[odc] Daily src changes for 2009-01-29

ODC auto at squish.net
Fri Jan 30 07:00:01 GMT 2009


OpenBSD src changes summary for 2009-01-29
==========================================

bin/ksh                                 lib/libc
regress/bin                             regress/sbin
sbin/ipsecctl                           sbin/isakmpd
sbin/route                              share/man
sys/arch/i386/isa                       sys/arch/macppc/dev
sys/arch/mvme88k/dev                    sys/arch/mvme88k/mvme88k
sys/arch/socppc/dev                     sys/compat/common
sys/compat/svr4                         sys/conf
sys/dev/ic                              sys/dev/pci
sys/kern                                sys/net
sys/net80211                            sys/netinet
sys/netmpls                             sys/sys
usr.sbin/cron                           usr.sbin/ntpd
usr.sbin/ospf6ctl                       usr.sbin/ospf6d
usr.sbin/smtpd                          usr.sbin/tcpdump
usr.sbin/wake                           usr.sbin/ypldap

== bin =============================================================== 01/07 ==

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

ksh

  ~ c_sh.c                                ~ eval.c
  ~ exec.c                                ~ jobs.c
  ~ ksh.1                                 ~ main.c
  ~ proto.h                               ~ sh.1

  > pass "xerrok" status across the execution call stack to more closely
  > match what both POSIX and ksh.1 already describe in regards to set
  > -e/errexit's behavior in determining when to exit from nonzero return
  > values.
  > specifically, the truth values tested as operands to `&&' and `||', as
  > well as the resulting compound expression itself, along with the truth
  > value resulting from a negated command (i.e. a pipeline prefixed `!'),
  > should not make the shell exit when -e is in effect.
  > issue reported by matthieu.
  > testing matthieu, naddy.
  > ok miod (earlier version), otto.
  > man page ok jmc. (jaredy@)

== lib =============================================================== 02/07 ==

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

libc

  ~ stdio/ferror.3                        

  > from Yoshihiko Sarumaru, freebsd pr #76333: fseek(3) can clear EOF too;
  > ok millert (jmc@)

  ~ sys/stat.2                            

  > Update to reflect renaming of the st_[acm]timespec members to st_[acm]tim
  > per the POSIX 1003.1-2008 spec.  Also, tweak the information about what
  > system calls hit which timestamps
  > ok jmc@ (guenther@)

== regress =========================================================== 03/07 ==

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

bin

  ~ ksh/seterror.sh                       

  > pass "xerrok" status across the execution call stack to more closely
  > match what both POSIX and ksh.1 already describe in regards to set
  > -e/errexit's behavior in determining when to exit from nonzero return
  > values.
  > specifically, the truth values tested as operands to `&&' and `||', as
  > well as the resulting compound expression itself, along with the truth
  > value resulting from a negated command (i.e. a pipeline prefixed `!'),
  > should not make the shell exit when -e is in effect.
  > issue reported by matthieu.
  > testing matthieu, naddy.
  > ok miod (earlier version), otto.
  > man page ok jmc. (jaredy@)

sbin

  - ipsecctl/ikefail10.in                 - ipsecctl/ikefail10.ok
  ~ ipsecctl/Makefile                     

  > Remove ikefail10 ipsecctl regression test as it always fails.  It
  > was expecting a certain parser error message.  Accepting the ikefail10
  > config file is not considered to be a bug anymore.
  > ok hshoexer@ (bluhm@)

== sbin ============================================================== 04/07 ==

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

ipsecctl

  ~ ipsec.conf.5                          

  > tweak previous; (jmc@)

