[owc] Weekly src changes ending 2006-06-25

OWC auto at squish.net
Mon Jun 26 08:00:20 BST 2006


OpenBSD src changes summary for 2006-06-18 to 2006-06-25 inclusive
==================================================================

bin/expr                                distrib/i386
distrib/miniroot                        distrib/notes
distrib/sets                            etc/etc.sparc64/sysctl.conf
etc/rc.shutdown                         lib/libc
lib/libkvm                              libexec/getNAME
libexec/ld.so                           regress/sbin
sbin/bioctl                             sbin/ifconfig
sbin/init                               sbin/ipsecctl
sbin/isakmpd                            share/man
sys/arch/alpha/alpha                    sys/arch/alpha/conf
sys/arch/alpha/include                  sys/arch/amd64/amd64
sys/arch/amd64/conf                     sys/arch/arm/xscale
sys/arch/armish/conf                    sys/arch/aviion/dev
sys/arch/hp300/hp300                    sys/arch/hp300/include
sys/arch/hppa/include                   sys/arch/hppa64/include
sys/arch/i386/conf                      sys/arch/i386/i386
sys/arch/i386/isa                       sys/arch/i386/pci
sys/arch/luna88k/conf                   sys/arch/m68k/include
sys/arch/m68k/m68k                      sys/arch/m88k/m88k
sys/arch/mac68k/dev                     sys/arch/mac68k/include
sys/arch/mac68k/mac68k                  sys/arch/macppc/conf
sys/arch/macppc/dev                     sys/arch/macppc/macppc
sys/arch/macppc/pci                     sys/arch/mips64/include
sys/arch/mips64/mips64                  sys/arch/mvme68k/conf
sys/arch/mvme68k/dev                    sys/arch/mvme68k/include
sys/arch/mvme68k/mvme68k                sys/arch/mvme88k/dev
sys/arch/solbourne/conf                 sys/arch/sparc/dev
sys/arch/sparc/sparc                    sys/arch/sparc64/conf
sys/arch/sparc64/dev                    sys/arch/sparc64/fpu
sys/arch/sparc64/include                sys/arch/sparc64/sparc64
sys/arch/vax/conf                       sys/arch/zaurus/zaurus
sys/compat/common                       sys/compat/freebsd
sys/compat/netbsd                       sys/conf
sys/dev                                 sys/dev/cardbus
sys/dev/gpio                            sys/dev/i2c
sys/dev/ic                              sys/dev/isa
sys/dev/onewire                         sys/dev/pci
sys/dev/sbus                            sys/dev/sdmmc
sys/dev/usb                             sys/isofs/udf
sys/kern                                sys/net
sys/net80211                            sys/netinet
sys/netinet6                            sys/nfs
sys/scsi                                sys/sys
sys/ufs/ext2fs                          sys/ufs/ffs
sys/ufs/mfs                             sys/ufs/ufs
sys/uvm                                 sys/xfs
usr.bin/calendar                        usr.bin/cdio
usr.bin/cvs                             usr.bin/fstat
usr.bin/ftp                             usr.bin/sudo
usr.bin/vi                              usr.sbin/bgpd
usr.sbin/hostapd                        usr.sbin/ntpd
usr.sbin/syslogd                        usr.sbin/tcpdump

== bin =============================================================== 01/11 ==

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

expr

  ~ expr.c                                

  > handle -- as POSIX requires; from kili at outback.escape.dek otto
  > (deraadt@)

== distrib =========================================================== 02/11 ==

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

i386

  ~ ramdisk_cd/list.local                 

  > we can fit all four ral firmwares (deraadt@)

  ~ ramdiskC/list.local                   

  > no more wicontrol on the media (get used to it)
  > add more firmwares that some drivers need (deraadt@)

miniroot

  ~ upgrade.sh                            ~ install.sh

  > Add a set siteXY-<hostname>.tgz and make it selected by default. This
  > allows host specific files to be trivially installed and makes
  > creating/maintaining large numbers of host specific configurations
  > much simpler.
  > Suggested by work being done by Alex Holst, brought to my attention by
  > mk@ during c2k6.
  > Name format suggested by henning@, tweaked by deraadt at .rw@)

notes

  ~ sparc64/hardware                      

  > More supported models. (miod@)

  ~ amd64/hardware                        ~ cats/hardware
  ~ i386/hardware                         

  > Support for Promise PDC205xx based SATA controllers; adapted
  > from NetBSD, with additions of some more devices from
  > Aaron Linville and Henrik Flodell.
  > ok grange@ (jsg@)

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.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/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.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@)

== etc =============================================================== 03/11 ==

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

etc.sparc64/sysctl.conf

  ~ etc.sparc64/sysctl.conf               

  > add the machdep.kbdreset logic to sparc64 (stuff to use it is coming)
  > (jason@)

rc.shutdown

  ~ rc.shutdown                           

  > make powerdown the default; discussed with jason (deraadt@)

  ~ rc.shutdown                           

  > ok there is a good reason not to do this yet (deraadt@)

== lib =============================================================== 04/11 ==

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

libc

  ~ sys/adjfreq.2                         

  > Mention that only the superuser may adjust the frequency.  Like in
  > adjtime(2), but less wordy.
  > OK otto@, jmc at .ay@)

libkvm

  ~ kvm_vax.c                             

  > Real kvm routines for vax; allows ``target kvm'' to work in gdb for kernel
  > crash dumps.
  > ok kettenis@ (miod@)

  ~ kvm_proc.c                            

  > from netbsd: make anons dynamically allocated from pool.
  > this results in lesse kva waste due to static preallocation of those
  > for every phys page and also every swap page.
  > tested by beck krw miod (mickey@)

== libexec =========================================================== 05/11 ==

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

getNAME

  ~ getNAME.c                             

  > minimal KNF and spacing
  > ok millert@, otto@ (jasper@)

ld.so

  ~ ldconfig/ldconfig.c                   

  > sort usage(); (jmc@)

  ~ ldconfig/ldconfig.8                   

  > tweak the -S stuff; (jmc@)

  ~ ld.so.1                               

  > These two env vars don't have a `=' in their name.
  > ok jmc (mk@)

