[owc] Weekly src changes ending 2006-11-05
OWC
auto at squish.net
Mon Nov 6 07:01:34 GMT 2006
OpenBSD src changes summary for 2006-10-29 to 2006-11-05 inclusive
==================================================================
bin/dd bin/df
bin/ksh bin/ps
distrib/landisk distrib/sets
etc/Makefile etc/changelist
etc/ftpusers etc/group
etc/mail/aliases etc/master.passwd
etc/mtree/special etc/rc
etc/rc.conf etc/security
games/atc games/backgammon
games/mille games/quiz
games/sail games/trek
gnu/usr.bin/gcc kerberosV
lib/libc lib/libevent
lib/libssl libexec/spamd
libexec/spamlogd regress/sbin
sbin/ipsecctl sbin/isakmpd
sbin/mount_nfs sbin/pfctl
sbin/raidctl sbin/restore
share/man share/misc
share/zoneinfo sys/arch/amd64/amd64
sys/arch/amd64/stand/boot sys/arch/amd64/stand/cdboot
sys/arch/arm/arm sys/arch/aviion/aviion
sys/arch/hp300/hp300 sys/arch/hppa/stand/boot
sys/arch/hppa64/stand/boot sys/arch/i386/conf
sys/arch/i386/i386 sys/arch/i386/stand/boot
sys/arch/i386/stand/cdboot sys/arch/landisk/dev
sys/arch/landisk/include sys/arch/landisk/landisk
sys/arch/landisk/stand/boot sys/arch/luna88k/luna88k
sys/arch/macppc/macppc sys/arch/mvmeppc/mvmeppc
sys/arch/sh/dev sys/arch/sh/include
sys/arch/sh/sh sys/arch/vax/dec
sys/arch/vax/vsa sys/arch/zaurus/stand/zboot
sys/conf sys/dev
sys/dev/acpi sys/dev/cardbus
sys/dev/hil sys/dev/i2c
sys/dev/ic sys/dev/mii
sys/dev/pci sys/dev/puc
sys/dev/usb sys/dev/wscons
sys/kern sys/net
sys/net80211 sys/netinet
sys/netinet6 sys/nfs
sys/scsi sys/sys
sys/ufs/ffs sys/ufs/ufs
usr.bin/awk usr.bin/cal
usr.bin/calendar usr.bin/cvs
usr.bin/encrypt usr.bin/file
usr.bin/ftp usr.bin/grep
usr.bin/mg usr.bin/sort
usr.bin/ssh usr.bin/talk
usr.bin/xlint usr.sbin
usr.sbin/acpidump usr.sbin/acpiscan
usr.sbin/bgpd usr.sbin/config
usr.sbin/httpd usr.sbin/ospfd
usr.sbin/pkg_add usr.sbin/ppp
usr.sbin/ripd
== bin =============================================================== 01/14 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/bin
dd
~ args.c
> Fix range checking for in.dbsz and out.dbsz.
> Add range checking for cbsz.
> Change type in format string to %zd and cast SSIZE_MAX to ssize_t.
> Fixes PR 5278 submitted by malaler at gmail dot com.
> OK otto@ and millert at .ay@)
df
~ df.c
> like -i, -h is incompatible with -P; suggested by jmc@; ok ray@ (otto@)
ksh
~ main.c
> Only check if last three letters are "/sh" if string is at least
> three characters long.
> OK moritz at .ay@)
ps
~ ps.1 ~ ps.c
> simplify synopsis and sync usage(); (jmc@)
~ ps.1
> note that -L may not be specified with other options; (jmc@)
~ ps.1
> give FILES a bit of space; (jmc@)
~ ps.1
> - no need for double markup
> - prefer .Sq for single letters (jmc@)
~ ps.1
> -.Xr sh 1 , (jmc@)
~ ps.1
> +.Ex -std ps (jmc@)
~ ps.1
> some minor adjustments to the options list; (jmc@)
~ ps.1
> overhaul the keywords section; (jmc@)
== distrib =========================================================== 02/14 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/distrib
landisk
~ ramdisk/Makefile
> build the right kernel... (drahn@)
sets
~ lists/man/mi
> sync (deraadt@)
~ lists/base/mi ~ lists/etc/mi
~ lists/man/mi
> sync (deraadt@)
~ lists/base/mi
> sync (deraadt@)
~ lists/man/mi
> sync (deraadt@)
~ lists/base/md.alpha ~ lists/base/md.amd64
~ lists/base/md.armish ~ lists/base/md.aviion
~ lists/base/md.cats ~ lists/base/md.hp300
~ lists/base/md.hppa ~ lists/base/md.hppa64
~ lists/base/md.i386 ~ lists/base/md.luna88k
~ lists/base/md.mac68k ~ lists/base/md.macppc
~ lists/base/md.mvme68k ~ lists/base/md.mvme88k
~ lists/base/md.mvmeppc ~ lists/base/md.sgi
~ lists/base/md.sparc ~ lists/base/md.sparc64
~ lists/base/md.vax ~ lists/base/md.zaurus
~ lists/comp/mi
> sync (deraadt@)
~ lists/base/md.alpha ~ lists/base/md.amd64
~ lists/base/md.armish ~ lists/base/md.aviion
~ lists/base/md.cats ~ lists/base/md.hp300
~ lists/base/md.hppa ~ lists/base/md.hppa64
~ lists/base/md.i386 ~ lists/base/md.luna88k
~ lists/base/md.mac68k ~ lists/base/md.macppc
~ lists/base/md.mvme68k ~ lists/base/md.mvme88k
~ lists/base/md.mvmeppc ~ lists/base/md.sgi
~ lists/base/md.sparc ~ lists/base/md.sparc64
~ lists/base/md.vax ~ lists/base/md.zaurus
~ lists/comp/mi
> sync (deraadt@)
== etc =============================================================== 03/14 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/etc
Makefile
~ Makefile
> Add the _ripd user and startup stuff.
> ok claudio@ (norby@)
changelist
~ changelist
> Add the _ripd user and startup stuff.
> ok claudio@ (norby@)
ftpusers
~ ftpusers
> Add the _ripd user and startup stuff.
> ok claudio@ (norby@)
group
~ group
> Add the _ripd user and startup stuff.
> ok claudio@ (norby@)
mail/aliases
~ mail/aliases
> Add the _ripd user and startup stuff.
> ok claudio@ (norby@)
master.passwd
~ master.passwd
> Add the _ripd user and startup stuff.
> ok claudio@ (norby@)
mtree/special
~ mtree/special
> Add the _ripd user and startup stuff.
> ok claudio@ (norby@)
rc
~ rc
> Add the _ripd user and startup stuff.
> ok claudio@ (norby@)
rc.conf
~ rc.conf
> Add the _ripd user and startup stuff.
> ok claudio@ (norby@)
security
~ security
> generate diffs for files listed in /etc/changelist as they are created
> and deleted. previously this script only generated diffs for existing
> files.
