[owc] Weekly src changes ending 2007-02-25
OWC
auto at squish.net
Mon Feb 26 07:08:00 GMT 2007
OpenBSD src changes summary for 2007-02-18 to 2007-02-25 inclusive
==================================================================
bin/chmod bin/ed
bin/pax bin/test
distrib/crunch distrib/landisk
distrib/sets distrib/sparc64
etc/Makefile etc/crontab
etc/etc.sparc64/disktab etc/pf.conf
etc/rc.conf etc/root/root.mail
etc/sasyncd.conf etc/spamd.conf
games/hack gnu/usr.bin/perl
lib/libc lib/libkvm
lib/libpthread libexec/spamd
libexec/spamd-setup regress/sbin
regress/usr.bin sbin/ccdconfig
sbin/dhclient sbin/disklabel
sbin/dump sbin/fsdb
sbin/fsirand sbin/ipsecctl
sbin/isakmpd sbin/modload
sbin/mount sbin/mount_nfs
sbin/pdisk sbin/pfctl
sbin/quotacheck sbin/raidctl
sbin/restore sbin/routed
sbin/sysctl share/man
sys/arch/alpha/alpha sys/arch/alpha/conf
sys/arch/alpha/include sys/arch/amd64/amd64
sys/arch/amd64/conf sys/arch/amd64/include
sys/arch/arm/arm sys/arch/arm/include
sys/arch/arm/xscale sys/arch/armish/include
sys/arch/aviion/aviion sys/arch/aviion/include
sys/arch/cats/include sys/arch/hp300/dev
sys/arch/hp300/include sys/arch/hppa/hppa
sys/arch/hppa/include sys/arch/hppa64/hppa64
sys/arch/hppa64/include sys/arch/i386/conf
sys/arch/i386/i386 sys/arch/i386/include
sys/arch/i386/pci sys/arch/landisk/include
sys/arch/landisk/landisk sys/arch/luna88k/include
sys/arch/m68k/include sys/arch/m88k/include
sys/arch/mac68k/include sys/arch/macppc/dev
sys/arch/macppc/include sys/arch/macppc/macppc
sys/arch/mips64/include sys/arch/mips64/mips64
sys/arch/mvme68k/include sys/arch/mvme88k/include
sys/arch/mvmeppc/include sys/arch/mvmeppc/mvmeppc
sys/arch/powerpc/include sys/arch/powerpc/powerpc
sys/arch/sgi/include sys/arch/sh/include
sys/arch/solbourne/include sys/arch/sparc/dev
sys/arch/sparc/include sys/arch/sparc64/conf
sys/arch/sparc64/dev sys/arch/sparc64/include
sys/arch/vax/include sys/arch/vax/vax
sys/arch/zaurus/include sys/conf
sys/crypto sys/dev
sys/dev/acpi sys/dev/ata
sys/dev/cardbus sys/dev/ic
sys/dev/pci sys/dev/sbus
sys/dev/usb sys/dev/vesa
sys/kern sys/net
sys/netinet sys/nfs
sys/scsi sys/stand
sys/sys sys/uvm
usr.bin/biff usr.bin/cmp
usr.bin/col usr.bin/cvs
usr.bin/dc usr.bin/diff
usr.bin/file usr.bin/from
usr.bin/locate usr.bin/mail
usr.bin/make usr.bin/mg
usr.bin/nc usr.bin/pkg-config
usr.bin/pr usr.bin/rcs
usr.bin/sdiff usr.bin/ssh
usr.bin/sudo usr.bin/systat
usr.bin/tip usr.bin/top
usr.bin/tset usr.bin/units
usr.bin/usbhidaction usr.bin/vacation
usr.sbin/acpidump usr.sbin/amd
usr.sbin/arp usr.sbin/authpf
usr.sbin/bgpctl usr.sbin/bgpd
usr.sbin/bind usr.sbin/cron
usr.sbin/dhcpd usr.sbin/fdformat
usr.sbin/hoststatectl usr.sbin/hoststated
usr.sbin/httpd usr.sbin/iostat
usr.sbin/kgmon usr.sbin/map-mbone
usr.sbin/memconfig usr.sbin/mrouted
usr.sbin/mtrace usr.sbin/ndp
usr.sbin/ntpd usr.sbin/ospfd
usr.sbin/pkg_add usr.sbin/pppd
usr.sbin/pwd_mkdb usr.sbin/quotaon
usr.sbin/repquota usr.sbin/rpc.bootparamd
usr.sbin/sensorsd usr.sbin/spamdb
usr.sbin/syslogd usr.sbin/tcpdump
usr.sbin/timed usr.sbin/trpt
usr.sbin/user usr.sbin/vnconfig
usr.sbin/ypbind usr.sbin/yppoll
usr.sbin/ypserv usr.sbin/ypset
usr.sbin/ztsscale
== bin =============================================================== 01/13 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/bin
chmod
~ chflags.1
> document the "opaque", "arch", and "nodump" flags more fully;
> closes documentation/5384 from Andrew Hamilton-Wright
> help from henning, otto, and weingart (jmc@)
~ chflags.1
> shuffle the text somewhat to make the page read better;
> use .Ex; (jmc@)
ed
~ ed.h
> There is no filter_lines function, remove its prototype. From Mark
> Lumsden. (millert@)
pax
~ cpio.1 ~ options.c
> sort options; sync usage(); from Igor Sobrado (jmc@)
~ cpio.1
> -F was missing a description; (jmc@)
~ pax.1 ~ options.c
> sort options; sync usage(); from Igor Sobrado (jmc@)
test
~ test.1
> order options; from Igor Sobrado (jmc@)
== distrib =========================================================== 02/13 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/distrib
crunch
~ crunchide/crunchide.c
> Use sizeof(buf) instead of hard coded numbers.
> From Charles Longeau <chl at tuxfamily dot org> long ago.
> OK moritz at .ay@)
landisk
~ miniroot/Makefile
> how about we use a release-specific output filename.... (deraadt@)
sets
~ 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.landisk
~ 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/man/mi
> sync (deraadt@)
~ 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.landisk
~ 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
> sync (deraadt@)
~ lists/etc/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.landisk
~ 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@)
sparc64
~ miniroot/Makefile
> crank size of sparc64 miniroot a wee bit (deraadt@)
== etc =============================================================== 03/13 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/etc
Makefile
~ Makefile
> sample sasyncd.conf, PR5387 sthen at symphytum.spacehopper.org, ok theo
> (henning@)
crontab
~ crontab
> Make greylisting the default when spamd is enabled. Uses the new -g flag
> for spamd-setup. OK beck@ (millert@)
etc.sparc64/disktab
~ etc.sparc64/disktab
> crank size of sparc64 miniroot a wee bit (deraadt@)
pf.conf
~ pf.conf
> Make greylisting the default when spamd is enabled. Uses the new -g flag
> for spamd-setup. OK beck@ (millert@)
rc.conf
~ rc.conf
> Make greylisting the default when spamd is enabled. Uses the new -g flag
> for spamd-setup. OK beck@ (millert@)
root/root.mail
~ root/root.mail
> we also want GENERIC.MP dmesgs to be submitted.
> ok deraadt@ (steven@)
sasyncd.conf
+ sasyncd.conf
> sample sasyncd.conf, PR5387 sthen at symphytum.spacehopper.org, ok theo
> (henning@)
spamd.conf
~ spamd.conf
> in in, spotted by toby (beck@)
~ spamd.conf
> spaces -> tab; "of course" deraadt@ (marc@)
== games ============================================================= 04/13 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/games
hack
~ hack.pager.c
> - Pass full buffer size to fgets.
> - No need to initialize buffer.
> - Spacing.
> Based on diff from Charles Longeau <chl at tuxfamily dot org> long ago.
> OK moritz at .ay@)
== gnu =============================================================== 05/13 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/gnu
usr.bin/perl
~ ext/Digest/MD5/MD5.xs
> replace 400 lines of md5 code in Digest::MD5's XS with 4 lines
> that use our system md5(3)
> idea by espie@, help and ok millert@ (simon@)
== lib =============================================================== 06/13 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/lib
libc
~ net/getaddrinfo.c
> strlen(3) returns size_t.
> OK itojun at .ay@)
~ net/gethostnamadr.c
> strlen(3) returns size_t, not int.
> Suggested by itojun@ in response to my getaddrinfo fixes.
> OK millert at .ay@)
~ string/strxfrm.3
> from hubertf at netbsd > Try to tell what this really does.
> Including a BUGS section on why it doesn't.
> *roff code by wiz@, with some input from
> Slava Semushin <slava.semushin at gmail.com > ok millert (jmc@)
~ string/strxfrm.3
> wording fix from wiz at netbsdr1.11; (jmc@)
~ crypt/bcrypt.c
> Change hard coded numbers to sizeof(buf). Also change some
> sizeof(buf) - 1 to sizeof(buf), since fgets takes the whole buffer size.
> Based on diff from Charles Longeau <chl at tuxfamily dot org> long ago.
> OK millert at .ay@)
~ rpc/getrpcent.c
> Change hard coded numbers to sizeof(buf). Also change some
> sizeof(buf) - 1 to sizeof(buf), since fgets takes the whole buffer size.
> Based on diff from Charles Longeau <chl at tuxfamily dot org> long ago.
> OK millert at .ay@)
~ gen/sysctl.3
> net.inet.esp.udpencap is enabled by default;
> from jared rr spiegel: documentation/5385 (jmc@)
libkvm
~ kvm_i386.c ~ shlib_version
> Revert PAE pmap for now, until the strange bug is found. This stops
> the freezes many of us are seeing (especially on amd64 machines running
> OpenBSD/i386).
> Much testing by nick@ (as always - thanks!), hugh@, ian@, kettenis@
> and Sam Smith (s (at) msmith (dot) net).
> Requested by, input from, and ok deraadt@ ok art@, kettenis@, miod@ (tom@)
libpthread
~ arch/sh/_atomic_lock.c ~ arch/sh/uthread_machdep.h
+ arch/sh/uthread_machdep.c + arch/sh/uthread_machdep_asm.S
> pthread bits for sh; help deraadt@, ok kettenis@ (miod@)
== libexec =========================================================== 07/13 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/libexec
spamd
~ grey.c ~ grey.h
~ spamd.8 ~ spamd.c
> greytrapping improvements
> 1) remove requirement for <> around spamtrap addresses
> 2) add support for /etc/spamd/alloweddomains to specify
> suffixes for which any destinations that don't match
> get trapped
> various knf's by theo, feedback from jmc, millert, deraadt (beck@)
~ grey.c
> spacing (deraadt@)
~ grey.c ~ sdl.c
~ spamd.c
> continue collecting lint to be sold as a parsley substitute; ok beck
> (deraadt@)
~ grey.c
> make dequotetolower use regular string functions instead of the manual
> walk with mutiple conditions in the loop.
