1115

(snj)

Pull up following revision(s) (requested by riastradh in ticket #1115):
sys/arch/x86/x86/pmap.c: patch
Use IPL_NONE for pserialized lock.  Assert sleepable.  (OOPS.)

(snj)

Fix the bitmask of MVXPE_PMACC0_FRAMESIZELIMIT. It did no harm.

(hikaru)

regen.

(mrg)

add several NVIDIA geforce and NVS cards.

(mrg)

Fix typo in comment.

(msaitoh)

tickets 1080-1088, 1090-1092, 1110-1113, 1116

(snj)

Pull up following revision(s) (requested by riastradh in ticket #1116):
sys/external/bsd/drm2/dist/drm/nouveau/core/engine/fifo/nouveau_engine_fifo_base.c: revision 1.4
sys/external/bsd/drm2/dist/drm/nouveau/core/include/engine/fifo.h: revision 1.4
dtor gets called if ctor fails.  Make it work in that case.

(snj)

Pull up following revision(s) (requested by riastradh in ticket #1113):
sys/external/bsd/drm2/include/asm/unaligned.h: revision 1.4
Don't return void.
Caught by pcc.  Thanks, pcc!

(snj)

Pull up following revision(s) (requested by riastradh in ticket #1112):
sys/external/bsd/drm2/dist/drm/nouveau/core/subdev/therm/nouveau_subdev_therm_temp.c: revision 1.2
More useful array bounds check for enum values.

(snj)

Pull up following revision(s) (requested by riastradh in ticket #1111):
sys/external/bsd/drm2/dist/drm/nouveau/nouveau_nv50_display.c: revision 1.5
Omit unused definition of readw.

(snj)

Pull up following revision(s) (requested by riastradh in ticket #1110):
sys/external/bsd/drm2/dist/drm/nouveau/core/engine/perfmon/nouveau_engine_perfmon_base.c: revision 1.3
Don't ask `size < 0' when size is unsigned.

(snj)

Pull up following revision(s) (requested by riastradh in ticket #1091):
sys/external/bsd/drm2/dist/drm/i915/i915_dma.c: revisions 1.17, 1.18
Zero out the guard for bus_space_unmap before calling i915_dma_cleanup() which
calls i915_free_hws(), which then tries to unmap. Perhaps this fixes PR/50060.
--
fix the same bug on the linux side, print the error, and return the -tive
error to mimick linux.

(snj)

Pull up following revision(s) (requested by riastradh in ticket #1090):
sys/external/bsd/drm2/dist/drm/nouveau/nouveau_nv50_display.c: revision 1.4
sys/external/bsd/drm2/dist/drm/nouveau/core/subdev/fb/nouveau_subdev_fb_nv50.c: revision 1.3
sys/external/bsd/drm2/dist/drm/nouveau/core/subdev/fb/nouveau_subdev_fb_nvc0.c: revision 1.3
sys/external/bsd/drm2/dist/drm/nouveau/core/subdev/vm/nouveau_subdev_vm_nv04.c: revision 1.3
sys/external/bsd/drm2/dist/drm/nouveau/core/subdev/vm/nouveau_subdev_vm_nv44.c: revision 1.3
use pci_dma64_available() to choose a valid DMA tag.

(snj)

Pull up following revision(s) (requested by riastradh in ticket #1088):
sys/external/bsd/drm2/drm/drmfb.c: revision 1.2
If the "is_console" property is already specified, do not overwrite it.

(snj)

Pull up following revision(s) (requested by riastradh in ticket #1087):
sys/external/bsd/drm2/drm/drm_drv.c: revision 1.17
only clear agp hook on last close when agp is in use, from riastradh@

(snj)

1086

(snj)

Pull up following revision(s) (requested by riastradh in ticket #1086):
sys/external/bsd/drm2/linux/linux_work.c: revision 1.12
Avoid NULL dereference in linux_worker_intr() in WORK_DELAYED_CANCELLED case.
Potentially fixes PR#49560.

(snj)

Pull up following revision(s) (requested by riastradh in ticket #1086):
sys/external/bsd/drm2/linux/linux_work.c: revision 1.12
Avoid NULL dereference in linux_worker_intr() in WORK_DELAYED_CANCELLED case.
Potentially fixes PR#49560.