> ok lots of people including millert@ msf@ mcbride@ todd@ and probably more.
> (dlg@)
== games ============================================================= 04/14 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/games
atc
~ main.c
> Check for valid character before truncating string.
> Initial patch from Charles Longeau <chl at tuxfamily dot org>.
> OK moritz at .ay@)
backgammon
~ backgammon/backgammon.6 ~ backgammon/main.c
~ backgammon/move.c ~ common_source/back.h
~ common_source/init.c ~ common_source/subs.c
> -d: make it possible playing without doubling as it is not commonly
> played everywhere in the world
> ok todd@ otto@ (martin@)
~ backgammon/backgammon.6
> Add new -d flag to SYNOPSIS and remove optional and unfunctional
> [-] flag, which, according to FreeBSD logs, is supposed to print
> the usage.
> Sort flags in SYNOPSIS and DESCRIPTION.
> OK martin@, jmc at .etective work and ok to kill [-] otto at .ay@)
~ backgammon/backgammon.6 ~ common_source/subs.c
> Remove -h flag and simplify usage output to one line, like most
> other programs.
> Vote to simplify usage jmc@, OK martin at .ay@)
~ backgammon/backgammon.6
> Sort flags and commands.
> Use correct capitalization and punctuation in flag descriptions.
> Mark up -n flag in description.
> Help and OK jmc at .ay@)
~ backgammon/backgammon.6
> tidy up the quick reference section; (jmc@)
mille
~ mille.c
> usage information goes to stderr (martin@)
quiz
~ quiz.c
> Don't check buf[strlen(buf) - 1] on empty strings.
> OK moritz at .ay@)
sail
~ pl_main.c
> Check fgets failure and clarify character checks.
> OK moritz at .ay@)
trek
~ out.c
> Don't access buf[strlen(buf) - 1] on empty strings.
> OK moritz at .ay@)
== gnu =============================================================== 05/14 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/gnu
usr.bin/gcc
~ gcc/config/i386/i386.c
> Fix infinite loop when passing object with 3+ base classes by value:
> http://gcc.gnu.org/bugzilla/show_bug.cgi?id=18300
> Needed by openoffice and boost on amd64
> ok espie@ miod@ (wilfried@)
~ gcc/config/sh/openbsd.h
> Use __start (the symbol we provide) not start. ok miod@ (drahn@)
== kerberosV ========================================================= 06/14 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/kerberosV
kerberosV
~ src/appl/login/limits_conf.c
> Report to syslog strings that start with NUL; prevents negative
> index array access.
> OK moritz at .ay@)
== lib =============================================================== 07/14 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/lib
libc
~ stdio/vfprintf.c ~ stdlib/ecvt.c
~ stdlib/gcvt.c ~ stdlib/strtod.c
> make __dtoa & strtod() thread-safe useing the same method as newer gdtoa
> codebase. tested mostly by ckuethe and myself. __dtoa() use now requires
> a call to __freedtoa() (deraadt@)
~ stdio/fgets.3
> suceed -> succeed; from chad dougherty (jmc@)
~ stdlib/strtod.c
> must pull in sys/types.h unconditionaly, because it is needed my pthreads
> includes used a bit later (for instance on the vax) (deraadt@)
~ sys/getfh.2
> +.Xr fhstat 2
> from mjf at netbsd.1.21
> ok otto (jmc@)
~ stdlib/strtol.3
> columns for underflow and overflow were swapped, noticed by Mathieu Olivier
> <molivier at users dot sourceforge dot net>
> ok jmc@ mickey@ (grunk@)
~ arch//sh/sys/fork.S
> Dont double fork, and since the assembles the same, use -1, instead of 0xff
> to subtract one. (drahn@)
libevent
~ event.c
> fix a potential memory leak in event_once().
> From Niels Provos via the libevent SVN repo, by Scott Lamb
> tested by niallo@
> ok deraadt@ (brad@)
~ kqueue.c
> typo in kqueue delete.
> From Niels Provos via the libevent SVN repo, by Bert JW. Regeer
> ok deraadt@ (brad@)
libssl
~ src/crypto/asn1/tasn_dec.c ~ src/crypto/dh/dh.h
~ src/crypto/dh/dh_err.c ~ src/crypto/dh/dh_key.c
~ src/crypto/dsa/dsa.h ~ src/crypto/dsa/dsa_err.c
~ src/crypto/dsa/dsa_ossl.c ~ src/crypto/rsa/rsa.h
~ src/crypto/rsa/rsa_eay.c ~ src/crypto/rsa/rsa_err.c
~ src/ssl/s2_clnt.c ~ src/ssl/ssl_lib.c
TAGGED OPENBSD_4_0
> MFC:
> Fix by pvalchev@
> openssl security fixes, diff from markus@, ok & "commit it" djm@
> http://www.openssl.org/news/secadv_20060928.txt for more (brad@)
== libexec =========================================================== 08/14 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/libexec
spamd
~ spamd.c
> use setresuid/gid and check for failure, ok bob (henning@)
spamlogd
~ Makefile ~ spamlogd.8
~ spamlogd.c
> donot for tcpdump, use libpcap directly.
> joint work with Berk D. Demir, ok beck deraadt (henning@)
~ spamlogd.8
> remove misplaced -l stuff; henning needs more caffeine... (jmc@)
== regress =========================================================== 09/14 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/regress
sbin
~ ipsecctl/Makefile + ipsecctl/ikefail10.in
+ ipsecctl/ikefail10.ok + ipsecctl/ikefail11.in
+ ipsecctl/ikefail11.ok + ipsecctl/ikefail12.in
+ ipsecctl/ikefail12.ok + ipsecctl/ikefail8.in
+ ipsecctl/ikefail8.ok + ipsecctl/ikefail9.in
+ ipsecctl/ikefail9.ok
> Add some regression tests for odd ipsecctl behaviour noticed by
> Prabhu Gurumurt. Test ikefail10 should fail, but does not and needs
> to be fixed. (hshoexer@)
~ ipsecctl/ikefail11.in ~ ipsecctl/ikefail11.ok
> Remove bogus input line. (hshoexer@)
~ ipsecctl/ikedel1.ok ~ ipsecctl/ikedel10.ok
~ ipsecctl/ikedel11.ok ~ ipsecctl/ikedel12.ok
~ ipsecctl/ikedel13.ok ~ ipsecctl/ikedel14.ok
~ ipsecctl/ikedel15.ok ~ ipsecctl/ikedel16.ok
~ ipsecctl/ikedel17.ok ~ ipsecctl/ikedel18.ok
~ ipsecctl/ikedel19.ok ~ ipsecctl/ikedel2.ok
~ ipsecctl/ikedel20.ok ~ ipsecctl/ikedel21.ok
~ ipsecctl/ikedel22.ok ~ ipsecctl/ikedel23.ok
~ ipsecctl/ikedel29.ok ~ ipsecctl/ikedel3.ok
~ ipsecctl/ikedel30.ok ~ ipsecctl/ikedel31.ok
~ ipsecctl/ikedel32.ok ~ ipsecctl/ikedel33.ok
~ ipsecctl/ikedel34.ok ~ ipsecctl/ikedel35.ok
~ ipsecctl/ikedel36.ok ~ ipsecctl/ikedel37.ok
~ ipsecctl/ikedel38.ok ~ ipsecctl/ikedel39.ok
~ ipsecctl/ikedel4.ok ~ ipsecctl/ikedel40.ok
~ ipsecctl/ikedel41.ok ~ ipsecctl/ikedel42.ok
~ ipsecctl/ikedel43.ok ~ ipsecctl/ikedel46.ok
~ ipsecctl/ikedel5.ok ~ ipsecctl/ikedel6.ok
~ ipsecctl/ikedel7.ok ~ ipsecctl/ikedel8.ok
~ ipsecctl/ikedel9.ok
> Adjust existing ikedel tests for aggressive mode support (we now
> delete both mainmode and aggressive mode phase 1 transforms) (mcbride@)
== sbin ============================================================== 10/14 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/sbin
ipsecctl
~ ike.c ~ ipsec.conf.5
~ ipsecctl.c ~ ipsecctl.h
~ parse.y
> Add support for aggressive mode (from the k2k6 IPsec hackathon).
