Fix a bad mov pc,lr -> RET

(matt)

Match arm*-*-netbsdelf*-*

(matt)

Add small helper script to keep lists sorted.

(joerg)

Sort entries in strcmp order.

(joerg)

Remove commented entries.

(joerg)

Add forgotten conversions of ja_JP for the COMPOUND_TEXT encoding.

(joerg)

Provide UTF-8 variants for all existing locales. The data is derived
from the Unicode Common Locale Data Repository.

Convert non-UTF-8 versions from the UTF-8 version using iconv and some
ad-hoc transliterations using sed.

Use EUR as currency_symbol in ISO8859-1.

Invert the Norwegian handling. no_NO is an alias for nb_NO as the latter
is used e.g. in CLDR.

Provide the Serbian locales in both Cyrilic and Latin script versions.
The alias is choosen based on the character set for the non-UTF-8 case
and Cyrillic is the default for UTF-8.

(joerg)

regen

(pooka)

Linux treats the data parameter as a pointer/long too.  The comments
in the sys/ptrace.h on Linux say it should be an int, but the code
glibc/musl/kernel disagrees.

(pooka)

Bump symtab space.

(riz)

RPI has an ARMv6 processor, so handle RPI_INSTALL properly when
building for the earmv6* MACHINE_ARCHes.

(riz)

A number of new earm MACHINE_ARCH have shown up recently; make sure
we can deduce their endianness.

(riz)

Don't try to estrdup() NULL; thanks martin@

(dholland)

Add the new xsh(4) driver to GENERIC.in template.

(rkujawa)

List all required libraries by libgssapi and libkrb5 for NOPIC builds.

(tsutsui)

Note X-Surf 100 support.

(rkujawa)

s/xsurfbus/xshbus/

(rkujawa)

Use the proper NE2000_TYPE for AX88796.

(rkujawa)

Add support for AX88796(B) variant.

(rkujawa)

regen

(pooka)

Use long instead of int for ptrace addr.  This makes things compile
better on powerpc64 with a compiler which warns about int-to-pointer
casts.
XXX: sys/ptrace.h on Linux says that the addr parameter is a "void *".

(pooka)

ioconf/pseudo-root have been in here unchanged for 3+ years and their use
has spread from rump kernels into dozens of kernel modules.  Let's pretend
we don't want to see config yell about them being experimental every time.

(pooka)

Use Va markup to show that the "suffix" part of ".PATH.suffix"
is a placeholder, not a literal string.
Also bump date.

(apb)

Fix another automatic variable that is life both pre- and post the siglongjmp
(which gcc calls "might be clobbered") by marking it volatile: on some
machines, this value might end up in a register loaded before the sigsetjmp,
but not preserved by the siglongjmp later.
Fixes the VAX build.

(martin)

Include powerpc+evbpcc.  Allows the buildrump.sh checkout to work on
powerpc machines.  Tested on Linux/ppc64 by Justin Cormack.

(pooka)

Allow to build in envs without USRSTACK (namely, rump kernels on powerpc).
TODO: all of the m*() calls are by definition incompatible with rump
kernels; including them is no great damage, but they're not useful either.

(pooka)

WARNS=5

(dholland)

Passes WARNS=5, so don't set WARNS=4.

(dholland)

Passes WARNS=5, so don't set WARNS=3.

(dholland)

add more static

(dholland)

add more const

(dholland)

WARNS=5

(dholland)

Remove redundant WARNS=5.

(dholland)

Remove redundant WARNS=5.

(dholland)

Pass WARNS=5.

(dholland)

Passes WARNS=5; don't set WARNS=4.

(dholland)

This passes WARNS=5.

(dholland)

These pass WARNS=5, so don't set WARNS=4

(dholland)

WARNS=5 is the default in sbin, don't need to set it explicitly.

(dholland)

Use siglongjmp to jump out of signal handlers.

