Merge tag 'v6.18-p1' of git://git.kernel.org/pub/scm/linux/kernel/git/herbert/crypto-2.6

Pull crypto updates from Herbert Xu:
"Drivers:
- Add ciphertext hiding support to ccp
- Add hashjoin, gather and UDMA data move features to hisilicon
- Add lz4 and lz77_only to hisilicon
- Add xilinx hwrng driver
- Add ti driver with ecb/cbc aes support
- Add ring buffer idle and command queue telemetry for GEN6 in qat

Others:
- Use rcu_dereference_all to stop false alarms in rhashtable
- Fix CPU number wraparound in padata"

* tag 'v6.18-p1' of git://git.kernel.org/pub/scm/linux/kernel/git/herbert/crypto-2.6: (78 commits)
dt-bindings: rng: hisi-rng: convert to DT schema
crypto: doc - Add explicit title heading to API docs
hwrng: ks-sa - fix division by zero in ks_sa_rng_init
KEYS: X.509: Fix Basic Constraints CA flag parsing
crypto: anubis - simplify return statement in anubis_mod_init
crypto: hisilicon/qm - set NULL to qm->debug.qm_diff_regs
crypto: hisilicon/qm - clear all VF configurations in the hardware
crypto: hisilicon - enable error reporting again
crypto: hisilicon/qm - mask axi error before memory init
crypto: hisilicon/qm - invalidate queues in use
crypto: qat - Return pointer directly in adf_ctl_alloc_resources
crypto: aspeed - Fix dma_unmap_sg() direction
rhashtable: Use rcu_dereference_all and rcu_dereference_all_check
crypto: comp - Use same definition of context alloc and free ops
crypto: omap - convert from tasklet to BH workqueue
crypto: qat - Replace kzalloc() + copy_from_user() with memdup_user()
crypto: caam - double the entropy delay interval for retry
padata: WQ_PERCPU added to alloc_workqueue users
padata: replace use of system_unbound_wq with system_dfl_wq
crypto: cryptd - WQ_PERCPU added to alloc_workqueue users
...

Documentation/ABI/testing/debugfs-driver-qat_telemetry

···
 		gp_lat_acc_avg		average get to put latency [ns]
 		bw_in			PCIe, write bandwidth [Mbps]
 		bw_out			PCIe, read bandwidth [Mbps]
+		re_acc_avg		average ring empty time [ns]
 		at_page_req_lat_avg	Address Translator(AT), average page
 					request latency [ns]
 		at_trans_lat_avg	AT, average page translation latency [ns]
···
 		exec_cph<N>		execution count of Cipher slice N
 		util_ath<N>		utilization of Authentication slice N [%]
 		exec_ath<N>		execution count of Authentication slice N
+		cmdq_wait_cnv<N>	wait time for cmdq N to get Compression and verify
+					slice ownership
+		cmdq_exec_cnv<N>	Compression and verify slice execution time while
+					owned by cmdq N
+		cmdq_drain_cnv<N>	time taken for cmdq N to release Compression and
+					verify slice ownership
+		cmdq_wait_dcprz<N>	wait time for cmdq N to get Decompression
+					slice N ownership
+		cmdq_exec_dcprz<N>	Decompression slice execution time while
+					owned by cmdq N
+		cmdq_drain_dcprz<N>	time taken for cmdq N to release Decompression
+					slice ownership
+		cmdq_wait_pke<N>	wait time for cmdq N to get PKE slice ownership
+		cmdq_exec_pke<N>	PKE slice execution time while owned by cmdq N
+		cmdq_drain_pke<N>	time taken for cmdq N to release PKE slice
+					ownership
+		cmdq_wait_ucs<N>	wait time for cmdq N to get UCS slice ownership
+		cmdq_exec_ucs<N>	UCS slice execution time while owned by cmdq N
+		cmdq_drain_ucs<N>	time taken for cmdq N to release UCS slice
+					ownership
+		cmdq_wait_ath<N>	wait time for cmdq N to get Authentication slice
+					ownership
+		cmdq_exec_ath<N>	Authentication slice execution time while owned
+					by cmdq N
+		cmdq_drain_ath<N>	time taken for cmdq N to release Authentication
+					slice ownership
 		=======================	========================================
 
 		The telemetry report file can be read with the following command::

Documentation/crypto/api-aead.rst

···
+Authenticated Encryption With Associated Data (AEAD)
+====================================================
+
 Authenticated Encryption With Associated Data (AEAD) Algorithm Definitions
 --------------------------------------------------------------------------
 

Documentation/crypto/api-akcipher.rst

···
+Asymmetric Cipher
+=================
+
 Asymmetric Cipher Algorithm Definitions
 ---------------------------------------
 

Documentation/crypto/api-digest.rst

···
+Message Digest
+==============
+
 Message Digest Algorithm Definitions
 ------------------------------------
 

Documentation/crypto/api-kpp.rst

···
+Key-agreement Protocol Primitives (KPP)
+=======================================
+
 Key-agreement Protocol Primitives (KPP) Cipher Algorithm Definitions
 --------------------------------------------------------------------
 

Documentation/crypto/api-rng.rst

···
+Random Number Generator (RNG)
+=============================
+
 Random Number Algorithm Definitions
 -----------------------------------
 

Documentation/crypto/api-sig.rst

···
+Asymmetric Signature
+====================
+
 Asymmetric Signature Algorithm Definitions
 ------------------------------------------
 

Documentation/crypto/api-skcipher.rst

···
+Symmetric Key Cipher
+====================
+
 Block Cipher Algorithm Definitions
 ----------------------------------
 

Documentation/devicetree/bindings/crypto/ti,am62l-dthev2.yaml

···
+# SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause)
+%YAML 1.2
+---
+$id: http://devicetree.org/schemas/crypto/ti,am62l-dthev2.yaml#
+$schema: http://devicetree.org/meta-schemas/core.yaml#
+
+title: K3 SoC DTHE V2 crypto module
+
+maintainers:
+  - T Pratham <[e-mail]>
+
+properties:
+  compatible:
+    enum:
+      - ti,am62l-dthev2
+
+  reg:
+    maxItems: 1
+
+  dmas:
+    items:
+      - description: AES Engine RX DMA Channel
+      - description: AES Engine TX DMA Channel
+      - description: SHA Engine TX DMA Channel
+
+  dma-names:
+    items:
+      - const: rx
+      - const: tx1
+      - const: tx2
+
+required:
+  - compatible
+  - reg
+  - dmas
+  - dma-names
+
+additionalProperties: false
+
+examples:
+  - |
+    crypto@40800000 {
+        compatible = "ti,am62l-dthev2";
+        reg = <0x40800000 0x10000>;
+
+        dmas = <&main_bcdma 0 0 0x4700 0>,
+               <&main_bcdma 0 0 0xc701 0>,
+               <&main_bcdma 0 0 0xc700 0>;
+        dma-names = "rx", "tx1", "tx2";
+    };

Documentation/devicetree/bindings/crypto/xlnx,versal-trng.yaml

···
+# SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause)
+%YAML 1.2
+---
+$id: http://devicetree.org/schemas/crypto/xlnx,versal-trng.yaml#
+$schema: http://devicetree.org/meta-schemas/core.yaml#
+
+title: Xilinx Versal True Random Number Generator Hardware Accelerator
+
+maintainers:
+  - Harsh Jain <[e-mail]>
+  - Mounika Botcha <[e-mail]>
+
+description:
+  The Versal True Random Number Generator consists of Ring Oscillators as
+  entropy source and a deterministic CTR_DRBG random bit generator (DRBG).
+
+properties:
+  compatible:
+    const: xlnx,versal-trng
+
+  reg:
+    maxItems: 1
+
+required:
+  - reg
+
+additionalProperties: false
+
+examples:
+  - |
+    rng@f1230000 {
+        compatible = "xlnx,versal-trng";
+        reg = <0xf1230000 0x1000>;
+    };
+...

Documentation/devicetree/bindings/rng/hisi-rng.txt

···
-Hisilicon Random Number Generator
-
-Required properties:
-- compatible : Should be "hisilicon,hip04-rng" or "hisilicon,hip05-rng"
-- reg : Offset and length of the register set of this block
-
-Example:
-
-rng@d1010000 {
-	compatible = "hisilicon,hip05-rng";
-	reg = <0xd1010000 0x100>;
-};

Documentation/devicetree/bindings/rng/hisi-rng.yaml

···
+# SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause)
+%YAML 1.2
+---
+$id: http://devicetree.org/schemas/rng/hisi-rng.yaml#
+$schema: http://devicetree.org/meta-schemas/core.yaml#
+
+title: Hisilicon Random Number Generator
+
+maintainers:
+  - Kefeng Wang <wangkefeng.wang@huawei>
+
+properties:
+  compatible:
+    enum:
+      - hisilicon,hip04-rng
+      - hisilicon,hip05-rng
+
+  reg:
+    maxItems: 1
+
+required:
+  - compatible
+  - reg
+
+additionalProperties: false
+
+examples:
+  - |
+    rng@d1010000 {
+      compatible = "hisilicon,hip05-rng";
+      reg = <0xd1010000 0x100>;
+    };

MAINTAINERS

···
 F:	drivers/clk/ti/
 F:	include/linux/clk/ti.h
 
+TI DATA TRANSFORM AND HASHING ENGINE (DTHE) V2 CRYPTO DRIVER
+M:	T Pratham <[e-mail]>
+L:	[e-mail]
+S:	Supported
+F:	Documentation/devicetree/bindings/crypto/ti,am62l-dthev2.yaml
+F:	drivers/crypto/ti/
+
 TI DAVINCI MACHINE SUPPORT
 M:	Bartosz Golaszewski <[e-mail]>
 L:	[e-mail] (moderated for non-subscribers)
···
 F:	Documentation/misc-devices/xilinx_sdfec.rst
 F:	drivers/misc/xilinx_sdfec.c
 F:	include/uapi/misc/xilinx_sdfec.h
+
+XILINX TRNG DRIVER
+M:	Mounika Botcha <[e-mail]>
+M:	Harsh Jain <[e-mail]>
+S:	Maintained
+F:	drivers/crypto/xilinx/xilinx-trng.c
 
 XILINX UARTLITE SERIAL DRIVER
 M:	Peter Korsgaard <[e-mail]>

arch/arm64/crypto/Kconfig

···
 config CRYPTO_AES_ARM64
 	tristate "Ciphers: AES, modes: ECB, CBC, CTR, CTS, XCTR, XTS"
 	select CRYPTO_AES
+	select CRYPTO_LIB_SHA256
 	help
 	  Block ciphers: AES cipher algorithms (FIPS-197)
 	  Length-preserving ciphers: AES with ECB, CBC, CTR, CTS,

arch/arm64/crypto/aes-glue.c

···
 struct crypto_aes_essiv_cbc_ctx {
 	struct crypto_aes_ctx key1;
 	struct crypto_aes_ctx __aligned(8) key2;
-	struct crypto_shash *hash;
 };
 
 struct mac_tfm_ctx {
···
 	if (ret)
 		return ret;
 
-	crypto_shash_tfm_digest(ctx->hash, in_key, key_len, digest);
+	sha256(in_key, key_len, digest);
 
 	return aes_expandkey(&ctx->key2, digest, sizeof(digest));
 }
···
 	kernel_neon_end();
 
 	return skcipher_walk_done(&walk, 0);
-}
-
-static int __maybe_unused essiv_cbc_init_tfm(struct crypto_skcipher *tfm)
-{
-	struct crypto_aes_essiv_cbc_ctx *ctx = crypto_skcipher_ctx(tfm);
-
-	ctx->hash = crypto_alloc_shash("sha256", 0, 0);
-
-	return PTR_ERR_OR_ZERO(ctx->hash);
-}
-
-static void __maybe_unused essiv_cbc_exit_tfm(struct crypto_skcipher *tfm)
-{
-	struct crypto_aes_essiv_cbc_ctx *ctx = crypto_skcipher_ctx(tfm);
-
-	crypto_free_shash(ctx->hash);
 }
 
 static int __maybe_unused essiv_cbc_encrypt(struct skcipher_request *req)
···
 	.setkey		= essiv_cbc_set_key,
 	.encrypt	= essiv_cbc_encrypt,
 	.decrypt	= essiv_cbc_decrypt,
-	.init		= essiv_cbc_init_tfm,
-	.exit		= essiv_cbc_exit_tfm,
 } };
 
 static int cbcmac_setkey(struct crypto_shash *tfm, const u8 *in_key,

arch/s390/crypto/sha.h

···
 #ifndef _CRYPTO_ARCH_S390_SHA_H
 #define _CRYPTO_ARCH_S390_SHA_H
 
+#include <crypto/hash.h>
 #include <crypto/sha2.h>
 #include <crypto/sha3.h>
+#include <linux/build_bug.h>
 #include <linux/types.h>
 
 /* must be big enough for the largest SHA variant */
 #define CPACF_MAX_PARMBLOCK_SIZE	SHA3_STATE_SIZE
 #define SHA_MAX_BLOCK_SIZE		SHA3_224_BLOCK_SIZE
-#define S390_SHA_CTX_SIZE		sizeof(struct s390_sha_ctx)
 
 struct s390_sha_ctx {
 	u64 count;		/* message length in bytes */
···
 			   unsigned int len);
 int s390_sha_finup(struct shash_desc *desc, const u8 *src, unsigned int len,
 		   u8 *out);
+
+static inline void __check_s390_sha_ctx_size(void)
+{
+	BUILD_BUG_ON(S390_SHA_CTX_SIZE != sizeof(struct s390_sha_ctx));
+}
 
 #endif

crypto/842.c

···
 }
 
 static struct scomp_alg scomp = {
-	.alloc_ctx		= crypto842_alloc_ctx,
-	.free_ctx		= crypto842_free_ctx,
+	.streams		= {
+		.alloc_ctx	= crypto842_alloc_ctx,
+		.free_ctx	= crypto842_free_ctx,
+	},
 	.compress		= crypto842_scompress,
 	.decompress		= crypto842_sdecompress,
 	.base			= {

crypto/anubis.c

···
 
 static int __init anubis_mod_init(void)
 {
-	int ret = 0;
-
-	ret = crypto_register_alg(&anubis_alg);
-	return ret;
+	return crypto_register_alg(&anubis_alg);
 }
 
 static void __exit anubis_mod_fini(void)

crypto/asymmetric_keys/x509_cert_parser.c

···
 		/*
 		 * Get hold of the basicConstraints
 		 * v[1] is the encoding size
-		 *	(Expect 0x2 or greater, making it 1 or more bytes)
+		 *	(Expect 0x00 for empty SEQUENCE with CA:FALSE, or
+		 *	0x03 or greater for non-empty SEQUENCE)
 		 * v[2] is the encoding type
 		 *	(Expect an ASN1_BOOL for the CA)
-		 * v[3] is the contents of the ASN1_BOOL
-		 *      (Expect 1 if the CA is TRUE)
+		 * v[3] is the length of the ASN1_BOOL
+		 *	(Expect 1 for a single byte boolean)
+		 * v[4] is the contents of the ASN1_BOOL
+		 *	(Expect 0xFF if the CA is TRUE)
 		 * vlen should match the entire extension size
 		 */
 		if (v[0] != (ASN1_CONS_BIT | ASN1_SEQ))
···
 			return -EBADMSG;
 		if (v[1] != vlen - 2)
 			return -EBADMSG;
-		if (vlen >= 4 && v[1] != 0 && v[2] == ASN1_BOOL && v[3] == 1)
+		/* Empty SEQUENCE means CA:FALSE (default value omitted per DER) */
+		if (v[1] == 0)
+			return 0;
+		if (vlen >= 5 && v[2] == ASN1_BOOL && v[3] == 1 && v[4] == 0xFF)
 			ctx->cert->pub->key_eflags |= 1 << KEY_EFLAG_CA;
+		else
+			return -EBADMSG;
 		return 0;
 	}
 

crypto/cryptd.c

···
 {
 	int err;
 
-	cryptd_wq = alloc_workqueue("cryptd", WQ_MEM_RECLAIM | WQ_CPU_INTENSIVE,
+	cryptd_wq = alloc_workqueue("cryptd",
+				    WQ_MEM_RECLAIM | WQ_CPU_INTENSIVE | WQ_PERCPU,
 				    1);
 	if (!cryptd_wq)
 		return -ENOMEM;

crypto/jitterentropy-kcapi.c

···
 		pr_warn_ratelimited("Unexpected digest size\n");
 		return -EINVAL;
 	}
+	kmsan_unpoison_memory(intermediary, sizeof(intermediary));
 
 	/*
 	 * This loop fills a buffer which is injected into the entropy pool.

crypto/lz4.c

···
 }
 
 static struct scomp_alg scomp = {
-	.alloc_ctx		= lz4_alloc_ctx,
-	.free_ctx		= lz4_free_ctx,
+	.streams		= {
+		.alloc_ctx	= lz4_alloc_ctx,
+		.free_ctx	= lz4_free_ctx,
+	},
 	.compress		= lz4_scompress,
 	.decompress		= lz4_sdecompress,
 	.base			= {

crypto/lz4hc.c

···
 }
 
 static struct scomp_alg scomp = {
-	.alloc_ctx		= lz4hc_alloc_ctx,
-	.free_ctx		= lz4hc_free_ctx,
+	.streams		= {
+		.alloc_ctx	= lz4hc_alloc_ctx,
+		.free_ctx	= lz4hc_free_ctx,
+	},
 	.compress		= lz4hc_scompress,
 	.decompress		= lz4hc_sdecompress,
 	.base			= {

crypto/lzo-rle.c

···
 }
 
 static struct scomp_alg scomp = {
-	.alloc_ctx		= lzorle_alloc_ctx,
-	.free_ctx		= lzorle_free_ctx,
+	.streams		= {
+		.alloc_ctx	= lzorle_alloc_ctx,
+		.free_ctx	= lzorle_free_ctx,
+	},
 	.compress		= lzorle_scompress,
 	.decompress		= lzorle_sdecompress,
 	.base			= {

crypto/lzo.c

···
 }
 
 static struct scomp_alg scomp = {
-	.alloc_ctx		= lzo_alloc_ctx,
-	.free_ctx		= lzo_free_ctx,
+	.streams		= {
+		.alloc_ctx	= lzo_alloc_ctx,
+		.free_ctx	= lzo_free_ctx,
+	},
 	.compress		= lzo_scompress,
 	.decompress		= lzo_sdecompress,
 	.base			= {

drivers/char/hw_random/Kconfig

···
 config HW_RANDOM_NOMADIK
 	tristate "ST-Ericsson Nomadik Random Number Generator support"
 	depends on ARCH_NOMADIK || COMPILE_TEST
+	depends on ARM_AMBA
 	default HW_RANDOM
 	help
 	  This driver provides kernel-side support for the Random Number

drivers/char/hw_random/cn10k-rng.c

···
 
 	rng->reg_base = pcim_iomap(pdev, 0, 0);
 	if (!rng->reg_base)
-		return dev_err_probe(&pdev->dev, -ENOMEM, "Error while mapping CSRs, exiting\n");
+		return -ENOMEM;
 
 	rng->ops.name = devm_kasprintf(&pdev->dev, GFP_KERNEL,
 				       "cn10k-rng-%s", dev_name(&pdev->dev));

drivers/char/hw_random/ks-sa-rng.c

···
 	if (IS_ERR(ks_sa_rng->regmap_cfg))
 		return dev_err_probe(dev, -EINVAL, "syscon_node_to_regmap failed\n");
 
+	ks_sa_rng->clk = devm_clk_get_enabled(dev, NULL);
+	if (IS_ERR(ks_sa_rng->clk))
+		return dev_err_probe(dev, PTR_ERR(ks_sa_rng->clk), "Failed to get clock\n");
+
 	pm_runtime_enable(dev);
 	ret = pm_runtime_resume_and_get(dev);
 	if (ret < 0) {

drivers/char/hw_random/timeriomem-rng.c

···
 		priv->rng_ops.quality = pdata->quality;
 	}
 
-	priv->period = ns_to_ktime(period * NSEC_PER_USEC);
+	priv->period = us_to_ktime(period);
 	init_completion(&priv->completion);
 	hrtimer_setup(&priv->timer, timeriomem_rng_trigger, CLOCK_MONOTONIC, HRTIMER_MODE_ABS);
 

drivers/crypto/Kconfig

···
 	  Select this to enable Tegra Security Engine which accelerates various
 	  AES encryption/decryption and HASH algorithms.
 
+config CRYPTO_DEV_XILINX_TRNG
+	tristate "Support for Xilinx True Random Generator"
+	depends on ZYNQMP_FIRMWARE || COMPILE_TEST
+	select CRYPTO_RNG
+	select HW_RANDOM
+	help
+	  Xilinx Versal SoC driver provides kernel-side support for True Random Number
+	  Generator and Pseudo random Number in CTR_DRBG mode as defined in NIST SP800-90A.
+
+	  To compile this driver as a module, choose M here: the module
+	  will be called xilinx-trng.
+
 config CRYPTO_DEV_ZYNQMP_AES
 	tristate "Support for Xilinx ZynqMP AES hw accelerator"
 	depends on ZYNQMP_FIRMWARE || COMPILE_TEST
···
 source "drivers/crypto/aspeed/Kconfig"
 source "drivers/crypto/starfive/Kconfig"
 source "drivers/crypto/inside-secure/eip93/Kconfig"
+source "drivers/crypto/ti/Kconfig"
 
 endif # CRYPTO_HW

drivers/crypto/Makefile

···
 obj-y += intel/
 obj-y += starfive/
 obj-y += cavium/
+obj-y += ti/

drivers/crypto/allwinner/sun8i-ce/sun8i-ce-cipher.c

···
 
 	if (IS_ENABLED(CONFIG_CRYPTO_DEV_SUN8I_CE_DEBUG)) {
 		struct skcipher_alg *alg = crypto_skcipher_alg(tfm);
-		struct sun8i_ce_alg_template *algt __maybe_unused;
+		struct sun8i_ce_alg_template *algt;
 
 		algt = container_of(alg, struct sun8i_ce_alg_template,
 				    alg.skcipher.base);
···
 	return err;
 }
 
-static int sun8i_ce_cipher_prepare(struct crypto_engine *engine, void *async_req)
+static int sun8i_ce_cipher_prepare(struct skcipher_request *areq,
+				   struct ce_task *cet)
 {
-	struct skcipher_request *areq = container_of(async_req, struct skcipher_request, base);
 	struct crypto_skcipher *tfm = crypto_skcipher_reqtfm(areq);
 	struct sun8i_cipher_tfm_ctx *op = crypto_skcipher_ctx(tfm);
 	struct sun8i_ce_dev *ce = op->ce;
 	struct sun8i_cipher_req_ctx *rctx = skcipher_request_ctx(areq);
 	struct skcipher_alg *alg = crypto_skcipher_alg(tfm);
 	struct sun8i_ce_alg_template *algt;
-	struct sun8i_ce_flow *chan;
-	struct ce_task *cet;
 	struct scatterlist *sg;
 	unsigned int todo, len, offset, ivsize;
 	u32 common, sym;
-	int flow, i;
+	int i;
 	int nr_sgs = 0;
 	int nr_sgd = 0;
 	int err = 0;
···
 	if (IS_ENABLED(CONFIG_CRYPTO_DEV_SUN8I_CE_DEBUG))
 		algt->stat_req++;
 
-	flow = rctx->flow;
-
-	chan = &ce->chanlist[flow];
-
-	cet = chan->tl;
 	memset(cet, 0, sizeof(struct ce_task));
 
-	cet->t_id = cpu_to_le32(flow);
+	cet->t_id = cpu_to_le32(rctx->flow);
 	common = ce->variant->alg_cipher[algt->ce_algo_id];
 	common |= rctx->op_dir | CE_COMM_INT;
 	cet->t_common_ctl = cpu_to_le32(common);
···
 	if (areq->iv && ivsize > 0) {
 		if (rctx->op_dir & CE_DECRYPTION) {
 			offset = areq->cryptlen - ivsize;
-			scatterwalk_map_and_copy(chan->backup_iv, areq->src,
+			scatterwalk_map_and_copy(rctx->backup_iv, areq->src,
 						 offset, ivsize, 0);
 		}
-		memcpy(chan->bounce_iv, areq->iv, ivsize);
-		rctx->addr_iv = dma_map_single(ce->dev, chan->bounce_iv, ivsize,
+		memcpy(rctx->bounce_iv, areq->iv, ivsize);
+		rctx->addr_iv = dma_map_single(ce->dev, rctx->bounce_iv, ivsize,
 					       DMA_TO_DEVICE);
 		if (dma_mapping_error(ce->dev, rctx->addr_iv)) {
 			dev_err(ce->dev, "Cannot DMA MAP IV\n");
···
 		goto theend_sgs;
 	}
 
-	chan->timeout = areq->cryptlen;
 	rctx->nr_sgs = ns;
 	rctx->nr_sgd = nd;
 	return 0;
···
 
 		offset = areq->cryptlen - ivsize;
 		if (rctx->op_dir & CE_DECRYPTION) {
-			memcpy(areq->iv, chan->backup_iv, ivsize);
-			memzero_explicit(chan->backup_iv, ivsize);
+			memcpy(areq->iv, rctx->backup_iv, ivsize);
+			memzero_explicit(rctx->backup_iv, ivsize);
 		} else {
 			scatterwalk_map_and_copy(areq->iv, areq->dst, offset,
 						 ivsize, 0);
 		}
-		memzero_explicit(chan->bounce_iv, ivsize);
+		memzero_explicit(rctx->bounce_iv, ivsize);
 	}
 
 	dma_unmap_single(ce->dev, rctx->addr_key, op->keylen, DMA_TO_DEVICE);
···
 	return err;
 }
 
