Add unused memory model argument to make GCC happy.

(joerg)

When building GCC with Clang, disable optimizations for tree.c until I
can figure out what is responsible for the miscompilation.

(joerg)

remove .if make(install)

(dholland)

remove .if make(install)
these are seriously bad juju

(dholland)

Remove stray .if make(insatll).

I would really like to know why this didn't fail the clean build I ran
last night.

(dholland)

Put /usr/share/doc/reference/ref1/make in the right list. The dirs are
always created.

(dholland)

Implement error checking in m_xbyte() and check for errors after m_xbyte() call.
Reuse (len - k) expression in m_xword() and m_xhalf() to give an optimization
hint to a compiler.

When m_xbyte() didn't exist, bpf_filter() handled out-of-bounds BPF_B loads
correctly because "return 0" inside MINDEX() was aborting filter programs.
After the change that added m_xbyte() zero values were passed to A or X
registers instead of aborting a filter program.

(alnsn)

Remove atomic hack for ARM, libc now provides the missing symbols.

(joerg)

Provide a basic implementation of __atomic_load_* and __atomic_store_*,
used by GCC and LLVM as backing for C11/C++11 atomics, if the hardware
is not known to have corresponding features. Include it on ARM as LLVM
and libc++ hit it when compiled for ARMv4.

(joerg)

Drop entries for LLVM/Clang older than 3.4.

(joerg)

Add a file to help keep track of what's where and what we never
actually had.

(dholland)

Update set lists for /usr/share/doc reorg.

(dholland)

Update the distrib-dirs mtree specification for the new /usr/share/doc
layout.

(dholland)

Remove HTMLDOCDIR; it is no longer used (and doesn't fit the docs
scheme) and it was never honored anyway except in a couple places.

(dholland)

Reorg docs, part 3:
Move the documents that are papers to /usr/share/doc/papers.
Give them suitable names (including the author and year).

The key property of papers that distinguishes them from documentation
is that they're historical: they're published at a particular time and
aren't updated or maintained. (Except cosmetically.)

We should only ship papers that are of interest to users, either for
historical perspective or because they're the original research
writeup of stuff that went into the system and is still pertinent.

The ffs papers clearly meet this standard; the other one here (about
passwords, in 1978) is probably past its sell-by date.

(dholland)

Reorg docs, part 2:
Rename the following reference documents to match their programs:
  shell -> sh
  viref -> vi

and rename the following to match their topic better:
  ipctut -> sockets
  ipc -> sockets-advanced

Also, the old "timed" and "timedop" docs are now ref5/timed and
ref8/timed respectively, as the first of these documented the
protocol.

(dholland)

Reorg docs, part 1:
Move all the reference manuals to subdirs of /usr/share/doc/reference.
We have subdirs ref1-ref9, corresponding to man page sections 1-9.

Everything that's the reference manual for a program (sections 1, 6,
8), C interface (sections 2, 3), driver or file system (section 4),
format or configuration (section 5), or kernel internal interface
(section 9) belongs in here.

Section 7 is a little less clear: some things that might go in section
7 if they were a man page aren't really reference manuals. So I'm only
putting things in reference section 7 that are (to me) clearly
reference material, rather than e.g. tutorials, guides, FAQs, etc.
This obviously leaves some room for debate, especially without first
editing the docs with this distinction in mind, but if people hate
what I've done things can always be moved again.

Note also that while roff macro man pages traditionally go in section
7, I have put all the roff documentation (macros, tools, etc.) in one
place in reference/ref1/roff. This will make it easier to find and
also easier to edit it into some kind of coherent form.

(dholland)

Rework /usr/share/doc.

Update the <bsd.doc.mk> infrastructure, and update the docs to match
the new infrastructure.

- Build and install text, ps, pdf, and/or html, not roff sources.

- Don't wire the chapter numbers into the build system, or use them in
the installed pathnames. This didn't matter much when the docs were a
museum, but now that we're theoretically going to start maintaining
them again, we're going to add and remove documents periodically and
having the chapter numbers baked in creates a lot of thrashing for no
purpose.

- Specify the document name explicitly, rather than implicitly in a
path. Use this name (instead of other random strings) as the name
of the installed files.