(snj)

Pull up following revision(s) (requested by riastradh in ticket #1085):
sys/external/bsd/drm2/dist/drm/nouveau/nouveau_bo.c: revision 1.6
sys/external/bsd/drm2/dist/drm/nouveau/nouveau_drm.h: revision 1.3
provide a ttm-backed set of nouveau_uvm_ops{} in the ttm_bo_driver{}.
fixes NULL pointer derefs inside UVM due to lack of pager ops.
copy the DRM_FILE_PAGE_OFFSET vs !_LP64 workaround.
with these, X starts up on my GTX 440.

(snj)

Pull up following revision(s) (requested by riastradh in ticket #1084):
sys/external/bsd/drm2/dist/drm/nouveau/core/subdev/therm/nouveau_subdev_therm_base.c: revision 1.2
sys/external/bsd/drm2/dist/drm/nouveau/core/subdev/therm/nouveau_subdev_therm_fan.c: revision 1.2
disable the once-a-second and many-times-a-minute logs about thermals.
these trigger on multiple cards for me and totally spam my logs in the
normal case.  one card switches from 41% to 42% and back again quite
regularly.

(snj)

Pull up following revision(s) (requested by riastradh in ticket #1083):
sys/external/bsd/drm2/radeon/radeondrmkmsfb.c: revision 1.6
set the "width" and "height" parameters from the surface_width and
surface_height members.  idea from the intel code and jmcneill@.
this makes the fb on my RV370 use the full display size, not the
(not actually connected) 800x600 the svideo output supports.
X still starts up at 800x600 mode, but xrandr can fix that.

(snj)

Pull up following revision(s) (requested by riastradh in ticket #1082):
sys/external/bsd/drm2/dist/drm/nouveau/core/engine/device/nouveau_engine_device_base.c: revisions 1.6-1.8
Fix bogus output parameter for bus_space_map.
We stored the bus space handle in nv_subdev(device)->mmioh, which we
then overwrote with whatever was in the mmioh local variable -- maybe
garbage, maybe an old mapping, almost certainly wrong either way.
Instead store it in mmioh so that the subsequent assignment
nv_subdev(device)->mmioh = mmioh actually works as intended.
--
Return error from bus_space_map.
--
Paranoia about mmio register mapping size.

(snj)

Pull up following revision(s) (requested by riastradh in ticket #1081):
sys/external/bsd/drm2/nouveau/nouveau_pci.c: revision 1.6
push the real attach down via config_mountroot, like radeon, as we
need firmware files.. which are currently missing:
drm kern error: nouveau E[  PGRAPH][nouveau0] failed to load fuc409c

(snj)

Pull up following revision(s) (requested by riastradh in ticket #1080):
sys/external/bsd/drm2/dist/drm/i915/intel_i2c.c: revision 1.14
sys/external/bsd/drm2/dist/drm/i915/intel_pm.c: revision 1.8
sys/external/bsd/drm2/dist/drm/i915/intel_sdvo.c: revision 1.7
sys/external/bsd/drm2/dist/drm/nouveau/nouveau_drm.c: revision 1.7
sys/external/bsd/drm2/dist/drm/nouveau/core/include/core/device.h: revision 1.5
sys/external/bsd/drm2/dist/drm/nouveau/core/subdev/bios/nouveau_subdev_bios_base.c: revision 1.3
sys/external/bsd/drm2/dist/drm/radeon/radeon_display.c: revision 1.5
sys/external/bsd/drm2/dist/drm/radeon/radeon_gem.c: revision 1.3
sys/external/bsd/drm2/dist/drm/radeon/radeon_i2c.c: revision 1.3
sys/external/bsd/drm2/dist/drm/radeon/radeon_kms.c: revision 1.3
sys/external/bsd/drm2/include/linux/pci.h: revision 1.21
sys/external/bsd/drm2/include/linux/platform_device.h: revision 1.7
sys/external/bsd/drm2/nouveau/nouveau_pci.c: revision 1.5
Use nv_device_base(dev) like other callers, not dev->pdev->dev.
First step in fixing the struct device nonsense.  This won't actually
fix anything yet until we also fix nv_device_base.
--
dev->dev, not &dev->pdev->dev.
Latter is a charlatan, soon to meet judgment for its sins.
--
dev->dev, not &dev->pdev->dev.
Latter is a mistake that is all my fault.  Oops.
--
Eliminate Linux struct pci_dev::dev, struct platform_device::dev.

