]> CyberLeo.Net >> Repos - FreeBSD/FreeBSD.git/log
FreeBSD/FreeBSD.git
22 years agoSparkling new implementation of linux_ptrace. Slight tweaking by
Marcel Moolenaar [Sun, 19 May 2002 01:21:55 +0000 (01:21 +0000)]
Sparkling new implementation of linux_ptrace. Slight tweaking by
yours truly.

PR: 33299
Submitted by: Alexander N. Kabaev <ak03@gte.com>

22 years agoChange p_can{debug,see,sched,signal}()'s first argument to be a thread
John Baldwin [Sun, 19 May 2002 00:14:50 +0000 (00:14 +0000)]
Change p_can{debug,see,sched,signal}()'s first argument to be a thread
pointer instead of a proc pointer and require the process pointed to
by the second argument to be locked.  We now use the thread ucred reference
for the credential checks in p_can*() as a result.  p_canfoo() should now
no longer need Giant.

22 years agoRemove IFS from 5.0-CURRENT. This facilitates introducing UFS2 as
Robert Watson [Sun, 19 May 2002 00:11:08 +0000 (00:11 +0000)]
Remove IFS from 5.0-CURRENT.  This facilitates introducing UFS2 as
IFS had its fingers deep in the belly of the UFS/FFS split.  IFS
will be reimplemented by the maintainer at a later date.

Requested by: adrian (maintainer)

22 years agoRemove IFS from 5.0-CURRENT. This facilitates introducing UFS2 as
Robert Watson [Sun, 19 May 2002 00:09:30 +0000 (00:09 +0000)]
Remove IFS from 5.0-CURRENT.  This facilitates introducing UFS2 as
IFS had its fingers deep in the belly of the UFS/FFS split.  IFS
will be reimplemented by the maintainer at a later date once the UFS
implementation is in place.

Requested by: adrian (maintainer)

22 years agoDocument RAD_MICROSOFT_MS_CHAP2_SUCCESS.
Brian Somers [Sat, 18 May 2002 23:58:15 +0000 (23:58 +0000)]
Document RAD_MICROSOFT_MS_CHAP2_SUCCESS.

22 years agoNow that daddr_t has grown up, use %lld to printf it and cast it to long
John Baldwin [Sat, 18 May 2002 23:46:04 +0000 (23:46 +0000)]
Now that daddr_t has grown up, use %lld to printf it and cast it to long
long.

22 years agoAdd an ext2_uninit() routine that undoes the actions performed by
Ian Dowse [Sat, 18 May 2002 22:18:17 +0000 (22:18 +0000)]
Add an ext2_uninit() routine that undoes the actions performed by
ext2_init(). This permits the ext2fs module to be unloaded without
causing panics and leaking memory.

22 years agoFix two off-by-one errors when sanity-checking inode numbers. In
Ian Dowse [Sat, 18 May 2002 21:33:07 +0000 (21:33 +0000)]
Fix two off-by-one errors when sanity-checking inode numbers. In
ext2fs, inode numbers start at 1, so the maximum valid inode number
is (s_inodes_per_group * s_groups_count), not one less. This is
just a minimal change to avoid unnecessary panics and errors; some
other related bugs that Bruce Evans mentioned to me are not addressed.

Reviewed by: bde (ages ago)

22 years agoUnconditionally update the maximum field width statistics when we
Ian Dowse [Sat, 18 May 2002 21:10:40 +0000 (21:10 +0000)]
Unconditionally update the maximum field width statistics when we
refetch the filesystem information in MNT_WAIT mode. This avoids
incorrect column alignment that sometimes occurs with NFS filesystems.

Submitted by: Ian <freebsd@damnhippie.dyndns.org>

22 years agoremove perl scripts from the build.
Mark Murray [Sat, 18 May 2002 19:47:57 +0000 (19:47 +0000)]
remove perl scripts from the build.

22 years agoUse explicitly-sized types where necessary to make ext2fs work again
Ian Dowse [Sat, 18 May 2002 19:12:38 +0000 (19:12 +0000)]
Use explicitly-sized types where necessary to make ext2fs work again
after the change to a 64-bit daddr_t.

