oweals/busybox.git
7 years agoSort some miscutils/ applets into coreutils or util-linux
Denys Vlasenko [Wed, 12 Apr 2017 11:58:40 +0000 (13:58 +0200)]
Sort some miscutils/ applets into coreutils or util-linux

No code changes

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
7 years agopartprobe: new applet
Denys Vlasenko [Tue, 11 Apr 2017 17:17:59 +0000 (19:17 +0200)]
partprobe: new applet

function                                             old     new   delta
partprobe_main                                         -      79     +79
packed_usage                                       31485   31511     +26
applet_names                                        2608    2618     +10
applet_main                                         1512    1516      +4
applet_install_loc                                   189     190      +1
------------------------------------------------------------------------------
(add/remove: 2/0 grow/shrink: 4/0 up/down: 120/0)             Total: 120 bytes

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
7 years agow: new applet, alias to "who -H"
Denys Vlasenko [Tue, 11 Apr 2017 16:17:03 +0000 (18:17 +0200)]
w: new applet, alias to "who -H"

function                                             old     new   delta
who_main                                             451     463     +12
packed_usage                                       31477   31485      +8
applet_main                                         1508    1512      +4
applet_names                                        2606    2608      +2
------------------------------------------------------------------------------
(add/remove: 0/0 grow/shrink: 4/0 up/down: 26/0)               Total: 26 bytes

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
7 years agofallocate: new applet
Denys Vlasenko [Tue, 11 Apr 2017 11:33:54 +0000 (13:33 +0200)]
fallocate: new applet

NAME
       fallocate - preallocate or deallocate space to a file
SYNOPSIS
       fallocate [-c|-p|-z] [-o offset] -l length [-n] filename
       fallocate -d [-o offset] [-l length] filename
DESCRIPTION
       fallocate  is  used  to manipulate the allocated disk space for a file,
       either to deallocate or preallocate it.  For filesystems which  support
       the  fallocate system call, preallocation is done quickly by allocating
       blocks and marking them as uninitialized, requiring no IO to  the  data
       blocks.   This  is  much faster than creating a file by filling it with
       zeroes.

function                                             old     new   delta
fallocate_main                                         -     179    +179
applet_names                                        2597    2606      +9
applet_main                                         1504    1508      +4
applet_suid                                           94      95      +1
applet_install_loc                                   188     189      +1

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
7 years agofsfreeze: new applet
Denys Vlasenko [Tue, 11 Apr 2017 09:53:05 +0000 (11:53 +0200)]
fsfreeze: new applet

NAME
       fsfreeze - suspend access to a filesystem (Ext3/4, ReiserFS, JFS, XFS)
SYNOPSIS
       fsfreeze --freeze|--unfreeze mountpoint
DESCRIPTION
       fsfreeze suspends or resumes access to a filesystem.
       fsfreeze halts any new access to the filesystem and creates a stable
       image on disk.
AVAILABILITY
       The fsfreeze command is part of the util-linux 2.28

function                                             old     new   delta
fsfreeze_main                                          -      81     +81
applet_names                                        2597    2606      +9
applet_main                                         1504    1508      +4
applet_suid                                           94      95      +1
applet_install_loc                                   188     189      +1

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
7 years agofactor: remove debug code
Denys Vlasenko [Tue, 11 Apr 2017 05:50:42 +0000 (07:50 +0200)]
factor: remove debug code

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
7 years agolibbb: move isqrt from factor, use it in diff too
Denys Vlasenko [Tue, 11 Apr 2017 05:34:56 +0000 (07:34 +0200)]
libbb: move isqrt from factor, use it in diff too

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
7 years agofactor: much faster, and very slightly larger isqrt()
Denys Vlasenko [Tue, 11 Apr 2017 05:02:42 +0000 (07:02 +0200)]
factor: much faster, and very slightly larger isqrt()

function                                             old     new   delta
isqrt_odd                                             70      88     +18

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
7 years agofactor: don't be too clever in isqrt - be small instead
Denys Vlasenko [Mon, 10 Apr 2017 16:30:35 +0000 (18:30 +0200)]
factor: don't be too clever in isqrt - be small instead

function                                             old     new   delta
isqrt_odd                                            111      70     -41

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
7 years agofactor: add code to test isqrt() correctness
Denys Vlasenko [Mon, 10 Apr 2017 11:53:57 +0000 (13:53 +0200)]
factor: add code to test isqrt() correctness

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
7 years agofactor: 25% faster sieving
Denys Vlasenko [Mon, 10 Apr 2017 10:10:46 +0000 (12:10 +0200)]
factor: 25% faster sieving

function                                             old     new   delta
factorize                                            287     260     -27

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
7 years agofactor: better comments, slightl more clever conversion even->odd
Denys Vlasenko [Mon, 10 Apr 2017 09:47:48 +0000 (11:47 +0200)]
factor: better comments, slightl more clever conversion even->odd

function                                             old     new   delta
isqrt_odd                                            114     111      -3

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
7 years agofactor: fix HALF_FMT usage
Denys Vlasenko [Mon, 10 Apr 2017 08:13:14 +0000 (10:13 +0200)]
factor: fix HALF_FMT usage

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
7 years agofactor: fix comments
Denys Vlasenko [Sun, 9 Apr 2017 22:41:11 +0000 (00:41 +0200)]
factor: fix comments

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
7 years agofactor: a bit more simple isqrt
Denys Vlasenko [Sun, 9 Apr 2017 22:28:26 +0000 (00:28 +0200)]
factor: a bit more simple isqrt

function                                             old     new   delta
isqrt_odd                                             87      80      -7

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
7 years agofactor: simpler isqrt
Denys Vlasenko [Sun, 9 Apr 2017 22:24:16 +0000 (00:24 +0200)]
factor: simpler isqrt

function                                             old     new   delta
isqrt_odd                                            102      87     -15

