This toollchain is almost 10 years old now, and not maintained anymore.
Let's therefore remove it from Buildroot. Newer maintained external
toolchains for MIPS are still available from Bootlin
Signed-off-by: Titouan Christophe <titouan.christophe@mind.be>
[Romain: remove leftover link to binutils 2.27 bug 19908]
Signed-off-by: Romain Naour <romain.naour@smile.fr>
For release note since version 38, see [1].
This commit introduces changes in package patches:
- 0001: Patch dropped. An similar change is included in this release.
See [2].
- 0002: Patch no longer needed since the package build recipe sets
CFLAGS without -Werror, and the package makefile sets -Werror only
by default. See [3], included since version 38.
- 0003: Rebased on version 39 and renamed to 0001. The patch is also
flagged as "Upstream: Not applicable".
- 0004: Patch dropped. Included in this release. See [4].
This commit also removes all patch entries in ".checkpackageignore"
(since the remaining patch has its "Upstream:" tag).
This version 39 also fixes few build failures. Those can be seen by
running the command "utils/test-pkg -a -p efivar". The first group
of build failures is:
br-arm-basic [28/45]: FAILED
br-i386-pentium4-full [31/45]: FAILED
br-mips64-n64-full [33/45]: FAILED
br-mips64r6-el-hf-glibc [34/45]: FAILED
linaro-aarch64 [38/45]: FAILED
linaro-arm [39/45]: FAILED
Fixes:
/buildroot/host/opt/ext-toolchain/bin/../lib/gcc/aarch64-linux-gnu/7.3.1/../../../../aarch64-linux-gnu/bin/ld: .data not found for insert
collect2: error: ld returned 1 exit status
efivar 38 was using linker scripts, which was not working in all
cases. Those issues are fixed by the upstream commit [5] which
removes the use of this linker script (included in this release).
The "test-pkg -a -p efivar" also caught another kind of build
failures:
br-i386-pentium-mmx-musl [32/45]: FAILED
Fixes:
In file included from efivar.h:18,
from efisec.h:24,
from secdb-dump.c:7:
list.h: In function 'list_sort':
list.h:152:2: warning: implicit declaration of function 'qsort_r'; did you mean 'qsort'? [-Wimplicit-function-declaration]
152 | qsort_r(array, nmemb, sizeof(*array), cmp, state);
| ^~~~~~~
| qsort
Those failures were introduced in commit f24029b561 "package/efivar:
bump to version 38". This is because efivar introduced a usage of the
qsort_r() libc function, in upstream commit [6], first included in
version 38.
Musl libc added the qsort_r() function in upstream commit [7], included in
version v1.2.3 (2022-04-07). So external toolchains including a Musl older
than this version will fail. But given how old this issue is (musql 1.2.3
is included in Buildroot since 2022.05), this issue is ignored.
uClibc-ng external toolchains are also not affected, since it added
the qsort_r() function in commit [8] included since its first version
v1.0.0 (2015-02-02). So there is no need to exclude external uclibc
toolchains.
[1] https://github.com/rhboot/efivar/releases/tag/39
[2] https://github.com/rhboot/efivar/commit/4f3da3dc351d7743d91327e74fcaaa13299eeb39
[3] https://github.com/rhboot/efivar/commit/998f617cec92d526e1fadb745673ceef63fa1483
[4] https://github.com/rhboot/efivar/commit/cece3ffd5be2f8641eb694513f2b73e5eb97ffd3
[5] https://github.com/rhboot/efivar/commit/cfd686de51494d3e34be896a91835657ccab37d4
[6] https://github.com/rhboot/efivar/commit/62afa2aa588fb0a6ff56acdd268b9f3c557028b8
[7] https://git.musl-libc.org/cgit/musl/commit/?id=b76f37fd5625d038141b52184956fb4b7838e9a5
[8] https://gogs.waldemar-brodkorb.de/oss/uclibc-ng/commit/515d54433138596e81267237542bd9168b8cc787
Signed-off-by: Julien Olivain <ju.o@free.fr>
[Peter: drop dependency on !external musl]
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
A RISC-V 64bit EFI defconfig was added in commit 82199551
"configs/qemu_riscv64_virt_efi: new defconfig". So efivar can now be
enabled for this architecture.
This commit adds BR2_RISCV_64 in BR2_PACKAGE_EFIVAR_ARCH_SUPPORTS.
Signed-off-by: Julien Olivain <ju.o@free.fr>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
For years already, $(HOST_DIR)/usr is deprecated. Install directly in
$(HOST_DIR). This will trigger a check-package error in the following
commit.
Signed-off-by: Arnout Vandecappelle <arnout@mind.be>
Drop backported patches and rebase the ones that are still relevant.