(snj)

Pull up following revision(s) (requested by riastradh in ticket #1092):
sys/external/bsd/drm2/nouveau/files.nouveau: revision 1.9
sys/external/bsd/drm2/nouveau/nouveau_module.c: revision 1.4
sys/external/bsd/drm2/nouveau/nouveau_pci.c: revision 1.4
sys/external/bsd/drm2/nouveau/nouveau_pci.h: revision 1.2
sys/external/bsd/drm2/nouveau/nouveaufb.c: revision 1.3
Split nouveau@pci from nouveau so we can attach it to other busses.
nouveaufb still depends on nouveau_pci, but it is not applicable to
non-PCI devices anyway.

(snj)

support screen blanking and hw cursor

(macallan)

- Add some more Char casts
- reduce ifdefs by providing empty defs for nls functions (Ingo Schwarze)

(christos)

remove unused wrapper (Ingo Schwarze)

(christos)

Remove utf8_islead() mbrtowc() handles this just fine (Ingo Schwarze)

(christos)

texinfo-6.1 and grep-2.23 out.

(wiz)

Pass a sensible device state pointer to nouveau suspend/resume ops.

Gives nouveau half a chance of suspending and resuming -- not that it
works on my test laptop yet, but it's a start.

(riastradh)

Set nouveau to a more reasonable debug level.

Match the compile-time and default run-time settings in Linux.

(riastradh)

Try xf86-video-nouveau for nvidia devices by default.

Fall back to nv (and then vesa and then wsfb) if it doesn't work.

(riastradh)

switch to 32bit colour in WSDISPLAYIO_MODE_MAPPED, now X with wsfb will work
Not optimal though - for some reason the framebuffer's endianness in 32bit
colour is wrong and I have no idea (yet) how to change that, so many apps
using xrender will crash.

(macallan)

dtor gets called if ctor fails.  Make it work in that case.

(riastradh)

Don't share struct work, instead have one per softc

Pointed out by riastradh@

(ozaki-r)

use pcpp front end rather than libexec/cpp directly, since commandline
options may not be handled by the latter

(plunky)

add pcpp binary, now in pcc-20160208. Also, p++ in debug set

(plunky)

update build machinery for pcc-20160208

(plunky)

fix merge conflicts from import of pcc-20160208

(plunky)

Fix build

(ozaki-r)

Document file format better. From Travis Paul and Matthew Bauer.

Bump date.
Addresses PR 50787.

(wiz)

Add more #warning message 'Please use mvxpe instead of this'.

(kiyohara)

Clear ATAC_CAP_DATA32, if dev is atabus@wdc@pcmcia@aupcmcia.  It is 32-bit transfers are not supported.

(kiyohara)

Actually, descsz should not contain the padding. The note still needs to
be padded out.

(skrll)

Introduce softint-based if_input

This change intends to run the whole network stack in softint context
(or normal LWP), not hardware interrupt context. Note that the work is
still incomplete by this change; to that end, we also have to softint-ify
if_link_state_change (and bpf) which can still run in hardware interrupt.

This change softint-ifies at ifp->if_input that is called from
each device driver (and ieee80211_input) to ensure Layer 2 runs
in softint (e.g., ether_input and bridge_input). To this end,
we provide a framework (called percpuq) that utlizes softint(9)
and percpu ifqueues. With this patch, rxintr of most drivers just
queues received packets and schedules a softint, and the softint
dequeues packets and does rest packet processing.

To minimize changes to each driver, percpuq is allocated in struct
ifnet for now and that is initialized by default (in if_attach).
We probably have to move percpuq to softc of each driver, but it's
future work. At this point, only wm(4) has percpuq in its softc
as a reference implementation.

Additional information including performance numbers can be found
in the thread at tech-kern@ and tech-net@:
http://mail-index.netbsd.org/tech-kern/2016/01/14/msg019997.html

Acknowledgment: riastradh@ greatly helped this work.
Thank you very much!

(ozaki-r)

Preserve the .text section alignment

(skrll)

Allocate struct cpu_info dynamically.  Saves quite a lot of BSS in the
common case and reduces rump kernel memory requirements by 10% or more
in really tiny deployments.

