[owc] Weekly src changes ending 2007-05-13
OWC
auto at squish.net
Mon May 14 07:00:02 BST 2007
OpenBSD src changes summary for 2007-05-06 to 2007-05-13 inclusive
==================================================================
bin/ls distrib/notes
distrib/sets games/fortune
gnu/usr.bin/cvs kerberosV
lib/libc regress/sbin
sbin/disklabel sbin/fsck
sbin/fsck_ffs sbin/ipsecctl
sbin/isakmpd sbin/mount
sbin/reboot sbin/restore
sbin/wsconsctl share/man
sys/arch/alpha/include sys/arch/amd64/amd64
sys/arch/amd64/conf sys/arch/amd64/include
sys/arch/arm/arm sys/arch/arm/include
sys/arch/arm/xscale sys/arch/armish/dev
sys/arch/aviion/aviion sys/arch/hppa/dev
sys/arch/hppa/stand/libkern sys/arch/hppa64/stand/libkern
sys/arch/i386/i386 sys/arch/i386/isa
sys/arch/luna88k/include sys/arch/luna88k/luna88k
sys/arch/m88k/m88k sys/arch/macppc/dev
sys/arch/macppc/macppc sys/arch/macppc/pci
sys/arch/mips64/include sys/arch/mips64/mips64
sys/arch/mvme68k/include sys/arch/mvme88k/include
sys/arch/mvme88k/mvme88k sys/arch/powerpc/powerpc
sys/arch/sgi/include sys/arch/sgi/sgi
sys/arch/sgi/stand/boot sys/arch/sh/include
sys/arch/sh/sh sys/arch/sparc/conf
sys/arch/sparc/include sys/arch/sparc/sparc
sys/arch/sparc64/sparc64 sys/arch/vax/bi
sys/arch/vax/mba sys/arch/vax/qbus
sys/arch/vax/vax sys/dev
sys/dev/acpi sys/dev/cardbus
sys/dev/eisa sys/dev/ic
sys/dev/isa sys/dev/pci
sys/dev/pcmcia sys/dev/sdmmc
sys/dev/usb sys/dev/wscons
sys/isofs/udf sys/kern
sys/net sys/netinet
sys/netinet6 sys/scsi
sys/sys sys/ufs/ufs
sys/uvm usr.bin/apply
usr.bin/cap_mkdb usr.bin/cvs
usr.bin/find usr.bin/rcs
usr.bin/sdiff usr.bin/sendbug
usr.bin/tftp usr.bin/tip
usr.sbin/bgpd usr.sbin/hoststated
usr.sbin/ospfd usr.sbin/pkg_add
usr.sbin/popa3d usr.sbin/tcpdump
== bin =============================================================== 01/12 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/bin
ls
~ ls.1 ~ ls.c
~ ls.h ~ print.c
> Change the -g flag from a no-op to be POSIX conforming. We allow the
> -l flag to override -g regardless of its position on the command line
> for backwards compat with 4.3BSD. From NetBSD.
> OK jmc@, tom@, sobrado@ (millert@)
~ ls.1
> small update, now that -g does something; ok millert (jmc@)
== distrib =========================================================== 02/12 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/distrib
notes
~ m4.common
> Typo; jakemsr@ (miod@)
sets
~ lists/comp/mi
> sync (deraadt@)
~ lists/base/md.alpha ~ lists/comp/md.alpha
> sync (deraadt@)
~ lists/base/md.hppa ~ lists/comp/md.hppa
> sync (deraadt@)
~ lists/base/md.hp300 ~ lists/base/md.mac68k
~ lists/base/md.mvme68k ~ lists/comp/md.hp300
~ lists/comp/md.mac68k ~ lists/comp/md.mvme68k
> sync (deraadt@)
~ lists/base/md.hppa64 ~ lists/base/md.mvmeppc
~ lists/comp/md.hppa64 ~ lists/comp/md.mvmeppc
> sync (deraadt@)
~ lists/comp/mi
> sync (deraadt@)
== games ============================================================= 03/12 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/games
fortune
~ datfiles/fortunes
> brigader -> brigadier; (jmc@)
== gnu =============================================================== 04/12 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/gnu
usr.bin/cvs
~ src/diff.c
> when comparing two revisions of a directory tree, don't show dead files
> as new in the 2nd revision
> 'makes sense' niallo, ok xsa (sturm@)
== kerberosV ========================================================= 05/12 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/kerberosV
kerberosV
~ src/lib/krb5/kerberos.8
> Xr passwd(8) not kpasswd(8); David Leonard (millert@)
== lib =============================================================== 06/12 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/lib
libc
~ net/getaddrinfo.c
> Use syslog_r instead of syslog in getaddrinfo. Slowly making
> getaddrinfo reentrant.
> OK millert@ and deraadt at . (ray@)
~ gen/sysctl.3
> document net.inet.tcp.keepinittime;
> spotted by Daniel Ouellet; help/ok claudio (jmc@)
~ stdio/mktemp.3
> grammar,
> okay jmc@ (espie@)
~ net/getaddrinfo.c
> Clean up macros according to style. From tbert <bret dot lambert at
> gmail>.
> OK moritz@ and jaredy at . (ray@)
~ net/getaddrinfo.c
> Use size_t for strlen results.
> OK moritz at . (ray@)
~ stdio/putc.3
> Document errno usage. From Tobias Stoeckmann.
> Earlier diff OK jmc@, with nit by otto at . (ray@)
== regress =========================================================== 07/12 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/regress
sbin
~ ipsecctl/Makefile + ipsecctl/ike58.in
+ ipsecctl/ike58.ok
> Do not crash when lists include the "any" keyword. Reported by
> <ralf.horstmann at gmx.net>, thanks!
> Slightly different fix. Also add a regression test.
> ok mpf@ (hshoexer@)
== sbin ============================================================== 08/12 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/sbin
disklabel
~ disklabel.c
> Sync editit() with sendbug.
