Module name: xenocara
Changes by: matthieu@cvs.openbsd.org 2014/08/20 14:23:42

Modified files:
driver/xf86-video-wsudl: configure.ac

Log message:
Remove duplicate AC_SUBST()

Module name: src
Changes by: bluhm@cvs.openbsd.org 2014/08/20 14:52:14

Log message:
Run syslogd regressions tests. As only one syslogd can run per
machine, each test kills any syslogd first. At the end the system's
syslogd gets restarted.
The test framework runs a client, and a server, and a syslogd. The
messages are passed via the log socket or via UDP from the client
to syslogd. From there UDP transport is used to reach the server.
All processes write log files where the message has to show up.
The test arguments are kept in the args-*.pl files.
The content of a log file, the data sent to a pipe process and what
the server received are checked. The invocation of the sendsyslog(2)
syscall is checked with ktrace, the open file descriptors of syslogd
are checked with fstat.

Status:

Vendor Tag: bluhm
Release Tags: bluhm_20140820

N src/regress/usr.sbin/syslogd/Proc.pm
N src/regress/usr.sbin/syslogd/README
N src/regress/usr.sbin/syslogd/LICENSE
N src/regress/usr.sbin/syslogd/Syslogd.pm
N src/regress/usr.sbin/syslogd/Makefile
N src/regress/usr.sbin/syslogd/args-default.pl
N src/regress/usr.sbin/syslogd/Client.pm
N src/regress/usr.sbin/syslogd/Server.pm
N src/regress/usr.sbin/syslogd/funcs.pl
N src/regress/usr.sbin/syslogd/syslogd.pl
N src/regress/usr.sbin/syslogd/args-client-udp.pl
N src/regress/usr.sbin/syslogd/args-client-udp4.pl
N src/regress/usr.sbin/syslogd/args-client-udp-nodns.pl
N src/regress/usr.sbin/syslogd/args-client-udp4-nodns.pl
N src/regress/usr.sbin/syslogd/args-sendsyslog.pl
N src/regress/usr.sbin/syslogd/args-socket.pl

No conflicts created by this import

Module name: src
Changes by: bluhm@cvs.openbsd.org 2014/08/20 14:54:27

Modified files:
regress/usr.sbin: Makefile

Log message:
Link ospfd and syslogd regression tests to the build.

Module name: src
Changes by: schwarze@cvs.openbsd.org 2014/08/20 16:13:24

Modified files:
share/man/man8 : rc.conf.8

Log message:
Move the examples to the EXAMPLES section and add some actual
reference manual style documentation to the DESCRIPTION.
Feedback and OK rpe@ and ajacoutot@, also using feedback from deraadt@.

Module name: src
Changes by: dlg@cvs.openbsd.org 2014/08/20 17:56:57

Modified files:
sys/dev/pci : if_nfe.c

Log message:
after allocating an mbuf and cluster you still need to init the length
fields.

found by steven roberts, who also tested this fix for me

> CVSROOT: /cvs
> Module name: src
> Changes by: bluhm@cvs.openbsd.org 2014/08/20 14:10:17
>
> Modified files:
> usr.sbin/syslogd: privsep.c syslogd.c syslogd.h
>
> Log message:
> Replace gethostbyaddr(3) with getnameinfo(3). Remove the sigprocmask()
> that was necessary for gethostbyaddr() because the latter is not
> signal safe. Change the return code semantics of priv_getnameinfo()
> to match getnameinfo(3).
> input and OK jca@

The man page for getnameinfo says getnameinfo is not thread safe,
which almost guarantees it is not signal safe either.

