[owc] Weekly src changes ending 2007-02-04
OWC
auto at squish.net
Mon Feb 5 07:18:13 GMT 2007
OpenBSD src changes summary for 2007-01-28 to 2007-02-04 inclusive
==================================================================
bin/df distrib/amd64
distrib/i386 distrib/sets
etc/Makefile etc/changelist
etc/etc.macppc/Makefile.inc etc/ftpusers
etc/group etc/hoststated.conf
etc/mail/aliases etc/master.passwd
etc/mklogin.conf etc/newsyslog.conf
etc/rc etc/rc.conf
etc/weekly gnu/usr.bin/perl
gnu/usr.sbin/sendmail lib/libc
lib/libevent lib/libpthread
libexec/ld.so regress/misc
sbin/dhclient sbin/init
sbin/pfctl sbin/shutdown
sbin/umount share/man
share/zoneinfo sys/arch/alpha/alpha
sys/arch/amd64/amd64 sys/arch/amd64/conf
sys/arch/amd64/include sys/arch/arm/arm
sys/arch/aviion/aviion sys/arch/hp300/hp300
sys/arch/hp300/include sys/arch/hppa/hppa
sys/arch/hppa64/hppa64 sys/arch/i386/conf
sys/arch/i386/i386 sys/arch/landisk/conf
sys/arch/landisk/landisk sys/arch/luna88k/dev
sys/arch/luna88k/include sys/arch/luna88k/luna88k
sys/arch/m68k/include sys/arch/m68k/m68k
sys/arch/mac68k/include sys/arch/mac68k/mac68k
sys/arch/macppc/macppc sys/arch/mips64/mips64
sys/arch/mvme68k/include sys/arch/mvme68k/mvme68k
sys/arch/mvmeppc/mvmeppc sys/arch/sparc/conf
sys/arch/sparc/include sys/arch/sparc/sparc
sys/arch/sparc64/conf sys/arch/sparc64/include
sys/arch/sparc64/sparc64 sys/arch/vax/include
sys/arch/vax/vax sys/dev
sys/dev/acpi sys/dev/ic
sys/dev/mii sys/dev/pci
sys/dev/pckbc sys/dev/sdmmc
sys/dev/usb sys/dev/vesa
sys/kern sys/net
sys/netinet sys/scsi
sys/sys sys/ufs/ffs
usr.bin/awk usr.bin/bc
usr.bin/compress usr.bin/cvs
usr.bin/ftp usr.bin/hexdump
usr.bin/pkg-config usr.bin/rcs
usr.bin/top usr.sbin
usr.sbin/ac usr.sbin/accton
usr.sbin/bgpctl usr.sbin/bgpd
usr.sbin/dhcpd usr.sbin/dvmrpd
usr.sbin/hoststatectl usr.sbin/hoststated
usr.sbin/httpd usr.sbin/ospfctl
usr.sbin/ospfd usr.sbin/pkg_add
usr.sbin/sa usr.sbin/vnconfig
== bin =============================================================== 01/12 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/bin
df
~ df.1
> statfs(2) and fstatfs(2) are one and the same, so don;t add both to SEE
> ALSO; (jmc@)
== distrib =========================================================== 02/12 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/distrib
amd64
~ ramdisk_cd/list.local
> Add zyd to cd ramdisks
> ok deraadt@ (jsg@)
i386
~ ramdisk_cd/list.local
> Add zyd to cd ramdisks
> ok deraadt@ (jsg@)
sets
~ lists/base/mi ~ lists/etc/mi
~ lists/man/mi
> Sync (miod@)
== etc =============================================================== 03/12 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/etc
Makefile
~ Makefile
> link hoststated to the builds.
> ok miod@, henning@ (pyr@)
changelist
~ changelist
> link hoststated to the builds.
> ok miod@, henning@ (pyr@)
etc.macppc/Makefile.inc
~ etc.macppc/Makefile.inc
> put boot.mac into snapshots, needed for booting old-world macs
> ok drahn@ gwk@ (martin@)
ftpusers
~ ftpusers
> link hoststated to the builds.
> ok miod@, henning@ (pyr@)
group
~ group
> link hoststated to the builds.
> ok miod@, henning@ (pyr@)
hoststated.conf
+ hoststated.conf
> link hoststated to the builds.
> ok miod@, henning@ (pyr@)
mail/aliases
~ mail/aliases
> link hoststated to the builds.
> ok miod@, henning@ (pyr@)
master.passwd
~ master.passwd
> link hoststated to the builds.
> ok miod@, henning@ (pyr@)
mklogin.conf
~ mklogin.conf
> bump DEF_OPENFILES_CUR to 128 so firefox has enough file descriptors
> for most uses. 'just do it' millert@ (kurt@)
newsyslog.conf
~ newsyslog.conf
> - enable weekly login accounting if LOGINACCOUNTING is set to 1 in the
> environment so that it matches the ROOTBACKUP and CHECKFILESYSTEMS way
> of doing things
> - make the sort command use "-k" instead of the deprecated "+" option
> (idea from mk@)
> - rotate /var/log/wtmp every week just after weekly ac(8) completes
> ok mk@ (ajacoutot@)
rc
~ rc
> link hoststated to the builds.
> ok miod@, henning@ (pyr@)
~ rc
> - add a new "accounting" variable (default to NO) to enable accouting
> (if the file /var/account/acct does not exist it will be created)
> ok mk@ (ajacoutot@)
rc.conf
~ rc.conf
> link hoststated to the builds.
> ok miod@, henning@ (pyr@)
~ rc.conf
> - add a new "accounting" variable (default to NO) to enable accouting
> (if the file /var/account/acct does not exist it will be created)
> ok mk@ (ajacoutot@)
weekly
~ weekly
> - enable weekly login accounting if LOGINACCOUNTING is set to 1 in the
> environment so that it matches the ROOTBACKUP and CHECKFILESYSTEMS way
> of doing things
> - make the sort command use "-k" instead of the deprecated "+" option
> (idea from mk@)
> - rotate /var/log/wtmp every week just after weekly ac(8) completes
> ok mk@ (ajacoutot@)
== gnu =============================================================== 04/12 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/gnu
usr.bin/perl
~ perl.h
> gcc 3.3 doesn't support __attribute__((unused)) in C++ mode. Patch
> taken from the perl devel tree. Required to build C++-based perl
> extensions. OK espie@ (who had a slightly different patch). (millert@)
~ XSUB.h
> Ignore HASATTRIBUTE_UNUSED for C++ as our g++ does not grok
> the __unused__ attribute. Also give XS functions C linkage.
> Patch taken from perl-current. (millert@)
~ pod/perlxstut.pod
> upgrade documentation so it reflects the way MakeMaker currently fills
> the test framework.
> okay millert@ (espie@)
usr.sbin/sendmail
~ cf/cf/corpse.mc ~ cf/cf/lucifier.mc
> Michael Shalayeff agreed to remove the advertising clauses from his
> licenses.
