Merge tag 'v6.10-p1' of git://git.kernel.org/pub/scm/linux/kernel/git/herbert/crypto-2.6

Pull crypto updates from Herbert Xu:
 "API:
   - Remove crypto stats interface

  Algorithms:
   - Add faster AES-XTS on modern x86_64 CPUs
   - Forbid curves with order less than 224 bits in ecc (FIPS 186-5)
   - Add ECDSA NIST P521

  Drivers:
   - Expose otp zone in atmel
   - Add dh fallback for primes > 4K in qat
   - Add interface for live migration in qat
   - Use dma for aes requests in starfive
   - Add full DMA support for stm32mpx in stm32
   - Add Tegra Security Engine driver

  Others:
   - Introduce scope-based x509_certificate allocation"

* tag 'v6.10-p1' of git://git.kernel.org/pub/scm/linux/kernel/git/herbert/crypto-2.6: (123 commits)
  crypto: atmel-sha204a - provide the otp content
  crypto: atmel-sha204a - add reading from otp zone
  crypto: atmel-i2c - rename read function
  crypto: atmel-i2c - add missing arg description
  crypto: iaa - Use kmemdup() instead of kzalloc() and memcpy()
  crypto: sahara - use 'time_left' variable with wait_for_completion_timeout()
  crypto: api - use 'time_left' variable with wait_for_completion_killable_timeout()
  crypto: caam - i.MX8ULP donot have CAAM page0 access
  crypto: caam - init-clk based on caam-page0-access
  crypto: starfive - Use fallback for unaligned dma access
  crypto: starfive - Do not free stack buffer
  crypto: starfive - Skip unneeded fallback allocation
  crypto: starfive - Skip dma setup for zeroed message
  crypto: hisilicon/sec2 - fix for register offset
  crypto: hisilicon/debugfs - mask the unnecessary info from the dump
  crypto: qat - specify firmware files for 402xx
  crypto: x86/aes-gcm - simplify GCM hash subkey derivation
  crypto: x86/aes-gcm - delete unused GCM assembly code
  crypto: x86/aes-xts - simplify loop in xts_crypt_slowpath()
  hwrng: stm32 - repair clock handling
  ...

74 files changed, 8191 insertions, 1179 deletions

diff --git a/drivers/crypto/Kconfig b/drivers/crypto/Kconfig
index bfe18ebc2a98..94f23c6fc93b 100644
--- a/drivers/crypto/Kconfig
+++ b/drivers/crypto/Kconfig
@@ -644,6 +644,14 @@ config CRYPTO_DEV_ROCKCHIP_DEBUG
 	  This will create /sys/kernel/debug/rk3288_crypto/stats for displaying
 	  the number of requests per algorithm and other internal stats.
 
+config CRYPTO_DEV_TEGRA
+	tristate "Enable Tegra Security Engine"
+	depends on TEGRA_HOST1X
+	select CRYPTO_ENGINE
+
+	help
+	  Select this to enable Tegra Security Engine which accelerates various
+	  AES encryption/decryption and HASH algorithms.
 
 config CRYPTO_DEV_ZYNQMP_AES
 	tristate "Support for Xilinx ZynqMP AES hw accelerator"
diff --git a/drivers/crypto/Makefile b/drivers/crypto/Makefile
index 95331bc6456b..ad4ccef67d12 100644
--- a/drivers/crypto/Makefile
+++ b/drivers/crypto/Makefile
@@ -41,6 +41,7 @@ obj-$(CONFIG_CRYPTO_DEV_SAHARA) += sahara.o
 obj-$(CONFIG_CRYPTO_DEV_SL3516) += gemini/
 obj-y += stm32/
 obj-$(CONFIG_CRYPTO_DEV_TALITOS) += talitos.o
+obj-$(CONFIG_CRYPTO_DEV_TEGRA) += tegra/
 obj-$(CONFIG_CRYPTO_DEV_VIRTIO) += virtio/
 #obj-$(CONFIG_CRYPTO_DEV_VMX) += vmx/
 obj-$(CONFIG_CRYPTO_DEV_BCM_SPU) += bcm/
diff --git a/drivers/crypto/atmel-i2c.c b/drivers/crypto/atmel-i2c.c
index 83a9093eff25..a895e4289efa 100644
--- a/drivers/crypto/atmel-i2c.c
+++ b/drivers/crypto/atmel-i2c.c
@@ -51,7 +51,7 @@ static void atmel_i2c_checksum(struct atmel_i2c_cmd *cmd)
 	*__crc16 = cpu_to_le16(bitrev16(crc16(0, data, len)));
 }
 
-void atmel_i2c_init_read_cmd(struct atmel_i2c_cmd *cmd)
+void atmel_i2c_init_read_config_cmd(struct atmel_i2c_cmd *cmd)
 {
 	cmd->word_addr = COMMAND;
 	cmd->opcode = OPCODE_READ;
@@ -68,7 +68,31 @@ void atmel_i2c_init_read_cmd(struct atmel_i2c_cmd *cmd)
 	cmd->msecs = MAX_EXEC_TIME_READ;
 	cmd->rxsize = READ_RSP_SIZE;
 }
-EXPORT_SYMBOL(atmel_i2c_init_read_cmd);
+EXPORT_SYMBOL(atmel_i2c_init_read_config_cmd);
+
+int atmel_i2c_init_read_otp_cmd(struct atmel_i2c_cmd *cmd, u16 addr)
+{
+	if (addr < 0 || addr > OTP_ZONE_SIZE)
+		return -1;
+
+	cmd->word_addr = COMMAND;
+	cmd->opcode = OPCODE_READ;
+	/*
+	 * Read the word from OTP zone that may contain e.g. serial
+	 * numbers or similar if persistently pre-initialized and locked
+	 */
+	cmd->param1 = OTP_ZONE;
+	cmd->param2 = cpu_to_le16(addr);
+	cmd->count = READ_COUNT;
+
+	atmel_i2c_checksum(cmd);
+
+	cmd->msecs = MAX_EXEC_TIME_READ;
+	cmd->rxsize = READ_RSP_SIZE;
+
+	return 0;
+}
+EXPORT_SYMBOL(atmel_i2c_init_read_otp_cmd);
 
 void atmel_i2c_init_random_cmd(struct atmel_i2c_cmd *cmd)
 {
@@ -301,7 +325,7 @@ static int device_sanity_check(struct i2c_client *client)
 	if (!cmd)
 		return -ENOMEM;
 
-	atmel_i2c_init_read_cmd(cmd);
+	atmel_i2c_init_read_config_cmd(cmd);
 
 	ret = atmel_i2c_send_receive(client, cmd);
 	if (ret)
diff --git a/drivers/crypto/atmel-i2c.h b/drivers/crypto/atmel-i2c.h
index c0bd429ee2c7..72f04c15682f 100644
--- a/drivers/crypto/atmel-i2c.h
+++ b/drivers/crypto/atmel-i2c.h
@@ -64,6 +64,10 @@ struct atmel_i2c_cmd {
 
 /* Definitions for eeprom organization */
 #define CONFIGURATION_ZONE		0
+#define OTP_ZONE			1
+
+/* Definitions for eeprom zone sizes */
+#define OTP_ZONE_SIZE			64
 
 /* Definitions for Indexes common to all commands */
 #define RSP_DATA_IDX			1 /* buffer index of data in response */
@@ -124,6 +128,7 @@ struct atmel_ecc_driver_data {
  * @wake_token          : wake token array of zeros
  * @wake_token_sz       : size in bytes of the wake_token
  * @tfm_count           : number of active crypto transformations on i2c client
+ * @hwrng               : hold the hardware generated rng
  *
  * Reads and writes from/to the i2c client are sequential. The first byte
  * transmitted to the device is treated as the byte size. Any attempt to send
@@ -177,7 +182,8 @@ void atmel_i2c_flush_queue(void);
 
 int atmel_i2c_send_receive(struct i2c_client *client, struct atmel_i2c_cmd *cmd);
 
-void atmel_i2c_init_read_cmd(struct atmel_i2c_cmd *cmd);
+void atmel_i2c_init_read_config_cmd(struct atmel_i2c_cmd *cmd);
+int atmel_i2c_init_read_otp_cmd(struct atmel_i2c_cmd *cmd, u16 addr);
 void atmel_i2c_init_random_cmd(struct atmel_i2c_cmd *cmd);
 void atmel_i2c_init_genkey_cmd(struct atmel_i2c_cmd *cmd, u16 keyid);
 int atmel_i2c_init_ecdh_cmd(struct atmel_i2c_cmd *cmd,
diff --git a/drivers/crypto/atmel-sha204a.c b/drivers/crypto/atmel-sha204a.c
index c77f482d2a97..24ffdf505023 100644
--- a/drivers/crypto/atmel-sha204a.c
+++ b/drivers/crypto/atmel-sha204a.c
@@ -91,6 +91,62 @@ static int atmel_sha204a_rng_read(struct hwrng *rng, void *data, size_t max,
 	return max;
 }
 
+static int atmel_sha204a_otp_read(struct i2c_client *client, u16 addr, u8 *otp)
+{
+	struct atmel_i2c_cmd cmd;
+	int ret = -1;
+
+	if (atmel_i2c_init_read_otp_cmd(&cmd, addr) < 0) {
+		dev_err(&client->dev, "failed, invalid otp address %04X\n",
+			addr);
+		return ret;
+	}
+
+	ret = atmel_i2c_send_receive(client, &cmd);
+
+	if (cmd.data[0] == 0xff) {
+		dev_err(&client->dev, "failed, device not ready\n");
+		return -ret;
+	}
+
+	memcpy(otp, cmd.data+1, 4);
+
+	return ret;
+}
+
+static ssize_t otp_show(struct device *dev,
+			struct device_attribute *attr, char *buf)
+{
+	u16 addr;
+	u8 otp[OTP_ZONE_SIZE];
+	char *str = buf;
+	struct i2c_client *client = to_i2c_client(dev);
+	int i;
+
+	for (addr = 0; addr < OTP_ZONE_SIZE/4; addr++) {
+		if (atmel_sha204a_otp_read(client, addr, otp + addr * 4) < 0) {
+			dev_err(dev, "failed to read otp zone\n");
+			break;
+		}
+	}
+
+	for (i = 0; i < addr*2; i++)
+		str += sprintf(str, "%02X", otp[i]);
+	str += sprintf(str, "\n");
+	return str - buf;
+}
+static DEVICE_ATTR_RO(otp);
+
+static struct attribute *atmel_sha204a_attrs[] = {
+	&dev_attr_otp.attr,
+	NULL
+};
+
+static const struct attribute_group atmel_sha204a_groups = {
+	.name = "atsha204a",
+	.attrs = atmel_sha204a_attrs,
+};
+
 static int atmel_sha204a_probe(struct i2c_client *client)
 {
 	struct atmel_i2c_client_priv *i2c_priv;
@@ -111,6 +167,16 @@ static int atmel_sha204a_probe(struct i2c_client *client)
 	if (ret)
 		dev_warn(&client->dev, "failed to register RNG (%d)\n", ret);
 
+	/* otp read out */
+	if (!i2c_check_functionality(client->adapter, I2C_FUNC_I2C))
+		return -ENODEV;
+
+	ret = sysfs_create_group(&client->dev.kobj, &atmel_sha204a_groups);
+	if (ret) {
+		dev_err(&client->dev, "failed to register sysfs entry\n");
+		return ret;
+	}
+
 	return ret;
 }
 
@@ -123,6 +189,8 @@ static void atmel_sha204a_remove(struct i2c_client *client)
 		return;
 	}
 
+	sysfs_remove_group(&client->dev.kobj, &atmel_sha204a_groups);
+
 	kfree((void *)i2c_priv->hwrng.priv);
 }
 
diff --git a/drivers/crypto/bcm/spu2.c b/drivers/crypto/bcm/spu2.c
index 07989bb8c220..3fdc64b5a65e 100644
--- a/drivers/crypto/bcm/spu2.c
+++ b/drivers/crypto/bcm/spu2.c
@@ -495,7 +495,7 @@ static void spu2_dump_omd(u8 *omd, u16 hash_key_len, u16 ciph_key_len,
 	if (hash_iv_len) {
 		packet_log("  Hash IV Length %u bytes\n", hash_iv_len);
 		packet_dump("  hash IV: ", ptr, hash_iv_len);
-		ptr += ciph_key_len;
+		ptr += hash_iv_len;
 	}
 
 	if (ciph_iv_len) {
diff --git a/drivers/crypto/caam/ctrl.c b/drivers/crypto/caam/ctrl.c
index bdf367f3f679..bd418dea586d 100644
--- a/drivers/crypto/caam/ctrl.c
+++ b/drivers/crypto/caam/ctrl.c
@@ -512,6 +512,7 @@ static const struct of_device_id caam_match[] = {
 MODULE_DEVICE_TABLE(of, caam_match);
 
 struct caam_imx_data {
+	bool page0_access;
 	const struct clk_bulk_data *clks;
 	int num_clks;
 };
@@ -524,6 +525,7 @@ static const struct clk_bulk_data caam_imx6_clks[] = {
 };
 
 static const struct caam_imx_data caam_imx6_data = {
+	.page0_access = true,
 	.clks = caam_imx6_clks,
 	.num_clks = ARRAY_SIZE(caam_imx6_clks),
 };
@@ -534,6 +536,7 @@ static const struct clk_bulk_data caam_imx7_clks[] = {
 };
 
 static const struct caam_imx_data caam_imx7_data = {
+	.page0_access = true,
 	.clks = caam_imx7_clks,
 	.num_clks = ARRAY_SIZE(caam_imx7_clks),
 };
@@ -545,6 +548,7 @@ static const struct clk_bulk_data caam_imx6ul_clks[] = {
 };
 
 static const struct caam_imx_data caam_imx6ul_data = {
+	.page0_access = true,
 	.clks = caam_imx6ul_clks,
 	.num_clks = ARRAY_SIZE(caam_imx6ul_clks),
 };
@@ -554,15 +558,19 @@ static const struct clk_bulk_data caam_vf610_clks[] = {
 };
 
 static const struct caam_imx_data caam_vf610_data = {
+	.page0_access = true,
 	.clks = caam_vf610_clks,
 	.num_clks = ARRAY_SIZE(caam_vf610_clks),
 };
 
+static const struct caam_imx_data caam_imx8ulp_data;
+
 static const struct soc_device_attribute caam_imx_soc_table[] = {
 	{ .soc_id = "i.MX6UL", .data = &caam_imx6ul_data },
 	{ .soc_id = "i.MX6*",  .data = &caam_imx6_data },
 	{ .soc_id = "i.MX7*",  .data = &caam_imx7_data },
 	{ .soc_id = "i.MX8M*", .data = &caam_imx7_data },
+	{ .soc_id = "i.MX8ULP", .data = &caam_imx8ulp_data },
 	{ .soc_id = "VF*",     .data = &caam_vf610_data },
 	{ .family = "Freescale i.MX" },
 	{ /* sentinel */ }
@@ -860,6 +868,7 @@ static int caam_probe(struct platform_device *pdev)
 	int pg_size;
 	int BLOCK_OFFSET = 0;
 	bool reg_access = true;
+	const struct caam_imx_data *imx_soc_data;
 
 	ctrlpriv = devm_kzalloc(&pdev->dev, sizeof(*ctrlpriv), GFP_KERNEL);
 	if (!ctrlpriv)
@@ -894,12 +903,20 @@ static int caam_probe(struct platform_device *pdev)
 			return -EINVAL;
 		}
 
+		imx_soc_data = imx_soc_match->data;
+		reg_access = reg_access && imx_soc_data->page0_access;
+		/*
+		 * CAAM clocks cannot be controlled from kernel.
+		 */
+		if (!imx_soc_data->num_clks)
+			goto iomap_ctrl;
+
 		ret = init_clocks(dev, imx_soc_match->data);
 		if (ret)
 			return ret;
 	}
 
-
+iomap_ctrl:
 	/* Get configuration properties from device tree */
 	/* First, get register page */
 	ctrl = devm_of_iomap(dev, nprop, 0, NULL);
diff --git a/drivers/crypto/ccp/sp-platform.c b/drivers/crypto/ccp/sp-platform.c
index 473301237760..ff6ceb4feee0 100644
--- a/drivers/crypto/ccp/sp-platform.c
+++ b/drivers/crypto/ccp/sp-platform.c
@@ -39,44 +39,38 @@ static const struct sp_dev_vdata dev_vdata[] = {
 	},
 };
 
-#ifdef CONFIG_ACPI
 static const struct acpi_device_id sp_acpi_match[] = {
 	{ "AMDI0C00", (kernel_ulong_t)&dev_vdata[0] },
 	{ },
 };
 MODULE_DEVICE_TABLE(acpi, sp_acpi_match);
-#endif
 
-#ifdef CONFIG_OF
 static const struct of_device_id sp_of_match[] = {
 	{ .compatible = "amd,ccp-seattle-v1a",
 	  .data = (const void *)&dev_vdata[0] },
 	{ },
 };
 MODULE_DEVICE_TABLE(of, sp_of_match);
-#endif
 
 static struct sp_dev_vdata *sp_get_of_version(struct platform_device *pdev)
 {
-#ifdef CONFIG_OF
 	const struct of_device_id *match;
 
 	match = of_match_node(sp_of_match, pdev->dev.of_node);
 	if (match && match->data)
 		return (struct sp_dev_vdata *)match->data;
-#endif
+
 	return NULL;
 }
 
 static struct sp_dev_vdata *sp_get_acpi_version(struct platform_device *pdev)
 {
-#ifdef CONFIG_ACPI
 	const struct acpi_device_id *match;
 
 	match = acpi_match_device(sp_acpi_match, &pdev->dev);
 	if (match && match->driver_data)
 		return (struct sp_dev_vdata *)match->driver_data;
-#endif
+
 	return NULL;
 }
 
@@ -212,12 +206,8 @@ static int sp_platform_resume(struct platform_device *pdev)
 static struct platform_driver sp_platform_driver = {
 	.driver = {
 		.name = "ccp",
-#ifdef CONFIG_ACPI
 		.acpi_match_table = sp_acpi_match,
-#endif
-#ifdef CONFIG_OF
 		.of_match_table = sp_of_match,
-#endif
 	},
 	.probe = sp_platform_probe,
 	.remove_new = sp_platform_remove,
diff --git a/drivers/crypto/hisilicon/debugfs.c b/drivers/crypto/hisilicon/debugfs.c
index cd67fa348ca7..1b9b7bccdeff 100644
--- a/drivers/crypto/hisilicon/debugfs.c
+++ b/drivers/crypto/hisilicon/debugfs.c
@@ -13,6 +13,7 @@
 #define QM_DFX_COMMON_LEN		0xC3
 #define QM_DFX_REGS_LEN			4UL
 #define QM_DBG_TMP_BUF_LEN		22
+#define QM_XQC_ADDR_MASK		GENMASK(31, 0)
 #define CURRENT_FUN_MASK		GENMASK(5, 0)
 #define CURRENT_Q_MASK			GENMASK(31, 16)
 #define QM_SQE_ADDR_MASK		GENMASK(7, 0)
@@ -167,7 +168,6 @@ static void dump_show(struct hisi_qm *qm, void *info,
 static int qm_sqc_dump(struct hisi_qm *qm, char *s, char *name)
 {
 	struct device *dev = &qm->pdev->dev;
-	struct qm_sqc *sqc_curr;
 	struct qm_sqc sqc;
 	u32 qp_id;
 	int ret;
@@ -183,6 +183,8 @@ static int qm_sqc_dump(struct hisi_qm *qm, char *s, char *name)
 
 	ret = qm_set_and_get_xqc(qm, QM_MB_CMD_SQC, &sqc, qp_id, 1);
 	if (!ret) {
+		sqc.base_h = cpu_to_le32(QM_XQC_ADDR_MASK);
+		sqc.base_l = cpu_to_le32(QM_XQC_ADDR_MASK);
 		dump_show(qm, &sqc, sizeof(struct qm_sqc), name);
 
 		return 0;
@@ -190,9 +192,10 @@ static int qm_sqc_dump(struct hisi_qm *qm, char *s, char *name)
 
 	down_read(&qm->qps_lock);
 	if (qm->sqc) {
-		sqc_curr = qm->sqc + qp_id;
-
-		dump_show(qm, sqc_curr, sizeof(*sqc_curr), "SOFT SQC");
+		memcpy(&sqc, qm->sqc + qp_id * sizeof(struct qm_sqc), sizeof(struct qm_sqc));
+		sqc.base_h = cpu_to_le32(QM_XQC_ADDR_MASK);
+		sqc.base_l = cpu_to_le32(QM_XQC_ADDR_MASK);
+		dump_show(qm, &sqc, sizeof(struct qm_sqc), "SOFT SQC");
 	}
 	up_read(&qm->qps_lock);
 
@@ -202,7 +205,6 @@ static int qm_sqc_dump(struct hisi_qm *qm, char *s, char *name)
 static int qm_cqc_dump(struct hisi_qm *qm, char *s, char *name)
 {
 	struct device *dev = &qm->pdev->dev;
-	struct qm_cqc *cqc_curr;
 	struct qm_cqc cqc;
 	u32 qp_id;
 	int ret;
@@ -218,6 +220,8 @@ static int qm_cqc_dump(struct hisi_qm *qm, char *s, char *name)
 
 	ret = qm_set_and_get_xqc(qm, QM_MB_CMD_CQC, &cqc, qp_id, 1);
 	if (!ret) {
+		cqc.base_h = cpu_to_le32(QM_XQC_ADDR_MASK);
+		cqc.base_l = cpu_to_le32(QM_XQC_ADDR_MASK);
 		dump_show(qm, &cqc, sizeof(struct qm_cqc), name);
 
 		return 0;
@@ -225,9 +229,10 @@ static int qm_cqc_dump(struct hisi_qm *qm, char *s, char *name)
 
 	down_read(&qm->qps_lock);
 	if (qm->cqc) {
-		cqc_curr = qm->cqc + qp_id;
-
-		dump_show(qm, cqc_curr, sizeof(*cqc_curr), "SOFT CQC");
+		memcpy(&cqc, qm->cqc + qp_id * sizeof(struct qm_cqc), sizeof(struct qm_cqc));
+		cqc.base_h = cpu_to_le32(QM_XQC_ADDR_MASK);
+		cqc.base_l = cpu_to_le32(QM_XQC_ADDR_MASK);
+		dump_show(qm, &cqc, sizeof(struct qm_cqc), "SOFT CQC");
 	}
 	up_read(&qm->qps_lock);
 
@@ -263,6 +268,10 @@ static int qm_eqc_aeqc_dump(struct hisi_qm *qm, char *s, char *name)
 	if (ret)
 		return ret;
 
+	aeqc.base_h = cpu_to_le32(QM_XQC_ADDR_MASK);
+	aeqc.base_l = cpu_to_le32(QM_XQC_ADDR_MASK);
+	eqc.base_h = cpu_to_le32(QM_XQC_ADDR_MASK);
+	eqc.base_l = cpu_to_le32(QM_XQC_ADDR_MASK);
 	dump_show(qm, xeqc, size, name);
 
 	return ret;
@@ -310,27 +319,26 @@ static int q_dump_param_parse(struct hisi_qm *qm, char *s,
 
 static int qm_sq_dump(struct hisi_qm *qm, char *s, char *name)
 {
-	u16 sq_depth = qm->qp_array->cq_depth;
-	void *sqe, *sqe_curr;
+	u16 sq_depth = qm->qp_array->sq_depth;
 	struct hisi_qp *qp;
 	u32 qp_id, sqe_id;
+	void *sqe;
 	int ret;
 
 	ret = q_dump_param_parse(qm, s, &sqe_id, &qp_id, sq_depth);
 	if (ret)
 		return ret;
 
-	sqe = kzalloc(qm->sqe_size * sq_depth, GFP_KERNEL);
+	sqe = kzalloc(qm->sqe_size, GFP_KERNEL);
 	if (!sqe)
 		return -ENOMEM;
 
 	qp = &qm->qp_array[qp_id];
-	memcpy(sqe, qp->sqe, qm->sqe_size * sq_depth);
-	sqe_curr = sqe + (u32)(sqe_id * qm->sqe_size);
-	memset(sqe_curr + qm->debug.sqe_mask_offset, QM_SQE_ADDR_MASK,
+	memcpy(sqe, qp->sqe + sqe_id * qm->sqe_size, qm->sqe_size);
+	memset(sqe + qm->debug.sqe_mask_offset, QM_SQE_ADDR_MASK,
 	       qm->debug.sqe_mask_len);
 
-	dump_show(qm, sqe_curr, qm->sqe_size, name);
+	dump_show(qm, sqe, qm->sqe_size, name);
 
 	kfree(sqe);
 
@@ -809,8 +817,14 @@ static void dfx_regs_uninit(struct hisi_qm *qm,
 {
 	int i;
 
+	if (!dregs)
+		return;
+
 	/* Setting the pointer is NULL to prevent double free */
 	for (i = 0; i < reg_len; i++) {
+		if (!dregs[i].regs)
+			continue;
+
 		kfree(dregs[i].regs);
 		dregs[i].regs = NULL;
 	}
@@ -860,14 +874,21 @@ alloc_error:
 static int qm_diff_regs_init(struct hisi_qm *qm,
 		struct dfx_diff_registers *dregs, u32 reg_len)
 {
+	int ret;
+
 	qm->debug.qm_diff_regs = dfx_regs_init(qm, qm_diff_regs, ARRAY_SIZE(qm_diff_regs));
-	if (IS_ERR(qm->debug.qm_diff_regs))
-		return PTR_ERR(qm->debug.qm_diff_regs);
+	if (IS_ERR(qm->debug.qm_diff_regs)) {
+		ret = PTR_ERR(qm->debug.qm_diff_regs);
+		qm->debug.qm_diff_regs = NULL;
+		return ret;
+	}
 
 	qm->debug.acc_diff_regs = dfx_regs_init(qm, dregs, reg_len);
 	if (IS_ERR(qm->debug.acc_diff_regs)) {
 		dfx_regs_uninit(qm, qm->debug.qm_diff_regs, ARRAY_SIZE(qm_diff_regs));
-		return PTR_ERR(qm->debug.acc_diff_regs);
+		ret = PTR_ERR(qm->debug.acc_diff_regs);
+		qm->debug.acc_diff_regs = NULL;
+		return ret;
 	}
 
 	return 0;
@@ -908,7 +929,9 @@ static int qm_last_regs_init(struct hisi_qm *qm)
 static void qm_diff_regs_uninit(struct hisi_qm *qm, u32 reg_len)
 {
 	dfx_regs_uninit(qm, qm->debug.acc_diff_regs, reg_len);
+	qm->debug.acc_diff_regs = NULL;
 	dfx_regs_uninit(qm, qm->debug.qm_diff_regs, ARRAY_SIZE(qm_diff_regs));
+	qm->debug.qm_diff_regs = NULL;
 }
 
 /**
@@ -1075,12 +1098,12 @@ static void qm_create_debugfs_file(struct hisi_qm *qm, struct dentry *dir,
 {
 	struct debugfs_file *file = qm->debug.files + index;
 
-	debugfs_create_file(qm_debug_file_name[index], 0600, dir, file,
-			    &qm_debug_fops);
-
 	file->index = index;
 	mutex_init(&file->lock);
 	file->debug = &qm->debug;
+
+	debugfs_create_file(qm_debug_file_name[index], 0600, dir, file,
+			    &qm_debug_fops);
 }
 
 static int qm_debugfs_atomic64_set(void *data, u64 val)
diff --git a/drivers/crypto/hisilicon/hpre/hpre_main.c b/drivers/crypto/hisilicon/hpre/hpre_main.c
index d93aa6630a57..10aa4da93323 100644
--- a/drivers/crypto/hisilicon/hpre/hpre_main.c
+++ b/drivers/crypto/hisilicon/hpre/hpre_main.c
@@ -106,7 +106,7 @@
 #define HPRE_SHAPER_TYPE_RATE		640
 #define HPRE_VIA_MSI_DSM		1
 #define HPRE_SQE_MASK_OFFSET		8
-#define HPRE_SQE_MASK_LEN		24
+#define HPRE_SQE_MASK_LEN		44
 #define HPRE_CTX_Q_NUM_DEF		1
 
 #define HPRE_DFX_BASE		0x301000
@@ -1074,41 +1074,40 @@ static int hpre_debugfs_init(struct hisi_qm *qm)
 	struct device *dev = &qm->pdev->dev;
 	int ret;
 
-	qm->debug.debug_root = debugfs_create_dir(dev_name(dev),
-						  hpre_debugfs_root);
-
-	qm->debug.sqe_mask_offset = HPRE_SQE_MASK_OFFSET;
-	qm->debug.sqe_mask_len = HPRE_SQE_MASK_LEN;
 	ret = hisi_qm_regs_debugfs_init(qm, hpre_diff_regs, ARRAY_SIZE(hpre_diff_regs));
 	if (ret) {
 		dev_warn(dev, "Failed to init HPRE diff regs!\n");
-		goto debugfs_remove;
+		return ret;
 	}
 
+	qm->debug.debug_root = debugfs_create_dir(dev_name(dev),
+							hpre_debugfs_root);
+	qm->debug.sqe_mask_offset = HPRE_SQE_MASK_OFFSET;
+	qm->debug.sqe_mask_len = HPRE_SQE_MASK_LEN;
+
 	hisi_qm_debug_init(qm);
 
 	if (qm->pdev->device == PCI_DEVICE_ID_HUAWEI_HPRE_PF) {
 		ret = hpre_ctrl_debug_init(qm);
 		if (ret)
-			goto failed_to_create;
+			goto debugfs_remove;
 	}
 
 	hpre_dfx_debug_init(qm);
 
 	return 0;
 
-failed_to_create:
-	hisi_qm_regs_debugfs_uninit(qm, ARRAY_SIZE(hpre_diff_regs));
 debugfs_remove:
 	debugfs_remove_recursive(qm->debug.debug_root);
+	hisi_qm_regs_debugfs_uninit(qm, ARRAY_SIZE(hpre_diff_regs));
 	return ret;
 }
 
 static void hpre_debugfs_exit(struct hisi_qm *qm)
 {
-	hisi_qm_regs_debugfs_uninit(qm, ARRAY_SIZE(hpre_diff_regs));
-
 	debugfs_remove_recursive(qm->debug.debug_root);
+
+	hisi_qm_regs_debugfs_uninit(qm, ARRAY_SIZE(hpre_diff_regs));
 }
 
 static int hpre_pre_store_cap_reg(struct hisi_qm *qm)
diff --git a/drivers/crypto/hisilicon/qm.c b/drivers/crypto/hisilicon/qm.c
index 92f0a1d9b4a6..3dac8d8e8568 100644
--- a/drivers/crypto/hisilicon/qm.c
+++ b/drivers/crypto/hisilicon/qm.c
@@ -645,6 +645,9 @@ int qm_set_and_get_xqc(struct hisi_qm *qm, u8 cmd, void *xqc, u32 qp_id, bool op
 		tmp_xqc = qm->xqc_buf.aeqc;
 		xqc_dma = qm->xqc_buf.aeqc_dma;
 		break;
+	default:
+		dev_err(&qm->pdev->dev, "unknown mailbox cmd %u\n", cmd);
+		return -EINVAL;
 	}
 
 	/* Setting xqc will fail if master OOO is blocked. */
@@ -2893,12 +2896,9 @@ void hisi_qm_uninit(struct hisi_qm *qm)
 	hisi_qm_set_state(qm, QM_NOT_READY);
 	up_write(&qm->qps_lock);
 
+	qm_remove_uacce(qm);
 	qm_irqs_unregister(qm);
 	hisi_qm_pci_uninit(qm);
-	if (qm->use_sva) {
-		uacce_remove(qm->uacce);
-		qm->uacce = NULL;
-	}
 }
 EXPORT_SYMBOL_GPL(hisi_qm_uninit);
 
diff --git a/drivers/crypto/hisilicon/sec2/sec_crypto.c b/drivers/crypto/hisilicon/sec2/sec_crypto.c
index 93a972fcbf63..0558f98e221f 100644
--- a/drivers/crypto/hisilicon/sec2/sec_crypto.c
+++ b/drivers/crypto/hisilicon/sec2/sec_crypto.c
@@ -481,8 +481,10 @@ static void sec_alg_resource_free(struct sec_ctx *ctx,
 
 	if (ctx->pbuf_supported)
 		sec_free_pbuf_resource(dev, qp_ctx->res);
-	if (ctx->alg_type == SEC_AEAD)
+	if (ctx->alg_type == SEC_AEAD) {
 		sec_free_mac_resource(dev, qp_ctx->res);
+		sec_free_aiv_resource(dev, qp_ctx->res);
+	}
 }
 
 static int sec_alloc_qp_ctx_resource(struct sec_ctx *ctx, struct sec_qp_ctx *qp_ctx)
diff --git a/drivers/crypto/hisilicon/sec2/sec_main.c b/drivers/crypto/hisilicon/sec2/sec_main.c
index c290d8937b19..75aad04ffe5e 100644
--- a/drivers/crypto/hisilicon/sec2/sec_main.c
+++ b/drivers/crypto/hisilicon/sec2/sec_main.c
@@ -99,8 +99,8 @@
 #define SEC_DBGFS_VAL_MAX_LEN		20
 #define SEC_SINGLE_PORT_MAX_TRANS	0x2060
 
-#define SEC_SQE_MASK_OFFSET		64
-#define SEC_SQE_MASK_LEN		48
+#define SEC_SQE_MASK_OFFSET		16
+#define SEC_SQE_MASK_LEN		108
 #define SEC_SHAPER_TYPE_RATE		400
 
 #define SEC_DFX_BASE		0x301000
@@ -152,7 +152,7 @@ static const struct hisi_qm_cap_info sec_basic_info[] = {
 	{SEC_CORE_TYPE_NUM_CAP, 0x313c, 16, GENMASK(3, 0), 0x1, 0x1, 0x1},
 	{SEC_CORE_NUM_CAP, 0x313c, 8, GENMASK(7, 0), 0x4, 0x4, 0x4},
 	{SEC_CORES_PER_CLUSTER_NUM_CAP, 0x313c, 0, GENMASK(7, 0), 0x4, 0x4, 0x4},
-	{SEC_CORE_ENABLE_BITMAP, 0x3140, 32, GENMASK(31, 0), 0x17F, 0x17F, 0xF},
+	{SEC_CORE_ENABLE_BITMAP, 0x3140, 0, GENMASK(31, 0), 0x17F, 0x17F, 0xF},
 	{SEC_DRV_ALG_BITMAP_LOW, 0x3144, 0, GENMASK(31, 0), 0x18050CB, 0x18050CB, 0x18670CF},
 	{SEC_DRV_ALG_BITMAP_HIGH, 0x3148, 0, GENMASK(31, 0), 0x395C, 0x395C, 0x395C},
 	{SEC_DEV_ALG_BITMAP_LOW, 0x314c, 0, GENMASK(31, 0), 0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF},
@@ -901,37 +901,36 @@ static int sec_debugfs_init(struct hisi_qm *qm)
 	struct device *dev = &qm->pdev->dev;
 	int ret;
 
-	qm->debug.debug_root = debugfs_create_dir(dev_name(dev),
-						  sec_debugfs_root);
-	qm->debug.sqe_mask_offset = SEC_SQE_MASK_OFFSET;
-	qm->debug.sqe_mask_len = SEC_SQE_MASK_LEN;
-
 	ret = hisi_qm_regs_debugfs_init(qm, sec_diff_regs, ARRAY_SIZE(sec_diff_regs));
 	if (ret) {
 		dev_warn(dev, "Failed to init SEC diff regs!\n");
-		goto debugfs_remove;
+		return ret;
 	}
 
+	qm->debug.debug_root = debugfs_create_dir(dev_name(dev),
+							sec_debugfs_root);
+	qm->debug.sqe_mask_offset = SEC_SQE_MASK_OFFSET;
+	qm->debug.sqe_mask_len = SEC_SQE_MASK_LEN;
+
 	hisi_qm_debug_init(qm);
 
 	ret = sec_debug_init(qm);
 	if (ret)
-		goto failed_to_create;
+		goto debugfs_remove;
 
 	return 0;
 
-failed_to_create:
-	hisi_qm_regs_debugfs_uninit(qm, ARRAY_SIZE(sec_diff_regs));
 debugfs_remove:
-	debugfs_remove_recursive(sec_debugfs_root);
+	debugfs_remove_recursive(qm->debug.debug_root);
+	hisi_qm_regs_debugfs_uninit(qm, ARRAY_SIZE(sec_diff_regs));
 	return ret;
 }
 
 static void sec_debugfs_exit(struct hisi_qm *qm)
 {
-	hisi_qm_regs_debugfs_uninit(qm, ARRAY_SIZE(sec_diff_regs));
-
 	debugfs_remove_recursive(qm->debug.debug_root);
+
+	hisi_qm_regs_debugfs_uninit(qm, ARRAY_SIZE(sec_diff_regs));
 }
 
 static int sec_show_last_regs_init(struct hisi_qm *qm)
@@ -1324,7 +1323,8 @@ static struct pci_driver sec_pci_driver = {
 	.probe = sec_probe,
 	.remove = sec_remove,
 	.err_handler = &sec_err_handler,
-	.sriov_configure = hisi_qm_sriov_configure,
+	.sriov_configure = IS_ENABLED(CONFIG_PCI_IOV) ?
+				hisi_qm_sriov_configure : NULL,
 	.shutdown = hisi_qm_dev_shutdown,
 	.driver.pm = &sec_pm_ops,
 };
diff --git a/drivers/crypto/hisilicon/sgl.c b/drivers/crypto/hisilicon/sgl.c
index 0beca257c20b..568acd0aee3f 100644
--- a/drivers/crypto/hisilicon/sgl.c
+++ b/drivers/crypto/hisilicon/sgl.c
@@ -161,9 +161,6 @@ static struct hisi_acc_hw_sgl *acc_get_sgl(struct hisi_acc_sgl_pool *pool,
 	struct mem_block *block;
 	u32 block_index, offset;
 
-	if (!pool || !hw_sgl_dma || index >= pool->count)
-		return ERR_PTR(-EINVAL);
-
 	block = pool->mem_block;
 	block_index = index / pool->sgl_num_per_block;
 	offset = index % pool->sgl_num_per_block;
@@ -230,7 +227,7 @@ hisi_acc_sg_buf_map_to_hw_sgl(struct device *dev,
 	struct scatterlist *sg;
 	int sg_n;
 
-	if (!dev || !sgl || !pool || !hw_sgl_dma)
+	if (!dev || !sgl || !pool || !hw_sgl_dma || index >= pool->count)
 		return ERR_PTR(-EINVAL);
 
 	sg_n = sg_nents(sgl);
diff --git a/drivers/crypto/hisilicon/zip/zip_main.c b/drivers/crypto/hisilicon/zip/zip_main.c
index c065fd867161..c94a7b20d07e 100644
--- a/drivers/crypto/hisilicon/zip/zip_main.c
+++ b/drivers/crypto/hisilicon/zip/zip_main.c
@@ -887,36 +887,34 @@ static int hisi_zip_ctrl_debug_init(struct hisi_qm *qm)
 static int hisi_zip_debugfs_init(struct hisi_qm *qm)
 {
 	struct device *dev = &qm->pdev->dev;
-	struct dentry *dev_d;
 	int ret;
 
-	dev_d = debugfs_create_dir(dev_name(dev), hzip_debugfs_root);
-
-	qm->debug.sqe_mask_offset = HZIP_SQE_MASK_OFFSET;
-	qm->debug.sqe_mask_len = HZIP_SQE_MASK_LEN;
-	qm->debug.debug_root = dev_d;
 	ret = hisi_qm_regs_debugfs_init(qm, hzip_diff_regs, ARRAY_SIZE(hzip_diff_regs));
 	if (ret) {
 		dev_warn(dev, "Failed to init ZIP diff regs!\n");
-		goto debugfs_remove;
+		return ret;
 	}
 
+	qm->debug.sqe_mask_offset = HZIP_SQE_MASK_OFFSET;
+	qm->debug.sqe_mask_len = HZIP_SQE_MASK_LEN;
+	qm->debug.debug_root = debugfs_create_dir(dev_name(dev),
+							hzip_debugfs_root);
+
 	hisi_qm_debug_init(qm);
 
 	if (qm->fun_type == QM_HW_PF) {
 		ret = hisi_zip_ctrl_debug_init(qm);
 		if (ret)
-			goto failed_to_create;
+			goto debugfs_remove;
 	}
 
 	hisi_zip_dfx_debug_init(qm);
 
 	return 0;
 
-failed_to_create:
-	hisi_qm_regs_debugfs_uninit(qm, ARRAY_SIZE(hzip_diff_regs));
 debugfs_remove:
-	debugfs_remove_recursive(hzip_debugfs_root);
+	debugfs_remove_recursive(qm->debug.debug_root);
+	hisi_qm_regs_debugfs_uninit(qm, ARRAY_SIZE(hzip_diff_regs));
 	return ret;
 }
 
@@ -940,10 +938,10 @@ static void hisi_zip_debug_regs_clear(struct hisi_qm *qm)
 
 static void hisi_zip_debugfs_exit(struct hisi_qm *qm)
 {
-	hisi_qm_regs_debugfs_uninit(qm, ARRAY_SIZE(hzip_diff_regs));
-
 	debugfs_remove_recursive(qm->debug.debug_root);
 
+	hisi_qm_regs_debugfs_uninit(qm, ARRAY_SIZE(hzip_diff_regs));
+
 	if (qm->fun_type == QM_HW_PF) {
 		hisi_zip_debug_regs_clear(qm);
 		qm->debug.curr_qm_qp_num = 0;
diff --git a/drivers/crypto/intel/iaa/iaa_crypto.h b/drivers/crypto/intel/iaa/iaa_crypto.h
index 2524091a5f70..56985e395263 100644
--- a/drivers/crypto/intel/iaa/iaa_crypto.h
+++ b/drivers/crypto/intel/iaa/iaa_crypto.h
@@ -49,10 +49,10 @@ struct iaa_wq {
 
 	struct iaa_device	*iaa_device;
 
-	u64			comp_calls;
-	u64			comp_bytes;
-	u64			decomp_calls;
-	u64			decomp_bytes;
+	atomic64_t		comp_calls;
+	atomic64_t		comp_bytes;
+	atomic64_t		decomp_calls;
+	atomic64_t		decomp_bytes;
 };
 
 struct iaa_device_compression_mode {
@@ -73,10 +73,10 @@ struct iaa_device {
 	int				n_wq;
 	struct list_head		wqs;
 
-	u64				comp_calls;
-	u64				comp_bytes;
-	u64				decomp_calls;
-	u64				decomp_bytes;
+	atomic64_t			comp_calls;
+	atomic64_t			comp_bytes;
+	atomic64_t			decomp_calls;
+	atomic64_t			decomp_bytes;
 };
 
 struct wq_table_entry {
diff --git a/drivers/crypto/intel/iaa/iaa_crypto_main.c b/drivers/crypto/intel/iaa/iaa_crypto_main.c
index b2191ade9011..e810d286ee8c 100644
--- a/drivers/crypto/intel/iaa/iaa_crypto_main.c
+++ b/drivers/crypto/intel/iaa/iaa_crypto_main.c
@@ -347,18 +347,16 @@ int add_iaa_compression_mode(const char *name,
 		goto free;
 
 	if (ll_table) {
-		mode->ll_table = kzalloc(ll_table_size, GFP_KERNEL);
+		mode->ll_table = kmemdup(ll_table, ll_table_size, GFP_KERNEL);
 		if (!mode->ll_table)
 			goto free;
-		memcpy(mode->ll_table, ll_table, ll_table_size);
 		mode->ll_table_size = ll_table_size;
 	}
 
 	if (d_table) {
-		mode->d_table = kzalloc(d_table_size, GFP_KERNEL);
+		mode->d_table = kmemdup(d_table, d_table_size, GFP_KERNEL);
 		if (!mode->d_table)
 			goto free;
-		memcpy(mode->d_table, d_table, d_table_size);
 		mode->d_table_size = d_table_size;
 	}
 
@@ -922,7 +920,7 @@ static void rebalance_wq_table(void)
 	for_each_node_with_cpus(node) {
 		node_cpus = cpumask_of_node(node);
 
-		for (cpu = 0; cpu < nr_cpus_per_node; cpu++) {
+		for (cpu = 0; cpu <  cpumask_weight(node_cpus); cpu++) {
 			int node_cpu = cpumask_nth(cpu, node_cpus);
 
 			if (WARN_ON(node_cpu >= nr_cpu_ids)) {
@@ -1079,8 +1077,8 @@ static void iaa_desc_complete(struct idxd_desc *idxd_desc,
 		update_total_comp_bytes_out(ctx->req->dlen);
 		update_wq_comp_bytes(iaa_wq->wq, ctx->req->dlen);
 	} else {
-		update_total_decomp_bytes_in(ctx->req->dlen);
-		update_wq_decomp_bytes(iaa_wq->wq, ctx->req->dlen);
+		update_total_decomp_bytes_in(ctx->req->slen);
+		update_wq_decomp_bytes(iaa_wq->wq, ctx->req->slen);
 	}
 
 	if (ctx->compress && compression_ctx->verify_compress) {
@@ -1498,7 +1496,6 @@ static int iaa_comp_acompress(struct acomp_req *req)
 	u32 compression_crc;
 	struct idxd_wq *wq;
 	struct device *dev;
-	u64 start_time_ns;
 	int order = -1;
 
 	compression_ctx = crypto_tfm_ctx(tfm);
@@ -1572,10 +1569,8 @@ static int iaa_comp_acompress(struct acomp_req *req)
 		" req->dlen %d, sg_dma_len(sg) %d\n", dst_addr, nr_sgs,
 		req->dst, req->dlen, sg_dma_len(req->dst));
 
-	start_time_ns = iaa_get_ts();
 	ret = iaa_compress(tfm, req, wq, src_addr, req->slen, dst_addr,
 			   &req->dlen, &compression_crc, disable_async);
-	update_max_comp_delay_ns(start_time_ns);
 	if (ret == -EINPROGRESS)
 		return ret;
 
@@ -1622,7 +1617,6 @@ static int iaa_comp_adecompress_alloc_dest(struct acomp_req *req)
 	struct iaa_wq *iaa_wq;
 	struct device *dev;
 	struct idxd_wq *wq;
-	u64 start_time_ns;
 	int order = -1;
 
 	cpu = get_cpu();
@@ -1679,10 +1673,8 @@ alloc_dest:
 	dev_dbg(dev, "dma_map_sg, dst_addr %llx, nr_sgs %d, req->dst %p,"
 		" req->dlen %d, sg_dma_len(sg) %d\n", dst_addr, nr_sgs,
 		req->dst, req->dlen, sg_dma_len(req->dst));
-	start_time_ns = iaa_get_ts();
 	ret = iaa_decompress(tfm, req, wq, src_addr, req->slen,
 			     dst_addr, &req->dlen, true);
-	update_max_decomp_delay_ns(start_time_ns);
 	if (ret == -EOVERFLOW) {
 		dma_unmap_sg(dev, req->dst, sg_nents(req->dst), DMA_FROM_DEVICE);
 		req->dlen *= 2;
@@ -1713,7 +1705,6 @@ static int iaa_comp_adecompress(struct acomp_req *req)
 	int nr_sgs, cpu, ret = 0;
 	struct iaa_wq *iaa_wq;
 	struct device *dev;
-	u64 start_time_ns;
 	struct idxd_wq *wq;
 
 	if (!iaa_crypto_enabled) {
@@ -1773,10 +1764,8 @@ static int iaa_comp_adecompress(struct acomp_req *req)
 		" req->dlen %d, sg_dma_len(sg) %d\n", dst_addr, nr_sgs,
 		req->dst, req->dlen, sg_dma_len(req->dst));
 
-	start_time_ns = iaa_get_ts();
 	ret = iaa_decompress(tfm, req, wq, src_addr, req->slen,
 			     dst_addr, &req->dlen, false);
-	update_max_decomp_delay_ns(start_time_ns);
 	if (ret == -EINPROGRESS)
 		return ret;
 
@@ -2014,7 +2003,7 @@ static int __init iaa_crypto_init_module(void)
 	int ret = 0;
 	int node;
 
-	nr_cpus = num_online_cpus();
+	nr_cpus = num_possible_cpus();
 	for_each_node_with_cpus(node)
 		nr_nodes++;
 	if (!nr_nodes) {
diff --git a/drivers/crypto/intel/iaa/iaa_crypto_stats.c b/drivers/crypto/intel/iaa/iaa_crypto_stats.c
index c9f83af4b307..f5cc3d29ca19 100644
--- a/drivers/crypto/intel/iaa/iaa_crypto_stats.c
+++ b/drivers/crypto/intel/iaa/iaa_crypto_stats.c
@@ -17,141 +17,117 @@
 #include "iaa_crypto.h"
 #include "iaa_crypto_stats.h"
 
-static u64 total_comp_calls;
-static u64 total_decomp_calls;
-static u64 total_sw_decomp_calls;
-static u64 max_comp_delay_ns;
-static u64 max_decomp_delay_ns;
-static u64 total_comp_bytes_out;
-static u64 total_decomp_bytes_in;
-static u64 total_completion_einval_errors;
-static u64 total_completion_timeout_errors;
-static u64 total_completion_comp_buf_overflow_errors;
+static atomic64_t total_comp_calls;
+static atomic64_t total_decomp_calls;
+static atomic64_t total_sw_decomp_calls;
+static atomic64_t total_comp_bytes_out;
+static atomic64_t total_decomp_bytes_in;
+static atomic64_t total_completion_einval_errors;
+static atomic64_t total_completion_timeout_errors;
+static atomic64_t total_completion_comp_buf_overflow_errors;
 
 static struct dentry *iaa_crypto_debugfs_root;
 
 void update_total_comp_calls(void)
 {
-	total_comp_calls++;
+	atomic64_inc(&total_comp_calls);
 }
 
 void update_total_comp_bytes_out(int n)
 {
-	total_comp_bytes_out += n;
+	atomic64_add(n, &total_comp_bytes_out);
 }
 
 void update_total_decomp_calls(void)
 {
-	total_decomp_calls++;
+	atomic64_inc(&total_decomp_calls);
 }
 
 void update_total_sw_decomp_calls(void)
 {
-	total_sw_decomp_calls++;
+	atomic64_inc(&total_sw_decomp_calls);
 }
 
 void update_total_decomp_bytes_in(int n)
 {
-	total_decomp_bytes_in += n;
+	atomic64_add(n, &total_decomp_bytes_in);
 }
 
 void update_completion_einval_errs(void)
 {
-	total_completion_einval_errors++;
+	atomic64_inc(&total_completion_einval_errors);
 }
 
 void update_completion_timeout_errs(void)
 {
-	total_completion_timeout_errors++;
+	atomic64_inc(&total_completion_timeout_errors);
 }
 
 void update_completion_comp_buf_overflow_errs(void)
 {
-	total_completion_comp_buf_overflow_errors++;
-}
-
-void update_max_comp_delay_ns(u64 start_time_ns)
-{
-	u64 time_diff;
-
-	time_diff = ktime_get_ns() - start_time_ns;
-
-	if (time_diff > max_comp_delay_ns)
-		max_comp_delay_ns = time_diff;
-}
-
-void update_max_decomp_delay_ns(u64 start_time_ns)
-{
-	u64 time_diff;
-
-	time_diff = ktime_get_ns() - start_time_ns;
-
-	if (time_diff > max_decomp_delay_ns)
-		max_decomp_delay_ns = time_diff;
+	atomic64_inc(&total_completion_comp_buf_overflow_errors);
 }
 
 void update_wq_comp_calls(struct idxd_wq *idxd_wq)
 {
 	struct iaa_wq *wq = idxd_wq_get_private(idxd_wq);
 
-	wq->comp_calls++;
-	wq->iaa_device->comp_calls++;
+	atomic64_inc(&wq->comp_calls);
+	atomic64_inc(&wq->iaa_device->comp_calls);
 }
 
 void update_wq_comp_bytes(struct idxd_wq *idxd_wq, int n)
 {
 	struct iaa_wq *wq = idxd_wq_get_private(idxd_wq);
 
-	wq->comp_bytes += n;
-	wq->iaa_device->comp_bytes += n;
+	atomic64_add(n, &wq->comp_bytes);
+	atomic64_add(n, &wq->iaa_device->comp_bytes);
 }
 
 void update_wq_decomp_calls(struct idxd_wq *idxd_wq)
 {
 	struct iaa_wq *wq = idxd_wq_get_private(idxd_wq);
 
-	wq->decomp_calls++;
-	wq->iaa_device->decomp_calls++;
+	atomic64_inc(&wq->decomp_calls);
+	atomic64_inc(&wq->iaa_device->decomp_calls);
 }
 
 void update_wq_decomp_bytes(struct idxd_wq *idxd_wq, int n)
 {
 	struct iaa_wq *wq = idxd_wq_get_private(idxd_wq);
 
-	wq->decomp_bytes += n;
-	wq->iaa_device->decomp_bytes += n;
+	atomic64_add(n, &wq->decomp_bytes);
+	atomic64_add(n, &wq->iaa_device->decomp_bytes);
 }
 
 static void reset_iaa_crypto_stats(void)
 {
-	total_comp_calls = 0;
-	total_decomp_calls = 0;
-	total_sw_decomp_calls = 0;
-	max_comp_delay_ns = 0;
-	max_decomp_delay_ns = 0;
-	total_comp_bytes_out = 0;
-	total_decomp_bytes_in = 0;
-	total_completion_einval_errors = 0;
-	total_completion_timeout_errors = 0;
-	total_completion_comp_buf_overflow_errors = 0;
+	atomic64_set(&total_comp_calls, 0);
+	atomic64_set(&total_decomp_calls, 0);
+	atomic64_set(&total_sw_decomp_calls, 0);
+	atomic64_set(&total_comp_bytes_out, 0);
+	atomic64_set(&total_decomp_bytes_in, 0);
+	atomic64_set(&total_completion_einval_errors, 0);
+	atomic64_set(&total_completion_timeout_errors, 0);
+	atomic64_set(&total_completion_comp_buf_overflow_errors, 0);
 }
 
 static void reset_wq_stats(struct iaa_wq *wq)
 {
-	wq->comp_calls = 0;
-	wq->comp_bytes = 0;
-	wq->decomp_calls = 0;
-	wq->decomp_bytes = 0;
+	atomic64_set(&wq->comp_calls, 0);
+	atomic64_set(&wq->comp_bytes, 0);
+	atomic64_set(&wq->decomp_calls, 0);
+	atomic64_set(&wq->decomp_bytes, 0);
 }
 
 static void reset_device_stats(struct iaa_device *iaa_device)
 {
 	struct iaa_wq *iaa_wq;
 
-	iaa_device->comp_calls = 0;
-	iaa_device->comp_bytes = 0;
-	iaa_device->decomp_calls = 0;
-	iaa_device->decomp_bytes = 0;
+	atomic64_set(&iaa_device->comp_calls, 0);
+	atomic64_set(&iaa_device->comp_bytes, 0);
+	atomic64_set(&iaa_device->decomp_calls, 0);
+	atomic64_set(&iaa_device->decomp_bytes, 0);
 
 	list_for_each_entry(iaa_wq, &iaa_device->wqs, list)
 		reset_wq_stats(iaa_wq);
@@ -160,10 +136,14 @@ static void reset_device_stats(struct iaa_device *iaa_device)
 static void wq_show(struct seq_file *m, struct iaa_wq *iaa_wq)
 {
 	seq_printf(m, "    name: %s\n", iaa_wq->wq->name);
-	seq_printf(m, "    comp_calls: %llu\n", iaa_wq->comp_calls);
-	seq_printf(m, "    comp_bytes: %llu\n", iaa_wq->comp_bytes);
-	seq_printf(m, "    decomp_calls: %llu\n", iaa_wq->decomp_calls);
-	seq_printf(m, "    decomp_bytes: %llu\n\n", iaa_wq->decomp_bytes);
+	seq_printf(m, "    comp_calls: %llu\n",
+		   atomic64_read(&iaa_wq->comp_calls));
+	seq_printf(m, "    comp_bytes: %llu\n",
+		   atomic64_read(&iaa_wq->comp_bytes));
+	seq_printf(m, "    decomp_calls: %llu\n",
+		   atomic64_read(&iaa_wq->decomp_calls));
+	seq_printf(m, "    decomp_bytes: %llu\n\n",
+		   atomic64_read(&iaa_wq->decomp_bytes));
 }
 
 static void device_stats_show(struct seq_file *m, struct iaa_device *iaa_device)
@@ -173,30 +153,41 @@ static void device_stats_show(struct seq_file *m, struct iaa_device *iaa_device)
 	seq_puts(m, "iaa device:\n");
 	seq_printf(m, "  id: %d\n", iaa_device->idxd->id);
 	seq_printf(m, "  n_wqs: %d\n", iaa_device->n_wq);
-	seq_printf(m, "  comp_calls: %llu\n", iaa_device->comp_calls);
-	seq_printf(m, "  comp_bytes: %llu\n", iaa_device->comp_bytes);
-	seq_printf(m, "  decomp_calls: %llu\n", iaa_device->decomp_calls);
-	seq_printf(m, "  decomp_bytes: %llu\n", iaa_device->decomp_bytes);
+	seq_printf(m, "  comp_calls: %llu\n",
+		   atomic64_read(&iaa_device->comp_calls));
+	seq_printf(m, "  comp_bytes: %llu\n",
+		   atomic64_read(&iaa_device->comp_bytes));
+	seq_printf(m, "  decomp_calls: %llu\n",
+		   atomic64_read(&iaa_device->decomp_calls));
+	seq_printf(m, "  decomp_bytes: %llu\n",
+		   atomic64_read(&iaa_device->decomp_bytes));
 	seq_puts(m, "  wqs:\n");
 
 	list_for_each_entry(iaa_wq, &iaa_device->wqs, list)
 		wq_show(m, iaa_wq);
 }
 
-static void global_stats_show(struct seq_file *m)
+static int global_stats_show(struct seq_file *m, void *v)
 {
 	seq_puts(m, "global stats:\n");
-	seq_printf(m, "  total_comp_calls: %llu\n", total_comp_calls);
-	seq_printf(m, "  total_decomp_calls: %llu\n", total_decomp_calls);
-	seq_printf(m, "  total_sw_decomp_calls: %llu\n", total_sw_decomp_calls);
-	seq_printf(m, "  total_comp_bytes_out: %llu\n", total_comp_bytes_out);
-	seq_printf(m, "  total_decomp_bytes_in: %llu\n", total_decomp_bytes_in);
+	seq_printf(m, "  total_comp_calls: %llu\n",
+		   atomic64_read(&total_comp_calls));
+	seq_printf(m, "  total_decomp_calls: %llu\n",
+		   atomic64_read(&total_decomp_calls));
+	seq_printf(m, "  total_sw_decomp_calls: %llu\n",
+		   atomic64_read(&total_sw_decomp_calls));
+	seq_printf(m, "  total_comp_bytes_out: %llu\n",
+		   atomic64_read(&total_comp_bytes_out));
+	seq_printf(m, "  total_decomp_bytes_in: %llu\n",
+		   atomic64_read(&total_decomp_bytes_in));
 	seq_printf(m, "  total_completion_einval_errors: %llu\n",
-		   total_completion_einval_errors);
+		   atomic64_read(&total_completion_einval_errors));
 	seq_printf(m, "  total_completion_timeout_errors: %llu\n",
-		   total_completion_timeout_errors);
+		   atomic64_read(&total_completion_timeout_errors));
 	seq_printf(m, "  total_completion_comp_buf_overflow_errors: %llu\n\n",
-		   total_completion_comp_buf_overflow_errors);
+		   atomic64_read(&total_completion_comp_buf_overflow_errors));
+
+	return 0;
 }
 
 static int wq_stats_show(struct seq_file *m, void *v)
@@ -205,8 +196,6 @@ static int wq_stats_show(struct seq_file *m, void *v)
 
 	mutex_lock(&iaa_devices_lock);
 
-	global_stats_show(m);
-
 	list_for_each_entry(iaa_device, &iaa_devices, list)
 		device_stats_show(m, iaa_device);
 
@@ -243,6 +232,18 @@ static const struct file_operations wq_stats_fops = {
 	.release = single_release,
 };
 
+static int global_stats_open(struct inode *inode, struct file *file)
+{
+	return single_open(file, global_stats_show, file);
+}
+
+static const struct file_operations global_stats_fops = {
+	.open = global_stats_open,
+	.read = seq_read,
+	.llseek = seq_lseek,
+	.release = single_release,
+};
+
 DEFINE_DEBUGFS_ATTRIBUTE(wq_stats_reset_fops, NULL, iaa_crypto_stats_reset, "%llu\n");
 
 int __init iaa_crypto_debugfs_init(void)
@@ -252,20 +253,8 @@ int __init iaa_crypto_debugfs_init(void)
 
 	iaa_crypto_debugfs_root = debugfs_create_dir("iaa_crypto", NULL);
 
-	debugfs_create_u64("max_comp_delay_ns", 0644,
-			   iaa_crypto_debugfs_root, &max_comp_delay_ns);
-	debugfs_create_u64("max_decomp_delay_ns", 0644,
-			   iaa_crypto_debugfs_root, &max_decomp_delay_ns);
-	debugfs_create_u64("total_comp_calls", 0644,
-			   iaa_crypto_debugfs_root, &total_comp_calls);
-	debugfs_create_u64("total_decomp_calls", 0644,
-			   iaa_crypto_debugfs_root, &total_decomp_calls);
-	debugfs_create_u64("total_sw_decomp_calls", 0644,
-			   iaa_crypto_debugfs_root, &total_sw_decomp_calls);
-	debugfs_create_u64("total_comp_bytes_out", 0644,
-			   iaa_crypto_debugfs_root, &total_comp_bytes_out);
-	debugfs_create_u64("total_decomp_bytes_in", 0644,
-			   iaa_crypto_debugfs_root, &total_decomp_bytes_in);
+	debugfs_create_file("global_stats", 0644, iaa_crypto_debugfs_root, NULL,
+			    &global_stats_fops);
 	debugfs_create_file("wq_stats", 0644, iaa_crypto_debugfs_root, NULL,
 			    &wq_stats_fops);
 	debugfs_create_file("stats_reset", 0644, iaa_crypto_debugfs_root, NULL,
diff --git a/drivers/crypto/intel/iaa/iaa_crypto_stats.h b/drivers/crypto/intel/iaa/iaa_crypto_stats.h
index c916ca83f070..3787a5f507eb 100644
--- a/drivers/crypto/intel/iaa/iaa_crypto_stats.h
+++ b/drivers/crypto/intel/iaa/iaa_crypto_stats.h
@@ -13,8 +13,6 @@ void	update_total_comp_bytes_out(int n);
 void	update_total_decomp_calls(void);
 void	update_total_sw_decomp_calls(void);
 void	update_total_decomp_bytes_in(int n);
-void	update_max_comp_delay_ns(u64 start_time_ns);
-void	update_max_decomp_delay_ns(u64 start_time_ns);
 void	update_completion_einval_errs(void);
 void	update_completion_timeout_errs(void);
 void	update_completion_comp_buf_overflow_errs(void);
@@ -24,8 +22,6 @@ void	update_wq_comp_bytes(struct idxd_wq *idxd_wq, int n);
 void	update_wq_decomp_calls(struct idxd_wq *idxd_wq);
 void	update_wq_decomp_bytes(struct idxd_wq *idxd_wq, int n);
 
-static inline u64	iaa_get_ts(void) { return ktime_get_ns(); }
-
 #else
 static inline int	iaa_crypto_debugfs_init(void) { return 0; }
 static inline void	iaa_crypto_debugfs_cleanup(void) {}
@@ -35,8 +31,6 @@ static inline void	update_total_comp_bytes_out(int n) {}
 static inline void	update_total_decomp_calls(void) {}
 static inline void	update_total_sw_decomp_calls(void) {}
 static inline void	update_total_decomp_bytes_in(int n) {}
-static inline void	update_max_comp_delay_ns(u64 start_time_ns) {}
-static inline void	update_max_decomp_delay_ns(u64 start_time_ns) {}
 static inline void	update_completion_einval_errs(void) {}
 static inline void	update_completion_timeout_errs(void) {}
 static inline void	update_completion_comp_buf_overflow_errs(void) {}
@@ -46,8 +40,6 @@ static inline void	update_wq_comp_bytes(struct idxd_wq *idxd_wq, int n) {}
 static inline void	update_wq_decomp_calls(struct idxd_wq *idxd_wq) {}
 static inline void	update_wq_decomp_bytes(struct idxd_wq *idxd_wq, int n) {}
 
-static inline u64	iaa_get_ts(void) { return 0; }
-
 #endif // CONFIG_CRYPTO_DEV_IAA_CRYPTO_STATS
 
 #endif
diff --git a/drivers/crypto/intel/qat/qat_420xx/adf_420xx_hw_data.c b/drivers/crypto/intel/qat/qat_420xx/adf_420xx_hw_data.c
index 1102c47f8293..78f0ea49254d 100644
--- a/drivers/crypto/intel/qat/qat_420xx/adf_420xx_hw_data.c
+++ b/drivers/crypto/intel/qat/qat_420xx/adf_420xx_hw_data.c
@@ -10,12 +10,14 @@
 #include <adf_fw_config.h>
 #include <adf_gen4_config.h>
 #include <adf_gen4_dc.h>
+#include <adf_gen4_hw_csr_data.h>
 #include <adf_gen4_hw_data.h>
 #include <adf_gen4_pfvf.h>
 #include <adf_gen4_pm.h>
 #include <adf_gen4_ras.h>
 #include <adf_gen4_timer.h>
 #include <adf_gen4_tl.h>
+#include <adf_gen4_vf_mig.h>
 #include "adf_420xx_hw_data.h"
 #include "icp_qat_hw.h"
 
@@ -296,7 +298,7 @@ static const u32 *adf_get_arbiter_mapping(struct adf_accel_dev *accel_dev)
 {
 	if (adf_gen4_init_thd2arb_map(accel_dev))
 		dev_warn(&GET_DEV(accel_dev),
-			 "Generate of the thread to arbiter map failed");
+			 "Failed to generate thread to arbiter mapping");
 
 	return GET_HW_DATA(accel_dev)->thd_to_arb_map;
 }
@@ -487,6 +489,7 @@ void adf_init_hw_data_420xx(struct adf_hw_device_data *hw_data, u32 dev_id)
 	adf_gen4_init_dc_ops(&hw_data->dc_ops);
 	adf_gen4_init_ras_ops(&hw_data->ras_ops);
 	adf_gen4_init_tl_data(&hw_data->tl_data);
+	adf_gen4_init_vf_mig_ops(&hw_data->vfmig_ops);
 	adf_init_rl_data(&hw_data->rl_data);
 }
 
diff --git a/drivers/crypto/intel/qat/qat_4xxx/adf_4xxx_hw_data.c b/drivers/crypto/intel/qat/qat_4xxx/adf_4xxx_hw_data.c
index 927506cf271d..9fd7ec53b9f3 100644
--- a/drivers/crypto/intel/qat/qat_4xxx/adf_4xxx_hw_data.c
+++ b/drivers/crypto/intel/qat/qat_4xxx/adf_4xxx_hw_data.c
@@ -10,12 +10,14 @@
 #include <adf_fw_config.h>
 #include <adf_gen4_config.h>
 #include <adf_gen4_dc.h>
+#include <adf_gen4_hw_csr_data.h>
 #include <adf_gen4_hw_data.h>
 #include <adf_gen4_pfvf.h>
 #include <adf_gen4_pm.h>
 #include "adf_gen4_ras.h"
 #include <adf_gen4_timer.h>
 #include <adf_gen4_tl.h>
+#include <adf_gen4_vf_mig.h>
 #include "adf_4xxx_hw_data.h"
 #include "icp_qat_hw.h"
 
