OpenBSD src changes summary for 2016-01-18
==========================================

bin/ls                                  distrib/special
games/morse                             lib/libc
lib/libtls                              sbin/pdisk
sys/dev/pv                              sys/kern
sys/net                                 usr.bin/colrm
usr.bin/sndiod                          usr.bin/ul
usr.sbin/smtpd                          

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

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

ls

  ~ utf8.c                                

  > Fix a regression (and POSIX violation) introduced with UTF-8 support:
  > When neither running on a terminal nor with -q, names must be passed
  > through as they are, nothing must be replaced with question marks.
  > Effectively, -q was always in effect.  SMALL was not affected.
  > Triggered by a different patch from Martijn van Duren <openbsd plus tech
  > at list dot imperialat dot at>, who confirmed that this version is better.
  > Identified as a regression by tedu@.
  > OK sthen@. (schwarze@)

== distrib =========================================================== 02/08 ==

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

special

  ~ pdisk/Makefile                        

  > sync to main (deraadt@)

== games ============================================================= 03/08 ==

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

morse

  ~ morse.6                               

  > Add a reference to the current ITU Morse code document.
  > Part of a diff that was ok abieber tb -.. . .-. .- .- -.. -;
  > other part being revised with pjanzen. (sthen@)

  ~ morse.c                               

  > There are now distinct codes for left and right parentheses.
  > Part of a diff that was ok tb@ abieber@ deraadt@ (sthen@)

== lib =============================================================== 04/08 ==

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

libc

  ~ sys/pledge.2                          

  > adjust some syscalls in proc & id; from Carlin Bingham (deraadt@)

libtls

  ~ tls.c                                 

  > Call BIO_sock_init() from tls_init() to ensure sockets are enabled on
  > Windows.
  > This is of course a no-op on other platforms.  Noted by equalsraf from
  > github. (bcook@)

== sbin ============================================================== 05/08 ==

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

pdisk

  ~ file_media.c                          ~ partition_map.c
  ~ partition_map.h                       ~ pdisk.c

  > Remove a grab bag of unused #define's, fields, enum's, variables,
  > functions. (krw@)

  ~ pdisk.c                               

  > Fold get_options() into main() and eliminate unneeded nFLAG_DEFAULT
  > #defines. Zero is a well known and acceptable number. (krw@)

  ~ pdisk.c                               

  > Run indent(1) over it and tweak the results a bit manually to look more
  > KNF'ish. (krw@)

  ~ validate.c                            

  > Run indent(1) over it and tweak the results a bit manually to look more
  > KNF'ish. (krw@)

  ~ partition_map.c                       

  > Run indent(1) over it and tweak the results a bit manually to look more
  > KNF'ish. (krw@)

  ~ convert.c                             

  > Run indent(1) over it and tweak the results a bit manually to look more
  > KNF'ish. (krw@)

  ~ file_media.c                          

  > Run indent(1) over it and tweak the results a bit manually to look more
  > KNF'ish. (krw@)

  ~ io.c                                  

  > Run indent(1) over it and tweak the results a bit manually to look more
  > KNF'ish. (krw@)

  ~ dump.c                                

  > Run indent(1) over it and tweak the results a bit manually to look more
  > KNF'ish. (krw@)

  ~ file_media.c                          ~ file_media.h
  ~ partition_map.c                       

  > os_reload_file_media() doesn't do anything except tell you if you passed it
  > a NULL pointer. And the result is ignored. Nuke it. This removes the only
  > use of the 'regular_file' field of struct file_media. So nuke that too.
  > (krw@)

  ~ file_media.c                          ~ file_media.h

  > Discard a whack of double and triple checking lseek/read/write parameters.
  > Just use pread/pwrite and report error emanating from them. No more
  > uses of size_in_byte field of struct file_media, so toss that too. (krw@)

  ~ dump.c                                ~ file_media.c
  ~ file_media.h                          ~ partition_map.c
  ~ partition_map.h                       ~ pdisk.c
  ~ validate.c                            

  > struct file_media had 1 field left. And one place where struct
  > file_media was actually stored. So move the last field (fd) into
  > the place formerly used to store pointer to the instance. As a
  > result we can just pass fd's around rather than pointers to a struct
  > containing a fd.
  > close_file_media() becomes empty but for a close(). So just use
  > close() and nuke close_file_media(). (krw@)

