[owc] Weekly src changes ending 2006-05-07

OWC auto at squish.net
Mon May 8 08:00:46 BST 2006


OpenBSD src changes summary for 2006-04-30 to 2006-05-07 inclusive
==================================================================

Makefile.cross                          bin/ps
bin/systrace                            distrib/sets
games/fortune                           gnu/usr.bin/binutils
include/math.h                          kerberosV
lib/libc                                lib/libdes
lib/libexpat                            lib/libkvm
lib/libm                                lib/libssl
libexec/ld.so                           regress/usr.bin
sbin/bioctl                             sbin/isakmpd
sbin/pfctl                              share/man
sys/arch/alpha/conf                     sys/arch/amd64/amd64
sys/arch/amd64/conf                     sys/arch/i386/conf
sys/arch/i386/i386                      sys/arch/i386/include
sys/arch/i386/isa                       sys/arch/luna88k/conf
sys/arch/luna88k/include                sys/arch/m88k/include
sys/arch/m88k/m88k                      sys/arch/macppc/conf
sys/arch/macppc/pci                     sys/arch/mvme88k/conf
sys/arch/mvme88k/dev                    sys/arch/mvme88k/include
sys/arch/mvme88k/mvme88k                sys/arch/sgi/sgi
sys/arch/sparc/include                  sys/arch/sparc64/include
sys/compat/common                       sys/compat/freebsd
sys/compat/netbsd                       sys/conf
sys/dev                                 sys/dev/acpi
sys/dev/i2c                             sys/dev/ic
sys/dev/pci                             sys/dev/usb
sys/kern                                sys/net
sys/nfs                                 sys/scsi
sys/sys                                 sys/ufs/mfs
sys/ufs/ufs                             sys/uvm
sys/xfs                                 usr.bin/cvs
usr.bin/diff                            usr.bin/getent
usr.bin/join                            usr.bin/mail
usr.bin/make                            usr.bin/mg
usr.bin/paste                           usr.bin/rcs
usr.bin/ssh                             usr.bin/xlint
usr.sbin/afs                            usr.sbin/bgpd
usr.sbin/bind                           usr.sbin/config
usr.sbin/ospfd                          usr.sbin/pkg_add

== Makefile.cross ==================================================== 01/15 ==

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

Makefile.cross


  > Install binutils objdump on a.out cross-tools; ok grange@ (miod@)

== bin =============================================================== 02/15 ==

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

ps

  ~ extern.h                              ~ keyword.c
  ~ print.c                               

  > Format nicelevel relative NZERO as other tools do.
  > Surely, says millert. (hugh@)

systrace

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

  > some type cleanup
  > with feedback from kettenis (sturm@)

  ~ systrace.1                            

  > be more clear about policy inheritance
  > help and ok jmc, provos (sturm@)

== distrib =========================================================== 03/15 ==

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

sets

  ~ lists/comp/mi                         ~ lists/man/mi

  > sync (deraadt@)

  ~ lists/base/md.alpha                   ~ lists/base/md.amd64
  ~ lists/base/md.cats                    ~ lists/base/md.hp300
  ~ lists/base/md.hppa                    ~ lists/base/md.hppa64
  ~ lists/base/md.i386                    ~ lists/base/md.luna88k
  ~ lists/base/md.mac68k                  ~ lists/base/md.macppc
  ~ lists/base/md.mvme68k                 ~ lists/base/md.mvme88k
  ~ lists/base/md.mvmeppc                 ~ lists/base/md.sgi
  ~ lists/base/md.sparc                   ~ lists/base/md.sparc64
  ~ lists/base/md.vax                     ~ lists/base/md.zaurus
  ~ lists/comp/mi                         ~ lists/man/mi

  > sync (deraadt@)

  ~ lists/base/md.alpha                   ~ lists/base/md.amd64
  ~ lists/base/md.cats                    ~ lists/base/md.hp300
  ~ lists/base/md.hppa                    ~ lists/base/md.hppa64
  ~ lists/base/md.i386                    ~ lists/base/md.mac68k
  ~ lists/base/md.macppc                  ~ lists/base/md.mvme68k
  ~ lists/base/md.mvmeppc                 ~ lists/base/md.sgi
  ~ lists/base/md.sparc                   ~ lists/base/md.sparc64
  ~ lists/base/md.zaurus                  

  > sync (deraadt@)

== games ============================================================= 04/15 ==

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

fortune

  ~ datfiles/fortunes                     

  > "The Wall" was released in 1979. This is not 'recent' anymore. feedback
  > from krw@ and jsg at .atthieu@)

== gnu =============================================================== 05/15 ==

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

usr.bin/binutils

  ~ Makefile.bsd-wrapper                  

  > Allow USING_NEW_BINUTILS to be overriden by whatever invokes make here.
  > (miod@)

== include =========================================================== 06/15 ==

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

math.h

  ~ math.h                                

  > delint; ok millert@ beck@ (otto@)

== kerberosV ========================================================= 07/15 ==

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

kerberosV

  - src/lib/krb5/krb5_build_principal.3   - src/lib/krb5/krb5_free_addresses.3
  - src/lib/krb5/krb5_free_principal.3    
  - src/lib/krb5/krb5_principal_get_realm.3
  - src/lib/krb5/krb5_sname_to_principal.3

  > Not used, niether by OpenBSD nor by Heimdal, anymore (biorn@)

  ~ src/lib/krb5/krb5.3                   ~ src/lib/krb5/krb5_address.3
  ~ src/lib/krb5/krb5_c_make_checksum.3   ~ src/lib/krb5/krb5_check_transited.3
  ~ src/lib/krb5/krb5_creds.3             ~ src/lib/krb5/krb5_expand_hostname.3
  ~ src/lib/krb5/krb5_generate_random_block.3
  ~ src/lib/krb5/krb5_get_init_creds.3    ~ src/lib/krb5/krb5_is_thread_safe.3
  ~ src/lib/krb5/krb5_keyblock.3          ~ src/lib/krb5/krb5_mk_safe.3
  ~ src/lib/krb5/krb5_parse_name.3        ~ src/lib/krb5/krb5_principal.3
  ~ src/lib/krb5/krb5_rcache.3            ~ src/lib/krb5/krb5_rd_safe.3
  ~ src/lib/krb5/krb5_ticket.3            ~ src/lib/krb5/krb5_timeofday.3
  ~ src/lib/krb5/krb5_verify_init_creds.3
  ~ src/lib/krb5/krb5_verify_user.3       ~ src/lib/krb5/krb5_warn.3

  > Minor tweaks, mostly eyeballed by jmc@ (biorn@)

  ~ src/lib/krb5/krb5_encrypt.3           ~ src/lib/krb5/krb5_is_thread_safe.3
  ~ src/lib/krb5/krb5_verify_init_creds.3

  > Fix long .Nd lines. Help from jmc@ (biorn@)