- Specify the document section, which is the subdirectory of
/usr/share/doc to install into.

- Allow multiple subdocuments. (That is, multiple documents in one
output directory.)

- Enumerate the .png files groff emits along with html so they can be
installed.

- Remove assorted hand-rolled rules for running roff and roff widgetry
and add enough variable settings to make these unnecessary. This
includes support for
  - explicit use of soelim
  - refer
  - tbl
  - pic
  - eqn

- Forcibly apply at least minimal amounts of sanity to certain
autogenerated roff files.

- Don't exclude USD.doc, SMM.doc, and PSD.doc directories from the
build, as they now actually do stuff.

Note: currently we can't generate pdf. This turns out to be a
nontrivial problem with no immediate solution forthcoming. So for now,
as a workaround, install compressed .ps as the printable form.

(dholland)

Use M_ZERO.  From OpenBSD if_wpi.c 1.53.

(jakllsch)

Reduce wpi(4) diff to OpenBSD circa September 2007; mostly by following KNF.
The only functional changes in this commit are some %d to %u debugging printf
format specifier adjustments so as to match the OpenBSD code.

(jakllsch)

PR/48964: In urndis(4), initialization of ifnet structure is not enough,
cause panic.

Update urndis_init to return and error and use as if_init based on the
patch in the PR with stylistic changes from me.

XXX IFF_RUNNING should be ignored here?

(skrll)

Review some SLJIT_MOV instructions with respect to width.

(alnsn)

Regen from GENERIC.in rev 1.125:

> Use COPTS="-O2 -fno-reorder-blocks" for kernels except INSTALL_CONFIGURATION.

(tsutsui)

Use COPTS="-O2 -fno-reorder-blocks" for kernels except INSTALL_CONFIGURATION.

(tsutsui)

Regen to reflect GENERIC.in rev 1.121-1.124:

rev 1.121:
> dd preliminary support for the Individual Computers ACA500 in the form of acahf
> driver. If ACA500 is equipped with supported accelerator with MMU, it allows
> running NetBSD on an Amiga 500.
>
> Since ACA is not autoconf-aware, and the probe procedure does not exist yet, the
> driver is activated with the ACA500_SUPPORT kernel option. The acahf driver
> is commented out in the GENERIC kernel for now (at least until it stops being
> very experimental and hackish).
>
> When it comes to ACA-specific hardware, for now the boot CF slot works. Support for everything else is incoming...

rev 1.122:
> Changes to ACA500 support:
> - Add support for second (aux) CF slot.
> - Rework wdc attachment to be more sane.
> - Add support for clockport on ACA500.
> - Fix style where appropriate.

rev 1.123:
> Implement the probe procedure for ACA500. Now the user just needs acafh0
> compiled in his kernel and it all should(tm) work.
>
> Due to lack of obio in amiga port, the probe is called directly from mbattach().
> This is going to change with the introduction of obio for all non-autoconfig
> expansions.

rev 1.124:
> Disable X-Surf devices for DRACO's configuration.
>
> - ne at xsh requires ax88190 and ax88190 requires mii(4),
>  but all mii devices are declared only inside AMIGA_CONFIGURATION
> - X-Surf also add wdc(4) but all children devices like atabus and
>  atapibus are declared only for AMIGA_CONFIGURATION
>
> XXX: DRACO kernel is not built during build.sh release.

(tsutsui)

Disable X-Surf devices for DRACO's configuration.

- ne at xsh requires ax88190 and ax88190 requires mii(4),
  but all mii devices are declared only inside AMIGA_CONFIGURATION
- X-Surf also add wdc(4) but all children devices like atabus and
  atapibus are declared only for AMIGA_CONFIGURATION

XXX: DRACO kernel is not built during build.sh release.

(tsutsui)

ne at xshbus seems to require ax88190, per if_ne_xsh.c.

On the other hand, dp8390nic is pulled by ne in sys/conf/files
so it isn't required here.

(tsutsui)

Appease gcc48 -Wunused-but-set-variable warning.

(tsutsui)

Use vcache_rekey_* for nfs_lookitup() in the "*npp != NULL" case.

