[owc] Weekly src changes ending 2008-01-20
OWC
auto at squish.net
Mon Jan 21 07:00:01 GMT 2008
OpenBSD src changes summary for 2008-01-13 to 2008-01-20 inclusive
==================================================================
distrib/notes distrib/sets
etc/netstart etc/rc
gnu/usr.bin/binutils lib/libc
sbin/bioctl sbin/disklabel
sbin/newfs sbin/pflogd
sbin/scsi share/man
sys/arch/alpha/alpha sys/arch/amd64/amd64
sys/arch/amd64/include sys/arch/aviion/aviion
sys/arch/i386/i386 sys/arch/i386/include
sys/arch/luna88k/luna88k sys/arch/m88k/include
sys/arch/macppc/conf sys/arch/mips64/mips64
sys/arch/mvme88k/include sys/arch/mvme88k/mvme88k
sys/arch/sgi/dev sys/arch/sparc64/conf
sys/arch/sparc64/dev sys/arch/sparc64/include
sys/arch/sparc64/sparc64 sys/arch/zaurus/conf
sys/conf sys/dev
sys/dev/cardbus sys/dev/ic
sys/dev/isa sys/dev/pci
sys/dev/usb sys/kern
sys/msdosfs sys/net
sys/sys sys/ufs/ufs
usr.bin/cvs usr.bin/mixerctl
usr.bin/ranlib usr.bin/ssh
usr.sbin/dhcpd usr.sbin/snmpctl
usr.sbin/snmpd usr.sbin/tcpdump
usr.sbin/usbdevs
== distrib =========================================================== 01/09 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/distrib
notes
~ sgi/hardware ~ sgi/prep
> - update the hardware list with reality
> - tweak a sentence in prep now that we have decent console support
> ok miod@ (jasper@)
~ zaurus/hardware
> Mention supported radio clocks.
> ok jmc. (mbalmer@)
sets
~ lists/base/md.i386 ~ lists/comp/md.i386
> sync (pvalchev@)
== etc =============================================================== 02/09 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/etc
netstart
~ netstart
> Execute rtsol after turning up trunk(4) and vlan(4) interfaces so they're
> taken into consideration for rtsol.
> ok reyk@ dlg@ (brad@)
rc
~ rc
> create pflog0 whenever pf is enabled, not just when pflogd_flags!=NO
> fixes spamlogd with pflogd disabled.
> ok henning (sthen@)
== gnu =============================================================== 03/09 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/gnu
usr.bin/binutils
~ Makefile.bsd-wrapper
> manpages need directories.
> okay kettenis@ (espie@)
~ Makefile.in
> Fix a race by making maybe-all-bfd depend on all-bfd and maybe-all-opcodes
> depend on all-opcodes. Problem found by espie at .
> ok espie@ (kettenis@)
== lib =============================================================== 04/09 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/lib
libc
~ gen/readpassphrase.c
> Use TCSAFLUSH instead of TCSANOW when turning echo back on so that
> there is no chance of output that was written but still pending to
> be displayed. This is what the original getpass(3) did. (millert@)
== sbin ============================================================== 05/09 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/sbin
bioctl
~ bioctl.c
> Add initial scaffold for RAID 0. No IO just yet.
> Much prodding todd (marco@)
disklabel
~ editor.c
> remove "unused variable" warning
> ok krw@ (chl@)
newfs
~ newfs.c
> s/fragements/fragments/ in usage().
> 'please fix' jmc (sthen@)
pflogd
~ pflogd.8
> clear up pidfile(-p) option form
> tweaked by jmc, ok henning (okan@)
scsi
~ libscsi.c ~ libscsi.h
> remove unused functions
> from tobias@
> ok miod@ tobias@ (chl@)
== share ============================================================= 06/09 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/share
man
~ man4/mpi.4
> Dell SAS 6/iR is an LSI SAS1068E which has been verified to use
> the generic LSI PCI ids so it should work fine.
> ok dlg@ (jsg@)
~ man4/upgt.4
> Replace firmware version 2.5.8.0 with 2.13.1.0. Firmware API slightly
> changes.
