Denis Vlasenko
da31fbc1b1
gzip cleanup part #5
2007-01-07 19:39:02 +00:00
Denis Vlasenko
f824136f6b
gzip cleanup part #4
2007-01-07 19:38:42 +00:00
Denis Vlasenko
ed0f6db35e
gzip cleanup part #3
2007-01-07 19:38:26 +00:00
Denis Vlasenko
30551fd6da
gzip cleanup part #2
2007-01-07 19:38:06 +00:00
Denis Vlasenko
ad403413c7
a ton of gzip changes, split up in compiled and
...
run-tested pieces. Code was rather messy.
It's not a bug fix, more like code cleanup.
This is the first part.
2007-01-07 19:37:42 +00:00
Denis Vlasenko
da98165ad6
kill unused variable
2007-01-06 01:27:24 +00:00
Denis Vlasenko
c7a4aa5c99
move [g]zip decompressor state into malloc'ed buffer. size:
...
text data bss dec hex
5256 0 108 5364 14f4 - old
4915 0 0 4915 1333 - new
2007-01-06 00:03:11 +00:00
Denis Vlasenko
d1a19affeb
split inflate_xx_setup() subroutines from inflate_xx()
2007-01-05 23:58:45 +00:00
Denis Vlasenko
cd42cb8df0
do not expose internal state of [g]zip unpacker.
...
fix memory leak in inflate_gunzip.
2007-01-05 23:56:53 +00:00
Denis Vlasenko
447b543eaf
Stopped doing assignments inside expressions.
...
Who wrote this gem, I wonder?
n -= (e = (e = gunzip_wsize - ((d &= gunzip_wsize - 1) > w ? d : w)) > n ? n : e);
2007-01-05 19:49:02 +00:00
Denis Vlasenko
cc33ef12d2
decompress_unzip: preparatory patch
2007-01-05 19:46:04 +00:00
Denis Vlasenko
3376298b59
Stop tracking buffer size - it is a constant.
...
Stop ignoring write errors.
Fix bugs in this line:
rc->buffer_size = read(rc->fd, RC_BUFFER, rc->buffer_size);
(a) should use safe_read()
(b) just ONE short read (e.g. 4 bytes) will make ALL future reads short!
2007-01-05 14:04:47 +00:00
Denis Vlasenko
98b8e9487d
simplify access to buffer, making code a bit smaller
2007-01-05 13:59:05 +00:00
Bernhard Reutner-Fischer
bfe1950c63
- spelling fix
2007-01-04 15:44:49 +00:00
Denis Vlasenko
bb3d0fab3b
extern variable declaration in a .c file is heresy - fixing it
2007-01-03 01:57:25 +00:00
Denis Vlasenko
3bba545a54
done a dozen of randconfig test. guess what? ALL failed...
...
these are resulting fixes
2006-12-30 17:57:03 +00:00
Denis Vlasenko
9a44c4f91c
bb_xget[pw/gr]nam were horribly misnamed - fixed.
...
uidgid_get -> get_uidgid, add additional param
(numeric_ok). Make chown use it.
chown: fix "chown user: ...."
install: fix incorrect use of bb_xget[pw/gr]nam
2006-12-28 05:44:47 +00:00
Denis Vlasenko
666da5e2c6
merge post-1.3.0 fixes
2006-12-26 18:17:42 +00:00
Denis Vlasenko
bf0a201008
style fixes
...
last xcalloc replaced by xzalloc
2006-12-26 10:42:51 +00:00
Denis Vlasenko
4cccc03768
remove useless casts (type*) xzalloc(...)
2006-12-22 18:37:07 +00:00
Denis Vlasenko
a6df5907d2
dpkg: getopt32-ification etc, -100 bytes
2006-12-22 18:32:40 +00:00
Denis Vlasenko
f4c52b32db
rpm: reformat. was really hard to read
2006-12-22 15:03:50 +00:00
Denis Vlasenko
afc9ff99fd
removal of commented-out cruft
2006-12-22 00:45:27 +00:00
Denis Vlasenko
714701c890
tar et al: die if bb_copyfd_size copies less than asked for.
