[owc] Weekly src changes ending 2008-04-20
OWC
auto at squish.net
Mon Apr 21 07:00:01 BST 2008
OpenBSD src changes summary for 2008-04-13 to 2008-04-20 inclusive
==================================================================
bin/mkdir distrib/miniroot
distrib/notes distrib/sets
etc/hosts.lpd etc/netstart
etc/security games/fortune
games/random gnu/usr.sbin/mkhybrid
lib/libc libexec/ftpd
libexec/identd libexec/tftp-proxy
regress/lib regress/sys
sbin sbin/disklabel
sbin/ifconfig sbin/mknod
sbin/routed sbin/wpa-psk
share/man sys/altq
sys/arch/amd64/amd64 sys/arch/amd64/conf
sys/arch/amd64/include sys/arch/amd64/stand/boot
sys/arch/amd64/stand/cdboot sys/arch/amd64/stand/libsa
sys/arch/amd64/stand/pxeboot sys/arch/armish/stand/boot
sys/arch/hppa/hppa sys/arch/hppa/stand/boot
sys/arch/hppa64/conf sys/arch/hppa64/dev
sys/arch/hppa64/hppa64 sys/arch/hppa64/include
sys/arch/hppa64/stand/boot sys/arch/i386/conf
sys/arch/i386/i386 sys/arch/i386/include
sys/arch/i386/stand/boot sys/arch/i386/stand/cdboot
sys/arch/i386/stand/libsa sys/arch/i386/stand/pxeboot
sys/arch/landisk/stand/boot sys/arch/macppc/dev
sys/arch/macppc/stand sys/arch/sgi/conf
sys/arch/sparc/conf sys/arch/sparc/dev
sys/arch/sparc64/conf sys/arch/sparc64/dev
sys/arch/sparc64/include sys/arch/sparc64/sparc64
sys/arch/zaurus/dev sys/conf
sys/ddb sys/dev/cardbus
sys/dev/i2c sys/dev/ic
sys/dev/microcode sys/dev/mii
sys/dev/pci sys/dev/sbus
sys/dev/usb sys/kern
sys/lib/libkern sys/lib/libsa
sys/miscfs/specfs sys/net
sys/net80211 sys/netatalk
sys/netinet sys/netinet6
sys/nfs usr.bin/aucat
usr.bin/awk usr.bin/cal
usr.bin/calendar usr.bin/cvs
usr.bin/mg usr.bin/pr
usr.bin/sendbug usr.bin/ssh
usr.sbin/config usr.sbin/dhcpd
usr.sbin/ftp-proxy usr.sbin/httpd
usr.sbin/lpr usr.sbin/ntpd
usr.sbin/ripd usr.sbin/rtadvd
usr.sbin/rtsold usr.sbin/tcpdump
== bin =============================================================== 01/13 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/bin
mkdir
~ mkdir.1
> mkfifo makes fifos, not directories; from Ben Calvert
> while here, make the description of -m more consistent for mkdir,
> mkfifo, and mknod.
> add a touch of .Ex for mkdir too. (jmc@)
== distrib =========================================================== 02/13 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/distrib
miniroot
~ install.sub
> force mode for hostname.* files to 600; ok krw (deraadt@)
notes
~ sparc/hardware ~ sparc64/hardware
> It turns out the mysterious chip with the heatsink glued on, found on mgx
> video boards, is a known entity, so add support for 2d acceleration with
> the help of the X code (which makes baby Jesus cry).
> The blitter is fast enough to make a cgsix feel as slow as a cgthree in
> comparison. (miod@)
sets
~ lists/base/mi ~ lists/man/mi
> sync (deraadt@)
~ lists/base/md.alpha ~ lists/base/md.amd64
~ lists/base/md.armish ~ lists/base/md.aviion
~ lists/base/md.hp300 ~ lists/base/md.hppa
~ lists/base/md.hppa64 ~ lists/base/md.i386
~ lists/base/md.landisk ~ lists/base/md.luna88k
~ lists/base/md.mac68k ~ lists/base/md.macppc
~ lists/base/md.mvme68k ~ lists/base/md.mvme88k
~ lists/base/md.mvmeppc ~ lists/base/md.sgi
~ lists/base/md.sparc ~ lists/base/md.sparc64
~ lists/base/md.vax ~ lists/base/md.zaurus
~ lists/comp/mi
> sync (deraadt@)
~ lists/man/mi
> sync (deraadt@)
== etc =============================================================== 03/13 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/etc
hosts.lpd
~ hosts.lpd
> - make it clear that hosts.lpd follows the same syntax that hosts.equiv(5)
> inputs from jmc@
> "looks ok" beck@ (similar older diff), "fine by me" jmc@ (ajacoutot@)
netstart
~ netstart
> before using them, force hostname.* files to be unreadable by world
> first version from todd, ok millert (deraadt@)
security
~ security
> Teach security(8) to check for world-readable hostname.if files.
> An increasing number of types of these files (e.g. ppp, carp and
> wlan adapters) may contain secrets.
> ok deraadt oga johan (sthen@)
== games ============================================================= 04/13 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/games
fortune
~ fortune/fortune.c
> Use arc4random_buf() when requesting more than a single word of output
> Use arc4random_uniform() when the desired random number upper bound
> is not a power of two
> ok deraadt@ millert@ (djm@)
random
~ random.c
> Use arc4random_buf() when requesting more than a single word of output
> Use arc4random_uniform() when the desired random number upper bound
> is not a power of two
> ok deraadt@ millert@ (djm@)
== gnu =============================================================== 05/13 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/gnu
usr.sbin/mkhybrid
~ src/match.c ~ src/match.h
> Fix return value of i_add_match() and remove some bogus APPLE_HYB
> #ifdefs. Found by ragge@ (millert@)
== lib =============================================================== 06/13 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/lib
libc
~ stdlib/malloc.c
> Use arc4random_buf() when requesting more than a single word of output
> Use arc4random_uniform() when the desired random number upper bound
> is not a power of two
> ok deraadt@ millert@ (djm@)
~ net/res_random.c
> Improve the libc DNS resolver ID generation algorithm to be more
> resistant to prediction atacks by wrapping the existing LCG in a
> random permutation generator based on a Luby-Rackoff block cipher.
