is an out param.
From Michael Cook mcook at bbn dot com

(christos)

Move filename to the scope it is being used.
From Michael Cook mcook at bbn dot com

(christos)

Make compat linux/linux32 getdents(2) fail with ENOTDIR instead of
EINVAL when file descriptor does not refer to a directory.

(njoly)

add a "public domain" note (there is nothing creative justifying a
long license)

(drochner)

Add usr/libdata/debug/usr/tests/lib/libc/string for MKDEBUG build.

(njoly)

Add comma in enumeration.

(wiz)

apply patch-am from pkgsrc modular-xorg-server:

date: 2009/06/24 17:25:36;  author: drochner;  state: Exp;  lines: +10 -89
zero out a local variable, fixes random behaviour in case a DRI hw module
cannot be loaded and the server falls back to software rendering, from
Rafal Boni per mail to tech-x11

(mrg)

apply the patch from http://bugs.freedesktop.org/attachment.cgi?id=27568.
it fixes various issues picked up by newer pixman.

(mrg)

typo in comment

(dholland)

regen: fh syscalls

(pooka)

+fhopen, +fhstatvfs1 RUMP

(pooka)

flag a 64-bit integer constant as "ULL" -- this is not clean but the
code around it assumes it anyway
fixes build on 32-bit

(drochner)

remove stray semicolons, makes lint happy and lets the libc build proceed

(drochner)

Extract a lot of code from dkwedge_del(), and move it to dkwedge_detach()
to create a comprehensive detachment hook.  Let that hook run at
shutdown.  Now, 'drvctl -d dk0' actually deletes a wedge if it is
not in-use (otherwise fails w/ EBUSY), and wedges are gracefully
detached from their "parent" at shutdown.

(dyoung)

Kill unreachable return statement.

(njoly)

Do reject unknown/invalid linux clockid.

(njoly)

Include limits.h.

(joerg)

Don't redefine uint16_t and uint8_t, just use the system includes.

(joerg)

Make the constant u_longlong too, lint is just too stupid...

(joerg)

Some parts of gcc fail with the direct include of sys/inttypes.h, so
fall back to using sys/types.h like stdlib.h does.

(joerg)

Add explicit casts to make lint happy.

(joerg)

device_t/softc split.  Compiled, but never run.

(dyoung)

No need to clear BSS in kernel itself if there is valid bootinfo,
i.e. it's loaded by our native bootloader.

(tsutsui)

Move popcount et al to src/common and add popcount32/popcount64.
Requested by rmind@. MD should now override popcount32/popcount64 and
provide the aliases as fitting.

(joerg)

Fix an #ifdef botch in rev 1.214 that causes
[ Kernel symbol table invalid! ] message at boot,
which means no ksyms(4) support even on GENERIC kernel.

Should be pulled up to netbsd-4 and netbsd-5.

(tsutsui)

+mount_nilfs.debug

(njoly)

Xref popcount(3).

(wiz)

Fix typo, add comma in enumeration.

(wiz)

Add popcount(3) and the long and long long version. Name is inspired by
gnulib, the implementation goes back to the AMD Software Optimizer
guide. A number of platforms will want to replace the C version with
assembler code using native instructions.

(joerg)

Add HISTORY.

(joerg)

Some improvements based on the work of Adam Ciarcinski.
Adapted everything to integrate with current 5.99 source.
Added an Amiga PIC for interrupt handling.

Status:
A CSPPC machine probably boots into single user, but only few devices
will attach without problems (e.g. keyboard, serial, parallel, floppy disk,
the RTC, CC graphics, sometimes Ariande ethernet and A4000 IDE.
BPPC support is prepared, but unfinished and untested (I don't own a BPPC).

A problem which has to be solved is interrupt acknowledging, which may happen
twice. Once in the amiga drivers from sys/arch/amiga/dev and then in the PPC
specific pic_amiga.c.

(phx)

device_t/softc split
CFATTACH_DECL -> CFATTACH_DECL_NEW
struct device * -> device_t
struct cfdata * -> cfdata_t
Use aprint*

(skrll)

protect C constructs from assembler source inclusion.

(nisimura)

Modify to support multibyte characters.

(ahoka)

Apply some ANSI and KNF.

(ahoka)

Change to support multibyte characters.

(ahoka)

do not need the postfix libraries here.

(mrg)

do not need the postfix libraries.

(mrg)

rebuild all of these based upon build-makefiles version 1.7.

XXX: i have re-applied the by-hand patch to
XXX: compat/gnu/lib/crtstuff4/Makefile that ensures it finds the right
XXX: tconfig.h.

(mrg)

updates for compat builds.

(mrg)

it's not quite upto libcrypto.so.6.1 yet...

(mrg)

what was usr.bin/ssh is now openssh/bin, not openssh/bin/ssh.

(mrg)

set SSHDIST to the new location.  HI CHRISTOS!

(mrg)

Tickets 857, 863, and 865.