Module name: src
Changes by: bluhm@cvs.openbsd.org 2014/08/20 18:04:58

Modified files:
usr.sbin/syslogd: privsep.c syslogd.c

Log message:
Parse loghost in a separate function. Allow [] around hostname,
needed for IPv6 addresses. Print full loghost specifier in case
of error or debug. Make string sizes more precise.
input henning@; input and OK deraadt@

> On Wed, Aug 20, 2014 at 14:10, Alexander Bluhm wrote:
> > CVSROOT: /cvs
> > Module name: src
> > Changes by: bluhm@cvs.openbsd.org 2014/08/20 14:10:17
> >
> > Modified files:
> > usr.sbin/syslogd: privsep.c syslogd.c syslogd.h
> >
> > Log message:
> > Replace gethostbyaddr(3) with getnameinfo(3). Remove the sigprocmask()
> > that was necessary for gethostbyaddr() because the latter is not
> > signal safe. Change the return code semantics of priv_getnameinfo()
> > to match getnameinfo(3).
> > input and OK jca@
>
> The man page for getnameinfo says getnameinfo is not thread safe,
> which almost guarantees it is not signal safe either.

That is right, almost nothing is signal safe. When sigprocmask(2)
was added, there was a gethostbyaddr(3) in the code to protect.
Now getnameinfo(3) is running in a separate prevsep process. All
signal handlers set integer flags or do a simple waitpid(2) as
calling privsep functions is not signal safe either. So I think
my change is safe.

bluhm

Module name: src
Changes by: doug@cvs.openbsd.org 2014/08/20 19:08:52

Modified files:
usr.bin/ssh : moduli.c ssh-keygen.c

Log message:
Free resources on error in mkstemp and fdopen

ok djm@

Module name: src
Changes by: ajacoutot@cvs.openbsd.org 2014/08/21 02:54:58

Modified files:
usr.sbin/rcctl : rcctl.sh

Log message:
If a service is disabled its flags are unconditionally set to "NO", so
there's not need to display "enabled" or "disabled" when calling "status",
we have the information already.

Module name: src
Changes by: ajacoutot@cvs.openbsd.org 2014/08/21 04:06:14

Modified files:
etc/rc.d : rc.subr

Log message:
Drop sanitation, _rc_parse_conf does this for us already.

ok schwarze@

Module name: src
Changes by: mpi@cvs.openbsd.org 2014/08/21 04:07:07

Modified files:
sys/net : route.c

Log message:
If an ifa has as NULL ifp pointer then is should be considered as
invalid. When such thing happens, it means that the address is no
longer configured on the system but still referenced by some routes.
So do not return such ifa in ifa_ifwithroute().

Fix a panic reported by Pierre Bardou.

ok mikeb@, henning@

Module name: src
Changes by: schwarze@cvs.openbsd.org 2014/08/21 04:15:11

Modified files:
usr.sbin/rcctl : rcctl.8

Log message:
Now that rc.conf(8) defines what a base system service, a base
system daemon, and a package daemon is, refer to it instead of
duplicating information. While here, clean up wording.
OK ajacoutot@

Module name: src
Changes by: mpi@cvs.openbsd.org 2014/08/21 04:23:47

Modified files:
usr.sbin/arp : arp.8 arp.c

Log message:
Change the output of arp(8) to match what ndp(8) does and include the
expire timer.

This will makes it easier to add further information in a coherent way
between these tools for local and broadcast entries.

This new view displays either symbolic names (by default) or numerical
addresses (with "-n") for hosts but not both at the same time, just
like ndp(8), route(8) or netstat(1) do.

ok henning@

Module name: src
Changes by: mpi@cvs.openbsd.org 2014/08/21 05:54:00

Modified files:
sys/netinet6 : in6_pcb.c

Log message:
Misleading comments about splnet().

Module name: src
Changes by: ajacoutot@cvs.openbsd.org 2014/08/21 06:47:22

Modified files:
usr.sbin/rcctl : rcctl.8 rcctl.sh

Log message:
Make "rcctl status" output match rc.conf(8) format.

ok jasper@

Module name: src
Changes by: schwarze@cvs.openbsd.org 2014/08/21 06:56:24

Modified files:
usr.bin/mandoc : mdoc_html.c mdoc_macro.c mdoc_man.c mdoc_term.c
regress/usr.bin/mandoc/mdoc/Ar: punct.in punct.out_ascii
regress/usr.bin/mandoc/mdoc/Fl: Makefile noarg.in
noarg.out_ascii punct.in
punct.out_ascii
regress/usr.bin/mandoc/mdoc/Nm: Makefile
regress/usr.bin/mandoc/mdoc/Pa: Makefile
Added files:
regress/usr.bin/mandoc/mdoc/Nm: punct.in punct.out_ascii
regress/usr.bin/mandoc/mdoc/Pa: punct.in punct.out_ascii

Log message:
Right after .Fl, a middle delimiter triggers an empty scope,
just like a closing delimiter. This didn't work in groff-1.15,
but it now works in groff-1.22.

