[owc] Weekly src changes ending 2006-05-21
OWC
auto at squish.net
Mon May 22 08:02:21 BST 2006
OpenBSD src changes summary for 2006-05-14 to 2006-05-21 inclusive
==================================================================
bin/date bin/ksh
bin/rcp distrib/miniroot
distrib/notes distrib/sets
gnu/usr.bin/binutils lib/libc
lib/libcurses lib/libssl
libexec/ftpd libexec/ld.so
libexec/spamd regress/lib
regress/sbin sbin/dhclient
sbin/disklabel sbin/ifconfig
sbin/ipsecctl sbin/pfctl
sbin/route sbin/sysctl
share/man sys/arch/alpha/dev
sys/arch/amd64/amd64 sys/arch/amd64/conf
sys/arch/amd64/include sys/arch/amd64/stand/boot
sys/arch/amd64/stand/libsa sys/arch/amd64/stand/pxeboot
sys/arch/aviion/aviion sys/arch/aviion/conf
sys/arch/aviion/dev sys/arch/aviion/include
sys/arch/aviion/stand sys/arch/aviion/stand/a2coff
sys/arch/aviion/stand/libprom sys/arch/aviion/stand/libsa
sys/arch/aviion/stand/libz sys/arch/aviion/stand/netboot
sys/arch/hp300/hp300 sys/arch/hppa/gsc
sys/arch/i386/conf sys/arch/i386/i386
sys/arch/i386/include sys/arch/i386/pci
sys/arch/i386/stand/boot sys/arch/i386/stand/libsa
sys/arch/i386/stand/pxeboot sys/arch/luna88k/luna88k
sys/arch/m88k/include sys/arch/m88k/m88k
sys/arch/mac68k/mac68k sys/arch/mips64/include
sys/arch/mvme68k/mvme68k sys/arch/mvme88k/include
sys/arch/mvme88k/stand sys/arch/mvme88k/stand/bootsd
sys/arch/mvme88k/stand/bootst sys/arch/mvme88k/stand/bootxx
sys/arch/mvme88k/stand/bugcrt sys/arch/mvme88k/stand/installboot
sys/arch/mvme88k/stand/libbug sys/arch/mvme88k/stand/libsa
sys/arch/mvme88k/stand/netboot sys/arch/mvme88k/stand/prtvid
sys/arch/mvme88k/stand/sboot sys/arch/mvme88k/stand/tftpboot
sys/arch/mvme88k/stand/wrtvid sys/arch/powerpc/include
sys/arch/powerpc/powerpc sys/arch/sparc/fpu
sys/arch/sparc64/conf sys/arch/sparc64/dev
sys/arch/sparc64/fpu sys/arch/sparc64/include
sys/arch/sparc64/sparc64 sys/arch/sparc64/stand/installboot
sys/arch/zaurus/include sys/ddb
sys/dev sys/dev/acpi
sys/dev/i2c sys/dev/ic
sys/dev/microcode sys/dev/mii
sys/dev/pci sys/dev/sbus
sys/dev/usb sys/dev/wscons
sys/kern sys/net
sys/net80211 sys/netinet
sys/netipx sys/scsi
sys/sys sys/ufs/ffs
sys/uvm usr.bin/bc
usr.bin/calendar usr.bin/ftp
usr.bin/kdump usr.bin/rcs
usr.bin/ssh usr.bin/time
usr.bin/vi usr.sbin
usr.sbin/apm usr.sbin/apmd
usr.sbin/hostapd usr.sbin/httpd
usr.sbin/ntpd usr.sbin/openssl
usr.sbin/pppoe usr.sbin/quotaon
usr.sbin/spppcontrol usr.sbin/tcpdump
== bin =============================================================== 01/11 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/bin
date
~ date.1
> better example;
> from siju george
> ok deraadt (jmc@)
ksh
~ tests/th
> paramter -> parameter (miod@)
~ var.c
> Set the index of item[0], since it might not have been set before
> (A=0; A[1]=1) and do not use a static buffer in str_val, since the
> results might be used in a loop. Report from jared r r spiegel; help
> from Matthias Kilian; ok beck@ (otto@)
rcp
~ rcp.c
> fix leak (markus@)
== distrib =========================================================== 02/11 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/distrib
miniroot
~ install.sub
> Don't ask for a password for ftp transfers if the username 'ftp' is used.
> Suggested by deraadt@, 'looks good to me' deraadt@ (krw@)
notes
~ sparc64/prep
> point to the correct FTP site that can be used to get firwmare patches
> and also note the reader that the SunSolve web interface can be used
> for the same purpose. (robert@)
sets
~ lists/base/md.alpha ~ lists/base/md.amd64
~ lists/base/md.aviion ~ lists/base/md.cats
~ lists/base/md.hp300 ~ lists/base/md.hppa
~ lists/base/md.hppa64 ~ lists/base/md.i386
~ lists/base/md.luna88k ~ lists/base/md.mac68k
~ lists/base/md.macppc ~ lists/base/md.mvme68k
~ lists/base/md.mvme88k ~ lists/base/md.mvmeppc
~ lists/base/md.sgi ~ lists/base/md.sparc
~ lists/base/md.sparc64 ~ lists/base/md.vax
~ lists/base/md.zaurus ~ lists/comp/mi
~ lists/man/mi
> sync (deraadt@)
~ lists/base/md.amd64
> sync (deraadt@)
~ lists/base/md.alpha ~ lists/base/md.amd64
~ lists/base/md.aviion ~ lists/base/md.cats
~ lists/base/md.hp300 ~ lists/base/md.hppa
~ lists/base/md.hppa64 ~ lists/base/md.i386
~ lists/base/md.luna88k ~ lists/base/md.mac68k
~ lists/base/md.macppc ~ lists/base/md.mvme68k
~ lists/base/md.mvme88k ~ lists/base/md.mvmeppc
~ lists/base/md.sgi ~ lists/base/md.sparc
~ lists/base/md.sparc64 ~ lists/base/md.vax
~ lists/base/md.zaurus ~ lists/comp/mi
> sync (deraadt@)
~ lists/man/mi
> sync (deraadt@)
~ lists/base/mi
> sync (deraadt@)
~ lists/base/mi ~ lists/man/mi
> sync (deraadt@)
~ lists/man/mi
> sync (deraadt@)
== gnu =============================================================== 03/11 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/gnu
usr.bin/binutils
~ gdb/ppc-tdep.h ~ gdb/ppcobsd-nat.c
~ gdb/ppcobsd-tdep.c ~ gdb/ppcobsd-tdep.h
> Bring over upstream changes that fix backtraces through signal handlers and
> make us use PT_{GET|SET}FPREGS. (kettenis@)
~ include/opcode/i386.h
> add some more xcrypt & co opcodes
> ok deraadt (hshoexer@)
== lib =============================================================== 04/11 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/lib
libc
~ stdlib/malloc.c
> Fix the second malloc_ulimit regression: maintaining the free list
> requires memory; try to make sure we have it. If all fails, leak
> instead of crash. Test case originally found by cloder@, fix tested
> by many. (otto@)
~ gen/getcap.c
> Handle empty terminal names more gracefully.
> Variously OK. (hugh@)
~ sys/ptrace.2
> Adjust now that powerpc has PT_{GET|SET}FPREGS.
> reminded by miod@ (kettenis@)
~ gen/sysctl.3
> Document recent sysctl additions.
> ok jmc@, deraadt@ (gwk@)
~ stdlib/strtod.c
> fix a !foo & bar.
> ok miod@, tedu@, pedro@ (thib@)
libcurses
~ base/keybound.c ~ base/keyok.c
> lint warns about some fishy type conversion.
> Indeed, some ints can be silently shortened to unsigned short.
> Since those functions can return errors, do not convert spurious
> parameters to the wrong values, but return an error code instead.
> Tested for ill effects without any noticeable problems.
