Enhance system call prints by printing out the execve() arguments

(reinoud)

Fix typo

(reinoud)

Return ZERO to the child on fork!

(reinoud)

Print lwp values on syscall entry

(reinoud)

No need for such big stack spaces on lwp_fork() with the new setup.

(reinoud)

Remove white-spaces.
Indents.

(kiyohara)

Call bus_space_mallocok earlier.  Moreover, don't bus_space_unmap call.

(kiyohara)

Fix arguments for initppc().

(kiyohara)

Fix hangs-up.  Remove "wrteei 1" in board's cpu_configure().  Interrupt
is enabled in powerpc-layer.

(kiyohara)

Give the child more time to timeout, etc.

(skrll)

Revert last change to elf32-hppa.c and pull across the fix to my binutils
PR 13476.

tests/lib/libc/tls on hppa still passes with this change.

(skrll)

Using BUS_DMA_NOCACHE for bus_dmamem_map() causes issues on (at least) sparc64.

(jakllsch)

remove duplicated code.

(christos)

Use RT_ADVANCE when available.
Fixes PR/44838

(roy)

Document rtadvd changes

(roy)

When adding or scrubbing a prefix, always notify userland even if the
prefix does not have IFA_ROUTE.
Don't scrub the interface in SIOCAIFADDR if the new address does't
have IFA_ROUTE. If more functions are added to in_ifscrub then this logic
might need to be revisited.

Fixes PR/26450.

(roy)

u_int -> uint

(christos)

We usually refer to RCS keywords as "Ids", so make it obvious in the
synopsis that this script strips RCS Ids, and cross reference
ident(1) - makes it easier to find this utility with apropos(1).

(agc)

fixup comments - no functional change

(agc)

make ttycons a proper tty device, now it can be the console

(jmcneill)

add thunk_pollchar

(jmcneill)

Align in the stack before calling _rtld in the way the x86_64 ABI
specifies. This is necessary to ensure that init functions are always
called with the correct alignment, e.g. in case they use SSE
instructions.

(joerg)

Free the systemcall and pagefault stacks on lwp destroy.

XXX what about the main ucps stackspace that is used for the trampoline?

(reinoud)

- sprinkle const
- malloc + memset -> calloc
- free + malloc -> realloc
- syslog("%s", strerror(errno)) -> syslog("%m")

(christos)

make sure to set cn_dev and cn_pri in our consdev struct

(jmcneill)

Rework the debug printing to use functions instead of clutting the system call
code.

(reinoud)

Use __dead. No prototype for main.

(joerg)

btsco works now: http://mail-index.netbsd.org/current-users/2011/12/11/msg018583.html

(jmcneill)

PR# kern/45701: hdaudio stops working after suspend/resume and then detach/reattach

restore BIOS pin widget config on detach, and selected pin widget config on
resume

(jmcneill)

Use new -mstack-alignment option to replace the removed internal option.

(joerg)

Update to LLVM/Clang r146325. This fixes a number of issues for i386.

(joerg)

Test LD_PRELOAD for stack alignment as well. This exposes issues on
AMD64.

(joerg)

- WARNS=4
- object size is more smaller.

(nonaka)

Mark xassert as __dead

(joerg)

Require arm3_cache_flush for ARM2/ARM250.

(kiyohara)

Split boot.h and memcreg.h for MKSYMS and MODULAR.

(kiyohara)

Add attributes arp to eca.

(kiyohara)

Add RDNSS and DNSSL support, RFC6106.
Replace custom lists with TAILQ lists.
Clean up plently of signed vs unsigned warnings and set WARNS=4.

Adapted from FreeBSD.

(roy)

Give my slow, old hardware more time to complete bsize_ffactor.

(skrll)

no need to acquire thread lock in btsco_open / btsco_close, as they are
already held by MI audio when these methods are called

(jmcneill)

Move determination of socket buffer sizes from startup to the first
time a socket is used, as the previous logic assumed AF_INET sockets
were available (which they may not be in an IPv6-only system).
Per discussion with Maxim Konovalov and the FreeBSD problem 162661.