@@ -208,7 +210,7 @@ static const u32 *adf_get_arbiter_mapping(struct adf_accel_dev *accel_dev)
 {
 	if (adf_gen4_init_thd2arb_map(accel_dev))
 		dev_warn(&GET_DEV(accel_dev),
-			 "Generate of the thread to arbiter map failed");
+			 "Failed to generate thread to arbiter mapping");
 
 	return GET_HW_DATA(accel_dev)->thd_to_arb_map;
 }
@@ -454,6 +456,8 @@ void adf_init_hw_data_4xxx(struct adf_hw_device_data *hw_data, u32 dev_id)
 	hw_data->get_ring_to_svc_map = adf_gen4_get_ring_to_svc_map;
 	hw_data->disable_iov = adf_disable_sriov;
 	hw_data->ring_pair_reset = adf_gen4_ring_pair_reset;
+	hw_data->bank_state_save = adf_gen4_bank_state_save;
+	hw_data->bank_state_restore = adf_gen4_bank_state_restore;
 	hw_data->enable_pm = adf_gen4_enable_pm;
 	hw_data->handle_pm_interrupt = adf_gen4_handle_pm_interrupt;
 	hw_data->dev_config = adf_gen4_dev_config;
@@ -469,6 +473,7 @@ void adf_init_hw_data_4xxx(struct adf_hw_device_data *hw_data, u32 dev_id)
 	adf_gen4_init_dc_ops(&hw_data->dc_ops);
 	adf_gen4_init_ras_ops(&hw_data->ras_ops);
 	adf_gen4_init_tl_data(&hw_data->tl_data);
+	adf_gen4_init_vf_mig_ops(&hw_data->vfmig_ops);
 	adf_init_rl_data(&hw_data->rl_data);
 }
 
diff --git a/drivers/crypto/intel/qat/qat_4xxx/adf_drv.c b/drivers/crypto/intel/qat/qat_4xxx/adf_drv.c
index 9762f2bf7727..d26564cebdec 100644
--- a/drivers/crypto/intel/qat/qat_4xxx/adf_drv.c
+++ b/drivers/crypto/intel/qat/qat_4xxx/adf_drv.c
@@ -197,7 +197,9 @@ module_pci_driver(adf_driver);
 MODULE_LICENSE("Dual BSD/GPL");
 MODULE_AUTHOR("Intel");
 MODULE_FIRMWARE(ADF_4XXX_FW);
+MODULE_FIRMWARE(ADF_402XX_FW);
 MODULE_FIRMWARE(ADF_4XXX_MMP);
+MODULE_FIRMWARE(ADF_402XX_MMP);
 MODULE_DESCRIPTION("Intel(R) QuickAssist Technology");
 MODULE_VERSION(ADF_DRV_VERSION);
 MODULE_SOFTDEP("pre: crypto-intel_qat");
diff --git a/drivers/crypto/intel/qat/qat_c3xxx/adf_c3xxx_hw_data.c b/drivers/crypto/intel/qat/qat_c3xxx/adf_c3xxx_hw_data.c
index a882e0ea2279..201f9412c582 100644
--- a/drivers/crypto/intel/qat/qat_c3xxx/adf_c3xxx_hw_data.c
+++ b/drivers/crypto/intel/qat/qat_c3xxx/adf_c3xxx_hw_data.c
@@ -6,6 +6,7 @@
 #include <adf_common_drv.h>
 #include <adf_gen2_config.h>
 #include <adf_gen2_dc.h>
+#include <adf_gen2_hw_csr_data.h>
 #include <adf_gen2_hw_data.h>
 #include <adf_gen2_pfvf.h>
 #include "adf_c3xxx_hw_data.h"
diff --git a/drivers/crypto/intel/qat/qat_c3xxxvf/adf_c3xxxvf_hw_data.c b/drivers/crypto/intel/qat/qat_c3xxxvf/adf_c3xxxvf_hw_data.c
index 84d9486e04de..a512ca4efd3f 100644
--- a/drivers/crypto/intel/qat/qat_c3xxxvf/adf_c3xxxvf_hw_data.c
+++ b/drivers/crypto/intel/qat/qat_c3xxxvf/adf_c3xxxvf_hw_data.c
@@ -4,6 +4,7 @@
 #include <adf_common_drv.h>
 #include <adf_gen2_config.h>
 #include <adf_gen2_dc.h>
+#include <adf_gen2_hw_csr_data.h>
 #include <adf_gen2_hw_data.h>
 #include <adf_gen2_pfvf.h>
 #include <adf_pfvf_vf_msg.h>
diff --git a/drivers/crypto/intel/qat/qat_c62x/adf_c62x_hw_data.c b/drivers/crypto/intel/qat/qat_c62x/adf_c62x_hw_data.c
index 48cf3eb7c734..6b5b0cf9c7c7 100644
--- a/drivers/crypto/intel/qat/qat_c62x/adf_c62x_hw_data.c
+++ b/drivers/crypto/intel/qat/qat_c62x/adf_c62x_hw_data.c
@@ -6,6 +6,7 @@
 #include <adf_common_drv.h>
 #include <adf_gen2_config.h>
 #include <adf_gen2_dc.h>
+#include <adf_gen2_hw_csr_data.h>
 #include <adf_gen2_hw_data.h>
 #include <adf_gen2_pfvf.h>
 #include "adf_c62x_hw_data.h"
diff --git a/drivers/crypto/intel/qat/qat_c62xvf/adf_c62xvf_hw_data.c b/drivers/crypto/intel/qat/qat_c62xvf/adf_c62xvf_hw_data.c
index 751d7aa57fc7..4aaaaf921734 100644
--- a/drivers/crypto/intel/qat/qat_c62xvf/adf_c62xvf_hw_data.c
+++ b/drivers/crypto/intel/qat/qat_c62xvf/adf_c62xvf_hw_data.c
@@ -4,6 +4,7 @@
 #include <adf_common_drv.h>
 #include <adf_gen2_config.h>
 #include <adf_gen2_dc.h>
+#include <adf_gen2_hw_csr_data.h>
 #include <adf_gen2_hw_data.h>
 #include <adf_gen2_pfvf.h>
 #include <adf_pfvf_vf_msg.h>
diff --git a/drivers/crypto/intel/qat/qat_common/Makefile b/drivers/crypto/intel/qat/qat_common/Makefile
index 5915cde8a7aa..6f9266edc9f1 100644
--- a/drivers/crypto/intel/qat/qat_common/Makefile
+++ b/drivers/crypto/intel/qat/qat_common/Makefile
@@ -14,16 +14,20 @@ intel_qat-objs := adf_cfg.o \
 	adf_hw_arbiter.o \
 	adf_sysfs.o \
 	adf_sysfs_ras_counters.o \
+	adf_gen2_hw_csr_data.o \
 	adf_gen2_hw_data.o \
 	adf_gen2_config.o \
 	adf_gen4_config.o \
+	adf_gen4_hw_csr_data.o \
 	adf_gen4_hw_data.o \
+	adf_gen4_vf_mig.o \
 	adf_gen4_pm.o \
 	adf_gen2_dc.o \
 	adf_gen4_dc.o \
 	adf_gen4_ras.o \
 	adf_gen4_timer.o \
 	adf_clock.o \
+	adf_mstate_mgr.o \
 	qat_crypto.o \
 	qat_compression.o \
 	qat_comp_algs.o \
@@ -52,6 +56,6 @@ intel_qat-$(CONFIG_DEBUG_FS) += adf_transport_debug.o \
 intel_qat-$(CONFIG_PCI_IOV) += adf_sriov.o adf_vf_isr.o adf_pfvf_utils.o \
 			       adf_pfvf_pf_msg.o adf_pfvf_pf_proto.o \
 			       adf_pfvf_vf_msg.o adf_pfvf_vf_proto.o \
-			       adf_gen2_pfvf.o adf_gen4_pfvf.o
+			       adf_gen2_pfvf.o adf_gen4_pfvf.o qat_mig_dev.o
 
 intel_qat-$(CONFIG_CRYPTO_DEV_QAT_ERROR_INJECTION) += adf_heartbeat_inject.o
diff --git a/drivers/crypto/intel/qat/qat_common/adf_accel_devices.h b/drivers/crypto/intel/qat/qat_common/adf_accel_devices.h
index 08658c3a01e9..7830ecb1a1f1 100644
--- a/drivers/crypto/intel/qat/qat_common/adf_accel_devices.h
+++ b/drivers/crypto/intel/qat/qat_common/adf_accel_devices.h
@@ -9,6 +9,7 @@
 #include <linux/pci.h>
 #include <linux/ratelimit.h>
 #include <linux/types.h>
+#include <linux/qat/qat_mig_dev.h>
 #include "adf_cfg_common.h"
 #include "adf_rl.h"
 #include "adf_telemetry.h"
@@ -140,6 +141,40 @@ struct admin_info {
 	u32 mailbox_offset;
 };
 
+struct ring_config {
+	u64 base;
+	u32 config;
+	u32 head;
+	u32 tail;
+	u32 reserved0;
+};
+
+struct bank_state {
+	u32 ringstat0;
+	u32 ringstat1;
+	u32 ringuostat;
+	u32 ringestat;
+	u32 ringnestat;
+	u32 ringnfstat;
+	u32 ringfstat;
+	u32 ringcstat0;
+	u32 ringcstat1;
+	u32 ringcstat2;
+	u32 ringcstat3;
+	u32 iaintflagen;
+	u32 iaintflagreg;
+	u32 iaintflagsrcsel0;
+	u32 iaintflagsrcsel1;
+	u32 iaintcolen;
+	u32 iaintcolctl;
+	u32 iaintflagandcolen;
+	u32 ringexpstat;
+	u32 ringexpintenable;
+	u32 ringsrvarben;
+	u32 reserved0;
+	struct ring_config rings[ADF_ETR_MAX_RINGS_PER_BANK];
+};
+
 struct adf_hw_csr_ops {
 	u64 (*build_csr_ring_base_addr)(dma_addr_t addr, u32 size);
 	u32 (*read_csr_ring_head)(void __iomem *csr_base_addr, u32 bank,
@@ -150,22 +185,49 @@ struct adf_hw_csr_ops {
 				  u32 ring);
 	void (*write_csr_ring_tail)(void __iomem *csr_base_addr, u32 bank,
 				    u32 ring, u32 value);
+	u32 (*read_csr_stat)(void __iomem *csr_base_addr, u32 bank);
+	u32 (*read_csr_uo_stat)(void __iomem *csr_base_addr, u32 bank);
 	u32 (*read_csr_e_stat)(void __iomem *csr_base_addr, u32 bank);
+	u32 (*read_csr_ne_stat)(void __iomem *csr_base_addr, u32 bank);
+	u32 (*read_csr_nf_stat)(void __iomem *csr_base_addr, u32 bank);
+	u32 (*read_csr_f_stat)(void __iomem *csr_base_addr, u32 bank);
+	u32 (*read_csr_c_stat)(void __iomem *csr_base_addr, u32 bank);
+	u32 (*read_csr_exp_stat)(void __iomem *csr_base_addr, u32 bank);
+	u32 (*read_csr_exp_int_en)(void __iomem *csr_base_addr, u32 bank);
+	void (*write_csr_exp_int_en)(void __iomem *csr_base_addr, u32 bank,
+				     u32 value);
+	u32 (*read_csr_ring_config)(void __iomem *csr_base_addr, u32 bank,
+				    u32 ring);
 	void (*write_csr_ring_config)(void __iomem *csr_base_addr, u32 bank,
 				      u32 ring, u32 value);
+	dma_addr_t (*read_csr_ring_base)(void __iomem *csr_base_addr, u32 bank,
+					 u32 ring);
 	void (*write_csr_ring_base)(void __iomem *csr_base_addr, u32 bank,
 				    u32 ring, dma_addr_t addr);
+	u32 (*read_csr_int_en)(void __iomem *csr_base_addr, u32 bank);
+	void (*write_csr_int_en)(void __iomem *csr_base_addr, u32 bank,
+				 u32 value);
+	u32 (*read_csr_int_flag)(void __iomem *csr_base_addr, u32 bank);
 	void (*write_csr_int_flag)(void __iomem *csr_base_addr, u32 bank,
 				   u32 value);
+	u32 (*read_csr_int_srcsel)(void __iomem *csr_base_addr, u32 bank);
 	void (*write_csr_int_srcsel)(void __iomem *csr_base_addr, u32 bank);
+	void (*write_csr_int_srcsel_w_val)(void __iomem *csr_base_addr,
+					   u32 bank, u32 value);
+	u32 (*read_csr_int_col_en)(void __iomem *csr_base_addr, u32 bank);
 	void (*write_csr_int_col_en)(void __iomem *csr_base_addr, u32 bank,
 				     u32 value);
+	u32 (*read_csr_int_col_ctl)(void __iomem *csr_base_addr, u32 bank);
 	void (*write_csr_int_col_ctl)(void __iomem *csr_base_addr, u32 bank,
 				      u32 value);
+	u32 (*read_csr_int_flag_and_col)(void __iomem *csr_base_addr,
+					 u32 bank);
 	void (*write_csr_int_flag_and_col)(void __iomem *csr_base_addr,
 					   u32 bank, u32 value);
+	u32 (*read_csr_ring_srv_arb_en)(void __iomem *csr_base_addr, u32 bank);
 	void (*write_csr_ring_srv_arb_en)(void __iomem *csr_base_addr, u32 bank,
 					  u32 value);
+	u32 (*get_int_col_ctl_enable_mask)(void);
 };
 
 struct adf_cfg_device_data;
@@ -197,6 +259,20 @@ struct adf_dc_ops {
 	void (*build_deflate_ctx)(void *ctx);
 };
 
+struct qat_migdev_ops {
+	int (*init)(struct qat_mig_dev *mdev);
+	void (*cleanup)(struct qat_mig_dev *mdev);
+	void (*reset)(struct qat_mig_dev *mdev);
+	int (*open)(struct qat_mig_dev *mdev);
+	void (*close)(struct qat_mig_dev *mdev);
+	int (*suspend)(struct qat_mig_dev *mdev);
+	int (*resume)(struct qat_mig_dev *mdev);
+	int (*save_state)(struct qat_mig_dev *mdev);
+	int (*save_setup)(struct qat_mig_dev *mdev);
+	int (*load_state)(struct qat_mig_dev *mdev);
+	int (*load_setup)(struct qat_mig_dev *mdev, int size);
+};
+
 struct adf_dev_err_mask {
 	u32 cppagentcmdpar_mask;
 	u32 parerr_ath_cph_mask;
@@ -244,6 +320,10 @@ struct adf_hw_device_data {
 	void (*enable_ints)(struct adf_accel_dev *accel_dev);
 	void (*set_ssm_wdtimer)(struct adf_accel_dev *accel_dev);
 	int (*ring_pair_reset)(struct adf_accel_dev *accel_dev, u32 bank_nr);
+	int (*bank_state_save)(struct adf_accel_dev *accel_dev, u32 bank_number,
+			       struct bank_state *state);
+	int (*bank_state_restore)(struct adf_accel_dev *accel_dev,
+				  u32 bank_number, struct bank_state *state);
 	void (*reset_device)(struct adf_accel_dev *accel_dev);
 	void (*set_msix_rttable)(struct adf_accel_dev *accel_dev);
 	const char *(*uof_get_name)(struct adf_accel_dev *accel_dev, u32 obj_num);
@@ -260,6 +340,7 @@ struct adf_hw_device_data {
 	struct adf_dev_err_mask dev_err_mask;
 	struct adf_rl_hw_data rl_data;
 	struct adf_tl_hw_data tl_data;
+	struct qat_migdev_ops vfmig_ops;
 	const char *fw_name;
 	const char *fw_mmp_name;
 	u32 fuses;
@@ -316,6 +397,7 @@ struct adf_hw_device_data {
 #define GET_CSR_OPS(accel_dev) (&(accel_dev)->hw_device->csr_ops)
 #define GET_PFVF_OPS(accel_dev) (&(accel_dev)->hw_device->pfvf_ops)
 #define GET_DC_OPS(accel_dev) (&(accel_dev)->hw_device->dc_ops)
+#define GET_VFMIG_OPS(accel_dev) (&(accel_dev)->hw_device->vfmig_ops)
 #define GET_TL_DATA(accel_dev) GET_HW_DATA(accel_dev)->tl_data
 #define accel_to_pci_dev(accel_ptr) accel_ptr->accel_pci_dev.pci_dev
 
@@ -330,11 +412,17 @@ struct adf_fw_loader_data {
 struct adf_accel_vf_info {
 	struct adf_accel_dev *accel_dev;
 	struct mutex pf2vf_lock; /* protect CSR access for PF2VF messages */
+	struct mutex pfvf_mig_lock; /* protects PFVF state for migration */
 	struct ratelimit_state vf2pf_ratelimit;
 	u32 vf_nr;
 	bool init;
 	bool restarting;
 	u8 vf_compat_ver;
+	/*
+	 * Private area used for device migration.
+	 * Memory allocation and free is managed by migration driver.
+	 */
+	void *mig_priv;
 };
 
 struct adf_dc_data {
diff --git a/drivers/crypto/intel/qat/qat_common/adf_common_drv.h b/drivers/crypto/intel/qat/qat_common/adf_common_drv.h
index 57328249c89e..3bec9e20bad0 100644
--- a/drivers/crypto/intel/qat/qat_common/adf_common_drv.h
+++ b/drivers/crypto/intel/qat/qat_common/adf_common_drv.h
@@ -248,6 +248,16 @@ static inline void __iomem *adf_get_pmisc_base(struct adf_accel_dev *accel_dev)
 	return pmisc->virt_addr;
 }
 
+static inline void __iomem *adf_get_etr_base(struct adf_accel_dev *accel_dev)
+{
+	struct adf_hw_device_data *hw_data = accel_dev->hw_device;
+	struct adf_bar *etr;
+
+	etr = &GET_BARS(accel_dev)[hw_data->get_etr_bar_id(hw_data)];
+
+	return etr->virt_addr;
+}
+
 static inline void __iomem *adf_get_aram_base(struct adf_accel_dev *accel_dev)
 {
 	struct adf_hw_device_data *hw_data = accel_dev->hw_device;
diff --git a/drivers/crypto/intel/qat/qat_common/adf_gen2_hw_csr_data.c b/drivers/crypto/intel/qat/qat_common/adf_gen2_hw_csr_data.c
new file mode 100644
index 000000000000..650c9edd8a66
--- /dev/null
+++ b/drivers/crypto/intel/qat/qat_common/adf_gen2_hw_csr_data.c
@@ -0,0 +1,101 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/* Copyright(c) 2024 Intel Corporation */
+#include <linux/types.h>
+#include "adf_gen2_hw_csr_data.h"
+
+static u64 build_csr_ring_base_addr(dma_addr_t addr, u32 size)
+{
+	return BUILD_RING_BASE_ADDR(addr, size);
+}
+
+static u32 read_csr_ring_head(void __iomem *csr_base_addr, u32 bank, u32 ring)
+{
+	return READ_CSR_RING_HEAD(csr_base_addr, bank, ring);
+}
+
+static void write_csr_ring_head(void __iomem *csr_base_addr, u32 bank, u32 ring,
+				u32 value)
+{
+	WRITE_CSR_RING_HEAD(csr_base_addr, bank, ring, value);
+}
+
+static u32 read_csr_ring_tail(void __iomem *csr_base_addr, u32 bank, u32 ring)
+{
+	return READ_CSR_RING_TAIL(csr_base_addr, bank, ring);
+}
+
+static void write_csr_ring_tail(void __iomem *csr_base_addr, u32 bank, u32 ring,
+				u32 value)
+{
+	WRITE_CSR_RING_TAIL(csr_base_addr, bank, ring, value);
+}
+
+static u32 read_csr_e_stat(void __iomem *csr_base_addr, u32 bank)
+{
+	return READ_CSR_E_STAT(csr_base_addr, bank);
+}
+
+static void write_csr_ring_config(void __iomem *csr_base_addr, u32 bank,
+				  u32 ring, u32 value)
+{
+	WRITE_CSR_RING_CONFIG(csr_base_addr, bank, ring, value);
+}
+
+static void write_csr_ring_base(void __iomem *csr_base_addr, u32 bank, u32 ring,
+				dma_addr_t addr)
+{
+	WRITE_CSR_RING_BASE(csr_base_addr, bank, ring, addr);
+}
+
+static void write_csr_int_flag(void __iomem *csr_base_addr, u32 bank, u32 value)
+{
+	WRITE_CSR_INT_FLAG(csr_base_addr, bank, value);
+}
+
+static void write_csr_int_srcsel(void __iomem *csr_base_addr, u32 bank)
+{
+	WRITE_CSR_INT_SRCSEL(csr_base_addr, bank);
+}
+
+static void write_csr_int_col_en(void __iomem *csr_base_addr, u32 bank,
+				 u32 value)
+{
+	WRITE_CSR_INT_COL_EN(csr_base_addr, bank, value);
+}
+
+static void write_csr_int_col_ctl(void __iomem *csr_base_addr, u32 bank,
+				  u32 value)
+{
+	WRITE_CSR_INT_COL_CTL(csr_base_addr, bank, value);
+}
+
+static void write_csr_int_flag_and_col(void __iomem *csr_base_addr, u32 bank,
+				       u32 value)
+{
+	WRITE_CSR_INT_FLAG_AND_COL(csr_base_addr, bank, value);
+}
+
+static void write_csr_ring_srv_arb_en(void __iomem *csr_base_addr, u32 bank,
+				      u32 value)
+{
+	WRITE_CSR_RING_SRV_ARB_EN(csr_base_addr, bank, value);
+}
+
+void adf_gen2_init_hw_csr_ops(struct adf_hw_csr_ops *csr_ops)
+{
+	csr_ops->build_csr_ring_base_addr = build_csr_ring_base_addr;
+	csr_ops->read_csr_ring_head = read_csr_ring_head;
+	csr_ops->write_csr_ring_head = write_csr_ring_head;
+	csr_ops->read_csr_ring_tail = read_csr_ring_tail;
+	csr_ops->write_csr_ring_tail = write_csr_ring_tail;
+	csr_ops->read_csr_e_stat = read_csr_e_stat;
+	csr_ops->write_csr_ring_config = write_csr_ring_config;
+	csr_ops->write_csr_ring_base = write_csr_ring_base;
+	csr_ops->write_csr_int_flag = write_csr_int_flag;
+	csr_ops->write_csr_int_srcsel = write_csr_int_srcsel;
+	csr_ops->write_csr_int_col_en = write_csr_int_col_en;
+	csr_ops->write_csr_int_col_ctl = write_csr_int_col_ctl;
+	csr_ops->write_csr_int_flag_and_col = write_csr_int_flag_and_col;
+	csr_ops->write_csr_ring_srv_arb_en = write_csr_ring_srv_arb_en;
+}
+EXPORT_SYMBOL_GPL(adf_gen2_init_hw_csr_ops);
diff --git a/drivers/crypto/intel/qat/qat_common/adf_gen2_hw_csr_data.h b/drivers/crypto/intel/qat/qat_common/adf_gen2_hw_csr_data.h
new file mode 100644
index 000000000000..55058b0f9e52
--- /dev/null
+++ b/drivers/crypto/intel/qat/qat_common/adf_gen2_hw_csr_data.h
@@ -0,0 +1,86 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/* Copyright(c) 2024 Intel Corporation */
+#ifndef ADF_GEN2_HW_CSR_DATA_H_
+#define ADF_GEN2_HW_CSR_DATA_H_
+
+#include <linux/bitops.h>
+#include "adf_accel_devices.h"
+
+#define ADF_BANK_INT_SRC_SEL_MASK_0	0x4444444CUL
+#define ADF_BANK_INT_SRC_SEL_MASK_X	0x44444444UL
+#define ADF_RING_CSR_RING_CONFIG	0x000
+#define ADF_RING_CSR_RING_LBASE		0x040
+#define ADF_RING_CSR_RING_UBASE		0x080
+#define ADF_RING_CSR_RING_HEAD		0x0C0
+#define ADF_RING_CSR_RING_TAIL		0x100
+#define ADF_RING_CSR_E_STAT		0x14C
+#define ADF_RING_CSR_INT_FLAG		0x170
+#define ADF_RING_CSR_INT_SRCSEL		0x174
+#define ADF_RING_CSR_INT_SRCSEL_2	0x178
+#define ADF_RING_CSR_INT_COL_EN		0x17C
+#define ADF_RING_CSR_INT_COL_CTL	0x180
+#define ADF_RING_CSR_INT_FLAG_AND_COL	0x184
+#define ADF_RING_CSR_INT_COL_CTL_ENABLE	0x80000000
+#define ADF_RING_BUNDLE_SIZE		0x1000
+#define ADF_ARB_REG_SLOT		0x1000
+#define ADF_ARB_RINGSRVARBEN_OFFSET	0x19C
+
+#define BUILD_RING_BASE_ADDR(addr, size) \
+	(((addr) >> 6) & (GENMASK_ULL(63, 0) << (size)))
+#define READ_CSR_RING_HEAD(csr_base_addr, bank, ring) \
+	ADF_CSR_RD(csr_base_addr, (ADF_RING_BUNDLE_SIZE * (bank)) + \
+		   ADF_RING_CSR_RING_HEAD + ((ring) << 2))
+#define READ_CSR_RING_TAIL(csr_base_addr, bank, ring) \
+	ADF_CSR_RD(csr_base_addr, (ADF_RING_BUNDLE_SIZE * (bank)) + \
+		   ADF_RING_CSR_RING_TAIL + ((ring) << 2))
+#define READ_CSR_E_STAT(csr_base_addr, bank) \
+	ADF_CSR_RD(csr_base_addr, (ADF_RING_BUNDLE_SIZE * (bank)) + \
+		   ADF_RING_CSR_E_STAT)
+#define WRITE_CSR_RING_CONFIG(csr_base_addr, bank, ring, value) \
+	ADF_CSR_WR(csr_base_addr, (ADF_RING_BUNDLE_SIZE * (bank)) + \
+		   ADF_RING_CSR_RING_CONFIG + ((ring) << 2), value)
+#define WRITE_CSR_RING_BASE(csr_base_addr, bank, ring, value) \
+do { \
+	u32 l_base = 0, u_base = 0; \
+	l_base = (u32)((value) & 0xFFFFFFFF); \
+	u_base = (u32)(((value) & 0xFFFFFFFF00000000ULL) >> 32); \
+	ADF_CSR_WR(csr_base_addr, (ADF_RING_BUNDLE_SIZE * (bank)) + \
+		   ADF_RING_CSR_RING_LBASE + ((ring) << 2), l_base); \
+	ADF_CSR_WR(csr_base_addr, (ADF_RING_BUNDLE_SIZE * (bank)) + \
+		   ADF_RING_CSR_RING_UBASE + ((ring) << 2), u_base); \
+} while (0)
+
+#define WRITE_CSR_RING_HEAD(csr_base_addr, bank, ring, value) \
+	ADF_CSR_WR(csr_base_addr, (ADF_RING_BUNDLE_SIZE * (bank)) + \
+		   ADF_RING_CSR_RING_HEAD + ((ring) << 2), value)
+#define WRITE_CSR_RING_TAIL(csr_base_addr, bank, ring, value) \
+	ADF_CSR_WR(csr_base_addr, (ADF_RING_BUNDLE_SIZE * (bank)) + \
+		   ADF_RING_CSR_RING_TAIL + ((ring) << 2), value)
+#define WRITE_CSR_INT_FLAG(csr_base_addr, bank, value) \
+	ADF_CSR_WR(csr_base_addr, (ADF_RING_BUNDLE_SIZE * (bank)) + \
+		   ADF_RING_CSR_INT_FLAG, value)
+#define WRITE_CSR_INT_SRCSEL(csr_base_addr, bank) \
+do { \
+	ADF_CSR_WR(csr_base_addr, (ADF_RING_BUNDLE_SIZE * (bank)) + \
+	ADF_RING_CSR_INT_SRCSEL, ADF_BANK_INT_SRC_SEL_MASK_0); \
+	ADF_CSR_WR(csr_base_addr, (ADF_RING_BUNDLE_SIZE * (bank)) + \
+	ADF_RING_CSR_INT_SRCSEL_2, ADF_BANK_INT_SRC_SEL_MASK_X); \
+} while (0)
+#define WRITE_CSR_INT_COL_EN(csr_base_addr, bank, value) \
+	ADF_CSR_WR(csr_base_addr, (ADF_RING_BUNDLE_SIZE * (bank)) + \
+		   ADF_RING_CSR_INT_COL_EN, value)
+#define WRITE_CSR_INT_COL_CTL(csr_base_addr, bank, value) \
+	ADF_CSR_WR(csr_base_addr, (ADF_RING_BUNDLE_SIZE * (bank)) + \
+		   ADF_RING_CSR_INT_COL_CTL, \
+		   ADF_RING_CSR_INT_COL_CTL_ENABLE | (value))
+#define WRITE_CSR_INT_FLAG_AND_COL(csr_base_addr, bank, value) \
+	ADF_CSR_WR(csr_base_addr, (ADF_RING_BUNDLE_SIZE * (bank)) + \
+		   ADF_RING_CSR_INT_FLAG_AND_COL, value)
+
+#define WRITE_CSR_RING_SRV_ARB_EN(csr_addr, index, value) \
+	ADF_CSR_WR(csr_addr, ADF_ARB_RINGSRVARBEN_OFFSET + \
+	(ADF_ARB_REG_SLOT * (index)), value)
+
+void adf_gen2_init_hw_csr_ops(struct adf_hw_csr_ops *csr_ops);
+
+#endif
diff --git a/drivers/crypto/intel/qat/qat_common/adf_gen2_hw_data.c b/drivers/crypto/intel/qat/qat_common/adf_gen2_hw_data.c
index d1884547b5a1..1f64bf49b221 100644
--- a/drivers/crypto/intel/qat/qat_common/adf_gen2_hw_data.c
+++ b/drivers/crypto/intel/qat/qat_common/adf_gen2_hw_data.c
@@ -111,103 +111,6 @@ void adf_gen2_enable_ints(struct adf_accel_dev *accel_dev)
 }
 EXPORT_SYMBOL_GPL(adf_gen2_enable_ints);
 
-static u64 build_csr_ring_base_addr(dma_addr_t addr, u32 size)
-{
-	return BUILD_RING_BASE_ADDR(addr, size);
-}
-
-static u32 read_csr_ring_head(void __iomem *csr_base_addr, u32 bank, u32 ring)
-{
-	return READ_CSR_RING_HEAD(csr_base_addr, bank, ring);
-}
-
-static void write_csr_ring_head(void __iomem *csr_base_addr, u32 bank, u32 ring,
-				u32 value)
-{
-	WRITE_CSR_RING_HEAD(csr_base_addr, bank, ring, value);
-}
-
-static u32 read_csr_ring_tail(void __iomem *csr_base_addr, u32 bank, u32 ring)
-{
-	return READ_CSR_RING_TAIL(csr_base_addr, bank, ring);
-}
-
-static void write_csr_ring_tail(void __iomem *csr_base_addr, u32 bank, u32 ring,
-				u32 value)
-{
-	WRITE_CSR_RING_TAIL(csr_base_addr, bank, ring, value);
-}
-
-static u32 read_csr_e_stat(void __iomem *csr_base_addr, u32 bank)
-{
-	return READ_CSR_E_STAT(csr_base_addr, bank);
-}
-
-static void write_csr_ring_config(void __iomem *csr_base_addr, u32 bank,
-				  u32 ring, u32 value)
-{
-	WRITE_CSR_RING_CONFIG(csr_base_addr, bank, ring, value);
-}
-
-static void write_csr_ring_base(void __iomem *csr_base_addr, u32 bank, u32 ring,
-				dma_addr_t addr)
-{
-	WRITE_CSR_RING_BASE(csr_base_addr, bank, ring, addr);
-}
-
-static void write_csr_int_flag(void __iomem *csr_base_addr, u32 bank, u32 value)
-{
-	WRITE_CSR_INT_FLAG(csr_base_addr, bank, value);
-}
-
-static void write_csr_int_srcsel(void __iomem *csr_base_addr, u32 bank)
-{
-	WRITE_CSR_INT_SRCSEL(csr_base_addr, bank);
-}
-
-static void write_csr_int_col_en(void __iomem *csr_base_addr, u32 bank,
-				 u32 value)
-{
-	WRITE_CSR_INT_COL_EN(csr_base_addr, bank, value);
-}
-
-static void write_csr_int_col_ctl(void __iomem *csr_base_addr, u32 bank,
-				  u32 value)
-{
-	WRITE_CSR_INT_COL_CTL(csr_base_addr, bank, value);
-}
-
-static void write_csr_int_flag_and_col(void __iomem *csr_base_addr, u32 bank,
-				       u32 value)
-{
-	WRITE_CSR_INT_FLAG_AND_COL(csr_base_addr, bank, value);
-}
-
-static void write_csr_ring_srv_arb_en(void __iomem *csr_base_addr, u32 bank,
-				      u32 value)
-{
-	WRITE_CSR_RING_SRV_ARB_EN(csr_base_addr, bank, value);
-}
-
-void adf_gen2_init_hw_csr_ops(struct adf_hw_csr_ops *csr_ops)
-{
-	csr_ops->build_csr_ring_base_addr = build_csr_ring_base_addr;
-	csr_ops->read_csr_ring_head = read_csr_ring_head;
-	csr_ops->write_csr_ring_head = write_csr_ring_head;
-	csr_ops->read_csr_ring_tail = read_csr_ring_tail;
-	csr_ops->write_csr_ring_tail = write_csr_ring_tail;
-	csr_ops->read_csr_e_stat = read_csr_e_stat;
-	csr_ops->write_csr_ring_config = write_csr_ring_config;
-	csr_ops->write_csr_ring_base = write_csr_ring_base;
-	csr_ops->write_csr_int_flag = write_csr_int_flag;
-	csr_ops->write_csr_int_srcsel = write_csr_int_srcsel;
-	csr_ops->write_csr_int_col_en = write_csr_int_col_en;
-	csr_ops->write_csr_int_col_ctl = write_csr_int_col_ctl;
-	csr_ops->write_csr_int_flag_and_col = write_csr_int_flag_and_col;
-	csr_ops->write_csr_ring_srv_arb_en = write_csr_ring_srv_arb_en;
-}
-EXPORT_SYMBOL_GPL(adf_gen2_init_hw_csr_ops);
-
 u32 adf_gen2_get_accel_cap(struct adf_accel_dev *accel_dev)
 {
 	struct adf_hw_device_data *hw_data = accel_dev->hw_device;
diff --git a/drivers/crypto/intel/qat/qat_common/adf_gen2_hw_data.h b/drivers/crypto/intel/qat/qat_common/adf_gen2_hw_data.h
index 6bd341061de4..708e9186127b 100644
--- a/drivers/crypto/intel/qat/qat_common/adf_gen2_hw_data.h
+++ b/drivers/crypto/intel/qat/qat_common/adf_gen2_hw_data.h
@@ -6,78 +6,9 @@
 #include "adf_accel_devices.h"
 #include "adf_cfg_common.h"
 
-/* Transport access */
-#define ADF_BANK_INT_SRC_SEL_MASK_0	0x4444444CUL
-#define ADF_BANK_INT_SRC_SEL_MASK_X	0x44444444UL
-#define ADF_RING_CSR_RING_CONFIG	0x000
-#define ADF_RING_CSR_RING_LBASE		0x040
-#define ADF_RING_CSR_RING_UBASE		0x080
-#define ADF_RING_CSR_RING_HEAD		0x0C0
-#define ADF_RING_CSR_RING_TAIL		0x100
-#define ADF_RING_CSR_E_STAT		0x14C
-#define ADF_RING_CSR_INT_FLAG		0x170
-#define ADF_RING_CSR_INT_SRCSEL		0x174
-#define ADF_RING_CSR_INT_SRCSEL_2	0x178
-#define ADF_RING_CSR_INT_COL_EN		0x17C
-#define ADF_RING_CSR_INT_COL_CTL	0x180
-#define ADF_RING_CSR_INT_FLAG_AND_COL	0x184
-#define ADF_RING_CSR_INT_COL_CTL_ENABLE	0x80000000
-#define ADF_RING_BUNDLE_SIZE		0x1000
 #define ADF_GEN2_RX_RINGS_OFFSET	8
 #define ADF_GEN2_TX_RINGS_MASK		0xFF
 
-#define BUILD_RING_BASE_ADDR(addr, size) \
-	(((addr) >> 6) & (GENMASK_ULL(63, 0) << (size)))
-#define READ_CSR_RING_HEAD(csr_base_addr, bank, ring) \
-	ADF_CSR_RD(csr_base_addr, (ADF_RING_BUNDLE_SIZE * (bank)) + \
-		   ADF_RING_CSR_RING_HEAD + ((ring) << 2))
-#define READ_CSR_RING_TAIL(csr_base_addr, bank, ring) \
-	ADF_CSR_RD(csr_base_addr, (ADF_RING_BUNDLE_SIZE * (bank)) + \
-		   ADF_RING_CSR_RING_TAIL + ((ring) << 2))
-#define READ_CSR_E_STAT(csr_base_addr, bank) \
-	ADF_CSR_RD(csr_base_addr, (ADF_RING_BUNDLE_SIZE * (bank)) + \
-		   ADF_RING_CSR_E_STAT)
-#define WRITE_CSR_RING_CONFIG(csr_base_addr, bank, ring, value) \
-	ADF_CSR_WR(csr_base_addr, (ADF_RING_BUNDLE_SIZE * (bank)) + \
-		   ADF_RING_CSR_RING_CONFIG + ((ring) << 2), value)
-#define WRITE_CSR_RING_BASE(csr_base_addr, bank, ring, value) \
-do { \
-	u32 l_base = 0, u_base = 0; \
-	l_base = (u32)((value) & 0xFFFFFFFF); \
-	u_base = (u32)(((value) & 0xFFFFFFFF00000000ULL) >> 32); \
-	ADF_CSR_WR(csr_base_addr, (ADF_RING_BUNDLE_SIZE * (bank)) + \
-		   ADF_RING_CSR_RING_LBASE + ((ring) << 2), l_base); \
-	ADF_CSR_WR(csr_base_addr, (ADF_RING_BUNDLE_SIZE * (bank)) + \
-		   ADF_RING_CSR_RING_UBASE + ((ring) << 2), u_base); \
-} while (0)
-
-#define WRITE_CSR_RING_HEAD(csr_base_addr, bank, ring, value) \
-	ADF_CSR_WR(csr_base_addr, (ADF_RING_BUNDLE_SIZE * (bank)) + \
-		   ADF_RING_CSR_RING_HEAD + ((ring) << 2), value)
-#define WRITE_CSR_RING_TAIL(csr_base_addr, bank, ring, value) \
-	ADF_CSR_WR(csr_base_addr, (ADF_RING_BUNDLE_SIZE * (bank)) + \
-		   ADF_RING_CSR_RING_TAIL + ((ring) << 2), value)
-#define WRITE_CSR_INT_FLAG(csr_base_addr, bank, value) \
-	ADF_CSR_WR(csr_base_addr, (ADF_RING_BUNDLE_SIZE * (bank)) + \
-		   ADF_RING_CSR_INT_FLAG, value)
-#define WRITE_CSR_INT_SRCSEL(csr_base_addr, bank) \
-do { \
-	ADF_CSR_WR(csr_base_addr, (ADF_RING_BUNDLE_SIZE * (bank)) + \
-	ADF_RING_CSR_INT_SRCSEL, ADF_BANK_INT_SRC_SEL_MASK_0); \
-	ADF_CSR_WR(csr_base_addr, (ADF_RING_BUNDLE_SIZE * (bank)) + \
-	ADF_RING_CSR_INT_SRCSEL_2, ADF_BANK_INT_SRC_SEL_MASK_X); \
-} while (0)
-#define WRITE_CSR_INT_COL_EN(csr_base_addr, bank, value) \
-	ADF_CSR_WR(csr_base_addr, (ADF_RING_BUNDLE_SIZE * (bank)) + \
-		   ADF_RING_CSR_INT_COL_EN, value)
-#define WRITE_CSR_INT_COL_CTL(csr_base_addr, bank, value) \
-	ADF_CSR_WR(csr_base_addr, (ADF_RING_BUNDLE_SIZE * (bank)) + \
-		   ADF_RING_CSR_INT_COL_CTL, \
-		   ADF_RING_CSR_INT_COL_CTL_ENABLE | (value))
-#define WRITE_CSR_INT_FLAG_AND_COL(csr_base_addr, bank, value) \
-	ADF_CSR_WR(csr_base_addr, (ADF_RING_BUNDLE_SIZE * (bank)) + \
-		   ADF_RING_CSR_INT_FLAG_AND_COL, value)
-
 /* AE to function map */
 #define AE2FUNCTION_MAP_A_OFFSET	(0x3A400 + 0x190)
 #define AE2FUNCTION_MAP_B_OFFSET	(0x3A400 + 0x310)
@@ -106,12 +37,6 @@ do { \
 #define ADF_ARB_OFFSET			0x30000
 #define ADF_ARB_WRK_2_SER_MAP_OFFSET	0x180
 #define ADF_ARB_CONFIG			(BIT(31) | BIT(6) | BIT(0))
-#define ADF_ARB_REG_SLOT		0x1000
-#define ADF_ARB_RINGSRVARBEN_OFFSET	0x19C
-
-#define WRITE_CSR_RING_SRV_ARB_EN(csr_addr, index, value) \
-	ADF_CSR_WR(csr_addr, ADF_ARB_RINGSRVARBEN_OFFSET + \
-	(ADF_ARB_REG_SLOT * (index)), value)
 
 /* Power gating */
 #define ADF_POWERGATE_DC		BIT(23)
@@ -158,7 +83,6 @@ u32 adf_gen2_get_num_aes(struct adf_hw_device_data *self);
 void adf_gen2_enable_error_correction(struct adf_accel_dev *accel_dev);
 void adf_gen2_cfg_iov_thds(struct adf_accel_dev *accel_dev, bool enable,
 			   int num_a_regs, int num_b_regs);
-void adf_gen2_init_hw_csr_ops(struct adf_hw_csr_ops *csr_ops);
 void adf_gen2_get_admin_info(struct admin_info *admin_csrs_info);
 void adf_gen2_get_arb_info(struct arb_info *arb_info);
 void adf_gen2_enable_ints(struct adf_accel_dev *accel_dev);
diff --git a/drivers/crypto/intel/qat/qat_common/adf_gen4_hw_csr_data.c b/drivers/crypto/intel/qat/qat_common/adf_gen4_hw_csr_data.c
new file mode 100644
index 000000000000..6609c248aaba
--- /dev/null
+++ b/drivers/crypto/intel/qat/qat_common/adf_gen4_hw_csr_data.c
@@ -0,0 +1,231 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/* Copyright(c) 2024 Intel Corporation */
+#include <linux/types.h>
+#include "adf_gen4_hw_csr_data.h"
+
+static u64 build_csr_ring_base_addr(dma_addr_t addr, u32 size)
+{
+	return BUILD_RING_BASE_ADDR(addr, size);
+}
+
+static u32 read_csr_ring_head(void __iomem *csr_base_addr, u32 bank, u32 ring)
+{
+	return READ_CSR_RING_HEAD(csr_base_addr, bank, ring);
+}
+
+static void write_csr_ring_head(void __iomem *csr_base_addr, u32 bank, u32 ring,
+				u32 value)
+{
+	WRITE_CSR_RING_HEAD(csr_base_addr, bank, ring, value);
+}
+
+static u32 read_csr_ring_tail(void __iomem *csr_base_addr, u32 bank, u32 ring)
+{
+	return READ_CSR_RING_TAIL(csr_base_addr, bank, ring);
+}
+
+static void write_csr_ring_tail(void __iomem *csr_base_addr, u32 bank, u32 ring,
+				u32 value)
+{
+	WRITE_CSR_RING_TAIL(csr_base_addr, bank, ring, value);
+}
+
+static u32 read_csr_stat(void __iomem *csr_base_addr, u32 bank)
+{
+	return READ_CSR_STAT(csr_base_addr, bank);
+}
+
+static u32 read_csr_uo_stat(void __iomem *csr_base_addr, u32 bank)
+{
+	return READ_CSR_UO_STAT(csr_base_addr, bank);
+}
+
+static u32 read_csr_e_stat(void __iomem *csr_base_addr, u32 bank)
+{
+	return READ_CSR_E_STAT(csr_base_addr, bank);
+}
+
+static u32 read_csr_ne_stat(void __iomem *csr_base_addr, u32 bank)
+{
+	return READ_CSR_NE_STAT(csr_base_addr, bank);
+}
+
+static u32 read_csr_nf_stat(void __iomem *csr_base_addr, u32 bank)
+{
+	return READ_CSR_NF_STAT(csr_base_addr, bank);
+}
+
+static u32 read_csr_f_stat(void __iomem *csr_base_addr, u32 bank)
+{
+	return READ_CSR_F_STAT(csr_base_addr, bank);
+}
+
+static u32 read_csr_c_stat(void __iomem *csr_base_addr, u32 bank)
+{
+	return READ_CSR_C_STAT(csr_base_addr, bank);
+}
+
+static u32 read_csr_exp_stat(void __iomem *csr_base_addr, u32 bank)
+{
+	return READ_CSR_EXP_STAT(csr_base_addr, bank);
+}
+
+static u32 read_csr_exp_int_en(void __iomem *csr_base_addr, u32 bank)
+{
+	return READ_CSR_EXP_INT_EN(csr_base_addr, bank);
+}
+
+static void write_csr_exp_int_en(void __iomem *csr_base_addr, u32 bank,
+				 u32 value)
+{
+	WRITE_CSR_EXP_INT_EN(csr_base_addr, bank, value);
+}
+
+static u32 read_csr_ring_config(void __iomem *csr_base_addr, u32 bank,
+				u32 ring)
+{
+	return READ_CSR_RING_CONFIG(csr_base_addr, bank, ring);
+}
+
+static void write_csr_ring_config(void __iomem *csr_base_addr, u32 bank, u32 ring,
+				  u32 value)
+{
+	WRITE_CSR_RING_CONFIG(csr_base_addr, bank, ring, value);
+}
+
+static dma_addr_t read_csr_ring_base(void __iomem *csr_base_addr, u32 bank,
+				     u32 ring)
+{
+	return READ_CSR_RING_BASE(csr_base_addr, bank, ring);
+}
+
+static void write_csr_ring_base(void __iomem *csr_base_addr, u32 bank, u32 ring,
+				dma_addr_t addr)
+{
+	WRITE_CSR_RING_BASE(csr_base_addr, bank, ring, addr);
+}
+
+static u32 read_csr_int_en(void __iomem *csr_base_addr, u32 bank)
+{
+	return READ_CSR_INT_EN(csr_base_addr, bank);
+}
+
+static void write_csr_int_en(void __iomem *csr_base_addr, u32 bank, u32 value)
+{
+	WRITE_CSR_INT_EN(csr_base_addr, bank, value);
+}
+
+static u32 read_csr_int_flag(void __iomem *csr_base_addr, u32 bank)
+{
+	return READ_CSR_INT_FLAG(csr_base_addr, bank);
+}
+
+static void write_csr_int_flag(void __iomem *csr_base_addr, u32 bank,
+			       u32 value)
+{
+	WRITE_CSR_INT_FLAG(csr_base_addr, bank, value);
+}
+
+static u32 read_csr_int_srcsel(void __iomem *csr_base_addr, u32 bank)
+{
+	return READ_CSR_INT_SRCSEL(csr_base_addr, bank);
+}
+
+static void write_csr_int_srcsel(void __iomem *csr_base_addr, u32 bank)
+{
+	WRITE_CSR_INT_SRCSEL(csr_base_addr, bank);
+}
+
+static void write_csr_int_srcsel_w_val(void __iomem *csr_base_addr, u32 bank,
+				       u32 value)
+{
+	WRITE_CSR_INT_SRCSEL_W_VAL(csr_base_addr, bank, value);
+}
+
+static u32 read_csr_int_col_en(void __iomem *csr_base_addr, u32 bank)
+{
+	return READ_CSR_INT_COL_EN(csr_base_addr, bank);
+}
+
+static void write_csr_int_col_en(void __iomem *csr_base_addr, u32 bank, u32 value)
+{
+	WRITE_CSR_INT_COL_EN(csr_base_addr, bank, value);
+}
+
+static u32 read_csr_int_col_ctl(void __iomem *csr_base_addr, u32 bank)
+{
+	return READ_CSR_INT_COL_CTL(csr_base_addr, bank);
+}
+
+static void write_csr_int_col_ctl(void __iomem *csr_base_addr, u32 bank,
+				  u32 value)
+{
+	WRITE_CSR_INT_COL_CTL(csr_base_addr, bank, value);
+}
+
+static u32 read_csr_int_flag_and_col(void __iomem *csr_base_addr, u32 bank)
+{
+	return READ_CSR_INT_FLAG_AND_COL(csr_base_addr, bank);
+}
+
+static void write_csr_int_flag_and_col(void __iomem *csr_base_addr, u32 bank,
+				       u32 value)
+{
+	WRITE_CSR_INT_FLAG_AND_COL(csr_base_addr, bank, value);
+}
+
+static u32 read_csr_ring_srv_arb_en(void __iomem *csr_base_addr, u32 bank)
+{
+	return READ_CSR_RING_SRV_ARB_EN(csr_base_addr, bank);
+}
+
+static void write_csr_ring_srv_arb_en(void __iomem *csr_base_addr, u32 bank,
+				      u32 value)
+{
+	WRITE_CSR_RING_SRV_ARB_EN(csr_base_addr, bank, value);
+}
+
+static u32 get_int_col_ctl_enable_mask(void)
+{
+	return ADF_RING_CSR_INT_COL_CTL_ENABLE;
+}
+
+void adf_gen4_init_hw_csr_ops(struct adf_hw_csr_ops *csr_ops)
+{
+	csr_ops->build_csr_ring_base_addr = build_csr_ring_base_addr;
+	csr_ops->read_csr_ring_head = read_csr_ring_head;
+	csr_ops->write_csr_ring_head = write_csr_ring_head;
+	csr_ops->read_csr_ring_tail = read_csr_ring_tail;
+	csr_ops->write_csr_ring_tail = write_csr_ring_tail;
+	csr_ops->read_csr_stat = read_csr_stat;
+	csr_ops->read_csr_uo_stat = read_csr_uo_stat;
+	csr_ops->read_csr_e_stat = read_csr_e_stat;
+	csr_ops->read_csr_ne_stat = read_csr_ne_stat;
+	csr_ops->read_csr_nf_stat = read_csr_nf_stat;
+	csr_ops->read_csr_f_stat = read_csr_f_stat;
+	csr_ops->read_csr_c_stat = read_csr_c_stat;
+	csr_ops->read_csr_exp_stat = read_csr_exp_stat;
+	csr_ops->read_csr_exp_int_en = read_csr_exp_int_en;
+	csr_ops->write_csr_exp_int_en = write_csr_exp_int_en;
+	csr_ops->read_csr_ring_config = read_csr_ring_config;
+	csr_ops->write_csr_ring_config = write_csr_ring_config;
+	csr_ops->read_csr_ring_base = read_csr_ring_base;
+	csr_ops->write_csr_ring_base = write_csr_ring_base;
+	csr_ops->read_csr_int_en = read_csr_int_en;
+	csr_ops->write_csr_int_en = write_csr_int_en;
+	csr_ops->read_csr_int_flag = read_csr_int_flag;
+	csr_ops->write_csr_int_flag = write_csr_int_flag;
+	csr_ops->read_csr_int_srcsel = read_csr_int_srcsel;
+	csr_ops->write_csr_int_srcsel = write_csr_int_srcsel;
+	csr_ops->write_csr_int_srcsel_w_val = write_csr_int_srcsel_w_val;
+	csr_ops->read_csr_int_col_en = read_csr_int_col_en;
+	csr_ops->write_csr_int_col_en = write_csr_int_col_en;
+	csr_ops->read_csr_int_col_ctl = read_csr_int_col_ctl;
+	csr_ops->write_csr_int_col_ctl = write_csr_int_col_ctl;
+	csr_ops->read_csr_int_flag_and_col = read_csr_int_flag_and_col;
+	csr_ops->write_csr_int_flag_and_col = write_csr_int_flag_and_col;
+	csr_ops->read_csr_ring_srv_arb_en = read_csr_ring_srv_arb_en;
+	csr_ops->write_csr_ring_srv_arb_en = write_csr_ring_srv_arb_en;
+	csr_ops->get_int_col_ctl_enable_mask = get_int_col_ctl_enable_mask;
+}
+EXPORT_SYMBOL_GPL(adf_gen4_init_hw_csr_ops);
diff --git a/drivers/crypto/intel/qat/qat_common/adf_gen4_hw_csr_data.h b/drivers/crypto/intel/qat/qat_common/adf_gen4_hw_csr_data.h
new file mode 100644
index 000000000000..6f33e7c87c2c
--- /dev/null
+++ b/drivers/crypto/intel/qat/qat_common/adf_gen4_hw_csr_data.h
@@ -0,0 +1,188 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/* Copyright(c) 2024 Intel Corporation */
+#ifndef ADF_GEN4_HW_CSR_DATA_H_
+#define ADF_GEN4_HW_CSR_DATA_H_
+
+#include <linux/bitops.h>
+#include "adf_accel_devices.h"
+
+#define ADF_BANK_INT_SRC_SEL_MASK	0x44UL
+#define ADF_RING_CSR_RING_CONFIG	0x1000
+#define ADF_RING_CSR_RING_LBASE		0x1040
+#define ADF_RING_CSR_RING_UBASE		0x1080
+#define ADF_RING_CSR_RING_HEAD		0x0C0
+#define ADF_RING_CSR_RING_TAIL		0x100
+#define ADF_RING_CSR_STAT		0x140
+#define ADF_RING_CSR_UO_STAT		0x148
+#define ADF_RING_CSR_E_STAT		0x14C
+#define ADF_RING_CSR_NE_STAT		0x150
+#define ADF_RING_CSR_NF_STAT		0x154
+#define ADF_RING_CSR_F_STAT		0x158
+#define ADF_RING_CSR_C_STAT		0x15C
+#define ADF_RING_CSR_INT_FLAG_EN	0x16C
+#define ADF_RING_CSR_INT_FLAG		0x170
+#define ADF_RING_CSR_INT_SRCSEL		0x174
+#define ADF_RING_CSR_INT_COL_EN		0x17C
+#define ADF_RING_CSR_INT_COL_CTL	0x180
+#define ADF_RING_CSR_INT_FLAG_AND_COL	0x184
+#define ADF_RING_CSR_EXP_STAT		0x188
+#define ADF_RING_CSR_EXP_INT_EN		0x18C
+#define ADF_RING_CSR_INT_COL_CTL_ENABLE	0x80000000
+#define ADF_RING_CSR_ADDR_OFFSET	0x100000
+#define ADF_RING_BUNDLE_SIZE		0x2000
+#define ADF_RING_CSR_RING_SRV_ARB_EN	0x19C
+
+#define BUILD_RING_BASE_ADDR(addr, size) \
+	((((addr) >> 6) & (GENMASK_ULL(63, 0) << (size))) << 6)
+#define READ_CSR_RING_HEAD(csr_base_addr, bank, ring) \
+	ADF_CSR_RD((csr_base_addr) + ADF_RING_CSR_ADDR_OFFSET, \
+		   ADF_RING_BUNDLE_SIZE * (bank) + \
+		   ADF_RING_CSR_RING_HEAD + ((ring) << 2))
+#define READ_CSR_RING_TAIL(csr_base_addr, bank, ring) \
+	ADF_CSR_RD((csr_base_addr) + ADF_RING_CSR_ADDR_OFFSET, \
+		   ADF_RING_BUNDLE_SIZE * (bank) + \
+		   ADF_RING_CSR_RING_TAIL + ((ring) << 2))
+#define READ_CSR_STAT(csr_base_addr, bank) \
+	ADF_CSR_RD((csr_base_addr) + ADF_RING_CSR_ADDR_OFFSET, \
+		   ADF_RING_BUNDLE_SIZE * (bank) + ADF_RING_CSR_STAT)
+#define READ_CSR_UO_STAT(csr_base_addr, bank) \
+	ADF_CSR_RD((csr_base_addr) + ADF_RING_CSR_ADDR_OFFSET, \
+		   ADF_RING_BUNDLE_SIZE * (bank) + ADF_RING_CSR_UO_STAT)
+#define READ_CSR_E_STAT(csr_base_addr, bank) \
+	ADF_CSR_RD((csr_base_addr) + ADF_RING_CSR_ADDR_OFFSET, \
+		   ADF_RING_BUNDLE_SIZE * (bank) + ADF_RING_CSR_E_STAT)
+#define READ_CSR_NE_STAT(csr_base_addr, bank) \
+	ADF_CSR_RD((csr_base_addr) + ADF_RING_CSR_ADDR_OFFSET, \
+		   ADF_RING_BUNDLE_SIZE * (bank) + ADF_RING_CSR_NE_STAT)
+#define READ_CSR_NF_STAT(csr_base_addr, bank) \
+	ADF_CSR_RD((csr_base_addr) + ADF_RING_CSR_ADDR_OFFSET, \
+		   ADF_RING_BUNDLE_SIZE * (bank) + ADF_RING_CSR_NF_STAT)
+#define READ_CSR_F_STAT(csr_base_addr, bank) \
+	ADF_CSR_RD((csr_base_addr) + ADF_RING_CSR_ADDR_OFFSET, \
+		   ADF_RING_BUNDLE_SIZE * (bank) + ADF_RING_CSR_F_STAT)
+#define READ_CSR_C_STAT(csr_base_addr, bank) \
+	ADF_CSR_RD((csr_base_addr) + ADF_RING_CSR_ADDR_OFFSET, \
+		   ADF_RING_BUNDLE_SIZE * (bank) + ADF_RING_CSR_C_STAT)
+#define READ_CSR_EXP_STAT(csr_base_addr, bank) \
+	ADF_CSR_RD((csr_base_addr) + ADF_RING_CSR_ADDR_OFFSET, \
+		   ADF_RING_BUNDLE_SIZE * (bank) + ADF_RING_CSR_EXP_STAT)
+#define READ_CSR_EXP_INT_EN(csr_base_addr, bank) \
+	ADF_CSR_RD((csr_base_addr) + ADF_RING_CSR_ADDR_OFFSET, \
+		   ADF_RING_BUNDLE_SIZE * (bank) + ADF_RING_CSR_EXP_INT_EN)
+#define WRITE_CSR_EXP_INT_EN(csr_base_addr, bank, value) \
+	ADF_CSR_WR((csr_base_addr) + ADF_RING_CSR_ADDR_OFFSET, \
+		   ADF_RING_BUNDLE_SIZE * (bank) + \
+		   ADF_RING_CSR_EXP_INT_EN, value)
+#define READ_CSR_RING_CONFIG(csr_base_addr, bank, ring) \
+	ADF_CSR_RD((csr_base_addr) + ADF_RING_CSR_ADDR_OFFSET, \
+		   ADF_RING_BUNDLE_SIZE * (bank) + \
+		   ADF_RING_CSR_RING_CONFIG + ((ring) << 2))
+#define WRITE_CSR_RING_CONFIG(csr_base_addr, bank, ring, value) \
+	ADF_CSR_WR((csr_base_addr) + ADF_RING_CSR_ADDR_OFFSET, \
+		   ADF_RING_BUNDLE_SIZE * (bank) + \
+		   ADF_RING_CSR_RING_CONFIG + ((ring) << 2), value)
+#define WRITE_CSR_RING_BASE(csr_base_addr, bank, ring, value)	\
+do { \
+	void __iomem *_csr_base_addr = csr_base_addr; \
+	u32 _bank = bank;						\
+	u32 _ring = ring;						\
+	dma_addr_t _value = value;					\
+	u32 l_base = 0, u_base = 0;					\
+	l_base = lower_32_bits(_value);					\
+	u_base = upper_32_bits(_value);					\
+	ADF_CSR_WR((_csr_base_addr) + ADF_RING_CSR_ADDR_OFFSET,		\
+		   ADF_RING_BUNDLE_SIZE * (_bank) +			\
+		   ADF_RING_CSR_RING_LBASE + ((_ring) << 2), l_base);	\
+	ADF_CSR_WR((_csr_base_addr) + ADF_RING_CSR_ADDR_OFFSET,		\
+		   ADF_RING_BUNDLE_SIZE * (_bank) +			\
+		   ADF_RING_CSR_RING_UBASE + ((_ring) << 2), u_base);	\
+} while (0)
+
+static inline u64 read_base(void __iomem *csr_base_addr, u32 bank, u32 ring)
+{
+	u32 l_base, u_base;
+
+	/*
+	 * Use special IO wrapper for ring base as LBASE and UBASE are
+	 * not physically contigious
+	 */
+	l_base = ADF_CSR_RD(csr_base_addr, (ADF_RING_BUNDLE_SIZE * bank) +
+			    ADF_RING_CSR_RING_LBASE + (ring << 2));
+	u_base = ADF_CSR_RD(csr_base_addr, (ADF_RING_BUNDLE_SIZE * bank) +
+			    ADF_RING_CSR_RING_UBASE + (ring << 2));
+
+	return (u64)u_base << 32 | (u64)l_base;
+}
+
+#define READ_CSR_RING_BASE(csr_base_addr, bank, ring) \
+	read_base((csr_base_addr) + ADF_RING_CSR_ADDR_OFFSET, (bank), (ring))
+
+#define WRITE_CSR_RING_HEAD(csr_base_addr, bank, ring, value) \
+	ADF_CSR_WR((csr_base_addr) + ADF_RING_CSR_ADDR_OFFSET, \
+		   ADF_RING_BUNDLE_SIZE * (bank) + \
+		   ADF_RING_CSR_RING_HEAD + ((ring) << 2), value)
+#define WRITE_CSR_RING_TAIL(csr_base_addr, bank, ring, value) \
+	ADF_CSR_WR((csr_base_addr) + ADF_RING_CSR_ADDR_OFFSET, \
+		   ADF_RING_BUNDLE_SIZE * (bank) + \
+		   ADF_RING_CSR_RING_TAIL + ((ring) << 2), value)
+#define READ_CSR_INT_EN(csr_base_addr, bank) \
+	ADF_CSR_RD((csr_base_addr) + ADF_RING_CSR_ADDR_OFFSET, \
+		   ADF_RING_BUNDLE_SIZE * (bank) + ADF_RING_CSR_INT_FLAG_EN)
+#define WRITE_CSR_INT_EN(csr_base_addr, bank, value) \
+	ADF_CSR_WR((csr_base_addr) + ADF_RING_CSR_ADDR_OFFSET, \
+		   ADF_RING_BUNDLE_SIZE * (bank) + \
+		   ADF_RING_CSR_INT_FLAG_EN, (value))
+#define READ_CSR_INT_FLAG(csr_base_addr, bank) \
+	ADF_CSR_RD((csr_base_addr) + ADF_RING_CSR_ADDR_OFFSET, \
+		   ADF_RING_BUNDLE_SIZE * (bank) + ADF_RING_CSR_INT_FLAG)
+#define WRITE_CSR_INT_FLAG(csr_base_addr, bank, value) \
+	ADF_CSR_WR((csr_base_addr) + ADF_RING_CSR_ADDR_OFFSET, \
+		   ADF_RING_BUNDLE_SIZE * (bank) + \
+		   ADF_RING_CSR_INT_FLAG, (value))
+#define READ_CSR_INT_SRCSEL(csr_base_addr, bank) \
+	ADF_CSR_RD((csr_base_addr) + ADF_RING_CSR_ADDR_OFFSET, \
+		   ADF_RING_BUNDLE_SIZE * (bank) + ADF_RING_CSR_INT_SRCSEL)
+#define WRITE_CSR_INT_SRCSEL(csr_base_addr, bank) \
+	ADF_CSR_WR((csr_base_addr) + ADF_RING_CSR_ADDR_OFFSET, \
+		   ADF_RING_BUNDLE_SIZE * (bank) + \
+		   ADF_RING_CSR_INT_SRCSEL, ADF_BANK_INT_SRC_SEL_MASK)
+#define WRITE_CSR_INT_SRCSEL_W_VAL(csr_base_addr, bank, value) \
+	ADF_CSR_WR((csr_base_addr) + ADF_RING_CSR_ADDR_OFFSET, \
+		   ADF_RING_BUNDLE_SIZE * (bank) + \
+		   ADF_RING_CSR_INT_SRCSEL, (value))
+#define READ_CSR_INT_COL_EN(csr_base_addr, bank) \
+	ADF_CSR_RD((csr_base_addr) + ADF_RING_CSR_ADDR_OFFSET, \
+		   ADF_RING_BUNDLE_SIZE * (bank) + ADF_RING_CSR_INT_COL_EN)
+#define WRITE_CSR_INT_COL_EN(csr_base_addr, bank, value) \
+	ADF_CSR_WR((csr_base_addr) + ADF_RING_CSR_ADDR_OFFSET, \
+		   ADF_RING_BUNDLE_SIZE * (bank) + \
+		   ADF_RING_CSR_INT_COL_EN, (value))
+#define READ_CSR_INT_COL_CTL(csr_base_addr, bank) \
+	ADF_CSR_RD((csr_base_addr) + ADF_RING_CSR_ADDR_OFFSET, \
+		   ADF_RING_BUNDLE_SIZE * (bank) + ADF_RING_CSR_INT_COL_CTL)
+#define WRITE_CSR_INT_COL_CTL(csr_base_addr, bank, value) \
+	ADF_CSR_WR((csr_base_addr) + ADF_RING_CSR_ADDR_OFFSET, \
+		   ADF_RING_BUNDLE_SIZE * (bank) + \
+		   ADF_RING_CSR_INT_COL_CTL, \
+		   ADF_RING_CSR_INT_COL_CTL_ENABLE | (value))
+#define READ_CSR_INT_FLAG_AND_COL(csr_base_addr, bank) \
+	ADF_CSR_RD((csr_base_addr) + ADF_RING_CSR_ADDR_OFFSET, \
+		   ADF_RING_BUNDLE_SIZE * (bank) + \
+		   ADF_RING_CSR_INT_FLAG_AND_COL)
+#define WRITE_CSR_INT_FLAG_AND_COL(csr_base_addr, bank, value) \
+	ADF_CSR_WR((csr_base_addr) + ADF_RING_CSR_ADDR_OFFSET, \
+		   ADF_RING_BUNDLE_SIZE * (bank) + \
+		   ADF_RING_CSR_INT_FLAG_AND_COL, (value))
+
+#define READ_CSR_RING_SRV_ARB_EN(csr_base_addr, bank) \
+	ADF_CSR_RD((csr_base_addr) + ADF_RING_CSR_ADDR_OFFSET, \
+		   ADF_RING_BUNDLE_SIZE * (bank) + \
+		   ADF_RING_CSR_RING_SRV_ARB_EN)
+#define WRITE_CSR_RING_SRV_ARB_EN(csr_base_addr, bank, value) \
+	ADF_CSR_WR((csr_base_addr) + ADF_RING_CSR_ADDR_OFFSET, \
+		   ADF_RING_BUNDLE_SIZE * (bank) + \
+		   ADF_RING_CSR_RING_SRV_ARB_EN, (value))
+
+void adf_gen4_init_hw_csr_ops(struct adf_hw_csr_ops *csr_ops);
+
+#endif
diff --git a/drivers/crypto/intel/qat/qat_common/adf_gen4_hw_data.c b/drivers/crypto/intel/qat/qat_common/adf_gen4_hw_data.c
index d28e1921940a..41a0979e68c1 100644
--- a/drivers/crypto/intel/qat/qat_common/adf_gen4_hw_data.c
+++ b/drivers/crypto/intel/qat/qat_common/adf_gen4_hw_data.c
@@ -1,6 +1,7 @@
 // SPDX-License-Identifier: (BSD-3-Clause OR GPL-2.0-only)
 /* Copyright(c) 2020 Intel Corporation */
 #include <linux/iopoll.h>
+#include <asm/div64.h>
 #include "adf_accel_devices.h"
 #include "adf_cfg_services.h"
 #include "adf_common_drv.h"
@@ -8,103 +9,6 @@
 #include "adf_gen4_hw_data.h"
 #include "adf_gen4_pm.h"
 
