[odc] Daily src changes for 2010-07-23

ODC auto at squish.net
Sat Jul 24 07:00:01 BST 2010


OpenBSD src changes summary for 2010-07-23
==========================================

etc/MAKEDEV.common                      etc/etc.alpha/MAKEDEV
etc/etc.amd64/MAKEDEV                   etc/etc.armish/MAKEDEV
etc/etc.hppa/MAKEDEV                    etc/etc.hppa64/MAKEDEV
etc/etc.i386/MAKEDEV                    etc/etc.landisk/MAKEDEV
etc/etc.loongson/MAKEDEV                etc/etc.macppc/MAKEDEV
etc/etc.palm/MAKEDEV                    etc/etc.sgi/MAKEDEV
etc/etc.socppc/MAKEDEV                  etc/etc.sparc64/MAKEDEV
etc/etc.zaurus/MAKEDEV                  etc/systrace/usr_sbin_named
sbin/reboot                             sys/arch/amd64/amd64
sys/arch/i386/conf                      sys/arch/i386/i386
sys/arch/i386/isa                       sys/dev
sys/dev/acpi                            sys/dev/ata
sys/dev/atapiscsi                       sys/dev/ic
sys/dev/isa                             sys/dev/microcode
sys/dev/mii                             sys/dev/pci
sys/dev/usb                             sys/kern
sys/ufs/ffs                             usr.bin/cvs
usr.bin/ftp                             usr.bin/rcs
usr.bin/ssh                             usr.sbin/rpc.lockd
usr.sbin/smtpd                          

== etc =============================================================== 01/05 ==

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

MAKEDEV.common

  ~ MAKEDEV.common                        

  > Create video1 by default.  Can be useful e.g. on many laptops with
  > integrated cams where you want to use an external cam.  Suggested by Brad.
  > OK todd@, jakemsr@ (mglocker@)

etc.alpha/MAKEDEV

  ~ etc.alpha/MAKEDEV                     

  > Sync. (mglocker@)

etc.amd64/MAKEDEV

  ~ etc.amd64/MAKEDEV                     

  > Sync. (mglocker@)

etc.armish/MAKEDEV

  ~ etc.armish/MAKEDEV                    

  > Sync. (mglocker@)

etc.hppa/MAKEDEV

  ~ etc.hppa/MAKEDEV                      

  > Sync. (mglocker@)

etc.hppa64/MAKEDEV

  ~ etc.hppa64/MAKEDEV                    

  > Sync. (mglocker@)

etc.i386/MAKEDEV

  ~ etc.i386/MAKEDEV                      

  > Sync. (mglocker@)

etc.landisk/MAKEDEV

  ~ etc.landisk/MAKEDEV                   

  > Sync. (mglocker@)

etc.loongson/MAKEDEV

  ~ etc.loongson/MAKEDEV                  

  > Sync. (mglocker@)

etc.macppc/MAKEDEV

  ~ etc.macppc/MAKEDEV                    

  > Sync. (mglocker@)

etc.palm/MAKEDEV

  ~ etc.palm/MAKEDEV                      

  > Sync. (mglocker@)

etc.sgi/MAKEDEV

  ~ etc.sgi/MAKEDEV                       

  > Sync. (mglocker@)

etc.socppc/MAKEDEV

  ~ etc.socppc/MAKEDEV                    

  > Sync. (mglocker@)

etc.sparc64/MAKEDEV

  ~ etc.sparc64/MAKEDEV                   

  > Sync. (mglocker@)

etc.zaurus/MAKEDEV

  ~ etc.zaurus/MAKEDEV                    

  > Sync. (mglocker@)

systrace/usr_sbin_named

  ~ systrace/usr_sbin_named               

  > Permit a few more syscalls for named to run.
  > OK deraadt (ray@)

== sbin ============================================================== 02/05 ==

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

reboot

  ~ reboot.c                              

  > Block all signals before sending SIGTERM to all non-system processes.
  > This ensures that we reach the call to reboot(2) without being killed
  > by some other process.  OK deraadt@ nicm@ (millert@)

== sys =============================================================== 03/05 ==

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

arch/amd64/amd64

  ~ lapic.c                               

  > Omit more IPI code from non-MP kernels.  Found using --gc-sections.
  > ok deraadt@ (matthew@)

