platform/upstream/busybox.git
18 years agoInstall links patch from Yann E. Morin. (Another thing hanging around in my
Rob Landley [Fri, 2 Dec 2005 18:27:39 +0000 (18:27 -0000)]
Install links patch from Yann E. Morin.  (Another thing hanging around in my
tree forever.  Tweaked the docs a bit.)

18 years agoIf we really _need_ an #ifdef like this, then we should be using the libbb
Rob Landley [Fri, 2 Dec 2005 17:57:23 +0000 (17:57 -0000)]
If we really _need_ an #ifdef like this, then we should be using the libbb
getopt infrastructure instead.

18 years agoAnother cleanup patch that's been in my tree for a while. Again I think it's
Rob Landley [Fri, 2 Dec 2005 17:55:45 +0000 (17:55 -0000)]
Another cleanup patch that's been in my tree for a while.  Again I think it's
originally from Bernhard Fischer...

18 years agoA cleanup patch I've had lying around in my tree for a while, I think it
Rob Landley [Fri, 2 Dec 2005 17:54:01 +0000 (17:54 -0000)]
A cleanup patch I've had lying around in my tree for a while, I think it
was from Bernhard Fischer?

18 years agoreduce signedness warning
"Vladimir N. Oleynik" [Fri, 2 Dec 2005 10:10:28 +0000 (10:10 -0000)]
reduce signedness warning

18 years agoreduce signedness warning
"Vladimir N. Oleynik" [Fri, 2 Dec 2005 10:06:04 +0000 (10:06 -0000)]
reduce signedness warning

18 years agoremove warnings if compile with -W, use ENABLE_FEATURE vs CONFIG_FEATURE
"Vladimir N. Oleynik" [Fri, 2 Dec 2005 09:46:04 +0000 (09:46 -0000)]
remove warnings if compile with -W, use ENABLE_FEATURE vs CONFIG_FEATURE

18 years agoMake char default to "unsigned" on all platforms so we have a consistent base
Rob Landley [Thu, 1 Dec 2005 17:01:43 +0000 (17:01 -0000)]
Make char default to "unsigned" on all platforms so we have a consistent base
to deal with all the new gcc 4.0.2 warnings from.

18 years agoFix losetup so that it A) actually works again, B) has much better error
Rob Landley [Tue, 29 Nov 2005 23:47:10 +0000 (23:47 -0000)]
Fix losetup so that it A) actually works again, B) has much better error
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...

18 years agogetopt -> getopt_ulflags. noticed by Rob Sullivan
"Vladimir N. Oleynik" [Tue, 29 Nov 2005 12:32:33 +0000 (12:32 -0000)]
getopt -> getopt_ulflags. noticed by Rob Sullivan

18 years agoupdate license
Mike Frysinger [Tue, 29 Nov 2005 02:53:52 +0000 (02:53 -0000)]
update license

18 years agoShaneAnderson writes in Bug 565:
Mike Frysinger [Tue, 29 Nov 2005 02:52:25 +0000 (02:52 -0000)]
ShaneAnderson writes in Bug 565:
need to include .config file before setting CROSS otherwise
TARGET_ARCH isnt set properly

18 years agoadd "-C" format to hexdump
Paul Fox [Mon, 28 Nov 2005 21:06:00 +0000 (21:06 -0000)]
add "-C" format to hexdump

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.)

18 years agofix bug #474:
Paul Fox [Mon, 28 Nov 2005 18:07:53 +0000 (18:07 -0000)]
fix bug #474:
     0000474: vi crashes often

problem was that the buffer used for "." command ("last_modifying_cmd")
wasn't being maintined correctly -- the recording code was walking back
over the front of that buffer when a repeatable insert command
included backspacing (e.g. "i\b\b\bfoo").  the fix is to simply
record the backspaces along with the rest of the command.

also, cleaned up start_new_cmd_q() slightly.

18 years agoeliminate mixed declaration, to support older compilers a little longer.
Paul Fox [Mon, 28 Nov 2005 16:29:45 +0000 (16:29 -0000)]
eliminate mixed declaration, to support older compilers a little longer.