> ok deraadt@ (beck@)
~ grey.c ~ spamd.c
> Make spamd include the HELO/EHLO identification string sent by
> the connecting hosts in the tuple key when greylisting. catches
> a few more bogus hosts and will let us trap based on HELO later.
> Changes spamdb(8) output to include the new field.
> ok deraadt@, jmc@ (beck@)
~ spamd.8
> to_suffixes -> alloweddomains, that got missed, noticed
> by okan at demirmen.comeck@)
~ spamd.8
> tweaks; (jmc@)
~ spamd.8
> When greylisting we don't actually need to use the <spamd> pf table.
> We just do no-rdr for things in <spamd-white> and rdr the rest to spamdb.
> OK beck@ (millert@)
~ spamd.8
> Attempt to clarify new greylisting rules. (millert@)
~ spamd.8
> - HELO/EHLO is part of the tuple
> - remove misleading sentence
> ok millert (jmc@)
spamd-setup
~ spamd-setup.8 ~ spamd-setup.c
> When greylisting we don't actually need to use the <spamd> pf table.
> We just do no-rdr for things in <spamd-white> and rdr the rest to spamdb.
> OK beck@ (millert@)
~ spamd-setup.c
> Add usage() and error out if given bogus arguments. (millert@)
== regress =========================================================== 08/13 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/regress
sbin
~ ipsecctl/ikefail8.ok ~ ipsecctl/ikefail9.ok
> previous commit to parse.y was undone. adopt these two regression tests.
> (hshoexer@)
~ ipsecctl/Makefile
> we have to use '-k' now to show keys. (hshoexer@)
~ ipsecctl/sa9.in ~ ipsecctl/sa9.ok
> add a test for null encryption (hshoexer@)
usr.bin
~ rcs/Makefile
> Fix regression tests. False positives happened whenever the date/time
> contained 1.3, 1.4, or 1.5.
> ``hahahahahaha'' joris at .ay@)
== sbin ============================================================== 09/13 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/sbin
ccdconfig
~ ccdconfig.8
> sort options and sync synopsis; from Igor Sobrado (jmc@)
dhclient
~ dhclient-script
> no need to quote/escape here; ok krw@ (stevesk@)
~ dhclient.c ~ dhcpd.h
> remove unused/dangling prototypes and comment reference;
> from Mark Lumsden (stevesk@)
~ dhclient.c
> No point in using zero'd fields in the parsed options. Use the options
> in the lease, whence the parsed info was moved.
> Fixes "Bogus servername" messages when processing an offer and stops
> bogus filename info from being stored in the lease. Found while
> testing a Tucson hotel's dhcp service.
> ok @stevesk "looks good" henning at .rw@)
disklabel
~ disklabel.c
> Change hard coded numbers to sizeof(buf). Also change some
> sizeof(buf) - 1 to sizeof(buf), since fgets takes the whole buffer size.
> Based on diff from Charles Longeau <chl at tuxfamily dot org> long ago.
> OK millert at .ay@)
~ editor.c
> Always write the disklabel to disk when asked, even if it appears to
> be unchanged. Fixes PR#5209. Proper behaviour of 'q'uit still under
> discussion.
> ok ray@ millert@ ckuethe@ nick@ (krw@)
~ disklabel.c
> Correct comment calling FILE *f a file descriptor.
> OK krw at .ay@)
dump
~ optr.c
> Change hard coded numbers to sizeof(buf). Also change some
> sizeof(buf) - 1 to sizeof(buf), since fgets takes the whole buffer size.
> Based on diff from Charles Longeau <chl at tuxfamily dot org> long ago.
> OK millert at .ay@)
~ dump.8
> igor sobrado spotted that although we have mlinks for rdump and
> rrestore, the pages they point to don;t actually mention them.
> so make it clear that their functionality is now integral in dump and
> restore, and sync the text a little between the two pages;
> help/ok otto (jmc@)
~ dump.8
> document the "opaque", "arch", and "nodump" flags more fully;
> closes documentation/5384 from Andrew Hamilton-Wright
> help from henning, otto, and weingart (jmc@)
~ dump.8
> move the text describing 4.3 syntax compatibility to HISTORY, rather
> than slap bang in SYNOPSIS; (jmc@)
~ dump.8 ~ main.c
> the -W and -w flags do not need a separate synopsis; (jmc@)
~ dump.8
> shuffle the text slightly into a more logical order; (jmc@)
fsdb
~ fsdb.8 ~ fsdb.c
> clarify that -f is not optional; ok otto (jmc@)
fsirand
~ fsirand.8 ~ fsirand.c
> simplify synopsis and usage(); from Igor Sobrado (jmc@)
ipsecctl
~ parse.y
> undo previous commit and keep the original behaviour of the parser.
> asked for by deraadt@ (hshoexer@)
~ ipsecctl.c
> do not display empty authkey/enckey line when -k option is not
> specified.
> ok markus@ (hshoexer@)
~ ike.c ~ parse.y
> Bits for ESP+NULL encryption. This is useful, when AH can not be
> used (when being behind NAT). With Martin Hedenfalk <martin.hedenfalk
> at gmail.com>, thanks!
> ok markus@ (hshoexer@)
~ ipsec.conf.5
> Document NULL encryption. (hshoexer@)
~ ipsec.conf.5
> tweak; (jmc@)
isakmpd
~ pf_key_v2.c ~ conf.c
> isakmpd bits for ESP+NULL encryption. This is useful, when AH can
> not be used (when being behind NAT). With Martin Hedenfalk
> <martin.hedenfalk at gmail.com>, thanks! (hshoexer@)
~ isakmpd.conf.5
> Document NULL encryption. (hshoexer@)
~ isakmpd.conf.5
> tweak; (jmc@)
~ nat_traversal.c
> Add a comment that explains, why the VID of draft 2 NAT-T includes
> a traling '\n'.
> suggested by and ok deraadt@, jmc@ (hshoexer@)
modload
~ modload.8 ~ modload.c
> sort options and sync usage(); from Igor Sobrado (jmc@)
mount
~ mount.8
> sort the -o options; from Igor Sobrado (jmc@)
mount_nfs
~ mount_nfs.8
> sort the -o options; from Igor Sobrado (jmc@)
pdisk
~ hfs_misc.c
> fix typo in function name, no functional change (martin@)
pfctl
~ pfctl.c
> if machine has more than 100MB of physmem, default the max table entries
> to 200,000 instead of the conservative 100,000; ok dhartmei beck
> tested by ckuethe (deraadt@)
quotacheck
~ quotacheck.8 ~ quotacheck.c
> simplify synopsis and usage(); from Igor Sobrado (jmc@)
raidctl
~ rf_configure.c
> Don't ignore buffer size argument.
> OK millert at .ay@)
~ rf_configure.c
> Pass correct buffer size. (sizeof(cfgPtr->spare_names[c]) != 256)
> OK moritz at .ay@)
~ rf_configure.c
> Use sizeof(buf) instead of magic numbers.
> From Charles Longeau <chl at tuxfamily dot org> to tech@ back in
> November, reminded by deraadt at . > OK moritz at .ay@)
~ raidctl.8 ~ raidctl.c
> sort options; sync synopsis and usage()
> from Igor Sobrado (jmc@)
restore
~ restore.8
> igor sobrado spotted that although we have mlinks for rdump and
> rrestore, the pages they point to don;t actually mention them.
> so make it clear that their functionality is now integral in dump and
> restore, and sync the text a little between the two pages;
> help/ok otto (jmc@)
~ restore.8
> move the text describing 4.3 syntax compatibility to HISTORY, rather
> than slap bang in SYNOPSIS; (jmc@)
~ restore.8 ~ main.c
> one application does not need five synopses; (jmc@)
~ restore.8
> reword some text to avoid a hideous Xr; (jmc@)
routed
~ parms.c
> Change hard coded numbers to sizeof(buf). Also change some
> sizeof(buf) - 1 to sizeof(buf), since fgets takes the whole buffer size.
> Based on diff from Charles Longeau <chl at tuxfamily dot org> long ago.
> OK millert at .ay@)
sysctl
~ sysctl.c
> knf found while fixing another bug (deraadt@)
== share ============================================================= 10/13 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/share
man
~ man9/aml_evalnode.9
> tweaks; (jmc@)
~ man4/udav.4
> Add support for ShanTou ST268.
> Tested by Paul de Weerd. (jsg@)
~ man4/man4.sparc/cgfourteen.4
> Provide a working example of changing resolution. (miod@)
~ man9/Makefile
> Add aml_evalnode.9 along with the required MLINKS.
> Prodded by deraadt which made me notice I'd forgotten to update MLINKS
> as I added new functions. (mk@)
~ man4/ubsa.4
> add Option GlobeTrotter 3G+ to the list of devices supported by ubsa(4)
> (fkr@)
~ man4/vlan.4
> from brad:
> Hardware VLAN stripping was removed by Camiel before 3.8.
> ok camiel (jmc@)
~ man4/tun.4
> Sync with reality. Initial diff by Ingo Schwarze additional help by jmc@
> (claudio@)
~ man4/man4.zaurus/zaudio.4 ~ man4/man4.zaurus/zrc.4
> mention support for C3100 and C3200; from Martynas Venckus
> help/ok deraadt (jmc@)
~ man4/aue.4
> Belkin USB to LAN -> Belkin F5D5050 (jsg@)
+ man4/cas.4
> cas(4) (kettenis@)
~ man4/Makefile ~ man4/pci.4
> make and Xr cas 4 (deraadt@)
~ man4/cdcef.4
> Add myself as author, since I completed the driver. (drahn@)
~ man4/man4.zaurus/pxaudc.4
> USB client is starting to work. (drahn@)
~ man4/man4.zaurus/pxaudc.4
> More complete info about usb client mode, remove broken url as intel sold
> pxa. (drahn@)
~ man4/man4.zaurus/pxaudc.4
> tweaks by me and dlg; drahn ok (deraadt@)
~ man4/cas.4
> tweak; (jmc@)
~ man4/pci.4
> only the initial macro is preceded by a dot; (jmc@)
~ man4/man4.sparc/pninek.4 ~ man4/man4.sparc/pnozz.4
> Disable acceleration by default; set device flags to a nonzero value to
> enable. (miod@)
~ man4/man4.sparc/pninek.4 ~ man4/man4.sparc/pnozz.4
> tweaks; (jmc@)
~ man4/usbf.4
> config lines for cdcef and an Xr (dlg@)
~ man4/cas.4
> cas(4) can receive packets now. (kettenis@)
== sys =============================================================== 11/13 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/sys
arch/alpha/alpha
~ disksubr.c
> If no MBR partitions are found when spoofing a disklabel, try looking
> for a bare FAT12/16/32 filesystem. If one is found, spoof it as 'i'.