> OK krw at . (ray@)
fsck
~ fsck.8
> remove references to mount_vnd.8 and fsck_vnd.8 as these are not hooked
> into the build yet.
> ok jmc@ (fkr@)
fsck_ffs
~ pass5.c
> pass the right size arg to memset, otherwise we end up with a
> partially initialized summary info struct. ok millert@ pedro@ (otto@)
ipsecctl
~ parse.y
> Do not crash when lists include the "any" keyword. Reported by
> <ralf.horstmann at gmx.net>, thanks!
> Slightly different fix. Also add a regression test.
> ok mpf@ (hshoexer@)
isakmpd
~ isakmpd.8
> Document "M active|passive" ui command.
> ok jmc@ mpf@ (joel@)
~ ike_phase_1.c
> It was possible for phase 1 negotiation to fail due to lifetime duration
> mismatch without any log message stating so. This diff makes sure that
> all phase 1 negotiation failures due to proposal attribute mismatch are
> logged. Also change these messages from LOG_NEGOTIATION debug level 70 to
> always be logged (not just with debug).
> General idea OK hshoexer, tested here in production. (cloder@)
~ crypto.c
> Bump crypto buffer logging (before crypto/after crypto) to level 70 from
> level 30. This was a huge cause of log spam at level 30 and below, and is
> really not that useful. (cloder@)
mount
~ mount.8
> remove references to mount_vnd.8 and fsck_vnd.8 as these are not hooked
> into the build yet.
> ok jmc@ (fkr@)
reboot
~ reboot.c
> it's silly for reboot to poweroff. adapted from diff by diana eichert
> (tedu@)
restore
~ restore.c
> g/c unused vars (millert@)
wsconsctl
~ mouse.c ~ util.c
~ wsconsctl.h
> Introduce the mouse.rawmode and mouse.scale fields for
> touchscreen calibration via ioctls.
> ok miod@ (robert@)
== share ============================================================= 09/12 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/share
man
~ man4/iha.4
> add HISTORY
> ok jmc@ krw@ (jasper@)
~ man5/pf.conf.5
> Document the fact that 'allow-opts' applies to IPv6 now as well.
> ok jmc@ dhartmei@ henning@ deraadt@ claudio@ (mcbride@)
~ man9/Makefile ~ man9/microtime.9
> getbintime and getbinuptime are no longer so update the documentation as
> well.
> OK deraadt@ (claudio@)
~ man4/pppoe.4
> PPP Over Ethernet (not "over"). More consistent with pppoe(8) and the rfc.
> ok jmc@ (kili@)
~ man4/pci.4
> use the right capitalization for `SMBus'
> ok jmc@ (sobrado@)
~ man4/acx.4
> use the right capitalization for `CardBus'
> ok jmc@ (sobrado@)
~ man4/rln.4
> use the right capitalization for `WaveLAN'
> ok jmc@ (sobrado@)
~ man9/pci_conf_read.9
> use the right capitalization for `CardBus'
> ok jmc@ (sobrado@)
~ man4/puc.4
> use the right capitalization for `CardBus'
> ok jmc@ (sobrado@)
~ man4/malo.4
> use the right capitalization for `CardBus'
> ok jmc@ (sobrado@)
~ man9/mutex.9
> reference rwlock(9) (jsg@)
== sys =============================================================== 10/12 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/sys
arch/alpha/include
~ cpu.h
> garbage collect release_fpu() and synchronize_fpstate() (martin@)
~ mutex.h
> Alpha needs (will need) MUTEX_OLDIPL as well. (art@)
arch/amd64/amd64
~ identcpu.c ~ ipifuncs.c
~ mainbus.c ~ powernow-k8.c
+ mp_setperf.c
> Add the mp setperf mechanism to AMD64, like its i386 counterpart it allows
> all cpus in a system supporting frequency and voltage scaling to be scaled
> by the same amount corresponding to the user (or apmd on their behalf)
> performance level.
> This diff also teaches amd64 about acpi_hasprocfvs (ACPI has processor
> frequency and voltage scaling).
> It also moves initilization of the underlying setperf mechanism such
> as powernow to mainbus from the cpu indentification and initilization
> code inspired by similar changes dim@ made to i386 durring h2k6. This
> is necessary to implement the AMD recommended method for retreiving
> p_state data from the ACPI _PSS object (a diff comming soon). It will
> also simplify the potential addition of enhanced speedstep as found
> on newer intel processors with EMT64 capable of running OpenBSD/amd64.
> MP setperf functionality verifed by myself and Johan M:son Lindman <tybolt
> AT solace DOT miun DOT se> on opteron 265 and 270 systems respectively.
> General testing done by many others thanks!
> ok tedu, dim (gwk@)
~ genassym.cf
> Garbage collect ci_astpending; it's no longer used.
> ok miod@, art@ (kettenis@)
~ genassym.cf ~ intr.c
~ ipi.c ~ vector.S
> evcnt & friends were replaced by a proper interrupt counting mechanism 3
> years ago; ok miod (deraadt@)
~ trap.c
> Don't use LK_CANRECURSE for the kernel lock, okay miod@ art@ (pedro@)
arch/amd64/conf
~ files.amd64
> Add the mp setperf mechanism to AMD64, like its i386 counterpart it allows
> all cpus in a system supporting frequency and voltage scaling to be scaled
> by the same amount corresponding to the user (or apmd on their behalf)
> performance level.
> This diff also teaches amd64 about acpi_hasprocfvs (ACPI has processor
> frequency and voltage scaling).
> It also moves initilization of the underlying setperf mechanism such
> as powernow to mainbus from the cpu indentification and initilization
> code inspired by similar changes dim@ made to i386 durring h2k6. This
> is necessary to implement the AMD recommended method for retreiving
> p_state data from the ACPI _PSS object (a diff comming soon). It will
> also simplify the potential addition of enhanced speedstep as found
> on newer intel processors with EMT64 capable of running OpenBSD/amd64.
