OpenBSD src changes summary for 2015-07-26 to 2015-08-02 inclusive ==================================================================
bin/ed bin/ksh distrib/armv7 distrib/miniroot distrib/notes distrib/sets etc/etc.amd64/disktab etc/etc.armv7/Makefile.inc etc/etc.armv7/disktab etc/etc.i386/disktab etc/services games/tetris gnu lib/libc lib/libcrypto lib/libradius lib/libsndio lib/libssl libexec/ld.so libexec/spamd regress/bin regress/lib regress/sys regress/usr.bin regress/usr.sbin sbin/fsck_ext2fs sbin/fsck_ffs sbin/fsck_msdos sbin/ifconfig sbin/kbd share/man share/termtypes sys/arch/alpha/pci sys/arch/hppa/include sys/arch/m88k/m88k sys/arch/macppc/macppc sys/arch/mips64/mips64 sys/arch/powerpc/include sys/arch/powerpc/powerpc sys/arch/socppc/socppc sys/conf sys/dev sys/dev/acpi sys/dev/ic sys/dev/pci sys/dev/pv sys/kern sys/net sys/netinet sys/netinet6 sys/netmpls sys/sys usr.bin/audioctl usr.bin/doas usr.bin/file usr.bin/ipcrm usr.bin/kdump usr.bin/make usr.bin/mandoc usr.bin/mixerctl usr.bin/nc usr.bin/openssl usr.bin/passwd usr.bin/patch usr.bin/rdist usr.bin/sndiod usr.bin/ssh usr.bin/su usr.bin/tmux usr.bin/tsort usr.bin/units usr.sbin/arp usr.sbin/dhcpd usr.sbin/dvmrpctl usr.sbin/dvmrpd usr.sbin/hotplugd usr.sbin/httpd usr.sbin/ikectl usr.sbin/iscsictl usr.sbin/iscsid usr.sbin/ldpctl usr.sbin/ldpd usr.sbin/ospf6ctl usr.sbin/ospf6d usr.sbin/ospfctl usr.sbin/ospfd usr.sbin/pkg_add usr.sbin/pppd usr.sbin/radiusctl usr.sbin/radiusd usr.sbin/rcctl usr.sbin/relayd usr.sbin/ripctl usr.sbin/ripd usr.sbin/sa usr.sbin/sensorsd usr.sbin/smtpd usr.sbin/snmpctl usr.sbin/sysmerge usr.sbin/ypldap == bin =============================================================== 01/13 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/bin ed ~ buf.c ~ main.c ~ re.c > remove gratuitous casts (deraadt@) ksh ~ lex.c > Fix the baskslash-escaped codes ('\nnn') usage in PS1. > Report and initial proposal from Vadim Ushakov on bugs@, tweaked by me. > okay deraadt@ (zhuk@) == distrib =========================================================== 02/13 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/distrib armv7 ~ miniroot/Makefile ~ miniroot/Makefile.inc ~ ramdisk/install.md + miniroot/wandboard/Makefile > Make use of u-boot 2015.07's unified wandboard config to provide a > miniroot to cover all current wandboard variations. > Unlike the other imx u-boot configs the wandboard config doesn't search > for ext2 only a fat fs. > Grow the ramdisk size to fit the extra u-boot files. > From Roman Kravchuk. (jsg@) ~ ramdisk/install.md > redirect dd standard output to /dev/null (jsg@) miniroot ~ install.sub > Change installer to cope with the new sshd_config(5) default for > the PermitRootLogin option. Additionally to 'yes' and 'no' allow > 'without-password' and make that the proposed default answer for > the "Allow root ssh login?" question. Modify sshd_config only if > the user choice is not the default. > OK deraadt > discussed with halex@, sthen@ and others (rpe@) ~ install.sub > - Be more specific as to what accept as answer to the "Allow root > ssh login?" question. Either the first letter or the whole word > - Use grep to check for the default in sshd_config > - Simplify sed expression for changing sshd_config > Based on feedback from and OK halex@ (rpe@) notes ~ armv7/contents > mention the wandboard miniroot (jsg@) ~ octeon/hardware ~ octeon/install ~ octeon/prep > long overdue overhaul to include instructions for miniroot and usb > (jasper@) sets ~ lists/comp/mi > sync (deraadt@) == etc =============================================================== 03/13 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/etc etc.amd64/disktab ~ etc.amd64/disktab > Fix ba# attribute in rdroot entry. It should be 4096 (a valid size for > a FFS block), not 4480 (the size of the partition). > ok millert@ (krw@) etc.armv7/Makefile.inc ~ etc.armv7/Makefile.inc > Make use of u-boot 2015.07's unified wandboard config to provide a > miniroot to cover all current wandboard variations. > Unlike the other imx u-boot configs the wandboard config doesn't search > for ext2 only a fat fs. > Grow the ramdisk size to fit the extra u-boot files. > From Roman Kravchuk. (jsg@) etc.armv7/disktab ~ etc.armv7/disktab > Make use of u-boot 2015.07's unified wandboard config to provide a > miniroot to cover all current wandboard variations. > Unlike the other imx u-boot configs the wandboard config doesn't search > for ext2 only a fat fs. > Grow the ramdisk size to fit the extra u-boot files. > From Roman Kravchuk. (jsg@) etc.i386/disktab ~ etc.i386/disktab > Fix geometry of rdroot entry so > nt# * ns# * nc# == pc# == pa# > which makes for a better disklabel via getdiskbyname(). > ok deraadt@ (krw@) services ~ services > Add services entries for BFD (RFC5880), and reorder MS RDP while there. > From Denis Fondras, ok deraadt (sthen@) == games ============================================================= 04/13 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/games tetris ~ tetris.6 > use file system path (.Pa) semantic markup macros where appropriate. > (sobrado@) == gnu =============================================================== 05/13 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/gnu gnu ~ gcc/gcc/c-common.h ~ gcc/gcc/c-parser.c > Implement support for __builtin_complex() to construct complex values, > required by the upcoming libm work. > OK miod@. (martynas@) == lib =============================================================== 06/13 