[owc] Weekly src changes ending 2007-09-16
OWC
auto at squish.net
Mon Sep 17 07:00:01 BST 2007
OpenBSD src changes summary for 2007-09-09 to 2007-09-16 inclusive
==================================================================
bin/chio bin/cp
bin/ln distrib/sets
games/atc games/battlestar
games/bcd games/fortune
games/hack games/monop
kerberosV lib/libc
lib/libssl lib/libutil
regress/lib sbin/dump
sbin/fsck_ffs sbin/ipsecctl
sbin/mount_vnd sbin/pfctl
sbin/raidctl sbin/restore
sbin/savecore share/man
sys/altq sys/arch/alpha/alpha
sys/arch/alpha/conf sys/arch/alpha/include
sys/arch/amd64/amd64 sys/arch/amd64/conf
sys/arch/amd64/include sys/arch/arm/include
sys/arch/armish/conf sys/arch/hppa/conf
sys/arch/hppa/dev sys/arch/hppa/hppa
sys/arch/hppa/include sys/arch/hppa64/hppa64
sys/arch/hppa64/include sys/arch/i386/conf
sys/arch/i386/i386 sys/arch/i386/include
sys/arch/i386/pci sys/arch/i386/stand/libsa
sys/arch/landisk/conf sys/arch/m68k/fpsp
sys/arch/m68k/include sys/arch/m88k/include
sys/arch/mac68k/dev sys/arch/mac68k/include
sys/arch/mac68k/mac68k sys/arch/macppc/conf
sys/arch/macppc/macppc sys/arch/mips64/include
sys/arch/powerpc/include sys/arch/powerpc/powerpc
sys/arch/sh/include sys/arch/sh/sh
sys/arch/solbourne/include sys/arch/sparc/include
sys/arch/sparc/sparc sys/arch/sparc64/conf
sys/arch/sparc64/fpu sys/arch/sparc64/include
sys/arch/sparc64/sparc64 sys/arch/vax/uba
sys/arch/vax/vax sys/arch/zaurus/conf
sys/compat/osf1 sys/conf
sys/crypto sys/ddb
sys/dev sys/dev/acpi
sys/dev/cardbus sys/dev/i2c
sys/dev/i2o sys/dev/ic
sys/dev/pci sys/dev/pcmcia
sys/dev/raidframe sys/dev/sbus
sys/dev/sdmmc sys/dev/usb
sys/dev/wscons sys/dev/wsfont
sys/kern sys/lib/libsa
sys/net sys/net80211
sys/netinet sys/netnatm
sys/nfs sys/scsi
sys/sys sys/ufs/ffs
sys/uvm usr.bin/bgplg
usr.bin/cdio usr.bin/cvs
usr.bin/diff usr.bin/diff3
usr.bin/file usr.bin/ftp
usr.bin/gprof usr.bin/gzsig
usr.bin/mail usr.bin/make
usr.bin/mg usr.bin/mkstr
usr.bin/msgs usr.bin/netstat
usr.bin/pcc usr.bin/rcs
usr.bin/rdist usr.bin/readlink
usr.bin/ssh usr.bin/sudo
usr.bin/sup usr.bin/systat
usr.bin/tail usr.bin/tftp
usr.bin/tip usr.bin/top
usr.bin/tset usr.bin/vacation
usr.bin/vi usr.bin/vmstat
usr.bin/xstr usr.sbin/bgpctl
usr.sbin/bgpd usr.sbin/config
usr.sbin/cron usr.sbin/dvmrpd
usr.sbin/eeprom usr.sbin/hostapd
usr.sbin/hoststated usr.sbin/httpd
usr.sbin/ifstated usr.sbin/inetd
usr.sbin/lpr usr.sbin/ntpd
usr.sbin/ospfctl usr.sbin/ospfd
usr.sbin/ppp usr.sbin/pppd
usr.sbin/ripctl usr.sbin/ripd
usr.sbin/sliplogin
== bin =============================================================== 01/11 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/bin
chio
~ parse.y
> extend lex to spot numbers in the stream, without impacting the parsing
> of ip addresses and such. this change is being pushed into all the pfctl
> derived parsers, starting with the easy ones. chio does not yet use
> the NUMBER token, but may one day. ok krw (deraadt@)
~ parse.y
> macro argument unused, using local instead; spotted by mpf (deraadt@)
cp
~ cp.1 ~ utils.c
> "source" and "target" can be either files or directories;
> cp(1) and ln(1) must have a synopsis consistent with mkdir(1),
> mv(1), rmdir(1) and others
> ok jmc@, millert@, otto@ (sobrado@)
ln
~ ln.1 ~ ln.c
> "source" and "target" can be either files or directories;
> cp(1) and ln(1) must have a synopsis consistent with mkdir(1),
> mv(1), rmdir(1) and others
> ok jmc@, millert@, otto@ (sobrado@)
== distrib =========================================================== 02/11 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/distrib
sets
~ lists/man/mi
> sync (deraadt@)
~ lists/base/md.hp300 ~ lists/base/md.mac68k
~ lists/base/md.mvme68k ~ lists/base/md.mvmeppc
~ lists/base/md.sparc
> sync (deraadt@)
== games ============================================================= 03/11 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/games
atc
~ main.c
> use strcspn to properly overwrite '\n' in fgets returned buffer
> ok pyr@, ray@, millert@, moritz@, chl@ (gilles@)
~ main.c
> remove some warnings:
> unused variable `variable'
> `variable' might be used uninitialized in this function
> ok gilles@ ray@ (chl@)
battlestar
~ cypher.c
> Move redundant prompt/fgets/sscanf lines into a function with
> appropriate checks, changing sscanf to strtonum in the process.
> Initial diff from Charles Longeau.
> OK moritz and chl. (ray@)
bcd
~ bcd.c
> use strcspn to properly overwrite '\n' in fgets returned buffer
> ok pyr@, ray@, millert@, moritz@, chl@ (gilles@)
fortune
~ fortune/fortune.c
> Proper use of fseek/fseeko macros.
> OK joris@, otto@ (tobias@)
hack
~ hack.pager.c ~ hack.rumors.c
> use strcspn to properly overwrite '\n' in fgets returned buffer
> ok pyr@, ray@, millert@, moritz@, chl@ (gilles@)
~ hack.pager.c
> remove some warnings:
> unused variable `variable'
> `variable' might be used uninitialized in this function
> ok gilles@ ray@ (chl@)
monop
~ monop.c
> use strcspn to properly overwrite '\n' in fgets returned buffer
> ok pyr@, ray@, millert@, moritz@, chl@ (gilles@)
== kerberosV ========================================================= 04/11 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/kerberosV
kerberosV
~ src/lib/kadm5/iprop.8
> - typo: ipropd->iprop
> ok millert@ (ajacoutot@)
== lib =============================================================== 05/11 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/lib
libc
~ sys/Makefile.inc + sys/Lint_environ.c
> Silence erroneous "environ used, but not defined" lint warning by
> providing a stub. ok deraadt@ (otto@)
~ sys/Makefile.inc + sys/Lint_longjmp.c
+ sys/Lint_setjmp.c + sys/Lint_siglongjmp.c
+ sys/Lint_sigsetjmp.c
> Add lint stubs for the longjmp family of functions which are defined in
> assembly on all platforms. ok deraadt@ (otto@)
~ rpc/xdr_stdio.c
> Proper use of fseek/fseeko macros.
> OK joris@, otto@ (tobias@)
~ rpc/clnt_perror.c
> Remove wrong snprintf(3) truncation checks, which can't be
> triggered and are off-by-one anyways. Instead replace them
> with correct checks and add all the missing cases.
> ok deraadt@ (moritz@)
~ db/man/dbopen.3
> for get the flags arg must be 0; from Tamas TEVESZ; ok millert@ (otto@)
libssl
~ src/crypto/bio/bss_file.c
> Proper use of fseek/fseeko macros.
> OK joris@, otto@ (tobias@)
libutil
~ login_fbtab.c
> comment inside comment (deraadt@)
== regress =========================================================== 06/11 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/regress
lib
~ libc/regex/main.c ~ libc/regex/split.c
> use strcspn to properly overwrite '\n' in fgets returned buffer
> ok moritz@ gilles@ (chl@)
~ libc/regex/main.c
> fix format strings, sizeof() does not return an int (otto@)
== sbin ============================================================== 07/11 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/sbin
dump
~ itime.c
> Proper use of fseek/fseeko macros.
> OK joris@, otto@ (tobias@)
fsck_ffs
~ dir.c
> Do not use DIP() as an lvalue; founc by pcc. ok millert@ ray@ (otto@)
ipsecctl
~ parse.y
> Here too: Add support to the lex for parsing number out of the stream.
> handle this in the parser. better range checks.
> with and ok deraadt@ (hshoexer@)
mount_vnd
~ mount_vnd.8
> rearrange the mount_vnd examples to show the equivalent entry for each
> vnconfig example, instead of having a bunch of different examples.
> discussed with and ok jmc@ grunk@ (aanriot@)
pfctl
~ parse.y
> Add support to the lex for parsing number out of the stream. handle
> this in the parser. because the new numbers are int64_t, many new
> range checks for < 0 are needed. re-check and improve all the
> existing rangechecks while at it. thanks for help by cloder and
> dhartmei (deraadt@)
~ parse.y
> add a missing range check for rtable ids; ok cloder henning (deraadt@)
raidctl
~ rf_configure.c
> use strcspn to properly overwrite '\n' in fgets returned buffer
> ok pyr@, ray@, millert@, moritz@, chl@ (gilles@)
~ rf_configure.c
> remove some warnings:
> unused variable `variable'
> `variable' might be used uninitialized in this function
> ok gilles@ ray@ (chl@)
restore
~ tape.c
> use strcspn to properly overwrite '\n' in fgets returned buffer
> use strtonum instead of atoi
> prompted by ray@
> ok moritz@ ray@ (chl@)
savecore
~ savecore.c
> use strcspn to properly overwrite '\n' in fgets returned buffer
> ok pyr@, ray@, millert@, moritz@, chl@ (gilles@)
~ savecore.c
> remove some warnings:
> unused variable `variable'
> `variable' might be used uninitialized in this function
> ok gilles@ ray@ (chl@)
== share ============================================================= 08/11 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/share
man
~ man4/Makefile + man4/thmc.4
> new driver for TI tmhc50 / adm1022 / adm1028; checked by kettenis
> (deraadt@)
~ man4/iic.4 ~ man4/thmc.4
> fix name, Xr (deraadt@)
~ man4/iic.4 ~ man4/thmc.4
> thmc vs tmhc screwup (deraadt@)
~ man9/uvm.9
> Introduce a md pmap hook, pmap_remove_holes(), which is supposed to mark
> the holes a MMU may have from a given vm_map. This will be automagically
> invoked for newly created vmspaces.