(pooka)

UTF-8 fixes from Ingo Schwarze:

1. Assume that errno is non-zero when entering read_char()
    and that read(2) returns 0 (indicating end of file).
    Then, the code will clear errno before returning.
    (Obviously, the statement "errno = 0" is almost always
    a bug unless there is save_errno = errno right before it
    and the previous value is properly restored later,
    in all reachable code paths.)

2. When encountering an invalid byte sequence, the code discards
    all following bytes until MB_LEN_MAX overflows; consider, for
    example, 0xc2 immediately followed by a few valid ASCII bytes.
    Three of those ASCII bytes will be discarded.

3. On a POSIX system, EILSEQ will always be set after reading a
    valid (yes, valid, not invalid!) UTF-8 character.  The reason
    is that mbtowc(3) will first be called with a length limit
    (third argument) of 1, which will fail, return -1, and - on
    a POSIX system - set errno to EILSEQ.
    This third bug is mitigated a bit because i couldn't find any
    system that actually conforms to POSIX in this respect:  None
    of OpenBSD, NetBSD, FreeBSD, Solaris 11, and glibc set errno
    when an incomplete character is passed to mbtowc(3), even though
    that is required by POSIX.
    Anyway, that mbtowc(3) bug will be fixed at least in OpenBSD
    after release unlock, so it would be good to fix this bug in
    libedit before fixing the bug in mbtowc(3).

How can these three bugs be fixed?

1. As far as i understand it, the intention of the bogus errno = 0
    is to undo the effects of failing system calls in el_wset(),
    sig_set(), and read__fixio() if the subsequent read(2) indicates
    end of file.  So, restoring errno has to be moved right after
    read__fixio().  Of course, neither 0 nor e is the right value
    to restore: 0 is wrong if errno happened to be set on entry, e
    would be wrong because if one read(2) fails but a second attempt
    succeeds after read__fixio(), errno should not be touched.  So,
    the errno to be restored in this case has to be saved before
    calling read(2) for the first time.

2. Solving the second issue requires distinguishing invalid and
    incomplete characters, but that is impossible with the function
    mbtowc(3) because it returns -1 in both cases and sets errno
    to EILSEQ in both cases (once properly implemented).

    It is vital that each input character is processed right away.
    It is not acceptable to wait for the next input character before
    processing the previous one because this is an interactive
    library, not a batch system.  Consequently, the only situation
    where it is acceptable to wait for the next byte without first
    processing the previous one(s) is when the previous one(s) form
    an incomplete sequence that can be continued to form a valid
    character.

    Consequently, short of reimplementing a full UTF-8 state machine
    by hand, the only correct way forward is to use mbrtowc(3).
    Even then, care is needed to always have the state object
    properly initialized before using it, and to not discard a valid
    ASCII or UTF-8 lead byte if it happens to follow an invalid
    sequence.

3. Fortunately, solution 2. also solves issue 3. as a side effect,
    by no longer using mbtowc(3) in the first place.

(christos)

PR/50783: David Binderman: Indent switch properly, add missing break.

(christos)

PR/50784: David Binderman: Add missing braces, this is not python.

(christos)

PR/50785: David Binderman: Remove dead code.

(christos)

Fix .note.netbsd.march by ensuring correct padding

(skrll)

Remove the fstd argument from uhci_reset_std_chain as it is always
ux_stds[0]

(skrll)

Remove the .SUNW_ctf sections when converting form ELF -> a.out by
updating OBJCOPY_ELF2AOUT_FLAGS

(skrll)

whitespace

(dholland)

don't re-define _KERNTYPES

(christos)

Use IPL_NONE for pserialized lock.  Assert sleepable.  (OOPS.)

(riastradh)

Simplify usbd_do_request/usbd_setup_default_xfer.

G/C usbd_do_request_flags_pipe as it's not required

(skrll)

G/C usbd_do_request_async declaration - it doesn't exist.

(skrll)

fix pasto

(christos)

New sentence, new line.

(wiz)

Unwrap line

(skrll)

Add the copyright notice which was missing from my initial version.

(jandberg)

Fix signing of in-memory data with SSH keys

(agc)

In module_do_load(), consolidate checking for a pre-existing module,
and return a single error value EEXIST.  When making a recursive
call (to load required modules), treat a pre-existing module as
success.