> okay millert@, otto@ (espie@)
libssl
~ src/apps/rand.c
> Add a -hex option to 'rand' to output hexadecimal output.
> ok djm@ miod@, man page help from jmc at .atthieu@)
== libexec =========================================================== 05/11 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/libexec
ftpd
~ monitor.c
TAGGED OPENBSD_3_9
> MFC:
> Fix by moritz@
> Allow anonymous logins without a password, which was broken
> by the last commit. Noticed by Armin Wolfermann.
> ok deraadt@ moritz@ (brad@)
ld.so
~ ldconfig/debug.c ~ ldconfig/prebind.c
> use __mips64__ (deraadt@)
~ ldconfig/prebind_delete.c
> prep prebind_newfile() for reuse by dale in prebind; ok drahn (deraadt@)
~ ldconfig/debug.c ~ ldconfig/ldconfig.c
~ ldconfig/prebind.c
> spacing (deraadt@)
spamd
~ spamd.8 ~ spamd.c
> add an -h option to override the hostname that is reported in the
> SMTP banner
> ok beck@ (jcs@)
== regress =========================================================== 06/11 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/regress
lib
~ libc/ieeefp/except/except.c
> Try harder to trigger an exception. On architectures where floating-point
> exceptions are deferred, we might not see the exception before another
> floating-point instruction is executed. This is especially true on the
> i387
> where we don't see underflow/overflow until the result is stored into
> memory. (kettenis@)
~ libc/malloc/malloc_ulimit1/malloc_ulimit1.c
~ libc/malloc/malloc_ulimit2/malloc_ulimit2.c
> Don't forget to return 0 on success, doh. (otto@)
sbin
~ ipsecctl/safail1.ok
> Update as a typo has been fixed there. Noticed by david@ (hshoexer@)
== sbin ============================================================== 07/11 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/sbin
dhclient
~ dhctoken.h
> whitespace cleanup
> ok claudio@, mbalmer@ (fkr@)
disklabel
~ disklabel.8
> change only use of slice to 'partition letter' to be consistent with
> use of partition everywhere else; ok jmc@ (stevesk@)
ifconfig
~ ifconfig.8 ~ ifconfig.c
> implement support for sppp(4) in ifconfig. have a look at the updated
> manual pages pppoe(4) and sppp(4) for examples.
> the spppcontrol(8) utility is not required anymore and will probably
> be removed in the near future.
> ok deraadt@ (reyk@)
~ ifconfig.8
> various fixes; (jmc@)
ipsecctl
~ ike.c
> delete weird C (deraadt@)
~ parse.y
> permit proto 0; ok hshoexer (deraadt@)
~ ipsec.conf.5
> paramter -> parameter (miod@)
pfctl
~ pfctl_altq.c
> better english to describe interfaces without bandwidth info; ok henning
> (deraadt@)
route
~ route.c
> Pretty print netmasks in monitor mode.
> From Alexander Bluhm <Alexander_Bluhm at genua.de>
> OK henning@ (mpf@)
sysctl
~ sysctl.8
> Document recent sysctl additions.
> ok jmc@, deraadt@ (gwk@)
== share ============================================================= 08/11 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/share
man
~ man5/pf.conf.5
> interface bandwidths can change; ok henning (deraadt@)
~ man4/Makefile + man4/wpi.4
> Initial import of wpi(4), a blob-free driver for Intel PRO/Wireless 3945ABG
> 802.11 adapters as found in Centrino Duo laptops.
> More to come but works well enough for me to commit this over a wpi(4).
> go for it deraadt@ (damien@)
~ man4/wpi.4
> wpi(4) adapters are Mini PCI Express. (damien@)
~ man4/wpi.4
> Fix date. (miod@)
~ man4/pci.4
> +wpi (miod@)
~ man4/pci.4
> add wpi(4)
> pointed out by deraadt@ (damien@)
~ man4/pci.4
> oops it got added twice (deraadt@)
~ man4/xge.4
> fix config syntax. (brad@)
~ man4/pppoe.4 ~ man4/sppp.4
> implement support for sppp(4) in ifconfig. have a look at the updated
> manual pages pppoe(4) and sppp(4) for examples.
> the spppcontrol(8) utility is not required anymore and will probably
> be removed in the near future.
> ok deraadt@ (reyk@)
~ man4/sppp.4
> various fixes; (jmc@)
~ man4/pppoe.4
> sort SEE ALSO; (jmc@)
~ man8/diskless.8
> use -enum lists instead of manually numbering them; (jmc@)
~ man8/diskless.8
> small cleanup; (jmc@)
~ man8/diskless.8
> remove a pile of pesky .Tn macros; (jmc@)
~ man8/diskless.8
> final -enum replacement with text adjusted accordingly; (jmc@)
~ man8/diskless.8
> make the phase 2 and 3 lists more consistent with each other; (jmc@)
~ man8/diskless.8
> section fixes; (jmc@)
~ man8/diskless.8
> otto's example of how to convert an IP address to hex; (jmc@)
~ man8/diskless.8
> make swap example work on diskless boots; (jmc@)
~ man8/diskless.8
> tidy up point 10; (jmc@)
~ man8/diskless.8
> document which server processes need to be running; (jmc@)
~ man8/diskless.8
> add some entries to FILES and SEE ALSO; (jmc@)
~ man8/diskless.8
> use miod's method for swapping; i had wanted to documented both ways,
> but it's just going to make the page overly complicated. this method
> has the advantage of making backups easier; (jmc@)
~ man8/diskless.8
> some wording/grammar fixes; (jmc@)
~ man8/diskless.8
> use miod's suggestion for making swap files; (jmc@)
~ man8/diskless.8
> back out last, after some discussion; (jmc@)
~ man8/diskless.8
> document that a smaller swap file may be used too, sometimes; (jmc@)
~ man8/diskless.8
> document diskless alpha boots, after some pointers from mickey; (jmc@)
~ man4/sppp.4 ~ man4/pppoe.4
> the physical interface must be marked UP
> ok naddy@ (reyk@)
~ man8/diskless.8
> cover diskless boots from more archs; some adjustments to come; (jmc@)
+ man4/mfi.4
> Add man page for mfi (marco@)
~ man4/Makefile
> Add mfi (marco@)
~ man8/diskless.8
> correct example awk expression; ok jmc@ (otto@)
~ man4/mfi.4
> extra .Ed (deraadt@)
~ man4/pci.4
> Xr mfi (deraadt@)
~ man4/mfi.4
> remove useless .Pp (jmc@)
~ man4/pci.4
> sort and add a full stop; (jmc@)
~ man4/pppoe.4
> tweak previous; ok reyk (jmc@)
~ man4/mfi.4
> Spello from <moorang at gmail dot com> (marco@)
~ man8/diskless.8
> point people to INSTALL.<arch> docs for details of net/diskless booting;
> (jmc@)
~ man8/diskless.8
> more stuff for alpha diskless; (jmc@)
~ man8/diskless.8
> hp 300 and hppa entries, gleaned from INSTALL; (jmc@)
~ man8/diskless.8
> mips64 bits; (jmc@)
~ man8/diskless.8
> make step 11 a little more readable; (jmc@)
~ man8/diskless.8
> kill some unneeded .Pp; (jmc@)
~ man8/diskless.8
> let's not discuss how to boot diskless for non-Ox machines; (jmc@)
~ man8/diskless.8
> fixes from miod:
> - MOPS -> MOP
> - .<arch> suffix is the exception, rather than the rule
> - better fstab entries (jmc@)
~ man8/afterboot.8
> - merge BOOTP and DHCP sections
> - point to diskless(8)
> based on a mail from alexey e. suslikov;
> ok henning mk (jmc@)
~ man8/diskless.8
> older alpha machines use mop; from miod (jmc@)
~ man8/diskless.8
> missing entry for hp 300; (jmc@)
~ man8/diskless.8
> mickey confirms some hppa64 bits... (jmc@)
~ man8/diskless.8
> tweak; (jmc@)
~ man4/trunk.4
> fix the documentation: failover mode only accepts incoming traffic on the
> active port and drops everything else. this has been changed to support
> failover mode on stupid devices like simple network hubs. (reyk@)
~ man4/ipw.4 ~ man4/iwi.4
~ man4/wpi.4
> - monitor mode for wpi(4)
> - document more DIAGNOSTICS messages for wpi(4)
> - the firmware is loaded when the interface is brought up, not during
> attachment of the device (ipw, iwi, wpi) (damien@)
~ man4/ddb.4
> document sh all poo (mickey@)
~ man4/wpi.4
> tweak; (jmc@)
~ man4/ddb.4
> tweak; (jmc@)
~ man8/diskless.8
> diskless bits for AMD64 and i386; (jmc@)
== sys =============================================================== 09/11 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/sys
arch/alpha/dev
~ sgmap_typedep.c
> Implement dmamap_load_uio for SGMAPs.
