Adolf Belka [Mon, 17 May 2021 12:29:44 +0000 (14:29 +0200)]
elfutils: Update to 0.184
- Update from 0.183 to 0.184
- Update rootfiles
- Changelog
2021-05-10 Mark Wielaard <mark@klomp.org>
* configure.ac (AC_INIT): Set version to 0.184.
* NEWS: Add libdw, translation and debuginfod-client entries.
2021-03-30 Frank Ch. Eigler <fche@redhat.com>
* configure.ac: Look for pthread_setname_np.
2021-02-17 Timm Bäder <tbaeder@redhat.com>
* configure.ac: Add -Wno-packed-not-aligned check.
2021-02-17 Timm Bäder <tbaeder@redhat.com>
* configure.ac: Add -Wtrampolines check.
Signed-off-by: Adolf Belka <adolf.belka@ipfire.org> Reviewed-by: Peter Müller <peter.mueller@ipfire.org> Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Adolf Belka [Mon, 17 May 2021 12:30:32 +0000 (14:30 +0200)]
gdb: Update to 10.2
- Update from 10.1 to 10.2
- Update rootfiles
- Changelog
GDB 10.2 brings the following fixes and enhancements over GDB 10.1:
* PR remote/26614 (AddressSanitizer: heap-use-after-free of extended_remote_target in remote_async_inferior_event_handler)
* PR gdb/26828 (SIGSEGV in follow_die_offset dwarf2/read.c:22950)
* PR gdb/26861 (internal-error: void target_mourn_inferior(ptid_t): Assertion `ptid == inferior_ptid' failed. OS: Mac OSX Catalina; Compiler: GCC; Language: C)
* PR gdb/26876 (gdb error: internal-error: Unknown CFA rule when debugging the linux kernel with qemu)
* PR breakpoints/26881 (infrun.c:6384: internal-error: void process_event_stop_test(execution_control_state*): Assertion `ecs->event_thread->control.exception_resume_breakpoint != NULL' failed)
* PR gdb/26901 (Array subscript fails with flexible array member without size)
* PR tui/26973 (gdb crashes when not including the status window in a new layout)
* PR python/26974 (Wrong Value.format_string docu for static members argument)
* PR breakpoints/27009 ([s390] GDB branches randomly for BC instruction while displaced stepping)
* PR tdep/27015 (ARC: "eret" value is collected from the wrong data in register cache)
* PR backtrace/27147 ([GNU/Linux, sparc64] GDB is unable to print full stack trace (got "previous frame inner to this frame" errors))
* PR rust/27194 (put rust demangler on 10.x branch)
* PR threads/27239 (gdb/cp-support.c:1619:(.text+0x5502): relocation truncated to fit: R_X86_64_PC32 against undefined symbol `TLS init function for thread_local_segv_handler')
* PR breakpoints/27330 (nextoverthrow.exp FAILs on arm-none-eabi)
* PR symtab/27333 ([dwarf-5] abort on unhandled DW_TAG_type_unit in process_psymtab_comp_unit)
* PR fortran/27341 ([dwarf-5] FAIL: gdb.fortran/function-calls.exp: p derived_types_and_module_calls::pass_cart_nd(c_nd))
* PR tdep/27369 (ARC: Stepping over atomic instruction sequences loops infinitely)
* PR build/27385 (Cannot compile arc.c with gcc-4.8 (error: no matching function for call to 'std::pair...'))
* PR gdb/27435 (Attach on solaris segfaults GDB)
* PR build/27535 (amd64-linux-siginfo.c fails to compile after updating to glibc-2.33 headers)
* PR build/27536 (aarch64-linux-hw-point.c fails to compile after updating to glibc-2.33)
* PR symtab/27541 (gdb crashes on "file -readnow")
* PR gdb/27750 (local variables have wrong address and values on sparc64)
* PR varobj/27757 (-var-list-children coredump)
Signed-off-by: Adolf Belka <adolf.belka@ipfire.org> Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Peter Müller [Mon, 17 May 2021 19:02:36 +0000 (21:02 +0200)]
DMA: do not ship a binary for creating mail boxes
This is only needed in case of bounces generated by locally emitted
messages. We neither store these, nor do we create mail boxes on a
firewall. Safe to drop.
