[odc] Daily src changes for 2008-06-09

ODC auto at squish.net
Tue Jun 10 07:00:01 BST 2008


OpenBSD src changes summary for 2008-06-09
==========================================

distrib/notes                           distrib/sets
etc/group                               etc/master.passwd
etc/netstart                            etc/rc
etc/rc.conf                             gnu/usr.bin/gcc
gnu/usr.sbin/sendmail                   lib/libarch
lib/libc                                regress/sys
regress/usr.bin                         sbin
sbin/fsck_ffs                           sbin/ldattach
sbin/nmeaattach                         sbin/slattach
share/man                               sys/arch/amd64/amd64
sys/arch/amd64/include                  sys/arch/armish/stand
sys/arch/armish/stand/boot              sys/arch/i386/i386
sys/arch/i386/include                   sys/arch/i386/pci
sys/arch/mips64/mips64                  sys/arch/socppc/stand
sys/arch/socppc/stand/boot              sys/arch/socppc/stand/mkboot
sys/arch/sparc/sparc                    sys/arch/sparc64/include
sys/arch/sparc64/sparc64                sys/arch/vax/vax
sys/conf                                sys/crypto
sys/dev                                 sys/dev/i2c
sys/dev/pci                             sys/dev/usb
sys/isofs/cd9660                        sys/isofs/udf
sys/kern                                sys/miscfs/procfs
sys/msdosfs                             sys/net
sys/net80211                            sys/netinet
sys/netinet6                            sys/nfs
sys/sys                                 sys/ufs/ext2fs
sys/ufs/ufs                             sys/uvm
usr.bin/cvs                             usr.bin/ssh
usr.sbin/bind                           usr.sbin/ndp
usr.sbin/ntpd                           usr.sbin/pkg_add
usr.sbin/procmap                        usr.sbin/rtadvd
usr.sbin/rtsold                         usr.sbin/sysmerge

== distrib =========================================================== 01/10 ==

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

notes

  ~ alpha/hardware                        ~ amd64/hardware
  ~ i386/hardware                         ~ sparc64/hardware

  > add IODATA WN-G54/US and MELCO WLI-U2-KAMG54 entries to the uath(4) list;
  > (jmc@)

sets

  ~ 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.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.socppc                  ~ lists/base/md.sparc
  ~ lists/base/md.sparc64                 ~ lists/base/md.vax
  ~ lists/base/md.zaurus                  ~ lists/comp/mi
  ~ lists/man/mi                          

  > sync (deraadt@)

== etc =============================================================== 02/10 ==

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

group

  ~ group                                 

  > drop root privileges in rtadvd to _rtadvd
  > ok deraadt@, reyk@, pyr@ (rainer@)

master.passwd

  ~ master.passwd                         

  > drop root privileges in rtadvd to _rtadvd
  > ok deraadt@, reyk@, pyr@ (rainer@)

netstart

  ~ netstart                              

  > Ensure that hostname.* files are also re-chowned to root.wheel at each
  > boot as discussed with claudio while eating tasty donairs.	ok todd
  > (deraadt@)

  ~ netstart                              

  > warn once not 3 times in case of a non existent file, discussed with
  > deraadt
  > originally pointed out by Johan Torin (todd@)

rc

  ~ rc                                    

  > Replace nmeattach (which will be removed) with ldattach. (mbalmer@)

rc.conf

  ~ rc.conf                               

  > Replace nmeattach (which will be removed) with ldattach. (mbalmer@)

== gnu =============================================================== 03/10 ==

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

usr.bin/gcc

  ~ gcc/config/alpha/openbsd.h            

  > Make gcc on alpha use -mieee by default.  This fixes many FPE problems
  > with several ports on alpha and does not produce a significant slowdown.
  > Suggestion to patch gcc directly from millert@
  > Diff and ok from miod@
  > ok pvalchev@, martin@, deraadt@ (merdely@)

usr.sbin/sendmail

  ~ cf/cf/openbsd-bulk.mc                 ~ cf/cf/openbsd-lists.mc

  > Disable ident queries
  > Crank max queue children in bulk config
  > Use the C flag in DAEMON_OPTIONS instead of nocanonify in bulk config
  > (millert@)

  ~ cf/cf/Makefile                        

  > Don't be overzealous when cleaning up, once is enough.
  > spotted by ajacoutot@, ok ajacoutot@, millert@ (pyr@)