(dholland)

Pass WARNS=5.

(dholland)

Convert some more conditional instructions to unified syntax

(matt)

Support thumb

(matt)

Switch to unified syntax
use RET/RETc
use push/pop

(matt)

Use foo{s} since it reduces the ifdefs for thumb

(matt)

Document .PATH.suffix:. Noted by apb back in December of 2008
(that's how long my queue is...)

(dholland)

unified syntax, make it play nice with thumb

(matt)

Switch to unified syntax
use it to get condition return on thumb

(matt)

Kill off uo_unmark_vnode/UFS_UNMARK_VNODE as it's now a leftover.

(dholland)

more dark ATF rituals are required

(dholland)

Turn off -fno-strict-aliasing for HAVE_GCC; it makes no difference to
the compiler output and is therefore ipso facto not necessary. I'll
leave it on for HAVE_LLVM as I can't check that tonight.

(If there is invalid code in here that actually requires it, let me
know so I can fix things properly.)

(dholland)

WARNS=5

(dholland)

typo in comment

(dholland)

this already passes WARNS=5

(dholland)

Use push/pop

(matt)

Use unified syntax for conditional instructions
Use RETc/RET

(matt)

fix conditional instructions to conform to unified syntax
use push/pop
use RET/RETc

(matt)

Change r12 -> ip
Change all condition instructions so they pass unified syntax

(matt)

Whitespace.

(dholland)

Add some cases pertaining to another bug I just fixed.

(dholland)

Fix another bug, relating to parsing the [=x=] syntax.

(dholland)

Tidy up more. This is pretty ratty code.

(dholland)

Tidy up.

(dholland)

Pass -Wpointer-sign.

(dholland)

avoid undefined behavior.

(dholland)

Don't explicitly zero bss variables that are already zeroed.

(dholland)

Add comments explaining how this works.

(dholland)

Simplify the parser handling.

(dholland)

Remove silly written-out initialization of string1.

(dholland)

Remove xfail for PR 48113.

(dholland)

Make tr -c work. Fixes PR 48113.

When -c is in effect, scan string1 first, complement it, and then
iterate over the results while scanning string2. Otherwise, scan
string1 and string2 together as before.

(dholland)

apply some CSE

(dholland)

Restore preexisting wrong behavior of tr -c (translates all characters
to the last letter in string2) instead of a different wrong behavior
(translates all characters to 0xff) accidentally just introduced.

(dholland)

sprinkle const

(dholland)

Make the parser state opaque. While here, fix a memory leak when using
the [:foo:] syntax.

(dholland)

Expose less of the parser state outside str.c.

(dholland)

Fix tool build under Mac OS X by using gettimeofday(2) instead of
clock_gettime(2). This needs a proper auto-config check.

(tron)

Tickets 925, 929.

(riz)

Pull up following revision(s) (requested by matt in ticket #929):
sys/dev/i2c/w83795g.c: revision 1.1
sys/dev/i2c/w83795greg.h: revision 1.1
sys/dev/i2c/files.i2c: revision 1.50
Add driver for Nuvoton W83795G voltage/temp/fan/gpio monitoring device.
As used in the HP Microserver N36L/N40L/N54L:
w83795g* at iic? addr 0x2f
gpio* at gpiobus?
Note that the gpio and watchdog functionality appears to be useless on
the Microserver, as the gpio pins don't affect the LED's and a watchdog
trigger does not cause a reset. Perhaps the reset pin just isn't wired.
There are quite flexible fan control options as well as just monitoring,
but it is not clear to me how best to export that function to userspace.

