[owc] Weekly src changes ending 2011-05-29
OWC
auto at squish.net
Mon May 30 07:00:01 BST 2011
OpenBSD src changes summary for 2011-05-22 to 2011-05-29 inclusive
==================================================================
bin/pax distrib/notes
distrib/sets etc/netstart
include/fenv.h include/math.h
include/wchar.h lib/libc
lib/libkvm lib/libm
lib/libssl lib/libz
libexec/ld.so libexec/security
regress/lib regress/usr.bin
regress/usr.sbin sbin/disklabel
sbin/ifconfig sbin/iked
sbin/newfs sbin/sysctl
share/man sys/arch/alpha/conf
sys/arch/alpha/include sys/arch/amd64/amd64
sys/arch/amd64/conf sys/arch/amd64/include
sys/arch/amd64/pci sys/arch/arm/include
sys/arch/armish/conf sys/arch/gumstix/conf
sys/arch/hppa/conf sys/arch/hppa/include
sys/arch/hppa64/conf sys/arch/hppa64/include
sys/arch/i386/conf sys/arch/i386/i386
sys/arch/i386/include sys/arch/i386/pci
sys/arch/landisk/conf sys/arch/loongson/conf
sys/arch/loongson/dev sys/arch/loongson/loongson
sys/arch/m68k/include sys/arch/m68k/m68k
sys/arch/m88k/include sys/arch/macppc/conf
sys/arch/macppc/dev sys/arch/macppc/include
sys/arch/macppc/macppc sys/arch/mips64/include
sys/arch/palm/conf sys/arch/powerpc/include
sys/arch/sgi/conf sys/arch/sh/include
sys/arch/socppc/include sys/arch/socppc/socppc
sys/arch/sparc/include sys/arch/sparc64/conf
sys/arch/sparc64/include sys/arch/zaurus/conf
sys/ddb sys/dev/acpi
sys/dev/ata sys/dev/atapiscsi
sys/dev/i2c sys/dev/ic
sys/dev/pci sys/dev/usb
sys/kern sys/lib/libsa
sys/net sys/netinet
sys/uvm usr.bin/aucat
usr.bin/man usr.bin/mandoc
usr.bin/pkg-config usr.bin/ssh
usr.bin/tmux usr.sbin
usr.sbin/ikectl usr.sbin/nsd
usr.sbin/ospfd usr.sbin/procmap
usr.sbin/relayd usr.sbin/smtpd
== bin =============================================================== 01/12 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/bin
pax
~ pax.c
> fix one signal race bug, and mark the others that remain much more
> specifically
> ok nicm (deraadt@)
== distrib =========================================================== 02/12 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/distrib
notes
~ m4.common
> sollicitation -> solicitation; (jmc@)
sets
~ lists/comp/mi
> sync (deraadt@)
~ lists/man/mi
> sync (deraadt@)
~ lists/comp/mi
> sync (deraadt@)
~ lists/base/md.alpha ~ lists/base/md.amd64
~ lists/base/md.armish ~ lists/base/md.hp300
~ lists/base/md.hppa ~ lists/base/md.hppa64
~ lists/base/md.i386 ~ lists/base/md.landisk
~ lists/base/md.loongson ~ lists/base/md.mac68k
~ lists/base/md.macppc ~ lists/base/md.mvme68k
~ lists/base/md.mvmeppc ~ lists/base/md.palm
~ lists/base/md.sgi ~ lists/base/md.socppc
~ lists/base/md.sparc ~ lists/base/md.sparc64
~ lists/base/md.zaurus ~ lists/comp/mi
> sync (deraadt@)
== etc =============================================================== 03/12 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/etc
netstart
~ netstart
> Add svlan(4) startup bits.
> From markus at . OK naddy, claudio, reyk. (mpf@)
== include =========================================================== 04/12 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/include
fenv.h
~ fenv.h
> Also enclose variable names with external linkage in __BEGIN_DECLS.
> Doesn't matter much since C++ ABI used by GCC doesn't mangle variable
> names; however technically is required by Section 7.5 of the C++ spec.
> Discussed with/OK guenther@, matthew at . (martynas@)
math.h
~ math.h
> Also enclose variable names with external linkage in __BEGIN_DECLS.
> Doesn't matter much since C++ ABI used by GCC doesn't mangle variable
> names; however technically is required by Section 7.5 of the C++ spec.
> Discussed with/OK guenther@, matthew at . (martynas@)
wchar.h
~ wchar.h
> wcs(n)casecmp support, manpage comments from jmc@, okay deraadt@ (espie@)
== lib =============================================================== 05/12 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/lib
libc
~ arch/powerpc/gen/fpsetsticky.c
> On PowerPC we cannot manipulate FPSCR[VX]--it's a summary of all
> the invalid exception bits, as described in Section 3.3.6.1.1 of
> PowerPC Architecture Programming Environments Manual.
> A proper way to cause an invalid operation exception is to set
> FPSCR[VXSOFT]. Similarly, we clear all the FPSCR[VX*] bits otherwise.
> (martynas@)
~ shlib_version ~ string/Makefile.inc
~ string/wmemchr.3 + string/wcscasecmp.c
> wcs(n)casecmp support, manpage comments from jmc@, okay deraadt@ (espie@)
~ string/wmemchr.3
> tweak previous (my fault); (jmc@)
libkvm
~ kvm_proc.c
> Reimplement uvm/uvm_map.
> vmmap is designed to perform address space randomized allocations,
> without letting fragmentation of the address space go through the roof.
> Some highlights:
> - kernel address space randomization
> - proper implementation of guardpages
> - roughly 10% system time reduction during kernel build
> Tested by alot of people on tech@ and developers.
