Backout the ED/QH change removal in [eo]hci_device_request.

Reports of the KASSERT firing mean this needs more thought/testing.

(skrll)

Fix build.

(htodd)

New sentence, new line.

(wiz)

nb_name_len does not increase size when nn_scope contains only a \0,
do the same on nb_name_encode to avoid a buffer overflow

(stacktic)

Make sure we keep enough space to encode the type

(stacktic)

XXXgcc Wuninitialized kludge

(pooka)

Allow overriding CTASSERT.  This helps with building NetBSD sources with
compilers that don't support __COUNTER__ -- shifting the CTASSERTs
around to avoid __LINE__ conflicts is a hopeless struggle.

(pooka)

Mention the new kern.module.autotime variable.

(pgoyette)

Make the auto-unload timeout configurable via sysctl, and if the timeout
is set to zero, disable all auto-unloads (even those that were "scheduled"
previously).

(pgoyette)

Fix output for -A -f ...

(mlelstv)

Correct a KASSERT's logic (SET vs. GET).

(jakllsch)

Move flags for ARM's APCS ABI into a common variable. Add some forces to
deal with limitations in the Clang <-> LLVM backend interaction to pick
the correct ABI and target assembler.

(joerg)

mark variable usage, fix non DIAGNOSTIC non DEBUG build

(para)

lcr is only used when COM_16650 is defined; unbreak the build

(mbalmer)

Only attempt to set 64byte fifo for 16750 when COM_16750 is defined.

Hi kiyohara.

PR/48359 NetBSD/cobalt 6.99.25 GENERIC kernel freeze at initializing com0

(skrll)

Coverity issues 996120 and 996121, Use after free

Use the M_CP_STROPT definition exclusive to servconf.c twice and
you have freed your original string.

servconf.h won copying authorized_keys_command and
authorized_keys_command_user in COPY_MATCH_STRING_OPTS in 1.107,
but servconf.c didn't drop its own, so it walks into this trap.
Remove the duplicate copies, and disarm the trap.

Note this is on a code path where authorized_keys_command and
authorized_keys_command_user don't actually get used except
for a debug dump of the config, and dump_cfg_string protects
itself against trying to print NULL pointers, so all
you get is sshd -T -C ... giving wrong results, which is rather
insignificant as far as security issues go.

(spz)

Now that usbd_new_device has been fixed there is no need to touch QHs/EDs
in [eo]hci_device_request.

KASSERT the address and maximum packet length just in case.

(skrll)

sync usage with man page

(wiz)

Sort SYNOPSIS and option descriptions. Bump date for previous.

(wiz)

Sync usage and option descriptions with manpage.

(wiz)

Standardize SYNOPSIS. Sort option descriptions and remove duplicates.

(wiz)

Trailing whitespace

(skrll)

Trailing whitespace.

(skrll)

New -A option. Add an output column with the address of the
kernel object (vnode or file) that can be matched with prstat(8) output.

(mlelstv)

Add -Wno-uninitialized with -Os so that we don't get spurious gcc warnings.

(christos)

PR/48448: David A. Holland: Avoid coredump by checking return code of
    localtime(3)

(christos)

Add extra rpcgen flags.

(christos)

Put back the extra prototype for main(); some programs #define main to
something else, so we need the extra prototype.

(christos)

Regen

(pgoyette)

Add ASMedia ASM1042 xHCI USB3 controller

(pgoyette)

Add linted comments.

(christos)

downgrade illegal bitfield from error to warning so we can ignore it.

(christos)

- remove support for pre-ansi output
- add -B (BSD) support for c++ output macros
- group functions to make output better looking

(christos)

centralize the rpcgen flags and set for BSD output

(christos)

only prevent autounload, not regular unload when we have exports

(christos)

Initialize capabilities in struct wsscreen_descr per omrasops settings.

Now REVERSE characters are drawn properly.

