Download nds32 prebuild toolchain from github
which is base on gcc 8.0.1 version for regression.
Signed-off-by: Rick Chen <rick@andestech.com>
Cc: Greentime Hu <greentime@andestech.com>
Vexpress ca15_tc2 is failing with 3.1.0 because of QEMU issue.
When this patch is applied
https://patchwork.kernel.org/patch/10754401/
Vexpress can be also turn to newer QEMU version.
Signed-off-by: Michal Simek <michal.simek@xilinx.com>
Reviewed-by: Tom Rini <trini@konsulko.com>
This change enables setting up specific Qemu version or sha1 for new
targets which are added after (current) v3.0.0 version.
This changes is preparation step for adding new Xilinx Versal Virt
platform which was merge after v3.0.0.
Signed-off-by: Michal Simek <michal.simek@xilinx.com>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
Reviewed-by: Tom Rini <trini@konsulko.com>
The single job for all Layerscape 10xx platforms is close to, and
sometimes exceeds the time limit for a single job configuration. Break
this down into jobs for LS101x, LS104x and LS108x instead. While in
here, in the name portion of these jobs, refer to them as NXP for ARM
and not Freescale as they've been NXP for quite some time.
Signed-off-by: Tom Rini <trini@konsulko.com>
The motivation for this is to allow distributions to distribute all
possible tools in a generic way, avoiding the need of specific tools
building for each machine.
Especially on OpenEmbedded / Yocto Project ecosystem, it is very
common each BSP to end providing their specific tools when they need
to generate images for some SoC (e.g MX23 / MX28 in meta-freescale
case).
Using this, we can package the tools doing:
$: make tools-only_defconfig
$: make tools-only
Signed-off-by: Otavio Salvador <otavio@ossystems.com.br>
[trini: Add MAINTAINERS entry for myself, add to .travis.yml, make
U-Boot itself buildable to not trip up other frameworks]
Signed-off-by: Tom Rini <trini@konsulko.com>
The genboardscfg.py script will emit a WARNING message if we have new
defconfig files that are not listed in a MAINTAINERS file. Make new
cases of this a failure we catch in Travis-CI.
Signed-off-by: Tom Rini <trini@konsulko.com>
The latest version of pytest (4.0.0) makes some of the code we have in
test/py/conftest.py a fatal error that needs to be migrated.
Unfortunately this in turn requires changes that don't exist in older
versions of pytest such as 2.8.7 that ships with Ubuntu 16.04. Force
travis to use this older version of pytest.
Signed-off-by: Tom Rini <trini@konsulko.com>
- Split the AArch64 LS10xx and LS20xx builds into their own jobs, and
then exclude only ls1/ls2 from the catch-all. This moves the S32V234
job (and future i.MX8*) to the catch-all.
- Split spear out from arm926ejs and exclude freescale, not mx from that
job. The older Freescale i.MX boards are caught by the catch-all job
for Freescale but now we build the non-Freescale older i.MX platforms.
Signed-off-by: Tom Rini <trini@konsulko.com>
Add qemu-x86_64 to the list of targets we use for test.py runs.
Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
To make testing with clang support easier, add sandbox/clang-7
combination to our testing matrix. To facilitate this, switch to using
the "sources" method that the travis.yml file supports to list
additional repositories and add the official one for llvm-7. Due to
buildman not supporting using clang at this time add logic to manually
build a single sandbox configuration in the expected output directory so
that we can still invoke all of our tests.
Cc: Simon Glass <sjg@chromium.org>
Signed-off-by: Tom Rini <trini@konsulko.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
Currently this uses x86_64 version toolchain for x86 build in
travis-ci. Change it to i386 version to avoid updating the
buildman toolchain path every time when the toolchain version
number is changed, eg: from 7.3.0 to 8.1.0.
Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
At present the tests run one after the other using a single CPU. This is
not very efficient. Bring in the concurrencytest module and run the tests
concurrently, using one process for each CPU by default. A -P option
allows this to be overridden, which is necessary for code-coverage to
function correctly.
This requires fixing a few tests which are currently not fully
independent.
At some point we might consider doing this across all pytests in U-Boot.
There is a pytest version that supports specifying the number of processes
to use, but it did not work for me.
Signed-off-by: Simon Glass <sjg@chromium.org>
Add support for compressing blob entries. This can help reduce image sizes
for many types of data. It requires that the firmware be able to
decompress the data at run-time.
Signed-off-by: Simon Glass <sjg@chromium.org>
Travis CI now supports giving jobs an explicit name. Do this for all jobs.