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
7 years agofactor: factor2 variable is unused now, drop it
Denys Vlasenko [Sun, 9 Apr 2017 21:19:47 +0000 (23:19 +0200)]
factor: factor2 variable is unused now, drop it

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
7 years agofactor: expand comments
Denys Vlasenko [Sun, 9 Apr 2017 21:12:38 +0000 (23:12 +0200)]
factor: expand comments

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
7 years agofactor: 30% faster code (estimate max possible factor just once)
Denys Vlasenko [Sun, 9 Apr 2017 20:54:57 +0000 (22:54 +0200)]
factor: 30% faster code (estimate max possible factor just once)

function                                             old     new   delta
factorize                                              -     161    +161
isqrt_odd                                              -     102    +102
factor_main                                          281     110    -171
------------------------------------------------------------------------------
(add/remove: 2/0 grow/shrink: 0/1 up/down: 263/-171)           Total: 92 bytes

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
7 years agonl: fix copyright message
Denys Vlasenko [Sun, 9 Apr 2017 19:19:29 +0000 (21:19 +0200)]
nl: fix copyright message

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
7 years agofactor: new applet
Denys Vlasenko [Sun, 9 Apr 2017 19:18:43 +0000 (21:18 +0200)]
factor: new applet

thus far only able to factor up to ULLONG_MAX

function                                             old     new   delta
factor_main                                            -     378    +378
packed_usage                                       31427   31502     +75
applet_names                                        2590    2597      +7
applet_main                                         1500    1504      +4
------------------------------------------------------------------------------
(add/remove: 2/0 grow/shrink: 3/0 up/down: 464/0)             Total: 464 bytes

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
7 years agonproc: new applet
Denys Vlasenko [Fri, 7 Apr 2017 19:47:53 +0000 (21:47 +0200)]
nproc: new applet

function                                             old     new   delta
nproc_main                                             -      98     +98
applet_names                                        2584    2590      +6
applet_main                                         1496    1500      +4
applet_install_loc                                   187     188      +1

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
7 years agoshred: smaller code
Denys Vlasenko [Fri, 7 Apr 2017 19:10:00 +0000 (21:10 +0200)]
shred: smaller code

function                                             old     new   delta
shred_main                                           361     356      -5

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
7 years agoshred: new applet
Denys Vlasenko [Fri, 7 Apr 2017 18:45:08 +0000 (20:45 +0200)]
shred: new applet

function                                             old     new   delta
shred_main                                             -     361    +361
packed_usage                                       31427   31467     +40
applet_names                                        2578    2584      +6
applet_main                                         1492    1496      +4
run_applet_and_exit                                  679     682      +3
------------------------------------------------------------------------------
(add/remove: 2/0 grow/shrink: 4/0 up/down: 414/0)             Total: 414 bytes

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
7 years agoinit: dont send "Sent SIGTERM to all" msg to syslog: we just TERMed it!
Denys Vlasenko [Fri, 7 Apr 2017 17:13:09 +0000 (19:13 +0200)]
init: dont send "Sent SIGTERM to all" msg to syslog: we just TERMed it!

While at it, explain FEATURE_INIT_SYSLOG

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
7 years agoipaddress: tweak help text more: mention "brd +" syntax
Denys Vlasenko [Fri, 7 Apr 2017 16:22:41 +0000 (18:22 +0200)]
ipaddress: tweak help text more: mention "brd +" syntax

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
7 years agoipaddress: tweak help text
Denys Vlasenko [Fri, 7 Apr 2017 16:14:46 +0000 (18:14 +0200)]
ipaddress: tweak help text

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
7 years agoiproute: add "a" command as a synonym to "add"
Denys Vlasenko [Fri, 7 Apr 2017 15:33:26 +0000 (17:33 +0200)]
iproute: add "a" command as a synonym to "add"

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
7 years agolibiproute: FACT_FUNCization
Denys Vlasenko [Fri, 7 Apr 2017 15:15:40 +0000 (17:15 +0200)]
libiproute: FACT_FUNCization

function                                             old     new   delta
get_addr_1                                           258     287     +29
next_arg                                               -      22     +22
print_neigh                                          885     902     +17
do_iplink                                           1269    1276      +7
get_unsigned                                          54      55      +1
get_u32                                               54      55      +1
get_u16                                               62      63      +1
rt_addr_n2a                                           53      52      -1
ipaddr_modify                                       1229    1226      -3
get_addr32                                            58      54      -4
invarg_1_to_2                                         18      12      -6
inet_addr_match                                      109     103      -6
duparg2                                               18      12      -6
duparg                                                18      12      -6
print_addrinfo                                      1231    1223      -8
ipneigh_list_or_flush                                724     714     -10
incomplete_command                                    10       -     -10
iproute_get                                          847     835     -12
ipaddr_list_or_flush                                1265    1253     -12
get_prefix                                           356     344     -12
do_add_or_delete                                    1162    1150     -12
get_addr                                              73      58     -15
iproute_list_or_flush                               1210    1191     -19
parse_args                                          1434    1412     -22
print_rule                                           634     610     -24
iprule_modify                                        864     816     -48
print_route                                         1621    1565     -56
iproute_modify                                      1221    1164     -57
------------------------------------------------------------------------------
(add/remove: 1/1 grow/shrink: 6/20 up/down: 78/-349)         Total: -271 bytes

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
7 years agoiproute: support advmss option
Denys Vlasenko [Fri, 7 Apr 2017 15:00:53 +0000 (17:00 +0200)]
iproute: support advmss option

function                                             old     new   delta
iproute_modify                                      1164    1221     +57
str_is_lock                                            -      22     +22
packed_usage                                       31372   31382     +10
do_iproute                                           157     132     -25
------------------------------------------------------------------------------
(add/remove: 1/0 grow/shrink: 2/1 up/down: 89/-25)             Total: 64 bytes

