OpenBSD src changes summary for 2015-12-20 to 2015-12-27 inclusive ==================================================================
bin/cp bin/csh bin/ksh bin/mv bin/pax distrib/armv7 distrib/miniroot distrib/sets distrib/sgi etc/MAKEDEV.common etc/etc.amd64/MAKEDEV etc/etc.amd64/MAKEDEV.md etc/etc.i386/MAKEDEV etc/etc.sgi/Makefile.inc etc/examples/eigrpd.conf etc/mail/Makefile etc/mail/smtpd.conf etc/rc.d/mountd etc/rc.d/rc.subr etc/rc.d/smtpd games/adventure games/arithmetic games/atc games/backgammon games/banner games/battlestar games/boggle games/bs games/canfield games/cribbage games/factor games/fish games/fortune games/gomoku games/grdc games/hack games/hangman games/hunt games/mille games/number games/phantasia games/pig games/pom games/ppt games/primes games/quiz games/random games/robots games/sail games/tetris games/worm games/wump gnu gnu/usr.bin/binutils-2.17 gnu/usr.bin/cxxfilt include/sndio.h lib/libc lib/libevent lib/libfuse lib/libkeynote lib/libpcap lib/libsndio lib/libsqlite3 lib/libssl libexec/getty libexec/ld.so libexec/login_token regress/sys regress/usr.bin sbin/disklabel sbin/dump sbin/fdisk sbin/init sbin/mountd share/man sys/arch/amd64/amd64 sys/arch/amd64/conf sys/arch/amd64/stand/efiboot sys/arch/amd64/stand/libsa sys/arch/i386/i386 sys/arch/mips64/include sys/arch/mips64/mips64 sys/arch/sgi/conf sys/arch/sgi/include sys/arch/sgi/localbus sys/arch/sgi/sgi sys/arch/sgi/xbow sys/arch/zaurus/include sys/ddb sys/dev/acpi sys/dev/isa sys/dev/pci sys/dev/pv sys/dev/usb sys/kern sys/net sys/netinet6 sys/nfs sys/sys usr.bin/aucat usr.bin/basename usr.bin/cvs usr.bin/file usr.bin/finger usr.bin/grep usr.bin/kdump usr.bin/less usr.bin/login usr.bin/make usr.bin/mandoc usr.bin/mg usr.bin/openssl usr.bin/rdist usr.bin/rdistd usr.bin/sndiod usr.bin/sqlite3 usr.bin/ssh usr.bin/telnet usr.bin/tmux usr.bin/uname usr.bin/usbhidctl usr.bin/vmstat usr.sbin/adduser usr.sbin/bgpctl usr.sbin/bgpd usr.sbin/dhcpd usr.sbin/eigrpd usr.sbin/ftp-proxy usr.sbin/gpioctl usr.sbin/hostapd usr.sbin/inetd usr.sbin/installboot usr.sbin/ldapd usr.sbin/memconfig usr.sbin/mtree usr.sbin/pppd usr.sbin/rdate usr.sbin/relayd usr.sbin/smtpd usr.sbin/snmpd usr.sbin/sysmerge usr.sbin/tcpdump usr.sbin/tokeninit usr.sbin/usbdevs == bin =============================================================== 01/14 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/bin cp ~ extern.h ~ utils.c > Set the times, mode, and flags on symlinks when doing cp -p (or mv across > filesystems) > problem noted by Carsten Kunze (carsten.kunze (at) arcor.de) (guenther@) csh ~ glob.c > Remove a needless allocation cast. No binary change. (mmcc@) ~ alloc.c ~ csh.c ~ csh.h ~ dir.c ~ dol.c ~ error.c ~ exec.c ~ exp.c ~ extern.h ~ file.c ~ func.c ~ glob.c ~ hist.c ~ lex.c ~ misc.c ~ parse.c ~ proc.c ~ sem.c ~ set.c > Replace handrolled xfree() function by directly using free(3) > OK mmcc@ (mestre@) ksh ~ misc.c > Replace an ugly bit of pointer arithmetic with a conditional expression. > "seems sound" deraadt@ (mmcc@) ~ vi.c > Assign pointer variables to NULL rather than 0. No functional change. > (mmcc@) ~ ksh.1 > remove special characters; diff from michael reed (jmc@) ~ main.c > unconditionally duplicate the argv array on initialization, to make it > resilient against being altered by a subsequent shift operation > tweak and ok semarie@ (halex@) ~ main.c > Tweak comment: we now always make a copy of argv. > ok halex@ (jca@) mv ~ cp.c > Set the times, mode, and flags on symlinks when doing cp -p (or mv across > filesystems) > problem noted by Carsten Kunze (carsten.kunze (at) arcor.de) (guenther@) ~ cp.c > remove unused variable > ok tedu@ (chl@) pax ~ cache.h > More adress -> address (mmcc@) == distrib =========================================================== 02/14 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/distrib armv7 ~ ramdisk/install.md > Simplify things by using grep -q. > OK krw@, halex@, jsg@ (rpe@) miniroot ~ install.sub > Ensure the installer asks for IPv4 default route in case an interface has > been configured via dhcp and then again manually. > While there, remove 'dhcp' as valid answer for the default route qeustion. > It was a no-op and was handled the same way as 'none'. > tested and OK krw@ > makes sense halex@ (rpe@) ~ install.sub > Ensure the installer asks for the dns domain and nameserver in case an > interface has been configured via dhcp and then again manually. > tested and OK krw@ > makes sense halex@ (rpe@) ~ install.sub > Change comment style for headings so they get removed by list2sh.awk and > don't end up in bsd.rd. > OK krw@ (rpe@) ~ upgrade.sh > - spacing > - use brace expansion in for-loop > OK krw@ (rpe@) ~ install.sh > - quote user supplied hostname > - in this context ((NIFS)) does the same as ((NIFS != 0)) > - use if-block for root ssh key question > - spacing > OK krw@ (rpe@) ~ install.sh ~ install.sub > Move the code, that lists uninitialized disk into a separate function. > This makes the disk initialization question code easier to understand. > OK krw@ (rpe@) ~ install.sub ~ install.sh > Refactor and move disk initialization code from install.sh to a separate > function configure_disk() in install.sub. > based on work from and OK halex@ (rpe@) sets ~ lists/base/md.alpha ~ lists/base/md.amd64 ~ lists/base/md.armish ~ lists/base/md.armv7 ~ lists/base/md.hppa ~ lists/base/md.hppa64 ~ lists/base/md.i386 ~ lists/base/md.landisk ~ lists/base/md.loongson ~ lists/base/md.luna88k ~ lists/base/md.macppc ~ lists/base/md.octeon ~ lists/base/md.sgi ~ lists/base/md.socppc ~ lists/base/md.sparc ~ lists/base/md.sparc64 ~ lists/base/md.zaurus > sync (deraadt@) ~ lists/comp/mi > sync (deraadt@) ~ lists/etc/mi ~ lists/man/mi > sync (deraadt@) ~ lists/base/md.alpha ~ lists/base/md.amd64 ~ lists/base/md.armish ~ lists/base/md.armv7 ~ lists/base/md.hppa ~ lists/base/md.hppa64 ~ lists/base/md.i386 ~ lists/base/md.landisk ~ lists/base/md.loongson ~ lists/base/md.luna88k ~ lists/base/md.macppc ~ lists/base/md.octeon ~ lists/base/md.sgi ~ lists/base/md.socppc ~ lists/base/md.sparc ~ lists/base/md.sparc64 ~ lists/base/md.zaurus > sync (deraadt@) sgi ~ iso/Makefile ~ ramdisk/install.md > welcome GENERIC.IP27 MP (dual cpu on O350 now) > ok visa (deraadt@) ~ ramdisk/install.md > Remove stray '}'. (rpe@) == etc =============================================================== 03/14 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/etc MAKEDEV.common ~ MAKEDEV.common > add virtio-console driver > This patch adds a new driver for use of virtio-console devices as > ttys. It's still in an early state and not compiled by default, yet. > Currently it is only wired into amd64. i386 is still missing. > Discussed with uebayasi@, deraadt@ (sf@) etc.amd64/MAKEDEV ~ etc.amd64/MAKEDEV > regen MAKEDEV (sf@) ~ etc.amd64/MAKEDEV > regen MAKEDEV (sf@) etc.amd64/MAKEDEV.md ~ etc.amd64/MAKEDEV.md > add virtio-console driver > This patch adds a new driver for use of virtio-console devices as > ttys. It's still in an early state and not compiled by default, yet. > Currently it is only wired into amd64. i386 is still missing. > Discussed with uebayasi@, deraadt@ (sf@) etc.i386/MAKEDEV ~ etc.i386/MAKEDEV > regen MAKEDEV (sf@) etc.sgi/Makefile.inc ~ etc.sgi/Makefile.inc > welcome GENERIC.IP27 MP (dual cpu on O350 now) > ok visa (deraadt@) examples/eigrpd.conf ~ examples/eigrpd.conf > add rcsid (gsoares@) mail/Makefile ~ mail/Makefile > Now that default smtpd.conf aliases table has been switched to > 'file' backend, we no longer have to generate the .db files. > Ok gilles@ (sunil@) mail/smtpd.conf ~ mail/smtpd.conf > Switch to 'file' backend for aliases table for default configuration. > Ok gilles@ jung@ millert@ (sunil@) rc.d/mountd ~ rc.d/mountd > mountd(8) changes require a pexp. > ok deraadt@ (krw@) ~ rc.d/mountd > Remove blank line for consistency with other rc.d scripts; no change in > behavior. (ajacoutot@) rc.d/rc.subr ~ rc.d/rc.subr > Comment the wait for start until I fix the x11 managers rc.d script. > (ajacoutot@) ~ rc.d/rc.subr > Revert, rc.d scripts have been fixed. (ajacoutot@) rc.d/smtpd ~ rc.d/smtpd > Remove blank line for consistency with other rc.d scripts; no change in > behavior. (ajacoutot@) == games ============================================================= 04/14 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/games adventure ~ extern.h > Set as __dead a few more functions, that don't return, on games/ > With precious tip, help and also OK from tb@ (mestre@) arithmetic ~ arithmetic.c > Declare usage() functions as __dead void, if they don't return, on games > section. > Found another one in arithmetic(6) which also didn't return, and removed a > function from number(6) which is not used anymore. > OK tb@ (mestre@) atc ~ extern.h > Set as __dead a few more functions, that don't return, on games/ > With precious tip, help and also OK from tb@ (mestre@) backgammon ~ common_source/back.h > Set as __dead a few more functions, that don't return, on games/ > With precious tip, help and also OK from tb@ (mestre@) banner ~ banner.c > - use strtonum(3) instead of atoi(3) > - while here switch main() to return instead of exit(3), makes stack > protector happy. > OK tb@ (gsoares@) battlestar ~ extern.h > Set as __dead a few more functions, that don't return, on games/ > With precious tip, help and also OK from tb@ (mestre@) boggle ~ boggle/extern.h > Declare usage() functions as __dead void, if they don't return, on games > section. > Found another one in arithmetic(6) which also didn't return, and removed a > function from number(6) which is not used anymore. > OK tb@ (mestre@) ~ boggle/mach.c > remove unused variables > ok mmcc@ (chl@) bs ~ bs.c > Function prototypes and a little bit of KNF. > Committing on behalf of mestre@ who has more to come. (tb@) ~ bs.c > Apply some more KNF to the declaration of functions, but keep style to the > rest of code for now. > OK tb@ (mestre@) ~ bs.c > Set as __dead a few more functions, that don't return, on games/ > With precious tip, help and also OK from tb@ (mestre@) canfield ~ canfield/canfield.c > Set as __dead a few more functions, that don't return, on games/ > With precious tip, help and also OK from tb@ (mestre@) cribbage ~ cribbage.h > Set as __dead a few more functions, that don't return, on games/ > With precious tip, help and also OK from tb@ (mestre@) factor ~ factor.c > Declare usage() functions as __dead void, if they don't return, on games > section. > Found another one in arithmetic(6) which also didn't return, and removed a > function from number(6) which is not used anymore. > OK tb@ (mestre@) fish ~ fish.c > Declare usage() functions as __dead void, if they don't return, on games > section. > Found another one in arithmetic(6) which also didn't return, and removed a > function from number(6) which is not used anymore. > OK tb@ (mestre@) fortune ~ fortune/fortune.c ~ strfile/strfile.c > Declare usage() functions as __dead void, if they don't return, on games > section. > Found another one in arithmetic(6) which also didn't return, and removed a > function from number(6) which is not used anymore. > OK tb@ (mestre@) gomoku ~ gomoku.h > Set as __dead a few more functions, that don't return, on games/ > With precious tip, help and also OK from tb@ (mestre@) grdc ~ grdc.c > Declare usage() functions as __dead void, if they don't return, on games > section. > Found another one in arithmetic(6) which also didn't return, and removed a > function from number(6) which is not used anymore. > OK tb@ (mestre@) hack ~ hack.h > Set as __dead a few more functions, that don't return, on games/ > With precious tip, help and also OK from tb@ (mestre@) hangman ~ main.c > Declare usage() functions as __dead void, if they don't return, on games > section. > Found another one in arithmetic(6) which also didn't return, and removed a > function from number(6) which is not used anymore. > OK tb@ (mestre@) ~ hangman.h > Set as __dead a few more functions, that don't return, on