Quote .Cd argument.

(njoly)

typo

(dholland)

no trailing comma on enum

(plunky)

add code to ensure that SO_SNDBUF is large enough on client connections,
as was previously done for server mode

(plunky)

Add missing ./usr/X11R7/lib/modules/libexa_g.a to fix MKDEBUGLIB=yes build.

(jmmv)

Improve the diagnostic messages for invalid/non-ELF files. Proposed a
while back; I forget the context. Has been sitting in one of my trees
awaiting a test run since, apparently, August 2011.

(dholland)

Pass -Wstrict-overflow

(dholland)

Pass -Wstrict-overflow; while here, use curses TRUE and FALSE only with
curses.

(dholland)

Pass -Wstrict-overflow... mostly.

(dholland)

Pass -Wstrict-overflow.

(dholland)

Pass -Wstrict-overflow, and while here, don't read from index -1 of
an array.

(dholland)

Factor out some common code; pass -Wstrict-overflow.

(dholland)

Pass -Wstrict-overflow.

(dholland)

Move scratch variables of main to main, instead of having them be
file-static.

(dholland)

Enable sljit in amd64 kernel and modules.

(alnsn)

Add strategic assertion, now passes -Wstrict-overflow.

(dholland)

Fix compilation.

(alnsn)

Remove ifdefs for V7 (as opposed to older) unix.

(dholland)

Pass the move info around instead of using it as a global.

(dholland)

Crack down on global variables with single-letter names.

(dholland)

PR/47068: unifdef(1) with cpp directive on last line without newline
Remove bogus test; if there is no newline, the state does not change.

(christos)

Add two flags to keyboard/mouse attachment:
  PCKBC_CANT_TRANSLATE for keyboards that cannot translate to XT scancodes
  PCKBC_NEED_AUXWRITE for mice that don't probe first time
These flags can be set by the port-specific attachments.

(jdc)

Add pckbc at ebus.

(jdc)

Adapt to the changed signature of pckbc_cnattach().

(jdc)

6.99.13 - namei NDAT() change and pckbc_cnattach()

(dholland)

Add the extra keys found on the Tadpole SPARCle to the keysym list.

(jdc)

Add two flags to keyboard/mouse attachment:
  PCKBC_CANT_TRANSLATE for keyboards that cannot translate to XT scancodes
  PCKBC_NEED_AUXWRITE for mice that don't probe first time
These flags can be set by the port-specific attachments.

Add the translation table and function to handle set 2 to set 1 keyboard
translation in software.

Based on OpenBSD sys/dev/ic/pckbc.c revisions 1.10, 1.16, 1.17, and
sys/dev/pckbc/pckbd.c revision 1.15, and 8042 scan code information at:

  http://www.computer-engineering.org/ps2keyboard/

Note, that this changes the signature of pckbc_cnattach(), so ride the
kernel version bump for namei.

(jdc)

regen

(dholland)

Replace hack implementation of NDAT() for "nameiat" with a proper one.
(This change requires a kernel bump.)

(dholland)

add KERN_USRSTACK (this is not dynamically defined for FreeBSD compatibility)

(christos)

Remove some more unneeded Pp macros.

(njoly)

Remove some unused includes

(skrll)

Remove a few unneeded Pp macros.

(njoly)

Change the mailbox handling so that the data value top 28 bits are used.

(skrll)

Unwrap short line. Other whitespace.

(skrll)

Fix variable name for message of printf.

(kiyohara)

struct device * -> device_t, use device_xname()  (from chs@)

(tsutsui)

Fix botch on device_t/softc split in debug printf macro. from chs@

(tsutsui)

struct device * -> device_t  (from chs@)

(tsutsui)

Normalize autoconf(9) function arg names (parent, self, and aux)
for clarification.  from chs@.

(tsutsui)

use device_xname()  (from chs@)

(tsutsui)

struct device * -> device_t, struct cfdata * -> cfdata_t
use device_xname()  (from chs@)

(tsutsui)

Add a cast to print dev_t. (broken since 64bit dev_t change?) from chs@
XXX: probably no one will use this...

(tsutsui)

Fix botch on device_t/softc split in #if 0'ed debug printf. from chs@

(tsutsui)

struct device * -> device_t, struct cfdata * -> cfdata_t  (from chs@)

(tsutsui)

Remove unused struct device menber from softc. from chs@

(tsutsui)

struct device * -> device_t, use device_xname()  (from chs@)

(tsutsui)

Fix botch on device_t/softc split inside #ifdef SNDEBUG part. from chs@

(tsutsui)

use device_xname()  (from chs@)

(tsutsui)

Ticket 607, tag to be moved.

