bsd.own.mk: rename GCC_NO_* to CC_WNO_*

Rename compiler-warning-disable variables from
GCC_NO_warning
to
CC_WNO_warning
where warning is the full warning name as used by the compiler.

GCC_NO_IMPLICIT_FALLTHRU is CC_WNO_IMPLICIT_FALLTHROUGH

Using the convention CC_compilerflag, where compilerflag
is based on the full compiler flag name.

introduce some common variables for use in GCC warning disables:

GCC_NO_FORMAT_TRUNCATION -Wno-format-truncation (GCC 7/8)
GCC_NO_STRINGOP_TRUNCATION -Wno-stringop-truncation (GCC 8)
GCC_NO_STRINGOP_OVERFLOW -Wno-stringop-overflow (GCC 8)
GCC_NO_CAST_FUNCTION_TYPE -Wno-cast-function-type (GCC 8)

use these to turn off warnings for most GCC-8 complaints. many
of these are false positives, most of the real bugs are already
commited, or are yet to come.


we plan to introduce versions of (some?) of these that use the
"-Wno-error=" form, which still displays the warnings but does
not make it an error, and all of the above will be re-considered
as either being "fix me" (warning still displayed) or "warning
is wrong."

branches: 1.14.42;
document non-literal format strings.

Enable WARNS=4 by default for usr.bin, except for:
awk bdes checknr compile_et error gss hxtool kgetcred kinit
klist ldd less lex locale login m4 man menuc mk_cmds
mklocale msgc openssl rpcgen rpcinfo sdiff spell ssh
string2key telnet tn3270 verify_krb5_conf xlint

branches: 1.12.26;
WFORMAT is no more...

protect libutil with hostprog

More programs using efun.

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.

Make hexdump a host tool.

WARNS=2

More format string cleanup by sommerfeld.

branches: 1.5.4; 1.5.12;
merge lite-2.

RCS ID police

specify man pages the new way.

Add RCS indentifiers.

branches: 1.1.1;
Initial revision

"dcl" is not "dc1".

Fixes PR#47547

branches: 1.13.6; 1.13.12;
remove agitation that od(1) was deprecated -- it is still POSIX

de-__P, ANSIfy prototypes

Overhaul how `build.sh tools' are used:

* Rename "config.h" to "nbtool_config.h" and
HAVE_CONFIG_H to HAVE_NBTOOL_CONFIG_H.
This makes in more obvious in the source when we're using
tools/compat/config.h versus "standard autoconf" config.h

* Consistently move the inclusion of nbtool_config.h to before
<sys/cdefs.h> so that the former can provide __RCSID() (et al),
and there's no need to protect those macros any more.

These changes should make it easier to "tool-ify" a program by adding:
#if HAVE_NBTOOL_CONFIG_H
#include "nbtool_config.h"
#endif
to the top of the source files (for the general case).

add
#if HAVE_CONFIG_H
#include "config.h"
#endif
to get this to build under solaris-2.6/sparc with gcc-2.95.1 as part of
the NetBSD cross toolchain build. Fixes PR23211. Patch approved by
Alistair Crooks.

Move UCB-licensed code from 4-clause to 3-clause licence.

Patches provided by Joel Baker in PR 22365, verified by myself.

More tests if __RCSID is defined.

WARNS=2

WARNSify, fix .Nm usage, deprecate register

merge lite-2.

add explicit return types to many functions
pull global variable declarations and function prototypes into hexdump.h
add #includes where necessary for prototypes
replace index() with strchr(), const-poison static string vars, etc.

RCS ID police

Add RCS identifiers.

branches: 1.1.1;
Initial revision

better check st_size == 0, than st_dev == -1

PR/57851: Ricardo Branco: stat(2) does not return correct size information
for kernfs/procfs files. Minor KNF.

Don't try to use stdin after clobbering it with a failed freopen().
Prevents an extra "Bad file descriptor" message when trying to hexdump
a single nonexistent file.

The intended behavior seems to have been to read from stdin if there
was one filename given and it wasn't valid. But this seems like a bad
idea, so prevent that case instead of hacking it up so it works.

Don't use one function for two different things. Two functions is the
ticket.

use "static"

don't read random garbage from the stack

branches: 1.21.8; 1.21.14;
Integer values, no matter what their width, are cast to int64_t or
uint64_t for printing. Because of this, parse.c:rewrite() should use
PRI[diouxX]64 as the printf format; the previous revision which used
"lld" and friends was a temporary measure. We assume that PRId64 ends
with 'd', PRIx64 ends with 'x', etc.

