Add new primaries -asince, -csince, and -since to compare file's
attributes against a user-specified timestamp (rather than the
attributes of a reference file).

Update the parse routines so they have access to the name of the
option being parsed.  This enables accurate error reporting for
"aliases" of primaries.

Now that aliases work, introduce some aliases for consistency with
Gnu findutils.

(pgoyette)

Use .Ic markup for consistency with the rest of the document

(pgoyette)

Further clarification of the treatment of numeric user and group names.

(pgoyette)

Extend the numeric handling for uids with -user to gids with -group, and
document it. Leftover bit of PR 46158.

(dholland)

Mark hp300/crash.{0,8,html} obsolete.
Discussed on: http://mail-index.netbsd.org/port-hp300/2016/06/07/msg000106.html

(abhinav)

Remove crash.8 for hp300 as there is already a man page for crash(8).
Discussed on port-hp300: http://mail-index.netbsd.org/port-hp300/2016/06/07/msg000106.html

(abhinav)

Remove KASSERTs that don't make sense.

(skrll)

Remove KASSERTs that don't make sense.

(skrll)

Add libutil to the DPADD variable.

(abhinav)

Fix a buffer overflow when reading from files containing extra long lines.
Instead of using a statically allocated buffer, manage the buffer at run-time
and reallocate as needed.

It was dumping core for following two cases:
man evrpc | ul
man xdm | ul

While there, also remove __P macro from function prototypes.

Ok from Christos.

(abhinav)

- fix setting of hdr_lba_alt which was broken in revision 1.10 on 2015/12/2
- switch to using gpt_last
- PR/51230 -- recreate the PMBR if it was lost

(jnemeth)

Fix m_reset_rcvif

m_reset_rcvif should zero whole m_pkthdr._rcvif union, which is
the original behavior.

(ozaki-r)

Don't refer getfsstat() using .Xr as we don't have man page for it anymore.
Ok from wiz@

(abhinav)

- add machdep.xen.version sysctl to easily get hypervisor version
- move machdep.xen_timepush_ticks to machdep.xen.timepush_ticks to
  consolidate all Xen related sysctls under machdep.xen

(jnemeth)

Oops... use the ipq we created.

(skrll)

Fix locking botch

(skrll)

First pass at making this driver and attachments NET_MPSAFE aware.

I've not tested this yet.

(skrll)

note nvmectl

(nonaka)

Change -{min,max}depth argument name from n to depth so that the
earlier statement concerning n being prefaced by a plus or minus.
(These exceptions to the +/- rule noted by kre@)

While here, fix markup error.  .Em for emphasis (stress) rather
than .Ar  :)

(pgoyette)

Make internal functions static to avoid conflicts with arc_* from zfs.

(christos)

Fix the out-of-memory behavior so the message doesn't disappear under
endwin(). Related to PR 3126.

(dholland)

small progress.

(christos)

fix compilation and getdents.

(christos)

Update to match change to dec_prom.h.

(dholland)

Sort options in SYNOPSIS too.

(wiz)

PR/51234: Onno van der Linden: syslogd sometimes incorrectly handles iso to
bsd time conversion

(christos)

Fix typo in the date.
While there, remove .Pp from inside of the list (mandoc -Tlint warns about it).

(abhinav)

More fixed from Rin Okuyama
fix waitpid, getdents, explain ptrace, cleanup redefinitions

(christos)

PR 51200 gets in libsa considered harmful: use kgets

(dholland)

PR 51200 gets in libsa considered harmful: use kgets

(dholland)

PR 51200 gets in libsa considered harmful: use kgets

(dholland)

Revert this one; I can't tell for sure if it's using libsa gets or its
own firmware one in ../libdos. If the latter, switching and bringing
in the libsa one (a) might not work and (b) might make the bootloader
overflow size limits.

PR 51200.

(dholland)

PR 51200 gets in libsa considered harmful: use kgets

(dholland)

PR 51200 gets in libsa considered harmful: change own gets() to kgets()
and use it.

