oweals/busybox.git
18 years ago2006-06-02 Shaun Jackman <sjackman@gmail.com>
Bernhard Reutner-Fischer [Sat, 3 Jun 2006 10:28:25 +0000 (10:28 -0000)]
2006-06-02  Shaun Jackman  <sjackman@gmail.com>

* shell/lash.c (pseudo_exec) [_NEWLIB_VERSION]: Remove the
__getopt_initialized hackery.

18 years ago- No options should set the current system time.
Bernhard Reutner-Fischer [Sat, 3 Jun 2006 10:24:20 +0000 (10:24 -0000)]
- No options should set the current system time.
Shaun Jackman writes:
A bug introduced in svn 11946 broke rdate. It no longer sets the
current system time when no options are specified. The options have
the opposite sense from what one might think, and, oddly enough, -ps
is intentionally a no-op.

Quoth rdate(8) from the BSD System Manager's Manual:

    -p      Do not set, just print the remote time
    -s      Do not print the time.

18 years agoPatch from Atsushi Nemoto:
Rob Landley [Fri, 2 Jun 2006 21:30:40 +0000 (21:30 -0000)]
Patch from Atsushi Nemoto:
This patch make depmod.pl more robust in cross environment.  If native
nm did not recognize target's object, you can override it with NM
environment variable.

18 years ago- move #include busybox.h to the very top so we pull in the config
Bernhard Reutner-Fischer [Fri, 2 Jun 2006 20:56:16 +0000 (20:56 -0000)]
- move #include busybox.h to the very top so we pull in the config
  and eventual platform specific includes in early.

18 years agoHeader cleanup: don't #include headers that libbb.h already includes.
Rob Landley [Fri, 2 Jun 2006 20:53:38 +0000 (20:53 -0000)]
Header cleanup: don't #include headers that libbb.h already includes.

18 years ago#ifdef reduction.
Rob Landley [Fri, 2 Jun 2006 19:03:01 +0000 (19:03 -0000)]
#ifdef reduction.

18 years ago- fix CONFIG_DEBUG_INIT compile failure
Bernhard Reutner-Fischer [Fri, 2 Jun 2006 10:43:17 +0000 (10:43 -0000)]
- fix CONFIG_DEBUG_INIT compile failure

18 years agouse xmalloc()
Mike Frysinger [Fri, 2 Jun 2006 06:30:30 +0000 (06:30 -0000)]
use xmalloc()

18 years agofix fdflish typo (reported by erik hovland)
Paul Fox [Fri, 2 Jun 2006 03:37:16 +0000 (03:37 -0000)]
fix fdflish typo (reported by erik hovland)

18 years ago- look at bb_applet_name. Should close bugs #892 and #893
Bernhard Reutner-Fischer [Thu, 1 Jun 2006 18:30:42 +0000 (18:30 -0000)]
- look at bb_applet_name. Should close bugs #892 and #893
  also restructure code a bit so the fix only adds 1 byte..

18 years ago- shrink logread a bit.
Bernhard Reutner-Fischer [Thu, 1 Jun 2006 14:36:14 +0000 (14:36 -0000)]
- shrink logread a bit.
   text    data     bss     dec     hex filename
    862      26     188    1076     434 logread.o.orig
    825      26     188    1039     40f logread.o

18 years agomake halt and poweroff messages more accurate (i.e., claim "we're about to",
Paul Fox [Thu, 1 Jun 2006 13:17:49 +0000 (13:17 -0000)]
make halt and poweroff messages more accurate (i.e., claim "we're about to",
rather than "it's done", and combine with reboot handler to save space.

18 years agoShaun Jackman pointed out that KERNEL_VERSION() is used without guards, so
Rob Landley [Thu, 1 Jun 2006 00:52:30 +0000 (00:52 -0000)]
Shaun Jackman pointed out that KERNEL_VERSION() is used without guards, so
defining it in a guard is silly.

18 years agoThe common case APPLET() macro only needs three arguments.
Rob Landley [Thu, 1 Jun 2006 00:51:38 +0000 (00:51 -0000)]
The common case APPLET() macro only needs three arguments.

