OpenBSD src changes summary for 2016-01-06 ==========================================
bin/expr distrib/sets etc/examples/vm.conf etc/mtree/4.4BSD.dist etc/signify/openbsd-55-base.pub etc/signify/openbsd-55-fw.pub etc/signify/openbsd-55-pkg.pub etc/signify/openbsd-56-base.pub etc/signify/openbsd-56-fw.pub etc/signify/openbsd-56-pkg.pub etc/signify/openbsd-57-base.pub etc/signify/openbsd-57-fw.pub etc/signify/openbsd-57-pkg.pub games/phantasia lib/libc sbin/ifconfig share/man sys/arch/amd64/stand/efiboot sys/arch/amd64/stand/libsa sys/conf sys/dev/acpi sys/dev/pci sys/kern sys/net80211 sys/netinet6 sys/sys usr.bin/kdump usr.bin/signify usr.bin/vi == bin =============================================================== 01/09 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/bin expr ~ expr.c > use int64_t for arithmetic. 64 bits ought to be enough for anyone. > ok deraadt millert schwarze (tedu@) == distrib =========================================================== 02/09 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/distrib sets ~ lists/base/mi > sync (deraadt@) ~ lists/base/mi > sync (deraadt@) == etc =============================================================== 03/09 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/etc examples/vm.conf ~ examples/vm.conf > This is just an example and we don't have to bump the set's version > number all the time, but let's use install59.fs as it will be the > first release that includes vmm. (reyk@) mtree/4.4BSD.dist ~ mtree/4.4BSD.dist > delete vi catalog space; ok millert (deraadt@) signify/openbsd-55-base.pub - signify/openbsd-55-base.pub > some old signify keys no longer have relevance to new releases. > prompted by comments from naddy (deraadt@) signify/openbsd-55-fw.pub - signify/openbsd-55-fw.pub > some old signify keys no longer have relevance to new releases. > prompted by comments from naddy (deraadt@) signify/openbsd-55-pkg.pub - signify/openbsd-55-pkg.pub > some old signify keys no longer have relevance to new releases. > prompted by comments from naddy (deraadt@) signify/openbsd-56-base.pub - signify/openbsd-56-base.pub > some old signify keys no longer have relevance to new releases. > prompted by comments from naddy (deraadt@) signify/openbsd-56-fw.pub - signify/openbsd-56-fw.pub > some old signify keys no longer have relevance to new releases. > prompted by comments from naddy (deraadt@) signify/openbsd-56-pkg.pub - signify/openbsd-56-pkg.pub > some old signify keys no longer have relevance to new releases. > prompted by comments from naddy (deraadt@) signify/openbsd-57-base.pub - signify/openbsd-57-base.pub > some old signify keys no longer have relevance to new releases. > prompted by comments from naddy (deraadt@) signify/openbsd-57-fw.pub - signify/openbsd-57-fw.pub > some old signify keys no longer have relevance to new releases. > prompted by comments from naddy (deraadt@) signify/openbsd-57-pkg.pub - signify/openbsd-57-pkg.pub > some old signify keys no longer have relevance to new releases. > prompted by comments from naddy (deraadt@) == games ============================================================= 04/09 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/games phantasia ~ map.c ~ setup.c > Use return instead of exit from main. > ok mestre@ (tb@) ~ map.c > Ansify map.c. > ok mestre@ (tb@) ~ fight.c ~ gamesupport.c ~ interplayer.c ~ io.c ~ main.c ~ misc.c > Ansify the bulk of phantasia. No binary change on amd64. Based > on an older diff by mestre. > ok mestre@ (tb@) ~ misc.c > Last step of ansification of phantasia: > Convert function() to function(void). > ok mestre@ (tb@) ~ setup.c > ansify > ok mestre@ (tb@) ~ fight.c ~ gamesupport.c ~ include.h ~ interplayer.c ~ io.c ~ main.c ~ misc.c ~ phantglobs.c ~ phantglobs.h ~ phantstruct.h ~ setup.c > Remove unnecessary header files from phantasia(6) > Lesson learnt: don't include an header sorted alphabetically just because. > The > external definition of variables was done on phantglobs.h but their > declaration > was on phantstruct.h and therefore the latter must be included before the > former. It's easier to just include phantstruct.h inside phantglobs.h since > it > always need it. > With help and OK from tb@ (mestre@) - include.h > Forgot to remove this file on my previous commit (mestre@) == lib =============================================================== 05/09 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/lib libc ~ stdlib/malloc.3 ~ stdlib/malloc.c > Long ago, malloc internally had two kinds of failures, warnings and errors. > The 'A' option elevated