Disable audit for host package to avoid getting the following error if
it is found on host:
[84/662] Generating audit_type-list.txt with a meson_exe.py custom command
In file included from <command-line>:32:
./../src/basic/missing_audit.h:7:10: fatal error: libaudit.h: No such file or directory
7 | #include <libaudit.h>
| ^~~~~~~~~~~~
Fixes:
- http://autobuild.buildroot.org/results/67782c225c08387c1bbcbea9eee3ca12bc6577cd
Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Build with cryptsetup and without libblkid will fail on:
../src/shared/dissect-image.c:1336:34: error: 'N_DEVICE_NODE_LIST_ATTEMPTS' undeclared (first use in this function)
1336 | for (unsigned i = 0; i < N_DEVICE_NODE_LIST_ATTEMPTS; i++) {
| ^~~~~~~~~~~~~~~~~~~~~~~~~~~
This bug has been reported upstream:
https://github.com/systemd/systemd/pull/16901
and is not an issue for the target variant as libblkid is select by
BR2_PACKAGE_UTIL_LINUX_MOUNT
As cryptsetup does not seem needed for host-systemd, just disable it
Fixes:
- http://autobuild.buildroot.org/results/67782c225c08387c1bbcbea9eee3ca12bc6577cd
Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
As 18f6c26118 just did to silence the file lists commands, switch to
using $(Q) instead of a plain @, to silence the commands.
Using $(Q) will allow to debug the commands with V=1.
We keep @ for the calls to MESSAGE, though.
The commands that are not currently silenced are left as-is, and they
can be converted to being silent in a followup patch, if need be,
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
Cc: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
If the modules directory that corresponds to the version of the kernel
being built has been deleted, don't try to run depmod, which will
obviously fail.
This can happen for instance when the modules are stripped from the main
root filesystem, and placed into a separate filesystem image, so that
the root filesystem and the kernel can be updated separately.
Signed-off-by: Paul Cercueil <paul@crapouillou.net>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
Commit 98a6f1fc02 (fs/cpio: make initramfs init script survive 'console='
kernel argument) dropped the explicit /dev/console execs for fd 0,1,2, as
they fail when booted with console= and aren't really needed as the kernel
will setup fd 0,1,2 from /dev/console before executing the initramfs anyway.
Not doing this unfortunately confuses glibc's ttyname_r(3) implementation
(used by E.G. busybox/coreutils 'tty'), causing it to fail with ENOENT as
it does a fstat on fd 0 and tries to match up st_ino / st_dev against the
entries in /dev (since glibc 2.26):
commit 15e9a4f378c8607c2ae1aa465436af4321db0e23
Author: Christian Brauner <christian.brauner@canonical.com>
Date: Fri Jan 27 15:59:59 2017 +0100
linux ttyname and ttyname_r: do not return wrong results
If a link (say /proc/self/fd/0) pointing to a device, say /dev/pts/2, in a
parent mount namespace is passed to ttyname, and a /dev/pts/2 exists (in a
different devpts) in the current namespace, then it returns /dev/pts/2.
But /dev/pts/2 is NOT the current tty, it is a different file and device.
Detect this case and return ENODEV. Userspace can choose to take this as a hint
that the fd points to a tty device but to act on the fd rather than the link.
Signed-off-by: Serge Hallyn <serge@hallyn.com>
Signed-off-by: Christian Brauner <christian.brauner@ubuntu.com>
The reason it fails is that we manually mount devtmpfs on /dev in /init, so
the /dev/console used by the kernel (in rootfs) is not the same file as
/dev/console at runtime (in devtmpfs).
Notice: Once logged in, tty does work correctly. Presumably login reopens
stdin/stdout/stderr.
To fix this, re-add the exec of /dev/console for fd 0,1,2, but only do so if
possible. Because of the above mentioned shell behaviour (specified by
POSIX [0]), perform this check in a subshell.
[0] https://pubs.opengroup.org/onlinepubs/9699919799/utilities/V3_chap02.html#tag_18_20_01
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
- Fix CVE-2020-14349: It was found that PostgreSQL versions before 12.4,
before 11.9 and before 10.14 did not properly sanitize the search_path
during logical replication. An authenticated attacker could use this
flaw in an attack similar to CVE-2018-1058, in order to execute
arbitrary SQL command in the context of the user used for replication.
