Not all systems have /bin/bash (e.g. NixOS[1] doesn't). Buildroot
already uses /usr/bin/env shebangs for other interpreters (perl,
python), so why not bash?
This changes only the shebangs used by Buildroot itself; stuff installed
to the target system is left unchanged.
With this applied I can run Buildroot unmodified on NixOS.
[1]: http://nixos.org/
Signed-off-by: Bjørn Forsman <bjorn.forsman@gmail.com>
Acked-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
The make "-s" option is used to enable the "Silent operation" so if that
option is used don't print anything as far as there isn't any error.
Add the "-s" option to "apply-patches.sh" to enable silent operation.
[Peter: use the existing QUIET variable]
Signed-off-by: Fabio Porcedda <fabio.porcedda@gmail.com>
Reviewed-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Let mkusers create groups alone, useful for supplementary permissions in
udev/systemd for example where users can be added to later at runtime.
Use a magic string "-" to signal that user creation should be skipped.
Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
Acked-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
As discussed during the Buildroot meeting, this commit extends the
pkg-stats script to include statistics about the number of packages
having vs. not having the hash file.
As of today, we have 104 packages with the hash file, and 1274
packages without.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Currently the graph-build-time script prints a python exception if a
needed module cannot be imported. Catch the exception and tell the user
which packages are missing, as we do for other missing dependencies.
Signed-off-by: Luca Ceresoli <luca@lucaceresoli.net>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
This instruction in the middle of 'import' lines looks very strange.
Also, it was not obvious to me what the 'Agg' backend is.
Both things are actually correct, but it took a while to find out why.
So clarify with a comment to save someone else's time.
[Peter: fix s/soe/some/ typo]
Signed-off-by: Luca Ceresoli <luca@lucaceresoli.net>
Cc: Sascha Arthur <sascha.arthur@gmail.com>
Cc: "Yann E. MORIN" <yann.morin.1998@free.fr>
Cc: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Signed-off-by: Luca Ceresoli <luca@lucaceresoli.net>
Acked-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Perl modules can have three different types of dependencies:
- configure/build time dependency which becomes host dependency
- runtime dependency which becomes target dependency
- test time dependency which is useless in a cross-compiling context like BR
Before this patch, test time dependencies are handled like runtime
dependencies.
After this patch, test time dependencies are ignored by default. The
newly added -test option allows to add them anyway if needed.
[Thomas: reword commit log using Francois proposal.]
Signed-off-by: Francois Perrad <francois.perrad@gadz.org>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
mktemp --tmpdir is not available on older Redhat RHEL5 machines. The
alternative that has the same behavior is 'mktemp -t'.
Signed-off-by: Thomas De Schampheleire <thomas.de.schampheleire@gmail.com>
Reviewed-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
The gconv libraries are used to translate between different character sets
('charsets', even 'csets' sometimes). Some packages need them to present
text to the user (eg. XBMC Gotham).
In (e)glibc they are implemented by the internal implemenation of iconv,
called gconv, and are provided as dlopen-able libraries.
Note that some gconv modules need extra libraries (shared by more than
one gconv module), so we must, when adding a subset of modules, scan the
installed modules in search of the missing libraries.
[Thomas: add general explanation in expunge-gconv-modules and fix
coding style.]
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Cc: Maxime Hadjinlian <maxime.hadjinlian@gmail.com>
Cc: Bernd Kuhls <bernd.kuhls@t-online.de>
Signed-off-by: Bernd Kuhls <bernd.kuhls@t-online.de>
Cc: Eric Limpens <limpens@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
In English, unlike in French, almost all usages of the word 'information'
are uncountable, meaning that 'informations' is invalid.
This patch fixes this typo throughout the tree, except in CHANGES and
docs/news.html (historic text).
Signed-off-by: Thomas De Schampheleire <thomas.de.schampheleire@gmail.com>
Acked-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
When searching for virtual package providers, there's no need to
handle legacy symbols at all, so just bail out early.
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Cc: Samuel Martin <s.martin49@gmail.com>
Acked-by: Samuel Martin <s.martin49@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
No need to pass as argument to a function, members of the class it's in.
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Cc: Samuel Martin <s.martin49@gmail.com>
Acked-by: Samuel Martin <s.martin49@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
- typo in comment
- remove trailing space in _HOST_DEPENCENCIES when no dependency
Signed-off-by: Francois Perrad <francois.perrad@gadz.org>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
This commit improves the scancpan script to automatically populate the
LICENSE_FILES variable using informations available in the Perl
package MANIFEST file.
Signed-off-by: Francois Perrad <francois.perrad@gadz.org>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Perl extensions are loaded at runtime with dlopen(), so it does not
make sense to even build extensions that are written in C when
BR2_PREFER_STATIC_LIB is enabled. A Perl module written in C or with a
dependency on a module written in C is not available when doing a
static build.