18 years agoBugfix from Shaun Jackman: don't attempt to write 64 bytes and then fail if
Rob Landley [Wed, 31 May 2006 23:54:50 +0000 (23:54 -0000)]
Bugfix from Shaun Jackman: don't attempt to write 64 bytes and then fail if
the write wasn't 192 bytes long.

18 years agoAdd CONFIG_NITPICK. (And tweak a couple other help entries while I'm there.)
Rob Landley [Wed, 31 May 2006 23:23:42 +0000 (23:23 -0000)]
Add CONFIG_NITPICK.  (And tweak a couple other help entries while I'm there.)

Nothing's actually using it yet, but we need it.

18 years agoPatch from Tito to remove long options from hdparm.
Rob Landley [Wed, 31 May 2006 22:52:57 +0000 (22:52 -0000)]
Patch from Tito to remove long options from hdparm.

18 years agoUse BB endianness macro.
Rob Landley [Wed, 31 May 2006 22:24:33 +0000 (22:24 -0000)]
Use BB endianness macro.

18 years agoHeader file cleanup. platform.h apparently needs to be included early on so it
Rob Landley [Wed, 31 May 2006 22:09:03 +0000 (22:09 -0000)]
Header file cleanup.  platform.h apparently needs to be included early on so it
can figure out what header files to include, and override stuff that comes
later.  But applets shouldn't include platform.h directly, they should include
busybox.h or libbb.h.  Since busybox.h already includes libbb.h, move libbb.h
to the top of busybox.h and platform.h near the top of libbb.h (right after
bbconfig.h, which is something platform.h also needs access to).

While we're at it, move some stuff from busybox.h to libbb.h so we have one
big file to audit/clean up/try to make sense of instead of many.

18 years agoAdd catv (separate applet instead of cat -v). Also cleanup cat.c comments,
Rob Landley [Wed, 31 May 2006 19:36:04 +0000 (19:36 -0000)]
Add catv (separate applet instead of cat -v).  Also cleanup cat.c comments,
the following of which (from cat.c) belongs in svn history instead of the
source code:

/* Mar 16, 2003      Manuel Novoa III   (mjn3@codepoet.org)
 *
 * This is a new implementation of 'cat' which aims to be SUSv3 compliant.
 *
 * Changes from the previous implementation include:
 * 1) Multiple '-' args are accepted as required by SUSv3.  The previous
 *    implementation would close stdin and segfault on a subsequent '-'.
 * 2) The '-u' options is required by SUSv3.  Note that the specified
 *    behavior for '-u' is done by default, so all we need do is accept
 *    the option.
 */

18 years ago- fix IMA compile error i introduced when adding this untested snippet.
Bernhard Reutner-Fischer [Wed, 31 May 2006 18:57:56 +0000 (18:57 -0000)]
- fix IMA compile error i introduced when adding this untested snippet.

18 years agoha! a scratch-buffer! ;)
Bernhard Reutner-Fischer [Wed, 31 May 2006 15:36:20 +0000 (15:36 -0000)]
ha! a scratch-buffer! ;)
put the isdigit hack into it to help landley out..

18 years ago- make getopt_ulflags aware of CONFIG_GETOPT_LONG
Bernhard Reutner-Fischer [Wed, 31 May 2006 14:12:51 +0000 (14:12 -0000)]
- make getopt_ulflags aware of CONFIG_GETOPT_LONG

18 years ago- add CONFIG_FEATURE_WGET_LONG_OPTIONS
Bernhard Reutner-Fischer [Wed, 31 May 2006 14:11:38 +0000 (14:11 -0000)]
- add CONFIG_FEATURE_WGET_LONG_OPTIONS

18 years ago- convert xcalloc(1,x) to xzalloc
Bernhard Reutner-Fischer [Wed, 31 May 2006 13:31:16 +0000 (13:31 -0000)]
- convert xcalloc(1,x) to xzalloc

18 years ago- shrink klogd a bit
Bernhard Reutner-Fischer [Wed, 31 May 2006 12:22:13 +0000 (12:22 -0000)]
- shrink klogd a bit
   text    data     bss     dec     hex filename
    569       0       0     569     239 sysklogd/klogd.o.orig
    537       0       0     537     219 sysklogd/klogd.o

18 years ago- s/defined(__uClinux__)/BB_NOMMU/
Bernhard Reutner-Fischer [Wed, 31 May 2006 10:19:51 +0000 (10:19 -0000)]
- s/defined(__uClinux__)/BB_NOMMU/
This needs a second pass to:
+ add bb_daemon(unsigned char no_chdir, unsigned char no_close, const char*flag)
+ eventually globally export argc and argv, so we don't need to pass it to
  bb_daemon().

