Implement device page struct vm_page * handling.

(uebayasi)

Define vm_physdev / vm_nphysdev, physical address segment data for managed
device pages.

(uebayasi)

vm_nphysseg -> vm_nphysmem

(uebayasi)

Bump date for previous.

(wiz)

Merge vm_physseg lookup routines.

(uebayasi)

Kill vm_page::phys_addr.

(uebayasi)

fix the example code to be closer to compilation and update for aibs(4)

(cnst)

Bump date for aibs.

(wiz)

Bump date for previous.

(wiz)

Nd argument does not need quotes (groff handles more than 9 arguments nicely);
add some commas.

(wiz)

use the latest envstat in the example output for aibs(4)

(cnst)

fix a typo resulting in warn-max being shown in two columns (one time instead of crit-min); ok pgoyette

(cnst)

regen.

(msaitoh)

Make some bge's name close to the real product names.

(msaitoh)

* s/misformed/malformed/;

* don't change .status to anything other than SVALID and SINVALID in refresh();

suggested by <pgoyette>

(cnst)

use ACPI_INTEGER and PRIx64, and check success of sensor_attach;    suggested by jruoho (Jukka Ruohonen)

(cnst)

New aibs(4) driver for ASUSTeK AI Booster (ACPI ATK0110) hardware monitor
with limit support.

        http://thread.gmane.org/gmane.os.netbsd.devel.kernel/35654

Reviewed by <pgoyette>, <jruoho> and <tech-kern>.

(cnst)

mention about mfi(4)'s GEN2 support.

(msaitoh)

Update the manual for mfi GEN2.

(msaitoh)

Add newer cars supports. Tested on MegaRAID SAS 9260-8i.
- Add MFI gen2 support from OpenBSD.
- Add entry for MegaRAID SAS 9260-8i

(msaitoh)

regen.

(msaitoh)

Add some MegaRAID Entries

(msaitoh)

Fix a bug that the system which has no logical volume panics in shutdown
hook. Without any logical volume, sysmon_envsys_register() fails. On such
case, sc->sc_sme must be NULL for the detach. Reviewed by dyoung and bouyer.

(msaitoh)

After reviewing the datasheets, most of the supported Winbond chips can
handles the new flags for selecting sensor type.  Update code and man
page accordingly.

(pgoyette)

Handle rump like the direct mapping case.

(joerg)

Remove trailing whitespace.

(wiz)

Add flags to all instances of lm(4), add a reference to man page

(pgoyette)

Document flags option for selecting temp sensor type

(pgoyette)

Add temp-sensor type selection (via autoconfig flags).  Currently only
enabled for W83627HF but easily extended to others if appropriate.

Fixes PR kern-42758

(pgoyette)

When a password is set for the bootloader ("installboot -o password=..."),
it currently complains about an unknown command and prints a usage if the
password is entered wrong:

        ...
        Choose an option; RETURN for default; SPACE to stop countdown.
        Option 1 will be chosen in 0 seconds.
        Password: *
        Password: *
        Password: *
        unknown command
        commands are:
        boot [xdNx:][filename] [-12acdqsvxz]
            (ex. "hd0a:netbsd.old -s"
        ls [path]
        dev xd[N[x]]:
        consdev {pc|com[0123]|com[0123]kbd|auto}
        modules {enabled|disabled}
        load {path_to_module}
        multiboot [xdNx:][filename] [<args>]
        help|?
        quit

        Choose an option; RETURN for default; SPACE to stop countdown.
        Option 1 will be chosen in 0 seconds.
        ...

This is confusing, plus someone may use it to determine bits of
information about the system. What should happen instead is that the user
is informed that the password is wrong:

        ...
        Choose an option; RETURN for default; SPACE to stop countdown.
        Option 1 will be chosen in 0 seconds.
        Password: ****
        Password: ****
        Password: ****
        Wrong password.

        Choose an option; RETURN for default; SPACE to stop countdown.
        ...

Implement the latter behaviour.

(hubertf)

Add ZTE MF626.  From Marc Vertes in PR kern/41667.

(snj)

Fix an off-by-one error.
Thanks to Matthias Drochner.

(roy)

Regen.

(snj)

Add ZTE MF626.  From Marc Vertes in PR kern/41667.

(snj)

Some bugfixes and enhancements.  Support for case .. of and fix calculation
of branch offsets.

(eeh)

Add entry for the mbuf cluster limit change.

(joerg)

Remove separate mb_map. The nmbclusters is computed at boot time based
on the amount of physical memory and limited by NMBCLUSTERS if present.
Architectures without direct mapping also limit it based on the kmem_map
size, which is used as backing store. On i386 and ARM, the maximum KVA
used for mbuf clusters is limited to 64MB by default.