== regress =========================================================== 06/11 ==

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

sbin

  ~ ipsecctl/Makefile                     ~ ipsecctl/ike1.ok
  ~ ipsecctl/ike10.ok                     ~ ipsecctl/ike11.ok
  ~ ipsecctl/ike12.ok                     ~ ipsecctl/ike13.ok
  ~ ipsecctl/ike14.ok                     ~ ipsecctl/ike15.ok
  ~ ipsecctl/ike17.ok                     ~ ipsecctl/ike18.ok
  ~ ipsecctl/ike19.ok                     ~ ipsecctl/ike2.ok
  ~ ipsecctl/ike20.ok                     ~ ipsecctl/ike21.ok
  ~ ipsecctl/ike22.ok                     ~ ipsecctl/ike23.ok
  ~ ipsecctl/ike29.ok                     ~ ipsecctl/ike3.ok
  ~ ipsecctl/ike30.ok                     ~ ipsecctl/ike31.ok
  ~ ipsecctl/ike32.ok                     ~ ipsecctl/ike33.ok
  ~ ipsecctl/ike34.ok                     ~ ipsecctl/ike35.ok
  ~ ipsecctl/ike36.ok                     ~ ipsecctl/ike37.ok
  ~ ipsecctl/ike39.ok                     ~ ipsecctl/ike4.ok
  ~ ipsecctl/ike40.ok                     ~ ipsecctl/ike41.ok
  ~ ipsecctl/ike42.ok                     ~ ipsecctl/ike43.ok
  ~ ipsecctl/ike46.ok                     ~ ipsecctl/ike47.ok
  ~ ipsecctl/ike6.ok                      ~ ipsecctl/ike7.ok
  ~ ipsecctl/ike8.ok                      ~ ipsecctl/ike9.ok

  > adopt to recent changes (hshoexer@)

  ~ ipsecctl/Makefile                     

  > The ike/ikedel tests 48 to 50 do net exist yet.  They will be needed
  > for stuff that will soon be commited.  In the meanwhile disable
  > them.  Noticed by david@, thanks! (hshoexer@)

== sbin ============================================================== 07/11 ==

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

bioctl

  ~ Makefile                              

  > append to CFLAGS instead of overwriting
  > ok kettenis@ miod@ (david@)

ifconfig

  ~ ifconfig.8                            

  > oops! i put -M in the wrong place; (jmc@)

  ~ ifconfig.8                            

  > -M clarity, ok jmc@ (todd@)

  ~ ifconfig.c                            

  > allow ifconfig to print the signal quality of the current ap. if the
  > driver reports a RSSI Max value, print the signal quality as a
  > percentage instead of the arbitrary "dB" value, this also applies to
  > the output of ifconfig -M (scan/node list).
  > ok damien@ jsg@ (reyk@)

init

  ~ init.8                                ~ init.c

  > accept SIGUSR2.  Like SIGUSR1, except it makes RB_POWERDOWN mandatory.
  > This will be used by kernel components which are trying to really really
  > power the system down.  suggested and tested by jason (deraadt@)

  ~ init.8                                

  > let's just use "getty" as the plural of `getty', because "getty's" is
  > very painful, and "gettys" is misleading; (jmc@)

ipsecctl

  ~ ike.c                                 ~ ipsec.conf.5
  ~ ipsecctl.h                            ~ parse.y

  > add group "none"; when choosen, pfs will be disabled.
  > ok david msf (hshoexer@)

isakmpd

  ~ monitor.c                             

  > clean up some gotos.  Originally from Andrey Matveev <evol at online
  > dot ptt dot ru>.  Ok and help moritz@ (hshoexer@)

== share ============================================================= 08/11 ==

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

man

  ~ man5/pf.conf.5                        

  > typo: queu -> queue
  > ok claudio@ (hshoexer@)

  ~ man4/ugen.4                           

  > Some markup/content errors.  Found by Marc Winiger <mw at msys.ch ok jmc@
  > (mbalmer@)

  ~ man4/usb.4                            

  > Xr zyd (deraadt@)

  ~ man4/Makefile                         + man4/zyd.4

  > slapped together a prelim man page based on atu.4 (deraadt@)

  ~ man4/zyd.4                            

  > can do g too (deraadt@)

  ~ man4/zyd.4                            

  > sort the hardware list, and zap some .Tn whilst there; (jmc@)

  ~ man4/usb.4                            

  > sync the zyd line; (jmc@)

  ~ man4/man4.mac68k/ae.4                 

  > sort hardware section (martin@)

  ~ man4/mpi.4                            

  > document that we run in vmware.
  > info from brad@ and lots of discussion with jmc@ (dlg@)

  ~ man9/namei.9                          

  > namei() only returns locked vnodes if LOCKLEAF is set, be more clear
  > about this
  > ok pedro (sturm@)

  ~ man4/pciide.4                         

  > Support for Promise PDC205xx based SATA controllers; adapted
  > from NetBSD, with additions of some more devices from
  > Aaron Linville and Henrik Flodell.
  > ok grange@ (jsg@)

  ~ man9/Makefile                         ~ man9/ieee80211_ioctl.9

  > ieee80211_cfgget() and ieee80211_cfgset() are dead; ok jsg (deraadt@)

  ~ man4/gre.4                            

  > add a note for MTU adjustment.
  > From NetBSD (brad@)

== sys =============================================================== 09/11 ==

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

arch/alpha/alpha

  ~ trap.c                                

  > Reset pcb_onfault around uvm_fault() calls. (miod@)

  ~ machdep.c                             

  > fix memory reporting on alphas with 2GB or more; ok miod (deraadt@)

arch/alpha/conf

  ~ files.alpha                           

  > move clock_subr.c to a better place, and now it is always in the kernel
  > so that things can use it; tested on all architectures; ok kettenis
  > (deraadt@)

  ~ GENERIC                               

  > mpi works here too.
  > ok deraadt@ (dlg@)

arch/alpha/include

  ~ disklabel.h                           

  > Zap Amiga bits of disklabel.h now that Amiga bits of readdisklabel() are
  > gone.
  > 'go ahead' deraadt@ (krw@)

arch/amd64/amd64

  ~ vm_machdep.c                          

  > consistantly count context switches on exit; miod@ ok (mickey@)

arch/amd64/conf

  ~ files.amd64                           

  > move clock_subr.c to a better place, and now it is always in the kernel
  > so that things can use it; tested on all architectures; ok kettenis
  > (deraadt@)

arch/arm/xscale

  ~ i80321_intr.c                         

  > Close a race in the interrupt handler code.
  > Inline functions when it makes sense. (drahn@)

arch/armish/conf

  ~ files.armish                          

  > move clock_subr.c to a better place, and now it is always in the kernel
  > so that things can use it; tested on all architectures; ok kettenis
  > (deraadt@)

arch/aviion/dev

  ~ nvram.c                               

  > Since a part of the nvram may be write-protected on some machines, check
  > the nvram writes and return EROFS if one or more bytes were found to be
  > immutable. (miod@)

arch/hp300/hp300

  ~ vm_machdep.c                          

  > consistantly count context switches on exit; miod@ ok (mickey@)

  ~ autoconf.c                            ~ bus_space.c
  ~ locore.s                              ~ vm_machdep.c

  > Use pmap_kenter_cache() for device memory mapping, instead of physacc(0
  > which bites the dust. (miod@)

arch/hp300/include

  ~ cpu.h                                 

  > Use pmap_kenter_cache() for device memory mapping, instead of physacc(0
  > which bites the dust. (miod@)

arch/hppa/include

  ~ disklabel.h                           

  > Zap Amiga bits of disklabel.h now that Amiga bits of readdisklabel() are
  > gone.
  > 'go ahead' deraadt@ (krw@)

arch/hppa64/include

  ~ disklabel.h                           

  > Zap Amiga bits of disklabel.h now that Amiga bits of readdisklabel() are
  > gone.
  > 'go ahead' deraadt@ (krw@)

arch/i386/conf

  ~ RAMDISKC                              

  > pcn no longer fits, sigh (deraadt@)