> Theo's machines are still happy. (ariane@)
libm
~ Makefile + man/feclearexcept.3
+ man/feenableexcept.3 + man/fegetenv.3
+ man/fegetround.3
> Document C99 floating-point environment. (martynas@)
~ Makefile ~ man/rint.3
> Document nearbyint, nearbyintf, nearbyintl. (martynas@)
~ man/feclearexcept.3 ~ man/feenableexcept.3
~ man/fegetenv.3 ~ man/fegetround.3
~ man/rint.3
> tweak previous; ok martynas (jmc@)
~ Makefile + arch/mc68881/e_acosf.S
+ arch/mc68881/e_asinf.S + arch/mc68881/e_atanhf.S
+ arch/mc68881/e_coshf.S + arch/mc68881/e_expf.S
+ arch/mc68881/e_log10f.S + arch/mc68881/e_logf.S
+ arch/mc68881/e_remainderf.S + arch/mc68881/e_scalbf.S
+ arch/mc68881/e_sinhf.S + arch/mc68881/e_sqrtf.S
+ arch/mc68881/s_atanf.S + arch/mc68881/s_ceilf.S
+ arch/mc68881/s_copysignf.S + arch/mc68881/s_cosf.S
+ arch/mc68881/s_expm1f.S + arch/mc68881/s_floorf.S
+ arch/mc68881/s_log1pf.S + arch/mc68881/s_logbf.S
+ arch/mc68881/s_rintf.S + arch/mc68881/s_scalbnf.S
+ arch/mc68881/s_sinf.S + arch/mc68881/s_tanf.S
+ arch/mc68881/s_tanhf.S
> Adapt m68k MD functions to single-precision.
> Looked over by miod at . (martynas@)
~ Makefile + arch/mc68881/e_acosl.S
+ arch/mc68881/e_asinl.S + arch/mc68881/e_sqrtl.S
+ arch/mc68881/s_atanl.S + arch/mc68881/s_copysignl.S
+ arch/mc68881/s_cosl.S + arch/mc68881/s_logbl.S
+ arch/mc68881/s_rintl.S + arch/mc68881/s_scalbnl.S
+ arch/mc68881/s_sinl.S + arch/mc68881/s_tanl.S
> Adapt m68k MD functions to extended-precision.
> Go for it miod at . (martynas@)
~ src/e_lgamma_r.c ~ src/e_lgammaf_r.c
> Store -1 into signgam in case of a special value of -0. (martynas@)
libssl
~ Makefile
> remove hack to test and create /usr/lib/pkgconfig/ if needed, it's been six
> months since it was introduced so it's safe to assume people have this dir
> now.
> ok deraadt@ (jasper@)
libz
~ Makefile
> remove hack to test and create /usr/lib/pkgconfig/ if needed, it's been six
> months since it was introduced so it's safe to assume people have this dir
> now.
> ok deraadt@ (jasper@)
== libexec =========================================================== 06/12 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/libexec
ld.so
~ dlfcn.c ~ loader.c
~ resolve.c ~ resolve.h
> Dynamic linker speed improvement diff. tested by several sinc k2k11.
> get it in tree now deraadt@, ok by several ports folks. Thanks for the
> testing. (drahn@)
~ resolve.c
> No need to call _dl_newsymsearch() twice; ok drahn@ (otto@)
security
~ security
> Fix two bugs where race conditions might cause stat(2) to fail,
> such that security(8) would output garbage on stderr.
> One reported by <RD at thrush dot com>, the other found by code audit.
> While here, remove dead code in two other places: stat(_) uses cashed
> data and cannot fail, not even if the file is removed in between.
> ok Andrew Fresh (schwarze@)
== regress =========================================================== 07/12 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/regress
lib
+ libm/lgamma/Makefile + libm/lgamma/lgamma.c
> Add some simple test cases for lgamma, and lgammaf.
> Currently the functions return correct results for the input of
> negative zero; however signgam is wrong in this case. (martynas@)
usr.bin
~ ssh/cfgmatch.sh
> include testing of multiple/overridden AuthorizedKeysFiles
> refactor to simply daemon start/stop and get rid of racy constructs (djm@)
usr.sbin
~ pkg_add/signatures.ref
> fix test output for new message (espie@)
~ pkg_add/Makefile + pkg_add/check-path
> unit-test @pkgpath before we add a lot of junk to it (espie@)
~ pkg_add/check-path
> flesh out flavors, future test that does not match yet (espie@)
~ pkg_add/check-path
> more extra stuff for later (espie@)
~ pkg_add/check-path
> rename (espie@)
~ pkg_add/check-path
> several more bizarre cases (espie@)
== sbin ============================================================== 08/12 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/sbin
disklabel
~ editor.c
> In addition to relative resizing, allow absolute resizing of partitions
> in auto-allocated labels. ok krw@ sthen@ (otto@)
~ disklabel.c
> zap unused var (otto@)
~ disklabel.8
> sync auto-allocation table with code (otto@)
~ editor.c
> Apply proper rounding for the R command (changing of auto-allocated
> partitions); ok krw@ (otto@)
~ editor.c
> Make /usr/obj a bit bigger, on small disks it ends up just a little
> bit too small. ok krw@ (otto@)
~ disklabel.8
> sync auto-allocation table (otto@)
~ Makefile ~ editor.c
> Allow for more room for /usr/obj on static linking only archs; ok krw@
> deraadt@ (otto@)
ifconfig
~ ifconfig.c
> Don't provide an af hint to getaddrinfo in settunnel(), this function
> already checks that families of source and destination addresses match
> and that's all we need.
> This allows "ifconfig foo tunnel 1::1 2::2" syntax rather than requiring
> "ifconfig foo inet6 tunnel 1::1 2::2", which in turn allows hostname.if
> files to create an IPv4-in-IPv6 tunnel without games with shell escapes.
