[odc] Daily src changes for 2005-04-08
ODC
auto at squish.net
Sat Apr 9 08:00:26 BST 2005
OpenBSD src changes summary for 2005-04-08
==========================================
bin/dd bin/mv
distrib/sets gnu/lib/libf2c
gnu/lib/libobjc gnu/lib/libstdc++
gnu/usr.bin/gcc gnu/usr.sbin/sendmail
regress/sbin sbin
sbin/dhclient sbin/ipsecadm
sbin/isakmpd sbin/mount
sbin/mount_ados sbin/mount_cd9660
sbin/mount_ext2fs sbin/mount_fdesc
sbin/mount_ffs sbin/mount_kernfs
sbin/mount_lfs sbin/mount_msdos
sbin/mount_nfs sbin/mount_ntfs
sbin/mount_null sbin/mount_portal
sbin/mount_procfs sbin/mount_udf
sbin/mount_umap sbin/mount_union
sbin/mount_xfs sbin/mountd
share/man sys/arch/amd64/amd64
sys/arch/amd64/include sys/arch/arm/xscale
sys/arch/i386/i386 sys/arch/i386/include
sys/arch/zaurus/dev sys/dev/ic
sys/dev/pci sys/dev/usb
sys/sys usr.bin/tip
usr.sbin/bgpd
== bin =============================================================== 01/09 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/bin
dd
~ args.c
> Correct comment about using m in blocksize (multiplies by 1048576 not 512).
> (jsg@)
mv
~ mv.c
> Sync with recent realpath(3) changes: on failure, don't use the second
> argument "resolved", since it is undefined.
> ok and help millert, otto (jaredy@)
== distrib =========================================================== 02/09 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/distrib
sets
~ lists/base/mi ~ lists/man/mi
> sync (deraadt@)
== gnu =============================================================== 03/09 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/gnu
lib/libf2c
~ config.guess
> amd64 canonicalization.
> ok deraadt@, pvalchev@ (kettenis@)
lib/libobjc
~ config.guess
> amd64 canonicalization.
> ok deraadt@, pvalchev@ (kettenis@)
lib/libstdc++
~ config.guess
> amd64 canonicalization.
> ok deraadt@, pvalchev@ (kettenis@)
usr.bin/gcc
~ Makefile.bsd-wrapper ~ config.guess
> amd64 canonicalization.
> ok deraadt@, pvalchev@ (kettenis@)
usr.sbin/sendmail
~ RELEASE_NOTES ~ cf/README
~ cf/m4/version.m4 ~ doc/op/op.me
~ include/sm/conf.h ~ include/sm/shm.h
~ libsm/Makefile ~ libsm/ldap.c
~ libsm/shm.c ~ libsm/t-shm.c
~ sendmail/README ~ sendmail/TRACEFLAGS
~ sendmail/collect.c ~ sendmail/conf.c
~ sendmail/daemon.c ~ sendmail/deliver.c
~ sendmail/domain.c ~ sendmail/envelope.c
~ sendmail/map.c ~ sendmail/mci.c
~ sendmail/milter.c ~ sendmail/parseaddr.c
~ sendmail/queue.c ~ sendmail/sendmail.h
~ sendmail/srvrsmtp.c ~ sendmail/tls.c
~ sendmail/usersmtp.c ~ sendmail/version.c
+ include/sm/sem.h + libsm/sem.c
+ libsm/t-sem.c
> update to sendmail 8.13.4 (millert@)
== regress =========================================================== 04/09 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/regress
sbin
+ isakmpd/exchange/README
> Forgot this file. Don't believe what it says in entirety.