> On platforms with MMU holes (e.g. sun4, sun4c and vax), this prevents
> mmap(2) hints which would end up being in the hole to be accepted as valid,
> causing unexpected signals when the process tries to access the hole
> (since pmap can not fill the hole anyway).
> Unfortunately, the logic mmap() uses to pick a valid address for anonymous
> mappings needs work, as it will only try to find an address higher than the
> hint, which causes all mmap() with a hint in the hole to fail on vax. This
> will be improved later. (miod@)
~ man4/wpi.4
> point to new 2.14.1.15 firmware (-current only). (damien@)
~ man4/wpi.4
> fix firmware package version number. it's 2.14.1.5 not 2.14.1.15.
> and yes 2.14.1.5 is more recent than 2.14.4.
> don't ask me why, i'm just following Intel's numbering scheme.
> pointed out by jmc@ (damien@)
~ man4/audio.4
> add missing description of AUDIO_PERROR ioctl
> reviewed (and corrected) by jakemsr@ (ratchov@)
~ man4/bluetooth.4
> sevice -> service; from a list from jung (jmc@)
~ man4/man4.amd64/Makefile + man4/man4.amd64/pctr.4
> port of i386 pctr code to amd64; Mike Belopuhov (deraadt@)
~ man9/aml_evalnode.9
> Update manpage to reflect new return code from function.
> Thanks to mk@ for reminding me. (weingart@)
~ man4/nviic.4
> Add the NVIDIA MCP73 SMBus controller.
> ok dlg@ (brad@)
~ man4/piixpm.4
> mention the AMD SB700 and SB800 chipsets.
> ok dlg@ (brad@)
~ man4/nviic.4
> please avoid line wrap in doc source; (jmc@)
- man9/sleep.9 ~ man9/Makefile
~ man9/autoconf.9 ~ man9/ctxsw.9
~ man9/delay.9 ~ man9/mbuf.9
~ man9/spl.9 ~ man9/timeout.9
+ man9/tsleep.9
> sleep(9) was removed aeons ago according to miod, so remove references
> to it. Because man pages are named after functions (at least they
> should be) and sleep(9) doesn't exist anymore, sleep.9 is renamed to
> tsleep.9.
> Input and reminders from jmc and ratchov. (mk@)
~ man4/pciide.4
> mention the NVIDIA MCP73 and MCP77.
> ok dlg@ (brad@)
~ man4/azalia.4
> mention newer Intel ICH and NVIDIA MCP chipsets.
> ok deanna@ dlg@ (brad@)
~ man4/Makefile ~ man4/cardbus.4
~ man4/pci.4 + man4/bwi.4
> Add initial man page for bwi(4). Points to the firmware package.
> OK jsg@ (mglocker@)
~ man4/bwi.4
> Fix docdate CVS tag. (mglocker@)
~ man4/bwi.4
> Adapt firmware images prefix to driver name. Needs bump to firmware
> package version 1.1. (mglocker@)
== sys =============================================================== 09/11 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/sys
altq
~ altq_cbq.c ~ altq_cdnr.c
~ altq_hfsc.c ~ altq_priq.c
~ altq_red.c ~ altq_rio.c
~ altq_rmclass.c ~ altq_subr.c
> MALLOC/FREE -> malloc/free and M_ZERO changes
> ok henning@ krw@ canacar@ ray@ (chl@)
arch/alpha/alpha
~ machdep.c
> replace ctob and btoc with ptoa and atop respectively
> help and ok miod@ thib@ (martin@)
arch/alpha/conf
~ GENERIC
> uchcom(4) has been verified to work by djm@, so add it to all
> USB capable archs. (jsg@)
arch/alpha/include
~ pmap.h
> Introduce a md pmap hook, pmap_remove_holes(), which is supposed to mark
> the holes a MMU may have from a given vm_map. This will be automagically
> invoked for newly created vmspaces.
> On platforms with MMU holes (e.g. sun4, sun4c and vax), this prevents
> mmap(2) hints which would end up being in the hole to be accepted as valid,
> causing unexpected signals when the process tries to access the hole
> (since pmap can not fill the hole anyway).
> Unfortunately, the logic mmap() uses to pick a valid address for anonymous
> mappings needs work, as it will only try to find an address higher than the
> hint, which causes all mmap() with a hint in the hole to fail on vax. This
> will be improved later. (miod@)
~ param.h
> replace ctob and btoc with ptoa and atop respectively
> help and ok miod@ thib@ (martin@)
arch/amd64/amd64
~ rbus_machdep.c
> s/atoin/ation/ typos
> ok ray@ cnst@ moritz@ sobrado@ millert@ mbalmer@ (chl@)
~ conf.c ~ locore.S
+ pctr.c
> port of i386 pctr code to amd64; Mike Belopuhov (deraadt@)
arch/amd64/conf
~ GENERIC
> uchcom(4) has been verified to work by djm@, so add it to all
> USB capable archs. (jsg@)
~ GENERIC ~ files.amd64
> port of i386 pctr code to amd64; Mike Belopuhov (deraadt@)
arch/amd64/include
~ pmap.h
> Introduce a md pmap hook, pmap_remove_holes(), which is supposed to mark
> the holes a MMU may have from a given vm_map. This will be automagically
> invoked for newly created vmspaces.
> On platforms with MMU holes (e.g. sun4, sun4c and vax), this prevents
> mmap(2) hints which would end up being in the hole to be accepted as valid,
> causing unexpected signals when the process tries to access the hole
> (since pmap can not fill the hole anyway).
> Unfortunately, the logic mmap() uses to pick a valid address for anonymous
> mappings needs work, as it will only try to find an address higher than the
> hint, which causes all mmap() with a hint in the hole to fail on vax. This
> will be improved later. (miod@)
~ conf.h + pctr.h
> port of i386 pctr code to amd64; Mike Belopuhov (deraadt@)
arch/arm/include
~ pmap.h
> Introduce a md pmap hook, pmap_remove_holes(), which is supposed to mark
> the holes a MMU may have from a given vm_map. This will be automagically
> invoked for newly created vmspaces.
> On platforms with MMU holes (e.g. sun4, sun4c and vax), this prevents
> mmap(2) hints which would end up being in the hole to be accepted as valid,
> causing unexpected signals when the process tries to access the hole
> (since pmap can not fill the hole anyway).
> Unfortunately, the logic mmap() uses to pick a valid address for anonymous
> mappings needs work, as it will only try to find an address higher than the
> hint, which causes all mmap() with a hint in the hole to fail on vax. This
> will be improved later. (miod@)
arch/armish/conf
~ GENERIC
> uchcom(4) has been verified to work by djm@, so add it to all
> USB capable archs. (jsg@)
arch/hppa/conf
~ GENERIC
> uchcom(4) has been verified to work by djm@, so add it to all
> USB capable archs. (jsg@)
arch/hppa/dev
~ sti_sgc.c
> [fF]uther -> [fF]urther in comments and man page. First one spotted on
> tech@ by Jung. (krw@)
arch/hppa/hppa
~ trap.c
> realy -> really in comments. Most reported on tech@ by Jung. (krw@)
arch/hppa/include
~ pmap.h
> Introduce a md pmap hook, pmap_remove_holes(), which is supposed to mark
> the holes a MMU may have from a given vm_map. This will be automagically
> invoked for newly created vmspaces.
> On platforms with MMU holes (e.g. sun4, sun4c and vax), this prevents
> mmap(2) hints which would end up being in the hole to be accepted as valid,
> causing unexpected signals when the process tries to access the hole
> (since pmap can not fill the hole anyway).
> Unfortunately, the logic mmap() uses to pick a valid address for anonymous
> mappings needs work, as it will only try to find an address higher than the
> hint, which causes all mmap() with a hint in the hole to fail on vax. This
> will be improved later. (miod@)
arch/hppa64/hppa64
~ trap.c
> realy -> really in comments. Most reported on tech@ by Jung. (krw@)
arch/hppa64/include
~ pmap.h
> Introduce a md pmap hook, pmap_remove_holes(), which is supposed to mark
> the holes a MMU may have from a given vm_map. This will be automagically
> invoked for newly created vmspaces.
> On platforms with MMU holes (e.g. sun4, sun4c and vax), this prevents
> mmap(2) hints which would end up being in the hole to be accepted as valid,
> causing unexpected signals when the process tries to access the hole
> (since pmap can not fill the hole anyway).
> Unfortunately, the logic mmap() uses to pick a valid address for anonymous
> mappings needs work, as it will only try to find an address higher than the
> hint, which causes all mmap() with a hint in the hole to fail on vax. This
> will be improved later. (miod@)
arch/i386/conf
~ GENERIC
> uchcom(4) has been verified to work by djm@, so add it to all
> USB capable archs. (jsg@)
~ GENERIC
> tmhc has only been seen on i386 (deraadt@)
~ GENERIC
> thmc vs tmhc screwup (deraadt@)
arch/i386/i386
~ via.c
> henric, tsk tsk, you missed stuff related to cryptosoft (deraadt@)
~ rbus_machdep.c
> s/atoin/ation/ typos
> ok ray@ cnst@ moritz@ sobrado@ millert@ mbalmer@ (chl@)
arch/i386/include
~ pmap.h
> Introduce a md pmap hook, pmap_remove_holes(), which is supposed to mark
> the holes a MMU may have from a given vm_map. This will be automagically
> invoked for newly created vmspaces.
> On platforms with MMU holes (e.g. sun4, sun4c and vax), this prevents
> mmap(2) hints which would end up being in the hole to be accepted as valid,
> causing unexpected signals when the process tries to access the hole
> (since pmap can not fill the hole anyway).
> Unfortunately, the logic mmap() uses to pick a valid address for anonymous
> mappings needs work, as it will only try to find an address higher than the
> hint, which causes all mmap() with a hint in the hole to fail on vax. This
> will be improved later. (miod@)
arch/i386/pci
~ glxsb.c
> Remove unused "extern" declaration.
> ok tom@ (henric@)
arch/i386/stand/libsa
~ apmprobe.c
> Don't bother bitching if APM does not exist. Pleanty
> enough systems today do not have apm. Also, the existance
> of apm is given on the probe line, absence means it ain't
> there. (weingart@)
arch/landisk/conf
~ GENERIC
> uchcom(4) has been verified to work by djm@, so add it to all
> USB capable archs. (jsg@)
arch/m68k/fpsp
~ x_unfl.sa
> s/atoin/ation/ typos
> ok ray@ cnst@ moritz@ sobrado@ millert@ mbalmer@ (chl@)
arch/m68k/include
~ pmap_motorola.h
> Introduce a md pmap hook, pmap_remove_holes(), which is supposed to mark
> the holes a MMU may have from a given vm_map. This will be automagically
> invoked for newly created vmspaces.