(dholland)

PR 51200 gets in libsa considered harmful: use kgets

(dholland)

PR 51200 gets in libsa considered harmful: use kgets

(dholland)

PR 51200 gets in libsa considered harmful: use kgets

(dholland)

PR 51200 gets in libsa considered harmful: use kgets

(dholland)

PR 51200 gets in libsa considered harmful: use kgets

(dholland)

PR 51200 gets in libsa considered harmful: use kgets

(dholland)

PR 51200 gets in libsa considered harmful: use kgets

(dholland)

PR 51200 gets in libsa considered harmful: use kgets

(dholland)

gets -> unsafe_gets; nothing uses this but let's not start by accident.
(related to PR 51200)

(dholland)

PR 51200 gets in libsa considered harmful: use kgets

(dholland)

PR 51200 gets in libsa considered harmful: use kgets

(dholland)

PR 51200 gets in libsa considered harmful: use kgets

(dholland)

PR 51200 gets in libsa considered harmful: use kgets

(dholland)

PR 51200 gets in libsa considered harmful: use kgets

(dholland)

PR 51200 gets in libsa considered harmful: use kgets

(dholland)

PR 51200 gets in libsa considered harmful: use kgets

(dholland)

PR 51200 gets in libsa considered harmful: use kgets

(dholland)

PR 51200 gets in libsa considered harmful: use kgets

(dholland)

PR 51200 gets in libsa considered harmful: use kgets

(dholland)

PR 51200 gets in libsa considered harmful: use kgets

(dholland)

PR 51200 gets in libsa considered harmful: use kgets

(dholland)

PR 51200 gets in libsa considered harmful: use kgets

(dholland)

Change "tgets_s" back to "tgets". We obviously want to keep the bounds
argument, but there's no need to mess with the function name.

(dholland)

gets_s -> kgets.

The exact name isn't that important; but it is important not to use
"gets_s" and thereby allow anyone to falsely get the impression we're
implementing Annex K. We aren't.

ok core.

(dholland)

Sort options. Use Nx. Uppercase start of sentence.

(wiz)

use the package name, not the set name, within the +CONTENTS file

(agc)

Add set2pkg to othersrc - a small script to convert an installed set to a
pkgsrc package.

set2pkg allows us to convert from a set-based installation to being
able to use pkgsrc tools to maintain the base system.  This has
nothing to do with syspkgs, which have been around for over 15 years
but have gained little, if any, traction.

Firstly, you can convert a single set, any number of sets, or all sets
using set2pkg:

% sudo set2pkg -a
Converting base to netbsd-base-20160610
Converting comp to netbsd-comp-20160610
Converting etc to netbsd-etc-20160610
Converting games to netbsd-games-20160610
Converting man to netbsd-man-20160610
Converting misc to netbsd-misc-20160610
Converting modules to netbsd-modules-20160610
Converting tests to netbsd-tests-20160610
Converting text to netbsd-text-20160610
Converting xbase to netbsd-xbase-20160610
Converting xcomp to netbsd-xcomp-20160610
Converting xetc to netbsd-xetc-20160610
Converting xfont to netbsd-xfont-20160610
Converting xserver to netbsd-xserver-20160610
% pkg_info -I netbsd-\*
netbsd-base-20160610 NetBSD system package for netbsd-base-20160610, created by set2pkg
netbsd-comp-20160610 NetBSD system package for netbsd-comp-20160610, created by set2pkg
netbsd-etc-20160610 NetBSD system package for netbsd-etc-20160610, created by set2pkg
netbsd-games-20160610 NetBSD system package for netbsd-games-20160610, created by set2pkg
netbsd-man-20160610 NetBSD system package for netbsd-man-20160610, created by set2pkg
netbsd-misc-20160610 NetBSD system package for netbsd-misc-20160610, created by set2pkg
netbsd-modules-20160610 NetBSD system package for netbsd-modules-20160610, created by set2pkg
netbsd-tests-20160610 NetBSD system package for netbsd-tests-20160610, created by set2pkg
netbsd-text-20160610 NetBSD system package for netbsd-text-20160610, created by set2pkg
netbsd-xbase-20160610 NetBSD system package for netbsd-xbase-20160610, created by set2pkg
netbsd-xcomp-20160610 NetBSD system package for netbsd-xcomp-20160610, created by set2pkg
netbsd-xetc-20160610 NetBSD system package for netbsd-xetc-20160610, created by set2pkg
netbsd-xfont-20160610 NetBSD system package for netbsd-xfont-20160610, created by set2pkg
netbsd-xserver-20160610 NetBSD system package for netbsd-xserver-20160610, created by set2pkg
%

The date in the package name is taken from the mtime on the /etc/mtree
metadata files which describe the installed sets on a machine.

