Rob Landley [Thu, 8 Jun 2006 14:11:36 +0000 (14:11 -0000)]
Some day, maybe gcc will be able to optimize out static functions that are
never used so we don't have to #ifdef them. Wouldn't that be nice?
Mike Frysinger [Wed, 7 Jun 2006 21:58:57 +0000 (21:58 -0000)]
need prototype for new vfork_daemon()
Mike Frysinger [Wed, 7 Jun 2006 21:48:43 +0000 (21:48 -0000)]
initial support for no-mmu systems
Mike Frysinger [Wed, 7 Jun 2006 21:48:30 +0000 (21:48 -0000)]
sometimes daemonizing and vfork()/exit() is ok
Mike Frysinger [Wed, 7 Jun 2006 21:47:34 +0000 (21:47 -0000)]
bb_xdaemon() isnt available on no-mmu systems
Mike Frysinger [Wed, 7 Jun 2006 21:37:59 +0000 (21:37 -0000)]
comment all fields
Mike Frysinger [Wed, 7 Jun 2006 21:37:49 +0000 (21:37 -0000)]
hint at where the Magic comes from
Bernhard Reutner-Fischer [Wed, 7 Jun 2006 21:26:09 +0000 (21:26 -0000)]
- do away with the comment that noted that the current implementation
unly supports linux.
Rob Landley [Wed, 7 Jun 2006 21:22:42 +0000 (21:22 -0000)]
Fix from Tito to read from stdin only when it's not a tty.
Bernhard Reutner-Fischer [Wed, 7 Jun 2006 20:30:01 +0000 (20:30 -0000)]
- spelling
Mike Frysinger [Wed, 7 Jun 2006 20:24:34 +0000 (20:24 -0000)]
add reference to proc(5)
Bernhard Reutner-Fischer [Wed, 7 Jun 2006 20:17:41 +0000 (20:17 -0000)]
- reuse strings and messages. Saves about 600B
Rob Landley [Wed, 7 Jun 2006 20:11:53 +0000 (20:11 -0000)]
Consolidate devfs garbage and mark it as obsolete.
Mike Frysinger [Wed, 7 Jun 2006 20:06:24 +0000 (20:06 -0000)]
use bb_xbind/bb_xlisten
Mike Frysinger [Wed, 7 Jun 2006 20:04:13 +0000 (20:04 -0000)]
whitespace
Mike Frysinger [Wed, 7 Jun 2006 20:03:13 +0000 (20:03 -0000)]
whitespace
Mike Frysinger [Wed, 7 Jun 2006 18:12:27 +0000 (18:12 -0000)]
add a --noclobber flag
Mike Frysinger [Wed, 7 Jun 2006 18:08:25 +0000 (18:08 -0000)]
add support for cleaning out previous busybox install
Mike Frysinger [Wed, 7 Jun 2006 17:27:46 +0000 (17:27 -0000)]
dont install emtpy libdirs
Mike Frysinger [Wed, 7 Jun 2006 17:24:29 +0000 (17:24 -0000)]
dont mix tabs and spaces, kthx
Mike Frysinger [Wed, 7 Jun 2006 16:31:05 +0000 (16:31 -0000)]
remove cvs $Id tag
Bernhard Reutner-Fischer [Wed, 7 Jun 2006 15:44:59 +0000 (15:44 -0000)]
- add applet taskset to set/retrieve the CPU affinity of a process
text data bss dec hex filename
584 0 0 584 248 taskset.o.gcc-2.95
509 0 0 509 1fd taskset.o.gcc-3.3
505 0 0 505 1f9 taskset.o.gcc-3.4
506 0 0 506 1fa taskset.o.gcc-4.0
498 0 0 498 1f2 taskset.o.gcc-4.1
495 0 0 495 1ef taskset.o.gcc-4.2-HEAD
Mike Frysinger [Wed, 7 Jun 2006 14:25:22 +0000 (14:25 -0000)]
Tito writes: this patch fixes two potential bugs with the -i -I flags that could be reset during option parsing
Bernhard Reutner-Fischer [Wed, 7 Jun 2006 13:31:59 +0000 (13:31 -0000)]
- fix the build if we have ftpput without ftpget (speeling reeoe)
Thanks to Stephane Billiart
Rob Landley [Wed, 7 Jun 2006 00:27:25 +0000 (00:27 -0000)]
Callers to identify() converted the endianness of the buffer. So did
identify(). This meant big endian systems had a NUXI problem. Removed
the redundant conversion from the callers, and made some in-passing cleanups
while I was there.
