Makefile to build gpio.so.

(mbalmer)

Comment out the '##### Libraries that modules may depend upon.' for now,
so that modules written in C can be built.

(mbalmer)

Support FAT filesystem on RDB partition.

(kiyohara)

Add voodoofb with comment-outed.  tested on PegasosII only.

(kiyohara)

Change get parameters(width,height...) from prop_dictionary, like genfb.  tested on ofppc only.

(kiyohara)

Fix panic when boot time.  Require attribute drm to voodoofb.

(kiyohara)

gnu style.

(christos)

P5PB_MATCH_CV643D is supposed to be disabled by default

(rkujawa)

Update from OpenBSD merging in 1.111 and 1.112 with log messages:

1.112 Differential gain calibration makes the 6005 firmware crap
out, so skip it for now until we figure out why.  This
probably means the device won't function optimally, but
that's better than not functioning at all.  Makes my "Intel
Centrinto Advanced-N 6205" work quite well.

1.111 The 6005 and 6050-based parts need DC calibration turned
on, otherwise the firmware will crap out, at least on the
6005.  First step to getting my "Intel Centrino Advanced-N
6205" to work.

Both are commited by kettenis@openbsd.org.

(elric)

Fix broken logic that created R_SPARC_64 relocations when -Bsymbolic.
When -Bsymbolic, all resolved symbols need to be created with relative
relocations. The dynamic linker depends on that because it only
knows how to perform relative relocations in the first pass to
relocate itself.  This bug ended up creating R_SPARC_64 symbols for
12 relocation symbols in ld.elf_so, one of them dlopen(), which
ended up breaking things using dlopen(), most notably PAM.

(christos)

Descent into lua.

(mbalmer)

Add stub Makefile, not yet called.

(mbalmer)

Interface gpio(4) from Lua.  Not linked to the build.

(mbalmer)

Stop calling ATH_LOCK_{INIT,DESTROY}(), they don't exist any more.

(dyoung)

disable building the .info file

(christos)

In umidi_attach(), adjust aprint functions match what uaudio(4) does,
so that '\n' is printed in the appropriate places.

(jakllsch)

update string pointer when input is partial escape sequence or multibyte.

(tnozaki)

Use bsd.clean.mk instead of defining our own "clean" target.

(apb)

Get rid of the "I don't know what I was thinking / somebody should have
stopped me / does anybody read source-changes?" ATH_LOCK()/ATH_UNLOCK()
and bracket with splnet()/splx() instead.  This is still not *good*,
since ifnet ioctls are not (yet) synchronized, but could be no worse
than what we have, now.  Survives light testing with my (forthcoming)
ifnet ioctl synchronization patch.

(dyoung)

Use atomic_ops(3) to increase/decrease node reference counts, just
like the upstream code did, because the current reference counting is
potentially racy.  This works fine in light testing.

(dyoung)

usr/lib/lua is base-sys-usr not base-sys-root

(plunky)

Cosmetic: remove whitespace at the end of line.

(dyoung)

remove casts

(christos)

install bsd.lua.mk into DESTDIR

(plunky)

Add a bsd.lua.mk for building Lua modules

(plunky)

Fix off-by-one error and simplify at the same time.

(joerg)

Refer to X11R7.

(mbalmer)

Simplify.

(joerg)

Simplify

(joerg)

Simplify.

(joerg)

Optimise by simplifying.

(joerg)

Use __dead

(joerg)

Simplify

(joerg)

Fix under-allocation-by-one

(joerg)

Simplify.

(joerg)

As vnalloc() always allocates with PR_WAITOK there is no longer the need
to test its result for NULL.

(hannken)

Remove trailing whitespace. Sort sections. New sentence, new line.
Break to about 80 characters per line.

(wiz)

remove NETBSD_LIBGCC_SPEC that is no longer necessary with ligcc
already doing pic code properly.

(mrg)

obsolete libgcc_pic and friends.  all libgcc's are supposed to be pic
for platforms with shlibs already (and are.)

this obsoletes our hacks for the libgcc specs to use libgcc_pic, and
fixes a couple of other issues reported to me directly.

(mrg)

Update to reflect p5pb(4) changes.

