Sort sections.

(wiz)

New sentence, new line.

(wiz)

Appease the extent routines and pass a u_long instead of a bus_addr_t.
(slightly bogus but then we really should kill extent and switch to vmem)

(matt)

Add/use PRIxBUS{ADDR,SIZE} and PRIxBSH

(matt)

Made mfcr/mtcf dependent on MIPS64

(matt)

fix some typos in asm()'s

(cliff)

set up some Processor Control registers at start of mach_init():
- disable all threads except #0
- disable Unaligned Access
- set single MMU Thread Mode, single TLB partition

(cliff)

Build P5064-64 if doing 64-bit builds

(matt)

remove debug printf

(matt)

Update for mdocml-1.9.14.

(joerg)

Import mdocml-1.9.14:
- fixed proper font cascading (no double-mode in -Tascii)
- added correct scoping of \f
- fixed handling of \c for all input/outputs

Status:

Vendor Tag: KRISTAPS
Release Tags: mdocml-1-9-14

(joerg)

kill a stray semicolon which made that the "-p <cmd>" option in
POSIX mode didn't work as expected

(drochner)

Add RTLD_LAZY to dlopen call, to make it work on Linux hosts.

ok by pooka.

(njoly)

Follow up commit to fix breakage that somehow passed by the compiler.

(joerg)

Merge changes between revision 1.16 and 1.19 of OpenBSD with
shutdownhook_establish(9) to pmf(9) conversion:

- LOMlite seems to get wedged from time to time; add some code to unwedge it.

- Make sure we don't insert and entry into the list of pending commends twice.

- Establish a shutdown hook to disable the watchdog timer to prevent watchdog
  triggers after the kernel has been halted.

- Handle LOMlite2 in an interrupt-driven way; avoids using delay(9) once the
  machine is up and running.

(nakayama)

Add SYSVMSG support for linux32.

(joerg)

Make sure to never leak padding space before copyout or copyin
uninitialized fields by explicitly using memset in the conversion
routines.

(joerg)

Note briefly that new amd64 kernels don't build with old binutils.

(dholland)

fix typo in previous

(dholland)

Updated fs-utils man pages

(stacktic)

add bus space for PCIe CFG and ECFG spaces

(cliff)

- use new obio bus space
- add bus space stuff for PCIe CFG and ECFG spaces

(cliff)

- abandon XKPHYS mapping for access PCIe CFG and ECFG space, useless for N32
- rmixl_pcie_conf_read() and rmixl_pcie_conf_write() now use
rmixl_pcie_conf_setup() to establish bus space mappings for
access to PCIe CFG, ECFG space.  The mappings are "lazy"
to allow successive accesses to the same (bus) in CFG space,
or to the same (bus, device) in ECFG space, to reuse the mapping.

(cliff)

- use PRIxPADDR and PRIxPSIZE as needed when printing
bus address and size following change in typedef
- use new obio bus space

(cliff)

- use new obio bus space

(cliff)

- use (intptr_t) as needed to make int to pointer casts work for N32 and N64
- in BS(map), when not _LP64, if the bus address cannot "fit"
in KSEG0/KSEG1, then use uvm_km_alloc() and pmap_kenter_pa()
to make page table mappings.

(cliff)

- need to be able to handle >32 bit bus addrs in N32 kernel
- bus_addr_t is now paddr_t
- bus_size_t is now psize_t
- bus_space_handle_t is now  intptr_t
- this will impact other MIPS ports, TBD

(cliff)

- make infop work for N32 or N64; firmware gives us a
non-sign-extended 32 bit pointer in a 64 bit variable
- in rmixlfw_init(), when PSB version not found,
fix rc_io_pbase initialization, and add the missing return MEMSIZE.
- use PRIx64 format to make some printfs work for N32 or N64
- use (intptr_t) as needed to make int to pointer casts work for N32 and N64
- init rc_obio_memt replaces init of rc_eb_memt, rc_el_memt

(cliff)

- make infop work for N32 or N64; firmware gives us a
non-sign-extended 32 bit pointer in a 64 bit variable
- in rmixlfw_init(), when PSB version not found,
fix rc_io_pbase initialization, and add the missing return MEMSIZE.
- use PRIx64 format to make some printfs work for N32 or N64
- use (intptr_t) as needed to make int to pointer casts work for N32 and N64
- init rc_obio_memt replaces init of rc_eb_memt, rc_el_memt

(cliff)

- delete un-needed makeoptions CPUFLAGS, AFLAGS ("-mips64")
- delete un-needed no options MIPS32

(cliff)

- delete un-needed options MIPS32

(cliff)

- we don't need -el bus space for obio, get rid of it and clean up the naming
- delete rmixl_eb_space.c, rmixl_el_space.c
- add mixl_obio_space.c, provides -eb bus space for obio devices

(cliff)

Don't define wchar_t for c++; it is native in the language. From Alexander
Nasonov

(christos)

Pull in fixes from upstream revision 1.146 to avoid string formatting
warnings.  Fixes build with newer GCC, like the one in Snow Leopard.