== lib =============================================================== 04/10 ==

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

libarch

  - amd64/amd64_get_mtrr.2                - amd64/amd64_get_mtrr.c
  - amd64/amd64_set_mtrr.c                ~ amd64/Makefile

  > Remove an arch-specific API that wasn't available in the first place, and
  > will be converted to a more generic one soon.
  > ok deraadt, oga (phessler@)

libc

  ~ crypt/arc4random.3                    

  > fix an Xr, and use a more appropriate macro; (jmc@)

  ~ sys/access.2                          

  > Update access(2) to have modern semantics with respect to X_OK and
  > the superuser.  access(2) will now only indicate success for X_OK on
  > non-directories if there is at least one execute bit set on the file.
  > OK deraadt@ thib@ otto@ (millert@)

== regress =========================================================== 05/10 ==

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

sys

  ~ crypto/Makefile                       

  > +aesxts subdir (djm@)

  + crypto/aesxts/Makefile                + crypto/aesxts/aes_xts.c

  > regress tests for crypto(4) CRYPTO_AES_XTS using test vectors from
  > IEEE P1619-2007 standards; ok hshoexer@ (djm@)

usr.bin

  ~ ssh/test-exec.sh                      

  > Don't generate putty keys if we're not going to use them.  ok djm
  > (dtucker@)

== sbin ============================================================== 06/10 ==

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

sbin

  ~ Makefile                              

  > Remove nmeaattch(8), which is superseeded by ldattach(8).
  > ok deraadt (mbalmer@)

fsck_ffs

  ~ setup.c                               

  > The new newfs(8) code causes alternate superblocks to end up in
  > different locations than before.  Actually, the disklabel does not
  > contain enough info to completely reconstruct the locations of all
  > alternate sb's.  So use a hardcoded list of all possible 1st alternate
  > sb locations, but don't forget to verify against the label.
  > ok millert@ thib@ (otto@)

ldattach

  ~ Makefile                              ~ ldattach.8
  ~ ldattach.c                            + atomicio.c
  + atomicio.h                            

  > Add the '-p' option to ldattach(8) to pass data received from the device
  > to the master device of a pty(4) pair.  The name of the slave device is
  > written to standard output.
  > This is useful for applications like e.g. gpsd from the misc/gpsd port
  > that also use the serial data stream (e.g. nmea(4) as a time source and
  > gpsd to get at positional data).
  > help and ok deraadt, makes ckuethe happy. (mbalmer@)

  ~ ldattach.8                            

  > zap trailing whitespace; (jmc@)

  ~ ldattach.c                            

  > Don't reverence slattach(8) or nmeaattach(8) in a comment. (mbalmer@)

nmeaattach

  ~ nmeaattach.8                          

  > nmeaattach(8) is now deprecated.  Use ldattach(8) instead. (mbalmer@)

  - Makefile                              - nmeaattach.8
  - nmeaattach.c                          

  > Remove nmeaattch(8), which is superseeded by ldattach(8).
  > ok deraadt (mbalmer@)

slattach

  ~ slattach.8                            

  > slattach(8) is now deprecated.  Use ldattach(8) instead. (mbalmer@)

  ~ slattach.8                            

  > - move the "this app is deprecated" blurb to the start, where it's more
  > likely to be read
  > - sort SEE ALSO (jmc@)

== share ============================================================= 07/10 ==

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

man

  ~ man4/pci.4                            ~ man4/ix.4

  > tweak previous; (jmc@)

  ~ man4/uvideo.4                         

  > - sort the hardware list
  > - knock out the video Xr until we get a video(4) man page (jmc@)

  ~ man9/Makefile                         ~ man9/random.9

  > man bits for arc4random_bytes => arc4random_buf rename; ok deraadt@ (djm@)

  ~ man9/crypto.9                         

  > mention CRYPTO_AES_CTR, CRYPTO_AES_XTS
  > describe special requirement for passing in IV/block number for
  > CRYPTO_AES_XTS (djm@)

  ~ man4/crypto.4                         

  > mention CRYPTO_AES_CTR and CRYPTO_AES_XTS (djm@)

  ~ man9/crypto.9                         

  > tweak previous; (jmc@)