> lots of discussion and final ok deraadt@ (djm@)
~ net/res_comp.c
> permit _ in the middle of a DNS name component (deraadt@)
~ net/res_send.c
> convert select() => poll(), saves a runtime malloc+free per retry
> feedback deraadt@ drahn@; ok deraadt@ (djm@)
== libexec =========================================================== 07/13 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/libexec
ftpd
~ ftpd.c
> Use arc4random_buf() when requesting more than a single word of output
> Use arc4random_uniform() when the desired random number upper bound
> is not a power of two
> ok deraadt@ millert@ (djm@)
identd
~ parse.c
> Use arc4random_buf() when requesting more than a single word of output
> Use arc4random_uniform() when the desired random number upper bound
> is not a power of two
> ok deraadt@ millert@ (djm@)
tftp-proxy
~ tftp-proxy.c
> Use arc4random_buf() when requesting more than a single word of output
> Use arc4random_uniform() when the desired random number upper bound
> is not a power of two
> ok deraadt@ millert@ (djm@)
== regress =========================================================== 08/13 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/regress
lib
~ libc/malloc/malloc0test/malloc0test.c
> Use arc4random_buf() when requesting more than a single word of output
> Use arc4random_uniform() when the desired random number upper bound
> is not a power of two
> ok deraadt@ millert@ (djm@)
sys
~ kern/signal-stress/signal-stress.c ~ sys/tree/rb/rb-test.c
~ sys/tree/splay/splay-test.c
> Use arc4random_buf() when requesting more than a single word of output
> Use arc4random_uniform() when the desired random number upper bound
> is not a power of two
> ok deraadt@ millert@ (djm@)
== sbin ============================================================== 09/13 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/sbin
sbin
~ Makefile
> link wpa-psk(8) to the build (damien@)
disklabel
~ editor.c
> Redo the '?' output in (E)ditor mode so it uses two columns and once
> again fits on one page. Suggested by, tweaked and ok deraadt at . (krw@)
ifconfig
~ ifconfig.8 ~ ifconfig.c
> Kernel implementation of the 4-way handshake and group-key
> handshake protocols (both supplicant and authenticator state
> machines) as defined in the IEEE 802.11i standard.
> Software implementation of the TKIP (Temporal Key Integrity
> Protocol) and CCMP (CTR with CBC-MAC Protocol) protocols.
> This diff doesn't implement any of the 802.1X authentication
> protocols and thus only PSK authentication (using pre-shared
> keys) is currently supported.
> In concrete terms, this adds support for WPA-PSK and WPA2-PSK
> protocols, both in station and hostap modes.
> The following drivers are marked as WPA-capable and should
> work: bwi(4), malo(4), ral(4), iwn(4), wpi(4), ural(4),
> rum(4), upgt(4), and zyd(4)
> The following options have been added to ifconfig(8):
> wpa, wpapsk, wpaprotos, wpaakms, wpaciphers, wpagroupcipher
> wpa-psk(8) can be used to generate keys from passphrases.
> tested by many@
> ok deraadt@ (damien@)
~ ifconfig.8
> - wpa-psk is section 8, not 1
> - one more passphrase tweak (jmc@)
~ ifconfig.8
> remove wparekey from the list of 802.11 ifconfig options.
> pointed out by drahn@ (damien@)
mknod
~ mkfifo.1 ~ mknod.8
> mkfifo makes fifos, not directories; from Ben Calvert
> while here, make the description of -m more consistent for mkdir,
> mkfifo, and mknod.
> add a touch of .Ex for mkdir too. (jmc@)
~ mkfifo.1 ~ mknod.8
> remove a word i misinterpreted; (jmc@)
routed
~ main.c
> Use arc4random_buf() when requesting more than a single word of output
> Use arc4random_uniform() when the desired random number upper bound
> is not a power of two
> ok deraadt@ millert@ (djm@)
wpa-psk
+ Makefile + wpa-psk.8
+ wpa-psk.c
> welcome wpa-psk(8).
> ok deraadt@ (damien@)
~ wpa-psk.8
> pass-phrase -> passphrase, in keeping with the rest of the docs; (jmc@)
== share ============================================================= 10/13 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/share
man
~ man4/uvideo.4
> some small fixes; (jmc@)
~ man9/mbuf.9
> MSIZE is defined in sys/param.h, not machine/param.h;
> from Jonathan Thornburg, documentation/5792 (jmc@)
~ man4/man4.sparc/mgx.4 ~ man4/man4.sparc64/mgx.4
> It turns out the mysterious chip with the heatsink glued on, found on mgx
> video boards, is a known entity, so add support for 2d acceleration with
> the help of the X code (which makes baby Jesus cry).
> The blitter is fast enough to make a cgsix feel as slow as a cgthree in
> comparison. (miod@)
~ man4/iic.4
> fix Xr (deraadt@)
~ man4/lmtemp.4
> mention LM76; stop being exact about increments (deraadt@)
~ man4/man4.sparc64/clock.4
> Mention m5819, m5823 and bq4802. (kettenis@)
~ man9/tsleep.9
> Correct a typo:
> tsleep -> msleep
> from Paul de Weerd, Thanks! (oga@)
~ man4/acx.4 ~ man4/an.4
~ man4/ath.4 ~ man4/atu.4
~ man4/atw.4 ~ man4/awi.4
~ man4/bwi.4 ~ man4/ipw.4
~ man4/iwi.4 ~ man4/iwn.4
~ man4/malo.4 ~ man4/pgt.4
~ man4/ral.4 ~ man4/ray.4
~ man4/rtw.4 ~ man4/rum.4
~ man4/uath.4 ~ man4/upgt.4
~ man4/ural.4 ~ man4/wi.4tbl
~ man4/wpi.4 ~ man4/zyd.4
> remove the CONFIGURATION section from the wireless pages:
> the addition of the wpa options have shown just how unwieldy it had
> become, it's already in ifconfig(8) anyway, and "ifconfig -m" will
> get the relevant ones.
> ok dlg jsg (jmc@)
~ man8/security.8
> Teach security(8) to check for world-readable hostname.if files.
> An increasing number of types of these files (e.g. ppp, carp and
> wlan adapters) may contain secrets.
