name resolution into IP or IPv6 only, otherwise
we take address family returned by host2sockaddr()
in lsa->sa.sa_family. IOW: "ping ::1" with do IPv6 ping,
"ping 127.0.0.1" will do IPv4 ping.
ping6 is aliased to "ping -6".
(will close all fd's > 2 on daemonization now)
getty: fix "getty -" support, and also do not try to chown/chmod "-"
telnetd: fix "lost ctty" bug
Yet another attempt on saner function names:
bb_sanitize_server_stdio(0/1) -> bb_sanitize_stdio() + bb_daemonize();
Since this is also needed for other applets like telnetd,
introduce generic driver for such things.
It even supports inetd-wait ('linger') mode, when inetd
hands out listen socket to child and waits to it to die,
instead of handing out accepted socket and continuing
listening itself (nowait mode).
Code growth ~200 bytes.
NB: our inetd doesn't support wait mode yet (or mabe it is buggy).
but entire applet can be orders of magnitude smaller
if written as an inetd service.
So did that (#ifdef'ed out entire old version).
inetd version is less than 10% of old one!
function old new delta
packed_usage 22083 22105 +22
nobodystr 4 - -4
bind_ip_address 4 - -4
ident_substr 20 - -20
chmatch 22 - -22
movefd 25 - -25
skipchars 49 - -49
handlexitsigs 51 - -51
replyError 70 - -70
.rodata 158120 158024 -96
deleteConn 102 - -102
G 524 388 -136
conns 560 - -560
fakeidentd_main 1457 143 -1314
------------------------------------------------------------------------------
(add/remove: 0/10 grow/shrink: 1/3 up/down: 22/-2453) Total: -2431 bytes
Convert telnet to it. Now this works:
telnetd -b [::1]:1234 - bind to IPv6 non-standard port
telnet [::1]:1234 - connect to IPv6 non-standard port
telnet ::1 1234 - same
This does not require ANY ipv6-specific code in applets
(no struct sockaddr_in6. In fact, no sockaddr_in, too).
uidgid_get -> get_uidgid, add additional param
(numeric_ok). Make chown use it.
chown: fix "chown user: ...."
install: fix incorrect use of bb_xget[pw/gr]nam
* unpack: handle tar header fields which are not NUL terminated
* pack: handle 4+GB files correctly
* pack: refuse to store 101+ softlinks (was truncating link
target name)
* pack: mask mode with 07777
Use it where appropriate.
Stop scanning /etc/passwd *for every process*!!! (uid->username)
top: reduce memory usage - we won't save unneeded fields
from /proc info anymore. Downside: ~+250 bytes of code
It is impossible to formulate sane ABI based on
size of ulong because it can be 32-bit or 64-bit.
Basically it means that you cannot portably use
more that 32 option chars in one call anyway...
Make it explicit.
generated binaries, to make copyright enforcement easier. Our liason with
them (Bradley Kuhn) suggested the following text:
> Copyright (C) YEAR-2006 Erik Andersen, Rob Landley, and others.
> Licensed under GPLv2. See source distribution for full notice.
And the busybox help message seems the least offensive place to put it. (At
some point in the future, I need to strip 128 bytes from the binary to feel
good about this, though. :)
Minor header file cleanup while I was there, since libbb.h already #includes
most of that stuff...
We've had -Werror in the tree for quite a while, so adding #warning
somewhat counterproductve way to make comments on the code that belong on the
mailing list anyway.
I don't know why patch decided to apply this even though it was already in
the tree. I thought the other hunks failed because they were totally
unrelated leakage from Bernhard's tree (which they are; was the a reason
for bundling them in with this fix? Do they have something to do with
the GCC 2.95 fix? I suspect they prevent me from backporting this
patch to 1.2.2 because the header consolidation into libbb.h hadn't
been done yet, and no I'm not fixing it up: if that's the case then
this patch won't be in 1.2.2 due to extraneous changes bundled with it that
prevent a clean backport without rolling a new patch).
option - it was going to return a special flag back to caller and
expecting caller to call it again with special parameter! Also
caller was charged with calling mount() syscall...
mount: mtab support was non-functional. Enabling it revealed serious bug
which is not fixed yet.
few new (unfinished) config options, which I intend to make hidden (but
enabled) when CONFIG_NITPICK is disabled. Getting the .config infrastructure
to do that is non-obvious, it seems...
teach scripts/individual new tricks. And while I'm at it, teach
scripts/individual other new tricks. Now builds 198 applets, some of which
I should teach it to hardlink together because they're really the same app...
were using "1" as one of the arguments anyway, and as for the rest a multiply
and a push isn't noticeably bigger than pushing two arguments on the stack.
things like xasprintf() into xfuncs.c, remove xprint_file_by_name() (it only
had one user), clean up lots of #includes... General cleanup pass. What I've
been doing for the last couple days.
