Commit Graph

226 Commits

Author SHA1 Message Date
thorpej e6d41917cc Newlines in string literals are not supported by ANSI C, and some
preprocessors complain bitterly when they are encountered.

For now, terminate all lines with \n (and make each line its own
string literal).  The author of this code can figure out how to
do the sed trick another way.
2001-12-23 22:41:27 +00:00
atatat 65ace4b729 Properly use $TMPDIR or /tmp for the temporary config file stuff,
instead of ".".  Read-only source trees are "good".
2001-12-21 19:09:43 +00:00
atatat eb0a09d704 Teach config how to store the config in a .h file such that it can be
embedded in the resulting kernel binary.
2001-12-17 15:39:43 +00:00
wiz 2f3bcd9810 Comment out now-empty .Rs/.Re. 2001-12-14 12:32:50 +00:00
gmcgarry 367be6a153 Sprinkle around the term "machine description file", since this is
the term used in config(9), autoconf(9) and in Torek's paper.

Also remove (comment out) reference to the 4.4BSD config docs,
since they are not installed in usr/share/doc/smm/02.config and
possibly aren't relevant any longer.  From comments made on
tech-kern by jhawk and tls.
2001-12-13 19:47:19 +00:00
atatat 6dce2e2bec Make the kernel depend on `vers.o' instead of `newvers'. This
eliminates the need to unconditionally rebuild vers.o and subsequently
to relink the kernel every time you type make, even if you just
finished one.
2001-12-09 16:04:03 +00:00
wiz 6a71504b07 Add a ',' between two Xrefs. 2001-12-06 10:57:20 +00:00
atatat 42e4429bad Since defopt has been totally eradicated from the tree and is
considered "bad form", let's tell people they shouldn't use it.
2001-12-05 23:42:37 +00:00
jhawk 22fc75e3b6 Xr config.9 2001-12-05 20:58:04 +00:00
atatat 73aae2a161 Stuff the target names (the kernel names) into a make variable so that
ports that build "alternate" versions of the same target can clean up
properly.
2001-12-04 17:53:40 +00:00
lukem 0e59a7f124 revert previous revert previous. now back to rev 1.27 behaviour:
- Don't bother emitting "#define NFOO 0" for defflag foo (whether or
  not "options FOO" was given); it's really useful as is.

  Until I recently started converting stuff from defopt to defflag,
  nothing used defflag, and because defopt doesn't provide NFOO for FOO,
  the NFOO behaviour shouldn't be missed. The NFOO stuff also added
  a lot of #define namespace pollution that we'd rather avoid.

- Remove a couple of unused vars
2001-11-28 09:02:16 +00:00
lukem 7b844b40dc ensure that enabled defflag options get a default value of "1" too 2001-11-28 05:03:31 +00:00
lukem 49379a0a84 revert previous (instead i'll work out why defflag isn't DTRT with Nxxx) 2001-11-28 03:23:35 +00:00
lukem cbe4dece17 - Don't bother emitting "#define NFOO 0" for defflag foo (whether or
not foo was provided); it's unnecessary #define namespace pollution
- Remove a couple of unused vars
2001-11-28 02:47:42 +00:00
atatat 3b7ac0bb5f Stuff the "swapxxx.o" object name into a make variable that can be
used in build rules.
2001-11-27 14:04:05 +00:00
thorpej b2cb6c26ef Don't set TARGET_MACHINE, set MACHINE, instead. MACHINE is consistent
with what other Makefiles use, and as a result, make cross-compiling
kernels a little easier.
2001-10-23 18:56:42 +00:00
tv 47ab6193ba Make this a little more portable to other NetBSD versions without
MAKE_BOOTSTRAP -- use our hand-rolled [gs]etprogname() at all times (er,
actually, the single time we need it...).
2001-10-02 21:31:01 +00:00
simonb c6d44b5eb4 ANSIfy the output of config(8). 2001-10-01 04:22:00 +00:00
tv 6042281606 Don't override the definition of LEX. 2001-09-25 01:40:09 +00:00
enami 54d0f0cc4b Fix off-by-one error; root on sd0i shoule be rejected if maxpartition is 8. 2001-09-11 05:11:59 +00:00
tv dba5d44670 Add hooks to allow toolchain bits to be reachover-built at the top level. 2001-08-14 10:18:26 +00:00
itojun a42562635c sync .Fl with reality 2001-07-03 08:52:00 +00:00
gmcgarry 32fc55d6d2 Don't pack locators. Introduce a new commandline option to
restore the previous behaviour of packing locators.

Results in a 2.5KB increase in size for the current i386 GENERIC
kernel.  My custom kernel resulted in a 12-byte increase.
2001-07-01 02:46:47 +00:00
fredette de800ed660 Added support for listing multiple additional names on
a `machine' line.  For each of these subarches, its
arch/${SUBARCH}/conf/files.${SUBARCH} is included, and
a symlink ${SUBARCH} -> arch/${SUBARCH}/include is made.
This will for a "richer structure" and easier code
sharing under sys/arch.
2001-06-08 12:47:06 +00:00
minoura 81e8dc3cac Correct a typo.
Pointed out by MAEKAWA Masahide <maekawa@kbug.gr.jp>.
2001-05-18 07:48:07 +00:00
cgd c2bdafab79 use getprogname() 2001-02-20 23:51:59 +00:00
bjh21 a9e45dc961 Put both .S and .s files in SFILES. Fixes bin/12051. Patch from Richard
Earnshaw via Chris Gilbert.