> On platforms with MMU holes (e.g. sun4, sun4c and vax), this prevents
> mmap(2) hints which would end up being in the hole to be accepted as valid,
> causing unexpected signals when the process tries to access the hole
> (since pmap can not fill the hole anyway).
> Unfortunately, the logic mmap() uses to pick a valid address for anonymous
> mappings needs work, as it will only try to find an address higher than the
> hint, which causes all mmap() with a hint in the hole to fail on vax. This
> will be improved later. (miod@)
arch/m88k/include
~ pmap.h
> Introduce a md pmap hook, pmap_remove_holes(), which is supposed to mark
> the holes a MMU may have from a given vm_map. This will be automagically
> invoked for newly created vmspaces.
> On platforms with MMU holes (e.g. sun4, sun4c and vax), this prevents
> mmap(2) hints which would end up being in the hole to be accepted as valid,
> causing unexpected signals when the process tries to access the hole
> (since pmap can not fill the hole anyway).
> Unfortunately, the logic mmap() uses to pick a valid address for anonymous
> mappings needs work, as it will only try to find an address higher than the
> hint, which causes all mmap() with a hint in the hole to fail on vax. This
> will be improved later. (miod@)
arch/mac68k/dev
~ grf_mv.c ~ if_ae_nubus.c
~ if_sm_nubus.c ~ if_sn_nubus.c
~ if_sn_obio.c
> Pass a real ipl level to add_nubus_intr(), and make sure the interrupt
> handler is invoked with spl raised to the given level, otherwise all nubus
> interrupts are processed at level 2. (miod@)
arch/mac68k/include
~ viareg.h
> Pass a real ipl level to add_nubus_intr(), and make sure the interrupt
> handler is invoked with spl raised to the given level, otherwise all nubus
> interrupts are processed at level 2. (miod@)
arch/mac68k/mac68k
~ via.c
> Pass a real ipl level to add_nubus_intr(), and make sure the interrupt
> handler is invoked with spl raised to the given level, otherwise all nubus
> interrupts are processed at level 2. (miod@)
arch/macppc/conf
~ GENERIC
> uchcom(4) has been verified to work by djm@, so add it to all
> USB capable archs. (jsg@)
arch/macppc/macppc
~ autoconf.c
> /enet apparently maps to mc0; gwk ok (deraadt@)
arch/mips64/include
~ pmap.h
> Introduce a md pmap hook, pmap_remove_holes(), which is supposed to mark
> the holes a MMU may have from a given vm_map. This will be automagically
> invoked for newly created vmspaces.
> On platforms with MMU holes (e.g. sun4, sun4c and vax), this prevents
> mmap(2) hints which would end up being in the hole to be accepted as valid,
> causing unexpected signals when the process tries to access the hole
> (since pmap can not fill the hole anyway).
> Unfortunately, the logic mmap() uses to pick a valid address for anonymous
> mappings needs work, as it will only try to find an address higher than the
> hint, which causes all mmap() with a hint in the hole to fail on vax. This
> will be improved later. (miod@)
arch/powerpc/include
~ pmap.h
> Introduce a md pmap hook, pmap_remove_holes(), which is supposed to mark
> the holes a MMU may have from a given vm_map. This will be automagically
> invoked for newly created vmspaces.
> On platforms with MMU holes (e.g. sun4, sun4c and vax), this prevents
> mmap(2) hints which would end up being in the hole to be accepted as valid,
> causing unexpected signals when the process tries to access the hole
> (since pmap can not fill the hole anyway).
> Unfortunately, the logic mmap() uses to pick a valid address for anonymous
> mappings needs work, as it will only try to find an address higher than the
> hint, which causes all mmap() with a hint in the hole to fail on vax. This
> will be improved later. (miod@)
arch/powerpc/powerpc
~ process_machdep.c
> Make sure ptrace(2) cannot change any priviliged bits in the Machine State
> Register. Fixes PR 5574. From Stefan Kempf.
> ok deraadt@ (kettenis@)
~ pmap.c
> [fF]uther -> [fF]urther in comments and man page. First one spotted on
> tech@ by Jung. (krw@)
arch/sh/include
~ pmap.h
> Introduce a md pmap hook, pmap_remove_holes(), which is supposed to mark
> the holes a MMU may have from a given vm_map. This will be automagically
> invoked for newly created vmspaces.
> On platforms with MMU holes (e.g. sun4, sun4c and vax), this prevents
> mmap(2) hints which would end up being in the hole to be accepted as valid,
> causing unexpected signals when the process tries to access the hole
> (since pmap can not fill the hole anyway).
> Unfortunately, the logic mmap() uses to pick a valid address for anonymous
> mappings needs work, as it will only try to find an address higher than the
> hint, which causes all mmap() with a hint in the hole to fail on vax. This
> will be improved later. (miod@)
arch/sh/sh
~ locore_c.c
> Disable page zeroing in the idle loop, as it blasts the entire cache.
> Spotted by kettenis (miod@)
arch/solbourne/include
~ pmap.h
> Introduce a md pmap hook, pmap_remove_holes(), which is supposed to mark
> the holes a MMU may have from a given vm_map. This will be automagically
> invoked for newly created vmspaces.
> On platforms with MMU holes (e.g. sun4, sun4c and vax), this prevents
> mmap(2) hints which would end up being in the hole to be accepted as valid,
> causing unexpected signals when the process tries to access the hole
> (since pmap can not fill the hole anyway).
> Unfortunately, the logic mmap() uses to pick a valid address for anonymous
> mappings needs work, as it will only try to find an address higher than the
> hint, which causes all mmap() with a hint in the hole to fail on vax. This
> will be improved later. (miod@)
arch/sparc/include
~ pmap.h
> Introduce a md pmap hook, pmap_remove_holes(), which is supposed to mark
> the holes a MMU may have from a given vm_map. This will be automagically
> invoked for newly created vmspaces.
> On platforms with MMU holes (e.g. sun4, sun4c and vax), this prevents
> mmap(2) hints which would end up being in the hole to be accepted as valid,
> causing unexpected signals when the process tries to access the hole
> (since pmap can not fill the hole anyway).
> Unfortunately, the logic mmap() uses to pick a valid address for anonymous
> mappings needs work, as it will only try to find an address higher than the
> hint, which causes all mmap() with a hint in the hole to fail on vax. This
> will be improved later. (miod@)
arch/sparc/sparc
~ pmap.c
> Introduce a md pmap hook, pmap_remove_holes(), which is supposed to mark
> the holes a MMU may have from a given vm_map. This will be automagically
> invoked for newly created vmspaces.
> On platforms with MMU holes (e.g. sun4, sun4c and vax), this prevents
> mmap(2) hints which would end up being in the hole to be accepted as valid,
> causing unexpected signals when the process tries to access the hole
> (since pmap can not fill the hole anyway).
> Unfortunately, the logic mmap() uses to pick a valid address for anonymous
> mappings needs work, as it will only try to find an address higher than the
> hint, which causes all mmap() with a hint in the hole to fail on vax. This
> will be improved later. (miod@)
arch/sparc64/conf
~ GENERIC
> uchcom(4) has been verified to work by djm@, so add it to all
> USB capable archs. (jsg@)
~ files.sparc64
> IPI implementation. Bits and pieces from NetBSD, but the interface has
> been
> changed to free up another 64-bit word in the interrupt transaction. This
> means we have two 64-bit words available for arguments, which means we
> probably don't need to pass arguments through structures. (kettenis@)
arch/sparc64/fpu
~ fpu.c
> Make fpproc per-cpu. This turns fpproc into a macro, so adjust fpu/fpu.c
> to avoid collisions. (kettenis@)
arch/sparc64/include
~ cpu.h
> Make handled_intr_level per-cpu. (kettenis@)
~ cpu.h
> Make fpproc per-cpu. This turns fpproc into a macro, so adjust fpu/fpu.c
> to avoid collisions. (kettenis@)
~ cpu.h ~ param.h
> Add a few missing bits to include/cpu.h required for MULTIPROCESSOR
> kernels. (kettenis@)
~ pmap.h
> Introduce a md pmap hook, pmap_remove_holes(), which is supposed to mark
> the holes a MMU may have from a given vm_map. This will be automagically
> invoked for newly created vmspaces.
> On platforms with MMU holes (e.g. sun4, sun4c and vax), this prevents
> mmap(2) hints which would end up being in the hole to be accepted as valid,
> causing unexpected signals when the process tries to access the hole
> (since pmap can not fill the hole anyway).
> Unfortunately, the logic mmap() uses to pick a valid address for anonymous
> mappings needs work, as it will only try to find an address higher than the
> hint, which causes all mmap() with a hint in the hole to fail on vax. This
> will be improved later. (miod@)
arch/sparc64/sparc64
~ genassym.cf ~ intr.c
~ locore.s
> Make handled_intr_level per-cpu. (kettenis@)
- cpuvar.h ~ emul.c
~ genassym.cf
> Remove some more goo left over from sparc. (kettenis@)
~ cpu.c
> Make fpproc per-cpu. This turns fpproc into a macro, so adjust fpu/fpu.c
> to avoid collisions. (kettenis@)
~ genassym.cf ~ pmap.c
> Add a few missing bits to include/cpu.h required for MULTIPROCESSOR
> kernels. (kettenis@)
~ locore.s + ipifuncs.c
> IPI implementation. Bits and pieces from NetBSD, but the interface has
> been
> changed to free up another 64-bit word in the interrupt transaction. This
> means we have two 64-bit words available for arguments, which means we
> probably don't need to pass arguments through structures. (kettenis@)
~ ipifuncs.c
> Missing $OpenBSD$ tag. (kettenis@)
arch/vax/uba
~ qv.c
> [fF]uther -> [fF]urther in comments and man page. First one spotted on
> tech@ by Jung. (krw@)
arch/vax/vax
~ pmap.c
> Introduce a md pmap hook, pmap_remove_holes(), which is supposed to mark
> the holes a MMU may have from a given vm_map. This will be automagically
> invoked for newly created vmspaces.
> On platforms with MMU holes (e.g. sun4, sun4c and vax), this prevents
> mmap(2) hints which would end up being in the hole to be accepted as valid,
> causing unexpected signals when the process tries to access the hole
> (since pmap can not fill the hole anyway).