In display.c:display(), change some u_intNN_t types to C99 standard
uintNN_t types, and consistently cast to [u]int64_t when printing
integer values (even in two cases where it's not really necessary but it
makes the code look more consistent).

More programs using efun.

de-__P, ANSIfy prototypes

Overhaul how `build.sh tools' are used:

* Rename "config.h" to "nbtool_config.h" and
HAVE_CONFIG_H to HAVE_NBTOOL_CONFIG_H.
This makes in more obvious in the source when we're using
tools/compat/config.h versus "standard autoconf" config.h

* Consistently move the inclusion of nbtool_config.h to before
<sys/cdefs.h> so that the former can provide __RCSID() (et al),
and there's no need to protect those macros any more.

These changes should make it easier to "tool-ify" a program by adding:
#if HAVE_NBTOOL_CONFIG_H
#include "nbtool_config.h"
#endif
to the top of the source files (for the general case).

add
#if HAVE_CONFIG_H
#include "config.h"
#endif
to get this to build under solaris-2.6/sparc with gcc-2.95.1 as part of
the NetBSD cross toolchain build. Fixes PR23211. Patch approved by
Alistair Crooks.

include <inttypes.h> to get uint64_t defined

This is a host tool: use {u,}int64_t instead of {u_,}quad_t.

Move UCB-licensed code from 4-clause to 3-clause licence.

Patches provided by Joel Baker in PR 22365, verified by myself.

Two more __RCSID checks that I missed.

WARNS=2

Enable support for printing 8-byte integers. For some reason, most of the
code for this was present, but disabled.
This is required for POSIX compliance on platforms with 8-byte longs.

PR/9195: Tanaka Akira: od does not display short files containing nulls. Fix
from Andrew Brown.

printf() pedant - even if we are sure that there's no %-formatted in a
varaiable, don't pass variable alone to printf().
From: kris@freebsd.org (FreeBSD-current usr.bin/hexdump/display.c 1.4 -> 1.6)

branches: 1.8.4;
Since our gcc doesn't warn about NULL format strings anymore, we can
fix the incorrect err(1, "%s", "") et al.
Closes PR bin/7592 by cgd.

branches: 1.7.8;
WARNSify, fix .Nm usage, deprecate register

merge lite-2.

add explicit return types to many functions
pull global variable declarations and function prototypes into hexdump.h
add #includes where necessary for prototypes
replace index() with strchr(), const-poison static string vars, etc.

RCS ID police

Include appropriate header files to bring prototypes into scope.
Removed explicit errno declarations.

Add RCS identifiers.

branches: 1.1.1;
Initial revision

Bug Hunting:
improve wording

From Bug Hunting:
- sort options in `SYNOPSIS' and `DESCRIPTION' sections;
- correct / improve macro usage;
- synchronize argument name to `-s' option;
- improve punctuation / wording;
- augment `SEE ALSO' section;
- bump date.

While here:
Use Sq for single characters instead of Dq.
Use Aq instead of \*[Lt]...\*[Gt].
Mark up NULL with Dv.

Remove unnecessary Bk/Ek pairs from SYNOPSIS.
No effective change except where I used the opportunity to sort options
and/or option descriptions.

branches: 1.21.6;
Replace "\t" with Ta in table.
Fixes output with mandoc. Noted by njoly.

DEL is 0x7F, not 0xFF. From FreeBSD via OpenBSD.

Do not use .Xo/.Xc to workaround ancient groff limits.

Move UCB-licensed code from 4-clause to 3-clause licence.

Patches provided by Joel Baker in PR 22365, verified by myself.

Fixes to make these man page look nice with groff-1.19.

.Nm does not need a dummy argument ("") before punctuation or
for correct formatting of the SYNOPSIS any longer.

Generate <>& symbolically. I'm avoiding .../dist/... directories for now.

Mention that eight-byte integers work.

POSIX.2 doesn't specify hexdump, so we can't claim our hexdump is compatible
with it.

Whitespace cleanup.

Improve usage of `comprise' according to Webster and misc/10893
by Jim Bernard.

Clean up SYNOPSIS formatting.