> We will update doc later. (cloder@)
+ isakmpd/Makefile
> Moving regress tests from sbin/isakmpd to regress/sbin/isakmpd. They are
> not run by default yet. (cloder@)
+ isakmpd/b2n/Makefile + isakmpd/b2n/b2ntest.c
+ isakmpd/crypto/Makefile + isakmpd/crypto/cryptotest.c
+ isakmpd/dh/Makefile + isakmpd/dh/dhtest.c
+ isakmpd/ec2n/Makefile + isakmpd/ec2n/ec2ntest.c
+ isakmpd/exchange/Makefile + isakmpd/exchange/def-i.1
+ isakmpd/exchange/def-r.1 + isakmpd/exchange/mm-1-setup.sh
+ isakmpd/exchange/mm-i-1.t + isakmpd/exchange/mm-r-1.t
+ isakmpd/exchange/run.pl + isakmpd/exchange/run.sh
+ isakmpd/group/Makefile + isakmpd/group/grouptest.c
+ isakmpd/hmac/Makefile + isakmpd/hmac/hmactest.c
+ isakmpd/msg/check.sh + isakmpd/prf/Makefile
+ isakmpd/prf/prftest.c + isakmpd/rsakeygen/.cvsignore
+ isakmpd/rsakeygen/Makefile + isakmpd/rsakeygen/rsakeygen.c
+ isakmpd/util/Makefile + isakmpd/util/utiltest.c
+ isakmpd/x509/Makefile + isakmpd/x509/x509test.c
> Moved these tests here from sbin/isakmpd/regress. They mostly work
> but are not enabled by default yet. (cloder@)
== sbin ============================================================== 05/09 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/sbin
sbin
~ Makefile
> connect ipsecctl, request by deraadt (hshoexer@)
dhclient
~ dhclient.c
> protect read() on rtsock against interruptions by signal, foudn with djm
> while looking for sth else (henning@)
ipsecadm
~ ipsecadm.8
> document ipcomp(4); (jmc@)
isakmpd
~ isakmpd.c
> use __progname (deraadt@)
~ isakmpd.c ~ sysdep.h
> more __progname (deraadt@)
- sysdep/bsdi/GNUmakefile.sysdep - sysdep/bsdi/Makefile.sysdep
- sysdep/bsdi/sysdep-os.h - sysdep/bsdi/sysdep.c
- sysdep/darwin/GNUmakefile.sysdep - sysdep/darwin/Makefile.sysdep
- sysdep/darwin/sysdep-os.h - sysdep/darwin/sysdep.c
- sysdep/freebsd/GNUmakefile.sysdep - sysdep/freebsd/Makefile.sysdep
- sysdep/freebsd/sysdep-os.h - sysdep/freebsd/sysdep.c
- sysdep/linux/bitstring.h - sysdep/linux/sys/queue.h
- sysdep/netbsd/GNUmakefile.sysdep - sysdep/netbsd/Makefile.sysdep
- sysdep/netbsd/sysdep-os.h - sysdep/netbsd/sysdep.c
> if someone wants a portable, create a -p project. clutter stops quality
> (deraadt@)
~ Makefile
> remove some portable junk (deraadt@)
~ sysdep/openbsd/sysdep-os.h
> remove portable stuff not needed (deraadt@)
~ sysdep/openbsd/sysdep.c
> sysdep_progname() is non-portable gunk (deraadt@)
- sysdep/linux/GNUmakefile.sysdep - sysdep/linux/sysdep-os.h
- sysdep/linux/sysdep.c
> if someone wants a portable, create a -p project. clutter stops quality
> (deraadt@)
~ conf.c ~ crypto.c
> un-ifdef USE_BLOWFISH (deraadt@)
- sysdep/common/blf.h - sysdep/common/cast.h
- sysdep/common/md5.h - sysdep/common/pcap.h
- sysdep/common/sha1.h - sysdep/common/libsysdep/GNUmakefile
- sysdep/common/libsysdep/Makefile
- sysdep/common/libsysdep/arc4random.c
- sysdep/common/libsysdep/blowfish.c - sysdep/common/libsysdep/cast.c
- sysdep/common/libsysdep/md5.c - sysdep/common/libsysdep/sha1.c
- sysdep/common/libsysdep/strlcat.c - sysdep/common/libsysdep/strlcpy.c
> if someone wants a portable, create a -p project. clutter stops quality
> (deraadt@)
~ Makefile ~ app.c
~ crypto.c ~ crypto.h
~ sysdep.h ~ sysdep/openbsd/sysdep.c
> app.c (deraadt@)
~ app.c
> need monitor.h (deraadt@)
~ log.c ~ sysdep/openbsd/Makefile.sysdep
> kill HAVE_PCAP (deraadt@)
~ math_group.c
