CVS log for src/share/mk/sys.mk

Up to [cvs.NetBSD.org] / src / share / mk

Request diff between arbitrary revisions

Keyword substitution: kv
Default branch: MAIN

Pull up following revision(s) (requested by wiz in ticket #526):

	share/mk/sys.mk: revision 1.149

Update ARFLAGS for new binutils.

In binutils 2.34, 'l' was a no-op.
In binutils 2.39, it is used to specify library dependencies.

Remove 'l' from ARFLAGS.

From Takahiro Kambe in PR 57565.

Update ARFLAGS for new binutils.

In binutils 2.34, 'l' was a no-op.
In binutils 2.39, it is used to specify library dependencies.

Remove 'l' from ARFLAGS.

From Takahiro Kambe in PR 57565.

Pass W with IDU for lint and deps.

sys.mk: fix duplicate character in character class of LINTFLAGS

The character class contained two times U+0020, which makes one of them
redundant, so remove the character class. PR toolchain/18248 already
contained these two spaces.

Apparently there is no need to support tabs in this place, otherwise
someone would have noticed it in the 17 years since this pattern had
been added in sys.mk 1.83 from 2004-01-27.

No functional change.

From Greg A. Woods:
- add LDSTATIC to all the LINK rules, replacing CPPFLAGS which should not be
  used
- create new COMPILE_LINK rules that are like LINK rules but have CPPFLAGS
  where appropriate

Fix the shuttle rules by defining a new EXEC_TARGET instead of re-using
OBJECT_TARGET which adds an extra .o sometimes. Noticed by Greg A. Woods

- make the CTFCONVERT command lines consistent and concise by putting all
  the implementation stuff in a variable CTFCONVERT_RUN
- introduce an OBJECT_TARGET variable that contains the output object file
- when we need to run ctfconvert, go through an intermediate ${.TARGET}.o
  file, instead of writing directly to ${.TARGET} and then overwriting
  ${.TARGET} with ctfconvert. This avoids build failures after a build
  got interrupted (the "partially built from C" scourge).

Merge changes from current as of 20200406

Pull up following revision(s) (requested by maya in ticket #689):

	share/mk/sys.mk: revision 1.143

Don't special case aarch64 and add -fomit-frame-pointer to builds.

This behaviour is probably due to a past behaviour of clang, where it
always emitted frame pointer code.

This is no longer true for clang on netbsd, and I don't think it was true
for GCC.

Meanwhile, this flag bleeds into pkgsrc where it breaks random packages,
requiring workarounds like lang/ruby*-base/hacks.mk.

Don't special case aarch64 and add -fomit-frame-pointer to builds.

This behaviour is probably due to a past behaviour of clang, where it
always emitted frame pointer code.
This is no longer true for clang on netbsd, and I don't think it was true
for GCC.

Meanwhile, this flag bleeds into pkgsrc where it breaks random packages,
requiring workarounds like lang/ruby*-base/hacks.mk.

Back out previous. This has no chance to work unless the make variable parsing
is not changed do that instead of scanning for a single character delim ':',
it scans for "?:". This is because !empty(COMPILE.c:M*-pg*) contains a ':'.

tweak the DBG rule that -fomit-frame-pointer for aarch64 when profiling
(because the compiler complains), to use a match with the compile flags
and *pg*, instead of using a match to a target suffix (which is NetBSD
build-specific). Pointed out by phone@.

Pull up following revision(s) (requested by christos in ticket #575):

	share/mk/bsd.sys.mk: revision 1.298
	share/mk/sys.mk: revision 1.140

move MV to sys.mk because it is used there. Pointed out by joerg@

move MV to sys.mk because it is used there. Pointed out by joerg@

Sync with HEAD

Add -r so that we get the location of redefinitions

Sync with HEAD

Most of the mv operations are to move temporary files to their final place.
Some use -f, others don't. This can lead to spurious build failures when
the user performing the build changes. Centralize, and always use -f.

Sync with HEAD

Resolve a couple of conflicts (result of the uimin/uimax changes)

remove random seed debugging

Sync with HEAD

Back to always using just the list of sources for the hash. Hopefully this
is reproducible now.

Build seed fixes:
    - use MKREPRO_TIMESTAMP as the seed if available.
    - skip relative paths that cannot be normalized if it is not
XXX: pullup-8

Debug hack: print out full random-seed for C++ compile/link.

