[owc] Weekly src changes ending 2008-03-30
OWC
auto at squish.net
Mon Mar 31 07:00:01 BST 2008
OpenBSD src changes summary for 2008-03-23 to 2008-03-30 inclusive
==================================================================
bin/ps distrib/amd64
distrib/armish distrib/hp300
distrib/i386 distrib/landisk
distrib/mac68k distrib/sets
distrib/zaurus games/boggle
gnu/egcs gnu/lib/libstdc++
gnu/usr.bin/gcc gnu/usr.bin/perl
kerberosV lib/libc
lib/libkvm lib/libpthread
libexec/ftpd libexec/login_skey
libexec/login_tis libexec/spamd
libexec/tftp-proxy libexec/tftpd
regress/sys regress/usr.bin
sbin/disklabel sbin/ifconfig
sbin/isakmpd sbin/mount_portal
sbin/pflogd sbin/ping6
sbin/raidctl sbin/scan_ffs
sbin/umount share/man
share/misc share/mk
sys/arch/amd64/amd64 sys/arch/amd64/conf
sys/arch/amd64/pci sys/arch/arm/arm
sys/arch/armish/armish sys/arch/i386/conf
sys/arch/i386/i386 sys/arch/i386/isa
sys/arch/i386/pci sys/arch/landisk/landisk
sys/arch/macppc/macppc sys/arch/macppc/pci
sys/arch/mips64/include sys/arch/mips64/mips64
sys/arch/mvme68k/stand/libsa sys/arch/mvme88k/stand/libsa
sys/arch/mvmeppc/mvmeppc sys/arch/mvmeppc/stand/libsa
sys/arch/sgi/include sys/arch/sgi/localbus
sys/arch/sgi/sgi sys/arch/sgi/stand/boot
sys/arch/solbourne/solbourne sys/arch/sparc/sparc
sys/arch/sparc64/conf sys/arch/sparc64/dev
sys/arch/sparc64/include sys/arch/sparc64/sparc64
sys/arch/vax/boot/boot sys/arch/vax/include
sys/arch/vax/vax sys/arch/zaurus/zaurus
sys/ddb sys/dev
sys/dev/ata sys/dev/i2c
sys/dev/pci sys/dev/raidframe
sys/kern sys/lib/libsa
usr.bin/make usr.bin/ssh
usr.bin/vi usr.sbin/bgpd
usr.sbin/bind usr.sbin/config
usr.sbin/eeprom usr.sbin/ospf6d
usr.sbin/ospfd usr.sbin/ppp
usr.sbin/relayd usr.sbin/ripd
usr.sbin/route6d usr.sbin/rtadvd
usr.sbin/rtsold usr.sbin/snmpd
usr.sbin/syslogd usr.sbin/tcpdump
usr.sbin/traceroute6
== bin =============================================================== 01/13 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/bin
ps
~ fmt.c
> - extern.h is already included by ps.h, so zap it.
> from Jacek Masiulaniec, via ray@
> ok ray@ (jasper@)
== distrib =========================================================== 02/13 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/distrib
amd64
~ common/install.md
> No point in warning users during install that the disklabel has a
> weird number of partitions, doesn't checksum, doesn't exist or was
> found at a particular sector. Brings all archs together in omerta by
> eliminating use of '-r' and examination of disklabel messages.
> ok deraadt@ (krw@)
armish
~ ramdisk/install.md
> No point in warning users during install that the disklabel has a
> weird number of partitions, doesn't checksum, doesn't exist or was
> found at a particular sector. Brings all archs together in omerta by
> eliminating use of '-r' and examination of disklabel messages.
> ok deraadt@ (krw@)
hp300
~ ramdisk/install.md
> No point in warning users during install that the disklabel has a
> weird number of partitions, doesn't checksum, doesn't exist or was
> found at a particular sector. Brings all archs together in omerta by
> eliminating use of '-r' and examination of disklabel messages.
> ok deraadt@ (krw@)
i386
~ common/install.md
> No point in warning users during install that the disklabel has a
> weird number of partitions, doesn't checksum, doesn't exist or was
> found at a particular sector. Brings all archs together in omerta by
> eliminating use of '-r' and examination of disklabel messages.
> ok deraadt@ (krw@)
landisk
~ ramdisk/install.md
> No point in warning users during install that the disklabel has a
> weird number of partitions, doesn't checksum, doesn't exist or was
> found at a particular sector. Brings all archs together in omerta by
> eliminating use of '-r' and examination of disklabel messages.
> ok deraadt@ (krw@)
mac68k
~ ramdisk/install.md
> Bring mac68k into the md_checkfordisklabel()-free world.
> Quick test by martin@ shows it seems to still work. (krw@)
sets
~ lists/comp/mi
> sync (deraadt@)
~ lists/man/mi
> sync (deraadt@)
~ lists/man/mi
> sync (deraadt@)
~ lists/base/mi
> sync (deraadt@)
~ lists/base/md.landisk ~ lists/comp/md.landisk
> Enable shared libraries on landisk.