Therefore, this commit adapts the scancpan script to automatically
generate a dependency on !BR2_PREFER_STATIC_LIB when the Perl module
would not work in a static-only configuration.
Signed-off-by: Francois Perrad <francois.perrad@gadz.org>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
- remove trailing space after perl when it's the only dependency
- license: substitution of perl name by BR name
- add a tabulation before source
Signed-off-by: Francois Perrad <francois.perrad@gadz.org>
Reviewed-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
This patch is the result of 2to3.
Signed-off-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Acked-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
Tested-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
It hides any error messages reported by make.
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
Reviewed-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Acked-by: Samuel Martin <s.martin49@gmail.com>
Tested-by: Samuel Martin <s.martin49@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
The graph-depends script calls make. If the outer make was called
recursively, or if it was called with '-C <somedir>', then the
environment will contain "MAKEFLAGS=w --". Therefore, the recursive
make prints 'Entering' and 'Leaving' messages, which clobbers the
output for dot.
To avoid this, add "--no-print-directory" to the recursive make
arguments. Since we require GNU make 3.81, we can be sure that this
option is available.
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
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>
This patch is the result of 2to3.
In addition, universal_newlines=True is added to the Popen calls. In
python3, this makes sure that the output is decoded so that we get a
string instead of a buffer object.
Cc: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
Reviewed-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
The transitive dependencies make the graphs barely readable for large
configs, with a large number of packages.
So, just switch to not drawing the transitive dependencies by default.
By popular demand... ;-)
[Peter: reword]
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Cc; Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Currently, the symlinks in the generated filesystems will have the
UID of the user running the build, because 'chown' does not change
the ownership of symlinks, by default.
Although the implications are limited, some may not want that UID
to leak in the generated filesystems.
So, use 'chown -h' so even symlinks get properly chowned.
Reported-by: Angelo Dureghello <angelo@barix.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>
Generate an asciidoc table that can be included in the manual, that
lists the existing virtual packages, the corresponding symbols, and
their providers (and sub-options thereof).
The core of this change is the addition of a new formatter for virtual
packages. This formatter is a bit tricky, as it has to catter for a
bunch of corner cases:
- provider is not a package, but is sub-options of a package
- such a sub-option may be itself 'select'-ed by one or more
other sub-options
- legacy packages should not be considered as a provider
Those cases are real:
- sub-options of mesa3d provide EGL or GLES
- selected sub-options of mesa3d provide GL
- udev is a legacy package, but it provides udev
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Cc: Samuel Martin <s.martin49@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Currently, we can generate two different tables of packages:
- a single-column table with the symbols' prompts,
- a two-column table with the symbols' prompts and locations in the
menuconfig.
For virtual packages, this is not enough, since we will have to display
more columns, with different content:
- the virtual package name (but such symbols do not have a prompt)
- the symbol name
- the providers for the virtual package
So, instead of having a single function that knows how to generate any
table, introduce a formatter function that is passed as argument to,
and called by format_asciidoc_table(). Such formatter functions are
responsible for providing:
- the layout of the table (number of columns, column arrangement),
- the formatted header line,
- a formatted line for a symbol.
What the formatter should ouput depends on its arguments:
- if none are passed, the layout is returned,
- if the header label is passed, it returns the formatted header line,
- otherwise, it returns the formatted line for a symbol.
Two formatter functions are introduced in this changeset, to replace the
current 'sub_menu' feature:
- _format_symbol_prompt() to display a one-column table with only the
symbols' prompts,
- _format_symbol_prompt_location() to display a two-column table with
the symbols' prompts and locations.
This will help us to later introduce a new formatter to generate a table
for virtual packages.
[Thanks to Samuel for his pythonistic help!]
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Cc: Samuel Martin <s.martin49@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
When generating the package lists, the responsibility to decide what is
actually a package symbol is currently split between the _is_package(),
the get_symbol_subset() and the format_asciidoc_table() functions.
The two latter functions check that an item is really a symbol, and that
is has a prompt.
While this is currently correct for real packages, this will no longer
be the case when we also generate a list of virtual packages, since they
do not have a prompt.
Move the responsibility to verify that a symbol is indeed a package symbol
to _is_package(), so it's all in one place, and makes it easier to change
for virtual packages.
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Cc: Samuel Martin <s.martin49@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
If a package has both a 'real' and a 'virtual' definition, consider it
is a virtual package and do not display it in the generated package list.
This is the case for jpeg and cryptodev, that are virtual packages, but
also real (but empty) packages used to provide a prompt to enable/disable
a choice to select an implementation. In this case, we do not want to
list the virtual packages, but only their implementations.