Cc: Michael Tremer <michael.tremer@ipfire.org> Signed-off-by: Peter Müller <peter.mueller@ipfire.org> Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Peter Müller [Mon, 17 May 2021 19:01:54 +0000 (21:01 +0200)]
/usr/bin/ping does not need a SUID bit if appropriate capabilities are set
Cc: Michael Tremer <michael.tremer@ipfire.org> Signed-off-by: Peter Müller <peter.mueller@ipfire.org> Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Peter Müller [Sun, 16 May 2021 20:48:58 +0000 (22:48 +0200)]
OpenSSH: do not ship ssh-keysign anymore
To my surprise, this binary comes with suid flag set, and since we do
not have SSH key signing enabled, there is no need to ship it with
IPFire.
Signed-off-by: Peter Müller <peter.mueller@ipfire.org> Reviewed-by: Adolf Belka <adolf.belka@ipfire.org> Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Adolf Belka [Sat, 15 May 2021 21:50:34 +0000 (23:50 +0200)]
python-distutils-extra: Removal of this python2 module
- python-distutils-extra is linked to python-distutils which is no longer
used as it has been replaced by setuptools.
- python-distutils-extra is currently from 2011 and the latest version
is from 2016. No development occurring on this.
- No problem on a clean build with this module being removed.
Signed-off-by: Adolf Belka <adolf.belka@ipfire.org> Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Adolf Belka [Sat, 15 May 2021 21:50:33 +0000 (23:50 +0200)]
python-distutils: Removal of this python2 module
- python-distutils has been replaced by setuptools.
- python-distutils was not being built anyway as it was not listed in
make.sh
- lfs has missing sections. There are no source and no build sections
Signed-off-by: Adolf Belka <adolf.belka@ipfire.org> Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Adolf Belka [Sat, 15 May 2021 21:50:32 +0000 (23:50 +0200)]
python-optional-src: Removal of this python2 module
- python-optional-src was not getting built anyway as it was not listed
in make.sh
- lfs file was missing most of the standard content. No source info
and no build instructions
- missing source file from IPFire source system
- grep on build/ found no dependencies on this module
Signed-off-by: Adolf Belka <adolf.belka@ipfire.org> Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Adolf Belka [Sat, 15 May 2021 21:50:31 +0000 (23:50 +0200)]
make.sh: Removal of three python2 modules
- Removal of python-distutils and python-distutils-extra as these have
been replaced by setuptools.
- Removal of python-optional-src
- Only python-distutils-extra line is removed from make.sh as
python-distutils and python-optional-src were not in make.sh
These two modules have not been getting built historically
Signed-off-by: Adolf Belka <adolf.belka@ipfire.org> Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Adolf Belka [Sat, 15 May 2021 21:50:03 +0000 (23:50 +0200)]
nmap: Migrate to python3
- Added PYTHON=python3 prior to configure. This then builds nmap with
python3.
- ndiff is written as python2 only and currently no patches to make it
work wih python3 have been accepted by the nmap team. It looks like ndiff
will stay as it is for some time so ndiff will be removed from the nmap
package install.
- Added --without-ndiff to configure so nmap is built without ndiff
- Updated rootfile
Signed-off-by: Adolf Belka <adolf.belka@ipfire.org> Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Peter Müller [Fri, 14 May 2021 21:11:49 +0000 (23:11 +0200)]
Tor: update to 0.4.5.8
Full changelog as per https://gitweb.torproject.org/tor.git/plain/ChangeLog?h=tor-0.4.5.8:
Changes in version 0.4.5.8 - 2021-05-10
Tor 0.4.5.8 fixes several bugs in earlier version, backporting fixes
from the 0.4.6.x series.
o Minor features (compatibility, Linux seccomp sandbox, backport from 0.4.6.3-rc):
- Add a workaround to enable the Linux sandbox to work correctly
with Glibc 2.33. This version of Glibc has started using the
fstatat() system call, which previously our sandbox did not allow.
Closes ticket 40382; see the ticket for a discussion of trade-offs.
o Minor features (compilation, backport from 0.4.6.3-rc):
- Make the autoconf script build correctly with autoconf versions
2.70 and later. Closes part of ticket 40335.
o Minor features (fallback directory list, backport from 0.4.6.2-alpha):
- Regenerate the list of fallback directories to contain a new set
of 200 relays. Closes ticket 40265.
o Minor features (geoip data):
- Update the geoip files to match the IPFire Location Database, as
retrieved on 2021/05/07.
o Minor features (onion services):
- Add warning message when connecting to now deprecated v2 onion
services. As announced, Tor 0.4.5.x is the last series that will
support v2 onions. Closes ticket 40373.