(hannken)

Add vcache operations to support key changes:

vcache_rekey_enter locks the old cache node and creates and locks the
  new cache node.  It is an error if the new cache node exists.

vcache_rekey_exit removes the old cache node and finalizes and
  unlocks the new cache node.

No objections on tech-kern@

Welcome to 6.99.46

(hannken)

Regen from MILAN.in rev 1.27 and GENERIC.in rev 1.106:

> Enable options WSDISPLAY_COMPAT_USL in template, not in generated files.

> Use COPTS="-O2 -fno-reorder-blocks" for kernels except SMALL030_KERNEL.
>
> Also remove COPTS="-Os" from std.* files and specify it in
> GENERIC.in template for SMALL030_KERNEL.

(tsutsui)

Regen from GENERIC.in rev 1.106:

> Use COPTS="-O2 -fno-reorder-blocks" for kernels except SMALL030_KERNEL.
>
> Also remove COPTS="-Os" from std.* files and specify it in
> GENERIC.in template for SMALL030_KERNEL.

(tsutsui)

PR/48963: kmem_free size mismatch causes panic when attaching urndis(4).

Fix the size passed in kmem_free in the urndis_ctrl_{query,set} functions

(skrll)

Use COPTS="-O2 -fno-reorder-blocks" for kernels except SMALL030_KERNEL.

Also remove COPTS="-Os" from std.* files and specify it in
GENERIC.in template for SMALL030_KERNEL.

(tsutsui)

Enable options WSDISPLAY_COMPAT_USL in template, not in generated files.

(tsutsui)

Use COPTS="-O2 -fno-reorder-blocks" as defined in sys.mk for userland.

Compile tests only.

(tsutsui)

Use COPTS="-O2 -fno-reorder-blocks" as defined in sys.mk for userland.

with "-O2":
  text    data    bss    dec    hex filename
2898020  54292  122728 3075040  2eebe0 netbsd

with "-O2 -fno-reorder-blocks":
  text    data    bss    dec    hex filename
2589888  54292  122728 2766908  2a383c netbsd

(tsutsui)

Use COPTS="-O2 -fno-reorder-blocks" as defined in sys.mk for userland.

with "-O2":
  text    data    bss    dec    hex filename
2064824  48784  85920 2199528  218fe8 netbsd

with "-O2 -fno-reorder-blocks":
  text    data    bss    dec    hex filename
1842944  48784  85920 1977648  1e2d30 netbsd

(tsutsui)

Use COPTS="-O2 -fno-reorder-blocks" as defined in sys.mk for userland.

with "-O2":
  text    data    bss    dec    hex filename
3717740  86092  152864 3956696  3c5fd8 netbsd

with "-O2 -fno-reorder-blocks":
  text    data    bss    dec    hex filename
3304404  86092  152864 3543360  361140 netbsd

(tsutsui)

Use COPTS="-O2 -fno-reorder-blocks" as defined in sys.mk for userland.

with "-O2":
  text    data    bss    dec    hex filename
3297898  65044  122584 3485526  352f56 netbsd

with "-O2 -fno-reorder-blocks":
  text    data    bss    dec    hex filename
2930782  65044  122584 3118410  2f954a netbsd

(tsutsui)

Add optimization hints. They replace nscratches and ncopfuncs and improve
readability.

(alnsn)

${.PARSEDIR} not .${PARSEDIR} for Makefile.compiler-rt include

(jmcneill)

Adapt to struct buf changes

(martin)

Adapt to new attach world order

(martin)

Add missing includes, fix C99 struct initialization

(martin)

Use .set arch=r5900 to enable r5900 specific instructions

(martin)

Fix extent_create() call

(martin)

Add missing <sys/device.h> include

(martin)

md_imask_update is not an inline function, fix declaration.

(martin)

Add missing OBJCOPY

(martin)

Fix pktq_enqueue for rump

Add _RUMP_NATIVE_ABI to the macro condition for i386 and x86_64 because
_RUMPKERNEL is not defined for them. See sys/rump/Makefile.rump.

Found by ATF

(ozaki-r)

Add some notes on using these shims to port Linux kernel code.

(riastradh)

Regen.

(msaitoh)

Add some Intel X38 devices.

