OpenBSD src changes summary for 2016-01-31 to 2016-02-07 inclusive
==================================================================

bin/domainname                          bin/hostname
bin/mv                                  bin/rm
distrib/miniroot                        distrib/notes
distrib/sets                            distrib/special
etc/MAKEDEV.common                      etc/etc.amd64/MAKEDEV
etc/etc.amd64/MAKEDEV.md                etc/etc.i386/MAKEDEV
etc/etc.i386/MAKEDEV.md                 etc/examples/pkg.conf
etc/rc.d/bgpd                           etc/rc.d/dvmrpd
etc/rc.d/eigrpd                         etc/rc.d/httpd
etc/rc.d/ldpd                           etc/rc.d/npppd
etc/rc.d/ntpd                           etc/rc.d/ospf6d
etc/rc.d/ospfd                          etc/rc.d/relayd
etc/rc.d/ripd                           etc/rc.d/smtpd
etc/rc.d/snmpd                          etc/rc.d/vmd
games/snake                             include/signal.h
lib/libarch                             lib/libc
lib/libcrypto                           lib/libedit
lib/libpcap                             libexec/talkd
sbin/bioctl                             sbin/dhclient
sbin/init                               sbin/pdisk
share/man                               share/misc
sys/arch/amd64/amd64                    sys/arch/amd64/conf
sys/arch/amd64/stand/efiboot            sys/arch/arm/arm
sys/arch/arm/cortex                     sys/arch/arm/include
sys/arch/arm/sa11x0                     sys/arch/arm/xscale
sys/arch/armv7/armv7                    sys/arch/armv7/exynos
sys/arch/armv7/imx                      sys/arch/armv7/omap
sys/arch/armv7/sunxi                    sys/arch/armv7/vexpress
sys/arch/i386/i386                      sys/arch/mips64/include
sys/arch/mips64/mips64                  sys/conf
sys/dev                                 sys/dev/i2c
sys/dev/pci                             sys/dev/pv
sys/isofs/udf                           sys/kern
sys/msdosfs                             sys/net
sys/net80211                            sys/netinet
sys/ntfs                                sys/scsi
sys/sys                                 sys/tmpfs
sys/ufs/ext2fs                          sys/ufs/ufs
usr.bin/cu                              usr.bin/doas
usr.bin/ftp                             usr.bin/getent
usr.bin/less                            usr.bin/locale
usr.bin/ssh                             usr.bin/tail
usr.bin/talk                            usr.bin/tmux
usr.bin/units                           usr.bin/vi
usr.bin/write                           usr.bin/xinstall
usr.sbin                                usr.sbin/bgpd
usr.sbin/dhcpd                          usr.sbin/dhcrelay
usr.sbin/dvmrpd                         usr.sbin/eigrpd
usr.sbin/hostapd                        usr.sbin/httpd
usr.sbin/identd                         usr.sbin/ldapctl
usr.sbin/ldapd                          usr.sbin/ldpd
usr.sbin/npppd                          usr.sbin/ntpd
usr.sbin/ospf6d                         usr.sbin/ospfd
usr.sbin/pkg_add                        usr.sbin/relayd
usr.sbin/ripd                           usr.sbin/rtadvd
usr.sbin/smtpd                          usr.sbin/snmpd
usr.sbin/sysmerge                       usr.sbin/tcpdump
usr.sbin/vmd                            

== bin =============================================================== 01/12 ==

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

domainname

  ~ domainname.c                          

  > - make usage() static and __dead
  > - while here switch main to return instead of exit(3)
  > OK tedu@ (gsoares@)

hostname

  ~ hostname.c                            

  > - make usage() static and __dead
  > - while here switch main to return instead of exit(3)
  > OK tedu@ (gsoares@)

mv

  ~ Makefile                              

  > stop linking mv(1) to libutil, since that isn't used.
  > OK guenther@ nicm@ (gsoares@)

rm

  ~ rm.c                                  

  > style nit - do not add a space between function name and bracket
  > from Edgar Pettijohn <edgar at pettijohn-web.com>
  > fix another similar space in line 127. (pointed out by/OK jung@)
  > OK tb@ jung@ (gsoares@)

== distrib =========================================================== 02/12 ==

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

miniroot

  ~ install.sub                           

  > - remove _d from local list which is a leftover from a previous change
  > - rename a -> _src and localize it
  > OK krw@, halex@ (rpe@)

  ~ install.sh                            

  > spacing
  > OK krw@ (rpe@)

notes

  ~ packages                              

  > update package version in pkg_add example (sthen@)

sets

  ~ lists/man/mi                          

  > sync (visa@)

  ~ lists/base/md.amd64                   ~ lists/base/md.i386
  ~ lists/man/mi                          

  > sync (jsg@)

  ~ lists/base/md.alpha                   ~ lists/base/md.amd64
  ~ lists/base/md.armish                  ~ lists/base/md.armv7
  ~ lists/base/md.hppa                    ~ lists/base/md.hppa64
  ~ lists/base/md.i386                    ~ lists/base/md.landisk
  ~ lists/base/md.loongson                ~ lists/base/md.luna88k
  ~ lists/base/md.macppc                  ~ lists/base/md.octeon
  ~ 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                         

  > sync (sthen@)

special

  ~ mv/Makefile                           

  > sync to main
  > reminded by guenther@ (gsoares@)

== etc =============================================================== 03/12 ==

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

MAKEDEV.common

  ~ MAKEDEV.common                        

  > Add ipmi(4) cdev entry on amd64 and i386 (uebayasi@)

etc.amd64/MAKEDEV

  ~ etc.amd64/MAKEDEV                     

  > Regen (uebayasi@)

etc.amd64/MAKEDEV.md

  ~ etc.amd64/MAKEDEV.md                  

  > Add ipmi(4) cdev entry on amd64 and i386 (uebayasi@)

etc.i386/MAKEDEV

  ~ etc.i386/MAKEDEV                      

  > Regen (uebayasi@)

etc.i386/MAKEDEV.md

  ~ etc.i386/MAKEDEV.md                   

  > Add ipmi(4) cdev entry on amd64 and i386 (uebayasi@)

examples/pkg.conf

  ~ examples/pkg.conf                     

  > sync (sthen@)

  ~ examples/pkg.conf                     

  > sync (sthen@)

  ~ examples/pkg.conf                     

  > sync (sthen@)

rc.d/bgpd

  ~ rc.d/bgpd                             

  > Remove setproctitle() for the parent process. Because rc.d(8) uses process
  > titles (including flags) to distinguish between daemons, this makes it
  > possible to manage multiple copies of a daemon using the normal
  > infrastructure
  > by symlinking rc.d scripts to a new name. ok jung@ ajacoutot@,
  > missed in previous commit, problem reported by mxb/alumni/chalmers/se.
  > (sthen@)

rc.d/dvmrpd

  ~ rc.d/dvmrpd                           

  > Remove setproctitle() for the parent process. Because rc.d(8) uses process
  > titles (including flags) to distinguish between daemons, this makes it
  > possible to manage multiple copies of a daemon using the normal
  > infrastructure
  > by symlinking rc.d scripts to a new name. ok jung@ ajacoutot@,
  > missed in previous commit noticed after re-checking following report in
  > bgpd. (sthen@)

rc.d/eigrpd

  ~ rc.d/eigrpd                           

  > Remove setproctitle() for the parent process. Because rc.d(8) uses process
  > titles (including flags) to distinguish between daemons, this makes it
  > possible to manage multiple copies of a daemon using the normal
  > infrastructure
  > by symlinking rc.d scripts to a new name. ok jung@ ajacoutot@, smtpd ok
  > gilles@ (sthen@)

