[odc] Daily src changes for 2008-05-21

ODC auto at squish.net
Thu May 22 07:00:02 BST 2008


OpenBSD src changes summary for 2008-05-21
==========================================

distrib/i386                            distrib/notes
distrib/sets                            lib/libc
lib/libm                                lib/librthread
share/man                               sys/arch/arm/include
sys/arch/armish/include                 sys/arch/hppa/hppa
sys/arch/hppa64/hppa64                  sys/arch/i386/conf
sys/arch/i386/i386                      sys/arch/i386/isa
sys/arch/macppc/include                 sys/arch/mvmeppc/include
sys/arch/powerpc/include                sys/arch/powerpc/powerpc
sys/arch/sh/include                     sys/arch/sh/sh
sys/arch/solbourne/solbourne            sys/arch/sparc/sparc
sys/arch/sparc64/conf                   sys/arch/sparc64/include
sys/arch/sparc64/sparc64                sys/arch/vax/boot/boot
sys/arch/vax/boot/common                sys/arch/vax/include
sys/arch/vax/vax                        sys/arch/zaurus/include
sys/dev/cardbus                         sys/dev/i2c
sys/dev/ic                              sys/dev/isa
sys/dev/pci                             sys/dev/pcmcia
sys/dev/puc                             sys/net
usr.bin                                 usr.bin/elf2aout
usr.bin/elf2ecoff                       usr.sbin/afs
usr.sbin/httpd                          usr.sbin/ppp

== distrib =========================================================== 01/06 ==

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

i386

  ~ common/install.md                     

  > pccom is dead. long live com. Bring install script into the new world.
  > (krw@)

notes

  ~ amd64/hardware                        ~ i386/hardware

  > Bye bye pccom (miod@)

sets

  ~ lists/man/mi                          

  > sync (deraadt@)

  ~ lists/base/md.armish                  ~ lists/base/md.macppc
  ~ lists/base/md.mvmeppc                 ~ lists/base/md.socppc
  ~ lists/base/md.sparc64                 ~ lists/base/md.zaurus
  ~ lists/comp/md.armish                  ~ lists/comp/md.macppc
  ~ lists/comp/md.mvmeppc                 ~ lists/comp/md.socppc
  ~ lists/comp/md.sparc64                 ~ lists/comp/md.zaurus

  > Remove <machine/elf_abi.h> on platforms which have it. Nothing in the tree
  > needs the few MD definitions they (sometimes) provide; only binutils is
  > interested in them, but binutils provide their own include files for that
  > purpose anyway.
  > ok deraadt@ kettenis@ (miod@)

== lib =============================================================== 02/06 ==

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

libc

  ~ arch/vax/sys/sigreturn.S              

  > Use _PROF_PROLOGUE insteaf of rolling our own. (miod@)

  ~ arch/vax/SYS.h                        ~ arch/vax/gen/_setjmp.S
  ~ arch/vax/gen/ldexp.S                  ~ arch/vax/gen/sigsetjmp.S
  ~ arch/vax/sys/Ovfork.S                 ~ arch/vax/sys/brk.S
  ~ arch/vax/sys/cerror.S                 ~ arch/vax/sys/exect.S
  ~ arch/vax/sys/sbrk.S                   ~ arch/vax/sys/sigprocmask.S
  ~ arch/vax/sys/sigsuspend.S             ~ arch/vax/sys/syscall.S

  > Use _C_LABEL when appropriate. (miod@)

libm

  ~ arch/vax/n_argred.S                   ~ arch/vax/n_atan2.S
  ~ arch/vax/n_cabs.S                     ~ arch/vax/n_cbrt.S
  ~ arch/vax/n_infnan.S                   ~ arch/vax/n_sincos.S
  ~ arch/vax/n_sqrt.S                     ~ arch/vax/n_support.S
  ~ arch/vax/n_tan.S                      

  > Use proper *ENTRY macros with symbolic register save masks, sprinkle
  > _C_LABEL and _ASM_LABEL, and add $OpenBSD$ RCS Ids.
  > Include <sys/errno.h> and use its values directly, instead of using .set
  > directives (which grow the individual .o files even)
  > Also, replace `$' in internal (but non static) symbols with `_', and prefix
  > those with `__'. (miod@)

librthread

  ~ arch/vax/rfork_thread.S               

  > Use _C_LABEL when appropriate. (miod@)