18 years ago- conditionally define BB_NOMMU
Bernhard Reutner-Fischer [Wed, 31 May 2006 10:04:03 +0000 (10:04 -0000)]
- conditionally define BB_NOMMU

18 years agoSince we have less than 100k of usage.h data anyway, using bzip -9 is silly.
Rob Landley [Tue, 30 May 2006 19:19:45 +0000 (19:19 -0000)]
Since we have less than 100k of usage.h data anyway, using bzip -9 is silly.
That says use 900k chunks when compressing, which needs about 4 megs of data
structures to undo the Burrows-Wheeler transform.  Switching it down to
bzip -1 (100k chunks) should have no impact on the compression (since it
still all fits in one chunk) but should reduce runtime decompression memory
requirements to something like 500k.  Still larger than gunzip, but not
egregiously so.

18 years agoTest to see if I can rename a file without blanking its history...
Rob Landley [Tue, 30 May 2006 18:49:05 +0000 (18:49 -0000)]
Test to see if I can rename a file without blanking its history...

18 years agoNote about BB_NOMMU.
Rob Landley [Tue, 30 May 2006 18:31:37 +0000 (18:31 -0000)]
Note about BB_NOMMU.

18 years ago- consolidate "Sending SIG%s to all processes"; untested..
Bernhard Reutner-Fischer [Tue, 30 May 2006 18:17:21 +0000 (18:17 -0000)]
- consolidate "Sending SIG%s to all processes"; untested..
   text    data     bss     dec     hex filename
   5379      32       8    5419    152b init/init.o.oorig
   5358      32       8    5398    1516 init/init.o

18 years ago- this could use a brother which deals with glibc based systems..
Bernhard Reutner-Fischer [Tue, 30 May 2006 15:49:10 +0000 (15:49 -0000)]
- this could use a brother which deals with glibc based systems..

18 years ago- remove useless variable "status" from init_main.
Bernhard Reutner-Fischer [Tue, 30 May 2006 12:16:54 +0000 (12:16 -0000)]
- remove useless variable "status" from init_main.
http://www.opengroup.org/onlinepubs/009695399/functions/waitpid.html suggests
that we need not specify a status if we don't want, and we don't.
"If wait() or waitpid() return because the status of a child process is available, these functions shall return a value equal to the process ID of the child process. In this case, if the value of the argument stat_loc is not a null pointer, information shall be stored in the location pointed to by stat_loc.  "
   text    data     bss     dec     hex filename
   5391      32       8    5431    1537 init/init.o.06
   5379      32       8    5419    152b init/init.o

18 years ago- use config_buffer for message()
Bernhard Reutner-Fischer [Tue, 30 May 2006 12:10:29 +0000 (12:10 -0000)]
- use config_buffer for message()
- add second argument to waitfor(*action,pid); if action==NULL then use pid tor
  wait for. If an action was given, we wait for the action to finish just as
  before. In run() remove second and third occurance of the same functionality
  the waitfor() call now provides.
  Adjust the former only caller of waitfor accordingly.

PS: Not using waitfor but creating a second function used a few bytes more than
simply extending and reusing waitfor.
   text    data     bss     dec     hex filename
   5426      32       8    5466    155a init/init.o.orig
   5391      32       8    5431    1537 init/init.o

18 years agoPut parentheses around some of the SWAP() macros, as pointed out by Peter
Rob Landley [Tue, 30 May 2006 06:28:03 +0000 (06:28 -0000)]
Put parentheses around some of the SWAP() macros, as pointed out by Peter
Kjellerstedt.