  ~ parse.y                               

  > After checking that peer == NULL do not assign peer = NULL a few
  > lines later.  No functional change.
  > ok grunk@, hshoexer@ (bluhm@)

isakmpd

  ~ ipsec.c                               

  > Improve logging:
  > - in ipsec_delete_spi_list() a log_verbose is added, when a remote peer
  > sends us a delete message for an SA.  However, to avoid spamming the log
  > when SAs are deleted during re-keying, I only log_verbose, when the soft
  > timeout of the SA is not expired yet.  Thus only deletion of live SAs
  > gets logged.
  > - in ipsec_decode_ids() I remove the additonal printing of IP-Adresses in
  > hex as the addresses are already printed in CIDR.
  > - while there, apply some KNF
  > ok todd@, mpf@, bluhm@ (hshoexer@)

route

  ~ keywords.h                            ~ route.c

  > Introduce -mplslabel to allow ipv4/ipv6 packets enter MPLS clouds.
  > This is also consistent with mplslabel in ifconfig(8).
  > OK claudio@ laurent@ (michele@)

== share ============================================================= 05/07 ==

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

man

  ~ man4/route.4                          

  > My try at documenting the route socket filtering. ja ja dlg@ (claudio@)

  ~ man4/route.4                          

  > tweak previous; (jmc@)

  ~ man9/Makefile                         + man9/socreate.9

  > Add some kernel socket documentation. Originally from FreeBSD with some
  > tweaks by me to match the OpenBSD reality. Additional help from jmc@
  > ja ja ja dlg@ and OK blambert@ (claudio@)

  ~ man9/socreate.9                       

  > - fix .Nm and .Dt
  > - remove a duplicate (jmc@)

  ~ man4/netintro.4                       

  > Document SIOCSIFPRIORITY and friend. Found on my HD. OK jmc@ (claudio@)

== sys =============================================================== 06/07 ==

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

arch/i386/isa

  ~ clock.c                               

  > Only start using the rtc for statclock after we've received the first
  > interrupt.	On some machines the rtc doesn't generate interrupts and we
  > would
  > end up not running statclock() at all.
  > ok miod@, art@ (kettenis@)

arch/macppc/dev

  ~ adb.c                                 

  > Some of the latest macppc laptops with no adb subsystem at all,
  > spontaneously
  > powerdown if we try to issue pmu commands. Skip these if no adb device
  > appears in the device tree.
  > Adapted from NetBSD. (miod@)

arch/mvme88k/dev

  ~ vs.c                                  

  > Add a few cache operations after I/O completes. (miod@)

arch/mvme88k/mvme88k

  ~ m88110.c                              

  > The ``flush before inval'' 88410 workaround needs to be done in
  > m88410_initialize_cpu() as well. Since we can't flush unknown data, just
  > make sure all 88410 lines are filled with valid data before flushing it.
  > This should fix the spurious ``197DP hangs or loops on osiop errors after
  > autoconf'' issue I've been seeing. (miod@)

arch/socppc/dev

  ~ if_tsec.c                             

  > Switch tsec(4) over to using MCLGETI when allocating mbufs for the rx ring.
  > (kettenis@)

  ~ if_tsec.c                             

  > Make multicast actually work this time. (kettenis@)

compat/common

  ~ vfs_syscalls_35.c                     

  > Switch struct stat's timespec members to the names standardized in
  > POSIX 1003.1-2008, with compatibility macros for the names used in
  > previous version of OpenBSD.  Update all the references in the
  > kernel to use the new, standard member names.
  > ok'ed by miod@, otto@; ports build test by naddy@ (guenther@)

compat/svr4

  ~ svr4_stat.c                           

  > Switch struct stat's timespec members to the names standardized in
  > POSIX 1003.1-2008, with compatibility macros for the names used in
  > previous version of OpenBSD.  Update all the references in the
  > kernel to use the new, standard member names.
  > ok'ed by miod@, otto@; ports build test by naddy@ (guenther@)

conf

  ~ files                                 

  > Split the address selection from pools away from pf.c and put it in
  > pf_lb.c. This will ease the process of adding more selection types
  > without bloatening pf.c even more.
  > ok and a weird death threat, henning@
  > raised eyebrow, dlg@ (pyr@)

dev/ic

  ~ sti.c                                 

  > Oops, correct logic when requested font index is out of bounds. Thankfully
  > it's a can't happen situation. (miod@)