  - lib/Makefile                          - lib/Makefile.inc
  - lib/45/Makefile                       - lib/45/shlib_version
  - lib/asn1/Makefile                     - lib/asn1/asn1_APOptions.c
  - lib/asn1/asn1_AP_REP.c                - lib/asn1/asn1_AP_REQ.c
  - lib/asn1/asn1_AS_REP.c                - lib/asn1/asn1_AS_REQ.c
  - lib/asn1/asn1_Authenticator.c         - lib/asn1/asn1_AuthorizationData.c
  - lib/asn1/asn1_CKSUMTYPE.c             - lib/asn1/asn1_Checksum.c
  - lib/asn1/asn1_ENCTYPE.c               - lib/asn1/asn1_ETYPE_INFO.c
  - lib/asn1/asn1_ETYPE_INFO_ENTRY.c      - lib/asn1/asn1_EncAPRepPart.c
  - lib/asn1/asn1_EncASRepPart.c          - lib/asn1/asn1_EncKDCRepPart.c
  - lib/asn1/asn1_EncKrbCredPart.c        - lib/asn1/asn1_EncKrbPrivPart.c
  - lib/asn1/asn1_EncTGSRepPart.c         - lib/asn1/asn1_EncTicketPart.c
  - lib/asn1/asn1_EncryptedData.c         - lib/asn1/asn1_EncryptionKey.c
  - lib/asn1/asn1_HostAddress.c           - lib/asn1/asn1_HostAddresses.c
  - lib/asn1/asn1_KDCOptions.c            - lib/asn1/asn1_KDC_REP.c
  - lib/asn1/asn1_KDC_REQ.c               - lib/asn1/asn1_KDC_REQ_BODY.c
  - lib/asn1/asn1_KRB_CRED.c              - lib/asn1/asn1_KRB_ERROR.c
  - lib/asn1/asn1_KRB_PRIV.c              - lib/asn1/asn1_KRB_SAFE.c
  - lib/asn1/asn1_KRB_SAFE_BODY.c         - lib/asn1/asn1_KerberosTime.c
  - lib/asn1/asn1_KrbCredInfo.c           - lib/asn1/asn1_LR_TYPE.c
  - lib/asn1/asn1_LastReq.c               - lib/asn1/asn1_MESSAGE_TYPE.c
  - lib/asn1/asn1_METHOD_DATA.c           - lib/asn1/asn1_NAME_TYPE.c
  - lib/asn1/asn1_PADATA_TYPE.c           - lib/asn1/asn1_PA_DATA.c
  - lib/asn1/asn1_PA_ENC_TS_ENC.c         - lib/asn1/asn1_Principal.c
  - lib/asn1/asn1_PrincipalName.c         - lib/asn1/asn1_Realm.c
  - lib/asn1/asn1_TGS_REP.c               - lib/asn1/asn1_TGS_REQ.c
  - lib/asn1/asn1_Ticket.c                - lib/asn1/asn1_TicketFlags.c
  - lib/asn1/asn1_TransitedEncoding.c     - lib/asn1/asn1_UNSIGNED.c
  - lib/asn1/krb5_asn1.h                  - lib/asn1/shlib_version
  - lib/gssapi/Makefile                   - lib/gssapi/shlib_version
  - lib/hdb/Makefile                      - lib/hdb/asn1_Event.c
  - lib/hdb/asn1_GENERATION.c             - lib/hdb/asn1_HDBFlags.c
  - lib/hdb/asn1_Key.c                    - lib/hdb/asn1_Salt.c
  - lib/hdb/asn1_files                    - lib/hdb/asn1_hdb_entry.c
  - lib/hdb/hdb_asn1.h                    - lib/kadm5clnt/Makefile
  - lib/kadm5srv/Makefile                 - lib/kafs/Makefile
  - lib/kafs/afssys_openbsd.c             - lib/kafs/dummy.c
  - lib/kafs/shlib_version                - lib/krb5/Makefile
  - lib/krb5/shlib_version                - lib/roken/Makefile
  ~ Makefile                              

  > This stuff isn't used anymore.
  > ok beck@ (biorn@)

  ~ src/lib/kafs/kafs.3                   

  > Add a missing , (biorn@)

  ~ src/lib/krb5/krb5_get_in_cred.3       

  > .Nd is needed.  Tree breaking is a disgusting thing, especially when ok'd
  > (deraadt@)

== lib =============================================================== 08/15 ==

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

libc

  ~ string/strlcpy.c                      ~ string/wcslcpy.c

  > Convert do {} while loop -> while {} for clarity.  No binary change
  > on most architectures.  From Oliver Smith.	OK deraadt@ and henning@
  > (millert@)

libdes

  ~ des_locl.h                            ~ set_key.c

  > delint; ok millert@ ray@ (otto@)

libexpat

  ~ COPYING                               ~ Changes
  ~ Makefile                              ~ README
  ~ shlib_version                         ~ doc/reference.html
  ~ doc/style.css                         ~ examples/elements.c
  ~ examples/outline.c                    ~ lib/expat.h
  ~ lib/internal.h                        ~ lib/winconfig.h
  ~ lib/xmlparse.c                        ~ lib/xmlrole.c
  ~ lib/xmltok.c                          ~ lib/xmltok.h
  ~ lib/xmltok_impl.c                     ~ lib/xmltok_ns.c
  + lib/amigaconfig.h                     + lib/expat_external.h
  + lib/macconfig.h                       

  > Update to 2.0.0; keep our local changes
  > ok espie@, djm@ (alek@)

libkvm

  ~ kvm_i386.c                            

  > Unbreak.  Fix some of the lint warnings that mickey reintroduced.
  > ok miod@ (kettenis@)

libm

  ~ src/e_j0f.c                           ~ src/e_j1f.c
  ~ src/s_cosf.c                          

  > delint; ok millert@ beck@ (otto@)

libssl

  ~ crypto/shlib_version                  ~ src/crypto/dh/dh.h
  ~ src/crypto/dh/dh_check.c              ~ src/crypto/dh/dh_err.c
  ~ src/crypto/dh/dh_key.c                

  > backport checks for degenerate Diffie-Hellman public exponents from
  > OpenSSL-0.9.8a, where they were added without a corresponding patch to
  > 0.9.7 or an advisory!  ok theo@ markus@ (djm@)

== libexec =========================================================== 09/15 ==

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

ld.so

  ~ i386/ldasm.S                          

  > Rewrite the ld.so syscall stubs as macros. no binary change. (drahn@)

  ~ dlfcn.c                               

  > Keep the state of the debugger constistant, let it know a the dynamic
  > linker is adding or removing a library before it does the act.
  > Discussed with kurt@ some time ago. (drahn@)

  ~ Makefile                              

  > Extra paranoid, do not allow the dynamic linker to have unresolved
  > references
  > as this causes much frustration as it fails to run after being installed
  > and is a paint to recover. (drahn@)

  ~ Makefile                              ~ ld.so.1
  ~ library.c                             ~ library_mquery.c
  ~ loader.c                              ~ resolve.c
  ~ resolve.h                             ~ alpha/ldasm.S
  ~ alpha/syscall.h                       ~ amd64/ldasm.S
  ~ amd64/syscall.h                       ~ arm/ldasm.S
  ~ arm/syscall.h                         ~ hppa/ldasm.S
  ~ hppa/rtld_machine.c                   ~ hppa/syscall.h
  ~ i386/ldasm.S                          ~ i386/syscall.h
  ~ mips64/syscall.h                      ~ powerpc/syscall.h
  ~ sparc/ldasm.S                         ~ sparc/syscall.h
  ~ sparc64/ldasm.S                       ~ sparc64/syscall.h
  + dl_prebind.c                          + dl_prebind.h
  + prebind.h                             + prebind/Makefile
  + prebind/debug.c                       + prebind/objarray.c
  + prebind/prebind.8                     + prebind/prebind.c
  + prebind/prebind_struct.h              + prebind/sod.c
  + prebind_strip/Makefile                + prebind_strip/prebind_strip.8
  + prebind_strip/prebind_strip.c         

  > prebind - how to prelink a binary without throwing security out the window
  > Prelink fixes the address of libraries making 'return to libc' attacks
  > trival,
  > prebind uses a different method to achieve most of the same gains, however
  > without adding any security conerns.
  > Still under development, now in-tree. (drahn@)

  ~ prebind/prebind.c                     

  > When loading 'binaries' ignore libraries. (drahn@)

  ~ prebind/prebind.c                     

  > checks that libraries are libaries and programs are programs. (needs
  > cleanup) (drahn@)

  ~ prebind/prebind.c                     

  > goodbye noisy debug. (drahn@)

  ~ prebind/debug.c                       ~ prebind/objarray.c
  ~ prebind/prebind.c                     

