[owc] Weekly src changes ending 2006-12-31
OWC
auto at squish.net
Mon Jan 1 07:10:50 GMT 2007
OpenBSD src changes summary for 2006-12-24 to 2006-12-31 inclusive
==================================================================
bin/pax distrib/crunch
distrib/notes distrib/ramdisk
etc/changelist etc/hostapd.conf
games/boggle games/gomoku
games/mille gnu/lib/libstdc++
gnu/usr.bin/gcc lib/libc
libexec libexec/ld.so
sbin/brconfig sbin/dhclient
sbin/dmesg sbin/dump
sbin/pfctl sbin/route
sbin/sysctl share/man
share/mk sys/arch/alpha/alpha
sys/arch/alpha/include sys/arch/amd64/include
sys/arch/amd64/stand/libsa sys/arch/arm/arm
sys/arch/arm/include sys/arch/hppa/hppa
sys/arch/i386/conf sys/arch/i386/i386
sys/arch/i386/include sys/arch/i386/isa
sys/arch/i386/stand/libsa sys/arch/m88k/include
sys/arch/m88k/m88k sys/arch/macppc/conf
sys/arch/macppc/dev sys/arch/mips64/include
sys/arch/mips64/mips64 sys/arch/powerpc/powerpc
sys/arch/solbourne/solbourne sys/arch/sparc/conf
sys/arch/sparc/dev sys/arch/sparc/include
sys/arch/sparc/sparc sys/arch/sparc64/dev
sys/arch/sparc64/include sys/arch/sparc64/sparc64
sys/arch/vax/include sys/arch/vax/vax
sys/compat/ibcs2 sys/crypto
sys/dev sys/dev/acpi
sys/dev/i2c sys/dev/ic
sys/dev/isa sys/dev/mii
sys/dev/pci sys/dev/usb
sys/isofs/cd9660 sys/kern
sys/lib/libsa sys/net
sys/net80211 sys/netinet
sys/netinet6 sys/nfs
sys/scsi sys/sys
sys/ufs/ffs sys/ufs/ufs
usr.bin/bgplg usr.bin/calendar
usr.bin/cvs usr.bin/find
usr.bin/grep usr.bin/join
usr.bin/mg usr.bin/netstat
usr.bin/rcs usr.bin/tail
usr.bin/top usr.sbin/adduser
usr.sbin/ftp-proxy usr.sbin/hostapd
usr.sbin/hostated usr.sbin/ntpd
usr.sbin/ripd usr.sbin/sasyncd
usr.sbin/sensorsd usr.sbin/tcpdump
usr.sbin/vnconfig
== bin =============================================================== 01/12 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/bin
pax
~ ftree.c
> fts_read returning NULL and errno set is an error. ok ray@ (otto@)
== distrib =========================================================== 02/12 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/distrib
crunch
~ crunchgen/crunchgen.c ~ crunchide/crunchide.c
> fix usage messages (deraadt@)
notes
~ alpha/hardware
> duplicated line. ok miod@
> duplicated line. ok miod@ (nick@)
~ i386/hardware ~ macppc/hardware
~ sparc64/hardware
> Add the newly supported Meinberg radio clocks. (mbalmer@)
ramdisk
~ Makefile
> remove some junk (deraadt@)
== etc =============================================================== 03/12 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/etc
changelist
~ changelist
> add the files in /etc/mtree to changelist
> ok krw@ (msf@)
hostapd.conf
~ hostapd.conf
> add examples for the optional channel hopper (reyk@)
== games ============================================================= 04/12 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/games
boggle
~ boggle/bog.c
> fix program name in usage, use __progname (martin@)
gomoku
~ main.c
> fix typos in comments (verses->versus) (martin@)
mille
~ mille.c
> fix program name in usage, use __progname (martin@)
== gnu =============================================================== 05/12 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/gnu
lib/libstdc++
~ libstdc++/config/cpu/i486/atomicity.h
> Fix inline asm. From GCC PR 17856. Fixes gnash on amd64.
> pointed out by deanna@ (kettenis@)
usr.bin/gcc
~ gcc/config.gcc
> Make exception handling work on alpha.
> tested by naddy@, deanna@; ok brad@ (kettenis@)
== lib =============================================================== 06/12 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/lib
libc
~ gen/sysctl.3
> reword a sentence and kill some trailing whitespace; (jmc@)
~ net/inet_pton.c
> RFC4291 declares that IPv6 numeric representation must have no more than
> 4 hex digits between colons. deraadt ok (itojun@)
== libexec =========================================================== 07/12 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/libexec
libexec
~ Makefile
> assume that new elf architectures will have ld.so (deraadt@)
ld.so
~ Makefile
> no more seperate prebind (deraadt@)
== sbin ============================================================== 08/12 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/sbin
brconfig
~ brconfig.8
> it is not permitted to change the "hellotime" is in RSTP, but in STP mode.
> thanks to Edy (email at edylie dot net)
> ok jmc@
> oh spanning tree, oh spanning tree, ... (reyk@)
~ brconfig.c
> add -A to usage().
