Lasse Collin [Mon, 20 May 2024 13:55:00 +0000 (16:55 +0300)]
Tests: Make test_compress.sh more flexible
Add a new optional second argument: directory of the xz and xzdec
executables. This is need with the CMake build where the binaries
end up in the top-level build directory.
If ../config.h doesn't exist, assume that all encoders and decoders
are available. This will make this script usable from CMake in the
most common build configuration.
NOTE: Since the existence of ../config.h is checked, the working
directory of the test script must be a subdir in the build tree!
Otherwise ../config.h would look outside the build tree.
Use the default check type instead of forcing CRC32 or CRC64.
Now the script doesn't need to check if CRC64 is available.
Lasse Collin [Mon, 20 May 2024 13:55:00 +0000 (16:55 +0300)]
CMake: Prepare to support the test_*.sh tests
This is a bit hacky since the scripts grep config.h to know which
features were built but the CMake build doesn't create config.h.
So instead those test scripts will be run only when all relevant
features have been enabled.
Lasse Collin [Sun, 12 May 2024 19:26:30 +0000 (22:26 +0300)]
xz: Simplify the memory usage scaling code
This is closer to what it was before the --filtersX support was added,
just extended to support for scaling all filter chains. The method
before this commit was an extended version of the original too but
it was done in a more complex way for no clear reason. In case of
an error, the complex version printed fewer informative messages
(a good thing) but it's not a sigificant benefit.
In the limit is too low even for single-threaded mode, the required
amount of memory is now reported like in 5.4.x instead of like in
5.5.1alpha - 5.6.1 which showed the original non-scaled usage. It
had been a FIXME in the old code but it's not clear what message
makes the most sense.
Lasse Collin [Sun, 12 May 2024 14:09:17 +0000 (17:09 +0300)]
xz: Rename "filters" to "chains"
The convention is that
lzma_filter filters[LZMA_FILTERS_MAX + 1];
contains the filters of a single filter chain.
It was so here as well before the commit d6af7f347077b22403133239592e478931307759.
It changes "filters" to a ten-element array of filter chains.
It's clearer to call this array-of-arrays "chains".
This also renames "filter_idx" to "chain_idx" which is used
as an index as in chains[chain_idx].
Lasse Collin [Fri, 10 May 2024 17:33:08 +0000 (20:33 +0300)]
xz: Remove unused code and simplify
opt_mode == MODE_COMPRESS isn't possible when HAVE_ENCODERS isn't
defined. Thus, when *encoding*, the message about *decoder* memory
usage is possible to show only when both encoder and decoder have
been built.
Since the message is shown only at V_DEBUG, skip the memusage
calculation if verbosity level isn't high enough.
Lasse Collin [Tue, 7 May 2024 12:05:21 +0000 (15:05 +0300)]
po4a/update-po: Delete the *.po.authors files
These are temporary files that are needed only when running po4a.
The top-level Makefile.am puts the whole po4a directory into
distribution tarball (it's simpler) so deleting these temporary
files is needed to prevent them from getting into tarballs.
The Landlock sandbox in xzdec could be stricter as now it's
active only for the last file being decompressed. In xz,
read-only sandbox is used for multi-file case. On the other hand,
xz doesn't go to the strictest mode when processing the last file
when more than one file was specified; xzdec does.
Lasse Collin [Tue, 30 Apr 2024 19:22:45 +0000 (22:22 +0300)]
liblzma: Fix incorrect function type error from sanitizer
Clang 17 with -fsanitize=address,undefined:
src/liblzma/common/filter_common.c:366:8: runtime error:
call to function encoder_find through pointer to incorrect
function type 'const lzma_filter_coder *(*)(unsigned long)'
src/liblzma/common/filter_encoder.c:187: note:
encoder_find defined here
Use a wrapper function to get the correct type neatly.
This reduces the number of casts needed too.
This issue could be a problem with control flow integrity (CFI)
methods that check the function type on indirect function calls.
Lasse Collin [Fri, 19 Apr 2024 12:14:02 +0000 (15:14 +0300)]
Doxygen: Don't strip JavaScript anymore
The stripping method worked well with Doxygen 1.8 and 1.9 but
it doesn't work with Doxygen 1.10 anymore. Since we won't ship
pre-generated liblzma API docs anymore, the extra bloat and
extra license info of the JavaScript files won't affect the
upstream source package anymore.
Lasse Collin [Sat, 20 Apr 2024 14:09:11 +0000 (17:09 +0300)]
CMake: Keep the build working if the "tests" directory is missing
This moves the tests section as is from CMakeLists.txt into
tests/tests.cmake. CMakeLists.txt now includes tests/tests.cmake
if the latter file exists.
Now it's possible to delete the whole "tests" directory and
building with CMake will still work normally, just without
the tests. This way the tests are readily available for those
who want them, and those who won't run the tests anyway have
a straightforward way to ensure that nothing from the "tests"
directory can affect the build process.
Lasse Collin [Sat, 20 Apr 2024 10:12:50 +0000 (13:12 +0300)]
Tests: Remove x86 and SPARC BCJ tests
These are very old but the exact test file isn't easy to reproduce
as it was compiled from a short C program (bcj_test.c) long ago.
These tests weren't very good anyway, just a little better than nothing.
Lasse Collin [Sat, 27 Apr 2024 11:56:16 +0000 (14:56 +0300)]
Tests: test_index: Make it clear that my_alloc() has no integer overflows
liblzma guarantees that the product of the allocation size arguments
will fit in size_t.
Putting the pre-increment in the if-statement was clearly wrong
although in practice it didn't matter here as the function is
called only a couple of times.
Lasse Collin [Sat, 27 Apr 2024 11:33:38 +0000 (14:33 +0300)]
liblzma: index_decoder: Fix missing initializations on LZMA_PROG_ERROR
If the arguments to lzma_index_decoder() or lzma_index_buffer_decode()
were such that LZMA_PROG_ERROR was returned, the lzma_index **i
argument wasn't touched even though the API docs say that *i = NULL
is done if an error occurs. This obviously won't be done even now
if i == NULL but otherwise it is best to do it due to the wording
in the API docs.
In practice this matters very little: The problem can occur only
if the functions are called with invalid arguments, that is,
the calling application must already have a bug.
Lasse Collin [Thu, 25 Apr 2024 11:00:57 +0000 (14:00 +0300)]
tuklib_integer: Rename bswapXX to byteswapXX
The __builtin_bswapXX from GCC and Clang are preferred when
they are available. This can allow compilers to emit the x86 MOVBE
instruction instead of doing a load + byteswap as two instructions
(which would happen if the byteswapping is done in inline asm).
bswap16, bswap32, and bswap64 exist in system headers on *BSDs
and Darwin. #defining bswap16 on NetBSD results in a warning about
macro redefinition. It's safest to avoid this namespace conflict
completely.
No OS supported by tuklib_integer.h uses byteswapXX names and
a web search doesn't immediately find any obvious danger of
namespace conflicts. So let's try these still-pretty-short names
for the macros.
Thanks to Sam James for pointing out the compiler warning on
NetBSD 10.0.
Lasse Collin [Mon, 22 Apr 2024 18:54:39 +0000 (21:54 +0300)]
liblzma: lzma_str_to_filters: Set *error_pos on all errors
The API docs clearly say that if error_pos isn't NULL then *error
is always set on any error. However, it wasn't touched if str == NULL
or filters == NULL or unsupported flags were specified.