-static u64 build_csr_ring_base_addr(dma_addr_t addr, u32 size)
-{
-	return BUILD_RING_BASE_ADDR(addr, size);
-}
-
-static u32 read_csr_ring_head(void __iomem *csr_base_addr, u32 bank, u32 ring)
-{
-	return READ_CSR_RING_HEAD(csr_base_addr, bank, ring);
-}
-
-static void write_csr_ring_head(void __iomem *csr_base_addr, u32 bank, u32 ring,
-				u32 value)
-{
-	WRITE_CSR_RING_HEAD(csr_base_addr, bank, ring, value);
-}
-
-static u32 read_csr_ring_tail(void __iomem *csr_base_addr, u32 bank, u32 ring)
-{
-	return READ_CSR_RING_TAIL(csr_base_addr, bank, ring);
-}
-
-static void write_csr_ring_tail(void __iomem *csr_base_addr, u32 bank, u32 ring,
-				u32 value)
-{
-	WRITE_CSR_RING_TAIL(csr_base_addr, bank, ring, value);
-}
-
-static u32 read_csr_e_stat(void __iomem *csr_base_addr, u32 bank)
-{
-	return READ_CSR_E_STAT(csr_base_addr, bank);
-}
-
-static void write_csr_ring_config(void __iomem *csr_base_addr, u32 bank, u32 ring,
-				  u32 value)
-{
-	WRITE_CSR_RING_CONFIG(csr_base_addr, bank, ring, value);
-}
-
-static void write_csr_ring_base(void __iomem *csr_base_addr, u32 bank, u32 ring,
-				dma_addr_t addr)
-{
-	WRITE_CSR_RING_BASE(csr_base_addr, bank, ring, addr);
-}
-
-static void write_csr_int_flag(void __iomem *csr_base_addr, u32 bank,
-			       u32 value)
-{
-	WRITE_CSR_INT_FLAG(csr_base_addr, bank, value);
-}
-
-static void write_csr_int_srcsel(void __iomem *csr_base_addr, u32 bank)
-{
-	WRITE_CSR_INT_SRCSEL(csr_base_addr, bank);
-}
-
-static void write_csr_int_col_en(void __iomem *csr_base_addr, u32 bank, u32 value)
-{
-	WRITE_CSR_INT_COL_EN(csr_base_addr, bank, value);
-}
-
-static void write_csr_int_col_ctl(void __iomem *csr_base_addr, u32 bank,
-				  u32 value)
-{
-	WRITE_CSR_INT_COL_CTL(csr_base_addr, bank, value);
-}
-
-static void write_csr_int_flag_and_col(void __iomem *csr_base_addr, u32 bank,
-				       u32 value)
-{
-	WRITE_CSR_INT_FLAG_AND_COL(csr_base_addr, bank, value);
-}
-
-static void write_csr_ring_srv_arb_en(void __iomem *csr_base_addr, u32 bank,
-				      u32 value)
-{
-	WRITE_CSR_RING_SRV_ARB_EN(csr_base_addr, bank, value);
-}
-
-void adf_gen4_init_hw_csr_ops(struct adf_hw_csr_ops *csr_ops)
-{
-	csr_ops->build_csr_ring_base_addr = build_csr_ring_base_addr;
-	csr_ops->read_csr_ring_head = read_csr_ring_head;
-	csr_ops->write_csr_ring_head = write_csr_ring_head;
-	csr_ops->read_csr_ring_tail = read_csr_ring_tail;
-	csr_ops->write_csr_ring_tail = write_csr_ring_tail;
-	csr_ops->read_csr_e_stat = read_csr_e_stat;
-	csr_ops->write_csr_ring_config = write_csr_ring_config;
-	csr_ops->write_csr_ring_base = write_csr_ring_base;
-	csr_ops->write_csr_int_flag = write_csr_int_flag;
-	csr_ops->write_csr_int_srcsel = write_csr_int_srcsel;
-	csr_ops->write_csr_int_col_en = write_csr_int_col_en;
-	csr_ops->write_csr_int_col_ctl = write_csr_int_col_ctl;
-	csr_ops->write_csr_int_flag_and_col = write_csr_int_flag_and_col;
-	csr_ops->write_csr_ring_srv_arb_en = write_csr_ring_srv_arb_en;
-}
-EXPORT_SYMBOL_GPL(adf_gen4_init_hw_csr_ops);
-
 u32 adf_gen4_get_accel_mask(struct adf_hw_device_data *self)
 {
 	return ADF_GEN4_ACCELERATORS_MASK;
@@ -321,8 +225,7 @@ static int reset_ring_pair(void __iomem *csr, u32 bank_number)
 int adf_gen4_ring_pair_reset(struct adf_accel_dev *accel_dev, u32 bank_number)
 {
 	struct adf_hw_device_data *hw_data = accel_dev->hw_device;
-	u32 etr_bar_id = hw_data->get_etr_bar_id(hw_data);
-	void __iomem *csr;
+	void __iomem *csr = adf_get_etr_base(accel_dev);
 	int ret;
 
 	if (bank_number >= hw_data->num_banks)
@@ -331,7 +234,6 @@ int adf_gen4_ring_pair_reset(struct adf_accel_dev *accel_dev, u32 bank_number)
 	dev_dbg(&GET_DEV(accel_dev),
 		"ring pair reset for bank:%d\n", bank_number);
 
-	csr = (&GET_BARS(accel_dev)[etr_bar_id])->virt_addr;
 	ret = reset_ring_pair(csr, bank_number);
 	if (ret)
 		dev_err(&GET_DEV(accel_dev),
@@ -489,3 +391,281 @@ set_mask:
 	return ring_to_svc_map;
 }
 EXPORT_SYMBOL_GPL(adf_gen4_get_ring_to_svc_map);
+
+/*
+ * adf_gen4_bank_quiesce_coal_timer() - quiesce bank coalesced interrupt timer
+ * @accel_dev: Pointer to the device structure
+ * @bank_idx: Offset to the bank within this device
+ * @timeout_ms: Timeout in milliseconds for the operation
+ *
+ * This function tries to quiesce the coalesced interrupt timer of a bank if
+ * it has been enabled and triggered.
+ *
+ * Returns 0 on success, error code otherwise
+ *
+ */
+int adf_gen4_bank_quiesce_coal_timer(struct adf_accel_dev *accel_dev,
+				     u32 bank_idx, int timeout_ms)
+{
+	struct adf_hw_device_data *hw_data = GET_HW_DATA(accel_dev);
+	struct adf_hw_csr_ops *csr_ops = GET_CSR_OPS(accel_dev);
+	void __iomem *csr_misc = adf_get_pmisc_base(accel_dev);
+	void __iomem *csr_etr = adf_get_etr_base(accel_dev);
+	u32 int_col_ctl, int_col_mask, int_col_en;
+	u32 e_stat, intsrc;
+	u64 wait_us;
+	int ret;
+
+	if (timeout_ms < 0)
+		return -EINVAL;
+
+	int_col_ctl = csr_ops->read_csr_int_col_ctl(csr_etr, bank_idx);
+	int_col_mask = csr_ops->get_int_col_ctl_enable_mask();
+	if (!(int_col_ctl & int_col_mask))
+		return 0;
+
+	int_col_en = csr_ops->read_csr_int_col_en(csr_etr, bank_idx);
+	int_col_en &= BIT(ADF_WQM_CSR_RP_IDX_RX);
+
+	e_stat = csr_ops->read_csr_e_stat(csr_etr, bank_idx);
+	if (!(~e_stat & int_col_en))
+		return 0;
+
+	wait_us = 2 * ((int_col_ctl & ~int_col_mask) << 8) * USEC_PER_SEC;
+	do_div(wait_us, hw_data->clock_frequency);
+	wait_us = min(wait_us, (u64)timeout_ms * USEC_PER_MSEC);
+	dev_dbg(&GET_DEV(accel_dev),
+		"wait for bank %d - coalesced timer expires in %llu us (max=%u ms estat=0x%x intcolen=0x%x)\n",
+		bank_idx, wait_us, timeout_ms, e_stat, int_col_en);
+
+	ret = read_poll_timeout(ADF_CSR_RD, intsrc, intsrc,
+				ADF_COALESCED_POLL_DELAY_US, wait_us, true,
+				csr_misc, ADF_WQM_CSR_RPINTSOU(bank_idx));
+	if (ret)
+		dev_warn(&GET_DEV(accel_dev),
+			 "coalesced timer for bank %d expired (%llu us)\n",
+			 bank_idx, wait_us);
+
+	return ret;
+}
+EXPORT_SYMBOL_GPL(adf_gen4_bank_quiesce_coal_timer);
+
+static int drain_bank(void __iomem *csr, u32 bank_number, int timeout_us)
+{
+	u32 status;
+
+	ADF_CSR_WR(csr, ADF_WQM_CSR_RPRESETCTL(bank_number),
+		   ADF_WQM_CSR_RPRESETCTL_DRAIN);
+
+	return read_poll_timeout(ADF_CSR_RD, status,
+				status & ADF_WQM_CSR_RPRESETSTS_STATUS,
+				ADF_RPRESET_POLL_DELAY_US, timeout_us, true,
+				csr, ADF_WQM_CSR_RPRESETSTS(bank_number));
+}
+
+void adf_gen4_bank_drain_finish(struct adf_accel_dev *accel_dev,
+				u32 bank_number)
+{
+	void __iomem *csr = adf_get_etr_base(accel_dev);
+
+	ADF_CSR_WR(csr, ADF_WQM_CSR_RPRESETSTS(bank_number),
+		   ADF_WQM_CSR_RPRESETSTS_STATUS);
+}
+
+int adf_gen4_bank_drain_start(struct adf_accel_dev *accel_dev,
+			      u32 bank_number, int timeout_us)
+{
+	void __iomem *csr = adf_get_etr_base(accel_dev);
+	int ret;
+
+	dev_dbg(&GET_DEV(accel_dev), "Drain bank %d\n", bank_number);
+
+	ret = drain_bank(csr, bank_number, timeout_us);
+	if (ret)
+		dev_err(&GET_DEV(accel_dev), "Bank drain failed (timeout)\n");
+	else
+		dev_dbg(&GET_DEV(accel_dev), "Bank drain successful\n");
+
+	return ret;
+}
+
+static void bank_state_save(struct adf_hw_csr_ops *ops, void __iomem *base,
+			    u32 bank, struct bank_state *state, u32 num_rings)
+{
+	u32 i;
+
+	state->ringstat0 = ops->read_csr_stat(base, bank);
+	state->ringuostat = ops->read_csr_uo_stat(base, bank);
+	state->ringestat = ops->read_csr_e_stat(base, bank);
+	state->ringnestat = ops->read_csr_ne_stat(base, bank);
+	state->ringnfstat = ops->read_csr_nf_stat(base, bank);
+	state->ringfstat = ops->read_csr_f_stat(base, bank);
+	state->ringcstat0 = ops->read_csr_c_stat(base, bank);
+	state->iaintflagen = ops->read_csr_int_en(base, bank);
+	state->iaintflagreg = ops->read_csr_int_flag(base, bank);
+	state->iaintflagsrcsel0 = ops->read_csr_int_srcsel(base, bank);
+	state->iaintcolen = ops->read_csr_int_col_en(base, bank);
+	state->iaintcolctl = ops->read_csr_int_col_ctl(base, bank);
+	state->iaintflagandcolen = ops->read_csr_int_flag_and_col(base, bank);
+	state->ringexpstat = ops->read_csr_exp_stat(base, bank);
+	state->ringexpintenable = ops->read_csr_exp_int_en(base, bank);
+	state->ringsrvarben = ops->read_csr_ring_srv_arb_en(base, bank);
+
+	for (i = 0; i < num_rings; i++) {
+		state->rings[i].head = ops->read_csr_ring_head(base, bank, i);
+		state->rings[i].tail = ops->read_csr_ring_tail(base, bank, i);
+		state->rings[i].config = ops->read_csr_ring_config(base, bank, i);
+		state->rings[i].base = ops->read_csr_ring_base(base, bank, i);
+	}
+}
+
+#define CHECK_STAT(op, expect_val, name, args...) \
+({ \
+	u32 __expect_val = (expect_val); \
+	u32 actual_val = op(args); \
+	(__expect_val == actual_val) ? 0 : \
+		(pr_err("QAT: Fail to restore %s register. Expected 0x%x, actual 0x%x\n", \
+			name, __expect_val, actual_val), -EINVAL); \
+})
+
+static int bank_state_restore(struct adf_hw_csr_ops *ops, void __iomem *base,
+			      u32 bank, struct bank_state *state, u32 num_rings,
+			      int tx_rx_gap)
+{
+	u32 val, tmp_val, i;
+	int ret;
+
+	for (i = 0; i < num_rings; i++)
+		ops->write_csr_ring_base(base, bank, i, state->rings[i].base);
+
+	for (i = 0; i < num_rings; i++)
+		ops->write_csr_ring_config(base, bank, i, state->rings[i].config);
+
+	for (i = 0; i < num_rings / 2; i++) {
+		int tx = i * (tx_rx_gap + 1);
+		int rx = tx + tx_rx_gap;
+
+		ops->write_csr_ring_head(base, bank, tx, state->rings[tx].head);
+		ops->write_csr_ring_tail(base, bank, tx, state->rings[tx].tail);
+
+		/*
+		 * The TX ring head needs to be updated again to make sure that
+		 * the HW will not consider the ring as full when it is empty
+		 * and the correct state flags are set to match the recovered state.
+		 */
+		if (state->ringestat & BIT(tx)) {
+			val = ops->read_csr_int_srcsel(base, bank);
+			val |= ADF_RP_INT_SRC_SEL_F_RISE_MASK;
+			ops->write_csr_int_srcsel_w_val(base, bank, val);
+			ops->write_csr_ring_head(base, bank, tx, state->rings[tx].head);
+		}
+
+		ops->write_csr_ring_tail(base, bank, rx, state->rings[rx].tail);
+		val = ops->read_csr_int_srcsel(base, bank);
+		val |= ADF_RP_INT_SRC_SEL_F_RISE_MASK << ADF_RP_INT_SRC_SEL_RANGE_WIDTH;
+		ops->write_csr_int_srcsel_w_val(base, bank, val);
+
+		ops->write_csr_ring_head(base, bank, rx, state->rings[rx].head);
+		val = ops->read_csr_int_srcsel(base, bank);
+		val |= ADF_RP_INT_SRC_SEL_F_FALL_MASK << ADF_RP_INT_SRC_SEL_RANGE_WIDTH;
+		ops->write_csr_int_srcsel_w_val(base, bank, val);
+
+		/*
+		 * The RX ring tail needs to be updated again to make sure that
+		 * the HW will not consider the ring as empty when it is full
+		 * and the correct state flags are set to match the recovered state.
+		 */
+		if (state->ringfstat & BIT(rx))
+			ops->write_csr_ring_tail(base, bank, rx, state->rings[rx].tail);
+	}
+
+	ops->write_csr_int_flag_and_col(base, bank, state->iaintflagandcolen);
+	ops->write_csr_int_en(base, bank, state->iaintflagen);
+	ops->write_csr_int_col_en(base, bank, state->iaintcolen);
+	ops->write_csr_int_srcsel_w_val(base, bank, state->iaintflagsrcsel0);
+	ops->write_csr_exp_int_en(base, bank, state->ringexpintenable);
+	ops->write_csr_int_col_ctl(base, bank, state->iaintcolctl);
+	ops->write_csr_ring_srv_arb_en(base, bank, state->ringsrvarben);
+
+	/* Check that all ring statuses match the saved state. */
+	ret = CHECK_STAT(ops->read_csr_stat, state->ringstat0, "ringstat",
+			 base, bank);
+	if (ret)
+		return ret;
+
+	ret = CHECK_STAT(ops->read_csr_e_stat, state->ringestat, "ringestat",
+			 base, bank);
+	if (ret)
+		return ret;
+
+	ret = CHECK_STAT(ops->read_csr_ne_stat, state->ringnestat, "ringnestat",
+			 base, bank);
+	if (ret)
+		return ret;
+
+	ret = CHECK_STAT(ops->read_csr_nf_stat, state->ringnfstat, "ringnfstat",
+			 base, bank);
+	if (ret)
+		return ret;
+
+	ret = CHECK_STAT(ops->read_csr_f_stat, state->ringfstat, "ringfstat",
+			 base, bank);
+	if (ret)
+		return ret;
+
+	ret = CHECK_STAT(ops->read_csr_c_stat, state->ringcstat0, "ringcstat",
+			 base, bank);
+	if (ret)
+		return ret;
+
+	tmp_val = ops->read_csr_exp_stat(base, bank);
+	val = state->ringexpstat;
+	if (tmp_val && !val) {
+		pr_err("QAT: Bank was restored with exception: 0x%x\n", val);
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+int adf_gen4_bank_state_save(struct adf_accel_dev *accel_dev, u32 bank_number,
+			     struct bank_state *state)
+{
+	struct adf_hw_device_data *hw_data = GET_HW_DATA(accel_dev);
+	struct adf_hw_csr_ops *csr_ops = GET_CSR_OPS(accel_dev);
+	void __iomem *csr_base = adf_get_etr_base(accel_dev);
+
+	if (bank_number >= hw_data->num_banks || !state)
+		return -EINVAL;
+
+	dev_dbg(&GET_DEV(accel_dev), "Saving state of bank %d\n", bank_number);
+
+	bank_state_save(csr_ops, csr_base, bank_number, state,
+			hw_data->num_rings_per_bank);
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(adf_gen4_bank_state_save);
+
+int adf_gen4_bank_state_restore(struct adf_accel_dev *accel_dev, u32 bank_number,
+				struct bank_state *state)
+{
+	struct adf_hw_device_data *hw_data = GET_HW_DATA(accel_dev);
+	struct adf_hw_csr_ops *csr_ops = GET_CSR_OPS(accel_dev);
+	void __iomem *csr_base = adf_get_etr_base(accel_dev);
+	int ret;
+
+	if (bank_number >= hw_data->num_banks  || !state)
+		return -EINVAL;
+
+	dev_dbg(&GET_DEV(accel_dev), "Restoring state of bank %d\n", bank_number);
+
+	ret = bank_state_restore(csr_ops, csr_base, bank_number, state,
+				 hw_data->num_rings_per_bank, hw_data->tx_rx_gap);
+	if (ret)
+		dev_err(&GET_DEV(accel_dev),
+			"Unable to restore state of bank %d\n", bank_number);
+
+	return ret;
+}
+EXPORT_SYMBOL_GPL(adf_gen4_bank_state_restore);
diff --git a/drivers/crypto/intel/qat/qat_common/adf_gen4_hw_data.h b/drivers/crypto/intel/qat/qat_common/adf_gen4_hw_data.h
index c6e80df5a85a..8b10926cedba 100644
--- a/drivers/crypto/intel/qat/qat_common/adf_gen4_hw_data.h
+++ b/drivers/crypto/intel/qat/qat_common/adf_gen4_hw_data.h
@@ -1,7 +1,7 @@
 /* SPDX-License-Identifier: (BSD-3-Clause OR GPL-2.0-only) */
 /* Copyright(c) 2020 Intel Corporation */
-#ifndef ADF_GEN4_HW_CSR_DATA_H_
-#define ADF_GEN4_HW_CSR_DATA_H_
+#ifndef ADF_GEN4_HW_DATA_H_
+#define ADF_GEN4_HW_DATA_H_
 
 #include <linux/units.h>
 
@@ -54,95 +54,6 @@
 #define ADF_GEN4_ADMINMSGLR_OFFSET	0x500578
 #define ADF_GEN4_MAILBOX_BASE_OFFSET	0x600970
 
-/* Transport access */
-#define ADF_BANK_INT_SRC_SEL_MASK	0x44UL
-#define ADF_RING_CSR_RING_CONFIG	0x1000
-#define ADF_RING_CSR_RING_LBASE		0x1040
-#define ADF_RING_CSR_RING_UBASE		0x1080
-#define ADF_RING_CSR_RING_HEAD		0x0C0
-#define ADF_RING_CSR_RING_TAIL		0x100
-#define ADF_RING_CSR_E_STAT		0x14C
-#define ADF_RING_CSR_INT_FLAG		0x170
-#define ADF_RING_CSR_INT_SRCSEL		0x174
-#define ADF_RING_CSR_INT_COL_CTL	0x180
-#define ADF_RING_CSR_INT_FLAG_AND_COL	0x184
-#define ADF_RING_CSR_INT_COL_CTL_ENABLE	0x80000000
-#define ADF_RING_CSR_INT_COL_EN		0x17C
-#define ADF_RING_CSR_ADDR_OFFSET	0x100000
-#define ADF_RING_BUNDLE_SIZE		0x2000
-
-#define BUILD_RING_BASE_ADDR(addr, size) \
-	((((addr) >> 6) & (GENMASK_ULL(63, 0) << (size))) << 6)
-#define READ_CSR_RING_HEAD(csr_base_addr, bank, ring) \
-	ADF_CSR_RD((csr_base_addr) + ADF_RING_CSR_ADDR_OFFSET, \
-		   ADF_RING_BUNDLE_SIZE * (bank) + \
-		   ADF_RING_CSR_RING_HEAD + ((ring) << 2))
-#define READ_CSR_RING_TAIL(csr_base_addr, bank, ring) \
-	ADF_CSR_RD((csr_base_addr) + ADF_RING_CSR_ADDR_OFFSET, \
-		   ADF_RING_BUNDLE_SIZE * (bank) + \
-		   ADF_RING_CSR_RING_TAIL + ((ring) << 2))
-#define READ_CSR_E_STAT(csr_base_addr, bank) \
-	ADF_CSR_RD((csr_base_addr) + ADF_RING_CSR_ADDR_OFFSET, \
-		   ADF_RING_BUNDLE_SIZE * (bank) + ADF_RING_CSR_E_STAT)
-#define WRITE_CSR_RING_CONFIG(csr_base_addr, bank, ring, value) \
-	ADF_CSR_WR((csr_base_addr) + ADF_RING_CSR_ADDR_OFFSET, \
-		   ADF_RING_BUNDLE_SIZE * (bank) + \
-		   ADF_RING_CSR_RING_CONFIG + ((ring) << 2), value)
-#define WRITE_CSR_RING_BASE(csr_base_addr, bank, ring, value)	\
-do { \
-	void __iomem *_csr_base_addr = csr_base_addr; \
-	u32 _bank = bank;						\
-	u32 _ring = ring;						\
-	dma_addr_t _value = value;					\
-	u32 l_base = 0, u_base = 0;					\
-	l_base = lower_32_bits(_value);					\
-	u_base = upper_32_bits(_value);					\
-	ADF_CSR_WR((_csr_base_addr) + ADF_RING_CSR_ADDR_OFFSET,		\
-		   ADF_RING_BUNDLE_SIZE * (_bank) +			\
-		   ADF_RING_CSR_RING_LBASE + ((_ring) << 2), l_base);	\
-	ADF_CSR_WR((_csr_base_addr) + ADF_RING_CSR_ADDR_OFFSET,		\
-		   ADF_RING_BUNDLE_SIZE * (_bank) +			\
-		   ADF_RING_CSR_RING_UBASE + ((_ring) << 2), u_base);	\
-} while (0)
-
-#define WRITE_CSR_RING_HEAD(csr_base_addr, bank, ring, value) \
-	ADF_CSR_WR((csr_base_addr) + ADF_RING_CSR_ADDR_OFFSET, \
-		   ADF_RING_BUNDLE_SIZE * (bank) + \
-		   ADF_RING_CSR_RING_HEAD + ((ring) << 2), value)
-#define WRITE_CSR_RING_TAIL(csr_base_addr, bank, ring, value) \
-	ADF_CSR_WR((csr_base_addr) + ADF_RING_CSR_ADDR_OFFSET, \
-		   ADF_RING_BUNDLE_SIZE * (bank) + \
-		   ADF_RING_CSR_RING_TAIL + ((ring) << 2), value)
-#define WRITE_CSR_INT_FLAG(csr_base_addr, bank, value) \
-	ADF_CSR_WR((csr_base_addr) + ADF_RING_CSR_ADDR_OFFSET, \
-		   ADF_RING_BUNDLE_SIZE * (bank) + \
-		   ADF_RING_CSR_INT_FLAG, (value))
-#define WRITE_CSR_INT_SRCSEL(csr_base_addr, bank) \
-	ADF_CSR_WR((csr_base_addr) + ADF_RING_CSR_ADDR_OFFSET, \
-		   ADF_RING_BUNDLE_SIZE * (bank) + \
-		   ADF_RING_CSR_INT_SRCSEL, ADF_BANK_INT_SRC_SEL_MASK)
-#define WRITE_CSR_INT_COL_EN(csr_base_addr, bank, value) \
-	ADF_CSR_WR((csr_base_addr) + ADF_RING_CSR_ADDR_OFFSET, \
-		   ADF_RING_BUNDLE_SIZE * (bank) + \
-		   ADF_RING_CSR_INT_COL_EN, (value))
-#define WRITE_CSR_INT_COL_CTL(csr_base_addr, bank, value) \
-	ADF_CSR_WR((csr_base_addr) + ADF_RING_CSR_ADDR_OFFSET, \
-		   ADF_RING_BUNDLE_SIZE * (bank) + \
-		   ADF_RING_CSR_INT_COL_CTL, \
-		   ADF_RING_CSR_INT_COL_CTL_ENABLE | (value))
-#define WRITE_CSR_INT_FLAG_AND_COL(csr_base_addr, bank, value) \
-	ADF_CSR_WR((csr_base_addr) + ADF_RING_CSR_ADDR_OFFSET, \
-		   ADF_RING_BUNDLE_SIZE * (bank) + \
-		   ADF_RING_CSR_INT_FLAG_AND_COL, (value))
-
-/* Arbiter configuration */
-#define ADF_RING_CSR_RING_SRV_ARB_EN 0x19C
-
-#define WRITE_CSR_RING_SRV_ARB_EN(csr_base_addr, bank, value) \
-	ADF_CSR_WR((csr_base_addr) + ADF_RING_CSR_ADDR_OFFSET, \
-		   ADF_RING_BUNDLE_SIZE * (bank) + \
-		   ADF_RING_CSR_RING_SRV_ARB_EN, (value))
-
 /* Default ring mapping */
 #define ADF_GEN4_DEFAULT_RING_TO_SRV_MAP \
 	(ASYM << ADF_CFG_SERV_RING_PAIR_0_SHIFT | \
@@ -166,10 +77,20 @@ do { \
 #define ADF_RPRESET_POLL_TIMEOUT_US	(5 * USEC_PER_SEC)
 #define ADF_RPRESET_POLL_DELAY_US	20
 #define ADF_WQM_CSR_RPRESETCTL_RESET	BIT(0)
+#define ADF_WQM_CSR_RPRESETCTL_DRAIN	BIT(2)
 #define ADF_WQM_CSR_RPRESETCTL(bank)	(0x6000 + ((bank) << 3))
 #define ADF_WQM_CSR_RPRESETSTS_STATUS	BIT(0)
 #define ADF_WQM_CSR_RPRESETSTS(bank)	(ADF_WQM_CSR_RPRESETCTL(bank) + 4)
 
+/* Ring interrupt */
+#define ADF_RP_INT_SRC_SEL_F_RISE_MASK	BIT(2)
+#define ADF_RP_INT_SRC_SEL_F_FALL_MASK	GENMASK(2, 0)
+#define ADF_RP_INT_SRC_SEL_RANGE_WIDTH	4
+#define ADF_COALESCED_POLL_TIMEOUT_US	(1 * USEC_PER_SEC)
+#define ADF_COALESCED_POLL_DELAY_US	1000
+#define ADF_WQM_CSR_RPINTSOU(bank)	(0x200000 + ((bank) << 12))
+#define ADF_WQM_CSR_RP_IDX_RX		1
+
 /* Error source registers */
 #define ADF_GEN4_ERRSOU0	(0x41A200)
 #define ADF_GEN4_ERRSOU1	(0x41A204)
@@ -197,6 +118,19 @@ do { \
 /* Arbiter threads mask with error value */
 #define ADF_GEN4_ENA_THD_MASK_ERROR	GENMASK(ADF_NUM_THREADS_PER_AE, 0)
 
+/* PF2VM communication channel */
+#define ADF_GEN4_PF2VM_OFFSET(i)	(0x40B010 + (i) * 0x20)
+#define ADF_GEN4_VM2PF_OFFSET(i)	(0x40B014 + (i) * 0x20)
+#define ADF_GEN4_VINTMSKPF2VM_OFFSET(i)	(0x40B00C + (i) * 0x20)
+#define ADF_GEN4_VINTSOUPF2VM_OFFSET(i)	(0x40B008 + (i) * 0x20)
+#define ADF_GEN4_VINTMSK_OFFSET(i)	(0x40B004 + (i) * 0x20)
+#define ADF_GEN4_VINTSOU_OFFSET(i)	(0x40B000 + (i) * 0x20)
+
+struct adf_gen4_vfmig {
+	struct adf_mstate_mgr *mstate_mgr;
+	bool bank_stopped[ADF_GEN4_NUM_BANKS_PER_VF];
+};
+
 void adf_gen4_set_ssm_wdtimer(struct adf_accel_dev *accel_dev);
 
 enum icp_qat_gen4_slice_mask {
@@ -230,11 +164,20 @@ u32 adf_gen4_get_num_aes(struct adf_hw_device_data *self);
 enum dev_sku_info adf_gen4_get_sku(struct adf_hw_device_data *self);
 u32 adf_gen4_get_sram_bar_id(struct adf_hw_device_data *self);
 int adf_gen4_init_device(struct adf_accel_dev *accel_dev);
-void adf_gen4_init_hw_csr_ops(struct adf_hw_csr_ops *csr_ops);
 int adf_gen4_ring_pair_reset(struct adf_accel_dev *accel_dev, u32 bank_number);
 void adf_gen4_set_msix_default_rttable(struct adf_accel_dev *accel_dev);
 void adf_gen4_set_ssm_wdtimer(struct adf_accel_dev *accel_dev);
 int adf_gen4_init_thd2arb_map(struct adf_accel_dev *accel_dev);
 u16 adf_gen4_get_ring_to_svc_map(struct adf_accel_dev *accel_dev);
+int adf_gen4_bank_quiesce_coal_timer(struct adf_accel_dev *accel_dev,
+				     u32 bank_idx, int timeout_ms);
+int adf_gen4_bank_drain_start(struct adf_accel_dev *accel_dev,
+			      u32 bank_number, int timeout_us);
+void adf_gen4_bank_drain_finish(struct adf_accel_dev *accel_dev,
+				u32 bank_number);
+int adf_gen4_bank_state_save(struct adf_accel_dev *accel_dev, u32 bank_number,
+			     struct bank_state *state);
+int adf_gen4_bank_state_restore(struct adf_accel_dev *accel_dev,
+				u32 bank_number, struct bank_state *state);
 
 #endif
diff --git a/drivers/crypto/intel/qat/qat_common/adf_gen4_pfvf.c b/drivers/crypto/intel/qat/qat_common/adf_gen4_pfvf.c
index 8e8efe93f3ee..21474d402d09 100644
--- a/drivers/crypto/intel/qat/qat_common/adf_gen4_pfvf.c
+++ b/drivers/crypto/intel/qat/qat_common/adf_gen4_pfvf.c
@@ -6,12 +6,10 @@
 #include "adf_accel_devices.h"
 #include "adf_common_drv.h"
 #include "adf_gen4_pfvf.h"
+#include "adf_gen4_hw_data.h"
 #include "adf_pfvf_pf_proto.h"
 #include "adf_pfvf_utils.h"
 
-#define ADF_4XXX_PF2VM_OFFSET(i)	(0x40B010 + ((i) * 0x20))
-#define ADF_4XXX_VM2PF_OFFSET(i)	(0x40B014 + ((i) * 0x20))
-
 /* VF2PF interrupt source registers */
 #define ADF_4XXX_VM2PF_SOU		0x41A180
 #define ADF_4XXX_VM2PF_MSK		0x41A1C0
@@ -29,12 +27,12 @@ static const struct pfvf_csr_format csr_gen4_fmt = {
 
 static u32 adf_gen4_pf_get_pf2vf_offset(u32 i)
 {
-	return ADF_4XXX_PF2VM_OFFSET(i);
+	return ADF_GEN4_PF2VM_OFFSET(i);
 }
 
 static u32 adf_gen4_pf_get_vf2pf_offset(u32 i)
 {
-	return ADF_4XXX_VM2PF_OFFSET(i);
+	return ADF_GEN4_VM2PF_OFFSET(i);
 }
 
 static void adf_gen4_enable_vf2pf_interrupts(void __iomem *pmisc_addr, u32 vf_mask)
diff --git a/drivers/crypto/intel/qat/qat_common/adf_gen4_tl.c b/drivers/crypto/intel/qat/qat_common/adf_gen4_tl.c
index 7fc7a77f6aed..c7ad8cf07863 100644
--- a/drivers/crypto/intel/qat/qat_common/adf_gen4_tl.c
+++ b/drivers/crypto/intel/qat/qat_common/adf_gen4_tl.c
@@ -149,5 +149,6 @@ void adf_gen4_init_tl_data(struct adf_tl_hw_data *tl_data)
 	tl_data->sl_exec_counters = sl_exec_counters;
 	tl_data->rp_counters = rp_counters;
 	tl_data->num_rp_counters = ARRAY_SIZE(rp_counters);
+	tl_data->max_sl_cnt = ADF_GEN4_TL_MAX_SLICES_PER_TYPE;
 }
 EXPORT_SYMBOL_GPL(adf_gen4_init_tl_data);
diff --git a/drivers/crypto/intel/qat/qat_common/adf_gen4_vf_mig.c b/drivers/crypto/intel/qat/qat_common/adf_gen4_vf_mig.c
new file mode 100644
index 000000000000..a62eb5e8dbe6
--- /dev/null
+++ b/drivers/crypto/intel/qat/qat_common/adf_gen4_vf_mig.c
@@ -0,0 +1,1010 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/* Copyright(c) 2024 Intel Corporation */
+#include <linux/delay.h>
+#include <linux/dev_printk.h>
+#include <linux/kernel.h>
+#include <linux/slab.h>
+#include <linux/string.h>
+#include <linux/types.h>
+#include <asm/errno.h>
+
+#include "adf_accel_devices.h"
+#include "adf_common_drv.h"
+#include "adf_gen4_hw_data.h"
+#include "adf_gen4_pfvf.h"
+#include "adf_pfvf_utils.h"
+#include "adf_mstate_mgr.h"
+#include "adf_gen4_vf_mig.h"
+
+#define ADF_GEN4_VF_MSTATE_SIZE		4096
+#define ADF_GEN4_PFVF_RSP_TIMEOUT_US	5000
+
+static int adf_gen4_vfmig_save_setup(struct qat_mig_dev *mdev);
+static int adf_gen4_vfmig_load_setup(struct qat_mig_dev *mdev, int len);
+
+static int adf_gen4_vfmig_init_device(struct qat_mig_dev *mdev)
+{
+	u8 *state;
+
+	state = kmalloc(ADF_GEN4_VF_MSTATE_SIZE, GFP_KERNEL);
+	if (!state)
+		return -ENOMEM;
+
+	mdev->state = state;
+	mdev->state_size = ADF_GEN4_VF_MSTATE_SIZE;
+	mdev->setup_size = 0;
+	mdev->remote_setup_size = 0;
+
+	return 0;
+}
+
+static void adf_gen4_vfmig_cleanup_device(struct qat_mig_dev *mdev)
+{
+	kfree(mdev->state);
+	mdev->state = NULL;
+}
+
+static void adf_gen4_vfmig_reset_device(struct qat_mig_dev *mdev)
+{
+	mdev->setup_size = 0;
+	mdev->remote_setup_size = 0;
+}
+
+static int adf_gen4_vfmig_open_device(struct qat_mig_dev *mdev)
+{
+	struct adf_accel_dev *accel_dev = mdev->parent_accel_dev;
+	struct adf_accel_vf_info *vf_info;
+	struct adf_gen4_vfmig *vfmig;
+
+	vf_info = &accel_dev->pf.vf_info[mdev->vf_id];
+
+	vfmig = kzalloc(sizeof(*vfmig), GFP_KERNEL);
+	if (!vfmig)
+		return -ENOMEM;
+
+	vfmig->mstate_mgr = adf_mstate_mgr_new(mdev->state, mdev->state_size);
+	if (!vfmig->mstate_mgr) {
+		kfree(vfmig);
+		return -ENOMEM;
+	}
+	vf_info->mig_priv = vfmig;
+	mdev->setup_size = 0;
+	mdev->remote_setup_size = 0;
+
+	return 0;
+}
+
+static void adf_gen4_vfmig_close_device(struct qat_mig_dev *mdev)
+{
+	struct adf_accel_dev *accel_dev = mdev->parent_accel_dev;
+	struct adf_accel_vf_info *vf_info;
+	struct adf_gen4_vfmig *vfmig;
+
+	vf_info = &accel_dev->pf.vf_info[mdev->vf_id];
+	if (vf_info->mig_priv) {
+		vfmig = vf_info->mig_priv;
+		adf_mstate_mgr_destroy(vfmig->mstate_mgr);
+		kfree(vfmig);
+		vf_info->mig_priv = NULL;
+	}
+}
+
+static int adf_gen4_vfmig_suspend_device(struct qat_mig_dev *mdev)
+{
+	struct adf_accel_dev *accel_dev = mdev->parent_accel_dev;
+	struct adf_hw_device_data *hw_data = accel_dev->hw_device;
+	struct adf_accel_vf_info *vf_info;
+	struct adf_gen4_vfmig *vf_mig;
+	u32 vf_nr = mdev->vf_id;
+	int ret, i;
+
+	vf_info = &accel_dev->pf.vf_info[vf_nr];
+	vf_mig = vf_info->mig_priv;
+
+	/* Stop all inflight jobs */
+	for (i = 0; i < hw_data->num_banks_per_vf; i++) {
+		u32 pf_bank_nr = i + vf_nr * hw_data->num_banks_per_vf;
+
+		ret = adf_gen4_bank_drain_start(accel_dev, pf_bank_nr,
+						ADF_RPRESET_POLL_TIMEOUT_US);
+		if (ret) {
+			dev_err(&GET_DEV(accel_dev),
+				"Failed to drain bank %d for vf_nr %d\n", i,
+				vf_nr);
+			return ret;
+		}
+		vf_mig->bank_stopped[i] = true;
+
+		adf_gen4_bank_quiesce_coal_timer(accel_dev, pf_bank_nr,
+						 ADF_COALESCED_POLL_TIMEOUT_US);
+	}
+
+	return 0;
+}
+
+static int adf_gen4_vfmig_resume_device(struct qat_mig_dev *mdev)
+{
+	struct adf_accel_dev *accel_dev = mdev->parent_accel_dev;
+	struct adf_hw_device_data *hw_data = accel_dev->hw_device;
+	struct adf_accel_vf_info *vf_info;
+	struct adf_gen4_vfmig *vf_mig;
+	u32 vf_nr = mdev->vf_id;
+	int i;
+
+	vf_info = &accel_dev->pf.vf_info[vf_nr];
+	vf_mig = vf_info->mig_priv;
+
+	for (i = 0; i < hw_data->num_banks_per_vf; i++) {
+		u32 pf_bank_nr = i + vf_nr * hw_data->num_banks_per_vf;
+
+		if (vf_mig->bank_stopped[i]) {
+			adf_gen4_bank_drain_finish(accel_dev, pf_bank_nr);
+			vf_mig->bank_stopped[i] = false;
+		}
+	}
+
+	return 0;
+}
+
+struct adf_vf_bank_info {
+	struct adf_accel_dev *accel_dev;
+	u32 vf_nr;
+	u32 bank_nr;
+};
+
+struct mig_user_sla {
+	enum adf_base_services srv;
+	u64 rp_mask;
+	u32 cir;
+	u32 pir;
+};
+
+static int adf_mstate_sla_check(struct adf_mstate_mgr *sub_mgr, u8 *src_buf,
+				u32 src_size, void *opaque)
+{
+	struct adf_mstate_vreginfo _sinfo = { src_buf, src_size };
+	struct adf_mstate_vreginfo *sinfo = &_sinfo, *dinfo = opaque;
+	u32 src_sla_cnt = sinfo->size / sizeof(struct mig_user_sla);
+	u32 dst_sla_cnt = dinfo->size / sizeof(struct mig_user_sla);
+	struct mig_user_sla *src_slas = sinfo->addr;
+	struct mig_user_sla *dst_slas = dinfo->addr;
+	int i, j;
+
+	for (i = 0; i < src_sla_cnt; i++) {
+		for (j = 0; j < dst_sla_cnt; j++) {
+			if (src_slas[i].srv != dst_slas[j].srv ||
+			    src_slas[i].rp_mask != dst_slas[j].rp_mask)
+				continue;
+
+			if (src_slas[i].cir > dst_slas[j].cir ||
+			    src_slas[i].pir > dst_slas[j].pir) {
+				pr_err("QAT: DST VF rate limiting mismatch.\n");
+				return -EINVAL;
+			}
+			break;
+		}
+
+		if (j == dst_sla_cnt) {
+			pr_err("QAT: SRC VF rate limiting mismatch - SRC srv %d and rp_mask 0x%llx.\n",
+			       src_slas[i].srv, src_slas[i].rp_mask);
+			return -EINVAL;
+		}
+	}
+
+	return 0;
+}
+
+static inline int adf_mstate_check_cap_size(u32 src_sz, u32 dst_sz, u32 max_sz)
+{
+	if (src_sz > max_sz || dst_sz > max_sz)
+		return -EINVAL;
+	else
+		return 0;
+}
+
+static int adf_mstate_compatver_check(struct adf_mstate_mgr *sub_mgr,
+				      u8 *src_buf, u32 src_sz, void *opaque)
+{
+	struct adf_mstate_vreginfo *info = opaque;
+	u8 compat = 0;
+	u8 *pcompat;
+
+	if (src_sz != info->size) {
+		pr_debug("QAT: State mismatch (compat version size), current %u, expected %u\n",
+			 src_sz, info->size);
+		return -EINVAL;
+	}
+
+	memcpy(info->addr, src_buf, info->size);
+	pcompat = info->addr;
+	if (*pcompat == 0) {
+		pr_warn("QAT: Unable to determine the version of VF\n");
+		return 0;
+	}
+
+	compat = adf_vf_compat_checker(*pcompat);
+	if (compat == ADF_PF2VF_VF_INCOMPATIBLE) {
+		pr_debug("QAT: SRC VF driver (ver=%u) is incompatible with DST PF driver (ver=%u)\n",
+			 *pcompat, ADF_PFVF_COMPAT_THIS_VERSION);
+		return -EINVAL;
+	}
+
+	if (compat == ADF_PF2VF_VF_COMPAT_UNKNOWN)
+		pr_debug("QAT: SRC VF driver (ver=%u) is newer than DST PF driver (ver=%u)\n",
+			 *pcompat, ADF_PFVF_COMPAT_THIS_VERSION);
+
+	return 0;
+}
+
+/*
+ * adf_mstate_capmask_compare() - compare QAT device capability mask
+ * @sinfo:	Pointer to source capability info
+ * @dinfo:	Pointer to target capability info
+ *
+ * This function compares the capability mask between source VF and target VF
+ *
+ * Returns: 0 if target capability mask is identical to source capability mask,
+ * 1 if target mask can represent all the capabilities represented by source mask,
+ * -1 if target mask can't represent all the capabilities represented by source
+ * mask.
+ */
+static int adf_mstate_capmask_compare(struct adf_mstate_vreginfo *sinfo,
+				      struct adf_mstate_vreginfo *dinfo)
+{
+	u64 src = 0, dst = 0;
+
+	if (adf_mstate_check_cap_size(sinfo->size, dinfo->size, sizeof(u64))) {
+		pr_debug("QAT: Unexpected capability size %u %u %zu\n",
+			 sinfo->size, dinfo->size, sizeof(u64));
+		return -1;
+	}
+
+	memcpy(&src, sinfo->addr, sinfo->size);
+	memcpy(&dst, dinfo->addr, dinfo->size);
+
+	pr_debug("QAT: Check cap compatibility of cap %llu %llu\n", src, dst);
+
+	if (src == dst)
+		return 0;
+
+	if ((src | dst) == dst)
+		return 1;
+
+	return -1;
+}
+
+static int adf_mstate_capmask_superset(struct adf_mstate_mgr *sub_mgr, u8 *buf,
+				       u32 size, void *opa)
+{
+	struct adf_mstate_vreginfo sinfo = { buf, size };
+
+	if (adf_mstate_capmask_compare(&sinfo, opa) >= 0)
+		return 0;
+
+	return -EINVAL;
+}
+
+static int adf_mstate_capmask_equal(struct adf_mstate_mgr *sub_mgr, u8 *buf,
+				    u32 size, void *opa)
+{
+	struct adf_mstate_vreginfo sinfo = { buf, size };
+
+	if (adf_mstate_capmask_compare(&sinfo, opa) == 0)
+		return 0;
+
+	return -EINVAL;
+}
+
+static int adf_mstate_set_vreg(struct adf_mstate_mgr *sub_mgr, u8 *buf,
+			       u32 size, void *opa)
+{
+	struct adf_mstate_vreginfo *info = opa;
+
+	if (size != info->size) {
+		pr_debug("QAT: Unexpected cap size %u %u\n", size, info->size);
+		return -EINVAL;
+	}
+	memcpy(info->addr, buf, info->size);
+
+	return 0;
+}
+
+static u32 adf_gen4_vfmig_get_slas(struct adf_accel_dev *accel_dev, u32 vf_nr,
+				   struct mig_user_sla *pmig_slas)
+{
+	struct adf_hw_device_data *hw_data = accel_dev->hw_device;
+	struct adf_rl *rl_data = accel_dev->rate_limiting;
+	struct rl_sla **sla_type_arr = NULL;
+	u64 rp_mask, rp_index;
+	u32 max_num_sla;
+	u32 sla_cnt = 0;
+	int i, j;
+
+	if (!accel_dev->rate_limiting)
+		return 0;
+
+	rp_index = vf_nr * hw_data->num_banks_per_vf;
+	max_num_sla = adf_rl_get_sla_arr_of_type(rl_data, RL_LEAF, &sla_type_arr);
+
+	for (i = 0; i < max_num_sla; i++) {
+		if (!sla_type_arr[i])
+			continue;
+
+		rp_mask = 0;
+		for (j = 0; j < sla_type_arr[i]->ring_pairs_cnt; j++)
+			rp_mask |= BIT(sla_type_arr[i]->ring_pairs_ids[j]);
+
+		if (rp_mask & GENMASK_ULL(rp_index + 3, rp_index)) {
+			pmig_slas->rp_mask = rp_mask;
+			pmig_slas->cir = sla_type_arr[i]->cir;
+			pmig_slas->pir = sla_type_arr[i]->pir;
+			pmig_slas->srv = sla_type_arr[i]->srv;
+			pmig_slas++;
+			sla_cnt++;
+		}
+	}
+
+	return sla_cnt;
+}
+
+static int adf_gen4_vfmig_load_etr_regs(struct adf_mstate_mgr *sub_mgr,
+					u8 *state, u32 size, void *opa)
+{
+	struct adf_vf_bank_info *vf_bank_info = opa;
+	struct adf_accel_dev *accel_dev = vf_bank_info->accel_dev;
+	struct adf_hw_device_data *hw_data = accel_dev->hw_device;
+	u32 pf_bank_nr;
+	int ret;
+
+	pf_bank_nr = vf_bank_info->bank_nr + vf_bank_info->vf_nr * hw_data->num_banks_per_vf;
+	ret = hw_data->bank_state_restore(accel_dev, pf_bank_nr,
+					  (struct bank_state *)state);
+	if (ret) {
+		dev_err(&GET_DEV(accel_dev),
+			"Failed to load regs for vf%d bank%d\n",
+			vf_bank_info->vf_nr, vf_bank_info->bank_nr);
+		return ret;
+	}
+
+	return 0;
+}
+
+static int adf_gen4_vfmig_load_etr_bank(struct adf_accel_dev *accel_dev,
+					u32 vf_nr, u32 bank_nr,
+					struct adf_mstate_mgr *mstate_mgr)
+{
+	struct adf_vf_bank_info vf_bank_info = {accel_dev, vf_nr, bank_nr};
+	struct adf_mstate_sect_h *subsec, *l2_subsec;
+	struct adf_mstate_mgr sub_sects_mgr;
+	char bank_ids[ADF_MSTATE_ID_LEN];
+
+	snprintf(bank_ids, sizeof(bank_ids), ADF_MSTATE_BANK_IDX_IDS "%x", bank_nr);
+	subsec = adf_mstate_sect_lookup(mstate_mgr, bank_ids, NULL, NULL);
+	if (!subsec) {
+		dev_err(&GET_DEV(accel_dev),
+			"Failed to lookup sec %s for vf%d bank%d\n",
+			ADF_MSTATE_BANK_IDX_IDS, vf_nr, bank_nr);
+		return -EINVAL;
+	}
+
+	adf_mstate_mgr_init_from_psect(&sub_sects_mgr, subsec);
+	l2_subsec = adf_mstate_sect_lookup(&sub_sects_mgr, ADF_MSTATE_ETR_REGS_IDS,
+					   adf_gen4_vfmig_load_etr_regs,
+					   &vf_bank_info);
+	if (!l2_subsec) {
+		dev_err(&GET_DEV(accel_dev),
+			"Failed to add sec %s for vf%d bank%d\n",
+			ADF_MSTATE_ETR_REGS_IDS, vf_nr, bank_nr);
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+static int adf_gen4_vfmig_load_etr(struct adf_accel_dev *accel_dev, u32 vf_nr)
+{
+	struct adf_accel_vf_info *vf_info = &accel_dev->pf.vf_info[vf_nr];
+	struct adf_hw_device_data *hw_data = accel_dev->hw_device;
+	struct adf_gen4_vfmig *vfmig = vf_info->mig_priv;
+	struct adf_mstate_mgr *mstate_mgr = vfmig->mstate_mgr;
+	struct adf_mstate_mgr sub_sects_mgr;
+	struct adf_mstate_sect_h *subsec;
+	int ret, i;
+
+	subsec = adf_mstate_sect_lookup(mstate_mgr, ADF_MSTATE_ETRB_IDS, NULL,
+					NULL);
+	if (!subsec) {
+		dev_err(&GET_DEV(accel_dev), "Failed to load sec %s\n",
+			ADF_MSTATE_ETRB_IDS);
+		return -EINVAL;
+	}
+
+	adf_mstate_mgr_init_from_psect(&sub_sects_mgr, subsec);
+	for (i = 0; i < hw_data->num_banks_per_vf; i++) {
+		ret = adf_gen4_vfmig_load_etr_bank(accel_dev, vf_nr, i,
+						   &sub_sects_mgr);
+		if (ret)
+			return ret;
+	}
+
+	return 0;
+}
+
+static int adf_gen4_vfmig_load_misc(struct adf_accel_dev *accel_dev, u32 vf_nr)
+{
+	struct adf_accel_vf_info *vf_info = &accel_dev->pf.vf_info[vf_nr];
+	struct adf_gen4_vfmig *vfmig = vf_info->mig_priv;
+	void __iomem *csr = adf_get_pmisc_base(accel_dev);
+	struct adf_mstate_mgr *mstate_mgr = vfmig->mstate_mgr;
+	struct adf_mstate_sect_h *subsec, *l2_subsec;
+	struct adf_mstate_mgr sub_sects_mgr;
+	struct {
+		char *id;
+		u64 ofs;
+	} misc_states[] = {
+		{ADF_MSTATE_VINTMSK_IDS, ADF_GEN4_VINTMSK_OFFSET(vf_nr)},
+		{ADF_MSTATE_VINTMSK_PF2VM_IDS, ADF_GEN4_VINTMSKPF2VM_OFFSET(vf_nr)},
+		{ADF_MSTATE_PF2VM_IDS, ADF_GEN4_PF2VM_OFFSET(vf_nr)},
+		{ADF_MSTATE_VM2PF_IDS, ADF_GEN4_VM2PF_OFFSET(vf_nr)},
+	};
+	int i;
+
+	subsec = adf_mstate_sect_lookup(mstate_mgr, ADF_MSTATE_MISCB_IDS, NULL,
+					NULL);
+	if (!subsec) {
+		dev_err(&GET_DEV(accel_dev), "Failed to load sec %s\n",
+			ADF_MSTATE_MISCB_IDS);
+		return -EINVAL;
+	}
+
+	adf_mstate_mgr_init_from_psect(&sub_sects_mgr, subsec);
+	for (i = 0; i < ARRAY_SIZE(misc_states); i++) {
+		struct adf_mstate_vreginfo info;
+		u32 regv;
+
+		info.addr = &regv;
+		info.size = sizeof(regv);
+		l2_subsec = adf_mstate_sect_lookup(&sub_sects_mgr,
+						   misc_states[i].id,
+						   adf_mstate_set_vreg,
+						   &info);
+		if (!l2_subsec) {
+			dev_err(&GET_DEV(accel_dev),
+				"Failed to load sec %s\n", misc_states[i].id);
+			return -EINVAL;
+		}
+		ADF_CSR_WR(csr, misc_states[i].ofs, regv);
+	}
+
+	return 0;
+}
+
+static int adf_gen4_vfmig_load_generic(struct adf_accel_dev *accel_dev, u32 vf_nr)
+{
+	struct adf_accel_vf_info *vf_info = &accel_dev->pf.vf_info[vf_nr];
+	struct mig_user_sla dst_slas[RL_RP_CNT_PER_LEAF_MAX] = { };
+	struct adf_gen4_vfmig *vfmig = vf_info->mig_priv;
+	struct adf_mstate_mgr *mstate_mgr = vfmig->mstate_mgr;
+	struct adf_mstate_sect_h *subsec, *l2_subsec;
+	struct adf_mstate_mgr sub_sects_mgr;
+	u32 dst_sla_cnt;
+	struct {
+		char *id;
+		int (*action)(struct adf_mstate_mgr *sub_mgr, u8 *buf, u32 size, void *opa);
+		struct adf_mstate_vreginfo info;
+	} gen_states[] = {
+		{ADF_MSTATE_IOV_INIT_IDS, adf_mstate_set_vreg,
+		{&vf_info->init, sizeof(vf_info->init)}},
+		{ADF_MSTATE_COMPAT_VER_IDS, adf_mstate_compatver_check,
+		{&vf_info->vf_compat_ver, sizeof(vf_info->vf_compat_ver)}},
+		{ADF_MSTATE_SLA_IDS, adf_mstate_sla_check, {dst_slas, 0}},
+	};
+	int i;
+
+	subsec = adf_mstate_sect_lookup(mstate_mgr, ADF_MSTATE_GEN_IDS, NULL, NULL);
+	if (!subsec) {
+		dev_err(&GET_DEV(accel_dev), "Failed to load sec %s\n",
+			ADF_MSTATE_GEN_IDS);
+		return -EINVAL;
+	}
+
+	adf_mstate_mgr_init_from_psect(&sub_sects_mgr, subsec);
+	for (i = 0; i < ARRAY_SIZE(gen_states); i++) {
+		if (gen_states[i].info.addr == dst_slas) {
+			dst_sla_cnt = adf_gen4_vfmig_get_slas(accel_dev, vf_nr, dst_slas);
+			gen_states[i].info.size = dst_sla_cnt * sizeof(struct mig_user_sla);
+		}
+
+		l2_subsec = adf_mstate_sect_lookup(&sub_sects_mgr,
+						   gen_states[i].id,
+						   gen_states[i].action,
+						   &gen_states[i].info);
+		if (!l2_subsec) {
+			dev_err(&GET_DEV(accel_dev), "Failed to load sec %s\n",
+				gen_states[i].id);
+			return -EINVAL;
+		}
+	}
+
+	return 0;
+}
+
+static int adf_gen4_vfmig_load_config(struct adf_accel_dev *accel_dev, u32 vf_nr)
+{
+	struct adf_accel_vf_info *vf_info = &accel_dev->pf.vf_info[vf_nr];
+	struct adf_hw_device_data *hw_data = accel_dev->hw_device;
+	struct adf_gen4_vfmig *vfmig = vf_info->mig_priv;
+	struct adf_mstate_mgr *mstate_mgr = vfmig->mstate_mgr;
+	struct adf_mstate_sect_h *subsec, *l2_subsec;
+	struct adf_mstate_mgr sub_sects_mgr;
+	struct {
+		char *id;
+		int (*action)(struct adf_mstate_mgr *sub_mgr, u8 *buf, u32 size, void *opa);
+		struct adf_mstate_vreginfo info;
+	} setups[] = {
+		{ADF_MSTATE_GEN_CAP_IDS, adf_mstate_capmask_superset,
+		{&hw_data->accel_capabilities_mask, sizeof(hw_data->accel_capabilities_mask)}},
+		{ADF_MSTATE_GEN_SVCMAP_IDS, adf_mstate_capmask_equal,
+		{&hw_data->ring_to_svc_map, sizeof(hw_data->ring_to_svc_map)}},
+		{ADF_MSTATE_GEN_EXTDC_IDS, adf_mstate_capmask_superset,
+		{&hw_data->extended_dc_capabilities, sizeof(hw_data->extended_dc_capabilities)}},
+	};
+	int i;
+
+	subsec = adf_mstate_sect_lookup(mstate_mgr, ADF_MSTATE_CONFIG_IDS, NULL, NULL);
+	if (!subsec) {
+		dev_err(&GET_DEV(accel_dev), "Failed to load sec %s\n",
+			ADF_MSTATE_CONFIG_IDS);
+		return -EINVAL;
+	}
+
+	adf_mstate_mgr_init_from_psect(&sub_sects_mgr, subsec);
+	for (i = 0; i < ARRAY_SIZE(setups); i++) {
+		l2_subsec = adf_mstate_sect_lookup(&sub_sects_mgr, setups[i].id,
+						   setups[i].action, &setups[i].info);
+		if (!l2_subsec) {
+			dev_err(&GET_DEV(accel_dev), "Failed to load sec %s\n",
+				setups[i].id);
+			return -EINVAL;
+		}
+	}
+
+	return 0;
+}
+
+static int adf_gen4_vfmig_save_etr_regs(struct adf_mstate_mgr *subs, u8 *state,
+					u32 size, void *opa)
+{
+	struct adf_vf_bank_info *vf_bank_info = opa;
+	struct adf_accel_dev *accel_dev = vf_bank_info->accel_dev;
+	struct adf_hw_device_data *hw_data = accel_dev->hw_device;
+	u32 pf_bank_nr;
+	int ret;
+
+	pf_bank_nr = vf_bank_info->bank_nr;
+	pf_bank_nr += vf_bank_info->vf_nr * hw_data->num_banks_per_vf;
+
+	ret = hw_data->bank_state_save(accel_dev, pf_bank_nr,
+				       (struct bank_state *)state);
+	if (ret) {
+		dev_err(&GET_DEV(accel_dev),
+			"Failed to save regs for vf%d bank%d\n",
+			vf_bank_info->vf_nr, vf_bank_info->bank_nr);
+		return ret;
+	}
+
+	return sizeof(struct bank_state);
+}
+
+static int adf_gen4_vfmig_save_etr_bank(struct adf_accel_dev *accel_dev,
+					u32 vf_nr, u32 bank_nr,
+					struct adf_mstate_mgr *mstate_mgr)
+{
+	struct adf_mstate_sect_h *subsec, *l2_subsec;
+	struct adf_vf_bank_info vf_bank_info;
+	struct adf_mstate_mgr sub_sects_mgr;
+	char bank_ids[ADF_MSTATE_ID_LEN];
+
+	snprintf(bank_ids, sizeof(bank_ids), ADF_MSTATE_BANK_IDX_IDS "%x", bank_nr);
+
+	subsec = adf_mstate_sect_add(mstate_mgr, bank_ids, NULL, NULL);
+	if (!subsec) {
+		dev_err(&GET_DEV(accel_dev),
+			"Failed to add sec %s for vf%d bank%d\n",
+			ADF_MSTATE_BANK_IDX_IDS, vf_nr, bank_nr);
+		return -EINVAL;
+	}
+
+	adf_mstate_mgr_init_from_parent(&sub_sects_mgr, mstate_mgr);
+	vf_bank_info.accel_dev = accel_dev;
+	vf_bank_info.vf_nr = vf_nr;
+	vf_bank_info.bank_nr = bank_nr;
+	l2_subsec = adf_mstate_sect_add(&sub_sects_mgr, ADF_MSTATE_ETR_REGS_IDS,
+					adf_gen4_vfmig_save_etr_regs,
+					&vf_bank_info);
+	if (!l2_subsec) {
+		dev_err(&GET_DEV(accel_dev),
+			"Failed to add sec %s for vf%d bank%d\n",
+			ADF_MSTATE_ETR_REGS_IDS, vf_nr, bank_nr);
+		return -EINVAL;
+	}
+	adf_mstate_sect_update(mstate_mgr, &sub_sects_mgr, subsec);
+
+	return 0;
+}
+
+static int adf_gen4_vfmig_save_etr(struct adf_accel_dev *accel_dev, u32 vf_nr)
+{
+	struct adf_accel_vf_info *vf_info = &accel_dev->pf.vf_info[vf_nr];
+	struct adf_hw_device_data *hw_data = accel_dev->hw_device;
+	struct adf_gen4_vfmig *vfmig = vf_info->mig_priv;
+	struct adf_mstate_mgr *mstate_mgr = vfmig->mstate_mgr;
+	struct adf_mstate_mgr sub_sects_mgr;
+	struct adf_mstate_sect_h *subsec;
+	int ret, i;
+
+	subsec = adf_mstate_sect_add(mstate_mgr, ADF_MSTATE_ETRB_IDS, NULL, NULL);
+	if (!subsec) {
+		dev_err(&GET_DEV(accel_dev), "Failed to add sec %s\n",
+			ADF_MSTATE_ETRB_IDS);
+		return -EINVAL;
+	}
+
+	adf_mstate_mgr_init_from_parent(&sub_sects_mgr, mstate_mgr);
+	for (i = 0; i < hw_data->num_banks_per_vf; i++) {
+		ret = adf_gen4_vfmig_save_etr_bank(accel_dev, vf_nr, i,
+						   &sub_sects_mgr);
+		if (ret)
+			return ret;
+	}
+	adf_mstate_sect_update(mstate_mgr, &sub_sects_mgr, subsec);
+
+	return 0;
+}
+
+static int adf_gen4_vfmig_save_misc(struct adf_accel_dev *accel_dev, u32 vf_nr)
+{
+	struct adf_accel_vf_info *vf_info = &accel_dev->pf.vf_info[vf_nr];
+	struct adf_gen4_vfmig *vfmig = vf_info->mig_priv;
+	struct adf_mstate_mgr *mstate_mgr = vfmig->mstate_mgr;
+	void __iomem *csr = adf_get_pmisc_base(accel_dev);
+	struct adf_mstate_sect_h *subsec, *l2_subsec;
+	struct adf_mstate_mgr sub_sects_mgr;
+	struct {
+		char *id;
+		u64 offset;
+	} misc_states[] = {
+		{ADF_MSTATE_VINTSRC_IDS, ADF_GEN4_VINTSOU_OFFSET(vf_nr)},
+		{ADF_MSTATE_VINTMSK_IDS, ADF_GEN4_VINTMSK_OFFSET(vf_nr)},
+		{ADF_MSTATE_VINTSRC_PF2VM_IDS, ADF_GEN4_VINTSOUPF2VM_OFFSET(vf_nr)},
+		{ADF_MSTATE_VINTMSK_PF2VM_IDS, ADF_GEN4_VINTMSKPF2VM_OFFSET(vf_nr)},
+		{ADF_MSTATE_PF2VM_IDS, ADF_GEN4_PF2VM_OFFSET(vf_nr)},
+		{ADF_MSTATE_VM2PF_IDS, ADF_GEN4_VM2PF_OFFSET(vf_nr)},
+	};
+	ktime_t time_exp;
+	int i;
+
+	subsec = adf_mstate_sect_add(mstate_mgr, ADF_MSTATE_MISCB_IDS, NULL, NULL);
+	if (!subsec) {
+		dev_err(&GET_DEV(accel_dev), "Failed to add sec %s\n",
+			ADF_MSTATE_MISCB_IDS);
+		return -EINVAL;
+	}
+
+	time_exp = ktime_add_us(ktime_get(), ADF_GEN4_PFVF_RSP_TIMEOUT_US);
+	while (!mutex_trylock(&vf_info->pfvf_mig_lock)) {
+		if (ktime_after(ktime_get(), time_exp)) {
+			dev_err(&GET_DEV(accel_dev), "Failed to get pfvf mig lock\n");
+			return -ETIMEDOUT;
+		}
+		usleep_range(500, 1000);
+	}
+
+	adf_mstate_mgr_init_from_parent(&sub_sects_mgr, mstate_mgr);
+	for (i = 0; i < ARRAY_SIZE(misc_states); i++) {
+		struct adf_mstate_vreginfo info;
+		u32 regv;
+
+		info.addr = &regv;
+		info.size = sizeof(regv);
+		regv = ADF_CSR_RD(csr, misc_states[i].offset);
+
+		l2_subsec = adf_mstate_sect_add_vreg(&sub_sects_mgr,
+						     misc_states[i].id,
+						     &info);
+		if (!l2_subsec) {
+			dev_err(&GET_DEV(accel_dev), "Failed to add sec %s\n",
+				misc_states[i].id);
+			mutex_unlock(&vf_info->pfvf_mig_lock);
+			return -EINVAL;
+		}
+	}
+
+	mutex_unlock(&vf_info->pfvf_mig_lock);
+	adf_mstate_sect_update(mstate_mgr, &sub_sects_mgr, subsec);
+
+	return 0;
+}
+
+static int adf_gen4_vfmig_save_generic(struct adf_accel_dev *accel_dev, u32 vf_nr)
+{
+	struct adf_accel_vf_info *vf_info = &accel_dev->pf.vf_info[vf_nr];
+	struct adf_gen4_vfmig *vfmig = vf_info->mig_priv;
+	struct adf_mstate_mgr *mstate_mgr = vfmig->mstate_mgr;
+	struct adf_mstate_mgr sub_sects_mgr;
+	struct adf_mstate_sect_h *subsec, *l2_subsec;
+	struct mig_user_sla src_slas[RL_RP_CNT_PER_LEAF_MAX] = { };
+	u32 src_sla_cnt;
+	struct {
+		char *id;
+		struct adf_mstate_vreginfo info;
+	} gen_states[] = {
+		{ADF_MSTATE_IOV_INIT_IDS,
+		{&vf_info->init, sizeof(vf_info->init)}},
+		{ADF_MSTATE_COMPAT_VER_IDS,
+		{&vf_info->vf_compat_ver, sizeof(vf_info->vf_compat_ver)}},
+		{ADF_MSTATE_SLA_IDS, {src_slas, 0}},
+	};
+	int i;
+
+	subsec = adf_mstate_sect_add(mstate_mgr, ADF_MSTATE_GEN_IDS, NULL, NULL);
+	if (!subsec) {
+		dev_err(&GET_DEV(accel_dev), "Failed to add sec %s\n",
+			ADF_MSTATE_GEN_IDS);
+		return -EINVAL;
+	}
+
+	adf_mstate_mgr_init_from_parent(&sub_sects_mgr, mstate_mgr);
+	for (i = 0; i < ARRAY_SIZE(gen_states); i++) {
+		if (gen_states[i].info.addr == src_slas) {
+			src_sla_cnt = adf_gen4_vfmig_get_slas(accel_dev, vf_nr, src_slas);
+			gen_states[i].info.size = src_sla_cnt * sizeof(struct mig_user_sla);
+		}
+
+		l2_subsec = adf_mstate_sect_add_vreg(&sub_sects_mgr,
+						     gen_states[i].id,
+						     &gen_states[i].info);
+		if (!l2_subsec) {
+			dev_err(&GET_DEV(accel_dev), "Failed to add sec %s\n",
+				gen_states[i].id);
+			return -EINVAL;
+		}
+	}
+	adf_mstate_sect_update(mstate_mgr, &sub_sects_mgr, subsec);
+
+	return 0;
+}
+
+static int adf_gen4_vfmig_save_config(struct adf_accel_dev *accel_dev, u32 vf_nr)
+{
+	struct adf_accel_vf_info *vf_info = &accel_dev->pf.vf_info[vf_nr];
+	struct adf_hw_device_data *hw_data = accel_dev->hw_device;
+	struct adf_gen4_vfmig *vfmig = vf_info->mig_priv;
+	struct adf_mstate_mgr *mstate_mgr = vfmig->mstate_mgr;
+	struct adf_mstate_mgr sub_sects_mgr;
+	struct adf_mstate_sect_h *subsec, *l2_subsec;
+	struct {
+		char *id;
+		struct adf_mstate_vreginfo info;
+	} setups[] = {
+		{ADF_MSTATE_GEN_CAP_IDS,
+		{&hw_data->accel_capabilities_mask, sizeof(hw_data->accel_capabilities_mask)}},
+		{ADF_MSTATE_GEN_SVCMAP_IDS,
+		{&hw_data->ring_to_svc_map, sizeof(hw_data->ring_to_svc_map)}},
+		{ADF_MSTATE_GEN_EXTDC_IDS,
+		{&hw_data->extended_dc_capabilities, sizeof(hw_data->extended_dc_capabilities)}},
+	};
+	int i;
+
+	subsec = adf_mstate_sect_add(mstate_mgr, ADF_MSTATE_CONFIG_IDS, NULL, NULL);
+	if (!subsec) {
+		dev_err(&GET_DEV(accel_dev), "Failed to add sec %s\n",
+			ADF_MSTATE_CONFIG_IDS);
+		return -EINVAL;
+	}
+
+	adf_mstate_mgr_init_from_parent(&sub_sects_mgr, mstate_mgr);
+	for (i = 0; i < ARRAY_SIZE(setups); i++) {
+		l2_subsec = adf_mstate_sect_add_vreg(&sub_sects_mgr, setups[i].id,
+						     &setups[i].info);
+		if (!l2_subsec) {
+			dev_err(&GET_DEV(accel_dev), "Failed to add sec %s\n",
+				setups[i].id);
+			return -EINVAL;
+		}
+	}
+	adf_mstate_sect_update(mstate_mgr, &sub_sects_mgr, subsec);
+
+	return 0;
+}
+
+static int adf_gen4_vfmig_save_state(struct qat_mig_dev *mdev)
+{
+	struct adf_accel_dev *accel_dev = mdev->parent_accel_dev;
+	struct adf_accel_vf_info *vf_info;
+	struct adf_gen4_vfmig *vfmig;
+	u32 vf_nr = mdev->vf_id;
+	int ret;
+
+	vf_info = &accel_dev->pf.vf_info[vf_nr];
+	vfmig = vf_info->mig_priv;
+
+	ret = adf_gen4_vfmig_save_setup(mdev);
+	if (ret) {
+		dev_err(&GET_DEV(accel_dev),
+			"Failed to save setup for vf_nr %d\n", vf_nr);
+		return ret;
+	}
+
+	adf_mstate_mgr_init(vfmig->mstate_mgr, mdev->state + mdev->setup_size,
+			    mdev->state_size - mdev->setup_size);
+	if (!adf_mstate_preamble_add(vfmig->mstate_mgr))
+		return -EINVAL;
+
+	ret = adf_gen4_vfmig_save_generic(accel_dev, vf_nr);
+	if (ret) {
+		dev_err(&GET_DEV(accel_dev),
+			"Failed to save generic state for vf_nr %d\n", vf_nr);
+		return ret;
+	}
+
+	ret = adf_gen4_vfmig_save_misc(accel_dev, vf_nr);
+	if (ret) {
+		dev_err(&GET_DEV(accel_dev),
+			"Failed to save misc bar state for vf_nr %d\n", vf_nr);
+		return ret;
+	}
+
+	ret = adf_gen4_vfmig_save_etr(accel_dev, vf_nr);
+	if (ret) {
+		dev_err(&GET_DEV(accel_dev),
+			"Failed to save etr bar state for vf_nr %d\n", vf_nr);
+		return ret;
+	}
+
+	adf_mstate_preamble_update(vfmig->mstate_mgr);
+
+	return 0;
+}
+
+static int adf_gen4_vfmig_load_state(struct qat_mig_dev *mdev)
+{
+	struct adf_accel_dev *accel_dev = mdev->parent_accel_dev;
+	struct adf_accel_vf_info *vf_info;
+	struct adf_gen4_vfmig *vfmig;
+	u32 vf_nr = mdev->vf_id;
+	int ret;
+
+	vf_info = &accel_dev->pf.vf_info[vf_nr];
+	vfmig = vf_info->mig_priv;
+
+	ret = adf_gen4_vfmig_load_setup(mdev, mdev->state_size);
+	if (ret) {
+		dev_err(&GET_DEV(accel_dev), "Failed to load setup for vf_nr %d\n",
+			vf_nr);
+		return ret;
+	}
+
+	ret = adf_mstate_mgr_init_from_remote(vfmig->mstate_mgr,
+					      mdev->state + mdev->remote_setup_size,
+					      mdev->state_size - mdev->remote_setup_size,
+					      NULL, NULL);
+	if (ret) {
+		dev_err(&GET_DEV(accel_dev), "Invalid state for vf_nr %d\n",
+			vf_nr);
+		return ret;
+	}
+
+	ret = adf_gen4_vfmig_load_generic(accel_dev, vf_nr);
+	if (ret) {
+		dev_err(&GET_DEV(accel_dev),
+			"Failed to load general state for vf_nr %d\n", vf_nr);
+		return ret;
+	}
+
+	ret = adf_gen4_vfmig_load_misc(accel_dev, vf_nr);
+	if (ret) {
+		dev_err(&GET_DEV(accel_dev),
+			"Failed to load misc bar state for vf_nr %d\n", vf_nr);
+		return ret;
+	}
+
+	ret = adf_gen4_vfmig_load_etr(accel_dev, vf_nr);
+	if (ret) {
+		dev_err(&GET_DEV(accel_dev),
+			"Failed to load etr bar state for vf_nr %d\n", vf_nr);
+		return ret;
+	}
+
+	return 0;
+}
+
+static int adf_gen4_vfmig_save_setup(struct qat_mig_dev *mdev)
+{
+	struct adf_accel_dev *accel_dev = mdev->parent_accel_dev;
+	struct adf_accel_vf_info *vf_info;
+	struct adf_gen4_vfmig *vfmig;
+	u32 vf_nr = mdev->vf_id;
+	int ret;
+
+	vf_info = &accel_dev->pf.vf_info[vf_nr];
+	vfmig = vf_info->mig_priv;
+
+	if (mdev->setup_size)
+		return 0;
+
+	adf_mstate_mgr_init(vfmig->mstate_mgr, mdev->state, mdev->state_size);
+	if (!adf_mstate_preamble_add(vfmig->mstate_mgr))
+		return -EINVAL;
+
+	ret = adf_gen4_vfmig_save_config(accel_dev, mdev->vf_id);
+	if (ret)
+		return ret;
+
+	adf_mstate_preamble_update(vfmig->mstate_mgr);
+	mdev->setup_size = adf_mstate_state_size(vfmig->mstate_mgr);
+
+	return 0;
+}
+
+static int adf_gen4_vfmig_load_setup(struct qat_mig_dev *mdev, int len)
+{
+	struct adf_accel_dev *accel_dev = mdev->parent_accel_dev;
+	struct adf_accel_vf_info *vf_info;
+	struct adf_gen4_vfmig *vfmig;
+	u32 vf_nr = mdev->vf_id;
+	u32 setup_size;
+	int ret;
+
+	vf_info = &accel_dev->pf.vf_info[vf_nr];
+	vfmig = vf_info->mig_priv;
+
+	if (mdev->remote_setup_size)
+		return 0;
+
+	if (len < sizeof(struct adf_mstate_preh))
+		return -EAGAIN;
+
+	adf_mstate_mgr_init(vfmig->mstate_mgr, mdev->state, mdev->state_size);
+	setup_size = adf_mstate_state_size_from_remote(vfmig->mstate_mgr);
+	if (setup_size > mdev->state_size)
+		return -EINVAL;
+
+	if (len < setup_size)
+		return -EAGAIN;
+
+	ret = adf_mstate_mgr_init_from_remote(vfmig->mstate_mgr, mdev->state,
+					      setup_size, NULL, NULL);
+	if (ret) {
+		dev_err(&GET_DEV(accel_dev), "Invalid setup for vf_nr %d\n",
+			vf_nr);
+		return ret;
+	}
+
+	mdev->remote_setup_size = setup_size;
+
+	ret = adf_gen4_vfmig_load_config(accel_dev, vf_nr);
+	if (ret) {
+		dev_err(&GET_DEV(accel_dev),
+			"Failed to load config for vf_nr %d\n", vf_nr);
+		return ret;
+	}
+
+	return 0;
+}
+
+void adf_gen4_init_vf_mig_ops(struct qat_migdev_ops *vfmig_ops)
+{
+	vfmig_ops->init = adf_gen4_vfmig_init_device;
+	vfmig_ops->cleanup = adf_gen4_vfmig_cleanup_device;
+	vfmig_ops->reset = adf_gen4_vfmig_reset_device;
+	vfmig_ops->open = adf_gen4_vfmig_open_device;
+	vfmig_ops->close = adf_gen4_vfmig_close_device;
+	vfmig_ops->suspend = adf_gen4_vfmig_suspend_device;
+	vfmig_ops->resume = adf_gen4_vfmig_resume_device;
+	vfmig_ops->save_state = adf_gen4_vfmig_save_state;
+	vfmig_ops->load_state = adf_gen4_vfmig_load_state;
+	vfmig_ops->load_setup = adf_gen4_vfmig_load_setup;
+	vfmig_ops->save_setup = adf_gen4_vfmig_save_setup;
+}
+EXPORT_SYMBOL_GPL(adf_gen4_init_vf_mig_ops);
diff --git a/drivers/crypto/intel/qat/qat_common/adf_gen4_vf_mig.h b/drivers/crypto/intel/qat/qat_common/adf_gen4_vf_mig.h
new file mode 100644
index 000000000000..72216d078ee1
--- /dev/null
+++ b/drivers/crypto/intel/qat/qat_common/adf_gen4_vf_mig.h
@@ -0,0 +1,10 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/* Copyright(c) 2024 Intel Corporation */
+#ifndef ADF_GEN4_VF_MIG_H_
+#define ADF_GEN4_VF_MIG_H_
+
+#include "adf_accel_devices.h"
+
+void adf_gen4_init_vf_mig_ops(struct qat_migdev_ops *vfmig_ops);
+
+#endif
diff --git a/drivers/crypto/intel/qat/qat_common/adf_mstate_mgr.c b/drivers/crypto/intel/qat/qat_common/adf_mstate_mgr.c
new file mode 100644
index 000000000000..41cc763a74aa
--- /dev/null
+++ b/drivers/crypto/intel/qat/qat_common/adf_mstate_mgr.c
@@ -0,0 +1,318 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/* Copyright(c) 2024 Intel Corporation */
+
+#include <linux/slab.h>
+#include <linux/types.h>
+#include "adf_mstate_mgr.h"
+
+#define ADF_MSTATE_MAGIC	0xADF5CAEA
+#define ADF_MSTATE_VERSION	0x1
+
+struct adf_mstate_sect_h {
+	u8 id[ADF_MSTATE_ID_LEN];
+	u32 size;
+	u32 sub_sects;
+	u8 state[];
+};
+
+u32 adf_mstate_state_size(struct adf_mstate_mgr *mgr)
+{
+	return mgr->state - mgr->buf;
+}
+
+static inline u32 adf_mstate_avail_room(struct adf_mstate_mgr *mgr)
+{
+	return mgr->buf + mgr->size - mgr->state;
+}
+
+void adf_mstate_mgr_init(struct adf_mstate_mgr *mgr, u8 *buf, u32 size)
+{
+	mgr->buf = buf;
+	mgr->state = buf;
+	mgr->size = size;
+	mgr->n_sects = 0;
+};
+
+struct adf_mstate_mgr *adf_mstate_mgr_new(u8 *buf, u32 size)
+{
+	struct adf_mstate_mgr *mgr;
+
+	mgr = kzalloc(sizeof(*mgr), GFP_KERNEL);
+	if (!mgr)
+		return NULL;
+
+	adf_mstate_mgr_init(mgr, buf, size);
+
+	return mgr;
+}
+
+void adf_mstate_mgr_destroy(struct adf_mstate_mgr *mgr)
+{
+	kfree(mgr);
+}
+
+void adf_mstate_mgr_init_from_parent(struct adf_mstate_mgr *mgr,
+				     struct adf_mstate_mgr *p_mgr)
+{
+	adf_mstate_mgr_init(mgr, p_mgr->state,
+			    p_mgr->size - adf_mstate_state_size(p_mgr));
+}
+
+void adf_mstate_mgr_init_from_psect(struct adf_mstate_mgr *mgr,
+				    struct adf_mstate_sect_h *p_sect)
+{
+	adf_mstate_mgr_init(mgr, p_sect->state, p_sect->size);
+	mgr->n_sects = p_sect->sub_sects;
+}
+
+static void adf_mstate_preamble_init(struct adf_mstate_preh *preamble)
+{
+	preamble->magic = ADF_MSTATE_MAGIC;
+	preamble->version = ADF_MSTATE_VERSION;
+	preamble->preh_len = sizeof(*preamble);
+	preamble->size = 0;
+	preamble->n_sects = 0;
+}
+
+/* default preambles checker */
+static int adf_mstate_preamble_def_checker(struct adf_mstate_preh *preamble,
+					   void *opaque)
+{
+	struct adf_mstate_mgr *mgr = opaque;
+
+	if (preamble->magic != ADF_MSTATE_MAGIC ||
+	    preamble->version > ADF_MSTATE_VERSION ||
+	    preamble->preh_len > mgr->size) {
+		pr_debug("QAT: LM - Invalid state (magic=%#x, version=%#x, hlen=%u), state_size=%u\n",
+			 preamble->magic, preamble->version, preamble->preh_len,
+			 mgr->size);
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+struct adf_mstate_preh *adf_mstate_preamble_add(struct adf_mstate_mgr *mgr)
+{
+	struct adf_mstate_preh *pre = (struct adf_mstate_preh *)mgr->buf;
+
+	if (adf_mstate_avail_room(mgr) < sizeof(*pre)) {
+		pr_err("QAT: LM - Not enough space for preamble\n");
+		return NULL;
+	}
+
+	adf_mstate_preamble_init(pre);
+	mgr->state += pre->preh_len;
+
+	return pre;
+}
+
+int adf_mstate_preamble_update(struct adf_mstate_mgr *mgr)
+{
+	struct adf_mstate_preh *preamble = (struct adf_mstate_preh *)mgr->buf;
+
+	preamble->size = adf_mstate_state_size(mgr) - preamble->preh_len;
+	preamble->n_sects = mgr->n_sects;
+
+	return 0;
+}
+
+static void adf_mstate_dump_sect(struct adf_mstate_sect_h *sect,
+				 const char *prefix)
+{
+	pr_debug("QAT: LM - %s QAT state section %s\n", prefix, sect->id);
+	print_hex_dump_debug("h-", DUMP_PREFIX_OFFSET, 16, 2, sect,
+			     sizeof(*sect), true);
+	print_hex_dump_debug("s-", DUMP_PREFIX_OFFSET, 16, 2, sect->state,
+			     sect->size, true);
+}
+
+static inline void __adf_mstate_sect_update(struct adf_mstate_mgr *mgr,
+					    struct adf_mstate_sect_h *sect,
+					    u32 size,
+					    u32 n_subsects)
+{
+	sect->size += size;
+	sect->sub_sects += n_subsects;
+	mgr->n_sects++;
+	mgr->state += sect->size;
+
+	adf_mstate_dump_sect(sect, "Add");
+}
+
+void adf_mstate_sect_update(struct adf_mstate_mgr *p_mgr,
+			    struct adf_mstate_mgr *curr_mgr,
+			    struct adf_mstate_sect_h *sect)
+{
+	__adf_mstate_sect_update(p_mgr, sect, adf_mstate_state_size(curr_mgr),
+				 curr_mgr->n_sects);
+}
+
+static struct adf_mstate_sect_h *adf_mstate_sect_add_header(struct adf_mstate_mgr *mgr,
+							    const char *id)
+{
+	struct adf_mstate_sect_h *sect = (struct adf_mstate_sect_h *)(mgr->state);
+
+	if (adf_mstate_avail_room(mgr) < sizeof(*sect)) {
+		pr_debug("QAT: LM - Not enough space for header of QAT state sect %s\n", id);
+		return NULL;
+	}
+
+	strscpy(sect->id, id, sizeof(sect->id));
+	sect->size = 0;
+	sect->sub_sects = 0;
+	mgr->state += sizeof(*sect);
+
+	return sect;
+}
+
+struct adf_mstate_sect_h *adf_mstate_sect_add_vreg(struct adf_mstate_mgr *mgr,
+						   const char *id,
+						   struct adf_mstate_vreginfo *info)
+{
+	struct adf_mstate_sect_h *sect;
+
+	sect = adf_mstate_sect_add_header(mgr, id);
+	if (!sect)
+		return NULL;
+
+	if (adf_mstate_avail_room(mgr) < info->size) {
+		pr_debug("QAT: LM - Not enough space for QAT state sect %s, requires %u\n",
+			 id, info->size);
+		return NULL;
+	}
+
+	memcpy(sect->state, info->addr, info->size);
+	__adf_mstate_sect_update(mgr, sect, info->size, 0);
+
+	return sect;
+}
+
+struct adf_mstate_sect_h *adf_mstate_sect_add(struct adf_mstate_mgr *mgr,
+					      const char *id,
+					      adf_mstate_populate populate,
+					      void *opaque)
+{
+	struct adf_mstate_mgr sub_sects_mgr;
+	struct adf_mstate_sect_h *sect;
+	int avail_room, size;
+
+	sect = adf_mstate_sect_add_header(mgr, id);
+	if (!sect)
+		return NULL;
+
+	if (!populate)
+		return sect;
+
+	avail_room = adf_mstate_avail_room(mgr);
+	adf_mstate_mgr_init_from_parent(&sub_sects_mgr, mgr);
+
+	size = (*populate)(&sub_sects_mgr, sect->state, avail_room, opaque);
+	if (size < 0)
+		return NULL;
+
+	size += adf_mstate_state_size(&sub_sects_mgr);
+	if (avail_room < size) {
+		pr_debug("QAT: LM - Not enough space for QAT state sect %s, requires %u\n",
+			 id, size);
+		return NULL;
+	}
+	__adf_mstate_sect_update(mgr, sect, size, sub_sects_mgr.n_sects);
+
+	return sect;
+}
+
+static int adf_mstate_sect_validate(struct adf_mstate_mgr *mgr)
+{
+	struct adf_mstate_sect_h *start = (struct adf_mstate_sect_h *)mgr->state;
+	struct adf_mstate_sect_h *sect = start;
+	u64 end;
+	int i;
+
+	end = (uintptr_t)mgr->buf + mgr->size;
+	for (i = 0; i < mgr->n_sects; i++) {
+		uintptr_t s_start = (uintptr_t)sect->state;
+		uintptr_t s_end = s_start + sect->size;
+
+		if (s_end < s_start || s_end > end) {
+			pr_debug("QAT: LM - Corrupted state section (index=%u, size=%u) in state_mgr (size=%u, secs=%u)\n",
+				 i, sect->size, mgr->size, mgr->n_sects);
+			return -EINVAL;
+		}
+		sect = (struct adf_mstate_sect_h *)s_end;
+	}
+
+	pr_debug("QAT: LM - Scanned section (last child=%s, size=%lu) in state_mgr (size=%u, secs=%u)\n",
+		 start->id, sizeof(struct adf_mstate_sect_h) * (ulong)(sect - start),
+		 mgr->size, mgr->n_sects);
+
+	return 0;
+}
+
+u32 adf_mstate_state_size_from_remote(struct adf_mstate_mgr *mgr)
+{
+	struct adf_mstate_preh *preh = (struct adf_mstate_preh *)mgr->buf;
+
+	return preh->preh_len + preh->size;
+}
+
+int adf_mstate_mgr_init_from_remote(struct adf_mstate_mgr *mgr, u8 *buf, u32 size,
+				    adf_mstate_preamble_checker pre_checker,
+				    void *opaque)
+{
+	struct adf_mstate_preh *pre;
+	int ret;
+
+	adf_mstate_mgr_init(mgr, buf, size);
+	pre = (struct adf_mstate_preh *)(mgr->buf);
+
+	pr_debug("QAT: LM - Dump state preambles\n");
+	print_hex_dump_debug("", DUMP_PREFIX_OFFSET, 16, 2, pre, pre->preh_len, 0);
+
+	if (pre_checker)
+		ret = (*pre_checker)(pre, opaque);
+	else
+		ret = adf_mstate_preamble_def_checker(pre, mgr);
+	if (ret)
+		return ret;
+
+	mgr->state = mgr->buf + pre->preh_len;
+	mgr->n_sects = pre->n_sects;
+
+	return adf_mstate_sect_validate(mgr);
+}
+
+struct adf_mstate_sect_h *adf_mstate_sect_lookup(struct adf_mstate_mgr *mgr,
+						 const char *id,
+						 adf_mstate_action action,
+						 void *opaque)
+{
+	struct adf_mstate_sect_h *sect = (struct adf_mstate_sect_h *)mgr->state;
+	struct adf_mstate_mgr sub_sects_mgr;
+	int i, ret;
+
+	for (i = 0; i < mgr->n_sects; i++) {
+		if (!strncmp(sect->id, id, sizeof(sect->id)))
+			goto found;
+
+		sect = (struct adf_mstate_sect_h *)(sect->state + sect->size);
+	}
+
+	return NULL;
+
+found:
+	adf_mstate_dump_sect(sect, "Found");
+
+	adf_mstate_mgr_init_from_psect(&sub_sects_mgr, sect);
+	if (sect->sub_sects && adf_mstate_sect_validate(&sub_sects_mgr))
+		return NULL;
+
+	if (!action)
+		return sect;
+
+	ret = (*action)(&sub_sects_mgr, sect->state, sect->size, opaque);
+	if (ret)
+		return NULL;
+
+	return sect;
+}
diff --git a/drivers/crypto/intel/qat/qat_common/adf_mstate_mgr.h b/drivers/crypto/intel/qat/qat_common/adf_mstate_mgr.h
new file mode 100644
index 000000000000..81d263a596c5
--- /dev/null
+++ b/drivers/crypto/intel/qat/qat_common/adf_mstate_mgr.h
@@ -0,0 +1,89 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/* Copyright(c) 2024 Intel Corporation */
+
+#ifndef ADF_MSTATE_MGR_H
+#define ADF_MSTATE_MGR_H
+
+#define ADF_MSTATE_ID_LEN		8
+
+#define ADF_MSTATE_ETRB_IDS		"ETRBAR"
+#define ADF_MSTATE_MISCB_IDS		"MISCBAR"
+#define ADF_MSTATE_EXTB_IDS		"EXTBAR"
+#define ADF_MSTATE_GEN_IDS		"GENER"
+#define ADF_MSTATE_CONFIG_IDS		"CONFIG"
+#define ADF_MSTATE_SECTION_NUM		5
+
+#define ADF_MSTATE_BANK_IDX_IDS		"bnk"
+
+#define ADF_MSTATE_ETR_REGS_IDS		"mregs"
+#define ADF_MSTATE_VINTSRC_IDS		"visrc"
+#define ADF_MSTATE_VINTMSK_IDS		"vimsk"
+#define ADF_MSTATE_SLA_IDS		"sla"
+#define ADF_MSTATE_IOV_INIT_IDS		"iovinit"
+#define ADF_MSTATE_COMPAT_VER_IDS	"compver"
+#define ADF_MSTATE_GEN_CAP_IDS		"gencap"
+#define ADF_MSTATE_GEN_SVCMAP_IDS	"svcmap"
+#define ADF_MSTATE_GEN_EXTDC_IDS	"extdc"
+#define ADF_MSTATE_VINTSRC_PF2VM_IDS	"vispv"
+#define ADF_MSTATE_VINTMSK_PF2VM_IDS	"vimpv"
+#define ADF_MSTATE_VM2PF_IDS		"vm2pf"
+#define ADF_MSTATE_PF2VM_IDS		"pf2vm"
+
+struct adf_mstate_mgr {
+	u8 *buf;
+	u8 *state;
+	u32 size;
+	u32 n_sects;
+};
+
+struct adf_mstate_preh {
+	u32 magic;
+	u32 version;
+	u16 preh_len;
+	u16 n_sects;
+	u32 size;
+};
+
+struct adf_mstate_vreginfo {
+	void *addr;
+	u32 size;
+};
+
+struct adf_mstate_sect_h;
+
+typedef int (*adf_mstate_preamble_checker)(struct adf_mstate_preh *preamble, void *opa);
+typedef int (*adf_mstate_populate)(struct adf_mstate_mgr *sub_mgr, u8 *buf,
+				   u32 size, void *opa);
+typedef int (*adf_mstate_action)(struct adf_mstate_mgr *sub_mgr, u8 *buf, u32 size,
+				 void *opa);
+
+struct adf_mstate_mgr *adf_mstate_mgr_new(u8 *buf, u32 size);
+void adf_mstate_mgr_destroy(struct adf_mstate_mgr *mgr);
+void adf_mstate_mgr_init(struct adf_mstate_mgr *mgr, u8 *buf, u32 size);
+void adf_mstate_mgr_init_from_parent(struct adf_mstate_mgr *mgr,
+				     struct adf_mstate_mgr *p_mgr);
+void adf_mstate_mgr_init_from_psect(struct adf_mstate_mgr *mgr,
+				    struct adf_mstate_sect_h *p_sect);
+int adf_mstate_mgr_init_from_remote(struct adf_mstate_mgr *mgr,
+				    u8 *buf, u32 size,
+				    adf_mstate_preamble_checker checker,
+				    void *opaque);
+struct adf_mstate_preh *adf_mstate_preamble_add(struct adf_mstate_mgr *mgr);
+int adf_mstate_preamble_update(struct adf_mstate_mgr *mgr);
+u32 adf_mstate_state_size(struct adf_mstate_mgr *mgr);
+u32 adf_mstate_state_size_from_remote(struct adf_mstate_mgr *mgr);
+void adf_mstate_sect_update(struct adf_mstate_mgr *p_mgr,
+			    struct adf_mstate_mgr *curr_mgr,
+			    struct adf_mstate_sect_h *sect);
+struct adf_mstate_sect_h *adf_mstate_sect_add_vreg(struct adf_mstate_mgr *mgr,
+						   const char *id,
+						   struct adf_mstate_vreginfo *info);
+struct adf_mstate_sect_h *adf_mstate_sect_add(struct adf_mstate_mgr *mgr,
+					      const char *id,
+					      adf_mstate_populate populate,
+					      void *opaque);
+struct adf_mstate_sect_h *adf_mstate_sect_lookup(struct adf_mstate_mgr *mgr,
+						 const char *id,
+						 adf_mstate_action action,
+						 void *opaque);
+#endif
diff --git a/drivers/crypto/intel/qat/qat_common/adf_pfvf_pf_proto.c b/drivers/crypto/intel/qat/qat_common/adf_pfvf_pf_proto.c
index 9ab93fbfefde..b9b5e744a3f1 100644
--- a/drivers/crypto/intel/qat/qat_common/adf_pfvf_pf_proto.c
+++ b/drivers/crypto/intel/qat/qat_common/adf_pfvf_pf_proto.c
@@ -242,13 +242,7 @@ static int adf_handle_vf2pf_msg(struct adf_accel_dev *accel_dev, u8 vf_nr,
 			"VersionRequest received from VF%d (vers %d) to PF (vers %d)\n",
 			vf_nr, vf_compat_ver, ADF_PFVF_COMPAT_THIS_VERSION);
 