- Fix CVE-2020-14350: It was found that some PostgreSQL extensions did
not use search_path safely in their installation script. An attacker
with sufficient privileges could use this flaw to trick an
administrator into executing a specially crafted script, during the
installation or update of such extension. This affects PostgreSQL
versions before 12.4, before 11.9, before 10.14, before 9.6.19, and
before 9.5.23.
https://www.postgresql.org/docs/12/release-12-4.html
Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Since commit 0e2be4db8a
("package/pkg-generic: make file list logic parallel build
compatible"), the commands executed at the every end of the build
to assemble the list of files installed by the different packages
are visible in the make output. They are quite noisy, and clutter
the output.
The other commands in target-finalize are also hidden using "@",
so we should also do the same for those commands. But that hurts
debuggability, so we use $(Q) (the existing '@'s can be changed
in a followup patch).
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
[yann.morin.1998@free.fr: use '$(Q)', not '@']
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
Fixes the following security issues:
CVE-2020-15810: HTTP(S) Request Smuggling
Due to incorrect data validation Squid is vulnerable to HTTP Request
Smuggling attacks against HTTP and HTTPS traffic. This leads to cache
poisoning.
https://github.com/squid-cache/squid/security/advisories/GHSA-3365-q9qx-f98m
CVE-2020-15811: HTTP(S) Request Splitting
Due to incorrect data validation Squid is vulnerable to HTTP Request
Splitting attacks against HTTP and HTTPS traffic. This leads to cache
poisoning.
https://github.com/squid-cache/squid/security/advisories/GHSA-c7p8-xqhm-49wv
CVE-2020-24606: Denial of Service processing Cache Digest Response
Due to Improper Input Validation Squid is vulnerable to a Denial of Service
attack against the machine operating Squid.
https://github.com/squid-cache/squid/security/advisories/GHSA-vvj7-xjgq-g2jg
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
wolfSSL version 4.5.0 contains 6 vulnerability fixes: 2 fixes for TLS 1.3,
2 side channel attack mitigations, 1 fix for a potential private key leak
in a specific use case, 1 fix for DTLS including those 3 CVEs:
- Fix CVE-2020-12457: An issue was discovered in wolfSSL before 4.5.0.
It mishandles the change_cipher_spec (CCS) message processing logic
for TLS 1.3. If an attacker sends ChangeCipherSpec messages in a
crafted way involving more than one in a row, the server becomes stuck
in the ProcessReply() loop, i.e., a denial of service.
- Fix CVE-2020-15309: An issue was discovered in wolfSSL before 4.5.0,
when single precision is not employed. Local attackers can conduct a
cache-timing attack against public key operations. These attackers may
already have obtained sensitive information if the affected system has
been used for private key operations (e.g., signing with a private
key).
- Fix CVE-2020-24585: An issue was discovered in the DTLS handshake
implementation in wolfSSL before 4.5.0. Clear DTLS application_data
messages in epoch 0 do not produce an out-of-order error. Instead,
these messages are returned to the application.
Also update hash of LICENSING as well as WOLF_LICENSE due to later
verbage update with
970391319bhttps://www.wolfssl.com/docs/security-vulnerabilities/
Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
Fix CVE-2020-17498: In Wireshark 3.2.0 to 3.2.5, the Kafka protocol
dissector could crash. This was addressed in
epan/dissectors/packet-kafka.c by avoiding a double free during LZ4
decompression.
https://www.wireshark.org/security/wnpa-sec-2020-10.html
Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
Fixes the following security issue:
CVE-2016-10228: An infinite loop has been fixed in the iconv program when
invoked with the -c option and when processing invalid multi-byte input
sequences. Reported by Jan Engelhardt.