rc.d/httpd

  ~ rc.d/httpd                            

  > Remove setproctitle() for the parent process. Because rc.d(8) uses process
  > titles (including flags) to distinguish between daemons, this makes it
  > possible to manage multiple copies of a daemon using the normal
  > infrastructure
  > by symlinking rc.d scripts to a new name. ok jung@ ajacoutot@, smtpd ok
  > gilles@ (sthen@)

rc.d/ldpd

  ~ rc.d/ldpd                             

  > Remove setproctitle() for the parent process. Because rc.d(8) uses process
  > titles (including flags) to distinguish between daemons, this makes it
  > possible to manage multiple copies of a daemon using the normal
  > infrastructure
  > by symlinking rc.d scripts to a new name. ok jung@ ajacoutot@, smtpd ok
  > gilles@ (sthen@)

rc.d/npppd

  ~ rc.d/npppd                            

  > Remove setproctitle() for the parent process. Because rc.d(8) uses process
  > titles (including flags) to distinguish between daemons, this makes it
  > possible to manage multiple copies of a daemon using the normal
  > infrastructure
  > by symlinking rc.d scripts to a new name. ok jung@ ajacoutot@, smtpd ok
  > gilles@ (sthen@)

rc.d/ntpd

  ~ rc.d/ntpd                             

  > Remove setproctitle() for the parent process. Because rc.d(8) uses process
  > titles (including flags) to distinguish between daemons, this makes it
  > possible to manage multiple copies of a daemon using the normal
  > infrastructure
  > by symlinking rc.d scripts to a new name. ok jung@ ajacoutot@, smtpd ok
  > gilles@ (sthen@)

rc.d/ospf6d

  ~ rc.d/ospf6d                           

  > Remove setproctitle() for the parent process. Because rc.d(8) uses process
  > titles (including flags) to distinguish between daemons, this makes it
  > possible to manage multiple copies of a daemon using the normal
  > infrastructure
  > by symlinking rc.d scripts to a new name. ok jung@ ajacoutot@, smtpd ok
  > gilles@ (sthen@)

rc.d/ospfd

  ~ rc.d/ospfd                            

  > Remove setproctitle() for the parent process. Because rc.d(8) uses process
  > titles (including flags) to distinguish between daemons, this makes it
  > possible to manage multiple copies of a daemon using the normal
  > infrastructure
  > by symlinking rc.d scripts to a new name. ok jung@ ajacoutot@, smtpd ok
  > gilles@ (sthen@)

rc.d/relayd

  ~ rc.d/relayd                           

  > Remove setproctitle() for the parent process. Because rc.d(8) uses process
  > titles (including flags) to distinguish between daemons, this makes it
  > possible to manage multiple copies of a daemon using the normal
  > infrastructure
  > by symlinking rc.d scripts to a new name. ok jung@ ajacoutot@, smtpd ok
  > gilles@ (sthen@)

rc.d/ripd

  ~ rc.d/ripd                             

  > Remove setproctitle() for the parent process. Because rc.d(8) uses process
  > titles (including flags) to distinguish between daemons, this makes it
  > possible to manage multiple copies of a daemon using the normal
  > infrastructure
  > by symlinking rc.d scripts to a new name. ok jung@ ajacoutot@, smtpd ok
  > gilles@ (sthen@)

rc.d/smtpd

  ~ rc.d/smtpd                            

  > Remove setproctitle() for the parent process. Because rc.d(8) uses process
  > titles (including flags) to distinguish between daemons, this makes it
  > possible to manage multiple copies of a daemon using the normal
  > infrastructure
  > by symlinking rc.d scripts to a new name. ok jung@ ajacoutot@, smtpd ok
  > gilles@ (sthen@)

rc.d/snmpd

  ~ rc.d/snmpd                            

  > Remove setproctitle() for the parent process. Because rc.d(8) uses process
  > titles (including flags) to distinguish between daemons, this makes it
  > possible to manage multiple copies of a daemon using the normal
  > infrastructure
  > by symlinking rc.d scripts to a new name. ok jung@ ajacoutot@, smtpd ok
  > gilles@ (sthen@)

rc.d/vmd

  ~ rc.d/vmd                              

  > Remove setproctitle() for the parent process. Because rc.d(8) uses process
  > titles (including flags) to distinguish between daemons, this makes it
  > possible to manage multiple copies of a daemon using the normal
  > infrastructure
  > by symlinking rc.d scripts to a new name. ok jung@ ajacoutot@, smtpd ok
  > gilles@ (sthen@)

== games ============================================================= 04/12 ==

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

snake

  ~ snake.c                               

  > - Swap atoi(3) for strtonum(3)
  > - Swap fputs(3) for fprintf(3)
  > - Use getprogname(3) instead of hardcoded string
  > OK and help from tb@ (mestre@)

== include =========================================================== 05/12 ==

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

signal.h

  ~ signal.h                              

  > Convert sigemptyset and sigfillset to inline functions to avoid
  > warnings with newer gcc versions and -Werror=unused-value.
  > OK guenther@ (millert@)

== lib =============================================================== 06/12 ==

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

libarch

  ~ amd64/amd64_iopl.c                    

  > ansify amd64_iopl() (jca@)

  ~ i386/i386_get_ioperm.c                ~ i386/i386_get_ldt.c
  ~ i386/i386_iopl.c                      ~ i386/i386_set_ioperm.c
  ~ i386/i386_set_ldt.c                   ~ i386/i386_vm86.c

  > ansify; ok sthen@ kettenis@ (jca@)

libc

  ~ locale/rune.c                         

  > avoid an uninitialised value in a return path added in rev 1.5
  > ok tobias@ (jsg@)

  ~ stdlib/malloc.3                       

  > Fix err(3) calls after allocation failures in examples.
  > There is long-standing consensus that err(1, NULL) is the best idiom
  > after failure of malloc(3) and friends.
  > Quirk in the manual noticed by tb@. (schwarze@)

  ~ stdlib/system.3                       ~ gen/popen.3

  > be more forceful about not using these.
  > improvements sthen@, jmc@. okay millert@, jca@ jmc@ (espie@)

  ~ stdlib/strtonum.3                     

  > compare pointer to NULL in example code (mmcc@)

libcrypto

  ~ cert.pem                              

  > Revamp cert.pem certificate information formatting. Skip headers which
  > aren't really useful (the information can be obtained by feeding the cert
  > into "openssl x509 -in filename -text") and add a separator between certs
  > showing the CA's CN or OU (similar to the display format in web browsers).
  > Include both SHA1 and SHA256 fingerprints for all certificates.
  > ok beck@ zhuk@ jung@ (sthen@)

  ~ cert.pem                              

  > Sort cert.pem alphabetically, first by organisation, then by CA name
  > (CN if available, otherwise OU).
  > Add a comment identifying the org. Now to get an easy-to-read list
  > of certificates in the file you can use "grep ^[#=] cert.pem".
  > Prepared with https://spacehopper.org/format-pem.20160201. If you would
  > like to verify this commit to ensure that I didn't sneak in any other
  > changes, it will be easier to use the script rather than do it by hand.
  > (sthen@)

libedit

  ~ filecomplete.c                        

  > trivial partial sync to NetBSD, no functional change:
  > * remove unused ifdef'ed header junk
  > * remove pointless cast of NULL
  > * wrap two excessively long lines (schwarze@)

  ~ keymacro.c                            ~ read.c

  > remove some whitespace differences with NetBSD; no change with diff -b
  > (schwarze@)

  ~ el.c                                  ~ sys.h

  > Get rid of an ugly #ifdef in the middle of el_init() by
  > hiding the system dependencies away in "sys.h" where they belong.
  > No binary change on OpenBSD.
  > Seems reasonable to christos at NetBSD, too. (schwarze@)

libpcap

  ~ optimize.c                            

  > Use the correct structure size when allocating the array for nodes.
  > ok tb@ millert@ (canacar@)