> ok hshoexer (mcbride@)
~ ike.c ~ ipsecctl.c
> KNF unrelated to previous commit. (mcbride@)
isakmpd
~ message.c
> Fix TAILQ usage, preventing crashes
> Okay henning@ krw@ millert@ hshoexer@ (pedro@)
mount_nfs
~ mount_nfs.c
> storing return value of strtol() in int variable was not safe,
> also strtol() result was not checked for under/overflow
> thus, rewrite getopt switch/cases with strtonum() and sensible bounds
> help from mickey@ millert@, ok millert@, no objections otto@ (grunk@)
pfctl
~ parse.y ~ pfctl.c
> - don't allow anchors with _* names to be cleared or loaded from the
> command line (but they can still be viewed)
> - don't allow users to specify _* as an anchor name in the ruleset
> - don't print _* anchor names with pfctl -sA unless -v is specified
> 'looks sensible' deraadt@ (mcbride@)
~ parse.y ~ pfctl.c
~ pfctl_optimize.c ~ pfctl_parser.h
> Allow pfctl ruleset optimizer to be controlled from the ruleset.
> "set" "ruleset-optimization" [ "none" | "basic" | "profile" ]
> You can optionally control ruleset optimization with these keywords on the
> command line with the -o option; the command line setting will override the
> pf.conf setting. The existing -o/-oo flags continue to work as expected.
> cleanup and ok henning@ (mcbride@)
~ pfctl.8
> Document new behaviour of the -o (ruleset optimization) flag. (mcbride@)
~ pfctl.c ~ pfctl_parser.c
~ pfctl_parser.h
> Allow a user to recursively print anchors including those without
> reserved names, if a trailing * is specified in the anchor name.
> e.g. recursively print the main ruleset:
> pfctl -a '*' -sr
> Recursively print the spam anchor:
> pfctl -a 'spam*'
> pfctl -a 'spam/*'
> Also fix a bug which prevented the contents of inline anchors with
> explicit names from being loaded into the kernel.
> ok henning@ (mcbride@)
~ pfctl.8
> Document recursive printing of anchors via -a '*' or -a 'anchor/*'.
> (mcbride@)
~ pfctl.c
> Don't recures ALL the time. (mcbride@)
~ pfctl.8
> tweaks; (jmc@)
~ pfctl.c
> sync usage(); ok mcbride (jmc@)
~ pfctl.c
> Don't open a transaction for a ruleset unless it's a brace ruleset that
> contains rules. Fixes DIOCXCOMMIT: Device busy when multiple anchors with
> the same name are specified.
> reported by ckuethe@ and mkb at crypt.org.rucbride@)
raidctl
~ raidctl.c
> Error out on empty string passed as device name.
> OK moritz at .ay@)
~ rf_configure.c
> Check for newline before truncating.
> OK moritz at .ay@)
restore
~ restore.8
> correctify example;
> from a mail posted to misc@ from uwe dippel;
> ok otto (jmc@)
== share ============================================================= 11/14 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/share
man
~ man4/acpi.4
> Add acpibtn based on initial submission from Dmitri Alenitchev and
> boiled down a bit by me after discussion with jmc.
> ok jmc (mk@)
~ man4/Makefile
> Add acpibtn.4. (mk@)
~ man8/security.8
> note that additions/removals of files in changelist will be reported too;
> ok dlg (jmc@)
~ man5/pf.conf.5
> Document set ruleset-optimization [ none | basic | profile ]. (mcbride@)
~ man5/pf.conf.5
> tweaks; (jmc@)
~ man7/ports.7 ~ man7/packages.7
> be more directive, assume people really want binary packages, and will
> resort to ports as a last resort.
> Also, unconfuse the part about sudo and make install. (espie@)
~ man7/ports.7 ~ man7/packages.7
> tweaks; (jmc@)
~ man4/man4.vax/lkkbd.4
> German LK-[24]01 keyboard layout;
> contributed by Sigi Rudzio (sigirudzio ; gmail . com) (miod@)
~ man4/vic.4 ~ man4/pci.4
> fix up vic a bit and add it to the pci man page. (brad@)
~ man4/vic.4
> add an example on how to add the virtual nic to a virtual machine.
> discussed with and ok'ed by jmc@ and dlg@ (fkr@)
~ man4/vic.4
> tweaks; (jmc@)
~ man7/ports.7
> say goodbye to BIN_PACKAGES (espie@)
~ man5/bsd.port.mk.5
> Document the new package repository structure, mark BIN_PACKAGES as
> obsolete,
> explain update and subupdate. (espie@)
~ man5/bsd.port.mk.5
> partially document the sub* stuff (espie@)
~ man5/bsd.port.mk.5
> Document print-plist/print-plist-contents (espie@)
~ man7/ports.7
> kill trailing space; (jmc@)
~ man5/bsd.port.mk.5
> tweaks; (jmc@)
~ man4/vic.4
> vmware can also emulate em these days, not just pcn.
> remove some verbosity that describes features we dont use or care about.
> im an author too.
> ok deraadt@ (dlg@)
~ man4/vic.4 ~ man4/pci.4
> vmxnet -> VMXnet
> ethernet -> Ethernet
> ok dlg (jmc@)
~ man4/vic.4
> improve EXAMPLES somewhat; from dlg and myself (jmc@)
~ man4/vic.4
> no need to capitalise "virtual"; (jmc@)
~ man4/vic.4
> move the hardware info to a more relevant part of the doc;
> no need for a separate HARDWARE section;
> ok dlg (jmc@)
~ man5/bsd.port.mk.5
> SCRIPTDIR and SCRIPTS_ENV are obsolete now. (espie@)
~ man5/bsd.port.mk.5
> tweaks; (jmc@)
~ man5/bsd.port.mk.5
> oops - forgot to kill trailing spaces; (jmc@)
~ man4/arc.4 ~ man4/mpi.4
> new devices and some sorting (dlg@)
~ man4/nfe.4
> add the MCP67 chipset. (brad@)
misc
~ airport
> Remove a double (and misspelled) entry for VCE.