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
openal uses std::max_align_t since version 1.20.0 and
585b0cf3be
As a result, it is affected by
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=56019
and the build with gcc <= 4.8 will fail on:
/home/buildroot/autobuild/instance-2/output-1/build/openal-1.20.1/common/almalloc.cpp: In function 'void* al_malloc(size_t, size_t)':
/home/buildroot/autobuild/instance-2/output-1/build/openal-1.20.1/common/almalloc.cpp:20:45: error: 'max_align_t' is not a member of 'std'
alignment = std::max(alignment, alignof(std::max_align_t));
^
Fixes:
- http://autobuild.buildroot.org/results/589c7853ce334c7502f7cd4cdbcaaf3c6840f43b
Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Many of the mender CLI commands use systemctl commands to get information about
the daemon, such as the PID (IE: systemctl show -p MainPID mender-client).
As seen above, these commands expect the service file to be named
"mender-client" instead of "mender."
As such, in the current state, running a forced update check in the CLI will
result in the following error:
failed to force updateCheck: could not find the PID of the mender daemon.
Changing the name of mender.service to mender-client.service fixes the issue.
Signed-off-by: Adam Duskett <Aduskett@gmail.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
On Ubuntu 18.04, make-4.1 emits spurious, incorrect "entering/leaving"
messages, which end up in the LINUX_VERSION_PROBED variable:
printf 'probed linux version: "%s"\n' "$(LINUX_VERSION_PROBED)"
probed linux version: "make[1]: Entering directory '/home/buildroot'
4.19.78-linux4sam-6.2
make[1]: Leaving directory '/home/buildroot/output/build/linux-linux4sam_6.2'"
First, the messages are displayed even though we do explicitly pass
--no-print-directory -s.
Second, the entering and leaving messages are not about the same
directory!
This *only* occurs in the following conditions:
- the user has the correct 0022 umask,
- top-level parallel is used (with or without PPD),
- initial -C is specified as well.
$ umask 0022
$ make -j16 -C $(pwd)
[...]
depmod: ERROR: Bad version passed make[1]:
[...]
(yes, 'make[1]:' is the string depmod is trying, and fails, to parse as
a version string).
If any of the three conditions above is removed, the problem no longer
occurs. Here's a table of the MAKEFLAGS:
| 0002 | 0022 |
----+-------+------------------------------------------------+--------------------------+
| no-j | --no-print-directory -- | |
noC | +------------------------------------------------+--------------------------+
| -j16 | -j --jobserver-fds=3,4 --no-print-directory -- | -j --jobserver-fds=3,4 |
----+-------+------------------------------------------------+--------------------------+
| no-j | --no-print-directory -- | w |
-C | +------------------------------------------------+--------------------------+
| -j16 | -j --jobserver-fds=3,4 --no-print-directory -- | w -j --jobserver-fds=3,4 |
----+-------+------------------------------------------------+--------------------------+
0002: umask == 0002
0022: umask == 0022
no-j: no -j flag
-j16: -j16 flag
noC: no -C flag
-C : -C /path/of/buildroot/
Only the bottom-right-most case fails...
This behaviour goes against what is documented:
https://www.gnu.org/software/make/manual/make.html#g_t_002dw-Option
5.7.4 The ‘--print-directory’ Option
[...]
you do not need to specify this option because ‘make’ does it for
you: ‘-w’ is turned on automatically when you use the ‘-C’ option,
and in sub-makes. make will not automatically turn on ‘-w’ if you
also use ‘-s’, which says to be silent, or if you use
‘--no-print-directory’ to explicitly disable it.
So this exactly describes our situation; yet 'w' is added to MAKEFLAGS.
Getting rid of the 'w' flag makes the build succeed again, so that's
what we do here (bleark, icky)...
Furthermore, the documented way to override MAKEFLAGS is to do so as a
make parameter:
https://www.gnu.org/software/make/manual/make.html#Options_002fRecursion
5.7.3 Communicating Options to a Sub-make
[...]
If you do not want to pass the other flags down, you must change the
value of MAKEFLAGS, like this:
subsystem:
cd subdir && $(MAKE) MAKEFLAGS=
However, doing so does not fix the issue. So we resort to pass the
modified MAKEFLAGS via the environment (bleark, icky)...