-static void sun8i_ce_cipher_unprepare(struct crypto_engine *engine,
-				      void *async_req)
+static void sun8i_ce_cipher_unprepare(struct skcipher_request *areq,
+				      struct ce_task *cet)
 {
-	struct skcipher_request *areq = container_of(async_req, struct skcipher_request, base);
 	struct crypto_skcipher *tfm = crypto_skcipher_reqtfm(areq);
 	struct sun8i_cipher_tfm_ctx *op = crypto_skcipher_ctx(tfm);
 	struct sun8i_ce_dev *ce = op->ce;
 	struct sun8i_cipher_req_ctx *rctx = skcipher_request_ctx(areq);
-	struct sun8i_ce_flow *chan;
-	struct ce_task *cet;
 	unsigned int ivsize, offset;
 	int nr_sgs = rctx->nr_sgs;
 	int nr_sgd = rctx->nr_sgd;
-	int flow;
 
-	flow = rctx->flow;
-	chan = &ce->chanlist[flow];
-	cet = chan->tl;
 	ivsize = crypto_skcipher_ivsize(tfm);
 
 	if (areq->src == areq->dst) {
···
 					 DMA_TO_DEVICE);
 		offset = areq->cryptlen - ivsize;
 		if (rctx->op_dir & CE_DECRYPTION) {
-			memcpy(areq->iv, chan->backup_iv, ivsize);
-			memzero_explicit(chan->backup_iv, ivsize);
+			memcpy(areq->iv, rctx->backup_iv, ivsize);
+			memzero_explicit(rctx->backup_iv, ivsize);
 		} else {
 			scatterwalk_map_and_copy(areq->iv, areq->dst, offset,
 						 ivsize, 0);
 		}
-		memzero_explicit(chan->bounce_iv, ivsize);
+		memzero_explicit(rctx->bounce_iv, ivsize);
 	}
 
 	dma_unmap_single(ce->dev, rctx->addr_key, op->keylen, DMA_TO_DEVICE);
 }
 
-static void sun8i_ce_cipher_run(struct crypto_engine *engine, void *areq)
-{
-	struct skcipher_request *breq = container_of(areq, struct skcipher_request, base);
-	struct crypto_skcipher *tfm = crypto_skcipher_reqtfm(breq);
-	struct sun8i_cipher_tfm_ctx *op = crypto_skcipher_ctx(tfm);
-	struct sun8i_ce_dev *ce = op->ce;
-	struct sun8i_cipher_req_ctx *rctx = skcipher_request_ctx(breq);
-	int flow, err;
-
-	flow = rctx->flow;
-	err = sun8i_ce_run_task(ce, flow, crypto_tfm_alg_name(breq->base.tfm));
-	sun8i_ce_cipher_unprepare(engine, areq);
-	local_bh_disable();
-	crypto_finalize_skcipher_request(engine, breq, err);
-	local_bh_enable();
-}
-
 int sun8i_ce_cipher_do_one(struct crypto_engine *engine, void *areq)
 {
-	int err = sun8i_ce_cipher_prepare(engine, areq);
+	struct skcipher_request *req = skcipher_request_cast(areq);
+	struct sun8i_cipher_req_ctx *rctx = skcipher_request_ctx(req);
+	struct crypto_skcipher *tfm = crypto_skcipher_reqtfm(req);
+	struct sun8i_cipher_tfm_ctx *ctx = crypto_skcipher_ctx(tfm);
+	struct sun8i_ce_dev *ce = ctx->ce;
+	struct sun8i_ce_flow *chan;
+	int err;
 
+	chan = &ce->chanlist[rctx->flow];
+
+	err = sun8i_ce_cipher_prepare(req, chan->tl);
 	if (err)
 		return err;
 
-	sun8i_ce_cipher_run(engine, areq);
+	err = sun8i_ce_run_task(ce, rctx->flow,
+				crypto_tfm_alg_name(req->base.tfm));
+
+	sun8i_ce_cipher_unprepare(req, chan->tl);
+
+	local_bh_disable();
+	crypto_finalize_skcipher_request(engine, req, err);
+	local_bh_enable();
+
 	return 0;
 }
 

drivers/crypto/allwinner/sun8i-ce/sun8i-ce-core.c

···
 	.trng = CE_ID_NOTSUPP,
 };
 
+static void sun8i_ce_dump_task_descriptors(struct sun8i_ce_flow *chan)
+{
+	print_hex_dump(KERN_INFO, "TASK: ", DUMP_PREFIX_NONE, 16, 4,
+		       chan->tl, sizeof(struct ce_task), false);
+}
+
 /*
  * sun8i_ce_get_engine_number() get the next channel slot
  * This is a simple round-robin way of getting the next channel
···
 {
 	u32 v;
 	int err = 0;
-	struct ce_task *cet = ce->chanlist[flow].tl;
 
 #ifdef CONFIG_CRYPTO_DEV_SUN8I_CE_DEBUG
 	ce->chanlist[flow].stat_req++;
···
 	mutex_unlock(&ce->mlock);
 
 	wait_for_completion_interruptible_timeout(&ce->chanlist[flow].complete,
-			msecs_to_jiffies(ce->chanlist[flow].timeout));
+			msecs_to_jiffies(CE_DMA_TIMEOUT_MS));
 
 	if (ce->chanlist[flow].status == 0) {
-		dev_err(ce->dev, "DMA timeout for %s (tm=%d) on flow %d\n", name,
-			ce->chanlist[flow].timeout, flow);
+		dev_err(ce->dev, "DMA timeout for %s on flow %d\n", name, flow);
 		err = -EFAULT;
 	}
 	/* No need to lock for this read, the channel is locked so
···
 		/* Sadly, the error bit is not per flow */
 		if (v) {
 			dev_err(ce->dev, "CE ERROR: %x for flow %x\n", v, flow);
+			sun8i_ce_dump_task_descriptors(&ce->chanlist[flow]);
 			err = -EFAULT;
-			print_hex_dump(KERN_INFO, "TASK: ", DUMP_PREFIX_NONE, 16, 4,
-				       cet, sizeof(struct ce_task), false);
 		}
 		if (v & CE_ERR_ALGO_NOTSUP)
 			dev_err(ce->dev, "CE ERROR: algorithm not supported\n");
···
 		v &= 0xF;
 		if (v) {
 			dev_err(ce->dev, "CE ERROR: %x for flow %x\n", v, flow);
+			sun8i_ce_dump_task_descriptors(&ce->chanlist[flow]);
 			err = -EFAULT;
-			print_hex_dump(KERN_INFO, "TASK: ", DUMP_PREFIX_NONE, 16, 4,
-				       cet, sizeof(struct ce_task), false);
 		}
 		if (v & CE_ERR_ALGO_NOTSUP)
 			dev_err(ce->dev, "CE ERROR: algorithm not supported\n");
···
 		v &= 0xFF;
 		if (v) {
 			dev_err(ce->dev, "CE ERROR: %x for flow %x\n", v, flow);
+			sun8i_ce_dump_task_descriptors(&ce->chanlist[flow]);
 			err = -EFAULT;
-			print_hex_dump(KERN_INFO, "TASK: ", DUMP_PREFIX_NONE, 16, 4,
-				       cet, sizeof(struct ce_task), false);
 		}
 		if (v & CE_ERR_ALGO_NOTSUP)
 			dev_err(ce->dev, "CE ERROR: algorithm not supported\n");
···
 			err = -ENOMEM;
 			goto error_engine;
 		}
-		ce->chanlist[i].bounce_iv = devm_kmalloc(ce->dev, AES_BLOCK_SIZE,
-							 GFP_KERNEL | GFP_DMA);
-		if (!ce->chanlist[i].bounce_iv) {
-			err = -ENOMEM;
-			goto error_engine;
-		}
-		ce->chanlist[i].backup_iv = devm_kmalloc(ce->dev, AES_BLOCK_SIZE,
-							 GFP_KERNEL);
-		if (!ce->chanlist[i].backup_iv) {
-			err = -ENOMEM;
-			goto error_engine;
-		}
 	}
 	return 0;
 error_engine:
···
 	pm_runtime_put_sync(ce->dev);
 
 	if (IS_ENABLED(CONFIG_CRYPTO_DEV_SUN8I_CE_DEBUG)) {
-		struct dentry *dbgfs_dir __maybe_unused;
+		struct dentry *dbgfs_dir;
 		struct dentry *dbgfs_stats __maybe_unused;
 
 		/* Ignore error of debugfs */

drivers/crypto/allwinner/sun8i-ce/sun8i-ce-hash.c

···
 static void sun8i_ce_hash_stat_fb_inc(struct crypto_ahash *tfm)
 {
 	if (IS_ENABLED(CONFIG_CRYPTO_DEV_SUN8I_CE_DEBUG)) {
-		struct sun8i_ce_alg_template *algt __maybe_unused;
+		struct sun8i_ce_alg_template *algt;
 		struct ahash_alg *alg = crypto_ahash_alg(tfm);
 
 		algt = container_of(alg, struct sun8i_ce_alg_template,
···
 
 	crypto_ahash_set_reqsize(tfm,
 				 sizeof(struct sun8i_ce_hash_reqctx) +
-				 crypto_ahash_reqsize(op->fallback_tfm));
+				 crypto_ahash_reqsize(op->fallback_tfm) +
+				 CRYPTO_DMA_PADDING);
 
 	if (IS_ENABLED(CONFIG_CRYPTO_DEV_SUN8I_CE_DEBUG))
 		memcpy(algt->fbname,
···
 
 int sun8i_ce_hash_init(struct ahash_request *areq)
 {
-	struct sun8i_ce_hash_reqctx *rctx = ahash_request_ctx(areq);
+	struct sun8i_ce_hash_reqctx *rctx = ahash_request_ctx_dma(areq);
 	struct crypto_ahash *tfm = crypto_ahash_reqtfm(areq);
 	struct sun8i_ce_hash_tfm_ctx *tfmctx = crypto_ahash_ctx(tfm);
 
···
 
 int sun8i_ce_hash_export(struct ahash_request *areq, void *out)
 {
-	struct sun8i_ce_hash_reqctx *rctx = ahash_request_ctx(areq);
+	struct sun8i_ce_hash_reqctx *rctx = ahash_request_ctx_dma(areq);
 	struct crypto_ahash *tfm = crypto_ahash_reqtfm(areq);
 	struct sun8i_ce_hash_tfm_ctx *tfmctx = crypto_ahash_ctx(tfm);
 
···
 
 int sun8i_ce_hash_import(struct ahash_request *areq, const void *in)
 {
-	struct sun8i_ce_hash_reqctx *rctx = ahash_request_ctx(areq);
+	struct sun8i_ce_hash_reqctx *rctx = ahash_request_ctx_dma(areq);
 	struct crypto_ahash *tfm = crypto_ahash_reqtfm(areq);
 	struct sun8i_ce_hash_tfm_ctx *tfmctx = crypto_ahash_ctx(tfm);
 
···
 
 int sun8i_ce_hash_final(struct ahash_request *areq)
 {
-	struct sun8i_ce_hash_reqctx *rctx = ahash_request_ctx(areq);
+	struct sun8i_ce_hash_reqctx *rctx = ahash_request_ctx_dma(areq);
 	struct crypto_ahash *tfm = crypto_ahash_reqtfm(areq);
 	struct sun8i_ce_hash_tfm_ctx *tfmctx = crypto_ahash_ctx(tfm);
 
···
 
 int sun8i_ce_hash_update(struct ahash_request *areq)
 {
-	struct sun8i_ce_hash_reqctx *rctx = ahash_request_ctx(areq);
+	struct sun8i_ce_hash_reqctx *rctx = ahash_request_ctx_dma(areq);
 	struct crypto_ahash *tfm = crypto_ahash_reqtfm(areq);
 	struct sun8i_ce_hash_tfm_ctx *tfmctx = crypto_ahash_ctx(tfm);
 
···
 
 int sun8i_ce_hash_finup(struct ahash_request *areq)
 {
-	struct sun8i_ce_hash_reqctx *rctx = ahash_request_ctx(areq);
+	struct sun8i_ce_hash_reqctx *rctx = ahash_request_ctx_dma(areq);
 	struct crypto_ahash *tfm = crypto_ahash_reqtfm(areq);
 	struct sun8i_ce_hash_tfm_ctx *tfmctx = crypto_ahash_ctx(tfm);
 
···
 
 static int sun8i_ce_hash_digest_fb(struct ahash_request *areq)
 {
-	struct sun8i_ce_hash_reqctx *rctx = ahash_request_ctx(areq);
+	struct sun8i_ce_hash_reqctx *rctx = ahash_request_ctx_dma(areq);
 	struct crypto_ahash *tfm = crypto_ahash_reqtfm(areq);
 	struct sun8i_ce_hash_tfm_ctx *tfmctx = crypto_ahash_ctx(tfm);
 
···
 int sun8i_ce_hash_digest(struct ahash_request *areq)
 {
 	struct crypto_ahash *tfm = crypto_ahash_reqtfm(areq);
-	struct ahash_alg *alg = __crypto_ahash_alg(tfm->base.__crt_alg);
-	struct sun8i_ce_hash_reqctx *rctx = ahash_request_ctx(areq);
-	struct sun8i_ce_alg_template *algt;
-	struct sun8i_ce_dev *ce;
+	struct sun8i_ce_hash_tfm_ctx *ctx = crypto_ahash_ctx(tfm);
+	struct sun8i_ce_hash_reqctx *rctx = ahash_request_ctx_dma(areq);
+	struct sun8i_ce_dev *ce = ctx->ce;
 	struct crypto_engine *engine;
 	int e;
 
 	if (sun8i_ce_hash_need_fallback(areq))
 		return sun8i_ce_hash_digest_fb(areq);
-
-	algt = container_of(alg, struct sun8i_ce_alg_template, alg.hash.base);
-	ce = algt->ce;
 
 	e = sun8i_ce_get_engine_number(ce);
 	rctx->flow = e;
···
 	return j;
 }
 
-int sun8i_ce_hash_run(struct crypto_engine *engine, void *breq)
+static int sun8i_ce_hash_prepare(struct ahash_request *areq, struct ce_task *cet)
 {
-	struct ahash_request *areq = container_of(breq, struct ahash_request, base);
 	struct crypto_ahash *tfm = crypto_ahash_reqtfm(areq);
 	struct ahash_alg *alg = __crypto_ahash_alg(tfm->base.__crt_alg);
-	struct sun8i_ce_hash_reqctx *rctx = ahash_request_ctx(areq);
+	struct sun8i_ce_hash_reqctx *rctx = ahash_request_ctx_dma(areq);
 	struct sun8i_ce_alg_template *algt;
 	struct sun8i_ce_dev *ce;
-	struct sun8i_ce_flow *chan;
-	struct ce_task *cet;
 	struct scatterlist *sg;
-	int nr_sgs, flow, err;
+	int nr_sgs, err;
 	unsigned int len;
 	u32 common;
 	u64 byte_count;
 	__le32 *bf;
-	void *buf, *result;
 	int j, i, todo;
 	u64 bs;
 	int digestsize;
-	dma_addr_t addr_res, addr_pad;
-	int ns = sg_nents_for_len(areq->src, areq->nbytes);
 
 	algt = container_of(alg, struct sun8i_ce_alg_template, alg.hash.base);
 	ce = algt->ce;
···
 	if (digestsize == SHA384_DIGEST_SIZE)
 		digestsize = SHA512_DIGEST_SIZE;
 
-	/* the padding could be up to two block. */
-	buf = kcalloc(2, bs, GFP_KERNEL | GFP_DMA);
-	if (!buf) {
-		err = -ENOMEM;
-		goto err_out;
-	}
-	bf = (__le32 *)buf;
-
-	result = kzalloc(digestsize, GFP_KERNEL | GFP_DMA);
-	if (!result) {
-		err = -ENOMEM;
-		goto err_free_buf;
-	}
-
-	flow = rctx->flow;
-	chan = &ce->chanlist[flow];
+	bf = (__le32 *)rctx->pad;
 
 	if (IS_ENABLED(CONFIG_CRYPTO_DEV_SUN8I_CE_DEBUG))
 		algt->stat_req++;
 
 	dev_dbg(ce->dev, "%s %s len=%d\n", __func__, crypto_tfm_alg_name(areq->base.tfm), areq->nbytes);
 
-	cet = chan->tl;
 	memset(cet, 0, sizeof(struct ce_task));
 
-	cet->t_id = cpu_to_le32(flow);
+	cet->t_id = cpu_to_le32(rctx->flow);
 	common = ce->variant->alg_hash[algt->ce_algo_id];
 	common |= CE_COMM_INT;
 	cet->t_common_ctl = cpu_to_le32(common);
···
 	cet->t_sym_ctl = 0;
 	cet->t_asym_ctl = 0;
 
-	nr_sgs = dma_map_sg(ce->dev, areq->src, ns, DMA_TO_DEVICE);
+	rctx->nr_sgs = sg_nents_for_len(areq->src, areq->nbytes);
+	nr_sgs = dma_map_sg(ce->dev, areq->src, rctx->nr_sgs, DMA_TO_DEVICE);
 	if (nr_sgs <= 0 || nr_sgs > MAX_SG) {
 		dev_err(ce->dev, "Invalid sg number %d\n", nr_sgs);
 		err = -EINVAL;
-		goto err_free_result;
+		goto err_out;
 	}
 
 	len = areq->nbytes;
···
 		err = -EINVAL;
 		goto err_unmap_src;
 	}
-	addr_res = dma_map_single(ce->dev, result, digestsize, DMA_FROM_DEVICE);
-	cet->t_dst[0].addr = desc_addr_val_le32(ce, addr_res);
-	cet->t_dst[0].len = cpu_to_le32(digestsize / 4);
-	if (dma_mapping_error(ce->dev, addr_res)) {
+
+	rctx->result_len = digestsize;
+	rctx->addr_res = dma_map_single(ce->dev, rctx->result, rctx->result_len,
+					DMA_FROM_DEVICE);
+	cet->t_dst[0].addr = desc_addr_val_le32(ce, rctx->addr_res);
+	cet->t_dst[0].len = cpu_to_le32(rctx->result_len / 4);
+	if (dma_mapping_error(ce->dev, rctx->addr_res)) {
 		dev_err(ce->dev, "DMA map dest\n");
 		err = -EINVAL;
 		goto err_unmap_src;
···
 		goto err_unmap_result;
 	}
 
-	addr_pad = dma_map_single(ce->dev, buf, j * 4, DMA_TO_DEVICE);
-	cet->t_src[i].addr = desc_addr_val_le32(ce, addr_pad);
+	rctx->pad_len = j * 4;
+	rctx->addr_pad = dma_map_single(ce->dev, rctx->pad, rctx->pad_len,
+					DMA_TO_DEVICE);
+	cet->t_src[i].addr = desc_addr_val_le32(ce, rctx->addr_pad);
 	cet->t_src[i].len = cpu_to_le32(j);
-	if (dma_mapping_error(ce->dev, addr_pad)) {
+	if (dma_mapping_error(ce->dev, rctx->addr_pad)) {
 		dev_err(ce->dev, "DMA error on padding SG\n");
 		err = -EINVAL;
 		goto err_unmap_result;
···
 	else
 		cet->t_dlen = cpu_to_le32(areq->nbytes / 4 + j);
 
-	chan->timeout = areq->nbytes;
-
-	err = sun8i_ce_run_task(ce, flow, crypto_ahash_alg_name(tfm));
-
-	dma_unmap_single(ce->dev, addr_pad, j * 4, DMA_TO_DEVICE);
+	return 0;
 
 err_unmap_result:
-	dma_unmap_single(ce->dev, addr_res, digestsize, DMA_FROM_DEVICE);
-	if (!err)
-		memcpy(areq->result, result, crypto_ahash_digestsize(tfm));
+	dma_unmap_single(ce->dev, rctx->addr_res, rctx->result_len,
+			 DMA_FROM_DEVICE);
 
 err_unmap_src:
-	dma_unmap_sg(ce->dev, areq->src, ns, DMA_TO_DEVICE);
-
-err_free_result:
-	kfree(result);
-
-err_free_buf:
-	kfree(buf);
+	dma_unmap_sg(ce->dev, areq->src, rctx->nr_sgs, DMA_TO_DEVICE);
 
 err_out:
+	return err;
+}
+
+static void sun8i_ce_hash_unprepare(struct ahash_request *areq,
+				    struct ce_task *cet)
+{
+	struct sun8i_ce_hash_reqctx *rctx = ahash_request_ctx_dma(areq);
+	struct crypto_ahash *tfm = crypto_ahash_reqtfm(areq);
+	struct sun8i_ce_hash_tfm_ctx *ctx = crypto_ahash_ctx(tfm);
+	struct sun8i_ce_dev *ce = ctx->ce;
+
+	dma_unmap_single(ce->dev, rctx->addr_pad, rctx->pad_len, DMA_TO_DEVICE);
+	dma_unmap_single(ce->dev, rctx->addr_res, rctx->result_len,
+			 DMA_FROM_DEVICE);
+	dma_unmap_sg(ce->dev, areq->src, rctx->nr_sgs, DMA_TO_DEVICE);
+}
+
+int sun8i_ce_hash_run(struct crypto_engine *engine, void *async_req)
+{
+	struct ahash_request *areq = ahash_request_cast(async_req);
+	struct crypto_ahash *tfm = crypto_ahash_reqtfm(areq);
+	struct sun8i_ce_hash_tfm_ctx *ctx = crypto_ahash_ctx(tfm);
+	struct sun8i_ce_hash_reqctx *rctx = ahash_request_ctx_dma(areq);
+	struct sun8i_ce_dev *ce = ctx->ce;
+	struct sun8i_ce_flow *chan;
+	int err;
+
+	chan = &ce->chanlist[rctx->flow];
+
+	err = sun8i_ce_hash_prepare(areq, chan->tl);
+	if (err)
+		return err;
+
+	err = sun8i_ce_run_task(ce, rctx->flow, crypto_ahash_alg_name(tfm));
+
+	sun8i_ce_hash_unprepare(areq, chan->tl);
+
+	if (!err)
+		memcpy(areq->result, rctx->result,
+		       crypto_ahash_digestsize(tfm));
+
 	local_bh_disable();
-	crypto_finalize_hash_request(engine, breq, err);
+	crypto_finalize_hash_request(engine, async_req, err);
 	local_bh_enable();
 
 	return 0;

drivers/crypto/allwinner/sun8i-ce/sun8i-ce-prng.c

···
 
 	cet->t_dst[0].addr = desc_addr_val_le32(ce, dma_dst);
 	cet->t_dst[0].len = cpu_to_le32(todo / 4);
-	ce->chanlist[flow].timeout = 2000;
 
 	err = sun8i_ce_run_task(ce, 3, "PRNG");
 	mutex_unlock(&ce->rnglock);

drivers/crypto/allwinner/sun8i-ce/sun8i-ce-trng.c

···
 
 	cet->t_dst[0].addr = desc_addr_val_le32(ce, dma_dst);
 	cet->t_dst[0].len = cpu_to_le32(todo / 4);
-	ce->chanlist[flow].timeout = todo;
 
 	err = sun8i_ce_run_task(ce, 3, "TRNG");
 	mutex_unlock(&ce->rnglock);

drivers/crypto/allwinner/sun8i-ce/sun8i-ce.h

···
 #define MAX_SG 8
 
 #define CE_MAX_CLOCKS 4
+#define CE_DMA_TIMEOUT_MS	3000
 
 #define MAXFLOW 4
+
+#define CE_MAX_HASH_DIGEST_SIZE		SHA512_DIGEST_SIZE
+#define CE_MAX_HASH_BLOCK_SIZE		SHA512_BLOCK_SIZE
 
 /*
  * struct ce_clock - Describe clocks used by sun8i-ce
···
  * @status:	set to 1 by interrupt if task is done
  * @t_phy:	Physical address of task
  * @tl:		pointer to the current ce_task for this flow
- * @backup_iv:		buffer which contain the next IV to store
- * @bounce_iv:		buffer which contain the IV
  * @stat_req:	number of request done by this flow
  */
 struct sun8i_ce_flow {
···
 	struct completion complete;
 	int status;
 	dma_addr_t t_phy;
-	int timeout;
 	struct ce_task *tl;
-	void *backup_iv;
-	void *bounce_iv;
 #ifdef CONFIG_CRYPTO_DEV_SUN8I_CE_DEBUG
 	unsigned long stat_req;
 #endif
···
  * @nr_sgd:		The number of destination SG (as given by dma_map_sg())
  * @addr_iv:		The IV addr returned by dma_map_single, need to unmap later
  * @addr_key:		The key addr returned by dma_map_single, need to unmap later
+ * @bounce_iv:		Current IV buffer
+ * @backup_iv:		Next IV buffer
  * @fallback_req:	request struct for invoking the fallback skcipher TFM
  */
 struct sun8i_cipher_req_ctx {
···
 	int nr_sgd;
 	dma_addr_t addr_iv;
 	dma_addr_t addr_key;
+	u8 bounce_iv[AES_BLOCK_SIZE] __aligned(sizeof(u32));
+	u8 backup_iv[AES_BLOCK_SIZE];
 	struct skcipher_request fallback_req;   // keep at the end
 };
 
