[odc] Daily src changes for 2007-09-03

ODC auto at squish.net
Tue Sep 4 07:00:01 BST 2007


OpenBSD src changes summary for 2007-09-03
==========================================

bin/ed                                  bin/ksh
etc/changelist                          gnu/egcs
gnu/lib/libstdc++                       gnu/usr.bin/gcc
include/stdlib.h                        include/string.h
lib/csu                                 lib/libc
lib/libusbhid                           regress/lib
sbin/ifconfig                           sbin/init
sys/arch/alpha/alpha                    sys/arch/amd64/amd64
sys/arch/i386/i386                      sys/arch/luna88k/include
sys/arch/macppc/macppc                  sys/arch/mips64/mips64
sys/arch/mvme88k/mvme88k                sys/arch/mvmeppc/mvmeppc
sys/arch/sparc64/include                sys/arch/sparc64/sparc64
sys/arch/vax/vax                        sys/dev/i2c
sys/dev/pci                             sys/kern
sys/net                                 sys/netinet
sys/sys                                 usr.bin/make
usr.bin/netstat                         usr.bin/pr
usr.bin/sudo                            usr.bin/yacc
usr.sbin/pppoe                          usr.sbin/pstat

== bin =============================================================== 01/10 ==

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

ed

  ~ glbl.c                                

  > malloc(n * m) -> calloc(n, m); ok espie (deraadt@)

ksh

  ~ c_sh.c                                

  > Unsetting a non-existent variable is not an error. See
  > http://www.opengroup.org/onlinepubs/009695399/utilities/unset.html
  > report from Arkadiusz Miskiewicz; fixed based on
  > http://cvs.pld-linux.org diff; ok millert@ fgsch@ (otto@)

== etc =============================================================== 02/10 ==

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

changelist

  ~ changelist                            

  > Only send hash of /etc/ipsec.conf if it changes since it can contain
  > a pre-shared key
  > ok hshoexer@ cloder@  (some time ago) (joel@)

== gnu =============================================================== 03/10 ==

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

egcs

  ~ libstdc++/shlib_version               

  > crank libc and libstdc++ majors (millert@)

lib/libstdc++

  ~ shlib_version                         

  > crank libc and libstdc++ majors (millert@)

usr.bin/gcc

  ~ Makefile.bsd-wrapper                  

  > Add __cxa_atexit() support for gcc3.  This provides support for shared
  > object destructors called at dlclose() time.  Inspired by similar changes
  > in FreeBSD and NetBSD. (millert@)

== include =========================================================== 04/10 ==

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

stdlib.h

  ~ stdlib.h                              

  > add recaloc(3) (millert@)

string.h

  ~ string.h                              

  > add memrchr(3) (millert@)

== lib =============================================================== 05/10 ==

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

csu

  ~ common_elf/crtbegin.c                 ~ common_elf/crtbeginS.c

  > Add __cxa_atexit() support for gcc3.  This provides support for shared
  > object destructors called at dlclose() time.  Inspired by similar changes
  > in FreeBSD and NetBSD. (millert@)

libc

  ~ stdlib/malloc.3                       ~ stdlib/malloc.c
  ~ stdlib/Makefile.inc                   

  > add recaloc(3) (millert@)

  ~ string/memchr.3                       ~ string/Makefile.inc
  + string/memrchr.c                      

  > add memrchr(3) (millert@)

  ~ sys/intro.2                           ~ sys/semop.2
  ~ sys/msgsnd.2                          ~ sys/msgrcv.2
  ~ gen/errlist.c                         

  > Add ENOMSG and EIDRM; from jsg@ (millert@)

  ~ stdlib/abort.c                        ~ stdlib/atexit.c
  ~ stdlib/atexit.h                       ~ stdlib/exit.c

  > Add __cxa_atexit() support for gcc3.  This provides support for shared
  > object destructors called at dlclose() time.  Inspired by similar changes
  > in FreeBSD and NetBSD. (millert@)