Fixes: #13141
Reported-by: Laurent <laurent@neko-labs.eu>
Reported-by: Asaf Kahlon <asafka7@gmail.com>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
Cc: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Fixes the following security issues:
CVE-2020-24332
If the tcsd daemon is started with root privileges,
the creation of the system.data file is prone to symlink attacks
CVE-2020-24330
If the tcsd daemon is started with root privileges,
it fails to drop the root gid after it is no longer needed
CVE-2020-24331
If the tcsd daemon is started with root privileges,
the tss user has read and write access to the /etc/tcsd.conf file
For details, see the advisory:
https://www.openwall.com/lists/oss-security/2020/05/20/3
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
- Fix CVE-2019-5163: An exploitable denial-of-service vulnerability
exists in the UDPRelay functionality of Shadowsocks-libev 3.3.2. When
utilizing a Stream Cipher and a local_address, arbitrary UDP packets
can cause a FATAL error code path and exit. An attacker can send
arbitrary UDP packets to trigger this vulnerability.
- Fix CVE-2019-5164: An exploitable code execution vulnerability exists
in the ss-manager binary of Shadowsocks-libev 3.3.2. Specially crafted
network packets sent to ss-manager can cause an arbitrary binary to
run, resulting in code execution and privilege escalation. An attacker
can send network packets to trigger this vulnerability.
Also update indentation in hash file (two spaces)
Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
Hopefully, this should fix the following error on one of the
autobuilders:
png: no [The C/C++ header for libpng (png.h) could not
be found. You may need to install the development
package.]
Fixes:
- http://autobuild.buildroot.org/results/afddcc44b2fb7983244f24542bfae921869e4ab8
Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
gnuradio-runtimeTargets.cmake and gnuradio-pmtTargets.cmake are filled
using CMAKE_INSTALL_PREFIX for INSTALL_INTERFACE.
Since CMAKE_INSTALL_PREFIX, in buildroot, is set to /usr, these files contains
path to host system.
With BR2_COMPILER_PARANOID_UNSAFE_PATH package using gnuradio fails with:
arm-linux-gnueabihf-g++: ERROR: unsafe header/library path used in cross-compilation: '-isystem' '/usr/include'
By simply providing 'include', produced .cmake contains:
INTERFACE_INCLUDE_DIRECTORIES "${_IMPORT_PREFIX}/include"
instead of
INTERFACE_INCLUDE_DIRECTORIES "/usr/include"
[Upstream status: https://github.com/gnuradio/gnuradio/pull/3737]
Fix (many) gr-osmosdr build failure:
- http://autobuild.buildroot.net/results/66b76c07f15bb3e6db697c47796ae3dd15ecf4b9/
Signed-off-by: Gwenhael Goavec-Merou <gwenhael.goavec-merou@trabucayre.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
Fix CVE-2020-15389: jp2/opj_decompress.c in OpenJPEG through 2.3.1 has a
use-after-free that can be triggered if there is a mix of valid and
invalid files in a directory operated on by the decompressor. Triggering
a double-free may also be possible. This is related to calling
opj_image_destroy twice.
Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
Fix CVE-2020-12762: json-c through 0.14 has an integer overflow and
out-of-bounds write via a large JSON file, as demonstrated by
printbuf_memappend.
Also update indentation in hash file (two spaces)
Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
Since the bump of assimp to 5.0.1, we have build failures on mips64el,
due to relocations being truncated. The issue seems to be quite
similar to the one on m68k coldfire, as both m68k and MIPS have this
-mxgot gcc option to switch to using a GOT that has no size limit (but
causes less efficient code to be produced).
Here as well, the overall relevance of assimp on mips64(el) platforms
being probably very limited, the incentive to search for a better
solution is pretty limited.
Fixes:
http://autobuild.buildroot.net/results/7df487d5117b2ee440a07dbff9cae1b181566748/
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
On m68k coldfire, we already pass -mxgot, but since the bump to assimp
5.0.1, this is no longer sufficient, and we have failures such as:
/tmp/ccqmJLil.s: Assembler messages:
/tmp/ccqmJLil.s:307948: Error: value -43420 out of range
/tmp/ccqmJLil.s:307985: Error: value -38606 out of range
/tmp/ccqmJLil.s:308010: Error: value -38626 out of range
/tmp/ccqmJLil.s:308056: Error: value -33280 out of range
Since these issues only arise when building with -O2, let's disable
the optimization for this package on m68k. The very relative relevance
of assimp on m68k coldfire makes the research of a better solution not
really useful (for the record, assimp is a "library to import various
well-known 3D model formats in a uniform manner").