> ok deraadt oga johan (sthen@)
~ man4/uvideo.4
> Add two cams. (mglocker@)
~ man4/Makefile
> install it (deraadt@)
~ man4/uvideo.4
> fix the list width; (jmc@)
~ man4/ahci.4
> add a driver flag to force the negotiation of SATA 1 transfers
> (1.5Gb/s). Useful where faster speeds are unstable; ok dlg@ (djm@)
~ man4/txp.4
> overhaul the page. (brad@)
~ man4/lge.4
> slight tweaking of the wording to make this consistent with other
> pages that mention checksum offload. (brad@)
~ man4/txp.4
> remove trailing whitespace; (jmc@)
== sys =============================================================== 11/13 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/sys
altq
~ altq_cdnr.c ~ altq_classq.h
~ altq_red.c ~ altq_rmclass.c
~ altq_var.h
> Do not use random(), which is supposed to be a scheduler-only API (we
> should rename it one day). Furthermore, do not do random() % value,
> but instead solve both problems using arc4random_uniform()
> Problem spotted by drahn, fix ok'd by djm (deraadt@)
arch/amd64/amd64
~ cpu.c
> garbage collect the cpus_running global variable;
> pointed out by and ok drahn@, toby@; (thib@)
~ ipifuncs.c
> lear CPUF_RUNNING when halting a CPU. Fixes problems with powering dowm
> MP amd64 systems for quite a few people.
> ok deraadt@, krw@ (kettenis@)
arch/amd64/conf
~ GENERIC
> add a driver flag to force the negotiation of SATA 1 transfers
> (1.5Gb/s). Useful where faster speeds are unstable; ok dlg@ (djm@)
arch/amd64/include
~ cpuvar.h
> garbage collect the cpus_running global variable;
> pointed out by and ok drahn@, toby@; (thib@)
arch/amd64/stand/boot
~ conf.c
> Change ELF loader to use the LMA as the load address for the
> various segments. Hopefully this will help remove various
> hacks in the boot loader in the future. This should have no
> effect on most architectures (as we tend to have LMA == VMA).
> ok drahn@, soft ok's various others. (weingart@)
arch/amd64/stand/cdboot
~ conf.c
> Change ELF loader to use the LMA as the load address for the
> various segments. Hopefully this will help remove various
> hacks in the boot loader in the future. This should have no
> effect on most architectures (as we tend to have LMA == VMA).
> ok drahn@, soft ok's various others. (weingart@)
arch/amd64/stand/libsa
~ bioscons.c
> rewrite the serial port handling to manipulate the hardware directly
> rather than relying on calls into the bios to work.
> this is a result of me getting pissed off with solaris and linux
> being able to cope the serial bios redirection on dracs, iloms, and
> ilos. trying to do the same thing with openbsds boot loader caused
> weird behaviour like machine hangs or no visible output. talking
> to the serial chips directly is more reliable.
> tested by many
> ok sthen@ deraadt@ (dlg@)
arch/amd64/stand/pxeboot
~ conf.c
> Change ELF loader to use the LMA as the load address for the
> various segments. Hopefully this will help remove various
> hacks in the boot loader in the future. This should have no
> effect on most architectures (as we tend to have LMA == VMA).
> ok drahn@, soft ok's various others. (weingart@)
arch/armish/stand/boot
~ conf.c
> Change ELF loader to use the LMA as the load address for the
> various segments. Hopefully this will help remove various
> hacks in the boot loader in the future. This should have no
> effect on most architectures (as we tend to have LMA == VMA).
> ok drahn@, soft ok's various others. (weingart@)
arch/hppa/hppa
~ pmap.c
> use arc4random_uniform() for random number requests that are not a
> power of two.
> use arc4random_bytes() when requesting more than a word of PRNG
> output.
> ok deraadt@ (djm@)
~ autoconf.c
> Print bootpath in the canonical place. (kettenis@)
arch/hppa/stand/boot
~ conf.c
> Change ELF loader to use the LMA as the load address for the
> various segments. Hopefully this will help remove various
> hacks in the boot loader in the future. This should have no
> effect on most architectures (as we tend to have LMA == VMA).
> ok drahn@, soft ok's various others. (weingart@)
arch/hppa64/conf
~ files.hppa64
> Sync console handling with hppa port:
> Delay switching to the real serial console until we attach the serial
> device.
> The variety of different serial devices is just too big to make a decision
> up
> front. This should make the second serial port on four-digit B/C/J-class
> workstation work as a serial console too.
> Last bit isn't true yet; the code to support com @ ssio isn't there yet on
> hppa64. (kettenis@)
~ GENERIC ~ files.hppa64
> crank maxusers (kettenis@)
~ files.hppa64
> Catch up with dismantled libkern. (kettenis@)
~ GENERIC ~ files.hppa64
> Bring over astro(4) from hppa, to replace pluto(4). (kettenis@)
arch/hppa64/dev
~ pdc.c
> Sync console handling with hppa port:
> Delay switching to the real serial console until we attach the serial
> device.
> The variety of different serial devices is just too big to make a decision
> up
> front. This should make the second serial port on four-digit B/C/J-class
> workstation work as a serial console too.
> Last bit isn't true yet; the code to support com @ ssio isn't there yet on
> hppa64. (kettenis@)
- pluto.c + astro.c
> Bring over astro(4) from hppa, to replace pluto(4). (kettenis@)
~ elroy.c ~ elroyreg.h
~ elroyvar.h
> Sync with hppa code. (kettenis@)
arch/hppa64/hppa64
~ machdep.c
> Sync console handling with hppa port:
> Delay switching to the real serial console until we attach the serial
> device.
> The variety of different serial devices is just too big to make a decision
> up
> front. This should make the second serial port on four-digit B/C/J-class
> workstation work as a serial console too.
> Last bit isn't true yet; the code to support com @ ssio isn't there yet on
> hppa64. (kettenis@)
~ conf.c
> Sync console handling with hppa port:
> Delay switching to the real serial console until we attach the serial
> device.
> The variety of different serial devices is just too big to make a decision
> up
> front. This should make the second serial port on four-digit B/C/J-class
> workstation work as a serial console too.
> Last bit isn't true yet; the code to support com @ ssio isn't there yet on
> hppa64. (kettenis@)
~ genassym.cf ~ locore.S
~ vm_machdep.c
> Catch up with cpu_switchto. Completely untested. (kettenis@)
~ mainbus.c
> Add proper bus_space_vaddr(9) impementation. (kettenis@)
~ autoconf.c
> Add missing function declaration. (kettenis@)
~ machdep.c
> Try to reset the machine through the firmware to reboot. (kettenis@)
arch/hppa64/include
~ param.h
> Sync console handling with hppa port:
> Delay switching to the real serial console until we attach the serial
> device.
> The variety of different serial devices is just too big to make a decision
> up
> front. This should make the second serial port on four-digit B/C/J-class
> workstation work as a serial console too.
