This patch introduces the new target to compare the current
configuration with the one derived from the defconfig + fragments (if
any). It helps identifying settings not yet inserted to the defconfig
or any fragment.
Signed-off-by: Marcel Patzlaff <m.patzlaff@pilz.de>
Reviewed-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Tested-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
This patch refactors savedefconfig and the configuration re-generation
to separate macros so that the functionality can be re-used in further
rules.
Signed-off-by: Marcel Patzlaff <m.patzlaff@pilz.de>
Reviewed-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Tested-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
Since linux-4.19, the kernel's build system internally touches its
.config file.
However, we currently used that file as a timestamp to detect whether
our kconfig fixups were to be (re)applied or not, which in turn is used
to decide whether we should (re)build the package or not.
But with latest kernel versions, this timestamp heuristic is now broken,
and we always rebuild the kernel on subsequent builds.
We fix that by introducing a separate timestamp file of our own, which
we know the kernel (or the kconfig-based packages, for that matters)
does not use.
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
Cc: Thomas De Schampheleire <patrickdepinguin@gmail.com>
Cc: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
Cc: Arnout Vandecappelle <arnout@mind.be>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
A patch was added to the Linux kernel in 5.1.0-rc3 which adds a
requirement that the host build environment include pkg-config. Add the
correct host-pkgconf dependency and environment variables to ensure
Linux picks up the correct libraries.
Move the existing LINUX_MAKE_ENV assignment earlier, to simplify the
append-assignment in the libelf conditional block.
Fixes: #11761
Cc: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
Cc: Yann E. MORIN <yann.morin.1998@free.fr>
Suggested-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
Signed-off-by: Stuart Summers <stuart.summers@intel.com>
Reviewed-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
[Arnout: extend commit message as suggested by Yann]
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
The kconfig build logic uses the HOSTCC variable to find the host
compiler. It makes sense to explicitly pass a value to this variable,
pointing to the host compiler used by Buildroot.
During the kconfig step, host-ccache is not ready (host-ccache is only
a dependency to the configure step of packages), so we use
$(HOSTCC_NOCCACHE).
Packages currently using the kconfig-package fell into two categories:
- Those not passing any HOSTCC value. For such packages, it was the
default host compiler detected by the kconfig build logic that was
used. ccache was therefore never used. With this commit, those
packages will now be using the host compiler detected by
Buildroot. Packages in this situation: at91bootstrap3, barebox,
busybox, swupdate, uclibc, xvisor.
- Those passing a HOSTCC value. Such packages were passing $(HOSTCC),
which doesn't work as host-ccache will not be ready. This commit
does not fix them, as they still override HOSTCC. It will be fixed
in followup commits. Packages in this situation: uboot and
linux. Note that linux was a bit special, because it has a
KCONFIG_DEPENDENCIES on the toolchain package, so in fact
host-ccache was ready.
So practically speaking, this commit does not fix anything, as the two
only problematic packages that use $(HOSTCC) are not fixed. However,
it makes things more correct by explicitly telling kconfig which
compiler to use.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
Acked-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Some users of kconfig need some packages to be built before their
kconfig infra be used.
For example, the linux kernel, starting with 4.16, needs flex and bison
to generate the parser code. Furthermore, starting with 4.18, it will
also need the cross-compiler before parsing the kconfig stuff, because
that calls the compiler to check its features.
Currently, this is broken, even the flex/bison ones, even though they
are listed, because there is no way to define dependencie that are
guaranteed before the (visual) configurators. For example:
$ make distclean
$ make menuconfig
--> enable the linux kernel, choose a defconfig, save, exit
$ make linux-menuconfig
[...]
HOSTCC scripts/basic/fixdep
HOSTCC scripts/kconfig/conf.o
YACC scripts/kconfig/zconf.tab.c
/bin/sh: bison: command not found
LEX scripts/kconfig/zconf.lex.c
scripts/Makefile.lib:196: recipe for target 'scripts/kconfig/zconf.tab.c' failed
make[3]: *** [scripts/kconfig/zconf.tab.c] Error 127
make[3]: *** Waiting for unfinished jobs....
