[odc] Daily src changes for 2008-06-22

[ODC]

OpenBSD src changes summary for 2008-06-22
==========================================

gnu/lib/libiberty                       share/man
sys/arch/mips64/mips64                  sys/arch/mvme68k/dev
sys/arch/sh/sh                          sys/dev/ic
sys/dev/usb                             sys/net
sys/scsi                                sys/sys
usr.bin/cdio                            

== gnu =============================================================== 01/04 ==

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

lib/libiberty

  ~ include/ansidecl.h                    

  > Enable the use of __attribute__ ((__nonnull__)) unconditionally under
  > OpenBSD,
  > as has been done for the __sentinel__ attribute, since all toolchains
  > support
  > it. (miod@)

== share ============================================================= 02/04 ==

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

man

  ~ man8/release.8                        

  > As discussed on ICB, do not invoke make with several targets to make
  > sure nothing blows up
  > ok krw@ espie@ jmc@ johan@ (who came up with the same diff) (ajacoutot@)

== sys =============================================================== 03/04 ==

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

arch/mips64/mips64

  ~ db_machdep.c                          

  > Like alpha, db_stack_dump() on mips and sh needs complete frame information
  > and can not work from a stack address only. Mention this and abort instead
  > of
  > printing a wrong trace. (prefer kern.splassert=3 to =2 on these platforms)
  > (miod@)

arch/mvme68k/dev

  ~ vs.c                                  

  > One last forgotten FREE(). (miod@)

arch/sh/sh

  ~ db_trace.c                            

  > Like alpha, db_stack_dump() on mips and sh needs complete frame information
  > and can not work from a stack address only. Mention this and abort instead
  > of
  > printing a wrong trace. (prefer kern.splassert=3 to =2 on these platforms)
  > (miod@)

dev/ic

  ~ acx.c                                 

  > Update watchdog timer code from DragonFly.
  > From sephe at DragonFly
  > ok mglocker@ (brad@)

dev/usb

  ~ uhci.c                                

  > correct function names on some debug and panics. (fgsch@)

  ~ uvideo.c                              

  > If no frame gets queue for a certain time, return EINVAL to the userland
  > application which will cause it to exit instead let it stuck forever.
  > (mglocker@)

  ~ usbdevs.h                             ~ usbdevs_data.h

  > regen (brad@)

  ~ usbdevs                               

  > Add a few more QuickCam cams and touch up some of the existing entries.
  > (brad@)

  ~ usb_quirks.c                          

  > Remove the QuickCam Pro 5000 from the quirks list now that there
  > is a UVC driver.
  > ok robert@ (brad@)

net

  ~ pf_ioctl.c                            

  > Revert 1.203; it's not safe to blindly walk the tailq instead of the rbtree
  > for picking states to unlink as the tailq may contain unlinked states.
  > (mcbride@)

scsi

  ~ cd.c                                  ~ ch.c
  ~ scsi_base.c                           ~ ss.c
  ~ ss_mustek.c                           ~ ss_scanjet.c
  ~ st.c                                  

  > Replace MUSTEK_RETRIES, SCANJET_RETRIES and ST_RETRIES with SCSI_RETRIES,
  > all defined to 4. Replace some magic number 4's with SCSI_RETRIES. Fix a
  > few lines that were now too long. (krw@)

sys

  ~ malloc.h                              

  > remove MALLOC and FREE macros, as they are no more used in the tree.
  > tested on non i386/amd64 archs by miod.
  > ok miod@ (chl@)

== usr.bin =========================================================== 04/04 ==

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

cdio

  ~ cdio.1                                ~ cdio.c
  ~ extern.h                              ~ mmc.c

  > Automatically distinguish between CD-DA track and WAVE audio file writing
  > them
  > in TAO mode. For WAVE files we should skip header.
  > ok and comments by fgsch (av@)

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