The 'done' files created by buildman may end up being empty if buildman
runs out of disk space while writing them. This error is then persistent,
since even if disk space is reclaimed and the build retries, the empty
file causes an exception in the builder thread.
Deal with this silently by doing a rebuild.
Signed-off-by: Simon Glass <sjg@chromium.org>
Unfortunately, for some releases the kernel.org toolchain tarball names adhere
to the following pattern:
<hostarch>-gcc-<ver>-nolib-<targetarch>-<type>.tar.xz
e.g.:
x86_64-gcc-8.1.0-nolibc-aarch64-linux.tar.xz
while others use the following pattern:
<hostarch>-gcc-<ver>-nolib_<targetarch>-<type>.tar.xz
e.g.:
x86_64-gcc-7.3.0-nolibc_aarch64-linux.tar.xz
Notice that the first pattern has dashes throughout, while the second has
dashes throughout except just before the target architecture which has an
underscore.
The "dash throughout" versions from kernel.org are:
8.1.0, 6.4.0, 5.5.0, 4.9.4, 4.8.5, 4.6.1
while the "dash and underscore" versions from kernel.org are:
7.3.0, 4.9.0, 4.8.0, 4.7.3, 4.6.3, 4.6.2, 4.5.1, 4.2.4
This tweak allows the code to handle both versions. Note that this tweak also
causes the architecture parsing to get confused and find the following two
bogus architectures, "2.0" and "64", which are explicitly checked for, and
removed.
Signed-off-by: Trevor Woerner <trevor@toganlabs.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
Change single quotes to double quotes:
Signed-off-by: Simon Glass <sjg@chromium.org>
The hexagon toolchain (4.6.1) from kernel.org, for example, was packaged in
a way that is different from most toolchains. The first entry when unpacking
most toolchain tarballs is:
gcc-<version>-nolib/<targetarch>-<system>
e.g.:
gcc-8.1.0-nolibc/aarch64-linux/
The first entry of the hexagon toolchain, however, is:
gcc-4.6.1-nolibc/
This causes the buildman logic in toolchain.py::ScanPath() to not be able to
find the "*gcc" executable since it looks in gcc-4.6.1-nolib/{.|bin|usr/bin}
instead of gcc-4.6.1/hexagon-linux/{.|bin|usr/bin}. Therefore when buildman
tries to download a set of toolchains that includes hexagon, the script fails.
This update takes the second line of the tarball unpacking (which works for
all the toolchains I've tested from kernel.org) and parses it to take the
first two elements, separated by '/'. It makes this logic a bit more robust.
Signed-off-by: Trevor Woerner <trevor@toganlabs.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
At present we should boards with warnings in the same way as those with
errors. This is not ideal. Add a new 'warn' state and show these listed
in yellow to match the actual warning lines printing with -e.
Signed-off-by: Simon Glass <sjg@chromium.org>
At present we don't distinguish between errors and warnings when printing
the architecture summary. Rename the variables to better describe their
purpose.
'Worse' at present means we got an error, so use that as the name.
Signed-off-by: Simon Glass <sjg@chromium.org>
At present messages from the device-tree compiler like this:
arch/arm/dts/socfpga_arria10_socdk_sdmmc.dtb: Warning
(avoid_unnecessary_addr_size): /clocks: unnecessary
#address-cells/#size-cells without "ranges" or child "reg" property
are detected as errors since they don't match the gcc warning regex. Add a
new one for dtc to fix this.
Signed-off-by: Simon Glass <sjg@chromium.org>
At present --list-tool-chains prints a lot of information about the
toolchain-probing process. This is generally not very interesting.
Update buildman to print this only if --list-tool-chains is given
with -v.
Signed-off-by: Simon Glass <sjg@chromium.org>
At present 'buildman sandbox' will build all 5 boards for the sandbox
architecture rather than the single board 'sandbox'. The only current way
to exclude sandbox_spl, sandbox_noblk, etc. is to use -x which is a bit
clumbsy.
Add a --boards option to allow individual build targets to be specified.
Signed-off-by: Simon Glass <sjg@chromium.org>
There are a few test cases which print output. Suppress this so that tests
can run silently in the normal case.
Signed-off-by: Simon Glass <sjg@chromium.org>
The filenames of the toolchains on kernel.org changes every now and then.
Fix it for the current change, and make the test use a regex so that it
has a better chance of passing with future changes too.
Signed-off-by: Simon Glass <sjg@chromium.org>
Something has changed in the last several month such that when buildman
builds U-Boot incrementally and a new CONFIG option has been added to the
Kconfig, the build hanges waiting for input:
Test new config (NEW_CONFIG) [N/y/?] (NEW)
Since binamn does not connect the build's stdin to anything this waits on
stdin to the build thread, which never comes. Eventually I suspect all the
threads end up in this state and the build does not progress.
Fix this by passing /dev/null as input to the build. That way, if there is
a new CONFIG, the build will stop (and fail):
Test new config (NEW_CONFIG) [N/y/?] (NEW)
Error in reading or end of file.
