Comment out EARLYCONS

(jmcneill)

initialize bs_cookie for generic_dsb tags

(jmcneill)

Fix rate calculation of fractional mode PLLs

(jmcneill)

PR/53371: Thomas Barabosch: Potential memory leak in usr.sbin/ndp/ndp.c

(christos)

Fix reading through /dev/gpio and use the same style for writing registers in both fdt and /dev/gpio cases

(jmcneill)

incldue -> include

(jmcneill)

Change aprint_error_dev to aprint_normal_dev for normal completion case.

(nakayama)

atomic_and/atomic_or do bitwise ops, not logical ops.

(dholland)

Actually, don't do anything if we switch to a kernel thread. When the cpu
switches back to a user thread the fpu is restored, so no point calling
fninit (which doesn't clear all the states anyway).

(maxv)

explicitly turn off tiling translation - now software rendered characters
look right again on r3xx hardware

(macallan)

PR/53368: Thomas Barabosch: Potential integer overflow in usr.bin/patch/inp.c

(christos)

arm: Add support for Rockchip RK3328 SoC.

(jmcneill)

Add initial support for Rockchip RK3328 SoC.

(jmcneill)

Add VERID reg bits

(jmcneill)

Add flag for disabling store-and-forward mode, and a callback for notifying
bus glue that the link speed has changed.

(jmcneill)

If clock-output-names is specified, use it for the clock name instead of the node name

(jmcneill)

fdtbus_clock_assign: simplify

(jmcneill)

Pull in arm/locore.h for arm_dsb() on aarch64

(jmcneill)

Fix Dd argument. Remove unnecessary whitespace.

(wiz)

Add missing word.

(wiz)

Sort SEE ALSO.

(wiz)

Fix fmttab value for L2CR_L2PE (parity enabled) so that we don't print
self-contradictory "no parity parity enabled".

(uwe)

Add missing .Fn mark-up

(pgoyette)

Add note that there is no way to delete an individual datum from a
container.

(pgoyette)

Bump date for previous changes

(pgoyette)

Improve synopsis, remove unnecessary .Nm markup

(pgoyette)

In mtmsr emulation ignore PSL_VEC we set in mfmsr emulation just
above, as it freaks out PSL_USEROK_P().  This is also congruent with
how we handle PSL_FP.

PR port-macppc/53360

(uwe)

Add POWER/PowerPC/PPC.
Leaving the original PPC entry as it's a valid case.

(sevan)

Fix pasto in .Dt title

(uwe)

cosmetics, no functional change

(macallan)

move 'utsname' to the main() function scope, so that the reference to
it outside the block remains valid.

should fix an asan reported issue.

(mrg)

Add basic ARM PL061 GPIO driver with FDT attachment.

Mostly for Qemu 'virt' platform; really needs interrupt support
in gpiokeys(4) and this driver to work efficently and reliably.

(jakllsch)

Add cfprint_t virtiobusprint() and use it for virtio_mmio.

(jakllsch)

Fix PSCI 0.1 detection.

(jakllsch)

Use correct value for PSCI 0.2+ PSCI_CPU_ON.

(jakllsch)

... -mcpu=armv7-a is actually spelled -march=armv7-a ...

(jakllsch)

PR/53367: Thomas Barabosch: Integer overflow in usr.sbin/dumplfs
While here use the "e" functions to always check for allocation errors.

(christos)

Whitespace nit.

(wiz)

Whitespace fixes.

(wiz)

Sort errors. Use \(em.

(wiz)

add FDT "qemu,fw-cfg-mmio" glue for qemufwcfg

(jakllsch)

'virt' platform adjustments:
- support broader CPU model range
- move kernel physical and virtual base to avoid bootloader,
  and keep KVA at or above 2GiB
- add relevant drivers

(jakllsch)

Big endian support for virt_platform_early_putchar and xputc.