o Minor bugfixes (bridge, pluggable transport, backport from 0.4.6.2-alpha):
- Fix a regression that made it impossible start Tor using a bridge
line with a transport name and no fingerprint. Fixes bug 40360;
bugfix on 0.4.5.4-rc.
o Minor bugfixes (build, cross-compilation, backport from 0.4.6.3-rc):
- Allow a custom "ar" for cross-compilation. Our previous build
script had used the $AR environment variable in most places, but
it missed one. Fixes bug 40369; bugfix on 0.4.5.1-alpha.
o Minor bugfixes (channel, DoS, backport from 0.4.6.2-alpha):
- Fix a non-fatal BUG() message due to a too-early free of a string,
when listing a client connection from the DoS defenses subsystem.
Fixes bug 40345; bugfix on 0.4.3.4-rc.
o Minor bugfixes (compiler warnings, backport from 0.4.6.3-rc):
- Fix an indentation problem that led to a warning from GCC 11.1.1.
Fixes bug 40380; bugfix on 0.3.0.1-alpha.
o Minor bugfixes (controller, backport from 0.4.6.1-alpha):
- Fix a "BUG" warning that would appear when a controller chooses
the first hop for a circuit, and that circuit completes. Fixes bug
40285; bugfix on 0.3.2.1-alpha.
o Minor bugfixes (onion service, client, memory leak, backport from 0.4.6.3-rc):
- Fix a bug where an expired cached descriptor could get overwritten
with a new one without freeing it, leading to a memory leak. Fixes
bug 40356; bugfix on 0.3.5.1-alpha.
o Minor bugfixes (testing, BSD, backport from 0.4.6.2-alpha):
- Fix pattern-matching errors when patterns expand to invalid paths
on BSD systems. Fixes bug 40318; bugfix on 0.4.5.1-alpha. Patch by
Daniel Pinto.
Signed-off-by: Peter Müller <peter.mueller@ipfire.org> Reviewed-by: Adolf Belka <adolf.belka@ipfire.org> Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Adolf Belka [Sat, 15 May 2021 21:49:32 +0000 (23:49 +0200)]
sudo: Update to 1.9.7
- Update from 1.9.6p1 to 1.9.7
- Update of rootfile not required
- Changelog
The fuzz Makefile target now runs all the fuzzers for 8192 passes (can be overridden via the FUZZ_RUNS variable). This makes it easier to run the fuzzers in-tree. To run a fuzzer indefinitely, set FUZZ_RUNS=-1, e.g. make FUZZ_RUNS=-1 fuzz.
Fixed fuzzing on FreeBSD where the ld.lld linker returns an error by default when a symbol is multiply-defined.
Added support for determining local IPv6 addresses on systems that lack the getifaddrs() function. This now works on AIX, HP-UX and Solaris (at least). Bug #969.
Fixed a bug introduced in sudo 1.9.6 that caused sudo -V to report a usage error. Also, when invoked as sudoedit, sudo now allows a more restricted set of options that matches the usage statement and documentation. GitHub Issue #95.
Fixed a crash in sudo_sendlog when the specified certificate or key does not exist or is invalid. Bug #970.
Fixed a compilation error when sudo is configured with the disable-log-clientoption.
Sudo's limited support for SUCCESS=return entries in nsswitch.conf is now documented. Bug #971.
Sudo now requires autoconf 2.70 or higher to regenerate the configure script. Bug #972.
sudo_logsrvd now has a relay mode which can be used to create a hierarchy of log servers. By default, when a relay server is defined, messages from the client are forwarded immediately to the relay. However, if the store_first setting is enabled, the log will be stored locally until the command completes and then relayed. Bug #965.
Sudo now links with OpenSSL by default if it is available unless the --disable-openssl configure option is used or both the --disable-log-client and --disable-log-server configure options are specified.
Fixed configure's Python version detection when the version minor number is more than a single digit, for example Python 3.10.
The sudo Python module tests now pass for Python 3.10.
Sudo will now avoid changing the datasize resource limit as long as the existing value is at least 1GB. This works around a problem on 64-bit HP-UX where it is not possible to exactly restore the original datasize limit. Bug #973.
Fixed a race condition that could result in a hang when sudo is executed by a process where the SIGCHLD handler is set to SIG_IGN. This fixes the bug described by GitHub PR #98.
