[owc] Weekly src changes ending 2011-12-04
OWC
auto at squish.net
Mon Dec 5 07:00:01 GMT 2011
OpenBSD src changes summary for 2011-11-27 to 2011-12-04 inclusive
==================================================================
etc/rc.d/rc.subr games/fortune
gnu kerberosV
lib/libc lib/libedit
lib/librthread libexec/ld.so
libexec/makewhatis regress/libexec
regress/sbin regress/usr.bin
regress/usr.sbin sbin/disklabel
sbin/fsdb sbin/ifconfig
sbin/pfctl share/man
sys/arch/amd64/stand/biosboot sys/arch/i386/pci
sys/arch/i386/stand/biosboot sys/compat/linux
sys/dev/ic sys/dev/pci
sys/dev/pckbc sys/dev/usb
sys/kern sys/lib/libkern
sys/net sys/netinet
sys/netinet6 sys/sys
sys/uvm usr.bin/aucat
usr.bin/jot usr.bin/mandoc
usr.bin/mg usr.bin/ssh
usr.bin/systat usr.bin/tcpbench
usr.bin/tmux usr.bin/units
usr.sbin/bind usr.sbin/pkg_add
usr.sbin/sensorsd usr.sbin/smtpd
usr.sbin/user
== etc =============================================================== 01/12 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/etc
rc.d/rc.subr
~ rc.d/rc.subr
> Fix an issue when uid != euid (e.g. when running shutdown(8) as a
> regular user, member of the operator group); rm(1) was waiting for
> interactive input to remove the runfiles which made no sense, so just
> use `-f'.
> issue spotted by weerd@
> ok weerd@ robert@ (ajacoutot@)
== games ============================================================= 02/12 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/games
fortune
~ datfiles/fortunes
> Fix grammar and spelling in a Latin fortune.
> ok jmc@ (pascal@)
== gnu =============================================================== 03/12 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/gnu
gnu
~ gcc/gcc/config/pa/openbsd.h
> Restore MAX_OFILE_ALIGNMENT define (lost during the gcc3 -> gcc4 switch),
> and change ASM_OUTPUT_ALIGNED_COMMON to output an explicit .align
> directive,
> as gas will not compute it from the size (despite what comments elsewhere
> in
> gcc code say).
> This effectively makes __attribute__ ((aligned (N))) constructs work as
> intended for .bss objects.
> ok kettenis@ (miod@)
== kerberosV ========================================================= 04/12 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/kerberosV
kerberosV
~ src/kdc/hprop.8
> Remove an OpenBSD-specific tweak regarding .Xr spacing
> and make it compatible with bsd.lv mandoc and with groff-1.21.
> This tweak was originally added for compatibility with groff-1.15,
> which is no longer needed.
> ok jmc@ kristaps@ (schwarze@)
== lib =============================================================== 05/12 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/lib
libc
~ sys/send.2
> document MSG_NOSIGNAL. (fgsch@)
~ sys/reboot.2
> Remove an OpenBSD-specific tweak regarding .Xr spacing
> and make it compatible with bsd.lv mandoc and with groff-1.21.
> This tweak was originally added for compatibility with groff-1.15,
> which is no longer needed.
> ok jmc@ kristaps@ (schwarze@)
libedit
~ eln.c ~ parse.c
> Don't segfault when trying to bind to an invalid command. While here,
> kill a C++-style comment.
> ok nicm@ (pascal@)
librthread
~ Makefile
> Set the initfirst and nodelete flags on the shared library, in anticipation
> of support in ld.so (guenther@)
== libexec =========================================================== 06/12 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/libexec
ld.so
~ dlfcn.c ~ library.c
~ library_mquery.c ~ library_subr.c
~ loader.c ~ resolve.c
~ resolve.h ~ ldconfig/prebind.c
> Add support for getting some flags from DT_FLAGS_1: new flags
> DF_1_NODELETE and DF_1_INITFIRST, as well as DF_1_NOW and DF_1_GLOBAL.
> Committing for kurt@ who worked out the final version; ok guenther@ drahn@
> (guenther@)
~ ld.so.1
> Remove an OpenBSD-specific tweak regarding .Xr spacing
> and make it compatible with bsd.lv mandoc and with groff-1.21.
> This tweak was originally added for compatibility with groff-1.15,
> which is no longer needed.
> ok jmc@ kristaps@ (schwarze@)
makewhatis
~ makewhatis.8
> Remove an OpenBSD-specific tweak regarding .Xr spacing
> and make it compatible with bsd.lv mandoc and with groff-1.21.
> This tweak was originally added for compatibility with groff-1.15,
> which is no longer needed.