== libexec =========================================================== 07/12 ==

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

talkd

  ~ announce.c                            ~ print.c
  ~ process.c                             ~ table.c
  ~ talkd.c                               

  > Remove unused headers and sort the remaining
  > OK tb@ (mestre@)

  ~ talkd.c                               

  > pledge(2) for talkd(8):
  > rpath: fopen(3) _PATH_UTMP in read mode
  > wpath/cpath: fopen(3) full_tty in write mode (w), which implies O_CREAT
  > inet/dns: the intervening hosts in the conversation may be remote, or not,
  > but
  > since we will never know beforehand then it'll always need inet and dns to
  > resolve the hostnames
  > ok jca@ and also discussed with tb@ (mestre@)

== sbin ============================================================== 08/12 ==

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

bioctl

  ~ bioctl.c                              

  > Fix an unitialized variable; pointed out by jsg@. (uebayasi@)

dhclient

  ~ packet.c                              

  > be very careful accepting packets via bpf. First check that the
  > fixed part of the IP header is completely present before using its
  > header length field.  Then use the data in the IP header to ensure
  > the entire IP packet is present. Then check that the entire UDP header
  > is present. Then use the data in the UDP header to ensure all the
  > data it thinks is present is actually present.
  > Started when tj@ and a few others noticed ISC "DHCP CVE-2015-8605:
  > UDP payload length not properly checked".
  > ok sthen@ henning@ (krw@)

  ~ bpf.c                                 ~ clparse.c
  ~ conflex.c                             ~ dhclient.c
  ~ dhcpd.h                               ~ dispatch.c
  ~ errwarn.c                             ~ kroute.c
  ~ options.c                             ~ packet.c
  ~ parse.c                               ~ privsep.c
  ~ privsep.h                             ~ tables.c

  > Eliminate #include inside *.h files and include only needed headers in
  > each *.c file.
  > Inspired by mention of header silliness by Edgar Pettijohn and mmcc@
  > on tech@. (krw@)

init

  ~ init.c                                

  > Remove variable unneeded since introduction of crypt_checkpass
  > ok millert@ (jca@)

pdisk

  ~ dump.c                                ~ partition_map.c

  > Apple sets lblock_start and lblocks to 0 in free space entries. Do
  > the same.
  > Don't use lblock_start or lblocks in calculation for length@base
  > display. We no longer display lblock info there. (krw@)

  ~ dump.c                                ~ dump.h
  ~ partition_map.h                       

  > Move dump_block() declaration from io.h to io.c since it is not
  > used anywhere else.
  > Remove declarations of non-existant kPatchType and init_partition_map().
  > (krw@)

  ~ dump.c                                ~ partition_map.c
  ~ partition_map.h                       

  > Just call contains_driver() when you need to know rather than carrying a
  > flag around. More obviously current data. Nuke unnecessary variables
  > when displaying a partition. (krw@)

  ~ partition_map.c                       

  > Don't use add_data_to_map() when reading from disk, since it resets some
  > of the dpme_ fields, overwriting data from disk.
  > Add some paranoia checks for map limits when reading from disk. (krw@)

  ~ partition_map.c                       

  > Fold add_data_to_map() into create_entry() as the former is always called
  > immediately after the latter, manipulating fields in the same structure.
  > Use create_entry() in create_partition() rather than doing the same
  > actions manually. (krw@)

  ~ partition_map.c                       

  > Use delete_entry()/create_entry() in delete_partition_from_map() instead of
  > trying to be clever and tweaking the existing entry into a free space
  > entry by manual manipulation of the fields. (krw@)

  ~ dump.c                                

  > Nuke djust_value_and_compute_prefix() and use fmt_scaled() to display
  > human readable sizes. (krw@)

  ~ dump.c                                ~ partition_map.c
  ~ partition_map.h                       ~ pdisk.c

  > Nuke pointless map field 'physical_block' and just use 'sbBlkSize' which is
  > set to the same value. (krw@)

  ~ io.c                                  

  > Use printf("%s",prompt) rather than printf(prompt) just for paranoia's
  > sake. (krw@)

  ~ partition_map.c                       

  > Un-shorten some lines that don't need breaking anymore. Remove a couple of
  > indent levels by shuffling code bit. (krw@)

  ~ pdisk.c                               

  > Pledge, based on what's done in fdisk. (krw@)

== share ============================================================= 09/12 ==

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

man

  ~ man7/ascii.7                          

  > Use standard name for ASCII LF and FF control codes.
  > As seen on FreeBSD; ok bentley@ millert@ (naddy@)

  ~ man4/man4.armv7/a1xintc.4             ~ man4/man4.armv7/sxiccmu.4
  ~ man4/man4.armv7/sxidog.4              ~ man4/man4.armv7/sxie.4
  ~ man4/man4.armv7/sxipio.4              ~ man4/man4.armv7/sxitimer.4

  > rapha mistakenly got the authorship of sunxi drivers wrong when adding
  > man pages.  Problem pointed out by Artturi Alm, who initially wrote
  > them.  Drop the AUTHORS sections instead of correcting them, as preferred
  > by Artturi. (jsg@)

  ~ man4/man4.armv7/sxidog.4              

  > sxidog attaches at sunxi not omap (jsg@)

  + man4/dwctwo.4                         

  > Manual page for dwctwo(4) (visa@)

  ~ man4/usb.4                            

  > Mention dwctwo(4).
  > Suggested by jmc@ (visa@)

  ~ man4/dwctwo.4                         ~ man4/usb.4

  > Octeon has at most one instance of dwctwo(4), at iobus? irq 56. (visa@)

  ~ man4/dwctwo.4                         

  > tweaks; (jmc@)

  ~ man4/Makefile                         

  > hookup dwctwo.4 (jasper@)

  ~ man8/man8.amd64/MAKEDEV.8             ~ man8/man8.i386/MAKEDEV.8

  > Regen (uebayasi@)

  ~ man4/pchtemp.4                        

  > Add support for the 9 series PCH.
  > From Dmitri Alenichev. (kettenis@)

misc

  ~ ascii                                 

  > Use standard name for ASCII LF and FF control codes.
  > As seen on FreeBSD; ok bentley@ millert@ (naddy@)

== sys =============================================================== 10/12 ==

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

arch/amd64/amd64

  ~ amd64_mem.c                           ~ cacheinfo.c
  ~ cpu.c                                 ~ identcpu.c

  > Test cpuid_level or ci->ci_pnfeatset before using a CPUID leaf; some BIOSes
  > can disable leaves that CPU feature flags would seem to imply.  Corrects
  > signal delivery on systems where the AVX leaf is disabled.
  > report and debugging help from Marcus MERIGHI (mcmer-openbsd (at) tor.at)
  > ok kettenis@ (guenther@)

  ~ conf.c                                

  > Implement FreeBSD-compatible IOCTL to access BMC in ipmi(4)
  > Initial help & testing by jmatthew@
  > Code review & input by mpi@
  > Final review & OK by jsg@ (uebayasi@)

arch/amd64/conf

  ~ RAMDISK_CD                            

  > enable xen devices here as well
  > tested by and ok mikeb@ (jsg@)

arch/amd64/stand/efiboot

  ~ efiboot.c                             

  > Some (buggy) UEFI implementations don't report the correct framebuffer
  > parameters unless SetMode() has been called, even if we don't actually
  > change the mode.  Other (equally buggy) UEFI implementations report an
  > error if we call SetMode() but don't actually change the mode.  Work around
  > these issues by always calling SetMode() (like we did before rev 1.10), but
  > don't report an error if the current mode is the same as the desired mode.
  > ok yasuoka@ (kettenis@)

arch/arm/arm

  ~ arm32_machdep.c                       ~ cpufunc_asm_sa1.S
  ~ cpufunc_asm_xscale.S                  ~ cpuswitch.S
  ~ cpuswitch7.S                          ~ fault.c
  ~ fiq.c                                 ~ locore.S
  ~ pmap.c                                ~ pmap7.c
  ~ process_machdep.c                     ~ sig_machdep.c
  ~ syscall.c                             ~ undefined.c