22 years agoFix two casts to "daddr_t *" that should have been "ufs_daddr_t *".
Ian Dowse [Sat, 18 May 2002 19:03:00 +0000 (19:03 +0000)]
Fix two casts to "daddr_t *" that should have been "ufs_daddr_t *".

22 years ago o Eliminate the acquisition and release of Giant from minherit(2).
Alan Cox [Sat, 18 May 2002 18:59:00 +0000 (18:59 +0000)]
 o Eliminate the acquisition and release of Giant from minherit(2).
    (vm_map_inherit() no longer requires Giant to be held.)

22 years agoFix a typo where sizeof(daddr_t) was specified instead of sizeof(doff_t).
Ian Dowse [Sat, 18 May 2002 18:58:27 +0000 (18:58 +0000)]
Fix a typo where sizeof(daddr_t) was specified instead of sizeof(doff_t).
Now that daddr_t is 64-bit, this caused hash blocks to be allocated
twice as large as they need to be.

22 years agoRemove um_i_effnlink_valid, i_spare[] and the ufsmount_u and inode_u
Ian Dowse [Sat, 18 May 2002 18:51:14 +0000 (18:51 +0000)]
Remove um_i_effnlink_valid, i_spare[] and the ufsmount_u and inode_u
unions, since these were only necessary when ext2fs used ufs code.

Reviewed by: mckusick

22 years agoNew release notes: Perl deletion (along with a caution that some work
Bruce A. Mah [Sat, 18 May 2002 18:44:24 +0000 (18:44 +0000)]
New release notes:  Perl deletion (along with a caution that some work
is still ongoing), catman(1) rewrite, makewhatis(1) rewrite,
spkrtest(8) rewrite.

Modified release note:  killall(1) for markup.

Deleted release notes:  Perl upgrade, style.perl(7) manpage, one note
about killall(1) was obsolete.

22 years agoThis is enough to get thru a cross-build.
David E. O'Brien [Sat, 18 May 2002 18:27:09 +0000 (18:27 +0000)]
This is enough to get thru a cross-build.

22 years agoelf2aout can be useful to have on all platforms
David E. O'Brien [Sat, 18 May 2002 18:25:58 +0000 (18:25 +0000)]
elf2aout can be useful to have on all platforms

22 years agoDon't build getconf if you don't have a full C++ installation.
David E. O'Brien [Sat, 18 May 2002 18:24:55 +0000 (18:24 +0000)]
Don't build getconf if you don't have a full C++ installation.

22 years agoBack out rev 1.278. Seems we grew a getconf(1) that requires it.
David E. O'Brien [Sat, 18 May 2002 18:23:04 +0000 (18:23 +0000)]
Back out rev 1.278.  Seems we grew a getconf(1) that requires it.

22 years agoRemove local patch that crept in.
David E. O'Brien [Sat, 18 May 2002 18:17:13 +0000 (18:17 +0000)]
Remove local patch that crept in.

22 years agoFix markup.
Ruslan Ermilov [Sat, 18 May 2002 15:39:56 +0000 (15:39 +0000)]
Fix markup.

Give Wolfram a credit for writing the original version in Perl
like in killall(1).

22 years agomdoc(7) police: fixed the whatis entry.
Ruslan Ermilov [Sat, 18 May 2002 15:20:28 +0000 (15:20 +0000)]
mdoc(7) police: fixed the whatis entry.

22 years agoMore whitespace fixes.
Ruslan Ermilov [Sat, 18 May 2002 15:15:30 +0000 (15:15 +0000)]
More whitespace fixes.

22 years agoRemove Perl and TCL/TK bits, neither of which are in the base system
Mark Murray [Sat, 18 May 2002 14:27:17 +0000 (14:27 +0000)]
Remove Perl and TCL/TK bits, neither of which are in the base system
(anymore).