  > several fixes dealing with determining if a program or library is being
  > processed. Also handles symlinks/hardlinks between binaries better.
  > (drahn@)

  ~ prebind/prebind.c                     

  > Using mmap for these is going to be trickier than I thought, go ahead
  > and do the allocation failure checks for now. (drahn@)

  ~ Makefile                              

  > Dont build/install prebind tools just yet. (drahn@)

  ~ ld.so.1                               

  > Fix wordo, pointed out by jmc@ (drahn@)

  ~ dl_prebind.c                          

  > Avoid a couple of potential crashes if the input data isn't right. ok beck@
  > (drahn@)

  ~ prebind/prebind.8                     ~ prebind/prebind_struct.h
  ~ prebind_strip/prebind_strip.8         

  > Add missing copyright on header file, OpenBSD -> .Ox in manpages, thanks
  > Dries. (drahn@)

  ~ prebind/prebind.c                     

  > When copying a prog reference from a lib do not put it in the common table
  > but put a 'invalid' marker there instead. (drahn@)

  ~ prebind/prebind.c                     

  > Work around a problem on sparc32 (for now), some debug cleanup. (drahn@)

  ~ prebind_strip/prebind_strip.c         

  > Add -o <output> mode in addition to strip in place, '-' is stdout. (drahn@)

  ~ dl_prebind.c                          

  > Make certain that it doesn't try to load prelink info once the primary link
  > is finished. prelink on dlopen cannot work, the symbol cache still works,
  > just
  > cannot be preloaded. fixes xchm. (drahn@)

  ~ prebind/prebind.c                     

  > Ignore binaries which are not OpenBSD binaries (.note section), some
  > cleanup (drahn@)

  ~ prebind/prebind.c                     ~ prebind/prebind.8
  ~ prebind_strip/prebind_strip.8         

  > various fixes up; some from michael knudsen, some from myself (jmc@)

  ~ dl_prebind.c                          

  > Prebind should not force lazy binding to be turned off during dlopen()
  > the presense of LD_BIND_NOW should control that. Fixes the X server.
  > (drahn@)

  ~ prebind/prebind.c                     

  > sigh, build on non gcc3 systems. (drahn@)

  ~ prebind/prebind.8                     

  > improve wording somewhat. (drahn@)

  ~ prebind/objarray.c                    

  > Do not write the same data to a library that is already there when
  > using -m also allows a user to run this on their binaries if the system
  > libraries already have prelink data. (drahn@)

  ~ prebind/prebind.c                     

  > If there is a conflict between two symbols, and one of the symbols is a
  > reference to the local object, go ahead and save save the local symbol
  > in the library cache, a fixup will be generated in any binary which
  > overrides
  > the symbol, however, the rest of the program which do not override the
  > symbol
  > will not have to pay for the fixup. Cuts about of prebind data here by 32%
  > (drahn@)

  ~ prebind/prebind.c                     

  > While os-note-elf.h uses '.long' on the openbsd note section, they
  > are really stored as ints. keep the variable name the same as the type.
  > (just a variable rename) (drahn@)

  ~ prebind/objarray.c                    

  > Initialize oprebind_data so that unintialized memory is not referenced.
  > caught by niallo (drahn@)

== regress =========================================================== 10/15 ==

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

usr.bin

  ~ rcs/Makefile                          

  > - add tests for -oranges.
  > ok joris@ (niallo@)

  ~ xlint/test-20.c                       ~ xlint/test-20.c.exp

  > more test cases (otto@)

  ~ xlint/Makefile                        + xlint/test-21.c
  + xlint/test-21.c.exp                   + xlint/test-22.c
  + xlint/test-22.c.exp                   

  > Add some regress tests for lint1 crashes which are not solved yet (cloder@)

  ~ rcs/Makefile                          

  > Output to a different filename so the tests work without an obj/
  > directory.	Also change diff -u to ${DIFF}.
  > OK xsa@ (ray@)

  ~ rcs/Makefile                          

  > o GNU ci returns 0 when a file is reverted (co -l file; ci file).
  > o Don't unlock file after revert.
  > o Fix spelling.
  > Passes extra tests in GNU rcstest.
  > OK niallo at .ay@)

  ~ xlint/test-21.c                       ~ xlint/test-21.c.exp
  ~ xlint/test-22.c.exp                   

  > SEGVs have been fixed, so include proper expected output (otto@)

  ~ xlint/Makefile                        + xlint/test-23.c
  + xlint/test-23.c.exp                   

  > test pointer casts (otto@)

== sbin ============================================================== 11/15 ==

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

bioctl

  ~ bioctl.c                              

  > Add a little debug to test BIOCINQ primitive. (marco@)

isakmpd

  ~ dh.c                                  ~ math_group.c
  ~ math_group.h                          

  > check for degenerate Diffie-Hellman public exponents;
  > ok markus@ hshoexer@ deraadt@ (djm@)

  ~ math_group.c                          

  > correct rfc reference (djm@)

  ~ math_group.c                          

  > correct correct rfc reference (djm@)

pfctl

  ~ parse.y                               

  > add support for "tagged {}" lists, from Pierre-Yves Ritschard (dhartmei@)

  ~ parse.y                               

  > fix creation of sub-anchors, e.g. if you create an anchor /foo/bar, create
  > only bar under foo, not /bar as well.
  > secondly, when using "load anchor from" from a sub-anchor, the loading
  > point should be relative to the sub-anchor doing the load (unless absolute
  > paths are used, of course).
  > from Boris Polevoy. probably a -stable candidate. (dhartmei@)

== share ============================================================= 12/15 ==

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

man

  ~ man4/bio.4                            

  > Bring the function list back into reality.
  > Several nits. (marco@)

  ~ man4/bio.4                            

  > fix SEE ALSO; (jmc@)

  ~ man4/maxtmp.4                         

  > mention support for Maxim 6657, Maxim 6658 and Maxium 6659 parts (djm@)

  ~ man5/pf.conf.5                        

  > - replace <> with .Aq
  > - replace OpenBSD with .Ox
  > from wiz at netbsdmc@)

  ~ man4/Makefile                         ~ man4/pci.4
  + man4/xge.4                            

  > and here comes a man page for xge(4). (brad@)

  ~ man4/ipw.4                            ~ man4/iwi.4

  > sync ipw man page w/ latest iwi changes.
  > move firmware packages to a better location. (damien@)

  ~ man4/ral.4                            

  > rephrase the CAVEATS section.  not all PCI ral adapters require a PCI 2.2
  > compliant motherboard.  the Edimax EW-7128g for instance works on PCI 2.1
  > boards.
  > submitted by Karel Gardas. (damien@)

  ~ man5/pf.conf.5                        

  > add support for "tagged {}" lists, from Pierre-Yves Ritschard (dhartmei@)

  ~ man4/xge.4                            

  > .An -noplit (jmc@)

  ~ man9/altq.9                           

  > make .Nm match the page name; (jmc@)

  ~ man5/pf.conf.5                        

  > update the "tagged" line; ok dhartmei (jmc@)

  ~ man4/ral.4                            

  > slight reword; (jmc@)

  ~ man8/vpn.8                            

  > correct command-line prompts; from daniel wade (jmc@)

  ~ man4/systrace.4                       

  > SYSTR_MSG_CHILD is also sent when a child exits, from pr 5088
  > ok jmc, provos (sturm@)

  ~ man4/ddb.4                            

  > Mention x/I behaviour on m88k. (miod@)

  ~ man4/lge.4                            

  > Xr xmphy (brad@)

  ~ man4/ral.4                            

  > small reading improvements; (jmc@)

  ~ man9/rwlock.9                         

  > name arg for rw_init (tedu@)

  ~ man9/Makefile                         ~ man9/pool.9

  > Pool drains are gone, okay tedu@ (pedro@)