> Last bit isn't true yet; the code to support com @ ssio isn't there yet on
> hppa64. (kettenis@)
~ bus.h
> Add proper bus_space_vaddr(9) impementation. (kettenis@)
arch/hppa64/stand/boot
~ conf.c
> Change ELF loader to use the LMA as the load address for the
> various segments. Hopefully this will help remove various
> hacks in the boot loader in the future. This should have no
> effect on most architectures (as we tend to have LMA == VMA).
> ok drahn@, soft ok's various others. (weingart@)
arch/i386/conf
~ GENERIC
> add a driver flag to force the negotiation of SATA 1 transfers
> (1.5Gb/s). Useful where faster speeds are unstable; ok dlg@ (djm@)
arch/i386/i386
~ ipifuncs.c
> lear CPUF_RUNNING when halting a CPU. Fixes problems with powering dowm
> MP amd64 systems for quite a few people.
> ok deraadt@, krw@ (kettenis@)
~ machdep.c
> Halt CPUs upon reboot/halt.
> ok krw@ (kettenis@)
~ apicvec.s ~ ipifuncs.c
~ lapic.c ~ machdep.c
> Now that i386 has a per-process astpending, we can garbage collect ipi_ast
> and do an ipi_nop cross-call from signotify() instead.
> ok miod@ (kettenis@)
~ machdep.c
> Call cnpollc() before cngetc() upon halt.
> ok miod@ (kettenis@)
arch/i386/include
~ cpu.h
> garbage collect the cpus_running global variable;
> pointed out by and ok drahn@, toby@; (thib@)
~ cpu.h ~ i82489var.h
~ intrdefs.h
> Now that i386 has a per-process astpending, we can garbage collect ipi_ast
> and do an ipi_nop cross-call from signotify() instead.
> ok miod@ (kettenis@)
arch/i386/stand/boot
~ conf.c
> Change ELF loader to use the LMA as the load address for the
> various segments. Hopefully this will help remove various
> hacks in the boot loader in the future. This should have no
> effect on most architectures (as we tend to have LMA == VMA).
> ok drahn@, soft ok's various others. (weingart@)
arch/i386/stand/cdboot
~ conf.c
> Change ELF loader to use the LMA as the load address for the
> various segments. Hopefully this will help remove various
> hacks in the boot loader in the future. This should have no
> effect on most architectures (as we tend to have LMA == VMA).
> ok drahn@, soft ok's various others. (weingart@)
arch/i386/stand/libsa
~ bioscons.c
> rewrite the serial port handling to manipulate the hardware directly
> rather than relying on calls into the bios to work.
> this is a result of me getting pissed off with solaris and linux
> being able to cope the serial bios redirection on dracs, iloms, and
> ilos. trying to do the same thing with openbsds boot loader caused
> weird behaviour like machine hangs or no visible output. talking
> to the serial chips directly is more reliable.
> tested by many
> ok sthen@ deraadt@ (dlg@)
arch/i386/stand/pxeboot
~ conf.c
> Change ELF loader to use the LMA as the load address for the
> various segments. Hopefully this will help remove various
> hacks in the boot loader in the future. This should have no
> effect on most architectures (as we tend to have LMA == VMA).
> ok drahn@, soft ok's various others. (weingart@)
arch/landisk/stand/boot
~ conf.c
> Change ELF loader to use the LMA as the load address for the
> various segments. Hopefully this will help remove various
> hacks in the boot loader in the future. This should have no
> effect on most architectures (as we tend to have LMA == VMA).
> ok drahn@, soft ok's various others. (weingart@)
arch/macppc/dev
~ kiic.c
> if an iic operation fails to move all the data, return error
> ok kettenis (deraadt@)
arch/macppc/stand
~ conf.c
> Change ELF loader to use the LMA as the load address for the
> various segments. Hopefully this will help remove various
> hacks in the boot loader in the future. This should have no
> effect on most architectures (as we tend to have LMA == VMA).
> ok drahn@, soft ok's various others. (weingart@)
arch/sgi/conf
~ RAMDISK
> typo in attachment (deraadt@)
arch/sparc/conf
~ Makefile.sparc
> Oops, it's a bit too early for -Wvariable-decl on sparc. (miod@)
arch/sparc/dev
~ mgx.c
> Fix braino in mgx_loadcmap() invocation. (miod@)
~ mgx.c
> It turns out the mysterious chip with the heatsink glued on, found on mgx
> video boards, is a known entity, so add support for 2d acceleration with
> the help of the X code (which makes baby Jesus cry).
> The blitter is fast enough to make a cgsix feel as slow as a cgthree in
> comparison. (miod@)
arch/sparc64/conf
~ GENERIC
> add a driver flag to force the negotiation of SATA 1 transfers
> (1.5Gb/s). Useful where faster speeds are unstable; ok dlg@ (djm@)
arch/sparc64/dev
~ rtc.c
> Add support for the Texas Instruments bq4802 real-time clock found on the
> Ultra 25 and Ultra 45. (kettenis@)
~ rtc.c
> Turns out getpropstring() returns an empty string if a property is missing
> instead of NULL. Makes rtc(4) really work on the u25.
> tested by mikeb@ (kettenis@)
~ pcfiic_ebus.c
> Support the pcf8584 variant found on the u25/45. (kettenis@)
arch/sparc64/include
~ cpu.h
> Use %g7 to store a pointer `struct cpu_info', and use it whereever
> possible. (kettenis@)
~ ctlreg.h
> Add workaround for UltraSPARC-II errata, where writes to %tick_cmpr would
> sometimes fail, which would result in the periodic clock interrupts on a
> CPU
> stop.
> Spotted in a NetBSD commit message, loosely based on code in OpenSolaris.
> (kettenis@)
~ cpu.h
> Really try to schedule clock ticks at fixed intervals. Make sure
> hardclock()
> gets called for every clock tick, even if we miss one. (kettenis@)
~ pte.h
> Implement locking of sun4u TSB entries for MULTIPROCESSOR kernels.
> Make sun4v code use the new TSB_TAG_LOCKED define. (kettenis@)
arch/sparc64/sparc64
~ locore.s ~ mutex.S
> Use %g7 to store a pointer `struct cpu_info', and use it whereever
> possible. (kettenis@)
~ autoconf.c ~ locore.s
> Introduce macros to switch to normal and alternate globals and switch to
> use the .section based mechanism to patch them up for sun4v. (kettenis@)
~ clock.c ~ locore.s
> Add workaround for UltraSPARC-II errata, where writes to %tick_cmpr would
> sometimes fail, which would result in the periodic clock interrupts on a
> CPU
> stop.