> ok dcoppa@, seems ok todd@, "Yes, yes and yes" claudio@ (sthen@)
iked
~ ca.c ~ chap_ms.c
~ eap.c ~ ikev2.c
~ parse.y ~ pfkey.c
~ timer.c ~ util.c
> spacing (reyk@)
newfs
~ mkfs.c ~ newfs.8
~ newfs.c
> Allow specifying k/m/g/... suffixes in newfs(8) -S and -s options.
> Useful for mount_mfs, now you can just say:
> # mount_mfs -s 50m swap /tmp
> And it will do what you want, taking into account sector size.
> Old behaviour of -s (specifying count of sectors) is, of course,
> preserved.
> All the work was done by Vadim Zhukov (persgray(at)gmail com).
> OK myself, otto@, thib@ (dcoppa@)
sysctl
~ sysctl.c
> allow ranges for entering net.inet.(tcp|udp).baddynamic lists, e.g.
> sysctl net.inet.tcp.baddynamic=1-32768
> sysctl net.inet.udp.baddynamic=+40000-65535
> ok mk@ (djm@)
~ sysctl.8
> mention ranges; prodding and feedback jmc@ (djm@)
== share ============================================================= 09/12 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/share
man
~ man9/Makefile
> fix pathname (deraadt@)
~ man9/pci_intr_map.9
> - new sentence, new line
> - space needed before punctuation (jmc@)
~ man8/rc.shutdown.8
> better document what happens at system shutdown;
> much discussion/help/ok ajacoutot schwarze (jmc@)
~ man4/man4.macppc/Makefile + man4/man4.macppc/dfs.4
> Manpage for dfs(4) with tweaks from jmc@
> ok sthen@, kettenis@, miod@ (mpi@)
~ man4/man4.macppc/dfs.4
> no comma desired for last Xr in SEE ALSO; (jmc@)
~ man4/man4.macppc/macgpio.4
> update the hardware list; (jmc@)
~ man8/rc.d.8
> BUGS: there is currently no way to know whether a background process was
> successfully started so rc_start will always succeed.
> ok robert@ (ajacoutot@)
~ man8/rc.d.8
> no .Pp before/after .S{s,h}; (jmc@)
~ man5/port-modules.5
> Adapt after recent dconf MODULE change. (ajacoutot@)
== sys =============================================================== 10/12 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/sys
arch/alpha/conf
~ GENERIC ~ RAMDISKBIG
> On OpenBSD, we've always attached scsibus(4) to umass(4), even for
> ATAPI devices. atapiscsi(4) is only for handling ATAPI devices on an
> ATA bus, so umass(4) shouldn't care about it.
> ok krw@, dlg@; no objections from deraadt@ (matthew@)
arch/alpha/include
~ fenv.h
> Also enclose variable names with external linkage in __BEGIN_DECLS.
> Doesn't matter much since C++ ABI used by GCC doesn't mangle variable
> names; however technically is required by Section 7.5 of the C++ spec.
> Discussed with/OK guenther@, matthew at . (martynas@)
arch/amd64/amd64
~ identcpu.c + k1x-pstate.c
> AMD K10/K11 pstate driver allows setperf and apm to change CPU
> frequencies on newer AMD systems.
> Driver written by Bryan Steele / brynet gmail.com
> Put it in deraadt@ (claudio@)
~ intr.c
> Remove out of data XXX comment. (kettenis@)
~ k1x-pstate.c
> Let this compile on kernels without acpicpu.
> Problem found by Nathanael Rensen (nathanael at polymorpheus dot com)
> (claudio@)
~ identcpu.c
> Use k1x cpu scaling on all families 0x10 and above (the trend is likely to
> continue); makes the AMD E-350 speed adjust (from slow to way slower).
> discussion with jsg. (deraadt@)
arch/amd64/conf
~ files.amd64
> AMD K10/K11 pstate driver allows setperf and apm to change CPU
> frequencies on newer AMD systems.
> Driver written by Bryan Steele / brynet gmail.com
> Put it in deraadt@ (claudio@)
~ GENERIC ~ RAMDISK
~ RAMDISK_CD
> On OpenBSD, we've always attached scsibus(4) to umass(4), even for
> ATAPI devices. atapiscsi(4) is only for handling ATAPI devices on an
> ATA bus, so umass(4) shouldn't care about it.
> ok krw@, dlg@; no objections from deraadt@ (matthew@)
arch/amd64/include
~ cpu.h
> AMD K10/K11 pstate driver allows setperf and apm to change CPU
> frequencies on newer AMD systems.
> Driver written by Bryan Steele / brynet gmail.com
> Put it in deraadt@ (claudio@)
~ fenv.h
> Also enclose variable names with external linkage in __BEGIN_DECLS.
> Doesn't matter much since C++ ABI used by GCC doesn't mangle variable
> names; however technically is required by Section 7.5 of the C++ spec.
> Discussed with/OK guenther@, matthew at . (martynas@)
arch/amd64/pci
~ pci_machdep.c
> Don't attempt to use MSI if we're not running in APIC mode. (kettenis@)
arch/arm/include
~ fenv.h
> Also enclose variable names with external linkage in __BEGIN_DECLS.
> Doesn't matter much since C++ ABI used by GCC doesn't mangle variable
> names; however technically is required by Section 7.5 of the C++ spec.
> Discussed with/OK guenther@, matthew at . (martynas@)
arch/armish/conf
~ GENERIC ~ RAMDISK
> On OpenBSD, we've always attached scsibus(4) to umass(4), even for
> ATAPI devices. atapiscsi(4) is only for handling ATAPI devices on an
> ATA bus, so umass(4) shouldn't care about it.
> ok krw@, dlg@; no objections from deraadt@ (matthew@)
arch/gumstix/conf
~ GENERIC
> On OpenBSD, we've always attached scsibus(4) to umass(4), even for
> ATAPI devices. atapiscsi(4) is only for handling ATAPI devices on an
> ATA bus, so umass(4) shouldn't care about it.