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/lib libc ~ sys/tame.2 > unsort: requested by deraadt (jmc@) ~ sys/tame.2 > describe CPATH before TMPPATH (deraadt@) ~ sys/execve.2 ~ sys/intro.2 > ARG_MAX is the standard name (not NCARGS); probab[l]y guenther (deraadt@) ~ gen/fnmatch.c > Fix a potential out-of-bounds read found via address-sanitizer while > fuzzing as reported by Elliott Hughes. OK doug@ stsp@ (millert@) ~ sys/ptrace.2 > spelling; ok deraadt (jmc@) libcrypto ~ man/BIO_s_bio.3 ~ man/BN_BLINDING_new.3 > spelling; ok deraadt (jmc@) libradius - radius.3 ~ Makefile + radius_new_request_packet.3 > Rename radius.3 to radius_new_request_packet.3 to be named after a > function. Also make all functions be listed in the NAME section and > have MLINKS in the makefile. > suggest jmc > ok deraadt (yasuoka@) ~ radius.c ~ radius_new_request_packet.3 > Add a description about radius_get_eap_msk(). (yasuoka@) ~ radius_new_request_packet.3 > some readability fixes; (jmc@) libsndio ~ sio_sun.c > Use new AUDIO_GETPOS ioctl instead of AUDIO_GETxOFFS and > AUDIO_xERROR. It returns a snapshot of the counters we need > and avoids inconsistencies. (ratchov@) ~ sndio.7 > Use "Ev" macro for AUCAT_COOKIE environment variable. From Michael > Reed <m.reed at mykolab.com>. Thanks. (ratchov@) libssl ~ src/crypto/asn1/asn1_lib.c ~ src/crypto/asn1/x_pkey.c > Less asn1_mac.h - asn1_lib.c does not need it and x_pkey.c should be > using ASN1err() instead of ASN1_MAC_H_err(). > ok miod@ (jsing@) ~ src/crypto/asn1/asn1_lib.c > Expand obsolete M_ASN1_STRING_* macros in asn1_lib.c, apply some style(9) > and drop an unnecessary return from a void function. > ok bcook@ doug@ (jsing@) ~ src/crypto/asn1/a_bitstr.c ~ src/crypto/asn1/a_int.c ~ src/crypto/asn1/a_octet.c ~ src/crypto/asn1/t_req.c ~ src/crypto/asn1/evp_asn1.c ~ src/crypto/ec/ec_asn1.c ~ src/crypto/pkcs12/p12_mutl.c ~ src/crypto/pkcs7/pk7_lib.c ~ src/crypto/pkcs7/pk7_doit.c ~ src/crypto/ts/ts_lib.c ~ src/crypto/ts/ts_rsp_utils.c ~ src/crypto/x509/x509_cmp.c ~ src/crypto/x509/x509_set.c ~ src/crypto/x509/x509cset.c ~ src/crypto/x509/x509_v3.c ~ src/crypto/x509v3/v3_prn.c ~ src/crypto/x509v3/v3_skey.c ~ src/crypto/x509v3/v3_akey.c ~ src/crypto/x509v3/v3_sxnet.c ~ src/crypto/x509v3/v3_alt.c > Expand obsolete M_ASN1.*(cmp|dup|print|set) macros - no change in generated > assembly. > ok bcook@ (jsing@) ~ src/crypto/x509v3/v3_akey.c ~ src/crypto/x509v3/v3_alt.c ~ src/crypto/x509v3/v3_bcons.c ~ src/crypto/x509v3/v3_bitst.c ~ src/crypto/x509v3/v3_cpols.c ~ src/crypto/x509v3/v3_crld.c ~ src/crypto/x509v3/v3_enum.c ~ src/crypto/x509v3/v3_extku.c ~ src/crypto/x509v3/v3_ia5.c ~ src/crypto/x509v3/v3_info.c ~ src/crypto/x509v3/v3_int.c ~ src/crypto/x509v3/v3_ncons.c ~ src/crypto/x509v3/v3_ocsp.c ~ src/crypto/x509v3/v3_pci.c ~ src/crypto/x509v3/v3_pcons.c ~ src/crypto/x509v3/v3_pku.c ~ src/crypto/x509v3/v3_pmaps.c ~ src/crypto/x509v3/v3_skey.c ~ src/crypto/x509v3/v3_sxnet.c > Use named initialisers for X509V3_EXT_METHOD structs (for the usual > reasons) - only change in generated assembly is due to line numbering. > (jsing@) ~ src/ssl/s3_clnt.c ~ src/ssl/s3_meth.c ~ src/ssl/s3_srvr.c > Add linker warnings in case SSLv3_{,client,server}_method are referenced. > Use of this symbols proves the existence of a code path willingly using > SSLv3, > even with OPENSSL_NO_SSL3 being defined, which hints that it needs fixing. > Discussed with the LibreSSL cabal during c2k15; ok deraadt@ (miod@) ~ src/ssl/d1_srtp.c > Fix SRTP parsing. > jsing@ noticed that during the CBS conversion, an extra CBS_len > comparison was introduced. It should be 0 after extracting MKI. > ok jsing@ bcook@ deraadt@ (doug@) == libexec =========================================================== 07/13 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/libexec ld.so ~ amd64/rtld_machine.c ~ i386/rtld_machine.c ~ sparc/rtld_machine.c > Copy relocations can't be jump-slot relocations, so delete the test for > that. > ok miod@ (guenther@) ~ Makefile > Reenable the ld.so pre-install test; the "am I cross-compiling" test was > too loose. Update the test to try to exercise the printf binding twice. > ok miod@ (guenther@) spamd ~ spamd.8 > use file system path (.Pa) semantic markup macros where appropriate. > ok jmc@ (sobrado@) == regress =========================================================== 08/13 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/regress bin ~ ksh/regress.t > Add regress test to exercise octal expansion (via blackslash) in > PS1 and PS2. OK deraadt@ (millert@) lib ~ libssl/bytestring/Makefile > Don't hardcode /usr/src, use BSDSRCDIR (tobiasu@) sys + kern/tame/Makefile + kern/tame/sigabrt/Makefile + kern/tame/sigabrt/sigabrt.c + kern/tame/sigabrt/sigabrt.out + kern/tame/sigkill/Makefile + kern/tame/sigkill/sigkill.c + kern/tame/sigkill/sigkill.out > add some regress for tame. just a starting point for now (semarie@) + kern/tame/sys_exit/Makefile + kern/tame/sys_exit/sys_exit.c > check that flags value of 0 restricts the process to the _exit(2) system > call. > note that it isn't the case currently. (semarie@) ~ net/pf_divert/LICENSE ~ net/pf_divert/Makefile ~ net/pf_divert/args-icmp-to.pl ~ net/pf_divert/args-rip-reply.pl ~ net/pf_divert/args-rip-to.pl ~ net/pf_divert/args-tcp-reply.pl ~ net/pf_divert/args-tcp-to.pl ~ net/pf_divert/args-udp-reply.pl ~ net/pf_divert/args-udp-to.pl ~ net/pf_divert/funcs.pl ~ net/pf_divert/remote.pl + net/pf_divert/README + net/pf_divert/args-icmp-reply-reuse.pl + net/pf_divert/args-icmp-reply-to.pl + net/pf_divert/args-rip-reply-to.pl + net/pf_divert/args-udp-reply-to.pl > Add more and deeper tests for pf divert-reply rules. Especially > the combination of sending and receiving multiple packets over one > socket is tested for UDP, raw IP and ICMP. (bluhm@) ~ kern/kqueue/kqueue-pipe.c > SPC -> TAB (uebayasi@) ~ kern/kqueue/kqueue-process.c > Refactor to prepare a future change; no functional changes. (uebayasi@) ~ kern/kqueue/kqueue-process.c > Kill a useless assignment. (uebayasi@) usr.bin + doas/Makefile + doas/t-fail-quotes.conf + doas/t-fail-quotes.err + doas/t-fail-quotes.expected.err + doas/t-fail-quotes.out + doas/t-okay.conf + doas/t-okay.err + doas/t-okay.out + doas/t-permit-1.conf + doas/t-permit-1.patterns > Import test suite for doas. Relies on recently committed command matching > functionlity with -C flag. (zhuk@) ~ Makefile > Hook up doas tests. (zhuk@) ~ doas/Makefile > Make test names printed like other test suites do. (zhuk@) ~ doas/Makefile + doas/master.passwd + doas/t-run-keepenv-path.conf > Add (currently failing) test for passing PATH via keepenv in doas. > Initial report by Ze Loff on misc@. (zhuk@) usr.sbin ~ syslogd/args-dropped-tcp.pl ~ syslogd/args-dropped-tls.pl ~ syslogd/args-sync-tcp.pl > Make syslogd tests more reliable. Allow TLS tests to drop a wider > rage of messages and make dropped regexes more specific. (bluhm@) ~ syslogd/args-client-bind-only4.pl ~ syslogd/args-client-bind-only6.pl ~ syslogd/args-client-bind-port.pl ~ syslogd/args-client-bind.pl ~ syslogd/args-client-bind4-port.pl ~ syslogd/args-client-bind4.pl ~ syslogd/args-client-bind6-port.pl ~ syslogd/args-client-bind6.pl ~ syslogd/args-client-tcp-maxline.pl ~ syslogd/args-client-tcp-multichunks.pl ~ syslogd/args-client-tcp-multilines.pl ~ syslogd/args-client-tcp-nontransp-maxline.pl ~ syslogd/args-client-tcp-nontransp.pl ~ syslogd/args-client-tcp-octet-badframe.pl ~ syslogd/args-client-tcp-octet-maxline.pl ~ syslogd/args-client-tcp-octet-nontransp.pl ~ syslogd/args-client-tcp-octet.pl ~ syslogd/args-client-tcp4.pl ~ syslogd/args-client-tcp6.pl ~ syslogd/args-fdexhaustion-tcp.pl ~ syslogd/args-rsyslog-client-tcp.pl ~ syslogd/args-rsyslog-client-udp.pl ~ syslogd/args-rsyslog-tcp.pl ~ syslogd/args-rsyslog-tls.pl ~ syslogd/args-rsyslog-udp.pl > Use Socket to make Perl syntax check of single test files pass. (bluhm@) == sbin ============================================================== 09/13 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/sbin fsck_ext2fs ~ fsck_ext2fs.8 > document "F" in interactive mode, like for fsck.8; > from michael mcconville > ok deraadt (jmc@) fsck_ffs ~ fsck_ffs.8 > document "F" in interactive mode, like for fsck.8; > from michael mcconville > ok deraadt (jmc@) fsck_msdos ~ fsck_msdos.8 > document "F" in interactive mode, like for fsck.8; > from michael mcconville > ok deraadt (jmc@) ifconfig ~ ifconfig.c > remove unused variable > ok mpi@ rzalamena@ (chl@) ~ ifconfig.8 > Document acceptable values for "carpdemote", from Julien Dhaille. > ok jmc@, deraadt@ (mpi@) kbd ~ kbd.8 > use file system path (.Pa) semantic markup macros where appropriate. > ok jmc@ (sobrado@) == share ============================================================= 10/13 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/share man ~ man4/audio.4 > Remove sections that describe features we don't support any longer. > (ratchov@) ~ man5/mixerctl.conf.5 ~ man5/mk.conf.5 ~ man8/diskless.8 > use file system path (.Pa) semantic markup macros where appropriate. > ok jmc@ (sobrado@) ~ man4/ppp.4 > Add a note to ppp(4)'s BUGS section where it talks about lack of IPv6 > support to make it clear that pppoe(4) does work with IPv6. Problem noted > by Simon Farnsworth. Add xrefs to pppoe(4) and pppx(4) to show that this > isn't the only PPP implementation on OpenBSD. ok/tweak stsp@ (sthen@) ~ man9/km_alloc.9 > kv_executable was removed (mikeb@) ~ man4/audio.4 > Add the AUDIO_GETPOS ioctl() to fetch a snapshot of the 4 counters > returned by AUDIO_GETxOFFS and AUDIO_xERROR ioctls. > ok deraadt (ratchov@) ~ man8/afterboot.8 ~ man8/release.8 ~ man5/bsd.port.mk.5 ~ man5/bsd.regress.mk.5 ~ man5/mk.conf.5 ~ man7/hier.7 ~ man7/ports.7 > switch references from sudo to doas; > ok deraadt (jmc@) ~ man7/hier.7 > two revisions ago i bungled the addition of radiusd - > fix that now; > ok deraadt (jmc@) ~ man3/intro.3 > update the radius reference; (jmc@) ~ man8/release.8 > need to use env here. from Mark Patruck (tedu@) termtypes ~ termtypes.master > Update to terminfo-20150725. Also add a note at the start about our > (two) local changes. (nicm@) ~ termtypes.master > Whoops, "screen" we also change kbs=^H to kbs=\177 now (this is why we > need a list of changes!) (nicm@) == sys =============================================================== 11/13 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/sys arch/alpha/pci ~ pci_1000.c ~ pci_1000a.c ~ pci_2100_a50.c ~ pci_550.c ~ pci_6600.c ~ pci_alphabook1.c ~ pci_axppci_33.c ~ pci_eb164.c ~ pci_eb64plus.c ~ pci_kn20aa.c ~ pci_kn300.c ~ pci_machdep.c ~ pci_machdep.h ~ pci_up1000.c > Make sure pci_intr_map() will perform proper interrupt swizzling for > devices > behind a bridge, if the SRM didn't pick an interrupt line for them. (miod@) arch/hppa/include ~ pmap.h > Always #include <sys/mutex.h>: need struct mutex for struct vm_page_md > problem noted by landry@ > ok dlg@ (guenther@) arch/m88k/m88k ~ pmap.c > Acquire the kernel lock in