Report TRAP_EXEC (for exec()) to a debugger in the PT_SYSCALL mode

Orignally exec() reporting was disabled in the NetBSD version as there
was no support for fine-grained reporting. Meanwhile PT_SYSCALL was broken
for years and there is no software that depends on this behavior.

There is need to catch exec() events in syscall tracers using ptrace(2).
Instead of adding workarounds of guessing that exec() happened, report the
event directly from the kernel.

All ATF ptrace(2) tests pass.

(kamil)

Invalidate the cache before updating the microcode. Some platforms require
this. Seen in Illumos and FreeBSD.

(maxv)

sti/cli are not allowed on Xen, we have to clear/set a bit in the
shared page. Revert x86_disable_intr/x86_enable_intr to plain function
calls on XENPV.
While there, clean up unused functions and macros, and change cli()/sti()
macros to x86_disable_intr/x86_enable_intr.
Makes Xen domU boot again
(http://www-soc.lip6.fr/~bouyer/NetBSD-tests/xen/HEAD/)

(bouyer)

fix backtrace. it was broken.
- use db_read_bytes() to avoid faults.
- quite a few functions do not use frame pointers,
  therefore always detect stack depth without a frame pointer.
  however, since the framepointer(=r14) is used as a trapframe,
  the code to detect the frame pointer is still needed.
- dump the contents of trapframe during backtracing.
- KNF

(ryo)

Revert -O0 for pmap, avoid RAW dependency compilation warning

(scole)

Fix USB_DEBUG build

(skrll)

Add ptrace() tests for reading mm* and xmm* registers

(mgorny)

toolify

PR/54182: Cross-building on Linux fails in rb.c

(skrll)

Sort options (no-argument options before ones demanding arguments).

Bump date.

(wiz)

Bump date.

(wiz)

Bump date.

(wiz)

Improve wording, bump date.

(wiz)

Improve wording, add more formatting.

(wiz)

Fix some typos, improve wording and formatting.

(wiz)

Remove trailing whitespace.

(wiz)

Fix build with MKYP=no

(roy)

Avoid KASSERT(!cpu_intr_p()) when breaking into ddb and issuing

show uvmexp

(skrll)

when dumping a kernhist history replace "%s" with "%p" so that the
kernhist in kernel can be a little more usable without making the
userland dumping code crash.

(mrg)

document 'devs' and 'procs' commands.

(mrg)

add cortex A-76 detection.

(mrg)

bio(4): .In dev/biovar.h in SYNOPSIS; .Xr envsys 4 et al in both DESCR && SEE ALSO

(cnst)

Avoid prepending a timestamp to lock debug outputs on ddb

Lock printer functions (lockops_t#lo_dump) use printf_nolog to print, but
printf_nolog now prepends a timestamp which is unnecessary for ddb:

    db{0}> show all locks/t
    [Locks tracked through LWPs]
    Locks held by an LWP (iperf):
    Lock 0 (initialized at soinit)
    lock address : 0xffffedeb84b06080 type    :    sleep/adaptive
    initialized  : 0xffffffff806d8c3f
    shared holds :                  0 exclusive:                  1
    shares wanted:                  0 exclusive:                11
    current cpu  :                  0 last held:                  1
    current lwp  : 0xffffedeb849ff040 last held: 0xffffedeb7dfdb240
    last locked* : 0xffffffff806d8335 unlocked : 0xffffffff806d8385
    [ 79103.0868574] owner field  : 0xffffedeb7dfdb240 wait/spin:                1/0

Fix it by passing a printer function to lo_dump functions, i.e., make the
functions use db_printf on ddb.