> The sh architecture required some different linkage than most platforms,
> This fixes those other requirements. (drahn@)
~ lists/man/mi
> sync (deraadt@)
zaurus
~ ramdisk/install.md
> No point in warning users during install that the disklabel has a
> weird number of partitions, doesn't checksum, doesn't exist or was
> found at a particular sector. Brings all archs together in omerta by
> eliminating use of '-r' and examination of disklabel messages.
> ok deraadt@ (krw@)
== games ============================================================= 03/13 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/games
boggle
~ boggle/bog.c
> remove dependency on recalloc() (millert@)
~ boggle/mach.c ~ boggle/timer.c
> Bomb out if we cannot read the input. This happens when select
> says there has been a status change to indicate that the tty has
> gone away. Prevents spinning on terminal detach. OK pvalchev@ (millert@)
== gnu =============================================================== 04/13 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/gnu
egcs
~ gcc/config/openbsd.h
> Remove unnecessary whitespace on .type foo, at object lines, which confuse gas
> on some (not in-tree) configurations.
> ok espie@ kettenis@ (miod@)
lib/libstdc++
~ ltcf-cxx.sh
> Enable shared libraries on landisk.
> The sh architecture required some different linkage than most platforms,
> This fixes those other requirements. (drahn@)
usr.bin/gcc
~ gcc/config/sh/lib1funcs.asm
> Enable shared libraries on landisk.
> The sh architecture required some different linkage than most platforms,
> This fixes those other requirements. (drahn@)
usr.bin/perl
~ perl.c
> msg_controllen has to be CMSG_SPACE so that the kernel can account for
> each cmsg_len (ie. msg_controllen = sum of CMSG_ALIGN(cmsg_len). This
> works now that kernel fd passing has been fixed to accept a bit of
> sloppiness because of this ABI repair.
> lots of discussion with kettenis (deraadt@)
~ hints/openbsd.sh
> Enable shared libraries on landisk.
> The sh architecture required some different linkage than most platforms,
> This fixes those other requirements. (drahn@)
== kerberosV ========================================================= 05/13 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/kerberosV
kerberosV
~ src/kcm/connect.c
> msg_controllen has to be CMSG_SPACE so that the kernel can account for
> each cmsg_len (ie. msg_controllen = sum of CMSG_ALIGN(cmsg_len). This
> works now that kernel fd passing has been fixed to accept a bit of
> sloppiness because of this ABI repair.
> lots of discussion with kettenis (deraadt@)
== lib =============================================================== 06/13 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/lib
libc
~ gen/auth_subr.c
> msg_controllen has to be CMSG_SPACE so that the kernel can account for
> each cmsg_len (ie. msg_controllen = sum of CMSG_ALIGN(cmsg_len). This
> works now that kernel fd passing has been fixed to accept a bit of
> sloppiness because of this ABI repair.
> lots of discussion with kettenis (deraadt@)
~ gen/statvfs.3
> section 3, Darrin Chandler (otto@)
~ gen/statvfs.3
> be more explicit about the unit used for the f_b* fields; prompted by
> and ok djm@ (otto@)
~ gen/statvfs.3
> HISTORY tweak; ok otto (jmc@)
libkvm
~ kvm_sparc64.c
> Unbreak build. (kettenis@)
libpthread
~ uthread/uthread_file.c
> Use fileno() instead of peeking into FILE *; Paul Stoeber (deraadt@)
== libexec =========================================================== 07/13 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/libexec
ftpd
~ monitor_fdpass.c
> msg_controllen has to be CMSG_SPACE so that the kernel can account for
> each cmsg_len (ie. msg_controllen = sum of CMSG_ALIGN(cmsg_len). This
> works now that kernel fd passing has been fixed to accept a bit of
> sloppiness because of this ABI repair.
> lots of discussion with kettenis (deraadt@)
login_skey
~ login_skey.c
> msg_controllen has to be CMSG_SPACE so that the kernel can account for
> each cmsg_len (ie. msg_controllen = sum of CMSG_ALIGN(cmsg_len). This
> works now that kernel fd passing has been fixed to accept a bit of
> sloppiness because of this ABI repair.
> lots of discussion with kettenis (deraadt@)
login_tis
~ login_tis.c
> msg_controllen has to be CMSG_SPACE so that the kernel can account for
> each cmsg_len (ie. msg_controllen = sum of CMSG_ALIGN(cmsg_len). This
> works now that kernel fd passing has been fixed to accept a bit of
> sloppiness because of this ABI repair.
> lots of discussion with kettenis (deraadt@)
spamd
~ spamd.8
> typo, ok beck@ (grunk@)
tftp-proxy
~ tftp-proxy.c
> msg_controllen has to be CMSG_SPACE so that the kernel can account for
> each cmsg_len (ie. msg_controllen = sum of CMSG_ALIGN(cmsg_len). This
> works now that kernel fd passing has been fixed to accept a bit of
> sloppiness because of this ABI repair.