And it conflicts! I've removed httpd.c from this checkin due to somebody else
touching that file. It builds for me. I have to catch a bus. (Now you know
why I'm looking forward to Mercurial.)
untangle them:
Rewrite u_signal_names() into get_signum() and get_signame(), plus trim the
signal list to that required by posix (they can specify the numbers for
the rest if they really need them). (This is preparatory cleanup for adding
a timeout applet like Roberto Foglietta wants.)
Export the itoa (added due to Denis Vlasenko, although it's not quite his
preferred implementation) from xfuncs.c so it's actually used, and remove
several other redundant implementations of itoa and utoa() in the tree.
what to exec. Add -f mode and a brief explanation of how to use it to replace
minicom. Add -l -l mode so you can turn any command into a server. And group
all of netcat's command line options under two CONFIG entries, so if you
disable both it doesn't use getopt at all.
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.
fallout due to the #include <sys/mount.h>. Removed that #include from various
applets and fixed up those that were unhappy when that #include was made
because they'd block copied stuff out of it. (Sigh.)
the start of the path. (This should be under the same config option as
the standalone shell, but right now that's buried in the shell menu.)
Also add the ability to specify CONFIG_BUSYBOX_EXEC_PATH with /proc/self/exe
as an overrideable default.
moved the contents of libbb/bb_echo.c back into coreutils/echo.c,
which is a more reasonable place for them than libbb. this
forces anyone who wants echo and test to be builtin to ash to
also have them available as applets. their cost is very small,
and the number of people who wouldn't want them as applets is
also very small.
added warning about shell builtins vs. CONFIG_FEATURE_SH_STANDALONE_SHELL,
which conflicts with their use.
thanks to nathanael copa for debugging help.
some string size optimization in test.c may have been lost with
this commit, but this is a good new baseline.
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.
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.
*/
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().
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.
- 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"
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...)
handle packets out of sequence if some data goes through the buffer and
some doesn't, B) it works on systems that can't handle aligned access,
C) we just have one code path to worry about.
While we're at it, sizeof() and RESERVE_CONFIG_BUFFER() really don't combine
well, which is why md5sum has been reading and processing data 4 bytes at a
time. I suspect that the existence of CONFIG_MD5_SIZE_VS_SPEED to do loop
unrolling and such in the algorithm was an attempt to work around that bug.
the side of the tree doesn't _COUNT_, and I will not ship it.
Udhcp was deleted shortly after I posted my philosophy for what should and
shouldn't go into busybox:
http://www.busybox.net/lists/busybox/2006-March/019484.html
I complained about the change t the time. I've complained repeatedly since.
But nobody felt like fixing it. External dependencies are something to be
minimized. I don't care about the ability for packages to build outside
busybox: something is either part of busybox, or it isn't. If I convert any
part of the external udhcp repository to use libbb, I've broken the external
package. Any random cleanups that touch that directory suddenly have to worry
about external dependencies that are NOT OUR PROBLEM. Therefore, that
directory is not and cannot be part of busybox. Wishful thinking isn't going
to change that. I will not ship something I can't maintain.
I'll try to get a new dhcp client and server in before the ship window closes,
but I have a half-dozen other projects pending. I'm sorry this happened, but
I'm not the one who removed it, and I'm not the one who ignored the project
maintainer's repeated complaints about the situation for the next month and a
half.
Here's my attempt at a mini diff applet - it's adapted from the code at
http://www.openbsd.org/cgi-bin/cvsweb/src/usr.bin/diff/, and only supports
unified diffs.
I've busyboxified everything to a reasonable degree, so I think the code is
suitable enough to be included, but there's still a fair bit of cleaning up
to be done.
For each CONFIG_SYMBOL, include/bb_config.h now has both ENABLE_SYMBOL
and USE_SYMBOL(x). ENABLE_SYMBOL is still always defined (1 or 0) so that
if(ENABLE) should optimize out when it's zero. The USE_SYMBOL(X) will only
splice in X if the symbol is defined, otherwise it'll be empty.
Thus we can convert this:
#ifdef CONFIG_ARGS
opt = bb_getopt_ulflags(argc, argv, "ab:c"
#ifdef CONFIG_THINGY
"d:"
#endif
, &bvalue
#ifdef CONFIG_THINGY
, &thingy
#endif
);
#endif
into this:
if (ENABLE_ARGS) {
opt = bb_getopt_ulflags(argc, argv, "ab:c" USE_THINGY("d:"), &bvalue
USE_THINGY(, &thingy));
}
And it should produce the same code.
Unlike the old versions in include/_usage.h, the new USE_SYMBOL(x) can handle
commas in its arguments (as shown above). (The _usage.h file is obsolete and
no longer generated.)