  ~ files.i386                            

  > move clock_subr.c to a better place, and now it is always in the kernel
  > so that things can use it; tested on all architectures; ok kettenis
  > (deraadt@)

  ~ RAMDISKC                              

  > we choose more drivers/firmwares over ext2fs (deraadt@)

arch/i386/i386

  ~ vm_machdep.c                          

  > consistantly count context switches on exit; miod@ ok (mickey@)

arch/i386/isa

  ~ clock.c                               

  > move clock_subr.c to a better place, and now it is always in the kernel
  > so that things can use it; tested on all architectures; ok kettenis
  > (deraadt@)

  ~ pccom.c                               

  > In a drivers activate() entry point, if on DVACT_DEACTIVATE it does
  > something, then it MUST return zero on DVACT_ACTIVATE, not EOPNOTSUPP;
  > this very popular bug has been cut and pasted a lot of times...
  > ok deraadt@ mickey@ (miod@)

arch/i386/pci

  ~ pci_machdep.c                         

  > reorganize the code to be more sane.
  > ok toby@ (brad@)

arch/luna88k/conf

  ~ files.luna88k                         

  > move clock_subr.c to a better place, and now it is always in the kernel
  > so that things can use it; tested on all architectures; ok kettenis
  > (deraadt@)

arch/m68k/include

  ~ pmap_motorola.h                       

  > Add pmap_enter_cache(), similar to pmap_kenter_cache() but for managed
  > pages, and implement pmap_enter() as a particular case of it. (miod@)

arch/m68k/m68k

  ~ pmap_motorola.c                       

  > Fix PMAP_DEBUG-only tests I botched in revision 1.13. (miod@)

  ~ pmap_motorola.c                       

  > Add pmap_enter_cache(), similar to pmap_kenter_cache() but for managed
  > pages, and implement pmap_enter() as a particular case of it. (miod@)

arch/m88k/m88k

  ~ vm_machdep.c                          

  > consistantly count context switches on exit; miod@ ok (mickey@)

arch/mac68k/dev

  ~ if_mc_obio.c                          ~ if_mcvar.h
  ~ if_sn.c                               ~ if_snvar.h

  > Allocate ethernet buffers as uvm pages, and map them with
  > pmap_enter_cache(),
  > instead of crossing fingers and expecting malloc() to return aligned and
  > contiguous memory. (miod@)

arch/mac68k/include

  ~ cpu.h                                 

  > Use pmap_enter_cache() instead of physacc() in bus_mem_add_mapping(), and
  > let
  > physacc() die. As a bonus, kvtop() dies too. (miod@)

arch/mac68k/mac68k

  ~ vm_machdep.c                          

  > consistantly count context switches on exit; miod@ ok (mickey@)

  ~ bus_space.c                           ~ locore.s
  ~ vm_machdep.c                          

  > Use pmap_enter_cache() instead of physacc() in bus_mem_add_mapping(), and
  > let
  > physacc() die. As a bonus, kvtop() dies too. (miod@)

arch/macppc/conf

  ~ files.macppc                          

  > move clock_subr.c to a better place, and now it is always in the kernel
  > so that things can use it; tested on all architectures; ok kettenis
  > (deraadt@)

  ~ GENERIC                               

  > this gpio should not be enabled (deraadt@)

  ~ GENERIC                               

  > do not enable the gpio parent either (deraadt@)

  ~ GENERIC                               ~ RAMDISK

  > Enable sk(4).
  > ok reyk@ (brad@)

arch/macppc/dev

  ~ uni_n.c                               

  > Remove redundant test in uni_n_config() as well as an unused softc member;
  > ok kettenis@ (miod@)

  ~ if_wi_obio.c                          ~ ki2c.c
  ~ macgpio.c                             ~ mediabay.c
  ~ mesh.c                                ~ openpic.c
  ~ wdc_obio.c                            ~ zs.c

  > Be more strict in the probe routines and actually check ca_nintr and
  > ca_nreg are large enough for our greedy needs; tweaks and ok kettenis@
  > (miod@)

arch/macppc/macppc

  ~ clock.c                               

  > move clock_subr.c to a better place, and now it is always in the kernel
  > so that things can use it; tested on all architectures; ok kettenis
  > (deraadt@)

arch/macppc/pci

  ~ macobio.c                             

  > Be more strict in the probe routines and actually check ca_nintr and
  > ca_nreg are large enough for our greedy needs; tweaks and ok kettenis@
  > (miod@)

arch/mips64/include

  ~ disklabel.h                           

  > Zap Amiga bits of disklabel.h now that Amiga bits of readdisklabel() are
  > gone.
  > 'go ahead' deraadt@ (krw@)

  ~ pmap.h                                

  > Define an empty pmap_collect(). (miod@)

arch/mips64/mips64

  ~ clock.c                               

  > move clock_subr.c to a better place, and now it is always in the kernel
  > so that things can use it; tested on all architectures; ok kettenis
  > (deraadt@)

  ~ busdma.c                              

  > Allocate bus_dma memory from kernel_map, not kmem_map. (miod@)

arch/mvme68k/conf

  ~ GENERIC                               ~ MVME147
  ~ MVME162                               ~ MVME167
  ~ MVME177                               ~ RAMDISK

  > Configure network devices at ipl 3, not ipl 1. (miod@)

arch/mvme68k/dev

  ~ nvram.c                               

  > let this compile after time.h changes (deraadt@)

arch/mvme68k/include

  ~ cpu.h                                 

  > Replace physaccess() usage with pmap_kenter_cache(). (miod@)

arch/mvme68k/mvme68k

  ~ vm_machdep.c                          

  > consistantly count context switches on exit; miod@ ok (mickey@)

  ~ autoconf.c                            ~ vm_machdep.c

  > Replace physaccess() usage with pmap_kenter_cache(). (miod@)

arch/mvme88k/dev

  ~ nvram.c                               

  > move clock_subr.c to a better place, and now it is always in the kernel
  > so that things can use it; tested on all architectures; ok kettenis
  > (deraadt@)

arch/solbourne/conf

  ~ files.solbourne                       

  > move clock_subr.c to a better place, and now it is always in the kernel
  > so that things can use it; tested on all architectures; ok kettenis
  > (deraadt@)

arch/sparc/dev

  ~ hme.c                                 

  > fix a typo in a comment and add the comment to the sparc hme driver.
  > (brad@)

arch/sparc/sparc

  ~ clock.c                               

  > move clock_subr.c to a better place, and now it is always in the kernel
  > so that things can use it; tested on all architectures; ok kettenis
  > (deraadt@)

