[odc] Daily src changes for 2005-04-05

ODC auto at squish.net
Wed Apr 6 08:00:26 BST 2005


OpenBSD src changes summary for 2005-04-05
==========================================

bin/cat                                 distrib/miniroot
distrib/sets                            lib/libc
libexec/ld.so                           sbin/bioctl
sbin/ipsecctl                           sbin/isakmpd
sbin/mount_udf                          sbin/quotacheck
share/man                               sys/arch/arm/xscale
sys/arch/sparc64/dev                    sys/dev/ic
sys/dev/isa                             sys/dev/pci
sys/isofs/udf                           sys/net
sys/netinet                             sys/scsi
usr.bin/ftp                             usr.bin/netstat
usr.bin/ssh                             usr.sbin/bgpd
usr.sbin/httpd                          usr.sbin/ospfd

== bin =============================================================== 01/09 ==

  http://www.openbsd.org/cgi-bin/cvsweb/src/bin

cat

  ~ cat.c                                 

  > change incorrect filename in output on fstat() failure
  > from Cedric Berger <cedric at berger.to  > ok otto (jaredy@)

== distrib =========================================================== 02/09 ==

  http://www.openbsd.org/cgi-bin/cvsweb/src/distrib

miniroot

  ~ install.sub                           

  > Use 'route -qn' for the IPv6 default route command in hostname.if.
  > Don't use local link address as possible IPv6 manual address.
  > From todd at .rw@)

sets

  ~ lists/man/mi                          

  > sync (deraadt@)

== lib =============================================================== 03/09 ==

  http://www.openbsd.org/cgi-bin/cvsweb/src/lib

libc

  ~ net/resolver.3                        

  > explain where type and class are defined; ok jmc@
  > CVS ----------------------------------------------------------------------
  > (fgsch@)

== libexec =========================================================== 04/09 ==

  http://www.openbsd.org/cgi-bin/cvsweb/src/libexec

ld.so

  ~ dlfcn.c                               ~ library.c
  ~ library_mquery.c                      ~ library_subr.c
  ~ loader.c                              ~ resolve.c
  ~ resolve.h                             

  > Do a better job of running destructors in the right order. (drahn@)

== sbin ============================================================== 05/09 ==

  http://www.openbsd.org/cgi-bin/cvsweb/src/sbin

bioctl

  ~ bioctl.8                              

  > add -b to synopsis;
  > use .Bk/.Ek to avoid line split; (jmc@)

  ~ bioctl.c                              ~ bioctl.h

  > clean up enumeration.
  > properly print inquiry data.
  > identify devices bioctl cares about. (marco@)

  ~ bioctl.c                              ~ bioctl.h

  > fix output from remaining passthrough functions (marco@)

ipsecctl

  ~ ipsec.conf.5                          ~ ipsecctl.8
  ~ ipsecctl.c                            

  > cleanup; ok hshoexer@ (jmc@)

  ~ ipsec.conf.5                          

  > The first release this will appear in is OpenBSD 3.8
  > ok jmc@ (tom@)

isakmpd

  ~ isakmpd.8                             ~ isakmpd.c
  ~ nat_traversal.c                       ~ nat_traversal.h
  ~ virtual.c                             

  > Add -T flag to isakmpd to disable NAT-T support from the command line.
  > This lets binat setups work again without having to recompile isakmpd.
  > OK ho, hshoexer. (cloder@)

  ~ crypto.c                              

  > Do not log before/after crypto blobs at log level 10.  Makes -DA=10
  > useable without enduring pages of hex. OK hshoexer (cloder@)

  ~ GNUmakefile                           ~ Makefile
  ~ cert.c                                ~ conf.c
  ~ ike_auth.c                            ~ ike_quick_mode.c
  ~ ipsec.c                               ~ key.c
  ~ libcrypto.c                           ~ libcrypto.h
  ~ pf_key_v2.c                           ~ x509.c

  > Always compile X509 support. Almost everyone uses it.  Makes the code
  > much easier to read and to maintain.
  > OK and testing by hshoexer@, more testing by me (cloder@)

  ~ Makefile                              ~ GNUmakefile

  > disable freeswan support
  > ok ho@ some time ago (hshoexer@)

