[odc] Daily src changes for 2002-06-21

ODC auto at squish.net
Sat Jun 22 08:00:16 BST 2002


OpenBSD src changes summary for 2002-06-21
==========================================

bin/systrace                            distrib/sets
lib/libutil                             libexec/talkd
share/man                               sys/arch/i386/stand/boot
sys/arch/macppc/macppc                  sys/arch/sparc64/conf
sys/dev                                 sys/dev/ic
sys/dev/pci                             sys/dev/pcmcia
sys/dev/sbus                            sys/dev/usb
sys/sys                                 sys/ufs/mfs
usr.bin/mg                              usr.bin/ssh
usr.bin/talk                            usr.sbin
usr.sbin/cron                           usr.sbin/dhcp
usr.sbin/ifmcstat                       usr.sbin/inetd

== bin =============================================================== 01/08 ==

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

systrace

  ~ systrace.1                            

  > change BUGS section to mention that only clone is a problem now.

  ~ intercept-translate.c                 ~ intercept.c
  ~ intercept.h                           ~ openbsd-syscalls.c
  ~ systrace.c                            

  > rewrite all system call arguments in the permit case.  use realpath
  > when we still have the root and we of the monitored process.  this
  > eliminates almost all race coniditions.

  ~ systrace.1                            

  > No .Pp before .Sh; from wiz at danbala.ifoer.tuwien.ac.at
== distrib =========================================================== 02/08 ==

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

sets

  ~ lists/base/mi                         ~ lists/man/mi

  > sync

== lib =============================================================== 03/08 ==

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

libutil

  ~ login_fbtab.3                         ~ login_fbtab.c
  ~ util.h                                

  > login_fbtab(3) fixes:
  > o make first arg const since we don't modify it
  > o use strsep() instead of strtok() as strtok() changes internal state
  > o add some bounds checking and use strlcat() instead of pointer arithmetic
  > o ANSI function headers
  > Originally based on a patch from Lars J. Buitinck but much modified.

== libexec =========================================================== 04/08 ==

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

talkd

  ~ talkd.c                               

  > o Use *_FILENO defines instead of numbers
  > o Kill sockt global, it is not needed
  > o Use STDOUT_FILENO as the 1st arg to sendto().
  > Previously we used sockt in sendto() which only worked because a) globals
  > are intialized to 0 and b) inetd sets up fds 0-2 dup'd to the same thing
  > so it didn't matter that we were sending to stdin, not stdout.

== share ============================================================= 05/08 ==

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

man

  + man9/hash.9                           + man9/hashinit.9
  ~ man9/Makefile                         

  > Add hash(9) and hashinit(9).

  ~ man4/wi.4tbl                          

  > add Netgear MA401RA
  > ok millert@

  ~ man9/hash.9                           ~ man9/hashinit.9

  > Alphabetise.  From Mike Pechkin.

== sys =============================================================== 06/08 ==

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

arch/i386/stand/boot

  ~ conf.c                                

  > Bump version to help distinquish new memory probe and fixes.

arch/macppc/macppc

  ~ cpu.c                                 

  > identify 0x7000 (IBM750FX).  snprintf.
  > do we need "IBM" or "Motorola" in cpu_model?

  ~ cpu.c                                 

  > configure HID0, and show cache status for IBM750FX.  from netbsd

arch/sparc64/conf

  ~ GENERIC                               

  > sbus(4) -> stp(4) -> pcmcia(4) -> wi(4).  cool shitz done by crazy fgsch

dev

  ~ systrace.c                            ~ systrace.h

  > system call argument rewriting framework via stackgap; okay art@

dev/ic

  ~ if_wireg.h                            

  > ugly kludge to deal correctly with endianess in sparc and sparc64.
  > millert@ ok.

  ~ if_wi.c                               ~ if_wi_hostap.c

  > Add some missing casts of sc->wi_txbuf -> caddr_t

  ~ if_wi.c                               

  > Don't set WI_FLAGS_ATTACHED in sc->wi_flags until after mi attach
  > routines have been called.  Avoids a panic when you insert & eject
  > a card very quickly on a slow machine.