···
  * struct sun8i_ce_hash_reqctx - context for an ahash request
  * @fallback_req:	pre-allocated fallback request
  * @flow:	the flow to use for this request
+ * @nr_sgs: number of entries in the source scatterlist
+ * @result_len: result length in bytes
+ * @pad_len: padding length in bytes
+ * @addr_res: DMA address of the result buffer, returned by dma_map_single()
+ * @addr_pad: DMA address of the padding buffer, returned by dma_map_single()
+ * @result: per-request result buffer
+ * @pad: per-request padding buffer (up to 2 blocks)
  */
 struct sun8i_ce_hash_reqctx {
 	int flow;
+	int nr_sgs;
+	size_t result_len;
+	size_t pad_len;
+	dma_addr_t addr_res;
+	dma_addr_t addr_pad;
+	u8 result[CE_MAX_HASH_DIGEST_SIZE] __aligned(CRYPTO_DMA_ALIGN);
+	u8 pad[2 * CE_MAX_HASH_BLOCK_SIZE];
 	struct ahash_request fallback_req; // keep at the end
 };
 

drivers/crypto/aspeed/aspeed-hace-crypto.c

···
 
 	} else {
 		dma_unmap_sg(hace_dev->dev, req->dst, rctx->dst_nents,
-			     DMA_TO_DEVICE);
+			     DMA_FROM_DEVICE);
 		dma_unmap_sg(hace_dev->dev, req->src, rctx->src_nents,
 			     DMA_TO_DEVICE);
 	}

drivers/crypto/atmel-tdes.c

···
 
 	if (err && (dd->flags & TDES_FLAGS_FAST)) {
 		dma_unmap_sg(dd->dev, dd->in_sg, 1, DMA_TO_DEVICE);
-		dma_unmap_sg(dd->dev, dd->out_sg, 1, DMA_TO_DEVICE);
+		dma_unmap_sg(dd->dev, dd->out_sg, 1, DMA_FROM_DEVICE);
 	}
 
 	return err;

drivers/crypto/caam/ctrl.c

···
 	int ret;
 
 	ctrlpriv->num_clks = data->num_clks;
-	ctrlpriv->clks = devm_kmemdup(dev, data->clks,
-				      data->num_clks * sizeof(data->clks[0]),
-				      GFP_KERNEL);
+	ctrlpriv->clks = devm_kmemdup_array(dev, data->clks,
+					    data->num_clks, sizeof(*data->clks),
+					    GFP_KERNEL);
 	if (!ctrlpriv->clks)
 		return -ENOMEM;
 
···
 			 */
 			if (needs_entropy_delay_adjustment())
 				ent_delay = 12000;
-			if (!(ctrlpriv->rng4_sh_init || inst_handles)) {
+			if (!inst_handles) {
 				dev_info(dev,
 					 "Entropy delay = %u\n",
 					 ent_delay);
 				kick_trng(dev, ent_delay);
-				ent_delay += 400;
+				ent_delay = ent_delay * 2;
 			}
 			/*
 			 * if instantiate_rng(...) fails, the loop will rerun

drivers/crypto/ccp/hsti.c

···
 	.is_visible = psp_security_is_visible,
 };
 
-static int psp_poulate_hsti(struct psp_device *psp)
+static int psp_populate_hsti(struct psp_device *psp)
 {
 	struct hsti_request *req;
 	int ret;
···
 		return 0;
 
 	/* Allocate command-response buffer */
-	req = kzalloc(sizeof(*req), GFP_KERNEL | __GFP_ZERO);
+	req = kzalloc(sizeof(*req), GFP_KERNEL);
 	if (!req)
 		return -ENOMEM;
 
-	req->header.payload_size = sizeof(req);
+	req->header.payload_size = sizeof(*req);
 
 	ret = psp_send_platform_access_msg(PSP_CMD_HSTI_QUERY, (struct psp_request *)req);
 	if (ret)
···
 	int ret;
 
 	if (PSP_FEATURE(psp, HSTI)) {
-		ret = psp_poulate_hsti(psp);
+		ret = psp_populate_hsti(psp);
 		if (ret)
 			return ret;
 	}

drivers/crypto/ccp/sev-dev.c

···
 	case SEV_CMD_SNP_GUEST_REQUEST:		return sizeof(struct sev_data_snp_guest_request);
 	case SEV_CMD_SNP_CONFIG:		return sizeof(struct sev_user_data_snp_config);
 	case SEV_CMD_SNP_COMMIT:		return sizeof(struct sev_data_snp_commit);
+	case SEV_CMD_SNP_FEATURE_INFO:		return sizeof(struct sev_data_snp_feature_info);
+	case SEV_CMD_SNP_VLEK_LOAD:		return sizeof(struct sev_user_data_snp_vlek_load);
 	default:				return 0;
 	}
 
···
 	struct sev_device *sev;
 	unsigned int cmdbuff_hi, cmdbuff_lo;
 	unsigned int phys_lsb, phys_msb;
-	unsigned int reg, ret = 0;
+	unsigned int reg;
 	void *cmd_buf;
 	int buf_len;
+	int ret = 0;
 
 	if (!psp || !psp->sev_data)
 		return -ENODEV;
···
 					 1 << entry->order, false);
 }
 
+bool sev_is_snp_ciphertext_hiding_supported(void)
+{
+	struct psp_device *psp = psp_master;
+	struct sev_device *sev;
+
+	if (!psp || !psp->sev_data)
+		return false;
+
+	sev = psp->sev_data;
+
+	/*
+	 * Feature information indicates if CipherTextHiding feature is
+	 * supported by the SEV firmware and additionally platform status
+	 * indicates if CipherTextHiding feature is enabled in the
+	 * Platform BIOS.
+	 */
+	return ((sev->snp_feat_info_0.ecx & SNP_CIPHER_TEXT_HIDING_SUPPORTED) &&
+		 sev->snp_plat_status.ciphertext_hiding_cap);
+}
+EXPORT_SYMBOL_GPL(sev_is_snp_ciphertext_hiding_supported);
+
+static int snp_get_platform_data(struct sev_device *sev, int *error)
+{
+	struct sev_data_snp_feature_info snp_feat_info;
+	struct snp_feature_info *feat_info;
+	struct sev_data_snp_addr buf;
+	struct page *page;
+	int rc;
+
+	/*
+	 * This function is expected to be called before SNP is
+	 * initialized.
+	 */
+	if (sev->snp_initialized)
+		return -EINVAL;
+
+	buf.address = __psp_pa(&sev->snp_plat_status);
+	rc = sev_do_cmd(SEV_CMD_SNP_PLATFORM_STATUS, &buf, error);
+	if (rc) {
+		dev_err(sev->dev, "SNP PLATFORM_STATUS command failed, ret = %d, error = %#x\n",
+			rc, *error);
+		return rc;
+	}
+
+	sev->api_major = sev->snp_plat_status.api_major;
+	sev->api_minor = sev->snp_plat_status.api_minor;
+	sev->build = sev->snp_plat_status.build_id;
+
+	/*
+	 * Do feature discovery of the currently loaded firmware,
+	 * and cache feature information from CPUID 0x8000_0024,
+	 * sub-function 0.
+	 */
+	if (!sev->snp_plat_status.feature_info)
+		return 0;
+
+	/*
+	 * Use dynamically allocated structure for the SNP_FEATURE_INFO
+	 * command to ensure structure is 8-byte aligned, and does not
+	 * cross a page boundary.
+	 */
+	page = alloc_page(GFP_KERNEL);
+	if (!page)
+		return -ENOMEM;
+
+	feat_info = page_address(page);
+	snp_feat_info.length = sizeof(snp_feat_info);
+	snp_feat_info.ecx_in = 0;
+	snp_feat_info.feature_info_paddr = __psp_pa(feat_info);
+
+	rc = sev_do_cmd(SEV_CMD_SNP_FEATURE_INFO, &snp_feat_info, error);
+	if (!rc)
+		sev->snp_feat_info_0 = *feat_info;
+	else
+		dev_err(sev->dev, "SNP FEATURE_INFO command failed, ret = %d, error = %#x\n",
+			rc, *error);
+
+	__free_page(page);
+
+	return rc;
+}
+
 static int snp_filter_reserved_mem_regions(struct resource *rs, void *arg)
 {
 	struct sev_data_range_list *range_list = arg;
···
 	return 0;
 }
 
-static int __sev_snp_init_locked(int *error)
+static int __sev_snp_init_locked(int *error, unsigned int max_snp_asid)
 {
 	struct psp_device *psp = psp_master;
 	struct sev_data_snp_init_ex data;
···
 		snp_add_hv_fixed_pages(sev, snp_range_list);
 
 		memset(&data, 0, sizeof(data));
+
+		if (max_snp_asid) {
+			data.ciphertext_hiding_en = 1;
+			data.max_snp_asid = max_snp_asid;
+		}
+
 		data.init_rmp = 1;
 		data.list_paddr_en = 1;
 		data.list_paddr = __psp_pa(snp_range_list);
···
 
 	sev = psp_master->sev_data;
 
-	if (sev->state == SEV_STATE_INIT)
+	if (sev->sev_plat_status.state == SEV_STATE_INIT)
 		return 0;
 
 	__sev_platform_init_handle_tmr(sev);
···
 		return rc;
 	}
 
-	sev->state = SEV_STATE_INIT;
+	sev->sev_plat_status.state = SEV_STATE_INIT;
 
 	/* Prepare for first SEV guest launch after INIT */
 	wbinvd_on_all_cpus();
···
 
 	sev = psp_master->sev_data;
 
-	if (sev->state == SEV_STATE_INIT)
+	if (sev->sev_plat_status.state == SEV_STATE_INIT)
 		return 0;
 
-	rc = __sev_snp_init_locked(&args->error);
+	rc = __sev_snp_init_locked(&args->error, args->max_snp_asid);
 	if (rc && rc != -ENODEV)
 		return rc;
 
···
 
 	sev = psp->sev_data;
 
-	if (sev->state == SEV_STATE_UNINIT)
+	if (sev->sev_plat_status.state == SEV_STATE_UNINIT)
 		return 0;
 
 	ret = __sev_do_cmd_locked(SEV_CMD_SHUTDOWN, NULL, error);
···
 		return ret;
 	}
 
-	sev->state = SEV_STATE_UNINIT;
+	sev->sev_plat_status.state = SEV_STATE_UNINIT;
 	dev_dbg(sev->dev, "SEV firmware shutdown\n");
 
 	return ret;
···
 {
 	int error, rc;
 
-	rc = __sev_snp_init_locked(&error);
+	rc = __sev_snp_init_locked(&error, 0);
 	if (rc) {
 		argp->error = SEV_RET_INVALID_PLATFORM_STATE;
 		return rc;
···
 	if (!writable)
 		return -EPERM;
 
-	if (sev->state == SEV_STATE_UNINIT) {
+	if (sev->sev_plat_status.state == SEV_STATE_UNINIT) {
 		rc = sev_move_to_init_state(argp, &shutdown_required);
 		if (rc)
 			return rc;
···
 	data.len = input.length;
 
 cmd:
-	if (sev->state == SEV_STATE_UNINIT) {
+	if (sev->sev_plat_status.state == SEV_STATE_UNINIT) {
 		ret = sev_move_to_init_state(argp, &shutdown_required);
 		if (ret)
 			goto e_free_blob;
···
 	struct sev_user_data_status status;
 	int error = 0, ret;
 
+	/*
+	 * Cache SNP platform status and SNP feature information
+	 * if SNP is available.
+	 */
+	if (cc_platform_has(CC_ATTR_HOST_SEV_SNP)) {
+		ret = snp_get_platform_data(sev, &error);
+		if (ret)
+			return 1;
+	}
+
 	ret = sev_platform_status(&status, &error);
 	if (ret) {
 		dev_err(sev->dev,
···
 		return 1;
 	}
 
+	/* Cache SEV platform status */
+	sev->sev_plat_status = status;
+
 	sev->api_major = status.api_major;
 	sev->api_minor = status.api_minor;
 	sev->build = status.build;
-	sev->state = status.state;
 
 	return 0;
 }
···
 	data.oca_cert_len = input.oca_cert_len;
 
 	/* If platform is not in INIT state then transition it to INIT */
-	if (sev->state != SEV_STATE_INIT) {
+	if (sev->sev_plat_status.state != SEV_STATE_INIT) {
 		ret = sev_move_to_init_state(argp, &shutdown_required);
 		if (ret)
 			goto e_free_oca;
···
 
 cmd:
 	/* If platform is not in INIT state then transition it to INIT. */
-	if (sev->state != SEV_STATE_INIT) {
+	if (sev->sev_plat_status.state != SEV_STATE_INIT) {
 		if (!writable) {
 			ret = -EPERM;
 			goto e_free_cert;

drivers/crypto/ccp/sev-dev.h

···
 
 	struct sev_vdata *vdata;
 
-	int state;
 	unsigned int int_rcvd;
 	wait_queue_head_t int_queue;
 	struct sev_misc_dev *misc;
···
 	bool cmd_buf_backup_active;
 
 	bool snp_initialized;
+
+	struct sev_user_data_status sev_plat_status;
+
+	struct sev_user_data_snp_status snp_plat_status;
+	struct snp_feature_info snp_feat_info_0;
 };
 
 int sev_dev_init(struct psp_device *psp);

drivers/crypto/chelsio/Kconfig

···
 	depends on CHELSIO_T4
 	select CRYPTO_LIB_AES
 	select CRYPTO_LIB_GF128MUL
-	select CRYPTO_SHA1
-	select CRYPTO_SHA256
-	select CRYPTO_SHA512
+	select CRYPTO_LIB_SHA1
+	select CRYPTO_LIB_SHA256
+	select CRYPTO_LIB_SHA512
 	select CRYPTO_AUTHENC
 	help
 	  The Chelsio Crypto Co-processor driver for T6 adapters.

drivers/crypto/chelsio/chcr_algo.c

···
 
 #include <crypto/aes.h>
 #include <crypto/algapi.h>
-#include <crypto/hash.h>
 #include <crypto/gcm.h>
 #include <crypto/sha1.h>
 #include <crypto/sha2.h>
···
 	}
 }
 
-static struct crypto_shash *chcr_alloc_shash(unsigned int ds)
+static int chcr_prepare_hmac_key(const u8 *raw_key, unsigned int raw_key_len,
+				 int digestsize, void *istate, void *ostate)
 {
-	struct crypto_shash *base_hash = ERR_PTR(-EINVAL);
+	__be32 *istate32 = istate, *ostate32 = ostate;
+	__be64 *istate64 = istate, *ostate64 = ostate;
+	union {
+		struct hmac_sha1_key sha1;
+		struct hmac_sha224_key sha224;
+		struct hmac_sha256_key sha256;
+		struct hmac_sha384_key sha384;
+		struct hmac_sha512_key sha512;
+	} k;
 
-	switch (ds) {
+	switch (digestsize) {
 	case SHA1_DIGEST_SIZE:
-		base_hash = crypto_alloc_shash("sha1", 0, 0);
+		hmac_sha1_preparekey(&k.sha1, raw_key, raw_key_len);
+		for (int i = 0; i < ARRAY_SIZE(k.sha1.istate.h); i++) {
+			istate32[i] = cpu_to_be32(k.sha1.istate.h[i]);
+			ostate32[i] = cpu_to_be32(k.sha1.ostate.h[i]);
+		}
 		break;
 	case SHA224_DIGEST_SIZE:
-		base_hash = crypto_alloc_shash("sha224", 0, 0);
+		hmac_sha224_preparekey(&k.sha224, raw_key, raw_key_len);
+		for (int i = 0; i < ARRAY_SIZE(k.sha224.key.istate.h); i++) {
+			istate32[i] = cpu_to_be32(k.sha224.key.istate.h[i]);
+			ostate32[i] = cpu_to_be32(k.sha224.key.ostate.h[i]);
+		}
 		break;
 	case SHA256_DIGEST_SIZE:
-		base_hash = crypto_alloc_shash("sha256", 0, 0);
+		hmac_sha256_preparekey(&k.sha256, raw_key, raw_key_len);
+		for (int i = 0; i < ARRAY_SIZE(k.sha256.key.istate.h); i++) {
+			istate32[i] = cpu_to_be32(k.sha256.key.istate.h[i]);
+			ostate32[i] = cpu_to_be32(k.sha256.key.ostate.h[i]);
+		}
 		break;
 	case SHA384_DIGEST_SIZE:
-		base_hash = crypto_alloc_shash("sha384", 0, 0);
+		hmac_sha384_preparekey(&k.sha384, raw_key, raw_key_len);
+		for (int i = 0; i < ARRAY_SIZE(k.sha384.key.istate.h); i++) {
+			istate64[i] = cpu_to_be64(k.sha384.key.istate.h[i]);
+			ostate64[i] = cpu_to_be64(k.sha384.key.ostate.h[i]);
+		}
 		break;
 	case SHA512_DIGEST_SIZE:
-		base_hash = crypto_alloc_shash("sha512", 0, 0);
+		hmac_sha512_preparekey(&k.sha512, raw_key, raw_key_len);
+		for (int i = 0; i < ARRAY_SIZE(k.sha512.key.istate.h); i++) {
+			istate64[i] = cpu_to_be64(k.sha512.key.istate.h[i]);
+			ostate64[i] = cpu_to_be64(k.sha512.key.ostate.h[i]);
+		}
 		break;
+	default:
+		return -EINVAL;
 	}
-
-	return base_hash;
-}
-
-static int chcr_compute_partial_hash(struct shash_desc *desc,
-				     char *iopad, char *result_hash,
-				     int digest_size)
-{
-	struct sha1_state sha1_st;
-	struct sha256_state sha256_st;
-	struct sha512_state sha512_st;
-	int error;
-
-	if (digest_size == SHA1_DIGEST_SIZE) {
-		error = crypto_shash_init(desc) ?:
-			crypto_shash_update(desc, iopad, SHA1_BLOCK_SIZE) ?:
-			crypto_shash_export_core(desc, &sha1_st);
-		memcpy(result_hash, sha1_st.state, SHA1_DIGEST_SIZE);
-	} else if (digest_size == SHA224_DIGEST_SIZE) {
-		error = crypto_shash_init(desc) ?:
-			crypto_shash_update(desc, iopad, SHA256_BLOCK_SIZE) ?:
-			crypto_shash_export_core(desc, &sha256_st);
-		memcpy(result_hash, sha256_st.state, SHA256_DIGEST_SIZE);
-
-	} else if (digest_size == SHA256_DIGEST_SIZE) {
-		error = crypto_shash_init(desc) ?:
-			crypto_shash_update(desc, iopad, SHA256_BLOCK_SIZE) ?:
-			crypto_shash_export_core(desc, &sha256_st);
-		memcpy(result_hash, sha256_st.state, SHA256_DIGEST_SIZE);
-
-	} else if (digest_size == SHA384_DIGEST_SIZE) {
-		error = crypto_shash_init(desc) ?:
-			crypto_shash_update(desc, iopad, SHA512_BLOCK_SIZE) ?:
-			crypto_shash_export_core(desc, &sha512_st);
-		memcpy(result_hash, sha512_st.state, SHA512_DIGEST_SIZE);
-
-	} else if (digest_size == SHA512_DIGEST_SIZE) {
-		error = crypto_shash_init(desc) ?:
-			crypto_shash_update(desc, iopad, SHA512_BLOCK_SIZE) ?:
-			crypto_shash_export_core(desc, &sha512_st);
-		memcpy(result_hash, sha512_st.state, SHA512_DIGEST_SIZE);
-	} else {
-		error = -EINVAL;
-		pr_err("Unknown digest size %d\n", digest_size);
-	}
-	return error;
-}
-
-static void chcr_change_order(char *buf, int ds)
-{
-	int i;
-
-	if (ds == SHA512_DIGEST_SIZE) {
-		for (i = 0; i < (ds / sizeof(u64)); i++)
-			*((__be64 *)buf + i) =
-				cpu_to_be64(*((u64 *)buf + i));
-	} else {
-		for (i = 0; i < (ds / sizeof(u32)); i++)
-			*((__be32 *)buf + i) =
-				cpu_to_be32(*((u32 *)buf + i));
-	}
+	memzero_explicit(&k, sizeof(k));
+	return 0;
 }
 
 static inline int is_hmac(struct crypto_tfm *tfm)
···
 	return 0;
 }
 
-static inline void chcr_free_shash(struct crypto_shash *base_hash)
-{
-		crypto_free_shash(base_hash);
-}
-
 /**
  *	create_hash_wr - Create hash work request
  *	@req: Cipher req base
···
 			     unsigned int keylen)
 {
 	struct hmac_ctx *hmacctx = HMAC_CTX(h_ctx(tfm));
-	unsigned int digestsize = crypto_ahash_digestsize(tfm);
-	unsigned int bs = crypto_tfm_alg_blocksize(crypto_ahash_tfm(tfm));
-	unsigned int i, err = 0, updated_digestsize;
-
-	SHASH_DESC_ON_STACK(shash, hmacctx->base_hash);
 
 	/* use the key to calculate the ipad and opad. ipad will sent with the
 	 * first request's data. opad will be sent with the final hash result
 	 * ipad in hmacctx->ipad and opad in hmacctx->opad location
 	 */
-	shash->tfm = hmacctx->base_hash;
-	if (keylen > bs) {
-		err = crypto_shash_digest(shash, key, keylen,
-					  hmacctx->ipad);
-		if (err)
-			goto out;
-		keylen = digestsize;
-	} else {
-		memcpy(hmacctx->ipad, key, keylen);
-	}
-	memset(hmacctx->ipad + keylen, 0, bs - keylen);
-	unsafe_memcpy(hmacctx->opad, hmacctx->ipad, bs,
-		      "fortified memcpy causes -Wrestrict warning");
-
-	for (i = 0; i < bs / sizeof(int); i++) {
-		*((unsigned int *)(&hmacctx->ipad) + i) ^= IPAD_DATA;
-		*((unsigned int *)(&hmacctx->opad) + i) ^= OPAD_DATA;
-	}
-
-	updated_digestsize = digestsize;
-	if (digestsize == SHA224_DIGEST_SIZE)
-		updated_digestsize = SHA256_DIGEST_SIZE;
-	else if (digestsize == SHA384_DIGEST_SIZE)
-		updated_digestsize = SHA512_DIGEST_SIZE;
-	err = chcr_compute_partial_hash(shash, hmacctx->ipad,
-					hmacctx->ipad, digestsize);
-	if (err)
-		goto out;
-	chcr_change_order(hmacctx->ipad, updated_digestsize);
-
-	err = chcr_compute_partial_hash(shash, hmacctx->opad,
-					hmacctx->opad, digestsize);
-	if (err)
-		goto out;
-	chcr_change_order(hmacctx->opad, updated_digestsize);
-out:
-	return err;
+	return chcr_prepare_hmac_key(key, keylen, crypto_ahash_digestsize(tfm),
+				     hmacctx->ipad, hmacctx->opad);
 }
 
 static int chcr_aes_xts_setkey(struct crypto_skcipher *cipher, const u8 *key,
···
 
 static int chcr_hmac_cra_init(struct crypto_tfm *tfm)
 {
-	struct chcr_context *ctx = crypto_tfm_ctx(tfm);
-	struct hmac_ctx *hmacctx = HMAC_CTX(ctx);
-	unsigned int digestsize =
-		crypto_ahash_digestsize(__crypto_ahash_cast(tfm));
-
 	crypto_ahash_set_reqsize(__crypto_ahash_cast(tfm),
 				 sizeof(struct chcr_ahash_req_ctx));
-	hmacctx->base_hash = chcr_alloc_shash(digestsize);
-	if (IS_ERR(hmacctx->base_hash))
-		return PTR_ERR(hmacctx->base_hash);
 	return chcr_device_init(crypto_tfm_ctx(tfm));
-}
-
-static void chcr_hmac_cra_exit(struct crypto_tfm *tfm)
-{
-	struct chcr_context *ctx = crypto_tfm_ctx(tfm);
-	struct hmac_ctx *hmacctx = HMAC_CTX(ctx);
-
-	if (hmacctx->base_hash) {
-		chcr_free_shash(hmacctx->base_hash);
-		hmacctx->base_hash = NULL;
-	}
 }
 
 inline void chcr_aead_common_exit(struct aead_request *req)
···
 	struct chcr_authenc_ctx *actx = AUTHENC_CTX(aeadctx);
 	/* it contains auth and cipher key both*/
 	struct crypto_authenc_keys keys;
-	unsigned int bs, subtype;
+	unsigned int subtype;
 	unsigned int max_authsize = crypto_aead_alg(authenc)->maxauthsize;
-	int err = 0, i, key_ctx_len = 0;
+	int err = 0, key_ctx_len = 0;
 	unsigned char ck_size = 0;
-	unsigned char pad[CHCR_HASH_MAX_BLOCK_SIZE_128] = { 0 };
-	struct crypto_shash *base_hash = ERR_PTR(-EINVAL);
 	struct algo_param param;
 	int align;
-	u8 *o_ptr = NULL;
 
 	crypto_aead_clear_flags(aeadctx->sw_cipher, CRYPTO_TFM_REQ_MASK);
 	crypto_aead_set_flags(aeadctx->sw_cipher, crypto_aead_get_flags(authenc)
···
 		get_aes_decrypt_key(actx->dec_rrkey, aeadctx->key,
 			    aeadctx->enckey_len << 3);
 	}
-	base_hash  = chcr_alloc_shash(max_authsize);
-	if (IS_ERR(base_hash)) {
-		pr_err("Base driver cannot be loaded\n");
+
+	align = KEYCTX_ALIGN_PAD(max_authsize);
+	err = chcr_prepare_hmac_key(keys.authkey, keys.authkeylen, max_authsize,
+				    actx->h_iopad,
+				    actx->h_iopad + param.result_size + align);
+	if (err)
 		goto out;
-	}
-	{
-		SHASH_DESC_ON_STACK(shash, base_hash);
 
-		shash->tfm = base_hash;
-		bs = crypto_shash_blocksize(base_hash);
-		align = KEYCTX_ALIGN_PAD(max_authsize);
-		o_ptr =  actx->h_iopad + param.result_size + align;
+	key_ctx_len = sizeof(struct _key_ctx) + roundup(keys.enckeylen, 16) +
+		      (param.result_size + align) * 2;
+	aeadctx->key_ctx_hdr = FILL_KEY_CTX_HDR(ck_size, param.mk_size, 0, 1,
+						key_ctx_len >> 4);
+	actx->auth_mode = param.auth_mode;
 
