[odc] Daily src changes for 2006-04-25

ODC auto at squish.net
Wed Apr 26 08:00:28 BST 2006


OpenBSD src changes summary for 2006-04-25
==========================================

bin/chio                                bin/chmod
bin/cp                                  bin/date
bin/ed                                  bin/ln
bin/ps                                  bin/test
lib/libc                                lib/libm
libexec/tftpd                           regress/usr.bin
sbin/sysctl                             share/man
sys/arch/i386/conf                      sys/arch/mvme68k/mvme68k
sys/dev/ic                              sys/dev/pci
sys/netatalk                            sys/netinet
sys/sys                                 sys/uvm
usr.bin/cal                             usr.bin/cvs
usr.bin/ftp                             usr.bin/make
usr.bin/rcs                             usr.bin/ssh
usr.bin/xlint                           usr.sbin/bgpd
usr.sbin/ospfd                          

== bin =============================================================== 01/09 ==

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

chio

  ~ chio.c                                

  > various small lint cleanups; ok otto (deraadt@)

chmod

  ~ chmod.c                               

  > various small lint cleanups; ok otto (deraadt@)

cp

  ~ cp.c                                  ~ extern.h

  > various small lint cleanups; ok otto (deraadt@)

date

  ~ netdate.c                             

  > various small lint cleanups; ok otto (deraadt@)

ed

  ~ buf.c                                 ~ cbc.c
  ~ main.c                                

  > various small lint cleanups; ok otto (deraadt@)

ln

  ~ ln.c                                  

  > mark usage() __dead (deraadt@)

ps

  ~ fmt.c                                 ~ ps.c

  > various small lint cleanups; ok otto (deraadt@)

test

  ~ test.c                                

  > mode_t is your friend (deraadt@)

== lib =============================================================== 02/09 ==

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

libc

  ~ stdlib/strtonum.3                     

  > strtonum does not require limits.h
  > spotted by ray some time ago (tedu@)

libm

  ~ src/e_hypot.c                         ~ src/e_hypotf.c
  ~ src/e_pow.c                           ~ src/e_powf.c
  ~ src/s_ceil.c                          ~ src/s_ceilf.c
  ~ src/s_floor.c                         ~ src/s_floorf.c
  ~ src/s_modff.c                         ~ src/s_rint.c
  ~ src/s_rintf.c                         ~ src/s_trunc.c
  ~ src/s_truncf.c                        

  > Avoid shadowing of external symbols by local variables; whoever
  > thought up the names of the bessel functions deserves a beating.
  > ok deraadt@ (otto@)

== libexec =========================================================== 03/09 ==

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

tftpd

  ~ tftpd.c                               

  > kill setjmp() and alarm() methods, which are almost always inverted signal
  > races.  use poll() instead.  by marcus at nazgul.cheraadt@)

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

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

usr.bin

  + xlint/test-19.c                       + xlint/test-19.c.exp

  > A new, very exhaustive type conversion regression test. (cloder@)

  ~ xlint/Makefile                        ~ xlint/test-17.c
  ~ xlint/test-17.c.exp                   ~ xlint/test-2.c.exp
  ~ xlint/test-5.c.exp                    

  > Reflect latest lint commit, which fixes many type conversion false
  > positives and negatives. (cloder@)

  ~ rcs/Makefile                          + rcs/keywords.in
  + rcs/keywords.out                      

  > Regression tests for keywords for ci.
  > OK xsa@ (ray@)

  - rcs/keywords.in                       - rcs/keywords.out
  ~ rcs/Makefile                          

  > Back out previous tests, the keywords all expanded and destroyed
  > my tests. (ray@)

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

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

sysctl

  ~ sysctl.c                              

  > if the timeval has data, print it at the end of the line (deraadt@)

  ~ sysctl.c                              

  > for SENSOR_TIMEDELTA, store time in nanoseconds.  discussion with kettenis
  > mbalmer will have to cope. (deraadt@)

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

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

man

  ~ man9/pci_intr_map.9                   

  > reference to spl(9) (mickey@)

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

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

arch/i386/conf

  ~ GENERIC                               

  > enable udcf, why not (deraadt@)

arch/mvme68k/mvme68k

  ~ machdep.c                             

  > that extern was needed for now, miod (deraadt@)