> MP setperf functionality verifed by myself and Johan M:son Lindman <tybolt
> AT solace DOT miun DOT se> on opteron 265 and 270 systems respectively.
> General testing done by many others thanks!
> ok tedu, dim (gwk@)
arch/amd64/include
~ cpu.h ~ intr.h
~ intrdefs.h
> Add the mp setperf mechanism to AMD64, like its i386 counterpart it allows
> all cpus in a system supporting frequency and voltage scaling to be scaled
> by the same amount corresponding to the user (or apmd on their behalf)
> performance level.
> This diff also teaches amd64 about acpi_hasprocfvs (ACPI has processor
> frequency and voltage scaling).
> It also moves initilization of the underlying setperf mechanism such
> as powernow to mainbus from the cpu indentification and initilization
> code inspired by similar changes dim@ made to i386 durring h2k6. This
> is necessary to implement the AMD recommended method for retreiving
> p_state data from the ACPI _PSS object (a diff comming soon). It will
> also simplify the potential addition of enhanced speedstep as found
> on newer intel processors with EMT64 capable of running OpenBSD/amd64.
> MP setperf functionality verifed by myself and Johan M:son Lindman <tybolt
> AT solace DOT miun DOT se> on opteron 265 and 270 systems respectively.
> General testing done by many others thanks!
> ok tedu, dim (gwk@)
~ cpu.h
> Garbage collect ci_astpending; it's no longer used.
> ok miod@, art@ (kettenis@)
~ cpu.h ~ intr.h
> evcnt & friends were replaced by a proper interrupt counting mechanism 3
> years ago; ok miod (deraadt@)
arch/arm/arm
~ softintr.c
> evcnt & friends were replaced by a proper interrupt counting mechanism 3
> years ago; ok miod (deraadt@)
arch/arm/include
~ softintr.h
> evcnt & friends were replaced by a proper interrupt counting mechanism 3
> years ago; ok miod (deraadt@)
arch/arm/xscale
~ i80321_intr.c
> diable -> disable (miod@)
arch/armish/dev
~ iq80321_pci.c
> evcnt & friends were replaced by a proper interrupt counting mechanism 3
> years ago; ok miod (deraadt@)
arch/aviion/aviion
~ av400_machdep.c
> Change the 88100 interrupt handlers to process DAEs with interrupts
> enabled,
> as done for DAEs not occuring during interrupts.
> Remove the check for unprocessed DAE on return from trap() in eh_common.S,
> since this can't happen. As a result, the return-from-trap code becomes
> identical on 88100 and 88110 systems. (miod@)
~ machdep.c
> Increment ncpus when hatching a secondary processor. (miod@)
arch/hppa/dev
~ mongoosevar.h
> evcnt & friends were replaced by a proper interrupt counting mechanism 3
> years ago; ok miod (deraadt@)
arch/hppa/stand/libkern
~ Makefile
> bcd.c died (deraadt@)
arch/hppa64/stand/libkern
~ Makefile
> bcd.c died (deraadt@)
arch/i386/i386
~ cpu.c
> cpu_init() is only used for the MP case (for now) (deraadt@)
~ process_machdep.c
> process_s87_to_xmm() is only needed if PTRACE (deraadt@)
~ gdt.c
> only supply ldt_alloc() and ldt_free() if USER_LDT (deraadt@)
~ mainbus.c
> remove 'struct sr_attach_args' as it's now elsewhere with softraid defined
> ok marco@ (todd@)
~ trap.c
> Don't use LK_CANRECURSE for the kernel lock, okay miod@ art@ (pedro@)
~ machdep.c
> Enable powernow on amd64 processors in the GENERIC.MP.
> ok tedu (gwk@)
arch/i386/isa
~ pccom.c
> put more foo_detach() routines where they are used (deraadt@)
~ pccom.c
> move com_activate() to where it belongs (deraadt@)
~ pccom.c
> backout while the amd64 puc situation is rethought (deraadt@)
~ pccom.c
> unused variable (deraadt@)
arch/luna88k/include
~ autoconf.h
> remove more junk in the setroot() code path (deraadt@)
~ board.h
> Remove unused code and definitions pasted over from mvme188 support, which
> do
> not apply to luna88k. (miod@)
arch/luna88k/luna88k
~ machdep.c
> Remove unused code and definitions pasted over from mvme188 support, which
> do
> not apply to luna88k. (miod@)
~ machdep.c
> Change the 88100 interrupt handlers to process DAEs with interrupts
> enabled,
> as done for DAEs not occuring during interrupts.
> Remove the check for unprocessed DAE on return from trap() in eh_common.S,
> since this can't happen. As a result, the return-from-trap code becomes
> identical on 88100 and 88110 systems. (miod@)
~ machdep.c
> Increment ncpus when hatching a secondary processor. (miod@)
arch/m88k/m88k
~ trap.c
> Don't use LK_CANRECURSE for the kernel lock, okay miod@ art@ (pedro@)
~ genassym.cf ~ process.S
> On multiprocessor kernels, update p->p_pcu on exit from the scheduler.
> (miod@)
~ eh_common.S
> Change the 88100 interrupt handlers to process DAEs with interrupts
> enabled,
> as done for DAEs not occuring during interrupts.
> Remove the check for unprocessed DAE on return from trap() in eh_common.S,
> since this can't happen. As a result, the return-from-trap code becomes
> identical on 88100 and 88110 systems. (miod@)
arch/macppc/dev
~ openpic.c
> The road to supporting bigendian openpic starts. (drahn@)
arch/macppc/macppc
~ autoconf.c
> uninitialized variable passed to setroot() (deraadt@)
~ autoconf.c
> no need to print boot device name twice (deraadt@)
arch/macppc/pci
~ ht.c
> correct dmesg output (deraadt@)
arch/mips64/include
~ _types.h ~ cpu.h
> Move sgo to __HAVE_CPUINFO.
