Style, and four fixes:

* Remove the (disabled) IPPROTO_ESP check. If the packet was decrypted it
  will have M_DECRYPTED, and this is already checked.

* Memory leaks in icmp6_error2. They seem hardly triggerable.

* Fix miscomputation in _icmp6_input, the ICMP6 header is not guaranteed
  to be located right after the IP6 header. ok mlelstv@

* Memory leak in _icmp6_input. This one seems to be impossible to trigger.

(maxv)

if 0 out unused code which is currently breaking my local builds

(maya)

Make the tests pass once more when __STDC_ISO_10646__ is not defined.

(perseant)

Add missing NULL-checking for m_pullup (CID 1427770: Null pointer dereferences (NULL_RETURNS))

(ozaki-r)

KNF: replace soft tabs with hard tabs

(ozaki-r)

Fix late NULL-checking (CID 1427782: Null pointer dereferences (REVERSE_INULL))

(ozaki-r)

Don't attempt to dereference cpu_infos if ncpus == 0.
Instead use the already initialized cpu_info_store.

(Also, now we assume all ncpus have cpu_infos initialized. seems to work.)

fixes PR port-mips/52940: ERLITE multiprocessor hangs early

(maya)

xen has separate list, update here also for x86/pmap_tlb.c -> x86/x86_tlb.c

(jdolecek)

Fix RCSID (hopefully)

(maya)

1550, 1554

(snj)

1550, 1554

(snj)

1550, 1554

(snj)

Pull up following revision(s) (requested by maya in ticket #1554):
bin/ksh/history.c: revision 1.18 via patch
Use 0600 as the mode for histfile here too.
pointed out by John D. Baker in PR bin/52480

(snj)

Pull up following revision(s) (requested by maya in ticket #1554):
bin/ksh/history.c: revision 1.18 via patch
Use 0600 as the mode for histfile here too.
pointed out by John D. Baker in PR bin/52480

(snj)

Pull up following revision(s) (requested by maya in ticket #1554):
bin/ksh/history.c: revision 1.18 via patch
Use 0600 as the mode for histfile here too.
pointed out by John D. Baker in PR bin/52480

(snj)

Pull up following revision(s) (requested by maxv in ticket #1550):
sys/arch/amd64/amd64/machdep.c: revision 1.280 via patch
sys/arch/amd64/include/segments.h: revision 1.34 via patch
sys/arch/i386/i386/machdep.c: revision 1.800 via patch
sys/arch/i386/include/segments.h: revision 1.64 via patch
sys/arch/x86/x86/vm_machdep.c: revision 1.30 via patch
Fix a huge privilege separation vulnerability in Xen-amd64.
On amd64 the kernel runs in ring3, like userland, and therefore SEL_KPL
equals SEL_UPL. While Xen can make a distinction between usermode and
kernelmode in %cs, it can't when it comes to iopl. Since we set SEL_KPL
in iopl, Xen sees SEL_UPL, and allows (unprivileged) userland processes
to read and write to the CPU ports.
It is easy, then, to completely escalate privileges; by reprogramming the
PIC, by reading the ATA disks, by intercepting the keyboard interrupts
(keylogger), etc.
Declare IOPL_KPL, set to 1 on Xen-amd64, which allows the kernel to use
the ports but not userland. I didn't test this change on i386, but it
seems fine enough.

(snj)

Pull up following revision(s) (requested by maxv in ticket #1550):
sys/arch/amd64/amd64/machdep.c: revision 1.280 via patch
sys/arch/amd64/include/segments.h: revision 1.34 via patch
sys/arch/i386/i386/machdep.c: revision 1.800 via patch
sys/arch/i386/include/segments.h: revision 1.64 via patch
sys/arch/x86/x86/vm_machdep.c: revision 1.30 via patch
Fix a huge privilege separation vulnerability in Xen-amd64.
On amd64 the kernel runs in ring3, like userland, and therefore SEL_KPL
equals SEL_UPL. While Xen can make a distinction between usermode and
kernelmode in %cs, it can't when it comes to iopl. Since we set SEL_KPL
in iopl, Xen sees SEL_UPL, and allows (unprivileged) userland processes
to read and write to the CPU ports.
It is easy, then, to completely escalate privileges; by reprogramming the
PIC, by reading the ATA disks, by intercepting the keyboard interrupts
(keylogger), etc.
Declare IOPL_KPL, set to 1 on Xen-amd64, which allows the kernel to use
the ports but not userland. I didn't test this change on i386, but it
seems fine enough.