18 years agouse libbb/get_kernel_revision(), reduce stack usage, add loses -w -f option for getop...
"Vladimir N. Oleynik" [Mon, 28 Nov 2005 15:54:22 +0000 (15:54 -0000)]
use libbb/get_kernel_revision(), reduce stack usage, add loses -w -f option for getopt, convert to bb_getopt_ulflags(), reduce memory usage - xmalloc to bb_common_bufsiz1, size reduce over 200 bytes

18 years agoPatch from Cristian Ionescu-Idbohrn to deal with _ vs - better.
Rob Landley [Sun, 27 Nov 2005 19:01:53 +0000 (19:01 -0000)]
Patch from Cristian Ionescu-Idbohrn to deal with _ vs - better.

18 years agoremove bug: fsck do not know path
"Vladimir N. Oleynik" [Sat, 26 Nov 2005 11:01:23 +0000 (11:01 -0000)]
remove bug: fsck do not know path

18 years agochange the interface of libbb/compare_string_array (unsigned short to int), usaging...
"Vladimir N. Oleynik" [Sat, 26 Nov 2005 10:45:26 +0000 (10:45 -0000)]
change the interface of libbb/compare_string_array (unsigned short to int), usaging for e2fsprogs/fsck

18 years agomore use const for interface of libbb/compare_string_array, example usage for e2fspro...
"Vladimir N. Oleynik" [Sat, 26 Nov 2005 10:33:55 +0000 (10:33 -0000)]
more use const for interface of libbb/compare_string_array, example usage for e2fsprogs/fsck

18 years agomore use libbb/bb_msg_read_error
"Vladimir N. Oleynik" [Fri, 25 Nov 2005 11:36:36 +0000 (11:36 -0000)]
more use libbb/bb_msg_read_error

18 years agoreduce bss
"Vladimir N. Oleynik" [Thu, 24 Nov 2005 12:10:13 +0000 (12:10 -0000)]
reduce bss

18 years agoUpdate to modprobe usage from Yann E. Morin
Rob Landley [Wed, 23 Nov 2005 04:55:02 +0000 (04:55 -0000)]
Update to modprobe usage from Yann E. Morin

18 years agoCan't store a pointer in an int on 64 bit platforms.
Rob Landley [Tue, 22 Nov 2005 17:09:14 +0000 (17:09 -0000)]
Can't store a pointer in an int on 64 bit platforms.

18 years agoMove interface.c to networking directory since it's only used by ifconfig.
Rob Landley [Tue, 22 Nov 2005 07:52:54 +0000 (07:52 -0000)]
Move interface.c to networking directory since it's only used by ifconfig.

18 years agoSupport # comments after s/// option list.
Rob Landley [Sun, 20 Nov 2005 07:44:35 +0000 (07:44 -0000)]
Support # comments after s/// option list.

18 years agoQuote the user we pass to addgroup so funky chars are preserved. Fix by dragan_marin...
Mike Frysinger [Tue, 15 Nov 2005 05:53:00 +0000 (05:53 -0000)]
Quote the user we pass to addgroup so funky chars are preserved.  Fix by dragan_marinkovic in Bug 544

18 years agoYann Morin's modprobe multiple options patch. There's more work to be done,
Rob Landley [Tue, 15 Nov 2005 00:08:29 +0000 (00:08 -0000)]
Yann Morin's modprobe multiple options patch.  There's more work to be done,
but let's ship 1.1 first...

18 years agoMy last change to this file broke tar x. Sigh. (Trying to make sure we work
Rob Landley [Sat, 12 Nov 2005 11:04:11 +0000 (11:04 -0000)]
My last change to this file broke tar x.  Sigh.  (Trying to make sure we work
on file sizes between 2 and 4 gigs on 32 bit machines.  For supporting >4 gigs,
just use a 64 bit machine already...)

18 years agoI noticed that "tar tvjf file.tbz" was segfaulting. This fixed it.
Rob Landley [Sat, 12 Nov 2005 02:54:06 +0000 (02:54 -0000)]
I noticed that "tar tvjf file.tbz" was segfaulting.  This fixed it.

18 years agotelnetd -b (bind to specific address) support from Iuri Gomes Diniz.
Rob Landley [Thu, 10 Nov 2005 22:37:40 +0000 (22:37 -0000)]
telnetd -b (bind to specific address) support from Iuri Gomes Diniz.