22 years agoFixed spacing.
Ruslan Ermilov [Sat, 18 May 2002 14:17:53 +0000 (14:17 +0000)]
Fixed spacing.

22 years agoLook into machine specific subdirs.
Ruslan Ermilov [Sat, 18 May 2002 13:56:53 +0000 (13:56 +0000)]
Look into machine specific subdirs.

22 years agoBack out part of previous commit: make -P the default again until I'm
Tim J. Robbins [Sat, 18 May 2002 13:52:07 +0000 (13:52 +0000)]
Back out part of previous commit: make -P the default again until I'm
convinced it's a good idea.

22 years agoSigh, this README is not a shell script.
Ruslan Ermilov [Sat, 18 May 2002 12:37:19 +0000 (12:37 +0000)]
Sigh, this README is not a shell script.

22 years agoRemind developers to update hier(7) if they make changes to these files.
Ruslan Ermilov [Sat, 18 May 2002 12:35:10 +0000 (12:35 +0000)]
Remind developers to update hier(7) if they make changes to these files.

22 years agoKeep up with BSD.usr.dist,v 1.260.
Ruslan Ermilov [Sat, 18 May 2002 12:32:41 +0000 (12:32 +0000)]
Keep up with BSD.usr.dist,v 1.260.

22 years agoThe AMD x86-64 Hammer bits.
David E. O'Brien [Sat, 18 May 2002 11:57:01 +0000 (11:57 +0000)]
The AMD x86-64 Hammer bits.

22 years agoAdd description for uma_zcreate().
Jeroen Ruigrok van der Werven [Sat, 18 May 2002 11:12:02 +0000 (11:12 +0000)]
Add description for uma_zcreate().

Submitted by: arr

22 years agoTry again: Make daddr_t 64 bits.
Poul-Henning Kamp [Sat, 18 May 2002 09:48:28 +0000 (09:48 +0000)]
Try again: Make daddr_t 64 bits.

Sponsored by: DARPA & NAI Labs.

22 years agoFix the sparc64 build and make the LIB1ASMSRC handling more robust.
David E. O'Brien [Sat, 18 May 2002 09:46:12 +0000 (09:46 +0000)]
Fix the sparc64 build and make the LIB1ASMSRC handling more robust.

22 years agoMove the hideously misnamed type "u_daddr_t" to <sys/blist.h> where it
Poul-Henning Kamp [Sat, 18 May 2002 09:38:20 +0000 (09:38 +0000)]
Move the hideously misnamed type "u_daddr_t" to <sys/blist.h> where it
belongs.

Sponsored by: DARPA & NAI Labs.

22 years agoUse btodb() macro.
Poul-Henning Kamp [Sat, 18 May 2002 09:34:09 +0000 (09:34 +0000)]
Use btodb() macro.

Sponsored by: DARPA & NAI Labs.

22 years agoRemove a check of blocknumbers/offsets which will be pointless with
Poul-Henning Kamp [Sat, 18 May 2002 09:32:56 +0000 (09:32 +0000)]
Remove a check of blocknumbers/offsets which will be pointless with
64 bit daddr_t.

Sponsored by: DARPA & NAI Labs.

22 years agoImprove chances that we correctly compile LIB1ASMSRC on all architectures.
Poul-Henning Kamp [Sat, 18 May 2002 09:21:42 +0000 (09:21 +0000)]
Improve chances that we correctly compile LIB1ASMSRC on all architectures.

sparc64 looked for the nonexistent sparc64/lb1spc.asm file instead
of the sparc/lb1spc.asm file.

arm probably looked for arm/arm/lib1funcs.asm instead of arm/lib1funcs.asm
ia64 probably looked for ia64/ia64/lib1funcs.asm instead of ia64/lib1funcs.asm

i386 and alpha don't seen to use the LIB1ASMSRC.

22 years agoAdd C rewrites of catman and makewhatis. These aren't quite as
Mark Murray [Sat, 18 May 2002 09:19:08 +0000 (09:19 +0000)]
Add C rewrites of catman and makewhatis. These aren't quite as
the submitter supplied them, as I did some WARNS=n fixups (mostly
const-ification).