Nobody should need to include config.h directly anymore, bb_config.h should
define all the configuration stuff we need. Someday, the CONFIG_SYMBOL
versions should go away in favor of ENABLE_SYMBOL and USE_SYMBOL().
Thanks to vodz for the new version of bb_mkdep.c that works with function
macros.
Somehow while applying the bb_do_delay patch a change slipped
in libbb.h that broke compilation.
libbb.h Line 355
extern char bb_path_mtab_file[];
This conflicts with mtab_file.c
#if defined(CONFIG_FEATURE_MTAB_SUPPORT)
const char bb_path_mtab_file[] = "/etc/mtab";
#else
const char bb_path_mtab_file[] = "/proc/mounts";
#endif
Not buying it, eh?
I know I said new features before 1.1, but, well... (I was weak!)
The config file and hotplug modes aren't implemented yet. Might take a stab at
those tomorrow. (I _should_ go back to focusing on the bug triage list.)
CONFIG_ symbol in the kernel, and this clashes with busybox's CONFIG_TR
and CONFIG_WATCHDOG, causing applets.h to barf if they're not switched
on (since the broken headers don't affect kconfig or the makefiles).
Since such broken kernel headers are common enough to crop up every few
months, a simple work around is to move TR and WATCHDOG from CONFIG_
to ENABLE_ early.
messages, C) can show the current association (if any) when called
with only one argument. Update the documentation a lot too.
Remind me to add a test suite for this thing. I think I've figured out
how to handle root-only testsuites...
fixes bug #113 and satisfies a personal need at the same time.
output compares identically to util-linux version. (with
exception of whitespace differences on last lines of output with
non-uniform length, which are neither fixed nor worsened by this
change.)
- change llist_add_to_end as proposed by vodz in http://busybox.net/lists/busybox/2005-September/016411.html
- remove unneeded includes, add short boilerplate and copyright to llist.c
- move COMM_LEN from find_pid_by_name to libbb.h and use it in procps_status_t
- add reverse_pidlist() to find_pid_by_name. Will be needed for pidof.
- move to ENABLE_ and use shorter boilerplate.
sizes without selinux-support:
text data bss dec hex filename
356 0 0 356 164 procps/ps.o.with-w
300 0 0 300 12c procps/ps.o.no-w
302 0 0 302 12e procps/ps.o.oorig
This fixes the warning, and makes the binary smaller out of sheer pique.
(Yes, since Manuel did this one it's nice tight code that took several
attempts to shrink, but I was ticked.)
Add the start of a test for uniq; this is about the first 1/3 of the
tests we need for full susv3 coverage of uniq.
the result of the ioctl so callers can tell if we have a tty. (0 means
we have a tty, nonzero means the ioctl couldn't find size info and we
fake 80x24. Really we should fake 80x25, but oh well...)
things down a bit, fixed a number of funky corner cases, added support for
several new features (things like mount --move, mount --bind, lazy unounts,
automatic detection of loop mounts, and so on). Probably broke several
other things, but it's fixable. (Bang on it, tell me what doesn't work for
you...)
Note: you no longer need to say "-o loop". It does that for you when
necessary.
Still need to add "user mount" support, which involves making mount suid. Not
too hard to do under the new infrastructure, just haven't done it yet...
The previous code had the following notes, that belong in the version
control comments:
- * 3/21/1999 Charles P. Wright <cpwright@cpwright.com>
- * searches through fstab when -a is passed
- * will try mounting stuff with all fses when passed -t auto
- *
- * 1999-04-17 Dave Cinege...Rewrote -t auto. Fixed ro mtab.
- *
- * 1999-10-07 Erik Andersen <andersen@codepoet.org>.
- * Rewrite of a lot of code. Removed mtab usage (I plan on
- * putting it back as a compile-time option some time),
- * major adjustments to option parsing, and some serious
- * dieting all around.
- *
- * 1999-11-06 mtab support is back - andersee
- *
- * 2000-01-12 Ben Collins <bcollins@debian.org>, Borrowed utils-linux's
- * mount to add loop support.
- *
- * 2000-04-30 Dave Cinege <dcinege@psychosis.com>
- * Rewrote fstab while loop and lower mount section. Can now do
- * single mounts from fstab. Can override fstab options for single
- * mount. Common mount_one call for single mounts and 'all'. Fixed
- * mtab updating and stale entries. Removed 'remount' default.
- *
0000072: Add applet to redirect console output via ioctl(..., TIOCCONS)
applet name changed to setconsole, since suse has a very similar
utility. better to treat differences as bugs than invent a new command.
since simply defining the prototype in an internal header file doesnt hurt anyone, lets not worry about the header file crap and just always prototype the inet6 functions
Rob Landley, and others.