> This enables iPod shuffles and other umass devices that are shipped
> with a bare FAT32 filesystem to be used as install media. And
> generally makes things easier for the user just trying to mount some
> media, e.g. floppies.
> Does not make it safe to write a disklabel or MBR on such a device!
> That will still obliterate the FAT filesystem.
> Help and suggestions from tom@ and deraadt at .edback and suggestions
> for future enhancements from espie@ mickey@ and peter at . > ok tom@ deraadt@ (krw@)
~ disksubr.c
> Remove #ifdef/#endif's around MBR checking code. Remove now unused
> option DISKLABEL_I386.
> ok deraadt@ (krw@)
arch/alpha/conf
~ GENERIC
> Remove #ifdef/#endif's around MBR checking code. Remove now unused
> option DISKLABEL_I386.
> ok deraadt@ (krw@)
arch/alpha/include
~ atomic.h
> only make this interface available to the kernel for now, discussed witha
> rt and such; tested and ok miod drahn (deraadt@)
arch/amd64/amd64
~ disksubr.c
> If no MBR partitions are found when spoofing a disklabel, try looking
> for a bare FAT12/16/32 filesystem. If one is found, spoof it as 'i'.
> This enables iPod shuffles and other umass devices that are shipped
> with a bare FAT32 filesystem to be used as install media. And
> generally makes things easier for the user just trying to mount some
> media, e.g. floppies.
> Does not make it safe to write a disklabel or MBR on such a device!
> That will still obliterate the FAT filesystem.
> Help and suggestions from tom@ and deraadt at .edback and suggestions
> for future enhancements from espie@ mickey@ and peter at . > ok tom@ deraadt@ (krw@)
~ lapic.c
> Remove unused variable lapic_timer
> ok art@ (tom@)
~ ioapic.c
> Print base so that we can make a correct mapping between apic and irq.
> ok jordan (marco@)
~ acpi_machdep.c
> Enable ACPI interrupts in APIC mode
> ok marco@ (jordan@)
~ ioapic.c
> Revert my previous commit. Kettenis no likey. (marco@)
arch/amd64/conf
~ GENERIC
> ACPIVERBOSE should be off, like i386 (deraadt@)
~ RAMDISK_CD
> Put ral* at pci? onto the install media. Noted by Sunnz (sunnzy (at)
> gmail (dot) com) and Stuart Henderson (stu (at) spacehopper (dot) org)
> on misc at . > ok deraadt@ (tom@)
arch/amd64/include
~ atomic.h
> only make this interface available to the kernel for now, discussed witha
> rt and such; tested and ok miod drahn (deraadt@)
arch/arm/arm
~ disksubr.c
> If no MBR partitions are found when spoofing a disklabel, try looking
> for a bare FAT12/16/32 filesystem. If one is found, spoof it as 'i'.
> This enables iPod shuffles and other umass devices that are shipped
> with a bare FAT32 filesystem to be used as install media. And
> generally makes things easier for the user just trying to mount some
> media, e.g. floppies.
> Does not make it safe to write a disklabel or MBR on such a device!
> That will still obliterate the FAT filesystem.
> Help and suggestions from tom@ and deraadt at .edback and suggestions
> for future enhancements from espie@ mickey@ and peter at . > ok tom@ deraadt@ (krw@)
arch/arm/include
~ atomic.h
> only make this interface available to the kernel for now, discussed witha
> rt and such; tested and ok miod drahn (deraadt@)
arch/arm/xscale
~ pxa27x_udc.c
> Use the maxpacket len from the pipe, not hardcoded. properly deal
> with end of frame when it happens to be at end of usb packet. (drahn@)
~ pxa27x_udc.c
> Add my copyright, since I took the skeleton driver and made it complete.
> More polishing, appears to be working fine without back-to-back transmits.
> (drahn@)
~ pxa27x_udc.c
> move debug print into debug block. (drahn@)
~ pxa27x_udcreg.h
> Defines to make isr status more useable. Bit patterns for debug prints.
> (drahn@)
~ pxa27x_udc.c
> add missing USBD_FORCE_SHORT_XFER checks, put back some code I had for a
> while but removed incorrectly, some minor cleanup. (drahn@)
arch/armish/include
~ atomic.h
> only make this interface available to the kernel for now, discussed witha
> rt and such; tested and ok miod drahn (deraadt@)
arch/aviion/aviion
~ disksubr.c
> If no MBR partitions are found when spoofing a disklabel, try looking
> for a bare FAT12/16/32 filesystem. If one is found, spoof it as 'i'.
> This enables iPod shuffles and other umass devices that are shipped
> with a bare FAT32 filesystem to be used as install media. And
> generally makes things easier for the user just trying to mount some
> media, e.g. floppies.
> Does not make it safe to write a disklabel or MBR on such a device!
> That will still obliterate the FAT filesystem.
> Help and suggestions from tom@ and deraadt at .edback and suggestions
> for future enhancements from espie@ mickey@ and peter at . > ok tom@ deraadt@ (krw@)
arch/aviion/include
~ atomic.h
> only make this interface available to the kernel for now, discussed witha
> rt and such; tested and ok miod drahn (deraadt@)
arch/cats/include
~ atomic.h
> only make this interface available to the kernel for now, discussed witha
> rt and such; tested and ok miod drahn (deraadt@)
arch/hp300/dev
~ hd.c ~ hdvar.h
> remove part of buf abuse and store errcnt in softc (instead of b_errcnt);
> more work can be done to lower buf abuse even more; miod@ testing and ok
> (mickey@)
~ hd.c ~ hdvar.h
> Long awaiting modernization:
> - reset the drive and fetch its identification strings during probe, and do
> not attach if they don't look good.
> - do not store synchronous command blocks (used by hdreset) in the softc,
> since they are not processed asynchronously. The stack will do.
> - cleanup the disklabel retrieval code.
> - use disk_{,un}lock instead of rolling our own equivalent.
> - use bounds_check_with_label() in hdstrategy() instead of a stripped-down
> inline version of it.
> Tested on 7957A. (miod@)
~ hd.c
> Comment out the errmsg printf in hdgetdisklabel() again. (miod@)
arch/hp300/include
~ atomic.h
> only make this interface available to the kernel for now, discussed witha
> rt and such; tested and ok miod drahn (deraadt@)
arch/hppa/hppa
~ disksubr.c
> If no MBR partitions are found when spoofing a disklabel, try looking
> for a bare FAT12/16/32 filesystem. If one is found, spoof it as 'i'.
> This enables iPod shuffles and other umass devices that are shipped
> with a bare FAT32 filesystem to be used as install media. And
> generally makes things easier for the user just trying to mount some
> media, e.g. floppies.
> Does not make it safe to write a disklabel or MBR on such a device!
> That will still obliterate the FAT filesystem.
> Help and suggestions from tom@ and deraadt at .edback and suggestions
> for future enhancements from espie@ mickey@ and peter at . > ok tom@ deraadt@ (krw@)
~ disksubr.c
> Remove #ifdef/#endif's around MBR checking code. Remove now unused
> option DISKLABEL_I386.
> ok deraadt@ (krw@)
arch/hppa/include
~ atomic.h
> only make this interface available to the kernel for now, discussed witha
> rt and such; tested and ok miod drahn (deraadt@)
arch/hppa64/hppa64
~ disksubr.c
> If no MBR partitions are found when spoofing a disklabel, try looking
> for a bare FAT12/16/32 filesystem. If one is found, spoof it as 'i'.
> This enables iPod shuffles and other umass devices that are shipped
> with a bare FAT32 filesystem to be used as install media. And
> generally makes things easier for the user just trying to mount some
> media, e.g. floppies.
> Does not make it safe to write a disklabel or MBR on such a device!
> That will still obliterate the FAT filesystem.
> Help and suggestions from tom@ and deraadt at .edback and suggestions
> for future enhancements from espie@ mickey@ and peter at . > ok tom@ deraadt@ (krw@)
~ disksubr.c
> Remove #ifdef/#endif's around MBR checking code. Remove now unused
> option DISKLABEL_I386.
> ok deraadt@ (krw@)
arch/hppa64/include
~ atomic.h
> only make this interface available to the kernel for now, discussed witha
> rt and such; tested and ok miod drahn (deraadt@)
arch/i386/conf
~ RAMDISK_CD
> Remove duplicate entry.
> ok miod@ (tom@)
~ files.i386
> Revert PAE pmap for now, until the strange bug is found. This stops
> the freezes many of us are seeing (especially on amd64 machines running
> OpenBSD/i386).
> Much testing by nick@ (as always - thanks!), hugh@, ian@, kettenis@
> and Sam Smith (s (at) msmith (dot) net).
> Requested by, input from, and ok deraadt@ ok art@, kettenis@, miod@ (tom@)
arch/i386/i386
~ disksubr.c
> If no MBR partitions are found when spoofing a disklabel, try looking
> for a bare FAT12/16/32 filesystem. If one is found, spoof it as 'i'.
> This enables iPod shuffles and other umass devices that are shipped
> with a bare FAT32 filesystem to be used as install media. And
> generally makes things easier for the user just trying to mount some
> media, e.g. floppies.
> Does not make it safe to write a disklabel or MBR on such a device!
> That will still obliterate the FAT filesystem.
> Help and suggestions from tom@ and deraadt at .edback and suggestions
> for future enhancements from espie@ mickey@ and peter at . > ok tom@ deraadt@ (krw@)
~ lapic.c
> Remove unused variable lapic_timer
> ok art@ (tom@)
~ ioapic.c
> Put warning about shared IPL interrupts under MPVERBOSE and don't talk
> about
> degraded performance.
