[odc] Daily src changes for 2006-03-05

ODC auto at squish.net
Mon Mar 6 07:14:25 GMT 2006


OpenBSD src changes summary for 2006-03-05
==========================================

distrib/notes                           lib/libm
regress/usr.bin                         share/man
sys/arch/amd64/amd64                    sys/arch/amd64/conf
sys/arch/i386/i386                      sys/compat/common
sys/compat/freebsd                      sys/compat/linux
sys/compat/netbsd                       sys/compat/osf1
sys/compat/svr4                         sys/compat/ultrix
sys/dev/acpi                            sys/dev/ata
sys/dev/ic                              sys/dev/isa
sys/dev/mii                             sys/dev/pci
sys/dev/raidframe                       sys/dev/sbus
sys/dev/usb                             sys/kern
sys/miscfs/procfs                       sys/miscfs/specfs
sys/net                                 sys/netinet
sys/netinet6                            sys/netipx
sys/netnatm                             sys/ntfs
sys/scsi                                sys/ufs/ffs
sys/ufs/lfs                             usr.bin/cvs
usr.bin/rcs                             usr.bin/sdiff
usr.sbin/pkg_add                        

== distrib =========================================================== 01/07 ==

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

notes

  ~ alpha/hardware                        ~ amd64/hardware
  ~ cats/hardware                         ~ i386/hardware
  ~ macppc/hardware                       ~ sparc64/hardware

  > mention newer chipsets and HP Intel-based Gigabit NICs (brad@)

== lib =============================================================== 02/07 ==

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

libm

  ~ man/atan2.3                           

  > improvements from alexey dobriyan;
  > ok miod kettenis (jmc@)

== regress =========================================================== 03/07 ==

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

usr.bin

  ~ sdiff/Makefile                        

  > Enable more tests now that sdiff handles stdin; from Ray (otto@)

== share ============================================================= 04/07 ==

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

man

  ~ man4/man4.i386/intro.4                ~ man4/man4.amd64/intro.4

  > Mention nfe(4). (jsg@)

  ~ man4/vic.4                            

  > better wording; (jmc@)

  ~ man4/iic.4                            

  > piixpm works on amd64 too; from olivier cherrier
  > ok grange (jmc@)

  ~ man4/gpioow.4                         

  > add some articles; (jmc@)

  ~ man4/owid.4                           

  > tweaks; (jmc@)

  ~ man4/owtemp.4                         

  > tweaks; (jmc@)

  ~ man4/options.4                        

  > document USER_LDT;
  > requested by mpech
  > ok deraadt (jmc@)

  ~ man4/options.4                        

  > better wording from weingart; (jmc@)

== sys =============================================================== 05/07 ==

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

arch/amd64/amd64

  ~ db_interface.c                        

  > Hook acpi to ddb since we need to be able to do some live debugging.
  > Lots and lots of help from miod@ and deraadt@
  > ok deraadt@ miod@ (marco@)

arch/amd64/conf

  ~ GENERIC                               

  > reference the correct man page;
  > ok deraadt miod tom (jmc@)

arch/i386/i386

  ~ db_interface.c                        

  > Hook acpi to ddb since we need to be able to do some live debugging.
  > Lots and lots of help from miod@ and deraadt@
  > ok deraadt@ miod@ (marco@)

compat/common

  ~ vfs_syscalls_25.c                     

  > Use more queue macros rather than doing it by hand; ok otto@ krw@ (miod@)

compat/freebsd

  ~ freebsd_file.c                        

  > Use more queue macros rather than doing it by hand; ok otto@ krw@ (miod@)

compat/linux

  ~ linux_socket.c                        

  > Use more queue macros rather than doing it by hand; ok otto@ krw@ (miod@)

compat/netbsd

  ~ netbsd_file.c                         

  > Use more queue macros rather than doing it by hand; ok otto@ krw@ (miod@)

compat/osf1

  ~ osf1_mount.c                          

  > Use more queue macros rather than doing it by hand; ok otto@ krw@ (miod@)

compat/svr4

  ~ svr4_misc.c                           ~ svr4_socket.c
  ~ svr4_sockio.c                         

  > Use more queue macros rather than doing it by hand; ok otto@ krw@ (miod@)

compat/ultrix

  ~ ultrix_fs.c                           

  > Use more queue macros rather than doing it by hand; ok otto@ krw@ (miod@)

dev/acpi

  ~ acpibat.c                             

  > KNF nit. (marco@)

  ~ acpicpu.c                             

  > Update tables after event.
  > Free memory while at it.
  > Register notify handler and use NULL for PNP id. (marco@)

