OpenBSD src changes summary for 2015-11-23 ==========================================
distrib/sets lib/libc sbin/disklabel sbin/dumpfs sbin/fsck sbin/fsck_ext2fs sbin/fsck_ffs sbin/fsck_msdos sbin/fsirand sbin/growfs sbin/ifconfig sbin/iked sbin/mount sbin/ncheck_ffs sbin/newfs sbin/newfs_ext2fs sbin/newfs_msdos sbin/scan_ffs sbin/tunefs share/man sys/arch/amd64/amd64 sys/arch/amd64/conf sys/arch/amd64/include sys/arch/amd64/stand/boot sys/arch/i386/stand/boot sys/arch/macppc/conf sys/arch/macppc/dev sys/dev/acpi sys/dev/pci sys/kern sys/net sys/netinet sys/sys usr.bin/bc usr.bin/less usr.bin/nc usr.bin/sndiod usr.bin/tmux usr.bin/vi usr.sbin/httpd usr.sbin/kvm_mkdb usr.sbin/relayd usr.sbin/smtpd usr.sbin/snmpd usr.sbin/vmd usr.sbin/vmmctl == distrib =========================================================== 01/07 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/distrib sets ~ lists/man/mi > sync (deraadt@) == lib =============================================================== 02/07 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/lib libc ~ sys/mount.2 > point to mount_mfs(8), not mfs(8); (jmc@) ~ asr/asr.c ~ asr/asr_private.h ~ asr/getaddrinfo_async.c ~ asr/gethostnamadr_async.c > Remove support for "lookup yp" in /etc/resolv.conf. This historical > wart is incompatible with pledge, because suddenly a "dns" operation > needs "getpw" access to ypbind/ypserv, etc. file + dns access is > enough for everyone, sorry if you were using that old SunOS 4.x style > mechanism, but it is now gone. > ok semarie millert florian (deraadt@) == sbin ============================================================== 03/07 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/sbin disklabel ~ disklabel.c ~ editor.c > Use pledge "disklabel" as needed. The theory here is these tools become > more > resistant against memory correctup, if a corrupt filesystem is given to > them. > ok krw (deraadt@) dumpfs ~ dumpfs.c > Use pledge "disklabel" as needed. The theory here is these tools become > more > resistant against memory correctup, if a corrupt filesystem is given to > them. > ok krw (deraadt@) fsck ~ fsck.c > Use pledge "disklabel" as needed. The theory here is these tools become > more > resistant against memory correctup, if a corrupt filesystem is given to > them. > ok krw (deraadt@) fsck_ext2fs ~ main.c > Use pledge "disklabel" as needed. The theory here is these tools become > more > resistant against memory correctup, if a corrupt filesystem is given to > them. > ok krw (deraadt@) ~ main.c > Cannot pledge() in the hotroot case > Discovered by krw (deraadt@) fsck_ffs ~ main.c > Use pledge "disklabel" as needed. The theory here is these tools become > more > resistant against memory correctup, if a corrupt filesystem is given to > them. > ok krw (deraadt@) ~ main.c > Cannot pledge() in the hotroot case > Discovered by krw (deraadt@) fsck_msdos ~ main.c > Use pledge "disklabel" as needed. The theory here is these tools become > more > resistant against memory correctup, if a corrupt filesystem is given to > them. > ok krw (deraadt@) fsirand ~ fsirand.8 > "random generation numbers" -> "randomly generated numbers" (mmcc@) ~ fsirand.c > Use pledge "disklabel" as needed. The theory here is these tools become > more > resistant against memory correctup, if a corrupt filesystem is given to > them. > ok krw (deraadt@) ~ fsirand.8 > Revert previous tweak, as discussed on source-changes@ (mmcc@) growfs ~ debug.c > Remove more useless comments. This whole file may go eventually, as per > deraadt@'s decree. (mmcc@) - debug.c - debug.h ~ Makefile ~ growfs.c > Remove the debugging infrastructure, which was ~1/3 of the code. Makes > pledging easier, and it was excessive complicated stuff. > "Good." deraadt@ (mmcc@) ifconfig ~ ifconfig.c > Initialize SIOCDIFADDR{_IN6,} argument as intended. > How this could have been unotified for so long? Thanks to sebastia@ > for reporting a breakage when re-configuring an IPv6 static address. > ok sebastia@ (mpi@) ~ ifconfig.c > Revert previous it introduces