(riz)

struct device * -> device_t, use device_xname()  (from chs@)

(tsutsui)

Pull up following revision(s) (requested by riz in ticket #607):
sys/arch/evbmips/conf/MALTA64: revision 1.6
sys/arch/algor/conf/P5064: revision 1.74
Bump SYMTAB_SPACE 475000->485000.
Bump SYMTAB_SPACE 480000->490000

(riz)

struct device * -> device_t, struct cfdata * -> cfdata_t
use device_xname()  (from chs@)

(tsutsui)

struct device * -> device_t, struct cfdata * -> cfdata_t  (from chs@)

(tsutsui)

Bump SYMTAB_SPACE 480000->490000

(riz)

Bump SYMTAB_SPACE 475000->485000.

(riz)

Add the next three /dev/pci nodes.
Marvell Orion already exposes two.

(jakllsch)

Add trailing NULL's to the varargs functions as required. (John Spencer)

(christos)

Add some defensive code to deal with "issues" of this interface.
Seems it can't do DMA updates of the rxsts for mbufs with addresses
>= 256MB.  It can't also seem to properly read from the descriptor rings
if they are < 256MB.  And I have no idea why either should matter.

(matt)

Welcome to 6.0!

(riz)

Welcome to NetBSD 6.0!

(riz)

Insert changes.mdoc, detailing the highlighted changes from 5->6.

(riz)

Update comment on vnode life-cycle a little.

(rmind)

Now we get working netboot loader, so enable options NFS_BOOT_DHCP by default.

(tsutsui)

Mention x68k new netboot.

(tsutsui)

Add new /usr/mdec/netboot.

(tsutsui)

First cut at x68k network bootloader via NE2000 on Neptune-X and Nereid.
Written by isaki@ back in 2003 based on x86's NE2000 netboot, and
misc debug and improvements by tsutsui@ and Y.Sugahara.

Tested on XM6i (emulating Nereid Ethernet) and X68030 + Neptune-X by me,
also on X68030 + Nereid and X68000 PRO-II + Xellent30 + Nereid by isaki@.

Note:
- X680x0 IPLROM doesn't support bootstrap from Neptune-X or Nereid,
  so this network bootloader needs to be loaded from floppy or harddisk.
- Currently netboot support is not merged into the secondary Multi-boot
  (/usr/mdec/boot) due to size restriction.
- Primary loaders assumes that the secondary bootloader name is /boot,
  so /usr/mdec/netboot needs to be installed as /boot into the target
  bootstrap device.

(tsutsui)

fix pre-5 stat system call structure which should have used the pre-64-time_t
timestamps!

(christos)

Remove OLD manunal to install correct manual.

(msaitoh)

Bump version to denote check_getchar() bug fix.

(tsutsui)

Fix a bug in check_getchar() (called from awaitkey_1sec()).
It didn't clear input key data when non ASCII keys (CTRL, OPT.1 etc)
were pressed so it couldn't detect further key inputs and
"any other key for boot menu" interrupt during boot didn't work
in that case.

The problem was tracked by Y.Sugahara,
and tested on X68030 and XM6i by me.

(tsutsui)

Add a way to force PCI to negotiate gen1 (setting flags 1 in the config file).

(matt)

Tweak SUBDIR build order and add comments about each dir.

(tsutsui)

fix previous, thanks asau@

(dholland)

Add support for various things:
- hddtemp sensor
- ambient light sensor on/off (via sysctl)
- display info
- docking station info

(cegger)

Explicitly cast to uint64_t instead of using PRIxPADDR because
sizeof(paddr_t) != sizeof(bus_addr_t) on some architecture.

See:
  http://mail-index.netbsd.org/source-changes-d/2012/10/09/msg005353.html

(msaitoh)

Don't use ++ to set flags to true. Now passes -Wstrict-overflow.

(dholland)

Pass -Wstrict-overflow.

(dholland)

- The MVGBE_RX_MAX_FRAME_LEN_ERROR bit is used to check whether a packet
  is fragmented or not, so define new MVGBE_RX_IP_FRAGMENT with the same
  value and use it.
- Remove the checking whether a packet length is lower than 72 octet.
  This check is not used in Linux and FreeBSD. Tested with me (for Kirkwood)
  and Kiyohara (for DiscoveryII).

