diff --git a/gatekeeper/1.0/default/OWNERS b/gatekeeper/1.0/default/OWNERS
new file mode 100644
index 0000000..335660d
--- /dev/null
+++ b/gatekeeper/1.0/default/OWNERS
@@ -0,0 +1,2 @@
+jdanis@google.com
+swillden@google.com
diff --git a/gatekeeper/1.0/software/Android.bp b/gatekeeper/1.0/software/Android.bp
new file mode 100644
index 0000000..148c989
--- /dev/null
+++ b/gatekeeper/1.0/software/Android.bp
@@ -0,0 +1,28 @@
+cc_binary {
+    name: "android.hardware.gatekeeper@1.0-service.software",
+    defaults: ["hidl_defaults"],
+    relative_install_path: "hw",
+    vendor: true,
+    init_rc: ["android.hardware.gatekeeper@1.0-service.software.rc"],
+
+    srcs: [
+        "service.cpp",
+        "SoftGateKeeperDevice.cpp",
+    ],
+
+    shared_libs: [
+        "android.hardware.gatekeeper@1.0",
+        "libbase",
+        "libhardware",
+        "libhidlbase",
+        "libhidltransport",
+        "libutils",
+        "liblog",
+        "libcrypto",
+        "libgatekeeper",
+    ],
+
+    static_libs: ["libscrypt_static"],
+
+    vintf_fragments: ["android.hardware.gatekeeper@1.0-service.software.xml"],
+}
diff --git a/gatekeeper/1.0/software/OWNERS b/gatekeeper/1.0/software/OWNERS
new file mode 100644
index 0000000..335660d
--- /dev/null
+++ b/gatekeeper/1.0/software/OWNERS
@@ -0,0 +1,2 @@
+jdanis@google.com
+swillden@google.com
diff --git a/gatekeeper/1.0/software/SoftGateKeeper.h b/gatekeeper/1.0/software/SoftGateKeeper.h
new file mode 100644
index 0000000..3276d1e
--- /dev/null
+++ b/gatekeeper/1.0/software/SoftGateKeeper.h
@@ -0,0 +1,171 @@
+/*
+ * Copyright 2015 The Android Open Source Project
+ *
+ * Licensed under the Apache License, Version 2.0 (the "License");
+ * you may not use this file except in compliance with the License.
+ * You may obtain a copy of the License at
+ *
+ *      http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ *
+ */
+
+#ifndef SOFT_GATEKEEPER_H_
+#define SOFT_GATEKEEPER_H_
+
+extern "C" {
+#include <openssl/rand.h>
+#include <openssl/sha.h>
+
+#include <crypto_scrypt.h>
+}
+
+#include <android-base/memory.h>
+#include <gatekeeper/gatekeeper.h>
+
+#include <iostream>
+#include <memory>
+#include <unordered_map>
+
+namespace gatekeeper {
+
+struct fast_hash_t {
+    uint64_t salt;
+    uint8_t digest[SHA256_DIGEST_LENGTH];
+};
+
+class SoftGateKeeper : public GateKeeper {
+  public:
+    static const uint32_t SIGNATURE_LENGTH_BYTES = 32;
+
+    // scrypt params
+    static const uint64_t N = 16384;
+    static const uint32_t r = 8;
+    static const uint32_t p = 1;
+
+    static const int MAX_UINT_32_CHARS = 11;
+
+    SoftGateKeeper() {
+        key_.reset(new uint8_t[SIGNATURE_LENGTH_BYTES]);
+        memset(key_.get(), 0, SIGNATURE_LENGTH_BYTES);
+    }
+
+    virtual ~SoftGateKeeper() {}
+
+    virtual bool GetAuthTokenKey(const uint8_t** auth_token_key, uint32_t* length) const {
+        if (auth_token_key == NULL || length == NULL) return false;
+        *auth_token_key = key_.get();
+        *length = SIGNATURE_LENGTH_BYTES;
+        return true;
+    }
+
+    virtual void GetPasswordKey(const uint8_t** password_key, uint32_t* length) {
+        if (password_key == NULL || length == NULL) return;
+        *password_key = key_.get();
+        *length = SIGNATURE_LENGTH_BYTES;
+    }
+
+    virtual void ComputePasswordSignature(uint8_t* signature, uint32_t signature_length,
+                                          const uint8_t*, uint32_t, const uint8_t* password,
+                                          uint32_t password_length, salt_t salt) const {
+        if (signature == NULL) return;
+        crypto_scrypt(password, password_length, reinterpret_cast<uint8_t*>(&salt), sizeof(salt), N,
+                      r, p, signature, signature_length);
+    }
+
+    virtual void GetRandom(void* random, uint32_t requested_length) const {
+        if (random == NULL) return;
+        RAND_pseudo_bytes((uint8_t*)random, requested_length);
+    }
+
+    virtual void ComputeSignature(uint8_t* signature, uint32_t signature_length, const uint8_t*,
+                                  uint32_t, const uint8_t*, const uint32_t) const {
+        if (signature == NULL) return;
+        memset(signature, 0, signature_length);
+    }
+
+    virtual uint64_t GetMillisecondsSinceBoot() const {
+        struct timespec time;
+        int res = clock_gettime(CLOCK_BOOTTIME, &time);
+        if (res < 0) return 0;
+        return (time.tv_sec * 1000) + (time.tv_nsec / 1000 / 1000);
+    }
+
+    virtual bool IsHardwareBacked() const { return false; }
+
+    virtual bool GetFailureRecord(uint32_t uid, secure_id_t user_id, failure_record_t* record,
+                                  bool /* secure */) {
+        failure_record_t* stored = &failure_map_[uid];
+        if (user_id != stored->secure_user_id) {
+            stored->secure_user_id = user_id;
+            stored->last_checked_timestamp = 0;
+            stored->failure_counter = 0;
+        }
+        memcpy(record, stored, sizeof(*record));
+        return true;
+    }
+
+    virtual bool ClearFailureRecord(uint32_t uid, secure_id_t user_id, bool /* secure */) {
+        failure_record_t* stored = &failure_map_[uid];
+        stored->secure_user_id = user_id;
+        stored->last_checked_timestamp = 0;
+        stored->failure_counter = 0;
+        return true;
+    }
+
+    virtual bool WriteFailureRecord(uint32_t uid, failure_record_t* record, bool /* secure */) {
+        failure_map_[uid] = *record;
+        return true;
+    }
+
+    fast_hash_t ComputeFastHash(const SizedBuffer& password, uint64_t salt) {
+        fast_hash_t fast_hash;
+        size_t digest_size = password.size() + sizeof(salt);
+        std::unique_ptr<uint8_t[]> digest(new uint8_t[digest_size]);
+        memcpy(digest.get(), &salt, sizeof(salt));
+        memcpy(digest.get() + sizeof(salt), password.Data<uint8_t>(), password.size());
+
+        SHA256(digest.get(), digest_size, (uint8_t*)&fast_hash.digest);
+
+        fast_hash.salt = salt;
+        return fast_hash;
+    }
+
+    bool VerifyFast(const fast_hash_t& fast_hash, const SizedBuffer& password) {
+        fast_hash_t computed = ComputeFastHash(password, fast_hash.salt);
+        return memcmp(computed.digest, fast_hash.digest, SHA256_DIGEST_LENGTH) == 0;
+    }
+
+    bool DoVerify(const password_handle_t* expected_handle, const SizedBuffer& password) {
+        uint64_t user_id = android::base::get_unaligned<secure_id_t>(&expected_handle->user_id);
+        FastHashMap::const_iterator it = fast_hash_map_.find(user_id);
+        if (it != fast_hash_map_.end() && VerifyFast(it->second, password)) {
+            return true;
+        } else {
+            if (GateKeeper::DoVerify(expected_handle, password)) {
+                uint64_t salt;
+                GetRandom(&salt, sizeof(salt));
+                fast_hash_map_[user_id] = ComputeFastHash(password, salt);
+                return true;
+            }
+        }
+
+        return false;
+    }
+
+  private:
+    typedef std::unordered_map<uint32_t, failure_record_t> FailureRecordMap;
+    typedef std::unordered_map<uint64_t, fast_hash_t> FastHashMap;
+
+    std::unique_ptr<uint8_t[]> key_;
+    FailureRecordMap failure_map_;
+    FastHashMap fast_hash_map_;
+};
+}  // namespace gatekeeper
+
+#endif  // SOFT_GATEKEEPER_H_
diff --git a/gatekeeper/1.0/software/SoftGateKeeperDevice.cpp b/gatekeeper/1.0/software/SoftGateKeeperDevice.cpp
new file mode 100644
index 0000000..d7a0b46
--- /dev/null
+++ b/gatekeeper/1.0/software/SoftGateKeeperDevice.cpp
@@ -0,0 +1,114 @@
+/*
+ * Copyright (C) 2015 The Android Open Source Project
+ *
+ * Licensed under the Apache License, Version 2.0 (the "License");
+ * you may not use this file except in compliance with the License.
+ * You may obtain a copy of the License at
+ *
+ *      http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+#include "SoftGateKeeperDevice.h"
+#include "SoftGateKeeper.h"
+
+using ::android::hardware::hidl_vec;
+using ::android::hardware::Return;
+using ::android::hardware::gatekeeper::V1_0::GatekeeperStatusCode;
+using ::gatekeeper::EnrollRequest;
+using ::gatekeeper::EnrollResponse;
+using ::gatekeeper::ERROR_INVALID;
+using ::gatekeeper::ERROR_MEMORY_ALLOCATION_FAILED;
+using ::gatekeeper::ERROR_NONE;
+using ::gatekeeper::ERROR_RETRY;
+using ::gatekeeper::SizedBuffer;
+using ::gatekeeper::VerifyRequest;
+using ::gatekeeper::VerifyResponse;
+
+#include <limits>
+
+namespace android {
+
+inline SizedBuffer hidl_vec2sized_buffer(const hidl_vec<uint8_t>& vec) {
+    if (vec.size() == 0 || vec.size() > std::numeric_limits<uint32_t>::max()) return {};
+    auto dummy = new uint8_t[vec.size()];
+    std::copy(vec.begin(), vec.end(), dummy);
+    return {dummy, static_cast<uint32_t>(vec.size())};
+}
+
+Return<void> SoftGateKeeperDevice::enroll(uint32_t uid,
+                                          const hidl_vec<uint8_t>& currentPasswordHandle,
+                                          const hidl_vec<uint8_t>& currentPassword,
+                                          const hidl_vec<uint8_t>& desiredPassword,
+                                          enroll_cb _hidl_cb) {
+    if (desiredPassword.size() == 0) {
+        _hidl_cb({GatekeeperStatusCode::ERROR_GENERAL_FAILURE, 0, {}});
+        return {};
+    }
+
+    EnrollRequest request(uid, hidl_vec2sized_buffer(currentPasswordHandle),
+                          hidl_vec2sized_buffer(desiredPassword),
+                          hidl_vec2sized_buffer(currentPassword));
+    EnrollResponse response;
+    impl_->Enroll(request, &response);
+
+    if (response.error == ERROR_RETRY) {
+        _hidl_cb({GatekeeperStatusCode::ERROR_RETRY_TIMEOUT, response.retry_timeout, {}});
+    } else if (response.error != ERROR_NONE) {
+        _hidl_cb({GatekeeperStatusCode::ERROR_GENERAL_FAILURE, 0, {}});
+    } else {
+        hidl_vec<uint8_t> new_handle(response.enrolled_password_handle.Data<uint8_t>(),
+                                     response.enrolled_password_handle.Data<uint8_t>() +
+                                             response.enrolled_password_handle.size());
+        _hidl_cb({GatekeeperStatusCode::STATUS_OK, response.retry_timeout, new_handle});
+    }
+    return {};
+}
+
+Return<void> SoftGateKeeperDevice::verify(
+        uint32_t uid, uint64_t challenge,
+        const ::android::hardware::hidl_vec<uint8_t>& enrolledPasswordHandle,
+        const ::android::hardware::hidl_vec<uint8_t>& providedPassword, verify_cb _hidl_cb) {
+    if (enrolledPasswordHandle.size() == 0) {
+        _hidl_cb({GatekeeperStatusCode::ERROR_GENERAL_FAILURE, 0, {}});
+        return {};
+    }
+
+    VerifyRequest request(uid, challenge, hidl_vec2sized_buffer(enrolledPasswordHandle),
+                          hidl_vec2sized_buffer(providedPassword));
+    VerifyResponse response;
+
+    impl_->Verify(request, &response);
+
+    if (response.error == ERROR_RETRY) {
+        _hidl_cb({GatekeeperStatusCode::ERROR_RETRY_TIMEOUT, response.retry_timeout, {}});
+    } else if (response.error != ERROR_NONE) {
+        _hidl_cb({GatekeeperStatusCode::ERROR_GENERAL_FAILURE, 0, {}});
+    } else {
+        hidl_vec<uint8_t> auth_token(
+                response.auth_token.Data<uint8_t>(),
+                response.auth_token.Data<uint8_t>() + response.auth_token.size());
+
+        _hidl_cb({response.request_reenroll ? GatekeeperStatusCode::STATUS_REENROLL
+                                            : GatekeeperStatusCode::STATUS_OK,
+                  response.retry_timeout, auth_token});
+    }
+    return {};
+}
+
+Return<void> SoftGateKeeperDevice::deleteUser(uint32_t /*uid*/, deleteUser_cb _hidl_cb) {
+    _hidl_cb({GatekeeperStatusCode::ERROR_NOT_IMPLEMENTED, 0, {}});
+    return {};
+}
+
+Return<void> SoftGateKeeperDevice::deleteAllUsers(deleteAllUsers_cb _hidl_cb) {
+    _hidl_cb({GatekeeperStatusCode::ERROR_NOT_IMPLEMENTED, 0, {}});
+    return {};
+}
+
+}  // namespace android
diff --git a/gatekeeper/1.0/software/SoftGateKeeperDevice.h b/gatekeeper/1.0/software/SoftGateKeeperDevice.h
new file mode 100644
index 0000000..17b474e
--- /dev/null
+++ b/gatekeeper/1.0/software/SoftGateKeeperDevice.h
@@ -0,0 +1,80 @@
+/*
+ * Copyright 2015 The Android Open Source Project
+ *
+ * Licensed under the Apache License, Version 2.0 (the "License");
+ * you may not use this file except in compliance with the License.
+ * You may obtain a copy of the License at
+ *
+ *      http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+#ifndef SOFT_GATEKEEPER_DEVICE_H_
+#define SOFT_GATEKEEPER_DEVICE_H_
+
+#include <android/hardware/gatekeeper/1.0/IGatekeeper.h>
+#include <hidl/Status.h>
+
+#include <memory>
+#include "SoftGateKeeper.h"
+
+namespace android {
+
+/**
+ * Software based GateKeeper implementation
+ */
+class SoftGateKeeperDevice : public ::android::hardware::gatekeeper::V1_0::IGatekeeper {
+  public:
+    SoftGateKeeperDevice() { impl_.reset(new ::gatekeeper::SoftGateKeeper()); }
+
+    // Wrappers to translate the gatekeeper HAL API to the Kegyuard Messages API.
+
+    /**
+     * Enrolls password_payload, which should be derived from a user selected pin or password,
+     * with the authentication factor private key used only for enrolling authentication
+     * factor data.
+     *
+     * Returns: 0 on success or an error code less than 0 on error.
+     * On error, enrolled_password_handle will not be allocated.
+     */
+    ::android::hardware::Return<void> enroll(
+            uint32_t uid, const ::android::hardware::hidl_vec<uint8_t>& currentPasswordHandle,
+            const ::android::hardware::hidl_vec<uint8_t>& currentPassword,
+            const ::android::hardware::hidl_vec<uint8_t>& desiredPassword,
+            enroll_cb _hidl_cb) override;
+
+    /**
+     * Verifies provided_password matches enrolled_password_handle.
+     *
+     * Implementations of this module may retain the result of this call
+     * to attest to the recency of authentication.
+     *
+     * On success, writes the address of a verification token to auth_token,
+     * usable to attest password verification to other trusted services. Clients
+     * may pass NULL for this value.
+     *
+     * Returns: 0 on success or an error code less than 0 on error
+     * On error, verification token will not be allocated
+     */
+    ::android::hardware::Return<void> verify(
+            uint32_t uid, uint64_t challenge,
+            const ::android::hardware::hidl_vec<uint8_t>& enrolledPasswordHandle,
+            const ::android::hardware::hidl_vec<uint8_t>& providedPassword,
+            verify_cb _hidl_cb) override;
+
+    ::android::hardware::Return<void> deleteUser(uint32_t uid, deleteUser_cb _hidl_cb) override;
+
+    ::android::hardware::Return<void> deleteAllUsers(deleteAllUsers_cb _hidl_cb) override;
+
+  private:
+    std::unique_ptr<::gatekeeper::SoftGateKeeper> impl_;
+};
+
+}  // namespace android
+
+#endif  // SOFT_GATEKEEPER_DEVICE_H_
diff --git a/gatekeeper/1.0/software/android.hardware.gatekeeper@1.0-service.software.rc b/gatekeeper/1.0/software/android.hardware.gatekeeper@1.0-service.software.rc
new file mode 100644
index 0000000..60cb96c
--- /dev/null
+++ b/gatekeeper/1.0/software/android.hardware.gatekeeper@1.0-service.software.rc
@@ -0,0 +1,4 @@
+service vendor.gatekeeper-1-0 /vendor/bin/hw/android.hardware.gatekeeper@1.0-service.software
+    class hal
+    user system
+    group system
diff --git a/gatekeeper/1.0/software/android.hardware.gatekeeper@1.0-service.software.xml b/gatekeeper/1.0/software/android.hardware.gatekeeper@1.0-service.software.xml
new file mode 100644
index 0000000..19714a8
--- /dev/null
+++ b/gatekeeper/1.0/software/android.hardware.gatekeeper@1.0-service.software.xml
@@ -0,0 +1,11 @@
+<manifest version="1.0" type="device">
+    <hal format="hidl">
+        <name>android.hardware.gatekeeper</name>
+        <transport>hwbinder</transport>
+        <version>1.0</version>
+        <interface>
+        <name>IGatekeeper</name>
+            <instance>default</instance>
+        </interface>
+    </hal>
+</manifest>
diff --git a/gatekeeper/1.0/software/service.cpp b/gatekeeper/1.0/software/service.cpp
new file mode 100644
index 0000000..a48a964
--- /dev/null
+++ b/gatekeeper/1.0/software/service.cpp
@@ -0,0 +1,39 @@
+/*
+ * Copyright (C) 2019 The Android Open Source Project
+ *
+ * Licensed under the Apache License, Version 2.0 (the "License");
+ * you may not use this file except in compliance with the License.
+ * You may obtain a copy of the License at
+ *
+ *      http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+#define LOG_TAG "android.hardware.gatekeeper@1.0-service"
+
+#include <android-base/logging.h>
+#include <android/hardware/gatekeeper/1.0/IGatekeeper.h>
+
+#include <hidl/LegacySupport.h>
+
+#include "SoftGateKeeperDevice.h"
+
+// Generated HIDL files
+using android::SoftGateKeeperDevice;
+using android::hardware::gatekeeper::V1_0::IGatekeeper;
+
+int main() {
+    ::android::hardware::configureRpcThreadpool(1, true /* willJoinThreadpool */);
+    android::sp<SoftGateKeeperDevice> gatekeeper(new SoftGateKeeperDevice());
+    auto status = gatekeeper->registerAsService();
+    if (status != android::OK) {
+        LOG(FATAL) << "Could not register service for Gatekeeper 1.0 (software) (" << status << ")";
+    }
+
+    android::hardware::joinRpcThreadpool();
+    return -1;  // Should never get here.
+}
diff --git a/gatekeeper/1.0/software/tests/Android.bp b/gatekeeper/1.0/software/tests/Android.bp
new file mode 100644
index 0000000..3f0300d
--- /dev/null
+++ b/gatekeeper/1.0/software/tests/Android.bp
@@ -0,0 +1,34 @@
+//
+// Copyright (C) 2015 The Android Open Source Project
+//
+// Licensed under the Apache License, Version 2.0 (the "License");
+// you may not use this file except in compliance with the License.
+// You may obtain a copy of the License at
+//
+//      http://www.apache.org/licenses/LICENSE-2.0
+//
+// Unless required by applicable law or agreed to in writing, software
+// distributed under the License is distributed on an "AS IS" BASIS,
+// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+// See the License for the specific language governing permissions and
+// limitations under the License.
+//
+
+cc_test {
+    name: "gatekeeper-software-device-unit-tests",
+
+    cflags: [
+        "-g",
+        "-Wall",
+        "-Werror",
+        "-Wno-missing-field-initializers",
+    ],
+    shared_libs: [
+        "libgatekeeper",
+        "libcrypto",
+        "libbase",
+    ],
+    static_libs: ["libscrypt_static"],
+
+    srcs: ["gatekeeper_test.cpp"],
+}
diff --git a/gatekeeper/1.0/software/tests/gatekeeper_test.cpp b/gatekeeper/1.0/software/tests/gatekeeper_test.cpp
new file mode 100644
index 0000000..bf4a8bc
--- /dev/null
+++ b/gatekeeper/1.0/software/tests/gatekeeper_test.cpp
@@ -0,0 +1,178 @@
+/*
+ * Copyright 2015 The Android Open Source Project
+ *
+ * Licensed under the Apache License, Version 2.0 (the "License");
+ * you may not use this file except in compliance with the License.
+ * You may obtain a copy of the License at
+ *
+ *      http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+#include <arpa/inet.h>
+#include <iostream>
+
+#include <gtest/gtest.h>
+#include <hardware/hw_auth_token.h>
+
+#include "../SoftGateKeeper.h"
+
+using ::gatekeeper::EnrollRequest;
+using ::gatekeeper::EnrollResponse;
+using ::gatekeeper::secure_id_t;
+using ::gatekeeper::SizedBuffer;
+using ::gatekeeper::SoftGateKeeper;
+using ::gatekeeper::VerifyRequest;
+using ::gatekeeper::VerifyResponse;
+using ::testing::Test;
+
+static SizedBuffer makePasswordBuffer(int init = 0) {
+    constexpr const uint32_t pw_buffer_size = 16;
+    auto pw_buffer = new uint8_t[pw_buffer_size];
+    memset(pw_buffer, init, pw_buffer_size);
+
+    return {pw_buffer, pw_buffer_size};
+}
+
+static SizedBuffer makeAndInitializeSizedBuffer(const uint8_t* data, uint32_t size) {
+    auto buffer = new uint8_t[size];
+    memcpy(buffer, data, size);
+    return {buffer, size};
+}
+
+static SizedBuffer copySizedBuffer(const SizedBuffer& rhs) {
+    return makeAndInitializeSizedBuffer(rhs.Data<uint8_t>(), rhs.size());
+}
+
+static void do_enroll(SoftGateKeeper& gatekeeper, EnrollResponse* response) {
+    EnrollRequest request(0, {}, makePasswordBuffer(), {});
+
+    gatekeeper.Enroll(request, response);
+}
+
+TEST(GateKeeperTest, EnrollSuccess) {
+    SoftGateKeeper gatekeeper;
+    EnrollResponse response;
+    do_enroll(gatekeeper, &response);
+    ASSERT_EQ(::gatekeeper::gatekeeper_error_t::ERROR_NONE, response.error);
+}
+
+TEST(GateKeeperTest, EnrollBogusData) {
+    SoftGateKeeper gatekeeper;
+    EnrollResponse response;
+
+    EnrollRequest request(0, {}, {}, {});
+
+    gatekeeper.Enroll(request, &response);
+
+    ASSERT_EQ(::gatekeeper::gatekeeper_error_t::ERROR_INVALID, response.error);
+}
+
+TEST(GateKeeperTest, VerifySuccess) {
+    SoftGateKeeper gatekeeper;
+    EnrollResponse enroll_response;
+
+    do_enroll(gatekeeper, &enroll_response);
+    ASSERT_EQ(::gatekeeper::gatekeeper_error_t::ERROR_NONE, enroll_response.error);
+    VerifyRequest request(0, 1, std::move(enroll_response.enrolled_password_handle),
+                          makePasswordBuffer());
+    VerifyResponse response;
+
+    gatekeeper.Verify(request, &response);
+
+    ASSERT_EQ(::gatekeeper::gatekeeper_error_t::ERROR_NONE, response.error);
+
+    auto auth_token = response.auth_token.Data<hw_auth_token_t>();
+
+    ASSERT_NE(nullptr, auth_token);
+    ASSERT_EQ((uint32_t)HW_AUTH_PASSWORD, ntohl(auth_token->authenticator_type));
+    ASSERT_EQ((uint64_t)1, auth_token->challenge);
+    ASSERT_NE(~((uint32_t)0), auth_token->timestamp);
+    ASSERT_NE((uint64_t)0, auth_token->user_id);
+    ASSERT_NE((uint64_t)0, auth_token->authenticator_id);
+}
+
+TEST(GateKeeperTest, TrustedReEnroll) {
+    SoftGateKeeper gatekeeper;
+    EnrollResponse enroll_response;
+
+    // do_enroll enrolls an all 0 password
+    do_enroll(gatekeeper, &enroll_response);
+    ASSERT_EQ(::gatekeeper::gatekeeper_error_t::ERROR_NONE, enroll_response.error);
+
+    // verify first password
+    VerifyRequest request(0, 0, copySizedBuffer(enroll_response.enrolled_password_handle),
+                          makePasswordBuffer());
+    VerifyResponse response;
+    gatekeeper.Verify(request, &response);
+    ASSERT_EQ(::gatekeeper::gatekeeper_error_t::ERROR_NONE, response.error);
+    auto auth_token = response.auth_token.Data<hw_auth_token_t>();
+    ASSERT_NE(nullptr, auth_token);
+
+    secure_id_t secure_id = auth_token->user_id;
+
+    // enroll new password
+    EnrollRequest enroll_request(0, std::move(enroll_response.enrolled_password_handle),
+                                 makePasswordBuffer(1) /* new password */,
+                                 makePasswordBuffer() /* old password */);
+    gatekeeper.Enroll(enroll_request, &enroll_response);
+    ASSERT_EQ(::gatekeeper::gatekeeper_error_t::ERROR_NONE, enroll_response.error);
+
+    // verify new password
+    VerifyRequest new_request(0, 0, std::move(enroll_response.enrolled_password_handle),
+                              makePasswordBuffer(1));
+    gatekeeper.Verify(new_request, &response);
+    ASSERT_EQ(::gatekeeper::gatekeeper_error_t::ERROR_NONE, response.error);
+    ASSERT_NE(nullptr, response.auth_token.Data<hw_auth_token_t>());
+    ASSERT_EQ(secure_id, response.auth_token.Data<hw_auth_token_t>()->user_id);
+}
+
+TEST(GateKeeperTest, UntrustedReEnroll) {
+    SoftGateKeeper gatekeeper;
+    SizedBuffer provided_password;
+    EnrollResponse enroll_response;
+
+    // do_enroll enrolls an all 0 password
+    provided_password = makePasswordBuffer();
+    do_enroll(gatekeeper, &enroll_response);
+    ASSERT_EQ(::gatekeeper::gatekeeper_error_t::ERROR_NONE, enroll_response.error);
+
+    // verify first password
+    VerifyRequest request(0, 0, std::move(enroll_response.enrolled_password_handle),
+                          std::move(provided_password));
+    VerifyResponse response;
+    gatekeeper.Verify(request, &response);
+    ASSERT_EQ(::gatekeeper::gatekeeper_error_t::ERROR_NONE, response.error);
+    auto auth_token = response.auth_token.Data<hw_auth_token_t>();
+    ASSERT_NE(nullptr, auth_token);
+
+    secure_id_t secure_id = auth_token->user_id;
+
+    EnrollRequest enroll_request(0, {}, makePasswordBuffer(1), {});
+    gatekeeper.Enroll(enroll_request, &enroll_response);
+    ASSERT_EQ(::gatekeeper::gatekeeper_error_t::ERROR_NONE, enroll_response.error);
+
+    // verify new password
+    VerifyRequest new_request(0, 0, std::move(enroll_response.enrolled_password_handle),
+                              makePasswordBuffer(1));
+    gatekeeper.Verify(new_request, &response);
+    ASSERT_EQ(::gatekeeper::gatekeeper_error_t::ERROR_NONE, response.error);
+    ASSERT_NE(nullptr, response.auth_token.Data<hw_auth_token_t>());
+    ASSERT_NE(secure_id, response.auth_token.Data<hw_auth_token_t>()->user_id);
+}
+
+TEST(GateKeeperTest, VerifyBogusData) {
+    SoftGateKeeper gatekeeper;
+    VerifyResponse response;
+
+    VerifyRequest request(0, 0, {}, {});
+
+    gatekeeper.Verify(request, &response);
+
+    ASSERT_EQ(::gatekeeper::gatekeeper_error_t::ERROR_INVALID, response.error);
+}
diff --git a/gatekeeper/1.0/vts/OWNERS b/gatekeeper/1.0/vts/OWNERS
new file mode 100644
index 0000000..738c710
--- /dev/null
+++ b/gatekeeper/1.0/vts/OWNERS
@@ -0,0 +1,3 @@
+jdanis@google.com
+swillden@google.com
+guangzhu@google.com
diff --git a/graphics/composer/2.1/utils/vts/Android.bp b/graphics/composer/2.1/utils/vts/Android.bp
index 846cfdf..fcb327f 100644
--- a/graphics/composer/2.1/utils/vts/Android.bp
+++ b/graphics/composer/2.1/utils/vts/Android.bp
@@ -25,6 +25,8 @@
     static_libs: [
         "VtsHalHidlTargetTestBase",
         "android.hardware.graphics.composer@2.1",
+        "android.hardware.graphics.mapper@2.0-vts",
+        "android.hardware.graphics.mapper@3.0-vts",
     ],
     header_libs: [
         "android.hardware.graphics.composer@2.1-command-buffer",
diff --git a/graphics/composer/2.1/utils/vts/ComposerVts.cpp b/graphics/composer/2.1/utils/vts/ComposerVts.cpp
index 7ba67d4..c5d5823 100644
--- a/graphics/composer/2.1/utils/vts/ComposerVts.cpp
+++ b/graphics/composer/2.1/utils/vts/ComposerVts.cpp
@@ -315,6 +315,77 @@
     writer->reset();
 }
 
