Avoid undefined behavior, from Chris Spiegel in PR 43860.

(dholland)

Bump date for previous

(apb)

Collapse <abckslash><newline> sequences in /etc/ifconfig.xxx files read
by /etc/rc.d/network.  Fixes PR 41662 by Christoph Badura, which was
also reported independently by Jeremy C. Reed.

(apb)

Add a collapse_backslash_newline function to rc.subr.

(apb)

Minor fixes.

(uebayasi)

Implement XIP "putpages".  Invalidate MMU mappings of pages at the
request of PGO_FREE.  PGO_DEACTIVATE and PGO_CLEANIT do nothing, because
XIP pages are neither queued nor writable.

Allocate read-only "zero" page per vnode.  Put it at offset 0 of vnode's
uvm_object.  This per-vnode "zero" page is mapped to all hole pages of
the vnode.  If one of its mapped pages are forced to be PGO_FREE'ed,
all the mappings are invalidated.

(uebayasi)

Wrap long lines.

(uebayasi)

ffs_vget: Mark XIP only for VREG vnodes.

(uebayasi)

locking changes

(yamt)

fix rewinddir on nfs.  fix PR/42879 (and probably PR/40229.)

(yamt)

Bump date for more supported I/O DATA devices.

(wiz)

New sentence, new line.

(wiz)

be more conservative about allocing.

(christos)

Remember the correct pointer which we will free later.

(tron)

Unlock the environment lock if __alocenv() fails.

(tron)

Remember memory used by allocated environment variables instead of
using a bitmap. This deals with the case where a variable is first
set via setenv(3) or putenv(3), then overwritten by changing
"environ" directory and afterwards overwritten with setenv(3) again.

This stops "zsh" from crashing under NetBSD-current.

Code reviewed by Christos Zoulas.

(tron)

Regen from usbdevs rev 1.562:
> Add support for I/O DATA ETX-US2.

(tsutsui)

Add support for I/O DATA ETX-US2.
Patch from PR kern/43040.

(tsutsui)

Treat empty or missing fs_passno field like it has a value of 0 as fstab(5)
specifies.
Related to PR misc/43905 but does not fix the underlying issues.

(bad)

Remove obsolete message:
'If you are using a SUN type 4 keyboard, please enter "sun-type4".'

(tsutsui)

Use vt100 rather than sun for $TERM.
The latter one is not supported by the hardcoded libterminfo fallback
without external database.

(tsutsui)

Change last commit to Use "ptrdiff_t" instead of "intptr_t" because the
former is more appropriate in this case.

(tron)

Add explicit cast to fix lint error on 32bit platforms.

(tron)

Mark /usr/share/doc/html/ntp/build.html as non-obsolete, since a build
installs it.

(wiz)

sort -u.

(wiz)

Remove obsoleted directory /usr/share/doc/html/ntp/build/scripts

(wiz)

Remove trailing whitespace, merge error sections for same error, nits.

(wiz)

Rename rb.h to rbtree.h, as it is more appropriate (c.f. ptree.h).  Also
helps find code that hasn't been updated to use the new rbtree API.

(matt)

Make groff happy.

(uwe)

Add nbuf_advfetch() and simplify some code slightly.

(rmind)

Fixes/improvements to RB-tree implementation:
1. Fix inverted node order, so that negative value from comparison operator
  would represent lower (left) node, and positive - higher (right) node.
2. Add an argument (i.e. "context"), passed to comparison operators.
3. Change rb_tree_insert_node() to return a node - either inserted one or
  already existing one.
4. Amend the interface to manipulate the actual object, instead of the
  rb_node (in a similar way as Patricia-tree interface does).
5. Update all RB-tree users accordingly.

XXX: Perhaps rename rb.h to rbtree.h, since cleaning-up..

1-3 address the PR/43488 by Jeremy Huddleston.

Passes RB-tree regression tests.
Reviewed by: matt@, christos@

(rmind)

add fmemopen

(christos)

add man page as promised.

(christos)

fix copy/paste/modify error in a diagnostic panic message

(jakllsch)

libc minor was bumped.

(skrll)

If not using O32 on mips, we need to read ELF64 symbols too.

(matt)

Make the GOT1_MASK 64-bit clean.

(matt)

fixed a fd leak, patch by getlaser (at) gmail.com

(vanhu)

protect the __allocenv call with the mutex; from Enami

(christos)

Fix memory leaks; thanks Enami!