pmap_remove(). The reasons for this can't be > stated here as I have been asked to be polite in this commit message. > ok deraadt@ (miod@) arch/macppc/macppc ~ locore.S > 17 years ago, setfault() was modified to save the status register in the > faultbuf. But 1/ sr was only restored for machine check exceptions, and 2/ > the > way it was saved was unsafe if interrupts were enabled, and could cause %r2 > to be lost. > Discussing this with deraadt@ at the end of c2k15, this was probably needed > for the old VI boards which were the target of the original powerpc port, > came with a worse-than-Genesi openfirmware. Since then, machine check > exceptions have been unheard of; or, if they happen, they do not need the > status register to be restored. > ok mpi@ deraadt@ (miod@) arch/mips64/mips64 ~ ipifuncs.c > Allow coalescing of IPI requests on mips64, to make IPI sending > non-blocking. This improves performance on systems that have more than > a few cores and where there is a heavy IPI load. Currently the only > place where coalescing must not happen is rendezvous processing, but > there invocations are serialized by the rendezvous mutex. > ok miod@ (visa@) ~ pmap.c > Make mips64 pmap prepared for the unlocked reaper. > ok miod@, deraadt@ (visa@) arch/powerpc/include ~ pcb.h > 17 years ago, setfault() was modified to save the status register in the > faultbuf. But 1/ sr was only restored for machine check exceptions, and 2/ > the > way it was saved was unsafe if interrupts were enabled, and could cause %r2 > to be lost. > Discussing this with deraadt@ at the end of c2k15, this was probably needed > for the old VI boards which were the target of the original powerpc port, > came with a worse-than-Genesi openfirmware. Since then, machine check > exceptions have been unheard of; or, if they happen, they do not need the > status register to be restored. > ok mpi@ deraadt@ (miod@) arch/powerpc/powerpc ~ trap.c > 17 years ago, setfault() was modified to save the status register in the > faultbuf. But 1/ sr was only restored for machine check exceptions, and 2/ > the > way it was saved was unsafe if interrupts were enabled, and could cause %r2 > to be lost. > Discussing this with deraadt@ at the end of c2k15, this was probably needed > for the old VI boards which were the target of the original powerpc port, > came with a worse-than-Genesi openfirmware. Since then, machine check > exceptions have been unheard of; or, if they happen, they do not need the > status register to be restored. > ok mpi@ deraadt@ (miod@) arch/socppc/socppc ~ locore.S > 17 years ago, setfault() was modified to save the status register in the > faultbuf. But 1/ sr was only restored for machine check exceptions, and 2/ > the > way it was saved was unsafe if interrupts were enabled, and could cause %r2 > to be lost. > Discussing this with deraadt@ at the end of c2k15, this was probably needed > for the old VI boards which were the target of the original powerpc port, > came with a worse-than-Genesi openfirmware. Since then, machine check > exceptions have been unheard of; or, if they happen, they do not need the > status register to be restored. > ok mpi@ deraadt@ (miod@) conf ~ GENERIC > As usual, turn off POOL_DEBUG for the release. > ok deraadt@ (benno@) dev ~ softraid.c ~ softraid_crypto.c > zap trailing linefeeds from sr_error() and panic() calls > ok jsing@ (halex@) ~ softraid.c > Tweak a couple of [SLIST|TAILQ]_REMOVE() usages in loops to a more > obvious idiom. > ok bluhm@ jsing@ (krw@) ~ audio.c > Add the AUDIO_GETPOS ioctl() to fetch a snapshot of the 4 counters > returned by AUDIO_GETxOFFS and AUDIO_xERROR ioctls. > ok deraadt (ratchov@) ~ audio.c > In case the system misses enough audio interrupts for DMA > pointers to wrap, recover by detecting and compensating > for the missed interrupts. Fixes certain audio hangs on > MP machines. > with help from armani@, typos fixed by Alexey Suslikov (ratchov@) ~ audio.c > add missing space, from Donovan Watteau <tsoomi at gmail.com> > thanks! (ratchov@) ~ softraid.c > Zap a SLIST in a more obviously correct/safe way. > Originally from pelikan@, recent prodding from bluhm@ and jsg@. > ok pelikan@ millert@ bluhm@ (krw@) ~ audio.c > Fix spacing and be more verbose in printfs under #ifdef AUDIO_DEBUG. > No binary change. (ratchov@) dev/acpi ~ acpi.c > Something subtle in previous revision is locking up suspend or booting > on some laptops. Mailing lists threads mention x220 and Asus machines. > Reverting for now. (deraadt@) dev/ic ~ i82365.c > Make pcic_ident_ok() completely silent in case of failure. (miod@) ~ vga.c ~ vgavar.h > When attaching on an alpha system, get the current textmode resolution by > querying the 6845 registers, and preserve it if it is larger than 80x25 yet > plausible. > This is necessary to properly operate on the Alphabook 1 where the vga > display > is running in a 100x37 textmode. > Other platforms are not affected. (miod@) dev/pci ~ azalia.c > Rework buffer position reporting code. > - use simpler data structures (makes the code shorter). > - if multiple interrupts are missed, call audio(4) as many times as we > missed interrupts (unless the pointer wrapped, which this is > undetectable, but upper layer handles it) > - fetch the link fifos size in the interrupt handler, as, according to > the spec it may change dynamically. > - certain devices expose the last offset of the link fifo size instead > of the fifo size itself; workaround this by incrementing the > reported