> lots of discussion with kettenis (deraadt@)
tftpd
~ tftpd.c
> msg_controllen has to be CMSG_SPACE so that the kernel can account for
> each cmsg_len (ie. msg_controllen = sum of CMSG_ALIGN(cmsg_len). This
> works now that kernel fd passing has been fixed to accept a bit of
> sloppiness because of this ABI repair.
> lots of discussion with kettenis (deraadt@)
== regress =========================================================== 08/13 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/regress
sys
~ kern/unfdpass/unfdpass.c
> msg_controllen has to be CMSG_SPACE so that the kernel can account for
> each cmsg_len (ie. msg_controllen = sum of CMSG_ALIGN(cmsg_len). This
> works now that kernel fd passing has been fixed to accept a bit of
> sloppiness because of this ABI repair.
> lots of discussion with kettenis (deraadt@)
usr.bin
~ ssh/sftp-badcmds.sh
> disable no-replace rename test now that we prefer a POSIX rename; spotted
> by dkrause@ (djm@)
== sbin ============================================================== 09/13 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/sbin
disklabel
~ disklabel.8 ~ disklabel.c
> Nuke disklabel -r. All labels come from the kernel via ioctl.
> "make sense" miod@ ok deraadt@ (krw@)
ifconfig
~ ifconfig.c
> fix a free(NULL) in setcarp_nodes().
> ok mpf@, chl@
> "i agree with the diagnosis" oga@ (damien@)
isakmpd
~ monitor_fdpass.c
> msg_controllen has to be CMSG_SPACE so that the kernel can account for
> each cmsg_len (ie. msg_controllen = sum of CMSG_ALIGN(cmsg_len). This
> works now that kernel fd passing has been fixed to accept a bit of
> sloppiness because of this ABI repair.
> lots of discussion with kettenis (deraadt@)
mount_portal
~ activate.c
> msg_controllen has to be CMSG_SPACE so that the kernel can account for
> each cmsg_len (ie. msg_controllen = sum of CMSG_ALIGN(cmsg_len). This
> works now that kernel fd passing has been fixed to accept a bit of
> sloppiness because of this ABI repair.
> lots of discussion with kettenis (deraadt@)
pflogd
~ privsep_fdpass.c
> msg_controllen has to be CMSG_SPACE so that the kernel can account for
> each cmsg_len (ie. msg_controllen = sum of CMSG_ALIGN(cmsg_len). This
> works now that kernel fd passing has been fixed to accept a bit of
> sloppiness because of this ABI repair.
> lots of discussion with kettenis (deraadt@)
ping6
~ ping6.c
> msg_controllen has to be CMSG_SPACE so that the kernel can account for
> each cmsg_len (ie. msg_controllen = sum of CMSG_ALIGN(cmsg_len). This
> works now that kernel fd passing has been fixed to accept a bit of
> sloppiness because of this ABI repair.
> lots of discussion with kettenis (deraadt@)
raidctl
~ raidctl.8
> Remove two stray references to the departed -r argument to disklabel.
> Pointed out by jmc@ (krw@)
scan_ffs
~ scan_ffs.8
> Remove two stray references to the departed -r argument to disklabel.
> Pointed out by jmc@ (krw@)
umount
~ umount.c
> zap undocumented 'F' flag; prompted by J.Thornburg. ok millert@ (otto@)
== share ============================================================= 10/13 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/share
man
~ man4/fins.4
> update sensor list; from Geoff Steckel (deraadt@)
~ man4/Makefile ~ man4/iic.4
+ man4/andl.4
> andl(4) manual page for Andigilog aSC7611 (cnst@)
~ man4/ddb.4
> show all mounts command for ddb. From mickey a long time ago. (miod@)
~ man4/fins.4
> .Pp is not needed before or after .Sh; (jmc@)
~ man4/andl.4
> tweak previous; (jmc@)
~ man4/man4.sparc64/Makefile ~ man4/man4.sparc64/agten.4
~ man4/man4.sparc64/bwtwo.4 ~ man4/man4.sparc64/cgsix.4
~ man4/man4.sparc64/cgthree.4 ~ man4/man4.sparc64/cgtwelve.4
~ man4/man4.sparc64/creator.4 ~ man4/man4.sparc64/mgx.4
~ man4/man4.sparc64/rfx.4 ~ man4/man4.sparc64/tvtwo.4
~ man4/man4.sparc64/vigra.4 ~ man4/man4.sparc64/zx.4
+ man4/man4.sparc64/ifb.4
> Not-functional-enough driver for the Sun Expert3D cards, put in tree as an
> incentive for further tinkering only; not for the faint of heart. (miod@)
~ man4/ddb.4
> Missing .El; spotted by jmc@ (miod@)
~ man4/fins.4
> 4.3 is fin; ok deraadt (cnst@)
~ man4/man4.sgi/power.4
> Correct the synopsis.
> ok jasper@ (jsing@)
~ man5/bsd.port.mk.5
> - since we are in control of shared library versions, advise people to
> use "0.0" for SHARED_LIBS version when importing a new port
> ok espie@ (ajacoutot@)
~ man3/CMSG_DATA.3
> msg_controllen has to be CMSG_SPACE so that the kernel can account for
> each cmsg_len (ie. msg_controllen = sum of CMSG_ALIGN(cmsg_len). This
> works now that kernel fd passing has been fixed to accept a bit of
> sloppiness because of this ABI repair.