Fixed an out-of-bounds read in sudoedit and visudo when the EDITOR, VISUAL or SUDO_EDITOR environment variables end in an unescaped backslash. Also fixed the handling of quote characters that are escaped by a backslash. GitHub Issue #99.
Fixed a bug that prevented the log_server_verify sudoers option from taking effect.
The sudo_sendlog utility has a new -s option to cause it to stop sending I/O records after a user-specified elapsed time. This can be used to test the I/O log restart functionality of sudo_logsrvd.
Fixed a crash introduced in sudo 1.9.4 in sudo_logsrvd when attempting to restart an interrupted I/O log transfer.
The TLS connection timeout in the sudoers log client was previously hard-coded to 10 seconds. It now uses the value of log_server_timeout.
The configure script now outputs a summary of the user-configurable options at the end, separate from output of configure script tests. Bug #820.
Corrected the description of which groups may be specified via the -g option in the Runas_Spec section. Bug #975.
Signed-off-by: Adolf Belka <adolf.belka@ipfire.org> Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Signed-off-by: Arne Fitzenreiter <arne_f@ipfire.org> Reviewed-by: Peter Müller <peter.mueller@ipfire.org> Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Adolf Belka [Fri, 14 May 2021 10:30:17 +0000 (12:30 +0200)]
python-ipaddress: Remove this python2 module
- python-ipaddress is the python2 backport of the python3 built in
ipaddress module. Therefore python-ipaddress is not needed with the
move to try and remove python2
- Remove the lfs and rootfiles and adjust make.sh
Signed-off-by: Adolf Belka <adolf.belka@ipfire.org> Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Adolf Belka [Thu, 13 May 2021 20:44:30 +0000 (22:44 +0200)]
python3-inotify: Update to 0.2.10 and convert to python3
- Update from 0.2.7 to 0.2.10
- Convert from python-inotify to python3-inotify
make.sh, lfs & rootfiles
- Update rootfiles
- Changelog
0.2.8: - We now just *skip* the event if not known
- Implement InotifyTree and InotifyTrees as sub-classes of new BaseTree
class
- Made InotifyTree and InotifyTrees sub-classes of new base class
BaseTree
- Recursively watch a list of paths/trees
0.2.9: - Added getter for Inotify object from tree objects
- Added note to docs about race-conditions. Added small change for
redundant adds.
- Slightly reorganized documentation. Updated example.
- Merge pull request #35 from dsoprea/dustin. Added extensive unit-test
coverage. Closes all bug requests.
- Added large amount of unit-test coverage.
- Now handle rename-specific events.
- Can now also ignore issues with new directories not existing if
you're created *and* deleted or renamed a folder since the last
time events were read.
- Adjusted requirements for simplicity.
- Added Python 3 compatibility.
- Fixed Unicode support.
- Can now provide `filter_predicate` to event_gen() to allow custom
loop termination based on events.
- We'll now terminate the loop when certain events are encountered.
These events are passed into event_gen() as `terminal_events`. By
default these are the IN_Q_OVERFLOW and IN_UNMOUNT types.
- Fixes #28
- Fixes #23
- Fixes #22
- Fixes #19
- Fixes #16
- Fixes #15
- Fixes #5
- Check presence of both glibc errno and musl libc err
- Support for musl libc (Alpine Linux)
- Merge pull request #27 from jessesuen/master. Support for musl libc
(Alpine Linux)
- Check presence of both glibc errno and musl libc err
- Merge pull request #26 from hathcock/hathcock/issue-25. resolves #25,
list of binary paths can't be logged with existing call
- Support for musl libc (Alpine Linux)
- Resolves #25, list of binary paths can't be logged with existing
call
0.2.10: - Merge pull request #34 from davidparsson/
feature/support-moved-directories
- Support MOVED_FROM and MOVED_TO in BaseTree
- events: Now log event types from epoll vs data stream.
- This release implicitly fixed the botched binary package released
in 0.2.9
Signed-off-by: Adolf Belka <adolf.belka@ipfire.org> Reviewed-by: Peter Müller <peter.mueller@ipfire.org> Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Adolf Belka [Thu, 13 May 2021 16:47:08 +0000 (18:47 +0200)]
python3-dateutil: Removal of python3-six as a dependency
- python3-dateutil is installed as a python3 module.
- It had python3-six defined as a dependency. Python3-six is a module that
allows a project to be capable of neing run under python2 or under
python3
- With the planned removal of python2 there is no need to have
python3-dateutil capable of working with python2.
