OpenBSD src changes summary for 2017-03-17 ==========================================
distrib/sets etc/Makefile etc/rc.d/dhcrelay6 lib/libc lib/libutil sys/dev/usb sys/kern sys/net sys/netinet sys/sys sys/uvm usr.bin/mandoc usr.bin/tmux usr.bin/whois usr.sbin usr.sbin/dhcrelay6 usr.sbin/smtpd == distrib =========================================================== 01/06 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/distrib sets ~ lists/base/mi ~ lists/man/mi > Enable dhcrelay6(8). > ok deraadt@ (rzalamena@) == etc =============================================================== 02/06 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/etc Makefile ~ Makefile > Enable dhcrelay6(8). > ok deraadt@ (rzalamena@) rc.d/dhcrelay6 + rc.d/dhcrelay6 > Enable dhcrelay6(8). > ok deraadt@ (rzalamena@) == lib =============================================================== 03/06 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/lib libc ~ stdio/asprintf.c ~ stdio/fgetln.c ~ stdio/open_memstream.c ~ stdio/vasprintf.c > Use recallocarray() to avoid leaving detritus in memory when resizing > buffers. We don't bother doing this for objects containing pointers, > but focus on controllable data. > ok millert (deraadt@) ~ gen/fts.c > recallocarray() the string buffer, to avoid leaving such contents > around in the address space. Don't bother doing so for the buffer > which contains aslr'd pointers... > ok millert (deraadt@) ~ stdio/fvwrite.c > Use recallocarray() to avoid leaving detritus in memory when resizing > the string buffer used by asprintf() and vasprintf(). OK deraadt@ > (millert@) ~ stdlib/malloc.3 > Strengthen description of recallocarray(3) behaviour, hoping that readers > make the behaviour -> use case connection. > help from jmc and jsing (deraadt@) ~ stdlib/malloc.3 > remove unneccessary macro; (jmc@) libutil ~ imsg-buffer.c > Grow buffers using recallocarray, to avoid the potential dribble that > the standard realloc*() functions can leave behind. imsg buffers are > sometimes used in protocol stacks which require some secrecy, and layering > violations would be needed to resolve this issue otherwise. > Discussed with many. (deraadt@) == sys =============================================================== 04/06 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/sys dev/usb ~ usbdi.c > ubcmtp(4) works because of a use-after-free. Many more drivers are > certainly doing the same. > So revert r1.85 for the moment, we're too close from the release and > these bugs have been there for years. > Regression reported by tb@ on bugs@ (mpi@) ~ ubcmtp.c > More debug goo to help tracking the use-after-free. (mpi@) kern ~ uipc_socket.c ~ uipc_socket2.c > Revert the NET_LOCK() and bring back pf's contention lock for release. > For the moment the NET_LOCK() is always taken by threads running under > KERNEL_LOCK(). That means it doesn't buy us anything except a possible > deadlock that we did not spot. So make sure this doesn't happen, we'll > have plenty of time in the next release cycle to stress test it. > ok visa@ (mpi@) net ~ if.c ~ if_pflow.c ~ pf.c ~ pf_ioctl.c ~ pf_norm.c ~ pfvar.h > Revert the NET_LOCK() and bring back pf's contention lock for release. > For the moment the NET_LOCK() is always taken by threads running under > KERNEL_LOCK(). That means it doesn't buy us anything except a possible > deadlock that we did not spot. So make sure this doesn't happen, we'll > have plenty of time in the next release cycle to stress test it. > ok visa@ (mpi@) netinet ~ ip_mroute.c > Be more strict on all route iterations, lets always make sure that we > are not going to get a unicast route by accident. > ok mpi@ (rzalamena@) ~ ip_carp.c > carp(4) code is always executed in the 'softnet' thread, so remove > unneeded splnet()/splx() dance. > ok mikeb@, bluhm@ (mpi@) sys ~ systm.h > Revert the NET_LOCK() and bring back pf's contention lock for release. > For the moment the NET_LOCK() is always taken by threads running under > KERNEL_LOCK(). That means it doesn't buy us anything except a possible > deadlock that we did not spot. So make sure this doesn't happen, we'll > have plenty of time in the next release cycle to stress test it. > ok visa@ (mpi@) uvm ~ uvm_vnode.c > Revert the NET_LOCK() and bring back pf's contention lock for release. > For the moment the NET_LOCK() is always taken by threads running under > KERNEL_LOCK(). That means it doesn't buy us anything except a possible > deadlock that we did not spot. So make sure this doesn't happen, we'll > have plenty of time in the next release cycle to stress test it. > ok visa@ (mpi@) == usr.bin =========================================================== 05/06 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/usr.bin mandoc ~ man_html.c ~ mdoc_html.c > Fix regression in mdoc_html.c 1.150, man_html 1.87: > For .Sh, .Ss, .SH, .SS, only write selflink if an id could be constructed. > Crash reported by Raf Czlonka <rczlonka at gmail dot com>, > analysis of root cause by natano@ (schwarze@) ~ man.cgi.8 > Update HISTORY and AUTHORS: we no longer use SQLite. > Outdated information reported by an Anonymous Coward on undeadly.org. > (schwarze@) ~ apropos.1 ~ man.1 > Document man(1) section selection priority, > and correct description of apropos(1) output search order. > Suggested by tb@. (schwarze@) tmux ~ window-choose.c > Add h and l for collapse and expand in choose mode with vi(1) keys, from > Gregory Pakosz. (nicm@) ~ utf8.c ~ cmd-show-messages.c > Fix a couple of argument types. (nicm@) whois ~ whois.c > choose_server() does not always return malloc'd memory so pass > in a pointer that gets filled in when there is something to free. > Noticed by sthen@ when looking up a numeric address. (millert@) == usr.sbin ========================================================== 06/06 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/usr.sbin usr.sbin ~ Makefile > Enable dhcrelay6(8). > ok deraadt@ (rzalamena@) dhcrelay6 + Makefile + bpf.c + dhcp.h + dhcpd.h + dhcrelay6.8 + dhcrelay6.c + dispatch.c + log.c + log.h + packet.c > Import the DHCPv6 relay implementation. > This code was based on the dhcrelay(8) daemon and shares a lot of the > structures and functions. This daemon implements the following RFCs: > * RFC 3315 Section 20: Relay Agent Behavior > * RFC 4649: Relay Agent Remote-ID option > * RFC 6221: Lightweight DHCPv6 Relay Agent (rzalamena@) ~ dhcrelay6.8 ~ dhcrelay6.c > various tweaks; ok rzalamena (jmc@) smtpd ~ iobuf.c > realloc() -> recallocarray(). > use calloc() for initial allocation. > prodded by deraadt@ ok gilles@ (eric@) ~ mproc.c > switch to recallocarray() and remove useless pre-allocation. > prodded by deraadt@ ok gilles@ (eric@) =============================================================================== _______________________________________________ odc mailing list [email protected] http://www.squish.net/mailman/listinfo/odc