Sync with HEAD

Add comment describing why we remove -Wsystem-headers, requested by mrg.

Strip -Wsystem-headers from CXXFLAGS.

GCC's C++ headers are not clean (yet). They are trying, but haven't got there
yet.

Necessary for including <atf-c++.hpp> which uses <sstream>.

Pull up following revision(s) (requested by christos in ticket #834):

	usr.bin/make/var.c: revision 1.219
	share/mk/bsd.obj.mk: revision 1.52
	usr.sbin/crash/Makefile: revision 1.40
	share/mk/sys.mk: revision 1.131
	share/mk/bsd.sys.mk: revision 1.276
	share/mk/bsd.sys.mk: revision 1.277

Prevent 'make obj' creating an objdir inside an objdir. This happens the
second time we call 'make obj' in /usr/src/rescue with MAKEOBJDIR set.
XXX: pullup-8
 -
Need to delete ${DESTDIR} for lint
XXX: pullup-8
 -
Fix reproducible builds
XXX: pullup-8
 -
Also substitute ${X11SRCDIR} for the random seed for reproducible builds so
that the random seed is consistent.
XXX: pullup-8
 -
Since ${MAKE} converts $$ -> $ during parsing we need to put it back to
preserve the original variable value with :Q.
XXX: pullup-8
 -
Be more selective about remapping object directories otherwise we pick
objdump and objcopy with the regex.
XXX: pullup-8

Sync with HEAD

Also substitute ${X11SRCDIR} for the random seed for reproducible builds so
that the random seed is consistent.
XXX: pullup-8

add DTRACE_OPTS and HOST_DTRACE_OPTS to allow disabling various optimizations
that interfere with using dtrace.  use them when MKDTRACE=yes.

Remove hacks for ancient vax gcc

Pull up following revision(s) (requested by chs in ticket #866):
	external/cddl/osnet/dist/tools/ctf/cvt/ctfmerge.c: revision 1.12 via patch
	share/mk/sys.mk: revision 1.128
reapply some fixes to make multi-threaded ctfmerge work
that were lost in the overzealous resync with freebsd.
remove the hacks which were applied to work around those problems.

reapply some fixes to make multi-threaded ctfmerge work
that were lost in the overzealous resync with freebsd.
remove the hacks which were applied to work around those problems.

Rebase to HEAD as of a few days ago.

Rebase.

Changes to existing files to enable building AARCH64 userland.
evbarm64-el
This is clang only.  While gcc4.8 supports aarch64, no netbsd support has
been written for aarch64 with gcc4.8.

Wrap CTFCONVERT invocations in ${defined(CTFCONVERT):? ... :}
so it performs the test at the time the rule is invoked.  We can't
use a ".if defined" test because that would be tested at the time the
sys.mk file is parsed, which may be before CTFCONVERT has been set by
bsd.own.mk.

Remove the now-unnecessary assignment "CTFCONVERT ?= : ctfconvert",
and the similar assignment to CTFMERGE.  Now the build logs are not
cluttered by many unnecessary invocations of the ": ctfconvert ..."

Use "-O2 -fno-reorder-blocks" instead of "-Os" for m68k default optimization.

"-Os" was specfied when gcc 4.5.x was imported and the commit log says
"-O2 produces much bigger code with gcc 4.5 than it did with gcc 4.1"
but "-Os" disables most inline declarations and makes some applications
much slower.  "-O2 -fno-reorder-blocks" seems enough to reduce sizes
and disabling -freorder-blocks wouldn't cause particular performance
impact on ancient m68k machines with small cache memories.
See my post on port-m68k@ for more details:
http://mail-index.NetBSD.org/port-m68k/2014/06/22/msg000488.html

No objection in the thread and "seems fine to me" from mrg@.

sync with head.

for a reference, the tree before this commit was tagged
as yamt-pagecache-tag8.

this commit was splitted into small chunks to avoid
a limitation of cvs.  ("Protocol error: too many arguments")

Add -g (gnu) so lint recognizes __attribute__

By default define CTFCONVERT and CTFMERGE to be no-ops in sys.mk.

Don't test for /usr/bin/ctfconvert or /usr/bin/ctfmerge -- that is
completely bonkers and breaks the build if the host's ctfconvert and
ctfmerge are too old.