  > Switch from PSR_X_bit and X32_bit PSR macro names to just PSR_X.
  > This matches FreeBSD and makes things a bit more consistent.
  > Discussed with Patrick. (jsg@)

  ~ pmap.c                                ~ pmap7.c

  > Fix the encoding of AP bits for large page second-level
  > short-descriptors with arm v7 (same as small page encoding, except XN is
  > in a different bit for the mask).
  > Expanded version of a diff from Patrick Wildt who also tested and
  > reviewed this. (jsg@)

arch/arm/cortex

  ~ ampintc.c                             

  > Switch from PSR_X_bit and X32_bit PSR macro names to just PSR_X.
  > This matches FreeBSD and makes things a bit more consistent.
  > Discussed with Patrick. (jsg@)

arch/arm/include

  ~ armreg.h                              ~ atomic.h
  ~ cpu.h                                 ~ cpufunc.h
  ~ frame.h                               

  > Switch from PSR_X_bit and X32_bit PSR macro names to just PSR_X.
  > This matches FreeBSD and makes things a bit more consistent.
  > Discussed with Patrick. (jsg@)

  ~ pmap.h                                

  > Fix the encoding of AP bits for large page second-level
  > short-descriptors with arm v7 (same as small page encoding, except XN is
  > in a different bit for the mask).
  > Expanded version of a diff from Patrick Wildt who also tested and
  > reviewed this. (jsg@)

arch/arm/sa11x0

  ~ sa11x0_ost.c                          

  > Switch from PSR_X_bit and X32_bit PSR macro names to just PSR_X.
  > This matches FreeBSD and makes things a bit more consistent.
  > Discussed with Patrick. (jsg@)

arch/arm/xscale

  ~ i80321_clock.c                        ~ i80321_intr.c
  ~ i80321_pci.c                          ~ pxa2x0.c
  ~ pxa2x0_apm.c                          ~ pxa2x0_apm_asm.S
  ~ pxa2x0_gpio.c                         ~ pxa2x0_intr.c
  ~ pxa2x0_lcd.c                          

  > Switch from PSR_X_bit and X32_bit PSR macro names to just PSR_X.
  > This matches FreeBSD and makes things a bit more consistent.
  > Discussed with Patrick. (jsg@)

arch/armv7/armv7

  ~ armv7_start.S                         ~ intr.c

  > Switch from PSR_X_bit and X32_bit PSR macro names to just PSR_X.
  > This matches FreeBSD and makes things a bit more consistent.
  > Discussed with Patrick. (jsg@)

  ~ armv7_machdep.c                       

  > Leonardo Guardati discovered a u-boot built with "Enable workarounds for
  > booting old kernels" was required to boot sunxi kernels with newer
  > versions of u-boot.
  > Unless built with OLD_SUNXI_KERNEL_COMPAT u-boot now uses the top four
  > bits of the board id for a 'compatibility revision'.  This was
  > supposedly for different handling of the pll5 dram clock, but we don't
  > touch that currently.  Mask these bits away so the kernel will boot from
  > a u-boot that doesn't have the option set.
  > Tested by Patrick Wildt on a Banana Pi R1 with current u-boot sources.
  > (jsg@)

arch/armv7/exynos

  ~ exynos.c                              

  > remove uneeded break statements (jsg@)

arch/armv7/imx

  ~ imx.c                                 

  > remove uneeded break statements (jsg@)

arch/armv7/omap

  ~ intc.c                                ~ omgpio.c

  > Switch from PSR_X_bit and X32_bit PSR macro names to just PSR_X.
  > This matches FreeBSD and makes things a bit more consistent.
  > Discussed with Patrick. (jsg@)

  ~ omap.c                                

  > remove uneeded break statements (jsg@)

arch/armv7/sunxi

  ~ a1xintc.c                             

  > Switch from PSR_X_bit and X32_bit PSR macro names to just PSR_X.
  > This matches FreeBSD and makes things a bit more consistent.
  > Discussed with Patrick. (jsg@)

  ~ sunxi.c                               ~ sunxireg.h
  ~ sxirtc.c                              

  > Don't use sxitimer on Allwinner sun7i/A20.
  > sxitimer isn't generating hardclock ticks on Allwinner A20.  Fortunately
  > the Cortex A7 based Allwinner A20 has the ARM Generic Timer (agtimer)
  > available so use that.
  > To allow this to work sxirtc gets it's own mapping instead of using a
  > subregion of sxitimer.
  > From Patrick Wildt. (jsg@)

  ~ sxirtc.c                              

  > Add A20 support to sxirtc.
  > From Artturi Alm. (jsg@)

  ~ sun7i.c                               ~ sxitimer.c

  > Remove A20 support in sxitimer, agtimer is now used instead.
  > From Artturi Alm. (jsg@)

  ~ sunxi.c                               

  > remove uneeded break statements (jsg@)

  ~ sunxireg.h                            ~ sxidog.c

  > give sxidog it's own mapping as well so it wont't require sxitimer
  > from Artturi Alm (jsg@)

arch/armv7/vexpress

  ~ vexpress.c                            

  > remove uneeded break statements (jsg@)

arch/i386/i386

  ~ cpu.c                                 

  > Test cpuid_level or ci->ci_pnfeatset before using a CPUID leaf; some BIOSes
  > can disable leaves that CPU feature flags would seem to imply.  Corrects
  > signal delivery on systems where the AVX leaf is disabled.
  > report and debugging help from Marcus MERIGHI (mcmer-openbsd (at) tor.at)
  > ok kettenis@ (guenther@)

  ~ conf.c                                

  > Implement FreeBSD-compatible IOCTL to access BMC in ipmi(4)
  > Initial help & testing by jmatthew@
  > Code review & input by mpi@
  > Final review & OK by jsg@ (uebayasi@)

arch/mips64/include

  ~ pmap.h                                

  > Move modify bit emulation into pmap.c to gather pmap C code in one place.
  > (visa@)

  ~ pmap.h                                

  > Zap stray pmap_kenter_cache() prototype. (visa@)

arch/mips64/mips64

  ~ pmap.c                                ~ trap.c

  > Move modify bit emulation into pmap.c to gather pmap C code in one place.
  > (visa@)

conf

  ~ newvers.sh                            

  > move to -release mode
  > requested by deraadt@ (jsg@)

dev

  ~ ipmi.c                                ~ ipmivar.h
  + ipmi.h                                

  > Implement FreeBSD-compatible IOCTL to access BMC in ipmi(4)
  > Initial help & testing by jmatthew@
  > Code review & input by mpi@
  > Final review & OK by jsg@ (uebayasi@)

  ~ ipmi.c                                

  > Remove a too strict assertion. (uebayasi@)

  ~ ipmi.c                                

  > Comment. (uebayasi@)

dev/i2c

  ~ ihidev.h                              

  > remove old link to documentation, it's in ihidev.c (jcs@)

dev/pci

  ~ drm/radeon/radeon_kms.c               

  > Prevent efifb(4) from attaching if we're the console.
  > Unfortunately, making this decision in radeondrm_attachhook() is too late
  > because at that point efifb(4) would have already been attached.  This
  > means
  > that if we decide to bail in radeondrm_attachhook() we may end up without
  > a glass console.  That may happen for example if the firmware package
  > has not been installed.  I'm still looking for a solution for that problem.
  > ok jsg@ (kettenis@)