> Unfortunately, the logic mmap() uses to pick a valid address for anonymous
> mappings needs work, as it will only try to find an address higher than the
> hint, which causes all mmap() with a hint in the hole to fail on vax. This
> will be improved later. (miod@)
arch/zaurus/conf
~ GENERIC
> uchcom(4) has been verified to work by djm@, so add it to all
> USB capable archs. (jsg@)
compat/osf1
~ osf1_misc.c
> replace ctob and btoc with ptoa and atop respectively
> help and ok miod@ thib@ (martin@)
conf
~ files
> PCI and Cardbus hookup for the Broadcom AirForce devices.
> OK jsg@ (mglocker@)
crypto
~ cryptosoft.c ~ cryptosoft.h
~ sha1.c ~ sha1.h
~ xform.c ~ xform.h
> Make the hmac ipad/opad globals "const" and fixup the crypto functions
> to match.
> ok deraadt@ (henric@)
~ arc4.c ~ arc4.h
> Add __bounded__ attributes to prototypes.
> Use (x & 0xff) rather than (x % 256) - compilers generate faster code.
> Add arc4_getbytes function to extract raw keystream
> ok markus@ tom@ (djm@)
~ cryptodev.c ~ cryptosoft.c
~ deflate.c ~ xform.c
> Here too: Convert MALLOC/FREE to malloc/free and use M_ZERO where
> applicable.
> error spotting and ok krw@ (hshoexer@)
~ cryptosoft.c
> fix error introduced by my previous commit:
> "MALLOC(*swd, ...)" vs. "swd = malloc(..."
> ok millert (hshoexer@)
ddb
~ db_examine.c ~ db_print.c
~ db_write_cmd.c
> A few %n/%r/%z format strings were missed, convert these to db_format().
> (miod@)
dev
~ audio.c
> don't try to set the recording gain or balance through /dev/audio[ctl]
> if the selected recording port is not a port the limited /dev/audio[ctl]
> interface to mixer controls supports.
> ok ratchov@ (jakemsr@)
~ audio.c
> initialize sc_outports->master and sc_inports->master. Fixes "set failed"
> errors when running audioctl(1) on devices without master controls
> ok jakemsr@ (ratchov@)
~ cninit.c
> Make function pointer calls look all alike; from Holger Mikolon.
> No change in generated code. (miod@)
~ systrace.c
> more MALLOC/FREE -> malloc/free and M_ZERO changes
> ok pyr@ (gilles@)
~ audio.c
> fix uninitialized linked list head
> ok jakemsr@ (ratchov@)
~ ccd.c ~ ksyms.c
~ midisyn.c ~ ramdisk.c
~ softraid.c ~ systrace.c
~ vnd.c
> KNF
> prompted and "much better" by marco@, ok pyr@ (gilles@)
~ ccd.c ~ vnd.c
> There is no such word as `detatch'.
> ok mglocker (mk@)
dev/acpi
~ acpi.c ~ acpidock.c
~ dsdt.c ~ dsdt.h
> Implement ACPI 6.5.1 spec tree walk for _STA and _INI.
> Fixes mk's laptop. No regressions so far (thank you to the
> testers).
> ok gwk@, mk@, marco@ (weingart@)
~ acpi.c
> Remove unnecessary memset() (see aml_evalnode(9)) and a bit of KNF.
> ok toby (mk@)
dev/cardbus
~ files.cardbus + if_bwi_cardbus.c
> PCI and Cardbus hookup for the Broadcom AirForce devices.
> OK jsg@ (mglocker@)
~ if_bwi_cardbus.c
> Enable access to PCI and Cardbus config space from the driver. Lets us
> power on the device. (mglocker@)
~ if_bwi_cardbus.c
> No variable names in prototypes. (mglocker@)
~ if_bwi_cardbus.c
> Properly fill in some of the pci information the driver uses internally.
> (jsg@)
dev/i2c
~ files.i2c + tmhc50.c
> new driver for TI tmhc50 / adm1022 / adm1028; checked by kettenis
> (deraadt@)
- tmhc50.c ~ files.i2c
+ thmc50.c
> thmc vs tmhc screwup (deraadt@)
dev/i2o
~ iop.c ~ iopsp.c
> Obvious M_ZERO changes. (krw@)
dev/ic
~ i82596.c
> Fix setting up the multicast filter. Fixes PR 5474.
> "looks right" deraadt@ (kettenis@)
~ ar5xxx.c ~ mpi.c
> KNF
> prompted and "much better" by marco@, ok pyr@ (gilles@)
+ bwi.c + bwivar.h
+ bwireg.h
> Initial port of Sepherosa Ziehau's DragonFlyBSD driver
> for Broadcom AirForce wireless devices.
> Not yet functional, things like DMA and firmware handling
> haven't been converted yet.
> ok mglocker@ (jsg@)
~ mpi.c
> always tag fibre channel commands.
> ok marco@ (dlg@)
~ bwi.c
> destatic/deinline (jsg@)
~ bwi.c
> Tweak bwi_attach() a little. (jsg@)
~ bwi.c
> For now just return directly from the interrupt handler, or my amd64 will
> crash badly. (mglocker@)
~ bwi.c ~ bwireg.h
~ bwivar.h
> Spacing run. (mglocker@)
~ mpireg.h
> define the fc port config page 1. (dlg@)
~ bwi.c ~ bwivar.h
> Enable access to PCI and Cardbus config space from the driver. Lets us
> power on the device. (mglocker@)
~ bwi.c
> Forgot to set psc->psc_pcitag. Replace all PCI/Cardbus config space
> functions with our new one. Attachment seems to work now mainly.
> (mglocker@)
~ bwi.c
> Ops, remove temporary debug line. (mglocker@)
~ bwivar.h
> Fix sc_dev tweak. Device attaches with correct name now. (mglocker@)
~ bwi.c
> Add if_detach(). Fixes kernel crash when detaching cardbus device.
> (mglocker@)
~ bwi.c
> Move all debug lines to DPRINTF. KNF. Fix some tweaks while there.
> (mglocker@)
~ bwi.c
> Another cleanup run. (mglocker@)
~ bwi.c ~ bwivar.h
> Sort prototypes. (mglocker@)
~ bwi.c
> Use our standard ratesets from ieee80211 instead. (mglocker@)
~ bwi.c ~ bwivar.h
> Get firmware load working. Adapt firmware routines that way so we can
> use our one file firmware. (mglocker@)
~ bwi.c
> define BWI_DEBUG to make things easier (jsg@)
~ bwi.c ~ bwivar.h
> Properly fill in some of the pci information the driver uses internally.
> (jsg@)
~ bwi.c
> convert bus_dmamap_load() calls (jsg@)
~ bwi.c
> convert bus_dmamap_unload() calls
> remove bus_dma_tag_destroy() calls as we don't use/require them. (jsg@)
~ aic7xxx_openbsd.h
> replace ctob and btoc with ptoa and atop respectively
> help and ok miod@ thib@ (martin@)
~ bwi.c
> suitible -> suitable
> ok mglocker@ dlg@ (brad@)
~ bwi.c
> Convert bus_dma_load_mbuf() calls and remove unrequired callback
> code previously used. (jsg@)
~ bwi.c
> Convert bus_dmamap_sync() calls. (jsg@)
~ bwi.c ~ bwivar.h
> most of the remaining bus_dma conversion (jsg@)
~ bwi.c
> Give hardware physical addresses and add some missing
> bus_dmamap_create() calls. (jsg@)
~ bwi.c ~ bwivar.h
> Get IV load working. Simplified IV loading routine (mainly a rewrite).
> Adapted IV loading routines so we can use it with our one file firmware.
> (mglocker@)
~ bwi.c
> Adapt firmware images prefix to driver name. Needs bump to firmware
> package version 1.1. (mglocker@)
~ bwi.c
> Remove most of the __unused due to to code being enabled. (jsg@)
~ bwi.c
> Convert most of the remaining non firmware handling code. (jsg@)
~ bwi.c
> Free ucode if an error happens while FW / IV load. Always return a proper
> code in case of error. (mglocker@)
~ bwi.c
> CardBus works much better when sc_enable is called.
> Remove some safety returns while here.
> ok mglocker@ (jsg@)
~ bwi.c
> Move some noisy debug printfs to higher levels. (jsg@)
~ bwi.c
> Clear error before starting m_defrag() equivalent. (jsg@)
~ bwi.c
> Do sc_disable when device gets stopped.
> OK jsg@ (mglocker@)
~ bwi.c
> Better version of fake KKASSERT until they are changed properly. (jsg@)
~ bwi.c ~ bwivar.h
> Remove obsolete firmware specific bits. (mglocker@)
~ bwi.c
> Remove more obsolete firmware specific bits. (mglocker@)
dev/pci
~ qli_pci.c ~ qlireg.h
> Add firmware control block structure. (marco@)
~ azalia_codec.c
> Set bidirectional pins to output, ignoring what the codec says is
> their purpose, since that's often incorrect.
> This should solve some no-sound problems for codecs using the generic
> functions.
> From kent at netbsd (deanna@)
~ azalia.c
> Some pin widgets default to off, so make sure they're all enabled, not
> just the bidirectional ones.
> This could fix some no-sound problems on codecs using the generic
> functions.
> Code from kent at netbsd (deanna@)
~ azalia.h ~ azalia_codec.c
> Add command verbs, a mixer target and mixer controls for EAPD control.
> Some codecs, like the AD1984 found in x60s, need this in order for the
> speakers to work.
> New mixer items with names *.eapd will show up for codecs that need it
> and that are using the generic functions.
> From kent at netbsd (deanna@)
~ azalia.c ~ azalia.h
> If 44.1kHz sample rate isn't supported, use 48kHz instead. (deanna@)
~ if_iwnreg.h
> cleanup. no binary change. (damien@)
~ if_iwn.c
> only send a sensitivity update command when the settings have changed.
> (damien@)
~ if_iwn.c ~ if_iwnreg.h
> if more than 5 consecutive beacons are lost, reinitialize sensitivity
> state machine. (damien@)
~ if_wpi.c ~ if_wpireg.h
> - add support for 802.11 headers with various sizes (will be required
> for EDCA support).
> - remove an unused function.
> - fix scan command.
> - detect if the hardware radio switch is on instead of failing in the
> calibration of the thermal sensor with an obscure message. (damien@)
~ if_iwn.c ~ if_iwnreg.h
> fix interrupts processing and scan command + cleanup (damien@)
~ if_wpireg.h
> fix boot code maximum size. (damien@)
~ azalia.c
> Fix a copy/pasto from last year. CMI9880 should work now.
> ok brad (copy/pasto-er) (deanna@)
~ azalia.c ~ azalia_codec.c
> Use M_ZERO.
> ok dlg@ (deanna@)
~ azalia_codec.c
> For STAC9221 Macs, create master volume and mute controls by binding
> the DAC volumes.