> lots of discussion with kettenis (deraadt@)
~ man4/it.4
> add it(4) at port 0x228, found on some motherboards.
> ok deraadt@ (form@)
~ man4/man4.sparc64/ifb.4
> add "sparc64" to .Dt; (jmc@)
~ man4/man4.hp300/apci.4 ~ man4/man4.hp300/dca.4
~ man4/man4.hp300/dcm.4 ~ man4/man4.hp300/dio.4
~ man4/man4.hp300/dvbox.4 ~ man4/man4.hp300/gbox.4
~ man4/man4.hp300/hyper.4 ~ man4/man4.hp300/rbox.4
~ man4/man4.hp300/topcat.4 ~ man4/man4.hp300/tvrx.4
> no sense for devices to reference other (same type) devices;
> requested by deraadt/miod
> help/ok miod (jmc@)
~ man4/man4.hppa/phantomas.4
> some readability fixes for DESCRIPTION; (jmc@)
~ man4/man4.hp300/intro.4
> add a couple missing devices; ok miod (jmc@)
~ man4/Makefile ~ man4/pci.4
+ man4/kate.4
> kate(4) manual page (cnst@)
~ man4/man4.sgi/gbe.4 ~ man4/man4.sgi/macebus.4
~ man4/man4.sgi/mavb.4 ~ man4/man4.sgi/mec.4
~ man4/man4.sgi/mkbc.4
> Bring the synopsis sections of various sgi man pages into line with direct
> configuration.
> ok miod@ (jsing@)
~ man4/man4.mvme68k/ie.4 ~ man4/man4.mvme68k/le.4
> no sense for devices to reference other (same type) devices;
> requested by deraadt/miod (jmc@)
~ man9/mbuf.9
> correct arg name; from Matthew Dempsky
> ok claudio (jmc@)
misc
~ airport
> Add PVG and change wording of SHA. (kettenis@)
mk
~ bsd.dep.mk
> check for emptiness of SRCS, since make warns otherwise.
> okay deraadt@ (espie@)
~ Makefile + bsd.xconf.mk
> Add a bsd.xconf.mk glue file, to be included at the start of Xenocara
> makefiles. ok oga at . (matthieu@)
~ bsd.own.mk
> Enable shared libraries on landisk.
> The sh architecture required some different linkage than most platforms,
> This fixes those other requirements. (drahn@)
== sys =============================================================== 11/13 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/sys
arch/amd64/amd64
~ machdep.c
> Fix amd64 and i386 serial console handling for non-default speeds. Bring
> i386 pccom into line with com. Problem reported and first diff by Markus
> Hennecke, who also tested and corrected subsequent diffs. Boot messages
> and console now work with his 57K serial connections.
> Suggestions and ok miod@ (krw@)
arch/amd64/conf
~ GENERIC
> enable andl; ok deraadt (cnst@)
~ GENERIC
> add it(4) at port 0x228, found on some motherboards.
> ok deraadt@ (form@)
~ GENERIC
> enable kate(4); ok deraadt@ (cnst@)
arch/amd64/pci
~ pchb.c
> Add AGP support for the Intel G33 chipset.
> Tested on i386 with a Gigabyte G33M-DS2R motherboard (Intel GMA 3100).
> Based on NetBSD code.
> ok oga@ (damien@)
arch/arm/arm
~ arm32_machdep.c
> setroot() is supposed to be safe in all cases, so kill the special cases
> for RB_DFLTROOT; ok miod (deraadt@)
arch/armish/armish
~ armish_machdep.c
> setroot() is supposed to be safe in all cases, so kill the special cases
> for RB_DFLTROOT; ok miod (deraadt@)
arch/i386/conf
~ GENERIC
> enable andl; ok deraadt (cnst@)
~ GENERIC
> add it(4) at port 0x228, found on some motherboards.
> ok deraadt@ (form@)
~ GENERIC
> enable kate(4); ok deraadt@ (cnst@)
arch/i386/i386
~ bios.c
> Fix amd64 and i386 serial console handling for non-default speeds. Bring
> i386 pccom into line with com. Problem reported and first diff by Markus
> Hennecke, who also tested and corrected subsequent diffs. Boot messages
> and console now work with his 57K serial connections.
> Suggestions and ok miod@ (krw@)
~ machdep.c
> Fix #ifdef NCOM collateral damage introduced in r1.210.
> ok miod@ kettenis@ (krw@)
arch/i386/isa
~ pccom.c
> Fix amd64 and i386 serial console handling for non-default speeds. Bring
> i386 pccom into line with com. Problem reported and first diff by Markus
> Hennecke, who also tested and corrected subsequent diffs. Boot messages
> and console now work with his 57K serial connections.
> Suggestions and ok miod@ (krw@)
arch/i386/pci
~ pchb.c
> Add AGP support for the Intel G33 chipset.