> ok krw@, dlg@; no objections from deraadt@ (matthew@)
arch/hppa/conf
~ GENERIC ~ RAMDISK
> On OpenBSD, we've always attached scsibus(4) to umass(4), even for
> ATAPI devices. atapiscsi(4) is only for handling ATAPI devices on an
> ATA bus, so umass(4) shouldn't care about it.
> ok krw@, dlg@; no objections from deraadt@ (matthew@)
arch/hppa/include
~ fenv.h
> Also enclose variable names with external linkage in __BEGIN_DECLS.
> Doesn't matter much since C++ ABI used by GCC doesn't mangle variable
> names; however technically is required by Section 7.5 of the C++ spec.
> Discussed with/OK guenther@, matthew at . (martynas@)
arch/hppa64/conf
~ GENERIC ~ RAMDISK
> On OpenBSD, we've always attached scsibus(4) to umass(4), even for
> ATAPI devices. atapiscsi(4) is only for handling ATAPI devices on an
> ATA bus, so umass(4) shouldn't care about it.
> ok krw@, dlg@; no objections from deraadt@ (matthew@)
arch/hppa64/include
~ fenv.h
> Also enclose variable names with external linkage in __BEGIN_DECLS.
> Doesn't matter much since C++ ABI used by GCC doesn't mangle variable
> names; however technically is required by Section 7.5 of the C++ spec.
> Discussed with/OK guenther@, matthew at . (martynas@)
arch/i386/conf
~ files.i386
> AMD K10/K11 pstate driver allows setperf and apm to change CPU
> frequencies on newer AMD systems.
> Driver written by Bryan Steele / brynet gmail.com adjusted for i386
> by myself. Put it in deraadt@ (claudio@)
~ GENERIC ~ RAMDISKC
~ RAMDISK_CD
> On OpenBSD, we've always attached scsibus(4) to umass(4), even for
> ATAPI devices. atapiscsi(4) is only for handling ATAPI devices on an
> ATA bus, so umass(4) shouldn't care about it.
> ok krw@, dlg@; no objections from deraadt@ (matthew@)
arch/i386/i386
~ machdep.c + k1x-pstate.c
> AMD K10/K11 pstate driver allows setperf and apm to change CPU
> frequencies on newer AMD systems.
> Driver written by Bryan Steele / brynet gmail.com adjusted for i386
> by myself. Put it in deraadt@ (claudio@)
~ pmap.c
> Reimplement uvm/uvm_map.
> vmmap is designed to perform address space randomized allocations,
> without letting fragmentation of the address space go through the roof.
> Some highlights:
> - kernel address space randomization
> - proper implementation of guardpages
> - roughly 10% system time reduction during kernel build
> Tested by alot of people on tech@ and developers.
> Theo's machines are still happy. (ariane@)
~ k1x-pstate.c
> Let this compile on kernels without acpicpu.
> Problem found by Nathanael Rensen (nathanael at polymorpheus dot com)
> (claudio@)
~ machdep.c
> Use k1x cpu scaling on all families 0x10 and above (the trend is likely to
> continue); makes the AMD E-350 speed adjust (from slow to way slower).
> discussion with jsg. (deraadt@)
arch/i386/include
~ cpu.h
> AMD K10/K11 pstate driver allows setperf and apm to change CPU
> frequencies on newer AMD systems.
> Driver written by Bryan Steele / brynet gmail.com adjusted for i386
> by myself. Put it in deraadt@ (claudio@)
~ fenv.h
> Also enclose variable names with external linkage in __BEGIN_DECLS.
> Doesn't matter much since C++ ABI used by GCC doesn't mangle variable
> names; however technically is required by Section 7.5 of the C++ spec.
> Discussed with/OK guenther@, matthew at . (martynas@)
arch/i386/pci
~ pci_machdep.c
> Don't attempt to use MSI if we're not running in APIC mode. (kettenis@)
arch/landisk/conf
~ GENERIC ~ RAMDISK
> On OpenBSD, we've always attached scsibus(4) to umass(4), even for
> ATAPI devices. atapiscsi(4) is only for handling ATAPI devices on an
> ATA bus, so umass(4) shouldn't care about it.
> ok krw@, dlg@; no objections from deraadt@ (matthew@)
arch/loongson/conf
~ GENERIC ~ RAMDISK
> On OpenBSD, we've always attached scsibus(4) to umass(4), even for
> ATAPI devices. atapiscsi(4) is only for handling ATAPI devices on an
> ATA bus, so umass(4) shouldn't care about it.
> ok krw@, dlg@; no objections from deraadt@ (matthew@)
~ RAMDISK
> Add the second serial port, might be the console device on 2E systems.
> (miod@)
arch/loongson/dev
~ smfb.c
> Do not dereference ri_devcmap[] twice; found the hard way by Gilbert
> Fernandes
> (firstname dot lastname at orange dot frogland). (miod@)
arch/loongson/loongson
~ machdep.c
> Set UX and SX in the system register early in boot. While not necessary on
> the real hardware, some emulators (such as those which name start with `qe'
> and end in `mu' and are four letters long) expect the cpu behaviour to
> honour the value of these bits. This is not worth doing but it's cheap.
> (miod@)
arch/m68k/include
~ pmap_motorola.h
> Reimplement uvm/uvm_map.
> vmmap is designed to perform address space randomized allocations,
> without letting fragmentation of the address space go through the roof.
> Some highlights:
> - kernel address space randomization
> - proper implementation of guardpages
> - roughly 10% system time reduction during kernel build
> Tested by alot of people on tech@ and developers.
> Theo's machines are still happy. (ariane@)
~ fenv.h
> Also enclose variable names with external linkage in __BEGIN_DECLS.
> Doesn't matter much since C++ ABI used by GCC doesn't mangle variable
> names; however technically is required by Section 7.5 of the C++ spec.