> OK Michael Shalayeff (mickey@) and millert at .ay@)
~ KNOWNBUGS ~ RELEASE_NOTES
~ cf/README ~ cf/cf/knecht.mc
~ cf/feature/access_db.m4 ~ cf/feature/dnsbl.m4
~ cf/feature/enhdnsbl.m4 ~ cf/m4/proto.m4
~ cf/m4/version.m4 ~ cf/mailer/smtp.m4
~ contrib/cidrexpand ~ doc/op/op.me
~ include/libmilter/mfapi.h ~ include/libmilter/mfdef.h
~ include/libmilter/milter.h ~ include/sm/conf.h
~ include/sm/heap.h ~ include/sm/ldap.h
~ libmilter/Makefile ~ libmilter/README
~ libmilter/comm.c ~ libmilter/engine.c
~ libmilter/handler.c ~ libmilter/libmilter.h
~ libmilter/listener.c ~ libmilter/main.c
~ libmilter/shlib_version ~ libmilter/smfi.c
~ libmilter/docs/api.html ~ libmilter/docs/design.html
~ libmilter/docs/index.html ~ libmilter/docs/installation.html
~ libmilter/docs/other.html ~ libmilter/docs/overview.html
~ libmilter/docs/sample.html ~ libmilter/docs/smfi_addheader.html
~ libmilter/docs/smfi_addrcpt.html ~ libmilter/docs/smfi_chgheader.html
~ libmilter/docs/smfi_delrcpt.html ~ libmilter/docs/smfi_getpriv.html
~ libmilter/docs/smfi_getsymval.html ~ libmilter/docs/smfi_insheader.html
~ libmilter/docs/smfi_main.html ~ libmilter/docs/smfi_opensocket.html
~ libmilter/docs/smfi_progress.html ~ libmilter/docs/smfi_quarantine.html
~ libmilter/docs/smfi_register.html
~ libmilter/docs/smfi_replacebody.html
~ libmilter/docs/smfi_setbacklog.html ~ libmilter/docs/smfi_setconn.html
~ libmilter/docs/smfi_setdbg.html ~ libmilter/docs/smfi_setmlreply.html
~ libmilter/docs/smfi_setpriv.html ~ libmilter/docs/smfi_setreply.html
~ libmilter/docs/smfi_settimeout.html ~ libmilter/docs/smfi_stop.html
~ libmilter/docs/xxfi_abort.html ~ libmilter/docs/xxfi_body.html
~ libmilter/docs/xxfi_close.html ~ libmilter/docs/xxfi_connect.html
~ libmilter/docs/xxfi_envfrom.html ~ libmilter/docs/xxfi_envrcpt.html
~ libmilter/docs/xxfi_eoh.html ~ libmilter/docs/xxfi_eom.html
~ libmilter/docs/xxfi_header.html ~ libmilter/docs/xxfi_helo.html
~ libsm/Makefile ~ libsm/exc.c
~ libsm/exc.html ~ libsm/findfp.c
~ libsm/flags.c ~ libsm/ldap.c
~ libsm/local.h ~ libsm/memstat.c
~ libsm/snprintf.c ~ libsm/t-memstat.c
~ libsm/vfscanf.c ~ mail.local/mail.local.c
~ sendmail/README ~ sendmail/TRACEFLAGS
~ sendmail/TUNING ~ sendmail/alias.c
~ sendmail/arpadate.c ~ sendmail/collect.c
~ sendmail/conf.c ~ sendmail/conf.h
~ sendmail/control.c ~ sendmail/daemon.c
~ sendmail/deliver.c ~ sendmail/domain.c
~ sendmail/envelope.c ~ sendmail/err.c
~ sendmail/headers.c ~ sendmail/macro.c
~ sendmail/main.c ~ sendmail/map.c
~ sendmail/mci.c ~ sendmail/milter.c
~ sendmail/mime.c ~ sendmail/parseaddr.c
~ sendmail/queue.c ~ sendmail/ratectrl.c
~ sendmail/readcf.c ~ sendmail/recipient.c
~ sendmail/sasl.c ~ sendmail/savemail.c
~ sendmail/sendmail.h ~ sendmail/sm_resolve.c
~ sendmail/srvrsmtp.c ~ sendmail/stab.c
~ sendmail/stats.c ~ sendmail/timers.c
~ sendmail/tls.c ~ sendmail/udb.c
~ sendmail/usersmtp.c ~ sendmail/util.c
~ sendmail/version.c ~ vacation/vacation.c
+ cf/feature/badmx.m4 + cf/feature/block_bad_helo.m4
+ cf/feature/require_rdns.m4 + include/sm/misc.h
+ include/sm/sendmail.h + include/sm/tailq.h
+ libmilter/example.c + libmilter/monitor.c
+ libmilter/worker.c
+ libmilter/docs/smfi_addrcpt_par.html
+ libmilter/docs/smfi_chgfrom.html + libmilter/docs/smfi_setsymlist.html
+ libmilter/docs/smfi_version.html + libmilter/docs/xxfi_data.html
+ libmilter/docs/xxfi_negotiate.html + libmilter/docs/xxfi_unknown.html
+ libsm/t-qic.c + libsm/util.c
+ sendmail/daemon.h + sendmail/map.h
> Update to sendmail-8.14.0. OK mbalmer@ (millert@)
== lib =============================================================== 05/12 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/lib
libc
~ stdio/tmpnam.3
> add an ENVIRONMENT section for TMPDIR;
> from freebsd pr# 108346, from Anton Yuzhaninov
> much tweaked by otto and myself; (jmc@)
~ stdio/vfprintf.c
> Remove and simplify an impossible case (if *p = memchr(cp, 0, prec),
> p - cp cannot be greater than prec).
> Prevent an integer overflow when printing a string with length
> greater than INT_MAX.
> Initial diff from millert at . > OK millert@, beck@, and otto at .ay@)
~ sys/statfs.2
> fix NAME, ok jmc@ (espie@)
libevent
~ buffer.c ~ event.c
~ event.h
> Two libevent fixes from the SVN repo:
> allow gotsig to terminate active event loop and evbuffer_find fix.
> OK brad@ pyr@ (millert@)
libpthread
~ uthread/uthread_kill.c
> Return ESRCH instead of segfaulting when a thread ID could not be found.
> ok marc@, krw@ (kettenis@)
== libexec =========================================================== 06/12 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/libexec
ld.so
~ loader.c
> revert 1.108, the diff allowed ldd to print the load address of the binary,
> however this modified variables used by C++ backtrace code which caused it
> to crash. ok kettenis (drahn@)
~ sh/rtld_machine.c
> Kludge around a horrible problem where gcc 'knows' that __udivsi3 will
> only clobber r0 and r4, and will not modify any other registers.