+Gralloc::Gralloc() {
+    [this] {
+        ASSERT_NO_FATAL_FAILURE(mGralloc3 = std::make_shared<Gralloc3>("default", "default",
+                                                                       /*errOnFailure=*/false));
+        if (mGralloc3->getAllocator() == nullptr || mGralloc3->getMapper() == nullptr) {
+            mGralloc3 = nullptr;
+            ASSERT_NO_FATAL_FAILURE(mGralloc2 = std::make_shared<Gralloc2>());
+        }
+    }();
+}
+
+const native_handle_t* Gralloc::allocate(uint32_t width, uint32_t height, uint32_t layerCount,
+                                         PixelFormat format, uint64_t usage, bool import,
+                                         uint32_t* outStride) {
+    if (mGralloc3) {
+        IMapper3::BufferDescriptorInfo info{};
+        info.width = width;
+        info.height = height;
+        info.layerCount = layerCount;
+        info.format = static_cast<android::hardware::graphics::common::V1_2::PixelFormat>(format);
+        info.usage = usage;
+        return mGralloc3->allocate(info, import, outStride);
+    } else {
+        IMapper2::BufferDescriptorInfo info{};
+        info.width = width;
+        info.height = height;
+        info.layerCount = layerCount;
+        info.format = format;
+        info.usage = usage;
+        return mGralloc2->allocate(info, import, outStride);
+    }
+}
+
+void* Gralloc::lock(const native_handle_t* bufferHandle, uint64_t cpuUsage,
+                    const AccessRegion& accessRegionRect, int acquireFence) {
+    if (mGralloc3) {
+        IMapper3::Rect accessRegion;
+        accessRegion.left = accessRegionRect.left;
+        accessRegion.top = accessRegionRect.top;
+        accessRegion.width = accessRegionRect.width;
+        accessRegion.height = accessRegionRect.height;
+        int32_t bytesPerPixel;
+        int32_t bytesPerStride;
+        return mGralloc3->lock(bufferHandle, cpuUsage, accessRegion, acquireFence, &bytesPerPixel,
+                               &bytesPerStride);
+    } else {
+        IMapper2::Rect accessRegion;
+        accessRegion.left = accessRegionRect.left;
+        accessRegion.top = accessRegionRect.top;
+        accessRegion.width = accessRegionRect.width;
+        accessRegion.height = accessRegionRect.height;
+        return mGralloc2->lock(bufferHandle, cpuUsage, accessRegion, acquireFence);
+    }
+}
+
+int Gralloc::unlock(const native_handle_t* bufferHandle) {
+    if (mGralloc3) {
+        return mGralloc3->unlock(bufferHandle);
+    } else {
+        return mGralloc2->unlock(bufferHandle);
+    }
+}
+
+void Gralloc::freeBuffer(const native_handle_t* bufferHandle) {
+    if (mGralloc3) {
+        mGralloc3->freeBuffer(bufferHandle);
+    } else {
+        mGralloc2->freeBuffer(bufferHandle);
+    }
+}
+
 }  // namespace vts
 }  // namespace V2_1
 }  // namespace composer
diff --git a/graphics/composer/2.1/utils/vts/include/composer-vts/2.1/ComposerVts.h b/graphics/composer/2.1/utils/vts/include/composer-vts/2.1/ComposerVts.h
index c97be76..7811048 100644
--- a/graphics/composer/2.1/utils/vts/include/composer-vts/2.1/ComposerVts.h
+++ b/graphics/composer/2.1/utils/vts/include/composer-vts/2.1/ComposerVts.h
@@ -25,8 +25,12 @@
 #include <android/hardware/graphics/composer/2.1/IComposer.h>
 #include <composer-command-buffer/2.1/ComposerCommandBuffer.h>
 #include <composer-vts/2.1/TestCommandReader.h>
+#include <mapper-vts/2.0/MapperVts.h>
+#include <mapper-vts/3.0/MapperVts.h>
 #include <utils/StrongPointer.h>
 
+#include "gtest/gtest.h"
+
 namespace android {
 namespace hardware {
 namespace graphics {
@@ -38,6 +42,10 @@
 using android::hardware::graphics::common::V1_0::Dataspace;
 using android::hardware::graphics::common::V1_0::Hdr;
 using android::hardware::graphics::common::V1_0::PixelFormat;
+using IMapper2 = android::hardware::graphics::mapper::V2_0::IMapper;
+using IMapper3 = android::hardware::graphics::mapper::V3_0::IMapper;
+using Gralloc2 = android::hardware::graphics::mapper::V2_0::vts::Gralloc;
+using Gralloc3 = android::hardware::graphics::mapper::V3_0::vts::Gralloc;
 
 class ComposerClient;
 
@@ -119,6 +127,34 @@
     const sp<IComposerClient> mClient;
 };
 
+class AccessRegion {
+  public:
+    int32_t left;
+    int32_t top;
+    int32_t width;
+    int32_t height;
+};
+
+class Gralloc {
+  public:
+    explicit Gralloc();
+
+    const native_handle_t* allocate(uint32_t width, uint32_t height, uint32_t layerCount,
+                                    PixelFormat format, uint64_t usage, bool import = true,
+                                    uint32_t* outStride = nullptr);
+
+    void* lock(const native_handle_t* bufferHandle, uint64_t cpuUsage,
+               const AccessRegion& accessRegionRect, int acquireFence);
+
+    int unlock(const native_handle_t* bufferHandle);
+
+    void freeBuffer(const native_handle_t* bufferHandle);
+
+  protected:
+    std::shared_ptr<Gralloc2> mGralloc2 = nullptr;
+    std::shared_ptr<Gralloc3> mGralloc3 = nullptr;
+};
+
 }  // namespace vts
 }  // namespace V2_1
 }  // namespace composer
diff --git a/graphics/composer/2.1/vts/functional/Android.bp b/graphics/composer/2.1/vts/functional/Android.bp
index c98cc0d..d54da60 100644
--- a/graphics/composer/2.1/vts/functional/Android.bp
+++ b/graphics/composer/2.1/vts/functional/Android.bp
@@ -26,10 +26,13 @@
     ],
     static_libs: [
         "android.hardware.graphics.allocator@2.0",
+        "android.hardware.graphics.allocator@3.0",
         "android.hardware.graphics.composer@2.1",
         "android.hardware.graphics.composer@2.1-vts",
         "android.hardware.graphics.mapper@2.0",
         "android.hardware.graphics.mapper@2.0-vts",
+        "android.hardware.graphics.mapper@3.0",
+        "android.hardware.graphics.mapper@3.0-vts",
     ],
     header_libs: [
         "android.hardware.graphics.composer@2.1-command-buffer",
diff --git a/graphics/composer/2.1/vts/functional/VtsHalGraphicsComposerV2_1TargetTest.cpp b/graphics/composer/2.1/vts/functional/VtsHalGraphicsComposerV2_1TargetTest.cpp
index 3c408b7..30b9694 100644
--- a/graphics/composer/2.1/vts/functional/VtsHalGraphicsComposerV2_1TargetTest.cpp
+++ b/graphics/composer/2.1/vts/functional/VtsHalGraphicsComposerV2_1TargetTest.cpp
@@ -21,6 +21,7 @@
 #include <composer-vts/2.1/GraphicsComposerCallback.h>
 #include <composer-vts/2.1/TestCommandReader.h>
 #include <mapper-vts/2.0/MapperVts.h>
+#include <mapper-vts/3.0/MapperVts.h>
 