(msaitoh)

Fix some entries:
- Desc 0x55 and 0xb1 are Instruction TLB but not fixed to 4K.
- Desc 0x5a and 0xc0 are Data TLB but not fixed to 4K.
- Desc 0x57 and 0x59 are 4K fixed DTLB.
- Fix string of desc 0xc2 and it's not fixed to 4K.
- Desc 0xca is 4K fixed L2 shared TLB.
- Add desc 0xa0.

BUG: A lot of CPUs have multiple CAI_DTLB and/or CAI_DTLB2 entries. Currently
TLB info is indexed in ci_cinfo[CAI_COUNT], so some info is overwritten.

Nowadays CPUs have very complexed TLBs. It's hard to manage with CAI_* index.
We should think to separate TLB info structure from ci_cinfo[CAI_COUNT]
in struct cpu_info.

(msaitoh)

Change the pattern of KMEM_REDZONE so that the first byte is never '\0'.

From me and lars@.

(maxv)

Trailing whitespace.

(skrll)

Exclude descriptor 0xff of CPUID leaf 2. 0xff means the cacheinfo is in leaf 4.

(msaitoh)

Add BCM5706.

(msaitoh)

Regen.

(msaitoh)

Add BCM5706 1000BASE-T/SX.

(msaitoh)

Don't use the PHY Auto Poll Mode on many chips. This fixes a bug that MII
Fiber NIC drop packet about 50%. Same as {Free,Open}BSD.
Tested on HP Moonshot. Thanks zafer@ for testing.

(msaitoh)

- 5720S and 5709S share the same PHY id. Assume 5720S PHY if parent device
  is bge(4). Same as FreeBSD. Tested on HP Moonshot.
- Fix media detect on some Fiber chips. Tested on my own BCM5709 card.
  Same as {Free,Open}BSD.

(msaitoh)

Don't check BNX_CHIP_NUM. Check MII PHY ID beacause this check is used
for both bge(4) and bnx(4).

(msaitoh)

- Remove extra delay in brgphy_mii_phy_auto. Same as {Free,Open}BSD.
- s/u_int/uint/
- KNF.

(msaitoh)

Regen.

(msaitoh)

Add Intel DH89xx devices.

(msaitoh)

Update mDNSResponder in-tree version

(pettai)

Update heimdal current and in-tree version

(pettai)

remember to call deregister.

(christos)

Register a null pmf handler even if we failed to attach.

Keeps ichsmb(4) from preventing suspend even if it's broken because
of ichlpcib(4) grodiness.

(riastradh)

If we register with pmf on attach, deregister on detach.

(riastradh)

Fix the KMEM_POISON check: it should check the whole buffer, otherwise some
write-after-free's wouldn't be detected (those occurring in the 8 last bytes
of the allocated buffer).

Was here before my changes, spotted by lars@.

(maxv)

Handle ENAMETOOLONG to return 404 error instead of 500.

OK mrg@

(shm)

Move t_bpfjit.debug entries back to mi and add "sljit" tag to them.

(alnsn)

Restore file for playstation2 accidently overlooked in the big revitilazation

(martin)

Add -msoft-float to AFLAGS as well

(martin)

The correct toolchain tuple is mipsel--netbsd (not netbsdelf)

(martin)

detect and (attempt to) deal with 8bit TCX with 2MB VRAM
TODO: get one of these VRAM modules...

(macallan)

Protect bridge_list with a mutex

(ozaki-r)

Remove obsolete codes for if_snd

(ozaki-r)

Restore RPS of pktq_enqueue unless _RUMPKERNEL

It's a workaround and would be fixed in rump soon.

ok pooka@

(ozaki-r)

Add NET_MPSAFE to ALL

(ozaki-r)

Move t_bpfjit.debug to md lists.

(htodd)

new acpica is out

(christos)

fix boot code maximum size.

From OpenBSD if_wpireg.h 1.19.

(jakllsch)

No need to duplicate the members of the wpi_cmd_data structure within
the wpi_scan_hdr structure when we can just put the wpi_cmd_data
structure within the wpi_scan_hdr structure.

This also brings our if_wpireg.h mostly in line with OpenBSD's 1.18.

(jakllsch)

On ARM the variable name 'delay' shadows a function here, rename to avoid
-Wshadow objecting.