  - sysdep/freeswan/GNUmakefile.sysdep    - sysdep/freeswan/Makefile.sysdep
  - sysdep/freeswan/README                - sysdep/freeswan/klips.c
  - sysdep/freeswan/klips.h               - sysdep/freeswan/sysdep-os.h
  - sysdep/freeswan/sysdep.c              - sysdep/freeswan/sys/queue.h

  > RIP freeswan (hshoexer@)

  ~ isakmpd.8                             

  > ipsecadm lives in section 8, not 4;
  > missing word; (jmc@)

  ~ policy.h                              ~ regress/rsakeygen/Makefile
  ~ regress/x509/Makefile                 ~ sysdep/linux/GNUmakefile.sysdep

  > Now that X509 is de-featurized, no need for USE_X509 in regress tests.
  > Start compiling the X509 regress test again, for the first time since
  > 2002 when DLOPEN stuff was removed.  Kill remnants of DLOPEN defines
  > left around in other Makefiles.  Allow isakmpd to compile if USE_KEYNOTE
  > is not defined. (cloder@)

mount_udf

  ~ mount_udf.8                           

  > - fix another 'an udf' mistake i made
  > - enhance 'history' and add an 'authors' section (pedro@)

  ~ mount_udf.8                           

  > whitespace and punctuation; (jmc@)

quotacheck

  ~ quotacheck.c                          

  > spacing found while digging for bugs (deraadt@)

== share ============================================================= 06/09 ==

  http://www.openbsd.org/cgi-bin/cvsweb/src/share

man

  ~ man4/man4.i386/Makefile               

  > -iy (brad@)

  - man4/man4.i386/iy.4                   

  > nuke a second driver for ISA Intel EtherExpress PRO/10 cards which
  > came from NetBSD and has never actually been used. (brad@)

  ~ man4/auich.4                          

  > ich6 support and deuglify long slash seqs (mickey@)

  ~ man4/fxp.4                            ~ man4/man4.amd64/intro.4
  ~ man4/man4.i386/ie.4                   ~ man4/man4.i386/intro.4

  > iy(4) has been removed; (jmc@)

  ~ man4/auich.4                          

  > add new nforce stuff (mickey@)

  ~ man8/afterboot.8                      

  > since csh is no longer the default root shell, and advice to not log in as
  > `root' was removed in dot.login -r1.11, remove text which says it happens;
  > tweak the rest while i'm in here;
  > spotted by david@ (jmc@)

== sys =============================================================== 07/09 ==

  http://www.openbsd.org/cgi-bin/cvsweb/src/sys

arch/arm/xscale

  ~ pxa2x0_ohci.c                         ~ pxa2x0reg.h

  > Argument to DELAY should be usecs, not hz.
  > ok dlg@ (pascoe@)

arch/sparc64/dev

  ~ comkbd_ebus.c                         

  > No extra newline if keyboard plugged but not console. (miod@)

  ~ creator.c                             ~ creator_mainbus.c
  ~ creator_upa.c                         

  > Off-by-ones in register comparisons. (miod@)

dev/ic

  ~ ath.c                                 

  > fix wep by adjusting the total packet length (reyk@)

dev/isa

  - if_iy.c                               

  > nuke a second driver for ISA Intel EtherExpress PRO/10 cards which
  > came from NetBSD and has never actually been used. (brad@)

dev/pci

  ~ if_sis.c                              ~ if_sisreg.h

  > - Fix for an issue with excessive collisions in half duplex mode.
  > - Nail the short cable problem the exact way National says it should be.
  > - Make sure to clear any pending interrupts when we stop the interface.
  > - Do not tell the hardware to send when there were no packets enqueued.
  > - Close a theoretical race:  By the time the watchdog comes around
  > the interface may have been stopped, so we should not restart it.
  > From FreeBSD
  > Tested by ho@ on a few Soekris boxes and martin@ on a 900 (brad@)

  ~ if_bgereg.h                           

  > fix parenthesis for BGE_RCB_MAXLEN_FLAGS
  > ok krw@ (brad@)

  ~ auich.c                               

  > ich6 support from freebsd via [Bjorn] Sandell <biorn at dce.chalmers.se  > (mickey@)