(ozaki-r)

Make _kernel_lock_dump static

(ozaki-r)

Add missing "static" declaration

(ozaki-r)

clean up ucom parents some more:
- it's always "bool sc_dying" now, with true/false
- heavy use of static functions
- remove all ucom parent ca_activate callbacks.  they're never called.
- callbacks should generally do little to nothing if sc_dying is set
- open resources should be released in detach after setting sc_dying
- don't complain about usbd_abort_pipe() or usbd_close_pipe() failure
- when releasing resources, zero the softc member as well
- remove ucom_methods members no longer destined to be filled in
- generally, DPRINTF() before sc_dying short circuit
- use EIO when dying, not ENXIO or 0
- add some ucom_open() callbacks that simply return EIO if dying

(mrg)

Protect ether_multi list

The list can be racy if NET_MPSAFE is enabled and the driver is executed without
KERNEL_LOCK.

Fix PR 54153

(ozaki-r)

Sync with reality.

(nat)

bio(4): document the state of setstate more thoroughly

(cnst)

fix acpica

(christos)

uvm_pagealloc() uses UVM_PGA_* flags, not UVM_KMF_* flags,
and it is always nowait.  fix uarea_poolpage_alloc() to not use
flags from the wrong collection for calling uvm_pagealloc()
and to wait itself if a page is not immediately available.

(chs)

Clarify that strunvisx() and strnunvisx() take the same flags as unvis().
Document VIS_NOESCAPE for unvis().
Bump date.

(bad)

Remove -p option.  AUDIO_SETCHAN is insecure and is obsoleted.

(isaki)

Update respond to isaki-audio2 branch.
- Extend list command to display supported hardware formats.
- Add set command to set hardware format.
- Use correct /dev/audioctl instead of /dev/audio.

(isaki)

Update manpages respond to isaki-audio2 branch.

(isaki)

mention truncation of the result if the size in optlen is less that the
size of the option to be returned in optval.

(christos)

PR/54176: Anthony Mallet:
getsockopt(2) does not silently truncate returned optval

(christos)

Fix a typo

Reported by Tobias Ulmer via netbsd-docs@ ML, thanks!

(leot)

Merge isaki-audio2 branch.

Welcome to 8.99.39.

(isaki)

Merge isaki-audio2 branch, the overhaul of audio subsystem.
- Interrupt-oriented system rather than thread-oriented.
- Improve stability, quality and performance.
- Split playback and record cleanly.  Improve halfduplex support.
- Many bugs are fixed including deadlocks, resource leaks, abuses, etc.
- Simplify audio filter mechanism.  The encoding/channels/frequency
  conversions are completely handled in the upper layer.  So the hard-
  ware driver only converts its hardware encoding (if necessary).
- audio_hw_if changes:
  - Obsoletes query_encoding and add query_format instead.
  - Obsoletes set_params and add set_format instead.
  - Remove drain, setfd, mappage.
  - The call sequences are changed.
- ioctl AUDIO_GETFD/SETFD, AUDIO_GETCHAN/SETCHAN are obsoleted.
- ioctl AUDIO_{QUERY,GET,SET}FORMAT are introduced.
- cleanup config attributes: au*conv and mulaw.
- All hardware drivers should follow it (I've done as much as possible).

Some file paths are changed:
- dev/audio.c        -> dev/audio/audio.c (rewritten)
- dev/audiovar.h      -> dev/audio/audiovar.h
- dev/audio_dai.h    -> dev/audio/audio_dai.h
- dev/audio_if.h      -> dev/audio/audio_if.h
- dev/audiobell.c    -> dev/audio/audiobell.c
- dev/audiobellvar.h  -> dev/audio/audiobellvar.h
- dev/mulaw.[ch]      -> dev/audio/mulaw.[ch] + dev/audio/alaw.c

(isaki)

Really go back to pre-whitespace change (1.52) arm32_kernel_vm_init

(skrll)

Add WARNS flag.  (It was left at my local in previous commit..)

(isaki)

Good bye stream_filter.

(isaki)

Remove commented out old filters.

(isaki)

Protect ether_multi list

PR 54153

(ozaki-r)

cac(4): Xr bio(4) and friends, and mention when it was first introduced

(cnst)

Make enumerations consistent, and use Dv for BIO*.

(wiz)

Revert previous, misunderstanding.