> Unfortunately
> calling thru the PLT clobbers other registers eg, r1, which then
> breaks because gcc doesn't reload the proper value. (drahn@)
== regress =========================================================== 07/12 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/regress
misc
- exceptions/exceptions.cc ~ exceptions/Makefile
+ exceptions/simple/Makefile + exceptions/simple/exceptions.cc
> Move test in its own directory. (kettenis@)
~ exceptions/Makefile + exceptions/foo/Makefile
+ exceptions/foo/foo.cc + exceptions/libbar/Makefile
+ exceptions/libbar/bar.cc + exceptions/libbar/shlib_version
> New test for exceptions. (kettenis@)
== sbin ============================================================== 08/12 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/sbin
dhclient
~ dhclient.c
> Don't malloc memory to store option data in leases, just appropriate
> the malloc'd memory from the parsed option buffer since it would be
> thrown away after the copy anyway.
> 'looks sane' millert@ (krw@)
~ dhcp.h
> Don't count ethernet as overhead for UDP packet constuction. Gives
> 14 more bytes of option space! A change ISC made a long time ago.
> ok stevesk@ (DHCP) canacar@ (bpf) (krw@)
init
~ init.8
> sort FILES; (jmc@)
~ init.8
> add /fastboot to FILES, since these pages discuss that file; (jmc@)
pfctl
~ pfctl.8
> document -sI -v;
> from Janne Johansson, tweaked by myself;
> ok dhartmei (jmc@)
~ parse.y
> in decide_address_family(), only limit a rule to a specific address family
> when ALL entries have this specific AF (when even just one entry has no
> specific AF, use any). found by Maurice Janssen, ok henning@ (dhartmei@)
shutdown
~ shutdown.8
> sort FILES; (jmc@)
umount
~ umount.c
> Remove #include <fstab.h> as there is no fstab in umount anymore.
> Always use defined type name. Comment clarification.
> No binary change. ok millert (bluhm@)
~ umount.8
> Remove cross references between fstab(5) and umount(8).
> ok millert jmc (bluhm@)
== share ============================================================= 09/12 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/share
man
~ man5/fstab.5
> Remove cross references between fstab(5) and umount(8).
> ok millert jmc (bluhm@)
~ man4/man4.macppc/macobio.4
> add Apple GC on the old world macs (martin@)
~ man8/rc.8
> avoid ambiguous wording; from Mark Lumsden (jmc@)
~ man8/rc.8
> add /fastboot to FILES, since these pages discuss that file; (jmc@)
~ man4/man4.landisk/Makefile + man4/man4.landisk/mem.4
+ man4/man4.landisk/obio.4
> add manpages for physical and kernel virtual memory, plus onboard I/O
> ok jmc@ (martin@)
~ man4/wdc.4
> add landisk synopsis (martin@)
~ man4/umsm.4
> Add some umsm devices found in Linux driver.
> "looks good" fkr@ (jsg@)
~ man4/man4.landisk/obio.4
> oops - no point putting stuff in SEE ALSO when we already have them
> in the supported devices list; (jmc@)
~ man4/umsm.4
> correct indent; (jmc@)
~ man8/daily.8
> - cope with recent weekly and newsyslog.conf changes
> some review by jason@, ok mk@ (ajacoutot@)
~ man5/fstab.5
> better example & description for mfs, parts from jmc, ok jmc simon
> (henning@)
~ man9/vnode.9
> Remove refrence to VLAYER, its gone.
> Diff from Iruata Souza, thanks!
> ok pedro@ (thib@)
~ man9/lock.9
> Remove references to LK_EXCLUPGRADE, LK_REENABLE, LK_SETRECURSE and
> LK_SPIN, okay miod@ (pedro@)
~ man8/daily.8
> Xr for crontab; (jmc@)
zoneinfo
~ datfiles/africa ~ datfiles/antarctica
~ datfiles/asia ~ datfiles/australasia
~ datfiles/backward ~ datfiles/europe
~ datfiles/iso3166.tab ~ datfiles/northamerica
~ datfiles/southamerica ~ datfiles/zone.tab
> Update to tzdata2007a from elsie.nci.nih.gov (millert@)
~ datfiles/africa ~ datfiles/antarctica
~ datfiles/asia ~ datfiles/australasia
~ datfiles/backward ~ datfiles/etcetera
~ datfiles/europe ~ datfiles/factory
~ datfiles/iso3166.tab ~ datfiles/leapseconds
~ datfiles/northamerica ~ datfiles/pacificnew
~ datfiles/solar87 ~ datfiles/solar88
~ datfiles/solar89 ~ datfiles/southamerica
~ datfiles/systemv ~ datfiles/yearistype.sh
~ datfiles/zone.tab
TAGGED OPENBSD_3_9
> MFC: update to tzdata2007a (millert@)
~ datfiles/africa ~ datfiles/antarctica
~ datfiles/asia ~ datfiles/australasia
~ datfiles/backward ~ datfiles/etcetera
~ datfiles/europe ~ datfiles/factory
~ datfiles/iso3166.tab ~ datfiles/leapseconds
~ datfiles/northamerica ~ datfiles/pacificnew
~ datfiles/solar87 ~ datfiles/solar88
~ datfiles/solar89 ~ datfiles/southamerica
~ datfiles/systemv ~ datfiles/yearistype.sh
~ datfiles/zone.tab
TAGGED OPENBSD_4_0
> MFC: update to tzdata2007a (millert@)
== sys =============================================================== 10/12 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/sys
arch/alpha/alpha
~ pmap.c
> Remove unused functionality from lockmgr():
> - LK_EXCLUPGRADE is never used.
> - LK_REENABLE is never used.
> - LK_SETRECURSE is never used. Because of this, the lk_recurselevel
> field is always zero, so it can be removed to.
> - the spinlock version (and LK_SPIN) is never used, since it was decided
> to use different locking structure for MP-safe protection.
> Tested by many (miod@)
~ disksubr.c
> Don't spoof an msdosfs partition if it claims to start after the end
> of the disk. Fixes the creation of spurious 'i' partitions on new iPod
> Shuffle. And other devices with no MBR, just a FAT32 filesystem.
> Pointed out most recently by Stephan A. Rickauer for his sixth
> generation, clip-on iPod shuffle.
> Fix tested by Stephan.
> ok pedro@ tom@ miod@ 'nice' deraadt@ (krw@)
arch/amd64/amd64
~ ipifuncs.c
> Rename the MICROSET IPI we inherited from NetBSD to simply NOP since
> it doesn't do anything (other than poking into the kernel a processing
> ASTs). (art@)
~ machdep.c
> When doing signotify, check if we're notifying a process on another
> cpu and if we do, send it an IPI that will force to get into the kernel
> and process the AST.
> This makes signals to processes on other CPUs to be actually delivered
> as soon as possible instead of some time in the future when some
> interupt happens.