warnings to errors, and has been the default for > some > time. Then warnings were effectively eliminated in favor of everything > being an error, but then the 'a' flag turned real errors into warnings! > Remove the 'a' option entirely. You shouldn't have used it anyway. > ok tb tdeval (tedu@) ~ sys/pledge.2 > add a bug that the path whitelist is not available at this time. (tedu@) == sbin ============================================================== 06/09 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/sbin ifconfig ~ ifconfig.c > correct format string. ok deraadt stsp (tedu@) == share ============================================================= 07/09 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/share man ~ man5/malloc.conf.5 > malloc option 'a' to go away (tedu@) == sys =============================================================== 08/09 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/sys arch/amd64/stand/efiboot ~ efidev.c > Bring efidev.c goodness into softraid.c by returning and displaying > the same error messages on GPT processing failures. > Add a message about failed GPT entry checksum error to both. This > was inadvertantly deleted in the switch to one entry at a time > processing. > tweaks & ok yasuoka@ jsing@ (krw@) arch/amd64/stand/libsa ~ softraid.c > Bring efidev.c goodness into softraid.c by returning and displaying > the same error messages on GPT processing failures. > Add a message about failed GPT entry checksum error to both. This > was inadvertantly deleted in the switch to one entry at a time > processing. > tweaks & ok yasuoka@ jsing@ (krw@) ~ softraid.c > We want the first, not the last, OpenBSD partition in the MBR. (krw@) conf ~ GENERIC > deactivate POOL_DEBUG for release > ok deraadt, jsg (benno@) ~ newvers.sh > document the signify command for the next release, so that users can > verify before the netx upgrade. > document that signify.1 needs an edit bump once in a while. > ok tedu@ florian@ (benno@) dev/acpi ~ acpi.c > Follow the spec and check the SCI_EN bit to see if ACPI has already > been enabled and don't bother enabling it again if it is. This should fix > systems that don't implement legacy mode and therefore set ACPI_ENABLE and > ACPI_DISABLE to zero. > ok mlarkin@, yasuoka@ (kettenis@) dev/pci ~ if_oce.c > revert 1.87, more work is needed here (mikeb@) ~ drm/drm_drv.c > Add pledge "drm", which allows a subset of the drm(4) ioctls. These are > basically only the ioctls that Linux allows on the so-called "render > nodes". > For now, it also allows DRM_IOCTL_GET_MAGIC and DRM_IOCTL_GEM_OPEN, as we > don't implement prime/dma-buf yet in OpenBSD. That still leaves a big > gaping > hole, so they will be removed as soon as we can. > Based on a diff by robert@, who did all the heavy lifting by studying the > behaviour of the chromium GPU process, with some further suggestions by > deraadt@. > ok jsg@, deraadt@, robert@ (kettenis@) ~ if_iwm.c > iwm(4) adjusts the firmware's tx rate retry table dynamically so we must > tell the firmware to always try the first rate in the table first. > ok kettenis@ (stsp@) ~ if_iwn.c > Ensure that iwn(4) always uses a CCK rate as last fallback in the > firmware's > tx rate retry table. This will hopefully make 11n mode work in very noisy > environments. iwm(4) already does the same. > ok kettenis@ (stsp@) ~ drm/radeon/radeon_kms.c > Add DRM_RENDER_ALLOW flag to the appropriate ioctls such that pledge "drm" > allows them. (kettenis@) ~ if_iwm.c ~ if_iwn.c > Initialize the A-MPDU parameters field in HT capability elements. > ok kettenis@ (stsp@) kern ~ kern_pledge.c > Add pledge "drm", which allows a subset of the drm(4) ioctls. These are > basically only the ioctls that Linux allows on the so-called "render > nodes". > For now, it also allows DRM_IOCTL_GET_MAGIC and DRM_IOCTL_GEM_OPEN, as we > don't implement prime/dma-buf yet in OpenBSD. That still leaves a big > gaping > hole, so they will be removed as soon as we can. > Based on a diff by robert@, who did all the heavy lifting by studying the > behaviour of the chromium GPU process, with some further suggestions by > deraadt@. > ok jsg@, deraadt@, robert@ (kettenis@) ~ uipc_socket.c > Prevent integer overflows in sosend() and soreceive() by converting > min()+uiomovei() to ulmin()+uiomove() and re-arranging space computations > in sosend(). The soreceive() part was also reported by Martin Natano. > ok bluhm@ and also discussed with tedu@ (stefan@) ~ kern_event.c > tidy up whitespace, etc. (tedu@) ~ kern_descrip.c ~ kern_pledge.c ~ sys_pipe.c ~ vfs_syscalls.c ~ vfs_vnops.c > remove unnecessary casts where the incoming type is void *. (tedu@) ~ kern_pledge.c > turn off the whitepath code for now; we're not sure it's ready for release > ok deraadt (tedu@) net80211 ~ ieee80211.h ~ ieee80211_output.c ~ ieee80211_var.h > Initialize the A-MPDU parameters field in HT capability elements. > ok kettenis@ (stsp@) netinet6 ~ ip6_input.c > fix comment typos (sthen@) sys ~ conf.h ~ pledge.h > Add pledge "drm", which allows a subset of the drm(4) ioctls. These are > basically only the ioctls that Linux allows on the so-called "render > nodes". > For now, it also allows DRM_IOCTL_GET_MAGIC and DRM_IOCTL_GEM_OPEN, as we > don't implement prime/dma-buf yet in OpenBSD. That still leaves a big > gaping > hole, so they will be removed as soon as we can. > Based on a diff by robert@, who did all the heavy lifting by studying the > behaviour of the chromium GPU process, with some further suggestions by > deraadt@. > ok jsg@, deraadt@, robert@ (kettenis@) == usr.bin =========================================================== 09/09 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/usr.bin kdump ~ ktrstruct.c > print more info about kevent (tedu@) signify ~ signify.1 > document the signify command for the next release, so that users can > verify before the netx upgrade. > document that signify.1 needs an edit bump once in a while. > ok tedu@ florian@ (benno@) vi ~ common/gs.h ~ common/main.c ~ common/msg.c ~ common/options.c ~ common/options_f.c ~ include/com_extern.h ~ include/options_def.h > Remove the message catalog DB. This removes the msg_open() and > msg_close() functions along with the msgcat command. > From Martijn van Duren (millert@) ~ cl/cl_screen.c ~ cl/cl_term.c ~ common/delete.c ~ common/exf.c ~ common/key.c ~ common/line.c ~ common/log.c ~ common/mark.c ~ common/msg.c ~ common/options.c ~ common/options_f.c ~ common/put.c ~ common/recover.c ~ common/search.c ~ ex/ex.c ~ ex/ex_abbrev.c ~ ex/ex_append.c ~ ex/ex_args.c ~ ex/ex_argv.c ~ ex/ex_bang.c ~ ex/ex_cd.c ~ ex/ex_display.c ~ ex/ex_filter.c ~ ex/ex_global.c ~ ex/ex_init.c ~ ex/ex_join.c ~ ex/ex_map.c ~ ex/ex_mark.c ~ ex/ex_mkexrc.c ~ ex/ex_move.c ~ ex/ex_open.c ~ ex/ex_preserve.c ~ ex/ex_read.c ~ ex/ex_screen.c ~ ex/ex_script.c ~ ex/ex_shift.c ~ ex/ex_subst.c ~ ex/ex_tag.c ~ ex/ex_util.c ~ ex/ex_visual.c ~ ex/ex_write.c ~ vi/getc.c ~ vi/v_ch.c ~ vi/v_ex.c ~ vi/v_increment.c ~ vi/v_match.c ~ vi/v_replace.c ~ vi/v_screen.c ~ vi/v_search.c ~ vi/v_txt.c ~ vi/v_util.c ~ vi/v_xchar.c ~ vi/vi.c ~ vi/vs_refresh.c ~ vi/vs_smap.c ~ vi/vs_split.c > Remove the numeric identifiers at the beginning of the messages > which used to be used as the message number to lookup in the > catalog. From Martijn van Duren (millert@) ~ common/exf.c ~ common/msg.c ~ ex/ex_append.c ~ ex/ex_bang.c ~ ex/ex_print.c ~ ex/ex_subst.c ~ include/com_extern.h ~ vi/vs_msg.c ~ vi/vs_refresh.c > Remove the msg_cat() function and adjust its former callers. > From Martijn van Duren (millert@) - catalog/Makefile - catalog/README - catalog/dump.c - catalog/dutch - catalog/dutch.base - catalog/dutch.check - catalog/dutch.owner - catalog/english - catalog/english.base - catalog/english.check - catalog/english.owner - catalog/french - catalog/french.base - catalog/french.check - catalog/german - catalog/german.base - catalog/german.check - catalog/german.owner - catalog/polish - catalog/polish.base - catalog/polish.check - catalog/polish.owner - catalog/ru_RU.KOI8-R - catalog/ru_RU.KOI8-R.base - catalog/ru_RU.KOI8-R.check - catalog/ru_RU.KOI8-R.owner - catalog/spanish - catalog/spanish.base - catalog/spanish.check - catalog/spell.ok - catalog/swedish - catalog/swedish.base - catalog/swedish.check - catalog/swedish.owner - catalog/uk_UA.KOI8-U - catalog/uk_UA.KOI8-U.base - catalog/uk_UA.KOI8-U.owner ~ LAYOUT ~ build/Makefile ~ build/pathnames.h > Remove the actual message catalogs. From Martijn van Duren (millert@) ~ include/com_extern.h > Remove prototype for now-deleted f_msgcat(). (millert@) ~ docs/help ~ docs/USD.doc/vi.man/vi.1 ~ docs/USD.doc/vi.ref/set.opt.roff ~ docs/USD.doc/vi.ref/spell.ok ~ docs/USD.doc/vi.ref/vi.ref > Remove msgcat from the documentation. (millert@) - build/README > We don't use configure so this file is full of lies and we are > better off without it. (millert@) ~ README > Remove mention of message catalog dir. (millert@) =============================================================================== _______________________________________________ odc mailing list [email protected] http://www.squish.net/mailman/listinfo/odc