So, consider packages that are both real and virtual as virtual packages.
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Cc: Samuel Martin <s.martin49@gmail.com>
Acked-by: Samuel Martin <s.martin49@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Prepare to tell apart real packages from virtual packages.
Currently, the code implicitly recognises only real packages, and
discards virtual packages, because of the heuristic used to recognise
whether a symbol is a package:
- for real package:
- symbols : BR2_PACKAGE_FOO
- .mk files: foo.mk
- for virtual packages:
- symbols : BR2_PACKAGE_HAS_FOO
- .mk files: foo.mk
The current heuristic is to check for each symbol if a corresponding .mk
file exists, by stripping 'BR2_PACKAGE_' from the beginning of the symbol,
converting the result to lowercase, and checking if a .mk file exists.
So, as a side effect, it completely misses the virtual packages [*], which
is pretty nice since we get a list with only real packages that the user
can indeed select and see in the menuconfig.
[*] Except for 'cryptodev' and 'jpeg' which are both virtual packages and
normal packages. Except they are not normal packages, they are used to
display a choice of the implementation to use. This case will be fixed in
follow-up patches.
Since we'll soon need to also output the table of virtual packages, we
need to teach the _is_package() function to recognise them as well.
This patch is the first step into that direction: it introduces a new
function _is_real_package() that is just a wrapper to _is_package(), which
gains a new parameter, being the type of packages to filter on.
No behavioural change is made in this patch, it is just a preparatory
patch.
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Cc: Samuel Martin <s.martin49@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Cc: Samuel Martin <s.martin49@gmail.com>
Acked-by: Samuel Martin <s.martin49@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Move to a function the code generating the package name from a
symbol's name, to avoid code duplication.
This is not used currently, but will be in a subsequent patch.
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Cc: Samuel Martin <s.martin49@gmail.com>
Acked-by: Samuel Martin <s.martin49@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
This is ugly, since Python does not have enum constructs, so by moving
the 'type' of the constant ('MODE' here) to the beginning, we get an
artificial 'namespace' for the constants.
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Cc: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Cc: Maxime Hadjinlian <maxime.hadjinlian@gmail.com>
Cc: Samuel Martin <s.martin49@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Although unnecessary (we already have initialisation via the parser),
initialise the 'transitive' option, and document it at the same time.
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Cc: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Cc: Maxime Hadjinlian <maxime.hadjinlian@gmail.com>
Cc: Samuel Martin <s.martin49@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Add some comment as well, enhance help text.
[thanks to Samuel for the hints to make it even more pythonic]
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Cc: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Cc: Maxime Hadjinlian <maxime.hadjinlian@gmail.com>
Cc: Samuel Martin <s.martin49@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Merge the redundant-dependencies elimination into the newly introduced
transitive-dependencies elimination.
This makes the code cleaner and much shorter, because:
- the ('all',pkg) redundant dependency is in fact a transitive
dependency, and we now have code to deal with that
- the (pkg,'toolchain') dependency is easy enough to deal with that
having a separate function for that is overkill
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Cc: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Cc: Maxime Hadjinlian <maxime.hadjinlian@gmail.com>
Cc: Samuel Martin <s.martin49@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Currently, all the dependencies of a package are drawn on the dependency
graph, including transitive dependencies (e.g. A->B->C and A->C).
For very big graphs, with lots of packages with lots of dependencies, the
dependency graph can be very dense, and transitive dependencies are
cluttering the graph.
In some cases, only getting the "build-order" dependencies is enough (e.g.
to see what impact a package rebuild would have).
Add a new environment variable to disable drawing transitive dependencies.
Basically, it would turn this graph:
pkg1 ---> pkg2 ---> pkg3 -------------------.
|\__________/ \ \
|\____________________ \ \
| \ \ \
`-> pkg4 ---> pkg5 ---> pkg6 ---> pkg7 ---> pkg8
\__________/
into that graph:
pkg1 ---> pkg2 ---> pkg3 -----------.
| \
`-> pkg4 ---> pkg5 ---> pkg6 ---> pkg7 ---> pkg8
[Thanks to Samuel for the parser hints]
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Cc: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Cc: Maxime Hadjinlian <maxime.hadjinlian@gmail.com>
Cc: Samuel Martin <s.martin49@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Do not use the same colors for toolchain, host and target packages.
Signed-off-by: Francois Perrad <francois.perrad@gadz.org>
[yann.morin.1998@free.fr rephrase commit log]
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
in order to avoid spurious diff when updating packages
Signed-off-by: Francois Perrad <francois.perrad@gadz.org>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Remove some spaces before tabs and add the empty line at end of file.