> kill USE_EC (deraadt@)
- GNUmakefile
> nothing uses this (deraadt@)
~ pf_key_v2.c ~ udp.c
~ udp_encap.c ~ virtual.c
> Remove references to LINUX_IPSEC (cloder@)
~ if.c ~ sysdep/openbsd/Makefile.sysdep
> kill HAVE_GETIFADDRS (deraadt@)
~ crypto.c
> kill USE_DES and USE_TRIPLEDES, why was it ever a choice even? (deraadt@)
~ isakmpd.c ~ sysdep/openbsd/Makefile.sysdep
> kill HAVE_CLOSEFROM (deraadt@)
~ ipsec.c ~ pf_key_v2.c
~ util.c ~ sysdep/openbsd/Makefile.sysdep
> kill USE_OLD_SOCKADDR (deraadt@)
- sysdep/openbsd/GNUmakefile.sysdep ~ isakmpd.c
~ log.c ~ monitor.h
~ x509.c
> privsep always (deraadt@)
~ Makefile
> remove cipher features (hshoexer@)
~ sysdep/openbsd/Makefile.sysdep
> the lovely thing about layers and layers of compatible stuff, is that
> deep down in the guts of it assumptions such as "this particular processors
> has no shared libraries yet" get hidden, only to be found about 2 years
> later. i hate code that hides stuff like that from me. (deraadt@)
- features/x509
> x509 is always on (deraadt@)
~ transport.c
> improve logging, original diff and ok cloder@ (hshoexer@)
- features/ec - features/privsep
~ Makefile
> ec and privsep always on (deraadt@)
~ Makefile
> remove crud (deraadt@)
- features/aggressive - features/dpd
- features/isakmp_cfg ~ Makefile
~ exchange.c ~ ike_phase_1.c
~ ipsec.c ~ ipsec.h
~ isakmp_doi.c ~ message.c
~ message.h ~ sa.c
~ sa.h
> always enable aggressive, dpd, and isakmp_cfg (deraadt@)
- features/nat_traversal ~ Makefile
~ exchange.c ~ ike_aggressive.c
~ ike_phase_1.c ~ init.c
~ ipsec.c ~ isakmpd.c
~ log.c ~ message.c
~ pf_key_v2.c ~ policy.c
~ sa.c ~ sa.h
~ virtual.c
> nat-traversal always (deraadt@)
- regress/Makefile - regress/check.sh
> Remove these as well. (cloder@)
~ Makefile
> regression tests have moved (hshoexer@)
~ sysdep/openbsd/Makefile.sysdep
> old static lib stuff is not even needed... (deraadt@)
- sysdep/openbsd/keynote_compat.c
> not needed anymore (deraadt@)
~ DESIGN-NOTES
> Refer to correct regress directory now that regress is moved. (cloder@)
- features/policy ~ Makefile
~ cert.c ~ conf.c
~ ike_auth.c ~ ike_quick_mode.c
~ init.c ~ isakmpd.c
~ libcrypto.c ~ monitor.c
~ pf_key_v2.c ~ sa.c
~ x509.c
> keynote and policy always compiled in (deraadt@)
- regress/b2n/.cvsignore - regress/b2n/Makefile
- regress/b2n/b2ntest.c - regress/crypto/.cvsignore
- regress/crypto/Makefile - regress/crypto/cryptotest.c
- regress/dh/.cvsignore - regress/dh/Makefile
- regress/dh/dhtest.c - regress/ec2n/.cvsignore
- regress/ec2n/Makefile - regress/ec2n/ec2ntest.c
- regress/exchange/.cvsignore - regress/exchange/Makefile
- regress/exchange/README - regress/exchange/def-i.1
- regress/exchange/def-r.1 - regress/exchange/mm-1-setup.sh
- regress/exchange/mm-i-1.t - regress/exchange/mm-r-1.t
- regress/exchange/run.pl - regress/exchange/run.sh
- regress/group/.cvsignore - regress/group/Makefile
- regress/group/grouptest.c - regress/hmac/.cvsignore
- regress/hmac/Makefile - regress/hmac/hmactest.c
- regress/prf/.cvsignore - regress/prf/Makefile
- regress/prf/prftest.c - regress/rsakeygen/.cvsignore
- regress/rsakeygen/Makefile - regress/rsakeygen/rsakeygen.c
- regress/util/Makefile - regress/util/utiltest.c
- regress/x509/.cvsignore - regress/x509/Makefile
- regress/x509/x509test.c
> Moved regression tests to src/regress/sbin/isakmpd where they belong.