(rkujawa)

Added experimental support for CV64/3D PCI bridge to p5pb(4). Approved by phx.

(rkujawa)

fix the sendmail, mailerconf and atf checks to use ${DEST_DIR}.
add a note about this to the top of the file.

(mrg)

Add HDMI/DP codec ID found on NVIDIA GTX 550 Ti

(jmcneill)

regen

(jmcneill)

add GeForce GTX 550 Ti

(jmcneill)

Provide symbol definitions for environ and __progname. These are duplicate
definitions to the ones provided by csu, but harmless since the linker resolves
them. There are 3 non weak data symbols provided in crt0.o:

    D __progname [initialized data, assigned to ""]
    B __ps_strings [BSS since assigned to 0]
    C environ [common, uninitialized]

__ps_strings had already a second definition in libc (in BSS), now we added
__progname and __ps_strings (in BSS).

To get rid of the duplicate definition, we can remove them from csu, and
move the assignments to libc.

This is done so that libc has no undefined symbols so that linker maps
that want to do:
    ...
local:
*;
    ...
don't end up producing link failures for libc.

(christos)

Include limits.h to get PTHREAD_KEYS_MAX, and move its definition there.

(christos)

Fix pthreads variables:
- Add missing stuff from Issue 6 (some of it commented out)
- Fix _POSIX_THREAD -> PTHREAD variable confusion
- Amend PTHREAD_KEYS_MAX to 256 since this is what libpthread defined internally

(christos)

Like the comment says, also MSR_IA32_EXT_CONFIG is unsafe; use rdmsr_safe().
Fixes the panic reported by njoly@.

(jruoho)

Bump version number to 20111006

(sborrill)

Provide function in libiscsi to return the compile-time limit on number of
targets. Use this function in the initiator, to ensure that the same limit
is used throughout (i.e. stop you attempting to compile the initiator with a
different limit to the library - the initiator uses the library for most of
its functionality).

If truncating the number of targets in the initiator, truncate it to the
correct amount, not half of what it should be.

Obey RFC3720 when determining targets. SendTargets=All MUST be supported
on a discovery session, and MUST NOT be supported on an operational session.
Previously, SendTargets=All was used both in the initial discovery session
and the later session. (from Daisuke Aoyama - author of istgt).

(sborrill)

Install i2cscan(8).

(wiz)

Correct pagermap emergva allocation.  From yamt@.

Tested by building i386 kernel with DTRACE defined which died 100%.

(uebayasi)

The crmfb device does not depend on iic; i2cbus and i2c_bitbang is
sufficient, says jmcneill@
Fixes a build problem for (at least) the GENERIC32_IP3x kernel.

(he)

Add man page for review.

(wiz)

Sort SEE ALSO.

(wiz)

Sort SEE ALSO.

(wiz)

Remove superfluous Pp.

(wiz)

enable building/installing audiocfg(1).  ok jmcneill.

(mrg)

remove a check against uvmexp.ncolors that is done inside uvm_page_recolor()
already anyway.

(mrg)

make struct page_size_map consistent across DEBUG/!DEBUG kernels.

(mrg)

Xref kcpuset(9).

(jruoho)

Add kcpuset(9). Ok rmind@.

(jruoho)

fix pasto, now SIOCS80211NWKEY and SIOCG80211NWKEY actually work

(macallan)

Fix lint unrelated "set but not used" warning, to unbreak test.

(njoly)

Remove extra space.

(njoly)

Include sys/syslog.h for log(9).

(njoly)

revert revision 1.291.  log(LOG_WARNING) is not strictly more
noisy than printf().

(apb)

forgot SIOCG80211NWKEY32 when adding SIOCS80211NWKEY32

(macallan)

When pipe1() calls pipe_create() and it fails, use the error
result from pipe_create(), don't assume it will always be ENOMEM.

From PR 45423 by Greg Woods.

(apb)

s/null terminate/NUL terminate/ in a comment.
From PR 45422 by Greg Woods.

(apb)

When killing a process due to RLIMIT_CPU, also log a message
with LOG_NOTICE, and print a message to the user with uprintf.

