OpenBSD src changes summary for 2016-09-06 ==========================================
distrib/sets gnu lib/libc libexec/spamd regress/lib regress/sys share/man sys/arch/mips64/include sys/arch/mips64/mips64 sys/conf sys/kern sys/netinet sys/netinet6 sys/sys usr.bin/mg usr.bin/ssh usr.sbin/acme-client usr.sbin/pkg_add usr.sbin/smtpd usr.sbin/sysmerge usr.sbin/syspatch usr.sbin/unbound == distrib =========================================================== 01/09 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/distrib sets ~ lists/comp/mi > sync (sthen@) == gnu =============================================================== 02/09 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/gnu gnu ~ llvm/tools/clang/lib/Driver/Tools.cpp > Link to -lcompiler_rt instead of -lgcc. > looks good to stefan@ (pascal@) == lib =============================================================== 03/09 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/lib libc ~ arch/amd64/SYS.h > Remove branch prediction hints from conditional branch instructions. These > hints are not recognized by clang's builtin assembler and the opcode > prefixes > they generate have been no-ops for all CPUs after the Pentium 4. > ok guenther@ (kettenis@) ~ dlfcn/dlfcn_stubs.c ~ dlfcn/init.c + dlfcn/init.h > Implement dl_iterate_phdr() for static PIE executables using info extracted > from the aux vector. > ok patrick@ kettenis@ (guenther@) ~ hidden/ctype.h ~ hidden/float.h ~ hidden/signal.h ~ hidden/stdio.h > Make sure attributes are applied to function declarations before they're > used or defined, as clang isn't as forgiving as gcc. Also, declare > __flt_rounds() so that libc can build it despire clang providing its own > <float.h> > ok stefan@ kettenis@ (guenther@) == libexec =========================================================== 04/09 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/libexec spamd ~ spamd.c > must close our fd unconditionally and not just in the !TLS case; tls_close > doesn't do that for us. found, analyzed and patch supplied by Martin Proyer > <martin at proyer dot net>, many thanks! ok bcook > funny since I just discussed a somewhat mysterious spamd-TLS issue with bob > and brent over breakfast yesterday in cambridge - not sure it is this one, > if it is I've been intensively digging in the wrong corner, almost "missing > the obvious" (henning@) == regress =========================================================== 05/09 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/regress lib ~ libedit/readline/Makefile ~ libedit/readline/history.c > These tests didn't work without setting READLINE because we don't > install <editline/readline.h> by default; make them work by using > the copy from the source tree. > Issue reported by otto@. (schwarze@) sys - dev/audio/Makefile - dev/audio/adpcm.c - dev/audio/adpcm.h - dev/audio/autest.1 - dev/audio/autest.c - dev/audio/law.c - dev/audio/law.h - dev/audio_info/Makefile - dev/audio_info/audiotest_gsinfo.1 - dev/audio_info/audiotest_gsinfo.c - dev/audio_rw/Makefile - dev/audio_rw/audiotest_rw.1 - dev/audio_rw/audiotest_rw.c ~ dev/Makefile > remove test programs for the audio driver, as we've better > converage with base tools and programs in regress/lib/libsndio > ok deraadt (ratchov@) == share ============================================================= 06/09 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/share man ~ man7/hier.7 > remove the Xr macro to syspatch (since it doesn;t get installed yet) > but leave the syspatch entry (since the directory does exist); > discussed with ajacoutot; (jmc@) ~ man5/bsd.port.mk.5 > byebye SIGNING_PARAMETERS (espie@) ~ man7/hier.7 > document /etc/acme/ > ok florian (jmc@) ~ man5/ruby-module.5 > Update documentation to reflect change in default ruby version (jeremy@) == sys =============================================================== 07/09 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/sys arch/mips64/include ~ pte.h > Define PTE_* macros in one place. Use the lwu instruction for 32-bit PTE > loads on all systems