Submitted by: John Rochester <john@jrochester.org>

22 years agoDon't try to build libc_r on sparc64, the _atomic_lock.S file isn't implemented.
Poul-Henning Kamp [Sat, 18 May 2002 09:14:36 +0000 (09:14 +0000)]
Don't try to build libc_r on sparc64, the _atomic_lock.S file isn't implemented.

22 years agoFix one of the va_arg() with less than int sized type probelms.
Poul-Henning Kamp [Sat, 18 May 2002 09:12:32 +0000 (09:12 +0000)]
Fix one of the va_arg() with less than int sized type probelms.

22 years agoUFS indirect blocks are size u_int32_t, not daddr_t
Poul-Henning Kamp [Sat, 18 May 2002 09:07:12 +0000 (09:07 +0000)]
UFS indirect blocks are size u_int32_t, not daddr_t

22 years agoDrop perl5 hard lock as perl5 is now gone.
Mark Murray [Sat, 18 May 2002 07:59:51 +0000 (07:59 +0000)]
Drop perl5 hard lock as perl5 is now gone.

22 years agoIn msgrcv(), set msgtyp correctly. Hardwiring 0 as the message type
Marcel Moolenaar [Sat, 18 May 2002 07:53:56 +0000 (07:53 +0000)]
In msgrcv(), set msgtyp correctly. Hardwiring 0 as the message type
yields incorrect behaviour. The hardwiring was present in the very
first commit that implemented msgrcv() (revision 1.4) and hasn't been
changed since. The native implementation was complete at that time,
so there doesn't seem to be a reason for the hardwiring from a
technical point of view.

Submitted by: Reinier Bezuidenhout <rbezuide@yahoo.com>

22 years ago o Remove GIANT_REQUIRED from vm_map_madvise(). Instead, acquire and
Alan Cox [Sat, 18 May 2002 07:48:06 +0000 (07:48 +0000)]
 o Remove GIANT_REQUIRED from vm_map_madvise().  Instead, acquire and
   release Giant around vm_map_madvise()'s call to pmap_object_init_pt().
 o Replace GIANT_REQUIRED in vm_object_madvise() with the acquisition
   and release of Giant.
 o Remove the acquisition and release of Giant from madvise().

22 years agoWe haven't needed libg++ since 1999-04-04 (pre 4.0).
David E. O'Brien [Sat, 18 May 2002 05:38:31 +0000 (05:38 +0000)]
We haven't needed libg++ since 1999-04-04 (pre 4.0).

22 years agoWARNS 6 this.
David E. O'Brien [Sat, 18 May 2002 05:33:28 +0000 (05:33 +0000)]
WARNS 6 this.

22 years agoAdd a perl wrapper. mergmaster(8) is broken with out it.
David E. O'Brien [Sat, 18 May 2002 05:31:48 +0000 (05:31 +0000)]
Add a perl wrapper.  mergmaster(8) is broken with out it.

Submitted by: DES

22 years agoSegregate MD weirdness a little bit more. Improve cross buildability.
David E. O'Brien [Sat, 18 May 2002 04:59:57 +0000 (04:59 +0000)]
Segregate MD weirdness a little bit more.  Improve cross buildability.

22 years agoStyle sync with other platforms.
David E. O'Brien [Sat, 18 May 2002 04:50:32 +0000 (04:50 +0000)]
Style sync with other platforms.

22 years agoUse a more API denoting way to handle what is in libc and what isn't.
David E. O'Brien [Sat, 18 May 2002 04:49:44 +0000 (04:49 +0000)]
Use a more API denoting way to handle what is in libc and what isn't.

22 years ago o Remove the acquisition and release of Giant from mprotect().
Alan Cox [Sat, 18 May 2002 03:58:16 +0000 (03:58 +0000)]
 o Remove the acquisition and release of Giant from mprotect().