a regression found by krw@ (mpi@) iked ~ control.c ~ iked.h ~ ocsp.c ~ proc.c ~ types.h ~ util.c > Replace socket_set_blockmode() and fcntl(fd, F_SETFL, O_NONBLOCK) calls > with the SOCK_NONBLOCK flag to socket() and accept4(). > OK claudio@ jung@ (reyk@) mount ~ mount.c > Use pledge "disklabel" as needed. The theory here is these tools become > more > resistant against memory correctup, if a corrupt filesystem is given to > them. > ok krw (deraadt@) ncheck_ffs ~ ncheck_ffs.8 > ncheck_ffs may be invoked more simply as ncheck, > so add an entry to NAME and document it; > ok schwarze (jmc@) ~ ncheck_ffs.c > Use pledge "disklabel" as needed. The theory here is these tools become > more > resistant against memory correctup, if a corrupt filesystem is given to > them. > ok krw (deraadt@) newfs ~ mkfs.c ~ newfs.c > Use pledge "disklabel" as needed. The theory here is these tools become > more > resistant against memory correctup, if a corrupt filesystem is given to > them. > ok krw (deraadt@) newfs_ext2fs ~ newfs_ext2fs.c > Use pledge "disklabel" as needed. The theory here is these tools become > more > resistant against memory correctup, if a corrupt filesystem is given to > them. > ok krw (deraadt@) newfs_msdos ~ newfs_msdos.c > Use pledge "disklabel" as needed. The theory here is these tools become > more > resistant against memory correctup, if a corrupt filesystem is given to > them. > ok krw (deraadt@) scan_ffs ~ scan_ffs.c > Use pledge "disklabel" as needed. The theory here is these tools become > more > resistant against memory correctup, if a corrupt filesystem is given to > them. > ok krw (deraadt@) tunefs ~ tunefs.c > Use pledge "disklabel" as needed. The theory here is these tools become > more > resistant against memory correctup, if a corrupt filesystem is given to > them. > ok krw (deraadt@) == share ============================================================= 04/07 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/share man ~ man5/bsd.port.mk.5 > point to mount_mfs(8), not mfs(8); (jmc@) ~ man4/man4.macppc/zs.4 > zs(4) is now called zs, from jmc@ (mpi@) ~ man9/Makefile ~ man9/ifq_enqueue.9 + man9/ifq_deq_begin.9 > Move ifq_deq_{begin,commit,rollback} appart to keep man pages short, > readable and to emphasis that they are not the prefered API for your > drivers. > ok dlg@ (mpi@) ~ man9/ifq_enqueue.9 ~ man9/ifq_deq_begin.9 > tweaks; (jmc@) ~ man9/VOP_LOOKUP.9 ~ man9/arc4random.9 ~ man9/autoconf.9 ~ man9/bemtoh32.9 ~ man9/bus_dma.9 ~ man9/copy.9 ~ man9/crypto.9 ~ man9/disk.9 ~ man9/file.9 ~ man9/idgen32.9 ~ man9/ieee80211.9 ~ man9/ieee80211_crypto.9 ~ man9/ieee80211_node.9 ~ man9/ieee80211_output.9 ~ man9/if_rxr_init.9 ~ man9/iic.9 ~ man9/kern.9 ~ man9/mbuf.9 ~ man9/namei.9 ~ man9/pmap.9 ~ man9/pool.9 ~ man9/route.9 ~ man9/spl.9 ~ man9/systrace.9 ~ man9/timeout.9 ~ man9/uiomove.9 ~ man9/uvm.9 > add missing NAME entries; > feedback/ok schwarze (jmc@) ~ man5/resolv.conf.5 > Remove support for "lookup yp" in /etc/resolv.conf. This historical > wart is incompatible with pledge, because suddenly a "dns" operation > needs "getpw" access to ypbind/ypserv, etc. file + dns access is > enough for everyone, sorry if you were using that old SunOS 4.x style > mechanism, but it is now gone. > ok semarie millert florian (deraadt@) == sys =============================================================== 05/07 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/sys arch/amd64/amd64 ~ cpu.c ~ identcpu.c ~ ipifuncs.c ~ mainbus.c > No longer need 'option VMM', declaring the vmm0 device is sufficient. > ok mlarkin (deraadt@) arch/amd64/conf ~ GENERIC ~ files.amd64 > No longer need 'option VMM', declaring the vmm0 device is sufficient. > ok mlarkin (deraadt@) arch/amd64/include ~ cpu.h > No longer need 'option VMM', declaring the vmm0 device is sufficient. > ok mlarkin (deraadt@) arch/amd64/stand/boot ~ boot.8 > downfalls -> pitfalls (mmcc@) arch/i386/stand/boot ~ boot.8 > downfalls -> pitfalls (mmcc@) arch/macppc/conf ~ GENERIC ~ files.macppc > Attaches zs(4) as "zs" not "zsc". > Incoherency brought by jmc@ and schwarze@ because the SYNOPSIS of the > corresponding manual do not match its NAME. > This is one step further in the zs(4) unification... Go dokey go! > Tested by jmatthew@, ok sebastia@, deraadt@ (mpi@) ~ RAMDISK > sync to GENERIC (deraadt@) arch/macppc/dev ~ zs.c > Attaches zs(4) as "zs" not "zsc". > Incoherency brought by jmc@ and schwarze@ because the SYNOPSIS of the > corresponding manual do not match its NAME. > This is one step further in the zs(4) unification... Go dokey go! > Tested by jmatthew@, ok sebastia@, deraadt@ (mpi@) dev/acpi ~ acpi.c > Back out the CRITICAL change, turn it back from 10% to 25%. Some > people with old/broken batteries didn't like this change, and the > underlying heuristics are wrong. (reyk@) dev/pci ~ if_bge.c ~ if_vmx.c > Include <sys/atomic.h> when atomic operations are used. > This has been masked because <sys/srp.h> is pulled unconditionally. > ok dlg@ (mpi@) ~ if_de.c > you cant touch m_pkthdr in between ifq_deq_begin and ifq_deq_commit. > TULIP_SETCTX() does though. this moves that call after ifq_deq_commit. > problem reported (with great detail) and fix tested by ricardo mestre > ok mpi@ (dlg@) kern ~ kern_pledge.c > need sys/device.h (deraadt@) ~ kern_srp.c ~ kern_synch.c > Do not include <sys/atomic.h> inside <sys/refcnt.h>. > Prevent lazy developers, like David and I, to use atomic operations > without including <sys/atomic.h>. > ok dlg@ (mpi@) ~ kern_pledge.c > the "getpw" test for /dev/tty is only needed for readpassphrase(3), > getpass(3), so don't specifically allow it for "rpath" (rpath will > accept it in the end, unless it is on the whitelist) (deraadt@) net ~ route.c > Include <sys/atomic.h> when atomic operations are used. > This has been masked because <sys/srp.h> is pulled unconditionally. > ok dlg@ (mpi@) ~ if_spppsubr.c > if_start needs splnet. help/ok dlg@ (sthen@) ~ if_var.h ~ pf_ioctl.c > There's no longer a need to include <net/hfsc.h> in <net/if_var.h> (mpi@) netinet ~ ip_input.c > Use if_get() rather than rt_ifp. > ok sashan@ (mpi@) sys ~ refcnt.h > Do not include <sys/atomic.h> inside <sys/refcnt.h>. > Prevent lazy developers, like David and I, to use atomic operations > without including <sys/atomic.h>. > ok dlg@ (mpi@) == usr.bin =========================================================== 06/07 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/usr.bin bc ~ bc.y > - fix printing of non-ascii chars in error message > - solve a couple of problems found by Michael McConville using > afl: printing of null nodes, and generating cyclic "trees". These > can only happen on invalid input, afaik. > ok benno@ (otto@) less ~ screen.c > calling ioctl() requires include sys/ioctl.h (tedu@) + Makefile.inc > less builds its programs one level lower, so we need a Makefile.inc here > to get -Werror-implicit-function-declaration as in ../Makefile.inc (tedu@) ~ Makefile.inc > $OpenBSD$ (nicm@) ~ option.h > Fix function type arguments. (nicm@) ~ less/Makefile ~ lesskey/Makefile > Just use the default CDIAGFLAGS rather than defining our own. (nicm@) ~ command.c ~ option.c > A couple more prototypes without arguments. (nicm@) ~ funcs.h > Missing prototype. (nicm@) ~ lesskey.c > Most of these functions can be static. (nicm@) ~ cmdbuf.c > Don't shadow a function argument. (nicm@) ~ lesskey.1 > Remove bug-less email address. From Michael Reed, thanks! > While there, remove CAVEAT describing DOS and OS/2 quirks. > ok nicm@ (tb@) ~ less.1 ~ lesskey.1 > Update less(1) and lesskey(1) manuals for LESSCHARSET removal > From Michael Reed < m ! reed () mykolab ! com >, thanks! > ok nicm@ (tb@) nc ~ netcat.c > rename variable 'sun' to allow building on Solaris > ok deraadt@ (bcook@) sndiod ~ midi.c > Insert new ports at the head of the list and stop assuming its > ordered. (ratchov@) ~ dev.c > use device numbers stored in the dev structure rather than > assuming dev_list is ordered (ratchov@) ~ sndiod.c > Save and restore umask when creating /tmp/aucat/ to ensure the > directory gets the right