-		if (vf_compat_ver == 0)
-			compat = ADF_PF2VF_VF_INCOMPATIBLE;
-		else if (vf_compat_ver <= ADF_PFVF_COMPAT_THIS_VERSION)
-			compat = ADF_PF2VF_VF_COMPATIBLE;
-		else
-			compat = ADF_PF2VF_VF_COMPAT_UNKNOWN;
-
+		compat = adf_vf_compat_checker(vf_compat_ver);
 		vf_info->vf_compat_ver = vf_compat_ver;
 
 		resp->type = ADF_PF2VF_MSGTYPE_VERSION_RESP;
diff --git a/drivers/crypto/intel/qat/qat_common/adf_pfvf_utils.h b/drivers/crypto/intel/qat/qat_common/adf_pfvf_utils.h
index 2be048e2287b..1a044297d873 100644
--- a/drivers/crypto/intel/qat/qat_common/adf_pfvf_utils.h
+++ b/drivers/crypto/intel/qat/qat_common/adf_pfvf_utils.h
@@ -28,4 +28,15 @@ u32 adf_pfvf_csr_msg_of(struct adf_accel_dev *accel_dev, struct pfvf_message msg
 struct pfvf_message adf_pfvf_message_of(struct adf_accel_dev *accel_dev, u32 raw_msg,
 					const struct pfvf_csr_format *fmt);
 
+static inline u8 adf_vf_compat_checker(u8 vf_compat_ver)
+{
+	if (vf_compat_ver == 0)
+		return ADF_PF2VF_VF_INCOMPATIBLE;
+
+	if (vf_compat_ver <= ADF_PFVF_COMPAT_THIS_VERSION)
+		return ADF_PF2VF_VF_COMPATIBLE;
+
+	return ADF_PF2VF_VF_COMPAT_UNKNOWN;
+}
+
 #endif /* ADF_PFVF_UTILS_H */
diff --git a/drivers/crypto/intel/qat/qat_common/adf_rl.c b/drivers/crypto/intel/qat/qat_common/adf_rl.c
index d4f2db3c53d8..346ef8bee99d 100644
--- a/drivers/crypto/intel/qat/qat_common/adf_rl.c
+++ b/drivers/crypto/intel/qat/qat_common/adf_rl.c
@@ -183,14 +183,14 @@ static enum adf_cfg_service_type srv_to_cfg_svc_type(enum adf_base_services rl_s
 }
 
 /**
- * get_sla_arr_of_type() - Returns a pointer to SLA type specific array
+ * adf_rl_get_sla_arr_of_type() - Returns a pointer to SLA type specific array
  * @rl_data: pointer to ratelimiting data
  * @type: SLA type
  * @sla_arr: pointer to variable where requested pointer will be stored
  *
  * Return: Max number of elements allowed for the returned array
  */
-static u32 get_sla_arr_of_type(struct adf_rl *rl_data, enum rl_node_type type,
+u32 adf_rl_get_sla_arr_of_type(struct adf_rl *rl_data, enum rl_node_type type,
 			       struct rl_sla ***sla_arr)
 {
 	switch (type) {
@@ -778,7 +778,7 @@ static void clear_sla(struct adf_rl *rl_data, struct rl_sla *sla)
 		rp_in_use[sla->ring_pairs_ids[i]] = false;
 
 	update_budget(sla, old_cir, true);
-	get_sla_arr_of_type(rl_data, sla->type, &sla_type_arr);
+	adf_rl_get_sla_arr_of_type(rl_data, sla->type, &sla_type_arr);
 	assign_node_to_parent(rl_data->accel_dev, sla, true);
 	adf_rl_send_admin_delete_msg(rl_data->accel_dev, node_id, sla->type);
 	mark_rps_usage(sla, rl_data->rp_in_use, false);
@@ -875,7 +875,7 @@ static int add_update_sla(struct adf_accel_dev *accel_dev,
 
 	if (!is_update) {
 		mark_rps_usage(sla, rl_data->rp_in_use, true);
-		get_sla_arr_of_type(rl_data, sla->type, &sla_type_arr);
+		adf_rl_get_sla_arr_of_type(rl_data, sla->type, &sla_type_arr);
 		sla_type_arr[sla->node_id] = sla;
 		rl_data->sla[sla->sla_id] = sla;
 	}
@@ -1065,7 +1065,7 @@ void adf_rl_remove_sla_all(struct adf_accel_dev *accel_dev, bool incl_default)
 
 	/* Unregister and remove all SLAs */
 	for (j = RL_LEAF; j >= end_type; j--) {
-		max_id = get_sla_arr_of_type(rl_data, j, &sla_type_arr);
+		max_id = adf_rl_get_sla_arr_of_type(rl_data, j, &sla_type_arr);
 
 		for (i = 0; i < max_id; i++) {
 			if (!sla_type_arr[i])
@@ -1125,7 +1125,7 @@ int adf_rl_start(struct adf_accel_dev *accel_dev)
 	}
 
 	if ((fw_caps & RL_CAPABILITY_MASK) != RL_CAPABILITY_VALUE) {
-		dev_info(&GET_DEV(accel_dev), "not supported\n");
+		dev_info(&GET_DEV(accel_dev), "feature not supported by FW\n");
 		ret = -EOPNOTSUPP;
 		goto ret_free;
 	}
diff --git a/drivers/crypto/intel/qat/qat_common/adf_rl.h b/drivers/crypto/intel/qat/qat_common/adf_rl.h
index 269c6656fb90..bfe750ea0e83 100644
--- a/drivers/crypto/intel/qat/qat_common/adf_rl.h
+++ b/drivers/crypto/intel/qat/qat_common/adf_rl.h
@@ -151,6 +151,8 @@ struct rl_sla {
 	u16 ring_pairs_cnt;
 };
 
+u32 adf_rl_get_sla_arr_of_type(struct adf_rl *rl_data, enum rl_node_type type,
+			       struct rl_sla ***sla_arr);
 int adf_rl_add_sla(struct adf_accel_dev *accel_dev,
 		   struct adf_rl_sla_input_data *sla_in);
 int adf_rl_update_sla(struct adf_accel_dev *accel_dev,
diff --git a/drivers/crypto/intel/qat/qat_common/adf_sriov.c b/drivers/crypto/intel/qat/qat_common/adf_sriov.c
index 87a70c00c41e..8d645e7e04aa 100644
--- a/drivers/crypto/intel/qat/qat_common/adf_sriov.c
+++ b/drivers/crypto/intel/qat/qat_common/adf_sriov.c
@@ -26,10 +26,12 @@ static void adf_iov_send_resp(struct work_struct *work)
 	u32 vf_nr = vf_info->vf_nr;
 	bool ret;
 
+	mutex_lock(&vf_info->pfvf_mig_lock);
 	ret = adf_recv_and_handle_vf2pf_msg(accel_dev, vf_nr);
 	if (ret)
 		/* re-enable interrupt on PF from this VF */
 		adf_enable_vf2pf_interrupts(accel_dev, 1 << vf_nr);
+	mutex_unlock(&vf_info->pfvf_mig_lock);
 
 	kfree(pf2vf_resp);
 }
@@ -62,6 +64,7 @@ static int adf_enable_sriov(struct adf_accel_dev *accel_dev)
 		vf_info->vf_nr = i;
 
 		mutex_init(&vf_info->pf2vf_lock);
+		mutex_init(&vf_info->pfvf_mig_lock);
 		ratelimit_state_init(&vf_info->vf2pf_ratelimit,
 				     ADF_VF2PF_RATELIMIT_INTERVAL,
 				     ADF_VF2PF_RATELIMIT_BURST);
@@ -138,8 +141,10 @@ void adf_disable_sriov(struct adf_accel_dev *accel_dev)
 	if (hw_data->configure_iov_threads)
 		hw_data->configure_iov_threads(accel_dev, false);
 
-	for (i = 0, vf = accel_dev->pf.vf_info; i < totalvfs; i++, vf++)
+	for (i = 0, vf = accel_dev->pf.vf_info; i < totalvfs; i++, vf++) {
 		mutex_destroy(&vf->pf2vf_lock);
+		mutex_destroy(&vf->pfvf_mig_lock);
+	}
 
 	if (!test_bit(ADF_STATUS_RESTARTING, &accel_dev->status)) {
 		kfree(accel_dev->pf.vf_info);
diff --git a/drivers/crypto/intel/qat/qat_common/adf_telemetry.c b/drivers/crypto/intel/qat/qat_common/adf_telemetry.c
index 2ff714d11bd2..74fb0c2ed241 100644
--- a/drivers/crypto/intel/qat/qat_common/adf_telemetry.c
+++ b/drivers/crypto/intel/qat/qat_common/adf_telemetry.c
@@ -41,6 +41,20 @@ static int validate_tl_data(struct adf_tl_hw_data *tl_data)
 	return 0;
 }
 
+static int validate_tl_slice_counters(struct icp_qat_fw_init_admin_slice_cnt *slice_count,
+				      u8 max_slices_per_type)
+{
+	u8 *sl_counter = (u8 *)slice_count;
+	int i;
+
+	for (i = 0; i < ADF_TL_SL_CNT_COUNT; i++) {
+		if (sl_counter[i] > max_slices_per_type)
+			return -EINVAL;
+	}
+
+	return 0;
+}
+
 static int adf_tl_alloc_mem(struct adf_accel_dev *accel_dev)
 {
 	struct adf_tl_hw_data *tl_data = &GET_TL_DATA(accel_dev);
@@ -214,6 +228,13 @@ int adf_tl_run(struct adf_accel_dev *accel_dev, int state)
 		return ret;
 	}
 
+	ret = validate_tl_slice_counters(&telemetry->slice_cnt, tl_data->max_sl_cnt);
+	if (ret) {
+		dev_err(dev, "invalid value returned by FW\n");
+		adf_send_admin_tl_stop(accel_dev);
+		return ret;
+	}
+
 	telemetry->hbuffs = state;
 	atomic_set(&telemetry->state, state);
 
diff --git a/drivers/crypto/intel/qat/qat_common/adf_telemetry.h b/drivers/crypto/intel/qat/qat_common/adf_telemetry.h
index 9be81cd3b886..e54a406cc1b4 100644
--- a/drivers/crypto/intel/qat/qat_common/adf_telemetry.h
+++ b/drivers/crypto/intel/qat/qat_common/adf_telemetry.h
@@ -40,6 +40,7 @@ struct adf_tl_hw_data {
 	u8 num_dev_counters;
 	u8 num_rp_counters;
 	u8 max_rp;
+	u8 max_sl_cnt;
 };
 
 struct adf_telemetry {
diff --git a/drivers/crypto/intel/qat/qat_common/adf_transport.c b/drivers/crypto/intel/qat/qat_common/adf_transport.c
index 630d0483c4e0..1efdf46490f1 100644
--- a/drivers/crypto/intel/qat/qat_common/adf_transport.c
+++ b/drivers/crypto/intel/qat/qat_common/adf_transport.c
@@ -474,7 +474,6 @@ err:
 int adf_init_etr_data(struct adf_accel_dev *accel_dev)
 {
 	struct adf_etr_data *etr_data;
-	struct adf_hw_device_data *hw_data = accel_dev->hw_device;
 	void __iomem *csr_addr;
 	u32 size;
 	u32 num_banks = 0;
@@ -495,8 +494,7 @@ int adf_init_etr_data(struct adf_accel_dev *accel_dev)
 	}
 
 	accel_dev->transport = etr_data;
-	i = hw_data->get_etr_bar_id(hw_data);
-	csr_addr = accel_dev->accel_pci_dev.pci_bars[i].virt_addr;
+	csr_addr = adf_get_etr_base(accel_dev);
 
 	/* accel_dev->debugfs_dir should always be non-NULL here */
 	etr_data->debug = debugfs_create_dir("transport",
diff --git a/drivers/crypto/intel/qat/qat_common/qat_asym_algs.c b/drivers/crypto/intel/qat/qat_common/qat_asym_algs.c
index 4128200a9032..85c682e248fb 100644
--- a/drivers/crypto/intel/qat/qat_common/qat_asym_algs.c
+++ b/drivers/crypto/intel/qat/qat_common/qat_asym_algs.c
@@ -110,6 +110,8 @@ struct qat_dh_ctx {
 	unsigned int p_size;
 	bool g2;
 	struct qat_crypto_instance *inst;
+	struct crypto_kpp *ftfm;
+	bool fallback;
 } __packed __aligned(64);
 
 struct qat_asym_request {
@@ -381,6 +383,36 @@ unmap_src:
 	return ret;
 }
 
+static int qat_dh_generate_public_key(struct kpp_request *req)
+{
+	struct kpp_request *nreq = kpp_request_ctx(req);
+	struct crypto_kpp *tfm = crypto_kpp_reqtfm(req);
+	struct qat_dh_ctx *ctx = kpp_tfm_ctx(tfm);
+
+	if (ctx->fallback) {
+		memcpy(nreq, req, sizeof(*req));
+		kpp_request_set_tfm(nreq, ctx->ftfm);
+		return crypto_kpp_generate_public_key(nreq);
+	}
+
+	return qat_dh_compute_value(req);
+}
+
+static int qat_dh_compute_shared_secret(struct kpp_request *req)
+{
+	struct kpp_request *nreq = kpp_request_ctx(req);
+	struct crypto_kpp *tfm = crypto_kpp_reqtfm(req);
+	struct qat_dh_ctx *ctx = kpp_tfm_ctx(tfm);
+
+	if (ctx->fallback) {
+		memcpy(nreq, req, sizeof(*req));
+		kpp_request_set_tfm(nreq, ctx->ftfm);
+		return crypto_kpp_compute_shared_secret(nreq);
+	}
+
+	return qat_dh_compute_value(req);
+}
+
 static int qat_dh_check_params_length(unsigned int p_len)
 {
 	switch (p_len) {
@@ -398,9 +430,6 @@ static int qat_dh_set_params(struct qat_dh_ctx *ctx, struct dh *params)
 	struct qat_crypto_instance *inst = ctx->inst;
 	struct device *dev = &GET_DEV(inst->accel_dev);
 
-	if (qat_dh_check_params_length(params->p_size << 3))
-		return -EINVAL;
-
 	ctx->p_size = params->p_size;
 	ctx->p = dma_alloc_coherent(dev, ctx->p_size, &ctx->dma_p, GFP_KERNEL);
 	if (!ctx->p)
@@ -454,6 +483,13 @@ static int qat_dh_set_secret(struct crypto_kpp *tfm, const void *buf,
 	if (crypto_dh_decode_key(buf, len, &params) < 0)
 		return -EINVAL;
 
+	if (qat_dh_check_params_length(params.p_size << 3)) {
+		ctx->fallback = true;
+		return crypto_kpp_set_secret(ctx->ftfm, buf, len);
+	}
+
+	ctx->fallback = false;
+
 	/* Free old secret if any */
 	qat_dh_clear_ctx(dev, ctx);
 
@@ -481,6 +517,9 @@ static unsigned int qat_dh_max_size(struct crypto_kpp *tfm)
 {
 	struct qat_dh_ctx *ctx = kpp_tfm_ctx(tfm);
 
+	if (ctx->fallback)
+		return crypto_kpp_maxsize(ctx->ftfm);
+
 	return ctx->p_size;
 }
 
@@ -489,11 +528,22 @@ static int qat_dh_init_tfm(struct crypto_kpp *tfm)
 	struct qat_dh_ctx *ctx = kpp_tfm_ctx(tfm);
 	struct qat_crypto_instance *inst =
 			qat_crypto_get_instance_node(numa_node_id());
+	const char *alg = kpp_alg_name(tfm);
+	unsigned int reqsize;
 
 	if (!inst)
 		return -EINVAL;
 
-	kpp_set_reqsize(tfm, sizeof(struct qat_asym_request) + 64);
+	ctx->ftfm = crypto_alloc_kpp(alg, 0, CRYPTO_ALG_NEED_FALLBACK);
+	if (IS_ERR(ctx->ftfm))
+		return PTR_ERR(ctx->ftfm);
+
+	crypto_kpp_set_flags(ctx->ftfm, crypto_kpp_get_flags(tfm));
+
+	reqsize = max(sizeof(struct qat_asym_request) + 64,
+		      sizeof(struct kpp_request) + crypto_kpp_reqsize(ctx->ftfm));
+
+	kpp_set_reqsize(tfm, reqsize);
 
 	ctx->p_size = 0;
 	ctx->g2 = false;
@@ -506,6 +556,9 @@ static void qat_dh_exit_tfm(struct crypto_kpp *tfm)
 	struct qat_dh_ctx *ctx = kpp_tfm_ctx(tfm);
 	struct device *dev = &GET_DEV(ctx->inst->accel_dev);
 
+	if (ctx->ftfm)
+		crypto_free_kpp(ctx->ftfm);
+
 	qat_dh_clear_ctx(dev, ctx);
 	qat_crypto_put_instance(ctx->inst);
 }
@@ -1265,8 +1318,8 @@ static struct akcipher_alg rsa = {
 
 static struct kpp_alg dh = {
 	.set_secret = qat_dh_set_secret,
-	.generate_public_key = qat_dh_compute_value,
-	.compute_shared_secret = qat_dh_compute_value,
+	.generate_public_key = qat_dh_generate_public_key,
+	.compute_shared_secret = qat_dh_compute_shared_secret,
 	.max_size = qat_dh_max_size,
 	.init = qat_dh_init_tfm,
 	.exit = qat_dh_exit_tfm,
@@ -1276,6 +1329,7 @@ static struct kpp_alg dh = {
 		.cra_priority = 1000,
 		.cra_module = THIS_MODULE,
 		.cra_ctxsize = sizeof(struct qat_dh_ctx),
+		.cra_flags = CRYPTO_ALG_NEED_FALLBACK,
 	},
 };
 
diff --git a/drivers/crypto/intel/qat/qat_common/qat_bl.c b/drivers/crypto/intel/qat/qat_common/qat_bl.c
index 76baed0a76c0..338acf29c487 100644
--- a/drivers/crypto/intel/qat/qat_common/qat_bl.c
+++ b/drivers/crypto/intel/qat/qat_common/qat_bl.c
@@ -81,7 +81,8 @@ static int __qat_bl_sgl_to_bufl(struct adf_accel_dev *accel_dev,
 		if (unlikely(!bufl))
 			return -ENOMEM;
 	} else {
-		bufl = &buf->sgl_src.sgl_hdr;
+		bufl = container_of(&buf->sgl_src.sgl_hdr,
+				    struct qat_alg_buf_list, hdr);
 		memset(bufl, 0, sizeof(struct qat_alg_buf_list));
 		buf->sgl_src_valid = true;
 	}
@@ -139,7 +140,8 @@ static int __qat_bl_sgl_to_bufl(struct adf_accel_dev *accel_dev,
 			if (unlikely(!buflout))
 				goto err_in;
 		} else {
-			buflout = &buf->sgl_dst.sgl_hdr;
+			buflout = container_of(&buf->sgl_dst.sgl_hdr,
+					       struct qat_alg_buf_list, hdr);
 			memset(buflout, 0, sizeof(struct qat_alg_buf_list));
 			buf->sgl_dst_valid = true;
 		}
diff --git a/drivers/crypto/intel/qat/qat_common/qat_bl.h b/drivers/crypto/intel/qat/qat_common/qat_bl.h
index d87e4f35ac39..85bc32a9ec0e 100644
--- a/drivers/crypto/intel/qat/qat_common/qat_bl.h
+++ b/drivers/crypto/intel/qat/qat_common/qat_bl.h
@@ -15,14 +15,17 @@ struct qat_alg_buf {
 } __packed;
 
 struct qat_alg_buf_list {
-	u64 resrvd;
-	u32 num_bufs;
-	u32 num_mapped_bufs;
+	/* New members must be added within the __struct_group() macro below. */
+	__struct_group(qat_alg_buf_list_hdr, hdr, __packed,
+		u64 resrvd;
+		u32 num_bufs;
+		u32 num_mapped_bufs;
+	);
 	struct qat_alg_buf buffers[];
 } __packed;
 
 struct qat_alg_fixed_buf_list {
-	struct qat_alg_buf_list sgl_hdr;
+	struct qat_alg_buf_list_hdr sgl_hdr;
 	struct qat_alg_buf descriptors[QAT_MAX_BUFF_DESC];
 } __packed __aligned(64);
 
diff --git a/drivers/crypto/intel/qat/qat_common/qat_mig_dev.c b/drivers/crypto/intel/qat/qat_common/qat_mig_dev.c
new file mode 100644
index 000000000000..892c2283a50e
--- /dev/null
+++ b/drivers/crypto/intel/qat/qat_common/qat_mig_dev.c
@@ -0,0 +1,130 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/* Copyright(c) 2024 Intel Corporation */
+#include <linux/dev_printk.h>
+#include <linux/export.h>
+#include <linux/pci.h>
+#include <linux/types.h>
+#include <linux/qat/qat_mig_dev.h>
+#include "adf_accel_devices.h"
+#include "adf_common_drv.h"
+
+struct qat_mig_dev *qat_vfmig_create(struct pci_dev *pdev, int vf_id)
+{
+	struct adf_accel_dev *accel_dev;
+	struct qat_migdev_ops *ops;
+	struct qat_mig_dev *mdev;
+
+	accel_dev = adf_devmgr_pci_to_accel_dev(pdev);
+	if (!accel_dev)
+		return ERR_PTR(-ENODEV);
+
+	ops = GET_VFMIG_OPS(accel_dev);
+	if (!ops || !ops->init || !ops->cleanup || !ops->reset || !ops->open ||
+	    !ops->close || !ops->suspend || !ops->resume || !ops->save_state ||
+	    !ops->load_state || !ops->save_setup || !ops->load_setup)
+		return ERR_PTR(-EINVAL);
+
+	mdev = kmalloc(sizeof(*mdev), GFP_KERNEL);
+	if (!mdev)
+		return ERR_PTR(-ENOMEM);
+
+	mdev->vf_id = vf_id;
+	mdev->parent_accel_dev = accel_dev;
+
+	return mdev;
+}
+EXPORT_SYMBOL_GPL(qat_vfmig_create);
+
+int qat_vfmig_init(struct qat_mig_dev *mdev)
+{
+	struct adf_accel_dev *accel_dev = mdev->parent_accel_dev;
+
+	return GET_VFMIG_OPS(accel_dev)->init(mdev);
+}
+EXPORT_SYMBOL_GPL(qat_vfmig_init);
+
+void qat_vfmig_cleanup(struct qat_mig_dev *mdev)
+{
+	struct adf_accel_dev *accel_dev = mdev->parent_accel_dev;
+
+	return GET_VFMIG_OPS(accel_dev)->cleanup(mdev);
+}
+EXPORT_SYMBOL_GPL(qat_vfmig_cleanup);
+
+void qat_vfmig_reset(struct qat_mig_dev *mdev)
+{
+	struct adf_accel_dev *accel_dev = mdev->parent_accel_dev;
+
+	return GET_VFMIG_OPS(accel_dev)->reset(mdev);
+}
+EXPORT_SYMBOL_GPL(qat_vfmig_reset);
+
+int qat_vfmig_open(struct qat_mig_dev *mdev)
+{
+	struct adf_accel_dev *accel_dev = mdev->parent_accel_dev;
+
+	return GET_VFMIG_OPS(accel_dev)->open(mdev);
+}
+EXPORT_SYMBOL_GPL(qat_vfmig_open);
+
+void qat_vfmig_close(struct qat_mig_dev *mdev)
+{
+	struct adf_accel_dev *accel_dev = mdev->parent_accel_dev;
+
+	GET_VFMIG_OPS(accel_dev)->close(mdev);
+}
+EXPORT_SYMBOL_GPL(qat_vfmig_close);
+
+int qat_vfmig_suspend(struct qat_mig_dev *mdev)
+{
+	struct adf_accel_dev *accel_dev = mdev->parent_accel_dev;
+
+	return GET_VFMIG_OPS(accel_dev)->suspend(mdev);
+}
+EXPORT_SYMBOL_GPL(qat_vfmig_suspend);
+
+int qat_vfmig_resume(struct qat_mig_dev *mdev)
+{
+	struct adf_accel_dev *accel_dev = mdev->parent_accel_dev;
+
+	return GET_VFMIG_OPS(accel_dev)->resume(mdev);
+}
+EXPORT_SYMBOL_GPL(qat_vfmig_resume);
+
+int qat_vfmig_save_state(struct qat_mig_dev *mdev)
+{
+	struct adf_accel_dev *accel_dev = mdev->parent_accel_dev;
+
+	return GET_VFMIG_OPS(accel_dev)->save_state(mdev);
+}
+EXPORT_SYMBOL_GPL(qat_vfmig_save_state);
+
+int qat_vfmig_save_setup(struct qat_mig_dev *mdev)
+{
+	struct adf_accel_dev *accel_dev = mdev->parent_accel_dev;
+
+	return GET_VFMIG_OPS(accel_dev)->save_setup(mdev);
+}
+EXPORT_SYMBOL_GPL(qat_vfmig_save_setup);
+
+int qat_vfmig_load_state(struct qat_mig_dev *mdev)
+{
+	struct adf_accel_dev *accel_dev = mdev->parent_accel_dev;
+
+	return GET_VFMIG_OPS(accel_dev)->load_state(mdev);
+}
+EXPORT_SYMBOL_GPL(qat_vfmig_load_state);
+
+int qat_vfmig_load_setup(struct qat_mig_dev *mdev, int size)
+{
+	struct adf_accel_dev *accel_dev = mdev->parent_accel_dev;
+
+	return GET_VFMIG_OPS(accel_dev)->load_setup(mdev, size);
+}
+EXPORT_SYMBOL_GPL(qat_vfmig_load_setup);
+
+void qat_vfmig_destroy(struct qat_mig_dev *mdev)
+{
+	kfree(mdev);
+}
+EXPORT_SYMBOL_GPL(qat_vfmig_destroy);
diff --git a/drivers/crypto/intel/qat/qat_dh895xcc/adf_dh895xcc_hw_data.c b/drivers/crypto/intel/qat/qat_dh895xcc/adf_dh895xcc_hw_data.c
index af14090cc4be..6e24d57e6b98 100644
--- a/drivers/crypto/intel/qat/qat_dh895xcc/adf_dh895xcc_hw_data.c
+++ b/drivers/crypto/intel/qat/qat_dh895xcc/adf_dh895xcc_hw_data.c
@@ -5,6 +5,7 @@
 #include <adf_common_drv.h>
 #include <adf_gen2_config.h>
 #include <adf_gen2_dc.h>
+#include <adf_gen2_hw_csr_data.h>
 #include <adf_gen2_hw_data.h>
 #include <adf_gen2_pfvf.h>
 #include "adf_dh895xcc_hw_data.h"
diff --git a/drivers/crypto/intel/qat/qat_dh895xccvf/adf_dh895xccvf_hw_data.c b/drivers/crypto/intel/qat/qat_dh895xccvf/adf_dh895xccvf_hw_data.c
index 70e56cc16ece..f4ee4c2e00da 100644
--- a/drivers/crypto/intel/qat/qat_dh895xccvf/adf_dh895xccvf_hw_data.c
+++ b/drivers/crypto/intel/qat/qat_dh895xccvf/adf_dh895xccvf_hw_data.c
@@ -4,6 +4,7 @@
 #include <adf_common_drv.h>
 #include <adf_gen2_config.h>
 #include <adf_gen2_dc.h>
+#include <adf_gen2_hw_csr_data.h>
 #include <adf_gen2_hw_data.h>
 #include <adf_gen2_pfvf.h>
 #include <adf_pfvf_vf_msg.h>
diff --git a/drivers/crypto/marvell/octeontx2/cn10k_cpt.c b/drivers/crypto/marvell/octeontx2/cn10k_cpt.c
index 79b4e74804f6..6bfc59e67747 100644
--- a/drivers/crypto/marvell/octeontx2/cn10k_cpt.c
+++ b/drivers/crypto/marvell/octeontx2/cn10k_cpt.c
@@ -138,6 +138,10 @@ int cn10k_cpt_hw_ctx_init(struct pci_dev *pdev,
 		return -ENOMEM;
 	cptr_dma = dma_map_single(&pdev->dev, hctx, CN10K_CPT_HW_CTX_SIZE,
 				  DMA_BIDIRECTIONAL);
+	if (dma_mapping_error(&pdev->dev, cptr_dma)) {
+		kfree(hctx);
+		return -ENOMEM;
+	}
 
 	cn10k_cpt_hw_ctx_set(hctx, 1);
 	er_ctx->hw_ctx = hctx;
diff --git a/drivers/crypto/nx/nx-842.c b/drivers/crypto/nx/nx-842.c
index 2ab90ec10e61..82214cde2bcd 100644
--- a/drivers/crypto/nx/nx-842.c
+++ b/drivers/crypto/nx/nx-842.c
@@ -251,7 +251,9 @@ int nx842_crypto_compress(struct crypto_tfm *tfm,
 			  u8 *dst, unsigned int *dlen)
 {
 	struct nx842_crypto_ctx *ctx = crypto_tfm_ctx(tfm);
-	struct nx842_crypto_header *hdr = &ctx->header;
+	struct nx842_crypto_header *hdr =
+				container_of(&ctx->header,
+					     struct nx842_crypto_header, hdr);
 	struct nx842_crypto_param p;
 	struct nx842_constraints c = *ctx->driver->constraints;
 	unsigned int groups, hdrsize, h;
@@ -490,7 +492,7 @@ int nx842_crypto_decompress(struct crypto_tfm *tfm,
 	}
 
 	memcpy(&ctx->header, src, hdr_len);
-	hdr = &ctx->header;
+	hdr = container_of(&ctx->header, struct nx842_crypto_header, hdr);
 
 	for (n = 0; n < hdr->groups; n++) {
 		/* ignore applies to last group */
diff --git a/drivers/crypto/nx/nx-842.h b/drivers/crypto/nx/nx-842.h
index 7590bfb24d79..25fa70b2112c 100644
--- a/drivers/crypto/nx/nx-842.h
+++ b/drivers/crypto/nx/nx-842.h
@@ -157,9 +157,11 @@ struct nx842_crypto_header_group {
 } __packed;
 
 struct nx842_crypto_header {
-	__be16 magic;		/* NX842_CRYPTO_MAGIC */
-	__be16 ignore;		/* decompressed end bytes to ignore */
-	u8 groups;		/* total groups in this header */
+	struct_group_tagged(nx842_crypto_header_hdr, hdr,
+		__be16 magic;		/* NX842_CRYPTO_MAGIC */
+		__be16 ignore;		/* decompressed end bytes to ignore */
+		u8 groups;		/* total groups in this header */
+	);
 	struct nx842_crypto_header_group group[];
 } __packed;
 
@@ -171,7 +173,7 @@ struct nx842_crypto_ctx {
 	u8 *wmem;
 	u8 *sbounce, *dbounce;
 
-	struct nx842_crypto_header header;
+	struct nx842_crypto_header_hdr header;
 	struct nx842_crypto_header_group group[NX842_CRYPTO_GROUP_MAX];
 
 	struct nx842_driver *driver;
diff --git a/drivers/crypto/sahara.c b/drivers/crypto/sahara.c
index 3423b5cde1c7..96d4af5d48a6 100644
--- a/drivers/crypto/sahara.c
+++ b/drivers/crypto/sahara.c
@@ -559,7 +559,7 @@ static int sahara_aes_process(struct skcipher_request *req)
 	struct sahara_ctx *ctx;
 	struct sahara_aes_reqctx *rctx;
 	int ret;
-	unsigned long timeout;
+	unsigned long time_left;
 
 	/* Request is ready to be dispatched by the device */
 	dev_dbg(dev->device,
@@ -597,15 +597,15 @@ static int sahara_aes_process(struct skcipher_request *req)
 	if (ret)
 		return -EINVAL;
 
-	timeout = wait_for_completion_timeout(&dev->dma_completion,
-				msecs_to_jiffies(SAHARA_TIMEOUT_MS));
+	time_left = wait_for_completion_timeout(&dev->dma_completion,
+						msecs_to_jiffies(SAHARA_TIMEOUT_MS));
 
 	dma_unmap_sg(dev->device, dev->out_sg, dev->nb_out_sg,
 		DMA_FROM_DEVICE);
 	dma_unmap_sg(dev->device, dev->in_sg, dev->nb_in_sg,
 		DMA_TO_DEVICE);
 
-	if (!timeout) {
+	if (!time_left) {
 		dev_err(dev->device, "AES timeout\n");
 		return -ETIMEDOUT;
 	}
@@ -931,7 +931,7 @@ static int sahara_sha_process(struct ahash_request *req)
 	struct sahara_dev *dev = dev_ptr;
 	struct sahara_sha_reqctx *rctx = ahash_request_ctx(req);
 	int ret;
-	unsigned long timeout;
+	unsigned long time_left;
 
 	ret = sahara_sha_prepare_request(req);
 	if (!ret)
@@ -963,14 +963,14 @@ static int sahara_sha_process(struct ahash_request *req)
 
 	sahara_write(dev, dev->hw_phys_desc[0], SAHARA_REG_DAR);
 
-	timeout = wait_for_completion_timeout(&dev->dma_completion,
-				msecs_to_jiffies(SAHARA_TIMEOUT_MS));
+	time_left = wait_for_completion_timeout(&dev->dma_completion,
+						msecs_to_jiffies(SAHARA_TIMEOUT_MS));
 
 	if (rctx->sg_in_idx)
 		dma_unmap_sg(dev->device, dev->in_sg, dev->nb_in_sg,
 			     DMA_TO_DEVICE);
 
-	if (!timeout) {
+	if (!time_left) {
 		dev_err(dev->device, "SHA timeout\n");
 		return -ETIMEDOUT;
 	}
diff --git a/drivers/crypto/starfive/Kconfig b/drivers/crypto/starfive/Kconfig
index cb59357b58b2..0fe389e9f932 100644
--- a/drivers/crypto/starfive/Kconfig
+++ b/drivers/crypto/starfive/Kconfig
@@ -14,6 +14,10 @@ config CRYPTO_DEV_JH7110
 	select CRYPTO_RSA
 	select CRYPTO_AES
 	select CRYPTO_CCM
+	select CRYPTO_GCM
+	select CRYPTO_ECB
+	select CRYPTO_CBC
+	select CRYPTO_CTR
 	help
 	  Support for StarFive JH7110 crypto hardware acceleration engine.
 	  This module provides acceleration for public key algo,
diff --git a/drivers/crypto/starfive/jh7110-aes.c b/drivers/crypto/starfive/jh7110-aes.c
index 1ac15cc4ef3c..86a1a1fa9f8f 100644
--- a/drivers/crypto/starfive/jh7110-aes.c
+++ b/drivers/crypto/starfive/jh7110-aes.c
@@ -78,7 +78,7 @@ static inline int is_gcm(struct starfive_cryp_dev *cryp)
 	return (cryp->flags & FLG_MODE_MASK) == STARFIVE_AES_MODE_GCM;
 }
 
-static inline int is_encrypt(struct starfive_cryp_dev *cryp)
+static inline bool is_encrypt(struct starfive_cryp_dev *cryp)
 {
 	return cryp->flags & FLG_ENCRYPT;
 }
@@ -103,16 +103,6 @@ static void starfive_aes_aead_hw_start(struct starfive_cryp_ctx *ctx, u32 hw_mod
 	}
 }
 
-static inline void starfive_aes_set_ivlen(struct starfive_cryp_ctx *ctx)
-{
-	struct starfive_cryp_dev *cryp = ctx->cryp;
-
-	if (is_gcm(cryp))
-		writel(GCM_AES_IV_SIZE, cryp->base + STARFIVE_AES_IVLEN);
-	else
-		writel(AES_BLOCK_SIZE, cryp->base + STARFIVE_AES_IVLEN);
-}
-
 static inline void starfive_aes_set_alen(struct starfive_cryp_ctx *ctx)
 {
 	struct starfive_cryp_dev *cryp = ctx->cryp;
@@ -261,7 +251,6 @@ static int starfive_aes_hw_init(struct starfive_cryp_ctx *ctx)
 
 	rctx->csr.aes.mode  = hw_mode;
 	rctx->csr.aes.cmode = !is_encrypt(cryp);
-	rctx->csr.aes.ie = 1;
 	rctx->csr.aes.stmode = STARFIVE_AES_MODE_XFB_1;
 
 	if (cryp->side_chan) {
@@ -279,7 +268,7 @@ static int starfive_aes_hw_init(struct starfive_cryp_ctx *ctx)
 	case STARFIVE_AES_MODE_GCM:
 		starfive_aes_set_alen(ctx);
 		starfive_aes_set_mlen(ctx);
-		starfive_aes_set_ivlen(ctx);
+		writel(GCM_AES_IV_SIZE, cryp->base + STARFIVE_AES_IVLEN);
 		starfive_aes_aead_hw_start(ctx, hw_mode);
 		starfive_aes_write_iv(ctx, (void *)cryp->req.areq->iv);
 		break;
@@ -300,52 +289,49 @@ static int starfive_aes_hw_init(struct starfive_cryp_ctx *ctx)
 	return cryp->err;
 }
 
-static int starfive_aes_read_authtag(struct starfive_cryp_dev *cryp)
+static int starfive_aes_read_authtag(struct starfive_cryp_ctx *ctx)
 {
-	int i, start_addr;
+	struct starfive_cryp_dev *cryp = ctx->cryp;
+	struct starfive_cryp_request_ctx *rctx = ctx->rctx;
+	int i;
 
 	if (starfive_aes_wait_busy(cryp))
 		return dev_err_probe(cryp->dev, -ETIMEDOUT,
 				     "Timeout waiting for tag generation.");
 
-	start_addr = STARFIVE_AES_NONCE0;
-
-	if (is_gcm(cryp))
-		for (i = 0; i < AES_BLOCK_32; i++, start_addr += 4)
-			cryp->tag_out[i] = readl(cryp->base + start_addr);
-	else
+	if ((cryp->flags & FLG_MODE_MASK) == STARFIVE_AES_MODE_GCM) {
+		cryp->tag_out[0] = readl(cryp->base + STARFIVE_AES_NONCE0);
+		cryp->tag_out[1] = readl(cryp->base + STARFIVE_AES_NONCE1);
+		cryp->tag_out[2] = readl(cryp->base + STARFIVE_AES_NONCE2);
+		cryp->tag_out[3] = readl(cryp->base + STARFIVE_AES_NONCE3);
+	} else {
 		for (i = 0; i < AES_BLOCK_32; i++)
 			cryp->tag_out[i] = readl(cryp->base + STARFIVE_AES_AESDIO0R);
+	}
 
 	if (is_encrypt(cryp)) {
-		scatterwalk_copychunks(cryp->tag_out, &cryp->out_walk, cryp->authsize, 1);
+		scatterwalk_map_and_copy(cryp->tag_out, rctx->out_sg,
+					 cryp->total_in, cryp->authsize, 1);
 	} else {
-		scatterwalk_copychunks(cryp->tag_in, &cryp->in_walk, cryp->authsize, 0);
-
 		if (crypto_memneq(cryp->tag_in, cryp->tag_out, cryp->authsize))
-			return dev_err_probe(cryp->dev, -EBADMSG, "Failed tag verification\n");
+			return -EBADMSG;
 	}
 
 	return 0;
 }
 
-static void starfive_aes_finish_req(struct starfive_cryp_dev *cryp)
+static void starfive_aes_finish_req(struct starfive_cryp_ctx *ctx)
 {
-	union starfive_aes_csr csr;
+	struct starfive_cryp_dev *cryp = ctx->cryp;
 	int err = cryp->err;
 
 	if (!err && cryp->authsize)
-		err = starfive_aes_read_authtag(cryp);
+		err = starfive_aes_read_authtag(ctx);
 
 	if (!err && ((cryp->flags & FLG_MODE_MASK) == STARFIVE_AES_MODE_CBC ||
 		     (cryp->flags & FLG_MODE_MASK) == STARFIVE_AES_MODE_CTR))
 		starfive_aes_get_iv(cryp, (void *)cryp->req.sreq->iv);
 
