This change enables host qemu support for or1k which among other
things allows to test this architecture in gitlab.
The or1k support was named or32 until Qemu 2.9.0 and then renamed or1k
in upstream commit [1]. Since we're already using Qemu 4.2.0, we use
the or1k name.
Tested using qemu_or1k_defconfig.
[1] https://git.qemu.org/?p=qemu.git;a=commit;h=4a09d0bb34ab030e09e87173b2e3ec0fd7616cff
Signed-off-by: Romain Naour <romain.naour@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
(cherry picked from commit a41fae16d6)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
There is no host variant for SDL2 library in Buildroot.
So the qemu build system will try to detect automatically the
external SDL2 libraries installed on the host.
$ ldd output/host/bin/qemu-system-aarch64
[...]
libSDL2-2.0.so.0 => /lib64/libSDL2-2.0.so.0
Disable explicitely sdl2 options (named sdl) to improve the
build reproducibility.
Signed-off-by: Romain Naour <romain.naour@gmail.com>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
(cherry picked from commit 44e5da60b7)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
bzip2 support is needed for reading bzip2-compressed dmg images.
But the host-bzip2 is missing from host qemu package dependency,
so the qemu build system will try to detect automatically the
external libbzip2 libraries installed on the host.
$ ldd output/host/bin/qemu-system-aarch64
[...]
libbz2.so.1 => /lib64/libbz2.so.1
or
libbz2.so.1.0 => output/host/lib/libbz2.so.1.0
if host-bzip2 is built before host-qemu.
Disable explicitely bzip2 options to improve the build
reproducibility.
Signed-off-by: Romain Naour <romain.naour@gmail.com>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
(cherry picked from commit 743fceb2ed)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
The vnc support is enabled by default and the build system
will try to detect automatically some external libraries
installed on the host for vnc-png, vnc-jpeg and vnc-sasl.
$ ldd output/host/bin/qemu-system-aarch64
[...]
libpng16.so.16 => /lib64/libpng16.so.16
or
libpng16.so.16 => output/host/lib/libpng16.so.16
if host-libpng is built before host-qemu.
Disable explicitely thoses options to improve the build
reproducibility.
Signed-off-by: Romain Naour <romain.naour@gmail.com>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
(cherry picked from commit ece36b9a46)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
There is no host-libssh in Buildroot, avoid qemu build system
to find libssh from the host.
Under certain circumstances (host distribution, openssl version), the
qemu-system binary fail to start:
host/bin/qemu-system-aarch64: symbol lookup error: /lib64/libssh.so.4: undefined symbol: EVP_KDF_ctrl, version OPENSSL_1_1_1b
$ ldd output/host/bin/qemu-system-aarch64
[...]
libssh.so.4 => /lib64/libssh.so.4
Explicitely disable libssh for the host variant.
Signed-off-by: Romain Naour <romain.naour@gmail.com>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
(cherry picked from commit 0c4a80c7c4)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
According to the depends used the package does not need C++.
Signed-off-by: Bernd Kuhls <bernd.kuhls@t-online.de>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
(cherry picked from commit 08935f2a86)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Previous address is still valid but not used as primary address any
longer
Signed-off-by: Davide Viti <zinosat@gmail.com>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
(cherry picked from commit 3d1583729c)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Without a source for randomness, qpdf will crash with 'no such file'
error. It's can me tested by command like 'qpd some_pdf.pdf -'.
This problem breaks cups printing.
This patch change configure options to '--with-random=/dev/urandom'
Signed-off-by: Alexey Lukyanchuk <skif@skif-web.ru>
[yann.morin.1998@free.fr: reword commit log]
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
(cherry picked from commit 69dfbbd33b)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
- This is the first release after Suricata joined the Oss-Fuzz program,
leading to discovery of a number of (potential) security issues:
https://suricata-ids.org/2020/04/28/suricata-4-1-8-released
- Drop first, second and fourth patches (already in version)
- Update indentation of hash file (two spaces)
Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
(cherry picked from commit 6138dee5e7)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
- ChangeLog:
- compression bomb protection
- memory handling issue found by Oss-Fuzz
- improve handling of anomalies in traffic
- Drop first patch (already in version)
- Update indentation of hash file (two spaces)
Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
(cherry picked from commit b3d5194696)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Commit b80712a16a ("configs/odroidc2:
remove the defconfig") has removed the odroidc2 defconfig, but left
behind a number of files in board/hardkernel/odroidc2, which are now
unused. Let's remove them.