(jakllsch)

fix xputc FIFO address for virt platform

(jakllsch)

Add "arm,arm-v7" to compatible strings.

(for `qemu-system-arm -M virt`)

(jakllsch)

Bump PIC_MAXSOURCES to 352.

(`qemu-system-arm -M virt` currently needs 288)

(jakllsch)

Add if_vlan directory to lists to fix the build failure

build.sh was failed when using "-V MKDEBUG=yes"

(yamaguchi)

simplify priq_class_create():  if the alloc for cl_q fails,
then only clean up required to to free cl.  delete all the
code at err_ret that was probably used in the past, and move
the free + return upon failure earlier.

(mrg)

Revert previous, mrg's fix is better (even with possible namespace issues,
which I guess are not great in compat includes) .... the perils of two
people doing fixes at around the same time.

(kre)

If we are going to use offsetof() we'd need to include <stddef.h> to
get it defined.  Rather than deal with potential namespace issues
with that, just clear the entire struct, rather than attempting to
stop after the potential padding field.  If the compiler is good enough
it should make no difference (there are just 3 fields, 2 named ones
are assigned to, immediately after the memset() - the compiler can
detect that, and not bother assigning (via memset()) to the unmamed
3rd padding field).  If the compiler is not smart enough to deal
with this, then I doubt writing 8 more zero bytes will make enough
difference to matter.

(kre)

include libkern.h or string.h & stddef.h, to get the offsetof()
and memset() definitions.

(mrg)

Regen for ticket #878.

(msaitoh)

Fix queue address setup bug introduced in 1.31.

This use-before-initialization problem resulted in non-zero queues never
being initialized with their virtqueue pointer, and queue zero's pointer
to be clobbered with the virtqueue pointer for the higest-indexed ring.

(jakllsch)

Add -n (don't send mail if command exited normally) option. From OpenBSD via
Job Snidjers.

(christos)

Replace with the OpenBSD man page. It removes some historical comparisons
that are not very useful (and trully if any they belong to a separate section
instead of being interspersed in the document), and organizes and formats
the information better.

(christos)

for consistency, print the statement that is true in the error case.
also add missing closing paren

(maya)

Test for FE_INVALID in a way that works for powerpc too.

powerpc seems to return FE_INVALID | FE_VXSOFT rather than just FE_INVALID.
XXX need extra careful reading of standards

(maya)

Tickets #877, #878, #879, #880, #881

(martin)

Pull up the following revisions, requestes by christos in ticket #881:

share/mk/bsd.sys.mk revision 1.280, 1.281

Create NETBSDOBJDIR to handle MAKEOBJDIRPREFIX, since that prepends the
prefix to the whole path.
Mention that we cannot handle MAKEOBJDIR

-

Handle more objdir variations for MKREPRO (untested)

(martin)

Pull up following revision(s) (requested by macallan in ticket #880):

sys/arch/macppc/conf/std.macppc: revision 1.24

set options ADBKBD_EMUL_USB by default

With this ADB and USB keyboards can coexist on the same mux, as needed by
built-in Bluetooth modules on some *Books

should fix PR53351

(martin)

Pull up following revision(s) (requested by ozaki-r in ticket #879):

sys/kern/subr_workqueue.c: revision 1.37

Don't wait on workqueue_wait if called from worker itself

Otherwise workqueue_wait never return in such a case.  This treatment
is the same as callout_halt.

(martin)

Pull up following revision(s) (requested by maya in ticket #878):

share/man/man4/rtsx.4: revision 1.5
sys/dev/pci/pcidevs: revision 1.1331
sys/dev/ic/rtsxreg.h: revision 1.3
sys/dev/ic/rtsxvar.h: revision 1.3
sys/dev/pci/rtsx_pci.c: revision 1.7
sys/dev/ic/rtsx.c: revision 1.3

add realtek RT525A PCI-E card reader
found in my Dell XPS 15 9550.

rtsx(4): Add support for RT525A, from openbsd.
ok rkujawa

Denote RTS525A, bump date

(martin)

Pull up following revision(s) (requested by zafer in ticket #877):

usr.sbin/user/user.c: revision 1.132

Set default homeperm values before writing usermgmt.conf file and
honor values in it instead of ignoring it.