> ok jmc@ kristaps@ (schwarze@)
== regress =========================================================== 07/12 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/regress
libexec
~ ld.so/initfirst/Makefile ~ ld.so/initfirst/prog1/Makefile
+ ld.so/initfirst/libaa/shlib_version + ld.so/initfirst/libaa/aa.C
+ ld.so/initfirst/libaa/Makefile + ld.so/initfirst/libab/shlib_version
+ ld.so/initfirst/libab/Makefile + ld.so/initfirst/libab/ab.C
+ ld.so/initfirst/libac/shlib_version + ld.so/initfirst/libac/Makefile
+ ld.so/initfirst/libac/ac.C + ld.so/initfirst/libad/shlib_version
+ ld.so/initfirst/libad/Makefile + ld.so/initfirst/libad/ad.C
+ ld.so/initfirst/libae/shlib_version + ld.so/initfirst/libae/Makefile
+ ld.so/initfirst/libae/ae.C + ld.so/initfirst/prog1/prog1.C
+ ld.so/initfirst/prog2/prog2.C + ld.so/initfirst/prog2/Makefile
> New import:
> ld.so initfirst regress test
~ ld.so/initfirst/prog1/Makefile ~ ld.so/initfirst/prog1/expected1
~ ld.so/initfirst/prog1/expected2 ~ ld.so/initfirst/prog1/expected3
~ ld.so/initfirst/prog1/expected4
> Check the destructor order too (guenther@)
- ld.so/initfirst/Makefile.inc - ld.so/initfirst/lib.c
- ld.so/initfirst/libaa/Makefile - ld.so/initfirst/libaa/aa.C
- ld.so/initfirst/libaa/shlib_version - ld.so/initfirst/libab/Makefile
- ld.so/initfirst/libab/ab.C - ld.so/initfirst/libab/shlib_version
- ld.so/initfirst/libac/Makefile - ld.so/initfirst/libac/ac.C
- ld.so/initfirst/libac/shlib_version - ld.so/initfirst/libad/Makefile
- ld.so/initfirst/libad/ad.C - ld.so/initfirst/libad/shlib_version
- ld.so/initfirst/libae/Makefile - ld.so/initfirst/libae/ae.C
- ld.so/initfirst/libae/shlib_version - ld.so/initfirst/libif1/Makefile
- ld.so/initfirst/libif1/lib.c
- ld.so/initfirst/libif1/shlib_version
- ld.so/initfirst/libif2/Makefile - ld.so/initfirst/libif2/lib.c
- ld.so/initfirst/libif2/shlib_version
- ld.so/initfirst/libif3/Makefile - ld.so/initfirst/libif3/lib.c
- ld.so/initfirst/libif3/shlib_version
- ld.so/initfirst/libnormal/Makefile - ld.so/initfirst/libnormal/lib.c
- ld.so/initfirst/libnormal/shlib_version
- ld.so/initfirst/prog1/Makefile - ld.so/initfirst/prog1/expected1
- ld.so/initfirst/prog1/expected2 - ld.so/initfirst/prog1/expected3
- ld.so/initfirst/prog1/expected4 - ld.so/initfirst/prog1/main.c
- ld.so/initfirst/prog1/prog1.C - ld.so/initfirst/prog2/Makefile
- ld.so/initfirst/prog2/prog2.C ~ ld.so/initfirst/Makefile
+ ld.so/initfirst/test1/Makefile + ld.so/initfirst/test1/Makefile.inc
+ ld.so/initfirst/test1/lib.c
+ ld.so/initfirst/test1/libif1/Makefile
+ ld.so/initfirst/test1/libif1/lib.c
+ ld.so/initfirst/test1/libif1/shlib_version
+ ld.so/initfirst/test1/libif2/Makefile
+ ld.so/initfirst/test1/libif2/lib.c
+ ld.so/initfirst/test1/libif2/shlib_version
+ ld.so/initfirst/test1/libif3/Makefile
+ ld.so/initfirst/test1/libif3/lib.c
+ ld.so/initfirst/test1/libif3/shlib_version
+ ld.so/initfirst/test1/libnormal/Makefile
+ ld.so/initfirst/test1/libnormal/lib.c
+ ld.so/initfirst/test1/libnormal/shlib_version
+ ld.so/initfirst/test1/prog1/Makefile
+ ld.so/initfirst/test1/prog1/expected1
+ ld.so/initfirst/test1/prog1/expected2
+ ld.so/initfirst/test1/prog1/expected3
+ ld.so/initfirst/test1/prog1/expected4
+ ld.so/initfirst/test1/prog1/main.c + ld.so/initfirst/test2/Makefile
+ ld.so/initfirst/test2/libaa/Makefile
+ ld.so/initfirst/test2/libaa/aa.C
+ ld.so/initfirst/test2/libaa/shlib_version
+ ld.so/initfirst/test2/libab/Makefile
+ ld.so/initfirst/test2/libab/ab.C
+ ld.so/initfirst/test2/libab/shlib_version
+ ld.so/initfirst/test2/libac/Makefile
+ ld.so/initfirst/test2/libac/ac.C
+ ld.so/initfirst/test2/libac/shlib_version
+ ld.so/initfirst/test2/libad/Makefile
+ ld.so/initfirst/test2/libad/ad.C
+ ld.so/initfirst/test2/libad/shlib_version
+ ld.so/initfirst/test2/libae/Makefile
+ ld.so/initfirst/test2/libae/ae.C
+ ld.so/initfirst/test2/libae/shlib_version
+ ld.so/initfirst/test2/prog1/Makefile
+ ld.so/initfirst/test2/prog1/prog1.C
+ ld.so/initfirst/test2/prog2/Makefile
+ ld.so/initfirst/test2/prog2/prog2.C
> clean up initfirst regress conflict created by my import. split tests
> into two sub dirs. (kurt@)
~ ld.so/Makefile
> enable initfirst regress (kurt@)
sbin
~ pfctl/pf1.loaded ~ pfctl/pf1.ok
~ pfctl/pf1.optimized ~ pfctl/pf101.loaded
~ pfctl/pf101.ok ~ pfctl/pf13.loaded
~ pfctl/pf13.ok ~ pfctl/pf13.optimized
~ pfctl/pf19.loaded ~ pfctl/pf19.ok