22 years agoMake -L the default, allow both -L and -P to be specified (last one used
Tim J. Robbins [Sat, 18 May 2002 02:47:25 +0000 (02:47 +0000)]
Make -L the default, allow both -L and -P to be specified (last one used
matters), fall back to -P mode if we can't get the logical directory.

22 years agoFix the order of shutdown scripts so it not only reverses the order of the
Gordon Tetlow [Sat, 18 May 2002 00:26:10 +0000 (00:26 +0000)]
Fix the order of shutdown scripts so it not only reverses the order of the
files in each ${local_startup} directory, it also reverses the order of the
directories.

Suggested by: jhb
Reviewed by: jake
Approved by: dougb
MFC after: 1 week

22 years agoDon't build gperf.
David E. O'Brien [Sat, 18 May 2002 00:20:01 +0000 (00:20 +0000)]
Don't build gperf.

22 years agoDon't depend on gperf.
David E. O'Brien [Sat, 18 May 2002 00:18:00 +0000 (00:18 +0000)]
Don't depend on gperf.

22 years agoBuild ports needed for release documentation WITHOUT_PYTHON.
Bruce A. Mah [Sat, 18 May 2002 00:00:41 +0000 (00:00 +0000)]
Build ports needed for release documentation WITHOUT_PYTHON.

22 years agoAdd devel/bison and textproc/xhtml ports to the set of ports required
Bruce A. Mah [Fri, 17 May 2002 23:59:53 +0000 (23:59 +0000)]
Add devel/bison and textproc/xhtml ports to the set of ports required
for building docs with a release.

22 years agoAdd TAG_VENDOR_INDENTIFIER (option 60) to our DHCP request done by the
Doug Ambrisko [Fri, 17 May 2002 20:18:48 +0000 (20:18 +0000)]
Add TAG_VENDOR_INDENTIFIER (option 60) to our DHCP request done by the
kernel BOOTP option.  The format will be:
FreeBSD:<MACHINE>:<osrelease>
this way people can tune their DHCP server to server up root file systems
via the OS, machine type and version.

Obtained from: NetBSD
MFC after: 3 weeks

22 years agothe processed hash file
David E. O'Brien [Fri, 17 May 2002 19:47:48 +0000 (19:47 +0000)]
the processed hash file

22 years agoCheck out the textproc/expat2 port for cases where we build docs
Bruce A. Mah [Fri, 17 May 2002 19:01:21 +0000 (19:01 +0000)]
Check out the textproc/expat2 port for cases where we build docs
in a release without a full ports tree.

22 years agoFix ufs_daddr_t/daddr_t type problems.
Poul-Henning Kamp [Fri, 17 May 2002 18:59:53 +0000 (18:59 +0000)]
Fix ufs_daddr_t/daddr_t type problems.

Sponsored by: DARPA & NAI labs.

22 years agoCall ufs_bmaparray() with right parameter type.
Poul-Henning Kamp [Fri, 17 May 2002 18:53:29 +0000 (18:53 +0000)]
Call ufs_bmaparray() with right parameter type.

Sponsored by: DARPA & NAI Labs.

22 years agoAdd my birthday to the list. Apparently, I share a birthday with dwhite.
Gordon Tetlow [Fri, 17 May 2002 17:49:32 +0000 (17:49 +0000)]
Add my birthday to the list. Apparently, I share a birthday with dwhite.

Reviewed by: jake
Approved by: jake

22 years agoFix a tiny bug in shlib support of ldd(1); if dlopen(lib, RTLD_TRACE)
Akinori MUSHA [Fri, 17 May 2002 17:06:56 +0000 (17:06 +0000)]
Fix a tiny bug in shlib support of ldd(1); if dlopen(lib, RTLD_TRACE)
returns, exit gracefully with 0.

This fixes the behavior you see when you specify libc.so.  It occurs
because ldd(1) itself is linked with libc.so.