18 years agoTeach bloatometer about .rodata, and tweak the display into something that
Rob Landley [Mon, 29 May 2006 20:56:27 +0000 (20:56 -0000)]
Teach bloatometer about .rodata, and tweak the display into something that
has a better chance of getting merged.

18 years ago- cleanup conftest temporary files.
Bernhard Reutner-Fischer [Mon, 29 May 2006 20:01:52 +0000 (20:01 -0000)]
- cleanup conftest temporary files.

18 years ago- DEBUG_INIT -> config CONFIG_DEBUG_INIT
Bernhard Reutner-Fischer [Mon, 29 May 2006 13:08:35 +0000 (13:08 -0000)]
- DEBUG_INIT -> config CONFIG_DEBUG_INIT

18 years ago- landley, remove this?
Bernhard Reutner-Fischer [Mon, 29 May 2006 13:01:37 +0000 (13:01 -0000)]
- landley, remove this?

18 years ago- set_term() doesn't need an argument. (Like r15220 this is just compile-tested)
Bernhard Reutner-Fischer [Mon, 29 May 2006 12:57:52 +0000 (12:57 -0000)]
- set_term() doesn't need an argument. (Like r15220 this is just compile-tested)
   text    data     bss     dec     hex filename
   5432      32       8    5472    1560 init/init.o.r15220
   5426      32       8    5466    155a init/init.o

18 years ago- move common code into new open_new_terminal() function;
Bernhard Reutner-Fischer [Mon, 29 May 2006 12:54:16 +0000 (12:54 -0000)]
- move common code into new open_new_terminal() function;
   text    data     bss     dec     hex filename
 864434   10244  645892 1520570  1733ba busybox_old
 864386   10244  645892 1520522  17338a busybox_unstripped

add/remove: 1/0 grow/shrink: 0/2 up/down: 92/-143 (-51)
function                                     old     new   delta
open_new_terminal                              -      92     +92
exec_signal                                  345     276     -69
run                                         1527    1453     -74
Again, 48 != 51 so let's trust the former..

18 years ago- just whitespace
Bernhard Reutner-Fischer [Mon, 29 May 2006 12:12:45 +0000 (12:12 -0000)]
- just whitespace

18 years ago- ls: remove unused variable
Bernhard Reutner-Fischer [Mon, 29 May 2006 12:10:23 +0000 (12:10 -0000)]
- ls: remove unused variable
- dpkg.c, diff: use xstat
   text    data     bss     dec     hex filename
 848823    9100  645216 1503139  16efa3 busybox_old
 848679    9100  645216 1502995  16ef13 busybox_unstripped
bloatcheck is completely useless as it sees -79 for this, which is bogus.

18 years agoMore size shrinkage.
Rob Landley [Mon, 29 May 2006 07:42:02 +0000 (07:42 -0000)]
More size shrinkage.

18 years agoSize reductions, mostly switching things to use libbb functions.
Rob Landley [Mon, 29 May 2006 06:43:55 +0000 (06:43 -0000)]
Size reductions, mostly switching things to use libbb functions.

18 years agoAdd SWAP_LE?? and SWAP_BE?? macros, and make things use them. Converts values
Rob Landley [Mon, 29 May 2006 05:51:12 +0000 (05:51 -0000)]
Add SWAP_LE?? and SWAP_BE?? macros, and make things use them.  Converts values
to/from little endian or big endian, which is a NOP if that's what the current
platform already is.

18 years agoThis is not C++.
Rob Landley [Mon, 29 May 2006 05:21:29 +0000 (05:21 -0000)]
This is not C++.

18 years agoA few new todo items from email with Bernhard.
Rob Landley [Mon, 29 May 2006 05:06:06 +0000 (05:06 -0000)]
A few new todo items from email with Bernhard.

18 years agoRemove _() and N_() from platform.h. #define them as NOP macros in the two
Rob Landley [Mon, 29 May 2006 05:00:44 +0000 (05:00 -0000)]
Remove _() and N_() from platform.h.  #define them as NOP macros in the two
files still using them.  I didn't remove them from e2fsck.c to avoid stomping
pending cleanup patches from Garrett, and I didn't bother to remove them from
fdisk.c because that entire file needs to be rewritten from scratch.