...
(we have bb_copyfd_exact_size now for that kind of usage)
2006-12-22 00:21:07 +00:00
Denis Vlasenko
b95636c52f
remove casts from xmalloc()
2006-12-19 23:36:04 +00:00
Denis Vlasenko
83e5d6f772
A bunch of defined(__GLIBC__) added. static-linking warning expanded
2006-12-18 21:49:06 +00:00
Denis Vlasenko
c88894602d
tar: report error if child dies while writing out the end of tarball
...
(e.g. out of disk space).
2006-12-17 19:08:20 +00:00
Denis Vlasenko
9f739445cd
inline strcmp(s, "-") [actually macro-ize it for now - gcc is too stupid]
2006-12-16 23:49:13 +00:00
Denis Vlasenko
a597aaddfa
s/extern inline/static ATTRIBUTE_ALWAYS_INLINE/g
...
xstrtou: disallow leading '+'
2006-12-16 23:48:13 +00:00
Bernhard Reutner-Fischer
6f0540e7ec
- remove functions marked as LEGACY in SUSv3 and use their modern counterparts.
2006-12-12 11:50:44 +00:00
Denis Vlasenko
c61852a02b
fix support for globally disabling --long-options.
...
(disabling them saves ~4K on fully configured bbox)
2006-11-29 11:09:43 +00:00
Denis Vlasenko
30d7a346e6
tar: fix misplaced --exclude long option definition
2006-11-27 20:24:40 +00:00
Denis Vlasenko
d9e15f2068
style cleanup: return(a) -> return a, part 2
2006-11-27 16:49:55 +00:00
Denis Vlasenko
079f8afa0a
style cleanup: return(a) -> return a, part 1
2006-11-27 16:49:31 +00:00
Denis Vlasenko
cf749bc10c
small fixes:
...
fix xstrdup to not grossly overallocate memory
use xopen instean of xopen3 in several places
etc.
2006-11-26 15:45:17 +00:00
Denis Vlasenko
c1660fea6d
tar: refuse to untar files with "/../" components
2006-11-26 15:42:03 +00:00
Denis Vlasenko
b596335868
tar: correctly skip (and warn about) pax headers.
...
plug memory leak. save 50 bytes.
Wooohooo! we finally can unpack kernel tarballs!
2006-11-26 01:46:59 +00:00
Denis Vlasenko
7725d0ff5c
tar: add support for FEATURE_TAR_GNU_EXTENSIONS so than we can save
...
long names now. We were able to read such tars, but not create.
+275 bytes. Without FEATURE_TAR_GNU_EXTENSIONS: -25 bytes.
We still cannot unpack Linux kernels, but not for long ;)
2006-11-25 23:56:50 +00:00
Denis Vlasenko
87cd4a87e3
tar: small fix and small optimization
2006-11-25 23:47:32 +00:00
Denis Vlasenko
f2408e6c3f
tar: abort if tarring up file larger that 64Gb
...
(otherwise we will produce garbled tarfile)
2006-11-25 14:48:09 +00:00
Denis Vlasenko
3feb2fc535
tar: fix multiple -t and/or -v options handling.
...
do not process list of files to tar up in reverse order.
2006-11-24 21:55:55 +00:00
Denis Vlasenko
d031ffa623
tar: sanitize option handling
2006-11-24 21:54:44 +00:00
Denis Vlasenko
b833ca9d2d
tar: buglet fix
2006-11-24 18:53:13 +00:00
Denis Vlasenko
d6772501db
tar: fix and sanitize handling of long filenames/linknames
...
(GNU extensions 'K' and 'L').
We correctly handle them when untarring now, but unfortunately
we still don't use them when tarring! That stupid 100 char limit
is still there!
The biggest problem is that we don't support 'pax' tar format.
Linux kernel tarballs are in this format... shame
2006-11-24 17:21:44 +00:00
Denis Vlasenko
b8a8e601df
tar: small fixes:
...
