OpenBSD src changes summary for 2017-03-08 ==========================================
distrib/arm64 lib/libc regress/sys regress/usr.bin sbin/dhclient sys/arch/arm64/arm64 sys/arch/arm64/conf sys/arch/arm64/include sys/arch/macppc/dev sys/arch/sgi/hpc sys/dev/ic sys/dev/pci sys/dev/usb sys/net sys/netinet6 usr.bin/mandoc usr.bin/ssh usr.bin/tmux usr.sbin/switchd usr.sbin/tcpdump == distrib =========================================================== 01/07 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/distrib arm64 ~ ramdisk/install.md > copy rpi3 u-boot to the installed disk > Even though this can't be the SD card at the moment, it is possible to > boot off USB if an OTP bit has been set. (jsg@) == lib =============================================================== 02/07 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/lib libc ~ arch/aarch64/gen/setjmp.S > Correct arm64 sigprocmask parameter passing in setjmp/longjmp calls. > Always return nonzero as the return code from longjmp. > ok guenther@ patrick@ (drahn@) == regress =========================================================== 03/07 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/regress sys ~ kern/fchdir/fchdir.c > fix comment block (deraadt@) ~ kern/wait/wait.c > don't use __syscall to emulate _exit, just call _exit (deraadt@) ~ kern/getpeereid/getpeereid_test.c ~ kern/nanosleep/nanosleep.c ~ kern/unixsock/unixsock_test.c > warning cleanups; ok guenther (deraadt@) ~ crypto/cmac/cmac_test.c > format string mishandling (deraadt@) usr.bin ~ mandoc/roff/de/infinite.in ~ mandoc/roff/de/infinite.out_ascii ~ mandoc/roff/de/infinite.out_lint > prevent infinite recursion while expanding the arguments > of a user-defined macro; issue found by tb@ with afl(1) (schwarze@) ~ mandoc/mdoc/Bd/Makefile ~ mandoc/mdoc/Bl/Makefile ~ mandoc/mdoc/D1/Makefile ~ mandoc/mdoc/Sm/Makefile ~ mandoc/mdoc/break/Makefile + mandoc/mdoc/Bd/badargs.out_markdown + mandoc/mdoc/Bd/beforeNAME.out_markdown + mandoc/mdoc/Bd/blank.out_markdown + mandoc/mdoc/Bd/break.out_markdown + mandoc/mdoc/Bd/broken.out_markdown + mandoc/mdoc/Bd/centered.out_markdown + mandoc/mdoc/Bd/empty.out_markdown + mandoc/mdoc/Bd/nested.out_markdown + mandoc/mdoc/Bd/offset-empty.out_markdown + mandoc/mdoc/Bd/offset-neg.out_markdown + mandoc/mdoc/Bd/spacing.out_markdown + mandoc/mdoc/Bd/unclosed.out_markdown + mandoc/mdoc/Bl/badargs.out_markdown + mandoc/mdoc/Bl/breakingIt.out_markdown + mandoc/mdoc/Bl/broken.out_markdown + mandoc/mdoc/Bl/emptyhead.out_markdown + mandoc/mdoc/Bl/emptytag.out_markdown + mandoc/mdoc/Bl/extend.out_markdown + mandoc/mdoc/Bl/hang.out_markdown + mandoc/mdoc/Bl/multitag.out_markdown + mandoc/mdoc/Bl/nested.out_markdown + mandoc/mdoc/Bl/noIt.out_markdown + mandoc/mdoc/Bl/offset.out_markdown + mandoc/mdoc/Bl/secstart.out_markdown + mandoc/mdoc/Bl/tag.out_markdown + mandoc/mdoc/D1/spacing.out_markdown + mandoc/mdoc/Sm/scope.out_markdown + mandoc/mdoc/break/tail.out_markdown > enable -T markdown tests of filled displays and tagged lists (schwarze@) ~ mandoc/mdoc/Bl/Makefile + mandoc/mdoc/Bl/colNoIt.out_markdown + mandoc/mdoc/Bl/column.out_markdown > .Bl -column never gets blank lines between rows (schwarze@) ~ mandoc/mdoc/Dv/font.out_markdown ~ mandoc/mdoc/Er/font.out_markdown ~ mandoc/mdoc/Ev/font.out_markdown ~ mandoc/mdoc/Li/font.out_markdown ~ mandoc/mdoc/Tn/font.out_markdown > According to the CommonMark specification, backslash escapes > and markdown markup do not work inside code spans. (schwarze@) ~ mandoc/Makefile.inc ~ mandoc/char/Makefile.inc ~ mandoc/eqn/Makefile.inc ~ mandoc/man/Makefile.inc ~ mandoc/mdoc/Ad/Makefile ~ mandoc/mdoc/An/Makefile ~ mandoc/mdoc/Ap/Makefile ~ mandoc/mdoc/Aq/Makefile ~ mandoc/mdoc/Ar/Makefile ~ mandoc/mdoc/At/Makefile ~ mandoc/mdoc/Bd/Makefile ~ mandoc/mdoc/Bf/Makefile ~ mandoc/mdoc/Bk/Makefile ~ mandoc/mdoc/Bl/Makefile ~ mandoc/mdoc/Brq/Makefile ~ mandoc/mdoc/Bx/Makefile ~ mandoc/mdoc/Cd/Makefile ~ mandoc/mdoc/Cm/Makefile ~ mandoc/mdoc/D1/Makefile ~ mandoc/mdoc/Db/Makefile ~ mandoc/mdoc/Dd/Makefile ~ mandoc/mdoc/Dl/Makefile ~ mandoc/mdoc/Dq/Makefile ~ mandoc/mdoc/Dt/Makefile ~ mandoc/mdoc/Dv/Makefile ~ mandoc/mdoc/Em/Makefile ~ mandoc/mdoc/Eo/Makefile ~ mandoc/mdoc/Er/Makefile ~ mandoc/mdoc/Ev/Makefile ~ mandoc/mdoc/Ex/Makefile ~ mandoc/mdoc/Fd/Makefile ~ mandoc/mdoc/Fl/Makefile ~ mandoc/mdoc/Fo/Makefile ~ mandoc/mdoc/Ft/Makefile ~ mandoc/mdoc/Ic/Makefile ~ mandoc/mdoc/In/Makefile ~ mandoc/mdoc/Lb/Makefile ~ mandoc/mdoc/Li/Makefile ~ mandoc/mdoc/Lk/Makefile ~ mandoc/mdoc/Ms/Makefile ~ mandoc/mdoc/Mt/Makefile ~ mandoc/mdoc/Nd/Makefile ~ mandoc/mdoc/Nm/Makefile ~ mandoc/mdoc/No/Makefile ~ mandoc/mdoc/Ns/Makefile ~ mandoc/mdoc/Oo/Makefile ~ mandoc/mdoc/Op/Makefile ~ mandoc/mdoc/Os/Makefile ~ mandoc/mdoc/Ox/Makefile ~ mandoc/mdoc/Pa/Makefile ~ mandoc/mdoc/Pf/Makefile ~ mandoc/mdoc/Pp/Makefile ~ mandoc/mdoc/Qq/Makefile ~ mandoc/mdoc/Rs/Makefile ~ mandoc/mdoc/Rv/Makefile ~ mandoc/mdoc/Sh/Makefile ~ mandoc/mdoc/Sm/Makefile ~ mandoc/mdoc/Sq/Makefile ~ mandoc/mdoc/St/Makefile ~ mandoc/mdoc/Sx/Makefile ~ mandoc/mdoc/Sy/Makefile ~ mandoc/mdoc/Tn/Makefile ~ mandoc/mdoc/Ud/Makefile ~ mandoc/mdoc/Ux/Makefile ~ mandoc/mdoc/Va/Makefile ~ mandoc/mdoc/Vt/Makefile ~ mandoc/mdoc/Xr/Makefile ~ mandoc/mdoc/blank/Makefile ~ mandoc/mdoc/break/Makefile ~ mandoc/roff/Makefile.inc ~ mandoc/tbl/Makefile.inc > Now that markdown output is tested for almost everything, test all > input files in -T markdown output mode by default and only mark > those files with SKIP_MARKDOWN that are not to be tested. > Much easier to read, and almost minus 40 lines of Makefile code. > (schwarze@) == sbin ============================================================== 04/07 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/sbin dhclient ~ dhclient.c ~ options.c ~ packet.c > Add comments to #endif's to allow easy differentiation between many > incoming #ifdef's. (krw@) ~ dhclient.c > Shuffle some declarations. Both more style(9)'ish and make future diff > smaller. (krw@) ~ dhclient.c > Split a declaration to make the later insertion of #ifdef easier. (krw@) ~ dhclient.c > Shuffle some functions around to group together functions that will > later be #ifdef'd in or out. (krw@) ~ kroute.c > Move ROUNDUP() to top of file, outside of warring #ifdef's to come. (krw@) ~ dhclient.c > Move a line of code to group together code that will be mediated > by #ifdef's. (krw@) ~ dispatch.c > Stray #endif w/o comment. (krw@) ~ privsep.c > Shuffle another bit of code closer to its friends. (krw@) ~ privsep.h > Shuffle, shuffle, shuffle. (krw@) ~ dhcpd.h > Shuffle, shuffle, shuffle. (krw@) ~ dhclient.c > Shuffle, shuffle, shuffle. (krw@) == sys =============================================================== 05/07 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/sys arch/arm64/arm64 ~ intr.c > Establish API to route interrupts to specific CPU cores. > Based on and ok drahn@ (patrick@) arch/arm64/conf ~ GENERIC ~ RAMDISK > Enable em(4), ahci(4), nvme(4) and USB controllers attached to pci(4). > (patrick@) ~ GENERIC ~ RAMDISK > Enable ix(4). Requested by deraadt@ (jsg@) ~ GENERIC ~ RAMDISK > Enable re(4). (patrick@) ~ GENERIC ~ RAMDISK > Enable eephy(4). (patrick@) ~ GENERIC ~ RAMDISK > Enable msk(4). (patrick@) arch/arm64/include ~ intr.h > Establish API to route interrupts to specific CPU cores. > Based on and ok drahn@ (patrick@) arch/macppc/dev ~ if_bm.c > Do not clear IFF_UP, even in the error path, clearing IFF_RUNNING > is enough. > This flag should only be set by the stack, drivers shouldn't mess > with it. > Discussed with dlg@ and mikeb@, ok mikeb@, stsp@ (mpi@) arch/sgi/hpc ~ if_sq.c > Fix logic in the driver preventing it to receive broadcast frames > before configuring an address. > Make dhclient(8) work on sq(4). > Problem reported by Frank Scheiner, diff from miod@ (mpi@) dev/ic ~ ar5008.c ~ ar9003.c > Do not clear