  ~ dump.c                                ~ dump.h
  ~ pdisk.c                               

  > The contents of the Apple_Patches partition is none of our business.
  > Stop pretending we care or know what's in there. Remove 'debugging'
  > mode command to examine the partition contents and the supporting
  > code.
  > ok deraadt@ jasper@ (krw@)

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

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

dev/pv

  ~ xenstore.c                            

  > Fixup a hang while performing a read operation on XenStore
  > Reyk has reported an issue that turned out to be an incorrect calculation
  > of bytes available for reading.  This diff also places missed semaphore
  > releases in the xs_start and makes 'left' a size_t in xs_ring_{get,put}.
  > (mikeb@)

  ~ xen.c                                 ~ xenreg.h
  ~ xenvar.h                              

  > Provide a Xen v3 API compatible fallback for event channel hypercalls
  > (mikeb@)

  ~ xen.c                                 

  > Log unhandled interrupts (mikeb@)

kern

  ~ kern_pledge.c                         

  > Add SYS_truncate as a "wpath" operation.  Omission noticed by tb and sthen.
  > (deraadt@)

net

  ~ route.h                               ~ rtable.c
  ~ rtable.h                              

  > Stop storing a backpointer to the corresponding ART node in each route
  > entry.
  > This pointer hasn't been used for some time and without it no external
  > reference count is needed to turn art_lookup() mpsafe. (mpi@)

  ~ art.c                                 ~ art.h
  ~ rtable.c                              

  > Pass the address length to art_alloc() and remove the hack abusing the
  > offset of the address in the sockaddr to initialize the stride lengths.
  > (mpi@)

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

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

colrm

  ~ colrm.1                               ~ colrm.c

  > UTF-8 support:
  > Cut by display columns rather than by character positions because
  > the latter would be useless in the presence of combining zero-width
  > characters.  Similarly, let tab advance to display columns rather
  > than to character positions.
  > For compatibility with nroff and man(1) output, let backspace back
  > up one character rather than on display column.  But for compatibility
  > with POSIX fold(1), *if* two backspaces follow a double-width character,
  > ignore the second one.
  > Fix some bugs while here:  Delete backspaces that immediately follow
  > deleted characters.  Expand tabs intersecting deletions, such that
  > part of the blanks can be removed.  Expand tabs following deletions,
  > or they would no longer align with adjacent lines without tabs.
  > OK jmc@ on a previous version of the manual.
  > No opposition when shown on tech@. (schwarze@)

sndiod

  ~ sndiod.8                              

  > Mention that "sndiod -d" doesn't daemonize and that
  > the -d flag can be specified multiple times. Diff
  > from Michael Reed <m.reed at mykolab.com>. Many thanks. (ratchov@)

  ~ file.c                                

  > Allow time differences between two clock_gettime() calls to
  > be up to 60s without logging a warning. (ratchov@)

ul

  ~ ul.1                                  ~ ul.c

  > UTF-8 support:
  > This is the first example of a program where doing the full char *
  > to wchar_t * to char * double conversion is actually simpler than
  > our usual approach of working with char * throughout.  In part
  > inspired by the FreeBSD version which is in turn based on Bruno
  > Haible's work in util-linux, but not sharing any code and avoiding
  > the almost half a dozen bugs that FreeBSD has.
  > As a bonus reimplement overstrike() and iattr() almost from scratch,
  > getting rid of useless malloc(3)ed local buffers.
  > Add lots of missing information to the manual.
  > No opposition when shown on tech@. (schwarze@)

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

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

smtpd

  ~ smtpd.conf.5                          

  > unify and add some more smtpd(8) references
  > ok sunil gilles eric (jung@)

  ~ parse.y                               

  > remove leftovers of (gone) curve option
  > ok gilles (jung@)

  ~ smtpd.c                               

  > Load pki keys before daemon(3). Keys with passphrase require access
  > to stdin.
  > Ok millert@ jung@ gilles@ (sunil@)

  ~ smtpd.conf.5                          

  > minor tweaks: add single quote to @ char, add .Ic markup to forward-only,
  > fix end of list marker, and add an 'are'
  > ok millert (to an earlier version without end of list fix and without
  > 'are') (jung@)

===============================================================================
_______________________________________________
odc mailing list
[email protected]
http://www.squish.net/mailman/listinfo/odc

Reply via email to