avoid a GCC warning with this:

+#if defined(__GNUC__) && (__GNUC__ >= 5) && defined(__OPTIMIZE__)
+      x.fp_sign = 0;
+#endif

(ridiculous, but seems better than disabling the warning entirely.)

(mrg)

workaround a problem in old binutils and GCC 5.3.

(mrg)

bump the image size.

(mrg)

Use LIST_FOREACH instead of LIST_FOREACH_SAFE

No need to use *_SAFE because we don't remove any items in the loop.

(ozaki-r)

slightly rearrange the generation of SRCS.

(mrg)

+HSM is also hierarchical storage management

(ginsbach)

add some stuff; PR 50992 from Nikolai Lifanov

(dholland)

mknative-gcc for coldfire and GCC 5.3.

(mrg)

1144

(snj)

Pull up following revision(s) (requested by martin in ticket #1144):
lib/librt/shm.c: revisions 1.2, 1.3
Allow /var/shm to be a symlink to a properly set up directory.
--
Simplify previous by just doing an open() on the directory and then using
the fstat* family to test for all relevant details.
Clean up buffer sizes and clarify a length check. Based on input from dholland.

(snj)

remove a bunch of "@echo done" from the tail of rules.  these messages
were vaguely useful back when we didn't run make -j, but now you end
up with a single line "done" every so often, with no idea what it is
for.  very few other targets claim they're done so just remove these.

(mrg)

updates to latest -current:
- sshd problem is libldap.so.4.3, at least on sparc64.  placing with 4.8 one works
- alpha mostly works fine, 7 new failures in atf, plus sshd problem
- update release build info for many *earm*, hppa, i386, amd64, mipsel, sh3*
- *mips* now at least completes mknative-gcc
- sh3eb has a mknative-gcc problem

(mrg)

mips64 does not get (all?) the non sw-float version.  fixes double symbol issues.

(mrg)

apply -Wno-error=stack-protector in the same places for GCC 5.3 as 4.8.
fixes sh3 and m68k issues, at least.  (vax and or1k are also included.)

(mrg)

mknative-gcc for m68k and m68000 and GCC 5.3.

(mrg)

mknative-gcc for mips64* and GCC 5.3.

(mrg)

mknative-gcc for GCC 5.3 and mipseb.

(mrg)

XXXGCC5: add symlinks for SH .cc files.

(mrg)

mknative-gcc for GCC 5.3 and mipseb.

(mrg)

hand-edit some not-yet-regenerated files that break the libstdc++
includes during mknative.

(mrg)

mknative-gcc for GCC 5.3 and mipsel.

(mrg)

revert part of rev 1.66 that converted a tri-state variable into
a bool while leaving a comparision against "< 3".

(mrg)

put back NUL check (Ingo Schwarze)

(christos)

Fix reversed condition in tty_end() (Ingo Schwarze)
Also don't succeed if calling setup twice.

(christos)

avoid some stupid GCC errors seen only with -Os and -O3.

(mrg)

regen with old assembly stubs.

(christos)

revert change from openssl-1.1.0-pre4, breaks gcc-4.8

(christos)

Install ftttdrv.h, which was apparently overlooked at some point
(probably because it contains little of substance...) PR 50902.

(dholland)

fix constness (from kre)

(christos)

Add command substitution test.

(christos)

new command substitution test from kre

(christos)

Move the command substitution "internal subroutine" part of
readtoken1() into a real function of its own (inspired by a
similar change made by FreeBSD - the other internal routines
they moved out are expected to move out here as well soon.)
This change helps avoid gcc 5.3 demanded (not required!) volatile
noise which would slow down parsing. (from kre)

(christos)

regen

(christos)

elide bug with new cpuid code.

(christos)

bring newer versions from 1.1.0-pre4

(christos)

re-gen to fix sha1. there were also improvements for montgomery multiplications
that we did not have from a previous change.

(christos)

move scratch stack offset from 64 to 72 bytes so that it is aligned properly.
We core-dump using xmm instructions with gcc-5.3 otherwise. This is all
fixed a lot better in openssl-1.1.0-pre4 and I am tempted....

(christos)

update alpha (mostly working) and hppa (mknative-gcc works)

(mrg)

mknative-gcc for HPPA and GCC 5.3.  thanks for fixing fenv.c, nick!

(mrg)

Sort errors. Improve markup.