(justin)

Implement i915drmkms wsdisplay blank/backlight/brightness controls.

(riastradh)

Print some information (ASCI revision, PCI status, etc). Ffrom FreeBSD.

(msaitoh)

Honour write-combining flag in drm_ioremap.

(riastradh)

Tweak debug printf directives.

(riastradh)

Move the main loop in bpfjit_generate_code() to a new function and make few
small changes.

(alnsn)

No functional change.
- s/u_int/uint/
- Modify comment a bit (sync with OpenBSD)

(msaitoh)

Disable DRI for Ivy Bridge (and later) until we update Mesa.

This is a stop-gap for PR 48916.

(riastradh)

KNF. No functional change.

(msaitoh)

Unbreak lib/libc/net/getifaddrs.c

  --- getifaddrs.o ---
  In file included from /tmp/bracket/build/2014.07.01.10.35.18-i386/src/lib/libc/net/getifaddrs.c:39:0:
  /tmp/bracket/build/2014.07.01.10.35.18-i386/src/sys/net/if.h:208:2: error: unknown type name 'kmutex_t'
    kmutex_t *ifq_lock;
      ^

(ozaki-r)

* bozo_clean_request free(3) clean up (removed needless checks)
* HEAD method no longer returns response body on error
* fixed bug with multiple bozo_http_error calls caused by fix_url_percent

OK @mrg

(shm)

Bump to version 6.99.45 for removal of struct lwp * parameter from pr_ioctl

(rtr)

1) Define a malloc(9)-like kmem_header structure for KMEM_SIZE. It is in
  fact more consistent, and more flexible (eg if we want to add new fields).
2) When I say "page" I actually mean "kmem page". It may not be clear, so
  replace it by "memory chunk" (suggested by lars@).
3) Minor changes for KMEM_REDZONE.

(maxv)

Make if_wm MPSAFE

- Make it MPSAFE only when NET_MPSAFE
  - otherwise, its instructions are almost same as before
  - the only change is IFQ_POLL/IFQ_DEQUEUE which
    is now single IFQ_DEQUEUE
- Protect driver operations with a lock
  - further work would make it separate
- Apply MPSAFE flag to
  - callout_init
  - pci_intr_establish
- Stop proceeding packets when the driver is likely
  to stop for graceful ifconfig down

Tested on Rangeley (I354) and KVM (e1000)
Reviewed by msaitoh@

(ozaki-r)

Lock IFQ operations when NET_MPSAFE

- Introduce NET_MPSAFE
  - not defined by default
- Add ifq_lock to protect ifnet#if_snd
- Initialize ifq_lock and lock IFQ operations
  when NET_MPSAFE

When NET_MPSAFE isn't defined, this modification
doesn't change its behavior and adds trivial
performance overheads.

Discussed with matt@ on tech-net

(ozaki-r)

Stop using callout randomly

nd6_dad_start uses callout when xtick > 0 while doesn't when
xtick == 0. So if we pass a random value ranging from 0 to N,
nd6_dad_start uses callout randomly. This behavior makes
debugging difficult.

Discussed in http://mail-index.netbsd.org/tech-kern/2014/06/25/msg017278.html

(ozaki-r)

fix parameter types in pr_ioctl, called xx_control() functions and remove
abuse of pointer to struct mbuf type.

param2 changed to u_long type and uses parameter name 'cmd' (ioctl command)
param3 changed to void * type and uses parameter name 'data'
param4 changed to struct ifnet * and uses parameter name 'ifp'
param5 has been removed (formerly struct lwp *) and uses of 'l' have been
      replaced with curlwp from curproc(9).

callers have had (now unnecessary) casts to struct mbuf * removed, called
code has had (now unnecessary) casts to u_long, void * and struct ifnet *
respectively removed.

reviewed by rmind@

(rtr)

Fix build for ipf tests.

(htodd)

Fix build for debug with bpfjit.

(htodd)

Update root.cache entry.

(taca)

Update root.cache to 2014060201.

Now B.ROOT-SERVERS.NET and C.ROOT-SERVERS.NET has an AAAA record.

(taca)

Revert previous commit since wrong branch.

