OpenBSD src changes summary for 2017-09-03 to 2017-09-10 inclusive ==================================================================
bin/ksh bin/pax distrib/sets distrib/syspatch gnu/usr.bin/perl include/ctype.h include/langinfo.h include/locale.h include/stdlib.h include/string.h include/strings.h include/time.h include/wchar.h include/wctype.h lib/libc lib/libcompiler_rt lib/libcrypto lib/libcxx lib/librthread libexec/ld.so regress/lib regress/sbin regress/sys sbin/dhclient sbin/pfctl sbin/pflogd sbin/route share/man sys/arch/alpha/alpha sys/arch/alpha/include sys/arch/alpha/isa sys/arch/alpha/pci sys/arch/alpha/stand/netboot sys/arch/amd64/amd64 sys/arch/amd64/include sys/arch/amd64/pci sys/arch/amd64/stand/boot sys/arch/amd64/stand/cdboot sys/arch/amd64/stand/efiboot sys/arch/amd64/stand/libsa sys/arch/amd64/stand/pxeboot sys/arch/arm/arm sys/arch/arm/include sys/arch/arm64/arm64 sys/arch/arm64/include sys/arch/arm64/stand/efiboot sys/arch/armv7/armv7 sys/arch/armv7/omap sys/arch/armv7/stand/efiboot sys/arch/armv7/sunxi sys/arch/hppa/hppa sys/arch/hppa/include sys/arch/hppa/stand/boot sys/arch/i386/eisa sys/arch/i386/i386 sys/arch/i386/include sys/arch/i386/isa sys/arch/i386/pci sys/arch/i386/stand/boot sys/arch/i386/stand/cdboot sys/arch/i386/stand/libsa sys/arch/i386/stand/pxeboot sys/arch/landisk/dev sys/arch/landisk/landisk sys/arch/landisk/stand/boot sys/arch/luna88k/cbus sys/arch/luna88k/dev sys/arch/m88k/include sys/arch/m88k/m88k sys/arch/macppc/dev sys/arch/macppc/macppc sys/arch/macppc/pci sys/arch/mips64/include sys/arch/octeon/dev sys/arch/powerpc/include sys/arch/sgi/dev sys/arch/sgi/gio sys/arch/sgi/stand/sgivol sys/arch/sgi/xbow sys/arch/sh/include sys/arch/socppc/socppc sys/arch/socppc/stand/boot sys/arch/sparc64/dev sys/arch/sparc64/include sys/ddb sys/dev sys/dev/acpi sys/dev/eisa sys/dev/fdt sys/dev/ic sys/dev/isa sys/dev/microcode sys/dev/pci sys/dev/pcmcia sys/dev/puc sys/dev/pv sys/dev/rasops sys/dev/sbus sys/dev/usb sys/dev/videomode sys/dev/wscons sys/dev/wsfont sys/isofs/cd9660 sys/isofs/udf sys/kern sys/lib/libkern sys/lib/libsa sys/net sys/net80211 sys/netinet sys/netinet6 sys/nfs sys/scsi sys/sys sys/tmpfs sys/ufs/ext2fs usr.bin/ctfconv usr.bin/ftp usr.bin/lock usr.bin/mandoc usr.bin/mg usr.bin/pctr usr.bin/sdiff usr.bin/ssh usr.bin/tftp usr.bin/tmux usr.sbin/ntpd usr.sbin/pwd_mkdb usr.sbin/quot usr.sbin/smtpd usr.sbin/tcpdump usr.sbin/vmctl usr.sbin/vmd == bin =============================================================== 01/12 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/bin ksh ~ c_ksh.c ~ path.c ~ sh.h > Use PATH_MAX instead of a local define > okk deraadt@ millert@ anton@ (jca@) ~ history.c > Keep histptr inside of the allocated history array > histptr == history -1 means that the history is empty, but the behavior > is then undefined. Allocate an extra array entry to work around this. > Input & ok tb@ (jca@) pax ~ cpio.c > intvar < sizeof(...) does not catch negative values since an int -> > unsigned > conversion is involved. Cast the sizeof to int to get a signed compare; > ok deraadt@ bluhm@ (otto@) ~ tar.c > Avoid clang warning and make code better by using a signed long; > with hint from millert@; ok millert@ guenther@ (otto@) ~ pat_rep.c > Backslash escapes the next character in filename patterns. > ok millert@ (guenther@) == distrib =========================================================== 02/12 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/distrib sets ~ lists/base/md.amd64 ~ lists/base/md.arm64 ~ lists/base/md.i386 ~ lists/base/mi ~ lists/comp/mi > New POSIX xlocale implementation written from scratch. > Complete in the sense that all POSIX *locale(3) and *_l(3) functions > are included, but in OpenBSD, we of course only really care about > LC_CTYPE and we only support ASCII and UTF-8. > With important help from kettenis@, guenther@, and jca@. > Repeated testing in ports bulk builds by naddy@. > Additional testing by jca@, sebastia@, dcoppa@, and others. > OK kettenis@ dcoppa@, and guenther@ on an earlier version. > Riding guenther@'s libc/librthread major bump. (schwarze@) ~ lists/base/mi > sync (deraadt@) syspatch ~ README > Previous fakeroot must be manually copied in case an errata does not apply > to the current architecture. (ajacoutot@) == gnu =============================================================== 03/12 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/gnu usr.bin/perl ~ cpan/OpenBSD-Pledge/Pledge.xs ~ cpan/OpenBSD-Pledge/lib/OpenBSD/Pledge.pm ~ cpan/OpenBSD-Pledge/t/OpenBSD-Pledge.t > Remove path support from perl OpenBSD::Pledge > ok guenther@ deraadt@ giovanni@ (afresh1@) == include =========================================================== 04/12 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/include ctype.h ~ ctype.h > New POSIX xlocale implementation written from scratch. > Complete in the sense that all POSIX *locale(3) and *_l(3) functions > are included, but in OpenBSD, we of course only really care about > LC_CTYPE and we only support ASCII and UTF-8. > With important help from kettenis@, guenther@, and jca@. > Repeated testing in ports bulk builds by naddy@. > Additional testing by jca@, sebastia@, dcoppa@, and others. > OK kettenis@ dcoppa@, and guenther@ on an earlier version. > Riding guenther@'s libc/librthread major bump. (schwarze@) langinfo.h ~ langinfo.h > New POSIX xlocale implementation written from scratch. > Complete in the sense that all POSIX *locale(3) and *_l(3) functions > are included, but in OpenBSD, we of course only really care about > LC_CTYPE and we only support ASCII and UTF-8. > With important help from kettenis@, guenther@, and jca@. > Repeated testing in ports bulk builds by naddy@. > Additional testing by jca@, sebastia@, dcoppa@, and others. > OK kettenis@ dcoppa@, and guenther@ on an earlier version. > Riding guenther@'s libc/librthread major bump. (schwarze@) locale.h ~ locale.h > New POSIX xlocale implementation written from scratch. > Complete in the sense that all POSIX *locale(3) and *_l(3) functions > are included, but in OpenBSD, we of course only really care about > LC_CTYPE and we only support ASCII and UTF-8. > With important help from kettenis@, guenther@, and jca@. > Repeated testing in ports bulk builds by naddy@. > Additional testing by jca@, sebastia@, dcoppa@, and others. > OK kettenis@ dcoppa@, and guenther@ on an earlier version. > Riding guenther@'s libc/librthread