games/ > With precious tip, help and also OK from tb@ (mestre@) hunt ~ hunt/hunt.c ~ huntd/server.h > Set as __dead a few more functions, that don't return, on games/ > With precious tip, help and also OK from tb@ (mestre@) mille ~ mille.h > Set as __dead a few more functions, that don't return, on games/ > With precious tip, help and also OK from tb@ (mestre@) number ~ number.c > Declare usage() functions as __dead void, if they don't return, on games > section. > Found another one in arithmetic(6) which also didn't return, and removed a > function from number(6) which is not used anymore. > OK tb@ (mestre@) phantasia ~ misc.c ~ setup.c > Set as __dead a few more functions, that don't return, on games/ > With precious tip, help and also OK from tb@ (mestre@) pig ~ pig.c > Declare usage() functions as __dead void, if they don't return, on games > section. > Found another one in arithmetic(6) which also didn't return, and removed a > function from number(6) which is not used anymore. > OK tb@ (mestre@) pom ~ pom.c > Set as __dead a few more functions, that don't return, on games/ > With precious tip, help and also OK from tb@ (mestre@) ppt ~ ppt.c > Declare usage() functions as __dead void, if they don't return, on games > section. > Found another one in arithmetic(6) which also didn't return, and removed a > function from number(6) which is not used anymore. > OK tb@ (mestre@) primes ~ primes.c > Declare usage() functions as __dead void, if they don't return, on games > section. > Found another one in arithmetic(6) which also didn't return, and removed a > function from number(6) which is not used anymore. > OK tb@ (mestre@) quiz ~ quiz.c > Declare usage() functions as __dead void, if they don't return, on games > section. > Found another one in arithmetic(6) which also didn't return, and removed a > function from number(6) which is not used anymore. > OK tb@ (mestre@) random ~ random.c > Declare usage() functions as __dead void, if they don't return, on games > section. > Found another one in arithmetic(6) which also didn't return, and removed a > function from number(6) which is not used anymore. > OK tb@ (mestre@) robots ~ robots.h > Declare usage() functions as __dead void, if they don't return, on games > section. > Found another one in arithmetic(6) which also didn't return, and removed a > function from number(6) which is not used anymore. > OK tb@ (mestre@) ~ robots.h > Set as __dead a few more functions, that don't return, on games/ > With precious tip, help and also OK from tb@ (mestre@) sail ~ extern.h > Set as __dead a few more functions, that don't return, on games/ > With precious tip, help and also OK from tb@ (mestre@) tetris ~ tetris.c > Declare usage() functions as __dead void, if they don't return, on games > section. > Found another one in arithmetic(6) which also didn't return, and removed a > function from number(6) which is not used anymore. > OK tb@ (mestre@) ~ scores.c > remove unused variable > ok tb@ (chl@) worm ~ worm.c > Set as __dead a few more functions, that don't return, on games/ > With precious tip, help and also OK from tb@ (mestre@) wump ~ wump.c > Declare usage() functions as __dead void, if they don't return, on games > section. > Found another one in arithmetic(6) which also didn't return, and removed a > function from number(6) which is not used anymore. > OK tb@ (mestre@) == gnu =============================================================== 05/14 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/gnu gnu ~ gcc/gcc/config/mips/sync.S > Fix branch delay slots. Found while making test builds of ports. (visa@) usr.bin/binutils-2.17 ~ include/opcode/i386.h ~ opcodes/i386-dis.c > Recognise the rdseed instruction introduced with Broadwell. > Similar to changes made in FreeBSD. > ok kettenis@ (jsg@) usr.bin/cxxfilt ~ cxxfilt.c > add missing header needed for pledge(2) > ok deraadt@ (chl@) == include =========================================================== 06/14 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/include sndio.h ~ sndio.h > Expose internal functions necessary to open audio devices and midi > ports using existing file descriptors. (ratchov@) == lib =============================================================== 07/14 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/lib libc ~ stdio/fgetwc.c > Both our manual and POSIX ask us to set the error indicator when an > encoding error occurs, so do it. > While here, do not set errno after mbrtowc(3) failure; mbrtowc(3) > already does that, and that behaviour is required by the standard. > ok jca@ guenther@ "nice find" deraadt@ (schwarze@) ~ sys/poll.2 > Add missing second argument to FD_SET macro in example code snippet. > Spotted and diff provided by d ! lowe () openmailbox ! org, thanks! > ok schwarze@ (tb@) ~ time/strftime.c > Simplify allocation logic based on free and realloc's behavior on NULL. > ok tedu@ (mmcc@) ~ stdlib/icdb.c > remove unused variable > ok tedu@ (chl@) libevent ~ kqueue.c > revert change to call kevent immediately. > tcpbench (at a minimum) relies on the old behavior of changes all happening > after all event handlers run. in particular, it resets the event for the > listening socket *before* calling accept(), when it is still readable. > kevent then (correctly) says it is readable on the next go through the > loop. > silly, subtle, and stupid. > problem reported by kettenis (tedu@) libfuse ~ fuse.c > remove NULL-check before free() (mmcc@) libkeynote ~ keynote-ver.l > Clean up a condition. No functional change. ok tb@ (mmcc@) ~ environment.c > remove a few NULL-checks before free() (mmcc@) ~ signature.c > Remove NULL casts and tweak style in a few places. A more invasive > version of this diff was reviewed by tb@ (mmcc@) ~ signature.c > Remove the rest of the scaffolding for nonexistent ElGamal support, > which has been disabled for sixteen years. > ok millert@, tb@, sthen@ (mmcc@) ~ keynote-ver.l > remove two NULL casts (mmcc@) libpcap ~ gencode.c ~ inet.c ~ optimize.c ~ pcap-bpf.c ~ savefile.c > remove NULL-checks before free() (mmcc@) libsndio ~ mio_rmidi.c ~ shlib_version ~ sio_sun.c > Expose internal functions necessary to open audio devices and midi > ports using existing file descriptors. (ratchov@) ~ aucat.c > remove NULL-checks before free() (mmcc@) libsqlite3 ~ VERSION ~ shlib_version ~ sqlite3.pc ~ ext/async/sqlite3async.c ~ ext/fts3/fts3.c ~ ext/fts3/fts3Int.h ~ ext/fts3/fts3_expr.c ~ ext/fts3/fts3_write.c ~ ext/misc/spellfix.c ~ ext/rtree/rtree1.test ~ src/alter.c ~ src/analyze.c ~ src/backup.c ~ src/btree.c ~ src/btree.h ~ src/btreeInt.h ~ src/build.c ~ src/ctime.c ~ src/date.c ~ src/dbstat.c ~ src/delete.c ~ src/expr.c ~ src/fkey.c ~ src/func.c ~ src/insert.c ~ src/lempar.c ~ src/loadext.c ~ src/main.c ~ src/malloc.c ~ src/mutex.c ~ src/mutex_unix.c ~ src/mutex_w32.c ~ src/os_unix.c ~ src/os_win.c ~ src/pager.c ~ src/pager.h ~ src/parse.y ~ src/pcache.c ~ src/pcache1.c ~ src/pragma.c ~ src/printf.c ~ src/resolve.c ~ src/select.c ~ src/sqlite.h.in ~ src/sqlite3.h ~ src/sqlite3ext.h ~ src/sqliteInt.h ~ src/test1.c ~ src/test_config.c ~ src/test_func.c ~ src/threads.c ~ src/tokenize.c ~ src/treeview.c ~ src/update.c ~ src/util.c ~ src/vdbe.c ~ src/vdbe.h ~ src/vdbeInt.h ~ src/vdbeapi.c ~ src/vdbeaux.c ~ src/vdbeblob.c ~ src/vdbemem.c ~ src/vtab.c ~ src/wal.c ~ src/walker.c ~ src/where.c ~ src/whereInt.h ~ src/wherecode.c ~ src/whereexpr.c ~ tool/build-all-msvc.bat ~ tool/fuzzershell.c ~ tool/lemon.c ~ tool/lempar.c ~ tool/mksqlite3c.tcl ~ tool/mksqlite3h.tcl ~ tool/mkvsix.tcl ~ tool/showdb.c ~ tool/spaceanal.tcl ~ tool/sqldiff.c ~ tool/tostr.awk > Update sqlite3 to 3.9.2. Bump major, regen .pc and header. Changes > available here: http://sqlite.org/changes.html > Tested in bulk by aja@. ok landry@ (jturner@) libssl ~ src/crypto/asn1/tasn_enc.c > initialize a pointer to NULL rather than 0 (mmcc@) ~ src/crypto/asn1/bio_asn1.c > assign pointer to NULL rather than 0 (mmcc@) ~ src/crypto/asn1/tasn_prn.c > assign pointer NULL rather than 0 (mmcc@) ~ src/crypto/asn1/asn_pack.c ~ src/crypto/bio/bss_bio.c ~ src/crypto/comp/c_zlib.c > remove NULL-checks before free() (mmcc@) ~ src/crypto/bn/asm/sparcv8plus.S ~ src/crypto/des/asm/des_enc.m4 > More adress -> address (mmcc@) == libexec =========================================================== 08/14 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/libexec getty ~ subr.c > assign pointers to NULL rather than 0 (mmcc@) ld.so ~ ldconfig/shlib.c > Remove a couple needless allocation casts. No binary change. (mmcc@) ~ dlfcn.c ~ library.c ~ library_mquery.c ~ alpha/rtld_machine.c > assign pointers to NULL rather than 0 (mmcc@) login_token ~ login_token.c > assign pointers to NULL rather than 0 (mmcc@) == regress =========================================================== 09/14 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/regress sys ~ netinet/in_pcbbind/Makefile ~ netinet/in_pcbbind/runtest.c > Correct expected results on v6 mcast bind on ANY, and add comments to > explain why they are expected. (vgross@) ~ netinet/arp/README > Add my machines' configuration files as an example to the readme. > That should make it easier for others to set up the test. (bluhm@) usr.bin ~ make/Makefile + make/mk40 + make/mk41 > add parsing tests that must not segfault based on Jonathan Gray initial > report. (espie@) + colrm/Makefile + colrm/colrm.sh > Test suite for colrm(1) including tests for single byte characters, > tabs, backspaces, multibyte characters of width 1, 0, and 2, > invalid bytes and non-printable characters. All tests are run > both with the UTF-8 and the C locale. > About half the tests are currently broken, and not only the multibyte > tests, but also half of the tab and backspace tests. The program is > expected to be fixed soon, and the suite will be hooked up after that. > (schwarze@) == sbin ============================================================== 10/14 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/sbin disklabel ~ disklabel.8 > tweak auto-allocation description, with and ok jmc@ (otto@) dump ~ itime.c > assign pointers to NULL rather than 0, remove a needless cast (mmcc@) fdisk ~ mbr.c > Use 'standard' gpt_chk_mbr() to check for protective MBR. (krw@) init ~ init.c > Use NULL rather than 0 for pointers. No binary change. (mmcc@) mountd ~ Makefile ~ mountd.c > Add privilege separation and pledge(). > A privileged (that is, unpledged) child is forked that makes the > "unpledgable" > getfh(2) and mount(2) calls. It also takes care of writing > /var/db/mountdtab so > that the parent does not have to pledge "cpath wpath". > This, then, allows the parent to pledge "stdio rpath inet dns". > Suggestions and OK deraadt@ (tim@) ~ mountd.c > Remove unnecessary SIGSYS/ENOSYS handling for getfh(2). > Encouragement from deraadt@ (tim@) == share ============================================================= 11/14 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/share man ~ man9/Makefile ~ man9/kern.9 > improve CTASSERT man page > * add CTASSERT.9 link > * add "void" to prototype (sf@) ~ man4/Makefile + man4/viocon.4 > add virtio-console driver > This patch adds a new driver for use of virtio-console devices as > ttys. It's still in an early state and not compiled by default, yet. > Currently it is only wired into amd64. i386 is still missing. > Discussed with uebayasi@, deraadt@ (sf@) ~ man8/man8.amd64/MAKEDEV.8 > regen MAKEDEV (sf@) ~ man4/viocon.4 > minor tweaks; (jmc@) ~ man1/gcc-local.1 > We support PIE on static executables now. > ok pascal@ (bentley@) ~ man5/ar.5 > mdoc macro cleanup. > - mark up types with Vt > - clean up some Tn > - Ql instead of ` ' > ok jmc@ schwarze@ (bentley@) ~ man4/drm.4 > Mark up paths with Pa, and adjust tag width to avoid ugly wrapping. > ok jmc@ schwarze@ (bentley@) ~ man8/starttls.8 > more e-mail -> email (mmcc@) ~ man9/bufq_init.9 > initialiases -> initialises (bentley@) == sys =============================================================== 12/14 