enable more applets

* chvt
* deallocvt
* fgconsole
* findfs
* hwclock
* lspci
* nmeter
* openvt
* powertop
* rdate
* setkeycodes
* smemcap
* switch_root
* tty

android-x86: fix building warnings of kitkat-x86

lsusb: add -v option to show bDeviceClass and DEVNAME

xlseek: large file support

This fixes the issue that mke2fs can't format large disks.

build: ignore selibs pkgconfig on android builds

This cause errors and is not required

Change-Id: If6cc6fb1595e83fac452d8a28cb4b8d250dff06d

update config, change version suffix

Change-Id: I3455772c65bbd2e1102ec1f2d9c55f4d91b702d8

android: add a script to regenerate prebuilt config

since x86 CROSS_COMPILER removal, it can be hard to do that properly

Change-Id: I6011e3b4dac5b5e670385df4812baf32f2e5ebc5

fstrim: use new-style config/kbuild/applet snippets; trim help text

function                                             old     new   delta
packed_usage                                       29546   29520     -26

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
Change-Id: I758c91f2054d1f2876fc1446f8f0afd8bffd2611

fstrim: sync with busybox master (now an official applet)

fstrim: Needs to fire at the mp, not bd

It's a filesystem thing from this perspective.

fstrim: Indicate failure on error

Signed-off-by: Bernhard Reutner-Fischer <rep.dot.nop@gmail.com>
Change-Id: Icffb851cbe1895523e67e5ff18c67f9109a46a28

Fix building on 4.4

Change-Id: I85030de3dea5f2d1a187cf2ced9517b3c820266a

bunzip2: fix off-by-one check

stage3-armv7a_hardfp-20130209.tar.bz2, 149189948 bytes long,
md5sum b29ce23312e14eb15a143377d4a38473, was failing to unpack.

It so happened that this file has a run which exactly fills
the 90k buffer. The check was "size >= bufsize", apparently
it has to be ">".

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
fix for bug https://jira.cyanogenmod.org/browse/CYAN-1896
Change-Id: I4e43fa2b84f2c00e627f75082d2250b4b4dff08e

[x86] Changes and Additions to fix broken x86 build

Add missing syscalls to android/libc/arch-x86/syscalls

Remove #undef HAVE_GETLINE from platform.h
Fixes definition conflict warnings for x86 but is also
suitable for arm as  bionic has had a getline implementation for a
while now

Remove hardcoded CONFIG_CROSS_COMPILER_PREFIX from .config-full
and .config-minimal. This is now appended to  .config as part of
the prepare target in Android.mk

It appears no other changes are needed to other files such as
bbconfigopts.h etc as these are autogenerated when make prepare
is run.

Android.mk:
Added missing syscalls source to BUSYBOX_SRC_FILES when
TARGET_ARCH=x86.
Added appending of the CONFIG_CROSS_COMPILER_PREFIX to the .config
file based on the TARGET_ARCH

Change-Id: I4f550757b3209b852cd5a017e88dce50678e5734

mktemp: go back to using mktemp() for mktemp -u

The -u option to mktemp(1) is inherently unsafe, and the fact it exists
in the first place is probably a mistake.  That said, it exists, and if
we want to support it, we should try to minimize the risk of using it.

The current code attempts to work around a build failure caused by
passing --fatal-warnings to the linker by using mkstemp() and unlinking
the resulting file, which eliminates the need for an attacker to guess
the filename (he can instead watch the temporary file directory using
inotify or some other similar mechanism).  It's better instead to use
mktemp() (which at least forces the attacker to guess what "XXXXXX" was
replaced with) and tell the linker to live with the warning.

Change-Id: I01e0901db39821b7d1ce06145d1199ecd929e689

Don't treat link-time warnings as errors

We want to use mktemp(3) in the mktemp applet (it's unsafe, but so is
that particular option of the applet, and mktemp() is the least unsafe
way of implementing that option), but forcing the linker to error out on
warnings breaks the build in this case.