> From NetBSD (brad@)
~ sgmap_typedep.c
> Pay attention to BUS_DMA_READ; don't need to allocate a spill
> page if it is set.
> From NetBSD (brad@)
~ bus_dma.c
> Fix a couple of comments.
> From NetBSD (brad@)
~ bus_dma.c
> - _bus_dmamap_load_buffer_direct_common -> _bus_dmamap_load_buffer_direct
> - fix _bus_dmamap_load_(uio/mbuf)_direct panic messages.
> - s/vm_page_alloc_memory/uvm_pglistalloc/ in panic message.
> From NetBSD (brad@)
~ bus_dma.c
> Check for stale flags in the DMA map.
> From thorpej NetBSD (brad@)
arch/amd64/amd64
~ bios.c
> Smbios cleanup
> Improve the heuristics a little and extend them to the version and serial
> number fields. Clean the strings of leading/trailing space and deal with
> strings which consist entirely of space. Use fixed buffers for strings
> instead of using pointers into the table.
> ok deraadt@, and tested by many. (gwk@)
~ machdep.c
> handle BOOTARG_BOOTMAC from boot block; tested by reyk (deraadt@)
~ autoconf.c
> Detect BOOTARG_BOOTMAC coming from PXE; search for interface and attempt
> to boot from it. At the same time add -a support for network interface
> selection; tested by reyk, mk, deraadt on i386, cloned code applied to
> amd64 and tested a little bit less (deraadt@)
arch/amd64/conf
~ GENERIC
> xge works as well here (ie. as well as i386, which is not yet perfect,
> but it will get fixed) (deraadt@)
~ GENERIC
> clsc works; tleslie at gmail.comeraadt@)
~ GENERIC
> enable azalia(4) so we can get some more feedback. (brad@)
~ GENERIC
> Just finished testing mfi on amd64 so add it here too. (marco@)
~ GENERIC
> disable azalia here too. (brad@)
~ RAMDISK ~ RAMDISK_CD
> since mfi is pretty small, it fits on some media (deraadt@)
~ RAMDISK_CD
> NFSCLIENT can obviously fit on here (deraadt@)
arch/amd64/include
~ smbiosvar.h
> Smbios cleanup
> Improve the heuristics a little and extend them to the version and serial
> number fields. Clean the strings of leading/trailing space and deal with
> strings which consist entirely of space. Use fixed buffers for strings
> instead of using pointers into the table.
> ok deraadt@, and tested by many. (gwk@)
~ biosvar.h
> add bios_bootmac_t and BOOTARG_BOOTMAC (deraadt@)
arch/amd64/stand/boot
~ boot.8
> better wording for the "memory" section; (jmc@)
arch/amd64/stand/libsa
~ pxe.c
> push a BOOTARG_BOOTMAC (what PXE says is the MAC address of the interface
> we are booting from) out towards the kernel; tested by reyk
> crank the version too (deraadt@)
arch/amd64/stand/pxeboot
~ conf.c
> push a BOOTARG_BOOTMAC (what PXE says is the MAC address of the interface
> we are booting from) out towards the kernel; tested by reyk
> crank the version too (deraadt@)
~ pxeboot.8
> some diskless pointers; ok deraadt (jmc@)
arch/aviion/aviion
~ machdep.c
> Print a final line before asking the PROM to halt or reboot, for the
> abruptness of its prompt almost looks like the kernel died horribly.
> (miod@)
~ m8820x.c
> Recognize braindead 6:1 CMMU to CPU ratio configurations (which we do not
> support yet), as found e.g. on some 530 models. (miod@)
~ av400_machdep.c ~ locore.S
~ prom.c
> Save the PROM VBR on startup, rather than assuming the PROM (and thus its
> VBR) will always reside at the same address.
> Also, when querying the PROM for the memory size, ask it for the ``user''
> size - i.e. with the PROM data area taken out. (miod@)
~ m8820x.c
> Try and use the CPUCONFIG PROM call even if we did not detect 88204 CMMUs,
> recent PROM might have it implemented, and for 6:1 processor boards on 530
> models, this is the only way we'll get to know the split scheme in use, for
> now. (miod@)
~ av400_machdep.c ~ machdep.c
> Introduce a ``struct board'' platform abstraction, to be able to support
> more AViiON designs in the future.
> Heuristics to tell designs apart in aviion_identify() are likely too weak,
> but this is a start and AV400 still gets identified as AV400 (-: (miod@)
~ locore.S ~ machdep.c
> Let the bootloader pass esym to the kernel, and let the kernel make good
> use
> of this if it finds out it has not been directly loaded by the PROM; not
> tested yet. (miod@)
~ av400_machdep.c ~ eh.S
~ m8820x.c ~ machdep.c
~ pmap_table.c
> Move common definitions shared by the various 88100-based designs to their
> own header file, and slowly make most of the code board-independent. No
> functional change. (miod@)
arch/aviion/conf
~ AV400
> This file was not supposed to be imported initially, but it might as well
> be a GENERIC-without-other-models support template (hence, for now,
> identical). (miod@)
~ GENERIC ~ files.aviion
> Hints and commented out lines about future work. (miod@)
arch/aviion/dev
~ if_le_syscon.c ~ mainbus.c
~ nvram.c ~ nvramreg.h
~ syscon.c ~ vme.c
~ vmevar.h
> Move common definitions shared by the various 88100-based designs to their
> own header file, and slowly make most of the code board-independent. No
> functional change. (miod@)
arch/aviion/include
~ prom.h
> Save the PROM VBR on startup, rather than assuming the PROM (and thus its
> VBR) will always reside at the same address.
> Also, when querying the PROM for the memory size, ask it for the ``user''
> size - i.e. with the PROM data area taken out. (miod@)
~ m8820x.h
> Try and use the CPUCONFIG PROM call even if we did not detect 88204 CMMUs,
> recent PROM might have it implemented, and for 6:1 processor boards on 530
> models, this is the only way we'll get to know the split scheme in use, for
> now. (miod@)
~ cpu.h ~ param.h
~ trap.h + board.h
> Introduce a ``struct board'' platform abstraction, to be able to support
> more AViiON designs in the future.
> Heuristics to tell designs apart in aviion_identify() are likely too weak,
> but this is a start and AV400 still gets identified as AV400 (-: (miod@)
~ av400.h ~ board.h
+ avcommon.h
> Move common definitions shared by the various 88100-based designs to their
> own header file, and slowly make most of the code board-independent. No
> functional change. (miod@)
~ avcommon.h
> Fix AV_IEN() computation for cpu #2 and #3. (miod@)
arch/aviion/stand
~ Makefile + Makefile.inc
> Network bootloader for AV400 series, will load a native kernel. Still some
> rough edges to sort out, but enough to successfully boot a kernel.
> Will not work on series 530 yet. (miod@)
~ Makefile.inc
> Lower load address 128KB, to cope with the PROM data area on 8MB machines.
> (miod@)
arch/aviion/stand/a2coff
~ a2coff.c
> Deal with arbitrary load addresses. Necessary for upcoming netboot code.