  ~ fpu.c                                 ~ machdep.c
  ~ process_machdep.c                     

  > Don't leak the contents of the FPU and SSE registers between processes.
  > ok deraadt@ (kettenis@)

  ~ fpu.c                                 

  > Change the strategy for using the FPU in the kernel.  The kernel FPU state
  > is now shared with all processes/threads.  As a result, you can now use the
  > FPU in true process context (instead of just in kernel threads), but you
  > need to make sure you restore the default FPU state before calling
  > fpu_kernel_exit() if you change rounding mode, precision or exception
  > masks.
  > Lots of discussion with thib@ and Mike Belopuhov.
  > ok thib@, deraadt@ (kettenis@)

arch/i386/conf

  ~ files.i386                            

  > AMD amd64 errara are not on the i386 install media, since there is often
  > not room.  we do place them on the amd64 install media though, since many
  > of them matter more for 64 bit mode. (deraadt@)

arch/i386/i386

  ~ lapic.c                               

  > Omit more IPI code from non-MP kernels.  Found using --gc-sections.
  > ok deraadt@ (matthew@)

  ~ machdep.c                             ~ process_machdep.c

  > Don't leak the contents of the FPU and SSE registers between processes.
  > ok deraadt@ (kettenis@)

arch/i386/isa

  ~ npx.c                                 

  > Don't leak the contents of the FPU and SSE registers between processes.
  > ok deraadt@ (kettenis@)

  ~ npx.c                                 

  > Change the strategy for using the FPU in the kernel.  The kernel FPU state
  > is now shared with all processes/threads.  As a result, you can now use the
  > FPU in true process context (instead of just in kernel threads), but you
  > need to make sure you restore the default FPU state before calling
  > fpu_kernel_exit() if you change rounding mode, precision or exception
  > masks.
  > Lots of discussion with thib@ and Mike Belopuhov.
  > ok thib@, deraadt@ (kettenis@)

dev

  ~ vscsi.c                               

  > matthew@ noted a possible misuse of the rwlock used to protect vscsi
  > from being opened by multiple processes at the same time. then he
  > pointed out that pool_get can sleep in the middle of scsi_cmd,
  > during which time the process processing the vscsi requests can go
  > away.
  > this uses a state variable to protect vscsi from multiple opens
  > which is checked by all the process and scsi midlayer entrypoints.
  > it avoids a potential sleep in the middle of the scsi_cmd handler
  > by moving to iopools.
  > lots of tweaks and feedback by matthew@ (dlg@)

dev/acpi

  ~ dsdt.c                                ~ acpidock.c

  > Fix searchname to handle short name segments, remove hack in acpidock.
  > (jordan@)

  ~ acpiec.c                              

  > Control/data registers were being mapped with swapped io space tags
  > Fix naming & order of tag to be clear
  > ok deraadt (jordan@)

  ~ acpiec.c                              

  > Don't initialize acpiec pointer in acpi_softc until registration succeeds
  > ok deraadt (jordan@)

dev/ata

  ~ ata.c                                 ~ ata_wdc.c
  ~ atareg.h                              ~ atavar.h
  ~ wd.c                                  ~ wdvar.h

  > Manuel Bouyer rescinded clauses 3 and 4 of his license text. (jsg@)

dev/atapiscsi

  ~ atapiscsi.c                           

  > Manuel Bouyer rescinded clauses 3 and 4 of his license text. (jsg@)

dev/ic

  ~ siop.c                                ~ siop_common.c
  ~ siopreg.h                             ~ siopvar.h
  ~ siopvar_common.h                      ~ wdc.c

  > Manuel Bouyer rescinded clauses 3 and 4 of his license text. (jsg@)

dev/isa

  ~ ad1848.c                              

  > ANSI/style(9), no binary change (jakemsr@)

dev/microcode

  ~ siop/siop.ss                          

  > Manuel Bouyer rescinded clauses 3 and 4 of his license text. (jsg@)

dev/mii

  ~ exphy.c                               ~ gentbi.c
  ~ icsphy.c                              ~ inphy.c
  ~ iophy.c                               ~ lxtphy.c
  ~ mii.h                                 ~ nsphy.c
  ~ nsphyter.c                            ~ qsphy.c
  ~ sqphy.c                               ~ tlphy.c
  ~ tlphyreg.h                            ~ tqphy.c
  ~ ukphy.c                               

