Include sys/ioccom.h directly because it's used.

(uebayasi)

recognize isysroot

(christos)

... and another.

(uebayasi)

Fix build caused by a last minute change.

(uebayasi)

Be a little more friendly to dynamic physical segment registration.

Maintain an array of pointer to struct vm_physseg, instead of struct
array.  So that VM subsystem can take its pointer safely.  Pointer
to this struct will replace raw paddr_t usage in the future.

Dynamic removal is not supported yet.

Only MD data structure changes, no kernel bump needed.

Tested on i386, amd64, powerpc/ibm40x, arm11.

(uebayasi)

Boot vs AP processors don't make sense for physical CPUs, these are
handled by the hypervisor and all CPUs are running when the dom0 is started.
In addition, we don't have a reliable way to determine the boot CPU as
- we may not be running on the boot CPU
- we don't have access to the lapic id
So simplify by ignoring the information and assign phycpu_info_primary to the
first attached CPU.

(bouyer)

rlphy fails to work without autonegotiation.

Reason is that ifm_data does not store BMCR data but a media index
that gets poked into the BMCR register.

Setting the BMCR and ANAR registers is correctly handled by the
generic function mii_phy_setmedia.

(mlelstv)

Explain why we hardwire lapic_cpu_number() to 0 on Xen.

(bouyer)

Move struct vm_page_md definition from vmparam.h to pmap.h, because
it's used only by pmap.  vmparam.h has definitions for wider
audience.

All GENERIC kernels build tested, except ia64.

powerpc/include/booke/vmparam.h has one too, but it has no pmap.h,
so it's left as is.

(uebayasi)

Add myself.

(kiyohara)

This actually uses only PAGE_SIZE, remove uvm/uvm_extern.h again.

(uebayasi)

According to nxr.netbsd.org nothing uses CTL_NAMES now.

(uebayasi)

Avoid pointless use of a sysctl definition, CTL_NAMES.  Clean up whitespace.

(uebayasi)

Avoid pointless use of a sysctl definition, CTL_NAMES.  Clean up whitespace.

(uebayasi)

Oops.  Fix thinko.

(uebayasi)

Platforms that dynamically set PAGE_{SIZE,MASK,SHIFT}, those values are
saved in struct uvmexp.  Expose only the relevant part for symbol users,
so that they don't need to include the whole uvm(9) API.

(uebayasi)

Yet another hidden uvmexp user.

(uebayasi)

Direct/indirect uses of sys/device.h and prop/proplib.h.

(uebayasi)

Include uvm/uvm_extern.h to access uvmexp.  gtmpsc.c doesn't need
uvm(9) API at all.

(uebayasi)

This uses autoconf(9) API, include sys/device.h.

(uebayasi)

prop/proplib.h is indirectly used here, include it.

XXX What is the conclusion about direct vs. indirect #include from headers?

(uebayasi)

Directly include sys/pmf.h because this uses it.

(uebayasi)

Include sys/pmf.h for pmf(9) API.

(uebayasi)

Don't rely on global APIs from internal.

(uebayasi)

Document RTM_CHGADDR and the newer RTM_IFINFO messages.

(roy)

Add a test for getenv(3) which fails at the moment because it doesn't
handle "=" in variable names properly.

I'll send out a diff for review soon which fixes this.

(tron)

quell some lint warnings

(plunky)

forgot to commit this one for two months!
add vdprintf.

(christos)

add want_agent to the commented line of pam_ssh

(christos)

Improve testing of environment handling:
1.) Always check the return value of setenv(3) and unsetenv(3).
2.) Test that calling setenv(..., ..., 0) doesn't overwrite environment
    variables.
3.) Add a new test which mixes putenv(3) and setenv(3).

(tron)

Some missing sys/device.h dependencies.

(uebayasi)

Add pseudo-device dmoverio(4) to ALL

(pgoyette)

dmover(4) really uses uvm(9).  (Wasn't revealed by i386/ALL...)