  ~ man9/uvm.9                            

  > Define a new flag, UVM_FLAG_HOLE, for uvm_map to create a vm_map_entry of
  > a new etype, UVM_ET_HOLE, meaning it has no backend.
  > UVM_ET_HOLE entries (which should be created as UVM_PROT_NONE and with
  > UVM_FLAG_NOMERGE and UVM_FLAG_HOLE) are skipped in uvm_unmap_remove(), so
  > that pmap_{k,}remove() is not called on the entry.
  > This is intended to save time, and behave better, on pmaps with MMU holes
  > at process exit time.
  > ok art@, kettenis@ provided feedback as well. (miod@)

  ~ man9/Makefile                         + man9/idgen.9

  > man page for crypto/idgen.h functions: idgen32() and idgen32_init()
  > ok deraadt@ (djm@)

== sys =============================================================== 08/10 ==

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

arch/amd64/amd64

  ~ copy.S                                

  > Remove fillw(). Bonus points if you do not remember what it was used for.
  > (miod@)

arch/amd64/include

  ~ cpu.h                                 

  > Remove fillw(). Bonus points if you do not remember what it was used for.
  > (miod@)

arch/armish/stand

  ~ Makefile.inc                          

  > Place manuals in the MD directories; from pierre.riteau at gmail (deraadt@)

arch/armish/stand/boot

  ~ Makefile                              

  > Place manuals in the MD directories; from pierre.riteau at gmail (deraadt@)

arch/i386/i386

  ~ via.c                                 

  > rename arc4random_bytes => arc4random_buf to match libc's nicer name;
  > ok deraadt@ (djm@)

  ~ locore.s                              

  > Remove fillw(). Bonus points if you do not remember what it was used for.
  > (miod@)

arch/i386/include

  ~ cpu.h                                 

  > Remove fillw(). Bonus points if you do not remember what it was used for.
  > (miod@)

arch/i386/pci

  ~ glxsb.c                               

  > rename arc4random_bytes => arc4random_buf to match libc's nicer name;
  > ok deraadt@ (djm@)

arch/mips64/mips64

  ~ lcore_access.S                        

  > Remove fillw(). Bonus points if you do not remember what it was used for.
  > (miod@)

arch/socppc/stand

  ~ Makefile.inc                          

  > Place manuals in the MD directories; from pierre.riteau at gmail (deraadt@)

arch/socppc/stand/boot

  ~ Makefile                              ~ boot.8

  > Place manuals in the MD directories; from pierre.riteau at gmail (deraadt@)

arch/socppc/stand/mkboot

  ~ Makefile                              

  > Place manuals in the MD directories; from pierre.riteau at gmail (deraadt@)

arch/sparc/sparc

  ~ pmap.c                                

  > Create a real holp by using uvm_map() with UVM_FLAG_HOLE in
  > pmap_remove_holes().
  > No functional change. (miod@)

arch/sparc64/include

  ~ pmap.h                                

  > Sparc64 MMUs have an address hole, too, and the pmap implementation we
  > are using has an even larger one, so implement pmap_remove_hole() to
  > prevent mmap() from ever reaching the hole.
  > feedback and ok kettenis@ (miod@)

arch/sparc64/sparc64

  ~ machdep.c                             

  > rename arc4random_bytes => arc4random_buf to match libc's nicer name;
  > ok deraadt@ (djm@)

  ~ pmap.c                                

  > Don't use u_int to store a 64-bit address.	Found with help from miod at .
  > Makes a Sun Fire V490 boot without spending ages in bus_dmamem_alloc(9).
  > (kettenis@)

  ~ pmap.c                                

  > Sparc64 MMUs have an address hole, too, and the pmap implementation we
  > are using has an even larger one, so implement pmap_remove_hole() to
  > prevent mmap() from ever reaching the hole.
  > feedback and ok kettenis@ (miod@)

arch/vax/vax

  ~ pmap.c                                

  > Create a real holp by using uvm_map() with UVM_FLAG_HOLE in
  > pmap_remove_holes().
  > No functional change. (miod@)

conf

  ~ files                                 

  > Introduce a facility to generate unpredictable 32 bit numbers with
  > near maximal (2^32) cycle times. These are useful for network
  > IDs in cases where there are negative consequences to ID prediction
  > and/or reuse.
  > Use the idgen32() functions to generate IPv6 IDs and NFS client/server
  > XIDs.
  > Pseudorandom permutation code in crypto/idgen.c based on public
  > domain skip32.c from Greg Rose.
  > feedback & ok thib@ deraadt@ (djm@)

crypto

  ~ cryptosoft.c                          

  > rename arc4random_bytes => arc4random_buf to match libc's nicer name;
  > ok deraadt@ (djm@)

