[owc] Weekly src changes ending 2006-07-30
OWC
auto at squish.net
Mon Jul 31 08:00:35 BST 2006
OpenBSD src changes summary for 2006-07-23 to 2006-07-30 inclusive
==================================================================
bin/ksh distrib/miniroot
distrib/notes distrib/sets
etc/Makefile etc/chio.conf
etc/etc.alpha/Makefile.inc etc/etc.amd64/Makefile.inc
etc/etc.armish/MAKEDEV.md etc/etc.armish/Makefile.inc
etc/etc.aviion/Makefile.inc etc/etc.cats/MAKEDEV.md
etc/etc.cats/Makefile.inc etc/etc.hp300/Makefile.inc
etc/etc.hppa/Makefile.inc etc/etc.hppa64/Makefile.inc
etc/etc.i386/Makefile.inc etc/etc.luna88k/Makefile.inc
etc/etc.mac68k/Makefile.inc etc/etc.macppc/Makefile.inc
etc/etc.mvme68k/Makefile.inc etc/etc.mvme88k/Makefile.inc
etc/etc.mvmeppc/Makefile.inc etc/etc.sgi/Makefile.inc
etc/etc.sparc/Makefile.inc etc/etc.sparc64/Makefile.inc
etc/etc.vax/MAKEDEV etc/etc.vax/MAKEDEV.md
etc/etc.vax/Makefile.inc etc/etc.zaurus/MAKEDEV.md
etc/etc.zaurus/Makefile.inc etc/pf.os
etc/ppp/ppp.conf.sample etc/rc.local
etc/root/root.mail gnu/usr.bin/binutils
include/arpa/tftp.h lib/libc
libexec/ld.so libexec/tftpd
regress/usr.bin sbin/brconfig
sbin/fdisk sbin/ifconfig
sbin/init sbin/isakmpd
sbin/mount_nfs sbin/scsi
sbin/wsconsctl share/man
share/mk share/tmac
sys/arch sys/arch/alpha/conf
sys/arch/amd64/conf sys/arch/arm/arm
sys/arch/armish/armish sys/arch/armish/conf
sys/arch/armish/include sys/arch/armish/stand
sys/arch/armish/stand/boot sys/arch/aviion/conf
sys/arch/cats/conf sys/arch/hp300/conf
sys/arch/hppa/conf sys/arch/hppa64/conf
sys/arch/i386/conf sys/arch/i386/i386
sys/arch/i386/include sys/arch/i386/isa
sys/arch/luna88k/conf sys/arch/mac68k/conf
sys/arch/macppc/conf sys/arch/macppc/stand/tbxidata
sys/arch/mips64/include sys/arch/mvme68k/conf
sys/arch/mvme88k/conf sys/arch/mvme88k/dev
sys/arch/mvmeppc/conf sys/arch/sgi/conf
sys/arch/solbourne/conf sys/arch/sparc/conf
sys/arch/sparc/dev sys/arch/sparc/sparc
sys/arch/sparc64/conf sys/arch/sparc64/stand/ofwboot
sys/arch/vax/conf sys/arch/vax/dec
sys/arch/vax/if sys/arch/vax/include
sys/arch/vax/qbus sys/arch/vax/uba
sys/arch/vax/vax sys/arch/vax/vsa
sys/arch/zaurus/conf sys/conf
sys/dev sys/dev/acpi
sys/dev/ic sys/dev/isa
sys/dev/mii sys/dev/pci
sys/dev/pcmcia sys/dev/sbus
sys/dev/sdmmc sys/dev/usb
sys/dev/wscons sys/dev/wsfont
sys/gnu sys/kern
sys/net80211 sys/scsi
sys/sys sys/ufs/ffs
sys/uvm usr.bin/cvs
usr.bin/mg usr.bin/rcs
usr.bin/rev usr.bin/ssh
usr.bin/tftp usr.bin/xargs
usr.sbin/bgpctl usr.sbin/bgpd
usr.sbin/dvmrpctl usr.sbin/gpioctl
usr.sbin/hotplugd usr.sbin/httpd
usr.sbin/ifstated usr.sbin/ospfctl
usr.sbin/wsconscfg
== bin =============================================================== 01/13 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/bin
ksh
~ sh.1 ~ ksh.1
> clarifications for OPTIND:
> hunk #1 from tamas tevesz;
> hunk #2 from otto;
> ok otto (jmc@)
== distrib =========================================================== 02/13 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/distrib
miniroot
~ install.sub
> crank to 4.0-beta (deraadt@)
notes
~ vax/hardware ~ vax/prep
> Update supported devices list and add a few words about console selection.
> (miod@)
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/comp/mi
> sync (deraadt@)
~ lists/man/mi
> sync (deraadt@)
~ lists/base/md.vax ~ lists/etc/md.vax
> sync (deraadt@)
~ lists/etc/md.vax
> 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/md.alpha ~ lists/comp/md.amd64
~ lists/comp/md.armish ~ lists/comp/md.aviion
~ lists/comp/md.cats ~ lists/comp/md.hp300
~ lists/comp/md.hppa ~ lists/comp/md.hppa64
~ lists/comp/md.i386 ~ lists/comp/md.luna88k
~ lists/comp/md.mac68k ~ lists/comp/md.macppc
~ lists/comp/md.mvme68k ~ lists/comp/md.mvme88k
~ lists/comp/md.mvmeppc ~ lists/comp/md.sgi
~ lists/comp/md.sparc ~ lists/comp/md.sparc64
~ lists/comp/md.vax ~ lists/comp/md.zaurus
> 3.9 -> 4.0 sync (deraadt@)
~ lists/base/md.armish
> sync (deraadt@)
~ lists/man/mi
> sync (deraadt@)
~ lists/base/mi ~ lists/man/mi
> sync (deraadt@)
== etc =============================================================== 03/13 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/etc
Makefile
~ Makefile
> Build wscons stuff on vax (but no for bsd.rd yet) (miod@)
~ Makefile
> remove "snapshot" code, now that our snapshots are always actually
> releases... (deraadt@)
~ Makefile
> remove unused WCS stuff (deraadt@)
chio.conf
~ chio.conf
> spacing (david@)
etc.alpha/Makefile.inc
~ etc.alpha/Makefile.inc
> unify these damnit! (deraadt@)
~ etc.alpha/Makefile.inc
> typo (deraadt@)
etc.amd64/Makefile.inc
~ etc.amd64/Makefile.inc
> unify these damnit! (deraadt@)
etc.armish/MAKEDEV.md
~ etc.armish/MAKEDEV.md
> Use the correct major for /dev/pci (instead of giving it the same major
> as /dev/raid).
> ok drahn@ (kettenis@)
etc.armish/Makefile.inc
~ etc.armish/Makefile.inc
> unify these damnit! (deraadt@)
etc.aviion/Makefile.inc
~ etc.aviion/Makefile.inc
> unify these damnit! (deraadt@)
etc.cats/MAKEDEV.md
~ etc.cats/MAKEDEV.md
> Use the correct major for /dev/pci (instead of giving it the same major
> as /dev/raid).