18 years agoRemove _().
Rob Landley [Mon, 29 May 2006 04:49:55 +0000 (04:49 -0000)]
Remove _().

18 years agoRemove _().
Rob Landley [Mon, 29 May 2006 04:37:28 +0000 (04:37 -0000)]
Remove _().

18 years agoI changed last10 to last25, and just bumped it up to 50 so I renamed it
Rob Landley [Sun, 28 May 2006 02:10:21 +0000 (02:10 -0000)]
I changed last10 to last25, and just bumped it up to 50 so I renamed it
"recent.html".  There's a symlink from each of the old names for the moment,
I'll zap those eventually...

18 years agoFix hdparm to use PRIu64 instead of typecasting to long long (which is 128 bits
Rob Landley [Sun, 28 May 2006 01:56:08 +0000 (01:56 -0000)]
Fix hdparm to use PRIu64 instead of typecasting to long long (which is 128 bits
on 64 bit platforms), and move #include <inttypes.h> to libbb.h.

18 years agoPut the ending quote on the help entry...
Rob Landley [Sun, 28 May 2006 01:40:26 +0000 (01:40 -0000)]
Put the ending quote on the help entry...

18 years agoFix some warnings in allbareconfig.
Rob Landley [Sun, 28 May 2006 01:19:06 +0000 (01:19 -0000)]
Fix some warnings in allbareconfig.

18 years agoMy first bout of untangling udhcp. Make lots of gratuitous #defines go
Rob Landley [Sun, 28 May 2006 01:06:36 +0000 (01:06 -0000)]
My first bout of untangling udhcp.  Make lots of gratuitous #defines go
away, substitutie BB_VER for an external VERSION, use busybox CONFIG symbols
rather than checking for them then defining others, etc.  Lots more cleanup
to do...

18 years agoMove portability stuff to platform.h, and clean up adjtimex.c a bit while
Rob Landley [Sat, 27 May 2006 22:08:01 +0000 (22:08 -0000)]
Move portability stuff to platform.h, and clean up adjtimex.c a bit while
I'm in the area.

18 years agoConsolidate #include <sys/time.h> so libbb.h does it.
Rob Landley [Sat, 27 May 2006 21:42:58 +0000 (21:42 -0000)]
Consolidate #include <sys/time.h> so libbb.h does it.

18 years agoUpdate FAQ to mention make baseline when talking about make bloatcheck.
Rob Landley [Sat, 27 May 2006 21:30:34 +0000 (21:30 -0000)]
Update FAQ to mention make baseline when talking about make bloatcheck.

18 years agoConsolidate the .PHONY targets and add "make baseline" for bloatcheck.
Rob Landley [Sat, 27 May 2006 21:24:16 +0000 (21:24 -0000)]
Consolidate the .PHONY targets and add "make baseline" for bloatcheck.

18 years ago- use the defines CURRENT_TTY and VC_1; also shrink xargs a tiny bit:
Bernhard Reutner-Fischer [Sat, 27 May 2006 09:36:43 +0000 (09:36 -0000)]
- use the defines CURRENT_TTY and VC_1; also shrink xargs a tiny bit:
add/remove: 0/0 grow/shrink: 2/6 up/down: 5/-37 (-32)
function                                     old     new   delta
static.add_interface                         271     274      +3
static.glob3                                  36      38      +2
static.renice_main                           409     408      -1
static.readcmd                              1083    1081      -2
static.changepath                            176     174      -2
static.parse_opts                             56      51      -5
static.rx_main                              1095    1088      -7
static.xargs_main                            759     739     -20
----------------------------------------------------------------
Result                                                :-)    -32

18 years agoChange llist_add_* to take the address of the list rather than returning the new
Rob Landley [Fri, 26 May 2006 23:44:51 +0000 (23:44 -0000)]
Change llist_add_* to take the address of the list rather than returning the new
head, and change all the callers.

18 years agoSlight sanity fix: data is void *, not char *. And it's called data
Rob Landley [Fri, 26 May 2006 23:00:10 +0000 (23:00 -0000)]
Slight sanity fix: data is void *, not char *.  And it's called data
almost everywhere, so be consistent.