  ~ drm/radeon/radeon_bios.c              

  > If the ROM size is 0, we won't be able to read the video BIOS from it.
  > Bail out early to prevent a panic from calling bus_space_map(9) with size
  > 0. (kettenis@)

  ~ if_iwn.c                              

  > In iwn(4), don't forget about restoring the CCMP key to firmware after
  > HT protection setting updates. Unbreaks WPA in 11n mode.
  > ok sthen@ (stsp@)

  ~ if_iwn.c                              

  > Make iwn(4) pass 802.11 control frames in monitor mode.
  > Now we can see rts/cts, ack, blockack etc. in tcpdump(8).
  > ok kettenis@ (stsp@)

  ~ drm/radeon/radeon_bios.c              

  > Fix a pci vendor/device conversion missed in rev 1.6. (kettenis@)

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

  > Improve Linux PCI compatibility code. (kettenis@)

  ~ drm/drm_linux.c                       ~ drm/drm_linux.h

  > Implement acpi_get_table_with_size().  Will soon be used to read VFCT
  > tables in radeondrm(4). (kettenis@)

  ~ if_iwm.c                              

  > In iwm(4), set max A-MPDU length to 64k instead of 4k and tell the firmware
  > about A-MPDU spacing. Makes BlockAck happy.
  > tested by myself and abieber@ (stsp@)

  ~ if_iwn.c                              ~ if_iwnreg.h

  > In iwn(4), set max A-MPDU length to 64k instead of 4k and tell the firmware
  > about A-MPDU spacing. Makes BlockAck happy.
  > tested by me, tb@, krw@, sthen@, and Henrik Friedrichsen (stsp@)

  ~ if_iwn.c                              

  > Fix iwn(4) CCMP replay detection so it does not drop out-of-order A-MPDU
  > subframes. Helps 11n mode with WPA.
  > tested by me, tb@, and krw@ (stsp@)

  ~ drm/radeon/radeon_bios.c              

  > Enable the code that reads the BIOS from the ACPI VFCT table on platforms
  > with ACPI.
  > ok jsg@ (kettenis@)

  ~ pchtemp.c                             

  > Add support for the 9 series PCH.
  > From Dmitri Alenichev. (kettenis@)

  ~ if_iwmreg.h                           

  > Make sure the maximum iwm(4) firmware command payload size fits into
  > the 12 bits available for it in struct iwm_tfd_tb.hi_n_len.
  > Patch by Imre Vadasz via tech@ (stsp@)

  ~ if_iwn.c                              

  > Fix off-by-one in loop termination condition.
  > ok stsp@ (tb@)

dev/pv

  ~ xen.c                                 ~ xenstore.c
  ~ xenvar.h                              

  > A few reliability improvements in the power management interface
  > Nathanael Rensen <nathanael at list ! polymorpheus ! com> came up with
  > a few improvements to the event watcher and power management interface,
  > namely:
  > o Make sure to put our watcher on a list before issuing an XS_WATCH
  > command since Xen will raise the event right after it's been set up.
  > o The first time xen_control is called the "control/shutdown" node
  > may not exist, so skip printing the error message in this case.
  > o Acknowledge requests by writing back an empty string.
  > o log(9) reboot and halt requests like vmt(4) does.
  > Huge thanks! (mikeb@)

  ~ vmt.c                                 

  > Increase the RPC buffer once more to 4096 bytes.  The guestinfo.ovfEnv
  > mechanism is used to configure VMs in the VMware vSphere world:
  > instead of using individual key-value guestinfo.* properties, it uses
  > the guestinfo.ovfEnv value to pass an enterprise-compliant XML file
  > that includes key-value properties.  This file can be rather large,
  > especially with comments, but 4k ought to be enough for anybody.
  > Also change a stack buffer to malloc'ed memory in the ioctl path.
  > OK mikeb@ (reyk@)

  ~ xenstore.c                            

  > Bail early if there's no token; found by jsg@ (mikeb@)

  ~ xen.c                                 ~ xenstore.c

  > Silence warnings from static analyzers; found by jsg@ (mikeb@)

  ~ if_xnf.c                              

  > Simple moderation of Tx completion notifications
  > Follow FreeBSD and schedule the next Tx completion event to fire
  > when about half of the packet segments scheduled for transmission
  > are consumed. (mikeb@)

isofs/udf

  ~ udf_vnops.c                           

  > Convert to uiomove(). From Martin Natano. (stefan@)

kern

  ~ uipc_mbuf.c                           

  > - m_pkthdr.pf.statekey changes are not ready for 5.9, I must back them out
  > OK sthen@ (sashan@)

  ~ kern_synch.c                          

  > add a DIAGNOSTIC for refcnt_take overflow.
  > ok mpi@ (dlg@)

msdosfs

  ~ msdosfs_vnops.c                       

  > Guard against integer overflow when checking whether
  > writing to a file on msdosfs stays within the max. file size.
  > ok kettenis@, krw@ (stefan@)

net

  ~ if_pfsync.c                           ~ pf.c
  ~ pfvar.h                               

  > - m_pkthdr.pf.statekey changes are not ready for 5.9, I must back them out
  > OK sthen@ (sashan@)

  ~ if_pppx.c                             ~ if_tun.c

  > Convert to ulmin and uiomove to prevent integer truncations.
  > Reviewed by Martin Natano. (stefan@)

  ~ bpf.c                                 

  > return if the bpf_if passed to bpf_tap and _bpf_mtap are NULL.
  > this works around a toctou bug in a very common idiom in our tree,
  > in between the two lines below:
  > if (ifp->if_bpf)
  > bpf_mtap(ifp->if_bpf, m, BPF_DIRECTION_OUT);
  > figured out by and diff from haesbart (dlg@)

net80211

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

  > Stop requiring a BlockAck session timeout. We still honour the timeout if
  > the AP requests it, though I don't think I've ever seen one that does.
  > Per the 802.11-2012 spec a value of zero disables the timeout and some
  > APs seem to have trouble with timeouts forced onto them.
  > Our behaviour now matches Linux in this regard.
  > This gets Apple Airport APs somewhat working in 11n mode but there
  > is still a remaining issue with receiving frames from them.
  > Doesn't seem to hurt any other APs I've tried.
  > ok sthen@ mpi@ millert@ (stsp@)

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

  > Restore the BlockAck session timer. It is still required to work around
  > stalled BA sessions observed with iwn(4). We can revisit this later once
  > the underlying problem in iwn(4) has been found.
  > Prompted by report from krw@, I could reproduce the issue.
  > ok krw@ (stsp@)

  ~ ieee80211_input.c                     ~ ieee80211_node.h
  ~ ieee80211_output.c                    ~ ieee80211_proto.c

  > Store ADDBA request and response parameters in the block ack record of
  > ieee80211_node. This way, we can keep track of the ACK policy and echo
  > it back to the AP as required by the standard. And use the correct bit
  > flag for the policy -- this code was confused between BlockAck and ADDBA,
  > both of which have a policy bit but in different places.
  > Fixes apple airport APs.
  > tested by tb@, krw@, sthen@, abieber@, and Henrik Friedrichsen (stsp@)

  ~ ieee80211_output.c                    

  > Define EDCA tables for 11n mode. Per 802.11-2012 they are the same as
  > the tables for 11a/11g modes.
  > Add a function to append a Microsoft WME parameter element to a frame,
  > using EDCA tables to construct the parameter records. Some 11n AP's I have
  > observed provide this element.
  > None of this code is used yet, so no functional change. I wrote this while
  > debugging BlockAck and then realized it was only needed for hostap mode.
  > Once we support 11n in hostap mode and send A-MPDUs, this code will be
  > needed. (stsp@)