> Tested by otto@, ajacoutot@, merdeley@ and Darrin Chandler. (deanna@)
~ azalia_codec.c
> For STAC9221 Macs - no need to max the DAC volumes now that a master
> is available. (deanna@)
~ pcidevs.h ~ pcidevs_data.h
> regen (martin@)
~ pcidevs
> add Innotek VirtualBox devices (martin@)
~ pcidevs.h ~ pcidevs_data.h
> sync (henning@)
~ pcidevs
> PLX 9016, found on a puc nick@ acquired somewhere (henning@)
~ if_wpi.c
> fix a bug in the scan command I introduced with my previous commit.
> (damien@)
~ if_wpi.c ~ if_wpireg.h
~ if_wpivar.h
> - add callbacks for setting EDCA parameters and pairwise crypto keys.
> - tweaks to make it look more like iwn(4). (damien@)
~ if_iwn.c
> s/iwn_edcaupdate/iwn_updateedca/ + hook it into net80211 (damien@)
~ if_nfe.c
> In nfe_start() do a fast return if IFF_OACTIVE is set, in
> this case we need a Tx interrupt to clean up the DMA ring
> before if_start can be properly called.
> Diff based on a comment by & ok claudio@ (jsg@)
~ files.pci + if_bwi_pci.c
> PCI and Cardbus hookup for the Broadcom AirForce devices.
> OK jsg@ (mglocker@)
~ if_san_front_end.h ~ if_san_xilinx.c
> s/atoin/ation/ typos
> ok ray@ cnst@ moritz@ sobrado@ millert@ mbalmer@ (chl@)
~ if_vr.c
> Remove the old software bit-banging MII interface, we started using
> the Rhine's shift registers four and a half years ago (1.31).
> From phk at FreeBSD
> ok dlg@
> Tested by brad@ and Jacob Yocom-Piatt. (brad@)
~ pcidevs
> - remove the bogus second AHCI PCI id for the AMD SB600 chipset.
> - add the AHCI PCI ids for the AMD SB700 chipset.
> From AMD via the Linux kernel
> ok dlg@ (brad@)
~ pcidevs
> - correct an unknown Marvell SATA entry to use the appropriate
> chipset name of 88SE6121
> - add the Marvell 88SX7042 SATA controller PCI id
> ok dlg@ (brad@)
~ pcidevs
> add the NVIDIA MCP73 SMBus controller PCI id.
> From Peer Chen @ NVIDIA
> ok dlg@ (brad@)
~ pcidevs.h ~ pcidevs_data.h
> regen (brad@)
~ if_bwi_pci.c
> Enable access to PCI and Cardbus config space from the driver. Lets us
> power on the device. (mglocker@)
~ if_bwi_pci.c
> Forgot to set psc->psc_pcitag. Replace all PCI/Cardbus config space
> functions with our new one. Attachment seems to work now mainly.
> (mglocker@)
~ nviic.c
> Add the NVIDIA MCP73 SMBus controller.
> ok dlg@ (brad@)
~ cmpci.c
> spacing nit.
> ok dlg@ (brad@)
~ pcidevs
> rename the AMD SB600 SMBus controller entry to X00 since this same
> PCI id is shared between the SB600, SB700 and SB800 chipsets.
> ok dlg@ (brad@)
~ pcidevs.h ~ pcidevs_data.h
> regen (brad@)
~ piixpm.c
> sync AMD SB600 PCI id with the new name.
> ok dlg@ (brad@)
~ pcidevs
> + Ralink Tech. RT2860 (damien@)
~ pcidevs.h ~ pcidevs_data.h
> sync (damien@)
~ if_bnxreg.h
> add two new chip revisions.
> From FreeBSD
> ok dlg@ (brad@)
~ pcidevs
> add the Marvell Yukon 2 88E8048 PCI id.
> From the Linux sky2 driver
> ok dlg@ (brad@)
~ pcidevs.h ~ pcidevs_data.h
> regen (brad@)
~ pciide.c
> add the NVIDIA MCP73/MCP77 IDE controller PCI ids.
> ok dlg@ (brad@)
~ cac_pci.c
> - correct error message if pci_mapreg_map() fails by not printing the
> device name in front.
> - remove printing of the name Compaq in front of the adapter model
> name. this is cac(4) it is pretty obvious this is a Compaq adapter.
> ok dlg@ (brad@)
~ if_bwi_pci.c
> Properly fill in some of the pci information the driver uses internally.
> (jsg@)
~ bktr/bktr_core.c
> replace ctob and btoc with ptoa and atop respectively
> help and ok miod@ thib@ (martin@)
dev/pcmcia
~ cfxga.c
> more M_ZERO conversions; ok krw at . (fgsch@)
~ cfxga.c ~ if_xe.c
~ pcmcia.c ~ pcmcia_cis.c
> KNF
> prompted and "much better" by marco@, ok pyr@ (gilles@)
dev/raidframe
~ rf_debugMem.h ~ rf_diskqueue.c
~ rf_engine.c ~ rf_openbsdkintf.c
> The obvious bzero/memset -> M_ZERO changes. (krw@)
dev/sbus
~ dma_sbus.c ~ isp_sbus.c
~ lebuffer.c ~ qec.c
> more MALLOC/FREE -> malloc/free and M_ZERO changes
> ok pyr@ (gilles@)
~ dma_sbus.c ~ isp_sbus.c
~ lebuffer.c ~ qec.c
> KNF
> prompted and "much better" by marco@, ok pyr@ (gilles@)
dev/sdmmc
~ sdhc.c ~ sdmmc.c
~ sdmmc_scsi.c
> more MALLOC/FREE -> malloc/free and M_ZERO changes
> ok pyr@ (gilles@)
~ sdmmc.c ~ sdmmc_scsi.c
> KNF
> prompted and "much better" by marco@, ok pyr@ (gilles@)
dev/usb
~ hid.c ~ uaudio.c
~ uhidev.c ~ usb_subr.c
~ usbf_subr.c
> more M_ZERO conversions; ok krw at . (fgsch@)
~ uchcom.c
> Properly prefix all functions with driver name. (jsg@)
~ usb_subr.c
> add string length arguments to some functions, and do not copy a string
> buffer to itself; checked by jsg and kettenis and others (deraadt@)
~ ucycom.c ~ uhci.c
> more M_ZERO usage. (fgsch@)
~ hid.c ~ uaudio.c
~ ucycom.c ~ uhci.c
~ uhidev.c ~ usb_subr.c
~ usbf_subr.c
> KNF
> prompted and "much better" by marco@, ok pyr@ (gilles@)
~ if_uath.c ~ if_uathvar.h
> undo rev 1.10 "implement a zero-copy RX data path".
> working with a fixed number of Rx buffers doesn't work well.
> fixes problems with blocked incoming traffic. (damien@)
~ if_cdce.c
> KNF. "go for it" deraadt@ (winiger@)
~ usbdevs
> new product (deraadt@)
~ usbdevs.h ~ usbdevs_data.h
> sync (deraadt@)
~ uplcom.c
> new product (deraadt@)
~ if_wi_usb.c
> There is no such word as `detatch'.
> ok mglocker (mk@)
~ uplcom.c
> adapt to new hx/non-hx type-selection algorithm from linux/dragonfly; ok
> jsg (deraadt@)
~ uplcom.c
> correct type printing (deraadt@)
~ umass.c
> Allow to pull out an usb stick with ffs filesystem while mounted
> and a file is written onto the stick. Without these fixes the
> machine panics or hangs.
> The usb fix calls the callback when the stick is pulled out to free
> the associated buffers. Otherwise we have busy buffers for ever
> and the automatic unmount will panic.
> The change in the scsi layer prevents passing down further dirty
> buffers to usb after the stick has been deactivated.
> In vfs the automatic unmount has moved from the function vgonel()
> to vop_generic_revoke(). Both are called when the sd device's vnode
> is removed. In vgonel() the VXLOCK is already held which can cause
> a deadlock. So call dounmount() earlier.
> ok krw@, I like this marco@, tested by ian@ (bluhm@)
~ uts.c
> some style(9). no binary change. (fgsch@)
dev/wscons
~ wsemul_vt100.c ~ wsevent.c
~ wskbd.c
> more MALLOC/FREE -> malloc/free and M_ZERO changes
> ok pyr@ (gilles@)
~ wsemul_vt100.c ~ wsevent.c
~ wskbd.c
> KNF
> prompted and "much better" by marco@, ok pyr@ (gilles@)
dev/wsfont
~ wsfont.c
> more MALLOC/FREE -> malloc/free and M_ZERO changes
> ok pyr@ (gilles@)
kern
~ init_main.c
> Introduce a md pmap hook, pmap_remove_holes(), which is supposed to mark
> the holes a MMU may have from a given vm_map. This will be automagically
> invoked for newly created vmspaces.
> On platforms with MMU holes (e.g. sun4, sun4c and vax), this prevents
> mmap(2) hints which would end up being in the hole to be accepted as valid,
> causing unexpected signals when the process tries to access the hole
> (since pmap can not fill the hole anyway).
> Unfortunately, the logic mmap() uses to pick a valid address for anonymous
> mappings needs work, as it will only try to find an address higher than the
> hint, which causes all mmap() with a hint in the hole to fail on vax. This
> will be improved later. (miod@)
~ kern_exec.c ~ kern_malloc.c
~ sysv_shm.c ~ vfs_bio.c
> replace ctob and btoc with ptoa and atop respectively
> help and ok miod@ thib@ (martin@)
~ vfs_default.c ~ vfs_subr.c
> Allow to pull out an usb stick with ffs filesystem while mounted
> and a file is written onto the stick. Without these fixes the
> machine panics or hangs.
> The usb fix calls the callback when the stick is pulled out to free
> the associated buffers. Otherwise we have busy buffers for ever
> and the automatic unmount will panic.
> The change in the scsi layer prevents passing down further dirty
> buffers to usb after the stick has been deactivated.
> In vfs the automatic unmount has moved from the function vgonel()
> to vop_generic_revoke(). Both are called when the sd device's vnode
> is removed. In vgonel() the VXLOCK is already held which can cause
> a deadlock. So call dounmount() earlier.
> ok krw@, I like this marco@, tested by ian@ (bluhm@)
lib/libsa
~ cons.c
> Make function pointer calls look all alike; from Holger Mikolon.
> No change in generated code. (miod@)
net
~ rtsock.c
> Only clear the message header if it is called from a sysctl walk. This
> unbreaks RTM_GET. Problem reported by fkr@ (claudio@)
~ pfkeyv2.c ~ pfkey.c
~ pfkeyv2_convert.c
> Convert MALLOC/FREE to malloc/free and use M_ZERO where applicable.