> requested by deraadt@ (kettenis@)
~ autoconf.c ~ cpu.c
~ db_memrw.c ~ genassym.cf
~ kgdb_machdep.c ~ lapic.c
~ locore.s ~ machdep.c
~ mpbios.c ~ mptramp.s
~ pmap.c ~ vm_machdep.c
~ kvm86.c
> Revert PAE pmap for now, until the strange bug is found. This stops
> the freezes many of us are seeing (especially on amd64 machines running
> OpenBSD/i386).
> Much testing by nick@ (as always - thanks!), hugh@, ian@, kettenis@
> and Sam Smith (s (at) msmith (dot) net).
> Requested by, input from, and ok deraadt@ ok art@, kettenis@, miod@ (tom@)
~ machdep.c
> part missed in PAE revert: ptoa() paddr_t size is no longer printed
> with %llu; ok miod (deraadt@)
~ ioapic.c
> Print base so that we can make a correct mapping between apic and irq.
> ok jordan (marco@)
~ ioapic.c
> Revert my previous commit. Kettenis no likey. (marco@)
~ vm_machdep.c
> Remove unused B_DIRTY, B_PAGET, B_PGIN, B_TAPE and B_UAREA b_flags values.
> (miod@)
arch/i386/include
~ atomic.h
> only make this interface available to the kernel for now, discussed witha
> rt and such; tested and ok miod drahn (deraadt@)
~ _types.h ~ bus.h
~ cpu.h ~ loadfile_machdep.h
~ param.h ~ pmap.h
~ pte.h ~ vmparam.h
> Revert PAE pmap for now, until the strange bug is found. This stops
> the freezes many of us are seeing (especially on amd64 machines running
> OpenBSD/i386).
> Much testing by nick@ (as always - thanks!), hugh@, ian@, kettenis@
> and Sam Smith (s (at) msmith (dot) net).
> Requested by, input from, and ok deraadt@ ok art@, kettenis@, miod@ (tom@)
arch/i386/pci
~ pci_addr_fixup.c ~ pci_machdep.c
~ pcibios.c
> Revert PAE pmap for now, until the strange bug is found. This stops
> the freezes many of us are seeing (especially on amd64 machines running
> OpenBSD/i386).
> Much testing by nick@ (as always - thanks!), hugh@, ian@, kettenis@
> and Sam Smith (s (at) msmith (dot) net).
> Requested by, input from, and ok deraadt@ ok art@, kettenis@, miod@ (tom@)
arch/landisk/include
~ atomic.h
> only make this interface available to the kernel for now, discussed witha
> rt and such; tested and ok miod drahn (deraadt@)
arch/landisk/landisk
~ disksubr.c
> If no MBR partitions are found when spoofing a disklabel, try looking
> for a bare FAT12/16/32 filesystem. If one is found, spoof it as 'i'.
> This enables iPod shuffles and other umass devices that are shipped
> with a bare FAT32 filesystem to be used as install media. And
> generally makes things easier for the user just trying to mount some
> media, e.g. floppies.
> Does not make it safe to write a disklabel or MBR on such a device!
> That will still obliterate the FAT filesystem.
> Help and suggestions from tom@ and deraadt at .edback and suggestions
> for future enhancements from espie@ mickey@ and peter at . > ok tom@ deraadt@ (krw@)
arch/luna88k/include
~ atomic.h
> only make this interface available to the kernel for now, discussed witha
> rt and such; tested and ok miod drahn (deraadt@)
arch/m68k/include
~ atomic.h
> only make this interface available to the kernel for now, discussed witha
> rt and such; tested and ok miod drahn (deraadt@)
arch/m88k/include
~ atomic.h
> only make this interface available to the kernel for now, discussed witha
> rt and such; tested and ok miod drahn (deraadt@)
arch/mac68k/include
~ atomic.h
> only make this interface available to the kernel for now, discussed witha
> rt and such; tested and ok miod drahn (deraadt@)
arch/macppc/dev
~ pm_direct.c ~ pm_direct.h
~ adb.c
> Correct yo-yo shutdowns:
> Apparently some machines don't detect a clean (i.e. software initiated)
> shutdown; and ignore the 'wake on AC loss' bit, meaning they wake up
> immediatly after being told to shutdown e.g. via shutdown -hp.
> So now we continue preserve the wake on AC loss behaivour, howerver if we
> initiate a clean shutdown we clear this bit so that it the machine does not
> immediately power back up again.
> ok kettenis@ (gwk@)
arch/macppc/include
~ atomic.h
> only make this interface available to the kernel for now, discussed witha
> rt and such; tested and ok miod drahn (deraadt@)
arch/macppc/macppc
~ disksubr.c
> If no MBR partitions are found when spoofing a disklabel, try looking
> for a bare FAT12/16/32 filesystem. If one is found, spoof it as 'i'.
> This enables iPod shuffles and other umass devices that are shipped
> with a bare FAT32 filesystem to be used as install media. And
> generally makes things easier for the user just trying to mount some
> media, e.g. floppies.
> Does not make it safe to write a disklabel or MBR on such a device!
> That will still obliterate the FAT filesystem.
> Help and suggestions from tom@ and deraadt at .edback and suggestions
> for future enhancements from espie@ mickey@ and peter at . > ok tom@ deraadt@ (krw@)
~ mainbus.c
> Change how we determin if were running on a new world versus old world
> machine for the purposes of printing model information durring boot and
> the hw.vendor/product sysctls. Turns out the apple didnt use the same
> format in the early imacs as they did in the latter new world machines
> resulting in NULL strings as found by drahn at .n this case and all future
> cases where we dont understand the format the default is to print whats
> found in the name property without interpertation.
> ok drahn@ (gwk@)
arch/mips64/include
~ atomic.h
> only make this interface available to the kernel for now, discussed witha
> rt and such; tested and ok miod drahn (deraadt@)
arch/mips64/mips64
~ disksubr.c
> If no MBR partitions are found when spoofing a disklabel, try looking
> for a bare FAT12/16/32 filesystem. If one is found, spoof it as 'i'.
> This enables iPod shuffles and other umass devices that are shipped
> with a bare FAT32 filesystem to be used as install media. And
> generally makes things easier for the user just trying to mount some
> media, e.g. floppies.
> Does not make it safe to write a disklabel or MBR on such a device!
> That will still obliterate the FAT filesystem.
> Help and suggestions from tom@ and deraadt at .edback and suggestions
> for future enhancements from espie@ mickey@ and peter at . > ok tom@ deraadt@ (krw@)
~ disksubr.c
> Remove #ifdef/#endif's around MBR checking code. Remove now unused
> option DISKLABEL_I386.
> ok deraadt@ (krw@)
~ vm_machdep.c
> Remove unused B_DIRTY, B_PAGET, B_PGIN, B_TAPE and B_UAREA b_flags values.
> (miod@)
arch/mvme68k/include
~ atomic.h
> only make this interface available to the kernel for now, discussed witha
> rt and such; tested and ok miod drahn (deraadt@)
arch/mvme88k/include
~ atomic.h
> only make this interface available to the kernel for now, discussed witha
> rt and such; tested and ok miod drahn (deraadt@)
arch/mvmeppc/include
~ atomic.h
> only make this interface available to the kernel for now, discussed witha
> rt and such; tested and ok miod drahn (deraadt@)
arch/mvmeppc/mvmeppc
~ disksubr.c
> If no MBR partitions are found when spoofing a disklabel, try looking
> for a bare FAT12/16/32 filesystem. If one is found, spoof it as 'i'.
> This enables iPod shuffles and other umass devices that are shipped
> with a bare FAT32 filesystem to be used as install media. And
> generally makes things easier for the user just trying to mount some
> media, e.g. floppies.
> Does not make it safe to write a disklabel or MBR on such a device!
> That will still obliterate the FAT filesystem.
> Help and suggestions from tom@ and deraadt at .edback and suggestions
> for future enhancements from espie@ mickey@ and peter at . > ok tom@ deraadt@ (krw@)
arch/powerpc/include
~ atomic.h
> only make this interface available to the kernel for now, discussed witha
> rt and such; tested and ok miod drahn (deraadt@)
arch/powerpc/powerpc
~ pmap.c
> Dont pass seemingly random numbers down as the flag
> parameter of pool_init()
> ok drahn@
> no objections miod@ (thib@)
arch/sgi/include
~ atomic.h
> only make this interface available to the kernel for now, discussed witha
> rt and such; tested and ok miod drahn (deraadt@)
arch/sh/include
~ atomic.h
> only make this interface available to the kernel for now, discussed witha
> rt and such; tested and ok miod drahn (deraadt@)
arch/solbourne/include
~ atomic.h
> only make this interface available to the kernel for now, discussed witha
> rt and such; tested and ok miod drahn (deraadt@)
arch/sparc/dev
~ mgx.c
> Make it more apparent these boards have (crippled version of)
> vga-compatible
> crtc. No functional change. (miod@)
~ cgfourteen.c ~ cgsix.c
~ cgthree.c ~ p9000.c
~ p9100.c ~ rfx.c
~ tcx.c ~ vigra.c
> Display the interrupt priority in dmesg, for frame buffers which register
> interrupt handlers. (miod@)
~ p9000.c ~ p9100.c
> Disable acceleration by default; set device flags to a nonzero value to
> enable. (miod@)
arch/sparc/include
~ atomic.h
> only make this interface available to the kernel for now, discussed witha
> rt and such; tested and ok miod drahn (deraadt@)
arch/sparc64/conf
~ GENERIC ~ RAMDISK
> Add cas(4). Ad a few missing comments while there.
> stolen from deraadt@ (kettenis@)
arch/sparc64/dev
~ schizo.c
> Establish pci error interrupt after we're done fiddling with schizo's
> registers
> instead of when we're halfway through. (kettenis@)
~ schizo.c
> Print safari error log register if we get a safari interrupt. (kettenis@)
arch/sparc64/include
~ atomic.h
> only make this interface available to the kernel for now, discussed witha
> rt and such; tested and ok miod drahn (deraadt@)
arch/vax/include
~ atomic.h
> only make this interface available to the kernel for now, discussed witha
> rt and such; tested and ok miod drahn (deraadt@)
arch/vax/vax
~ disksubr.c
> In writedisklabel(), read the disklabel sector before rewriting it, to
> preserve the non-disklabel contents; otherwise altering your disklabel may
> render your machine unbootable.