-		if (keys.authkeylen > bs) {
-			err = crypto_shash_digest(shash, keys.authkey,
-						  keys.authkeylen,
-						  o_ptr);
-			if (err) {
-				pr_err("Base driver cannot be loaded\n");
-				goto out;
-			}
-			keys.authkeylen = max_authsize;
-		} else
-			memcpy(o_ptr, keys.authkey, keys.authkeylen);
+	memzero_explicit(&keys, sizeof(keys));
+	return 0;
 
-		/* Compute the ipad-digest*/
-		memset(pad + keys.authkeylen, 0, bs - keys.authkeylen);
-		memcpy(pad, o_ptr, keys.authkeylen);
-		for (i = 0; i < bs >> 2; i++)
-			*((unsigned int *)pad + i) ^= IPAD_DATA;
-
-		if (chcr_compute_partial_hash(shash, pad, actx->h_iopad,
-					      max_authsize))
-			goto out;
-		/* Compute the opad-digest */
-		memset(pad + keys.authkeylen, 0, bs - keys.authkeylen);
-		memcpy(pad, o_ptr, keys.authkeylen);
-		for (i = 0; i < bs >> 2; i++)
-			*((unsigned int *)pad + i) ^= OPAD_DATA;
-
-		if (chcr_compute_partial_hash(shash, pad, o_ptr, max_authsize))
-			goto out;
-
-		/* convert the ipad and opad digest to network order */
-		chcr_change_order(actx->h_iopad, param.result_size);
-		chcr_change_order(o_ptr, param.result_size);
-		key_ctx_len = sizeof(struct _key_ctx) +
-			roundup(keys.enckeylen, 16) +
-			(param.result_size + align) * 2;
-		aeadctx->key_ctx_hdr = FILL_KEY_CTX_HDR(ck_size, param.mk_size,
-						0, 1, key_ctx_len >> 4);
-		actx->auth_mode = param.auth_mode;
-		chcr_free_shash(base_hash);
-
-		memzero_explicit(&keys, sizeof(keys));
-		return 0;
-	}
 out:
 	aeadctx->enckey_len = 0;
 	memzero_explicit(&keys, sizeof(keys));
-	if (!IS_ERR(base_hash))
-		chcr_free_shash(base_hash);
 	return -EINVAL;
 }
 
···
 
 			if (driver_algs[i].type == CRYPTO_ALG_TYPE_HMAC) {
 				a_hash->halg.base.cra_init = chcr_hmac_cra_init;
-				a_hash->halg.base.cra_exit = chcr_hmac_cra_exit;
 				a_hash->init = chcr_hmac_init;
 				a_hash->setkey = chcr_ahash_setkey;
 				a_hash->halg.base.cra_ctxsize = SZ_AHASH_H_CTX;

drivers/crypto/chelsio/chcr_crypto.h

···
 };
 
 struct hmac_ctx {
-	struct crypto_shash *base_hash;
 	u8 ipad[CHCR_HASH_MAX_BLOCK_SIZE_128];
 	u8 opad[CHCR_HASH_MAX_BLOCK_SIZE_128];
 };

drivers/crypto/hisilicon/debugfs.c

···
 		dfx_regs_uninit(qm, qm->debug.qm_diff_regs, ARRAY_SIZE(qm_diff_regs));
 		ret = PTR_ERR(qm->debug.acc_diff_regs);
 		qm->debug.acc_diff_regs = NULL;
+		qm->debug.qm_diff_regs = NULL;
 		return ret;
 	}
 

drivers/crypto/hisilicon/hpre/hpre_main.c

···
 #define HPRE_HAC_RAS_NFE_ENB		0x301414
 #define HPRE_HAC_RAS_FE_ENB		0x301418
 #define HPRE_HAC_INT_SET		0x301500
+#define HPRE_AXI_ERROR_MASK		GENMASK(21, 10)
 #define HPRE_RNG_TIMEOUT_NUM		0x301A34
 #define HPRE_CORE_INT_ENABLE		0
 #define HPRE_RDCHN_INI_ST		0x301a00
···
 #define HPRE_PREFETCH_ENABLE		(~(BIT(0) | BIT(30)))
 #define HPRE_PREFETCH_DISABLE		BIT(30)
 #define HPRE_SVA_DISABLE_READY		(BIT(4) | BIT(8))
+#define HPRE_SVA_PREFTCH_DFX4		0x301144
+#define HPRE_WAIT_SVA_READY		500000
+#define HPRE_READ_SVA_STATUS_TIMES	3
+#define HPRE_WAIT_US_MIN		10
+#define HPRE_WAIT_US_MAX		20
 
 /* clock gate */
 #define HPRE_CLKGATE_CTL		0x301a10
···
 	return NULL;
 }
 
+static int hpre_wait_sva_ready(struct hisi_qm *qm)
+{
+	u32 val, try_times = 0;
+	u8 count = 0;
+
+	/*
+	 * Read the register value every 10-20us. If the value is 0 for three
+	 * consecutive times, the SVA module is ready.
+	 */
+	do {
+		val = readl(qm->io_base + HPRE_SVA_PREFTCH_DFX4);
+		if (val)
+			count = 0;
+		else if (++count == HPRE_READ_SVA_STATUS_TIMES)
+			break;
+
+		usleep_range(HPRE_WAIT_US_MIN, HPRE_WAIT_US_MAX);
+	} while (++try_times < HPRE_WAIT_SVA_READY);
+
+	if (try_times == HPRE_WAIT_SVA_READY) {
+		pci_err(qm->pdev, "failed to wait sva prefetch ready\n");
+		return -ETIMEDOUT;
+	}
+
+	return 0;
+}
+
 static void hpre_config_pasid(struct hisi_qm *qm)
 {
 	u32 val1, val2;
···
 	writel(PEH_AXUSER_CFG_ENABLE, qm->io_base + QM_PEH_AXUSER_CFG_ENABLE);
 }
 
-static void hpre_open_sva_prefetch(struct hisi_qm *qm)
-{
-	u32 val;
-	int ret;
-
-	if (!test_bit(QM_SUPPORT_SVA_PREFETCH, &qm->caps))
-		return;
-
-	/* Enable prefetch */
-	val = readl_relaxed(qm->io_base + HPRE_PREFETCH_CFG);
-	val &= HPRE_PREFETCH_ENABLE;
-	writel(val, qm->io_base + HPRE_PREFETCH_CFG);
-
-	ret = readl_relaxed_poll_timeout(qm->io_base + HPRE_PREFETCH_CFG,
-					 val, !(val & HPRE_PREFETCH_DISABLE),
-					 HPRE_REG_RD_INTVRL_US,
-					 HPRE_REG_RD_TMOUT_US);
-	if (ret)
-		pci_err(qm->pdev, "failed to open sva prefetch\n");
-}
-
 static void hpre_close_sva_prefetch(struct hisi_qm *qm)
 {
 	u32 val;
···
 					 HPRE_REG_RD_TMOUT_US);
 	if (ret)
 		pci_err(qm->pdev, "failed to close sva prefetch\n");
+
+	(void)hpre_wait_sva_ready(qm);
+}
+
+static void hpre_open_sva_prefetch(struct hisi_qm *qm)
+{
+	u32 val;
+	int ret;
+
+	if (!test_bit(QM_SUPPORT_SVA_PREFETCH, &qm->caps))
+		return;
+
+	/* Enable prefetch */
+	val = readl_relaxed(qm->io_base + HPRE_PREFETCH_CFG);
+	val &= HPRE_PREFETCH_ENABLE;
+	writel(val, qm->io_base + HPRE_PREFETCH_CFG);
+
+	ret = readl_relaxed_poll_timeout(qm->io_base + HPRE_PREFETCH_CFG,
+					 val, !(val & HPRE_PREFETCH_DISABLE),
+					 HPRE_REG_RD_INTVRL_US,
+					 HPRE_REG_RD_TMOUT_US);
+	if (ret) {
+		pci_err(qm->pdev, "failed to open sva prefetch\n");
+		hpre_close_sva_prefetch(qm);
+		return;
+	}
+
+	ret = hpre_wait_sva_ready(qm);
+	if (ret)
+		hpre_close_sva_prefetch(qm);
 }
 
 static void hpre_enable_clock_gate(struct hisi_qm *qm)
···
 
 	/* Config data buffer pasid needed by Kunpeng 920 */
 	hpre_config_pasid(qm);
+	hpre_open_sva_prefetch(qm);
 
 	hpre_enable_clock_gate(qm);
 
···
 	val1 = readl(qm->io_base + HPRE_AM_OOO_SHUTDOWN_ENB);
 	if (enable) {
 		val1 |= HPRE_AM_OOO_SHUTDOWN_ENABLE;
-		val2 = hisi_qm_get_hw_info(qm, hpre_basic_info,
-					   HPRE_OOO_SHUTDOWN_MASK_CAP, qm->cap_ver);
+		val2 = qm->err_info.dev_err.shutdown_mask;
 	} else {
 		val1 &= ~HPRE_AM_OOO_SHUTDOWN_ENABLE;
 		val2 = 0x0;
···
 
 static void hpre_hw_error_disable(struct hisi_qm *qm)
 {
-	u32 ce, nfe;
-
-	ce = hisi_qm_get_hw_info(qm, hpre_basic_info, HPRE_CE_MASK_CAP, qm->cap_ver);
-	nfe = hisi_qm_get_hw_info(qm, hpre_basic_info, HPRE_NFE_MASK_CAP, qm->cap_ver);
+	struct hisi_qm_err_mask *dev_err = &qm->err_info.dev_err;
+	u32 err_mask = dev_err->ce | dev_err->nfe | dev_err->fe;
 
 	/* disable hpre hw error interrupts */
-	writel(ce | nfe | HPRE_HAC_RAS_FE_ENABLE, qm->io_base + HPRE_INT_MASK);
+	writel(err_mask, qm->io_base + HPRE_INT_MASK);
 	/* disable HPRE block master OOO when nfe occurs on Kunpeng930 */
 	hpre_master_ooo_ctrl(qm, false);
 }
 
 static void hpre_hw_error_enable(struct hisi_qm *qm)
 {
-	u32 ce, nfe, err_en;
-
-	ce = hisi_qm_get_hw_info(qm, hpre_basic_info, HPRE_CE_MASK_CAP, qm->cap_ver);
-	nfe = hisi_qm_get_hw_info(qm, hpre_basic_info, HPRE_NFE_MASK_CAP, qm->cap_ver);
+	struct hisi_qm_err_mask *dev_err = &qm->err_info.dev_err;
+	u32 err_mask = dev_err->ce | dev_err->nfe | dev_err->fe;
 
 	/* clear HPRE hw error source if having */
-	writel(ce | nfe | HPRE_HAC_RAS_FE_ENABLE, qm->io_base + HPRE_HAC_SOURCE_INT);
+	writel(err_mask, qm->io_base + HPRE_HAC_SOURCE_INT);
 
 	/* configure error type */
-	writel(ce, qm->io_base + HPRE_RAS_CE_ENB);
-	writel(nfe, qm->io_base + HPRE_RAS_NFE_ENB);
-	writel(HPRE_HAC_RAS_FE_ENABLE, qm->io_base + HPRE_RAS_FE_ENB);
+	writel(dev_err->ce, qm->io_base + HPRE_RAS_CE_ENB);
+	writel(dev_err->nfe, qm->io_base + HPRE_RAS_NFE_ENB);
+	writel(dev_err->fe, qm->io_base + HPRE_RAS_FE_ENB);
 
 	/* enable HPRE block master OOO when nfe occurs on Kunpeng930 */
 	hpre_master_ooo_ctrl(qm, true);
 
 	/* enable hpre hw error interrupts */
-	err_en = ce | nfe | HPRE_HAC_RAS_FE_ENABLE;
-	writel(~err_en, qm->io_base + HPRE_INT_MASK);
+	writel(~err_mask, qm->io_base + HPRE_INT_MASK);
 }
 
 static inline struct hisi_qm *hpre_file_to_qm(struct hpre_debugfs_file *file)
···
 	size_t i, size;
 
 	size = ARRAY_SIZE(hpre_cap_query_info);
-	hpre_cap = devm_kzalloc(dev, sizeof(*hpre_cap) * size, GFP_KERNEL);
+	hpre_cap = devm_kcalloc(dev, size, sizeof(*hpre_cap), GFP_KERNEL);
 	if (!hpre_cap)
 		return -ENOMEM;
 
···
 
 static void hpre_disable_error_report(struct hisi_qm *qm, u32 err_type)
 {
-	u32 nfe_mask;
+	u32 nfe_mask = qm->err_info.dev_err.nfe;
 
-	nfe_mask = hisi_qm_get_hw_info(qm, hpre_basic_info, HPRE_NFE_MASK_CAP, qm->cap_ver);
 	writel(nfe_mask & (~err_type), qm->io_base + HPRE_RAS_NFE_ENB);
+}
+
+static void hpre_enable_error_report(struct hisi_qm *qm)
+{
+	u32 nfe_mask = qm->err_info.dev_err.nfe;
+	u32 ce_mask = qm->err_info.dev_err.ce;
+
+	writel(nfe_mask, qm->io_base + HPRE_RAS_NFE_ENB);
+	writel(ce_mask, qm->io_base + HPRE_RAS_CE_ENB);
 }
 
 static void hpre_open_axi_master_ooo(struct hisi_qm *qm)
···
 
 	err_status = hpre_get_hw_err_status(qm);
 	if (err_status) {
-		if (err_status & qm->err_info.ecc_2bits_mask)
+		if (err_status & qm->err_info.dev_err.ecc_2bits_mask)
 			qm->err_status.is_dev_ecc_mbit = true;
 		hpre_log_hw_error(qm, err_status);
 
-		if (err_status & qm->err_info.dev_reset_mask) {
+		if (err_status & qm->err_info.dev_err.reset_mask) {
 			/* Disable the same error reporting until device is recovered. */
 			hpre_disable_error_report(qm, err_status);
 			return ACC_ERR_NEED_RESET;
 		}
 		hpre_clear_hw_err_status(qm, err_status);
+		/* Avoid firmware disable error report, re-enable. */
+		hpre_enable_error_report(qm);
 	}
 
 	return ACC_ERR_RECOVERED;
···
 	u32 err_status;
 
 	err_status = hpre_get_hw_err_status(qm);
-	if (err_status & qm->err_info.dev_shutdown_mask)
+	if (err_status & qm->err_info.dev_err.shutdown_mask)
 		return true;
 
 	return false;
 }
 
+static void hpre_disable_axi_error(struct hisi_qm *qm)
+{
+	struct hisi_qm_err_mask *dev_err = &qm->err_info.dev_err;
+	u32 err_mask = dev_err->ce | dev_err->nfe | dev_err->fe;
+	u32 val;
+
+	val = ~(err_mask & (~HPRE_AXI_ERROR_MASK));
+	writel(val, qm->io_base + HPRE_INT_MASK);
+
+	if (qm->ver > QM_HW_V2)
+		writel(dev_err->shutdown_mask & (~HPRE_AXI_ERROR_MASK),
+		       qm->io_base + HPRE_OOO_SHUTDOWN_SEL);
+}
+
+static void hpre_enable_axi_error(struct hisi_qm *qm)
+{
+	struct hisi_qm_err_mask *dev_err = &qm->err_info.dev_err;
+	u32 err_mask = dev_err->ce | dev_err->nfe | dev_err->fe;
+
+	/* clear axi error source */
+	writel(HPRE_AXI_ERROR_MASK, qm->io_base + HPRE_HAC_SOURCE_INT);
+
+	writel(~err_mask, qm->io_base + HPRE_INT_MASK);
+
+	if (qm->ver > QM_HW_V2)
+		writel(dev_err->shutdown_mask, qm->io_base + HPRE_OOO_SHUTDOWN_SEL);
+}
+
 static void hpre_err_info_init(struct hisi_qm *qm)
 {
 	struct hisi_qm_err_info *err_info = &qm->err_info;
+	struct hisi_qm_err_mask *qm_err = &err_info->qm_err;
+	struct hisi_qm_err_mask *dev_err = &err_info->dev_err;
 
-	err_info->fe = HPRE_HAC_RAS_FE_ENABLE;
-	err_info->ce = hisi_qm_get_hw_info(qm, hpre_basic_info, HPRE_QM_CE_MASK_CAP, qm->cap_ver);
-	err_info->nfe = hisi_qm_get_hw_info(qm, hpre_basic_info, HPRE_QM_NFE_MASK_CAP, qm->cap_ver);
-	err_info->ecc_2bits_mask = HPRE_CORE_ECC_2BIT_ERR | HPRE_OOO_ECC_2BIT_ERR;
-	err_info->dev_shutdown_mask = hisi_qm_get_hw_info(qm, hpre_basic_info,
-			HPRE_OOO_SHUTDOWN_MASK_CAP, qm->cap_ver);
-	err_info->qm_shutdown_mask = hisi_qm_get_hw_info(qm, hpre_basic_info,
-			HPRE_QM_OOO_SHUTDOWN_MASK_CAP, qm->cap_ver);
-	err_info->qm_reset_mask = hisi_qm_get_hw_info(qm, hpre_basic_info,
-			HPRE_QM_RESET_MASK_CAP, qm->cap_ver);
-	err_info->dev_reset_mask = hisi_qm_get_hw_info(qm, hpre_basic_info,
-			HPRE_RESET_MASK_CAP, qm->cap_ver);
+	qm_err->fe = HPRE_HAC_RAS_FE_ENABLE;
+	qm_err->ce = hisi_qm_get_hw_info(qm, hpre_basic_info, HPRE_QM_CE_MASK_CAP, qm->cap_ver);
+	qm_err->nfe = hisi_qm_get_hw_info(qm, hpre_basic_info, HPRE_QM_NFE_MASK_CAP, qm->cap_ver);
+	qm_err->shutdown_mask = hisi_qm_get_hw_info(qm, hpre_basic_info,
+						    HPRE_QM_OOO_SHUTDOWN_MASK_CAP, qm->cap_ver);
+	qm_err->reset_mask = hisi_qm_get_hw_info(qm, hpre_basic_info,
+						 HPRE_QM_RESET_MASK_CAP, qm->cap_ver);
+	qm_err->ecc_2bits_mask = QM_ECC_MBIT;
+
+	dev_err->fe = HPRE_HAC_RAS_FE_ENABLE;
+	dev_err->ce = hisi_qm_get_hw_info(qm, hpre_basic_info, HPRE_CE_MASK_CAP, qm->cap_ver);
+	dev_err->nfe = hisi_qm_get_hw_info(qm, hpre_basic_info, HPRE_NFE_MASK_CAP, qm->cap_ver);
+	dev_err->shutdown_mask = hisi_qm_get_hw_info(qm, hpre_basic_info,
+						     HPRE_OOO_SHUTDOWN_MASK_CAP, qm->cap_ver);
+	dev_err->reset_mask = hisi_qm_get_hw_info(qm, hpre_basic_info,
+						  HPRE_RESET_MASK_CAP, qm->cap_ver);
+	dev_err->ecc_2bits_mask = HPRE_CORE_ECC_2BIT_ERR | HPRE_OOO_ECC_2BIT_ERR;
+
 	err_info->msi_wr_port = HPRE_WR_MSI_PORT;
 	err_info->acpi_rst = "HRST";
 }
···
 	.err_info_init		= hpre_err_info_init,
 	.get_err_result		= hpre_get_err_result,
 	.dev_is_abnormal	= hpre_dev_is_abnormal,
+	.disable_axi_error	= hpre_disable_axi_error,
+	.enable_axi_error	= hpre_enable_axi_error,
 };
 
 static int hpre_pf_probe_init(struct hpre *hpre)
···
 	ret = hpre_set_user_domain_and_cache(qm);
 	if (ret)
 		return ret;
-
-	hpre_open_sva_prefetch(qm);
 
 	hisi_qm_dev_err_init(qm);
 	ret = hpre_show_last_regs_init(qm);

drivers/crypto/hisilicon/qm.c

···
 
 #define QM_SQ_TYPE_MASK			GENMASK(3, 0)
 #define QM_SQ_TAIL_IDX(sqc)		((le16_to_cpu((sqc).w11) >> 6) & 0x1)
+#define QM_SQC_DISABLE_QP		(1U << 6)
+#define QM_XQC_RANDOM_DATA		0xaaaa
 
 /* cqc shift */
 #define QM_CQ_HOP_NUM_SHIFT		0
···
 #define QM_RAS_CE_TIMES_PER_IRQ		1
 #define QM_OOO_SHUTDOWN_SEL		0x1040f8
 #define QM_AXI_RRESP_ERR		BIT(0)
-#define QM_ECC_MBIT			BIT(2)
 #define QM_DB_TIMEOUT			BIT(10)
 #define QM_OF_FIFO_OF			BIT(11)
+#define QM_RAS_AXI_ERROR		(BIT(0) | BIT(1) | BIT(12))
 
 #define QM_RESET_WAIT_TIMEOUT		400
 #define QM_PEH_VENDOR_ID		0x1000d8
···
 #define ACC_MASTER_TRANS_RETURN		0x300150
 #define ACC_MASTER_GLOBAL_CTRL		0x300000
 #define ACC_AM_CFG_PORT_WR_EN		0x30001c
-#define QM_RAS_NFE_MBIT_DISABLE		~QM_ECC_MBIT
 #define ACC_AM_ROB_ECC_INT_STS		0x300104
 #define ACC_ROB_ECC_ERR_MULTPL		BIT(1)
 #define QM_MSI_CAP_ENABLE		BIT(16)
···
 		return false;
 
 	err_status = qm_get_hw_error_status(pf_qm);
-	if (err_status & pf_qm->err_info.qm_shutdown_mask)
+	if (err_status & pf_qm->err_info.qm_err.shutdown_mask)
 		return true;
 
 	if (pf_qm->err_ini->dev_is_abnormal)
···
 
 static void qm_hw_error_cfg(struct hisi_qm *qm)
 {
-	struct hisi_qm_err_info *err_info = &qm->err_info;
+	struct hisi_qm_err_mask *qm_err = &qm->err_info.qm_err;
 
-	qm->error_mask = err_info->nfe | err_info->ce | err_info->fe;
+	qm->error_mask = qm_err->nfe | qm_err->ce | qm_err->fe;
 	/* clear QM hw residual error source */
 	writel(qm->error_mask, qm->io_base + QM_ABNORMAL_INT_SOURCE);
 
 	/* configure error type */
-	writel(err_info->ce, qm->io_base + QM_RAS_CE_ENABLE);
+	writel(qm_err->ce, qm->io_base + QM_RAS_CE_ENABLE);
 	writel(QM_RAS_CE_TIMES_PER_IRQ, qm->io_base + QM_RAS_CE_THRESHOLD);
-	writel(err_info->nfe, qm->io_base + QM_RAS_NFE_ENABLE);
-	writel(err_info->fe, qm->io_base + QM_RAS_FE_ENABLE);
+	writel(qm_err->nfe, qm->io_base + QM_RAS_NFE_ENABLE);
+	writel(qm_err->fe, qm->io_base + QM_RAS_FE_ENABLE);
 }
 
 static void qm_hw_error_init_v2(struct hisi_qm *qm)
···
 	qm_hw_error_cfg(qm);
 
 	/* enable close master ooo when hardware error happened */
-	writel(qm->err_info.qm_shutdown_mask, qm->io_base + QM_OOO_SHUTDOWN_SEL);
+	writel(qm->err_info.qm_err.shutdown_mask, qm->io_base + QM_OOO_SHUTDOWN_SEL);
 
 	irq_unmask = ~qm->error_mask;
 	irq_unmask &= readl(qm->io_base + QM_ABNORMAL_INT_MASK);
···
 
 static enum acc_err_result qm_hw_error_handle_v2(struct hisi_qm *qm)
 {
+	struct hisi_qm_err_mask *qm_err = &qm->err_info.qm_err;
 	u32 error_status;
 
 	error_status = qm_get_hw_error_status(qm);
···
 			qm->err_status.is_qm_ecc_mbit = true;
 
 		qm_log_hw_error(qm, error_status);
-		if (error_status & qm->err_info.qm_reset_mask) {
+		if (error_status & qm_err->reset_mask) {
 			/* Disable the same error reporting until device is recovered. */
-			writel(qm->err_info.nfe & (~error_status),
-			       qm->io_base + QM_RAS_NFE_ENABLE);
+			writel(qm_err->nfe & (~error_status), qm->io_base + QM_RAS_NFE_ENABLE);
 			return ACC_ERR_NEED_RESET;
 		}
 