From PR 45421 by Greg Woods, but I changed the log priority (the user
might think it's an error, but the kernel is just doing its job) and the
wording of the message, and I edited a nearby comment.

(apb)

Merge duplicate information.

(christos)

Print "WARNING: negative runtime; monotonic clock has gone backwards\n"
using log(LOG_WARNING, ...), not just printf(...).

From PR 45421 by Greg Woods.

(apb)

Add MKCLEANSRC and MKCLEANVERIFY variables, defaulting to "yes".
You may set them to "no" to get the old behaviour of "make clean"
and "make cleandir".

(apb)

Use terminfo and the non-standard ttytype rather than termcap.

(roy)

There is no standard way of getting a list of aliases for the
terminal. However, some applications such as telnet want to know this.
ncurses dumps the terminfo header into an undefined variable ttytype
and these applications then parse it to work out the aliases.
We should do the same for now, until a standard mechanism for getting
the information is available or the need for it goes away.

(roy)

fujitsu(4) -> fujbp(4) && fujhk(4).

(jruoho)

Limit wd(4) transfers to 128 (512-byte) logical sectors, as the traditional
MAXPHYS value has for at least the past decade.

There are issues in wd(4) and possibly in host adapter drivers that need
work before this should be raised.

(jakllsch)

Make parts of this this somewhat less gross by using ilog2() from
<sys/bitops.h>.  Additionally, concoct shorter wchan names for >=1 MiB
buf pools using the irrelevantly-more-correct format specifier for u_int.

(jakllsch)

PR/43968 -- add 'segqlen' of TCPCB to 'netstat -P'.

OK by wiz@

(shattered)

add new test

(christos)

.Dq Dv \&:

(apb)

Mention what happens when we don't include :. It would be nice to use
.Dv :
but it produces ``'':

(christos)

bump more

(christos)

Fixed reversed/child parent and check the right variable for failure from
gson@

(christos)

Add new test.

(christos)

Handle c99 declarations in the middle of blocks.

(christos)

fixed warnings where wint_t is unsigned.

(christos)

Ensure we only pass cap names of one or two characters to tgetstr
and friends so we don't mistakenly try and convert a terminfo code to a
termcap code.

(roy)

Fix typo in diagnostic output

(martin)

tput now uses longname(3) to derive the terminal description instead
of trying to parse it from termpcap.
Fixes PR/43386.

(roy)

Move longname(3) from curses to terminfo, pre-cursor to fix PR/43386.

(roy)

Create the sysmon device node for the software watchdog

(hauke)

Make ffb(4) buildable without iic(4) driver.

(nakayama)

Regen for 3WARE 9750 Series RAID.

(njoly)

Add 3WARE 9750 Series RAID.

(njoly)

Fix thinko about halt vs. poweroff; from Moritz Wilhelmy in PR 45141.

(dholland)

expand the attr array to hold all the bytes we store in it.
not that the code actually does anything with the extra bytes,
but at least it builds with gcc 4.5.

(chs)

build machdep.c with -O1 to work around a gcc 4.5 issue.

(chs)

build cread.c with -O1 to work around a gcc 4.5 bug.

(chs)

iic_ioctl_exec: if the i2c controller's "exec" function uses -1 instead of
an error code to report failure, convert it to EIO.

(jmcneill)

Remove an invalid include path from CPPFLAGS

(jmcneill)

avoid printing an empty line in the sb800 case

(jmcneill)

remove iic, I fixed the i2c code

(jmcneill)

fix build for kernels with i2c host controller drivers but no iic(4)
driver present

(jmcneill)

Fix tparm.
As we now go via tiparm, store nums as int to conserve memory.

(roy)

Add the MI software watchdog to GENERIC (not to SMALLRAM, as usual).

(hauke)

Remove _ti_freeterm as consumers should just use del_curterm.

(roy)

Correct args

(roy)

Do not redefine CPUID_LAHF.

(njoly)

Note functions failure for unknown timer type.

(njoly)

Fix magic_descriptor/magic_file quoting in SYNOPSIS.

(njoly)

Fix the exceed test.