$ ldd /usr/lib/libc.so
/usr/lib/libc.so:
ldd: /usr/lib/libc.so: (null)
/usr/lib/libc.so: exit status 1

Reviewed by: silence of audit@

22 years agoUpdate the EINVAL description again - it turns out that EINVAL
Peter Pentchev [Fri, 17 May 2002 16:39:21 +0000 (16:39 +0000)]
Update the EINVAL description again - it turns out that EINVAL
may be returned by Q_SETQUOTA as well as the originally listed
Q_GETQUOTA.

Noticed by: dd
Approved by: silence from dd
MFC after: 3 days

22 years agoMFC noted: cdcontrol(1) speed.
Bruce A. Mah [Fri, 17 May 2002 16:11:36 +0000 (16:11 +0000)]
MFC noted:  cdcontrol(1) speed.

Minor wording tweaks in ports infrastructure section.

22 years agoGet byte swapping primitives from sys/endian.h now that they are exported
Jake Burkholder [Fri, 17 May 2002 15:46:08 +0000 (15:46 +0000)]
Get byte swapping primitives from sys/endian.h now that they are exported
to userland.  This makes elf2aout work as a cross tool.

22 years agoTemporarily change our umask to 066 so that the potential creation
Brian Somers [Fri, 17 May 2002 14:05:08 +0000 (14:05 +0000)]
Temporarily change our umask to 066 so that the potential creation
of wtmp.0 is done as mode 600.

This ensures that tight permissions set in /etc/newsyslog.conf for
wtmp logging aren't ``betrayed''.

Suggested by: lumpy <lumpy@the.whole.net>
MFC after: 3 days

22 years agoChange ``dmesg -a'' to ``dmesg''.
Brian Somers [Fri, 17 May 2002 13:38:36 +0000 (13:38 +0000)]
Change ``dmesg -a'' to ``dmesg''.

The change was introduced in src/etc/security 1.53 almost a year ago
in an attempt to see ipfw deny message logs.

However, ipfw deny/reject logs have been displayed since version 1.13
of the same file as a separate ``job'' and have since moved to
src/etc/periodic/security/500.ipfwdenied.

MFC after: 3 days

22 years agoRegenerate with new output grouping in uuencode -m.
Juli Mallett [Fri, 17 May 2002 12:27:31 +0000 (12:27 +0000)]
Regenerate with new output grouping in uuencode -m.

22 years agoGroup the output chunks differently, to try to match BSD/OS output. Switch
Juli Mallett [Fri, 17 May 2002 12:27:02 +0000 (12:27 +0000)]
Group the output chunks differently, to try to match BSD/OS output.  Switch
to using arithmatic to determine buffer sizes to encode into.

Diane Bruce pointed out to me that BSD/OS did MIME too, so I want to match
their output, too, since my decision of 8 output groups was wholly arbitrary.

22 years agoOptimize prev. commit code a bit
Andrey A. Chernov [Fri, 17 May 2002 12:24:19 +0000 (12:24 +0000)]
Optimize prev. commit code a bit

22 years agoTrailing slash fixes.
Andrey A. Chernov [Fri, 17 May 2002 11:52:48 +0000 (11:52 +0000)]
Trailing slash fixes.

Fix the case:
        cp file nonexistent/
which create nonextstent as file while trailing slash clearly indicates
that nonexistent must be a directory.
Also fix the case:
cp file1 file2/
which should produce error.

22 years agoMove _PATH_WALL from dump and shutdown's local pathnames.h to paths.h.
Juli Mallett [Fri, 17 May 2002 11:47:12 +0000 (11:47 +0000)]
Move _PATH_WALL from dump and shutdown's local pathnames.h to paths.h.

22 years agoRemove local definition of _PATH_CP, and use <paths.h>, _PATH_CP is in the
Juli Mallett [Fri, 17 May 2002 11:38:48 +0000 (11:38 +0000)]
Remove local definition of _PATH_CP, and use <paths.h>, _PATH_CP is in the
CURRENT <paths.h>.