(christos)

Add support for envsys(4) version 2 ioctls.

(njoly)

Comment reformat.

(skrll)

Wrap long line.

(skrll)

Whitespace.

(skrll)

Wrap long lines

Reformat comments.

(skrll)

add fmemopen(3) derrived from POSIX:2008.
libc minor bump.

(tnozaki)

Do not discard ACPI PCI addresses with function number 0xFFFF: the
ACPI specification allows them (ACPI 4.0a, p. 200).

ok jruoho@

(gsutre)

GNU grep-2.7 out.

(wiz)

prevent integer oveflow. From Maksymilian Arciemowicz

(christos)

new environment test

(christos)

unit test for environment. Grr, someone should make fixing the sets easier
otherwise people are not going to want to write unit-tests.

(christos)

Use a bit array to keep track of malloced environment entries so we can
free them.

(christos)

PR/43899: Nicolas Joly: setenv(3)/unsetenv(3) memory leak.
Partial fix: Don't allocate a new string if the length is equal to the
old length, because presumably the old string was also nul terminated
so it has the extra byte needed.
The real fix is to keep an adjunct array of bits, one for each environment
variable and keep track if the entry was allocated or not so that we can
free it in unsetenv.

(christos)

== file close operations ==
- use PUFFS_KFLAG_WTCACHE to puffs_init so that all writes are
immediatly send to the filesystem, and we do not have anymore write
after inactive. As a consequence, we can close files at inactive
stage, and there is not any concern left with files opened at
create time. We also do not have anymore to open ourselves in readdir and
fsync.

- Fsync on close (inactive stage). That makes sure we will not need to
do these operations once the file is closed (FUSE want an open file).
short sircuit the request that come after the close, bu not fsinc'ing
closed files,

- Use PUFFS_KFLAG_IAONDEMAND to get less inactive calls

== Removed nodes ==
- more ENOENT retunred for operations on removed node (but there
are probably some still missing): getattr, ooen, setattr, fsync

- set PND_REMOVE before sending the UNLINK/RMDIR operations so that we avoid
races during UNLINK completion. Also set PND_REMOVED on node we overwirte
in rename

== Filehandle fixes ==
- queue open operation to avoid getting two fh for one file

- set FH in getattr, if the file is open

- Just requires a read FH for fsyncdir, as we always opendir in read
mode. Ok, this is misleading :-)

== Misc ==
- do not set FUSE_FATTR_ATIME_NOW in setattr, as we provide the time

- short circuit nilpotent operations in setattr

- add a filename diagnostic flag to dump file names

(manu)

Avoid casting from d_val to plttype of different signedness.
Just two booleans for the same purpose.

(joerg)

Add empty line in ERRORS section.

(njoly)

Add Marvell Secure Digital Input/Output (SDIO) Interface Driver.

(kiyohara)

Print bus width and clock for feature(High speed support).

(kiyohara)

Move shared library bits of the tests out to shl.mi, so that sun2,
which only does static linking, can complete the build.

(he)

Don't depend on shared library dependencies; we also need -lpuffs here.
This allows us to link statically.

(he)

Reorder the library specifications, so that we don't depend on the
library dependencies recorded in shared libraries, allowing these
to be linked statically as well.

(he)

Do symbol renaming for these bits as well, so that we don't get
link-time errors caused by conflicts with normal user-space libraries
when we link statically.

The tests still pass for i386 after this set of changes, and this now
builds for sun2 (after a few more changes).

(he)

Replace some whitespaces to Tab.

(kiyohara)

+ Remove crypto, swcrypto.  No need.
+ Enable tap.  It needs Bluetooth PAN.
+ Enable bridge for PAN/netDUO.
+ Enable dmoverio.

(kiyohara)

Move 'include files.sdmmc' to files.evbarm.

(kiyohara)

Remove #ifdef NPRCM.  Always include omap2_prcm.h.

(kiyohara)

If GUMSTIX_NETBSD_ARGS_CONSOLE is defined, argument 'console=ttyS0,<speed>'
that u-boot passes Linux will be interpreted.

(kiyohara)

Remove #if !defined(OVERO) - #endif.  Overo can call delay().

