On Fri, Dec 9, 2022 at 5:48 PM Sean Christopherson <seanjc@xxxxxxxxxx> wrote: > > On Fri, Dec 09, 2022, David Matlack wrote: > > On Mon, Dec 05, 2022 at 11:14:29AM -0800, Vipin Sharma wrote: > > > Make ASSERT_EXIT_REASON() macro and replace all exit reason test assert > > > statements with it. > > > > > > No functional changes intended. > > > > > > Suggested-by: David Matlack <dmatlack@xxxxxxxxxx> > > > Signed-off-by: Vipin Sharma <vipinsh@xxxxxxxxxx> > > > > Reviewed-by: David Matlack <dmatlack@xxxxxxxxxx> > > > > > --- > > > .../testing/selftests/kvm/aarch64/psci_test.c | 4 +-- > > > .../testing/selftests/kvm/include/test_util.h | 10 ++++++++ > > [...] > > > .../selftests/kvm/x86_64/xapic_ipi_test.c | 6 +---- > > > .../selftests/kvm/x86_64/xen_shinfo_test.c | 7 +----- > > > .../selftests/kvm/x86_64/xen_vmcall_test.c | 5 +--- > > > 44 files changed, 71 insertions(+), 293 deletions(-) > > > > Nice diff stat :) > > I like the diffstat too, but I think we need a slightly different name for the > macro. "EXIT_REASON" can be interpreted as "hardware exit reason" or "KVM exit > (to userspace) reason". Most usage will be fairly obvious, but I'd like to avoid > confusion when swapping between selftests and KVM-unit-tests (which often asserts > on the hardware exit reason). The name will be a bit longer, but I don't think > line length with ever be a problem. And most people will just copy+paste so finger > fatigue likeliy won't be an issue either. > > I also vote to prefix it with TEST, i.e. TEST_ASSERT_KVM_EXIT_REASON(), for > consistency and to make it very clear it's a TEST_ASSERT() wrapper. ASSERT_EQ() > makes me twitch every time I see it. This is definitely a personal preference > (or problem) I can survive with though :-) I will send a patch with TEST_ASSERT_KVM_EXIT_REASON. No one should twitch while reading code!