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
7 years agoiproute: update help text, add commented-out code for unsupported options
Denys Vlasenko [Thu, 6 Apr 2017 15:54:38 +0000 (17:54 +0200)]
iproute: update help text, add commented-out code for unsupported options

function                                             old     new   delta
packed_usage                                       31327   31372     +45
do_iproute                                           132     157     +25
iproute_modify                                      1162    1164      +2
------------------------------------------------------------------------------
(add/remove: 0/0 grow/shrink: 3/0 up/down: 72/0)               Total: 72 bytes

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
7 years agoiplink: implement "set promisc on|off". Closes 4682
Denys Vlasenko [Thu, 6 Apr 2017 14:06:04 +0000 (16:06 +0200)]
iplink: implement "set promisc on|off". Closes 4682

function                                             old     new   delta
do_iplink                                           1232    1269     +37
packed_usage                                       31337   31327     -10

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
7 years agoip link: tidying up
Denys Vlasenko [Thu, 6 Apr 2017 13:55:29 +0000 (15:55 +0200)]
ip link: tidying up

Misplaced comment. --help was wrong: dynamic on|off is not supported.

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
7 years agomodprobe: do not descend into /etc/modprobe.d/DIR/. Closes 8686
Denys Vlasenko [Thu, 6 Apr 2017 13:22:24 +0000 (15:22 +0200)]
modprobe: do not descend into /etc/modprobe.d/DIR/. Closes 8686

Also expanded comments in recursive_action.c

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
7 years agomodprobe-small: fix "modprobe non-existing-module" exitcode (should be 1)
Denys Vlasenko [Thu, 6 Apr 2017 11:45:00 +0000 (13:45 +0200)]
modprobe-small: fix "modprobe non-existing-module" exitcode (should be 1)

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
7 years agomodprobe: read modules.builtin
Ben Hutchings [Thu, 6 Apr 2017 09:54:04 +0000 (11:54 +0200)]
modprobe: read modules.builtin

This allows explicit probing to succeed when the requested module
is actually built-in, and corrects the error message for removal.

Signed-off-by: Ben Hutchings <ben@decadent.org.uk>
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
7 years agoexamples: make udhcpc script handle /32 netmasks
Denys Vlasenko [Wed, 5 Apr 2017 22:53:43 +0000 (00:53 +0200)]
examples: make udhcpc script handle /32 netmasks

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
7 years agoAdd help text for 'uniq -i'
Jody Bruchon [Thu, 30 Mar 2017 16:24:56 +0000 (12:24 -0400)]
Add help text for 'uniq -i'

Signed-off-by: Jody Bruchon <jody@jodybruchon.com>
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
7 years agonl: use CONFIG_NL, not CONFIG_UNIQ
Denys Vlasenko [Wed, 5 Apr 2017 16:31:19 +0000 (18:31 +0200)]
nl: use CONFIG_NL, not CONFIG_UNIQ

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
7 years agonl: new applet; also implement cat -nb (similar functionality to nl)
Denys Vlasenko [Wed, 5 Apr 2017 16:17:17 +0000 (18:17 +0200)]
nl: new applet; also implement cat -nb (similar functionality to nl)

function                                             old     new   delta
nl_main                                                -     201    +201
print_numbered_lines                                   -     115    +115
cat_main                                              36     149    +113
static.nl_longopts                                     -     106    +106
packed_usage                                       31081   31182    +101
applet_main                                         1488    1492      +4
applet_names                                        2575    2578      +3
applet_suid                                           93      94      +1
applet_install_loc                                   186     187      +1
------------------------------------------------------------------------------
(add/remove: 4/0 grow/shrink: 6/0 up/down: 645/0)             Total: 645 bytes

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
7 years agotls: do not compile in TLS_RSA_WITH_NULL_SHA256 code if unreachable
Denys Vlasenko [Mon, 3 Apr 2017 23:41:15 +0000 (01:41 +0200)]
tls: do not compile in TLS_RSA_WITH_NULL_SHA256 code if unreachable

function                                             old     new   delta
tls_handshake                                       1595    1588      -7
xwrite_encrypted                                     244     209     -35

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
7 years agotls: avoid using int16 in pstm code
Denys Vlasenko [Mon, 3 Apr 2017 19:53:29 +0000 (21:53 +0200)]
tls: avoid using int16 in pstm code

function                                             old     new   delta
pstm_div                                            1472    1522     +50
psRsaEncryptPub                                      403     413     +10
pstm_2expt                                            91      96      +5
pstm_clear                                            68      72      +4
pstm_init                                             39      42      +3
pstm_unsigned_bin_size                                36      37      +1
pstm_montgomery_reduce                               398     399      +1
pstm_init_size                                        45      46      +1
pstm_zero                                             39      38      -1
pstm_set                                              35      34      -1
pstm_read_unsigned_bin                               112     109      -3
pstm_mulmod                                          123     120      -3
pstm_mod                                             116     113      -3
pstm_cmp                                              57      54      -3
pstm_sub                                             107     102      -5
pstm_to_unsigned_bin                                 157     151      -6
pstm_clamp                                            63      57      -6
pstm_add                                             116     108      -8
pstm_grow                                             81      72      -9
pstm_count_bits                                       57      48      -9
pstm_init_copy                                        84      72     -12
pstm_cmp_mag                                          93      78     -15
pstm_sqr_comba                                       567     551     -16
pstm_montgomery_calc_normalization                   158     140     -18
pstm_copy                                            115      92     -23
pstm_lshd                                            133     109     -24
pstm_mul_comba                                       525     500     -25
pstm_mul_d                                           251     224     -27
s_pstm_sub                                           256     228     -28
s_pstm_add                                           370     337     -33
pstm_div_2d                                          444     409     -35
pstm_mul_2                                           195     156     -39
pstm_rshd                                            154     104     -50
pstm_mul_2d                                          247     186     -61
pstm_exptmod                                        1524    1463     -61
------------------------------------------------------------------------------
(add/remove: 0/0 grow/shrink: 8/27 up/down: 75/-524)         Total: -449 bytes