  ~ siop_common.c                         

  > ``beginning'' is spelled with 2 `g' and 3 `n'. (miod@)

dev/pci

  ~ pcidevs                               

  > Soundblaster stuff. from brad. (marco@)

  ~ pcidevs.h                             ~ pcidevs_data.h

  > regen (marco@)

  - drm/drm_drawable.c                    - drm/drm_internal.h
  ~ drm/drmP.h                            ~ drm/drm_drv.c
  ~ drm/files.drm                         

  > Rip out the drm_drawable support. There is no need for the xserver to
  > push cliprects into the kernel anymore so no op the ioctls. The only
  > consumer of this information died a couple of months ago.
  > When i've got removal of these calls from userland upstream i'm going to
  > remove them from the interface entirely. (oga@)

  ~ drm/drmP.h                            ~ drm/drm_drv.c
  ~ drm/drm_ioctl.c                       

  > drm_getmap and interface can die too. (oga@)

  - drm/drm_vm.c                          ~ drm/drmP.h
  ~ drm/drm_drv.c                         ~ drm/files.drm

  > Move drmmmap to the main driver file.  It's stupid to have a file with
  > one (not particlarly large) function in it. (oga@)

  - drm/drm_ioctl.c                       ~ drm/drmP.h
  ~ drm/drm_drv.c                         ~ drm/files.drm

  > Similar to my last commit, it's stupid to have a file with two very
  > small functions in it. Move them to main driver file. (oga@)

kern

  ~ sys_pipe.c                            ~ uipc_usrreq.c
  ~ vfs_vnops.c                           

  > Switch struct stat's timespec members to the names standardized in
  > POSIX 1003.1-2008, with compatibility macros for the names used in
  > previous version of OpenBSD.  Update all the references in the
  > kernel to use the new, standard member names.
  > ok'ed by miod@, otto@; ports build test by naddy@ (guenther@)

  ~ kern_sig.c                            

  > Remove KASSERT() that could trigger when an rthread in sigwait()
  > got multiple signals before tsleep() could wakeup.	Also, POSIX
  > says that sigwait() should never return EINTR, so map that to
  > ERESTART.
  > ok kurt@, tedu@
  > fixes the panic encountered by ariane@ with kaffe (guenther@)

net

  ~ radix_mpath.c                         

  > Fix a crash found by bluhm@ in rn_mpath_reprio(). The rn_mklist handling
  > was
  > wrong which resulted in a NULL rm_leaf which caussed a NULL deref a bit
  > later.
  > Check against the right route nodes to figure out if the rm_leaf needs to
  > be
  > updated. Tested by bluhm@, OK dlg@ (claudio@)

  ~ pf_norm.c                             

  > move some code around in preparation for future work:
  > break out the code that doesn't deal with fragment reassembly and only
  > modifies stuff in the ip header to their own functions. pass them what they
  > need instead of making them get the info from a rule ptr.
  > ok dlg ryan (henning@)

  ~ pf.c                                  ~ pfvar.h
  + pf_lb.c                               

  > Split the address selection from pools away from pf.c and put it in
  > pf_lb.c. This will ease the process of adding more selection types
  > without bloatening pf.c even more.
  > ok and a weird death threat, henning@
  > raised eyebrow, dlg@ (pyr@)

net80211

  ~ ieee80211_node.c                      

  > fix a panic that occurs when IEEE80211_STA_ONLY is defined (SMALL_KERNEL).
  > noticed by ray@, more investigation from sthen@
  > fix tested by sthen@ (damien@)

netinet

  ~ ip_output.c                           

  > Always zero the IP checksum field for packets and packet fragments
  > being passed down if using HW checksum offload.
  > From Brad, inspired by NetBSD/FreeBSD.  ok markus@ (naddy@)

netmpls

  ~ mpls.h                                

  > mpls_operation have to be 2 bytes long to store
  > operations correctly
  > ok claudio@ laurent@ (michele@)

sys

  ~ stat.h                                

  > Switch struct stat's timespec members to the names standardized in
  > POSIX 1003.1-2008, with compatibility macros for the names used in
  > previous version of OpenBSD.  Update all the references in the
  > kernel to use the new, standard member names.
  > ok'ed by miod@, otto@; ports build test by naddy@ (guenther@)