Without this change, when a module was loaded by specific request
(as opposed to being loaded as a requirement of some other module),
we would always load the module from the file-system, and then
after making various sanity/compatability checks we would destroy
the new copy if there was a pre-existing copy.

Fixes PR kern/40764

XXX Note that if the module exists, we bypass all of the various
XXX "compatability" checks, including whether or not the existing
XXX module is of any particular class!  (In the previous code, we
XXX checked to see if the newly-loaded copy had the correct class,
XXX but not the pre-existing copy, which could have been loaded
XXX from a different path/filename.)

(pgoyette)

109[3489]

(snj)

109[3479]

(snj)

use sizeof() and array notation.

(christos)

Pull up following revision(s) (requested by christos in ticket #1099):
external/gpl2/xcvs/dist/src/import.c: revision 1.4
Remove trailing whitespace from imported messages when adding new files.
This is done in RCS_checkin() but we have a "fast path" for new files here
that does not do it. Hooray for specialized code.

(snj)

Pull up following revision(s) (requested by christos in ticket #1099):
external/gpl2/xcvs/dist/src/import.c: revision 1.4
Remove trailing whitespace from imported messages when adding now files.
This is done in RCS_checkin() but we have a "fast path" for new files here
that does not do it. Hooray for specialized code.

(snj)

Pull up following revision(s) (requested by bouyer in ticket #1098):
lib/libquota/quota_nfs.c: revision 1.5
Some NFS servers return RPC_PROGNOTREGISTERED instead of RPC_PROGVERSMISMATCH
when they don't support EXT_RQUOTAVERS but support RQUOTAVERS, so
retry with RQUOTAVERS for both RPC_PROGNOTREGISTERED and RPC_PROGVERSMISMATCH
return code.
Fix issue against a NetApp server reported by 6bone on current-users

(snj)

Pull up following revision(s) (requested by bouyer in ticket #1098):
lib/libquota/quota_nfs.c: revision 1.5
Some NFS servers return RPC_PROGNOTREGISTERED instead of RPC_PROGVERSMISMATCH
when they don't support EXT_RQUOTAVERS but support RQUOTAVERS, so
retry with RQUOTAVERS for both RPC_PROGNOTREGISTERED and RPC_PROGVERSMISMATCH
return code.
Fix issue against a NetApp server reported by 6bone on current-users

(snj)

Pull up following revision(s) (requested by skrll in ticket #1097):
sys/dev/usb/usb.c: revision 1.161
sys/dev/usb/usb_subr.c: revisions 1.207, 1.208
sys/dev/usb/usbdivar.h: revision 1.111
sys/dev/usb/xhci.c: revision 1.33
Get the iManufacturer, iProduct, and iSerialNumber strings before probing
for drivers and cache them for later use.  This reduces bus transactions
and fixes attachment for at least two of my umass(4)s.
--
Need sys/kmem.h

(snj)

Pull up following revision(s) (requested by hannken in ticket #1094):
sys/nfs/nfs_vnops.c: revision 1.309
Return an error if NFSPROC_LOOKUP returns the file handle of the current
directory.  Treating it as DOT lookup would put garbage into the name
cache and could panic on future lookups.
Seen with ZFS file system exported from OmniOS, an OpenSolaris derivative.
Fixes PR kern/50664 "cd .." over NFS/ZFS can panic kernel

(snj)

Pull up following revision(s) (requested by hannken in ticket #1094):
sys/nfs/nfs_vnops.c: revision 1.309
Return an error if NFSPROC_LOOKUP returns the file handle of the current
directory.  Treating it as DOT lookup would put garbage into the name
cache and could panic on future lookups.
Seen with ZFS file system exported from OmniOS, an OpenSolaris derivative.
Fixes PR kern/50664 "cd .." over NFS/ZFS can panic kernel

(snj)

Pull up following revision(s) (requested by tsutsui in ticket #1093):
sys/arch/hp300/hp300/intr.c: revision 1.42
Fix fatal "intr_dispatch: stray level 6 interrupt" panic by the GENERIC
kernel on machines without arcofi(4) audio (i.e. all hp300 except HP425e).
I failed to pull all necessary changes for arcofi(4) from OpenBSD
and I should have tested 7.0_RC kernels on my HP382 too :-(
Should be pulled up to netbsd-7 and netbsd-7-0.