Signed-off-by: Fabio Porcedda <fabio.porcedda@gmail.com>
Cc: Jerzy Grzegorek <jerzy.grzegorek@trzebnica.net>
Cc: Luca Ceresoli <luca@lucaceresoli.net>
Cc: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Acked-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
As Samuel said:
In Python, None is a singleton, and it is recommended to use "is" or
"is not" for testing them [1].
[1] http://legacy.python.org/dev/peps/pep-0008/#programming-recommendations
Reported-by: Samuel Martin <s.martin49@gmail.com>
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>
Some magic numbers obtained with trial-and-error and successive
iterations, to eventually get a nice graph.
[Thomas: remove excessive spaces in expressions.]
Reported-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Al packages depend on 'toolchain'. Currently, 'graph-depends' graphs this
dependency. The resulting graph is thus cluttered with less-than-useful
information.
Instead, do not graph the 'toolchain' dependency for any package, save
for the fake 'all' package. The graph is now a bit more readable.
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 complete dependency chain of a package is used to
generate the dependency graph. When this dependency chain is long,
the generated graph becomes almost unreadable.
However, it is often sufficient to get the first few levels of
dependency of a package.
Add a new variable BR2_GRAPH_DEPTH, that the user can set to limit
the depth of the dependency list.
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, we are using a crude, ad-hoc parsing of argv[].
This is a limiting factor to adding new options.
Use argparse instead, and introduce a single argument for now:
--package, -p PACKAGE
In the (near) future, we'll be able to add more option arguments,
such as depth-limiting for big graphs.
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>
Old toolchains, with old gcc that do not support -print-sysroot, break the
kernel-headers version check script: it fails to find the sysroot of the
toolchain, and thus ends up including the host's linux/version.h.
Most of the time, this will break early, since the host's kernel headers
will not match the toolchain settings.
But it can happen that the check is succesful, although the configuration
of the toolchain is wrong:
- the custom toolchain has kernel headers vX.Y
- the user selected vX.Z (Z!=Y)
- the host has headers vX.Y
In this case, the check passes OK, but the build of some packages later on
will break (which is exactly what those _AT_LEAST_XXX options were added to
avoid).
Fix that by passing the sysroot to the check script, instead of the cross
compiler.
We get the sysroot as thus:
- for custom toolchains, we use the macro toolchain_find_sysroot. We can
do that, because we already have a complete sysroot with libc.a at that
time.
- for internal toolchain using a custom kernel headers version, we just
use $(STAGING_DIR). We can't use the macro as for custom toolchains
above, because at the time we install the kernel headers, we do not yet
have a complete sysroot with a libc.a. But we can just use
$(STAGING_DIR), since we're only interested in the kernel headers.
For all other types of toolchains, we already have the _AT_LEAST_XXX options
properly set, so we need not add a check in this case.
Fixes:
http://autobuild.buildroot.net/results/f33/f331a6eff0b0b93c73af52db3a6b43e4e598577e/http://autobuild.buildroot.net/results/a57/a5797c025bec50c10efdcff74945aab4021d05e4/
[...]
[Thanks to Thomas for pointing out the toolchain_find_sysroot macro!]
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Cc: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Instead of creating a temporary files with a dubious scheme, use mktemp,
which purpose is exactly that: creating temporary files
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>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Since we introduced the _AT_LEAST_XXX for the kernel headers, people
using pre-built custom toolchain now have to specify the version of
the kernel headers their custom toolchain uses.
So, when we detect that there is a mismatch between the selection in
the menuconfig, and the actual version of the headers, we currently
only bail out with a terse message "Incorrect selection of kernel
headers".
This could be confusing some, and getting the version of the headers
used by the toolchain is not trivial (well, it's very easy, but not
trivial.)
This patch changes the way we report the error by moving the message
into the test-code, and by printing the expected and actual versions
of the kernel headers.
BUT! To get this pretty error message, we need to run the
test-program, so we can not use the cross-toolchain, we have to use
the native one.
BUT! The native one has its own linux/version.h header, so we can not
simply include it.
So, we ask the cross-compiler where its default sysroot is, and use
that to then force-feed the cross linux/version.h to the native
toolchain.
[Thomas: augment commit log with a message provided by Yann, fix
coding style to not have spaces after opening parenthesis and before
closing parenthesis, reformatted the message "Incorrect selection..."
to make it fit on one line.]
Reported-by: Maxime Hadjinlian <maxime.hadjinlian@gmail.com>
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Cc: Maxime Hadjinlian <maxime.hadjinlian@gmail.com>
Cc: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
When adding a new user (or a new group), we would get warnings, like:
[...]/support/scripts/mkusers: line 145: [: too many arguments
This is because we're checking if a UID (or a GID) is already defined,
and/or is different from the requested one, both checks in the same
test.
