Commit Graph

113068 Commits

Author SHA1 Message Date
fvdl
b2a54b691e Back out previous for now. I already suspected part of it was wrong,
and there has been one report of problems in the mean time.
2003-05-19 09:26:15 +00:00
scw
cd2d76d833 Fix an obvious pasto. 2003-05-19 08:10:03 +00:00
lukem
76a3147d38 Remove dependency on ${RPCGEN}:
*	Unnecessarily causes lib/librpcsvc (etc) to be rebuilt every time
	rpcgen is updated.
    *	No other "generated" file (.l, .y, ...) depends upon its tool
    	like this
    *	As <bsd.own.mk> wasn't being pulled in, the tools/ version
    	wasn't being used, so a lot of times the dependency was wrong.

Fixes [toolchain/11568] by Bernd Ernesti.
2003-05-19 07:19:10 +00:00
dyoung
9d9a86120a Make ppsratecheck conform with its man page, which says, "If maxpps
is set to 0, the function will always return 0 (no packets/events
are permitted)." Before this patch, ppsratecheck returned 1 once
a second when maxpps was 0.
2003-05-19 03:23:37 +00:00
seb
e18fdad4d9 Note str{vis,unvis} support in db(1). 2003-05-19 00:52:57 +00:00
seb
8875d36598 Bump date for last change. 2003-05-19 00:28:37 +00:00
seb
c4cd748ebd Add option -S to specify strvis(3) encoding of keys and/or values
when reading a database.  Add options -T and -X to tweak encoding.

Add option -U to specify strunvis(3) decoding of keys and/or values
when writing a database or deleting entries from a database.

Kill TODO item 'add option to strunvis(3) key (& value) from infile
?': unvis(1) can be used to achieve that.

Approved by lukem@. And thanks to him for suggestions.
2003-05-19 00:27:06 +00:00
christos
56a0a03629 From Martin Blapp mb at imp dot ch, should fix PR/20768
Remove the special treatment off non-blocking mode in
xdrrec_eof(). This change broke seriously reading serialized
xdr-records. A end of a request could sometimes not
be recognized and the socket got closed too early.

Initialize in_reclen and in_received in a second case
if we get more outstanding records.
2003-05-18 23:59:28 +00:00
kristerw
8e46e7bae0 Prevent use of stale pointer for determining return type of ({ })
expressions.

Details:
lint did fail on constructs like

   struct foo *x;
   x = ({ struct foo *y; /* do stuff */; y;});

since it gave the whole ({ }) the same structure representing the
type as y, but that structure is reclaimed when y goes out of scope.
2003-05-18 23:30:00 +00:00
martin
a47f54c6b8 Make kernels with options KGDB at least compile.
Still needs some work to make it working.
2003-05-18 22:11:31 +00:00
martin
ce380782ec My reading of the gdb code seems to indicate it expects 125 registers
from sparc64, not 72 (sparc value). Bump kgdb buffer size a bit too.
2003-05-18 21:00:57 +00:00
martin
6b8eb7fd4f Remove reference to non-existing file 2003-05-18 20:56:54 +00:00
dsl
09746cd0ad Stop double-inclusion being an error.
Adjust whitespace and remove names from prototypes
2003-05-18 20:40:09 +00:00
dsl
1d6ff8632d White-space, buffer-overrun and long-line police,
Correct code that selects one of the available disks.
2003-05-18 20:04:21 +00:00
uwe
c7e8d1f928 Oops, correct platids for 680/690 with European English (ABB)
keyboard.  We still use hpcboot's "Hungarian" entry for them.
2003-05-18 19:06:11 +00:00
dsl
33180ff44d Use MD menus and msgs for mbr code.
Remove a couple of unused variables.
Use actual raw partition when looking for disklabel in sparc code
2003-05-18 18:54:05 +00:00
martin
c189434481 Make kernels with options KGDB compile. (Didn't test kgdb though) 2003-05-18 16:45:21 +00:00
martin
f799459a60 List more supported systems. 2003-05-18 16:29:57 +00:00
fvdl
385c785fd2 Inialize sc_tty->t_dev early for serial console, so that cn_check_magic
will match in the case of a break before the console was opened for the
first time. Makes BREAKing into DDB work (again) as soon as interrupts
are enabled.
2003-05-18 15:10:08 +00:00
yamt
ab2238cfad make is_sequential a callback in order to achieve better lfs write clustering.
since lfs always rewrite blocks into the new segment,
current on-disk place of the block doesn't affect to write clustering.