major bump. (schwarze@) stdlib.h ~ stdlib.h > New POSIX xlocale implementation written from scratch. > Complete in the sense that all POSIX *locale(3) and *_l(3) functions > are included, but in OpenBSD, we of course only really care about > LC_CTYPE and we only support ASCII and UTF-8. > With important help from kettenis@, guenther@, and jca@. > Repeated testing in ports bulk builds by naddy@. > Additional testing by jca@, sebastia@, dcoppa@, and others. > OK kettenis@ dcoppa@, and guenther@ on an earlier version. > Riding guenther@'s libc/librthread major bump. (schwarze@) string.h ~ string.h > New POSIX xlocale implementation written from scratch. > Complete in the sense that all POSIX *locale(3) and *_l(3) functions > are included, but in OpenBSD, we of course only really care about > LC_CTYPE and we only support ASCII and UTF-8. > With important help from kettenis@, guenther@, and jca@. > Repeated testing in ports bulk builds by naddy@. > Additional testing by jca@, sebastia@, dcoppa@, and others. > OK kettenis@ dcoppa@, and guenther@ on an earlier version. > Riding guenther@'s libc/librthread major bump. (schwarze@) strings.h ~ strings.h > New POSIX xlocale implementation written from scratch. > Complete in the sense that all POSIX *locale(3) and *_l(3) functions > are included, but in OpenBSD, we of course only really care about > LC_CTYPE and we only support ASCII and UTF-8. > With important help from kettenis@, guenther@, and jca@. > Repeated testing in ports bulk builds by naddy@. > Additional testing by jca@, sebastia@, dcoppa@, and others. > OK kettenis@ dcoppa@, and guenther@ on an earlier version. > Riding guenther@'s libc/librthread major bump. (schwarze@) ~ strings.h > str[n]casecmp_l(3) appeared in POSIX 2008, not in XPG 4; > phessler@ reported that this oversight broke an armv7 bulk build; > OK guenther@ (schwarze@) time.h ~ time.h > New POSIX xlocale implementation written from scratch. > Complete in the sense that all POSIX *locale(3) and *_l(3) functions > are included, but in OpenBSD, we of course only really care about > LC_CTYPE and we only support ASCII and UTF-8. > With important help from kettenis@, guenther@, and jca@. > Repeated testing in ports bulk builds by naddy@. > Additional testing by jca@, sebastia@, dcoppa@, and others. > OK kettenis@ dcoppa@, and guenther@ on an earlier version. > Riding guenther@'s libc/librthread major bump. (schwarze@) wchar.h ~ wchar.h > New POSIX xlocale implementation written from scratch. > Complete in the sense that all POSIX *locale(3) and *_l(3) functions > are included, but in OpenBSD, we of course only really care about > LC_CTYPE and we only support ASCII and UTF-8. > With important help from kettenis@, guenther@, and jca@. > Repeated testing in ports bulk builds by naddy@. > Additional testing by jca@, sebastia@, dcoppa@, and others. > OK kettenis@ dcoppa@, and guenther@ on an earlier version. > Riding guenther@'s libc/librthread major bump. (schwarze@) wctype.h ~ wctype.h > New POSIX xlocale implementation written from scratch. > Complete in the sense that all POSIX *locale(3) and *_l(3) functions > are included, but in OpenBSD, we of course only really care about > LC_CTYPE and we only support ASCII and UTF-8. > With important help from kettenis@, guenther@, and jca@. > Repeated testing in ports bulk builds by naddy@. > Additional testing by jca@, sebastia@, dcoppa@, and others. > OK kettenis@ dcoppa@, and guenther@ on an earlier version. > Riding guenther@'s libc/librthread major bump. (schwarze@) == lib =============================================================== 05/12 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/lib libc ~ shlib_version ~ Symbols.list ~ hidden/pthread.h ~ include/cancel.h ~ include/thread_private.h ~ sys/canceled.c ~ thread/Makefile.inc ~ thread/callbacks.c ~ thread/rthread.c ~ thread/rthread.h ~ thread/rthread_cb.h ~ thread/rthread_cond.c ~ thread/rthread_condattr.c ~ thread/rthread_debug.c ~ thread/rthread_libc.c ~ thread/rthread_sync.c ~ thread/rthread_tls.c ~ thread/synch.h + hidden/machine/spinlock.h > Move mutex, condvar, and thread-specific data routes, pthread_once, and > pthread_exit from libpthread to libc, along with low-level bits to > support them. Major bump to both libc and libpthread. > Requested by libressl team. Ports testing by naddy@ > ok kettenis@ (guenther@) ~ gen/errlist.c ~ sys/intro.2 > Add additional errno values required by POSIX. > ok jca@ kettenis@ deraadt@ (jsg@) - locale/setrunelocale.c ~ Symbols.list ~ gen/isalnum.3 ~ gen/isalpha.3 ~ gen/isblank.3 ~ gen/iscntrl.3 ~ gen/isdigit.3 ~ gen/isgraph.3 ~ gen/islower.3 ~ gen/isprint.3 ~ gen/ispunct.3 ~ gen/isspace.3 ~ gen/isupper.3 ~ gen/isxdigit.3 ~ gen/tolower.3 ~ gen/toupper.3 ~ hidden/ctype.h ~ hidden/langinfo.h ~ hidden/locale.h ~ hidden/stdlib.h ~ hidden/string.h ~ hidden/time.h ~ hidden/wchar.h ~ hidden/wctype.h ~ locale/Makefile.inc ~ locale/___runetype_mb.c ~ locale/__mb_cur_max.c ~ locale/iswalnum.3 ~ locale/iswctype.3 ~ locale/iswctype.c ~ locale/multibyte_citrus.c ~ locale/nl_langinfo.3 ~ locale/nl_langinfo.c ~ locale/rune.h ~ locale/rune_local.h ~ locale/runetable.c ~ locale/setlocale.c ~ locale/towctrans.3 ~ locale/towlower.3 ~ locale/wcsxfrm.c ~ locale/wctrans.3 ~ locale/wctype.3 ~ string/Makefile.inc ~ string/strcasecmp.3 ~ string/strcoll.3 ~ string/strerror.3 ~ string/strxfrm.3 ~ string/wcscasecmp.3 ~ time/Makefile.inc ~ time/strftime.3 + locale/_CurrentRuneLocale.c + locale/_get_locname.c + locale/duplocale.c + locale/freelocale.c + locale/isctype_l.c + locale/iswctype_l.c + locale/newlocale.3 + locale/newlocale.c + locale/nl_langinfo_l.c + locale/uselocale.3 + locale/uselocale.c + locale/wcscoll.3 + locale/wcscoll_l.c + locale/wcsxfrm.3 + locale/wcsxfrm_l.c + string/strcasecmp_l.c + string/strcoll_l.c + string/strerror_l.c + string/strxfrm_l.c + string/wcscasecmp_l.c + time/strftime_l.c > New POSIX xlocale implementation written from scratch. > Complete in the sense that all POSIX *locale(3) and *_l(3) functions > are included, but in OpenBSD, we of course only really care about > LC_CTYPE and we only support ASCII and UTF-8. > With important help from kettenis@, guenther@, and jca@. > Repeated testing in ports bulk builds by naddy@. > Additional testing by jca@, sebastia@, dcoppa@, and others. > OK kettenis@ dcoppa@, and guenther@ on an earlier version. > Riding guenther@'s libc/librthread major