  ~ clock.c                               ~ clockreg.h

  > Use MI clock conversion code for "new" clock.
  > ok deraadt@, miod@ (kettenis@)

arch/sparc64/conf

  ~ GENERIC                               

  > enable udcf(4) on sparc64, works on my Blade 100; ok mbalmer@ (naddy@)

  ~ files.sparc64                         

  > move clock_subr.c to a better place, and now it is always in the kernel
  > so that things can use it; tested on all architectures; ok kettenis
  > (deraadt@)

  ~ GENERIC                               ~ RAMDISK
  ~ RAMDISKU1                             ~ RAMDISKU5

  > Don't need option SUN4U.  It's not -really- an option on sparc64, duh.
  > (jason@)

  ~ GENERIC                               

  > if machdep.kbdreset==1 and the front panel button interrupt fires, tell
  > init
  > that it's time to go sleepy-bye. (jason@)

  ~ GENERIC                               ~ RAMDISK

  > Enable sk(4).
  > ok reyk@ (brad@)

arch/sparc64/dev

  ~ power.c                               

  > if machdep.kbdreset==1 and the front panel button interrupt fires, tell
  > init
  > that it's time to go sleepy-bye. (jason@)

  ~ pcfiic_ebus.c                         

  > BBC-based machines swap the two registers (good going Sun -- that is why we
  > don't get documentation, right?), and the clock fed into the chip can vary
  > from model to model; help & testing from dlg (deraadt@)

  ~ schizo.c                              

  > match tomatillo varient (not yet fully working) (deraadt@)

arch/sparc64/fpu

  ~ fpu.c                                 ~ fpu_emu.h
  ~ fpu_explode.c                         ~ fpu_extern.h
  ~ fpu_implode.c                         

  > There was hope, at one time, of sharing fpu/ with sparc.  That's still
  > possible, but not with #ifdef SUN4U all over the place.  The define should
  > be _v9 or __sparc64__ since it will equally well apply to all v9 machines.
  > We'll burn that bridge before we cross it. (jason@)

arch/sparc64/include

  ~ frame.h                               

  > grr... missed this... nuke SUN4U stuff here, too (jason@)

  ~ cpu.h                                 

  > add the machdep.kbdreset logic to sparc64 (stuff to use it is coming)
  > (jason@)

arch/sparc64/sparc64

  ~ pmap.c                                

  > Fix tsbsize computation; from NetBSD (miod@)

  ~ clock.c                               

  > Do not print hostid while booting, there's a sysctl for this. (miod@)

  ~ machdep.c                             

  > add the machdep.kbdreset logic to sparc64 (stuff to use it is coming)
  > (jason@)

  ~ clock.c                               

  > Only attach timer to the first counter-timer found. (miod@)

arch/vax/conf

  ~ files.vax                             

  > move clock_subr.c to a better place, and now it is always in the kernel
  > so that things can use it; tested on all architectures; ok kettenis
  > (deraadt@)

arch/zaurus/zaurus

  ~ zaurus_machdep.c                      

  > revert the interrupt vector diff, this fixes zaurus suspend; ok drahn@
  > (todd@)

compat/common

  ~ vfs_syscalls_25.c                     

  > rename vfs_busy() flags VB_UMIGNORE/VB_UMWAIT to VB_NOWAIT/VB_WAIT
  > requested by and ok pedro (sturm@)

compat/freebsd

  ~ freebsd_file.c                        

  > rename vfs_busy() flags VB_UMIGNORE/VB_UMWAIT to VB_NOWAIT/VB_WAIT
  > requested by and ok pedro (sturm@)

compat/netbsd

  ~ netbsd_file.c                         

  > rename vfs_busy() flags VB_UMIGNORE/VB_UMWAIT to VB_NOWAIT/VB_WAIT
  > requested by and ok pedro (sturm@)

conf

  ~ files                                 

  > move clock_subr.c to a better place, and now it is always in the kernel
  > so that things can use it; tested on all architectures; ok kettenis
  > (deraadt@)

dev

  - clock_subr.c                          

  > and this goes away (deraadt@)

  ~ clock_subr.h                          

  > move clock_subr.c to a better place, and now it is always in the kernel
  > so that things can use it; tested on all architectures; ok kettenis
  > (deraadt@)

  ~ audio.c                               ~ radio.c

  > In a drivers activate() entry point, if on DVACT_DEACTIVATE it does
  > something, then it MUST return zero on DVACT_ACTIVATE, not EOPNOTSUPP;
  > this very popular bug has been cut and pasted a lot of times...
  > ok deraadt@ mickey@ (miod@)

dev/cardbus

  ~ cardbus.c                             ~ cardbus_map.c
  ~ cardbusvar.h                          ~ ehci_cardbus.c
  ~ if_ath_cardbus.c                      ~ if_atw_cardbus.c
  ~ if_ral_cardbus.c                      ~ if_re_cardbus.c
  ~ if_rl_cardbus.c                       ~ if_rtw_cardbus.c
  ~ if_xl_cardbus.c                       ~ ohci_cardbus.c
  ~ rbus.h                                ~ uhci_cardbus.c

  > removal of '#ifdef rbus ... #else ...' directives, we use rbus in all
  > cases anyways, so we might as well remove code (the !rbus case) thats been
  > dead throughout the last years and just use the rbus code without ifdefs.
  > brad, miod "I am all for this diff"
  > ok deraadt, mickey (fkr@)

  ~ if_re_cardbus.c                       

  > de-register, ANSI functions and a little cleaning. (brad@)

dev/gpio

  ~ gpioow.c                              

  > In a drivers activate() entry point, if on DVACT_DEACTIVATE it does
  > something, then it MUST return zero on DVACT_ACTIVATE, not EOPNOTSUPP;
  > this very popular bug has been cut and pasted a lot of times...
  > ok deraadt@ mickey@ (miod@)

dev/i2c

  ~ w83l784r.c                            

  > Avoid using the rfact member of 'struct sensor' which is on its way out.
  > (kettenis@)

  ~ rs5c372.c                             

  > pretty code, shorten it, etc, ok drahn (deraadt@)

dev/ic

  ~ mpireg.h                              ~ mpi.c

  > Print World Wide Node Name and World Wide Port Name during dmesg so that
  > we can actually find the drives on the fabric.  Requested by kettenis krw
  > and brad.
  > ok dlg (marco@)

  ~ mpi.c                                 

  > Don't walk memory whenever there is nothing there.	I ran into this while
  > debugging FC stuff. (marco@)

  ~ rt2661.c                              ~ rt2560.c

  > enable packet bursting when operating as a STA.
  > limit bursts to 8 frames. (damien@)

  ~ rt2560.c                              ~ rt2560var.h
  ~ rt2661.c                              ~ rt2661reg.h
  ~ rt2661var.h                           

  > - implement new ic_updateslot() callback.
  > - in hostap mode, we defer update of the slot time until all associated
  > STAs are notified with updated beacons. (damien@)

  ~ mpi.c                                 

  > Make mpi not spit out WWNN and WWPN as requested by deraadt and dlg.  Do
  > store these values in the scsi_link structure for each device.
  > ok dlg. (marco@)