(snj)

Pull in fixes from upstream revision 1.406 to avoid string formatting
warnings.  Fixes build with newer GCC, like the one in Snow Leopard.

(snj)

PR/42320: Alexander Nasonov: According to:
http://www.opengroup.org/onlinepubs/7990989775/xcu/awk.html
the LC_NUMERIC decimal point recognized is always period. Make it so.

(christos)

Fix unbalanced parentheses to make this build again.

(snj)

Pull in fixes from upstream revision 1.108 to avoid string formatting
warnings.  Fixes build with newer GCC, like the one in Snow Leopard.

(snj)

Argh.  Missed another now-unnecessary mhitch@ acknowledgement.

(snj)

Three ad clauses down: one for mhitch@ && is@, and two for mhitch@.

(snj)

Drop 3rd and 4th clauses (except in one case, where copyright is shared
with UCB and only the 3rd clause is dropped).  Approved by mhitch@ and is@
(copyright holders).

(snj)

mDNSResponder-214 out.

(wiz)

Pull in fixes from upstream revision 1.362 to avoid string formatting
warnings.  Fixes build with newer GCC, like the one in Snow Leopard.

(snj)

Pull in fixes from upstream revision 1.83 to avoid string formatting
warnings.  Fixes build with newer GCC, like the one in Snow Leopard.

(snj)

Fix all formats for 64bit builds

(dsl)

Declare trap name array as extern const char *const [] to match how
it's defined in trap.c. (Should really be in a header, of course.)
Compile-tested.

(dholland)

Fix borkeage

(dsl)

Remove PROTO(), which is like _P() only spelled differently.

The change causes gcc to harmlessly permute two instructions in the output
for some reason, but otherwise has no effect on the generated code.

(dholland)

Since this is a host program, fix some 64-bit cleanliness issues.

(dholland)

Need %zu for size_t

(dsl)

Declare trap name array as extern const char *[] to match how it's defined
in trap.c. (Should really be in a header, of course.) Compile-tested.

(dholland)

Reserve WSDISPLAY_TYPE for DisplayLink DL-1x0/DL-1x5 USB LCDs and
USB-VGA adaptors (which may require a new ioctl for DUMBFB ops). From
FUKAUMI Naoki.

(tsutsui)

Regen from usbdevs rev 1.537:
> Add IDs for DisplayLink DL-1x0/1x5 based USB LCDs and USB-VGA adaptors.
> Info from FUKAUMI Naoki's udl(4) driver patch.

(tsutsui)

Add IDs for DisplayLink DL-1x0/1x5 based USB LCDs and USB-VGA adaptors.
Info from FUKAUMI Naoki's udl(4) driver patch.

(tsutsui)

Reformat comment.

(skrll)

Add a period at the end of a sentence.

(pgoyette)

Get patch part of version from comment.

(dsl)

Use shell builtins instead of awk and sed.
Avoids problems with awk processing floating point numbers when LC_NUMERIC
give a decimal point of ','.
Fixes PR/42321

(dsl)

Put PMAP_NOCACHE in the MD space.

(skrll)

Pull across most of the changes from matt-nb5-mips64.

Only remaining change is bfd/elf.c:_bfd_elf_is_local_label_name which
should probably be done via mips_elf_is_local_label_name.

(skrll)

sync to NetBSD-ftp 20090915

(lukem)

merge NetBSD-20090915

(lukem)

Sync to 20090915:
* rename getline() to get_line() to avoid libc conflict with latter
* avoid NULL deref in error message

Status:

Vendor Tag: TNF
Release Tags: NetBSD-20090915

(lukem)

Crank version to 20090915:
* rename internal getline() to get_line() to avoid conflict with libc
* avoid NULL deref in verbose output in an error path
* improve man page markup

(lukem)

Add a commented out "encoding us.swapctrlcaps" line as a popular example.

(tsutsui)

Tickets 1134-1138, 1142, 1144, and 1146.

(snj)

Tickets 1134 and 1137.