link fifo size if it's odd. Fixes recovery after missed > interrupts on asus eeepcs, maybe others. > ok deraadt, help from armani@ (ratchov@) ~ envy.c ~ envyvar.h > Use DMA pointer determine the number of times the audio(4) layer has > to advance in time. This is needed to properly recover, when > interrupts are blocked for too long. Fixes permanent distortion on MP > systems. > ok deraadt (ratchov@) dev/pv ~ pvbus.c ~ pvreg.h ~ pvvar.h ~ vmt.c > Add callbacks for supported hypervisors in pvbus(4) to print more > information and for some generic CPUID/MSR-level initialization. This > changes the attach/match path a bit. > Hypervisor informatiom may include a version number, for example: > pvbus0 at mainbus0: Hyper-V 6.3.9600 > OK mlarkin@ (reyk@) ~ pvbus.c ~ pvreg.h > Determine and print Xen hypervisor version. > ok reyk, with deraadt@'s permission (mikeb@) kern ~ subr_disk.c > Always initialize the b_error field and B_ERROR flag before using > a buf. Otherwise stale data might cause a successful I/O to be > seen as a failed I/O. > ok tedu@ (krw@) ~ uipc_syscalls.c > tame check of msghdr should be done after checking for copyin failure > ok deraadt@ (guenther@) ~ kern_tame.c > enforce tame by disabling all TAME_ flags if tame_fail() is reach, and not > only > if TAME_ABORT is setted. > ok deraadt@ (semarie@) ~ kern_exec.c TAGGED OPENBSD_5_7 > Index: sys/kern/kern_exec.c > =================================================================== > RCS file: /cvs/src/sys/kern/kern_exec.c,v > retrieving revision 1.160 > diff -u -p -r1.160 kern_exec.c > --- sys/kern/kern_exec.c 9 Feb 2015 11:52:47 -0000 1.160 > +++ sys/kern/kern_exec.c 27 Jul 2015 00:23:35 -0000 > @@ -785,12 +785,12 @@ exec_abort: > free_pack_abort: > free(pack.ep_hdr, M_EXEC, 0); > + if (pathbuf != NULL) > + pool_put(&namei_pool, pathbuf); > exit1(p, W_EXITCODE(0, SIGABRT), EXIT_NORMAL); > /* NOTREACHED */ > atomic_clearbits_int(&pr->ps_flags, PS_INEXEC); > - if (pathbuf != NULL) > - pool_put(&namei_pool, pathbuf); > return (0); > } (krw@) ~ kern_exec.c TAGGED OPENBSD_5_6 > Backport rev 1.163 by deraadt: > memory leak in execve with systrace, plus some unreachable code > spotted by Maxime Villard > ok guenther millert (guenther@) ~ kern_sig.c ~ kern_tame.c TAGGED OPENBSD_5_6 > Rather than disabling tame to coredump, leave it enabled but flag that > a coredump is happening. This improves behaviour while threaded. > ok semarie (deraadt@) ~ sys_generic.c ~ uipc_syscalls.c TAGGED OPENBSD_5_6 > Add ktracing of structs iovec, msghdr, and cmsghdr for {,p}{read,write}v(), > sendmsg(), and recvmsg(). For cmsghdr, the len, level, and type are always > shown, and for SOL_SOCKET,SCM_RIGHTS the fd numbers being passed are shown. > ok millert@ deraadt@ (guenther@) ~ uipc_usrreq.c TAGGED OPENBSD_5_6 > For unix domain sequenced packet socket pairs the ruby regression > tests reported an EMSGSIZE error although the sent message was not > too large. The way backpressure was implemented for unix domain > sockets confused the check in sosend(). > Unix domain sockets append data only to the recv buffer. To report > the amount of content to the sender, the high watermark of the send > buffer was reduced. This happend for SOCK_STREAM and SOCK_SEQPACKET. > Sosend checks wether atomic chunks could ever fit into the send > buffer which is limited by the high watermark. This happens for > SOCK_DGRAM and SOCK_SEQPACKET. For SOCK_SEQPACKET the combination > of these mechanisms resulted in an EMSGSIZE error when the buffer > got filled. This also happened when space could be created by > reading from the other end in contradiction to the semantics of > EMSGSIZE. > Do not emulate a send buffer that has no space. It is better to > fill the buffer with fake data than to reduce its size. Thus the > high watermark always contains the real value. When disconnecting, > reset the counters. Otherwise the socket layer would try to flush > non existing data in the send buffer. > Tested by jeremy@ with a C program and the ruby tests. > OK markus@ jeremy@ (bluhm@) ~ kern_tame.c TAGGED OPENBSD_5_6 > normalize disclaimer to license.template; noted by reyk (deraadt@) ~ subr_disk.c TAGGED OPENBSD_5_6 > Convert sectors to DEV_BSIZE block values were necessary to find, > read and validate disklabels on GPT partitioned disks. Makes it much > more likely GPT partitioned disks with non-512-byte sectors will > work. > Part of the larger reverted GPT work. > ok deraadt@ (krw@) ~ subr_disk.c TAGGED OPENBSD_5_6 > Zap a couple of extra blank lines that keep showing up in diffs I'm > trying to build. (krw@) ~ kern_tame.c TAGGED OPENBSD_5_6 > Neuter tame(2) with ENOSYS for upcoming release cycle, in a fairly > simple and gross fashion, but the least impact on other parts... (deraadt@) ~ kern_ktrace.c TAGGED OPENBSD_5_6 > Fix free() of uninitialized variable introduced in previous commit. > Eliminate the goto that I tripped on. > problem noted by Mark Latimer (mark.latimer (at) gmail.com) > ok miod@ millert@ (guenther@) net ~ if.c ~ if_ethersubr.c ~ if_loop.c > Don't use mpls_input() as input handler anymore and instead call it > directly. Also protect non mp-safe functions while at it. > ok mpi@. (rzalamena@) netinet ~ raw_ip.c > Implement pf divert-reply for raw sockets. Note that an empty > divert->addr means that the rule has a divert-reply option. With > divert-to the divert->addr is always set. The divert-reply rules > should compare the original packet with the socket addresses. So > skip the step that compares the inp->inp_laddr with the empty > divert->addr. > OK mikeb@ (bluhm@) ~ ip_ether.c ~ ip_gre.c > Don't use mpls_input() as input handler anymore and instead call it > directly. Also protect non mp-safe functions while at it. > ok mpi@. (rzalamena@) ~ ip_ether.c > Removed uninitialized pointer usage, we didn't need to use it there anyway. > Thanks to Patrick Wild for noticing that! > ok mpi@, deraadt@. (rzalamena@) netinet6 ~ raw_ip6.c > iDo not link an ICMP6 socket to the pf state. When multiple ICMP > packets with a different ICMP ID are sent over the same raw socket, > multiple states should be created. Put a similar check into > rip6_output() like in the IPv4 case. > OK mikeb@ (bluhm@) ~ icmp6.c ~ raw_ip6.c > Implement pf divert-reply for raw sockets. Note that an empty > divert->addr means that the rule has a divert-reply option. With > divert-to the divert->addr is always set. The divert-reply rules > should compare the original packet with the socket addresses. So > skip the step that compares the inp->inp_laddr with the empty > divert->addr. > OK mikeb@ (bluhm@) netmpls ~ mpls.h ~ mpls_input.c > Don't use mpls_input() as input handler anymore and instead call it > directly. Also protect non mp-safe functions while at it. > ok mpi@. (rzalamena@) sys ~ tame.h > typo (deraadt@) ~ proc.h > Rather than disabling tame to coredump, leave it enabled but flag that > a coredump is happening. This improves behaviour while threaded. > ok semarie (deraadt@) ~ ktrace.h > Add ktracing of structs iovec, msghdr, and cmsghdr for {,p}{read,write}v(), > sendmsg(), and recvmsg(). For cmsghdr, the len, level, and type are always > shown, and for SOL_SOCKET,SCM_RIGHTS the fd numbers being passed are shown. > ok millert@ deraadt@ (guenther@) ~ tame.h > normalize disclaimer to license.template; noted by reyk (deraadt@) ~ audioio.h > Add the AUDIO_GETPOS ioctl() to fetch a snapshot of the 4 counters > returned by AUDIO_GETxOFFS and AUDIO_xERROR ioctls. > ok deraadt (ratchov@) == usr.bin =========================================================== 12/13 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/usr.bin audioctl ~ audioctl.c > Use new AUDIO_GETPOS ioctl() to ensure "bytes" and "errors" counters > are consistent. (ratchov@) doas ~ doas.1 ~ doas.c > Implement command matching without execution. This just extends > functionality of the -C flag, so we are not introducing more garbage. > Input and okay from jmc@ (documentation) and tedu@ (everything). (zhuk@) ~ doas.c > Stop exiting on cmdline overflow: it's used only for logging, so aborting > the whole process is stupid, and actually breaks things. > Noticed and analyzed by as well as input from nigel@. > Okay tedu@, espie@ and (if I understood correctly) hall@ (zhuk@) ~ doas.c > tweak config checking slightly (tedu@) ~ doas.c > Oops, CVS mismerged changes, resulting in compilable and mostly working, > but somewhat wrong code. Well, the CVS mismerged but I just missed. (zhuk@) ~ doas.1 ~ doas.c > nflag (as in sudo, force non-interactive mode) > as discussed with ted@ (espie@) ~ doas.c > checkconfig doesn't return anymore, noted by zhuk (tedu@) ~ doas.1 > small clarifications (tedu@) ~ doas.c > Add -n to usage. As noticed by Theo Buehler. (espie@) ~ doas.1 ~ doas.conf.5 > some improvements from michael reed; (jmc@) ~ doas.conf.5 > default permitted target is all users, not root. (tedu@) ~ doas.c > Fix keepenv handling. Initially reported by Ze Loff on misc@. > tedu@ agrees. (zhuk@) ~ doas.c > Rename some variables and add few comments in keepenv handling code. > Makes the code more readable. > okay tedu@ (zhuk@) ~ doas.c ~ parse.y > wrap some exceedingly long lines (deraadt@) ~ doas.c > refine a comment (tedu@) ~ doas.conf.5 > Fix usage examples. > From Theo Buehler on misc@, thanks! > okay jmc@, deraadt@. (zhuk@) ~ doas.c > make gid parsing look like uid parsing. from Martijn van Duren > ack deraadt (tedu@) file ~ file.1 > note that symlinks are not followed by default; > from tilo stritzky > ok nicm and discussed with deraadt (jmc@) ipcrm ~ ipcrm.c > fix format string > ok deraadt@ (chl@) kdump ~ ktrstruct.c > Add ktracing of structs iovec, msghdr, and cmsghdr for {,p}{read,write}v(), > sendmsg(), and recvmsg(). For cmsghdr, the len, level, and type are always > shown, and for SOL_SOCKET,SCM_RIGHTS the fd numbers being passed are shown. > ok millert@ deraadt@ (guenther@) make ~ job.c > let the hack that lets make kill jobs after sudo also work with doas. > okay tedu@, millert@, tweaks by millert@ (espie@) mandoc ~ main.c ~ tag.c ~ tag.h > Remove the hack of scrolling forward and backward with +G1G that > many (jmc@, millert@, espie@, deraadt@) considered revolting. > Instead, when using a pager, since we are using a temporary file > for tags anyway, use another temporary file for the formatted > page(s), as suggested by millert@ and similar to what the traditional > BSD man(1) did, except that we use only one single temporary output > file rather than one for each formatted manual page, such that > searching (both with / and :t) works across all the displayed files. > (schwarze@) mixerctl ~ mixerctl.1 > use file system path (.Pa) semantic markup macros where appropriate. > ok jmc@ (sobrado@) nc ~ netcat.c > remove unused variable > ok tedu@ (chl@) openssl ~ openssl.1 > use file system path (.Pa) semantic markup macros where appropriate. > ok jmc@ (sobrado@) ~ openssl.1 > remove ssl3 bits; ok doug (jmc@) passwd ~ passwd.1 > use file