[odc] Daily src changes for 2006-03-20

ODC auto at squish.net
Tue Mar 21 07:00:25 GMT 2006


OpenBSD src changes summary for 2006-03-20
==========================================

bin/ksh                                 distrib/notes
lib/libkvm                              regress/usr.bin
sbin/brconfig                           sbin/fsck
sbin/fsck_ext2fs                        sbin/fsck_ffs
sbin/fsck_msdos                         sbin/ifconfig
sbin/ipsecctl                           sbin/isakmpd
sbin/restore                            sys/arch/alpha/dev
sys/arch/alpha/pci                      sys/arch/i386/i386
sys/arch/macppc/dev                     sys/dev/ic
sys/dev/pci                             sys/net
usr.bin/cvs                             usr.bin/m4
usr.bin/rcs                             usr.bin/ssh
usr.bin/vi                              usr.bin/xlint
usr.sbin/sensorsd                       

== bin =============================================================== 01/08 ==

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

ksh

  ~ c_sh.c                                ~ syn.c

  > Fix potential null deref and out-of-bound access.
  > ok millert@ deraadt@ (otto@)

== distrib =========================================================== 02/08 ==

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

notes

  ~ alpha/hardware                        

  > sync (brad@)

== lib =============================================================== 03/08 ==

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

libkvm

  ~ kvm.c                                 ~ kvm_alpha.c
  ~ kvm_amd64.c                           ~ kvm_arm.c
  ~ kvm_hppa.c                            ~ kvm_i386.c
  ~ kvm_m68k.c                            ~ kvm_m88k.c
  ~ kvm_mips64.c                          ~ kvm_powerpc.c
  ~ kvm_private.h                         ~ kvm_sparc.c
  ~ kvm_sparc64.c                         ~ kvm_vax.c

  > for some of internal functions use paddr_t where phys addr is intended
  > insteadd of u_long; miod@ ok (mickey@)

== regress =========================================================== 04/08 ==

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

usr.bin

  ~ xlint/Makefile                        + xlint/test-16.c
  + xlint/test-16.c.exp                   

  > Add a regression test for false positive with ((x == y) == z), fixed in
  > r1.31 of usr.bin/xlint/lint1/tree.c (cloder@)

  ~ m4/Makefile                           + m4/gnuformat.m4
  + m4/gnuformat.out                      

  > add limited support for format builtin in gnu-m4 mode, because I'm fed
  > up of patching it away in various autoconf derivatives.
  > okay miod@ (espie@)

== sbin ============================================================== 05/08 ==

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

brconfig

  ~ brconfig.c                            

  > NetBSD Coverity CID 774: Don't increment a pointer *before* testing it for
  > NULL!
  > ok otto@ (dhill@)

fsck

  ~ fsck.c                                

  > NetBSD Coverity CID 1745: Fix memory leak.
  > yes otto@ (dhill@)

fsck_ext2fs

  ~ dir.c                                 ~ pass5.c

  > NetBSD Coverity CID 2298: Fix memory leak.
  > NetBSD Coverity CID 2299: Fix memory leak.
  > NetBSD Coverity CID 2301: Fix memory leak.
  > ok ray@ (dhill@)

fsck_ffs

  ~ utilities.c                           

  > NetBSD Coverity CID 2074: Fix memory leak.
  > ok deraadt@ (dhill@)

fsck_msdos

  ~ fat.c                                 

  > NetBSD Coverity CID 2302: Free fat if fatal error to avoid leak.
  > ok otto@ deraadt@ (dhill@)

ifconfig

  ~ ifconfig.c                            

  > Don't increment a pointer *before* testing it for NULL
  > ok deraadt@ (dhill@)

ipsecctl

  ~ ike.c                                 

  > When adding a connection, do not explicitly start that connection
  > using "t" and "c" fifo commands.  This is prone to a race when
  > adding several tunnels between the same peers.  Just let isakmpd
  > start that connection on its own (using the connection checker).
  > (hshoexer@)

  ~ ike.c                                 

  > When being verbose while deleting ike rules (-dv), print deletions instead
  > of
  > additions.	Suggested by david@ (hshoexer@)

isakmpd

  ~ init.c                                ~ isakmpd.c
  ~ monitor.c                             ~ monitor.h

  > make sure the command fifo is ready before isakmpd returns.  This
  > resolves a startup race when interacting with ipsecctl.  Suggested
  > by and discussed with moritz@
  > ok moritz@ (hshoexer@)

restore

  ~ main.c                                

  > NetBSD Coverity CID 2305: Fix memory leak.
  > ok deraadt@ (dhill@)