* size-optimize mapping code
* kill double close
2006-11-24 14:59:45 +00:00
Denis Vlasenko
c50f370f98
tar: cache [ug]id->username/groupname mappings. Cuts down amount
...
of open/read/close of /etc/passwd and /etc/group
dramatically (we were rereading those for each untarred file!!!)
2006-11-24 14:57:31 +00:00
Denis Vlasenko
4fbb584a0e
tar: cry murder and bail out if file shrinks under us while we tar it up
2006-11-24 14:55:23 +00:00
Denis Vlasenko
cf30cc82a3
header_verbose_list: stop truncating file size in listing
2006-11-24 14:53:18 +00:00
Denis Vlasenko
376ce1e775
tar:
...
* unpack: handle tar header fields which are not NUL terminated
* pack: handle 4+GB files correctly
* pack: refuse to store 101+ softlinks (was truncating link
target name)
* pack: mask mode with 07777
2006-11-24 14:51:01 +00:00
Denis Vlasenko
9229794ab3
insmod_ng_main: -80 bytes. Stopp mmapping, use xmalloc_open_read_close().
2006-11-21 11:58:14 +00:00
Denis Vlasenko
3eb91c2e35
cpio: (allegedly) fix bug 0001095:
...
"cpio hardlink support and possible memory leak fix"
2006-11-21 00:55:46 +00:00
Denis Vlasenko
5af906e7c8
rename: compare_string_array -> index_in_str_array
...
introduce index_in_substr_array and use it in
iproute2
2006-11-05 18:05:09 +00:00
Denis Vlasenko
8c35d65c43
recursive_action: add depth param
...
chmod: match coreutils versus following links
2006-10-27 23:42:25 +00:00
Denis Vlasenko
d3d004dd35
last nail into error_msg() (de)capitalization
2006-10-27 09:02:31 +00:00
Denis Vlasenko
ddec5af6b0
rename functions to more understandable names
2006-10-26 23:25:17 +00:00
Denis Vlasenko
e1a0d486e4
message string changes, mostly for consistency, also -32 bytes in .rodata
2006-10-20 13:28:22 +00:00
Denis Vlasenko
ea62077b85
add open_read_close() and similar stuff
2006-10-14 02:23:43 +00:00
Denis Vlasenko
2d5ca60bfb
bb_get_[chomped]line_from_file wasn't descriptive enough.
...
Renaming...
2006-10-12 22:43:20 +00:00
Denis Vlasenko
9c267b851e
style fixes
2006-10-12 20:06:18 +00:00
Denis Vlasenko
a6dbb08a48
small style fixes
2006-10-12 19:29:44 +00:00
Denis Vlasenko
1114de7219
od: provide full-blown od from coreutils if CONFIG_DESKTOP
2006-10-10 23:26:05 +00:00
Denis Vlasenko
cba9ef5523
fixes from Vladimir Dronnikov <dronnikov@gmail.ru>
2006-10-10 21:00:47 +00:00
Denis Vlasenko
7039a66b58
correct largefile support, add comments about it.
2006-10-08 17:54:47 +00:00
Denis Vlasenko
1385899416
attempt to regularize atoi mess.
2006-10-08 12:49:22 +00:00
Denis Vlasenko
7d219aab70
build system overhaul
2006-10-05 10:17:08 +00:00
Denis Vlasenko
8f8f268cfd
bb_applet_name -> applet_name
2006-10-03 21:00:43 +00:00
Denis Vlasenko
67b23e6043
getopt_ulflags -> getopt32.
...
It is impossible to formulate sane ABI based on
size of ulong because it can be 32-bit or 64-bit.
Basically it means that you cannot portably use
more that 32 option chars in one call anyway...
Make it explicit.
2006-10-03 21:00:06 +00:00
Denis Vlasenko
cf94446af7
random style fixes
2006-10-03 19:02:20 +00:00
Denis Vlasenko
bb119d059a
gzip: add a bit of comment. The code itself
...
is too bloated to work on it...