> heavy testing by krw@, miod@ ok (in principle, since he doesn't do amd64)
> (art@)
~ pmap.c
> Remove unused functionality from lockmgr():
> - LK_EXCLUPGRADE is never used.
> - LK_REENABLE is never used.
> - LK_SETRECURSE is never used. Because of this, the lk_recurselevel
> field is always zero, so it can be removed to.
> - the spinlock version (and LK_SPIN) is never used, since it was decided
> to use different locking structure for MP-safe protection.
> Tested by many (miod@)
~ disksubr.c
> Don't spoof an msdosfs partition if it claims to start after the end
> of the disk. Fixes the creation of spurious 'i' partitions on new iPod
> Shuffle. And other devices with no MBR, just a FAT32 filesystem.
> Pointed out most recently by Stephan A. Rickauer for his sixth
> generation, clip-on iPod shuffle.
> Fix tested by Stephan.
> ok pedro@ tom@ miod@ 'nice' deraadt@ (krw@)
arch/amd64/conf
~ RAMDISK_CD
> Add zyd to cd ramdisks
> ok deraadt@ (jsg@)
arch/amd64/include
~ intrdefs.h
> Rename the MICROSET IPI we inherited from NetBSD to simply NOP since
> it doesn't do anything (other than poking into the kernel a processing
> ASTs). (art@)
~ cpu.h
> When doing signotify, check if we're notifying a process on another
> cpu and if we do, send it an IPI that will force to get into the kernel
> and process the AST.
> This makes signals to processes on other CPUs to be actually delivered
> as soon as possible instead of some time in the future when some
> interupt happens.
> heavy testing by krw@, miod@ ok (in principle, since he doesn't do amd64)
> (art@)
arch/arm/arm
~ pmap.c
> Remove unused functionality from lockmgr():
> - LK_EXCLUPGRADE is never used.
> - LK_REENABLE is never used.
> - LK_SETRECURSE is never used. Because of this, the lk_recurselevel
> field is always zero, so it can be removed to.
> - the spinlock version (and LK_SPIN) is never used, since it was decided
> to use different locking structure for MP-safe protection.
> Tested by many (miod@)
~ disksubr.c
> Don't spoof an msdosfs partition if it claims to start after the end
> of the disk. Fixes the creation of spurious 'i' partitions on new iPod
> Shuffle. And other devices with no MBR, just a FAT32 filesystem.
> Pointed out most recently by Stephan A. Rickauer for his sixth
> generation, clip-on iPod shuffle.
> Fix tested by Stephan.
> ok pedro@ tom@ miod@ 'nice' deraadt@ (krw@)
arch/aviion/aviion
~ disksubr.c
> Don't spoof an msdosfs partition if it claims to start after the end
> of the disk. Fixes the creation of spurious 'i' partitions on new iPod
> Shuffle. And other devices with no MBR, just a FAT32 filesystem.
> Pointed out most recently by Stephan A. Rickauer for his sixth
> generation, clip-on iPod shuffle.
> Fix tested by Stephan.
> ok pedro@ tom@ miod@ 'nice' deraadt@ (krw@)
arch/hp300/hp300
~ trap.c
> optimize m68k writeback():
> - it can really only be invoked from trap(), not from other userret()
> callers,
> so it is safe to hardcode its docachepush parameter to 1.
> - use pmap_kenter_pa()/pmap_kremove() for the temporary mapping instead of
> pmap_enter()/pmap_remove().
> optimize m68k userret():
> - define PROC_PC for m68k systems.
> - only check want_resched when processing T_ASTFLT traps.
> - provide two version of userret(), one which will also invoke writeback()
> on 68040 if required, which is only invoked from trap(), and regular
> userret(). This speeds up system call returns. (miod@)
arch/hp300/include
+ mutex.h
> Simple single-processor mutex implementation, simpler than the MI code by
> use of MD spl code bowels. No functional change. (miod@)
arch/hppa/hppa
~ disksubr.c
> Don't spoof an msdosfs partition if it claims to start after the end
> of the disk. Fixes the creation of spurious 'i' partitions on new iPod
> Shuffle. And other devices with no MBR, just a FAT32 filesystem.
> Pointed out most recently by Stephan A. Rickauer for his sixth
> generation, clip-on iPod shuffle.
> Fix tested by Stephan.
> ok pedro@ tom@ miod@ 'nice' deraadt@ (krw@)
arch/hppa64/hppa64
~ disksubr.c
> Don't spoof an msdosfs partition if it claims to start after the end
> of the disk. Fixes the creation of spurious 'i' partitions on new iPod
> Shuffle. And other devices with no MBR, just a FAT32 filesystem.
> Pointed out most recently by Stephan A. Rickauer for his sixth
> generation, clip-on iPod shuffle.
> Fix tested by Stephan.
> ok pedro@ tom@ miod@ 'nice' deraadt@ (krw@)
arch/i386/conf
~ RAMDISK_CD
> Add zyd to cd ramdisks
> ok deraadt@ (jsg@)
arch/i386/i386
~ mainbus.c
> Introduce acpi_hasprocfvs thats "ACPI Has Processor Frequency and Voltage
> Scalling" a global flag set by acicpu attach to indicate that ACPI is
> handling fvs through the _PSS objects. This will fix disappearing
> EST/powernow when booting kernels with acpi enabled, as acpicpu isnt
> compiled into GENERIC yet. Also in cases where acpicpu is built in and
> legacy EST/powernow work but acpi is missing the _PSS object, the legacy
> routines will be used.
> tested by dim@, and Nick Nauwelaerts <nick AT nauwelaerts.net>
> ok marco@, dim@ (gwk@)
~ machdep.c ~ pmap.c
~ pmapae.c
> Remove unused functionality from lockmgr():
> - LK_EXCLUPGRADE is never used.
> - LK_REENABLE is never used.
> - LK_SETRECURSE is never used. Because of this, the lk_recurselevel
> field is always zero, so it can be removed to.
> - the spinlock version (and LK_SPIN) is never used, since it was decided
> to use different locking structure for MP-safe protection.
> Tested by many (miod@)
~ disksubr.c
> Don't spoof an msdosfs partition if it claims to start after the end
> of the disk. Fixes the creation of spurious 'i' partitions on new iPod
> Shuffle. And other devices with no MBR, just a FAT32 filesystem.
> Pointed out most recently by Stephan A. Rickauer for his sixth
> generation, clip-on iPod shuffle.
> Fix tested by Stephan.
> ok pedro@ tom@ miod@ 'nice' deraadt@ (krw@)
arch/landisk/conf
~ GENERIC ~ RAMDISK
> There is no good reason to disable half of the 7751 cache, so don't.
> (miod@)
arch/landisk/landisk
~ disksubr.c
> Don't spoof an msdosfs partition if it claims to start after the end
> of the disk. Fixes the creation of spurious 'i' partitions on new iPod
> Shuffle. And other devices with no MBR, just a FAT32 filesystem.