  ~ lm78.c                                

  > Avoid using the rfact member of 'struct sensor' which is on its way out.
  > (kettenis@)

  ~ mfi.c                                 ~ mfireg.h
  ~ mfivar.h                              

  > Add hot spare support. (marco@)

  ~ mfi.c                                 

  > Make a BGI (Back Ground Init) a scrub instead of rebuild. (marco@)

  ~ mfi.c                                 ~ mpi.c

  > Everytime one forgets an argument in a printf-like function call, God kills
  > a kitten. Commiters, please think of the kittens when working on code.
  > (miod@)

  ~ mfi.c                                 

  > Factor for disk size is /2.  Odd but tested on all mfis I have. (marco@)

  ~ pcf8584.c                             

  > properly handle timeout failure in while loop (deraadt@)

  ~ pcf8584.c                             ~ pcf8584var.h

  > BBC-based machines swap the two registers (good going Sun -- that is why we
  > don't get documentation, right?), and the clock fed into the chip can vary
  > from model to model; help & testing from dlg (deraadt@)

  ~ ath.c                                 ~ atw.c
  ~ awi.c                                 ~ com.c
  ~ smc91cxx.c                            

  > In a drivers activate() entry point, if on DVACT_DEACTIVATE it does
  > something, then it MUST return zero on DVACT_ACTIVATE, not EOPNOTSUPP;
  > this very popular bug has been cut and pasted a lot of times...
  > ok deraadt@ mickey@ (miod@)

  ~ ar5xxx.h                              ~ ath.c
  ~ athvar.h                              

  > set the RSSI Max value in ath(4) and use the new RSSI radiotap header
  > instead of the old db signal header. also allow tcpdump and hostapd to
  > print the new RSSI radiotap header values current/max rssi.
  > ok damien@ jsg@ (reyk@)

  ~ re.c                                  

  > some comment fixes.
  > From FreeBSD (brad@)

  ~ re.c                                  

  > de-register, ANSI functions and a little cleaning. (brad@)

  ~ an.c                                  

  > put the 1.32 change back; tested by many (mickey@)

  ~ hme.c                                 

  > fix a typo in a comment and add the comment to the sparc hme driver.
  > (brad@)

dev/isa

  ~ aps.c                                 

  > Do not bother initializing the rfact member of `struct sensor'; it is on
  > its
  > way out anyway. (kettenis@)

  ~ it.c                                  

  > Add abstraction for resistor factors and avoid using the rfact member of
  > 'struct sensor' which is on its way out.
  > tested by & ok jsg@ (kettenis@)

dev/onewire

  ~ owid.c                                ~ owtemp.c

  > In a drivers activate() entry point, if on DVACT_DEACTIVATE it does
  > something, then it MUST return zero on DVACT_ACTIVATE, not EOPNOTSUPP;
  > this very popular bug has been cut and pasted a lot of times...
  > ok deraadt@ mickey@ (miod@)

dev/pci

  ~ pcidevs                               

  > add some additional LSI Logic PCI ids.
  > From the LSI Linux MPT driver. (brad@)

  ~ pcidevs.h                             ~ pcidevs_data.h

  > regen (brad@)

  ~ pcidevs                               

  > add some more Broadcom Gig PCI ids.
  > From the Linux tg3 driver. (brad@)

  ~ pcidevs.h                             ~ pcidevs_data.h

  > regen (brad@)

  ~ if_bgereg.h                           

  > some new ASIC revisions. (brad@)

  ~ if_bge.c                              

  > add new PCI ids. (brad@)

  ~ if_bge.c                              

  > add the new ASIC revs to the bge_majorrevs table. (brad@)

  ~ wdt.c                                 

  > Remove freebsdism.
  > input and ok mickey. (mk@)

  ~ viaenv.c                              

  > Do not bother initializing the rfact member of `struct sensor'; it is on
  > its
  > way out anyway. (kettenis@)

  ~ if_bge.c                              

  > add the BCM5755 and BCM5787 ASICs to the appropriate macros. (brad@)

  ~ pcidevs.h                             ~ pcidevs_data.h

  > regen (jsg@)

  ~ pcidevs                               

  > Promise PDC20775, from Aaron Linville. (jsg@)

  ~ if_bge.c                              

  > - check for revision B2 intead of B1 in one more spot.
  > - fix no link check in bge_start(). (brad@)

  ~ if_bge.c                              

  > have bge_start() check for an empty send queue instead. (brad@)

  ~ azalia.c                              

  > make sure to return and indicate an error when an invalid PCM format
  > is found. (brad@)

  ~ if_ixgb.c                             ~ if_xge.c

  > use the 10Gb SR media type for now. (brad@)

  ~ pccbb.c                               ~ pccbbvar.h

  > removal of '#ifdef rbus ... #else ...' directives, we use rbus in all
  > cases anyways, so we might as well remove code (the !rbus case) thats been
  > dead throughout the last years and just use the rbus code without ifdefs.
  > brad, miod "I am all for this diff"
  > ok deraadt, mickey (fkr@)

  ~ pcidevs                               ~ if_bge.c

  > add the BCM5786 PCI id.
  > From the Linux tg3 driver. (brad@)

  ~ pcidevs.h                             ~ pcidevs_data.h

  > regen (brad@)

  ~ if_bge.c                              

  > remove some unused code.
  > ok reyk@ (brad@)

  ~ if_bge.c                              ~ if_bgereg.h

  > recognize the BCM5703 B0. (brad@)

  ~ ixgb_ids.h                            ~ ixgb_hw.c

  > identify CX4 based hardware.
  > From Intel's FreeBSD driver via their web-site. (brad@)

  ~ if_ixgb.c                             

  > better media handling. (brad@)

  ~ pcidevs                               

  > The DGE560T_2 PCI id was a guess and more of a place holder as I did not
  > know what hardware this really was at the time. It looks as if this is
  > really a DGE530 B1. (brad@)

  ~ if_sk.c                               

  > adjust PCI ids. (brad@)

  ~ pcidevs.h                             ~ pcidevs_data.h

  > regen (brad@)

  ~ if_skreg.h                            

  > lower the RX ring count and Jumbo slots on sparc64 as a workaround.
  > ok reyk@ (brad@)

  ~ if_sk.c                               

  > Add appropriate swapping for big-endian systems. Allows sk(4) to work
  > on macppc and sparc64.
  > ok reyk@ (brad@)

  ~ if_skreg.h                            

  > bump up the RX ring count on sparc64 a little bit. (brad@)

  ~ azalia.c                              

  > In a drivers activate() entry point, if on DVACT_DEACTIVATE it does
  > something, then it MUST return zero on DVACT_ACTIVATE, not EOPNOTSUPP;
  > this very popular bug has been cut and pasted a lot of times...
  > ok deraadt@ mickey@ (miod@)

  ~ pcidevs                               

  > add a few more Broadcom BCM43xx wireless PCI ids. (brad@)