-	/* reset irq flags*/
-	csr.v = 0;
-	csr.aesrst = 1;
-	writel(csr.v, cryp->base + STARFIVE_AES_CSR);
-
 	if (cryp->authsize)
 		crypto_finalize_aead_request(cryp->engine, cryp->req.areq, err);
 	else
@@ -353,39 +339,6 @@ static void starfive_aes_finish_req(struct starfive_cryp_dev *cryp)
 						 err);
 }
 
-void starfive_aes_done_task(unsigned long param)
-{
-	struct starfive_cryp_dev *cryp = (struct starfive_cryp_dev *)param;
-	u32 block[AES_BLOCK_32];
-	u32 stat;
-	int i;
-
-	for (i = 0; i < AES_BLOCK_32; i++)
-		block[i] = readl(cryp->base + STARFIVE_AES_AESDIO0R);
-
-	scatterwalk_copychunks(block, &cryp->out_walk, min_t(size_t, AES_BLOCK_SIZE,
-							     cryp->total_out), 1);
-
-	cryp->total_out -= min_t(size_t, AES_BLOCK_SIZE, cryp->total_out);
-
-	if (!cryp->total_out) {
-		starfive_aes_finish_req(cryp);
-		return;
-	}
-
-	memset(block, 0, AES_BLOCK_SIZE);
-	scatterwalk_copychunks(block, &cryp->in_walk, min_t(size_t, AES_BLOCK_SIZE,
-							    cryp->total_in), 0);
-	cryp->total_in -= min_t(size_t, AES_BLOCK_SIZE, cryp->total_in);
-
-	for (i = 0; i < AES_BLOCK_32; i++)
-		writel(block[i], cryp->base + STARFIVE_AES_AESDIO0R);
-
-	stat = readl(cryp->base + STARFIVE_IE_MASK_OFFSET);
-	stat &= ~STARFIVE_IE_MASK_AES_DONE;
-	writel(stat, cryp->base + STARFIVE_IE_MASK_OFFSET);
-}
-
 static int starfive_aes_gcm_write_adata(struct starfive_cryp_ctx *ctx)
 {
 	struct starfive_cryp_dev *cryp = ctx->cryp;
@@ -451,60 +404,165 @@ static int starfive_aes_ccm_write_adata(struct starfive_cryp_ctx *ctx)
 	return 0;
 }
 
-static int starfive_aes_prepare_req(struct skcipher_request *req,
-				    struct aead_request *areq)
+static void starfive_aes_dma_done(void *param)
 {
-	struct starfive_cryp_ctx *ctx;
-	struct starfive_cryp_request_ctx *rctx;
-	struct starfive_cryp_dev *cryp;
+	struct starfive_cryp_dev *cryp = param;
 
-	if (!req && !areq)
-		return -EINVAL;
+	complete(&cryp->dma_done);
+}
 
-	ctx = req ? crypto_skcipher_ctx(crypto_skcipher_reqtfm(req)) :
-		    crypto_aead_ctx(crypto_aead_reqtfm(areq));
+static void starfive_aes_dma_init(struct starfive_cryp_dev *cryp)
+{
+	cryp->cfg_in.direction = DMA_MEM_TO_DEV;
+	cryp->cfg_in.src_addr_width = DMA_SLAVE_BUSWIDTH_16_BYTES;
+	cryp->cfg_in.dst_addr_width = DMA_SLAVE_BUSWIDTH_4_BYTES;
+	cryp->cfg_in.src_maxburst = cryp->dma_maxburst;
+	cryp->cfg_in.dst_maxburst = cryp->dma_maxburst;
+	cryp->cfg_in.dst_addr = cryp->phys_base + STARFIVE_ALG_FIFO_OFFSET;
 
-	cryp = ctx->cryp;
-	rctx = req ? skcipher_request_ctx(req) : aead_request_ctx(areq);
+	dmaengine_slave_config(cryp->tx, &cryp->cfg_in);
 
-	if (req) {
-		cryp->req.sreq = req;
-		cryp->total_in = req->cryptlen;
-		cryp->total_out = req->cryptlen;
-		cryp->assoclen = 0;
-		cryp->authsize = 0;
-	} else {
-		cryp->req.areq = areq;
-		cryp->assoclen = areq->assoclen;
-		cryp->authsize = crypto_aead_authsize(crypto_aead_reqtfm(areq));
-		if (is_encrypt(cryp)) {
-			cryp->total_in = areq->cryptlen;
-			cryp->total_out = areq->cryptlen;
-		} else {
-			cryp->total_in = areq->cryptlen - cryp->authsize;
-			cryp->total_out = cryp->total_in;
-		}
-	}
+	cryp->cfg_out.direction = DMA_DEV_TO_MEM;
+	cryp->cfg_out.src_addr_width = DMA_SLAVE_BUSWIDTH_4_BYTES;
+	cryp->cfg_out.dst_addr_width = DMA_SLAVE_BUSWIDTH_16_BYTES;
+	cryp->cfg_out.src_maxburst = 4;
+	cryp->cfg_out.dst_maxburst = 4;
+	cryp->cfg_out.src_addr = cryp->phys_base + STARFIVE_ALG_FIFO_OFFSET;
 
-	rctx->in_sg = req ? req->src : areq->src;
-	scatterwalk_start(&cryp->in_walk, rctx->in_sg);
+	dmaengine_slave_config(cryp->rx, &cryp->cfg_out);
 
-	rctx->out_sg = req ? req->dst : areq->dst;
-	scatterwalk_start(&cryp->out_walk, rctx->out_sg);
+	init_completion(&cryp->dma_done);
+}
 
-	if (cryp->assoclen) {
-		rctx->adata = kzalloc(cryp->assoclen + AES_BLOCK_SIZE, GFP_KERNEL);
-		if (!rctx->adata)
-			return dev_err_probe(cryp->dev, -ENOMEM,
-					     "Failed to alloc memory for adata");
+static int starfive_aes_dma_xfer(struct starfive_cryp_dev *cryp,
+				 struct scatterlist *src,
+				 struct scatterlist *dst,
+				 int len)
+{
+	struct dma_async_tx_descriptor *in_desc, *out_desc;
+	union starfive_alg_cr alg_cr;
+	int ret = 0, in_save, out_save;
+
+	alg_cr.v = 0;
+	alg_cr.start = 1;
+	alg_cr.aes_dma_en = 1;
+	writel(alg_cr.v, cryp->base + STARFIVE_ALG_CR_OFFSET);
+
+	in_save = sg_dma_len(src);
+	out_save = sg_dma_len(dst);
 
-		scatterwalk_copychunks(rctx->adata, &cryp->in_walk, cryp->assoclen, 0);
-		scatterwalk_copychunks(NULL, &cryp->out_walk, cryp->assoclen, 2);
+	writel(ALIGN(len, AES_BLOCK_SIZE), cryp->base + STARFIVE_DMA_IN_LEN_OFFSET);
+	writel(ALIGN(len, AES_BLOCK_SIZE), cryp->base + STARFIVE_DMA_OUT_LEN_OFFSET);
+
+	sg_dma_len(src) = ALIGN(len, AES_BLOCK_SIZE);
+	sg_dma_len(dst) = ALIGN(len, AES_BLOCK_SIZE);
+
+	out_desc = dmaengine_prep_slave_sg(cryp->rx, dst, 1, DMA_DEV_TO_MEM,
+					   DMA_PREP_INTERRUPT | DMA_CTRL_ACK);
+	if (!out_desc) {
+		ret = -EINVAL;
+		goto dma_err;
 	}
 
-	ctx->rctx = rctx;
+	out_desc->callback = starfive_aes_dma_done;
+	out_desc->callback_param = cryp;
+
+	reinit_completion(&cryp->dma_done);
+	dmaengine_submit(out_desc);
+	dma_async_issue_pending(cryp->rx);
+
+	in_desc = dmaengine_prep_slave_sg(cryp->tx, src, 1, DMA_MEM_TO_DEV,
+					  DMA_PREP_INTERRUPT | DMA_CTRL_ACK);
+	if (!in_desc) {
+		ret = -EINVAL;
+		goto dma_err;
+	}
+
+	dmaengine_submit(in_desc);
+	dma_async_issue_pending(cryp->tx);
+
+	if (!wait_for_completion_timeout(&cryp->dma_done,
+					 msecs_to_jiffies(1000)))
+		ret = -ETIMEDOUT;
+
+dma_err:
+	sg_dma_len(src) = in_save;
+	sg_dma_len(dst) = out_save;
+
+	alg_cr.v = 0;
+	alg_cr.clear = 1;
+	writel(alg_cr.v, cryp->base + STARFIVE_ALG_CR_OFFSET);
+
+	return ret;
+}
+
+static int starfive_aes_map_sg(struct starfive_cryp_dev *cryp,
+			       struct scatterlist *src,
+			       struct scatterlist *dst)
+{
+	struct scatterlist *stsg, *dtsg;
+	struct scatterlist _src[2], _dst[2];
+	unsigned int remain = cryp->total_in;
+	unsigned int len, src_nents, dst_nents;
+	int ret;
+
+	if (src == dst) {
+		for (stsg = src, dtsg = dst; remain > 0;
+		     stsg = sg_next(stsg), dtsg = sg_next(dtsg)) {
+			src_nents = dma_map_sg(cryp->dev, stsg, 1, DMA_BIDIRECTIONAL);
+			if (src_nents == 0)
+				return dev_err_probe(cryp->dev, -ENOMEM,
+						     "dma_map_sg error\n");
+
+			dst_nents = src_nents;
+			len = min(sg_dma_len(stsg), remain);
+
+			ret = starfive_aes_dma_xfer(cryp, stsg, dtsg, len);
+			dma_unmap_sg(cryp->dev, stsg, 1, DMA_BIDIRECTIONAL);
+			if (ret)
+				return ret;
+
+			remain -= len;
+		}
+	} else {
+		for (stsg = src, dtsg = dst;;) {
+			src_nents = dma_map_sg(cryp->dev, stsg, 1, DMA_TO_DEVICE);
+			if (src_nents == 0)
+				return dev_err_probe(cryp->dev, -ENOMEM,
+						     "dma_map_sg src error\n");
+
+			dst_nents = dma_map_sg(cryp->dev, dtsg, 1, DMA_FROM_DEVICE);
+			if (dst_nents == 0)
+				return dev_err_probe(cryp->dev, -ENOMEM,
+						     "dma_map_sg dst error\n");
+
+			len = min(sg_dma_len(stsg), sg_dma_len(dtsg));
+			len = min(len, remain);
+
+			ret = starfive_aes_dma_xfer(cryp, stsg, dtsg, len);
+			dma_unmap_sg(cryp->dev, stsg, 1, DMA_TO_DEVICE);
+			dma_unmap_sg(cryp->dev, dtsg, 1, DMA_FROM_DEVICE);
+			if (ret)
+				return ret;
+
+			remain -= len;
+			if (remain == 0)
+				break;
+
+			if (sg_dma_len(stsg) - len) {
+				stsg = scatterwalk_ffwd(_src, stsg, len);
+				dtsg = sg_next(dtsg);
+			} else if (sg_dma_len(dtsg) - len) {
+				dtsg = scatterwalk_ffwd(_dst, dtsg, len);
+				stsg = sg_next(stsg);
+			} else {
+				stsg = sg_next(stsg);
+				dtsg = sg_next(dtsg);
+			}
+		}
+	}
 
-	return starfive_aes_hw_init(ctx);
+	return 0;
 }
 
 static int starfive_aes_do_one_req(struct crypto_engine *engine, void *areq)
@@ -513,35 +571,42 @@ static int starfive_aes_do_one_req(struct crypto_engine *engine, void *areq)
 		container_of(areq, struct skcipher_request, base);
 	struct starfive_cryp_ctx *ctx =
 		crypto_skcipher_ctx(crypto_skcipher_reqtfm(req));
+	struct starfive_cryp_request_ctx *rctx = skcipher_request_ctx(req);
 	struct starfive_cryp_dev *cryp = ctx->cryp;
-	u32 block[AES_BLOCK_32];
-	u32 stat;
-	int err;
-	int i;
+	int ret;
 
-	err = starfive_aes_prepare_req(req, NULL);
-	if (err)
-		return err;
+	cryp->req.sreq = req;
+	cryp->total_in = req->cryptlen;
+	cryp->total_out = req->cryptlen;
+	cryp->assoclen = 0;
+	cryp->authsize = 0;
 
-	/*
-	 * Write first plain/ciphertext block to start the module
-	 * then let irq tasklet handle the rest of the data blocks.
-	 */
-	scatterwalk_copychunks(block, &cryp->in_walk, min_t(size_t, AES_BLOCK_SIZE,
-							    cryp->total_in), 0);
-	cryp->total_in -= min_t(size_t, AES_BLOCK_SIZE, cryp->total_in);
+	rctx->in_sg = req->src;
+	rctx->out_sg = req->dst;
+
+	ctx->rctx = rctx;
+
+	ret = starfive_aes_hw_init(ctx);
+	if (ret)
+		return ret;
 
-	for (i = 0; i < AES_BLOCK_32; i++)
-		writel(block[i], cryp->base + STARFIVE_AES_AESDIO0R);
+	if (!cryp->total_in)
+		goto finish_req;
 
-	stat = readl(cryp->base + STARFIVE_IE_MASK_OFFSET);
-	stat &= ~STARFIVE_IE_MASK_AES_DONE;
-	writel(stat, cryp->base + STARFIVE_IE_MASK_OFFSET);
+	starfive_aes_dma_init(cryp);
+
+	ret = starfive_aes_map_sg(cryp, rctx->in_sg, rctx->out_sg);
+	if (ret)
+		return ret;
+
+finish_req:
+	starfive_aes_finish_req(ctx);
 
 	return 0;
 }
 
-static int starfive_aes_init_tfm(struct crypto_skcipher *tfm)
+static int starfive_aes_init_tfm(struct crypto_skcipher *tfm,
+				 const char *alg_name)
 {
 	struct starfive_cryp_ctx *ctx = crypto_skcipher_ctx(tfm);
 
@@ -549,12 +614,26 @@ static int starfive_aes_init_tfm(struct crypto_skcipher *tfm)
 	if (!ctx->cryp)
 		return -ENODEV;
 
+	ctx->skcipher_fbk = crypto_alloc_skcipher(alg_name, 0,
+						  CRYPTO_ALG_NEED_FALLBACK);
+	if (IS_ERR(ctx->skcipher_fbk))
+		return dev_err_probe(ctx->cryp->dev, PTR_ERR(ctx->skcipher_fbk),
+				     "%s() failed to allocate fallback for %s\n",
+				     __func__, alg_name);
+
 	crypto_skcipher_set_reqsize(tfm, sizeof(struct starfive_cryp_request_ctx) +
-				    sizeof(struct skcipher_request));
+				    crypto_skcipher_reqsize(ctx->skcipher_fbk));
 
 	return 0;
 }
 
+static void starfive_aes_exit_tfm(struct crypto_skcipher *tfm)
+{
+	struct starfive_cryp_ctx *ctx = crypto_skcipher_ctx(tfm);
+
+	crypto_free_skcipher(ctx->skcipher_fbk);
+}
+
 static int starfive_aes_aead_do_one_req(struct crypto_engine *engine, void *areq)
 {
 	struct aead_request *req =
@@ -562,79 +641,99 @@ static int starfive_aes_aead_do_one_req(struct crypto_engine *engine, void *areq
 	struct starfive_cryp_ctx *ctx =
 		crypto_aead_ctx(crypto_aead_reqtfm(req));
 	struct starfive_cryp_dev *cryp = ctx->cryp;
-	struct starfive_cryp_request_ctx *rctx;
-	u32 block[AES_BLOCK_32];
-	u32 stat;
-	int err;
-	int i;
+	struct starfive_cryp_request_ctx *rctx = aead_request_ctx(req);
+	struct scatterlist _src[2], _dst[2];
+	int ret;
+
+	cryp->req.areq = req;
+	cryp->assoclen = req->assoclen;
+	cryp->authsize = crypto_aead_authsize(crypto_aead_reqtfm(req));
+
+	rctx->in_sg = scatterwalk_ffwd(_src, req->src, cryp->assoclen);
+	if (req->src == req->dst)
+		rctx->out_sg = rctx->in_sg;
+	else
+		rctx->out_sg = scatterwalk_ffwd(_dst, req->dst, cryp->assoclen);
+
+	if (is_encrypt(cryp)) {
+		cryp->total_in = req->cryptlen;
+		cryp->total_out = req->cryptlen;
+	} else {
+		cryp->total_in = req->cryptlen - cryp->authsize;
+		cryp->total_out = cryp->total_in;
+		scatterwalk_map_and_copy(cryp->tag_in, req->src,
+					 cryp->total_in + cryp->assoclen,
+					 cryp->authsize, 0);
+	}
 
-	err = starfive_aes_prepare_req(NULL, req);
-	if (err)
-		return err;
+	if (cryp->assoclen) {
+		rctx->adata = kzalloc(cryp->assoclen + AES_BLOCK_SIZE, GFP_KERNEL);
+		if (!rctx->adata)
+			return dev_err_probe(cryp->dev, -ENOMEM,
+					     "Failed to alloc memory for adata");
+
+		if (sg_copy_to_buffer(req->src, sg_nents_for_len(req->src, cryp->assoclen),
+				      rctx->adata, cryp->assoclen) != cryp->assoclen)
+			return -EINVAL;
+	}
+
+	if (cryp->total_in)
+		sg_zero_buffer(rctx->in_sg, sg_nents(rctx->in_sg),
+			       sg_dma_len(rctx->in_sg) - cryp->total_in,
+			       cryp->total_in);
 
-	rctx = ctx->rctx;
+	ctx->rctx = rctx;
+
+	ret = starfive_aes_hw_init(ctx);
+	if (ret)
+		return ret;
 
 	if (!cryp->assoclen)
 		goto write_text;
 
 	if ((cryp->flags & FLG_MODE_MASK) == STARFIVE_AES_MODE_CCM)
-		cryp->err = starfive_aes_ccm_write_adata(ctx);
+		ret = starfive_aes_ccm_write_adata(ctx);
 	else
-		cryp->err = starfive_aes_gcm_write_adata(ctx);
+		ret = starfive_aes_gcm_write_adata(ctx);
 
 	kfree(rctx->adata);
 
-	if (cryp->err)
-		return cryp->err;
+	if (ret)
+		return ret;
 
 write_text:
 	if (!cryp->total_in)
 		goto finish_req;
 
-	/*
-	 * Write first plain/ciphertext block to start the module
-	 * then let irq tasklet handle the rest of the data blocks.
-	 */
-	scatterwalk_copychunks(block, &cryp->in_walk, min_t(size_t, AES_BLOCK_SIZE,
-							    cryp->total_in), 0);
-	cryp->total_in -= min_t(size_t, AES_BLOCK_SIZE, cryp->total_in);
-
-	for (i = 0; i < AES_BLOCK_32; i++)
-		writel(block[i], cryp->base + STARFIVE_AES_AESDIO0R);
-
-	stat = readl(cryp->base + STARFIVE_IE_MASK_OFFSET);
-	stat &= ~STARFIVE_IE_MASK_AES_DONE;
-	writel(stat, cryp->base + STARFIVE_IE_MASK_OFFSET);
+	starfive_aes_dma_init(cryp);
 
-	return 0;
+	ret = starfive_aes_map_sg(cryp, rctx->in_sg, rctx->out_sg);
+	if (ret)
+		return ret;
 
 finish_req:
-	starfive_aes_finish_req(cryp);
+	starfive_aes_finish_req(ctx);
 	return 0;
 }
 
-static int starfive_aes_aead_init_tfm(struct crypto_aead *tfm)
+static int starfive_aes_aead_init_tfm(struct crypto_aead *tfm,
+				      const char *alg_name)
 {
 	struct starfive_cryp_ctx *ctx = crypto_aead_ctx(tfm);
-	struct starfive_cryp_dev *cryp = ctx->cryp;
-	struct crypto_tfm *aead = crypto_aead_tfm(tfm);
-	struct crypto_alg *alg = aead->__crt_alg;
 
 	ctx->cryp = starfive_cryp_find_dev(ctx);
 	if (!ctx->cryp)
 		return -ENODEV;
 
-	if (alg->cra_flags & CRYPTO_ALG_NEED_FALLBACK) {
-		ctx->aead_fbk = crypto_alloc_aead(alg->cra_name, 0,
-						  CRYPTO_ALG_NEED_FALLBACK);
-		if (IS_ERR(ctx->aead_fbk))
-			return dev_err_probe(cryp->dev, PTR_ERR(ctx->aead_fbk),
-					     "%s() failed to allocate fallback for %s\n",
-					     __func__, alg->cra_name);
-	}
+	ctx->aead_fbk = crypto_alloc_aead(alg_name, 0,
+					  CRYPTO_ALG_NEED_FALLBACK);
+	if (IS_ERR(ctx->aead_fbk))
+		return dev_err_probe(ctx->cryp->dev, PTR_ERR(ctx->aead_fbk),
+				     "%s() failed to allocate fallback for %s\n",
+				     __func__, alg_name);
 
-	crypto_aead_set_reqsize(tfm, sizeof(struct starfive_cryp_ctx) +
-				sizeof(struct aead_request));
+	crypto_aead_set_reqsize(tfm, sizeof(struct starfive_cryp_request_ctx) +
+				crypto_aead_reqsize(ctx->aead_fbk));
 
 	return 0;
 }
@@ -646,6 +745,46 @@ static void starfive_aes_aead_exit_tfm(struct crypto_aead *tfm)
 	crypto_free_aead(ctx->aead_fbk);
 }
 
+static bool starfive_aes_check_unaligned(struct starfive_cryp_dev *cryp,
+					 struct scatterlist *src,
+					 struct scatterlist *dst)
+{
+	struct scatterlist *tsg;
+	int i;
+
+	for_each_sg(src, tsg, sg_nents(src), i)
+		if (!IS_ALIGNED(tsg->offset, sizeof(u32)) ||
+		    (!IS_ALIGNED(tsg->length, AES_BLOCK_SIZE) &&
+		     !sg_is_last(tsg)))
+			return true;
+
+	if (src != dst)
+		for_each_sg(dst, tsg, sg_nents(dst), i)
+			if (!IS_ALIGNED(tsg->offset, sizeof(u32)) ||
+			    (!IS_ALIGNED(tsg->length, AES_BLOCK_SIZE) &&
+			     !sg_is_last(tsg)))
+				return true;
+
+	return false;
+}
+
+static int starfive_aes_do_fallback(struct skcipher_request *req, bool enc)
+{
+	struct starfive_cryp_ctx *ctx =
+		crypto_skcipher_ctx(crypto_skcipher_reqtfm(req));
+	struct skcipher_request *subreq = skcipher_request_ctx(req);
+
+	skcipher_request_set_tfm(subreq, ctx->skcipher_fbk);
+	skcipher_request_set_callback(subreq, req->base.flags,
+				      req->base.complete,
+				      req->base.data);
+	skcipher_request_set_crypt(subreq, req->src, req->dst,
+				   req->cryptlen, req->iv);
+
+	return enc ? crypto_skcipher_encrypt(subreq) :
+		     crypto_skcipher_decrypt(subreq);
+}
+
 static int starfive_aes_crypt(struct skcipher_request *req, unsigned long flags)
 {
 	struct crypto_skcipher *tfm = crypto_skcipher_reqtfm(req);
@@ -660,32 +799,54 @@ static int starfive_aes_crypt(struct skcipher_request *req, unsigned long flags)
 		if (req->cryptlen & blocksize_align)
 			return -EINVAL;
 
+	if (starfive_aes_check_unaligned(cryp, req->src, req->dst))
+		return starfive_aes_do_fallback(req, is_encrypt(cryp));
+
 	return crypto_transfer_skcipher_request_to_engine(cryp->engine, req);
 }
 
+static int starfive_aes_aead_do_fallback(struct aead_request *req, bool enc)
+{
+	struct starfive_cryp_ctx *ctx =
+		crypto_aead_ctx(crypto_aead_reqtfm(req));
+	struct aead_request *subreq = aead_request_ctx(req);
+
+	aead_request_set_tfm(subreq, ctx->aead_fbk);
+	aead_request_set_callback(subreq, req->base.flags,
+				  req->base.complete,
+				  req->base.data);
+	aead_request_set_crypt(subreq, req->src, req->dst,
+			       req->cryptlen, req->iv);
+	aead_request_set_ad(subreq, req->assoclen);
+
+	return enc ? crypto_aead_encrypt(subreq) :
+		     crypto_aead_decrypt(subreq);
+}
+
 static int starfive_aes_aead_crypt(struct aead_request *req, unsigned long flags)
 {
 	struct starfive_cryp_ctx *ctx = crypto_aead_ctx(crypto_aead_reqtfm(req));
 	struct starfive_cryp_dev *cryp = ctx->cryp;
+	struct scatterlist *src, *dst, _src[2], _dst[2];
 
 	cryp->flags = flags;
 
-	/*
-	 * HW engine could not perform CCM tag verification on
-	 * non-blocksize aligned text, use fallback algo instead
+	/* aes-ccm does not support tag verification for non-aligned text,
+	 * use fallback for ccm decryption instead.
 	 */
-	if (ctx->aead_fbk && !is_encrypt(cryp)) {
-		struct aead_request *subreq = aead_request_ctx(req);
+	if (((cryp->flags & FLG_MODE_MASK) == STARFIVE_AES_MODE_CCM) &&
+	    !is_encrypt(cryp))
+		return starfive_aes_aead_do_fallback(req, 0);
 
-		aead_request_set_tfm(subreq, ctx->aead_fbk);
-		aead_request_set_callback(subreq, req->base.flags,
-					  req->base.complete, req->base.data);
-		aead_request_set_crypt(subreq, req->src,
-				       req->dst, req->cryptlen, req->iv);
-		aead_request_set_ad(subreq, req->assoclen);
+	src = scatterwalk_ffwd(_src, req->src, req->assoclen);
 
-		return crypto_aead_decrypt(subreq);
-	}
+	if (req->src == req->dst)
+		dst = src;
+	else
+		dst = scatterwalk_ffwd(_dst, req->dst, req->assoclen);
+
+	if (starfive_aes_check_unaligned(cryp, src, dst))
+		return starfive_aes_aead_do_fallback(req, is_encrypt(cryp));
 
 	return crypto_transfer_aead_request_to_engine(cryp->engine, req);
 }
@@ -706,7 +867,7 @@ static int starfive_aes_setkey(struct crypto_skcipher *tfm, const u8 *key,
 	memcpy(ctx->key, key, keylen);
 	ctx->keylen = keylen;
 
-	return 0;
+	return crypto_skcipher_setkey(ctx->skcipher_fbk, key, keylen);
 }
 
 static int starfive_aes_aead_setkey(struct crypto_aead *tfm, const u8 *key,
@@ -725,16 +886,20 @@ static int starfive_aes_aead_setkey(struct crypto_aead *tfm, const u8 *key,
 	memcpy(ctx->key, key, keylen);
 	ctx->keylen = keylen;
 
-	if (ctx->aead_fbk)
-		return crypto_aead_setkey(ctx->aead_fbk, key, keylen);
-
-	return 0;
+	return crypto_aead_setkey(ctx->aead_fbk, key, keylen);
 }
 
 static int starfive_aes_gcm_setauthsize(struct crypto_aead *tfm,
 					unsigned int authsize)
 {
-	return crypto_gcm_check_authsize(authsize);
+	struct starfive_cryp_ctx *ctx = crypto_aead_ctx(tfm);
+	int ret;
+
+	ret = crypto_gcm_check_authsize(authsize);
+	if (ret)
+		return ret;
+
+	return crypto_aead_setauthsize(ctx->aead_fbk, authsize);
 }
 
 static int starfive_aes_ccm_setauthsize(struct crypto_aead *tfm,
@@ -820,9 +985,35 @@ static int starfive_aes_ccm_decrypt(struct aead_request *req)
 	return starfive_aes_aead_crypt(req, STARFIVE_AES_MODE_CCM);
 }
 
+static int starfive_aes_ecb_init_tfm(struct crypto_skcipher *tfm)
+{
+	return starfive_aes_init_tfm(tfm, "ecb(aes-generic)");
+}
+
+static int starfive_aes_cbc_init_tfm(struct crypto_skcipher *tfm)
+{
+	return starfive_aes_init_tfm(tfm, "cbc(aes-generic)");
+}
+
+static int starfive_aes_ctr_init_tfm(struct crypto_skcipher *tfm)
+{
+	return starfive_aes_init_tfm(tfm, "ctr(aes-generic)");
+}
+
+static int starfive_aes_ccm_init_tfm(struct crypto_aead *tfm)
+{
+	return starfive_aes_aead_init_tfm(tfm, "ccm_base(ctr(aes-generic),cbcmac(aes-generic))");
+}
+
+static int starfive_aes_gcm_init_tfm(struct crypto_aead *tfm)
+{
+	return starfive_aes_aead_init_tfm(tfm, "gcm_base(ctr(aes-generic),ghash-generic)");
+}
+
 static struct skcipher_engine_alg skcipher_algs[] = {
 {
-	.base.init			= starfive_aes_init_tfm,
+	.base.init			= starfive_aes_ecb_init_tfm,
+	.base.exit			= starfive_aes_exit_tfm,
 	.base.setkey			= starfive_aes_setkey,
 	.base.encrypt			= starfive_aes_ecb_encrypt,
 	.base.decrypt			= starfive_aes_ecb_decrypt,
@@ -832,7 +1023,8 @@ static struct skcipher_engine_alg skcipher_algs[] = {
 		.cra_name		= "ecb(aes)",
 		.cra_driver_name	= "starfive-ecb-aes",
 		.cra_priority		= 200,
-		.cra_flags		= CRYPTO_ALG_ASYNC,
+		.cra_flags		= CRYPTO_ALG_ASYNC |
+					  CRYPTO_ALG_NEED_FALLBACK,
 		.cra_blocksize		= AES_BLOCK_SIZE,
 		.cra_ctxsize		= sizeof(struct starfive_cryp_ctx),
 		.cra_alignmask		= 0xf,
@@ -842,7 +1034,8 @@ static struct skcipher_engine_alg skcipher_algs[] = {
 		.do_one_request = starfive_aes_do_one_req,
 	},
 }, {
-	.base.init			= starfive_aes_init_tfm,
+	.base.init			= starfive_aes_cbc_init_tfm,
+	.base.exit			= starfive_aes_exit_tfm,
 	.base.setkey			= starfive_aes_setkey,
 	.base.encrypt			= starfive_aes_cbc_encrypt,
 	.base.decrypt			= starfive_aes_cbc_decrypt,
@@ -853,7 +1046,8 @@ static struct skcipher_engine_alg skcipher_algs[] = {
 		.cra_name		= "cbc(aes)",
 		.cra_driver_name	= "starfive-cbc-aes",
 		.cra_priority		= 200,
-		.cra_flags		= CRYPTO_ALG_ASYNC,
+		.cra_flags		= CRYPTO_ALG_ASYNC |
+					  CRYPTO_ALG_NEED_FALLBACK,
 		.cra_blocksize		= AES_BLOCK_SIZE,
 		.cra_ctxsize		= sizeof(struct starfive_cryp_ctx),
 		.cra_alignmask		= 0xf,
@@ -863,7 +1057,8 @@ static struct skcipher_engine_alg skcipher_algs[] = {
 		.do_one_request = starfive_aes_do_one_req,
 	},
 }, {
-	.base.init			= starfive_aes_init_tfm,
+	.base.init			= starfive_aes_ctr_init_tfm,
+	.base.exit			= starfive_aes_exit_tfm,
 	.base.setkey			= starfive_aes_setkey,
 	.base.encrypt			= starfive_aes_ctr_encrypt,
 	.base.decrypt			= starfive_aes_ctr_decrypt,
@@ -874,7 +1069,8 @@ static struct skcipher_engine_alg skcipher_algs[] = {
 		.cra_name		= "ctr(aes)",
 		.cra_driver_name	= "starfive-ctr-aes",
 		.cra_priority		= 200,
-		.cra_flags		= CRYPTO_ALG_ASYNC,
+		.cra_flags		= CRYPTO_ALG_ASYNC |
+					  CRYPTO_ALG_NEED_FALLBACK,
 		.cra_blocksize		= 1,
 		.cra_ctxsize		= sizeof(struct starfive_cryp_ctx),
 		.cra_alignmask		= 0xf,
@@ -892,7 +1088,7 @@ static struct aead_engine_alg aead_algs[] = {
 	.base.setauthsize		= starfive_aes_gcm_setauthsize,
 	.base.encrypt			= starfive_aes_gcm_encrypt,
 	.base.decrypt			= starfive_aes_gcm_decrypt,
-	.base.init			= starfive_aes_aead_init_tfm,
+	.base.init			= starfive_aes_gcm_init_tfm,
 	.base.exit			= starfive_aes_aead_exit_tfm,
 	.base.ivsize			= GCM_AES_IV_SIZE,
 	.base.maxauthsize		= AES_BLOCK_SIZE,
@@ -900,7 +1096,8 @@ static struct aead_engine_alg aead_algs[] = {
 		.cra_name               = "gcm(aes)",
 		.cra_driver_name        = "starfive-gcm-aes",
 		.cra_priority           = 200,
-		.cra_flags              = CRYPTO_ALG_ASYNC,
+		.cra_flags              = CRYPTO_ALG_ASYNC |
+					  CRYPTO_ALG_NEED_FALLBACK,
 		.cra_blocksize          = 1,
 		.cra_ctxsize            = sizeof(struct starfive_cryp_ctx),
 		.cra_alignmask          = 0xf,
@@ -914,7 +1111,7 @@ static struct aead_engine_alg aead_algs[] = {
 	.base.setauthsize		= starfive_aes_ccm_setauthsize,
 	.base.encrypt			= starfive_aes_ccm_encrypt,
 	.base.decrypt			= starfive_aes_ccm_decrypt,
-	.base.init			= starfive_aes_aead_init_tfm,
+	.base.init			= starfive_aes_ccm_init_tfm,
 	.base.exit			= starfive_aes_aead_exit_tfm,
 	.base.ivsize			= AES_BLOCK_SIZE,
 	.base.maxauthsize		= AES_BLOCK_SIZE,
diff --git a/drivers/crypto/starfive/jh7110-cryp.c b/drivers/crypto/starfive/jh7110-cryp.c
index 425fddf3a8ab..e4dfed7ee0b0 100644
--- a/drivers/crypto/starfive/jh7110-cryp.c
+++ b/drivers/crypto/starfive/jh7110-cryp.c
@@ -89,34 +89,10 @@ static void starfive_dma_cleanup(struct starfive_cryp_dev *cryp)
 	dma_release_channel(cryp->rx);
 }
 
-static irqreturn_t starfive_cryp_irq(int irq, void *priv)
-{
-	u32 status;
-	u32 mask;
-	struct starfive_cryp_dev *cryp = (struct starfive_cryp_dev *)priv;
-
-	mask = readl(cryp->base + STARFIVE_IE_MASK_OFFSET);
-	status = readl(cryp->base + STARFIVE_IE_FLAG_OFFSET);
-	if (status & STARFIVE_IE_FLAG_AES_DONE) {
-		mask |= STARFIVE_IE_MASK_AES_DONE;
-		writel(mask, cryp->base + STARFIVE_IE_MASK_OFFSET);
-		tasklet_schedule(&cryp->aes_done);
-	}
-
-	if (status & STARFIVE_IE_FLAG_HASH_DONE) {
-		mask |= STARFIVE_IE_MASK_HASH_DONE;
-		writel(mask, cryp->base + STARFIVE_IE_MASK_OFFSET);
-		tasklet_schedule(&cryp->hash_done);
-	}
-
-	return IRQ_HANDLED;
-}
-
 static int starfive_cryp_probe(struct platform_device *pdev)
 {
 	struct starfive_cryp_dev *cryp;
 	struct resource *res;
-	int irq;
 	int ret;
 
 	cryp = devm_kzalloc(&pdev->dev, sizeof(*cryp), GFP_KERNEL);
@@ -131,9 +107,6 @@ static int starfive_cryp_probe(struct platform_device *pdev)
 		return dev_err_probe(&pdev->dev, PTR_ERR(cryp->base),
 				     "Error remapping memory for platform device\n");
 
-	tasklet_init(&cryp->aes_done, starfive_aes_done_task, (unsigned long)cryp);
-	tasklet_init(&cryp->hash_done, starfive_hash_done_task, (unsigned long)cryp);
-
 	cryp->phys_base = res->start;
 	cryp->dma_maxburst = 32;
 	cryp->side_chan = side_chan;
@@ -153,16 +126,6 @@ static int starfive_cryp_probe(struct platform_device *pdev)
 		return dev_err_probe(&pdev->dev, PTR_ERR(cryp->rst),
 				     "Error getting hardware reset line\n");
 
-	irq = platform_get_irq(pdev, 0);
-	if (irq < 0)
-		return irq;
-
-	ret = devm_request_irq(&pdev->dev, irq, starfive_cryp_irq, 0, pdev->name,
-			       (void *)cryp);
-	if (ret)
-		return dev_err_probe(&pdev->dev, ret,
-				     "Failed to register interrupt handler\n");
-
 	clk_prepare_enable(cryp->hclk);
 	clk_prepare_enable(cryp->ahb);
 	reset_control_deassert(cryp->rst);
@@ -219,9 +182,6 @@ err_dma_init:
 	clk_disable_unprepare(cryp->ahb);
 	reset_control_assert(cryp->rst);
 
-	tasklet_kill(&cryp->aes_done);
-	tasklet_kill(&cryp->hash_done);
-
 	return ret;
 }
 
@@ -233,9 +193,6 @@ static void starfive_cryp_remove(struct platform_device *pdev)
 	starfive_hash_unregister_algs();
 	starfive_rsa_unregister_algs();
 
-	tasklet_kill(&cryp->aes_done);
-	tasklet_kill(&cryp->hash_done);
-
 	crypto_engine_stop(cryp->engine);
 	crypto_engine_exit(cryp->engine);
 
diff --git a/drivers/crypto/starfive/jh7110-cryp.h b/drivers/crypto/starfive/jh7110-cryp.h
index 6cdf6db5d904..494a74f52706 100644
--- a/drivers/crypto/starfive/jh7110-cryp.h
+++ b/drivers/crypto/starfive/jh7110-cryp.h
@@ -91,6 +91,7 @@ union starfive_hash_csr {
 #define STARFIVE_HASH_KEY_DONE			BIT(13)
 		u32 key_done			:1;
 		u32 key_flag			:1;
+#define STARFIVE_HASH_HMAC_DONE			BIT(15)
 		u32 hmac_done			:1;
 #define STARFIVE_HASH_BUSY			BIT(16)
 		u32 busy			:1;
@@ -168,6 +169,7 @@ struct starfive_cryp_ctx {
 	struct crypto_akcipher			*akcipher_fbk;
 	struct crypto_ahash			*ahash_fbk;
 	struct crypto_aead			*aead_fbk;
+	struct crypto_skcipher			*skcipher_fbk;
 };
 
 struct starfive_cryp_dev {
@@ -185,11 +187,8 @@ struct starfive_cryp_dev {
 	struct dma_chan				*rx;
 	struct dma_slave_config			cfg_in;
 	struct dma_slave_config			cfg_out;
-	struct scatter_walk			in_walk;
-	struct scatter_walk			out_walk;
 	struct crypto_engine			*engine;
-	struct tasklet_struct			aes_done;
-	struct tasklet_struct			hash_done;
+	struct completion			dma_done;
 	size_t					assoclen;
 	size_t					total_in;
 	size_t					total_out;
@@ -236,7 +235,4 @@ void starfive_rsa_unregister_algs(void);
 
 int starfive_aes_register_algs(void);
 void starfive_aes_unregister_algs(void);
-
-void starfive_hash_done_task(unsigned long param);
-void starfive_aes_done_task(unsigned long param);
 #endif
diff --git a/drivers/crypto/starfive/jh7110-hash.c b/drivers/crypto/starfive/jh7110-hash.c
index b6d1808012ca..2c60a1047bc3 100644
--- a/drivers/crypto/starfive/jh7110-hash.c
+++ b/drivers/crypto/starfive/jh7110-hash.c
@@ -36,15 +36,22 @@
 #define STARFIVE_HASH_BUFLEN		SHA512_BLOCK_SIZE
 #define STARFIVE_HASH_RESET		0x2
 
-static inline int starfive_hash_wait_busy(struct starfive_cryp_ctx *ctx)
+static inline int starfive_hash_wait_busy(struct starfive_cryp_dev *cryp)
 {
-	struct starfive_cryp_dev *cryp = ctx->cryp;
 	u32 status;
 
 	return readl_relaxed_poll_timeout(cryp->base + STARFIVE_HASH_SHACSR, status,
 					  !(status & STARFIVE_HASH_BUSY), 10, 100000);
 }
 
+static inline int starfive_hash_wait_hmac_done(struct starfive_cryp_dev *cryp)
+{
+	u32 status;
+
+	return readl_relaxed_poll_timeout(cryp->base + STARFIVE_HASH_SHACSR, status,
+					  (status & STARFIVE_HASH_HMAC_DONE), 10, 100000);
+}
+
 static inline int starfive_hash_wait_key_done(struct starfive_cryp_ctx *ctx)
 {
 	struct starfive_cryp_dev *cryp = ctx->cryp;
@@ -84,64 +91,26 @@ static int starfive_hash_hmac_key(struct starfive_cryp_ctx *ctx)
 	return 0;
 }
 
-static void starfive_hash_start(void *param)
+static void starfive_hash_start(struct starfive_cryp_dev *cryp)
 {
-	struct starfive_cryp_ctx *ctx = param;
-	struct starfive_cryp_request_ctx *rctx = ctx->rctx;
-	struct starfive_cryp_dev *cryp = ctx->cryp;
-	union starfive_alg_cr alg_cr;
 	union starfive_hash_csr csr;
-	u32 stat;
-
-	dma_unmap_sg(cryp->dev, rctx->in_sg, rctx->in_sg_len, DMA_TO_DEVICE);
-
-	alg_cr.v = 0;
-	alg_cr.clear = 1;
-
-	writel(alg_cr.v, cryp->base + STARFIVE_ALG_CR_OFFSET);
 
 	csr.v = readl(cryp->base + STARFIVE_HASH_SHACSR);
 	csr.firstb = 0;
 	csr.final = 1;
-
-	stat = readl(cryp->base + STARFIVE_IE_MASK_OFFSET);
-	stat &= ~STARFIVE_IE_MASK_HASH_DONE;
-	writel(stat, cryp->base + STARFIVE_IE_MASK_OFFSET);
 	writel(csr.v, cryp->base + STARFIVE_HASH_SHACSR);
 }
 
-static int starfive_hash_xmit_dma(struct starfive_cryp_ctx *ctx)
+static void starfive_hash_dma_callback(void *param)
 {
-	struct starfive_cryp_request_ctx *rctx = ctx->rctx;
-	struct starfive_cryp_dev *cryp = ctx->cryp;
-	struct dma_async_tx_descriptor	*in_desc;
-	union  starfive_alg_cr alg_cr;
-	int total_len;
-	int ret;
-
-	if (!rctx->total) {
-		starfive_hash_start(ctx);
-		return 0;
-	}
+	struct starfive_cryp_dev *cryp = param;
 
-	writel(rctx->total, cryp->base + STARFIVE_DMA_IN_LEN_OFFSET);
-
-	total_len = rctx->total;
-	total_len = (total_len & 0x3) ? (((total_len >> 2) + 1) << 2) : total_len;
-	sg_dma_len(rctx->in_sg) = total_len;
-
-	alg_cr.v = 0;
-	alg_cr.start = 1;
-	alg_cr.hash_dma_en = 1;
-
-	writel(alg_cr.v, cryp->base + STARFIVE_ALG_CR_OFFSET);
-
-	ret = dma_map_sg(cryp->dev, rctx->in_sg, rctx->in_sg_len, DMA_TO_DEVICE);
-	if (!ret)
-		return dev_err_probe(cryp->dev, -EINVAL, "dma_map_sg() error\n");
+	complete(&cryp->dma_done);
+}
 
-	cryp->cfg_in.direction = DMA_MEM_TO_DEV;
-	cryp->cfg_in.src_addr_width = DMA_SLAVE_BUSWIDTH_4_BYTES;
+static void starfive_hash_dma_init(struct starfive_cryp_dev *cryp)
+{
+	cryp->cfg_in.src_addr_width = DMA_SLAVE_BUSWIDTH_16_BYTES;
 	cryp->cfg_in.dst_addr_width = DMA_SLAVE_BUSWIDTH_4_BYTES;
 	cryp->cfg_in.src_maxburst = cryp->dma_maxburst;
 	cryp->cfg_in.dst_maxburst = cryp->dma_maxburst;
@@ -149,50 +118,48 @@ static int starfive_hash_xmit_dma(struct starfive_cryp_ctx *ctx)
 
 	dmaengine_slave_config(cryp->tx, &cryp->cfg_in);
 
-	in_desc = dmaengine_prep_slave_sg(cryp->tx, rctx->in_sg,
-					  ret, DMA_MEM_TO_DEV,
-					  DMA_PREP_INTERRUPT  |  DMA_CTRL_ACK);
-
-	if (!in_desc)
-		return -EINVAL;
-
-	in_desc->callback = starfive_hash_start;
-	in_desc->callback_param = ctx;
-
-	dmaengine_submit(in_desc);
-	dma_async_issue_pending(cryp->tx);
-
-	return 0;
+	init_completion(&cryp->dma_done);
 }
 
-static int starfive_hash_xmit(struct starfive_cryp_ctx *ctx)
+static int starfive_hash_dma_xfer(struct starfive_cryp_dev *cryp,
+				  struct scatterlist *sg)
 {
-	struct starfive_cryp_request_ctx *rctx = ctx->rctx;
-	struct starfive_cryp_dev *cryp = ctx->cryp;
+	struct dma_async_tx_descriptor *in_desc;
+	union starfive_alg_cr alg_cr;
 	int ret = 0;
 
-	rctx->csr.hash.v = 0;
-	rctx->csr.hash.reset = 1;
-	writel(rctx->csr.hash.v, cryp->base + STARFIVE_HASH_SHACSR);
-
-	if (starfive_hash_wait_busy(ctx))
-		return dev_err_probe(cryp->dev, -ETIMEDOUT, "Error resetting engine.\n");
+	alg_cr.v = 0;
+	alg_cr.start = 1;
+	alg_cr.hash_dma_en = 1;
+	writel(alg_cr.v, cryp->base + STARFIVE_ALG_CR_OFFSET);
 
-	rctx->csr.hash.v = 0;
-	rctx->csr.hash.mode = ctx->hash_mode;
-	rctx->csr.hash.ie = 1;
+	writel(sg_dma_len(sg), cryp->base + STARFIVE_DMA_IN_LEN_OFFSET);
+	sg_dma_len(sg) = ALIGN(sg_dma_len(sg), sizeof(u32));
 
-	if (ctx->is_hmac) {
-		ret = starfive_hash_hmac_key(ctx);
-		if (ret)
-			return ret;
-	} else {
-		rctx->csr.hash.start = 1;
-		rctx->csr.hash.firstb = 1;
-		writel(rctx->csr.hash.v, cryp->base + STARFIVE_HASH_SHACSR);
+	in_desc = dmaengine_prep_slave_sg(cryp->tx, sg, 1, DMA_MEM_TO_DEV,
+					  DMA_PREP_INTERRUPT | DMA_CTRL_ACK);
+	if (!in_desc) {
+		ret = -EINVAL;
+		goto end;
 	}
 
-	return starfive_hash_xmit_dma(ctx);
+	reinit_completion(&cryp->dma_done);
+	in_desc->callback = starfive_hash_dma_callback;
+	in_desc->callback_param = cryp;
+
+	dmaengine_submit(in_desc);
+	dma_async_issue_pending(cryp->tx);
+
+	if (!wait_for_completion_timeout(&cryp->dma_done,
+					 msecs_to_jiffies(1000)))
+		ret = -ETIMEDOUT;
+
+end:
+	alg_cr.v = 0;
+	alg_cr.clear = 1;
+	writel(alg_cr.v, cryp->base + STARFIVE_ALG_CR_OFFSET);
+
+	return ret;
 }
 
 static int starfive_hash_copy_hash(struct ahash_request *req)
@@ -215,58 +182,74 @@ static int starfive_hash_copy_hash(struct ahash_request *req)
 	return 0;
 }
 
-void starfive_hash_done_task(unsigned long param)
+static void starfive_hash_done_task(struct starfive_cryp_dev *cryp)
 {
-	struct starfive_cryp_dev *cryp = (struct starfive_cryp_dev *)param;
 	int err = cryp->err;
 
 	if (!err)
 		err = starfive_hash_copy_hash(cryp->req.hreq);
 
-	/* Reset to clear hash_done in irq register*/
-	writel(STARFIVE_HASH_RESET, cryp->base + STARFIVE_HASH_SHACSR);
-
 	crypto_finalize_hash_request(cryp->engine, cryp->req.hreq, err);
 }
 
-static int starfive_hash_check_aligned(struct scatterlist *sg, size_t total, size_t align)
+static int starfive_hash_one_request(struct crypto_engine *engine, void *areq)
 {
-	int len = 0;
+	struct ahash_request *req = container_of(areq, struct ahash_request,
+						 base);
+	struct starfive_cryp_ctx *ctx = crypto_ahash_ctx(crypto_ahash_reqtfm(req));
+	struct starfive_cryp_request_ctx *rctx = ctx->rctx;
+	struct starfive_cryp_dev *cryp = ctx->cryp;
+	struct scatterlist *tsg;
+	int ret, src_nents, i;
 
-	if (!total)
-		return 0;
+	writel(STARFIVE_HASH_RESET, cryp->base + STARFIVE_HASH_SHACSR);
 
-	if (!IS_ALIGNED(total, align))
-		return -EINVAL;
+	if (starfive_hash_wait_busy(cryp))
+		return dev_err_probe(cryp->dev, -ETIMEDOUT, "Error resetting hardware\n");
 
-	while (sg) {
-		if (!IS_ALIGNED(sg->offset, sizeof(u32)))
-			return -EINVAL;
+	rctx->csr.hash.v = 0;
+	rctx->csr.hash.mode = ctx->hash_mode;
 
-		if (!IS_ALIGNED(sg->length, align))
-			return -EINVAL;
+	if (ctx->is_hmac) {
+		ret = starfive_hash_hmac_key(ctx);
+		if (ret)
+			return ret;
+	} else {
+		rctx->csr.hash.start = 1;
+		rctx->csr.hash.firstb = 1;
+		writel(rctx->csr.hash.v, cryp->base + STARFIVE_HASH_SHACSR);
+	}
+
+	/* No input message, get digest and end. */
+	if (!rctx->total)
+		goto hash_start;
+
+	starfive_hash_dma_init(cryp);
+
+	for_each_sg(rctx->in_sg, tsg, rctx->in_sg_len, i) {
+		src_nents = dma_map_sg(cryp->dev, tsg, 1, DMA_TO_DEVICE);
+		if (src_nents == 0)
+			return dev_err_probe(cryp->dev, -ENOMEM,
+					     "dma_map_sg error\n");
 
-		len += sg->length;
-		sg = sg_next(sg);
+		ret = starfive_hash_dma_xfer(cryp, tsg);
+		dma_unmap_sg(cryp->dev, tsg, 1, DMA_TO_DEVICE);
+		if (ret)
+			return ret;
 	}
 
-	if (len != total)
-		return -EINVAL;
+hash_start:
+	starfive_hash_start(cryp);
 
-	return 0;
-}
+	if (starfive_hash_wait_busy(cryp))
+		return dev_err_probe(cryp->dev, -ETIMEDOUT, "Error generating digest\n");
 
-static int starfive_hash_one_request(struct crypto_engine *engine, void *areq)
-{
-	struct ahash_request *req = container_of(areq, struct ahash_request,
-						 base);
-	struct starfive_cryp_ctx *ctx = crypto_ahash_ctx(crypto_ahash_reqtfm(req));
-	struct starfive_cryp_dev *cryp = ctx->cryp;
+	if (ctx->is_hmac)
+		cryp->err = starfive_hash_wait_hmac_done(cryp);
 
-	if (!cryp)
-		return -ENODEV;
+	starfive_hash_done_task(cryp);
 
-	return starfive_hash_xmit(ctx);
+	return 0;
 }
 
 static int starfive_hash_init(struct ahash_request *req)
@@ -337,22 +320,6 @@ static int starfive_hash_finup(struct ahash_request *req)
 	return crypto_ahash_finup(&rctx->ahash_fbk_req);
 }
 
-static int starfive_hash_digest_fb(struct ahash_request *req)
-{
-	struct starfive_cryp_request_ctx *rctx = ahash_request_ctx(req);
-	struct crypto_ahash *tfm = crypto_ahash_reqtfm(req);
-	struct starfive_cryp_ctx *ctx = crypto_ahash_ctx(tfm);
-
-	ahash_request_set_tfm(&rctx->ahash_fbk_req, ctx->ahash_fbk);
-	ahash_request_set_callback(&rctx->ahash_fbk_req, req->base.flags,
-				   req->base.complete, req->base.data);
-
-	ahash_request_set_crypt(&rctx->ahash_fbk_req, req->src,
-				req->result, req->nbytes);
-
-	return crypto_ahash_digest(&rctx->ahash_fbk_req);
-}
-
 static int starfive_hash_digest(struct ahash_request *req)
 {
 	struct crypto_ahash *tfm = crypto_ahash_reqtfm(req);
@@ -370,9 +337,6 @@ static int starfive_hash_digest(struct ahash_request *req)
 	rctx->in_sg_len = sg_nents_for_len(rctx->in_sg, rctx->total);
 	ctx->rctx = rctx;
 
-	if (starfive_hash_check_aligned(rctx->in_sg, rctx->total, rctx->blksize))
-		return starfive_hash_digest_fb(req);
-
 	return crypto_transfer_hash_request_to_engine(cryp->engine, req);
 }
 
@@ -406,7 +370,8 @@ static int starfive_hash_import(struct ahash_request *req, const void *in)
 
 static int starfive_hash_init_tfm(struct crypto_ahash *hash,
 				  const char *alg_name,
-				  unsigned int mode)
+				  unsigned int mode,
+				  bool is_hmac)
 {
 	struct starfive_cryp_ctx *ctx = crypto_ahash_ctx(hash);
 
@@ -426,7 +391,7 @@ static int starfive_hash_init_tfm(struct crypto_ahash *hash,
 	crypto_ahash_set_reqsize(hash, sizeof(struct starfive_cryp_request_ctx) +
 				 crypto_ahash_reqsize(ctx->ahash_fbk));
 
-	ctx->keylen = 0;
+	ctx->is_hmac = is_hmac;
 	ctx->hash_mode = mode;
 
 	return 0;
@@ -529,81 +494,61 @@ static int starfive_hash_setkey(struct crypto_ahash *hash,
 static int starfive_sha224_init_tfm(struct crypto_ahash *hash)
 {
 	return starfive_hash_init_tfm(hash, "sha224-generic",
-				      STARFIVE_HASH_SHA224);
+				      STARFIVE_HASH_SHA224, 0);
 }
 
 static int starfive_sha256_init_tfm(struct crypto_ahash *hash)
 {
 	return starfive_hash_init_tfm(hash, "sha256-generic",
-				      STARFIVE_HASH_SHA256);
+				      STARFIVE_HASH_SHA256, 0);
 }
 
 static int starfive_sha384_init_tfm(struct crypto_ahash *hash)
 {
 	return starfive_hash_init_tfm(hash, "sha384-generic",
-				      STARFIVE_HASH_SHA384);
+				      STARFIVE_HASH_SHA384, 0);
 }
 
 static int starfive_sha512_init_tfm(struct crypto_ahash *hash)
 {
 	return starfive_hash_init_tfm(hash, "sha512-generic",
-				      STARFIVE_HASH_SHA512);
+				      STARFIVE_HASH_SHA512, 0);
 }
 
 static int starfive_sm3_init_tfm(struct crypto_ahash *hash)
 {
 	return starfive_hash_init_tfm(hash, "sm3-generic",
-				      STARFIVE_HASH_SM3);
+				      STARFIVE_HASH_SM3, 0);
 }
 
 static int starfive_hmac_sha224_init_tfm(struct crypto_ahash *hash)
 {
-	struct starfive_cryp_ctx *ctx = crypto_ahash_ctx(hash);
-
-	ctx->is_hmac = true;
-
 	return starfive_hash_init_tfm(hash, "hmac(sha224-generic)",
-				      STARFIVE_HASH_SHA224);
+				      STARFIVE_HASH_SHA224, 1);
 }
 
 static int starfive_hmac_sha256_init_tfm(struct crypto_ahash *hash)
 {
-	struct starfive_cryp_ctx *ctx = crypto_ahash_ctx(hash);
-
-	ctx->is_hmac = true;
-
 	return starfive_hash_init_tfm(hash, "hmac(sha256-generic)",
-				      STARFIVE_HASH_SHA256);
+				      STARFIVE_HASH_SHA256, 1);
 }
 
 static int starfive_hmac_sha384_init_tfm(struct crypto_ahash *hash)
 {
-	struct starfive_cryp_ctx *ctx = crypto_ahash_ctx(hash);
-
-	ctx->is_hmac = true;
-
 	return starfive_hash_init_tfm(hash, "hmac(sha384-generic)",
-				      STARFIVE_HASH_SHA384);
+				      STARFIVE_HASH_SHA384, 1);
 }
 
 static int starfive_hmac_sha512_init_tfm(struct crypto_ahash *hash)
 {
-	struct starfive_cryp_ctx *ctx = crypto_ahash_ctx(hash);
-
-	ctx->is_hmac = true;
-
 	return starfive_hash_init_tfm(hash, "hmac(sha512-generic)",
-				      STARFIVE_HASH_SHA512);
+				      STARFIVE_HASH_SHA512, 1);
 }
 
 static int starfive_hmac_sm3_init_tfm(struct crypto_ahash *hash)
 {
-	struct starfive_cryp_ctx *ctx = crypto_ahash_ctx(hash);
-
-	ctx->is_hmac = true;
-
 	return starfive_hash_init_tfm(hash, "hmac(sm3-generic)",
-				      STARFIVE_HASH_SM3);
+				      STARFIVE_HASH_SM3, 1);
 }
 
 static struct ahash_engine_alg algs_sha2_sm3[] = {
diff --git a/drivers/crypto/starfive/jh7110-rsa.c b/drivers/crypto/starfive/jh7110-rsa.c
index cf8bda7f0855..33093ba4b13a 100644
--- a/drivers/crypto/starfive/jh7110-rsa.c
+++ b/drivers/crypto/starfive/jh7110-rsa.c
@@ -45,6 +45,9 @@ static inline int starfive_pka_wait_done(struct starfive_cryp_ctx *ctx)
 
 static void starfive_rsa_free_key(struct starfive_rsa_key *key)
 {
+	if (!key->key_sz)
+		return;
+
 	kfree_sensitive(key->d);
 	kfree_sensitive(key->e);
 	kfree_sensitive(key->n);
@@ -273,7 +276,6 @@ static int starfive_rsa_enc_core(struct starfive_cryp_ctx *ctx, int enc)
 
 err_rsa_crypt:
 	writel(STARFIVE_RSA_RESET, cryp->base + STARFIVE_PKA_CACR_OFFSET);
-	kfree(rctx->rsa_data);
 	return ret;
 }
 
@@ -534,16 +536,14 @@ static int starfive_rsa_init_tfm(struct crypto_akcipher *tfm)
 {
 	struct starfive_cryp_ctx *ctx = akcipher_tfm_ctx(tfm);
 
+	ctx->cryp = starfive_cryp_find_dev(ctx);
+	if (!ctx->cryp)
+		return -ENODEV;
+
 	ctx->akcipher_fbk = crypto_alloc_akcipher("rsa-generic", 0, 0);
 	if (IS_ERR(ctx->akcipher_fbk))
 		return PTR_ERR(ctx->akcipher_fbk);
 
-	ctx->cryp = starfive_cryp_find_dev(ctx);
-	if (!ctx->cryp) {
-		crypto_free_akcipher(ctx->akcipher_fbk);
-		return -ENODEV;
-	}
-
 	akcipher_set_reqsize(tfm, sizeof(struct starfive_cryp_request_ctx) +
 			     sizeof(struct crypto_akcipher) + 32);
 
diff --git a/drivers/crypto/stm32/stm32-hash.c b/drivers/crypto/stm32/stm32-hash.c
index 34e0d7e381a8..351827372ea6 100644
--- a/drivers/crypto/stm32/stm32-hash.c
+++ b/drivers/crypto/stm32/stm32-hash.c
@@ -94,6 +94,7 @@
 #define HASH_FLAGS_ERRORS		BIT(21)
 #define HASH_FLAGS_EMPTY		BIT(22)
 #define HASH_FLAGS_HMAC			BIT(23)
+#define HASH_FLAGS_SGS_COPIED		BIT(24)
 
 #define HASH_OP_UPDATE			1
 #define HASH_OP_FINAL			2
@@ -145,7 +146,7 @@ struct stm32_hash_state {
 	u16			bufcnt;
 	u16			blocklen;
 
-	u8 buffer[HASH_BUFLEN] __aligned(4);
+	u8 buffer[HASH_BUFLEN] __aligned(sizeof(u32));
 
 	/* hash state */
 	u32			hw_context[3 + HASH_CSR_NB_MAX];
@@ -158,8 +159,8 @@ struct stm32_hash_request_ctx {
 	u8 digest[SHA512_DIGEST_SIZE] __aligned(sizeof(u32));
 	size_t			digcnt;
 
-	/* DMA */
 	struct scatterlist	*sg;
+	struct scatterlist	sgl[2]; /* scatterlist used to realize alignment */
 	unsigned int		offset;
 	unsigned int		total;
 	struct scatterlist	sg_key;
@@ -184,6 +185,7 @@ struct stm32_hash_pdata {
 	size_t					algs_info_size;
 	bool					has_sr;
 	bool					has_mdmat;
+	bool					context_secured;
 	bool					broken_emptymsg;
 	bool					ux500;
 };
@@ -195,6 +197,7 @@ struct stm32_hash_dev {
 	struct reset_control	*rst;
 	void __iomem		*io_base;
 	phys_addr_t		phys_base;
+	u8			xmit_buf[HASH_BUFLEN] __aligned(sizeof(u32));
 	u32			dma_mode;
 	bool			polled;
 
@@ -220,6 +223,8 @@ static struct stm32_hash_drv stm32_hash = {
 };
 
 static void stm32_hash_dma_callback(void *param);
+static int stm32_hash_prepare_request(struct ahash_request *req);
+static void stm32_hash_unprepare_request(struct ahash_request *req);
 
 static inline u32 stm32_hash_read(struct stm32_hash_dev *hdev, u32 offset)
 {
@@ -232,6 +237,11 @@ static inline void stm32_hash_write(struct stm32_hash_dev *hdev,
 	writel_relaxed(value, hdev->io_base + offset);
 }
 
+/**
+ * stm32_hash_wait_busy - wait until hash processor is available. It return an
+ * error if the hash core is processing a block of data for more than 10 ms.
+ * @hdev: the stm32_hash_dev device.
+ */
 static inline int stm32_hash_wait_busy(struct stm32_hash_dev *hdev)
 {
 	u32 status;
@@ -245,6 +255,11 @@ static inline int stm32_hash_wait_busy(struct stm32_hash_dev *hdev)
 				   !(status & HASH_SR_BUSY), 10, 10000);
 }
 
+/**
+ * stm32_hash_set_nblw - set the number of valid bytes in the last word.
+ * @hdev: the stm32_hash_dev device.
+ * @length: the length of the final word.
+ */
 static void stm32_hash_set_nblw(struct stm32_hash_dev *hdev, int length)
 {
 	u32 reg;
@@ -282,6 +297,11 @@ static int stm32_hash_write_key(struct stm32_hash_dev *hdev)
 	return 0;
 }
 
+/**
+ * stm32_hash_write_ctrl - Initialize the hash processor, only if
+ * HASH_FLAGS_INIT is set.
+ * @hdev: the stm32_hash_dev device
+ */
 static void stm32_hash_write_ctrl(struct stm32_hash_dev *hdev)
 {
 	struct stm32_hash_request_ctx *rctx = ahash_request_ctx(hdev->req);
@@ -469,9 +489,7 @@ static int stm32_hash_update_cpu(struct stm32_hash_dev *hdev)
 {
 	struct stm32_hash_request_ctx *rctx = ahash_request_ctx(hdev->req);
 	struct stm32_hash_state *state = &rctx->state;
-	u32 *preg = state->hw_context;
 	int bufcnt, err = 0, final;
-	int i, swap_reg;
 
 	dev_dbg(hdev->dev, "%s flags %x\n", __func__, state->flags);
 
@@ -495,34 +513,23 @@ static int stm32_hash_update_cpu(struct stm32_hash_dev *hdev)
 		return stm32_hash_xmit_cpu(hdev, state->buffer, bufcnt, 1);
 	}
 
-	if (!(hdev->flags & HASH_FLAGS_INIT))
-		return 0;
-
-	if (stm32_hash_wait_busy(hdev))
-		return -ETIMEDOUT;
-
-	swap_reg = hash_swap_reg(rctx);
-
-	if (!hdev->pdata->ux500)
-		*preg++ = stm32_hash_read(hdev, HASH_IMR);
-	*preg++ = stm32_hash_read(hdev, HASH_STR);
-	*preg++ = stm32_hash_read(hdev, HASH_CR);
-	for (i = 0; i < swap_reg; i++)
-		*preg++ = stm32_hash_read(hdev, HASH_CSR(i));
-
-	state->flags |= HASH_FLAGS_INIT;
-
 	return err;
 }
 
 static int stm32_hash_xmit_dma(struct stm32_hash_dev *hdev,
-			       struct scatterlist *sg, int length, int mdma)
+			       struct scatterlist *sg, int length, int mdmat)
 {
 	struct dma_async_tx_descriptor *in_desc;
 	dma_cookie_t cookie;
 	u32 reg;
 	int err;
 
+	dev_dbg(hdev->dev, "%s mdmat: %x length: %d\n", __func__, mdmat, length);
+
+	/* do not use dma if there is no data to send */
+	if (length <= 0)
+		return 0;
+
 	in_desc = dmaengine_prep_slave_sg(hdev->dma_lch, sg, 1,
 					  DMA_MEM_TO_DEV, DMA_PREP_INTERRUPT |
 					  DMA_CTRL_ACK);
@@ -535,13 +542,12 @@ static int stm32_hash_xmit_dma(struct stm32_hash_dev *hdev,
 	in_desc->callback = stm32_hash_dma_callback;
 	in_desc->callback_param = hdev;
 
-	hdev->flags |= HASH_FLAGS_FINAL;
 	hdev->flags |= HASH_FLAGS_DMA_ACTIVE;
 
 	reg = stm32_hash_read(hdev, HASH_CR);
 
 	if (hdev->pdata->has_mdmat) {
-		if (mdma)
+		if (mdmat)
 			reg |= HASH_CR_MDMAT;
 		else
 			reg &= ~HASH_CR_MDMAT;
@@ -550,7 +556,6 @@ static int stm32_hash_xmit_dma(struct stm32_hash_dev *hdev,
 
 	stm32_hash_write(hdev, HASH_CR, reg);
 