18 years agoOk, I've converted the contents of the "testing/sed" directory into a
Rob Landley [Thu, 10 Nov 2005 06:26:40 +0000 (06:26 -0000)]
Ok, I've converted the contents of the "testing/sed" directory into a
sed.tests file.  My brain hurts now.  (Lots of boggling at sed minutiae and
corner cases and going "why is gnu giving that output".  The behavior of N
and n with regard to EOF are only understandable if you read the Open Group
spec, not if you read the sed info page, by the way...)

Some of the existing sed tests are just nuts.  For example, sed-next-line is
testing for our behavior (which is wrong), and would fail if run against gnu
sed (which was getting it right.  Again, this was a spec-boggling moment,
with much head scratching.  I've got to add a debug mode where the stuff
output by the p command is a different color from the stuff output by normal
end of script printing (when not suppressed by -n).)

As for sed-handles-unsatisifed-backrefs: what is this test trying to _do_?  I
ran it against gnu sed and got an error message, and this behavior sounds
perfectly reasonable.  (It _is_ an unsatisfied backref.)  The fact we
currently ignore this case (and treat \1 as an empty string) isn't really
behavior we should have a test depend on for success.

The remaining one is sed-aic-commands, which is long and complicated.  I'm
trying to figure out if I should chop this into a number of smaller tests, or
if having one big "does-many-things" test is a good idea.

In any case, the _next_ step is to go through the Open Group standard and
make tests for every case not yet covered.  (And there are plenty.  There
are few comments in the file already.)  Plus I have notes about corner
cases from development that I need to collate and put into here.  This file
is maybe the first 1/3 of a truly comprehensive sed test.

Rob

18 years agochange osbolete sigpause() to sigsuspend()
"Vladimir N. Oleynik" [Wed, 9 Nov 2005 09:19:29 +0000 (09:19 -0000)]
change osbolete sigpause() to sigsuspend()

18 years agoPatch from Tomasz Mateja: don't try to CWD to a _filename_, just send the full
Rob Landley [Tue, 8 Nov 2005 00:52:31 +0000 (00:52 -0000)]
Patch from Tomasz Mateja: don't try to CWD to a _filename_, just send the full
path to the STOR command.

18 years agoIt was a bit out of date.
Rob Landley [Mon, 7 Nov 2005 09:06:34 +0000 (09:06 -0000)]
It was a bit out of date.

18 years agoFix the test suite so that individual *.tests files can be run ala
Rob Landley [Mon, 7 Nov 2005 08:50:53 +0000 (08:50 -0000)]
Fix the test suite so that individual *.tests files can be run ala
  COMMAND=sort ./sort.tests
So we can compare against non-busybox versions, and possibly our testsuite
will be useful to somebody like the Linux Test Project someday.

Redid testing.sh to add new command, "optional", to skip tests that require
certain features.  (use: `optional FEATURE_SORT_BIG`, or `optional ""` to
stop skipping.)  Note that optional is a NOP if the environment variable
"OPTIONFLAGS" is blank, so although we're marking up the tests with busybox
specific knowledge, it doesn't interfere with running the tests without
busybox.

Moved setting the "OPTIONFLAGS" environment variable to runtest.  Philosophy:
busybox-specific stuff belongs in runtest; both testing.sh and the tests
themselves should be as busybox-agnostic as possible.

Moved detecting that a command isn't in busybox at all (hence skipping the
entire command.tests file) to runtests.  Rationale: optional can't currently
test for more than one feature at a time, so if we clear anything with
optional "" we might perform tests we don't want to.

Marked up busybox.tests to know which tests need CAT enabled.  Fixed up other
tests to be happy with new notation.

I suspect egrep should be appended to grep.  It's a sub-feature, really...

18 years agoYann E. Morin pointed out that rpcsvc/nfs_prot.h isn't needed with any
Rob Landley [Sun, 6 Nov 2005 23:37:47 +0000 (23:37 -0000)]
Yann E. Morin pointed out that rpcsvc/nfs_prot.h isn't needed with any
reasonably recent glibc, and breaks with uClibc.  So yank it.

18 years agoSize I zapped the bb_close() error, might as well make a TODO item.
Rob Landley [Fri, 4 Nov 2005 20:45:54 +0000 (20:45 -0000)]
Size I zapped the bb_close() error, might as well make a TODO item.

18 years agoFix bug 424: doing full_read breaks things like cat which should return a
Rob Landley [Fri, 4 Nov 2005 01:54:15 +0000 (01:54 -0000)]
Fix bug 424: doing full_read breaks things like cat which should return a
chunk of data when they get it and not block until they've buffered 4k.