(uebayasi)

sys/device.h is not suitable for such low-level definitions.

(uebayasi)

Don't pull in the whole uvm(9) API to access only PAGE_SIZE and
some other constants.  These are provided by sys/param.h now.

(uebayasi)

Don't pull in huge sys/device.h definitions.

(uebayasi)

Include sys/device_if.h instead of sys/device.h.  machine/cpu.h is
one of the lower-most definitions.  Relying on the huge sys/device.h
is not a good idea.

(uebayasi)

UVM constants should not rely on sys/lock.h.

(uebayasi)

Include sys/evcnt.h for struct evcnt.

(uebayasi)

Note new et(4).

(jnemeth)

Add new et(4)/etphy(4) devices.

(jnemeth)

Don't redefine famous round_page() / trunc_page() locally.

(uebayasi)

Include sys/proc.h for tsleep, wakeup.

(uebayasi)

Include sys/proc.h for tsleep, wakeup.

(uebayasi)

Include sys/proc.h for curproc.

(uebayasi)

Include uvm/uvm_param.h from here to provide PAGE_SIZE and other
VM related constants.  Many users of uvm/uvm_extern.h in subsystems
are interested in only PAGE_SIZE.  The whole uvm/uvm_extern.h is
not necessary at all.

(uebayasi)

i.MX51 and NetWalker

(bsh)

Add some verbose output.

From OpenBSD.

(skrll)

Sharp NetWalker PC-Z1 support by SHIMIZU Ryo, Hashimoto Kenichi, and me.
http://www.sharp.co.jp/netwalker/

(bsh)

i.MX51 support by SHIMIZU Ryo, Hashimoto Kenichi, and me.

(bsh)

driver to use EPIT as system clock for i.MX31 and i.MX51,
by Hashimoto Kenichi.

(bsh)

extract GPIO register definitions from imx31reg.h to share them with
i.MX51

(bsh)

make this compile again.  Need to revisit for UART support.

(bsh)

UART driver for i.MX31 and 51.

(bsh)

UART driver for i.MX31 and 51.

(bsh)

back out previous. That was an mistake.

(bsh)

Hide uvm/uvm_page.h again to ensure its internal structures are MD.

GENERIC or at least one kernel compile tested for:
acorn26, acorn32, algor, all, alpha, amd64, amiga, amigappc,
arc, bebox, bighill, cats, cobalt, dreamcast, ews4800mips,
hp300, hp700, hpcarm, hpcmips, hpcsh, i386, ibmnws,
integrator, ixm1200, iyonix, landisk, luna68k, mac68k,
macppc, mipsco, mmeye, mvme68k, mvmeppc, netwinder, news68k,
newsmips, next68k, obs266a, ofppc, pmax, pmppc, prep,
rs6000, sandpoint, sbmips, shark, sidebeach, sparc, sparc64,
sun2, sun3, usermode, vax, x68k, zaurus

(uebayasi)

Add INSTALL.{html,more,ps,txt} for bebox.

(kiyohara)

rename imx31_bs_tag as imx_bs_tag to share it with imx51.

(bsh)

rename imx31_space.c as imx_space.c

(bsh)

rename arm/imx/files.imx as files.imx31

(bsh)

rename imx31_space.c as imx_space.c

(bsh)

rename imx31_space.c as imx_space.c

(bsh)

rename files.imx as files.imx31

(bsh)

Typon.

(uebayasi)

Bump symtab space (350000 -> 360000).

(uebayasi)

Pull in uvm/uvm.h for vm_nphysseg.

(uebayasi)

Implement CAS using a RAS for non-MP VAX systems.  This technique uses the
atomicity of indirect addressing for correctness.  It also uses the knowledge
of dereferencing an address outside the page table length will cause a PTELEN
trap.  [Tested on netbsd-5]

(matt)

Add sample xorg.conf files for hpcarm, hpcmips, and hpcsh.