> ok miod@ (kettenis@)
~ limits.h
> Remove symbols already defined in <sys/limits.h>; unbreaks build on sgi.
> ok miod (kettenis@)
arch/mips64/mips64
~ context.S ~ cpu.c
~ interrupt.c ~ lcore_float.S
~ trap.c
> Move sgo to __HAVE_CPUINFO.
> ok miod@ (kettenis@)
~ exception.S
> Check for astpending in u_general (userland traps) in addition to u_intr
> (hardware interrupts occuring while cpu is in usermode), fixes
> regress/sys/kern/sig-stop on sgi. (miod@)
~ context.S
> When entering the idle loop, do a real spl0 (processing soft interrupts)
> instead of simply lowering cpl to zero. (miod@)
~ interrupt.c
> Comment out ``generic'' interrupt routines, which are not used at the
> moment. (miod@)
~ vm_machdep.c
> In cpu_fork(), enable all known interrupt sources in cop0 sr/ic, so that
> kthreads start with interrupts enabled. (miod@)
~ exception.S ~ trap.c
> Change trap() into a void function, and restore pc from the trapframe
> instead
> of having trap() return it. This makes trap() consistent with other
> platforms, and its code smaller. (miod@)
arch/mvme68k/include
~ autoconf.h
> remove more junk in the setroot() code path (deraadt@)
arch/mvme88k/include
~ autoconf.h
> remove more junk in the setroot() code path (deraadt@)
arch/mvme88k/mvme88k
~ m187_machdep.c ~ m188_machdep.c
~ m197_machdep.c
> Change the 88100 interrupt handlers to process DAEs with interrupts
> enabled,
> as done for DAEs not occuring during interrupts.
> Remove the check for unprocessed DAE on return from trap() in eh_common.S,
> since this can't happen. As a result, the return-from-trap code becomes
> identical on 88100 and 88110 systems. (miod@)
~ machdep.c
> Increment ncpus when hatching a secondary processor. (miod@)
arch/powerpc/powerpc
~ pmap.c
> Ansi prototypes, not K&R. no binary difference. (drahn@)
arch/sgi/include
~ intr.h
> Move sgo to __HAVE_CPUINFO.
> ok miod@ (kettenis@)
arch/sgi/sgi
~ genassym.cf
> Move sgo to __HAVE_CPUINFO.
> ok miod@ (kettenis@)
arch/sgi/stand/boot
~ boot.c
> simple misspell, miod@ ok (jj@)
arch/sh/include
~ intr.h
> evcnt & friends were replaced by a proper interrupt counting mechanism 3
> years ago; ok miod (deraadt@)
arch/sh/sh
~ interrupt.c
> evcnt & friends were replaced by a proper interrupt counting mechanism 3
> years ago; ok miod (deraadt@)
arch/sparc/conf
~ Makefile.sparc
> Switch sparc to __HAVE_CPUINFO.
> miod@ tested (since I hacked it up blindly) and ok. (art@)
~ Makefile.sparc
> Remove cross-compiling stuff that is neither correct nor should have been
> committed. (art@)
arch/sparc/include
~ _types.h ~ cpu.h
~ psl.h
> Switch sparc to __HAVE_CPUINFO.
> miod@ tested (since I hacked it up blindly) and ok. (art@)
arch/sparc/sparc
~ cpu.c ~ cpuvar.h
~ genassym.cf ~ locore.s
~ trap.c
> Switch sparc to __HAVE_CPUINFO.
> miod@ tested (since I hacked it up blindly) and ok. (art@)
~ cpu.c
> Don't zero out the cpu_info in cpu_attach. We've already set a field in
> it (curproc) and it's allocated from zeroed memory anyway.
> deraadt@ ok and test. (art@)
arch/sparc64/sparc64
~ clock.c
> Whack "No counter-timer" message. These days, most UltraSPARC cpu's don't
> have one.
> ok deraadt@ (kettenis@)
~ locore.s
> Don't clear %tick register. Should not be necessary and undesirable since
> %tick is hyperprivileged on sun4v.
> ok jason@ (kettenis@)
~ autoconf.c
> Put names we get from the prom between double quotes when we print them.
> ok deraadt@ (kettenis@)
arch/vax/bi
~ bivar.h ~ if_ni.c
~ kdb.c
> evcnt & friends were replaced by a proper interrupt counting mechanism 3
> years ago; ok miod (deraadt@)
~ if_ni.c ~ kdb.c
> Don't use LK_CANRECURSE for the kernel lock, okay miod@ art@ (pedro@)
arch/vax/mba
~ mbavar.h
> evcnt & friends were replaced by a proper interrupt counting mechanism 3
> years ago; ok miod (deraadt@)
arch/vax/qbus
~ if_de.c
> evcnt & friends were replaced by a proper interrupt counting mechanism 3
> years ago; ok miod (deraadt@)
arch/vax/vax
~ intvec.s ~ subr.s
> evcnt & friends were replaced by a proper interrupt counting mechanism 3
> years ago; ok miod (deraadt@)
dev
~ softraid.c ~ softraidvar.h
> Assemble disk from user supplied metadata.