> Spotted in a NetBSD commit message, loosely based on code in OpenSolaris.
> (kettenis@)
~ locore.s
> Make sure tickcmpr_set() always writes a time in the future. (kettenis@)
~ clock.c
> Really try to schedule clock ticks at fixed intervals. Make sure
> hardclock()
> gets called for every clock tick, even if we miss one. (kettenis@)
~ machdep.c
> use arc4random_uniform() for random number requests that are not a
> power of two.
> use arc4random_bytes() when requesting more than a word of PRNG
> output.
> ok deraadt@ (djm@)
~ locore.s ~ genassym.cf
> Implement locking of sun4u TSB entries for MULTIPROCESSOR kernels.
> Make sun4v code use the new TSB_TAG_LOCKED define. (kettenis@)
~ locore.s ~ machdep.c
> Remove the random() function from locore.s (which used sparcv7
> instructions,
> including mulscc to do multiplications) and switch to the generic random.c
> code. (kettenis@)
arch/zaurus/dev
~ zaurus_flash.c
> use arc4random_uniform() for random number requests that are not a
> power of two.
> use arc4random_bytes() when requesting more than a word of PRNG
> output.
> ok deraadt@ (djm@)
conf
~ files
> Kernel implementation of the 4-way handshake and group-key
> handshake protocols (both supplicant and authenticator state
> machines) as defined in the IEEE 802.11i standard.
> Software implementation of the TKIP (Temporal Key Integrity
> Protocol) and CCMP (CTR with CBC-MAC Protocol) protocols.
> This diff doesn't implement any of the 802.1X authentication
> protocols and thus only PSK authentication (using pre-shared
> keys) is currently supported.
> In concrete terms, this adds support for WPA-PSK and WPA2-PSK
> protocols, both in station and hostap modes.
> The following drivers are marked as WPA-capable and should
> work: bwi(4), malo(4), ral(4), iwn(4), wpi(4), ural(4),
> rum(4), upgt(4), and zyd(4)
> The following options have been added to ifconfig(8):
> wpa, wpapsk, wpaprotos, wpaakms, wpaciphers, wpagroupcipher
> wpa-psk(8) can be used to generate keys from passphrases.
> tested by many@
> ok deraadt@ (damien@)
ddb
~ db_hangman.c
> use arc4random_uniform() for random number requests that are not a
> power of two.
> use arc4random_bytes() when requesting more than a word of PRNG
> output.
> ok deraadt@ (djm@)
dev/cardbus
~ if_re_cardbus.c
> Remove unused flags.
> ok dlg@ (brad@)
dev/i2c
~ lm75.c
> Take a shot at LM76 support (13 bit temperature, 0.065 degC per LSB) for
> when we finally find one of these devices. looked at by kettenis..
> (deraadt@)
~ lm75.c
> Use ntohs() to swap the raw i2c data to native format; tested by sthen
> (deraadt@)
~ pca9532.c ~ rs5c372.c
> use sizeof(thing) instead of hard-coding it (deraadt@)
~ i2c_exec.c
> use sizeof(thing) instead of hard-coding it (deraadt@)
~ ad741x.c ~ asb100.c
~ ds1631.c ~ i2c_scan.c
~ lm75.c ~ sdtemp.c
~ spdmem.c
> make it more apparent that when we are dealing with 16 bit registers
> (not a series of 8 bit registers), the bytes come off the wire in
> big-endian order (deraadt@)
~ tsl2560.c
> correct sizeof; palmeida (deraadt@)
~ i2c_scan.c
> adm1033 has a rev register to check as well (deraadt@)
~ i2c_scan.c
> detect adm1034 (deraadt@)
~ i2c_scan.c
> two more i2c probe checks. adt has split off or been sold to ON
> semi, so now even more data sheets are available at
> http://www.onsemi.com/PowerSolutions/parametrics.do?id=2127
> so if people want to help write drivers... (deraadt@)
dev/ic
+ atxxreg.h
> It turns out the mysterious chip with the heatsink glued on, found on mgx
> video boards, is a known entity, so add support for 2d acceleration with
> the help of the X code (which makes baby Jesus cry).
> The blitter is fast enough to make a cgsix feel as slow as a cgthree in
> comparison. (miod@)
~ acx.c ~ atw.c
~ bwi.c ~ malo.c
~ pgt.c ~ rt2560.c
~ rt2661.c ~ rt2860.c
~ rt2860reg.h ~ rtw.c
> Kernel implementation of the 4-way handshake and group-key
> handshake protocols (both supplicant and authenticator state
> machines) as defined in the IEEE 802.11i standard.
> Software implementation of the TKIP (Temporal Key Integrity
> Protocol) and CCMP (CTR with CBC-MAC Protocol) protocols.
> This diff doesn't implement any of the 802.1X authentication
> protocols and thus only PSK authentication (using pre-shared
> keys) is currently supported.
> In concrete terms, this adds support for WPA-PSK and WPA2-PSK
> protocols, both in station and hostap modes.
> The following drivers are marked as WPA-capable and should
> work: bwi(4), malo(4), ral(4), iwn(4), wpi(4), ural(4),
> rum(4), upgt(4), and zyd(4)
> The following options have been added to ifconfig(8):
> wpa, wpapsk, wpaprotos, wpaakms, wpaciphers, wpagroupcipher
> wpa-psk(8) can be used to generate keys from passphrases.
> tested by many@
> ok deraadt@ (damien@)
~ rt2860.c
> do not blindly call ieee80211_get_hdrlen() in rt2860_rx_intr().
> we may end up passing control frames (ps-poll or others) which
> is not supported by ieee80211_get_hdrlen().
> first found by pedro la peu, reminded by jsg@
> closes kernel/5750 (damien@)
~ if_wi.c
> save some space by making use of the WEP CRC table from net80211
> and use the ieee80211_crc_update() function too.
> not tested as I do not have a wi(4) but I see no reason why it
> would not work.
> "Just be brave" deraadt@ (damien@)
~ if_wi.c
> extend the if_ethersubr.c crc functions to support updating a running
> crc in addition to the existing "oneshot" mode and use them to replace
> ieee80211_crc_update() with the new ether_crc32_le_update(). Saves 1k
> kernel bss + some code.
> Mark the new ether_crc32_[lb]e_update functions as __pure for a
> ~25x speedup (on my i386 at least).