Fixes PR bin/40324 by Guy Yur

(martin)

Tickets #1612, #1615 and #1616

(martin)

Pull up following revision(s) (requested by maya in ticket #1616):

share/i18n/esdb/ISO-8859/ISO-8859.alias: revision 1.4

Add more aliases for Hebrew and Arabic ISO-8859-... encodings.

ISO-8859-8 is supposed to be visual order (i.e. legible if displayed ltr)
ISO-8859-8-i is supposed to be implicit logic order
ISO-8859-8-e is supposed to be explicit about order

In practice, ISO-8859-8 implying visual order is rare, and logic
order is used. ISO-8859-8-e is rarely used.

Same for Arabic, which uses ISO-8859-6-...

Mentioned in RFC 1555, RFC 1556.

(martin)

Tickets #1612, #1615 and #1616

(martin)

Pull up following revision(s) (requested by maya in ticket #1616):

share/i18n/esdb/ISO-8859/ISO-8859.alias: revision 1.4

Add more aliases for Hebrew and Arabic ISO-8859-... encodings.

ISO-8859-8 is supposed to be visual order (i.e. legible if displayed ltr)
ISO-8859-8-i is supposed to be implicit logic order
ISO-8859-8-e is supposed to be explicit about order

In practice, ISO-8859-8 implying visual order is rare, and logic
order is used. ISO-8859-8-e is rarely used.

Same for Arabic, which uses ISO-8859-6-...

Mentioned in RFC 1555, RFC 1556.

(martin)

Ticket #1616

(martin)

Pull up following revision(s) (requested by maya in ticket #1616):

share/i18n/esdb/ISO-8859/ISO-8859.alias: revision 1.4

Add more aliases for Hebrew and Arabic ISO-8859-... encodings.

ISO-8859-8 is supposed to be visual order (i.e. legible if displayed ltr)
ISO-8859-8-i is supposed to be implicit logic order
ISO-8859-8-e is supposed to be explicit about order

In practice, ISO-8859-8 implying visual order is rare, and logic
order is used. ISO-8859-8-e is rarely used.

Same for Arabic, which uses ISO-8859-6-...

Mentioned in RFC 1555, RFC 1556.

(martin)

Pull up following revision(s) (requested by maya in ticket #1615):

usr.sbin/makefs/ffs.c: revision 1.70

PR/52828: Mark Johnston: makefs UFS2 lazy inode initialization is buggy
makefs(8) emulates UFS2 in performing lazy initialization of inode
blocks when allocating and writing inodes. However, it only ever
initializes one inode block at a time, which may be insufficient.

If so, a later initialization may clobber an inode, resulting in
an inconsistent filesystem.

I committed a minimal fix for the problem to FreeBSD:
https://svnweb.freebsd.org/changeset/base/326912

(martin)

Pull up following revision(s) (requested by maya in ticket #1615):

usr.sbin/makefs/ffs.c: revision 1.70

PR/52828: Mark Johnston: makefs UFS2 lazy inode initialization is buggy
makefs(8) emulates UFS2 in performing lazy initialization of inode
blocks when allocating and writing inodes. However, it only ever
initializes one inode block at a time, which may be insufficient.

If so, a later initialization may clobber an inode, resulting in
an inconsistent filesystem.