> (cloder@)
~ Makefile
> de-feature raw rsa keys, cleanup leftovers (hshoexer@)
~ ike_auth.c
> zap USE_RAWKEY (deraadt@)
~ policy.c
> we have IPPROTO_ETHERIP (deraadt@)
~ crypto.h ~ hash.c
> we do not care about __APPLE__ (deraadt@)
~ exchange.c
> remove leftovers (hshoexer@)
~ crypto.c
> simplify (deraadt@)
~ exchange.c
> missing includes (hshoexer@)
~ util.c
> zap USE_DEFAULT_ROUTE (hshoexer@)
~ sysdep/openbsd/Makefile.sysdep
> zap USE_DEFAULT_ROUTE (hshoexer@)
- sysdep/openbsd/Makefile.sysdep ~ Makefile
~ apps/certpatch/Makefile ~ sysdep/openbsd/sysdep.c
> remove more conditional crud (deraadt@)
- sysdep/openbsd/sysdep-os.h ~ Makefile
~ sysdep.h
> forget about sysdep/openbsd/sysdep-os.h (deraadt@)
~ Makefile ~ gmp_util.c
~ math_group.c ~ math_mp.h
~ apps/certpatch/Makefile
> remove gmp, no binary change (hshoexer@)
~ Makefile ~ connection.c
~ doi.h ~ dpd.c
~ exchange.c ~ ipsec.c
~ isakmp_doi.c ~ isakmpd.c
~ log.c ~ log.h
~ message.c ~ pf_key_v2.c
~ ui.c ~ virtual.c
> USE_DEBUG is bye bye (deraadt@)
- gmp_util.c - gmp_util.h
~ Makefile ~ ike_auth.c
~ math_group.c
> kill gmp (hshoexer@)
~ message.c
> move __inline to right place (deraadt@)
~ apps/certpatch/Makefile
> remove junk not needed to build certpatch (deraadt@)
- features/dnssec ~ Makefile
> whack last feature, put in main Makefile (deraadt@)
~ dpd.c
> Make DPD vendor ID u_int8_t, not char. Gets rid of a warning in
> -pedantic mode. OK hshoexer (cloder@)
~ apps/certpatch/certpatch.c
> knf; ok cloder hshoexunpunouncable (deraadt@)
~ attribute.c ~ cert.c
~ conf.c ~ constants.c
~ crypto.c ~ dh.c
~ dnssec.c ~ doi.c
~ exchange.c ~ field.c
~ hash.c ~ if.c
~ ike_aggressive.c ~ ike_auth.c
~ ike_main_mode.c ~ ike_phase_1.c
~ ike_quick_mode.c ~ init.c
~ isakmp_cfg.c ~ isakmp_doi.c
~ isakmpd.8 ~ isakmpd.c
~ key.c ~ libcrypto.c
~ math_2n.c ~ math_ec2n.c
~ math_group.c ~ nat_traversal.c
~ policy.c ~ prf.c
~ sa.c ~ sysdep.h
~ timer.c ~ udp.c
~ ui.c ~ util.c
~ util.h ~ x509.c
~ sysdep/openbsd/sysdep.c
> Make deterministic randomness (only ever used for testing) a compile-time
> option. Reduces chances of somehow setting regrand when it's not supposed
> to be set. Remove "-r" option from man page. Also xref certpatch(8) while
> we are in there. And remove some include sysdep.h where it is no longer
> needed.