(tsutsui)

PR/39094 - Kaspar Brand -- Add et (Agere ET1310/ET1301) network driver

The et(4) driver supports PCI Express Ethernet adapters based on
the Agere/LSI ET1310/ET1301 integrated MAC/PHY.

The et(4) driver was written by Sepherosa Ziehau for DragonFlyBSD,
ported to OpenBSD by Jonathan Gray and subsequently ported to NetBSD
by Kaspar Brand.

(jnemeth)

Oops - don't expose unnecessary data.

(uebayasi)

Fix debug code.

(uebayasi)

+tp

(pooka)

Allow clients to reuse a "park".

Patch from <yamt>, fixes PR kern/44086 by him.

(pooka)

Remove accidentally committed part (which I need to run these tests
for a -current rump kernel on NetBSD 5.1)

(pooka)

Add test case for problem described by yamt in PR kern/44086.
"large writes with PUFFS_KFLAG_NOCACHE_PAGE ends up with EIO"

(pooka)

Fix last 2 commits.

(skrll)

Pull in uvm/uvm.h for uvm_pageismanaged().

(uebayasi)

Provide port-specific sample xorg.conf files in xserver.tgz.

No particular comments on tech-x11.

(tsutsui)

Note about X11FLAVOUR=Xorg on dreamcast.

(tsutsui)

kloader(4) accesses vm_page, pull in uvm/uvm.h.

(uebayasi)

Support the new RTM_IFINFO and RTM_CHGADDR messages.

(roy)

Add RTM_CHGADDR to signal that an address on the interface has changed.
This is mainly used for notifying userland about active link address changes.

(roy)

Pull in uvm/uvm.h for vm_page_zero_enable.

(uebayasi)

Switch NetBSD/dreamcast to X11FLAVOUR=Xorg.

No particular comments on port-dreamcast@ and
tested with JP (HKT-7600) and US (HKT-7620) keyboards.

(tsutsui)

Pull in uvm/uvm.h for VM_PAGE_TO_PHYS().

(uebayasi)

Pull in uvm/uvm.h for VM_PAGE_TO_PHYS().

(uebayasi)

Pull in uvm/uvm.h where UVM's page level interface is used.

(uebayasi)

Include uvm/uvm.h for UVM internal definitions.

(uebayasi)

Get the dma registers from the soundbus property. This should work with all
snapper-compatible hardware.
Tested on PowerBook G4, MacMini and iBook G4. Failed on the PowerBook before.

(phx)

Put back uvm_page.h for now.  Sorry for mess.

(uebayasi)

Include uvm.h not uvm_extern.h following recent changes.

(njoly)

isakmp_post_acquire is now called from admin commands too, add a flag so
admin commands can be used to establish even passive links on demand.

(tteras)

Purge all IPsec-SA's if the last main ISAKMP-SA for the node is deleted
by remote request and the phase1 rekeying is enabled (this will also
trigger the new phase1_dead script hook).

(tteras)

Improve DPD sequence checks to allow any reply within valid sequence window
to be proof of livelyness. This can improves things if there's random
packet delays, or if racoon is not getting enough CPU time.

(tteras)

Extern admin protocol to allow reply packets to exceed 64kb. E.g SA dumps
with many established SAs can be easily over the limit.

(tteras)

Move MD member in struct vm_physseg to the tail, in case this struct
can be shared among architectures with only difference of the MD
part.

(uebayasi)

Put VM_PAGE_TO_MD() definition in one place.  No functional changes.

(uebayasi)

Add some more machines that need help with autoconf.

Slight variation on patch from Sergey Svishchev

(skrll)

Abstraction fix; move physical address -> per-page metadata (struct
vm_page *) "reverse" lookup code from uvm_page.h to uvm_page.c, to
help migration to not do that.

Likewise move per-page metadata (struct vm_page *) -> physical
address "forward" conversion code into *.c too.  This is called
only low-layer VM and MD code.