Rob Landley [Tue, 6 Jun 2006 22:59:37 +0000 (22:59 -0000)]
Remove a config entry that leaked in from my tree before I decided to
implement cat -v as a separate applet.
Rob Landley [Tue, 6 Jun 2006 19:50:16 +0000 (19:50 -0000)]
Whack the one last warning in make allbareconfig...
Mike Frysinger [Tue, 6 Jun 2006 06:30:32 +0000 (06:30 -0000)]
import support for microblaze relocations from uClinux-dist
Mike Frysinger [Tue, 6 Jun 2006 06:26:12 +0000 (06:26 -0000)]
make the default prompt suck less when using fancy prompts
Mike Frysinger [Tue, 6 Jun 2006 06:19:19 +0000 (06:19 -0000)]
fix from uClinux-dist for proper exit status if reading /proc/modules failed
Mike Frysinger [Tue, 6 Jun 2006 06:15:52 +0000 (06:15 -0000)]
merge blackfin/microblaze from uClinux-dist
Mike Frysinger [Tue, 6 Jun 2006 06:08:34 +0000 (06:08 -0000)]
use vfork() instead of vork() since the only thing we do is execve() an app after fork (allows time to work on no-mmu)
Mike Frysinger [Tue, 6 Jun 2006 06:00:20 +0000 (06:00 -0000)]
only check __GNU_LIBRARY__ if it is actually defined
Rob Landley [Mon, 5 Jun 2006 17:35:24 +0000 (17:35 -0000)]
Header cleanup on two more networking files (move libbb.h to the top and
remove #includes that libbb.h already does), plus a minor cleanup of
libbb.h to move #includes towards the top of the file where we can see 'em.
Rob Landley [Mon, 5 Jun 2006 17:32:44 +0000 (17:32 -0000)]
Confirming it's not needed.
Bernhard Reutner-Fischer [Mon, 5 Jun 2006 16:00:12 +0000 (16:00 -0000)]
- add note about need to unify the 4 itoa() implementations.
Bernhard Reutner-Fischer [Sun, 4 Jun 2006 18:40:48 +0000 (18:40 -0000)]
- add testcase for grep bug (busybox.net/bugs/view.php?id=887)
The patch for bug #887 seems to work for me..
Bernhard Reutner-Fischer [Sun, 4 Jun 2006 17:58:19 +0000 (17:58 -0000)]
- add note about ingroup and is_a_group_member
Bernhard Reutner-Fischer [Sun, 4 Jun 2006 16:05:02 +0000 (16:05 -0000)]
- add note about removing superfluous fmt specifiers
Bernhard Reutner-Fischer [Sun, 4 Jun 2006 15:37:59 +0000 (15:37 -0000)]
- note to cleanup fprintf(2,...) -> bb_error_msg(...) and
perror("bar") -> bb_perror_msg("bar");
Bernhard Reutner-Fischer [Sat, 3 Jun 2006 22:45:37 +0000 (22:45 -0000)]
- use bb_msg_{read,write}_error where appropriate.
text data bss dec hex filename
825015 9100 645216
1479331 1692a3 busybox.old
824919 9100 645216
1479235 169243 busybox
Bernhard Reutner-Fischer [Sat, 3 Jun 2006 21:40:11 +0000 (21:40 -0000)]
- use bb functions instead of handrolling stuff.
- add XXX question about supposedly superfluous else clause
text data bss dec hex filename
5918 0 0 5918 171e editors/sed.o.orig
5901 0 0 5901 170d editors/sed.o
Rob Landley [Sat, 3 Jun 2006 21:23:20 +0000 (21:23 -0000)]
Remove #ifdeffed out code, both KEEP_UNUSED and #if 0.
Bernhard Reutner-Fischer [Sat, 3 Jun 2006 21:19:09 +0000 (21:19 -0000)]
- remove superfluous call of stack_machine(0) upon exit.
Bernhard Reutner-Fischer [Sat, 3 Jun 2006 20:53:18 +0000 (20:53 -0000)]
- no need to have timer_duration in the data section.
- cleanup a little bit while at it.
Bernhard Reutner-Fischer [Sat, 3 Jun 2006 20:09:02 +0000 (20:09 -0000)]
- use bb_msg_write_error in 3 places
- whitespace
Bernhard Reutner-Fischer [Sat, 3 Jun 2006 19:49:21 +0000 (19:49 -0000)]
- move #include busybox.h to the very top so we pull in the config
and eventual platform specific includes in early.