/bin/sh: flex: command not found
scripts/Makefile.lib:188: recipe for target 'scripts/kconfig/zconf.lex.c' failed
make[3]: *** [scripts/kconfig/zconf.lex.c] Error 127
Makefile:528: recipe for target 'rpc_defconfig' failed
make[2]: *** [rpc_defconfig] Error 2
linux/linux.mk:511: recipe for target '/home/ymorin/dev/buildroot/buildroot/output/build/linux-4.17.11/.config' failed
make[1]: *** [/home/ymorin/dev/buildroot/buildroot/output/build/linux-4.17.11/.config] Error 2
Makefile:79: recipe for target '_all' failed
make: *** [_all] Error 2
So, we introduce a new type of dependencies for kconfig-based packages,
that are guaranteed to be built and installed before the (visual)
configurators are called.
Since those dependencies are phony targets and therefore always out of
date, a normal dependency would cause the .config target to be rebuilt
on each invocation of make. So we use an order-only pre-requisite, like
is done for the patch dependency.
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Cc: Arnout Vandecappelle <arnout@mind.be>
Cc: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
Cc: Jan Kundrát <jan.kundrat@cesnet.cz>
Tested-by: Jan Kundrát <jan.kundrat@cesnet.cz>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
Currently, if the user specifies the path to an existing directory as
the path to the custom (def)config file (FOO_KCONFIG_FILE), then we
happily create a file in there, either 'defconfig' or '.config' (or
whatever the .config is named for that package), depending on whether
we're saving a defconfig or a full config.
So, we could save the file, but then the Buildroot defconfig file that
contains that path would no longer be reusable as-is, because we
interpret that path as a path to a file.
Furthermore, if the directory-portion of FOO_KCONFIG_FILE does not
exist yet, the update would fail, because cp does not create missing
directory components.
So we fix that by adding an explicit test for the directory-ness of the
target file, and then an explicit mkdir to create missing directory
components.
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
When updating back the (def)config file, the touch command is not very
useful by default, so hide it away.
We do not hide away the cp command, as that could confuse the user under
some circunmstances. For example, when the toolchain does not yet exist,
the uClibc buildsystem will complain multiple times about gcc not being
found, like so:
make[2]: /home/ymorin/dev/buildroot/O/host/bin/i686-buildroot-linux-uclibc-gcc: Command not found
(Note that we can not suppress those warnings, as they are on stderr,
and we still want to see stderr in case of real errors).
So, if we were to hide the cp command, the user could be left confused,
even though we were sucessful in updating back the (def)config file.
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
Except for the nice human-friendly reminder of the command that the user
was just running, and the name of the file to copy from, those two rules
are exactly the same.
Make that a common macro that is shared, so that it's easier to add more
checks, and to simplify maintenance.
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
Kconfig clones, such as openconf used by xvisor [1], do not look for
.config at the root of the build directory, but in a subdirectory (e.g.
build/openconf).
This patch introduces a new Makefile variable named
$(2)_KCONFIG_DOTCONFIG, which contains the path to .config relative to
the source directory of the package. It defaults to .config and can be
overridden in the package Makefile.
It also creates the full directory hierarchy where the .config file is
stored, in case it does not exist (e.g. in xvisor).
This allows the use of the kconfig-package infrastructure with packages
relying on such clones.
[1] https://github.com/xvisor/xvisor/tree/master/tools/openconf
Signed-off-by: Eric Le Bihan <eric.le.bihan.dev@free.fr>
Reviewed-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
[Thomas: use "install -D" instead of "mkdir + cp", as suggested by
Yann.]
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Currently, calling foo-reconfigure for a kconfig-based package will not
re-trigger the configuration (kconfig-wise) step for the package.
This can be problematic when using an override-srcdir suring development
and the content of the veride-srcdir changes (e.g. on a rebase, a merge,
or during a bisect).
This is because the configuration (kconfig-wise) of the package is not
done in the _CONFIGURE_CMDS block, but as a separate action that is not
part of any step [0].
So, be sure to remove the stamp file .stamp_kconfig_fixup_done during
the foo-clean-for-reconfigure rule, so that the configuration is applied
again with the new source tree.