Fixes:
http://autobuild.buildroot.net/results/a7d4fb2653b0f1be4d036ee46a44e72da0ed4376/
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
Our patch
0003-Add-option-to-make-the-rpath-relative-under-a-specif.patch adds
an option --make-rpath-relative, which we use to tweak RPATH of target
binaries.
However, one of the effect of this option is that it drops RPATH
entries if the corresponding directory does not contain a library that
is referenced by a DT_NEEDED entry of the binary.
This unfortunately isn't correct, as RPATH entries are not only used
by the dynamic linker to resolve the location of libraries listed
through DT_NEEDED entries: RPATH entries are also used by dlopen()
when resolving the location of libraries that are loaded at runtime.
Therefore, the removal of RPATH entries that don't correspond to
directories containing libraries referenced by DT_NEEDED entries break
legitimate uses of RPATH for dlopen()ed libraries.
This issue was even pointed out during the review of the upstream pull
request:
https://github.com/NixOS/patchelf/pull/118#discussion_r329660138
This fixes tst-origin uClibc-ng unit test:
https://github.com/wbx-github/uclibc-ng-test/blob/master/test/dlopen/Makefile.in#L25https://github.com/wbx-github/uclibc-ng-test/blob/master/test/dlopen/tst-origin.c#L15
Without this patch:
$ gcc -o toto toto.c -Wl,-rpath,/tmp/test/bar
$ readelf -d toto | grep PATH
0x000000000000000f (RPATH) Library rpath: [/tmp/test/bar]
$ ./output/host/bin/patchelf --debug --make-rpath-relative /tmp/
toto
patching ELF file `toto'
Kernel page size is 4096 bytes
removing directory '/tmp/test/bar' from RPATH because it does not contain needed libs
new rpath is `'
$ readelf -d toto | grep PATH
0x000000000000001d (RUNPATH) Library runpath: []
With the patch applied:
$ gcc -o toto toto.c -Wl,-rpath,/tmp/test/bar
$ readelf -d toto | grep PATH
0x000000000000000f (RPATH) Library rpath: [/tmp/test/bar]
$ ./output/host/bin/patchelf --debug --make-rpath-relative /tmp/ toto
patching ELF file `toto'
Kernel page size is 4096 bytes
keeping relative path of /tmp/test/bar
new rpath is `test/bar'
$ readelf -d toto | grep PATH
0x000000000000001d (RUNPATH) Library runpath: [test/bar]
Signed-off-by: Yann Sionneau <ysionneau@kalray.eu>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
As we recently stopped testing the x86-64 Sourcery toolchain, it means
we no longer have any x86-64 glibc based toolchain in our
autobuilders. Since this is a pretty common configuration, it makes
sense to test it, which this commit does by adding a config fragment
to use the x86-64 glibc bleeding edge Bootlin toolchain.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
This toolchain uses an old gcc 6.2.0 compiler (not even the latest gcc
from the 6.x series), which fails to build the recent Boost
package. Since newer versions of this toolchain are no longer made
publicly available from Mentor Graphics, our only option is to drop
the toolchain.
Fixes:
http://autobuild.buildroot.net/results/10edaed22c15b9d0f7de187085aeebc96e5ebe6c/
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
This toolchain uses an old gcc 6.2.0, and newer versions of the
toolchain are no longer publicly available. This old gcc 6.2.0 causes
build issues of Boost, which are unfixable without updating the
toolchain. As we're about to drop support for this toolchain entirely,
we must stop testing it in our autobuilder infrastructure.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
This bump is needed to fix a build failure with gcc 10:
https://savannah.nongnu.org/support/?func=detailitem&item_id=110186#options
Here is an extract of the bug report:
"It is not a bug to have variables with the same name in different source
files. The bug was the missing keyword "static".
But there was a different bug that was not tolerated by GCC 10. It was
same strange data type conversions in dav_coda.c. The resolution was to
drop coda altogether because there is still fuse which is better suited
anyway.
The new release 1.6.0 should fix all these problems. Please tell me if
there are still problems with GCC 10."