Of course, if a UID (or a GID) is not defined, it does not have a value,
so we can not compare it to an integer.
Fix that by splitting the test in two, so the second is only executed if
the first is sucessful.
Reported-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
This fixes the spurious "[: too many arguments" errors from mkusers.
Signed-off-by: Philip Paeps <philip@paeps.cx>
Acked-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Ensure the kernel headers version used in the custom external toolchain,
or the manually-specified kernel headers version, matches exactly the one
selected by the user.
We do not care about the patch-level, since headers are not supposed to
change between patchlevels. This applies only to kernels >= 3.0, but
those are actually the ones we do care about; we treat all 2.6.x kernels
as being a single version, since we do not support any 2.6 kernels for
packages with kernel-dependant features.
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Cc: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Cc: Arnout Vandecappelle <arnout@mind.be>
Cc: Thomas De Schampheleire <patrickdepinguin@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
To make the naming consistent (all user-visible options should be
prefixed with BR2_).
An entry is added to Makefile.legacy to warn users who have set
BUILDROOT_CONFIG but not BR2_CONFIG.
Still export BUILDROOT_CONFIG but pointing to some phony value, to
make sure that scripts that still use it fail in a predictable way.
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
Tested-by: Jérémy Rosen <jeremy.rosen@openwide.fr>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
scan_patchdir is called recursively. For this to work properly, the
variable path which is set to $1 at the very beginning must be local not
global.
A test case is to set BR2_GLOBAL_PATCH_DIR to 'mypatches' and having the
following tree in the buildroot root:
$ find mypatches/
mypatches/
mypatches/busybox
mypatches/busybox/subdir.patch
mypatches/busybox/subdir.patch/busybox-0001-abc.patch
mypatches/busybox/busybox-0002-def.patch
mypatches/busybox/asubdir.patch
mypatches/busybox/asubdir.patch/busybox-0003-xyz.patch
When running 'make busybox-dirclean busybox-patch' originally, you'd get:
Applying busybox-0003-xyz.patch using patch:
Applying busybox-0002-def.patch using patch:
Error: missing patch file
mypatches/busybox/asubdir.patch/busybox-0002-def.patch
While with this fix:
Applying busybox-0003-xyz.patch using patch:
Applying busybox-0002-def.patch using patch:
Applying busybox-0001-abc.patch using patch:
This fixes bug #6434 (https://bugs.busybox.net/show_bug.cgi?id=6434)
Signed-off-by: Daniel Mentz <daniel@exxm.de>
[Thomas: update commit message with test case]
Signed-off-by: Thomas De Schampheleire <thomas.de.schampheleire@gmail.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
the initial implementation assumes that when a version found in
buildroot is different from the one in the X11 release, it
requires an upgrade. even though this is most likely the case, it
could be a downgrade too, and it's probably worth highlighting
such cases when it (rarely) happens.
LooseVersion from distutils is doing the low level job of sorting
version numbers represented in strings...
[Thomas & Thomas:
- do not count packages more recent in Buildroot than in the latest
X.org release as to be downgraded. If we have more recent version,
it's generally for a good reason, so we want to keep them as
is. Such packages are counted as "nothing to do", but for
information, we indicate that there are "More recent"
- also remove the "nothing to do" action indicator. It used to be a
simple dash, which was not really useful.
]
Signed-off-by: Nicolas Dechesne <ndec13@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Signed-off-by: Thomas De Schampheleire <thomas.de.schampheleire@gmail.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
In order to keep better track of when a feature got deprecated, and hence
when it can be removed, a new set of symbols BR2_DEPRECATED_SINCE_xxxx_xx is
introduced. These symbols are automatically selected when BR2_DEPRECATED is
selected, and thus are transparent to the user.
A deprecated feature will no longer depend on BR2_DEPRECATED directly, but
rather on the appropriate BR2_DEPRECATED_SINCE_xxxx_xx. If that symbol does
not yet exist, it has to be created in Config.in.
When removing a deprecated feature, one should also check whether this was
the last feature using the BR2_DEPRECATED_SINCE_xxxx_xx symbol, in which
case the latter can be removed from Config.in.
A followup patch will make sure the overview is added to the list of
deprecated features in the manual, so that a buildroot core developer can
easily determine which features to remove in a given development cycle.
Signed-off-by: Thomas De Schampheleire <thomas.de.schampheleire@gmail.com>
Acked-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
PDF files can not be easily embedded in other documents (eg. ODT, or HTML).
Add support for generating PNG graphs, by setting the GRAPH_OUT=pdf|png on
the command line:
make GRAPH_OUT=png graph-build graph-depends
The default is still to generate PDF graphs.