> Tested on i386 with a Gigabyte G33M-DS2R motherboard (Intel GMA 3100).
> Based on NetBSD code.
> ok oga@ (damien@)
arch/landisk/landisk
~ machdep.c
> setroot() is supposed to be safe in all cases, so kill the special cases
> for RB_DFLTROOT; ok miod (deraadt@)
arch/macppc/macppc
~ machdep.c
> setroot() is supposed to be safe in all cases, so kill the special cases
> for RB_DFLTROOT; ok miod (deraadt@)
~ autoconf.c
> nam2blk[] clearly must contain "rd" ... found by krw (deraadt@)
arch/macppc/pci
~ mpcpcibus.c
> Make pci_intr_map(9) fail if the interrupt line register is set to 0xff.
> Prevents crashes later on when we try to estabish the interrupt.
> ok miod@, deraadt@ (kettenis@)
arch/mips64/include
~ memconf.h ~ vmparam.h
> Allow the machine-specific headers to overwrite the number of memory
> segments and the number of freelists. (miod@)
arch/mips64/mips64
~ mem.c
> Remove APERTURE remains. (miod@)
arch/mvme68k/stand/libsa
~ parse_args.c
> setroot() is supposed to be safe in all cases, so kill the special cases
> for RB_DFLTROOT; ok miod (deraadt@)
arch/mvme88k/stand/libsa
~ parse_args.c
> setroot() is supposed to be safe in all cases, so kill the special cases
> for RB_DFLTROOT; ok miod (deraadt@)
arch/mvmeppc/mvmeppc
~ autoconf.c
> nam2blk[] clearly must contain "rd" ... found by krw (deraadt@)
arch/mvmeppc/stand/libsa
~ parse_args.c
> setroot() is supposed to be safe in all cases, so kill the special cases
> for RB_DFLTROOT; ok miod (deraadt@)
arch/sgi/include
~ bus.h
> Implement bus_space_*8 and bus_space_vaddr. (miod@)
arch/sgi/localbus
~ com_lbus.c
> Majorly cleanup/overhaul com @ macebus code.
> ok miod@ (jsing@)
~ macebus.c ~ macebus.h
> Implement bus_space_*8 and bus_space_vaddr. (miod@)
arch/sgi/sgi
~ machdep.c
> setroot() is supposed to be safe in all cases, so kill the special cases
> for RB_DFLTROOT; ok miod (deraadt@)
~ clock_md.c
> Data mode can only be changed when updating the clock, so don't bother
> trying to change to BCD mode before reading. Also, don't assume that the
> clock is in 24 hour mode when updating.
> tested jasper@, ok miod@ (jsing@)
~ locore.S
> Remove former idle_stack. (miod@)
~ autoconf.c
> nam2blk[] clearly must contain "rd" ... found by krw (deraadt@)
arch/sgi/stand/boot
~ arcbios.c
> Remove unwanted code and sync with mips64/arcbios.c.
> ok miod@ (jsing@)
~ boot.c
> If we fail to load the kernel, drop the user into interactive mode of ARCS,
> rather than just rebooting.
> ok deraadt@, miod@, jasper@ (jsing@)
arch/solbourne/solbourne
~ autoconf.c
> setroot() is supposed to be safe in all cases, so kill the special cases
> for RB_DFLTROOT; ok miod (deraadt@)
arch/sparc/sparc
~ locore.s
> Make ovbcopy() the default bcopy() flavour; requested by deraadt@ some time
> ago. (miod@)
~ autoconf.c
> setroot() is supposed to be safe in all cases, so kill the special cases
> for RB_DFLTROOT; ok miod (deraadt@)
arch/sparc64/conf
~ files.sparc64
> Not-functional-enough driver for the Sun Expert3D cards, put in tree as an
> incentive for further tinkering only; not for the faint of heart. (miod@)
~ GENERIC ~ RAMDISK
> More sun4v support. GENERIC and RAMDISK kernels will now boot on both
> sun4u and sun4v. GENERIC.MP won't work yet though. (kettenis@)
arch/sparc64/dev
~ vgafb.c
> Use new defines from pcidevs to match Expert3D. (miod@)
+ ifb.c
> Not-functional-enough driver for the Sun Expert3D cards, put in tree as an
> incentive for further tinkering only; not for the faint of heart. (miod@)
~ ifb.c
> typo; deraadt (miod@)
~ pci_machdep.c
> Remove KASSERT that's causing grieve on sun4v, and I've never seen
> triggered. (kettenis@)
arch/sparc64/include
~ pte.h
> Sigh! The sun4v TTEs have a different layout than sun4u TTEs. Rename the
> existing sun4u defines and add sun4v. For now, decide which set to use
> at compile time. Change the sun4u-specific code in locore.s to use the
> sun4u
> defines. (kettenis@)
~ cpu.h
> Add code to initialize CPUs on sun4v. (kettenis@)
~ pte.h
> More sun4v support. GENERIC and RAMDISK kernels will now boot on both
> sun4u and sun4v. GENERIC.MP won't work yet though. (kettenis@)
arch/sparc64/sparc64
~ locore.s
> Fix ovbcopy() operation when copying shorts backwards, similar to sparc
> locore.s r1.76 (miod@)
~ genassym.cf ~ locore.s
> Sigh! The sun4v TTEs have a different layout than sun4u TTEs. Rename the
> existing sun4u defines and add sun4v. For now, decide which set to use
> at compile time. Change the sun4u-specific code in locore.s to use the
> sun4u
> defines. (kettenis@)
~ cpu.c ~ pmap.c
> Add code to initialize CPUs on sun4v. (kettenis@)
~ cpu.c
> Use correct cpuid on sun4v. (kettenis@)
~ trap.c
> Disable some code that reads some sun4u-specific registers for diagnostic
> reasons, but isn't otherwise used. (kettenis@)
~ clock.c
> Don't double count clock and prof interrupts. (kettenis@)
~ autoconf.c ~ cache.h
~ genassym.cf ~ locore.s
~ pmap.c
> More sun4v support. GENERIC and RAMDISK kernels will now boot on both
> sun4u and sun4v. GENERIC.MP won't work yet though. (kettenis@)
~ locore.s
> Remove for #if 0'ed out code I left behind with the previous commit.