After being closed by delimiters, .Nm scopes do not reopen.

Do not suppress white space after .Fl if the next node is a text node
on the same input line; that can happen for middle delimiters.

Fixing an issue reported by jmc@.

Module name: src
Changes by: mpi@cvs.openbsd.org 2014/08/21 08:24:08

Modified files:
sys/arch/alpha/alpha: in_cksum.c
sys/arch/alpha/stand/netboot: if_prom.c
sys/arch/amd64/stand/libsa: pxe.c pxe_net.c
sys/arch/arm/arm: genassym.cf
sys/arch/armv7/imx: imxenet.c
sys/arch/armv7/sunxi: sxie.c
sys/arch/aviion/stand/boot: conf.c dev_net.c if_le.c
sys/arch/i386/stand/libsa: pxe.c pxe_net.c
sys/arch/landisk/stand/boot: getsecs.c
sys/arch/luna88k/stand/boot: conf.c if_le.c
sys/arch/m88k/m88k: in_cksum.c
sys/arch/macppc/stand: net.c netif_of.c
sys/arch/sparc/stand/common: net.c netif_sun.c
sys/arch/sparc64/sparc64: in4_cksum.c
sys/arch/sparc64/stand/ofwboot: net.c netif_of.c
sys/arch/vax/if: if_de.c
sys/arch/vax/stand/boot: if_de.c if_le.c if_qe.c if_ze.c netio.c
sys/arch/vax/vax: in4_cksum.c

Log message:
Kill the remaining <netinet/in_systm.h> inclusion!

Module name: src
Changes by: jsg@cvs.openbsd.org 2014/08/21 08:26:16

Modified files:
usr.sbin/rcctl : rcctl.sh

Log message:
add $OpenBSD$

Module name: src
Changes by: brad@cvs.openbsd.org 2014/08/21 08:30:21

Modified files:
sys/dev/pci : if_bge.c

Log message:
Fix a copy and pasto with the standard ring setup with calling if_rxr_init()
to use BGE_STD_RX_RING_CNT instead of BGE_JUMBO_RX_RING_CNT.

ok dlg@

Module name: src
Changes by: mpi@cvs.openbsd.org 2014/08/21 08:52:56

Modified files:
sys/dev/usb : uaudio.c ukbd.c umass.c ums.c

Log message:
Now that DVACT_DEACTIVATE is propagated to the children of a device when
a driver does not implement a specific *_activate() handler and that our
USB stack sets the dying flag before detaching a device, these specific
handlers can die.

Module name: src
Changes by: mikeb@cvs.openbsd.org 2014/08/21 09:09:27

Modified files:
sbin/pfctl : parse.y

Log message:
deny "once" flags for match rules; ok henning

Module name: src
Changes by: schwarze@cvs.openbsd.org 2014/08/21 10:03:50

Modified files:
usr.bin/mandoc : cgi.c

Log message:
limit CGI process execution time to make REDoS attacks less effective;
attack surface pointed out by Sebastien Marie

Module name: src
Changes by: jca@cvs.openbsd.org 2014/08/21 10:46:48

Modified files:
usr.bin/ftp : fetch.c

Log message:
Fix double free. ok guenther@

Module name: src
Changes by: ajacoutot@cvs.openbsd.org 2014/08/21 10:50:11

Modified files:
usr.sbin/sysmerge: sysmerge.sh

Log message:
Unbreak when "SRCDIR=."