== sys =============================================================== 13/15 ==

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

arch/alpha/conf

  ~ GENERIC                               ~ RAMDISKBIG

  > add ciss(4) to the untested section, 530[24]'s exist with Alpha specific
  > firmware for SRM setup. (brad@)

  ~ GENERIC                               ~ RAMDISKBIG
  ~ files.alpha                           

  > - Add vge(4), lge(4), hme(4), mtd(4), luphy(4) and ciphy(4)
  > - Move tl(4) and ste(4) out of the untested section (brad@)

  ~ GENERIC                               ~ RAMDISKBIG

  > fix the untested comment. (brad@)

arch/amd64/amd64

  ~ fpu.c                                 

  TAGGED OPENBSD_3_9
  > MFC:
  > Fix by mickey@
  > from todays freebsd advisory: fxrstor on amd cpu does not restore
  > fip,fdp,fop thus leaking other proc'sexecution history (brad@)

  ~ fpu.c                                 

  TAGGED OPENBSD_3_8
  > MFC:
  > Fix by mickey@
  > from todays freebsd advisory: fxrstor on amd cpu does not restore
  > fip,fdp,fop thus leaking other proc'sexecution history (brad@)

  ~ fpu.c                                 

  TAGGED OPENBSD_3_7
  > MFC:
  > Fix by mickey@
  > from todays freebsd advisory: fxrstor on amd cpu does not restore
  > fip,fdp,fop thus leaking other proc'sexecution history (brad@)

arch/amd64/conf

  ~ GENERIC                               ~ RAMDISK
  ~ RAMDISK_CD                            

  > - add I2O to RAMDISK_CD
  > - add I2O SCSI pass-through driver (brad@)

  ~ GENERIC                               ~ RAMDISK_CD

  > add commented out entries for the Neterion Xframe-I 10Gb ethernet adapters.
  > the driver is currently untested. (brad@)

arch/i386/conf

  ~ GENERIC                               ~ RAMDISKB
  ~ RAMDISK_CD                            

  > sync the I2O section with macppc config, adding the I2O SCSI pass-through
  > driver. (brad@)

  ~ GENERIC                               ~ RAMDISK_CD

  > add commented out entries for the Neterion Xframe-I 10Gb ethernet adapters.
  > the driver is currently untested. (brad@)

  ~ files.i386                            

  > Fixup broken mpbios'es on VT8237 and nForce4 chipsets.  Fixes interrupt
  > routing for several integrated devices on those chipsets in GENERIC.MP.
  > ok brad@, mickey@ (kettenis@)

arch/i386/i386

  ~ mpbios.c                              

  > Do not leave behind half-initialized data structures so we don't stumble
  > over
  > corrupt interrupt table entries later on.
  > ok brad@ (kettenis@)

  ~ mpbios.c                              + mpbios_intr_fixup.c

  > Fixup broken mpbios'es on VT8237 and nForce4 chipsets.  Fixes interrupt
  > routing for several integrated devices on those chipsets in GENERIC.MP.
  > ok brad@, mickey@ (kettenis@)

  ~ pmap.c                                

  TAGGED OPENBSD_3_8
  > MFC:
  > Fix by tom@
  > Don't leave a dangling pointer to the process-specific LDT after
  > we free it.  Instead set it to the default value for a new process. (brad@)

  ~ pmap.c                                

  TAGGED OPENBSD_3_9
  > MFC:
  > Fix by tom@
  > Don't leave a dangling pointer to the process-specific LDT after
  > we free it.  Instead set it to the default value for a new process. (brad@)

  ~ pmap.c                                

  TAGGED OPENBSD_3_7
  > MFC:
  > Fix by tom@
  > Don't leave a dangling pointer to the process-specific LDT after
  > we free it.  Instead set it to the default value for a new process. (brad@)

  ~ powernow-k7.c                         

  TAGGED OPENBSD_3_7
  > * Plug memory leak if init fails; ok gwk@ (pat@)

  ~ locore.s                              

  TAGGED OPENBSD_3_7
  > somehow PTDsize got lost from the pae diff; found by kettenis@ (mickey@)

  ~ locore.s                              

  TAGGED OPENBSD_3_7
  > .globl PTDsize; mickey obviously doesn't test before he commits (kettenis@)

arch/i386/include

  ~ mpbiosvar.h                           

  > Fixup broken mpbios'es on VT8237 and nForce4 chipsets.  Fixes interrupt
  > routing for several integrated devices on those chipsets in GENERIC.MP.
  > ok brad@, mickey@ (kettenis@)

arch/i386/isa

  ~ npx.c                                 

  TAGGED OPENBSD_3_9
  > MFC:
  > Fix by mickey@
  > from todays freebsd advisory: fxrstor on amd cpu does not restore
  > fip,fdp,fop thus leaking other proc'sexecution history (brad@)

  ~ npx.c                                 

  TAGGED OPENBSD_3_8
  > MFC:
  > Fix by mickey@
  > from todays freebsd advisory: fxrstor on amd cpu does not restore
  > fip,fdp,fop thus leaking other proc'sexecution history (brad@)

  ~ npx.c                                 

  TAGGED OPENBSD_3_7
  > MFC:
  > Fix by mickey@
  > from todays freebsd advisory: fxrstor on amd cpu does not restore
  > fip,fdp,fop thus leaking other proc'sexecution history (brad@)

arch/luna88k/conf

  ~ Makefile.luna88k                      

  > Now that .S files have proper dependencies, do not hardcode dependencies
  > on assym.h in Makefile, especially since some files have been renamed...
  > (miod@)

arch/luna88k/include

  ~ board.h                               

  > eneble -> enable (miod@)

arch/m88k/include

  ~ db_machdep.h                          

  > Replace db_{disable,enable}_interrupts with expanded version of themselves,
  > having macros for this is not worth doing. (miod@)

  ~ db_machdep.h                          

  > Overhaul of the disassembler code, with random fixes, better conditionnal
  > branches decoding, correct control register descriptions, and (the reason
  > for these changes in the first place) 88110 instruction support.
  > ddb will now disassemble by default for the processor it is running for,
  > and using the alternate format (x/I) disassembles for the other. (miod@)

  ~ ieee.h                                

  > Document the 88110 extended (80-bit) floating point format. (miod@)

  ~ m8820x.h                              

  > On 88100-based systems, take the two hardwired BATC into account in
  > pmap_extract(), and do not bother creating regular page table mappings
  > for obio regions which are covered by these BATC entries. (miod@)

  ~ vmparam.h                             

  > Move the userland stack at the top of the userland address space. (miod@)

arch/m88k/m88k

  ~ trap.c                                

  > Replace db_{disable,enable}_interrupts with expanded version of themselves,
  > having macros for this is not worth doing. (miod@)

  ~ db_sstep.c                            

  > Fix a (documented) logic botch in inst_load(). (miod@)

  ~ db_disasm.c                           

  > Overhaul of the disassembler code, with random fixes, better conditionnal
  > branches decoding, correct control register descriptions, and (the reason
  > for these changes in the first place) 88110 instruction support.
  > ddb will now disassemble by default for the processor it is running for,
  > and using the alternate format (x/I) disassembles for the other. (miod@)

  ~ db_disasm.c                           

  > Even better control register name display logic. (miod@)

  ~ trap.c                                

  > scaled version of ld.d and st.d is obviously not a 64 bit transaction,
  > so there is no need to take care of it in double_reg_fixup(), as it will
  > never trigger the kind of misalignment faults this function recovers from.
  > Doh! (miod@)

  ~ pmap.c                                

  > On 88100-based systems, take the two hardwired BATC into account in
  > pmap_extract(), and do not bother creating regular page table mappings
  > for obio regions which are covered by these BATC entries. (miod@)

