Disable SSL V3 session renegotiation since the protocol parameters of the
old session are not cryptographically tied to the new session ones.
NB: Applications that require session re-negotiation will fail after this
update.

(christos)

fix unused var warning when _LP64 is defined and CHIP_EXTENT is undefined

(cliff)

Apply errata workarounds.

From OpenBSD.

(skrll)

Enumerate ACPI PCI devices. Allows to link PCI with ACPI devices.
Patch presented on tech-kern@
http://mail-index.netbsd.org/tech-kern/2009/11/28/msg006552.html

'nice work' Jukka Ruohonen

(cegger)

Add MKZFS to MKEXTRAVARS just to make sure.

(uebayasi)

Ticket 1178.

(snj)

Apply patch (requested by tron in ticket 1178):
Update BIND server and tools to version 9.5.2-P1.

(snj)

Ticket 1178.

(snj)

Apply patch (requested by tron in ticket 1178):
Update BIND server and tools to version 9.5.2-P1.

(snj)

Add "MKZFS" to either "_MKVARS.yes" or "_MKVARS.no" depending on the
default of this platform. This fixed the build for NetBSD/amd64
and NetBSD/i386.

Yes, this horrible. There should be a better way to handle these
settings in general.

(tron)

Make sure "error" gets initialised.

(tron)

one more path adjustment for rump_vfs_makedevnodes() change

(pooka)

* create /dev on the rump rootfs automatically
* make rump_vfs_makedevnodes() take a full basepath instead of hardcoding
  an assumption that the device node is created in /dev
  + the caller is responsible that the directory the nodes are being
    created in exists

(pooka)

Make compile with new ukfs_part interfaces (doesn't use them, though).

(pooka)

Have as a backup at least *some* usable WWN.

(mjacob)

Interim workaround for modern Xeons that don't have the simplistic view of
bus speed and therefore do not support MSR_FSB_FREQ (e.g. X3400). In the
long-term, ACPI should be used for this (c.f. FreeBSD).

(sborrill)

CPU model and CPU extended model cannot simply be summed; the extended model
differentiates different CPUs within a given model type (i.e. model 0xe with
extended model 0x1 is NOT the same as a model 0xf).
Modern Xeons do not support MSR_IA32_EXT_CONFIG, so use model and extended
model correctly to avoid it

(sborrill)

vrele, not vput for unlocked devvp

(pooka)

Adapt ukfs users to the new ukfs_part interface.

(pooka)

remove unnecessary LDADD

(pooka)

Improve the ukfs magicpath interface a bit:
* rename PART to DISKLABEL
* allow to take an OFFSET range to make it possible to access an image
  where there is no disklabel present
* make future extensions easier and less intrusive for the callers

(pooka)

Do bounds-checking before adding host offset.  Otherwise the host
offset would be counted in with "size" and incorrect operation
would ensue.

(pooka)

Put ${MKXORG} back to MKVARS.  It doesn't exist in bsd.own.mk but used here
to generate th set list keywork "xorg" in list_set_files().  Hopefully fix
some of reported build failures.

(uebayasi)

Remove last remnants of the long-ago-properly-fixed RUMP_LMUTEX_MAGIC hack.

(pooka)

Make sure node lengths match in addition to strncmp().
(I thought i fixed this already once?)

(pooka)

Call spec_init() to avoid device open lossage.

(pooka)

Exit with a sensible error message if makeflist output is empty.
This should be better than an incomprehensible error message from
mtree.

(apb)

Decide it's not worth the fuss to have rumpfs as a module and just
hardcode attach into rump_vfs_init().  Saves us from a lot of
pingpong init bouncing from one component to another to get the order
right.

(pooka)

Remove a superfluous base-extsrc-* entry.

(uebayasi)

Soft-fail xcall thread creation to make RUMP_THREADS=0 work again.

(pooka)

Make the pdc_pat_io_num and pdc_pat_pci_rt structures static so they're
definitely VA==PA. Being on a stack doesn't guarantee this.

(skrll)

Add udl(4)

(sborrill)

Fix subst so that the subcribers can restart their daemon.