We can now try them out, check entries, etc:

% pkg_info netbsd-base
Information for netbsd-base-20160610:

*** PACKAGE MAY NOT BE DELETED ***
Comment:
NetBSD system package for netbsd-base-20160610, created by set2pkg

Required by:

Description:
NetBSD system package for netbsd-base-20160610.

This was automatically created from the base set
by the set2pkg utility.

This package may not be deleted, and has been created to allow
better manipulation of systems by using the pkgsrc tools.

More information can be found by browsing the mailing lists at
the homepage.

Hompage:
https://www.NetBSD.org/

*** PACKAGE MAY NOT BE DELETED ***

% pkg_info -qL netbsd-base
/bin/[
/bin/cat
/bin/chio
/bin/chmod
/bin/cp
/bin/cpio
/bin/csh
/bin/date
/bin/dd
/bin/df
/bin/domainname
/bin/echo
/bin/ed
/bin/expr
/bin/hostname
/bin/kill
/bin/ksh
/bin/ln
/bin/ls
/bin/mkdir
/bin/mt
/bin/mv
/bin/pax
/bin/ps
/bin/pwd
/bin/rcmd
/bin/rcp
/bin/rm
/bin/rmdir
/bin/rump.dd
/bin/sh
/bin/sleep
/bin/stty
/bin/sync
/bin/tar
/bin/test
/etc/release
/etc/mtree/set.base
...
% sudo pkg_admin check netbsd-base
Password:
pkg_admin: base: File `//dev/log' is in +CONTENTS but not on filesystem!
.
Checked 4870 files from 1 package.
Done.
%

Alistair Crooks
Fri Jun 10 18:53:12 PDT 2016

(agc)

GSoC 2016: Charles Cui: add SIGPOLL

(christos)

GSoC 2016: Charles Cui: Add timer related macros
    _POSIX_CPUTIME
    _POSIX_THREAD_CPUTIME
    _POSIX_DELAYTIMER_MAX

(christos)

Bump date for previous.

(wiz)

GSoC 2016: Charles Cui: add SEM_NSEMS_MAX

(christos)

Avoid trying to create a tag for locating AMD HyperTransport bridge that will
panic a machine that uses Configuration Mechanism 2.

(jakllsch)

add driver for 'platinum' onboard video found in Power Mac 7200 and possibly
other machines
from s_cole

(macallan)

add WSDISPLAY_TYPE_PLATINUM, for platinumfb

(macallan)

Fix spelling of the month in the date (sent pull request upstream).

(abhinav)

Add callout_stop while clearing stall.  From t-hash.

(skrll)

Add callout_stop while clearing stall.  From t-hash.

(skrll)

Don't forget to destroy sc_cmdbusy_cv.  From t-hash.

(skrll)

Only call sc_vendor_init once. Spotted by t-hash.

(skrll)

Avoid storing a pointer of an interface in a mbuf

Having a pointer of an interface in a mbuf isn't safe if we remove big
kernel locks; an interface object (ifnet) can be destroyed anytime in any
packet processing and accessing such object via a pointer is racy. Instead
we have to get an object from the interface collection (ifindex2ifnet) via
an interface index (if_index) that is stored to a mbuf instead of an
pointer.

The change provides two APIs: m_{get,put}_rcvif_psref that use psref(9)
for sleep-able critical sections and m_{get,put}_rcvif that use
pserialize(9) for other critical sections. The change also adds another
API called m_get_rcvif_NOMPSAFE, that is NOT MP-safe and for transition
moratorium, i.e., it is intended to be used for places where are not
planned to be MP-ified soon.

The change adds some overhead due to psref to performance sensitive paths,
however the overhead is not serious, 2% down at worst.

Proposed on tech-kern and tech-net.