  ~ db_trace.c                            

  > Nuke TRACE_DEBUG code. Half of it was useless, and the other half was
  > either
  > not smart enough to cope with gcc -O2 or with the no longer fixed address
  > for the kernels stack. (miod@)

arch/macppc/conf

  ~ GENERIC                               ~ RAMDISK

  > - Enable fxp and rl at CardBus GENERIC and RAMDISK
  > - Enable dc at CardBus RAMDISK, already in GENERIC
  > - Enable wdc at PCMCIA RAMDISK, already in GENERIC (brad@)

  ~ GENERIC                               

  > Enable rtw at pcirad@)

  ~ RAMDISK                               

  > add I2O to the RAMDISK (brad@)

arch/macppc/pci

  ~ vgafb.c                               

  > The r128 driver also need the 0xb0000-0xc0000 address range. found by
  > Jolan at .atthieu@)

arch/mvme88k/conf

  ~ GENERIC                               ~ M187
  ~ M188                                  ~ M197
  ~ RAMDISK                               ~ files.mvme88k

  > Wave goodbye to bugtty on mvme88k as well. (miod@)

  ~ Makefile.mvme88k                      

  > Now that .S files have proper dependencies, do not hardcode dependencies
  > on assym.h in Makefile, especially since some files have been renamed...
  > (miod@)

arch/mvme88k/dev

  - bugtty.c                              - bugttyfunc.h
  ~ bugio.c                               

  > Wave goodbye to bugtty on mvme88k as well. (miod@)

  ~ mainbus.c                             

  > On 88100-based systems, take the two hardwired BATC into account in
  > pmap_extract(), and do not bother creating regular page table mappings
  > for obio regions which are covered by these BATC entries. (miod@)

arch/mvme88k/include

  ~ bugio.h                               

  > Wave goodbye to bugtty on mvme88k as well. (miod@)

arch/mvme88k/mvme88k

  ~ autoconf.c                            ~ conf.c
  ~ m188_machdep.c                        ~ m1x7_machdep.c
  ~ machdep.c                             

  > Wave goodbye to bugtty on mvme88k as well. (miod@)

  ~ pmap_table.c                          

  > On 88100-based systems, take the two hardwired BATC into account in
  > pmap_extract(), and do not bother creating regular page table mappings
  > for obio regions which are covered by these BATC entries. (miod@)

  ~ locore.S                              

  > Decide whether we are running on 88100 or 88110 upon the processor
  > identification register value, rather than upon the MVME board number.
  > (miod@)

  ~ m187_machdep.c                        ~ m188_machdep.c

  > Sync VM games on startup with earlier pmap_table[] cleanups. (miod@)

  ~ m188_machdep.c                        

  > Do not panic on MVME188 if the abort switch is pressed while we are at ipl
  > 7
  > (fallout from always enabling the abort switch). (miod@)

  ~ m88410.c                              

  > No need to mask misaligned address exceptions for global flush or
  > invalidate
  > operations (those do not take a page number in the low address bits of the
  > 88410 control register). (miod@)

arch/sgi/sgi

  ~ machdep.c                             

  > fix pmap debug code and only compile it when PMAPDEBUG is defined.
  > this change also unbreaks builds with DEBUG defined.
  > ok miod@ (robert@)

arch/sparc/include

  ~ reloc.h                               

  > Long due update from a.out to ELF (slightly different) relocation defines,
  > from sparc64 with love. Spotted by drahn@ (miod@)

arch/sparc64/include

  ~ reloc.h                               

  > Missing license block; from art@ (miod@)

compat/common

  ~ vfs_syscalls_25.c                     

  > remove the simplelock argument from vfs_busy() which is currently not
  > used and will never be used this way in VFS
  > requested by and ok pedro, ok krw, biorn (sturm@)

compat/freebsd

  ~ freebsd_file.c                        

  > remove the simplelock argument from vfs_busy() which is currently not
  > used and will never be used this way in VFS
  > requested by and ok pedro, ok krw, biorn (sturm@)

compat/netbsd

  ~ netbsd_file.c                         

  > remove the simplelock argument from vfs_busy() which is currently not
  > used and will never be used this way in VFS
  > requested by and ok pedro, ok krw, biorn (sturm@)

conf

  ~ newvers.sh                            

  TAGGED OPENBSD_3_9
  > -stable (brad@)

dev

  ~ ipmivar.h                             

  > remove a block of duplicated defines, ok marco (wilfried@)

  ~ ipmi.c                                

  > if the fan data rate is per second convert it to per minute, ok marco
  > (wilfried@)

dev/acpi

  ~ acpibat.c                             

  > rv not set.  Sloppy but didn't matter.
  > pointed out by joris@ (marco@)

dev/i2c

  ~ maxim6690.c                           

  > support the Maxim 6657, 6658 and 6659 parts, which are register-wise
  > identical to the Maxim 6690. Also fixes a bug which caused Maxim 6642
  > parts to be mishandled; ok deraadt@ (djm@)

  ~ maxim6690.c                           

  > incorporate feedback from kettenis@: fix init of sc_temp_invalid for LM90,
  > make sc_temp_invalid an array and set 0x7f as an invalid reading for the
  > Maxim 6690 too (djm@)

dev/ic

  ~ rt2661.c                              

  > remove an unused prototype. (damien@)

  ~ wdc.c                                 ~ wdcvar.h

  > Provide a function pointer that allows for the wdc reset
  > method to be changed as required.
  > This will be required for dealing with several chips in pciide. (jsg@)

  ~ rt2560.c                              ~ rt2661.c

  TAGGED OPENBSD_3_9
  > MFC:
  > Fix by damien@
  > fixes interrupts processing.
  > should fix a panic reported by Karel Gardas. (brad@)

  ~ ral.c                                 

  TAGGED OPENBSD_3_8
  > MFC:
  > Fix by damien@
  > allows dhcpd to work when ral/ural is acting as a standalone AP (without
  > bridging).	it appears that ieee80211_input() enqueues packets in if_snd
  > without calling if_start().
  > ok damien@ (brad@)

  ~ wdc.c                                 

  TAGGED OPENBSD_3_8
  > Default to Ultra DMA Mode 6 not 5 when we know we have a native
  > SATA drive.
  > ok grange@ (jsg@)

  ~ dc.c                                  

  TAGGED OPENBSD_3_8
  > - Return if there is nothing to do in the interrupt handler.
  > - Check for IFF_RUNNING in the interrupt loop. (brad@)

  ~ lm78.c                                ~ lm78var.h

  TAGGED OPENBSD_3_8
  > Add abstraction for resistor factor; makes it easier to compare them to
  > the datasheets.  Fix a few typos too.
  > From Constantine Murenin <mureninc at gmail.org(kettenis@)

  ~ aacvar.h                              

  TAGGED OPENBSD_3_8
  > add a name to rwlock so that we can tell where procs are getting stuck
  > without breaking into ddb.	doubles the size of rwlock [1], but moving
  > forward this really helps.	ok/tested pedro fgsch millert krw
  > [1 - next person to add a field to this struct gets whipped with a wet
  > noodle] (tedu@)

  ~ gdt_common.c                          ~ gdtvar.h

  TAGGED OPENBSD_3_8
  > Add bio framework (not functional yet).
  > Replace index into struct for device name with DEVNAME macro.
  > Started translating linux driver back into swedish for gdt->sc_more_proc
  > part.
  > Added initial scsi passthrough magic.
  > Added 2 new debug bits.
  > ok krw (marco@)

  ~ gdt_common.c                          

  TAGGED OPENBSD_3_8
  > oops, use correct size for GET_IOC_CHAN_DESC (marco@)

  ~ wdcreg.h                              

  TAGGED OPENBSD_3_8
  > Fix cut'n'paste abuse in previous typo fix; spotted by Constantine A.
  > Murenin,
  > thanks! (miod@)