 		/* Clear error source if not need reset. */
 		writel(error_status, qm->io_base + QM_ABNORMAL_INT_SOURCE);
-		writel(qm->err_info.nfe, qm->io_base + QM_RAS_NFE_ENABLE);
-		writel(qm->err_info.ce, qm->io_base + QM_RAS_CE_ENABLE);
+		writel(qm_err->nfe, qm->io_base + QM_RAS_NFE_ENABLE);
+		writel(qm_err->ce, qm->io_base + QM_RAS_CE_ENABLE);
 	}
 
 	return ACC_ERR_RECOVERED;
···
 	}
 }
 
+static void qm_uacce_api_ver_init(struct hisi_qm *qm)
+{
+	struct uacce_device *uacce = qm->uacce;
+
+	switch (qm->ver) {
+	case QM_HW_V1:
+		uacce->api_ver = HISI_QM_API_VER_BASE;
+		break;
+	case QM_HW_V2:
+		uacce->api_ver = HISI_QM_API_VER2_BASE;
+		break;
+	case QM_HW_V3:
+	case QM_HW_V4:
+		uacce->api_ver = HISI_QM_API_VER3_BASE;
+		break;
+	default:
+		uacce->api_ver = HISI_QM_API_VER5_BASE;
+		break;
+	}
+}
+
 static int qm_alloc_uacce(struct hisi_qm *qm)
 {
 	struct pci_dev *pdev = qm->pdev;
···
 	uacce->priv = qm;
 
 	if (qm->ver == QM_HW_V1)
-		uacce->api_ver = HISI_QM_API_VER_BASE;
-	else if (qm->ver == QM_HW_V2)
-		uacce->api_ver = HISI_QM_API_VER2_BASE;
-	else
-		uacce->api_ver = HISI_QM_API_VER3_BASE;
-
-	if (qm->ver == QM_HW_V1)
 		mmio_page_nr = QM_DOORBELL_PAGE_NR;
 	else if (!test_bit(QM_SUPPORT_DB_ISOLATION, &qm->caps))
 		mmio_page_nr = QM_DOORBELL_PAGE_NR +
···
 	uacce->qf_pg_num[UACCE_QFRT_DUS]  = dus_page_nr;
 
 	qm->uacce = uacce;
+	qm_uacce_api_ver_init(qm);
 	INIT_LIST_HEAD(&qm->isolate_data.qm_hw_errs);
 	mutex_init(&qm->isolate_data.isolate_lock);
 
···
 
 	qm_init_eq_aeq_status(qm);
 
+	/* Before starting the dev, clear the memory and then configure to device using. */
+	memset(qm->qdma.va, 0, qm->qdma.size);
+
 	ret = qm_eq_ctx_cfg(qm);
 	if (ret) {
 		dev_err(dev, "Set eqc failed!\n");
···
 
 static int __hisi_qm_start(struct hisi_qm *qm)
 {
+	struct device *dev = &qm->pdev->dev;
 	int ret;
 
-	WARN_ON(!qm->qdma.va);
+	if (!qm->qdma.va) {
+		dev_err(dev, "qm qdma is NULL!\n");
+		return -EINVAL;
+	}
 
 	if (qm->fun_type == QM_HW_PF) {
 		ret = hisi_qm_set_vft(qm, 0, qm->qp_base, qm->qp_num);
···
 	for (i = 0; i < qm->qp_num; i++) {
 		qp = &qm->qp_array[i];
 		if (atomic_read(&qp->qp_status.flags) == QP_STOP &&
-		    qp->is_resetting == true) {
+		    qp->is_resetting == true && qp->is_in_kernel == true) {
 			ret = qm_start_qp_nolock(qp, 0);
 			if (ret < 0) {
 				dev_err(dev, "Failed to start qp%d!\n", i);
···
 }
 
 /**
- * qm_clear_queues() - Clear all queues memory in a qm.
- * @qm: The qm in which the queues will be cleared.
+ * qm_invalid_queues() - invalid all queues in use.
+ * @qm: The qm in which the queues will be invalidated.
  *
- * This function clears all queues memory in a qm. Reset of accelerator can
- * use this to clear queues.
+ * This function invalid all queues in use. If the doorbell command is sent
+ * to device in user space after the device is reset, the device discards
+ * the doorbell command.
  */
-static void qm_clear_queues(struct hisi_qm *qm)
+static void qm_invalid_queues(struct hisi_qm *qm)
 {
 	struct hisi_qp *qp;
+	struct qm_sqc *sqc;
+	struct qm_cqc *cqc;
 	int i;
+
+	/*
+	 * Normal stop queues is no longer used and does not need to be
+	 * invalid queues.
+	 */
+	if (qm->status.stop_reason == QM_NORMAL)
+		return;
+
+	if (qm->status.stop_reason == QM_DOWN)
+		hisi_qm_cache_wb(qm);
 
 	for (i = 0; i < qm->qp_num; i++) {
 		qp = &qm->qp_array[i];
-		if (qp->is_in_kernel && qp->is_resetting)
+		if (!qp->is_resetting)
+			continue;
+
+		/* Modify random data and set sqc close bit to invalid queue. */
+		sqc = qm->sqc + i;
+		cqc = qm->cqc + i;
+		sqc->w8 = cpu_to_le16(QM_XQC_RANDOM_DATA);
+		sqc->w13 = cpu_to_le16(QM_SQC_DISABLE_QP);
+		cqc->w8 = cpu_to_le16(QM_XQC_RANDOM_DATA);
+		if (qp->is_in_kernel)
 			memset(qp->qdma.va, 0, qp->qdma.size);
 	}
-
-	memset(qm->qdma.va, 0, qm->qdma.size);
 }
 
 /**
···
 		}
 	}
 
-	qm_clear_queues(qm);
+	qm_invalid_queues(qm);
 	qm->status.stop_reason = QM_NORMAL;
 
 err_unlock:
···
 	return 0;
 }
 
-static int qm_clear_vft_config(struct hisi_qm *qm)
+static void qm_clear_vft_config(struct hisi_qm *qm)
 {
-	int ret;
 	u32 i;
 
-	for (i = 1; i <= qm->vfs_num; i++) {
-		ret = hisi_qm_set_vft(qm, i, 0, 0);
-		if (ret)
-			return ret;
-	}
-	qm->vfs_num = 0;
+	/*
+	 * When disabling SR-IOV, clear the configuration of each VF in the hardware
+	 * sequentially. Failure to clear a single VF should not affect the clearing
+	 * operation of other VFs.
+	 */
+	for (i = 1; i <= qm->vfs_num; i++)
+		(void)hisi_qm_set_vft(qm, i, 0, 0);
 
-	return 0;
+	qm->vfs_num = 0;
 }
 
 static int qm_func_shaper_enable(struct hisi_qm *qm, u32 fun_index, u32 qos)
···
 	}
 
 	pdev = container_of(dev, struct pci_dev, dev);
+	if (pci_physfn(pdev) != qm->pdev) {
+		pci_err(qm->pdev, "the pdev input does not match the pf!\n");
+		return -EINVAL;
+	}
 
 	*fun_index = pdev->devfn;
 
···
 		goto err_put_sync;
 	}
 
+	qm->vfs_num = num_vfs;
 	ret = pci_enable_sriov(pdev, num_vfs);
 	if (ret) {
 		pci_err(pdev, "Can't enable VF!\n");
 		qm_clear_vft_config(qm);
 		goto err_put_sync;
 	}
-	qm->vfs_num = num_vfs;
 
 	pci_info(pdev, "VF enabled, vfs_num(=%d)!\n", num_vfs);
 
···
 	}
 
 	pci_disable_sriov(pdev);
-
-	qm->vfs_num = 0;
+	qm_clear_vft_config(qm);
 	qm_pm_put_sync(qm);
 
-	return qm_clear_vft_config(qm);
+	return 0;
 }
 EXPORT_SYMBOL_GPL(hisi_qm_sriov_disable);
 
···
 		   !qm->err_status.is_qm_ecc_mbit &&
 		   !qm->err_ini->close_axi_master_ooo) {
 		nfe_enb = readl(qm->io_base + QM_RAS_NFE_ENABLE);
-		writel(nfe_enb & QM_RAS_NFE_MBIT_DISABLE,
+		writel(nfe_enb & ~qm->err_info.qm_err.ecc_2bits_mask,
 		       qm->io_base + QM_RAS_NFE_ENABLE);
-		writel(QM_ECC_MBIT, qm->io_base + QM_ABNORMAL_INT_SET);
+		writel(qm->err_info.qm_err.ecc_2bits_mask, qm->io_base + QM_ABNORMAL_INT_SET);
 	}
 }
 
···
 {
 	u32 value;
 
-	if (qm->err_ini->open_sva_prefetch)
-		qm->err_ini->open_sva_prefetch(qm);
-
 	if (qm->ver >= QM_HW_V3)
 		return;
 
···
 	       qm->io_base + ACC_AM_CFG_PORT_WR_EN);
 
 	/* clear dev ecc 2bit error source if having */
-	value = qm_get_dev_err_status(qm) & qm->err_info.ecc_2bits_mask;
+	value = qm_get_dev_err_status(qm) & qm->err_info.dev_err.ecc_2bits_mask;
 	if (value && qm->err_ini->clear_dev_hw_err_status)
 		qm->err_ini->clear_dev_hw_err_status(qm, value);
 
 	/* clear QM ecc mbit error source */
-	writel(QM_ECC_MBIT, qm->io_base + QM_ABNORMAL_INT_SOURCE);
+	writel(qm->err_info.qm_err.ecc_2bits_mask, qm->io_base + QM_ABNORMAL_INT_SOURCE);
 
 	/* clear AM Reorder Buffer ecc mbit source */
 	writel(ACC_ROB_ECC_ERR_MULTPL, qm->io_base + ACC_AM_ROB_ECC_INT_STS);
···
 clear_flags:
 	qm->err_status.is_qm_ecc_mbit = false;
 	qm->err_status.is_dev_ecc_mbit = false;
+}
+
+static void qm_disable_axi_error(struct hisi_qm *qm)
+{
+	struct hisi_qm_err_mask *qm_err = &qm->err_info.qm_err;
+	u32 val;
+
+	val = ~(qm->error_mask & (~QM_RAS_AXI_ERROR));
+	writel(val, qm->io_base + QM_ABNORMAL_INT_MASK);
+	if (qm->ver > QM_HW_V2)
+		writel(qm_err->shutdown_mask & (~QM_RAS_AXI_ERROR),
+		       qm->io_base + QM_OOO_SHUTDOWN_SEL);
+
+	if (qm->err_ini->disable_axi_error)
+		qm->err_ini->disable_axi_error(qm);
+}
+
+static void qm_enable_axi_error(struct hisi_qm *qm)
+{
+	/* clear axi error source */
+	writel(QM_RAS_AXI_ERROR, qm->io_base + QM_ABNORMAL_INT_SOURCE);
+
+	writel(~qm->error_mask, qm->io_base + QM_ABNORMAL_INT_MASK);
+	if (qm->ver > QM_HW_V2)
+		writel(qm->err_info.qm_err.shutdown_mask, qm->io_base + QM_OOO_SHUTDOWN_SEL);
+
+	if (qm->err_ini->enable_axi_error)
+		qm->err_ini->enable_axi_error(qm);
 }
 
 static int qm_controller_reset_done(struct hisi_qm *qm)
···
 
 	qm_restart_prepare(qm);
 	hisi_qm_dev_err_init(qm);
+	qm_disable_axi_error(qm);
 	if (qm->err_ini->open_axi_master_ooo)
 		qm->err_ini->open_axi_master_ooo(qm);
 
···
 	ret = qm_wait_vf_prepare_finish(qm);
 	if (ret)
 		pci_err(pdev, "failed to start by vfs in soft reset!\n");
-
+	qm_enable_axi_error(qm);
 	qm_cmd_init(qm);
 	qm_restart_done(qm);
 
···
 }
 EXPORT_SYMBOL_GPL(hisi_qm_reset_done);
 
+static irqreturn_t qm_rsvd_irq(int irq, void *data)
+{
+	struct hisi_qm *qm = data;
+
+	dev_info(&qm->pdev->dev, "Reserved interrupt, ignore!\n");
+
+	return IRQ_HANDLED;
+}
+
 static irqreturn_t qm_abnormal_irq(int irq, void *data)
 {
 	struct hisi_qm *qm = data;
···
 	ret = hisi_qm_stop(qm, QM_DOWN);
 	if (ret)
 		dev_err(&pdev->dev, "Fail to stop qm in shutdown!\n");
-
-	hisi_qm_cache_wb(qm);
 }
 EXPORT_SYMBOL_GPL(hisi_qm_dev_shutdown);
 
···
 	struct pci_dev *pdev = qm->pdev;
 	u32 irq_vector, val;
 
-	if (qm->fun_type == QM_HW_VF)
+	if (qm->fun_type == QM_HW_VF && qm->ver < QM_HW_V3)
 		return;
 
 	val = qm->cap_tables.qm_cap_table[QM_ABNORMAL_IRQ].cap_val;
···
 	u32 irq_vector, val;
 	int ret;
 
-	if (qm->fun_type == QM_HW_VF)
-		return 0;
-
 	val = qm->cap_tables.qm_cap_table[QM_ABNORMAL_IRQ].cap_val;
 	if (!((val >> QM_IRQ_TYPE_SHIFT) & QM_ABN_IRQ_TYPE_MASK))
 		return 0;
-
 	irq_vector = val & QM_IRQ_VECTOR_MASK;
+
+	/* For VF, this is a reserved interrupt in V3 version. */
+	if (qm->fun_type == QM_HW_VF) {
+		if (qm->ver < QM_HW_V3)
+			return 0;
+
+		ret = request_irq(pci_irq_vector(pdev, irq_vector), qm_rsvd_irq,
+				  IRQF_NO_AUTOEN, qm->dev_name, qm);
+		if (ret) {
+			dev_err(&pdev->dev, "failed to request reserved irq, ret = %d!\n", ret);
+			return ret;
+		}
+		return 0;
+	}
+
 	ret = request_irq(pci_irq_vector(pdev, irq_vector), qm_abnormal_irq, 0, qm->dev_name, qm);
 	if (ret)
-		dev_err(&qm->pdev->dev, "failed to request abnormal irq, ret = %d", ret);
+		dev_err(&qm->pdev->dev, "failed to request abnormal irq, ret = %d!\n", ret);
 
 	return ret;
 }
···
 	pci_set_master(pdev);
 
 	num_vec = qm_get_irq_num(qm);
+	if (!num_vec) {
+		dev_err(dev, "Device irq num is zero!\n");
+		ret = -EINVAL;
+		goto err_get_pci_res;
+	}
+	num_vec = roundup_pow_of_two(num_vec);
 	ret = pci_alloc_irq_vectors(pdev, num_vec, num_vec, PCI_IRQ_MSI);
 	if (ret < 0) {
 		dev_err(dev, "Failed to enable MSI vectors!\n");

drivers/crypto/hisilicon/sec/sec_drv.c

···
 	struct iommu_domain *domain;
 	u32 sec_ipv4_mask = 0;
 	u32 sec_ipv6_mask[10] = {};
-	u32 i, ret;
+	int ret;
+	u32 i;
 
 	domain = iommu_get_domain_for_dev(info->dev);
 

drivers/crypto/hisilicon/sec2/sec_crypto.c

···
 static int sec_request_init(struct sec_ctx *ctx, struct sec_req *req)
 {
 	struct sec_qp_ctx *qp_ctx;
-	int i;
+	int i = 0;
 
-	for (i = 0; i < ctx->sec->ctx_q_num; i++) {
+	do {
 		qp_ctx = &ctx->qp_ctx[i];
 		req->req_id = sec_alloc_req_id(req, qp_ctx);
-		if (req->req_id >= 0)
-			break;
-	}
+	} while (req->req_id < 0 && ++i < ctx->sec->ctx_q_num);
 
 	req->qp_ctx = qp_ctx;
 	req->backlog = &qp_ctx->backlog;

drivers/crypto/hisilicon/sec2/sec_main.c

···
 #define SEC_RAS_FE_ENB_MSK		0x0
 #define SEC_OOO_SHUTDOWN_SEL		0x301014
 #define SEC_RAS_DISABLE		0x0
+#define SEC_AXI_ERROR_MASK		(BIT(0) | BIT(1))
+
 #define SEC_MEM_START_INIT_REG	0x301100
 #define SEC_MEM_INIT_DONE_REG		0x301104
 
···
 #define SEC_PREFETCH_ENABLE		(~(BIT(0) | BIT(1) | BIT(11)))
 #define SEC_PREFETCH_DISABLE		BIT(1)
 #define SEC_SVA_DISABLE_READY		(BIT(7) | BIT(11))
+#define SEC_SVA_PREFETCH_INFO		0x301ED4
+#define SEC_SVA_STALL_NUM		GENMASK(23, 8)
+#define SEC_SVA_PREFETCH_NUM		GENMASK(2, 0)
+#define SEC_WAIT_SVA_READY		500000
+#define SEC_READ_SVA_STATUS_TIMES	3
+#define SEC_WAIT_US_MIN			10
+#define SEC_WAIT_US_MAX			20
+#define SEC_WAIT_QP_US_MIN		1000
+#define SEC_WAIT_QP_US_MAX		2000
+#define SEC_MAX_WAIT_TIMES		2000
 
 #define SEC_DELAY_10_US			10
 #define SEC_POLL_TIMEOUT_US		1000
···
 	writel_relaxed(reg, qm->io_base + SEC_CONTROL_REG);
 }
 
+static int sec_wait_sva_ready(struct hisi_qm *qm, __u32 offset, __u32 mask)
+{
+	u32 val, try_times = 0;
+	u8 count = 0;
+
+	/*
+	 * Read the register value every 10-20us. If the value is 0 for three
+	 * consecutive times, the SVA module is ready.
+	 */
+	do {
+		val = readl(qm->io_base + offset);
+		if (val & mask)
+			count = 0;
+		else if (++count == SEC_READ_SVA_STATUS_TIMES)
+			break;
+
+		usleep_range(SEC_WAIT_US_MIN, SEC_WAIT_US_MAX);
+	} while (++try_times < SEC_WAIT_SVA_READY);
+
+	if (try_times == SEC_WAIT_SVA_READY) {
+		pci_err(qm->pdev, "failed to wait sva prefetch ready\n");
+		return -ETIMEDOUT;
+	}
+
+	return 0;
+}
+
+static void sec_close_sva_prefetch(struct hisi_qm *qm)
+{
+	u32 val;
+	int ret;
+
+	if (!test_bit(QM_SUPPORT_SVA_PREFETCH, &qm->caps))
+		return;
+
+	val = readl_relaxed(qm->io_base + SEC_PREFETCH_CFG);
+	val |= SEC_PREFETCH_DISABLE;
+	writel(val, qm->io_base + SEC_PREFETCH_CFG);
+
+	ret = readl_relaxed_poll_timeout(qm->io_base + SEC_SVA_TRANS,
+					 val, !(val & SEC_SVA_DISABLE_READY),
+					 SEC_DELAY_10_US, SEC_POLL_TIMEOUT_US);
+	if (ret)
+		pci_err(qm->pdev, "failed to close sva prefetch\n");
+
+	(void)sec_wait_sva_ready(qm, SEC_SVA_PREFETCH_INFO, SEC_SVA_STALL_NUM);
+}
+
+static void sec_open_sva_prefetch(struct hisi_qm *qm)
+{
+	u32 val;
+	int ret;
+
+	if (!test_bit(QM_SUPPORT_SVA_PREFETCH, &qm->caps))
+		return;
+
+	/* Enable prefetch */
+	val = readl_relaxed(qm->io_base + SEC_PREFETCH_CFG);
+	val &= SEC_PREFETCH_ENABLE;
+	writel(val, qm->io_base + SEC_PREFETCH_CFG);
+
+	ret = readl_relaxed_poll_timeout(qm->io_base + SEC_PREFETCH_CFG,
+					 val, !(val & SEC_PREFETCH_DISABLE),
+					 SEC_DELAY_10_US, SEC_POLL_TIMEOUT_US);
+	if (ret) {
+		pci_err(qm->pdev, "failed to open sva prefetch\n");
+		sec_close_sva_prefetch(qm);
+		return;
+	}
+
+	ret = sec_wait_sva_ready(qm, SEC_SVA_TRANS, SEC_SVA_PREFETCH_NUM);
+	if (ret)
+		sec_close_sva_prefetch(qm);
+}
+
 static void sec_engine_sva_config(struct hisi_qm *qm)
 {
 	u32 reg;
···
 		writel_relaxed(reg, qm->io_base +
 				SEC_INTERFACE_USER_CTRL1_REG);
 	}
-}
-
-static void sec_open_sva_prefetch(struct hisi_qm *qm)
-{
-	u32 val;
-	int ret;
-
-	if (!test_bit(QM_SUPPORT_SVA_PREFETCH, &qm->caps))
-		return;
-
-	/* Enable prefetch */
-	val = readl_relaxed(qm->io_base + SEC_PREFETCH_CFG);
-	val &= SEC_PREFETCH_ENABLE;
-	writel(val, qm->io_base + SEC_PREFETCH_CFG);
-
-	ret = readl_relaxed_poll_timeout(qm->io_base + SEC_PREFETCH_CFG,
-					 val, !(val & SEC_PREFETCH_DISABLE),
-					 SEC_DELAY_10_US, SEC_POLL_TIMEOUT_US);
-	if (ret)
-		pci_err(qm->pdev, "failed to open sva prefetch\n");
-}
-
-static void sec_close_sva_prefetch(struct hisi_qm *qm)
-{
-	u32 val;
-	int ret;
-
-	if (!test_bit(QM_SUPPORT_SVA_PREFETCH, &qm->caps))
-		return;
-
-	val = readl_relaxed(qm->io_base + SEC_PREFETCH_CFG);
-	val |= SEC_PREFETCH_DISABLE;
-	writel(val, qm->io_base + SEC_PREFETCH_CFG);
-
-	ret = readl_relaxed_poll_timeout(qm->io_base + SEC_SVA_TRANS,
-					 val, !(val & SEC_SVA_DISABLE_READY),
-					 SEC_DELAY_10_US, SEC_POLL_TIMEOUT_US);
-	if (ret)
-		pci_err(qm->pdev, "failed to close sva prefetch\n");
+	sec_open_sva_prefetch(qm);
 }
 
 static void sec_enable_clock_gate(struct hisi_qm *qm)
···
 	val1 = readl(qm->io_base + SEC_CONTROL_REG);
 	if (enable) {
 		val1 |= SEC_AXI_SHUTDOWN_ENABLE;
-		val2 = hisi_qm_get_hw_info(qm, sec_basic_info,
-					   SEC_OOO_SHUTDOWN_MASK_CAP, qm->cap_ver);
+		val2 = qm->err_info.dev_err.shutdown_mask;
 	} else {
 		val1 &= SEC_AXI_SHUTDOWN_DISABLE;
 		val2 = 0x0;
···
 
 static void sec_hw_error_enable(struct hisi_qm *qm)
 {
-	u32 ce, nfe;
+	struct hisi_qm_err_mask *dev_err = &qm->err_info.dev_err;
+	u32 err_mask = dev_err->ce | dev_err->nfe | dev_err->fe;
 
 	if (qm->ver == QM_HW_V1) {
 		writel(SEC_CORE_INT_DISABLE, qm->io_base + SEC_CORE_INT_MASK);
···
 		return;
 	}
 
-	ce = hisi_qm_get_hw_info(qm, sec_basic_info, SEC_CE_MASK_CAP, qm->cap_ver);
-	nfe = hisi_qm_get_hw_info(qm, sec_basic_info, SEC_NFE_MASK_CAP, qm->cap_ver);
-
 	/* clear SEC hw error source if having */
-	writel(ce | nfe | SEC_RAS_FE_ENB_MSK, qm->io_base + SEC_CORE_INT_SOURCE);
+	writel(err_mask, qm->io_base + SEC_CORE_INT_SOURCE);
 
 	/* enable RAS int */
-	writel(ce, qm->io_base + SEC_RAS_CE_REG);
-	writel(SEC_RAS_FE_ENB_MSK, qm->io_base + SEC_RAS_FE_REG);
-	writel(nfe, qm->io_base + SEC_RAS_NFE_REG);
+	writel(dev_err->ce, qm->io_base + SEC_RAS_CE_REG);
+	writel(dev_err->fe, qm->io_base + SEC_RAS_FE_REG);
+	writel(dev_err->nfe, qm->io_base + SEC_RAS_NFE_REG);
 
 	/* enable SEC block master OOO when nfe occurs on Kunpeng930 */
 	sec_master_ooo_ctrl(qm, true);
 
 	/* enable SEC hw error interrupts */
-	writel(ce | nfe | SEC_RAS_FE_ENB_MSK, qm->io_base + SEC_CORE_INT_MASK);
+	writel(err_mask, qm->io_base + SEC_CORE_INT_MASK);
 }
 
 static void sec_hw_error_disable(struct hisi_qm *qm)
···
 
 static void sec_disable_error_report(struct hisi_qm *qm, u32 err_type)
 {
-	u32 nfe_mask;
+	u32 nfe_mask = qm->err_info.dev_err.nfe;
 
-	nfe_mask = hisi_qm_get_hw_info(qm, sec_basic_info, SEC_NFE_MASK_CAP, qm->cap_ver);
 	writel(nfe_mask & (~err_type), qm->io_base + SEC_RAS_NFE_REG);
+}
+
+static void sec_enable_error_report(struct hisi_qm *qm)
+{
+	u32 nfe_mask = qm->err_info.dev_err.nfe;
+	u32 ce_mask = qm->err_info.dev_err.ce;
+
+	writel(nfe_mask, qm->io_base + SEC_RAS_NFE_REG);
+	writel(ce_mask, qm->io_base + SEC_RAS_CE_REG);
 }
 
 static void sec_open_axi_master_ooo(struct hisi_qm *qm)
···
 
 	err_status = sec_get_hw_err_status(qm);
 	if (err_status) {
-		if (err_status & qm->err_info.ecc_2bits_mask)
+		if (err_status & qm->err_info.dev_err.ecc_2bits_mask)
 			qm->err_status.is_dev_ecc_mbit = true;
 		sec_log_hw_error(qm, err_status);
 