~ pfctl/pf19.optimized ~ pfctl/pf2.loaded
~ pfctl/pf2.ok ~ pfctl/pf2.optimized
~ pfctl/pf20.loaded ~ pfctl/pf20.ok
~ pfctl/pf20.optimized ~ pfctl/pf24.loaded
~ pfctl/pf24.ok ~ pfctl/pf24.optimized
~ pfctl/pf33.loaded ~ pfctl/pf33.ok
~ pfctl/pf35.loaded ~ pfctl/pf35.ok
~ pfctl/pf4.loaded ~ pfctl/pf4.ok
~ pfctl/pf40.loaded ~ pfctl/pf40.ok
~ pfctl/pf41.ok ~ pfctl/pf43.loaded
~ pfctl/pf43.ok ~ pfctl/pf5.loaded
~ pfctl/pf5.ok ~ pfctl/pf5.optimized
~ pfctl/pf56.loaded ~ pfctl/pf56.ok
~ pfctl/pf56.optimized ~ pfctl/pf68.loaded
~ pfctl/pf68.ok ~ pfctl/pf68.optimized
~ pfctl/pf7.loaded ~ pfctl/pf7.ok
~ pfctl/pf7.optimized ~ pfctl/pf71.loaded
~ pfctl/pf71.ok ~ pfctl/pf71.optimized
~ pfctl/pf75.loaded ~ pfctl/pf75.ok
~ pfctl/pf75.optimized ~ pfctl/pf80.loaded
~ pfctl/pf80.ok ~ pfctl/pf80.optimized
~ pfctl/pf84.loaded ~ pfctl/pf84.ok
~ pfctl/pf84.optimized ~ pfctl/pf87.loaded
~ pfctl/pf87.ok ~ pfctl/pf87.optimized
~ pfctl/pf88.loaded ~ pfctl/pf88.ok
~ pfctl/pf89.loaded ~ pfctl/pf89.ok
~ pfctl/pf89.optimized ~ pfctl/pf97.ok
~ pfctl/pfaltq5.ok ~ pfctl/pfaltq6.ok
> adjust for port numbers instead of names by default
> From: Lawrence Teo <lteo at devio at us> (henning@)
usr.bin
~ mandoc/mdoc/Bd/Makefile + mandoc/mdoc/Bd/spacing.in
+ mandoc/mdoc/Bd/spacing.out_ascii
> test vertical spacing around display blocks (schwarze@)
~ mandoc/char/N/basic.out_ascii ~ mandoc/char/hyphen/man.out_ascii
~ mandoc/char/space/leading-man.out_ascii
~ mandoc/char/space/tab-man.out_ascii ~ mandoc/man/BI/literal.out_ascii
~ mandoc/man/IP/literal.out_ascii ~ mandoc/man/IP/manyargs.out_ascii
~ mandoc/man/IP/spacing.out_ascii ~ mandoc/man/IP/width.out_ascii
~ mandoc/man/PP/empty.out_ascii ~ mandoc/man/TH/baddate.out_ascii
~ mandoc/man/TH/emptydate.out_ascii ~ mandoc/man/TH/isodate.out_ascii
~ mandoc/man/TH/mdocdate.out_ascii ~ mandoc/man/TH/onlyyear.out_ascii
~ mandoc/man/TP/literal.out_ascii ~ mandoc/man/TP/macrotag.out_ascii
~ mandoc/man/TP/manyargs.out_ascii ~ mandoc/man/TP/width.out_ascii
~ mandoc/man/nf/args.out_ascii ~ mandoc/man/nf/indent.out_ascii
~ mandoc/roff/args/man.out_ascii ~ mandoc/roff/args/roff.out_ascii
~ mandoc/roff/br/args.out_ascii ~ mandoc/roff/cond/ie.out_ascii
~ mandoc/roff/cond/if.out_ascii ~ mandoc/roff/na/args.out_ascii
~ mandoc/roff/ps/ignore.out_ascii
~ mandoc/roff/sp/badargs-man.out_ascii
~ mandoc/roff/string/escape.out_ascii
~ mandoc/roff/string/infinite.out_ascii
~ mandoc/roff/string/zerolength.out_ascii
> After unifying volume titles in the page headers for man(7) and mdoc(7),
> adjust the expected man(7) output to the new convention.
> Related to src/usr.bin/mandoc/man_validate.c rev. 1.51. (schwarze@)
~ mandoc/mdoc/Xr/Makefile ~ mandoc/mdoc/Xr/args.in
~ mandoc/mdoc/Xr/args.out_ascii
> Remove an OpenBSD-specific tweak regarding .Xr spacing
> and make it compatible with bsd.lv mandoc and with groff-1.21.
> This tweak was originally added for compatibility with groff-1.15,
> which is no longer needed.
> ok jmc@ kristaps@ (schwarze@)
+ mandoc/mdoc/Ap/Makefile + mandoc/mdoc/Ap/middle.out_ascii
+ mandoc/mdoc/Ap/middle.in + mandoc/mdoc/Aq/Makefile
+ mandoc/mdoc/Aq/empty.in + mandoc/mdoc/Aq/empty.out_ascii
+ mandoc/mdoc/Brq/Makefile + mandoc/mdoc/Brq/empty.in
+ mandoc/mdoc/Brq/empty.out_ascii + mandoc/mdoc/Dq/Makefile
+ mandoc/mdoc/Dq/empty.in + mandoc/mdoc/Dq/empty.out_ascii
+ mandoc/mdoc/Dv/Makefile + mandoc/mdoc/Dv/font.out_ascii
+ mandoc/mdoc/Dv/font.in + mandoc/mdoc/Ev/Makefile
+ mandoc/mdoc/Ev/font.out_ascii + mandoc/mdoc/Ev/font.in
+ mandoc/mdoc/Qq/Makefile + mandoc/mdoc/Qq/empty.in
+ mandoc/mdoc/Qq/empty.out_ascii + mandoc/mdoc/Sq/Makefile
+ mandoc/mdoc/Sq/empty.in + mandoc/mdoc/Sq/empty.out_ascii
+ mandoc/mdoc/Ux/Makefile + mandoc/mdoc/Ux/spacing.out_ascii
+ mandoc/mdoc/Ux/spacing.in
> New import:
> test some simple macros; written while working on -Tman
~ mandoc/mdoc/Makefile
> link new tests to the build: Ap Aq Brq Dq Dv Ev Qq Sq Ux (schwarze@)
usr.sbin
~ pkg_add/Makefile
> pure display check, not linked since it fails, but the reported message
> should be better. (espie@)
== sbin ============================================================== 08/12 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/sbin
disklabel
~ editor.c
> Fix negating of unsigned d_secsize. FIRST, cast to signed type,
> THEN negate. Fixes issues with 'b' and 'k' size suffixes. Make error
> message more clear while here.