  ~ ieee80211_input.c                     

  > Count A-MPDU subframes with sequence number below the current BA window as
  > duplicates rather than input errors. These subframes have either already
  > been received, or the window was moved by the gap timeout which should only
  > happen with buggy APs. Neither condition indicates a severe problem.
  > Perhaps we will introduce a separate counter for this later. (stsp@)

  ~ ieee80211_input.c                     

  > Hide some excessive net80211 debug printfs behind ieee80211_debug >= 2.
  > (stsp@)

netinet

  ~ ip_input.c                            

  > - m_pkthdr.pf.statekey changes are not ready for 5.9, I must back them out
  > OK sthen@ (sashan@)

ntfs

  ~ ntfs_subr.c                           

  > Convert to uiomove. While there, use size_t consistently for
  > variables 'left' and 'tocopy' and adapt printf and DPRINTF format
  > strings accordingly.
  > From Martin Natano. (stefan@)

scsi

  ~ sd.c                                  

  > When accessing the scsi link of a scsi disk, use a variable "sc_link"
  > everywhere.  This is the first step to fix a use after free of the
  > sc_link when the disk detaches.  If a function gets a scsi transfer,
  > the scsi link is always valid.  Call this variable "link" consistently.
  > OK krw@ (bluhm@)

  ~ sd.c                                  

  > Do not access the scsi link of the disk at the beginning of sdopen()
  > and sdminphys() if the scsi disk is dying.  The memory of the link
  > may have been freed already.
  > OK krw@ (bluhm@)

sys

  ~ mbuf.h                                

  > - m_pkthdr.pf.statekey changes are not ready for 5.9, I must back them out
  > OK sthen@ (sashan@)

  ~ conf.h                                

  > Implement FreeBSD-compatible IOCTL to access BMC in ipmi(4)
  > Initial help & testing by jmatthew@
  > Code review & input by mpi@
  > Final review & OK by jsg@ (uebayasi@)

tmpfs

  ~ tmpfs_subr.c                          ~ tmpfs_vnops.c

  > Convert to uiomove. From Martin Natano. (stefan@)

ufs/ext2fs

  ~ ext2fs_subr.c                         

  > Prevent signed overflow when calculating i_modrev initial value
  > Reported and fixed by Martin Natano <natano at natano ! net>, thanks!
  > ok stefan (mikeb@)

ufs/ufs

  ~ ufs_vnops.c                           

  > Prevent signed overflow when calculating i_modrev initial value
  > Reported and fixed by Martin Natano <natano at natano ! net>, thanks!
  > ok stefan (mikeb@)

== usr.bin =========================================================== 11/12 ==

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

cu

  ~ xmodem.c                              

  > Include stdint.h for the uint{8,16}_t types.  OK nicm@ (millert@)

doas

  ~ doas.c                                

  > require a tty for the password.
  > sudo regression noted by miod, fix by millert (tedu@)

ftp

  ~ fetch.c                               

  TAGGED OPENBSD_5_8
  > Backport fix for crash when non-standard newline is returned.
  > OK sthen@ and beck@ (martijn@)

  ~ fetch.c                               

  TAGGED OPENBSD_5_7
  > Backport fix for crash when non-standard newline is returned.
  > OK sthen@ and beck@ (martijn@)

getent

  ~ getent.c                              

  > Pointless #define
  > was:
  > #define SHELLSPRINT     printf("%s\n", sh) (jca@)

  ~ getent.c                              

  > Hmm, SHELLSPRINT may be pointless, but keep it for consistency with other
  > dbs. (jca@)

less

  ~ optfunc.c                             

  > Rename a local var so that we can call the global message buffer msg.
  > ok nicm, tb (mmcc@)

locale

  ~ locale.c                              

  > No need for pledge("rpath") (jca@)

  ~ locale.c                              

  > Sort includes. (jca@)

ssh

  ~ clientloop.c                          

  > Avoid ugly "DISPLAY "(null)" invalid; disabling X11 forwarding"
  > message when DISPLAY is not set.  This could also result in a crash
  > on systems with a printf that doesn't handle NULL.  OK djm@ (millert@)

  ~ packet.c                              

  > printf argument casts to avoid warnings on strict compilers (djm@)

  ~ sshd_config.5                         

  > better description for MaxSessions; bz#2531 (djm@)

  ~ sshd.8                                

  > mention internal DH-GEX fallback groups; bz#2302 (djm@)

  ~ sshconnect2.c                         

  > avoid an uninitialised value when NumberOfPasswordPrompts is 0
  > ok markus@ djm@ (jsg@)

  ~ channels.c                            

  > Only check errno if read() has returned an error.  EOF is not an error.
  > This fixes a problem where the mux master would sporadically fail to
  > notice that the client had exited.
  > ok mikeb@ djm@ (naddy@)

tail

  ~ tail.c                                

  > fix off-by-one in argument parsing
  > ok martijn@ (halex@)

talk

  ~ ctl.c                                 ~ ctl_transact.c
  ~ display.c                             ~ get_addrs.c
  ~ get_names.c                           ~ init_disp.c
  ~ invite.c                              ~ io.c
  ~ look_up.c                             ~ msgs.c
  ~ talk.c                                ~ talk.h

  > Remove unused headers and sort the remaining
  > Remove lint-style comments
  > Swap hardcoded string on usage() for __progname
  > OK tb@ (mestre@)

  ~ talk.c                                

  > pledge(2) for talk(1):
  > At the beginning the largest pledge is the following:
  > rpath: read ~/.terminfo (the reason was changed pointed out by semarie@)
  > inet/dns: talk may need to connect to a remote host and resolve it
  > getpw: if getlogin(2) fails then it needs getpwuid(3) as a fallback
  > tty: this is a typical tty application, so it'll always need this
  > annotation
  > Then just before the application main loop check if the talk is with local
  > user
  > so it only needs "stdio tty", if it's remote then it needs "stdio inet
  > tty".
  > I couldn't test this with a remote host to confirm if it needs inet or not
  > but
  > as per jca@'s comment "tighter settings - if possible - can happen later"
  > ok jca@ and also discussed with tb@ (mestre@)

tmux

  ~ cmd-new-session.c                     

  > Fix new-session with -t after command flags changes, reported by Michael
  > Graczyk. (nicm@)

  ~ format.c                              ~ tmux.1

  > Support negative trim values (#{=-10:pane_title}) to trim from the end,
  > suggested by Kevin Brubeck Unhammer. (nicm@)

  ~ grid.c                                

  > Add RGB escape sequences for capture-pane -e. (nicm@)

  ~ tmux.h                                ~ utf8.c

  > Whoops, need this for the previous reverse trim commit too. (nicm@)

  ~ screen-write.c                        

  > Clear RGB flags during selection. (nicm@)

  ~ cmd-new-session.c                     

  > Set up -t flag properly when passing new-session -A off to
  > attach-session, GitHub issue 295. (nicm@)

  ~ window-copy.c                         

  > Do not wrap cursor at start or end of history, from Michal Mazurek. (nicm@)

units

  ~ units.lib                             

  > update currency exchange rates; (jmc@)

vi

  ~ cl/cl_main.c                          ~ common/main.c
  ~ common/mem.h                          

  > Remove needless alias macros for malloc and calloc. No binary change. I
  > got this upstreamed a few weeks ago.
  > ok tb (less a few style tweaks), martijn (who suggested style tweaks)
  > (mmcc@)

write

  ~ write.1                               ~ write.c

  > Make write explicitly ASCII only by transforming UTF-8 characters and
  > non-ASCII
  > bytes to a single '?'. This prevents sending of potentially harmful bytes
  > to
  > terminals who don't support UTF-8.
  > written with help from schwarze@
  > text suggestions by jmc@
  > OK schwarze@ and semarie@ (martijn@)

