seperate RMI CPU revision codes from RMI CPU processor codes
and improve comment

(cliff)

usr/libdata/debug directories for X11R7

(lukem)

Clean up <dev/acpi/acpireg.h>. While documenting the control methods is an
admirable goal, it is pretty much mission impossible; the specifications are
nearly thousand pages each and the amount of methods is counted in hundreds.

In addition, use ACPICA's native constants from <actypes.h> when possible.
Also move ACPI_STA_OK from "mpacpi.c" to <dev/acpi/acpireg.h> to simplify
the evaluation of device status.

(jruoho)

Use OHARE_NIRQ instead of NIRQ.
  NIRQ defined in powerpc/intr.h sinse 'Support 64-bit imask for powerpc/pic.'

(kiyohara)

Clean up comments.

(uebayasi)

Describe in a comment what the emit_dist_file target is for.

Under some circumstances, ${TOOL_CAT} may refer to an executable
that does not exist.  As a stopgap fix, use cat(1) instead of
${TOOL_CAT} in emit_dist_file.

(dyoung)

add usr/libdata/debug/usr/lib and usr/libdata/debug/usr/lib/i18n

(lukem)

Apparently exec.h is needed at least for i386/xen.

(pooka)

Undo infamous "mad \"scientist\"" experiment.  There is no need
for complicated application frobbing to get a kernel module loaded
into rump -- it will be loaded from the host automatically now.

(pooka)

Register module_base as an etfs recursive subdirectory.  This means
that rump will autoload kernel modules from the host
/stand/arch/vers/modules directory.

(pooka)

Add directory support to etfs: host directories and their contents
will be mapped to rumpfs based on the given key.  The directory
can be mapped either for a single level or recursively down the
entire subtree.

(pooka)

Allow everyone in the kernel to see module_base.

(pooka)

Implement kobj_renamespace() for rump.  Support for a few archs is
missing, but that doesn't really matter, since they are living in
their own "everything is a macro" happyland and don't support the
native kernel ABI anyway.

(pooka)

Give a cpp hint if the native kernel abi is supported.

(pooka)

Add an optional hook for adjusting module string table after load.
Currently to be used for rump "SAOS" namespace trick.

(pooka)

Change machine from "rump" to MACHINE so that module_base goes
right.

Now, any guesses on how many archs this seemingly trivial change
will break the principal metric, i.e. The Build(tm)?  Keep the bets
coming in.

(pooka)

Slight simplification to TLB shootdown handler on i386 and amd64.

(rmind)

Revert rev1.6, it shall return ETIMEDOUT for pthread calls.

(rmind)

Provide hppa_cpu_hastlbu_p

(skrll)

SANITY output tweak.

(skrll)

lwp_uc_pool has been replaced with kmem(9)
so also replace <sys/pool.h> with <sys/kmem.h>.

(tsutsui)

Fix typo. Doh.

(jruoho)

Improve and rearrange comments.

(jruoho)

AE_CTRL_TERMINATE -> AE_CTRL_FALSE.

(jruoho)

KNF

(skrll)

Remove used BSS allocation.

(skrll)

ungrossify

(pooka)

Avoid using demap_lock early in pmap_bootstrap before it is initialized.

(martin)

Describe the struct iovec better.

(jruoho)

Small markup improvements.

(jruoho)

Collect a garbage.

(uebayasi)

Clean up: move memory segment specific code from uvm_page_physload_common()
to uvm_page_physload().

(uebayasi)

Remove the unfinished code to add a memory segment after uvm_page_init().
It doesn't even compile.

(In the future, we should allocate struct vm_page [] on the added memory
segment for NUMA's sake.)

(uebayasi)

Partly rewrite amd64 TLB shutdown handler for the changes in x86 pmap.
At this point, branch seems to pass preliminar stress tests on amd64.

(rmind)

Move the ACPIVERBOSE blocks to a single place.