(kiyohara)

1406

(snj)

Pull up following revision(s) (requested by christos in ticket #1406):
dist/bzip2/decompress.c: revision 1.2
avoid integer overflow that can lead to buffer overflow

(snj)

1406

(snj)

Pull up following revision(s) (requested by christos in ticket #1406):
dist/bzip2/decompress.c: revision 1.2
avoid integer overflow that can lead to buffer overflow

(snj)

1455

(snj)

Pull up following revision(s) (requested by christos in ticket #1455):
dist/bzip2/decompress.c: revision 1.2
avoid integer overflow that can lead to buffer overflow

(snj)

1455

(snj)

Pull up following revision(s) (requested by christos in ticket #1455):
dist/bzip2/decompress.c: revision 1.2
avoid integer overflow that can lead to buffer overflow

(snj)

Punctuation improvement, bump date for FAST_IPSEC description.

(wiz)

Bump date for O_DIRECTORY, more markup, merge two error sections.

(wiz)

Avoid fault if acpi_softc is NULL at attempted power-off.
XXX at least some of this should be factored off into arch/x86.

(jakllsch)

fix typo. From Anon Ymous

(christos)

get the correct length of username when processing ADMIN_LOGOUT_USER, patch by rweikusat (at) mssgmbh.com

(vanhu)

fixed a typo in macros, reported by marisp (at) mt.lv

(vanhu)

Our API just uses typedefs, so don't use the struct symbol which will avoid
any potential conflicts with userland.

(roy)

Back out previous change.  "automagically" is NOT a typo.

(kiyohara)

1. Don't try to call the driver's get_limits routine if it doesn't have
  one.  Fixes a panic reported by njoly@ in private Email

2. Acquire the device before checking whether or not the get_limits
  routine exists.

XXX There's still a bug in here somewhere that prevents setting of
XXX limit values after using 'envstat -S'

(pgoyette)

implement O_DIRECTORY as standardized in POSIX-2008,
for both native and linux emulations.
this fixes the rest of PR 43695.

(chs)

Mentions that IPSEC_DEBUG works with FAST_IPSEC implementation

(degroote)

Mentions the FAST_IPSEC options

(degroote)

Fix ipcomp input counter

Reported Wolfgang Stukenbrock in pr/43250.

(degroote)

moved from utmp.h to utmpx.h (patch provided by marcin.cieslak (at) gmail.com)

(vanhu)

O2W is "off to work", from mrg@ and spz@.

(soda)

Fix typo.

(wiz)

note bozohttpd 20100920

(mrg)

+ Add macro SMC_CAPS_8BIT_MODE for 8bit mode support.
+ Typo.  s/automagically/automatically/.

(kiyohara)

Fix build failure.
  A new variable width and value had been forgotten since r1.7.

(kiyohara)

add separate flags for putchar() based copycols() and copyrows() methods
for hw that can accelerate one but not the other, like Sun's Creator series
VCONS_DONT_READ does the same as before

(macallan)

* implement hw acceleration for putchar()
* use VCONS for copycols()
* don't map the framebuffer now that we don't access it anymore
* make a bunch of drawing engine syncs optional

(macallan)

defflag some flags for the ffb driver

(macallan)

_cpuset_create: initialize size argument for sysctl call.
From PR/43837 by Sandy Snaman.

(rmind)

Promote paddr_t to 64 bits for i386 userland (non PAE and PAE).

paddr_t has limited interest in userland, and should only be used by specific
code like kvm(3) (for VA => PA translations). To support non PAE and PAE
kernels simultaneously in kvm(3), promote paddr_t to 64 bits.

No objection on current-users@. See also:

http://mail-index.netbsd.org/current-users/2010/09/07/msg014249.html

Kernel remains untouched by this change. In-kernel 64 bits for paddr_t is for
a future commit (Yes, I do not forget about that :) )

(jym)

Change kvm_pa2off() and kvm_kvatop() prototypes (private to kvm(3)):

-int    _kvm_kvatop(kvm_t *, u_long, u_long *);
-off_t  _kvm_pa2off(kvm_t *, u_long);
+int    _kvm_kvatop(kvm_t *, vaddr_t, paddr_t *);
+off_t  _kvm_pa2off(kvm_t *, paddr_t);

Basically, use vaddr_t for VA and paddr_t for PA. In addition, for variables
representing addresses, use paddr_t or vaddr_t, depending on the context.