> ok jmc (mbalmer@)
~ inter.phone
> Correct some swiss entries and mispellings.
> ok claudio (mbalmer@)
zoneinfo
~ datfiles/northamerica
> Uncomment changes made for the Energy Policy Act of 2005; it looks
> like they will actually happen. (millert@)
~ datfiles/asia ~ datfiles/europe
~ datfiles/iso3166.tab ~ datfiles/northamerica
~ datfiles/southamerica ~ datfiles/zone.tab
> Update to tzdata2006n from elsie.nci.nih.gov (millert@)
== sys =============================================================== 12/14 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/sys
arch/amd64/amd64
~ disksubr.c
> Remove redundant if (msg) {} block around invocations of
> iso_disklabelspoof() and udf_disklabelspoof(). Use (msg &&
> iso_disklabelspoof()) instead. Makes more arch's textually consistant.
> No functional change. (krw@)
~ disksubr.c
> Don't need to 'goto done;' when 'done:' is the next line. (krw@)
~ autoconf.c
> The default for swap should be partition 'b' on the same unit as root is,
> not partition 'b' on unit 0.
> ok deraadt@ (kettenis@)
arch/amd64/stand/boot
~ Makefile
> install boot.conf as a section 5 page, not section 8;
> from mikolaj kucharski;
> ok mk deraadt krw (jmc@)
arch/amd64/stand/cdboot
~ cdboot.8
> Mention that cdboot won't find boot.conf if you create the CD
> with 8.3-style filenames. Issue raised by Mikolaj Kucharski,
> eth0 (at) o2.pl.
> Jointly written with, and ok, jmc at .om@)
arch/arm/arm
~ disksubr.c
> Remove redundant if (msg) {} block around invocations of
> iso_disklabelspoof() and udf_disklabelspoof(). Use (msg &&
> iso_disklabelspoof()) instead. Makes more arch's textually consistant.
> No functional change. (krw@)
arch/aviion/aviion
~ disksubr.c
> Remove redundant if (msg) {} block around invocations of
> iso_disklabelspoof() and udf_disklabelspoof(). Use (msg &&
> iso_disklabelspoof()) instead. Makes more arch's textually consistant.
> No functional change. (krw@)
arch/hp300/hp300
~ disksubr.c
> Don't need to 'goto done;' when 'done:' is the next line. (krw@)
arch/hppa/stand/boot
~ Makefile
> install boot.conf as a section 5 page, not section 8;
> from mikolaj kucharski;
> ok mk deraadt krw (jmc@)
arch/hppa64/stand/boot
~ Makefile
> install boot.conf as a section 5 page, not section 8;
> from mikolaj kucharski;
> ok mk deraadt krw (jmc@)
arch/i386/conf
~ GENERIC ~ RAMDISK_CD
> enable vic(4) now that it works. (dlg@)
arch/i386/i386
~ disksubr.c
> Remove redundant if (msg) {} block around invocations of
> iso_disklabelspoof() and udf_disklabelspoof(). Use (msg &&
> iso_disklabelspoof()) instead. Makes more arch's textually consistant.
> No functional change. (krw@)
~ autoconf.c
> The default for swap should be partition 'b' on the same unit as root is,
> not partition 'b' on unit 0.
> ok deraadt@ (kettenis@)
~ est.c
> Add deduced power state table for 1.00 GHz VIA Eden 90nm 'Esther'.
> Tested & confirmed by Greg Mortensen, thevision at pobox.com (dim@)
arch/i386/stand/boot
~ Makefile
> install boot.conf as a section 5 page, not section 8;
> from mikolaj kucharski;
> ok mk deraadt krw (jmc@)
arch/i386/stand/cdboot
~ cdboot.8
> Mention that cdboot won't find boot.conf if you create the CD
> with 8.3-style filenames. Issue raised by Mikolaj Kucharski,
> eth0 (at) o2.pl.
> Jointly written with, and ok, jmc at .om@)
arch/landisk/dev
~ obio.c
> Do not map obio mappings twice. (miod@)
arch/landisk/include
~ conf.h
> add scif (mickey@)
~ conf.h
> cleanup some devices (mickey@)
arch/landisk/landisk
~ disksubr.c
> Remove redundant if (msg) {} block around invocations of
> iso_disklabelspoof() and udf_disklabelspoof(). Use (msg &&
> iso_disklabelspoof()) instead. Makes more arch's textually consistant.
> No functional change. (krw@)
~ conf.c
> need an entry for scif (mickey@)
~ conf.c
> cleanup some devices (mickey@)
arch/landisk/stand/boot
~ devs.c ~ libsa.h
~ srt0.S
> Disable cache before jumping into the kernel, makes bootloader more
> reliable.
> ok miod@ (drahn@)
~ devs.c
> These printfs appear unnecessary now that the cache is disabled. (drahn@)
arch/luna88k/luna88k
~ m8820x.c
> Changed CMMU probing method. On the luna88k, badaddr() returns 'good'
> on unpopulated slots, so we check the CMMU type value for each CMMU
> register address.
> ok miod@ (aoyama@)
~ machdep.c
> Now the primary CPU starts the CMMUs which tied to the secondary CPUs.
> Without this, the performance is not good on multi-CPU machine with
> non-MULTIPROCESSOR kernel, because the secondary CPUs run on the
> never-ending loop without cache.
> ok miod@ (aoyama@)
arch/macppc/macppc
~ disksubr.c
> Remove redundant if (msg) {} block around invocations of
> iso_disklabelspoof() and udf_disklabelspoof(). Use (msg &&
> iso_disklabelspoof()) instead. Makes more arch's textually consistant.
> No functional change. (krw@)
arch/mvmeppc/mvmeppc
~ disksubr.c
> Remove redundant if (msg) {} block around invocations of
> iso_disklabelspoof() and udf_disklabelspoof(). Use (msg &&
> iso_disklabelspoof()) instead. Makes more arch's textually consistant.
> No functional change. (krw@)
arch/sh/dev
~ scif.c
> cleanup some devices (mickey@)
arch/sh/include
~ db_machdep.h
> proper brkpt detection so continue does restart proper (mickey@)
~ param.h
> For now use 3 pages like netbsd, allows the kernel to get _MUCH_ farther.