  ~ socketvar.h                           

  > KNF (whitespace fixes).  ok grunk@ (bluhm@)

== usr.sbin ========================================================== 07/07 ==

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

cron

  ~ cron.c                                ~ crontab.c

  > tweak usage() (sobrado@)

ntpd

  ~ ntpd.c                                

  > log warning when can't read drift file; allows detection of
  > corrupt contents. ok henning@ (stevesk@)

ospf6ctl

  ~ ospf6ctl.c                            

  > Fix offset calculation for lsa_prefix structs.
  > ok claudio@ (stsp@)

  ~ ospf6ctl.c                            

  > Don't print a header line above every individual Intra-Area-Prefix LSA
  > in the detailed DB output. Now behaves the same as with other LSA types.
  > (stsp@)

  ~ ospf6ctl.c                            

  > One more step on the long road of beautifying ospf6ctl output:
  > The Link State ID in OSPFv3 is just a unique identifier and has
  > lost its semantic meaning for a most LSA types.
  > Print the correct semantic meaning for those that still have one,
  > and don't print any additional information for those that don't.
  > ok claudio@ (stsp@)

ospf6d

  ~ rde.c                                 

  > Improve debugging message in orig_intra_lsa_net().
  > Print not just the area, but also the interface to the link the LSA
  > is generated for.
  > ok claudio@ (stsp@)

  ~ rde.c                                 

  > Hard-code the Link State ID of Intra-Area-Prefix LSAs referencing Network
  > LSAs to zero. We were using the interface index, which is not quite right
  > for this type of LSA.
  > This is part of a greater scheme:
  > Intra-Area-Prefix LSAs referencing Router LSAs (not yet implemented)
  > will initially have their Link State ID hard-coded to one.
  > Then, as soon as we implement fragmentation of Intra-Area-Prefix LSAs,
  > Link State IDs for both types of Intra-Area-Prefix LSA will have to be
  > generated dynamically in a non-overlapping fashion.
  > discussed with claudio@ (stsp@)

  ~ rde_lsdb.c                            

  > lsa_get_prefix() should return the number of bytes consumed and not the
  > number of bytes left in the buffer. OK stsp@ (claudio@)

  ~ rde_lsdb.c                            

  > In Link LSAs, #prefixes is a 32bit value.
  > But in Intra-Area-Prefix LSAs, #prefixes is a 16bit value.
  > lsa_intra_a_pref_check() was failing on little-endian machines because
  > it was trying to byte-swap a 32bit value.
  > ok claudio@ (stsp@)

  ~ interface.c                           ~ ospfe.c
  ~ ospfe.h                               

  > Originate Intra-Area-Prefix LSAs which reference a router LSA.
  > Routers can now advertise prefixes with global scope throughout
  > an area. This paves the way towards doing SPF calculations to
  > generate routing tables for a single area.
  > ok claudio@ (stsp@)

  ~ ospfe.c                               

  > Correctly byte-swap the link state ID when building Intra-Area-Prefix
  > LSAs which refer to Router LSAs.
  > ok claudio@ (stsp@)

smtpd

  ~ enqueue.c                             

  > - remove a very annoying debug printf() (gilles@)

  - showqueue.c                           ~ queue.c
  ~ queue_shared.c                        ~ runner.c
  ~ smtpctl.c                             ~ smtpd.h
  ~ smtpctl/Makefile                      

  > Common queue walking code for smtpd and smtpctl. Kills majority of
  > showqueue.c,
  > the remaining code was moved to queue_shared.c; ok gilles@ (jacekm@)