(snj)

Pull up following revision(s) (requested by maxv in ticket #1550):
sys/arch/amd64/amd64/machdep.c: revision 1.280 via patch
sys/arch/amd64/include/segments.h: revision 1.34 via patch
sys/arch/i386/i386/machdep.c: revision 1.800 via patch
sys/arch/i386/include/segments.h: revision 1.64 via patch
sys/arch/x86/x86/vm_machdep.c: revision 1.30 via patch
Fix a huge privilege separation vulnerability in Xen-amd64.
On amd64 the kernel runs in ring3, like userland, and therefore SEL_KPL
equals SEL_UPL. While Xen can make a distinction between usermode and
kernelmode in %cs, it can't when it comes to iopl. Since we set SEL_KPL
in iopl, Xen sees SEL_UPL, and allows (unprivileged) userland processes
to read and write to the CPU ports.
It is easy, then, to completely escalate privileges; by reprogramming the
PIC, by reading the ATA disks, by intercepting the keyboard interrupts
(keylogger), etc.
Declare IOPL_KPL, set to 1 on Xen-amd64, which allows the kernel to use
the ports but not userland. I didn't test this change on i386, but it
seems fine enough.

(snj)

rename sys/arch/x86/x86/pmap_tlb.c to sys/arch/x86/x86/x86_tlb.c, so that
x86 can eventually use uvm/pmap/pmap_tlb.c; step to future PCID support

(jdolecek)

Use right variable as revealed by previous typo...

(flxd)

- Coverity does not like -std=gnu99
- Another lose is that you need to manually edit the compiler XML to add
  all the kernel options because cov-configure barfs because it tries to
  test the compilation environment thinking that we are userland even
  when options like -ffreestanding are present.

(christos)

white space for readability.

(christos)

Disgusting and evil hack to elide coverity options that break production
of coverity output when present [userland portion]

    - --sysroot
    - -fstack-protector
    - -std=gnu99
    - --nostdinc
    - -fPIE

The critical one here being --sysroot because this means that we need to
use the actual system headers while building (or chroot/sandbox appropriately)

(christos)

Style and clarify.

(maxv)

Fix memory leak, looks like there is still something wrong here.

(maxv)

Pass PMAP_WRITE_COMBINE as appropriate in armv7_generic_bs_map, i.e.
same fix as bcm283x_bs_map just got.

(skrll)

Trailing whitespace

(skrll)

Pass PMAP_WRITE_COMBINE as appropriate in bcm283x_bs_map.

RPI2 FB console now updates properly again

(skrll)

Add PMAP_WRITE_COMBINE: to the list of flags supported by pmap_kenter_pa

(skrll)

Trailing whitespace

(skrll)

Tickets #505 - #508

(martin)

Pull up following revision(s) (requested by mlelstv in ticket #508):
sys/fs/msdosfs/msdosfs_vfsops.c: revision 1.129
relax sanity check. It's ok to have more FAT sectors than needed.

(martin)

Pull up following revision(s) (requested by skrll in ticket #507):
sys/arch/arm/arm/cpufunc_asm_arm11x6.S: revision 1.10
PR/52934: Yasushi Oshima: Apply the erratum fix that was applied to wbinv_range
to isync_range so that we don't hang when we try to sync from execcmd_readvn().
XXX: pullup 8

(martin)

Pull up following revision(s) (requested by jdolecek in ticket #506):
sys/kern/kern_softint.c: revision 1.45
sys/rump/librump/rumpkern/rump.c: revision 1.331
sys/kern/subr_pserialize.c: revision 1.10
sys/kern/subr_psref.c: revision 1.10
Prevent panic or hangup in softint_disestablish(), pserialize_perform() or
psref_target_destroy() while mp_online == false.
See http://mail-index.netbsd.org/tech-kern/2017/12/25/msg022829.html
Set mp_online = true. This change might fix PR#52886.