18 years ago- provide fallback defines for non-linux
Bernhard Reutner-Fischer [Fri, 26 May 2006 20:34:02 +0000 (20:34 -0000)]
- provide fallback defines for non-linux

18 years ago- provide fallback defines for non-linux
Bernhard Reutner-Fischer [Fri, 26 May 2006 20:33:12 +0000 (20:33 -0000)]
- provide fallback defines for non-linux

18 years ago- add central knob to turn off getopt_long everywhere. EXPERIMENTAL!
Bernhard Reutner-Fischer [Fri, 26 May 2006 20:19:22 +0000 (20:19 -0000)]
- add central knob to turn off getopt_long everywhere. EXPERIMENTAL!
  Adds "Enable getopt long" under "General options", default y.
  Send patches to fix getopt_ulflags and run_parts.c if you turn this off..
  See http://busybox.net/lists/busybox/2006-May/021828.html for a start to run-parts

18 years ago- use strtol instead of strtoll if the latter does not exist
Bernhard Reutner-Fischer [Fri, 26 May 2006 19:58:58 +0000 (19:58 -0000)]
- use strtol instead of strtoll if the latter does not exist
- add and use wrapper for attribute
- add and use replacement for vasprintf if it is unavailable

18 years ago- introduce and use bb_path_wtmp_file for portability (saves 11 Bytes).
Bernhard Reutner-Fischer [Fri, 26 May 2006 14:41:40 +0000 (14:41 -0000)]
- introduce and use bb_path_wtmp_file for portability (saves 11 Bytes).
- fix last.c to also look at the double-underscore UT_ defines.

18 years ago- use portability wrapper define bb_setpgrp.
Bernhard Reutner-Fischer [Fri, 26 May 2006 14:24:56 +0000 (14:24 -0000)]
- use portability wrapper define bb_setpgrp.

18 years ago- work around bug in gcc-3.4.x on ARM
Bernhard Reutner-Fischer [Fri, 26 May 2006 14:05:48 +0000 (14:05 -0000)]
- work around bug in gcc-3.4.x on ARM

18 years ago- move llist to top of file so we can potentially use it in all prototypes easily
Bernhard Reutner-Fischer [Fri, 26 May 2006 13:54:07 +0000 (13:54 -0000)]
- move llist to top of file so we can potentially use it in all prototypes easily

18 years ago- make find_mount_point conditional on df and eject; Fixes `make hosttools´ on
Bernhard Reutner-Fischer [Fri, 26 May 2006 13:38:43 +0000 (13:38 -0000)]
- make find_mount_point conditional on df and eject; Fixes `make hosttools´ on
  hosts where mntent.h does not exist.

18 years ago- add workaround for tar being broken since it uses a non-portable constant.
Bernhard Reutner-Fischer [Fri, 26 May 2006 13:34:25 +0000 (13:34 -0000)]
- add workaround for tar being broken since it uses a non-portable constant.

18 years ago- use the SED given by the user
Bernhard Reutner-Fischer [Fri, 26 May 2006 13:28:39 +0000 (13:28 -0000)]
- use the SED given by the user
- use the portable `` instead of $(). There is no bbsh, so this is needed.

18 years ago- add target hosttools (see make help)
Bernhard Reutner-Fischer [Fri, 26 May 2006 13:22:57 +0000 (13:22 -0000)]
- add target hosttools (see make help)
- rename check_gcc to check_cc and pass the CC to use as arg#1; peruse check_cc for HOSTCFLAGS
- add and use check_strip
- add checks for {,no-}whole-archive and {start,end}-group LD flags

18 years ago- use ATTRIBUTE_ALWAYS_INLINE if requested
Bernhard Reutner-Fischer [Fri, 26 May 2006 13:13:11 +0000 (13:13 -0000)]
- use ATTRIBUTE_ALWAYS_INLINE if requested
- use shorter boilerplate while at it

18 years ago- move libc checks from busybox.h to platform.h
Bernhard Reutner-Fischer [Fri, 26 May 2006 13:10:10 +0000 (13:10 -0000)]
- move libc checks from busybox.h to platform.h
- add ATTRIBUTE_ALWAYS_INLINE, endian handling for DEC UNIX, some more
  compiler dependent defines to platform.h
- add conditional bb_setpgrp define to platform.h
- remove superfluous specifying args from "#define fdprintf dprintf"