(lukem)

Fixes PR kern/20700 reported by Darcy Cain

Make 1000baset connections work even when the user specifies media 1000baset
connections with ifconfig(8) rather then only when media auto is specified.

For a further discussion of this fix, see:
http://mail-index.NetBSD.org/current-users/2011/12/07/msg018561.html

(buhrow)

Fix typo. Use more markup.

(wiz)

only use PG_G on leaf PTEs.
go back to tlbflush(), all the global entries
that we create in pmap_bootstrap() are permanent.

(chs)

Disabled copyin/copyout debugging messages since its swamping the output on
debug output.

(reinoud)

Restructure dev/clock to run from the signal stack since its safe to do now.

(reinoud)

Disable aio usage in thunkbus_ld since mixing ucontext_t and threads is a NONO
according to the manpage.

Why is AIO using a pthread?

(reinoud)

Enhance printing of system callss

(reinoud)

- add a CAVEATS
- fix a warning

(yamt)

- fix the section
- fix a warning

(yamt)

a utility to show in-core status of file pages

(yamt)

Move events/interrupt subsystem initialisation to before when ipis are used

(cherry)

Kill extra El macro.

(njoly)

re-apply merge loss

(cegger)

add AMD ucode MSRs

(cegger)

auvia reported to work

(mrg)

Remove empty lines in ttycons_consinit

(reinoud)

provide a lock for mpu_attach.

(mrg)

Use random seed file (saved entropy) from default location by default
in i386 and amd64 boot.cfg.

(tls)

drop the thread lock in a couple of error cases that were missing.

(mrg)

update a little:  zaudio failure, uaudio/umidi success

(mrg)

from the usbmp branch:
- drop the thread lock when usbd might sleep on us.

this also makes mixerctl -s vs play work without usbmp.

(mrg)

Use mutex(9) instead of spl(9). Add further locks around grant table access.

(cherry)

- make pipe->close method take the thread lock

- convert usb_taskq to use mutex/cv

- convert needs_explore usage into a cv on the thread lock

- remove KERNEL_*LOCK from uaudio and umidi, since we're supposedly
  MPSAFE here now

- use IPL_SCHED instead of IPL_USB (aka biglocked) interrupts

- drop the audio thread lock when calling into usb when it may sleep,
  avoiding a deadlock between audiowrite and audioioctl.  this fixes
  mixerctl -a vs. playing hanging the system
  XXX probably need to check this in a bunch more places.

(mrg)

- Explain the magic in npf_tcpfl2case().
- Use __unused instead of (void)cast; fix comment.

(rmind)

make ohci mostly work again.

(mrg)

pmap_bootstrap: use tlbflushg(), not tlbflush(), since we use global pages.

(rmind)

Uses uintptr_t and therefore needs stdint.h.

(dholland)

sync usb_subr.c and usbdivar.h with the branch entirely, and most of
usbdi.c as well.

(mrg)

Track if we saved l_private in the mcontext and don't blindly restore it.

This fixes the tests in /usr/tests/lib/libc/setjmp/t_threadjmp

(skrll)

pull across a few more of the changes from the main branch.

(mrg)

allow building without direct-map support in non-XEN kernels.

(chs)

build fix for keylock secmodel(9).

(jym)

merge a few more changes from the main branch.

(mrg)

merge a few more things from the main branch.  uaudio@ohci still works.

(mrg)

mostly in sync with the branch here now.

(mrg)

at little closer to the main branch.

(mrg)

partial sync with main branch

(mrg)

sync with main branch

(mrg)

sync a couple more changes from the main branch, uaudio@ohci still works.

(mrg)

slightly closer, and still working with uaudio.

(mrg)

there's something wrong with uaudio@ohci on the usbmp branch.
this is the minimally working port of ohci to mpusb, and it works
with uaudio but is lacking many other changes, yet.

(mrg)

kmem_free() the appropriate size.