  ~ dsdt.c                                ~ dsdt.h
  ~ files.acpi                            + acpidebug.c
  + acpidebug.h                           

  > Hook acpi to ddb since we need to be able to do some live debugging.
  > Lots and lots of help from miod@ and deraadt@
  > ok deraadt@ miod@ (marco@)

dev/ata

  ~ wd.c                                  

  > Remove more NRND NetBSDisms.
  > ok marco@ (krw@)

dev/ic

  ~ ncr53c9x.c                            

  > Use more queue macros rather than doing it by hand; ok otto@ krw@ (miod@)

dev/isa

  ~ if_trtcm_isa.c                        

  > Use more queue macros rather than doing it by hand; ok otto@ krw@ (miod@)

dev/mii

  ~ ukphy.c                               

  > remove whitespace at the end of line. (brad@)

dev/pci

  ~ if_ixgb.c                             

  > tabs are holier than spaces. (brad@)

  ~ if_em.h                               ~ if_em_osdep.h

  > Sprinkle some tabs and a little cleaning. (brad@)

dev/raidframe

  ~ rf_openbsdkintf.c                     

  > Use more queue macros rather than doing it by hand; ok otto@ krw@ (miod@)

dev/sbus

  ~ esp_sbus.c                            

  > Use more queue macros rather than doing it by hand; ok otto@ krw@ (miod@)

dev/usb

  ~ usbdevs                               

  > Add another two EVDO devices. (jsg@)

  ~ usbdevs.h                             ~ usbdevs_data.h

  > regen (jsg@)

kern

  ~ uipc_mbuf.c                           

  > splimp -> splvm (brad@)

miscfs/procfs

  ~ procfs_subr.c                         ~ procfs_vnops.c

  > Use more queue macros rather than doing it by hand; ok otto@ krw@ (miod@)

miscfs/specfs

  ~ spec_vnops.c                          

  > Use more queue macros rather than doing it by hand; ok otto@ krw@ (miod@)

net

  ~ if_tun.c                              

  > change the interface type from IFF_POINTOPOINT to the more correct
  > type of IFF_TUNNEL (Encapsulation interface).
  > ok djm@ (brad@)

netinet

  ~ ip_output.c                           

  > revert unrelated change that snuck into the last commit. (brad@)

  ~ ip_input.c                            

  TAGGED OPENBSD_3_7
  > MFC:
  > Fix by mpf@
  > m_pullup the same amount we copied, not just the IP header.
  > Any pf(4) translation that modifies more than IP addresses,
  > was happening on the shared mbuf cluster. Thus we were
  > sending icmp errors with corrupted payload.
  > ok mpf@ markus@ dhartmei@ (brad@)

  ~ ip_input.c                            

  TAGGED OPENBSD_3_8
  > MFC:
  > Fix by mpf@
  > m_pullup the same amount we copied, not just the IP header.
  > Any pf(4) translation that modifies more than IP addresses,
  > was happening on the shared mbuf cluster. Thus we were
  > sending icmp errors with corrupted payload.
  > ok mpf@ markus@ dhartmei@ (brad@)

  ~ if_ether.h                            ~ in.c
  ~ in_pcb.c                              ~ in_var.h
  ~ ip_icmp.c                             ~ ip_input.c
  ~ ip_ipip.c                             ~ ip_output.c
  ~ raw_ip.c                              

  TAGGED OPENBSD_3_8
  > Use more queue macros rather than doing it by hand; ok otto@ krw@ (miod@)

netinet6

  ~ icmp6.c                               ~ in6.c
  ~ in6_ifattach.c                        ~ in6_var.h
  ~ ip6_input.c                           ~ ip6_output.c
  ~ mld6.c                                ~ nd6_nbr.c
  ~ nd6_rtr.c                             ~ raw_ip6.c

  > Use more queue macros rather than doing it by hand; ok otto@ krw@ (miod@)

netipx

  ~ ipx.c                                 ~ ipx_input.c
  ~ ipx_ip.c                              ~ ipx_outputfl.c
  ~ ipx_pcb.c                             ~ ipx_usrreq.c
  ~ spx_usrreq.c                          