(snj)

Pull up following revision(s) (requested by tsutsui in ticket #1093):
sys/arch/hp300/hp300/intr.c: revision 1.42
Fix fatal "intr_dispatch: stray level 6 interrupt" panic by the GENERIC
kernel on machines without arcofi(4) audio (i.e. all hp300 except HP425e).
I failed to pull all necessary changes for arcofi(4) from OpenBSD
and I should have tested 7.0_RC kernels on my HP382 too :-(
Should be pulled up to netbsd-7 and netbsd-7-0.

(snj)

Pull up following revision(s) (requested by hannken in ticket #1073):
sys/arch/x86/x86/errata.c: revision 1.23
sys/arch/x86/include/cpufunc.h: revision 1.19
Adapt prototypes and usage of rdmsr_locked() and wrmsr_locked() to
their implementation.  Both functions don't take the passcode as
argument.
As wrmsr_locked() no longer writes the passcode to the msr the
erratum 721 on my Opteron 2356 really gets patched and cc1 no longer
crashes with SIGSEGV.

(snj)

No <immintrin.h> for pcc even on x86.

(riastradh)

METEORGSIGNAL and METEORSFMT overlap with dtrace, prefer the latter.

(joerg)

Fix some awkward language in manpage and code comments.

(jandberg)

Add $NetBSD$ tag.

(riastradh)

Avoid shadowing global.

(riastradh)

Override default LIB_PATH for i386 emulation. Unlike older binutils, we
now have to check the emulation and not that the target as the change
otherwise leaks into the native 64bit linker scripts as well.

(joerg)

We don't want to target-specific search directories on NetBSD in the
linker scripts, they just add noise.

(joerg)

Avoid empty loop.

(joerg)

SDT is not a disease. Use const consistently.

(joerg)

Back out unintended changes committed with last.

(riz)

Don't test dtrace-related variables until after they've been set.

(riz)

Don't do _BSD_SIZE_T_ dance here.  Use <stddef.h> in userland.

This file is used in tools, where _BSD_SIZE_T_ is unreliable.  In the
kernel, <sys/types.h> will define size_t anyway.  Outside the kernel,
<stddef.h> is the standard way to get size_t.

(riastradh)

Toolify.

(joerg)

KNF comments with whitespace

(skrll)

Split case folding table into separate source file and add full
copyright and permission notice from http://www.unicode.org/copyright.html

(mlelstv)

fix usage message

(mlelstv)

Missed hunk from previous

(skrll)

Make interrupt handling MP safe by not dropping the bus lock while
traversing the active transfer list.  Instead move the complete transfers
from the active list to a complete list and do callbacks on this complete
list.

usbd_transfer_complete can safely drop the bus lock while traversing this
private list.

(skrll)

Improve DEBUG

(skrll)

Whitespace

(skrll)

Remove unintended commit

(skrll)

#include "opt_*" corrections

(skrll)

Remove uhci_device_request declaration as it doesn't exist anymore

(skrll)

Use if (expr) { ... } instead of a if (expr) goto label; ... ; label:

(skrll)

Fix the fix... look a the correct list of TDs when aborting a transfer

(skrll)

More debug

(skrll)

More debug

(skrll)

No need to use mutex_obj_alloc here

(skrll)

Pull across fix from HEAD

Use the port status straight after the reset and not after the

usbd_clear_port_feature(dev, port, UHF_C_PORT_RESET)

The Gadget USB HUB in the IBM x3250 sborrill sent me returns the wrong
status and this would appear to leave the attached cdce(4) in the default
state.  The next device probed fails as a result.

(skrll)

Avoid shadowing global `min'.

(riastradh)

Convert bool to target type before shift.

(riastradh)

Avoid shadowing global `min'.

(riastradh)

Missed a spot.

(riastradh)

Avoid shadowing global `max' from libkern.

(riastradh)

Don't return void.

Caught by pcc.  Thanks, pcc!

(riastradh)

More useful array bounds check for enum values.

(riastradh)

Omit unused definition of readw.

(riastradh)

Don't ask `size < 0' when size is unsigned.

