elfcore_grok_netbsd_note: sync SuperH PT_GETREGS value with reality.
Recognize PT_GETFPREGS that we don't yet actually have, but for which
the number is now reserved.

(uwe)

Minor cosmetics to minimize diffs with vendor branch.

(uwe)

Since sh3 has got PT_STEP, nuke software single-stepping entirely.

(uwe)

shnbsd_software_single_step: fix logic botch.  Now that removal of
software single step breakpoints is handled by the caller we no longer
need to keep static next_pc around.

(uwe)

PR/45545 Yui NARUSE: pipe2's return value is wrong

(christos)

make this work

(christos)

struct reg has GBR since NetBSD 5, so bump up SHNBSD_SIZEOF_GREGS.
It would be nice to detect this dynamically, but gdb7 wouldn't work on
NetBSD 4 anyway.

(uwe)

regen

(christos)

get_netbsd_elfcore_note_type: sync SuperH PT_GETREGS value with
reality.  Recognize old PT___GETREGS40 (for struct __reg40 without
GBR).  Recognize PT_GETFPREGS that we don't yet actually have, but for
which the number is now reserved.

(uwe)

#if-0'ed defines for PT_GETFPREGS and PT_SETFPREGS.  Not used yet, but
reserve the numbers as they are "leaked" to readelf(1), that has to know
them as NT_NETBSDCORE_FIRSTMACH + x (see get_netbsd_elfcore_note_type).

(uwe)

We need to allocate an inferior with our fake pid so that kernel core dumps
work.

(christos)

enable kernel stack frames now that I have tested them.

(christos)

No need to keep on trying when this fails in the expected manner.  We
really don't need to have 32k failures.

OK jruoho@

(pgoyette)

Move all pipe2 tests to a separate file.

(christos)

add pipe2 tests

(christos)

Constify structure pointed to by getataparams() return value.

(jakllsch)

Remove write-only variable.

(jakllsch)

Wrap or shorten lines longer than 80 columns.

(jakllsch)

switch everything except vax to gcc 4.5.
switch m68k to -Os since -O2 produces much bigger code
with gcc 4.5 than it did with gcc 4.1.

(chs)

annotate for lint

(yamt)

fix a type in a printf message

(yamt)

replace a non us-ascii character in a comment

(yamt)

comment

(yamt)

redo ip_output.c rev.1.206 and 1.207 differently.  PR/43664.
ok'ed by martin@

(yamt)

fix a double unlock bug introduced by tcp_input.c rev.1.312.

(yamt)

tcp_drain: grab softnet_lock where appropriate

(yamt)

tcp_reass_unlock: assertion

(yamt)

remove an unnecessary cast

(yamt)

typo in a comment

(yamt)

hpet:
- fix an integer overflow bug introduced by hpet.c rev.1.11.
- a workaround for AMD SB700 based systems, inspired from linux.

(yamt)

dumpsys_seg: don't overwrite the previous mapping

(yamt)

dumpsys_seg: don't overwrite the previous mapping

(yamt)

- make lendpri/changepri similar.
- make common code a subroutine.

(yamt)

Add support for the WSDISPLAYIO_GET_EDID ioctl (requires passing a device_t
to ffb_attach).

(jdc)

obsolete now useless (and wrongly set to "219" :-) HAVE_BINUTILS.
switch the sense of the gdb and gcc tests to list the platforms not
switched to gcc 4.5 or gdb 7, which is now:

GCC 4.1:
.if ${MACHINE_ARCH} == "alpha" || \
    ${MACHINE_CPU}  == "arm" || \
    ${MACHINE_CPU}  == "m68k" || \
    ${MACHINE_CPU}  == "vax"

GDB 6:
.if ${MACHINE_ARCH} == "alpha" || \
    ${MACHINE_ARCH} == "hppa" || \
    ${MACHINE_CPU}  == "mips" || \
    ${MACHINE_CPU}  == "sh3" || \
    ${MACHINE_ARCH} == "vax"

let's go, people!  (-:

(mrg)

PR/45517: Henning Petersen: Add restrict keywords in the definition of glob
to match the declaration.

(christos)

Use boolean AND instead of bitwise one in _DIAGASSERT.
From Henning Petersen in PR 45518.

(wiz)

regenerate without sim support for now.  (the rest isn't setup.)

mips64el can't debug n32 at all and n64 can't "run" any app.

(mrg)

Added in8() and out8() to access a byte with reorder-protection.
Use it in all drivers instead of (volatile uint8_t *).

(phx)

Make sure to disable interrupts, as U-Boot may call us with interrupts
enabled, causing quite unpleasant effects (e.g. by decrementer interrupts).

When DEBUG is defined, an exception handler will be installed, which reports
about all exceptions while altboot is running, including register dump and
stack frame backtrace.