(msaitoh)

Disentangle do_sys_rename.

Elide the fs-wide rename lock for single-directory renames.  This
required changing the order of lookups, so that we know what the
directories are before we lock the nodes.

Clean up error branches, explain why various nonsense happens and
what it does and doesn't do, and note some of what needs to change.

(riastradh)

PR/47058: Antti Kantee: If the ipv6 flow code modifies the mbuf, pass the
change up to the caller.

(christos)

Note recent BCM57762 support to bge(4).

(tsutsui)

A few natty acronyms.

(ginsbach)

make libhack's syslog.c produce exactly the same symbols as libc's syslog.c,
so that in the future we can keep them synced. Avoid strong_alias since it
does not play well with symbol renaming.

(christos)

take care of the signal safe syslog calls which are not exposed.

(christos)

Change "=r" to "=qQ" in a register constraint in an asm statement
for a register that is used with the "xchgb" instruction in the
definition of __cpu_simple_lock_try().  This fixes PR 45673, or at
least works around the gcc bug that might be behind PR 45673.

The output from "objdump -d" before and after this change is
identical, for the amd64 GENERIC kernel, the i386 GENERIC kernel,
and the i386 MONOLITHIC kernel.

(apb)

Pass up all multicast addresses, not just broadcast.  Among other
things, makes IPv6 work over this interface.

(pooka)

Complete the device_t/softc split I attempted previously.

From chuq.

(skrll)

Remove files not used since Sep 2006

(skrll)

Clean-ups in usage message

(cheusov)

No need to include files more than once.

(njoly)

Fix in OSVER initialization

(cheusov)

add more locking (when we read and write fields)

(christos)

Add compat functions for syslog.

(christos)

PR/47055: David Holland: Add hostname to syslog_data, version number, rename
fields consistently, and version library calls affected.

(christos)

compat hook.

(christos)

PR/47055: David Holland: Remove static hostname and store the hostname in
syslog_data. Initialize hostname before it is used. While here, add a version
to syslog_data.

(christos)

Second part of PR/46998. The following is taken from the PR with a slight
edit from me.

The TX buffer size is not a function of the SEEQ chip, it is an arbitrary
driver threshold, so I've renamed the define accordingly and added a new
one to specify how many buffers are used (in my port I allow multiple
packets to be in flight at once, so have used #ifndef, such that the
makefile can override).

Comment corrected, and make use of the above defines.

Missing delay(1) added, otherwise the 20,000 timeout loop is dependent
on the speed of your processor. Matches ea_stoptx logic now.

The FIFO empty check does nothing if the previous mode was 'read', but is
required before changing the BUFCODE (per 80C04 datasheet page 19,
note [2]). Then the mode is set to write, so a second FIFO empty check is
needed incase the previous mode was read.

Treat m0 as a pointer not an integer.

Remove double write of the NULL packet header. Either do
memset/ea_writebuf or two writes to SEEQ_BUFWIN, but not both.

The calculation of nextpacket (for hdr[]) assumes bufstart = 0, and puts
the packet header pointing in the wrong place when it isn't.

The setting of CFG2_OUTPUT is done in ea_init(), so doing it in ea_rxinit
is duplicated code.

(skrll)

Split softc/device_t
Use device_xname

From chuq

(skrll)

Rename eatxpacket to ea_txpacket for consistency.

First part of PR/46998

(skrll)

Use device_xname.

From chuq.

(skrll)

Fix #if 0'ed code.

From chuq

(skrll)

Update comments to reflect the real function prototypes.

From chuq.

(skrll)

Replace .sp with .Pp macro.

(njoly)

Remove empty BUGS section.

(njoly)

Adjust column width to match reality.

(njoly)

Note the replacement of kame_ipsec by fast_ipsec;  this change was
originally submitted as a patch to the netbsd-6 branch, but should have
been committed on the trunk first.

(riz)

Fix device_t/softc split bothces.  From Chuck Silvers' patch posted on
current-users and tech-kern:
http://mail-index.netbsd.org/current-users/2012/10/09/msg021233.html
and several more fixes by me.

Not sure if these are fatal (at least 6.0_RC2 works without thise fixes),
but needs more investigation.

(tsutsui)

TIcket 599.

(riz)

Pull up following revision(s) (requested by apb in ticket #599):
distrib/notes/common/main: revision 1.490
Refer to the list of mirrors at http://www.NetBSD.org/mirrors
in "Quick install notes for the impatient".  Previously it
said "Check the NetBSD website for details" without giving an URL.