> OK hshoexer (cloder@)
~ cookie.c ~ ipsec.c
~ message.c ~ monitor.c
~ pf_key_v2.c ~ sa.c
~ sysdep.h ~ transport.c
~ udp.c ~ udp_encap.c
~ util.c ~ virtual.c
~ sysdep/openbsd/sysdep.c
> get rid of sysdep_sa_len
> ok cloder@ (hshoexer@)
mount
~ mount.c
> Sync with recent realpath(3) changes: on failure, don't use the second
> argument "resolved", since it is undefined.
> ok and help millert, otto (jaredy@)
mount_ados
~ mount_ados.c
> Sync with recent realpath(3) changes: on failure, don't use the second
> argument "resolved", since it is undefined.
> ok and help millert, otto (jaredy@)
mount_cd9660
~ mount_cd9660.c
> Sync with recent realpath(3) changes: on failure, don't use the second
> argument "resolved", since it is undefined.
> ok and help millert, otto (jaredy@)
mount_ext2fs
~ mount_ext2fs.c
> Sync with recent realpath(3) changes: on failure, don't use the second
> argument "resolved", since it is undefined.
> ok and help millert, otto (jaredy@)
mount_fdesc
~ mount_fdesc.c
> Sync with recent realpath(3) changes: on failure, don't use the second
> argument "resolved", since it is undefined.
> ok and help millert, otto (jaredy@)
mount_ffs
~ mount_ffs.c
> Sync with recent realpath(3) changes: on failure, don't use the second
> argument "resolved", since it is undefined.
> ok and help millert, otto (jaredy@)
mount_kernfs
~ mount_kernfs.c
> Sync with recent realpath(3) changes: on failure, don't use the second
> argument "resolved", since it is undefined.
> ok and help millert, otto (jaredy@)
mount_lfs
~ mount_lfs.c
> Sync with recent realpath(3) changes: on failure, don't use the second
> argument "resolved", since it is undefined.
> ok and help millert, otto (jaredy@)
mount_msdos
~ mount_msdos.c
> Sync with recent realpath(3) changes: on failure, don't use the second
> argument "resolved", since it is undefined.
> ok and help millert, otto (jaredy@)
mount_nfs
~ mount_nfs.c
> Sync with recent realpath(3) changes: on failure, don't use the second
> argument "resolved", since it is undefined.
> ok and help millert, otto (jaredy@)
mount_ntfs
~ mount_ntfs.c
> Sync with recent realpath(3) changes: on failure, don't use the second
> argument "resolved", since it is undefined.
> ok and help millert, otto (jaredy@)
mount_null
~ mount_null.c
> Sync with recent realpath(3) changes: on failure, don't use the second
> argument "resolved", since it is undefined.
> ok and help millert, otto (jaredy@)
mount_portal
~ mount_portal.c
> Sync with recent realpath(3) changes: on failure, don't use the second
> argument "resolved", since it is undefined.
> ok and help millert, otto (jaredy@)
mount_procfs
~ mount_procfs.c
> Sync with recent realpath(3) changes: on failure, don't use the second
> argument "resolved", since it is undefined.
> ok and help millert, otto (jaredy@)
mount_udf
~ mount_udf.c
> Sync with recent realpath(3) changes: on failure, don't use the second
> argument "resolved", since it is undefined.
> ok and help millert, otto (jaredy@)
mount_umap
~ mount_umap.c
> Sync with recent realpath(3) changes: on failure, don't use the second
> argument "resolved", since it is undefined.
> ok and help millert, otto (jaredy@)
mount_union
~ mount_union.c
> Sync with recent realpath(3) changes: on failure, don't use the second
> argument "resolved", since it is undefined.