The use case was cat /proc/psaux, but you can also reproduce this by
running non-busybox cat by itself and typing things at the command line.
Then run busybox cat.  Notice how cat is _supposed_ to echo each line back
to us as we hit enter?

18 years agoWhen cp ran out of space it didn't return a nonzero error code. Fixes bug 493.
Rob Landley [Fri, 4 Nov 2005 01:20:46 +0000 (01:20 -0000)]
When cp ran out of space it didn't return a nonzero error code.  Fixes bug 493.

18 years agoIt's no longer a temporary location. Smooth out the text a bit.
Rob Landley [Fri, 4 Nov 2005 00:50:48 +0000 (00:50 -0000)]
It's no longer a temporary location.  Smooth out the text a bit.

18 years agopoint to the official links
Eric Andersen [Thu, 3 Nov 2005 23:32:14 +0000 (23:32 -0000)]
point to the official links

18 years agoAnnounce 1.1.0-pre1 already. Erik, please move links to the downloads directory
Rob Landley [Thu, 3 Nov 2005 22:11:00 +0000 (22:11 -0000)]
Announce 1.1.0-pre1 already.  Erik, please move links to the downloads directory
once you've got the signed tarballs up.

18 years agoFix cp /dev/null filename, and a few in-passing cleanups.
Rob Landley [Tue, 1 Nov 2005 21:55:14 +0000 (21:55 -0000)]
Fix cp /dev/null filename, and a few in-passing cleanups.

18 years agoWe've got fuser, fix some typos, and move Vodz's comment about UTF8 into
Rob Landley [Mon, 31 Oct 2005 23:52:02 +0000 (23:52 -0000)]
We've got fuser, fix some typos, and move Vodz's comment about UTF8 into
a new "internationalization" section, with some related concerns.

18 years ago\0 is a synonym for &. Our own "make allbaseconfig" breaks without this.
Rob Landley [Sun, 30 Oct 2005 10:08:13 +0000 (10:08 -0000)]
\0 is a synonym for &.  Our own "make allbaseconfig" breaks without this.

18 years agoadd ZyXEL to the shame list
Mike Frysinger [Sat, 29 Oct 2005 08:14:34 +0000 (08:14 -0000)]
add ZyXEL to the shame list

18 years agofix <p> and <a> tags
Mike Frysinger [Sat, 29 Oct 2005 08:03:54 +0000 (08:03 -0000)]
fix <p> and <a> tags

18 years agoupdate doctype and add a charset encoding
Mike Frysinger [Sat, 29 Oct 2005 07:57:51 +0000 (07:57 -0000)]
update doctype and add a charset encoding

18 years ago- add BB_APPLET_RUNLEVEL used by emdebian via /etc/init.d/rc.
Bernhard Reutner-Fischer [Fri, 28 Oct 2005 20:37:03 +0000 (20:37 -0000)]
- add BB_APPLET_RUNLEVEL used by emdebian via /etc/init.d/rc.
  Note that we leave the buggy CONFIG_* namespace now, so please fix any applet you thouch.

18 years agoupdate
"Vladimir N. Oleynik" [Fri, 28 Oct 2005 16:17:09 +0000 (16:17 -0000)]
update

18 years agoPlease don't use if(asprintf(&t)<0) free(t); Use concat_subpath_file() for scan dir...
"Vladimir N. Oleynik" [Fri, 28 Oct 2005 16:08:47 +0000 (16:08 -0000)]
Please don't use if(asprintf(&t)<0) free(t); Use concat_subpath_file() for scan dir. Added losed chdir(".."). Removed warnings for 64-bit inode size. Why use memcpy and do not let? Very strange code for me

18 years agobb_opt_complementally for rdate applet. Add losed record to util-linux/Makefile.in...
"Vladimir N. Oleynik" [Fri, 28 Oct 2005 15:43:41 +0000 (15:43 -0000)]
bb_opt_complementally for rdate applet. Add losed record to util-linux/Makefile.in for switch_root applet

18 years agoNew applet, fuser, from Tony J. White. (Needs some cleanup.)
Rob Landley [Fri, 28 Oct 2005 13:05:12 +0000 (13:05 -0000)]
New applet, fuser, from Tony J. White.  (Needs some cleanup.)