The right solution here is to have every port use .S, but that might take a
while...
2001-01-31 00:15:40 +00:00
thorpej ed7ec7c142 Make this work again for the case where there are devices
configured, but none of those devices happen to have locators.
2001-01-18 07:09:47 +00:00
cgd fdfcf1962d labels (even default:) _must_ be followed by statements! usr.bin/fgen/fgen.l 2001-01-16 02:43:37 +00:00
lukem 1d0a6f62b8 revert 1.51 and 1.52 (_KERNEL_foo stuff), due to loud objections from
a couple of people.  we'll just have to go back to adding this manually
to our config files...
2000-12-14 22:14:45 +00:00
lukem ea03d7ba8e change KERNEL_foo -> _KERNEL_foo, as suggested by matt green 2000-12-12 17:49:20 +00:00
lukem f125ced5a1 Add option `KERNEL_foo', where `foo' is the basename of the build directory.
This will get added to the Makefile's IDENT= as -DKERNEL_foo.
2000-12-12 08:46:02 +00:00
matt a2877b394c Don't emit MACHINE/MACHINE_ARCH. Instead emit a TARGET_MACHINE=machine
instead.
2000-12-03 07:06:15 +00:00
matt 5cde4cf735 If a port-specific Makefile.<port> doesn't exist, try a arch-specific
Makefile.<arch>.
Emit MACHINE=<port> and MACHINE_ARCH=<arch> to the emitted Makefile.
2000-12-02 23:01:54 +00:00
cgd a1a91fd28f if there are no locators or parent, don't output those arrays. (Sometimes
you really want no devices whatsoever.  In that case, you get unused
static var warnings from the compile.)  Reviewed by sommerfeld.
2000-11-18 00:51:29 +00:00
is d8302e2d73 More format string cleanups by sommerfeld. 2000-10-11 20:23:46 +00:00
bjh21 23dc98ae44 Don't try to use yydebug when we're building on a foreign system. Not all
yaccs provide it (notably GNU Bison 1.28).
Fixes the final part of PR#9112 (building under GNU/Linux).
2000-10-08 11:33:40 +00:00
cgd 1e69382368 clean up dev_t-related portability defines (and actually #define dev_t now).
Make major() et al match current NetBSD implementation.  In reality,
the code here in config which plays with these bits should be cleaned up
to treat major and minor separately.  (They only reason they're being mashed
together is storage convenience.)
2000-10-02 19:59:42 +00:00
cgd da723c733f include copyright notice in binary even if bootstrapping. 2000-10-02 19:57:23 +00:00
cgd 506c287193 bump ALIGNBYTES up to 7 -- that way this'll have more of a hope
of working on non-NetBSD systems with 64-bit pointers.  (affects only
the MAKE_BOOTSTRAP case, and then only if BSD not defied.)
2000-10-02 19:51:07 +00:00
cgd 463d864ef0 nuke __P, and convert all functions to use ANSI-style declarations.
Also, do a few trivial KNF cleanups (e.g. newline at start of fn if no
locals).  Verified to have no effect via diff on new and old compiled
binaries.
2000-10-02 19:48:34 +00:00
cgd a755a0d943 remove vestiges of checks for __STDC__ 2000-10-02 18:59:03 +00:00
cgd c026189f76 do that previous differently. only allow redefinitions if: in different
file, make command specified, and no flags or attrs-which-cause-inclusion
are spec'd.  The notion is, if you change either of the last 2, it will
probably have very undesirable results, so only allow the make command to
be changed.  override by clobbering the make command in the previous entry.
also, fix a bug where line number of original entry would get clobbered on
dup entry, so that if you had multiple dups the later ones would get bogus
initial definition info.
2000-06-09 05:06:12 +00:00
eeh e5cf411171 Allow MD files files to override settings in MI files files. (From Chris Torek). 2000-06-08 21:22:55 +00:00
hubertf 13afcc6b33 If maxusers is out of bounds, show the allowed bounts in the error msg. 2000-05-09 00:34:58 +00:00
simonb fd4ede242b Don't declare 'extern opt*' getopt variables. 2000-04-14 06:26:52 +00:00
enami 1b1b19c5c5 Emit `.MAIN: all' to force the target `all' as default target. 2000-02-01 05:13:17 +00:00
enami 0d45936ccf Please pay an attention to the existing code and not to introduce
different coding style.
2000-01-25 01:16:00 +00:00
hubertf 764307a5a1 Keeping my kernel config files under RCS control, I always wished to
have a way to embed the revision number into the kernel's "uname -v"
output. The patch below does this, by generating a new keyword "ident"
that can be followed by any string, e.g.

        ident   "NOON-$Revision$"

will lead to
                                           vvvvvvvvvvvvvvvvvvvvvv
        char version[] =     "NetBSD 1.4P (NOON-$Revision: 1.21 $) #37: Thu Jan
20 02:01:23 MET 2000\n    feyrer@noon:/usr/cvs.local/src-current/sys/arch/i386/c
ompile/NOON\n";

This will lead to a version of "MYMACHINE-$Revision$" instead of the
kernel config file name. If "ident" is not present, the current behaviour
of using the kernel config file's name as identifier is used.

Implement by writing the ident to a file ("ident") in the compile dir,
which newvers.sh will pick up for generating the ident.
2000-01-23 23:37:42 +00:00
thorpej aa1aaac03e Allow abolute pathnames as prefixes. 2000-01-20 00:08:08 +00:00