(snj)

Pull up following revision(s) (requested by dholland in ticket #1146):
etc/rc.d/named: revision 1.20
PR/41273: Martin Mersberger: missing '/' in /etc/rc.d/named in section
named_migrate

(snj)

Pull up following revision(s) (requested by mbalmer in ticket #1144):
sys/dev/pckbport/wskbdmap_mfii.c: revision 1.18
Add support for Swiss german kezboard to wskbd.
Contributed by Marc Balmer.

(snj)

Pull up following revision(s) (requested by mlelstv in ticket #1142):
sys/netkey/keysock.c: revision 1.53
Replace splsoftnet() with proper mutexes as suggested by ad@.
Fixes PR kren/40940.

(snj)

Pull up following revision(s) (requested by jdc in ticket #1138):
sys/arch/sparc/sparc/pmap.c: revision 1.336
Avoid unused variable warnings on Sun4c-only kernels.
Fix applied from PR#42249 -- thanks!

(snj)

Pull up following revision(s) (requested by mhitch in ticket #1137):
sys/arch/alpha/alpha/pmap.c: revision 1.250
The tlb shootdown IPI can interrupt the pool_cache_get() in places that
can cause a deadlock or pool cache corruption.  Take the shootdown job
queue mutex before calling pool_cache_get(), which will block the IPI
interrupts and seems to fix the remaining tlb shootdown deadlocks and
pool cache corruption I've been seeing.  Should address both
PR port-alpha/38335 and PR port-alpha/42174.

(snj)

Pull up following revision(s) (requested by mhitch in ticket #1137):
sys/arch/alpha/alpha/pmap.c: revision 1.250
The tlb shootdown IPI can interrupt the pool_cache_get() in places that
can cause a deadlock or pool cache corruption.  Take the shootdown job
queue mutex before calling pool_cache_get(), which will block the IPI
interrupts and seems to fix the remaining tlb shootdown deadlocks and
pool cache corruption I've been seeing.  Should address both
PR port-alpha/38335 and PR port-alpha/42174.

(snj)

Pull up following revision(s) (requested by mhitch in ticket #1136):
distrib/alpha/instkernel/cd-hd-tape/Makefile: revision 1.20
The tarfs image in cdhdinstall used to get padded out to an 18KB boundary
(the size of the read used to read it) with the alpha-specific image
creation.  This got lost when moved to using common routines, so that
burning the cdhdinstall image onto a CDROM resulted in a failure to read
the last block of the tarfs image.  Use the provided option to have the
image padded out to the full image size.  It's more than actually needed,
but extra isn't likely to cause any problems.

(snj)

Pull up following revision(s) (requested by mhitch in ticket #1135):
sys/arch/alpha/conf/Makefile.alpha: revision 1.82
Loading the kernel at 0xfffffc0000300000 doesn't work on my ES40 when I've
got raid drives configured on the iop(4) adapter and the mlx(4) adapter.
Change the kernel text to 0xfffffc0000430000 (which is where Tru64 has its
kernel).

(snj)

Pull up following revision(s) (requested by rmind in ticket #1135):
sys/dev/rnd.c: revision 1.77
rnd_sample_allocate_isr: pass correct flag to pool_get().
Should fix PR/39325.

(snj)

Pull up following revision(s) (requested by rmind in ticket #1135):
sys/dev/rnd.c: revision 1.77
rnd_sample_allocate_isr: pass correct flag to pool_get().
Should fix PR/39325.

(snj)

Include miscfs/specfs/specdev.h for spec_init().

(elad)

Fix typo.

(matt)

one more printf like function, and fix the lossage.

(christos)

Add a gcc printf attribute to the printf functions and fix the lossage caught.

(christos)

- rmixls_subr.S is replaced by rmixl_subr.S

(cliff)

Switch from fu*/su* to ufetch_*/ustore_*.  This make netbsd32 compat root
on a LP64 BE kernel.

(matt)

switch from fu*/su* to ufetch_*/ustore_*.

(matt)

Fix comment.

(matt)

Add MIPS_SR_PX

(matt)

fix format string bug, filename may contain % character, don't use it as format string.

(tnozaki)

kpsignal2: do not make the signal pending twice when tracing the process,
also update a comment and add an assert.  Fixes PR/42309 by Nicolas Joly.

(rmind)

- Move kauth_init() a little bit higher.

- Add spec_init() to authorize special device actions (and passthru too for
  the time being). Move policy out of secmodel_suser.

(elad)

Sprinkle static.