> ok drahn@ (kettenis@)
etc.cats/Makefile.inc
~ etc.cats/Makefile.inc
> unify these damnit! (deraadt@)
etc.hp300/Makefile.inc
~ etc.hp300/Makefile.inc
> unify these damnit! (deraadt@)
etc.hppa/Makefile.inc
~ etc.hppa/Makefile.inc
> unify these damnit! (deraadt@)
etc.hppa64/Makefile.inc
~ etc.hppa64/Makefile.inc
> unify these damnit! (deraadt@)
etc.i386/Makefile.inc
~ etc.i386/Makefile.inc
> rearrange to be more like the others (deraadt@)
etc.luna88k/Makefile.inc
~ etc.luna88k/Makefile.inc
> unify these damnit! (deraadt@)
etc.mac68k/Makefile.inc
~ etc.mac68k/Makefile.inc
> unify these damnit! (deraadt@)
etc.macppc/Makefile.inc
~ etc.macppc/Makefile.inc
> unify these damnit! (deraadt@)
etc.mvme68k/Makefile.inc
~ etc.mvme68k/Makefile.inc
> unify these damnit! (deraadt@)
etc.mvme88k/Makefile.inc
~ etc.mvme88k/Makefile.inc
> unify these damnit! (deraadt@)
etc.mvmeppc/Makefile.inc
~ etc.mvmeppc/Makefile.inc
> unify (deraadt@)
~ etc.mvmeppc/Makefile.inc
> sorry, but proper boot cd building should be done in distrib/mvmeppc
> when the time comes (not here!) (deraadt@)
etc.sgi/Makefile.inc
~ etc.sgi/Makefile.inc
> do a make depend phase (deraadt@)
~ etc.sgi/Makefile.inc
> unify these damnit! (deraadt@)
etc.sparc/Makefile.inc
~ etc.sparc/Makefile.inc
> make these more similar (few steps to that) (deraadt@)
etc.sparc64/Makefile.inc
~ etc.sparc64/Makefile.inc
> make these more similar (few steps to that) (deraadt@)
etc.vax/MAKEDEV
~ etc.vax/MAKEDEV
> regen (miod@)
~ etc.vax/MAKEDEV
> Regen (miod@)
etc.vax/MAKEDEV.md
~ etc.vax/MAKEDEV.md
> wsdisplay nodes (miod@)
~ etc.vax/MAKEDEV.md
> wskbd and wsmouse nodes. (miod@)
etc.vax/Makefile.inc
~ etc.vax/Makefile.inc
> make these more similar (few steps to that) (deraadt@)
etc.zaurus/MAKEDEV.md
~ etc.zaurus/MAKEDEV.md
> Remove /dev/iop and /dev/pci.
> ok drahn@ (kettenis@)
etc.zaurus/Makefile.inc
~ etc.zaurus/Makefile.inc
> make these more similar (few steps to that) (deraadt@)
pf.os
~ pf.os
> crank the OpenBSD version to 4.0; Opera 9.0 on OpenBSD verified by sturm@
> (david@)
ppp/ppp.conf.sample
~ ppp/ppp.conf.sample
> spacing (david@)
rc.local
~ rc.local
> remove port startup code, it's all moved to the respective port
> ok deraadt, toby, pedro, david (sturm@)
root/root.mail
~ root/root.mail
> crank to 4.0-beta (deraadt@)
== gnu =============================================================== 04/13 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/gnu
usr.bin/binutils
~ gdb/Makefile.in
> Do not build & install gdbtui. (kettenis@)
== include =========================================================== 05/13 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/include
arpa/tftp.h
~ arpa/tftp.h
> Fixing several timeout quirks at tftpd and tftp:
> - move TIMEOUT* defines to arpa/tftp.h, as they are used several times
> in tftpd and tftp, and the values are part of the RFC definition.
> - tftpd and tftp did count the total retransmission time in retries
> instead in seconds. fixed.
> - tftpd rexmt timeout was hardcoded by a define and therefore didn't
> changed when the timeout option was sent. fixed.
> - limit total retransmission timeout in tftp to also 255 seconds.
> - replace obvious atoi()'s by strtonum().
> ok claudio@ (mglocker@)
~ arpa/tftp.h
> TIMEOUT* values are not part of the protocol. tftp.h is a namespace
> export of the protocol. you shall not add non-protocol stuff to such
> a file, period. (deraadt@)
== lib =============================================================== 06/13 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/lib
libc
~ time/ctime.3
> remove tzset(3) from SYNOPSIS: it has its own man page;
> spotted by stevesk; ok otto (jmc@)
~ crypt/blowfish.3
> updates from nicholas marriott;
> re-worded and ok djm (jmc@)
~ crypt/blowfish.3
> document `iv';
> from eren elci; tweaked by djm
> ok djm (jmc@)
~ crypt/blowfish.3
> better macros; (jmc@)
~ crypt/blowfish.3
> wording/grammar tweaks; (jmc@)
== libexec =========================================================== 07/13 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/libexec
ld.so
~ ldconfig/prebind_delete.c
> Only stop processing prebind deletion if an actual error occurs. (drahn@)
tftpd
~ tftpd.c
> Fixing several timeout quirks at tftpd and tftp:
> - move TIMEOUT* defines to arpa/tftp.h, as they are used several times
> in tftpd and tftp, and the values are part of the RFC definition.
> - tftpd and tftp did count the total retransmission time in retries
> instead in seconds. fixed.
> - tftpd rexmt timeout was hardcoded by a define and therefore didn't
> changed when the timeout option was sent. fixed.
> - limit total retransmission timeout in tftp to also 255 seconds.
> - replace obvious atoi()'s by strtonum().
> ok claudio@ (mglocker@)
~ tftpd.c
> Mini KNF. No binary changes. (mglocker@)
~ tftpd.c
> TIMEOUT* values are not part of the protocol. tftp.h is a namespace
> export of the protocol. you shall not add non-protocol stuff to such
> a file, period. (deraadt@)
~ tftpd.c
> Make tftpd atoi() free.
> ok claudio@ (mglocker@)
== regress =========================================================== 08/13 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/regress
usr.bin
~ rcs/Makefile ~ rcs/keywords.in
~ rcs/keywords.out
> Remove $Log$ check from keyword expansion test, which is the only
> keyword we don't do yet. Now we can enable the test.
> OK niallo@ (ray@)
~ rcs/Makefile
> Add regression test for files with lots of keywords.
> OK xsa@ (ray@)
== sbin ============================================================== 09/13 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/sbin
brconfig
~ brconfig.c
> make "brconfig" with no arguments equivalent to "brconfig -a",
> following the similar change to ifconfig; 'nice' deraadt@ (djm@)
~ brconfig.c
> better yet, make this use getopt for argument processing;
> "looks good" kettenis@ (djm@)
~ brconfig.8
> -a flag is now optional; prompted by steven@ (djm@)
~ brconfig.8
> tweak; (jmc@)
fdisk
~ cmd.c ~ fdisk.c
~ user.c
> Add checks to each write(2) attempt and try to keep fdisk in a
> consistent state upon write failure.
> OK krw@ and deraadt at .ay@)
~ fdisk.c
> Change some obvious atoi() to strtonum().