  ~ pcidevs.h                             ~ pcidevs_data.h

  > regen (brad@)

  ~ pcidevs                               

  > add the Promise PDC40518 SATA-II chipset PCI id.
  > From Henrik Flodell (brad@)

  ~ pcidevs.h                             ~ pcidevs_data.h

  > regen (brad@)

  ~ azalia.h                              

  > add some constant macros for unsolicited response and jack sense.
  > From kent NetBSD (brad@)

  ~ if_em.c                               

  > make em_fixup_rx() a void function. (brad@)

  ~ if_re_pci.c                           

  > de-register, ANSI functions and a little cleaning. (brad@)

  ~ pciide.c                              ~ pciide_pdc202xx_reg.h

  > Support for Promise PDC205xx based SATA controllers; adapted
  > from NetBSD, with additions of some more devices from
  > Aaron Linville and Henrik Flodell.
  > ok grange@ (jsg@)

  ~ pciide.c                              

  > Somehow managed to miss including the part of Henrik's diff where
  > we actually match the PCI ID of the 40518 controller.
  > Pointed out by Henrik Flodell. (jsg@)

  ~ pcidevs                               

  > Correct a typo in one of the CN700 host bridge product ids. (jsg@)

  ~ pcidevs.h                             ~ pcidevs_data.h

  > regen (jsg@)

  ~ azalia.c                              

  > If the encoding of an audio converter is 0, refer to
  > the audio function node.  CMI9880 0.2 has such ADCs.
  > From kent NetBSD (brad@)

  ~ azalia_codec.c                        

  > DAC&ADC group and custom mixer table for the CMedia CMI9880.
  > From kent NetBSD (brad@)

  ~ pcidevs                               

  > add Parallels PCI ids. (brad@)

  ~ pcidevs.h                             ~ pcidevs_data.h

  > regen (brad@)

  ~ pcidevs                               

  > fix Surecom NE-34 dmesg, found on misc@
  > ok brad@ (martin@)

  ~ pcidevs.h                             ~ pcidevs_data.h

  > regen (martin@)

dev/sbus

  ~ qe.c                                  

  > don't need #ifdef SUN4U any more (jason@)

  ~ sbusvar.h                             

  > use __sparc__ and __sparc64__ instead of SUN4|SUN4C|SUN4M vs SUN4U to
  > figure
  > out which includes to use. (jason@)

  ~ esp_sbus.c                            

  > Pick the sbus clock frequency from the attach args instead of peeking at
  > the
  > parent softc. (miod@)

dev/sdmmc

  ~ sdmmc_io.c                            

  > Appease the printf gods here too (miod@)

dev/usb

  ~ udcf.c                                

  > move clock_subr.c to a better place, and now it is always in the kernel
  > so that things can use it; tested on all architectures; ok kettenis
  > (deraadt@)

  ~ if_atu.c                              

  > Don't sleep and then say we're going to sleep -- do it the other way
  > around because the other thing is just silly and confusing.
  > ok dlg (mk@)

  ~ udcf.c                                

  > remove the no longer needed time calculations now that we have them in
  > sys/kern/clock_subr.c (mbalmer@)

  ~ usbdevs                               

  > Aceeca MEZ1000, from Mike Durian in FreeBSD PR 82839 (jsg@)

  ~ usbdevs.h                             ~ usbdevs_data.h

  > regen (jsg@)

  ~ uvisor.c                              

  > match Aceeca MEZ1000, from Mike Durian in FreeBSD PR 82839 (jsg@)

  ~ usbdevs                               

  > Add a few RT73 devices that are known to exist in the wild. (jsg@)

  ~ usbdevs.h                             ~ usbdevs_data.h

  > regen (jsg@)

  ~ if_rum.c                              

  > Add a few RT73 devices that are known to exist in the wild. (jsg@)

  ~ files.usb                             + if_zyd.c
  + if_zydreg.h                           

  > a good start at a driver for the Zydas ZD1211 802.11 devices by
  > ich at florian-stoehr.de.pparently it can receive now.  Contact him
  > if you want to help, or need the firmware image. (deraadt@)

  ~ if_zydreg.h                           

  > minimal cleaning (deraadt@)

  ~ if_zyd.c                              

  > minimal knf and cleanup (deraadt@)

  ~ ehci.c                                ~ if_atu.c
  ~ if_aue.c                              ~ if_axe.c
  ~ if_cdce.c                             ~ if_cue.c
  ~ if_kue.c                              ~ if_ral.c
  ~ if_udav.c                             ~ if_upl.c
  ~ if_url.c                              ~ if_wi_usb.c
  ~ ohci.c                                ~ uaudio.c
  ~ ubsa.c                                ~ ucom.c
  ~ udcf.c                                ~ udsbr.c
  ~ ueagle.c                              ~ uftdi.c
  ~ ugen.c                                ~ uhci.c
  ~ uhid.c                                ~ uhidev.c
  ~ uhub.c                                ~ uipaq.c
  ~ ukbd.c                                ~ ulpt.c
  ~ umass.c                               ~ umct.c
  ~ umidi.c                               ~ umodem.c
  ~ ums.c                                 ~ umsm.c
  ~ uplcom.c                              ~ urio.c
  ~ usb.c                                 ~ uscanner.c
  ~ uslcom.c                              ~ usscanner.c
  ~ uvisor.c                              ~ uvscom.c

  > In a drivers activate() entry point, if on DVACT_DEACTIVATE it does
  > something, then it MUST return zero on DVACT_ACTIVATE, not EOPNOTSUPP;
  > this very popular bug has been cut and pasted a lot of times...
  > ok deraadt@ mickey@ (miod@)

  ~ usb.h                                 

  > A typo a day, keeps the dictionnary away. (miod@)

isofs/udf

  ~ ecma167-udf.h                         ~ udf_vfsops.c

  > Make udf_find_partmaps() correctly spot virtual partition maps in
  > medias that have it, testing jmc@ and myself. (pedro@)

  ~ ecma167-udf.h                         ~ udf_vfsops.c

  > Roll in some defines for the file types used in udf_vget() and add one
  > for the VAT, no binary change (pedro@)

  ~ udf_vfsops.c                          

  > fix two nested loops sharing the same control variable, plug two memory
  > leaks and be consistent about error values returned to the caller (pedro@)

  ~ udf.h                                 ~ udf_vfsops.c
  ~ udf_vnops.c                           

  > rearrange some comments (pedro@)

  ~ udf_vfsops.c                          

  > replace unneeded bzero() (pedro@)

  ~ udf.h                                 

  > Shave 8 bytes off struct udf_mnt (pedro@)

kern

  + clock_subr.c                          

  > move clock_subr.c to a better place, and now it is always in the kernel
  > so that things can use it; tested on all architectures; ok kettenis
  > (deraadt@)

  ~ tty_nmea.c                            

  > Remove the no longer needed date/time calculations. (mbalmer@)

  ~ tty_nmea.c                            

  > pruning; ok mbalmer (deraadt@)