  ~ rijndael.c                            ~ rijndael.h

  > constify arguments in wrapper functions; the lower level functions
  > were already done (djm@)

  ~ xform.c                               

  > add myself to copyright (djm@)

  ~ cryptodev.c                           ~ cryptodev.h
  ~ cryptosoft.c                          ~ xform.c
  ~ xform.h                               

  > Implement the AES XTS mode of operation for the crypto(9) framework.
  > XTS is a "tweaked" AES mode that has properties that are desirable
  > for block device encryption and it is specified in the IEEE
  > P1619-2007 standard for this purpose.
  > prodded by & ok hshoexer@ (djm@)

  + idgen.c                               + idgen.h

  > Introduce a facility to generate unpredictable 32 bit numbers with
  > near maximal (2^32) cycle times. These are useful for network
  > IDs in cases where there are negative consequences to ID prediction
  > and/or reuse.
  > Use the idgen32() functions to generate IPv6 IDs and NFS client/server
  > XIDs.
  > Pseudorandom permutation code in crypto/idgen.c based on public
  > domain skip32.c from Greg Rose.
  > feedback & ok thib@ deraadt@ (djm@)

dev

  ~ video.c                               ~ video_if.h

  > Allocate the video buffer based on the maximum frame size that's reported
  > by the device instead of using the statically set VIDEO_BUF_SIZE.
  > ok mglocker@ (robert@)

  ~ rnd.c                                 ~ rndvar.h
  ~ softraid.c                            

  > rename arc4random_bytes => arc4random_buf to match libc's nicer name;
  > ok deraadt@ (djm@)

  ~ video.c                               

  > according to miod pmap_update() is not need after pmap_extract() at all,
  > so remove it. (robert@)

  ~ video.c                               ~ video_if.h
  ~ videovar.h                            

  > Let the driver switch between read() and mmap() depending on what
  > the userland application wants.
  > OK robert@ (mglocker@)

  ~ rnd.c                                 

  > simplify math for arc4random_uniform() suggested by
  > Jinmei_Tatuya AT isc.org via jakob@
  > empirically verified for entire domain of upper_bound (djm@)

dev/i2c

  ~ adm1026.c                             

  > * fix the order of precedence in the divisor bits calculation
  > * correct the constant in the fanrpm formula
  > ok deraadt (cnst@)

dev/pci

  ~ hifn7751.c                            ~ if_ix.c
  ~ noct.c                                ~ safe.c
  ~ ubsec.c                               

  > rename arc4random_bytes => arc4random_buf to match libc's nicer name;
  > ok deraadt@ (djm@)

dev/usb

  ~ uvideo.c                              

  > Allocate the video buffer based on the maximum frame size that's reported
  > by the device instead of using the statically set VIDEO_BUF_SIZE.
  > ok mglocker@ (robert@)

  ~ uvideo.h                              

  > oops. this change was missing from the previous commit (robert@)

  ~ uvideo.c                              ~ uvideo.h

  > Let the driver switch between read() and mmap() depending on what
  > the userland application wants.
  > OK robert@ (mglocker@)

  ~ ehci.c                                ~ ehcivar.h

  > Don't play games with the PCD interrupt.  It delays the detection of
  > connect/disconnect events.	From NetBSD.
  > ok deraadt@ (kettenis@)

isofs/cd9660

  ~ cd9660_vnops.c                        

  > Update access(2) to have modern semantics with respect to X_OK and
  > the superuser.  access(2) will now only indicate success for X_OK on
  > non-directories if there is at least one execute bit set on the file.
  > OK deraadt@ thib@ otto@ (millert@)

isofs/udf

  ~ udf_vnops.c                           

  > Update access(2) to have modern semantics with respect to X_OK and
  > the superuser.  access(2) will now only indicate success for X_OK on
  > non-directories if there is at least one execute bit set on the file.
  > OK deraadt@ thib@ otto@ (millert@)

kern

  ~ init_main.c                           ~ kern_sysctl.c

  > rename arc4random_bytes => arc4random_buf to match libc's nicer name;
  > ok deraadt@ (djm@)

