[odc] Daily src changes for 2004-02-08
ODC
auto at squish.net
Mon Feb 9 07:06:49 GMT 2004
OpenBSD src changes summary for 2004-02-08
==========================================
bin/ksh bin/ps
distrib/amd64 distrib/miniroot
distrib/notes distrib/sets
etc/etc.amd64/MAKEDEV gnu/usr.bin/perl
lib/libc lib/libm
sbin/dhclient share/man
share/mk share/tmac
sys/arch/amd64/amd64 sys/arch/amd64/compile
sys/arch/amd64/conf sys/arch/i386/i386
sys/arch/sparc/include sys/arch/sparc/sparc
sys/crypto sys/dev/ic
sys/dev/microcode sys/dev/raidframe
sys/kern sys/net
usr.sbin/bgpd usr.sbin/bind
usr.sbin/pppd
== bin =============================================================== 01/09 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/bin
ksh
~ config.h ~ sh.h
> forget about 16 bit machines; millert ok (deraadt@)
ps
~ ps.c
> 'ps' should not act like 'ps a'. Resolves PR 3676.
> ok tdeval@ millert@ (otto@)
== distrib =========================================================== 02/09 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/distrib
amd64
~ common/list
> XXX Why does amd64 require a bin/-sh to boot correctly?? Is this because
> of some flags being passed in? Anyone? (deraadt@)
~ common/list
> workaround for awk breakage no longer needed (deraadt@)
miniroot
~ install.sub
> Fix typo: 2>&2 -> 2>&1. (krw@)
~ install.sub
> Add loopback routes late during upgrades, as /etc/netstart does now.
> Scavange a couple of characters while in the area. (krw@)
~ install.sub
> Replace 'localhost' with '127.0.0.1' and add -q to route invocations
> in enable_network(). From /etc/netstart. (krw@)
~ install.sub
> Shave a few more characters. (krw@)
notes
+ cats/contents + cats/hardware
+ cats/install + cats/prep
+ cats/upgrade + cats/whatis
+ cats/xfer
> Add cats documentation, still needs lots of work. (drahn@)
~ m4.common
> Oops, InstallPart4 would eat its args. (miod@)
~ cats/contents ~ cats/install
~ cats/prep ~ cats/upgrade
~ cats/whatis ~ cats/xfer
> First pass at a whacking. (miod@)
~ cats/install
> Even better (miod@)
sets
~ lists/man/mi
> sync (deraadt@)
~ lists/comp/md.amd64
> sync (deraadt@)
~ lists/man/md.amd64
> sync (deraadt@)
== etc =============================================================== 03/09 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/etc
etc.amd64/MAKEDEV
~ etc.amd64/MAKEDEV
> fix, including ptm, ahead of time (deraadt@)
== gnu =============================================================== 04/09 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/gnu
usr.bin/perl
~ Makefile.bsd-wrapper
> shave build time.
> okay millert@ (espie@)
== lib =============================================================== 05/09 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/lib
libc
~ arch/x86_64/gen/modf.S
> from freebsd (and appears to make our awk work better)
> Tidy up modf.S and make it actually work. It wasn't extracting
> the value out of ST(0) before copying it to %xmm0. Also remove bogus stack
> frame and work in the red zone. (deraadt@)
~ arch/x86_64/gen/fabs.S
> from freebsd, helps awk too:
> Fix fabs(). This commit brought to you by the letter 'l'.
> (fstp stores a mem32 value, fstpl stores a mem64 value) (deraadt@)
~ arch/x86_64/gen/infinity.c
> 16 byte align for performance, as on other architectures (deraadt@)
~ sys/select.2
> FD_SETSIZE has been cranked up to 1024
> ok deraadt@ millert@ (otto@)
libm
~ Makefile + arch/x86_64/abi.h
+ arch/x86_64/e_acos.S + arch/x86_64/e_asin.S
+ arch/x86_64/e_atan2.S + arch/x86_64/e_atan2f.S
+ arch/x86_64/e_exp.S + arch/x86_64/e_fmod.S
+ arch/x86_64/e_log.S + arch/x86_64/e_log10.S
+ arch/x86_64/e_remainder.S + arch/x86_64/e_remainderf.S
+ arch/x86_64/e_scalb.S + arch/x86_64/e_sqrt.S
+ arch/x86_64/e_sqrtf.S + arch/x86_64/s_atan.S
+ arch/x86_64/s_atanf.S + arch/x86_64/s_ceil.S
+ arch/x86_64/s_ceilf.S + arch/x86_64/s_copysign.S
+ arch/x86_64/s_copysignf.S + arch/x86_64/s_cos.S
+ arch/x86_64/s_cosf.S + arch/x86_64/s_finite.S
+ arch/x86_64/s_finitef.S + arch/x86_64/s_floor.S
+ arch/x86_64/s_floorf.S + arch/x86_64/s_ilogb.S
+ arch/x86_64/s_ilogbf.S + arch/x86_64/s_log1p.S
+ arch/x86_64/s_log1pf.S + arch/x86_64/s_logb.S
+ arch/x86_64/s_logbf.S + arch/x86_64/s_rint.S
+ arch/x86_64/s_rintf.S + arch/x86_64/s_scalbn.S
+ arch/x86_64/s_scalbnf.S + arch/x86_64/s_significand.S
+ arch/x86_64/s_significandf.S + arch/x86_64/s_sin.S
+ arch/x86_64/s_sinf.S + arch/x86_64/s_tan.S
+ arch/x86_64/s_tanf.S
> add libm parts for amd64; much from netbsd, integration by mickey.