(tsutsui)

tickets 990, 992, 993, 994, 996

(bouyer)

tickets 994, 996

(bouyer)

tickets 994, 996

(bouyer)

Pull up following revision(s) (requested by spz in ticket #996):
sys/kern/uipc_syscalls.c: revision 1.163
PR/47591: Michael Plass: If the unix socket is closed before accept,
unp->unp_conn will be NULL in PRU_ACCEPT, as called from
sys_accept->so_accept. This will cause the usrreq to return with
no error, leaving the mbuf gotten from m_get() with an uninitialized
length, containing junk from a previous call. Initialize m_len to
be 0 to handle this case. This is yet another reason why Beverly's
idea of setting m_len = 0 in m_get() makes a lot of sense. Arguably
this could be an error, since the data we return now has 0 family
and length.

(bouyer)

Pull up following revision(s) (requested by spz in ticket #996):
sys/kern/uipc_syscalls.c: revision 1.163
PR/47591: Michael Plass: If the unix socket is closed before accept,
unp->unp_conn will be NULL in PRU_ACCEPT, as called from
sys_accept->so_accept. This will cause the usrreq to return with
no error, leaving the mbuf gotten from m_get() with an uninitialized
length, containing junk from a previous call. Initialize m_len to
be 0 to handle this case. This is yet another reason why Beverly's
idea of setting m_len = 0 in m_get() makes a lot of sense. Arguably
this could be an error, since the data we return now has 0 family
and length.

(bouyer)

Pull up following revision(s) (requested by spz in ticket #996):
sys/kern/uipc_syscalls.c: revision 1.163
PR/47591: Michael Plass: If the unix socket is closed before accept,
unp->unp_conn will be NULL in PRU_ACCEPT, as called from
sys_accept->so_accept. This will cause the usrreq to return with
no error, leaving the mbuf gotten from m_get() with an uninitialized
length, containing junk from a previous call. Initialize m_len to
be 0 to handle this case. This is yet another reason why Beverly's
idea of setting m_len = 0 in m_get() makes a lot of sense. Arguably
this could be an error, since the data we return now has 0 family
and length.

(bouyer)

Pull up following revision(s) (requested by nakayama in ticket #994):
sys/arch/sparc64/sparc64/locore.s: revision 1.350
Increase an interrupt depth only in the case of hardware interrupts,
and remove the ci_idepth trick in softint_fastintr.
Fixes the following diagnostic panic reported in port-sparc64.
  panic: kernel diagnostic assertion "!cpu_intr_p()" failed: file
  "../../../../kern/subr_xcall.c", line 351

(bouyer)

Pull up following revision(s) (requested by nakayama in ticket #994):
sys/arch/sparc64/sparc64/locore.s: revision 1.350
Increase an interrupt depth only in the case of hardware interrupts,
and remove the ci_idepth trick in softint_fastintr.
Fixes the following diagnostic panic reported in port-sparc64.
  panic: kernel diagnostic assertion "!cpu_intr_p()" failed: file
  "../../../../kern/subr_xcall.c", line 351

(bouyer)

Pull up following revision(s) (requested by nakayama in ticket #994):
sys/arch/sparc64/sparc64/locore.s: revision 1.350
Increase an interrupt depth only in the case of hardware interrupts,
and remove the ci_idepth trick in softint_fastintr.
Fixes the following diagnostic panic reported in port-sparc64.
  panic: kernel diagnostic assertion "!cpu_intr_p()" failed: file
  "../../../../kern/subr_xcall.c", line 351

(bouyer)

Pull up following revision(s) (requested by hauke in ticket #993):
sys/compat/linux/common/linux_misc.c: revision 1.227
On linux_sys_getdents, insert d_type at the end of each record.
Fixes PR kern/47806.

(bouyer)

Pull up following revision(s) (requested by kefren in ticket #992):
sys/netinet/tcp_usrreq.c: revision 1.170
Update TCP CB with new values on rfc1323 and mssdflt sysctl updates
=46rom yasuoka-cj7TXg5MjN14Eiagz67IpQ@public.gmane.org in kern/44254

(bouyer)

Make omrasops copycols() op work even if columns are not 32 bit aligned.

This means command line editing works properly in any cases.