> Pointed out most recently by Stephan A. Rickauer for his sixth
> generation, clip-on iPod shuffle.
> Fix tested by Stephan.
> ok pedro@ tom@ miod@ 'nice' deraadt@ (krw@)
arch/luna88k/dev
~ lcd.c
> LCD device driver support for luna88k, first step.
> ok miod@ (aoyama@)
arch/luna88k/include
~ conf.h + lcd.h
> LCD device driver support for luna88k, first step.
> ok miod@ (aoyama@)
arch/luna88k/luna88k
~ conf.c
> LCD device driver support for luna88k, first step.
> ok miod@ (aoyama@)
arch/m68k/include
~ cpu.h
> optimize m68k writeback():
> - it can really only be invoked from trap(), not from other userret()
> callers,
> so it is safe to hardcode its docachepush parameter to 1.
> - use pmap_kenter_pa()/pmap_kremove() for the temporary mapping instead of
> pmap_enter()/pmap_remove().
> optimize m68k userret():
> - define PROC_PC for m68k systems.
> - only check want_resched when processing T_ASTFLT traps.
> - provide two version of userret(), one which will also invoke writeback()
> on 68040 if required, which is only invoked from trap(), and regular
> userret(). This speeds up system call returns. (miod@)
~ _types.h + mutex.h
> Simple single-processor mutex implementation, simpler than the MI code by
> use of MD spl code bowels. No functional change. (miod@)
arch/m68k/m68k
~ m68k_machdep.c
> optimize m68k writeback():
> - it can really only be invoked from trap(), not from other userret()
> callers,
> so it is safe to hardcode its docachepush parameter to 1.
> - use pmap_kenter_pa()/pmap_kremove() for the temporary mapping instead of
> pmap_enter()/pmap_remove().
> optimize m68k userret():
> - define PROC_PC for m68k systems.
> - only check want_resched when processing T_ASTFLT traps.
> - provide two version of userret(), one which will also invoke writeback()
> on 68040 if required, which is only invoked from trap(), and regular
> userret(). This speeds up system call returns. (miod@)
+ mutex.c
> Simple single-processor mutex implementation, simpler than the MI code by
> use of MD spl code bowels. No functional change. (miod@)
arch/mac68k/include
+ mutex.h
> Simple single-processor mutex implementation, simpler than the MI code by
> use of MD spl code bowels. No functional change. (miod@)
arch/mac68k/mac68k
~ trap.c
> optimize m68k writeback():
> - it can really only be invoked from trap(), not from other userret()
> callers,
> so it is safe to hardcode its docachepush parameter to 1.
> - use pmap_kenter_pa()/pmap_kremove() for the temporary mapping instead of
> pmap_enter()/pmap_remove().
> optimize m68k userret():
> - define PROC_PC for m68k systems.
> - only check want_resched when processing T_ASTFLT traps.
> - provide two version of userret(), one which will also invoke writeback()
> on 68040 if required, which is only invoked from trap(), and regular
> userret(). This speeds up system call returns. (miod@)
arch/macppc/macppc
~ disksubr.c
> Don't spoof an msdosfs partition if it claims to start after the end
> of the disk. Fixes the creation of spurious 'i' partitions on new iPod
> Shuffle. And other devices with no MBR, just a FAT32 filesystem.
> Pointed out most recently by Stephan A. Rickauer for his sixth
> generation, clip-on iPod shuffle.
> Fix tested by Stephan.
> ok pedro@ tom@ miod@ 'nice' deraadt@ (krw@)
arch/mips64/mips64
~ disksubr.c
> Don't spoof an msdosfs partition if it claims to start after the end
> of the disk. Fixes the creation of spurious 'i' partitions on new iPod
> Shuffle. And other devices with no MBR, just a FAT32 filesystem.
> Pointed out most recently by Stephan A. Rickauer for his sixth
> generation, clip-on iPod shuffle.
> Fix tested by Stephan.
> ok pedro@ tom@ miod@ 'nice' deraadt@ (krw@)
arch/mvme68k/include
+ mutex.h
> Simple single-processor mutex implementation, simpler than the MI code by
> use of MD spl code bowels. No functional change. (miod@)
arch/mvme68k/mvme68k
~ trap.c
> optimize m68k writeback():
> - it can really only be invoked from trap(), not from other userret()
> callers,
> so it is safe to hardcode its docachepush parameter to 1.
> - use pmap_kenter_pa()/pmap_kremove() for the temporary mapping instead of
> pmap_enter()/pmap_remove().
> optimize m68k userret():
> - define PROC_PC for m68k systems.
> - only check want_resched when processing T_ASTFLT traps.
> - provide two version of userret(), one which will also invoke writeback()
> on 68040 if required, which is only invoked from trap(), and regular
> userret(). This speeds up system call returns. (miod@)
arch/mvmeppc/mvmeppc
~ disksubr.c
> Don't spoof an msdosfs partition if it claims to start after the end
> of the disk. Fixes the creation of spurious 'i' partitions on new iPod
> Shuffle. And other devices with no MBR, just a FAT32 filesystem.
> Pointed out most recently by Stephan A. Rickauer for his sixth
> generation, clip-on iPod shuffle.
> Fix tested by Stephan.
> ok pedro@ tom@ miod@ 'nice' deraadt@ (krw@)
arch/sparc/conf
~ files.sparc
> Simple single-processor mutex implementation, simpler than the MI code by
> use of MD spl code bowels. No functional change. (miod@)
arch/sparc/include
~ _types.h + mutex.h
> Simple single-processor mutex implementation, simpler than the MI code by
> use of MD spl code bowels. No functional change. (miod@)
arch/sparc/sparc
+ mutex.c
> Simple single-processor mutex implementation, simpler than the MI code by
> use of MD spl code bowels. No functional change. (miod@)
arch/sparc64/conf
~ files.sparc64
> Simple single-processor mutex implementation, simpler than the MI code by
> use of MD spl code bowels. No functional change. (miod@)
arch/sparc64/include
~ _types.h + mutex.h
> Simple single-processor mutex implementation, simpler than the MI code by
> use of MD spl code bowels. No functional change. (miod@)
arch/sparc64/sparc64
+ mutex.c
> Simple single-processor mutex implementation, simpler than the MI code by
> use of MD spl code bowels. No functional change. (miod@)
arch/vax/include
~ _types.h + mutex.h
> Simple single-processor mutex implementation, simpler than the MI code by
> use of MD spl code bowels. No functional change. (miod@)
arch/vax/vax
+ mutex.c
> Simple single-processor mutex implementation, simpler than the MI code by
> use of MD spl code bowels. No functional change. (miod@)
dev
~ vnd.c
> Don't throttle and serialize I/O operations, okay miod@ krw@
> Fixes kern/5347 (pedro@)
dev/acpi
~ acpimadt.c
> Properly route lapic NMIs. (kettenis@)
~ acpimadt.c
> Fix pasto. (kettenis@)
~ acpi.c ~ acpivar.h
~ acpicpu.c
> Introduce acpi_hasprocfvs thats "ACPI Has Processor Frequency and Voltage
> Scalling" a global flag set by acicpu attach to indicate that ACPI is
> handling fvs through the _PSS objects. This will fix disappearing
> EST/powernow when booting kernels with acpi enabled, as acpicpu isnt
> compiled into GENERIC yet. Also in cases where acpicpu is built in and
> legacy EST/powernow work but acpi is missing the _PSS object, the legacy
> routines will be used.
