.It, not .IT. Noticed by Henning Petersen in PR 43445.

(dholland)

seperate initializing spl is from initizlizing the PIC:
setting ipl_sr_map is moved from sb1250_icu_init to sb1250_ipl_map_init.

(cliff)

insert missing */
thanks matt

(cliff)

insert missing */
thanks matt

(cliff)

Wrap a call to tls_level_lookup() in #if USE_TLS to allow
postfix to build with MKCRYPTO=no.

OK tron@

(riz)

Catch up with whitespace changes in df.c rev 1.79 from over 2 years ago!

(pooka)

Pull up revision 2f7a426c0f4149d59a7f3717ebedd6c55998e8bc from upstream:

----------
Fix detection of crashed test cases

Prevent cross-test-case contamination that occured when a first test case
passed and the second crashed.  The second could pick up the result of the
first test case and not be reported as failed.

Similarly, change the way timed out test cases are reported back to the
caller.  The creation of a temporary results file was just a really stupid
way of passing information around and introduced false positives if the
test case creates a results file before timing out.

Fixes ticket #35.
----------

Problem originally reported by pooka@.

(jmmv)

Kill empty .TP

(joerg)

Make ioctl(CG14_SET_PIXELMODE) a proper NetBSD ioctl() ( as in use _IOW ),
also fix some bogosity in colour weights and depth.
With this and the kernel changes committed earlier the cg14 in my SS20 works
with the suncg14 driver.
TODO:
- beat some sanity in the order in which the Xserver issues various ioctl()s
  and opens or closes devices to reduce unnecessary flicker and make sure we
  end up in the right colour mode when entering and exiting X
- add hw cursor support
- support 8 bit visuals along with 24 bit or alone for high resolutions

(macallan)

Add a couple more hw offset definitions, add an ioctl() to set colour depth
which works like Solaris and Linux, and finally make mmap() behave like an
actual CG14. This should be all we need to get Xorg's suncg14 driver going.

(macallan)

sprinkle some const

(dholland)

ANSIfy function declarations

(dholland)

de-__P()

(dholland)

Allow -f -, as per PR 10898.

(dholland)

Move "space on 256MB sbmips" comment into alternate of #ifdef where it applies.
Matthew Green noticed this one, thanks.

(cliff)

After consultation with jdc@, apply the patch from PR 42412 even though
the crash went away -- it's not clear that it went away for any reason
besides by accident.

The change is: use calloc instead of malloc for allocating SCREENs.

(dholland)

- remove rmixl_spl_init_cpu calls from rmixl_intr_init_cpu
  because rmixl_intr_init_cpu only runs on cpu#0.
  rmixl_spl_init_cpu calls are now done in cpu_rmixl_atach (for cpu#0)
  or cpu_rmixl_hatch (for other cpus).
- add cpu_xls616_erratum to avoid count/compare conflict on XLS616
  XXX VERIFY WITH RMI
- rmixl_spl_init_cpu CPU#0 in attach, others in hatch
- remove splhi/splx for non-cpu#0 work in attach
  interrupts arent enabled yet anyway

(cliff)

- in findroot(), add "sd" to search for root
we are kind of lying about 'booted_device' here,
until we figure out how to get booted dev info from the firmware.

(cliff)

- in setroot(), in the `root on <dev> ...' case, if the lookup failed
and we have no rootdev, then we have to ASK.

(cliff)

- remove rmixl_spl_init_cpu calls from rmixl_intr_init_cpu
because rmixl_intr_init_cpu only runs on cpu#0.
rmixl_spl_init_cpu calls are now done in cpu_rmixl_atach (for cpu#0)
or cpu_rmixl_hatch (for other cpus).
- in evbmips_iointr, be sure to mask out count/compare interrupt
along with softints (these are handled elsewhere)
and improve coments
- fix bug in rmixl_send_ipi: 'cpus_running' is bit-indexed
by ci_index, not ci_cpuid

(cliff)

- fix typo in evcnt string

(cliff)

in pmap_bootstrap:
- increase Sysmapsize to allow mapping all memory (e.g. for big tmpfs)
- when calculating mips_virtual_end, cast Sysmapsize to vaddr_t
before multiplying by NBPG to avoid overflow

(cliff)

in MIPSX(kern_intr), before "Call the interrupt handler",
store s1 to TF_BASE+TF_REG_SR(sp) to save STATUS there.
no big deal, it seems to be unused;  at least now it is unused correctly.

(cliff)

- set mips_locoresw.lsw_bus_error = sibyte_bus_watch_check
- call sb1250_ipl_map_init() to set the ipl_sr_map for sbmips.