I committed a minimal fix for the problem to FreeBSD:
https://svnweb.freebsd.org/changeset/base/326912

(martin)

Pull up following revision(s) (requested by maya in ticket #1612):

sbin/fsck_lfs/bufcache.h: revision 1.14
sbin/fsck_lfs/bufcache.c: revision 1.20

PR/51418: Jose Luis Rodriguez Garcia: Fix incore src/sbin/fsck_lfs/bufcache.c
XXX: pullup-8, pullup-7

(martin)

Pull up following revision(s) (requested by maya in ticket #1612):

sbin/fsck_lfs/bufcache.h: revision 1.14
sbin/fsck_lfs/bufcache.c: revision 1.20

PR/51418: Jose Luis Rodriguez Garcia: Fix incore src/sbin/fsck_lfs/bufcache.c
XXX: pullup-8, pullup-7

(martin)

Install the FPU state on the current CPU in setregs (execve).

(maxv)

Eager FPU on i386.

(maxv)

make this work on rv350 chips found in last generation *Books
Needs testing on non-Apple hardware.

(macallan)

PR/53365: Thomas Barabosch: memory leak in route6d
XXX: pullup 7, 8

(christos)

SpectreV4, backports in NetBSD-8, no XSAVEOPT

(maxv)

Add some code to support eager fpu switch, INTEL-SA-00145. We restore the
FPU state of the lwp right away during context switches. This guarantees
that when the CPU executes in userland, the FPU doesn't contain secrets.

Maybe we also need to clear the FPU in setregs(), not sure about this one.

Can be enabled/disabled via:

machdep.fpu_eager = {0/1}

Not yet turned on automatically on affected CPUs (Intel Family 6).

More generally it would be good to turn it on automatically when XSAVEOPT
is supported, because in this case there is probably a non-negligible
performance gain; but we need to fix PR/52966.

(maxv)

arm: Add support for QEMU ARM Virtual Machine ("virt").

(jmcneill)

Add support for QEMU ARM Virtual Machine ("virt").

(jmcneill)

Increase storage space for FDT data

(jmcneill)

Keep up with HEAD.  The original change in the wsevent_50 compat code
has been moved to sys/time_types.h and will get picked up during the
next sync-with-HEAD

(pgoyette)

timespec_to_timespec50 - don't leak kernel memory in struct padding.

On 64-bit machines struct timespec50 has padding between 32-bit tv_sec
and long tv_nsec that is not affected by normal assignment.  Scrub it.
The result is most likely about to be passed to the user space by the
caller, so do it here to take care of them all.

(uwe)

Make the alignof() usage more portable in grep(1)

Remove NetBSD specific check and replace it with GCC extension __alignof__
that is available since GCC 2.x.

Requested by <kre>

(kamil)

Remove extra tabs. No functional change.

(msaitoh)

Update the error message in t_vlan (ENXIO => EINVAL)

ok ozaki-r@

(yamaguchi)

Fix to check whether the address has been added before delete

The list named ifv_mc_listhead saves multicast addresses that
are added through SIOCADDMULTI. Each nodes added to the list
are used for deleting the related address from a parent I/F
when remove the configuration of parent I/F.
In carp(4) and OpenBSD's vlan(4), the lists is used to check
a parameter of SIOCDELMULTI in addition to the use.
Based on them, the check is added to vlan(4)

ok ozaki-r@

(yamaguchi)

Widen shift to the LHS type.
same as aarch64/db_interface.c r1.4, PR/53338.

(ryo)

Add test cases for multicast address handling of vlan(4)

ok ozaki-r@

(yamaguchi)

Add the lock to refer the list included in ethercom for safety

The lock is already held while adding and deleting
ok ozaki-r@

(yamaguchi)

Remove ETHER_LOOKUP_MULTI()

The macro has been replaced with a function.
ok ozaki-r@

(yamaguchi)

Use ether_lookup_multi() instead of the macro

ok ozaki-r@

(yamaguchi)

Replace macros related to multicast address with inline functions

ok ozaki-r@

(yamaguchi)

Move macros related to multicast address into #ifdef _KERNEL

Those macros and structure are only used in the kernel.
reviewed by ozaki-r@n.o, thanks.