(ozaki-r)

Introduce m_set_rcvif and m_reset_rcvif

The API is used to set (or reset) a received interface of a mbuf.
They are counterpart of m_get_rcvif, which will come in another
commit, hide internal of rcvif operation, and reduce the diff of
the upcoming change.

No functional change.

(ozaki-r)

switch sparc to GCC 5.4 and powerpc to GDB 7.10

(mrg)

Use SOFTINT_SERIAL for the USB stack (the 'S' stands for serial).

This is a step in the right direction towards fixing

kern/50491: unkillable wait in usbd_transfer while using usmsc0 on
raspberry pi 2

kern/49065: ifconfig tun0 ... sequence locks up system

but further analysis of locking of the usb drivers is required.

(skrll)

New sentence, new line. Remove Pp before section header.

(wiz)

Bump date for -r.

XXX: this file has no license.

(wiz)

Sort option descriptions. Use Mt.

(wiz)

Add missing \n (I triggered the systbl message on a qemu virtual machine!)

(pgoyette)

needs <stdbool.h>

(dholland)

re-enable MKCOMPAT for arm eabi platforms, as it now works again.

(mrg)

switch all arm (vs just evbarm) to GDB 7.10.

(mrg)

switch to using xorg-server.old for everyone.

(mrg)

Add some patches which were in an older tree, from January 29 2015:

+ add a -r argument to agrep(1) and document it. This has the same
effect as the -r flag to grep(1) - perform a recursive search
through sub-directories.

+ if a match is found in a "binary" file, then do the same as grep(1)
and just print the message that a match was found in the file, without
attempting to print the match in full.

(agc)

minor clean ups and use ${XORG_SERVER_SUBDIR} where necessary

(mrg)

define XORG_SERVER_SUBDIR to either xorg-server or xorg-server.old.

(mrg)

for the eabi targetted compiler, if not compiling libgcc for eabi ensure
that libgcc doesn't try to define __[CD]TOR_LIST__[].

(mrg)

add missed file for ppc64 gcc5.4 mknative.

(mrg)

switch powerpc64 to GCC 5.

(mrg)

update powerpc64 info

(mrg)

mknative-gcc for powerpc64 and GCC 5.4.

(mrg)

enable checks for no dot syms on netbsd ppc64, fixes build issues
in libutil where the opd() function conflicts with the .opd section.

remove a couple of tainted defines that are handled differently now.

this, with mknative-gcc as well, enables ppc64 world to build again.

(mrg)

Fix typo: degress -> degrees.

(abhinav)

improve on ipfilter rules by Cy Schubert

(christos)

Rename 'a' to set the active PMBR flag to 'A'

(christos)

Support ipfw and ipf from FreeBSD

(christos)

Make "gpt show" add an "(active)" annotation to the PMBR output line
when it has been marked active for legacy (non UEFI) booting.
This was (kind of obviously) intended by the previous commit, but...

(kre)

Fix rendering of biosboot command synopsis.

(kre)

PR/51230: Add the ability to set the active flag in the PMBR.

(christos)

Since sun{2,3} don't have any working X server, don't use .Xr to refer X(1)
Discussed on port-sparc@: http://mail-index.netbsd.org/port-sparc/2016/06/07/msg002002.html
Ok from wiz@

(abhinav)

There is no X(1) man page, change the references to Xorg(1).
Discussed on port-sparc@: http://mail-index.netbsd.org/port-sparc/2016/06/07/msg002002.html
Ok from wiz@

(abhinav)

PR kern/32830 (from 10 years ago, pre NetBSD 4...)
Reference correct include files needed for the ioctl calls mentioned.

(kre)

Fix syscall argument passing from Ryn Okuyama (with minor changes; hope I did
not make it worse :-)

(christos)

Disable glibc workaround for non linux. Rin Okuyama

(christos)

Fix typo: processess -> processes

(abhinav)

add netbsd padding to 'struct addrinfo' copy.

eeew, this is really horrible

(mrg)

mknative-gcc for sparc64 and GCC 5.4 (and a bit of missed vax, too)

(mrg)

mknative-gcc for vax and GCC 5.4.

(mrg)

apply -O1 more liberally for vax.  fixes build.