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/sys arch/amd64/amd64 ~ conf.c > add virtio-console driver > This patch adds a new driver for use of virtio-console devices as > ttys. It's still in an early state and not compiled by default, yet. > Currently it is only wired into amd64. i386 is still missing. > Discussed with uebayasi@, deraadt@ (sf@) ~ vmm.c > Make sure we don't overflow a page during vm_readpage/vm_writepage. > Noticed over a month ago by Stefan Kempf <sn.kempf at t-online.de>, and I > shamefully just got around to committing it. Thanks Stefan. (mlarkin@) ~ vmm.c > Normalize return values from various exit functions. > Another diff from Stefan Kempf <sn.kempf at t-online.de>. (mlarkin@) ~ autoconf.c ~ cpu.c ~ identcpu.c > If available prefer the rdseed instruction over rdrand when adding entropy > to the kernel rng. If the rdseed source is empty fallback to rdrand > as suggested by naddy. rdrand output comes from a prng that is > periodically reseeded. rdseed should give us more bits of entropy. > ok naddy@ djm@ deraadt@ (jsg@) arch/amd64/conf ~ GENERIC > add virtio-console driver > This patch adds a new driver for use of virtio-console devices as > ttys. It's still in an early state and not compiled by default, yet. > Currently it is only wired into amd64. i386 is still missing. > Discussed with uebayasi@, deraadt@ (sf@) arch/amd64/stand/efiboot ~ efidev.c > Add prototype for gpt_chk_mbr() to i386_installboot.c. Add 'static' to > declarations of gpt_chk_mbr() to match prototypes in efidev.c and > softraid.c. (krw@) ~ efidev.c > Make all instances of gpt_chk_mbr() identical (bar static vs > non-static) by passing disk size as 2nd parameter instead of the > different structures holding the disk size info. The fifth copy of > gpt_chk_mbr() in fdisk is a little specialer. No functional change. (krw@) ~ efidev.c > If the GPT header checksum check fails, do *not* say the OpenBSD partition > starts at sector 1. Instead return -1, as for all other failures during > the header check. (krw@) arch/amd64/stand/libsa ~ softraid.c > Find OpenBSD disklabel on GPT partitioned softraid volumes. Now you can > boot from such volumes to match being able to install to them. > Problem spotted by jcs@, fix tested by rpe@, ok jsing@ (krw@) ~ softraid.c > Ooops. Wrong version of the diff committed, with reversed check for > MBR/GPT. > Problem found the hard way by Glenn Faustino on tech@ (krw@) ~ softraid.c > Argh. The other part the wrong diff was missing. > Use the size info in struct sr_boot_volume, and not the uninitialized > info in the disklabel, when checking the GPT protective MBR. Now > both GPT and MBR formatted softraid volumes actually boot. > As a bonus add prototype for gpt_chk_mbr(). (krw@) ~ softraid.c > Add prototype for gpt_chk_mbr() to i386_installboot.c. Add 'static' to > declarations of gpt_chk_mbr() to match prototypes in efidev.c and > softraid.c. (krw@) ~ softraid.c > Make all instances of gpt_chk_mbr() identical (bar static vs > non-static) by passing disk size as 2nd parameter instead of the > different structures holding the disk size info. The fifth copy of > gpt_chk_mbr() in fdisk is a little specialer. No functional change. (krw@) ~ softraid.c > If the GPT header checksum check fails, do *not* say the OpenBSD partition > starts at sector 1. Instead return -1, as for all other failures during > the header check. (krw@) arch/i386/i386 ~ autoconf.c ~ cpu.c ~ machdep.c > If available prefer the rdseed instruction over rdrand when adding entropy > to the kernel rng. If the rdseed source is empty fallback to rdrand > as suggested by naddy. rdrand output comes from a prng that is > periodically reseeded. rdseed should give us more bits of entropy. > ok naddy@ djm@ deraadt@ (jsg@) arch/mips64/include ~ cpu.h > Make interrupt masking MP-aware. Linux IP27 and IP35 ports served as a > substitute for hardware documentation. (visa@) arch/mips64/mips64 ~ cpu.c > Make interrupt masking MP-aware. Linux IP27 and IP35 ports served as a > substitute for hardware documentation. (visa@) arch/sgi/conf + GENERIC-IP27.MP > welcome GENERIC.IP27 MP (dual cpu on O350 now) > ok visa (deraadt@) arch/sgi/include ~ asm.h > Use the ErrorEPC register for curcpu() on Origin, for now. (visa@) ~ autoconf.h > Make interrupt masking MP-aware. Linux IP27 and IP35 ports served as a > substitute for hardware documentation. (visa@) ~ intr.h > Add IPI logic. Assign two additional interrupts for inter-processor > signalling as a workaround to a limitation in the hub interrupt code, > to allow four CPUs per node. At the moment, multi-node setups are not > supported. (visa@) arch/sgi/localbus ~ tcc.c > Execution of the `sync' instruction is not supported by the T-Bus and > raises a machine check exception. Do not report anything on console in > that case, so that userland does not suffer too much. > Diff from miod@ (visa@) arch/sgi/sgi ~ wscons_machdep.c > commited -> committed (mmcc@) ~ locore.S > Use the ErrorEPC register for curcpu() on Origin, for now. (visa@) ~ ip27_machdep.c > Add timecounter for MP. (visa@) ~ ip27.h ~ ip27_machdep.c > Make interrupt masking MP-aware. Linux IP27 and IP35 ports served as a > substitute for hardware documentation. (visa@) ~ ip27.h ~ ip27_machdep.c ~ locore.S > Add launch logic for secondary CPUs. The PROM's launch entry point address > and function signature are from Linux. (visa@) ~ intr_template.c ~ ip27_machdep.c > Add IPI logic. Assign two additional interrupts for inter-processor > signalling as a workaround to a limitation in the hub interrupt code, > to allow four CPUs per node. At the moment, multi-node setups are not > supported. (visa@) arch/sgi/xbow ~ hub.h > Add timecounter for MP. (visa@) ~ hub.h > Add IPI logic. Assign two additional interrupts for inter-processor > signalling as a workaround to a limitation in the hub interrupt code, > to allow four CPUs per node. At the moment, multi-node setups are not > supported. (visa@) arch/zaurus/include ~ machine_reg.h > More adress -> address (mmcc@) ddb ~ db_command.c ~ db_sym.c > assign pointers NULL rather than 0 (mmcc@) dev/acpi ~ acpicpu.c > add aml_freevalue in a couple places, avoid storing a result that > would need to be freed in a couple other places > with deraadt (jcs@) dev/isa ~ asmc.c > The Apple SMC is implemented on a H8S/2117 chip. Communication with the > chip > becomes much more reliably if we pay attention to the IBF and OBF flags. > This removes the need for excessive delays and retries, making operations > complete in microseconds instead of seconds. > This diff also removes flushing the read buffer after a read operations. > This > doesn't seem to be necessary, but we might need to put that back if the > length of a reply changes for some keys. > ok jung@, deraadt@ (kettenis@) ~ asmc.c > with Mark's fix in previous commit the extra taskq for delayed > initialization > and updates is no longer needed and can be removed, so simplify things and > initialize the sensors on attach and update them through the sensors task > (jung@) ~ asmc.c > simplify and unify keyboard backlight hooks, doing the same kassert instead > of > for-loop as acpithinkpad does (jung@) ~ asmc.c > extra taskq is gone, so also remove the no longer needed init variable > (jung@) ~ asmc.c > rename kbdled to backlight and mention chip model in comment (jung@) ~ asmc.c > fold for loops back into a wait function in a similar manner as acpiec does > (jung@) ~ asmc.c > set sensor type later on init (jung@) dev/pci ~ drm/drm_drv.c > Advertise support for DRM version 1.4. We have support for all the > necessary > features, and some code (such as the Xorg modesetting driver) actually > checks > the version. With this change the PCI-based probe method in that driver > works. > ok jsg@ (kettenis@) ~ xspd.c > Cleanup hypercall subsystem type defines (mikeb@) ~ pciide.c > Goto a loop_end label rather than using a continue statement for an > error condition in an inner loop. Found with Coccinelle. > Pointed out by tedu@, ok kettenis@ and deraadt@ about three months ago. > (mmcc@) ~ files.pci + viocon.c > add virtio-console driver > This patch adds a new driver for use of virtio-console devices as > ttys. It's still in an early state and not compiled by default, yet. > Currently it is only wired into amd64. i386 is still missing. > Discussed with uebayasi@, deraadt@ (sf@) ~ agp.c > Remove obsolete comment. (kettenis@) ~ viocon.c > Remove a couple NULL-checks before free(). > ok sf@ (mmcc@) ~ pcidevs > Add Perle Speed8 LE. (kettenis@) ~ pcidevs.h ~ pcidevs_data.h > regen (kettenis@) ~ pucdata.c > Add support for the Perle Speed8 LE. (kettenis@) dev/pv ~ xen.c > Don't unmask the port in xen_intr_establish (mikeb@) ~ xen.c ~ xenstore.c ~ xenvar.h > Cleanup hypercall subsystem type defines (mikeb@) ~ xen.c ~ xenvar.h > Introduce xen_intr_mask and xen_intr_unmask primitives > Mask the event port during xen_intr_establish, but don't set the > masked flag in the intsrc. By providing mask and unmask routines > we allow the device to decide when to perform these actions. The > port will still be unmasked during xen_intr_enable. This allows > netfront to fulfil the intr_barrier pattern requirements fairly > easily and at the same time should be sufficient for diskfront > that doesn't need to fiddle with interrupt masking. (mikeb@) ~ xen.c ~ xenreg.h ~ xenstore.c ~ xenvar.h > Implement a bus_dma(9) abstraction on top of Grant Table API (mikeb@) ~ xenstore.c ~ xenvar.h > Make xs_setprop a bit more useful by removing property nodes > when NULL or zero length value was specified. (mikeb@) dev/usb ~ uvideo.c > Do not delay video(4) attachment if the device does not need to load > a firmware. > Regression spotted by deraadt@ (mpi@) ~ dwc2/dwc2.c > Drop unused goto label. (visa@) ~ umcs.h > more e-mail -> email (mmcc@) ~ ukbd.c > enable iso keyboard munge fix for macbookair6,2 > ok jcs (jung@) kern ~ subr_autoconf.c > Simple free(9) sizes, from Matthieu <naabed AT poolp DOT org>. (mpi@) ~ kern_event.c > size for free, sent and reminded by Mathieu. also delete null check. > (tedu@) ~ tty_conf.c ~ tty_endrun.c ~ tty_msts.c ~ tty_nmea.c > Move ppp, nmea, endrun, and msts prototypes to tty.h > Fix inconsistent arguments for pppopen/pppclose/pppstart. > Use passed in argument p in pppopen instead of curproc. > "Looks good to me" deraadt@ (sf@) ~ uipc_mbuf.c > - yet another tiny step towards MP PF. This time we need to make sure > statekey attached to packet stays around, while accepted packet is > routed through IP stack. > OK mpi@, henning@ (sashan@) ~ tty_conf.c > Re-add config includes > These got lost in previous commit. This broke ppp, nmea, msts, endrun. > Found the hard way by David Coppa (sf@) ~ vfs_lockf.c > spliting -> splitting (mmcc@) ~ kern_sched.c > One "sbar" taskq is enough. > ok visa@ (kettenis@) ~ kern_pledge.c > If PLEDGE_AUDIO is set, allow audio(4) ioctls necessary > to use raw audio devices. > ok deraadt, semarie (ratchov@) ~ uipc_mbuf.c > revert previous: > ---------------------------------------------------------------------- > revision 1.961 > date: 2015/12/22 13:33:26; author: sashan; state: Exp; lines: +153 -44; > commitid: oBRhtWcDV0ThviVT; > - yet another tiny step towards MP PF. This time we need to make sure > statekey attached to packet stays around, while accepted packet is > routed through IP stack. > OK mpi@, henning@ > ---------------------------------------------------------------------- > there have been multiple reports of KASSERT(!pf_state_key_isvalid(sk)) > being > triggered without much effort, so back this out for now. (jasper@) ~ kern_pledge.c > pledge "audio" code block must be !SMALL_KERNEL (deraadt@) ~ subr_disk.c > Make all instances of gpt_chk_mbr() identical (bar static vs > non-static) by passing disk size as 2nd parameter instead of the > different structures holding the disk size info. The fifth copy of > gpt_chk_mbr() in fdisk is a little specialer. No functional change. (krw@) ~ subr_disk.c > Make