Strip the broken conditionals in the .c.o &c. rules.

bsd.own.mk will override CTFCONVERT and CTFMERGE with TOOL_CTFCONVERT
and TOOL_CTFMERGE if MKCTF=yes.

XXX The way CTFCONVERT and CTFMERGE are integrated into the rules is
bogus -- it is not safe to interrupt them.  We need some intermediate
object to do this right.

Please let me know if I broke anything with this -- I am not an elvin
or dwarven wizard, nor a dtrace expert.  But this has been breaking
builds for ages now and nobody else seems to want to do anything
about it.

XXX: use only thread for ctfmerge

Elide ctfmerge bug.

disable ctf handling for the shuttle .c -> '' rule since we don't create
an object file to work on.

redo the ctf stuff to handle the USETOOLS=never case correctly and explain
how.

ignore all -std=gnu patterns not just 99

sync with HEAD

Test BSD makefile about coldfire.
XXX still hate m68k--netbsdelf-coldfire

resync from head

don't make ctfconvert remove original debug info

resync with head

CTFCONVERT complains and coredumps for c++

CTFCONVERT me harder.

sync with head

Now that ctf is fixed for at least amd64, go back to an equivalent to the
original functionality, use the tool only if it exists.

Disable CTF Conversion for now until it is fixed.
- ERROR: ctrl_iface_ap.c: die 2767: failed to get unsigned: No entry found
  [dwarf _attrval_unsigned(219)]

Remove bogus conditionals that were also copied to other files presumably
because they did not work from here, and explain why.

Resync to 2012-11-19 00:00:00 UTC

sync with head

old-style-definitions and no-format-zero-length are invalid for c++

Switch -current vax back to gcc 4.1 to match netbsd-6 - until someone has
time to look at the gcc 4.5 codegen issues we should at least have a usable
-current for vax.

Apply patches (requested by martin in ticket #153):

share/mk/bsd.own.mk				patch
share/mk/sys.mk					patch
external/gpl3/gdb/Makefile.inc			patch
external/gpl3/binutils/lib/libbfd/Makefile	patch

	Switch the vax port back to GCC 4.1.
	[martin, ticket #153]

sync with head

Switch VAX to use GCC 4.5 and GDB 7.
Change VAX to use DBG=-O2 like most other ports.

switch everything except vax to gcc 4.5.
switch m68k to -Os since -O2 produces much bigger code
with gcc 4.5 than it did with gcc 4.1.

Fully support per-target AFLAGS.

Compute the build seed using :hash.

Sync with HEAD

Switch remaining platforms to modern CPP for assembler.

Move SPARC and SPARC64 to modern CPP. Update UPDATING note for this and
Alpha.

Alpha doesn't need -traditional-cpp for assembler.

Allow use of traditional CPP to be set on a per platform base in sys.mk.
Honour this for dependency processing in bsd.dep.mk.  Switch i386 and
amd64 assembly to use ISO C90 preprocessor concat and drop the
-traditional-cpp on this platform.

DTrace: Add CTF generation to the make framework.  Only enabled when
MKDTRACE=yes.

Sync with HEAD.

Sync with wrstuden-revivesa-base-2.

Remove -Wno-pointer-sign and -ffreestanding from CXXFLAGS.  g++
doesn't support for them.

When compiling C++ sources and BUILDSEED is set, pass down
-frandom-seed with a reduced version of .ALLSRC and .TARGET
as argument.  This allows bit-identical builds of C++ files.

Sync with HEAD.

sync with head.

For sh3 set default DBG to "-Os -freorder-blocks" which is almost like -O2
but without -falign-* that is too zealous for low-memory sh3 machines.

There is only one more file which requires HACKS for m68000,
so handle it in each Makefile rather than sys.mk.

These ICEs might be related with GCC Bugzilla Bug 32424
which is not resolved yet even in the upstream.

Enable gcc optimization on m68000 (sun2) again with minimum hacks.
gcc -O0 generates a bit fat binaries.

sync with HEAD

make all sun2 use -O0 and move most of the hacks out into just 3 files.

Remove support for NetBSD/pc532.