(wiz)

bio(4): document which commands are supported by which drivers;

(cnst)

mpt(4): remove ENOTTY / noop implementation of bio(4) setstate

* All other bio(4) drivers either do or don't implement each command.

* This changes the returned error code, but it does not seem to matter,
  as some bio(4) drivers always return ENOTTY in the default case, whilst
  others always return EINVAL (e.g., both cac(4) and ciss(4) for setstate).

(cnst)

new comment explains this hack:

* XXXMRG: Stall task can run after slot is disabled when yanked.
* This hack notices that the xs has been memset() in
* xhci_disable_slot() and returns.  Both xhci_reset_endpoint()
* and xhci_set_dequeue() rely upon a valid ring setup for correct
* operation, and the latter will fault, as would
* usb_transfer_complete() if it got that far.

this truly fixes yanking active ucom devices (uchcom, uplcom and
umcs all tested.)  prior efforts largely worked by allowing the
stall task to run first, as a side effect to their actual work.

ok @skrll.

XXX: pullup.

(mrg)

Sort SEE ALSO.

(wiz)

Fix typo.

(wiz)

bio(4) drivers: remove noop fallthrough-to-default bio(4) cmd cases

* Makes it easier to do code/feature analysis by not having
  extra noop code/symbols spattered around.

* Only an issue in cac(4) and ciss(4); other drivers don't do this.

* No binary changes.

(cnst)

Use posix_madvise() rather than the legacy madvise() call, and
wrapp the calls in #ifdef for the advice we're giving.  Should
address reports of host tool build issues.

(thorpej)

bioctl(8): Xr ataraid(4), mpii(4) && mfii(4);

(cnst)

mpii(4): Xr bio(4) and mention first appearance in Nx;

(cnst)

Add DTrace.

(msaitoh)

bio(4): enlist ataraid(4), mpii(4) and mfii(4) in bio(4);

(cnst)

Fix building libprop as a host tool library on platforms that don't have
the Matt Thomas rbtree:

- Include rb.c in libnbcompat, and provide a nbcompat sys/rbtree.h
  header.
- Make sure libprop's source file include prop_object_impl.h before
  anything else, and pull in nbtool_config.h from there.

Tested by simulating such a host system by renaming the host's
<sys/rbtree.h> out of the way (which reproduced the build failure)
and verifying that the host-tool installboot contained the rb_*
functions in its own .text segment.

(thorpej)

add more

(christos)

Add slop of 1000 and explain why.

(christos)

obsolete moscom(4).  it was never in any default configs, did
not work for some cases.

umcs(4) supports everything it does plus one additional chipset,
has been in default configurations since netbsd 7.

(mrg)

TMG

(sevan)

Use the max limit (aka maxfiles or the moral equivalent of OPEN_MAX) which
makes poll(2) align with the Posix documentation (which allows EINVAL if
nfds > OPEN_MAX). From: Anthony Mallet

(christos)

Ticket #1696

(martin)

Pull up following revision(s) (requested by christos in ticket #1696):

sys/netsmb/smb_conn.c: revision 1.30

Prevent a NULL pointer dereference when the local endpoint is not defined.

From Andy Nguyen, many thanks.

(martin)

Ticket #1696

(martin)

Pull up following revision(s) (requested by christos in ticket #1696):

sys/netsmb/smb_conn.c: revision 1.30

Prevent a NULL pointer dereference when the local endpoint is not defined.

From Andy Nguyen, many thanks.

(martin)

Ticket #1696

(martin)

Pull up following revision(s) (requested by christos in ticket #1696):

sys/netsmb/smb_conn.c: revision 1.30

Prevent a NULL pointer dereference when the local endpoint is not defined.

From Andy Nguyen, many thanks.

(martin)

Remove backed out ticket #1240, add ticket #1258

(martin)

Pull up following revision(s) (requested by christos in ticket #1258):

sys/netsmb/smb_conn.c: revision 1.30

Prevent a NULL pointer dereference when the local endpoint is not defined.

From Andy Nguyen, many thanks.