(riastradh)

Need nbtool_config.h for __dead in tools.

(riastradh)

Need nbtools_config.h for __unused.

Fixes

.../src/tools/ctfmerge/../../external/cddl/osnet/dist/tools/ctf/common/list.c:73:47: error: expected ';', ',' or ')' before '__unused'
list_defcmp(void *d1, void *d2, void *private __unused)
                                              ^~~~~~~~

on non-NetBSD systems, where <assert.h>, <stdlib.h>, <stdio.h>, and
<sys/types.h> don't define __unused.

(riastradh)

World+dog want to treat rdl_path as char, so make it so.

(roy)

Use the port status straight after the reset and not after the

usbd_clear_port_feature(dev, port, UHF_C_PORT_RESET)

The Gadget USB HUB in the IBM x3250 sborrill sent me returns the wrong
status and this would appear to leave the attached cdce(4) in the default
state.  The next device probed fails as a result.

(skrll)

Fix is defined in wpa_common.h which the driver already pulls in.

(roy)

s/max/ralmax/. No functional change.

(msaitoh)

Only build tests if MKATF != no

(roy)

Fix the actual issue rather than silencing the warning.

Thanks to gson@

(roy)

PR/50744: NONAKA Kimihiro: Protect more stuff with _KERNEL && _KMEMUSER to
make uvm_extern.h compile standalone again for net-snmp.

(christos)

Initialize the token match pointer.

(nonaka)

File watt in the right group, and accept W as well, as that's quite
standard.

(dholland)

Improve previous slightly; while we accept e.g. "meter2" to mean "meter^2",
don't allow "meter-2" to be interpreted as "meter^-2" as that's very
confusing.

(dholland)

Improve the parser. Now we understand negative exponents; fixes PR 50768.

Also handle negative numbers better in general (don't randomly drop
the sign in a number of cases) and don't choke on exponents > 9.

This commit alters the meaning of a few previously valid but marginal
inputs (e.g. "3 foot-5 pound" is now treated as "3*-5 foot-pound"
rather than "3*5 foot-pound"; if you want the latter insert another
space) but corrects obviously wrong handling of many more.

(dholland)

fix: "vmstat -C" CpuLayer showed only the last cpu values.

(knakahara)

Add -Wno-format-nonliteral

(roy)

dtrace needs Wno-format-security for clang it seems.

(roy)

Add some __printflike and __dead love. Init some structs fully.

(roy)

Fix build.

(roy)

Mark more functions as __printflike and __dead.

(roy)

Mark more functions as __dead.

(roy)

Disable -Wformat-nonliteral here to build.

(roy)

Disable -Wformat-nonliteral here to build.

(roy)

More printflike and dead fixes.

(roy)

Fix printf formats and casts.

(roy)

Correct a prior patch by moving __printflike marking from the .c files
to the .h ones.

(roy)

Mark __printflike and __dead.

(roy)

Mark printflike functions as __printflike.

(roy)

- Avoid uninitialized variable usage in do_nologin() when HAVE_LOGIN_CAP
  isn't defined (which doesn't apply to NetBSD but...) and a root
  login is being evaluated.
- From upstream.

(seanb)

Declare in6_tmpaddrtimer_ch in in6_var.h.

Do not declare extern variables in .c files!

(riastradh)

Regen the rest.

(christos)

regen some more.

(christos)

sun4v: extend dmmu_set_secondary_context() to handle both sun4u and sun4v. ok martin@

(palle)

define _KERNTYPES

(christos)

define _KMEMUSER earlier for mips

(christos)

Adapt getmajor and getminor signatures for NetBSD.

(riz)

switch i386 to gdb-7.10.1

(christos)

regen i386

(christos)

New rawrite32 version out (small bugfixes)

(martin)

Fix mergo

(skrll)

PR/50751: David Binderman: check bounds before dereferencing.

(christos)

PR/50750: David Binderman: Check bounds before dereference

(christos)

PR/50749: David Binderman: check bounds before dereference.

(christos)

PR/50748: David Binderman: check bounds before dereference

(christos)

PR/50747: David Binderman: check bounds before dereference.
While here add some continues before semicolons.

(christos)

PR/50746: David Binderman: Check bounds before dereferencing.

(christos)

switch amd64 to gdb-7.10.1

(christos)