  ~ sysv_ipc.c                            ~ vfs_subr.c

  > Update access(2) to have modern semantics with respect to X_OK and
  > the superuser.  access(2) will now only indicate success for X_OK on
  > non-directories if there is at least one execute bit set on the file.
  > OK deraadt@ thib@ otto@ (millert@)

miscfs/procfs

  ~ procfs_vnops.c                        

  > Update access(2) to have modern semantics with respect to X_OK and
  > the superuser.  access(2) will now only indicate success for X_OK on
  > non-directories if there is at least one execute bit set on the file.
  > OK deraadt@ thib@ otto@ (millert@)

msdosfs

  ~ msdosfs_vnops.c                       

  > Update access(2) to have modern semantics with respect to X_OK and
  > the superuser.  access(2) will now only indicate success for X_OK on
  > non-directories if there is at least one execute bit set on the file.
  > OK deraadt@ thib@ otto@ (millert@)

net

  ~ if_spppsubr.c                         ~ pf.c

  > rename arc4random_bytes => arc4random_buf to match libc's nicer name;
  > ok deraadt@ (djm@)

net80211

  ~ ieee80211_input.c                     ~ ieee80211_node.c
  ~ ieee80211_proto.c                     

  > rename arc4random_bytes => arc4random_buf to match libc's nicer name;
  > ok deraadt@ (djm@)

netinet

  ~ ip_carp.c                             ~ ip_esp.c
  ~ ip_id.c                               ~ tcp_subr.c

  > rename arc4random_bytes => arc4random_buf to match libc's nicer name;
  > ok deraadt@ (djm@)

netinet6

  ~ ip6_id.c                              ~ ip6_input.c
  ~ ip6_output.c                          ~ ip6_var.h

  > Introduce a facility to generate unpredictable 32 bit numbers with
  > near maximal (2^32) cycle times. These are useful for network
  > IDs in cases where there are negative consequences to ID prediction
  > and/or reuse.
  > Use the idgen32() functions to generate IPv6 IDs and NFS client/server
  > XIDs.
  > Pseudorandom permutation code in crypto/idgen.c based on public
  > domain skip32.c from Greg Rose.
  > feedback & ok thib@ deraadt@ (djm@)

nfs

  ~ krpc_subr.c                           ~ nfs_subs.c
  ~ nfs_var.h                             

  > Introduce a facility to generate unpredictable 32 bit numbers with
  > near maximal (2^32) cycle times. These are useful for network
  > IDs in cases where there are negative consequences to ID prediction
  > and/or reuse.
  > Use the idgen32() functions to generate IPv6 IDs and NFS client/server
  > XIDs.
  > Pseudorandom permutation code in crypto/idgen.c based on public
  > domain skip32.c from Greg Rose.
  > feedback & ok thib@ deraadt@ (djm@)

  ~ nfs_subs.c                            ~ nfs_var.h

  > nfsm_get_xid() => nfs_get_xid() after thib@ informed me that the
  > nfsm_* functions are for mbufs (djm@)

  ~ nfs_vnops.c                           

  > Update access(2) to have modern semantics with respect to X_OK and
  > the superuser.  access(2) will now only indicate success for X_OK on
  > non-directories if there is at least one execute bit set on the file.
  > OK deraadt@ thib@ otto@ (millert@)

sys

  ~ vnode.h                               

  > Update access(2) to have modern semantics with respect to X_OK and
  > the superuser.  access(2) will now only indicate success for X_OK on
  > non-directories if there is at least one execute bit set on the file.
  > OK deraadt@ thib@ otto@ (millert@)

ufs/ext2fs

  ~ ext2fs_vnops.c                        

  > Update access(2) to have modern semantics with respect to X_OK and
  > the superuser.  access(2) will now only indicate success for X_OK on
  > non-directories if there is at least one execute bit set on the file.
  > OK deraadt@ thib@ otto@ (millert@)

ufs/ufs

  ~ ufs_vnops.c                           

  > Update access(2) to have modern semantics with respect to X_OK and
  > the superuser.  access(2) will now only indicate success for X_OK on
  > non-directories if there is at least one execute bit set on the file.
  > OK deraadt@ thib@ otto@ (millert@)

uvm

  ~ uvm.h                                 ~ uvm_extern.h
  ~ uvm_map.c                             

  > Define a new flag, UVM_FLAG_HOLE, for uvm_map to create a vm_map_entry of
  > a new etype, UVM_ET_HOLE, meaning it has no backend.
  > UVM_ET_HOLE entries (which should be created as UVM_PROT_NONE and with
  > UVM_FLAG_NOMERGE and UVM_FLAG_HOLE) are skipped in uvm_unmap_remove(), so
  > that pmap_{k,}remove() is not called on the entry.
  > This is intended to save time, and behave better, on pmaps with MMU holes
  > at process exit time.
  > ok art@, kettenis@ provided feedback as well. (miod@)