> ok deraadt@ krw@ (miod@)
arch/zaurus/include
~ atomic.h
> only make this interface available to the kernel for now, discussed witha
> rt and such; tested and ok miod drahn (deraadt@)
conf
~ files
> pull in atascsi, a scsi to ata translation layer, so i can work on it in
> tree. ive lost two large changes today thanks to my own stupidity which
> could have been avoided if they were in the tree.
> discussed with deraadt@
> ok jsg@ grange@ an older version was ok marco@ (dlg@)
crypto
~ xform.c
> minimum blocksize for ESP is 32 bit, so adjust blocksize of NULL
> encryption accordingly. Makes NULL encryption useable with ESP.
> Noticed by Martin Hedenfalk <martin.hedenfalk at gmail.com>.
> ok markus@ (hshoexer@)
~ blf.h
> Don't use arbitrary 128 bytes as size of the blf key array if we know
> exactly
> how much blowfish takes at max, which is 72 bytes.
> Also define a constant for this in the include file, suggested by ray at . > ok pedro@ thib@ tedu@ (grunk@)
dev
~ vnd.c
> Don't use arbitrary 128 bytes as size of the blf key array if we know
> exactly
> how much blowfish takes at max, which is 72 bytes.
> Also define a constant for this in the include file, suggested by ray at . > ok pedro@ thib@ tedu@ (grunk@)
dev/acpi
~ acpi.c
> Shave off another K pre compression.
> Shuffle SMALL_KERNEL functions around so that we only have a single #ifndef
> (marco@)
~ dsdt.h ~ dsdt.c
~ acpi.c
> Added dynamic perfect hash for opcodes
> ok marco@ (jordan@)
~ acpi.c ~ acpihpet.c
~ acpitimer.c ~ acpivar.h
> Added acpi_map_address function for mapping GAS/IO registers
> ok marco@ (jordan@)
~ dsdt.c
> Move AML_OBJDEBUG into ACPI_DEBUG.
> KNF while in here too. (marco@)
~ acpitimer.c
> KNF while in here. (marco@)
~ acpihpet.c
> KNF while here. (marco@)
~ acpihpet.c
> It helps to enable hpet... This fixes bnx not working on amd64. (marco@)
~ dsdt.c
> Fixed deref for storing to local after CondRef op
> ok marco@ (jordan@)
~ acpiac.c
> Some Sony VAIOs use the wrong eventing type. Work around that by honering
> it.
> Found by giovanni <qgiovanni @ gmail . com>
> ok jordan (marco@)
~ acpiec.c
> KNF (marco@)
~ acpiprt.c
> Fix a few memory leaks.
> ok jordan@ (marco@)
~ acpi.c ~ acpimadt.c
~ acpiprt.c
> Make an acpi-enabled GENERIC.MP work on systems that only support PIC mode.
> ok marco@ (kettenis@)
~ acpiec.c
> Fix stuttering audio playback as reported by Igor Grabin.
> Tested by several folks and ok canacar (marco@)
~ dsdt.c ~ dsdt.h
> New resource mapper interface and structures
> ok marco@ (jordan@)
~ acpitimer.c
> Fixed ACPITIMER header revision for ACPI2.0 only
> ok marco@ (jordan@)
~ acpi.c
> We were getting no interrupts in .MP. There are 2 fixes; one in
> acpi_machdep.c and move the interrupt establishment down after ioapic
> has been initialized.
> ok jordan (marco@)
~ acpiprt.c
> Added support for getting host-bridge _BBN from _CRS resources
> suggestions by kettenis
> ok marco@ (jordan@)
dev/ata
~ atascsi.c
> code to set up a global pool for allocating ata_xfers out of, and for
> accessing that pool. (dlg@)
~ atascsi.h
> i think ata_xfers should look more like this. stash the number of the port
> in the ata_port struct so a hba can find its own info for that port. (dlg@)
~ files.ata + atascsi.c
+ atascsi.h
> pull in atascsi, a scsi to ata translation layer, so i can work on it in
> tree. ive lost two large changes today thanks to my own stupidity which
> could have been avoided if they were in the tree.
> discussed with deraadt@
> ok jsg@ grange@ an older version was ok marco@ (dlg@)
~ atascsi.c
> actually init and destroy the shared ata stuff when an atascsi instance
> is created or destroyed. (dlg@)
~ atascsi.c
> oops, i should have tried to compile the REQUEST_SENSE handling before
> committing it. (dlg@)
~ atascsi.c ~ atascsi.h
> as is short for atascsi, not ab (dlg@)
~ atascsi.c
> REQUEST_SENSE will say there is no sense data. (dlg@)
~ atascsi.c ~ atascsi.h
> implement the start of the framework for scsi emulation of ata disks (dlg@)
~ atascsi.h
> precomp -> features as per ATA spec
> ok dlg@ (jsg@)
dev/cardbus
~ if_bcw_cardbus.c
> Adjust var name so cardbus attaches again. (mglocker@)
~ if_bcw_cardbus.c
> No need to fillup chip_rev here. (mglocker@)
~ if_bcw_cardbus.c
> Power on/off cardbus socket at attach/detach time. Makes firmware upload
> work on my BCM4318 cardbus device now. (mglocker@)
dev/ic
~ bcw.c ~ bcwreg.h
~ bcwvar.h
> Turn radio (version 2050) on. (mglocker@)
~ bcw.c ~ bcwreg.h
~ bcwvar.h
> Start to add radio PHY initialization routines.
> Rename some radio variable names to names which doesn't confuse me
> everytime I have to use them. (mglocker@)
~ bcw.c
> Sort routines (phy, radio). (mglocker@)
~ bcw.c
> Add bcw_shm_write16(). (mglocker@)
~ bcw.c
> Comment corrections from Bret Lambert. (mglocker@)
~ bcw.c
> Add bcw_phy_set_baseband_attenuation(). (mglocker@)
~ bcw.c ~ bcwreg.h
~ bcwvar.h
> ~1052 lines of new PHY radio initialization code and no end in sight.
> (mglocker@)
~ acx.c
> Fix a memory leak in acx_load_radio_firmware() -- free the ucode after use.
> OK mglocker@ (claudio@)
~ acx.c
> Make the firmware load on big endian systems and while there make sure that
> the right value is printed in case of an error. OK mglocker@ (claudio@)
~ bcw.c ~ bcwreg.h
~ bcwvar.h
> Halftime for PHY init code, another ~1000 new lines. (mglocker@)
~ bcw.c ~ bcwvar.h
> Add bcw_phy_initb2() and bcw_phy_initb4(). Activate bcw_phy_init() in the
> bcw_init() path. (mglocker@)
~ lm78.c ~ lm78var.h
> Add support for Winbond W83627EHF chips. (kettenis@)
~ bcw.c ~ bcwvar.h
> Add bcw_phy_lo_b_measure(). (mglocker@)
~ bcw.c ~ bcwreg.h
~ bcwvar.h
> Complement bcw_phy_inita(), add bcw_radio_init2060(). (mglocker@)
~ bcw.c
> Activate bcw_phy_init2060() (type A radio) in bcw_radio_on(). (mglocker@)
~ bcw.c
> Add and activate bcw_radio_init2050 (radio type B). (mglocker@)
~ bcw.c
> Complement bcw_radio_set_txpower_a(). (mglocker@)
~ bcw.c
> Add missing routine to just allow a number in a given range. Replaces
> some /* XXX */ lines. (mglocker@)
~ bcw.c ~ bcwvar.h
> Add bcw_phy_xmitpower(). (mglocker@)
~ bcw.c
> Complement and fix a bunch of things in bcw_radio_select_channel().
> (mglocker@)
~ bcw.c
> Add bcw_radio_freq_r3a_value() and complement bcw_phy_initg(). (mglocker@)
~ acx.c
> Another two missing htoleX() for big endian support. Still more missing
> but now I'm able to receive packets on macppc. OK mglocker@ (claudio@)
~ bcw.c
> Add bcw_phy_lo_g_state(). (mglocker@)
~ bcw.c ~ bcwvar.h
> Fix value typos, sync radio vars with phy vars, add radio default
> attenuation routine, add temporary debug printfs. (mglocker@)
~ bcw.c ~ bcwvar.h
> Get correct chip revision and also get chip package. (mglocker@)
~ re.c
> dmesg should look like devN at busN locators: device info
> the locators are printed by the bus drivers, rest by the device's driver.
> (deraadt@)
~ bcw.c
> Calculate number of cores correctly. This is done based on the common
> core revision. Now my BCM4318 and BCM4306 both get the right number of
> cores. (mglocker@)
~ bcw.c
> Power on/off cardbus socket at attach/detach time. Makes firmware upload
> work on my BCM4318 cardbus device now. (mglocker@)
dev/pci
~ ahci.c
> reorder allocation of the ports resources so register manipulation can be
> done earlier. (dlg@)
~ files.pci
> ahci will depend on atascsi and scsi (dlg@)
~ ahci.c
> stick atascsi onto ahci, which in turn attaches a scsibus. while this
> compiles, it definitely doesnt work. (dlg@)
~ arc.c
> No use checking malloc return when called with M_WAITOK.
> ok dlg@, tested by jolan@ (thib@)
~ ahci.c
> fill some macros for ahci structures. fix a spello while there. (dlg@)
~ ahci.c
> oops, missed one in the cmd list entry (dlg@)
~ ahci.c
> rename ccb_id to ccb_slot so its more clear what its used for. stash a
> pointer to the port and the ata_xfer in there too. (dlg@)
~ devlist2h.awk ~ pci_subr.c
> put the structs defining the content of pcidevs_data.h into that header
> file, rather than in pci_subr.c before theyre used. this allows the header
> to be used elsewhere without needing to copy a chunk out of pci_subr.c.