(uebayasi)

Update a couple of comments.

(skrll)

Christos was worried about clrbits() being called with a length of zero.
This can't happen, but rework so it doesn't matter.
Remove 'optimisation' for length 1, that doesn't happen often enough.

(dsl)

automake

(lukem)

Ticket 1366

(sborrill)

Pull up the following revisions(s) (requested by joerg in ticket #1366):
sys/netinet6/ip6_forward.c: revision 1.67

Clear cksum flags before any further processing like ip_forward does.
Many drivers set the UDP/TCP v4 flags even for v6 traffic and if the
packet is encapsulated with gif, the IPv6 header would get corrupted by
ip_output.

(sborrill)

Ticket 1366

(sborrill)

Pull up the following revisions(s) (requested by joerg in ticket #1366):
sys/netinet6/ip6_forward.c: revision 1.67

Clear cksum flags before any further processing like ip_forward does.
Many drivers set the UDP/TCP v4 flags even for v6 traffic and if the
packet is encapsulated with gif, the IPv6 header would get corrupted by
ip_output.

(sborrill)

Use __packed, not __attribute__((__packed__)).

(njoly)

use __arraycount

(cegger)

Ticket 1139

(sborrill)

Pull up the following revisions(s) (requested by joerg in ticket #1139):
sys/netinet6/ip6_forward.c: revision 1.67

Clear cksum flags before any further processing like ip_forward does.
Many drivers set the UDP/TCP v4 flags even for v6 traffic and if the
packet is encapsulated with gif, the IPv6 header would get corrupted by
ip_output.

(sborrill)

Ticket 1139

(sborrill)

Pull up the following revisions(s) (requested by joerg in ticket #1139):
sys/netinet6/ip6_forward.c: revision 1.67

Clear cksum flags before any further processing like ip_forward does.
Many drivers set the UDP/TCP v4 flags even for v6 traffic and if the
packet is encapsulated with gif, the IPv6 header would get corrupted by
ip_output.

(sborrill)

include <sys/device_if.h> for device_t. There is no use of struct device.

(cegger)

Fix a wrong index value for edid_products[] inside #ifdef EDIDVERBOSE in
edid_findproduct(). From Yasushi Oshima via FUKAUMI Naoki in udl(4) patch.

(tsutsui)

regen

(lukem)

Preserve wscons(4) device numbers for forthcoming udl(4) USB display.

(tsutsui)

Convert to automake/libtool

(lukem)

Fix pppd for ipv6 link-local addresses. Change the prefix length to 72 bits.
Thanks to Nagae Hidetake for providing patches and testing.

(christos)

comcons in dev/ic/com.c is now declared as static, so don't peek it
directly but use it via cn_tab.

Tested on my Netra X1.

(nakayama)

1. PR/42209 various problems with EBCDIC support in iconv(3)
part3: added CP114[0-9] support.
2. fix csmapper name separator.

(tnozaki)

Simplify ifreq_setaddr:
- Drop the INET6 block. The commands are never given to this function
  and truncating the sockaddr is arguably not the desired result anyway.
- Clear the address before copying. This fixes SIOCGIFNETMASK and possible
  other ioctls for users that don't check sa_len. This includes
  COMPAT_43 and Linux emulation.

OK dyoung@

(joerg)

tlbstate is 'int', so use 'cmpl' not 'cmpq'.
Fixes gprof on amd64 PR/40960.

(dsl)

Return the result of copyout. Reminded by Niolas Joly.

(joerg)

Provide SIOCGIFNAME.

(joerg)

Ticket 1133

(sborrill)

Pull up the following revisions(s) (requested by jakllsch in ticket #1133):
sys/dev/ic/hme.c: revision 1.84 via patch

Use hme_init() to provide a non-NULL ifp->if_init.

(sborrill)

Ticket #1132.
Fix grammar error in previous ticket.

(sborrill)

Pull up the following revisions(s) (requested by mrg in ticket #1132):
external/mit/xorg/lib/dri/i915/Makefile: revision 1.4
external/mit/xorg/lib/dri/i965/Makefile: revision 1.5

Link in -ldrm_intel on i915 and i965 drivers.
Build i915 driver with -DI915 -DDRM_VBLANK_FLIP=DRM_VBLANK_FLIP.

(sborrill)

Ticket 1140

(sborrill)

Pull up the following revisions(s) (requested by cube in ticket #1140):
sys/arch/amd64/include/Makefile.inc: revision 1.1

Build kernel modules with -mno-red-zone like kernel is build. This fixes
frequent panics in amd64 zfs module, plus other reported problems.