  ~ if_wivar.h                            

  > put the objects in a better order, so that they align correctly.
  > whoever made struct ieee80211_nwid an odd-sized structure and created
  > this particular little corner of alignment hell surely deserves a boot
  > to the head.

  ~ if_wi.c                               

  > don't convert these to little-endian. now ifconfig shows the nwid on
  > sparc64 and macppc; tested by me and drahn@ on i386, macppc and sparc64.

  ~ if_wi.c                               

  > Clear wi_hostap_info.apflags in wi_attach() to guarantee wihap_shutdown()
  > doesn't do the wrong thing with quick card insert/ejects.

dev/pci

  ~ noct.c                                ~ noctreg.h
  ~ noctvar.h                             

  > skeleton for queue handling on the pk side of the chip.  For now it just
  > enqueues nops and handles them.

  ~ pcidevs.h                             ~ pcidevs_data.h

  > regen

  ~ pcidevs                               

  > add some ICH4 devices; ok deraadt@

  ~ noct.c                                ~ noctreg.h
  ~ noctvar.h                             

  > skeleton queue handling for the enc/auth piece of the chip (fed with nops v
  > ia
  > timeout for now).

dev/pcmcia

  ~ pcmciadevs.h                          

  > regen.

  ~ if_wi_pcmcia.c                        ~ pcmciadevs
  ~ pcmciadevs.h                          

  > add support for Netgear MA401RA (Prism-2.5)

dev/sbus

  ~ stp4020reg.h                          

  > missing tab.

  ~ stp4020.c                             

  > make this more quiet on card insertion.

dev/usb

  ~ umass.c                               

  > Trial and error gives me this tweak for the trumpion 0x1100 device.

  ~ usbdevs.h                             ~ usbdevs_data.h

  > sync

  ~ usbdevs                               

  > Add Trumpion Microelectronics (Who comes up with those names?)
  > and a product that I actually don't know what it is (could be one of
  > three possible chips).

sys

  ~ malloc.h                              

  > tabify

ufs/mfs

  ~ mfs_vfsops.c                          

  > Remove pointless macro.

== usr.bin =========================================================== 07/08 ==

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

mg

  ~ def.h                                 

  > remove the "extern" for a variable that does not exist anymore.

  ~ extend.c                              ~ fileio.c
  ~ help.c                                

  > fix bad usage of strlcpy()'s return value. (made pointers point
  > beyond the boundaries of buffers)
  > ok deraadt@

ssh

  ~ monitor.c                             

  > Don't initialise compression buffers when compression=no in sshd_config;
  > ok Niels@

  ~ version.h                             

  > OpenSSH_3.3

talk

  ~ get_names.c                           ~ init_disp.c
  ~ invite.c                              ~ io.c
  ~ msgs.c                                ~ talk.h

  > o Resize nicely when we receive SIGWINCH
  > o Also change some 0-2 to *_FILENO

== usr.sbin ========================================================== 08/08 ==

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

usr.sbin

  ~ Makefile                              - ifmcstat/Makefile
  - ifmcstat/ifmcstat.8                   - ifmcstat/ifmcstat.c
  - ifmcstat/Makefile                     - ifmcstat/ifmcstat.8
  - ifmcstat/ifmcstat.c                   

  > bye bye, use netstat -inalv; itojun ok

cron

  ~ do_command.c                          

  > Correctly close the writing end of the pipe when dup'ing the fd to
  > stdout.  This was a pasto/thinko on my part.  Problem found by
  > jean-philippe wan-hoi.

dhcp

  ~ common/tree.c                         ~ relay/dhcrelay.c
  ~ statmsg/main.c                        ~ sysconfd/sysconfd.c

  > more strlcpy

ifmcstat

  - Makefile                              - ifmcstat.8
  - ifmcstat.c                            - Makefile
  - ifmcstat.8                            - ifmcstat.c

  > bye bye, use netstat -inalv; itojun ok

inetd

  ~ inetd.c                               

  > o change some 0-2 to *_FILENO
  > o don't #ifdef RLIMIT_NOFILE
  > o dup2() safety
  > o some NULL vs. 0 changes
  > deraadt@ OK

  ~ inetd.c                               

  > kill gratuitous hex; millert ok

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




More information about the odc mailing list