gpt_chk_mbr() instances even more identicaller by always > returning 0 or 1 since only boolean checks of the result are done. > No need for EINVAL as a return value. (krw@) ~ kern_pledge.c > wrap "audio" pledge code in NAUDIO > 0, because there are a few > GENERIC kernels which lack audio drivers. (deraadt@) ~ kern_pledge.c > Add missing #include "audio.h". Fixes sndiod pledge issue reported by > timo.myyra () wickedbsd ! net, thanks! > While there, remove extraneous "pty.h". > ok deraadt@ (tb@) net ~ rtable.c ~ rtable.h ~ route.c > Pass the destination and mask to rtable_mpath_reprio() in order to not > use ``rt_node'' with ART. (mpi@) ~ ppp_tty.c > Move ppp, nmea, endrun, and msts prototypes to tty.h > Fix inconsistent arguments for pppopen/pppclose/pppstart. > Use passed in argument p in pppopen instead of curproc. > "Looks good to me" deraadt@ (sf@) ~ pf.c ~ pfvar.h > - yet another tiny step towards MP PF. This time we need to make sure > statekey attached to packet stays around, while accepted packet is > routed through IP stack. > OK mpi@, henning@ (sashan@) ~ pf.c ~ pfvar.h > revert previous: > ---------------------------------------------------------------------- > revision 1.961 > date: 2015/12/22 13:33:26; author: sashan; state: Exp; lines: +153 -44; > commitid: oBRhtWcDV0ThviVT; > - yet another tiny step towards MP PF. This time we need to make sure > statekey attached to packet stays around, while accepted packet is > routed through IP stack. > OK mpi@, henning@ > ---------------------------------------------------------------------- > there have been multiple reports of KASSERT(!pf_state_key_isvalid(sk)) > being > triggered without much effort, so back this out for now. (jasper@) netinet6 ~ in6.c > Do not return EEXIST if the same address is added twice. > Reported by sebastisa@. > ok sebasitia@, stsp@, florian@, vgross@ (mpi@) nfs ~ nfs_vnops.c > commited -> committed (mmcc@) sys ~ tty.h > Move ppp, nmea, endrun, and msts prototypes to tty.h > Fix inconsistent arguments for pppopen/pppclose/pppstart. > Use passed in argument p in pppopen instead of curproc. > "Looks good to me" deraadt@ (sf@) == usr.bin =========================================================== 13/14 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/usr.bin aucat ~ aucat.1 > follwing -> following (bentley@) basename ~ basename.c > Order includes. From [email protected] (mmcc@) ~ basename.c > Use return rather than exit to enable the stack protector. From > [email protected]. > ok tb@ (mmcc@) cvs ~ tag.c > commited -> committed (mmcc@) ~ cvs.1 > more e-mail -> email (mmcc@) file ~ file.1 ~ file.c > Add --brief and --dereference, used by xdg-open/xdg-mime. > From Ralf Horstmann, ok tb@ deraadt@ nicm@ sthen@ (jca@) finger ~ util.c > Use pread/pwrite instead separate lseek+read/write for lastlog. > Cast to off_t before multiplication to avoid truncation on ILP32 > ok kettenis@ mmcc@ (guenther@) grep ~ util.c > The loop index in grep_cmp() should be size_t to match the type of > the length parameter. The return value of grep_cmp() is only used > in a boolean context so make it return bool instead of the index. > OK mmcc@ (millert@) ~ util.c > cast toupper's argument to unsigned char > ok millert@ (mmcc@) kdump ~ ktrstruct.c > very basic kevent printing (tedu@) less ~ linenum.c > inline a once-used function that just prints a warning > ok benno@ (mmcc@) ~ ch.c > Put macro arg in parentheses. > ok nicm@ (mmcc@) login ~ login.c > Use pread/pwrite instead separate lseek+read/write for lastlog. > Cast to off_t before multiplication to avoid truncation on ILP32 > ok kettenis@ mmcc@ (guenther@) make ~ parse.c > in case we have a parse error, dispell the target list early to avoid > various problems. > Problem noticed by jsg@, deeper patch than he suggested > okay jsg@ and tb@. (espie@) mandoc ~ term_ps.c > remove NULL-checks before free() (mmcc@) ~ cgi.c > Don't retain the search query in the resulting manual links. > Clean, simple URLs are best. > ok schwarze@ (bentley@) ~ html.c ~ mdoc_html.c > Generate simpler in-page links: just replace spaces with underscores. > So > http://example.com/OpenBSD-current/man1/ls.1#x546865204c6f6e6720466f726d617 > 4 > becomes > http://example.com/OpenBSD-current/man1/ls.1#The_Long_Format > ok schwarze@ (bentley@) mg ~ tutorial > Mention dired. (lum@) ~ mg.1 > Add RET to dired-find-file. ok jasper@ > (and jasper@ ok'd recent tutorial diff, forgot to mention) (lum@) ~ mg.1 > Remove commands that are not implemented, at the moment. Added by > accident in previous diff. (lum@) ~ theo.c > Do undo. (lum@) ~ mg.1 ~ main.c > Add -R option to allow files specified on the command line to be > opened read-only. ok jasper@ phessler@ and man page advice from jmc@ (lum@) openssl ~ s_server.c > remove NULL-check before free() (mmcc@) ~ ca.c ~ openssl.1 > more e-mail -> email (mmcc@) rdist ~ client.h ~ common.c > remove needless allocation casts (mmcc@) rdistd ~ server.c > Remove a couple needless allocation casts. No binary change. (mmcc@) ~ server.c > remove needless allocation casts (mmcc@) sndiod ~ Makefile ~ listen.c ~ miofile.c ~ siofile.c ~ sndiod.c + fdpass.c + fdpass.h > In case of a bug in sndiod, an attacker (a local user) could run > arbitrary code as user _sndio, i.e. get a second uid. > Mitigate the risk by implementing initial privilege separation as > follows. Break sndiod in two processes: a chroot()ed "worker" process > processing input, and a non-chroot()ed "helper" process opening > devices and passing descriptors to the worker. > With help from benno, claudio, semarie and gilles. > ok benno, semarie and tb (ratchov@) ~ listen.c ~ sndiod.c > Don't attempt to remove unix socket as it's owned by root and > we've already dropped root privileges. (ratchov@) ~ fdpass.c > hide forgotten debug printfs (ratchov@) ~ sndiod.c > Call getpwname() earlier, and if it fails, return the proper > exit code, so that the failure gets reported by /etc/rc. (ratchov@) ~ sndiod.c > fix false-positive "uninitialized" gcc warning (ratchov@) ~ sndiod.c > Add pledge calls to both "helper" and "worker" processes. > ok deraadt, semarie (ratchov@) ~ listen.c > Don't set (unused) IPV6_V6ONLY option. > suggested by deraadt (ratchov@) ~ sndiod.1 > follwing -> following (bentley@) sqlite3 ~ shell.c > Update sqlite3 to 3.9.2. Bump major, regen .pc and header. Changes > available here: http://sqlite.org/changes.html > Tested in bulk by aja@. ok landry@ (jturner@) ssh ~ clientloop.c ~ mux.c > adjust pledge promises for ControlMaster: when using "ask" or "autoask", > the process will use ssh-askpass for asking confirmation. > problem found by halex@ > ok halex@ (semarie@) ~ sshlogin.c > Use pread/pwrite instead separate lseek+read/write for lastlog. > Cast to off_t before multiplication to avoid truncation on ILP32 > ok kettenis@ mmcc@ (guenther@) telnet ~ telnet.c > Add rpath to telnet's more restrictive pledge; otherwise if the server > requests > the client to send the next terminal-type (RFC-1091), the client tries to > read > terminfo files and is killed by pledge. Noticed when connecting to 4.3 BSD. > ok semarie@ tb@ (sthen@) tmux ~ cmd-attach-session.c > Detach the right session with -d. (nicm@) ~ cmd-switch-client.c > Repair switch-client -l and switch-client with a window target. (nicm@) uname ~ uname.c > KNF cleanup: sort #includes and cases in switch statements. > Fix whitespace issues with if-statements. > Use return instead of exit. > From fritjof () alokat ! org, thanks! > ok benno@ (tb@) usbhidctl ~ usbhid.c > remove a NULL-check before free() (mmcc@) vmstat ~ dkstats.c ~ vmstat.c > Replace bzero with memset. No binary change. > ok tb@ (mmcc@) == usr.sbin ========================================================== 14/14 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/usr.sbin adduser ~ adduser.8 > more e-mail -> email (mmcc@) bgpctl ~ mrtparser.c > Remove NULL-checks before free(). I think I've almost scrubbed the > entire tree of these... (mmcc@) bgpd ~ rde.c > commited -> committed (mmcc@) ~ pftable.c > Remove NULL-checks before free(). I think I've almost scrubbed the > entire tree of these... (mmcc@) dhcpd ~ alloc.c ~ dhcp.c ~ tree.c > remove NULL-checks before free() (mmcc@) eigrpd ~ eigrpd.conf.5 > adress -> address (bentley@) ftp-proxy ~ ftp-proxy.c > commiting -> committing (mmcc@) gpioctl ~ gpioctl.c > bzero -> memset. No binary change. > ok tb@ (mmcc@) hostapd ~ hostapd.c > remove a NULL-check before free() (mmcc@) inetd ~ inetd.c > remove a NULL-check before free() (mmcc@) installboot ~ i386_installboot.c > Remove unnecessary options: > - in fsck_msdos, '-f' option is a no-op > - newfs_msdos does not need '-t msdos' > OK krw@ (rpe@) ~ i386_installboot.c > Add prototype for gpt_chk_mbr() to i386_installboot.c. Add 'static' to > declarations of gpt_chk_mbr() to match prototypes in efidev.c and > softraid.c. (krw@) ~ i386_installboot.c > Make all instances of gpt_chk_mbr() identical (bar static vs > non-static) by passing disk size as 2nd parameter instead of the > different structures holding the disk size info. The fifth copy of > gpt_chk_mbr() in fdisk is a little specialer. No functional change. (krw@) ~ i386_installboot.c > Make gpt_chk_mbr() instances even more identicaller by always > returning 0 or 1 since only boolean checks of the result are done. > No need for EINVAL as a return value. (krw@) ldapd ~ btree.c > commiting -> committing (mmcc@) ~ search.c > use strndup instead of malloc/strncpy/nul > ok krw@ (mmcc@) ~ auth.c ~ ber.c ~ btest.c ~ btree.c ~ control.c ~ index.c ~ ldapd.c ~ ldape.c ~ modify.c ~ namespace.c ~ search.c ~ util.c > bzero -> memset. No binary change. (mmcc@) memconfig ~ memconfig.c > remove NULL-check before free() (mmcc@) mtree ~ mtree.c > need to include err.h for err(). > ok deraadt@ (benno@) ~ compare.c > Remove NULL-checks before free(). (mmcc@) pppd ~ auth.c > Use pread/pwrite instead separate lseek+read/write for lastlog. > Cast to off_t before multiplication to avoid truncation on ILP32 > ok kettenis@ mmcc@ (guenther@) rdate ~ rdate.c > expand message between processes to 2048 bytes. Andre Smagin found > a case where it can reach 637 bytes of output, so use almost 4x > (he suggested 1024) (deraadt@) relayd ~ relay.c > completly -> completely (mmcc@) ~ check_tcp.c > revert check_tcp rev 1.41 by yasuoka@ > This part was not meant to be commited, only the bits in check_icmp.c > and relayd.h are ok. > ok yasuoka@ (benno@) smtpd ~ smtpd.c > remove redundant log (gilles@) ~ smtpctl.c > Start accepting some sendmail compatible commandline arguments. > This diff teaches smtpctl to understand '-bi -O AliasFile=<path_aliases>' > and generate aliases.db. > Ok gilles@ jung@ millert@ (sunil@) ~ table_static.c > Wrap long lines. > Ok gilles@ jung@ (sunil@) ~ delivery_mbox.c ~ ioev.c > Wrap long lines. > Ok gilles@ jung@ (sunil@) ~ smtpctl.c > Kill a trailing whitespace that snuck in with previous commit. (sunil@) ~ newaliases.8 > make it clear this needs to be run only for db files > since smtpd has switched to using plain files by default; > ok sunil gilles (jmc@) ~ smtpd.conf.5 > update examples > diff from Serguey Parkhomovsky <[email protected]> > ok sunil@ (gilles@) ~ smtpctl.c > fix sendmail options parsing > diff from Serguey Parkhomovsky <[email protected]> (gilles@) ~ bounce.c > more e-mail -> email (mmcc@) snmpd ~ snmpd.conf.5 > more e-mail -> email (mmcc@) sysmerge ~ sysmerge.sh > Only run newaliases(8) if we already have an aliases db. (ajacoutot@) ~ sysmerge.sh > Tidy up output. (ajacoutot@) tcpdump ~ print-bootp.c ~ print-gtp.c ~ print-udp.c ~ tcpdump.c > assign pointers to NULL rather than 0 (mmcc@) tokeninit ~ tokeninit.c > assign pointers to NULL rather than 0 (mmcc@) usbdevs ~ usbdevs.c > Initialize a char* to NULL rather than 0. (mmcc@) =============================================================================== _______________________________________________ owc mailing list [email protected] http://www.squish.net/mailman/listinfo/owc