> ok miod@ deraadt@ (dlg@)
~ pcidevs.h ~ pcidevs_data.h
> regen for inclusion of the structs used in pcidevs_data.h (dlg@)
~ pcidevs.h ~ pcidevs_data.h
> sync (deraadt@)
~ devlist2h.awk
> put structs after copyright notice; ok dlg (deraadt@)
~ pcivar.h
> hide all the things userland should not see, so that this file can be
> pulled into get at pcireg_t. otherwise it would hit nasty _KERNEL stuff;
> discussed with miod and dlg (deraadt@)
+ if_cas.c + if_casreg.h
+ if_casvar.h
> Start of a driver for Sun Cassini gigabit ethernet interfaces. Driver can
> send packets but not receive any yet. (kettenis@)
~ files.pci
> Glue for cas(4). (kettenis@)
~ if_bcw_pci.c
> Get correct chip revision and also get chip package. (mglocker@)
~ if_cas.c ~ if_casreg.h
~ if_casvar.h
> Make receiving packets work. Driver works well enough to do the commit
> over. (kettenis@)
dev/sbus
~ mgx.c
> Make it more apparent these boards have (crippled version of)
> vga-compatible
> crtc. No functional change. (miod@)
dev/usb
~ usbdevs
> another udav(4) (jsg@)
~ usbdevs.h ~ usbdevs_data.h
> regen (jsg@)
~ if_udav.c
> Add support for ShanTou ST268.
> Tested by Paul de Weerd. (jsg@)
~ if_ral.c ~ if_rum.c
~ if_uath.c
> do not do a curproc test for interrupt context, because it is plainly
> wrong.
> this lets these work on macppc, for instance
> diagnosed by kettenis, but damien is not around, so ok jsg and others
> (deraadt@)
~ usbdevs
> usb device id for Globetrotter 3G+ from Victor Meghesan through bugs@
> (fkr@)
~ usbdevs.h ~ usbdevs_data.h
> regen (fkr@)
~ ubsa.c
> let the Globetrotter 3G+ attach to ubsa, from Victor Meghesan through bugs@
> (fkr@)
~ if_cdce.c
> force usb to send a zero length packet on the end of usb packets
> which happen to be a multple of 64 bytes, to indicate end of packet.
> (drahn@)
~ if_cdcef.c
> Use FORCE_SHORT_XFER on tx, hardware driver doesn't use it yet, but will.
> (drahn@)
~ usbdevs
> Belkin USB to LAN -> Belkin F5D5050 (jsg@)
~ usbdevs.h ~ usbdevs_data.h
> regen (jsg@)
~ if_aue.c
> Belkin USB to LAN -> Belkin F5D5050 (jsg@)
~ if_cdcef.c
> Add myself to copyright since I took the skeleton driver and made it
> complete.
> Work around a transmit hang problem by setting a delay on transmitting
> back-to-back packets. Count dropped packets as input errors.
> Abort pipe if a watchdog occurs to clean up state better. (drahn@)
dev/vesa
~ vesabios.c ~ vesafb.c
+ vbe.h
> Add currently unused functions for getting EDID information (max supported
> resolution, hsync, vsync etc) from the monitor using DDC.
> Cleanup a whole bunch of magic,
> ok deraadt (gwk@)
kern
~ kern_sysctl.c
> for sensors do not leak kernel pointers when copying out to userland;
> spotted by art, ok dlg art (deraadt@)
~ vfs_subr.c
> for vfsconf sysctl, do not leak kernel sensors out to userland
> ok art thib (deraadt@)
net
~ pf.c
> add handling of skip steps for urpf-failed addresses.
> ok dharthmei@, henning@ (pyr@)
~ if_tun.c
> For TUNSIFMODE protect the if_flags similar to the TUNSIFINFO case.
> Don't allow the userland to fiddle with flags reserved by the driver.
> Noticed by Ingo Schwarze. (claudio@)
~ pf.c
> make urpf-failed work with multipath routes.
> ok claudio@ (pyr@)
~ if_gif.c
> Make gif(4) altq aware. This simplifies setting up traffic shaping on
> gif(4)
> tunnels.
> Additional testing by Marc Winiger. OK kjc@ mbalmer@ (claudio@)
~ pf_ioctl.c ~ pfvar.h
> if machine has more than 100MB of physmem, default the max table entries
> to 200,000 instead of the conservative 100,000; ok dhartmei beck
> tested by ckuethe (deraadt@)
netinet
~ ip_ether.c
> Fix the bpf hook in etherip_input() by using the right address family and
> moving the m_adj() down to prevent so that the ethernet header is
> preserved.
> This fixes tcpdump on bridged gif(4) interfaces. OK mglocker@ (claudio@)
~ in_pcb.h
> Remove TCP ports 760 and 761 from DEFBADDYNAMICPORTS_TCP; they are
> not used with Heimdal. Add UDP port 749 to DEFBADDYNAMICPORTS_UDP
> for consistency with DEFBADDYNAMICPORTS_TCP. We retain some Kerberos
> 4 ports for people running Heimdal in Kerberos 4 compat mode.
> OK deraadt@ beck@ (millert@)
nfs
~ nfs_vnops.c
> backout previous revision -- on nfs2 mounts (perhaps only automounted?) it
> tends to .nfs silly files lying around during high filesystem activity.
> whatever bug it was trying to fix, it made things worse. (deraadt@)
scsi
~ sd.c
> Don't unconditionally spin up SDEV_UMASS devices. They almost
> certainly don't need it, or are capable of reporting themselves in
> need of a spin up. Feedback from various agreeing it is unlikely such
> devices exist.
> Fixes devices found by Markus Bergkvist and Johan M:son Lindman.
> ok dlg@ (krw@)
~ safte.c
> Add missing sensor description for temperature.
> ok dlg (deanna@)
stand
~ boot/cmd.c
> hand-cruft an isprint() for deciding which characters to echo to the
> console; ignore all others. based on idea by didickman at gmail.com > ok tom (deraadt@)
sys
~ sensors.h
> add missing copyright; Constantine (deraadt@)
~ buf.h
> Remove unused B_DIRTY, B_PAGET, B_PGIN, B_TAPE and B_UAREA b_flags values.
> (miod@)
uvm
~ uvm_mmap.c
> Make integer wrap checks the same for mmap, munmap, msync, etc
> by factoring most of the checks into a macro. OK otto@ (millert@)
== usr.bin =========================================================== 12/13 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/usr.bin
biff
~ biff.1 ~ biff.c
> sort options; from Igor Sobrado (jmc@)
cmp
~ extern.h
> g/c OK_EXIT, it was never used. From Mark Lumsden. (millert@)
col
~ col.1
> sort options; from Igor Sobrado (jmc@)
cvs
~ checkout.c
> an erroneous free() crept in due to a merge error (otto@)
~ buf.c ~ buf.h
~ cmd.c ~ cvs.h
~ entries.c ~ file.c
~ rcs.c ~ rcs.h
~ util.c ~ util.h
> Fix const inconsistencies, void pointer artithmetic and a few other
> things. now gcc -Wall and lint are more happy. ok joris@ xsa@ (otto@)
~ rcsnum.c
> Replace a manual xrealloc with rcsnum_setsize.
> From Charles Longeau <chl at tuxfamily dot org>.
> OK joris at .ay@)
~ buf.c ~ buf.h
~ file.c ~ import.c
> Simplify cvs_buf_differ(). Also cvs_buf_differ(), a comparison
> function, should not do any resource freeing.
> OK joris at .ay@)
- includes.h ~ add.c
~ admin.c ~ annotate.c
~ buf.c ~ buf.h
~ checkout.c ~ client.c
~ cmd.c ~ commit.c
~ compress.c ~ config.c
~ cvs.c ~ cvs.h
~ date.y ~ diff.c
~ diff3.c ~ diff_internals.c
~ edit.c ~ entries.c
~ fatal.c ~ file.c
~ file.h ~ getlog.c
~ import.c ~ init.c
~ log.c ~ log.h
~ logmsg.c ~ rcs.c
~ rcsnum.c ~ remote.c
~ remove.c ~ repository.c
~ root.c ~ server.c
~ status.c ~ tag.c
~ update.c ~ util.c
~ version.c ~ watch.c
~ worklist.c ~ worklist.h
~ xmalloc.c
> general includes cleanup sweep. ok joris@ niallo@ (otto@)
~ cvs.c
> fix comment; from jared r r spiegel (otto@)
dc
~ dc.1
> put `I' in the right place; from Igor Sobrado (jmc@)
diff
~ diffreg.c
> Defer printing of the per-file diff header until after the regexp
> "ignore" processing has finished. This way we only print the header
> for files that have diffs. The new behavior matches GNU diff (which
> is where the -I flag comes from). OK otto@ espie@ (millert@)
~ diffreg.c
> print the header only once.
> okay millert@ (espie@)
file
~ file.1 ~ file.c
> Our version of file(1) doesn't have the MIME magic file, so remove
> references to -i/--mime and no longer accept them on the command line.
> Update the HISTORY section to be less confusing.
> Based on a diff to tech@ by Mikolaj Kucharski, mikolaj (dot) kucharski
> (at) gmail (dot) com.
> Prompted by jmc@; input from and ok jmc@ and otto@ (tom@)
~ file.1
> Remove several incorrect and/or unmaintained emails, prompted by correction
> from bernd@ (ian@)
from
~ from.1 ~ from.c
> sort options and avoid line wrap; from Igor Sobrado (jmc@)
locate
~ code/locate.code.c
> - Check for fgets failure.
> - Pass sizeof(buf) to fgets.
> - Only print what was read.
> - Simplify error checking while printing bigrams.
> Inspired by diff from Charles Longeau <chl at tuxfamily dot org> long ago.
> OK millert@ and moritz at .ay@)
mail
~ mail.1
> remove bad comma; from Igor Sobrado (jmc@)
make
~ PSD.doc/tutorial.ms
> typos; from Daniel Dickman (jmc@)
mg
~ mg.1
> Document -h option. (cloder@)
~ mg.1
> Back out -h doc after discussion with theo (cloder@)
~ main.c
> Remove useless -h option (if you don't know the synopsis, how do you know
> -h?). OK deraadt (cloder@)
~ def.h
> Fix a reference to WFHARD in a comment; it was renamed WFFULL months
> ago.
> ok kjell (deanna@)
nc
~ netcat.c
> usage(): add -P and tweak -X; from Igor Sobrado (jmc@)
pkg-config
~ pkg-config.1
> missing full stop; from Daniel Matic (jmc@)
pr
~ pr.c
> Some cleanup including:
> - use sig_atomic_t signal handler flag
> - atoi() -> strtonum() conversion
> - remove vararg without matching fmt string
> - some more lint/gcc silencing
> ok millert@ ray@ (moritz@)
rcs
~ co.c
> keep correct permissions on the RCS file when locking/unlocking the file
> with ci and co. fixes PR 5379.
