Date:   Thu, 16 Mar 2023 00:30:57 +0000
In-Reply-To: <cover.1678926164.git.ackerleytng@google.com>
Mime-Version: 1.0
References: <cover.1678926164.git.ackerleytng@google.com>
Message-ID: <a80f3bcce5b5612af9ddd42badbac57eb464f6b4.1678926164.git.ackerleytng@google.com>
Subject: [RFC PATCH 04/10] KVM: selftests: Exercise restrictedmem allocation
 and truncation code after KVM invalidation code has been unbound
From:   Ackerley Tng <ackerleytng@google.com>
To:     kvm@vger.kernel.org, linux-api@vger.kernel.org,
        linux-arch@vger.kernel.org, linux-doc@vger.kernel.org,
        linux-fsdevel@vger.kernel.org, linux-kernel@vger.kernel.org,
        linux-mm@kvack.org, qemu-devel@nongnu.org
Cc:     aarcange@redhat.com, ak@linux.intel.com, akpm@linux-foundation.org,
        arnd@arndb.de, bfields@fieldses.org, bp@alien8.de,
        chao.p.peng@linux.intel.com, corbet@lwn.net, dave.hansen@intel.com,
        david@redhat.com, ddutile@redhat.com, dhildenb@redhat.com,
        hpa@zytor.com, hughd@google.com, jlayton@kernel.org,
        jmattson@google.com, joro@8bytes.org, jun.nakajima@intel.com,
        kirill.shutemov@linux.intel.com, linmiaohe@huawei.com,
        luto@kernel.org, mail@maciej.szmigiero.name, mhocko@suse.com,
        michael.roth@amd.com, mingo@redhat.com, naoya.horiguchi@nec.com,
        pbonzini@redhat.com, qperret@google.com, rppt@kernel.org,
        seanjc@google.com, shuah@kernel.org, steven.price@arm.com,
        tabba@google.com, tglx@linutronix.de, vannapurve@google.com,
        vbabka@suse.cz, vkuznets@redhat.com, wanpengli@tencent.com,
        wei.w.wang@intel.com, x86@kernel.org, yu.c.zhang@linux.intel.com,
        Ackerley Tng <ackerleytng@google.com>
Content-Type: text/plain; charset="UTF-8"
Precedence: bulk

The kernel interfaces restrictedmem_bind and restrictedmem_unbind are
used by KVM to bind/unbind kvm functions to restrictedmem's
invalidate_start and invalidate_end callbacks.

After the KVM VM is freed, the KVM functions should have been unbound
from the restrictedmem_fd's callbacks.

In this test, we exercise fallocate to back and unback memory using
the restrictedmem fd, and we expect no problems (crashes) after the
KVM functions have been unbound.

Signed-off-by: Ackerley Tng <ackerleytng@google.com>
---
 .../kvm/x86_64/private_mem_conversions_test.c | 26 ++++++++++++++++++-
 1 file changed, 25 insertions(+), 1 deletion(-)

diff --git a/tools/testing/selftests/kvm/x86_64/private_mem_conversions_test.c b/tools/testing/selftests/kvm/x86_64/private_mem_conversions_test.c
index f2c1e4450b0e..7741916818db 100644
--- a/tools/testing/selftests/kvm/x86_64/private_mem_conversions_test.c
+++ b/tools/testing/selftests/kvm/x86_64/private_mem_conversions_test.c
@@ -203,6 +203,30 @@ static void handle_exit_hypercall(struct kvm_vcpu *vcpu)
 	run->hypercall.ret = 0;
 }
 
+static void test_invalidation_code_unbound(struct kvm_vm *vm)
+{
+	uint32_t fd;
+	uint64_t offset;
+	struct userspace_mem_region *region;
+
+	region = memslot2region(vm, DATA_SLOT);
+	fd = region->region.restrictedmem_fd;
+	offset = region->region.restrictedmem_offset;
+
+	kvm_vm_free(vm);
+
+	/*
+	 * At this point the KVM invalidation code should have been unbound from
+	 * the vm. We do allocation and truncation to exercise the restrictedmem
+	 * code. There should be no issues after the unbinding happens.
+	 */
+	if (fallocate(fd, 0, offset, DATA_SIZE))
+		TEST_FAIL("Unexpected error in fallocate");
+	if (fallocate(fd, FALLOC_FL_PUNCH_HOLE | FALLOC_FL_KEEP_SIZE,
+		      offset, DATA_SIZE))
+		TEST_FAIL("Unexpected error in fallocate");
+}
+
 static void test_mem_conversions(enum vm_mem_backing_src_type src_type)
 {
 	struct kvm_vcpu *vcpu;
@@ -270,7 +294,7 @@ static void test_mem_conversions(enum vm_mem_backing_src_type src_type)
 	}
 
 done:
-	kvm_vm_free(vm);
+	test_invalidation_code_unbound(vm);
 }
 
 int main(int argc, char *argv[])
-- 
2.40.0.rc2.332.ga46443480c-goog