Also print non-devices in the EISAID/HID/UID/ADR-block

(jruoho)

fix comment (so bracket matching works as intended)

(lukem)

Apply renovated patch to significantly reduce TLB shootdowns in x86 pmap,
also provide TLBSTATS option to measure and track TLB shootdowns.  Details:

http://mail-index.netbsd.org/port-i386/2009/01/11/msg001018.html

Patch from Andrew Doran, proposed on tech-x86 [sic], in January 2009.

XXX: amd64 and xen are not yet; work in progress.

(rmind)

Add ubc_purge() and purge/deassociate any related UBC entries during
object (usually, vnode) destruction.  Since locking (and thus object)
is required to enter/remove mappings - object is not allowed anymore
to disappear with any UBC entries left.

From original patch by ad@ with some modifications.

(rmind)

Prevent ntpd instances from sending MODE_PRIVATE error responses back and
forth. This resolves CVE-2009-3563, but it should be noted that nobody uses
this code so far, so this is more of a preventive update than a security
one.

(tonnerre)

ubc_alloc: when replacing a cache entry, lock the old object from which we
are deassociating and removing the old mapping.

(rmind)

- Drop vmmap and its reserved page on hp700, sparc and x86.
- mm_init: use UVM_KMF_WAITVA when allocating a VA.

(rmind)

Fix hp300, missed in previous.

(rmind)

A round of compile and other fixes to previous.

(rmind)

Remove the incorrect table. Fix grammar.

(jruoho)

Mention the ACPICA debugger and the way to enter to it from ddb(4).

(jruoho)

Returning instead of breaking out if the sleep state is not available.

(jruoho)

As noted by jmcneill@, the specifications specifically mention that the _TTS
should be invoked before the system has notified any native mode device
drivers. Thus, do the call before pmf_system_suspend(9).

(jruoho)

Nothing uses x86_multicast_ipi() right now and it complicates many
CPU support, so remove it.

(ad)

Allocate the cpu_infos array dynamically.

(ad)

- console spam.
- atomic_dec_uint -> vfs_destroy.

(ad)

Make select/poll work with more than 32 CPUs.
No ABI change.

(ad)

Reduce memory spent on bookkeeping for large values of MAXCPUS.

(ad)

Pull-up rev 1.94

(rmind)

Drop per-"MD page" (i.e. struct pmap_page) locking i.e. pp_lock/pp_unlock
and rely on locking provided by upper layer, UVM.  Sprinkle asserts.

(rmind)

Fix KASSERTMSG() to be consistent with KASSERT() logic, not inverted.
Hi matt@!

(rmind)

- Invent mm_md_getva() and mm_md_relva() routines, provided by MD and
  indicated with __HAVE_MM_MD_PREFER_VA.  It will be used to deal with
  cache aliasing issues and thus fix little MIPS, ARM and friends.

- Convert dev_mem_readwrite() to use unmanaged mappings.  Fix a missed
  offset addition in a case of direct map.  Sprinkle various comments in
  the memory device driver.

- Add missing direct map handling on hp700 and vax.  Make checks across
  m68k ports more consistent, reduce the diffs.  Fix kernacc check miss
  on news68k.  Minor off-by-one fix for alpha.  Add MEMC_PHYS_BASE for
  mmap() case check on acorn26.  Misc clean-up.

(rmind)

Refactor uvm_page_physload_common().  Memory allocation failure here is
critical; panic if it happens.

(uebayasi)

Fix forgot to commit.
  Support 64-bit imask for powerpc/pic.

(kiyohara)

MAXCPUS -> maxcpus

(ad)

MAXCPUS -> maxcpus

(ad)

MAXCPUS -> __arraycount

(ad)

console spam

(ad)

console spam

(ad)

Sort sections, make HTML-ready.

(wiz)

Various fixes, mostly for typos.

(wiz)

Make uvm_page_physload() return the registered struct vm_physseg *.