18 years ago- move GETXXKEY_R_FUNC into an internal helper file to allow for compiling
Bernhard Reutner-Fischer [Fri, 28 Oct 2005 11:21:40 +0000 (11:21 -0000)]
- move GETXXKEY_R_FUNC into an internal helper file to allow for compiling
  with IMA.

18 years ago- remove warning about "'static' is not at beginning of declaration".
Bernhard Reutner-Fischer [Fri, 28 Oct 2005 09:45:07 +0000 (09:45 -0000)]
- remove warning about "'static' is not at beginning of declaration".

18 years agoRob Sullivan sent in some cleanups, which I beat on slightly.
Rob Landley [Fri, 28 Oct 2005 09:24:33 +0000 (09:24 -0000)]
Rob Sullivan sent in some cleanups, which I beat on slightly.

18 years agoAdd a switch_root utility (like kconfig's utils/run_init.c, although not
Rob Landley [Thu, 27 Oct 2005 22:55:50 +0000 (22:55 -0000)]
Add a switch_root utility (like kconfig's utils/run_init.c, although not
actuall using any of that code).  This is needed because pivot_root doesn't
work right under initramfs.  (See the menuconfig help.)

18 years agoSomebody suggested this, but I've forgotten who.
Rob Landley [Thu, 27 Oct 2005 22:50:04 +0000 (22:50 -0000)]
Somebody suggested this, but I've forgotten who.

18 years agoAdd --no-same-owner and --no-same-permissions options to tar.
Rob Landley [Thu, 27 Oct 2005 22:49:08 +0000 (22:49 -0000)]
Add --no-same-owner and --no-same-permissions options to tar.

18 years ago- Matthias Wenzel writes:
Bernhard Reutner-Fischer [Thu, 27 Oct 2005 14:43:27 +0000 (14:43 -0000)]
- Matthias Wenzel writes:
without the fix below md5sum will always report a correct md5 on _any_
wrongly formattet input files.
- use short boilerplate and remove superfluous keyword extern.

18 years ago- correct boilerplate
Bernhard Reutner-Fischer [Thu, 27 Oct 2005 06:59:05 +0000 (06:59 -0000)]
- correct boilerplate
- use bb_xgetlarg for the patch_level to catch invalid arguments.

18 years agoReference for new patch feature we'll need to be compatible with...
Rob Landley [Thu, 27 Oct 2005 00:57:50 +0000 (00:57 -0000)]
Reference for new patch feature we'll need to be compatible with...

18 years ago * uniq.c: remove unneeded include and use short boilerplate.
Bernhard Reutner-Fischer [Wed, 26 Oct 2005 10:47:26 +0000 (10:47 -0000)]
* uniq.c: remove unneeded include and use short boilerplate.
* coreutils.h: remove prototype of non-existing xgetoptfile_sort_uniq
and add boilerplate.
* networking/{ipaddr,ip,iplink,iproute,iptunnel}.c: touch includes
and use short boilerplate.
* libiproute/iproute.c: rename round to avoid clashes with older
SuSE gcc and use short boilerplate.

18 years agoUpdate the TODO list to reflect stuff we've already done.
Rob Landley [Tue, 25 Oct 2005 20:22:50 +0000 (20:22 -0000)]
Update the TODO list to reflect stuff we've already done.

18 years agoShaun Jackman noted that limits.h defines path_max and isn't included from
Rob Landley [Tue, 25 Oct 2005 02:22:13 +0000 (02:22 -0000)]
Shaun Jackman noted that limits.h defines path_max and isn't included from
find_root_device.c.  (We #include it in busybox.h but not libbb.h, it seems.
Someday, someone's going to have to clarify for me the difference between
those two...)

18 years agoTwo changes. One from Shaun Jackman makes libbb/interface.c compile only if
Rob Landley [Tue, 25 Oct 2005 02:19:55 +0000 (02:19 -0000)]
Two changes.  One from Shaun Jackman makes libbb/interface.c compile only if
it's going to be used.  (I'm guessing it doesn't work with newlib.)

The other one is from me: allyesconfig shouldn't enable devfs because that
changes all sorts of unrelated stuff (like /dev/loop0->dev/loop/0), which
can come as a bit of a surprise.  (It's still there, but you have to go into
menuconfig and select it manually.)