(taca)

Update root.cache to 2014060201.

Now B.ROOT-SERVERS.NET and C.ROOT-SERVERS.NET has an AAAA record.

(taca)

Use MAP_ANONYNOUS (POSIX).  Define MAP_ANON as MAP_ANONYMOUS.

(matt)

Fix test directory.

(alnsn)

Apply OpenBSD src/sys/dev/pci/if_wpireg.h 1.17.

(jakllsch)

Add new net/t_bpfjit test.

(alnsn)

Add bpfjit kernel tests for loading from mbuf chain.

(alnsn)

Sync with OpenBSD src/sys/dev/pci/if_wpireg.h 1.16.

(jakllsch)

IAS is clang specific.

(joerg)

This is weird; 'abort' already does all this, so simply use goto abort.

(maxv)

Reorder two variables and fix some comments.

(maxv)

If the interpreter is "", do not keep loading the script (which will later
fail), but return ENOEXEC directly.

ok christos@

(maxv)

New IPv6 tests for ipnat parsing need to be built by make so that they can
be executed.

(darrenr)

NetBSD has an entry for protocol #255 in /etc/protocols labeled "reserved"
so expect that in the parsing output instead of 255.

(darrenr)

Undo the mac68k specific previous change and instead move all mkisofs-based
images into ${RELEASEDIR}/images (just like the makefs ones).

(martin)

Use NZERO instead of hard-coded "20" value.

(njoly)

sun4v: The maximum number of register windows is a constant and thus not available via the %ver register (sun4u only). Introduce a macro GET_MAXCWP that handles this, so cpu_switchto() works properly on sun4v. Other relevant reference to %ver can be adapted to use GET_MAXCWP as the sun4v port progresses. OK martin@

(palle)

Cleanup ppp_inproc

- Remove unnecessary variable isr
- Use pktq instead of rv to switch between inet/inet6 and other protocols

ok msaitoh@ and rmind@

(ozaki-r)

Ticket 1090.

(msaitoh)

Ticket 1090.

(msaitoh)

Ticket 1090.

(msaitoh)

Pull up following revision(s) (requested by maxv in ticket #1090):
sys/compat/netbsd32/netbsd32_compat_50.c: revision 1.24
sys/compat/netbsd32/netbsd32_event.c: revision 1.11
Allocate directly KQ_NEVENTS bytes. Otherwise a user can panic the system.
ok christos@

(msaitoh)

Pull up following revision(s) (requested by maxv in ticket #1090):
sys/compat/netbsd32/netbsd32_compat_50.c: revision 1.24
sys/compat/netbsd32/netbsd32_event.c: revision 1.11
Allocate directly KQ_NEVENTS bytes. Otherwise a user can panic the system.
ok christos@

(msaitoh)

Pull up following revision(s) (requested by maxv in ticket #1090):
sys/compat/netbsd32/netbsd32_compat_50.c: revision 1.24
sys/compat/netbsd32/netbsd32_event.c: revision 1.11
Allocate directly KQ_NEVENTS bytes. Otherwise a user can panic the system.
ok christos@

(msaitoh)

Ticket 1909.

(msaitoh)

Pull up following revision(s) (requested by maxv in ticket #1909):
        sys/compat/freebsd/freebsd_sched.c 1.20-1.21

Avoid NULL dereference and fix sched param conversion (at least make it
do something). Pointed out by Maxime Villard.

(msaitoh)

Ticket 1909.

(msaitoh)

Pull up following revision(s) (requested by maxv in ticket #1909):
        sys/compat/freebsd/freebsd_sched.c 1.20-1.21

Avoid NULL dereference and fix sched param conversion (at least make it
do something). Pointed out by Maxime Villard.

(msaitoh)

Ticket 1909.

(msaitoh)

Pull up following revision(s) (requested by maxv in ticket #1909):
        sys/compat/freebsd/freebsd_sched.c 1.20-1.21

Avoid NULL dereference and fix sched param conversion (at least make it
do something). Pointed out by Maxime Villard.