(uebayasi)

Evaluate _TTS on sleep state transitions. This is mainly to pro-actively
support some weird BIOS, which may require the evaluation. Ok jmcneill@.

(jruoho)

Fix for PR#42572. I was running with this patch for several month without
any problems.

(matthias)

Add a note about AcpiEnterSleepStatePrep().

(jruoho)

Update.

(jruoho)

No need to spread bad shell script practices in manual pages...

(jruoho)

Add also a paragraph about the platform-specific nature of the thermal
management.

(jruoho)

Update this too. As the page was so laconic, and as there is little to write
about the driver, add an example on howto put the system into sleep when the
lid is closed.

(jruoho)

Update this.

Remove the nonsense about passive cooling. As there is no known bugs, speak
about CAVEATS instead, and add few items to this section as a TODO.

(jruoho)

Added entry about cdb

(joerg)

Add the constant database reader (cdbr(3)) and writer ((cdbw(3)).
They implement a space efficent write-once database with fast access
path. Switch the services(5) database to use cdb. The size of the
database file decreases from 2.1MB disk space to 307KB. Access
performance is about the same if setservent(0) is used and about an
order of magnitude faster otherwise. services_mkdb defaults to the new
format, but can optionally create the old db(3) format as well for
statically linked legacy applications.

(joerg)

Fix typo in comments.

(tsutsui)

The capacity of RAM is decided according to the model.

(nonaka)

Oops, the direction of the shift is opposite.

(nonaka)

ANSIfy, KNF

(cegger)

Amend previous.

(rmind)

nuke sc_stable_count.

(nonaka)

Another one bites the dust.

(snj)

Define _PATH_POWER and use it.

(jruoho)

switch from 4 clause to 2 clause BSD license.

(dbj)

Embarrassing typos: _PI_ -> _SI_, _PB_ -> _SB_.

(jruoho)

Do not use __func__ with ACPI_DEBUG_PRINT (it already adds it).

(jruoho)

Remove retrieving the zone name via something called "REGN".

We really should not add code that aims to satisfy some oddball firmware,
not in the generic drivers that aim to comply with the specifications.
In the long-term this is even worse than quirk tables.

(jruoho)

Some purely cosmetic editing in the name of readability: clean up the softc
and remove unused variables therein, remove unused constants, use
ACPI_DEBUG_PRINT, add more detailed comments, et cetera.

(jruoho)

pass LDSTATIC to recursive makes.

(christos)

Fix a bug involving acpitz(4) accidentally attaching to the _TZ_ scope.

This in turn was caused by a bug in ACPICA, which reports the types of _SB_
and _TZ_ scopes incorrectly for its own internal reasons (utglobal.c):

  /*
  * Predefined ACPI Names (Built-in to the Interpreter)
  *
  * NOTES:
  * 1) _SB_ is defined to be a device to allow \_SB_._INI to be run
  *    during the initialization sequence.
  * 2) _TZ_ is defined to be a thermal zone in order to allow ASL code to
  *    perform a Notify() operation on it.
  */

Thanks to cegger@ for noticing the bug and testing a fix.

(jruoho)

Support detach by drvctl(8).

(kiyohara)

Only post EOI once for shared interrupts.  From Mark Kettenis.

(skrll)

Support 64-bit imask for powerpc/pic.

(kiyohara)

Support 64-bit imask for powerpc/pic.

(kiyohara)

add missing .debug suffix

(lukem)

When extracting separate debug symbols with MKDEBUG, there's no need
to leave the original file unstripped...

(lukem)

Mention the last couple changes.

(snj)

Welcome to 5.1_RC1!

(snj)

Add utility function acpi_get_node().

This retrieves a struct acpi_devnode from a handle. Since this requires
accessing the global softc, it is better to do it in one place alone. The
same goes for possible locking of the node-queue; it is better not to
publicize such a lock for generic device drivers.

(jruoho)

Update for 5.1.

