[odc] Daily src changes for 2004-02-21

ODC auto at squish.net
Sun Feb 22 07:00:27 GMT 2004


OpenBSD src changes summary for 2004-02-21
==========================================

distrib/notes                           distrib/sets
gnu/egcs                                gnu/usr.bin/gcc
lib/libpthread                          lib/libutil
regress                                 regress/usr.sbin
sbin/mount_ntfs                         share/man
share/mk                                sys/arch/amd64/amd64
sys/arch/amd64/conf                     sys/arch/amd64/stand/installboot
sys/arch/i386/include                   sys/arch/i386/stand/installboot
sys/dev/atapiscsi                       sys/dev/ic
sys/dev/pci                             sys/dev/usb
sys/kern                                sys/miscfs/genfs
sys/miscfs/portal                       sys/scsi
sys/sys                                 sys/uvm
sys/xfs                                 usr.bin/wall
usr.sbin/bgpd                           usr.sbin/pkg_add

== distrib =========================================================== 01/09 ==

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

notes

  ~ i386/prep                             

  > Note that some BIOSes have a 128 GB hard disk addressing limit.
  > Found by Marco Peereboom.
  > ok jmc@, miod@, nick@, weingart@ (tom@)

sets

  ~ lists/base/md.alpha                   ~ lists/base/md.amd64
  ~ lists/base/md.cats                    ~ lists/base/md.hp300
  ~ lists/base/md.hppa                    ~ lists/base/md.i386
  ~ lists/base/md.mac68k                  ~ lists/base/md.macppc
  ~ lists/base/md.mvme68k                 ~ lists/base/md.mvme88k
  ~ lists/base/md.mvmeppc                 ~ lists/base/md.pegasos
  ~ lists/base/md.sparc                   ~ lists/base/md.sparc64
  ~ lists/base/md.vax                     ~ lists/comp/md.alpha
  ~ lists/comp/md.hp300                   ~ lists/comp/md.hppa
  ~ lists/comp/md.i386                    ~ lists/comp/md.mac68k
  ~ lists/comp/md.macppc                  ~ lists/comp/md.mvme68k
  ~ lists/comp/md.mvme88k                 ~ lists/comp/md.mvmeppc
  ~ lists/comp/md.pegasos                 ~ lists/comp/md.sparc
  ~ lists/comp/md.sparc64                 ~ lists/comp/md.vax
  ~ lists/comp/mi                         ~ lists/man/md.amd64
  ~ lists/man/mi                          

  > sync (deraadt@)

  ~ lists/man/md.i386                     ~ lists/man/mi

  > sync (deraadt@)

  ~ lists/comp/md.amd64                   ~ lists/comp/md.i386
  ~ lists/man/md.amd64                    ~ lists/man/md.cats
  ~ lists/man/md.i386                     

  > sync (deraadt@)

== gnu =============================================================== 02/09 ==

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

egcs

  ~ gcc/Makefile.bsd-wrapper              

  > zap cccp.1, install phantom info.
  > ok theo (espie@)

  - gcc/gcc-local.1                       

  > kill old copies of gcc-local (espie@)

usr.bin/gcc

  ~ Makefile.bsd-wrapper                  

  > lose gcc-local as well. (espie@)

  - gcc-local.1                           

  > kill old copies of gcc-local (espie@)

== lib =============================================================== 03/09 ==

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

libpthread

  ~ arch/arm/uthread_machdep.c            ~ arch/arm/uthread_machdep_asm.S

  > Implement the arm pthreads pieces, only two regress fails, preemption_float
  > and sigmask. (drahn@)

  ~ arch/i386/uthread_machdep.c           ~ arch/i386/uthread_machdep.h

  > if cpu has fxsr, use fxsave instead; ok marc (deraadt@)

libutil

  ~ openpty.3                             

  > typo (espie@)

== regress =========================================================== 04/09 ==

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

regress

  ~ Makefile                              

  > First regression test for pkg_add, checking packages-specs lists.
  > (the nice thing about having this as perl modules is that we CAN write
  > regression tests) (espie@)

usr.sbin

  + Makefile                              + pkg_add/Makefile
  + pkg_add/check-name                    

  > First regression test for pkg_add, checking packages-specs lists.
  > (the nice thing about having this as perl modules is that we CAN write
  > regression tests) (espie@)

== sbin ============================================================== 05/09 ==

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

mount_ntfs

  ~ Makefile                              

  > commit man page always, on all architectures (deraadt@)

== share ============================================================= 06/09 ==

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

man

  ~ man1/Makefile                         

  > +gcc-local, common to both gcc versions. (espie@)

mk

  ~ bsd.own.mk                            

  > move sparc64 to gcc3. to move up please install a snapshot (or, install
  > this, build the right gcc, do a make build and pray) (deraadt@)