(msaitoh)

Schedule pppoe_softintr only when a packet is enqueued

(ozaki-r)

Acquire SW semaphore in wm_get_swsm_semaphore

It is required to serialize device drivers running on different CPUs.
The basic instructions are same as the official Intel driver.

ok msaitoh@

(ozaki-r)

Note micphy(4)

(ozaki-r)

Ticket 1086.

(msaitoh)

Ticket 1086.

(msaitoh)

NPF: use BPF JIT by default.

(rmind)

drop trailing whitespace

(jakllsch)

Pass the scan result RSSI to the WPA code in a way that it understands.

(jakllsch)

Make sure the iso target directory is created in-time for the mkisofs
invocation

(martin)

set TOOLCHAIN_MISSING for playstation2.
default EXTERNAL_TOOLCHAIN to choosing to be GCC.

ps2 pkgsrc almost works now..

(mrg)

Remove an unused variable

(martin)

Change assertion as rump_component_load may be called before curlwp available

(justin)

NBPFILTER is no more.

(mrg)

allow i2c address 0x01-0x7F
igore responses if the sensor is in command mode

(kardel)

make implementation match gpio(4) man entry. do no change configuration
if no flags are be set (allows non-destructive inquiry via GPIOSET again).

(kardel)

PR kern/47462

(darrenr)

Add in test cases for testing ipnat's parsing of IPv6 NAT rules

(darrenr)

Add in test cases for testing ipnat's parsing of IPv6 NAT rules

(darrenr)

remove artifacts inspired by misleading documentation
fix address caculation macros
fix typoes

(kardel)

#552 destination port not zero after parsing nat rule
3561691 gethost never returns an ipv6 address

(darrenr)

#551 ipf.conf address structure not properly zero filled

(darrenr)

#536 ipnat can try to print rule as dstlist incorrectly

(darrenr)

Use "-O2 -fno-reorder-blocks" instead of "-Os" for m68k default optimization.

"-Os" was specfied when gcc 4.5.x was imported and the commit log says
"-O2 produces much bigger code with gcc 4.5 than it did with gcc 4.1"
but "-Os" disables most inline declarations and makes some applications
much slower.  "-O2 -fno-reorder-blocks" seems enough to reduce sizes
and disabling -freorder-blocks wouldn't cause particular performance
impact on ancient m68k machines with small cache memories.
See my post on port-m68k@ for more details:
http://mail-index.NetBSD.org/port-m68k/2014/06/22/msg000488.html

No objection in the thread and "seems fine to me" from mrg@.

(tsutsui)

3561690 ipv6 address for test.hosts.dots in wrong byte order

(darrenr)

Pull the simliar workaround from gcc 4.5.4:

http://mail-index.NetBSD.org/source-changes/2014/06/29/msg055885.html
---
Tweak LIB1ASMFUNCS order to avoid linker warnings on libgcc_s build with -O2.

Without this change, ld complains as the following:
>> libgcc_s_pic.a(_float.pico):(.text+0x8): relocation truncated to fit:
>> R_68K_PC16 against symbol `$_exception_handler' defined in .text section in
>> libgcc_s_pic.a(_floatex.pico)

_float.S and _double.S refer `$_exception_handler' declared in _floatex.S
and linking the _floatex.S first seems to work around these warnings
(probably caused by pic relative jump addresses).
See port-m68k@ posts for more details:
http://mail-index.NetBSD.org/port-m68k/2014/06/22/msg000488.html
---

Note m68k/defs.mk is manually edited to avoid extra diffs.

(tsutsui)

Tweak LIB1ASMFUNCS order to avoid linker warnings on libgcc_s build with -O2.

Without this change, ld complains as the following:
>> libgcc_s_pic.a(_float.pico):(.text+0x8): relocation truncated to fit:
>> R_68K_PC16 against symbol `$_exception_handler' defined in .text section in
>> libgcc_s_pic.a(_floatex.pico)

_float.S and _double.S refer `$_exception_handler' declared in _floatex.S
and linking the _floatex.S first seems to work around these warnings
(probably caused by pic relative jump addresses).
See port-m68k@ posts for more details:
http://mail-index.NetBSD.org/port-m68k/2014/06/22/msg000488.html

Note m68k.mk is manually edited to avoid extra diffs.

(tsutsui)

Export accessops functions to allow overriding them in MD attachment.

(tsutsui)