The old default limits and limits based on GATEWAY have been removed.
key_registered_sb_max is hard-wired to a value derived from 2048
clusters.

(joerg)

Take another stab at fixing the LOCKDEBUG panic reported in PR
kern/39940 and by Martti Kuparinen on current-users@:  replace the
ioctl lock with finer-grained locking.  Lock the ports list and
wait to if_clone_destroy() until all threads are out of the softc.

Thanks to Martti Kuparinen for testing these changes.

(dyoung)

Update netpgp to version 1.99.17/20100208

Changes to 1.99.17/20100208

+ get rid of last 2 static variables - use the __ops_printstate_t struct
  passed down, and add the indent variable here too
+ get rid of 3 occurrences in reader.c where an automatic buffer was
  addressed (as part of a subsequent callback) by a struct field from
  a calling scope, and only valid within the callback.  Found by
  Flexelint and phk - many thanks.
+ print filename/"memory" when time problems occur when validating signatures

(agc)

Re-apply:

Invert the sense of the bit to mark if LOCKDEBUG is enabled to
disabled.

This will help my fellow developers spot "use before initialised"
problems that hppa picks up very well.

but fix the !LOCKDEBUG case by defining the "no debug" bits to zero so
they have no effect on lock stubs.

(skrll)

Add USERCONF as per all other GENERIC kernels

(skrll)

use

.if defined(HAVE_GCC) && ${HAVE_GCC} == 4

rather than

.if ${HAVE_GCC} == 4

as HAVE_GCC may be undefined

(plunky)

use

.if defined(HAVE_GCC) && ${HAVE_GCC} == 4

rather than

.if ${HAVE_GCC} == 4

as HAVE_GCC may be undefined

(plunky)

Abstract vm_physseg_find() to handle struct vm_page *.

(uebayasi)

Make vm_physseg lookup into a real function.

(uebayasi)

Make vm_physseg::lastpg exclusive end.

(uebayasi)

Add XIP flag and globally enable it on this branch.

(uebayasi)

Describe a new branch, uebayasi-xip.

(uebayasi)

pgo_get needs the page array to be initialized.

(mlelstv)

Move assertion to make check more clear.

(mlelstv)

Make UVMHIST build again.

(mlelstv)

Recognize the i2c bus on the AMD768 PCI-ISA/LPC Bridge

Resolves PR/42759 - thanks Michael Stapelberg

(pgoyette)

Use http: for links to files on ftp.NetBSD.org.

(snj)

Update version number examples for the post-2.0I world.

(snj)

Note the import of uthum driver from openbsd

(tonio)

Fix typo in Email address, spotted by pgoyette@

(tonio)

- ufs_balloc_range(): on error, only PG_RELEASED the pages that were
  allocated to extend the file to the new size. Releasing all pages
  may release pages that contains previously-written data not yet flushed
  to disk. Should fix PR kern/35704
- {ffs,lfs,ext2fs}_truncate(): Even if the inode's size is the same as
  the new length, call uvm_vnp_setsize(). *_truncate() may have been
  called by *_write() in the error path (e.g. block allocation failure
  because of quota of file system full), and at this point v_writesize
  has been set to the desired size of the file and not reverted to the
  old size. Not adjusting v_writesize to the real size cause
  genfs_do_io() to write to disk past the real end of the file.

(bouyer)

d_psize routine returns a number of blocks or -1 on error.
d_dump routine returns 0 or an error code.

(mlelstv)

Use filesystem blocks to address filesystem objects. f_iosize just
happens to be the same for current filesystems.

(mlelstv)

Remove a now-unused variable declaration.

(he)

Xref i386/splraise.9 and bump date for previous.

(wiz)

KNF.

XXX This should use bus_dma(9).

(uebayasi)

When using -lcurses, you also need -lterminfo.
This fixes the build for sun2, and also builds with LDSTATIC=-static,
since archive libraries don't record inter-library dependencies.

(he)

scsi_sense_data: add comments taken from src/sys/dev/ieee1394/sbp.h

(cegger)

Move MD spllower(9) and splraise(9) doco from the MI spl(9) to an i386
manual page.

(dyoung)

Fix date.

(wiz)

Enable acpismbus(4) for ALL kernel, per discussion on tech-kern

(pgoyette)

Improve the error message that results when you have a multi-variable .for
and the substitution list doesn't divide evenly.