(martin)

Backout the following changes (requested by mrg in ticket #1240):

sys/dev/usb/umodem_common.c: revision 1.27
sys/dev/usb/umodemvar.h: revision 1.10
sys/dev/usb/ucom.c: revision 1.122

fix umodem(4) detach.
There are different fixes upcoming.

(martin)

Make CLEANFILES actually work. .TARGET is not defined when not in a target
rule.

Thanks xtos for the heads up.

(maya)

Replace the link command for libpthread.a so that we create a single section
with all the libpthread symbols in it.
This makes -lpthread behave like to -Wl,--whole-archive -lpthread.

This avoids a situation where threaded static binaries use some libc thread
stubs, which are racy.

Fixes PR lib/54001: call_once2_32, call_once2_static test cases failing on
amd64 since gcc7 import.

Suggested by Jonathan Wakely, thanks!

(maya)

- Use bus_dmamem_unmap() correctly in txp_dma_free() to prevent panic.
  The code was wrong since rev. 1.1. This panic was found by adding KASSERT
  in uvm_map.c rev. 1.351. This bug may be related to PR kern/26204.
- Use aprint_normal() for non-error message.

(msaitoh)

Fix paths and flags.
- aurateconv is gone.  mulaw is a part of audio inseparably.

(isaki)

Fix build, "pax" must be built before "libprop".

Ok: Matthew Green mrg@

(hannken)

Cleanup modules "solaris" and "zfs":

- Defer spa_config_load() until root is mounted.
- Restore the config path to "/etc/zfs/zpool.cache".
- Module "zfs" is type MODULE_CLASS_VFS and no longer depends on "rootvnode".
- Module "solaris" no longer depends on "mp_online".
- Fix rump component registration to not detach "/dev/zfs" if
  it didn't attach it.

(hannken)

This implementation of Solaris taskq API is incomplete and doesn't track
Solaris upstream.  FreeBSD already replaced it with a glue to their
taskqueue API.

Replace it with a glue component that queues Solaris taskq requests to
threadpool jobs.

(hannken)

Fix typos. Fix link. Use Pa for path.

(wiz)

convert to USBHIST style debugging.

(mrg)

Note installboot(8) for evbarm.

(thorpej)

Add installboot(8) support for evbarm (and, in the future, other evb*)
boards that use u-boot.  A known board database lists boards and their
respective u-boot packages.  u-boot packages are discovered at run-time
(in /usr/pkg/share/u-boot, by default).  These packages contain board
database overlays that describe u-boot installation procedure that's
specific for that board.

Support this as a native tool and as a host tool.  The native tool
will attempt to determine the running board type using OpenFirmware
calls.  Host tool and native tool alike may also specify a board type
directly using the "-o board=xxx option" or have installboot(8) determine
the board type from a device tree blob using "-o dtb=/path/to/board.dtb".
A "-o media=xxx" option is provided for boards that have different u-boot
binaries and/or installation procedures for different media types (e.g.
SDMMC, eMMC, or USB).

This is trivial to extend to other evb* platforms that use u-boot, even if
they don't use FDT for autoconfiguration.

(thorpej)

Use designated initializers for struct ib_mach.

(thorpej)

Add support for libprop as a host tool library.

(thorpej)

Add "nextthing,pocketchip" to the head of the 'compatible' property
so that this device tree can be distinguished from that of the base
CHIP.  "nextthing,chip" remains in the list, but one match-priority
notch down.

(thorpej)

Switch all users (except ia64) of custom machine/ansi.h to common_ansi.h

Deduplicate the code among ports and poll definitions of types
directly from a compiler.

This fixes miscompilation of certain programs that instruct compilers
to generate code for different types. This bug has been detected with
-fshort-wchar in EFI firmware.

Proposed and discussed on a mailing list (twice).

Itanium uses custom !ELF fallback switch, temporarily leave it as it is.