(riz)

Split softc/device_t

(skrll)

Fix device_t/softc split botches that could be fatal.  From
Chuck Silvers' patch posted current-users and tech-kern:
http://mail-index.netbsd.org/current-users/2012/10/09/msg021233.html

Should be pulled up to netbsd-6.

(tsutsui)

Fix device_t/softc split botches that could be fatal.  From
Chuck Silvers' patch posted current-users and tech-kern:
http://mail-index.netbsd.org/current-users/2012/10/09/msg021233.html

Should be pulled up to netbsd-6.

(tsutsui)

Remove the (never used) struct device from ssio_softc.

From chuq.

(skrll)

The pagedaemon ca emit write as large as the underlying device's maxphys,
so emergva size needs to be MACHINE_MAXPHYS, MAXPHYS is not enough.

(bouyer)

We know vp is not NULL, no need to check

(bouyer)

Refer to the list of mirrors at http://www.NetBSD.org/mirrors
in "Quick install notes for the impatient".  Previously it
said "Check the NetBSD website for details" without giving an URL.

(apb)

Use device_xname.

From chuq

(skrll)

Implement the writewatchfile routines for a Linux hypervisor.  As a
result, the shmif network driver now works when hosted on Linux.

(pooka)

In layer_lookup(), clear *vpp before returning EROFS, as otherwise a
stale value can be returned and this causes a diagnostic panic in
namei.

In relookup(), clear *vpp before calling VOP_LOOKUP, as is done in
lookup_once(), as an additional precautionary measure.

(in theory both of these fixes are not required together)

Should fix PR 47040.

(dholland)

Split out modsi3 and umodsi3 from the divsi3 file. This is so that
we don't get re-defined symbols in libc from libgcc in static linking.
Example: cc -pthread -static main-calls-pthread-create.c

(christos)

add split files.

(christos)

add split files

(christos)

Tickets 593-595, 597-598.

(riz)

Pull up following revision(s) (requested by christos in ticket #598):
external/bsd/bind/dist/version: revision 1.4
external/bsd/bind/dist/CHANGES: revision 1.7
external/bsd/bind/dist/bin/named/query.c: revision 1.9
Apply P4:
3383.  [security]      A certain combination of records in the RBT could
                      cause named to hang while populating the additional
                      section of a response. [RT #31090]

(riz)

external/bsd/wpa/dist/src/eap_server/eap_server_tls_common.c patch

Fix overflow in hostapd.  PRE-SA-2012-07.
[christos, ticket #597]

(riz)

Pull up following revision(s) (requested by jnemeth in ticket #595):
distrib/notes/macppc/install: revision 1.39
Add some text to "Common Problems" about Open Firmware 3.x not
setting up the console and how to work around the problem.  This
is an initial stab at solving PR/35348 for NetBSD 6.0.  After 6.0
is released, a better solution can be done for 6.1.

(riz)

Pull up following revision(s) (requested by rmind in ticket #594):
sys/net/npf/npf_handler.c: revision 1.23
npf_packet_handler: drop the packet if IPv6 reassembly did not work.

(riz)

Pull up following revision(s) (requested by christos in ticket #593):
sys/kern/uipc_usrreq.c: revision 1.140
Avoid crash dereferencing a NULL fp in fd_affix() in unp_externalize
caused by the sequence of passing two fd's with two sendmsg()'s,
then doing a read() and a recvmsg(). The read() calls dom_dispose()
which discards both messages in the mbuf, and sets the fp's in the
array to NULL. Linux dequeues only one message per read() so the
second recvmsg() gets the fd from the second message.  This fix
just avoids the NULL pointer de-reference, making the second
recvmsg() to fail. It is dubious to pass fd's with stream sockets
and expect mixing read() and recvmsg() to work. Plus processing
one control message per read() changes the current semantics and
should be examined before applied. In addition there is a race between
dom_externalize() and dom_dispose(): what happens in a multi-threaded
network stack when one thread disposes where the other externalizes
the same array?
NB: Pullup to 6.

(riz)

Tickets 587-590.

(riz)

Pull up following revision(s) (requested by rmind in ticket #590):
usr.sbin/npf/npfctl/npf_parse.y: revision 1.14
npfctl/yyerror(): print the right line number if we already parsed the line.