Added -C option for hexadecimal+ASCII display. [PR/5017]
Fixed minor inconsistency in option descriptions in the man page.

WARNSify, fix .Nm usage, deprecate register

merge lite-2.

kill xref to nonexistent adb(1)

RCS ID police

fix the the typos

Fix spelling errors.

Add RCS indentifiers.

branches: 1.1.1;
Initial revision

Don't use one function for two different things. Two functions is the
ticket.

From Bug Hunting:
- de-capitalize ``usage'' in `usage' message;
- synchronize `usage' message with man pages.

branches: 1.17.6;
properly detect that we aren't od(1) and act as hexdump. fixes hexdump,
which also broke startx (second time this week!)

from mlelstv.

revert and fix the usage to be consistent with other programs.

remove agitation that od(1) was deprecated -- it is still POSIX

Remove the \n and tabs from the __COPYRIGHT() strings.
Tweak to use a consistent format.

branches: 1.13.22;
de-__P, ANSIfy prototypes

Overhaul how `build.sh tools' are used:

* Rename "config.h" to "nbtool_config.h" and
HAVE_CONFIG_H to HAVE_NBTOOL_CONFIG_H.
This makes in more obvious in the source when we're using
tools/compat/config.h versus "standard autoconf" config.h

* Consistently move the inclusion of nbtool_config.h to before
<sys/cdefs.h> so that the former can provide __RCSID() (et al),
and there's no need to protect those macros any more.

These changes should make it easier to "tool-ify" a program by adding:
#if HAVE_NBTOOL_CONFIG_H
#include "nbtool_config.h"
#endif
to the top of the source files (for the general case).

add
#if HAVE_CONFIG_H
#include "config.h"
#endif
to get this to build under solaris-2.6/sparc with gcc-2.95.1 as part of
the NetBSD cross toolchain build. Fixes PR23211. Patch approved by
Alistair Crooks.

Move UCB-licensed code from 4-clause to 3-clause licence.

Patches provided by Joel Baker in PR 22365, verified by myself.

Do the defined(__RCSID) song and dance to enable cross compilation.

setlocale(LC_ALL, "") on startup.

branches: 1.7.14;
WARNSify, fix .Nm usage, deprecate register

merge lite-2.

add explicit return types to many functions
pull global variable declarations and function prototypes into hexdump.h
add #includes where necessary for prototypes
replace index() with strchr(), const-poison static string vars, etc.

RCS ID police

Include appropriate header files to bring prototypes into scope.
Removed explicit errno declarations.

Add RCS identifiers.

branches: 1.1.1;
Initial revision

Don't use one function for two different things. Two functions is the
ticket.

use "static"

static + __dead

revert and fix the usage to be consistent with other programs.

remove agitation that od(1) was deprecated -- it is still POSIX

More programs using efun.

de-__P, ANSIfy prototypes

Move UCB-licensed code from 4-clause to 3-clause licence.

Patches provided by Joel Baker in PR 22365, verified by myself.

WARNS=2

constify

merge lite-2.

add explicit return types to many functions
pull global variable declarations and function prototypes into hexdump.h
add #includes where necessary for prototypes
replace index() with strchr(), const-poison static string vars, etc.

RCS ID police

Add RCS identifiers.

branches: 1.1.1;
Initial revision

revert and fix the usage to be consistent with other programs.

de-__P, ANSIfy prototypes

Overhaul how `build.sh tools' are used:

* Rename "config.h" to "nbtool_config.h" and
HAVE_CONFIG_H to HAVE_NBTOOL_CONFIG_H.
This makes in more obvious in the source when we're using
tools/compat/config.h versus "standard autoconf" config.h

* Consistently move the inclusion of nbtool_config.h to before
<sys/cdefs.h> so that the former can provide __RCSID() (et al),
and there's no need to protect those macros any more.

These changes should make it easier to "tool-ify" a program by adding:
#if HAVE_NBTOOL_CONFIG_H
#include "nbtool_config.h"
#endif
to the top of the source files (for the general case).

add
#if HAVE_CONFIG_H
#include "config.h"
#endif
to get this to build under solaris-2.6/sparc with gcc-2.95.1 as part of
the NetBSD cross toolchain build. Fixes PR23211. Patch approved by
Alistair Crooks.

Move UCB-licensed code from 4-clause to 3-clause licence.