> (miod@)
arch/aviion/stand/libprom
+ Makefile + Makefile.inc
+ getc.c + getenaddr.c
+ halt.c + prom.h
+ putc.c
> Network bootloader for AV400 series, will load a native kernel. Still some
> rough edges to sort out, but enough to successfully boot a kernel.
> Will not work on series 530 yet. (miod@)
arch/aviion/stand/libsa
+ Makefile + Makefile.inc
+ clock.c + exec_aout.c
+ libsa.h + nvramreg.h
+ parse_args.c
> Network bootloader for AV400 series, will load a native kernel. Still some
> rough edges to sort out, but enough to successfully boot a kernel.
> Will not work on series 530 yet. (miod@)
~ exec_aout.c
> Let the bootloader pass esym to the kernel, and let the kernel make good
> use
> of this if it finds out it has not been directly loaded by the PROM; not
> tested yet. (miod@)
arch/aviion/stand/libz
+ Makefile + Makefile.inc
> Network bootloader for AV400 series, will load a native kernel. Still some
> rough edges to sort out, but enough to successfully boot a kernel.
> Will not work on series 530 yet. (miod@)
arch/aviion/stand/netboot
+ Makefile + boot.c
+ conf.c + crt.c
+ dev_net.c + dev_net.h
+ devopen.c + if_le.c
+ if_lereg.h + version.c
> Network bootloader for AV400 series, will load a native kernel. Still some
> rough edges to sort out, but enough to successfully boot a kernel.
> Will not work on series 530 yet. (miod@)
~ crt.c
> Lower load address 128KB, to cope with the PROM data area on 8MB machines.
> (miod@)
arch/hp300/hp300
~ locore.s ~ machdep.c
~ pmap_bootstrap.c
> Get rid of ``maxmem'' and fix the descriptive comment for ``physmem''.
> Either maxmem is not used (mac68k), or the position of the physical memory
> is set in stone so we don't need to know the top of the memory (MAXADDR
> on hp300, physmem on mvme68k) it was pointing to.
> Plus this gets rid of unused lowram on mvme68k - all the m68k world is not
> an
> hp300, after all. (miod@)
~ locore.s
> Do not wait until Lstart2 to reset the HP MMU command register on models
> which need it. This allows up to free a1 for general use much earlier.
> (miod@)
arch/hppa/gsc
~ harmony.c
> do not even match on unsupported hw (instead of return from attach); from
> 710 (mickey@)
arch/i386/conf
~ GENERIC
> add wpi(4) so i can get feedback.
> ok deraadt@ (damien@)
~ GENERIC
> xge might as well be enabled (deraadt@)
~ GENERIC
> enable azalia(4) so we can get some more feedback. (brad@)
~ GENERIC
> Enable mfi. (marco@)
~ GENERIC
> no, azalia may not be on. Why? You did not ASK ME, brad. And there
> is an outstanding PR which makes it obvious that it should not be
> enabled yet! (deraadt@)
~ RAMDISKB ~ RAMDISK_CD
> since mfi is pretty small, it fits on some media (deraadt@)
~ GENERIC
> Add acpitz, commented out (canacar@)
~ RAMDISK_CD
> NFSCLIENT can obviously fit on here (deraadt@)
arch/i386/i386
~ mpbios.c
> Bring over workaround for broken BIOSes that use ACPI global interrupt
> numbers
> instead of APIC pin numbers from amd64.
> ok brad@ (kettenis@)
~ machdep.c
> Enable detection of SpeedStep support for the VIA Esther CPU's. Note
> this won't identify the possible speeds yet. Add some KNF while there.
> ok deraadt (dim@)
~ bios.c
> Smbios cleanup
> Improve the heuristics a little and extend them to the version and serial
> number fields. Clean the strings of leading/trailing space and deal with
> strings which consist entirely of space. Use fixed buffers for strings
> instead of using pointers into the table.
> ok deraadt@, and tested by many. (gwk@)
~ est.c ~ machdep.c
> Reimplement the Enhanced SpeedStep CPU detection by using the
> quintuplet { vendor, MHz_hi, mV_hi, MHz_lo, mV_lo }, which works much
> better than parsing the branding string. It also makes it easier to add
> new models later.
> Shouldn't break any CPU's which previously worked okay with EST; if
> there's any trouble, please let me know!
> Also adds quite a number of VIA C7-M models, and a few more Pentium M
> models, all from FreeBSD.
> "go for it" deraadt@ (dim@)
~ pmap.c ~ pmapae.c
> clean out some NetBSD cruft from the pmap code.
> ok mickey@ deraadt@ (brad@)
~ autoconf.c
> delete old #ifdef INSTALL chunk (deraadt@)
~ est.c
> Some more EST improvements:
> - Add Pentium M 710, 730, 740, 750, 760; preliminary, not all table
> values are known yet, thanks to incomplete Intel documentation.
> - Change Pentium M 770 to 133 MHz system bus clock.
> - Fix rounding, so 133 MHz bus clock based CPU speeds like 1733, 1866,
> 2133 show up correcly, instead of 1729, 1862, 2128 as before.
> If you have any of the 7[1-7]0 models, and it doesn't get detected
> properly yet, please let me know. It might be you have a slightly
> different variant from what is in the tables.
> Tested by a lot of people; ok brad@, slap it in deraadt@ (dim@)
~ bios.c
> handle BOOTARG_BOOTMAC from boot block; tested by reyk (deraadt@)
~ autoconf.c
> Detect BOOTARG_BOOTMAC coming from PXE; search for interface and attempt
> to boot from it. At the same time add -a support for network interface
> selection; tested by reyk, mk, deraadt on i386, cloned code applied to
> amd64 and tested a little bit less (deraadt@)
arch/i386/include
~ smbiosvar.h
> Smbios cleanup
> Improve the heuristics a little and extend them to the version and serial
> number fields. Clean the strings of leading/trailing space and deal with
> strings which consist entirely of space. Use fixed buffers for strings
> instead of using pointers into the table.
> ok deraadt@, and tested by many. (gwk@)
~ cpu.h
> Reimplement the Enhanced SpeedStep CPU detection by using the
> quintuplet { vendor, MHz_hi, mV_hi, MHz_lo, mV_lo }, which works much
> better than parsing the branding string. It also makes it easier to add
> new models later.
> Shouldn't break any CPU's which previously worked okay with EST; if
> there's any trouble, please let me know!
> Also adds quite a number of VIA C7-M models, and a few more Pentium M
> models, all from FreeBSD.
> "go for it" deraadt@ (dim@)
~ biosvar.h
> add bios_bootmac_t and BOOTARG_BOOTMAC (deraadt@)
arch/i386/pci
~ agp_machdep.c
> Add AGP support for Intel 945G/GM chipsets. The 945GM (mobile) variant
> also needs a X.org update, which will go in later.
> Tested by a bunch of people; ok deraadt@ (dim@)
arch/i386/stand/boot
~ boot.8
> better wording for the "memory" section; (jmc@)
arch/i386/stand/libsa
~ pxe.c
> push a BOOTARG_BOOTMAC (what PXE says is the MAC address of the interface
> we are booting from) out towards the kernel; tested by reyk
> crank the version too (deraadt@)
arch/i386/stand/pxeboot
~ conf.c
> push a BOOTARG_BOOTMAC (what PXE says is the MAC address of the interface
> we are booting from) out towards the kernel; tested by reyk
> crank the version too (deraadt@)
~ pxeboot.8
> some diskless pointers; ok deraadt (jmc@)
arch/luna88k/luna88k
~ locore.S ~ m8820x.c
~ machdep.c
> Clean secondary processors initialization code, similar to mvme88k;
> tested aoyama@ (miod@)
arch/m88k/include
~ vmparam.h
> INTSTACK_SIZE is not used anymore. (miod@)
~ mmu.h
> Make sure pmap copes with address wraparounds when operating on ranges.