> Instead of mixing the i387 & xmm register components, we decide to
> explicitly seperate them. libm is already confusing enough as it is,
> thank you very much (deraadt@)
== sbin ============================================================== 06/09 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/sbin
dhclient
~ osdep.h
> fix preprocessor syntax; ok henning@ (naddy@)
== share ============================================================= 07/09 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/share
man
~ man5/bsd.port.mk.5
> kill whitespace at EOL;
> new sentence, new line; (jmc@)
~ man4/man4.cats/footbridge.4 ~ man4/man4.cats/fcom.4
> mark .Dt as being cats pages;
> ok miod@ (jmc@)
~ man5/types.5
> Update FD_SETSIZE value to 1024. Spotted by and ok jmc@ (otto@)
mk
~ bsd.man.mk
> 3p suffix, formatting.
> okay millert@
> should have committed before the perl change, except I checked it thru a
> make build first, and then forgot to commit. (espie@)
tmac
~ mdoc/doc-common
> add .Dt entries for amd64 and cats;
> ok miod@ (jmc@)
== sys =============================================================== 08/09 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/sys
arch/amd64/amd64
~ autoconf.c
> correct ramdisk support (deraadt@)
arch/amd64/compile
~ .cvsignore
> ignore RAMDISK_CD too (deraadt@)
arch/amd64/conf
+ RAMDISK_CD
> ramdisk; from mickey (deraadt@)
~ GENERIC
> better one; from mickey (deraadt@)
~ Makefile.amd64
> minimum CPPFLAGS (deraadt@)
arch/i386/i386
~ machdep.c
> for the via c3 use software aes keys, thereby permitting 192 and 256.
> this also speed things up substantially. ipsec tested by markus (deraadt@)
arch/sparc/include
~ ptrace.h
> restored & repaired wcookie support; kettenis at chello.nleraadt@)
arch/sparc/sparc
~ process_machdep.c
> restored & repaired wcookie support; kettenis at chello.nleraadt@)
crypto
~ rijndael.c
> save kernel space; full unroll is not needed; from freebsd (deraadt@)
dev/ic
~ aic7xxx.c ~ aic7xxx.h
> Pull in a few recent changes from FreeBSD, plus a couple of local
> tweaks.
> ok deraadt at .rw@)
dev/microcode
~ aic7xxx/aic7xxx.reg ~ aic7xxx/aic7xxx.seq
~ aic7xxx/aic7xxx_reg.h ~ aic7xxx/aic7xxx_seq.h
> Pull in a few recent changes from FreeBSD, plus a couple of local
> tweaks.
> ok deraadt at .rw@)
dev/raidframe
- rf_netbsd.h - rf_netbsdkintf.c
> Remove unused files.
> ok tdeval@ millert@ drahn@ (grange@)
kern
~ sys_process.c
> restored & repaired wcookie support; kettenis at chello.nleraadt@)
net
~ if_pfsync.c
> Fix kernel panic which occurs under very high load:
> - Make sure we calculate the correct maximum size for PFSYNC_ACT_UREQ.
> - Make pfsync_sendout() return immediately if there is nothing to send.
> (mcbride@)
~ if.c
> if_detach_rtdelete(): abort and restart rn_walktree() if a cloning route
> gets deleted; fixes pr 3649; ok henning, deraadt, dhartmei (markus@)
== usr.sbin ========================================================== 09/09 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/usr.sbin
bgpd
~ Makefile ~ parse.y
~ session.h + printconf.c
> factor out functions to print the configuration (rules only for now)
> will become more and needed by bgpctl too (henning@)
bind
~ lib/isccfg/parser.c
> 64-bit fix; ok/idea jakob (pvalchev@)
~ configure.in
> no matter what size "long long" is, the printf format is always %lld. ok
> jakob (pvalchev@)
~ configure
> regenerate (pvalchev@)
~ lib/isccfg/parser.c
> back out last change since it was unnecessary, the last bind merge fixed
> the deeper issue (pvalchev@)
pppd
~ main.c
> amd64 needs the same hack as powerpc; but this code really needs to be
> deleted/replaced. anyone want to step up? (deraadt@)
===============================================================================
More information about the odc
mailing list