> Verify all metada to make sure it is correct; this includes sorting the
> chunks
> Add 2 new functions for printing uuid and metadata. (marco@)
~ softraid.c
> Add initial bits for assembly at boot time. (marco@)
~ vnd.c
> Use VNDLABELDEV() and not DISKLABELDEV() in writedisklabel() since
> VNDLABELDEV() is used in readdisklabel(). VNDLABELDEV() preserves the
> vndsimple() property encoded in the device. (krw@)
dev/acpi
~ acpicpu.c
> at least always print a new line, even if NOPSS (tedu@)
dev/cardbus
~ if_dc_cardbus.c
> dc_detach() is only used by cardbus code, so move it there; ok jsg
> (deraadt@)
~ if_rl_cardbus.c
> rl_detach() is only used for cardbus case (deraadt@)
~ com_cardbus.c ~ if_fxp_cardbus.c
> put more foo_detach() routines where they are used (deraadt@)
~ com_cardbus.c ~ if_fxp_cardbus.c
> repair com/fxp foo_detach() stuff, the hardest part being the pccom mess
> (deraadt@)
~ com_cardbus.c
> move com_activate() to where it belongs (deraadt@)
~ com_cardbus.c
> backout while the amd64 puc situation is rethought (deraadt@)
dev/eisa
~ aha1742.c
> ahb_debug should be inside #ifdef AHBDEBUG (deraadt@)
dev/ic
~ dp8390.c
> floating unused variable (except in debug case) (deraadt@)
~ dc.c
> dc_detach() is only used by cardbus code, so move it there; ok jsg
> (deraadt@)
~ siop.c
> turn off SIOP_STATS; ok krw (deraadt@)
~ ami.c
> ami chips have no mechanism to take commands off them once we've submitted
> them. this means that we cant reliably complete an io before the chip says
> we can because it can possibly complete later and overwrite memory it no
> longer owns, or write garbage to disk.
> so this diff forces the timeout on a scsi io to be as long as the chip
> should ever take, which is sixty seconds. after much discussion with marco
> we also decided to add a bit more so the time spent in the software runq
> wouldnt affect the ios lifetime too much.
> instead of completing ios out of the timeout, this simply warns the
> operator to check the state of the volumes if things are starting to slow
> down.
> ok marco@ (dlg@)
~ wdc.c ~ wdcvar.h
> atapi_print() is never used (deraadt@)
~ wdc.c ~ wdcvar.h
> wdcrestart() is never used (deraadt@)
~ rtl81x9.c ~ rtl81x9reg.h
> rl_detach() is only used for cardbus case (deraadt@)
~ fxp.c ~ fxpvar.h
~ com.c ~ comvar.h
> put more foo_detach() routines where they are used (deraadt@)
~ com.c ~ comvar.h
> move com_activate() to where it belongs (deraadt@)
~ re.c
> non-static inline functions waste space (deraadt@)
~ com.c ~ comvar.h
> backout while the amd64 puc situation is rethought (deraadt@)
~ ath.c ~ ar5xxx.c
~ ar5xxx.h
> remove the internal wrapper functions for ieee80211_mhz2ieee and
> ieee80211_ieee2mhz which also removes some dead code.
> figured out by deraadt@ (reyk@)
~ aic79xx.c
> diable -> disable (miod@)
~ malo.c
> Remove an obsolete comment which guesses about a register write an never
> was used anyway.
> OK claudio@ (mglocker@)
dev/isa
~ if_we.c
> old school media handling is no longer used; ok jsg (deraadt@)
~ fd.c
> 2nd one found: If you really wish to make a function __inline (which will
> be called once), please also make it static so that the compiler does not
> generate a 2nd one to export to outside callers! that is a stupid waste
> of space. (deraadt@)
dev/pci
~ if_tht.c
> typo (dlg@)
~ if_tht.c
> point the macros for the interrupt coalescing registers at the right
> location. they moved from what the doco says to somewhere else.
> pointed out by Nadav Shemer (thanks) (dlg@)
~ if_sk.c
> shrink code by not using __inline without static; ok kettenis (deraadt@)
~ pcidevs
> add Radeon RX1650 XT, Radeon RX1650 XT Secondary and SoundBlaster X-Fi
> (robert@)
~ pcidevs.h ~ pcidevs_data.h
> regen (robert@)
~ if_tht.c
> write the mac back to the chip when we bring it up. (dlg@)
~ if_tht.c
> add the multicast filter registers (dlg@)
~ if_tht.c
> make a start at dealing with interface flags. this toggles the hardware
> promiscuity depending on the if_flags IFF_PROMISC bit. (dlg@)
~ if_tl.c
> #if 0 a function that is only ever called in #if 0 (deraadt@)
~ if_em.c
> Revert rev 1.116. For some reason this caused the device to do DMA from/to
> random addresses in some cases, causing watchdog timeouts.
> tested by many; ok deraadt@ (kettenis@)
~ if_em.c ~ if_em_hw.c
> delete unused functions. typical vendor garbage driver..; ok kettenis
> (deraadt@)
~ ahci.c
> remove and hide code that isn't used.
> found by deraadt@ (dlg@)
~ pciide.c
> Don't skip channels that don't have the WDCF_IRQ_WAIT flag set. Apparently
> some chips generate interrupts we don't expect, but we need to ack those.
> Fixes the cdrom hangs on various sparc64 systems.
> tested by some, ok deraadt@ (kettenis@)
~ pciide.c
> Ditch acard_pci_intr() as nothing currently uses it.
> Pointed out by deraadt@ (jsg@)
~ if_msk.c ~ if_mskvar.h
> Take alignment restrictions into account when setting up RAM buffer
> addresses.
> This makes Yukon-2 FE chips work reliably.
> Pointed out by giovanni at gmail dot com. (kettenis@)
dev/pcmcia
~ if_ne_pcmcia.c
> places they get called are #ifdef notdef, so the code itself can be too
> (deraadt@)
~ if_ep_pcmcia.c
> #ifdef notdef functions which are only called in #ifdef notdef contexts
> (deraadt@)
dev/sdmmc
~ sdmmcdevs.h
> regen (mglocker@)
~ sdmmcdevs
> Add vendor and product; AboCom Systems, Inc. SDW11G. (mglocker@)
dev/usb
~ if_aue.c ~ if_auereg.h
~ if_udav.c ~ if_udavreg.h
~ if_url.c ~ if_urlreg.h
> More lockmgr -> rwlock low hanging fruit.