(kamil)

Avoid some RAW dependency compilation warnings

(scole)

revert most of:

>fix umodem(4) detach:
>
>- ucom(4) needs kpreempt disabled around softint_schedule()
>- switch a copied printf() to aprint_error_dev()
>- use static normally in umodem_common.c
>- remove unused sc_openings in softc, convert sc_dying to real bool
>- add sc_refcnt, sc_lock and sc_detach_cv to softc.  usage is:
>  - sc_dying is protected by sc_lock
>  - sc_detach_cv is matched with sc_lock for cv operations
>  - sc_refcnt is increased in open and decreased in close, any time
>    it is decreased, it is checked for less than zero, and a broadcast
>    performed on sc_detach_cv.  detach waits for sc_refcnt.
>- umodem_param() and umodem_set() check for sc_dying
>
>this fixes pullout out an open ucom@umodem.
>
>@skrll.
>
>XXX: pullup

it only fixes the issue by chance (slightly delays, which
allows task to run, but there is no guarantee.  real fix
is incoming for all ucom parents.

(mrg)

revert most of:

>fix uchcom(4) detach, like umodem(4) recently:
>
>- use static normally in umodem_common.c
>- add sc_refcnt, sc_lock and sc_detach_cv to softc.  usage is:
>  - sc_dying is protected by sc_lock
>  - sc_detach_cv is matched with sc_lock for cv operations
>  - sc_refcnt is increased in open and decreased in close, any time
>    it is decreased, it is checked for less than zero, and a broadcast
>    performed on sc_detach_cv.  detach waits for sc_refcnt
>  - uchcom_param() and uchcom_set() check for sc_dying
>
>this fixes pullout out an open ucom@uchcom.

it only fixes the issue by chance (slightly delays, which
allows task to run, but there is no guarantee.  real fix
is incoming for all ucom parents.

(mrg)

Avoid clang bug on earm with SSP/FORTIFY:

rt_libelftc_dem_gnu3.c:3567:3: warning: '__builtin___memset_chk' will always
      overflow destination buffer [-Wbuiltin-memcpy-chk-size]
      memset(&f, 0, FLOAT_EXTENED_BYTES);

(christos)

You don't have to be _LP64 to run 32 bit binaries. You can be mips (64 bit
kernel/32 bit userland).

(christos)

Bump size to resolve build.

(sevan)

cv_init() must be done before calling maple_scanbus().
ok tsutsui@

(ryo)

don't use tsleep() in interrupt context.
when the audio starts playing, initialization is performed on the aica arm driver side.

(ryo)

Revive clonify.
XXX should clean code more.

(isaki)

Oops, disable FILECORE as intended.
heads up <leot> <wiz>

(sevan)

Add support for SIGRTMIN+1..SIGRTMAX-1 signals in ktruss(1)

Generate misc.c and misc.h with entries for signals between SIGRTMIN+1
and SIGRTMAX-1.

(kamil)

siginfo(2): Ship with MI syscall information of a debugged process for
SIGTRAP and si_code TRAP_SCE/TRAP_SCX.

(kamil)

Welcome to NetBSD 8.99.38!

Signal code struct and API changes.

(kamil)

Ship with syscall information with SIGTRAP TRAP_SCE/TRAP_SCX for tracers

Expand siginfo_t (struct size not changed) to new values for
SIGTRAP TRAP_SCE/TRAP_SCX events.

- si_sysnum  -- syscall number (int)
- si_retval  -- return value (2 x int)
- si_error  -- error code (int)
- si_args    -- syscall arguments (8 x uint64_t)

TRAP_SCE delivers si_sysnum and si_args.

TRAP_SCX delivers si_sysnum, si_retval, si_error and si_args.

Users: debuggers (like GDB) and syscall tracers (like strace, truss).

This MI interface is similar to the Linux kernel proposal of
PTRACE_GET_SYSCALL_INFO by the strace developer team.

(kamil)

Remove extra word.

(wiz)

Sort SEE ALSO.

(wiz)

add a 'devs' script to display a list of all devices by device_t,
name, parent, private data.

for procs, put the printf \n into the same line.  non trivial
speed up with large process table.