  ~ auich.c                               

  > from netbsd:
  > Some SiS 7012 chips come muted by default, so un-mute all SiS 7012 chips at
  > attach time the way Linux does it.	Reported and tested by Jonathan
  > Schleifer, I checked it didn't break my own 7012 device which doesn't need
  > that manipulation. (mickey@)

  ~ pcidevs                               

  > nforce4 devs and nforce2-400 ac97 (mickey@)

  ~ pcidevs.h                             ~ pcidevs_data.h

  > regen (mickey@)

  ~ auich.c                               

  > match on nforce4 and nforce2-400 ac97 as well; form netbsd (mickey@)

  ~ if_san_common.c                       ~ if_san_xilinx.c

  > prevent non-root users to invoke wanpipe ioctls.
  > noticed by uwe@, looks ok deraadt@ (canacar@)

  ~ auich.c                               

  > Don't clobber the returned sample rate when the AC97 changes it.
  > This lets an application know, should it bother testing, that a
  > requested bit rate is not available.   'looks ok to me' mickey@ (marc@)

  ~ pcidevs.h                             ~ pcidevs_data.h

  > regen. (fgsch@)

  ~ pcidevs                               

  > winmodem found in vaio PCGFR315M; ok deraadt@ (fgsch@)

isofs/udf

  ~ udf_vnops.c                           

  > make sure *vpp == NULL if udf_lookup() returns an error (pedro@)

net

  ~ if.c                                  

  > Return EINVAL in SIOCSIFLLADDR if new lladdr is a multicast address.
  > OK derradt@ tedu@ (claudio@)

netinet

  ~ tcp_input.c                           ~ tcp_output.c
  ~ tcp_var.h                             

  > add tcp sack stats, similar to freebsd; ok deraadt (markus@)

scsi

  ~ sd.c                                  ~ cd.c

  > Clean up some logic. Use better error handling for sd devices. Be
  > quiet if raw partition access tried. Be more careful with
  > SDEV_MEDIA_LOADED. Inspired by some NetBSD changes.
  > ok marco@ (krw@)

  ~ cd_scsi.c                             ~ ch.c
  ~ scsi_base.c                           ~ scsiconf.h
  ~ sd_scsi.c                             ~ ss_mustek.c
  ~ ss_mustek.h                           ~ st.c

  > Factor out scsi_mode_sense() and scsi_mode_select() like NetBSD does.
  > Eliminate some duplicate structures and defines along the way. SCSI CD
  > drives will now pay attention to the 'flags' parameter for MODE SENSE
  > and MODE SELECT commands. No other functional change.
  > ok tdeval@ (krw@)

== usr.bin =========================================================== 08/09 ==

  http://www.openbsd.org/cgi-bin/cvsweb/src/usr.bin

ftp

  ~ fetch.c                               

  > a : in the path is not a port # indicator so don't take ones into
  > account after the first / behind the hostname, ok jaredy (henning@)

netstat

  ~ inet.c                                

  > add tcp sack stats, similar to freebsd; ok deraadt (markus@)

ssh

  ~ ssh-keygen.c                          

  > sync and sort usage(). ok jmc@ markus@ (otto@)

== usr.sbin ========================================================== 09/09 ==

  http://www.openbsd.org/cgi-bin/cvsweb/src/usr.sbin

bgpd

  ~ session.h                             

  > magic police, Jason Ackley <jason at ackley.nethenning@)

httpd

  ~ src/include/ap_compat.h               

  > ap_compat.h defines MD5Update to one thing, but now that we use functions
  > from md5.h this breaks things which use the compatibility header, such as
  > some 3rd party modules.
  > so simply remove this define from ap_compat.h
  > original breakage found and reported in ports/www/mod_log_sql by naddy@
  > ok henning@ (niallo@)

ospfd

  ~ buffer.c                              

  > Set errno in case buf_realloc() fails because the limit of the buffer is
  > reached. (claudio@)

  ~ auth.c                                ~ database.c
  ~ hello.c                               ~ lsack.c
  ~ lsreq.c                               ~ lsupdate.c
  ~ ospfd.h                               ~ ospfe.h
  ~ packet.c                              

  > Use the dynamic buffer API for packet generation and sending.
  > OK norby@ (claudio@)

===============================================================================



More information about the odc mailing list