(phx)

remove some lint

(christos)

PR/45542: Henning Petersen: Misplaced parenthesis in rshd.c

(christos)

Improve wording. Remove trailing whitespace.

(wiz)

Add a comment that pn_sizemtx should be useless as VOP_GETATTR now
needs a shared lock at least.

(hannken)

Import of the virtio driver written by MINOURA Makoto <minoura@netbsd.org>
with minor changes to make it compile an run on -current.  This driver
speeds up disk and network access in virtual environments like KVM.

Enabled on i386 and amd64.  Tested with a CentOS 5.7 x86_64 host.

See http://ozlabs.org/~rusty/virtio-spec/virtio.pdf for the specification.

(hannken)

VOP_GETATTR() needs a shared lock at least.

(hannken)

regen kernel configs

(rkujawa)

Enable efa in GENERIC.

(rkujawa)

Improve probe procedure in efa(4). Update man apge to reflect this.

(rkujawa)

switch powerpc64 to GCC 4.5 and GDB 7.  untested, but it builds which
is a step forward from a week ago.

(mrg)

Update comment (mention added 32-bit methods).

(rkujawa)

Fix the confusion between nodeid and inode numbers

(manu)

- Fix the confusion between fileno (opaque FUSE reference) and inode
  numbers. fileno must be used when exchanging FUSE messages.
- Do not use kernel name cache anymore, as it caused modification from
  other machines to be invisible.
- Honour name and attribute cache directive from FUSE filesystem

(manu)

Back out previous; this should not be done in statclock but when the
allocations change or getrusage(). Plus the += seems wrong, we cannot
keep adding to the stats.

(christos)

makefile frontend to run gmp configure under nbmake-$MACHINE.

(mrg)

add mips64el, the last one

(christos)

As both of the "ns-wide.wide.ad.jp" and "light.imasy.or.jp" IPv6 resolvers
now refuse my queries, replace them with "google-public-dns-a.google.com"
and "google-public-dns-b.google.com".

(jakllsch)

use LIBISPRIVATE instead of home-grown.

(mrg)

expand the POWERPC_LINUX checks to also apply to POWERPC_NETBSD.

(mrg)

"regenerate" after fixing link-time issues.

(mrg)

a copy of the ppc one.

(christos)

add ppc64 and armeb

(christos)

psim on ppc64 needs libz

(christos)

add support for powerpc64-netbsd.
fix support for sh*-netbsd.

(mrg)

shnbsd_get_next_pc: declare displacement as LONGEST and use
regcache_cooked_read_signed() to fetch displacement for braf/bsrf.

(uwe)

Another binutils workaround

(christos)

- make noat-at noreorder-reorder sequences consistent.
- add nop hack to fix assembler internal error.

(christos)

Typo fix.

(jym)

Allow 32-bit transfers for drives which support it. Add missing bus_space methods needed for such transfers.

(rkujawa)

Add Broadcom BCM3302 CPU to the table.

(jakllsch)

Some comment improvements; spelling and grammar.

(jakllsch)

Use pool(9) for struct pv_entry allocations rather than
uvm_km_alloc(9)/uvm_km_free(9) and ancient homegrown pv_page_info structures.

Calling uvm_km_free(9) during pmap_remove(9) could cause rw_lock error
in uvm_unmap1() as noted in PR port-m68k/45519.

NetBSD/x68k (both real X68030 and XM6i emulator) still gets weird panic
(corrupted kernel stack pointer?) on some heavy load:
---
panic: MMU fault
Stopped in pid 363.1 (X68k) at  netbsd:cpu_Debugger+0x6: unlk a6
db> tr
cpu_Debugger(4012004,8,1cbb528,2a618e0,2a5b000) + 6
db>
---
but it also occurs without this change so there might be some more bugs
in m68k pmap...

(tsutsui)

Make the mips softloat special case the same as in the libgcc Makefile.

(christos)

cleanup some messy casts and type violations.
XXX: There is one piece of code I marked that does not make sense.
Someone should take a look at it.

(christos)

make all bitfields unsigned int

(christos)

Everything else uses PIC branches to __cerror except those two...

(christos)

Sync with other ports: in PIC mode make __cerror hidden; it is not supposed
to be used outside libc.

(christos)

PR/45539: Greg A. Woods: add support for getrusage(2) memory size statistics

(christos)

mention another ppc64 hack.

(christos)

mark all these modules with "compatmodules" as evbppc64 doesn't want them.

(mrg)

make sure to use powerpc64/elf.m4, not powerpc32/elf.m4 here!

(mrg)

Fix comment.

(jym)

Fix style (pointed out by phx).

(rkujawa)

regenerate binutils for powerpc64.