> feedback and ok damien@ (djm@)
~ rtl81x9.c
> Don't bother to call rl_rxeof() twice if both the RX Ok and Err bits are
> set.
> Based on the same change to re(4) quite some time ago.
> ok dlg@ (brad@)
~ rtl81x9reg.h
> Remove unused flags.
> ok dlg@ (brad@)
~ re.c ~ rtl81x9reg.h
> - Add some more defines.
> - Use one of the new defines in the re(4) code to remove a magic number.
> - Correct a comment.
> - Fix a typo.
> Most of this is from NetBSD.
> ok dlg@ (brad@)
~ re.c
> Remove redundant checks for the 8139C+ chipset, this code can only
> be reached if using the 8139C+ chipset.
> ok dlg@ (brad@)
~ re.c ~ rtl81x9reg.h
> - Recognize the 8168C chipset.
> - Add some more defines for the 8168C chipset.
> From NetBSD
> ok dlg@ (brad@)
~ re.c ~ rtl81x9reg.h
> Move the threshold fix from the re(4) code back into the header as
> it was going to be before the release. No functional change.
> prodded by kettenis@ ok dlg@ (brad@)
dev/microcode
~ ral/microcode.h
> fix a comment. no binary change. (damien@)
dev/mii
~ brgphy.c
> Add a workaround for a CRC bug errata with BCM5701 A0 and B0 chipset
> revisions.
> From Linux via FreeBSD.
> ok dlg@ (brad@)
dev/pci
~ drm/drm_drv.c
> bring back a free call which was left out when moving from handrolled lists
> to tailqs. this plugs a tiny memleak.
> ``nice catch'' and ok oga@ (pyr@)
~ isp_pci.c
> Disable ISP10160 and ISP12160. They don't quite work yet, and people with
> another (working) isp(4) get hurt by this.
> ok krw@, deraadt@ (kettenis@)
~ drm/drmP.h ~ drm/drm_auth.c
~ drm/drm_drv.c
> Replace handrolled hashes with a splay tree, while there plug a memleak
> inherited from the original drm driver.
> input and ok oga at . (pyr@)
~ drm/drm_auth.c
> Some formatting nits. No functional change. (oga@)
~ if_ipw.c ~ if_ipwvar.h
~ if_iwn.c ~ if_wpi.c
> Kernel implementation of the 4-way handshake and group-key
> handshake protocols (both supplicant and authenticator state
> machines) as defined in the IEEE 802.11i standard.
> Software implementation of the TKIP (Temporal Key Integrity
> Protocol) and CCMP (CTR with CBC-MAC Protocol) protocols.
> This diff doesn't implement any of the 802.1X authentication
> protocols and thus only PSK authentication (using pre-shared
> keys) is currently supported.
> In concrete terms, this adds support for WPA-PSK and WPA2-PSK
> protocols, both in station and hostap modes.
> The following drivers are marked as WPA-capable and should
> work: bwi(4), malo(4), ral(4), iwn(4), wpi(4), ural(4),
> rum(4), upgt(4), and zyd(4)
> The following options have been added to ifconfig(8):
> wpa, wpapsk, wpaprotos, wpaakms, wpaciphers, wpagroupcipher
> wpa-psk(8) can be used to generate keys from passphrases.
> tested by many@
> ok deraadt@ (damien@)
~ pcidevs
> shorten name (deraadt@)
~ pcidevs.h ~ pcidevs_data.h
> sync (deraadt@)
~ ahci.c
> add a driver flag to force the negotiation of SATA 1 transfers
> (1.5Gb/s). Useful where faster speeds are unstable; ok dlg@ (djm@)
~ if_re_pci.c
> Remove unused flags.
> ok dlg@ (brad@)
~ if_bge.c ~ if_bgereg.h
> Add a workaround for a CRC bug errata with BCM5701 A0 and B0 chipset
> revisions.
> From Linux via FreeBSD.
> ok dlg@ (brad@)
~ if_bge.c
> rev 1.35 added special register settings for BCM5704 chipsets when
> in TBI mode for fiber adapters which causes problems for some systems.
> Add code to detect when the special settings are necessary which helps
> with some blade systems.
> From FreeBSD
> ok dlg@ (brad@)
~ ichiic.c
> be quiet even when errors happen; ok kettenis (deraadt@)
dev/sbus
~ mgx.c
> Fix braino in mgx_loadcmap() invocation. (miod@)
~ mgx.c
> It turns out the mysterious chip with the heatsink glued on, found on mgx
> video boards, is a known entity, so add support for 2d acceleration with
> the help of the X code (which makes baby Jesus cry).
> The blitter is fast enough to make a cgsix feel as slow as a cgthree in
> comparison. (miod@)
dev/usb
~ usbdevs
> add my logitech camera
> "add the product id to usbdevs" robert@ (ckuethe@)
~ usbdevs.h ~ usbdevs_data.h
> sync (ckuethe@)
~ uvideo.c ~ uvideo.h
> Start to make device initialization automatic instead all the static
> games.
> Tested by robert@ and myself (mglocker@)
~ ehci.c ~ ukbd.c
> Use the timeout(9) API the right way: Don't initialize the timeout
> cookie all the time, and only call timeout_del if we're not going to
> timeout_add right away.
> Testing by johan@, oga@, sthen@, and possibly some more people (thanks).
> Ok krw@, oga@, sthen at . (mk@)
~ uvideo.c
> Oops, decrease debug log level back to 1. (mglocker@)
~ if_ral.c ~ if_rum.c
~ if_upgt.c ~ if_zyd.c
> Kernel implementation of the 4-way handshake and group-key
> handshake protocols (both supplicant and authenticator state
> machines) as defined in the IEEE 802.11i standard.
> Software implementation of the TKIP (Temporal Key Integrity
> Protocol) and CCMP (CTR with CBC-MAC Protocol) protocols.
> This diff doesn't implement any of the 802.1X authentication
> protocols and thus only PSK authentication (using pre-shared
> keys) is currently supported.
> In concrete terms, this adds support for WPA-PSK and WPA2-PSK
> protocols, both in station and hostap modes.
> The following drivers are marked as WPA-capable and should
> work: bwi(4), malo(4), ral(4), iwn(4), wpi(4), ural(4),
> rum(4), upgt(4), and zyd(4)
> The following options have been added to ifconfig(8):
> wpa, wpapsk, wpaprotos, wpaakms, wpaciphers, wpagroupcipher
> wpa-psk(8) can be used to generate keys from passphrases.