Change-Id: I6e976e9b467d2241be4bb0f3628991d098c492f0

mktemp: unlink mkstemp alternative to (unsafe) dry-run option

Change-Id: I19eeca4e8a915418527630105405c4d74d542ead

mktemp: switch to mkstemp

The compiler warning was breaking the build due to Werror, and it's
an easy fix

Change-Id: Ic109aa017c02596e9cf0aae3c6f82d72849b90cf

Fix syscalls on 4.3's bionic

Change-Id: Ibaeaa30ce88880338cd0236791f7db4cac8289b5

Merge remote branch 'busybox_net/1_21_stable' into cm-10.1

Apply post-1.21.0 patches, bump version to 1.21.1

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
update config for 1.21.1
Signed-off-by: Tanguy Pruvot <tanguy.pruvot@gmail.com>
Change-Id: Ibfd9124cbc9be744709dfe07a1a5b30df545e806

volume_id: fix ntfs attribute search

len is declared as uint32_t but le16_to_cpu macro is used

Signed-off-by: Oliver Metz <oliver.metz@gmx.de>
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
Change-Id: Ibc342fcf970d98003fac15c7008b87622e38647f

Apply post-1.21.0 patches, bump version to 1.21.1

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>

xz: support concatenated .xz streams

function                                             old     new   delta
xz_dec_reset                                           -      77     +77
unpack_xz_stream                                    2402    2397      -5

Signed-off-by: Lasse Collin <lasse.collin@tukaani.org>
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>

xz: fix put_unaligned_{l,b}e32

Signed-off-by: Leonid Lisovskiy <lly.dev@gmail.com>
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
Sync code with busybox master fix
Signed-off-by: Tanguy Pruvot <tanguy.pruvot@gmail.com>

Busybox 1.21.0 squashed commit for jellybean

Change-Id: I423c7fc1254050c6495126b1b18dd33af07fed6b
Signed-off-by: Tanguy Pruvot <tanguy.pruvot@gmail.com>

Build with -fno-strict-aliasing

Change-Id: I4117a526cdc93c96a9b7102727f5b33542810990
Signed-off-by: Evan McClain <aeroevan@gmail.com>

update config (fstrim addition)

Change-Id: I5c872afbcd071d389d0db3f10862677c08212e2c

Add fstrim support to restore NAND/eMMC write performance. fstrim applet is a port from util-linux.

"Trimming" your NAND/eMMC storage will restore the write performance back to normal
after having slow down issues on sequential write and random write due to usage over time.

Good reading on subject: http://forum.xda-developers.com/showthread.php?t=1971852

Change-Id: I0ba903cb7248b288152736f2e3392a5bdbe77ba3

Bump version to 1.21.0

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>

fix testsuite false positive

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>

mdev: fix mode of dir1 in =dir1/dir2/file rule

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>

ifconfig: do not accept "ifconfig eth0 up 1.2.3.4/17" (ip with mask). Closes 5786

function                                             old     new   delta
ifconfig_main                                       1221    1237     +16

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>

grep: fix grep -Fw not respecting the -w option. Closes 5792

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>

syslogd: do not segfault on parse error when using default config. Closes 5762

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>

sha3: code shrink

function                                             old     new   delta
sha3_hash                                            155     101     -54

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>

syslogd: convert dummy functions to statics and get rid of IF_FEATURE_* checks

As suggested by Mike. No bloat-o-meter difference, but a bit nicer to look at.
We cannot convert the call to log_to_shmem() as it checks for G.shbuf outside
the function, and G.shbuf is only available when IPC support is enabled.

Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>

dc: code shrink

function                                             old     new   delta
stack_machine                                        103     101      -2
operators                                            176     168      -8

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>

ash: revert wrong "fix" for an apparent memory leak. Closes 5822

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>

code shrink

function                                             old     new   delta
applet_name_compare                                   36      31      -5
find_applet_by_name                                   43      25     -18

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>

fix a typo in config help text. Closes 5714

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>

mount: fix -o user=foo mishandling, fix unc= generation, add prefixpath=

function                                             old     new   delta
singlemount                                         1019    1049     +30
packed_usage                                       29252   29257      +5
parse_mount_options                                  230     232      +2
mount_option_str                                     337     338      +1
------------------------------------------------------------------------------
(add/remove: 0/0 grow/shrink: 4/0 up/down: 38/0)               Total: 38 bytes

Signed-off-by: Bernhard Reutner-Fischer <rep.dot.nop@gmail.com>
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>

sha3: s/sha3_process_block76/sha3_process_block72/

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>

sha3: rename KeccakF->sha3_process_block76.