(yamaguchi)

s/regisrer/register/. No functional change.

(msaitoh)

The revision number of Yukon FE A2 is not 0x3 but 0x2. Same as other OSes.

(msaitoh)

Regen.

(msaitoh)

Add Mobility Radeon HD 4300(0x9552) and Mobility Radeon HD 4500(0x9553)
from OpenBSD.

(msaitoh)

remove notreached

(christos)

return to caller instead of exit(2)

(christos)

sort the models, and add several more variants; data found in OpenBSD eephy.c

this also makes makphy match model 88E3016, which adresses PR kern/53301
by Chris Humphries

(jdolecek)

regen

(jdolecek)

add E1000 with model id 0x0006, it exists according to OpenBSD
rename E1000 model 0x0000 to E1000_0 for consistency

(jdolecek)

fix device name in printf when not matched

(jdolecek)

actually recognize the Yukon 2 variants up to SK_YUKON_OPTIMA2 as such

(jdolecek)

Split out error reporting and make it compile without SCSIVERBOSE

(reinoud)

sync and adapt with OpenBSD if_msk.c rev. 1.43-1.65 (modulo some
cosmetics, stuff we already had, or stuff which does not apply due
to different frameworks):

- Add support for fiber on 88E8021/22
- Avoid unnecessary resets.  This should make fiber 88E8021/22 work completely
- Only probe phy's at address 0; gets rid of bogus phy's on the
  Marvell Yukon 88E8036
- Remove an unused function msk_setfilt() which was copied from sk(4)
- Make msk(4) detachable
- Simplify the combination use of pci_mapreg_type()/pci_mapreg_map()

(jdolecek)

revert rev. 1.29 'Only tick mii if the phy had a status change', as neither
OpenBSD nor FreeBSD drivers have this conditional

(jdolecek)

Detect properly overflow in expr(1) for 0 + INT

(kamil)

Avoid Undefind Behavior in DEFAULT_ALIGNMENT in GNU grep(1)

Replace homegrown logic of calculating alignment with alignof, a C11 header
feature.

Add a check for NetBSD >= 8.0 just in case that someone will want to build
it on older NetBSD.

Issue detected with UBSan.

(kamil)

revert previous MAKEOBJDIRPREFIX massaging

(christos)

revert previous MAKEOBJDIRPREFIX massaging

(christos)

revert previous MAKEOBJDIRPREFIX massaging

(christos)

revert previous MAKEOBJDIRPREFIX massaging

(christos)

Create NETBSDOBJDIR to handle MAKEOBJDIRPREFIX, since that prepends the
prefix to the whole path.
Mention that we cannot handle MAKEOBJDIR

(christos)

PR/53362: Thomas Barabosch: Use strlcpy() to prevent theoretical buffer
overflow.

(christos)

Revert "use len=0 for max length transfers" part of previous commit.

PR: port-evbarm/53359

(jmcneill)

libc: remove explicit cast NULL in atoi

There isn't any reason to cast NULL so just remove it. A similar change
was already made in DragonFly and FreeBSD (by me).

ok fly@

(eadler)

Disable sanitizers for libm and librt

The sanitizers are designed to use pristine OS versions of these libraries,
install interceptors for public symbols in them and link with them in every
sanitized executable.

Having unmodified base libraries (c, m, rt, pthread) also makes the system
usable with programs linking with them and without sanitization, as there
are no fatal issues during the resolving of missing symbols.

The remaining libraries like libutil, libintl etc are prebuilt with
sanitization and all their users in the MKSANITIZER=yes distribution are
required to use the same sanitizer.

libc and libpthread are already marked with the NOSANITIZER flag.

(kamil)

Don't wait on workqueue_wait if called from worker itself

Otherwise workqueue_wait never return in such a case.  This treatment
is the same as callout_halt.

(ozaki-r)

Avoid declaring ulonglong_t.