(snj)

Now that we are able to remove references to power resources, provide a
detachment routine for acpitz(4). Add comments explaining the relationship
between power resources and active cooling. Also some improvements to code
readability.

(jruoho)

Remove trailing comma in enum, per PR kern/43200.

(trailing commas in enums are allowed in C99, but not C89, and possibly
not in C++, so let's be nice to older and deviant compilers)

(dholland)

Edit previous HISTORY patch for style; note our 64-bit time_t change.
Bump date.

(dholland)

Document the options that actually exist, instead of some other
completely unrelated set. Bump date. (first time since 1993!)

(dholland)

Handle the "diedtime" field of the player log (which is not the high
scores file, the append-only log of all games) as int32_t instead of
time_t. Log files from before the 64-bit time_t change can now be read
again; however, log files from the last year of -current are hosed.
All none of you who play larn, take note...

(dholland)

Be explicit about sizeof(*var) being preferred over sizeof(type).

(christos)

fflush the score file while signals are blocked to decrease the chance of
it getting corrupted.

(dholland)

Fix compilation.

(joerg)

\\ -> \e
Drop redundant .Ic.
Add explicit empty argument for intentionally empty bracket and
parenthesis.

(joerg)

Use consistent naming - uvm_obj_*().

(rmind)

Align usage of *free and *alloc and clean up the firmware read code.
This fixes two panics that occur for kernels compiled with DIAGNOSTIC and
DEBUG. From Sverre Froyen

(christos)

Update build glue for pkg_install-20100421

(joerg)

Merge pkg_install-20100421

(joerg)

pkg_install-20100421:
Fix an off-by-one in the check for properly sized pkgdb entries.
It rejected perfectly valid entries.

Extract dependencies of libarchive from the pkgconfig file and thereby
drop knowledge of the needed libraries.

At least some versions of HP-UX are known to not support %zu, add a
workaround. This is using the black list approach for now.

Recognize xz as compression type for pkg_create.

The first time an error is hit while fetching packages, try to reget
from the same position. This works around the server closing the
connection while fetching dependencies.

Try to detect common forms of pkgdb corruption and issue a warning in
that case.

Refactor the pkg_vulnerabilities logic to use the compression support
from libarchive. This reduces the amount zlib/bzip2 interaction to
the linkage.

Add man pages for audit-packages and download-vulnerability-list which
point to pkg_admin and notes that the scripts are obsolete.

Status:

Vendor Tag: PKGSRC
Release Tags: pkg_install-20100421

(joerg)

Add MKBSDTAR, defaulting to no. It switching cpio and tar to the
libarchive based frontends.

(joerg)

mq_timed{send,receive}: as required by POSIX, return EINVAL on invalid
timeout and thread would have blocked, instead of ETIMEDOUT.  Change is
to abstimeout2timo(), thus also affects _lwp_park(2).

Reported by Stathis Kamperis some months ago.

(rmind)

don't try to print a bus_space_handle_t
now this at least compiles on sparc64

(macallan)

Do not try in postinstall(8) to replicate the code in etc/mtree/Makefile
that assembles /etc/mtree/NetBSD.dist.  Instead, use the Makefile's
new target, emit_dist_file, to assemble the correct NetBSD.dist.

Previously, 'postinstall -m amd64 -s $SRC_TOP' would install a
NetBSD.dist that was missing /usr/lib/i386/ et cetera.

(dyoung)

Remove lwp_uc_pool, replace it with kmem(9), plus add some consistency.
As discussed, a while ago, with ad@.

(rmind)

-build ldexp/frexp/modf locally again rather than relying on libc
reaching over, to get things into a sane state,
-do the usual namespace protection game with scalbn(f) because it
is used internally
(more symbol renamings are necessary)

The weak_alias change for mc68881 is untested.

(drochner)

Remove M_IOV (missed in previous commit) and M_1394DATA types.

