On Wed, Jan 24, 2024 at 10:45:29AM +0100, Thomas Huth wrote: > On 24/01/2024 08.18, Andrew Jones wrote: > > Add build/run tests for riscv64. We would also add riscv32, but Fedora > > doesn't package what we need for that. > > > > Signed-off-by: Andrew Jones <andrew.jones@xxxxxxxxx> > > --- > > .gitlab-ci.yml | 16 ++++++++++++++++ > > 1 file changed, 16 insertions(+) > > > > diff --git a/.gitlab-ci.yml b/.gitlab-ci.yml > > index 273ec9a7224b..f3ec551a50f2 100644 > > --- a/.gitlab-ci.yml > > +++ b/.gitlab-ci.yml > > @@ -87,6 +87,22 @@ build-ppc64le: > > | tee results.txt > > - if grep -q FAIL results.txt ; then exit 1 ; fi > > +# build-riscv32: > > +# Fedora doesn't package a riscv32 compiler for QEMU. Oh, well. > > + > > +build-riscv64: > > + extends: .intree_template > > + script: > > + - dnf install -y qemu-system-riscv gcc-riscv64-linux-gnu > > + - ./configure --arch=riscv64 --cross-prefix=riscv64-linux-gnu- > > + - make -j2 > > + - printf "FOO=foo\nBAR=bar\nBAZ=baz\nMVENDORID=0\n" >test-env > > + - ACCEL=tcg KVM_UNIT_TESTS_ENV=test-env ./run_tests.sh > > + selftest > > + sbi > > + | tee results.txt > > + - if grep -q FAIL results.txt ; then exit 1 ; fi > > IIRC it's better to make sure that at least one test passed: > > - grep -q PASS results.txt && ! grep -q FAIL results.txt > > Otherwise all tests could be SKIP which indicates that something went wrong, > too. Yeah, for sure the basic selftests should pass. > We're using the check for PASS for some tests in the gitlab-ci.yml file, but > not for all ... we should maybe update the remaining ones to use that, > too... Yup, almost all of them are the FAIL pattern which I copy+pasted. If I respin this series I'll change riscv now, but since I think we should change the other ones in a separate patch, then we could change riscv at the same time as the rest. Thanks, drew