(riz)

Pull up following revision(s) (requested by he in ticket #925):
sys/arch/i386/stand/lib/bootmenu.c: revision 1.11
sys/arch/i386/stand/lib/bootmenu.h: revision 1.3
sys/arch/i386/stand/boot/boot2.c: revision 1.59
Two changes for the i386 boot loader related to the boot menu which
can be defined in boot.cfg:
* Add a "menu" command which re-displays the menu and initiates
  the timed countdown
* Use any default command defined in boot.cfg as default args
  if the user runs "boot" with no arguments
This is useful in circumstances where you e.g. need to interrupt
the normal boot process to switch to serial console, and where
simply "boot netbsd" is no longer sufficient (e.g. as with install
media which needs the miniroot kernel module loaded).

(riz)

+tp

(dholland)

+tp

(dholland)

Add tests for tr, including an xfail one for PR 48113.

(dholland)

PR/48112 - Kai-Uwe Eckhardt -- add support for Sitecom N300 USB WIFI adapter

(jnemeth)

ParseGetLine: don't treat a zero byte as end of buffer if P_end says it isn't.
Consume up to next newline, and issue a parse warning.
If no newline found before P_end, carry on as before.

(sjg)

PR/48112 - Kai-Uwe Eckhardt -- add support for Sitecom N300 usb wifi adapter

(jnemeth)

punctuation usage

(dholland)

Make these under Thumb2

(matt)

Never modify NLO/NHI (r2,r3) only LO/HI (r0,r1) so subsequent loops will work.

(matt)

Use r2 instead of ip

(matt)

Silence -Wpointer-sign on crypt-sha1.c.
This requires casts, which is not entirely desirable; however, this way
at least no *more* pointer sign issues can creep in.

Output object files are unchanged on amd64.

(dholland)

Use subs/sbc
Use push/pop
Don't adjust return for atomic_dec_64 since it's void and returns nothing

(matt)

use push/pop

(matt)

Use push/pop

(matt)

use push/pop

(matt)

cmpne -> cmp
Use push/pop

(matt)

Don't adjust return of atomic_inc_64 since it's void and doesn't return
anything.

(matt)

Fix add -> adds

(matt)

Redo -r1.53 correctly.

(dholland)

Make sure to check if the driver has a valid intr handler in virtio_detach().

Fixes a panic during shutdown on KVM under ubuntu 13.04 with virtio,
as reported in PR kern/48105 by Richard Hansen.

Should be pulled up to netbsd-6 branches.

(tsutsui)

thanks to mhitch, we got all the interrupt issues sorted out for now,
so turn off PDB_INTR in the debug kernel.

(mrg)

I don't have the resources to maintain OpenLDAP in NetBSD anymore.

(lukem)

Comment out pseudo-device md so that ARMADAXP_INSTALL passes config(1).
Fixes evbarm releases.

(aymeric)

regen

(jmcneill)

Fix 32 bit issue in main file read-in function. On both 32 bit and 64 bit
hosts a missing cast would result in `garbage' after the 4Gbyte limit.

(reinoud)

makefs_udf uses clock_gettime, and on some versions of linux that lives in librt

(jmcneill)

bump command timeout to 5 minutes. several
types of changers (Overland PowerLoader, Dell
PowerVault) have been exceeding the 100 sec
limit aborting a perfectly (slowly) progressing
operation.

(kardel)

fix issues when reading variable block sized tapes.

symptoms:
generic HBA error on console when reading
with a larger blocksize. blocks read
are padded to requested block size with
a 5a... a5... pattern.

problems fixed:
- controller scsi_status values did not match
  the ones used by the spsipi layer.
  a mapping function was introduced.
- when experiencing an underrun (read 64k and
  get a 63k block) the controller posted
  not a SUCCESS status but CHECK status. handle
  that like SUCCESS adjusting xs->resid and set
  XS_SENSE.
  now the correct data amount is returned and
  nothing is 'added' and no 'generic HBA error'
  occurs.
- make decisions using variables and constants
  from the controller domain.

(kardel)

Make the xsh man pages to unbreak the build (sets lists were updated to
expect these man pages, but they were not being built)

(pgoyette)

Fix possible 32 bit issues. In practice hard to reach but better be safe than
sorry.