> This removes the dependency for VM_MAX_ADDRESS to be aligned on a segment
> boundary.
> While there, remove a few internal pmap macros from mmu.h. (miod@)
~ vmparam.h
> Crank VM_MAX_ADDRESS to 4GB minus one page now that pmap will accept it.
> (miod@)
~ mmu.h
> kvtopte() is not used anymore. (miod@)
arch/m88k/m88k
~ eh_common.S
> Missing semicolon, found by deraadt (miod@)
~ vectors_88110.S
> Wrong routine for reserved vector 0x0f. (miod@)
~ pmap.c
> Directly pass a pmap to flush_atc_entry() instead of individual fields of
> it.
> No functional change. (miod@)
~ pmap.c
> Make sure pmap copes with address wraparounds when operating on ranges.
> This removes the dependency for VM_MAX_ADDRESS to be aligned on a segment
> boundary.
> While there, remove a few internal pmap macros from mmu.h. (miod@)
arch/mac68k/mac68k
~ machdep.c
> Get rid of ``maxmem'' and fix the descriptive comment for ``physmem''.
> Either maxmem is not used (mac68k), or the position of the physical memory
> is set in stone so we don't need to know the top of the memory (MAXADDR
> on hp300, physmem on mvme68k) it was pointing to.
> Plus this gets rid of unused lowram on mvme68k - all the m68k world is not
> an
> hp300, after all. (miod@)
arch/mips64/include
~ stdarg.h
> use __mips64__ (robert@)
arch/mvme68k/mvme68k
~ locore.s ~ machdep.c
~ mem.c ~ pmap_bootstrap.c
> Get rid of ``maxmem'' and fix the descriptive comment for ``physmem''.
> Either maxmem is not used (mac68k), or the position of the physical memory
> is set in stone so we don't need to know the top of the memory (MAXADDR
> on hp300, physmem on mvme68k) it was pointing to.
> Plus this gets rid of unused lowram on mvme68k - all the m68k world is not
> an
> hp300, after all. (miod@)
arch/mvme88k/include
~ prom.h
> Cleanup of mvme88k standalone code, with rewritten startup code with
> fewer assembly statements and much less magic, a few display artefacts
> removed, and private definitions moved out of the main prom.h. (miod@)
- vid.h
> Move otherwise unused <machine/vid.h> to stand/ (miod@)
arch/mvme88k/stand
~ Makefile.inc
> Make standalone binaries compile with -Wall -Werror (miod@)
arch/mvme88k/stand/bootsd
~ boot.c
> Make standalone binaries compile with -Wall -Werror (miod@)
~ version.c
> Crank bootblocks version for today's changes. (miod@)
~ Makefile
> Nuke whitespace at EOL (miod@)
arch/mvme88k/stand/bootst
~ boot.c ~ dev_tape.c
~ rawfs.c
> Make standalone binaries compile with -Wall -Werror (miod@)
~ version.c
> Crank bootblocks version for today's changes. (miod@)
~ Makefile
> Nuke whitespace at EOL (miod@)
arch/mvme88k/stand/bootxx
~ Makefile
> Cleanup of mvme88k standalone code, with rewritten startup code with
> fewer assembly statements and much less magic, a few display artefacts
> removed, and private definitions moved out of the main prom.h. (miod@)
~ bootxx.c
> Make standalone binaries compile with -Wall -Werror (miod@)
~ version.c
> Crank bootblocks version for today's changes. (miod@)
arch/mvme88k/stand/bugcrt
~ Makefile.inc ~ crt.c
> Cleanup of mvme88k standalone code, with rewritten startup code with
> fewer assembly statements and much less magic, a few display artefacts
> removed, and private definitions moved out of the main prom.h. (miod@)
arch/mvme88k/stand/installboot
~ installboot.c
> Make standalone binaries compile with -Wall -Werror (miod@)
arch/mvme88k/stand/libbug
~ libbug.h ~ putchar.c
~ return.c
> Cleanup of mvme88k standalone code, with rewritten startup code with
> fewer assembly statements and much less magic, a few display artefacts
> removed, and private definitions moved out of the main prom.h. (miod@)
~ libbug.h
> Delay board type retrieval. (miod@)
arch/mvme88k/stand/libsa
- SRT0.S - SRT1.c
~ Makefile ~ Makefile.inc
~ exec_mvme.c
> Cleanup of mvme88k standalone code, with rewritten startup code with
> fewer assembly statements and much less magic, a few display artefacts
> removed, and private definitions moved out of the main prom.h. (miod@)
~ bugdev.c ~ parse_args.c
> Make standalone binaries compile with -Wall -Werror (miod@)
~ clock.c ~ libsa.h
> Nuke whitespace at EOL (miod@)
~ exec_mvme.c
> Delay board type retrieval. (miod@)
arch/mvme88k/stand/netboot
~ Makefile ~ boot.c
> Cleanup of mvme88k standalone code, with rewritten startup code with
> fewer assembly statements and much less magic, a few display artefacts
> removed, and private definitions moved out of the main prom.h. (miod@)
~ dev_net.c ~ devopen.c
~ if_ie.c
> Make standalone binaries compile with -Wall -Werror (miod@)
~ version.c
> Crank bootblocks version for today's changes. (miod@)
~ if_le.c
> Nuke whitespace at EOL (miod@)
arch/mvme88k/stand/prtvid
~ Makefile ~ prtvid.c
> Nuke whitespace at EOL (miod@)
- chklabel.c ~ prtvid.c
> Use standard swap16() and swap32() functions instead of abusing <db.h> to
> get
> similar M_xx_SWAP(). (miod@)
+ vid.h
> Move otherwise unused <machine/vid.h> to stand/ (miod@)
arch/mvme88k/stand/sboot
~ TODO ~ XBUG.S
~ XSRT0.S ~ clock.c
~ etherfun.h ~ if_le.c
~ if_lereg.h ~ oc_cksum.S
~ sboot.c ~ sboot.h
~ srec.c
> Nuke whitespace at EOL (miod@)
arch/mvme88k/stand/tftpboot
~ Makefile
> Cleanup of mvme88k standalone code, with rewritten startup code with
> fewer assembly statements and much less magic, a few display artefacts
> removed, and private definitions moved out of the main prom.h. (miod@)
~ netdev.c ~ tftpfs.c
> Make standalone binaries compile with -Wall -Werror (miod@)
~ version.c
> Crank bootblocks version for today's changes. (miod@)
arch/mvme88k/stand/wrtvid
~ Makefile ~ disklabel.h
> Nuke whitespace at EOL (miod@)
~ wrtvid.c
> Use standard swap16() and swap32() functions instead of abusing <db.h> to
> get
> similar M_xx_SWAP(). (miod@)
arch/powerpc/include
~ ptrace.h
> Implement PT_{GET|SET}FPREGS.
> "looks good to me" miod@, ok drahn@ (kettenis@)
~ reg.h
> Add struct fpreg. (kettenis@)
arch/powerpc/powerpc
~ process_machdep.c
> Implement PT_{GET|SET}FPREGS.
> "looks good to me" miod@, ok drahn@ (kettenis@)
arch/sparc/fpu
~ fpu.c
> Make the array of fpu exception types/codes big enough.
> Fix a typo too, such that we actually read beyond element 0 in all cases.
> ok miod@ (kettenis@)
arch/sparc64/conf
~ files.sparc64
> Merge creator attachment code again, it turns out the upa/mainbus
> attachment
> code split is not necessary after all; ok jason@ (miod@)
arch/sparc64/dev
- creator_mainbus.c - creator_upa.c
~ creator.c ~ creatorvar.h
> Merge creator attachment code again, it turns out the upa/mainbus
> attachment
> code split is not necessary after all; ok jason@ (miod@)
arch/sparc64/fpu
~ fpu.c
> Make the array of fpu exception types/codes big enough.