 #include <VtsHalHidlTargetTestBase.h>
 #include <VtsHalHidlTargetTestEnvBase.h>
@@ -47,8 +48,6 @@
 using android::hardware::graphics::common::V1_0::Dataspace;
 using android::hardware::graphics::common::V1_0::PixelFormat;
 using android::hardware::graphics::common::V1_0::Transform;
-using android::hardware::graphics::mapper::V2_0::IMapper;
-using android::hardware::graphics::mapper::V2_0::vts::Gralloc;
 using GrallocError = android::hardware::graphics::mapper::V2_0::Error;
 
 // Test environment for graphics.composer
@@ -669,7 +668,6 @@
         ASSERT_NO_FATAL_FAILURE(GraphicsComposerHidlTest::SetUp());
 
         ASSERT_NO_FATAL_FAILURE(mGralloc = std::make_unique<Gralloc>());
-
         Config activeConfig = mComposerClient->getActiveConfig(mPrimaryDisplay);
         mDisplayWidth = mComposerClient->getDisplayAttribute(mPrimaryDisplay, activeConfig,
                                                              IComposerClient::Attribute::WIDTH);
@@ -685,16 +683,10 @@
     }
 
     const native_handle_t* allocate() {
-        IMapper::BufferDescriptorInfo info{};
-        info.width = mDisplayWidth;
-        info.height = mDisplayHeight;
-        info.layerCount = 1;
-        info.format = PixelFormat::RGBA_8888;
-        info.usage =
-            static_cast<uint64_t>(BufferUsage::CPU_WRITE_OFTEN | BufferUsage::CPU_READ_OFTEN |
-                                  BufferUsage::COMPOSER_OVERLAY);
-
-        return mGralloc->allocate(info);
+        uint64_t usage =
+                static_cast<uint64_t>(BufferUsage::CPU_WRITE_OFTEN | BufferUsage::CPU_READ_OFTEN |
+                                      BufferUsage::COMPOSER_OVERLAY);
+        return mGralloc->allocate(mDisplayWidth, mDisplayHeight, 1, PixelFormat::RGBA_8888, usage);
     }
 
     void execute() { mComposerClient->execute(mReader.get(), mWriter.get()); }
@@ -705,7 +697,7 @@
     int32_t mDisplayHeight;
 
    private:
-    std::unique_ptr<Gralloc> mGralloc;
+     std::unique_ptr<Gralloc> mGralloc;
 };
 
 /**
diff --git a/graphics/composer/2.2/utils/vts/Android.bp b/graphics/composer/2.2/utils/vts/Android.bp
index c6b524d..56b12f5 100644
--- a/graphics/composer/2.2/utils/vts/Android.bp
+++ b/graphics/composer/2.2/utils/vts/Android.bp
@@ -19,12 +19,17 @@
     defaults: ["hidl_defaults"],
     srcs: [
         "ComposerVts.cpp",
+        "ReadbackVts.cpp",
     ],
     static_libs: [
         "VtsHalHidlTargetTestBase",
         "android.hardware.graphics.composer@2.1",
         "android.hardware.graphics.composer@2.1-vts",
         "android.hardware.graphics.composer@2.2",
+        "android.hardware.graphics.mapper@2.1",
+        "android.hardware.graphics.mapper@2.1-vts",
+        "android.hardware.graphics.mapper@3.0",
+        "android.hardware.graphics.mapper@3.0-vts",
     ],
     export_static_lib_headers: [
         "android.hardware.graphics.composer@2.1-vts",
diff --git a/graphics/composer/2.2/utils/vts/ComposerVts.cpp b/graphics/composer/2.2/utils/vts/ComposerVts.cpp
index da99460..a380fc0 100644
--- a/graphics/composer/2.2/utils/vts/ComposerVts.cpp
+++ b/graphics/composer/2.2/utils/vts/ComposerVts.cpp
@@ -180,6 +180,48 @@
     return matrix;
 }
 
+Gralloc::Gralloc() {
+    [this] {
+        ALOGD("Attempting to initialize gralloc3");
+        ASSERT_NO_FATAL_FAILURE(mGralloc3 = std::make_shared<Gralloc3>("default", "default",
+                                                                       /*errOnFailure=*/false));
+        if (mGralloc3->getMapper() == nullptr || mGralloc3->getAllocator() == nullptr) {
+            mGralloc3 = nullptr;
+            ALOGD("Failed to initialize gralloc3, initializing gralloc2_1");
+            mGralloc2_1 = std::make_shared<Gralloc2_1>(/*errOnFailure*/ false);
+            if (!mGralloc2_1->getMapper()) {
+                mGralloc2_1 = nullptr;
+                ALOGD("Failed to initialize gralloc2_1, initializing gralloc2");
+                ASSERT_NO_FATAL_FAILURE(mGralloc2 = std::make_shared<Gralloc2>());
+            }
+        }
+    }();
+}
+
+bool Gralloc::validateBufferSize(const native_handle_t* bufferHandle, uint32_t width,
+                                 uint32_t height, uint32_t layerCount, PixelFormat format,
+                                 uint64_t usage, uint32_t stride) {
+    if (mGralloc3) {
+        IMapper3::BufferDescriptorInfo info{};
+        info.width = width;
+        info.height = height;
+        info.layerCount = layerCount;
+        info.format = static_cast<android::hardware::graphics::common::V1_2::PixelFormat>(format);
+        info.usage = usage;
+        return mGralloc3->validateBufferSize(bufferHandle, info, stride);
+    } else if (mGralloc2_1) {
+        IMapper2_1::BufferDescriptorInfo info{};
+        info.width = width;
+        info.height = height;
+        info.layerCount = layerCount;
+        info.format = static_cast<android::hardware::graphics::common::V1_1::PixelFormat>(format);
+        info.usage = usage;
+        return mGralloc2_1->validateBufferSize(bufferHandle, info, stride);
+    } else {
+        return true;
+    }
+}
+
 }  // namespace vts
 }  // namespace V2_2
 }  // namespace composer
diff --git a/graphics/composer/2.2/utils/vts/ReadbackVts.cpp b/graphics/composer/2.2/utils/vts/ReadbackVts.cpp
new file mode 100644
index 0000000..81a6452
--- /dev/null
+++ b/graphics/composer/2.2/utils/vts/ReadbackVts.cpp
@@ -0,0 +1,251 @@
+/*
+ * Copyright 2019 The Android Open Source Project
+ *
+ * Licensed under the Apache License, Version 2.0 (the "License");
+ * you may not use this file except in compliance with the License.
+ * You may obtain a copy of the License at
+ *
+ *      http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+#include <composer-vts/2.2/ReadbackVts.h>
+
+namespace android {
+namespace hardware {
+namespace graphics {
+namespace composer {
+namespace V2_2 {
+namespace vts {
+
+void TestLayer::write(const std::shared_ptr<CommandWriterBase>& writer) {
+    writer->selectLayer(mLayer);
+    writer->setLayerDisplayFrame(mDisplayFrame);
+    writer->setLayerSourceCrop(mSourceCrop);
+    writer->setLayerZOrder(mZOrder);
+    writer->setLayerSurfaceDamage(mSurfaceDamage);
+    writer->setLayerTransform(mTransform);
+    writer->setLayerPlaneAlpha(mAlpha);
+    writer->setLayerBlendMode(mBlendMode);
+}
+
+int32_t ReadbackHelper::GetBytesPerPixel(PixelFormat pixelFormat) {
+    switch (pixelFormat) {
+        case PixelFormat::RGBA_8888:
+            return 4;
+        case PixelFormat::RGB_888:
+            return 3;
+        default:
+            return -1;
+    }
+}
+
+void ReadbackHelper::fillBuffer(int32_t width, int32_t height, uint32_t stride, void* bufferData,
+                                PixelFormat pixelFormat,
+                                std::vector<IComposerClient::Color> desiredPixelColors) {
+    ASSERT_TRUE(pixelFormat == PixelFormat::RGB_888 || pixelFormat == PixelFormat::RGBA_8888);
+    int32_t bytesPerPixel = GetBytesPerPixel(pixelFormat);
+    ASSERT_NE(-1, bytesPerPixel);
+    for (int row = 0; row < height; row++) {
+        for (int col = 0; col < width; col++) {
+            int pixel = row * width + col;
+            IComposerClient::Color srcColor = desiredPixelColors[pixel];
+
+            int offset = (row * stride + col) * bytesPerPixel;
+            uint8_t* pixelColor = (uint8_t*)bufferData + offset;
+            pixelColor[0] = srcColor.r;
+            pixelColor[1] = srcColor.g;
+            pixelColor[2] = srcColor.b;
+
+            if (bytesPerPixel == 4) {
+                pixelColor[3] = srcColor.a;
+            }
+        }
+    }
+}
+
+void ReadbackHelper::clearColors(std::vector<IComposerClient::Color>& expectedColors, int32_t width,
+                                 int32_t height, int32_t displayWidth) {
+    for (int row = 0; row < height; row++) {
+        for (int col = 0; col < width; col++) {
+            int pixel = row * displayWidth + col;
+            expectedColors[pixel] = BLACK;
+        }
+    }
+}
+
+void ReadbackHelper::fillColorsArea(std::vector<IComposerClient::Color>& expectedColors,
+                                    int32_t stride, IComposerClient::Rect area,
+                                    IComposerClient::Color color) {
+    for (int row = area.top; row < area.bottom; row++) {
+        for (int col = area.left; col < area.right; col++) {
+            int pixel = row * stride + col;
+            expectedColors[pixel] = color;
+        }
+    }
+}
+
+bool ReadbackHelper::readbackSupported(const PixelFormat& pixelFormat, const Dataspace& dataspace,
+                                       const Error error) {
+    if (error != Error::NONE) {
+        return false;
+    }
+    // TODO: add support for RGBA_1010102
+    if (pixelFormat != PixelFormat::RGB_888 && pixelFormat != PixelFormat::RGBA_8888) {
+        return false;
+    }
+    if (dataspace != Dataspace::V0_SRGB) {
+        return false;
+    }
+    return true;
+}
+
+ReadbackBuffer::ReadbackBuffer(Display display, const std::shared_ptr<ComposerClient>& client,
+                               const std::shared_ptr<Gralloc>& gralloc, uint32_t width,
+                               uint32_t height, PixelFormat pixelFormat, Dataspace dataspace) {
+    mDisplay = display;
+
+    mComposerClient = client;
+    mGralloc = gralloc;
+
+    mPixelFormat = pixelFormat;
+    mDataspace = dataspace;
+
+    mWidth = width;
+    mHeight = height;
+    mLayerCount = 1;
+    mFormat = mPixelFormat;
+    mUsage = static_cast<uint64_t>(BufferUsage::CPU_READ_OFTEN | BufferUsage::GPU_TEXTURE);
+
+    mAccessRegion.top = 0;
+    mAccessRegion.left = 0;
+    mAccessRegion.width = width;
+    mAccessRegion.height = height;
+}
+
+ReadbackBuffer::~ReadbackBuffer() {
+    if (mBufferHandle != nullptr) {
+        mGralloc->freeBuffer(mBufferHandle);
+    }
+}
+
+void ReadbackBuffer::setReadbackBuffer() {
+    if (mBufferHandle != nullptr) {
+        mGralloc->freeBuffer(mBufferHandle);
+        mBufferHandle = nullptr;
+    }
+    mBufferHandle = mGralloc->allocate(mWidth, mHeight, mLayerCount, mFormat, mUsage,
+                                       /*import*/ true, &mStride);
+    ASSERT_NE(false, mGralloc->validateBufferSize(mBufferHandle, mWidth, mHeight, mLayerCount,
+                                                  mFormat, mUsage, mStride));
+    ASSERT_NO_FATAL_FAILURE(mComposerClient->setReadbackBuffer(mDisplay, mBufferHandle, -1));
+}
+
+void ReadbackBuffer::checkReadbackBuffer(std::vector<IComposerClient::Color> expectedColors) {
+    // lock buffer for reading
+    int32_t fenceHandle;
+    ASSERT_NO_FATAL_FAILURE(mComposerClient->getReadbackBufferFence(mDisplay, &fenceHandle));
+
+    void* bufData = mGralloc->lock(mBufferHandle, mUsage, mAccessRegion, fenceHandle);
+    ASSERT_TRUE(mPixelFormat == PixelFormat::RGB_888 || mPixelFormat == PixelFormat::RGBA_8888);
+    int32_t bytesPerPixel = ReadbackHelper::GetBytesPerPixel(mPixelFormat);
+    ASSERT_NE(-1, bytesPerPixel);
+    for (int row = 0; row < mHeight; row++) {
+        for (int col = 0; col < mWidth; col++) {
+            int pixel = row * mWidth + col;
+            int offset = (row * mStride + col) * bytesPerPixel;
+            uint8_t* pixelColor = (uint8_t*)bufData + offset;
+
+            ASSERT_EQ(expectedColors[pixel].r, pixelColor[0]);
+            ASSERT_EQ(expectedColors[pixel].g, pixelColor[1]);
+            ASSERT_EQ(expectedColors[pixel].b, pixelColor[2]);
+        }
+    }
+    int32_t unlockFence = mGralloc->unlock(mBufferHandle);
+    if (unlockFence != -1) {
+        sync_wait(unlockFence, -1);
+        close(unlockFence);
+    }
+}
+
+void TestColorLayer::write(const std::shared_ptr<CommandWriterBase>& writer) {
+    TestLayer::write(writer);
+    writer->setLayerCompositionType(IComposerClient::Composition::SOLID_COLOR);
+    writer->setLayerColor(mColor);
+}
+
+TestBufferLayer::TestBufferLayer(const std::shared_ptr<ComposerClient>& client,
+                                 const std::shared_ptr<Gralloc>& gralloc, Display display,
+                                 int32_t width, int32_t height, PixelFormat format,
+                                 IComposerClient::Composition composition)
+    : TestLayer{client, display} {
+    mGralloc = gralloc;
+    mComposition = composition;
+    mWidth = width;
+    mHeight = height;
+    mLayerCount = 1;
+    mFormat = format;
+    mUsage = static_cast<uint64_t>(BufferUsage::CPU_READ_OFTEN | BufferUsage::CPU_WRITE_OFTEN |
+                                   BufferUsage::COMPOSER_OVERLAY);
+
+    mAccessRegion.top = 0;
+    mAccessRegion.left = 0;
+    mAccessRegion.width = width;
+    mAccessRegion.height = height;
+
+    setSourceCrop({0, 0, (float)width, (float)height});
+}
+
+TestBufferLayer::~TestBufferLayer() {
+    if (mBufferHandle != nullptr) {
+        mGralloc->freeBuffer(mBufferHandle);
+    }
+}
+
+void TestBufferLayer::write(const std::shared_ptr<CommandWriterBase>& writer) {
+    TestLayer::write(writer);
+    writer->setLayerCompositionType(mComposition);
+    writer->setLayerDataspace(Dataspace::UNKNOWN);
+    writer->setLayerVisibleRegion(std::vector<IComposerClient::Rect>(1, mDisplayFrame));
+    if (mBufferHandle != nullptr) writer->setLayerBuffer(0, mBufferHandle, mFillFence);
+}
+
+void TestBufferLayer::fillBuffer(std::vector<IComposerClient::Color> expectedColors) {
+    void* bufData = mGralloc->lock(mBufferHandle, mUsage, mAccessRegion, -1);
+    ASSERT_NO_FATAL_FAILURE(
+            ReadbackHelper::fillBuffer(mWidth, mHeight, mStride, bufData, mFormat, expectedColors));
+    mFillFence = mGralloc->unlock(mBufferHandle);
+    if (mFillFence != -1) {
+        sync_wait(mFillFence, -1);
+        close(mFillFence);
+    }
+}
+void TestBufferLayer::setBuffer(std::vector<IComposerClient::Color> colors) {
+    if (mBufferHandle != nullptr) {
+        mGralloc->freeBuffer(mBufferHandle);
+        mBufferHandle = nullptr;
+    }
+    mBufferHandle = mGralloc->allocate(mWidth, mHeight, mLayerCount, mFormat, mUsage,
+                                       /*import*/ true, &mStride);
+    ASSERT_NE(nullptr, mBufferHandle);
+    ASSERT_NO_FATAL_FAILURE(fillBuffer(colors));
+    ASSERT_NE(false, mGralloc->validateBufferSize(mBufferHandle, mWidth, mHeight, mLayerCount,
+                                                  mFormat, mUsage, mStride));
+}
+
+void TestBufferLayer::setToClientComposition(const std::shared_ptr<CommandWriterBase>& writer) {
+    writer->selectLayer(mLayer);
+    writer->setLayerCompositionType(IComposerClient::Composition::CLIENT);
+}
+
+}  // namespace vts
+}  // namespace V2_2
+}  // namespace composer
+}  // namespace graphics
+}  // namespace hardware
+}  // namespace android
diff --git a/graphics/composer/2.2/utils/vts/include/composer-vts/2.2/ComposerVts.h b/graphics/composer/2.2/utils/vts/include/composer-vts/2.2/ComposerVts.h
index 2633021..8fa9b7b 100644
--- a/graphics/composer/2.2/utils/vts/include/composer-vts/2.2/ComposerVts.h
+++ b/graphics/composer/2.2/utils/vts/include/composer-vts/2.2/ComposerVts.h
@@ -27,6 +27,7 @@
 #include <android/hardware/graphics/composer/2.2/IComposerClient.h>
 #include <composer-command-buffer/2.2/ComposerCommandBuffer.h>
 #include <composer-vts/2.1/ComposerVts.h>
+#include <mapper-vts/2.1/MapperVts.h>
 #include <utils/StrongPointer.h>
 
 namespace android {
@@ -41,6 +42,11 @@
 using common::V1_1::Dataspace;
 using common::V1_1::PixelFormat;
 using common::V1_1::RenderIntent;
+using IMapper2_1 = android::hardware::graphics::mapper::V2_1::IMapper;
+using IMapper3 = android::hardware::graphics::mapper::V3_0::IMapper;
+using Gralloc2 = android::hardware::graphics::mapper::V2_0::vts::Gralloc;
+using Gralloc2_1 = android::hardware::graphics::mapper::V2_1::vts::Gralloc;
+using Gralloc3 = android::hardware::graphics::mapper::V3_0::vts::Gralloc;
 
 class ComposerClient;
 