For most arches, vaddr_t and paddr_t are equivalent to unsigned long. However,
the change was needed for exotic situations, like i386 PAE, were unsigned long
is not suitable for PA which are 64 bits long. As this required a complete
change of the function prototypes, all arches had to be adapted accordingly.

Core files from before this commit should still work with the new code; I did
not see any direct dependency between core's structure and kvatop/pa2off.

The change was compile tested for all arches, as it impacts all of them.

See also:

http://mail-index.netbsd.org/current-users/2010/09/07/msg014249.html

(jym)

merge bozohttpd 20100920

(mrg)

initial import of bozohttpd 20100920.  the only change missing in here is:

        o  fix dynamic CGI content maps, from rudolf

Status:

Vendor Tag: bozohttpd
Release Tags: bozohttpd-20100920

(mrg)

fix an error in the previous.

(mrg)

fix a serious error in virtual hosting support, noticed by seanb@netbsd,
and disallow ".." as a virtual host name!  also ".".

patch from sean.

(mrg)

- provide lockless interrupt dispatch by eliminating use of LIST(9)
for interrupt handles.  Handles are now managed in variable size arrays.
Establishing a new interrupt causes allocation of a new array, pointer
to which is changed atomically.  Old arrays are allowed to persist for
some time before free, allowing any CPU working with that data
to safely finish using it.
- interrupt events are now managed per-CPU, avoid need for atomic adds.

(cliff)

- rmixl_intrhand_t is valid only if ih_func is non-NULL;
set valid at end of rmixl_vec_establish and
set invalid at start of rmixl_vec_disestablish
to allow lockless check if valid in dispatch.

(cliff)

- .set noreorder up top to avoid instruction reordering
- adopt bugfix suggested by Manuel Boyer for mips/spl.S:
in _splraise and _splsw_splhigh, reload L_CPU in case we were
preempted prior to interrupts being blocked (thanks).

(cliff)

avoid integer overflow that can lead to buffer overflow

(christos)

- remove MULTIPROCESSOR condition for rmixl_cpuinfo_print and
rmixl_cpu_data_print functions, so we can call them
e.g. from ddb in non-MP kernels

(cliff)

deal with infinity on the vax.

(christos)

no more ieee ifdefs

(christos)

get rid of ieee dependencies.

(christos)

cargo cult the x86 changes.

(christos)

shouldn't need ieee.h here.

(christos)

Use the control (not data) interface descriptor when looking at the
control interface endpoints! (from Anon Ymous)

(christos)

Avoid NaN when calculating std-dev, which fixes a crash on VAX.

(ahoka)

Prepare sparc optimized (by 32 bit add-with-carry insns) cpu_in_cksum()
which is reusing in_cksum_internal() in old sparc/in_cksum.c.

~4% improvements of ttcp(1) xfer rates on SS1+ with le(4).

(tsutsui)

Clean up return/error in sdmem_mem_send_scr().

(kiyohara)

Fix SCR data.

(kiyohara)

Support MMC 4,8-bit mode.
  It tested only 4bit on Marvell Sheevaplug.

(kiyohara)

Sync with POSTREAD before memcpy.

(kiyohara)

Clear SMF_CARD_PRESENT, if failed to attach.

(kiyohara)

Print product name.  like sd(4), wd(4) and etc...

(kiyohara)

- performance improvement for read, readdir and write. Now we use
SOCK_DGRAM, we can send many pages at once without hitting any bug

- when creating a file, it is open for FUSE, but not for the kernel.
If the kernel does not do a subsequent open, we have a leak. We fight
against this by trying to close such file that the kernel left unopen
for some time.

- some code refactoring to make message exchange debug easier (more to come)

(manu)

Check disk bounds, in RAW_PART.

(kiyohara)

- Add debug santity checks
- Do not request peer creds anymore once we have them.

(manu)

Multiple channel devices *should* be working. Remove the debug code
that was avoiding finding out.

(mjacob)

use '/etc/rc.d/apmd onestatus' rather than '/etc/rc.d/apmd status',
so we DTRT if apmd is set to NO in rc.conf

(abs)

Make pci(4) device nodes root:wheel 0640 by default.
Mortals do not need to be able to generate PCI Configuration Space
read transactions, which are not entirely without side effect, as
reported in PR#16300.

(jakllsch)

KNF

(jym)

AUDIO_WSEEK uses a u_long, not a int.