> Fix a typo too, such that we actually read beyond element 0 in all cases.
> ok miod@ (kettenis@)
arch/sparc64/include
~ psl.h
> Put explicit barriers in in-line spl functions. (miod@)
arch/sparc64/sparc64
~ trap.c
> Fix SIGFPE handling. The code doing an ADVANCE after calling trapsignal(),
> which made is skip the first instruction of the signal trampoline in
> certain
> cases. That in turn truly hosed the stack.
> ok miod@, jason@ (kettenis@)
arch/sparc64/stand/installboot
~ installboot.8
> - /usr/mdec is usually not in path, invoke installboot with full path
> - /ofwboot does not get updated by installboot as one might think from
> the example, it has to be copied in place by hand
> - Xr. boot(8)
> ok jmc@ (martin@)
arch/zaurus/include
~ vmparam.h
> No need to define two page freelists since we'll only populate one; tested
> by robert@ (miod@)
ddb
~ db_interface.h
> fix comments (mickey@)
~ db_command.c ~ db_interface.h
> add show all pools command listing all pools as vmstat -m does; miod@ ok
> (mickey@)
dev
~ ipmivar.h ~ ipmi.c
> Make _bmc_io_wait 1 second instead of 500ms.
> Prevent multiple reads and writes to the bmc at the same time which could
> interrupt a complete bmc transaction.
> tested by various people. (marco@)
~ ipmivar.h
> Kill evil spacing. (marco@)
~ ipmi.c
> Add device name when printing out errors. Pointed out by deraadt@ (marco@)
~ ipmi.c ~ ipmivar.h
> Replace lockmgr with rwlock; help and ok marco@ (alek@)
dev/acpi
~ acpibat.c
> Use rwlock instead of lockmgr locks.
> ok marco@ (canacar@)
~ acpi.c ~ files.acpi
+ acpitz.c
> Minimal thermal zone driver. Monitors thermal zone temperature,
> shuts down the system if the 'critical temperature' is reached.
> ok marco@ (canacar@)
dev/i2c
~ w83l784r.c
> Add abstraction for resistor factors; makes it easier to compare them to
> the datasheets. Fix a few bugs in the handling of voltage sensors.
> From Constantine Murenin <mureninc at gmail.com(kettenis@)
dev/ic
~ mfi.c ~ mfireg.h
~ mfivar.h
> Add INQUIRY command.
> Add sgl magic. (marco@)
~ mfi.c
> Oops don't send command twice to firmware when polling. (marco@)
~ mfi.c
> Clear all relevant ccb fields when mfi_put_ccb frees a ccb. (marco@)
~ mfi.c
> Add TUR and make it send it directly to firmware.
> Sprinkle some extra debug. (marco@)
~ mfi.c
> Oops PREREAD not PREWRITE. (marco@)
~ mfi.c
> Add READ CAPACITY. (marco@)
~ re.c
> only allow setting Jumbo MTU frame sizes with the 8169 chipset. (brad@)
~ mfi.c
> Fix some htole32 I missed.
> Sprinkle more debug. (marco@)
~ mfi.c ~ mfivar.h
> Fix some more htole32 screw-ups.
> Colapse all little SCSI functions into a single generic one.
> Fix a bug in the SGE calculation.
> This brings the driver up to post:
> mfi0 at pci12 dev 14 function 0 "Dell PERC SAS" rev 0x00: Dell PERC 5/e irq
> 5
> scsibus3 at mfi0: 2 targets
> sd1 at scsibus3 targ 0 lun 0: <DELL, PERC 5/E Adapter, 1.00> SCSI5 0/direct
> fixed
> sd1: 237824MB, 237824 cyl, 64 head, 32 sec, 512 bytes/sec, 487063552 sec
> total
> sd2 at scsibus3 targ 1 lun 0: <DELL, PERC 5/E Adapter, 1.00> SCSI5 0/direct
> fixed
> sd2: 34176MB, 34176 cyl, 64 head, 32 sec, 512 bytes/sec, 69992448 sec total
> (marco@)
~ mfi.c
> Move !xs->data case into mfi_start_xs.
> Return proper sense data when a polling command fails. (marco@)
~ mfi.c
> Nuke some { } (marco@)
~ mfi.c ~ mfivar.h
> Refactor scsi path so that we can have a single sgl function for io and
> non-io
> traffic.
> Redo parameters on functions that were passing too many parameters around.
> (marco@)
~ mfi.c ~ mfivar.h
~ mfireg.h
> Add mfi_get_info to obtain all kinds of fun information from the fw.
> Print out whole info structure when debug is enabled.
> Add a nice print with useful info during attach. (marco@)
~ mfi.c ~ mfivar.h
~ mfireg.h
> Add completion routines.
> Fix some bugs inside the interrupt handler. (marco@)
~ mfi.c ~ mfivar.h
> And we have lift off! Was able to create a filesystem and write to the
> filesystem.
> Collapse both completion routines into one.
> Fixup scsi error reporting in the completion routine. (marco@)
~ mfi.c
> Kill splbio/splx in the interrupt handler. (marco@)
~ mfi.c
> Remove some debug cruft and pretty up some debug messages. (marco@)
~ re.c
> set if_jumbo_mtu and the IFCAP_JUMBO_MTU capabilities flag where
> appropriate.
> ok reyk@ (brad@)
~ ami.c
> The SCSI command on the passthrough bus is handed to the firmware via a
> structure not included in the MegaRAID command, so it needs to be synced
> before we submit it. Previously this was done only when the SCSI command
> had an associated data buffer. The syncing of the passthrough command is
> now done unconditionally.
> ok krw@ marco@ (dlg@)
~ ami.c
> remove a chunk of code that was ifdefed out due to compat issues with
> older firmwares. it was stagnating and now completely broken. if we need it
> back then we can poke around in cvs for it. (dlg@)
~ mfi.c
> Add first BIOCINQ primitive. (marco@)
~ mfi.c
> Oops, helps to use the right structure field.
> Use actual nr of physical disks present. (marco@)
~ mfireg.h
> Add alarm function opcodes. (marco@)
~ amivar.h ~ ami.c
> the maximum possible number of ccbs that a megaraid can use (126 of them)
> is preallocated in the softc, even if the driver is only ever going to use
> one of them when running. this allocates them separately during attach as a
> first step toward having the driver only allocate as many ccbs as the
> firmware can support.
> ok krw@ (dlg@)
~ ami.c
> during attach use a static, spartan ccb with ami_poll to query to firmare
> for its params. this means we can ask it how many ccbs it can deal with
> before we allocate them.
> ok krw@ (dlg@)
~ ami.c
> move variable declarations at the start of a block in ami_attach to the top
> of the function and unindent that block.
> ok krw@ (dlg@)
~ ami.c
> move the allocation of ccb out of ami_attach into a function of its own
> (aptly called ami_alloc_ccbs). allocate the ccbs after the firmware has
> been queried, but still for the full number of AMI_MAXCMDS.
> ok krw@ (dlg@)
~ mfi.c ~ mfivar.h
> Add generic management function.
> Redo ccb_state flags so that we can actually sleep on it.
> Redo mfi_get_info to use new mgmt function.
> Add rw_locks for mgmt interface.
> Make BIOCINQ dynamic. (marco@)
~ mfi.c
> Add bio alarm function.
> Redo mfi_mgmt to deal with zero data commands. (marco@)
~ mfi.c
> pretty up some debug statements. (marco@)
~ mfi.c
> Do a giant flush of disk and raid-card memory when SYNCHONIZE_CACHE is
> called. (marco@)
dev/microcode
~ cirruslogic/Makefile
> amd64 can handle this too; tleslie at gmail.comeraadt@)
dev/mii
~ rlphy.c
> consider re(4) as a possible 8139 chipset. (brad@)
~ brgphy.c
> also attach to Broadcom BCM5706C and BCM5708C PHY. (brad@)
dev/pci
~ files.pci + if_wpi.c
+ if_wpireg.h + if_wpivar.h
> Initial import of wpi(4), a blob-free driver for Intel PRO/Wireless 3945ABG
> 802.11 adapters as found in Centrino Duo laptops.