> I suspect this is really a bug in the P1_STACK code path. (drahn@)
~ psl.h
> define the bit that sh4 uses for FPU disable. (drahn@)
+ fpu.h
> Some basic fpu trap handling. (miod@)
~ locore.h
> Use TF_SPC instead of hardcoding its value. (miod@)
arch/sh/sh
~ locore_subr.S
> Remove unused code and move the knowledge of struct switchframe to a single
> place. (miod@)
~ trap.c
> step and stuff works in ddb now too (mickey@)
~ vm_machdep.c
> mdproc has to be initialised (mickey@)
~ pmap.c
> must steal pages from the end of the physseg (mickey@)
~ trap.c
> Return address is way too useful to not print here. (drahn@)
~ db_trace.c
> make stack trace look as other flavoured stack traces look like (mickey@)
~ clock.c
> make compile w/ debug (mickey@)
~ vm_machdep.c
> Fix a couple of comments. (miod@)
~ trap.c
> Some basic fpu trap handling. (miod@)
~ interrupt.c
> Correctly release resources in intc_free_ih(). (miod@)
arch/vax/dec
~ wskbdmap_lk201.c
> German LK-[24]01 keyboard layout;
> contributed by Sigi Rudzio (sigirudzio ; gmail . com) (miod@)
arch/vax/vsa
~ gpx.c
> Fix Bt458 programming; screen burner now works on 8bpp flavour. (miod@)
~ gpx.c
> Do not enable hardware cursor in the Brooktree DAC on the 8bpp model as
> well
> (it would only appear when booting after an UNJAM). (miod@)
~ ncr.c
> Wait two seconds after reset before probing devices, lets RX23 get detected
> again. (miod@)
arch/zaurus/stand/zboot
~ Makefile
> install boot.conf as a section 5 page, not section 8;
> from mikolaj kucharski;
> ok mk deraadt krw (jmc@)
conf
~ newvers.sh
TAGGED OPENBSD_4_0
> -stable (brad@)
dev
~ audio.c ~ cons.c
~ midi.c
> poll errors should be POLLERR, not some random E value
> from alexandre ratchov. ok claudio (tedu@)
~ systrace.c ~ systrace.h
TAGGED OPENBSD_4_0
> MFC:
> Fix by djm@
> fix an exploitable integer overflow found by Chris Evans of Google
> Security; ok deraadt (brad@)
dev/acpi
~ dsdt.c
> Fix nameref -> objref for forward definition evaluator
> ok marco@ (jordan@)
~ dsdt.c
> add missing prototype for aml_evalinteger()
> ok macro@ (thib@)
~ acpitz.c
> Ooops, forgot a debug printf. (marco@)
~ acpitz.c
> Add active cooling support mostly by me with jordan help.
> Add initial passive cooling trending formulas by jordan.
> ok jordan@ (marco@)
~ acpi.c
> Fix comments to reflect reality. (marco@)
~ acpi.c
> Kill acpiec attach function and add its functionality to the generic _HID
> device attach
> prodded and ok jordan (marco@)
dev/cardbus
~ if_malo_cardbus.c
> Add first TX path and active scan bits. Don't expect to much yet.
> ok claudio@ (mglocker@)
dev/hil
~ hil.c ~ hilvar.h
> Handle loop reconfiguration in a kernel thread, instead of doing it from
> interrupt context. (miod@)
dev/i2c
~ i2c_scan.c
> lm75 scoring should not show up in verbose (deraadt@)
dev/ic
~ malo.c ~ malo.h
> Add first TX path and active scan bits. Don't expect to much yet.
> ok claudio@ (mglocker@)
~ malo.c
> Fix some nasty bugs in malo_reset() and malo_cmd_set_channel() and simplify
> the code a bit. Still impossible to correctly set a channel but it is no
> longer crashing. OK mglocker@ (claudio@)
~ malo.c
> MALO_CMD_RESPONSE is acting just as a flag so move it to the end of the
> list.
> OK mglocker@ (claudio@)
~ malo.c
> In malo_ioctl() if the returned error is ENETRESET reset the device.
> This is net80211 way of telling the driver that settings have changed.
> OK mglocker@ (claudio@)
~ malo.c
> Reorganize malo_intr() a bit. Check for unhandled interrupt flags and
> do not call wakeup() for each and every command sent. I think more and more
> will not tsleep(). OK mglocker@ (claudio@)
~ rt2661.c
> Fix dynamic tuning of the radio sensitivity I broke when moving from
> rssadapt(9) to AMRR.
> pointed out by Erki Malling on misc@ (damien@)
~ re.c ~ rtl81x9reg.h
> - Make DMA descripter members volatile
> - Remove usage of BUS_DMA_ALLOCNOW with bus_dmamap_create
> - Use BUS_DMA_COHERENT on mapping DMA memory for TX/RX descriptors
> - Adjust m_data and m_len directly rather than calling m_adj
> - Proper use of bus_dmamap_sync against DMA descriptors
> From tsutsui at NetBSD > Tested by kettenis@ brad@ otto@
> Tested on armish/amd64/i386/sparc64 (brad@)
~ re.c ~ rtl81x9reg.h
> - Rename a variable to clarify meaning of index.
> - Rename RL_[TR]X_DESC_INC() macro to RL_NEXT_[TR]X_DESC() and
> change them to return rvalues to avoid possible side effects.
> - Also define RL_NEXT_TXQ() for rl_txq index and use it as well.
> From tsutsui at NetBSDrad@)
~ re.c ~ rtl81x9reg.h
> Put common data for each RX DMA descriptor into a new rxsoft structure.
> From tsutsui at NetBSD > Tested by brad@ otto@
> Tested on amd64/i386/sparc64 (brad@)
~ fxp.c
> ANSI (brad@)
dev/mii
~ rlphy.c
> simplify parent driver matching code a bit. no-op change. (brad@)
dev/pci
~ pcidevs
> add an ALI PCI-ISA bridge and an NEC PCIe to PCI-X bridge. (brad@)
~ pcidevs.h ~ pcidevs_data.h
> regen (brad@)
~ if_em.c
> removed unused variable.
> ok brad@ (dlg@)
~ pcidevs
> fix a typo in the comment above the SNI entries, SNC -> SNI.
> From Igor Sobrado <igor at string1 dot ciencias dot uniovi dot es> (brad@)
~ pcidevs.h ~ pcidevs_data.h
> regen (brad@)
~ pcidevs.h ~ pcidevs_data.h
> regen (brad@)
~ pcidevs
> add PCI ids for newer Intel PRO/1000 chipsets. (brad@)
- if_vicvar.h ~ if_vic.c
~ if_vicreg.h
> absolutely huge reworking on this driver (sorry reyk). so far attach,
> resource mapping, mem allocation, init, and rx is working. the tx path
> needs work, but that can happen in tree. overall it's closer to working
> than the old code. if_vicvar.h goes away since nothing else uses it.
> im getting this in so other people can play with it too.
> ok brad@ (dlg@)
~ if_vic.c
> we're passing physical addresses to the adapter, so say so. add some magic
> that lets me see packets coming out of the box. (dlg@)
~ arc.c
> first cut at sensors
> help from & ok dlg@ (jolan@)
~ if_vic.c
> fix the barriers in setlladdr. (dlg@)
~ if_vic.c
> vmware frames are slightly short, but that is ok. done with clues from
> brad@ (dlg@)
~ if_vic.c
> let the tx completion path clean up by recording which mbuf we're
> sending in the tx forward path. gotta love one line fixes...