IFF_UP, even in the error path, clearing IFF_RUNNING > is enough. > This flag should only be set by the stack, drivers shouldn't mess > with it. > Discussed with dlg@ and mikeb@, ok mikeb@, stsp@ (mpi@) dev/pci ~ if_de.c ~ if_ipw.c ~ if_iwi.c ~ if_iwm.c ~ if_iwn.c ~ if_wpi.c > Do not clear IFF_UP, even in the error path, clearing IFF_RUNNING > is enough. > This flag should only be set by the stack, drivers shouldn't mess > with it. > Discussed with dlg@ and mikeb@, ok mikeb@, stsp@ (mpi@) dev/usb ~ if_umb.c > Do not clear IFF_UP, even in the error path. > This flag should only be set by the stack, drivers shouldn't > mess with it. > Discussed with dlg@ and mikeb@, ok gerhard@ (mpi@) net ~ if_etherip.h > Sync struct etheripstat decl with netinet/ip_ether.h (jca@) ~ if.c > Do not unset IFF_UP twice. > ok stsp@, claudio@, mikeb@, dlg@ (mpi@) ~ if_ppp.c > No need to clear IFF_UP after calling if_down(). (mpi@) netinet6 ~ nd6.c > Always invalidate ND entries in nd6_free(). > This way RTF_CACHED entries are properly flushed. Fix a regression > reported by weerd@ and also tested by matthieu@. > ok bluhm@ (mpi@) == usr.bin =========================================================== 06/07 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/usr.bin mandoc ~ roff.c > prevent infinite recursion while expanding the arguments > of a user-defined macro; issue found by tb@ with afl(1) (schwarze@) ~ mdoc_markdown.c > Add a blank after ">" when quoting. > If is allowed by markdown syntax and more human-readable. (schwarze@) ~ mdoc_markdown.c > .Bl -column never gets blank lines between rows (schwarze@) ~ mdoc_markdown.c > Do not increment .Bl -enum list markers beyond two digits. > Otherwise, we would indent subsequent paragraphs less than > the CommonMark specification requires, harming portability. (schwarze@) ~ mdoc_markdown.c > The CommonMark specification allows list markers fo the form "number) " > as well as "number. ", so escape closing parentheses after leading digits > to improve portability. (schwarze@) ~ mdoc_markdown.c > According to the CommonMark specification, backslash escapes > and markdown markup do not work inside code spans. (schwarze@) ~ mdoc_markdown.c > The CommonMark specification wants that opening parentheses > inside link destinations be escaped. > While here, remove the obsolete ESC_PAR. (schwarze@) ~ mandoc.1 > Document that -T markdown produces ASCII output, and the implied > limitations. Of course, we could write UTF-8 output instead, > but even the CommonMark specification doesn't require parsers > to support that, so portability would be doubtful. > While here, provide a link to the CommonMark specification. (schwarze@) ssh ~ ssh.c > quote [host]:port in generated ProxyJump commandline; the [ / ] > characters can confuse some shells (e.g. zsh). > Reported by Lauri Tirkkonen via bugs@ (djm@) tmux ~ cmd-attach-session.c ~ cmd-break-pane.c ~ cmd-if-shell.c ~ cmd-new-session.c ~ cmd-new-window.c ~ cmd-run-shell.c ~ cmd-split-window.c ~ control-notify.c ~ format.c ~ tmux.h ~ window-copy.c > Add a helper function for the most common format_create/defaults/expand > pattern. (nicm@) ~ tty.c > Always send smkx to the terminal outside, the keys we get from terminfo > are the keys when it is on. (nicm@) ~ options.c > Handle empty options correctly. (nicm@) ~ window-copy.c > Skip over padding cells when moving the cursor left or right. (nicm@) == usr.sbin ========================================================== 07/07 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/usr.sbin switchd ~ switchd.conf.5 > argument to the document description (Nd) macro should be lowercase. > ok jmc@ (sobrado@) tcpdump ~ print-etherip.c > Fix etherip version parsing, ok dlg@ (jca@) =============================================================================== _______________________________________________ odc mailing list [email protected] http://www.squish.net/mailman/listinfo/odc