  > Use more queue macros rather than doing it by hand; ok otto@ krw@ (miod@)

netnatm

  ~ natm_pcb.c                            

  > Use more queue macros rather than doing it by hand; ok otto@ krw@ (miod@)

ntfs

  ~ ntfs_subr.c                           

  > Use more queue macros rather than doing it by hand; ok otto@ krw@ (miod@)

scsi

  ~ cd.h                                  ~ sd.c
  ~ sdvar.h                               

  > Remove NetBSDism NRND. Pointed out by Mickey.
  > ok miod@ marco@ (krw@)

ufs/ffs

  ~ ffs_alloc.c                           

  > Simplify ffs_alloccgblk() a lot by not caring about allocating
  > rotationally optimal blocks. There's no point in doing that anymore.
  > Okay beck@, marco@, tedu@ and krw at .rious testing for a while. (pedro@)

ufs/lfs

  ~ lfs_inode.c                           ~ lfs_segment.c
  ~ lfs_vfsops.c                          

  > Use more queue macros rather than doing it by hand; ok otto@ krw@ (miod@)

== usr.bin =========================================================== 06/07 ==

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

cvs

  ~ rcs.c                                 ~ rcs.h

  > our deltatexts were sometimes different from gnu.  rcs_getrev() was
  > unconditionally doing keyword expansion, even when this wasn't what we
  > wanted e.g. in checkout_rev() and cvs_checkout_rev(). so:
  > - change semantics of rcs_getrev() to not do keyword expansion
  > - add an rcs.c api function rcs_kwexp_buf() which does this
  > - change both checkout_rev() and cvs_checkout_rev() to use this function.
  > eyeballed by xsa, joris and Ray Lai (niallo@)

  ~ rcs.c                                 

  > fatal() instead of returning -1 on failure in patch functions
  > ed_patch_lines() and rcs_patch_lines().  this can avoid segfault in certain
  > circumstances.
  > From: Ray Lai <ray at cyth.net  > discussed with xsa@ and joris@ (niallo@)

  ~ rcs.c                                 

  > in rcs_patch_lines(), ensure that rcs patch delete operations don't
  > leave the dlp pointing at a non-existant line, but instead point it to
  > the last one in our TAILQ.
  > this fixes an annoying bug in rcs_patch_lines() where dlp would be
  > sometimes set to NULL when really it shouldn't be, resulting in an
  > "invalid line specification in RCS patch" error.
  > ok xsa@ joris@ (niallo@)

  ~ diff3.c                               

  > oops, forgot the diff3.c portion of this diff:
  > fatal() instead of returning -1 on failure in patch functions
  > ed_patch_lines() and rcs_patch_lines().  this can avoid segfault in certain
  > circumstances.
  > From: Ray Lai <ray at cyth.net  > discussed with xsa@ and joris@ (niallo@)

rcs

  ~ ci.1                                  

  > wording fix from ray lai;
  > ok xsa (jmc@)

  ~ co.1                                  

  > add a description; from ray lai and myself
  > ok xsa (jmc@)

  ~ co.c                                  

  > our deltatexts were sometimes different from gnu.  rcs_getrev() was
  > unconditionally doing keyword expansion, even when this wasn't what we
  > wanted e.g. in checkout_rev() and cvs_checkout_rev(). so:
  > - change semantics of rcs_getrev() to not do keyword expansion
  > - add an rcs.c api function rcs_kwexp_buf() which does this
  > - change both checkout_rev() and cvs_checkout_rev() to use this function.
  > eyeballed by xsa, joris and Ray Lai (niallo@)

  ~ rcsdiff.c                             

  > now that rcs_getrev() isn't doing keyword expansion anymore and rcsdiff(1)
  > expects to be comparing files with expanded keywords, we need to add
  > a couple of calls to rcs_kwexp_buf() to rcsdiff.c.	two minor output
  > fixes in this diff also.
  > ok xsa@ eyeballed by Rai Lai (niallo@)

  ~ ci.c                                  

  > honour -m on initial revision.
  > ok xsa@ (niallo@)

  ~ ci.c                                  

  > GNU ci(1) doesn't allow dates younger than HEAD, neither on the command
  > line via -d<date> nor via stat mtime on the working file (plain -d).
  > add a check for this in checkin_update(), abort and warn the user
  > should we encounter such a situation. (niallo@)

  ~ ci.c                                  

  > checkin_parsekeyword() returns void so zap unnecessary return. (niallo@)

  ~ ci.c                                  

  > plug some potential resource leaks in error paths. (niallo@)

sdiff

  ~ sdiff.c                               

  > Handle stdin as arg; from Ray Lai; ok jaredy@ (otto@)

== usr.sbin ========================================================== 07/07 ==

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

pkg_add

  ~ pkg_add.1                             

  > - kill whitespace at eol
  > - kill double .Pp (jmc@)

  ~ Makefile.PL                           

  > Synch (espie@)

  ~ OpenBSD/PackingElement.pm             

  > avoid running mkfontdir/fc-cache if they don't exist. (espie@)

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



More information about the odc mailing list