-		if (err_status & qm->err_info.dev_reset_mask) {
+		if (err_status & qm->err_info.dev_err.reset_mask) {
 			/* Disable the same error reporting until device is recovered. */
 			sec_disable_error_report(qm, err_status);
 			return ACC_ERR_NEED_RESET;
 		}
 		sec_clear_hw_err_status(qm, err_status);
+		/* Avoid firmware disable error report, re-enable. */
+		sec_enable_error_report(qm);
 	}
 
 	return ACC_ERR_RECOVERED;
···
 	u32 err_status;
 
 	err_status = sec_get_hw_err_status(qm);
-	if (err_status & qm->err_info.dev_shutdown_mask)
+	if (err_status & qm->err_info.dev_err.shutdown_mask)
 		return true;
 
 	return false;
 }
 
+static void sec_disable_axi_error(struct hisi_qm *qm)
+{
+	struct hisi_qm_err_mask *dev_err = &qm->err_info.dev_err;
+	u32 err_mask = dev_err->ce | dev_err->nfe | dev_err->fe;
+
+	writel(err_mask & ~SEC_AXI_ERROR_MASK, qm->io_base + SEC_CORE_INT_MASK);
+
+	if (qm->ver > QM_HW_V2)
+		writel(dev_err->shutdown_mask & (~SEC_AXI_ERROR_MASK),
+		       qm->io_base + SEC_OOO_SHUTDOWN_SEL);
+}
+
+static void sec_enable_axi_error(struct hisi_qm *qm)
+{
+	struct hisi_qm_err_mask *dev_err = &qm->err_info.dev_err;
+	u32 err_mask = dev_err->ce | dev_err->nfe | dev_err->fe;
+
+	/* clear axi error source */
+	writel(SEC_AXI_ERROR_MASK, qm->io_base + SEC_CORE_INT_SOURCE);
+
+	writel(err_mask, qm->io_base + SEC_CORE_INT_MASK);
+
+	if (qm->ver > QM_HW_V2)
+		writel(dev_err->shutdown_mask, qm->io_base + SEC_OOO_SHUTDOWN_SEL);
+}
+
 static void sec_err_info_init(struct hisi_qm *qm)
 {
 	struct hisi_qm_err_info *err_info = &qm->err_info;
+	struct hisi_qm_err_mask *qm_err = &err_info->qm_err;
+	struct hisi_qm_err_mask *dev_err = &err_info->dev_err;
 
-	err_info->fe = SEC_RAS_FE_ENB_MSK;
-	err_info->ce = hisi_qm_get_hw_info(qm, sec_basic_info, SEC_QM_CE_MASK_CAP, qm->cap_ver);
-	err_info->nfe = hisi_qm_get_hw_info(qm, sec_basic_info, SEC_QM_NFE_MASK_CAP, qm->cap_ver);
-	err_info->ecc_2bits_mask = SEC_CORE_INT_STATUS_M_ECC;
-	err_info->qm_shutdown_mask = hisi_qm_get_hw_info(qm, sec_basic_info,
-				     SEC_QM_OOO_SHUTDOWN_MASK_CAP, qm->cap_ver);
-	err_info->dev_shutdown_mask = hisi_qm_get_hw_info(qm, sec_basic_info,
-			SEC_OOO_SHUTDOWN_MASK_CAP, qm->cap_ver);
-	err_info->qm_reset_mask = hisi_qm_get_hw_info(qm, sec_basic_info,
-			SEC_QM_RESET_MASK_CAP, qm->cap_ver);
-	err_info->dev_reset_mask = hisi_qm_get_hw_info(qm, sec_basic_info,
-			SEC_RESET_MASK_CAP, qm->cap_ver);
+	qm_err->fe = SEC_RAS_FE_ENB_MSK;
+	qm_err->ce = hisi_qm_get_hw_info(qm, sec_basic_info, SEC_QM_CE_MASK_CAP, qm->cap_ver);
+	qm_err->nfe = hisi_qm_get_hw_info(qm, sec_basic_info, SEC_QM_NFE_MASK_CAP, qm->cap_ver);
+	qm_err->shutdown_mask = hisi_qm_get_hw_info(qm, sec_basic_info,
+						    SEC_QM_OOO_SHUTDOWN_MASK_CAP, qm->cap_ver);
+	qm_err->reset_mask = hisi_qm_get_hw_info(qm, sec_basic_info,
+						 SEC_QM_RESET_MASK_CAP, qm->cap_ver);
+	qm_err->ecc_2bits_mask = QM_ECC_MBIT;
+
+	dev_err->fe = SEC_RAS_FE_ENB_MSK;
+	dev_err->ce = hisi_qm_get_hw_info(qm, sec_basic_info, SEC_CE_MASK_CAP, qm->cap_ver);
+	dev_err->nfe = hisi_qm_get_hw_info(qm, sec_basic_info, SEC_NFE_MASK_CAP, qm->cap_ver);
+	dev_err->shutdown_mask = hisi_qm_get_hw_info(qm, sec_basic_info,
+						     SEC_OOO_SHUTDOWN_MASK_CAP, qm->cap_ver);
+	dev_err->reset_mask = hisi_qm_get_hw_info(qm, sec_basic_info,
+						  SEC_RESET_MASK_CAP, qm->cap_ver);
+	dev_err->ecc_2bits_mask = SEC_CORE_INT_STATUS_M_ECC;
+
 	err_info->msi_wr_port = BIT(0);
 	err_info->acpi_rst = "SRST";
 }
···
 	.err_info_init		= sec_err_info_init,
 	.get_err_result		= sec_get_err_result,
 	.dev_is_abnormal        = sec_dev_is_abnormal,
+	.disable_axi_error	= sec_disable_axi_error,
+	.enable_axi_error	= sec_enable_axi_error,
 };
 
 static int sec_pf_probe_init(struct sec_dev *sec)
···
 	if (ret)
 		return ret;
 
-	sec_open_sva_prefetch(qm);
 	hisi_qm_dev_err_init(qm);
 	sec_debug_regs_clear(qm);
 	ret = sec_show_last_regs_init(qm);
···
 	size_t i, size;
 
 	size = ARRAY_SIZE(sec_cap_query_info);
-	sec_cap = devm_kzalloc(&pdev->dev, sizeof(*sec_cap) * size, GFP_KERNEL);
+	sec_cap = devm_kcalloc(&pdev->dev, size, sizeof(*sec_cap), GFP_KERNEL);
 	if (!sec_cap)
 		return -ENOMEM;
 

drivers/crypto/hisilicon/zip/dae_main.c

···
 #define DAE_REG_RD_TMOUT_US		USEC_PER_SEC
 
 #define DAE_ALG_NAME			"hashagg"
+#define DAE_V5_ALG_NAME			"hashagg\nudma\nhashjoin\ngather"
 
 /* error */
 #define DAE_AXI_CFG_OFFSET		0x331000
···
 
 int hisi_dae_set_alg(struct hisi_qm *qm)
 {
+	const char *alg_name;
 	size_t len;
 
 	if (!dae_is_support(qm))
···
 	if (!qm->uacce)
 		return 0;
 
+	if (qm->ver >= QM_HW_V5)
+		alg_name = DAE_V5_ALG_NAME;
+	else
+		alg_name = DAE_ALG_NAME;
+
 	len = strlen(qm->uacce->algs);
 	/* A line break may be required */
-	if (len + strlen(DAE_ALG_NAME) + 1 >= QM_DEV_ALG_MAX_LEN) {
+	if (len + strlen(alg_name) + 1 >= QM_DEV_ALG_MAX_LEN) {
 		pci_err(qm->pdev, "algorithm name is too long!\n");
 		return -EINVAL;
 	}
···
 	if (len)
 		strcat((char *)qm->uacce->algs, "\n");
 
-	strcat((char *)qm->uacce->algs, DAE_ALG_NAME);
+	strcat((char *)qm->uacce->algs, alg_name);
 
 	return 0;
 }
···
 	writel(DAE_ERR_NFE_MASK & (~err_type), qm->io_base + DAE_ERR_NFE_OFFSET);
 }
 
+static void hisi_dae_enable_error_report(struct hisi_qm *qm)
+{
+	writel(DAE_ERR_CE_MASK, qm->io_base + DAE_ERR_CE_OFFSET);
+	writel(DAE_ERR_NFE_MASK, qm->io_base + DAE_ERR_NFE_OFFSET);
+}
+
 static void hisi_dae_log_hw_error(struct hisi_qm *qm, u32 err_type)
 {
 	const struct hisi_dae_hw_error *err = dae_hw_error;
···
 		return ACC_ERR_NEED_RESET;
 	}
 	hisi_dae_clear_hw_err_status(qm, err_status);
+	/* Avoid firmware disable error report, re-enable. */
+	hisi_dae_enable_error_report(qm);
 
 	return ACC_ERR_RECOVERED;
 }

drivers/crypto/hisilicon/zip/zip_main.c

···
 #define HZIP_SRAM_ECC_ERR_NUM_SHIFT	16
 #define HZIP_SRAM_ECC_ERR_ADDR_SHIFT	24
 #define HZIP_CORE_INT_MASK_ALL		GENMASK(12, 0)
+#define HZIP_AXI_ERROR_MASK		(BIT(2) | BIT(3))
 #define HZIP_SQE_SIZE			128
 #define HZIP_PF_DEF_Q_NUM		64
 #define HZIP_PF_DEF_Q_BASE		0
···
 #define HZIP_ALG_GZIP_BIT		GENMASK(3, 2)
 #define HZIP_ALG_DEFLATE_BIT		GENMASK(5, 4)
 #define HZIP_ALG_LZ77_BIT		GENMASK(7, 6)
+#define HZIP_ALG_LZ4_BIT		GENMASK(9, 8)
 
 #define HZIP_BUF_SIZE			22
 #define HZIP_SQE_MASK_OFFSET		64
···
 #define HZIP_PREFETCH_ENABLE		(~(BIT(26) | BIT(17) | BIT(0)))
 #define HZIP_SVA_PREFETCH_DISABLE	BIT(26)
 #define HZIP_SVA_DISABLE_READY		(BIT(26) | BIT(30))
+#define HZIP_SVA_PREFETCH_NUM		GENMASK(18, 16)
+#define HZIP_SVA_STALL_NUM		GENMASK(15, 0)
 #define HZIP_SHAPER_RATE_COMPRESS	750
 #define HZIP_SHAPER_RATE_DECOMPRESS	140
-#define HZIP_DELAY_1_US		1
-#define HZIP_POLL_TIMEOUT_US	1000
+#define HZIP_DELAY_1_US			1
+#define HZIP_POLL_TIMEOUT_US		1000
+#define HZIP_WAIT_SVA_READY		500000
+#define HZIP_READ_SVA_STATUS_TIMES	3
+#define HZIP_WAIT_US_MIN		10
+#define HZIP_WAIT_US_MAX		20
 
 /* clock gating */
 #define HZIP_PEH_CFG_AUTO_GATE		0x3011A8
···
 
 /* zip comp high performance */
 #define HZIP_HIGH_PERF_OFFSET		0x301208
+
+#define HZIP_LIT_LEN_EN_OFFSET		0x301204
+#define HZIP_LIT_LEN_EN_EN		BIT(4)
 
 enum {
 	HZIP_HIGH_COMP_RATE,
···
 	}, {
 		.alg_msk = HZIP_ALG_LZ77_BIT,
 		.alg = "lz77_zstd\n",
+	}, {
+		.alg_msk = HZIP_ALG_LZ77_BIT,
+		.alg = "lz77_only\n",
+	}, {
+		.alg_msk = HZIP_ALG_LZ4_BIT,
+		.alg = "lz4\n",
 	},
 };
 
···
 	return false;
 }
 
-static int hisi_zip_set_high_perf(struct hisi_qm *qm)
+static void hisi_zip_literal_set(struct hisi_qm *qm)
 {
 	u32 val;
-	int ret;
+
+	if (qm->ver < QM_HW_V3)
+		return;
+
+	val = readl_relaxed(qm->io_base + HZIP_LIT_LEN_EN_OFFSET);
+	val &= ~HZIP_LIT_LEN_EN_EN;
+
+	/* enable literal length in stream mode compression */
+	writel(val, qm->io_base + HZIP_LIT_LEN_EN_OFFSET);
+}
+
+static void hisi_zip_set_high_perf(struct hisi_qm *qm)
+{
+	u32 val;
 
 	val = readl_relaxed(qm->io_base + HZIP_HIGH_PERF_OFFSET);
 	if (perf_mode == HZIP_HIGH_COMP_PERF)
···
 
 	/* Set perf mode */
 	writel(val, qm->io_base + HZIP_HIGH_PERF_OFFSET);
-	ret = readl_relaxed_poll_timeout(qm->io_base + HZIP_HIGH_PERF_OFFSET,
-					 val, val == perf_mode, HZIP_DELAY_1_US,
-					 HZIP_POLL_TIMEOUT_US);
-	if (ret)
-		pci_err(qm->pdev, "failed to set perf mode\n");
-
-	return ret;
 }
 
-static void hisi_zip_open_sva_prefetch(struct hisi_qm *qm)
+static int hisi_zip_wait_sva_ready(struct hisi_qm *qm, __u32 offset, __u32 mask)
 {
-	u32 val;
-	int ret;
+	u32 val, try_times = 0;
+	u8 count = 0;
 
-	if (!test_bit(QM_SUPPORT_SVA_PREFETCH, &qm->caps))
-		return;
+	/*
+	 * Read the register value every 10-20us. If the value is 0 for three
+	 * consecutive times, the SVA module is ready.
+	 */
+	do {
+		val = readl(qm->io_base + offset);
+		if (val & mask)
+			count = 0;
+		else if (++count == HZIP_READ_SVA_STATUS_TIMES)
+			break;
 
-	/* Enable prefetch */
-	val = readl_relaxed(qm->io_base + HZIP_PREFETCH_CFG);
-	val &= HZIP_PREFETCH_ENABLE;
-	writel(val, qm->io_base + HZIP_PREFETCH_CFG);
+		usleep_range(HZIP_WAIT_US_MIN, HZIP_WAIT_US_MAX);
+	} while (++try_times < HZIP_WAIT_SVA_READY);
 
-	ret = readl_relaxed_poll_timeout(qm->io_base + HZIP_PREFETCH_CFG,
-					 val, !(val & HZIP_SVA_PREFETCH_DISABLE),
-					 HZIP_DELAY_1_US, HZIP_POLL_TIMEOUT_US);
-	if (ret)
-		pci_err(qm->pdev, "failed to open sva prefetch\n");
+	if (try_times == HZIP_WAIT_SVA_READY) {
+		pci_err(qm->pdev, "failed to wait sva prefetch ready\n");
+		return -ETIMEDOUT;
+	}
+
+	return 0;
 }
 
 static void hisi_zip_close_sva_prefetch(struct hisi_qm *qm)
···
 					 HZIP_DELAY_1_US, HZIP_POLL_TIMEOUT_US);
 	if (ret)
 		pci_err(qm->pdev, "failed to close sva prefetch\n");
+
+	(void)hisi_zip_wait_sva_ready(qm, HZIP_SVA_TRANS, HZIP_SVA_STALL_NUM);
+}
+
+static void hisi_zip_open_sva_prefetch(struct hisi_qm *qm)
+{
+	u32 val;
+	int ret;
+
+	if (!test_bit(QM_SUPPORT_SVA_PREFETCH, &qm->caps))
+		return;
+
+	/* Enable prefetch */
+	val = readl_relaxed(qm->io_base + HZIP_PREFETCH_CFG);
+	val &= HZIP_PREFETCH_ENABLE;
+	writel(val, qm->io_base + HZIP_PREFETCH_CFG);
+
+	ret = readl_relaxed_poll_timeout(qm->io_base + HZIP_PREFETCH_CFG,
+					 val, !(val & HZIP_SVA_PREFETCH_DISABLE),
+					 HZIP_DELAY_1_US, HZIP_POLL_TIMEOUT_US);
+	if (ret) {
+		pci_err(qm->pdev, "failed to open sva prefetch\n");
+		hisi_zip_close_sva_prefetch(qm);
+		return;
+	}
+
+	ret = hisi_zip_wait_sva_ready(qm, HZIP_SVA_TRANS, HZIP_SVA_PREFETCH_NUM);
+	if (ret)
+		hisi_zip_close_sva_prefetch(qm);
 }
 
 static void hisi_zip_enable_clock_gate(struct hisi_qm *qm)
···
 	void __iomem *base = qm->io_base;
 	u32 dcomp_bm, comp_bm;
 	u32 zip_core_en;
+	int ret;
 
 	/* qm user domain */
 	writel(AXUSER_BASE, base + QM_ARUSER_M_CFG_1);
···
 		writel(AXUSER_BASE, base + HZIP_DATA_WUSER_32_63);
 		writel(AXUSER_BASE, base + HZIP_SGL_RUSER_32_63);
 	}
+	hisi_zip_open_sva_prefetch(qm);
 
 	/* let's open all compression/decompression cores */
 
···
 	       CQC_CACHE_WB_ENABLE | FIELD_PREP(SQC_CACHE_WB_THRD, 1) |
 	       FIELD_PREP(CQC_CACHE_WB_THRD, 1), base + QM_CACHE_CTL);
 
+	hisi_zip_set_high_perf(qm);
+	hisi_zip_literal_set(qm);
 	hisi_zip_enable_clock_gate(qm);
 
-	return hisi_dae_set_user_domain(qm);
+	ret = hisi_dae_set_user_domain(qm);
+	if (ret)
+		goto close_sva_prefetch;
+
+	return 0;
+
+close_sva_prefetch:
+	hisi_zip_close_sva_prefetch(qm);
+	return ret;
 }
 
 static void hisi_zip_master_ooo_ctrl(struct hisi_qm *qm, bool enable)
···
 	val1 = readl(qm->io_base + HZIP_SOFT_CTRL_ZIP_CONTROL);
 	if (enable) {
 		val1 |= HZIP_AXI_SHUTDOWN_ENABLE;
-		val2 = hisi_qm_get_hw_info(qm, zip_basic_cap_info,
-				ZIP_OOO_SHUTDOWN_MASK_CAP, qm->cap_ver);
+		val2 = qm->err_info.dev_err.shutdown_mask;
 	} else {
 		val1 &= ~HZIP_AXI_SHUTDOWN_ENABLE;
 		val2 = 0x0;
···
 
 static void hisi_zip_hw_error_enable(struct hisi_qm *qm)
 {
-	u32 nfe, ce;
+	struct hisi_qm_err_mask *dev_err = &qm->err_info.dev_err;
+	u32 err_mask = dev_err->ce | dev_err->nfe | dev_err->fe;
 
 	if (qm->ver == QM_HW_V1) {
 		writel(HZIP_CORE_INT_MASK_ALL,
···
 		return;
 	}
 
-	nfe = hisi_qm_get_hw_info(qm, zip_basic_cap_info, ZIP_NFE_MASK_CAP, qm->cap_ver);
-	ce = hisi_qm_get_hw_info(qm, zip_basic_cap_info, ZIP_CE_MASK_CAP, qm->cap_ver);
-
 	/* clear ZIP hw error source if having */
-	writel(ce | nfe | HZIP_CORE_INT_RAS_FE_ENB_MASK, qm->io_base + HZIP_CORE_INT_SOURCE);
+	writel(err_mask, qm->io_base + HZIP_CORE_INT_SOURCE);
 
 	/* configure error type */
-	writel(ce, qm->io_base + HZIP_CORE_INT_RAS_CE_ENB);
-	writel(HZIP_CORE_INT_RAS_FE_ENB_MASK, qm->io_base + HZIP_CORE_INT_RAS_FE_ENB);
-	writel(nfe, qm->io_base + HZIP_CORE_INT_RAS_NFE_ENB);
+	writel(dev_err->ce, qm->io_base + HZIP_CORE_INT_RAS_CE_ENB);
+	writel(dev_err->fe, qm->io_base + HZIP_CORE_INT_RAS_FE_ENB);
+	writel(dev_err->nfe, qm->io_base + HZIP_CORE_INT_RAS_NFE_ENB);
 
 	hisi_zip_master_ooo_ctrl(qm, true);
 
 	/* enable ZIP hw error interrupts */
-	writel(0, qm->io_base + HZIP_CORE_INT_MASK_REG);
+	writel(~err_mask, qm->io_base + HZIP_CORE_INT_MASK_REG);
 
 	hisi_dae_hw_error_enable(qm);
 }
 
 static void hisi_zip_hw_error_disable(struct hisi_qm *qm)
 {
-	u32 nfe, ce;
+	struct hisi_qm_err_mask *dev_err = &qm->err_info.dev_err;
+	u32 err_mask = dev_err->ce | dev_err->nfe | dev_err->fe;
 
 	/* disable ZIP hw error interrupts */
-	nfe = hisi_qm_get_hw_info(qm, zip_basic_cap_info, ZIP_NFE_MASK_CAP, qm->cap_ver);
-	ce = hisi_qm_get_hw_info(qm, zip_basic_cap_info, ZIP_CE_MASK_CAP, qm->cap_ver);
-	writel(ce | nfe | HZIP_CORE_INT_RAS_FE_ENB_MASK, qm->io_base + HZIP_CORE_INT_MASK_REG);
+	writel(err_mask, qm->io_base + HZIP_CORE_INT_MASK_REG);
 
 	hisi_zip_master_ooo_ctrl(qm, false);
 
···
 
 static void hisi_zip_disable_error_report(struct hisi_qm *qm, u32 err_type)
 {
-	u32 nfe_mask;
+	u32 nfe_mask = qm->err_info.dev_err.nfe;
 
-	nfe_mask = hisi_qm_get_hw_info(qm, zip_basic_cap_info, ZIP_NFE_MASK_CAP, qm->cap_ver);
 	writel(nfe_mask & (~err_type), qm->io_base + HZIP_CORE_INT_RAS_NFE_ENB);
+}
+
+static void hisi_zip_enable_error_report(struct hisi_qm *qm)
+{
+	u32 nfe_mask = qm->err_info.dev_err.nfe;
+	u32 ce_mask = qm->err_info.dev_err.ce;
+
+	writel(nfe_mask, qm->io_base + HZIP_CORE_INT_RAS_NFE_ENB);
+	writel(ce_mask, qm->io_base + HZIP_CORE_INT_RAS_CE_ENB);
 }
 
 static void hisi_zip_open_axi_master_ooo(struct hisi_qm *qm)
···
 	/* Get device hardware new error status */
 	err_status = hisi_zip_get_hw_err_status(qm);
 	if (err_status) {
-		if (err_status & qm->err_info.ecc_2bits_mask)
+		if (err_status & qm->err_info.dev_err.ecc_2bits_mask)
 			qm->err_status.is_dev_ecc_mbit = true;
 		hisi_zip_log_hw_error(qm, err_status);
 
-		if (err_status & qm->err_info.dev_reset_mask) {
+		if (err_status & qm->err_info.dev_err.reset_mask) {
 			/* Disable the same error reporting until device is recovered. */
 			hisi_zip_disable_error_report(qm, err_status);
-			return ACC_ERR_NEED_RESET;
+			zip_result = ACC_ERR_NEED_RESET;
 		} else {
 			hisi_zip_clear_hw_err_status(qm, err_status);
+			/* Avoid firmware disable error report, re-enable. */
+			hisi_zip_enable_error_report(qm);
 		}
 	}
 
···
 	u32 err_status;
 
 	err_status = hisi_zip_get_hw_err_status(qm);
-	if (err_status & qm->err_info.dev_shutdown_mask)
+	if (err_status & qm->err_info.dev_err.shutdown_mask)
 		return true;
 
 	return hisi_dae_dev_is_abnormal(qm);
···
 	return hisi_dae_close_axi_master_ooo(qm);
 }
 
+static void hisi_zip_disable_axi_error(struct hisi_qm *qm)
+{
+	struct hisi_qm_err_mask *dev_err = &qm->err_info.dev_err;
+	u32 err_mask = dev_err->ce | dev_err->nfe | dev_err->fe;
+	u32 val;
+
+	val = ~(err_mask & (~HZIP_AXI_ERROR_MASK));
+	writel(val, qm->io_base + HZIP_CORE_INT_MASK_REG);
+
+	if (qm->ver > QM_HW_V2)
+		writel(dev_err->shutdown_mask & (~HZIP_AXI_ERROR_MASK),
+		       qm->io_base + HZIP_OOO_SHUTDOWN_SEL);
+}
+
+static void hisi_zip_enable_axi_error(struct hisi_qm *qm)
+{
+	struct hisi_qm_err_mask *dev_err = &qm->err_info.dev_err;
+	u32 err_mask = dev_err->ce | dev_err->nfe | dev_err->fe;
+
+	/* clear axi error source */
+	writel(HZIP_AXI_ERROR_MASK, qm->io_base + HZIP_CORE_INT_SOURCE);
+
+	writel(~err_mask, qm->io_base + HZIP_CORE_INT_MASK_REG);
+
+	if (qm->ver > QM_HW_V2)
+		writel(dev_err->shutdown_mask, qm->io_base + HZIP_OOO_SHUTDOWN_SEL);
+}
+
 static void hisi_zip_err_info_init(struct hisi_qm *qm)
 {
 	struct hisi_qm_err_info *err_info = &qm->err_info;
+	struct hisi_qm_err_mask *qm_err = &err_info->qm_err;
+	struct hisi_qm_err_mask *dev_err = &err_info->dev_err;
 