The bitcopy strategies for 1bpp copycols() op are taken from
recently fixed MI sys/dev/rasops/rasops_bitops.h.
GETBITS() and PUTBITS() m68k asm macro are taken from hp300.

Tested on both 1bpp (on LUNA-II) and 4bpp (on LUNA) framebuffers.

(tsutsui)

Pull up following revision(s) (requested by skrll in ticket #990):
sys/dev/usb/ehci.c: revision 1.221
In ehci_check_qh_intr don't treat a short control transfer as done if the
status phase is still inflight.  Let the hardware complete it.
PR/48358: Repeated low-speed USB control transfers returning short data
  fail on EHCI
PR/46696: uhub disables port where USB keyboard attached

(bouyer)

It's been many years since realloc(NULL, size) == malloc(size). Also don't
destroy pointers on allocation errors so someone can free them later.

(christos)

knf, reduce pointer gymnastics

(christos)

Regen.

(njoly)

Use native/rump syscalls for pread/pwrite instead of linux ones
which only exist to deal with PAD argument.

(njoly)

don't allow the nfs server module to unload if it has exported filesystems.

(christos)

In xhci_init() use aprint_debug_dev() instead of device_printf().

(jakllsch)

If we can't enable VFP/VFP2 via the CPACCESS register, bail since there
isn't a VFP.

(matt)

Regen.

(njoly)

Add syscall wrapper for mknodat(2) to deal with PAD argument difference
between rump and linux versions.

(njoly)

Comment out unsed variable. Left for reference.

(skrll)

Remove unused variables.

(skrll)

Remove unused variables.

(skrll)

Sprinkle __unused.

(skrll)

Add SYS_compat_60__lwp_park to the list of syscalls that can be resolved by loading kernel modules.

This seems to address my PR kern/48027

(pgoyette)

Remove duplicate/unused declarations.

(nakayama)

Put previous (RTK_HWREV_8168F) in numeric sort order like its neighbors.

(jakllsch)

Ticket #1891

(sborrill)

Pull up the following revisions(s) (requested by spz in ticket #1891):
sys/kern/uipc_syscalls.c: revision 1.163

If the unix socket is closed before accept, the mbuf returned by
m_get() will have an uninitialized length and contain junk from a
previous call. Initialize m_len to be 0 to handle this case.
Fixes PR/47591

(sborrill)

time_t_{min,max} are only used in localtime.c

(christos)

add a cast for sparc64 where int_fast32_t is long (should it be?), and
time_t is int32_t (for compat)

(christos)

use CPU_INFO_ITERATOR instead of int.

(christos)

it's called netbsd-INSTALL64.gz for pmax64.

(mrg)

update to reality for gcc 4.8.2.

(mrg)

Ticket #1890

(sborrill)

Pull up the following revisions(s) (requested by bouyer in ticket #1890):
sys/arch/xen/xen/evtchn.c: revision 1.70

Remove the "evtchn_do_event: handler %p didn't lower ipl %d %d\n" printf
as analysis shows it actually isn't a bug in the handler, but related to
spin mutexes. Fixes port-xen/46313

(sborrill)

Ticket #1890

(sborrill)

Pull up the following revisions(s) (requested by bouyer in ticket #1890):
sys/arch/xen/xen/evtchn.c: revision 1.70

Remove the "evtchn_do_event: handler %p didn't lower ipl %d %d\n" printf
as analysis shows it actually isn't a bug in the handler, but related to
spin mutexes. Fixes port-xen/46313

(sborrill)

Cosmetics: use "msg" instead of "echo"

(martin)

regen

(pedro)

Add entries for Atheros AR9462 and a new flavour of Intel Haswell
Integrated Graphics Device; OK rmind@ mbalmer@

(pedro)

Don't strip local symbols for earmv7*eb images since that will break linking.

(matt)

Since regs->fxstate is actually of type 'struct fxsave64' use a structure
  copy instead of a 'not obviously incorrect' memcpy().

(dsl)

Don't define ptrdiff_t since that's done in the NetBSD <sys/types.h>

(matt)

Fix a bug in the divby0 case.