Also update indentation in hash file (two spaces)
Fixes:
- http://autobuild.buildroot.org/results/42beafade6fd31927c8db14bc52110c0fc5b17c2
Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
Fixes the following security vulnerabilities:
CVE-2020-12695: The Open Connectivity Foundation UPnP specification before
2020-04-17 does not forbid the acceptance of a subscription request with a
delivery URL on a different network segment than the fully qualified
event-subscription URL, aka the CallStranger issue.
For details, see the advisory:
https://w1.fi/security/2020-1/upnp-subscribe-misbehavior-wps-ap.txt
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
In commit 26e37cef16, we started using
__WORDSIZE to get the size of longs on the given architecture, in
order to support all CPU architectures.
Unfortunately, __WORDSIZE is not enabled in musl, so in
19bd089004, we switched to using
LONG_BIT instead of __WORDSIZE.
However, LONG_BIT is not readily available on glibc, you need
_XOPEN_SOURCE to be defined, which was done in
a34e7f88f6.
However, in a34e7f88f6, _XOPEN_SOURCE
was just defined, with no specific value. This caused the build to
break again on uClibc-ng and older glibcs, because clock_gettime() and
CLOCK_MONOTONIC were no longer defined. In both uClibc-ng and glibc,
CLOCK_MONOTONIC is only defined if __USE_POSIX199309 is defined. It
turns out that simply defining _XOPEN_SOURCE with no value does not
lead to __USE_POSIX199309 being defined in uClibc-ng and old glibcs,
while it is defined in newer glibcs.
The difference comes from the following snippet of code, which is
present in recent enough glibc's <feature.h> but not uClibc-ng's or
older glibc's <feature.h>:
/* If none of the ANSI/POSIX macros are defined, or if _DEFAULT_SOURCE
is defined, use POSIX.1-2008 (or another version depending on
_XOPEN_SOURCE). */
So the fact that we are defining _DEFAULT_SOURCE makes it assume that
we're using POSIX 2008.09, which obviously includes POSIX 1993.09.
Due to the lack of this code snippet, uClibc-ng <features.h> only
enables:
!defined _POSIX_SOURCE && !defined _POSIX_C_SOURCE)
but not:
So we need an _XOPEN_SOURCE level of at least 500 for POSIX 1993.09
definitions to be available.
This is confirmed by the feature_test_macros man page, which states:
_XOPEN_SOURCE < 500
_POSIX_C_SOURCE is defined with the value 2.
500 <= _XOPEN_SOURCE < 600
_POSIX_C_SOURCE is defined with the value 199506L.
When this is fixed, another issue arises with older glibc toolchains
(such as Sourcery ARM), where fd_set is no longer defined. Inded, with
POSIX-1.2001 being enabled, we need to include <sys/select.h> to
access the fd_set definition and friends (see man fd_set for details).
This commit was tested with two glibc toolchains (recent and old), one
uClibc-ng toolchain and one musl toolchain.
Fixes:
http://autobuild.buildroot.net/results/e20f9474fc0217036faa6561df33fa983466ddfe/
(uClibc-ng)
http://autobuild.buildroot.net/results/b5d944389fc96ef2c5e0608fe4ac34149e5f9739/
(glibc)
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
Acked-by: Vincent Fazio <vfazio@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
When building on a host that has no name specified for the used group in
/etc/group the script in install step will fail due to missing group name.
CMake Error at CMakeLists.txt:80 (message):
Learning server module group failed: id: cannot find name for group ID 8000
The patch was taken from upstream and modified manually because of merge
conflicts.
Fixes:
http://autobuild.buildroot.net/results/f197ca1def9dc1292e1e784757f2da9d95484431/
Signed-off-by: Heiko Thiery <heiko.thiery@gmail.com>
Cc: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Cc: Matthew Weber <matthew.weber@rockwellcollins.com>
Reviewed-by: Matthew Weber <matthew.weber@rockwellcollins.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
As Thomas said:
> In this sort of situation, we generally don't add any legacy
> handling. Indeed, since the feature is now mandatory... the
> default behavior will always be OK.
> People who could be annoyed are people who had this feature
> disabled... which is now always enabled. But the legacy handling
> will anyway not help those people.
Reported-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