(riz)

sys/arch/powerpc/powerpc/trap.c patch

Silence trap printfs caused by OpenSSL.
[christos, ticket #589]

(riz)

Pull up following revision(s) (requested by roy in ticket #588):
share/man/man5/ifconfig.if.5: revision 1.16
share/man/man5/rc.conf.5: revision 1.154
Discourage the use of ifconfig_bge0=dhcp and prefer setting dhcpcd=YES
in rc.conf(5) instead.

(riz)

Pull up following revision(s) (requested by christos in ticket #587):
etc/rc.d/named: revision 1.23
PR/47024: Taylor R Campbell: handle "keys" directory and directory
permissions in general

(riz)

Apply P4:

3383.  [security]      A certain combination of records in the RBT could
                      cause named to hang while populating the additional
                      section of a response. [RT #31090]

(christos)

add acceleration support, for Sun XVR-1200 and compatible only so far
adapted from OpenBSD with various improvements, no support for Expert3D has
been added
this is still suboptimal since we have to sync the engine all the time but
it's an improvement over software-only operation

(macallan)

Use mnt_maxphys not MAXPHYS to limit the size of I/O to disk.
Now the read-ahead code does issue 512k requests to disk.

(bouyer)

Redo previous: it seems that the point of the bytelen computation was to
get transfers aligned to chunksz. So reintroduce the code, but using chunksz
instead of chunksize (if the readahead is trucated there's no point in
trying to align it anyway).
Now I get 64k read requests at the drive level again.

(bouyer)

remove wrong ntohl (from Aran Clauson)

(christos)

Driver for pckbc at ebus, as found on Tadpole SPARCle.
Largely based on sparc's pckbc_js.c.

(jdc)

Fix panic "bad chunksize ..." in read-ahead code:
- off comes from the pager, so should already be page-aligned.
  KASSERT() that it is, and remove the off = trunc_page(off)
- as off is not changed any more, the size of the transfer is chunksize.
  Don't compute bytelen any more,  which is what required chunksize
  to be a power of 2. KASSERT() that chunksize is a multiple of page size.

(bouyer)

Support transfers of up to MACHINE_MAXPHYS in all pciide variants, and ahci.
wd(4) limits its maxphys depending on the drives's capability (64k sectors
for LBA48, 256 sectors for LBA and 128 sectors for older devices).

I assumed all pciide controllers could do MACHINE_MAXPHYS transfers, but
this may not be true. The capabilities of each controller variants should be
looked at more closely.

(bouyer)

Gather some statistics about biglock usage.

(pooka)

Use PRIxPADDR to print a DMA address. This fix a problem that if_wm.c
can't compile with WM_DEBUG on non-64bit platforms.

(msaitoh)

Add {INET,TCP,UDP}_CSUM_COUNTERS and LOCKDEBUG

(msaitoh)

regen

(dholland)

Add namei-level support for openat() and friends. The way you do it is
by calling NDAT(&nd, dirvp) after NDINIT().

Right now the implementation is vile and unspeakable to avoid changing
the kernel ABI; this way we can get openat() and friends into 6.1. I
will rectify the mess and bump the kernel once things are working.

(dholland)

Tidy up namei internals to allow openat() and friends without getting
tangled in nfsd's special cases.

(dholland)

Fix compilation when DIAGNOSTIC is defined.

(alnsn)

Initialize the workqueue to use IPL_NET for its mutex.
Prefer softints if the current lwp is the idle lwp.

(matt)

put all kern socket sysctls in the same place

(pooka)

Remove a tab character in non-literal context.
While here, replace all sp macros with Pp.

(njoly)

Kill some extra spaces in function arguments.

(njoly)

Function argument Fa macro expects its name on the same line.

(njoly)

call pthread_setname_np() on Linux too.

(pooka)

In columnated list, adjust address column size to match reality and
remove extra Ta macro in first It line.
While here kill a superfuous Pp.

(njoly)

Left and Right LEDs on in end of cpu_reboot().

(kiyohara)

It is Sprite, not Sprint; pointed out by Todd Miller.

(christos)

EXP-TLS server: Fix TLS Message Length validation:

    http://w1.fi/gitweb/gitweb.cgi?p=hostap.git;a=commitdiff;\
    h=586c446e0ff42ae00315b014924ec669023bd8de

http://www.pre-cert.de/advisories/PRE-SA-2012-07.txt

(christos)

Style fix: keep lists of names sorted.

Reviewed by: christos

(jkoshy)

mention new wpa code.

(christos)

merge v1.0

(christos)