-	stm32_hash_set_nblw(hdev, length);
 
 	cookie = dmaengine_submit(in_desc);
 	err = dma_submit_error(cookie);
@@ -590,7 +595,7 @@ static int stm32_hash_hmac_dma_send(struct stm32_hash_dev *hdev)
 	struct stm32_hash_ctx *ctx = crypto_ahash_ctx(tfm);
 	int err;
 
-	if (ctx->keylen < rctx->state.blocklen || hdev->dma_mode == 1) {
+	if (ctx->keylen < rctx->state.blocklen || hdev->dma_mode > 0) {
 		err = stm32_hash_write_key(hdev);
 		if (stm32_hash_wait_busy(hdev))
 			return -ETIMEDOUT;
@@ -655,18 +660,20 @@ static int stm32_hash_dma_send(struct stm32_hash_dev *hdev)
 	struct scatterlist sg[1], *tsg;
 	int err = 0, reg, ncp = 0;
 	unsigned int i, len = 0, bufcnt = 0;
+	bool final = hdev->flags & HASH_FLAGS_FINAL;
 	bool is_last = false;
+	u32 last_word;
 
-	rctx->sg = hdev->req->src;
-	rctx->total = hdev->req->nbytes;
+	dev_dbg(hdev->dev, "%s total: %d bufcnt: %d final: %d\n",
+		__func__, rctx->total, rctx->state.bufcnt, final);
 
-	rctx->nents = sg_nents(rctx->sg);
 	if (rctx->nents < 0)
 		return -EINVAL;
 
 	stm32_hash_write_ctrl(hdev);
 
-	if (hdev->flags & HASH_FLAGS_HMAC) {
+	if (hdev->flags & HASH_FLAGS_HMAC && (!(hdev->flags & HASH_FLAGS_HMAC_KEY))) {
+		hdev->flags |= HASH_FLAGS_HMAC_KEY;
 		err = stm32_hash_hmac_dma_send(hdev);
 		if (err != -EINPROGRESS)
 			return err;
@@ -677,22 +684,36 @@ static int stm32_hash_dma_send(struct stm32_hash_dev *hdev)
 		len = sg->length;
 
 		if (sg_is_last(sg) || (bufcnt + sg[0].length) >= rctx->total) {
-			sg->length = rctx->total - bufcnt;
-			is_last = true;
-			if (hdev->dma_mode == 1) {
-				len = (ALIGN(sg->length, 16) - 16);
-
-				ncp = sg_pcopy_to_buffer(
-					rctx->sg, rctx->nents,
-					rctx->state.buffer, sg->length - len,
-					rctx->total - sg->length + len);
-
-				sg->length = len;
+			if (!final) {
+				/* Always manually put the last word of a non-final transfer. */
+				len -= sizeof(u32);
+				sg_pcopy_to_buffer(rctx->sg, rctx->nents, &last_word, 4, len);
+				sg->length -= sizeof(u32);
 			} else {
-				if (!(IS_ALIGNED(sg->length, sizeof(u32)))) {
-					len = sg->length;
-					sg->length = ALIGN(sg->length,
-							   sizeof(u32));
+				/*
+				 * In Multiple DMA mode, DMA must be aborted before the final
+				 * transfer.
+				 */
+				sg->length = rctx->total - bufcnt;
+				if (hdev->dma_mode > 0) {
+					len = (ALIGN(sg->length, 16) - 16);
+
+					ncp = sg_pcopy_to_buffer(rctx->sg, rctx->nents,
+								 rctx->state.buffer,
+								 sg->length - len,
+								 rctx->total - sg->length + len);
+
+					if (!len)
+						break;
+
+					sg->length = len;
+				} else {
+					is_last = true;
+					if (!(IS_ALIGNED(sg->length, sizeof(u32)))) {
+						len = sg->length;
+						sg->length = ALIGN(sg->length,
+								   sizeof(u32));
+					}
 				}
 			}
 		}
@@ -706,43 +727,67 @@ static int stm32_hash_dma_send(struct stm32_hash_dev *hdev)
 
 		err = stm32_hash_xmit_dma(hdev, sg, len, !is_last);
 
+		/* The last word of a non final transfer is sent manually. */
+		if (!final) {
+			stm32_hash_write(hdev, HASH_DIN, last_word);
+			len += sizeof(u32);
+		}
+
+		rctx->total -= len;
+
 		bufcnt += sg[0].length;
 		dma_unmap_sg(hdev->dev, sg, 1, DMA_TO_DEVICE);
 
-		if (err == -ENOMEM)
+		if (err == -ENOMEM || err == -ETIMEDOUT)
 			return err;
 		if (is_last)
 			break;
 	}
 
-	if (hdev->dma_mode == 1) {
-		if (stm32_hash_wait_busy(hdev))
-			return -ETIMEDOUT;
-		reg = stm32_hash_read(hdev, HASH_CR);
-		reg &= ~HASH_CR_DMAE;
-		reg |= HASH_CR_DMAA;
-		stm32_hash_write(hdev, HASH_CR, reg);
+	/*
+	 * When the second last block transfer of 4 words is performed by the DMA,
+	 * the software must set the DMA Abort bit (DMAA) to 1 before completing the
+	 * last transfer of 4 words or less.
+	 */
+	if (final) {
+		if (hdev->dma_mode > 0) {
+			if (stm32_hash_wait_busy(hdev))
+				return -ETIMEDOUT;
+			reg = stm32_hash_read(hdev, HASH_CR);
+			reg &= ~HASH_CR_DMAE;
+			reg |= HASH_CR_DMAA;
+			stm32_hash_write(hdev, HASH_CR, reg);
+
+			if (ncp) {
+				memset(buffer + ncp, 0, 4 - DIV_ROUND_UP(ncp, sizeof(u32)));
+				writesl(hdev->io_base + HASH_DIN, buffer,
+					DIV_ROUND_UP(ncp, sizeof(u32)));
+			}
 
-		if (ncp) {
-			memset(buffer + ncp, 0,
-			       DIV_ROUND_UP(ncp, sizeof(u32)) - ncp);
-			writesl(hdev->io_base + HASH_DIN, buffer,
-				DIV_ROUND_UP(ncp, sizeof(u32)));
+			stm32_hash_set_nblw(hdev, ncp);
+			reg = stm32_hash_read(hdev, HASH_STR);
+			reg |= HASH_STR_DCAL;
+			stm32_hash_write(hdev, HASH_STR, reg);
+			err = -EINPROGRESS;
 		}
-		stm32_hash_set_nblw(hdev, ncp);
-		reg = stm32_hash_read(hdev, HASH_STR);
-		reg |= HASH_STR_DCAL;
-		stm32_hash_write(hdev, HASH_STR, reg);
-		err = -EINPROGRESS;
-	}
 
-	if (hdev->flags & HASH_FLAGS_HMAC) {
-		if (stm32_hash_wait_busy(hdev))
-			return -ETIMEDOUT;
-		err = stm32_hash_hmac_dma_send(hdev);
+		/*
+		 * The hash processor needs the key to be loaded a second time in order
+		 * to process the HMAC.
+		 */
+		if (hdev->flags & HASH_FLAGS_HMAC) {
+			if (stm32_hash_wait_busy(hdev))
+				return -ETIMEDOUT;
+			err = stm32_hash_hmac_dma_send(hdev);
+		}
+
+		return err;
 	}
 
-	return err;
+	if (err != -EINPROGRESS)
+		return err;
+
+	return 0;
 }
 
 static struct stm32_hash_dev *stm32_hash_find_dev(struct stm32_hash_ctx *ctx)
@@ -765,33 +810,6 @@ static struct stm32_hash_dev *stm32_hash_find_dev(struct stm32_hash_ctx *ctx)
 	return hdev;
 }
 
-static bool stm32_hash_dma_aligned_data(struct ahash_request *req)
-{
-	struct scatterlist *sg;
-	struct stm32_hash_request_ctx *rctx = ahash_request_ctx(req);
-	struct stm32_hash_ctx *ctx = crypto_ahash_ctx(crypto_ahash_reqtfm(req));
-	struct stm32_hash_dev *hdev = stm32_hash_find_dev(ctx);
-	int i;
-
-	if (!hdev->dma_lch || req->nbytes <= rctx->state.blocklen)
-		return false;
-
-	if (sg_nents(req->src) > 1) {
-		if (hdev->dma_mode == 1)
-			return false;
-		for_each_sg(req->src, sg, sg_nents(req->src), i) {
-			if ((!IS_ALIGNED(sg->length, sizeof(u32))) &&
-			    (!sg_is_last(sg)))
-				return false;
-		}
-	}
-
-	if (req->src->offset % 4)
-		return false;
-
-	return true;
-}
-
 static int stm32_hash_init(struct ahash_request *req)
 {
 	struct crypto_ahash *tfm = crypto_ahash_reqtfm(req);
@@ -802,8 +820,10 @@ static int stm32_hash_init(struct ahash_request *req)
 	bool sha3_mode = ctx->flags & HASH_FLAGS_SHA3_MODE;
 
 	rctx->hdev = hdev;
+	state->flags = 0;
 
-	state->flags = HASH_FLAGS_CPU;
+	if (!(hdev->dma_lch &&  hdev->pdata->has_mdmat))
+		state->flags |= HASH_FLAGS_CPU;
 
 	if (sha3_mode)
 		state->flags |= HASH_FLAGS_SHA3_MODE;
@@ -857,6 +877,7 @@ static int stm32_hash_init(struct ahash_request *req)
 		dev_err(hdev->dev, "Error, block too large");
 		return -EINVAL;
 	}
+	rctx->nents = 0;
 	rctx->total = 0;
 	rctx->offset = 0;
 	rctx->data_type = HASH_DATA_8_BITS;
@@ -874,6 +895,9 @@ static int stm32_hash_update_req(struct stm32_hash_dev *hdev)
 	struct stm32_hash_request_ctx *rctx = ahash_request_ctx(hdev->req);
 	struct stm32_hash_state *state = &rctx->state;
 
+	dev_dbg(hdev->dev, "update_req: total: %u, digcnt: %zd, final: 0",
+		rctx->total, rctx->digcnt);
+
 	if (!(state->flags & HASH_FLAGS_CPU))
 		return stm32_hash_dma_send(hdev);
 
@@ -887,6 +911,11 @@ static int stm32_hash_final_req(struct stm32_hash_dev *hdev)
 	struct stm32_hash_state *state = &rctx->state;
 	int buflen = state->bufcnt;
 
+	if (!(state->flags & HASH_FLAGS_CPU)) {
+		hdev->flags |= HASH_FLAGS_FINAL;
+		return stm32_hash_dma_send(hdev);
+	}
+
 	if (state->flags & HASH_FLAGS_FINUP)
 		return stm32_hash_update_req(hdev);
 
@@ -968,15 +997,21 @@ static int stm32_hash_finish(struct ahash_request *req)
 static void stm32_hash_finish_req(struct ahash_request *req, int err)
 {
 	struct stm32_hash_request_ctx *rctx = ahash_request_ctx(req);
+	struct stm32_hash_state *state = &rctx->state;
 	struct stm32_hash_dev *hdev = rctx->hdev;
 
+	if (hdev->flags & HASH_FLAGS_DMA_ACTIVE)
+		state->flags |= HASH_FLAGS_DMA_ACTIVE;
+	else
+		state->flags &= ~HASH_FLAGS_DMA_ACTIVE;
+
 	if (!err && (HASH_FLAGS_FINAL & hdev->flags)) {
 		stm32_hash_copy_hash(req);
 		err = stm32_hash_finish(req);
 	}
 
-	pm_runtime_mark_last_busy(hdev->dev);
-	pm_runtime_put_autosuspend(hdev->dev);
+	/* Finalized request mist be unprepared here */
+	stm32_hash_unprepare_request(req);
 
 	crypto_finalize_hash_request(hdev->engine, req, err);
 }
@@ -1006,6 +1041,10 @@ static int stm32_hash_one_request(struct crypto_engine *engine, void *areq)
 
 	pm_runtime_get_sync(hdev->dev);
 
+	err = stm32_hash_prepare_request(req);
+	if (err)
+		return err;
+
 	hdev->req = req;
 	hdev->flags = 0;
 	swap_reg = hash_swap_reg(rctx);
@@ -1030,6 +1069,12 @@ static int stm32_hash_one_request(struct crypto_engine *engine, void *areq)
 		if (state->flags & HASH_FLAGS_HMAC)
 			hdev->flags |= HASH_FLAGS_HMAC |
 				       HASH_FLAGS_HMAC_KEY;
+
+		if (state->flags & HASH_FLAGS_CPU)
+			hdev->flags |= HASH_FLAGS_CPU;
+
+		if (state->flags & HASH_FLAGS_DMA_ACTIVE)
+			hdev->flags |= HASH_FLAGS_DMA_ACTIVE;
 	}
 
 	if (rctx->op == HASH_OP_UPDATE)
@@ -1054,6 +1099,284 @@ static int stm32_hash_one_request(struct crypto_engine *engine, void *areq)
 	return 0;
 }
 
+static int stm32_hash_copy_sgs(struct stm32_hash_request_ctx *rctx,
+			       struct scatterlist *sg, int bs,
+			       unsigned int new_len)
+{
+	struct stm32_hash_state *state = &rctx->state;
+	int pages;
+	void *buf;
+
+	pages = get_order(new_len);
+
+	buf = (void *)__get_free_pages(GFP_ATOMIC, pages);
+	if (!buf) {
+		pr_err("Couldn't allocate pages for unaligned cases.\n");
+		return -ENOMEM;
+	}
+
+	if (state->bufcnt)
+		memcpy(buf, rctx->hdev->xmit_buf, state->bufcnt);
+
+	scatterwalk_map_and_copy(buf + state->bufcnt, sg, rctx->offset,
+				 min(new_len, rctx->total) - state->bufcnt, 0);
+	sg_init_table(rctx->sgl, 1);
+	sg_set_buf(rctx->sgl, buf, new_len);
+	rctx->sg = rctx->sgl;
+	state->flags |= HASH_FLAGS_SGS_COPIED;
+	rctx->nents = 1;
+	rctx->offset += new_len - state->bufcnt;
+	state->bufcnt = 0;
+	rctx->total = new_len;
+
+	return 0;
+}
+
+static int stm32_hash_align_sgs(struct scatterlist *sg,
+				int nbytes, int bs, bool init, bool final,
+				struct stm32_hash_request_ctx *rctx)
+{
+	struct stm32_hash_state *state = &rctx->state;
+	struct stm32_hash_dev *hdev = rctx->hdev;
+	struct scatterlist *sg_tmp = sg;
+	int offset = rctx->offset;
+	int new_len;
+	int n = 0;
+	int bufcnt = state->bufcnt;
+	bool secure_ctx = hdev->pdata->context_secured;
+	bool aligned = true;
+
+	if (!sg || !sg->length || !nbytes) {
+		if (bufcnt) {
+			bufcnt = DIV_ROUND_UP(bufcnt, bs) * bs;
+			sg_init_table(rctx->sgl, 1);
+			sg_set_buf(rctx->sgl, rctx->hdev->xmit_buf, bufcnt);
+			rctx->sg = rctx->sgl;
+			rctx->nents = 1;
+		}
+
+		return 0;
+	}
+
+	new_len = nbytes;
+
+	if (offset)
+		aligned = false;
+
+	if (final) {
+		new_len = DIV_ROUND_UP(new_len, bs) * bs;
+	} else {
+		new_len = (new_len - 1) / bs * bs; // return n block - 1 block
+
+		/*
+		 * Context save in some version of HASH IP can only be done when the
+		 * FIFO is ready to get a new block. This implies to send n block plus a
+		 * 32 bit word in the first DMA send.
+		 */
+		if (init && secure_ctx) {
+			new_len += sizeof(u32);
+			if (unlikely(new_len > nbytes))
+				new_len -= bs;
+		}
+	}
+
+	if (!new_len)
+		return 0;
+
+	if (nbytes != new_len)
+		aligned = false;
+
+	while (nbytes > 0 && sg_tmp) {
+		n++;
+
+		if (bufcnt) {
+			if (!IS_ALIGNED(bufcnt, bs)) {
+				aligned = false;
+				break;
+			}
+			nbytes -= bufcnt;
+			bufcnt = 0;
+			if (!nbytes)
+				aligned = false;
+
+			continue;
+		}
+
+		if (offset < sg_tmp->length) {
+			if (!IS_ALIGNED(offset + sg_tmp->offset, 4)) {
+				aligned = false;
+				break;
+			}
+
+			if (!IS_ALIGNED(sg_tmp->length - offset, bs)) {
+				aligned = false;
+				break;
+			}
+		}
+
+		if (offset) {
+			offset -= sg_tmp->length;
+			if (offset < 0) {
+				nbytes += offset;
+				offset = 0;
+			}
+		} else {
+			nbytes -= sg_tmp->length;
+		}
+
+		sg_tmp = sg_next(sg_tmp);
+
+		if (nbytes < 0) {
+			aligned = false;
+			break;
+		}
+	}
+
+	if (!aligned)
+		return stm32_hash_copy_sgs(rctx, sg, bs, new_len);
+
+	rctx->total = new_len;
+	rctx->offset += new_len;
+	rctx->nents = n;
+	if (state->bufcnt) {
+		sg_init_table(rctx->sgl, 2);
+		sg_set_buf(rctx->sgl, rctx->hdev->xmit_buf, state->bufcnt);
+		sg_chain(rctx->sgl, 2, sg);
+		rctx->sg = rctx->sgl;
+	} else {
+		rctx->sg = sg;
+	}
+
+	return 0;
+}
+
+static int stm32_hash_prepare_request(struct ahash_request *req)
+{
+	struct crypto_ahash *tfm = crypto_ahash_reqtfm(req);
+	struct stm32_hash_ctx *ctx = crypto_ahash_ctx(tfm);
+	struct stm32_hash_request_ctx *rctx = ahash_request_ctx(req);
+	struct stm32_hash_dev *hdev = stm32_hash_find_dev(ctx);
+	struct stm32_hash_state *state = &rctx->state;
+	unsigned int nbytes;
+	int ret, hash_later, bs;
+	bool update = rctx->op & HASH_OP_UPDATE;
+	bool init = !(state->flags & HASH_FLAGS_INIT);
+	bool finup = state->flags & HASH_FLAGS_FINUP;
+	bool final = state->flags & HASH_FLAGS_FINAL;
+
+	if (!hdev->dma_lch || state->flags & HASH_FLAGS_CPU)
+		return 0;
+
+	bs = crypto_ahash_blocksize(tfm);
+
+	nbytes = state->bufcnt;
+
+	/*
+	 * In case of update request nbytes must correspond to the content of the
+	 * buffer + the offset minus the content of the request already in the
+	 * buffer.
+	 */
+	if (update || finup)
+		nbytes += req->nbytes - rctx->offset;
+
+	dev_dbg(hdev->dev,
+		"%s: nbytes=%d, bs=%d, total=%d, offset=%d, bufcnt=%d\n",
+		__func__, nbytes, bs, rctx->total, rctx->offset, state->bufcnt);
+
+	if (!nbytes)
+		return 0;
+
+	rctx->total = nbytes;
+
+	if (update && req->nbytes && (!IS_ALIGNED(state->bufcnt, bs))) {
+		int len = bs - state->bufcnt % bs;
+
+		if (len > req->nbytes)
+			len = req->nbytes;
+		scatterwalk_map_and_copy(state->buffer + state->bufcnt, req->src,
+					 0, len, 0);
+		state->bufcnt += len;
+		rctx->offset = len;
+	}
+
+	/* copy buffer in a temporary one that is used for sg alignment */
+	if (state->bufcnt)
+		memcpy(hdev->xmit_buf, state->buffer, state->bufcnt);
+
+	ret = stm32_hash_align_sgs(req->src, nbytes, bs, init, final, rctx);
+	if (ret)
+		return ret;
+
+	hash_later = nbytes - rctx->total;
+	if (hash_later < 0)
+		hash_later = 0;
+
+	if (hash_later && hash_later <= state->blocklen) {
+		scatterwalk_map_and_copy(state->buffer,
+					 req->src,
+					 req->nbytes - hash_later,
+					 hash_later, 0);
+
+		state->bufcnt = hash_later;
+	} else {
+		state->bufcnt = 0;
+	}
+
+	if (hash_later > state->blocklen) {
+		/* FIXME: add support of this case */
+		pr_err("Buffer contains more than one block.\n");
+		return -ENOMEM;
+	}
+
+	rctx->total = min(nbytes, rctx->total);
+
+	return 0;
+}
+
+static void stm32_hash_unprepare_request(struct ahash_request *req)
+{
+	struct stm32_hash_request_ctx *rctx = ahash_request_ctx(req);
+	struct stm32_hash_state *state = &rctx->state;
+	struct stm32_hash_ctx *ctx = crypto_ahash_ctx(crypto_ahash_reqtfm(req));
+	struct stm32_hash_dev *hdev = stm32_hash_find_dev(ctx);
+	u32 *preg = state->hw_context;
+	int swap_reg, i;
+
+	if (hdev->dma_lch)
+		dmaengine_terminate_sync(hdev->dma_lch);
+
+	if (state->flags & HASH_FLAGS_SGS_COPIED)
+		free_pages((unsigned long)sg_virt(rctx->sg), get_order(rctx->sg->length));
+
+	rctx->sg = NULL;
+	rctx->offset = 0;
+
+	state->flags &= ~(HASH_FLAGS_SGS_COPIED);
+
+	if (!(hdev->flags & HASH_FLAGS_INIT))
+		goto pm_runtime;
+
+	state->flags |= HASH_FLAGS_INIT;
+
+	if (stm32_hash_wait_busy(hdev)) {
+		dev_warn(hdev->dev, "Wait busy failed.");
+		return;
+	}
+
+	swap_reg = hash_swap_reg(rctx);
+
+	if (!hdev->pdata->ux500)
+		*preg++ = stm32_hash_read(hdev, HASH_IMR);
+	*preg++ = stm32_hash_read(hdev, HASH_STR);
+	*preg++ = stm32_hash_read(hdev, HASH_CR);
+	for (i = 0; i < swap_reg; i++)
+		*preg++ = stm32_hash_read(hdev, HASH_CSR(i));
+
+pm_runtime:
+	pm_runtime_mark_last_busy(hdev->dev);
+	pm_runtime_put_autosuspend(hdev->dev);
+}
+
 static int stm32_hash_enqueue(struct ahash_request *req, unsigned int op)
 {
 	struct stm32_hash_request_ctx *rctx = ahash_request_ctx(req);
@@ -1070,16 +1393,26 @@ static int stm32_hash_update(struct ahash_request *req)
 	struct stm32_hash_request_ctx *rctx = ahash_request_ctx(req);
 	struct stm32_hash_state *state = &rctx->state;
 
-	if (!req->nbytes || !(state->flags & HASH_FLAGS_CPU))
+	if (!req->nbytes)
 		return 0;
 
-	rctx->total = req->nbytes;
-	rctx->sg = req->src;
-	rctx->offset = 0;
 
-	if ((state->bufcnt + rctx->total < state->blocklen)) {
-		stm32_hash_append_sg(rctx);
-		return 0;
+	if (state->flags & HASH_FLAGS_CPU) {
+		rctx->total = req->nbytes;
+		rctx->sg = req->src;
+		rctx->offset = 0;
+
+		if ((state->bufcnt + rctx->total < state->blocklen)) {
+			stm32_hash_append_sg(rctx);
+			return 0;
+		}
+	} else { /* DMA mode */
+		if (state->bufcnt + req->nbytes <= state->blocklen) {
+			scatterwalk_map_and_copy(state->buffer + state->bufcnt, req->src,
+						 0, req->nbytes, 0);
+			state->bufcnt += req->nbytes;
+			return 0;
+		}
 	}
 
 	return stm32_hash_enqueue(req, HASH_OP_UPDATE);
@@ -1098,20 +1431,18 @@ static int stm32_hash_final(struct ahash_request *req)
 static int stm32_hash_finup(struct ahash_request *req)
 {
 	struct stm32_hash_request_ctx *rctx = ahash_request_ctx(req);
-	struct stm32_hash_ctx *ctx = crypto_ahash_ctx(crypto_ahash_reqtfm(req));
-	struct stm32_hash_dev *hdev = stm32_hash_find_dev(ctx);
 	struct stm32_hash_state *state = &rctx->state;
 
 	if (!req->nbytes)
 		goto out;
 
 	state->flags |= HASH_FLAGS_FINUP;
-	rctx->total = req->nbytes;
-	rctx->sg = req->src;
-	rctx->offset = 0;
 
-	if (hdev->dma_lch && stm32_hash_dma_aligned_data(req))
-		state->flags &= ~HASH_FLAGS_CPU;
+	if ((state->flags & HASH_FLAGS_CPU)) {
+		rctx->total = req->nbytes;
+		rctx->sg = req->src;
+		rctx->offset = 0;
+	}
 
 out:
 	return stm32_hash_final(req);
@@ -1215,7 +1546,6 @@ static int stm32_hash_cra_sha3_hmac_init(struct crypto_tfm *tfm)
 					HASH_FLAGS_HMAC);
 }
 
-
 static void stm32_hash_cra_exit(struct crypto_tfm *tfm)
 {
 	struct stm32_hash_ctx *ctx = crypto_tfm_ctx(tfm);
@@ -1228,14 +1558,9 @@ static irqreturn_t stm32_hash_irq_thread(int irq, void *dev_id)
 {
 	struct stm32_hash_dev *hdev = dev_id;
 
-	if (HASH_FLAGS_CPU & hdev->flags) {
-		if (HASH_FLAGS_OUTPUT_READY & hdev->flags) {
-			hdev->flags &= ~HASH_FLAGS_OUTPUT_READY;
-			goto finish;
-		}
-	} else if (HASH_FLAGS_DMA_ACTIVE & hdev->flags) {
-		hdev->flags &= ~HASH_FLAGS_DMA_ACTIVE;
-			goto finish;
+	if (HASH_FLAGS_OUTPUT_READY & hdev->flags) {
+		hdev->flags &= ~HASH_FLAGS_OUTPUT_READY;
+		goto finish;
 	}
 
 	return IRQ_HANDLED;
@@ -1984,6 +2309,7 @@ static const struct stm32_hash_pdata stm32_hash_pdata_stm32mp13 = {
 	.algs_info_size	= ARRAY_SIZE(stm32_hash_algs_info_stm32mp13),
 	.has_sr		= true,
 	.has_mdmat	= true,
+	.context_secured = true,
 };
 