  > Manuel Bouyer rescinded clauses 3 and 4 of his license text. (jsg@)

dev/pci

  ~ pciide.c                              

  > For suspend/reume, ATI IXP chips (might) need 0x40-0x56 saved.. but
  > these the 6 registers already saved (deraadt@)

  ~ if_iwn.c                              ~ if_iwnvar.h

  > Add suspend/resume logic.
  > based heavily on the wpi change by kettenis@
  > OK deraadt@, kettenis@, damien@ (phessler@)

  ~ pciide.c                              ~ pciide_acer_reg.h
  ~ pciide_apollo_reg.h                   ~ pciide_cmd_reg.h
  ~ pciide_cy693_reg.h                    ~ pciide_hpt_reg.h
  ~ pciide_pdc202xx_reg.h                 ~ pciide_piix_reg.h
  ~ pciide_sis_reg.h                      ~ siop_pci.c
  ~ siop_pci_common.c                     ~ siop_pci_common.h

  > Manuel Bouyer rescinded clauses 3 and 4 of his license text. (jsg@)

  ~ pcidevs                               

  > SIS 86C202 is not a VGA device, it's an AGP bridge. Use a better
  > description for this device.
  > ok kettenis@ (mlarkin@)

  ~ pcidevs_data.h                        ~ pcidevs.h

  > regen (mlarkin@)

dev/usb

  ~ uaudio.c                              

  > support for playback sync endpoints
  > sync endpoints allow for the sample clock on USB audio devices to not
  > be synchronized to the USB clock.  the sync endpoint gives information
  > on how much data to send to the data endpoint.  although devices that
  > require sync endpoints will "work" when the sync endpoint inormation
  > is ignored, there is a possibility of static, echoing, or any other
  > type of clock desync error.
  > requires some small changes to existing code:
  > * tighten up checks for whether a setting needs/supplies a sync
  > endpoint.  ignore settings that need a sync endpoint but don't
  > supply one.  previously all settings that require a sync endpoint
  > were ignored.
  > * if a sync endpoint is being used, use the same denominator for
  > fractional samples as the sync endpoint uses (2**16).
  > this only implements playback sync endpoints.  recording sync
  > endpoints are an odd concept: the driver would control the clock
  > rate of the device.  there may be such devices, but I can't even
  > imagine how that could be reliably implemented.  I guess you would
  > sync to the USB clock or the system clock ... but then, what's the
  > point? (jakemsr@)

  ~ uaudio.c                              

  > revert last commit, unbreaking the tree; was missing a macro definition
  > presumably from a missing commit to a header file. go ahead kettenis@ todd@
  > (sthen@)

  ~ uaudio.c                              ~ uaudioreg.h

  > bring back changes in uaudio.c r1.83, this time remembering to commit
  > uaudioreg.h.  sorry about that.
  > sync endpoints allow for the sample clock on USB audio devices to not
  > be synchronized to the USB clock.  the sync endpoint gives information
  > on how much data to send to the data endpoint.  although devices that
  > require sync endpoints will "work" when the sync endpoint inormation
  > is ignored, there is a possibility of static, echoing, or any other
  > type of clock desync error.
  > requires some small changes to existing code:
  > * tighten up checks for whether a setting needs/supplies a sync
  > endpoint.  ignore settings that need a sync endpoint but don't
  > supply one.  previously all settings that require a sync endpoint
  > were ignored.
  > * if a sync endpoint is being used, use the same denominator for
  > fractional samples as the sync endpoint uses (2**16).
  > this only implements playback sync endpoints.  recording sync
  > endpoints are an odd concept: the driver would control the clock
  > rate of the device.  there may be such devices, but I can't even
  > imagine how that could be reliably implemented.  I guess you would
  > sync to the USB clock or the system clock ... but then, what's the
  > point? (jakemsr@)

kern

  ~ kern_fork.c                           

  > Make sure the u area of new processes is zero-filled; this got lost in
  > 1.119. (miod@)

ufs/ffs

  ~ ffs_vfsops.c                          

  > Improve English.
  > Help by jmc, earlier version OK thib krw (ray@)