We use another rule, foo-clean-kconfig-for-reconfigure, because we do
not want to override the default foo-clean-for-reconfigure rule, and we
have no way to add conditional commands to it.
[0] The reasons it was not done are not entirely clear in my head, but
IIRC that was not working at the time we tried with Thomas DS.
Reported-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Cc: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Cc: Thomas De Schampheleire <patrickdepinguin@gmail.com>
Cc: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Tested-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Reviewed-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
This is basically the same change as in 0515fe4566
("Makefile: pass host PKG_CONFIG_PATH at "make menuconfig" time"). That
commit made sure to pass host PKG_CONFIG_PATH when invoking Buildroot's
own menuconfig program. This change ensures that the same is true for
third party menuconfig programs (i.e. Linux, uClibc and Busybox).
This unbreaks "make {linux,uclibc}-menuconfig" for host platforms which
rely on PKG_CONFIG_PATH to find .pc files (e.g. NixOS). (When Busybox
updates to a more recent Kconfig snapshot, one that uses pkg-config to
find ncurses, "make busybox-menuconfig" will also start working.)
Tested on Ubuntu and NixOS:
$ make qemu_arm_versatile_defconfig
$ make linux-menuconfig
$ make
Signed-off-by: Bjørn Forsman <bjorn.forsman@gmail.com>
Acked-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Using 'yes "" | make oldconfig' to regenerate a .config file from an
existing defconfig does not work reliably in all cases. Specifically, it
does not work well with tristate choice entries.
The correct way to do it is to use 'make olddefconfig', but this target
is not supported in projects that use an old version of kconfig. As
most projects do not use tristate entries, there is no need for them to
upgrade.
'oldnoconfig' is an alias for 'olddefconfig', but it also is the only
name for this target between Linux versions 2.6.36 and 3.6.
Check which targets are supported by the current project, and use the
appropriate method to regenerate the .config file.
Signed-off-by: Romain Izard <romain.izard.pro@gmail.com>
Reviewed-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Tested-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Reviewed-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
This reverts commit 638ea2d715, since it
breaks the build of busybox in many configurations, with the following
issues:
/bin/bash: : command not found
make[1]: Entering directory `/home/test/autobuild/instance-0/output/build/busybox-1.24.2'
[...]
GEN miscutils/Config.in
make[2]: *** No rule to make target `olddefconfig'. Stop.
make[1]: *** [olddefconfig] Error 2
make[1]: Leaving directory `/home/test/autobuild/instance-0/output/build/busybox-1.24.2'
make: *** [/home/test/autobuild/instance-0/output/build/busybox-1.24.2/.config] Error 1
make: Leaving directory `/home/test/autobuild/instance-0/buildroot'
See:
http://autobuild.buildroot.org/results/48481bb3531927be35cda3df57d02f02cec709aa/
for example.
A minimal configuration with just Busybox doesn't exhibit the problem,
but adding libtirpc in the configuration is enough to trigger the
problem, so the following defconfig allows to reproduce the issue:
BR2_arm=y
BR2_TOOLCHAIN_EXTERNAL=y
BR2_TOOLCHAIN_EXTERNAL_CUSTOM=y
BR2_TOOLCHAIN_EXTERNAL_DOWNLOAD=y
BR2_TOOLCHAIN_EXTERNAL_URL="http://autobuild.buildroot.org/toolchains/tarballs/br-arm-full-2016.02-3-g762b7c9.tar.bz2"
BR2_TOOLCHAIN_EXTERNAL_GCC_4_7=y
BR2_TOOLCHAIN_EXTERNAL_HEADERS_3_10=y
BR2_TOOLCHAIN_EXTERNAL_LOCALE=y
BR2_TOOLCHAIN_EXTERNAL_INET_RPC=y
BR2_TOOLCHAIN_EXTERNAL_CXX=y
BR2_PACKAGE_LIBTIRPC=y
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Using 'yes "" | make oldconfig' to regenerate a .config file from an
existing defconfig does not work reliably in all cases. Specifically, it
does not work well with tristate choice entries.
The correct way to do it is to use 'make olddefconfig', but this target
is not supported in projects that use an old version of kconfig. As
most projects do not use tristate entries, there is no need for them to
upgrade.