  ~ machdep.c                             

  > Missed an include chunk. (miod@)

dev/ic

  ~ ami.c                                 

  > the error path for teh sync cache command should return, not try to blunder
  > on with the second half of the command. (dlg@)

dev/pci

  ~ pciide.c                              ~ pciide_piix_reg.h

  > Match PCI ids for Intel ICH7 SATA in AHCI mode.
  > Then ask for SATA IDE mode of operation.
  > X60/T60 Thinkpads are shipped in AHCI configuration by default,
  > this makes them work without changing a BIOS option. (jsg@)

netatalk

  ~ aarp.c                                

  > fix endian issue.  netbsd pr31051 via ian mcwilliam (tedu@)

netinet

  ~ ip_ipip.c                             ~ ip_mroute.c
  ~ ip_mroute.h                           

  > Remove virtual tunnel support from the mrouting code. The virtual tunnel
  > code breaks multicast on gif(4) interfaces and it is far better to
  > configure
  > a real gif(4) tunnel instead of a multicast tunnel as the latter is almost
  > not manageable. OK norby@, mblamer@ (claudio@)

sys

  ~ sensors.h                             

  > add a struct timeval to struct sensor.  any sensor which wishes the user
  > to know when it was last adjusted, can set a timestamp in here.  if tv_sec
  > is 0, the timestamp is invalid (maintaining the norm); ok dlg kettenis
  > (deraadt@)

uvm

  ~ uvm_km.c                              

  > limit pool backend preallocation to 2048 pages max (which only affects >2g
  > physmem); miod@ toby@ ok (mickey@)

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

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

cal

  ~ cal.c                                 

  > atoi/strtonum from han boetes (tedu@)

cvs

  ~ rcs.c                                 

  > check rcs_head_set() return value. (xsa@)

ftp

  ~ cmds.c                                ~ cmdtab.c
  ~ complete.c                            ~ domacro.c
  ~ fetch.c                               ~ ftp.c
  ~ main.c                                ~ ruserpass.c
  ~ stringlist.c                          ~ util.c

  > two strtol calls that were begging to be converted to strtonum (tedu@)

make

  ~ Makefile                              

  > same hack for .ln as .o (espie@)

rcs

  ~ rcsclean.c                            

  > The -r flag without an argument signifies HEAD for rcsclean.
  > OK xsa@ and joris@ (ray@)

  ~ rcsmerge.c                            

  > The -r flag without an argument signifies HEAD for rcsmerge as well.
  > OK joris@ (ray@)

  ~ ci.c                                  

  > style; (xsa@)

  ~ ci.c                                  ~ co.c
  ~ rcsclean.c                            ~ rcsdiff.c
  ~ rcsmerge.c                            ~ rcsprog.c
  ~ rcsutil.c                             ~ rlog.c

  > fatal() -> err()/errx() as we try to not depend on heavy cvs_log().
  > OK joriski. (xsa@)

  ~ ci.c                                  ~ rcsprog.c
  ~ rcsutil.c                             

  > quote strings consistently. (xsa@)

ssh

  ~ authfile.c                            ~ authfile.h
  ~ ssh.c                                 ~ sshconnect1.c
  ~ sshconnect2.c                         

  > Prevent ssh from trying to open private keys with bad permissions more than
  > once or prompting for their passphrases (which it subsequently ignores
  > anyway), similar to a previous change in ssh-add.  bz #1186, ok djm@
  > (dtucker@)

xlint

  ~ lint1/decl.c                          ~ lint1/lint.h
  ~ lint1/tree.c                          ~ lint2/chk.c

  > Rather than use the psize() macro everywhere (portable size macro which
  > only makes sense when -p option is specified to lint), add a new concept
  > and macro called rank(), which corresponds to the C99 definition of rank
  > for integral types. This has one main benefit, which is that on ILP32
  > platforms, if -p was not specified, lint did not warn when the code
  > converted from long to int, because they are the same size. This is
  > fixed.
  > While here, get rid of a false positive with warning 259, when converting
  > from an unsigned char to an int, for example. (cloder@)

  ~ lint1/err.c                           

  > Prettier error messages (cloder@)

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

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

bgpd

  ~ bgpd.h                                

  > Mark fatal and fatalx as __dead from deraadt@ (claudio@)

ospfd

  ~ kroute.c                              

  > ARGSUSED for event handler. (claudio@)

  ~ control.c                             

  > Mark event handlers with ARGSUSED to please lint. (claudio@)

  ~ ospfe.c                               

  > More lint cleanup. Kill unneeded stuff and sprinkle some ARGSUSED.
  > (claudio@)

  ~ rde.h                                 ~ rde_lsdb.c

  > The return value of lsa_num_links is an u_int16_t tnd not int. Found by
  > lint. (claudio@)

  ~ in_cksum.c                            ~ ospfd.h

  > Path the length of the packet as size_t to in_cksum(). Enforce that the
  > passed size is not bigger than 2^16 (limit of the used algorithm).
  > This removes some more lint warnings and makes sense. (claudio@)

  ~ ospfe.h                               ~ lsupdate.c

  > Kill unused function and add a ARGSUSED. (claudio@)

  ~ lsack.c                               ~ ospfe.h

  > Use size_t for buffer lenght passed to send_ls_ack() plus an ARGSUSED.
  > (claudio@)

  ~ log.h                                 

  > Mark fatal and fatalx as __dead from a similar diff for bgpd from deraadt@
  > (claudio@)

  ~ rde_spf.c                             

  > kill unused old debug code.
  > ok claudio@ (norby@)

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



More information about the odc mailing list