This brings the naming more in line with other hashes.
Pulled most statics and constants into it.
Also noticed that two byte arrays are 1 element too big.

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>

mount: add unc option to CIFS mount (needed for Linux 3.4+)

unc option for CIFS mount is mandatory after CIFS option parsing
was rewritten in Linux 3.4

Signed-off-by: Martin Santesson <martinsn@axis.com>
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>

sha3: cache ctx->bytes_queued

function                                             old     new   delta
sha3_hash                                            171     155     -16

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>

sha3: remove two "small code" codepaths: I can't reproduce code size win on them anymore

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>

sha3: tweak choice of a fast code path for 64-bit

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>

sha3: cosmetic tweaks to various names, comments. No logic changes.

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>

sha3: code shrink

function                                             old     new   delta
KeccakF                                             1053    1078     +25
KeccakF_RoundConstants                               192      48    -144

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>

sha3: code shrink

function                                             old     new   delta
KeccakF                                             1064    1053     -11

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>

sha3: another speedup for SHA3_SMALL=0 case

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>

whitespace fixes. no code changes

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>

sha3: make size/speed optimization decision configurable

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>

whitespace cleanup. no code changes

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>

sha3sum: new applet

function                                             old     new   delta
KeccakF                                                -     496    +496
KeccakF_RoundConstants                                 -     192    +192
sha3_hash                                              -     171    +171
sha3_end                                               -      40     +40
hash_file                                            274     299     +25
KeccakF_RotationConstants                              -      25     +25
KeccakF_PiLane                                         -      25     +25
packed_usage                                       29213   29232     +19
sha3_begin                                             -      18     +18
KeccakF_Mod5                                           -      10     +10
applet_names                                        2445    2453      +8
applet_main                                         1420    1424      +4
applet_nameofs                                       710     712      +2
------------------------------------------------------------------------------
(add/remove: 8/0 grow/shrink: 9/7 up/down: 1049/-54)         Total: ~995 bytes

Signed-off-by: Lauri Kasanen <curaga@operamail.com>
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>

ntpd: fix incorrect m_status field in outgoing packets. Closes 5120

When using busybox ntpd with an NTPv3 client and NTPv4 server (or vice
versa), the version numbers can be incorrectly ORed together, yielding
the bogus value of "NTPv7".  This makes ntpd unusable with clients
such as Chrony and Windows "Internet Time".

This patch avoids the version mangling, by copying only the Leap
Indicator bits from the server's status field.

Signed-off-by: Paul Marks <paul@pmarks.net>
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>

Remove redundant 'config FEATURE_VOLUMEID_SQUASHFS'

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>

volume_id: add squashfs detection

function                                             old     new   delta
volume_id_probe_squashfs                               -      74     +74
fs1                                                   12      16      +4

Signed-off-by: Sven-Göran Bergh <sgb@systemasis.org>
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>

whitespace fixes. no code changes

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>

volume_id: uuid_format small code shrink

function                                             old     new   delta
volume_id_set_uuid                                   318     300     -18

Signed-off-by: Sven-Göran Bergh <sgb@systemasis.org>
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>

volume_id: display hfs[+] 128-bit UUID properly

Signed-off-by: Sven-Göran Bergh <sgb@systemasis.org>
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>

blkid: add type display for hfsplus

Signed-off-by: Sven-Göran Bergh <sgb@systemasis.org>
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>

syslogd: add option to log to Linux kernel printk buffer

Why invent our own shared memory circular buffer when the kernel has a
perfectly fine one already?

This can be used as a smaller/simpler alternative to the syslogd IPC support
(as IPC shmem/klogd/logread aren't needed), while also allowing centralised
logging of everything (kernel messages, userspace bootup and syslog)
when used together with ttyprintk.