> OK tedu@ (ray@)
ifconfig
~ ifconfig.c
> ifconfig output contained trailing whites in inet and groups: lines
> fix that by rearranging spaces in printf format strings
> ok claudio@ mpf@ mickey@ (grunk@)
init
~ init.8
> correct .Xr; from marcus popp (jmc@)
isakmpd
~ dpd.c ~ math_group.c
~ monitor.c ~ util.c
~ vendor.c
> Style; return is not a function. hshoexer@ ok. (ho@)
mount_nfs
~ mount_nfs.c
> permit synchronous mounts
> ok pedro (sturm@)
scsi
~ scsi.c
> remove some scsi ioctls, most importantly the ones that can reprobe the
> bus. supporting these complicates the midlayer unnecessarily.
> ok krw@ (dlg@)
~ scsi.8
> update to reflect reality (dlg@)
~ scsi.8
> userland shouldnt care about kernel issues (dlg@)
wsconsctl
~ util.c
> Driver for the VS4000/90 frame buffer, adapted from NetBSD (ragge) and
> Blaz Antonic's work, no acceleration yet, untested due to the lack of
> hardware, but enough sacrifices were made to the RAMDAC Gods. (miod@)
~ Makefile
> Build wscons stuff on vax (but no for bsd.rd yet) (miod@)
== share ============================================================= 10/13 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/share
man
~ man4/nviic.4 ~ man4/nfe.4
> add newer NVidia chipsets. (brad@)
~ man4/pci.4
> update nviic(4) entry. (brad@)
~ man8/man8.vax/MAKEDEV.8
> regen (miod@)
~ man4/pcfadc.4
> add HISTORY; OK jmc at .sa@)
~ man5/pf.conf.5
> document "tos": pointed out by maxim bourmistrov
> diff from jared r r spiegel
> ok dhartmei (jmc@)
~ man4/man4.i386/npx.4
> Remove text about GPL_MATH_EMULATE and floating-point emulator. (kettenis@)
~ man8/man8.alpha/boot_alpha.8
> Alpha boot has defaulted to multiuser for a long time, so better stop
> lying now.
> ok martin (mk@)
~ man3/dlfcn.3
> Sync with current behavior:
> - clarify how paths and parital shared library versions are treated
> with dlopen()
> - dlopen() mode parameter is supported now - describe how to use it.
> - describe how visiblity effects dlsym() with special handles
> - remove bugs section
> - formatting improvements
> With input from drahn@ and jmc at .ay drahn@, that's fine jmc@ (kurt@)
~ man4/opl.4 ~ man4/cmpci.4
> update cmpci(4) man page a bit. (brad@)
~ man4/pci.4
> honestly, puc(4) belongs in the "serial port" block not "misc", because
> most people when they want serial should see it (even if maybe 0.1% of
> our users use a parallel port puc) (deraadt@)
~ man3/dlfcn.3
> Explain the effect of the --export-dynamic link option on the visibility
> of the executable's symbols as well as improve the visibility paragraph a
> bit. okay drahn@ (kurt@)
~ man4/wsdisplay.4 ~ man4/man4.vax/Makefile
~ man4/man4.vax/intro.4 ~ man4/man4.vax/smg.4
~ man4/man4.vax/vsbus.4 + man4/man4.vax/gpx.4
+ man4/man4.vax/lcg.4 + man4/man4.vax/lcspx.4
> Document the recent VAX frame buffer changes, especially the way smg(4)
> attaches (or refuses to) now. (miod@)
~ man4/Makefile ~ man4/pci.4
+ man4/arc.4
> document the existence of a driver for areca raid controllers. (dlg@)
~ man8/man8.vax/MAKEDEV.8
> Regen (miod@)
mk
~ sys.mk
> crank to 4.0-beta (deraadt@)
tmac
~ mdoc/doc-common
> crank to 4.0-beta (deraadt@)
== sys =============================================================== 11/13 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/sys
arch
~ i386/Makefile
> remove gnu stuff from the 'tags' target.
> ok miod@ (thib@)
arch/alpha/conf
~ GENERIC
> + opl* at cmpci? (brad@)
~ Makefile.alpha
> Compile all kernels with -Wstack-usage-larger-than-2047, now that all
> offending code has been taken out and shot. ok deraadt@ (miod@)
arch/amd64/conf
~ GENERIC
> enable cmpci(4) and enable midi at oplrad@)
~ Makefile.amd64
> Compile all kernels with -Wstack-usage-larger-than-2047, now that all
> offending code has been taken out and shot. ok deraadt@ (miod@)
arch/arm/arm
~ conf.c
> Make option USER_PCICONF work.
> ok drahn@ (kettenis@)
arch/armish/armish
~ armish_machdep.c ~ autoconf.c
> With new kernel and bootloader, argument passing now works, version 1.0.
> (and yes this does work on iodata and thecus) (drahn@)
~ autoconf.c
> spacing (deraadt@)
arch/armish/conf
~ Makefile.armish
> Compile all kernels with -Wstack-usage-larger-than-2047, now that all
> offending code has been taken out and shot. ok deraadt@ (miod@)
arch/armish/include
~ bootconfig.h
> With new kernel and bootloader, argument passing now works, version 1.0.
> (and yes this does work on iodata and thecus) (drahn@)
arch/armish/stand
~ Makefile
> enter boot (deraadt@)
+ Makefile.inc
> should install into /usr/mdec/ (deraadt@)
arch/armish/stand/boot
+ Makefile + boot.c
+ boot.h + clock.c
+ conf.c + devopen.c
+ exec.c + ldscript
+ ns16550.c + pciide.c
+ start.S + wd.c
+ wdc.c + wdvar.h
> First stab at an armish bootloader. It will load /bsd from the drive in
> the
> upper slot of the Thecus n2100. Glued together from bits from the NetBSD
> cobalt and evbarm ports. (kettenis@)
~ Makefile
> no man page for now (deraadt@)
~ exec.c
> Pass the end of the symbol table to the kernel. (kettenis@)
~ Makefile
> only enter for armish (deraadt@)
- boot.c - boot.h
~ Makefile ~ clock.c
~ conf.c ~ devopen.c
~ ns16550.c ~ pciide.c
~ wd.c ~ wdc.c
~ wdvar.h + dev_armish.c
+ libsa.h + machdep.c
+ time.c
> ework armish bootloader into an interactive one, much like i386 has.
> Make sure disk names match those used by the kernel. This means that now
> it boots from wd0 by default, whatever slot it is in. (kettenis@)
~ machdep.c
> Make _rtt reset the machine. (kettenis@)
~ machdep.c ~ pciide.c
~ wd.c
> Perform a limited PCI probe to locate the wdc on iodata/thecus, boot loader
> will not run on those systems. (drahn@)
~ conf.c ~ exec.c
> With new kernel and bootloader, argument passing now works, version 1.0.
> (and yes this does work on iodata and thecus) (drahn@)
arch/aviion/conf
~ Makefile.aviion
> Compile all kernels with -Wstack-usage-larger-than-2047, now that all
> offending code has been taken out and shot. ok deraadt@ (miod@)
arch/cats/conf
~ Makefile.cats
> Compile all kernels with -Wstack-usage-larger-than-2047, now that all
> offending code has been taken out and shot. ok deraadt@ (miod@)
arch/hp300/conf
~ Makefile.hp300
> Compile all kernels with -Wstack-usage-larger-than-2047, now that all
> offending code has been taken out and shot. ok deraadt@ (miod@)
arch/hppa/conf
~ Makefile.hppa
> Compile all kernels with -Wstack-usage-larger-than-2047, now that all
> offending code has been taken out and shot. ok deraadt@ (miod@)
arch/hppa64/conf
~ Makefile.hppa64
> Compile all kernels with -Wstack-usage-larger-than-2047, now that all
> offending code has been taken out and shot. ok deraadt@ (miod@)
arch/i386/conf
~ GENERIC ~ RAMDISK
~ RAMDISKB ~ RAMDISKC
~ RAMDISK_CD ~ files.i386
> Kill option GPL_MATH_EMULATE.