@@ -84,6 +90,26 @@
     const sp<IComposerClient> mClient;
 };
 
+class Gralloc : public V2_1::vts::Gralloc {
+  public:
+    Gralloc();
+    const native_handle_t* allocate(uint32_t width, uint32_t height, uint32_t layerCount,
+                                    PixelFormat format, uint64_t usage, bool import = true,
+                                    uint32_t* outStride = nullptr) {
+        return V2_1::vts::Gralloc::allocate(
+                width, height, layerCount,
+                static_cast<android::hardware::graphics::common::V1_0::PixelFormat>(format), usage,
+                import, outStride);
+    }
+
+    bool validateBufferSize(const native_handle_t* bufferHandle, uint32_t width, uint32_t height,
+                            uint32_t layerCount, PixelFormat format, uint64_t usage,
+                            uint32_t stride);
+
+  protected:
+    std::shared_ptr<Gralloc2_1> mGralloc2_1 = nullptr;
+};
+
 }  // namespace vts
 }  // namespace V2_2
 }  // namespace composer
diff --git a/graphics/composer/2.2/utils/vts/include/composer-vts/2.2/ReadbackVts.h b/graphics/composer/2.2/utils/vts/include/composer-vts/2.2/ReadbackVts.h
new file mode 100644
index 0000000..7e93167
--- /dev/null
+++ b/graphics/composer/2.2/utils/vts/include/composer-vts/2.2/ReadbackVts.h
@@ -0,0 +1,184 @@
+/*
+ * Copyright 2019 The Android Open Source Project
+ *
+ * Licensed under the Apache License, Version 2.0 (the "License");
+ * you may not use this file except in compliance with the License.
+ * You may obtain a copy of the License at
+ *
+ *      http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+#include <android-base/unique_fd.h>
+#include <android/hardware/graphics/composer/2.2/IComposerClient.h>
+#include <composer-command-buffer/2.2/ComposerCommandBuffer.h>
+#include <composer-vts/2.1/GraphicsComposerCallback.h>
+#include <composer-vts/2.1/TestCommandReader.h>
+#include <composer-vts/2.2/ComposerVts.h>
+#include <mapper-vts/2.1/MapperVts.h>
+
+namespace android {
+namespace hardware {
+namespace graphics {
+namespace composer {
+namespace V2_2 {
+namespace vts {
+
+using android::hardware::hidl_handle;
+using common::V1_1::BufferUsage;
+using common::V1_1::Dataspace;
+using common::V1_1::PixelFormat;
+using IMapper2_1 = mapper::V2_1::IMapper;
+using Gralloc2_1 = mapper::V2_1::vts::Gralloc;
+using V2_1::Display;
+using V2_1::Layer;
+using V2_1::vts::AccessRegion;
+using V2_1::vts::TestCommandReader;
+
+static const IComposerClient::Color BLACK = {0, 0, 0, 0xff};
+static const IComposerClient::Color RED = {0xff, 0, 0, 0xff};
+static const IComposerClient::Color TRANSLUCENT_RED = {0xff, 0, 0, 0x33};
+static const IComposerClient::Color GREEN = {0, 0xff, 0, 0xff};
+static const IComposerClient::Color BLUE = {0, 0, 0xff, 0xff};
+
+class TestLayer {
+  public:
+    TestLayer(const std::shared_ptr<ComposerClient>& client, Display display)
+        : mLayer(client->createLayer(display, kBufferSlotCount)), mComposerClient(client) {}
+
+    // ComposerClient will take care of destroying layers, no need to explicitly
+    // call destroyLayers here
+    virtual ~TestLayer(){};
+
+    virtual void write(const std::shared_ptr<CommandWriterBase>& writer);
+
+    void setDisplayFrame(IComposerClient::Rect frame) { mDisplayFrame = frame; }
+    void setSourceCrop(IComposerClient::FRect crop) { mSourceCrop = crop; }
+    void setZOrder(uint32_t z) { mZOrder = z; }
+
+    void setSurfaceDamage(std::vector<IComposerClient::Rect> surfaceDamage) {
+        mSurfaceDamage = surfaceDamage;
+    }
+
+    void setTransform(Transform transform) { mTransform = transform; }
+    void setAlpha(float alpha) { mAlpha = alpha; }
+    void setBlendMode(IComposerClient::BlendMode blendMode) { mBlendMode = blendMode; }
+
+    static constexpr uint32_t kBufferSlotCount = 64;
+
+    IComposerClient::Rect mDisplayFrame = {0, 0, 0, 0};
+    uint32_t mZOrder = 0;
+    std::vector<IComposerClient::Rect> mSurfaceDamage;
+    Transform mTransform = static_cast<Transform>(0);
+    IComposerClient::FRect mSourceCrop = {0, 0, 0, 0};
+    float mAlpha = 1.0;
+    IComposerClient::BlendMode mBlendMode = IComposerClient::BlendMode::NONE;
+
+  protected:
+    Layer mLayer;
+
+  private:
+    std::shared_ptr<ComposerClient> const mComposerClient;
+};
+
+class TestColorLayer : public TestLayer {
+  public:
+    TestColorLayer(const std::shared_ptr<ComposerClient>& client, Display display)
+        : TestLayer{client, display} {}
+
+    void write(const std::shared_ptr<CommandWriterBase>& writer) override;
+
+    void setColor(IComposerClient::Color color) { mColor = color; }
+
+  private:
+    IComposerClient::Color mColor = {0xff, 0xff, 0xff, 0xff};
+};
+
+class TestBufferLayer : public TestLayer {
+  public:
+    TestBufferLayer(
+            const std::shared_ptr<ComposerClient>& client, const std::shared_ptr<Gralloc>& gralloc,
+            Display display, int32_t width, int32_t height, PixelFormat format,
+            IComposerClient::Composition composition = IComposerClient::Composition::DEVICE);
+
+    ~TestBufferLayer();
+
+    void write(const std::shared_ptr<CommandWriterBase>& writer) override;
+
+    void fillBuffer(std::vector<IComposerClient::Color> expectedColors);
+
+    void setBuffer(std::vector<IComposerClient::Color> colors);
+
+    void setToClientComposition(const std::shared_ptr<CommandWriterBase>& writer);
+
+    uint32_t mWidth;
+    uint32_t mHeight;
+    uint32_t mLayerCount;
+    PixelFormat mFormat;
+    uint64_t mUsage;
+    AccessRegion mAccessRegion;
+    uint32_t mStride;
+
+  protected:
+    IComposerClient::Composition mComposition;
+    std::shared_ptr<Gralloc> mGralloc;
+    int32_t mFillFence;
+    const native_handle_t* mBufferHandle = nullptr;
+};
+
+class ReadbackHelper : public ::testing::VtsHalHidlTargetTestBase {
+  public:
+    static int32_t GetBytesPerPixel(PixelFormat pixelFormat);
+
+    static void fillBuffer(int32_t width, int32_t height, uint32_t stride, void* bufferData,
+                           PixelFormat pixelFormat,
+                           std::vector<IComposerClient::Color> desiredPixelColors);
+
+    static void clearColors(std::vector<IComposerClient::Color>& expectedColors, int32_t width,
+                            int32_t height, int32_t displayWidth);
+
+    static void fillColorsArea(std::vector<IComposerClient::Color>& expectedColors, int32_t stride,
+                               IComposerClient::Rect area, IComposerClient::Color color);
+
+    static bool readbackSupported(const PixelFormat& pixelFormat, const Dataspace& dataspace,
+                                  const Error error);
+};
+
+class ReadbackBuffer {
+  public:
+    ReadbackBuffer(Display display, const std::shared_ptr<ComposerClient>& client,
+                   const std::shared_ptr<Gralloc>& gralloc, uint32_t width, uint32_t height,
+                   PixelFormat pixelFormat, Dataspace dataspace);
+    ~ReadbackBuffer();
+
+    void setReadbackBuffer();
+
+    void checkReadbackBuffer(std::vector<IComposerClient::Color> expectedColors);
+
+  protected:
+    uint32_t mWidth;
+    uint32_t mHeight;
+    uint32_t mLayerCount;
+    PixelFormat mFormat;
+    uint64_t mUsage;
+    AccessRegion mAccessRegion;
+    uint32_t mStride;
+    const native_handle_t* mBufferHandle = nullptr;
+    PixelFormat mPixelFormat;
+    Dataspace mDataspace;
+    Display mDisplay;
+    std::shared_ptr<Gralloc> mGralloc;
+    std::shared_ptr<ComposerClient> mComposerClient;
+};
+
+}  // namespace vts
+}  // namespace V2_2
+}  // namespace composer
+}  // namespace graphics
+}  // namespace hardware
+}  // namespace android
diff --git a/graphics/composer/2.2/vts/functional/Android.bp b/graphics/composer/2.2/vts/functional/Android.bp
index 4cb0bdf..9f7e1cd 100644
--- a/graphics/composer/2.2/vts/functional/Android.bp
+++ b/graphics/composer/2.2/vts/functional/Android.bp
@@ -31,6 +31,7 @@
     ],
     static_libs: [
         "android.hardware.graphics.allocator@2.0",
+        "android.hardware.graphics.allocator@3.0",
         "android.hardware.graphics.common@1.1",
         "android.hardware.graphics.composer@2.1",
         "android.hardware.graphics.composer@2.1-vts",
@@ -40,6 +41,8 @@
         "android.hardware.graphics.mapper@2.0-vts",
         "android.hardware.graphics.mapper@2.1",
         "android.hardware.graphics.mapper@2.1-vts",
+        "android.hardware.graphics.mapper@3.0",
+        "android.hardware.graphics.mapper@3.0-vts",
     ],
     header_libs: [
         "android.hardware.graphics.composer@2.1-command-buffer",
diff --git a/graphics/composer/2.2/vts/functional/VtsHalGraphicsComposerV2_2ReadbackTest.cpp b/graphics/composer/2.2/vts/functional/VtsHalGraphicsComposerV2_2ReadbackTest.cpp
index da8858e..eef0433 100644
--- a/graphics/composer/2.2/vts/functional/VtsHalGraphicsComposerV2_2ReadbackTest.cpp
+++ b/graphics/composer/2.2/vts/functional/VtsHalGraphicsComposerV2_2ReadbackTest.cpp
@@ -18,13 +18,10 @@
 
 #include <VtsHalHidlTargetTestBase.h>
 #include <VtsHalHidlTargetTestEnvBase.h>
-#include <android-base/unique_fd.h>
-#include <android/hardware/graphics/composer/2.2/IComposerClient.h>
 #include <composer-command-buffer/2.2/ComposerCommandBuffer.h>
 #include <composer-vts/2.1/GraphicsComposerCallback.h>
 #include <composer-vts/2.1/TestCommandReader.h>
-#include <composer-vts/2.2/ComposerVts.h>
-#include <mapper-vts/2.1/MapperVts.h>
+#include <composer-vts/2.2/ReadbackVts.h>
 
 namespace android {
 namespace hardware {
@@ -39,17 +36,10 @@
 using common::V1_1::Dataspace;
 using common::V1_1::PixelFormat;
 using mapper::V2_1::IMapper;
-using mapper::V2_1::vts::Gralloc;
 using V2_1::Display;
 using V2_1::Layer;
 using V2_1::vts::TestCommandReader;
 
-static const IComposerClient::Color BLACK = {0, 0, 0, 0xff};
-static const IComposerClient::Color RED = {0xff, 0, 0, 0xff};
-static const IComposerClient::Color TRANSLUCENT_RED = {0xff, 0, 0, 0x33};
-static const IComposerClient::Color GREEN = {0, 0xff, 0, 0xff};
-static const IComposerClient::Color BLUE = {0, 0, 0xff, 0xff};
-
 // Test environment for graphics.composer
 class GraphicsComposerHidlEnvironment : public ::testing::VtsHalHidlTargetTestEnvBase {
    public:
@@ -65,92 +55,7 @@
     GTEST_DISALLOW_COPY_AND_ASSIGN_(GraphicsComposerHidlEnvironment);
 };
 
-class TestLayer {
-   public:
-    TestLayer(const std::shared_ptr<ComposerClient>& client, Display display)
-        : mLayer(client->createLayer(display, kBufferSlotCount)), mComposerClient(client) {}
-
-    // ComposerClient will take care of destroying layers, no need to explicitly
-    // call destroyLayers here
-    virtual ~TestLayer(){};
-
-    virtual void write(const std::shared_ptr<CommandWriterBase>& writer) {
-        writer->selectLayer(mLayer);
-        writer->setLayerDisplayFrame(mDisplayFrame);
-        writer->setLayerSourceCrop(mSourceCrop);
-        writer->setLayerZOrder(mZOrder);
-        writer->setLayerSurfaceDamage(mSurfaceDamage);
-        writer->setLayerTransform(mTransform);
-        writer->setLayerPlaneAlpha(mAlpha);
-        writer->setLayerBlendMode(mBlendMode);
-    }
-
-    void setDisplayFrame(IComposerClient::Rect frame) { mDisplayFrame = frame; }
-    void setSourceCrop(IComposerClient::FRect crop) { mSourceCrop = crop; }
-    void setZOrder(uint32_t z) { mZOrder = z; }
-
-    void setSurfaceDamage(std::vector<IComposerClient::Rect> surfaceDamage) {
-        mSurfaceDamage = surfaceDamage;
-    }
-
-    void setTransform(Transform transform) { mTransform = transform; }
-    void setAlpha(float alpha) { mAlpha = alpha; }
-    void setBlendMode(IComposerClient::BlendMode blendMode) { mBlendMode = blendMode; }
-
-    static constexpr uint32_t kBufferSlotCount = 64;
-
-    IComposerClient::Rect mDisplayFrame = {0, 0, 0, 0};
-    uint32_t mZOrder = 0;
-    std::vector<IComposerClient::Rect> mSurfaceDamage;
-    Transform mTransform = static_cast<Transform>(0);
-    IComposerClient::FRect mSourceCrop = {0, 0, 0, 0};
-    float mAlpha = 1.0;
-    IComposerClient::BlendMode mBlendMode = IComposerClient::BlendMode::NONE;
-
-   protected:
-    Layer mLayer;
-
-   private:
-    std::shared_ptr<ComposerClient> const mComposerClient;
-};
-
 class GraphicsComposerReadbackTest : public ::testing::VtsHalHidlTargetTestBase {
-   public:
-    static int32_t GetBytesPerPixel(PixelFormat pixelFormat) {
-        switch (pixelFormat) {
-            case PixelFormat::RGBA_8888:
-                return 4;
-            case PixelFormat::RGB_888:
-                return 3;
-            default:
-                return -1;
-        }
-    }
-
-    static void fillBuffer(int32_t width, int32_t height, uint32_t stride, void* bufferData,
-                           PixelFormat pixelFormat,
-                           std::vector<IComposerClient::Color> desiredPixelColors) {
-        ASSERT_TRUE(pixelFormat == PixelFormat::RGB_888 || pixelFormat == PixelFormat::RGBA_8888);
-        int32_t bytesPerPixel = GetBytesPerPixel(pixelFormat);
-        ASSERT_NE(-1, bytesPerPixel);
-        for (int row = 0; row < height; row++) {
-            for (int col = 0; col < width; col++) {
-                int pixel = row * width + col;
-                IComposerClient::Color srcColor = desiredPixelColors[pixel];
-
-                int offset = (row * stride + col) * bytesPerPixel;
-                uint8_t* pixelColor = (uint8_t*)bufferData + offset;
-                pixelColor[0] = srcColor.r;
-                pixelColor[1] = srcColor.g;
-                pixelColor[2] = srcColor.b;
-
-                if (bytesPerPixel == 4) {
-                    pixelColor[3] = srcColor.a;
-                }
-            }
-        }
-    }
-
    protected:
     using PowerMode = V2_1::IComposerClient::PowerMode;
     void SetUp() override {
@@ -183,12 +88,13 @@
         mGralloc = std::make_shared<Gralloc>();
 
         mComposerClient->getRaw()->getReadbackBufferAttributes(
-            mPrimaryDisplay,
-            [&](const auto& tmpError, const auto& tmpPixelFormat, const auto& tmpDataspace) {
-                mHasReadbackBuffer = readbackSupported(tmpPixelFormat, tmpDataspace, tmpError);
-                mPixelFormat = tmpPixelFormat;
-                mDataspace = tmpDataspace;
-            });
+                mPrimaryDisplay,
+                [&](const auto& tmpError, const auto& tmpPixelFormat, const auto& tmpDataspace) {
+                    mHasReadbackBuffer = ReadbackHelper::readbackSupported(tmpPixelFormat,
+                                                                           tmpDataspace, tmpError);
+                    mPixelFormat = tmpPixelFormat;
+                    mDataspace = tmpDataspace;
+                });
         ASSERT_NO_FATAL_FAILURE(mComposerClient->setPowerMode(mPrimaryDisplay, PowerMode::ON));
     }
 
@@ -209,10 +115,6 @@
         mReader->mErrors.clear();
     }
 
-    void execute() {
-        ASSERT_NO_FATAL_FAILURE(mComposerClient->execute(mReader.get(), mWriter.get()));
-    }
-
     void writeLayers(const std::vector<std::shared_ptr<TestLayer>>& layers) {
         for (auto layer : layers) {
             layer->write(mWriter);
@@ -220,42 +122,10 @@
         execute();
     }
 
-    void clearColors(std::vector<IComposerClient::Color>& expectedColors, int32_t width,
-                     int32_t height) {
-        for (int row = 0; row < height; row++) {
-            for (int col = 0; col < width; col++) {
-                int pixel = row * mDisplayWidth + col;
-                expectedColors[pixel] = BLACK;
-            }
-        }
+    void execute() {
+        ASSERT_NO_FATAL_FAILURE(mComposerClient->execute(mReader.get(), mWriter.get()));
     }
 
-    void fillColorsArea(std::vector<IComposerClient::Color>& expectedColors, int32_t stride,
-                        IComposerClient::Rect area, IComposerClient::Color color) {
-        for (int row = area.top; row < area.bottom; row++) {
-            for (int col = area.left; col < area.right; col++) {
-                int pixel = row * stride + col;
-                expectedColors[pixel] = color;
-            }
-        }
-    }
-
-    bool readbackSupported(const PixelFormat& pixelFormat, const Dataspace& dataspace,
-                           const Error error) {
-        if (error != Error::NONE) {
-            return false;
-        }
-        // TODO: add support for RGBA_1010102
-        if (pixelFormat != PixelFormat::RGB_888 && pixelFormat != PixelFormat::RGBA_8888) {
-            return false;
-        }
-        if (dataspace != Dataspace::V0_SRGB) {
-            return false;
-        }
-        return true;
-    }
-
-
     std::unique_ptr<Composer> mComposer;
     std::shared_ptr<ComposerClient> mComposerClient;
 