> You need to update the firmware package to upgt-firmware-1.1.tgz.
> (mglocker@)
~ man4/xf86.4
> macpppc -> macppc
> ok jmc (sthen@)
~ man4/man4.sparc64/fhc.4
> Mention environ(4). (kettenis@)
~ man4/man4.sparc64/Makefile + man4/man4.sparc64/environ.4
> Add environ(4). (kettenis@)
~ man4/man4.sparc64/central.4 ~ man4/man4.sparc64/clkbrd.4
~ man4/man4.sparc64/environ.4 ~ man4/man4.sparc64/fhc.4
> english nouns do not need to be capitalised;
> sync fhc.4; (jmc@)
~ man4/man4.sparc64/ebus.4
> kill strange workaround; (jmc@)
~ man4/upgt.4
> Add more potentially supported devices. Reports are welcome. (mglocker@)
~ man4/upgt.4
> Add one more device which could work. Fix comment typo in usbdevs while
> there. (mglocker@)
== sys =============================================================== 07/09 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/sys
arch/alpha/alpha
~ debug.s ~ locore.s
~ multiproc.s ~ pal.s
~ prom_disp.s
> Replace STABS debug info (which causes problems with truncated relocations)
> with some limited DWARF2 generating .file and .line symbols. Makes it
> possible
> to build kernels with debug information again.
> ok miod@ (kettenis@)
arch/amd64/amd64
~ dkcsum.c
> Remove extra whitespace after comment; makes file identical to i386 version
> (except for RCS string of course). (kettenis@)
arch/amd64/include
~ mpbiosreg.h ~ mpconfig.h
> Move some definitions from mpconfig.h to mpbiosreg.h (where they are on
> i396).
> ok marco@, miod@ (kettenis@)
arch/aviion/aviion
~ machdep.c
> Add a machdep.cputype sysctl, which returns the processor type (0 for
> 88100,
> 1 for 88110), for userland to have an easy way to figure out. (miod@)
arch/i386/i386
~ machdep.c
> Fix build with DEBUG defined.
> ok kettenis (mikeb@)
~ pmap.c
> Don't reserve space for msgbuf and bootargs in pmap_bootstrap(), since we
> do those reservations again (and now for real) in init386().
> ok weingart@ (kettenis@)
~ mainbus.c
> Kill <machine/mp.h>.
> ok marco@ (kettenis@)
~ machdep.c
> Work around an intel BIOS bug where it locates the ACPI tables in
> a region of memory marked free for use by the OS.
> Ok beck@, kettenis@, and --ok && ++!!ok tedu at . (weingart@)
arch/i386/include
- mp.h
> Kill <machine/mp.h>.
> ok marco@ (kettenis@)
arch/luna88k/luna88k
~ machdep.c
> Add a machdep.cputype sysctl, which returns the processor type (0 for
> 88100,
> 1 for 88110), for userland to have an easy way to figure out. (miod@)
arch/m88k/include
~ cpu.h
> Add a machdep.cputype sysctl, which returns the processor type (0 for
> 88100,
> 1 for 88110), for userland to have an easy way to figure out. (miod@)
arch/macppc/conf
~ GENERIC
> Activate upgt(4) for macppc. Tested for a while on my PowerBook.
> OK dlg@ (mglocker@)
arch/mips64/mips64
~ pmap.c
> Revert r1.15 and flush pages (if cached) in pmap_remove_pv on every
> invocation,
> instead of when removing the last page. It looked correct, but is defeated
> by ``index'' cache invalidates. (miod@)
arch/mvme88k/include
~ param.h
> Remove unused cpumod variable. (miod@)
arch/mvme88k/mvme88k
~ machdep.c
> Remove unused cpumod variable. (miod@)
~ machdep.c
> Add a machdep.cputype sysctl, which returns the processor type (0 for
> 88100,
> 1 for 88110), for userland to have an easy way to figure out. (miod@)
arch/sgi/dev
~ if_mec.c
> Clean up comments. (jsing@)
arch/sparc64/conf
~ GENERIC
> Enable umbg(4).