- python3-six addon is being removed as there is no need for any python3
module in IPFire to also be capable o0f running under python2
Signed-off-by: Adolf Belka <adolf.belka@ipfire.org> Reviewed-by: Peter Müller <peter.mueller@ipfire.org> Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Adolf Belka [Thu, 13 May 2021 16:47:07 +0000 (18:47 +0200)]
python-six: Removal of python2 & 3 addon versions of six
- six is a python compatibility module to enable modules to run on
both python2 and python3. The code from six has to be copied into
any other module/project that is intending to use it.
- With the planned removal of python2 then neither version of this
compatibility module is needed.
- Removal of the lfs and rootfiles. Although python-six is an addon
its rootfile was installed into the common folder rather than the
packages folder.
- Removal of the python-six and python3-six entries in make.sh
Signed-off-by: Adolf Belka <adolf.belka@ipfire.org> Reviewed-by: Peter Müller <peter.mueller@ipfire.org> Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Adolf Belka [Thu, 13 May 2021 11:43:31 +0000 (13:43 +0200)]
make.sh: Removal of four python2 modules
- Removal of python-clientform, python-feedparser, python-mechanize
and python-rssdler addons - lfs and rootfiles
- python-clientform was made obsolete in 2008 and its functions taken
over by python-mechanize
- python-rssdler is an RSS feed downloader to facilitate downloading of
podcasts, videocasts and torrents. Current IPFire version is 0.4.0a
from 2008. The most recent version available is 0.4.2 from 2009. No
further development is being done with this module. An RSS feed
downloader is not appropriate for use in a firewall system, even
less so when it is 12 years old.
- python-feedparser and python-mechanize are both dependencies for
python-rssdler. They are not dependencies for any other addon.
- Full clean build with these four modules fully removed gave no problems.
Signed-off-by: Adolf Belka <adolf.belka@ipfire.org> Reviewed-by: Peter Müller <peter.mueller@ipfire.org> Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Adolf Belka [Tue, 11 May 2021 19:16:39 +0000 (21:16 +0200)]
libxslt: Update to remove python2 modules and disable static libraries
- Add --without-python and --disable-static to the configure section
python2 modules not required for libxslt use in IPFire
disable build of static libraries
- Update rootfile
Signed-off-by: Adolf Belka <adolf.belka@ipfire.org> Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Adolf Belka [Sat, 27 Mar 2021 21:14:50 +0000 (22:14 +0100)]
libcdada: New addon - dependency for pmacct
- This package is required for its library which is a dependency for pmacct
- url for developer is https://msune.github.io/libcdada/
- No change for this patch compared to the V1 version
Signed-off-by: Jon Murphy <jon.murphy@ipfire.org> Signed-off-by: Erik Kapfer <ummeegge@ipfire.org> Signed-off-by: Adolf Belka <adolf.belka@ipfire.org> Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Adolf Belka [Sat, 27 Mar 2021 21:14:49 +0000 (22:14 +0100)]
pmacct: New addon
- What is it?
pmacct is a monitoring tool for network management tasks. Data collected
can be used for analysis and troubleshooting purposes to maintain the
health of the network. pmacct can collect, replicate and export network
information. It can cache in memory tables, store persistently to SQLite3
and output to flat-files like CSV, formatted, and JSON.
- Why is it needed?
To monitor data usage (IP-based or MAC-based data accounting) down to the
client level. Net-Traffic will monitor traffic for the entire RED, GREEN,
etc. networks, but it cannot pinpoint which client is using lots of data.
Connections will take a snapshot but not show day by day sums. pmacct can
help admins keep tabs on users that use too much data.
- What are the use cases?
An ISP may implement data caps and if the limit is over-run then you have
to pay for every additional xxGB of data used. Typical charges can be
around $10 per 50GB. With pmacct you can identify the high users and take
action, hopefully before the limit is breached.
- This is being introduced as a command line only tool. However, at a later
date, if it is useful to enough additional users a WUI page could be
developed as discussed in the development mailing list
https://lists.ipfire.org/pipermail/development/2021-January/009174.html
- Changes in V2 version
- Initscript is using IPFire template and installed with IPFire method.
- All other daemons except pmacct and pmacctd have been removed from the install.
- Example conf files have been removed from /etc/pmacct
Both example conf files are described in the pmacct wiki draft.