> Found, and fix provided, by David Imhoff via bugs at . Thanks! (krw@)
fsdb
~ fsdb.8
> Remove an OpenBSD-specific tweak regarding .Xr spacing
> and make it compatible with bsd.lv mandoc and with groff-1.21.
> This tweak was originally added for compatibility with groff-1.15,
> which is no longer needed.
> ok jmc@ kristaps@ (schwarze@)
ifconfig
~ ifconfig.8
> Manpage bits for the recent changes in vlan(4) prio handling.
> ok jmc henning sthen claudio (haesbaert@)
~ ifconfig.c ~ ifconfig.8
> Cleanup recently removed flags from ifconfig.c and its manpage.
> ok jmc@ mikeb@ (haesbaert@)
pfctl
~ pfctl.c
> Avoid loading garbage hostid and other values not always initialised,
> by bzero()ing the 'struct pfsync' properly.
> ok dlg mpf (mcbride@)
~ parse.y ~ pfctl.c
~ pfctl_parser.h
> pfctl_set_hostid always returns 0; don't pretend otherwise and make it a
> void function instead.
> ok dlg (mcbride@)
== share ============================================================= 09/12 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/share
man
~ man4/vlan.4 ~ man5/pf.conf.5
> Manpage bits for the recent changes in vlan(4) prio handling.
> ok jmc henning sthen claudio (haesbaert@)
~ man5/moduli.5
> fix formula
> ok djm@ (eric@)
~ man9/mbuf.9
> this diff introduces the MAXMCLBYTES macro to describe the largest
> cluster the generic network stack will be able to give you.
> it also recognises that external storage on an mbuf may be bigger than
> MCLBYTES. its only when m_pullup or m_pulldown need to allocate
> another cluster that they now check the len argument, and now they
> do it against MAXMCLBYTES.
> this is required for me to do pfsync on jumbo frames as the m_pulldown
> for the subregions fail beyond MCLBYTES into the packet.
> ok deraadt@ mikeb@ henning@ blambert@
> manpage changes ok jmc@ (dlg@)
~ man7/man.7
> In man(7), when no explicit volume name is given, use the default
> volume name for the respective manual section, just like in mdoc(7).
> This gives us nicer page headers for cvs(1), lynx(1), tic(1),
> mkhybrid(8), and many curses(3) manuals.
> ok kristaps@
> To not break compatibility, i wrote a corresponding patch for GNU troff
> which Werner Lemberg accepted upstream at rev. 1.65 of:
> http://cvs.savannah.gnu.org/viewvc/groff/tmac/an-old.tmac?root=groff
> (schwarze@)
~ man9/mbuf.9
> it's MAXMCLBYTES, not MAXCLBYTES. (dlg@)
~ man5/bsd.port.mk.5
> a few pieces of gnu.port.mk were not actually documented/documented
> correctly,
> as noticed by Amit Kulkarni. (espie@)
~ man5/port-modules.5
> Add MODGNOME_ICON_CACHE and MODXFCE_ICON_CACHE. (ajacoutot@)
~ man5/port-modules.5
> better format (.Ev, .Ar, .Li)... rephrase a few odd things.
> Synch with current reality: a few variables got lost since bsd.port.arch.mk
> is included after modules (to be fixed ?) (espie@)
~ man5/port-modules.5
> clean-up description of lang/ghc a bit more (espie@)
~ man5/port-modules.5
> format most modules, exempts ruby since jeremy has a pending diff (espie@)
~ man7/packages.7
> fussing around (espie@)
~ man5/bsd.port.mk.5 ~ man5/port-modules.5
> tweak previous; (jmc@)
~ man5/bsd.port.mk.5
> more OBSOLETE stuff near the end, as it's less relevant
> build-depends is less important and may vanish.
> prepare is more important... (espie@)
~ man5/bsd.port.mk.5
> Builds -> Build.
> flesh out prepare with reference to MULTI_PACKAGES. (espie@)
~ man5/bsd.port.mk.5
> thinko (espie@)
~ man5/bsd.port.mk.5
> more .Ev -> .Ar
> explicit link to THE FAKE FRAMEWORK
> reflect reality: *-fetch hooks are gone. (espie@)
~ man5/bsd.port.mk.5
> fix typo (espie@)
~ man7/ports.7
> document prepare instead of depends (espie@)
~ man5/bsd.port.mk.5
> Nits (espie@)
~ man4/gif.4 ~ man4/lc.4
~ man4/netintro.4 ~ man4/sm.4
~ man4/we.4 ~ man4/man4.sparc/mgx.4
~ man4/man4.sparc/tvtwo.4 ~ man4/man4.sparc64/mgx.4
~ man4/man4.sparc64/tvtwo.4 ~ man5/bsd.port.mk.5
> Remove an OpenBSD-specific tweak regarding .Xr spacing
> and make it compatible with bsd.lv mandoc and with groff-1.21.