  ~ mta.c                                 ~ smtpd.h
  ~ ssl.c                                 

  > bring initial support for SSL in the mta part of smtpd, allowing for:
  > accept for domain "openbsd.org" relay via ssmtp "mx1.example.org"
  > to ensure that deliveries for whatever at openbsd.org goes through an SSL
  > session
  > to mx1.example.org (gilles@)

  ~ mta.c                                 ~ smtpd.h
  ~ ssl.c                                 

  > initial starttls support in mta, this allows:
  > accept for domain "openbsd.org" relay via tls "mx.example.org"
  > to ensure the relaying of mail for whoever at openbsd.org will happen through
  > a secure tls (STARTTLS) session. failure to establish a tls session will be
  > considered as a permanent failure. As a side effect:
  > accept for domain "openbsd.org" relay via ssl "mx.example.org"
  > can now work as well and ensure that the relaying happens through ssmtp OR
  > tls, but never through an unsafe channel. no need to specify a port, they
  > are automatically detected if not specified.
  > still a work in progress, don't expect that it will work flawlessly.
  > (gilles@)

  ~ mta.c                                 

  > fix a bug in mta's event masking (gilles@)

  ~ enqueue.c                             ~ mta.c
  ~ smtp_session.c                        ~ smtpd.h
  ~ store.c                               

  > Received header line was incomplete for mail submitted through the enqueuer
  > as well as for some outgoing messages, this is fixed now (gilles@)

  ~ smtp_session.c                        

  > - remove debug message (gilles@)

  ~ mta.c                                 ~ smtp_session.c
  ~ smtpd.h                               

  > missing prototype for session_respond() in mta, move session_respond()'s
  > prototype to smtpd.h
  > spotted and diff from Oleg Safiullin <form at pdp-11.org.ru> (gilles@)

  ~ forward.c                             

  > better parsing of .forward files
  > ok gilles@ (form@)

  ~ control.c                             ~ parser.c
  ~ parser.h                              ~ queue.c
  ~ runner.c                              ~ smtp.c
  ~ smtp_session.c                        ~ smtpctl.c
  ~ smtpd.c                               ~ smtpd.h

  > Implement "smtpctl show stats"; ok gilles@ (jacekm@)

  ~ smtpctl.c                             

  > fix compiler warning (jacekm@)

tcpdump

  ~ print-bgp.c                           ~ print-icmp6.c

  > In tcpdump some printf() had an additional \n at the end.  Removing
  > that new line restores the one-line -> one-packet semantics.
  > ok hshoexer@, henning@, markus@ (bluhm@)

wake

  ~ wake.c                                

  > Round brackets are not required. Back out previous commit. (michele@)

  ~ wake.c                                

  > simplify the bpf opening loop a bit.
  > ``shorter is always better'' and ok mbalmer@ (pyr@)

  ~ wake.c                                

  > Report error if wake on lan frame is not sent correctly.
  > Loop in the really unlikely event of a non atomic write.
  > ok mbalmer@ (pyr@)

  ~ wake.c                                

  > detect truncation with strlcpy, not strlen.
  > ok mbalmer@ (pyr@)

  ~ Makefile                              ~ wake.c

  > * Add more CFLAGS commonly found in src/usr.sbin/
  > * Add prototype for wake()
  > * Add __dead to usage (for lint)
  > * Do not rely on pointer value for asprintf return as advised
  > * Make sure lint is happy
  > some comments and ok mbalmer@ (pyr@)

  ~ wake.8                                ~ wake.c

  > Correct copyright years. (mbalmer@)

  ~ wake.c                                

  > Correct capitalization.
  > From Thomas Pfaff. (pyr@)

  ~ wake.c                                

  > Better error handling. (mbalmer@)

  ~ wake.8                                

  > various improvements; help/ok mbalmer (jmc@)

ypldap

  ~ ypldap.c                              

  > No need to use RB_FIND here, RB_INSERT does that job for us.
  > ok aschrijver@ (form@)

  ~ aldap.c                               

  > Make anonymous binds work properly. (aschrijver@)

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


More information about the odc mailing list