> thanks to Sven-Volker Nowarra (reyk@)
~ brconfig.c
> -A before -a; (jmc@)
dhclient
~ dhclient.c
> There must be no alias address set, so check ifi->client->alias
> for NULL before dereferencing it. OK henning@ krw@ deraadt@ (moritz@)
~ bpf.c ~ clparse.c
~ dhclient.c ~ dhcpd.h
~ dispatch.c ~ options.c
~ packet.c
> Our dhclient only handles one interface. Simplify code by using the
> single, global, instance of 'struct interface_info' (ifi) rather than
> passing around pointers to it.
> "I agree" henning@
> Testing moritz@ stevesk@, ok stevesk@ (krw@)
~ dhclient.c
TAGGED OPENBSD_4_0
> MFC:
> Fix by moritz@
> There must be no alias address set, so check ifi->client->alias
> for NULL before dereferencing it. OK henning@ krw@ deraadt@ (moritz@)
~ clparse.c ~ dhclient.c
~ dhcpd.h
TAGGED OPENBSD_4_0
> Since we only have the single global struct interface_info, we
> obviously have single global struct client_state and struct
> client_config nested inside interface_info and client_state
> respectively.
> Pull them out to their own global variables and eliminate a bunch
> (i.e. hundreds) of multi-level dereferences. Make it much nicer to
> read.
> No functional change.
> "makes sense" henning@
> Testing stevesk@, ok stevesk@ (krw@)
~ dhclient.c
TAGGED OPENBSD_4_0
> no extra "\n" in note(); ok krw@ henning@ (stevesk@)
~ dhclient.c ~ packet.c
TAGGED OPENBSD_4_0
> small knf (extra space, newline) (stevesk@)
~ dhclient.c
TAGGED OPENBSD_4_0
> parenthesis not needed here; consistent with usage in other areas.
> no binary change. (stevesk@)
~ dhclient.c
TAGGED OPENBSD_4_0
> please lint (deraadt@)
dmesg
~ dmesg.c
> zap incorrect comment (djm@)
dump
~ traverse.c
> fts_read returning NULL and errno set is an error. ok ray@ (otto@)
pfctl
~ Makefile
> Remove m88k compiler flags tweak which is no longer necessary since a long
> time
> already. (miod@)
route
~ show.c
> Route warns about a illegal prefixlen when a default inet6 route is
> present.
> Need to typecast the sizeof() to int so that the compare works. Having a
> negativ lim is OK. Reported by Bjvrn Ketelaars. OK deraadt@ (claudio@)
sysctl
~ sysctl.c
> Show more digits in timedelta sensor output as the timedelta sensors become
> increasingly more precise.
> ok deraadt (mbalmer@)
== share ============================================================= 09/12 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/share
man
~ man9/sensor_attach.9
> typos; Murenin (deraadt@)
~ man4/acpi.4
> Uncomment a sentence about acpitz(4) shutting down the system if a
> critical temperature is detected. (mk@)
~ man4/adt.4
> support adt7475; tested by dhartmei (deraadt@)
~ man5/bsd.port.mk.5
> FULL_DISTDIR -> FULLDISTDIR (steven@)
~ man4/umsm.4
> sync with reality, add Sierra Wireless mc8755, while here clean up the
> listing.
> discussed and ok jsg@ (fkr@)
~ man4/usb.4
> umsm(4) sync; (jmc@)
~ man9/sensor_attach.9
> tweaks; (jmc@)
~ man9/namei.9
> s/vfs_lookup/VOP_LOOKUP/ since the former does not
> exist and we are talking about the latter.
> ok pedro@ (thib@)
~ man4/mbg.4
> Add support for the Meinberg PCI511 and GPS170PCI cards.