(martin)

Pull up following revision(s) (requested by maya in ticket #505):
bin/ksh/history.c: revision 1.18
Use 0600 as the mode for histfile here too.
pointed out by John D. Baker in PR bin/52480

(martin)

Sync usage with man page.

(wiz)

m_split does not 'attempt' to restore the chain, it just restores it plain
and simple.

(maxv)

Add RCS Id, simplify wording, sort.

The option arguments need to be properly described.

(wiz)

Improve the mount_autofs(8) stub documentation

Improve DESCRIPTION and NAME.

(kamil)

m_prepend does not tolerate being given len > MHLEN, so add a panic, and
document this behavior.

(maxv)

Fix null deref, m could be NULL if M_PREPEND fails.

(maxv)

Add mount_autofs(8) from DragonFly BSD

(kamil)

Style, no functional change.

(maxv)

Ah, remove duplicate SVS_LEAVE. Fixes 32bit binaries. While here remove
duplicate 'cli', but that's harmless.

(maxv)

Fix m_prepend(). If 'm' is not a pkthdr, it doesn't make sense to use
MH_ALIGN, it should rather be M_ALIGN.

I'm wondering whether there should not be a KASSERT to make sure 'm' is
always a pkthdr.

(maxv)

Add KASSERTs in *_ALIGN: ensure the mbuf is of the correct type, and
also make sure m->m_data points at the beginning of the mbuf.

(maxv)

Adapt previous, reintroduce MH_ALIGN. It's used as an optimization - we
can later prepend something to the current mbuf without having to allocate
a new mbuf.

(maxv)

Ignore pragma warnings with clang for the zfs code.

(joerg)

use __typeof__ instead of typeof

(christos)

use __typeof__ instead of typeof.

(christos)

PR/40491: From Tobias Ulmer in tech-kern@:
1. Protect the nfs request queue with its own mutex
2. make the nfs_receive queue check for signals so that intr mounts
  can be interrupted.
XXX: pullup-8

(christos)

Add files missing from previous commit

(perseant)

regen

(sevan)

Add the GeForce4 MX440 Go, as found on the 17" flat panel G4 iMac (PowerMac4,5).

Thanks to Jasper Wallace <jasper AT pointless DOT net> for the loan of the hardware.

(sevan)

CID-1427736: Appease coverity using KASSERT

(christos)

CID-1427737: Pacify coverity using KASSERT

(christos)

CID-1427745: kill possible buffer overflows.

(christos)

CID-1427760: While overrun can't happen, make it obvious that it can't

(christos)

CID-1427768: Off by one

(christos)

CID-1364351: Fix uninitialized warnings.

(christos)

fix typo

(flxd)

Switch sp_timoff to u_int16_t, to prevent possible overflow in
ieee80211_recv_mgmt_beacon(). Actually this field is unused.

(maxv)

Appease the overflow check, 4 is enough.

(maxv)

Add KASLR.

(alnsn)

PR kern/52931 Kernel panics with Atheros usb wireless interface

Audit the flags to usbd_create_xfer so that USBD_FORCE_SHORT_XFER is
supplied wherever such a transfer is setup.  We can drop
USBD_SHORT_XFER_OK as it has not bearing on number of TDs

(skrll)

Unmap the kernel from userland in SVS, and leave only the needed
trampolines. As explained below, SVS should now completely mitigate
Meltdown on GENERIC kernels, even though it needs some more tweaking
for GENERIC_KASLR.

Until now the kernel entry points looked like:

FUNC(intr)
pushq $ERR
pushq $TRAPNO
INTRENTRY
... handle interrupt ...
INTRFASTEXIT
END(intr)

With this change they are split and become:

FUNC(handle)
... handle interrupt ...
INTRFASTEXIT
END(handle)

TEXT_USER_BEGIN
FUNC(intr)
pushq $ERR
pushq $TRAPNO
INTRENTRY
jmp handle
END(intr)
TEXT_USER_END

A new section is introduced, .text.user, that contains minimal kernel
entry/exit points. In order to choose what to put in this section, two
macros are introduced, TEXT_USER_BEGIN and TEXT_USER_END.