> tested by dim@, and Nick Nauwelaerts <nick AT nauwelaerts.net>
> ok marco@, dim@ (gwk@)
~ acpidock.c
> Change sensor description to `docked'/`not docked' depending on actual
> state. This is more in line with acpiac and acpibat, and it confuses me
> less when I look at the sensor output. (mk@)
dev/ic
~ pckbc.c ~ pckbcvar.h
> run a timeout (calling intr handler) to recover from in buffer oflows that
> make no ints; w/ help from miod@ and ok miod@ (mickey@)
~ pckbc.c
> oops. now is the miodski version of the diff (; (mickey@)
~ rtl81x9.c
> up the accepted length to ETHER_MAX_LEN + ETHER_VLAN_ENCAP_LEN on rx; from
> brad@ (jason@)
~ re.c
> Don't set the RL_CFG1_FULLDUPLEX bit. The RL_CFG1_FULLDUPLEX bit in
> config register 1 is only valid with the 8129 chipset.
> The rtl81x9reg.h header has this bit marked with a comment indicating
> it is for the 8129 chipset and the 8168/8169 datasheets confirm that
> this bit is not valid for the newer chipsets.
> From yongari at FreeBSDa brad. Tested by brad, kettenis and otto. (krw@)
dev/mii
~ brgphy.c ~ brgphyreg.h
> Rough in more support for 5787/5755 chips so far known to not work or
> unreported in the wild. In this case add a PHY workaround for an eventual
> mobile version of the chipset. No change to existing functionality.
> From Michael Chan (mchan at broadcomvia Linux tg3 and brad. (krw@)
dev/pci
~ vga_pci.c
> Fix a number of issues with vesabios discovered by miod@
> 1. Add bounds checking to vga_pci_mmap, fixing a potential security issue.
> Limit the amount of memory to what the vbe info structure says the card
> card has. This should be nicely refined to limit to the amount of memory
> needed by the active resolution and depth but this will do for the time
> being.
> 2. Fix several places in vesafb.c where the return of kvm86_bios_addpage()
> isnt being checked. Also ensure that if vesafb_get_mode_info() fails that
> it
> cleans up after itself by releasing the page it added.
> 3. Correct the range checks in vesafb_putcmap and vesafb_getcmap, harmonize
> code with similar code found else where.
> ok miod@ (gwk@)
~ pcidevs
> Add nVIDIA GeForce 7800 GS. From Jim Razmus. (kettenis@)
~ pcidevs.h ~ pcidevs_data.h
> regen (kettenis@)
~ if_bge.c ~ if_bgereg.h
> Rough in more support for 5787/5755 chips so far known to not work or
> unreported in the wild. In this case add a PHY workaround for an eventual
> mobile version of the chipset. No change to existing functionality.
> From Michael Chan (mchan at broadcomvia Linux tg3 and brad. (krw@)
~ if_bnx.c
> Allow the bnx(4) driver to make use of all of the available hardware
> multicast hash slots. The bnx(4) hardware supports 8 slots instead of
> 4 like the bge(4) hardware.
> From Mike Karels via FreeBSD
> Tested by Brad, biorn@ and Johan M:son Lindman (krw@)
~ if_vic.c
> update vic(4) to use the LINK_STATE_IS_UP() macro (reyk@)
~ agp.c
> Simplify agp_generic_bind_memory by not trying to allocate big chunks
> of contiguous memory in a loop. This should save contiguous memory
> for other drivers, and AGP doesn't need it.
> Tested by many, ok miod (dim@)
~ ahci.c
> macros describing a ports sata status register (dlg@)
~ if_sk.c ~ if_msk.c
> Fix typo; clear tx FIFO underrun IRQ as intended instead of rx FIFO overrun
> IRQ. From brad@ (kettenis@)
dev/pckbc
~ pckbd.c
> default to have pckbd_set_xtscancode() instruct the 8042 to use
> table 3, still falling back on table 2 or 1 after carefully checking
> that the table set request took
> fixes the keyboard on the oqo model 01/01+
> in snaps for a while, tested by many
> ok miod@ (jcs@)
dev/sdmmc
~ sdhc.c
> Restore spl level before exiting on error.
> Noted by Matthew R. Dempsky. (krw@)
~ sdhc.c
> Workaround for the problem where the SD card reader stops responding after
> a card is ejected. Problem identified and patch provided by
> Matthew R. Dempsky < mrd at alkemio dot org >:
> > According to the Simplified SD Host Controller spec, this change
> > shouldn't affect anything. All of the bits in the Software Reset
> > register are labeled as ignoring 0-bit writes, and yet somehow it
> > changes behavior.
> This seems to reliably fix the problems seen on many X40 laptops.
> Tested by many, OK beck@, "should be good to go" henning@ (claudio@)
dev/usb
~ usbdevs
> Add some additional umsm products found in Linux driver. (jsg@)
~ usbdevs_data.h
> regen (jsg@)
~ umsm.c
> Add some umsm devices found in Linux driver.
> "looks good" fkr@ (jsg@)
~ usbdevs.h
> Oops, missed this in the regen commit.
> Pointed out by pedro@ (jsg@)
dev/vesa
~ vesabiosvar.h ~ vesabios.c
~ vesafb.c
> Fix a number of issues with vesabios discovered by miod@
> 1. Add bounds checking to vga_pci_mmap, fixing a potential security issue.
> Limit the amount of memory to what the vbe info structure says the card
> card has. This should be nicely refined to limit to the amount of memory
> needed by the active resolution and depth but this will do for the time
> being.
> 2. Fix several places in vesafb.c where the return of kvm86_bios_addpage()
> isnt being checked. Also ensure that if vesafb_get_mode_info() fails that
> it
> cleans up after itself by releasing the page it added.
> 3. Correct the range checks in vesafb_putcmap and vesafb_getcmap, harmonize
> code with similar code found else where.
> ok miod@ (gwk@)
kern
~ kern_lock.c
> Remove unused functionality from lockmgr():
> - LK_EXCLUPGRADE is never used.
> - LK_REENABLE is never used.
> - LK_SETRECURSE is never used. Because of this, the lk_recurselevel
> field is always zero, so it can be removed to.
> - the spinlock version (and LK_SPIN) is never used, since it was decided
> to use different locking structure for MP-safe protection.