> ok claudio (mbalmer@)
~ files.sparc64
> Add environ(4). (kettenis@)
~ GENERIC
> Add commented out entry for environ(4). (kettenis@)
~ GENERIC
> When I say commented out, I should comment it out before comitting.
> (kettenis@)
arch/sparc64/dev
~ sab.c
> Make sabtty(4) adopt the speed configured in OpenFirmware when it is the
> console.
> ok miod@ (kettenis@)
~ sbus.c ~ central.c
~ fhc.c ~ upa.c
> Print names we get from the prom within double quotes.
> ok marco@ (kettenis@)
+ environ.c
> First attempt to write a driver for the environment device found on
> Enterprise 3000/3500/4000/4500/5000/5500/6000/6500. (kettenis@)
~ z8530tty.c
> Print consoleness the canonical way. (kettenis@)
~ pci_machdep.c ~ psycho.c
~ pyro.c ~ schizo.c
> Make host bridges provide their own implementation of pci_conf_read() and
> pci_conf_write() and give pyro(4) an implementation suitable for PCIe. For
> psycho(4) and schizo(4), go back to the origional implementation. This
> gets
> rid of the 'tagshift' member of pci_chipset_tag_t, and clears the way for
> sun4v. (kettenis@)
~ clkbrd.c ~ clkbrdvar.h
> Add temperature sensor. (kettenis@)
arch/sparc64/include
~ pcb.h
> Simplify spilling register windows into the pcb by storing the stack
> pointer
> seperately from the window and copying out data back to the stack by using
> the
> T_RWRET softtrap and rwindow_save(). (kettenis@)
~ pci_machdep.h
> Make host bridges provide their own implementation of pci_conf_read() and
> pci_conf_write() and give pyro(4) an implementation suitable for PCIe. For
> psycho(4) and schizo(4), go back to the origional implementation. This
> gets
> rid of the 'tagshift' member of pci_chipset_tag_t, and clears the way for
> sun4v. (kettenis@)
~ cpu.h
> Simplify CLKF_INTR by making it look at the saved interrupt level,
> instead of the stack pointer. (kettenis@)
arch/sparc64/sparc64
~ genassym.cf ~ locore.s
~ trap.c
> Simplify spilling register windows into the pcb by storing the stack
> pointer
> seperately from the window and copying out data back to the stack by using
> the
> T_RWRET softtrap and rwindow_save(). (kettenis@)
arch/zaurus/conf
~ GENERIC
> Enable umbg(4) on the Zaurus.
> ok mglocker, claudio. (mbalmer@)
conf
~ files
> Add initial scaffold for RAID 0. No IO just yet.
> Much prodding todd (marco@)
dev
~ softraid.c
> Remove printf that doesn't make sense. (marco@)
~ softraid.c ~ softraidvar.h
+ softraid_raid0.c
> Add initial scaffold for RAID 0. No IO just yet.
> Much prodding todd (marco@)
~ softraid.c
> debug kruft (marco@)
dev/cardbus
~ if_dc_cardbus.c
> - Remove a unwanted slash from a error message printf
> - Fix interrupt string printing in the dmesg output
> ok dlg@ (brad@)
dev/ic
~ ac97.c
> Check for a negative devinfo index, which could get here through
> mixer_ioctl.
> ok jakemsr@ (deanna@)
~ ac97.c ~ ac97.h
> bring in lots of ac97(4) and auvia(4) updates from NetBSD
> user visible changes:
> - when setting up ac97(4) mixer items, for some features like surround
> sound, check if the feature if supported by the codec and only add
> items for that feature if it is
> - auvia(4) now supports multichannel playback with at least some
> multichannel ac97(4) codecs
> fixes problems for me and ajacoutot with different auvia(4)
> devices/codecs. no regressions reported by testers.
> if you have an ac97(4) audio device and are using mixerctl.conf(5),
> you may want/need to update it.