> tested by many@
> ok deraadt@ (damien@)
~ uvideo.h
> Tune isoc maximal transfer size to a better value. Makes all available
> resolutions on my cam work. (mglocker@)
~ if_uath.c
> Add missing monitor mode capabilities flag.
> ok damien@ (brad@)
~ uftdi.c
> attach FTDI 2232L (fkr@)
~ uvideo.c
> Add a function which can identify an USB descriptor explicitly by it's
> size, since the UVC spec uses different USB descriptors with the same
> bDescriptorSubtype ID. (mglocker@)
~ uvideo.c ~ uvideo.h
> Add VS_INPUT_HEADER descriptor. (mglocker@)
~ uvideo.c
> Sort functions and remove obsolete bits. (mglocker@)
~ uvideo.c ~ uvideo.h
> Parse VS_INPUT_HEADER descriptor. (mglocker@)
kern
~ uipc_usrreq.c
> describe why fd passing is now so fucked up, and who's fault it is
> (deraadt@)
lib/libkern
- arch/sparc64/random.S
> Remove the random() function from locore.s (which used sparcv7
> instructions,
> including mulscc to do multiplications) and switch to the generic random.c
> code. (kettenis@)
+ arch/hppa64/Makefile + arch/hppa64/bcopy.S
+ arch/hppa64/memcpy.S + arch/hppa64/memmove.S
+ arch/hppa64/ovbcopy.S + arch/hppa64/spcopy.S
> Catch up with dismantled libkern. (kettenis@)
lib/libsa
~ loadfile_elf.c
> Change ELF loader to use the LMA as the load address for the
> various segments. Hopefully this will help remove various
> hacks in the boot loader in the future. This should have no
> effect on most architectures (as we tend to have LMA == VMA).
> ok drahn@, soft ok's various others. (weingart@)
miscfs/specfs
~ spec_vnops.c ~ specdev.h
> garbage collect the now unused spec_ebadf();
> ok dlg@ (thib@)
net
~ if_ethersubr.c
> extend the if_ethersubr.c crc functions to support updating a running
> crc in addition to the existing "oneshot" mode and use them to replace
> ieee80211_crc_update() with the new ether_crc32_le_update(). Saves 1k
> kernel bss + some code.
> Mark the new ether_crc32_[lb]e_update functions as __pure for a
> ~25x speedup (on my i386 at least).
> feedback and ok damien@ (djm@)
net80211
~ ieee80211.c ~ ieee80211.h
~ ieee80211_crypto.c ~ ieee80211_crypto.h
~ ieee80211_input.c ~ ieee80211_ioctl.c
~ ieee80211_ioctl.h ~ ieee80211_node.c
~ ieee80211_node.h ~ ieee80211_output.c
~ ieee80211_proto.c ~ ieee80211_proto.h
~ ieee80211_var.h + ieee80211_crypto_ccmp.c
+ ieee80211_crypto_tkip.c + ieee80211_crypto_wep.c
> Kernel implementation of the 4-way handshake and group-key
> handshake protocols (both supplicant and authenticator state
> machines) as defined in the IEEE 802.11i standard.
> Software implementation of the TKIP (Temporal Key Integrity
> Protocol) and CCMP (CTR with CBC-MAC Protocol) protocols.
> This diff doesn't implement any of the 802.1X authentication
> protocols and thus only PSK authentication (using pre-shared
> keys) is currently supported.
> In concrete terms, this adds support for WPA-PSK and WPA2-PSK
> protocols, both in station and hostap modes.
> The following drivers are marked as WPA-capable and should
> work: bwi(4), malo(4), ral(4), iwn(4), wpi(4), ural(4),
> rum(4), upgt(4), and zyd(4)
> The following options have been added to ifconfig(8):
> wpa, wpapsk, wpaprotos, wpaakms, wpaciphers, wpagroupcipher
> wpa-psk(8) can be used to generate keys from passphrases.
> tested by many@
> ok deraadt@ (damien@)
~ ieee80211_crypto.c
> call ieee80211_crc_init() only once, when the first 802.11 device
> attaches instead of at every attach.
> discussed with deraadt@ (damien@)
~ ieee80211_crypto.c ~ ieee80211_crypto.h
~ ieee80211_crypto_tkip.c ~ ieee80211_crypto_wep.c
> extend the if_ethersubr.c crc functions to support updating a running
> crc in addition to the existing "oneshot" mode and use them to replace
> ieee80211_crc_update() with the new ether_crc32_le_update(). Saves 1k
> kernel bss + some code.
> Mark the new ether_crc32_[lb]e_update functions as __pure for a
> ~25x speedup (on my i386 at least).
> feedback and ok damien@ (djm@)
netatalk
~ at_control.c
> use arc4random_uniform() for random number requests that are not a
> power of two.
> use arc4random_bytes() when requesting more than a word of PRNG
> output.
> ok deraadt@ (djm@)
netinet
~ ip_ipsp.c ~ in_pcb.c
~ igmp_var.h
> use arc4random_uniform() for random number requests that are not a
> power of two.
> use arc4random_bytes() when requesting more than a word of PRNG
> output.
> ok deraadt@ (djm@)
~ if_ether.h
> extend the if_ethersubr.c crc functions to support updating a running
> crc in addition to the existing "oneshot" mode and use them to replace
> ieee80211_crc_update() with the new ether_crc32_le_update(). Saves 1k
> kernel bss + some code.
> Mark the new ether_crc32_[lb]e_update functions as __pure for a
> ~25x speedup (on my i386 at least).
> feedback and ok damien@ (djm@)
netinet6
~ nd6_nbr.c ~ mld6_var.h
~ ip6_id.c ~ in6_pcb.c
> use arc4random_uniform() for random number requests that are not a
> power of two.
> use arc4random_bytes() when requesting more than a word of PRNG
> output.
> ok deraadt@ (djm@)
nfs
~ nfs_srvcache.c
> move the cache lookup code into a function, instead repeating
> it twice, makes this nicer as a bonus;
> From NetBSD.
> tested by johan@, ok blambert at . (thib@)
~ nfs_subs.c ~ nfs_var.h
~ nfs_vfsops.c ~ nfs_vnops.c
~ nfsm_subs.h
> Rework NFS mbuf write routines:
> nfsm_uiotombuf():
> - no meaningful return, function becomes void
> - replace frankensteined uiomove/MGET with actual call to uiomove
> nfsm_strtmbuf():
> - no meaningful return, becomes void
> - rename and reorder parameters to be consistent with other functions
> - becomes wrapper to nfsm_uiotombuf
> nfsm_buftombuf():
> - new function; unused but slated for future use
> - wrapper to nfsm_uiotombuf
> As a bonus, making these functions void removes much unneeded code and
> garbage collects `magic' variable t2 which was only used (invisibly)
> by the macros from nfsm_subs.h
> Preps the way for removing much more NFS goo...