(mrg)

Use filesystems.config to reduce duplication

(sevan)

Add ADOSFS & FILECORE (both disabled)

(sevan)

Add V7FS (disabled)

(sevan)

Add autofs pseudo device (disabled)
Add a description for existing pseudo devices

(sevan)

document the EINVAL caused by a large nfds.

(christos)

Include EFS support.
Tested on amd64 & macppc

(sevan)

EFS

(sevan)

fix unused variable warnings.

(christos)

Also fix the error message when -T is used with something other than A or D

(pgoyette)

Fix the BUGS section.  There is no ``-t A'' or ``-t D'' option ``-t''
requires an interval specification.  The correct bug info is for the
``-T A'' and ``-T D'' options.

(pgoyette)

Ignore 0 size'd symbols when looking for CMSE veneers.
XXX: Is this correct?

(christos)

PTYFS is no longer considered experimental.
Reviewed by <kamil>

(sevan)

Everyone has libexpat so enable it.

(christos)

Remove the slop code. Suggested by mrg@

(christos)

sync with reality

(christos)

RK3399: Skip setting RGMII TX/RX clock delayline parameters for now and
rely on the settings configured by firmware.

(jmcneill)

Introduce tests for reading r8..r15 registers.

(mgorny)

Reintroduce i386 tests for general-purpose registers, without MMX req

Reintroduce the test for reading i386 general-purpose register values
from the debugged program.  This time it's split into one test for the six
registers not normally used by the compiler (eax..edx, esi, edi), and another
for ebp, esp.  This leaves enough free registers for the latter to work
without MMX.

(mgorny)

PR/54133: Sevan Janiyan: Binaries fail to execute
Prefer MAX_PAGE_SHIFT over PAGE_SHIFT

(christos)

PR/54133: Sevan Janiyan: Binaries fail to execute
Define M{IN,AX}_PAGE_SHIFT to cover all page possibilities

(christos)

Now the real number for ipmi

(mlelstv)

reservation for IPMI driver

(mlelstv)

a little bit of clean up, sort ski items, and some notes about libunwind

(scole)

Add more comments to explain what we are doing.

(christos)

Revert i386_regs_gp_read as it fails on non-MMX systems.

(mgorny)

Add a -z flag to zero out the up to 4 bytes of padding in directory entry
names (including the terminating NUL), as well as directory entries with
extra free space (d->d_reclen > UFS_DIRSIZ(d)).

Inspired from FreeBSD:
    https://svnweb.freebsd.org/base?view=revision&revision=347066

While the kernel has been fixed to deal with the padding bytes (new
kernels will correctly zero out all the padding after the name), it
appears that there is still an issue with directory entries with extra
free space, since a newly created and populated filesystem gets modified
with "fsck_ffs -z".

(christos)

simplify the endian byte-swapping code.

(christos)

Add tests for reading pre-set general purpose register values via PT_GETREGS.

Add new tests verifying that PT_GETREGS provides correct register values
from the underlying program.  The test uses inline assembly in the child program
to set registers to well-known constants, then compares the results obtained
via PT_GETREGS against them.

Reviewed by kamil.

(mgorny)

Tickets #1693 and #1694

(martin)

Apply patch, requested by roy in ticket #1695:

        external/bsd/dhcpcd/dist/src/dhcp6.c

DHCPv6: Fix a potential read overflow with D6_OPTION_PD_EXCLUDE

(martin)

Tickets #1693 and #1694

(martin)

Apply patch, requested by roy in ticket #1695:

        external/bsd/dhcpcd/dist/src/dhcp6.c

DHCPv6: Fix a potential read overflow with D6_OPTION_PD_EXCLUDE

(martin)

Tickets #1693 and #1695

(martin)

Apply patch, requested by roy in ticket #1695:

external/bsd/dhcpcd/dist/src/dhcp6.c

DHCPv6: Fix a potential read overflow with D6_OPTION_PD_EXCLUDE

(martin)

Pull up following revision(s) (requested by maxv in ticket #1693):

sys/arch/amd64/amd64/copy.S: revision 1.33
sys/arch/i386/i386/copy.S: revision 1.31

Hum. Fix a potentially catastrophic bug: kcopy() sets DF=1 if the areas
overlap, but doesn't clear it if the copy faults. If this happens, we
return to the caller with DF=1, and each future memory copy will be
backwards.