(cliff)

- in trap(), if traptype is bus error, call chip-specific bus error
handler in locoresw: (*mips_locoresw.lsw_bus_error)(cause)

(cliff)

- add lsw_bus_error to struct locoresw, provides hook to call
for chip-specific bus error handling/decode from e.g. trap()

(cliff)

add bus watcher support for sibyte

(cliff)

we can have ldap without crypto

(christos)

Allow syslogd to compile when MKCRYPTO=no.  No promises as to whether
it works or not, as this has been broken since late 2008, when the
"encrypted connections" support was added to syslogd.

(riz)

Take responsability for Postfix.

(tron)

Fix a typo.

(wiz)

Add missing .debug extension.

(njoly)

If the definition of a static function is #if DES, the declaration
of it should be too.  Fixes build of ed where MKCRYPTO=no.

(riz)

Tag some manpages with "complex" so a build with "MKCOMPLEX=no" completes
successfully.

(riz)

only copy PK_32, left rest of p_flag along.

(matt)

Fix socket passing in netbsd32 compat.  unfdpasses and incoming ssh
interactive session now work.

(matt)

Adapted sysinst and installation docs for modern NAS boxes.

(phx)

The sandpoint port is now focussed on 824x based NAS boxes.
Replaced GENERIC by GENERIC.NAS.
Added INSTALL_KURO for KuroBox.
GENERIC.NAS and RAMDISK are no longer needed.

(phx)

minor clarification

(dholland)

Explain variable expansion better. Requested by Aleksey Cheusov

(christos)

Fix build with DIAGNOSTIC.

(uebayasi)

Don't clear SOFT_INT bits in spl0.

(matt)

Add a _IPL_NAMES(pfx) which is a list of strings corresponding to the
IPL names.

(matt)

Update to the device NWO:
Use CFATTACH_DECL_NEW
struct device * -> device_t
struct cfdata * -> cfdata_t
printf -> aprint_normal_*
Use device_* accessors

(matt)

Add support for setting/clearing PK_32 on _LP64 kernels.  Make cpu_proc_fork
a real function and add it to vm_machdep.c and let it copy PK_32 on fork.
Properly clear/set PK_32 depending on ABI in setregs.  Lack of PX_32 use
tracked down by Cliff Neighbors.  [Ya! ps now works!]

(matt)

Version rumpuser ABI with a lightweight mechanism.

(pooka)

initialize specificdata for lwp0 too

(pooka)

Reference aps(4) and thinkpad(4) with each other.

In aps(4) note that sensor values are refreshed every 0.5 seconds, which may
have a negative effect on battery life.

(Since the sensors are purely descriptive and no time-sensitive protective
measures are being carried out, it would probably be a good idea to remove
the whole callout(9) from the driver and let the sysmon's poll cycle take
over.)

(jruoho)

take care of the usual pastos invisible to the human eye

(pooka)

Revert previous - it was committed unintentionally along with other changes.

(pgoyette)

+test

(pooka)

arbor days are here again

(pooka)

Add a test which checks autoloading modules from the host's
/stand/arch/vers/kmods works in rump (and that the result is usable ;).

On i386 this "just works".  For amd64, due to -mcmodel=kernel,
things are a little more complicated.  We must have the entire rump
kernel loaded in the lower 2GB.  Currently, this can be done either
by using the non-PIC version for the rump kernel compiled with
-mcmodel=small, or, as njoly pointed out, using netbsd32, which
causes vm_default_addr() to give something in the lower 2GB and
therefore shared libs "magically" getting loaded there.  I guess
it would be possible to put a suggested vaddr into the rump kernel
libs and make ld.elf_so map memory from the suggested address if
present ... but that's another show.

Also thanks to tron for access to an amd64 so that I could verify
the test works.

(pooka)

On amd64, allocate module_map memory from the lowest 2GB.

(pooka)

Add the ability to specify a preferred address the "map anon memory"
hypercall.

(pooka)

``twistymount'' regression test for scenario described in PR kern/43439

(pooka)

Similarly to cv_wait, fail if trying to cv_wait_sig() without threads.

(pooka)

ofppc: Support Gigabit Ethernet Controller for MV64361 on PegasosII by mvgbe@gt.

(kiyohara)

Remove _PATH_NOLOGIN before reboot, as well as before exit(), to prevent
a root file system which has been marked read-only in fstab from continuing
to prevent logins after reboot.  From Ian D. Leroux in PR#43390.

(riz)

Add mvgbe and gtpci.
Change gt devices to MI from evbppc.

(kiyohara)

Indent.