Tested-by: Jon Murphy <jon.murphy@ipfire.org> Signed-off-by: Jon Murphy <jon.murphy@ipfire.org> Signed-off-by: Erik Kapfer <ummeegge@ipfire.org> Signed-off-by: Adolf Belka <adolf.belka@ipfire.org> Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Adolf Belka [Mon, 10 May 2021 17:03:08 +0000 (19:03 +0200)]
iotop: Update to work with python3
- v2 version has updated rootfile. One line was accidentally missed out of
original patch submission.
- Change from building with python2 to python3
- iotop setup.py used "itervalues" which is no longer used by python3
In python3 this has been changed to "values". Patch created to update
this in the source tarball setup.py
- Update lfs file with patch application and use of python3
- Update of rootfile
- Installed updated version into vm testbed and iotop confirmed working
Signed-off-by: Adolf Belka <adolf.belka@ipfire.org> Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
In file included from ./boost/regex/v5/perl_matcher_non_recursive.hpp:23,
from ./boost/regex/v5/perl_matcher.hpp:572,
from ./boost/regex/v5/regex.hpp:45,
from ./boost/regex.hpp:34,
from libs/regex/build/../src/wide_posix_api.cpp:25:
./boost/regex/v5/mem_block_cache.hpp:91:11: error: 'static_mutex' in namespace 'boost' does not name a type
91 | boost::static_mutex mut;
| ^~~~~~~~~~~~
./boost/regex/v5/mem_block_cache.hpp: In member function 'void* boost::re_detail_500::mem_block_cache::get()':
./boost/regex/v5/mem_block_cache.hpp:106:37: error: 'mut' was not declared in this scope; did you mean 'put'?
106 | std::lock_guard<std::mutex> g(mut);
| ^~~
| put
./boost/regex/v5/mem_block_cache.hpp: In member function 'void boost::re_detail_500::mem_block_cache::put(void*)':
./boost/regex/v5/mem_block_cache.hpp:120:37: error: 'mut' was not declared in this scope; did you mean 'put'?
120 | std::lock_guard<std::mutex> g(mut);
| ^~~
| put
./boost/regex/v5/mem_block_cache.hpp: In static member function 'static boost::re_detail_500::mem_block_cache& boost::re_detail_500::mem_block_cache::instance()':
./boost/regex/v5/mem_block_cache.hpp:137:52: error: 'BOOST_STATIC_MUTEX_INIT' was not declared in this scope; did you mean 'BOOST_STATIC_CONSTANT'?
137 | static mem_block_cache block_cache = { 0, 0, BOOST_STATIC_MUTEX_INIT, };
| ^~~~~~~~~~~~~~~~~~~~~~~
| BOOST_STATIC_CONSTANT
./boost/regex/v5/mem_block_cache.hpp:137:77: error: too many initializers for 'boost::re_detail_500::mem_block_cache'
137 | static mem_block_cache block_cache = { 0, 0, BOOST_STATIC_MUTEX_INIT, };
| ^
...failed updating 2 targets...
make: *** [boost:102: /usr/src/log/boost_1_76_0] Error 1
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Adolf Belka [Fri, 7 May 2021 16:22:10 +0000 (18:22 +0200)]
boost: Update to 1_76_0 and using python3
- Update from 1_71_0 to 1_76_0
- Update rootfile for x86_64 and copy for other architectures by replacing
x64 with x32 for i586, a32 for armv5tel and a64 for aarch64
- Make build use python3
- add link=shared to build to only have shared libraries created, except for
libboost_exception and libboost_test_exec_monitor which are only
created as static versions
- Changelog
Fixes
algorithm::reduce with crop now does not remove the counts in flow
bins anymore if the selected range actually overlaps with the flow
bins, making the treatment of flow bins consistent with inner bins
accumulators::mean and accumulators::weighted_mean now compute the
correct variance when operator+= was called, for example, when
histograms with accumulators are added; this was wrong before
leading to too small variances
detection of weight support in user-defined accumulators was broken
at compile-time if accumulator used operator+= instead of
operator(); tests only accidentally passed for builtin weighted_sum
Signed-off-by: Adolf Belka <adolf.belka@ipfire.org> Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Adolf Belka [Fri, 7 May 2021 11:44:25 +0000 (13:44 +0200)]
poppler: Update to 21.05.0
- Update from 21.04.0 to 21.05.0
- Update of rootfile
- Ran find-dependencies - nothing found
- Changelog
Release 21.05.0:
core:
* Fix crashes in malformed files
* Export SplashFont* symbols used by Scribus
* Minor code improvements
glib:
* Enhance find to support multi-line matching
qt5/qt6:
* Make sure new signatures are always properly oriented
* Allow to pass the border width when signing
utils:
* pdftoppm: Fix regression when using single scaleTo. Issue #1062
build system:
* Allow to disable building manual tests
Signed-off-by: Adolf Belka <adolf.belka@ipfire.org> Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Adolf Belka [Fri, 7 May 2021 11:44:11 +0000 (13:44 +0200)]
nano: Update to 5.7
- Update from 5.6 to 5.7