The section is mapped in userland with normal 4K pages.

In GENERIC, the section is 4K-page-aligned and embedded in .text, which
is mapped with large pages. That is to say, when an interrupt comes in,
the CPU has the user page tables loaded and executes the 'intr' functions
on 4K pages; after calling SVS_ENTER (in INTRENTRY) these 4K pages become
2MB large pages, and remain so when executing in kernel mode.

In GENERIC_KASLR, the section is 4K-page-aligned and independent from the
other kernel texts. The prekern just picks it up and maps it at a random
address.

In GENERIC, SVS should now completely mitigate Meltdown: what we put in
.text.user is not secret.

In GENERIC_KASLR, SVS would have to be improved a bit more: the
'jmp handle' instruction is actually secret, since it leaks the address
of the section we are jumping into. By exploiting Meltdown on Intel, this
theoretically allows a local user to reconstruct the address of the first
text section. But given that our KASLR produces several texts, and that
each section is not correlated with the others, the level of protection
KASLR provides is still good.

(maxv)

Increase the size of the initial mapping of the kernel. KASLR kernels are
bigger than their GENERIC counterparts, and the limit will soon be hit on
them.

(maxv)

Try to fix previous to make it compile.

(martin)

fix some logic errors in the previous and fix non-kernel builds.

(mrg)

fix ofppc/pegasosII (and maybe others).

don't assume PPC_OEA64_BRIDGE means we have a 64 bit cpu (code
for 64 bit in bridge and normal 32 bit can co-exist due to
fixups the early boot code does has, and ofppc builds GENERIC
this way):
- fix mtmsr()/mfmsr() to use the right method based upon the
  actually cpu booted on.
- fix cpu_setup() to have 32 bit and 64 bit hid0 variables
  and operate on the right one based upon the current cpu.
  restore a minor optimisation of not writing hid0 if it
  didn't change.

in set_timebase() check if OF_finddevice("/cpus/@0") failed
and returned -1 before using it for OF_getprop().

(mrg)

Fix the build, on Xen too amd64_trap.S needs to be compiled independently.

(maxv)

Make it possible for SVS to map in the user page tables a 4K kernel page
contained in a 2MB large page. Will be used soon.

(maxv)

CID-1427771: Remove incorrect pre-condition (index cannot be 256!)

(christos)

CID-1427779: Fix uninitialized variables

(christos)

make check cleaner, and simplify the code.

(christos)

Correct wrong assertion code introduced by rev 1.41:
http://cvsweb.netbsd.org/bsdweb.cgi/src/sys/dev/wscons/wsemul_vt100.c#rev1.41

Fix kernel panic reported in PR kern/52935.

(rin)

Ack EOI for IRQ_SLAVE of the master PIC after ack for IRQ of the slave PIC.

I'm not sure if there are possible races in the original code, but
this is what i8259_asm_ack2() in sys/arch/x86/include/i8259.h does
and it looks this change makes a Milan kernel a bit stable.

(tsutsui)

Use multilevel tables (tries) to convert kuten <-> unicode.
Get rid of dchains, replace with chain/rchain conversions through keys
allocated from the PUA.

(perseant)

Skip NVRAM checksum check and re-initialization on Milan.

Milan's firmware seems to use different check method.

(tsutsui)

Regen from MILAN.in rev 1.28.

> Restore piixide(4) for Milan and disable other pciide devices.

(tsutsui)

Restore piixide(4) for Milan and disable other pciide devices.

piixide was removed in rev 1.18 and the log message said
"because Intel IDE disk controllers only exist as part of
Intel chipsets for x86 systems" but the Milan actually has
the Intel 82371FB southbridge on its board.
Other pciide devices are unlikely necessary for the default
kernel for such a rare machine.
Also fix pasto in comment.

(tsutsui)

Use a proper PSL value to be passed to splx(9) functions.

This should have been changed on yamt-splraiseipl branch merge
back in 2006, which made MI IPL_xxx values independent from
m68k MD PSL values for the %sr register.

(tsutsui)

Handle yet another atari specific quirk in the MI ISA fdc(4)/fd(4) driver.