 static const struct of_device_id stm32_hash_of_match[] = {
diff --git a/drivers/crypto/tegra/Makefile b/drivers/crypto/tegra/Makefile
new file mode 100644
index 000000000000..a32001e58eb2
--- /dev/null
+++ b/drivers/crypto/tegra/Makefile
@@ -0,0 +1,9 @@
+// SPDX-License-Identifier: GPL-2.0-only
+// SPDX-FileCopyrightText: Copyright (c) 2023 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
+
+tegra-se-objs := tegra-se-key.o tegra-se-main.o
+
+tegra-se-y += tegra-se-aes.o
+tegra-se-y += tegra-se-hash.o
+
+obj-$(CONFIG_CRYPTO_DEV_TEGRA) += tegra-se.o
diff --git a/drivers/crypto/tegra/tegra-se-aes.c b/drivers/crypto/tegra/tegra-se-aes.c
new file mode 100644
index 000000000000..ae7a0f8435fc
--- /dev/null
+++ b/drivers/crypto/tegra/tegra-se-aes.c
@@ -0,0 +1,1933 @@
+// SPDX-License-Identifier: GPL-2.0-only
+// SPDX-FileCopyrightText: Copyright (c) 2023 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
+/*
+ * Crypto driver to handle block cipher algorithms using NVIDIA Security Engine.
+ */
+
+#include <linux/clk.h>
+#include <linux/dma-mapping.h>
+#include <linux/module.h>
+#include <linux/of_device.h>
+#include <linux/platform_device.h>
+
+#include <crypto/aead.h>
+#include <crypto/aes.h>
+#include <crypto/engine.h>
+#include <crypto/gcm.h>
+#include <crypto/scatterwalk.h>
+#include <crypto/xts.h>
+#include <crypto/internal/aead.h>
+#include <crypto/internal/hash.h>
+#include <crypto/internal/skcipher.h>
+
+#include "tegra-se.h"
+
+struct tegra_aes_ctx {
+	struct tegra_se *se;
+	u32 alg;
+	u32 ivsize;
+	u32 key1_id;
+	u32 key2_id;
+};
+
+struct tegra_aes_reqctx {
+	struct tegra_se_datbuf datbuf;
+	bool encrypt;
+	u32 config;
+	u32 crypto_config;
+	u32 len;
+	u32 *iv;
+};
+
+struct tegra_aead_ctx {
+	struct tegra_se *se;
+	unsigned int authsize;
+	u32 alg;
+	u32 keylen;
+	u32 key_id;
+};
+
+struct tegra_aead_reqctx {
+	struct tegra_se_datbuf inbuf;
+	struct tegra_se_datbuf outbuf;
+	struct scatterlist *src_sg;
+	struct scatterlist *dst_sg;
+	unsigned int assoclen;
+	unsigned int cryptlen;
+	unsigned int authsize;
+	bool encrypt;
+	u32 config;
+	u32 crypto_config;
+	u32 key_id;
+	u32 iv[4];
+	u8 authdata[16];
+};
+
+struct tegra_cmac_ctx {
+	struct tegra_se *se;
+	unsigned int alg;
+	u32 key_id;
+	struct crypto_shash *fallback_tfm;
+};
+
+struct tegra_cmac_reqctx {
+	struct scatterlist *src_sg;
+	struct tegra_se_datbuf datbuf;
+	struct tegra_se_datbuf residue;
+	unsigned int total_len;
+	unsigned int blk_size;
+	unsigned int task;
+	u32 crypto_config;
+	u32 config;
+	u32 key_id;
+	u32 *iv;
+	u32 result[CMAC_RESULT_REG_COUNT];
+};
+
+/* increment counter (128-bit int) */
+static void ctr_iv_inc(__u8 *counter, __u8 bits, __u32 nums)
+{
+	do {
+		--bits;
+		nums += counter[bits];
+		counter[bits] = nums & 0xff;
+		nums >>= 8;
+	} while (bits && nums);
+}
+
+static void tegra_cbc_iv_copyback(struct skcipher_request *req, struct tegra_aes_ctx *ctx)
+{
+	struct tegra_aes_reqctx *rctx = skcipher_request_ctx(req);
+	unsigned int offset;
+
+	offset = req->cryptlen - ctx->ivsize;
+
+	if (rctx->encrypt)
+		memcpy(req->iv, rctx->datbuf.buf + offset, ctx->ivsize);
+	else
+		scatterwalk_map_and_copy(req->iv, req->src, offset, ctx->ivsize, 0);
+}
+
+static void tegra_aes_update_iv(struct skcipher_request *req, struct tegra_aes_ctx *ctx)
+{
+	int num;
+
+	if (ctx->alg == SE_ALG_CBC) {
+		tegra_cbc_iv_copyback(req, ctx);
+	} else if (ctx->alg == SE_ALG_CTR) {
+		num = req->cryptlen / ctx->ivsize;
+		if (req->cryptlen % ctx->ivsize)
+			num++;
+
+		ctr_iv_inc(req->iv, ctx->ivsize, num);
+	}
+}
+
+static int tegra234_aes_crypto_cfg(u32 alg, bool encrypt)
+{
+	switch (alg) {
+	case SE_ALG_CMAC:
+	case SE_ALG_GMAC:
+	case SE_ALG_GCM:
+	case SE_ALG_GCM_FINAL:
+		return 0;
+	case SE_ALG_CBC:
+		if (encrypt)
+			return SE_CRYPTO_CFG_CBC_ENCRYPT;
+		else
+			return SE_CRYPTO_CFG_CBC_DECRYPT;
+	case SE_ALG_ECB:
+		if (encrypt)
+			return SE_CRYPTO_CFG_ECB_ENCRYPT;
+		else
+			return SE_CRYPTO_CFG_ECB_DECRYPT;
+	case SE_ALG_XTS:
+		if (encrypt)
+			return SE_CRYPTO_CFG_XTS_ENCRYPT;
+		else
+			return SE_CRYPTO_CFG_XTS_DECRYPT;
+
+	case SE_ALG_CTR:
+		return SE_CRYPTO_CFG_CTR;
+	case SE_ALG_CBC_MAC:
+		return SE_CRYPTO_CFG_CBC_MAC;
+
+	default:
+		break;
+	}
+
+	return -EINVAL;
+}
+
+static int tegra234_aes_cfg(u32 alg, bool encrypt)
+{
+	switch (alg) {
+	case SE_ALG_CBC:
+	case SE_ALG_ECB:
+	case SE_ALG_XTS:
+	case SE_ALG_CTR:
+		if (encrypt)
+			return SE_CFG_AES_ENCRYPT;
+		else
+			return SE_CFG_AES_DECRYPT;
+
+	case SE_ALG_GMAC:
+		if (encrypt)
+			return SE_CFG_GMAC_ENCRYPT;
+		else
+			return SE_CFG_GMAC_DECRYPT;
+
+	case SE_ALG_GCM:
+		if (encrypt)
+			return SE_CFG_GCM_ENCRYPT;
+		else
+			return SE_CFG_GCM_DECRYPT;
+
+	case SE_ALG_GCM_FINAL:
+		if (encrypt)
+			return SE_CFG_GCM_FINAL_ENCRYPT;
+		else
+			return SE_CFG_GCM_FINAL_DECRYPT;
+
+	case SE_ALG_CMAC:
+		return SE_CFG_CMAC;
+
+	case SE_ALG_CBC_MAC:
+		return SE_AES_ENC_ALG_AES_ENC |
+		       SE_AES_DST_HASH_REG;
+	}
+	return -EINVAL;
+}
+
+static unsigned int tegra_aes_prep_cmd(struct tegra_aes_ctx *ctx,
+				       struct tegra_aes_reqctx *rctx)
+{
+	unsigned int data_count, res_bits, i = 0, j;
+	struct tegra_se *se = ctx->se;
+	u32 *cpuvaddr = se->cmdbuf->addr;
+	dma_addr_t addr = rctx->datbuf.addr;
+
+	data_count = rctx->len / AES_BLOCK_SIZE;
+	res_bits = (rctx->len % AES_BLOCK_SIZE) * 8;
+
+	/*
+	 * Hardware processes data_count + 1 blocks.
+	 * Reduce 1 block if there is no residue
+	 */
+	if (!res_bits)
+		data_count--;
+
+	if (rctx->iv) {
+		cpuvaddr[i++] = host1x_opcode_setpayload(SE_CRYPTO_CTR_REG_COUNT);
+		cpuvaddr[i++] = se_host1x_opcode_incr_w(se->hw->regs->linear_ctr);
+		for (j = 0; j < SE_CRYPTO_CTR_REG_COUNT; j++)
+			cpuvaddr[i++] = rctx->iv[j];
+	}
+
+	cpuvaddr[i++] = se_host1x_opcode_nonincr(se->hw->regs->last_blk, 1);
+	cpuvaddr[i++] = SE_LAST_BLOCK_VAL(data_count) |
+			SE_LAST_BLOCK_RES_BITS(res_bits);
+
+	cpuvaddr[i++] = se_host1x_opcode_incr(se->hw->regs->config, 6);
+	cpuvaddr[i++] = rctx->config;
+	cpuvaddr[i++] = rctx->crypto_config;
+
+	/* Source address setting */
+	cpuvaddr[i++] = lower_32_bits(addr);
+	cpuvaddr[i++] = SE_ADDR_HI_MSB(upper_32_bits(addr)) | SE_ADDR_HI_SZ(rctx->len);
+
+	/* Destination address setting */
+	cpuvaddr[i++] = lower_32_bits(addr);
+	cpuvaddr[i++] = SE_ADDR_HI_MSB(upper_32_bits(addr)) |
+			SE_ADDR_HI_SZ(rctx->len);
+
+	cpuvaddr[i++] = se_host1x_opcode_nonincr(se->hw->regs->op, 1);
+	cpuvaddr[i++] = SE_AES_OP_WRSTALL | SE_AES_OP_LASTBUF |
+			SE_AES_OP_START;
+
+	cpuvaddr[i++] = se_host1x_opcode_nonincr(host1x_uclass_incr_syncpt_r(), 1);
+	cpuvaddr[i++] = host1x_uclass_incr_syncpt_cond_f(1) |
+			host1x_uclass_incr_syncpt_indx_f(se->syncpt_id);
+
+	dev_dbg(se->dev, "cfg %#x crypto cfg %#x\n", rctx->config, rctx->crypto_config);
+
+	return i;
+}
+
+static int tegra_aes_do_one_req(struct crypto_engine *engine, void *areq)
+{
+	struct skcipher_request *req = container_of(areq, struct skcipher_request, base);
+	struct tegra_aes_ctx *ctx = crypto_skcipher_ctx(crypto_skcipher_reqtfm(req));
+	struct tegra_aes_reqctx *rctx = skcipher_request_ctx(req);
+	struct tegra_se *se = ctx->se;
+	unsigned int cmdlen;
+	int ret;
+
+	rctx->datbuf.buf = dma_alloc_coherent(se->dev, SE_AES_BUFLEN,
+					      &rctx->datbuf.addr, GFP_KERNEL);
+	if (!rctx->datbuf.buf)
+		return -ENOMEM;
+
+	rctx->datbuf.size = SE_AES_BUFLEN;
+	rctx->iv = (u32 *)req->iv;
+	rctx->len = req->cryptlen;
+
+	/* Pad input to AES Block size */
+	if (ctx->alg != SE_ALG_XTS) {
+		if (rctx->len % AES_BLOCK_SIZE)
+			rctx->len += AES_BLOCK_SIZE - (rctx->len % AES_BLOCK_SIZE);
+	}
+
+	scatterwalk_map_and_copy(rctx->datbuf.buf, req->src, 0, req->cryptlen, 0);
+
+	/* Prepare the command and submit for execution */
+	cmdlen = tegra_aes_prep_cmd(ctx, rctx);
+	ret = tegra_se_host1x_submit(se, cmdlen);
+
+	/* Copy the result */
+	tegra_aes_update_iv(req, ctx);
+	scatterwalk_map_and_copy(rctx->datbuf.buf, req->dst, 0, req->cryptlen, 1);
+
+	/* Free the buffer */
+	dma_free_coherent(ctx->se->dev, SE_AES_BUFLEN,
+			  rctx->datbuf.buf, rctx->datbuf.addr);
+
+	crypto_finalize_skcipher_request(se->engine, req, ret);
+
+	return 0;
+}
+
+static int tegra_aes_cra_init(struct crypto_skcipher *tfm)
+{
+	struct tegra_aes_ctx *ctx = crypto_skcipher_ctx(tfm);
+	struct skcipher_alg *alg = crypto_skcipher_alg(tfm);
+	struct tegra_se_alg *se_alg;
+	const char *algname;
+	int ret;
+
+	se_alg = container_of(alg, struct tegra_se_alg, alg.skcipher.base);
+
+	crypto_skcipher_set_reqsize(tfm, sizeof(struct tegra_aes_reqctx));
+
+	ctx->ivsize = crypto_skcipher_ivsize(tfm);
+	ctx->se = se_alg->se_dev;
+	ctx->key1_id = 0;
+	ctx->key2_id = 0;
+
+	algname = crypto_tfm_alg_name(&tfm->base);
+	ret = se_algname_to_algid(algname);
+	if (ret < 0) {
+		dev_err(ctx->se->dev, "invalid algorithm\n");
+		return ret;
+	}
+
+	ctx->alg = ret;
+
+	return 0;
+}
+
+static void tegra_aes_cra_exit(struct crypto_skcipher *tfm)
+{
+	struct tegra_aes_ctx *ctx = crypto_tfm_ctx(&tfm->base);
+
+	if (ctx->key1_id)
+		tegra_key_invalidate(ctx->se, ctx->key1_id, ctx->alg);
+
+	if (ctx->key2_id)
+		tegra_key_invalidate(ctx->se, ctx->key2_id, ctx->alg);
+}
+
+static int tegra_aes_setkey(struct crypto_skcipher *tfm,
+			    const u8 *key, u32 keylen)
+{
+	struct tegra_aes_ctx *ctx = crypto_skcipher_ctx(tfm);
+
+	if (aes_check_keylen(keylen)) {
+		dev_dbg(ctx->se->dev, "invalid key length (%d)\n", keylen);
+		return -EINVAL;
+	}
+
+	return tegra_key_submit(ctx->se, key, keylen, ctx->alg, &ctx->key1_id);
+}
+
+static int tegra_xts_setkey(struct crypto_skcipher *tfm,
+			    const u8 *key, u32 keylen)
+{
+	struct tegra_aes_ctx *ctx = crypto_skcipher_ctx(tfm);
+	u32 len = keylen / 2;
+	int ret;
+
+	ret = xts_verify_key(tfm, key, keylen);
+	if (ret || aes_check_keylen(len)) {
+		dev_dbg(ctx->se->dev, "invalid key length (%d)\n", keylen);
+		return -EINVAL;
+	}
+
+	ret = tegra_key_submit(ctx->se, key, len,
+			       ctx->alg, &ctx->key1_id);
+	if (ret)
+		return ret;
+
+	return tegra_key_submit(ctx->se, key + len, len,
+			       ctx->alg, &ctx->key2_id);
+
+	return 0;
+}
+
+static int tegra_aes_kac_manifest(u32 user, u32 alg, u32 keylen)
+{
+	int manifest;
+
+	manifest = SE_KAC_USER_NS;
+
+	switch (alg) {
+	case SE_ALG_CBC:
+	case SE_ALG_ECB:
+	case SE_ALG_CTR:
+		manifest |= SE_KAC_ENC;
+		break;
+	case SE_ALG_XTS:
+		manifest |= SE_KAC_XTS;
+		break;
+	case SE_ALG_GCM:
+		manifest |= SE_KAC_GCM;
+		break;
+	case SE_ALG_CMAC:
+		manifest |= SE_KAC_CMAC;
+		break;
+	case SE_ALG_CBC_MAC:
+		manifest |= SE_KAC_ENC;
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	switch (keylen) {
+	case AES_KEYSIZE_128:
+		manifest |= SE_KAC_SIZE_128;
+		break;
+	case AES_KEYSIZE_192:
+		manifest |= SE_KAC_SIZE_192;
+		break;
+	case AES_KEYSIZE_256:
+		manifest |= SE_KAC_SIZE_256;
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	return manifest;
+}
+
+static int tegra_aes_crypt(struct skcipher_request *req, bool encrypt)
+
+{
+	struct crypto_skcipher *tfm;
+	struct tegra_aes_ctx *ctx;
+	struct tegra_aes_reqctx *rctx;
+
+	tfm = crypto_skcipher_reqtfm(req);
+	ctx  = crypto_skcipher_ctx(tfm);
+	rctx = skcipher_request_ctx(req);
+
+	if (ctx->alg != SE_ALG_XTS) {
+		if (!IS_ALIGNED(req->cryptlen, crypto_skcipher_blocksize(tfm))) {
+			dev_dbg(ctx->se->dev, "invalid length (%d)", req->cryptlen);
+			return -EINVAL;
+		}
+	} else if (req->cryptlen < XTS_BLOCK_SIZE) {
+		dev_dbg(ctx->se->dev, "invalid length (%d)", req->cryptlen);
+		return -EINVAL;
+	}
+
+	if (!req->cryptlen)
+		return 0;
+
+	rctx->encrypt = encrypt;
+	rctx->config = tegra234_aes_cfg(ctx->alg, encrypt);
+	rctx->crypto_config = tegra234_aes_crypto_cfg(ctx->alg, encrypt);
+	rctx->crypto_config |= SE_AES_KEY_INDEX(ctx->key1_id);
+
+	if (ctx->key2_id)
+		rctx->crypto_config |= SE_AES_KEY2_INDEX(ctx->key2_id);
+
+	return crypto_transfer_skcipher_request_to_engine(ctx->se->engine, req);
+}
+
+static int tegra_aes_encrypt(struct skcipher_request *req)
+{
+	return tegra_aes_crypt(req, true);
+}
+
+static int tegra_aes_decrypt(struct skcipher_request *req)
+{
+	return tegra_aes_crypt(req, false);
+}
+
+static struct tegra_se_alg tegra_aes_algs[] = {
+	{
+		.alg.skcipher.op.do_one_request	= tegra_aes_do_one_req,
+		.alg.skcipher.base = {
+			.init = tegra_aes_cra_init,
+			.exit = tegra_aes_cra_exit,
+			.setkey	= tegra_aes_setkey,
+			.encrypt = tegra_aes_encrypt,
+			.decrypt = tegra_aes_decrypt,
+			.min_keysize = AES_MIN_KEY_SIZE,
+			.max_keysize = AES_MAX_KEY_SIZE,
+			.ivsize	= AES_BLOCK_SIZE,
+			.base = {
+				.cra_name = "cbc(aes)",
+				.cra_driver_name = "cbc-aes-tegra",
+				.cra_priority = 500,
+				.cra_flags = CRYPTO_ALG_TYPE_SKCIPHER | CRYPTO_ALG_ASYNC,
+				.cra_blocksize = AES_BLOCK_SIZE,
+				.cra_ctxsize = sizeof(struct tegra_aes_ctx),
+				.cra_alignmask = 0xf,
+				.cra_module = THIS_MODULE,
+			},
+		}
+	}, {
+		.alg.skcipher.op.do_one_request	= tegra_aes_do_one_req,
+		.alg.skcipher.base = {
+			.init = tegra_aes_cra_init,
+			.exit = tegra_aes_cra_exit,
+			.setkey	= tegra_aes_setkey,
+			.encrypt = tegra_aes_encrypt,
+			.decrypt = tegra_aes_decrypt,
+			.min_keysize = AES_MIN_KEY_SIZE,
+			.max_keysize = AES_MAX_KEY_SIZE,
+			.base = {
+				.cra_name = "ecb(aes)",
+				.cra_driver_name = "ecb-aes-tegra",
+				.cra_priority = 500,
+				.cra_flags = CRYPTO_ALG_TYPE_SKCIPHER | CRYPTO_ALG_ASYNC,
+				.cra_blocksize = AES_BLOCK_SIZE,
+				.cra_ctxsize = sizeof(struct tegra_aes_ctx),
+				.cra_alignmask = 0xf,
+				.cra_module = THIS_MODULE,
+			},
+		}
+	}, {
+		.alg.skcipher.op.do_one_request	= tegra_aes_do_one_req,
+		.alg.skcipher.base = {
+			.init = tegra_aes_cra_init,
+			.exit = tegra_aes_cra_exit,
+			.setkey = tegra_aes_setkey,
+			.encrypt = tegra_aes_encrypt,
+			.decrypt = tegra_aes_decrypt,
+			.min_keysize = AES_MIN_KEY_SIZE,
+			.max_keysize = AES_MAX_KEY_SIZE,
+			.ivsize	= AES_BLOCK_SIZE,
+			.base = {
+				.cra_name = "ctr(aes)",
+				.cra_driver_name = "ctr-aes-tegra",
+				.cra_priority = 500,
+				.cra_flags = CRYPTO_ALG_TYPE_SKCIPHER | CRYPTO_ALG_ASYNC,
+				.cra_blocksize = 1,
+				.cra_ctxsize = sizeof(struct tegra_aes_ctx),
+				.cra_alignmask = 0xf,
+				.cra_module = THIS_MODULE,
+			},
+		}
+	}, {
+		.alg.skcipher.op.do_one_request	= tegra_aes_do_one_req,
+		.alg.skcipher.base = {
+			.init = tegra_aes_cra_init,
+			.exit = tegra_aes_cra_exit,
+			.setkey	= tegra_xts_setkey,
+			.encrypt = tegra_aes_encrypt,
+			.decrypt = tegra_aes_decrypt,
+			.min_keysize = 2 * AES_MIN_KEY_SIZE,
+			.max_keysize = 2 * AES_MAX_KEY_SIZE,
+			.ivsize	= AES_BLOCK_SIZE,
+			.base = {
+				.cra_name = "xts(aes)",
+				.cra_driver_name = "xts-aes-tegra",
+				.cra_priority = 500,
+				.cra_blocksize = AES_BLOCK_SIZE,
+				.cra_ctxsize	   = sizeof(struct tegra_aes_ctx),
+				.cra_alignmask	   = (__alignof__(u64) - 1),
+				.cra_module	   = THIS_MODULE,
+			},
+		}
+	},
+};
+
+static unsigned int tegra_gmac_prep_cmd(struct tegra_aead_ctx *ctx,
+					struct tegra_aead_reqctx *rctx)
+{
+	unsigned int data_count, res_bits, i = 0;
+	struct tegra_se *se = ctx->se;
+	u32 *cpuvaddr = se->cmdbuf->addr;
+
+	data_count = (rctx->assoclen / AES_BLOCK_SIZE);
+	res_bits = (rctx->assoclen % AES_BLOCK_SIZE) * 8;
+
+	/*
+	 * Hardware processes data_count + 1 blocks.
+	 * Reduce 1 block if there is no residue
+	 */
+	if (!res_bits)
+		data_count--;
+
+	cpuvaddr[i++] = se_host1x_opcode_nonincr(se->hw->regs->last_blk, 1);
+	cpuvaddr[i++] = SE_LAST_BLOCK_VAL(data_count) |
+			SE_LAST_BLOCK_RES_BITS(res_bits);
+
+	cpuvaddr[i++] = se_host1x_opcode_incr(se->hw->regs->config, 4);
+	cpuvaddr[i++] = rctx->config;
+	cpuvaddr[i++] = rctx->crypto_config;
+	cpuvaddr[i++] = lower_32_bits(rctx->inbuf.addr);
+	cpuvaddr[i++] = SE_ADDR_HI_MSB(upper_32_bits(rctx->inbuf.addr)) |
+			SE_ADDR_HI_SZ(rctx->assoclen);
+
+	cpuvaddr[i++] = se_host1x_opcode_nonincr(se->hw->regs->op, 1);
+	cpuvaddr[i++] = SE_AES_OP_WRSTALL | SE_AES_OP_FINAL |
+			SE_AES_OP_INIT | SE_AES_OP_LASTBUF |
+			SE_AES_OP_START;
+
+	cpuvaddr[i++] = se_host1x_opcode_nonincr(host1x_uclass_incr_syncpt_r(), 1);
+	cpuvaddr[i++] = host1x_uclass_incr_syncpt_cond_f(1) |
+			host1x_uclass_incr_syncpt_indx_f(se->syncpt_id);
+
+	return i;
+}
+
+static unsigned int tegra_gcm_crypt_prep_cmd(struct tegra_aead_ctx *ctx,
+					     struct tegra_aead_reqctx *rctx)
+{
+	unsigned int data_count, res_bits, i = 0, j;
+	struct tegra_se *se = ctx->se;
+	u32 *cpuvaddr = se->cmdbuf->addr, op;
+
+	data_count = (rctx->cryptlen / AES_BLOCK_SIZE);
+	res_bits = (rctx->cryptlen % AES_BLOCK_SIZE) * 8;
+	op = SE_AES_OP_WRSTALL | SE_AES_OP_FINAL |
+	     SE_AES_OP_LASTBUF | SE_AES_OP_START;
+
+	/*
+	 * If there is no assoc data,
+	 * this will be the init command
+	 */
+	if (!rctx->assoclen)
+		op |= SE_AES_OP_INIT;
+
+	/*
+	 * Hardware processes data_count + 1 blocks.
+	 * Reduce 1 block if there is no residue
+	 */
+	if (!res_bits)
+		data_count--;
+
+	cpuvaddr[i++] = host1x_opcode_setpayload(SE_CRYPTO_CTR_REG_COUNT);
+	cpuvaddr[i++] = se_host1x_opcode_incr_w(se->hw->regs->linear_ctr);
+	for (j = 0; j < SE_CRYPTO_CTR_REG_COUNT; j++)
+		cpuvaddr[i++] = rctx->iv[j];
+
+	cpuvaddr[i++] = se_host1x_opcode_nonincr(se->hw->regs->last_blk, 1);
+	cpuvaddr[i++] = SE_LAST_BLOCK_VAL(data_count) |
+			SE_LAST_BLOCK_RES_BITS(res_bits);
+
+	cpuvaddr[i++] = se_host1x_opcode_incr(se->hw->regs->config, 6);
+	cpuvaddr[i++] = rctx->config;
+	cpuvaddr[i++] = rctx->crypto_config;
+
+	/* Source Address */
+	cpuvaddr[i++] = lower_32_bits(rctx->inbuf.addr);
+	cpuvaddr[i++] = SE_ADDR_HI_MSB(upper_32_bits(rctx->inbuf.addr)) |
+			SE_ADDR_HI_SZ(rctx->cryptlen);
+
+	/* Destination Address */
+	cpuvaddr[i++] = lower_32_bits(rctx->outbuf.addr);
+	cpuvaddr[i++] = SE_ADDR_HI_MSB(upper_32_bits(rctx->outbuf.addr)) |
+			SE_ADDR_HI_SZ(rctx->cryptlen);
+
+	cpuvaddr[i++] = se_host1x_opcode_nonincr(se->hw->regs->op, 1);
+	cpuvaddr[i++] = op;
+
+	cpuvaddr[i++] = se_host1x_opcode_nonincr(host1x_uclass_incr_syncpt_r(), 1);
+	cpuvaddr[i++] = host1x_uclass_incr_syncpt_cond_f(1) |
+			host1x_uclass_incr_syncpt_indx_f(se->syncpt_id);
+
+	dev_dbg(se->dev, "cfg %#x crypto cfg %#x\n", rctx->config, rctx->crypto_config);
+	return i;
+}
+
+static int tegra_gcm_prep_final_cmd(struct tegra_se *se, u32 *cpuvaddr,
+				    struct tegra_aead_reqctx *rctx)
+{
+	unsigned int i = 0, j;
+	u32 op;
+
+	op = SE_AES_OP_WRSTALL | SE_AES_OP_FINAL |
+	     SE_AES_OP_LASTBUF | SE_AES_OP_START;
+
+	/*
+	 * Set init for zero sized vector
+	 */
+	if (!rctx->assoclen && !rctx->cryptlen)
+		op |= SE_AES_OP_INIT;
+
+	cpuvaddr[i++] = se_host1x_opcode_incr(se->hw->regs->aad_len, 2);
+	cpuvaddr[i++] = rctx->assoclen * 8;
+	cpuvaddr[i++] = 0;
+
+	cpuvaddr[i++] = se_host1x_opcode_incr(se->hw->regs->cryp_msg_len, 2);
+	cpuvaddr[i++] = rctx->cryptlen * 8;
+	cpuvaddr[i++] = 0;
+
+	cpuvaddr[i++] = host1x_opcode_setpayload(SE_CRYPTO_CTR_REG_COUNT);
+	cpuvaddr[i++] = se_host1x_opcode_incr_w(se->hw->regs->linear_ctr);
+	for (j = 0; j < SE_CRYPTO_CTR_REG_COUNT; j++)
+		cpuvaddr[i++] = rctx->iv[j];
+
+	cpuvaddr[i++] = se_host1x_opcode_incr(se->hw->regs->config, 6);
+	cpuvaddr[i++] = rctx->config;
+	cpuvaddr[i++] = rctx->crypto_config;
+	cpuvaddr[i++] = 0;
+	cpuvaddr[i++] = 0;
+
+	/* Destination Address */
+	cpuvaddr[i++] = lower_32_bits(rctx->outbuf.addr);
+	cpuvaddr[i++] = SE_ADDR_HI_MSB(upper_32_bits(rctx->outbuf.addr)) |
+			SE_ADDR_HI_SZ(0x10); /* HW always generates 128-bit tag */
+
+	cpuvaddr[i++] = se_host1x_opcode_nonincr(se->hw->regs->op, 1);
+	cpuvaddr[i++] = op;
+
+	cpuvaddr[i++] = se_host1x_opcode_nonincr(host1x_uclass_incr_syncpt_r(), 1);
+	cpuvaddr[i++] = host1x_uclass_incr_syncpt_cond_f(1) |
+			host1x_uclass_incr_syncpt_indx_f(se->syncpt_id);
+
+	dev_dbg(se->dev, "cfg %#x crypto cfg %#x\n", rctx->config, rctx->crypto_config);
+
+	return i;
+}
+
+static int tegra_gcm_do_gmac(struct tegra_aead_ctx *ctx, struct tegra_aead_reqctx *rctx)
+{
+	struct tegra_se *se = ctx->se;
+	unsigned int cmdlen;
+
+	scatterwalk_map_and_copy(rctx->inbuf.buf,
+				 rctx->src_sg, 0, rctx->assoclen, 0);
+
+	rctx->config = tegra234_aes_cfg(SE_ALG_GMAC, rctx->encrypt);
+	rctx->crypto_config = tegra234_aes_crypto_cfg(SE_ALG_GMAC, rctx->encrypt) |
+			      SE_AES_KEY_INDEX(ctx->key_id);
+
+	cmdlen = tegra_gmac_prep_cmd(ctx, rctx);
+
+	return tegra_se_host1x_submit(se, cmdlen);
+}
+
+static int tegra_gcm_do_crypt(struct tegra_aead_ctx *ctx, struct tegra_aead_reqctx *rctx)
+{
+	struct tegra_se *se = ctx->se;
+	int cmdlen, ret;
+
+	scatterwalk_map_and_copy(rctx->inbuf.buf, rctx->src_sg,
+				 rctx->assoclen, rctx->cryptlen, 0);
+
+	rctx->config = tegra234_aes_cfg(SE_ALG_GCM, rctx->encrypt);
+	rctx->crypto_config = tegra234_aes_crypto_cfg(SE_ALG_GCM, rctx->encrypt) |
+			      SE_AES_KEY_INDEX(ctx->key_id);
+
+	/* Prepare command and submit */
+	cmdlen = tegra_gcm_crypt_prep_cmd(ctx, rctx);
+	ret = tegra_se_host1x_submit(se, cmdlen);
+	if (ret)
+		return ret;
+
+	/* Copy the result */
+	scatterwalk_map_and_copy(rctx->outbuf.buf, rctx->dst_sg,
+				 rctx->assoclen, rctx->cryptlen, 1);
+
+	return 0;
+}
+
+static int tegra_gcm_do_final(struct tegra_aead_ctx *ctx, struct tegra_aead_reqctx *rctx)
+{
+	struct tegra_se *se = ctx->se;
+	u32 *cpuvaddr = se->cmdbuf->addr;
+	int cmdlen, ret, offset;
+
+	rctx->config = tegra234_aes_cfg(SE_ALG_GCM_FINAL, rctx->encrypt);
+	rctx->crypto_config = tegra234_aes_crypto_cfg(SE_ALG_GCM_FINAL, rctx->encrypt) |
+			      SE_AES_KEY_INDEX(ctx->key_id);
+
+	/* Prepare command and submit */
+	cmdlen = tegra_gcm_prep_final_cmd(se, cpuvaddr, rctx);
+	ret = tegra_se_host1x_submit(se, cmdlen);
+	if (ret)
+		return ret;
+
+	if (rctx->encrypt) {
+		/* Copy the result */
+		offset = rctx->assoclen + rctx->cryptlen;
+		scatterwalk_map_and_copy(rctx->outbuf.buf, rctx->dst_sg,
+					 offset, rctx->authsize, 1);
+	}
+
+	return 0;
+}
+
+static int tegra_gcm_do_verify(struct tegra_se *se, struct tegra_aead_reqctx *rctx)
+{
+	unsigned int offset;
+	u8 mac[16];
+
+	offset = rctx->assoclen + rctx->cryptlen;
+	scatterwalk_map_and_copy(mac, rctx->src_sg, offset, rctx->authsize, 0);
+
+	if (crypto_memneq(rctx->outbuf.buf, mac, rctx->authsize))
+		return -EBADMSG;
+
+	return 0;
+}
+
+static inline int tegra_ccm_check_iv(const u8 *iv)
+{
+	/* iv[0] gives value of q-1
+	 * 2 <= q <= 8 as per NIST 800-38C notation
+	 * 2 <= L <= 8, so 1 <= L' <= 7. as per rfc 3610 notation
+	 */
+	if (iv[0] < 1 || iv[0] > 7) {
+		pr_debug("ccm_check_iv failed %d\n", iv[0]);
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+static unsigned int tegra_cbcmac_prep_cmd(struct tegra_aead_ctx *ctx,
+					  struct tegra_aead_reqctx *rctx)
+{
+	unsigned int data_count, i = 0;
+	struct tegra_se *se = ctx->se;
+	u32 *cpuvaddr = se->cmdbuf->addr;
+
+	data_count = (rctx->inbuf.size / AES_BLOCK_SIZE) - 1;
+
+	cpuvaddr[i++] = se_host1x_opcode_nonincr(se->hw->regs->last_blk, 1);
+	cpuvaddr[i++] = SE_LAST_BLOCK_VAL(data_count);
+
+	cpuvaddr[i++] = se_host1x_opcode_incr(se->hw->regs->config, 6);
+	cpuvaddr[i++] = rctx->config;
+	cpuvaddr[i++] = rctx->crypto_config;
+
+	cpuvaddr[i++] = lower_32_bits(rctx->inbuf.addr);
+	cpuvaddr[i++] = SE_ADDR_HI_MSB(upper_32_bits(rctx->inbuf.addr)) |
+			SE_ADDR_HI_SZ(rctx->inbuf.size);
+
+	cpuvaddr[i++] = lower_32_bits(rctx->outbuf.addr);
+	cpuvaddr[i++] = SE_ADDR_HI_MSB(upper_32_bits(rctx->outbuf.addr)) |
+			SE_ADDR_HI_SZ(0x10); /* HW always generates 128 bit tag */
+
+	cpuvaddr[i++] = se_host1x_opcode_nonincr(se->hw->regs->op, 1);
+	cpuvaddr[i++] = SE_AES_OP_WRSTALL |
+			SE_AES_OP_LASTBUF | SE_AES_OP_START;
+
+	cpuvaddr[i++] = se_host1x_opcode_nonincr(host1x_uclass_incr_syncpt_r(), 1);
+	cpuvaddr[i++] = host1x_uclass_incr_syncpt_cond_f(1) |
+			host1x_uclass_incr_syncpt_indx_f(se->syncpt_id);
+
+	return i;
+}
+
+static unsigned int tegra_ctr_prep_cmd(struct tegra_aead_ctx *ctx,
+				       struct tegra_aead_reqctx *rctx)
+{
+	unsigned int i = 0, j;
+	struct tegra_se *se = ctx->se;
+	u32 *cpuvaddr = se->cmdbuf->addr;
+
+	cpuvaddr[i++] = host1x_opcode_setpayload(SE_CRYPTO_CTR_REG_COUNT);
+	cpuvaddr[i++] = se_host1x_opcode_incr_w(se->hw->regs->linear_ctr);
+	for (j = 0; j < SE_CRYPTO_CTR_REG_COUNT; j++)
+		cpuvaddr[i++] = rctx->iv[j];
+
+	cpuvaddr[i++] = se_host1x_opcode_nonincr(se->hw->regs->last_blk, 1);
+	cpuvaddr[i++] = (rctx->inbuf.size / AES_BLOCK_SIZE) - 1;
+	cpuvaddr[i++] = se_host1x_opcode_incr(se->hw->regs->config, 6);
+	cpuvaddr[i++] = rctx->config;
+	cpuvaddr[i++] = rctx->crypto_config;
+
+	/* Source address setting */
+	cpuvaddr[i++] = lower_32_bits(rctx->inbuf.addr);
+	cpuvaddr[i++] = SE_ADDR_HI_MSB(upper_32_bits(rctx->inbuf.addr)) |
+			SE_ADDR_HI_SZ(rctx->inbuf.size);
+
+	/* Destination address setting */
+	cpuvaddr[i++] = lower_32_bits(rctx->outbuf.addr);
+	cpuvaddr[i++] = SE_ADDR_HI_MSB(upper_32_bits(rctx->outbuf.addr)) |
+			SE_ADDR_HI_SZ(rctx->inbuf.size);
+
+	cpuvaddr[i++] = se_host1x_opcode_nonincr(se->hw->regs->op, 1);
+	cpuvaddr[i++] = SE_AES_OP_WRSTALL | SE_AES_OP_LASTBUF |
+			SE_AES_OP_START;
+
+	cpuvaddr[i++] = se_host1x_opcode_nonincr(host1x_uclass_incr_syncpt_r(), 1);
+	cpuvaddr[i++] = host1x_uclass_incr_syncpt_cond_f(1) |
+			host1x_uclass_incr_syncpt_indx_f(se->syncpt_id);
+
+	dev_dbg(se->dev, "cfg %#x crypto cfg %#x\n",
+		rctx->config, rctx->crypto_config);
+
+	return i;
+}
+
+static int tegra_ccm_do_cbcmac(struct tegra_aead_ctx *ctx, struct tegra_aead_reqctx *rctx)
+{
+	struct tegra_se *se = ctx->se;
+	int cmdlen;
+
+	rctx->config = tegra234_aes_cfg(SE_ALG_CBC_MAC, rctx->encrypt);
+	rctx->crypto_config = tegra234_aes_crypto_cfg(SE_ALG_CBC_MAC,
+						      rctx->encrypt) |
+						      SE_AES_KEY_INDEX(ctx->key_id);
+
+	/* Prepare command and submit */
+	cmdlen = tegra_cbcmac_prep_cmd(ctx, rctx);
+
+	return tegra_se_host1x_submit(se, cmdlen);
+}
+
+static int tegra_ccm_set_msg_len(u8 *block, unsigned int msglen, int csize)
+{
+	__be32 data;
+
+	memset(block, 0, csize);
+	block += csize;
+
+	if (csize >= 4)
+		csize = 4;
+	else if (msglen > (1 << (8 * csize)))
+		return -EOVERFLOW;
+
+	data = cpu_to_be32(msglen);
+	memcpy(block - csize, (u8 *)&data + 4 - csize, csize);
+
+	return 0;
+}
+
+static int tegra_ccm_format_nonce(struct tegra_aead_reqctx *rctx, u8 *nonce)
+{
+	unsigned int q, t;
+	u8 *q_ptr, *iv = (u8 *)rctx->iv;
+
+	memcpy(nonce, rctx->iv, 16);
+
+	/*** 1. Prepare Flags Octet ***/
+
+	/* Encode t (mac length) */
+	t = rctx->authsize;
+	nonce[0] |= (((t - 2) / 2) << 3);
+
+	/* Adata */
+	if (rctx->assoclen)
+		nonce[0] |= (1 << 6);
+
+	/*** Encode Q - message length ***/
+	q = iv[0] + 1;
+	q_ptr = nonce + 16 - q;
+
+	return tegra_ccm_set_msg_len(q_ptr, rctx->cryptlen, q);
+}
+
+static int tegra_ccm_format_adata(u8 *adata, unsigned int a)
+{
+	int len = 0;
+
+	/* add control info for associated data
+	 * RFC 3610 and NIST Special Publication 800-38C
+	 */
+	if (a < 65280) {
+		*(__be16 *)adata = cpu_to_be16(a);
+		len = 2;
+	} else	{
+		*(__be16 *)adata = cpu_to_be16(0xfffe);
+		*(__be32 *)&adata[2] = cpu_to_be32(a);
+		len = 6;
+	}
+
+	return len;
+}
+
+static int tegra_ccm_add_padding(u8 *buf, unsigned int len)
+{
+	unsigned int padlen = 16 - (len % 16);
+	u8 padding[16] = {0};
+
+	if (padlen == 16)
+		return 0;
+
+	memcpy(buf, padding, padlen);
+
+	return padlen;
+}
+
+static int tegra_ccm_format_blocks(struct tegra_aead_reqctx *rctx)
+{
+	unsigned int alen = 0, offset = 0;
+	u8 nonce[16], adata[16];
+	int ret;
+
+	ret = tegra_ccm_format_nonce(rctx, nonce);
+	if (ret)
+		return ret;
+
+	memcpy(rctx->inbuf.buf, nonce, 16);
+	offset = 16;
+
+	if (rctx->assoclen) {
+		alen = tegra_ccm_format_adata(adata, rctx->assoclen);
+		memcpy(rctx->inbuf.buf + offset, adata, alen);
+		offset += alen;
+
+		scatterwalk_map_and_copy(rctx->inbuf.buf + offset,
+					 rctx->src_sg, 0, rctx->assoclen, 0);
+
+		offset += rctx->assoclen;
+		offset += tegra_ccm_add_padding(rctx->inbuf.buf + offset,
+					 rctx->assoclen + alen);
+	}
+
+	return offset;
+}
+
+static int tegra_ccm_mac_result(struct tegra_se *se, struct tegra_aead_reqctx *rctx)
+{
+	u32 result[16];
+	int i, ret;
+
+	/* Read and clear Result */
+	for (i = 0; i < CMAC_RESULT_REG_COUNT; i++)
+		result[i] = readl(se->base + se->hw->regs->result + (i * 4));
+
+	for (i = 0; i < CMAC_RESULT_REG_COUNT; i++)
+		writel(0, se->base + se->hw->regs->result + (i * 4));
+
+	if (rctx->encrypt) {
+		memcpy(rctx->authdata, result, rctx->authsize);
+	} else {
+		ret = crypto_memneq(rctx->authdata, result, rctx->authsize);
+		if (ret)
+			return -EBADMSG;
+	}
+
+	return 0;
+}
+
+static int tegra_ccm_ctr_result(struct tegra_se *se, struct tegra_aead_reqctx *rctx)
+{
+	/* Copy result */
+	scatterwalk_map_and_copy(rctx->outbuf.buf + 16, rctx->dst_sg,
+				 rctx->assoclen, rctx->cryptlen, 1);
+
+	if (rctx->encrypt)
+		scatterwalk_map_and_copy(rctx->outbuf.buf, rctx->dst_sg,
+					 rctx->assoclen + rctx->cryptlen,
+					 rctx->authsize, 1);
+	else
+		memcpy(rctx->authdata, rctx->outbuf.buf, rctx->authsize);
+
+	return 0;
+}
+
+static int tegra_ccm_compute_auth(struct tegra_aead_ctx *ctx, struct tegra_aead_reqctx *rctx)
+{
+	struct tegra_se *se = ctx->se;
+	struct scatterlist *sg;
+	int offset, ret;
+
+	offset = tegra_ccm_format_blocks(rctx);
+	if (offset < 0)
+		return -EINVAL;
+
+	/* Copy plain text to the buffer */
+	sg = rctx->encrypt ? rctx->src_sg : rctx->dst_sg;
+
+	scatterwalk_map_and_copy(rctx->inbuf.buf + offset,
+				 sg, rctx->assoclen,
+				 rctx->cryptlen, 0);
+	offset += rctx->cryptlen;
+	offset += tegra_ccm_add_padding(rctx->inbuf.buf + offset, rctx->cryptlen);
+
+	rctx->inbuf.size = offset;
+
+	ret = tegra_ccm_do_cbcmac(ctx, rctx);
+	if (ret)
+		return ret;
+
+	return tegra_ccm_mac_result(se, rctx);
+}
+
+static int tegra_ccm_do_ctr(struct tegra_aead_ctx *ctx, struct tegra_aead_reqctx *rctx)
+{
+	struct tegra_se *se = ctx->se;
+	unsigned int cmdlen, offset = 0;
+	struct scatterlist *sg = rctx->src_sg;
+	int ret;
+
+	rctx->config = tegra234_aes_cfg(SE_ALG_CTR, rctx->encrypt);
+	rctx->crypto_config = tegra234_aes_crypto_cfg(SE_ALG_CTR, rctx->encrypt) |
+			      SE_AES_KEY_INDEX(ctx->key_id);
+
+	/* Copy authdata in the top of buffer for encryption/decryption */
+	if (rctx->encrypt)
+		memcpy(rctx->inbuf.buf, rctx->authdata, rctx->authsize);
+	else
+		scatterwalk_map_and_copy(rctx->inbuf.buf, sg,
+					 rctx->assoclen + rctx->cryptlen,
+					 rctx->authsize, 0);
+
+	offset += rctx->authsize;
+	offset += tegra_ccm_add_padding(rctx->inbuf.buf + offset, rctx->authsize);
+
+	/* If there is no cryptlen, proceed to submit the task */
+	if (rctx->cryptlen) {
+		scatterwalk_map_and_copy(rctx->inbuf.buf + offset, sg,
+					 rctx->assoclen, rctx->cryptlen, 0);
+		offset += rctx->cryptlen;
+		offset += tegra_ccm_add_padding(rctx->inbuf.buf + offset, rctx->cryptlen);
+	}
+
+	rctx->inbuf.size = offset;
+
+	/* Prepare command and submit */
+	cmdlen = tegra_ctr_prep_cmd(ctx, rctx);
+	ret = tegra_se_host1x_submit(se, cmdlen);
+	if (ret)
+		return ret;
+
+	return tegra_ccm_ctr_result(se, rctx);
+}
+
+static int tegra_ccm_crypt_init(struct aead_request *req, struct tegra_se *se,
+				struct tegra_aead_reqctx *rctx)
+{
+	struct crypto_aead *tfm = crypto_aead_reqtfm(req);
+	u8 *iv = (u8 *)rctx->iv;
+	int ret, i;
+
+	rctx->src_sg = req->src;
+	rctx->dst_sg = req->dst;
+	rctx->assoclen = req->assoclen;
+	rctx->authsize = crypto_aead_authsize(tfm);
+
+	memcpy(iv, req->iv, 16);
+
+	ret = tegra_ccm_check_iv(iv);
+	if (ret)
+		return ret;
+
+	/* Note: rfc 3610 and NIST 800-38C require counter (ctr_0) of
+	 * zero to encrypt auth tag.
+	 * req->iv has the formatted ctr_0 (i.e. Flags || N || 0).
+	 */
+	memset(iv + 15 - iv[0], 0, iv[0] + 1);
+
+	/* Clear any previous result */
+	for (i = 0; i < CMAC_RESULT_REG_COUNT; i++)
+		writel(0, se->base + se->hw->regs->result + (i * 4));
+
+	return 0;
+}
+
+static int tegra_ccm_do_one_req(struct crypto_engine *engine, void *areq)
+{
+	struct aead_request *req = container_of(areq, struct aead_request, base);
+	struct tegra_aead_reqctx *rctx = aead_request_ctx(req);
+	struct crypto_aead *tfm = crypto_aead_reqtfm(req);
+	struct tegra_aead_ctx *ctx = crypto_aead_ctx(tfm);
+	struct tegra_se *se = ctx->se;
+	int ret;
+
+	/* Allocate buffers required */
+	rctx->inbuf.buf = dma_alloc_coherent(ctx->se->dev, SE_AES_BUFLEN,
+					     &rctx->inbuf.addr, GFP_KERNEL);
+	if (!rctx->inbuf.buf)
+		return -ENOMEM;
+
+	rctx->inbuf.size = SE_AES_BUFLEN;
+
+	rctx->outbuf.buf = dma_alloc_coherent(ctx->se->dev, SE_AES_BUFLEN,
+					      &rctx->outbuf.addr, GFP_KERNEL);
+	if (!rctx->outbuf.buf) {
+		ret = -ENOMEM;
+		goto outbuf_err;
+	}
+
+	rctx->outbuf.size = SE_AES_BUFLEN;
+
+	ret = tegra_ccm_crypt_init(req, se, rctx);
+	if (ret)
+		goto out;
+
+	if (rctx->encrypt) {
+		rctx->cryptlen = req->cryptlen;
+
+		/* CBC MAC Operation */
+		ret = tegra_ccm_compute_auth(ctx, rctx);
+		if (ret)
+			goto out;
+
+		/* CTR operation */
+		ret = tegra_ccm_do_ctr(ctx, rctx);
+		if (ret)
+			goto out;
+	} else {
+		rctx->cryptlen = req->cryptlen - ctx->authsize;
+		if (ret)
+			goto out;
+
+		/* CTR operation */
+		ret = tegra_ccm_do_ctr(ctx, rctx);
+		if (ret)
+			goto out;
+
+		/* CBC MAC Operation */
+		ret = tegra_ccm_compute_auth(ctx, rctx);
+		if (ret)
+			goto out;
+	}
+
+out:
+	dma_free_coherent(ctx->se->dev, SE_AES_BUFLEN,
+			  rctx->outbuf.buf, rctx->outbuf.addr);
+
+outbuf_err:
+	dma_free_coherent(ctx->se->dev, SE_AES_BUFLEN,
+			  rctx->inbuf.buf, rctx->inbuf.addr);
+
+	crypto_finalize_aead_request(ctx->se->engine, req, ret);
+
+	return 0;
+}
+
+static int tegra_gcm_do_one_req(struct crypto_engine *engine, void *areq)
+{
+	struct aead_request *req = container_of(areq, struct aead_request, base);
+	struct crypto_aead *tfm = crypto_aead_reqtfm(req);
+	struct tegra_aead_ctx *ctx = crypto_aead_ctx(tfm);
+	struct tegra_aead_reqctx *rctx = aead_request_ctx(req);
+	int ret;
+
+	/* Allocate buffers required */
+	rctx->inbuf.buf = dma_alloc_coherent(ctx->se->dev, SE_AES_BUFLEN,
+					     &rctx->inbuf.addr, GFP_KERNEL);
+	if (!rctx->inbuf.buf)
+		return -ENOMEM;
+
+	rctx->inbuf.size = SE_AES_BUFLEN;
+
+	rctx->outbuf.buf = dma_alloc_coherent(ctx->se->dev, SE_AES_BUFLEN,
+					      &rctx->outbuf.addr, GFP_KERNEL);
+	if (!rctx->outbuf.buf) {
+		ret = -ENOMEM;
+		goto outbuf_err;
+	}
+
+	rctx->outbuf.size = SE_AES_BUFLEN;
+
+	rctx->src_sg = req->src;
+	rctx->dst_sg = req->dst;
+	rctx->assoclen = req->assoclen;
+	rctx->authsize = crypto_aead_authsize(tfm);
+
+	if (rctx->encrypt)
+		rctx->cryptlen = req->cryptlen;
+	else
+		rctx->cryptlen = req->cryptlen - ctx->authsize;
+
+	memcpy(rctx->iv, req->iv, GCM_AES_IV_SIZE);
+	rctx->iv[3] = (1 << 24);
+
+	/* If there is associated data perform GMAC operation */
+	if (rctx->assoclen) {
+		ret = tegra_gcm_do_gmac(ctx, rctx);
+		if (ret)
+			goto out;
+	}
+
+	/* GCM Encryption/Decryption operation */
+	if (rctx->cryptlen) {
+		ret = tegra_gcm_do_crypt(ctx, rctx);
+		if (ret)
+			goto out;
+	}
+
+	/* GCM_FINAL operation */
+	ret = tegra_gcm_do_final(ctx, rctx);
+	if (ret)
+		goto out;
+
+	if (!rctx->encrypt)
+		ret = tegra_gcm_do_verify(ctx->se, rctx);
+
+out:
+	dma_free_coherent(ctx->se->dev, SE_AES_BUFLEN,
+			  rctx->outbuf.buf, rctx->outbuf.addr);
+
+outbuf_err:
+	dma_free_coherent(ctx->se->dev, SE_AES_BUFLEN,
+			  rctx->inbuf.buf, rctx->inbuf.addr);
+
+	/* Finalize the request if there are no errors */
+	crypto_finalize_aead_request(ctx->se->engine, req, ret);
+
+	return 0;
+}
+
+static int tegra_aead_cra_init(struct crypto_aead *tfm)
+{
+	struct tegra_aead_ctx *ctx = crypto_aead_ctx(tfm);
+	struct aead_alg *alg = crypto_aead_alg(tfm);
+	struct tegra_se_alg *se_alg;
+	const char *algname;
+	int ret;
+
+	algname = crypto_tfm_alg_name(&tfm->base);
+
+	se_alg = container_of(alg, struct tegra_se_alg, alg.aead.base);
+
+	crypto_aead_set_reqsize(tfm, sizeof(struct tegra_aead_reqctx));
+
+	ctx->se = se_alg->se_dev;
+	ctx->key_id = 0;
+
+	ret = se_algname_to_algid(algname);
+	if (ret < 0) {
+		dev_err(ctx->se->dev, "invalid algorithm\n");
+		return ret;
+	}
+
+	ctx->alg = ret;
+
+	return 0;
+}
+
+static int tegra_ccm_setauthsize(struct crypto_aead *tfm,  unsigned int authsize)
+{
+	struct tegra_aead_ctx *ctx = crypto_aead_ctx(tfm);
+
+	switch (authsize) {
+	case 4:
+	case 6:
+	case 8:
+	case 10:
+	case 12:
+	case 14:
+	case 16:
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	ctx->authsize = authsize;
+
+	return 0;
+}
+
+static int tegra_gcm_setauthsize(struct crypto_aead *tfm,  unsigned int authsize)
+{
+	struct tegra_aead_ctx *ctx = crypto_aead_ctx(tfm);
+	int ret;
+
+	ret = crypto_gcm_check_authsize(authsize);
+	if (ret)
+		return ret;
+
+	ctx->authsize = authsize;
+
+	return 0;
+}
+
+static void tegra_aead_cra_exit(struct crypto_aead *tfm)
+{
+	struct tegra_aead_ctx *ctx = crypto_tfm_ctx(&tfm->base);
+
+	if (ctx->key_id)
+		tegra_key_invalidate(ctx->se, ctx->key_id, ctx->alg);
+}
+
+static int tegra_aead_crypt(struct aead_request *req, bool encrypt)
+{
+	struct crypto_aead *tfm = crypto_aead_reqtfm(req);
+	struct tegra_aead_ctx *ctx = crypto_aead_ctx(tfm);
+	struct tegra_aead_reqctx *rctx = aead_request_ctx(req);
+
+	rctx->encrypt = encrypt;
+
+	return crypto_transfer_aead_request_to_engine(ctx->se->engine, req);
+}
+
+static int tegra_aead_encrypt(struct aead_request *req)
+{
+	return tegra_aead_crypt(req, true);
+}
+
+static int tegra_aead_decrypt(struct aead_request *req)
+{
+	return tegra_aead_crypt(req, false);
+}
+
+static int tegra_aead_setkey(struct crypto_aead *tfm,
+			     const u8 *key, u32 keylen)
+{
+	struct tegra_aead_ctx *ctx = crypto_aead_ctx(tfm);
+
+	if (aes_check_keylen(keylen)) {
+		dev_dbg(ctx->se->dev, "invalid key length (%d)\n", keylen);
+		return -EINVAL;
+	}
+
+	return tegra_key_submit(ctx->se, key, keylen, ctx->alg, &ctx->key_id);
+}
+
+static unsigned int tegra_cmac_prep_cmd(struct tegra_cmac_ctx *ctx,
+					struct tegra_cmac_reqctx *rctx)
+{
+	unsigned int data_count, res_bits = 0, i = 0, j;
+	struct tegra_se *se = ctx->se;
+	u32 *cpuvaddr = se->cmdbuf->addr, op;
+
+	data_count = (rctx->datbuf.size / AES_BLOCK_SIZE);
+
+	op = SE_AES_OP_WRSTALL | SE_AES_OP_START | SE_AES_OP_LASTBUF;
+
+	if (!(rctx->task & SHA_UPDATE)) {
+		op |= SE_AES_OP_FINAL;
+		res_bits = (rctx->datbuf.size % AES_BLOCK_SIZE) * 8;
+	}
+
+	if (!res_bits && data_count)
+		data_count--;
+
+	if (rctx->task & SHA_FIRST) {
+		rctx->task &= ~SHA_FIRST;
+
+		cpuvaddr[i++] = host1x_opcode_setpayload(SE_CRYPTO_CTR_REG_COUNT);
+		cpuvaddr[i++] = se_host1x_opcode_incr_w(se->hw->regs->linear_ctr);
+		/* Load 0 IV */
+		for (j = 0; j < SE_CRYPTO_CTR_REG_COUNT; j++)
+			cpuvaddr[i++] = 0;
+	}
+
+	cpuvaddr[i++] = se_host1x_opcode_nonincr(se->hw->regs->last_blk, 1);
+	cpuvaddr[i++] = SE_LAST_BLOCK_VAL(data_count) |
+			SE_LAST_BLOCK_RES_BITS(res_bits);
+
+	cpuvaddr[i++] = se_host1x_opcode_incr(se->hw->regs->config, 6);
+	cpuvaddr[i++] = rctx->config;
+	cpuvaddr[i++] = rctx->crypto_config;
+
+	/* Source Address */
+	cpuvaddr[i++] = lower_32_bits(rctx->datbuf.addr);
+	cpuvaddr[i++] = SE_ADDR_HI_MSB(upper_32_bits(rctx->datbuf.addr)) |
+			SE_ADDR_HI_SZ(rctx->datbuf.size);
+	cpuvaddr[i++] = 0;
+	cpuvaddr[i++] = SE_ADDR_HI_SZ(AES_BLOCK_SIZE);
+
+	cpuvaddr[i++] = se_host1x_opcode_nonincr(se->hw->regs->op, 1);
+	cpuvaddr[i++] = op;
+
+	cpuvaddr[i++] = se_host1x_opcode_nonincr(host1x_uclass_incr_syncpt_r(), 1);
+	cpuvaddr[i++] = host1x_uclass_incr_syncpt_cond_f(1) |
+			host1x_uclass_incr_syncpt_indx_f(se->syncpt_id);
+
+	return i;
+}
+
+static void tegra_cmac_copy_result(struct tegra_se *se, struct tegra_cmac_reqctx *rctx)
+{
+	int i;
+
+	for (i = 0; i < CMAC_RESULT_REG_COUNT; i++)
+		rctx->result[i] = readl(se->base + se->hw->regs->result + (i * 4));
+}
+
+static void tegra_cmac_paste_result(struct tegra_se *se, struct tegra_cmac_reqctx *rctx)
+{
+	int i;
+
+	for (i = 0; i < CMAC_RESULT_REG_COUNT; i++)
+		writel(rctx->result[i],
+		       se->base + se->hw->regs->result + (i * 4));
+}
+
+static int tegra_cmac_do_update(struct ahash_request *req)
+{
+	struct tegra_cmac_reqctx *rctx = ahash_request_ctx(req);
+	struct crypto_ahash *tfm = crypto_ahash_reqtfm(req);
+	struct tegra_cmac_ctx *ctx = crypto_ahash_ctx(tfm);
+	struct tegra_se *se = ctx->se;
+	unsigned int nblks, nresidue, cmdlen;
+	int ret;
+
+	if (!req->nbytes)
+		return 0;
+
+	nresidue = (req->nbytes + rctx->residue.size) % rctx->blk_size;
+	nblks = (req->nbytes + rctx->residue.size) / rctx->blk_size;
+
+	/*
+	 * Reserve the last block as residue during final() to process.
+	 */
+	if (!nresidue && nblks) {
+		nresidue += rctx->blk_size;
+		nblks--;
+	}
+
+	rctx->src_sg = req->src;
+	rctx->datbuf.size = (req->nbytes + rctx->residue.size) - nresidue;
+	rctx->total_len += rctx->datbuf.size;
+	rctx->config = tegra234_aes_cfg(SE_ALG_CMAC, 0);
+	rctx->crypto_config = SE_AES_KEY_INDEX(ctx->key_id);
+
+	/*
+	 * Keep one block and residue bytes in residue and
+	 * return. The bytes will be processed in final()
+	 */
+	if (nblks < 1) {
+		scatterwalk_map_and_copy(rctx->residue.buf + rctx->residue.size,
+					 rctx->src_sg, 0, req->nbytes, 0);
+
+		rctx->residue.size += req->nbytes;
+		return 0;
+	}
+
+	/* Copy the previous residue first */
+	if (rctx->residue.size)
+		memcpy(rctx->datbuf.buf, rctx->residue.buf, rctx->residue.size);
+
+	scatterwalk_map_and_copy(rctx->datbuf.buf + rctx->residue.size,
+				 rctx->src_sg, 0, req->nbytes - nresidue, 0);
+
+	scatterwalk_map_and_copy(rctx->residue.buf, rctx->src_sg,
+				 req->nbytes - nresidue, nresidue, 0);
+
+	/* Update residue value with the residue after current block */
+	rctx->residue.size = nresidue;
+
+	/*
+	 * If this is not the first 'update' call, paste the previous copied
+	 * intermediate results to the registers so that it gets picked up.
+	 * This is to support the import/export functionality.
+	 */
+	if (!(rctx->task & SHA_FIRST))
+		tegra_cmac_paste_result(ctx->se, rctx);
+
+	cmdlen = tegra_cmac_prep_cmd(ctx, rctx);
+
+	ret = tegra_se_host1x_submit(se, cmdlen);
+	/*
+	 * If this is not the final update, copy the intermediate results
+	 * from the registers so that it can be used in the next 'update'
+	 * call. This is to support the import/export functionality.
+	 */
+	if (!(rctx->task & SHA_FINAL))
+		tegra_cmac_copy_result(ctx->se, rctx);
+
+	return ret;
+}
+
+static int tegra_cmac_do_final(struct ahash_request *req)
+{
+	struct tegra_cmac_reqctx *rctx = ahash_request_ctx(req);
+	struct crypto_ahash *tfm = crypto_ahash_reqtfm(req);
+	struct tegra_cmac_ctx *ctx = crypto_ahash_ctx(tfm);
+	struct tegra_se *se = ctx->se;
+	u32 *result = (u32 *)req->result;
+	int ret = 0, i, cmdlen;
+
+	if (!req->nbytes && !rctx->total_len && ctx->fallback_tfm) {
+		return crypto_shash_tfm_digest(ctx->fallback_tfm,
+					rctx->datbuf.buf, 0, req->result);
+	}
+
+	memcpy(rctx->datbuf.buf, rctx->residue.buf, rctx->residue.size);
+	rctx->datbuf.size = rctx->residue.size;
+	rctx->total_len += rctx->residue.size;
+	rctx->config = tegra234_aes_cfg(SE_ALG_CMAC, 0);
+
+	/* Prepare command and submit */
+	cmdlen = tegra_cmac_prep_cmd(ctx, rctx);
+	ret = tegra_se_host1x_submit(se, cmdlen);
+	if (ret)
+		goto out;
+
+	/* Read and clear Result register */
+	for (i = 0; i < CMAC_RESULT_REG_COUNT; i++)
+		result[i] = readl(se->base + se->hw->regs->result + (i * 4));
+
+	for (i = 0; i < CMAC_RESULT_REG_COUNT; i++)
+		writel(0, se->base + se->hw->regs->result + (i * 4));
+
+out:
+	dma_free_coherent(se->dev, SE_SHA_BUFLEN,
+			  rctx->datbuf.buf, rctx->datbuf.addr);
+	dma_free_coherent(se->dev, crypto_ahash_blocksize(tfm) * 2,
+			  rctx->residue.buf, rctx->residue.addr);
+	return ret;
+}
+
+static int tegra_cmac_do_one_req(struct crypto_engine *engine, void *areq)
+{
+	struct ahash_request *req = ahash_request_cast(areq);
+	struct tegra_cmac_reqctx *rctx = ahash_request_ctx(req);
+	struct crypto_ahash *tfm = crypto_ahash_reqtfm(req);
+	struct tegra_cmac_ctx *ctx = crypto_ahash_ctx(tfm);
+	struct tegra_se *se = ctx->se;
+	int ret;
+
+	if (rctx->task & SHA_UPDATE) {
+		ret = tegra_cmac_do_update(req);
+		rctx->task &= ~SHA_UPDATE;
+	}
+
+	if (rctx->task & SHA_FINAL) {
+		ret = tegra_cmac_do_final(req);
+		rctx->task &= ~SHA_FINAL;
+	}
+
+	crypto_finalize_hash_request(se->engine, req, ret);
+
+	return 0;
+}
+
+static void tegra_cmac_init_fallback(struct crypto_ahash *tfm, struct tegra_cmac_ctx *ctx,
+				     const char *algname)
+{
+	unsigned int statesize;
+
+	ctx->fallback_tfm = crypto_alloc_shash(algname, 0, CRYPTO_ALG_NEED_FALLBACK);
+
+	if (IS_ERR(ctx->fallback_tfm)) {
+		dev_warn(ctx->se->dev, "failed to allocate fallback for %s\n", algname);
+		ctx->fallback_tfm = NULL;
+		return;
+	}
+
+	statesize = crypto_shash_statesize(ctx->fallback_tfm);
+
+	if (statesize > sizeof(struct tegra_cmac_reqctx))
+		crypto_ahash_set_statesize(tfm, statesize);
+}
+
+static int tegra_cmac_cra_init(struct crypto_tfm *tfm)
+{
+	struct tegra_cmac_ctx *ctx = crypto_tfm_ctx(tfm);
+	struct crypto_ahash *ahash_tfm = __crypto_ahash_cast(tfm);
+	struct ahash_alg *alg = __crypto_ahash_alg(tfm->__crt_alg);
+	struct tegra_se_alg *se_alg;
+	const char *algname;
+	int ret;
+
+	algname = crypto_tfm_alg_name(tfm);
+	se_alg = container_of(alg, struct tegra_se_alg, alg.ahash.base);
+
+	crypto_ahash_set_reqsize(ahash_tfm, sizeof(struct tegra_cmac_reqctx));
+
+	ctx->se = se_alg->se_dev;
+	ctx->key_id = 0;
+
+	ret = se_algname_to_algid(algname);
+	if (ret < 0) {
+		dev_err(ctx->se->dev, "invalid algorithm\n");
+		return ret;
+	}
+
+	ctx->alg = ret;
+
+	tegra_cmac_init_fallback(ahash_tfm, ctx, algname);
+
+	return 0;
+}
+
+static void tegra_cmac_cra_exit(struct crypto_tfm *tfm)
+{
+	struct tegra_cmac_ctx *ctx = crypto_tfm_ctx(tfm);
+
+	if (ctx->fallback_tfm)
+		crypto_free_shash(ctx->fallback_tfm);
+
+	tegra_key_invalidate(ctx->se, ctx->key_id, ctx->alg);
+}
+
+static int tegra_cmac_init(struct ahash_request *req)
+{
+	struct tegra_cmac_reqctx *rctx = ahash_request_ctx(req);
+	struct crypto_ahash *tfm = crypto_ahash_reqtfm(req);
+	struct tegra_cmac_ctx *ctx = crypto_ahash_ctx(tfm);
+	struct tegra_se *se = ctx->se;
+	int i;
+
+	rctx->total_len = 0;
+	rctx->datbuf.size = 0;
+	rctx->residue.size = 0;
+	rctx->task = SHA_FIRST;
+	rctx->blk_size = crypto_ahash_blocksize(tfm);
+
+	rctx->residue.buf = dma_alloc_coherent(se->dev, rctx->blk_size * 2,
+					       &rctx->residue.addr, GFP_KERNEL);
+	if (!rctx->residue.buf)
+		goto resbuf_fail;
+
+	rctx->residue.size = 0;
+
+	rctx->datbuf.buf = dma_alloc_coherent(se->dev, SE_SHA_BUFLEN,
+					      &rctx->datbuf.addr, GFP_KERNEL);
+	if (!rctx->datbuf.buf)
+		goto datbuf_fail;
+
+	rctx->datbuf.size = 0;
+
+	/* Clear any previous result */
+	for (i = 0; i < CMAC_RESULT_REG_COUNT; i++)
+		writel(0, se->base + se->hw->regs->result + (i * 4));
+
+	return 0;
+
+datbuf_fail:
+	dma_free_coherent(se->dev, rctx->blk_size, rctx->residue.buf,
+			  rctx->residue.addr);
+resbuf_fail:
+	return -ENOMEM;
+}
+
+static int tegra_cmac_setkey(struct crypto_ahash *tfm, const u8 *key,
+			     unsigned int keylen)
+{
+	struct tegra_cmac_ctx *ctx = crypto_ahash_ctx(tfm);
+
+	if (aes_check_keylen(keylen)) {
+		dev_dbg(ctx->se->dev, "invalid key length (%d)\n", keylen);
+		return -EINVAL;
+	}
+
+	if (ctx->fallback_tfm)
+		crypto_shash_setkey(ctx->fallback_tfm, key, keylen);
+
+	return tegra_key_submit(ctx->se, key, keylen, ctx->alg, &ctx->key_id);
+}
+
+static int tegra_cmac_update(struct ahash_request *req)
+{
+	struct crypto_ahash *tfm = crypto_ahash_reqtfm(req);
+	struct tegra_cmac_ctx *ctx = crypto_ahash_ctx(tfm);
+	struct tegra_cmac_reqctx *rctx = ahash_request_ctx(req);
+
+	rctx->task |= SHA_UPDATE;
+
+	return crypto_transfer_hash_request_to_engine(ctx->se->engine, req);
+}
+
+static int tegra_cmac_final(struct ahash_request *req)
+{
+	struct crypto_ahash *tfm = crypto_ahash_reqtfm(req);
+	struct tegra_cmac_ctx *ctx = crypto_ahash_ctx(tfm);
+	struct tegra_cmac_reqctx *rctx = ahash_request_ctx(req);
+
+	rctx->task |= SHA_FINAL;
+
+	return crypto_transfer_hash_request_to_engine(ctx->se->engine, req);
+}
+
+static int tegra_cmac_finup(struct ahash_request *req)
+{
+	struct crypto_ahash *tfm = crypto_ahash_reqtfm(req);
+	struct tegra_cmac_ctx *ctx = crypto_ahash_ctx(tfm);
+	struct tegra_cmac_reqctx *rctx = ahash_request_ctx(req);
+
+	rctx->task |= SHA_UPDATE | SHA_FINAL;
+
+	return crypto_transfer_hash_request_to_engine(ctx->se->engine, req);
+}
+
+static int tegra_cmac_digest(struct ahash_request *req)
+{
+	struct crypto_ahash *tfm = crypto_ahash_reqtfm(req);
+	struct tegra_cmac_ctx *ctx = crypto_ahash_ctx(tfm);
+	struct tegra_cmac_reqctx *rctx = ahash_request_ctx(req);
+
+	tegra_cmac_init(req);
+	rctx->task |= SHA_UPDATE | SHA_FINAL;
+
+	return crypto_transfer_hash_request_to_engine(ctx->se->engine, req);
+}
+
+static int tegra_cmac_export(struct ahash_request *req, void *out)
+{
+	struct tegra_cmac_reqctx *rctx = ahash_request_ctx(req);
+
+	memcpy(out, rctx, sizeof(*rctx));
+
+	return 0;
+}
+
+static int tegra_cmac_import(struct ahash_request *req, const void *in)
+{
+	struct tegra_cmac_reqctx *rctx = ahash_request_ctx(req);
+
+	memcpy(rctx, in, sizeof(*rctx));
+
+	return 0;
+}
+
+static struct tegra_se_alg tegra_aead_algs[] = {
+	{
+		.alg.aead.op.do_one_request = tegra_gcm_do_one_req,
+		.alg.aead.base = {
+			.init = tegra_aead_cra_init,
+			.exit = tegra_aead_cra_exit,
+			.setkey = tegra_aead_setkey,
+			.setauthsize = tegra_gcm_setauthsize,
+			.encrypt = tegra_aead_encrypt,
+			.decrypt = tegra_aead_decrypt,
+			.maxauthsize = AES_BLOCK_SIZE,
+			.ivsize	= GCM_AES_IV_SIZE,
+			.base = {
+				.cra_name = "gcm(aes)",
+				.cra_driver_name = "gcm-aes-tegra",
+				.cra_priority = 500,
+				.cra_blocksize = 1,
+				.cra_ctxsize = sizeof(struct tegra_aead_ctx),
+				.cra_alignmask = 0xf,
+				.cra_module = THIS_MODULE,
+			},
+		}
+	}, {
+		.alg.aead.op.do_one_request = tegra_ccm_do_one_req,
+		.alg.aead.base = {
+			.init = tegra_aead_cra_init,
+			.exit = tegra_aead_cra_exit,
+			.setkey	= tegra_aead_setkey,
+			.setauthsize = tegra_ccm_setauthsize,
+			.encrypt = tegra_aead_encrypt,
+			.decrypt = tegra_aead_decrypt,
+			.maxauthsize = AES_BLOCK_SIZE,
+			.ivsize	= AES_BLOCK_SIZE,
+			.chunksize = AES_BLOCK_SIZE,
+			.base = {
+				.cra_name = "ccm(aes)",
+				.cra_driver_name = "ccm-aes-tegra",
+				.cra_priority = 500,
+				.cra_blocksize = 1,
+				.cra_ctxsize = sizeof(struct tegra_aead_ctx),
+				.cra_alignmask = 0xf,
+				.cra_module = THIS_MODULE,
+			},
+		}
+	}
+};
+
+static struct tegra_se_alg tegra_cmac_algs[] = {
+	{
+		.alg.ahash.op.do_one_request = tegra_cmac_do_one_req,
+		.alg.ahash.base = {
+			.init = tegra_cmac_init,
+			.setkey	= tegra_cmac_setkey,
+			.update = tegra_cmac_update,
+			.final = tegra_cmac_final,
+			.finup = tegra_cmac_finup,
+			.digest = tegra_cmac_digest,
+			.export = tegra_cmac_export,
+			.import = tegra_cmac_import,
+			.halg.digestsize = AES_BLOCK_SIZE,
+			.halg.statesize = sizeof(struct tegra_cmac_reqctx),
+			.halg.base = {
+				.cra_name = "cmac(aes)",
+				.cra_driver_name = "tegra-se-cmac",
+				.cra_priority = 300,
+				.cra_flags = CRYPTO_ALG_TYPE_AHASH,
+				.cra_blocksize = AES_BLOCK_SIZE,
+				.cra_ctxsize = sizeof(struct tegra_cmac_ctx),
+				.cra_alignmask = 0,
+				.cra_module = THIS_MODULE,
+				.cra_init = tegra_cmac_cra_init,
+				.cra_exit = tegra_cmac_cra_exit,
+			}
+		}
+	}
+};
+
+int tegra_init_aes(struct tegra_se *se)
+{
+	struct aead_engine_alg *aead_alg;
+	struct ahash_engine_alg *ahash_alg;
+	struct skcipher_engine_alg *sk_alg;
+	int i, ret;
+
+	se->manifest = tegra_aes_kac_manifest;
+
+	for (i = 0; i < ARRAY_SIZE(tegra_aes_algs); i++) {
+		sk_alg = &tegra_aes_algs[i].alg.skcipher;
+		tegra_aes_algs[i].se_dev = se;
+
+		ret = crypto_engine_register_skcipher(sk_alg);
+		if (ret) {
+			dev_err(se->dev, "failed to register %s\n",
+				sk_alg->base.base.cra_name);
+			goto err_aes;
+		}
+	}
+
+	for (i = 0; i < ARRAY_SIZE(tegra_aead_algs); i++) {
+		aead_alg = &tegra_aead_algs[i].alg.aead;
+		tegra_aead_algs[i].se_dev = se;
+
+		ret = crypto_engine_register_aead(aead_alg);
+		if (ret) {
+			dev_err(se->dev, "failed to register %s\n",
+				aead_alg->base.base.cra_name);
+			goto err_aead;
+		}
+	}
+
+	for (i = 0; i < ARRAY_SIZE(tegra_cmac_algs); i++) {
+		ahash_alg = &tegra_cmac_algs[i].alg.ahash;
+		tegra_cmac_algs[i].se_dev = se;
+
+		ret = crypto_engine_register_ahash(ahash_alg);
+		if (ret) {
+			dev_err(se->dev, "failed to register %s\n",
+				ahash_alg->base.halg.base.cra_name);
+			goto err_cmac;
+		}
+	}
+
+	return 0;
+
+err_cmac:
+	while (i--)
+		crypto_engine_unregister_ahash(&tegra_cmac_algs[i].alg.ahash);
+
+	i = ARRAY_SIZE(tegra_aead_algs);
+err_aead:
+	while (i--)
+		crypto_engine_unregister_aead(&tegra_aead_algs[i].alg.aead);
+
+	i = ARRAY_SIZE(tegra_aes_algs);
+err_aes:
+	while (i--)
+		crypto_engine_unregister_skcipher(&tegra_aes_algs[i].alg.skcipher);
+
+	return ret;
+}
+
+void tegra_deinit_aes(struct tegra_se *se)
+{
+	int i;
+
+	for (i = 0; i < ARRAY_SIZE(tegra_aes_algs); i++)
+		crypto_engine_unregister_skcipher(&tegra_aes_algs[i].alg.skcipher);
+
+	for (i = 0; i < ARRAY_SIZE(tegra_aead_algs); i++)
+		crypto_engine_unregister_aead(&tegra_aead_algs[i].alg.aead);
+
+	for (i = 0; i < ARRAY_SIZE(tegra_cmac_algs); i++)
+		crypto_engine_unregister_ahash(&tegra_cmac_algs[i].alg.ahash);
+}
diff --git a/drivers/crypto/tegra/tegra-se-hash.c b/drivers/crypto/tegra/tegra-se-hash.c
new file mode 100644
index 000000000000..4d4bd727f498
--- /dev/null
+++ b/drivers/crypto/tegra/tegra-se-hash.c
@@ -0,0 +1,1060 @@
+// SPDX-License-Identifier: GPL-2.0-only
+// SPDX-FileCopyrightText: Copyright (c) 2023 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
+/*
+ * Crypto driver to handle HASH algorithms using NVIDIA Security Engine.
+ */
+
+#include <linux/clk.h>
+#include <linux/dma-mapping.h>
+#include <linux/module.h>
+#include <linux/of_device.h>
+#include <linux/platform_device.h>
+
+#include <crypto/aes.h>
+#include <crypto/sha1.h>
+#include <crypto/sha2.h>
+#include <crypto/sha3.h>
+#include <crypto/internal/des.h>
+#include <crypto/engine.h>
+#include <crypto/scatterwalk.h>
+#include <crypto/internal/hash.h>
+
+#include "tegra-se.h"
+
+struct tegra_sha_ctx {
+	struct tegra_se *se;
+	unsigned int alg;
+	bool fallback;
+	u32 key_id;
+	struct crypto_ahash *fallback_tfm;
+};
+
+struct tegra_sha_reqctx {
+	struct scatterlist *src_sg;
+	struct tegra_se_datbuf datbuf;
+	struct tegra_se_datbuf residue;
+	struct tegra_se_datbuf digest;
+	unsigned int alg;
+	unsigned int config;
+	unsigned int total_len;
+	unsigned int blk_size;
+	unsigned int task;
+	u32 key_id;
+	u32 result[HASH_RESULT_REG_COUNT];
+	struct ahash_request fallback_req;
+};
+
+static int tegra_sha_get_config(u32 alg)
+{
+	int cfg = 0;
+
+	switch (alg) {
+	case SE_ALG_SHA1:
+		cfg |= SE_SHA_ENC_ALG_SHA;
+		cfg |= SE_SHA_ENC_MODE_SHA1;
+		break;
+
+	case SE_ALG_HMAC_SHA224:
+		cfg |= SE_SHA_ENC_ALG_HMAC;
+		fallthrough;
+	case SE_ALG_SHA224:
+		cfg |= SE_SHA_ENC_ALG_SHA;
+		cfg |= SE_SHA_ENC_MODE_SHA224;
+		break;
+
+	case SE_ALG_HMAC_SHA256:
+		cfg |= SE_SHA_ENC_ALG_HMAC;
+		fallthrough;
+	case SE_ALG_SHA256:
+		cfg |= SE_SHA_ENC_ALG_SHA;
+		cfg |= SE_SHA_ENC_MODE_SHA256;
+		break;
+
+	case SE_ALG_HMAC_SHA384:
+		cfg |= SE_SHA_ENC_ALG_HMAC;
+		fallthrough;
+	case SE_ALG_SHA384:
+		cfg |= SE_SHA_ENC_ALG_SHA;
+		cfg |= SE_SHA_ENC_MODE_SHA384;
+		break;
+
+	case SE_ALG_HMAC_SHA512:
+		cfg |= SE_SHA_ENC_ALG_HMAC;
+		fallthrough;
+	case SE_ALG_SHA512:
+		cfg |= SE_SHA_ENC_ALG_SHA;
+		cfg |= SE_SHA_ENC_MODE_SHA512;
+		break;
+
+	case SE_ALG_SHA3_224:
+		cfg |= SE_SHA_ENC_ALG_SHA;
+		cfg |= SE_SHA_ENC_MODE_SHA3_224;
+		break;
+	case SE_ALG_SHA3_256:
+		cfg |= SE_SHA_ENC_ALG_SHA;
+		cfg |= SE_SHA_ENC_MODE_SHA3_256;
+		break;
+	case SE_ALG_SHA3_384:
+		cfg |= SE_SHA_ENC_ALG_SHA;
+		cfg |= SE_SHA_ENC_MODE_SHA3_384;
+		break;
+	case SE_ALG_SHA3_512:
+		cfg |= SE_SHA_ENC_ALG_SHA;
+		cfg |= SE_SHA_ENC_MODE_SHA3_512;
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	return cfg;
+}
+
+static int tegra_sha_fallback_init(struct ahash_request *req)
+{
+	struct tegra_sha_reqctx *rctx = ahash_request_ctx(req);
+	struct crypto_ahash *tfm = crypto_ahash_reqtfm(req);
+	struct tegra_sha_ctx *ctx = crypto_ahash_ctx(tfm);
+
+	ahash_request_set_tfm(&rctx->fallback_req, ctx->fallback_tfm);
+	rctx->fallback_req.base.flags = req->base.flags &
+					CRYPTO_TFM_REQ_MAY_SLEEP;
+
+	return crypto_ahash_init(&rctx->fallback_req);
+}
+
+static int tegra_sha_fallback_update(struct ahash_request *req)
+{
+	struct tegra_sha_reqctx *rctx = ahash_request_ctx(req);
+	struct crypto_ahash *tfm = crypto_ahash_reqtfm(req);
+	struct tegra_sha_ctx *ctx = crypto_ahash_ctx(tfm);
+
+	ahash_request_set_tfm(&rctx->fallback_req, ctx->fallback_tfm);
+	rctx->fallback_req.base.flags = req->base.flags &
+					CRYPTO_TFM_REQ_MAY_SLEEP;
+	rctx->fallback_req.nbytes = req->nbytes;
+	rctx->fallback_req.src = req->src;
+
+	return crypto_ahash_update(&rctx->fallback_req);
+}
+
+static int tegra_sha_fallback_final(struct ahash_request *req)
+{
+	struct tegra_sha_reqctx *rctx = ahash_request_ctx(req);
+	struct crypto_ahash *tfm = crypto_ahash_reqtfm(req);
+	struct tegra_sha_ctx *ctx = crypto_ahash_ctx(tfm);
+
+	ahash_request_set_tfm(&rctx->fallback_req, ctx->fallback_tfm);
+	rctx->fallback_req.base.flags = req->base.flags &
+					CRYPTO_TFM_REQ_MAY_SLEEP;
+	rctx->fallback_req.result = req->result;
+
+	return crypto_ahash_final(&rctx->fallback_req);
+}
+
+static int tegra_sha_fallback_finup(struct ahash_request *req)
+{
+	struct tegra_sha_reqctx *rctx = ahash_request_ctx(req);
+	struct crypto_ahash *tfm = crypto_ahash_reqtfm(req);
+	struct tegra_sha_ctx *ctx = crypto_ahash_ctx(tfm);
+
+	ahash_request_set_tfm(&rctx->fallback_req, ctx->fallback_tfm);
+	rctx->fallback_req.base.flags = req->base.flags &
+					CRYPTO_TFM_REQ_MAY_SLEEP;
+
+	rctx->fallback_req.nbytes = req->nbytes;
+	rctx->fallback_req.src = req->src;
+	rctx->fallback_req.result = req->result;
+
+	return crypto_ahash_finup(&rctx->fallback_req);
+}
+
+static int tegra_sha_fallback_digest(struct ahash_request *req)
+{
+	struct tegra_sha_reqctx *rctx = ahash_request_ctx(req);
+	struct crypto_ahash *tfm = crypto_ahash_reqtfm(req);
+	struct tegra_sha_ctx *ctx = crypto_ahash_ctx(tfm);
+
+	ahash_request_set_tfm(&rctx->fallback_req, ctx->fallback_tfm);
+	rctx->fallback_req.base.flags = req->base.flags &
+					CRYPTO_TFM_REQ_MAY_SLEEP;
+
+	rctx->fallback_req.nbytes = req->nbytes;
+	rctx->fallback_req.src = req->src;
+	rctx->fallback_req.result = req->result;
+
+	return crypto_ahash_digest(&rctx->fallback_req);
+}
+
+static int tegra_sha_fallback_import(struct ahash_request *req, const void *in)
+{
+	struct tegra_sha_reqctx *rctx = ahash_request_ctx(req);
+	struct crypto_ahash *tfm = crypto_ahash_reqtfm(req);
+	struct tegra_sha_ctx *ctx = crypto_ahash_ctx(tfm);
+
+	ahash_request_set_tfm(&rctx->fallback_req, ctx->fallback_tfm);
+	rctx->fallback_req.base.flags = req->base.flags &
+					CRYPTO_TFM_REQ_MAY_SLEEP;
+
+	return crypto_ahash_import(&rctx->fallback_req, in);
+}
+
+static int tegra_sha_fallback_export(struct ahash_request *req, void *out)
+{
+	struct tegra_sha_reqctx *rctx = ahash_request_ctx(req);
+	struct crypto_ahash *tfm = crypto_ahash_reqtfm(req);
+	struct tegra_sha_ctx *ctx = crypto_ahash_ctx(tfm);
+
+	ahash_request_set_tfm(&rctx->fallback_req, ctx->fallback_tfm);
+	rctx->fallback_req.base.flags = req->base.flags &
+					CRYPTO_TFM_REQ_MAY_SLEEP;
+
+	return crypto_ahash_export(&rctx->fallback_req, out);
+}
+
+static int tegra_sha_prep_cmd(struct tegra_se *se, u32 *cpuvaddr,
+			      struct tegra_sha_reqctx *rctx)
+{