(mrg)

Add a small lib and utility program to convert a basic regular expression into
an extended one. For more information on regexp specifications, please see
re_format(7).

The library implements a single routine called regextend(3), which implements
a simple codec-style transformation from basic to extended regular expression.
This can be useful for regular expression implementations which do not
implement basic regular expressions.

(agc)

Sync register number for curlwp with the kernel

(martin)

Open the rx interrupt pipe in umct_open when ucom is ready.  Also, tidyup
in umct_close.

(skrll)

Tweak debug

(skrll)

Fix indent

(skrll)

Update sets lists for new ipl (aka ipfilter) module.

(pgoyette)

New module for ipl (aka ipfilter).

(pgoyette)

Enable building of ipfilter code as a separately-loaded module.

(pgoyette)

regen (minor version updates)

(mrg)

Mark phone so we rebuild every time.

(christos)

Don't run .BEGIN if we are cleaning or making obj.

(christos)

Avoid most of the .CURDIR hackery by letting make locate the files.

(christos)

return something

(christos)

allow earm* platforms to *try* to enable MKCOMPAT=yes, since it only
is a build failure, not an entirely unavailable feature.

(mrg)

make sure the build dist files are build before we try to use them.

(mrg)

fix variable name

(christos)

ignore EACCES

(christos)

sort.

(mrg)

Revert previous (rev. 1.98).  It breaks at least one user's card.

PR kern/44433

(pgoyette)

don't give NetBSD.dist.tmp any inputs, as then it won't rebuild ever
unless they change, and we want to rebuild based upon what inputs are
present, not when they were last touched.

this fixes update builds that switch options that change the dirlist
like MKX11 or MKCOMPAT, restoring a portion of rev 1.14.

note that some opertions like turning off MKX11=yes will also require
a fresh DESTDIR, in addition to this fix.  there may be more issues
remaining, but i am now able to enable MKX11=yes successfully without
any other change.

(mrg)

Grr, gcc-5.4 creates DW_AT_typedef without DW_AT_type for HARD_REG_SET!

<1><c26b>: Abbrev Number: 4 (DW_TAG_typedef)
    <c26c>  DW_AT_name        : (indirect string, offset: 0x16e30): HARD_REG_ELT_TYPE
    <c270>  DW_AT_decl_file  : 57
    <c271>  DW_AT_decl_line  : 43
    <c272>  DW_AT_type        : <0x70>
<1><c276>: Abbrev Number: 102 (DW_TAG_typedef)
    <c277>  DW_AT_name        : (indirect string, offset: 0x2f954): HARD_REG_SET
    <c27b>  DW_AT_decl_file  : 57
    <c27c>  DW_AT_decl_line  : 54

(christos)

switch all arm platforms to GCC 5, not just evbarm.

(mrg)

Fix typo. From Kurt Jaeger via Brooks Davis and FreeBSD.

(wiz)

.Fx should only have version numbers as argument.

(wiz)

Re-sort

(pgoyette)

Bump date for previous.

(wiz)

Sync STANDARDS of iconv(3) with current reality

GNU libiconv is no longer incompatible with POSIX since 2007.

Solaris 11 moved to POSIX.
Illumos moved to POSIX http://illumos.org/man/3c/iconv

New software is aligned to POSIX not to the old mistake in the standard.

(kamil)

Correct misspellings in a few Thai airport names

(kre)

delete generated file

(christos)

more -Wno-stack-protector

(christos)

elide stack protector warning

(christos)

Feeding uninitialised garbage to the hypervisor is likely to be a bad idea.

(jnemeth)

stack protector

(christos)

fix variable length stack allocation from upstream

(christos)

Port to linux

(christos)

port to linux

(christos)

Port to linux

(christos)

compile on linux

(christos)

portability macros.

(christos)

Quick+Dirty port to linux

(christos)

elide stack protector warning

(christos)

sprinke -Wno-stack-protector

(christos)

BER is not an operational airport

(mbalmer)

add bits/uniform_int_dist.h for gcc=53 (*)

(*) gcc=53 now means gcc 5.4, and is hard to fix until we remove gcc=48
    support and can go to eg, gcc=5, gcc=6, etc.  the same applies for
    the HAVE_GCC variable in the build system.

(mrg)