This makes fd(4) drive(s) (which is necessary for installation)
properly attached on Milan.

atari uses "fdcisa" and "fdisa" for ISA fdc to co-exist other fd(4)
drivers, on-board (atari/dev/fd.c) one and Hades (atari/dev/hdfd.c) one.

(tsutsui)

Fix silent hang after isa_intr_establish() on Milan.

The problems (wrong macro replacements) were slipped in rev 1.107:
http://cvsweb.netbsd.org/bsdweb.cgi/src/sys/arch/atari/atari/locore.s#rev1.107

(tsutsui)

Fix silent hang during config_console() (before consinit()) on Milan.

config_console() was a dirty hack used by ancient m68k ports
to probe and initialize console devices before "real" configure(9),
using subset of configure(9) functions.

In that case, most device specific data (except I/O access method)
are not initialized so we must not access device specific device_t and
softc structures in config_console() cases.

(tsutsui)

Fix I/O access failures for regions allocated by bus_space_map(9) on Milan.

Passing to PMAP_WIRED against I/O spaces seems problematic,
probably after yamt-km branch merge, which was committed
between NetBSD 3.0 and NetBSD 4.0.
(i.e. ISA and PCI devices on Milan didn't work after 4.0 release)

XXX:
According to pmap(9) man page, the "flags" arg for pmap_enter(9)
doesn't take VM_PROT_READ and VM_PROT_WRITE, but pmap_enter()
implementation in sys/arch/m68k/m68k/pmap_motolora.c historically
checks them.

(tsutsui)

Regenerate config files to update RCSIds in "Created from:" lines.

(tsutsui)

PR/52934: Yasushi Oshima: Apply the erratum fix that was applied to wbinv_range
to isync_range so that we don't hang when we try to sync from execcmd_readvn().
XXX: pullup 8

(christos)

Use .pushsection/.popsection, we will soon embed macros in several layers
of nested sections.

(maxv)

Compile amd64_trap.S as a file instead of including it.

(maxv)

Start with .text not to inherit the last section of amd64_trap.S, and
remove outdated #define.

(maxv)

Switch cobalt to common bus_space.

Tested on raq 2

(skrll)

Eliminate a '.text'.

(maxv)

Don't declare exceptions[] with IDTVEC, it's an array, not a function.
Rename it to x86_exceptions[], and move it to .rodata.

(maxv)

Mmh, restore PG_G on the direct map, we still want that in the non-SVS
case.

(maxv)

Fix the double-fault handler. We're executing on ist1 and must not jump
out of it, so don't enable interrupts. And use the SVS_*_ALTSTACK macros.

While here, fix the NMI handler too: it should use SVS_LEAVE_ALTSTACK.

(maxv)

Improve two comments and a KASSERT.

(maxv)

Add a copyright (approx 16.5 years later).

(simonb)

fixes for the previous, noted by nakayama@.

- RAIDFRAME_CONFIGURE needs to be versioned as the rows was removed,
  adding RAIDFRAME_CONFIGURE80, rf_config80() etc.
- RAIDFRAME_CONFIGURE32 changes to match
- rf_get_info80() passed the wrong source to copyout()

some fixes to my original change were independantly made by nakayama@
who confirmed the changes work properly now.

(mrg)

add conversion goop for WSDISPLAYIO_LDFONT and _SFONT
tested on mips64

(macallan)

enable font loading

(macallan)

don't forget to eat the high bits.

(christos)

Several changes:

* Declare TRIM_LABEL as a function.

* In mpls_unlabel_inet, copy the label locally. It's not incorrect to
  keep a pointer on the mbuf, but it's bug-friendly.

* In mpls_label_inetX, fix the length check. Meanwhile add an XXX: we
  just want to make sure that m_copydata won't fail, but if we were
  guaranteed that m has M_PKTHDR set, we could simply check the length
  against m->m_pkthdr.len.

(maxv)

Fix build failure, the structure is already defined now.

(maxv)

Add XXX.

(maxv)

Fix a buffer overflow in icmp_error. We create in 'm' a packet that must
contain:

  IPv4 header | Fixed part of ICMP header | Variable part of ICMP header

But we perform length checks on 'totlen', which does not count the IPv4
header.