> Discussed with/OK guenther@, matthew at . (martynas@)
arch/m68k/m68k
~ pmap_motorola.c
> Reimplement uvm/uvm_map.
> vmmap is designed to perform address space randomized allocations,
> without letting fragmentation of the address space go through the roof.
> Some highlights:
> - kernel address space randomization
> - proper implementation of guardpages
> - roughly 10% system time reduction during kernel build
> Tested by alot of people on tech@ and developers.
> Theo's machines are still happy. (ariane@)
~ pmap_motorola.c
> Thou shall not use uninitialized TAILQs (miod@)
arch/m88k/include
~ fenv.h
> Also enclose variable names with external linkage in __BEGIN_DECLS.
> Doesn't matter much since C++ ABI used by GCC doesn't mangle variable
> names; however technically is required by Section 7.5 of the C++ spec.
> Discussed with/OK guenther@, matthew at . (martynas@)
arch/macppc/conf
~ GENERIC
> Enable ath(4)
> ok deraadt@ (mpi@)
~ GENERIC
> On OpenBSD, we've always attached scsibus(4) to umass(4), even for
> ATAPI devices. atapiscsi(4) is only for handling ATAPI devices on an
> ATA bus, so umass(4) shouldn't care about it.
> ok krw@, dlg@; no objections from deraadt@ (matthew@)
~ GENERIC ~ files.macppc
> Add dfs(4), a driver to support the Dynamic Frequency Switching feature
> found on some G4 PowerBook.
> Tested by many, thanks.
> ok sthen@, kettenis@, miod@ (mpi@)
arch/macppc/dev
+ dfs.c
> Add dfs(4), a driver to support the Dynamic Frequency Switching feature
> found on some G4 PowerBook.
> Tested by many, thanks.
> ok sthen@, kettenis@, miod@ (mpi@)
arch/macppc/include
~ cpu.h
> Add dfs(4), a driver to support the Dynamic Frequency Switching feature
> found on some G4 PowerBook.
> Tested by many, thanks.
> ok sthen@, kettenis@, miod@ (mpi@)
arch/macppc/macppc
~ cpu.c
> Add dfs(4), a driver to support the Dynamic Frequency Switching feature
> found on some G4 PowerBook.
> Tested by many, thanks.
> ok sthen@, kettenis@, miod@ (mpi@)
arch/mips64/include
~ fenv.h
> Also enclose variable names with external linkage in __BEGIN_DECLS.
> Doesn't matter much since C++ ABI used by GCC doesn't mangle variable
> names; however technically is required by Section 7.5 of the C++ spec.
> Discussed with/OK guenther@, matthew at . (martynas@)
arch/palm/conf
~ GENERIC
> On OpenBSD, we've always attached scsibus(4) to umass(4), even for
> ATAPI devices. atapiscsi(4) is only for handling ATAPI devices on an
> ATA bus, so umass(4) shouldn't care about it.
> ok krw@, dlg@; no objections from deraadt@ (matthew@)
arch/powerpc/include
~ fenv.h
> Also enclose variable names with external linkage in __BEGIN_DECLS.
> Doesn't matter much since C++ ABI used by GCC doesn't mangle variable
> names; however technically is required by Section 7.5 of the C++ spec.
> Discussed with/OK guenther@, matthew at . (martynas@)
arch/sgi/conf
~ GENERIC-IP27 ~ GENERIC-IP30
~ GENERIC-IP32 ~ RAMDISK-IP27
~ RAMDISK-IP30 ~ RAMDISK-IP32
> On OpenBSD, we've always attached scsibus(4) to umass(4), even for
> ATAPI devices. atapiscsi(4) is only for handling ATAPI devices on an
> ATA bus, so umass(4) shouldn't care about it.
> ok krw@, dlg@; no objections from deraadt@ (matthew@)
arch/sh/include
~ fenv.h
> Also enclose variable names with external linkage in __BEGIN_DECLS.
> Doesn't matter much since C++ ABI used by GCC doesn't mangle variable
> names; however technically is required by Section 7.5 of the C++ spec.
> Discussed with/OK guenther@, matthew at . (martynas@)
arch/socppc/include
~ disklabel.h
> do not need the DPME parts anymore (deraadt@)
arch/socppc/socppc
~ disksubr.c
> socppc is a pure MBR architecture (just like landisk and armish), so
> it does not need macppc DPME logic
> ok kettenis (deraadt@)
arch/sparc/include
~ fenv.h
> Also enclose variable names with external linkage in __BEGIN_DECLS.
> Doesn't matter much since C++ ABI used by GCC doesn't mangle variable
> names; however technically is required by Section 7.5 of the C++ spec.
> Discussed with/OK guenther@, matthew at . (martynas@)
arch/sparc64/conf
~ GENERIC ~ RAMDISK
> On OpenBSD, we've always attached scsibus(4) to umass(4), even for
> ATAPI devices. atapiscsi(4) is only for handling ATAPI devices on an
> ATA bus, so umass(4) shouldn't care about it.
> ok krw@, dlg@; no objections from deraadt@ (matthew@)
arch/sparc64/include
~ fenv.h
> Also enclose variable names with external linkage in __BEGIN_DECLS.
> Doesn't matter much since C++ ABI used by GCC doesn't mangle variable
> names; however technically is required by Section 7.5 of the C++ spec.
> Discussed with/OK guenther@, matthew at . (martynas@)
arch/zaurus/conf
~ GENERIC
> On OpenBSD, we've always attached scsibus(4) to umass(4), even for
> ATAPI devices. atapiscsi(4) is only for handling ATAPI devices on an
> ATA bus, so umass(4) shouldn't care about it.
> ok krw@, dlg@; no objections from deraadt@ (matthew@)
ddb
~ db_expr.c
> In ddb, when dereferencing a pointer, use sizeof a pointer.