Notice that kernel 3.5+ is needed to store syslog facility in printk buffer,
otherwise only the priority is stored.

bloat-o-meter compared to IPC+klogd+logread:
function                                             old     new   delta
get_linux_version_code                                 -      84     +84
lbb_prepare                                           25      90     +65
applet_nameofs                                         6       -      -6
static.stdout@@GLIBC_2                                 8       -      -8
applet_names                                          23       9     -14
bb_msg_standard_output                                16       -     -16
init_sem                                              18       -     -18
xatoull_range                                         19       -     -19
overlapping_strcpy                                    21       -     -21
init_data                                             56      32     -24
applet_main                                           24       -     -24
main                                                 124      99     -25
full_write2_str                                       26       -     -26
error_exit                                            26       -     -26
bb_basename                                           30       -     -30
sem_up                                                32       -     -32
interrupted                                           35       -     -35
fflush_stdout_and_exit                                38       -     -38
bb_banner                                             46       -     -46
find_applet_by_name                                   59       -     -59
bb_signals_recursive_norestart                        90       -     -90
run_applet_no_and_exit                               104       -    -104
timestamp_and_log                                    651     523    -128
syslogd_main                                         798     581    -217
xstrtoull_range_sfx                                  267       -    -267
run_applet_and_exit                                  432       -    -432
klogd_main                                           490       -    -490
logread_main                                         508       -    -508
.rodata                                             1870     937    -933
bb_common_bufsiz1                                   8193       -   -8193
------------------------------------------------------------------------------
(add/remove: 2/26 grow/shrink: 1/6 up/down: 149/-11829)    Total: -11680 bytes

Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>
Signed-off-by: Mike Frysinger <vapier@gentoo.org>

klogd: handle multi-char log levels

Since Linux 3.5 (7ff9554bb5: printk: convert byte-buffer to variable-length
record buffer), klog buffer can now contain log lines with multi-char
loglevel indicators (<[0-9]+>) - So use strtoul to parse it.

function                                             old     new   delta
klogd_main                                           490     525     +35
------------------------------------------------------------------------------
(add/remove: 0/0 grow/shrink: 1/0 up/down: 35/0)               Total: 35 bytes

Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>
Signed-off-by: Mike Frysinger <vapier@gentoo.org>

dmesg: handle multi-char log levels

Since Linux 3.5 (7ff9554bb5: printk: convert byte-buffer to variable-length
record buffer), klog buffer can now contain log lines with multi-char
loglevel indicators (<[0-9]+>) - So we can no longer just skip 3 bytes.

Instead skip past the terminating '>' like util-linux does.

function                                             old     new   delta
dmesg_main                                           266     280     +13
------------------------------------------------------------------------------
(add/remove: 0/0 grow/shrink: 1/0 up/down: 13/0)               Total: 13 bytes

Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>
Signed-off-by: Mike Frysinger <vapier@gentoo.org>

powertop: fix error message

Application tries to use timer_stats module instead of cpufreq_stats.
Error message is printed if opening of the file /proc/timer_stats fails.

Signed-off-by: Lauri Hintsala <lauri.hintsala@bluegiga.com>
Signed-off-by: Mike Frysinger <vapier@gentoo.org>

ifenslave: fix missing close paren

Reported-by: David J Cozatt <ygdrasil@comcast.net>
Signed-off-by: Mike Frysinger <vapier@gentoo.org>

CONFIG_PID_FILE_PATH: new configuration option for pidfile paths

We set a default path for the directory where pidfiles are create
when FEATURE_PIDFILE is selected.  The default has no effect on
applets which must specify a pidfile path on the command line to
run, and it can be overridden by applets which optionally allow
the user to specify the pidfile path.

We also add pidfile write/remove support for klogd, ntpd and watchdog.
For syslogd, we add a missing remove_pidfile() for better cleanup
on daemon exit.

Signed-off-by: Anthony G. Basile <blueness@gentoo.org>
Signed-off-by: Mike Frysinger <vapier@gentoo.org>

du: document incompatibility with standard tool

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>

decompress_uncompress: comment out a bigger chunk of debug code

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>

decompress_uncompress: comment out debug printout on corrupted data

99% plus of all people who'll get corrupted archive wouldn't bother
debugging it. The rest can uncomment the code.

function                                             old     new   delta
unpack_Z_stream                                     1304    1234     -70

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>

decompress_uncompress: move 'code' variable into loop - sole user

Apparently, gcc does this optimization itself, since generated code is the same.