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
7 years agotls: merge sha1 and sha256 hmac functions
Denys Vlasenko [Mon, 3 Apr 2017 15:43:44 +0000 (17:43 +0200)]
tls: merge sha1 and sha256 hmac functions

function                                             old     new   delta
hmac_begin                                             -     196    +196
hmac_sha256                                           61      68      +7
hmac                                                 250      87    -163
hmac_sha256_begin                                    190       -    -190
------------------------------------------------------------------------------
(add/remove: 1/1 grow/shrink: 1/1 up/down: 203/-353)         Total: -150 bytes

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
7 years agotls: replace aes encryption/decryption by much smaller one
Denys Vlasenko [Sat, 1 Apr 2017 18:11:59 +0000 (20:11 +0200)]
tls: replace aes encryption/decryption by much smaller one

The replacement code is ~6 times slower,
but drastically decreases size of tls_aes.o:

text data bss dec hex filename
8050 0 0 8050 1f72 tls_aes_OLD.o
2461 0 0 2461 99d tls_aes.o

function                                             old     new   delta
sbox                                                   -     256    +256
rsbox                                                  -     256    +256
KeyExpansion                                           -     197    +197
Subword                                                -      66     +66
AddRoundKey                                            -      61     +61
static.Rcon                                            -      10     +10
rcon                                                  40       -     -40
setup_mix                                             80       -     -80
setup_mix2                                           123       -    -123
aes_cbc_decrypt                                     1377     971    -406
aes_cbc_encrypt                                     1375     644    -731
psAesInit                                            848       -    -848
Te4                                                 1024       -   -1024
TE0                                                 1024       -   -1024
TD0                                                 1024       -   -1024
Td4                                                 1040       -   -1040
------------------------------------------------------------------------------
(add/remove: 6/8 grow/shrink: 0/2 up/down: 846/-6340)       Total: -5494 bytes

This code is based on public domain "tiny-AES128-C" code.

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
7 years agouniq: add -i option to ignore case
Jody Bruchon [Thu, 30 Mar 2017 16:15:54 +0000 (18:15 +0200)]
uniq: add -i option to ignore case

Signed-off-by: Jody Bruchon <jody@jodybruchon.com>
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
7 years agoUpdate depmod & modprobe upstream help text in comments
Kang-Che Sung [Thu, 9 Feb 2017 14:48:51 +0000 (22:48 +0800)]
Update depmod & modprobe upstream help text in comments

No code changes.

Signed-off-by: Kang-Che Sung <explorer09@gmail.com>
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
7 years agoudhcp6: move misplaced comment
Denys Vlasenko [Mon, 27 Mar 2017 20:49:12 +0000 (22:49 +0200)]
udhcp6: move misplaced comment

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
7 years agoudhcp6: fix releasing
Denys Vlasenko [Mon, 27 Mar 2017 20:32:44 +0000 (22:32 +0200)]
udhcp6: fix releasing

Patch is based on work by tiggerswelt.net. They say:
"
We wanted udhcpc6 to release its IPv6-Addresses on
quit (-R-commandline-option) which turned out to generate once again
kind of garbage on the network-link.
We tracked this down to two issues:

 - udhcpc6 uses a variable called "srv6_buf" to send packets to
   the dhcp6-server, but this variable is never initialized correctly
   and contained kind of a garbage-address

 - The address of the dhcp6-server is usually a link-local-address,
   that requires an interface-index when using connect() on an AF_INET6-
   socket

We added an
additional parameter for ifindex to d6_send_kernel_packet() and made
d6_recv_raw_packet() to capture the address of the dhcp6-server and
forward it to its callee.
"

Three last patches together:

function                                             old     new   delta
d6_read_interface                                      -     454    +454
d6_recv_raw_packet                                     -     283    +283
option_to_env                                        249     504    +255
.rodata                                           165226  165371    +145
send_d6_discover                                     195     237     +42
send_d6_select                                       118     159     +41
send_d6_renew                                        173     186     +13
send_d6_release                                      162     173     +11
opt_req                                                -      10     +10
d6_send_kernel_packet                                304     312      +8
opt_fqdn_req                                           -       6      +6
d6_mcast_from_client_config_ifindex                   48      51      +3
d6_find_option                                        63      61      -2
udhcpc6_main                                        2416    2411      -5
static.d6_recv_raw_packet                            266       -    -266
------------------------------------------------------------------------------
(add/remove: 5/1 grow/shrink: 8/2 up/down: 1271/-273)         Total: 998 bytes

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
7 years agoudhcp6: fix problems found running against dnsmasq
Denys Vlasenko [Mon, 27 Mar 2017 20:22:09 +0000 (22:22 +0200)]
udhcp6: fix problems found running against dnsmasq

Patch is based on work by tiggerswelt.net. They say:
"
But when we tried to use dnsmasq on server-side, udhcpc6 was unable to
forward the acquired address to its setup-script although the
IPv6-Address had been assigned by the server as we could see via
tcpdump. We traced this issue down to a problem on how udhcpc6 parses
DHCPv6-Options: When moving to next option, a pointer-address is
increased and a length buffer is decreased by the length of the option.
The problem is that it is done in this order:

  option += 4 + option[3];
  len_m4 -= 4 + option[3];

But this has to be switched as the length is decreased by the length of
the *next* option, not the current one. This affected both - internal
checks if a required option is present and the function to expose
options to the environment of the setup-script.
There was also a bug parsing D6_OPT_STATUS_CODE Options, that made
dnsmasq not work as udhcpc6 thought it is receiving a non-positive
status-code (because it did not parse the status-code as required in RFC
3315).
In addition we introduced basic support for RFC 3646 (OPTION_DNS_SERVERS
and OPTION_DOMAIN_LIST) and RFC 4704 (OPTION_CLIENT_FQDN).
"

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
7 years agoudhcp6: read_interface should save link-local ipv6 address
Denys Vlasenko [Mon, 27 Mar 2017 20:10:15 +0000 (22:10 +0200)]
udhcp6: read_interface should save link-local ipv6 address

Patch is based on work by tiggerswelt.net. They say:
"Using this patch it was no problem to acquire an IPv6-Address via DHCPv6
using ISC DHCPD6 on server-side."