(wiz)

PR/50460: Abhinav Upadhyay: Fix legacy apropos query to match both the name
and the one line description and delete extra args.

(christos)

simplify.

(christos)

use foo - 0 instead of testing for defined(foo); from joerg@

(christos)

Let PaX ASLR know about the current emulation

This effectively fixes PaX ASLR with 32-bits emulation on 64-bits
platforms. Without this knowledge, the offset applied for 32-bits
programs was really meant for a 64-bits address space - thus
shifting the address up to 12 bits, with a success rate of about
1/4096. This offset is calculated once in the lifetime of the
process, which therefore behaved normally when able to start.

Fixes kern/50469, probably also kern/50986

Tested on NetBSD/amd64 (emul_netbsd32)

(khorben)

Update {read,write}fpsr to work everywhere

(skrll)

PR/50983: David Shao: Protect stddef;
also make the pointer void *.

(christos)

Pass a correct residual size to wdcbit_bucket() that drains extra bytes.

Seems broken (but almost harmless?) since rev 1.101 committed ~10 years ago:
http://mail-index.netbsd.org/source-changes/2006/09/07/0043.html

(tsutsui)

Remove extra spaces.

(tsutsui)

Whitespace

(skrll)

Whitespace

(skrll)

Whitespace

(skrll)

Provide and use DPRINTF{,N} macros in the same way as all other HCDs.

No functional change.

(skrll)

Comment out unused logic. PR 50973

(dholland)

Disable uscanner in all kernel configs

As discussed on current-users@, SANE uses ugen via libusb and not
uscanner, so users are not well served by having uscanner.  Consensus
is that addressing how to adjust permissions for scanners should not
block restoring basic functionionality.

(Compile-tested only, but there are multiple reports of this being the
right approach.)

(gdt)

Save and restore all the registers we need instead of playing tricks and
"knowing" which registers the compiler clobbers. gcc-5.3 clobbers both
rcx and rdx...

(christos)

remove some dead code.

(mrg)

disable -fdelete-null-pointer-checks by default for now.

(mrg)

note alpha mostly works (sshd problem as well)
add that sparc64 test results are about 60 additional failures

(mrg)

Allow enabling and disabling PaX ASLR debug

(christos)

Default kernel builds to USE_SSP=yes.

(martin)

Add MAP_FMT

(christos)

Sync documentation of struct ktr_header with reality:
  - there is a ktr_version field
  - the ordering of ktr_type and ktr_version is byte-order dependent
  - ktr_time is now a timespec, not a timeval
  - there's now a ktr_lid field
  - there is no such thing as a ktr_buf field and it's certainly not
    a caddr_t.

XXX: I would appreciate it if someone who knows how this stuff works
XXX: could review this file in case it's still lying in some more
XXX: subtle way.

(dholland)

Update documented prerequisite includes. PR 50633 and PR 50634.

Remove sys/uio.h, and swap sys/param.h for sys/types.h as sys/ktrace.h
depends on MAXCOMLEN.

Bump date; first time since 1993.

(dholland)

PR 50633 and PR 50634 from Brad Harder: add missing includes:
sys/signal.h and sys/time.h. Also add sys/uio.h which is needed and
for some probably historical/obsolete reason has been documented as a
prerequisite instead of included directly.

This also requires sys/param.h for MAXCOMLEN (blah) but for now at
least shift that to the documentation instead of including it
directly as it's a sewer.

(dholland)

Adapt to branch

(skrll)

Adapt to the branch.

(skrll)

Sync with HEAD

(skrll)

Initialise up_serialise to true and add a KASSERT.

(skrll)

add missing insn-modes.h files from earlier regen.

(mrg)

sprinkle more volatile (distribution build with gcc-5.3)

(christos)

kill more mudflap for gcc-5.3

(christos)

sprinkle more volatile (needed for the rescue build with gcc-5.3)

(christos)

Add more stack-protector elides

(christos)

write references as pointers XXX:

(christos)

treat class line struct and union

(christos)

Add support for c++ classes.

(christos)

spell restrict

(christos)

fix printf format

(christos)

Add references, comments.

(christos)

add references.

(christos)

Add references.

(christos)

match the updated <sys/types.h> definition

(christos)

tidy up: parse DW_AT_specification like DW_AT_abstract_origin

(christos)

- Add processing for c++ references.
- Make sure we load the DIE that contains the types of array elements so
  we can resolve them later.
- Print t_id (die offsets) in hex.

