Rob Landley
ef10d52745
Upgrade mdev to allow commands to be run on create/delete.
...
Both Jason Schoon and Giuseppe Ciotta deserve credit for this, I used elements
of both. It's been upgraded so that you can specify that a given command
should run at create, at delete, or at both using different special characters
(@, $, and * respectively). It uses the system() method of running command
lines which means you can use environment variables on the command line (it
sets $MDEV to the name of the current device being created/deleted, which is
useful if you matched it via regex), and the documentation warns that you need
a /bin/sh to make that work, so you probably want to pick a default shell.
2006-06-26 14:11:33 +00:00
Rob Landley
cf7577d417
Missed one. Based on suggestions from Erik Hovland.
2006-06-25 22:59:31 +00:00
Rob Landley
11c7a7bed6
Various cleanups I made while going through Erik Hovland's patch submissions,
...
some of which are even from him. :)
2006-06-25 22:39:24 +00:00
Bernhard Reutner-Fischer
f087798e8b
- convert old-style parameter declarations (K&R) to modern style.
2006-06-25 22:08:53 +00:00
Rob Landley
a892bf8284
Whitespace. (Vim's auto-indent is kind of annoying at times, like "daytime"
...
and "nighttime"...)
2006-06-25 15:37:28 +00:00
Rob Landley
934da82913
Fix a possible race condition if two processes try to claim the same loop
...
device at the same time. We should only CLR_FD if the set status fails,
not if the SET_FD fails.
2006-06-25 15:29:12 +00:00
Rob Landley
768945b762
A few patches from Erik Hovland, turning strncpy() into safe_strncpy() and
...
removing some unnecessary code.
2006-06-25 00:34:52 +00:00
Rob Landley
641c3537c0
Add a test from Glenn.
2006-06-24 22:55:13 +00:00
Rob Landley
e1ee482684
Document mount's new error return conventions.
2006-06-24 22:39:19 +00:00
Rob Landley
c983274565
The logic to make cp -d or -P treat things like regular files should only
...
trigger for symlinks, not for device nodes. This should fix "cp -a /dev ."
to work as expected (when run by root, anyway).
While I was there, cleanup headers and make an #ifdef go away...
2006-06-24 21:27:36 +00:00
Rob Landley
5d9c42aadc
Slowly pulling more common headers into libbb.h. Andre pointed out that
...
older uClibc didn't include sys/socket.h from netinet/in.h, so add an
explicit #include for that too...
2006-06-24 17:55:02 +00:00
Rob Landley
53433b3ca1
Clean up.
2006-06-22 22:28:29 +00:00
Rob Landley
1fa4a9448b
Patch from Shaun Jackman to replace CFLAGS_EXTRA with .config.mak
2006-06-22 22:05:00 +00:00
Rob Landley
67d4943393
CONFIG_DHCP -> CONFIG_APP_DHCP.
2006-06-22 18:48:34 +00:00
Mike Frysinger
135cee3741
Jean Wolter writes: modprobe checks, whether a module is already loaded. The function used for this currently always returns 0.
2006-06-21 23:03:37 +00:00
Bernhard Reutner-Fischer
35492137f8
- add note about strict ordering
2006-06-21 18:19:53 +00:00
Mike Frysinger
c2ad4f5d27
Jean Wolter writes: ash uses bsearch to search in the list of built-in commands, therefore the list of built-in commands has to be ordered. Revision 15077 introduced the source built-in and destroyed this order. And now ash doesn't find the shift command anymore.
2006-06-21 18:04:49 +00:00
Rob Landley
0cc2c2c289
Patch from Paul Fox to improve mount's error handling behavior, which I beat
...
on a bit.
2006-06-21 03:53:33 +00:00
Rob Landley
dbfb5a3cb7
Andre (armcc2200@yahoo) patched a bug where successful mounts could sometimes
...
produce an error, due to a missing rc assignment.
2006-06-21 02:39:57 +00:00
Rob Landley
399d45f1c6
Aurelien Jacobs pointed out inline functions that could be removed. Doing so
...
shrank the code a bit.
2006-06-21 01:49:17 +00:00
Rob Landley
e5e1a10cd3
Remove pointless "inline" wrapper around get_terminal_width_height().
2006-06-21 01:15:36 +00:00
Rob Landley
2d6af16035
Zubicaray reported a bug in vi that causes it to eat 100% cpu when you close
...
an xterm it's running in. The vi signal behavior would catch and restarts lots
of signals, like SIGHUP, that should just kill the thing. (Leftover behavior
from when it would segfault all the time.) Filtered out the more obviously
bad ones. If it segfaults, we should find and fix the problem.
2006-06-21 00:52:31 +00:00
Mike Frysinger
b049c0ec19
only enable count_bit functions when actually needed to fix warning reported by Bernhard Fischer
2006-06-20 23:03:27 +00:00
Rob Landley
b13fee4cbb
Since rangecoder is just a bunch of C functions, move it into the one user
...
(decompress_unlzma.c). Also a slight #include cleanup, and I've been meaning
to put #include <unistd.h> into libbb.h since it's so darn common...