> "put it in" deraadt, ok fkr (mbalmer@)
~ man4/mbg.4
> tweaks; ok mbalmer (jmc@)
~ man4/malo.4
> Place right firmware names in FILES section for now. We still need to
> find a better name somewhen. (mglocker@)
~ man4/bcw.4
> Add URL and names for the new BCM43xx firmware package. Also here,
> probably not final naming. (mglocker@)
~ man8/netstart.8
> - newly-created -> newly created
> - hostname.if(5) and bridgename.if(5) are the same files, so don;t list
> them both in SEE ALSO
> - +.Xr ifconfig 8 (jmc@)
mk
~ bsd.xorg.mk
> - XORG_PREFIX is now X11BASE
> - define X11ETC here to avoid bootstrap problems noticed by Pierre-Yves
> Dampure and Frederick C. Druseikis. (matthieu@)
== sys =============================================================== 10/12 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/sys
arch/alpha/alpha
~ trap.c
> Define PROC_PC. Then, since profiling information is being reported in
> statclock(), do not bother doing this in userret() anymore. As a result,
> userret() does not need its pc and ticks arguments, simplify. (miod@)
arch/alpha/include
~ cpu.h
> Define PROC_PC. Then, since profiling information is being reported in
> statclock(), do not bother doing this in userret() anymore. As a result,
> userret() does not need its pc and ticks arguments, simplify. (miod@)
arch/amd64/include
~ cpu.h
> Define PROC_PC. Then, since profiling information is being reported in
> statclock(), do not bother doing this in userret() anymore. As a result,
> userret() does not need its pc and ticks arguments, simplify. (miod@)
arch/amd64/stand/libsa
~ gidt.S
> Fix another 'tripple' (not surprising, since this was copied from the
> i386 code that I fixed earlier). Noticed by Gregory Steuck (greg (at)
> nest (dot) cx); thanks. (tom@)
arch/arm/arm
~ ast.c ~ fault.c
~ syscall.c ~ undefined.c
> Define PROC_PC. Then, since profiling information is being reported in
> statclock(), do not bother doing this in userret() anymore. As a result,
> userret() does not need its pc and ticks arguments, simplify. (miod@)
~ sigcode.S ~ syscall.c
> Switch new new syscall ABI, now passes the syscall number in r12 instead of
> in the swi instruction. elminates an illegal userland reference from the
> kernel
> also elminates a bunch of code. This is a flag day for binaries from
> before the middle of November. (drahn@)
arch/arm/include
~ cpu.h
> Define PROC_PC. Then, since profiling information is being reported in
> statclock(), do not bother doing this in userret() anymore. As a result,
> userret() does not need its pc and ticks arguments, simplify. (miod@)
arch/hppa/hppa
~ trap.c
> Check for want_resched when processing AST and nowhere else. But then, when
> doing so, do not check for signals - userret() will do this. (miod@)
arch/i386/conf
~ GENERIC
> Change the comment for mbg(4): Use "radio clocks" instead of "time base".
> ok fkr (mbalmer@)
arch/i386/i386
~ trap.c
> Check for want_resched when processing AST and nowhere else. But then, when
> doing so, do not check for signals - userret() will do this. (miod@)
~ trap.c
> Define PROC_PC. Then, since profiling information is being reported in
> statclock(), do not bother doing this in userret() anymore. As a result,
> userret() does not need its pc and ticks arguments, simplify. (miod@)
~ freebsd_machdep.c ~ linux_machdep.c
~ mem.c ~ svr4_machdep.c
> Avoid void * arithmetic, okay deraadt@, suggestions from millert@ (pedro@)
arch/i386/include
~ cpu.h
> Define PROC_PC. Then, since profiling information is being reported in
> statclock(), do not bother doing this in userret() anymore. As a result,
> userret() does not need its pc and ticks arguments, simplify. (miod@)
arch/i386/isa
~ pccom.c ~ pccomvar.h
> Support for ST16C654 chips, however these aren't detected as such, so the
> attachment code has to know better for now;
> from Alexei G. Malinin (alexei.malinin at inetcomm.rumiod@)
~ isa_machdep.c
> Avoid void * arithmetic, okay deraadt@, suggestions from millert@ (pedro@)
arch/i386/stand/libsa
~ gidt.S
> There is a single 'p', not a double, in 'triple'. (tom@)
arch/m88k/include
~ cpu.h ~ db_machdep.h
> Define PROC_PC. Then, since profiling information is being reported in
> statclock(), do not bother doing this in userret() anymore. As a result,
> userret() does not need its pc and ticks arguments, simplify. (miod@)
arch/m88k/m88k
~ trap.c
> Check for want_resched when processing AST and nowhere else. But then, when
> doing so, do not check for signals - userret() will do this. (miod@)
~ trap.c
> Define PROC_PC. Then, since profiling information is being reported in
> statclock(), do not bother doing this in userret() anymore. As a result,
> userret() does not need its pc and ticks arguments, simplify. (miod@)
arch/macppc/conf
~ GENERIC
> Change the comment for mbg(4) to "radio clocks" here, too. (mbalmer@)
arch/macppc/dev
~ if_mc.c
> Some spaces to tabs and cosmetic tweaking of the code,
> From brad at .ts more spacing cleanup by myself.
> Tested on Power Macintosh 9600 (gwk@)
~ esp.c
> Cover the other case we need to unload the dmamap, in espdmaintr(),
> While here fix some spacing too.
> Bus probing and some commands such as eject tested on a Power
> Macintosh 9600, still unable to mount a cdrom on a SONY CD-ROM
> model CDU-8003A. (gwk@)
arch/mips64/include
~ cpu.h
> Define PROC_PC. Then, since profiling information is being reported in
> statclock(), do not bother doing this in userret() anymore. As a result,
> userret() does not need its pc and ticks arguments, simplify. (miod@)
arch/mips64/mips64
~ interrupt.c ~ trap.c
> Define PROC_PC. Then, since profiling information is being reported in
> statclock(), do not bother doing this in userret() anymore. As a result,
> userret() does not need its pc and ticks arguments, simplify. (miod@)
arch/powerpc/powerpc
~ trap.c
> Check for want_resched when processing AST and nowhere else. But then, when
> doing so, do not check for signals - userret() will do this. (miod@)
arch/solbourne/solbourne
~ trap.c
> Check for want_resched when processing AST and nowhere else. But then, when
> doing so, do not check for signals - userret() will do this. (miod@)
~ trap.c
> Define PROC_PC. Then, since profiling information is being reported in
> statclock(), do not bother doing this in userret() anymore. As a result,
> userret() does not need its pc and ticks arguments, simplify. (miod@)
arch/sparc/conf
~ files.sparc
> pnozz is initialized by the prom in 8bpp mode, so don't bring high-depth
> rasops code; 3.5KB off RAMDISK kernels. (miod@)
arch/sparc/dev
~ p9100.c
> Add code to change video mode (resolution and color depth) on the fly for
> pnozz(4), when switching between emulation (i.e. text console) and mapped
> (i.e. X11) modes.