> committed over vic (dlg@)
~ pcidevs.h ~ pcidevs_data.h
> sync (deraadt@)
~ pcidevs
> amd 8111 has an ethernet, but we have never seen anyone use it yet
> (deraadt@)
~ if_vic.c
> fix the tx path so it can use more than one scatter gather entry when
> sending the packet. this makes it less likely that we'll have to repack
> fragmented packets for transmission. (dlg@)
~ if_wpi.c
> fix firmware fatal errors on re-associations.
> reported and tested by Marc Winiger and dhartmei@ (damien@)
~ arc.c
> whitespace fixes (dlg@)
~ arc.c
> dont walk the device list looking for our scsibus, it is stashed when we
> attach it. (dlg@)
~ if_vic.c
> Use correct error message as return value. OK dlg@ (claudio@)
~ if_vic.c
> whitespaces cleanup, no functional change
> ok dlg@ (fkr@)
~ if_vic.c
> rewrite the tx path so it doesnt feel so lopsided.
> now i can have coffee! (dlg@)
~ arc.c
TAGGED OPENBSD_4_0
> MFC:
> Fix by dlg@
> well, this is embarassing.
> it turns out a drivers detach code isnt called when the machine is shutting
> down which meant my bits to stop bgrb and sync the cache werent being
> run. without those bits being run the filesystems on these controllers
> were never clean on reboot, so we'd always get an fsck.
> so now we provide a shutdown hook which does get run on shutdown to sync
> the cache properly.
> found by claudio@
> ok dlg@ (brad@)
~ if_vic.c
TAGGED OPENBSD_4_0
> more whitespace fixes (dlg@)
~ if_em.c
TAGGED OPENBSD_4_0
> MFC:
> Fix by brad@
> revert revision 1.131, the code in question was later found to not ensure
> the proper alignment requirement for the VLAN layer on strict alignment
> architectures. This would result in Jumbo's working fine as long as VLANs
> were not in use. If VLANs were in use and a packet comes in with a size
> of 2046 bytes or larger, it would be corrupted as it came up through the
> VLAN layer. Also check the hw max frame size, instead of the MTU, so the
> alignment fixup is done as appropriate.
> Fixes PR 5185.
> Tested by Rui DeSousa with macppc and myself with alpha/sparc64 (brad@)
~ if_vic.c
TAGGED OPENBSD_4_0
> VIC_DEBUG is useless, so it goes away (dlg@)
~ if_vic.c
TAGGED OPENBSD_4_0
> we sync regions in the dma area anymore, we just sync the whole thing. this
> means the insane macro used to calculate the offset for syncing can go
> away. (dlg@)
~ if_vic.c
TAGGED OPENBSD_4_0
> set the proper maximum queue length. (brad@)
~ if_vic.c
TAGGED OPENBSD_4_0
> split the initialisation of the dma area up between the attach path and the
> interface init path. the init path now does the allocation of the dmamaps
> for both tx and rx, and allocates the mbufs for the rx ring.
> the interface stop is now uncommented. it waits for the hardware to stop
> working, and cleans up the bits allocated in the init path.
> ifconfig vic0 up and ifconfig vic0 down both work now. (dlg@)
~ if_vic.c
TAGGED OPENBSD_4_0
> remove cruft (dlg@)
~ if_vic.c
TAGGED OPENBSD_4_0
> set up a tick to check the link state every second. (dlg@)
~ if_vic.c
TAGGED OPENBSD_4_0
> reyk said i can put my copyright on this file too. (dlg@)
~ if_vic.c
TAGGED OPENBSD_4_0
> pointers in hardware structures are stupid. shame on you vmware.
> this should have a chance at running on amd64 now. can anyone test? (dlg@)
- if_vicreg.h ~ if_vic.c
TAGGED OPENBSD_4_0
> merge the register description into the c file. nothing else will ever
> use the reg descriptions, so why pollute the tree with it? (dlg@)
~ if_vic.c
TAGGED OPENBSD_4_0
> dont set tx_stopped when we actually want to do tx. (dlg@)
~ if_vic.c
TAGGED OPENBSD_4_0
> comments like "schedule timeout" before going timeout_add are dumb. learn
> to read the code. (dlg@)
~ if_vic.c
TAGGED OPENBSD_4_0
> if the hardware doesnt say it can do scatter gather lists for tx, then
> limit it to one dma segment so it only fills one scatter gather entry.
> (dlg@)
~ pcidevs.h ~ pcidevs_data.h
TAGGED OPENBSD_4_0
> regen (brad@)
~ pcidevs
TAGGED OPENBSD_4_0
> correct the Intel 108e id, its not a Gig MAC PCI id but an IPMI
> interface for the NIC and add 108f which is an AMT interface for
> the NIC. (brad@)
~ if_em.c ~ if_em_hw.c
~ if_em_hw.h
TAGGED OPENBSD_4_0
> the 8257E KCS PCI id is not the MAC but an IPMI interface being provided
> from the chipset, so remove it. (brad@)
~ if_vr.c
TAGGED OPENBSD_4_0
> vr_intr(): remove duplicate check for VR_ISR_RX_NOBUF and the "packet lost"
> printf which will never be reached. (brad@)
~ if_vr.c
TAGGED OPENBSD_4_0
> vr_rxeof(): stick receive error messages under VR_DEBUG.
> vr_init(): add debug printfs for some error conditions. (brad@)
~ pcidevs
TAGGED OPENBSD_4_0
> add the Broadcom BCM5825 chipset. (brad@)
~ pcidevs.h ~ pcidevs_data.h
TAGGED OPENBSD_4_0
> regen (brad@)
~ pcidevs
TAGGED OPENBSD_4_0
> Add Radeon X1300 Pro (with secondary).
> input and ok brad (mk@)
~ pcidevs.h ~ pcidevs_data.h
TAGGED OPENBSD_4_0
> regen (mk@)
~ pcidevs
TAGGED OPENBSD_4_0
> add a few more TI and O2 Micro devices. (brad@)
~ pcidevs.h ~ pcidevs_data.h
TAGGED OPENBSD_4_0
> regen (brad@)
~ pcidevs.h ~ pcidevs_data.h
TAGGED OPENBSD_4_0
> regen (brad@)
~ pcidevs
TAGGED OPENBSD_4_0
> add the NVIDIA MCP67 IDE PCI id.
> From Peer Chen @ NVIDIA via Linux (brad@)
~ arc.c
TAGGED OPENBSD_4_0
> query the firmware for the number of volumes, and then iterate through them
> to get the right names for sensor creation. (dlg@)
~ pcidevs.h ~ pcidevs_data.h
TAGGED OPENBSD_4_0
> regen (brad@)
~ pcidevs
TAGGED OPENBSD_4_0
> add the NVIDIA MCP67 SATA PCI ids.