> ok ratchov (jakemsr@)
~ re.c ~ rtl81x9reg.h
> - Store the hw revision into a new softc field in re_attach() and reference
> the softc field wherever it is necessary.
> - Use a switch case in re_setmulti() to make things a bit easier to read.
> ok henning@ dlg@ (brad@)
~ z8530tty.c
> Print consoleness the canonical way. (kettenis@)
dev/isa
~ sbdsp.c
> return earlier if the device index is negative
> ok ratchov (jakemsr@)
dev/pci
~ envy.c ~ sv.c
> protect against an invalid device index possibly leading to kernel
> crash. inspired by deanna's similar bug fix in ac97.
> ok ratchov (jakemsr@)
~ if_tht.c
> Zap two usless printfs.
> ok dlg@ (thib@)
~ auich.c ~ auixp.c
~ auvia.c ~ auviavar.h
> bring in lots of ac97(4) and auvia(4) updates from NetBSD
> user visible changes:
> - when setting up ac97(4) mixer items, for some features like surround
> sound, check if the feature if supported by the codec and only add
> items for that feature if it is
> - auvia(4) now supports multichannel playback with at least some
> multichannel ac97(4) codecs
> fixes problems for me and ajacoutot with different auvia(4)
> devices/codecs. no regressions reported by testers.
> if you have an ac97(4) audio device and are using mixerctl.conf(5),
> you may want/need to update it.
> ok ratchov (jakemsr@)
~ if_myx.c
> Set the baudrate with IF_Gbps(10); and remove an
> XXX comment now that if_baudrate is 64bits.
> ok reyk@ (thib@)
~ if_sk.c
> Add splnet around mii_tick() in sk_yukon_tick().
> ok kettenis@ dlg@ (brad@)
~ if_myxreg.h
> casts so this compiles on i386.
> ok reyk (thib@)
~ if_tht.c
> Set if_baudrate with IF_Gbps(10);
> ok dlg@ (thib@)
~ if_cas.c
> Remove some #if;ed out code. (kettenis@)
dev/usb
~ if_upgt.c ~ if_upgtvar.h
> Replace firmware version 2.5.8.0 with 2.13.1.0. Firmware API slightly
> changes.
> You need to update the firmware package to upgt-firmware-1.1.tgz.
> (mglocker@)
~ if_upgt.c ~ if_upgtvar.h
> Fix RX queue stalling by regulary reading the device statistics (with a
> firmware command). Doh! We can now reliable communicate in 11g.
> Took me about 3 weeks to find out (just had to say that). (mglocker@)
~ if_upgt.c ~ if_upgtvar.h
> Add LED support. (mglocker@)
~ if_upgt.c
> Make the device work again on big endian archs (after firmware upgrade).
> Tested on macppc. (mglocker@)
~ usbdevs.h ~ usbdevs_data.h
> regen (mglocker@)
~ if_upgt.c ~ usbdevs
> Add one more device which could work. Fix comment typo in usbdevs while
> there. (mglocker@)
~ if_upgt.c
> Adjust level 1 debug messages. (mglocker@)
~ if_upgt.c
> Read EEPROM content the other way around (from start to end). (mglocker@)
~ if_upgt.c ~ if_upgtvar.h
> Cleanup rates:
> - The device (firmware) does automatic rate control for 11b/g.
> - Allow user to set fixed rates.
> - Send management frames always at DS1. (mglocker@)
~ if_upgt.c
> Remove obsolete rateset variables. (mglocker@)
kern
~ uipc_mbuf.c
> Dont use PR_LIMITFAIL with PR_WAITOK when M_WAIT is passed with
> the flag argument to the m_*get* functions. Since PR_LIMITFAIL
> can cause us to return NULL if hit the limits.
> ok claudio@ (thib@)
~ kern_timeout.c
> Revert 1.25 and do not use _Q_INVALIDATE on timeout structures; being
> subtly different from CIRCLEQ, it is possible, when emptying the whole
> timeout chain, to end up with CIRCQ_EMPTY being false, and bad things
> happen. Back to the drawing board... (miod@)
msdosfs
~ msdosfs_denode.c
> Return pcbmap() errors from deget() rather than ignoring them. EIO at
> least has been seen from ian@'s new iPod, causing inappropriate
> mounting.