(snj)

don't need absolute paths for fs component names

(pooka)

Pull up following revision(s) (requested by rmind in ticket #865):
sys/arch/amd64/amd64/trap.c: revisions 1.56, 1.57
handle protection fault properly.

(snj)

Tickets 857, 863, and 865.

(snj)

Pull up following revision(s) (requested by rmind in ticket #865):
sys/arch/amd64/amd64/trap.c: revisions 1.56, 1.57
handle protection fault properly.

(snj)

Make rump_lfs work again: load fs components dynamically instead
of trying to link them in on the command line and get pounded by
link set failage.

(pooka)

the consumers expect the aiodone workqueue to be mpsafe.  make it so.

(pooka)

Pull up following revision(s) (requested by rmind in ticket #863):
sys/sys/vnode.h: revision 1.207
sys/kern/vfs_subr.c: revision 1.379
put a flag bit into v_usecount to prevent vtryget during getcleanvnode.
this fixes the following deadlock.
        a thread doing getcleanvnode:
        pick a vnode
        acqure v_interlock
        v_usecount++
        call vclean
                now, another thread doing cache_lookup:
                picks the vnode
                vtryget succeed
                vn_lock succeed
        now in vclean:
        set VI_XLOCK (too late to be noticed by the competing thread)
        wait on the vnode lock (this might violate locking order)
the use of a flag bit was suggested by Andrew Doran.  PR/41374.

(snj)

Pull up following revision(s) (requested by rmind in ticket #863):
sys/sys/vnode.h: revision 1.207
sys/kern/vfs_subr.c: revision 1.379
put a flag bit into v_usecount to prevent vtryget during getcleanvnode.
this fixes the following deadlock.
        a thread doing getcleanvnode:
        pick a vnode
        acqure v_interlock
        v_usecount++
        call vclean
                now, another thread doing cache_lookup:
                picks the vnode
                vtryget succeed
                vn_lock succeed
        now in vclean:
        set VI_XLOCK (too late to be noticed by the competing thread)
        wait on the vnode lock (this might violate locking order)
the use of a flag bit was suggested by Andrew Doran.  PR/41374.

(snj)

enter the right bin/ssh subdir for "configinstall".

HI CHRISTOS!

(mrg)

fix the generated from id.
don't generate some subdir makefiles.

(mrg)

Pull up following revision(s) (requested by rmind in ticket #857):
sys/kern/sys_mqueue.c: revision 1.21 via patch
mq_send/mq_receive: while permission may allow that, return EBADF if sending
to read-only queue, or receiving from write-only queue.
From Stathis Kamperis, thanks!

(snj)

Pull up following revision(s) (requested by rmind in ticket #857):
sys/kern/sys_mqueue.c: revision 1.21 via patch
mq_send/mq_receive: while permission may allow that, return EBADF if sending
to read-only queue, or receiving from write-only queue.
From Stathis Kamperis, thanks!

(snj)

Can't see any reason to stat() the .so before dlopen(), so get rid
of that call.  uses normal ldlib paths now

(pooka)

bring this much closer to reality:

- add many subdirs of libs missing
- output a literal $NetBSD$ as well as the expanded version of this script
- special case a bunch of the external subdirs, since some of the library
  subdirs exist in toplevel directory, not the "lib" directory

(mrg)

Don't lets this linger around forever. Causes hidden bugs.

(christos)

ansi prototypes for yyinput.

(christos)

moduli moved with openssh

(christos)

fix for new openssl

(christos)

make tests compile!

(christos)

reference the regular Makefile directly instead of copying it.

(christos)

use the new ssh

(christos)

In ipmi_match(), initialize the condition variable sc_cmd_sleep.
Fixes a bug in previous, exposed by

# drvctl -d ipmi0
# drvctl -r -a ipmibus mainbus0
*lockdebug panic here*

(dyoung)

Overhaul synchronization in ipmi(4): synchronize all access to
device registers with a mutex.  Convert tsleep/wakeup calls to
cv_wait/cv_signal.

Do not repeatedly malloc/free tiny buffers for sending/receiving
commands, but reserve a command buffer in the softc.

Tickle the watchdog in the sensors-refreshing thread.

I am fairly certain that after the device is attached, every register
access happens in the sensors-refreshing thread.  Moreover, no
software interrupt touches any register, now.  So I may get rid of
the mutex that protects register accesses, sc_cmd_mtx.

(dyoung)

Use __constfunc instead of __attribute__((__const__)).

(joerg)

constify for new openssl

(christos)

Use __RCSID.

(joerg)

describe & attach raidframe rump kernel component

(pooka)

Add cgd rump kernel component.

(pooka)

device subroutines and autoconfig support for rump.  work in progress.

(pooka)

add crypto support for rump

(pooka)

use new openssh tree

(christos)

Add __constfunc and explain how it differs from __pure.

(joerg)

ssh has moved (a long time ago)

(christos)

ssh has moved

(christos)