(uebayasi)

Fix build.

(uebayasi)

Abstraction fix; move physical address -> physical segment "reverse"
lookup code from uvm_page.h to uvm_page.c.

This code is used by some pmaps to lookup per-page state (PV) from
per-segment metadata (struct vm_physseg).  This is not needed if
UVM looks up physical segment once in fault handler, then directly
passes it to pmap.  This change helps transition to that model.

The only users of vm_physseg_find() are pmap_motorola.c and
powerpc/ibm4xx/pmap.c.

Tested By: Compiling and running powerpc/ibm4xx/pmap.c
(evbppc/conf/OPENBLOCKS266)

(uebayasi)

Abstraction fix; don't pull in physical segment/page definitions
in UVM external API, uvm_extern.h.  Because most users care only
virtual memory.

Device drivers use bus_dma(9) to manage physical memory.  Device
drivers pull in bus_dma(9) API, bus_dma.h.  bus_dma(9) implementations
pull in UVM internal API, uvm.h.

Tested By: Compiling i386 ALL kernel

(uebayasi)

Build fix for xen domu + PCI, from Juho Salminen in PR 44083.

(dholland)

Add readme explaining which grue has ok'd the eating of some code
until recently located here.

(pooka)

Remove some programs which live as atf tests now.

(pooka)

add compile-conditional rumpclient support

(pooka)

build system worship for new tests.

(pooka)

fix

(pooka)

Add rudimentary cgd tests.  The tests use cgd to transform a
plaintext into into an encrypted image and back into plaintext by
doing rump I/O on /dev/cgd.  There is one test to check that giving
the same password for both encryption and decryption produces the
same plaintext and another to check that giving a different passwords
does not produce the same plaintext.

This could be fairly easily extended to test all feature of cgd
(hint hint).  For example, now cgd.conf is included in cvs, but
the only reason for that is that without further hacking cgdconfig
uses /dev/random quality random to generate the salt for a
pkcsetcetc_kdf2 cgconfig -g, and making an automated test block on
the entropy pool is just not good form.  Details are everything.

(pooka)

The libevent tests are fairly lengthy - each test case actually contains
about 20 or so mini-cases.  Increase the timeout for these tests to let
them run to completion.

(pgoyette)

Apparently swwdog reboot hasn't worked in several years since it
tried to cpu_reboot() from a callout.  Make it reboot from a workq
instead.

problem made manifest
by atf test
rumpfs unmount flush
caused issue to un-shush

tested: rump kernel (tests/dev/sysmon) and qemu

(pooka)

help me if you can i'm gcc
and i do appreciate work'round-eee'e
help me, get my head out of the ground
won't you please, please help me?

(pooka)

skip tests which use features which rumpfs does not support
(namely: vop_rename and a file system size limit)

(pooka)

Add rumpfs to list of file systems to be autotested.

(pooka)

add necessary libs for t_posix_fadvise

(pooka)

+VOP_RENAME

(pooka)

support read/write & ubc

(pooka)

s/managed device page/device page/

(uebayasi)

Use vm_physseg accessors.  Remove confusing comments.

(uebayasi)

support vop_pathconf

(pooka)

use atf interfaces for error reportage

(pooka)

support vfs_mount/unmount

(pooka)

C style; make a sentinel pointer have an exclusive value; no
functional changes.

(uebayasi)

Fix usb_port.h-removal fallout with the patch by Brett Slager in
port-arm/44081, "evbarm/SMDK2410 kernel fails to compile".

Use device_t instead of device_ptr_t.  Use device_t and cfdata_t
typedefs.  Insert missing aprint_naive().

(dyoung)

Typo in a comment.

(uebayasi)

Minor clean up.

(uebayasi)

Actually, add a full tc for operating on fd's backed by regular
files (as opposed to pipes and devices).