> ok and help millert, otto (jaredy@)
mount_xfs
~ mount_xfs.c
> Sync with recent realpath(3) changes: on failure, don't use the second
> argument "resolved", since it is undefined.
> ok and help millert, otto (jaredy@)
mountd
~ mountd.c
> Sync with recent realpath(3) changes: on failure, don't use the second
> argument "resolved", since it is undefined.
> ok and help millert, otto (jaredy@)
== share ============================================================= 06/09 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/share
man
~ man8/release.8
> make distrib-dirs, ok otto (henning@)
~ man4/ath.4 ~ man4/atu.4
~ man4/atw.4 ~ man4/ipw.4
~ man4/ral.4 ~ man4/ray.4
~ man4/rtw.4 ~ man4/wi.4tbl
> grammar; from jaredy@ (jmc@)
~ man4/ipcomp.4
> minor tweaks; (jmc@)
~ man4/ipsec.4
> document ipcomp(4); (jmc@)
~ man4/tty.4
> no point telling people to read tip(1) *and* cu(1), since they
> are one and the same; (jmc@)
~ man4/man4.sparc64/sab.4
> grammar; (jmc@)
== sys =============================================================== 07/09 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/sys
arch/amd64/amd64
~ mutex.S
> add rcs ids (jolan@)
arch/amd64/include
~ mutex.h
> add rcs ids (jolan@)
arch/arm/xscale
~ pxa2x0_ohci.c
> clean up properly in the event of a failure during attach.
> ok uwe@ (dlg@)
~ pxa2x0_gpio.c ~ pxa2x0_gpio.h
> - manipulate the proper event counter in pxagpio_intr1
> - pxa2x0_gpio_intr_mask/unmask as discussed with dale (uwe@)
arch/i386/i386
~ mutex.S
> add rcs ids (jolan@)
arch/i386/include
~ mutex.h
> add rcs ids (jolan@)
arch/zaurus/dev
~ zaurus_ssp.c ~ zaurus_sspvar.h
> new functions for a unified interface to all chips. (uwe@)
~ zts.c
> Synchronize ADS7846 communication and LCD refresh so as to avoid jitter
> in the touch screen data, and enable the existing IRQ handling code
> (even though polling is still mandatory). (uwe@)
dev/ic
~ re.c
> - enable reception of VLAN sized frames
> - enable HW receive checksum offload
> ok pvlachev@ (brad@)
~ ar5210.c ~ ar5210var.h
~ ar5211.c ~ ar5211var.h
~ ar5212.c ~ ar5212var.h
~ ar5xxx.h
> bye bye unportable bitfields (reyk@)
~ ar5210.c ~ ar5211.c
~ ar5212.c ~ ath.c
> move reset of HAL descriptors to ath.c (reyk@)
dev/pci
~ pcidevs.h ~ pcidevs_data.h
> regen (brad@)
~ pcidevs
> add QLogic 2312 and 6312 cards (brad@)
~ isp_pci.c
> recognize Dell OEM 2312 card.
> From FreeBSD (brad@)
~ if_vge.c
> - enable reception of VLAN sized frames
> - enable HW receive checksum offload
> ok pvlachev@ (brad@)
~ if_vge.c
> fix parenthesis (brad@)
~ if_ste.c
> remove dmesg spewage when link changes, ok deraadt, brad, canacar (beck@)
dev/usb
~ uftdi.c
> handle at least the first port of a dual-port ufdti; from freebsd
> (deraadt@)
~ ucom.c
> no need to spit out portno twice (deraadt@)
sys
~ sensors.h
> interger->integer (jolan@)
== usr.bin =========================================================== 08/09 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/usr.bin
tip
~ tip.1
> ~. was not being marked up properly; (jmc@)
== usr.sbin ========================================================== 09/09 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/usr.sbin
bgpd
~ kroute.c
> don't whine when we can't find interface with index 0
> (that is a rare corner case in that context, but still) (henning@)
===============================================================================
More information about the odc
mailing list