> More to come but works well enough for me to commit this over a wpi(4).
> go for it deraadt@ (damien@)
~ if_xge.c
> comment out link up/down printf's. (brad@)
~ if_xge.c
> re-enable link up/down messages because they spray a lot right now
> (clearly indicating some bug)
> copy mac address correctly (deraadt@)
~ if_xge.c
> account for interrupts in the interrupt handler. (brad@)
~ if_xge.c
> set the maximum MRU by default. (brad@)
~ if_xge.c
> set the multicast filter in xge_init(). (brad@)
~ if_xge.c
> only call init if not running. (brad@)
~ if_ixgb.h
> up the TX descriptors to the max. (brad@)
~ if_xge.c
> add a shutdown hook. (brad@)
~ agp_i810.c
> Add AGP support for Intel 945G/GM chipsets. The 945GM (mobile) variant
> also needs a X.org update, which will go in later.
> Tested by a bunch of people; ok deraadt@ (dim@)
~ if_re_pci.c ~ if_rl_pci.c
> attach 8139's capable of C+ mode to the re(4) driver, instead of the rl(4)
> driver. (brad@)
~ if_wpi.c
> Fix tsleep wait channel name; ok damien@ (miod@)
~ mfi_pci.c
> Labeling was the wrong way around. Make 5e 5i and vice versa. (marco@)
~ if_wpi.c ~ if_wpivar.h
> mega-commit of unrelated changes:
> - factor out contiguous DMA allocations (the same code was duplicated
> in several places and was sometimes incorrect)
> - s/EIO/ETIMEDOUT where it makes sense
> - s/memset/bzero/g
> - don't always bzero the ring descriptors
> - scan channels 1-14 instead of 1-13
> - remove the auhmode stuff since mgt frames are sent by net80211
> - remove the SCANALL flag since the f/w notifies us when it switches
> to another channel
> - don't re-enable interrupts on f/w panics
> - KASSERT that commands fit in a descriptor (damien@)
~ mfi_pci.c
> tweak the dmesg printing a bit.
> ok marco@ (brad@)
~ pcidevs
> remove 10GbE from the Xframe entries. (brad@)
~ pcidevs.h ~ pcidevs_data.h
> regen (brad@)
~ if_ipw.c ~ if_iwi.c
> sync handling of fatal firmware errors w/ wpi(4) (damien@)
~ if_em_hw.c
> Fix static array overrun.
> Coverity ID: 916
> From FreeBSD (brad@)
~ azalia_codec.c
> de-static (brad@)
~ azalia_codec.c
> disable the Analog Devices AD1981HD codec widget init routine for now.
> (brad@)
~ ahd_pci.c
> paramter -> parameter (miod@)
~ if_wpi.c
> use ieee80211_add_ssid() in ieee80211_beacon_alloc() and wpi(4) instead
> of duplicating the code.
> last commit was ok reyk@ (damien@)
~ if_wpi.c
> don't leak a mbuf with each scan command (damien@)
~ if_wpi.c
> simplify finalization of mbuf in wpi_rx_intr() (damien@)
~ if_wpi.c
> read channels calibration settings from EEPROM only once in wpi_attach()
> instead of doing it in wpi_init().
> remove a double definition of wpi_ioctl() while i'm here. (damien@)
~ if_vgevar.h
> garbage collect VGE_JUMBO_MTU (brad@)
~ if_em.c
> simplify MTU ioctl switch case. (brad@)
~ if_bge.c ~ if_em.c
~ if_ixgb.c ~ if_lge.c
~ if_nfe.c ~ if_nge.c
~ if_sk.c ~ if_ti.c
~ if_vge.c ~ if_xge.c
> set if_jumbo_mtu and the IFCAP_JUMBO_MTU capabilities flag where
> appropriate.
> ok reyk@ (brad@)
~ if_bge.c
> remove commented out and unused capabilities flags. (brad@)
~ if_em.c
> fix pci resource allocation in em, don't crash!
> fix by brad@ (reyk@)
~ if_wpi.c ~ if_wpireg.h
~ if_wpivar.h
> - fix short slot time and short preamble settings
> - fix basic rates setting
> - fix check for WEP encryption (authreq frames might require encryption
> in shared mode)
> - cleanup a few things while i'm here (damien@)
~ if_wpi.c ~ if_wpireg.h
> - fix CCK mask for 802.11b/only
> - add beacon command for future IBSS or HostAP mode support (damien@)
~ if_wpi.c ~ if_wpireg.h
> - fix monitor mode
> - cosmetic in wpi_dma_contig_alloc() (damien@)
~ if_wpi.c ~ if_wpivar.h
~ if_wpireg.h
> more rx radiotap stats (damien@)
~ if_wpi.c ~ if_wpivar.h
> more radiotap. (damien@)
~ ami_pci.c
> recognise the perc4/dc firmware (dlg@)
~ mfi_pci.c
> Add generic management function.
> Redo ccb_state flags so that we can actually sleep on it.
> Redo mfi_get_info to use new mgmt function.
> Add rw_locks for mgmt interface.
> Make BIOCINQ dynamic. (marco@)
dev/sbus
~ if_le_lebuffer.c ~ if_le_ledma.c
> Sprinkle bus_space_barrier()s after LANCE register writes, as already done
> in if_le.c; ok jason@ (miod@)
dev/usb
~ uhidev.c ~ usb_quirks.c
~ usb_quirks.h
> Add a quirk to force the Wi-spy spectrum analyser to attach as ugen.
> It claims to be a HID device, but it's not really one. ok miod@, jsg at . > (matthieu@)
~ if_aue.c ~ if_auereg.h
> some devices can do weird things upon a warm reboot so add a shutdown
> hook which explicitly resets and stops the device; fixes pr/5001
> ok dlg@ (jolan@)
dev/wscons
~ wsksymdef.h
> Add the missing Latin-2 glyphs, not used yet. (miod@)
kern
~ subr_autoconf.c
> When assigning a unit number, treat running off the end of the
> cd_devs[] array the same as finding a NULL entry.
> Fixes inconsistant assignment of unit numbers to media slots on the
> Neodio 8-in-1 USB card reader donated by Manuel Pata. Previously the
> unit numbers would vary depending on whether it had attached an even
> or odd number of times.
> 'reads ok' marco@ ok miod@ (krw@)
~ kern_ktrace.c
> change ktr_len to size_t. put ktr_type after comm for better alignment
> ok deraadt (tedu@)
~ subr_autoconf.c
> Detach children devices in reverse order. This gives better display, and
> allows cf_unit to be decremented correctly.
> 'put it in' deraadt@ "i think it's good' tedu@ (krw@)
~ vfs_getcwd.c
> Don't cast malloc() (pedro@)
~ subr_autoconf.c
> cf_unit should have a value between 0 and cd_ndevs. So set cf_unit to 0
> when
> cd_ndevs is set to 0.
> "No kidding!" deraadt@ "Seems to make sense to me." beck@ "Seems right."
> miod@ (krw@)
~ sys_process.c
> paramter -> parameter (miod@)
~ subr_pool.c
> add show all pools command listing all pools as vmstat -m does; miod@ ok
> (mickey@)
~ kern_sysctl.c
> It is possible for a sensor to be detached and freed while it is being
> copied out to userland. If this happens we get a use after free and things
> blow up. Prevent this by making a local copy of the sensor that cannot be
> freed elsewhere in the kernel and copying that out instead.