(reinoud)

Fix style.

(rkujawa)

Add xsh(4) man page to dist list.

(rkujawa)

Add man page for xsh(4).

(rkujawa)

Oops, used size_t when off_t was meant. This results in the 4G file size
modulation seen on 32 bit machines due to size_t being 32 bit there.

While here, also fix compilation errors he@ discovered while compiling on
NetBSD/vax.

(reinoud)

Fix debug kernel builds when the automatic symbol table resizing kicks in.
Previously, the TAIL/HEAD link steps were being executed twice and the
(relinked/stripped) kernel would overwrite the debug version.

XXX this is all ugly.

(skrll)

Use cpu_idcache_wbinv_range so we don't nuke the entire idcache.
Add a test to make sure we are only modifying movw/movt

(matt)

Decode movw/movt

(matt)

Fix addend manipulation

(matt)

Add a INSTALL kernel for ARMADAXP

(matt)

Ignore REL/RELA if they aren't against a PROGBIT section.
(this causes EXIDX sections to be ignored for ARM).

(matt)

ldscript doesn't need -Ttext

(matt)

Now that all of the macros are enclosed in "do { ... } while 0" we can
remove unneded parenthesization of their invocations.

OK mrg@

(pgoyette)

926

(snj)

Pull up following revision(s) (requested by msaitoh in ticket #926):
sys/netinet6/nd6_rtr.c: revision 1.86
PR/47576: Takahiro HAYASHI: Avoid crash destroying tap0 after deleting
it's link-local address.

(snj)

926

(snj)

Pull up following revision(s) (requested by msaitoh in ticket #926):
sys/netinet6/nd6_rtr.c: revision 1.86
PR/47576: Takahiro HAYASHI: Avoid crash destroying tap0 after deleting
it's link-local address.

(snj)

926

(snj)

Pull up following revision(s) (requested by msaitoh in ticket #926):
sys/netinet6/nd6_rtr.c: revision 1.86
PR/47576: Takahiro HAYASHI: Avoid crash destroying tap0 after deleting
it's link-local address.

(snj)

Note the new xsh(4) driver.

(rkujawa)

Add preliminary support for Individual Computers new X-Surf 100 network card.

(rkujawa)

resort entry about ipsec kernel option change so it's chronologically
fitting

(spz)

add hints about FAST_IPSEC no longer being around

(spz)

Fix reverted condition. Fix panic reported by Patrick Welche on
current-users with an ATAPI device connected to siisata.

(bouyer)

Get the correct inttypes.h header, so this can build on !NetBSD.

(agc)

Bump RAMDISK size so that the contents fits again.

(he)

Allow 8 luns instead of 1. This enables access to the changer device on
a Dell PV-124T:
ch0 at scsibus0 target 9 lun 1: <DELL, PV-124T, 0086> changer removable
ch0: 16 slots, 1 drive, 1 picker, 0 portals

(kardel)

Build kobj_machdep.c

(matt)

Don't sync idcache/tlb when _RUMPKERNEL is defined.

(matt)

Move the shared libraries for librumpnet_npf to base from comp.

(matt)

Arghhh!  Typo.

(pgoyette)

Add a new test case for snprintb_m(3) to catch previous boundary bug.

(pgoyette)

apply some do { ... } while (0) so that macros aren't broken,
completing the previous change.

(mrg)

Remove the note about ticket #652 (bge) from CHANGES-6.1 because the change
wasn't added.  Fixes PR#48069 for netbsd-6-1 branch.

(msaitoh)

The PUT* things are macros.  Use { ... } to make sure they expand correctly.

(pgoyette)

Update example using hex-escape-sequences to conform to gcc parsing rules.
(See http://gcc.gnu.org/bugzilla/show_bug.cgi?id=33167 for details.)