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>
This script generates graphs of packages build time, from the timing
data generated by Buildroot in the $(O)/build-time.log file.
Example usage:
./support/scripts/graph-build-time \
--type=histogram --input=$(O)/build-time.log --output=foobar.pdf
Three graph types are available :
* histogram, which creates an histogram of the build time for each
package, decomposed by each step (extract, patch, configure,
etc.). The order in which the packages are shown is
configurable: by package name, by build order, or by duration
order. See the --order option.
* pie-packages, which creates a pie chart of the build time of
each package (without decomposition in steps). Packages that
contributed to less than 1% of the overall build time are all
grouped together in an "Other" entry.
* pie-steps, which creates a pie chart of the time spent globally
on each step (extract, patch, configure, etc...)
The default is to generate an histogram ordered by package name.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
[yann.morin.1998@free.fr: adapt to the format of the step-hooks build-time.log,
add sort order by name, default to name-ordered histogram, use our colours
for pie-charts, add alternate color-scheme, add short-options, add
--input/-i]
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
This patch fixes an issue that occurs during the manual build process
which will occur when BR2_EXTERNAL is introduced.
During the package list generation, the python script using kconfiglib
module reads and parses the Config.in files. So, symbols, including
environment variables, got expanded and/or resolved. In
kconfiglib.py, this patch fixes the regex that did not allow to use
numbers in the environment variable names, so '$BR2_EXTERNAL' got
wrongly expanded like it was '${BR}2_EXTERNAL':
<snip>
>>> Updating the manual lists...
Traceback (most recent call last):
File "/opt/buildroot/master/support/scripts/gen-manual-lists.py", line 375, in <module>
buildroot = Buildroot()
File "/opt/buildroot/master/support/scripts/gen-manual-lists.py", line 216, in __init__
self.root_config))
File "/opt/buildroot/master/support/scripts/kconfiglib.py", line 214, in __init__
self.top_block = self._parse_file(filename, None, None, None)
File "/opt/src/buildroot/master/support/scripts/kconfiglib.py", line 919, in _parse_file
return self._parse_block(line_feeder, None, parent, deps, visible_if_deps, res)
File "/opt/buildroot/master/support/scripts/kconfiglib.py", line 1114, in _parse_block
self.base_dir))
IOError: /opt/buildroot/master/Config.in:490: sourced file "$BR2_EXTERNAL/Config.in" (expands to
"2_EXTERNAL/Config.in") not found. Perhaps base_dir
(argument to Config.__init__(), currently
"/opt/buildroot/master") is set to the wrong value.
docs/manual/manual.mk:2: recipe for target 'manual-update-lists' failed
make: *** [manual-update-lists] Error 1
</snip>
Reported-by: Ryan Barnett <rjbarnet@rockwellcollins.com>
Signed-off-by: Samuel Martin <s.martin49@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Reviewed-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Tested-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Acked-by: Ryan Barnett <rjbarnet@rockwellcollins.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Python saves a pre-compiled support/scripts/kconfiglib.pyc file
side-to-side with the corresponding .py file.
This does not work if the Buildroot source tree is read-only (but
this is not an error for Python, which keep going OK).
But this may cause issues for out-of-tree builds in case the same
Buildroot source tree is shared by many builds.
Also, 'make clean' currently does not clean this file, and out-of-tree
builds can remove it either, at the risk of causing issues for other
out-of-tree builds running at the same time.
Just tell Python not to generate .pyc files:
- call the script via python, don't use the sha-bang
- thus, make the script non-executable, and remove the sha-bang
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Cc: Arnout Vandecappelle <arnout@mind.be>
Cc: Samuel Martin <s.martin49@gmail.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
This patch fixes typos in the 'encode_password' function calls.
Signed-off-by: Samuel Martin <s.martin49@gmail.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
For out-of-tree builds, this use-case fails to build:
$ make clean all
This is because 'all' is filtered-out in the Makefile wrapper, since
the wrapper itself has a 'all' target.
The 'all' target is just the usual naming for the default target in a
Makefile. In fact, the first target is the default one, so we can name
it whatever we want.
Rename the Makefile wrapper 'all' target to avoid name-clashing.
Fixes#6644.
Reported-by: Ryan Barnett <rjbarnet@rockwellcollins.com>
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Cc: Ryan Barnett <rjbarnet@rockwellcollins.com>
Cc: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Tested-by: Ryan Barnett <rjbarnet@rockwellcollins.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
[Thomas: added Thomas DS Acked-by, given at
http://patchwork.ozlabs.org/patch/284719/, and made the additional
typo fixes suggested by Thomas DS.]