  ~ shlib_version                         

  > crank libc and libstdc++ majors (millert@)

  ~ string/memchr.3                       

  > remove trailing whitespace; (jmc@)

  ~ stdlib/malloc.3                       

  > Update free(3) section--we use mmap()/munmap(), not brk().
  > Based on a diff from Mike Belopuhov.  OK jmc@ (millert@)

libusbhid

  ~ usage.c                               

  > malloc(n * m) -> calloc(n, m); ok espie (deraadt@)

== regress =========================================================== 06/10 ==

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

lib

  ~ libc/Makefile                         ~ libc/atexit/atexit_test.c
  + libc/cxa-atexit/Makefile              + libc/cxa-atexit/libgd1/Makefile
  + libc/cxa-atexit/libgd1/gd1.C          
  + libc/cxa-atexit/libgd1/shlib_version
  + libc/cxa-atexit/libgd2/Makefile       + libc/cxa-atexit/libgd2/gd2.C
  + libc/cxa-atexit/libgd2/shlib_version
  + libc/cxa-atexit/test1/Makefile        + libc/cxa-atexit/test1/test1.C

  > Adapt atexit() regress to recent changes and add __cxa_atexit() regress.
  > __cxa_atexit() regress from kurt@ (millert@)

== sbin ============================================================== 07/10 ==

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

ifconfig

  ~ ifconfig.8                            

  > Typos from miod. 'specificed' -> 'specified' in some comments and man page.
  > (krw@)

init

  ~ init.c                                

  > malloc(n * m) -> calloc(n, m); ok espie (deraadt@)

== sys =============================================================== 08/10 ==

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

arch/alpha/alpha

  ~ vm_machdep.c                          

  > Typos from miod. 'specificed' -> 'specified' in some comments and man page.
  > (krw@)

  ~ pmap.c                                

  > Remove the pv and pte stealing code. Neither art@ and I like it, the
  > theoretical gain of it is close to zero, and our moms told us stealing is
  > bad. (miod@)

arch/amd64/amd64

  ~ bus_dma.c                             

  > Typos from miod. 'functin' -> 'functin' in some comments. (krw@)

arch/i386/i386

  ~ machdep.c                             

  > Typos from miod. 'functin' -> 'functin' in some comments. (krw@)

arch/luna88k/include

  ~ board.h                               

  > Typos from miod. 'sfotware' -> 'software' in some comments. (krw@)

arch/macppc/macppc

  ~ dma.c                                 

  > Typos from miod. 'functin' -> 'functin' in some comments. (krw@)

arch/mips64/mips64

  ~ db_machdep.c                          ~ exception.S

  > Typos from miod. 'requiered' -> 'required' in some comments. (krw@)

  ~ vm_machdep.c                          

  > call savectx() in cpu_fork(), so that the new process can take signals
  > correctly in child_return()'s userret(); makes the sys/kern/signal/earlysig
  > regression test pass. From a 2.5 years old diff I should have commited much
  > earlier. (miod@)

arch/mvme88k/mvme88k

  ~ bus_dma.c                             

  > Typos from miod. 'functin' -> 'functin' in some comments. (krw@)

arch/mvmeppc/mvmeppc

  ~ bus_dma.c                             

  > Typos from miod. 'functin' -> 'functin' in some comments. (krw@)

arch/sparc64/include

  ~ ctlreg.h                              

  > Fix a few interrupt dispatch/receive register definitions. (kettenis@)

arch/sparc64/sparc64

  ~ machdep.c                             

  > Typos from miod. 'functin' -> 'functin' in some comments. (krw@)

arch/vax/vax

  ~ bus_dma.c                             

  > Typos from miod. 'functin' -> 'functin' in some comments. (krw@)

dev/i2c

  ~ i2c_scan.c                            

  > Recognize Andigilog aSC7621.
  > ok deraadt@ (kettenis@)