(sborrill)

Ticket 1140

(sborrill)

Pull up the following revisions(s) (requested by cube in ticket #1140):
sys/arch/amd64/include/Makefile.inc: revision 1.1

Build kernel modules with -mno-red-zone like kernel is build. This fixes
frequent panics in amd64 zfs module, plus other reported problems.

(sborrill)

Ticket 1128

(sborrill)

Pull up the following revisions(s) (requested by bouyer in ticket #1128):
sys/nfs/nfs_socket.c: revision 1.182

Handle EWOULDBLOCK the same way as EPIPE. It seems the TCP socket layer
can return EWOULDBLOCK on some occasion when the connection is broken.

(sborrill)

reapply 64bit fixes

(macallan)

Fix clrbits() so that it doesn't mask no bits out of the byte after the
range (when the last bit to be cleared is the msb of a byte).
Fixes PR/42312 in a slightly better way than proposed.

(dsl)

Change args to clrbits() to be unsigned for efficiency.

(dsl)

Removed return statement in parser action that caused a memory leak with
the new yacc, making "sh MAKEDEV -MM init" consume 27 MB of virtual memory
with the result that NetBSD could no longer be installed on a 32 MB system.

(gson)

The ippool(5) manpage mentions ippool.conf (which is also more common
name for config files) - add it to distribution. Per <hubertf> suggestion.

(mishka)

Remove reference to now-moved binutils src directory, it's not needed
anymore since strings is commented out and has been for a long time.
... and remove the commented-out strings as well.

(he)

Point to the new ncdcs directory.

(he)

- add inline to simplify recognizing RMI Company ID
- add inlines to simplify use of RMI company specific flags in pridtab entry

(cliff)

- KASSERT this interrupt code being used on XLS family CPU

(cliff)

- define used to prevent recursive include gets renamed according to convention

(cliff)

- pridtab definition is removed to cpu.h
- pritab entries now initialize cpu_cp0flags and cpu_cidflags
except for RMI XLS entries, these are all 0 for now.
- pridtab entries for RMI XLS use those new fields and CPU_MIPS_HAVE_MxCR

(cliff)

- KSEG and XK labels, useful for address representation in ddb,
moved here from RMI specific code
- replace .word XXX with proper mnemonics for mfcr ops in cacheException
now that binutils knows these ops

(cliff)

- in SHOW64 macro remove extraneous 'x' in printf format
- in MIPS64_SHOW32 macro add push and pop directives
- use mycpu->cpu_cp0flags to determine whether various CP0 regs exist
- use CPU_MIPS_HAVE_MxCR to determine whether mfcr, mtcr insns are available
- use asm() to inline mfcr, mtcr ops instead of calling
MIPS-implementation-specific subroutines

(cliff)

- move #ifndef LOCORE up a few lines to wrap more XSEG, XKSEG stuff

(cliff)

- struct pridtab definition is moved here from mips_machdep.c
- 'mycpu' is now global (was static); uh, a name change might be nice?
- new bit for cpu_flags 'CPU_MIPS_HAVE_MxCR' denotes
mfcr, mtcr instructions are available on this CPU
- new field 'cpu_cp0flags' in pridtab tracks whether (some) CP0 regs exist
- define bits in cpu_cp0flags, including a "USE" bit that,
if clear means cpu_cp0flags can be ignored. common CP0 regs do not
need to be represented here, only newer optional ones are.
- new field 'cpu_cidflags' in pridtab allows defining company-specific flags
- some RMI company specific flags are defined to track chip family

(cliff)

- rmixls_subr.S is replaced by rmixl_subr.S
- those subroutine names changed accordingly

(cliff)

- (re-) configure ksyms
- SYMTAB_SPACE can now shrink by almost 10x now that local symbols
are properly stripped  (!)

(cliff)

Reduce the diff a little.

(uebayasi)

Don't use com_activate(), it's gone away.  Thanks to Andreas Wrede
for pointing out that these drivers still used com_activate().

Use device_private().  Join some lines.

Remove superfluous activation hooks.

Add child-detachment hooks (not used, yet).

TBD: device_t/softc split.

(dyoung)

Use TAILQ_FOREACH() instead of open-coding it.

I applied this patch with Coccinelle's semantic patch tool, spatch(1).
I installed Coccinelle from pkgsrc: devel/coccinelle/.  I wrote
tailq.spatch and kdefs.h (see below) and ran this command,

spatch -debug -macro_file_builtins ./kdefs.h -outplace \
    -sp_file sys/kern/tailq.spatch sys/kern/subr_autoconf.c

which wrote the transformed source file to /tmp/subr_autoconf.c.  Then I
used indent(1) to fix the indentation.