Signed-off-by: Dagg Stompler <daggs@gmx.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
(cherry picked from commit 197da62866)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
The package was removed a long time ago.
Signed-off-by: Asaf Kahlon <asafka7@gmail.com>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
(cherry picked from commit 427697f297)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
- I fixed build problems on cups and cups-filters but don't use them.
- gtest, libpam-radius-auth, libpam-tacplus and perl-file-util were
used in my previous job. I don't have access to the packages that
use them neither to the corresponding test infrastructure anymore.
Signed-off-by: Carlos Santos <unixmania@gmail.com>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
(cherry picked from commit 1c3ffaed57)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Back in 2014, about 6 years ago now, in my infinite wisdom, I
decided that extracting the tzdata source was not necessary for
the target variant, because we would be installing the files
generated by the host variant, in commit 7aad5daa5d (package/tzdata:
only compile the zoneinfo once).
However, that did not account for the fact that we would eventually
like to have the licensing information for tzdata, later added in
2019, in commit 60889ccdf0 (package/tzdata: bump to version 2019b).
However, that last comit only added the license file to the host
variant, without explanations why that was so. It turns out that the
reason it was not added to the target variant is, probably, that he
source code for the target variant is not extracted, and thus saving
the license file fails.
But we really want the license file for what goes on into the target.
So, do extract the source code for the target variant, even if only to
get the license file.
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
Cc: Richard Braun <rbraun@sceen.net>
Cc: Martin Bark <martin@barkynet.com>
Cc: Christopher McCrory <chrismcc@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
(cherry picked from commit 5c0c4861be)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Signed-off-by: Alexander Dahl <post@lespocky.de>
[yann.morin.1998@free.fr: two spaces in hash file]
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
(cherry picked from commit 2f976c31b0)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Signed-off-by: Alexander Dahl <post@lespocky.de>
[yann.morin.1998@free.fr: two spaces in hash file]
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
(cherry picked from commit 7656cae01e)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
This is a minor release which provides fixes for CVE-2020-11793,
CVE-2020-3887, CVE-2020-3894, and CVE-2020-3899.
Updating from 2.28.0 also brings a few rendering fixes, a build fix
on MIPS64, a build fix for GStreamer 1.12, and solves a couple of
crashes. The full release notes covering 2.28.1 and 2.28.2 can be
found at:
https://webkitgtk.org/2020/04/13/webkitgtk2.28.1-released.htmlhttps://webkitgtk.org/2020/04/24/webkitgtk2.28.2-released.html
A detailed security advisory can be found at:
https://webkitgtk.org/security/WSA-2020-0004.html
Note that the above does not cover all the CVEs, and a new advisory
including them is expected to be published in the next days.
Signed-off-by: Adrian Perez de Castro <aperez@igalia.com>
[yann.morin.1998@free.fr: two spaces in hash file]
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
(cherry picked from commit 080f4251ad)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
This is a minor release which provides fixes for CVE-2020-11793,
CVE-2020-3887, CVE-2020-3894, and CVE-2020-3899.
Updating from 2.28.0 also brings a few rendering fixes, a build fix
on MIPS64, a build fix for GStreamer 1.12, and solves a couple of
crashes. The full release notes covering 2.28.1 and 2.28.2 can be
found at:
https://wpewebkit.org/release/wpewebkit-2.28.1.htmlhttps://wpewebkit.org/release/wpewebkit-2.28.2.html
A detailed security advisory can be found at:
https://wpewebkit.org/security/WSA-2020-0004.html
Note that the above does not cover all the CVEs, and a new advisory
including them is expected to be published in the next days.
Signed-off-by: Adrian Perez de Castro <aperez@igalia.com>
[yann.morin.1998@free.fr: two spaces in hash file]
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
(cherry picked from commit e028d52b7e)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Signed-off-by: Adrian Perez de Castro <aperez@igalia.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
(cherry picked from commit 6ebd152853)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Signed-off-by: James Hilliard <james.hilliard1@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
(cherry picked from commit 60992a8ce1)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Signed-off-by: James Hilliard <james.hilliard1@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
(cherry picked from commit 90e735f6d9)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Adjust fix build with musl patch for 2.28.0.
Signed-off-by: James Hilliard <james.hilliard1@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
(cherry picked from commit cd75229976)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Fixes a number of out of bounds memory accesses. For details, see the
release notes:
https://github.com/tytso/e2fsprogs/blob/v1.45.6/doc/RelNotes/v1.45.6.txt
Signed-off-by: Ryan Barnett <ryanbarnett3@gmail.com>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
(cherry picked from commit 3938f87e41)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
mesa3d-headers is a little bit special: it uses the same sources as
the mesa3d package, named just 'mesa' upstream. mesa uses the meson
buildsystem, an that is what we use in the mesa3d package.