bump. (schwarze@) - gen/elf_hash.c ~ Symbols.list ~ gen/Makefile.inc > Remove unused 32bit version of elf_hash(). > Riding previous libc bump. > ok kettenis@ (mpi@) ~ gen/sysctl.3 > sysctl strings include the termingating NUL character in both in and out > lengths > Prompted by Nan Xiao noticing the redundancy of bzero+termination - thanks. > ok deraadt@ (tom@) ~ gen/getcwd.3 > Fix spelling of 4.3BSD-Net/2. (schwarze@) ~ termios/tcflow.c ~ termios/tcsendbreak.c ~ gen/sysconf.c ~ gen/shm_open.c > shm_open(), sysconf(), tcflow(), and tcsendbreak() are not permitted to be > cancellation points in POSIX, so change them to invoke the non-cancellation > point versions of open(), close(), nanosleep(), and write() > ok deraadt@ millert@ (guenther@) libcompiler_rt ~ Makefile > In compiler-rt, a few assembler implementations for i386 floating point > conversion functions use SSE2 instructions, but these are not guarded by > #ifdef __SSE2__, and there is no implementation using general purpose > registers. For these functions, use the generic C variants instead, > otherwise they will cause SIGILL on older processors. > From FreeBSD; ok deraadt@ mlarkin@ patrick@ (naddy@) libcrypto ~ modes/gcm128.c > Checking sizeof size_t by SIZE_MAX instead of _LP64 > ok bcook@ (inoguchi@) libcxx ~ shlib_version ~ include/support/newlib/xlocale.h > New POSIX xlocale implementation written from scratch. > Complete in the sense that all POSIX *locale(3) and *_l(3) functions > are included, but in OpenBSD, we of course only really care about > LC_CTYPE and we only support ASCII and UTF-8. > With important help from kettenis@, guenther@, and jca@. > Repeated testing in ports bulk builds by naddy@. > Additional testing by jca@, sebastia@, dcoppa@, and others. > OK kettenis@ dcoppa@, and guenther@ on an earlier version. > Riding guenther@'s libc/librthread major bump. (schwarze@) librthread - rthread_cond.c - rthread_condattr.c - rthread_debug.c - rthread_file.c - rthread_libc.c - rthread_mutex.c - rthread_once.c - rthread_sync.c - rthread_tls.c - synch.h - arch/aarch64/_atomic_lock.c - arch/alpha/_atomic_lock.S - arch/amd64/_atomic_lock.c - arch/arm/_atomic_lock.c - arch/hppa/_atomic_lock.c - arch/i386/_atomic_lock.c - arch/m88k/_atomic_lock.c - arch/mips64/_atomic_lock.c - arch/powerpc/_atomic_lock.c - arch/sh/_atomic_lock.c - arch/sparc64/_atomic_lock.c ~ shlib_version ~ Makefile ~ Symbols.map ~ pthread.h ~ rthread.c ~ rthread.h ~ rthread_attr.c ~ rthread_cb.h ~ rthread_fork.c ~ rthread_np.c ~ rthread_sched.c ~ rthread_sem.c ~ rthread_stack.c > Move mutex, condvar, and thread-specific data routes, pthread_once, and > pthread_exit from libpthread to libc, along with low-level bits to > support them. Major bump to both libc and libpthread. > Requested by libressl team. Ports testing by naddy@ > ok kettenis@ (guenther@) == libexec =========================================================== 06/12 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/libexec ld.so ~ m88k/SYS.h ~ m88k/ldasm.S > Fix m88k problem in recent ld.so changes. > - move _dl_cerror function from SYS.h to ldasm.S to avoid duplicate > definition when linking > - [SYS.h] pull some macros from lib/libc/arch/m88k/DEFS.h > - [ldasm.S] add "'never hit' but needed" br again in _dl_cacheflush > (thanks to Miod Vallat) > - [ldasm.S] change ld.hu into ld.h in order to get a correctly signed > and sign-extended value (also thanks to Miod Vallat) > ok deraadt@ (aoyama@) ~ m88k/rtld_machine.c > Remove superfluous mprotect()ing of the PLT that's been obsoleted by > the generic handling of RWX segments. > ok aoyama@ (guenther@) == regress =========================================================== 07/12 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/regress lib ~ libpthread/dlopen/dlopen.c > Try both libpthread and RTLD_DEFAULT when looking up functions > so the test can find everything after the libpthread->libc move (guenther@) sbin ~ route/Makefile + route/rttest30.ok > Test that all MPATH routes attached to an interface are taken down when > the link state change. > Issue reported by gerhard@ (mpi@) ~ route/Makefile > Document new test. (mpi@) ~ route/Makefile + route/rttest31.ok > Test that MPATH routes with different priorities are insterted respecting > their priority order. (mpi@) sys ~ nfs/Makefile ~ ffs/nfs/Makefile > Before mounting NFS, wait until mountd(8) has exported the directory. > Fixes a race in the test setup. > OK mpi@. (bluhm@) ~ net/pf_divert/Makefile ~ net/pf_divert/Packet.pm > Run divert-packet tests also for IPv6. (bluhm@) ~ kern/ptrace/ptrace.c > Fix declarations > ptrace(PT_IO) memory protection faults return EACCES, not EFAULT > (guenther@) + netinet/frag/Makefile + netinet/frag/LICENSE + netinet/frag/pf.conf + netinet/frag/frag.py + netinet/frag/frag_mf0middle.py + netinet/frag/frag_opt.py + netinet/frag/frag_mf0long.py + netinet/frag/frag_mf0short.py + netinet/frag/frag_mf1end.py + netinet/frag/frag_padding.py + netinet/frag/frag_permute.py + netinet/frag/frag_timeout.py + netinet/frag/frag_udpheader.py + netinet/frag/frag_udppayload.py + netinet/frag/frag_zerosecond.py + netinet/frag/frag_zerofirst.py + netinet/frag/frag_refrag.py + netinet/frag/frag_overhead0.py + netinet/frag/frag_overtail.py + netinet/frag/frag_overtail0.py + netinet/frag/frag_overhead.py > New import: > Copy IPv6 fragment test cases to IPv4 and adapt to the different ~ netinet/Makefile ~ netinet/frag/frag.py ~ netinet/frag/frag_mf0long.py ~ netinet/frag/frag_mf0middle.py ~ netinet/frag/frag_mf0short.py ~ netinet/frag/frag_mf1end.py ~ netinet/frag/frag_opt.py ~ netinet/frag/frag_overhead.py ~ netinet/frag/frag_overtail.py ~ netinet/frag/frag_padding.py ~ netinet/frag/frag_permute.py ~ netinet/frag/frag_refrag.py ~ netinet/frag/frag_timeout.py ~ netinet/frag/frag_udpheader.py ~ netinet/frag/frag_udppayload.py ~ netinet/frag/frag_zerofirst.py ~ netinet/frag/frag_zerosecond.py ~ netinet6/frag6/frag6_mf0long.py ~ netinet6/frag6/frag6_mf0short.py ~ netinet6/frag6/frag6_mf1end.py ~ netinet6/frag6/frag6_overhead.py ~ netinet6/frag6/frag6_overhead0.py ~ netinet6/frag6/frag6_overtail.py ~ netinet6/frag6/frag6_permute.py + netinet/frag/frag_overhead1.py + netinet/frag/frag_overtail1.py + netinet6/frag6/frag6_overhead1.py + netinet6/frag6/frag6_overtail0.py + netinet6/frag6/frag6_overtail1.py > Link regress/sys/netinet/frag to build. Get IPv4 and IPv6 fragment > tests into common shape. Add some missing