(mrg)

remove the duplicated -mcall-netbsd.
adjust the __PIC__ definition to match common.opt.

(mrg)

remove -I- from CPPFLAGS.

(mrg)

Don't run off the beginning of an array from Maurizio Lombardi.

(jnemeth)

mknative-gcc for powerpc64.

(mrg)

now that ppc64 sort of builds, i can generate at least gmp configury.
thanks christos!

(mrg)

Remove the #if 1 / #endif around some code that appears to be
responsible deleting the 'first' AF_INET address on the interface if the
target address has family == AF_UNSPEC.

(dyoung)

For these interfaces, the implementation of SIOCSIFDSTADDR is identical
to SIOCINITIFADDR, and SIOCSIFDSTADDR callers always fall back to
SIOCINITIFADDR, so just get rid of the SIOCSIFDSTADDR case.

(dyoung)

Only advance the time if we are skipping the gap.

(christos)

Don't kauth-orize SIOCSIFMTU in pppsioctl() and stf_ioctl(), ifioctl()
has already done that for us.

(dyoung)

Userland may not change the IFF_CANTCHANGE flags, however, the kernel
may, so make sure if_flags_set() takes care of them.  Fixes a regression
in ifpromisc().

(dyoung)

Don't kauth-orize SIOCDIFPHYADDR, SIOCSIFFLAGS, SIOCSIFMTU, or
SIOCSLIFPHYADDR, in gif_ioctl() or in gre_ioctl(), because those
operations are ordinarily kauth-orized already in ifioctl().

Kauth-orizing SIOCSIFFLAGS in gre_ioctl() caused a panic ("panic:
bpf_detachd: ifpromisc failed: 1") when tcpdump(8) was interrupted.
Somehow bpf(4) enables promiscuous mode using different credentials than
it uses to disable promiscuous mode, hence the ifpromisc failure.  This
may have something to do with privilege-separation in tcpdump(8).  I.e.,
an LWP with SIOCSIFFLAGS privilege opens /dev/bpf, but an LWP without
SIOCSIFFLAGS privilege closes it.

(dyoung)

kauth isn't used in here, so don't #include <sys/kauth.h>.

(dyoung)

rename data() -> parse_data() to appease the powerpc64 toolchain:
{standard input}: Assembler messages:
{standard input}:105: Error: symbol `.data' is already defined

(christos)

prevent underflow from giving us a false positive test in the first iteration.

(christos)

Update to version 1.0.2.2

(martin)

Note briefly what MSRs are. Requested by wiz@.

(jruoho)

Postfix 2.8.6 was imported.

(tron)

Resolve conflicts from last import.

(tron)

Import Postfix 2.8.6. Changes since version 2.8.5:
- The Postfix SMTP daemon sent "bare" newline characters instead of
  <CR><LF> when a header_checks REJECT pattern matched multi-line
  header. This bug was introduced with Postfix 1.1.
- The Postfix SMTP daemon sent "bare" newline characters instead of
  <CR><LF> when an smtpd_proxy_filter returned a multi-line
  response. This bug was introduced with Postfix 2.1.
- For compatibility with future EAI (email address
  internationalization) implementations, the Postfix MIME processor no
  longer enforces the strict_mime_encoding_domain check on unknown
  message subtypes such as message/global*. This check is disabled by
  default.
- The Postfix master daemon could report a panic error ("master_spawn:
  at process limit") after the process limit for some service was
  reduced with "postfix reload". This bug existed in all Postfix
  versions.

Status:

Vendor Tag: VENEMA
Release Tags: PFIX-2-8-6

(tron)

switch m68* to GDB 7.3.1.  tested on sun2 and sun3 in TME.  works
at least reasonably on sun3.  GCC 4.1 barely runs on sun2 and GCC
4.5 doesn't run...

(mrg)

Treat emips specially as we did with libgcc. Needs some of the softfloat
functions, even if it does not MKSOFTFLOAT.
XXX: still bad text relocations in eh_frame.

(christos)

shouldn't need .size for ppc64.

(christos)

Whitespace. New sentence, new line. Fix a typo.

(wiz)

Add info about FastATA 1200

(rkujawa)

Mention the new driver for FastATA 1200.

(rkujawa)

regen kernel configs

(rkujawa)

Add driver for ELBOX FastATA 1200 Mk-III/Mk-IV (and the man page).

(rkujawa)

Use PR_SLOWHZ, PR_FASTHZ defines in a couple spots.

(seanb)

regen powerpc64!

(christos)

Fix gif(4)/gre(4) operation over interfaces such as wm(4) that do IPv4
checksum-offload.  Note well: it really is necessary to clear the
csum_data.

While I'm here, remove the do-nothing case for SIOCSIFDSTADDR and let
ifioctl_common() or the protocol handle it.