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
7 years agovi: avoid touching a new file with ZZ when no editing has been done
Yousong Zhou [Fri, 24 Mar 2017 20:13:10 +0000 (21:13 +0100)]
vi: avoid touching a new file with ZZ when no editing has been done

This is the behaviour observed with standard vim and busybox vi of at
least 1.22.1.  It was changed with commit "32afd3a vi: some
simplifications" which happened before 1.23.0.

Mistyping filename on command line happens fairly often and it's better
we restore the old behaviour to avoid a few unnecessary flash writes and
sometimes efforts of debugging bugs caused by those unneeded stray
files.

Signed-off-by: Yousong Zhou <yszhou4tech@gmail.com>
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
7 years agoswitch_root: don't bail out when console doesn't exist
Andrei Gherzan [Fri, 24 Mar 2017 15:39:08 +0000 (16:39 +0100)]
switch_root: don't bail out when console doesn't exist

Busybox is very often used in initramfs at the end of which usually
there is a switch_root to the actual rootfs. There are many cases where
the console kernel argument is either just a placeholder (for example
RaspberryPi uses serial0 and serial1) or configured as null to avoid any
console messages - usually you would see such of a setup in production
environments.

Currently busybox bails out if can't open the console argument. If this
happenes in initramfs and if the console=null for example, you get in a
blind kernel panic. Avoid this by only warning instead of dying.

function                                             old     new   delta
switch_root_main                                     371     368      -3

Signed-off-by: Andrei Gherzan <andrei@gherzan.com>
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
7 years agowhitespace fix
Denys Vlasenko [Fri, 24 Mar 2017 14:00:12 +0000 (15:00 +0100)]
whitespace fix

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
7 years agopaste: delimiter list use should restart for each new output line
Denys Vlasenko [Thu, 23 Mar 2017 16:58:32 +0000 (17:58 +0100)]
paste: delimiter list use should restart for each new output line

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
7 years agotypo fix
Denys Vlasenko [Thu, 23 Mar 2017 16:41:59 +0000 (17:41 +0100)]
typo fix

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
7 years agopaste: new applet
Maxime Coste [Thu, 23 Mar 2017 16:35:20 +0000 (17:35 +0100)]
paste: new applet

function                                             old     new   delta
paste_main                                             -     493    +493
packed_usage                                       31019   31070     +51
applet_names                                        2569    2575      +6
applet_main                                         1484    1488      +4
------------------------------------------------------------------------------
(add/remove: 2/0 grow/shrink: 3/0 up/down: 554/0)             Total: 554 bytes

Signed-off-by: Maxime Coste <mawww@kakoune.org>
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
7 years agodd: call fsync() only once before exiting if conv=fsync is specified
Rostislav Skudnov [Thu, 16 Mar 2017 19:54:35 +0000 (20:54 +0100)]
dd: call fsync() only once before exiting if conv=fsync is specified

Signed-off-by: Rostislav Skudnov <rostislav@tuxera.com>
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
7 years agoRevert "umount: make -d always active, add -D to suppress it"
Denys Vlasenko [Thu, 16 Mar 2017 16:45:07 +0000 (17:45 +0100)]
Revert "umount: make -d always active, add -D to suppress it"

This reverts commit 86a03bee1d3d6990c03bf500836b19ec8a1c1f12.

Since now our "mount -oloop" creates AUTOCLEARed loopdevs, we no longer
need our umount to destroy loopdevs to match the usual util-linux behaviour.

Now this revert fixes another, opposite bug: "explicit" mount /dev/loopN
and then umount must not drop loopdevs!

User complaint is as follows:

It seems LOOP_CLR_FD called on a loop-*partition* removes the mapping of
the whole *device* - which results in the following:

root@LEDE:/# loop=$(losetup -f)
root@LEDE:/# echo ${loop}
/dev/loop2
root@LEDE:/# losetup ${loop} /IMAGE
root@LEDE:/# ls -l ${loop}*
brw-------  1 root root     7,   2 Mar  6 20:09 /dev/loop2
root@LEDE:/# partprobe ${loop}
root@LEDE:/# ls -l ${loop}*
brw-------  1 root  root    7,   2 Mar  6 20:09 /dev/loop2
brw-------  1 root  root  259,   8 Mar  6 21:59 /dev/loop2p1
brw-------  1 root  root  259,   9 Mar  6 21:59 /dev/loop2p2
brw-------  1 root  root  259,  10 Mar  6 21:59 /dev/loop2p3
brw-------  1 root  root  259,  11 Mar  6 21:59 /dev/loop2p4
brw-------  1 root  root  259,  12 Mar  6 21:59 /dev/loop2p5
brw-------  1 root  root  259,  13 Mar  6 21:59 /dev/loop2p6
brw-------  1 root  root  259,  14 Mar  6 21:59 /dev/loop2p7
brw-------  1 root  root  259,  15 Mar  6 21:59 /dev/loop2p8
root@LEDE:/# mount ${loop}p8 /MOUNT       # mount loop partition
root@LEDE:/# losetup -a | grep $loop      # loop dev mapping still there
/dev/loop2: 0 /mnt/IMAGE
root@LEDE:/# strace umount /MOUNT 2> /log # unmount loop partition
root@LEDE:/# losetup -a | grep ${loop}    # loop device mapping is gone
root@LEDE:/# grep -i loop /log
open("/dev/loop2p7", O_RDONLY|O_LARGEFILE) = 3
ioctl(3, LOOP_CLR_FD)                   = 0
root@LEDE:/#

The strace was done to figure out, if maybe umount wrongly ioctl()'s the
parent device instead of the partition - it doesn't.