22 years agoTighten up temporary file permissions and move them to ${TMPDIR:-/tmp}
Brian Somers [Fri, 17 May 2002 11:34:12 +0000 (11:34 +0000)]
Tighten up temporary file permissions and move them to ${TMPDIR:-/tmp}

Problem reported by: lumpy <lumpy@the.whole.net>
MFC after: 3 days

22 years agoReturn 3 unless $daily_status_security_enable != YES.
Brian Somers [Fri, 17 May 2002 11:31:45 +0000 (11:31 +0000)]
Return 3 unless $daily_status_security_enable != YES.
Returning $? masks security output when ``periodic security'' is successful !

MFC after: 3 days

22 years agoRevert part of the previous revision. Changing exit() to return() in main()
Juli Mallett [Fri, 17 May 2002 11:16:49 +0000 (11:16 +0000)]
Revert part of the previous revision.  Changing exit() to return() in main()
is wrong, even though our C compiler doesn't understand that exit() eats
control of the program, and as such can sometimes complain that main() reaches
the bottom of its body without an explicit return().

Aside from that, the functional usefulness of changing it is effectively nil,
so back it out.

Submitted by: mike

22 years agoMFS: Extend a buffer size so that packages with particularily deep dependency
Brian Somers [Fri, 17 May 2002 09:46:46 +0000 (09:46 +0000)]
MFS: Extend a buffer size so that packages with particularily deep dependency
     lists install properly.

22 years agoMove LIB1ASMFUNCS from the SYMS variable and explicitly add it to OBJS
Peter Wemm [Fri, 17 May 2002 08:59:13 +0000 (08:59 +0000)]
Move LIB1ASMFUNCS from the SYMS variable and explicitly add it to OBJS
later.  Otherwise make will try and build the supposedly assembler .o
files from libgcc2.c - which does not work too well (the .o's have no
content)

Reviewed by: obrien

22 years agoMake the output tidier when multiple integer formats are requested by
Tim J. Robbins [Fri, 17 May 2002 08:54:32 +0000 (08:54 +0000)]
Make the output tidier when multiple integer formats are requested by
attempting to line up values into columns.

Obtained from: NetBSD (idea)

22 years agoSimplify prerun() in the case where there are no arguments. Can't do -I or -J
Juli Mallett [Fri, 17 May 2002 07:55:24 +0000 (07:55 +0000)]
Simplify prerun() in the case where there are no arguments.  Can't do -I or -J
substitution if `utility' is not specified and we're using the buildin echo;
It has no arguments for us to abuse.

22 years agoPut on peril-sensitive sunglasses and risk touching gcc related build glue
Peter Wemm [Fri, 17 May 2002 07:55:08 +0000 (07:55 +0000)]
Put on peril-sensitive sunglasses and risk touching gcc related build glue
again.  Try and deal with platforms that provide their own crtbegin/end asm
files (ia64 for example).  crtstuff.c does not actually work on ia64 since
libgcc.a doesn't have a few key support functions when built on ia64 so it
is compulsory to use crtbegin.asm and crtend.asm.

Reviewed by: obrien

22 years agoOverhaul hexdump's od syntax code to handle the -s -A -j -N -t options that
Tim J. Robbins [Fri, 17 May 2002 07:14:55 +0000 (07:14 +0000)]
Overhaul hexdump's od syntax code to handle the -s -A -j -N -t options that
SUSv3 requires and give od a proper manual page.

PR: 36783

22 years agoRemove some WIP bits that I didn't fully clean out before merging to HEAD.
David E. O'Brien [Fri, 17 May 2002 06:35:44 +0000 (06:35 +0000)]
Remove some WIP bits that I didn't fully clean out before merging to HEAD.

22 years agoDeclare variables that were extern'd in multiple places in hexdump.h.
Tim J. Robbins [Fri, 17 May 2002 06:32:00 +0000 (06:32 +0000)]
Declare variables that were extern'd in multiple places in hexdump.h.

PR: 36783

22 years agoAdd support for printing long doubles.
Tim J. Robbins [Fri, 17 May 2002 06:12:25 +0000 (06:12 +0000)]
Add support for printing long doubles.