2006-10-01 16:44:04 +00:00
Denis Vlasenko
97a8dd3857
g[un]zip: add support for -v (verbose).
...
Add CONFIG_DESKTOP, almost all bloat from this change
is hidden under that.
2006-10-01 15:55:11 +00:00
Denis Vlasenko
aecabffb8b
dpkg: fix use-after-free (bug 939)
2006-09-30 21:05:25 +00:00
Denis Vlasenko
3038ac9c19
vconfig: fix bug 701 (I be damned if I understand it)
2006-09-30 19:37:25 +00:00
Denis Vlasenko
a959588b80
Yet another silly little byte saving. couldn't -> cannot
2006-09-29 21:30:43 +00:00
Denis Vlasenko
5492884311
dpkg: style fixes; short -> int (generated code is smaller)
2006-09-28 22:35:42 +00:00
Denis Vlasenko
57308afb5b
dpkg: reduce bss usage by ~130 kbytes (yes, kilobytes!)
...
at the cost of ~100 bytes of text.
Improves friendliness to nommu systems.
(Dunno whether nommu people ever use dpkg, though...)
2006-09-28 22:34:46 +00:00
Denis Vlasenko
ce97960a6e
usage.h: remove ugly tricks. We have USE_FEATURE_xxx now.
...
Few bugs are now exposed...
2006-09-27 23:31:08 +00:00
Denis Vlasenko
c16bd212e3
silly switch style fix
2006-09-27 19:51:06 +00:00
Denis Vlasenko
6429aabbf1
bb_askpass: shorten static password buffer. 256 is way too large.
...
simplify code a bit.
2006-09-23 12:22:11 +00:00
Denis Vlasenko
01e88f0339
gzip: getopt_ulflags'isation
2006-09-22 15:13:38 +00:00
Denis Vlasenko
9213a9e0f2
whitespace cleanup
2006-09-17 16:28:10 +00:00
Rob Landley
20cc6d567f
Remove pointless "const". Bloatcheck says 0 bytes difference.
2006-09-12 21:42:17 +00:00
Bernhard Reutner-Fischer
64d7e93081
- convert a few xopen3(,,0) into xopen(,). Also peruse the fact that xopen defaults to 0777 on it's own, so we don't need to xopen3(,,0777).
...
Saves a few bytes:
$ size busybox.old busybox text data bss dec hex filename
839676 8780 243592 1092048 10a9d0 busybox.old
839660 8780 243592 1092032 10a9c0 busybox
$ make CC=gcc-4.2-HEAD bloatcheck
function old new delta
static.vconfig_main 281 279 -2
static.rx_main 1076 1074 -2
static.mt_main 294 292 -2
static.gzip_main 753 751 -2
static.mkfs_minix_main 4034 4030 -4
static.unzip_main 1771 1766 -5
static.crontab_main 1513 1507 -6
------------------------------------------------------------------------------
(add/remove: 0/0 grow/shrink: 0/7 up/down: 0/-23) Total: -23 bytes
2006-09-11 16:01:40 +00:00
Rob Landley
1bfca7bac7
Cleanup headers and replace the zero padding with an even smaller version.
...
One of the pads turned out to be unnecessary: sizeof(struct TarHeader) is
TAR_BLOCK_SIZE, the padding's in the struct. The others could be done inline
with bb_common_bufsiz1.
This is a cleanup I did to Denis' patch long ago, but got sidetracked by
what turned into svn 15660.
2006-09-10 03:20:37 +00:00
Denis Vlasenko
0c45bb23d2
tar: fix "xopen with O_CREAT" warning, improve zero padding write
...
(was doing zillions of 1-byte write syscalls)
2006-09-09 12:49:03 +00:00
Denis Vlasenko
9cac521f07
using [xa]sprintf for string concatenation is neat and saves
...