I already wasn't a fan of umount implicitly removing the mapping in the
first place (as I usually setup and release loop devices with `losetup`
and scripts needed to call umount differently in order to work and
outside busybox).

However taking above (kernel-)behaviour into account - umount calling
ioctl(LOOP_CLR_FD) unconditionally potentially causes some nasty side
effects

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
7 years agomount: create loop devices with LO_FLAGS_AUTOCLEAR flag
Denys Vlasenko [Thu, 16 Mar 2017 15:49:37 +0000 (16:49 +0100)]
mount: create loop devices with LO_FLAGS_AUTOCLEAR flag

The "autolooped" mount (mount [-oloop] IMAGE /DIR/DIR)
always creates AUTOCLEARed loopdevs, so that umounting
drops them (and this does not require any code in the
umount userspace).
This happens since circa linux-2.6.25:
commit 96c5865559cee0f9cbc5173f3c949f6ce3525581
Date:    Wed Feb 6 01:36:27 2008 -0800
Subject: Allow auto-destruction of loop devices
IOW: in this case, umount does not have to use -d
to drop the loopdev.

The explicit loop mount (mount /dev/loopN /DIR/DIR)
does not do this. In this case, umount without -d
should not drop loopdev.
Unfortunately, bbox umount currently always implies -d,
this probably needs fixing.

function                                             old     new   delta
set_loop                                             537     597     +60
singlemount                                         1101    1138     +37
losetup_main                                         419     432     +13
------------------------------------------------------------------------------
(add/remove: 0/0 grow/shrink: 3/0 up/down: 110/0)             Total: 110 bytes

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
7 years agorunit: fix chpst -n -N -u USER
Denys Vlasenko [Mon, 13 Mar 2017 21:32:05 +0000 (22:32 +0100)]
runit: fix chpst -n -N -u USER

busybox's chpst first switches user/group and then tries to call nice().
Once the root priviledges are dropped, process priority can only be lowered.
So negative nice values don't work anymore.
Upstream version of chpst correctly calls nice() before switching user.

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
7 years agovi: don't touch file with :x when modified_count == 0
Denys Vlasenko [Mon, 13 Mar 2017 19:50:42 +0000 (20:50 +0100)]
vi: don't touch file with :x when modified_count == 0

Along with it, there are other changes

 - Check for uppercase X is removed as the expression will be always false and
   :X itself is another totally different command in standard vim
 - The status line will show number of written lines instead of lines requested
   by the colon command.  This is also how the standard vim is doing, though
   the difference is that '!' has to be explicitly specified in vim to allow
   partial writes

Signed-off-by: Yousong Zhou <yszhou4tech@gmail.com>
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
7 years agoreset: before calling execvp(), reset needs to flush stdout
Glenn Matthews [Fri, 17 Feb 2017 22:01:13 +0000 (23:01 +0100)]
reset: before calling execvp(), reset needs to flush stdout

Signed-off-by: Glenn Matthews <glmatthe@cisco.com>
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
7 years agoFix FEATURE_{GZIP,BZIP2}_DECOMPRESS link error
Kang-Che Sung [Fri, 17 Feb 2017 11:04:24 +0000 (19:04 +0800)]
Fix FEATURE_{GZIP,BZIP2}_DECOMPRESS link error

... when gzip is selected but not gunzip nor zcat, or when bzip2 is
selected but not bunzip2 nor bzcat.

This regression is introduced in b130f9f758b6404c6d0911a1c120937ae6ab47f8
("Allow 'gzip -d' and 'bzip2 -d' without gunzip or bunzip2")

Signed-off-by: Kang-Che Sung <explorer09@gmail.com>
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
7 years agoudhcpc: make sure we do not overflow poll timeout
Denys Vlasenko [Fri, 17 Feb 2017 18:20:32 +0000 (19:20 +0100)]
udhcpc: make sure we do not overflow poll timeout

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
7 years agonc_bloaty: use poll() instead of select()
Denys Vlasenko [Fri, 17 Feb 2017 16:48:59 +0000 (17:48 +0100)]
nc_bloaty: use poll() instead of select()

function                                             old     new   delta
readwrite                                            829     715    -114

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
7 years agoudhcp: use poll() instead of select()
Denys Vlasenko [Thu, 16 Feb 2017 22:25:44 +0000 (23:25 +0100)]
udhcp: use poll() instead of select()

function                                             old     new   delta
udhcp_sp_read                                         65      46     -19
udhcp_sp_fd_set                                       79      54     -25
udhcpd_main                                         1530    1482     -48
udhcpc_main                                         2780    2730     -50
------------------------------------------------------------------------------
(add/remove: 0/0 grow/shrink: 0/4 up/down: 0/-142)           Total: -142 bytes

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
7 years agoudhcp: do not clobber errno by signal handler
Denys Vlasenko [Thu, 16 Feb 2017 19:04:19 +0000 (20:04 +0100)]
udhcp: do not clobber errno by signal handler

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
7 years agonc: use poll() instead of select()
Denys Vlasenko [Thu, 16 Feb 2017 16:17:17 +0000 (17:17 +0100)]
nc: use poll() instead of select()

function                                             old     new   delta
nc_main                                              943     866     -77

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
7 years agotls: covert i/o loop from using select() to poll()
Denys Vlasenko [Thu, 16 Feb 2017 15:27:39 +0000 (16:27 +0100)]
tls: covert i/o loop from using select() to poll()

function                                             old     new   delta
tls_run_copy_loop                                    377     282     -95