> ok miod@ (krw@)
net
~ if_ethersubr.c
> In ether_input() scheduling the soft interrupt and queueing the
> packet have to be protected by the same splnet. Otherwise on
> architectures where soft interrupts are executed immediately, the
> interrupt is processed before the packet is in the queue.
> On real hardware interfaces this was not a problem as ether_input()
> is called at splnet anyway. But when sending to a tun interface
> the packet got delayed.
> ok beck@ miod@ henning@ mpf@ markus@ (bluhm@)
~ netisr.h
> Fix add_net_randomness() not being called because the wrong
> bit in netisr is set. 1 != (1 << 1).
> Reported by mickey, fix by me.
> OK markus@, miod@, claudio@ (mpf@)
sys
~ disklabel.h
> Fix max disk size: whe have 48 bits for size in 512 bytes units.
> 1 bit allowance to avoid possible sign extend issues for now.
> This results in a max partition or disk size of 64PB. ok krw@ miod@
> (otto@)
ufs/ufs
~ ufs_vnops.c
> Notify us of NOTE_LINK on mkdir(2) as well.
> ok thib and miod. (fgsch@)
== usr.bin =========================================================== 08/09 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/usr.bin
cvs
~ commit.c ~ rcs.c
> Typos fixed.
> OK xsa@
> > Diff from Igor Zinovik. (tobias@)
~ commit.c
> Revision buffers are supposed to be of size CVS_REV_BUFSZ (32) instead
> of 24.
> OK xsa@
> > Diff from Igor Zinovik. (tobias@)
~ rcs.c
> Changed xmalloc(n * m) pattern into xcalloc(n, m).
> OK ray@, xsa@
> > Diff from Igor Zinovik. (tobias@)
mixerctl
~ mixerctl.c
> the mixer_devinfo structure contains next and prev pointers; they are used
> to create doubly linked "chains" of mixer devices. Currently mixerctl(1)
> supposes that the 'next' index is never smaller than the index of the
> device; so it fails to handle such mixers. The fix, allows 'next' indexes
> to
> be smaller than the index of the device.
> ok jakemsr@ (ratchov@)
ranlib
~ extern.h
> revert 1.4, a.out toolchain needs this not-so-unused-after-all function.
> ok miod@ (chl@)
ssh
~ ssh.1
> satisfy the pedants: -q does not suppress all diagnostic messages (e.g.
> some commandline parsing warnings go unconditionally to stdout). (djm@)
~ scp.1
> scp -q implies ssh -q for the underlying connection, it doesn't just
> hush the progress meter (djm@)
~ clientloop.c
> fd leak on session multiplexing error path. Report and patch from
> gregory_shively AT fanniemae.com (djm@)
~ ssh.c
> ignore SIGPIPE in multiplex client mode - we can receive this if the
> server runs out of fds on us midway. Report and patch from
> gregory_shively AT fanniemae.com (djm@)
~ sftp-client.c
> fix remote handle leak in do_download() local file open error path;
> report and fix from sworley AT chkno.net (djm@)
~ ssh-keygen.c
> when hashing individual hosts (ssh-keygen -Hf hostname), make sure we
> hash just the specified hostname and not the entire hostspec from the
> keyfile. It may be of the form "hostname,ipaddr", which would lead to
> a hash that never matches. report and fix from jp AT devnull.cz (djm@)
~ ssh-keygen.c
> unbreak line numbering (broken in revision 1.164), fix error message (djm@)
~ channels.c
> When we added support for specified bind addresses for port forwards, we
> added a quirk SSH_OLD_FORWARD_ADDR. There is a bug in our handling of
> this for -L port forwards that causes the client to listen on both v4
> and v6 addresses when connected to a server with this quirk, despite
> having set 0.0.0.0 as a bind_address.