test cases. (bluhm@) ~ netinet6/frag6/Makefile ~ netinet6/frag6/frag6_refrag.py ~ netinet6/frag6/pf.conf > Get IPv4 and IPv6 fragment tests into common shape. Some chunks > were forgotten in previous commit. (bluhm@) == sbin ============================================================== 08/12 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/sbin dhclient ~ dhclient.c > Unbind (a.k.a. delete) interface address and thus routes when a NAK is > received. When a lease expires (a.k.a. no server responds to renewal > attempts), wait for a server to provide a new address via DISCOVER > before unbinding the existing address. > ok mpi@ (krw@) ~ kroute.c > Tweak route_in_rtstatic() to return the position of the > route if found. Cleans up some int vs unsigned int > confusion and makes the function more generally useful > for future changes. (krw@) ~ dhcpd.h ~ kroute.c ~ options.c > Tweak extract_classless_route() to return unsigned int > and accept NULL pointers for unwanted components of > the route. (krw@) pfctl ~ pfctl.c > - split pf_find_or_create_ruleset() to smaller chunks. > tested by Hrvoje > OK mpi@, OK bluhm@ (sashan@) pflogd ~ pflogd.c ~ pflogd.h ~ privsep.c > fork+exec model for pflogd(8); move pcap init to the re-exec'd privsep > parent and use 'legit' fdpassing primitives to send the bpf fd to the > unprivileged child process. > Also reduces the pledge(2) promises in the unpriv child to just > "stdio recvfd" > with help from deraadt, pcap feedback from canacar > ok deraadt@ (brynet@) ~ pflogd.c ~ privsep.c > pflogd(8) currently spams the console on shutdown if syslogd(8) wins the > race to die, so just stop logging pflogd exits. > This logging probably comes from the fact that pflogd was largely based > on syslogd. > Removes the annoying "pflogd[23954]: Exiting" messages pointed out by > deraadt@ > Also cleanup some missed SIGCHLD handling code that is no longer needed. > "LGTM" mikeb@ (brynet@) ~ pflogd.c > Bring back the "Exiting" log message in pflogd(8), but do not set > the LOG_CONS flag in openlog(3). This avoids output to the console > on shutdown. Discussed with benno@ who wants to see when a deamon > stops during normal system operation. > OK brynet@ (bluhm@) ~ privsep.c ~ pflogd.c ~ pflogd.h > Rework pflogd(8)'s fork+exec model; re-exec the unpriv child, not the > privileged parent. > Based on feedback from deraadt@ and bluhm@ (worked on syslogd). > ok deraadt@ (brynet@) route ~ route.8 ~ route.c > backout previous diff: > autodetect AF when setting inet6 default route > From Denis Fondras, thanks! > ok phessler@ bluhm@ > it breaks with ipv6 link-local routes, e.g. > /sbin/route -inet6 default -ifp pppoe0 fe80::%pppoe0 > found by jmc@ > ok florian (benno@) == share ============================================================= 09/12 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/share man ~ man4/rtwn.4 > Document rtwn(4) RTL8192CE support. (stsp@) ~ man4/pci.4 > update entry for rtwn; (jmc@) ~ man7/eqn.7 > fix grammar; while here, improve spacing. > ok schwarze@ (sobrado@) ~ man9/mi_switch.9 > bring mi_switch() manpage up to date > ok mpi@ guenther@ (jmatthew@) ~ man4/rtsx.4 > add support for the Realtek RTS525A > ok stsp (jcs@) ~ man8/afterboot.8 > rework "Errata" to mention syspatch(8); > based on a diff from jesper wallin > ok aja (jmc@) ~ man4/man4.octeon/cnmac.4 > Update hardware list. (visa@) ~ man4/urng.4 > fix capitalization of "Altus Metrum" > from Raf Czlonka (jasper@) == sys =============================================================== 10/12 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/sys arch/alpha/alpha ~ db_trace.c ~ promcons.c > If you use sys/param.h, you don't need sys/types.h (deraadt@) arch/alpha/include ~ spinlock.h > Move mutex, condvar, and thread-specific data routes, pthread_once, and > pthread_exit from libpthread to libc, along with low-level bits to > support them. Major bump to both libc and libpthread. > Requested by libressl team. Ports testing by naddy@ > ok kettenis@ (guenther@) arch/alpha/isa ~ isa_machdep.c > If you use sys/param.h, you don't need sys/types.h (deraadt@) arch/alpha/pci ~ pci_1000.c ~ pci_1000a.c ~ pci_2100_a50.c ~ pci_550.c ~ pci_alphabook1.c ~ pci_axppci_33.c ~ pci_eb164.c ~ pci_eb64plus.c ~ pci_kn20aa.c ~ pci_kn300.c ~ pci_machdep.c ~ pci_up1000.c > If you use sys/param.h, you don't need sys/types.h (deraadt@) arch/alpha/stand/netboot ~ if_prom.c > If you use sys/param.h, you don't need sys/types.h (deraadt@) arch/amd64/amd64 ~ aesni.c ~ amd64errata.c ~ k1x-pstate.c ~ pctr.c ~ powernow-k8.c ~ vmm.c > If you use sys/param.h, you don't need sys/types.h (deraadt@) arch/amd64/include ~ spinlock.h > Move mutex, condvar, and thread-specific data routes, pthread_once, and > pthread_exit from libpthread to libc, along with low-level bits to > support them. Major bump to both libc and libpthread. > Requested by libressl team. Ports testing by naddy@ > ok kettenis@ (guenther@) ~ param.h > bump UPAGES to 6. > deraadt@s stack base randomisation may take away up to a whole page > in space, and it is possible that i will take another whole page > away in the future as a guard page. this means we continue to provide > roughtly the same amount of stack space that has previously been > in effect. > ok deraadt@ (dlg@) arch/amd64/pci ~ pci_machdep.c > If you use sys/param.h, you don't need sys/types.h (deraadt@) arch/amd64/stand/boot ~ conf.c > If you use sys/param.h, you don't need sys/types.h (deraadt@) arch/amd64/stand/cdboot ~ conf.c > If you use sys/param.h, you don't need sys/types.h (deraadt@) arch/amd64/stand/efiboot ~ conf.c > If you use sys/param.h, you don't need sys/types.h (deraadt@) arch/amd64/stand/libsa ~ gateA20.c > If you use sys/param.h, you don't need sys/types.h (deraadt@) arch/amd64/stand/pxeboot ~ conf.c > If you use sys/param.h, you don't need sys/types.h (deraadt@) arch/arm/arm ~ cpufunc.c ~ fault.c ~ pmap7.c > If you use sys/param.h, you don't need sys/types.h (deraadt@) arch/arm/include ~ spinlock.h > Move mutex, condvar, and thread-specific data routes, pthread_once, and > pthread_exit from libpthread to libc, along with low-level bits to > support them. Major bump to both libc and libpthread. > Requested by libressl team. Ports testing by naddy@ > ok kettenis@ (guenther@) arch/arm64/arm64 ~ ast.c ~ intr.c ~ machdep.c > If you use sys/param.h, you don't need sys/types.h (deraadt@) arch/arm64/include ~ spinlock.h > Move mutex, condvar, and thread-specific data routes, pthread_once, and > pthread_exit from libpthread to libc, along with low-level bits to > support them. Major bump to both libc and libpthread. > Requested by libressl team. Ports testing by naddy@ > ok kettenis@ (guenther@) arch/arm64/stand/efiboot ~ conf.c > If you use sys/param.h, you don't need sys/types.h (deraadt@) arch/armv7/armv7 ~ armv7.c ~ intr.c ~ platform.c > If you use sys/param.h, you don't need sys/types.h (deraadt@) arch/armv7/omap ~ am335x.c ~ dmtimer.c ~ edma.c ~ gptimer.c ~ omap3.c ~ omap_machdep.c ~ ompinmux.c ~ prcm.c ~ sitara_cm.c > If you use sys/param.h, you don't need sys/types.h (deraadt@) arch/armv7/stand/efiboot ~ conf.c > If you use sys/param.h, you don't need sys/types.h (deraadt@) arch/armv7/sunxi ~ sxitimer.c > If you use sys/param.h, you don't need sys/types.h (deraadt@) arch/hppa/hppa ~ ipi.c ~ mem.c > If you use sys/param.h, you don't need sys/types.h (deraadt@) arch/hppa/include ~ spinlock.h > Move mutex, condvar, and thread-specific data routes, pthread_once, and > pthread_exit from libpthread to libc, along with low-level bits to > support them. Major bump to both libc and libpthread. > Requested by libressl team. Ports testing by naddy@ > ok kettenis@ (guenther@) arch/hppa/stand/boot ~ conf.c > If you use sys/param.h, you don't need sys/types.h (deraadt@) arch/i386/eisa ~ eisa_machdep.c > If you use sys/param.h, you don't need sys/types.h (deraadt@) arch/i386/i386 ~ machdep.c > fix boot on older 486DX CPUs that lack CPUID. > ok tom, kettenis (mlarkin@) ~ amd64errata.c ~ k1x-pstate.c ~ pctr.c ~ powernow-k7.c ~ powernow-k8.c ~ vmm.c > If you use sys/param.h, you don't need sys/types.h (deraadt@) arch/i386/include ~ spinlock.h > Move mutex, condvar, and thread-specific data routes, pthread_once, and > pthread_exit from libpthread to libc, along with low-level bits to > support them. Major bump to both libc and libpthread. > Requested by libressl team. Ports testing by naddy@ > ok kettenis@ (guenther@) arch/i386/isa ~ clock.c > If you use sys/param.h, you don't need sys/types.h (deraadt@) arch/i386/pci ~ glxsb.c ~ pci_machdep.c > If you use sys/param.h, you don't need sys/types.h (deraadt@) arch/i386/stand/boot ~ conf.c > If you use sys/param.h, you don't need sys/types.h (deraadt@) arch/i386/stand/cdboot ~ conf.c > If you use sys/param.h, you don't need sys/types.h (deraadt@) arch/i386/stand/libsa ~ gateA20.c > If you use sys/param.h, you don't need sys/types.h (deraadt@) arch/i386/stand/pxeboot ~ conf.c > If you use sys/param.h, you don't need sys/types.h (deraadt@) arch/landisk/dev ~ power.c > If you use sys/param.h, you don't need sys/types.h (deraadt@) arch/landisk/landisk ~ shpcic_machdep.c > If you use sys/param.h, you don't need sys/types.h (deraadt@) arch/landisk/stand/boot ~ conf.c > If you use sys/param.h, you don't need sys/types.h (deraadt@) arch/luna88k/cbus ~ i82365_cbus.c > If you use sys/param.h, you don't need sys/types.h (deraadt@) arch/luna88k/dev ~ xp.c > If you use sys/param.h, you don't need sys/types.h (deraadt@) arch/m88k/include ~ spinlock.h > Move mutex, condvar, and thread-specific data routes, pthread_once, and > pthread_exit from libpthread to libc, along with low-level bits to > support them. Major bump to both libc and libpthread. > Requested by libressl team. Ports testing by naddy@ > ok kettenis@ (guenther@) arch/m88k/m88k ~ fpu.c ~ m88100_fp.c ~ m88110_fp.c ~ trap.c > If you use sys/param.h, you don't need sys/types.h (deraadt@) arch/macppc/dev ~ esp.c ~ macgpio.c ~ pgs.c ~ sysbutton.c ~ xlights.c > If you use sys/param.h, you don't need sys/types.h (deraadt@) arch/macppc/macppc ~ mem.c > If you use sys/param.h, you don't need sys/types.h (deraadt@) arch/macppc/pci ~ macobio.c ~ pchb.c > If you use sys/param.h, you don't need sys/types.h (deraadt@) arch/mips64/include ~ spinlock.h > Move mutex, condvar, and thread-specific data routes, pthread_once, and > pthread_exit from libpthread to libc, along with low-level bits to > support them. Major bump to both libc and libpthread. > Requested by libressl team. Ports testing by naddy@ > ok kettenis@ (guenther@) arch/octeon/dev ~ octmmc.c ~ octmmcreg.h > Let octmmc(4) attach on revision 7890 controller. (visa@) ~ octeon_iobus.c > Allow two GMX instances on CN70xx/CN71xx to enable all ports > on the E300 boards. (visa@) ~ cn30xxfpa.c ~ cn30xxgmx.c ~ cn30xxpow.c > If you use sys/param.h, you don't need sys/types.h (deraadt@) arch/powerpc/include ~ spinlock.h > Move mutex, condvar, and thread-specific data routes, pthread_once, and > pthread_exit from libpthread to libc, along with low-level bits to > support them. Major bump to both libc and libpthread. > Requested by libressl team. Ports testing by naddy@ > ok kettenis@ (guenther@) arch/sgi/dev ~ impact.c > If you use sys/param.h, you don't need sys/types.h (deraadt@) arch/sgi/gio ~ impact_gio.c > If you use sys/param.h, you don't need sys/types.h (deraadt@) arch/sgi/stand/sgivol ~ sgivol.c > If you use sys/param.h, you don't need sys/types.h (deraadt@) arch/sgi/xbow ~ impact_xbow.c ~ odyssey.c > If you use sys/param.h, you don't need sys/types.h (deraadt@) arch/sh/include ~ spinlock.h > Move mutex, condvar, and thread-specific data routes, pthread_once, and > pthread_exit from libpthread to libc, along with low-level bits to > support them. Major bump to both libc and libpthread. > Requested by libressl team. Ports testing by naddy@ > ok kettenis@ (guenther@) arch/socppc/socppc ~ mem.c > If you use sys/param.h, you don't need sys/types.h (deraadt@) arch/socppc/stand/boot ~ fdt.c ~ wd.c > If you use sys/param.h, you don't need sys/types.h (deraadt@) arch/sparc64/dev ~ beep.c ~ beeper.c ~ central.c ~ com_ebus.c ~ creator.c ~ fhc.c ~ fhc_central.c ~ fhc_mainbus.c ~ lpt_ebus.c ~ pci_machdep.c ~ pckbc_ebus.c ~ power.c ~ ppm.c ~ sab.c ~ upa.c ~ uperf.c ~ uperf_ebus.c > If you use sys/param.h, you don't need sys/types.h (deraadt@) arch/sparc64/include ~ spinlock.h > Move mutex, condvar, and thread-specific data routes, pthread_once, and > pthread_exit from libpthread to libc, along with low-level bits to > support them. Major bump to both libc and libpthread. > Requested by libressl team. Ports testing by naddy@ > ok kettenis@ (guenther@) ddb ~ db_command.c ~ db_ctf.c ~ db_sym.h > replace the DDB_STRUCT backend for "show struct" with ctf code. > this lets you inspect arbitrary memory in the kernel