-	err_info->fe = HZIP_CORE_INT_RAS_FE_ENB_MASK;
-	err_info->ce = hisi_qm_get_hw_info(qm, zip_basic_cap_info, ZIP_QM_CE_MASK_CAP, qm->cap_ver);
-	err_info->nfe = hisi_qm_get_hw_info(qm, zip_basic_cap_info,
-					    ZIP_QM_NFE_MASK_CAP, qm->cap_ver);
-	err_info->ecc_2bits_mask = HZIP_CORE_INT_STATUS_M_ECC;
-	err_info->qm_shutdown_mask = hisi_qm_get_hw_info(qm, zip_basic_cap_info,
-							 ZIP_QM_OOO_SHUTDOWN_MASK_CAP, qm->cap_ver);
-	err_info->dev_shutdown_mask = hisi_qm_get_hw_info(qm, zip_basic_cap_info,
-							  ZIP_OOO_SHUTDOWN_MASK_CAP, qm->cap_ver);
-	err_info->qm_reset_mask = hisi_qm_get_hw_info(qm, zip_basic_cap_info,
-						      ZIP_QM_RESET_MASK_CAP, qm->cap_ver);
-	err_info->dev_reset_mask = hisi_qm_get_hw_info(qm, zip_basic_cap_info,
-						       ZIP_RESET_MASK_CAP, qm->cap_ver);
+	qm_err->fe = HZIP_CORE_INT_RAS_FE_ENB_MASK;
+	qm_err->ce = hisi_qm_get_hw_info(qm, zip_basic_cap_info, ZIP_QM_CE_MASK_CAP, qm->cap_ver);
+	qm_err->nfe = hisi_qm_get_hw_info(qm, zip_basic_cap_info,
+					  ZIP_QM_NFE_MASK_CAP, qm->cap_ver);
+	qm_err->ecc_2bits_mask = QM_ECC_MBIT;
+	qm_err->reset_mask = hisi_qm_get_hw_info(qm, zip_basic_cap_info,
+						 ZIP_QM_RESET_MASK_CAP, qm->cap_ver);
+	qm_err->shutdown_mask = hisi_qm_get_hw_info(qm, zip_basic_cap_info,
+						    ZIP_QM_OOO_SHUTDOWN_MASK_CAP, qm->cap_ver);
+
+	dev_err->fe = HZIP_CORE_INT_RAS_FE_ENB_MASK;
+	dev_err->ce = hisi_qm_get_hw_info(qm, zip_basic_cap_info, ZIP_CE_MASK_CAP, qm->cap_ver);
+	dev_err->nfe = hisi_qm_get_hw_info(qm, zip_basic_cap_info, ZIP_NFE_MASK_CAP, qm->cap_ver);
+	dev_err->ecc_2bits_mask = HZIP_CORE_INT_STATUS_M_ECC;
+	dev_err->shutdown_mask = hisi_qm_get_hw_info(qm, zip_basic_cap_info,
+						     ZIP_OOO_SHUTDOWN_MASK_CAP, qm->cap_ver);
+	dev_err->reset_mask = hisi_qm_get_hw_info(qm, zip_basic_cap_info,
+						  ZIP_RESET_MASK_CAP, qm->cap_ver);
+
 	err_info->msi_wr_port = HZIP_WR_PORT;
 	err_info->acpi_rst = "ZRST";
 }
···
 	.get_err_result		= hisi_zip_get_err_result,
 	.set_priv_status	= hisi_zip_set_priv_status,
 	.dev_is_abnormal	= hisi_zip_dev_is_abnormal,
+	.disable_axi_error	= hisi_zip_disable_axi_error,
+	.enable_axi_error	= hisi_zip_enable_axi_error,
 };
 
 static int hisi_zip_pf_probe_init(struct hisi_zip *hisi_zip)
···
 	if (ret)
 		return ret;
 
-	ret = hisi_zip_set_high_perf(qm);
-	if (ret)
-		return ret;
-
-	hisi_zip_open_sva_prefetch(qm);
 	hisi_qm_dev_err_init(qm);
 	hisi_zip_debug_regs_clear(qm);
 
···
 	size_t i, size;
 
 	size = ARRAY_SIZE(zip_cap_query_info);
-	zip_cap = devm_kzalloc(&pdev->dev, sizeof(*zip_cap) * size, GFP_KERNEL);
+	zip_cap = devm_kcalloc(&pdev->dev, size, sizeof(*zip_cap), GFP_KERNEL);
 	if (!zip_cap)
 		return -ENOMEM;
 

drivers/crypto/intel/keembay/keembay-ocs-hcu-core.c

···
 	struct device *dev = rctx->hcu_dev->dev;
 	unsigned int remainder = 0;
 	unsigned int total;
-	size_t nents;
+	int nents;
 	size_t count;
 	int rc;
 	int i;
···
 
 	/* Determine the number of scatter gather list entries to process. */
 	nents = sg_nents_for_len(req->src, rctx->sg_data_total - remainder);
+
+	if (nents < 0)
+		return nents;
 
 	/* If there are entries to process, map them. */
 	if (nents) {

drivers/crypto/intel/qat/Kconfig

···
 	select CRYPTO_SKCIPHER
 	select CRYPTO_AKCIPHER
 	select CRYPTO_DH
-	select CRYPTO_HMAC
 	select CRYPTO_RSA
-	select CRYPTO_SHA1
-	select CRYPTO_SHA256
-	select CRYPTO_SHA512
 	select CRYPTO_LIB_AES
+	select CRYPTO_LIB_SHA1
+	select CRYPTO_LIB_SHA256
+	select CRYPTO_LIB_SHA512
 	select FW_LOADER
 	select CRC8
 

drivers/crypto/intel/qat/qat_common/adf_ctl_drv.c

···
 	return -EFAULT;
 }
 
-static int adf_ctl_alloc_resources(struct adf_user_cfg_ctl_data **ctl_data,
-				   unsigned long arg)
+static struct adf_user_cfg_ctl_data *adf_ctl_alloc_resources(unsigned long arg)
 {
 	struct adf_user_cfg_ctl_data *cfg_data;
 
-	cfg_data = kzalloc(sizeof(*cfg_data), GFP_KERNEL);
-	if (!cfg_data)
-		return -ENOMEM;
-
-	/* Initialize device id to NO DEVICE as 0 is a valid device id */
-	cfg_data->device_id = ADF_CFG_NO_DEVICE;
-
-	if (copy_from_user(cfg_data, (void __user *)arg, sizeof(*cfg_data))) {
+	cfg_data = memdup_user((void __user *)arg, sizeof(*cfg_data));
+	if (IS_ERR(cfg_data))
 		pr_err("QAT: failed to copy from user cfg_data.\n");
-		kfree(cfg_data);
-		return -EIO;
-	}
-
-	*ctl_data = cfg_data;
-	return 0;
+	return cfg_data;
 }
 
 static int adf_add_key_value_data(struct adf_accel_dev *accel_dev,
···
 static int adf_ctl_ioctl_dev_config(struct file *fp, unsigned int cmd,
 				    unsigned long arg)
 {
-	int ret;
 	struct adf_user_cfg_ctl_data *ctl_data;
 	struct adf_accel_dev *accel_dev;
+	int ret = 0;
 
-	ret = adf_ctl_alloc_resources(&ctl_data, arg);
-	if (ret)
-		return ret;
+	ctl_data = adf_ctl_alloc_resources(arg);
+	if (IS_ERR(ctl_data))
+		return PTR_ERR(ctl_data);
 
 	accel_dev = adf_devmgr_get_dev_by_id(ctl_data->device_id);
 	if (!accel_dev) {
···
 	int ret;
 	struct adf_user_cfg_ctl_data *ctl_data;
 
-	ret = adf_ctl_alloc_resources(&ctl_data, arg);
-	if (ret)
-		return ret;
+	ctl_data = adf_ctl_alloc_resources(arg);
+	if (IS_ERR(ctl_data))
+		return PTR_ERR(ctl_data);
 
 	if (adf_devmgr_verify_id(ctl_data->device_id)) {
 		pr_err("QAT: Device %d not found\n", ctl_data->device_id);
···
 	struct adf_user_cfg_ctl_data *ctl_data;
 	struct adf_accel_dev *accel_dev;
 
-	ret = adf_ctl_alloc_resources(&ctl_data, arg);
-	if (ret)
-		return ret;
+	ctl_data = adf_ctl_alloc_resources(arg);
+	if (IS_ERR(ctl_data))
+		return PTR_ERR(ctl_data);
 
 	ret = -ENODEV;
 	accel_dev = adf_devmgr_get_dev_by_id(ctl_data->device_id);

drivers/crypto/intel/qat/qat_common/adf_gen6_tl.c

···
 
 #define SLICE_IDX(sl) offsetof(struct icp_qat_fw_init_admin_slice_cnt, sl##_cnt)
 
+#define ADF_GEN6_TL_CMDQ_WAIT_COUNTER(_name)                     \
+	ADF_TL_COUNTER("cmdq_wait_" #_name, ADF_TL_SIMPLE_COUNT, \
+		       ADF_TL_CMDQ_REG_OFF(_name, reg_tm_cmdq_wait_cnt, gen6))
+#define ADF_GEN6_TL_CMDQ_EXEC_COUNTER(_name)                     \
+	ADF_TL_COUNTER("cmdq_exec_" #_name, ADF_TL_SIMPLE_COUNT, \
+		       ADF_TL_CMDQ_REG_OFF(_name, reg_tm_cmdq_exec_cnt, gen6))
+#define ADF_GEN6_TL_CMDQ_DRAIN_COUNTER(_name)                            \
+	ADF_TL_COUNTER("cmdq_drain_" #_name, ADF_TL_SIMPLE_COUNT,        \
+		       ADF_TL_CMDQ_REG_OFF(_name, reg_tm_cmdq_drain_cnt, \
+					   gen6))
+
+#define CPR_QUEUE_COUNT		5
+#define DCPR_QUEUE_COUNT	3
+#define PKE_QUEUE_COUNT		1
+#define WAT_QUEUE_COUNT		7
+#define WCP_QUEUE_COUNT		7
+#define USC_QUEUE_COUNT		3
+#define ATH_QUEUE_COUNT		2
+
 /* Device level counters. */
 static const struct adf_tl_dbg_counter dev_counters[] = {
 	/* PCIe partial transactions. */
···
 	/* Maximum uTLB used. */
 	ADF_TL_COUNTER(AT_MAX_UTLB_USED_NAME, ADF_TL_SIMPLE_COUNT,
 		       ADF_GEN6_TL_DEV_REG_OFF(reg_tl_at_max_utlb_used)),
+	/* Ring Empty average[ns] across all rings */
+	ADF_TL_COUNTER_LATENCY(RE_ACC_NAME, ADF_TL_COUNTER_NS_AVG,
+			       ADF_GEN6_TL_DEV_REG_OFF(reg_tl_re_acc),
+			       ADF_GEN6_TL_DEV_REG_OFF(reg_tl_re_cnt)),
 };
 
 /* Accelerator utilization counters */
···
 	[SLICE_IDX(ath)] = ADF_GEN6_TL_SL_EXEC_COUNTER(ath),
 };
 
+static const struct adf_tl_dbg_counter cnv_cmdq_counters[] = {
+	ADF_GEN6_TL_CMDQ_WAIT_COUNTER(cnv),
+	ADF_GEN6_TL_CMDQ_EXEC_COUNTER(cnv),
+	ADF_GEN6_TL_CMDQ_DRAIN_COUNTER(cnv)
+};
+
+#define NUM_CMDQ_COUNTERS ARRAY_SIZE(cnv_cmdq_counters)
+
+static const struct adf_tl_dbg_counter dcprz_cmdq_counters[] = {
+	ADF_GEN6_TL_CMDQ_WAIT_COUNTER(dcprz),
+	ADF_GEN6_TL_CMDQ_EXEC_COUNTER(dcprz),
+	ADF_GEN6_TL_CMDQ_DRAIN_COUNTER(dcprz)
+};
+
+static_assert(ARRAY_SIZE(dcprz_cmdq_counters) == NUM_CMDQ_COUNTERS);
+
+static const struct adf_tl_dbg_counter pke_cmdq_counters[] = {
+	ADF_GEN6_TL_CMDQ_WAIT_COUNTER(pke),
+	ADF_GEN6_TL_CMDQ_EXEC_COUNTER(pke),
+	ADF_GEN6_TL_CMDQ_DRAIN_COUNTER(pke)
+};
+
+static_assert(ARRAY_SIZE(pke_cmdq_counters) == NUM_CMDQ_COUNTERS);
+
+static const struct adf_tl_dbg_counter wat_cmdq_counters[] = {
+	ADF_GEN6_TL_CMDQ_WAIT_COUNTER(wat),
+	ADF_GEN6_TL_CMDQ_EXEC_COUNTER(wat),
+	ADF_GEN6_TL_CMDQ_DRAIN_COUNTER(wat)
+};
+
+static_assert(ARRAY_SIZE(wat_cmdq_counters) == NUM_CMDQ_COUNTERS);
+
+static const struct adf_tl_dbg_counter wcp_cmdq_counters[] = {
+	ADF_GEN6_TL_CMDQ_WAIT_COUNTER(wcp),
+	ADF_GEN6_TL_CMDQ_EXEC_COUNTER(wcp),
+	ADF_GEN6_TL_CMDQ_DRAIN_COUNTER(wcp)
+};
+
+static_assert(ARRAY_SIZE(wcp_cmdq_counters) == NUM_CMDQ_COUNTERS);
+
+static const struct adf_tl_dbg_counter ucs_cmdq_counters[] = {
+	ADF_GEN6_TL_CMDQ_WAIT_COUNTER(ucs),
+	ADF_GEN6_TL_CMDQ_EXEC_COUNTER(ucs),
+	ADF_GEN6_TL_CMDQ_DRAIN_COUNTER(ucs)
+};
+
+static_assert(ARRAY_SIZE(ucs_cmdq_counters) == NUM_CMDQ_COUNTERS);
+
+static const struct adf_tl_dbg_counter ath_cmdq_counters[] = {
+	ADF_GEN6_TL_CMDQ_WAIT_COUNTER(ath),
+	ADF_GEN6_TL_CMDQ_EXEC_COUNTER(ath),
+	ADF_GEN6_TL_CMDQ_DRAIN_COUNTER(ath)
+};
+
+static_assert(ARRAY_SIZE(ath_cmdq_counters) == NUM_CMDQ_COUNTERS);
+
+/* CMDQ drain counters. */
+static const struct adf_tl_dbg_counter *cmdq_counters[ADF_TL_SL_CNT_COUNT] = {
+	/* Compression accelerator execution count. */
+	[SLICE_IDX(cpr)] = cnv_cmdq_counters,
+	/* Decompression accelerator execution count. */
+	[SLICE_IDX(dcpr)] = dcprz_cmdq_counters,
+	/* PKE execution count. */
+	[SLICE_IDX(pke)] = pke_cmdq_counters,
+	/* Wireless Authentication accelerator execution count. */
+	[SLICE_IDX(wat)] = wat_cmdq_counters,
+	/* Wireless Cipher accelerator execution count. */
+	[SLICE_IDX(wcp)] = wcp_cmdq_counters,
+	/* UCS accelerator execution count. */
+	[SLICE_IDX(ucs)] = ucs_cmdq_counters,
+	/* Authentication accelerator execution count. */
+	[SLICE_IDX(ath)] = ath_cmdq_counters,
+};
+
 /* Ring pair counters. */
 static const struct adf_tl_dbg_counter rp_counters[] = {
 	/* PCIe partial transactions. */
···
 	/* Payload DevTLB miss rate. */
 	ADF_TL_COUNTER(AT_PAYLD_DTLB_MISS_NAME, ADF_TL_SIMPLE_COUNT,
 		       ADF_GEN6_TL_RP_REG_OFF(reg_tl_at_payld_devtlb_miss)),
+	/* Ring Empty average[ns]. */
+	ADF_TL_COUNTER_LATENCY(RE_ACC_NAME, ADF_TL_COUNTER_NS_AVG,
+			       ADF_GEN6_TL_RP_REG_OFF(reg_tl_re_acc),
+			       ADF_GEN6_TL_RP_REG_OFF(reg_tl_re_cnt)),
 };
 
 void adf_gen6_init_tl_data(struct adf_tl_hw_data *tl_data)
 {
 	tl_data->layout_sz = ADF_GEN6_TL_LAYOUT_SZ;
 	tl_data->slice_reg_sz = ADF_GEN6_TL_SLICE_REG_SZ;
+	tl_data->cmdq_reg_sz = ADF_GEN6_TL_CMDQ_REG_SZ;
 	tl_data->rp_reg_sz = ADF_GEN6_TL_RP_REG_SZ;
 	tl_data->num_hbuff = ADF_GEN6_TL_NUM_HIST_BUFFS;
 	tl_data->max_rp = ADF_GEN6_TL_MAX_RP_NUM;
···
 	tl_data->num_dev_counters = ARRAY_SIZE(dev_counters);
 	tl_data->sl_util_counters = sl_util_counters;
 	tl_data->sl_exec_counters = sl_exec_counters;
+	tl_data->cmdq_counters = cmdq_counters;
+	tl_data->num_cmdq_counters = NUM_CMDQ_COUNTERS;
 	tl_data->rp_counters = rp_counters;
 	tl_data->num_rp_counters = ARRAY_SIZE(rp_counters);
 	tl_data->max_sl_cnt = ADF_GEN6_TL_MAX_SLICES_PER_TYPE;
+
+	tl_data->multiplier.cpr_cnt = CPR_QUEUE_COUNT;
+	tl_data->multiplier.dcpr_cnt = DCPR_QUEUE_COUNT;
+	tl_data->multiplier.pke_cnt = PKE_QUEUE_COUNT;
+	tl_data->multiplier.wat_cnt = WAT_QUEUE_COUNT;
+	tl_data->multiplier.wcp_cnt = WCP_QUEUE_COUNT;
+	tl_data->multiplier.ucs_cnt = USC_QUEUE_COUNT;
+	tl_data->multiplier.ath_cnt = ATH_QUEUE_COUNT;
 }
 EXPORT_SYMBOL_GPL(adf_gen6_init_tl_data);

drivers/crypto/intel/qat/qat_common/adf_telemetry.c

···
 	return ret;
 }
 
+static void adf_set_cmdq_cnt(struct adf_accel_dev *accel_dev,
+			     struct adf_tl_hw_data *tl_data)
+{
+	struct icp_qat_fw_init_admin_slice_cnt *slice_cnt, *cmdq_cnt;
+
+	slice_cnt = &accel_dev->telemetry->slice_cnt;
+	cmdq_cnt = &accel_dev->telemetry->cmdq_cnt;
+
+	cmdq_cnt->cpr_cnt = slice_cnt->cpr_cnt * tl_data->multiplier.cpr_cnt;
+	cmdq_cnt->dcpr_cnt = slice_cnt->dcpr_cnt * tl_data->multiplier.dcpr_cnt;
+	cmdq_cnt->pke_cnt = slice_cnt->pke_cnt * tl_data->multiplier.pke_cnt;
+	cmdq_cnt->wat_cnt = slice_cnt->wat_cnt * tl_data->multiplier.wat_cnt;
+	cmdq_cnt->wcp_cnt = slice_cnt->wcp_cnt * tl_data->multiplier.wcp_cnt;
+	cmdq_cnt->ucs_cnt = slice_cnt->ucs_cnt * tl_data->multiplier.ucs_cnt;
+	cmdq_cnt->ath_cnt = slice_cnt->ath_cnt * tl_data->multiplier.ath_cnt;
+}
+
 int adf_tl_run(struct adf_accel_dev *accel_dev, int state)
 {
 	struct adf_tl_hw_data *tl_data = &GET_TL_DATA(accel_dev);
···
 		adf_send_admin_tl_stop(accel_dev);
 		return ret;
 	}
+
+	adf_set_cmdq_cnt(accel_dev, tl_data);
 
 	telemetry->hbuffs = state;
 	atomic_set(&telemetry->state, state);

drivers/crypto/intel/qat/qat_common/adf_telemetry.h

···
 struct adf_tl_hw_data {
 	size_t layout_sz;
 	size_t slice_reg_sz;
+	size_t cmdq_reg_sz;
 	size_t rp_reg_sz;
 	size_t msg_cnt_off;
 	const struct adf_tl_dbg_counter *dev_counters;
 	const struct adf_tl_dbg_counter *sl_util_counters;
 	const struct adf_tl_dbg_counter *sl_exec_counters;
+	const struct adf_tl_dbg_counter **cmdq_counters;
 	const struct adf_tl_dbg_counter *rp_counters;
 	u8 num_hbuff;
 	u8 cpp_ns_per_cycle;
 	u8 bw_units_to_bytes;
 	u8 num_dev_counters;
 	u8 num_rp_counters;
+	u8 num_cmdq_counters;
 	u8 max_rp;
 	u8 max_sl_cnt;
+	struct icp_qat_fw_init_admin_slice_cnt multiplier;
 };
 
 struct adf_telemetry {
···
 	struct mutex wr_lock;
 	struct delayed_work work_ctx;
 	struct icp_qat_fw_init_admin_slice_cnt slice_cnt;
+	struct icp_qat_fw_init_admin_slice_cnt cmdq_cnt;
 };
 
 #ifdef CONFIG_DEBUG_FS

drivers/crypto/intel/qat/qat_common/adf_tl_debugfs.c

···
 	return 0;
 }
 
+static int tl_print_cmdq_counter(struct adf_telemetry *telemetry,
+				 const struct adf_tl_dbg_counter *ctr,
+				 struct seq_file *s, u8 cnt_id, u8 counter)
+{
+	size_t cmdq_regs_sz = GET_TL_DATA(telemetry->accel_dev).cmdq_reg_sz;
+	size_t offset_inc = cnt_id * cmdq_regs_sz;
+	struct adf_tl_dbg_counter slice_ctr;
+	char cnt_name[MAX_COUNT_NAME_SIZE];
+
+	slice_ctr = *(ctr + counter);
+	slice_ctr.offset1 += offset_inc;
+	snprintf(cnt_name, MAX_COUNT_NAME_SIZE, "%s%d", slice_ctr.name, cnt_id);
+
+	return tl_calc_and_print_counter(telemetry, s, &slice_ctr, cnt_name);
+}
+
+static int tl_calc_and_print_cmdq_counters(struct adf_accel_dev *accel_dev,
+					   struct seq_file *s, u8 cnt_type,
+					   u8 cnt_id)
+{
+	struct adf_tl_hw_data *tl_data = &GET_TL_DATA(accel_dev);
+	struct adf_telemetry *telemetry = accel_dev->telemetry;
+	const struct adf_tl_dbg_counter **cmdq_tl_counters;
+	const struct adf_tl_dbg_counter *ctr;
+	u8 counter;
+	int ret;
+
+	cmdq_tl_counters = tl_data->cmdq_counters;
+	ctr = cmdq_tl_counters[cnt_type];
+
+	for (counter = 0; counter < tl_data->num_cmdq_counters; counter++) {
+		ret = tl_print_cmdq_counter(telemetry, ctr, s, cnt_id, counter);
+		if (ret) {
+			dev_notice(&GET_DEV(accel_dev),
+				   "invalid slice utilization counter type\n");
+			return ret;
+		}
+	}
+
+	return 0;
+}
+
 static void tl_print_msg_cnt(struct seq_file *s, u32 msg_cnt)
 {
 	seq_printf(s, "%-*s", TL_KEY_MIN_PADDING, SNAPSHOT_CNT_MSG);
···
 	struct adf_telemetry *telemetry = accel_dev->telemetry;
 	const struct adf_tl_dbg_counter *dev_tl_counters;
 	u8 num_dev_counters = tl_data->num_dev_counters;
+	u8 *cmdq_cnt = (u8 *)&telemetry->cmdq_cnt;
 	u8 *sl_cnt = (u8 *)&telemetry->slice_cnt;
 	const struct adf_tl_dbg_counter *ctr;
 	unsigned int i;
···
 	for (i = 0; i < ADF_TL_SL_CNT_COUNT; i++) {
 		for (j = 0; j < sl_cnt[i]; j++) {
 			ret = tl_calc_and_print_sl_counters(accel_dev, s, i, j);
+			if (ret)
+				return ret;
+		}
+	}
+
+	/* Print per command queue telemetry. */
+	for (i = 0; i < ADF_TL_SL_CNT_COUNT; i++) {
+		for (j = 0; j < cmdq_cnt[i]; j++) {
+			ret = tl_calc_and_print_cmdq_counters(accel_dev, s, i, j);
 			if (ret)
 				return ret;
 		}

drivers/crypto/intel/qat/qat_common/adf_tl_debugfs.h

···
 #define LAT_ACC_NAME		"gp_lat_acc_avg"
 #define BW_IN_NAME		"bw_in"
 #define BW_OUT_NAME		"bw_out"
+#define RE_ACC_NAME		"re_acc_avg"
 #define PAGE_REQ_LAT_NAME	"at_page_req_lat_avg"
 #define AT_TRANS_LAT_NAME	"at_trans_lat_avg"
 #define AT_MAX_UTLB_USED_NAME	"at_max_tlb_used"
···
 #define ADF_TL_SLICE_REG_OFF(slice, reg, qat_gen)		\
 	(ADF_TL_DEV_REG_OFF(slice##_slices[0], qat_gen) +	\
 	offsetof(struct adf_##qat_gen##_tl_slice_data_regs, reg))
+
+#define ADF_TL_CMDQ_REG_OFF(slice, reg, qat_gen)        \
+	(ADF_TL_DEV_REG_OFF(slice##_cmdq[0], qat_gen) + \
+	 offsetof(struct adf_##qat_gen##_tl_cmdq_data_regs, reg))
 
 #define ADF_TL_RP_REG_OFF(reg, qat_gen)					\
 	(ADF_TL_DATA_REG_OFF(tl_ring_pairs_data_regs[0], qat_gen) +	\

drivers/crypto/intel/qat/qat_common/qat_algs.c

···
 #include <linux/crypto.h>
 #include <crypto/internal/aead.h>
 #include <crypto/internal/cipher.h>
-#include <crypto/internal/hash.h>
 #include <crypto/internal/skcipher.h>
 #include <crypto/aes.h>
 #include <crypto/sha1.h>
 #include <crypto/sha2.h>
-#include <crypto/hmac.h>
 #include <crypto/algapi.h>
 #include <crypto/authenc.h>
 #include <crypto/scatterwalk.h>
···
 	dma_addr_t dec_cd_paddr;
 	struct icp_qat_fw_la_bulk_req enc_fw_req;
 	struct icp_qat_fw_la_bulk_req dec_fw_req;
-	struct crypto_shash *hash_tfm;
 	enum icp_qat_hw_auth_algo qat_hash_alg;
+	unsigned int hash_digestsize;
+	unsigned int hash_blocksize;
 	struct qat_crypto_instance *inst;
-	union {
-		struct sha1_state sha1;
-		struct sha256_state sha256;
-		struct sha512_state sha512;
-	};
-	char ipad[SHA512_BLOCK_SIZE]; /* sufficient for SHA-1/SHA-256 as well */
-	char opad[SHA512_BLOCK_SIZE];
 };
 
 struct qat_alg_skcipher_ctx {
···
 	int mode;
 };
 