2006-06-20 22:38:00 +00:00
Rob Landley
5483de1cb7
Cleanup patch from Shaun Jackman converting %m to perror.
2006-06-20 21:35:26 +00:00
Rob Landley
9fe801e0d2
Patch by Robert P. Day to remove obsolete alias code that the 2.4 and 2.6
...
Linux kernels don't do anymore.
2006-06-20 21:13:29 +00:00
Mike Frysinger
14ff19b28a
assign default debug level to the MSHDEBUG define
2006-06-20 20:37:01 +00:00
Rob Landley
2b8a05a775
I broke allbareconfig with half-finished cleanups I'd forgotten weren't done.
...
Got both to a decent stopping point. (There's a lot more cleanup to do on ls.c
but now is not the time...)
2006-06-20 17:43:01 +00:00
Rob Landley
8437ede6c2
The type changed too.
2006-06-20 16:43:05 +00:00
Peter Kjellerstedt
5ab8f7de4f
Use the new names for tty_baud_to_value() and tty_value_to_baud() again.
2006-06-20 16:35:37 +00:00
Rob Landley
57c1f73dd5
Revert the last two patches to go back to a state before this file was
...
controversial.
2006-06-20 16:17:24 +00:00
Rob Landley
2818b292fb
Patch by Robert P. Day to make structs use c99 initializers.
2006-06-20 15:52:52 +00:00
Rob Landley
2a613d7360
Patch from Robert P. J. Day cleaning up some unnecessary preprocessor
...
directives.
2006-06-20 14:57:00 +00:00
Rob Landley
352e117c6b
Amrein-Marie Christophe pointed out that Mandrake changed its name.
2006-06-20 14:09:20 +00:00
Rob Landley
7a260f01ce
Make some 64 bit warnings go away on x86-64.
2006-06-19 03:20:03 +00:00
Rob Landley
290fcb4213
Undo all of the ugliness and some of the bloat from 15412.
2006-06-18 23:59:03 +00:00
Rob Landley
ea224be6aa
skip_whitespace() shouldn't claim its return value is const, it doesn't know
...
that and callers wind up typecasting it back.
2006-06-18 20:20:07 +00:00
Rob Landley
14d7065ef1
Make this slightly more readable, and expose the hypocrasy of a comment that
...
says "our implementation makes it impossible to use blocksizes smaller than
22 octets" right above a check for blocksize < 8.
2006-06-18 15:23:13 +00:00
Mike Frysinger
b000586d02
danf writes: stty doesn't support baud rates > 460800 bps, the attached patch adds support for all baud rates supported by Linux 2.6.16 on x86.
2006-06-17 22:41:28 +00:00
Rob Landley
fe3d844bb1
Make it so Erik has to forward me less email.
2006-06-16 22:29:03 +00:00
Bernhard Reutner-Fischer
fa61ca7591
- add cross-reference generation target.
...
Experimental.
2006-06-16 17:19:39 +00:00
Bernhard Reutner-Fischer
6d82f94681
- do away with the fwd-decls of functions. Should not contain any code changes
...
AFAICT.
2006-06-16 16:37:07 +00:00
Rob Landley
d9a761d9c3
Revert svn 15333. If you vfork, you must either exec or exit. (The parent is
...
blocked until you do so.)
2006-06-16 16:35:53 +00:00
Rob Landley
1449a2014a
This directory was fallout from the great feature freeze of 2003, which led
...
up to the 1.00 release. I just moved what was left of it to
http://busybox.net/~landley/pending because it does _not_ belong in the
tree anymore.
2006-06-16 15:08:59 +00:00
Bernhard Reutner-Fischer
747041955e
- hide CONFIG_GETOPT_LONG for the 1.2.0 release. Turned on per default.
...
I will revisit this and make sure that run-parts and ifupdown are fixed; hdparm needs treatment too..
2006-06-16 14:03:23 +00:00
Bernhard Reutner-Fischer
f5efa7bfeb
- add patch to provide a target 'html' to build a cross-reference of the source
2006-06-16 12:45:29 +00:00
Bernhard Reutner-Fischer
fd4b61d5f3
- add needed includes and include busybox.h needed for the ATTRIBUTE define.
2006-06-16 12:28:39 +00:00
Rob Landley
bc8c50351f
Remove warnings caused by #include <sys/mount.h> in platform.h. Apparently
...
we redefine a lot of stuff from standard header files...
2006-06-16 04:25:19 +00:00
Rob Landley
9947a244d4
Patch from Rich Felker:
...
- duplicated format bit 14 forced LIST_EXEC always-on
- -p option was behaving just like -F (SUSv3 says -p only shows / for
dirs, not other special symbols)
- tests for LIST_EXEC in color functions were nonsense (constant
tests). i assume they were supposed to be (all_fmt & LIST_EXEC)
however having coloring of executable files depend on -F seems
undesirable.
2006-06-15 22:11:10 +00:00
Bernhard Reutner-Fischer
1761b9dabe
- use shorter boilerplate, more accurate header guard
...
- don't use multi-line string literals
2006-06-15 21:56:28 +00:00