(roy)

Ticket 1368

(sborrill)

Pull up the following revisions(s) (requested by plunky in ticket #1368):
sys/net/if_tap.c: revision 1.60

Fix a potential leak on tap device close; purging the send queue did not
actually release the dequeued mbufs.

(sborrill)

Ticket 1173

(sborrill)

Pull up the following revisions(s) (requested by plunky in ticket #1173):
sys/net/if_tap.c: revision 1.60

Fix a potential leak on tap device close; purging the send queue
did not actually release the dequeued mbufs.

(sborrill)

Ticket 1179

(sborrill)

Pull up the following revisions(s) (requested by tron in ticket #1179):
libexec/ld.elf_so/rtld.c: revision 1.127

If the environment is corrupt, die. From FreeBSD.

(sborrill)

Ticket 1168

(sborrill)

Pull up the following revisions(s) (requested by scw in ticket #1168):
sys/arch/arm/arm32/pmap.c: revision 1.203

Work-around a possible process exit corner case which can leave stale
data in the cache after a context-switch. Addresses kern/41058.

(sborrill)

s/DEBUG/NOT_DEBUG/ on the intr_establish() re-used message.
it triggers all the time on more-modern systems.

(mrg)

fix a segfault in xf86EnterServerState() which would fire whenever there is
more than one graphics device
This is a bandaid, we ( or the Xorg people ) need to fix it properly by
getting rid of all the now useless RAC-related code.

(macallan)

base this on GENERIC.MP.  seems just as stable.
turn on esiop/siop now that their interrupts match.
turn off admtemp, since we get a false match.
XXX admtemp_match needs help.

(mrg)

Oops.  Remove debug echo again.

(uebayasi)

Convert make variables ${_MKVARS.yes} / ${_MKVARS.no} to environments
${_MKVARS_yes} / ${_MKVARS_no} to make sure assignments done right both in
make and shell.

Extract important make variables not listed in ${_MKVARS.*}.

(uebayasi)

Ticket 1167

(sborrill)

Pull up the following revisions(s) (requested by nonaka in ticket #1167):
sys/dev/sdmmc/ld_sdmmc.c: revision 1.4
sys/dev/sdmmc/sdmmc_mem.c: revision 1.3
sys/dev/sdmmc/sdmmcvar.h: revision 1.2

Always treat sector size as 512 bytes.

(sborrill)

fix inverted comparison

(pooka)

Use common pmap_bootstrap_finalize() to initialize lwp0 uarea etc.
Tested on NWS-1750.

(tsutsui)

Use common pmap_bootstrap_finalize() to initialize lwp0 uarea etc.
Tested on TT030.

(tsutsui)

Patches from Lloyd Parkes in PR 41684, plus my changes:
* ${HOSTNAME} is special to bash, so rename the variable to HOSTNAME_CMD.
* Don't use "cd -".
* Don't use only comments in the else part of an if statement.

(apb)

Use common pmap_bootstrap_finalize() to initialize lwp0 uarea etc.
Tested on HP9000/382.

(tsutsui)

If the environment is corrupt, die. From FreeBSD.

(christos)

Add a common pmap_bootstrap_finalize() function which initializes
lwp0 uarea, curlwp, and curpcb after MMU is turned on, using
lwp0uarea variable saved during pmap_bootstrap() (or equivalent functions)
for pmap_motorola based ports.  Discussed with rmind@.

(tsutsui)

Removed obsolete files.

(martti)

Add EXTSRCS to the list of syspkgs aware variables.  Remove debug code.

(uebayasi)

Initial version.

(martti)

Pass the correct ${MKVARS} equivalent value to list_set_files().

XXX list_set_files() needs to be rewritten.

(uebayasi)

Quote variable values from make in case it has spaces.  Remove debug print.

(uebayasi)

Fix previous; don't forget extra vars other than MK*.

(uebayasi)

Don't hardcode the list of MK* variables but use ${_MKVARS.yes} and
${_MKVARS.no} provided by bsd.own.mk.

(uebayasi)

Define lists of MK* variable names as _MKVARS.yes and _MKVARS.no.  They're
referred to by syspkg.

(uebayasi)

Define DPRINTF and use it.