(rmind)

use the local versions of ldexp/frexp/modf again rather than pulling
in libm source code. The libm functions depend on other libm functions,
this requires symbol renaming, and with the reachover method this
is going to be a mess. Also, bundling the dependencies into one .o
file has the potential to cause symbol conflicts on static linking.

(drochner)

Forgot that the firmware may change the power state behind our back,
so actively query the state when the sysctl(8) is invoked.

(jruoho)

Add hw.acpi.power sysctl-node.

(jruoho)

+ B/W for bandwidth, ATM and PIN, TBA

(dholland)

+ JBOD, NAS, OQL, OWL, RDF, SPARQL, TOS (type of service), UUID

(dholland)

minor edits, mostly capitalization.
remove FUD since it's in the main acronyms file already.
Don't use {a,b} notation for semantically unrelated expansions.

(dholland)

Add a missing parentheses.

(ahoka)

Mention that kmem_free(9) may block, just in a case it is not clear.

(rmind)

Use struct segment_descriptor for pcb_fsd and pcb_gsd instead of int[2].

(joerg)

One malloc(9) type is enough for ACPI; use M_ACPI.

(jruoho)

Simplify acpi_pcidev_find(): directly return a pointer to the device node.

(jruoho)

PR: 42850
Reviewed by:

Reduced the "expected to exist" dirs in path to just
/ /bin and /tmp
and change the "not expected to exist" dir to something
even less likely.
Add a comment to suggest why.

(sjg)

Enforce RLIMIT_FSIZE before VOP_WRITE.  This adds support to file
system drivers where it was missing from and fixes one buggy
implementation.  The arguably weird semantics of the check are
maintained (v_size vs. va_bytes, overwrite).

(pooka)

Cosmetics: be pedantic and use AE_CTRL_TERMINATE instead of AE_ABORT_METHOD
as a condition to break out of a loop.

(jruoho)

no [longer?] need to define _KERNEL when including <net/route.h>

(plunky)

When setting the power, always mark the state as invalid on failure, even if
going to D0. Also fix a potential NULL pointer dereference in an aprint_().

(jruoho)

Replace M_IOV and some malloc(9)s with kmem(9), and while there:
- Fix invalid free (M_TEMP vs M_IOV) in do_sys_recvmsg(), spotted by jakllsch@.
  Also, same fix in osf1_sys_sendmsg_xopen().
- Fix attempt to free non-allocated memory in error path in netbsd32___getfh30().
- Plug a memory leak in compat_43_netbsd32_orecvmsg().

(rmind)

Don't round the bus space map/unmap requests to page size, but keep track
of the exact regions requested.

Only ever map the whole FLEX area in when mapped and don't unmap.

G/C the BTLB stuff - it doesn't work. Maybe one day it will come back.

(skrll)

Um, as has been noted, INSTALL_FILE records what's installed if it's
done unprived, and that doesn't go down well on a re-build.  So use
${INSTALL} ${COPY} instead to fix this problem.

(he)

Remove a couple of zero length kmem_frees.

It should fix at least one crash when unloading the dtrace module,
possibly many others.

(ahoka)

Uppercase some proper names and abbreviations (probably more to do here).

(wiz)

Add some common ones.

(jruoho)

Make the ACPIVERBOSE output even prettier by removing cruft.

(jruoho)

vgonel() went away a long time ago, so stop declaring it.

(dholland)

Ticket 1380.

(snj)

Apply patch (requested by jym in ticket #1380):
Fix the NX regression issue observed on amd64 kernels, where per-page
execution right was disabled (therefore leading to the inability
of the kernel to detect fraudulent use of memory mappings marked as not
being executable).

(snj)

regen

(chs)

add missing argument to clone().  the symptom of this was that pthread_join()
would sometimes get stuck, such as in our "mutex2" regression test.

(chs)

mention crash ldscript.

(christos)

workaround for MKPIE=yes until binutils-2.20.1 is imported. From Piotr Meyer

(christos)