However, mesa3d-headers does not install the whole of mesa; it only
installs a select set of headers for those binary blobs that do not
provide them.
mesa does not provide such a feature (only installing headers) with
its meson buildsystem. As a consequence, we've made mesa3d-headers a
generic package, that basically only copies headers over.
Additionally, mesa3d-headers also provides the dri.pc file for when
Xorg is enabled; see 7468b60e7c (package/mesa3d-headers: also install
dri header and .pc file).
We used to manually generate that file from a .in template that was
present in mesa source code at the time it was still using autotools.
But when they switched over to using meson, the template was dropped
[0], and the dri.pc is now entirely generated using meson internals
[1].
So we now have no template present in the source code, so we must
come up with our own. This simplifies the replacement pattern to
just inject the version string.
[0] https://cgit.freedesktop.org/mesa/mesa/commit/?id=158758618264eac113025a86a360dc305ed4498b
[1] https://cgit.freedesktop.org/mesa/mesa/tree/src/mesa/drivers/dri/meson.build?h=19.2#n93
Signed-off-by: Bernd Kuhls <bernd.kuhls@t-online.de>
Tested-by: Vincent Fazio <vfazio@xes-inc.com>
[yann.morin.1998@free.fr:
- entirely rework the commit log
]
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
(cherry picked from commit 9014c21cac)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
paho-mqtt-c by default enables the building of test materials and
install of CPack documentation:
PAHO_ENABLE_TESTING - "Build tests and run"
PAHO_ENABLE_CPACK - "Enable CPack"
Let's disable these to save a couple megabytes and time. This is
in keeping with the generic settings in pkg-cmake.mk.
Signed-off-by: Danomi Manchego <danomimanchego123@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
(cherry picked from commit 5686d69eef)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
For details, see the announcement:
https://lists.zx2c4.com/pipermail/wireguard/2020-April/005295.html
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
(cherry picked from commit 66962e9bb8)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Force Release otherwise libraries will be suffixed by _debug which will raise
unexpected build failures with packages that use gflags (e.g. rocksdb)
Fixes:
- http://autobuild.buildroot.org/results/3545774258babc3584f69e7d523efdf7fff1acb5
Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
(cherry picked from commit 198e42d6eb)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
When 4bcc344464 was applied, it was not
noticed that 96dc6701af (from another
contributor) had already been
applied. 4bcc344464 essentially did the
same thing as 96dc6701af, except it also
disable zstd support explicitly in the host-libarchive package.
Let's drop the part of 4bcc344464 that
duplicates 96dc6701af.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
(cherry picked from commit 37e853d63c)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
zstd is available since version 3.3.3 and
26838cf5c1
Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
(cherry picked from commit 4bcc344464)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
lz4 is available since version 3.2.0 and
724f3f918e
Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
(cherry picked from commit 9a514c72cd)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Signed-off-by: Christian Stewart <christian@paral.in>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
(cherry picked from commit e505bc4777)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Signed-off-by: Christian Stewart <christian@paral.in>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
(cherry picked from commit 93fecb7ed5)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
When booting with 'console=<empty>' in the kernel command line (as e.g.
U-Boot does with silent flags in effect), opening /dev/console fails.
As per POSIX [0], when iany redirection fails, the shell running exec
shal exit in error. So, when 'console=<empty>' is specified.
/dev/console can't be opened, and the redirection fails, and /init is
killed.
That behaviour was fixed on the kernel side with commit 2bd3a997befc2
(Open /dev/console from rootfs), present since 2.6.34, released in May
2010, so any [dr]ecent kernel will have that fix.
Furthermore, busybox will fix things up anyway (in bb_sanitize_stdio()),
falling back to opening /dev/null if no console is availble. systemd
does a similar thing (in make_console_stdio()), and sysvinit again has
a similar approach (in console_init()).
The archealogy search turned up those relevant commits:
2011-08-04 10a130f91e initramfs/init: make sure that 0, 1, 2 fds are available
introduces the three exec redirections in initramfs
2011-09-06 3fac21ef8d cpio: fix boot with dynamic /dev
introduces the three exec redirections in cpio
2011-09-06 13a3afc536 fs/initramfs: refactor with fs/cpio
dropped the initramfs tweaks to reuse the cpio ones
2012-11-04 e1ebae700a fs/common: Create initial console device
introduces the /dev/console char,5,1 pseudo device creation in
cpio
2018-03-31 dec061adce fs/cpio: don't extend packages' permissions table
switched from the permission-table to a manual mknod to create
/dev/console
The redirections were added before we could guarantee there was a
/dev/console in the rootfs.