> Tested by many (miod@)
net
~ if_trunk.c ~ if_trunk.h
> handle the full duplex link state in trunk(4). load sharing trunks
> with at least two ports are always handled as full fuplex links. this
> change will allow trunks as edge ports in a rstp bridge(4).
> ok brad@ pyr@ (reyk@)
netinet
~ tcp_var.h
> correct rfc; from Kris Katterjohn (jmc@)
scsi
~ cd.c ~ sd.c
> Make comment match code. ok pedro (bluhm@)
sys
~ exec.h
> Bump fixed part of the stackgap to 2k. Allows two full paths to be
> stored, needed by systrace. ok deraadt@ miod@ (otto@)
~ lock.h ~ sched.h
> Remove unused functionality from lockmgr():
> - LK_EXCLUPGRADE is never used.
> - LK_REENABLE is never used.
> - LK_SETRECURSE is never used. Because of this, the lk_recurselevel
> field is always zero, so it can be removed to.
> - the spinlock version (and LK_SPIN) is never used, since it was decided
> to use different locking structure for MP-safe protection.
> Tested by many (miod@)
ufs/ffs
~ ffs_softdep.c
> Correctly fake dinode for the FFS2 case in handle_workitem_freeblocks()
> Okay millert@ (pedro@)
== usr.bin =========================================================== 11/12 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/usr.bin
awk
~ main.c
> sync usage(); spotted by Igor Sobrado (jmc@)
~ awk.1
> sync the -f description w/ its synopsis; (jmc@)
bc
~ bc.library
> whitespace, no change in code produced by bc -dl (otto@)
~ bc.library
> Save and restore not only scale, but ibase as well. Problem noted my
> Karel Kulhavy. (otto@)
compress
~ gzip.1
> Document $GZIP environment variable.
> OK jmc@ and millert at .ay@)
cvs
~ checkout.c ~ client.c
~ diff.h ~ diff3.c
~ file.c ~ update.c
> add merging support in both local and remote sides.
> tested by many, thanks. (joris@)
~ file.c
> remove debug printfs that sneaked in. (joris@)
~ update.c
> do not overwrite a file that is modified but has a sticky tag set
> when running update -A, instead merge in the changes.
> 'kay xsa@ (joris@)
~ checkout.c
> hey look, i found another debug printf from my merging stuff!
> kill-zap-nuke-remove-delete it. (joris@)
~ file.c
> no longer spit out that you can add new files using 'cvs add',
> it is pretty obvious on itself.
> ok niallo@ xsa@ (joris@)
~ xmalloc.c ~ xmalloc.h
> introduce xsnprintf(); OK otto@ joris at .sa@)
~ add.c ~ checkout.c
~ client.c ~ commit.c
~ diff3.c ~ file.c
~ import.c ~ log.c
~ remote.c ~ remove.c
~ server.c ~ status.c
~ update.c
> snprintf() -> xsnprintf() (xsa@)
~ admin.c
> add support for [-k mode]; tests/input/ok otto at .sa@)
~ import.c
> add support for [-k mode]; tests/ok otto at .sa@)
~ commit.c
> change mode of a newly created rcs file in the repository to 0444
> ok jorix@ xsa@ (otto@)
~ rcs.c
> When writing an rcs file, create the temp file next to the target
> file and then rename. Avoids inter-filesystem copy and fixes a
> potential rcs file truncation when the target fileystem becomes full.
> While there, use fchmod instead of chmod. ok xsa@ joris@ (otto@)
~ buf.c ~ buf.h
> Remove unused and buggy {cvs,rcs}_buf_set functions.
> Bugginess pointed out and fix provided by Pierre Riteau.
> OK joris at .ay@)
~ buf.c
> Remove cb_cur or replace with cb_buf; cb_cur was the same as cb_buf
> most of the time and was never of much use.
> OK niallo at .ay@)
~ entries.c
> dont bother trying to parse empty lines as dates.
> cuts away some cpu time.
> from Tobias Ulmer, thanks a million! (joris@)
~ rcs.c
> Use the right HEADNAME. Sometimes compiler warnings should not be
> ignored. ok xsa@ joris@ (otto@)
ftp
~ fetch.c
> Plug memory leak.
> OK otto at .ay@)
hexdump
~ display.c
> In doskip(), do not access an uninitialized variable in the statok == 0
> case
> (introduced in r1.9) (miod@)
~ hexdump.1 ~ hexdump.c
~ hexdump.h ~ hexsyntax.c
> Accept a leading 0x for the -n length parameter.
> Correctly check for out-of-bounds values for -s skip parameter, and use
> strtoll to parse it instead of strtol since this is an off_t value. (miod@)
pkg-config
~ pkg-config
> handle errors reading .pc files in --list gracefully. (espie@)
rcs
~ co.c
> When checking to see if a checked-out copy of the file is the same
> as what is in the repo, expand keywords before doing the diff.
> OK xsa@ (millert@)
~ buf.c ~ buf.h
> Remove unused and buggy {cvs,rcs}_buf_set functions.
> Bugginess pointed out and fix provided by Pierre Riteau.
> OK joris at .ay@)
~ buf.c
> Remove cb_cur or replace with cb_buf; cb_cur was the same as cb_buf
> most of the time and was never of much use.
> OK niallo at .ay@)
top
~ machine.c
> long and long long is not the same on 32 bits archs (otto@)
~ top.c
> actully use a default value when we say so. from Mark Lumsden (otto@)
~ commands.c ~ top.1
~ top.c
> Introducing the '+' interactive command to reset all filters.
> From Mark Lumsden. ok millert@ deraadt@ simon@ (otto@)
~ top.c
> Complain of the user specifies a negative number for the 's' command.
> From Mark Lumsden. (otto@)
== usr.sbin ========================================================== 12/12 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/usr.sbin
usr.sbin
~ Makefile
> link hoststated to the builds.
> ok miod@, henning@ (pyr@)
ac
~ ac.8
> - cope with recent weekly and newsyslog.conf changes
> some review by jason@, ok mk@ (ajacoutot@)
~ ac.8
> point to weekly(8);
> ok/help ajacoutot (jmc@)
~ ac.8
> use .Ex; (jmc@)
accton
~ accton.8
> document accounting=; ok ajacoutot (jmc@)
bgpctl
~ parser.c
> Switch from strtoul() to strtonum() making the code easier to read.
> OK henning@ (claudio@)
~ parser.c
> Correct function name in err() message. (claudio@)
~ bgpctl.c
> Print neighbor templates with a state of Template. Makes the difference
> between templates and real sessions more obvious. OK henning@ (claudio@)
bgpd
~ config.c
> Replace another strtol() with strtonum() making the error checking easier
> to understand. This is modeled after bgpctl's parse_prefix function.
> OK henning@ (claudio@)
~ rde_update.c
> Do not check that the aspath is loopfree for the peer when sending updates.