(kiyohara)

Support MV64361 on PegasosII by gt(4).
  + Support gtpci@gt instead of pegasospci more smartly.
  + Support Gigabit Ethernet by mvgbe@gt.

(kiyohara)

pci_chipset_tag_t is opaque - don't compare it to NULL, just use it.
shouldn't be a problem because "acpi_softc" variable isn't setup until
the "sc_pc" member is set.

fixes ia64 build.  tested on amd64.  ok jruoho.

(mrg)

document the acorn26, i386, ia64 and vax ddb commands in both the builtin
help and in ddb.4.  fixes PR#38304, PR#38305, PR#38309 and PR#38313.

(mrg)

declare 'struct pci_attach_args' before it is used.  fixes ia64 build.

(mrg)

Fix debug messages.
  - s/DEBUG/GT_DEBUG/.
  - Indent.
  - The old one is 20 bits in the mask of the address register, and the
    new one is 16 bits.

(kiyohara)

Note that MKZFS is actually "yes" on i386/amd64.

(riz)

The previous revision worked brilliantly if all you wanted to test
was autoload *not* working with an alternate path.  This revision
make the code double as good in the sense that it now works also
in case you *do* want it to work.

(pooka)

Only check the parent's power state if the parent has power management
capabilities. Based on a short DSDT survey, not all systems adhere to the
model noted in the comment.

(jruoho)

Add quirks for Creative Nomad MuVo, it chokes when told to prevent
media removal or synchronize cache.  Should fix PR#30389.

(jakllsch)

Add FUNCTIONS. List only complete prototypes in SYNOPSIS. Sort SYNOPSIS
according to the appearance in FUNCTIONS. Fix typo; cnm_trap() -> cn_trap().
Remove the .Va typedef struct cnm_state cnm_state_t; in SYNOPSIS. And some
miscellaneous markup improvements.

(jruoho)

Offset indent the example.

(jruoho)

Fix a comment and some debug printing.

(jruoho)

pulled fix from -current (1.207) back to setroot():
when there is a rootspec, but it's a disk, don't say it's "not configured"
just on the basis of finddevice() returning NULL -- that will always happen
when e.g. rootspec is "sd0a" and the device xname is "sd0".

(cliff)

Error out if we try to set power to a higher-power state than the parent's
state. This follows the "bus power state" -logic noted in the spec:

        bus: D1
                device A : D1 -> D0 -> error
                device B : D1 -> D2 -> success

                -> bus must remain in D1 due device A

Following this scheme, it is easy to derive the "bus power state" now that
we have the device nodes in a tree-like structure. If required, separate
acpi_power_get_bus() and acpi_power_set_bus() can be added in the future.

(jruoho)

use modulo instead of logical and. from pooka

(christos)

Allow to set a non-standard module_base for host kernel module loading.

(pooka)

fix build breakage by multiple definition of __mb_cur_max.

(tnozaki)

Add PNP0C33, "a hardware error device". Part of APEI.

(jruoho)

Add and organize some comments. Add also ACPI_DEVICE_EJECT constant to
indicate "ejectable" devices such as docking stations.

(jruoho)

Use the same notation with the PNP identifiers. Specifically note that
PNP0C80 is the device that is required for memory hot-plug on x86.

(jruoho)

Allow loading the symbols from objects which are not directly mapped
into memory, namely the main object (i.e. map->l_addr is NULL).

(pooka)

explain why this code is there.

(christos)

Regen

(pgoyette)

Add a couple entries for ATI Radeon 4250 HD

(pgoyette)

GDT, IDT, ISR, IST, IVT, LDT, LTR, MSI, SCI, SMI, TR, TSS.

(jruoho)

Use rump component libs on all platforms.  I'll add kernel module
autoloading as a separate test later.

(pooka)

Remove the unnecessary example.

(jruoho)

Put also powerof2(9) to SEE ALSO.

(jruoho)

Xref ilog2(3) and imax(9).

(jruoho)

Xref ilog2(3).

(jruoho)

Document ilog2() from <sys/bitops.h>.

(jruoho)

Procfs_lookup() does not lookup directory descriptors in the fd/
subdirectory.  There is no need for recursive vnode locking here.

Ok: Christos Zoulas <christos@netbsd.org>

(hannken)

Three small markup improvements.

(jruoho)

Fix xref.

(wiz)

get rid of the cg3/cg8 'emulation' kludge since we have the xf86-video-suncg14
driver now
this will probably need some work in cgfourteenmmap() to match what the
driver expects from a cg14

(macallan)

properly document alpha ddb "mach" commands.  PR#38303.

(mrg)

Xref mq(3).

(jruoho)

Ups. Fix wrong filename suffix.

(jruoho)

Xref bintime_add(9).