> ok krw@ (hshoexer@)
~ bpf.c ~ bridgestp.c
~ bsd-comp.c ~ if.c
~ if_bridge.c ~ if_ethersubr.c
~ if_faith.c ~ if_gif.c
~ if_gre.c ~ if_loop.c
~ if_pfsync.c ~ if_ppp.c
~ if_pppoe.c ~ if_sl.c
~ if_trunk.c ~ if_tun.c
~ if_vlan.c ~ pf_if.c
~ pf_ioctl.c ~ ppp-deflate.c
~ route.c ~ rtsock.c
> malloc sweep:
> -remove useless casts
> -MALLOC/FREE -> malloc/free
> -use M_ZERO where appropriate instead of seperate bzero
> feedback & ok krw, hshoexer (henning@)
~ if_pflog.c
> another M_ZERO diff. (pyr@)
net80211
~ ieee80211_crypto.c
> remove horrid casts now that SHA1Update() has been constified. (damien@)
~ ieee80211.c
> realy -> really in comments. Most reported on tech@ by Jung. (krw@)
netinet
~ ip_input.c
> Remove the ipq locking, it isn't strictly needed right now
> and is actually wrong in some cases, since we can enter
> functions without taking the lock because the return value
> of ipq_lock() isn't checked properly.
> However, this needs to be revisited when we start calling
> ip_drain() from the pool code when we are running out of
> memory, but this isn't done currently.
> OK art@, henning@ (thib@)
netnatm
~ natm.c
> There is no such word as `detatch'.
> ok mglocker (mk@)
nfs
~ nfsm_subs.h ~ nfs_serv.c
~ nfs_vfsops.c ~ nfs_vnops.c
> Replace nfsm_reqdone and nfsm_srvdone macros inline. No binary change.
> ok thib@ (blambert@)
scsi
~ sdvar.h ~ sd.c
> Allow to pull out an usb stick with ffs filesystem while mounted
> and a file is written onto the stick. Without these fixes the
> machine panics or hangs.
> The usb fix calls the callback when the stick is pulled out to free
> the associated buffers. Otherwise we have busy buffers for ever
> and the automatic unmount will panic.
> The change in the scsi layer prevents passing down further dirty
> buffers to usb after the stick has been deactivated.
> In vfs the automatic unmount has moved from the function vgonel()
> to vop_generic_revoke(). Both are called when the sd device's vnode
> is removed. In vgonel() the VXLOCK is already held which can cause
> a deadlock. So call dounmount() earlier.
> ok krw@, I like this marco@, tested by ian@ (bluhm@)
~ cd.c ~ ch.c
~ safte.c ~ scsi_base.c
~ scsi_ioctl.c ~ ses.c
~ ss.c ~ ss_mustek.c
~ ss_scanjet.c
> MALLOC/FREE -> malloc/free, M_ZERO, extraneous casts,
> extraneous #include <malloc.h> (krw@)
~ scsiconf.c ~ st.c
> A couple of obvious bzero() -> M_ZERO changes I missed. (krw@)
sys
~ protosw.h
> There is no such word as `detatch'.
> ok mglocker (mk@)
ufs/ffs
~ ffs_vfsops.c
> Lock the vnode in ffs_unmount() before passing it too vinvalbuf()
> and VOP_CLOSE() since both of them expected it too be locked.
> ok art@ (thib@)
~ ffs_alloc.c
> When selecting cylinder groups as candidates for new directory creation
> make sure that theres a minimum amount of free blocks in that cylinder
> group by tweaking the minbfree calculation. Without this, cylinder groups
> with no free blocks could be chosen which resaults in expensive searches
> for free blocks each time a new file is created in the directory.
> Tweak the calculation of minifree (minimum free inodes) in the same
> way. Also decrease the maxcontigdirs as the file system fills up to
> lessen the chance of directory clusters overflowing the available
> space in the cylinder group.
> From FreeBSD (Rev 1.116 of sys/ufs/ffs/ffs_alloc.c);
> ok miod@ (thib@)
uvm
~ uvm_extern.h ~ uvm_map.c
~ uvm_pmap.h
> Introduce a md pmap hook, pmap_remove_holes(), which is supposed to mark
> the holes a MMU may have from a given vm_map. This will be automagically
> invoked for newly created vmspaces.
> On platforms with MMU holes (e.g. sun4, sun4c and vax), this prevents
> mmap(2) hints which would end up being in the hole to be accepted as valid,
> causing unexpected signals when the process tries to access the hole
> (since pmap can not fill the hole anyway).
> Unfortunately, the logic mmap() uses to pick a valid address for anonymous
> mappings needs work, as it will only try to find an address higher than the
> hint, which causes all mmap() with a hint in the hole to fail on vax. This
> will be improved later. (miod@)
~ uvm_map.c ~ uvm_unix.c
> replace ctob and btoc with ptoa and atop respectively
> help and ok miod@ thib@ (martin@)
== usr.bin =========================================================== 10/11 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/usr.bin
bgplg
~ bgplg.c
> Fix missing free statement due to extraneous semicolon spotted by a
> modified lint. (cloder@)
cdio
~ cddb.c
> Make sure we don't skip the code which sends the QUIT command to the CDDB
> server and validates the returned track names. Unreachable code spotted
> by lint.
> OK krw@, deraadt@, espie@ (cloder@)
cvs
~ checkout.c ~ file.c
~ util.c
> Added support for checkout -p with local repository.
> OK joris@ (tobias@)
~ cvs.c
> Fixed segmentation fault that occured during .cvsrc parsing if an invalid
> cvs command has been specified.
> Spotted by Mathias De Belder <mathias at woei dot be> who also supplied a
> patch, although I took his second suggestion.
> OK joris@ (tobias@)
~ history.c
> Properly display list of commands in default 80 column terminal.
> OK joris@ (tobias@)
~ client.c ~ cvs.h
~ root.c
> Allow port specification in our CVSROOT which is used for our
> ssh connection to our server.
> Example: joris at somehost.com:2222:/cvs
> OK tobias@ (joris@)
~ diff3.c
> Proper use of fseek/fseeko macros.
> OK joris@, otto@ (tobias@)
~ client.c ~ cvs.h
~ root.c
> remove my addition of portnumber for rsh in CVSROOT.
> while it sounds like a "good" idea some people convinced me otherwise.
> mostly because this breaks compat with GNU cvs. (joris@)
~ annotate.c ~ rcs.c
~ rcs.h ~ util.h
> Added annotate support for local and remote repositories. Behaves like
> GNU cvs but is a little faster.
> OK joris@, ray@, xsa@ (tobias@)
diff
~ diffreg.c
> use strcspn to properly overwrite '\n' in fgets returned buffer
> ok pyr@, ray@, millert@, moritz@, chl@ (gilles@)
diff3
~ diff3prog.c
> Proper use of fseek/fseeko macros.
> OK joris@, otto@ (tobias@)
file
~ print.c ~ softmagic.c
> use strcspn to properly overwrite '\n' in fgets returned buffer
> ok pyr@, ray@, millert@, moritz@, chl@ (gilles@)
~ print.c
> remove some warnings:
> unused variable `variable'
> `variable' might be used uninitialized in this function
> ok gilles@ ray@ (chl@)
ftp
~ cmds.c ~ util.c
> use strcspn to properly overwrite '\n' in fgets returned buffer
> ok pyr@, ray@, millert@, moritz@, chl@ (gilles@)
gprof
~ aout.c
> Proper use of fseek/fseeko macros.
> OK joris@, otto@ (tobias@)
gzsig
~ sign.c
> use strcspn to properly overwrite '\n' in fgets returned buffer
> ok pyr@, ray@, millert@, moritz@, chl@ (gilles@)
mail
~ collect.c ~ edit.c
~ fio.c ~ quit.c
> Proper use of fseek/fseeko macros.
> OK joris@, otto@ (tobias@)
make
~ arch.c ~ compat.c
~ config.h ~ for.c
~ main.c ~ make.c
~ parsevar.c ~ suff.c
~ var.c ~ varmodifiers.c
~ varmodifiers.h ~ lst.lib/lstAddNew.c
~ lst.lib/lstAppend.c ~ lst.lib/lstConcat.c
~ lst.lib/lstConcatDestroy.c ~ lst.lib/lstDeQueue.c
~ lst.lib/lstDestroy.c ~ lst.lib/lstDupl.c
~ lst.lib/lstFindFrom.c ~ lst.lib/lstForEachFrom.c
~ lst.lib/lstInit.c ~ lst.lib/lstInsert.c
~ lst.lib/lstMember.c ~ lst.lib/lstRemove.c
~ lst.lib/lstReplace.c ~ lst.lib/lstSucc.c
> remove dead code: #define RECHECK is always on (espie@)
~ arch.c ~ config.h
~ suff.c
> more dead code: we always define LIBRARIES and INCLUDES (espie@)
~ Makefile ~ arch.c
~ dir.c ~ dir.h
~ main.c ~ parse.c
~ suff.c + direxpand.c
+ direxpand.h
> cut up dir.c into dir.c/direxpand.c as there is very little interface
> between
> two modules that do different things. (espie@)
~ Makefile ~ generate.c
+ node_int.h
> scaffolding to recognize special targets/sources using ohash functions.
> (espie@)
~ Makefile ~ compat.c
~ job.c ~ make.c
+ engine.c + engine.h
> first step towards sanity: take the functions common to parallel/not
> parallel
> make outside of make.c and job.c, and create an engine.c file to hold them.
> (espie@)
~ dir.c ~ memory.c
~ memory.h
> move free_ohash function so that we can reuse it. (espie@)
~ job.c ~ parse.c
~ PSD.doc/tutorial.ms
> kill .SHELL, make doesn't even support it in non-parallel mode. (espie@)
~ arch.c
> reindent arch.c (espie@)
~ arch.c
> reuse free_hash (espie@)
~ job.c ~ var.c
~ varmodifiers.c
> remove warnings (espie@)
~ varmodifiers.c
> reindent varmodifiers.c, no binary change (espie@)
~ arch.c ~ compat.c
~ dir.c ~ engine.c
~ make.c ~ targ.c
~ targ.h ~ timestamp.c
~ timestamp.h
> rename Targ_FmtTime into time_to_string and move it, as it's not related
> to target nodes at all (reduces modules inter-dependencies) (espie@)
~ dir.c ~ dir.h
~ direxpand.c
> rename Path to less misleading PathEntry. Get rid of typedef. (espie@)
~ arch.c
> minor reindent issues (espie@)
~ arch.c ~ cond.c
~ dir.c ~ dir.h
~ main.c ~ parse.c
~ suff.c
> rename dirSearchPath -> defaultPath, and openDirectories ->
> knownDirectories (espie@)
~ targ.c
> reindent targ.c (espie@)
~ compat.c
> reindent compat.c (espie@)
~ dir.c ~ dir.h
~ targ.c
> remove old debug code nothing uses (espie@)
~ engine.c
> reindent (espie@)
~ job.c
> reindent (espie@)
~ str.c
> remove some of the complexity of Str_Matchi by separating code into
> range_match and star_match to handle special characters.