(mrg)

add (runtime untested) support for AUDIO_WSEEK.

(mrg)

add support for the netbsd 5.0 versions of the VND* ioctls.

(mrg)

actually, put the old definitions back into vndvar.h, under _KERNEL,
as netbsd32 wants access to them.

(mrg)

support VND* ioctls.

(mrg)

fix the vnd_osize changes on 32 bit platforms with 64 bit alignment for
64 bit integers (eg, sparc).  the problem was that the new 64 bit
element on the end was used for the offsetof() (aka size) for the old
structure, but this includes the padding required, thus the ioctl number
was set wrongly.

move all the supporting code for this inside COMPAT_50, with some renaming
to suit, and kill all the external definitions related to it.

tested on i386, amd64 and sparc.

(mrg)

add support for COMPAT_50 ioctls.  struct vnd_user has a dev_t component
which grew since netbsd 5.0 (hi christos!)

fix a few issues/problems:
- the COMPAT_30 code wasn't used since opt_compat_netbsd.h wasn't included
- move 'struct vnd_ouser' (for COMPAT_30) into vnd.c itself, and call it
  'struct vnd_user30'
- same for VNDIOOCGET -> VNDIOCGET30

now 'vnconfig -l' works on -current with a netbsd-5 binary, using i386.

XXX: there is still a potential problem with the old VNDIOOCSET and
VNDIOOCCLR macros on some platforms like sparc.  there is padding
between the old vnd_osize member and the new vnd_size member on
platforms that want 64 bit values 64 bit aligned, but are 32 bit
otherwise (like sparc.)  64 bit systems already end up with this
member 64 bit aligned, and should be fine.

this most likely results in the old ioctl numbers being wrong and
the code won't match/run ever (ENOTTY.)

(mrg)

enable COMPAT_30 and COMPAT_50 in the module.

(mrg)

- add m68k/conf/std.m68k, which includes common standard options for m68k,
  options EXEC_ELF32, EXEC_SCRIPTS, and CPU_IN_CKSUM for md cpu_in_cksum.c
- make all m68k ports include common arch/m68k/conf/std.m68k
  from MD std.${MACHINE}

(tsutsui)

Ansify and KNF all functions within kvm(3). No objection on current-users@.

Used as ground for u_long => vaddr_t/paddr_t replacements in kvm(3)
private functions.

Compile tested for all arches. See also:

http://mail-index.netbsd.org/current-users/2010/09/07/msg014249.html

(jym)

Remove MD m68k/in_cksum.c which is obsolete since cpu_in_cksum refactoring
in 2008 and now superseded by m68k/cpu_in_cksum.c.

(tsutsui)

Add m68k MD cpu_in_cksum.c.
Based on sys/netinet/cpu_in_cksum.c but just modified
to use traditional MD m68k/oc_cksum.s.

Makes ~5% improvements of ttcp(1) xfer rates on ATARI TT030 + SMC_TT.

(tsutsui)

Fix wrong checksum calculations of 32 bit unaligned two byte payloads.
Analyzed in PR port-mips/43882, but applied slightly different patch.
Also put some changes for readability.

Note netbsd-5 doesn't use this MD version but netbsd-4 needs a pullup.
(though it's unclear if it's really faster even on modern aggressive gcc4)

(tsutsui)

AMD publication 25759 rev 3.69 says that DisIOReqLock in NB_CFG is "bit 3".
They probably mean "bit 3" and not "the third bit" (or bit 2).
This change should prevent superfluous warnings of errata 89.

(jakllsch)

don't compile long double support for the havenots.

(christos)

fix an obvious botch which made eg conjl(3) a no-op

(drochner)

unbreak readline history.

(christos)

NPF checkpoint:
- Add support for bi-directional NAT and redirection / port forwarding.
- Finish filtering on ICMP type/code and add filtering on TCP flags.
- Add support for TCP reset (RST) or ICMP destination unreachable on block.
- Fix a bunch of bugs; misc cleanup.

(rmind)

- remove a wrong _DIAGASSERT
- update comments
- whitespace

(yamt)

Use standard ELF types for ELF code, instead of paddr_t. paddr_t should
only be used for low level code, like virtual memory internals.

(jym)

Use PRIxVADDR to print vaddr_t elements. Wrap lines.

(jym)