(dholland)

Add man page for acpismbus(4)

(pgoyette)

Mention acpismbus(4) (expermiental)

(pgoyette)

Add acpismbus enries - commented out!

(pgoyette)

Add man page and cross-refs for experimental acpismbus(4)

(pgoyette)

Import my experimental ACPI SMBus Control Method Interface driver

XXX Should not be enabled in any configuration file until you have
XXX disabled corresponding "native" i2c driver!  Read the acpismbus(4)
XXX man page!

(pgoyette)

add entry for mips/spl.S so kernel can compile

(cliff)

Enlarge temporary buffer, so that event counter's name is not truncated
for ioapic interrupts.

(bouyer)

Welcome to 5.0.2!

(snj)

Give device node examples.

(snj)

Improve description of -mno-red-zone entry and use http for a couple
links.

(snj)

Split spl functions into their own sources.
Make sure fast softints clear l_ctxswtch in the softint thread, not their own.

(matt)

Add some more KASSERTs

(matt)

wait idle also needs to see if at least one interrupt is unmasked.

(matt)

Attach to SMBus on Intel P55 chipset.

(tnn)

Fix Dd, use Dq, and remove trailing comma.

(wiz)

__inline -> inline

(uebayasi)

Add uthum(4) man page

(tonio)

Add uthum(4) man page

(tonio)

Add uthum(4) driver from openbsd, adapted to handle both
TEMPerHUM and TEMPer devices

(tonio)

Add uthum driver, ported from openbsd

Handles Tenx temperature sensors TEMPer and TEMPerHUM
and declares them to envsys

(tonio)

Fix printf formats on amd64 (and probably other 64bit systems).

(dsl)

regen

(tonio)

Add TENX vendor and two devices

(tonio)

Sync with terminfo(5): Remove /etc/terminfo.db.

(wiz)

Add some prelim poolpage debugging code.
Don't use ptoa to expand pfns to paddrs since it's cast with a vaddr_t.

(matt)

Note notes.

(snj)

Update for 5.0.2

(snj)

Allow uvm_km_alloc to allocate from a specific vm freelist if the port wants
it to.

(matt)

When choosing a msgbuf for a LP32 kernel, make sure it's mappable via KSEG0

(matt)

Revert commit from Fri Feb  5 06:43:17 UTC 2010 by skrll:

      Invert the sense of the bit to mark if LOCKDEBUG is enabled to disabled.

      This will help my fellow developers spot "use before initialised" problems
      that hppa picks up very well.

It has to be done differently, because the semantics of mtx_owner in the non-
LOCKDEBUG case can vary significantly between archs, and thus it is not
possible to simply flip a bit to 1.

Ok core@, as at least i386 is unbootable right now.

(cube)

- when establishing an intr, if malloc fails, be sure to splx on the way out
- in interrupt dispatch, when ack-ing EIRR, preserve the softint bits

(cliff)

Typo in a wchan string.

(uebayasi)

A little constification and remove some old softintr cruft.

(matt)

<mips/softintr.h> is no longer needed.

(matt)

Save curlwp in context (even though it should already be there).

(matt)

Make vm_physseg lookup routines take the target vm_physseg.  This is for the
coming "managed" device segments.

(uebayasi)

Changes to 1.99.16/20100205

+ minor simplifications to netpgp(1) internally
+ fix a bug in netpgp_verify_file where a non-existent file while listing
  packets would cause a SIGSEGV
+ add duration arg to netpgp(1), and check for validity when verifying
  signatures
+ add birthtime arg to netpgp(1), and check for validity when verifying
  signatures
+ add netpgp commands to print pubkey, if desired
+ allow the passphrase for the signature to be taken from --pass-fd
+ get rid of static indent value when printing packet contents
+ print signature validity times when verifying a file's signature

(agc)

Allow __HAVE_FAST_SOFTINTS to be optional

(matt)

- properly map the schizo interrupt registers, so we can use the normal
  virtual mappings to access them.
- with the above fxied, kill SCHIZO_BUS_SPACE_BROKEN, and introduce
  an options CHEETAH for now, that disables the SPITFIRE code in locore.
- merge schizo_init() into schizo_attach() and clean up the softc a bit.

with this we are only one more step away from GENERIC booting on USIII
machines.

(mrg)

The rumpdev_net80211 until-recently-wip component is now rumpnet_net80211.

(pooka)

add rumpnet_net80211 to lists

(pooka)

Move the net80211 support component into the proper place in "net"
and install as rumpnet_net80211.

(pooka)

Don't touch past the end of allocated region.  It results segmentation
violation.

(enami)

Include locators.h for SSIOCF_ADDR_DEFAULT.