> This tweak was originally added for compatibility with groff-1.15,
> which is no longer needed.
> ok jmc@ kristaps@ (schwarze@)
~ man5/bsd.port.mk.5
> more nits (espie@)
~ man5/bsd.port.mk.5
> a few more nits, such as more consistent annotations (espie@)
~ man5/bsd.port.mk.5
> nit, nit, nit (espie@)
~ man5/bsd.port.mk.5
> more nits (espie@)
~ man5/port-modules.5
> document qts (espie@)
~ man5/port-modules.5
> Update ruby section for recent and not-so-recent changes to ruby.port.mk.
> Reformat ruby section to be consistent with other recent changes.
> OK espie@ (jeremy@)
== sys =============================================================== 10/12 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/sys
arch/amd64/stand/biosboot
~ biosboot.8
> Remove an OpenBSD-specific tweak regarding .Xr spacing
> and make it compatible with bsd.lv mandoc and with groff-1.21.
> This tweak was originally added for compatibility with groff-1.15,
> which is no longer needed.
> ok jmc@ kristaps@ (schwarze@)
arch/i386/pci
~ pci_machdep.c
> Do not check for an HyperTransport bus if we are using mode #2. Prevents a
> safety panic from occuring.
> Found the hard way by nick@ (miod@)
~ pci_machdep.c
> Slightly different fix for the problem fixed by the previous commit for
> better
> diffability with the amd64 codebase. (kettenis@)
arch/i386/stand/biosboot
~ biosboot.8
> Remove an OpenBSD-specific tweak regarding .Xr spacing
> and make it compatible with bsd.lv mandoc and with groff-1.21.
> This tweak was originally added for compatibility with groff-1.15,
> which is no longer needed.
> ok jmc@ kristaps@ (schwarze@)
compat/linux
~ linux_socket.c ~ linux_socket.h
> add support for MSG_NOSIGNAL.
> linux bits compiled on i386 by sebastia@, mikeb@ ok (fgsch@)
dev/ic
~ pgt.c
> Be sure to splx() before return in pgt_media_status(); from Jan Klemkow on
> tech@ (miod@)
~ trm.c
> Make sure splx() is called before exiting trm_StartWaitingSRB().
> Found by Jan Klemkow. Thanks! (krw@)
dev/pci
~ if_ix.c
> Checksum fix from reyk (rev1.31) that was lost during update:
> Fix the IP ckecksum offloading logic that disables and breaks offloading
> if the packet is neither TCP nor UDP because of an erroneous "default"
> case.
> No functional change in the default build because IP checksum offloading
> is currently disabled in ix(4).
> Tested on 82598 and 82599 (though checksums are still disabled by default).
> ok claudio (mikeb@)
~ if_ix.c ~ ixgbe_type.h
> Bring back low latency interrupt moderation for 82599
> lost during update. Originally from claudio, rev1.46.
> ok claudio (mikeb@)
~ if_myx.c
> Fix reversed error-handling gotos in myx_buf_fill(), which would lead to
> either an mbuf leak or a NULL pointer dereference.
> ok sthen@ claudio@ dlg@
> testing claudio@ dlg@ (blambert@)
~ if_em_hw.c
> Limit the kumeran errata to a specific mac/phy combo as done in the Intel
> code instead of testing only the phy. Fixes autoneg on ich9.
> From Jonathan Perkin (jsg@)
~ if_vic.c
> Do not blindly return 1 from the interrupt handler since we do not know
> if we actually did any work. Otherwise devices sharing this interrupt will
> not have their interrupt handler run, possibly leading to the kernel
> spinning in the vic(4) interrupt handler.
> ok dlg@ (jsing@)
dev/pckbc
~ pms.c ~ pmsreg.h
> Correctly handle clitpad packets for some ALPS models.
> reported by espie@, not reported by miod@
> with and ok shadchin@, ok espie@ (mpi@)
~ pms.c
> renaming synaptics_pt_* -> pms_sec_*
> (synaptics and alps uses this for trackpoint)
> ok mpi@ (shadchin@)
dev/usb
~ ueagle.c
> This needs sys/kthread.h to build (claudio@)
~ ucom.c
> the code for managing tty vs cua usage collision was free'ing all sorts
> of critical structures when it hit a collision. Also, ttyopen is ldisc
> 0, so it should not be called twice (and furthermore, with the wrong dev_t)
> original report from mikeb, ok dlg mikeb and discussed with miod (deraadt@)
~ usbdevs
> Add a bunch of Apple iPod and iPhone devices (dhill@)
~ usbdevs.h ~ usbdevs_data.h
> regen (dhill@)
~ uaudio.c ~ usb_quirks.c
> Mark Apple iPod and iPhone products with UQ_BAD_HID and
> UAUDIO_FLAG_BAD_AUDIO.
> This allows these devices to attach as ugen, so at least we can use some
> apps,
> like gphoto2 to retrieve media.
> requested by a few, ok mikeb@ (dhill@)
~ if_wi_usb.c
> Make sure splx() is called before exiting wi_usb_tx_lock_try().
> Found by Jan Klemkow. Thanks! (krw@)
~ ukbd.c ~ usbhid.h
> Add support for Apple fn keys
> based on a diff from jcs@, ok miod@ (mpi@)
kern
~ vfs_vnops.c
> When checking for offset wrap around in vn_read(), compare against
> LLONG_MAX instead of SSIZE_MAX
> ok deraadt@ (guenther@)
~ subr_hibernate.c
> Only free the piglet on hibernate fail if we are suspending. (mlarkin@)
~ subr_hibernate.c
> ready this for big-endian support later on
> ok mlarkin (deraadt@)
~ uipc_mbuf.c
> whitespace fixes. no binary change. (dlg@)
~ uipc_mbuf.c
> correctly calculate the space available in external storage in m_pullup.