== share ============================================================= 03/06 ==

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

man

  ~ man4/Makefile                         ~ man4/addcom.4
  ~ man4/ast.4                            ~ man4/boca.4
  ~ man4/cardbus.4                        ~ man4/com.4
  ~ man4/hsq.4                            ~ man4/isa.4
  ~ man4/isapnp.4                         ~ man4/pcmcia.4
  ~ man4/puc.4                            ~ man4/rtfps.4

  > Bye, bye pccom(4). (kettenis@)

  ~ man4/com.4                            

  > Update sgi synopsis, add armish and socppc. (miod@)

== sys =============================================================== 04/06 ==

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

arch/arm/include

  - elf_abi.h                             

  > Remove <machine/elf_abi.h> on platforms which have it. Nothing in the tree
  > needs the few MD definitions they (sometimes) provide; only binutils is
  > interested in them, but binutils provide their own include files for that
  > purpose anyway.
  > ok deraadt@ kettenis@ (miod@)

arch/armish/include

  - elf_abi.h                             

  > Remove <machine/elf_abi.h> on platforms which have it. Nothing in the tree
  > needs the few MD definitions they (sometimes) provide; only binutils is
  > interested in them, but binutils provide their own include files for that
  > purpose anyway.
  > ok deraadt@ kettenis@ (miod@)

arch/hppa/hppa

  ~ locore.S                              

  > ddb expects the kernel longjmp() to only take a single parameter and always
  > return 1 since 12+ years, it's about time to fix the offending ports.
  > Reported by Pierre Riteau (firstname.lastname at gmail) (miod@)

arch/hppa64/hppa64

  ~ locore.S                              

  > ddb expects the kernel longjmp() to only take a single parameter and always
  > return 1 since 12+ years, it's about time to fix the offending ports.
  > Reported by Pierre Riteau (firstname.lastname at gmail) (miod@)

arch/i386/conf

  ~ GENERIC                               ~ RAMDISK
  ~ RAMDISKB                              ~ RAMDISKC
  ~ RAMDISK_CD                            ~ files.i386

  > Switch i386 from pccom to com.  Welcomed by many.
  > ok dlg@, jsing@, deraadt@ (kettenis@)

arch/i386/i386

  ~ apicvec.s                             ~ bios.c
  ~ conf.c                                ~ machdep.c

  > Switch i386 from pccom to com.  Welcomed by many.
  > ok dlg@, jsing@, deraadt@ (kettenis@)

arch/i386/isa

  ~ icu.s                                 

  > Switch i386 from pccom to com.  Welcomed by many.
  > ok dlg@, jsing@, deraadt@ (kettenis@)

arch/macppc/include

  - elf_abi.h                             

  > Remove <machine/elf_abi.h> on platforms which have it. Nothing in the tree
  > needs the few MD definitions they (sometimes) provide; only binutils is
  > interested in them, but binutils provide their own include files for that
  > purpose anyway.
  > ok deraadt@ kettenis@ (miod@)

arch/mvmeppc/include

  - elf_abi.h                             

  > Remove <machine/elf_abi.h> on platforms which have it. Nothing in the tree
  > needs the few MD definitions they (sometimes) provide; only binutils is
  > interested in them, but binutils provide their own include files for that
  > purpose anyway.
  > ok deraadt@ kettenis@ (miod@)

arch/powerpc/include

  - elf_abi.h                             

  > Remove <machine/elf_abi.h> on platforms which have it. Nothing in the tree
  > needs the few MD definitions they (sometimes) provide; only binutils is
  > interested in them, but binutils provide their own include files for that
  > purpose anyway.
  > ok deraadt@ kettenis@ (miod@)

arch/powerpc/powerpc

  ~ setjmp.S                              

  > ddb expects the kernel longjmp() to only take a single parameter and always
  > return 1 since 12+ years, it's about time to fix the offending ports.
  > Reported by Pierre Riteau (firstname.lastname at gmail) (miod@)

arch/sh/include

  ~ cpu.h                                 

  > Not all cache operations need to be run from P2, so don't do this unless
  > necessary. Also, let the P2 functions return to P1 addresses, instead of
  > jumping to their own P1 image before returning. This gives a ~15% speedup.
  > From NetBSD, thanks uwe at netbsd for spotting this in the sh4 docs! (miod@)