reported by phessler@

Module name: src
Changes by: bluhm@cvs.openbsd.org 2014/08/21 11:00:34

Modified files:
usr.sbin/syslogd: privsep.c syslogd.c

Log message:
Send and receive UDP syslog packets on the IPv6 socket.
OK henning@

Module name: src
Changes by: bluhm@cvs.openbsd.org 2014/08/21 11:16:37

Modified files:
usr.sbin/syslogd: syslog.conf.5

Log message:
Document square brackets for IPv6 addresses. From FreeBSD.
OK logan@ henning@

Module name: src
Changes by: chrisz@cvs.openbsd.org 2014/08/21 13:23:11

Modified files:
usr.sbin/httpd : httpd.h server_fcgi.c server_file.c
server_http.c

Log message:
Add Last-Modified: HTTP header.

OK reyk@

Module name: src
Changes by: schwarze@cvs.openbsd.org 2014/08/21 14:27:03

Modified files:
usr.bin/mandoc : mansearch.c

Log message:
Bugfix: make whatis(1) case-insensitive again.
The traditional whatis(1) was case-insensitve and it's still documented
that way, that but apparently got broken with or after the switch.

Module name: src
Changes by: kspillner@cvs.openbsd.org 2014/08/21 15:49:37

Modified files:
sys/dev/acpi : acpidock.c

Log message:
Set the sensor's status when docking and undocking, not just its value.
With this change sensorsd(8) now correctly detects state changes when
docking and undocking.

ok mlarkin@

Module name: src
Changes by: deraadt@cvs.openbsd.org 2014/08/21 19:28:19

Modified files:
distrib/sets/lists/man: mi

Log message:
sync

Module name: src
Changes by: jsg@cvs.openbsd.org 2014/08/21 22:36:16

Modified files:
usr.sbin/npppd/npppd: npppd_auth.c

Log message:
remove an uneeded test
ok yasuoka@

Module name: src
Changes by: ajacoutot@cvs.openbsd.org 2014/08/21 23:47:08

Modified files:
usr.sbin : Makefile

Log message:
Enable rcctl(8).

prodded by deraadt@

Module name: src
Changes by: deraadt@cvs.openbsd.org 2014/08/22 01:41:48

Modified files:
distrib/sets/lists/base: mi
distrib/sets/lists/man: mi

Log message:
sync

Module name: src
Changes by: doug@cvs.openbsd.org 2014/08/22 01:59:52

Modified files:
sbin/isakmpd : ike_phase_1.c

Log message:
fix memory leak in isakmpd

ok gerhard@ (also corrected first version)

Module name: src
Changes by: jsg@cvs.openbsd.org 2014/08/22 02:10:38

Modified files:
etc/rc.d : nfsd

Log message:
Set the default nfsd flags to "-tun 4". Matches the comment
in rc.conf and the behaviour of the backwards compatibility
code in rc.subr for nfs_server=YES.
ok ajacoutot@

Module name: src
Changes by: schwarze@cvs.openbsd.org 2014/08/22 09:49:45

Modified files:
usr.bin/mandoc : man.cgi.8

Log message:
typo; noticed by jmc@ some time ago

Module name: src
Changes by: bluhm@cvs.openbsd.org 2014/08/22 10:14:11

Modified files:
usr.sbin/syslogd: syslogd.c

Log message:
Instead of getting a nasty error message from privsep
"syslogd: priv_getaddrinfo: overflow attempt in hostname"
check the host and port length when parsing the config.
OK henning@

Module name: src
Changes by: tedu@cvs.openbsd.org 2014/08/22 10:29:27

Removed files:
usr.bin/signify: ge25519.c

Log message:
don't need this file, only the mod version

Module name: xenocara
Changes by: okan@cvs.openbsd.org 2014/08/22 13:04:01

Modified files:
app/cwm : group.c mousefunc.c search.c xutil.c

Log message:
Fix nogroup regression, where nogroup became an actual group - the
symantics between cwm groups and ewmh got in the way. Ensure a client
that wants to be in nogroup stays in nogroup (thus stays in view), even
when (re)reading NET_WM_DESKTOP. Paritially reverts patchset 644
(2014-02-07 13:09 PST) which deals with a NULL cc->group. All to be
revisited when NET_WM_STATE_STICKY hits cwm.