> From Peer Chen @ NVIDIA via Linux (brad@)
~ pcidevs
TAGGED OPENBSD_4_0
> add the NVIDIA MCP67 AHCI and HD Audio PCI ids.
> From Peer Chen @ NVIDIA via Linux (brad@)
~ pcidevs.h ~ pcidevs_data.h
TAGGED OPENBSD_4_0
> regen (brad@)
~ if_fxp_pci.c
TAGGED OPENBSD_4_0
> ANSI (brad@)
~ pcidevs.h ~ pcidevs_data.h
TAGGED OPENBSD_4_0
> regen (brad@)
~ pcidevs
TAGGED OPENBSD_4_0
> add the NVIDIA MCP67 LAN PCI ids.
> From Peer Chen @ NVIDIA via Linux (brad@)
~ if_nfe.c
TAGGED OPENBSD_4_0
> add the NVIDIA MCP67 LAN PCI ids. (brad@)
~ pciide.c
TAGGED OPENBSD_4_0
> add the MCP67 PATA and SATA PCI ids as well as fixing nforce_chip_map() to
> set the proper UDMA capability for MCP61 and MCP65 chipsets.
> ok jsg@ (brad@)
dev/puc
~ com_puc.c
> kill leftover debug printf
> ok mickey (martin@)
dev/usb
~ ugen.c ~ uhid.c
~ usb.c ~ uscanner.c
> poll errors should be POLLERR, not some random E value
> from alexandre ratchov. ok claudio (tedu@)
~ usbdevs
> BWCT serial (jolan@)
~ usbdevs.h ~ usbdevs_data.h
> regen (jolan@)
~ if_zyd.c
> abort bulk-in and bulk-out pipes in zyd_stop() before freeing the
> Rx/Tx xfer lists. (damien@)
~ if_zyd.c ~ if_zydreg.h
> add new code to handle the AL7230B radios.
> code written by Florian Stoehr. not tested due to lack of hardware.
> if you happen to have an adapter with this radio please report success/
> failure. (damien@)
dev/wscons
~ wsdisplay.c ~ wsmouse.c
~ wsmux.c
> poll errors should be POLLERR, not some random E value
> from alexandre ratchov. ok claudio (tedu@)
kern
~ vfs_subr.c
> use vp->v_type to index into vtypes rather then vp->v_tag,
> fixing odd output in the 'show vnode' ddb code.
> ok mickey@ (thib@)
~ kern_tc.c ~ kern_time.c
> Timecounter based implementation of adjfreq(2). Largely from art@
> Tested by various using not (yet) committed amd64 timecounter code.
> ok deraadt@ (otto@)
~ uipc_usrreq.c
> undo unp_internalize in case of errors; report Christian Biere; ok claudio
> (markus@)
net
~ if.h ~ if_ethersubr.c
> ether_input_mbuf() isn't necessary, turn it into a macro and deal with
> it's "special" case in ether_input(). Based on similiar idea in FreeBSD.
> ok brad (jason@)
~ if_pfsync.c ~ if_pfsync.h
~ pf.c
> make pfsync a clonable too, but prevent more than one instance from
> beeing created for now - much more work would be required to change that
> input & ok ryan (henning@)
~ if_pfsync.c
> hard to believe people still manage to commit non-compiling code once in a
> while (deraadt@)
~ if_pfsync.c
> in pfsync_update_tdb, when there is no pfsync interface, we must return
> without trying to free the (in that case nonexistant) tdb mbuf
> found out the hard way by pedro (henning@)
~ if_pfsync.c
> slightly improve consustency and readability, no functional change
> (henning@)
~ if_pfsync.c
> remove redundant null check, ok ryan (henning@)
~ if_tun.c
> poll errors should be POLLERR, not some random E value
> from alexandre ratchov. ok claudio (tedu@)
~ if_pfsync.c ~ if_pflog.c
> Attach pfsync0 and pflog0 by default like they used to, /etc/rc depends on
> them being there.
> diff & ok deraadt (mcbride@)
net80211
~ ieee80211_output.c ~ ieee80211_proto.h
> Add two new functions:
> - ieee80211_get_rts
> - ieee80211_get_cts_to_self
> that can be use by drivers for chipsets that don't offer hardware assisted
> RTS/CTS protection (like ral/ural/rum).
> "no objections here" jsg@ (damien@)
netinet
~ tcp_input.c
> do not re-generate the timestamp modulation offset for SYN-ACK retransmits;
> this unbreaks TCP for high RTT (~3s); ok mcbride@ (markus@)
netinet6
~ in6.c
> Only run address hooks on SIOCAIFADDR_IN6 and SIOCDIFADDR_IN6.
> Before a normal user running ifconfig(8) could trigger up to three
> address hook calls per interface.
> OK mcbride@, henning@ (mpf@)
nfs
~ nfs_bio.c ~ nfs_subs.c
> move the declaration of nfsstats from nfs_bio.c to
> nfs_subs.c so it gets pulled in for NFSSERVER only
> kernels.
> ok deraadt@,krw@ (thib@)
scsi
~ scsi_base.c
> kthread_create returns an int, not a pointer. compare the return value to
> 0, not NULL. (dlg@)
sys
~ timetc.h
> Timecounter based implementation of adjfreq(2). Largely from art@
> Tested by various using not (yet) committed amd64 timecounter code.
> ok deraadt@ (otto@)
ufs/ffs
~ ffs_vnops.c
> LFS and FFS shared the same read and write vnops, through
> the macro games in ufs_readwrite.c (LFS_READWRITE) now that
> LFS has gone, make these functions ffs specific and move them
> into ffs/ffs_vnops.c zapping ufs_readwrite.c in the process.
> Also, remove lfs fields from the inode and ufsmount structs.
> ok toby@,deraadt@,pedro@
> testing and ok krw@ (thib@)
ufs/ufs
- ufs_readwrite.c ~ inode.h
~ ufsmount.h
> LFS and FFS shared the same read and write vnops, through
> the macro games in ufs_readwrite.c (LFS_READWRITE) now that
> LFS has gone, make these functions ffs specific and move them
> into ffs/ffs_vnops.c zapping ufs_readwrite.c in the process.
> Also, remove lfs fields from the inode and ufsmount structs.
> ok toby@,deraadt@,pedro@
> testing and ok krw@ (thib@)
== usr.bin =========================================================== 13/14 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/usr.bin
awk
~ tran.c
> from ru at freebsd > Fix a bug converting a variable from the numeric type to a string.
> ok otto deraadt (jmc@)
cal
~ cal.c
> Optimise a couple of printf()s; originally from Han Boetes.
> Reminded by, and ok, otto at .om@)
calendar
~ calendars/calendar.space
> other galaxies exist! (jasper@)
~ calendars/calendar.history
> fill november and december; now every date has an event.