> reindent. (espie@)
~ targ.c
> remove some totally useless comments (espie@)
~ str.c
> finish reindenting str.c (espie@)
mg
~ dired.c
> use strcspn to properly overwrite '\n' in fgets returned buffer
> ok pyr@, ray@, millert@, moritz@, chl@ (gilles@)
~ undo.c
> I can't think of any good reason to inline this.
> Makes pcc happier, to boot. (kjell@)
mkstr
~ mkstr.c
> Proper use of fseek/fseeko macros.
> OK joris@, otto@ (tobias@)
msgs
~ msgs.c
> check fgets return value
> use strcspn to properly overwrite '\n' in fgets returned buffer
> with help and ok from moritz@ and ray@ (chl@)
~ msgs.c
> Avoid setting a global pointer variable from signal handlers
> OK deraadt (cloder@)
netstat
~ if.c
> More 64 bits counters fixes.
> From brad.
> ok mk claudio (mk@)
~ inet6.c ~ mroute6.c
> u_quad_t -> u_int64_t and %qu -> %llu, from brad (henning@)
pcc
+ Makefile.in + config.guess
+ config.h.in + config.sub
+ configure + configure.ac
+ install-sh + arch/m16c/local.c
+ arch/m16c/TODO + arch/m16c/code.c
+ arch/m16c/local2.c + arch/m16c/macdefs.h
+ arch/m16c/order.c + arch/m16c/table.c
+ arch/mips/local.c + arch/mips/TODO
+ arch/mips/code.c + arch/mips/local2.c
+ arch/mips/macdefs.h + arch/mips/order.c
+ arch/mips/table.c + arch/nova/local.c
+ arch/nova/README + arch/nova/code.c
+ arch/nova/local2.c + arch/nova/macdefs.h
+ arch/nova/order.c + arch/nova/table.c
+ arch/pdp10/local.c + arch/pdp10/code.c
+ arch/pdp10/local2.c + arch/pdp10/macdefs.h
+ arch/pdp10/order.c + arch/pdp10/table.c
+ arch/vax/local.c + arch/vax/code.c
+ arch/vax/local2.c + arch/vax/macdefs.h
+ arch/vax/order.c + arch/vax/table.c
+ arch/x86/local.c + arch/x86/code.c
+ arch/x86/local2.c + arch/x86/macdefs.h
+ arch/x86/order.c + arch/x86/table.c
+ cc/Makefile.in + cc/cc/Makefile.in
+ cc/cc/cc.1 + cc/cc/cc.c
+ cc/cpp/Makefile.in + cc/cpp/cpp.c
+ cc/cpp/cpp.h + cc/cpp/cpy.y
+ cc/cpp/scanner.l + cc/cpp/token.c
+ cc/cpp/tests/res1 + cc/cpp/tests/res2
+ cc/cpp/tests/res3 + cc/cpp/tests/res4
+ cc/cpp/tests/res5 + cc/cpp/tests/res6
+ cc/cpp/tests/res7 + cc/cpp/tests/res8
+ cc/cpp/tests/res9 + cc/cpp/tests/test1
+ cc/cpp/tests/test2 + cc/cpp/tests/test3
+ cc/cpp/tests/test4 + cc/cpp/tests/test5
+ cc/cpp/tests/test6 + cc/cpp/tests/test7
+ cc/cpp/tests/test8 + cc/cpp/tests/test9
+ cc/ccom/Makefile.in + cc/ccom/ccom.1
+ cc/ccom/cgram.y + cc/ccom/gcc_compat.c
+ cc/ccom/init.c + cc/ccom/inline.c
+ cc/ccom/main.c + cc/ccom/optim.c
+ cc/ccom/pass1.h + cc/ccom/pftn.c
+ cc/ccom/scan.l + cc/ccom/stabs.c
+ cc/ccom/symtabs.c + cc/ccom/trees.c
+ f77/Makefile.in + f77/f77/Makefile.in
+ f77/fcom/Makefile.in + mip/common.c
+ mip/manifest.h + mip/match.c
+ mip/mkext.c + mip/node.h
+ mip/optim2.c + mip/pass2.h
+ mip/protos.h + mip/reader.c
+ mip/regs.c + os/dragonfly/ccconfig.h
+ os/linux/ccconfig.h + os/netbsd/ccconfig.h
+ os/netbsd/f77config.h + os/none/ccconfig.h
+ os/openbsd/ccconfig.h + os/openbsd/f77config.h
> New import:
> Import ragge's version of PCC into our tree, so we can hack on it more
+ Makefile + config.h
+ cc/Makefile + cc/cc/Makefile
+ cc/ccom/Makefile + cc/cpp/Makefile
+ cc/cpp/cpp.1
> "un"configure, to make it possible to build in obj dir. No need to run
> configure any more. (otto@)
- Makefile.in - config.guess
- config.h.in - config.sub
- configure - configure.ac
- install-sh
> remove files we do not need anymore since we moved to fixed Makefiles
> (otto@)
~ cc/cc/Makefile ~ cc/ccom/Makefile
~ cc/cpp/Makefile
> fix man installation path (otto@)
- cc/Makefile.in - cc/cc/Makefile.in
- cc/ccom/Makefile.in - cc/cpp/Makefile.in
> more files we do not need anymore (otto@)
~ cc/ccom/Makefile
> cleanup some files generated in obj (otto@)
~ Makefile ~ config.h
~ cc/Makefile ~ cc/cpp/cpp.1
> $OpenBSD$
> OK otto. (ray@)
~ config.h
> Lost the last $ in $OpenBSD$. (ray@)
~ arch/m16c/TODO ~ arch/m16c/code.c
~ arch/m16c/local.c ~ arch/m16c/local2.c
~ arch/m16c/macdefs.h ~ arch/m16c/order.c
~ arch/m16c/table.c ~ arch/mips/TODO
~ arch/mips/code.c ~ arch/mips/local.c
~ arch/mips/local2.c ~ arch/mips/macdefs.h
~ arch/mips/order.c ~ arch/mips/table.c
~ arch/nova/README ~ arch/nova/code.c
~ arch/nova/local.c ~ arch/nova/local2.c
~ arch/nova/macdefs.h ~ arch/nova/order.c
~ arch/nova/table.c ~ arch/pdp10/code.c
~ arch/pdp10/local.c ~ arch/pdp10/local2.c
~ arch/pdp10/macdefs.h ~ arch/pdp10/order.c
~ arch/pdp10/table.c ~ arch/vax/code.c
~ arch/vax/local.c ~ arch/vax/local2.c
~ arch/vax/macdefs.h ~ arch/vax/order.c
~ arch/vax/table.c ~ arch/x86/code.c
~ arch/x86/local.c ~ arch/x86/local2.c
~ arch/x86/macdefs.h ~ arch/x86/order.c
~ arch/x86/table.c ~ cc/cc/cc.1
~ cc/cc/cc.c ~ cc/ccom/ccom.1
~ cc/ccom/cgram.y ~ cc/ccom/gcc_compat.c
~ cc/ccom/init.c ~ cc/ccom/inline.c
~ cc/ccom/main.c ~ cc/ccom/optim.c
~ cc/ccom/pass1.h ~ cc/ccom/pftn.c
~ cc/ccom/scan.l ~ cc/ccom/stabs.c
~ cc/ccom/symtabs.c ~ cc/ccom/trees.c
~ cc/cpp/cpp.c ~ cc/cpp/cpp.h
~ cc/cpp/cpy.y ~ cc/cpp/scanner.l
~ cc/cpp/token.c ~ f77/Makefile.in
~ f77/f77/Makefile.in ~ f77/fcom/Makefile.in
~ mip/common.c ~ mip/manifest.h
~ mip/match.c ~ mip/mkext.c
~ mip/node.h ~ mip/optim2.c
~ mip/pass2.h ~ mip/protos.h
~ mip/reader.c ~ mip/regs.c
~ os/dragonfly/ccconfig.h ~ os/linux/ccconfig.h
~ os/netbsd/ccconfig.h ~ os/netbsd/f77config.h
~ os/none/ccconfig.h ~ os/openbsd/ccconfig.h
~ os/openbsd/f77config.h
> $OpenBSD$
> Requested by otto. (ray@)
~ cc/ccom/Makefile
> remove cgram.h from CLEANFILES, so make depend; make; make clean; make
> works; instead reomve cgram.h when doing make cleandir (otto@)
~ cc/cc/cc.c
> Process (and ignore for now) -pipe (otto@)
~ cc/ccom/pftn.c ~ cc/ccom/scan.l
~ cc/ccom/trees.c ~ cc/cpp/cpp.c
~ cc/cpp/scanner.l ~ mip/common.c
~ mip/regs.c
> sync to ragge's tree to fix #include<file.h> and a declaration parsing
> error. (otto@)
~ cc/cc/Makefile ~ cc/ccom/Makefile
~ cc/cpp/Makefile
> change MANDIR to ${PREFIX}/man. ok otto@ (kjell@)
~ arch/x86/local2.c
> Merge from ragge's tree:
> Needed an extra check for leal to avoid --. Reported by otto. (otto@)
rcs
~ rlog.1 ~ rlog.c
> Have rlog exit 1 if any of the files could not be opened.
> Don't mention -q flag, which is there for compatibility.
> From Pierre Riteau.
> OK joris for first part, OK niallo and jmc for second. (ray@)
~ rlog.1
> Mention -q in compatibility section.
> OK jmc. (ray@)
~ ident.c
> From Pierre Riteau:
> Sync ident behavior with GNU RCS:
> - return 1 to the shell if any file could not be opened.
> - print an empty line after processing a file unless we couldn't
> open it (and not for the last one).
> OK niallo. (ray@)
~ diff3.c
> Proper use of fseek/fseeko macros.
