Testcase:
21 01 01 00 00 00 00 00 e7 01 01 01 ef 00 df b6
00 17 02 10 11 0f ff 00 16 00 00
Unfortunately, the bug is not reliably causing a segfault,
the behavior depends on what's in memory before the buffer.
function old new delta
unpack_lzma_stream 2762 2768 +6
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
On certain corrupt gzip files, huft_build will set the error bit on
the result pointer. If afterwards abort_unzip is called huft_free
might run into a segmentation fault or an invalid pointer to
free(p).
In order to mitigate this, we check in huft_free if the error bit
is set and clear it before the linked list is freed.
Signed-off-by: Samuel Sapalski <samuel.sapalski@nokia.com>
Signed-off-by: Peter Kaestle <peter.kaestle@nokia.com>
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
Avoid using same include file naming pattern as available in GLIBC.
Renamed to match grp_.h et al.
Signed-off-by: Joachim Nilsson <troglobit@gmail.com>
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
busybox will remove directory symlinks, which is at
odds with common layouts that have some of
bin/lib/lib32/lib64 symlinked.
this adds a exludelist for critcal and often symlinked
directories.
Fixes: Bug 12551
function old new delta
remove_file_array 139 231 +92
Signed-off-by: Norbert Lange <nolange79@gmail.com>
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
The official Info-Zip unzip creates the dir if it doesn't exist.
Signed-off-by: Lauri Kasanen <cand@gmx.com>
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
huft_build() has way too many params
function old new delta
inflate_block 1293 1281 -12
huft_build 1085 1058 -27
------------------------------------------------------------------------------
(add/remove: 0/0 grow/shrink: 0/2 up/down: 0/-39) Total: -39 bytes
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
Converted a few 16-bit variables and small arrays to 32-bit.
Stopped pulling desc->FOO members into temporary local variables
in gen_bitlen(): on register-starved arches, this is a loss,
temporaries go into stack slots.
Sprinkled a few "const" on pointer arguments.
function old new delta
pack_gzip 742 745 +3
gen_codes 101 97 -4
build_tree 886 833 -53
------------------------------------------------------------------------------
(add/remove: 0/0 grow/shrink: 1/2 up/down: 3/-57) Total: -54 bytes
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
With this change, GNU gzip -n and BusyBox gzip now produce identical output
assuming that CONFIG_GZIP_FAST=2.
>> Excuse me, but I wonder one thing: Why should we follow
>> strictly with gzip on the no-options default behavior?
> First, the default 6 compression level is a de-facto standard. BSD gzip
> and Apple gzip (on macOS) use this default as well. So there is a
> reasonable expectation that different gzip implementations act the same.
> For instance, if the default for busybox gzip becomes 9, then someone
> writing a script using busybox gzip could reasonably expect that the
> compression level will still be 9 when the same script is run on another
> system. That would be wrong. Implementations should not deviate from
> de-facto standards without a strong reason.
>
> Second, the inherent reason for this default has not gone away. While
> processor speeds have exploded since the default was set, so has the
> typical size of compressed files. Multiple gigabytes are nothing unusual
> these days. And gzip is often used for compression on the fly, precisely
> because it offers a good compromise between speed and compression ratio.
> So I believe 6 continues to be a reasonable default.
function old new delta
deflate 939 927 -12
Signed-off-by: Daniel Edgecumbe <git@esotericnonsense.com>
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
With this change and CONFIG_GZIP_FAST=2, CONFIG_FEATURE_GZIP_LEVELS=y,
GNU gzip and BusyBox gzip now produce identical output at each compression
level (excluding 1..3, as BusyBox does not implement these levels).
Signed-off-by: Daniel Edgecumbe <git@esotericnonsense.com>
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
Fixes an off-by-one that actually resulted in level 7 being used
Signed-off-by: Daniel Edgecumbe <git@esotericnonsense.com>
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
Back in 2007, commit 0c97c9d437 ("'simple' error message functions by
Loic Grenie") introduced bb_simple_perror_msg() to allow for a lower
overhead call to bb_perror_msg() when only a string was being printed
with no parameters. This saves space for some CPU architectures because
it avoids the overhead of a call to a variadic function. However there
has never been a simple version of bb_error_msg(), and since 2007 many
new calls to bb_perror_msg() have been added that only take a single
parameter and so could have been using bb_simple_perror_message().
This changeset introduces 'simple' versions of bb_info_msg(),
bb_error_msg(), bb_error_msg_and_die(), bb_herror_msg() and
bb_herror_msg_and_die(), and replaces all calls that only take a
single parameter, or use something like ("%s", arg), with calls to the
corresponding 'simple' version.
Since it is likely that single parameter calls to the variadic functions
may be accidentally reintroduced in the future a new debugging config
option WARN_SIMPLE_MSG has been introduced. This uses some macro magic
which will cause any such calls to generate a warning, but this is
turned off by default to avoid use of the unpleasant macros in normal
circumstances.
This is a large changeset due to the number of calls that have been
replaced. The only files that contain changes other than simple
substitution of function calls are libbb.h, libbb/herror_msg.c,
libbb/verror_msg.c and libbb/xfuncs_printf.c. In miscutils/devfsd.c,
networking/udhcp/common.h and util-linux/mdev.c additonal macros have
been added for logging so that single parameter and multiple parameter
logging variants exist.
The amount of space saved varies considerably by architecture, and was
found to be as follows (for 'defconfig' using GCC 7.4):
Arm: -92 bytes
MIPS: -52 bytes
PPC: -1836 bytes
x86_64: -938 bytes
Note that for the MIPS architecture only an exception had to be made
disabling the 'simple' calls for 'udhcp' (in networking/udhcp/common.h)
because it made these files larger on MIPS.
Signed-off-by: James Byrne <james.byrne@origamienergy.com>
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