> Geometries different than 800x600 are only available on the external video
> port, with the internal panel blanked.
> Currently this mode is compiled in, until an interface allows the X server
> and the kernel to settle on which one to use.
> Due to the internal panel blanking requirements, all of this is only
> available
> if tctrl(4) is configured in, which is the case for GENERIC but not for
> installation media kernels (who runs X11 with installation media anyway?)
> Most of this has been written 18 months ago, it was just lacking a final
> touch... (miod@)
arch/sparc/include
~ cpu.h
> Define PROC_PC. Then, since profiling information is being reported in
> statclock(), do not bother doing this in userret() anymore. As a result,
> userret() does not need its pc and ticks arguments, simplify. (miod@)
arch/sparc/sparc
~ trap.c
> Check for want_resched when processing AST and nowhere else. But then, when
> doing so, do not check for signals - userret() will do this. (miod@)
~ trap.c
> Define PROC_PC. Then, since profiling information is being reported in
> statclock(), do not bother doing this in userret() anymore. As a result,
> userret() does not need its pc and ticks arguments, simplify. (miod@)
arch/sparc64/dev
~ psycho.c ~ schizo.c
> change schizo/psycho interrupts to be xname:what instead of ugly names
> seen previously; discussed with kettenis (deraadt@)
arch/sparc64/include
~ cpu.h
> Define PROC_PC. Then, since profiling information is being reported in
> statclock(), do not bother doing this in userret() anymore. As a result,
> userret() does not need its pc and ticks arguments, simplify. (miod@)
~ bus.h
> Remove __BUS_SPACE_HAS_STREAM_METHODS, the stream methods are a netbsd-ism.
> discussed with miod@ who has the same changes hiding in one of his trees.
> (claudio@)
arch/sparc64/sparc64
~ trap.c
> Check for want_resched when processing AST and nowhere else. But then, when
> doing so, do not check for signals - userret() will do this. (miod@)
~ trap.c
> Define PROC_PC. Then, since profiling information is being reported in
> statclock(), do not bother doing this in userret() anymore. As a result,
> userret() does not need its pc and ticks arguments, simplify. (miod@)
~ locore.s
> Patch a few more D-cache flush instructions on UltraSPARC-III.
> ok miod@ (kettenis@)
~ cache.h ~ cpu.c
~ locore.s
> Give the UltraSPARC-III its own version of dcache_flush_page(). Fixes
> problems
> with svnd(4). Now you can do a full mkr on a v210 (and a blade1k if you're
> lucky). (kettenis@)
~ machdep.c
> Save pcb in dumppcb when dumping core.
> ok miod@ (kettenis@)
arch/vax/include
~ cpu.h
> Define PROC_PC. Then, since profiling information is being reported in
> statclock(), do not bother doing this in userret() anymore. As a result,
> userret() does not need its pc and ticks arguments, simplify. (miod@)
arch/vax/vax
~ trap.c
> Check for want_resched when processing AST and nowhere else. But then, when
> doing so, do not check for signals - userret() will do this. (miod@)
~ trap.c
> Define PROC_PC. Then, since profiling information is being reported in
> statclock(), do not bother doing this in userret() anymore. As a result,
> userret() does not need its pc and ticks arguments, simplify. (miod@)
compat/ibcs2
~ ibcs2_exec.c
> Avoid void * arithmetic, okay deraadt@, suggestions from millert@ (pedro@)
crypto
~ criov.c ~ cryptosoft.c
> Avoid void * arithmetic, okay deraadt@, suggestions from millert@ (pedro@)
dev
~ vnd.c
> ansify, no binary change, from Bret Lambert (pedro@)
dev/acpi
~ acpi.c ~ acpiac.c
~ acpibat.c ~ acpibtn.c
~ acpicpu.c ~ acpidev.h
~ acpitz.c ~ acpivar.h
~ dsdt.c ~ dsdt.h
> Add polling to devices that require it such as acpibat and acpitz.
> Use the same tsleep/wakeup mechanism as the interrupt code to keep it non-
> concurrent. (marco@)
~ acpiac.c
> Prettify dmesg a bit for attach with power state being either offline or
> unknown.
> ok marco (who already had this change in his tree but said I should
> commit) (mk@)
~ acpicpu.c
> Replace setperf algorithm with a linear version. Also use the aml provided
> gasio structure when appropriate instead of always assuming 32 bit access.