@@ -286,178 +156,6 @@
         }
     }
 };
-class ReadbackBuffer {
-   public:
-    ReadbackBuffer(Display display, const std::shared_ptr<ComposerClient>& client,
-                   const std::shared_ptr<Gralloc>& gralloc, uint32_t width, uint32_t height,
-                   PixelFormat pixelFormat, Dataspace dataspace) {
-        mDisplay = display;
-
-        mComposerClient = client;
-        mGralloc = gralloc;
-
-        mPixelFormat = pixelFormat;
-        mDataspace = dataspace;
-
-        mInfo.width = width;
-        mInfo.height = height;
-        mInfo.layerCount = 1;
-        mInfo.format = mPixelFormat;
-        mInfo.usage = static_cast<uint64_t>(BufferUsage::CPU_READ_OFTEN | BufferUsage::GPU_TEXTURE);
-
-        mAccessRegion.top = 0;
-        mAccessRegion.left = 0;
-        mAccessRegion.width = width;
-        mAccessRegion.height = height;
-    };
-
-    ~ReadbackBuffer() {
-        if (mBufferHandle != nullptr) {
-            mGralloc->freeBuffer(mBufferHandle);
-        }
-    }
-
-    void setReadbackBuffer() {
-        if (mBufferHandle != nullptr) {
-            mGralloc->freeBuffer(mBufferHandle);
-            mBufferHandle = nullptr;
-        }
-        mBufferHandle = mGralloc->allocate(mInfo, /*import*/ true, &mStride);
-        ASSERT_NE(false, mGralloc->validateBufferSize(mBufferHandle, mInfo, mStride));
-        ASSERT_NO_FATAL_FAILURE(mComposerClient->setReadbackBuffer(mDisplay, mBufferHandle, -1));
-    }
-
-    void checkReadbackBuffer(std::vector<IComposerClient::Color> expectedColors) {
-        // lock buffer for reading
-        int32_t fenceHandle;
-        ASSERT_NO_FATAL_FAILURE(mComposerClient->getReadbackBufferFence(mDisplay, &fenceHandle));
-
-        void* bufData = mGralloc->lock(mBufferHandle, mInfo.usage, mAccessRegion, fenceHandle);
-        ASSERT_TRUE(mPixelFormat == PixelFormat::RGB_888 || mPixelFormat == PixelFormat::RGBA_8888);
-        int32_t bytesPerPixel = GraphicsComposerReadbackTest::GetBytesPerPixel(mPixelFormat);
-        ASSERT_NE(-1, bytesPerPixel);
-        for (int row = 0; row < mInfo.height; row++) {
-            for (int col = 0; col < mInfo.width; col++) {
-                int pixel = row * mInfo.width + col;
-                int offset = (row * mStride + col) * bytesPerPixel;
-                uint8_t* pixelColor = (uint8_t*)bufData + offset;
-
-                ASSERT_EQ(expectedColors[pixel].r, pixelColor[0]);
-                ASSERT_EQ(expectedColors[pixel].g, pixelColor[1]);
-                ASSERT_EQ(expectedColors[pixel].b, pixelColor[2]);
-            }
-        }
-        int32_t unlockFence = mGralloc->unlock(mBufferHandle);
-        if (unlockFence != -1) {
-            sync_wait(unlockFence, -1);
-            close(unlockFence);
-        }
-    }
-
-   protected:
-    IMapper::BufferDescriptorInfo mInfo;
-    IMapper::Rect mAccessRegion;
-    uint32_t mStride;
-    const native_handle_t* mBufferHandle = nullptr;
-    PixelFormat mPixelFormat;
-    Dataspace mDataspace;
-    Display mDisplay;
-    std::shared_ptr<Gralloc> mGralloc;
-    std::shared_ptr<ComposerClient> mComposerClient;
-};
-
-class TestColorLayer : public TestLayer {
-   public:
-    TestColorLayer(const std::shared_ptr<ComposerClient>& client, Display display)
-        : TestLayer{client, display} {}
-
-    void write(const std::shared_ptr<CommandWriterBase>& writer) override {
-        TestLayer::write(writer);
-        writer->setLayerCompositionType(IComposerClient::Composition::SOLID_COLOR);
-        writer->setLayerColor(mColor);
-    }
-
-    void setColor(IComposerClient::Color color) { mColor = color; }
-
-   private:
-    IComposerClient::Color mColor = {0xff, 0xff, 0xff, 0xff};
-};
-
-class TestBufferLayer : public TestLayer {
-   public:
-    TestBufferLayer(const std::shared_ptr<ComposerClient>& client,
-                    const std::shared_ptr<Gralloc>& gralloc, Display display, int32_t width,
-                    int32_t height, PixelFormat format,
-                    IComposerClient::Composition composition = IComposerClient::Composition::DEVICE)
-        : TestLayer{client, display} {
-        mGralloc = gralloc;
-        mComposition = composition;
-        mInfo.width = width;
-        mInfo.height = height;
-        mInfo.layerCount = 1;
-        mInfo.format = format;
-        mInfo.usage =
-            static_cast<uint64_t>(BufferUsage::CPU_READ_OFTEN | BufferUsage::CPU_WRITE_OFTEN |
-                                  BufferUsage::COMPOSER_OVERLAY);
-
-        mAccessRegion.top = 0;
-        mAccessRegion.left = 0;
-        mAccessRegion.width = width;
-        mAccessRegion.height = height;
-
-        setSourceCrop({0, 0, (float)width, (float)height});
-    }
-
-    ~TestBufferLayer() {
-        if (mBufferHandle != nullptr) {
-            mGralloc->freeBuffer(mBufferHandle);
-        }
-    }
-
-    void write(const std::shared_ptr<CommandWriterBase>& writer) override {
-        TestLayer::write(writer);
-        writer->setLayerCompositionType(mComposition);
-        writer->setLayerDataspace(Dataspace::UNKNOWN);
-        writer->setLayerVisibleRegion(std::vector<IComposerClient::Rect>(1, mDisplayFrame));
-        if (mBufferHandle != nullptr) writer->setLayerBuffer(0, mBufferHandle, mFillFence);
-    }
-
-    void fillBuffer(std::vector<IComposerClient::Color> expectedColors) {
-        void* bufData = mGralloc->lock(mBufferHandle, mInfo.usage, mAccessRegion, -1);
-        ASSERT_NO_FATAL_FAILURE(GraphicsComposerReadbackTest::fillBuffer(
-            mInfo.width, mInfo.height, mStride, bufData, mInfo.format, expectedColors));
-        mFillFence = mGralloc->unlock(mBufferHandle);
-        if (mFillFence != -1) {
-            sync_wait(mFillFence, -1);
-            close(mFillFence);
-        }
-    }
-    void setBuffer(std::vector<IComposerClient::Color> colors) {
-        if (mBufferHandle != nullptr) {
-            mGralloc->freeBuffer(mBufferHandle);
-            mBufferHandle = nullptr;
-        }
-        mBufferHandle = mGralloc->allocate(mInfo, /*import*/ true, &mStride);
-        ASSERT_NE(nullptr, mBufferHandle);
-        ASSERT_NO_FATAL_FAILURE(fillBuffer(colors));
-        ASSERT_NE(false, mGralloc->validateBufferSize(mBufferHandle, mInfo, mStride));
-    }
-
-    void setToClientComposition(const std::shared_ptr<CommandWriterBase>& writer) {
-        writer->selectLayer(mLayer);
-        writer->setLayerCompositionType(IComposerClient::Composition::CLIENT);
-    }
-
-    IMapper::BufferDescriptorInfo mInfo;
-    IMapper::Rect mAccessRegion;
-    uint32_t mStride;
-
-   protected:
-    IComposerClient::Composition mComposition;
-    std::shared_ptr<Gralloc> mGralloc;
-    int32_t mFillFence;
-    const native_handle_t* mBufferHandle = nullptr;
-};
 
 TEST_F(GraphicsComposerReadbackTest, SingleSolidColorLayer) {
     if (!mHasReadbackBuffer) {
@@ -478,7 +176,7 @@
 
     // expected color for each pixel
     std::vector<IComposerClient::Color> expectedColors(mDisplayWidth * mDisplayHeight);
-    fillColorsArea(expectedColors, mDisplayWidth, coloredSquare, BLUE);
+    ReadbackHelper::fillColorsArea(expectedColors, mDisplayWidth, coloredSquare, BLUE);
 
     ReadbackBuffer readbackBuffer(mPrimaryDisplay, mComposerClient, mGralloc, mDisplayWidth,
                                   mDisplayHeight, mPixelFormat, mDataspace);
@@ -515,11 +213,13 @@
                                   mDisplayHeight, mPixelFormat, mDataspace);
     ASSERT_NO_FATAL_FAILURE(readbackBuffer.setReadbackBuffer());
     std::vector<IComposerClient::Color> expectedColors(mDisplayWidth * mDisplayHeight);
-    fillColorsArea(expectedColors, mDisplayWidth, {0, 0, mDisplayWidth, mDisplayHeight / 4}, RED);
-    fillColorsArea(expectedColors, mDisplayWidth,
-                   {0, mDisplayHeight / 4, mDisplayWidth, mDisplayHeight / 2}, GREEN);
-    fillColorsArea(expectedColors, mDisplayWidth,
-                   {0, mDisplayHeight / 2, mDisplayWidth, mDisplayHeight}, BLUE);
+    ReadbackHelper::fillColorsArea(expectedColors, mDisplayWidth,
+                                   {0, 0, mDisplayWidth, mDisplayHeight / 4}, RED);
+    ReadbackHelper::fillColorsArea(expectedColors, mDisplayWidth,
+                                   {0, mDisplayHeight / 4, mDisplayWidth, mDisplayHeight / 2},
+                                   GREEN);
+    ReadbackHelper::fillColorsArea(expectedColors, mDisplayWidth,
+                                   {0, mDisplayHeight / 2, mDisplayWidth, mDisplayHeight}, BLUE);
 
     auto layer =
         std::make_shared<TestBufferLayer>(mComposerClient, mGralloc, mPrimaryDisplay, mDisplayWidth,
@@ -568,19 +268,16 @@
     layer->write(mWriter);
 
     // This following buffer call should have no effect
-    IMapper::BufferDescriptorInfo bufferInfo{};
-    bufferInfo.width = mDisplayWidth;
-    bufferInfo.height = mDisplayHeight;
-    bufferInfo.layerCount = 1;
-    bufferInfo.format = PixelFormat::RGBA_8888;
-    bufferInfo.usage =
-        static_cast<uint64_t>(BufferUsage::CPU_READ_OFTEN | BufferUsage::CPU_WRITE_OFTEN);
-    const native_handle_t* bufferHandle = mGralloc->allocate(bufferInfo);
+    PixelFormat format = PixelFormat::RGBA_8888;
+    uint64_t usage =
+            static_cast<uint64_t>(BufferUsage::CPU_READ_OFTEN | BufferUsage::CPU_WRITE_OFTEN);
+    const native_handle_t* bufferHandle =
+            mGralloc->allocate(mDisplayWidth, mDisplayHeight, 1, format, usage);
     mWriter->setLayerBuffer(0, bufferHandle, -1);
 
     // expected color for each pixel
     std::vector<IComposerClient::Color> expectedColors(mDisplayWidth * mDisplayHeight);
-    fillColorsArea(expectedColors, mDisplayWidth, coloredSquare, BLUE);
+    ReadbackHelper::fillColorsArea(expectedColors, mDisplayWidth, coloredSquare, BLUE);
 
     ReadbackBuffer readbackBuffer(mPrimaryDisplay, mComposerClient, mGralloc, mDisplayWidth,
                                   mDisplayHeight, mPixelFormat, mDataspace);
@@ -611,11 +308,13 @@
     mWriter->selectDisplay(mPrimaryDisplay);
 
     std::vector<IComposerClient::Color> expectedColors(mDisplayWidth * mDisplayHeight);
-    fillColorsArea(expectedColors, mDisplayWidth, {0, 0, mDisplayWidth, mDisplayHeight / 4}, RED);
-    fillColorsArea(expectedColors, mDisplayWidth,
-                   {0, mDisplayHeight / 4, mDisplayWidth, mDisplayHeight / 2}, GREEN);
-    fillColorsArea(expectedColors, mDisplayWidth,
-                   {0, mDisplayHeight / 2, mDisplayWidth, mDisplayHeight}, BLUE);
+    ReadbackHelper::fillColorsArea(expectedColors, mDisplayWidth,
+                                   {0, 0, mDisplayWidth, mDisplayHeight / 4}, RED);
+    ReadbackHelper::fillColorsArea(expectedColors, mDisplayWidth,
+                                   {0, mDisplayHeight / 4, mDisplayWidth, mDisplayHeight / 2},
+                                   GREEN);
+    ReadbackHelper::fillColorsArea(expectedColors, mDisplayWidth,
+                                   {0, mDisplayHeight / 2, mDisplayWidth, mDisplayHeight}, BLUE);
 
     auto layer =
         std::make_shared<TestBufferLayer>(mComposerClient, mGralloc, mPrimaryDisplay, mDisplayWidth,
@@ -642,23 +341,21 @@
 
         // create client target buffer
         uint32_t clientStride;
-        IMapper::BufferDescriptorInfo clientInfo;
-        clientInfo.width = layer->mInfo.width;
-        clientInfo.height = layer->mInfo.height;
-        clientInfo.layerCount = layer->mInfo.layerCount;
-        clientInfo.format = PixelFormat::RGBA_8888;
-        clientInfo.usage =
-            static_cast<uint64_t>(BufferUsage::CPU_READ_OFTEN | BufferUsage::CPU_WRITE_OFTEN |
-                                  BufferUsage::COMPOSER_CLIENT_TARGET);
+        PixelFormat clientFormat = PixelFormat::RGBA_8888;
+        uint64_t clientUsage =
+                static_cast<uint64_t>(BufferUsage::CPU_READ_OFTEN | BufferUsage::CPU_WRITE_OFTEN |
+                                      BufferUsage::COMPOSER_CLIENT_TARGET);
         const native_handle_t* clientBufferHandle =
-            mGralloc->allocate(clientInfo, /*import*/ true, &clientStride);
+                mGralloc->allocate(layer->mWidth, layer->mHeight, layer->mLayerCount, clientFormat,
+                                   clientUsage, /*import*/ true, &clientStride);
         ASSERT_NE(nullptr, clientBufferHandle);
 
         void* clientBufData =
-            mGralloc->lock(clientBufferHandle, clientInfo.usage, layer->mAccessRegion, -1);
+                mGralloc->lock(clientBufferHandle, clientUsage, layer->mAccessRegion, -1);
 
-        ASSERT_NO_FATAL_FAILURE(fillBuffer(clientInfo.width, clientInfo.height, clientStride,
-                                           clientBufData, clientInfo.format, expectedColors));
+        ASSERT_NO_FATAL_FAILURE(ReadbackHelper::fillBuffer(layer->mWidth, layer->mHeight,
+                                                           clientStride, clientBufData,
+                                                           clientFormat, expectedColors));
         int clientFence = mGralloc->unlock(clientBufferHandle);
         if (clientFence != -1) {
             sync_wait(clientFence, -1);
@@ -695,9 +392,10 @@
         mComposerClient->setClientTargetSlotCount(mPrimaryDisplay, kClientTargetSlotCount));
 
     std::vector<IComposerClient::Color> expectedColors(mDisplayWidth * mDisplayHeight);
-    fillColorsArea(expectedColors, mDisplayWidth, {0, 0, mDisplayWidth, mDisplayHeight / 2}, GREEN);
-    fillColorsArea(expectedColors, mDisplayWidth,
-                   {0, mDisplayHeight / 2, mDisplayWidth, mDisplayHeight}, RED);
+    ReadbackHelper::fillColorsArea(expectedColors, mDisplayWidth,
+                                   {0, 0, mDisplayWidth, mDisplayHeight / 2}, GREEN);
+    ReadbackHelper::fillColorsArea(expectedColors, mDisplayWidth,
+                                   {0, mDisplayHeight / 2, mDisplayWidth, mDisplayHeight}, RED);
 
     ReadbackBuffer readbackBuffer(mPrimaryDisplay, mComposerClient, mGralloc, mDisplayWidth,
                                   mDisplayHeight, mPixelFormat, mDataspace);
@@ -706,14 +404,13 @@
     auto deviceLayer =
         std::make_shared<TestBufferLayer>(mComposerClient, mGralloc, mPrimaryDisplay, mDisplayWidth,
                                           mDisplayHeight / 2, PixelFormat::RGBA_8888);
-    std::vector<IComposerClient::Color> deviceColors(deviceLayer->mInfo.width *
-                                                     deviceLayer->mInfo.height);
-    fillColorsArea(deviceColors, deviceLayer->mInfo.width,
-                   {0, 0, static_cast<int32_t>(deviceLayer->mInfo.width),
-                    static_cast<int32_t>(deviceLayer->mInfo.height)},
-                   GREEN);
-    deviceLayer->setDisplayFrame({0, 0, static_cast<int32_t>(deviceLayer->mInfo.width),
-                                  static_cast<int32_t>(deviceLayer->mInfo.height)});
+    std::vector<IComposerClient::Color> deviceColors(deviceLayer->mWidth * deviceLayer->mHeight);
+    ReadbackHelper::fillColorsArea(deviceColors, deviceLayer->mWidth,
+                                   {0, 0, static_cast<int32_t>(deviceLayer->mWidth),
+                                    static_cast<int32_t>(deviceLayer->mHeight)},
+                                   GREEN);
+    deviceLayer->setDisplayFrame({0, 0, static_cast<int32_t>(deviceLayer->mWidth),
+                                  static_cast<int32_t>(deviceLayer->mHeight)});
     deviceLayer->setZOrder(10);
     ASSERT_NO_FATAL_FAILURE(deviceLayer->setBuffer(deviceColors));
     deviceLayer->write(mWriter);
@@ -728,30 +425,26 @@
     execute();
     ASSERT_EQ(0, mReader->mErrors.size());
 