> ok deraadt@ and many others. (kettenis@)
~ GENERIC
> Newer cmcpci(4) sound driver, which adds S/PDIF, OPL and MPU support, some
> fixes for 48KHz sound and adds the C-Media CMI8738B PCI id.
> By ITOH Yasufumi
> From NetBSD
> Thanks to kurt@ for providing me with a CMI8738 board. (brad@)
~ GENERIC
> disable shit that will break the tree soon (deraadt@)
~ GENERIC
> ok, opl/cmpci can go in now, got the workaround for upcoming issues
> (deraadt@)
~ Makefile.i386
> Compile all kernels with -Wstack-usage-larger-than-2047, now that all
> offending code has been taken out and shot. ok deraadt@ (miod@)
arch/i386/i386
~ trap.c
> Kill option GPL_MATH_EMULATE.
> ok deraadt@ and many others. (kettenis@)
arch/i386/include
~ cpu.h ~ npx.h
> Kill option GPL_MATH_EMULATE.
> ok deraadt@ and many others. (kettenis@)
arch/i386/isa
~ npx.c
> Kill option GPL_MATH_EMULATE.
> ok deraadt@ and many others. (kettenis@)
arch/luna88k/conf
~ Makefile.luna88k
> Compile all kernels with -Wstack-usage-larger-than-2047, now that all
> offending code has been taken out and shot. ok deraadt@ (miod@)
arch/mac68k/conf
~ Makefile.mac68k
> Compile all kernels with -Wstack-usage-larger-than-2047, now that all
> offending code has been taken out and shot. ok deraadt@ (miod@)
arch/macppc/conf
~ Makefile.macppc
> Compile all kernels with -Wstack-usage-larger-than-2047, now that all
> offending code has been taken out and shot. ok deraadt@ (miod@)
arch/macppc/stand/tbxidata
~ bsd.tbxi
> crank to 4.0-beta (deraadt@)
arch/mips64/include
~ archtype.h
> fix typo in comment
> ok miod (jasper@)
arch/mvme68k/conf
~ Makefile.mvme68k
> Compile all kernels with -Wstack-usage-larger-than-2047, now that all
> offending code has been taken out and shot. ok deraadt@ (miod@)
arch/mvme88k/conf
~ Makefile.mvme88k
> Compile all kernels with -Wstack-usage-larger-than-2047, now that all
> offending code has been taken out and shot. ok deraadt@ (miod@)
arch/mvme88k/dev
~ vx.c
> 4KB buffer on the stack, and would you like fries with that? (miod@)
arch/mvmeppc/conf
~ Makefile.mvmeppc
> Compile all kernels with -Wstack-usage-larger-than-2047, now that all
> offending code has been taken out and shot. ok deraadt@ (miod@)
arch/sgi/conf
~ Makefile.sgi
> Compile all kernels with -Wstack-usage-larger-than-2047, now that all
> offending code has been taken out and shot. ok deraadt@ (miod@)
arch/solbourne/conf
~ Makefile.solbourne
> Compile all kernels with -Wstack-usage-larger-than-2047, now that all
> offending code has been taken out and shot. ok deraadt@ (miod@)
arch/sparc/conf
~ Makefile.sparc
> Compile all kernels with -Wstack-usage-larger-than-2047, now that all
> offending code has been taken out and shot. ok deraadt@ (miod@)
arch/sparc/dev
~ cgeight.c ~ cgfour.c
~ cgsix.c ~ cgthree.c
~ tcx.c
> Do not bother reading the existing colormap on attach, since we will
> override it with the rasops one. (miod@)
arch/sparc/sparc
~ autoconf.c
> move large local array out of a "called once, only at boot" function
> into being a global, to reduce stack size; ok miod (deraadt@)
arch/sparc64/conf
~ Makefile.sparc64
> Compile all kernels with -Wstack-usage-larger-than-2047, now that all
> offending code has been taken out and shot. ok deraadt@ (miod@)
arch/sparc64/stand/ofwboot
~ boot.c ~ ofdev.c
> remove weird floppyboot garbage; ok miod (deraadt@)
arch/vax/conf
~ files.vax
> Make smg(4) on par with the current state of wsdisplay drivers: make it use
> a rasops backend and wsfont, better ioctl conformance, and too many other
> changes to mention. (miod@)
~ GENERIC ~ files.vax
> Add a wsdisplay driver for the frame buffer found on VS4000/60 and VLC
> systems, based on Blaz Antonic's work and adjusted to fit modern wscons
> requirements; currently running as a dumb frame buffer, acceleration code
> going in soon. (miod@)
~ GENERIC ~ files.vax
> Driver for the VS4000/90 frame buffer, adapted from NetBSD (ragge) and
> Blaz Antonic's work, no acceleration yet, untested due to the lack of
> hardware, but enough sacrifices were made to the RAMDAC Gods. (miod@)
~ Makefile.vax
> Compile all kernels with -Wstack-usage-larger-than-2047, now that all
> offending code has been taken out and shot. ok deraadt@ (miod@)
~ GENERIC
> Driver for the color frame buffer option on VAXstation 3100 models 3x/4x.
> Still a few rough edges to polish, but nevertheless usable; 4bpp flavour
> untested. (miod@)
~ files.vax
> Rewrite glass console detection and initialization code, it's simpler and
> will not allocate frame buffer resources at probe time. A few buglets fixed
> while doing so. (miod@)
~ GENERIC
> options SCSIVERBOSE is long dead (miod@)
~ RAMDISK
> Add wscons stuff to the installation media as well. (miod@)
arch/vax/dec
~ dzkbd.c ~ dzkbdvar.h
~ dzms.c ~ lk201.h
~ lk201_ws.c ~ lk201var.h
~ wskbdmap_lk201.c ~ wskbdmap_lk201.h
> Sync dz and lk{kbd,ms} code with NetBSD, this allows keyboards to work when
> initialized by the PROM. (miod@)
~ lk201_ws.c
> Disable keyclick by default. (miod@)
~ dzkbd.c ~ lk201_ws.c
~ lk201var.h
> Always invoke the lk201 input handler, even when the wskbd device is
> disabled,
> so that the keyboard gets configured correctly if plugged late. (miod@)
arch/vax/if
- if_zereg.h
> More files which have changed name or location since some time already and
> are only confusing things. (miod@)
arch/vax/include
~ pte.h
> Assembly versions of kvtopte() and kvtophys(), better than what gcc
> outputs; from NetBSD (miod@)
~ ka410.h ~ ka420.h
~ ka43.h
> Remove cursor register address constants, code should (and does) use
> dc503reg.h
> instead. (miod@)
arch/vax/qbus
~ dz.c ~ dz_uba.c
~ dzreg.h
> Sync dz and lk{kbd,ms} code with NetBSD, this allows keyboards to work when
> initialized by the PROM. (miod@)
arch/vax/uba
- uba.c - uba_mainbus.c
- ubavar.h - uda.c
> More files which have changed name or location since some time already and
> are only confusing things. (miod@)
arch/vax/vax
~ autoconf.c
> Do not print an odd message on machines where led0 is configured but did
> not attach. (miod@)
~ sbi.c
> Recent change to the way mainbus children attach requires sbi_match to be
> split between mainbus and abus attachments. Untested (and this is not part
> of the supported hardware at the moment), but at least this will allow
> sbi to attach to abus (this was broken, even before the mainbus attachment
> changes). (miod@)
~ vm_machdep.c
> Simplify ioaccess() by computing the pte template outside the loop, and
> modify iounaccess() loop style accordingly for consistency.