(matt)

Make ptrdiff_t available in the kernel.

(matt)

regen

(nonaka)

Added some MediaTek wifi adapters.

(nonaka)

regen

(nonaka)

Added Radeon HD6320 Graphics.

(nonaka)

Add stack alignement test for sparc

(martin)

regen

(pooka)

need to grab KTRFACv[n] too

(pooka)

Remove the fields that were used to save the i387 fp state on interrupt.
They were written but never read.
Possibly they should be saved for 32 bit processes, but that might be a relic
  from real i387 where the fpu was actully asynchronous.

(dsl)

Add stack alignment checks for alpha, arm, mips, powerpc, and vax.
Fix makefile to support MACHINE_CPU

(matt)

Ticket #1890

Fix typo in previous ticket

(sborrill)

Pull up the following revisions(s) (requested by bouyer in ticket #1890):
sys/arch/xen/xen/evtchn.c: revision 1.70

Remove the "evtchn_do_event: handler %p didn't lower ipl %d %d\n" printf
as analysis shows it actually isn't a bug in the handler, but related to
spin mutexes. Fixes port-xen/46313

(sborrill)

remove duplicate

(dholland)

another minor fix

(dholland)

Reinstate typo since kre claims it's by purpose.
Discussion in PR 48442.

(wiz)

Error and typo fixes from OpenBSD and FreeBSD via Henning Petersen
in PR 48443.

While here, remove a duplication, fix another typo, and
don't claim that Mr. Sheperd was the first man in space.

(wiz)

Error and typo fixes from OpenBSD and FreeBSD via Henning Petersen
in PR 48442.

(wiz)

reregengen

(pooka)

create rump kernel wrapper for utrace() too

(pooka)

Make sure the __CTOR_LIST__ is just aligned to a pointer boundary.
By default, mips N32 will aligned to a 64-bit boundary not 32-bit
which causes an extra NULL entry to be added.

(matt)

Remove entirely bogus /* NOTREACHED */ annotation.

(dholland)

Make new function named tsc_is_invariant() to avoid code duplication.
The behavior of acpicpu_md_flags() will change on some CPUs because
the detecting code of invariant TSC is replaced with newer code.

(msaitoh)

Moved to src/common.

(joerg)

Allow kernel code to access constant databases by moving cdbr(3) and the
required mi_vector_hash(3) into src/common.

(joerg)

Repeat after me: '\0' is *not* a pointer.  Use NULL instead.

(riz)

CID 1135779: Fix resource leak

(christos)

Add a few comments about some perversities of the xhci specification.
XXX: I can't see any code that ensures there is space in the rings.
XXX: Nothing stops buffer fragments crossing 64kB boundaries.

(dsl)

CID 975205: Unchecked return value

(christos)

Fix bad cut'n'paste in previous commit.
This should fix the remaining loogson build issues.

(bouyer)

shut up an unused variable warning that shows up in UP kernels

(macallan)

Move alignment attribute onto the types, not the variable instances.
Drop (unused) packed attribute.

(joerg)

Regen.

(njoly)

Add *at() syscalls like other arches.

(njoly)

Fix the build setlists for MKLINT=yes + MKRUMP=yes

(riz)

Explicitly state which partition was added/modified.

(jnemeth)

regen

(pooka)

one newline is enough

(pooka)

need string.h now for !_KERNEL

(pooka)

Now that Makefile.rump is included here first, do not override any
SRCS it adds.  This is required for linksyms_sun.c used when linking
with the Solaris linker.

(pooka)

Adjust revision list for #1889

(bouyer)

Adjust revision list for #982

(bouyer)

Pullup the following revisions, requested by pettai in ticket #982:
usr.bin/gzip/zmore: revision 1.5
fix == compatibility problem

(bouyer)

Pullup the following revisions, requested by pettai in ticket #1889:
usr.bin/gzip/zmore: revision 1.5
fix == compatibility problem

(bouyer)

Reenable calls to KPREEMPT_EN/DISABLE().  I'm almost certain I commented
them out accidentally in rev 1.30.