18 years ago- patch from Tito to pretty print the overall delta
Bernhard Reutner-Fischer [Fri, 26 May 2006 10:29:40 +0000 (10:29 -0000)]
- patch from Tito to pretty print the overall delta

18 years agoYann E. MORIN. writes:
Mike Frysinger [Fri, 26 May 2006 02:05:19 +0000 (02:05 -0000)]
Yann E. MORIN. writes:
hdparm has a uint64_t passed to printf with a %llu format (which is ok, because
it's the same size), but without first being casted(?) to unsigned long long int.
Which is kinda funny because it's done OK on a few lines above...

18 years agoerrno and h_errno values are not the same #876 by rfelker
Mike Frysinger [Fri, 26 May 2006 01:56:42 +0000 (01:56 -0000)]
errno and h_errno values are not the same #876 by rfelker

18 years agofix spelling mistakes
Mike Frysinger [Fri, 26 May 2006 01:48:17 +0000 (01:48 -0000)]
fix spelling mistakes

18 years agocheck for queueing dependancy dependancies typos
Mike Frysinger [Fri, 26 May 2006 01:41:04 +0000 (01:41 -0000)]
check for queueing dependancy dependancies typos

18 years agoqueuing is spelled correctly
Mike Frysinger [Fri, 26 May 2006 01:38:19 +0000 (01:38 -0000)]
queuing is spelled correctly

18 years agoRich Felker suggested removing dprintf() from watch, and one thing led to
Rob Landley [Thu, 25 May 2006 23:02:40 +0000 (23:02 -0000)]
Rich Felker suggested removing dprintf() from watch, and one thing led to
another... This adds bb_xspawn() support, which does vfork/exec.  (I don't
know why using a static instead of a local adds ~40 bytes, but using
the local doesn't work...)

18 years agoAnd fix the error I just introduced. :)
Rob Landley [Thu, 25 May 2006 21:13:30 +0000 (21:13 -0000)]
And fix the error I just introduced. :)

18 years agoRemove pointless #define.
Rob Landley [Thu, 25 May 2006 21:12:40 +0000 (21:12 -0000)]
Remove pointless #define.

18 years agoPatch from Tito:
Rob Landley [Thu, 25 May 2006 18:53:06 +0000 (18:53 -0000)]
Patch from Tito:
Remove -v verbose -V version -q quiet flags
Remove parse_opt_v2();
Rename parse_opt_v3() to parse_xfermode()
Remove if_strcat()
Rearrange code in  dump_identity() to reduce size
Modify some strings so that they could be optimized by the compiler
Minor code clean up
Remove unneeded #define <shm.h>

#if BB_BIG_ENDIAN && !defined(__USE_XOPEN)
# define __USE_XOPEN
#endif    
Must be before #include <unistd.h> and #include "busybox.h"
(noticed by Bernhard Fischer <rep.nop@aon.at>)

Remove duplicate code in do_time()

18 years agoRemove a pointless #define.
Rob Landley [Thu, 25 May 2006 18:31:11 +0000 (18:31 -0000)]
Remove a pointless #define.

18 years agomake the testsuite a little less brittle:
Bernhard Reutner-Fischer [Thu, 25 May 2006 13:24:02 +0000 (13:24 -0000)]
make the testsuite a little less brittle:
- 'function fn_name\n{' breaks on older FreeBSD default shells, so use the more
  widely supported 'fn_name () {'. This needs more fixing..
- test for integers ought to use the proper operators
- test for strings ought to use quoting of the strings to be fair to strange
  implementations of test(1)
- make sure not to ignore return-codes != 0 from commands; Some shells exit
  immediately on this (much like explicitely requesting set -e in e.g. bash)

TODO:
*) Some older shells do not allow a space after the test-condition in an "if"
statement. This doesn't work:
if [ $status -ne 0 ] ; then
as opposed to this:
if [ $status -ne 0 ]; then
or this
if [ $status -ne 0 ]
then

*) strict spacing between commands. In some shells you have to say:
foo ; bar ; baz
The affected shells barf on stuff like ommitting the space, so this doesn't
work:
foo; bar   ;baz

*) $() vs. ``
The former isn't really portable as opposed to the latter.