From version 37 to 38, there is a new build system in place, which is
still custom Makefiles.
This introduces a host-efivar package to account for the 'makeguids'
tool. Previously it was not too cumbersome to work around building that
tool for the host, but with the new build system it is increasingly
coupled with the target build that the least hacky way to work around it
is with a host package and a Makefile patch.
Signed-off-by: Erico Nunes <nunes.erico@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
It's time to finally switch over globally to the new spacing format
that we have agreed on for the hash file, with 2 spaces as a separator
between fields.
This commit was mechanically generated using:
find . -type f -name '*.hash' | xargs sed -i 's%^md5[ \t]*\([^ \t]*\)[ \t]*\(.*\)$%md5 \1 \2%'
find . -type f -name '*.hash' | xargs sed -i 's%^sha1[ \t]*\([^ \t]*\)[ \t]*\(.*\)$%sha1 \1 \2%'
find . -type f -name '*.hash' | xargs sed -i 's%^sha256[ \t]*\([^ \t]*\)[ \t]*\(.*\)$%sha256 \1 \2%'
find . -type f -name '*.hash' | xargs sed -i 's%^sha512[ \t]*\([^ \t]*\)[ \t]*\(.*\)$%sha512 \1 \2%'
This commit can easily be backported on the LTS branch by re-running
the same commands, if needed.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
Fix the following build failure with gcc 10:
/home/buildroot/autobuild/run/instance-1/output-1/host/bin/aarch64-none-linux-gnu-gcc -D_LARGEFILE_SOURCE -D_LARGEFILE64_SOURCE -D_FILE_OFFSET_BITS=64 -O2 -I/home/buildroot/autobuild/run/instance-1/output-1/build/efivar-37/src/include/ -specs=/home/buildroot/autobuild/run/instance-1/output-1/build/efivar-37/gcc.specs -L. -fPIC -Wl,-z,muldefs -D_LARGEFILE_SOURCE -D_LARGEFILE64_SOURCE -D_FILE_OFFSET_BITS=64 -o efivar efivar.c -lefivar -ldl
In file included from efivar.h:28,
from efivar.c:40:
In function 'text_to_guid',
inlined from 'parse_name.constprop' at efivar.c:157:8:
guid.h:106:2: error: 'strncpy' output may be truncated copying 8 bytes from a string of length 38 [-Werror=stringop-truncation]
106 | strncpy(eightbytes, text, 8);
| ^~~~~~~~~~~~~~~~~~~~~~~~~~~~
cc1: all warnings being treated as errors
Fixes:
- http://autobuild.buildroot.org/results/fcba72d359f4128515560e9105384cd4deff5043
Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
Drop 0001-Use-z-muldefs-to-avoid-the-multiple-definitions-bug-.patch and
0003-Remove-some-extra-const-that-gcc-complains-about.patch as they are now
upstream.
The upstream repo moved to the 'rhboot' github project, so adjust upstream
URL in .mk and help text to match.
Drop dependency on !musl as it is now support since e04281e60cf0d
(makeguids: Ensure compatibility with other libcs).
Drop BINTARGETS workaround as this is fixed since 6c674283697 (Don't build
static by default).
Drop popt dependency as it is no longer needed since 1aec5e7891 (Replace
popt usage with getopt_long in efivar.c).
While we are at it, also add a hash for the license file.
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
The check-package script when ran gives warnings on ordering issues
on all of these Config files. This patch cleans up all warnings
related to the ordering in the Config files for packages starting with
the letter e in the package directory.
The appropriate ordering is: type, default, depends on, select, help
See http://nightly.buildroot.org/#_config_files for more information.
Signed-off-by: Adam Duskett <Adamduskett@outlook.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
We want to use SPDX identifier for license string as much as possible.
SPDX short identifier for LGPLv2.1/LGPLv2.1+ is LGPL-2.1/LGPL-2.1+.
This change is done using following command.
find . -name "*.mk" | xargs sed -ri '/LICENSE( )?[\+:]?=/s/LGPLv2.1(\+)?/LGPL-2.1\1/g'
Signed-off-by: Rahul Bedarkar <rahulbedarkar89@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
efivar only makes sense on platforms that support UEFI.
UEFI is only supported by some architectures at the moment, being mostly
employed on platforms such as x86, x86_64 and aarch64. Some other
platforms such as MIPS and PowerPC may have some unofficial UEFI
support. UEFI is also limited to little endian architectures.
efivar was being supported in Buildroot without architecture
restrictions so far, however this has led to the creation of a number of
hacks in the recipes, mostly for architectures that are not supported by
UEFI.