> (kettenis@)
arch/vax/boot/boot
~ Makefile ~ boot.c
~ version
> Switch bootloader code to loadfile, and enable a.out and ELF loaders
> within.
> Crank the second stage boot blocks version. (miod@)
arch/vax/include
~ exec.h
> Let the kernel load forthcoming ELF binaries. (miod@)
~ db_machdep.h
> Nuke common. (miod@)
+ loadfile_machdep.h
> Switch bootloader code to loadfile, and enable a.out and ELF loaders
> within.
> Crank the second stage boot blocks version. (miod@)
arch/vax/vax
~ db_machdep.c
> Nuke common. (miod@)
~ locore.c ~ pmap.c
> Use vaddr_t instead of void * for some initial virtual memory layout
> arithmetic. (miod@)
arch/zaurus/zaurus
~ zaurus_machdep.c
> setroot() is supposed to be safe in all cases, so kill the special cases
> for RB_DFLTROOT; ok miod (deraadt@)
ddb
~ db_command.c ~ db_command.h
> show all mounts command for ddb. From mickey a long time ago. (miod@)
dev
~ vnd.c
> Change XXgetdisklabel() to the 'normal' four parameter model to
> prepare for adding missing DIOC* ioctl's.
> No functional change. (krw@)
dev/ata
~ atascsi.c
> plug a mem leak. the array of ports wasnt freed on detach. (dlg@)
dev/i2c
~ files.i2c + asc7611.c
> New andl(4) driver for Andigilog aSC7611 Hardware Monitor.
> Tested in my i2c sandbox against an i2c dump from D946GZIS,
> a dmesg for which was kindly donated to our dmesg at .
> 'commit it' deraadt@ (cnst@)
~ i2c_scan.c
> test the Vdd registers of Andigilog chips against the minimum voltage
> readings;
> 'good plan' deraadt; ok kettenis, mikeb (cnst@)
dev/pci
~ agp.c ~ agp_i810.c
~ agpreg.h
> Add AGP support for the Intel G33 chipset.
> Tested on i386 with a Gigabyte G33M-DS2R motherboard (Intel GMA 3100).
> Based on NetBSD code.
> ok oga@ (damien@)
~ pcidevs
> More 3DLabs devices. (miod@)
~ pcidevs.h ~ pcidevs_data.h
> regen (miod@)
~ agp.c ~ agp_i810.c
~ agpvar.h
> Sometimes the registers claiming the aperture size on 9XX chipsets lie.
> So instead, use the size of the aperture BAR to determine the aperture
> size for agp_i810, and don't let the 855 and above have the aperture
> set. Next to nothing uses it anyway so no harm.
> introduce generic_{get,set} aperture which has this behaviour so other
> drivers may use it if useful.
> Change inspired by what freebsd did after i noticed my aperture size was
> wrong.
> Tested by many, ok kettenis at . (oga@)
~ auvia.c
> auvia(4) doesn't record properly, so don't claim full-duplex
> capabilities. (jakemsr@)
~ auich.c
> 8-bit has been emulated for some time now. quit lying. (jakemsr@)
~ files.pci + kate.c
> welcome kate(4), a new driver supporting AMD K8 temperature sensors
> discussed with deraadt and others
> tested by jasper, deraadt, todd and others (cnst@)
~ kate.c
> update the revision cpuid table, removing three duplicate ids from "JH-F"
> and adding two missing ids for Mobile AMD Sempron rev DH-G2; tested todd
> (cnst@)
~ jmb.c
> add detach support. this has to be the easiest diff for hotplug ever, go
> read it :)
> ok kettenis@ jsg@ (dlg@)
~ auvia.c
> todd@ has 82c686 based auvia(4) devices that do full-duplex
> properly. restore full-duplex capabilities for 82c686 based
> devices. (jakemsr@)
~ if_lii.c
> Pad the RXD buffer so that packets are aligned on a 128-byte boundary. This
> prevents the chip from hard locking the machine when receiving packets.