(christos)

swap first, then write.

(christos)

Use the correct upper bounds for the types array.  The correct upper
bound is typecnt not timecnt.  Now perpetual 'standard' time zones
will work correctly as they have a typecnt of 1 but a timecnt of 0.

(ginsbach)

update info for:
hppa: libc build failes in fenv.c
m68*: libc fails with stack protector issues
mips*: libgcc fails
sh*: initial stuff works
sparc*: sshd is broken for both, perhaps kernels are SMP problematic
vax: tools/gcc failes

(mrg)

move the compiler hack closer to the source of the fail

(mrg)

minimal changes necessary to link into an INET6-less kernel.

(mrg)

find all libgcc sources

(mrg)

disable our local hack to check alloc alignment on arm, as the define
is no longer compile-time constant and CTASSERT() fails.

(mrg)

mknative-gcc for GCC 5.3 and sh3*, hppa, m68*, mips*.

none of them are complete yet.

(mrg)

update info for alpha, powerpc, earmv4, i386, and sparc*

(mrg)

initial mknative-gcc for GCC 5.3 and alpha/powerpc.

(mrg)

set fixes for GCC 5.3.

(only arm is tested so far, but there's a pattern i can see :-)

(mrg)

regen.

(mrg)

delete the GCC 4.8 status, and update the 5.3 one a little.

(mrg)

set NETBSD_GCC_VERSION for GCC 5.3 separately.
don't collect crtstuff anymore.

(mrg)

generate mknative-gcc for GCC 5.3 and all the arm targets.
minor regen for x86 and sparc*

(mrg)

document the state of gcc 5.3 in src.

(mrg)

remove unused label.

(christos)

PR/50974: David Binderman: Remove unused code.

(christos)

Handle DW_AT_specification better; also don't core-dump if at == NULL.
XXX: Is that right?

(christos)

Add DPADD's

(christos)

Added ldexpl to ldexp manpage since it is missing.
Also added ldexpf manpage link since it is missing.

(nros)

undo the dso protected for clang

(christos)

disable dso protected for clang now, so that packages can be built.
(again this is the binutils warning about copy relocations on protected objects)

(christos)

Regen without #line directives.
- nbyacc -l -d -o initparse.c parse.y
- nblex -L -t -p scan.l > initscan.c

(nakayama)

Expose debugging

(christos)

Apply our changes dropped in previous merge.

(nakayama)

s/sed/${SED}/

(christos)

put back the complex sed/awk since the code can't handle unsorted or repeated
entries (Rin Okuyama)

(christos)

Fix typo, add whitespace for better formatting, bump date for previous.

(wiz)

Bump date for previous.

(wiz)

All HCDs were fighting the seriaisation of transfers in usbdi.c for isoc
transfers.  Instead allow the HCDs to specify which pipes can handle
removing this serialisation and apply it appropriately.

dwctwo(4) can handle this for all transfer types, but only enable
bulk/isoc for now.

(skrll)

mknative bootstrap-libgcc GCC 5.3 for most arm platforms.  having
minor issue with the next step.

regen minor bits for amd64.

(mrg)

don't build a fake insn-modes.h for arm.

(mrg)

Protect read of isoc->next with bus lock

(skrll)

Fix matching of ICMP queries when NAT'd through IPF

This notably fixes MTU updates for hosts issueing ICMP queries through a
NAT performed by NetBSD with IPF.

(khorben)

strtod yields ERANGE for below-subnormal magnitudes, not underflow.

For a floating-point computation, in the language of IEEE 754,
`underflow' means the output was rounded and is too small to be
represented *normally*.

There are many nonzero floating-point numbers to which the exact
output may have been rounded -- namely subnormals.  The condition
under which strtod returns ERANGE for small magnitudes is when the
magnitude of the exact result is so small it is rounded to zero, not
even to a subnormal.

While here, use parallel language about large magnitudes instead of
the (albeit correct) word `overflow', to avoid temptation to treat
`underflow' as the opposite notion with zero instead of infinity.

(riastradh)

Avoid longjmp clobbering

(christos)

s/auto_ptr/unique_ptr/

(christos)

fix const issues (Q+D)

(christos)

Fix off-by-one.

(christos)

more casting fixes

(christos)