  ~ gdt_common.c                          ~ gdtreg.h
  ~ gdtvar.h                              

  TAGGED OPENBSD_3_8
  > Add remaining glue for physical disk count.
  > Fix a bugglet in the size of GDT_IOCHAN_DESC structure. (marco@)

dev/pci

  ~ pcscp.c                               

  > Free one more resource on attach failure.
  > From NetBSD
  > ok martin@ (brad@)

  ~ if_sk.c                               ~ if_skreg.h

  > - Added/corrected register definitions for Yukon.
  > (Register information from Linux skge driver.)
  > From FreeBSD (brad@)

  ~ if_sk.c                               

  > Update if_oerrors if we encounter watchdog error.
  > From FreeBSD (brad@)

  ~ if_sk.c                               

  > - Interrupt handler now checks shared interrupt source and protects
  > the interrupt handler from NULL pointer dereference which was caused
  > by odd status word value. The status word can return 0xffffffff if the
  > cable is unplugged while Rx/Tx/auto-negotiation is in progress.
  > From FreeBSD (brad@)

  ~ if_sk.c                               ~ if_skreg.h

  > - Use SK_IS_GENESIS/SK_IS_YUKON where appropriate
  > - Remove SK_YUKON_FAMILY (brad@)

  ~ if_lge.c                              

  > - check for IFF_RUNNING being set before calling lge_init().
  > - call lge_setmulti() when enabling/disabling promisc mode or
  > if the IFF_ALLMULTI flag is set. (brad@)

  ~ if_bge.c                              

  > style (brad@)

  ~ files.pci                             + if_xge.c
  + if_xgereg.h                           

  > initial port of a driver for the Neterion Xframe-I 10Gb Ethernet adapter.
  > From NetBSD (brad@)

  ~ if_xge.c                              

  > OpenBSD-ify the attach function printf's. (brad@)

  ~ if_xge.c                              

  > always call xge_init() here, for the time being. (brad@)

  ~ if_iwi.c                              

  > call iwi_stop() at beginning of iwi_init() so that all rings are properly
  > reset when iwi_init() is called from iwi_ioctl() on ENETRESET.
  > this fixes a firmware panic when setting a WEP key multiple times for
  > instance.  fix some printf while i'm here.
  > pointed out by Kurt Miller (kurt@). (damien@)

  ~ if_nfereg.h                           

  > increase the Tx ring count from 64 to 256.
  > ok jsg@ damien@ (brad@)

  ~ if_em.c                               

  > replace magic value of 32 with EM_MAX_SCATTER. (brad@)

  ~ if_sk.c                               

  > A new function sk_yukon_tick(), to handle auto-negotiation properly for
  > Yukon-based adapters.
  > From FreeBSD (brad@)

  ~ if_sk.c                               

  > Added Rx/Tx FIFO flush support for Yukon
  > From FreeBSD (brad@)

  ~ if_sk.c                               

  > Activate Tx descriptor poll timer in order to protect possible loss
  > of SK_TXBMU_TX_START command. Previously the driver continuously issued
  > SK_TXBMU_TX_START when it notices pending Tx descriptors not processed
  > yet in interrupt handler. That approach would add additional PCI
  > write access overhead under high Tx load situations and it might fail
  > if the first SK_TXBMU_TX_START was lost and no interrupt is generated
  > from the first SK_TXBMU_TX_START command.
  > From FreeBSD (brad@)

  ~ if_ixgb.h                             ~ if_ixgb_osdep.h

  > sprinkle some tabs. (brad@)

  ~ if_ixgb.c                             

  > - when setting the interface address, only call ixgb_init() if the
  > interface
  > is not already running.
  > - remove splnet usage from ixgb_intr().
  > - replace magic value of 32 with IXGB_MAX_SCATTER. (brad@)

  ~ if_fxp_pci.c                          

  TAGGED OPENBSD_3_9
  > MFC:
  > Fix by miod@
  > Fix device id typo introduced in r1.41; found the hard way by Daniel
  > Hamlin. (brad@)

  ~ pcidevs.h                             ~ pcidevs_data.h

  TAGGED OPENBSD_3_9
  > regen (brad@)

  ~ pcidevs                               

  TAGGED OPENBSD_3_9
  > add the Neterion Xframe-II 10GbE product id. (brad@)

  ~ if_nfe.c                              

  TAGGED OPENBSD_3_9
  > MFC:
  > Fix by jsg@
  > Add work around for mbuf leak in the tx path until we
  > can come up with a better guess as to how the hardware works.
  > From Chuck Silvers. (brad@)

  ~ if_de.c                               ~ if_devar.h

  TAGGED OPENBSD_3_9
  > restore ALTQ support which was lost with rev 1.87 (brad@)

  ~ if_em.c                               

  TAGGED OPENBSD_3_9
  > - Remove unreachable bus_dmamap_unload() in em_dma_malloc().
  > - Set the dma_tag to NULL upon failure in em_dma_malloc().
  > - In em_dma_free(), return if dma_tag is NULL. (brad@)

  ~ if_em.c                               

  TAGGED OPENBSD_3_9
  > fix a typo and some KNF. (brad@)

dev/usb

  ~ usb_subr.c                            

  > Zero the memory allocated for subdevs[], preventing a quick trip to
  > ddb>. Discovered by trying 'usbdevs -v' while a broken card reader was
  > trying to attach. Broken device provided by Manuel Pata.
  > ok pedro@ deraadt@ (krw@)

  ~ usbdevs.h                             ~ usbdevs_data.h

  > Regen (pedro@)

  ~ usbdevs                               

  > Add Neodio 8-in-1 Flash Device Controller, okay dlg@ (pedro@)

  ~ ohci.c                                

  > fix interrupt out transfers
  > from netbsd, ok dlg (jolan@)

  ~ uaudio.c                              

  > free resources on close; found by Alexandre Ratchov <alex at caoua.org  > (mickey@)

  ~ if_ral.c                              

  TAGGED OPENBSD_3_8
  > MFC:
  > Fix by damien@
  > allows dhcpd to work when ral/ural is acting as a standalone AP (without
  > bridging).	it appears that ieee80211_input() enqueues packets in if_snd
  > without calling if_start().
  > ok damien@ (brad@)

kern

  ~ vfs_getcwd.c                          

  > Prune proc_isunder() and sys___getcwd(), no binary change (pedro@)

  ~ vfs_getcwd.c                          

  > Prune getcwd_common() and vn_isunder(), no binary change (pedro@)

  ~ vfs_getcwd.c                          

  > Prune remaining of the code, no binary change (pedro@)

  ~ vfs_lookup.c                          ~ vfs_subr.c
  ~ vfs_syscalls.c                        ~ vfs_sync.c

  > remove the simplelock argument from vfs_busy() which is currently not
  > used and will never be used this way in VFS
  > requested by and ok pedro, ok krw, biorn (sturm@)

  ~ kern_fork.c                           

  > Backout last commit, it makes us crash when a ptraced program forks.
  > (kettenis@)

  ~ vfs_getcwd.c                          

  > Make static functions global so DDB can see them (pedro@)

  ~ vfs_getcwd.c                          

  > Rename functions and move prototypes around (pedro@)

  ~ vfs_getcwd.c                          ~ vfs_vnops.c

  > Move vn_isunder() to vfs_vnops.c, that's the place for vn_* functions
  > (pedro@)

  ~ init_main.c                           

  > initialize the propolice cookie more carefully so -fstack-protector-all
  > will work; ok miod (deraadt@)

  ~ vfs_subr.c                            

  > forgot to remove this sentence from the comment
  > ok pedro (sturm@)