Sync w/ NetBSD-4-RC_1

Pull up following revision(s) (requested by soren in ticket #774):
	build.sh: revision 1.169
	gnu/usr.bin/gcc4/backend/Makefile: revision 1.9
	sys/arch/evbsh5/README: revision 1.1
	distrib/notes/common/contents: revision 1.124
	usr.sbin/ndbootd/ndbootd.c: revision 1.12
	gnu/lib/libopcodes/arch/sh3eb/defs.mk: revision 1.4
	gnu/lib/libbfd/arch/sh3eb/defs.mk: revision 1.8
	libexec/ld.elf_so/Makefile: revision 1.78
	gnu/dist/binutils/bfd/config.bfd: revision 1.14
	doc/HACKS: revision 1.93
	distrib/cdrom/current.conf: revision 1.8
	gnu/lib/libgcc4/Makefile.inc: revision 1.6
	share/mk/sys.mk: revision 1.90
	distrib/sets/lists/base/mi: revision 1.700
	libexec/ld.elf_so/rtld.c: revision 1.112
	distrib/sets/lists/base/mi: revision 1.702
	sys/arch/sh5/README: revision 1.1
	gnu/lib/libbfd/arch/sh3el/bfd.h: revision 1.7
	usr.sbin/traceroute/traceroute.c: revision 1.69
	distrib/notes/common/main: revision 1.330
	distrib/notes/common/list-setsizes.sh: revision 1.2
	sys/arch/Makefile: revision 1.35
	gnu/lib/libopcodes/arch/sh3el/defs.mk: revision 1.4
	gnu/lib/libbfd/arch/sh3el/defs.mk: revision 1.8
	share/mk/bsd.own.mk: revision 1.495
	gnu/lib/libbfd/arch/sh3eb/bfd.h: revision 1.7
	x11/lib/OSmesa/Makefile: revision 1.20
	sys/dev/dkwedge/dkwedge_bsdlabel.c: revision 1.13
	gnu/dist/binutils/opcodes/configure.in: revision 1.2
	gnu/dist/binutils/opcodes/configure: revision 1.2
	gnu/usr.bin/send-pr/categories: revision 1.53
	etc/mtree/NetBSD.dist: revision 1.339
	sys/arch/README: revision 1.42
	lib/libc/gdtoa/Makefile.inc: revision 1.6
	usr.bin/crunch/crunchide/Makefile: revision 1.16
	sys/lkm/net/bsdcomp/Makefile: revision 1.5
	regress/sys/uvm/Makefile: revision 1.6
	etc/MAKEDEV.awk: revision 1.18
	x11/Xserver/GL/GLcore/Makefile.GLcore: revision 1.10
	sys/contrib/dev/ath/netbsd/Makefile.ath.inc: revision 1.14
	gnu/lib/libbfd/Makefile: revision 1.42
	doc/RESPONSIBLE: revision 1.78
Remove evbsh5.

Pull up following revision(s) (requested by mhitch in ticket #606):
	share/mk/sys.mk: revision 1.91-1.92 via patch
For now, make vax default to -O not -O2

-O1 on vax isn't quite enough, add -fgcse -fstrength-reduce  -fgcse-after-reload
too.  We can build again.

This addresses PR#34783.

-O1 on vax isn't quite enough, add -fgcse -fstrength-reduce -fgcse-after-reload
too.  We can build again.

For now, make vax default to -O not -O2

Remove support for NetBSD/{,evb}sh5.

delete "-std=gnu99" from c++ flags. From mrg.

Make sure to compile things at -O2 for mips64e[bl]

Move LDFLAGS to the CC section instead of the LD section.

like we do for -Wno-traditional, remove -Wstrict-prototypes and
-Wmissing-prototypes from C++ since those are always requird for C++.

add "S" to LINTFLAGS so that "inline" is accepted by lint

turn on -O2 for hppa.

Replace
	:M-[IDU]*
with
	:C/-([IDU])[  ]*/-\1/Wg:M-[IDU]*
so that arguments with whitespace after them work correctly.
Resolves [toolchain/18248] from Andrew White at Motorola.

Be consisent in this syntax; previously we'd sometimes we'd just have :M-[ID]*
which would ignore -Ufoo ...

(This needs :C///W support in make(1) that Simon Gerraty added for me :)

no-traditional is not a valid warning for c++ programs.

Use -O2 for ns32k now.

The order of items in .SUFFIXES is extremely important!
Revert part of previous and put `.s .S' before `.c' again.

(Fixes recent libkern lossage.)

Trim .SUFFIXES to only contain those used in this file.

Reorder some variables appropriately.

Add chmod a+x in .sh:, as per SUSv3.