~100 bytes according to bloatcheck. Also this fixes bug in rpm
2006-09-09 12:24:19 +00:00
Denis Vlasenko
6d655be5df
removed a lot of trailing \n in bb_msg() calls. It is added
...
automatically by function itself.
2006-09-06 19:02:46 +00:00
Bernhard Reutner-Fischer
5d725462d4
- strip 399424 off the bss by making decompress_uncompress buffers config buffers.
...
Compile tested (too lazy to look for a small .Z on the net).
$ size busybox.old busybox
text data bss dec hex filename
859555 10232 645732 1515519 171fff busybox.old
859683 10232 246308 1116223 11083f busybox
$ make bloatcheck
function old new delta
uncompress 1036 1160 +124
inbuf 2116 4 -2112
outbuf 4100 4 -4096
htab 131072 - -131072
codetab 262144 - -262144
------------------------------------------------------------------------------
(add/remove: 0/2 grow/shrink: 1/2 up/down: 124/-399424) Total: -399300 bytes
2006-09-06 15:28:32 +00:00
Rob Landley
02496aa69e
The bb_ prefixes were removed from xasprintf() and xopen() a month ago (in
...
svn 15767).
2006-09-05 13:48:21 +00:00
Denis Vlasenko
1da6a21666
dpkg: fix buffer overflow (bug 983)
2006-09-03 16:33:58 +00:00
Denis Vlasenko
22dca23d52
archival: added O_TRUNC so that when we overwrite files on unpack,
...
we truncate them. Also spotted & fixed hard to trigger bug
with extension handling.
2006-09-03 14:23:29 +00:00
Denis Vlasenko
314908da9e
tar: revert older fix (non-portable), added new one.
...
testsuite tar-extracts-all-subdirs now passes.
2006-09-03 14:04:33 +00:00
Denis Vlasenko
06ab5fb6b9
tar: tar xf foo.tar dir/dir did not extract all subdirs.
...
Added testsuite entry for this
2006-09-02 18:40:10 +00:00
Rob Landley
88621d7398
Svn 16007 broke the build under gcc 4.0.3. This fixes up some of the damage
...
(the e2fsprogs directory is too twisty and evil to easily fix, but I plan
to rewrite it anyway so I'll just bump that up in priority a bit).
2006-08-29 19:41:06 +00:00
Bernhard Reutner-Fischer
d591a360d3
- merge -r15463:15564 from busybox_scratch branch through these changesets:
...
------------------------------------------------------------------------
r15465 | aldot | 2006-06-21 20:48:06 +0200 (Wed, 21 Jun 2006) | 3 lines
- use CONFIG_BUSYBOX_EXEC_PATH as before it one was broken by a recent revert.
- use xchdir() since all is invain if it fails there anyways, supposedly
------------------------------------------------------------------------
r15466 | aldot | 2006-06-21 20:55:16 +0200 (Wed, 21 Jun 2006) | 2 lines
- adjust docs to take CONFIG_BUSYBOX_EXEC_PATH into account.
------------------------------------------------------------------------
r15467 | aldot | 2006-06-21 21:31:24 +0200 (Wed, 21 Jun 2006) | 18 lines
- partial fallout of my TREE_USED touchup against gcc-4.2: rip unused vars, save
s 144 bytes
text data bss dec hex filename
862434 10156 645924 1518514 172bb2 busybox.old
862322 10156 645892 1518370 172b22 busybox
function old new delta
z_len 4 - -4
textend 4 - -4
part_nb 4 - -4
insize 4 - -4
ifile_size 4 - -4
do_link 4 - -4
new_text 70 60 -10
ipaddr_list_link 33 23 -10
gzip_main 898 822 -76
------------------------------------------------------------------------------
(add/remove: 0/6 grow/shrink: 0/3 up/down: 0/-120) Total: -120 bytes
------------------------------------------------------------------------
r15468 | aldot | 2006-06-21 21:43:05 +0200 (Wed, 21 Jun 2006) | 19 lines
- remove useless global exports
function old new delta
rpm_main 940 1601 +661
rpm_getstring 107 112 +5
rpm_getint 148 153 +5
loop_through_files 103 106 +3
fileaction_dobackup 115 113 -2
fileaction_list 5 - -5
rpm_getcount 42 - -42
extract_cpio_gz 161 - -161
rpm_gettags 504 - -504
------------------------------------------------------------------------------
(add/remove: 0/4 grow/shrink: 4/1 up/down: 674/-714) Total: -40 bytes
text data bss dec hex filename
862322 10156 645892 1518370 172b22 busybox.old
862290 10156 645892 1518338 172b02 busybox
------------------------------------------------------------------------
r15555 | aldot | 2006-06-30 14:10:11 +0200 (Fri, 30 Jun 2006) | 22 lines
- shrink syslog a little bit, move a big buffer (for 'line') off the bss, fold s
emaphore stuff into single caller manually.