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
7 years agosendmail: allow "+" symbol in recipient. Closes 9646
Denys Vlasenko [Sun, 12 Feb 2017 20:57:22 +0000 (21:57 +0100)]
sendmail: allow "+" symbol in recipient. Closes 9646

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
7 years agomodprobe-small: define and use DEPMOD_OPT_n (option mask)
Kang-Che Sung [Thu, 9 Feb 2017 14:48:52 +0000 (22:48 +0800)]
modprobe-small: define and use DEPMOD_OPT_n (option mask)

Signed-off-by: Kang-Che Sung <explorer09@gmail.com>
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
7 years agomodprobe-small: document '-n' in depmod usage
Kang-Che Sung [Thu, 9 Feb 2017 14:48:50 +0000 (22:48 +0800)]
modprobe-small: document '-n' in depmod usage

Signed-off-by: Kang-Che Sung <explorer09@gmail.com>
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
7 years agomodprobe_small: if only MODPROBE and DEPMOD are selected, no need to test for them
Denys Vlasenko [Tue, 7 Feb 2017 15:41:25 +0000 (16:41 +0100)]
modprobe_small: if only MODPROBE and DEPMOD are selected, no need to test for them

function                                             old     new   delta
modprobe_main                                        321     306     -15

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
7 years agoReorder modutils config options & fix yet more dependency
Kang-Che Sung [Wed, 1 Feb 2017 11:43:54 +0000 (19:43 +0800)]
Reorder modutils config options & fix yet more dependency

- modprobe can indirectly benefit from FEATURE_2_4_MODULES and
  FEATURE_INSMOD_TRY_MAP options.
- The position of config FEATURE_INSMOD_TRY_MMAP prevented some other
  config options from indenting under FEATURE_2_4_MODULES. Reorder to
  fix this.
- FEATURE_MODPROBE_SMALL_CHECK_ALREADY_LOADED is now moved to
  Config.src under "Common options" section. (I wished to edit this
  config so that it also work with "big" modutils, but it's not done at
  the moment. Sorry.)

Signed-off-by: Kang-Che Sung <explorer09@gmail.com>
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
7 years agomodutils: fix config options dependency (2)
Kang-Che Sung [Sun, 5 Feb 2017 18:02:34 +0000 (19:02 +0100)]
modutils: fix config options dependency (2)

- The modprobe-small implementation of rmmod no longer chdir's to
  "/lib/modules/`uname -r`" as it was not necessary for rmmod's
  operation. (And it no longer need to die if such modules directory
  doesn't exist.)
- Configs DEFAULT_MODULES_DIR and DEFAULT_DEPMOD_FILE no longer depend
  on MODPROBE_SMALL as the latter may not enable depmod or modprobe
  that requires these configs.
- Clarify DEFAULT_DEPMOD_FILE's description regarding the ".bb" name
  suffix.

Signed-off-by: Kang-Che Sung <explorer09@gmail.com>
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
7 years agohttpd: use "Content-Length", not "-length"
Denys Vlasenko [Sat, 4 Feb 2017 22:19:30 +0000 (23:19 +0100)]
httpd: use "Content-Length", not "-length"

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
7 years agoReplace int -> uint to avoid signed integer overflow
Rostislav Skudnov [Wed, 1 Feb 2017 18:35:13 +0000 (18:35 +0000)]
Replace int -> uint to avoid signed integer overflow

An example of such an error (should be compiled with DEBUG_SANITIZE):

runtime error: left shift of 1 by 31 places cannot be represented in
type 'int'

Signed-off-by: Rostislav Skudnov <rostislav@tuxera.com>
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
7 years agotls: fold AES CBC en/decryption into single functions
Denys Vlasenko [Sat, 4 Feb 2017 15:23:49 +0000 (16:23 +0100)]
tls: fold AES CBC en/decryption into single functions

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
7 years agowget/tls: session_id of zero length is ok (arxiv.org responds with such)
Denys Vlasenko [Fri, 3 Feb 2017 17:19:59 +0000 (18:19 +0100)]
wget/tls: session_id of zero length is ok (arxiv.org responds with such)

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
7 years agoash: add INT_OFF/ON around allocations
Denys Vlasenko [Fri, 3 Feb 2017 12:00:06 +0000 (13:00 +0100)]
ash: add INT_OFF/ON around allocations

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
7 years agops: avoid -o stat to contain spaces. Closes 9631
Denys Vlasenko [Thu, 2 Feb 2017 12:04:30 +0000 (13:04 +0100)]
ps: avoid -o stat to contain spaces. Closes 9631

function                                             old     new   delta
procps_scan                                         1227    1236      +9

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
7 years agocmdline module options can be disabled on "big" modutils
Kang-Che Sung [Tue, 31 Jan 2017 13:09:17 +0000 (21:09 +0800)]
cmdline module options can be disabled on "big" modutils

Allow module options on command line to be disabled on "big" modutils.

Config FEATURE_MODPROBE_SMALL_OPTIONS_ON_CMDLINE is renamed to
FEATURE_CMDLINE_MODULE_OPTIONS and no longer depends on !MODPROBE_SMALL

(I'm not sure if disabling this is useful on "big" modutils, but at
least the macro can serve as a marker and ensure both implementations
of same feature have consistent behavior.)

Signed-off-by: Kang-Che Sung <explorer09@gmail.com>
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
7 years agomodutils: remove redundant "select PLATFORM_LINUX" configs
Kang-Che Sung [Wed, 1 Feb 2017 00:43:32 +0000 (01:43 +0100)]
modutils: remove redundant "select PLATFORM_LINUX" configs

It is enough to have only applets' configs select PLATFORM_LINUX.