== usr.bin =========================================================== 09/10 ==

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

cvs

  ~ annotate.c                            

  > Check if RCS file exists before trying to retrieve revisions for annotate,
  > fixes segmentation fault on added but not yet committed files for example.
  > Based on diff by Igor Zinovik. (tobias@)

  ~ file.c                                

  > Stop fooling ourselfs by freeing cvs_directory_tag at the wrong position:
  > cvs_directory_tag is reinitalized every time we enter a new directory, so
  > make it clear by freeing before going into a new directory.
  > discussed with and ok joris (tobias@)

  ~ commit.c                              ~ entries.c
  ~ file.c                                ~ update.c

  > Properly handle merged files and conflicts which may arrise while merge.
  > Instead of ignoring all files which contain possible conflict markers,
  > only watch out for files which have actually been merged.
  > With input by and ok joris. (tobias@)

  ~ getlog.c                              

  > log already gets it FILE_UNKNOWN warning spewed out from cvs_file_classify,
  > so do not mention it again in cvs_log_local(); (joris@)

  ~ update.c                              

  > unused var (joris@)

ssh

  ~ sftp-server.c                         

  > Extend 32bit -> 64bit values for statvfs extension missed in previous
  > commit. (dtucker@)

  ~ PROTOCOL                              

  > Use a $OpenBSD tag so our scripts will sync changes. (dtucker@)

== usr.sbin ========================================================== 10/10 ==

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

bind

  ~ lib/isc/random.c                      

  > simplify math for arc4random_uniform() suggested by
  > Jinmei_Tatuya AT isc.org via jakob@
  > empirically verified for entire domain of upper_bound (djm@)

ndp

  ~ ndp.c                                 

  > Correctly round the sockaddrs so that ndp works on 64bit machines.
  > Stupid sockaddr_in6 has the worst size possible.
  > OK henning@ deraadt@ (claudio@)

ntpd

  ~ ntpd.conf.5                           ~ ntpd.h
  ~ parse.y                               ~ sensors.c

  > Allow outgoing replies from sensor-driven servers to have a
  > user-configurable reference ID, eg. "GPS" or "DCF"...
  > ok mbalmer (ckuethe@)

  ~ sensors.c                             

  > spacing (deraadt@)

  ~ parse.y                               

  > do not leak memory on failure in refid production; ok ckuethe henning
  > (deraadt@)

  ~ ntpd.h                                ~ sensors.c

  > rename refstr to refid since it is an int32; ok henning (deraadt@)

  ~ sensors.c                             

  > do not copy up to two garbage characters from a 1 char string into a 4 char
  > output buffer; ok henning (deraadt@)

pkg_add

  ~ OpenBSD/Subst.pm                      

  > allow direct copy of fh to another (espie@)

procmap

  ~ procmap.c                             

  > Define a new flag, UVM_FLAG_HOLE, for uvm_map to create a vm_map_entry of
  > a new etype, UVM_ET_HOLE, meaning it has no backend.
  > UVM_ET_HOLE entries (which should be created as UVM_PROT_NONE and with
  > UVM_FLAG_NOMERGE and UVM_FLAG_HOLE) are skipped in uvm_unmap_remove(), so
  > that pmap_{k,}remove() is not called on the entry.
  > This is intended to save time, and behave better, on pmaps with MMU holes
  > at process exit time.
  > ok art@, kettenis@ provided feedback as well. (miod@)

rtadvd

  ~ rtadvd.8                              

  > some syslog Xr; (jmc@)

  ~ rtadvd.c                              ~ rtadvd.h

  > drop root privileges in rtadvd to _rtadvd
  > ok deraadt@, reyk@, pyr@ (rainer@)

rtsold

  ~ if.c                                  

  > also handle wireless interfaces (IFM_IEEE80211 in addition to IFM_ETHER)
  > From kame
  > ok rainer@ (reyk@)

sysmerge

  ~ sysmerge.8                            

  > - encourage people to use the same source as was used to upgrade system
  > binaries so that some etc files don't get forgotten in the process
  > discussed with kurt@
  > tweaks and ok jmc@ (ajacoutot@)

  ~ sysmerge.sh                           

  > - add a sanity check on tarballs name to ensure we're not feeding
  > sysmerge with unrelated sets
  > ok sthen@ pyr@ jdixon@ (ajacoutot@)

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


More information about the odc mailing list