> No functional change. (miod@)
~ conf.c
> Add a wsdisplay driver for the frame buffer found on VS4000/60 and VLC
> systems, based on Blaz Antonic's work and adjusted to fit modern wscons
> requirements; currently running as a dumb frame buffer, acceleration code
> going in soon. (miod@)
~ conf.c
> Driver for the VS4000/90 frame buffer, adapted from NetBSD (ragge) and
> Blaz Antonic's work, no acceleration yet, untested due to the lack of
> hardware, but enough sacrifices were made to the RAMDAC Gods. (miod@)
~ conf.c
> Driver for the color frame buffer option on VAXstation 3100 models 3x/4x.
> Still a few rough edges to polish, but nevertheless usable; 4bpp flavour
> untested. (miod@)
~ led.c
> Use a less bright pattern. (miod@)
~ conf.c + wscons_machdep.c
> Rewrite glass console detection and initialization code, it's simpler and
> will not allocate frame buffer resources at probe time. A few buglets fixed
> while doing so. (miod@)
~ conf.c
> wskbd and wsmouse nodes. (miod@)
~ locore.c
> Report MicroVAX 3100 m{10,20}{,e} as such instead of bare ``MicroVAX
> 3100'';
> ok hugh@ (miod@)
arch/vax/vsa
~ smg.c
> Make smg(4) on par with the current state of wsdisplay drivers: make it use
> a rasops backend and wsfont, better ioctl conformance, and too many other
> changes to mention. (miod@)
+ lcg.c + lcgreg.h
> Add a wsdisplay driver for the frame buffer found on VS4000/60 and VLC
> systems, based on Blaz Antonic's work and adjusted to fit modern wscons
> requirements; currently running as a dumb frame buffer, acceleration code
> going in soon. (miod@)
~ lcg.c
> Make sure the mapped frame buffer area is a multiple of the logical page
> size, for mmap to be happy. (miod@)
+ lcspx.c
> Driver for the VS4000/90 frame buffer, adapted from NetBSD (ragge) and
> Blaz Antonic's work, no acceleration yet, untested due to the lack of
> hardware, but enough sacrifices were made to the RAMDAC Gods. (miod@)
- ncr.h
> More files which have changed name or location since some time already and
> are only confusing things. (miod@)
~ smg.c
> Better dmesg (miod@)
+ gpx.c
> Driver for the color frame buffer option on VAXstation 3100 models 3x/4x.
> Still a few rough edges to polish, but nevertheless usable; 4bpp flavour
> untested. (miod@)
~ dz_ibus.c
> Make sure we never attach a dz console to a keyboard port on VAXstations,
> regardless of the L3 switch (so that machines with unsupported frame
> buffers
> can work with serial console). (miod@)
~ gpx.c ~ lcg.c
~ lcspx.c ~ smg.c
> Rewrite glass console detection and initialization code, it's simpler and
> will not allocate frame buffer resources at probe time. A few buglets fixed
> while doing so. (miod@)
~ gpx.c ~ lcg.c
~ lcspx.c ~ smg.c
> At attach() time, do not simply trust the L3 switch to know if we are the
> console, check for cn_tab pointing to wsdisplay as well. (miod@)
~ dz_ibus.c
> Better ``no dz console to a keyboard connector'' fix which does not
> interfere
> with the dz console priority. (miod@)
~ dz_ibus.c
> Attach keyboard and mouse on VAXstations. Non-keyboard capable machines are
> not affected. (miod@)
~ gpx.c
> Force gpx to attach if it is the console, the interrupt test is unreliable
> after the dragon has been initialized to our needs. (miod@)
~ gpx.c
> Fix depth detection when gpx is the console. (miod@)
~ gpx.c
> And correctly map the ramdac when console, too. Doh. Somehow I was
> expecting
> the vax pagesize to be 256 bytes... (miod@)
~ dz_ibus.c
> Allow keyboard attachment on VS 4k/90 as well, spotted by deraadt@ (miod@)
~ dz_ibus.c
> Use DZ_CSR_xxx symbolic constants. (miod@)
arch/zaurus/conf
~ Makefile.zaurus
> Compile all kernels with -Wstack-usage-larger-than-2047, now that all
> offending code has been taken out and shot. ok deraadt@ (miod@)
conf
~ newvers.sh
> crank to 4.0-beta (deraadt@)
dev
~ ipmi.c
> Fix BT interface by waiting for BMC_BUSY on writes.
> Found and ok marco (jordan@)
~ ipmi.c ~ ipmivar.h
> Only read 1 sensor at a time instead of all of them. This should help
> systems that have a burst of activity every 10 seconds.
> Also lower the polling frequency from 10 to 5 seconds.
> ok jordan@ (marco@)
~ ipmi.c
> KNF (marco@)
dev/acpi
~ acpitz.c
> Donot pass a uninitialized pointer to aml_register_notify(), should fix
> PR5153
> awaiting confirmation of this from the submitter, but its wrong regardless.
> ok marco@, canacar@ (gwk@)
dev/ic
+ dc503reg.h
> Make smg(4) on par with the current state of wsdisplay drivers: make it use
> a rasops backend and wsfont, better ioctl conformance, and too many other
> changes to mention. (miod@)
~ opl.c ~ oplvar.h
> Rework opl_find() so that it does not need to take a proto softc, shaves
> more than 2.5KB of stack for its callers (and even more on 64 bit arches).
> (miod@)
~ mfi.c
> make dmesg printing look a little bit nicer. (brad@)
~ ne2000.c
> Do not redefine ALIGNED_POINTER, it comes from <machine/param.h> (miod@)
~ aic79xx.c
> Be consistant and always check for non-NULL pointer before assigning into
> it. (krw@)
~ aic79xx.c ~ aic7xxx.c
> Make [ahc|ahd]_print_register() consistant, since they are both called
> by the same aicasm-generated code. ahc_print_register() now just
> tracks the return values of printf() as ahd_print_register() was
> doing.
> Eliminates a 1024 character buffer on the stack that
> ahc_print_register() was using to keep track of the number of
> characters printed. (krw@)
dev/isa
~ opl_ess.c ~ opl_sb.c
> Rework opl_find() so that it does not need to take a proto softc, shaves
> more than 2.5KB of stack for its callers (and even more on 64 bit arches).