== usr.bin =========================================================== 04/05 ==

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

cvs

  ~ buf.c                                 ~ cvs.c
  ~ cvs.h                                 ~ diff.c
  ~ diff3.c                               ~ logmsg.c
  ~ rcs.c                                 ~ repository.c
  ~ repository.h                          ~ util.h
  ~ worklist.c                            ~ worklist.h

  > Rename a bunch of variables and functions, removing the cvs_/rcs_
  > prefixes to reduce differences between cvs and rcs files.
  > "yes!" nicm (ray@)

  ~ date.y                                

  > Clean up struct tm handling. Instead of calling gmtime/localtime
  > and making copies (among other weird dances), use reentrant versions.
  > Not being able to tell the time is a fatal error, so die if that
  > happens.
  > Diff originally from nicm.
  > OK nicm xsa (ray@)

  ~ add.c                                 ~ annotate.c
  ~ buf.c                                 ~ buf.h
  ~ checkout.c                            ~ client.c
  ~ commit.c                              ~ cvs.h
  ~ date.y                                ~ diff.c
  ~ diff.h                                ~ diff3.c
  ~ diff_internals.c                      ~ file.c
  ~ getlog.c                              ~ history.c
  ~ import.c                              ~ init.c
  ~ logmsg.c                              ~ rcs.c
  ~ rcs.h                                 ~ remote.c
  ~ server.c                              ~ trigger.c
  ~ update.c                              ~ util.c
  ~ util.h                                

  > Reduce variable/function name and whitespace differences between
  > cvs/rcs.
  > OK xsa zinovik (ray@)

ftp

  ~ fetch.c                               

  > Handle redirection to relative url's in the Location: header of http
  > responses. Yes, they are violating the rfc's. Yes they do exist anyway.
  > Also fix a memory leak when url_encode fails to malloc, by simply
  > err'ing out if so.
  > feedback and ok phessler@ (halex@)

rcs

  ~ buf.c                                 ~ diff3.c
  ~ rcs.c                                 ~ rcsprog.c
  ~ rcsprog.h                             ~ rcsutil.h
  ~ worklist.c                            ~ worklist.h

  > Rename a bunch of variables and functions, removing the cvs_/rcs_
  > prefixes to reduce differences between cvs and rcs files.
  > "yes!" nicm (ray@)

  ~ date.y                                

  > Sync with cvs:
  > Clean up struct tm handling. Instead of calling gmtime/localtime
  > and making copies (among other weird dances), use reentrant versions.
  > Not being able to tell the time is a fatal error, so die if that
  > happens.
  > Diff originally from nicm.
  > OK nicm xsa (ray@)

  ~ buf.c                                 ~ buf.h
  ~ ci.c                                  ~ co.c
  ~ date.y                                ~ diff.c
  ~ diff.h                                ~ diff3.c
  ~ ident.c                               ~ merge.c
  ~ rcs.c                                 ~ rcs.h
  ~ rcsclean.c                            ~ rcsdiff.c
  ~ rcsmerge.c                            ~ rcsnum.c
  ~ rcsprog.h                             ~ rcsutil.c
  ~ rlog.c                                

  > Reduce variable/function name and whitespace differences between
  > cvs/rcs.
  > OK xsa zinovik (ray@)

  ~ ci.c                                  

  > Plug mem leak.
  > OK nicm (ray@)

ssh

  ~ ssh.1                                 

  > Ciphers is documented in ssh_config(5) these days (dtucker@)

== usr.sbin ========================================================== 05/05 ==

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

rpc.lockd

  ~ lockd_lock.c                          

  > Manuel Bouyer rescinded clauses 3 and 4 of his license text. (jsg@)

smtpd

  ~ queue.c                               

  > smtpd should NOT fatal when it permanently fails a bounce delivery as this
  > can be a normal situation and will allow a local/authenticated user to
  > trigger the fatal on purpose ...
  > ignore permanently failed bounce deliveries since there's not much smtpd
  > can do anyway, that's what the previous queue code did.
  > experienced and reported by pirofti@ (gilles@)

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


More information about the odc mailing list