stats:
function old new delta
logMessage 395 427 +32
message 1245 1257 +12
opts - 4 +4
small 1 - -1
local_logging 4 - -4
doRemoteLog 4 - -4
circular_logging 4 - -4
syslogd_main 1299 1285 -14
static.res 36 16 -20
.rodata 186650 186586 -64
static.line 1025 - -1025
------------------------------------------------------------------------------
(add/remove: 1/5 grow/shrink: 2/3 up/down: 48/-1136) Total: -1088 bytes
cow@s37:~/src/busybox_scratch$ size sysklogd/syslogd.o{.orig,}
text data bss dec hex filename
3723 348 5242 9313 2461 sysklogd/syslogd.o.orig
3697 348 4188 8233 2029 sysklogd/syslogd.o
==============================================================================
Overall bloatcheck for the changeset mentioned above:
function old new delta
rpm_main 953 1608 +655
logMessage 395 427 +32
message 1245 1257 +12
opts - 4 +4
rpm_getstring 107 110 +3
rpm_getint 148 151 +3
loop_through_files 103 104 +1
small 1 - -1
fileaction_dobackup 115 113 -2
z_len 4 - -4
textend 4 - -4
part_nb 4 - -4
local_logging 4 - -4
insize 4 - -4
ifile_size 4 - -4
do_link 4 - -4
doRemoteLog 4 - -4
circular_logging 4 - -4
fileaction_list 5 - -5
new_text 70 60 -10
ipaddr_list_link 33 23 -10
clear_bufs 31 21 -10
syslogd_main 1287 1273 -14
builtin_help 190 176 -14
static.res 36 16 -20
builtin_source 229 199 -30
rpm_getcount 42 - -42
gzip_main 842 786 -56
.rodata 227176 227112 -64
lash_main 609 527 -82
busy_loop 3883 3739 -144
extract_cpio_gz 155 - -155
rpm_gettags 501 - -501
static.line 1025 - -1025
------------------------------------------------------------------------------
(add/remove: 1/15 grow/shrink: 6/12 up/down: 710/-2221) Total: -1511 bytes
2006-08-20 17:35:13 +00:00
Rob Landley
f4e50b0bcd
We haven't got a CONFIG_APT_GET.
2006-08-04 21:12:14 +00:00
Rob Landley
74bb70cf54
Ok, that's not a typo, that's just me not thinking.
2006-08-04 20:19:01 +00:00
Rob Landley
3629d95a24
Typo.
2006-08-04 20:15:18 +00:00
Rob Landley
27d07c9ce2
Now that we have xopen3(), it's just plain unclean to have xopen() with
...
O_CREAT. Two users are still doing it (and thus getting permissions 777),
I'm not sure what permissions they should be using but here they are
changed to xopen3(). This costs us a dozen or so bytes, but removing the
777 from xopen() should get some of that back.
2006-08-04 19:16:31 +00:00
Rob Landley
daf58ef61a
Use xopen3() instead of an after-the-fact chown.
2006-08-04 17:26:58 +00:00
Rob Landley
86b4d64aa3
These should have been part of 15767 too.
2006-08-03 17:58:17 +00:00