arch/sh/sh

  ~ cache_sh4.c                           

  > Not all cache operations need to be run from P2, so don't do this unless
  > necessary. Also, let the P2 functions return to P1 addresses, instead of
  > jumping to their own P1 image before returning. This gives a ~15% speedup.
  > From NetBSD, thanks uwe at netbsd for spotting this in the sh4 docs! (miod@)

arch/solbourne/solbourne

  ~ locore.s                              

  > ddb expects the kernel longjmp() to only take a single parameter and always
  > return 1 since 12+ years, it's about time to fix the offending ports.
  > Reported by Pierre Riteau (firstname.lastname at gmail) (miod@)

arch/sparc/sparc

  ~ locore.s                              

  > ddb expects the kernel longjmp() to only take a single parameter and always
  > return 1 since 12+ years, it's about time to fix the offending ports.
  > Reported by Pierre Riteau (firstname.lastname at gmail) (miod@)

arch/sparc64/conf

  ~ GENERIC                               ~ RAMDISK

  > First step towards supporting sun4us machines with Fujitsu's SPARC64-V CPU.
  > (kettenis@)

arch/sparc64/include

  ~ param.h                               

  > First step towards supporting sun4us machines with Fujitsu's SPARC64-V CPU.
  > (kettenis@)

  - elf_abi.h                             

  > Remove <machine/elf_abi.h> on platforms which have it. Nothing in the tree
  > needs the few MD definitions they (sometimes) provide; only binutils is
  > interested in them, but binutils provide their own include files for that
  > purpose anyway.
  > ok deraadt@ kettenis@ (miod@)

arch/sparc64/sparc64

  ~ autoconf.c                            ~ cpu.c
  ~ pmap.c                                

  > First step towards supporting sun4us machines with Fujitsu's SPARC64-V CPU.
  > (kettenis@)

  ~ locore.s                              

  > ddb expects the kernel longjmp() to only take a single parameter and always
  > return 1 since 12+ years, it's about time to fix the offending ports.
  > Reported by Pierre Riteau (firstname.lastname at gmail) (miod@)

arch/vax/boot/boot

  ~ autoconf.c                            

  > ddb expects the kernel longjmp() to only take a single parameter and always
  > return 1 since 12+ years, it's about time to fix the offending ports.
  > Reported by Pierre Riteau (firstname.lastname at gmail) (miod@)

arch/vax/boot/common

  ~ str.S                                 

  > ddb expects the kernel longjmp() to only take a single parameter and always
  > return 1 since 12+ years, it's about time to fix the offending ports.
  > Reported by Pierre Riteau (firstname.lastname at gmail) (miod@)

arch/vax/include

  ~ asm.h                                 

  > Make _PROF_PROLOGUE visible if PROF is defined; will be needed by libc
  > shortly. (miod@)

arch/vax/vax

  ~ subr.s                                

  > ddb expects the kernel longjmp() to only take a single parameter and always
  > return 1 since 12+ years, it's about time to fix the offending ports.
  > Reported by Pierre Riteau (firstname.lastname at gmail) (miod@)

arch/zaurus/include

  - elf_abi.h                             

  > Remove <machine/elf_abi.h> on platforms which have it. Nothing in the tree
  > needs the few MD definitions they (sometimes) provide; only binutils is
  > interested in them, but binutils provide their own include files for that
  > purpose anyway.
  > ok deraadt@ kettenis@ (miod@)

dev/cardbus

  ~ com_cardbus.c                         ~ files.cardbus

  > Switch i386 from pccom to com.  Welcomed by many.
  > ok dlg@, jsing@, deraadt@ (kettenis@)

dev/i2c

  ~ spdmem.c                              

  > Print a specific message for non i2c attachments like macppc
  > memc, in the situation where Apple cheaps out by having no SPD
  > on integrated RAM. (jsg@)

dev/ic

  ~ com_subr.c                            

  > Switch i386 from pccom to com.  Welcomed by many.
  > ok dlg@, jsing@, deraadt@ (kettenis@)

dev/isa

  ~ files.isapnp                          

  > Switch i386 from pccom to com.  Welcomed by many.
  > ok dlg@, jsing@, deraadt@ (kettenis@)

  ~ isapnp.c                              ~ pnpdevs

  > More pccom references (miod@)