Patches provided by Joel Baker in PR 22365, verified by myself.

Two more __RCSID checks that I missed.

Show -C option in usage. PR/5266

Added -C option for hexadecimal+ASCII display. [PR/5017]
Fixed minor inconsistency in option descriptions in the man page.

WARNSify, fix .Nm usage, deprecate register

merge lite-2.

add explicit return types to many functions
pull global variable declarations and function prototypes into hexdump.h
add #includes where necessary for prototypes
replace index() with strchr(), const-poison static string vars, etc.

RCS ID police

Add RCS identifiers.

branches: 1.1.1;
Initial revision

From Bug Hunting:
- correct / improve macro usage;
- (re-)add an actual description in `DESCRIPTION' section;
- synchronize argument name to `-j' option;
- improve wording;
- bump date.

Remove unnecessary Bk/Ek pairs from SYNOPSIS.
No effective change except where I used the opportunity to sort options
and/or option descriptions.

branches: 1.25.6;
Bump date for previous.

remove agitation that od(1) was deprecated -- it is still POSIX

Support -the "-A addressformat" flag for od(1), from FreeBSD.
(I found a script which tries to figure out the byteorder of a box
by "od -An | grep", but it is POSIX anyway.)
There is a little behavioural change: The whitespace without any arguments
is like "-to2" now; before it was custom. It's not worse imho.

branches: 1.22.2;
Convert TNF licenses to new 2 clause variant

branches: 1.21.32;
.Nm does not need a dummy argument ("") before punctuation or
for correct formatting of the SYNOPSIS any longer.

Correct typos (from OpenBSD).

Wrong word; from jmc@prioris.mini.pw.edu.pl via openbsd.

New sentence, new line.

Generate <>& symbolically. I'm avoiding .../dist/... directories for now.

Further POSIX progress: The C, S, I and L modifiers now behave correctly,
specifying output in units of a char, short, int and long (as defined by the
host system) respectively. This made the POSIX -t code more than complicated
enough to merit its own function, so I did that.

Further POSIXification: od now supports -N, which is equivalent to hexdump's
-n (format only COUNT bytes of input).

Add a -j option to od, which does the same as the -s option to
hexdump (skipping some of the input). This brings our od slightly
closer to POSIX.2 conformance.

Hmm...apparently I left out a word.

Lowercase name in xref.

fix fix typo from @@@

new man page from @@@

PR/12147: TheMan: add -t option to od.

Clean up SYNOPSIS formatting.

Add a real date.

everything has bugs; either document them or don't bother with a BUGS section

WARNSify, fix .Nm usage, deprecate register

merge lite-2.

RCS ID police

Add RCS indentifiers.

branches: 1.1.1;
Initial revision

revert and fix the usage to be consistent with other programs.

PR 44156: od -? prints usage for "hexdump" rather than "od"

remove agitation that od(1) was deprecated -- it is still POSIX

Support -the "-A addressformat" flag for od(1), from FreeBSD.
(I found a script which tries to figure out the byteorder of a box
by "od -An | grep", but it is POSIX anyway.)
There is a little behavioural change: The whitespace without any arguments
is like "-to2" now; before it was custom. It's not worse imho.

branches: 1.24.18;
More programs using efun.

de-__P, ANSIfy prototypes

appease gcc -Wuninitialized

Add (unsigned char) cast to ctype functions

Overhaul how `build.sh tools' are used:

* Rename "config.h" to "nbtool_config.h" and
HAVE_CONFIG_H to HAVE_NBTOOL_CONFIG_H.
This makes in more obvious in the source when we're using
tools/compat/config.h versus "standard autoconf" config.h

* Consistently move the inclusion of nbtool_config.h to before
<sys/cdefs.h> so that the former can provide __RCSID() (et al),
and there's no need to protect those macros any more.

These changes should make it easier to "tool-ify" a program by adding:
#if HAVE_NBTOOL_CONFIG_H
#include "nbtool_config.h"
#endif
to the top of the source files (for the general case).

add
#if HAVE_CONFIG_H
#include "config.h"
#endif
to get this to build under solaris-2.6/sparc with gcc-2.95.1 as part of
the NetBSD cross toolchain build. Fixes PR23211. Patch approved by
Alistair Crooks.

Move UCB-licensed code from 4-clause to 3-clause licence.

Patches provided by Joel Baker in PR 22365, verified by myself.