> Enables show map /f *kernel_map to succeed without nullpointer errors on
> 64-bit archs.
> ok miod@ (ariane@)
dev/acpi
~ acpivout.c
> Do not check malloc return value against NULL, as M_WAITOK is used.
> (pirofti@)
~ acpivideo.c
> Remove leftover include. (pirofti@)
~ acpithinkpad.c
> Ignore errors from the UCMS method. Some newer (really crappy, stay
> away from them) thinkpads are returning garbage for the return value.
> We never did anything special about the return value except spit out
> some noise, so silence it.
> ok mlarkin (deraadt@)
dev/ata
~ ata_wdc.c
> Move wdc_cd from ata/ata_wdc.c to ic/wdc.c so that you can compile a
> kernel without the former. Also, eliminate two dead stores in
> wdcattach().
> ok dlg@ (matthew@)
dev/atapiscsi
~ files.atapiscsi
> wdc.c doesn't use NATAPISCSI, so no point in including atapiscsi.h.
> It's the last file referencing atapiscsi.h too, so might as well get
> rid of the need-flags in files.atapiscsi. (matthew@)
dev/i2c
~ i2c_scan.c
> The check for the mcp98243 sdtemp(4) device was too strict. Relax it.
> Tested by Joe Gidi (deraadt@)
dev/ic
~ com.c
> Fix a problem found where one SOC has a uart implementation that causes bus
> errors if the fifo is read when the fifo is empty. Dont read an empty fifo.
> ok deraaadt@ fgsch@ (drahn@)
~ cac.c
> Fix after poolification. For reasons not completely understood, the
> fifo full check has to be protected by the mutex. ok dlg@ krw@ (otto@)
~ wdc.c
> Move wdc_cd from ata/ata_wdc.c to ic/wdc.c so that you can compile a
> kernel without the former. Also, eliminate two dead stores in
> wdcattach().
> ok dlg@ (matthew@)
~ wdc.c
> wdc.c doesn't use NATAPISCSI, so no point in including atapiscsi.h.
> It's the last file referencing atapiscsi.h too, so might as well get
> rid of the need-flags in files.atapiscsi. (matthew@)
dev/pci
~ pcidevs.h ~ pcidevs_data.h
> regen (kettenis@)
~ pcidevs
> The RD890 entries here are really SR5690/5670/5650 devices.
> ok deraadt@, krw@ (kettenis@)
~ drm/i915_drv.c
> Reimplement uvm/uvm_map.
> vmmap is designed to perform address space randomized allocations,
> without letting fragmentation of the address space go through the roof.
> Some highlights:
> - kernel address space randomization
> - proper implementation of guardpages
> - roughly 10% system time reduction during kernel build
> Tested by alot of people on tech@ and developers.
> Theo's machines are still happy. (ariane@)
~ pcidevs
> a few more devices (deraadt@)
~ pcidevs.h ~ pcidevs_data.h
> sync (deraadt@)
~ mpii.c
> try to map an msi style interrupt. if that doesnt work then try to map a
> normal intx style interrupt.
> tested on a box with three mpii(4) controllers:
> mpii0 at pci5 dev 0 function 0 "Symbios Logic SAS2008" rev 0x03: msi
> mpii1 at pci8 dev 0 function 0 "Symbios Logic SAS2008" rev 0x03: msi
> mpii2 at pci14 dev 0 function 0 "Symbios Logic SAS2008" rev 0x03: apic 2
> int 8
> no functional change, ie, everything still works.
> it is ok to do this in mpii(4) since it is only enabled on i386 and
> amd64, which are the only archs to get msi code so far.
> discussed with and ok kettenis@ (dlg@)
~ if_alc.c ~ if_alcreg.h
> - For AR8132 fast ethernet controller, do not report 1000baseT
> capability to mii(4).
> - Limit DMA burst size to be less than or equal to 1024 bytes.
> Controller does not seem to support more than 1024 bytes DMA burst.
> - Do not touch CMB TX threshold register when CMB is not used.
> From FreeBSD via Brad
> - Properly initialize sc_product and alc_rev early enough in
> alc_attach() with the PCI product id and PCI revision.
> - Further sync if_alcvar.h changes from FreeBSD for L2C/L1D
> addition commit which in OpenBSD has been merged into if_alcreg.h
> From Brad; tested by Gabriel Linder and edd at . (kevlo@)
~ if_alc.c
> Whitespace nits; from Brad (kevlo@)
~ if_age.c
> age_newbuf is called from the interrupt context so it can't sleep.
> From Brad. Tested by Thomas Pfaff (kevlo@)
~ if_vte.c ~ if_vtereg.h
> Sync MCR0_BROADCAST register naming from FreeBSD now that its
> known the register is to disable broadcast instead of enabling
> broadcast packets.
> From Brad (kevlo@)
~ piixpm.c ~ piixreg.h
> Add support for AMD SB800, where the SMBus control registers have been
> hidden
> away. Based on a diff from Bryan Steele.
> ok deraadt@ (kettenis@)
~ pcidevs
> Add AMD Family 14h devices. (kettenis@)
~ pcidevs.h ~ pcidevs_data.h
> regen (kettenis@)
~ pcidevs
> Fix typo in previous commit. (kettenis@)
~ pcidevs.h ~ pcidevs_data.h
> regen (kettenis@)
~ pcidevs
> Intels calls this the 5400 chipset series instead of the E5400 chipset
> series
> (the latter is the name of the associated Xeon processors). (kettenis@)
~ pcidevs.h ~ pcidevs_data.h
> regen (kettenis@)
~ pcidevs
> two ATI bridges found in the x201 (deraadt@)
~ pcidevs_data.h
> pcidevs.h (deraadt@)
~ pcidevs.h
> sync (deraadt@)
~ pcidevs
> Add a few missing ATI chipset entries (and fix one since we prefer the chip
> names used in the documentation over marketing names). (kettenis@)
~ pcidevs.h ~ pcidevs_data.h
> regen (kettenis@)
~ pcidevs
> Shorten a couple of old Intel chip names for consistency. Should save a
> few
> bytes as a bonus. (kettenis@)
~ pcidevs.h ~ pcidevs_data.h
> regen (kettenis@)
~ if_re_pci.c
> Add MSI support.