> Fixes my Dell laptop and works on others tested. (marco@)
dev/i2c
~ fintek.c
> normally hardware engineers call it Vcc not VCC (deraadt@)
~ i2c_scan.c
> identify adt7475 (deraadt@)
~ adt7460.c
> support adt7475; tested by dhartmei (deraadt@)
~ i2c_scan.c
> The adt7475 has a fixed address of 0x2e.
> ok deraadt@ (kettenis@)
~ adt7460.c
> correct the code to handle the adt7475, which stupidly is missing some
> of the registers.... tested by dhartmei and others (deraadt@)
dev/ic
~ fxp.c
> Don't re-initialize the card every time the IP address is changed.
> Brings fxp into line with many other drivers.
> From brad@, tested by martin@, thib@, pyr@, mk@ and Johan M:son
> Lindman.
> ok martin@ (krw@)
~ bcw.c ~ bcwreg.h
> Wrap bus space access in similiar macros to other drivers
> making things much more readable. (jsg@)
~ bcw.c
> spacing (jsg@)
~ p9000.h
> Add code to change video mode (resolution and color depth) on the fly for
> pnozz(4), when switching between emulation (i.e. text console) and mapped
> (i.e. X11) modes.
> Geometries different than 800x600 are only available on the external video
> port, with the internal panel blanked.
> Currently this mode is compiled in, until an interface allows the X server
> and the kernel to settle on which one to use.
> Due to the internal panel blanking requirements, all of this is only
> available
> if tctrl(4) is configured in, which is the case for GENERIC but not for
> installation media kernels (who runs X11 with installation media anyway?)
> Most of this has been written 18 months ago, it was just lacking a final
> touch... (miod@)
~ com_subr.c ~ comvar.h
> Support for ST16C654 chips, however these aren't detected as such, so the
> attachment code has to know better for now;
> from Alexei G. Malinin (alexei.malinin at inetcomm.rumiod@)
~ bcw.c
> spacing. (mglocker@)
~ bcw.c ~ bcwreg.h
~ bcwvar.h
> Finish core enumeration, break out core change and reset functions.
> From Jon Simola <simola at mecha.com Diff merged together by me with
> last changes from jsg at .glocker@)
~ twe.c
> Avoid void * arithmetic, okay deraadt@, suggestions from millert@ (pedro@)
~ malo.c
> No colon after "address". (mglocker@)
~ malo.c
> Why doing something as complex as for () bus_space_write_1(); when there
> is bus_space_write_region_1() that does the same. OK mglocker@ (claudio@)
~ re.c
> Add flow control support.
> ok brad@ (kettenis@)
~ acx.c ~ malo.c
~ pgt.c
> Don't use M_DUP_PKTHDR() on static mbufs. M_DUP_PKTHDR() copies the mtag
> chain and so a later MFREE() is needed to free the chain again.
> Just initialize a minimal mbuf header for bpf_mtap().
> See earlier rum(4) commit for the full story.
> OK mglocker@ Sounds good jsg@ (claudio@)
~ bcwreg.h
> spacing. (mglocker@)
~ malo.c
> Fix another printf where the format string did not match with the varargs.
> (claudio@)
~ acx.c
> Bye bye unused acx_shutdown() we won't miss you. (claudio@)
~ malo.c
> Use #ifdef MALO_DEBUG else a simple #define MALO_DEBUG will not work.
> (claudio@)
~ malo.c
> malo_stop() never fails so no need to return an error. OK mglocker@
> (claudio@)
~ malo.c
> Use a bus_space_barrier instead of those stupid 0x0c14 reads to make sure
> that access is done in correct order. OK mglocker@ (claudio@)
~ malo.c
> Add a debug printf to warn when malo_init() fails and resets the card.
> (claudio@)
~ malo.c
> In malo_init() check for firmware loading errors and in case of an error
> reset the card so that we have a chance that a later malo_init() works.
> OK mglocker@ (claudio@)
~ malo.c
> There is no need to busy wait in malo_send_cmd(). 0x0c14 only stores the
> current state of the card (not running, loading FW, running). Now instead
> wait after loading the boot firmware for the magical 0x5 to appear.
> Also add a few missing bus_dmamap_sync() calls. OK mglocker@ (claudio@)
dev/isa
~ sbdsp.c
> Avoid void * arithmetic, okay deraadt@, suggestions from millert@ (pedro@)
dev/mii
~ acphy.c ~ amphy.c
~ bmtphy.c ~ ciphy.c
~ dcphy.c ~ icsphy.c
~ inphy.c ~ iophy.c
~ lxtphy.c ~ nsgphy.c
~ nsphy.c ~ nsphyter.c
~ qsphy.c ~ rgephy.c
~ rlphy.c ~ sqphy.c
~ tlphy.c ~ tqphy.c
~ ukphy_subr.c ~ urlphy.c
> Always explicitly set IFM_HDX for half-duplex.
> From brad@ (kettenis@)
~ mii_physubr.c ~ miivar.h
> Add function to detect flow control status. From NetBSD.
> ok brad@ (kettenis@)
~ eephy.c
> Use mii_phy_flowstatus() to set flow control status bits.