> found by mbalmer@
> ok deraadt@ krw@ (dlg@)
net
~ if_trunk.c
> the ifp->if_linkstatehooks may be NULL, add an extra check to avoid
> possible kernel panic. this happened to me by using tun(4) in layer 2
> mode (link0 flag) as a trunk port for testing. (reyk@)
~ pf.c
> missing rtlabel support in pf_addr_wrap_neq()
> PR5122 Jann Fischer <jfi at openbsd.dethanks! (henning@)
~ if.h ~ if_ethersubr.c
> add a if_jumbo_mtu field to the interface structure for drivers
> supporting ethernet jumbo frames. there's no standard for the size of
> jumbo MTUs, so either let the driver set it's own value or use 9000
> byte jumbo frames by default.
> ok brad@ (reyk@)
~ route.c
> don't panic if the address familiy is not found in the rt_tables
> ok henning@ (reyk@)
~ if_trunk.c
> drop packets received on inactive failover ports without increasing the
> error counter. just silently drop them...
> figured out by todd@, ok brad@ (reyk@)
~ if_trunk.c ~ if_trunk.h
> bump copyright (reyk@)
net80211
~ ieee80211_proto.h ~ ieee80211_output.c
> - export ieee80211_add_ssid() just like ieee80211_add_(x)rates() so I don't
> need to duplicate the code in wpi(4)
> - fix ieee80211_alloc_beacon()
> - clean two comments while i'm here (damien@)
~ ieee80211_output.c
> use ieee80211_add_ssid() in ieee80211_beacon_alloc() and wpi(4) instead
> of duplicating the code.
> last commit was ok reyk@ (damien@)
~ ieee80211_output.c
> account for the size of a 802.11 header in ieee80211_getmbuf().
> fix some typos while i'm here.
> ok reyk@ brad@ (damien@)
netinet
~ udp_usrreq.c
> - pcbflags is not modified in pcb_(dis)connect, so there's no
> need to save/restore it.
> - inp->inp_laddr should be set _before_ calling pcb_disconnect().
> otherwise pcb_disconnect() calls pcb_rehash() with the old laddr.
> [ currently this is not a problem since the hash ignores laddr ]
> tested by pedro (markus@)
~ ip_carp.c
> Add a duplicate check for our own advertisements. This is necessary
> for dumb non simplex interfaces, that receive packets they've just
> sent. Fixes bug 5121.
> OK deraadt@, mcbride@ (mpf@)
netipx
~ ipx_pcb.c
> paramter -> parameter (miod@)
scsi
~ sd.c
> Test the state of a disk *after* spinning it up and not before. This
> prevents the issuing of sizing inquiries to non-existant or non-responsive
> devices like empty slots on a USB card reader.
> Also try to avoid *extremely* unlikely loop of scsi_start() commands, by
> not using sd_interpret_sense() until after the disk is spinning.
> ok dlg@ (krw@)
~ ss.c
> paramter -> parameter (miod@)
~ safte.c
> replace lockmgr with rwlock. (dlg@)
sys
~ ktrace.h
> change ktr_len to size_t. put ktr_type after comm for better alignment
> ok deraadt (tedu@)
~ ioccom.h
> paramter -> parameter (miod@)
ufs/ffs
~ ffs_vfsops.c
> Don't mark a mount point read-only if we are about to sync it, otherwise
> stuff that relies on ffs_update() to flush an inode will fail.
> Fixes PR 4997, okay tedu@ (pedro@)
uvm
~ uvm_pdaemon.c
> tpyo (mickey@)
== usr.bin =========================================================== 10/11 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/usr.bin
bc
~ bc.y
> Avoid segv when a syntax error occurs while reading from command line.
> found by jmc@ (otto@)
calendar
~ calendars/fr_FR.ISO8859-1/calendar.fetes
~ calendars/fr_FR.ISO8859-1/calendar.jferies
> fix Mother's/Father's day and another entry.
> from FreeBSD;
> ok miod@ (aanriot@)
ftp
~ Makefile ~ fetch.c
~ ftp.1 ~ ftp.c
~ ftp_var.h ~ main.c
> https URL support; rototilled by a few people including me; originally
> from Rainer_Giedat at genua.deeraadt@)
~ cmds.c ~ domacro.c
~ fetch.c ~ ftp.c
~ ruserpass.c ~ util.c
> Remove shadowing variables and properly use /* FALLTHROUGH */
> comments. No binary change.
> Found by lint.
> OK beck@, deraadt@ (ray@)
~ cmds.c
> Remove ``sanity check'' that is already done in loop.
> OK beck@ (ray@)
~ ftp.1
> be more explicit about HTTPS; (jmc@)
kdump
~ kdump.c
> use lots more size_t instead of int running around
> ok deraadt (tedu@)
rcs
~ rcsmerge.c
> warnx(); return(); --> errx(); (xsa@)
~ Makefile ~ diff.h
~ diff3.c ~ merge.c
~ rcsprog.c ~ rcsprog.h
> missing bits for a working merge(1). Built and enabled.
> OK joris@ ray at .sa@)
~ Makefile
> oopps, missing \ at EOL. (xsa@)
~ rcsmerge.1
> .Xr merge(1) (xsa@)
~ ci.c ~ co.c
> fix printf's (xsa@)
~ date.y
> yyerror() little cleanup (xsa@)
~ rcsprog.c
> fix printf + add missing printf when unlocking file (-u). (xsa@)
ssh
~ clientloop.c
> missing free; from Kylene Hall (markus@)
~ scp.c ~ sftp.c
~ ssh-agent.c ~ ssh-keygen.c
~ sshconnect.c
> fix leak; coverity via Kylene Jo Hall (markus@)
~ kexdhc.c ~ kexgexc.c
> paramter -> parameter (miod@)
time
~ time.c
> remove useless case that is handled by default anyway
> zap trailing blanks
> ok otto@ (grunk@)
vi
~ vi/vs_relative.c
> Backport fix from nvi 1.81.5: do not go into loop if :set number and
> :set leftright and the cursor moves to an empty line; PR 3154; ok beck@
> (otto@)
== usr.sbin ========================================================== 11/11 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/usr.sbin
usr.sbin
~ Makefile
> do not build spppcontrol (deraadt@)
apm
~ apm.8
> same typo fix that ian just fixed in apmd.8; (jmc@)
apmd
~ apmd.8
> typo (ian@)
hostapd
~ print-802_11.c
> make lint happy (reyk@)
~ Makefile ~ apme.c
~ hostapd.8 ~ hostapd.conf.5
~ hostapd.h ~ iapp.c
~ parse.y ~ privsep.c
+ roaming.c
> initial implementation of "IP Roaming" in hostapd, see hostapd.conf(5).
> ok dlg@ (reyk@)
~ hostapd.8 ~ hostapd.conf.5
> mention the lack of authenticated management frames in IEEE 802.11. (reyk@)
~ hostapd.8
> .Sx for section references; (jmc@)
httpd
~ src/main/http_main.c
> set cur & max limits with MaxFOOPerChild instead of cur twice
> From: John Wong <johnw at wonghome.netok ckuethe (henning@)
ntpd
~ ntp.c
> PFD_MAX betterer than harcoded 1 (henning@)
openssl
~ openssl.1
> Add a -hex option to 'rand' to output hexadecimal output.
> ok djm@ miod@, man page help from jmc at .atthieu@)
~ openssl.1
> move an option to the right place; (jmc@)
pppoe
~ pppoe.8
> Reference pppoe(4).
> ``it makes sense'' jmc@ (ray@)
quotaon
~ quotaon.c
> add ()'s to improve readability.
> ok ray (dhill@)
spppcontrol
- Makefile - spppcontrol.8
- spppcontrol.c
> remove spppcontrol
> sppp(4) configuration is now done with ifconfig(8)
> ok canacar@ deraadt@ and others (reyk@)
tcpdump
~ tcpdump.8
> better description for -X;
> based on a diff from jean raby;
> ok deraadt (jmc@)
~ tcpdump.8
> add a little bit more to -X; (jmc@)
~ print-bootp.c
> Fix size checks for bootp packets. Problem found and initial patch
> from pedro at . moritz@ (canacar@)
===============================================================================
More information about the owc
mailing list