== sys =============================================================== 07/09 ==

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

arch/amd64/amd64

  ~ sys_machdep.c                         ~ machdep.c

  > aperture stuff (deraadt@)

arch/amd64/conf

  ~ GENERIC                               

  > much more goo (deraadt@)

arch/amd64/stand/installboot

  ~ installboot.c                         

  > Tidy up installboot -v output.
  > requested by and ok deraadt@ (tom@)

arch/i386/include

  ~ cpu.h                                 

  TAGGED SMP
  > only get sys/systm in _KERNEL (niklas@)

arch/i386/stand/installboot

  ~ installboot.c                         

  > Tidy up installboot -v output.
  > requested by and ok deraadt@ (tom@)

dev/atapiscsi

  ~ atapiscsi.c                           

  > SCSI_DATA_OUT doesn't mean transfer from host in all cases, when
  > requesting sense we reuse the same scsi xfer so flags lie and we should
  > check for C_SENSE too.
  > Eliminates pciide timeouts at the end of every burning with cdrecord.
  > Bug introduced, found and then fixed by costa at .  > ok costa@ krw@ (grange@)

dev/ic

  ~ ramdac.h                              

  > Remove left-over dependencies on the old rcons code. (miod@)

dev/pci

  ~ tga.c                                 

  > Remove left-over dependencies on the old rcons code. (miod@)

dev/usb

  ~ umass_quirks.c                        ~ umass_scsi.c
  ~ umassvar.h                            

  > Eliminate the quirks SDEV_NOSTARTUNIT, UMASS_QUIRK_NO_START_STOP, and
  > UMASS_QUIRK_FORCE_SHORT_INQUIRY. Fixes a bunch of USB devices. Based
  > on work by Mycroft in NetBSD.
  > ok tdeval@ deraadt at .rw@)

kern

  ~ vfs_subr.c                            ~ vfs_vnops.c

  TAGGED UBC
  > add VEXECMAP.  also make sure to modify filepages count only in the not
  > execpages case in uvm_pageremove().
  > this actually appears to solve the swap freak out problems.  sitting on it
  > for
  > a long time, never checked if it worked.  sigh. (tedu@)

miscfs/genfs

  ~ genfs_vnops.c                         

  TAGGED UBC
  > trick gcc into non-DIAG compile (tedu@)

miscfs/portal

  ~ portal_vnops.c                        

  > fiddle with filedesc lock to prevent recursion.  crash and fix confirmed
  > by marc balmer (tedu@)

scsi

  ~ scsi_base.c                           

  > Fix a race. scsi_done() can free a scsi_xfer, so relying on the values
  > of fields in the scsi_xfer after scsi_done() could have been called is
  > bad.
  > cdrecord can now safely burn cd's at high speeds.
  > Found by grange@, original diff by costa@, cleaned up by grange@ and
  > further polished by me following input from deraadt at .  > ok grange@ costa@ deraadt at .rw@)

  ~ scsi_all.h                            ~ scsi_base.c
  ~ scsiconf.c                            ~ scsiconf.h
  ~ sd.c                                  

  > Eliminate the quirks SDEV_NOSTARTUNIT, UMASS_QUIRK_NO_START_STOP, and
  > UMASS_QUIRK_FORCE_SHORT_INQUIRY. Fixes a bunch of USB devices. Based
  > on work by Mycroft in NetBSD.
  > ok tdeval@ deraadt at .rw@)

sys

  ~ vnode.h                               

  TAGGED UBC
  > add VEXECMAP.  also make sure to modify filepages count only in the not
  > execpages case in uvm_pageremove().
  > this actually appears to solve the swap freak out problems.  sitting on it
  > for
  > a long time, never checked if it worked.  sigh. (tedu@)

  ~ sched.h                               

  TAGGED SMP
  > export some defns to non-_KERNEL land (niklas@)

uvm

  ~ uvm_map.c                             ~ uvm_object.h
  ~ uvm_page.c                            ~ uvm_pdaemon.c
  ~ uvm_vnode.c                           

  TAGGED UBC
  > add VEXECMAP.  also make sure to modify filepages count only in the not
  > execpages case in uvm_pageremove().
  > this actually appears to solve the swap freak out problems.  sitting on it
  > for
  > a long time, never checked if it worked.  sigh. (tedu@)

xfs

  ~ xfs_vfsops-openbsd.c                  

  TAGGED UBC
  > compile (tedu@)

== usr.bin =========================================================== 08/09 ==

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

wall

  ~ wall.c                                

  > Correctness is also not using err(1, "out of memory") if malloc()
  > fails (per err(3)).  Ditto for strdup().
  > ok henning@ (tom@)

== usr.sbin ========================================================== 09/09 ==

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

bgpd

  ~ session.c                             

  > keep O_NONBLOCK after teh tcp sessions are established, claudio ok
  > (henning@)

pkg_add

  ~ OpenBSD/PackageInfo.pm                

  > distinguish between unreadable directories and non packages.
  > problem found by John L. Scarfone, who runs a paranoid root. (espie@)

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



More information about the odc mailing list