::::::::::::::::::::
::: tailq.spatch :::
::::::::::::::::::::

@@
identifier I, N;
expression H;
statement S;
iterator name TAILQ_FOREACH;
@@

- for (I = TAILQ_FIRST(H); I != NULL; I = TAILQ_NEXT(I, N)) S
+ TAILQ_FOREACH(I, H, N) S

:::::::::::::::
::: kdefs.h :::
:::::::::::::::

#define MAXUSERS 64
#define _KERNEL
#define _KERNEL_OPT
#define i386

/*
* Tail queue definitions.
*/
#define _TAILQ_HEAD(name, type, qual) \
struct name { \
qual type *tqh_first; /* first element */ \
qual type *qual *tqh_last; /* addr of last next element */ \
}
#define TAILQ_HEAD(name, type) _TAILQ_HEAD(name, struct type,)

#define TAILQ_HEAD_INITIALIZER(head) \
{ NULL, &(head).tqh_first }

#define _TAILQ_ENTRY(type, qual) \
struct { \
qual type *tqe_next; /* next element */ \
qual type *qual *tqe_prev; /* address of previous next element */\
}
#define TAILQ_ENTRY(type) _TAILQ_ENTRY(struct type,)

#define PMF_FN_PROTO1 pmf_qual_t
#define PMF_FN_ARGS1 pmf_qual_t qual
#define PMF_FN_CALL1 qual

#define PMF_FN_PROTO , pmf_qual_t
#define PMF_FN_ARGS , pmf_qual_t qual
#define PMF_FN_CALL , qual

#define __KERNEL_RCSID(a, b)

(dyoung)

Don't use com_activate(), it's gone away.  Delete superfluous
activation hook, pcmcom_activate().

Add a child-detachment hook and use it.

Use device_private().

Cosmetic: use __arraycount().  Join a couple of lines.  Delete a
set of superfluous parentheses.

(dyoung)

reapply 64bit fixes

(macallan)

Remove some dead code.

(dyoung)

Delete definition of com_activate(), it's been removed.

(dyoung)

Move the code in com_activate() to com_detach(), where it always
belonged.  Remove com_activate().

Consolidate information about the console on com(4) in a new struct
comcons_info.

Support detachment & re-attachment of a system console on com(4).
Re-attachment is somehow incomplete.  Ideally, if some other device
could take over as console, it would, but we're not quite there,
yet.

(dyoung)

Don't use com_activate(), it's going away.

(dyoung)

Don't use com_activate(), it's going away.

(dyoung)

Call com_detach() before doing any bus-specific detachment.

Don't use com_activate(), it's going away.

(dyoung)

Call the common detach routine, com_detach(), and get out on error,
before starting the bus-specific detachment.

com_activate() is going away, so don't use it any more.

(dyoung)

Remove definition of spc_activate(), it has gone away.

(dyoung)

Remove superfluous activation hook.

Add a child-detachment hook.

(dyoung)

Re-order operations in usb_detach() so that if a usb(4) instance's
children will not detach, the instance is not left in an inconsistent
state.

If uhub(4) port is disconnected, forcefully detach the children on
that port.

Simplify detachment hooks.  (sc_dying must die!)

Pass along and respect detachment flags, esp. DETACH_FORCE,
throughout.

(dyoung)

Simplify activation hooks.  (sc_dying must die!)

(dyoung)

Simplify activation hooks.  (sc_dying must die!)

(dyoung)

Simplify activation hook.  (sc_dying must die!)

unifdef -D__NetBSD__ -U__FreeBSD__ -U__OpenBSD__.

(dyoung)

Simplify activation hook.  (sc_dying must die!)

(dyoung)

Remove superfluous activation hook.

(dyoung)

Simplify activation hooks.  (sc_dying must die!)

(dyoung)

Simplify activation hook.  (sc_dying must die!)

s/device_ptr_t/device_t/.

(dyoung)

Simplify activation hook.  (sc_dying must die!)

unifdef -D__NetBSD__ -U__FreeBSD__ -U__OpenBSD__.

(dyoung)

Simplify activation hook.

(sc_dying must die!)

(dyoung)

Remove superfluous activation hooks.

(dyoung)

Remove superfluous activation hook.

(dyoung)

Remove superfluous activation hook.

(dyoung)