[odc] Daily src changes for 2008-03-02
ODC
auto at squish.net
Mon Mar 3 07:00:01 GMT 2008
OpenBSD src changes summary for 2008-03-02
==========================================
distrib/miniroot distrib/sets
sbin/isakmpd share/man
sys/arch/i386/i386 sys/dev
sys/dev/mii sys/dev/pci
sys/netinet sys/uvm
usr.bin/cvs usr.bin/make
usr.bin/rcs usr.bin/ssh
usr.sbin/bind usr.sbin/ntpd
usr.sbin/pcidump usr.sbin/pkg_add
usr.sbin/ppp
== distrib =========================================================== 01/06 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/distrib
miniroot
~ install.sh
> Keep resolv.conf.shadow around when restarting an install. This
> means manually entered DNS server addresses are not lost when
> an install is restarted. Restores behaviour unintentionally lost
> with r1.127.
> ok deraadt@ (krw@)
sets
~ lists/comp/mi
> sync (deraadt@)
== sbin ============================================================== 02/06 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/sbin
isakmpd
~ monitor_fdpass.c
> Use a union to ensure alignment of the cmsg.
> ok deraadt (hshoexer@)
== share ============================================================= 03/06 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/share
man
~ man9/Makefile ~ man9/random.9
> Add a arc4random_uniform() that returns a uniformly distributed number
> in the range 0 <= x < upper_bound
> Please use this new API instead of "arc4random() % upper_bound", as it
> avoids the "modulo bias" that favours small results when upper_bound is
> not a power of two.
> feedback deraadt@ mcbride@; ok deraadt@ (djm@)
~ man9/random.9
> improve wording - from deraadt@ (djm@)
== sys =============================================================== 04/06 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/sys
arch/i386/i386
~ bios.c
> if there is no smbios, there cannot be acpi
> ok kettenis gwk, tested by a lot of people, and checked from the logs
> (deraadt@)
dev
~ rnd.c ~ rndvar.h
> Add a arc4random_uniform() that returns a uniformly distributed number
> in the range 0 <= x < upper_bound
> Please use this new API instead of "arc4random() % upper_bound", as it
> avoids the "modulo bias" that favours small results when upper_bound is
> not a power of two.
> feedback deraadt@ mcbride@; ok deraadt@ (djm@)
dev/mii
~ brgphy.c ~ nsphy.c
~ qsphy.c
> Add missing half duplex flag for 100baseT4.
> ok kettenis@ (brad@)
~ amphy.c
> Correct a status flag which could cause half duplex to be reported for
> a 10 Mbps full duplex connection but only when not using autoneg.
> ok kettenis@ (brad@)
dev/pci
~ if_em.c
> If bus_dmamap_load_mbuf() fails in em_get_buf() use m_freem() intead of
> m_free() to free the mbuf cluster.
> ok krw@ mglocker@ (brad@)
~ if_em.c
> At the moment em_flowstatus() does not deal with fiber interfaces and
> will not return any flow control status so just return no status (0)
> instead of reading copper PHY registers and not returning anything
> anyway.
> ok kettenis@ (brad@)
~ if_ixgb.c
> If bus_dmamap_load_mbuf() fails in ixgb_get_buf() use m_freem() intead of
> m_free() to free the mbuf cluster.
> ok mglocker@ (brad@)
~ if_bgereg.h
> Correct the comment describing the BGE_PCISTATE_PCI_BUSSPEED bit.
> It had the sense of the bit reversed.
> From jdp at FreeBSD
> ok kettenis@ (brad@)
~ azalia.c
> Clear interrupts on detach. Fixes interrupt storm on digital-only
> controllers, like the HDMI found on Radeon cards.
> Tested by claudio@ and kettenis at . (deanna@)
~ if_sk.c ~ if_msk.c
> Change some spaces to tabs in the transmit functions.
> ok kettenis@ (brad@)
netinet
~ ip_id.c
> Add a arc4random_uniform() that returns a uniformly distributed number
> in the range 0 <= x < upper_bound
> Please use this new API instead of "arc4random() % upper_bound", as it
> avoids the "modulo bias" that favours small results when upper_bound is
> not a power of two.
> feedback deraadt@ mcbride@; ok deraadt@ (djm@)
~ ip_id.c
> because arc4random_uniform is being used, the modulo bias is taken care of
> (deraadt@)
uvm
~ uvm_unix.c
> Include sigcode in core dumps. This makes it possible to backtrace
> through signal handlers with gdb.
> ok miod@ (kettenis@)
== usr.bin =========================================================== 05/06 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/usr.bin
cvs
~ Makefile
> - Remove LIBZ from DPADD.
> - Garbage collect DPADD and LDADD.
> ok joris@ (brad@)
~ config.c ~ config.h
~ modules.c
> allow our parsing functions to stop parsing a config file at
> any given time.
> will come in handy for CVSROOT/loginfo and CVSROOT/commitinfo (joris@)
~ annotate.c ~ diff.c
~ getlog.c ~ tag.c
~ update.c
> Print the right synopsis for commands on error (especially if commands
> share the same function: (r)tag, etc.).
> ok joris (tobias@)
~ annotate.c
> rannotate needs at least one module as argument.
> ok joris (tobias@)
~ rcs.c ~ rcs.h
> rcs_errno is gone: zap rcs_errstr() and rcs_errstrs[] as well.
> ok joris (tobias@)
make
~ make.1
> document the two big bugs left that somewhat hinder parallel mode. (espie@)
rcs
~ diff3.c
> Make sure that temporary file has been successfully opened.
> ok joris (tobias@)
ssh
~ monitor_fdpass.c
> use a union to ensure alignment of the cmsg (pay attention: various other
> parts of the tree need this treatment too); ok djm (deraadt@)
== usr.sbin ========================================================== 06/06 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/usr.sbin
bind
~ lib/isc/random.c ~ lib/isc/shuffle.c
~ lib/isc/include/isc/random.h
> introduce a isc_random_uniform() function to return a uniformly distributed
> number 0 < x <= upper_bound and use it to correct the last tiny bias in the
> shuffle initialisation
> feedback & ok deraadt@ (djm@)
ntpd
~ sensors.c
> set the refid for v4 queries too. this allows sensor-driven ntpd to says
> its
> reference is "HARD" (or "GPS", "UMBG", "UDCF" ... in theory) rather than
> "^@^@^@^@"
> "why not" henning@ (ckuethe@)
pcidump
~ pcidump.8 ~ pcidump.c
> PCI devices are specified by bus:dev:func, not dev:bus:func. Adjust usage
> and man page. (kettenis@)
pkg_add
~ pod/OpenBSD::Intro.pod
> fix bugs formatting (espie@)
~ OpenBSD/Replace.pm
> fix pkg_add -r -B
> problem noticed by winiger@ (espie@)
ppp
~ ppp/command.c ~ ppp/systems.c
~ ppp/systems.h
> Pass a buffer size to InterpretArg() and do not let it overflow it, saves
> propolice some work in command_Expand_Interpret().
> Inspired by FreeBSD; ok deraadt@ (miod@)
~ ppp/systems.c
> another extra endpwent() not needed (deraadt@)
===============================================================================
More information about the odc
mailing list