This allows more direct control over jobs names than the previous
automatic or implicit naming based on the environment variables or script
text.
Signed-off-by: Stephen Warren <swarren@nvidia.com>
[trini: Update names for jobs added/changed since posting]
Signed-off-by: Tom Rini <trini@konsulko.com>
The openrd platforms are currently orphaned, and are constantly on-edge
or overflowing their binary limit. Exclude them from travis for now.
Cc: Vagrant Cascadian <vagrant@debian.org>
Cc: Chris Packham <judge.packham@gmail.com>
Signed-off-by: Tom Rini <trini@konsulko.com>
The various Aries Embedded boards have been orphaned for a year and no
one has come forward to take care of them. Remove.
Signed-off-by: Tom Rini <trini@konsulko.com>
Fix riscv: ax25-ae350 build fail problem
https://travis-ci.org/trini/u-boot/jobs/385147373
...
Building current source for 1 boards (1 thread, 2 jobs per thread)
riscv: + ax25-ae350
+arch/riscv/cpu/ax25/start.S: Assembler messages:
+arch/riscv/cpu/ax25/start.S:48: Error: unrecognized opcode `sd a2,0(t0)'
+arch/riscv/cpu/ax25/start.S:112: Error: unrecognized opcode `ld t5,0(t0)'
...
After apply the commit
configs: ax25-ae350: Set 64-bit as default configuration
Toolchain shall be also setuped with 64-bit in .travis.yml.
Signed-off-by: Rick Chen <rick@andestech.com>
Signed-off-by: Rick Chen <rickchen36@gmail.com>
Reviewed-by: Chih-Mao Chen <cmchen@andestech.com>
Cc: Greentime Hu <green.hu@gmail.com>
- Xilinx aarch64 is caught in the general xilinx arm job, exclude from
the general aarch64 job.
- Give the generic aarch64 job a better name
- Re-sort the PowerPC jobs so that we can complete them a bit quicker.
Signed-off-by: Tom Rini <trini@konsulko.com>
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 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>
This allows running tests on emulated KC705 board with DC233C xtensa
core. It expects to find conf.xtfpga_qemu in the uboot-test-hooks.
Signed-off-by: Max Filippov <jcmvbkbc@gmail.com>
xtensa toolchains are core-specific, so give full toolchain name and
download corresponding prebuilt toolchain from the github release.
Signed-off-by: Max Filippov <jcmvbkbc@gmail.com>
The build script should not manipulate shell flags (especially '-e').
A non-zero exit value can also be catched with 'cmd || ret=$?'.
Signed-off-by: Daniel Schwierzeck <daniel.schwierzeck@gmail.com>
Reviewed-by: Tom Rini <trini@konsulko.com>
The corresponding changes in the uboot-test-hooks repo are:
https://github.com/swarren/uboot-test-hooks/pull/15
Signed-off-by: Tuomas Tynkkynen <tuomas.tynkkynen@iki.fi>
Reviewed-by: Tom Rini <trini@konsulko.com>
Enable travis-ci support with a link having built.
Signed-off-by: Chih-Mao Chen <cmchen@andestech.com>
Signed-off-by: Rick Chen <rick@andestech.com>
Signed-off-by: Rick Chen <rickchen36@gmail.com>
Signed-off-by: Greentime Hu <green.hu@gmail.com>
The 't208xrdb t4qds t102*' job is close to the time limit and
sometimes fails, so this splits it into 3 separate jobs.
Signed-off-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
Given how we handle the ARM toolchain we can't easily combine these two
jobs, so don't. Give xilinx/ARM a separate build.
Signed-off-by: Tom Rini <trini@konsulko.com>
- Move SoCFPGA and K2 boards to their own job
- Expand the microblaze job to cover ARM boards from Xilinx as well.
Signed-off-by: Tom Rini <trini@konsulko.com>
This makes us act like the Linux Kernel does and allow for dtc to be
provided externally but otherwise we use the version of dtc that is
included in the sources. This in turn means that we can drop the
checkdtc logic. We select DTC in the cases where we will need the dtc
tool provided.
Signed-off-by: Tom Rini <trini@konsulko.com>
In a0f3e3df4a we switched to using the Ubuntu-provided dtc as travis
was having a problem with the number of warnings that were generated by
the newer dtc. This is no longer a concern as we now have the same
logic as Linux to enable/disable additional more stringent warnings. Go
back to building dtc from source.
Signed-off-by: Tom Rini <trini@konsulko.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
Tested on travis-ci:
Tested-by: Simon Glass <sjg@chromium.org>