'oldnoconfig' is an alias for 'olddefconfig', but it also is the only
name for this target between Linux versions 2.6.36 and 3.6.
Check which targets are supported by the current project, and use the
appropriate method to regenerate the .config file.
Signed-off-by: Romain Izard <romain.izard.pro@gmail.com>
[yann.morin.1998@free.fr: use a shell construct instead of makefile code]
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
Cc: Arnout Vandecappelle <arnout@mind.be>
Cc: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
[Thomas: fix minor typos.]
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
When there are no parameters, simply use $(var) instead of $(call var)
Signed-off-by: Romain Izard <romain.izard.pro@gmail.com>
Reviewed-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Use a variable to factorize the calls to make, as we use the same set
of options multiple times.
Signed-off-by: Romain Izard <romain.izard.pro@gmail.com>
Reviewed-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Ensure that each command line in the .config rule is prefixed by $$(Q)
Signed-off-by: Romain Izard <romain.izard.pro@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Since the introduction of _KCONFIG_DEFCONFIG in 8ef62b99, the package's
.config file no longer depends on anything (unless a fragment is
defined). Therefore, there is no dependency anymore between .config
and <pkg>-patch. Thus, it is possible that the .config file is
attempted to be built before the package is extracted and patched.
Usually this works out OK because <pkg>-patch will always be done
before <pkg>-configure, but it will fail when the user calls
<pkg>-menuconfig explicitly. It will also fail when we enable
top-level parallel build.
To solve this, just add an explicit order-only dependency on
<pkg>-patch. It really is only necessary when _KCONFIG_DEFCONFIG is
defined and _KCONFIG_FRAGMENT_FILES is not, but it doesn't hurt to
add it unconditionally.
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
Reported-by: FrAnKenStEiN MC <chfakht@gmail.com>
Acked-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
As of Linux kernel commit ea4d1a8 "powerpc/configs: Replace
pseries_le_defconfig with a Makefile target using merge_config" some
kconfig defconfigs (one so far: "pseries_le_defconfig") can be
generated using "make <defconfig>" and they do not exist on disk as a
single defconfig file.
This causes buildroot's build to fail for those configs with:
'arch/powerpc/configs/pseries_le_defconfig' for 'linux' does not exist
To handle this case and keep the makefile steps as simple as possible,
introduce a new package variable, *_KCONFIG_DEFCONFIG, that can be
used to indicate that a defconfig rule is being used, rather than a
file.
This allows the rule that generates the .config file to use either the
provided file (by copying) or a generated defconfig (by running "make
<defconfig>") as its starting point. merge_config.sh can then be run the
same way in either case, using .config as both input and output.
Note that merge_config.sh is now modifying .config in-place but this
is safe because it uses a temporary copy while making changes.
This patch introduces the new variable but does not make use of it.
Use of the new variable will be introduced in separate patches.
Signed-off-by: Sam Bobroff <sam.bobroff@au1.ibm.com>
[yann.morin.1998@free.fr:
- include all the other kconfig-package hunks into this one patch
- do not transform the 'echo "..."; exit 1' into $(error ...) calls
- use a make $(if)-block instead of a shell if-block to copy the file
or run make (like is done to check the kconfig snippets)
- misc typoes and rephrasing in the commit log
- do not force the _defconfig suffix in the infra (Thomas)
]
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Cc: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Currently, the configurators are using $($(2)_MAKE_ENV), often derived
from $(TARGE_MAKE_ENV), as the environment to be set when calling the
various configurators.
This means that our host tools are used first, most notably pkg-config
(from host-pkgconf).
However, this is inherently flawed. Our pkg-config, when set for the
host, only searches .pc files in $(HOST_DIR) and never ever uses the
ones from the host. For example, since we do not build a host-qt, our
pkg-config would not find the host's QtCore.pc et al.
Consequently, on some systems (but not on others?) most of the
configurators fail to build, especially the latest kernel versions, as
they have been starting to use pkg-config two years ago.
Fix that by filtering-out sensible values out of the environment, but
only when calling the configurators.
[Thomas: rewrap comment to appropriate length.]