as the TLB code does not need sign extension. (visa@) arch/mips64/mips64 ~ context.S ~ exception_tfp.S ~ tlbhandler.S > Define PTE_* macros in one place. Use the lwu instruction for 32-bit PTE > loads on all systems as the TLB code does not need sign extension. (visa@) conf ~ files > zap bha, hiding outside dev/ tree (tedu@) kern ~ tty_tty.c > it's not wrong, but it's not clear what the verauth check is for. > add a comment to this effect, and explain the intention. (tedu@) netinet ~ if_ether.c ~ in_pcb.c ~ ip_input.c ~ ip_spd.c ~ tcp_subr.c > pool_setipl for various netinet and netinet6 bits > thank you to everyone who helped reviewed these diffs > ok mpi@ (dlg@) netinet6 ~ nd6.c > pool_setipl for various netinet and netinet6 bits > thank you to everyone who helped reviewed these diffs > ok mpi@ (dlg@) sys ~ tree.h > store the offset of an rb_entry in a struct as an unsigned int. > this means users of the api dont need to know about size_t. > found by jmc@ (dlg@) == usr.bin =========================================================== 08/09 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/usr.bin mg ~ paragraph.c > If you have a paragraph: > 123 > 456 > With the cursor on either the 4, 5 or 6 and no newline after the '6', > and then execute forward-paragraph (M-}), the cursor sits still and > does not move to the end of the second line (after the 6), which is in > effect the end of parapraph. This diff fixes that behaviour. (lum@) ssh ~ kex.c > update ext_info_c every time we receive a kexinit msg; fixes sending of > ext_info if privsep is disabled; report Aris Adamantiadis & Mancha; ok djm@ > (markus@) ~ packet.c > ssh_set_newkeys: print correct block counters on rekeying; ok djm@ > (markus@) == usr.sbin ========================================================== 09/09 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/usr.sbin acme-client ~ acme-client.1 > some indent; from michael reed (jmc@) pkg_add ~ pkg_create.1 ~ OpenBSD/PkgCreate.pm ~ OpenBSD/PkgSign.pm ~ OpenBSD/Signer.pm > remove code that allows pkg_create to sign (espie@) ~ OpenBSD/PackageRepository.pm > heck, have new signatures be -Dnewsign until we flip the switch (espie@) smtpd ~ smtpd.c > use closefrom() before reexec to make sure the new process only gets > the imsg socket. > ok gilles@ sunil@ (eric@) sysmerge ~ sysmerge.8 > Mention PAGER behavior when undefined/empty. > from Michael Reed > ok jmc@ (ajacoutot@) syspatch ~ syspatch.sh > Sort and set readonly global vars. (ajacoutot@) ~ syspatch.sh > Greatly simplify patch type detection and make things more generic. > (ajacoutot@) ~ syspatch.sh > Drop useless variable and check. (ajacoutot@) unbound ~ iterator/iterator.c ~ services/mesh.c > Backport unbound r3830/3831 from upstreadm, fixing an issue where it can > become unresponsive and need to be restarted following a connectivity > outage. > From Brad, ok florian@ > r3831 | wouter | 2016-08-23 04:49:01 -0400 (Tue, 23 Aug 2016) | 3 lines > - Fix #804: lower num_target_queries for iterator also for failed > lookups. > r3830 | wouter | 2016-08-23 04:30:12 -0400 (Tue, 23 Aug 2016) | 3 lines > - Fix #804: unbound stops responding after outage. Fixes queries > that attempt to wait for an empty list of subqueries. (sthen@) ~ iterator/iterator.c ~ services/mesh.c TAGGED OPENBSD_6_0 > MFC: Backport unbound r3830/3831 from upstreadm, fixing an issue where it > can > become unresponsive and need to be restarted following a connectivity > outage. > From Brad, ok florian@ > r3831 | wouter | 2016-08-23 04:49:01 -0400 (Tue, 23 Aug 2016) | 3 lines > - Fix #804: lower num_target_queries for iterator also for failed > lookups. > r3830 | wouter | 2016-08-23 04:30:12 -0400 (Tue, 23 Aug 2016) | 3 lines > - Fix #804: unbound stops responding after outage. Fixes queries > that attempt to wait for an empty list of subqueries. (sthen@) =============================================================================== _______________________________________________ odc mailing list [email protected] http://www.squish.net/mailman/listinfo/odc