  ~ pnpdevs.h                             

  > regen (miod@)

dev/pci

  ~ pcidevs.h                             ~ pcidevs_data.h

  > sync (deraadt@)

  ~ pcidevs                               

  > intel 5100 chipset; mitja (deraadt@)

  ~ pcidevs.h                             ~ pcidevs_data.h

  > regen (kettenis@)

  ~ pcidevs                               

  > Add Fujitsu PRIMEPOWER250/450 secundary LAN. (kettenis@)

  ~ if_bge.c                              

  > Fujitsu PRIMEPOWER250/450 secondary LAN is a BCM5703C in disguise.
  > (kettenis@)

dev/pcmcia

  ~ com_pcmcia.c                          ~ files.pcmcia

  > Switch i386 from pccom to com.  Welcomed by many.
  > ok dlg@, jsing@, deraadt@ (kettenis@)

dev/puc

  ~ com_puc.c                             ~ files.puc

  > Switch i386 from pccom to com.  Welcomed by many.
  > ok dlg@, jsing@, deraadt@ (kettenis@)

net

  ~ bridgestp.c                           

  > Just clear the timeout since timeout_del() can cope with
  > unscheduled/already triggered timeouts.
  > ok brad claudio (mk@)

  ~ if_bridge.c                           

  > Only cancel (timeout_del()) the timeout if we're not going to delay it
  > (timeout_add()) anyway.
  > ok brad claudio (mk@)

== usr.bin =========================================================== 05/06 ==

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

usr.bin

  ~ Makefile                              

  > Remove elf2aout and elf2ecoff. They are no longer built since they were
  > only
  > used by the long defunct OpenBSD/arc and OpenBSD/pmax ports, and nowadays
  > using objcopy or a linker script will perform such conversions if
  > necessary.
  > ok deraadt@ kettenis@ (miod@)

elf2aout

  - Makefile                              - elf2aout.1
  - elf2aout.c                            

  > Remove elf2aout and elf2ecoff. They are no longer built since they were
  > only
  > used by the long defunct OpenBSD/arc and OpenBSD/pmax ports, and nowadays
  > using objcopy or a linker script will perform such conversions if
  > necessary.
  > ok deraadt@ kettenis@ (miod@)

elf2ecoff

  - Makefile                              - elf2ecoff.1
  - elf2ecoff.c                           

  > Remove elf2aout and elf2ecoff. They are no longer built since they were
  > only
  > used by the long defunct OpenBSD/arc and OpenBSD/pmax ports, and nowadays
  > using objcopy or a linker script will perform such conversions if
  > necessary.
  > ok deraadt@ kettenis@ (miod@)

== usr.sbin ========================================================== 06/06 ==

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

afs

  ~ src/lwp/process.vax.S                 

  > Use a symbolic entry register save mask, and sprinkle _C_LABEL.
  > No functional change. (miod@)

httpd

  ~ src/main/http_vhost.c                 

  > The default address family to be used is stored in a global variable.  If
  > no options are given on the commandline, it is set to PF_INET.
  > The configuration file parser did not use this variable in all cases, but
  > used PF_UNSPEC for getaddrinfo/getnameinfo, leading to bogus error messages
  > in some cases (but httpd operated as expected).  Use the global variable
  > instead of the hardcode PF_UNSPEC in the cases.
  > Add a new commandline flag, -U, to set the default address family to
  > PF_UNSPEC for ambigous directives.
  > Discussed with sthen. (mbalmer@)

  ~ README.IPv6                           ~ httpd.8
  ~ htdocs/manual/ipv6.html               ~ src/main/http_main.c

  > Document and implement -U option, fix usage information for '-?'.
  > Documentation corrections and spelling by jmc. (mbalmer@)

  ~ src/main/http_core.c                  ~ src/main/http_main.c
  ~ src/main/http_vhost.c                 ~ src/main/rfc1413.c

  > Remove #ifndef SIN6_LEN/#else/#endif constructs.
  > Suggested by djm a while ago.  No binary changes. (mbalmer@)

  ~ src/main/util.c                       

  > KNF and readability changes.
  > No binary changes. (mbalmer@)

ppp

  ~ ppp/ppp.8.m4                          

  > Do not mention pccom any longer. (miod@)

  ~ ppp/ppp.8.m4                          

  > punctuation fix; (jmc@)

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


More information about the odc mailing list