18 years ago- fix bug to avoid aliasing busybox' main to zcip_main in IMA mode.
Bernhard Reutner-Fischer [Fri, 21 Oct 2005 10:43:11 +0000 (10:43 -0000)]
- fix bug to avoid aliasing busybox' main to zcip_main in IMA mode.

18 years agoanother more const
"Vladimir N. Oleynik" [Thu, 20 Oct 2005 11:17:48 +0000 (11:17 -0000)]
another more const

18 years agomore const
"Vladimir N. Oleynik" [Thu, 20 Oct 2005 09:48:28 +0000 (09:48 -0000)]
more const

18 years agomore const, attribute_noreturn saved 200 bytes
"Vladimir N. Oleynik" [Wed, 19 Oct 2005 09:21:51 +0000 (09:21 -0000)]
more const, attribute_noreturn saved 200 bytes

18 years ago- consume space between functionname and opening brackets
Bernhard Reutner-Fischer [Mon, 17 Oct 2005 14:21:06 +0000 (14:21 -0000)]
- consume space between functionname and opening brackets
  Fixes Rob's issue using busybox awk for building gcc-4_0 optionlist (http://busybox.net/lists/busybox/2005-October/016659.html)

18 years agomore const, use bb_getopt_ulflags, insert XXX for show place of problems
"Vladimir N. Oleynik" [Mon, 17 Oct 2005 10:47:19 +0000 (10:47 -0000)]
more const, use bb_getopt_ulflags, insert XXX for show place of problems

18 years agodata --> text
"Vladimir N. Oleynik" [Mon, 17 Oct 2005 09:48:57 +0000 (09:48 -0000)]
data --> text

18 years agoThe fact "tar tvjf thing.tbz" didn't work was due to the "-1" in -r11859,
Rob Landley [Sun, 16 Oct 2005 08:42:18 +0000 (08:42 -0000)]
The fact "tar tvjf thing.tbz" didn't work was due to the "-1" in -r11859,
which was apparently never tested.  That meant that there always had
to be at least one unparsed argument left over, which is not the case for
tar.

18 years agoWhitespace and curly bracket cleanup (our tabstop is 4 in busybox),
Rob Landley [Sun, 16 Oct 2005 03:54:49 +0000 (03:54 -0000)]
Whitespace and curly bracket cleanup (our tabstop is 4 in busybox),
and switch more stuff from CONFIG to ENABLE.

18 years ago- don't update copyright years. He might release it, but apparently will have
Bernhard Reutner-Fischer [Sat, 15 Oct 2005 20:56:31 +0000 (20:56 -0000)]
- don't update copyright years. He might release it, but apparently will have
  to update the year himself.

18 years agomore const, all data to text
"Vladimir N. Oleynik" [Sat, 15 Oct 2005 14:41:08 +0000 (14:41 -0000)]
more const, all data to text

18 years ago20 bytes data to bss
"Vladimir N. Oleynik" [Sat, 15 Oct 2005 14:24:01 +0000 (14:24 -0000)]
20 bytes data to bss

18 years ago- an empty middle term in ?: violates ISO C
Bernhard Reutner-Fischer [Sat, 15 Oct 2005 14:13:09 +0000 (14:13 -0000)]
- an empty middle term in ?: violates ISO C
- use shorter boilerplate and use C89 style comments

18 years agodata --> code
"Vladimir N. Oleynik" [Sat, 15 Oct 2005 14:10:36 +0000 (14:10 -0000)]
data --> code

18 years agodata --> bss
"Vladimir N. Oleynik" [Sat, 15 Oct 2005 13:56:17 +0000 (13:56 -0000)]
data --> bss

18 years agocomplex RESERVE_CONFIG_BUFFER --> bb_common_bufsiz1
"Vladimir N. Oleynik" [Sat, 15 Oct 2005 13:49:21 +0000 (13:49 -0000)]
complex RESERVE_CONFIG_BUFFER --> bb_common_bufsiz1

18 years agoRESERVE_CONFIG_BUFFER --> bb_common_bufsiz1
"Vladimir N. Oleynik" [Sat, 15 Oct 2005 13:45:32 +0000 (13:45 -0000)]
RESERVE_CONFIG_BUFFER --> bb_common_bufsiz1