> ok deraadt@ claudio@ blambert@ mikeb@ (dlg@)
~ uipc_mbuf.c ~ uipc_mbuf2.c
> this diff introduces the MAXMCLBYTES macro to describe the largest
> cluster the generic network stack will be able to give you.
> it also recognises that external storage on an mbuf may be bigger than
> MCLBYTES. its only when m_pullup or m_pulldown need to allocate
> another cluster that they now check the len argument, and now they
> do it against MAXMCLBYTES.
> this is required for me to do pfsync on jumbo frames as the m_pulldown
> for the subregions fail beyond MCLBYTES into the packet.
> ok deraadt@ mikeb@ henning@ blambert@
> manpage changes ok jmc@ (dlg@)
~ uipc_mbuf.c
> dont put MAXMCLBYTES in the mclsizes array with a comment saying its 64k
> when the macro can change without automatically fixing the comment.
> instead add a diagnostic that checks that the biggest cluster size is
> always MAXMCLBYTES.
> requested by and ok kettenis@ (dlg@)
~ uipc_syscalls.c
> add support for MSG_NOSIGNAL.
> linux bits compiled on i386 by sebastia@, mikeb@ ok (fgsch@)
lib/libkern
~ arch/hppa64/bcopy.m4
> Replace a few more 32-bit instructions with the proper 64-bit equivalent.
> Also fix the return value of memcpy. With these changes, this seems to
> work as advertised now. (kettenis@)
~ arch/hppa64/bcopy.S ~ arch/hppa64/spcopy.S
> Regen. (kettenis@)
~ arch/hppa/bcopy.m4
> Oops, jsing@ changed spcopy.S without realizing that that file is generated
> from bcopy.m4. Fix that. (kettenis@)
~ arch/hppa/bcopy.S
> Regen. (kettenis@)
~ arch/hppa/bcopy.m4
> Fix return value of memcpy. (kettenis@)
~ arch/hppa64/bcopy.m4
> Switch to proper PA-RISC 2.0 mnemonics. Works around a bug in the
> assembler
> getting rid of the (hopefully) last bug in this code. (kettenis@)
~ arch/hppa64/bcopy.S ~ arch/hppa64/spcopy.S
> Regen. (kettenis@)
net
~ if_vlan.c
> Kill old comment, we already do 802.1p tagging. (haesbaert@)
~ if_pfsync.c
> Protect more operations in the pfsync_clone_destroy to prevent
> accidental race conditions. From Erik Lax, thanks! ok dlg (mikeb@)
~ pf.c ~ pf_ioctl.c
~ pfvar.h
> deprecate PFTM_UNTIL_PACKET. nothing in the tree uses it, and
> pf_state_expires() handled it incorrectly.
> discussed with mikeb@
> ok henning@ (dlg@)
~ if_pfsync.c ~ pf_ioctl.c
~ pfvar.h
> use a u_int64_t for the state id in pfsync_state. this makes it consistent
> with every other thing that stores the state id (including other pfsync
> messages).
> includes improvements to the systat code to consider the creatorid as well
> as the state id in its cache to avoid collisions between states created on
> different hosts.
> tested by me in production and on amd64 talking to sparc64.
> ok henning@ (dlg@)
~ if_pfsync.c
> Make sure we only enter pf_route() when undefering in the PF_ROUTETO case.
> ok dlg claudio (mcbride@)
~ if.h ~ pf.c
> Kill unused IFCAP_IPSEC and IFCAP_IPCOMP.
> ok claudio@ henning@ mikeb@ (haesbaert@)
netinet
~ ip_output.c
> Kill unused IFCAP_IPSEC and IFCAP_IPCOMP.
> ok claudio@ henning@ mikeb@ (haesbaert@)
netinet6
~ nd6.c
> Kill unused IFCAP_IPSEC and IFCAP_IPCOMP.
> ok claudio@ henning@ mikeb@ (haesbaert@)
sys
~ hibernate.h
> ready this for big-endian support later on
> ok mlarkin (deraadt@)
~ param.h
> this diff introduces the MAXMCLBYTES macro to describe the largest
> cluster the generic network stack will be able to give you.
> it also recognises that external storage on an mbuf may be bigger than
> MCLBYTES. its only when m_pullup or m_pulldown need to allocate
> another cluster that they now check the len argument, and now they
> do it against MAXMCLBYTES.
> this is required for me to do pfsync on jumbo frames as the m_pulldown
> for the subregions fail beyond MCLBYTES into the packet.
> ok deraadt@ mikeb@ henning@ blambert@
> manpage changes ok jmc@ (dlg@)
~ socket.h
> add support for MSG_NOSIGNAL.
> linux bits compiled on i386 by sebastia@, mikeb@ ok (fgsch@)
uvm
~ uvm_pmemrange.c
> Be sure not to access the vm_page array out of bounds in
> uvm_pmr_freepages().
> Among other things, this fixes early panics on hppa system which memory
> size
> is exactly 128MB.