Drop explicit medlow model and default to -O2 for sparc64.

Conditions for sh5 when setting DBG to be empty were deleted (moved)
in 1.72, but brought back accidentally in 1.73.  That was harmless,
anyway, as sh5 conditions in their new place get matched earlier.

be consistent: use ${.TARGET} for $@ and ${.PREFIX} for $*

whitespace cleanup

Default to -O2 for x86_64.

gcc for x86_64 should be safe enough these days to enable -O by default.

The sh5 toolchain works fine with -O2 now.

The traditional C preprocessor is fixed in GCC 3.3 now, so no need
to skip -traditional-cpp for HAVE_GCC3.

SH5 Makefile Magic.

Like on x86_64, don't use -O at all with the hppa compiler.

Pull up revision 1.67 (requested by thorpej in ticket #158):
Add case for armeb.

Don't use -traditional-cpp if HAVE_GCC3.

Add case for armeb.

No more arm32.

Reserve %g4 for backwards compatibility with the old ABI.

MACHINE_ARCH == "arm26" hasn't been valid for some time, so don't check
for it.

Remove -Wa,-Av9a hack from <sys.mk> and relegate it to the places that need
it (kernel and libc).

The current version of the gas assembler in the tree (2.11.2) already
defaults to generating object files for "-Av9 -64", supporting V9
instructions in ELF64 object format.  "-Av9a" is only needed for specific
parts of the NetBSD base sources, and not for all third-party code.

Use -O2 on PowerPC.

Make arm use -O2

Tell the assembler to accept UltraSPARC extensions on sparc64.

Add some extra flags for the x86-64 port. Disable -O in sys.mk
(which is where all ports set it); there seem to be some lurking
optimization bugs for the x86-64.

Switch arm26 to -O2.  Most of my kernels are compiled with this anyway,
and gcc's code generation should be the same as for arm32, so I think this
is safe.

Allow optimizer/debug flags to be changed on command line without
breaking everything.  Eg. use DBG="-O0 -g" when debugging something.

Pull up to last week's -current.

VAX should use -O2.  (why don't all ports?)

merge build rules for files ending .cc and .C
add support for C++ source files ending .cpp and .cxx

alpha seems to work with -O2...

add missing parts of the "mips"->"mipse?" transition

* Add support for MKDOC MKLINKLIB MKLINT MKMAN MKNLS MKOBJ MKPIC MKPICINSTALL
  MKPROFILE MKSHARE, which default to `yes' unless an equivalent NOxxx
  variable is defined.
* Add MKCATPAGES, which if "no" prevents catpages from being built or installed

Users should use "MKxxx=no" instead of "NOxxx=" in /etc/mk.conf, because
this allows easy override on the make(1) command line (e.g 'make MKSHARE=yes'
to override MKSHARE=no in /etc/mk.conf).

Back out previous backout of -O2 in mips.  Bug in strtod.c fixed via
Niklas Hallqvist.

_Don't_ use -O2 on mips, libc/stdlib/strtod.c breaks

Use -O2 on arm32.

Add -F to LINTFLAGS so that the full pathname is printed.

Use -O2 on mips.

Use -O2 on m68k.

Use -O2 on SPARC.

Use -O2 on i386.

Fix the .c.a brokenness with libbfd the "right" way - nuke everything but
*o from .ALLSRC.

Apply same .c.a workaround from alpha to arm32

The C++ compiler is "c++", not "g++"; that is more compatible with non-GCC
based C++ compilers.

.y.c <sys.mk> rule fixes.  Don't create a y.tab.h file unless asked for,
and use smarter creation of the header file.

Revert.  This broke too much in the build tree, and a better solution
needs to be created.  (The existing situation breaks a lot too.)

Fix .y rules to DTRT with possibly generated .h files.  See mail on
tech-toolchain for a very detailed explanation of this.  In short, add
a .y.h rule identical to the .y.c rule and make the .y.c rule rename
the .h file; also, clean the .h file in CLEANFILES.

Update from trunk: Fix .c.ln rule: use CPPFLAGS, not CFLAGS.

Fix .c.ln rule: use CPPFLAGS, not CFLAGS.

Pull up from trunk: add objective-c stuff, fix lint flags.