Comment here already says it isn't necessary and will be removed.
Hasten the process for us and avoid the problematic typedef.

This is useful because u_longlong_t would otherwise need to be
provided by compat headers.

(maya)

Revert my previous config test inclusion for u_longlong_t.

Fixing this differently by avoiding the need for the definition.
It also caused a problem in OS X, as I neglected to include a correct header

(maya)

Refer ipsec(4) instead of fast_ipsec(4)

Because fast_ipsec(4) is retired and IPSEC is described in ipsec(4) now.

(ozaki-r)

Retire fast_ipsec.4

We switched to Fast IPsec at NetBSD 6.0 and that's the IPsec implementation of
us now.  So we don't need to have a separate manual.  Merge fast_ipsec.4 into
ipsec.4 and remove fast_ipsec.4.

(ozaki-r)

Keep up with -HEAD

(pgoyette)

wsevent_copyout_events50 - don't leak garbage from the kernel stack.

On 64-bit machines struct timespec50 has padding between 32-bit tv_sec
and long tv_nsec that is not affected by normal assignment.  Scrub it
before we uiomove struct owscons_event.

(uwe)

macppc IPsec change from last week

(sevan)

Move the typedefs before the header inclusions (except <features.h>) and
explain why.

(christos)

If setting rate on a clock with no set_rate function, and the desired rate
matches the current rate, don't return an error.

(jmcneill)

move SK_HASH_BITS to if_skreg.h and reuse in if_msk.c (from OpenBSD)

(jdolecek)

Set NOSANITIZER in i386 mbr files

This is required to build these files and keep then functional with
enabled MKSANITIZER.

Sponsored by <The NetBSD Foundation>

(kamil)

Prevent Undefined Behavior in shift of signed integer in grep(1)

There is an interface in grep: dfa.c with functions to test bit, set bit,
clear bit etc. They operate over the INT-wide mask (INTBITS). Use unsigned
shift in these interfaces to prevent UB.

Detected with MKSANITIZER with the undefined behavior sanitizer option.

Sponsored by <The NetBSD Foundation>

(kamil)

match three more variants - DGE550T_B1, 8040T, 8042

sync more sk_type conditionals from OpenBSD

(jdolecek)

regen

(jdolecek)

add two more Yukon variants (from OpenBSD)

(jdolecek)

Remove duplicate assignments.
Fixes PR bin/51512 by Jose Luis Rodriguez Garcia

(zafer)

fix whitespace

(jdolecek)

Add 2 new expr(1) ATF tests

Assert that -9223372036854775808 % -1 and -9223372036854775808 / -1 return
message about overflow / underflow detected.

These tests pass correctly.

Sponsored by <The NetBSD Foundation>

(kamil)

Rework perform_arith_op() in expr(1) to omit Undefined Behavior

The current implementation of operations - + * / % could cause Undefined
Behavior and in narrow cases (INT64_MIN / -1 and INT64_MIN % -1) SIGFPE
and crash duping core.

Detected with MKSANITIZER enabled for the Undefined Behavior variation:
# eval expr '4611686018427387904 + 4611686018427387904'
/public/src.git/bin/expr/expr.y:315:12: runtime error: signed integer overflow: 4611686018427387904 + 4611686018427387904 cannot be represented in type 'long'

All bin/t_expr ATF tests pass now in a sanitized userland.

Sponsored by <The NetBSD Foundation>

(kamil)

876

(snj)

Pull up following revision(s) (requested by ozaki-r in ticket #876):
sys/net/if_vlan.c: 1.126
tests/net/if_vlan/t_vlan.sh: 1.9
vlan: call ether_ifdetach without IFNET_LOCK
Fix PR kern/53357
--
Add tests of vlan with bridge
The tests trigger a panic reported in PR kern/53357.

(snj)

remove extra quotes.

(christos)

- instead of hard-coding the include paths in mkioctls, pass them in
  from the Makefile so that they are consistent.
- do more sed so that destination paths are not hard-coded inside #include
  statements.