- Update of rootfile not required
- Changelog
Changes between v5.6.1 and v5.7:
build: drop the check for two functions that we don't use any more
build: fix compilation for --enable-tiny plus --enable-multibuffer
build: fix compilation when configured with --disable-multibuffer
build: fix compilation when configured with --enable-tiny
bump version numbers and add a news item for the 5.7 release
chars: implement mblen() ourselves, for efficiency
chars: implement mbtowc() ourselves, for more efficiency
chars: work around a UTF-8 bug in glibc, to display invalid codes right
chars: work around the wrong private-use-character widths on OpenBSD
display: avoid determining twice from and until where to draw each row
display: make the output of --constantshow less jittery
editing: prevent the pointer for the top row from becoming dangling
feedback: upon first switch to a buffer, show its error message (if any)
files: always register the format, also when the file is unwritable
files: create a new buffer earlier, so that error messages can be stored
files: when Mac format has been detected, stay with it
gnulib: pull in the fix for a build problem on older Debian
gnulib: update to its current upstream state
indicator: adjust the size to the number of visible lines, not chunks
input: accept Unicode codes for non-characters as valid, since they are
memory: do not allocate space for multidata when it's already allocated
memory: fix an off-by-one error to free also the last line in a group
memory: prevent a use-after-free when the user respects a lock file
oops: that doesn't work -- you can't break out of two for loops at once
options: retire the obsolete 'smooth', 'morespace', and 'nopauses'
softwrap: avoid time-consuming computations, to burden large files less
startup: do not crash when trying to open a device or directory
startup: do not store an error message in the record of another buffer
startup: save the compiled file-matching regexes, to avoid recompiling
startup: show the helpful message only when ^G has not been rebound
syntax: c: colorize also labels that contain digits, and uncolorize colon
syntax: po: improve the coloring of format specifiers
syntaxes: replace [[:space:]] with [[:blank:]] to exclude carriage return
tweaks: adjust and improve one comment, and frob another
tweaks: adjust two comments, and reshuffle two fragments
tweaks: avoid a warning on newer compilers, by writing an extra byte
tweaks: avoid calling extra_chunks_in() when not softwrapping
tweaks: avoid converting a file name for more than will fit on screen
tweaks: avoid parsing a multibyte character twice
tweaks: condense three comments, drop another, and rewrap a line
tweaks: drop unneeded braces and adjust indentation after previous change
tweaks: elide a call of strlen() for every row
tweaks: elide a function that is now basically just two lines
tweaks: elide an unneeded resetting NULL call to wctomb()
tweaks: elide a small function that is used just once
tweaks: elide the pointless is_valid_unicode() function
tweaks: elide two more instances of useless character copying
tweaks: improve a couple of comments
tweaks: morph a function into what it is actually used for
tweaks: normalize the indentation after an earlier change
tweaks: put the most likely condition first, for a quicker return
tweaks: reduce the maximum character length from six bytes to four
tweaks: remove a misplaced (and nested) #ifdef
tweaks: rename a variable, away from an abbreviation
tweaks: rename a variable, for contrast with another
tweaks: reshuffle a comment, and put the main extension first
tweaks: reshuffle a fragment of code, to prepare for the next change
tweaks: reshuffle two conditions, to have the most unlikely one first
tweaks: set the file format only when unset, so it doesn't need saving
tweaks: shorten a comment and trim an #ifdef
tweaks: simplify two fragments of code
tweaks: simplify two fragments of code, eliding useless character copying
syntax: c: make the highlighting of '#include <...>' more compliant
syntax: tcl: support Expect scripts too
Changes between v5.6 and v5.6.1:
bump version numbers and add a news item for the 5.6.1 release
options: rename 'highlightcolor' to the more distinct 'spotlightcolor'
search: correctly colorize a match also when softwrapping is active
tweaks: rename a symbol, to better match the corresponding option
Signed-off-by: Adolf Belka <adolf.belka@ipfire.org> Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Adolf Belka [Fri, 7 May 2021 11:43:56 +0000 (13:43 +0200)]
meson: Update to 0.58.0
- Update from 0.57.2 to 0.58.0
- Updated rootfile
- Changelog is too long to include here.