> ok dlg@, oga@ (kettenis@)
dev/usb
~ files.usb ~ umass.c
~ umass_scsi.c ~ usb_port.h
> On OpenBSD, we've always attached scsibus(4) to umass(4), even for
> ATAPI devices. atapiscsi(4) is only for handling ATAPI devices on an
> ATA bus, so umass(4) shouldn't care about it.
> ok krw@, dlg@; no objections from deraadt@ (matthew@)
kern
~ exec_elf.c ~ kern_exec.c
~ kern_malloc.c ~ sysv_shm.c
> Reimplement uvm/uvm_map.
> vmmap is designed to perform address space randomized allocations,
> without letting fragmentation of the address space go through the roof.
> Some highlights:
> - kernel address space randomization
> - proper implementation of guardpages
> - roughly 10% system time reduction during kernel build
> Tested by alot of people on tech@ and developers.
> Theo's machines are still happy. (ariane@)
~ sys_pipe.c
> Fix the return values from pipe_kqfilter - kqfilters should return an
> errno not 0/1.
> ok guenther (nicm@)
lib/libsa
~ ufs.c
> Do not leak memory if ufs_open() fails; reported by and ok martynas@
> (miod@)
net
~ if_pflog.c ~ if_pflog.h
~ pf.c ~ pfvar.h
> Do not pass AF specific information to pf_test_rule() and PFLOG_PACKET()
> because either the info is already available in struct pd or easy
> to figure out. Makes pf_test() and pf_test6() even more similar
> (with the target to remove one of them in the near future).
> OK henning@ (claudio@)
~ pf.c ~ pf_norm.c
~ pfvar.h
> Merge pf_scrub_ip() and pf_scrub_ip6() into a single function. Call
> pf_scrub with the right arugments in the rule case so that match
> rules will work as expected. As a benefit allow setting the tos
> on IPv6 packets as well.
> OK henning@ (claudio@)
~ pf.c
> Don't do last minute changes to diffs. Revert the argument change to
> pf_scrub in the rule (no-state) case. Since the action (a) may be NULL
> for e.g. the implicit pass rule. Should fix the panics seen by other
> people. (claudio@)
netinet
~ ip_output.c
> Do not allow traffic to be sent with a destination address in 0/8;
> this is not allowed according to Stevens and RFCs 5735 and 1122.
> Suggestion to use ENETUNREACH from claudio.
> OK phessler@, claudio@ (weerd@)
~ in_pcb.c
> Move the source address selection for multicast destinations that
> specify the outgoing interface with a multicast option up. Doing
> this before the route lookup allows multicast traffic to work even
> when the default 224/4 reject route is installed. Raw IP and IPv6
> already had this behaviour.
> Based on work by Christiano F. Haesbaert. OK sthen@ (claudio@)
uvm
~ uvm_extern.h ~ uvm_fault.c
~ uvm_io.c ~ uvm_km.c
~ uvm_map.c ~ uvm_map.h
~ uvm_mmap.c ~ uvm_unix.c
> Reimplement uvm/uvm_map.
> vmmap is designed to perform address space randomized allocations,
> without letting fragmentation of the address space go through the roof.
> Some highlights:
> - kernel address space randomization
> - proper implementation of guardpages
> - roughly 10% system time reduction during kernel build
> Tested by alot of people on tech@ and developers.
> Theo's machines are still happy. (ariane@)
~ uvm_map.c
> Fix parameter range clamping in vmmap routines.
> The old VM_MAP_RANGE_CHECK macro was wrong and caused code to be unreadable
> (argument altering macros are harmful).
> Each function now treats the memory range outside the map as it would treat
> free memory: if it would error on being given free memory, it'll error
> in a similar fashion when the start,end parameters fall outside the map.
> If it would accept free memory in its argument range, it'll silently accept
> the outside-map memory too.
> Confirmed to help ports build machines. (ariane@)
~ uvm_map.c
> Replace the lower bound PAGE_SIZE with VMMAP_MIN_ADDR.
> This makes writing a diff that makes 64-bit unclean applications explode
> a one-line diff.
> ok deraadt (ariane@)
== usr.bin =========================================================== 11/12 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/usr.bin
aucat
~ aucat.1
> postion -> position; (jmc@)
~ aproc.c ~ aproc.h
~ aucat.1 ~ aucat.c
~ dev.c ~ dev.h
> add a new -w flag to control whether master volume is automatically
> adjusted when new streams are connected and disconnected. Disabling
> automatic volume adjustment makes sense if all streams are recorded
> with properly lowered volumes. (ratchov@)
~ aproc.c
> make clipping less ugly, from Sviatoslav Chagaev (ratchov@)
~ aucat.1
> slightly clarify -v option description (ratchov@)
~ aucat.1
> tweak previous; (jmc@)
man
~ man.c
> add one more signal block for handling '_intmp' (the list of files
> scheduled
> to be deleted at termination), and then make the signal handler race safe.
> ok nicm (deraadt@)
mandoc
~ Makefile ~ chars.c
~ chars.in ~ html.c
~ html.h ~ libmandoc.h
~ libmdoc.h ~ main.c
~ main.h ~ man_html.c
~ man_term.c ~ man_validate.c
~ mandoc.1 ~ mandoc.c
~ mandoc.h ~ mdoc_argv.c
~ mdoc_html.c ~ mdoc_macro.c
~ mdoc_term.c ~ mdoc_validate.c
~ out.c ~ out.h
~ read.c ~ roff.c
~ tbl_layout.c ~ term.c
~ term.h ~ term_ascii.c
~ term_ps.c + predefs.in
> Merge release 1.11.3, almost all code by kristaps@:
> * Unicode output support (no Unicode input yet, though).