Reported by many; testing and ok phessler.

Module name: src
Changes by: deraadt@cvs.openbsd.org 2014/08/22 13:19:25

Modified files:
etc : Makefile changelist group master.passwd rc
rc.conf
etc/mail : aliases
etc/mtree : 4.4BSD.dist
Removed files:
etc/bind : db.localhost db.loopback db.loopback6.arpa
named-dual.conf named-simple.conf root.hint
etc/rc.d : named
etc/systrace : usr_sbin_named

Log message:
disable use of bind in base; in the base use nsd/unbound instead.
a proper & complete bind port will show up.
discussed with many for years

Module name: src
Changes by: brad@cvs.openbsd.org 2014/08/22 13:28:25

Modified files:
usr.sbin/bind : Makefile.bsd-wrapper Makefile.in configure
configure.in
usr.sbin/bind/bin: Makefile.in

Log message:
Strip the BIND code down to just building and installing dig, host and nslookup.

Module name: src
Changes by: deraadt@cvs.openbsd.org 2014/08/22 14:03:56

Modified files:
sys/arch/armish/conf: RAMDISK
sys/arch/armv7/conf: RAMDISK-IMX RAMDISK-OMAP RAMDISK-SUNXI
sys/arch/landisk/conf: RAMDISK
sys/arch/sgi/conf: GENERIC-IP22 GENERIC-IP26 GENERIC-IP27
GENERIC-IP28 GENERIC-IP30 GENERIC-IP32
sys/arch/zaurus/conf: RAMDISK

Log message:
pseudo-device crypto was accidentally left for some architectures.

Module name: www
Changes by: sthen@cvs.openbsd.org 2014/08/22 14:53:44

Modified files:
faq : current.html

Log message:
bind->ports

Module name: src
Changes by: sthen@cvs.openbsd.org 2014/08/22 16:14:53

Modified files:
usr.sbin/arp : arp.c

Log message:
print leading 0's in MAC addresses again, ok deraadt

> CVSROOT: /cvs
> Module name: src
> Changes by: sthen@cvs.openbsd.org 2014/08/22 16:14:53
>
> Modified files:
> usr.sbin/arp : arp.c
>
> Log message:
> print leading 0's in MAC addresses again, ok deraadt

and mpi

Module name: src
Changes by: sthen@cvs.openbsd.org 2014/08/22 16:37:32

Modified files:
usr.sbin/ndp : ndp.c

Log message:
switch ndp to display MAC addresses in 00:00:00:00:00:00 format, ok deraadt@

Module name: src
Changes by: krw@cvs.openbsd.org 2014/08/22 17:05:15

Modified files:
lib/libc/sys : stat.2
sys/sys : _types.h stat.h types.h

Log message:
POSIX/FreeBSD/NetBSD/Linux/etc agree -- blk[size|cnt]_t is the bomb.

So add the types blksize_t (a.k.a. int32_t) and blkcnt_t (a.k.a.
int64_t). Use blkcnt_t in stat.h since the base type (int64_t) does
not change. blksize_t in stat.h will follow after the tree is audited
for signed issues, since the base type will change from u_int32_t
to a POSIX compliant int32_t.

Guidance and feedback from guenther@

ok millert@

Module name: www
Changes by: brad@cvs.openbsd.org 2014/08/22 17:52:00

Modified files:
faq : current.html

Log message:
Also clear out the man pages for BIND.

Module name: src
Changes by: pelikan@cvs.openbsd.org 2014/08/22 18:11:03

Modified files:
regress/sbin/pfctl: Makefile
sbin/pfctl : parse.y pfctl.c pfctl_parser.h
Added files:
regress/sbin/pfctl: pfail54.in pfail54.ok

Log message:
when you specify queues in a rule, make sure they have been defined.

DIOCADDRULE EBUSY turns into an error message that pfctl -n catches.
DIOCXCOMMIT EINVAL after the kernel rejected the rules was reported
to occur, possibly from hfsc.c: this should be fixed as well.

ok henning mikeb sthen