PR: 36783

22 years ago- Apparently, the Alpha ABI mandates that arguments be passed sign-extended
John Baldwin [Fri, 17 May 2002 05:45:39 +0000 (05:45 +0000)]
- Apparently, the Alpha ABI mandates that arguments be passed sign-extended
  regardless of if they are signed or unsigned since it is easier to work
  with sign-extended values.  Thus, remove the disabled zapnot to
  zero-extend the sign-extended value we read from *p in atomic_cmpset_32()
  since the cmpval we are comparing against should already be
  sign-extended.
- To ensure that the compiler knows to sign-extend the upper 32 bits of
  cmpval rather than leaving garbage in there, cast the appropriately in
  the constraints section.

Help from: Richard Henderson <rth@redhat.com>

22 years agoPrint signed single-byte decimal integers correctly instead of implicitly
Tim J. Robbins [Fri, 17 May 2002 05:43:00 +0000 (05:43 +0000)]
Print signed single-byte decimal integers correctly instead of implicitly
converting them to unsigned bytes.

PR: 36783

22 years agoBack out revision 1.30 change.
Ruslan Ermilov [Fri, 17 May 2002 05:41:47 +0000 (05:41 +0000)]
Back out revision 1.30 change.

cc1plus can apparently be built if you happen to have
/usr/bin/gperf, or set CXX to point to a C++ compiler
that can build gperf(1) in the bootstrap-tools stage
of buildworld.

22 years agoStop this program's abuse of malloc(3). Its return value doesn't need these
Juli Mallett [Fri, 17 May 2002 05:27:52 +0000 (05:27 +0000)]
Stop this program's abuse of malloc(3).  Its return value doesn't need these
ugly explicit casts, and its argument doesn't need explicitly cast to u_int,
especially if sizeof() is being used.

22 years agoUn-deprecate od(1): rename the `deprecated' variable to `odmode', remove the
Tim J. Robbins [Fri, 17 May 2002 05:20:30 +0000 (05:20 +0000)]
Un-deprecate od(1): rename the `deprecated' variable to `odmode', remove the
deprecation warning from the utility and manual page. Since this utility
is required by POSIX, it's not likely to be removed any time soon.

This is leading up to the addition of the P1003.1-2001 -s -A -j -N -t options.

PR: 36783

22 years ago_C_LABEL -> CNAME
Benno Rice [Fri, 17 May 2002 05:11:13 +0000 (05:11 +0000)]
_C_LABEL -> CNAME

Forgotten by: benno

22 years agoClean up malloc(3)'s argument. Remove casts which do nothing when we're
Juli Mallett [Fri, 17 May 2002 05:11:07 +0000 (05:11 +0000)]
Clean up malloc(3)'s argument.  Remove casts which do nothing when we're
using sizeof() anyway.  Use slightly more consistent (per-file) error
reporting for malloc(3) returning NULL.  If "malloc failed" was being printed,
don't use err(3).  If a NULL format is being used, use err(3).  In one case
errx(3) was being used with strerror(3), so just use err(3).

22 years agoPost rev 1.39, the PowerPC specific additions to OBJS was getting lost.
David E. O'Brien [Fri, 17 May 2002 04:18:33 +0000 (04:18 +0000)]
Post rev 1.39, the PowerPC specific additions to OBJS was getting lost.

22 years agoUpdate to Binutils 2.12.0 and stick to the native emulation for now.
David E. O'Brien [Fri, 17 May 2002 03:43:47 +0000 (03:43 +0000)]
Update to Binutils 2.12.0 and stick to the native emulation for now.
This lets us get on the with the task at hand -- porting FreeBSD.
We can get fancy later on.

22 years agoUpdate to Binutils 2.12.0 release.
David E. O'Brien [Fri, 17 May 2002 03:43:09 +0000 (03:43 +0000)]
Update to Binutils 2.12.0 release.

22 years agoBump __FreeBSD_version to note that Perl is not in /usr/src any more.
David E. O'Brien [Fri, 17 May 2002 03:13:08 +0000 (03:13 +0000)]
Bump __FreeBSD_version to note that Perl is not in /usr/src any more.