Signed-off-by: Samuel Martin <s.martin49@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Acked-by: Thomas De Schampheleire <thomas.de.schampheleire@gmail.com>
Disable reversed/already applied patches fallout from commit
5871b79199
Reverse patches are bad, they may unfix things with version bumps and
just sneak under the radar with pure batch mode.
Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>
Fixes http://autobuild.buildroot.net/results/868/8687be8ec029486d9c5e2224cde542134f72884b/
The recent (d245fbb41d: apply-patches.sh: detect missing patches)
change to apply-patches.sh causes a number of regressions with packages
using downloadable tarballs of patches (typically from Debian), as
those contain additional files besides just the patches (ChangeLog's,
debian/rules, ..).
This use case is arguably abusing the _PATCH handling, but it used to
work so people might rely on it so go back to only warn about this
instead of erroring out.
At the same time reword the warning message.
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>
The "patch" command returns an error code only if patches fail
to apply. Therefore the pipleline "cat <patchfile> | patch ..."
does not fail, even if <patchfile> is missing. Fix this by
adding an explicit check for patch file existence.
Based on feedback from buildroot mailing list, also change the
existing check for unsupported patch format into a fatal error.
Signed-off-by: Ralph Siemsen <ralphs@netwinder.org>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>
In BR sub-directory boot/ linux/ and package/ there are a few .mk files which
aren't <package>.mk files. These files shouldn't be taken into account
in package statistics.
Signed-off-by: Jerzy Grzegorek <jerzy.grzegorek@trzebnica.net>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>
Brings a number of fixes, and most importantly no longer tries to figure
out if the tree contains uncommitted changes when using svn, as that can
be very slow.
This only syncs with setlocalversion as of 2.6.34 as later kernel versions
aren't directly compatible with our use cases since 09155120c (kbuild:
Clean up and speed up the localversion logic).
We still have one delta from the kernel version (setlocalversion: fix i18n
issue with svn), as that has only later been fixed in the kernel version.
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>
If the file to be patched is missing, then `patch' will interactively
ask for a file to be patched. This is annoying in e.g. the autobuilders
because they have to wait for a timeout instead of failing.
Giving the '-t' (batch mode) option to patch fixes this: it will skip the
missing file, and return a non-zero exit code. So the build cleanly
fails.
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
Acked-by: Luca Ceresoli <luca@lucaceresoli.net>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>
This finally removes the BR2_HAVE_DEVFILES option, that was used to
install/keep development files on target. With the recent migration of
the internal backend to the package infrastructure, we had anyway lost
the ability to build gcc for the target, and install the uClibc
development files on the target.
[Peter: also remove support/scripts/copy.sh]
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>
We ran into a "Login incorrect" problem when running the same rootfs
image across platforms with different loging ports ttyS0/1/2/3.
Simply assignning "console" to BR2_TARGET_GENERIC_GETTY_PORT, which in
turn modifies the /etc/inittab, is not enough because the "console" device
was missing in the /etc/securetty.
While current securetty has enumerated a lot of ttys, this patch should save
some efforts to enumerate more.
[Peter: guard with single quotes]
Signed-off-by: Tzu-Jung Lee <tjlee@ambarella.com>
Signed-off-by: Spenser Gilliland <spenser@gillilanding.com>
Acked-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>
Otherwise, graph-depends tries to call 'make target-purgelocales-show-depends',
which does not exist, as 'target-purgelocales' is not an actual package.
Signed-off-by: Danomi Manchego <danomimanchego123@gmail.com>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>
The graph-depends script tries to call 'make target-generic-dont-remount-rw',
which doesn't exist since 'target-generic-dont-remount-rw' is not a package.
See also the comments for commit 72bd61e5b8c2094378.
Signed-off-by: Luca Ceresoli <luca@lucaceresoli.net>
Acked-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>
Script generating the target and host package tables, and the deprecated
stuff list as well. These tables and lists are generated parsing the
Config.in files.
Signed-off-by: Samuel Martin <s.martin49@gmail.com>
[yann.morin.1998@free.fr: no leading dot, no menu path for host-utils]
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>
Signed-off-by: Samuel Martin <s.martin49@gmail.com>
[yann.morin.1998@free.fr: rename readme so it is obvious it's about kconfiglib]
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>
Packages that install daemons may need those daemons to run as a non-root,
or an otherwise non-system (eg. 'daemon'), user.
Add infrastructure for packages to create users, by declaring the FOO_USERS
variable that contain a makedev-syntax-like description of the user(s) to
add.