So now, add sizeof(struct ip) in totlen, and stop doing this m_data
nonsense, just get the pointers as usual.

(maxv)

Clarify icmp_error:

* Rename (and constify) oiplen -> oiphlen.

* Rename icmplen -> datalen, it's the size of the variable part of
  the ICMP header, not the total size of the ICMP header itself.

* Introduce totlen, this is the total size of the ICMP header (icmp_ip
  included).

No real functional change.

(maxv)

sun4v doesn't have %ver register.

(nakayama)

Fix inverted logic.

(nakayama)

Move the ICMP Extension structures from mpls_ttl.c to ip_icmp.h; that's
part of the ICMP protocol (per RFC4884), and not specific to MPLS. Also
add ih_exthdr in struct icmp, the 'length' field appeared.

While here, style in MPLS.

(maxv)

Style, explain a bit, and fix icmp_radv, it should be icmp_dun.id_radv.

(maxv)

Fix some ARM kernel builds.  ARM abuses compat32 for ABI compatibility and
this means some ioctls overlap.

(skrll)

Suppress noisy debugging outputs

Even if DEBUG they are too noisy under load.

(ozaki-r)

Style, no functional change.

(maxv)

Style, and check the return value of m_append.

(maxv)

Style, no functional change.

(maxv)

Style, no functional change.

(maxv)

Run tcp_slowtimo in workqueue if NET_MPSAFE

If NET_MPSAFE is enabled, we have to avoid taking softnet_lock in softint as
much as possible to prevent any softint handlers including callout handlers
such as tcp_slowtimo from sticking on softnet_lock because it results in
undesired delays of executing subsequent softint handlers.

NFCI for !NET_MPSAFE

(ozaki-r)

Style, and make sure that there is a header+trailer included in the
packet. The crypto functions can touch the trailer, but they don't check
whether it's there in the first place.

(maxv)

Release rt_so_mtx on updating a rtentry to avoid a deadlock with route_intr

The deadlock happened only if NET_MPSAFE on.

(ozaki-r)

Use usbdevices. add midi needed to build umidi.
Remove now redundant USB entries.

So we will stop missing random USB drivers.

(maya)

Style, no functional change.

(maxv)

Style, and zero out 'ns' entirely, otherwise some bytes get leaked to
userland (eg ns_rsvd0).

(maxv)

Several changes:

* Make the code more readable.

* Add a panic in ieee80211_compute_duration(). I'm not sure there's
  a bug here - I don't have the hardware -, but looking at the code, it
  may be possible for 'paylen' to go negative. Obviously that's not the
  correct way to fix it, but at least we'll see if it happens.

(maxv)

Fix test for SMI_IMB{,range} that was broken in 1.46

From Ramakrishna Rao Desetti

(skrll)

Don't return the address of the kernel modules if the user is not
privileged. Discussed on tech-kern@.

(maxv)

Several changes:

* Make the code more readable. In particular, declare variables as const
  along the way.

* Explain what we're doing in ieee80211_send_mgmt(). The
  IEEE80211_FC0_SUBTYPE_PROBE_RESP case has some inconsistencies, but
  they are not inherently wrong so I'm not changing that.

* When sending IEEE80211_FC0_SUBTYPE_REASSOC_RESP frames, make sure to
  zero out the 'association ID', otherwise two bytes are leaked.

* Fix a possible memory leak in ieee80211_send_probereq().

(maxv)

Use r0 in the tlbiall instruction to avoid confusion.

From Ramakrishna Rao Desetti

(skrll)

More markup.

(wiz)

add WM_TX_PROCESS_LIMIT_DEFAULT and WM_TX_INTR_PROCESS_LIMIT_DEFAULT man.

(knakahara)

wm_txeof() can limit the loop count the same as wm_rxeof() now.

(knakahara)

typo in comment

(skrll)

Unmap the kernel heap from the user page tables (SVS).

This implementation is optimized and organized in such a way that we
don't need to copy the kernel stack to a safe place during user<->kernel
transitions. We create two VAs that point to the same physical page; one
will be mapped in userland and is offset in order to contain only the
trapframe, the other is mapped in the kernel and maps the entire stack.