> ok dlg@ (krw@)
~ uts.c
> Make use of struct wsmouse_calibcoords and add ioctl support
> for getting and passing calibration values. Move the scale
> struct to uts_softc so that each device can have it's own values
> instead of using the global one.
> ok miod@ (robert@)
~ umsm.c
> sync comment with reality (fkr@)
dev/wscons
~ wsconsio.h
> add swapxy, resx and resy to struct wsmouse_calibcoords,
> so we can store these values here; ok miod@ (robert@)
isofs/udf
~ udf_vnops.c
> lockmgr_printinfo() calls only if DIAGNOSTIC (deraadt@)
kern
~ kern_sig.c
> First step in making the SCHED_LOCK not recursive.
> - Defer the sending of SIGCHLD and waking the parent when a process goes
> to sleep. We set a flag in the process "P_STOPPED" and schedule a
> timeout that checks all processes for this flag and sends SIGCHLD
> and wakes the parent on those processes that have this flag set.
> - Delay the waking of the parent in psignal on SIGCONT until we've
> released the SCHED_LOCK at the end.
> - In issignal, check for P_SINTR as a way to see we were called from
> the tsleep path with SCHED_LOCK locked and don't skip locking
> in that case. (This is slightly ugly, but it works until we have
> some better way).
> miod@ ok (after finding and fixing an MD bug on sgi) (art@)
~ kern_sig.c
> Needs an extern decl for generic soft interrupts platforms. (miod@)
~ vfs_conf.c
> struct mount *rootfs is never used (deraadt@)
~ kern_lock.c
> lockmgr_printinfo() is only called from #ifdef DIAGNOSTIC positions, so
> #ifdef DIAGNOSTIC it too (deraadt@)
~ vfs_subr.c
> kinfo_vgetfailed has not been used for > 8 years (deraadt@)
~ subr_autoconf.c
> evcnt_attach() is no longer used; ok miod (deraadt@)
~ kern_tc.c
> unused apis, very dangerous: getbinuptime() getbintime(), ok art
> (deraadt@)
~ subr_disk.c
> also print where (primary) swap and dumps are; ok miod (deraadt@)
~ kern_lock.c
> Don't use LK_CANRECURSE for the kernel lock, okay miod@ art@ (pedro@)
~ kern_rwlock.c
> print lock name in some panic messages so we know what's up
> ok krw marco pedro (tedu@)
net
~ route.c
> rtalloc2() and rtalloc_noclone() inside NPF > 0 (deraadt@)
~ route.c
> wildcard is nowhere used anymore. Figured out by deraadt@ (claudio@)
~ pf.c
> Routing headers are dangerous. Deal with them the same way as IPv4 options:
> drop all packets with routing headers unless the matching rule explicitly
> specifies 'allow-opts'.
> ok dhartmei@ henning@ deraadt@ claudio@ (mcbride@)
~ pf.c
> block ALL packets with rthdr0 in pf_test6(). We already do this
> in ip6_input(), but pf_test6() is called from bridge code as well.
> ok dhartmei@ henning@ deraadt@ claudio@ (mcbride@)
netinet
~ ip_carp.c
> do not fill RTAX_GATEWAY for RTM_DELETE, ok claudio (henning@)
~ tcp_subr.c
> variables used by #ifdef code should be inside #ifdef too (deraadt@)
~ tcp_subr.c
> tcp_iss usage is ifdef TCP_COMPAT_42, so the variable decl can be too
> (deraadt@)
~ ip_icmp.c
> totally unused legacy variable; ok claudio (deraadt@)
netinet6
~ ip6_input.c
> KILL all IPv6 packets with the rthdr0 extension header or multiple routing
> headers, regardless of forwarding path. It's the sane thing to do.
> ip6_check_rthdr0() function from claudio@
> ok deraadt@ claudio@ henning@ (mcbride@)
scsi
~ scsiconf.c
> all scsidebug_*-using code is under #ifdef, so the variables themselves
> should be too (deraadt@)
sys
~ limits.h
> UCHAR_MAX and USHRT_MAX should not have a 'U' suffix.
> That is only needed for constants that would not fit in a signed int.
> This is also consistent with UINT8_MAX and UINT16_MAX.
> Fixes PR 5467, submitted by ikz isr. (millert@)
~ proc.h
> First step in making the SCHED_LOCK not recursive.
> - Defer the sending of SIGCHLD and waking the parent when a process goes
> to sleep. We set a flag in the process "P_STOPPED" and schedule a
> timeout that checks all processes for this flag and sends SIGCHLD
> and wakes the parent on those processes that have this flag set.
> - Delay the waking of the parent in psignal on SIGCONT until we've
> released the SCHED_LOCK at the end.
> - In issignal, check for P_SINTR as a way to see we were called from
> the tsleep path with SCHED_LOCK locked and don't skip locking
> in that case. (This is slightly ugly, but it works until we have
> some better way).
> miod@ ok (after finding and fixing an MD bug on sgi) (art@)
~ time.h
> unused apis, very dangerous: getbinuptime() getbintime(), ok art
> (deraadt@)
~ device.h
> evcnt & friends were replaced by a proper interrupt counting mechanism 3
> years ago; ok miod (deraadt@)
~ systm.h
> Don't use LK_CANRECURSE for the kernel lock, okay miod@ art@ (pedro@)
ufs/ufs
~ ufs_extern.h ~ ufs_vnops.c
> unused function; ok pedro (deraadt@)
uvm
~ uvm_init.c
> Truncate the addresses for the deadbeef values so that they don't need
> to be page aligned and can contain more "noise".
> From mickey art@ ok (art@)
== usr.bin =========================================================== 11/12 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/usr.bin
apply
~ apply.c
> Remove shadowed, redundant variable. From Bret Lambert.
> OK jaredy at . (ray@)
cap_mkdb
~ getinfo.c
> Grammar. (ray@)
~ getinfo.c
> Replace manual loops with strcspn and plug realloc memory leaks.