  ~ tty_nmea.c                            

  > If tty timestamping is enabled, copy the tty timestamp early to preserve
  > the value if it gets overwritten by other events. (mbalmer@)

  ~ kern_exit.c                           

  > consistantly count context switches on exit; miod@ ok (mickey@)

  ~ vfs_lookup.c                          

  > namei() only returns locked vnodes if LOCKLEAF is set, be more clear
  > about this
  > ok pedro (sturm@)

  ~ vfs_lookup.c                          ~ vfs_subr.c
  ~ vfs_sync.c                            ~ vfs_syscalls.c

  > rename vfs_busy() flags VB_UMIGNORE/VB_UMWAIT to VB_NOWAIT/VB_WAIT
  > requested by and ok pedro (sturm@)

net

  ~ radix.c                               ~ radix_mpath.c
  ~ radix_mpath.h                         ~ route.h

  > Add support for equal-cost multipath IP.
  > To minimise path disruptions, this implements recommendations made in
  > RFC2992 -
  > the hash-threshold mechanism to select paths based on source/destination IP
  > address pairs, and inserts multipath routes in the middle of the route
  > table.
  > To enable multipath distribution, use:
  > sysctl net.inet.ip.multipath=1
  > and/or:
  > sysctl net.inet6.ip6.multipath=1
  > testing norby@
  > ok claudio@ henning@ hshoexer@ (pascoe@)

  ~ radix_mpath.c                         

  > Whitespace, oops. (pascoe@)

  ~ route.c                               

  > Unbreak the tree. The code to set and unset the RTF_MPATH bit on all
  > multipath
  > routes did not carefully check if the route lookup succeded or not and so
  > rn_mpath_next(rn) blowed up because rn was NULL. Check if rnh_lookup
  > succeded
  > before touching rn in anyway. OK norby@ initial diff by hshoexer@
  > (claudio@)

  ~ if_media.h                            

  > add media types for 10Gb Ethernet, though only using a baudrate
  > of 1Gbps, until the size of the baudrate field has been increased. (brad@)

net80211

  ~ ieee80211.c                           ~ ieee80211_crypto.c
  ~ ieee80211_input.c                     ~ ieee80211_node.c
  ~ ieee80211_node.h                      ~ ieee80211_output.c
  ~ ieee80211_proto.c                     ~ ieee80211_proto.h
  ~ ieee80211_var.h                       

  > Improve 802.11b/g interoperability and move toward better compliance
  > with IEEE Std 802.11g-2003 standard:
  > - add ERP Information Element in probe responses and beacons
  > - keep track of the number of associated non-ERP STAs and non-short slot
  > time capable STAs in the BSS
  > - enable use of RTS/CTS or CTS-to-self protection when required by the BSS
  > - add a ic_updateslot() callback to notify drivers of slot time changes
  > - cleanup computation of mgmt frames sizes in ieee80211_output.c
  > - nuke unnecessary <sys/cdefs.h> includes
  > - remove an unused macro (LOGICALLY_EQUAL) while i'm here
  > From {free,net}bsd, with additional fixes.
  > ok brad@, reyk@ (damien@)

  ~ ieee80211_input.c                     

  > s/memcmp/IEEE80211_ADDR_EQ/ (damien@)

  ~ ieee80211_node.c                      

  > unbreak the code if compiled with IEEE80211_DEBUG by using the right
  > node attribute.
  > ok damien@, figured out by claudio@ (reyk@)

  ~ ieee80211_ioctl.c                     ~ ieee80211_ioctl.h
  ~ ieee80211_radiotap.h                  ~ ieee80211_var.h

  > add an optional max_rssi attribute to the ieee80211com structure and
  > allow to export the RSSI Max value with ioctls and by radiotap
  > headers.
  > ok damien@ jsg@
  > description:
  > we currently use "dB" as an indication for the signal strength in
  > ifconfig and in the radiotap headers. it means "decibel difference
  > from an arbitrary, fixed reference". this is quite confusing, because
  > different chipsets have different references for the dB/rssi values.
  > we can use the plain RSSI which is described in IEEE 802.11: "The
  > receive signal strength indicator (RSSI) is an optional parameter that
  > has a value of 0 through RSSI Max.". all wireless chipsets have
  > something like a RSSI (normally as a Rx descriptor field), but the
  > value for RSSI Max is chipset-specific.
  > if we know the RSSI Max, we can calculate a percentage which is much
  > easier to understand for the user. we even don't have to use the
  > absolute RSSI Max, we can use an average RSSI Max, figured out by
  > monitoring and tuning the RSSI Max of the drivers. if the user gets a
  > signal of 110%, it would mean "better than the average Max signal".
  > there's no need to do any RSSI calculations in the kernel, it just
  > passes the the relative rssi and max_rssi values to userspace. this is
  > done in the ieee80211_nodereq ioctl structure and possible with a new
  > radiotap header. the radiotap RSSI header allows to get a flexible but
  > common signal indicator instead of the complex and unrelated dB/dBm
  > signal fields. it must include two 8bit values current rssi and RSSI
  > max. (reyk@)

  ~ ieee80211_var.h                       

  > ieee80211_cfgget() and ieee80211_cfgset() are dead; ok jsg (deraadt@)

netinet

  ~ in.h                                  ~ in_pcb.c
  ~ ip_input.c                            ~ ip_output.c

  > Add support for equal-cost multipath IP.
  > To minimise path disruptions, this implements recommendations made in
  > RFC2992 -
  > the hash-threshold mechanism to select paths based on source/destination IP
  > address pairs, and inserts multipath routes in the middle of the route
  > table.
  > To enable multipath distribution, use:
  > sysctl net.inet.ip.multipath=1
  > and/or:
  > sysctl net.inet6.ip6.multipath=1
  > testing norby@
  > ok claudio@ henning@ hshoexer@ (pascoe@)

  ~ ip_input.c                            

  > Whitespace, oops. (pascoe@)

netinet6

  ~ frag6.c                               ~ in6.h
  ~ in6_proto.c                           ~ in6_src.c
  ~ ip6_forward.c                         ~ ip6_input.c
  ~ ip6_output.c                          ~ ip6_var.h

  > Add support for equal-cost multipath IP.
  > To minimise path disruptions, this implements recommendations made in
  > RFC2992 -
  > the hash-threshold mechanism to select paths based on source/destination IP
  > address pairs, and inserts multipath routes in the middle of the route
  > table.
  > To enable multipath distribution, use:
  > sysctl net.inet.ip.multipath=1
  > and/or:
  > sysctl net.inet6.ip6.multipath=1
  > testing norby@
  > ok claudio@ henning@ hshoexer@ (pascoe@)

nfs

  ~ nfs_syscalls.c                        

  > rename vfs_busy() flags VB_UMIGNORE/VB_UMWAIT to VB_NOWAIT/VB_WAIT
  > requested by and ok pedro (sturm@)

scsi

  ~ scsiconf.h                            

  > pack wwn fields into struct better; ok marco krw (deraadt@)

sys

  ~ time.h                                

  > move clock_subr.c to a better place, and now it is always in the kernel
  > so that things can use it; tested on all architectures; ok kettenis
  > (deraadt@)