(snj)

Correct addressing of superblock updates.

(mlelstv)

Fix a typo in one DPRINTF (s/edata->des/edata->desc/).

(jruoho)

Don't look in /etc for a terminfo database.
Don't report the rescue database if we did not find the terminal in
there AND we were able to access prior databases.

(roy)

add PCC commands and manpages [pcc(1), pcpp(1) and ccom(1)] and libpcc

(plunky)

amd64 provides MACHINE_ARCH=x86_64 but pcc wants TARGMACH=amd64. handle that

(plunky)

add some missing (mostly hungarian or polish) dead accent combinations,
from lkundrak's netbsd-sk_cz-howto

(drochner)

One more callback of sme_refresh() routine needs to check first that
there is a routine to be called!

(pgoyette)

Mark up PKGDB_DIR with Dv (same as in pkg_install.conf).

(wiz)

when making VERSSTR, provide the target os/machine rather than the host.
have the CVS checkout date in the PACKAGE_STRING rather than in a separate
definition

(plunky)

remove a debugging printf.  cleanup the reinit of the lwp.

(matt)

Fix year.

(roy)

Fix infocmp when both -a and -x are given as arguments.

(roy)

Note that -a no longer sets -x.

(roy)

Retain commented out capabilities by default.

(roy)

Note that pkgdb_dir is deprecated and point to the alternative setting.

(jmmv)

Deprecate the pkgdb_dir settings from daily.conf and security.conf in
favor of the PKG_DBDIR variable in /etc/pkg_install.conf.  The purpose
of this is to only have to define the location of the packages database
in a single place and have all other system components pick it up.

pkgdb_dir is still honored if defined and the scripts will spit out a
warning in that case, asking the administrator to migrate to the
PKG_DBDIR setting.  We can't remove this compatibility workaround until,
at least, after NetBSD 6 is released.

(jmmv)

More adaptation for ports where char is unsigned char.

(he)

More adaptation for ports with unsigned char as char.

(he)

+infocmp.debug +tic.debug

(njoly)

Store the aliases against the entry, so that infocmp can reproduce them.
Handy for creating smaller terminfo databases.

(roy)

Added two new macros to device.h:
amiga_membarrier() enforces a reorder protection on memory read/writes.
amiga_cpu_sync() makes sure the instruction pipelines are flushed.
Both macros may be used in all amiga device drivers for compatibility with
amigappc. For the start I fixed some SCSI drivers.
cbiiisc and ahsc were tested and seem to work really fast with DMA now.
Some stability problems with amigappc remain nevertheless (spontaneous
kernel DSI traps with high CPU/SCSI load).

(phx)

dma_cachectl() flushes the data cache for Amiga SCSI drivers.
Drop into the debugger after oea_dumpsys(), which doesn't dump anything ATM.

(phx)

Avoid taking a (not yet) initialized mutex early during bootstrap.

(martin)

fix LOCKDEBUG panic 'uninitialized lock'.
seminit() calls exithook_establish(). exithook_establish() uses the exec_lock.
exec_lock is initialzed by exec_init(1).
Call exec_init(1) before seminit().

(cegger)

Make our unsigned char ports happy.

(roy)

Use the rescue terminfo.db in our images instead of termcap.

(roy)

Move the emergancy terminfo database from /root to /rescue.

(roy)

Allow loading of a terminfo database in /rescue.

(roy)

be more adaptable when choosing the target arch, try to relate
the MACHINE_ARCH to the supported architectures before failing.

(plunky)

just use the actual date in PACKAGE_CHECKOUT definition

(plunky)

don't build and install the crtbegin.o and crtend.o files at this stage,
they conflict with the system provided ones (lib/csu) and cause problems.

(plunky)

use SUBDIR= not SUBDIR+=

(plunky)

don't conditionally include the Makefile.inc, we need the BINDIR etc
definitions for installation

(plunky)

only go to WARNS=2 on these by default

(plunky)

Change to deal with new method of invoking softints.  Remove mips/softintr.c
from config files.  Fix SYMTAB_SPACE for WGT62V43

(matt)

Add __HAVE_FAST_SOFTINTS support.
Add routine to remap an uarea via a direct-mapped address.  This avoids
TLB machinations when swtching to/from the softint thread.  This can only
be done for lwp which won't exit.

(matt)

Invert the sense of the bit to mark if LOCKDEBUG is enabled to disabled.

This will help my fellow developers spot "use before initialised" problems
that hppa picks up very well.

(skrll)

vnode.h is not used here.

(uebayasi)

Cosmetic.  Shorten some long names.

(uebayasi)