Signed-off-by: Simon Glass <sjg@chromium.org>
You do not need to use the typedefs provided by compiler.
Our compilers are either IPL32 or LP64. Hence, U-Boot can/should
always use int-ll64.h typedefs like Linux kernel, whatever the
typedefs the compiler internally uses.
Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
When summarising the builds, add the -U option to emit delta lines for
the default environment built into U-Boot at each commit.
Signed-off-by: Alex Kiernan <alex.kiernan@gmail.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
As we're building the boards, extract the default U-Boot environment to
uboot.env so we can interrogate it later.
Signed-off-by: Alex Kiernan <alex.kiernan@gmail.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
Add support for gcc versions 7.3.0, 6.4.0 and 4.9.4.
Also use a regex for matching the tarball names. Some gcc versions
use '-ARCH-' instead of '_ARCH-'.
As part of this, we switch TravisCI to also using these toolchains for
all platforms.
Signed-off-by: Daniel Schwierzeck <daniel.schwierzeck@gmail.com>
Signed-off-by: Tom Rini <trini@konsulko.com>
When the U-Boot base directory happens to have the same name as the branch
that buildman is directed to use via the '-b' option and no output
directory is specified with '-o', buildman happily starts removing the
whole U-Boot sources eventually only stopped with the error message:
OSError: [Errno 20] Not a directory: '../<branch-name>/boards.cfg
Add a check to avoid this and also deal with the case where '-o' points
to the source directory, or any subdirectory of it.
Finally, tidy up the confusing logic for removing the old tree when using
-b. This is only done when building a branch.
Signed-off-by: Lothar Waßmann <LW@KARO-electronics.de>
Signed-off-by: Simon Glass <sjg@chromium.org>
Tested-by: Lothar Waßmann <LW@KARO-electronics.de>
When U-Boot started using SPDX tags we were among the early adopters and
there weren't a lot of other examples to borrow from. So we picked the
area of the file that usually had a full license text and replaced it
with an appropriate SPDX-License-Identifier: entry. Since then, the
Linux Kernel has adopted SPDX tags and they place it as the very first
line in a file (except where shebangs are used, then it's second line)
and with slightly different comment styles than us.
In part due to community overlap, in part due to better tag visibility
and in part for other minor reasons, switch over to that style.
This commit changes all instances where we have a single declared
license in the tag as both the before and after are identical in tag
contents. There's also a few places where I found we did not have a tag
and have introduced one.
Signed-off-by: Tom Rini <trini@konsulko.com>
Add a new option '-E' for treating all compiler warnings as errors.
Eventually this will pass 'KCFLAGS=-Werror' to Kbuild.
Signed-off-by: Daniel Schwierzeck <daniel.schwierzeck@gmail.com>
All of these host tools are apparently written for Python2,
not Python3.
Use 'python2' in the shebang line according to PEP 394
(https://www.python.org/dev/peps/pep-0394/).
Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
In some cases when "more" is told to page a given file it will prepend
the output with:
::::::::::::::
/PATH/TO/THE/FILE
::::::::::::::
And when this happens the output will not match the expected length.
Further, if we use a different pager we will instead fail the coverage
tests as we will not have 100% coverage. Update the help test to remove
the string in question.
Reviewed-by: Simon Glass <sjg@chromium.org>
Signed-off-by: Tom Rini <trini@konsulko.com>
The tests were broken by two separate commits which adjusted the output
when boards are listed. Fix this by adding back a PowerPC board and
putting the name of each board in the test.
Fixes: b9f7d881 (powerpc, 5xx: remove some "5xx" remains)
Fixes: 8d7523c5 (buildman: Allow showing the list of boards with -n)
Signed-off-by: Simon Glass <sjg@chromium.org>
Accessing the network slows down the test and limits the environment in
which it can be run. Add an option to disable network tests.
Signed-off-by: Simon Glass <sjg@chromium.org>
Corresponds to 375506d (File writing nit) from upstream
(https://github.com/ulfalizer/Kconfiglib).
Adds proper 'imply' support and fixes a few minor issues, one of which
previously triggered the following weird warning:
configs/taurus_defconfig: /tmp/tmpisI45S:6: warning: assignment to SPL_LDSCRIPT changes mode of containing choice from "arch/$(ARCH)/cpu/u-boot-spl.lds" to "y"
The change in 8639f69 (genconfig.py: Print defconfig next to warnings)
was reapplied.
tools/moveconfig.py previously depended on a hack that merged 'select's
with 'imply's. It was modified to look at the union of
Symbol.get_selected_symbols() and Symbol.get_implied_symbols(), which
should give the same behavior.
tools/genboardscfg.py was verified to produce identical board.cfg's
before and after the change.
Signed-off-by: Ulf Magnusson <ulfalizer@gmail.com>
At present we sometimes see warnings of the form:
/tmp/tmpMA89kB:36: warning: overriding the value of CMD_SPL.
Old value: "y", new value: "y".
This is not very useful as it does not show whch defconfig file it relates
to. Update the tool to show this.