18 years agocommon BUFSIZ BSS buffer, small reduce code, data and bss
"Vladimir N. Oleynik" [Sat, 15 Oct 2005 10:23:55 +0000 (10:23 -0000)]
common BUFSIZ BSS buffer, small reduce code, data and bss

18 years agollist must initialize, add loses ifdef, small indent correction
"Vladimir N. Oleynik" [Sat, 15 Oct 2005 06:32:38 +0000 (06:32 -0000)]
llist must initialize, add loses ifdef, small indent correction

18 years agoAdd --exclude option (to make uClibc-0.9.28 headers install using busybox tar).
Rob Landley [Sat, 15 Oct 2005 03:06:21 +0000 (03:06 -0000)]
Add --exclude option (to make uClibc-0.9.28 headers install using busybox tar).

I have no idea how to apply bb_getopt_complementally to a --longopt that
has no short option.  The documentation from vodz has a bad case of
babelfish poisoning, and I can't understand it.  It sort of seems to
suggest there is a way, but what it is I have no idea.  So I used \n as
the short option, which is fairly unlikely to be used for something else. :)

18 years ago- new bb_opt_complementally syntax, use [-:?] only - 'free' chars
"Vladimir N. Oleynik" [Fri, 14 Oct 2005 09:56:52 +0000 (09:56 -0000)]
- new bb_opt_complementally syntax, use [-:?] only - 'free' chars
- new bb_getopt_ulflags features: check max and min args, convert first argv to options special for ar and tar applets
- use bb_default_error_retval for env applet

18 years ago- split VERSION into parts
Bernhard Reutner-Fischer [Thu, 13 Oct 2005 10:40:18 +0000 (10:40 -0000)]
- split VERSION into parts
- add some filesystem operation wrapper variables for use in the makefiles and
  pull them in early in the toplevel makefile
- use the cross-toolchain for "make sizes"

18 years ago- initialize the correct variables
Bernhard Reutner-Fischer [Thu, 13 Oct 2005 10:24:44 +0000 (10:24 -0000)]
- initialize the correct variables

18 years ago- typo in makefile variable: s/archvial/archival/
Bernhard Reutner-Fischer [Thu, 13 Oct 2005 10:19:35 +0000 (10:19 -0000)]
- typo in makefile variable: s/archvial/archival/

18 years agoshow_usage should use bb_default_error_retval.
Rob Landley [Wed, 12 Oct 2005 21:50:02 +0000 (21:50 -0000)]
show_usage should use bb_default_error_retval.

18 years agoDid you know the uclibc IRC channel has daily logs on the web going back to
Rob Landley [Wed, 12 Oct 2005 19:10:34 +0000 (19:10 -0000)]
Did you know the uclibc IRC channel has daily logs on the web going back to
2002?

18 years agoperror correction
"Vladimir N. Oleynik" [Wed, 12 Oct 2005 16:45:21 +0000 (16:45 -0000)]
perror correction

18 years agocreate e2fs_set_sbin_path(), noticed by Tito, Thanks
"Vladimir N. Oleynik" [Wed, 12 Oct 2005 16:22:19 +0000 (16:22 -0000)]
create e2fs_set_sbin_path(), noticed by Tito, Thanks

18 years agousage bb_dev_null
"Vladimir N. Oleynik" [Wed, 12 Oct 2005 15:34:25 +0000 (15:34 -0000)]
usage bb_dev_null

18 years agobb_dev_null
"Vladimir N. Oleynik" [Wed, 12 Oct 2005 15:21:32 +0000 (15:21 -0000)]
bb_dev_null

18 years agoremove 1 CR, correct strange bb_opt_complementally, add #if 0 and XXX-comment: pleace...
"Vladimir N. Oleynik" [Wed, 12 Oct 2005 14:36:42 +0000 (14:36 -0000)]
remove 1 CR, correct strange bb_opt_complementally, add #if 0 and XXX-comment: pleace examine this\!

18 years agomake fsck.h as common, e2fsck: includes correct
"Vladimir N. Oleynik" [Wed, 12 Oct 2005 12:24:01 +0000 (12:24 -0000)]
make fsck.h as common, e2fsck: includes correct

18 years agomore busyboxes, remove 1 extern function (mainstream also have for one fsck)
"Vladimir N. Oleynik" [Wed, 12 Oct 2005 12:11:42 +0000 (12:11 -0000)]
more busyboxes, remove 1 extern function (mainstream also have for one fsck)