as a specified > struct. > ok mpi@ jasper@ (dlg@) ~ db_ctf.c > when pretty printing a pointer, display its value instead of address. > this makes it consistent with printing of other values. > ok mpi@ (dlg@) ~ db_ctf.c > use sizeof(variable) instead of sizeof(type) to shorten some code (dlg@) ~ db_ctf.c ~ db_dwarf.c ~ db_elf.c ~ db_usrreq.c > If you use sys/param.h, you don't need sys/types.h (deraadt@) dev ~ softraid.c > In the case that we want to force a new RAID assembly and the metadata > type on the chunks is not the same, we don't actually care because we > are going to clear their metadata anyway, so don't error out. This > allows assembling a new RAID based on a drive that is zeroed and one > that used to be part of a softraid in its previous life. > ok jsing@ (patrick@) ~ diskmap.c ~ ipmi.c > If you use sys/param.h, you don't need sys/types.h (deraadt@) dev/acpi ~ acpibat.c > Don't require exactly 20 elements in the _BIX package if at least 20 > are returned since extras may be OEM-specific values we end up > ignoring anyway, such as on the Lenovo X1C 5th Gen which returns 21 > elements. > ok mpi (jcs@) ~ acpidev.h > fix comment (jcs@) ~ acpitoshiba.c > In toshiba_hotkey() initialize the ret value to failure as it is > not set in all code paths. > found by clang -Wuninitialized; OK deraadt@ (bluhm@) dev/eisa ~ cac_eisa.c ~ dpt_eisa.c ~ uha_eisa.c > If you use sys/param.h, you don't need sys/types.h (deraadt@) dev/fdt ~ if_mvneta.c ~ sxiccmu.c > If you use sys/param.h, you don't need sys/types.h (deraadt@) dev/ic ~ r92creg.h > When I merged rtwn(4) and urtwn(4) register definitions, an extra > line snuck into the rtl8192ce_bb_vals_2t array. Remove it again. > This broke baseband initialization for RTL8192CE which went unnoticed > because at the time nobody had tested rtwn(4) with that chip and the > driver didn't attach to it. > Problem reported by Martin Wanvik who eventually did this test. > ok kevlo@ (stsp@) ~ rtsx.c ~ rtsxreg.h ~ rtsxvar.h > add support for the Realtek RTS525A > ok stsp (jcs@) ~ rtsx.c > Check return of rtsx_write in previous commit as RTSX_WRITE macro > does > Coverity CID 1455437 (jcs@) ~ adv.c ~ advlib.c ~ adw.c ~ adwlib.c ~ aic6250.c ~ aic6360.c ~ bwi.c ~ cy.c ~ i82365.c ~ lsi64854.c ~ malo.c ~ mfi.c ~ ncr5380sbc.c ~ ncr53c9x.c ~ rtw.c ~ tcic2.c ~ uha.c > If you use sys/param.h, you don't need sys/types.h (deraadt@) dev/isa ~ fdc.c ~ fd.c > fdc: defer probing of floppy drives > Defer probing of the drives to a kthread. This avoids several seconds of > delay() during boot. However, drives may now appear only after init has > started. But this does not affect installs from floppy. In that case the > root > ramdisk is linked into the kernel and loaded by the boot loader. > Approach suggested by deraadt@, tested by miod@ (sf@) ~ aic_isa.c ~ i82365_isa.c ~ i82365_isapnp.c ~ i82365_isasubr.c ~ if_eg.c ~ sch311x.c ~ tcic2_isa.c ~ uha_isa.c ~ wdc_isa.c ~ wdc_isapnp.c ~ wds.c > If you use sys/param.h, you don't need sys/types.h (deraadt@) ~ fdc.c > fdc: Print warning if kthread_create fails > Other drivers do this, too. Fixes coverity CID 1455436. (sf@) dev/microcode ~ adw/adwmcode.c ~ tusb3410/build.c > If you use sys/param.h, you don't need sys/types.h (deraadt@) dev/pci ~ drm/drm_gem.c > drm: Release driver tracking before making the object available again > From Chris Wilson > 2a9d7664ffb2c223c488058ee6bee61512db9396 in linux-4.4.y/4.4.85 > fe4600a548f2763dec91b3b27a1245c370ceee2a in mainline linux (jsg@) ~ drm/drm_atomic.c > drm/atomic: If the atomic check fails, return its value first > From Maarten Lankhorst > 3416ee45a8cbeb5bc4b13a754873fbb26a27dccb in linux-4.4.y/4.4.85 > a0ffc51e20e90e0c1c2491de2b4b03f48b6caaba in mainline linux (jsg@) ~ drm/i915/intel_uncore.c > drm/i915: fix compiler warning in drivers/gpu/drm/i915/intel_uncore.c > From Greg Kroah-Hartman > c81c4d453edf242af644f7e71f673fc7632f337c in linux-4.4.y/4.4.86 (jsg@) ~ if_rtwn.c > Attach rtwn(4) to RTL8192CE. Tested by me. > ok kevlo@ (stsp@) ~ if_iwn.c > Make sure iwn(4) firmware gets to see a beacon before we send an auth > request. > Otherwise, the Tx attempt can fail due to the firmware's built-in > regulatory > domain enforcement. > Because this code runs in interrrupt context we use a DELAY(). It's > currently > set to 3 times the beacon interval. This is a bit long as far as DELAYs go, > but since this driver runs state transitions in interrupt context there is > no better way for now. Lower delays weren't reliable enough in my testing. > Makes association work reliably for me on 5GHz. My original plan was to > only > add this hack for 5GHz but Holger Mikolon reported that it also helps on > 2GHz > channels with regulatory restrictions (e.g. 13), so do it always. > ok mpi@ (stsp@) ~ drm/drm_linux.h > Add additional errno values required by POSIX. > ok jca@ kettenis@ deraadt@ (jsg@) ~ rtsx_pci.c > add support for the Realtek RTS525A > ok stsp (jcs@) ~ adv_pci.c ~ adw_pci.c ~ auixp.c ~ azalia.c ~ berkwdt.c ~ emuxki.c ~ esa.c ~ i82365_pci.c ~ if_age.c ~ if_alc.c ~ if_ale.c ~ if_et.c ~ if_jme.c ~ if_nfe.c ~ if_vte.c ~ mbg.c ~ mfii.c ~ pccbb.c ~ pwdog.c ~ wdt.c > If you use sys/param.h, you don't need sys/types.h (deraadt@) dev/pcmcia ~ com_pcmcia.c ~ pcmcia.c ~ pcmcia_cis.c ~ pcmcia_cis_quirks.c > If you use sys/param.h, you don't need sys/types.h (deraadt@) dev/puc ~ lpt_puc.c > If you use sys/param.h, you don't need sys/types.h (deraadt@) dev/pv ~ vioscsi.c ~ virtiovar.h > If you use sys/param.h, you don't need sys/types.h (deraadt@) dev/rasops ~ rasops_bitops.h > In copycols() remove useless code that modifies the uninitialized > sp variable. Its value is not used, it is overwritten later on. > found by clang -Wuninitialized; OK deraadt@ (bluhm@) dev/sbus ~ apio.c ~ asio.c ~ dma_sbus.c ~ esp_sbus.c ~ qec.c ~ uperf_sbus.c > If you use sys/param.h, you don't need sys/types.h (deraadt@) dev/usb ~ xhci.c > Revert previous: "Fix a remaining length miscalculation in xhci(4)." > It made one of my machines get stuck during boot. (stsp@) ~ udl.c > Correct a bounds test. Coverity CID 1452930. > ok mpi@ (jsg@) ~ dwc2/dwc2var.h > EPROTO is now in sys/errno.h (otto@) ~ urng.c > explicitly clear the output pipe > ok mpi@ (jasper@) ~ ulpt.c ~ dwc2/dwc2_core.c ~ dwc2/dwc2_hcd.c ~ dwc2/dwc2_hcdddma.c ~ dwc2/dwc2_hcdintr.c ~ dwc2/dwc2_hcdqueue.c > If you use sys/param.h, you don't need sys/types.h (deraadt@) ~ xhci.c ~ xhcireg.h > Add support for isochronous transfers to xhci(4). > This is just a step forward which allows further progress to happen > in-tree. > The isochronous code path remains disabled for now. Playing audio over > xhci(4) does not work properly yet, and I haven't even tested video input. > Based on a work-in-progress diff by mpi@ from 2015. > ok mpi@ (stsp@) ~ urng.c > - call usbd_deactivate() rather than simply returning in case of errors > - explicitly clear sc_xfer upon detach > - fix spacing in urng_softc definition (jasper@) dev/videomode ~ vesagtf.c > If you use sys/param.h, you don't need sys/types.h (deraadt@) dev/wscons ~ wskbdutil.c > If you use sys/param.h, you don't need sys/types.h (deraadt@) dev/wsfont ~ wsfont.c > If you use sys/param.h, you don't need sys/types.h (deraadt@) isofs/cd9660 ~ cd9660_rrip.c > If you use sys/param.h, you don't need sys/types.h (deraadt@) isofs/udf ~ udf_vfsops.c > If you use sys/param.h, you don't need sys/types.h (deraadt@) kern ~ vfs_getcwd.c > Do not pass an uninitialized size value to free(9) even if the addr > pointer is NULL as it may generate false positive warnings. > requested by markus@ (bluhm@) ~ exec_elf.c > In elf_load_file() to not call free(9) with an uninitialized size > even if the pointer is NULL. This is not a real bug as free(9) > checks the addr pointer before the size value, but the compiler > cannot know that. > found by clang -Wuninitialized; OK deraadt@ (bluhm@) ~ kern_pledge.c ~ kern_srp.c ~ subr_percpu.c ~ subr_poison.c > If you use sys/param.h, you don't need sys/types.h (deraadt@) lib/libkern ~ random.c > If you use sys/param.h, you don't need sys/types.h (deraadt@) lib/libsa ~ netif.c > If you use sys/param.h, you don't need sys/types.h (deraadt@) net ~ rtsock.c > Replace NET_ASSERT_LOCKED() by soassertlocked() in *_usrreq(). > Not all of them need the NET_LOCK(). > ok bluhm@ (mpi@) ~ route.c ~ rtable.c > Restart the iteration when a multipath list is re-ordered to make sure > no entry are missed. > While here do not re-ordered or send messages for route entries that are > already in the expected state. > Make rttest30 pass. > ok gerhard@ (mpi@) ~ rtable.c > Simplify rtable_mpath_insert(). > ok jmatthew@ (mpi@) ~ pf_ruleset.c ~ pfvar.h > - split pf_find_or_create_ruleset() to smaller chunks. > tested by Hrvoje > OK mpi@, OK bluhm@ (sashan@) ~ bfd.c ~ bpf_filter.c ~ bsd-comp.c ~ if_pflow.c ~ if_pppoe.c ~ pfkeyv2.c ~ pfkeyv2_convert.c ~ ppp-deflate.c > If you use sys/param.h, you don't need sys/types.h (deraadt@) net80211 ~ ieee80211_input.c > If a beacon is received in RUN state, reset the management timer. > Some wifi drivers send a probe request if the hardware reports "missed > beacon" > events. If the AP replies with a probe response it is still servicing us > and > there is no need to search for a new AP. However, the management timer was > not > reset if a beacon was received while in RUN state. So the interface > watchdog > always ended up putting the driver into SCAN state after a missed beacon > event, > even if the AP did respond to our probe request. Under some conditions this > bug would cause spurious disconnects. > Problem reported and fix tested by mlarkin@ > (Using the management timer in RUN state is a new convention. Before > support > for missed beacons was added, this timer was only used during the > association > sequence to handle APs which don't respond to our assoc requests and such.) > (stsp@) ~ ieee80211.c > When selecting the next wireless mode during the scan loop, always select > AUTO mode if the driver scans all bands at once. Otherwise the net80211 > layer unnecessarily filters out some of the beacons received by the device. > ok phessler@ mpi@ kevlo@ (stsp@) ~ ieee80211_node.c > When starting a new scan always set the mode to AUTO if the driver scans > all bands at once. Fixes a problem where e.g. 5GHz APs were filtered out > if we were previously associated to an 11g-only AP. > ok mpi@ phessler@ (stsp@) netinet ~ ip_input.c ~ ip_var.h > Serialize access to IP reassembly queue with a mutex. This lets > ip_local(), ip_slowtimo() and ip_drain() run without KERNEL_LOCK() > and NET_LOCK(). > Input and OK mpi@, bluhm@ (visa@) ~ ip_divert.c ~ raw_ip.c ~ tcp_usrreq.c ~ udp_usrreq.c > Replace NET_ASSERT_LOCKED() by soassertlocked() in *_usrreq(). > Not all of them need the NET_LOCK(). > ok bluhm@ (mpi@) ~ ip_divert.c > Replace the call to ifa_ifwithaddr() in divert_output() with a route > lookup to make it MP safe. Only set the mbuf header fields that > are needed. Validate the name input. > OK mpi@ (bluhm@) ~ ip_divert.c > Replace the call to ifa_ifwithaddr() in divert6_output() with a > route lookup to make it MP safe. Only set the mbuf header fields > that are needed. Validate the name input. Also use the same > variables in IPv4 and IPv6 functions and avoid unneccessary > initialization. > OK mpi@ (bluhm@) ~ ip_input.c > Replace a goto found in the ipq foreach loop with a simple break. > This is a common idiom when a list element has been found. > OK visa@ mpi@ (bluhm@) netinet6 ~ ip6_divert.c ~ raw_ip6.c > Replace NET_ASSERT_LOCKED() by soassertlocked() in *_usrreq(). > Not all of them need the NET_LOCK(). > ok bluhm@ (mpi@) ~ ip6_divert.c > Replace the call to ifa_ifwithaddr() in divert6_output() with a > route lookup to make it MP safe. Only set the mbuf header fields > that are needed. Validate the name input. Also use the same > variables in IPv4 and IPv6 functions and avoid unneccessary > initialization. > OK mpi@ (bluhm@) ~ ip6_id.c > If you use sys/param.h, you don't need sys/types.h (deraadt@) nfs ~ nfs_socket.c > Preallocate option mbufs in order to reduce solock()/sounlock() dances. > Finally protect the last `so_rcv' and `so_snd' accesses with the socket > lock. > ok visa@, bluhm@ (mpi@) ~ nfs_socket.c > nfs_connect() returns EINVAL at the beginning if nm_sotype is > invalid. But the compiler cannot know whether it has changed in > the meantime, so in the else case a bunch of variables would not > be initialized. Add a panic() there to change the compiler's > assumptions, the code should not be reached anyway. > found by clang -Wuninitialized; OK deraadt@ (bluhm@) scsi ~ cd.c ~ scsi_base.c ~ scsi_ioctl.c ~ scsiconf.c ~ sd.c ~ st.c ~ uk.c > If you use sys/param.h, you don't need sys/types.h (deraadt@) sys ~ errno.h > Add additional errno values required by POSIX. > ok jca@ kettenis@ deraadt@ (jsg@) ~ exec_elf.h > Remove unused 32bit version of elf_hash(). > Riding previous libc bump. > ok kettenis@ (mpi@) tmpfs ~ tmpfs_vfsops.c > If you use sys/param.h, you don't need sys/types.h (deraadt@) ufs/ext2fs ~ ext2fs_extents.c > If you use sys/param.h, you don't need sys/types.h (deraadt@) == usr.bin =========================================================== 11/12 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/usr.bin ctfconv ~ dw.c > Fix typo in comment; ok mpi@ (anton@) ftp ~ util.c > Avoid a possible leak in progressmeter > From Scott Cheloha, ok bluhm@ (jca@) lock ~ lock.c > After recent changes, we need to clear hash, not s1. > From Scott Cheloha > ok deraadt (tb@) mandoc ~ html.c > fix a NULL pointer access on deroff() failure; > could be triggered with '.SS ""'; > reported by Michael <Stapelberg at debian> (schwarze@) ~ mandoc.1 > typo: convertion -> convention > OK schwarze@ (dcoppa@) mg ~ line.c > When inserting a newline, only increment dotline for windows that are > displaying the same buffer as the current one. > From Martin Wanvik (martin.kr.wanvik AT gmail), thanks! > OK bcallah (florian@) pctr ~ pctr.c > sysctl strings include the termingating NUL character in both in and out > lengths > Prompted by Nan Xiao noticing the redundancy of bzero+termination - thanks. > ok deraadt@ (tom@) sdiff ~ Makefile > Add -Wno-sign-compare, the single warning produced is a false positive; > ok deraadt@ (otto@) ssh ~ readconf.c ~ readconf.h ~ ssh_config.5 ~ sshconnect.c > Expand ssh_config's StrictModes option with two new settings: > StrictModes=accept-new will automatically accept hitherto-unseen keys > but will refuse connections for changed or invalid hostkeys. > StrictModes=off is the same as StrictModes=no > Motivation: > StrictModes=no combines two behaviours for host key processing: > automatically learning new hostkeys and continuing to connect to hosts > with invalid/changed hostkeys. The latter behaviour is quite dangerous > since it removes most of the protections the SSH protocol is supposed to > provide. > Quite a few users want to automatically learn hostkeys however, so > this makes that feature available with less danger. > At some point in the future, StrictModes=no will change to be a synonym > for accept-new, with its current behaviour remaining available via > StrictModes=off. > bz#2400, suggested by Michael Samuel; ok markus (djm@) ~ ssh_config.5 > tweak previous; (jmc@) ~ auth2-pubkey.c > typo in comment (djm@) tftp ~ tftp.c > Use monotonic clock for statistics display. From Scott Cheloha. > ok dlg (tb@) tmux ~ tmux.1 ~ window-copy.c > Move to current mouse position not last when clcking in copy mode; fixes > GitHub issue 1055. Also a man page fix from jmc. (nicm@) ~ server-client.c > Do not attempt to use TIOCSWINSZ on a -1 file descriptor (possible if > the pane has already died). (nicm@) ~ cmd-set-option.c > Do not fail if unset an option that is already unset, reported by Thomas > Sattler. (nicm@) ~ key-bindings.c > When removing a key table clear it out of clients, fixes issue with > unbind -a reported by Thomas Sattler. (nicm@) ~ mode-tree.c > Fix a few errors in how the selected line is chosen after resize, > reported by Felix Rosencrantz in GitHub issue 1059. (nicm@) ~ input.c > Apply timeout to CAN and RS which also wait for ST. (nicm@) ~ grid.c > Previously, extended cell data was never reduced in size even when the > cell was overwritten. With a large history this can be a substantial > amount of memory. To reduce this, compact each extended cell list to > only cells in use as it is scrolled off the visible screen into the > history. From Dan Aloni in GitHub issue 1062. (nicm@) == usr.sbin ========================================================== 12/12 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/usr.sbin ntpd ~ ntpd.8 > ntpd has been on by default for over two years now, so rework > the text about enabling it; > ok deraadt (jmc@) pwd_mkdb ~ pwd_mkdb.8 > Assorted simple improvements: > - Considerably simplify the first paragraph; suggested by Raf Czlonka. > - No need to say twice that "file" must be in the correct format. > - Clarify which file is checked by -c. > - Clarify that -p creates passwd(5) in addition, not instead. > - Simplify reference to the master.passwd(5) format. > - Turn one sentence into a standard EXIT STATUS section. > - New HISTORY and AUTHORS sections, based on the CSRG archive CDs. > (schwarze@) quot ~ quot.8 > Add HISTORY and AUTHORS, using hints from miod@ and my own research. > Sources: http://minnie.tuhs.org/ and the NetBSD CVS. (schwarze@) ~ quot.c > unifdef some ancient COMPAT code that's not used. 500 byte blocks! > noticed by miod (tedu@) smtpd ~ smtp_session.c > use plain values for transaction error codes rather than a bitmask. > rename ill-named "msgflags" to "error" while there. > ok gilles@ (eric@) ~ parse.y ~ smtp_session.c ~ smtpd-api.h ~ smtpd.c ~ smtpd.h > remove more filter-related cruft > ok gilles@ (eric@) tcpdump ~ privsep.c ~ privsep.h ~ tcpdump.c > fork+exec model for tcpdump(8); re-exec the privileged child after fork > While tcpdump isn't a daemon in the traditional sense, it's not uncommon > for people to have long running sessions. At least on OpenBSD, this is > even safe thanks to the existing privsep design by otto@, canacar@ and > pledge(2) work done by deraadt. > ok deraadt@ (brynet@) ~ privsep.c > Whoops, remove extra include. (brynet@) vmctl ~ vmctl.8 > Detail how VM guests acquire a nameserver for local interfaces. > OK mlarkin@, krw@. Thanks! (edd@) ~ vmctl.c > vmctl: style cleanup - spaces vs tabs. > Noticed by Carlos Cardenas (mlarkin@) ~ vmctl.c > better VM termination handling. > diff provided by Carlos Cardenas, thanks (mlarkin@) vmd ~ virtio.c > vmd: handle queue index wraparound in viornd. Without this, openbsd guests > used to crash after roughly 68 hours uptime. > ok deraadt (mlarkin@) ~ config.c ~ control.c ~ proc.c ~ vmd.c ~ vmm.c > vmd: add more explanatory log_debug messages > From Carlos Cardenas, many thanks! (mlarkin@) ~ vmm.c > fix a wrong comment. > Noticed by Carlos Cardenas (mlarkin@) ~ vmd.h ~ vmm.c > better VM termination handling. > diff provided by Carlos Cardenas, thanks (mlarkin@) =============================================================================== _______________________________________________ owc mailing list [email protected] http://www.squish.net/mailman/listinfo/owc