Bernhard Reutner-Fischer [Sat, 3 Jun 2006 19:35:15 +0000 (19:35 -0000)]
- move #include busybox.h to the very top so we pull in the config
and eventual platform specific includes in early.
- remove two supposedly superfluous newlines from ...error_msg() in modprobe
and use shorter boilerplate while at it.
Bernhard Reutner-Fischer [Sat, 3 Jun 2006 19:08:49 +0000 (19:08 -0000)]
- patch from Yann E. Morin: makes modprobe understand shell patterns
(especially '*') in module aliases, such as:
"alias usb:v0582p0075d*dc*dsc*dp*ic*isc*ip* snd_usb_audio"
Fixes bug #889
842162 10244 645924
1498330 16dcda busybox.old-4.1.
20060603-1948
842178 10244 645924
1498346 16dcea busybox.new-4.1.
20060603-1948
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.
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.
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.
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.
Rob Landley [Fri, 2 Jun 2006 20:53:38 +0000 (20:53 -0000)]
Header cleanup: don't #include headers that libbb.h already includes.
Rob Landley [Fri, 2 Jun 2006 19:03:01 +0000 (19:03 -0000)]
#ifdef reduction.
Bernhard Reutner-Fischer [Fri, 2 Jun 2006 10:43:17 +0000 (10:43 -0000)]
- fix CONFIG_DEBUG_INIT compile failure
Mike Frysinger [Fri, 2 Jun 2006 06:30:30 +0000 (06:30 -0000)]
use xmalloc()
Paul Fox [Fri, 2 Jun 2006 03:37:16 +0000 (03:37 -0000)]
fix fdflish typo (reported by erik hovland)
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..
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
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.
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.
Rob Landley [Thu, 1 Jun 2006 00:51:38 +0000 (00:51 -0000)]
The common case APPLET() macro only needs three arguments.
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.
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.
Rob Landley [Wed, 31 May 2006 22:52:57 +0000 (22:52 -0000)]
Patch from Tito to remove long options from hdparm.
Rob Landley [Wed, 31 May 2006 22:24:33 +0000 (22:24 -0000)]
Use BB endianness macro.
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.
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.
*/
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.
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..
Bernhard Reutner-Fischer [Wed, 31 May 2006 14:12:51 +0000 (14:12 -0000)]
- make getopt_ulflags aware of CONFIG_GETOPT_LONG
Bernhard Reutner-Fischer [Wed, 31 May 2006 14:11:38 +0000 (14:11 -0000)]
- add CONFIG_FEATURE_WGET_LONG_OPTIONS
Bernhard Reutner-Fischer [Wed, 31 May 2006 13:31:16 +0000 (13:31 -0000)]
- convert xcalloc(1,x) to xzalloc
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
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().
Bernhard Reutner-Fischer [Wed, 31 May 2006 10:04:03 +0000 (10:04 -0000)]
- conditionally define BB_NOMMU
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.
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...
Rob Landley [Tue, 30 May 2006 18:31:37 +0000 (18:31 -0000)]
Note about BB_NOMMU.
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
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..
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
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
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.
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.
Bernhard Reutner-Fischer [Mon, 29 May 2006 20:01:52 +0000 (20:01 -0000)]
- cleanup conftest temporary files.
Bernhard Reutner-Fischer [Mon, 29 May 2006 13:08:35 +0000 (13:08 -0000)]
- DEBUG_INIT -> config CONFIG_DEBUG_INIT
Bernhard Reutner-Fischer [Mon, 29 May 2006 13:01:37 +0000 (13:01 -0000)]
- landley, remove this?
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
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..
Bernhard Reutner-Fischer [Mon, 29 May 2006 12:12:45 +0000 (12:12 -0000)]
- just whitespace
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.
Rob Landley [Mon, 29 May 2006 07:42:02 +0000 (07:42 -0000)]
More size shrinkage.
Rob Landley [Mon, 29 May 2006 06:43:55 +0000 (06:43 -0000)]
Size reductions, mostly switching things to use libbb functions.
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.
Rob Landley [Mon, 29 May 2006 05:21:29 +0000 (05:21 -0000)]
This is not C++.
Rob Landley [Mon, 29 May 2006 05:06:06 +0000 (05:06 -0000)]
A few new todo items from email with Bernhard.
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.
Rob Landley [Mon, 29 May 2006 04:49:55 +0000 (04:49 -0000)]
Remove _().