> (miod@)
~ if_ec.c ~ if_we.c
> Do not redefine ALIGNED_POINTER, it comes from <machine/param.h> (miod@)
dev/mii
~ brgphy.c
> revert back to the old brgphy_mii_phy_auto() which simplifies the autoneg
> code. (brad@)
~ rgephy.c ~ rlphy.c
> add sys/timeout.h (brad@)
~ miidevs.h
> regen (brad@)
~ miidevs
> another Marvell Gig PHY. (brad@)
~ eephy.c
> another Marvell Gig PHY. (brad@)
~ eephy.c
> Simplify autoneg code a bit. Tested on a few sk(4)'s and an axe(4). (brad@)
dev/pci
~ pcidevs
> add PCI ids for the NVidia MCP61/65 chipsets and additional ids for the
> MCP51/55 chipsets.
> From The Linux PCI ID Repository (brad@)
~ pcidevs.h ~ pcidevs_data.h
> regen (brad@)
~ pcidevs
> add two more MCP61 Ethernet MAC PCI ids.
> From Linux's pci_ids.h (brad@)
~ pcidevs.h ~ pcidevs_data.h
> regen (brad@)
~ pcidevs.h ~ pcidevs_data.h
> regen (brad@)
~ pcidevs
> more consistent naming for some of the NVidia entries. (brad@)
~ pcidevs.h ~ pcidevs_data.h
> regen (brad@)
~ pcidevs
> add NVidia MCP61 IDE and SMBus PCI ids. (brad@)
~ pciide.c ~ nviic.c
~ if_nfe.c
> add NVidia MCP61/65 ids. (brad@)
~ if_sk.c
> improved Yukon init routine with proper workaround for the PHY COMA mode.
> From yongari at FreeBSDrad@)
~ azalia_codec.c
> Make the default value of outputs.linein.dir "output" for the ThinkPad T60.
> From kent at NetBSDrad@)
~ files.pci ~ cmpci.c
~ cmpcireg.h ~ cmpcivar.h
+ opl_cmpci.c
> Newer cmcpci(4) sound driver, which adds S/PDIF, OPL and MPU support, some
> fixes for 48KHz sound and adds the C-Media CMI8738B PCI id.
> By ITOH Yasufumi
> From NetBSD
> Thanks to kurt@ for providing me with a CMI8738 board. (brad@)
~ opl_cmpci.c ~ opl_yds.c
> Rework opl_find() so that it does not need to take a proto softc, shaves
> more than 2.5KB of stack for its callers (and even more on 64 bit arches).
> (miod@)
~ opl_cmpci.c
> typo (deraadt@)
~ pcidevs
> oops (mickey@)
~ pcidevs
> new twe id (mickey@)
~ pcidevs.h ~ pcidevs_data.h
~ twe_pci.c
> new twe; reported by Randal Harisch <azar at eyeonvideo.commickey@)
~ if_vge.c
> do not bother playing tag with the PCI_COMMAND_STATUS_REG
> tested by jolan@, ok brad@ (kettenis@)
~ pcidevs
> add some Broadcom, NVIDIA and Ricoh PCI ids. (brad@)
~ pcidevs.h ~ pcidevs_data.h
> regen (brad@)
~ pcidevs
> add Intel IOP333, 5000 and 6321ESB PCI ids. (brad@)
~ pcidevs.h ~ pcidevs_data.h
> regen (brad@)
~ pcidevs
> more areca raid controller ids (dlg@)
~ pcidevs.h ~ pcidevs_data.h
> regen (dlg@)
~ files.pci + arc.c
> here's the real reason i removed the if_arc network goo: i wanted its name.
> this is the start of a driver supporting the areca raid controllers. so far
> it probes, attaches, establishes the interrupt, allocates all the necessary
> memory and queries the firmware for operating params. the io path is yet to
> be done. that is going to be hard to do since i dont have any sata disks to
> plug into it.
> thanks go to Billion Wu and Erich Chen from Areca for their assistance.
> ok deraadt@ (dlg@)
~ arc.c
> not keeping the dma tag around means we panic when we try to use it during
> allocation of dmaable memory. (dlg@)
~ if_sk.c
> Add more strict receive path packet validation routine. Previously
> the driver could pass corrupted packets to the upper layer under
> certain conditions.
> From yongari at FreeBSDrad@)
~ if_skreg.h
> missed this part of the more strict receive path packet validation
> routine commit. (brad@)
dev/pcmcia
~ pcmciadevs.h
> regen (brad@)
~ pcmciadevs
> move the Siemens vendor id to the right location. (brad@)
dev/sbus
~ cgsix.c ~ cgthree.c
> Do not bother reading the existing colormap on attach, since we will
> override it with the rasops one. (miod@)
dev/sdmmc
~ sdhc.c
> o Change sdhc_soft_reset() to pass the bits we want to reset and wait
> until all bits are cleared.
> o Change sdhc_host_reset() to use the above function and use RESET_ALL.
> This avoid timeouts on TI controllers when initializing the controller but
> does not fix the command timeouts.
> uwe@ ok. (fgsch@)
~ sdhcreg.h
> add some more definitions. (fgsch@)
~ sdhcreg.h
> more. (fgsch@)
dev/usb
~ umodem.c
> Crank buffer sizes to the 1024 limit imposed by ttymalloc()
> which should allow high speed devices to work better.
> tested by fkr@ (jsg@)
dev/wscons
~ wsconsio.h
> Driver for the VS4000/90 frame buffer, adapted from NetBSD (ragge) and
> Blaz Antonic's work, no acceleration yet, untested due to the lack of
> hardware, but enough sacrifices were made to the RAMDAC Gods. (miod@)
~ wsdisplay_compat_usl.c
> Do not fail KDENABIO and KDDISABIO ioctls on non-i386 platforms, this
> allows
> older X servers to run; ok matthieu@ (miod@)
dev/wsfont
~ wsfont.c
> Select the 8x15 font by default on vax. (miod@)
~ wsfont.c
> Add the 12x22 font to vax defaults. (miod@)
~ qvss8x15.h
> Insert empty glyphs to make the font data really contiguous, as rasops
> expects it. (miod@)
gnu
- arch/i386/fpemul/Changelog - arch/i386/fpemul/README
- arch/i386/fpemul/README.OpenBSD - arch/i386/fpemul/control_w.h
- arch/i386/fpemul/div_small.s - arch/i386/fpemul/errors.c
- arch/i386/fpemul/exception.h - arch/i386/fpemul/files.fpemul
- arch/i386/fpemul/fpu_arith.c - arch/i386/fpemul/fpu_asm.h
- arch/i386/fpemul/fpu_aux.c - arch/i386/fpemul/fpu_emu.h
- arch/i386/fpemul/fpu_entry.c - arch/i386/fpemul/fpu_etc.c
- arch/i386/fpemul/fpu_proto.h - arch/i386/fpemul/fpu_system.h
- arch/i386/fpemul/fpu_trig.c - arch/i386/fpemul/get_address.c
- arch/i386/fpemul/load_store.c - arch/i386/fpemul/math_emu.h
- arch/i386/fpemul/poly_2xm1.c - arch/i386/fpemul/poly_atan.c
- arch/i386/fpemul/poly_div.s - arch/i386/fpemul/poly_l2.c
- arch/i386/fpemul/poly_mul64.s - arch/i386/fpemul/poly_sin.c
- arch/i386/fpemul/poly_tan.c - arch/i386/fpemul/polynomial.s
- arch/i386/fpemul/reg_add_sub.c - arch/i386/fpemul/reg_compare.c
- arch/i386/fpemul/reg_constant.c - arch/i386/fpemul/reg_constant.h
- arch/i386/fpemul/reg_div.s - arch/i386/fpemul/reg_ld_str.c
- arch/i386/fpemul/reg_mul.c - arch/i386/fpemul/reg_norm.s
- arch/i386/fpemul/reg_round.s - arch/i386/fpemul/reg_u_add.s
- arch/i386/fpemul/reg_u_div.s - arch/i386/fpemul/reg_u_mul.s
- arch/i386/fpemul/reg_u_sub.s - arch/i386/fpemul/status_w.h
- arch/i386/fpemul/version.h - arch/i386/fpemul/wm_shrx.s
- arch/i386/fpemul/wm_sqrt.s
> Remove.