> Diff committed from an ASUS EeePC using lii(4).
> From NetBSD.
> ok dlg@ (jsing@)
~ if_lii.c
> Enable setting of interrupt timers.
> ok dlg@ (jsing@)
~ if_lii.c
> Add code to configure the vendor specific opcodes of the SPI Flash.
> From NetBSD.
> ok dlg@ (jsing@)
dev/raidframe
~ rf_openbsdkintf.c
> Change XXgetdisklabel() to the 'normal' four parameter model to
> prepare for adding missing DIOC* ioctl's.
> No functional change. (krw@)
~ rf_openbsdkintf.c
> Add DIOCRLDINFO support. (krw@)
kern
~ vfs_subr.c
> Wrong printf construct. (miod@)
~ subr_disk.c
> setroot() is supposed to be safe in all cases, so kill the special cases
> for RB_DFLTROOT; ok miod (deraadt@)
~ uipc_usrreq.c
> We were led astray (like many others before us) to believe that
> msg_controllen should be CMSG_LEN() instead of CMSG_SPACE() because
> the kernel fd passing code was erroring out when
> "cm->cmsg_len != control->m_len"
> instead of
> "CMSG_ALIGN(cm->cmsg_len) != control->m_len".
> On machines with 16-byte alignment, when one thinks about how the ALIGN
> padding happens, it is clear that msg_controllen has to be CMSG_SPACE()
> or the kernel cannot hope to bounds check the messages correctly.
> For now, change the check to cm->cmsg_len > control->m_len to permit the
> old ABI to continue working. Later perhaps when all the old binaries
> are gone we can stop permitting their use.
> lots of discussion with kettenis (deraadt@)
~ subr_userconf.c
> Make it possible to disable/enable pseudo devices in UKC, config file and
> at config -e. -moj ok deraadt@ (maja@)
~ subr_disk.c
> Do not do the ``swap generic'' behaviour if rootdev != NODEV when entering
> setroot(), as this neuters RAIDframe autoconfiguration; found the hard way
> by naddy@ (miod@)
lib/libsa
~ loadfile.c
> Let this compile with BOOT_AOUT defined. (miod@)
== usr.bin =========================================================== 12/13 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/usr.bin
make
~ job.c ~ memory.c
~ memory.h
> bye, bye recalloc. Bad interface for various reasons.
> discussed with deraadt@ and otto@ and millert@ (espie@)
ssh
~ sftp-client.c
> prefer POSIX-style file renaming over filexfer rename behaviour if the
> server
> supports the posix-rename at openssh.com extension.
> Note that the old (filexfer) behaviour would refuse to clobber an existing
> file. Users who depended on this should adjust their sftp(1) usage.
> ok deraadt@ markus@ (djm@)
~ monitor_fdpass.c
> msg_controllen has to be CMSG_SPACE so that the kernel can account for
> each cmsg_len (ie. msg_controllen = sum of CMSG_ALIGN(cmsg_len). This
> works now that kernel fd passing has been fixed to accept a bit of
> sloppiness because of this ABI repair.
> lots of discussion with kettenis (deraadt@)
~ session.c ~ sshd_config.5
> ignore ~/.ssh/rc if a sshd_config ForceCommand is specified;
> from dtucker@ ok deraadt@ djm@ (djm@)
~ session.c
> last patch had backwards test; spotted by termim AT gmail.com (djm@)
~ auth-options.c ~ auth-options.h
~ session.c ~ sshd.8
> add no-user-rc authorized_keys option to disable execution of ~/.ssh/rc
> (djm@)
~ sftp-server-main.c
> Fix RCS ident (djm@)
~ version.h
> openssh-4.9 (djm@)
~ sshd.8
> remove trailing whitespace; (jmc@)
vi
~ cl/cl_main.c ~ catalog/dump.c
> - minor ansification
> from Gleydson Soares
> ok otto@ (jasper@)
== usr.sbin ========================================================== 13/13 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/usr.sbin
bgpd
~ buffer.c ~ imsg.c
> msg_controllen has to be CMSG_SPACE so that the kernel can account for
> each cmsg_len (ie. msg_controllen = sum of CMSG_ALIGN(cmsg_len). This
> works now that kernel fd passing has been fixed to accept a bit of
> sloppiness because of this ABI repair.
> lots of discussion with kettenis (deraadt@)
bind
~ lib/isc/unix/privsep_fdpass.c ~ lib/isc/unix/socket.c
> msg_controllen has to be CMSG_SPACE so that the kernel can account for
> each cmsg_len (ie. msg_controllen = sum of CMSG_ALIGN(cmsg_len). This
> works now that kernel fd passing has been fixed to accept a bit of
> sloppiness because of this ABI repair.
> lots of discussion with kettenis (deraadt@)
config
~ ukcutil.c
> Done mark a disable of a disabled device as an change (and same with
> enable).
