OpenBSD src changes summary for 2015-12-17 ==========================================
etc/netstart gnu lib/libcrypto libexec/ld.so sys/arch/amd64/amd64 sys/arch/amd64/include sys/arch/i386/i386 sys/dev/acpi sys/dev/ic sys/dev/pci sys/dev/pv sys/dev/usb sys/kern sys/netinet sys/sys usr.bin/nc usr.bin/tmux usr.sbin/npppd usr.sbin/nsd usr.sbin/rebound usr.sbin/unbound usr.sbin/vmd == etc =============================================================== 01/07 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/etc netstart ~ netstart > Simplify multicast option handling (10 less lines) by matching /etc/rc > behavior > towards other YES|NO options and drop the error warning. > with and ok tim@, ok rpe@ on an earlier diff (ajacoutot@) == gnu =============================================================== 02/07 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/gnu gnu ~ gcc/gcc/config/mips/mips.md > Implement a memory barrier for mips. Basically this makes > __sync_synchronize() > emit a "sync" instruction. > ok visa@ (kettenis@) == lib =============================================================== 03/07 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/lib libcrypto ~ cert.pem TAGGED OPENBSD_5_8 > remove O=VeriSign, Inc., OU=Class 3 Public Primary Certification Authority > from cert.pem in 5.8-stable, requested by Josh Grosse (sthen@) == libexec =========================================================== 04/07 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/libexec ld.so ~ ldconfig/prebind.c ~ ldconfig/prebind_delete.c > s/failes/failed/g (mmcc@) == sys =============================================================== 05/07 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/sys arch/amd64/amd64 ~ vmm.c > Move vcpu register state init to vmd. Allows vmd bootloader to make the > decision as to how the vcpu should be set up for initial start and > reset. Also removes some hardcoded register constants from vmm(4). > ok jsing@, mpi@ (mlarkin@) ~ bus_dma.c > add a size to free. from Mathieu (tedu@) arch/amd64/include ~ vmmvar.h > Move vcpu register state init to vmd. Allows vmd bootloader to make the > decision as to how the vcpu should be set up for initial start and > reset. Also removes some hardcoded register constants from vmm(4). > ok jsing@, mpi@ (mlarkin@) arch/i386/i386 ~ bus_dma.c > add a size to free. from Mathieu (tedu@) dev/acpi ~ acpithinkpad.c > Add display brightness support. This is available on the last few > generations > of Thinkpads (x220 and later) and using means the brightness level is > coordinated properly with the firmware. This gets rid of the surprising > brightness changes that would sometimes happen if you used the brightness > keys or if the firmware decided to reset the brightness level for some > other reason. > ok tedu@, jung@, naddy@ (kettenis@) dev/ic ~ lance.c > s/expresion/expression/g (mmcc@) ~ aic79xx.c ~ aic79xx.h ~ aic7xxxvar.h > sprinkling ifndef __linux__ around two variable declarations could not > possibly make this driver portable (tedu@) dev/pci ~ azalia.c > If CORB/RIRB are not running, log error message unconditonnaly > and return error. From Alexey Suslikov <alexey.suslikov at gmail.com> > (ratchov@) dev/pv ~ vmt.c > Use config_mountroot(9) instead of startuphook_establish(9). > Fix a regression reported by Ian Mcwilliam on tech@ (mpi@) dev/usb ~ umass.c > Missing include, spotted by benoit@ (mpi@) kern ~ kern_descrip.c > add size to free. from Mathieu (tedu@) ~ kern_event.c > add ktrace to kevent. ok guenther (tedu@) ~ kern_descrip.c > no need to check null before free (tedu@) ~ kern_event.c > arrays decay to pointers without needing & (tedu@) ~ kern_sched.c > Make the cost of moving a process to the primary cpu a bit higher. This is > the CPU that handles most hardware interrupts but we don't account for that > in any way in the scheduler. So processes (and kernel threads) that are > unlucky enough to end up on this CPU will get less CPU cycles than those > running on other CPUs. This is especially true for the softnet taskq. > There network interrupts will prevent the softnet taskq from running. This > means that the more packets we receive, the less packets we can actually > process and/or forward. This is why "unlocking" network drivers actually > decreases the forwarding performance. This diff restores most of the lost > performance by making it less likely that the softnet taskq ends up on the > same CPU that handles network interrupts. > Tested by Hrvoje Popovski > ok mpi@, deraadt@ (kettenis@) netinet ~ if_ether.c > rm a bit more trailers code. no longer accept them as alias for ethernet. > ok mpi (tedu@) sys ~ ktrace.h > add ktrace to kevent. ok guenther (tedu@) == usr.bin =========================================================== 06/07 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/usr.bin nc ~ netcat.c > Add missing colon after "Peer name" in verbose output. Mentioned on the > lists recently. (mmcc@) tmux ~ cmd-find.c > As well as setting up the state, actually use it in cmd_find_target. > (nicm@) == usr.sbin ========================================================== 07/07 == http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/usr.sbin npppd ~ common/slist.c ~ pptp/pptpd.c > Replace arc4random() % i by arc4random_uniform(i) to avoid modulo bias. > Part of a diff by Matthew Martin, reviewed by deraadt@ and me. > ok deraadt@ (tb@) ~ common/debugmacro.h ~ common/debugutil.c ~ common/debugutil.h ~ common/hash.c ~ common/hash.h ~ common/ipsec_util.c ~ common/ipsec_util.h ~ common/ipsec_util_local.h ~ common/net_utils.c ~ common/net_utils.h ~ common/radish.c ~ common/radish.h ~ common/recvfromto.c ~ common/recvfromto.h ~ common/slist.h ~ common/slist_test.c ~ common/time_utils.c > Add missing $OpenBSD$ tags. (tb@) ~ l2tp/l2tpd.c ~ pppoe/pppoed.c > Replace 'arc4random() % (2^k - 1)' by 'arc4random() & (2^k - 1)' to > avoid modulo bias. Part of a diff by Matthew Martin, reviewed by > deraadt@ and me. > ok deraadt@ (tb@) ~ common/addr_range.c > s/failes/failed/g (mmcc@) nsd ~ configure > regen NSD's autoconf script, forgotten with the update - noticed by Brad > (sthen@) rebound ~ rebound.c > add return code to newrequest to distinguish between cache hit and error. > we want to keep looping for more requests after a hit, not stop. > (though i'm reconsidering if the looping is worthwhile. maybe should just > return to kevent() after each request.) (tedu@) unbound ~ iterator/iter_hints.c TAGGED OPENBSD_5_8 > update hints for h.root in 5.8-stable, requested by phessler (sthen@) vmd ~ loadfile.h ~ loadfile_elf.c ~ vmm.c > Move vcpu register state init to vmd. Allows vmd bootloader to make the > decision as to how the vcpu should be set up for initial start and > reset. Also removes some hardcoded register constants from vmm(4). > ok jsing@, mpi@ (mlarkin@) =============================================================================== _______________________________________________ odc mailing list [email protected] http://www.squish.net/mailman/listinfo/odc
