| Message ID | 20210604155312.15902-9-alex.bennee@linaro.org |
|---|---|
| State | New |
| Headers | show |
| Series | arm tcg/kvm refactor and split with kvm only support | expand |
On 6/4/21 8:51 AM, Alex Bennée wrote: > if (g_str_equal(qtest_get_arch(), "ppc64") && > (access("/sys/module/kvm_hv", F_OK) || > - access("/dev/kvm", R_OK | W_OK))) { > + access("/dev/kvm", R_OK | W_OK) || !qtest_has_accel("kvm"))) { > g_test_message("Skipping test: kvm_hv not available"); > return g_test_run(); > } > @@ -1398,7 +1398,7 @@ int main(int argc, char **argv) > */ > if (g_str_equal(qtest_get_arch(), "s390x")) { > #if defined(HOST_S390X) > - if (access("/dev/kvm", R_OK | W_OK)) { > + if (access("/dev/kvm", R_OK | W_OK) || !qtest_has_accel("kvm")) { > g_test_message("Skipping test: kvm not available"); I would have sorted the kvm test first. For s390x, we has the HOST test, but ppc doesn't. So we're doing the access() on any host, e.g. x86_64, where kvm cannot true for this test. r~
On 04/06/2021 17.51, Alex Bennée wrote: > From: Philippe Mathieu-Daudé <philmd@redhat.com> > > We might have a s390x/ppc64 QEMU binary built without the KVM > accelerator (configured with --disable-kvm). > Checking for /dev/kvm accessibility isn't enough, also check for the > accelerator in the binary. > > Reviewed-by: David Gibson <david@gibson.dropbear.id.au> > Reviewed-by: Greg Kurz <groug@kaod.org> > Reviewed-by: Cornelia Huck <cohuck@redhat.com> > Reviewed-by: Alex Bennée <alex.bennee@linaro.org> > Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com> > Signed-off-by: Alex Bennée <alex.bennee@linaro.org> > Message-Id: <20210505125806.1263441-9-philmd@redhat.com> > --- > tests/qtest/migration-test.c | 4 ++-- > 1 file changed, 2 insertions(+), 2 deletions(-) Reviewed-by: Thomas Huth <thuth@redhat.com>
On 6/4/21 11:11 PM, Richard Henderson wrote: > On 6/4/21 8:51 AM, Alex Bennée wrote: >> if (g_str_equal(qtest_get_arch(), "ppc64") && >> (access("/sys/module/kvm_hv", F_OK) || >> - access("/dev/kvm", R_OK | W_OK))) { >> + access("/dev/kvm", R_OK | W_OK) || !qtest_has_accel("kvm"))) { >> g_test_message("Skipping test: kvm_hv not available"); >> return g_test_run(); >> } >> @@ -1398,7 +1398,7 @@ int main(int argc, char **argv) >> */ >> if (g_str_equal(qtest_get_arch(), "s390x")) { >> #if defined(HOST_S390X) >> - if (access("/dev/kvm", R_OK | W_OK)) { >> + if (access("/dev/kvm", R_OK | W_OK) || >> !qtest_has_accel("kvm")) { >> g_test_message("Skipping test: kvm not available"); > > I would have sorted the kvm test first. access() is a simple syscall from the qtest, while qtest_has_accel() spawn a whole QEMU process to exec the QMP request. > For s390x, we has the HOST test, but ppc doesn't. So we're doing the > access() on any host, e.g. x86_64, where kvm cannot true for this test. Hmm I suppose the issue you described predate this patch?
diff --git a/tests/qtest/migration-test.c b/tests/qtest/migration-test.c index 2b028df687..102bc36b91 100644 --- a/tests/qtest/migration-test.c +++ b/tests/qtest/migration-test.c @@ -1387,7 +1387,7 @@ int main(int argc, char **argv) */ if (g_str_equal(qtest_get_arch(), "ppc64") && (access("/sys/module/kvm_hv", F_OK) || - access("/dev/kvm", R_OK | W_OK))) { + access("/dev/kvm", R_OK | W_OK) || !qtest_has_accel("kvm"))) { g_test_message("Skipping test: kvm_hv not available"); return g_test_run(); } @@ -1398,7 +1398,7 @@ int main(int argc, char **argv) */ if (g_str_equal(qtest_get_arch(), "s390x")) { #if defined(HOST_S390X) - if (access("/dev/kvm", R_OK | W_OK)) { + if (access("/dev/kvm", R_OK | W_OK) || !qtest_has_accel("kvm")) { g_test_message("Skipping test: kvm not available"); return g_test_run(); }