-    IMapper::BufferDescriptorInfo clientInfo;
-    clientInfo.width = mDisplayWidth;
-    clientInfo.height = mDisplayHeight;
-    clientInfo.layerCount = 1;
-    clientInfo.format = PixelFormat::RGBA_8888;
-    clientInfo.usage =
-        static_cast<uint64_t>(BufferUsage::CPU_READ_OFTEN | BufferUsage::CPU_WRITE_OFTEN |
-                              BufferUsage::COMPOSER_CLIENT_TARGET);
-
+    uint64_t clientUsage =
+            static_cast<uint64_t>(BufferUsage::CPU_READ_OFTEN | BufferUsage::CPU_WRITE_OFTEN |
+                                  BufferUsage::COMPOSER_CLIENT_TARGET);
     uint32_t clientStride;
     const native_handle_t* clientBufferHandle =
-        mGralloc->allocate(clientInfo, /*import*/ true, &clientStride);
+            mGralloc->allocate(mDisplayWidth, mDisplayHeight, 1, PixelFormat::RGBA_8888,
+                               clientUsage, /*import*/ true, &clientStride);
     ASSERT_NE(nullptr, clientBufferHandle);
 
-    IMapper::Rect clientAccessRegion;
+    AccessRegion clientAccessRegion;
     clientAccessRegion.left = 0;
     clientAccessRegion.top = 0;
     clientAccessRegion.width = mDisplayWidth;
     clientAccessRegion.height = mDisplayHeight;
-    void* clientData = mGralloc->lock(clientBufferHandle, clientInfo.usage, clientAccessRegion, -1);
-    std::vector<IComposerClient::Color> clientColors(clientInfo.width * clientInfo.height);
-    fillColorsArea(clientColors, clientInfo.width, clientFrame, RED);
-    ASSERT_NO_FATAL_FAILURE(fillBuffer(clientInfo.width, clientInfo.height, clientStride,
-                                       clientData, clientInfo.format, clientColors));
+    void* clientData = mGralloc->lock(clientBufferHandle, clientUsage, clientAccessRegion, -1);
+    std::vector<IComposerClient::Color> clientColors(mDisplayWidth * mDisplayHeight);
+    ReadbackHelper::fillColorsArea(clientColors, mDisplayWidth, clientFrame, RED);
+    ASSERT_NO_FATAL_FAILURE(ReadbackHelper::fillBuffer(mDisplayWidth, mDisplayHeight, clientStride,
+                                                       clientData, PixelFormat::RGBA_8888,
+                                                       clientColors));
     int clientFence = mGralloc->unlock(clientBufferHandle);
     if (clientFence != -1) {
         sync_wait(clientFence, -1);
@@ -787,7 +480,7 @@
     IComposerClient::Rect redRect = {0, 0, mDisplayWidth / 4, mDisplayHeight / 4};
 
     std::vector<IComposerClient::Color> expectedColors(mDisplayWidth * mDisplayHeight);
-    fillColorsArea(expectedColors, mDisplayWidth, redRect, RED);
+    ReadbackHelper::fillColorsArea(expectedColors, mDisplayWidth, redRect, RED);
 
     auto layer =
         std::make_shared<TestBufferLayer>(mComposerClient, mGralloc, mPrimaryDisplay, mDisplayWidth,
@@ -820,8 +513,8 @@
 
     // update surface damage and recheck
     redRect = {mDisplayWidth / 4, mDisplayHeight / 4, mDisplayWidth / 2, mDisplayHeight / 2};
-    clearColors(expectedColors, mDisplayWidth, mDisplayHeight);
-    fillColorsArea(expectedColors, mDisplayWidth, redRect, RED);
+    ReadbackHelper::clearColors(expectedColors, mDisplayWidth, mDisplayHeight, mDisplayWidth);
+    ReadbackHelper::fillColorsArea(expectedColors, mDisplayWidth, redRect, RED);
 
     ASSERT_NO_FATAL_FAILURE(layer->fillBuffer(expectedColors));
     layer->setSurfaceDamage(
@@ -895,9 +588,10 @@
     mWriter->selectDisplay(mPrimaryDisplay);
 
     std::vector<IComposerClient::Color> expectedColors(mDisplayWidth * mDisplayHeight);
-    fillColorsArea(expectedColors, mDisplayWidth, {0, 0, mDisplayWidth, mDisplayHeight / 4}, RED);
-    fillColorsArea(expectedColors, mDisplayWidth,
-                   {0, mDisplayHeight / 2, mDisplayWidth, mDisplayHeight}, BLUE);
+    ReadbackHelper::fillColorsArea(expectedColors, mDisplayWidth,
+                                   {0, 0, mDisplayWidth, mDisplayHeight / 4}, RED);
+    ReadbackHelper::fillColorsArea(expectedColors, mDisplayWidth,
+                                   {0, mDisplayHeight / 2, mDisplayWidth, mDisplayHeight}, BLUE);
 
     auto layer =
         std::make_shared<TestBufferLayer>(mComposerClient, mGralloc, mPrimaryDisplay, mDisplayWidth,
@@ -911,7 +605,8 @@
     std::vector<std::shared_ptr<TestLayer>> layers = {layer};
 
     // update expected colors to match crop
-    fillColorsArea(expectedColors, mDisplayWidth, {0, 0, mDisplayWidth, mDisplayHeight}, BLUE);
+    ReadbackHelper::fillColorsArea(expectedColors, mDisplayWidth,
+                                   {0, 0, mDisplayWidth, mDisplayHeight}, BLUE);
     ReadbackBuffer readbackBuffer(mPrimaryDisplay, mComposerClient, mGralloc, mDisplayWidth,
                                   mDisplayHeight, mPixelFormat, mDataspace);
     ASSERT_NO_FATAL_FAILURE(readbackBuffer.setReadbackBuffer());
@@ -959,8 +654,8 @@
     redLayer->setZOrder(10);
 
     // fill blue first so that red will overwrite on overlap
-    fillColorsArea(expectedColors, mDisplayWidth, blueRect, BLUE);
-    fillColorsArea(expectedColors, mDisplayWidth, redRect, RED);
+    ReadbackHelper::fillColorsArea(expectedColors, mDisplayWidth, blueRect, BLUE);
+    ReadbackHelper::fillColorsArea(expectedColors, mDisplayWidth, redRect, RED);
 
     ReadbackBuffer readbackBuffer(mPrimaryDisplay, mComposerClient, mGralloc, mDisplayWidth,
                                   mDisplayHeight, mPixelFormat, mDataspace);
@@ -982,9 +677,9 @@
     ASSERT_NO_FATAL_FAILURE(readbackBuffer.checkReadbackBuffer(expectedColors));
 
     redLayer->setZOrder(1);
-    clearColors(expectedColors, mDisplayWidth, mDisplayHeight);
-    fillColorsArea(expectedColors, mDisplayWidth, redRect, RED);
-    fillColorsArea(expectedColors, mDisplayWidth, blueRect, BLUE);
+    ReadbackHelper::clearColors(expectedColors, mDisplayWidth, mDisplayHeight, mDisplayWidth);
+    ReadbackHelper::fillColorsArea(expectedColors, mDisplayWidth, redRect, RED);
+    ReadbackHelper::fillColorsArea(expectedColors, mDisplayWidth, blueRect, BLUE);
 
     ASSERT_NO_FATAL_FAILURE(readbackBuffer.setReadbackBuffer());
 
@@ -1019,8 +714,8 @@
     void setUpLayers(IComposerClient::BlendMode blendMode) {
         mLayers.clear();
         std::vector<IComposerClient::Color> topLayerPixelColors(mDisplayWidth * mDisplayHeight);
-        fillColorsArea(topLayerPixelColors, mDisplayWidth, {0, 0, mDisplayWidth, mDisplayHeight},
-                       mTopLayerColor);
+        ReadbackHelper::fillColorsArea(topLayerPixelColors, mDisplayWidth,
+                                       {0, 0, mDisplayWidth, mDisplayHeight}, mTopLayerColor);
 
         auto backgroundLayer = std::make_shared<TestColorLayer>(mComposerClient, mPrimaryDisplay);
         backgroundLayer->setDisplayFrame({0, 0, mDisplayWidth, mDisplayHeight});
@@ -1043,7 +738,7 @@
 
     void setExpectedColors(std::vector<IComposerClient::Color>& expectedColors) {
         ASSERT_EQ(2, mLayers.size());
-        clearColors(expectedColors, mDisplayWidth, mDisplayHeight);
+        ReadbackHelper::clearColors(expectedColors, mDisplayWidth, mDisplayHeight, mDisplayWidth);
 
         auto layer = mLayers[1];
         IComposerClient::BlendMode blendMode = layer->mBlendMode;
@@ -1216,8 +911,8 @@
         mLayer->setZOrder(10);
 
         std::vector<IComposerClient::Color> baseColors(mSideLength * mSideLength);
-        fillColorsArea(baseColors, mSideLength, redRect, RED);
-        fillColorsArea(baseColors, mSideLength, blueRect, BLUE);
+        ReadbackHelper::fillColorsArea(baseColors, mSideLength, redRect, RED);
+        ReadbackHelper::fillColorsArea(baseColors, mSideLength, blueRect, BLUE);
         ASSERT_NO_FATAL_FAILURE(mLayer->setBuffer(baseColors));
 
         mLayers = {backgroundLayer, mLayer};
@@ -1240,10 +935,10 @@
     ASSERT_NO_FATAL_FAILURE(readbackBuffer.setReadbackBuffer());
     mLayer->setTransform(Transform::FLIP_H);
     std::vector<IComposerClient::Color> expectedColors(mDisplayWidth * mDisplayHeight);
-    fillColorsArea(expectedColors, mDisplayWidth,
-                   {mSideLength / 2, 0, mSideLength, mSideLength / 2}, RED);
-    fillColorsArea(expectedColors, mDisplayWidth,
-                   {0, mSideLength / 2, mSideLength / 2, mSideLength}, BLUE);
+    ReadbackHelper::fillColorsArea(expectedColors, mDisplayWidth,
+                                   {mSideLength / 2, 0, mSideLength, mSideLength / 2}, RED);
+    ReadbackHelper::fillColorsArea(expectedColors, mDisplayWidth,
+                                   {0, mSideLength / 2, mSideLength / 2, mSideLength}, BLUE);
 
     writeLayers(mLayers);
     ASSERT_EQ(0, mReader->mErrors.size());
@@ -1274,10 +969,10 @@
     mLayer->setTransform(Transform::FLIP_V);
 
     std::vector<IComposerClient::Color> expectedColors(mDisplayWidth * mDisplayHeight);
-    fillColorsArea(expectedColors, mDisplayWidth,
-                   {0, mSideLength / 2, mSideLength / 2, mSideLength}, RED);
-    fillColorsArea(expectedColors, mDisplayWidth,
-                   {mSideLength / 2, 0, mSideLength, mSideLength / 2}, BLUE);
+    ReadbackHelper::fillColorsArea(expectedColors, mDisplayWidth,
+                                   {0, mSideLength / 2, mSideLength / 2, mSideLength}, RED);
+    ReadbackHelper::fillColorsArea(expectedColors, mDisplayWidth,
+                                   {mSideLength / 2, 0, mSideLength, mSideLength / 2}, BLUE);
 
     writeLayers(mLayers);
     ASSERT_EQ(0, mReader->mErrors.size());
@@ -1307,9 +1002,11 @@
     mLayer->setTransform(Transform::ROT_180);
 
     std::vector<IComposerClient::Color> expectedColors(mDisplayWidth * mDisplayHeight);
-    fillColorsArea(expectedColors, mDisplayWidth,
-                   {mSideLength / 2, mSideLength / 2, mSideLength, mSideLength}, RED);
-    fillColorsArea(expectedColors, mDisplayWidth, {0, 0, mSideLength / 2, mSideLength / 2}, BLUE);
+    ReadbackHelper::fillColorsArea(expectedColors, mDisplayWidth,
+                                   {mSideLength / 2, mSideLength / 2, mSideLength, mSideLength},
+                                   RED);
+    ReadbackHelper::fillColorsArea(expectedColors, mDisplayWidth,
+                                   {0, 0, mSideLength / 2, mSideLength / 2}, BLUE);
 
     writeLayers(mLayers);
     ASSERT_EQ(0, mReader->mErrors.size());
diff --git a/graphics/composer/2.2/vts/functional/VtsHalGraphicsComposerV2_2TargetTest.cpp b/graphics/composer/2.2/vts/functional/VtsHalGraphicsComposerV2_2TargetTest.cpp
index 9c80f4d..51832f9 100644
--- a/graphics/composer/2.2/vts/functional/VtsHalGraphicsComposerV2_2TargetTest.cpp
+++ b/graphics/composer/2.2/vts/functional/VtsHalGraphicsComposerV2_2TargetTest.cpp
@@ -40,7 +40,6 @@
 using common::V1_1::PixelFormat;
 using common::V1_1::RenderIntent;
 using mapper::V2_0::IMapper;
-using mapper::V2_0::vts::Gralloc;
 
 // Test environment for graphics.composer
 class GraphicsComposerHidlEnvironment : public ::testing::VtsHalHidlTargetTestEnvBase {
@@ -171,15 +170,10 @@
     }
 
     const native_handle_t* allocate() {
-        IMapper::BufferDescriptorInfo info{};
-        info.width = 64;
-        info.height = 64;
-        info.layerCount = 1;
-        info.format = static_cast<common::V1_0::PixelFormat>(PixelFormat::RGBA_8888);
-        info.usage =
-            static_cast<uint64_t>(BufferUsage::CPU_WRITE_OFTEN | BufferUsage::CPU_READ_OFTEN);
-
-        return mGralloc->allocate(info);
+        uint64_t usage =
+                static_cast<uint64_t>(BufferUsage::CPU_WRITE_OFTEN | BufferUsage::CPU_READ_OFTEN);
+        return mGralloc->allocate(/*width*/ 64, /*height*/ 64, /*layerCount*/ 1,
+                                  PixelFormat::RGBA_8888, usage);
     }
 
     void execute() { mComposerClient->execute(mReader.get(), mWriter.get()); }
@@ -456,18 +450,15 @@
         return;
     }
 
-    IMapper::BufferDescriptorInfo info{};
-    info.width = mDisplayWidth;
-    info.height = mDisplayHeight;
-    info.layerCount = 1;
-    info.format = static_cast<common::V1_0::PixelFormat>(mReadbackPixelFormat);
     // BufferUsage::COMPOSER_OUTPUT is missing
-    info.usage = static_cast<uint64_t>(BufferUsage::COMPOSER_OVERLAY | BufferUsage::CPU_READ_OFTEN);
+    uint64_t usage =
+            static_cast<uint64_t>(BufferUsage::COMPOSER_OVERLAY | BufferUsage::CPU_READ_OFTEN);
 
     std::unique_ptr<Gralloc> gralloc;
     const native_handle_t* buffer;
     ASSERT_NO_FATAL_FAILURE(gralloc = std::make_unique<Gralloc>());
-    ASSERT_NO_FATAL_FAILURE(buffer = gralloc->allocate(info));
+    ASSERT_NO_FATAL_FAILURE(buffer = gralloc->allocate(mDisplayWidth, mDisplayHeight, 1,
+                                                       mReadbackPixelFormat, usage));
 
     mComposerClient->setReadbackBuffer(mPrimaryDisplay, buffer, -1);
 }
@@ -483,17 +474,14 @@
         return;
     }
 
-    IMapper::BufferDescriptorInfo info{};
-    info.width = mDisplayWidth;
-    info.height = mDisplayHeight;
-    info.layerCount = 1;
-    info.format = static_cast<common::V1_0::PixelFormat>(mReadbackPixelFormat);
-    info.usage = static_cast<uint64_t>(BufferUsage::COMPOSER_OVERLAY | BufferUsage::CPU_READ_OFTEN);
+    uint64_t usage =
+            static_cast<uint64_t>(BufferUsage::COMPOSER_OVERLAY | BufferUsage::CPU_READ_OFTEN);
 
     std::unique_ptr<Gralloc> gralloc;
     const native_handle_t* buffer;
     ASSERT_NO_FATAL_FAILURE(gralloc = std::make_unique<Gralloc>());
-    ASSERT_NO_FATAL_FAILURE(buffer = gralloc->allocate(info));
+    ASSERT_NO_FATAL_FAILURE(buffer = gralloc->allocate(mDisplayWidth, mDisplayHeight, 1,
+                                                       mReadbackPixelFormat, usage));
 
     Error error = mComposerClient->getRaw()->setReadbackBuffer(mInvalidDisplayId, buffer, nullptr);
     ASSERT_EQ(Error::BAD_DISPLAY, error);
diff --git a/graphics/composer/2.3/utils/vts/Android.bp b/graphics/composer/2.3/utils/vts/Android.bp
index 19438cb..036ef69 100644
--- a/graphics/composer/2.3/utils/vts/Android.bp
+++ b/graphics/composer/2.3/utils/vts/Android.bp
@@ -27,6 +27,12 @@
         "android.hardware.graphics.composer@2.2",
         "android.hardware.graphics.composer@2.2-vts",
         "android.hardware.graphics.composer@2.3",
+        "android.hardware.graphics.mapper@2.0",
+        "android.hardware.graphics.mapper@2.0-vts",
+        "android.hardware.graphics.mapper@2.1",
+        "android.hardware.graphics.mapper@2.1-vts",
+        "android.hardware.graphics.mapper@3.0",
+        "android.hardware.graphics.mapper@3.0-vts",
     ],
     header_libs: [
         "android.hardware.graphics.composer@2.1-command-buffer",
diff --git a/graphics/composer/2.3/vts/functional/Android.bp b/graphics/composer/2.3/vts/functional/Android.bp
index ce675ee..2766638 100644
--- a/graphics/composer/2.3/vts/functional/Android.bp
+++ b/graphics/composer/2.3/vts/functional/Android.bp
@@ -28,6 +28,7 @@
     ],
     static_libs: [
         "android.hardware.graphics.allocator@2.0",
+        "android.hardware.graphics.allocator@3.0",
         "android.hardware.graphics.composer@2.1",
         "android.hardware.graphics.composer@2.1-vts",
         "android.hardware.graphics.composer@2.2",
@@ -37,6 +38,9 @@
         "android.hardware.graphics.mapper@2.0",
         "android.hardware.graphics.mapper@2.0-vts",
         "android.hardware.graphics.mapper@2.1",
+        "android.hardware.graphics.mapper@2.1-vts",
+        "android.hardware.graphics.mapper@3.0",
+        "android.hardware.graphics.mapper@3.0-vts",
     ],
     header_libs: [
         "android.hardware.graphics.composer@2.1-command-buffer",
diff --git a/graphics/composer/2.3/vts/functional/VtsHalGraphicsComposerV2_3TargetTest.cpp b/graphics/composer/2.3/vts/functional/VtsHalGraphicsComposerV2_3TargetTest.cpp
index d51c30a..dafe587 100644
--- a/graphics/composer/2.3/vts/functional/VtsHalGraphicsComposerV2_3TargetTest.cpp
+++ b/graphics/composer/2.3/vts/functional/VtsHalGraphicsComposerV2_3TargetTest.cpp
@@ -41,7 +41,7 @@
 using common::V1_2::Dataspace;
 using common::V1_2::PixelFormat;
 using mapper::V2_0::IMapper;
-using mapper::V2_0::vts::Gralloc;
+using V2_2::vts::Gralloc;
 