(pooka)

be friends with genfs

(pooka)

Add expected failure test for posix_fadvise() panic reported by yamt
on source-changes-d.

(pooka)

Minor clean up.

(uebayasi)

+posix_fadvise

(pooka)

rump posix_fadvise()

(pooka)

Support non-hostbacked regular files, at least just a little bit.

(pooka)

regen.

(cegger)

add AMD RD880, ATI Radeon HD5600, ATI Radeon HD4200 and Atheros AR9285

(cegger)

Calibrate the amount of time that a sleep() requires, and use that
interval instead of assuming that there are exactly 1000 real-time-clock
milliseconds per second!  On some ports when running under qemu, there
can be twice as many RTC milliseconds as expected.

This is part 2 of the changes required to make the libevent tests work
on port-amd64 under qemu.

(pgoyette)

Adjust timers a bit so that the http timer can get a chance to expire,
even when running under qemu on platforms with a clock-skew problem.

The original 3-second timer was intended to be "longer than the http
timeout" (which is 2 seconds), and the updated 5-second value still meets
this requirement.  The updated value also meets the requirement even when
the http timeout stretches to 4-seconds under qemu.

This is part 1 of getting the libevent tests working on port-amd64 with
qemu.

(pgoyette)

vclean: fix a bug which makes getcleanvnode always cause VOP_INACTIVE.

(yamt)

Change md(4) to:
- create md devices on first open and destroy on last close.
- add enough disk label support to make DIOCGDINFO and DIOCGPART work.
- add disk_busy()/disk_unbusy() instrumentation.

Ok: David Young <dyoung@netbsd.org>

(hannken)

NPF checkpoint:
- Add proper TCP state tracking as described in Guido van Rooij paper,
  plus handle TCP Window Scaling option.
- Completely rework npf_cache_t, reduce granularity, simplify code.
- Add npf_addr_t as an abstraction, amend session handling code, as well
  as NAT code et al, to use it.  Now design is prepared for IPv6 support.
- Handle IPv4 fragments i.e. perform packet reassembly.
- Add support for IPv4 ID randomization and minimum TTL enforcement.
- Add support for TCP MSS "clamping".
- Random bits for IPv6.  Various fixes and clean-up.

(rmind)

Bring the netpgpverify(1) manual page into line with current output, etc.

With thanks to Jeremy Reed for the fixes.

(agc)

make this compile on amd64: clean up a debug statement, pointed out by jak

(agc)

Changes to 3.99.15/20101110

+ add support for partial blocks, defined in rfc 4880, and used fairly
extensively by gnupg where the input size may not be known in advance
(e.g. for encrypted compressed data, as produced by default by gpg -e)

(agc)

Note nside(4).

(skrll)

Add nside(4) - a driver for the National Semiconductor PC87415 IDE
controller as found in many HP PA-RISC machines.

From OpenBSD.

Reviewed by Manuel Bouyer.

(skrll)

PR/44077: Dr. Wolfgang Stukenbrock: redundant xdr_passwd implementation should
be removed, and the one from -lrpcsvc should be used.

(christos)

Update pa-netbsd.h to match pa-linux.h where it should. That is, remove
local definitions of

ASM_PREFERRED_EH_DATA_FORMAT
ASM_MAYBE_OUTPUT_ENCODED_ADDR_RTX

and make it obvious we're not using local ASM_OUTPUT_INTERNAL_LABEL

This fixes the current build problems (and probably more)

(skrll)

Don't puff sigpipe if the connection has been severed.  May happen
e.g. when a client executes a blocking call such a poll() and
decides to exit before the result is ready.

(pooka)

Put correct init.d hooks into dhcpcd-run-hooks.

(roy)

Now that copyinstr() correctly returns error, trim the semaphore
name to fit into SEM_MAX_NAMELEN.

(pooka)

<space><tab>

I went and wrote a gizmo to automate this a while back, but I seem to
have *lost* the thing. grr.

(dholland)