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>

Fix building under 4.2

Resolve overflow in move_to_unaligned16 and add missing header to
route.c

Change-Id: If1c4d30ac298faecea4493b839354cb891a92837

ash: fix a memory leak

The script which triggers the leak:

while true
  do
    while true
      do
        break;
    done</dev/null
done

Signed-off-by: Jon Tollefson <kniht@linux.vnet.ibm.com>
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>

volume_id: add exFAT detection

function                                             old     new   delta
volume_id_probe_exfat                                  -     294    +294

Signed-off-by: S-G Bergh <sgb@systemasis.org>
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>

ifupdown: code shrink

function                                             old     new   delta
execute                                              571     548     -23

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>

ifupdown: remove unused errno assignments

function                                             old     new   delta
execute                                              607     571     -36

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>

ifupdown: simple code shrink

function                                             old     new   delta
execute                                              631     607     -24

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>

blkid: show filesystem when both label and UUID are missing, but type is known

Signed-off-by: S-G Bergh <sgb@systemasis.org>
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>

android ndk defconfig: fix default location of modules

The default location of modules is currently /lib/modules which
we never use on android.  Some kernel modules need to be loaded on
boot with /system/bin or init.d scripts.

Revert "dd: support skip and seek values greater than 2G on android"

This reverts commit 6a025ba7107abe0c46a29c023ea54262c0dda26a, breaks
dealing with large storages, and it's making simple operations like
"/dd if=/dev/zero of=/dev/block/mmcblk0p3 seek=6144 count=75 bs=1"
fail in recovery

Change-Id: I246072ea7e614ee19fc421a8b11f9d11504c9f45

build system: use pkg-config to look up selinux libs

Newer versions of libselinux has started linking against more libs.
Rather than continuing hardcoding things, switch to using pkg-config
to query for its dependencies.

Signed-off-by: Mike Frysinger <vapier@gentoo.org>

fbsplash: fix regression from e4fa7b7

commit e4fa7b7 (limit progress bar flicker) made the progress bar
counter unsigned causing i < 0 to never come true.

Signed-off-by: Timo Teräs <timo.teras@iki.fi>
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>

nanddump: invert the meaning of the -o parameter to match upstream

Signed-off-by: Baruch Siach <baruch@tkos.co.il>
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>

nanddump: skip bad blocks when instructed to do so

Signed-off-by: Baruch Siach <baruch@tkos.co.il>
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>

nanddump: use the right operator of logic AND

Signed-off-by: Baruch Siach <baruch@tkos.co.il>
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>

mktemp: fix mktemp -u temp.XXXXXX returning garbage when TMPDIR is set

Use mktemp instead of tempnam for compatibility with real mktemp.
Don't let mktemp fail silently, print some simple error messages.
Don't ignore -q.

Signed-off-by: Tito Ragusa <farmatito@tiscali.it>
Signed-off-by: Ron Yorston <rmy@tigress.co.uk>
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>

adduser: make it accept "adduser USER GROUP" form

Signed-off-by: Tito Ragusa <farmatito@tiscali.it>
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>

mark get_shell_name FAST_FUNC

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>

top/ps: argv0:"gdm-session-worker [pam/gdm-password]" == comm:"gdm-session-wor"

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>

ash: implement export -n

function                                             old     new   delta
exportcmd                                            129     175     +46

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>

brctl: fix description and usage

brctl uncorrectly displays the "showmacs" command as being supported while
it is not, remove it from the usage and Config.src file.

Signed-off-by: Nicolas Thill <nico@openwrt.org>
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>

udhcpc[6]: allow discover_retries == 0 (infinite)

Signed-off-by: Felix Fietkau <nbd@openwrt.org>
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>

flashcp: remove unused variable opts

Signed-off-by: Manuel Zerpies <manuel.f.zerpies@ww.stud.uni-erlangen.de>
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>

lineedit: in !EDITING config, return -1 on fgets error

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>

fix unsafe bb_perror_msg(filename) calls

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>

scripts/kconfig/mconf: work on systems w/o SIGWINCH

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>

libbb: correctness/size tweaks in signal-related helpers

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>

top: do not touch stdin if -b

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>

modprobe_small: make rmmod to NOT remove dependencies. Closes 5162

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>

ash: trivial fixes for compile failures

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>