(jruoho)

Briefly document the bintime_add() family of static inline functions.

(jruoho)

Create a manpage for igphy(4).  kern/35398.

(riz)

Don't unnecessarily clutter the SEE ALSO.

(jruoho)

Support big endian machines.

(kiyohara)

attach on Sil240 and IXP 300 chips as well, as if they are 3112
variants.  should fix PR#32577, id's found in linux driver.

(mrg)

regenerate

(mrg)

add two pciids for the satalink driver, from the linux driver, for PR#32577.

(mrg)

look for and find config.h.  part of PR#43433.

(mrg)

include a config.h for openchrome as well.  part of PR#43433.

(mrg)

Remove SIOCGIFADDR. Noted by Ross Patterson in PR kern/21513.

(jruoho)

Grammar, markup.

(jruoho)

Comment.

(uebayasi)

Fix spelling of "whether", from Ryo HAYASAKA in PR 43432.

(riz)

Add a missing word.

(jruoho)

Now that we have 1:1 mapping between handles and nodes, use the former for
all power resource operations. Eliminates the wrapper functions.

(jruoho)

Use the new ACPICA functions AcpiAttachData() and AcpiGetData() to associate
all "struct acpi_devnodes" to their corresponding ACPI_HANDLEs. Anywhere in
the acpi(4) subtree, the node-structure can be obtained from a handle via
acpi_get_node(). The idea is similar to e.g. device_private().

Benefits: (a) simplifies code, (b) avoids issues with locking as ACPICA does
the serialization for us, (c) avoids the need to access the glocal softc, and
(d) avoids the O(n) loop required to search for a handle from the node queue.

(jruoho)

Comment.

(uebayasi)

Fix xref.

(wiz)

Grammar and spacing fixes.

(wiz)

don't use dv_xname directly.

(bsh)

Fix a bug: the last elements in the _PRW package are direct references to
power resources, not devices. Thus, we can not use the function that sets
the power state of a device.

(jruoho)

Refactor slightly: we need to expose the function that turns individual
power resource on or off.

(jruoho)

Rump components do not, at least for now, get to tell what's NEEDED.

(pooka)

1. MB_LEN_MAX switch MD to MI.
2. unfortunately hppa's MB_LEN_MAX is defined incorrectly 6 instead of 32
  so we have to add more setlocale(3) __RENAME func, __setlocale50.
3. move setlocale1.c and setlocale32.c to lib/libc/compat/locale/*
  prepareing for next libc major crunk.
4. bump libc minor version.

(tnozaki)

Now that bus_handle_t is signed, avoid % operations on it. Convert all
% PAGE_SIZE to & PAGE_MASK. We don't expect to support non power of two
page sizes anytime soon ;-)
This makes PCI work again on my au1550.

(martin)

Build the power resource queue dynamically. This eliminates the need to call
acpi_power_res_add() in the main scan function, also saving few bytes of
memory.

(jruoho)

no need to typecast in pathconf anymore

(pooka)

Reenable pathconf now that the retval argument is register_t.

(pooka)

Make retval argument for pathconf a register_t to match VOP_PATHCONF.
This makes the size the same on 64bit archs.  Don't bother bumping
any version, since you'd have explicitly had to jump through some
hoops to use pathconf before.

(pooka)

Note ACPIVERBOSE. Use lower case letters in .Ss subtitles.

(jruoho)

Remove some leftover .\" comments.

The sysctl variables will be described in sysctl(7).

(jruoho)

Build links also for mq_timedreceive(3) and mq_timedsend(3).

(jruoho)

Mostly refer to the main page, mq(3), in SEE ALSO.

(jruoho)

Add mq(3), a manual page for the POSIX message queues. This tries to act as
a short introduction to the rationale and API, noting also some pros and cons.

rmind@: basically ok. Please feel free to adjust, correct, and extend.

(jruoho)

Add "optional" keyword to rc.d/xdm and rc.d/xfs. Fixes PR misc/43307.

(jruoho)

Improve wording describing the behavior of the attributes argument, and
add reference to pthread_attr(3), as requested in PR 42871. The text
added is different from but based on a suggestion by Matthew Mondor.

(dholland)

Instead of disabling all GPEs for non-wakedevs, set both runtime and wake
GPEs, but unset only wake GPEs. The old behavior was questionable, as the
intention was only to disable a device's ability to wake the system.
Maintaining the runtime GPE generation is more important now as ACPICA
supports sharing a single GPE across multiple devices.

Discussed with jmcneill@.

(jruoho)

Remove a test (that a created directory had a predictable inode number)
that no longer makes sense since the vmlocking2 branch was merged.