> Reported by jj@ a long time ago. -moj (maja@)
~ gram.y ~ mkioconf.c
~ sem.c ~ sem.h
~ ukcutil.c
> Make it possible to disable/enable pseudo devices in UKC, config file and
> at config -e. -moj ok deraadt@ (maja@)
eeprom
~ optree.c
> Do not abort walking the tree if a node's contents is too large; allows
> eeprom -p to be used on more macppc models.
> ok fgsch@ (miod@)
ospf6d
~ packet.c
> msg_controllen has to be CMSG_SPACE so that the kernel can account for
> each cmsg_len (ie. msg_controllen = sum of CMSG_ALIGN(cmsg_len). This
> works now that kernel fd passing has been fixed to accept a bit of
> sloppiness because of this ABI repair.
> lots of discussion with kettenis (deraadt@)
ospfd
~ packet.c
> msg_controllen has to be CMSG_SPACE so that the kernel can account for
> each cmsg_len (ie. msg_controllen = sum of CMSG_ALIGN(cmsg_len). This
> works now that kernel fd passing has been fixed to accept a bit of
> sloppiness because of this ABI repair.
> lots of discussion with kettenis (deraadt@)
ppp
~ ppp/bundle.c
> msg_controllen has to be CMSG_SPACE so that the kernel can account for
> each cmsg_len (ie. msg_controllen = sum of CMSG_ALIGN(cmsg_len). This
> works now that kernel fd passing has been fixed to accept a bit of
> sloppiness because of this ABI repair.
> lots of discussion with kettenis (deraadt@)
relayd
~ buffer.c ~ imsg.c
> msg_controllen has to be CMSG_SPACE so that the kernel can account for
> each cmsg_len (ie. msg_controllen = sum of CMSG_ALIGN(cmsg_len). This
> works now that kernel fd passing has been fixed to accept a bit of
> sloppiness because of this ABI repair.
> lots of discussion with kettenis (deraadt@)
ripd
~ packet.c
> msg_controllen has to be CMSG_SPACE so that the kernel can account for
> each cmsg_len (ie. msg_controllen = sum of CMSG_ALIGN(cmsg_len). This
> works now that kernel fd passing has been fixed to accept a bit of
> sloppiness because of this ABI repair.
> lots of discussion with kettenis (deraadt@)
route6d
~ route6d.c
> msg_controllen has to be CMSG_SPACE so that the kernel can account for
> each cmsg_len (ie. msg_controllen = sum of CMSG_ALIGN(cmsg_len). This
> works now that kernel fd passing has been fixed to accept a bit of
> sloppiness because of this ABI repair.
> lots of discussion with kettenis (deraadt@)
rtadvd
~ rtadvd.c
> msg_controllen has to be CMSG_SPACE so that the kernel can account for
> each cmsg_len (ie. msg_controllen = sum of CMSG_ALIGN(cmsg_len). This
> works now that kernel fd passing has been fixed to accept a bit of
> sloppiness because of this ABI repair.
> lots of discussion with kettenis (deraadt@)
rtsold
~ probe.c ~ rtsol.c
> msg_controllen has to be CMSG_SPACE so that the kernel can account for
> each cmsg_len (ie. msg_controllen = sum of CMSG_ALIGN(cmsg_len). This
> works now that kernel fd passing has been fixed to accept a bit of
> sloppiness because of this ABI repair.
> lots of discussion with kettenis (deraadt@)
snmpd
~ buffer.c ~ imsg.c
> msg_controllen has to be CMSG_SPACE so that the kernel can account for
> each cmsg_len (ie. msg_controllen = sum of CMSG_ALIGN(cmsg_len). This
> works now that kernel fd passing has been fixed to accept a bit of
> sloppiness because of this ABI repair.
> lots of discussion with kettenis (deraadt@)
~ snmpd.8 ~ snmpd.conf.5
> .Xr snmpctl
> ok jmc@ (martin@)
syslogd
~ privsep_fdpass.c
> msg_controllen has to be CMSG_SPACE so that the kernel can account for
> each cmsg_len (ie. msg_controllen = sum of CMSG_ALIGN(cmsg_len). This
> works now that kernel fd passing has been fixed to accept a bit of
> sloppiness because of this ABI repair.
> lots of discussion with kettenis (deraadt@)
tcpdump
~ privsep_fdpass.c
> msg_controllen has to be CMSG_SPACE so that the kernel can account for
> each cmsg_len (ie. msg_controllen = sum of CMSG_ALIGN(cmsg_len). This
> works now that kernel fd passing has been fixed to accept a bit of
> sloppiness because of this ABI repair.
> lots of discussion with kettenis (deraadt@)
traceroute6
~ traceroute6.c
> msg_controllen has to be CMSG_SPACE so that the kernel can account for
> each cmsg_len (ie. msg_controllen = sum of CMSG_ALIGN(cmsg_len). This
> works now that kernel fd passing has been fixed to accept a bit of
> sloppiness because of this ABI repair.
> lots of discussion with kettenis (deraadt@)
===============================================================================
More information about the owc
mailing list