  ~ i2c_scan.c                            

  > there is a prototype version of the lm93 (plus spacing fixes) (deraadt@)

dev/pci

  ~ if_devar.h                            ~ if_lmcvar.h

  > Typos from miod. 'intrrupt' -> 'interrupt' in some comments. (krw@)

  ~ pcidevs                               

  > Add some more PCI ids for mpi; pointed out by Stephan Eisvogel
  > <eisvogel at embinet dot de> (marco@)

  ~ pcidevs.h                             ~ pcidevs_data.h

  > sync (marco@)

  ~ mpi_pci.c                             

  > Add more PCI ids; taken from the linux driver.  Pointed out by Stephan
  > Eisvogel <eisvogel at embinet dot de> (marco@)

  ~ mpi_pci.c                             

  > unbreak. (mbalmer@)

  ~ cmpci.c                               

  > __inline removal
  > tested on my amd64
  > from bret.lambert at gmail.com via art@, thanks (jakemsr@)

kern

  ~ subr_log.c                            

  > unsupported ioctl's should be returned with ENOTTY
  > not -1;
  > ok miod@ (thib@)

net

  ~ if_pfsync.c                           

  > Make use of the pfsync 'badval' and 'stale' counters instead of using
  > 'badstate' everywhere.
  > ok henning@ (joel@)

  ~ if.h                                  ~ route.h
  ~ rtsock.c                              

  > Bump RTM_VERSION to 4 and start a new aera of routing in OpenBSD :)
  > Changes include 64bit counters instead of u_long, routing table id in the
  > header
  > of most messages, reserved routing priority field, added a hdrlen field to
  > skip
  > over the header so that binary compatibility becomes easier.
  > A minimal backward support for old binaries is included to ease upgrades
  > but
  > don't expect anything more than ifconfig, route and dhclient to correctly
  > work.
  > OK henning@ mglocker@ (claudio@)

netinet

  ~ ip_carp.c                             ~ ip_carp.h

  > Make use of the carp preempt counter to signal number of transitions of
  > any carp group to master status.
  > ok dhartmei@ (joel@)

sys

  ~ errno.h                               

  > Add ENOMSG and EIDRM; from jsg@ (millert@)

== usr.bin =========================================================== 09/10 ==

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

make

  ~ generate.c                            

  > use calloc() for the sake of regularity, prompted by deraadt@
  > Note that I *still* fill every pointer with explicit NULL, for portability.
  > filling with 0 bytes is not necessarily the same as NULL pointers. (espie@)

netstat

  ~ inet.c                                

  > Make use of the carp preempt counter to signal number of transitions of
  > any carp group to master status.
  > ok dhartmei@ (joel@)

pr

  ~ pr.c                                  

  > sizeof char is 1; ok djm (deraadt@)

  ~ pr.c                                  

  > last commit broke the tree because of missing parentheses.
  > ok henning@ (moritz@)

sudo

  ~ sudoers                               

  > - break "Defaults env_keep" into several lines so it is more readable
  > - add OWNER and GROUP to env_keep so that make-plist creates pkg/P* with
  > the correct uid/gid (hopefully this would be the last ports related change)
  > ok millert@ (ajacoutot@)

  ~ config.h                              ~ sudo/Makefile

  > OpenBSD now has memrchr(3) (millert@)

yacc

  ~ skeleton.c                            

  > move back to using malloc() instead of calloc(), because the yacc
  > skeleton really should only call malloc/realloc/free, no other external
  > APIs at all.  theefore, add a pre-check for the overflow case, thus
  > protecting realloc too; tested mblamer, ok millert, help from kettenis
  > (deraadt@)

== usr.sbin ========================================================== 10/10 ==

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

pppoe

  ~ client.c                              ~ server.c

  > malloc(n * m) -> calloc(n, m); ok espie (deraadt@)

pstat

  ~ pstat.c                               

  > malloc(n * m) -> calloc(n, m); ok espie (deraadt@)

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


More information about the odc mailing list