Signed-off-by: Kang-Che Sung <explorer09@gmail.com>
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
7 years agomodprobe-small: move lsmod code out of modprobe_main()
Kang-Che Sung [Tue, 31 Jan 2017 09:06:43 +0000 (17:06 +0800)]
modprobe-small: move lsmod code out of modprobe_main()

Having lsmod code inside modprobe_main() makes some of the applet name
checking code awkward. Besides, this make busybox x86_64 binary a few
bytes smaller. :)

function                                  old     new   delta
lsmod_main                                  -      23     +23
modprobe_main                             599     564     -35
-------------------------------------------------------------------
(add/remove: 1/0 grow/shrink: 0/1 up/down: 23/-35) Total: -12 bytes

Signed-off-by: Kang-Che Sung <explorer09@gmail.com>
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
7 years agomodprobe-small: improve config help text
Denys Vlasenko [Wed, 1 Feb 2017 00:22:37 +0000 (01:22 +0100)]
modprobe-small: improve config help text

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
7 years agoash: improve / fix glob expansion
Felix Fietkau [Tue, 31 Jan 2017 20:58:55 +0000 (21:58 +0100)]
ash: improve / fix glob expansion

When using musl libc glob() a very long string can cause glob() to fail,
which leads to an out of memory error being raised by ash.

This can happen easily if a very long quoted string contains *, even
though no glob expansion should ever be performed on it (since it's
quoted).

Fix this by properly parsing control characters and escaping and only
accept unquoted metacharacters. While we're at it, unify this check for
libc and built-in glob expansion

Signed-off-by: Felix Fietkau <nbd@nbd.name>
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
7 years agotaskset: simplify code a bit; tweak --help
Denys Vlasenko [Mon, 30 Jan 2017 21:20:06 +0000 (22:20 +0100)]
taskset: simplify code a bit; tweak --help

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
7 years agomodprobe-small: fix --help texts, they are from "big" modutils
Denys Vlasenko [Mon, 30 Jan 2017 18:21:12 +0000 (19:21 +0100)]
modprobe-small: fix --help texts, they are from "big" modutils

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
7 years agomodutils: fix config options dependency
Kang-Che Sung [Mon, 30 Jan 2017 18:14:50 +0000 (19:14 +0100)]
modutils: fix config options dependency

module.aliases and module.symbols files have no use in modprobe-small
implementation. So FEATURE_MODUTILS_ALIAS and FEATURE_MODUTILS_SYMBOLS
will depend on !MODPROBE_SMALL.

The try_to_mmap_module() function is not called in modprobe-small.c,
so I will let FEATURE_INSMOD_TRY_MMAP depend on !MODPROBE_SMALL for
now.

Signed-off-by: Kang-Che Sung <explorer09@gmail.com>
Likewise, FEATURE_2_4_MODULES is not used by modprobe-small.

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
7 years agowget: add a big explanation what TLS code implements and what does not
Denys Vlasenko [Mon, 30 Jan 2017 15:27:37 +0000 (16:27 +0100)]
wget: add a big explanation what TLS code implements and what does not

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
7 years agolibbb: match_fstype() is unreadable in the extreme, fixing it
Denys Vlasenko [Sun, 29 Jan 2017 23:45:05 +0000 (00:45 +0100)]
libbb: match_fstype() is unreadable in the extreme, fixing it

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
7 years agovi,fsck: do not use build timestamp unconditionally. Closes 9626
Denys Vlasenko [Sun, 29 Jan 2017 18:14:26 +0000 (19:14 +0100)]
vi,fsck: do not use build timestamp unconditionally. Closes 9626

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
7 years agotaskset: separate "current" and "new" strings
Denys Vlasenko [Sun, 29 Jan 2017 17:59:38 +0000 (18:59 +0100)]
taskset: separate "current" and "new" strings

Better for constant sharing:

   text    data     bss     dec     hex filename
 912997     485    6856  920338   e0b12 busybox_old
 912988     485    6856  920329   e0b09 busybox_unstripped

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
7 years agotypo in comment
Denys Vlasenko [Sun, 29 Jan 2017 17:48:58 +0000 (18:48 +0100)]
typo in comment

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
7 years agotaskset: rewrite to be task size-agnostic
Denys Vlasenko [Sun, 29 Jan 2017 17:19:29 +0000 (18:19 +0100)]
taskset: rewrite to be task size-agnostic

function                                             old     new   delta
packed_usage                                       31130   31190     +60
taskset_main                                         623     525     -98
------------------------------------------------------------------------------
(add/remove: 0/0 grow/shrink: 1/1 up/down: 60/-98)            Total: -38 bytes

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
7 years ago*: add comment about APPLET_ODDNAME format
Denys Vlasenko [Sun, 29 Jan 2017 13:57:33 +0000 (14:57 +0100)]
*: add comment about APPLET_ODDNAME format

It confused me more than once

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
7 years agohttpd: defend against attempts to OOM us. Closes 9611
Denys Vlasenko [Thu, 26 Jan 2017 18:35:40 +0000 (19:35 +0100)]
httpd: defend against attempts to OOM us. Closes 9611

We were strdup'ing "Cookie: foo" every time we saw it.

function                                             old     new   delta
handle_incoming_and_exit                            2733    2821     +88

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
7 years agoadduser: help for -G GRP is misleading, fix it
Denys Vlasenko [Thu, 26 Jan 2017 17:54:12 +0000 (18:54 +0100)]
adduser: help for -G GRP is misleading, fix it

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
7 years agosh: fix FEATURE_SH_STANDALONE help text: it does not enable NOFORK
Denys Vlasenko [Thu, 26 Jan 2017 17:31:51 +0000 (18:31 +0100)]
sh: fix FEATURE_SH_STANDALONE help text: it does not enable NOFORK

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
7 years agolibbb: spawn_and_wait() fflushes before forking NOEXEC; child reinits logmode
Denys Vlasenko [Thu, 26 Jan 2017 00:13:58 +0000 (01:13 +0100)]
libbb: spawn_and_wait() fflushes before forking NOEXEC; child reinits logmode

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