(skrll)

Whitespace.

(skrll)

Call pxa2x0_clkman_config() instead of ioreg_write().

(kiyohara)

Enable latest bluetooth module (Infineon PBA31308).

(kiyohara)

Maybe supported LAN9218 Family.

(kiyohara)

KNF

(stacktic)

Maybe supported LAN9218 Family.

(kiyohara)

tyop de fix

(pooka)

Document that strtouq can return UQUAD_MAX.

(pooka)

Added RCS id.

(stacktic)

Pass lint.

(roy)

In calculating initial tstate, cast to 64-bit first to avoid sign extension.

(nakayama)

Retain old variables when failed to allocate memory.

(enami)

Reinstate __getdelim which does no locking.
Callers are now required to FLOCKFILE so they can operate on fp as well.

(roy)

struct _locale_time_t _global_locale is a datum.  Declare it as extern.

(uebayasi)

Protect getline.

(roy)

Declare global pointers as extern, otherwise they become common symbols.
Fix mips build in lib/librumpnet.

(uebayasi)

print the fplwp in "mach cpu" list.

(mrg)

count IPI's as interrupts, not misc counters.

(mrg)

print pid, lid and p_comm in some diag messages that indicate killed processes.

(mrg)

Mention lvm2tools update.

(haad)

Mention lvm2tools update.

(haad)

Fix typo in ticket 1160's description.  Noted by rafal.

(snj)

Add two more needed files dev-swap.c and unknown.c.

(haad)

Fix build, add missing symbols, remove unknown ioctls etc.

(haad)

Fix comment about running ./configure script.

(haad)

Fix cinflicts created by import of latest lvm2tools LVM2-2-02-56.

(haad)

Add python bindings for netpgp, via swig.

When using python, always add the dumb symbolic link to the library name.

(agc)

Import of new version of lvm2tools LVM2-2-02-56.

Status:

Vendor Tag: REDHAT
Release Tags: LVM2-02-56

(haad)

Ticket 1177.

(snj)

Pull up following revision(s) (requested by mrg in ticket #1177):
external/mit/xorg/lib/pixman/Makefile: revision 1.10
disable using sse2 for i386.  from drochner@ and veego@.

(snj)

Apply patch (requested by mrg in ticket 1177):
Update pixman to 0.16.2.

(snj)

Apply patch (requested by mrg in ticket 1177):
Update pixman to 0.16.2.

(snj)

Spell busy correctly.

(skrll)

Fix build. Hi rmind.

(skrll)

Remove not needed debugging CFLAGS.

(haad)

Revert my commit which have added knowledge about dm targets to libdevmapper,
this breaks abstraction. Because only lvmtools/lvmlib and device-mapper can
have knowledge about target mapping and libdevmapper only passes requests
from lvmtools to kernel and back. Bump major library and driver version.

Requested by: yamt@

(haad)

Revert my commit which added knowledge about dm targets to libdevmapper,
this breaks abstraction. Because only lvmtools/lvmlib and device-mapper can
have knowledge about target mapping and libdevmapper only passes requests
from lvmtools to kernel and back. Bump major library and driver version.

Requested by: yamt@

(haad)

Typo in comment.

(skrll)

Re-instate perl taint checking by re-defining the possible taint check in
swig. Fix a bug whereby the generated shlib_version file got appended to,
rather than rewritten.

(agc)

Tickets 1158, 1160, 1161, 1166, 1174, and 1176.

(snj)

Pull up following revision(s) (requested by tron in ticket #1174):
usr.sbin/rpc.lockd/lockd_lock.c: revision 1.31
allow comparing v2 and v3 filehandles.

(snj)

Apply patch (requested by rafal in ticket 1176):
Regenerate xf86-video-ati's config.h.

(snj)

Pull up following revision(s) (requested by mhitch in ticket #1166):
sys/arch/amiga/amiga/machdep.c: revision 1.222
Remove old legacy IPL_* settings in ipl2spl_table - they are redundant
and in the case of IPL_AUDIO, results in IPL_VM blocking all interrupts.
[IPL_AUDIO is now the same as IPL_VM, and was setting the IPL_VM entry
to PSL_IPL6.  IPL_SERIAL had a duplicate entry for some reason, and
serial drivers that check and modify IPL_SERIAL now shouldn't be affected
since IPL_SERIAL is now the same as IPL_HIGH.  I'm hoping this will fix
or at least help the problem of serial interrupts being blocked too long.