+	u64 msg_len, msg_left;
+	int i = 0;
+
+	msg_len = rctx->total_len * 8;
+	msg_left = rctx->datbuf.size * 8;
+
+	/*
+	 * If IN_ADDR_HI_0.SZ > SHA_MSG_LEFT_[0-3] to the HASH engine,
+	 * HW treats it as the last buffer and process the data.
+	 * Therefore, add an extra byte to msg_left if it is not the
+	 * last buffer.
+	 */
+	if (rctx->task & SHA_UPDATE) {
+		msg_left += 8;
+		msg_len += 8;
+	}
+
+	cpuvaddr[i++] = host1x_opcode_setpayload(8);
+	cpuvaddr[i++] = se_host1x_opcode_incr_w(SE_SHA_MSG_LENGTH);
+	cpuvaddr[i++] = lower_32_bits(msg_len);
+	cpuvaddr[i++] = upper_32_bits(msg_len);
+	cpuvaddr[i++] = 0;
+	cpuvaddr[i++] = 0;
+	cpuvaddr[i++] = lower_32_bits(msg_left);
+	cpuvaddr[i++] = upper_32_bits(msg_left);
+	cpuvaddr[i++] = 0;
+	cpuvaddr[i++] = 0;
+	cpuvaddr[i++] = host1x_opcode_setpayload(6);
+	cpuvaddr[i++] = se_host1x_opcode_incr_w(SE_SHA_CFG);
+	cpuvaddr[i++] = rctx->config;
+
+	if (rctx->task & SHA_FIRST) {
+		cpuvaddr[i++] = SE_SHA_TASK_HASH_INIT;
+		rctx->task &= ~SHA_FIRST;
+	} else {
+		cpuvaddr[i++] = 0;
+	}
+
+	cpuvaddr[i++] = rctx->datbuf.addr;
+	cpuvaddr[i++] = (u32)(SE_ADDR_HI_MSB(upper_32_bits(rctx->datbuf.addr)) |
+				SE_ADDR_HI_SZ(rctx->datbuf.size));
+	cpuvaddr[i++] = rctx->digest.addr;
+	cpuvaddr[i++] = (u32)(SE_ADDR_HI_MSB(upper_32_bits(rctx->digest.addr)) |
+				SE_ADDR_HI_SZ(rctx->digest.size));
+	if (rctx->key_id) {
+		cpuvaddr[i++] = host1x_opcode_setpayload(1);
+		cpuvaddr[i++] = se_host1x_opcode_nonincr_w(SE_SHA_CRYPTO_CFG);
+		cpuvaddr[i++] = SE_AES_KEY_INDEX(rctx->key_id);
+	}
+
+	cpuvaddr[i++] = host1x_opcode_setpayload(1);
+	cpuvaddr[i++] = se_host1x_opcode_nonincr_w(SE_SHA_OPERATION);
+	cpuvaddr[i++] = SE_SHA_OP_WRSTALL |
+			SE_SHA_OP_START |
+			SE_SHA_OP_LASTBUF;
+	cpuvaddr[i++] = se_host1x_opcode_nonincr(host1x_uclass_incr_syncpt_r(), 1);
+	cpuvaddr[i++] = host1x_uclass_incr_syncpt_cond_f(1) |
+			host1x_uclass_incr_syncpt_indx_f(se->syncpt_id);
+
+	dev_dbg(se->dev, "msg len %llu msg left %llu cfg %#x",
+		msg_len, msg_left, rctx->config);
+
+	return i;
+}
+
+static void tegra_sha_copy_hash_result(struct tegra_se *se, struct tegra_sha_reqctx *rctx)
+{
+	int i;
+
+	for (i = 0; i < HASH_RESULT_REG_COUNT; i++)
+		rctx->result[i] = readl(se->base + se->hw->regs->result + (i * 4));
+}
+
+static void tegra_sha_paste_hash_result(struct tegra_se *se, struct tegra_sha_reqctx *rctx)
+{
+	int i;
+
+	for (i = 0; i < HASH_RESULT_REG_COUNT; i++)
+		writel(rctx->result[i],
+		       se->base + se->hw->regs->result + (i * 4));
+}
+
+static int tegra_sha_do_update(struct ahash_request *req)
+{
+	struct tegra_sha_ctx *ctx = crypto_ahash_ctx(crypto_ahash_reqtfm(req));
+	struct tegra_sha_reqctx *rctx = ahash_request_ctx(req);
+	unsigned int nblks, nresidue, size, ret;
+	u32 *cpuvaddr = ctx->se->cmdbuf->addr;
+
+	nresidue = (req->nbytes + rctx->residue.size) % rctx->blk_size;
+	nblks = (req->nbytes + rctx->residue.size) / rctx->blk_size;
+
+	/*
+	 * If nbytes is a multiple of block size and there is no residue,
+	 * then reserve the last block as residue during final() to process.
+	 */
+	if (!nresidue && nblks) {
+		nresidue = rctx->blk_size;
+		nblks--;
+	}
+
+	rctx->src_sg = req->src;
+	rctx->datbuf.size = (req->nbytes + rctx->residue.size) - nresidue;
+	rctx->total_len += rctx->datbuf.size;
+
+	/*
+	 * If nbytes are less than a block size, copy it residue and
+	 * return. The bytes will be processed in final()
+	 */
+	if (nblks < 1) {
+		scatterwalk_map_and_copy(rctx->residue.buf + rctx->residue.size,
+					 rctx->src_sg, 0, req->nbytes, 0);
+
+		rctx->residue.size += req->nbytes;
+		return 0;
+	}
+
+	/* Copy the previous residue first */
+	if (rctx->residue.size)
+		memcpy(rctx->datbuf.buf, rctx->residue.buf, rctx->residue.size);
+
+	scatterwalk_map_and_copy(rctx->datbuf.buf + rctx->residue.size,
+				 rctx->src_sg, 0, req->nbytes - nresidue, 0);
+
+	scatterwalk_map_and_copy(rctx->residue.buf, rctx->src_sg,
+				 req->nbytes - nresidue, nresidue, 0);
+
+	/* Update residue value with the residue after current block */
+	rctx->residue.size = nresidue;
+
+	rctx->config = tegra_sha_get_config(rctx->alg) |
+			SE_SHA_DST_HASH_REG;
+
+	/*
+	 * If this is not the first 'update' call, paste the previous copied
+	 * intermediate results to the registers so that it gets picked up.
+	 * This is to support the import/export functionality.
+	 */
+	if (!(rctx->task & SHA_FIRST))
+		tegra_sha_paste_hash_result(ctx->se, rctx);
+
+	size = tegra_sha_prep_cmd(ctx->se, cpuvaddr, rctx);
+
+	ret = tegra_se_host1x_submit(ctx->se, size);
+
+	/*
+	 * If this is not the final update, copy the intermediate results
+	 * from the registers so that it can be used in the next 'update'
+	 * call. This is to support the import/export functionality.
+	 */
+	if (!(rctx->task & SHA_FINAL))
+		tegra_sha_copy_hash_result(ctx->se, rctx);
+
+	return ret;
+}
+
+static int tegra_sha_do_final(struct ahash_request *req)
+{
+	struct tegra_sha_reqctx *rctx = ahash_request_ctx(req);
+	struct crypto_ahash *tfm = crypto_ahash_reqtfm(req);
+	struct tegra_sha_ctx *ctx = crypto_ahash_ctx(tfm);
+	struct tegra_se *se = ctx->se;
+	u32 *cpuvaddr = se->cmdbuf->addr;
+	int size, ret = 0;
+
+	memcpy(rctx->datbuf.buf, rctx->residue.buf, rctx->residue.size);
+	rctx->datbuf.size = rctx->residue.size;
+	rctx->total_len += rctx->residue.size;
+
+	rctx->config = tegra_sha_get_config(rctx->alg) |
+		       SE_SHA_DST_MEMORY;
+
+	size = tegra_sha_prep_cmd(se, cpuvaddr, rctx);
+
+	ret = tegra_se_host1x_submit(se, size);
+	if (ret)
+		goto out;
+
+	/* Copy result */
+	memcpy(req->result, rctx->digest.buf, rctx->digest.size);
+
+out:
+	dma_free_coherent(se->dev, SE_SHA_BUFLEN,
+			  rctx->datbuf.buf, rctx->datbuf.addr);
+	dma_free_coherent(se->dev, crypto_ahash_blocksize(tfm),
+			  rctx->residue.buf, rctx->residue.addr);
+	dma_free_coherent(se->dev, rctx->digest.size, rctx->digest.buf,
+			  rctx->digest.addr);
+	return ret;
+}
+
+static int tegra_sha_do_one_req(struct crypto_engine *engine, void *areq)
+{
+	struct ahash_request *req = ahash_request_cast(areq);
+	struct tegra_sha_reqctx *rctx = ahash_request_ctx(req);
+	struct crypto_ahash *tfm = crypto_ahash_reqtfm(req);
+	struct tegra_sha_ctx *ctx = crypto_ahash_ctx(tfm);
+	struct tegra_se *se = ctx->se;
+	int ret = 0;
+
+	if (rctx->task & SHA_UPDATE) {
+		ret = tegra_sha_do_update(req);
+		rctx->task &= ~SHA_UPDATE;
+	}
+
+	if (rctx->task & SHA_FINAL) {
+		ret = tegra_sha_do_final(req);
+		rctx->task &= ~SHA_FINAL;
+	}
+
+	crypto_finalize_hash_request(se->engine, req, ret);
+
+	return 0;
+}
+
+static void tegra_sha_init_fallback(struct crypto_ahash *tfm, struct tegra_sha_ctx *ctx,
+				    const char *algname)
+{
+	unsigned int statesize;
+
+	ctx->fallback_tfm = crypto_alloc_ahash(algname, 0, CRYPTO_ALG_ASYNC |
+						CRYPTO_ALG_NEED_FALLBACK);
+
+	if (IS_ERR(ctx->fallback_tfm)) {
+		dev_warn(ctx->se->dev,
+			 "failed to allocate fallback for %s\n", algname);
+		ctx->fallback_tfm = NULL;
+		return;
+	}
+
+	statesize = crypto_ahash_statesize(ctx->fallback_tfm);
+
+	if (statesize > sizeof(struct tegra_sha_reqctx))
+		crypto_ahash_set_statesize(tfm, statesize);
+
+	/* Update reqsize if fallback is added */
+	crypto_ahash_set_reqsize(tfm,
+				 sizeof(struct tegra_sha_reqctx) +
+			crypto_ahash_reqsize(ctx->fallback_tfm));
+}
+
+static int tegra_sha_cra_init(struct crypto_tfm *tfm)
+{
+	struct tegra_sha_ctx *ctx = crypto_tfm_ctx(tfm);
+	struct crypto_ahash *ahash_tfm = __crypto_ahash_cast(tfm);
+	struct ahash_alg *alg = __crypto_ahash_alg(tfm->__crt_alg);
+	struct tegra_se_alg *se_alg;
+	const char *algname;
+	int ret;
+
+	algname = crypto_tfm_alg_name(tfm);
+	se_alg = container_of(alg, struct tegra_se_alg, alg.ahash.base);
+
+	crypto_ahash_set_reqsize(ahash_tfm, sizeof(struct tegra_sha_reqctx));
+
+	ctx->se = se_alg->se_dev;
+	ctx->fallback = false;
+	ctx->key_id = 0;
+
+	ret = se_algname_to_algid(algname);
+	if (ret < 0) {
+		dev_err(ctx->se->dev, "invalid algorithm\n");
+		return ret;
+	}
+
+	if (se_alg->alg_base)
+		tegra_sha_init_fallback(ahash_tfm, ctx, algname);
+
+	ctx->alg = ret;
+
+	return 0;
+}
+
+static void tegra_sha_cra_exit(struct crypto_tfm *tfm)
+{
+	struct tegra_sha_ctx *ctx = crypto_tfm_ctx(tfm);
+
+	if (ctx->fallback_tfm)
+		crypto_free_ahash(ctx->fallback_tfm);
+
+	tegra_key_invalidate(ctx->se, ctx->key_id, ctx->alg);
+}
+
+static int tegra_sha_init(struct ahash_request *req)
+{
+	struct tegra_sha_reqctx *rctx = ahash_request_ctx(req);
+	struct crypto_ahash *tfm = crypto_ahash_reqtfm(req);
+	struct tegra_sha_ctx *ctx = crypto_ahash_ctx(tfm);
+	struct tegra_se *se = ctx->se;
+
+	if (ctx->fallback)
+		return tegra_sha_fallback_init(req);
+
+	rctx->total_len = 0;
+	rctx->datbuf.size = 0;
+	rctx->residue.size = 0;
+	rctx->key_id = ctx->key_id;
+	rctx->task = SHA_FIRST;
+	rctx->alg = ctx->alg;
+	rctx->blk_size = crypto_ahash_blocksize(tfm);
+	rctx->digest.size = crypto_ahash_digestsize(tfm);
+
+	rctx->digest.buf = dma_alloc_coherent(se->dev, rctx->digest.size,
+					      &rctx->digest.addr, GFP_KERNEL);
+	if (!rctx->digest.buf)
+		goto digbuf_fail;
+
+	rctx->residue.buf = dma_alloc_coherent(se->dev, rctx->blk_size,
+					       &rctx->residue.addr, GFP_KERNEL);
+	if (!rctx->residue.buf)
+		goto resbuf_fail;
+
+	rctx->datbuf.buf = dma_alloc_coherent(se->dev, SE_SHA_BUFLEN,
+					      &rctx->datbuf.addr, GFP_KERNEL);
+	if (!rctx->datbuf.buf)
+		goto datbuf_fail;
+
+	return 0;
+
+datbuf_fail:
+	dma_free_coherent(se->dev, rctx->blk_size, rctx->residue.buf,
+			  rctx->residue.addr);
+resbuf_fail:
+	dma_free_coherent(se->dev, SE_SHA_BUFLEN, rctx->datbuf.buf,
+			  rctx->datbuf.addr);
+digbuf_fail:
+	return -ENOMEM;
+}
+
+static int tegra_hmac_fallback_setkey(struct tegra_sha_ctx *ctx, const u8 *key,
+				      unsigned int keylen)
+{
+	if (!ctx->fallback_tfm) {
+		dev_dbg(ctx->se->dev, "invalid key length (%d)\n", keylen);
+		return -EINVAL;
+	}
+
+	ctx->fallback = true;
+	return crypto_ahash_setkey(ctx->fallback_tfm, key, keylen);
+}
+
+static int tegra_hmac_setkey(struct crypto_ahash *tfm, const u8 *key,
+			     unsigned int keylen)
+{
+	struct tegra_sha_ctx *ctx = crypto_ahash_ctx(tfm);
+
+	if (aes_check_keylen(keylen))
+		return tegra_hmac_fallback_setkey(ctx, key, keylen);
+
+	ctx->fallback = false;
+
+	return tegra_key_submit(ctx->se, key, keylen, ctx->alg, &ctx->key_id);
+}
+
+static int tegra_sha_update(struct ahash_request *req)
+{
+	struct tegra_sha_reqctx *rctx = ahash_request_ctx(req);
+	struct crypto_ahash *tfm = crypto_ahash_reqtfm(req);
+	struct tegra_sha_ctx *ctx = crypto_ahash_ctx(tfm);
+
+	if (ctx->fallback)
+		return tegra_sha_fallback_update(req);
+
+	rctx->task |= SHA_UPDATE;
+
+	return crypto_transfer_hash_request_to_engine(ctx->se->engine, req);
+}
+
+static int tegra_sha_final(struct ahash_request *req)
+{
+	struct tegra_sha_reqctx *rctx = ahash_request_ctx(req);
+	struct crypto_ahash *tfm = crypto_ahash_reqtfm(req);
+	struct tegra_sha_ctx *ctx = crypto_ahash_ctx(tfm);
+
+	if (ctx->fallback)
+		return tegra_sha_fallback_final(req);
+
+	rctx->task |= SHA_FINAL;
+
+	return crypto_transfer_hash_request_to_engine(ctx->se->engine, req);
+}
+
+static int tegra_sha_finup(struct ahash_request *req)
+{
+	struct tegra_sha_reqctx *rctx = ahash_request_ctx(req);
+	struct crypto_ahash *tfm = crypto_ahash_reqtfm(req);
+	struct tegra_sha_ctx *ctx = crypto_ahash_ctx(tfm);
+
+	if (ctx->fallback)
+		return tegra_sha_fallback_finup(req);
+
+	rctx->task |= SHA_UPDATE | SHA_FINAL;
+
+	return crypto_transfer_hash_request_to_engine(ctx->se->engine, req);
+}
+
+static int tegra_sha_digest(struct ahash_request *req)
+{
+	struct tegra_sha_reqctx *rctx = ahash_request_ctx(req);
+	struct crypto_ahash *tfm = crypto_ahash_reqtfm(req);
+	struct tegra_sha_ctx *ctx = crypto_ahash_ctx(tfm);
+
+	if (ctx->fallback)
+		return tegra_sha_fallback_digest(req);
+
+	tegra_sha_init(req);
+	rctx->task |= SHA_UPDATE | SHA_FINAL;
+
+	return crypto_transfer_hash_request_to_engine(ctx->se->engine, req);
+}
+
+static int tegra_sha_export(struct ahash_request *req, void *out)
+{
+	struct tegra_sha_reqctx *rctx = ahash_request_ctx(req);
+	struct crypto_ahash *tfm = crypto_ahash_reqtfm(req);
+	struct tegra_sha_ctx *ctx = crypto_ahash_ctx(tfm);
+
+	if (ctx->fallback)
+		return tegra_sha_fallback_export(req, out);
+
+	memcpy(out, rctx, sizeof(*rctx));
+
+	return 0;
+}
+
+static int tegra_sha_import(struct ahash_request *req, const void *in)
+{
+	struct tegra_sha_reqctx *rctx = ahash_request_ctx(req);
+	struct crypto_ahash *tfm = crypto_ahash_reqtfm(req);
+	struct tegra_sha_ctx *ctx = crypto_ahash_ctx(tfm);
+
+	if (ctx->fallback)
+		return tegra_sha_fallback_import(req, in);
+
+	memcpy(rctx, in, sizeof(*rctx));
+
+	return 0;
+}
+
+static struct tegra_se_alg tegra_hash_algs[] = {
+	{
+		.alg.ahash.op.do_one_request = tegra_sha_do_one_req,
+		.alg.ahash.base = {
+			.init = tegra_sha_init,
+			.update = tegra_sha_update,
+			.final = tegra_sha_final,
+			.finup = tegra_sha_finup,
+			.digest = tegra_sha_digest,
+			.export = tegra_sha_export,
+			.import = tegra_sha_import,
+			.halg.digestsize = SHA1_DIGEST_SIZE,
+			.halg.statesize = sizeof(struct tegra_sha_reqctx),
+			.halg.base = {
+				.cra_name = "sha1",
+				.cra_driver_name = "tegra-se-sha1",
+				.cra_priority = 300,
+				.cra_flags = CRYPTO_ALG_TYPE_AHASH,
+				.cra_blocksize = SHA1_BLOCK_SIZE,
+				.cra_ctxsize = sizeof(struct tegra_sha_ctx),
+				.cra_alignmask = 0,
+				.cra_module = THIS_MODULE,
+				.cra_init = tegra_sha_cra_init,
+				.cra_exit = tegra_sha_cra_exit,
+			}
+		}
+	}, {
+		.alg.ahash.op.do_one_request = tegra_sha_do_one_req,
+		.alg.ahash.base = {
+			.init = tegra_sha_init,
+			.update = tegra_sha_update,
+			.final = tegra_sha_final,
+			.finup = tegra_sha_finup,
+			.digest = tegra_sha_digest,
+			.export = tegra_sha_export,
+			.import = tegra_sha_import,
+			.halg.digestsize = SHA224_DIGEST_SIZE,
+			.halg.statesize = sizeof(struct tegra_sha_reqctx),
+			.halg.base = {
+				.cra_name = "sha224",
+				.cra_driver_name = "tegra-se-sha224",
+				.cra_priority = 300,
+				.cra_flags = CRYPTO_ALG_TYPE_AHASH,
+				.cra_blocksize = SHA224_BLOCK_SIZE,
+				.cra_ctxsize = sizeof(struct tegra_sha_ctx),
+				.cra_alignmask = 0,
+				.cra_module = THIS_MODULE,
+				.cra_init = tegra_sha_cra_init,
+				.cra_exit = tegra_sha_cra_exit,
+			}
+		}
+	}, {
+		.alg.ahash.op.do_one_request = tegra_sha_do_one_req,
+		.alg.ahash.base = {
+			.init = tegra_sha_init,
+			.update = tegra_sha_update,
+			.final = tegra_sha_final,
+			.finup = tegra_sha_finup,
+			.digest = tegra_sha_digest,
+			.export = tegra_sha_export,
+			.import = tegra_sha_import,
+			.halg.digestsize = SHA256_DIGEST_SIZE,
+			.halg.statesize = sizeof(struct tegra_sha_reqctx),
+			.halg.base = {
+				.cra_name = "sha256",
+				.cra_driver_name = "tegra-se-sha256",
+				.cra_priority = 300,
+				.cra_flags = CRYPTO_ALG_TYPE_AHASH,
+				.cra_blocksize = SHA256_BLOCK_SIZE,
+				.cra_ctxsize = sizeof(struct tegra_sha_ctx),
+				.cra_alignmask = 0,
+				.cra_module = THIS_MODULE,
+				.cra_init = tegra_sha_cra_init,
+				.cra_exit = tegra_sha_cra_exit,
+			}
+		}
+	}, {
+		.alg.ahash.op.do_one_request = tegra_sha_do_one_req,
+		.alg.ahash.base = {
+			.init = tegra_sha_init,
+			.update = tegra_sha_update,
+			.final = tegra_sha_final,
+			.finup = tegra_sha_finup,
+			.digest = tegra_sha_digest,
+			.export = tegra_sha_export,
+			.import = tegra_sha_import,
+			.halg.digestsize = SHA384_DIGEST_SIZE,
+			.halg.statesize = sizeof(struct tegra_sha_reqctx),
+			.halg.base = {
+				.cra_name = "sha384",
+				.cra_driver_name = "tegra-se-sha384",
+				.cra_priority = 300,
+				.cra_flags = CRYPTO_ALG_TYPE_AHASH,
+				.cra_blocksize = SHA384_BLOCK_SIZE,
+				.cra_ctxsize = sizeof(struct tegra_sha_ctx),
+				.cra_alignmask = 0,
+				.cra_module = THIS_MODULE,
+				.cra_init = tegra_sha_cra_init,
+				.cra_exit = tegra_sha_cra_exit,
+			}
+		}
+	}, {
+		.alg.ahash.op.do_one_request = tegra_sha_do_one_req,
+		.alg.ahash.base = {
+			.init = tegra_sha_init,
+			.update = tegra_sha_update,
+			.final = tegra_sha_final,
+			.finup = tegra_sha_finup,
+			.digest = tegra_sha_digest,
+			.export = tegra_sha_export,
+			.import = tegra_sha_import,
+			.halg.digestsize = SHA512_DIGEST_SIZE,
+			.halg.statesize = sizeof(struct tegra_sha_reqctx),
+			.halg.base = {
+				.cra_name = "sha512",
+				.cra_driver_name = "tegra-se-sha512",
+				.cra_priority = 300,
+				.cra_flags = CRYPTO_ALG_TYPE_AHASH,
+				.cra_blocksize = SHA512_BLOCK_SIZE,
+				.cra_ctxsize = sizeof(struct tegra_sha_ctx),
+				.cra_alignmask = 0,
+				.cra_module = THIS_MODULE,
+				.cra_init = tegra_sha_cra_init,
+				.cra_exit = tegra_sha_cra_exit,
+			}
+		}
+	}, {
+		.alg.ahash.op.do_one_request = tegra_sha_do_one_req,
+		.alg.ahash.base = {
+			.init = tegra_sha_init,
+			.update = tegra_sha_update,
+			.final = tegra_sha_final,
+			.finup = tegra_sha_finup,
+			.digest = tegra_sha_digest,
+			.export = tegra_sha_export,
+			.import = tegra_sha_import,
+			.halg.digestsize = SHA3_224_DIGEST_SIZE,
+			.halg.statesize = sizeof(struct tegra_sha_reqctx),
+			.halg.base = {
+				.cra_name = "sha3-224",
+				.cra_driver_name = "tegra-se-sha3-224",
+				.cra_priority = 300,
+				.cra_flags = CRYPTO_ALG_TYPE_AHASH,
+				.cra_blocksize = SHA3_224_BLOCK_SIZE,
+				.cra_ctxsize = sizeof(struct tegra_sha_ctx),
+				.cra_alignmask = 0,
+				.cra_module = THIS_MODULE,
+				.cra_init = tegra_sha_cra_init,
+				.cra_exit = tegra_sha_cra_exit,
+			}
+		}
+	}, {
+		.alg.ahash.op.do_one_request = tegra_sha_do_one_req,
+		.alg.ahash.base = {
+			.init = tegra_sha_init,
+			.update = tegra_sha_update,
+			.final = tegra_sha_final,
+			.finup = tegra_sha_finup,
+			.digest = tegra_sha_digest,
+			.export = tegra_sha_export,
+			.import = tegra_sha_import,
+			.halg.digestsize = SHA3_256_DIGEST_SIZE,
+			.halg.statesize = sizeof(struct tegra_sha_reqctx),
+			.halg.base = {
+				.cra_name = "sha3-256",
+				.cra_driver_name = "tegra-se-sha3-256",
+				.cra_priority = 300,
+				.cra_flags = CRYPTO_ALG_TYPE_AHASH,
+				.cra_blocksize = SHA3_256_BLOCK_SIZE,
+				.cra_ctxsize = sizeof(struct tegra_sha_ctx),
+				.cra_alignmask = 0,
+				.cra_module = THIS_MODULE,
+				.cra_init = tegra_sha_cra_init,
+				.cra_exit = tegra_sha_cra_exit,
+			}
+		}
+	}, {
+		.alg.ahash.op.do_one_request = tegra_sha_do_one_req,
+		.alg.ahash.base = {
+			.init = tegra_sha_init,
+			.update = tegra_sha_update,
+			.final = tegra_sha_final,
+			.finup = tegra_sha_finup,
+			.digest = tegra_sha_digest,
+			.export = tegra_sha_export,
+			.import = tegra_sha_import,
+			.halg.digestsize = SHA3_384_DIGEST_SIZE,
+			.halg.statesize = sizeof(struct tegra_sha_reqctx),
+			.halg.base = {
+				.cra_name = "sha3-384",
+				.cra_driver_name = "tegra-se-sha3-384",
+				.cra_priority = 300,
+				.cra_flags = CRYPTO_ALG_TYPE_AHASH,
+				.cra_blocksize = SHA3_384_BLOCK_SIZE,
+				.cra_ctxsize = sizeof(struct tegra_sha_ctx),
+				.cra_alignmask = 0,
+				.cra_module = THIS_MODULE,
+				.cra_init = tegra_sha_cra_init,
+				.cra_exit = tegra_sha_cra_exit,
+			}
+		}
+	}, {
+		.alg.ahash.op.do_one_request = tegra_sha_do_one_req,
+		.alg.ahash.base = {
+			.init = tegra_sha_init,
+			.update = tegra_sha_update,
+			.final = tegra_sha_final,
+			.finup = tegra_sha_finup,
+			.digest = tegra_sha_digest,
+			.export = tegra_sha_export,
+			.import = tegra_sha_import,
+			.halg.digestsize = SHA3_512_DIGEST_SIZE,
+			.halg.statesize = sizeof(struct tegra_sha_reqctx),
+			.halg.base = {
+				.cra_name = "sha3-512",
+				.cra_driver_name = "tegra-se-sha3-512",
+				.cra_priority = 300,
+				.cra_flags = CRYPTO_ALG_TYPE_AHASH,
+				.cra_blocksize = SHA3_512_BLOCK_SIZE,
+				.cra_ctxsize = sizeof(struct tegra_sha_ctx),
+				.cra_alignmask = 0,
+				.cra_module = THIS_MODULE,
+				.cra_init = tegra_sha_cra_init,
+				.cra_exit = tegra_sha_cra_exit,
+			}
+		}
+	}, {
+		.alg_base = "sha224",
+		.alg.ahash.op.do_one_request = tegra_sha_do_one_req,
+		.alg.ahash.base = {
+			.init = tegra_sha_init,
+			.update = tegra_sha_update,
+			.final = tegra_sha_final,
+			.finup = tegra_sha_finup,
+			.digest = tegra_sha_digest,
+			.export = tegra_sha_export,
+			.import = tegra_sha_import,
+			.setkey = tegra_hmac_setkey,
+			.halg.digestsize = SHA224_DIGEST_SIZE,
+			.halg.statesize = sizeof(struct tegra_sha_reqctx),
+			.halg.base = {
+				.cra_name = "hmac(sha224)",
+				.cra_driver_name = "tegra-se-hmac-sha224",
+				.cra_priority = 300,
+				.cra_flags = CRYPTO_ALG_TYPE_AHASH | CRYPTO_ALG_NEED_FALLBACK,
+				.cra_blocksize = SHA224_BLOCK_SIZE,
+				.cra_ctxsize = sizeof(struct tegra_sha_ctx),
+				.cra_alignmask = 0,
+				.cra_module = THIS_MODULE,
+				.cra_init = tegra_sha_cra_init,
+				.cra_exit = tegra_sha_cra_exit,
+			}
+		}
+	}, {
+		.alg_base = "sha256",
+		.alg.ahash.op.do_one_request = tegra_sha_do_one_req,
+		.alg.ahash.base = {
+			.init = tegra_sha_init,
+			.update = tegra_sha_update,
+			.final = tegra_sha_final,
+			.finup = tegra_sha_finup,
+			.digest = tegra_sha_digest,
+			.export = tegra_sha_export,
+			.import = tegra_sha_import,
+			.setkey = tegra_hmac_setkey,
+			.halg.digestsize = SHA256_DIGEST_SIZE,
+			.halg.statesize = sizeof(struct tegra_sha_reqctx),
+			.halg.base = {
+				.cra_name = "hmac(sha256)",
+				.cra_driver_name = "tegra-se-hmac-sha256",
+				.cra_priority = 300,
+				.cra_flags = CRYPTO_ALG_TYPE_AHASH | CRYPTO_ALG_NEED_FALLBACK,
+				.cra_blocksize = SHA256_BLOCK_SIZE,
+				.cra_ctxsize = sizeof(struct tegra_sha_ctx),
+				.cra_alignmask = 0,
+				.cra_module = THIS_MODULE,
+				.cra_init = tegra_sha_cra_init,
+				.cra_exit = tegra_sha_cra_exit,
+			}
+		}
+	}, {
+		.alg_base = "sha384",
+		.alg.ahash.op.do_one_request = tegra_sha_do_one_req,
+		.alg.ahash.base = {
+			.init = tegra_sha_init,
+			.update = tegra_sha_update,
+			.final = tegra_sha_final,
+			.finup = tegra_sha_finup,
+			.digest = tegra_sha_digest,
+			.export = tegra_sha_export,
+			.import = tegra_sha_import,
+			.setkey = tegra_hmac_setkey,
+			.halg.digestsize = SHA384_DIGEST_SIZE,
+			.halg.statesize = sizeof(struct tegra_sha_reqctx),
+			.halg.base = {
+				.cra_name = "hmac(sha384)",
+				.cra_driver_name = "tegra-se-hmac-sha384",
+				.cra_priority = 300,
+				.cra_flags = CRYPTO_ALG_TYPE_AHASH | CRYPTO_ALG_NEED_FALLBACK,
+				.cra_blocksize = SHA384_BLOCK_SIZE,
+				.cra_ctxsize = sizeof(struct tegra_sha_ctx),
+				.cra_alignmask = 0,
+				.cra_module = THIS_MODULE,
+				.cra_init = tegra_sha_cra_init,
+				.cra_exit = tegra_sha_cra_exit,
+			}
+		}
+	}, {
+		.alg_base = "sha512",
+		.alg.ahash.op.do_one_request = tegra_sha_do_one_req,
+		.alg.ahash.base = {
+			.init = tegra_sha_init,
+			.update = tegra_sha_update,
+			.final = tegra_sha_final,
+			.finup = tegra_sha_finup,
+			.digest = tegra_sha_digest,
+			.export = tegra_sha_export,
+			.import = tegra_sha_import,
+			.setkey = tegra_hmac_setkey,
+			.halg.digestsize = SHA512_DIGEST_SIZE,
+			.halg.statesize = sizeof(struct tegra_sha_reqctx),
+			.halg.base = {
+				.cra_name = "hmac(sha512)",
+				.cra_driver_name = "tegra-se-hmac-sha512",
+				.cra_priority = 300,
+				.cra_flags = CRYPTO_ALG_TYPE_AHASH | CRYPTO_ALG_NEED_FALLBACK,
+				.cra_blocksize = SHA512_BLOCK_SIZE,
+				.cra_ctxsize = sizeof(struct tegra_sha_ctx),
+				.cra_alignmask = 0,
+				.cra_module = THIS_MODULE,
+				.cra_init = tegra_sha_cra_init,
+				.cra_exit = tegra_sha_cra_exit,
+			}
+		}
+	}
+};
+
+static int tegra_hash_kac_manifest(u32 user, u32 alg, u32 keylen)
+{
+	int manifest;
+
+	manifest = SE_KAC_USER_NS;
+
+	switch (alg) {
+	case SE_ALG_HMAC_SHA224:
+	case SE_ALG_HMAC_SHA256:
+	case SE_ALG_HMAC_SHA384:
+	case SE_ALG_HMAC_SHA512:
+		manifest |= SE_KAC_HMAC;
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	switch (keylen) {
+	case AES_KEYSIZE_128:
+		manifest |= SE_KAC_SIZE_128;
+		break;
+	case AES_KEYSIZE_192:
+		manifest |= SE_KAC_SIZE_192;
+		break;
+	case AES_KEYSIZE_256:
+	default:
+		manifest |= SE_KAC_SIZE_256;
+		break;
+	}
+
+	return manifest;
+}
+
+int tegra_init_hash(struct tegra_se *se)
+{
+	struct ahash_engine_alg *alg;
+	int i, ret;
+
+	se->manifest = tegra_hash_kac_manifest;
+
+	for (i = 0; i < ARRAY_SIZE(tegra_hash_algs); i++) {
+		tegra_hash_algs[i].se_dev = se;
+		alg = &tegra_hash_algs[i].alg.ahash;
+
+		ret = crypto_engine_register_ahash(alg);
+		if (ret) {
+			dev_err(se->dev, "failed to register %s\n",
+				alg->base.halg.base.cra_name);
+			goto sha_err;
+		}
+	}
+
+	return 0;
+
+sha_err:
+	while (i--)
+		crypto_engine_unregister_ahash(&tegra_hash_algs[i].alg.ahash);
+
+	return ret;
+}
+
+void tegra_deinit_hash(struct tegra_se *se)
+{
+	int i;
+
+	for (i = 0; i < ARRAY_SIZE(tegra_hash_algs); i++)
+		crypto_engine_unregister_ahash(&tegra_hash_algs[i].alg.ahash);
+}
diff --git a/drivers/crypto/tegra/tegra-se-key.c b/drivers/crypto/tegra/tegra-se-key.c
new file mode 100644
index 000000000000..ac14678dbd30
--- /dev/null
+++ b/drivers/crypto/tegra/tegra-se-key.c
@@ -0,0 +1,156 @@
+// SPDX-License-Identifier: GPL-2.0-only
+// SPDX-FileCopyrightText: Copyright (c) 2023 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
+/*
+ * Crypto driver file to manage keys of NVIDIA Security Engine.
+ */
+
+#include <linux/bitops.h>
+#include <linux/module.h>
+#include <crypto/aes.h>
+
+#include "tegra-se.h"
+
+#define SE_KEY_FULL_MASK		GENMASK(SE_MAX_KEYSLOT, 0)
+
+/* Reserve keyslot 0, 14, 15 */
+#define SE_KEY_RSVD_MASK		(BIT(0) | BIT(14) | BIT(15))
+#define SE_KEY_VALID_MASK		(SE_KEY_FULL_MASK & ~SE_KEY_RSVD_MASK)
+
+/* Mutex lock to guard keyslots */
+static DEFINE_MUTEX(kslt_lock);
+
+/* Keyslot bitmask (0 = available, 1 = in use/not available) */
+static u16 tegra_se_keyslots = SE_KEY_RSVD_MASK;
+
+static u16 tegra_keyslot_alloc(void)
+{
+	u16 keyid;
+
+	mutex_lock(&kslt_lock);
+	/* Check if all key slots are full */
+	if (tegra_se_keyslots == GENMASK(SE_MAX_KEYSLOT, 0)) {
+		mutex_unlock(&kslt_lock);
+		return 0;
+	}
+
+	keyid = ffz(tegra_se_keyslots);
+	tegra_se_keyslots |= BIT(keyid);
+
+	mutex_unlock(&kslt_lock);
+
+	return keyid;
+}
+
+static void tegra_keyslot_free(u16 slot)
+{
+	mutex_lock(&kslt_lock);
+	tegra_se_keyslots &= ~(BIT(slot));
+	mutex_unlock(&kslt_lock);
+}
+
+static unsigned int tegra_key_prep_ins_cmd(struct tegra_se *se, u32 *cpuvaddr,
+					   const u32 *key, u32 keylen, u16 slot, u32 alg)
+{
+	int i = 0, j;
+
+	cpuvaddr[i++] = host1x_opcode_setpayload(1);
+	cpuvaddr[i++] = se_host1x_opcode_incr_w(se->hw->regs->op);
+	cpuvaddr[i++] = SE_AES_OP_WRSTALL | SE_AES_OP_DUMMY;
+
+	cpuvaddr[i++] = host1x_opcode_setpayload(1);
+	cpuvaddr[i++] = se_host1x_opcode_incr_w(se->hw->regs->manifest);
+	cpuvaddr[i++] = se->manifest(se->owner, alg, keylen);
+	cpuvaddr[i++] = host1x_opcode_setpayload(1);
+	cpuvaddr[i++] = se_host1x_opcode_incr_w(se->hw->regs->key_dst);
+
+	cpuvaddr[i++] = SE_AES_KEY_DST_INDEX(slot);
+
+	for (j = 0; j < keylen / 4; j++) {
+		/* Set key address */
+		cpuvaddr[i++] = host1x_opcode_setpayload(1);
+		cpuvaddr[i++] = se_host1x_opcode_incr_w(se->hw->regs->key_addr);
+		cpuvaddr[i++] = j;
+
+		/* Set key data */
+		cpuvaddr[i++] = host1x_opcode_setpayload(1);
+		cpuvaddr[i++] = se_host1x_opcode_incr_w(se->hw->regs->key_data);
+		cpuvaddr[i++] = key[j];
+	}
+
+	cpuvaddr[i++] = host1x_opcode_setpayload(1);
+	cpuvaddr[i++] = se_host1x_opcode_incr_w(se->hw->regs->config);
+	cpuvaddr[i++] = SE_CFG_INS;
+
+	cpuvaddr[i++] = host1x_opcode_setpayload(1);
+	cpuvaddr[i++] = se_host1x_opcode_incr_w(se->hw->regs->op);
+	cpuvaddr[i++] = SE_AES_OP_WRSTALL | SE_AES_OP_START |
+			SE_AES_OP_LASTBUF;
+
+	cpuvaddr[i++] = se_host1x_opcode_nonincr(host1x_uclass_incr_syncpt_r(), 1);
+	cpuvaddr[i++] = host1x_uclass_incr_syncpt_cond_f(1) |
+			host1x_uclass_incr_syncpt_indx_f(se->syncpt_id);
+
+	dev_dbg(se->dev, "key-slot %u key-manifest %#x\n",
+		slot, se->manifest(se->owner, alg, keylen));
+
+	return i;
+}
+
+static bool tegra_key_in_kslt(u32 keyid)
+{
+	bool ret;
+
+	if (keyid > SE_MAX_KEYSLOT)
+		return false;
+
+	mutex_lock(&kslt_lock);
+	ret = ((BIT(keyid) & SE_KEY_VALID_MASK) &&
+		(BIT(keyid) & tegra_se_keyslots));
+	mutex_unlock(&kslt_lock);
+
+	return ret;
+}
+
+static int tegra_key_insert(struct tegra_se *se, const u8 *key,
+			    u32 keylen, u16 slot, u32 alg)
+{
+	const u32 *keyval = (u32 *)key;
+	u32 *addr = se->cmdbuf->addr, size;
+
+	size = tegra_key_prep_ins_cmd(se, addr, keyval, keylen, slot, alg);
+
+	return tegra_se_host1x_submit(se, size);
+}
+
+void tegra_key_invalidate(struct tegra_se *se, u32 keyid, u32 alg)
+{
+	u8 zkey[AES_MAX_KEY_SIZE] = {0};
+
+	if (!keyid)
+		return;
+
+	/* Overwrite the key with 0s */
+	tegra_key_insert(se, zkey, AES_MAX_KEY_SIZE, keyid, alg);
+
+	tegra_keyslot_free(keyid);
+}
+
+int tegra_key_submit(struct tegra_se *se, const u8 *key, u32 keylen, u32 alg, u32 *keyid)
+{
+	int ret;
+
+	/* Use the existing slot if it is already allocated */
+	if (!tegra_key_in_kslt(*keyid)) {
+		*keyid = tegra_keyslot_alloc();
+		if (!(*keyid)) {
+			dev_err(se->dev, "failed to allocate key slot\n");
+			return -ENOMEM;
+		}
+	}
+
+	ret = tegra_key_insert(se, key, keylen, *keyid, alg);
+	if (ret)
+		return ret;
+
+	return 0;
+}
diff --git a/drivers/crypto/tegra/tegra-se-main.c b/drivers/crypto/tegra/tegra-se-main.c
new file mode 100644
index 000000000000..9955874b3dc3
--- /dev/null
+++ b/drivers/crypto/tegra/tegra-se-main.c
@@ -0,0 +1,437 @@
+// SPDX-License-Identifier: GPL-2.0-only
+// SPDX-FileCopyrightText: Copyright (c) 2023 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
+/*
+ * Crypto driver for NVIDIA Security Engine in Tegra Chips
+ */
+
+#include <linux/clk.h>
+#include <linux/dma-mapping.h>
+#include <linux/module.h>
+#include <linux/platform_device.h>
+#include <linux/mod_devicetable.h>
+
+#include <crypto/engine.h>
+
+#include "tegra-se.h"
+
+static struct host1x_bo *tegra_se_cmdbuf_get(struct host1x_bo *host_bo)
+{
+	struct tegra_se_cmdbuf *cmdbuf = container_of(host_bo, struct tegra_se_cmdbuf, bo);
+
+	kref_get(&cmdbuf->ref);
+
+	return host_bo;
+}
+
+static void tegra_se_cmdbuf_release(struct kref *ref)
+{
+	struct tegra_se_cmdbuf *cmdbuf = container_of(ref, struct tegra_se_cmdbuf, ref);
+
+	dma_free_attrs(cmdbuf->dev, cmdbuf->size, cmdbuf->addr,
+		       cmdbuf->iova, 0);
+
+	kfree(cmdbuf);
+}
+
+static void tegra_se_cmdbuf_put(struct host1x_bo *host_bo)
+{
+	struct tegra_se_cmdbuf *cmdbuf = container_of(host_bo, struct tegra_se_cmdbuf, bo);
+
+	kref_put(&cmdbuf->ref, tegra_se_cmdbuf_release);
+}
+
+static struct host1x_bo_mapping *
+tegra_se_cmdbuf_pin(struct device *dev, struct host1x_bo *bo, enum dma_data_direction direction)
+{
+	struct tegra_se_cmdbuf *cmdbuf = container_of(bo, struct tegra_se_cmdbuf, bo);
+	struct host1x_bo_mapping *map;
+	int err;
+
+	map = kzalloc(sizeof(*map), GFP_KERNEL);
+	if (!map)
+		return ERR_PTR(-ENOMEM);
+
+	kref_init(&map->ref);
+	map->bo = host1x_bo_get(bo);
+	map->direction = direction;
+	map->dev = dev;
+
+	map->sgt = kzalloc(sizeof(*map->sgt), GFP_KERNEL);
+	if (!map->sgt) {
+		err = -ENOMEM;
+		goto free;
+	}
+
+	err = dma_get_sgtable(dev, map->sgt, cmdbuf->addr,
+			      cmdbuf->iova, cmdbuf->words * 4);
+	if (err)
+		goto free_sgt;
+
+	err = dma_map_sgtable(dev, map->sgt, direction, 0);
+	if (err)
+		goto free_sgt;
+
+	map->phys = sg_dma_address(map->sgt->sgl);
+	map->size = cmdbuf->words * 4;
+	map->chunks = err;
+
+	return map;
+
+free_sgt:
+	sg_free_table(map->sgt);
+	kfree(map->sgt);
+free:
+	kfree(map);
+	return ERR_PTR(err);
+}
+
+static void tegra_se_cmdbuf_unpin(struct host1x_bo_mapping *map)
+{
+	if (!map)
+		return;
+
+	dma_unmap_sgtable(map->dev, map->sgt, map->direction, 0);
+	sg_free_table(map->sgt);
+	kfree(map->sgt);
+	host1x_bo_put(map->bo);
+
+	kfree(map);
+}
+
+static void *tegra_se_cmdbuf_mmap(struct host1x_bo *host_bo)
+{
+	struct tegra_se_cmdbuf *cmdbuf = container_of(host_bo, struct tegra_se_cmdbuf, bo);
+
+	return cmdbuf->addr;
+}
+
+static void tegra_se_cmdbuf_munmap(struct host1x_bo *host_bo, void *addr)
+{
+}
+
+static const struct host1x_bo_ops tegra_se_cmdbuf_ops = {
+	.get = tegra_se_cmdbuf_get,
+	.put = tegra_se_cmdbuf_put,
+	.pin = tegra_se_cmdbuf_pin,
+	.unpin = tegra_se_cmdbuf_unpin,
+	.mmap = tegra_se_cmdbuf_mmap,
+	.munmap = tegra_se_cmdbuf_munmap,
+};
+
+static struct tegra_se_cmdbuf *tegra_se_host1x_bo_alloc(struct tegra_se *se, ssize_t size)
+{
+	struct tegra_se_cmdbuf *cmdbuf;
+	struct device *dev = se->dev->parent;
+
+	cmdbuf = kzalloc(sizeof(*cmdbuf), GFP_KERNEL);
+	if (!cmdbuf)
+		return NULL;
+
+	cmdbuf->addr = dma_alloc_attrs(dev, size, &cmdbuf->iova,
+				       GFP_KERNEL, 0);
+	if (!cmdbuf->addr)
+		return NULL;
+
+	cmdbuf->size = size;
+	cmdbuf->dev  = dev;
+
+	host1x_bo_init(&cmdbuf->bo, &tegra_se_cmdbuf_ops);
+	kref_init(&cmdbuf->ref);
+
+	return cmdbuf;
+}
+
+int tegra_se_host1x_submit(struct tegra_se *se, u32 size)
+{
+	struct host1x_job *job;
+	int ret;
+
+	job = host1x_job_alloc(se->channel, 1, 0, true);
+	if (!job) {
+		dev_err(se->dev, "failed to allocate host1x job\n");
+		return -ENOMEM;
+	}
+
+	job->syncpt = host1x_syncpt_get(se->syncpt);
+	job->syncpt_incrs = 1;
+	job->client = &se->client;
+	job->class = se->client.class;
+	job->serialize = true;
+	job->engine_fallback_streamid = se->stream_id;
+	job->engine_streamid_offset = SE_STREAM_ID;
+
+	se->cmdbuf->words = size;
+
+	host1x_job_add_gather(job, &se->cmdbuf->bo, size, 0);
+
+	ret = host1x_job_pin(job, se->dev);
+	if (ret) {
+		dev_err(se->dev, "failed to pin host1x job\n");
+		goto job_put;
+	}
+
+	ret = host1x_job_submit(job);
+	if (ret) {
+		dev_err(se->dev, "failed to submit host1x job\n");
+		goto job_unpin;
+	}
+
+	ret = host1x_syncpt_wait(job->syncpt, job->syncpt_end,
+				 MAX_SCHEDULE_TIMEOUT, NULL);
+	if (ret) {
+		dev_err(se->dev, "host1x job timed out\n");
+		return ret;
+	}
+
+	host1x_job_put(job);
+	return 0;
+
+job_unpin:
+	host1x_job_unpin(job);
+job_put:
+	host1x_job_put(job);
+
+	return ret;
+}
+
+static int tegra_se_client_init(struct host1x_client *client)
+{
+	struct tegra_se *se = container_of(client, struct tegra_se, client);
+	int ret;
+
+	se->channel = host1x_channel_request(&se->client);
+	if (!se->channel) {
+		dev_err(se->dev, "host1x channel map failed\n");
+		return -ENODEV;
+	}
+
+	se->syncpt = host1x_syncpt_request(&se->client, 0);
+	if (!se->syncpt) {
+		dev_err(se->dev, "host1x syncpt allocation failed\n");
+		ret = -EINVAL;
+		goto channel_put;
+	}
+
+	se->syncpt_id =  host1x_syncpt_id(se->syncpt);
+
+	se->cmdbuf = tegra_se_host1x_bo_alloc(se, SZ_4K);
+	if (!se->cmdbuf) {
+		ret = -ENOMEM;
+		goto syncpt_put;
+	}
+
+	ret = se->hw->init_alg(se);
+	if (ret) {
+		dev_err(se->dev, "failed to register algorithms\n");
+		goto cmdbuf_put;
+	}
+
+	return 0;
+
+cmdbuf_put:
+	tegra_se_cmdbuf_put(&se->cmdbuf->bo);
+syncpt_put:
+	host1x_syncpt_put(se->syncpt);
+channel_put:
+	host1x_channel_put(se->channel);
+
+	return ret;
+}
+
+static int tegra_se_client_deinit(struct host1x_client *client)
+{
+	struct tegra_se *se = container_of(client, struct tegra_se, client);
+
+	se->hw->deinit_alg(se);
+	tegra_se_cmdbuf_put(&se->cmdbuf->bo);
+	host1x_syncpt_put(se->syncpt);
+	host1x_channel_put(se->channel);
+
+	return 0;
+}
+
+static const struct host1x_client_ops tegra_se_client_ops = {
+	.init = tegra_se_client_init,
+	.exit = tegra_se_client_deinit,
+};
+
+static int tegra_se_host1x_register(struct tegra_se *se)
+{
+	INIT_LIST_HEAD(&se->client.list);
+	se->client.dev = se->dev;
+	se->client.ops = &tegra_se_client_ops;
+	se->client.class = se->hw->host1x_class;
+	se->client.num_syncpts = 1;
+
+	host1x_client_register(&se->client);
+
+	return 0;
+}
+
+static int tegra_se_probe(struct platform_device *pdev)
+{
+	struct device *dev = &pdev->dev;
+	struct tegra_se *se;
+	int ret;
+
+	se = devm_kzalloc(dev, sizeof(*se), GFP_KERNEL);
+	if (!se)
+		return -ENOMEM;
+
+	se->dev = dev;
+	se->owner = TEGRA_GPSE_ID;
+	se->hw = device_get_match_data(&pdev->dev);
+
+	se->base = devm_platform_ioremap_resource(pdev, 0);
+	if (IS_ERR(se->base))
+		return PTR_ERR(se->base);
+
+	dma_set_mask_and_coherent(dev, DMA_BIT_MASK(39));
+	platform_set_drvdata(pdev, se);
+
+	se->clk = devm_clk_get_enabled(se->dev, NULL);
+	if (IS_ERR(se->clk))
+		return dev_err_probe(dev, PTR_ERR(se->clk),
+				"failed to enable clocks\n");
+
+	if (!tegra_dev_iommu_get_stream_id(dev, &se->stream_id))
+		return dev_err_probe(dev, -ENODEV,
+				"failed to get IOMMU stream ID\n");
+
+	writel(se->stream_id, se->base + SE_STREAM_ID);
+
+	se->engine = crypto_engine_alloc_init(dev, 0);
+	if (!se->engine)
+		return dev_err_probe(dev, -ENOMEM, "failed to init crypto engine\n");
+
+	ret = crypto_engine_start(se->engine);
+	if (ret) {
+		crypto_engine_exit(se->engine);
+		return dev_err_probe(dev, ret, "failed to start crypto engine\n");
+	}
+
+	ret = tegra_se_host1x_register(se);
+	if (ret) {
+		crypto_engine_stop(se->engine);
+		crypto_engine_exit(se->engine);
+		return dev_err_probe(dev, ret, "failed to init host1x params\n");
+	}
+
+	return 0;
+}
+
+static void tegra_se_remove(struct platform_device *pdev)
+{
+	struct tegra_se *se = platform_get_drvdata(pdev);
+
+	crypto_engine_stop(se->engine);
+	crypto_engine_exit(se->engine);
+	iommu_fwspec_free(se->dev);
+	host1x_client_unregister(&se->client);
+}
+
+static const struct tegra_se_regs tegra234_aes1_regs = {
+	.config = SE_AES1_CFG,
+	.op = SE_AES1_OPERATION,
+	.last_blk = SE_AES1_LAST_BLOCK,
+	.linear_ctr = SE_AES1_LINEAR_CTR,
+	.aad_len = SE_AES1_AAD_LEN,
+	.cryp_msg_len = SE_AES1_CRYPTO_MSG_LEN,
+	.manifest = SE_AES1_KEYMANIFEST,
+	.key_addr = SE_AES1_KEY_ADDR,
+	.key_data = SE_AES1_KEY_DATA,
+	.key_dst = SE_AES1_KEY_DST,
+	.result = SE_AES1_CMAC_RESULT,
+};
+
+static const struct tegra_se_regs tegra234_hash_regs = {
+	.config = SE_SHA_CFG,
+	.op = SE_SHA_OPERATION,
+	.manifest = SE_SHA_KEYMANIFEST,
+	.key_addr = SE_SHA_KEY_ADDR,
+	.key_data = SE_SHA_KEY_DATA,
+	.key_dst = SE_SHA_KEY_DST,
+	.result = SE_SHA_HASH_RESULT,
+};
+
+static const struct tegra_se_hw tegra234_aes_hw = {
+	.regs = &tegra234_aes1_regs,
+	.kac_ver = 1,
+	.host1x_class = 0x3b,
+	.init_alg = tegra_init_aes,
+	.deinit_alg = tegra_deinit_aes,
+};
+
+static const struct tegra_se_hw tegra234_hash_hw = {
+	.regs = &tegra234_hash_regs,
+	.kac_ver = 1,
+	.host1x_class = 0x3d,
+	.init_alg = tegra_init_hash,
+	.deinit_alg = tegra_deinit_hash,
+};
+
+static const struct of_device_id tegra_se_of_match[] = {
+	{
+		.compatible = "nvidia,tegra234-se-aes",
+		.data = &tegra234_aes_hw
+	}, {
+		.compatible = "nvidia,tegra234-se-hash",
+		.data = &tegra234_hash_hw,
+	},
+	{ },
+};
+MODULE_DEVICE_TABLE(of, tegra_se_of_match);
+
+static struct platform_driver tegra_se_driver = {
+	.driver = {
+		.name	= "tegra-se",
+		.of_match_table = tegra_se_of_match,
+	},
+	.probe		= tegra_se_probe,
+	.remove_new	= tegra_se_remove,
+};
+
+static int tegra_se_host1x_probe(struct host1x_device *dev)
+{
+	return host1x_device_init(dev);
+}
+
+static int tegra_se_host1x_remove(struct host1x_device *dev)
+{
+	host1x_device_exit(dev);
+
+	return 0;
+}
+
+static struct host1x_driver tegra_se_host1x_driver = {
+	.driver = {
+		.name = "tegra-se-host1x",
+	},
+	.probe = tegra_se_host1x_probe,
+	.remove = tegra_se_host1x_remove,
+	.subdevs = tegra_se_of_match,
+};
+
+static int __init tegra_se_module_init(void)
+{
+	int ret;
+
+	ret = host1x_driver_register(&tegra_se_host1x_driver);
+	if (ret)
+		return ret;
+
+	return platform_driver_register(&tegra_se_driver);
+}
+
+static void __exit tegra_se_module_exit(void)
+{
+	host1x_driver_unregister(&tegra_se_host1x_driver);
+	platform_driver_unregister(&tegra_se_driver);
+}
+
+module_init(tegra_se_module_init);
+module_exit(tegra_se_module_exit);
+
+MODULE_DESCRIPTION("NVIDIA Tegra Security Engine Driver");
+MODULE_AUTHOR("Akhil R <akhilrajeev@nvidia.com>");
+MODULE_LICENSE("GPL");
diff --git a/drivers/crypto/tegra/tegra-se.h b/drivers/crypto/tegra/tegra-se.h
new file mode 100644
index 000000000000..b9dd7ceb8783
--- /dev/null
+++ b/drivers/crypto/tegra/tegra-se.h
@@ -0,0 +1,560 @@
+/* SPDX-License-Identifier: GPL-2.0-only
+ * SPDX-FileCopyrightText: Copyright (c) 2023 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
+ *
+ * Header file for NVIDIA Security Engine driver.
+ */
+
+#ifndef _TEGRA_SE_H
+#define _TEGRA_SE_H
+
+#include <linux/bitfield.h>
+#include <linux/iommu.h>
+#include <linux/host1x.h>
+#include <crypto/aead.h>
+#include <crypto/engine.h>
+#include <crypto/hash.h>
+#include <crypto/sha1.h>
+#include <crypto/sha3.h>
+#include <crypto/skcipher.h>
+
+#define SE_OWNERSHIP					0x14
+#define SE_OWNERSHIP_UID(x)				FIELD_GET(GENMASK(7, 0), x)
+#define TEGRA_GPSE_ID					3
+
+#define SE_STREAM_ID					0x90
+
+#define SE_SHA_CFG					0x4004
+#define SE_SHA_KEY_ADDR					0x4094
+#define SE_SHA_KEY_DATA					0x4098
+#define SE_SHA_KEYMANIFEST				0x409c
+#define SE_SHA_CRYPTO_CFG				0x40a4
+#define SE_SHA_KEY_DST					0x40a8
+#define SE_SHA_SRC_KSLT					0x4180
+#define SE_SHA_TGT_KSLT					0x4184
+#define SE_SHA_MSG_LENGTH				0x401c
+#define SE_SHA_OPERATION				0x407c
+#define SE_SHA_HASH_RESULT				0x40b0
+
+#define SE_SHA_ENC_MODE(x)				FIELD_PREP(GENMASK(31, 24), x)
+#define SE_SHA_ENC_MODE_SHA1				SE_SHA_ENC_MODE(0)
+#define SE_SHA_ENC_MODE_SHA224				SE_SHA_ENC_MODE(4)
+#define SE_SHA_ENC_MODE_SHA256				SE_SHA_ENC_MODE(5)
+#define SE_SHA_ENC_MODE_SHA384				SE_SHA_ENC_MODE(6)
+#define SE_SHA_ENC_MODE_SHA512				SE_SHA_ENC_MODE(7)
+#define SE_SHA_ENC_MODE_SHA_CTX_INTEGRITY		SE_SHA_ENC_MODE(8)
+#define SE_SHA_ENC_MODE_SHA3_224			SE_SHA_ENC_MODE(9)
+#define SE_SHA_ENC_MODE_SHA3_256			SE_SHA_ENC_MODE(10)
+#define SE_SHA_ENC_MODE_SHA3_384			SE_SHA_ENC_MODE(11)
+#define SE_SHA_ENC_MODE_SHA3_512			SE_SHA_ENC_MODE(12)
+#define SE_SHA_ENC_MODE_SHAKE128			SE_SHA_ENC_MODE(13)
+#define SE_SHA_ENC_MODE_SHAKE256			SE_SHA_ENC_MODE(14)
+#define SE_SHA_ENC_MODE_HMAC_SHA256_1KEY		SE_SHA_ENC_MODE(0)
+#define SE_SHA_ENC_MODE_HMAC_SHA256_2KEY		SE_SHA_ENC_MODE(1)
+#define SE_SHA_ENC_MODE_SM3_256				SE_SHA_ENC_MODE(0)
+
+#define SE_SHA_CFG_ENC_ALG(x)				FIELD_PREP(GENMASK(15, 12), x)
+#define SE_SHA_ENC_ALG_NOP				SE_SHA_CFG_ENC_ALG(0)
+#define SE_SHA_ENC_ALG_SHA_ENC				SE_SHA_CFG_ENC_ALG(1)
+#define SE_SHA_ENC_ALG_RNG				SE_SHA_CFG_ENC_ALG(2)
+#define SE_SHA_ENC_ALG_SHA				SE_SHA_CFG_ENC_ALG(3)
+#define SE_SHA_ENC_ALG_SM3				SE_SHA_CFG_ENC_ALG(4)
+#define SE_SHA_ENC_ALG_HMAC				SE_SHA_CFG_ENC_ALG(7)
+#define SE_SHA_ENC_ALG_KDF				SE_SHA_CFG_ENC_ALG(8)
+#define SE_SHA_ENC_ALG_KEY_INVLD			SE_SHA_CFG_ENC_ALG(10)
+#define SE_SHA_ENC_ALG_KEY_INQUIRE			SE_SHA_CFG_ENC_ALG(12)
+#define SE_SHA_ENC_ALG_INS				SE_SHA_CFG_ENC_ALG(13)
+
+#define SE_SHA_OP_LASTBUF				FIELD_PREP(BIT(16), 1)
+#define SE_SHA_OP_WRSTALL				FIELD_PREP(BIT(15), 1)
+
+#define SE_SHA_OP_OP(x)					FIELD_PREP(GENMASK(2, 0), x)
+#define SE_SHA_OP_START					SE_SHA_OP_OP(1)
+#define SE_SHA_OP_RESTART_OUT				SE_SHA_OP_OP(2)
+#define SE_SHA_OP_RESTART_IN				SE_SHA_OP_OP(4)
+#define SE_SHA_OP_RESTART_INOUT				SE_SHA_OP_OP(5)
+#define SE_SHA_OP_DUMMY					SE_SHA_OP_OP(6)
+
+#define SE_SHA_CFG_DEC_ALG(x)				FIELD_PREP(GENMASK(11, 8), x)
+#define SE_SHA_DEC_ALG_NOP				SE_SHA_CFG_DEC_ALG(0)
+#define SE_SHA_DEC_ALG_AES_DEC				SE_SHA_CFG_DEC_ALG(1)
+#define SE_SHA_DEC_ALG_HMAC				SE_SHA_CFG_DEC_ALG(7)
+#define SE_SHA_DEC_ALG_HMAC_VERIFY			SE_SHA_CFG_DEC_ALG(9)
+
+#define SE_SHA_CFG_DST(x)				FIELD_PREP(GENMASK(4, 2), x)
+#define SE_SHA_DST_MEMORY				SE_SHA_CFG_DST(0)
+#define SE_SHA_DST_HASH_REG				SE_SHA_CFG_DST(1)
+#define SE_SHA_DST_KEYTABLE				SE_SHA_CFG_DST(2)
+#define SE_SHA_DST_SRK					SE_SHA_CFG_DST(3)
+
+#define SE_SHA_TASK_HASH_INIT				BIT(0)
+
+/* AES Configuration */
+#define SE_AES0_CFG					0x1004
+#define SE_AES0_CRYPTO_CONFIG				0x1008
+#define SE_AES0_KEY_DST					0x1030
+#define SE_AES0_OPERATION				0x1038
+#define SE_AES0_LINEAR_CTR				0x101c
+#define SE_AES0_LAST_BLOCK				0x102c
+#define SE_AES0_KEY_ADDR				0x10bc
+#define SE_AES0_KEY_DATA				0x10c0
+#define SE_AES0_CMAC_RESULT				0x10c4
+#define SE_AES0_SRC_KSLT				0x1100
+#define SE_AES0_TGT_KSLT				0x1104
+#define SE_AES0_KEYMANIFEST				0x1114
+#define SE_AES0_AAD_LEN					0x112c
+#define SE_AES0_CRYPTO_MSG_LEN				0x1134
+
+#define SE_AES1_CFG					0x2004
+#define SE_AES1_CRYPTO_CONFIG				0x2008
+#define SE_AES1_KEY_DST					0x2030
+#define SE_AES1_OPERATION				0x2038
+#define SE_AES1_LINEAR_CTR				0x201c
+#define SE_AES1_LAST_BLOCK				0x202c
+#define SE_AES1_KEY_ADDR				0x20bc
+#define SE_AES1_KEY_DATA				0x20c0
+#define SE_AES1_CMAC_RESULT				0x20c4
+#define SE_AES1_SRC_KSLT				0x2100
+#define SE_AES1_TGT_KSLT				0x2104
+#define SE_AES1_KEYMANIFEST				0x2114
+#define SE_AES1_AAD_LEN					0x212c
+#define SE_AES1_CRYPTO_MSG_LEN				0x2134
+
+#define SE_AES_CFG_ENC_MODE(x)				FIELD_PREP(GENMASK(31, 24), x)
+#define SE_AES_ENC_MODE_GMAC				SE_AES_CFG_ENC_MODE(3)
+#define SE_AES_ENC_MODE_GCM				SE_AES_CFG_ENC_MODE(4)
+#define SE_AES_ENC_MODE_GCM_FINAL			SE_AES_CFG_ENC_MODE(5)
+#define SE_AES_ENC_MODE_CMAC				SE_AES_CFG_ENC_MODE(7)
+#define SE_AES_ENC_MODE_CBC_MAC				SE_AES_CFG_ENC_MODE(12)
+
+#define SE_AES_CFG_DEC_MODE(x)				FIELD_PREP(GENMASK(23, 16), x)
+#define SE_AES_DEC_MODE_GMAC				SE_AES_CFG_DEC_MODE(3)
+#define SE_AES_DEC_MODE_GCM				SE_AES_CFG_DEC_MODE(4)
+#define SE_AES_DEC_MODE_GCM_FINAL			SE_AES_CFG_DEC_MODE(5)
+#define SE_AES_DEC_MODE_CBC_MAC				SE_AES_CFG_DEC_MODE(12)
+
+#define SE_AES_CFG_ENC_ALG(x)				FIELD_PREP(GENMASK(15, 12), x)
+#define SE_AES_ENC_ALG_NOP				SE_AES_CFG_ENC_ALG(0)
+#define SE_AES_ENC_ALG_AES_ENC				SE_AES_CFG_ENC_ALG(1)
+#define SE_AES_ENC_ALG_RNG				SE_AES_CFG_ENC_ALG(2)
+#define SE_AES_ENC_ALG_SHA				SE_AES_CFG_ENC_ALG(3)
+#define SE_AES_ENC_ALG_HMAC				SE_AES_CFG_ENC_ALG(7)
+#define SE_AES_ENC_ALG_KDF				SE_AES_CFG_ENC_ALG(8)
+#define SE_AES_ENC_ALG_INS				SE_AES_CFG_ENC_ALG(13)
+
+#define SE_AES_CFG_DEC_ALG(x)				FIELD_PREP(GENMASK(11, 8), x)
+#define SE_AES_DEC_ALG_NOP				SE_AES_CFG_DEC_ALG(0)
+#define SE_AES_DEC_ALG_AES_DEC				SE_AES_CFG_DEC_ALG(1)
+
+#define SE_AES_CFG_DST(x)				FIELD_PREP(GENMASK(4, 2), x)
+#define SE_AES_DST_MEMORY				SE_AES_CFG_DST(0)
+#define SE_AES_DST_HASH_REG				SE_AES_CFG_DST(1)
+#define SE_AES_DST_KEYTABLE				SE_AES_CFG_DST(2)
+#define SE_AES_DST_SRK					SE_AES_CFG_DST(3)
+
+/* AES Crypto Configuration */
+#define SE_AES_KEY2_INDEX(x)				FIELD_PREP(GENMASK(31, 28), x)
+#define SE_AES_KEY_INDEX(x)				FIELD_PREP(GENMASK(27, 24), x)
+
+#define SE_AES_CRYPTO_CFG_SCC_DIS			FIELD_PREP(BIT(20), 1)
+
+#define SE_AES_CRYPTO_CFG_CTR_CNTN(x)			FIELD_PREP(GENMASK(18, 11), x)
+
+#define SE_AES_CRYPTO_CFG_IV_MODE(x)			FIELD_PREP(BIT(10), x)
+#define SE_AES_IV_MODE_SWIV				SE_AES_CRYPTO_CFG_IV_MODE(0)
+#define SE_AES_IV_MODE_HWIV				SE_AES_CRYPTO_CFG_IV_MODE(1)
+
+#define SE_AES_CRYPTO_CFG_CORE_SEL(x)			FIELD_PREP(BIT(9), x)
+#define SE_AES_CORE_SEL_DECRYPT				SE_AES_CRYPTO_CFG_CORE_SEL(0)
+#define SE_AES_CORE_SEL_ENCRYPT				SE_AES_CRYPTO_CFG_CORE_SEL(1)
+
+#define SE_AES_CRYPTO_CFG_IV_SEL(x)			FIELD_PREP(GENMASK(8, 7), x)
+#define SE_AES_IV_SEL_UPDATED				SE_AES_CRYPTO_CFG_IV_SEL(1)
+#define SE_AES_IV_SEL_REG				SE_AES_CRYPTO_CFG_IV_SEL(2)
+#define SE_AES_IV_SEL_RANDOM				SE_AES_CRYPTO_CFG_IV_SEL(3)
+
+#define SE_AES_CRYPTO_CFG_VCTRAM_SEL(x)			FIELD_PREP(GENMASK(6, 5), x)
+#define SE_AES_VCTRAM_SEL_MEMORY			SE_AES_CRYPTO_CFG_VCTRAM_SEL(0)
+#define SE_AES_VCTRAM_SEL_TWEAK				SE_AES_CRYPTO_CFG_VCTRAM_SEL(1)
+#define SE_AES_VCTRAM_SEL_AESOUT			SE_AES_CRYPTO_CFG_VCTRAM_SEL(2)
+#define SE_AES_VCTRAM_SEL_PREV_MEM			SE_AES_CRYPTO_CFG_VCTRAM_SEL(3)
+
+#define SE_AES_CRYPTO_CFG_INPUT_SEL(x)			FIELD_PREP(GENMASK(4, 3), x)
+#define SE_AES_INPUT_SEL_MEMORY				SE_AES_CRYPTO_CFG_INPUT_SEL(0)
+#define SE_AES_INPUT_SEL_RANDOM				SE_AES_CRYPTO_CFG_INPUT_SEL(1)
+#define SE_AES_INPUT_SEL_AESOUT				SE_AES_CRYPTO_CFG_INPUT_SEL(2)
+#define SE_AES_INPUT_SEL_LINEAR_CTR			SE_AES_CRYPTO_CFG_INPUT_SEL(3)
+#define SE_AES_INPUT_SEL_REG				SE_AES_CRYPTO_CFG_INPUT_SEL(1)
+
+#define SE_AES_CRYPTO_CFG_XOR_POS(x)			FIELD_PREP(GENMASK(2, 1), x)
+#define SE_AES_XOR_POS_BYPASS				SE_AES_CRYPTO_CFG_XOR_POS(0)
+#define SE_AES_XOR_POS_BOTH				SE_AES_CRYPTO_CFG_XOR_POS(1)
+#define SE_AES_XOR_POS_TOP				SE_AES_CRYPTO_CFG_XOR_POS(2)
+#define SE_AES_XOR_POS_BOTTOM				SE_AES_CRYPTO_CFG_XOR_POS(3)
+
+#define SE_AES_CRYPTO_CFG_HASH_EN(x)			FIELD_PREP(BIT(0), x)
+#define SE_AES_HASH_DISABLE				SE_AES_CRYPTO_CFG_HASH_EN(0)
+#define SE_AES_HASH_ENABLE				SE_AES_CRYPTO_CFG_HASH_EN(1)
+
+#define SE_LAST_BLOCK_VAL(x)				FIELD_PREP(GENMASK(19, 0), x)
+#define SE_LAST_BLOCK_RES_BITS(x)			FIELD_PREP(GENMASK(26, 20), x)
+
+#define SE_AES_OP_LASTBUF				FIELD_PREP(BIT(16), 1)
+#define SE_AES_OP_WRSTALL				FIELD_PREP(BIT(15), 1)
+#define SE_AES_OP_FINAL					FIELD_PREP(BIT(5), 1)
+#define SE_AES_OP_INIT					FIELD_PREP(BIT(4), 1)
+
+#define SE_AES_OP_OP(x)					FIELD_PREP(GENMASK(2, 0), x)
+#define SE_AES_OP_START					SE_AES_OP_OP(1)
+#define SE_AES_OP_RESTART_OUT				SE_AES_OP_OP(2)
+#define SE_AES_OP_RESTART_IN				SE_AES_OP_OP(4)
+#define SE_AES_OP_RESTART_INOUT				SE_AES_OP_OP(5)
+#define SE_AES_OP_DUMMY					SE_AES_OP_OP(6)
+
+#define SE_KAC_SIZE(x)					FIELD_PREP(GENMASK(15, 14), x)
+#define SE_KAC_SIZE_128					SE_KAC_SIZE(0)
+#define SE_KAC_SIZE_192					SE_KAC_SIZE(1)
+#define SE_KAC_SIZE_256					SE_KAC_SIZE(2)
+
+#define SE_KAC_EXPORTABLE				FIELD_PREP(BIT(12), 1)
+
+#define SE_KAC_PURPOSE(x)				FIELD_PREP(GENMASK(11, 8), x)
+#define SE_KAC_ENC					SE_KAC_PURPOSE(0)
+#define SE_KAC_CMAC					SE_KAC_PURPOSE(1)
+#define SE_KAC_HMAC					SE_KAC_PURPOSE(2)
+#define SE_KAC_GCM_KW					SE_KAC_PURPOSE(3)
+#define SE_KAC_HMAC_KDK					SE_KAC_PURPOSE(6)
+#define SE_KAC_HMAC_KDD					SE_KAC_PURPOSE(7)
+#define SE_KAC_HMAC_KDD_KUW				SE_KAC_PURPOSE(8)
+#define SE_KAC_XTS					SE_KAC_PURPOSE(9)
+#define SE_KAC_GCM					SE_KAC_PURPOSE(10)
+
+#define SE_KAC_USER_NS					FIELD_PREP(GENMASK(6, 4), 3)
+
+#define SE_AES_KEY_DST_INDEX(x)				FIELD_PREP(GENMASK(11, 8), x)
+#define SE_ADDR_HI_MSB(x)				FIELD_PREP(GENMASK(31, 24), x)
+#define SE_ADDR_HI_SZ(x)				FIELD_PREP(GENMASK(23, 0), x)
+
+#define SE_CFG_AES_ENCRYPT				(SE_AES_ENC_ALG_AES_ENC | \
+							 SE_AES_DEC_ALG_NOP | \
+							 SE_AES_DST_MEMORY)
+
+#define SE_CFG_AES_DECRYPT				(SE_AES_ENC_ALG_NOP | \
+							 SE_AES_DEC_ALG_AES_DEC | \
+							 SE_AES_DST_MEMORY)
+
+#define SE_CFG_GMAC_ENCRYPT				(SE_AES_ENC_ALG_AES_ENC | \
+							 SE_AES_DEC_ALG_NOP | \
+							 SE_AES_ENC_MODE_GMAC | \
+							 SE_AES_DST_MEMORY)
+
+#define SE_CFG_GMAC_DECRYPT				(SE_AES_ENC_ALG_NOP | \
+							 SE_AES_DEC_ALG_AES_DEC | \
+							 SE_AES_DEC_MODE_GMAC | \
+							 SE_AES_DST_MEMORY)
+
+#define SE_CFG_GCM_ENCRYPT				(SE_AES_ENC_ALG_AES_ENC | \
+							 SE_AES_DEC_ALG_NOP | \
+							 SE_AES_ENC_MODE_GCM | \
+							 SE_AES_DST_MEMORY)
+
+#define SE_CFG_GCM_DECRYPT				(SE_AES_ENC_ALG_NOP | \
+							 SE_AES_DEC_ALG_AES_DEC | \
+							 SE_AES_DEC_MODE_GCM | \
+							 SE_AES_DST_MEMORY)
+
+#define SE_CFG_GCM_FINAL_ENCRYPT			(SE_AES_ENC_ALG_AES_ENC | \
+							 SE_AES_DEC_ALG_NOP | \
+							 SE_AES_ENC_MODE_GCM_FINAL | \
+							 SE_AES_DST_MEMORY)
+
+#define SE_CFG_GCM_FINAL_DECRYPT			(SE_AES_ENC_ALG_NOP | \
+							 SE_AES_DEC_ALG_AES_DEC | \
+							 SE_AES_DEC_MODE_GCM_FINAL | \
+							 SE_AES_DST_MEMORY)
+
+#define SE_CFG_CMAC					(SE_AES_ENC_ALG_AES_ENC | \
+							 SE_AES_ENC_MODE_CMAC | \
+							 SE_AES_DST_HASH_REG)
+
+#define SE_CFG_CBC_MAC					(SE_AES_ENC_ALG_AES_ENC | \
+							 SE_AES_ENC_MODE_CBC_MAC)
+
+#define SE_CFG_INS					(SE_AES_ENC_ALG_INS | \
+							 SE_AES_DEC_ALG_NOP)
+
+#define SE_CRYPTO_CFG_ECB_ENCRYPT			(SE_AES_INPUT_SEL_MEMORY | \
+							 SE_AES_XOR_POS_BYPASS | \
+							 SE_AES_CORE_SEL_ENCRYPT)
+
+#define SE_CRYPTO_CFG_ECB_DECRYPT			(SE_AES_INPUT_SEL_MEMORY | \
+							 SE_AES_XOR_POS_BYPASS | \
+							 SE_AES_CORE_SEL_DECRYPT)
+
+#define SE_CRYPTO_CFG_CBC_ENCRYPT			(SE_AES_INPUT_SEL_MEMORY | \
+							 SE_AES_VCTRAM_SEL_AESOUT | \
+							 SE_AES_XOR_POS_TOP | \
+							 SE_AES_CORE_SEL_ENCRYPT | \
+							 SE_AES_IV_SEL_REG)
+
+#define SE_CRYPTO_CFG_CBC_DECRYPT			(SE_AES_INPUT_SEL_MEMORY | \
+							 SE_AES_VCTRAM_SEL_PREV_MEM | \
+							 SE_AES_XOR_POS_BOTTOM | \
+							 SE_AES_CORE_SEL_DECRYPT | \
+							 SE_AES_IV_SEL_REG)
+
+#define SE_CRYPTO_CFG_CTR				(SE_AES_INPUT_SEL_LINEAR_CTR | \
+							 SE_AES_VCTRAM_SEL_MEMORY | \
+							 SE_AES_XOR_POS_BOTTOM | \
+							 SE_AES_CORE_SEL_ENCRYPT | \
+							 SE_AES_CRYPTO_CFG_CTR_CNTN(1) | \
+							 SE_AES_IV_SEL_REG)
+
+#define SE_CRYPTO_CFG_XTS_ENCRYPT			(SE_AES_INPUT_SEL_MEMORY | \
+							 SE_AES_VCTRAM_SEL_TWEAK | \
+							 SE_AES_XOR_POS_BOTH | \
+							 SE_AES_CORE_SEL_ENCRYPT | \
+							 SE_AES_IV_SEL_REG)
+
+#define SE_CRYPTO_CFG_XTS_DECRYPT			(SE_AES_INPUT_SEL_MEMORY | \
+							 SE_AES_VCTRAM_SEL_TWEAK | \
+							 SE_AES_XOR_POS_BOTH | \
+							 SE_AES_CORE_SEL_DECRYPT | \
+							 SE_AES_IV_SEL_REG)
+
+#define SE_CRYPTO_CFG_XTS_DECRYPT			(SE_AES_INPUT_SEL_MEMORY | \
+							 SE_AES_VCTRAM_SEL_TWEAK | \
+							 SE_AES_XOR_POS_BOTH | \
+							 SE_AES_CORE_SEL_DECRYPT | \
+							 SE_AES_IV_SEL_REG)
+
+#define SE_CRYPTO_CFG_CBC_MAC				(SE_AES_INPUT_SEL_MEMORY | \
+							 SE_AES_VCTRAM_SEL_AESOUT | \
+							 SE_AES_XOR_POS_TOP | \
+							 SE_AES_CORE_SEL_ENCRYPT | \
+							 SE_AES_HASH_ENABLE | \
+							 SE_AES_IV_SEL_REG)
+
+#define HASH_RESULT_REG_COUNT				50
+#define CMAC_RESULT_REG_COUNT				4
+
+#define SE_CRYPTO_CTR_REG_COUNT			4
+#define SE_MAX_KEYSLOT				15
+#define SE_MAX_MEM_ALLOC			SZ_4M
+#define SE_AES_BUFLEN				0x8000
+#define SE_SHA_BUFLEN				0x2000
+
+#define SHA_FIRST	BIT(0)
+#define SHA_UPDATE	BIT(1)
+#define SHA_FINAL	BIT(2)
+
+/* Security Engine operation modes */
+enum se_aes_alg {
+	SE_ALG_CBC,		/* Cipher Block Chaining (CBC) mode */
+	SE_ALG_ECB,		/* Electronic Codebook (ECB) mode */
+	SE_ALG_CTR,		/* Counter (CTR) mode */
+	SE_ALG_XTS,		/* XTS mode */
+	SE_ALG_GMAC,		/* GMAC mode */
+	SE_ALG_GCM,		/* GCM mode */
+	SE_ALG_GCM_FINAL,	/* GCM FINAL mode */
+	SE_ALG_CMAC,	/* Cipher-based MAC (CMAC) mode */
+	SE_ALG_CBC_MAC,	/* CBC MAC mode */
+};
+
+enum se_hash_alg {
+	SE_ALG_RNG_DRBG,	/* Deterministic Random Bit Generator */
+	SE_ALG_SHA1,		/* Secure Hash Algorithm-1 (SHA1) mode */
+	SE_ALG_SHA224,		/* Secure Hash Algorithm-224  (SHA224) mode */
+	SE_ALG_SHA256,		/* Secure Hash Algorithm-256  (SHA256) mode */
+	SE_ALG_SHA384,		/* Secure Hash Algorithm-384  (SHA384) mode */
+	SE_ALG_SHA512,		/* Secure Hash Algorithm-512  (SHA512) mode */
+	SE_ALG_SHA3_224,	/* Secure Hash Algorithm3-224 (SHA3-224) mode */
+	SE_ALG_SHA3_256,	/* Secure Hash Algorithm3-256 (SHA3-256) mode */
+	SE_ALG_SHA3_384,	/* Secure Hash Algorithm3-384 (SHA3-384) mode */
+	SE_ALG_SHA3_512,	/* Secure Hash Algorithm3-512 (SHA3-512) mode */
+	SE_ALG_SHAKE128,	/* Secure Hash Algorithm3 (SHAKE128) mode */
+	SE_ALG_SHAKE256,	/* Secure Hash Algorithm3 (SHAKE256) mode */
+	SE_ALG_HMAC_SHA224,	/* Hash based MAC (HMAC) - 224 */
+	SE_ALG_HMAC_SHA256,	/* Hash based MAC (HMAC) - 256 */
+	SE_ALG_HMAC_SHA384,	/* Hash based MAC (HMAC) - 384 */
+	SE_ALG_HMAC_SHA512,	/* Hash based MAC (HMAC) - 512 */
+};
+
+struct tegra_se_alg {
+	struct tegra_se *se_dev;
+	const char *alg_base;
+
+	union {
+		struct skcipher_engine_alg skcipher;
+		struct aead_engine_alg aead;
+		struct ahash_engine_alg ahash;
+	} alg;
+};
+
+struct tegra_se_regs {
+	u32 op;
+	u32 config;
+	u32 last_blk;
+	u32 linear_ctr;
+	u32 out_addr;
+	u32 aad_len;
+	u32 cryp_msg_len;
+	u32 manifest;
+	u32 key_addr;
+	u32 key_data;
+	u32 key_dst;
+	u32 result;
+};
+
+struct tegra_se_hw {
+	const struct tegra_se_regs *regs;
+	int (*init_alg)(struct tegra_se *se);
+	void (*deinit_alg)(struct tegra_se *se);
+	bool support_sm_alg;
+	u32 host1x_class;
+	u32 kac_ver;
+};
+
+struct tegra_se {
+	int (*manifest)(u32 user, u32 alg, u32 keylen);
+	const struct tegra_se_hw *hw;
+	struct host1x_client client;
+	struct host1x_channel *channel;
+	struct tegra_se_cmdbuf *cmdbuf;
+	struct crypto_engine *engine;
+	struct host1x_syncpt *syncpt;
+	struct device *dev;
+	struct clk *clk;
+	unsigned int opcode_addr;
+	unsigned int stream_id;
+	unsigned int syncpt_id;
+	void __iomem *base;
+	u32 owner;
+};
+
+struct tegra_se_cmdbuf {
+	dma_addr_t iova;
+	u32 *addr;
+	struct device *dev;
+	struct kref ref;
+	struct host1x_bo bo;
+	ssize_t size;
+	u32 words;
+};
+
+struct tegra_se_datbuf {
+	u8 *buf;
+	dma_addr_t addr;
+	ssize_t size;
+};
+
+static inline int se_algname_to_algid(const char *name)
+{
+	if (!strcmp(name, "cbc(aes)"))
+		return SE_ALG_CBC;
+	else if (!strcmp(name, "ecb(aes)"))
+		return SE_ALG_ECB;
+	else if (!strcmp(name, "ctr(aes)"))
+		return SE_ALG_CTR;
+	else if (!strcmp(name, "xts(aes)"))
+		return SE_ALG_XTS;
+	else if (!strcmp(name, "cmac(aes)"))
+		return SE_ALG_CMAC;
+	else if (!strcmp(name, "gcm(aes)"))
+		return SE_ALG_GCM;
+	else if (!strcmp(name, "ccm(aes)"))
+		return SE_ALG_CBC_MAC;
+
+	else if (!strcmp(name, "sha1"))
+		return SE_ALG_SHA1;
+	else if (!strcmp(name, "sha224"))
+		return SE_ALG_SHA224;
+	else if (!strcmp(name, "sha256"))
+		return SE_ALG_SHA256;
+	else if (!strcmp(name, "sha384"))
+		return SE_ALG_SHA384;
+	else if (!strcmp(name, "sha512"))
+		return SE_ALG_SHA512;
+	else if (!strcmp(name, "sha3-224"))
+		return SE_ALG_SHA3_224;
+	else if (!strcmp(name, "sha3-256"))
+		return SE_ALG_SHA3_256;
+	else if (!strcmp(name, "sha3-384"))
+		return SE_ALG_SHA3_384;
+	else if (!strcmp(name, "sha3-512"))
+		return SE_ALG_SHA3_512;
+	else if (!strcmp(name, "hmac(sha224)"))
+		return SE_ALG_HMAC_SHA224;
+	else if (!strcmp(name, "hmac(sha256)"))
+		return SE_ALG_HMAC_SHA256;
+	else if (!strcmp(name, "hmac(sha384)"))
+		return SE_ALG_HMAC_SHA384;
+	else if (!strcmp(name, "hmac(sha512)"))
+		return SE_ALG_HMAC_SHA512;
+	else
+		return -EINVAL;
+}
+
+/* Functions */
+int tegra_init_aes(struct tegra_se *se);
+int tegra_init_hash(struct tegra_se *se);
+void tegra_deinit_aes(struct tegra_se *se);
+void tegra_deinit_hash(struct tegra_se *se);
+int tegra_key_submit(struct tegra_se *se, const u8 *key,
+		     u32 keylen, u32 alg, u32 *keyid);
+void tegra_key_invalidate(struct tegra_se *se, u32 keyid, u32 alg);
+int tegra_se_host1x_submit(struct tegra_se *se, u32 size);
+
+/* HOST1x OPCODES */
+static inline u32 host1x_opcode_setpayload(unsigned int payload)
+{
+	return (9 << 28) | payload;
+}
+
+static inline u32 host1x_opcode_incr_w(unsigned int offset)
+{
+	/* 22-bit offset supported */
+	return (10 << 28) | offset;
+}
+
+static inline u32 host1x_opcode_nonincr_w(unsigned int offset)
+{
+	/* 22-bit offset supported */
+	return (11 << 28) | offset;
+}
+
+static inline u32 host1x_opcode_incr(unsigned int offset, unsigned int count)
+{
+	return (1 << 28) | (offset << 16) | count;
+}
+
+static inline u32 host1x_opcode_nonincr(unsigned int offset, unsigned int count)
+{
+	return (2 << 28) | (offset << 16) | count;
+}
+
+static inline u32 host1x_uclass_incr_syncpt_cond_f(u32 v)
+{
+	return (v & 0xff) << 10;
+}
+
+static inline u32 host1x_uclass_incr_syncpt_indx_f(u32 v)
+{
+	return (v & 0x3ff) << 0;
+}
+
+static inline u32 host1x_uclass_wait_syncpt_r(void)
+{
+	return 0x8;
+}
+
+static inline u32 host1x_uclass_incr_syncpt_r(void)
+{
+	return 0x0;
+}
+
+#define se_host1x_opcode_incr_w(x) host1x_opcode_incr_w((x) / 4)
+#define se_host1x_opcode_nonincr_w(x) host1x_opcode_nonincr_w((x) / 4)
+#define se_host1x_opcode_incr(x, y) host1x_opcode_incr((x) / 4, y)
+#define se_host1x_opcode_nonincr(x, y) host1x_opcode_nonincr((x) / 4, y)
+
+#endif /*_TEGRA_SE_H*/