  ~ write.1                               

  > Explain why we intentionally violate POSIX here, to prevent people from
  > being sad that they can't use UTF-8 in write(1), and to prevent that
  > they waste their time trying to fix it.
  > While here, fix some broken escaping.
  > OK jmc@ martijn@ (schwarze@)

xinstall

  ~ install.1                             

  > missing word; from jan stary (jmc@)

== usr.sbin ========================================================== 12/12 ==

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

usr.sbin

  ~ Makefile                              

  > Enable hostctl (only i386 and amd64 for now).
  > OK mikeb@ jsg@ krw@ ajacoutot@ (reyk@)

bgpd

  ~ bgpd.c                                

  > Remove setproctitle() for the parent process. Because rc.d(8) uses process
  > titles (including flags) to distinguish between daemons, this makes it
  > possible to manage multiple copies of a daemon using the normal
  > infrastructure
  > by symlinking rc.d scripts to a new name. ok jung@ ajacoutot@, smtpd ok
  > gilles@ (sthen@)

dhcpd

  ~ packet.c                              

  > be very careful accepting packets via bpf. First check that the
  > fixed part of the IP header is completely present before using its
  > header length field.  Then use the data in the IP header to ensure
  > the entire IP packet is present. Then check that the entire UDP header
  > is present. Then use the data in the UDP header to ensure all the
  > data it thinks is present is actually present.
  > Started when tj@ and a few others noticed ISC "DHCP CVE-2015-8605:
  > UDP payload length not properly checked".
  > ok sthen@ henning@ (krw@)

  ~ alloc.c                               ~ bootp.c
  ~ bpf.c                                 ~ conflex.c
  ~ confpars.c                            ~ convert.c
  ~ db.c                                  ~ dhcp.c
  ~ dhcpd.c                               ~ dhcpd.h
  ~ dispatch.c                            ~ errwarn.c
  ~ hash.c                                ~ icmp.c
  ~ inet.c                                ~ memory.c
  ~ options.c                             ~ packet.c
  ~ parse.c                               ~ pfutils.c
  ~ print.c                               ~ sync.c
  ~ tables.c                              ~ tree.c
  ~ udpsock.c                             

  > Eliminate #include inside *.h files and include only needed headers in
  > each *.c file.
  > Inspired by mention of header silliness by Edgar Pettijohn and mmcc@
  > on tech@. (krw@)

  ~ memory.c                              

  > avoid a use after free when parsing address ranges from a config
  > ok krw@ (jsg@)

dhcrelay

  ~ bpf.c                                 ~ dhcpd.h
  ~ packet.c                              

  > Remove unused (a.k.a. always passed NULL) parameter 'data' from
  > decode_udp_ip_header() and the useless check of it. Part of original
  > diff from pelikan about udp length errors.
  > From dhcpd: bpf.c r1.9, dhcpd.h r1.46, packet.c r1.5
  > ok jca (sthen@)

  ~ packet.c                              

  > Nuke local #define ETHER_HEADER_SIZE and use the identical value'd
  > ETHER_HDR_LEN from if_ether.h.
  > From dhcpd packet.c r1.6.  ok jca (sthen@)

  ~ packet.c                              

  > Variables to count seen packets should be unsigned.
  > From dhcpd/packet.c r1.8   ok jca (sthen@)

  ~ packet.c                              

  > be very careful accepting packets via bpf. First check that the
  > fixed part of the IP header is completely present before using its
  > header length field.  Then use the data in the IP header to ensure
  > the entire IP packet is present. Then check that the entire UDP header
  > is present. Then use the data in the UDP header to ensure all the
  > data it thinks is present is actually present.
  > Started when tj@ and a few others noticed ISC "DHCP CVE-2015-8605:
  > UDP payload length not properly checked".
  > ok sthen@ henning@ (krw@)

  ~ bpf.c                                 ~ dhcpd.h
  ~ dhcrelay.c                            ~ dispatch.c
  ~ errwarn.c                             ~ packet.c

  > Eliminate #include inside *.h files and include only needed headers in
  > each *.c file.
  > Inspired by mention of header silliness by Edgar Pettijohn and mmcc@
  > on tech@. (krw@)

dvmrpd

  ~ dvmrpd.c                              

  > Remove setproctitle() for the parent process. Because rc.d(8) uses process
  > titles (including flags) to distinguish between daemons, this makes it
  > possible to manage multiple copies of a daemon using the normal
  > infrastructure
  > by symlinking rc.d scripts to a new name. ok jung@ ajacoutot@, smtpd ok
  > gilles@ (sthen@)

eigrpd

  ~ eigrpd.c                              

  > Remove setproctitle() for the parent process. Because rc.d(8) uses process
  > titles (including flags) to distinguish between daemons, this makes it
  > possible to manage multiple copies of a daemon using the normal
  > infrastructure
  > by symlinking rc.d scripts to a new name. ok jung@ ajacoutot@, smtpd ok
  > gilles@ (sthen@)

hostapd

  ~ privsep.c                             

  > Remove setproctitle() for the parent process. Because rc.d(8) uses process
  > titles (including flags) to distinguish between daemons, this makes it
  > possible to manage multiple copies of a daemon using the normal
  > infrastructure
  > by symlinking rc.d scripts to a new name. ok jung@ ajacoutot@, smtpd ok
  > gilles@ (sthen@)

httpd

  ~ httpd.c                               

  > Remove setproctitle() for the parent process. Because rc.d(8) uses process
  > titles (including flags) to distinguish between daemons, this makes it
  > possible to manage multiple copies of a daemon using the normal
  > infrastructure
  > by symlinking rc.d scripts to a new name. ok jung@ ajacoutot@, smtpd ok
  > gilles@ (sthen@)

identd

  ~ Makefile                              

  > No need for SRCS, identd.c is deduced from PROG and make(1) rules. (jca@)

ldapctl

  ~ ldapctl.8                             ~ ldapctl.c

  > Implement ldapctl -r datadir; ok sthen@ landry@ (jca@)

  ~ ldapctl.8                             

  > tweak previous; (jmc@)

  ~ ldapctl.c                             

  > Try to be user-friendly and check early that the datadir is somewhat sane.
  > Inspired by a diff by gsoares@, ok landry@ gsoares@ (jca@)

ldapd

  ~ ldapd.8                               ~ ldapd.c
  ~ namespace.c                           

  > Add -r argument to ldapd, to specify an alternative directory to
  > store/read the database, still defaulting to /var/db/ldap.
  > This will allow running totally separate instances, to be used by an
  > upcoming regress suite for example.
  > With a tweak from gsoares@ to check that the directory exists.
  > ok dlg@ semarie@ jca@ (landry@)

  ~ ldapd.8                               

  > some -r fixes; (jmc@)

  ~ ldapd.c                               

  > use stat(2) instead of chdir(2) to check if given the directory is valid.
  > OK landry@ jca@ (gsoares@)