> ok brad@ (kettenis@)
~ eephy.c
> No need to read ANAR and ANLPAR anymore.
> From brad@ (kettenis@)
~ ipgphy.c
> Use mii_phy_flowstatus() to replace the code that sets flag0/flag1 to
> report flow control status.
> From brad@ (kettenis@)
~ rgephy.c
> Add flow control support.
> ok brad@ (kettenis@)
~ mii_physubr.c
> Make mii_phy_auto() advertise flow control if MIIF_DOPAUSE is set.
> ok jsg@, brad@ (kettenis@)
~ ipgphy.c
> Only advertise flow control if MIIF_DOPAUSE is set.
> From brad@ (kettenis@)
~ eephy.c
> Rewrite eephy(4) to take advantage of generic phy code.
> Enable auto MDI/MDIX on PHY's that support it.
> Disable energy detect on PHY's that need it.
> tested on gem(4), msk(4), nfe(4), sk(4), stge(4) (kettenis@)
~ brgphy.c ~ ciphy.c
~ ipgphy.c ~ rgephy.c
> Set MASTER based on IFM_ETH_MASTER instead of LINK0.
> ok brad@ (kettenis@)
~ amphy.c ~ brgphy.c
~ ciphy.c ~ ipgphy.c
~ rgephy.c ~ xmphy.c
> Bring last few phys into line by calling their XXX_status() functions
> through mii_phy_status() rather than directly. No functional change.
> from brad@ ok mglocker@ (krw@)
dev/pci
~ if_bnx.c
> use the right size when loading the rx/tx descriptor bus dma maps.
> from the NetBSD port
> tested by bion@ and others from tech@
> ok marco@ brad@ (reyk@)
~ if_msk.c
> In msk_marv_miibus_writereg, wait for busy flag to clear instead of
> continuing when busy flag set. (kettenis@)
~ if_msk.c
> Make non-autonegotiated 10baseT and 100baseTX work. (kettenis@)
~ if_em.c
> Set IFM_ETH_MASTER if local PHY configuration resolved to MASTER.
> From brad@ (kettenis@)
~ if_msk.c
> Pass MIIF_DOPAUSE to mii_attach(). (kettenis@)
~ if_msk.c ~ if_mskvar.h
> Remove useless code dealing with media type. Simplify PHY reset code.
> (kettenis@)
~ pcidevs.h ~ pcidevs_data.h
> Regen (miod@)
~ pcidevs
> Cronyx Omega serial ports card. (miod@)
~ puc.c ~ pucdata.c
> Add support for the Cronyx Omega serial ports card; this particular card
> forces the com(4) uart type to ST16C654 since that's what's on it;
> from Alexei G. Malinin (alexei.malinin at inetcomm.rumiod@)
~ if_malo_pci.c
> Don't wrap MAC address to newline. (mglocker@)
~ mbg.c
> Add support for the Meinberg PCI511 and GPS170PCI cards.
> "put it in" deraadt, ok fkr (mbalmer@)
~ auich.c ~ maestro.c
~ ubsec.c ~ bktr/bktr_core.c
> Avoid void * arithmetic, okay deraadt@, suggestions from millert@ (pedro@)
~ mbg.c
> Sort product ids alphabetically like jmc did in the manpage. fkr also
> suggested that. (mbalmer@)
~ if_stge.c
> Pass MIIF_DOPAUSE to mii_attach().
> ok brad@ (kettenis@)
~ pcidevs.h ~ pcidevs_data.h
> Sync (marco@)
~ pcidevs
> Add new marvell sata controller. from brad@ (marco@)
~ agp.c
> In agp_generic_alloc_memory(), make sure the age_mem structure freshly
> malloc'ed is completely bzeroed before use.
> ok deraadt@ (miod@)
~ agp_i810.c
> In agp_i810_alloc_memory(), do not check for sc_maxmem limit reached, as
> the
> caller has already done this.
> Test matthieu@ ok deraadt@ (miod@)
~ if_msk.c
> Pass MIIF_FORCEANEG to mii_attack() until we stop doing unnecessary PHY
> hardware resets. (kettenis@)
~ if_msk.c
> Enable jumbo frames (except on the Yukon-2 FE).
> From brad@ (kettenis@)
~ mbg.c
> Use the same PCI product name strings we have in pcidevs_data.h for the
> timedelta sensor desc field and not the full description produced by
> pci_devinfo() (which gets truncated in this case). There is no need to
> have the PCi vendor string in the description as this is apparent from the
> driver name already.
> problem noticed by deraadt. (mbalmer@)
dev/usb
~ usbdevs.h
> regen with id from Sierra Wireless MC8755 (fkr@)
~ usbdevs
> add device id for Sierra Wireless MC8755, from janek at openbug.org > ok jsg (fkr@)
~ umsm.c
> make the Sierra Wireless MC8755 attach to umsm(4). this is a MiniPCI
> adapter that hosts a usb controller. from janek at openbug.org. > ok jsg@ (fkr@)
~ usbdevs.h ~ usbdevs_data.h
> regen (martin@)
~ usbdevs
> add IDs for Integrated System Solution Corp. KY-BT100 and another Belkin
> bluetooth (martin@)
~ if_rum.c
> Don't use M_DUP_PKTHDR() on static mbufs. M_DUP_PKTHDR() copies the mtag
> chain and so a later MFREE() is needed to free the chain again.