Full details can be found at
https://mesonbuild.com/Release-notes-for-0-58-0.html
Signed-off-by: Adolf Belka <adolf.belka@ipfire.org> Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Adolf Belka [Fri, 7 May 2021 11:43:36 +0000 (13:43 +0200)]
libxcrypt: Update to 4.4.20
- Update from 4.4.19 to 4.4.20
- Update of rootfile not required
- Changelog
Version 4.4.20
* Fix build when the CFLAGS variable, that is passed into the
configure script, has a leading whitespace character in it
(issue #125).
Signed-off-by: Adolf Belka <adolf.belka@ipfire.org> Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Adolf Belka [Fri, 30 Apr 2021 16:13:32 +0000 (18:13 +0200)]
samba: Update to 4.14.4
- Update from 4.14.3 to 4.14.4
- Update of rootfile not required
- Changelog
Release Notes for Samba 4.14.4 April 29, 2021
This is a security release in order to address the following defect:
o CVE-2021-20254: Negative idmap cache entries can cause incorrect
group entries in the Samba file server process token.
Details
o CVE-2021-20254:
The Samba smbd file server must map Windows group identities (SIDs) into unix
group ids (gids). The code that performs this had a flaw that could allow it
to read data beyond the end of the array in the case where a negative cache
entry had been added to the mapping cache. This could cause the calling code
to return those values into the process token that stores the group
membership for a user.
Most commonly this flaw caused the calling code to crash, but an alert user
(Peter Eriksson, IT Department, Linköping University) found this flaw by
noticing an unprivileged user was able to delete a file within a network
share that they should have been disallowed access to.
Analysis of the code paths has not allowed us to discover a way for a
remote user to be able to trigger this flaw reproducibly or on demand,
but this CVE has been issued out of an abundance of caution.
Signed-off-by: Adolf Belka <adolf.belka@ipfire.org> Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Adolf Belka [Fri, 30 Apr 2021 16:13:18 +0000 (18:13 +0200)]
cmake: Update to 3.20.2
- Update from 3.20.0 to 3.20.2
- Update rootfile
- Changelog
3.20.1
-The FindIntl module in CMake 3.20.0 added checks
Intl_HAVE_GETTEXT_BUILTIN, Intl_HAVE_DCGETTEXT_BUILTIN, and
Intl_IS_BUILTIN, but they were not implemented correctly. These have
been removed and replaced with a single Intl_IS_BUILT_IN check, whose
name is consistent with the FindIconv module.
-The -rpath linker flag is now specified as supported on all Apple
platforms, not just macOS. The install_name_dir used for iOS, tvOS
and watchOS should now default to @rpath instead of using a full
absolute path and failing at runtime when the library or framework
is embedded in an application bundle (see XCODE_EMBED_<type>).
3.20.2
-The Intel Classic 2021 compiler version numbers are now detected
correctly as having major version 2021. CMake 3.20.1 and below were
not aware of a change to the identification macro version scheme
made by Intel starting in version 2021, and detected the version
as 20.2.
-The Intel oneAPI Fortran compiler is now identified as IntelLLVM.
The oneAPI 2021.1 Fortran compiler is missing an identification
macro, so CMake 3.20.1 and below identified it as Intel. CMake now
has a special case to recognize oneAPI 2021.1 Fortran as IntelLLVM.
The oneAPI 2021.2 Fortran compiler defines the proper identification
macro and so is identified as IntelLLVM by all CMake 3.20 versions.
Signed-off-by: Adolf Belka <adolf.belka@ipfire.org> Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Adolf Belka [Wed, 5 May 2021 21:43:06 +0000 (23:43 +0200)]
fetchmail: Update to 6.4.19 and using python3
- Update from 6.3.26 (2013-04-23) to 6.4.19 (2021-04-24)
- Update rootfile
- Delete fetchmail-6.3.26-permit-build-without-ssl3.patch as it is not
needed with version 6.4.19
- Added command to use python3 to lfs
- Changelog is too large to include here
Full details can be found in NEWS file in the source tarball
Signed-off-by: Adolf Belka <adolf.belka@ipfire.org> Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>