== sys =============================================================== 06/08 ==

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

arch/alpha/dev

  ~ sgmap_common.c                        ~ sgmapvar.h

  > factorize SGMAP-mapped DMA map creation and destroy code
  > ok miod@, additional testing jsg@
  > from NetBSD (martin@)

arch/alpha/pci

  ~ apecs_dma.c                           ~ cia_dma.c
  ~ lca_dma.c                             ~ tsp_dma.c

  > factorize SGMAP-mapped DMA map creation and destroy code
  > ok miod@, additional testing jsg@
  > from NetBSD (martin@)

arch/i386/i386

  ~ powernow-k7.c                         ~ powernow-k8.c
  ~ machdep.c                             

  > a bit of demagification. this moves the checks for the cpu scaling features
  > out of machdep and into powernow-k7 and -k8. machdep now just figures out
  > if its the right type of cpu before calling the powernow code which
  > figures out if the scaling is supported.
  > from gwk (dlg@)

arch/macppc/dev

  ~ smu.c                                 

  > Don't forget to unlock of reading the RTC fails.
  > Should fix panic on iMac G5 (iSight). (kettenis@)

dev/ic

  ~ ami.c                                 

  > shorten the setup of the xs timeout (dlg@)

  ~ ami.c                                 

  > the command queues are important so wrap them up in splbio (dlg@)

  ~ ami.c                                 

  > get rid of the CCB_READY check in ami_start, it is subtly broken and slows
  > things down. this path is easy to verify without this code so im just
  > dropping it. (dlg@)

  ~ ami.c                                 

  > move the drv_inq ioctl path to its own completion routine. all the cleanup
  > of the dmammem is now done in the ioctl, the completion routing simply
  > wakes it up. (dlg@)

  ~ ami.c                                 

  > switch the mgmt ioctl path to use the ioctl completion function. (dlg@)

  ~ ami.c                                 ~ amivar.h

  > remove the bits in the ccb for generalised handling of dmaable memory.
  > now that each code path that puts commands on the hardware has specific
  > completion routines, we dont have to deal with the memory magic in a
  > generic fashion. (dlg@)

  ~ ami.c                                 ~ amivar.h

  > shrink the ccb even more. instead of having a "wakeup" member to say when
  > the ioctl paths can wake up, check the state of the ccb instead. when the
  > ccb becomes READY again then the command has completed. (dlg@)

  ~ ami.c                                 ~ amivar.h

  > finer grained use of splbio in the ioctl paths. only use it when dealing
  > with the ccb queues rather than over the whole ioctl.
  > also get rid of the CMD_WAIT flag which is used to restrict the ioctl to
  > only one caller at a time. this is actually protecting from overuse of
  > ccbs for management tasks (like bioctl) by making sure the scsi paths
  > will always have their ccbs. replace the flag with a real lock so that
  > callers can sleep till the mgmt ccb is free, rather than returning EBUSY
  > (dlg@)

  ~ ami.c                                 

  > remove kernel printfs that say "unable to allocate memory" just before
  > returning ENOMEM and the like. userland can say that for us while keeping
  > our dmesg clean. (dlg@)

dev/pci

  ~ pci.c                                 

  > de-register. (brad@)

  ~ if_lge.c                              ~ if_sf.c
  ~ if_sis.c                              ~ if_tl.c
  ~ if_vge.c                              ~ if_vr.c
  ~ if_wb.c                               

  > de-register. (brad@)

net

  ~ if.c                                  ~ if_tun.c
  ~ route.c                               ~ route.h

  > introduce rt_if_remove which takes care of routing table updates for an
  > interface that is removed. use that from if.c and if_tun.c instead of
  > re-implementing in the latter case. ok claudio (henning@)

== usr.bin =========================================================== 07/08 ==

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

cvs

  ~ buf.c                                 ~ buf.h

  > delint; cvs_buf_write_stmp() will never return non-zero:
  > - remove superfluous error handling for cvs_buf_write_stmp()
  > - change cvs_buf_write_stmp() return type to void
  > ok joris@ xsa@ (niallo@)

m4

  ~ eval.c                                ~ extern.h
  ~ gnum4.c                               ~ main.c
  ~ misc.c                                ~ trace.c

  > report function m4errx, that shows app-specific data (filename/linenumber)
  > before the error message. Use it to simplify code.
  > okay miod@ (espie@)

  ~ eval.c                                ~ extern.h
  ~ gnum4.c                               ~ m4.1
  ~ main.c                                ~ mdef.h