(christos)

fix broken change (parenthesis in the wrong place). From Nirbhay Choubey

(christos)

David Laight's script to convert K&R functions to ANSI

BUGS:
      Dpesn't handle functions with no arguments
      Doesn't handle va_list args
      Doesn't like blank lines or extra line breaks
      Generated ANSI definitions are often longer than 80 cols.

Added here so it won't get lost.

(wiz)

Fix man pages for tiparm and ti_tiparm.

(roy)

Install correct man page links.

(roy)

Use tiparm instead of vtparm.

(roy)

Use tiparm instead of vtparm.

(roy)

Use tiparm instead of vtparm.

(roy)

Correctly use ti_ instead of t_ for our extensions as per the man page.
Replace vtparm with tiparm.
tiparm is also non standard, but has been proposed at least.

(roy)

Fix non-COMPAT_50 compilation.

(mbalmer)

Unbreak and move the list of child devices to COMPAT_50.

(mbalmer)

Remove the GPIODETACH gpio(4) ioctl (it is still available in COMPAT_50) and
the 'detach' command line option from gpioctl(8).  Drivers that are
attached to gpio pins can be detached using the drvctl(8) command.

(mbalmer)

As getnewvnode() is prepared to wait for an allocation change vnalloc()
to always use PR_WAITOK.

No more 'WARNING: unable to allocate new vnode, retrying...' messages.

(hannken)

+i2cscan.debug

(njoly)

Don't cast void pointers.

(mbalmer)

Build i2cscan, since it is the list files.

(mbalmer)

Enable an iic(4) device to prevent a linker warning.

(mbalmer)

Lamport signatures were invented in 1979, and are used as one-time
signatures (since verification exposes parts of the public key, which
is derived from the secret key).  Lamport signatures can be built from
any cryptographically-secure hash function.  It is posited that
Lamport signatures are still secure, depending on the hash function
used, in the case of quantum computers; the same cannot be said of RSA
or DSA, for example.

In combination with hashtrees or merkle trees, Lamport signatures can
be generalised to work by exposing only the top element of the
hashtree, and by providing the elements of the hash tree that make up
the hash values of the other blocks.  Whilst this exposes one hash
value (out of 256 in the case of SHA256), the other hash values making
up the key are not exposed, and the key can be used again.

+ to create a secret Lamport key, a random number generator is used to
generate 256 pairs of random numbers, each number being 32 bytes in
length.  Each of these pairs has a "0" value and a "1" value.  A
secret key is thus 16KB.  The public key is derived from the secret
key by hashing each of the 512 32-byte random numbers to give 512
32-byte numbers, also 16 KB.

+ to create a lamport signature, the message itself is hashed to give
a 256-bit digest value.  For each of the bits in this hash value, the
corresponding hash from the private key is used.  If this bit is a 0,
the "0" value from the 256 pairs is used; if the bit is 1, the "1"
value from the 256 pairs is used.  In this way, 256 numbers of 256
bits each constitute the signature, making it 8KB in length.  The
unused values are discarded, and the key cannot be used again, since
the private key is exposed during verification.

+ to verify a signature, the same message is hashed to give a 256-bit
digest value, and the corresponding bits in the public key are used to
generate an 8KB computed signature.  The hashed signature is
calculated by hashing each of the 256 32-byte numbers in the signature
itself.  The resulting 8KB signature must match the calculated
signature for a match to be made.

This implementation provides the simple, one-time key, and uses the
SHA256 algorithm as the hash function.

Status:

Vendor Tag: CROOKS
Release Tags: lamport-signatures-base

(agc)

Newer ATI SB800 SMBus controllers don't report the base address in PCI
config space. According to the SB800-Series Southbridges Register Reference
Guide, we can still read this value from PM config space using indirect I/O.

(jmcneill)

_ACx is for active cooling policy, using it for envsys temp 'warn max' doesn't
make sense

(jmcneill)

No point on keeping hp700 on the old gcc; it is produces broken assembly
according to the new binutils.

(christos)

Attansic L2 is 10/100 but the status registers report 1000 Mbps, so don't
bother reading MII_EXTSR for this part. Also use MII_ANEGTICKS instead of
MII_ANEGTICKS_GIGE for this part.