Reported-by: Mauro Condarelli <mc5686@mclink.it>
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Cc: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Cc: Mauro Condarelli <mc5686@mclink.it>
Tested-by: Mauro Condarelli <mc5686@mclink.it>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Currently, we override @D for the kconfig configurators, so the fixup
commands can use $(@D); otherwise @D would be simply '.' because it is
not a real file in the package build dir.
However, this breaks the soon-to-be-introduced linux-backports package,
which needs to have a dependency on the linux package to be configured.
The underlying reason is that @D is an automatic variable that is always
set by make to the directory part of the target of the rule. However,
automatic variables loose their "automatic" property when they are
manually set.
Furthermore, a variable that is defined for a rule is inherited by all
dependencies of that rule, so our manually-set @D is inherited all the
way down the dependency chain of linux-backports, down to the linux' own
.config rule, which is thus run with @D pointing to linux-backports'
build dir, not linux'.
Fix that by using a "static pattern rule", redirecting the configurators
to an intermediate stamp-like file which path is in the package build
dir, so we get a valid @D from the onset, without having to manually
fiddle with it.
Thanks to Arnout for suggesting that in the first place. Sorry I did
reject it as "too complex" when it was in fact the best solution.
Suggested-by: Arnout Vandecappelle <arnout@mind.be>
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Cc: Arnout Vandecappelle <arnout@mind.be>
Cc: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Cc: Thomas De Schampheleire <patrickdepinguin@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
When debugging packages that use the pkg-kconfig infrastructure, it
is sometime interesting to see the exact command line that is being
executed when running oldconfig.
But currently, those command lines are prefixed with '@' which makes
them silent, always.
Instead, use $(Q) so they are silent by default, and verbose with V=1.
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Cc: Thomas De Schampheleire <patrickdepinguin@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Currently, this is triggering the error message:
make randconfig
make source
Only limit the check to enforce a config file being set to when we are
actually building, like is done in the various kconfig-using packages.
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
A very interesting use-case for a kconfig-based package is to create a
custom (def)config file based on one bundled with the package itself,
like described in PR-8156:
make menuconfig
-> enable kernel, use an in-tree defconfig, save and exit
make linux-menuconfig
-> enable/disable whatever option, save and exit
make menuconfig
-> change to use a custom defconfig file, set a path, save and exit
make linux-update-config
-> should save to the new custom defconfig file
However, that is currently not possible, because the dependency chain
when saving the configuration goes back up to the (newly-set!) custom
(def)config file, which does not exist.
So, we break the dependency chain so that saving the configuration does
not depend on that file. Instead, we use a terminal rule that checks
that the configuration has indeed been done, and fails if not.
Closes#8156.
Reported-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Cc: Thomas De Schampheleire <patrickdepinguin@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
After we exit the configurators, we need to re-run the kconfig fixups to
ensure the user is not able to override them in the configurators.
Currently, we schedule that "for later", by removing the corresponding
stamp file, so make will run the fixups "later".
This means the user has access to the un-fixed .config file, which he
might decide to copy and use as a reference (not too bad, since we'd run
the fixups anyway; but not clean either).
Note that we still remove the stamp file before running the fixups, in
case any one of those fixups breaks, so we don't want to believe the
fixups have been applied; the fixup macro will touch that file anyway.
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Cc: Thomas De Schampheleire <patrickdepinguin@gmail.com>
Cc: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
The same fixups will have to be done after leaving the configurators,
so we want to commonalise that code.
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Cc: Thomas De Schampheleire <patrickdepinguin@gmail.com>
Cc: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Even though we do have a dependency chain back to each of the kconfig
base and fragment files:
$$($(2)_DIR)/.config: $$($(2)_KCONFIG_FILE) $$($(2)_KCONFIG_FRAGMENT_FILES)
we can't rely on it to ensure they are all present, because they all have
this rule:
$$($(2)_KCONFIG_FILE) $$($(2)_KCONFIG_FRAGMENT_FILES): | $(1)-patch
but since this rule has no prerequisite (only build-order, but that does
not count in this case) and no recipe, make will believe each missing
file to be a PHONY target, and will always run targets that depend on
it:
https://www.gnu.org/software/make/manual/make.html#Force-Targets
So, that means a missing kconfig base or fragment file would always
cause the rule to generate .config to be run at each invocation, which
in turn would cause a rebuild of the kernel, which is clearly not what
we want.