*) fix frong assumption that the testsuite is run from the source-dir.
This is a complete misconception and renders the testsuite completely useless.

That said, i note that IMO a test-harness ought to do it's best to work in
a wide variety of environments, everything else defeats it's purpose.

18 years agoTwo more issues pointed out by Bernhard.
Rob Landley [Wed, 24 May 2006 17:58:00 +0000 (17:58 -0000)]
Two more issues pointed out by Bernhard.

18 years agoYann Morin spotted some leftover flotsam from halfway through the recent
Rob Landley [Wed, 24 May 2006 17:45:47 +0000 (17:45 -0000)]
Yann Morin spotted some leftover flotsam from halfway through the recent
llist.c interface change that accidentally got checked in.  My bad.

18 years agoIt takes _talent_ to implement a version of nohup that does everything except
Rob Landley [Tue, 23 May 2006 00:28:06 +0000 (00:28 -0000)]
It takes _talent_ to implement a version of nohup that does everything except
the actual blocking of HUP.

18 years ago- set x bit
Bernhard Reutner-Fischer [Sun, 21 May 2006 22:06:59 +0000 (22:06 -0000)]
- set x bit

18 years agoConvert to a global struct, the way sed was. Now I have two, I can work out
Rob Landley [Sun, 21 May 2006 18:33:27 +0000 (18:33 -0000)]
Convert to a global struct, the way sed was.  Now I have two, I can work out
the infrastructure to merge global structs into a union...

18 years agoMinor cleanups: Convert a few calloc() calls to xzalloc, remove unnecessary
Rob Landley [Sun, 21 May 2006 18:30:35 +0000 (18:30 -0000)]
Minor cleanups: Convert a few calloc() calls to xzalloc, remove unnecessary
memset, collate variable declarations...

18 years agoNew version of nohup that's much smaller, less paranoid, consistent,
Rob Landley [Sun, 21 May 2006 18:28:13 +0000 (18:28 -0000)]
New version of nohup that's much smaller, less paranoid, consistent,
vaguely portable, and licensed GPLv2 "or later".

18 years ago- fix wrong path to scripts/usage; It is in the builddir and not in the srcdir.
Bernhard Reutner-Fischer [Sun, 21 May 2006 15:05:49 +0000 (15:05 -0000)]
- fix wrong path to scripts/usage; It is in the builddir and not in the srcdir.
- flag .depend as .NOTPARALLEL to workaround bug in make-3.80 and make-3.81
Some more details on the second workaround (i.e. kludge):
make-3.80 silently fails, while make-3.81 is at least self-conscious enough to
admit:
include/usage_compressed.h: .config scripts/usage
#  Implicit rule search has not been done.
#  File does not exist.
#  File has not been updated.
#  Dependencies commands running (THIS IS A BUG).
#  commands to execute (from `/home/cow/src/bb2/Makefile', line 430):
        $(Q)$(SHELL) $(top_srcdir)/scripts/usage_compressed "$(top_builddir)/scripts" > $@

I'm giving up on trying to run the rest of the prerequisites of .depend in
parallel and just (try to) indicate that the prerequisites of .depend are
not to be run in parallel.

While this makes `make distclean defconfig;make -j´ work for me, it is not a
real -- longterm -- solution; FIXME

18 years ago- check for existing usage binary, so it get's more obvious that the toplevel
Bernhard Reutner-Fischer [Sun, 21 May 2006 10:45:01 +0000 (10:45 -0000)]
- check for existing usage binary, so it get's more obvious that the toplevel
  Makefile uses the wrong path to the binary (it's in top_builddir and not in top_srcdir)

18 years ago- there is no _usage.h anymore, so don't rm it in distclean.
Bernhard Reutner-Fischer [Sun, 21 May 2006 10:04:46 +0000 (10:04 -0000)]
- there is no _usage.h anymore, so don't rm it in distclean.