ok'ed by Konrad Schroder.
2003-05-18 12:59:05 +00:00
lukem
91618e1c90 -N also prevents NUL termination of value 2003-05-18 12:26:18 +00:00
lukem
20b9064eac Don't force "make -j 1" any more, since David Laight appears to have
fixed the parallel-make issues in crunchgen generated Makefiles.
2003-05-18 11:25:33 +00:00
lukem
0c4720f396 For normal builds, MKOBJDIRS default to no' (not yes').
It's only build.sh that sets this to `yes' by default.
Noted by OGAWA Takaya in [toolchain/19823]
2003-05-18 11:18:02 +00:00
lukem
6019a973a8 * Display the start time again just before the end time, to make it
easier to work out how long things took.  Suggested by Alistair Crooks.
* Explicitly display the makewrapper path
2003-05-18 10:57:11 +00:00
dsl
1473d232c3 Don't do __stopwin() if it is already stopped.
Allows error messages from programs like sysinst() to be seen.
(There is an explicit stopwin() before the error message is output,
and a second in the atexit() called cleanup() routine)
2003-05-18 10:02:07 +00:00
yamt
cc3f08f723 define HAVE_XDR_MOUNTRES3 in pre-generated config.h as we have mountres3 now.
this makes amd buildable for archs without shlibs support like hpcsh.
(i've tested i386->hpcsh cross-compile)

pointed by Torsten Harenberg on tech-hpcsh
and analyzed by Valeriy E. Ushakov.
2003-05-18 09:44:28 +00:00
wiz
11d31abb7e Use standard section header; add nokeep to table to avoid page/line break
in table. From jmc@openbsd.
2003-05-18 09:30:08 +00:00
wiz
be82955f10 New sentence, new line; fix xrefs; sort SEE ALSO; punctuation nits;
uppercase beginning of sentence; bump date for last two.
2003-05-18 09:10:51 +00:00
wiz
a5016fab1f New sentence, new line. Bump date for last. Add missing "the". 2003-05-18 09:04:52 +00:00
lukem
6087671da1 * Always include <bsd.endian.mk>, and include it much earlier so that
it's available for the MD etc.${MACHINE}/Makefile.inc
* Set PWD_MKDB_ENDIAN to the cap_mkdb(8) endian flag, rather than
  overriding TARGET_ENDIANNESS.
2003-05-18 08:28:05 +00:00
lukem
f5dd9322ee Set CAP_MKDB_ENDIAN to the cap_mkdb(8) endian flag, rather than
overriding TARGET_ENDIANNESS.
2003-05-18 08:20:34 +00:00
lukem
4ae3ede358 Only determine TARGET_ENDIANNESS if it's not already defined. 2003-05-18 08:19:03 +00:00
lukem
65e312f75f Add ${TOOL_DB}; db(1) as a (host) tool.
Note: this is the first tool using a "TOOL_" prefix in the make(1) variable;
other similar "non-standard" variable names will be converted in the future.
2003-05-18 08:09:25 +00:00
lukem
fc80f6c55f Add db(1) as a host tool 2003-05-18 08:05:32 +00:00
dsl
b1389a6fef md menus and messages are taken from the i386 directory.
mbr stuff is now mi
2003-05-18 08:00:38 +00:00
dsl
04a7d9497e Use same code as i386 2003-05-18 07:58:10 +00:00
lukem
826a14607a Now that <bsd.prog.mk> DTRT if HOSTPROG is defined (i.e, it is a no-op),
there's no need to special-case .include-ing it.
2003-05-18 07:57:31 +00:00
dsl
2de0c528b3 Remove some unused definitions 2003-05-18 07:57:28 +00:00
dsl
949c967305 Pull md menus and msgs from i386 port.
Enable BOOTSEL stuff from mbr.c
2003-05-18 07:56:30 +00:00
dsl
9a5b2e3d2a Use snprintf just in case target_expand() generates some long strings.
#if the bootmodel stuff so that the amd64 port can use the same file
2003-05-18 07:52:57 +00:00
dsl
633389164c Fix typo in previous 2003-05-18 07:44:41 +00:00
lukem
03a26b2c80 make this a no-op if HOSTPROG is defined 2003-05-18 07:36:23 +00:00
lukem
cd46c56ae6 missing rcsid 2003-05-18 06:45:14 +00:00
thorpej
02ad493d76 Fail icpioctl() if securelevel > 1. 2003-05-18 06:18:25 +00:00
tsutsui
30f71213c8 Remove unused file since asc.c switched to MI ncr53c9x. 2003-05-18 04:28:33 +00:00
itojun
f5878476d9 simplify by strlcpy 2003-05-18 02:11:13 +00:00
itojun
8138cbd1bf simplify more 2003-05-18 02:07:20 +00:00
itojun
ebf587df1a simplify by strlcpy/cat 2003-05-18 02:06:28 +00:00
thorpej
55f89430cf Decrease the verbosity level unless VERBOSE_INIT_ARM is defiend. 2003-05-17 23:46:59 +00:00
itojun
262bdee727 do not hardcode buffer size 2003-05-17 23:16:47 +00:00