> OK jaredy@ and moritz at . (ray@)
~ getinfo.c
> Compare chars with '\0', not NULL. (ray@)
~ getinfo.c
> typo in license found by ray; verified repair using my old archives...
> (deraadt@)
cvs
~ cvs.c
> - we do not have -H, zap it.
> - add missing `x' flag in getopt() and sort flags.
> - sync usage().
> Based on initial diff from Tobias Stoeckmann. (xsa@)
~ util.c ~ util.h
> cvs_splitpath() isn't used anymore; from Tobias Stoeckmann. (xsa@)
~ log.c
> Simplify cvs_vlog() to avoid endless fatal() loop on some error case.
> Problem spotted and fix submitted by Tobias Stoeckmann.
> OK niallo at . (xsa@)
~ logmsg.c
> Sync with editit from sendbug. (ray@)
~ logmsg.c
> cvs_logmsg_edit no longer fails with ECHILD.
> Add comment describing cvs_logmsg_edit usage. (ray@)
~ root.c
> fix format of the CVSROOT string in comments.
> from Tobias Stoeckmann. OK niallo at . (xsa@)
~ root.c
> typo in previous commit; (xsa@)
~ rcsnum.c
> Change 0x30 to more readable '0'.
> OK xsa at . (ray@)
find
~ find.c ~ main.c
> ftsopen -> fts_open; partly from Alexander Holupirek, documentation/5472
> ok simon art (jmc@)
rcs
~ rcsnum.c
> Change 0x30 to more readable '0'.
> OK xsa at . (ray@)
sdiff
~ edit.c
> Sync with editit from sendbug. (ray@)
sendbug
~ sendbug.c
> Save and restore signals once instead of each time fork fails with
> EAGAIN. (ray@)
~ sendbug.c
> Restore signals before returning. (ray@)
~ sendbug.c
> Simplify some loops.
> OK jaredy@ and moritz at . (ray@)
~ sendbug.c
> Be more consistent about when dmesg is attached to template.
> Specifically, -V. (ray@)
~ sendbug.c
> Instead of returning -1 and setting errno to ECHILD if the editor
> has a non-zero exit status, just return the exit status on success.
> Hard errors still return -1. If the editor is killed, return -1
> and set EINTR. (ray@)
~ sendbug.c
> If fork fails with EAGAIN, ignore errno and fail. Looping just
> worsens the problem.
> cloder@ and deraadt@ agree. (ray@)
tftp
~ tftp.c ~ main.c
> Delinting from tbert <bret dot lambert at gmail dot com>
> OK moritz@ (ray@)
tip
~ acu.c
> Prevent core dump if strpbrk() returns NULL. Reported to bugs@ by
> Pav Lucistnik <pav at FreeBSD>, thanks!
> OK otto at . (ray@)
== usr.sbin ========================================================== 12/12 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/usr.sbin
bgpd
~ bgpd.c ~ kroute.c
~ rde.c ~ rde_attr.c
~ rde_decide.c ~ rde_filter.c
~ rde_prefix.c ~ rde_rib.c
~ rde_update.c ~ util.c
> Various spelling fixes from Stuart Henderson. (claudio@)
hoststated
~ pfe.c
> fix potential NULL deref.
> spotted by Ching-Feng Wang <cfw at telepaq.com>.
> ok reyk@ (pyr@)
ospfd
~ ospfd.conf.5
> transposition; ok jmc@ (stevesk@)
pkg_add
~ pkg_create
> let pkg_create generate the correct description if the HOMEPAGE is empty.
> (espie@)
~ pkg_add
> finish removing old libdepends checks. (espie@)
~ OpenBSD/Dependencies.pm
> share common code for system libs. (espie@)
~ pkg_add ~ OpenBSD/Dependencies.pm
> zap unused parameters. (espie@)
~ pkg_add.1 ~ OpenBSD/Dependencies.pm
~ OpenBSD/SharedLibs.pm
> remove -Fboguslibs, we haven't had unmarked libraries for years (espie@)
~ OpenBSD/Dependencies.pm ~ OpenBSD/SharedLibs.pm
> use one single list for system directories. (espie@)
~ OpenBSD/Delete.pm ~ OpenBSD/PkgCfl.pm
~ OpenBSD/Replace.pm
> kill old code to handle deprecated pkgcfl. (espie@)
~ OpenBSD/Delete.pm
> name code from keep_old_files, don't set pkgname directly. (espie@)
~ OpenBSD/PackingElement.pm
> remove non sensical eval {} (espie@)
~ OpenBSD/PackingElement.pm
> split common code into run_if_exists (espie@)
~ OpenBSD/Delete.pm ~ OpenBSD/PackageRepository.pm
~ OpenBSD/Replace.pm ~ OpenBSD/Temp.pm
> make sure all temporary file handling goes through OpenBSD::Temp (espie@)
~ pkg_create ~ OpenBSD/Temp.pm
> zap old unused code (espie@)
~ OpenBSD/Temp.pm
> take over tempfile/dir cleaning entirely (espie@)
~ OpenBSD/PkgCfl.pm
> PkgSpec::match does the right thing, so use it there too. (espie@)
~ OpenBSD/PkgCfl.pm
> since we always use PkgSpec::Match, there's no need to build anon subs.
> (espie@)
~ pkg_add ~ OpenBSD/Dependencies.pm
> refactor: create a solver object to handle state issues in a simpler way.
> (espie@)
~ OpenBSD/Dependencies.pm
> restore lost behavior (espie@)
~ OpenBSD/Dependencies.pm ~ OpenBSD/PackageLocator.pm
> we want to no longer expose PackageLocator::available(), lift interfaces
> up there, starting with match_spec. (espie@)
~ pkg_add ~ OpenBSD/PackageName.pm
~ OpenBSD/Update.pm
> make keep_most_recent explicit, do not try to pre-compute it at the
> stemlist
> level. (espie@)
~ pkg_info ~ OpenBSD/PackageName.pm
> do not expose stemlist internals, create find_partialstem function.
