* Move more environment variables into GoPackage/Environment
* Split GoPackage/Environment into target and build sections
* Do not set GOROOT_FINAL for Go packages (setting it should only affect
the Go compiler and not Go packages)
* Set CGO_LDFLAGS to $(TARGET_LDFLAGS)
* Move GO_TARGET_* variables from golang-values.mk, and GO_VERSION_*
variables from golang-version.mk, into golang/Makefile
This also updates runc, containerd, and docker-ce to reflect the changes
in GoPackage/Environment.
Signed-off-by: Jeffery To <jeffery.to@gmail.com>
gettid happens to be supported by all glibc, musl and uclibc, but
configure cannot detect it when cross-compiling. This adds a configure
variable for it.
Signed-off-by: Sebastian Kemper <sebastian_ml@gmx.net>
Adds configure variables for features that the compiler or the libc
supports and which cannot be detected by apr-util's configure script
when cross-compiling.
Also removes one call to INSTALL_DIR (no functional change here).
Signed-off-by: Sebastian Kemper <sebastian_ml@gmx.net>
apr's configure script uses lots of AC_TRY_RUNs and when cross-compiling
needs a leg-up. This commit adds more configure variables (and removes
one, too).
Notable changes:
- apr_cv_use_lfs64=yes is removed (again) after having a discussion
about it with the musl developers. The conclusion was that
_LARGEFILE64_SOURCE is a horrible thing that we don't want and should
avoid (hence the removal of the variable, because defining
_LARGEFILE64_SOURCE is all it does).
- ap_cv_atomic_builtins is set to "yes" for 64-bit platforms. If
anybody is interested in enabling this for other targets feel free to
send a patch after testing it.
- configure doesn't add -lpthread to LIBS when cross-compiling. This is
not a problem for musl but might be for other libcs. This commit adds
to related variable.
- configure caches are added (via patches) for strerror and /dev/zero
mmap tests. The former fixes a warning on musl (where strerror
returns int instead of a pointer) and the latter is required for
pthread pshared mutex apr_lock implementation and cannot be detected
during cross-compile either.
Signed-off-by: Sebastian Kemper <sebastian_ml@gmx.net>
Today CVE-2020-7221 was publicly discussed on oss-sec [1]. MariaDB
upstream had not mentioned this CVE in their last release notes. The CVE
is related to auth-pam and the possibility of a local mariadb to root
user exploit in the mysql_install_db script.
Upstream has made amendments to the script, but according to the oss-sec
posts the folder permissions were not updated as they should have been.
In OpenWrt the script mysql_install_db is actually patched to never run
the commands in question. This has been this way since MariaDB 10.4 was
made available.
Still, the directory permissions set by the postinstall script are too
lax. To quote the discoverer of the issue, Matthias Gerstner from Suse,
they exhibit "the dangerous situation of a setuid-root binary residing
in a directory owned by an unprivileged user".
This commit fixes this by changing the permissions to the following:
root:mariadb 0750 /usr/lib/mariadb/plugin/auth_pam_tool_dir
This way the setuid-root binary is only available to root and the
mariadb user, while at the same time the mariadb user has no ownership
of the directory.
[1] https://seclists.org/oss-sec/2020/q1/55
Signed-off-by: Sebastian Kemper <sebastian_ml@gmx.net>
Added missing limits header. This is normally included in fortify-headers,
which I have disabled locally.
Signed-off-by: Rosen Penev <rosenp@gmail.com>
With mpd's build system, it requires either an iconv built into the libc
or icu. Since uClibc-ng as configured by OpenWrt currently has no iconv,
use icu for it to work around the problem. This is the simplest solution.
Added a patch to use boost's rounding functions. They are more appropriate
and work with uClibc-ng.
Signed-off-by: Rosen Penev <rosenp@gmail.com>
Rosen Penev