[odc] Daily src changes for 2009-03-19

[ODC]

OpenBSD src changes summary for 2009-03-19
==========================================

sbin/bioctl                             sbin/pfctl
share/man                               sys/dev/pci
sys/nfs                                 usr.bin/cvs
usr.bin/ssh                             usr.sbin/bgpd
usr.sbin/smtpd                          

== sbin ============================================================== 01/05 ==

  http://www.openbsd.org/cgi-bin/cvsweb/src/sbin

bioctl

  ~ bioctl.8                              

  > correct an example in the manpage:	after creating a new crypto volume,
  > it is sufficient to zero the first megabyte of the disk, not the whole
  > disk.
  > ok marco@ mpf@ rainer@ jmc@ (grunk@)

  ~ bioctl.8                              

  > minor tweaks (sorry grunk!); (jmc@)

pfctl

  ~ pf_print_state.c                      

  > pfctl -ss printed state levels for ICMPv6.	Disable this the same
  > way it has already been done for ICMPv4.
  > ok mcbride@ (bluhm@)

== share ============================================================= 02/05 ==

  http://www.openbsd.org/cgi-bin/cvsweb/src/share

man

  ~ man9/uvm.9                            

  > no .Pp before or after sections; (jmc@)

  ~ man4/mpu.4                            ~ man4/usbf.4

  > comment out some macros until they are needed; (jmc@)

== sys =============================================================== 03/05 ==

  http://www.openbsd.org/cgi-bin/cvsweb/src/sys

dev/pci

  ~ ips.c                                 

  > Remove unused code. (grange@)

  ~ ips.c                                 

  > Make ips_timeout to be usable not only with scsi xfers. (grange@)

  ~ ips.c                                 

  > Unify command id printing. (grange@)

  ~ ips.c                                 

  > For pass-through commands adjust timeout value to what controller
  > supports. Make sure our timeout will be fired after controller
  > gives up. (grange@)

  ~ ips.c                                 

  > For rebuilds use a special REBUILD command instead of a generic
  > SETSTATE, for some reason it works better on lpinto's machine. (grange@)

nfs

  ~ nfs_bio.c                             

  > We don't count buffercache stats in the B_PHYS case, so fix nfs to not
  > increment the num{read,write} and pending{read,write} statistics in that
  > case, since biodone won't change them on completion.
  > On another note, I'm not sure that we use physical buffers for swapping
  > over nfs anymore, so this chunk may be superfluous.
  > beck@ came up with the same diff "So anyway rather than me commiting it
  > from my copy, I'm giving you the OK and the commit. since it officially
  > makes you a buffer cache and NFS hacker };-)" (oga@)

== usr.bin =========================================================== 04/05 ==

  http://www.openbsd.org/cgi-bin/cvsweb/src/usr.bin

cvs

  ~ trigger.c                             

  > only pass the real $var to expand_vars() instead of the entire
  > line, this way commitinfo and stuff works again.
  > found by myself the hard way, fix by tobias@
  > who's not around to commit. (joris@)

  ~ cvs.h                                 ~ entries.c
  ~ file.c                                ~ remote.h
  ~ server.c                              

  > properly register Questionable files using CVS/Entries server-side
  > in order for them to show up as "? foobar" when opencvs is in server mode.
  > (joris@)

  ~ server.c                              

  > kill unused vars (joris@)

  ~ diff.c                                

  > do not diff files server-side that are uptodate but require a patch,
  > as the client does not provide these files at all because they are
  > not modified in anyway. otherwise this results in bad mojo.
  > found by myself and sthen@ (joris@)

ssh

  ~ ssh.1                                 

  > for "Ciphers", just point the reader to the keyword in ssh_config(5), just
  > as we do for "MACs": this stops us getting out of sync when the lists
  > change;
  > fixes documentation/6102, submitted by Peter J. Philipp
  > alternative fix proposed by djm
  > ok markus (jmc@)

== usr.sbin ========================================================== 05/05 ==

  http://www.openbsd.org/cgi-bin/cvsweb/src/usr.sbin

bgpd

  ~ rde.h                                 ~ rde_attr.c

  > Implement a attr_writebuf() function that works on a struct buf instead of
  > a pre allocated piece of memory. Will be used by newer mrt code.
  > OK henning@ (claudio@)

  ~ mrt.c                                 ~ mrt.h
  ~ session.c                             

  > First big chunk of mrt rewrite. Simplifies code hopefully and fixes table
  > dumps that were wrong because of the 4byte AS support. Dumps I took seemed
  > to work so far. "Put it in" henning@ (claudio@)

smtpd

  ~ mta.c                                 

  > in the event of a timeout, mxhost is passed to TAILQ_REMOVE before it is
  > initialized.
  > spotted and diff by Matthew Haub <matthew.haub at alumni.adelaide.edu.au>
  > (gilles@)

  ~ parse.y                               

  > when listen was declared without a port parameter, it would get it wrong
  > because of a missing htons() (gilles@)

  ~ smtpd.h                               

  > make action_type == 0 mean A_INVALID, not A_RELAY; ok gilles@ (jacekm@)

  ~ makemap.c                             

  > since maps may contain secrets, carry ownership and perms from source file
  > to db file; ok gilles@ (jacekm@)

===============================================================================