(jmcneill)

- Replace calc_order() with ilog2(), wrap into SIZE2ORDER().
- Inline some functions, reduce some #ifdefs.

(rmind)

Add termname(3) entries

(pgoyette)

rename ioconf file to match module name, and remove an invalid include path
from cppflags

(jmcneill)

iic_ioctl_exec: initialize cmd before using it

(jmcneill)

fix broken sets.

(christos)

eat the last d in double constants.

(christos)

Also re-add usr/include/dev/i2c to the mtree file

(pgoyette)

Now that we're installing i2c/io.h usr/include/dev/i2c/ is no longer
obsolete!

(pgoyette)

Add termname(3) as defined by POSIX.

(roy)

Attempt to load the xc5k firmware at attach time using config_mountroot. If
it is unavailable the driver will try again on the next open of either the
video or dtv device.

(jmcneill)

now that iic is a module, add "iic" dependency to iic drivers

(jmcneill)

add support for detaching iic(4)

(jmcneill)

remove stray underscore.

(christos)

Bump SYMTAB_SPACE

(christos)

Bump SYMTAB_SPACE

(christos)

don't mark vers.c phony.

(christos)

put back pthread__dbg variable; this is set to no zero by td_open() when
debugging to avoid multiple td_open() mess.

(christos)

constify.

(christos)

enable thread debugging support.

(christos)

Enable thread debugging support (not completely working yet).

(christos)

- add objfile hook.
- constify.
- we cannot run things, so don't lie.

(christos)

I2C_IOCTL_EXEC: only allow i2c write ops if the device was opened for writing

(jmcneill)

add userland implementation of I2C_SCAN code

(jmcneill)

Install dev/i2c/i2c_io.h and implement the API in the iic(4) driver.
Obsolete the I2C_SCAN option as this can now be done from userland.

(jmcneill)

allow attaching an iic to auvitek if AUVITEK_I2C_DEBUG is defined, and
change the auvitek i2c lock from IPL_VM to IPL_NONE

(jmcneill)

undo previous, my sources were wrong.

(christos)

add a missing file, found with the new binutils. No chance we are going
to run mknative again on this, so the manual fix.

(christos)

The path getnewvnode()->getcleanvnode()->vclean()->VOP_LOCK() will panic
if the vnode we want to clean is a layered vnode and the caller already
locked its lower vnode.

Change getnewvnode() to always allocate a fresh vnode and add a helper
thread (vdrain) to keep the number of allocated vnodes within desiredvnodes.

Rename getcleanvnode() to cleanvnode() and let it take a vnode from the
lists, clean and free it.

Reviewed by: David Holland <dholland@netbsd.org>

Should fix:
PR #19110 (nullfs mounts over NFS cause lock manager problems)
PR #34102 (ffs panic in NetBSD 3.0_STABLE)
PR #45115 (lock error panic when build.sh*3 and daily script is running)
PR #45355 (Reader/writer lock error:  rw_vector_enter: locking against myself)

(hannken)

It's actually 'the flag locator', not 'flags' (as 'flags' can not be used
as a locator name for reasons unknown to me.)

(mbalmer)

Consistent use of 'return'.

(mbalmer)

Document gpioiic flag in a comment.

(mbalmer)

Build iic(4) and gpioiic(4) as modules.

(mbalmer)

Modules for iic(4) and gpioiic(4).

(mbalmer)

ii2c can be built as module.

(mbalmer)

Add the flag locator (missed in the previous commit).

(mbalmer)

Add a ga_flags field to the gpio_attach structure to hand driver
specific flags to drivers being attached at gpio pins.  gpioiic(4)
uses this to reverse the SDA/SCL signal order.  gpioctl(8) accepts
the flag values as optional argument to the attach command.
While here, make sure we retain backwards compatability and wrap compat
code in #ifdef COMPAT_50/#endif.

(mbalmer)

Describe the output more accurately, and in the correct order. Noted by
Snader_LB on freenode.

(dholland)

Fix ambiguous/confusing wording.

(dholland)

add a missing E

(christos)

The arm .arm.atpcs section does not have segment information, so skip it.

(christos)

disable test on the vax until we fix it.

(christos)

Undo previous, Checking for vax is more appropriate.

(christos)