> In this special case we get away by initializing a minimal mbuf header
> instead of the M_DUP_PKTHDR() because bpf_mtap() does not access the
> pkthdr.
> This fixes kettenis@ mtag memory leak on armish.
> Tested by kettenis@ OK mglocker@ Sounds good jsg@ (claudio@)
isofs/cd9660
~ cd9660_vnops.c
> Avoid void * arithmetic, okay deraadt@, suggestions from millert@ (pedro@)
kern
~ subr_prof.c
> Keep a counter for addupc_intr() invocations, and pass its value, instead
> of 1,
> to addupc_task() in ADDUPROF(). From NetBSD via art at .iod@)
~ kern_clock.c
> If <machine/cpu.h> defines PROC_PC to compute the userland pc from an
> exception
> frame, use it to report profiling information via addupc_intr(). (miod@)
~ exec_elf.c ~ uipc_mbuf.c
> Avoid void * arithmetic, okay deraadt@, suggestions from millert@ (pedro@)
lib/libsa
~ loadfile.c
> do not load unneeded stringtab sections; avoids backward seeking; from
> netbsd; tested by miod@ and martin@; deraadt@ ok (mickey@)
net
~ if_media.h
> Add IFM_ETH_RXPAUSE and IFM_ETH_TXPAUSE. From NetBSD.
> ok brad@, deraadt@ (kettenis@)
~ if_bridge.c ~ if_gif.c
~ if_ppp.c ~ if_pppoe.c
> check if ifqueue has anything queued before doing the dance of
> splnet/IF_DEQUEUE/splx; ok various people (deraadt@)
~ if_pppoe.c
> knf, from bred@ (todd@)
~ if_media.h
> Add IFM_ETH_FMASK. From NetBSD.
> Requested by brad@ (kettenis@)
net80211
~ ieee80211.c ~ ieee80211_input.c
~ ieee80211_output.c ~ ieee80211_proto.c
~ ieee80211_proto.h
> print the phy mode in IFF_DEBUG status output which helps to debug
> dual/multi mode chipset operation.
> ok jsg@ (reyk@)
~ ieee80211_ioctl.c ~ ieee80211_crypto.c
> fix the key buffer size used for software wep, this could cause
> problems with non-standard wep keys >= 104 bits.
> thanks to Alexander Bluhm
> ok mglocker@ jsg@ (reyk@)
netinet
~ ip_input.c
> check if ifqueue has anything queued before doing the dance of
> splnet/IF_DEQUEUE/splx; ok various people (deraadt@)
netinet6
~ route6.c
> correct rhlen computation (if the "if" clause on the next line is taken,
> we will use rhlen uninitialized). checked with kame (itojun@)
~ ip6_input.c
> check if ifqueue has anything queued before doing the dance of
> splnet/IF_DEQUEUE/splx; ok various people (deraadt@)
nfs
~ nfs_subs.c ~ nfs_vnops.c
> Avoid void * arithmetic, okay deraadt@, suggestions from millert@ (pedro@)
scsi
~ scsiconf.c
> Avoid void * arithmetic, okay deraadt@, suggestions from millert@ (pedro@)
sys
~ resourcevar.h
> Keep a counter for addupc_intr() invocations, and pass its value, instead
> of 1,
> to addupc_task() in ADDUPROF(). From NetBSD via art at .iod@)
~ sensors.h
> consistent naming of units, percentage -> percent.
> ok deraadt (mbalmer@)
ufs/ffs
~ ffs_vfsops.c ~ ffs_alloc.c
> Zap nextgennumber, unused since LFS went up into the attic.
> ok pedro@ (thib@)
ufs/ufs
~ ufs_inode.c
> Zap nextgennumber, unused since LFS went up into the attic.
> ok pedro@ (thib@)
~ ufs_vnops.c
> Avoid void * arithmetic, okay deraadt@, suggestions from millert@ (pedro@)
== usr.bin =========================================================== 11/12 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/usr.bin
bgplg
~ bgplg.h
> "show ip bgp source-as" should issue a bgpctl show ip bgp source-as and not
> just a show ip bgp as. OK reyk@ (claudio@)
calendar
~ calendars/ru_RU.KOI8-R/calendar.history
> Typo. (grange@)
cvs
~ commit.c
> minor nits; (xsa@)
~ admin.c
> add support for -m option. (xsa@)
~ admin.c
> client-side bits might be useful too for -m... (xsa@)
find
~ find.c
> fts_read returning NULL and errno set is an error. ok ray@ (otto@)
grep
~ util.c
> fts_read returning NULL and errno set is an error. ok ray@ (otto@)
join
~ join.1
> - rearrange text
> - sort options
> - some markup fixes (jmc@)
mg
~ display.c
> Add column-number indicator to status line (enabled when line-number-mode
> is
> enabled; i.e. by default). This is essentially free, and has been requested
> a few times now (kjell@)
~ file.c ~ fileio.c
~ line.c
> Fix a bug where inserting a file resulted in an incorrect
> line-number count for a buffer (M-X insert-file, M-> to reproduce).