I wonder if there really are places where kcopy() is called with
overlapping areas.

(martin)

Pull up following revision(s) (requested by maxv in ticket #1693):

sys/arch/amd64/amd64/copy.S: revision 1.33
sys/arch/i386/i386/copy.S: revision 1.31

Hum. Fix a potentially catastrophic bug: kcopy() sets DF=1 if the areas
overlap, but doesn't clear it if the copy faults. If this happens, we
return to the caller with DF=1, and each future memory copy will be
backwards.

I wonder if there really are places where kcopy() is called with
overlapping areas.

(martin)

Pull up following revision(s) (requested by maxv in ticket #1693):

sys/arch/amd64/amd64/copy.S: revision 1.33
sys/arch/i386/i386/copy.S: revision 1.31

Hum. Fix a potentially catastrophic bug: kcopy() sets DF=1 if the areas
overlap, but doesn't clear it if the copy faults. If this happens, we
return to the caller with DF=1, and each future memory copy will be
backwards.

I wonder if there really are places where kcopy() is called with
overlapping areas.

(martin)

Tickets #1255 - #1257

(martin)

Pull up following revision(s) (requested by maxv in ticket #1257):

sys/arch/amd64/amd64/copy.S: revision 1.33
sys/arch/i386/i386/copy.S: revision 1.31

Hum. Fix a potentially catastrophic bug: kcopy() sets DF=1 if the areas
overlap, but doesn't clear it if the copy faults. If this happens, we
return to the caller with DF=1, and each future memory copy will be
backwards.

I wonder if there really are places where kcopy() is called with
overlapping areas.

(martin)

Update dhcpcd entry for ticket #1256

(martin)

Sync to current, requested by roy in ticket #1256:

external/bsd/dhcpcd/dist/src/bpf.c              up to 1.9
external/bsd/dhcpcd/dist/src/defs.h            up to 1.1.1.21
external/bsd/dhcpcd/dist/src/dhcp.c            up to 1.19
external/bsd/dhcpcd/dist/src/dhcp6.c            up to 1.7
external/bsd/dhcpcd/dist/src/dhcpcd.c          up to 1.19
external/bsd/dhcpcd/dist/src/dhcpcd.h          up to 1.1.1.10
external/bsd/dhcpcd/dist/src/if-bsd.c          up to 1.7
external/bsd/dhcpcd/dist/src/if-linux.c        up to 1.1.1.13
external/bsd/dhcpcd/dist/src/if-options.c      up to 1.14
external/bsd/dhcpcd/dist/src/if-sun.c          up to 1.1.1.8
external/bsd/dhcpcd/dist/src/if.c              up to 1.1.1.13
external/bsd/dhcpcd/dist/src/if.h              up to 1.1.1.8

Import dhcpcd-7.2.2 with the following changes:
  *  DHCP: Ensure dhcp is running on the interface received from
  *  BSD: Link handling has been simplified, however it is expected
    that if an interface supports SIOCGIFMEDIA then it reports
    the correct link status via route(4) for reliable operations
  *  BPF: ARP filter is more robust
  *  BSD: Validate RTM message lengths

This security issue has been addressed:
  *  DHCPv6: Fix a potential read overflow with D6_OPTION_PD_EXCLUDE

Many thanks to Maxime Villard <max@m00nbsd.net> for discovering this issue.

(martin)

Pull up following revision(s) (requested by jdolecek in ticket #1255):

sys/dev/pci/pucdata.c: revision 1.104

add NetMos NM9912 entries

PR kern/54149 by Mouse

(martin)

Remove encoding conversions on recording.
These are handled in the upper layer now.

(isaki)