Thanks cegger@

(cherry)

convert the remaining wakeup(xfer)'s to use the xfer->cv.

utoppy - untested.

(mrg)

- convert usbd_bus_methods{} and usbd_pipe_methods{} to use
  c99 struct initialisers

- move the locks from the pipe to the bus, since we'll need
  access to them from bus-level ops

- remove dead-for-years SPLUSBCHECK and replaced it with
  asserts that the thread lock is held

- begin to document the locking scheme

- convert usbd_*lock_pipe() into real function-like macros

(mrg)

Use a constant array for the MIB. Newer LLVM decided that mib[] warranted
stack protections, with the obvious crash after the setup was done.
As a positive side effect, code size shrinks a bit.

(joerg)

According to /etc/rc.d/mdnsd, /var/run/mdnsd is owned by _mdnsd:_mdnsd
if it created.  So, set owner/group of /var/run/mdnsd.

It stop a useless warning in /etc/security.

(taca)

ISC DHCP 4.2.3-P1 and 4.1-ESV-R4 has released.

(taca)

Bump SYMTAB_SPACE.

(he)

be more portable, explain what we are doing, simplify.

(christos)

kill some #ifdef USB_DEBUG with some compiler smarts.

(mrg)

Add the secmodel_extensions variants of the kernel modules.

(he)

Initialize the opl softc lock pointer to the parent's interrupt lock.
Briefly tested only on wss@isapnp.

(jakllsch)

We need to hold the lock during mpu_find().

(jakllsch)

replace malloc() with kmem_zalloc() per request from cherry.
While here fix error path.

XXX: The size allocated and freed for interrupt handlers looks
inconsistent to me.

(cegger)

remove xen3-public

(cegger)

switch from xen3-public to xen-public.

(cegger)

switch from xen3-public to xen-public.

(cegger)

merge. (forgot in previous)

(cegger)

merge.

(cegger)

Welcome to Xen 4.1.2 headers.

New interfaces for PV drivers:
- Xen transcedent memory
- USB IO
- SCSI IO

PCI IO improvements:
- PCI MSI support
- PCI Express AER support

New features:
- xen honors flags to be placed into guest kernel available pte bits
  if enabled (for grant table)
- support for 128 vcpus
  (old interface is still present and supports up to 32 vcpus)
- PCI passthrough: new hypercalls to support SR-IOV
- new hypercall for physical cpu hotplugging
- new hypercall for physical page offlining
- fixes to compile with clang
- machine check recovery mechanism

Status:

Vendor Tag: Xen
Release Tags: xen-4_1_2

(cegger)

build fix: add back <sys/malloc.h>. malloc(9) is still in use.

(cegger)

merge. The 'conflicts' happened because xen-public was once used for xen2 headers.

(cegger)

re-import xen3-public to rename this to xen-public

Status:

Vendor Tag: Xen
Release Tags: xen-3_3_0

(cegger)

Move to kmem_zalloc() instead of malloc().

(cherry)

consolidate bit twiddling functions into one generic case and a bunch of
macros, use them

(macallan)

add a bunch of register definitions

(macallan)

make this build without DIAGNOSTIC

(macallan)

Deal with __MACHINE_STACK_GROWS_UP machines.

constify the other psize while I'm here.

Thanks to Christos for looking.

(skrll)

Per [1] the speed of light in a vaccum is exactly 299792458 m/s.
Per [2] a furlong is 220 yards and a yard is exactly 0.9144 m.
Per [3] a fortnight is 14 days.

As I didn't find a good authority for what definition of a day a fortnight is
measured in, I'll assume here a day is 86400 SI seconds.

Thus, the speed of light in a vaccum is approximately
1.80*10^12 furlongs per fortnight.

1.80*10^12 = 299792458*86400*14/(220*0.9144)

[1] Resolution 1 of the 17th meeting of the CGPM (1983)
http://www.bipm.org/en/CGPM/db/17/1/
[2] Weights and Measures Act 1985
http://www.legislation.gov.uk/ukpga/1985/72
[3] The Concise Oxford Dictionary, 5th Edition, 1964, p. 480

(jakllsch)

Make break handling (where supported by the hardware/firmware) actually work.