(dyoung)

remove second _errno.h

(christos)

librt needs the ppc64 hack too.

(christos)

Add the same ppc64 hack that libposix needs.

(christos)

mention ppc64 hack.

(christos)

HACK: Include _errno.c here for powerpc64 build. Otherwise __errno does not
resolve. Why?

(christos)

There is no reason not to support CLOCK_MONOTONIC in {g,s}etitimer() since
the underlying implementation already supports it, so add it.

(christos)

document ITIMER_MONOTONIC.

(christos)

Change mktime*(3) so that it does not always return EOVERFLOW when it cannot
perform the conversion, but returns EINVAL when the time requested would fall
in the DST gap, or is not representable in the timezone requested, and document
this.

(christos)

Attempt to complete the WD_QUIRK_SPLIT_MOD15_WRITE list.
Per Seagate Publication number 100221381, Rev. B, the ST360015AS
is the only other drive in the Seagate Barracuda Serial ATA V
family that was not already listed.

Also, correct spelling of "globing".

(jakllsch)

- Postfix 2.8.6 is out.
- Add 4.2.3 to ISC DHCP.

(taca)

Sort SEE ALSO.

(wiz)

Revert the revision 1.5 in cpufreq(9). Instead, document that the KPI can
not be used before interrupts have been enabled. Suggested by macallan@.

(jruoho)

Bump CPUFREQ_LATENCY_MAX to UINT32_MAX.

(jruoho)

Obvious copy&pasto, PR port-vax/45525 from Henning Petersen

(martin)

avoid linker warnings with the new binutils by not forcing rump modules
to be soft-float.

(chs)

avoid linker warnings with the new binutils by not forcing ld.elf_so
to be soft-float.  the floating-point usage in ld.elf_so is only
because of stdarg functions like printf(), and gcc 4.5 is good
about not actually executing the FP instructions in such functions
unless FP values are passed to them, which ld.elf_so doesn't do.

(chs)

more device/softc splitting

(macallan)

Change the pattern for matching ttys to allow any number of digits,
(tty[0-9]*).  This should fix PR 45514.

(apb)

actually split device and softc

(macallan)

- Indent ifdefs, comment endif and else cpp tags
- Add missing END macro for LP64
- Make whitespace consistent

(christos)

revert previous.

(mrg)

catch up with other zs ports, original commit was:

>add two new functions for z8530tty: zs_chan_lock() and zs_chan_unlock(),
>and use them instead of various spl's in the zs.c's.
>
>reviewed by ad and martin.

(mrg)

add mips64eb

(christos)

Move getuid() out of the loop, uid should not change behind our back. KNF.

(jym)

add mc68000

(christos)

Document the ifioctl locking in comments.

Add a missing percpu_free(9) call.

(dyoung)

kill(2) might not be a good test after all, users can send signals
to setuid binaries (like SIGINT for ping(8)). So apply the exact same
rights as procfs, and test uid against proc p_uid (real owner of process).

Bail out if we are not root and we attempt to print mappings of a process
that the user does not really own.

(jym)

age_init() is called from age_watchdog() which is interrupt context, we
can't sleep here or we get a DIAGNOSTIC panic when age_watchdog() fires.
More correct bus_dma(9) usage in age_encap()
Introduce a age_shutdown() to be called by pmf(9) at shutdown time,
to stop the DMA engine. My system would't properly reboot without this.
Be consistent in WAIT/NOWAIT use in init routines
Use BUS_DMA_COHERENT where appropriate
Rework the interrupt routine a bit, and ACK but do not disable interrupts
here. There seems to be a race where interrupts would not be properly
reenabled after this, leading do watchdog timeouts.

With these changes, the on-board age(4) on ftp.fr.netbsd.org seems to
be finally working properly.

(bouyer)

Make this build without options MULTIPROCESSOR

(bouyer)

back to using timespec, but convert to microseconds to prevent overflow.

(christos)

Fix bogus KASSERT: if there is a xbdi_io, xbdi_pendingreqs must *NOT* be 0.
Not sure why it has stayed unoticed for so long ...

(bouyer)

Add all the missing HAS_ so that we get back login_cap functionality.

(christos)

enable Sony's Six Axis and DualShock 3 USB controllers

(aymeric)

regen

(aymeric)

add Sony PLAYSTATION(R)3 Controller entry

(aymeric)

Add mipseb,mipsel,sh3el

(christos)

Per existing x86 GENERIC kernels, add:
#options BUFQ_READPRIO
options BUFQ_PRIOCSCAN
to Xen kernels.

(jakllsch)

add radix_tree_empty_tagged_tree_p, a "tagged" variant of
radix_tree_empty_tree_p.

(yamt)

Use microsecond rather than nanosecond resolution when sampling latencies.

(jruoho)