  > add limited support for format builtin in gnu-m4 mode, because I'm fed
  > up of patching it away in various autoconf derivatives.
  > okay miod@ (espie@)

rcs

  ~ ci.c                                  ~ rcsdiff.c

  > delint; cvs_buf_write_stmp() will never return non-zero:
  > - remove superfluous error handling for cvs_buf_write_stmp()
  > - change cvs_buf_write_stmp() return type to void
  > ok joris@ xsa@ (niallo@)

  ~ rlog.c                                

  > add missing xfree()'s after we are done with cvs_strsplit();
  > spotted by ray@, OK joris at .sa@)

  ~ rcsprog.c                             

  > make `rcs -a' use cvs_strsplit() for easiness; OK joris at .sa@)

ssh

  ~ auth2-gss.c                           

  > GSSAPI related leaks detected by Coverity via elad AT netbsd.org;
  > reviewed by simon AT sxw.org.uk; deraadt@ ok (djm@)

  ~ gss-genr.c                            

  > more GSSAPI related leaks detected by Coverity via elad AT netbsd.org;
  > reviewed by simon AT sxw.org.uk; deraadt@ ok (djm@)

  ~ monitor.c                             

  > memory leaks detected by Coverity via elad AT netbsd.org;
  > deraadt@ ok
  > that should be all of them now (djm@)

  ~ gss-serv.c                            

  > last lot of GSSAPI related leaks detected by Coverity via
  > elad AT netbsd.org; reviewed by simon AT sxw.org.uk; deraadt@ ok (djm@)

  ~ key.c                                 

  > (really) last of the Coverity diffs: avoid possible NULL deref in
  > key_free. via elad AT netbsd.org; markus@ ok (djm@)

  ~ auth.c                                ~ key.c
  ~ misc.c                                ~ packet.c
  ~ ssh-add.c                             

  > in a switch (), break after return or goto is stupid (deraadt@)

  ~ key.c                                 

  > djm did a typo (deraadt@)

  ~ ssh-rsa.c                             

  > in a switch (), break after return or goto is stupid (deraadt@)

  ~ channels.c                            ~ clientloop.c
  ~ monitor_wrap.c                        ~ monitor_wrap.h
  ~ serverloop.c                          ~ ssh.c
  ~ sshpty.c                              ~ sshpty.h

  > sprinkle u_int throughout pty subsystem, ok markus (deraadt@)

  ~ auth1.c                               ~ auth2.c
  ~ sshd.c                                

  > sprinkle some ARGSUSED for table driven functions (which sometimes must
  > ignore their args) (deraadt@)

  ~ channels.c                            ~ monitor.c
  ~ session.c                             ~ session.h
  ~ ssh-agent.c                           ~ ssh-keygen.c
  ~ ssh-rsa.c                             ~ ssh.c
  ~ sshlogin.c                            

  > annoying spacing fixes getting in the way of real diffs (deraadt@)

  ~ monitor.c                             

  > spacing (deraadt@)

  ~ channels.c                            

  > x11_fake_data is only ever used as u_char * (deraadt@)

  ~ deattack.c                            

  > spacing (deraadt@)

  ~ dns.c                                 

  > cast xstrdup to propert u_char * (deraadt@)

  ~ canohost.c                            ~ match.c
  ~ ssh.c                                 ~ sshconnect.c

  > be strict with tolower() casting (deraadt@)

  ~ channels.c                            ~ fatal.c
  ~ kex.c                                 ~ packet.c
  ~ serverloop.c                          

  > spacing (deraadt@)

  ~ ttymodes.c                            

  > spacing (deraadt@)

vi

  ~ vi/v_ex.c                             

  > if we're in visual mode reading a command, check the termination value
  > of v_tcmd() and bail if it's not TERM_OK as opposed to in a more
  > specific case.  this is based on the NetBSD ^C fix but after
  > discussion with otto at .hile it did not affect the specific crash
  > it is more correct. (pvalchev@)

xlint

  ~ lint1/tree.c                          

  > Lint should not complain about expressions of the form: ((x == y) == z)
  > because this is perfectly legal in C.  deraadt@ was seeing lint false
  > positives with the S_ISREG macro from stat.h.  So set m_eqwarn to 0 for
  > the EQ operator.  See xlint regress case #16. (cloder@)

== usr.sbin ========================================================== 08/08 ==

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

sensorsd

  ~ sensorsd.c                            

  > add two missing function declarations
  > change sensor_status to sensorsd_status
  > ok deraadt@ (dhill@)

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



More information about the odc mailing list