Since this is expected make behaviour, we can well end up with a missing
Kconfig base or fragment. To avoid continuously rebuilding the kernel in
that case, we must check those files exist by ourselves, and error out
if any one of them is missing.
One would expect we check for them right in their dependency rule, like
so:
$$($(2)_KCONFIG_FILE) $$($(2)_KCONFIG_FRAGMENT_FILES): | $(1)-patch
[ -f $(@) ] || {echo Missing $(@) >&2; exit 1; }
but that does not work, as only the first target is tested for. That
check msut be turned into a loop explicitly testing all files, like so:
$$($(2)_KCONFIG_FILE) $$($(2)_KCONFIG_FRAGMENT_FILES): | $(1)-patch
for f in $$($(2)_KCONFIG_FILE) $$($(2)_KCONFIG_FRAGMENT_FILES); do \
[ -f $(@) ] || {echo Missing $$$${f} >&2; exit 1; }; \
done
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Cc: Floris Bos <bos@je-eigen-domein.nl>
Cc: Thomas De Schampheleire <patrickdepinguin@gmail.com>
Cc: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Kconfig fragments may be present in the package, so we need to extract
and patch it before we can use the fragments.
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Cc: Floris Bos <bos@je-eigen-domein.nl>
Cc: Thomas De Schampheleire <patrickdepinguin@gmail.com>
Cc: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Currently, we only check that the variable is defined, which is not
enough since we really want it to be non-empty.
We however can't check it points to an existing file, because the
package might well not be extracted yet, and we may use an internal
defconfig. If that file does not eventually exist, there will be a
failure down the road at build time when we try to copy it...
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Cc: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Cc: Thomas De Schampheleire <patrickdepinguin@gmail.com>
Reviewed-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Adds functionality to the kconfig infrastructure to merge additional
configuration fragment files to the main configuration file of
kconfig packages, using support/kconfig/merge_config.sh
Typical use-case is when you want your configuration to be
kept in sync with an upstream (def)config file, but do require
some minor local modifications.
Disables -update-config and -update-defconfig targets when
fragment files are set.
[Thomas: take into account comments made by Arnout:
- Minor fixes in the documentation changes
- Add @ before the tests done in the $(1)-update-config and
$(1)-update-defconfig targets.]
Signed-off-by: Floris Bos <bos@je-eigen-domein.nl>
Tested-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
Tested-by: Gergely Imreh <imrehg@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Just like the previous commit did for the pkg-generic infrastructure,
this commit improves the pkg-kconfig infrastructure to declare its
custom <pkg>-<something> targets as PHONY.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Reviewed-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
When a package is not enabled, we need not check that it provides
a .config/defconfig file.
So far, all our pkg-kconfig packages unconditionally set their
_KCONFIG_FILE, even to an empty value. But some packages (e.g.
at91bootstrap3) wanted to set it only when enabled, and broke.
So, only do the check when the package is enabled.
Reported-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Cc: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Cc: Thomas De Schampheleire <patrickdepinguin@gmail.com>
Acked-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Signed-off-by: "Yann E. Morin" <yann.morin.1998@free.fr>
Signed-off-by: Thomas De Schampheleire <thomas.de.schampheleire@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Some packages (like linux) may install things inside images/ as well, so
remove the associated stamp file after running the configuration editor.
Signed-off-by: Thomas De Schampheleire <thomas.de.schampheleire@gmail.com>
Signed-off-by: "Yann E. Morin" <yann.morin.1998@free.fr>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Replace a 'cp --preserve' with a 'touch --reference' so that the code
for foo-update-config and foo-update-defconfig is symmetrical to ease
maintainability and increase clarity.
Signed-off-by: Thomas De Schampheleire <thomas.de.schampheleire@gmail.com>
Signed-off-by: "Yann E. Morin" <yann.morin.1998@free.fr>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
This commit adds support for using a defconfig file instead of a full
.config. This is a precondition to migrate packages like linux and
barebox to the kconfig infrastructure.