We're now guaranteed to have /dev/console in an initramfs, and any recent
kernel will automatically open /dev/console before spawning /init.
The three redirections are useless now, and cause harm under certain
conditions. Drop them.
[0] https://pubs.opengroup.org/onlinepubs/9699919799/utilities/V3_chap02.html#tag_18_20_01
Signed-off-by: Timo Ketola <timo.ketola@exertus.fi>
Cc: Peter Korsgaard <peter@korsgaard.com>
[yann.morin.1998@free.fr:
- extend commit log with the analysis done with Peter
]
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
(cherry picked from commit 98a6f1fc02)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
The reason why the external wireguard kernel module is not allowed with
kernel headers >= 5.6 is that wireguard is included in the upstream kernel
since 5.6 rather than some kind of (fixable) incompatibility issue. Adjust
the comment to make that clear.
While we're at it, drop the redundant !5.6 dependency on the kernel headers
dependency comment. If headers are older than 3.10, then they are also
older than 5.6, so the statement is redundant.
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
(cherry picked from commit 74a865b1fc)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Since wireguard is built into kernels 5.6 and later we can't build
wireguard-linux-compat on them, so we need to depend on
!BR2_TOOLCHAIN_HEADERS_AT_LEAST_5_6.
Signed-off-by: James Hilliard <james.hilliard1@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
(cherry picked from commit da5afc10a4)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Latest version of Apache introduce a new apxs with a slightly modified
path handling logic. In order to simplify the crosscompilation, the
software removes the common prefix from bin install dir and build
install dir, but for this to work they both should have a common prefix.
So we introduce a new regexp to fix /usr/bin to staging dir, the regexps
are also fixed to replace only the exact path between double quotes, to
avoid replacing the she-bang line.
Fixes:
http://autobuild.buildroot.net/results/c41f31566974209897a3a1ec35afe2536fb248cchttp://autobuild.buildroot.net/results/b93f19976ce96e79ea159c25ed74a7377c78f334
Signed-off-by: Angelo Compagnucci <angelo@amarulasolutions.com>
[yann.morin.1998@free.fr:
- add the last few words about the she-bang blurb
- do not use quotes in the existing /sur/bin regexp
]
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
(cherry picked from commit aa04edab77)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Downloading from the cgit repo is now broken and it is a mirror of the
Gitlab repo so use the Gitlab repo.
Signed-off-by: Sébastien Szymanski <sebastien.szymanski@armadeus.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
(cherry picked from commit 9ea9dd021f)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Fix for xtensa PR ld/25861 introduced a regression in handling negative
symbol differences resulting in linker performing incorrect relaxation
or failing to link. Fix XTENSA_NDIFF relocation handling.
Backported from:
https://sourceware.org/git/gitweb.cgi?p=binutils-gdb.git;a=commitdiff;h=d548f47df4d2e3d117d504a4c9977982c78a0556
Fixes: f0291ef4ab ("package/binutils: fix xtensa PR ld/25861")
Signed-off-by: Max Filippov <jcmvbkbc@gmail.com>
Tested-by: Romain Naour <romain.naour@gmail.com>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
(cherry picked from commit f31db17a8d)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
NXP apparently changed file naming convention for this particular file.
Fixes: 1d0ea8d433 (package/freescale-imx/imx-gpu-g2d: bump to version 6.4.0.p1.0)
Signed-off-by: Timo Ketola <timo.ketola@exertus.fi>
Reviewed-by: Gary Bisson <gary.bisson@boundarydevices.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
(cherry picked from commit cd3dc5b4fa)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
python-jedi bundles its own copy of typeshed since version 0.14.0 and
7d2b7bb3c1
So add it to the license files (and update indentation of hash file to
two spaces while at it)
Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
(cherry picked from commit ab98c1ffb6)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Notice: 5.5.x is now EOL, so should be dropped at the next version bump.
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
[Peter: drop 5.5.x / 5.6.x bump]
(cherry picked from commit 72a6e50da9)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Sparc support is deprecated and may be removed in future releases. There are
two choices to fix this issue:
1) Set --enable-deprecated-ports=yes in the CONF_OPTS to supress the error.
2) Remove support for Sparc.
Because this port is deprecated, it's safer to remove support alltogether.
