[odc] Daily src changes for 2005-08-18
ODC
auto at squish.net
Fri Aug 19 08:00:38 BST 2005
OpenBSD src changes summary for 2005-08-18
==========================================
distrib/notes distrib/sets
lib/libc sbin/bioctl
share/man sys/arch/sparc/sparc
sys/arch/zaurus/dev sys/crypto
sys/dev sys/dev/cardbus
sys/dev/ic sys/dev/pci
sys/net sys/scsi
usr.bin/hexdump usr.bin/sort
== distrib =========================================================== 01/06 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/distrib
notes
~ i386/hardware ~ amd64/hardware
> nforce SATA is supported now. (jsg@)
~ amd64/hardware ~ i386/hardware
> sync with nforce portion of pciide(4). (jsg@)
sets
~ lists/man/mi
> sync (deraadt@)
== lib =============================================================== 02/06 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/lib
libc
~ sys/intro.2
> Update a few error descriptions w.r.t. obtaining the limits
> that cause the errors to reality.
> ok millert, jmc (jaredy@)
== sbin ============================================================== 03/06 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/sbin
bioctl
~ bioctl.8 ~ bioctl.c
> Add "create hot spare" ok dlg@ (marco@)
~ bioctl.8
> - mark up argument
> - missing full stop at end of sentence (jmc@)
~ bioctl.8 ~ bioctl.c
> very basic blink support.
> some manpage tweaks from jmc@ (dlg@)
~ bioctl.8 ~ bioctl.c
> a step towards where dlg is heading next + knf (deraadt@)
~ bioctl.8
> tweak SYNOPSIS; (jmc@)
~ bioctl.8
> if you show an example, BE CORRECT DAMNIT (deraadt@)
~ bioctl.c
> rework the addressing for blinking to make it like the hotspares
> help from and ok deraadt@ (dlg@)
~ bioctl.c
> better diagnostics in case of bad arguments; ok dlg (deraadt@)
~ bioctl.8
> sync with new functionality, and add a bit more meat (deraadt@)
~ bioctl.c
> add support to stop blinking (-u) or use the alarm LED instead (-B)
> (deraadt@)
== share ============================================================= 04/06 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/share
man
~ man7/environ.7
> note that current environment can be displayed using env(1) and
> printenv(1);
> this is based on a freebsd pr (#85008) from gary w. swearingen,
> and the following commit (-r1.24) from keramida at freebsd.mc@)
~ man4/ath.4
> Comment out AR5212 entries where it is unknown if they work.
> ok reyk@, jmc@ (jsg@)
~ man8/Makefile
> afterboot(8) MLINK to faq(8) is pointless;
> ok deraadt@ (jmc@)
~ man4/pciide.4
> nforce SATA is supported now. (jsg@)
~ man8/release.8
> rearrange some sections to avoid having instructions and explanations
> blocked together: it made the page difficult to read;
> ok mpf@ deraadt@ (jmc@)
~ man8/release.8
> no need to wrap (sub-)section headers in quotes; (jmc@)
~ man9/malloc.9
> document M_WAITOK as a separate flag rather than a note to M_NOWAIT
> help from and ok jmc@ ok deraadt@ no objections art@ (dlg@)
~ man8/release.8
> make sure dependencies are picked up;
> from mpf@, tweaked by millert@ (jmc@)
== sys =============================================================== 05/06 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/sys
arch/sparc/sparc
~ locore.s ~ machdep.c
~ vm_machdep.c
> Skip (trap) instruction in cpu_fork() instead of proc_trampoline().
> Add special handling for init(8) in setregs().
> Fixes returning from fork(2) in the child with a pending signal.
> ok deraadt@, art@ (screaming in agony) (kettenis@)
arch/zaurus/dev
~ zaurus_audio.c
> Make zaudio(4) work like a laptop, so that when you plug your headphones
> into the plug it changes the mute values so the speaker will be muted and
> the headphones will be unmuted. If you unplug the headphones, it will
> change
> everything back to the default state.
> help and ok from uwe@ pascoe@; tested by others (robert@)
crypto
~ cryptodev.c
> do not allow 0-length transforms; from freebsd (deraadt@)
dev
~ biovar.h
> Add "create hot spare" ok dlg@ (marco@)
~ biovar.h
> fix a pasto by marco@ (dlg@)
~ biovar.h
> blinking doesnt care about volid and diskid at the moment. use target
> instead (dlg@)
dev/cardbus
~ cardbus.c
> on debug, print the tuple number along with the name to help debugging in
> case of reserved tuples and pad to 0 the byte number. deraadt@ ok. (fgsch@)
dev/ic
~ ami.c
> Add "create hot spare" ok dlg@ (marco@)
dev/pci
~ pciide.c
> Several boards incorporating SIS SATA chips have dellusions of granduer
> so we have to override the PCI class they claim to be. ok deraadt@ (jsg@)
~ pciide.c
> Match nforce 2/3/4 SATA using generic sata_chip_map().
> Tested by dlg@ on nforce 3, and several people on
> nforce 4. ok deraadt@ (jsg@)
net
~ if_pfsync.c ~ pf.c
~ pf_if.c ~ pf_ioctl.c
~ pfvar.h
> Rearrange pf_state and pfi_kif so that the parts of the structure needed
> to search for a particular entry in the RB trees are at the start of the
> structure.
> This permits us to place a much smaller structure on the stack in the
> interrupt paths that match packets against state entries.
> ok mcbride (pascoe@)
~ pf_ioctl.c
> Malloc temporary buffers in pfioctl rather than having several large
> pf_state buffers on the stack.
> ok henning mcbride (pascoe@)
scsi
~ safte.c ~ safte.h
> das blinkenlighten! uses /dev/bio to allow an ioctl to turn identify on
> for a particular target (dlg@)
~ ses.c ~ ses.h
> infrastructure to support blinking controlled via /dev/bio ioctls. the
> code to do the actual blink is ifdefed out till marco can figure out what
> im doing wrong. (dlg@)
~ ses.h ~ ses.c
> Silly dlg did not read my old code careful enough. Missed 2 bits that are
> necessary for the blinkenlight magic.
> Also removed some XXX cruft that was misleading. (marco@)
~ sd.c
> Only allow blocksizes that are powers of 2 between 512 and 64K. An sd
> device that returns any other value via READ CAPACITY or MODE SENSE
> will be marked 'drive offline'. Prevents divide by zero when
> calculating disk size in MB, amoung other interesting possibilities.
> ok mickey@ dlg@ deraadt@ millert@ marco@ (krw@)
== usr.bin =========================================================== 06/06 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/usr.bin
hexdump
~ display.c
> Oops, previous fix would prevent standard input from being processed at all
> since it is not stat(2)'ed; spotted by fgs@, better fix by tom@ (miod@)
sort
~ files.c
> Properly initialize 'end' in seq(), fixing a crash
> on input with long lines when the -n flag is given.
> ok millert (jaredy@)
===============================================================================
More information about the odc
mailing list