... not that they do much in rump kernels which are always
run-to-completion, but maybe there's some driver debugging value.

(pooka)

regen

(pooka)

include ktrace constants

(pooka)

regen

(pooka)

Initialize the syscall "frame" to 0 before setting parameters.
Otherwise things which treat syscall parameters as register_t (like
ktrace) will encounter garbage for parameters which are of smaller size
than register_t.  Using memset is probably not the most optimal way,
but oh well.

(pooka)

As long as we're using #ifdef to decide which stubs get generated, might
as well use __strong_alias instead of __weak_alias.  Some toolchains
such as the cygwin pecoff one get weak aliases a bit wrong, so avoiding
unnecessary weak alises helps there.

(pooka)

Make ktrace a compile-time option

(pooka)

include necessary headers

(pooka)

Include machine/npx.h here.
The only file that needs 'struct freebsd_save87' includes it already.

(dsl)

regen

(pooka)

Create rump kernel wrappers for {f,}ktrace

(pooka)

make !LOCKDEBUG work too

(pooka)

need ktrace.h now

(pooka)

Support ktrace for rump kernels.

Requested by Justin Cormack on rumpkernel-users.

(pooka)

Group more syscall related routines together (kern_subr -> kern_syscall)

(pooka)

Put vfs bits of ktrace into kern_ktrace_vfs.c per convention.

(pooka)

regen

(pooka)

Echo generated file names so that you can just copypaste & cvs ci
without having to resort to the blazingly fast help of cvs diff.

(pooka)

Rump kernels now use sy_invoke(), which initializes the return
value registers, so no need to do it individually in each wrapper.

(pooka)

Use sy_invoke() instead of sy_call() directly.

(pooka)

stop ktrace at process exit

(pooka)

Fix typo ("then" instead of "than")

(wiz)

Fix typo.

(wiz)

Missing change to the fxsave structure rename

(dsl)

Add two new subcommands, "set" and "unset".  These were inspired by
FreeBSD's gpart(8), but the code is all mine.  The purpose of these
is to set and unset partition attributes.

(jnemeth)

For the "show -i <entry>" subcommand, print Start and Size both in
terms of number of sectors and bytes.

(jnemeth)

add note about gets()

(dholland)

Rename the fields of 'struct envxmm' (itself badly named) so that they
  are different from those used for the x87 fpu.
Use a union to describe the 64bit 'address' field so that it correctly
  spilt for 32bit binaries.
The freebsd emulation defined its own 'struct freebsd_env87' that the
  code just assumed matched the netbsd one.
  Since this is a hardware structure, just use the netbsd one.
This shouldn't change the binaries except that 'en_fos' used to be
  incorrectly a 32bit field.

(dsl)

Remove the now-unused CPU_MAXMODEL and CPU_DEFMODEL

(dsl)

Add some definitions for cpu 'extended state'.
These are needed for support of the AVX SIMD instructions.
Nothing yet uses them.

(dsl)

Regen.

(njoly)

Sprinkle a few more linux_umode_t where appropriate.
Reduce diffs between archs.

(njoly)

Add missing syscall argument fd in comment.

(njoly)

Add support for running ofwboot on sun4v. Please note that the kernel on sun4v will be loaded properly but when the kernel takes over it will crash immediately since the kernel is not sun4v ready yet

(palle)

Mark rv as unused (for now)

(skrll)

Cosmectic: use the same define in all places when referencing the BAR.
No functional change.

(martin)

Use a double linked list with a static head to track MMU entries.
Code from martin@.
Tested on SUN4 (4/330), SUN4C (SS2), and SUN4M (Krups, 4/630).

(jdc)

For the add and resize subcommands, change the -s option.  If there is
no suffix, or the suffix is 's' or 'S', size is in sectors (as before)
otherwise size is in bytes.

(jnemeth)

Regen.

(njoly)

Add support for *at syscalls.

(njoly)

more complete fix for Coverity issue 976653

(spz)

fix off-by-one error

(jnemeth)

regen.

(msaitoh)