 // Test environment for graphics.composer
 class GraphicsComposerHidlEnvironment : public ::testing::VtsHalHidlTargetTestEnvBase {
@@ -156,15 +156,9 @@
     }
 
     const native_handle_t* allocate() {
-        IMapper::BufferDescriptorInfo info{};
-        info.width = 64;
-        info.height = 64;
-        info.layerCount = 1;
-        info.format = static_cast<common::V1_0::PixelFormat>(PixelFormat::RGBA_8888);
-        info.usage =
-            static_cast<uint64_t>(BufferUsage::CPU_WRITE_OFTEN | BufferUsage::CPU_READ_OFTEN);
-
-        return mGralloc->allocate(info);
+        return mGralloc->allocate(
+                64, 64, 1, static_cast<common::V1_1::PixelFormat>(PixelFormat::RGBA_8888),
+                static_cast<uint64_t>(BufferUsage::CPU_WRITE_OFTEN | BufferUsage::CPU_READ_OFTEN));
     }
 
     void execute() { mComposerClient->execute(mReader.get(), mWriter.get()); }
diff --git a/graphics/mapper/2.1/utils/vts/MapperVts.cpp b/graphics/mapper/2.1/utils/vts/MapperVts.cpp
index 078068e..36f9cbb 100644
--- a/graphics/mapper/2.1/utils/vts/MapperVts.cpp
+++ b/graphics/mapper/2.1/utils/vts/MapperVts.cpp
@@ -43,24 +43,25 @@
                       offsetof(IMapper::BufferDescriptorInfo, usage),
               "");
 
-Gralloc::Gralloc() : V2_0::vts::Gralloc() {
+Gralloc::Gralloc(bool errOnFailure) : V2_0::vts::Gralloc() {
     if (::testing::Test::HasFatalFailure()) {
         return;
     }
-    init();
+    init(errOnFailure);
 }
 
-Gralloc::Gralloc(const std::string& allocatorServiceName, const std::string& mapperServiceName)
+Gralloc::Gralloc(const std::string& allocatorServiceName, const std::string& mapperServiceName,
+                 bool errOnFailure)
     : V2_0::vts::Gralloc(allocatorServiceName, mapperServiceName) {
     if (::testing::Test::HasFatalFailure()) {
         return;
     }
-    init();
+    init(errOnFailure);
 }
 
-void Gralloc::init() {
+void Gralloc::init(bool errOnFailure) {
     mMapperV2_1 = IMapper::castFrom(V2_0::vts::Gralloc::getMapper());
-    ASSERT_NE(nullptr, mMapperV2_1.get()) << "failed to get mapper 2.1 service";
+    if (errOnFailure) ASSERT_NE(nullptr, mMapperV2_1.get()) << "failed to get mapper 2.1 service";
 }
 
 sp<IMapper> Gralloc::getMapper() const {
diff --git a/graphics/mapper/2.1/utils/vts/include/mapper-vts/2.1/MapperVts.h b/graphics/mapper/2.1/utils/vts/include/mapper-vts/2.1/MapperVts.h
index 423d4b3..bc683bb 100644
--- a/graphics/mapper/2.1/utils/vts/include/mapper-vts/2.1/MapperVts.h
+++ b/graphics/mapper/2.1/utils/vts/include/mapper-vts/2.1/MapperVts.h
@@ -32,25 +32,26 @@
 // A wrapper to IAllocator and IMapper.
 class Gralloc : public V2_0::vts::Gralloc {
    public:
-    Gralloc();
-    Gralloc(const std::string& allocatorServiceName, const std::string& mapperServiceName);
+     Gralloc(bool errOnFailure = true);
+     Gralloc(const std::string& allocatorServiceName, const std::string& mapperServiceName,
+             bool errOnFailure = true);
 
-    sp<IMapper> getMapper() const;
+     sp<IMapper> getMapper() const;
 
-    bool validateBufferSize(const native_handle_t* bufferHandle,
-                            const IMapper::BufferDescriptorInfo& descriptorInfo, uint32_t stride);
-    void getTransportSize(const native_handle_t* bufferHandle, uint32_t* outNumFds,
-                          uint32_t* outNumInts);
+     bool validateBufferSize(const native_handle_t* bufferHandle,
+                             const IMapper::BufferDescriptorInfo& descriptorInfo, uint32_t stride);
+     void getTransportSize(const native_handle_t* bufferHandle, uint32_t* outNumFds,
+                           uint32_t* outNumInts);
 
-    BufferDescriptor createDescriptor(const IMapper::BufferDescriptorInfo& descriptorInfo);
+     BufferDescriptor createDescriptor(const IMapper::BufferDescriptorInfo& descriptorInfo);
 
-    const native_handle_t* allocate(const IMapper::BufferDescriptorInfo& descriptorInfo,
-                                    bool import = true, uint32_t* outStride = nullptr);
+     const native_handle_t* allocate(const IMapper::BufferDescriptorInfo& descriptorInfo,
+                                     bool import = true, uint32_t* outStride = nullptr);
 
    protected:
-    void init();
+     void init(bool errOnFailure = true);
 
-    sp<IMapper> mMapperV2_1;
+     sp<IMapper> mMapperV2_1;
 };
 
 }  // namespace vts
diff --git a/graphics/mapper/3.0/utils/vts/MapperVts.cpp b/graphics/mapper/3.0/utils/vts/MapperVts.cpp
index f2b7594..c94e8db 100644
--- a/graphics/mapper/3.0/utils/vts/MapperVts.cpp
+++ b/graphics/mapper/3.0/utils/vts/MapperVts.cpp
@@ -25,8 +25,13 @@
 namespace V3_0 {
 namespace vts {
 
-Gralloc::Gralloc(const std::string& allocatorServiceName, const std::string& mapperServiceName) {
-    init(allocatorServiceName, mapperServiceName);
+Gralloc::Gralloc(const std::string& allocatorServiceName, const std::string& mapperServiceName,
+                 bool errOnFailure) {
+    if (errOnFailure) {
+        init(allocatorServiceName, mapperServiceName);
+    } else {
+        initNoErr(allocatorServiceName, mapperServiceName);
+    }
 }
 
 void Gralloc::init(const std::string& allocatorServiceName, const std::string& mapperServiceName) {
@@ -38,6 +43,16 @@
     ASSERT_FALSE(mMapper->isRemote()) << "mapper is not in passthrough mode";
 }
 
+void Gralloc::initNoErr(const std::string& allocatorServiceName,
+                        const std::string& mapperServiceName) {
+    mAllocator = ::testing::VtsHalHidlTargetTestBase::getService<IAllocator>(allocatorServiceName);
+
+    mMapper = ::testing::VtsHalHidlTargetTestBase::getService<IMapper>(mapperServiceName);
+    if (mMapper.get()) {
+        ASSERT_FALSE(mMapper->isRemote()) << "mapper is not in passthrough mode";
+    }
+}
+
 Gralloc::~Gralloc() {
     for (auto bufferHandle : mClonedBuffers) {
         auto buffer = const_cast<native_handle_t*>(bufferHandle);
diff --git a/graphics/mapper/3.0/utils/vts/include/mapper-vts/3.0/MapperVts.h b/graphics/mapper/3.0/utils/vts/include/mapper-vts/3.0/MapperVts.h
index ba79ca4..1141a88 100644
--- a/graphics/mapper/3.0/utils/vts/include/mapper-vts/3.0/MapperVts.h
+++ b/graphics/mapper/3.0/utils/vts/include/mapper-vts/3.0/MapperVts.h
@@ -36,62 +36,66 @@
 // A wrapper to IAllocator and IMapper.
 class Gralloc {
    public:
-    Gralloc(const std::string& allocatorServiceName = "default",
-            const std::string& mapperServiceName = "default");
-    ~Gralloc();
+     Gralloc(const std::string& allocatorServiceName = "default",
+             const std::string& mapperServiceName = "default", bool errOnFailure = true);
+     ~Gralloc();
 
-    // IAllocator methods
+     // IAllocator methods
 
-    sp<IAllocator> getAllocator() const;
+     sp<IAllocator> getAllocator() const;
 
-    std::string dumpDebugInfo();
+     std::string dumpDebugInfo();
 
-    // When import is false, this simply calls IAllocator::allocate. When import
-    // is true, the returned buffers are also imported into the mapper.
-    //
-    // Either case, the returned buffers must be freed with freeBuffer.
-    std::vector<const native_handle_t*> allocate(const BufferDescriptor& descriptor, uint32_t count,
-                                                 bool import = true, uint32_t* outStride = nullptr);
-    const native_handle_t* allocate(const IMapper::BufferDescriptorInfo& descriptorInfo,
-                                    bool import = true, uint32_t* outStride = nullptr);
+     // When import is false, this simply calls IAllocator::allocate. When import
+     // is true, the returned buffers are also imported into the mapper.
+     //
+     // Either case, the returned buffers must be freed with freeBuffer.
+     std::vector<const native_handle_t*> allocate(const BufferDescriptor& descriptor,
+                                                  uint32_t count, bool import = true,
+                                                  uint32_t* outStride = nullptr);
+     const native_handle_t* allocate(const IMapper::BufferDescriptorInfo& descriptorInfo,
+                                     bool import = true, uint32_t* outStride = nullptr);
 
-    // IMapper methods
+     // IMapper methods
 
-    sp<IMapper> getMapper() const;
+     sp<IMapper> getMapper() const;
 
-    BufferDescriptor createDescriptor(const IMapper::BufferDescriptorInfo& descriptorInfo);
+     BufferDescriptor createDescriptor(const IMapper::BufferDescriptorInfo& descriptorInfo);
 
-    const native_handle_t* importBuffer(const hidl_handle& rawHandle);
-    void freeBuffer(const native_handle_t* bufferHandle);
+     const native_handle_t* importBuffer(const hidl_handle& rawHandle);
+     void freeBuffer(const native_handle_t* bufferHandle);
 
-    // We use fd instead of hidl_handle in these functions to pass fences
-    // in and out of the mapper.  The ownership of the fd is always transferred
-    // with each of these functions.
-    void* lock(const native_handle_t* bufferHandle, uint64_t cpuUsage,
-               const IMapper::Rect& accessRegion, int acquireFence, int32_t* outBytesPerPixel,
-               int32_t* outBytesPerStride);
-    YCbCrLayout lockYCbCr(const native_handle_t* bufferHandle, uint64_t cpuUsage,
-                          const IMapper::Rect& accessRegion, int acquireFence);
-    int unlock(const native_handle_t* bufferHandle);
+     // We use fd instead of hidl_handle in these functions to pass fences
+     // in and out of the mapper.  The ownership of the fd is always transferred
+     // with each of these functions.
+     void* lock(const native_handle_t* bufferHandle, uint64_t cpuUsage,
+                const IMapper::Rect& accessRegion, int acquireFence, int32_t* outBytesPerPixel,
+                int32_t* outBytesPerStride);
+     YCbCrLayout lockYCbCr(const native_handle_t* bufferHandle, uint64_t cpuUsage,
+                           const IMapper::Rect& accessRegion, int acquireFence);
+     int unlock(const native_handle_t* bufferHandle);
 
-    bool validateBufferSize(const native_handle_t* bufferHandle,
-                            const IMapper::BufferDescriptorInfo& descriptorInfo, uint32_t stride);
-    void getTransportSize(const native_handle_t* bufferHandle, uint32_t* outNumFds,
-                          uint32_t* outNumInts);
+     bool validateBufferSize(const native_handle_t* bufferHandle,
+                             const IMapper::BufferDescriptorInfo& descriptorInfo, uint32_t stride);
+     void getTransportSize(const native_handle_t* bufferHandle, uint32_t* outNumFds,
+                           uint32_t* outNumInts);
 
-    bool isSupported(const IMapper::BufferDescriptorInfo& descriptorInfo);
+     bool isSupported(const IMapper::BufferDescriptorInfo& descriptorInfo);
 
    private:
-    void init(const std::string& allocatorServiceName, const std::string& mapperServiceName);
-    const native_handle_t* cloneBuffer(const hidl_handle& rawHandle);
+     void init(const std::string& allocatorServiceName, const std::string& mapperServiceName);
 
-    sp<IAllocator> mAllocator;
-    sp<IMapper> mMapper;
+     // initialize without checking for failure to get service
+     void initNoErr(const std::string& allocatorServiceName, const std::string& mapperServiceName);
+     const native_handle_t* cloneBuffer(const hidl_handle& rawHandle);
 
-    // Keep track of all cloned and imported handles.  When a test fails with
-    // ASSERT_*, the destructor will free the handles for the test.
-    std::unordered_set<const native_handle_t*> mClonedBuffers;
-    std::unordered_set<const native_handle_t*> mImportedBuffers;
+     sp<IAllocator> mAllocator;
+     sp<IMapper> mMapper;
+
+     // Keep track of all cloned and imported handles.  When a test fails with
+     // ASSERT_*, the destructor will free the handles for the test.
+     std::unordered_set<const native_handle_t*> mClonedBuffers;
+     std::unordered_set<const native_handle_t*> mImportedBuffers;
 };
 
 }  // namespace vts
diff --git a/keymaster/4.0/support/Keymaster.cpp b/keymaster/4.0/support/Keymaster.cpp
index 1eb9a68..f20f951 100644
--- a/keymaster/4.0/support/Keymaster.cpp
+++ b/keymaster/4.0/support/Keymaster.cpp
@@ -80,8 +80,7 @@
 }
 
 template <typename Wrapper>
-std::vector<std::unique_ptr<Keymaster>> enumerateDevices(
-    const sp<IServiceManager>& serviceManager) {
+Keymaster::KeymasterSet enumerateDevices(const sp<IServiceManager>& serviceManager) {
     Keymaster::KeymasterSet result;
 
     bool foundDefault = false;
@@ -92,7 +91,7 @@
             auto device = Wrapper::WrappedIKeymasterDevice::getService(name);
             CHECK(device) << "Failed to get service for " << descriptor << " with interface name "
                           << name;
-            result.push_back(std::unique_ptr<Keymaster>(new Wrapper(device, name)));
+            result.push_back(new Wrapper(device, name));
         }
     });
 
@@ -100,7 +99,7 @@
         // "default" wasn't provided by listManifestByInterface.  Maybe there's a passthrough
         // implementation.
         auto device = Wrapper::WrappedIKeymasterDevice::getService("default");
-        if (device) result.push_back(std::unique_ptr<Keymaster>(new Wrapper(device, "default")));
+        if (device) result.push_back(new Wrapper(device, "default"));
     }
 
     return result;
diff --git a/keymaster/4.0/support/include/keymasterV4_0/Keymaster.h b/keymaster/4.0/support/include/keymasterV4_0/Keymaster.h
index 43a34b0..ad83f17 100644
--- a/keymaster/4.0/support/include/keymasterV4_0/Keymaster.h
+++ b/keymaster/4.0/support/include/keymasterV4_0/Keymaster.h
@@ -39,8 +39,8 @@
  * while still having to use only the latest interface.
  */
 class Keymaster : public IKeymasterDevice {
-   public:
-    using KeymasterSet = std::vector<std::unique_ptr<Keymaster>>;
+  public:
+    using KeymasterSet = std::vector<android::sp<Keymaster>>;
 
     Keymaster(const hidl_string& descriptor, const hidl_string& instanceName)
         : descriptor_(descriptor), instanceName_(instanceName) {}
@@ -86,7 +86,7 @@
      */
     static void performHmacKeyAgreement(const KeymasterSet& keymasters);
 
-   private:
+  private:
     hidl_string descriptor_;
     hidl_string instanceName_;
 };
diff --git a/neuralnetworks/1.2/vts/functional/GeneratedTests.cpp b/neuralnetworks/1.2/vts/functional/GeneratedTests.cpp
index 2c3287a..5af3255 100644
--- a/neuralnetworks/1.2/vts/functional/GeneratedTests.cpp
+++ b/neuralnetworks/1.2/vts/functional/GeneratedTests.cpp
@@ -44,6 +44,21 @@
 // in frameworks/ml/nn/runtime/tests/generated/
 #include "all_generated_V1_2_vts_tests.cpp"
 
+// Generated from spec/strided_slice_invalid_output_dims.mod.py.
+// TODO(b/132155416): Make this part of all_generated_V1_2_vts_tests.cpp.
+namespace strided_slice_invalid_output_dims {
+#include "generated/strided_slice_invalid_output_dims.example.cpp"
+#include "generated/strided_slice_invalid_output_dims.model.cpp"
+}  // namespace strided_slice_invalid_output_dims
+
+// TODO(b/132155416): Make this part of all_generated_V1_2_vts_tests.cpp.
+TEST_F(ValidationTest, strided_slice_invalid_output_dims) {
+    const Model model = strided_slice_invalid_output_dims::createTestModel();
+    const std::vector<Request> requests =
+            createRequests(strided_slice_invalid_output_dims::get_examples());
+    validateFailure(model, requests);
+}
+
 }  // namespace functional
 }  // namespace vts
 }  // namespace V1_2
diff --git a/neuralnetworks/1.2/vts/functional/ValidateModel.cpp b/neuralnetworks/1.2/vts/functional/ValidateModel.cpp
index f87ca89..a0b6d9a 100644
--- a/neuralnetworks/1.2/vts/functional/ValidateModel.cpp
+++ b/neuralnetworks/1.2/vts/functional/ValidateModel.cpp
@@ -508,15 +508,15 @@
                 }
             }
         }
-        // BIDIRECTIONAL_SEQUENCE_LSTM and BIDIRECTIONAL_SEQUENCE_RNN can have
-        // either one or two outputs depending on their mergeOutputs parameter.
+        // BIDIRECTIONAL_SEQUENCE_LSTM and BIDIRECTIONAL_SEQUENCE_RNN can have either one or two
+        // outputs depending on their mergeOutputs parameter.
         if (operation.type == OperationType::BIDIRECTIONAL_SEQUENCE_LSTM ||
             operation.type == OperationType::BIDIRECTIONAL_SEQUENCE_RNN) {
-          for (const size_t outOprand : operation.outputs) {
-            if (operand == outOprand) {
-              return true;
+            for (const size_t outOprand : operation.outputs) {
+                if (operand == outOprand) {
+                    return true;
+                }
             }
-          }
         }
     }
     return false;
diff --git a/neuralnetworks/1.2/vts/functional/ValidateRequest.cpp b/neuralnetworks/1.2/vts/functional/ValidateRequest.cpp
index 9703c2d..e935aaa 100644
--- a/neuralnetworks/1.2/vts/functional/ValidateRequest.cpp
+++ b/neuralnetworks/1.2/vts/functional/ValidateRequest.cpp
@@ -274,6 +274,22 @@
     }
 }
 