(riz)

Use atf_check to su in a couple cases, to make the test failure
easier to understand when the unprivileged user has /sbin/nologin
for a shell.

(riz)

Four more quirky umass(4)es.
These include the Philips SA235, and iRiver iFP-series audio players.
Mostly from PR#25440 and PR#43264.

(jakllsch)

Update acpiverbose module to use module_autoload() rather than module_load().
Load the module right before each attempt to use its features, and let the
module subsystem handle unloading.

(pgoyette)

Update scsiverbose module to use module_autoload() rather than module_load().
Load the module right before each attempt to use its features, and let the
module subsystem handle unloading.

(pgoyette)

Actually, disable pathconf for a while longer: the retval needs a
little finetuning (not that everyone is jumping right to using it,
but fixing will have to wait until tomorrow, and who knows what
i'll forget during the night and how long it will eventually take).

(pooka)

Regenerate.

(jakllsch)

Add Philips SA235 Digital Audio Player.

This USB Mass Storage Class device stops responding after a
SCSI PREVENT ALLOW MEDIUM REMOVAL command.

(jakllsch)

Sort Sigmatel products.

(jakllsch)

Pass VOP_PATHCONF to backend file server.  inspired by njoly.

(pooka)

signal pathconf in the vector of supported ops

(pooka)

Kill empty .TP.

(joerg)

Make vers.c depend on sys/param.h too, to ensure that this file is
regenerated for on kernel version bump. Avoids __NetBSD_Version__ and
osrelease out of sync problem for mkupdate builds.

ok from pooka@.

(njoly)

Fix a build failure.

Apparently upstream does not bother to test-build their own releases.

(jruoho)

Add missing files required for iASL from ACPICA 20100528.

(jruoho)

Note the ACPICA update to 20100528.

(jruoho)

Update usbverbose module to use module_autoload() rather than module_load().
Load the module right before each attempt to use its features, and let the
module subsystem handle unloading.

(pgoyette)

Update pciverbose module to use module_autoload() rather than module_load().
Load the module right before each attempt to use its features, and let the
module subsystem handle unloading.

(pgoyette)

Update miiverbose module to use module_autoload() rather than module_load().
Load the module right before each attempt to use its features, and let the
module subsystem handle unloading.

(pgoyette)

We can no longer explicitly clear a GPE by calling AcpiClearGpe() in the
interrupt handler. However, all edge-triggered GPEs should already be
cleared before our GPE handler has a chance to run.

The reason can be found from the changes in the locking primitives of
ACPICA. All GPE operations now use a spin mutex on AcpiGbl_GpeLock, acquired
via AcpiOsAcquireLock(). This same lock is now acquired unconditionally in
the AcpiClearGpe() function. This causes a deadlock of the following form:

  ...

  AcpiEvGpeDetect() : acquire AcpiGbl_GpeLock;

  -> AcpiEvGpeDispatch();

  -> acpiec_gpe_handler();

  -> AcpiClearGpe() : acquire AcpiGbl_GpeLock;

  -> panic.

(jruoho)

ACPICA 20100331:

Completed a major update for the GPE support in order to improve
support for shared GPEs and to simplify both host OS and ACPICA
code. Added a reference count mechanism to support shared GPEs that
require multiple device drivers.  Several external interfaces have
changed. One external interface has been removed. One new external
interface was added. Most of the GPE external interfaces now use the
GPE spinlock instead of the events mutex (and the Flags parameter
for many GPE interfaces has been removed.) See the updated ACPICA
Programmer Reference for details. Matthew Garrett, Bob Moore, Rafael
Wysocki. ACPICA BZ 831.

Changed:
    AcpiEnableGpe, AcpiDisableGpe, AcpiClearGpe, AcpiGetGpeStatus
Removed:
    AcpiSetGpeType
New:
    AcpiSetGpe

(jruoho)

Add new files from ACPICA 20100528.

(jruoho)

ACPICA does not understand the const type qualifier.

(jruoho)

Merge ACPICA 20100528

(jruoho)

Import ACPICA 20100528.


Status:

Vendor Tag: intel
Release Tags: acpica-20100528

(jruoho)

Bump date for previous.

(wiz)

fix conditional, from dholland.

(christos)

MULTIPROCESSOR fixes

- provide curcpu in a control register instead of curlwp
- define {GET,SET}_CURLWP and GET_CURCPU and use whereever possible.
- define a cpu count and use it in CPU_INFO_FOREACH

XXX hppa_ncpus isn't valid yet.

(skrll)

Only include opt_m68k_arch.h if _KERNEL_OPT is defined, so that librump
builds again.  OK'ed by mrg@.

(he)