(jakllsch)

Remove expected failure now that PR/44708 is fixed.

(njoly)

Remove the unneeded rump component; the library already includes the
module code that will be initialised by rump.

Fix PR/44708, t_zpool:create test failure for RUMP_LOCKDEBUG=yes
builds.

(njoly)

Do not protect wrong KASSERT by LOCKEDBUG ifdef/endif, the latter uses
its own mecanism. Kill them both.
From discussion with pooka@.

(njoly)

Allocate the locks from the same page we allocate the cpu_info and
pmap_tlb_info structure.  This assures they will be in KSEG0.

(matt)

Use MIPS_COP_0_OSSCRATCH instead $22

(matt)

Add missing KSEGX support.

(matt)

don't take the sc_lock in uhci_device_request() since it is already
taken (as verified by the assert at the top of this function.)

(mrg)

kassert the right lock.

(mrg)

my compiles-but-un-runtime-tested port of uhci to usbmp branch.
based upon the ohci and ehci changes.

(mrg)

don't forget to drop the lock in ohci_intr if we got there when
polling.  also don't try to drop a lock we don't have in ohci_open.

XXX: these might apply to ehci as well, i found them when porting
XXX: to uhci.

(mrg)

need to ensure kpreempt_disable() for softint_schedule().

fix the locking in ohci_timeout_task(), ohci_device_isoc_start(),
ohci_device_isoc_abort() and ohci_device_isoc_close().

uaudio(4) can still play with these, but mixerctl -a against it will
hang the writer.  however, mixerctl continues to run in a tight loop
and the system isn't soft-locked up at this point, an advance from
how it is in -current.

(mrg)

do_sys_sync() is part of vfs, hence rumpvfs. Used by pmf(9) during
device suspend, so explicitly link against rumpvfs to resolve that
symbol when linking rumpdev.

Unbreaks "rumptest" target.

(jym)

close the comment, there is a draft.

(christos)

PR/45687: Mike Pumford: savecore: (null): _ksyms_hdr not in namelist
Make it non-static and add a comment so that they don't become static again.

(christos)

make joy code consistent (avoid games with pointers), and fix the isa softc
declaration

(christos)

Fix mm_md_kernacc to report correct status for the direct mapped pages
between kernel start and end of physical memory.  Report everything else
as unhandled.

(skrll)

The union file system is as stable as other layered file systems so
no longer print a warning to the console.

Gnats is waiting ...

(hannken)

Remove ".gitignore" files which blacklist some files that actually are
in the NetBSD sources. This should fix build problems for people using
GIT to track current.

Problem reported by Evgeniy Ivanov on "current-users" mailing list.

(tron)

Remove trailing whitespace and fix a typo.

(wiz)

Do not xref to itself.

(wiz)

Remove unused file.

(skrll)

Whitespace.

(skrll)

Set list fix...

(jym)

- Add npf_tcpfl2case() and make TCP state table more compact.
- Adjust the state for FIN case on sim-SYN and SYN-RECEIVED.

(rmind)

secmodel_eval(9) may want to access securelevel before it is set
to the right value, so init it first before registering secmodel(9).

(jym)

KAUTH_GENERIC_CANSEE is no more.

(jym)

Update secmodel_examples to better describe the secmodel(9) API.

(jym)

Hook secmodel_extensions(9) to modules build.

(jym)

GETPARAMS => GETPARAM
SETPARAMS => SETPARAM

(jym)

Improvements in secmodel(9). Document secmodel_register(9), _deregister(9)
and _eval(9).

Add secmodel_extensions(9), and indicate the new sysctl(7) to let
ordinary users control the CPU affinity (user_set_cpu_affinity).

(jym)

When user_set_cpu_affinity is non-zero, only allow users to modify
the CPU affinity of the LWPs they own.