> This is an undocumented optimisation that fails with the introduction of
> AS 23456 as placeholder for 4-byte ASN.
> OK henning@ (claudio@)
dhcpd
~ dhcp.h
> Don't count ethernet as overhead for UDP packet constuction. Gives
> 14 more bytes of option space! A change ISC made a long time ago.
> ok stevesk@ (DHCP) canacar@ (bpf) (krw@)
dvmrpd
~ rde_srt.c
> Introduce new checks against the received metric to update or delete the
> route.
> OK norby@ (michele@)
hoststatectl
~ parser.c
> Correctly check for missing arguments, as in ospfctl/bgpctl. OK pyr@
> (claudio@)
~ hoststatectl.c ~ parser.c
> Add SSL support to hoststated.
> with help and OK reyk@
> with help and advice by claudio@ and Srebrenko Sehic (pyr@)
~ hoststatectl.8 ~ hoststatectl.c
~ parser.c ~ parser.h
> add a monitor mode to hoststatectl to continuously report changes in
> hoststated.
> ok reyk@, "looks nice and clean" niallo@ (pyr@)
~ hoststatectl.c
> modify the imsg monitor to look even nicer and to use a more flexible
> API (inspired by the ipsec pfkey monitor).
> ok pyr@ niallo@ (reyk@)
~ hoststatectl.c
> print id name in the monitor output if available (reyk@)
~ hoststatectl.c
> wait for hoststated to become available in monitor mode.
> ok pyr@ (reyk@)
hoststated
~ Makefile ~ buffer.c
~ check_http.c ~ check_icmp.c
~ check_send_expect.c ~ check_tcp.c
~ control.c ~ hce.c
~ hoststated.c ~ hoststated.conf.5
~ hoststated.h ~ imsg.c
~ parse.y ~ pfe.c
~ pfe_filter.c + ssl.c
> Add SSL support to hoststated.
> with help and OK reyk@
> with help and advice by claudio@ and Srebrenko Sehic (pyr@)
~ hoststated.conf.5
> manpage tweaks.
> advised by and ok jmc@ (pyr@)
~ check_tcp.c
> style changes.
> ok reyk@ (pyr@)
~ parse.y
> small memleak plugged and style changes.
> ok reyk@ (pyr@)
~ ssl.c
> provide more accurate error messages.
> ok reyk@ (pyr@)
~ Makefile
> this diff changes "-lcrypto -lssl" to "-lssl -lcrypto" to let it
> compile on vax. on some architectures the order of the libraries
> matters for the linker.
> ok pyr@ (reyk@)
~ control.c ~ hoststated.h
~ pfe.c
> add a monitor mode to hoststatectl to continuously report changes in
> hoststated.
> ok reyk@, "looks nice and clean" niallo@ (pyr@)
~ check_tcp.c
> check_tcp.c (reyk@)
- check_http.c - check_send_expect.c
~ Makefile ~ check_tcp.c
> merge tcp-based checks into one file, no functional changes
> ok pyr@ (reyk@)
httpd
~ src/support/apachectl
> apachectl picks up httpd_flags from rc.conf for some time now - and that
> is good! so let's do it in all cases...
> missed case: on restart and graceful, when apachectl figures out that no
> httpd is running, it tries to start one.
> found out the hard way by yours truly (henning@)
~ src/Configure ~ src/ap/ap_snprintf.c
~ src/lib/expat-lite/xmldef.h
> remove varying parts from generated configure info:
> - ap_snprintf can grab needed types from stdint.h
> - expat-lite can grab byte-order from system includes.
> no breakage in modules in the ports tree.
> work by me and millert@, ok miod at .spie@)
ospfctl
~ parser.c
> Enable reload support in ospfctl. This is still a bit experimental so try
> it out in your test lab before using it on the core infrastructure.
> (claudio@)
ospfd
~ rde.c ~ rde.h
~ rde_lsdb.c ~ rde_spf.c
> Do not link from the LS DB to outside structures that may vanish before
> the actual LS DB entry is removed. In particular a neighbor may be removed
> at any time -- we were lucky because we kept down neighbors around for
> another 24h. Reload support unhided this problem again. Just copy the
> needed info into the vertex. Pointing to the area is save as the vertex is
> part of the area itself and removed when the area is removed.
> OK norby@ (claudio@)
~ rde_spf.c
> Remove useless debug message when calculating the SPF tree.
> ok claudio@ (norby@)
~ auth.c ~ ospfe.h
> Implement md_list_send() function to send auth crypt keys to the OSPF
> engine.
> Needed for reload support. OK norby@ pyr@ (claudio@)
~ rde.c
> There is no need to check if the LSA is matched by an net LSA in
> rde_asext_put(). Just try to remove the as-ext LSA, if the prefix is
> announced as stub net LSA lsa_find() will fail later and nothing will
> happen. OK norby@ pyr@ (claudio@)
~ kroute.c ~ ospfd.h
> kr_reload() -- another helper function for reload support. This one
> takes care that redistribution is correct after reload. OK norby@ pyr@
> (claudio@)
~ interface.c
> If an interface is going down send out an empty OSPF hello packet so that
> all other routers will drop the session immediatly instead of waiting for
> the timeout that is normaly 40 sec. This reduces fail-over time massivly
> and does not hurt anymore since we are now able to reload the configuration
> on the fly. OK norby@ pyr@ (claudio@)
~ ospfd.c ~ ospfd.h
~ ospfe.c ~ rde.c
> Preliminary configuration reload support. Seems to work reasonably well
> after fixing many use-after-free errors a couple of days ago.
> Most cases -- adding/removing of interfaces, adding/removing of areas --
> seem to work. Still merge_config() is a behemoth of a function and some
> edge cases may not be covered correctly (yet).
> OK norby@ pyr@ (claudio@)
~ auth.c ~ interface.c
~ ospfd.c ~ parse.y
~ rde.c
> whitespace cleanup, no binary change. (claudio@)
pkg_add
~ pkg_info ~ pkg_info.1
> Three lines in SYNOPSIS are too much. Use only one line and sync usage.
> prodded by & ok jmc@ (bernd@)
~ OpenBSD/Dependencies.pm ~ OpenBSD/Interactive.pm
> streamline messages for interactive mode, no need to repeat the same list
> twice. (espie@)
~ Makefile ~ pkg_add
> protect `normal' installs against archive burps. (espie@)
~ Makefile
> back-out extra commit (espie@)
sa
~ sa.8
> document accton(8); ok ajacoutot (jmc@)
~ sa.8
> use .Ex; (jmc@)
vnconfig
~ vnconfig.c
> Make -c, -l and -u mutually exclusive, they indicate the mode of operation.
> Point out that using -S with anything else than -K is useless.
> "looks good" pedro@, no objections otto@ (grunk@)
===============================================================================
More information about the owc
mailing list