(christos)

Change typ of tilde_ok from int to unsigned int in ksh(1)

UBSan can detect that during switching a login to root there is unportable
left shift operation:

$ su -
Password:
/public/src.git/bin/ksh/eval.c:598:13: runtime error: left shift of 1073741824 by 1 places cannot be represented in type 'int'
#

Sponsored by <The NetBSD Foundation>

(kamil)

Do not use index out of bounds in nawk

$ awk '{w=$1}' < /dev/null
/public/src.git/external/historical/nawk/bin/../dist/lex.c:476:16: runtime error: index -1 out of bounds for type 'const Keyword [46]'

There used to be documented a bug in the code that index ouf of bounds
can in theory fault (by daniel barrett).

Before assigning the pointer, first check for the index whether it's not
not -1. This was a suggested solution in the comment in the code.

The sanitizer is overcautious as this pointer wasn't dereferenced, but
fix is nonetheless.

Sponsored by <The NetBSD Foundation>

(kamil)

When initiating a transfer, if a device isn't present, we won't
get an ACK after sending the address.  Check for this alternate
state and suppress the error message when it occurs.

Fixes PR kern/53356.

(thorpej)

Process assigned clock parents and rates on clock provider nodes.

(jmcneill)

Check if ether_ifdetach is called without INET_LOCK

(ozaki-r)

Add tests of vlan with bridge

The tests trigger a panic reported in PR kern/53357.

(ozaki-r)

vlan: call ether_ifdetach without IFNET_LOCK

Fix PR kern/53357

(ozaki-r)

Correct Undefined Behavior in gzip(1)

Unportable left shift reported with MKSANITIZER=yes USE_SANITIZER=undefined:

# progress -zf ./games.tgz  tar -xp -C "./" -f -
/public/src.git/usr.bin/gzip/gzip.c:2126:33: runtime error: left shift of 251 by 24 places cannot be represented in type 'int'
100% |****************************************************************************************************************| 44500 KiB  119.69 MiB/s    00:00 ETA

Refactor the following code into something that is more clear
and fix signed integer shift, by casting all buf[] elements to
(unsigned int):

unsigned char buf[8];
uint32_t usize;
[...]
else {
    usize = buf[4] | buf[5] << 8 |
            buf[6] << 16 | buf[7] << 24;
[...]

New version:

    usize = buf[4];
    usize |= (unsigned int)buf[5] << 8;
    usize |= (unsigned int)buf[6] << 16;
    usize |= (unsigned int)buf[7] << 24;

Only the "<< 24" part needs explicit cast, but for consistency make the
integer promotion explicit and clear to a code reader.

Sponsored by <The NetBSD Foundation>

(kamil)

If we don't have a native driver for a node and it has a simple-mfd
compatible string, treat it the same as a simple-bus.

(jmcneill)

switch from MAKEOBJDIRPREFIX to MAKEOBJDIR.

(christos)

switch from MAKEOBJDIRPREFIX to MAKEOBJDIR.

(christos)

Switch from using MAKEOBJDIRPREFIX to MAKEOBJDIR

(christos)

switch CFLAGS from using MAKEOBJDIRPREFIX to MAKEOBJDIR. This whole thing
is disgusting, but fixing it properly. requires restructuring

(christos)

Avoid left-shifting a negative number in the non-compat case, too.

(maya)

Fix fdtbus_get_cells(, 2) when running big endian.

