On Wed, Jun 08, 2022 at 11:20:06PM +0000, Sean Christopherson wrote:
> On Wed, Jun 08, 2022, Marc Zyngier wrote:
> > On 2022-06-07 16:27, Paolo Bonzini wrote:
> > > Marc, Christian, Anup, can you please give this a go?
> >
> > Can you please, pretty please, once and for all, kill that alias you
> > seem to have for me and email me on an address I actually can read?
> >
> > I can't remember how many times you emailed me on my ex @arm.com address
> > over the past 2+years...
> >
> > The same thing probably applies to Sean, btw.
>
> Ha! I was wondering how my old @intel address snuck in...
>
> On the aarch64 side, with the following tweaks, courtesy of Raghu, all tests
> pass. I'll work these into the next version, and hopefully also learn how to
> run on aarch64 myself...
>
> Note, the i => 0 "fix" in test_v3_typer_accesses() is a direct revert of patch 3,
> "KVM: selftests: Fix typo in vgic_init test". I'll just drop that patch unless
> someone figures out why doing the right thing causes the test to fail.
CCing Eric for that one.
>
> diff --git a/tools/testing/selftests/kvm/aarch64/vgic_init.c b/tools/testing/selftests/kvm/aarch64/vgic_init.c
> index b91ea02a8a80..66b7e9c76370 100644
> --- a/tools/testing/selftests/kvm/aarch64/vgic_init.c
> +++ b/tools/testing/selftests/kvm/aarch64/vgic_init.c
> @@ -317,7 +317,7 @@ static void test_vgic_then_vcpus(uint32_t gic_dev_type)
>
> /* Add the rest of the VCPUs */
> for (i = 1; i < NR_VCPUS; ++i)
> - vm_vcpu_add(v.vm, i, guest_code);
> + vcpus[i] = vm_vcpu_add(v.vm, i, guest_code);
>
> ret = run_vcpu(vcpus[3]);
> TEST_ASSERT(ret == -EINVAL, "dist/rdist overlap detected on 1st vcpu run");
> @@ -424,7 +424,7 @@ static void test_v3_typer_accesses(void)
> KVM_DEV_ARM_VGIC_CTRL_INIT, NULL);
>
> for (i = 0; i < NR_VCPUS ; i++) {
> - ret = v3_redist_reg_get(v.gic_fd, i, GICR_TYPER, &val);
> + ret = v3_redist_reg_get(v.gic_fd, 0, GICR_TYPER, &val);
> TEST_ASSERT(!ret && !val, "read GICR_TYPER before rdist region setting");
> }
>
> @@ -654,11 +654,12 @@ static void test_v3_its_region(void)
> */
> int test_kvm_device(uint32_t gic_dev_type)
> {
> + struct kvm_vcpu *vcpus[NR_VCPUS];
> struct vm_gic v;
> uint32_t other;
> int ret;
>
> - v.vm = vm_create_with_vcpus(NR_VCPUS, guest_code, NULL);
> + v.vm = vm_create_with_vcpus(NR_VCPUS, guest_code, vcpus);
>
> /* try to create a non existing KVM device */
> ret = __kvm_test_create_device(v.vm, 0);
> diff --git a/tools/testing/selftests/kvm/aarch64/get-reg-list.c b/tools/testing/selftests/kvm/aarch64/get-reg-list.c
> index b3116c151d1c..17f7ef975d5c 100644
> --- a/tools/testing/selftests/kvm/aarch64/get-reg-list.c
> +++ b/tools/testing/selftests/kvm/aarch64/get-reg-list.c
> @@ -419,7 +419,7 @@ static void run_test(struct vcpu_config *c)
>
> check_supported(c);
>
> - vm = vm_create_barebones();
> + vm = vm_create(1);
Hmm, looks like something, somewhere for AArch64 needs improving to avoid
strangeness like this. I'll look into it after we get this series merged.
> prepare_vcpu_init(c, &init);
> vcpu = aarch64_vcpu_add(vm, 0, &init, NULL);
> finalize_vcpu(vcpu, c);
>
Thanks,
drew