+void ValidationTest::validateRequestFailure(const sp<IPreparedModel>& preparedModel,
+                                            const std::vector<Request>& requests) {
+    for (const Request& request : requests) {
+        SCOPED_TRACE("Expecting request to fail [executeSynchronously]");
+        Return<void> executeStatus = preparedModel->executeSynchronously(
+                request, MeasureTiming::NO,
+                [](ErrorStatus error, const hidl_vec<OutputShape>& outputShapes,
+                   const Timing& timing) {
+                    ASSERT_NE(ErrorStatus::NONE, error);
+                    EXPECT_EQ(outputShapes.size(), 0);
+                    EXPECT_TRUE(badTiming(timing));
+                });
+        ASSERT_TRUE(executeStatus.isOk());
+    }
+}
+
 }  // namespace functional
 }  // namespace vts
 }  // namespace V1_2
diff --git a/neuralnetworks/1.2/vts/functional/VtsHalNeuralnetworks.cpp b/neuralnetworks/1.2/vts/functional/VtsHalNeuralnetworks.cpp
index 4ddefe8..666f9b5 100644
--- a/neuralnetworks/1.2/vts/functional/VtsHalNeuralnetworks.cpp
+++ b/neuralnetworks/1.2/vts/functional/VtsHalNeuralnetworks.cpp
@@ -140,6 +140,20 @@
     validateBurst(preparedModel, requests);
 }
 
+void ValidationTest::validateFailure(const Model& model, const std::vector<Request>& requests) {
+    // TODO: Should this always succeed?
+    //       What if the invalid input is part of the model (i.e., a parameter).
+    validateModel(model);
+
+    sp<IPreparedModel> preparedModel;
+    ASSERT_NO_FATAL_FAILURE(createPreparedModel(device, model, &preparedModel));
+    if (preparedModel == nullptr) {
+        return;
+    }
+
+    validateRequestFailure(preparedModel, requests);
+}
+
 sp<IPreparedModel> getPreparedModel_1_2(
     const sp<V1_2::implementation::PreparedModelCallback>& callback) {
     sp<V1_0::IPreparedModel> preparedModelV1_0 = callback->getPreparedModel();
diff --git a/neuralnetworks/1.2/vts/functional/VtsHalNeuralnetworks.h b/neuralnetworks/1.2/vts/functional/VtsHalNeuralnetworks.h
index 8d1acbe..80e810a 100644
--- a/neuralnetworks/1.2/vts/functional/VtsHalNeuralnetworks.h
+++ b/neuralnetworks/1.2/vts/functional/VtsHalNeuralnetworks.h
@@ -73,11 +73,14 @@
 class ValidationTest : public NeuralnetworksHidlTest {
    protected:
      void validateEverything(const Model& model, const std::vector<Request>& requests);
+     void validateFailure(const Model& model, const std::vector<Request>& requests);
 
    private:
      void validateModel(const Model& model);
      void validateRequests(const sp<IPreparedModel>& preparedModel,
                            const std::vector<Request>& requests);
+     void validateRequestFailure(const sp<IPreparedModel>& preparedModel,
+                                 const std::vector<Request>& requests);
      void validateBurst(const sp<IPreparedModel>& preparedModel,
                         const std::vector<Request>& requests);
 };
diff --git a/neuralnetworks/1.2/vts/functional/generated/strided_slice_invalid_output_dims.example.cpp b/neuralnetworks/1.2/vts/functional/generated/strided_slice_invalid_output_dims.example.cpp
new file mode 100644
index 0000000..0640833
--- /dev/null
+++ b/neuralnetworks/1.2/vts/functional/generated/strided_slice_invalid_output_dims.example.cpp
@@ -0,0 +1,116 @@
+// clang-format off
+// Generated file (from: strided_slice_invalid_output_dims.mod.py). Do not edit
+std::vector<MixedTypedExample>& get_examples() {
+static std::vector<MixedTypedExample> examples = {
+// Begin of an example
+{
+.operands = {
+//Input(s)
+{ // See tools/test_generator/include/TestHarness.h:MixedTyped
+  // int -> Dimensions map
+  .operandDimensions = {{0, {2, 3}}},
+  // int -> FLOAT32 map
+  .float32Operands = {{0, {1.0f, 2.0f, 3.0f, 4.0f, 5.0f, 6.0f}}},
+  // int -> INT32 map
+  .int32Operands = {},
+  // int -> QUANT8_ASYMM map
+  .quant8AsymmOperands = {},
+  // int -> QUANT16_SYMM map
+  .quant16SymmOperands = {},
+  // int -> FLOAT16 map
+  .float16Operands = {},
+  // int -> BOOL8 map
+  .bool8Operands = {},
+  // int -> QUANT8_SYMM_PER_CHANNEL map
+  .quant8ChannelOperands = {},
+  // int -> QUANT16_ASYMM map
+  .quant16AsymmOperands = {},
+  // int -> QUANT8_SYMM map
+  .quant8SymmOperands = {},
+},
+//Output(s)
+{ // See tools/test_generator/include/TestHarness.h:MixedTyped
+  // int -> Dimensions map
+  .operandDimensions = {{0, {3}}},
+  // int -> FLOAT32 map
+  .float32Operands = {{0, {1.0f, 2.0f, 3.0f}}},
+  // int -> INT32 map
+  .int32Operands = {},
+  // int -> QUANT8_ASYMM map
+  .quant8AsymmOperands = {},
+  // int -> QUANT16_SYMM map
+  .quant16SymmOperands = {},
+  // int -> FLOAT16 map
+  .float16Operands = {},
+  // int -> BOOL8 map
+  .bool8Operands = {},
+  // int -> QUANT8_SYMM_PER_CHANNEL map
+  .quant8ChannelOperands = {},
+  // int -> QUANT16_ASYMM map
+  .quant16AsymmOperands = {},
+  // int -> QUANT8_SYMM map
+  .quant8SymmOperands = {},
+}
+},
+}, // End of an example
+};
+return examples;
+};
+
+std::vector<MixedTypedExample>& get_examples_dynamic_output_shape() {
+static std::vector<MixedTypedExample> examples_dynamic_output_shape = {
+// Begin of an example
+{
+.operands = {
+//Input(s)
+{ // See tools/test_generator/include/TestHarness.h:MixedTyped
+  // int -> Dimensions map
+  .operandDimensions = {{0, {2, 3}}},
+  // int -> FLOAT32 map
+  .float32Operands = {{0, {1.0f, 2.0f, 3.0f, 4.0f, 5.0f, 6.0f}}},
+  // int -> INT32 map
+  .int32Operands = {},
+  // int -> QUANT8_ASYMM map
+  .quant8AsymmOperands = {},
+  // int -> QUANT16_SYMM map
+  .quant16SymmOperands = {},
+  // int -> FLOAT16 map
+  .float16Operands = {},
+  // int -> BOOL8 map
+  .bool8Operands = {},
+  // int -> QUANT8_SYMM_PER_CHANNEL map
+  .quant8ChannelOperands = {},
+  // int -> QUANT16_ASYMM map
+  .quant16AsymmOperands = {},
+  // int -> QUANT8_SYMM map
+  .quant8SymmOperands = {},
+},
+//Output(s)
+{ // See tools/test_generator/include/TestHarness.h:MixedTyped
+  // int -> Dimensions map
+  .operandDimensions = {{0, {3}}},
+  // int -> FLOAT32 map
+  .float32Operands = {{0, {1.0f, 2.0f, 3.0f}}},
+  // int -> INT32 map
+  .int32Operands = {},
+  // int -> QUANT8_ASYMM map
+  .quant8AsymmOperands = {},
+  // int -> QUANT16_SYMM map
+  .quant16SymmOperands = {},
+  // int -> FLOAT16 map
+  .float16Operands = {},
+  // int -> BOOL8 map
+  .bool8Operands = {},
+  // int -> QUANT8_SYMM_PER_CHANNEL map
+  .quant8ChannelOperands = {},
+  // int -> QUANT16_ASYMM map
+  .quant16AsymmOperands = {},
+  // int -> QUANT8_SYMM map
+  .quant8SymmOperands = {},
+}
+},
+}, // End of an example
+};
+return examples_dynamic_output_shape;
+};
+
diff --git a/neuralnetworks/1.2/vts/functional/generated/strided_slice_invalid_output_dims.model.cpp b/neuralnetworks/1.2/vts/functional/generated/strided_slice_invalid_output_dims.model.cpp
new file mode 100644
index 0000000..106655a
--- /dev/null
+++ b/neuralnetworks/1.2/vts/functional/generated/strided_slice_invalid_output_dims.model.cpp
@@ -0,0 +1,216 @@
+// clang-format off
+// Generated file (from: strided_slice_invalid_output_dims.mod.py). Do not edit
+// Create the model
+Model createTestModel() {
+    const std::vector<Operand> operands = {
+        {
+            .type = OperandType::TENSOR_FLOAT32,
+            .dimensions = {2, 3},
+            .numberOfConsumers = 1,
+            .scale = 0.0f,
+            .zeroPoint = 0,
+            .lifetime = OperandLifeTime::MODEL_INPUT,
+            .location = {.poolIndex = 0, .offset = 0, .length = 0},
+        },
+        {
+            .type = OperandType::TENSOR_INT32,
+            .dimensions = {2},
+            .numberOfConsumers = 1,
+            .scale = 0.0f,
+            .zeroPoint = 0,
+            .lifetime = OperandLifeTime::CONSTANT_COPY,
+            .location = {.poolIndex = 0, .offset = 0, .length = 8},
+        },
+        {
+            .type = OperandType::TENSOR_INT32,
+            .dimensions = {2},
+            .numberOfConsumers = 1,
+            .scale = 0.0f,
+            .zeroPoint = 0,
+            .lifetime = OperandLifeTime::CONSTANT_COPY,
+            .location = {.poolIndex = 0, .offset = 8, .length = 8},
+        },
+        {
+            .type = OperandType::TENSOR_INT32,
+            .dimensions = {2},
+            .numberOfConsumers = 1,
+            .scale = 0.0f,
+            .zeroPoint = 0,
+            .lifetime = OperandLifeTime::CONSTANT_COPY,
+            .location = {.poolIndex = 0, .offset = 16, .length = 8},
+        },
+        {
+            .type = OperandType::INT32,
+            .dimensions = {},
+            .numberOfConsumers = 1,
+            .scale = 0.0f,
+            .zeroPoint = 0,
+            .lifetime = OperandLifeTime::CONSTANT_COPY,
+            .location = {.poolIndex = 0, .offset = 24, .length = 4},
+        },
+        {
+            .type = OperandType::INT32,
+            .dimensions = {},
+            .numberOfConsumers = 1,
+            .scale = 0.0f,
+            .zeroPoint = 0,
+            .lifetime = OperandLifeTime::CONSTANT_COPY,
+            .location = {.poolIndex = 0, .offset = 28, .length = 4},
+        },
+        {
+            .type = OperandType::INT32,
+            .dimensions = {},
+            .numberOfConsumers = 1,
+            .scale = 0.0f,
+            .zeroPoint = 0,
+            .lifetime = OperandLifeTime::CONSTANT_COPY,
+            .location = {.poolIndex = 0, .offset = 32, .length = 4},
+        },
+        {
+            .type = OperandType::TENSOR_FLOAT32,
+            .dimensions = {3},
+            .numberOfConsumers = 0,
+            .scale = 0.0f,
+            .zeroPoint = 0,
+            .lifetime = OperandLifeTime::MODEL_OUTPUT,
+            .location = {.poolIndex = 0, .offset = 0, .length = 0},
+        }
+    };
+
+    const std::vector<Operation> operations = {
+        {
+            .type = OperationType::STRIDED_SLICE,
+            .inputs = {0, 1, 2, 3, 4, 5, 6},
+            .outputs = {7},
+        }
+    };
+
+    const std::vector<uint32_t> inputIndexes = {0};
+    const std::vector<uint32_t> outputIndexes = {7};
+    std::vector<uint8_t> operandValues = {
+      0, 0, 0, 0, 0, 0, 0, 0, 2, 0, 0, 0, 3, 0, 0, 0, 1, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0
+    };
+    const std::vector<hidl_memory> pools = {};
+
+    return {
+        .operands = operands,
+        .operations = operations,
+        .inputIndexes = inputIndexes,
+        .outputIndexes = outputIndexes,
+        .operandValues = operandValues,
+        .pools = pools,
+    };
+}
+
+inline bool is_ignored(int i) {
+  static std::set<int> ignore = {};
+  return ignore.find(i) != ignore.end();
+}
+
+// Create the model
+Model createTestModel_dynamic_output_shape() {
+    const std::vector<Operand> operands = {
+        {
+            .type = OperandType::TENSOR_FLOAT32,
+            .dimensions = {2, 3},
+            .numberOfConsumers = 1,
+            .scale = 0.0f,
+            .zeroPoint = 0,
+            .lifetime = OperandLifeTime::MODEL_INPUT,
+            .location = {.poolIndex = 0, .offset = 0, .length = 0},
+        },
+        {
+            .type = OperandType::TENSOR_INT32,
+            .dimensions = {2},
+            .numberOfConsumers = 1,
+            .scale = 0.0f,
+            .zeroPoint = 0,
+            .lifetime = OperandLifeTime::CONSTANT_COPY,
+            .location = {.poolIndex = 0, .offset = 0, .length = 8},
+        },
+        {
+            .type = OperandType::TENSOR_INT32,
+            .dimensions = {2},
+            .numberOfConsumers = 1,
+            .scale = 0.0f,
+            .zeroPoint = 0,
+            .lifetime = OperandLifeTime::CONSTANT_COPY,
+            .location = {.poolIndex = 0, .offset = 8, .length = 8},
+        },
+        {
+            .type = OperandType::TENSOR_INT32,
+            .dimensions = {2},
+            .numberOfConsumers = 1,
+            .scale = 0.0f,
+            .zeroPoint = 0,
+            .lifetime = OperandLifeTime::CONSTANT_COPY,
+            .location = {.poolIndex = 0, .offset = 16, .length = 8},
+        },
+        {
+            .type = OperandType::INT32,
+            .dimensions = {},
+            .numberOfConsumers = 1,
+            .scale = 0.0f,
+            .zeroPoint = 0,
+            .lifetime = OperandLifeTime::CONSTANT_COPY,
+            .location = {.poolIndex = 0, .offset = 24, .length = 4},
+        },
+        {
+            .type = OperandType::INT32,
+            .dimensions = {},
+            .numberOfConsumers = 1,
+            .scale = 0.0f,
+            .zeroPoint = 0,
+            .lifetime = OperandLifeTime::CONSTANT_COPY,
+            .location = {.poolIndex = 0, .offset = 28, .length = 4},
+        },
+        {
+            .type = OperandType::INT32,
+            .dimensions = {},
+            .numberOfConsumers = 1,
+            .scale = 0.0f,
+            .zeroPoint = 0,
+            .lifetime = OperandLifeTime::CONSTANT_COPY,
+            .location = {.poolIndex = 0, .offset = 32, .length = 4},
+        },
+        {
+            .type = OperandType::TENSOR_FLOAT32,
+            .dimensions = {0},
+            .numberOfConsumers = 0,
+            .scale = 0.0f,
+            .zeroPoint = 0,
+            .lifetime = OperandLifeTime::MODEL_OUTPUT,
+            .location = {.poolIndex = 0, .offset = 0, .length = 0},
+        }
+    };
+
+    const std::vector<Operation> operations = {
+        {
+            .type = OperationType::STRIDED_SLICE,
+            .inputs = {0, 1, 2, 3, 4, 5, 6},
+            .outputs = {7},
+        }
+    };
+
+    const std::vector<uint32_t> inputIndexes = {0};
+    const std::vector<uint32_t> outputIndexes = {7};
+    std::vector<uint8_t> operandValues = {
+      0, 0, 0, 0, 0, 0, 0, 0, 2, 0, 0, 0, 3, 0, 0, 0, 1, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0
+    };
+    const std::vector<hidl_memory> pools = {};
+
+    return {
+        .operands = operands,
+        .operations = operations,
+        .inputIndexes = inputIndexes,
+        .outputIndexes = outputIndexes,
+        .operandValues = operandValues,
+        .pools = pools,
+    };
+}
+
+inline bool is_ignored_dynamic_output_shape(int i) {
+  static std::set<int> ignore = {};
+  return ignore.find(i) != ignore.end();
+}
+
diff --git a/neuralnetworks/1.2/vts/functional/spec/strided_slice_invalid_output_dims.mod.py b/neuralnetworks/1.2/vts/functional/spec/strided_slice_invalid_output_dims.mod.py
new file mode 100644
index 0000000..e8d30f3
--- /dev/null
+++ b/neuralnetworks/1.2/vts/functional/spec/strided_slice_invalid_output_dims.mod.py
@@ -0,0 +1,43 @@
+#
+# Copyright (C) 2019 The Android Open Source Project
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#      http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+#
+
+# This test makes sure that executing STRIDED_SLICE results in a failure when
+# the output dimensions do not match shrinkAxisMask.
+#
+# The test generator does not support generating tests resulting in execution
+# failure, so the gTest part of this test has been written by hand.
+# TODO(b/132155416): Move this under frameworks/ml/nn/runtime/test/specs/V1_2.
+#
+# Based on strided_slice_float_11.mod.py.
+
+model = Model()
+i1 = Input("input", "TENSOR_FLOAT32", "{2, 3}")
+begins = Parameter("begins", "TENSOR_INT32", "{2}", [0, 0])
+# The value "2" below makes the test invalid. See http://b/79856511#comment2.
+ends = Parameter("ends", "TENSOR_INT32", "{2}", [2, 3])
+strides = Parameter("strides", "TENSOR_INT32", "{2}", [1, 1])
+beginMask = Int32Scalar("beginMask", 0)
+endMask = Int32Scalar("endMask", 0)
+shrinkAxisMask = Int32Scalar("shrinkAxisMask", 1)
+
+output = Output("output", "TENSOR_FLOAT32", "{3}")
+
+model = model.Operation("STRIDED_SLICE", i1, begins, ends, strides, beginMask, endMask, shrinkAxisMask).To(output)
+
+Example({
+    i1: [1, 2, 3, 4, 5, 6],
+    output: [1, 2, 3],
+})