> OK millert@ joris at .sa@)
~ co.c
> make sure checkout_rev() sets correct file permissions via file->rf_mode
> in case where fstat() of file->rf_fd won't give us right mode.
> fixes PR #5383
> ok xsa@ (niallo@)
~ buf.c ~ buf.h
~ diff.c ~ diff3.c
~ rcsutil.c ~ rcsutil.h
> Fix const inconsistencies, void pointer artithmetic;
> Based on diff from otto@ in OpenCVS. OK otto at .sa@)
~ ci.c ~ co.c
~ rcsutil.c ~ rcsutil.h
> If a ,suffix file is given as an arg to ci and co, strip it. Avoids
> potential disasters. Initial diff from niallo@, ok niallo@ joris@ (otto@)
sdiff
~ sdiff.1
> Expand description.
> Help and okay jmc at .ay@)
ssh
~ monitor.c ~ monitor_wrap.c
~ servconf.c ~ servconf.h
~ sshd_config.5
> Teach Match how handle config directives that are used before
> authentication.
> This allows configurations such as permitting password authentication from
> the local net only while requiring pubkey from offsite. ok djm@, man page
> bits ok jmc@ (dtucker@)
~ clientloop.c
> set maximum packet and window sizes the same for multiplexed clients
> as normal connections; ok markus@ (djm@)
~ sshd.c
> Clear alarm() before restarting sshd on SIGHUP. Without this, if there's
> a SIGALRM pending (for SSH1 key regeneration) when sshd is SIGHUP'ed, the
> newly exec'ed sshd will get the SIGALRM and not have a handler for it,
> and the default action will terminate the listening sshd. Analysis and
> patch from andrew at gaul.org. (dtucker@)
~ servconf.c
> Check activep so Match and GatewayPorts work together; ok markus@
> (dtucker@)
~ moduli.c
> - strlen returns size_t, not int.
> - Pass full buffer size to fgets.
> OK djm@, millert@, and moritz at .ay@)
sudo
~ sudo.8
> Fix typo; from Stephane Chausson. (millert@)
systat
~ Makefile ~ cmdtab.c
~ extern.h ~ systat.1
+ sensors.c
> Add a display for hardware sensors, available with :sensors.
> ok deraadt, jmc (deanna@)
~ sensors.c
> knf (deraadt@)
~ systat.1
> forgot to list the display in one place (deraadt@)
~ sensors.c
> leave unspecified fields blank; ok deanna
> also, fix two logic errors in switches. (deraadt@)
~ sensors.c
> add missing copyright; Constantine (deraadt@)
~ systat.1
> more logical placing for discussion of minimum prefixes accepted;
> started by deraadt (jmc@)
~ systat.1
> consistency fix which i should have spotted earlier... (jmc@)
~ systat.1
> sort; (jmc@)
~ systat.1
> mving the description of "mim prefix" to the display description was not
> quite correct - simplify it, and document it both for command line and
> interactive; (jmc@)
~ systat.1
> one final sort; (jmc@)
~ sensors.c
> some cleanups, i agree; Murenin (deraadt@)
~ extern.h
> unused vars; from Mark Lumsden (otto@)
~ cmdtab.c ~ if.c
~ iostat.c ~ main.c
~ mbufs.c ~ netstat.c
~ pigs.c ~ sensors.c
~ swap.c ~ vmstat.c
> use the 1 line load line from 'systat vm' on all the other displays, giving
> them more room. much cleaner, and it is less code too; ok deanna
> (deraadt@)
~ sensors.c
> add missing include, and int return value.
> ok deraadt, deanna (steven@)
tip
~ value.c
> Change hard coded numbers to sizeof(buf). Also change some
> sizeof(buf) - 1 to sizeof(buf), since fgets takes the whole buffer size.
> Based on diff from Charles Longeau <chl at tuxfamily dot org> long ago.
> OK millert at .ay@)
~ value.c
> Remove wrong length check, which is off-by-one, and
> let snprintf(3) decide, if the whole string fittet
> into the buffer.
> ok ray@ (moritz@)
top
~ layout.h ~ top.h
> an unused define and an unused decl. From Mark Lumsden. (otto@)
- sigconv.awk ~ Makefile
~ commands.c
> Use sys_signame[] from libc instead of creating a top-specific
> signame name to number table. Also use strcasecmp() when matching
> signal names for consistency with /bin/kill.
> OK deraadt@ henning@ simon@ (millert@)
tset
~ tset.c
> Change hard coded numbers to sizeof(buf). Also change some
> sizeof(buf) - 1 to sizeof(buf), since fgets takes the whole buffer size.
> Based on diff from Charles Longeau <chl at tuxfamily dot org> long ago.
> OK millert at .ay@)
units
~ units.c
> Change hard coded numbers to sizeof(buf). Also change some
> sizeof(buf) - 1 to sizeof(buf), since fgets takes the whole buffer size.
> Based on diff from Charles Longeau <chl at tuxfamily dot org> long ago.
> OK millert at .ay@)
~ units.c
> - Don't access buffers with negative indexes.
> - Save a ton of strlen calls.
> - Return NULL pointer, not 0.
> ``boy there is a lot of strlen()'' millert@ and OK moritz at .ay@)
usbhidaction
~ usbhidaction.c
> signal handler flag must be volatile sig_atomic_t (deraadt@)
vacation
~ vacation.c
> support for "Auto-Submitted: no" from Tamas TEVESZ; small tweak by millert,
> ok millert (man page update being done by jmc) (deraadt@)
~ vacation.c
> add back in character count check by Tamas TEVESZ, though this does
> not truly handle trailing whitespace on the line. but that is handled
> very weakly for other patterns, too. (deraadt@)
~ vacation.1
> support for "Auto-Submitted: no" from Tamas TEVESZ
> plus use a list for things not replied to; (jmc@)
~ vacation.c
> Ignore trailing space after "Auto-Submitted: no"
> Remove an unneeded check for !*p
> OK deraadt@ (millert@)
~ vacation.1
> show flag argument (deraadt@)
== usr.sbin ========================================================== 13/13 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/usr.sbin
acpidump
~ acpidump.8
> Added new prefix dump option for obtaining all ACPI files
> ok marco@ (jordan@)
~ acpi.c ~ aml_dump.c
> Added changes to dump all tables not just DSDT
> -o generates files of form <prefix>.<sig>.<num>
> eg. -o foo generates foo.DSDT.1 foo.FACP.0 foo.MCFG.2 etc
> ok marco@ (jordan@)
~ acpidump.8
> tweaks; (jmc@)
amd
~ amd/amd.8 ~ amd/get_args.c
> sort options and sync usage(); from Igor Sobrado
> also fix a misplaced .El (jmc@)
~ amq/amq.8 ~ amq/amq.c
> simplify synopsis/usage; mostly from Igor Sobrado (jmc@)
~ amd/amd.8
> document -C option; ok jmc (deraadt@)
~ amd/amd.8 ~ amd/get_args.c
> oops, let synopsis and usage agree; (jmc@)
arp
~ arp.c
> Use sizeof(buf) instead of hard coded numbers.
> From Charles Longeau <chl at tuxfamily dot org> long ago.
> OK moritz at .ay@)
authpf
~ authpf.c
> close 5389 and 5390,
> unused variable and a chance to unlink the pidfile without lock
> if we couldn't kill a preexisting authpf process.
> spotted by Stefan Krah <sfk1 at bigfoot.com(beck@)
~ authpf.c
> this ftruncate is really not needed now, if we're just unlinking.
> ok millert@ (beck@)
~ authpf.8 ~ authpf.c
> license + copyright (beck@)
~ authpf.c
> exit right away if the config file isn't there, rather than
> doing a whole bunch of needless screwing around
> noticed by Stefan Krah <stefan-usenet at bytereef.orgbeck@)
bgpctl
~ bgpctl.c ~ parser.c
> minor KNF bits here too (henning@)
bgpd
~ session.c
> do not try to set IP_TTL on non-INET listening sockets, it's not going
> to work on INET6...
> non-fatal error, so in pratice this just avoids a misleading error msg
> (henning@)
~ carp.c ~ kroute.c
~ parse.y ~ rde.c
~ rde_attr.c ~ rde_prefix.c
~ rde_rib.c ~ session.c
> KNF (henning@)
~ log.h
> tiny typos hiding, uncovered Leen Besselink <leen at consolejunkie.net > (henning@)
~ log.c
> we do have a table with suberror codes for cease and their meaning;
> use it, instead of always telling "unknown suberror code X"
> From: Leen Besselink <leen at consolejunkie.net > if memory serves, cease didn't have suberror codes defined in the original
> RFC 1771 and they were added later (henning@)
bind
~ bin/dig/dighost.c
> Pass sizeof(buf) to fgets instead of hard-coded number.
> From Charles Longeau.
> OK moritz@ and jakob at .ay@)
cron
~ cron.8 ~ cron.c
> tidy up synopsis and usage(); (jmc@)
~ crontab.c ~ crontab.1
> tidy up synopsis and usage(); (jmc@)
dhcpd
~ dhcpd.h
> unused extern decl
> From: Mark Lumsden <mark at cyodesigns.comhenning@)
~ dhcpd.h
> another dead extern; Mark Lumsden <mark at cyodesigns.comhenning@)
fdformat
~ fdformat.1 ~ fdformat.c
> tidy up synopsis and usage(); reported by Igor Sobrado (jmc@)
~ fdformat.1
> put -n in the right place; from Igor Sobrado (jmc@)
~ fdformat.c
> - Pass full buffer size to fgets.
> - No need to initialize buffer.
> - Spacing.
> Based on diff from Charles Longeau <chl at tuxfamily dot org> long ago.
> OK moritz at .ay@)
hoststatectl
~ hoststatectl.8 ~ hoststatectl.c
~ parser.c ~ parser.h
> Add layer 7 functionality to hoststated used for layer 7
> loadbalancing, SSL acceleration, general-purpose TCP relaying, and
> transparent proxying.
> see hoststated.conf(5) and my upcoming article on undeadly.org for
> details.