Sent on tech-kern@ a week ago.

(maxv)

note netbsd32 support for raid(4).

(mrg)

implement 32-bit compat support for raidframe.

convert several raidframe ioctls to be bitsize idempotent so that
they work the same in 32 and 64 bit worlds, allowing netbsd32 to
configure and query raid properly.  remove useless 'row' in a few
places.  add COMPAT_80 and put the old ioctls there.

raidframeio.h:
  RAIDFRAME_TEST_ACC
  - remove, unused
  RAIDFRAME_GET_COMPONENT_LABEL
  - convert to label not pointer to label
  RAIDFRAME_CHECK_RECON_STATUS_EXT
  RAIDFRAME_CHECK_PARITYREWRITE_STATUS_EXT
  RAIDFRAME_CHECK_COPYBACK_STATUS_EXT
  - convert to progress info not pointer to info
  RAIDFRAME_GET_INFO
  - version entirely.
raidframevar.h:
  - rf_recon_req{} has row, flags and raidPtr removed (they're
    not a useful part of this interface.)
  - RF_Config_s{} and RF_DeviceConfig_s{} have numRow/rows removed.
  - RF_RaidDisk_s{} is re-ordered slightly to fix alignment
    padding - the actual data was already OK.
  - InstallSpareTable() loses row argument

rf_compat32.c has code for RF_Config_s{} in 32 bit mode, used
by RAIDFRAME_CONFIGURE and RAIDFRAME_GET_INFO32.

rf_compat80.c has code for rf_recon_req{}, RF_RaidDisk_s{} and
RF_DeviceConfig_s{} to handle RAIDFRAME_FAIL_DISK,
RAIDFRAME_GET_COMPONENT_LABEL, RAIDFRAME_CHECK_RECON_STATUS_EXT,
RAIDFRAME_CHECK_PARITYREWRITE_STATUS_EXT,
RAIDFRAME_CHECK_COPYBACK_STATUS_EXT, RAIDFRAME_GET_INFO.

move several of the per-ioctl code blocks into separate functions.

add rf_recon_req_internal{} to replace old usage of global
rf_recon_req{} that had unused void * in the structure, ruining
it's 32/64 bit ABI.

add missing case for RAIDFRAME_GET_INFO50.

adjust raid tests to use the new .conf format, and add a case to
test the old method as well.

raidctl:
deal with lack of 'row' members in a couple of places.
fail request no longer takes row.
handle "START array" sections with just "numCol numSpare", ie
no "numRow" specified.  for now, generate old-style configuration
but update raidctl.8 to specify the new style (keeping reference
to the old style.)

note that: RF_ComponentLabel_s::{row,num_rows} and
RF_SingleComponent_s::row are obsolete but not removed yet.

(mrg)

merge the ebus@pci attach output to be like most PCI drivers:

ebus0 at pci1 dev 1 function 0
ebus0: Sun Microsystems PCIO Ebus2, revision 0x01

becomes:

ebus0 at pci1 dev 1 function 0: Sun Microsystems PCIO Ebus2, revision 0x01

(mrg)

G/C ARM32_NEW_VM_LAYOUT

(skrll)

Style, and fix two pretty bad mistakes in the crypto functions:

* They call M_PREPEND, but don't pass the updated pointer back to the
  caller.

* They use memmove on the mbuf data, but they don't ensure that the
  area they touch is contiguous.

This fix is not complete, ieee80211_crypto_encap too needs to pass back
the updated pointer. This will be done in another commit.

(maxv)

Track 3rd party firmware in our tree.

(mlelstv)

Several changes:

* Style in several places, to make the code more readable or easier to
  understand.

* Instead of checking m->m_pkthdr.len, check m->m_len. m_pkthdr.len is
  the total size of the packet, not the size of the current mbuf (which
  may be smaller).

* Add a missing length check when handling QoS frames.

* Cast the lengths passed in IEEE80211_VERIFY_LENGTH to size_t.

* Remove the length check on scan.sp_xrates, that I added yesterday.
  xrates gets silently truncated in ieee80211_setup_rates().