> While here, fix a number of bugs with incorrect line numbers
> after swap point-and-mark
> Originally reported via debian's bug tracking system. Fix tested by
> Han Boetes and Deanna Phillips. (kjell@)
~ mg.1
> mention meta-key-mode in mg(1) to enable 8-bit character input. this
> is required if you need to type characters like german umlauts. the mg
> documentation is still very incomplete and will need some more work.
> also split the manpage into sections to make it a bit more readable.
> ok jmc@ (reyk@)
~ extend.c
> one more 'verses'->'versus' typo in comment (martin@)
netstat
~ show.c
> Netstat has the same issue problem as route when printing a default inet6
> route. Need to typecast the sizeof() to int so that the compare works.
> Having a negativ lim is OK. Reported by Bjvrn Ketelaars. OK deraadt@
> (claudio@)
rcs
~ rcs.1
> - sync with usage (niallo@)
~ rcsprog.c
> - no need to mention -e option twice in usage.
> pointed out by Igor Sobrado <igor at string1.ciencias.uniovi.es> (niallo@)
~ Makefile
> - remove un-needed trailing backslash from LINKS declaration.
> noticed by Igor Sobrado <igor at string1.ciencias.uniovi.es> (niallo@)
tail
~ tail.1
> - less macros needed in SYNOPSIS
> - sort options (jmc@)
top
~ top.1 ~ top.c
> Also provide 'C' as a command line arg. From Mark Lumsden <mark at
> cyodesigns
> dot com>. (otto@)
== usr.sbin ========================================================== 12/12 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/usr.sbin
adduser
~ adduser.perl
> s/form/from/
> ``Sure.'' millert@, ``Sure'' espie at .ay@)
ftp-proxy
~ ftp-proxy.8 ~ ftp-proxy.c
> Convert three instances of atoi() to strtonum() and apply sane upper
> bounds.
> Triggered by Rik/harry Bobbaers on bugs at . > ok mbalmer@ ray@ (camield@)
~ ftp-proxy.c
> Remove stub write callback functions now that libevent allows them to be
> NULL. (camield@)
hostapd
~ apme.c ~ hostapd.conf.5
~ hostapd.h ~ parse.y
~ privsep.c
> add a channel hopper for wireless interfaces. the channel hopper will
> jump to the next available channel after a configurable delay for the
> specified list of wireless "hostap" interfaces, see the new "set
> hostap hopper" commands in hostapd.conf(5).
> with help by jsg@ (reyk@)
hostated
~ check_http.c ~ check_icmp.c
~ check_tcp.c ~ hce.c
~ hostated.conf.5 ~ hostated.h
~ parse.y
> partial rewrite of the check_* routines to use libevent everywhere
> instead of nested select() calls and to handle the non-blocking
> sockets properly.
> From Pierre-Yves Ritschard (pyr at spootnik dot org)
> (with a little help by me) (reyk@)
~ hostated.c
> knf (reyk@)
~ check_http.c ~ check_icmp.c
~ check_tcp.c ~ hostated.h
~ parse.y
> fix the conversion from milliseconds to struct timeval, which uses
> seconds (tv_sec) and microseconds (tv_usec), but the code assumed
> seconds and milliseconds... (reyk@)
~ hostated.conf.5
> the global timeout for checks is specified in milliseconds (reyk@)
~ hostated.h
> Add missing $OpenBSD$ (jsg@)
ntpd
~ sensors.c
> forgot a dereference in the previous conversion; ok deraadt@ (naddy@)
ripd
~ neighbor.c
> typo (michele@)
sasyncd
~ conf.y ~ sasyncd.c
~ sasyncd.h
> first pass cleanup of sasyncd, based on some discussion with deraadt@
> inline conf_init into main() and remove it from conf.y. add usage().
> small amount of whitespace nits in sasync.h
> ok deraadt@ mcbride@ (msf@)
~ sasyncd.h ~ monitor.c
> spacing (deraadt@)
~ sasyncd.c
> good day to remove a some poo (deraadt@)
~ sasyncd.c
> make option processing happen first.
> ok deraadt@ (msf@)
sensorsd
~ sensorsd.conf.5
> tweak; (jmc@)
tcpdump
~ print-ipsec.c
> print icmpv6 inside security header (ah/esp). (itojun@)
~ print-ipsec.c
> Fix truncation check by directly comparing both values and don't
> compare the unsigned value of the subtraction against <= 0.
> ok krw@ ray@ (moritz@)
vnconfig
~ vnconfig.c
> make -k and -K mutually exclusive, and ignore multiple invocations of them.
> put pcks key initialization into a separate routine instead of bloating the
> getopt() loop.
> ok pedro@ deraadt@ tedu@ (grunk@)
===============================================================================
More information about the owc
mailing list