system path (.Pa) semantic markup macros where appropriate. > ok jmc@ (sobrado@) patch ~ common.h ~ inp.c ~ patch.1 ~ pch.c ~ util.c ~ util.h > Remove support for automatically checking files out of RCS. The > behavior is confusing to users and even GNU patch no longer does > it by default. OK krw@ guenther@ (millert@) ~ common.h ~ inp.c ~ patch.1 ~ pch.c ~ util.c ~ util.h TAGGED OPENBSD_5_6 > Backport commit of 2014-11-22 15:49 by tobias: > Remove SCCS support. For this to work, we would need "get", which we don't > even have in ports. XSI functionality of POSIX, therefore optional. > ok deraadt, schwarze, tedu > ---------------------------- > Backport commit of 2015-07-26 14:32 by millert > Remove support for automatically checking files out of RCS. The > behavior is confusing to users and even GNU patch no longer does > it by default. OK krw@ guenther@ (guenther@) ~ common.h ~ inp.c ~ patch.1 ~ pch.c ~ util.c ~ util.h TAGGED OPENBSD_5_7 > Backport commit of 2015-07-26 14:32 by millert > Remove support for automatically checking files out of RCS. The > behavior is confusing to users and even GNU patch no longer does > it by default. OK krw@ guenther@ (krw@) ~ pch.c TAGGED OPENBSD_5_7 > Account for newlines in substitution (s///) commands. Substitution > commands might contain a newline in the replacement pattern (escaped > with a backslash before it), causing patch's understanding of the > state the ed child process is in to diverge from reality. This can > lead to patch unwillingly feeding '!' (execute shell command) lines > to ed. From Martin Natano. OK deraadt@ (millert@) ~ pch.c TAGGED OPENBSD_5_6 > Backport rev 1.53 by millert: > Account for newlines in substitution (s///) commands. Substitution > commands might contain a newline in the replacement pattern (escaped > with a backslash before it), causing patch's understanding of the > state the ed child process is in to diverge from reality. This can > lead to patch unwillingly feeding '!' (execute shell command) lines > to ed. From Martin Natano. OK deraadt@ (krw@) ~ pch.c TAGGED OPENBSD_5_7 > Backport rev 1.53 by millert: > Account for newlines in substitution (s///) commands. Substitution > commands might contain a newline in the replacement pattern (escaped > with a backslash before it), causing patch's understanding of the > state the ed child process is in to diverge from reality. This can > lead to patch unwillingly feeding '!' (execute shell command) lines > to ed. From Martin Natano. OK deraadt@ (krw@) rdist ~ rdist.1 > use file system path (.Pa) semantic markup macros where appropriate. > ok jmc@ (sobrado@) sndiod ~ file.c ~ file.h > Move processing of polled files in its own function, and call it > twice: once for files that need immediate handling, and once for files > that would block. No behavior change. (ratchov@) ssh ~ sandbox-systrace.c > Permit kbind(2) use in the sandbox now, to ease testing of ld.so work using > it > reminded by miod@, ok deraadt@ (guenther@) ~ compat.c > add Cisco to the list of clients that choke on the hostkeys update > extension. Pointed out by Howard Kash (djm@) ~ kex.c ~ kex.h ~ packet.c > include the peer's offer when logging a failure to negotiate a > mutual set of algorithms (kex, pubkey, ciphers, etc.) > ok markus@ (djm@) ~ kex.c > fix bug in previous; was printing incorrect string for failed > host key algorithms negotiation (djm@) ~ kex.c ~ kex.h ~ readconf.c ~ servconf.c ~ ssh.c ~ ssh_config.5 ~ sshconnect2.c ~ sshd.c ~ sshd_config.5 > Allow ssh_config and sshd_config kex parameters options be prefixed > by a '+' to indicate that the specified items be appended to the > default rather than replacing it. > approach suggested by dtucker@, feedback dlg@, ok markus@ (djm@) ~ servconf.c ~ sshd_config ~ sshd_config.5 > change default: PermitRootLogin without-password > matching install script changes coming as well > ok djm markus (deraadt@) ~ sshpty.c > fix pty permissions; patch from Nikolay Edigaryev; ok deraadt (djm@) ~ servconf.c > Allow PermitRootLogin to be overridden by config > ok markus@ deeradt@ (chris@) ~ version.h > openssh 7.0; ok deraadt@ (djm@) su ~ su.1 > switch references from sudo to doas; > ok deraadt (jmc@) tmux ~ cmd-set-option.c ~ tmux.1 > Make -q suppress ambiguous option warnings too, from Cam Hutchison. (nicm@) ~ cmd-show-messages.c ~ status.c ~ tmux.h ~ tty-term.c > Tidy up the way terminals are described and move some structs out of > tmux.h. (nicm@) ~ server-client.c ~ status.c ~ tmux.h > status_out and associated data structures are no longer used. (nicm@) tsort ~ tsort.c > gc macro that's no longer used since the move to reallocarray (espie@) units ~ units.lib > update currency exchange rates; (jmc@) == usr.sbin ========================================================== 13/13 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/usr.sbin arp ~ arp.8 > use file system path (.Pa) semantic markup macros where appropriate. > ok jmc@ (sobrado@) dhcpd ~ dhcpd.leases.5 > use file system path (.Pa) semantic markup macros where appropriate. > (sobrado@) dvmrpctl ~ dvmrpctl.8 > use file system path (.Pa) semantic markup macros where appropriate. > (sobrado@) dvmrpd ~ dvmrpd.8 > use file system path (.Pa) semantic markup macros where appropriate. > ok jmc@ (sobrado@) hotplugd ~ hotplugd.8 > use file system path (.Pa) semantic markup macros where appropriate. > ok jmc@ (sobrado@) httpd ~ server_fcgi.c > add HSTS to fcgi responses > OK reyk (florian@) ~ httpd.h ~ server.c ~ server_fcgi.c ~ server_http.c > Read fcgi response records until we have the whole http header and can > parse it. Otherwise http headers can leak into the body. > Pointed out by Jean-Philippe Ouellet on bugs@ Thanks! > OK reyk, commit ASAP deraadt@ (florian@) ~ httpd.h ~ server.c ~ server_fcgi.c ~ server_http.c > backout the previous: it broke wordpress somehow. > we need more care to find a proper fix for the fastcgi headers. > acknowledged by deraadt@ (reyk@) ~ httpd.h ~ server_fcgi.c ~ server_http.c > repair hsts header output, wrong format strings caused broken > Strict-Transport-Security headers. Add __format__ attribute to > kv_set() and kv_setkey() to make it easier to spot such problems. > Found by and fix from Donovan Watteau <tsoomi -AT- gmail -DOT- com>, > thanks for your help. > ok deraadt@ (benno@) ikectl ~ ikectl.8 > use file system path (.Pa) semantic markup macros where appropriate. > ok jmc@ (sobrado@) iscsictl ~ iscsictl.8 > use file system path (.Pa) semantic markup macros where appropriate. > ok jmc@ (sobrado@) iscsid ~ iscsid.8 > use file system path (.Pa) semantic markup macros where appropriate. > ok jmc@ (sobrado@) ldpctl ~ ldpctl.8 > use file system path (.Pa) semantic markup macros where appropriate. > (sobrado@) ldpd ~ ldpd.8 > use file system path (.Pa) semantic markup macros where appropriate. > ok jmc@ (sobrado@) ospf6ctl ~ ospf6ctl.8 > use file system path (.Pa) semantic markup macros where appropriate. > (sobrado@) ospf6d ~ ospf6d.8 > use file system path (.Pa) semantic markup macros where appropriate. > ok jmc@ (sobrado@) ospfctl ~ ospfctl.8 > use file system path (.Pa) semantic markup macros where appropriate. > (sobrado@) ospfd ~ ospfd.8 > use file system path (.Pa) semantic markup macros where appropriate. > ok jmc@ (sobrado@) pkg_add ~ OpenBSD/Delete.pm > make -B with bad checksums cope. Reported by millert@ (espie@) pppd ~ pppd.8 > use file system path (.Pa) semantic markup macros where appropriate. > ok jmc@ (sobrado@) radiusctl ~ radiusctl.c > Fix styles. Also delete -h option since any other ctl command doesn't have > it. (yasuoka@) radiusd ~ radiusd.c > Use log_warn() instead of warn() in radiusd_module_load(). Also fix style. > (yasuoka@) ~ radiusd.h ~ radiusd_bsdauth.c ~ radiusd_local.h ~ radiusd_module.c ~ radiusd_module.h ~ radiusd_radius.c > Drop the privilege from modules. "radiusd_radius" could simply run > without root. "radiusd_bsdauth" uses some functions which needs root. > So separate its process into a privileged process and a non-privileged > process. (yasuoka@) ~ radiusd_bsdauth.c ~ radiusd_radius.c > Stop using _this in radiusd modules. (yasuoka@) ~ radiusd.8 > add a FILES section to make clear where RADIUS configuration file > is placed by default. > ok yasuoka@ (sobrado@) ~ radiusd.conf.5 > add a FILES section to make clear where RADIUS configuration file > is placed by default. > ok yasuoka@ (sobrado@) ~ radiusd_radius.c > q_id didn't pass to radiusd_radius properly. It should be u_int not u_int. > (yasuoka@) ~ radiusd_bsdauth.c > Fix typo in comment. (yasuoka@) ~ radiusd_radius.c > Fix the variable name for struct radius_server. It was confused. > (yasuoka@) ~ radiusd_radius.c > Use the dedicated log function instead of log_warn(). Also, fix the > variable name again. (yasuoka@) ~ radiusd.c > Fix radiusd to start without -d. Also stop using event_initialized() > to check whether the event handler is set. (yasuoka@) ~ radiusd_radius.c > Fix radiusd_radius to use syslog(3). It had used stderr for debug. > (yasuoka@) ~ radiusd_module.c > radiusd_module with libevent didn't stop when the daemon stops. Fix > it to delete its event handler properly. (yasuoka@) ~ parse.y ~ radiusd.c > Tweak XXX comments. (yasuoka@) ~ radiusd.c > Fire pending events when the module starts. (yasuoka@) ~ radiusd_module.c > Check the received packet length properly. (yasuoka@) rcctl ~ rcctl.sh > Tweak needs_root() output. > When using "ls" with a root action, properly output the action and flags > in the error output; from Theo Buehler (ajacoutot@) relayd ~ relayd.8 > use file system path (.Pa) semantic markup macros where appropriate. > ok jmc@ (sobrado@) ~ relay.c ~ relay_http.c > spacing (reyk@) ~ relayd.c > fix bug where other than the last of multiple forward rules in http > protocols would be ignored, reported and fixed by J. Fischer (lists > -AT- mistrust -DOT- net) and reminded by (trondd -AT- kagu-tsuchi -DOT- > com), thanks! > ok deraadt@ (benno@) ripctl ~ ripctl.8 > use file system path (.Pa) semantic markup macros where appropriate. > (sobrado@) ripd ~ ripd.8 > use file system path (.Pa) semantic markup macros where appropriate. > ok jmc@ (sobrado@) sa ~ main.c > fix format string > ok deraadt@ (chl@) sensorsd ~ sensorsd.8 ~ sensorsd.conf.5 > use file system path (.Pa) semantic markup macros where appropriate. > ok jmc@ (sobrado@) smtpd ~ smtpctl.8 > use file system path (.Pa) semantic markup macros where appropriate. > (sobrado@) snmpctl ~ snmpctl.8 > use file system path (.Pa) semantic markup macros where appropriate. > (sobrado@) sysmerge ~ sysmerge.8 > use file system path (.Pa) semantic markup macros where appropriate. > ok jmc@ (sobrado@) ypldap ~ ypldap.8 > use file system path (.Pa) semantic markup macros where appropriate. > ok jmc@ (sobrado@) =============================================================================== _______________________________________________ owc mailing list [email protected] http://www.squish.net/mailman/listinfo/owc