> OK joris@, otto@ (tobias@)
~ diff.c
> use strcspn to properly overwrite '\n' in fgets returned buffer
> ok pyr@, ray@, millert@, moritz@, chl@ (gilles@)
rdist
~ message.c
> use strcspn to properly overwrite '\n' in fgets returned buffer
> ok pyr@, ray@, millert@, moritz@, chl@ (gilles@)
~ message.c
> remove some warnings:
> unused variable `variable'
> `variable' might be used uninitialized in this function
> ok gilles@ ray@ (chl@)
readlink
~ readlink.c
> optional flags must be sorted and enclosed in single brackets,
> as style(9) suggests; usage() should be a function; minor KNF fix
> ok otto@ (sobrado@)
ssh
~ ssh-add.c ~ ssh-agent.1
~ ssh-agent.c ~ ssh-keygen.c
> sort synopsis and options in ssh-agent(1); usage is lowercase
> ok jmc@ (sobrado@)
~ sshpty.c
> sort #include (stevesk@)
~ session.c ~ ssh-keygen.c
~ sshlogin.c
> use strcspn to properly overwrite '\n' in fgets returned buffer
> ok pyr@, ray@, millert@, moritz@, chl@ (gilles@)
~ sshpty.c
> remove #if defined block not needed; ok markus@ dtucker@ (stevesk@)
~ umac.c
> use xmalloc() and xfree(); ok markus@ pvalchev@ (stevesk@)
~ sftp-server.c
> fix incorrect test when setting syslog facility; from Jan Pechanec (djm@)
~ sftp-client.c
> use off_t instead of u_int64_t for file offsets, matching what the
> progressmeter code expects; bz #842 (djm@)
sudo
~ parse.yacc
> Use erealloc3() in more_aliases() instead of malloc/realloc. (millert@)
sup
~ src/scan.c ~ src/supcmain.c
~ src/supcmeat.c ~ src/supcname.c
~ src/supfilesrv.c ~ src/supscan.c
> use strcspn to properly overwrite '\n' in fgets returned buffer
> ok pyr@, ray@, millert@, moritz@, chl@ (gilles@)
~ src/supcmeat.c ~ src/supfilesrv.c
> remove some warnings:
> unused variable `variable'
> `variable' might be used uninitialized in this function
> ok gilles@ ray@ (chl@)
systat
~ disks.c ~ netcmds.c
> use strcspn to properly overwrite '\n' in fgets returned buffer
> ok pyr@, ray@, millert@, moritz@, chl@ (gilles@)
tail
~ read.c
> Plug mem leak. From Igor Zinovik; ok millert@ (otto@)
~ read.c
> Another mem leak plugged. With help from Igor Zinovi. ok millert@ (otto@)
tftp
~ tftpsubs.c
> Proper use of fseek/fseeko macros.
> OK joris@, otto@ (tobias@)
tip
~ tip.h
> Use sig_atomic_t where required. OK moritz, deraadt (cloder@)
top
~ top.c
> typo in message, from Mark Lumsden. (otto@)
tset
~ tset.c
> use strcspn to properly overwrite '\n' in fgets returned buffer
> ok pyr@, ray@, millert@, moritz@, chl@ (gilles@)
~ tset.c
> remove some warnings:
> unused variable `variable'
> `variable' might be used uninitialized in this function
> ok gilles@ ray@ (chl@)
vacation
~ vacation.c
> use strcspn to properly overwrite '\n' in fgets returned buffer
> ok pyr@, ray@, millert@, moritz@, chl@ (gilles@)
vi
~ common/recover.c ~ ex/ex_cscope.c
> use strcspn to properly overwrite '\n' in fgets returned buffer
> ok pyr@, ray@, millert@, moritz@, chl@ (gilles@)
~ common/recover.c
> remove some warnings:
> unused variable `variable'
> `variable' might be used uninitialized in this function
> ok gilles@ ray@ (chl@)
vmstat
~ vmstat.c
> Use a sig_atomic_t instead of an int when setting it from a signal handler
> (cloder@)
xstr
~ xstr.c
> Proper use of fseek/fseeko macros.
> OK joris@, otto@ (tobias@)
== usr.sbin ========================================================== 11/11 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/usr.sbin
bgpctl
~ bgpctl.c
> baudrate is 64bit now, plus print w/ %llu (henning@)
bgpd
~ bgpd.h
> baudrate u_long -> u_int64_t (henning@)
~ parse.y
> Move parser to use NUMBER as all other parse.y do know. A bit tricky
> because
> all the relative metrics need some special handling. OK henning@ deraadt@
> (claudio@)
~ rde.c
> log_fmt_peer() does not use a static buffer so it is necessary to free the
> string that it returned later. rde_update_log() did not do that.
> Memory leak found and diff provided by Mike Belopuhov. (claudio@)
config
~ misc.c
> use strcspn to properly overwrite '\n' in fgets returned buffer
> ok pyr@, ray@, millert@, moritz@, chl@ (gilles@)
cron
~ env.c
> Proper use of fseek/fseeko macros.
> OK joris@, otto@ (tobias@)
dvmrpd
~ dvmrpd.h
> baudrate 64bit, two more questionable marked for norby (henning@)
~ dvmrpd.h ~ interface.c
> The same dance as in all other routing daemons. baudrate is 64bit plus
> initialize iface->baudrate when the interface is created. An additional
> gimmick don't set the ictl->baudrate twice when building the control
> message. (claudio@)
~ parse.y
> extend lex to spot numbers in the stream, without impacting the parsing
> of ip addresses and such. this change is being pushed into all the pfctl
> derived parsers, starting with the easy ones; ok claudio michele
> (deraadt@)
~ parse.y
> macro argument unused, using local instead; spotted by mpf (deraadt@)
~ parse.y
> spacing (deraadt@)
eeprom
~ optree.c
> better printing code, fixing certain properties; reported by kettenis,
> tested by him and Johan Mson Lindman. ok kettenis. (fgsch@)
hostapd
~ parse.y
> Add NUMBER support for signed and unsigned 64bit numbers to the lexer
> instead of passing numbers as STRINGs. Add careful range checks
> whenever NUMBERs are used in the grammar.
> From deraadt@ (except the specific range checks) (reyk@)
hoststated
~ Makefile ~ hoststated.conf.5
~ hoststated.h ~ parse.y
~ relay.c + relay_udp.c
> add support for relaying DNS traffic (with a little bit of packet
> header randomization). this adds an infrastructure to support
> UDP-based protocols.
> ok gilles@, tested by some (reyk@)
httpd
~ conf/mime.types
> add application/x-bittorrent
> From: Alex Holst <a at mongers.org> (henning@)
~ src/modules/standard/mod_mime_magic.c
~ src/support/logresolve.c
> use strcspn to properly overwrite '\n' in fgets returned buffer
> ok pyr@, ray@, millert@, moritz@, chl@ (gilles@)
~ src/main/buff.c
> [fF]uther -> [fF]urther in comments and man page. First one spotted on
> tech@ by Jung. (krw@)
ifstated
~ parse.y
> extend lex to spot numbers in the stream, without impacting the parsing
> of ip addresses and such. this change is being pushed into all the pfctl
> derived parsers, starting with the easier ones; range check written by
> mpf; ok mpf (deraadt@)
~ parse.y
> unchecked conversion from signed 64 to unsigned 32 (deraadt@)
inetd
~ inetd.c
> use strcspn to properly overwrite '\n' in fgets returned buffer
> ok pyr@, ray@, millert@, moritz@, chl@ (gilles@)
lpr
~ lpd/printjob.c
> Proper use of fseek/fseeko macros.
> OK joris@, otto@ (tobias@)
ntpd
~ parse.y
> this is where it all started, since future ntpd.conf commands will require
> negative parameters. extend lex to spot numbers in the stream. as well,
> make it easier to add parameters on command line in any order later
> ok otto ckuethe (deraadt@)
~ parse.y
> default weight has to remain 1; seen by Maurice Janssen (deraadt@)
~ ntpd.conf.5 ~ ntpd.h
~ parse.y ~ sensors.c
> Add a knob to compensate for a refclock that is early or late. Based on a
> diff from Maurice Janssen. Manpage help from jmc and Maurice, other nits
> from deraadt and otto.
> ok deraadt, otto (ckuethe@)
~ ntpd.conf.5
> Small style tweak, from jmc and Maurice Janssen (ckuethe@)
~ ntpd.conf.5
> one more missed change; (jmc@)
~ ntpd.8 ~ ntpd.c
~ ntpd.h
> Provide the -n switch like in the other imsg daemons for testing
> the configuration file.
> "yes please, ok" henning@ (pyr@)
~ ntpd.c
> add -n to usage(); (jmc@)
~ parse.y
> Correctly assign a default weight of 1 to sensors and servers.
> ok beck (ckuethe@)
~ parse.y
> use a setup function for options, cleaner; ok cloder (deraadt@)
ospfctl
~ ospfctl.c
> Baudrate is 64bit so use %llu in printf(). Reminded by chl@ (claudio@)
ospfd
~ ospfd.8
> Improve the ospfd(8) manpage.
> Hooray! claudio@ (norby@)
~ interface.c ~ ospfd.h
> Switch some more baudrates to 64 bit and initialize iface->baudrate.
> (claudio@)
~ ospfd.8
> tweak previous; (jmc@)
~ parse.y
> Another parser.y that switches to parse numbers directly in the lexer.
> Most complex part was "redistribute 10/8" which need special handling.
> With and OK deraadt@, OK norby@ (claudio@)
~ ospfd.8
> [fF]uther -> [fF]urther in comments and man page. First one spotted on
> tech@ by Jung. (krw@)
~ rde_spf.c
> Change the way nexthops are calculated on the root level. Instead of
> looking
> from the target back and trying to figure the nexthop out, the link is
> searched
> in the interface list and the info from the matiching interface is used.
> This should solve the nexthop issues with setups having multiple
> point-to-point
> links between two routers as reported on misc@ some time ago.
> tested and OK norby@ (claudio@)
ppp
~ ppp/auth.c
> use strcspn to properly overwrite '\n' in fgets returned buffer
> ok pyr@, ray@, millert@, moritz@, chl@ (gilles@)
pppd
~ chat/chat.c
> use strcspn to properly overwrite '\n' in fgets returned buffer
> ok pyr@, ray@, millert@, moritz@, chl@ (gilles@)
ripctl
~ ripctl.c
> 64bit baudrate and %llu to print (henning@)
ripd
~ ripd.h
> baudrate, two non-obvious ones marked for michele (henning@)
~ interface.c ~ ripd.h
> More baudrate fixes. Again don't forget to initialize the iface->baudrate.
> (claudio@)
~ parse.y
> extend lex to spot numbers in the stream, without impacting the parsing
> of ip addresses and such. this change is being pushed into all the pfctl
> derived parsers, starting with the easy ones; ok claudio michele
> (deraadt@)
~ parse.y
> macro argument unused, using local instead; spotted by mpf (deraadt@)
sliplogin
~ sliplogin.c
> use strcspn to properly overwrite '\n' in fgets returned buffer
> ok pyr@, ray@, millert@, moritz@, chl@ (gilles@)
~ sliplogin.c
> remove some warnings:
> unused variable `variable'
> `variable' might be used uninitialized in this function
> ok gilles@ ray@ (chl@)
===============================================================================
More information about the owc
mailing list