(jym)

adapt ohci, from mrg with some changes by me

(jmcneill)

correct some typos in comments, whitespace adjustments

(jakllsch)

change printed indicator and battery charge strings from ON and OFF to TRUE
and FALSE, because messages like "battery present: ON" are difficult for my
brain to parse.

(jmcneill)

Implement the register/deregister/evaluation API for secmodel(9). It
allows registration of callbacks that can be used later for
cross-secmodel "safe" communication.

When a secmodel wishes to know a property maintained by another
secmodel, it has to submit a request to it so the other secmodel can
proceed to evaluating the request. This is done through the
secmodel_eval(9) call; example:

    bool isroot;
    error = secmodel_eval("org.netbsd.secmodel.suser", "is-root",
        cred, &isroot);
    if (error == 0 && !isroot)
            result = KAUTH_RESULT_DENY;

This one asks the suser module if the credentials are assumed to be root
when evaluated by suser module. If the module is present, it will
respond. If absent, the call will return an error.

Args and command are arbitrarily defined; it's up to the secmodel(9) to
document what it expects.

Typical example is securelevel testing: when someone wants to know
whether securelevel is raised above a certain level or not, the caller
has to request this property to the secmodel_securelevel(9) module.
Given that securelevel module may be absent from system's context (thus
making access to the global "securelevel" variable impossible or
unsafe), this API can cope with this absence and return an error.

We are using secmodel_eval(9) to implement a secmodel_extensions(9)
module, which plugs with the bsd44, suser and securelevel secmodels
to provide the logic behind curtain, usermount and user_set_cpu_affinity
modes, without adding hooks to traditional secmodels. This solves a
real issue with the current secmodel(9) code, as usermount or
user_set_cpu_affinity are not really tied to secmodel_suser(9).

The secmodel_eval(9) is also used to restrict security.models settings
when securelevel is above 0, through the "is-securelevel-above"
evaluation:
- curtain can be enabled any time, but cannot be disabled if
securelevel is above 0.
- usermount/user_set_cpu_affinity can be disabled any time, but cannot
be enabled if securelevel is above 0.

Regarding sysctl(7) entries:
curtain and usermount are now found under security.models.extensions
tree. The security.curtain and vfs.generic.usermount are still
accessible for backwards compat.

Documentation is incoming, I am proof-reading my writings.

Written by elad@, reviewed and tested (anita test + interact for rights
tests) by me. ok elad@.

See also
http://mail-index.netbsd.org/tech-security/2011/11/29/msg000422.html

XXX might consider va0 mapping too.

XXX Having a secmodel(9) specific printf (like aprint_*) for reporting
secmodel(9) errors might be a good idea, but I am not sure on how
to design such a function right now.

(jym)

change callout_stop + usbd_delay_ms to callout_halt + callout_destroy in
ehci_detach

(jmcneill)

- add the bool type for IMMEDIATE flag.
- minor tweak to the handler example: it leaks 't' (on stack)
  when passed to sysctl_lookup(9), as it copyout its content via
  sysctl_data. That would not be the case if CTLFLAG_IMMEDIATE flag
  was set for this node but the example does not preclude that.

(jym)

add info on L2 TLBs and 1GB pages.

(chs)

Use the tlsoffset from the correct object.

(skrll)

map all of physical memory using large pages.
ported from openbsd years ago by Murray Armfield,
updated for changes since then by me.

(chs)

[merging from cherry-xenmp]
Make MP aware: use mutex(9) instead of spl(9)

(cherry)

Revert Christos's accidental changes.

(dholland)

Make ehci mpsafe.

(jmcneill)

default powerd=YES if acpi is present

(jmcneill)

exit 0 on success

(jmcneill)

Split out the cross-CPU tlb flushing code between XEN and non-XEN.
x86 tlb flushing is asynchronous and uses x86_ipi()
XEN tlb flushing uses synchronous hypercalls.

(cherry)

Extend the mppb(4) man page.

(rkujawa)