(snj)

Pull up following revision(s) (requested by cube in ticket #1161):
sys/arch/i386/pnpbios/fdc_pnpbios.c: revision 1.15
Properly fix fdc@pnpbios as shown by Brian Buhrow in PR#42374.
I think it tells a lot about the relevancy of that attachment when it took
20 months for someone to experience the issue and even then, acpi had to be
disabled.

(snj)

Apply patch (requested by rafal in ticket 1160):
Regenerate xf86-video-radeonhd's config.h.

(snj)

Apply patch (requested by bouyer in ticket 1158):
On amd64, add a third free list distinct from the default free list, holding
RAM between 16Mb and 4Gb. This helps preventing bus_dma(9) memory
allocation failures for 32bit DMA on large-memory machines.

(snj)

Remove restriction of "mach prom" to cpu0 - it does not make sense now
that cpu0 is not always the bootstrap cpu, and in the majority of cases
the prom abort will just work fine on any cpu.

(martin)

Revert previous, which caused tools build to fail.

(uebayasi)

Add a missing '\n' from a printf.

(skrll)

There's no point in adding _PKGTOOLS_VER=$(${PKG_CREATE} -V)
to any syspkg +BUILD_INFO files, since we don't use the pkg_* tools
to create syspkgs.  Also remove the definition of PKG_CREATE
from sets.subr, since nothing uses it now.

(apb)

Add note saying that special must be a raw device.
PR bin/42391

(pooka)

call cache_cpu_init() for all cpus

(pooka)

Include cpu crosscall support (instead of stubbing it out).

(pooka)

Almost there for virtual CPU MP support:
* support bound kernel threads
* bind softint threads to specific virtual cpus
  + remove now-unnecessary locks from softint code

Now, if we only had MI CPU_INFO_FOREACH() .... (hi rmind ;)

(pooka)

Call VOP_CLOSE() in mount error branch with the device locked to
prevent crash when specfs attempts to unlock the vnode.

(pooka)

Revert part of previous - the removal of the "first cpu to attach is
bootstrap cpu" requirement already fixed everything - no need to use
cf_unit for a bogus test here.

(martin)

Pass ${TOOL_PKG_CREATE} to scripts for syspkg creation.

(uebayasi)

Add ${TOOL_PKG_CREATE}.

(uebayasi)

Use ${NETBSDSRCDIR} for deep tools/* directories.

(uebayasi)

Remove U_PCB as requested by rmind.

Same code before and after.

(skrll)

whitespace

(pooka)

Add pkg_install as a tool.

(uebayasi)

Remove vinum-removal sentence fragment, similarly to newfs(8).

(pooka)

Make HTML-ready, use standard section headers, fix Xr.

(wiz)

Turns out that swig and tainted don't play well together - perl has no way
of knowing whether the memory will be modified. For now, the gross hack is
to switch off tainting

(agc)

Import PF from OpenBSD 4.2

Status:

Vendor Tag: OPENBSD
Release Tags: v4_2

(martti)

Add language bindings for tcl and perl

(agc)

Use the right field for the prefix

(agc)

Initial version.

(martti)

Add a swig interface file, and a wrapper script, for calling swig for
various language bindings for netpgp.

(agc)

Skip parameters for multilib build for now.

(uebayasi)

Handle cases where *.asm is listed in LIB2FUNCS_EXTRA (gcc/config/rs6000).

(uebayasi)

Remove two different explorer@ acknowledgements.

(snj)

Recognise the hash algorithm in a case-insensitive manner.

(agc)

Top level reach-over directory for extsrc.

(uebayasi)

Assign copyright to TNF

(explorer)

Assign copyright to TNF

(explorer)

KNF: define struct pareg near the top of the file.

(dyoung)

Delete wdcactivate() prototype, it's gone away for good.

(dyoung)

Simplify the device activation routine.