In order to avoid spending more time to debug these failures and
maintaining more hacks for unsupported architectures, efivar can be
restricted to that platforms where it makes sense and where it is more
likely to receive some testing and actual usage.
The existing hacks for the now unsupported architectures are removed,
and the dependency is propagated to efibootmgr as it depends on efivar.
Signed-off-by: Erico Nunes <nunes.erico@gmail.com>
Cc: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
uClibc support was recently added to efivar through a small
compatibility patch.
This commit updates a comment in the efivar recipe to reflect this, as
we no longer have glibc as the only supported C library.
Signed-off-by: Erico Nunes <nunes.erico@gmail.com>
Cc: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Upstream has changed version numbering scheme since version 25, which
came after version 0.24.
The two patches that we had in package/efivar/ were backports and so can
be dropped with this bump.
This new version also fixes the build failure with gcc 6.x, so:
Fixes:
http://autobuild.buildroot.net/results/ea6/ea6968d5d0fd0d8ee64e045e2aa56c8e8a0b7b14
[Peter: require GCC 4.9+, adjust Config.in nvme.h comment]
Signed-off-by: Erico Nunes <nunes.erico@gmail.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
The efivar internal build system uses flags such as -std=gnu11 ,
-Wmaybe-uninitialized , -flto which are available in gcc >= 4.7 only.
These flags are passed both to target build and to build a host tool.
For the host part, this has been worked around by overriding
'gcc_flags', but doing that for the target would remove all flags that
upstream intended for the target build.
Buildroot doesn't support building gcc 4.6 in its own toolchain anymore
but it's possible to use an external toolchain with gcc 4.6 which would
be unable to build this package.
This patch adds a limitation on toolchains with target gcc >= 4.7 to
make sure that the flags are available in the chosen toolchain.
Signed-off-by: Erico Nunes <nunes.erico@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Following the removal of eglibc support, this commit replaces all
occurences of "(e)glibc" by just "glibc". Most of the occurences are in
package Config.in comments.
In addition, when the form "an (e)glibc ..." was used, it is replaced by
"a glibc ...".
[Peter: add new efi* packages, s/uclibc/uClibc as suggested by Romain,
systemd / liquid-dsp tweaks as suggested by Yann]
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Reviewed-by: Romain Naour <romain.naour@gmail.com>
Acked-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
The efivar build process starts by building one tool for the host,
which is needed for the rest of the build. This tool currently fails
to build with old gcc versions because the gcc.specs used by efivar
specifies -std=gnu11. To address this, this patch:
- passes 'gcc_flags=' to the host build, so that the custom gcc specs
are not passed. They are in practice not needed for the build of
the simple makeguids host utility.
- passes -std=gnu99 instead of -std=c99 in the build of host
makeguids, because the source code uses anonymous structs and
unions, which requires std=gnu99 and not just std=c99
In addition, the build by default assumes that the target toolchain is
LTO capable, and that therefore you can call gcc-ar, gcc-nm and
gcc-ranlib. This fails short when the target toolchain is for example
gcc 4.7. To address this, we explicitly specify AR, NM and RANLIB to
be used, but pass them as make options instead of in the environment,
in order to override the values specified in the package Makefile.
Fixes:
http://autobuild.buildroot.net/results/fe40c1d139ba8ddeef3dafd5c1818a946f014d7c/
Cc: Erico Nunes <nunes.erico@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
efivar contains tools and libraries to manipulate EFI variables.
This package has some restrictions to build. It needs uchar.h which
apparently does not come in uClibc, and it also needs
__bswap_constant_*() definitions which are only present in glibc.
So, this package was limited to build with glibc toolchains.
Although it probably makes sense mostly on x86, x86-64 and aarch64
architectures, there are no architecture specific limitations which
would prevent it to build for other architectures, so this restriction
was not imposed. This package has been build-tested additionaly on many
architectures which support glibc, such as mips, MicroBlaze, sparc64,
ppc64. There were some build flag issues found when building to some of
these some architectures, although the flags can be also added in the
general case, so they were added by default.
It has also been found that in some host systems it failed to build due
to not generating the .pc files. This has been tracked to the use of
make 3.81, so a patch was prepared for it and was submitted upstream.
There's also a dependency on the linux/nvme.h header, which is somewhat
conturbed, as it has appeared in user space linux headers 3.12 and in
4.4 it was renamed. This has been solved by restricting it to build with
linux headers >= 3.12 and a patch from upstream was picked which fixes
it for linux headers >= 4.4.
Signed-off-by: Erico Nunes <nunes.erico@gmail.com>
[Thomas:
- Pass TARGET_LDFLAGS in LDFLAGS.
- Pass -shared-libgcc only on Microblaze.]
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Titouan Christophe