> testing by myself, thib@ and merdely@
> ok thib@ (blambert@)
~ nfs_subs.c
> more efficient use of mbufs in nfsm_uiotombuf().
> allocate a mbuf cluster only if length is greater than MLEN
> instead of MINCLSIZE.
> ok blambert@, thib@ (damien@)
~ nfs_subs.c
> use arc4random_uniform() for random number requests that are not a
> power of two.
> use arc4random_bytes() when requesting more than a word of PRNG
> output.
> ok deraadt@ (djm@)
~ nfs_subs.c
> in nfsm_reqh(), only allocate an mbuf cluster if the
> len exceeds MLEN, not MINCLSIZE; inspired by a similar
> commit from damien.
> ok damien@, blambert@ (thib@)
== usr.bin =========================================================== 12/13 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/usr.bin
aucat
~ aucat.c
> - remove code that was rendered pointless by the last commit
> - minor knf
> from Pierre Riteau (jakemsr@)
awk
~ run.c
> Use arc4random_buf() when requesting more than a single word of output
> Use arc4random_uniform() when the desired random number upper bound
> is not a power of two
> ok deraadt@ millert@ (djm@)
cal
~ cal.1 ~ cal.c
> Allow cal to show week numbers. Switch between sunday based week numbering
> and ISO 8601 monday based week numbering with overlapping weeks if the -m
> flag is present.
> many positive feedback received.
> ok mbalmer@, jasper@, simon@
> input from todd@ (pyr@)
~ cal.1
> tweaks; (jmc@)
calendar
~ day.c
> Use arc4random_buf() when requesting more than a single word of output
> Use arc4random_uniform() when the desired random number upper bound
> is not a power of two
> ok deraadt@ millert@ (djm@)
cvs
~ client.c ~ server.c
> Properly check memory allocation in client code. While at it, unified
> vasprintf() check across OpenCVS code base.
> Based on a diff by Jacek Masiulaniec.
> ok (and unification requested by) xsa (tobias@)
mg
~ theo.c
> Use arc4random_buf() when requesting more than a single word of output
> Use arc4random_uniform() when the desired random number upper bound
> is not a power of two
> ok deraadt@ millert@ (djm@)
pr
~ pr.c
> If memory allocation fails during error message buffering, print all hold
> back messages and exit.
> ok millert (tobias@)
sendbug
~ sendbug.c
> Decrement len variable after removing newline, prevents copying the
> NUL at the end of a string into the mail. Discovered by dasn.
> Move newline printing into outer loop, prevents multiple newlines
> from appearing if there are multiple comments in a line. Discovered
> by okan.
> OK okan. (ray@)
ssh
~ dh.c ~ sshd.c
> Use arc4random_buf() when requesting more than a single word of output
> Use arc4random_uniform() when the desired random number upper bound
> is not a power of two
> ok deraadt@ millert@ (djm@)
~ sftp-client.c ~ sftp-client.h
~ sftp-server.c ~ sftp.1
~ sftp.c ~ sftp.h
~ sftp/Makefile
> introduce sftp extension methods statvfs at openssh.com and
> fstatvfs at openssh.com that implement statvfs(2)-like operations,
> based on a patch from miklos AT szeredi.hu (bz#1399)
> also add a "df" command to the sftp client that uses the
> statvfs at openssh.com to produce a df(1)-like display of filesystem
> space and inode utilisation
> ok markus@ (djm@)
~ sftp.1
> macro fixage; (jmc@)
~ session.c
> remove unneccessary parentheses (djm@)
== usr.sbin ========================================================== 13/13 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/usr.sbin
config
~ mkmakefile.c
> remove build warnings:
> - unused variables
> - missing return
> ok espie@ (chl@)
dhcpd
~ options.c
> Redo/clarify the logic for storing options in the various buffers.
> While there, bail out quickly if no options can be stored, and avoid
> overflow buffers if building a bootp reply.
> Henning looked and didn't get upset. millert@ tested it for a few
> days and liked the look of it. canacar@ glanced at it. (krw@)
ftp-proxy
~ ftp-proxy.c
> Use arc4random_buf() when requesting more than a single word of output
> Use arc4random_uniform() when the desired random number upper bound
> is not a power of two
> ok deraadt@ millert@ (djm@)
httpd
~ src/modules/standard/mod_rewrite.c
> Use arc4random_buf() when requesting more than a single word of output
> Use arc4random_uniform() when the desired random number upper bound
> is not a power of two
> ok deraadt@ millert@ (djm@)
lpr
~ lpd/lpd.8
> - make it clear that hosts.lpd follows the same syntax that hosts.equiv(5)
> inputs from jmc@
> "looks ok" beck@ (similar older diff), "fine by me" jmc@ (ajacoutot@)
ntpd
~ ntp.c
> Use arc4random_buf() when requesting more than a single word of output
> Use arc4random_uniform() when the desired random number upper bound
> is not a power of two
> ok deraadt@ millert@ (djm@)
ripd
~ message.c
> Use arc4random_buf() when requesting more than a single word of output
> Use arc4random_uniform() when the desired random number upper bound
> is not a power of two
> ok deraadt@ millert@ (djm@)
rtadvd
~ rtadvd.c
> Use arc4random_buf() when requesting more than a single word of output
> Use arc4random_uniform() when the desired random number upper bound
> is not a power of two
> ok deraadt@ millert@ (djm@)
~ advcap.c ~ config.c
> Take failed system calls more serious than LOG_INFO.
> This is mainly in preparation for more changes to logging.
> Ok bluhm@, pyr@ (rainer@)
rtsold
~ rtsold.c
> Use arc4random_buf() when requesting more than a single word of output
> Use arc4random_uniform() when the desired random number upper bound
> is not a power of two
> ok deraadt@ millert@ (djm@)
tcpdump
~ privsep.c ~ privsep.h
~ privsep_pcap.c ~ tcpdump.8
~ tcpdump.c
> add filtering on direction; ok mpf@ deraadt@ feedback jmc@ (djm@)
===============================================================================
More information about the owc
mailing list