permissions, as we do in other places. (ratchov@) ~ dev.c > If the clock is ticking because of a MMC start message, then stop > it on exit. This avoids holding an extra ref to the device, > in turn preventing sndiod from exitting cleanly. (ratchov@) ~ sndiod.c > Remove unused -M option (2+ years get used to it is enough) (ratchov@) ~ midi.c > define port_log even without -DDEBUG (ratchov@) tmux ~ input-keys.c ~ input.c ~ screen-write.c ~ server-client.c ~ tmux.h ~ tty-keys.c ~ tty.c > Remove support for the UTF-8 mouse extension. This was a briefly used, > poor idea that was fairly quickly replaced by SGR mouse input (which is > now widespread). It is impossible to tell the difference between UTF-8 > and non-UTF-8 mouse input; since the mouse-utf8 option was removed tmux > has not handled it correctly in any case; and it is ridiculous to have > three different forms of mouse input. (nicm@) ~ input-keys.c ~ input.c ~ screen-write.c ~ tmux.h > Partly revert previous, it is harmless to keep support for UTF-8 mouse > mode inside tmux, just no longer support it for tmux itself. (nicm@) vi ~ docs/USD.doc/vi.man/spell.ok ~ docs/USD.doc/vi.ref/ex.cmd.roff ~ docs/USD.doc/vi.ref/spell.ok ~ docs/USD.doc/vi.ref/vi.ref > Remove cscope references in documentation. > requested by and ok jmc@ (bentley@) == usr.sbin ========================================================== 07/07 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/usr.sbin httpd ~ proc.c > Sync proc.c with iked. (reyk@) ~ control.c ~ httpd.c ~ httpd.h ~ proc.c ~ server.c ~ server_fcgi.c > Retire socket_set_blockmode() in favor of the SOCK_NONBLOCK type flag. > As done in iked and snmpd. > OK jung@ (reyk@) kvm_mkdb ~ kvm_mkdb.c > missing pledge "getpw" for getgrnam(3) (deraadt@) relayd ~ proc.c > Sync proc.c with iked: the default proc_dispatch_null allows to remove > some empty imsg callback stubs in the privsep processes. > iked's proc.c -r1.24 was OK benno@ deraadt@ (reyk@) smtpd ~ bounce.c ~ queue.c > Restructure bounce content as a multi-part MIME message. > Content-Type header diff from Philipp Takacs <philipp<at>bureaucracy.de> > Ok gilles@ jung@ (sunil@) ~ Makefile > cleanup (gilles@) ~ ber.c > strings.h -> string.h > ok sunil@ (gilles@) ~ smtpd-api.h ~ smtpd.h ~ table.c ~ makemap/Makefile ~ smtpd/Makefile > allow table API to lookup for mailaddr mappings > ok sunil@, ok jung@ (gilles@) snmpd ~ proc.c > Sync proc.c with iked (reyk@) ~ control.c ~ proc.c ~ snmpd.c ~ snmpd.h > Use SOCK_NONBLOCK to replace socket_set_blockmode() and > fcntl(..O_NONBLOCK). > (SOCK_CLOEXEC should also be added where it is appropriate, but this is > OBnot > done in this commit yet.) > OK claudio@ (reyk@) vmd ~ Makefile > install manpages global, since two architecture (will) use them. (deraadt@) ~ Makefile ~ loadfile_elf.c ~ pci.c ~ virtio.c ~ vmd.8 ~ vmd.c ~ vmd.h + log.c > Add support for logging to stderr or syslog, and to run vmd in > foreground with -d. > OK mlarkin@ jung@ (reyk@) ~ vmd.c > I accidentally removed a newline in usage() when converting the log > messages to log_*. > From Cesare Gargano (reyk@) ~ vmd.c > accept4() is restarted after signals which prevents vmd from exiting > in the current control socket loop. Add a poll before the accept that > is not restarted and allows to escape the loop. This code is kind of > temporary, as we're planning to replace the event handling, but it > allows to kill (or Ctrl+c) vmd for now. > OK tedu@, discussed with many (reyk@) ~ Makefile > Do not need YFLAGS= (deraadt@) vmmctl ~ Makefile > install manpages global, since two architecture (will) use them. (deraadt@) ~ vmmctl.8 ~ vmm.conf.5 > Clarify the size argument, you can specify it as bytes or kilobytes, > but it is rounded to megabytes. > Pointed out by deraadt@ (reyk@) ~ vmm.conf.5 > spelling fix; from frank groeneveld (jmc@) =============================================================================== _______________________________________________ odc mailing list [email protected] http://www.squish.net/mailman/listinfo/odc