Signed-off-by: Simon Glass <sjg@chromium.org>
AVR32 is gone. It's already more than two years for no support in Buildroot,
even longer there is no support in GCC (last version is heavily patched 4.2.4).
Linux kernel v4.12 got rid of it (and v4.11 didn't build successfully).
There is no good point to keep this support in U-Boot either.
Reviewed-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Tom Rini <trini@konsulko.com>
Reviewed-by: Heiko Schocher <hs@denx.de>
Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
The build output can still produce unicode encoded output. But in
the buildman's log and err files we only want plain ASCII characters.
To handle all situations with unicode and non-unicode output, encode
the stdout and stderr strings to UTF-8 and afterwards to ASCII with
replacing all special characters.
Signed-off-by: Daniel Schwierzeck <daniel.schwierzeck@gmail.com>
Build tools like Make, gcc or binutils support localized output
or unicode encoded output dependent on the default system locale.
This is not useful for buildman, where we want reproducible
warning or error messages or where the output of binutils is
further processed.
Signed-off-by: Daniel Schwierzeck <daniel.schwierzeck@gmail.com>
In the case where a new build only decreases sizes and does not increase
any size we still want to report what functions have been dropped when
doing a bloat comparison.
Cc: Simon Glass <sjg@chromium.org>
Signed-off-by: Tom Rini <trini@konsulko.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
Many toolchains for ARM use the 'gnueabihf' suffix rather than just
'gnueabi', so allow these to be used, but with a lower priority than
'gnueabi' ones.
Cc: Simon Glass <sjg@chromium.org>
Signed-off-by: Tom Rini <trini@konsulko.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
When writing out some of our results we may now have UTF-8 characters
in there as well. Translate these to latin-1 and ignore any errors (as
this is for diagnostic and given the githash anything else can be
reconstructed by the user.
Reviewed-by: Simon Glass <sjg@chromium.org>
Signed-off-by: Tom Rini <trini@konsulko.com>
One of these has crept in in this commit:
40a808f1 ARCv2: SLC: Make sure busy bit is set properly on SLC flushing
Adjust buildman to handle it.
Signed-off-by: Simon Glass <sjg@chromium.org>
Currently upstream does not yet understand the imply keyword. For what
we use kconfiglib.py for today, this is OK. We only need to be able to
evaluate in order to make boards.cfg and none of those choices will
depend on how imply evaluates out.
Signed-off-by: Tom Rini <trini@konsulko.com>
As well as showing the number of boards, allow showing the actual list of
boards that would be built, if -v is provided.
Signed-off-by: Simon Glass <sjg@chromium.org>
To troubleshoot unexpected bhavior during building and what's more
important during execution it is strongly recommended to use recent
ARC toolchain, and so we're now referring to arc-2016.09 which is the
latest as of today.
Signed-off-by: Alexey Brodkin <abrodkin@synopsys.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
When using #define CONFIG_SOME_OPTION, the value it set to '1'. When using
defconfig (i.e. CONFIG_SOME_OPTION=y) the value is set to 'y'. This results
in differences showing up with -K. These differences are seldom useful.
Adjust buildman to suppress these differences by default.
Signed-off-by: Simon Glass <sjg@chromium.org>
Normally buildman does a full build of a board. This includes creating the
u-boot.cfg file which contains all the configuration options. Buildman uses
this file with the -K option, to show differences in effective configuration
for each commit.
Doing a full build of U-Boot just to create the u-boot.cfg file is wasteful.
Add a -D option which causes buildman to only create the configuration. This
is enough to support use of -K and can be done much more quickly (typically
5-10 times faster).
Signed-off-by: Simon Glass <sjg@chromium.org>
The README for buildman says that we can use any field in boards.cfg to
decide what to build. However, we were not saving the options field
correctly.
Cc: Simon Glass <sjg@chromium.org>
Signed-off-by: Tom Rini <trini@konsulko.com>
Acked-by: Simon Glass <sjg@chromium.org>
Now we can use compiler wrapper such as ccache or distcc for buildman.
Signed-off-by: York Sun <york.sun@nxp.com>
CC: Simon Glass <sjg@chromium.org>
Acked-by: Simon Glass <sjg@chromium.org>
Signed-off-by: York Sun <york.sun@nxp.com>
CC: Simon Glass <sjg@chromium.org>
Fixed commit subject:
Signed-off-by: Simon Glass <sjg@chromium.org>
Acked-by: Simon Glass <sjg@chromium.org>
When Ctrl-C is pressed, just exited quietly. There is no sense in displaying
a stack trace since buildman will always be in the same place: waiting for
threads to complete building all the jobs on the queue.
Signed-off-by: Simon Glass <sjg@chromium.org>
It is annoying that buildman does not respond cleanly to Ctrl-C or SIGINT,
particularly on machines with lots of CPUS. Unfortunately queue.join()
blocks the main thread and does not allow it to see the signal. Use a
separate thread instead,
Signed-off-by: Simon Glass <sjg@chromium.org>