  ~ ldapd.c                               ~ namespace.c

  > Minor ldapd -r tweaks
  > - fix style
  > - the string pointed to by datadir shouldn't be modified, use const
  > - initialize datadir at compile time
  > - in namespace.c, move the extern datadir decl above local decls (jca@)

ldpd

  ~ ldpd.c                                

  > Remove setproctitle() for the parent process. Because rc.d(8) uses process
  > titles (including flags) to distinguish between daemons, this makes it
  > possible to manage multiple copies of a daemon using the normal
  > infrastructure
  > by symlinking rc.d scripts to a new name. ok jung@ ajacoutot@, smtpd ok
  > gilles@ (sthen@)

npppd

  ~ npppd/privsep.c                       

  > Remove setproctitle() for the parent process. Because rc.d(8) uses process
  > titles (including flags) to distinguish between daemons, this makes it
  > possible to manage multiple copies of a daemon using the normal
  > infrastructure
  > by symlinking rc.d scripts to a new name. ok jung@ ajacoutot@, smtpd ok
  > gilles@ (sthen@)

ntpd

  ~ ntpd.c                                

  > Remove setproctitle() for the parent process. Because rc.d(8) uses process
  > titles (including flags) to distinguish between daemons, this makes it
  > possible to manage multiple copies of a daemon using the normal
  > infrastructure
  > by symlinking rc.d scripts to a new name. ok jung@ ajacoutot@, smtpd ok
  > gilles@ (sthen@)

ospf6d

  ~ ospf6d.c                              

  > Remove setproctitle() for the parent process. Because rc.d(8) uses process
  > titles (including flags) to distinguish between daemons, this makes it
  > possible to manage multiple copies of a daemon using the normal
  > infrastructure
  > by symlinking rc.d scripts to a new name. ok jung@ ajacoutot@, smtpd ok
  > gilles@ (sthen@)

ospfd

  ~ ospfd.c                               

  > Remove setproctitle() for the parent process. Because rc.d(8) uses process
  > titles (including flags) to distinguish between daemons, this makes it
  > possible to manage multiple copies of a daemon using the normal
  > infrastructure
  > by symlinking rc.d scripts to a new name. ok jung@ ajacoutot@, smtpd ok
  > gilles@ (sthen@)

pkg_add

  ~ OpenBSD/Delete.pm                     ~ OpenBSD/PackingElement.pm

  > update the font cache by running fc-cache after removing packages
  > with @fontdir markers
  > ok espie@ (robert@)

relayd

  ~ relayd.c                              

  > Remove setproctitle() for the parent process. Because rc.d(8) uses process
  > titles (including flags) to distinguish between daemons, this makes it
  > possible to manage multiple copies of a daemon using the normal
  > infrastructure
  > by symlinking rc.d scripts to a new name. ok jung@ ajacoutot@, smtpd ok
  > gilles@ (sthen@)

ripd

  ~ ripd.c                                

  > Remove setproctitle() for the parent process. Because rc.d(8) uses process
  > titles (including flags) to distinguish between daemons, this makes it
  > possible to manage multiple copies of a daemon using the normal
  > infrastructure
  > by symlinking rc.d scripts to a new name. ok jung@ ajacoutot@, smtpd ok
  > gilles@ (sthen@)

rtadvd

  ~ config.c                              ~ if.c

  > rtadvd used a strncpy with a src interface name from the command line
  > without checking its length.  Replace it with strlcpy, check for
  > truncation and move the check before using the user-specified
  > interface name.  While here, replace another strncpy of a interface
  > name with memset-zero + strlcpy and check for truncation in
  > if_getmtu() (should not happen as we validated the length before, but
  > it's better to be safe and generally a better practice).
  > OK jsg@ (reyk@)

  ~ rtadvd.c                              

  > Bail out if an unknown option is passed.
  > There is no reason for rtadvd to be different from other daemons.
  > Input from and ok bluhm@ florian@ (jca@)

  ~ rtadvd.c                              

  > select -> poll in log message; ok bluhm@ florian@ (jca@)

  ~ log.h                                 ~ rtadvd.c

  > Mark a few others functions as __dead; ok bluhm@ (jca@)

smtpd

  ~ control.c                             ~ mda.c

  > Check imsg data size before use.
  > Ok eric@ millert@ (sunil@)

  ~ smtpd.c                               

  > Remove setproctitle() for the parent process. Because rc.d(8) uses process
  > titles (including flags) to distinguish between daemons, this makes it
  > possible to manage multiple copies of a daemon using the normal
  > infrastructure
  > by symlinking rc.d scripts to a new name. ok jung@ ajacoutot@, smtpd ok
  > gilles@ (sthen@)

  ~ ca.c                                  

  > in RSA privsep engine, do not provide methods for rsa_sign / rsa_verify,
  > they are unused in OpenSMTPD and lead to crashes in -portable when we're
  > linked to OpenSSL starting with 1.0.2f
  > ok reyk@ (gilles@)

  ~ bounce.c                              

  > Use "esc_class" to classify bounce type instead of "errorline" as
  > we no longer prepend status code to "errorline". Fixes mismatch
  > between DSN's subject line and its content.
  > Ok jung@ gilles@ millert@ (sunil@)

  ~ enqueue.c                             ~ smtpctl.c

  > NULL initialize two file pointers to make it more obvious when reading that
  > they can't end up uninitialized after the mkstemp/fdopen dance (gilles@)

  ~ filter.c                              

  > When a filter rejects a message content, report immediatly to the
  > smtp session. Simplify code while here.
  > ok gilles@ jung@ (eric@)

  ~ smtp_session.c                        

  > tell filters to rollback the current transaction if MAIL or EOM fails
  > ok gilles@ jung@ (eric@)

  ~ smtp_session.c                        

  > Add helpers function to talk to the queue. Make the code simpler.
  > ok gilles@ sunil@ (eric@)

  ~ queue_backend.c                       

  > When deleting a message, remove associated envelopes from the cache,
  > if any (it happens when a message transaction is cancelled).
  > ok gilles@ sunil@ (eric@)

  ~ smtp_session.c                        

  > Make code more readable by putting the message parsing logic in a single
  > place.
  > ok gilles@ sunil@ jung@ (eric@)

  ~ rfc2822.c                             

  > fix memory leak
  > ok gilles@ jung@ (eric@)

  ~ smtp_session.c                        ~ smtpd.c

  > use SMTPD_NAME define
  > ok gilles (jung@)

  ~ smtpd.h                               

  > remove no longer relevant ifndef
  > suggested by gilles (jung@)

snmpd

  ~ snmpd.c                               

  > Remove setproctitle() for the parent process. Because rc.d(8) uses process
  > titles (including flags) to distinguish between daemons, this makes it
  > possible to manage multiple copies of a daemon using the normal
  > infrastructure
  > by symlinking rc.d scripts to a new name. ok jung@ ajacoutot@, smtpd ok
  > gilles@ (sthen@)

sysmerge

  ~ sysmerge.sh                           

  > Uppecase. (ajacoutot@)

tcpdump

  ~ print-802_11.c                        

  > Make tcpdump show 802.11 QoS frames properly.
  > ok kettenis@ (stsp@)

  ~ print-802_11.c                        

  > Make tcpdump show 802.11 control frames.
  > ok kettenis@, and help from David Vasek (thanks!) (stsp@)

  ~ print-802_11.c                        

  > Fix an infinite loop when printing a country element in a management
  > frame in case we hit channel Tx power limits we can't pretty-print.
  > Also ensure we consume the last item in this list.
  > ok sthen@ (stsp@)

vmd

  ~ vmd.c                                 

  > Remove setproctitle() for the parent process. Because rc.d(8) uses process
  > titles (including flags) to distinguish between daemons, this makes it
  > possible to manage multiple copies of a daemon using the normal
  > infrastructure
  > by symlinking rc.d scripts to a new name. ok jung@ ajacoutot@, smtpd ok
  > gilles@ (sthen@)

  ~ vmd.c                                 

  > Fix a possible use-after-free in vmd, forward the result to the
  > control socket before free'ing the vm.
  > Found by and OK jsg@ (reyk@)

  ~ virtio.c                              

  > avoid a double free in an error path
  > ok stefan@ (jsg@)

===============================================================================
_______________________________________________
owc mailing list
[email protected]
http://www.squish.net/mailman/listinfo/owc

Reply via email to