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Cc: Samuel Martin <s.martin49@gmail.com>
Cc: Cam Hutchison <camh@xdna.net>
Cc: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Acked-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>
Signed-off-by: Jerzy Grzegorek <jerzy.grzegorek@trzebnica.net>
[yann.morin.1998@free.fr: space-damage]
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>
Since the introduction of the post-image mechanism, the graph-depends
script is broken: it tries to call 'make
target-post-image-show-depends', which doesn't exist since
'target-post-image' is not a package.
So we should simply ignore this 'target-post-image'.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>
This mechanism of root filesystem customization has been deprecated
since a long time, so let's remove it now.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Acked-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
Acked-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Acked-by: Luca Ceresoli <luca@lucaceresoli.net>
Acked-by: "Samuel Martin" <s.martin49@gmail.com>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>
As requested by Peter, add a bit of documentation in the
eclipse-register-toolchain script, and add a few more checks (even
though this script is not intended to be executed manually, which is
also now mentionned in the documentation).
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>
The Eclipse plugin at
https://github.com/mbats/eclipse-buildroot-toolchain-plugin allows
users of Eclipse to easily use the toolchain available in
Buildroot. To do so, this plugin reads
~/.buildroot-eclipse.toolchains, which contains the list of Buildroot
toolchains available on the system, and then offer those toolchains to
compile Eclipse projects.
In order to interface with this plugin, this commit adds an option
that allows the user to tell whether (s)he wants the Buildroot project
toolchain to be visible under this Eclipse plugin. It simply adds a
line in this ~/.buildroot-eclipse.toolchains file.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>
Following Gustavo's removal of two X.org drivers for old hardware
unlikely to be used in embedded contexts, the xorg-release script now
reports those two X.org packages as "to be added": they exist in
X.org, but not in Buildroot.
So, we add a small list, XORG_EXCEPTIONS, in our xorg-release script,
to list the X.org packages we don't want to hear about. Of course,
packages that exist in X.org, and that are not part of this exception
list, and are not packaged in Buildroot are still listed as "to be
added".
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>
This script generates a report on the packaging status of X.org
releases in Buildroot. It does so by downloading the list of tarballs
that are part of a given X.org release, and compare that with the
packages that are available in Buildroot.
[Peter: drop .py suffix, make executable]
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>
Add the root-password internal target to the exclusion list.
Fixes failures like:
Getting dependencies for [... 'target-root-passwd' ...]
Error getting dependencies [... 'target-root-passwd' ...]
Which is easily singled out with:
$ make target-root-passwd-show-depends
make[1]: *** No rule to make target `target-root-passwd-show-depends'.
Signed-off-by: "Yann E. MORIN" <yann.morin.1998@free.fr>
Cc: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>
The "unknown" packages mechanism was used to render packages that did
not implement the make <pkg>-show-depends target, i.e the packages
that were not yet converted to one of the package infrastructures.
Since now all packages have been converted, we can remove this
"unknown" packages feature.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>
Since 9bc7b1d4ae, all X.org .mk files
are parsed unconditionally, even if BR2_PACKAGE_XORG7 is disabled.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>
Until now, graph-depends was calling "make <pkg>-show-depends"
individually for eack package, which was very slow. Now, it calls
"make <pkg1>-show-depends <pkg2>-show-depends ... <pkgN>-show-depends"
for all packages it knows, and then does that recursively. It reduces
the number of make invocations to the deepest dependency chain in the
current configuration, instead of having a number of make invocations
equal to the number of enabled packages.
For a configuration with xvkbd enabled (which brings a significant
number of X.org dependencies) and a tar root filesystem, the time to
execute graph-depends was:
real 5m14.944s
user 4m53.590s
sys 0m14.069s
After our optimizations, it is now:
real 0m33.096s
user 0m30.878s
sys 0m1.472s
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>
In preparation for more graph-depends improvements, use a
TARGET_EXCEPTIONS list to list all the targets that should be ignored
while building the dependency graph.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>
When doing a full graph of the dependencies, graph-depends starts by
doing a "make show-targets", which lists all the packages registered
in the $(TARGETS) variable. This variable contains all packages that
are enabled according to the .config file. Then, for each of those
packages, we used to create a "all" -> "package" dependency, even if
in fact most of some packages are already dependencies of other
packages. This creates a needlessly complex dependency graph.
This patch modifies graph-depends so that it filters out the unneeded
"all" -> "package" dependencies when "package" is already the
dependency of another package.
For example, if you have a configuration with libpng (which selects
zlib), "make show-targets" displays "libpng zlib", so graph-depends
used to create the following dependencies: (all -> libpng, all ->
zlib, libpng -> zlib). However, the (all -> zlib) dependency is not
really needed, as zlib is already the dependency of libpng. Those
dependencies are now filtered out.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Signed-off-by: Peter Korsgaard <jacmet@sunsite.dk>