More tests if __RCSID is defined.

More POSIX overhaul:
Support the F, D and L modifiers to the f format correctly.
Generate hexdump byte-counts, precisions and spacings automatically.

WARNS=2

Further POSIX progress: The C, S, I and L modifiers now behave correctly,
specifying output in units of a char, short, int and long (as defined by the
host system) respectively. This made the POSIX -t code more than complicated
enough to merit its own function, so I did that.

Further POSIXification: od now supports -N, which is equivalent to hexdump's
-n (format only COUNT bytes of input).

Add a -j option to od, which does the same as the -s option to
hexdump (skipping some of the input). This brings our od slightly
closer to POSIX.2 conformance.

PR/12147: TheMan: add -t option to od.

char -> unsigned char

Add { and } to shut up egcs. Reformat the more questionable code.

WARNSify, fix .Nm usage, deprecate register

merge lite-2.

add explicit return types to many functions
pull global variable declarations and function prototypes into hexdump.h
add #includes where necessary for prototypes
replace index() with strchr(), const-poison static string vars, etc.

RCS ID police

Add RCS identifiers.

Previous patch was incorrect.

Fix a NULL-dereferencing, that caused a segfault with QMAGIC (yay).

branches: 1.1.1;
Initial revision

static + __dead

Integer values, no matter what their width, are cast to int64_t or
uint64_t for printing. Because of this, parse.c:rewrite() should use
PRI[diouxX]64 as the printf format; the previous revision which used
"lld" and friends was a temporary measure. We assume that PRId64 ends
with 'd', PRIx64 ends with 'x', etc.

In display.c:display(), change some u_intNN_t types to C99 standard
uintNN_t types, and consistently cast to [u]int64_t when printing
integer values (even in two cases where it's not really necessary but it
makes the code look more consistent).

Use ll instead of non-standard q as length modifier in format strings.
Makes this work on Solaris. OK by apb.

branches: 1.24.24;
use ecalloc in previous

Fix core dump caused by non-zerod memory.
From Auster in current_users.

PR/28157: KAMADA Ken'ichi: escape sequence of hexdump(1) is broken
(SIGSEGV at worst)
Applied patch, thanks for the report!

Fix uninitialized variables.

More programs using efun.

Simplify the way the end of a singly linked list is followed (for adding
items) so it is more obvious that we aren't going to indirect through
a null pointer.
Fixes coverty SID:101

de-__P, ANSIfy prototypes

Overhaul how `build.sh tools' are used:

* Rename "config.h" to "nbtool_config.h" and
HAVE_CONFIG_H to HAVE_NBTOOL_CONFIG_H.
This makes in more obvious in the source when we're using
tools/compat/config.h versus "standard autoconf" config.h

* Consistently move the inclusion of nbtool_config.h to before
<sys/cdefs.h> so that the former can provide __RCSID() (et al),
and there's no need to protect those macros any more.

These changes should make it easier to "tool-ify" a program by adding:
#if HAVE_NBTOOL_CONFIG_H
#include "nbtool_config.h"
#endif
to the top of the source files (for the general case).

add
#if HAVE_CONFIG_H
#include "config.h"
#endif
to get this to build under solaris-2.6/sparc with gcc-2.95.1 as part of
the NetBSD cross toolchain build. Fixes PR23211. Patch approved by
Alistair Crooks.

Move UCB-licensed code from 4-clause to 3-clause licence.

Patches provided by Joel Baker in PR 22365, verified by myself.

More tests if __RCSID is defined.

err/errx/warn/warnx do not need \n at the end

Enable support for printing 8-byte integers. For some reason, most of the
code for this was present, but disabled.
This is required for POSIX compliance on platforms with 8-byte longs.

constify

Don't core dump with an empty format string. Fixes PR#11218.
Patch supplied by Launey Thomas.

string length computation bug. PR 8676.

branches: 1.8.2; 1.8.10;
char -> unsigned char

WARNSify, fix .Nm usage, deprecate register

merge lite-2.

add explicit return types to many functions
pull global variable declarations and function prototypes into hexdump.h
add #includes where necessary for prototypes
replace index() with strchr(), const-poison static string vars, etc.

RCS ID police

Display addresses -- which are off_t's -- with proper format string.

Add RCS identifiers.

branches: 1.1.1;
Initial revision