> ok deraadt@ & many others. (kettenis@)
kern
~ exec_subr.c
> don't call vn_rdwr() with IO_NODELOCKED when the vnode actually isn't
> locked; with this diff VFSDEBUG kernels can boot again
> ok pedro, thib (who came up with the same diff) (sturm@)
net80211
~ ieee80211_input.c
> ALIGNED_POINTER is always defined. (miod@)
scsi
~ cd.c ~ scsi_ioctl.c
~ sd.c
> remove some scsi ioctls, most importantly the ones that can reprobe the
> bus. supporting these complicates the midlayer unnecessarily.
> ok krw@ (dlg@)
~ scsi_all.h ~ scsi_base.c
~ scsiconf.c ~ scsiconf.h
> Use REPORT LUNS to get the list of LUNs to probe. If such a list is
> obtained probe the LUNs given without checking for duplicate INQUIRY
> data.
> For non-USB, non-ATAPI, devices claiming to be SCSI-3 compliant. And
> the target must have something attached at LUN 0.
> If REPORT LUNS can't be used or isn't supported, the old scan process
> is used.
> Fixes Fibre Channel and SCSI enclosure devices that provide identical
> INQUIRY data for all LUNs and were thus being misprobed as having
> only LUN 0.
> Tested by Bob Kitella, dlg@, beck at .ggestions from deraadt at . > ok dlg@ beck@ (krw@)
~ cd.c ~ safte.c
~ scsi_base.c ~ scsiconf.c
~ scsiconf.h ~ sd.c
~ ses.c
> The version field of scsi_inquiry_data is not a simple numeric value
> that specifies the version of SCSI being supported. Even the ANSI part
> that we use is complex. 4 means 2, 5 means 3 and 6 means 4. Translate
> and use the value correctly. Fixes SCSI5 and SCSI6 in dmesg. And
> properly protects SCSI2 devices from getting SCSI3 commands.
> "seems like an elegant solution to me" millert@ ok dlg@ marco@ (krw@)
~ scsi_all.h
> Zap a couple of unused defines - SID_ANSI_SCSI2 and SID_ANSI_SCSI3. (krw@)
sys
~ scsiio.h
> remove some scsi ioctls, most importantly the ones that can reprobe the
> bus. supporting these complicates the midlayer unnecessarily.
> ok krw@ (dlg@)
~ param.h
> crank to 4.0-beta (deraadt@)
ufs/ffs
~ ffs_softdep.c
> missing \n in ddb printf (mickey@)
uvm
~ uvm_amap.c ~ uvm_amap_i.h
~ uvm_anon.c ~ uvm_aobj.c
~ uvm_device.c ~ uvm_fault.c
~ uvm_km.c ~ uvm_map.c
~ uvm_map_i.h ~ uvm_page.c
~ uvm_pager.c ~ uvm_swap.c
~ uvm_vnode.c
> fix fmts for UVMHIST_LOG() entries making it more useful on 64bit archs;
> miod@ ok (mickey@)
== usr.bin =========================================================== 12/13 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/usr.bin
cvs
~ rcsnum.c
> Pull in improper memory allocation from RCS:
> > Not were we allocating memory for a pointer array, it wasn't even for
> > the right variable! Fixes make regress with MALLOC_OPTIONS=PR.
> >
> > Found by and fix worked on with Joris. (ray@)
~ rcs.c
> Sync some whitespace changes from rcs. No binary change. (ray@)
mg
~ basic.c ~ buffer.c
~ def.h ~ dired.c
~ display.c ~ extend.c
~ file.c ~ fileio.c
~ grep.c ~ line.c
~ main.c ~ match.c
~ paragraph.c ~ random.c
~ re_search.c ~ region.c
~ search.c ~ undo.c
~ window.c ~ yank.c
> Rename the header line of a buffer to b_headp, from the remarkably
> unintuitive b_linep. No binary change. (kjell@)
~ basic.c ~ buffer.c
~ def.h ~ dired.c
~ extend.c ~ file.c
~ grep.c ~ line.c
~ random.c ~ search.c
> Add bfirstlp(), blastlp() macros, returning the first and last lines
> of a buffer respectively. Removes an ugly construction than necessitated
> "go to first line"-type comments throughout the code.
> No binary change (kjell@)
~ kbd.h ~ keymap.c
> unused function (deraadt@)
rcs
~ diff.c ~ rcs.c
> spacing (deraadt@)
~ ci.c
> unused variable (deraadt@)
~ rcs.c
> unused variable (deraadt@)
~ rcsnum.c
> Not were we allocating memory for a pointer array, it wasn't even for
> the right variable! Fixes make regress with MALLOC_OPTIONS=PR.
> Found by and fix worked on with Joris.
> OK niallo@, tested by pedro at .ay@)
~ rcs.c
> Second try at adding binary file support to rcs_expand_keywords(),
> using BUFs instead of C strings. Also remove initial keyword search,
> which is a premature optimization, slower, and apparently too hard for
> me to get right.
> OK niallo@ (ray@)
~ rcs.c
> Recalculate buffer length after each expansion.
> OK niallo@ (ray@)
rev
~ rev.c
> Remove double variable assignment.