> * Refactoring: completely handle predefined strings in roff.c.
> - New function mandoc_escape() replaces a2roffdeco() and mandoc_special().
> - Start using mandoc_getarg() in mdoc_argv.c.
> - Clean up parsing of delimiters in mdoc(7).
> * And many minor fixes and lots of cleanup. (schwarze@)
~ chars.c
> Fix two regressions introduced in 1.11.3:
> * Do not pass integers outside the ASCII range to isprint().
> * Make sure escaped characters are really printed verbatim
> when the escape sequence has no special meaning. (schwarze@)
~ predefs.in
> Fix more regressions introduced in 1.11.3:
> Some predefined strings got changed in -Tascii mode.
> With this fix, they change in other output modes, but -Tascii is what
> matters. (schwarze@)
pkg-config
~ pkg-config
> memoize handle_config; is not a slow routine per se, but we spend
> a lot of time in it, so cache the return values. The most gain
> is from configure scripts/build commands which query several modules
> at the same time. this results in as much as a 40% speedup for
> those pkg-config invocations.
> tested in a bulk/ok landry@ (jasper@)
ssh
~ auth-rsa.c ~ auth.c
~ auth.h ~ auth2-pubkey.c
~ monitor.c ~ monitor_wrap.c
~ pathnames.h ~ servconf.c
~ servconf.h ~ sshd.8
~ sshd_config ~ sshd_config.5
> allow AuthorizedKeysFile to specify multiple files, separated by spaces.
> Bring back authorized_keys2 as a default search path (to avoid breaking
> existing users of this file), but override this in sshd_config so it will
> be no longer used on fresh installs. Maybe in 2015 we can remove it
> entierly :)
> feedback and ok markus@ dtucker@ (djm@)
~ auth.c
> make secure_filename() spam debug logs less (djm@)
~ sshconnect.c
> remove extra newline (djm@)
~ sshd.8 ~ sshd_config.5
> tweak previous; ok djm (jmc@)
~ authfile.c
> read in key comments for v.2 keys (though note that these are not
> passed over the agent protocol); bz#439, based on patch from binder
> AT arago.de; ok markus@ (djm@)
~ readconf.c ~ readconf.h
~ ssh.c ~ ssh_config.5
~ sshconnect.c ~ sshconnect2.c
> Remove undocumented legacy options UserKnownHostsFile2 and
> GlobalKnownHostsFile2 by making UserKnownHostsFile/GlobalKnownHostsFile
> accept multiple paths per line and making their defaults include
> known_hosts2; ok markus (djm@)
tmux
~ cmd-if-shell.c
> Nuke a redundant if statement, from Tiago Cunha. (nicm@)
~ cmd-set-buffer.c
> Memory leak in error path, from Tiago Cunha. (nicm@)
~ tty.c
> Trim another useless if statement, from Ailin Nemui. (nicm@)
== usr.sbin ========================================================== 12/12 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/usr.sbin
usr.sbin
~ Makefile
> disable nsd until the yoyos get their shit together (deraadt@)
~ Makefile
> re-enable nsd; if it happens again nsd in-tree will be abandoned (deraadt@)
ikectl
~ ikeca.c ~ ikectl.c
~ parser.c ~ parser.h
> spacing (reyk@)
nsd
- nsd-checkconf.8 - nsd-notify.8
- nsd-patch.8 - nsd-xfer.8
- nsd-zonec.8 - nsd.8
- nsd.conf.5 - nsdc.8
> remove old man-pages (jakob@)
~ nsd-checkconf.8.in ~ nsd-notify.8.in
~ nsd-patch.8.in ~ nsd-xfer.8.in
~ nsd.8.in ~ nsd.conf.5.in
~ nsdc.8.in ~ zonec.8.in
> make sure nsd-zonec is used instead of zonec (jakob@)
~ Makefile.in
> nsd-zonec.8 (jakob@)
~ Makefile.in
> fix manpage build; from theo (jakob@)
ospfd
~ rde_spf.c
> Fix nexthop calculation for directly connected P2P links. connected
> may only be set on directly connected networks but not on router
> nodes which are used by the P2P links. Fix for PR 6597. OK sthen@
> (claudio@)
procmap
~ procmap.c
> Reimplement uvm/uvm_map.
> vmmap is designed to perform address space randomized allocations,
> without letting fragmentation of the address space go through the roof.
> Some highlights:
> - kernel address space randomization
> - proper implementation of guardpages
> - roughly 10% system time reduction during kernel build
> Tested by alot of people on tech@ and developers.
> Theo's machines are still happy. (ariane@)
relayd
~ parse.y ~ relayd.conf.5
> Support interface groups in address specifications for tables or
> directives like "listen on egress".
> Based on gilles@' code for smtpd and an idea from Mikolaj Kucharski.
> (reyk@)
~ check_script.c ~ relayd.h
> fix "check script" by sending all required information to the parent.
> (reyk@)
~ check_script.c ~ parse.y
~ relayd.h
> Add additional check to prevent running scripts when not configured.
> (reyk@)
smtpd
~ smtpd.conf.5
> fix examples so they stay do not use external utilities and do not refer
> to external domains
> prompted by deraadt@ (gilles@)
~ parse.y ~ smtpd.conf.5
> teach smtpd how to listen on an interface group so that we can do:
> listen on egress
> listen on wlan
> idea unvoluntarily suggested by Mikolaj Kucharski a few weeks ago,
> unslacked after theo suggested it again. (gilles@)
~ smtpd.conf.5
> tweak previous; (jmc@)
===============================================================================
More information about the owc
mailing list