* Fix several buffer overflows in the parsers of the MANAGEMENT frames.

(maxv)

Note veriexec being enabed by default on macppc port

(sevan)

Note fileassoc(9) support is a dependency Veriexec.

Resolves PR kern/52909

(sevan)

Enable veriexec(4) support by default on the macppc port and update the manual to mention it.

Closes PR port-powerpc/52908

(sevan)

Use more semantic markup.

(uwe)

fix previous:
CPUFLG_SUN4CACHEBUG was renamed to CACHE_TRAPPAGEBUG and moved
into the cacheinfo.  don't revive the no longer used value, but
fix the non-sun4m code to match cpu.c 1.251.

(mrg)

make lint compile again.

(christos)

Avoid applying .Dv to the left parenthesis, since it is not applied to
the closing right parenthesis.

(pgoyette)

PR/47615: Dr. W. Stukenbrock: Always zero out the result structs in the
svc procs to avoid returning stale request data to the client.
XXX: pullup-6,7,8

(christos)

Don't skip the superfluous "is" argument if it's not superfluous!

This allows us to use `wtf is` and get information for the acronym "is"
and produces the same output as `wtf is is` withough requiring the extra
typing by the user.

(pgoyette)

document pfil_{add,remove}_ihook, pfil_run_{addr,if}hooks

from ryo shimizu in PR bin/51941

changes by myself: fix mdoc -Tlint, add missing set lists, adjust date
to today, reword history addition.

(maya)

fix compilation

(christos)

rename DCACHE_SIZE to XSCALE_DCACHE_SIZE to avoid conflict with genassym
field offset with the same name.

(christos)

Fix duplicated "rxintr" evcnt counting. Pointed out by ozaki-r@n.o, thanks.

(knakahara)

Bring eawk up to date, it's had no real love since 2013.

1. Don't use bison, just use standard yacc to build
2. Use pure-parser directive, rather than %pure_parser
3. libarchive has changed its API since 2013 - adapt
4. netpgpverify has also changed its API since 2013 - adapt also
5. There is no sqlite3 sub-directory, so don't descend into it

With thanks to yukonbob for the nudge.

(agc)

Revert to commits as they broke few rump tests

"Include namespace.h in a few of libc source files
[...]
    This change finishes elimination of usage of the global name of the
    following symbols:
      - close -> _close
      - execve -> _execve
      - fcntl -> _fcntl
      - setcontext -> _setcontext
      - wait6 -> _wait6
      - write -> _write
      - writev -> _writev"

"Register more syscalls in namespace.h (of libc)

    Add weak symbols for:
    - fcntl
    - close
    - execve
    - setcontext
    - wait6
    - write
    - writev"

These changes broke:

    fs/nfs/t_rquotad:get_nfs_be_1_both
    fs/nfs/t_rquotad:get_nfs_be_1_group
    fs/nfs/t_rquotad:get_nfs_be_1_user
    fs/nfs/t_rquotad:get_nfs_le_1_both
    fs/nfs/t_rquotad:get_nfs_le_1_group
    fs/nfs/t_rquotad:get_nfs_le_1_user
    lib/librumphijack/t_config:fdoff
    lib/librumphijack/t_tcpip:http
    lib/librumphijack/t_tcpip:nfs
    lib/librumphijack/t_vfs:cpcopy
    lib/librumphijack/t_vfs:mv_x
    lib/librumphijack/t_vfs:paxcopy
    net/net/t_forwarding:ipforwarding_fastforward_v4
    net/net/t_forwarding:ipforwarding_fastforward_v6
    net/net/t_forwarding:ipforwarding_fragment_v4
    net/net/t_forwarding:ipforwarding_misc
    net/net/t_mtudisc6:mtudisc6_basic

This revert fixes the failures, except lib/librumphijack/t_vfs.

The original changes were added in order to facilitate a usage of
sanitizers against programs linked with NetBSD's libc. It is no longer
needed, so abandon these changes.

Sponsored by <The NetBSD Foundation>

(kamil)

Use 0600 as the mode for histfile here too.

pointed out by John D. Baker in PR bin/52480

(maya)