> ok mickey (jasper@)
cvs
~ add.c ~ client.c
~ cvs.h ~ remote.h
~ remove.c ~ server.c
> a step ahead in opencvs add|remove remote support. (xsa@)
encrypt
~ encrypt.c
> There is no need to trim an empty string any further, just return
> it.
> OK moritz at .ay@)
file
~ apprentice.c
> Don't blindly truncate buf[strlen(buf) - 1] without checking for
> newline.
> Initial patch from Charles Longeau <chl at tuxfamily dot org>, fix
> inspired by comment by Anonymous Coward on undeadly.
> OK moritz@, ian at .ay@)
ftp
~ cmds.c ~ util.c
> Add checks for fgets and properly overwrite newline.
> Initial patch from Charles Longeau <chl at tuxfamily dot org>.
> OK moritz@ and jaredy at .ay@)
~ cmds.c ~ util.c
> Pass full buffer size to fgets.
> OK moritz@ and jaredy at .ay@)
~ ftp.1
> ftp(1) sends EPSV by default, not PASV;
> from daniel wade
> ok deraadt (jmc@)
grep
~ grep.c
> Check length before checking index of len - 1.
> OK moritz at .ay@)
mg
~ dired.c
> Don't overwrite line[strlen(line) - 1] when line is zero-length.
> Initial patch by Charles Longeau <chl at tuxfamily dot org>.
> OK kjell at .ay@)
sort
~ fsort.c ~ fsort.h
~ msort.c ~ sort.c
> Remove useless code that can cause a SEGV if a buffer is realloc()ed.
> Allocate buffers early and in one place instead of two.
> Make bufend global so it can be updated along with buffer.
> OK deraadt@ (millert@)
ssh
~ kexdhc.c ~ kexdhs.c
~ kexgexc.c ~ kexgexs.c
> check DH_compute_key() for -1 even if it should not happen because of
> earlier calls to dh_pub_is_valid(); report krahmer at suse.de; ok djm
> (markus@)
talk
~ get_names.c
> Plug a mem leak. Found in one of my trees. Forgat where it came from; most
> likely from a NetBSD Coverity fix. ok ray@ (otto@)
xlint
~ lint1/Makefile
> ouT makefile rules are smarter now, kill a redundant CLEANFILES (otto@)
== usr.sbin ========================================================== 14/14 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/usr.sbin
usr.sbin
~ Makefile
> Connect ripd and ripctl to the build.
> ok claudio@ deraadt@ (norby@)
~ Makefile
> Add Jordan's acpiscan acpi table dump tool. The current tool, acpidump,
> does not dump all relevant tables and will eventually be deprecated by
> this.
> It needs some more cleanup but we need to start getting acpi dumps from
> the field using this instead of acpidump.
> Code written by jordan, cleanup by me.
> ok jordan@ (marco@)
~ Makefile
> delete marco bullshit (deraadt@)
acpidump
~ aml/aml_parse.c
> Remove dead code; len is always >= 1.
> Convert a strncpy() to strlcpy() since it is followed by strlcat().
> OK marco@ mk@ (millert@)
acpiscan
+ Makefile + acpiscan.8
+ acpiscan.c + oslib.c
+ oslib.h
> Add Jordan's acpiscan acpi table dump tool. The current tool, acpidump,
> does not dump all relevant tables and will eventually be deprecated by
> this.
> It needs some more cleanup but we need to start getting acpi dumps from
> the field using this instead of acpidump.
> Code written by jordan, cleanup by me.
> ok jordan@ (marco@)
- Makefile - acpiscan.8
- acpiscan.c - oslib.c
- oslib.h
> Revert acpiscan commit. This should have replaced acpidump instead of
> being a new tool. My bad.
> prompted by deraadt. (marco@)
bgpd
~ pfkey.c ~ session.h
TAGGED OPENBSD_4_0
> MFC:
> Fix by henning@
> rev 1.33
> make sure we keep copies of everything we need to
> remove SAs and flows later again, even if the
> configuration info changed due to reload.
> rev 1.34
> storing the dynamically acquired SPIs for tcpmd5 inside the conf struct
> is not such a good idea - it gets nulled on config reloads, and thus
> we fail to clear the old SAs when the session is restarted after a config
> reload occured. obvious solution: store the SPIs outside the config
> area.
> ok henning@ (brad@)
~ pfkey.c ~ session.h
TAGGED OPENBSD_3_9
> MFC:
> Fix by henning@
> rev 1.33
> make sure we keep copies of everything we need to
> remove SAs and flows later again, even if the
> configuration info changed due to reload.
> rev 1.34
> storing the dynamically acquired SPIs for tcpmd5 inside the conf struct
> is not such a good idea - it gets nulled on config reloads, and thus
> we fail to clear the old SAs when the session is restarted after a config
> reload occured. obvious solution: store the SPIs outside the config
> area.
> ok henning@ (brad@)
config
~ Makefile
> Now that we have smarter lex and yacc rules, remove redundant lines
> from Makefile. ok millert@ (otto@)
httpd
~ src/main/http_protocol.c
TAGGED OPENBSD_4_0
> MFC:
> Fix by djm@
> escape error response to unsupported Expect: header, fixes
> CVE-2006-3918 (brad@)
ospfd
~ parse.y
> Allow braces to fit on a single line if only one or no argument is used.
> This allows construct like these:
> area 0.0.0.1 {
> interface lo0
> interface em1 { passive }
> }
> From Pierre-Yves Ritschard (claudio@)
~ ospfd.c
> Fix copy pasto, use SIGCHLD for the sigchild handler not SIGINT twice.
> Spotted by janus <at> errornet.de (claudio@)
pkg_add
~ pkg_create
> warn for any weird stuff in packages.
> Namely:
> - anything that doesn't belong to owner root/bin, group bin/wheel
> must be annotated in the packing-list.
> - any file setuid, setgid, writable by others must be annotated in
> the packing-list.
> These warnings must be fixed. They will become errors in the near futur.
> (espie@)
ppp
~ ppp/auth.c
> Check that last character is a newline before truncating; prevents
> negative index array access.
> OK moritz at .ay@)
ripd
~ auth.c ~ ripd.h
> Time is not a secret.
> Modulate the timer with an MD5 hash of the key to generate the sequence
> number.
> Ensure that the most significant bit of the modulated value is 0 to prevent
> wrapping of the sequence number.
> ok michele@ (mcbride@)
~ log.c ~ message.c
~ neighbor.c ~ packet.c
~ ripd.h ~ ripe.c
~ ripe.h
> create a temporary neighbor instead of a permanent one when a request is
> received (michele@)
~ neighbor.c
> Stop the running timer in case a neighbor is deleted. This fixes a crash
> found by Martin Boehme. ok michele@ (claudio@)
~ message.c
> if the nexthop is not reachable through outgoing interface set it to
> INADDR_ANY. Spotted out by Martin Boehme.
> OK claudio@ henning@ (michele@)
===============================================================================
More information about the owc
mailing list