Currently CONFIG options are defined or undefined, so we chop out code with
#ifdefs, ala:
#ifdef CONFIG_THING
stuff();
#endif
This creates a new header file, bb_config.h, which sets the CONFIG entry to 1
or 0, and lets us do:
if(CONFIG_THING) stuff();
And let the compiler do dead code elimination to get rid of it. (Note: #ifdef
will still work because for the 1 case it's a static const int, not a #define.)
with tweaks from Mike Frysinger and Rob Landley.
Note: this will still fail to umount a path that contains an ' or \ character.
Is it worth the extra size to filter for that?
To: busybox@mail.codepoet.org
Wrap sockaddr_in6 in a ifdef CONFIG_FEATURE_IPV6.
Include sys/types.h instead of asm/types.h.
Include netinet/if_ether.h instead of linux/if_ether.h
The ioctl request argument is an int, not an unsigned short.
> Hi,
> this is a first attempt of size optimization for zcip taking into account all
> the hints given so far on the list.
> I've applied just the more obvious busyboxifications so maybe it could be
> optimized more.
BTW: I've ripped out a lot of debug code and changed c++ // comments to /* */
as both were rather confusing for a newbie like me. ;-)
Sorry to the author for that.
I know that this makes mantaining the code easier, but I'm simple minded....
The second patch contains:
1) a size optimization for adduser.c
2) removes a warning about an unused variable in syslogd.c if CONFIG_FEATURE_REMOTE_LOG is not set
3)cosmetic fixes for addgroup_full_usage and adduser_full_usage
Ciao,
Tito
Hi Erik,
Hi to all,
This is part five of the my_get*id story.
I've tweaked a bit this two functions to make them more flexible,
but this changes will not affect existing code.
Now they work so:
1) my_getpwuid( char *user, uid_t uid, int bufsize)
if bufsize is > 0 char *user cannot be set to NULL
on success username is written on static allocated buffer
on failure uid as string is written to buffer and NULL is returned
if bufsize is = 0 char *user can be set to NULL
on success username is returned
on failure NULL is returned
if bufsize is < 0 char *user can be set to NULL
on success username is returned
on failure an error message is printed and the program exits
2) 1) my_getgrgid( char *group, uid_t uid, int bufsize)
if bufsize is > 0 char *group cannot be set to NULL
on success groupname is written on static allocated buffer
on failure gid as string is written to buffer and NULL is returned
if bufsize is = 0 char *group can be set to NULL
on success groupname is returned
on failure NULL is returned
if bufsize is < 0 char *group can be set to nULL
on success groupname is returned
on failure an error message is printed and the program exits
This changes were needed mainly for my new id applet.
It is somewhat bigger then the previous but matches the behaviour of GNU id
and is capable to handle usernames of whatever length.
BTW: at a first look it seems to me that it will integrate well (with just a few changes)
with the pending patch in patches/id_groups_alias.patch.
The increase in size is balanced by the removal of my_getpwnamegid.c
from libbb as this was used only in previous id applet and by size optimizations
made possible in whoami.c and in passwd.c.
I know that we are in feature freeze but I think that i've tested it enough
(at least I hope so.......).
Hi,
I've spent the half night staring at the devilish my_getpwuid and my_getgrgid functions
trying to find out a way to avoid actual and future potential buffer overflow problems
without breaking existing code.
Finally I've found a not intrusive way to do this that surely doesn't break existing code
and fixes a couple of problems too.
The attached patch:
1) changes the behaviour of my_getpwuid and my_getgrgid to avoid potetntial buffer overflows
2) fixes all occurences of this function calls in tar.c , id.c , ls.c, whoami.c, logger.c, libbb.h.
3) The behaviour of tar, ls and logger is unchanged.
4) The behavior of ps with somewhat longer usernames messing up output is fixed.
5) The only bigger change was the increasing of size of the buffers in id.c to avoid
false negatives (unknown user: xxxxxx) with usernames longer than 8 chars.
The value i used ( 32 chars ) was taken from the tar header ( see gname and uname).
Maybe this buffers can be reduced a bit ( to 16 or whatever ), this is up to you.
6) The increase of size of the binary is not so dramatic:
size busybox
text data bss dec hex filename
239568 2300 36816 278684 4409c busybox
size busybox_fixed
text data bss dec hex filename
239616 2300 36816 278732 440cc busybox
7) The behaviour of whoami changed:
actually it prints out an username cut down to the size of the buffer.
This could be fixed by increasing the size of the buffer as in id.c or
avoid the use of my_getpwuid and use getpwuid directly instead.
Maybe this colud be also remain unchanged......
Please apply if you think it is ok to do so.
The diff applies on today's cvs tarball (2004-08-25).
Thanks in advance,
Ciao,
Tito