(jakllsch)

fill in vexpress_platform_early_putchar(); make comment about CONSADDR

(jakllsch)

fix xputc for big endian

(jakllsch)

Factual correction: actually old code also used number of octents, but set
sc_if->sk_tx_ramend to invalid value, and assumed 128k buffer when
SK_EPROM0 read returned zero

(no code change)

(jdolecek)

Now we use timestamp info, use direct console output instead of kernel
timestamped output for usage() reporting

(reinoud)

Prevent nested SIGIOs ruining the system stack

(reinoud)

adopt tx/rx buffer size computation from OpenBSD rev. 1.51 to fix
"random" msk failures

the crucial fix is that the start/end adresses are computed and
written to hw registers in number of octets (i.e. size /
sizeof(u_int64_t)), as that is actually what the hardware expects

fixes PR kern/36454 (which had fix which highlighted the trouble),
and seems likely to also fix kern/35711 and port-i386/42514

(jdolecek)

Add configure check and fallback definition for u_longlong_t

Should help linux tools compilation of dtrace tools, and not
affect NetBSD.

From Chuck Zmudzinski in current-users, with light modification
by myself.

(maya)

Correct Undefined Behavior in libc/citrus

Unportable left shift reported with MKSANITIZER=yes USE_SANITIZER=undefined:

# nm /usr/lib/libc.so|grep sanit
/public/src.git/lib/libc/citrus/modules/citrus_mapper_std.c:173:8: runtime error: left shift of 1 by 31 places cannot be represented in type 'int'

Sponsored by <The NetBSD Foundation>

(kamil)

Correct Undefined Behavior in ifconfig(8)

Unportable left shift reported with MKSANITIZER=yes USE_SANITIZER=undefined:

# ifconfig
alc0: flags=0x8843<UP,BROADCAST,RUNNING,SIMPLEX,MULTICAST> mtu 1500
        ec_capabilities=3<VLAN_MTU,VLAN_HWTAGGING>
        ec_enabled=0
        address: xx:xx:xx:xx:xx:xx
/public/src.git/sbin/ifconfig/af_inet.c:102:34: runtime error: left shift of 16777215 by 8 places cannot be represented in type 'int'
        inet 192.168.0.38/24 broadcast 192.168.0.255 flags 0x0
        inet6 xxxx::xxxx:xxxx:xxxx:xxx%alc0/64 flags 0x0 scopeid 0x1
lo0: flags=0x8049<UP,LOOPBACK,RUNNING,MULTICAST> mtu 33624
        inet 127.0.0.1/8 flags 0x0
        inet6 ::1/128 flags 0x20<NODAD>
        inet6 fe80::1%lo0/64 flags 0x0 scopeid 0x2

Change shifting left 1 to shifting 1U. This corrects the issue.

        if (cidr < 32) {                /* more than 1 bit in mask */
                /* check for non-contig netmask */
                if ((mask ^ (((1 << cidr) - 1) << (32 - cidr))) != 0) // <- here
                        return -1;      /* noncontig, no pfxlen */
        }

Solution suggested by <uwe>

Sponsored by <The NetBSD Foundation>

(kamil)

use LIBISCXX = yes, instead of hard-coding the c++ library. Suggested by joerg

(christos)

Restore the MKGROFF=bo MKCXX=yes build

Mark the documentation in dc(1), gprof(1), rogue(6) and fsck_ffs(8) with
the .roff flag in SUBDIR.

Sponsored by <The NetBSD Foundation>

(kamil)

Add libstdc++ again; needed for:
undefined reference to `typeinfo for __cxxabiv1::__class_type_info'
undefined reference to `typeinfo for __cxxabiv1::__si_class_type_info'
undefined reference to `typeinfo for __cxxabiv1::__vmi_class_type_info'
undefined reference to `typeinfo for std::type_info'
undefined reference to `__dynamic_cast'

(christos)

Apply changes of FreeBSD ix-3.2.18.tgz:
- Add IXGBE_DEV_ID_82599_LS(0x154f) support.
- Increase timeout for newer than ixgbe_mac_X550 (i.e. Xeon-D and Denverton)
  in ixgbe_acquire_swfw_sync_X540().
- Add some unused macros (IXGBE_FW_LESM_*).

(msaitoh)

Ticket #875

(martin)