> report and patch from Jan.Pechanec AT Sun.COM; ok dtucker@ (djm@)
~ readconf.c ~ readconf.h
~ sshconnect2.c
> promote rekeylimit to a int64 so it can hold the maximum useful limit
> of 2^32; report and patch from Jan.Pechanec AT Sun.COM, ok dtucker@ (djm@)
~ sftp.c
> When uploading, correctly handle the case of an unquoted filename with
> glob metacharacters that match a file exactly but not as a glob, e.g. a
> file called "[abcd]". report and test cases from duncan2nd AT gmx.de (djm@)
== usr.sbin ========================================================== 09/09 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/usr.sbin
dhcpd
~ dhcpd.h ~ options.c
> Call store_options() with correct cutoff values, i.e. ones not 7 bytes
> too large. Fixes a problem where a nicely crafted packet could crash
> dhcpd. Nuke a superfluous declaration of store_options() while here.
> Problem found and fix tested by Peter Hessler.
> ok beck@ millert@ henning@ (krw@)
snmpctl
~ parser.h ~ snmpctl.8
~ snmpctl.c
> bump copyright (reyk@)
~ parser.c ~ parser.h
~ snmpctl.8 ~ snmpctl.c
> add a snmpctl command to send traps to registered receivers using
> snmpd's imsg trap interface. (reyk@)
~ parser.c
> fix stringval length (reyk@)
~ parser.c
> parse ip addresses for traps (reyk@)
~ parser.c
> fix description (reyk@)
snmpd
~ control.c ~ snmp.h
~ snmpd.h ~ snmpe.c
> start working on an interface to send traps via snmpd.sock. userland
> applications will be able to send imsgs defining the trap to snmpd and
> the daemon will do the ASN.1/BER encoding before sending traps to the
> registered receivers. there are two advantages of this approach that
> a) the applications do not need to handle any ASN.1/BER encoding and
> b) snmpd will provide a central interface to define trap receivers.
> discussed with thib and others (reyk@)
~ Makefile ~ control.c
~ snmpd.h ~ snmpe.c
+ trap.c
> move trap code to an own file; there will be more (reyk@)
~ trap.c
> tweak (reyk@)
~ kroute.c ~ mib.c
~ mib.h ~ mps.c
~ parse.y ~ smi.c
~ snmp.h ~ snmpd.8
~ snmpd.c ~ snmpd.conf.5
~ snmpd.h ~ snmpe.c
> bump copyright (reyk@)
~ snmp.h
> define correct snmpv3 version (reyk@)
~ snmp.h
> typo (reyk@)
~ control.c ~ parse.y
~ snmpd.c ~ snmpd.h
~ snmpe.c ~ trap.c
> implementation of the SNMP trap sender interface (reyk@)
~ snmpd.conf.5
> document the "trap receiver" option. (reyk@)
~ snmpd.h ~ snmpe.c
~ trap.c
> send a coldStart trap to registered receivers on startup and cleanup
> the trap code a little bit. (reyk@)
~ parse.y ~ snmpd.conf.5
~ snmpd.h ~ trap.c
> allow to configure optional per trap receiver communities and to
> restrict the receivers to a specified mib. (reyk@)
~ trap.c
> the optional trap varbind elements may be null (reyk@)
~ trap.c
> swap arguments (reyk@)
~ imsg.c ~ snmpd.h
~ trap.c
> add a snmpctl command to send traps to registered receivers using
> snmpd's imsg trap interface. (reyk@)
~ trap.c
> trap.c (reyk@)
~ mib.c ~ mib.h
> implement very basic support of the BRIDGE-MIB which is enough to get
> recognized as a network device by some NMS. (reyk@)
~ mps.c
> free oid data before overwriting it (reyk@)
tcpdump
~ print-snmp.c ~ mib.h
> add basic snmpv2 support (snmpv2, trapv2, getbulk)
> ok canacar@ (reyk@)
usbdevs
~ usbdevs.c
> some usb devices stale (getstring: ioctl: Input/output error) if a
> string is requested with language id 0, so we get the language table
> first and use the language id from there
> ok mglocker@ (winiger@)
===============================================================================
More information about the owc
mailing list