  ~ subr_pool.c                           ~ uipc_mbuf.c

  > remove drain hooks from pool.
  > 1.	drain hooks and lists of allocators make the code complicated
  > 2.	the only hooks in the system are the mbuf reclaim routines
  > 3.	if reclaim is actually able to put a meaningful amount of memory back
  > in the system, i think something else is dicked up.  ie, if reclaiming
  > your ip fragment buffers makes the difference thrashing swap and not,
  > your system is in a load of trouble.
  > 4.	it's a scary amount of code running with very weird spl requirements
  > and i'd say it's pretty much totally untested.  raise your hand if your
  > router is running at the edge of swap.
  > 5.	the reclaim stuff goes back to when mbufs lived in a tiny vm_map and
  > you could run out of va.  that's very unlikely (like impossible) now.
  > ok/tested pedro krw sturm (tedu@)

  ~ kern_descrip.c                        ~ kern_rwlock.c

  > add a name to rwlock so that we can tell where procs are getting stuck
  > without breaking into ddb.	doubles the size of rwlock [1], but moving
  > forward this really helps.	ok/tested pedro fgsch millert krw
  > [1 - next person to add a field to this struct gets whipped with a wet
  > noodle] (tedu@)

net

  ~ pf_table.c                            

  > fix creation of sub-anchors, e.g. if you create an anchor /foo/bar, create
  > only bar under foo, not /bar as well.
  > secondly, when using "load anchor from" from a sub-anchor, the loading
  > point should be relative to the sub-anchor doing the load (unless absolute
  > paths are used, of course).
  > from Boris Polevoy. probably a -stable candidate. (dhartmei@)

  ~ pf_norm.c                             

  TAGGED OPENBSD_3_9
  > MFC:
  > Fix by dhartmei@
  > fixup IP checksum when modifying IP header fields, based on a patch in
  > fbsd PR 93849 from Max Laier (brad@)

  ~ pf_norm.c                             

  TAGGED OPENBSD_3_8
  > MFC:
  > Fix by dhartmei@
  > fixup IP checksum when modifying IP header fields, based on a patch in
  > fbsd PR 93849 from Max Laier (brad@)

  ~ pf.c                                  

  TAGGED OPENBSD_3_8
  > MFC:
  > Fix by frantzen@
  > the TCP SACK option needs sequence number modulation (brad@)

  ~ pfkeyv2.c                             

  TAGGED OPENBSD_3_8
  > Fix bracketing messed up in KNF commit 1.86, allows sasyncd to reliably
  > set up pfkey promiscuous mode.
  > Diff from Nathanael <list-openbsd-tech at polymorpheus dot com> (mcbride@)

  ~ if_pfsync.c                           

  TAGGED OPENBSD_3_8
  > The SPI in a TDB is actually stored in network order. Make sa
  > synchronisation
  > work between little-endian and big-endian machines, and compare the spi
  > against SPI_RESERVED_MAX correctly.
  > Fix from Nathanael <list-openbsd-tech at polymorpheus dot com> (mcbride@)

nfs

  ~ nfs_syscalls.c                        

  > remove the simplelock argument from vfs_busy() which is currently not
  > used and will never be used this way in VFS
  > requested by and ok pedro, ok krw, biorn (sturm@)

  ~ nfs_node.c                            

  > add a name to rwlock so that we can tell where procs are getting stuck
  > without breaking into ddb.	doubles the size of rwlock [1], but moving
  > forward this really helps.	ok/tested pedro fgsch millert krw
  > [1 - next person to add a field to this struct gets whipped with a wet
  > noodle] (tedu@)

scsi

  ~ scsi_base.c                           

  > Zap a bunch of trailing whitespace. (krw@)

  ~ scsi_base.c                           

  > If a device says a MODE SENSE command worked fine, do *not* try a MODE
  > SENSE BIG command even if no page data was returned.  Just accept the
  > fact that the device does not have any such page data.
  > This is more in line with the spec, and unbreaks devices (like the
  > Neodio 8-in-1 USB Card Reader donated by Manuel Pata) which freak out
  > when getting a MODE SENSE BIG.
  > Should not break any 'working' device.
  > ok pedro@ dlg@ marco@ 'Looks correct.' miod@ (krw@)

sys

  ~ mount.h                               

  > remove the simplelock argument from vfs_busy() which is currently not
  > used and will never be used this way in VFS
  > requested by and ok pedro, ok krw, biorn (sturm@)

  ~ vnode.h                               

  > Rename functions and move prototypes around (pedro@)

  ~ pool.h                                

  > remove drain hooks from pool.
  > 1.	drain hooks and lists of allocators make the code complicated
  > 2.	the only hooks in the system are the mbuf reclaim routines
  > 3.	if reclaim is actually able to put a meaningful amount of memory back
  > in the system, i think something else is dicked up.  ie, if reclaiming
  > your ip fragment buffers makes the difference thrashing swap and not,
  > your system is in a load of trouble.
  > 4.	it's a scary amount of code running with very weird spl requirements
  > and i'd say it's pretty much totally untested.  raise your hand if your
  > router is running at the edge of swap.
  > 5.	the reclaim stuff goes back to when mbufs lived in a tiny vm_map and
  > you could run out of va.  that's very unlikely (like impossible) now.
  > ok/tested pedro krw sturm (tedu@)

  ~ rwlock.h                              

  > add a name to rwlock so that we can tell where procs are getting stuck
  > without breaking into ddb.	doubles the size of rwlock [1], but moving
  > forward this really helps.	ok/tested pedro fgsch millert krw
  > [1 - next person to add a field to this struct gets whipped with a wet
  > noodle] (tedu@)

ufs/mfs

  ~ mfs_vfsops.c                          

  > remove the simplelock argument from vfs_busy() which is currently not
  > used and will never be used this way in VFS
  > requested by and ok pedro, ok krw, biorn (sturm@)

ufs/ufs

  ~ ufs_quota.c                           

  > remove the simplelock argument from vfs_busy() which is currently not
  > used and will never be used this way in VFS
  > requested by and ok pedro, ok krw, biorn (sturm@)

uvm

  ~ uvm_mmap.c                            

  TAGGED OPENBSD_3_8
  > MFC:
  > Fix by kurt@
  > Fix a process datasize leak with MAP_FIXED. When zapping old mappings
  > call uvm_unmap_p instead of uvm_unmap so that it has the process
  > information
  > and can adjust vm_dused. (brad@)

  ~ uvm_mmap.c                            

  TAGGED OPENBSD_3_9
  > MFC:
  > Fix by kurt@
  > Fix a process datasize leak with MAP_FIXED. When zapping old mappings
  > call uvm_unmap_p instead of uvm_unmap so that it has the process
  > information
  > and can adjust vm_dused. (brad@)

  ~ uvm_pdaemon.c                         

  TAGGED OPENBSD_3_9
  > remove drain hooks from pool.
  > 1.	drain hooks and lists of allocators make the code complicated
  > 2.	the only hooks in the system are the mbuf reclaim routines
  > 3.	if reclaim is actually able to put a meaningful amount of memory back
  > in the system, i think something else is dicked up.  ie, if reclaiming
  > your ip fragment buffers makes the difference thrashing swap and not,
  > your system is in a load of trouble.
  > 4.	it's a scary amount of code running with very weird spl requirements
  > and i'd say it's pretty much totally untested.  raise your hand if your
  > router is running at the edge of swap.
  > 5.	the reclaim stuff goes back to when mbufs lived in a tiny vm_map and
  > you could run out of va.  that's very unlikely (like impossible) now.
  > ok/tested pedro krw sturm (tedu@)

xfs

  ~ xfs_config.h                          ~ xfs_dev-common.c

  > remove the simplelock argument from vfs_busy() which is currently not
  > used and will never be used this way in VFS
  > requested by and ok pedro, ok krw, biorn (sturm@)

== usr.bin =========================================================== 14/15 ==

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

cvs

  ~ rcs.c                                 

  > - fix a case in rcs_getrev() handling of branches where variant was not
  > being
  > increased and thus no progress could be made in the loop.
  > bug noticed in rcsdiff(1) by sturm@
  > ok joris@ (niallo@)