> Found the hard way and reported by fries@, not reported by beck@ (miod@)
== usr.bin =========================================================== 11/12 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/usr.bin
aucat
~ wav.c
> don't add .wav header to .syx files (ratchov@)
~ dev.h
> remove unused 'reqrate' parameter (ratchov@)
~ dev.c
> initialize device reference counter, unbreaks -aoff
> with MALLOC_OPTIONS=J (ratchov@)
~ aproc.h ~ aucat.c
~ dev.c ~ dev.h
~ midi.c ~ midi.h
> reuse midi-control code to implement midi thru boxes and remove
> the old midithru implementation; less code, less bugs. As a side
> effect, midi output doesn't implement running status "compression"
> any more. (ratchov@)
~ aucat.c
> don't use a "server" flag, just check if we expose sub-devices (ratchov@)
jot
~ jot.1
> Remove an OpenBSD-specific tweak regarding .Xr spacing
> and make it compatible with bsd.lv mandoc and with groff-1.21.
> This tweak was originally added for compatibility with groff-1.15,
> which is no longer needed.
> ok jmc@ kristaps@ (schwarze@)
mandoc
~ mandocdb.c
> Rudimentary handling of formatted manuals ("cat pages").
> Coded on the train back from p2k11 in Budapest.
> Kristaps has seen the patch and agreed with the direction. (schwarze@)
~ apropos_db.c ~ apropos_db.h
~ mandocdb.c
> Save the manual type (mdoc, man, or cat) in the index file
> of the mandoc databases, as suggested by kristaps at .
> Given the well-structured code, this is surprisingly simple.
> This changes the mandoc.index database format.
> Run "sudo mandocdb" to regenerate your databases. (schwarze@)
~ mandocdb.c
> Reimplement the global command line options -a and -v
> as static global variables, reducing the maze of arguments
> passed around among various static functions.
> Suggested by kristaps at . (schwarze@)
~ apropos.c ~ apropos_db.c
~ apropos_db.h ~ main.c
> Implement whatis(1) as a special apropos(1) mode as a part of
> the mandoc(1) binary; not yet enabled for the general public.
> Intended to replace src/usr.bin/whatis at a later time.
> Coded by kristaps@, with a few tweaks by me.
> To test this:
> $ mandocdb # unless you have already done so earlier
> $ sudo ln -s /usr/bin/mandoc /usr/bin/whatis.m
> $ whatis.m mandoc apropos whatis
> $ whatis.m man (schwarze@)
~ mandocdb.c
> Put back the chdir(2) to the right man page tree before parsing manuals;
> this got lost during the bsd.lv sync in rev. 1.10; oops.
> This is required for processing .so links: otherwise, several files
> in Xenocara fail to parse and cause ugly complaints instead. (schwarze@)
~ mandocdb.8
> Discuss the default behaviour up front before talking about options
> modifying it; based on a remark by kristaps at .
> While here, mention parsing of unformatted files
> and the changed index format and fix a few minor issues. (schwarze@)
~ apropos.1
> tables of matchable keys; from kristaps@ (schwarze@)
~ apropos.1 ~ apropos.c
~ apropos_db.c ~ mandocdb.8
~ mandocdb.c
> Sync to bsd.lv, mostly from kristaps at .
> Make `-i' only apply to regular expressions.
> For substring matches, always use strcasestr().
> Also, improve some manuals and comments. (schwarze@)
~ mandocdb.8 ~ mandocdb.c
> Fix mandocdb(8) to pass over the type when pruning the database.
> This fixed `-d' perpetually adding the same files.
> While here, remove -vv, clean up the code and document it.
> From kristaps at . (schwarze@)
~ libmandoc.h ~ libmdoc.h
~ man_validate.c ~ mdoc_validate.c
~ msec.c
> In man(7), when no explicit volume name is given, use the default
> volume name for the respective manual section, just like in mdoc(7).
> This gives us nicer page headers for cvs(1), lynx(1), tic(1),
> mkhybrid(8), and many curses(3) manuals.
> ok kristaps@
> To not break compatibility, i wrote a corresponding patch for GNU troff
> which Werner Lemberg accepted upstream at rev. 1.65 of:
> http://cvs.savannah.gnu.org/viewvc/groff/tmac/an-old.tmac?root=groff
> (schwarze@)
~ apropos_db.c ~ mandocdb.8
~ mandocdb.c
> Make the mandocdb(8) format endian-neutral by storing integer data
> in network byte order; from kristaps at .
> Hopefully, this finishes the database format.
> This commit requires another rebuild of your mandoc databases
> by running "sudo mandocdb". (schwarze@)
~ apropos_db.c ~ mdoc_macro.c
~ mdoc_validate.c
> remove useless "#ifdef __linux__" that crept in,
> and trivial sync to bsd.lv (two new comments) (schwarze@)
~ mdoc_macro.c
> Remove an OpenBSD-specific tweak regarding .Xr spacing
> and make it compatible with bsd.lv mandoc and with groff-1.21.
> This tweak was originally added for compatibility with groff-1.15,
> which is no longer needed.
> ok jmc@ kristaps@ (schwarze@)
~ man_html.c ~ man_macro.c
~ man_term.c
> When a man document contains nothing at all except one or more invalid
> macros, do not die on an assertion, but show correct error messages.
> Assertions of meta data validity suggested by joerg at .
> ok joerg@ (schwarze@)
~ mandocdb.c
> Fix parsing of file names given on the command line; i broke it
> when adding support for formatted manual pages. (schwarze@)
mg
~ Makefile ~ cinfo.c
~ def.h ~ funmap.c
~ keymap.c ~ main.c
~ mg.1 + tags.c
> Add some ctags support to mg. From Sunil Nimmagadda.