> ok to commit deraadt@ pyr@ (reyk@)
~ hoststatectl.c
> spacing (reyk@)
hoststated
~ pfe_filter.c
> use HOST_UP instead of an hard coded integer (reyk@)
~ Makefile ~ hce.c
~ hoststated.8 ~ hoststated.c
~ hoststated.conf.5 ~ hoststated.h
~ log.c ~ parse.y
~ pfe.c ~ pfe_filter.c
~ ssl.c + carp.c
+ relay.c
> Add layer 7 functionality to hoststated used for layer 7
> loadbalancing, SSL acceleration, general-purpose TCP relaying, and
> transparent proxying.
> see hoststated.conf(5) and my upcoming article on undeadly.org for
> details.
> ok to commit deraadt@ pyr@ (reyk@)
~ hoststated.conf.5
> document the new options to manipulate carp demotion counters. (reyk@)
~ hoststated.conf.5
> document the retry option before setting the state to down for hosts
> in tables. (reyk@)
~ carp.c ~ check_tcp.c
~ hoststated.c ~ hoststated.h
~ pfe.c ~ pfe_filter.c
~ ssl.c
> spacing (reyk@)
~ hoststated.8
> +.An -nosplit (jmc@)
~ hoststated.conf.5
> various language/macro fixes; (jmc@)
~ hoststated.conf.5
> put `check ssl' in the right place; (jmc@)
~ relay.c
> read the exact length for POST requests as specified by the
> content-length header. (reyk@)
~ Makefile ~ check_icmp.c
~ hoststated.c ~ hoststated.h
~ pfe.c
> knf (deraadt@)
~ hoststated.conf.5
> i.e. -> e.g.; ok reyk (jmc@)
~ hoststated.conf.5 ~ hoststated.h
~ parse.y ~ relay.c
> - allow to specify the SSL cipher suite and the SSL protocols
> (as required by the PCI DSS)
> - increase the default listen backlog to 10, allow to modify the
> backlog as a per-protocol tcp option to improve the performance
> on busy systems (to get less connection failures on heavy load)
> - close the connection if SSL_accept returned an error
> - instead of logging _new_ relay sessions to syslog, log the
> sessions in relay_close() after they have been _finished_.
> this will allow to collect some additional information
> - add a new log keyword to log specified header/url entities (useful
> to track "bad guys" using many session ids or multiple user agents)
> - some minor fixes, manpage bits, and bump the copyright (by some
> reason, i didn't realize that we already have 2007...). (reyk@)
~ hoststated.conf.5 ~ hoststated.h
~ relay.c
> disable SSLv2 and use "HIGH" crypto cipher suites by default.
> suggested by dlg@ (reyk@)
~ hoststated.conf.5 ~ hoststated.h
> disable anonymous DH by default (cipher suite HIGH:!ADH instead of HIGH).
> (reyk@)
~ hoststated.conf.5
> tweaks; (jmc@)
~ hoststated.h
> remove unused variable (reyk@)
~ Makefile
> delete -W flags that reyk uses to break the tree (deraadt@)
~ relay.c
> one example (of two) of tree breaking the "other gcc" (deraadt@)
httpd
~ Makefile.bsd-wrapper
> unbreak by allowing clean in !objdir case to fail (no Makefile there in
> that
> case), ok espie, also noticed by simon (henning@)
~ src/modules/standard/mod_cern_meta.c
> Change hard coded numbers to sizeof(buf). Also change some
> sizeof(buf) - 1 to sizeof(buf), since fgets takes the whole buffer size.
> Based on diff from Charles Longeau <chl at tuxfamily dot org> long ago.
> OK millert at .ay@)
iostat
~ iostat.c ~ iostat.8
> - sync SYNOPSIS and usage()
> - add args to the option list
> - I/O does not need .Tn
> - .Nm does not need an argument
> based on a diff from Igor Sobrado (jmc@)
kgmon
~ kgmon.8 ~ kgmon.c
> order options and usage(); from Igor Sobrado (jmc@)
map-mbone
~ mapper.c ~ map-mbone.8
> tidy up synopsis and usage(); from Igor Sobrado (jmc@)
~ map-mbone.8
> tweaks from Igor Sobrado and myself; (jmc@)
memconfig
~ memconfig.8
> unneeded .Op; from Igor Sobrado (jmc@)
mrouted
~ mrouted.8 ~ main.c
> tidy up synopsis and usage; from Igor Sobrado (jmc@)
mtrace
~ mtrace.8 ~ mtrace.c
> tidy up synopsis and usage; from Igor Sobrado (jmc@)
~ mtrace.8
> put -n in the right place; from Igor Sobrado (jmc@)
ndp
~ ndp.c
> Use sizeof(buf) instead of hard coded numbers.
> From Charles Longeau <chl at tuxfamily dot org> long ago.
> OK moritz at .ay@)
ntpd
~ ntpd.8
> xref adjfreq(2); from Igor Zinovik (otto@)
ospfd
~ kroute.c
> spelling; sthen at spacehoppereraadt@)
~ kroute.c
> copy mis-aligned routing socket message into a properly aligned struct, so
> that we can access it safely; ok claudio
> in the long term we must deal with the 32/64 bit issues of if_buadrate and
> friends (deraadt@)
pkg_add
~ OpenBSD/PackageInfo.pm ~ pkg_delete
> Problem noticed by Henrik Hellerstedt <henrik at kohina.net > Fix a really fun bug in pkg_delete: solve_installed_names has to
> normalize the list of names so that comparing its size vs. its
> transitive closure is meaningful. Use the classical hash + list to
> preserve the order. While we're there, use the hash to avoid asking
> the same questions again and again.
> After some back&forth with bernd. okay bernd@ (espie@)
~ OpenBSD/ProgressMeter.pm
> avoid comparing undef to 0: perl's ioctl returns `true' 0s, so that this
> idiom works. (espie@)
~ OpenBSD/Ustar.pm
> avoid undefined. closes pr5386. (espie@)
~ OpenBSD/Add.pm ~ OpenBSD/Vstat.pm
> UI improvement: in case the disk overflows, let tally give information
> about the extra size needed. In validate, mark the overflow and call tally
> later.
> .e.g.,
> Error: /dev/wd0f is not large enough (/usr/local/lib/libkparts.so.4.0.0)
> Error: /dev/wd0f is not large enough (/usr/local/lib/libkrosscore.so.4.0.0)
> Error: /dev/wd0f is not large enough
> (/usr/local/lib/libktexteditor.so.4.0.0)
> Error: ... more files do not fit on /dev/wd0f
> /dev/wd0h: 97 bytes
> /dev/wd0f: 337090852 bytes (missing 78473 blocks) (espie@)
pppd
~ pppstats/pppstats.c ~ pppstats/pppstats.8
> tidy up synopsis and usage; from Igor Sobrado (jmc@)
pwd_mkdb
~ pwd_mkdb.8
> tidy up synopsis and options list; from Igor Sobrado (jmc@)
quotaon
~ quotaon.8 ~ quotaon.c
> simplify synopsis and usage(); based on a diff from Igor Sobrado (jmc@)
repquota
~ repquota.c ~ repquota.8
> tidy up synopsis and usage(); partly from Igor Sobrado (jmc@)
rpc.bootparamd
~ rpc.bootparamd.8 ~ bootparamd.c
> sync usage() and synopsis; spotted by Igor Sobrado (jmc@)
sensorsd
~ sensorsd.c
> add missing copyright; Constantine (deraadt@)
spamdb
~ spamdb.8
> greytrapping improvements
> 1) remove requirement for <> around spamtrap addresses
> 2) add support for /etc/spamd/alloweddomains to specify
> suffixes for which any destinations that don't match
> get trapped
> various knf's by theo, feedback from jmc, millert, deraadt (beck@)
~ spamdb.8 ~ spamdb.c
> Make spamd include the HELO/EHLO identification string sent by
> the connecting hosts in the tuple key when greylisting. catches
> a few more bogus hosts and will let us trap based on HELO later.
> Changes spamdb(8) output to include the new field.
> ok deraadt@, jmc@ (beck@)
syslogd
~ privsep.c ~ syslogd.c
> implement logging to other program's stdin.
> if the target is like "| /path/to/program", syslogd forks and execs program
> and sends the selected log messages to program's stdin.
> uses a socketpair, grows the receive buffer on the reader side and has the
> socket nonblocking on syslog'd side to prevent syslogd blocking.
> I'm using that here to feed logsurfer from ports for automated log
> analysis,
> werks beautifully.
> lots of input & help mpf, ok mpf djm "no objections" millert (henning@)
~ syslog.conf.5
> document syslog to other programs, ok jmc (henning@)
~ syslogd.c
> Create "syslogd: restart" message after the setup of signal
> handlers. This prevents a race where syslogd could die
> on SIGPIPE at startup. Also distinguish between start and restart.
> OK henning@ (mpf@)
tcpdump
~ print-enc.c
> register is so 1980, plus knf (deraadt@)
timed
~ timed/timed.8 ~ timed/timed.c
> sort options and sync usage(); from Igor Sobrado (jmc@)
trpt
~ trpt.8 ~ trpt.c
> tidy up synopsis and usage(); from Igor Sobrado (jmc@)
user
~ user.8
> put -G in the right place; from Igor Sobrado (jmc@)
~ useradd.8
> put -L and -G in the right place; (jmc@)
vnconfig
~ vnconfig.c
> use correct variable in error message: print the filename, not the key
> ok ray@ thib@ (grunk@)
ypbind
~ ypbind.c ~ ypbind.8
> tidy up synopsis and usage(); from Igor Sobrado (jmc@)
yppoll
~ yppoll.8 ~ yppoll.c
> tidy up synopsis and usage(); from Igor Sobrado (jmc@)
ypserv
~ ypserv/ypserv.8 ~ ypserv/ypserv.c
> tidy up synopsis and usage(); from Igor Sobrado (jmc@)
~ makedbm/makedbm.8 ~ makedbm/makedbm.c
> - simply synopsis and usage()
> - order options
> mostly from Igor Sobrado (jmc@)
~ revnetgroup/parse_netgroup.c
> Change hard coded numbers to sizeof(buf). Also change some
> sizeof(buf) - 1 to sizeof(buf), since fgets takes the whole buffer size.
> Based on diff from Charles Longeau <chl at tuxfamily dot org> long ago.
> OK millert at .ay@)
ypset
~ ypset.8 ~ ypset.c
> tidy up synopsis and usage(); from Igor Sobrado (jmc@)
ztsscale
~ ztsscale.c
> remove some lint (deraadt@)
===============================================================================
More information about the owc
mailing list