> (espie@)
~ pkg_add ~ pkg_info
~ OpenBSD/PackageLocator.pm ~ OpenBSD/Update.pm
> remove dependency on available_stems(), make it go thru PackageLocator
> (espie@)
~ pkg_add ~ OpenBSD/PackageRepository.pm
> lift up findstem to the PackageRepository (espie@)
~ OpenBSD/PackageName.pm
> zap old code (espie@)
~ pkg_info
> use repo->findstem (espie@)
~ OpenBSD/PackageInfo.pm
> avoid calling compile_stemlist directly (espie@)
~ OpenBSD/PackageRepository.pm
> cheap singleton for installed packages (espie@)
~ OpenBSD/PackageLocator.pm
> spec match works better with a list (espie@)
~ OpenBSD/PackageRepository.pm
> allow match_spec on repository (espie@)
~ OpenBSD/PackageRepository.pm ~ OpenBSD/PkgSpec.pm
> pass refs around instead of extending them. (espie@)
~ pkg_create
> strip extra quotes from all defines. (espie@)
~ OpenBSD/PkgSpec.pm
> somewhat sneaky: allow PkgSpecs to be objects, so that we can perform
> optimizations on them. For now, we just do the subpattern split once.
> (espie@)
~ OpenBSD/PackageLocator.pm ~ OpenBSD/PackageRepository.pm
> also allow Spec objects in Repository matches. (espie@)
~ OpenBSD/PkgSpec.pm
> oops, make the object variant functional (espie@)
~ OpenBSD/Dependencies.pm
> use OO version of spec match (espie@)
~ pkg ~ pkg_add
~ pkg_info ~ OpenBSD/PackingElement.pm
~ OpenBSD/PkgCfl.pm
> use PkgSpec objects most of the time (espie@)
~ OpenBSD/PackageLocator.pm ~ OpenBSD/PackageRepository.pm
> use match_ref consistently (espie@)
~ pod/OpenBSD::PkgSpec.pod
> document the object interface to PkgSpec. (espie@)
~ OpenBSD/PackageLocator.pm ~ OpenBSD/PackageRepository.pm
~ OpenBSD/PkgSpec.pm
> kill old PkgSpec interface (espie@)
~ OpenBSD/Dependencies.pm ~ OpenBSD/PackingElement.pm
> lift PkgSpec creation up into PackingElement::Dependency. (espie@)
~ OpenBSD/Replace.pm
> oops. forgot this one. Noticed by M.Pereboom. (espie@)
~ OpenBSD/PkgSpec.pm
> remove obsolete comment. (espie@)
~ OpenBSD/PackageRepository.pm ~ OpenBSD/PackageRepositoryList.pm
> provide match_spec, findstem, find_partialstem at the repository list
> level,
> with ld -L like semantics: we stop at the first repository which has the
> corresponding packages.. (espie@)
~ OpenBSD/PackingElement.pm
> normalize quoting in extra info. (espie@)
~ OpenBSD/PackageRepositoryList.pm
> tweak _first_of so it can take an extra $filter routine, so that we
> get the first set of results that are non empty after filtering
> properly. (espie@)
~ OpenBSD/PackageLocator.pm
> allow filtering while calling the Locator. (espie@)
~ pkg_add ~ OpenBSD/Update.pm
> turn the updater into an object (espie@)
~ OpenBSD/Update.pm
> move most of the code into one process_package method (espie@)
~ OpenBSD/Update.pm
> reorg the update code some more, so that it goes through a filtering
> of candidates. (espie@)
~ OpenBSD/PackageLocator.pm
> have PackageLocator defer to PackageRepositoryList, so that package
> availability uses normal linker rules. (espie@)
~ OpenBSD/PackageName.pm ~ OpenBSD/PkgSpec.pm
> move towards unifying stem searches and pkgspec searches. (espie@)
~ OpenBSD/PackageRepository.pm
> make caching of stemlists implicit (espie@)
~ OpenBSD/PackageLocator.pm ~ OpenBSD/PackageName.pm
~ OpenBSD/PackageRepositoryList.pm
> also create PartialStem objects, and create a generic
> PackageLocator->match($o), where $o can be a spec, a stem, a partialstem.
> (espie@)
~ pkg_add ~ pkg_info
~ OpenBSD/Dependencies.pm ~ OpenBSD/PackageInfo.pm
~ OpenBSD/PackageName.pm ~ OpenBSD/PackageRepository.pm
~ OpenBSD/PackageRepositoryList.pm ~ OpenBSD/Update.pm
> unify object lookup as locator->match($o) or $repo->match($o) (espie@)
~ OpenBSD/PackageLocator.pm ~ OpenBSD/PackageRepository.pm
~ OpenBSD/PackageRepositoryList.pm
> kill unused stuff (espie@)
~ OpenBSD/PackageName.pm
> kill more stuff (espie@)
~ OpenBSD/Update.pm
> next -> return (espie@)
~ OpenBSD/PackageLocator.pm ~ OpenBSD/PackageRepository.pm
~ OpenBSD/PackageRepositoryList.pm
> remove some specialness of the Filtering stuff. (espie@)
popa3d
~ standalone.c
> Delinting from tbert <bret dot lambert at gmail dot com>
> OK moritz@ (ray@)
tcpdump
~ print-ip6.c
> Fix the out of bounds check when parsing IPv6 headers. Fixes a SIGSEGV
> when parsing IPv6 headers with unknown or corrupted header options.
> OK henning@ mcbride@ (claudio@)
~ print-rt6.c
> Don't fail hard on non RH0 routing headers. The calculation of the length
> of RH headers is always the same. Now tcpdump just prints the type, length
> and segleft headers for unknown headers.
> OK henning@ mcbride@ (claudio@)
===============================================================================
More information about the owc
mailing list