(pgoyette)

Add an example using snprintb_m()

Replace \*[Gt] and \*[Lt] with the simple characters > and < (OK wiz)

XXX Note that the examples currently do not compile with GCC!  The hex
XXX character sequences such as \x10CACHE are being parsed as longer
XXX than 2-hex-digit strings!

(pgoyette)

Fix boundary condition in multi-line format.  When the '=' following a
field name exactly filled a line, the back-up logic failed and really
messed up the formatting!

(pgoyette)

Switch alpha to USE_COMPILERCRTSTUFF= no

(skrll)

fix another typo

(matt)

Allow i2c addr wildcard matching. Use with care!

(soren)

Add R_ARM_V4BX, R_ARM_MOVW_ABS_NC, R_ARM_MOVT_ABS
Fix R_ARM_PC24 and match R_ARM_CALL and R_ARM_JUMP24

(matt)

Add more relocs

(matt)

Revert previous change to tools/Makefile.gnuhost.  When getenv("M4")
contains multiple words, flex fails.

Adjust tools/autoconf/Makefile to pass the necessary extra arguments via
M4=... in CONFIGURE_ENV.  This is similar to what used to be in
revision 1.4 of tools/autoconf/Makefile, but it's now after instead of
before the inclusion of Makefile.gnuhost, because it has to override the
value of M4=...  that Makefile.gnuhost stores in CONFIGURE_ENV.

(apb)

Add support for generating tramponlines in the module at link time.
This is done using some awk scripts and KMODTRAMPOLINE in <machine/asm.h>
It's not as efficient as having the kloader do it but it is a lot simpler.

(matt)

Keep track of undefined symbols and then use to determine whether a wrap
is required.

(matt)

Add KMODTRAMPOLINE(name)
There are 3 variations:
one for armv7 using movw ip,#:lower16:func; movt ip,#:upper16:func; bx ip
one for armv4t+ using ldr ip, pc; bx ip; .word func
one for everything else using ldr pc, [pc, #-4]; .word func

(matt)

awk support scripts to generate trampolines in modules.
(adapted from previous LKM method).
<machine/asm.h> now has a KMODTRAMPOLINE macro to generate the trampoline

(matt)

Fix stale comment

(skrll)

Make sure a channel has drives attached before resetting it otherwise
wdc_reset will panic the kernel due to a null derefrence of ch_drives.

(blymn)

reserve space for ICMP header accordingly to RFC4884

(kefren)

Fix a typo

(matt)

Add some more (cryptography-related) acronyms

(agc)

Build orphaned man pages.

(soren)

Compile snprintb.c into libnbcompat.  The source file is found in the
common/lib/libutil directory.

(apb)

If HAVE_SNPRINTB_M is defined, then do not define the snprintb() and
snprintb_m() functions.  This should allow this file to be compiled
from a tools build regardless of whether or not the host defines these
functions.

(apb)

declare snptintb_m() and snprintb() if HAVE_SNPRINTB is not defined.

(apb)

regen, to pick up the test for snprintb_m.

(apb)

Add a configure test for snprintb_m.  We will assume that
the rpesence of snprintb_m implies the presence of snprintb.

(apb)

Add the "-g -D__gnu__" flags when invoking M4 via CONFIGURE_ARGS
in tools/Makefile.gnuhost.  This makes the tools version of m4
behave like GNU m4.

Remove similar code from tools/autoconf/Makefile, where it didn't
work because Makefile.gnuhost would previously have overridden it.

This fixes a problem in which the tools version of autoconf complained
"M4sugar requires GNU M4.
Install it before installing M4sugar or set the M4 environment variable to its path name."

(apb)

db_stack_trace_print: support trace/[at], starting trace from switchframe.

(uwe)

Only use MACHINE_ARCH if !_STANDALONE and CPPFLAGS/CFLAGS/CPUFLAGS don't
contain -mcpu or -march

(matt)

Select ldrex/strex if ${MACHINE_ARCH:Mearmv[67]*}

(matt)