Signed-off-by: Thomas De Schampheleire <thomas.de.schampheleire@gmail.com>
Signed-off-by: "Yann E. Morin" <yann.morin.1998@free.fr>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
When the configuration file of a package is located inside of the
package sources, a make dependency can only be expressed after the
package has been extracted (and patched).
Signed-off-by: Thomas De Schampheleire <thomas.de.schampheleire@gmail.com>
Signed-off-by: "Yann E. Morin" <yann.morin.1998@free.fr>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
When saving (aka updating) the configuration file of a kconfig package,
a subsequent call to "make" would rebuild the package, even though the
configuration did not actually change.
It took quite a while to understand why. But the reason is so simple and
obvious, when you think about it:
- $(@D)/.config depends on the config file $(BUSYBOX_CONFIG_FILE)
- busybox-update-config then copies $(@D)/.config back to the config
file $(BUSYBOX_CONFIG_FILE)
- so the config file is newer than $(@D)/.config
So, in the next run, pkg-kconfig believes that it has to rebuild
busybox. Bummer... :-/
So, the fix is very trivial, and just requires asking "cp" to preserve
timestamps, so the dependency does not kick in at the next run.
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Cc: Thomas De Schampheleire <patrickdepinguin@gmail.com>
Reviewed-by: Thomas De Schampheleire <thomas.de.schampheleire@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
To be consistent with the recent change of FOO_MAKE_OPT into FOO_MAKE_OPTS,
make the same change for FOO_KCONFIG_OPT.
Sed command used:
find * -type f | xargs sed -i 's#_KCONFIG_OPT\>#&S#g'
Signed-off-by: Thomas De Schampheleire <thomas.de.schampheleire@gmail.com>
Reviewed-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
The configure step of the busybox package performs 'make oldconfig', which
causes the .config file to be updated. Thus, the .config file is more recent
than our stamp file .stamp_kconfig_fixup_done. On a subsequent build, our
dependency rules would kick in, and run the config fixup again, thus
kicking in the package's configure, build and install steps yet once
more, that, ad infinitum.
One solution is to modify kconfig-package to introduce an explicit touch of
the .kconfig_fixup_config_done stamp file, as post-configure hook.
Another solution, implemented by this patch, is to move the oldconfig call
from the package's .mk file to the kconfig-package infrastructure and make
sure it is done as part of the fixup commands. This way, the stamp file will
only be touched once, after the full fixup (including oldconfig) and no
endless rebuilds will occur.
Reported-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Suggested-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Signed-off-by: Thomas De Schampheleire <thomas.de.schampheleire@gmail.com>
Tested-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Acked-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
The current layout of comments wrt. to the related code-blocks is a bit
hard to handle visually.
Remove empty lines between the comments and their respective related
code-blocks.
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Acked-by: Thomas De Schampheleire <thomas.de.schampheleire@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
There are several packages that have a configuration file managed by
kconfig: uclibc, busybox, linux and barebox. All these packages need some
make targets to handle the kconfig specificities: creating a configuration
(menuconfig, ...) and saving it back (update-config, ...)
These targets should be the same for each of these packages, but
unfortunately they are not. Especially with respect to saving back the
configuration to the original config file, there are many differences.
A previous set of patches fixed these targets for the uclibc package.
This patch extracts these targets into a common kconfig-package
infrastructure, with the goals of:
- aligning the behavior of all kconfig-based packages
- removing code duplication
In order to use this infrastructure, a package should at a minimum specify
FOO_KCONFIG_FILE and eval the kconfig-package macro. The supported
configuration editors can be set with FOO_KCONFIG_EDITORS and defaults to
menuconfig only.
Additionally, a package can specify FOO_KCONFIG_OPT for extra options to
pass to the invocation of the kconfig editors, and FOO_KCONFIG_FIXUP_CMDS
for a list of shell commands used to fixup the .config file after a
configuration has been created/edited.
Signed-off-by: Thomas De Schampheleire <thomas.de.schampheleire@gmail.com>
[yann.morin.1998@free.fr: add missing 4th argument when calling to
inner-kconfig-package (namely, 'target']
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