> Man page review and suggestions from jmc@
> Revewied and tested by myself, and ok deraadt@ (lum@)
~ paragraph.c
> Change behaviour of M-q (fill-paragraph) to allow
> double-space after /[.?!]\)?/
> From a diff sent to tech@ by Matthew Dempsky some
> months ago. (lum@)
~ mg.1
> jmc's suggested text for the ctags man page update was slighty
> different in the committed version. Change to his original text,
> requested by him. (lum@)
~ paragraph.c
> Fix a few grating comment typos. (matthew@)
~ paragraph.c
> Missing wordlen check from version 1.20. Spotted by matthew@ (lum@)
ssh
~ mux.c
> fix bz#1948: ssh -f doesn't fork for multiplexed connection.
> ok dtucker@ (djm@)
~ mac.c
> fix bz#1934: newer OpenSSL versions will require HMAC_CTX_Init before
> HMAC_init (this change in policy seems insane to me)
> ok dtucker@ (djm@)
~ mux.c
> revert:
> > revision 1.32
> > date: 2011/12/02 00:41:56; author: djm; state: Exp; lines: +4 -1
> > fix bz#1948: ssh -f doesn't fork for multiplexed connection.
> > ok dtucker@
> it interacts badly with ControlPersist (djm@)
systat
~ cache.c ~ cache.h
> use a u_int64_t for the state id in pfsync_state. this makes it consistent
> with every other thing that stores the state id (including other pfsync
> messages).
> includes improvements to the systat code to consider the creatorid as well
> as the state id in its cache to avoid collisions between states created on
> different hosts.
> tested by me in production and on amd64 talking to sparc64.
> ok henning@ (dlg@)
~ cache.h
> minor whitespace fix (dlg@)
tcpbench
~ tcpbench.c
> Fix some warnx() calls which should be fprintf plus KNF. (haesbaert@)
tmux
~ status.c
> Make M-f and M-b work the same at the command prompt as in copy mode,
> pointed out by Romain Francoise. (nicm@)
~ mode-key.c
> Add home and end keys for command prompt, from Matthias Lederhofer. (nicm@)
~ tty-term.c
> Log terminal overrides. (nicm@)
~ mode-key.c ~ tmux.1
~ tmux.h ~ window-copy.c
> Support "jump to" like vi in copy mode using t and T keys. Also add x
> and X for delete in edit mode.
> From Ben Boeckel, thanks. (nicm@)
units
~ units.lib
> update currency exchange rates; (jmc@)
== usr.sbin ========================================================== 12/12 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/usr.sbin
bind
~ bin/named/query.c ~ lib/dns/rbtdb.c
TAGGED OPENBSD_5_0
> MFC
> ok deraadt@
> ----------------------------------------------
> query.c
> revision 1.9
> date: 2011/11/17 13:11:20; author: sthen; state: Exp; lines: +8 -11
> Apply the patches from BIND upstream to mitigate cve-2011-4313 (crashes
> with INSIST(! dns_rdataset_isassociated(sigrdataset))).
> rbtdb.c
> revision 1.7
> date: 2011/11/17 13:11:20; author: sthen; state: Exp; lines: +5 -2
> Apply the patches from BIND upstream to mitigate cve-2011-4313 (crashes
> with INSIST(! dns_rdataset_isassociated(sigrdataset))). (pea@)
~ bin/named/query.c ~ lib/dns/rbtdb.c
TAGGED OPENBSD_4_9
> MFC
> ok sthen@
> ----------------------------------------------
> query.c
> revision 1.9
> date: 2011/11/17 13:11:20; author: sthen; state: Exp; lines: +8 -11
> Apply the patches from BIND upstream to mitigate cve-2011-4313 (crashes
> with INSIST(! dns_rdataset_isassociated(sigrdataset))).
> rbtdb.c
> revision 1.7
> date: 2011/11/17 13:11:20; author: sthen; state: Exp; lines: +5 -2
> Apply the patches from BIND upstream to mitigate cve-2011-4313 (crashes
> with INSIST(! dns_rdataset_isassociated(sigrdataset))). (pea@)
pkg_add
~ pkg_delete.1 ~ OpenBSD/PkgDelete.pm
> -X exclude option: negates the packages we give (remove them instead of
> adding them). (espie@)
~ pkg_delete.1 ~ OpenBSD/PkgDelete.pm
> -X before -x; (jmc@)
~ OpenBSD/PkgDelete.pm
> small logic error (espie@)
~ OpenBSD/CollisionReport.pm ~ OpenBSD/Add.pm
> In a collision, retrofit storage for 2nd pkgname.
> (in large updatesets, it may not be obvious where the 2nd copy comes from,
> as exemplified by ajacoutot@). (espie@)
~ OpenBSD/CollisionReport.pm
> simplify/generalize (espie@)
sensorsd
~ sensorsd.conf.5
> Remove an OpenBSD-specific tweak regarding .Xr spacing
> and make it compatible with bsd.lv mandoc and with groff-1.21.
> This tweak was originally added for compatibility with groff-1.15,
> which is no longer needed.
> ok jmc@ kristaps@ (schwarze@)
smtpd
~ smtpd.h
> fix STATE_COUNT count
> ok eric@ gilles@ (chl@)
~ filter.c ~ filter.h
~ mfa_session.c
> use STATUS_* filter codes, instead of -1/0/1 hard coded values
> ok gilles@ (chl@)
user
~ user.c ~ usermod.8
> Add a new '-S' option to _set_ the user's secondary group(s) (which also
> permits to remove a user from a group).
> It uses the same syntax as `-G'.
> In practice this works like `-G' on Linux; it was decided not to modify
> our `-G' option because people rely on the existing behavior (which
> works like `-aG' on Linux, i.e. append to group).
> Reworked from scratch from a submission by Frank Brodbeck.
> man page bits reviewed by jmc@
> ok miod@ (ajacoutot@)
===============================================================================
More information about the owc
mailing list