  ~ param.h                               

  > permissable -> permissible.   ok jmc (steven@)

  ~ mount.h                               

  > rename vfs_busy() flags VB_UMIGNORE/VB_UMWAIT to VB_NOWAIT/VB_WAIT
  > requested by and ok pedro (sturm@)

ufs/ext2fs

  ~ ext2fs_inode.c                        

  > Under certain circumstances, ext2fs_inactive() can be called without a
  > backing dinode, in which case we just want to recycle the vnode.
  > Fixes a crash reported by reyk@, okay krw@, mickey@ and pat at .edro@)

ufs/ffs

  ~ ffs_softdep_stub.c                    

  > Shave 900 bytes off ramdisk kernels (before compression) by not including
  > the softdep stubs which get optimized out when DOINGSOFTDEP evaluates to
  > zero
  > at compile time and thus are never ever invoked; ok deraadt@ (miod@)

  ~ ffs_extern.h                          ~ ffs_softdep.c
  ~ ffs_softdep_stub.c                    ~ ffs_vnops.c

  > propagate the wait flag from fsync down to softdep_fsync_mountdev()
  > and do not perform synchronous sync there is no wait requested by
  > skipping the drain_output() call. this fixes a problem where
  > update kthread would sleep forever on some vnode since work is created
  > faster than it can be flushed. pedro@ ok; tested by many (mickey@)

ufs/mfs

  ~ mfs_vfsops.c                          

  > rename vfs_busy() flags VB_UMIGNORE/VB_UMWAIT to VB_NOWAIT/VB_WAIT
  > requested by and ok pedro (sturm@)

ufs/ufs

  ~ ufs_dirhash.c                         

  > do not wait in pool_get() here as we can recover from no memory; tedu@
  > pedro@ ok; tested by many (mickey@)

  ~ ufs_quota.c                           

  > rename vfs_busy() flags VB_UMIGNORE/VB_UMWAIT to VB_NOWAIT/VB_WAIT
  > requested by and ok pedro (sturm@)

uvm

  ~ uvm.h                                 ~ uvm_amap.c
  ~ uvm_amap.h                            ~ uvm_amap_i.h
  ~ uvm_anon.c                            ~ uvm_anon.h
  ~ uvm_fault.c                           ~ uvm_init.c
  ~ uvm_loan.c                            ~ uvm_map.c
  ~ uvm_mmap.c                            ~ uvm_page.c
  ~ uvm_pdaemon.c                         ~ uvm_swap.c

  > from netbsd: make anons dynamically allocated from pool.
  > this results in lesse kva waste due to static preallocation of those
  > for every phys page and also every swap page.
  > tested by beck krw miod (mickey@)

xfs

  ~ xfs_dev-common.c                      

  > rename vfs_busy() flags VB_UMIGNORE/VB_UMWAIT to VB_NOWAIT/VB_WAIT
  > requested by and ok pedro (sturm@)

== usr.bin =========================================================== 10/11 ==

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

calendar

  ~ calendars/calendar.birthday           ~ calendars/calendar.wissenschaft

  > fix zuse and turing (mickey@)

cdio

  ~ mmc.c                                 

  > retry if sense code indicates not ready. fixes writing for drives
  > that issue long write in progress if the write buffer is full.
  > ok deraadt@ (mjc@)

  ~ cdio.1                                

  > - tweak the -f description
  > - move DISC into an ENVIRONMENT section
  > - make the tao option list -compact
  > - uppercase `cddb' (jmc@)

cvs

  ~ add.c                                 ~ checkout.c
  ~ commit.c                              ~ cvs.h
  ~ diff.c                                ~ file.c
  ~ getlog.c                              ~ import.c
  ~ remove.c                              ~ status.c
  ~ tag.c                                 ~ update.c

  > kill local and remote callback and just have one called fileproc.
  > we let the commands pass the correct function.
  > all in preparation for remote. (joris@)

fstat

  ~ fstat.c                               

  > oops, forgot \n at end of kqueue lines; Nicholas Marriott (deraadt@)

ftp

  ~ complete.c                            ~ fetch.c
  ~ ftp.c                                 

  > free and sl_free already check against NULL, remove a few unneeded ifs.
  > ok otto (steven@)

sudo

  ~ sudoers.5                             

  > Typo.
  > From Alexander Hall (alexander at hakan dot beard dot se). (ray@)

vi

  ~ common/exf.c                          

  > Fix memleak; From Coverity Scan, CID 3135.
  > From simonb NetBSD
  > ok naddy@, otto@ (jasper@)

== usr.sbin ========================================================== 11/11 ==

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

bgpd

  ~ session.c                             

  > fix a comment (henning@)

  ~ session.c                             

  > make cloned sessions appear with RECONF_REINIT so they are treated like
  > peers added at runtime (henning@)

  ~ carp.c                                ~ bgpd.8

  > if at startup a given group has carpdemote == 0, we don't do carp demotion.
  > now, once all sessions went to established, we start doing demotion, so
  > that when one drops and has dmeotion enabled, we actually demote.
  > ok claudio (henning@)

  ~ bgpd.c                                ~ bgpd.8

  > add -c to usage() and synopsis; (jmc@)

  ~ bgpd.conf.5                           

  > a little .Xr; (jmc@)

hostapd

  ~ print-802_11.c                        

  > set the RSSI Max value in ath(4) and use the new RSSI radiotap header
  > instead of the old db signal header. also allow tcpdump and hostapd to
  > print the new RSSI radiotap header values current/max rssi.
  > ok damien@ jsg@ (reyk@)

ntpd

  ~ ntpd.h                                

  > tsk, tsk, tsk... the rule is simple: any define consisting of
  > more than one token MUST be put in parentheses! (otto@)

  ~ ntp.c                                 ~ ntpd.c

  > avoid a race by installing SIGCHLD handler before fork() is called.
  > ok henning@ ckuethe@ (otto@)

  ~ ntpd.c                                ~ ntpd.h

  > Save the computed clock drift and use it on startup. ok deraadt@
  > henning@ (otto@)

  ~ ntpd.8                                

  > Document /var/db/ntpd.drift; ok jmc@ (otto@)

syslogd

  ~ syslogd.8                             

  > CAVEATS
  > syslogd does not create files, it only logs to existing ones.
  > From millert@: http://marc.theaimsgroup.com/?m=106329875409953
  > ``Good idea'' millert@, ``fine by me too'' jmc@ (ray@)

tcpdump

  ~ print-802_11.c                        

  > set the RSSI Max value in ath(4) and use the new RSSI radiotap header
  > instead of the old db signal header. also allow tcpdump and hostapd to
  > print the new RSSI radiotap header values current/max rssi.
  > ok damien@ jsg@ (reyk@)

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



More information about the owc mailing list