Fixes:
http://autobuild.buildroot.net/results/692820b4b6d4da42cd557fa7badbbd11806bbeba/
Signed-off-by: Adam Duskett <Aduskett@gmail.com>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
(cherry picked from commit 6ee7de3d2f)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Several directories and files are currently not installed during the
target installation, these include:
- conf
Several configuration files, including security configuration files which
may be necessary for running various java applications.
- legal
This directory contains legal notices that some java applications may
require, as they may print legal information and will throw exceptions at
runtime if the legal files are not present on the system.
- release
This file contains a list of modules included in the image.
Because these directories take up less than of megabyte extra, it is not an
issue to install all of them.
Signed-off-by: Adam Duskett <Aduskett@gmail.com>
Reviewed-by: Ryan Barnett <ryan.barnett@rockwellcollins.com>
Tested-by: Ryan Barnett <ryan.barnett@rockwellcollins.com>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
(cherry picked from commit 63b576095b)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Currently, Buildroot installs the jre libraries using
cp -dprf /build/linux-*-release/images/jre/lib/* $(TARGET_DIR)/usr/lib/
However, if a system has a merged /usr directory, and there is a built kernel
before installing OpenJDK, the installation fails because jre/lib has binary
modules file, which causes the following error: cp: cannot overwrite directory
'/usr/lib/modules with non-directory
The obvious fix is to install the modules to /usr/lib/jvm/ and set the
appropriate rpaths via the --with-extra-ldflags conf option. However, this fix
does not work because the built binaries themselves do not link against
libjava.so
Indeed, running readelf on the built java binary reports the following:
"(RUNPATH) Library runpath: [/usr/lib/jvm]" and /usr/lib/jvm/libjava.so exists.
However, when running the Java binary on the target, the following error
occurs: "Error: could not find libjava.so."
The following is the result of "strace java" ran on the target:
faccessat(AT_FDCWD, "/usr/lib/libjava.so", F_OK) = -1 ENOENT
faccessat(AT_FDCWD, "/usr/jre/lib/libjava.so", F_OK) = -1 ENOENT
newfstatat(AT_FDCWD, "/usr/lib/libjava.so", 0x7ffe7b4af8, 0) = -1 ENOENT
newfstatat(AT_FDCWD, "/usr/lib/jvm/libjli.so", [sic] AT_SYMLINK_NOFOLLOW) = 0
As seen above, the java binary searches for libjli.so in /usr/lib/jvm,
which demonstrates that the java binary searches for some of the
DT_NEEDED libraries using the correct rpath. But libjava.so is not
searched from the rpath; it is instead dl-opened manually, looked for in
the search paths hardcoded to the following directories:
- /usr/lib/
- /usr/jre/lib/
- $(dirname $0)/../lib/
The reason behind the hardcoded paths given by the maintainers is due to
historical purposes for the need to support several java versions at the
same time on a single system, and that changing the above behavior is not
likely to ever happen.
As such, most distributions such as Redhat do the following:
- Create the directory /usr/lib/jvm/java-$(JAVA_VERSION)/
- Install all directories and files found in images/jre to that directory.
- Symlink the binaries to in /usr/lib/jvm/java-$(JAVA_VERSION)/bin to
/usr/bin.
However, because Buildroot does not need to support multiple versions of java
concurrently, there is no need for the additional java-$(JAVA_VERSION)
directory.
To fix the above issue, the following changes are performed:
- Introduce the variable "OPENJDK_INSTALL_BASE" which points to /usr/lib/jvm
- Set the --with-extra-ldflags conf_opt to
"-Wl,-rpath,$(OPENJDK_INSTALL_BASE)/lib,-rpath,
$(OPENJDK_INSTALL_BASE)/lib/$(OPENJDK_JVM_VARIANT)"
- Run "mkdir -p $(TARGET_DIR)/usr/lib/jvm/" in the INSTALL_TARGET_CMDS step.
- Copy both the lib and bin directories to /usr/lib/jvm/
- Symlink the binaries in /usr/lib/jvm/bin/ to /usr/bin.
Fixes: https://bugs.busybox.net/show_bug.cgi?id=12751
Signed-off-by: Adam Duskett <Aduskett@gmail.com>
Reviewed-by: Ryan Barnett <ryan.barnett@rockwellcollins.com>
Tested-by: Ryan Barnett <ryan.barnett@rockwellcollins.com>
[yann.morin.1998@free.fr: fix two remaining mis-placed '/']
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
(cherry picked from commit 3edb915709)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