> From Bret Lambert <bret dot lambert at gmail dot com>
> OK krw@ and deraadt at .ay@)
ssh
~ auth.h ~ dispatch.c
~ kex.h ~ sftp-client.c
> #include <signal.h> for sig_atomic_t; need this prior to <sys/param.h>
> move (stevesk@)
~ sshconnect.c
> disable tunnel forwarding when no strict host key checking
> and key changed; ok djm@ markus@ dtucker@ (stevesk@)
~ scard.c
> need #include <string.h> (stevesk@)
~ includes.h ~ channels.c
~ clientloop.c ~ packet.c
~ scp.c ~ serverloop.c
~ sftp-client.c ~ sftp-server.c
~ ssh-agent.c ~ ssh-keyscan.c
~ sshconnect.c ~ sshd.c
> move #include <sys/time.h> out of includes.h (stevesk@)
~ includes.h ~ atomicio.c
~ auth.c ~ authfile.c
~ buffer.c ~ clientloop.c
~ dh.c ~ groupaccess.c
~ gss-genr.c ~ kex.c
~ kexgexs.c ~ misc.c
~ monitor.c ~ monitor_mm.c
~ packet.c ~ scp.c
~ serverloop.c ~ session.c
~ sftp-client.c ~ sftp-common.c
~ sftp-server.c ~ sftp.c
~ ssh-add.c ~ ssh-agent.c
~ ssh-keygen.c ~ sshlogin.c
~ uidswap.c ~ xmalloc.c
> move #include <sys/param.h> out of includes.h (stevesk@)
~ includes.h ~ authfd.c
~ authfile.c ~ canohost.c
~ channels.c ~ clientloop.c
~ compat.c ~ dh.c
~ hostfile.c ~ kex.c
~ log.c ~ misc.c
~ moduli.c ~ monitor.c
~ packet.c ~ readpass.c
~ scp.c ~ servconf.c
~ session.c ~ sftp-server.c
~ sftp.c ~ ssh-add.c
~ ssh-agent.c ~ ssh-keygen.c
~ ssh-keyscan.c ~ ssh-keysign.c
~ ssh.c ~ sshconnect.c
~ sshconnect1.c ~ sshd.c
~ xmalloc.c
> move #include <stdlib.h> out of includes.h (stevesk@)
~ ssh_config.5
> avoid confusing wording in HashKnownHosts:
> originally spotted by alan amesbury;
> ok deraadt (jmc@)
~ atomicio.h
> order includes to KNF (stevesk@)
tftp
~ main.c ~ tftp.1
~ tftp.c
> Adding TFTP Option Extension to the tftp client according to RFC 2347.
> Implemented options are:
> - TFTP Blocksize Option, RFC 2348
> - TFTP Timeout Interval and Transfer Size Options, RFC 2349
> We have now on the tftp client side the same options supported as in
> our tftp server.
> ok claudio@ (mglocker@)
~ tftp.1
> tweaks; (jmc@)
~ tftp.1
> eliminate hyphen madness; (jmc@)
~ main.c ~ tftp.1
~ tftp.c
> Fixing several timeout quirks at tftpd and tftp:
> - move TIMEOUT* defines to arpa/tftp.h, as they are used several times
> in tftpd and tftp, and the values are part of the RFC definition.
> - tftpd and tftp did count the total retransmission time in retries
> instead in seconds. fixed.
> - tftpd rexmt timeout was hardcoded by a define and therefore didn't
> changed when the timeout option was sent. fixed.
> - limit total retransmission timeout in tftp to also 255 seconds.
> - replace obvious atoi()'s by strtonum().
> ok claudio@ (mglocker@)
~ tftpsubs.c
> Mini KNF. No binary changes. (mglocker@)
~ extern.h
> TIMEOUT* values are not part of the protocol. tftp.h is a namespace
> export of the protocol. you shall not add non-protocol stuff to such
> a file, period. (deraadt@)
~ main.c
> Make tftp atoi() free. (mglocker@)
xargs
~ xargs.1
> Use 'cp -R' instead of 'cp -r' in the example of the -J option.
> ok jmc@ (bernd@)
== usr.sbin ========================================================== 13/13 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/usr.sbin
bgpctl
~ bgpctl.c
> kill redundant prototypes (henning@)
~ Makefile
> amazing how far a typo of mine (CFLAGS vs CLFAGS) from early 2004 spread
> (and how long it went unnoticed)
> From: Alexey Dobriyan <adobriyan at gmail.comhenning@)
bgpd
~ session.c ~ session.h
> factor out the code to build bgp messages, ok claudio a long time ago
> (henning@)
~ session.c
> there's a nasty little race condition when the neigbor reached max-prefix
> and at the same time there is messages from him in the socket buffer,
> because we process the imsgs from the RDE (which tells us max-prefix was
> reached) first, and put the session to IDLE, close connection and
> deallocate buffers. if we then try to read from the socket and write to the
> deallocated buffer we crash, of course. so check wether we have a buffer
> before reading.
> crash seen and fix tested by "Sylwester S. Biernacki" <obeer at obeer.com > (henning@)
dvmrpctl
~ Makefile
> amazing how far a typo of mine (CFLAGS vs CLFAGS) from early 2004 spread
> (and how long it went unnoticed)
> From: Alexey Dobriyan <adobriyan at gmail.comhenning@)
gpioctl
~ Makefile
> amazing how far a typo of mine (CFLAGS vs CLFAGS) from early 2004 spread
> (and how long it went unnoticed)
> From: Alexey Dobriyan <adobriyan at gmail.comhenning@)
hotplugd
~ Makefile
> amazing how far a typo of mine (CFLAGS vs CLFAGS) from early 2004 spread
> (and how long it went unnoticed)
> From: Alexey Dobriyan <adobriyan at gmail.comhenning@)
httpd
~ src/modules/standard/mod_rewrite.c
> A vulnerability has been reported in Apache HTTP Server, which potentially
> can
> be exploited by malicious people to compromise a vulnerable system.
> The vulnerability is caused by a off-by-one error in mod_rewrite and can be
> exploited to cause a one-byte buffer overflow.
> http://secunia.com/advisories/21197/
> report "Alexey E. Suslikov" <cruel at texnika.com.uathe obvious fix
> mailed in by Stuart Henderson <stu at spacehopper.orghenning@)
~ src/modules/example/mod_example.c ~ src/modules/standard/mod_info.c
~ src/modules/standard/mod_status.c
> avoid printing the server built date (henning@)
~ src/main/http_main.c
> avoid printing server built date, we set that to "unknown" anyway
> leave the function there of course, it's part of the API (henning@)
~ src/modules/standard/mod_rewrite.c
TAGGED OPENBSD_3_9
> MFC:
> Fix by henning@
> A vulnerability has been reported in Apache HTTP Server, which potentially
> can
> be exploited by malicious people to compromise a vulnerable system.
> The vulnerability is caused by a off-by-one error in mod_rewrite and can be
> exploited to cause a one-byte buffer overflow.
> http://secunia.com/advisories/21197/
> report "Alexey E. Suslikov" <cruel at texnika.com.uathe obvious fix
> mailed in by Stuart Henderson <stu at spacehopper.orgbrad@)
~ src/modules/standard/mod_rewrite.c
TAGGED OPENBSD_3_8
> MFC:
> Fix by henning@
> A vulnerability has been reported in Apache HTTP Server, which potentially
> can
> be exploited by malicious people to compromise a vulnerable system.
> The vulnerability is caused by a off-by-one error in mod_rewrite and can be
> exploited to cause a one-byte buffer overflow.
> http://secunia.com/advisories/21197/
> report "Alexey E. Suslikov" <cruel at texnika.com.uathe obvious fix
> mailed in by Stuart Henderson <stu at spacehopper.orgbrad@)
ifstated
~ Makefile
> amazing how far a typo of mine (CFLAGS vs CLFAGS) from early 2004 spread
> (and how long it went unnoticed)
> From: Alexey Dobriyan <adobriyan at gmail.comhenning@)
ospfctl
~ Makefile
> amazing how far a typo of mine (CFLAGS vs CLFAGS) from early 2004 spread
> (and how long it went unnoticed)
> From: Alexey Dobriyan <adobriyan at gmail.comhenning@)
wsconscfg
~ Makefile
> Build wscons stuff on vax (but no for bsd.rd yet) (miod@)
===============================================================================
More information about the owc
mailing list