  ~ date.y                                

  > Use xasprintf and xfree.
  > ``But i thought it already went in...'' xsa@ (ray@)

diff

  ~ diff.1                                

  > - no need to escape these (jmc@)

getent

  ~ getent.c                              

  > getservbyport() wants network byte order; from NetBSD (otto@)

join

  ~ join.1                                

  > Spread some .Xr for the hard to stumble across lam(1).
  > jmc yeps. (hugh@)

mail

  ~ extern.h                              

  > Nuke a couple orphaned prototypes.
  > quoth deraadt: no problem (hugh@)

make

  ~ init.c                                ~ main.c

  > do not bother defining End if !CLEANUP, reminded by lint. (espie@)

mg

  ~ buffer.c                              ~ def.h
  ~ dir.c                                 ~ dired.c
  ~ file.c                                ~ grep.c

  > Make buffers store their own working directory. This makes things like
  > grep, compile, lint work as expected (act on current buffer's cwd).
  > Display this path when opening or replacing a file, rather than
  > relying on the user to guess, or remember. (kjell@)

  ~ fileio.c                              

  > len = strlen(foo); if foo[len - 1]... will do bad things if len == 0.
  > Avoid, and rearrange a test so it can't (hypothetically) overflow. (kjell@)

  ~ fileio.c                              ~ sysdef.h

  > Do some KNF, clean up some unused junk that has lying around for 40-odd
  > cvs revisions, and fix a few comments to match reality (kjell@)

  ~ fileio.c                              

  > correct one more comment (kjell@)

paste

  ~ paste.c                               

  > Use queue(3) macros instead of custom queue.
  > OK beck@ (ray@)

rcs

  ~ rcs.1                                 

  > rearrange the page to read better, and document branches/revisions/state;
  > ok xsa niallo ray (jmc@)

  ~ rcs.1                                 

  > better description for -b; ok joris (jmc@)

  ~ rcsprog.c                             

  > using *_FOREACH macros is really bad when we are actually
  > removing elements from the list inside the loop.
  > this has been documented *clearly* in queue(3).
  > okay niallo@ (joris@)

  ~ rcs.c                                 

  > - fix a case in rcs_getrev() handling of branches where variant was not
  > being
  > increased and thus no progress could be made in the loop.
  > bug noticed in rcsdiff(1) by sturm@
  > ok joris@ (niallo@)

  ~ ident.c                               

  > prevent a very, very stupid buffer overflow in ident_line().
  > fix for PR 5091
  > okay niallo@ (joris@)

  ~ diff3.c                               

  > respect TMPDIR; OK ray at .sa@)

  ~ date.y                                

  > Use xasprintf and xfree.
  > ``But i thought it already went in...'' xsa@ (ray@)

  ~ ci.c                                  ~ diff3.c
  ~ rcs.c                                 ~ rcsdiff.c

  > finish work wrt TMPDIR; use xasprintf() to simplify code while in there;
  > "looks fine" ray at .sa@)

  ~ ci.c                                  ~ co.c

  > o GNU ci returns 0 when a file is reverted (co -l file; ci file).
  > o Don't unlock file after revert.
  > o Fix spelling.
  > Passes extra tests in GNU rcstest.
  > OK niallo at .ay@)

  ~ diff.c                                ~ diff.h
  ~ rcsdiff.1                             ~ rcsdiff.c

  > - fix rcsdiff exit status on success/errors; now matches gnu/usr.bin/rcs.
  > - add DIAGNOSTICS section to rcsdiff.1 wrt exit codes.
  > ok ray@ and jmc@ for man page bits. (xsa@)

ssh

  - README.dns                            - README.tun

  > these files should not exist.  the developers have been given ample time
  > and warnings to integrate this into the manual page proper, but users who
  > find documentation missing keep being pointed at these files in the src
  > tree. we now delete the files, so that they will document these things
  > in the correct place.  you know who you are, and btw, jmc will help you
  > integrate the information into the man page if you just wrote simple bits
  > of text and asked nicely.. (deraadt@)

  ~ dh.c                                  

  > tighter DH exponent checks here too; feedback and ok markus@ (djm@)

  ~ auth-krb5.c                           

  > Add $OpenBSD$ in comment here too (dtucker@)

xlint

  ~ lint1/tree.c                          

  > be smarter about warning for null effects; kills a lot of spurious
  > warnings. ok cloder@ (otto@)

  ~ lint1/decl.c                          

  > Fix SEGV on prototypes containing functions; from NetBSD rev 1.18;
  > ok cloder@ (otto@)

  ~ lint1/init.c                          

  > Fix SEGV on scalar initialization with braces; also found in NetBSD
  > rev 1.23 as part of a much larger diff. ok cloder@ (otto@)

  ~ lint1/tree.c                          

  > Do not warn when casting a pointer to [unsigned] char *;
  > discussion with espie@; ok cloder@ (otto@)

  ~ xlint/xlint.c                         

  > use mkstemp, remove races wrt lint1 and cpp.
  > (check lint1 and cpp code, they use fopen/freopen to open that temp
  > file, and thus will happily overwrite a symlink)
  > okay cloder@ (espie@)

== usr.sbin ========================================================== 15/15 ==

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

afs

  ~ include/config.h                      ~ src/cf/bsd-vfs-busy.m4
  ~ src/include/config.h                  ~ src/include/config.h.in
  ~ src/nnpfs/bsd/nnpfs_dev-common.c      

  > remove the simplelock argument from vfs_busy() which is currently not
  > used and will never be used this way in VFS
  > requested by and ok pedro, ok krw, biorn (sturm@)

bgpd

  ~ name2id.c                             

  > The pftable name to ID functions should use the pftable_labels list and not
  > the rt_labels list. Luckily this bug was not visible because of the way the
  > pftable and rtlabel code works. Found by Thomas E. Spanjaard. (claudio@)

  ~ rde_attr.c                            

  TAGGED OPENBSD_3_9
  > MFC:
  > Fix by claudio@
  > Fix a memory leak when communities are set or deleted. This seems to fix
  > Sylvain Coutant memory issues. A lot of patience and testing by Sylvain.
  > ok claudio@ deraadt@ (brad@)

  ~ kroute.c                              

  TAGGED OPENBSD_3_9
  > MFC:
  > Fix by henning@
  > messages on the routing socket have different headers - rt_msghdr,
  > if_msghdr, etc. they only share the first 3 fields, and then differ.
  > this leads to a bug in RTM_IFINFO handling. We did abort on
  > rtm->rtm_errno != 0,
  > but if_msghdr has no errno, so we look at something in the data part
  > instead. Surprising that this didn't bite us before!
  > So we must only do these checks for RTM_ADD/CHANGE/DELETE that actually
  > use rt_msghdr. (brad@)

bind

  ~ bin/named/named.8                     

  > - document -i
  > - remove some whitespace whilst here
  > fixes pr #5087 (james strandboge)
  > ok jakob (jmc@)

config

  ~ mkmakefile.c                          

  > lint can't do assembly files, put them directly in OBJS.
  > okay theo (espie@)

ospfd

  ~ rde_spf.c                             

  TAGGED OPENBSD_3_9
  > MFC:
  > Fix by claudio@
  > Invalidating the full RIB and then only recalculating part of it is not
  > that smart. Kill the global rt_invalidate() instead clear only that part of
  > the RIB that is actually recalculated.
  > ok claudio@ deraadt@ (brad@)

  ~ kroute.c                              

  TAGGED OPENBSD_3_9
  > MFC:
  > Fix by henning@
  > only look at later rt_msghdr fields for those messages actually using
  > rt_msghdr, same as in bgpd (brad@)

pkg_add

  ~ OpenBSD/CollisionReport.pm            

  > more clues for people who update their system. (espie@)

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



More information about the owc mailing list