(dyoung)

KNF: whitespace.

(dyoung)

Cosmetic: fix indentation, change some spaces to tabs.

(dyoung)

Protect getdelim when used internally.

(roy)

Remove restriction that cpu0 has to attach first.
Make sure cpu0 attaches to the bootstrap cpu only (equivalent of previous
change but not restricted to UP kernels).

(martin)

Remove __getdelim and just use getdelim.
fgetstr now works with strings up to SSIZE_MAX as a result, but may
reallocate buffers needlessly just like it used to when the buffer size
exceeds INT_MAX.
fgetstr converts errno EOVERFLOW to EINVAL on getdelim error.

(roy)

Note that callers should use feof(3) or ferror(3) to distinguish between
EOF or an error.

(roy)

Note that EOF returns -1 when no characters are read.
Add code example.

(roy)

Move relocation address to lower memory (0x1000) further away from the
BIOS 40:0 segment than previously but giving more room for heap that 0x8800.

(dsl)

Update 'oemname' to NetBSD60.

(dsl)

Use getprogname() instead of hardcoding "mount_nfs" to get the
right info in rump_nfs.

(pooka)

KNF

(skrll)

Mention udl(4) USB display device support.

(tsutsui)

Add udl(4) and wsdisplay(4) as its child.

(tsutsui)

Add a driver for DisplayLink DL-1x0/1x5 based USB LCDs and USB-VGA adaptors.
Ported by FUKAUMI Naoki from OpenBSD with many modification.
No particular comments (except from me) on tech-kern@.

There are still many TODO even in MI wscons(4) API to handle this device:

* No detach function for wsdisplay(9).
  Unpluging a device causes a panic. (should be trivial?)

* ioctl() for X server support is currently commented out. ("notyet")
  OpenBSD allows device depedent ioctl()s and they introduced
  UDLIO_DAMAGE ioctl for the damage extension ops of X servers for udl(4).
  Before blindly pulling such ioctl(), probably we should discuss
  how such specific operations should be handled in MI wscons(4) API.

* Screen text of wsemul tty could be mangled during large scroll ops.
  All tty output operations are invoked via ttstart() with the giant
  tty_lock mutex held, so we can't call cv_wait(9) to wait resources
  for data xfers via usbdi(9).h, then text output is silently discarded
  on resource shortage. To handle this without tty_lock reorganization,
  we have to change wsdisplay(9) APIs (especially wsdisplaystart()) to
  return a number of actually handled characters as OpenBSD does, but
  it may require whole API changes around child rasops(9) etc.

* No MI API definition to convert mmap(9) cookie to physical address.
  The conversion is required to create a cookie which will be passed to
  pmap_phys_address(9) in uvm/uvm_device.c:udv_fault(). Most other
  drivers use bus_dmamem_mmap(9) or bus_space_mmap(9), but udl(4) uses
  kmem_alloc(9)'ed memory for bitmap data.
  Furthermore, pmap(9) man page says about pmap_phys_address(9):

    "This function is provided to accommodate systems which have
      physical address spaces larger than can be directly addressed
      by the platform's paddr_t type. The existence of this function is
      highly dubious, and it is expected that this function will be
      removed from the pmap API in a future release of NetBSD."

  As the man page says we have already had split paddr_t and vaddr_t,
  so it's time to remove such old ugly cookie and change all mmap(4)
  functions (mostly in MD bus_dma(9) and bus_space(9) APIs) to return
  simple physical address in paddr_t?

* We need proper device names for wsdisplay1 (and more devices).
  Currently wsdisplay0 uses ttyE0 through ttyE253 (minor 0 to 253)
  for screens, ttyEstat (254) for status, and ttyEcfg (255) for config.
  The next wsdisplay1 will use 256 through 509 for screens, 510 for stat,
  and 511 for config but what names should we use for them? ttyFxxx?

* How to handle multiple sets of wskbd/wsdisplay on a single machine.
  rc.d/wscons doesn't provide method to specify wscons control devices.
  There is no proper interface to specify which keyboard should be connected
  to which wsdisplay, etc.

* And maybe more...

(tsutsui)

Note "extsrc" support.

(uebayasi)