Move definitions of: NM, LORDER, TSORT from bsd.sys.mk to sys.mk
so they are more easly used by things like sys/libkern/Makefile.inc
(Besides, all the other tool definitions are in sys.mk)

add '-z' to LINTFLAGS, so that forward structure definions (for pointer
args, etc.), don't cause warnings if the structure isn't ever completely
defined.

add default rules for Objective C sources (`.m' suffix):
- sys.mk:
	add ${OBJC} and ${OBJCFLAGS} (equivalent to ${CC} and
	${CFLAGS} respectively), and ${COMPILE.m} and ${LINK.m} rules
- bsd.sys.mk:
	add .m, .m.o, and .m.ln rules (as per the C rules in sys.mk).
	it's here, rather than in sys.mk, because `.m' isn't
	exclusively used for Objective C files.
- bsd.lib.mk:
	add .m.o, .m.po, .m.so, and .m.ln rules (as per C)

Back out BUILDDIR and NOINSTALL changes.

Put include information in CPPFLAGS only (not in CFLAGS or MKDEPFLAGS).
Add note that CPPFLAGS is (or should be) passed to all programs that
pass input through cpp.

Remove unneeded .SUFFIXES lines, and move .s and .S closer to .o to give the
right preferences.

merge lite-2 changes (mostly SCCSids)

import 4.4BSD-Lite2 versions

Define an INSTALL make variable, and default it to "install".  Suggested
by Simon J. Gerraty <sjg@zen.void.oz.au>, PR #1458.

Add "-traditional-cpp" to the COMPILE.S command, so that assembler
syntax (specifically, use of # to indicate an immediate value) isn't
misinterpreted as a CPP directive.

Remove broken hack that attempted to do the same as the previous.

Fix COMPILE.s so the output file is named correctly.

Removed special handling of CPPFLAGS if DESTDIR is set.  Third party
makefiles may use DESTDIR for something completely different.

Default .s.o and .S.o left output in wrong file

don't include the .c.a rule on the alpha.  see p.r. 1488 for an
example of how make breaks when it's present.  In particular, it causes
build of the compiler toolchain to fail if present, which keeps
a complete system from building, on the alpha.

- In preparation for parallel makes fix the library rules so that they
  don't use a.out as intermediate output.
- Note in the README file that we cannot fix the xstr stuff.
  We still need to fix lex and yacc.
- Allow overriding of the variables in sys.mk

update lint rules, but keep linting disabled

from trunk.

Move the inclusion of bsd.own.mk to a better place.

Use CPPFLAGS (previously defined but unused) to yank in include files from
${DESTDIR}/usr/include.

.m4 files are sources too

RCSID frobbing, and minor sanity adjustments.

add back inclusion of bsd.own.mk

Added various COMPILE.? & LINK.? macros and changed suffix rules to use
them (Like Sun & GNU make).

Now that single suffix rules work, uncomment them.

Moved manpage formatting rules from sys.mk to bsd.man.mk.
Some programs (like ispell) use *.[0-8] for non-manpage things, and having
the manpage rules in sys.mk really messes things up.

oops; kill two extra spaces.

do the right thing if nroff doesn't work

Comment out single suffix rules (until I can get a debugger that works).

Added single suffix implicit rules.  Rules for *.c & *.f files are required
by POSIX.2 --- I added the others (*.p, *.l, *.y, etc.) to be consistant
with all the other default rules.

Add .S.o rule

Makefile: Use `install -d' to create directory.
bsd.dep.mk: Add defaults for `beforedepend' and `afterdepend'.
sys.mk: 8 tabs --> space

Move man rules to sys.mk.
Use bsd.subdir.mk for building subdirectories.
Don't forget to remove x.cc and x.C when using SHAREDSTRINGS.
Make ${PROG} always depend on ${LIBCRT0}.
Don't duplicate code between cleandir and clean.
Move tags target to bsd.dep.mk.
Move STRIP to bsd.own.mk.
Remove redundant definitions of {BIN,MAN}{DIR,GRP,OWN,MODE}.
Move {LIB,DOC}{DIR,GRP,OWN,MODE} defaults to bsd.own.mk.

Combine some suffix rules.

Add suffix rules for C++ files.

Add global CXX and CXXFLAGS for compiling C++ programs.  (These names are used
by some GNU stuff; if you have a better suggestion, please tell me.)

various changes to default make rules.  from patch 101.
note that only the "sensible" changes in patch 101 are implemented.

initial import of 386bsd-0.1 sources

Initial revision