-static int qat_get_inter_state_size(enum icp_qat_hw_auth_algo qat_hash_alg)
-{
-	switch (qat_hash_alg) {
-	case ICP_QAT_HW_AUTH_ALGO_SHA1:
-		return ICP_QAT_HW_SHA1_STATE1_SZ;
-	case ICP_QAT_HW_AUTH_ALGO_SHA256:
-		return ICP_QAT_HW_SHA256_STATE1_SZ;
-	case ICP_QAT_HW_AUTH_ALGO_SHA512:
-		return ICP_QAT_HW_SHA512_STATE1_SZ;
-	default:
-		return -EFAULT;
-	}
-}
-
 static int qat_alg_do_precomputes(struct icp_qat_hw_auth_algo_blk *hash,
 				  struct qat_alg_aead_ctx *ctx,
 				  const u8 *auth_key,
 				  unsigned int auth_keylen)
 {
-	SHASH_DESC_ON_STACK(shash, ctx->hash_tfm);
-	int block_size = crypto_shash_blocksize(ctx->hash_tfm);
-	int digest_size = crypto_shash_digestsize(ctx->hash_tfm);
-	__be32 *hash_state_out;
-	__be64 *hash512_state_out;
-	int i, offset;
-
-	memset(ctx->ipad, 0, block_size);
-	memset(ctx->opad, 0, block_size);
-	shash->tfm = ctx->hash_tfm;
-
-	if (auth_keylen > block_size) {
-		int ret = crypto_shash_digest(shash, auth_key,
-					      auth_keylen, ctx->ipad);
-		if (ret)
-			return ret;
-
-		memcpy(ctx->opad, ctx->ipad, digest_size);
-	} else {
-		memcpy(ctx->ipad, auth_key, auth_keylen);
-		memcpy(ctx->opad, auth_key, auth_keylen);
-	}
-
-	for (i = 0; i < block_size; i++) {
-		char *ipad_ptr = ctx->ipad + i;
-		char *opad_ptr = ctx->opad + i;
-		*ipad_ptr ^= HMAC_IPAD_VALUE;
-		*opad_ptr ^= HMAC_OPAD_VALUE;
-	}
-
-	if (crypto_shash_init(shash))
-		return -EFAULT;
-
-	if (crypto_shash_update(shash, ctx->ipad, block_size))
-		return -EFAULT;
-
-	hash_state_out = (__be32 *)hash->sha.state1;
-	hash512_state_out = (__be64 *)hash_state_out;
-
 	switch (ctx->qat_hash_alg) {
-	case ICP_QAT_HW_AUTH_ALGO_SHA1:
-		if (crypto_shash_export_core(shash, &ctx->sha1))
-			return -EFAULT;
-		for (i = 0; i < digest_size >> 2; i++, hash_state_out++)
-			*hash_state_out = cpu_to_be32(ctx->sha1.state[i]);
-		break;
-	case ICP_QAT_HW_AUTH_ALGO_SHA256:
-		if (crypto_shash_export_core(shash, &ctx->sha256))
-			return -EFAULT;
-		for (i = 0; i < digest_size >> 2; i++, hash_state_out++)
-			*hash_state_out = cpu_to_be32(ctx->sha256.state[i]);
-		break;
-	case ICP_QAT_HW_AUTH_ALGO_SHA512:
-		if (crypto_shash_export_core(shash, &ctx->sha512))
-			return -EFAULT;
-		for (i = 0; i < digest_size >> 3; i++, hash512_state_out++)
-			*hash512_state_out = cpu_to_be64(ctx->sha512.state[i]);
-		break;
+	case ICP_QAT_HW_AUTH_ALGO_SHA1: {
+		struct hmac_sha1_key key;
+		__be32 *istate = (__be32 *)hash->sha.state1;
+		__be32 *ostate = (__be32 *)(hash->sha.state1 +
+					    round_up(sizeof(key.istate.h), 8));
+
+		hmac_sha1_preparekey(&key, auth_key, auth_keylen);
+		for (int i = 0; i < ARRAY_SIZE(key.istate.h); i++) {
+			istate[i] = cpu_to_be32(key.istate.h[i]);
+			ostate[i] = cpu_to_be32(key.ostate.h[i]);
+		}
+		memzero_explicit(&key, sizeof(key));
+		return 0;
+	}
+	case ICP_QAT_HW_AUTH_ALGO_SHA256: {
+		struct hmac_sha256_key key;
+		__be32 *istate = (__be32 *)hash->sha.state1;
+		__be32 *ostate = (__be32 *)(hash->sha.state1 +
+					    sizeof(key.key.istate.h));
+
+		hmac_sha256_preparekey(&key, auth_key, auth_keylen);
+		for (int i = 0; i < ARRAY_SIZE(key.key.istate.h); i++) {
+			istate[i] = cpu_to_be32(key.key.istate.h[i]);
+			ostate[i] = cpu_to_be32(key.key.ostate.h[i]);
+		}
+		memzero_explicit(&key, sizeof(key));
+		return 0;
+	}
+	case ICP_QAT_HW_AUTH_ALGO_SHA512: {
+		struct hmac_sha512_key key;
+		__be64 *istate = (__be64 *)hash->sha.state1;
+		__be64 *ostate = (__be64 *)(hash->sha.state1 +
+					    sizeof(key.key.istate.h));
+
+		hmac_sha512_preparekey(&key, auth_key, auth_keylen);
+		for (int i = 0; i < ARRAY_SIZE(key.key.istate.h); i++) {
+			istate[i] = cpu_to_be64(key.key.istate.h[i]);
+			ostate[i] = cpu_to_be64(key.key.ostate.h[i]);
+		}
+		memzero_explicit(&key, sizeof(key));
+		return 0;
+	}
 	default:
 		return -EFAULT;
 	}
-
-	if (crypto_shash_init(shash))
-		return -EFAULT;
-
-	if (crypto_shash_update(shash, ctx->opad, block_size))
-		return -EFAULT;
-
-	offset = round_up(qat_get_inter_state_size(ctx->qat_hash_alg), 8);
-	if (offset < 0)
-		return -EFAULT;
-
-	hash_state_out = (__be32 *)(hash->sha.state1 + offset);
-	hash512_state_out = (__be64 *)hash_state_out;
-
-	switch (ctx->qat_hash_alg) {
-	case ICP_QAT_HW_AUTH_ALGO_SHA1:
-		if (crypto_shash_export_core(shash, &ctx->sha1))
-			return -EFAULT;
-		for (i = 0; i < digest_size >> 2; i++, hash_state_out++)
-			*hash_state_out = cpu_to_be32(ctx->sha1.state[i]);
-		break;
-	case ICP_QAT_HW_AUTH_ALGO_SHA256:
-		if (crypto_shash_export_core(shash, &ctx->sha256))
-			return -EFAULT;
-		for (i = 0; i < digest_size >> 2; i++, hash_state_out++)
-			*hash_state_out = cpu_to_be32(ctx->sha256.state[i]);
-		break;
-	case ICP_QAT_HW_AUTH_ALGO_SHA512:
-		if (crypto_shash_export_core(shash, &ctx->sha512))
-			return -EFAULT;
-		for (i = 0; i < digest_size >> 3; i++, hash512_state_out++)
-			*hash512_state_out = cpu_to_be64(ctx->sha512.state[i]);
-		break;
-	default:
-		return -EFAULT;
-	}
-	memzero_explicit(ctx->ipad, block_size);
-	memzero_explicit(ctx->opad, block_size);
-	return 0;
 }
 
 static void qat_alg_init_common_hdr(struct icp_qat_fw_comn_req_hdr *header)
···
 		ICP_QAT_HW_AUTH_CONFIG_BUILD(ICP_QAT_HW_AUTH_MODE1,
 					     ctx->qat_hash_alg, digestsize);
 	hash->sha.inner_setup.auth_counter.counter =
-		cpu_to_be32(crypto_shash_blocksize(ctx->hash_tfm));
+		cpu_to_be32(ctx->hash_blocksize);
 
 	if (qat_alg_do_precomputes(hash, ctx, keys->authkey, keys->authkeylen))
 		return -EFAULT;
···
 	struct icp_qat_hw_cipher_algo_blk *cipher =
 		(struct icp_qat_hw_cipher_algo_blk *)((char *)dec_ctx +
 		sizeof(struct icp_qat_hw_auth_setup) +
-		roundup(crypto_shash_digestsize(ctx->hash_tfm), 8) * 2);
+		roundup(ctx->hash_digestsize, 8) * 2);
 	struct icp_qat_fw_la_bulk_req *req_tmpl = &ctx->dec_fw_req;
 	struct icp_qat_fw_comn_req_hdr_cd_pars *cd_pars = &req_tmpl->cd_pars;
 	struct icp_qat_fw_comn_req_hdr *header = &req_tmpl->comn_hdr;
···
 					     ctx->qat_hash_alg,
 					     digestsize);
 	hash->sha.inner_setup.auth_counter.counter =
-		cpu_to_be32(crypto_shash_blocksize(ctx->hash_tfm));
+		cpu_to_be32(ctx->hash_blocksize);
 
 	if (qat_alg_do_precomputes(hash, ctx, keys->authkey, keys->authkeylen))
 		return -EFAULT;
···
 	cipher_cd_ctrl->cipher_state_sz = AES_BLOCK_SIZE >> 3;
 	cipher_cd_ctrl->cipher_cfg_offset =
 		(sizeof(struct icp_qat_hw_auth_setup) +
-		 roundup(crypto_shash_digestsize(ctx->hash_tfm), 8) * 2) >> 3;
+		 roundup(ctx->hash_digestsize, 8) * 2) >> 3;
 	ICP_QAT_FW_COMN_CURR_ID_SET(cipher_cd_ctrl, ICP_QAT_FW_SLICE_CIPHER);
 	ICP_QAT_FW_COMN_NEXT_ID_SET(cipher_cd_ctrl, ICP_QAT_FW_SLICE_DRAM_WR);
 
···
 }
 
 static int qat_alg_aead_init(struct crypto_aead *tfm,
-			     enum icp_qat_hw_auth_algo hash,
-			     const char *hash_name)
+			     enum icp_qat_hw_auth_algo hash_alg,
+			     unsigned int hash_digestsize,
+			     unsigned int hash_blocksize)
 {
 	struct qat_alg_aead_ctx *ctx = crypto_aead_ctx(tfm);
 
-	ctx->hash_tfm = crypto_alloc_shash(hash_name, 0, 0);
-	if (IS_ERR(ctx->hash_tfm))
-		return PTR_ERR(ctx->hash_tfm);
-	ctx->qat_hash_alg = hash;
+	ctx->qat_hash_alg = hash_alg;
+	ctx->hash_digestsize = hash_digestsize;
+	ctx->hash_blocksize = hash_blocksize;
 	crypto_aead_set_reqsize(tfm, sizeof(struct qat_crypto_request));
 	return 0;
 }
 
 static int qat_alg_aead_sha1_init(struct crypto_aead *tfm)
 {
-	return qat_alg_aead_init(tfm, ICP_QAT_HW_AUTH_ALGO_SHA1, "sha1");
+	return qat_alg_aead_init(tfm, ICP_QAT_HW_AUTH_ALGO_SHA1,
+				 SHA1_DIGEST_SIZE, SHA1_BLOCK_SIZE);
 }
 
 static int qat_alg_aead_sha256_init(struct crypto_aead *tfm)
 {
-	return qat_alg_aead_init(tfm, ICP_QAT_HW_AUTH_ALGO_SHA256, "sha256");
+	return qat_alg_aead_init(tfm, ICP_QAT_HW_AUTH_ALGO_SHA256,
+				 SHA256_DIGEST_SIZE, SHA256_BLOCK_SIZE);
 }
 
 static int qat_alg_aead_sha512_init(struct crypto_aead *tfm)
 {
-	return qat_alg_aead_init(tfm, ICP_QAT_HW_AUTH_ALGO_SHA512, "sha512");
+	return qat_alg_aead_init(tfm, ICP_QAT_HW_AUTH_ALGO_SHA512,
+				 SHA512_DIGEST_SIZE, SHA512_BLOCK_SIZE);
 }
 
 static void qat_alg_aead_exit(struct crypto_aead *tfm)
···
 	struct qat_alg_aead_ctx *ctx = crypto_aead_ctx(tfm);
 	struct qat_crypto_instance *inst = ctx->inst;
 	struct device *dev;
-
-	crypto_free_shash(ctx->hash_tfm);
 
 	if (!inst)
 		return;

drivers/crypto/intel/qat/qat_common/qat_uclo.c

···
 	if (sobj_hdr)
 		sobj_chunk_num = sobj_hdr->num_chunks;
 
-	mobj_hdr = kzalloc((uobj_chunk_num + sobj_chunk_num) *
+	mobj_hdr = kcalloc(size_add(uobj_chunk_num, sobj_chunk_num),
 			   sizeof(*mobj_hdr), GFP_KERNEL);
 	if (!mobj_hdr)
 		return -ENOMEM;

drivers/crypto/marvell/octeontx2/otx2_cptpf_ucode.c

···
 		return -EINVAL;
 	}
 	err_msg = "Invalid engine group format";
-	strscpy(tmp_buf, ctx->val.vstr, strlen(ctx->val.vstr) + 1);
+	strscpy(tmp_buf, ctx->val.vstr);
 	start = tmp_buf;
 
 	has_se = has_ie = has_ae = false;

drivers/crypto/nx/nx-common-powernv.c

···
 	.base.cra_priority	= 300,
 	.base.cra_module	= THIS_MODULE,
 
-	.alloc_ctx		= nx842_powernv_crypto_alloc_ctx,
-	.free_ctx		= nx842_crypto_free_ctx,
+	.streams		= {
+		.alloc_ctx	= nx842_powernv_crypto_alloc_ctx,
+		.free_ctx	= nx842_crypto_free_ctx,
+	},
 	.compress		= nx842_crypto_compress,
 	.decompress		= nx842_crypto_decompress,
 };

drivers/crypto/nx/nx-common-pseries.c

···
 	.base.cra_priority	= 300,
 	.base.cra_module	= THIS_MODULE,
 
-	.alloc_ctx		= nx842_pseries_crypto_alloc_ctx,
-	.free_ctx		= nx842_crypto_free_ctx,
+	.streams		= {
+		.alloc_ctx	= nx842_pseries_crypto_alloc_ctx,
+		.free_ctx	= nx842_crypto_free_ctx,
+	},
 	.compress		= nx842_crypto_compress,
 	.decompress		= nx842_crypto_decompress,
 };

drivers/crypto/omap-aes.c

···
 #include <linux/pm_runtime.h>
 #include <linux/scatterlist.h>
 #include <linux/string.h>
+#include <linux/workqueue.h>
 
 #include "omap-crypto.h"
 #include "omap-aes.h"
···
 	struct omap_aes_dev *dd = data;
 
 	/* dma_lch_out - completed */
-	tasklet_schedule(&dd->done_task);
+	queue_work(system_bh_wq, &dd->done_task);
 }
 
 static int omap_aes_dma_init(struct omap_aes_dev *dd)
···
 		((u32 *)ivbuf)[i] = omap_aes_read(dd, AES_REG_IV(dd, i));
 }
 
-static void omap_aes_done_task(unsigned long data)
+static void omap_aes_done_task(struct work_struct *t)
 {
-	struct omap_aes_dev *dd = (struct omap_aes_dev *)data;
+	struct omap_aes_dev *dd = from_work(dd, t, done_task);
 
 	pr_debug("enter done_task\n");
 
···
 
 		if (!dd->total)
 			/* All bytes read! */
-			tasklet_schedule(&dd->done_task);
+			queue_work(system_bh_wq, &dd->done_task);
 		else
 			/* Enable DATA_IN interrupt for next block */
 			omap_aes_write(dd, AES_REG_IRQ_ENABLE(dd), 0x2);
···
 		 (reg & dd->pdata->major_mask) >> dd->pdata->major_shift,
 		 (reg & dd->pdata->minor_mask) >> dd->pdata->minor_shift);
 
-	tasklet_init(&dd->done_task, omap_aes_done_task, (unsigned long)dd);
+	INIT_WORK(&dd->done_task, omap_aes_done_task);
 
 	err = omap_aes_dma_init(dd);
 	if (err == -EPROBE_DEFER) {
···
 
 	omap_aes_dma_cleanup(dd);
 err_irq:
-	tasklet_kill(&dd->done_task);
+	cancel_work_sync(&dd->done_task);
 err_pm_disable:
 	pm_runtime_disable(dev);
 err_res:
···
 
 	crypto_engine_exit(dd->engine);
 
-	tasklet_kill(&dd->done_task);
+	cancel_work_sync(&dd->done_task);
 	omap_aes_dma_cleanup(dd);
 	pm_runtime_disable(dd->dev);
 }

drivers/crypto/omap-aes.h

···
 	unsigned long		flags;
 	int			err;
 
-	struct tasklet_struct	done_task;
+	struct work_struct	done_task;
 	struct aead_queue	aead_queue;
 	spinlock_t		lock;
 

drivers/crypto/omap-des.c

···
 #include <linux/pm_runtime.h>
 #include <linux/scatterlist.h>
 #include <linux/string.h>
+#include <linux/workqueue.h>
 
 #include "omap-crypto.h"
 
···
 	unsigned long		flags;
 	int			err;
 
-	struct tasklet_struct	done_task;
+	struct work_struct	done_task;
 
 	struct skcipher_request	*req;
 	struct crypto_engine		*engine;
···
 	struct omap_des_dev *dd = data;
 
 	/* dma_lch_out - completed */
-	tasklet_schedule(&dd->done_task);
+	queue_work(system_bh_wq, &dd->done_task);
 }
 
 static int omap_des_dma_init(struct omap_des_dev *dd)
···
 	       omap_des_crypt_dma_start(dd);
 }
 
-static void omap_des_done_task(unsigned long data)
+static void omap_des_done_task(struct work_struct *t)
 {
-	struct omap_des_dev *dd = (struct omap_des_dev *)data;
+	struct omap_des_dev *dd = from_work(dd, t, done_task);
 	int i;
 
 	pr_debug("enter done_task\n");
···
 
 		if (!dd->total)
 			/* All bytes read! */
-			tasklet_schedule(&dd->done_task);
+			queue_work(system_bh_wq, &dd->done_task);
 		else
 			/* Enable DATA_IN interrupt for next block */
 			omap_des_write(dd, DES_REG_IRQ_ENABLE(dd), 0x2);
···
 		 (reg & dd->pdata->major_mask) >> dd->pdata->major_shift,
 		 (reg & dd->pdata->minor_mask) >> dd->pdata->minor_shift);
 
-	tasklet_init(&dd->done_task, omap_des_done_task, (unsigned long)dd);
+	INIT_WORK(&dd->done_task, omap_des_done_task);
 
 	err = omap_des_dma_init(dd);
 	if (err == -EPROBE_DEFER) {
···
 
 	omap_des_dma_cleanup(dd);
 err_irq:
-	tasklet_kill(&dd->done_task);
+	cancel_work_sync(&dd->done_task);
 err_get:
 	pm_runtime_disable(dev);
 err_res:
···
 			crypto_engine_unregister_skcipher(
 					&dd->pdata->algs_info[i].algs_list[j]);
 
-	tasklet_kill(&dd->done_task);
+	cancel_work_sync(&dd->done_task);
 	omap_des_dma_cleanup(dd);
 	pm_runtime_disable(dd->dev);
 }

drivers/crypto/omap-sham.c

···
 #include <linux/scatterlist.h>
 #include <linux/slab.h>
 #include <linux/string.h>
+#include <linux/workqueue.h>
 
 #define MD5_DIGEST_SIZE			16
 
···
 	int			irq;
 	int			err;
 	struct dma_chan		*dma_lch;
-	struct tasklet_struct	done_task;
+	struct work_struct	done_task;
 	u8			polling_mode;
 	u8			xmit_buf[BUFLEN] OMAP_ALIGNED;
 
···
 	struct omap_sham_dev *dd = param;
 
 	set_bit(FLAGS_DMA_READY, &dd->flags);
-	tasklet_schedule(&dd->done_task);
+	queue_work(system_bh_wq, &dd->done_task);
 }
 
 static int omap_sham_xmit_dma(struct omap_sham_dev *dd, size_t length,
···
 },
 };
 
-static void omap_sham_done_task(unsigned long data)
+static void omap_sham_done_task(struct work_struct *t)
 {
-	struct omap_sham_dev *dd = (struct omap_sham_dev *)data;
+	struct omap_sham_dev *dd = from_work(dd, t, done_task);
 	int err = 0;
 
 	dev_dbg(dd->dev, "%s: flags=%lx\n", __func__, dd->flags);
···
 static irqreturn_t omap_sham_irq_common(struct omap_sham_dev *dd)
 {
 	set_bit(FLAGS_OUTPUT_READY, &dd->flags);
-	tasklet_schedule(&dd->done_task);
+	queue_work(system_bh_wq, &dd->done_task);
 
 	return IRQ_HANDLED;
 }
···
 	platform_set_drvdata(pdev, dd);
 
 	INIT_LIST_HEAD(&dd->list);
-	tasklet_init(&dd->done_task, omap_sham_done_task, (unsigned long)dd);
+	INIT_WORK(&dd->done_task, omap_sham_done_task);
 	crypto_init_queue(&dd->queue, OMAP_SHAM_QUEUE_LENGTH);
 
 	err = (dev->of_node) ? omap_sham_get_res_of(dd, dev, &res) :
···
 					&dd->pdata->algs_info[i].algs_list[j]);
 			dd->pdata->algs_info[i].registered--;
 		}
-	tasklet_kill(&dd->done_task);
+	cancel_work_sync(&dd->done_task);
 	pm_runtime_dont_use_autosuspend(&pdev->dev);
 	pm_runtime_disable(&pdev->dev);
 

drivers/crypto/rockchip/rk3288_crypto_ahash.c

···
 	struct rk_ahash_rctx *rctx = ahash_request_ctx(areq);
 	struct rk_crypto_info *rkc = rctx->dev;
 
-	dma_unmap_sg(rkc->dev, areq->src, rctx->nrsg, DMA_TO_DEVICE);
+	dma_unmap_sg(rkc->dev, areq->src, sg_nents(areq->src), DMA_TO_DEVICE);
 }
 
 static int rk_hash_run(struct crypto_engine *engine, void *breq)

drivers/crypto/starfive/jh7110-aes.c

···
 		     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");
+				return -ENOMEM;
 
 			dst_nents = src_nents;
 			len = min(sg_dma_len(stsg), remain);
···
 		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");
+				return -ENOMEM;
 
 			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");
+				return -ENOMEM;
 
 			len = min(sg_dma_len(stsg), sg_dma_len(dtsg));
 			len = min(len, remain);
···
 	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");
+			return -ENOMEM;
 
 		if (sg_copy_to_buffer(req->src, sg_nents_for_len(req->src, cryp->assoclen),
 				      rctx->adata, cryp->assoclen) != cryp->assoclen)

drivers/crypto/starfive/jh7110-hash.c

···
 	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");
+			return -ENOMEM;
 
 		ret = starfive_hash_dma_xfer(cryp, tsg);
 		dma_unmap_sg(cryp->dev, tsg, 1, DMA_TO_DEVICE);

drivers/crypto/stm32/stm32-cryp.c

···
 module_platform_driver(stm32_cryp_driver);
 
 MODULE_AUTHOR("Fabien Dessenne <[e-mail]>");
-MODULE_DESCRIPTION("STMicrolectronics STM32 CRYP hardware driver");
+MODULE_DESCRIPTION("STMicroelectronics STM32 CRYP hardware driver");
 MODULE_LICENSE("GPL");

drivers/crypto/tegra/tegra-se-hash.c

···
 	struct tegra_sha_ctx *ctx = crypto_ahash_ctx(crypto_ahash_reqtfm(req));
 	struct tegra_sha_reqctx *rctx = ahash_request_ctx(req);
 	struct tegra_se *se = ctx->se;
-	unsigned int nblks, nresidue, size, ret;
+	unsigned int nblks, nresidue, size;
 	u32 *cpuvaddr = se->cmdbuf->addr;
+	int ret;
 
 	nresidue = (req->nbytes + rctx->residue.size) % rctx->blk_size;
 	nblks = (req->nbytes + rctx->residue.size) / rctx->blk_size;

drivers/crypto/tegra/tegra-se-main.c

···
 
 	se->engine = crypto_engine_alloc_init(dev, 0);
 	if (!se->engine)
-		return dev_err_probe(dev, -ENOMEM, "failed to init crypto engine\n");
+		return -ENOMEM;
 
 	ret = crypto_engine_start(se->engine);
 	if (ret) {

drivers/crypto/ti/Kconfig

···
+# SPDX-License-Identifier: GPL-2.0-only
+config CRYPTO_DEV_TI_DTHEV2
+	tristate "Support for TI DTHE V2 cryptography engine"
+	depends on ARCH_K3 || COMPILE_TEST
+	select CRYPTO_ENGINE
+	select CRYPTO_SKCIPHER
+	select CRYPTO_ECB
+	select CRYPTO_CBC
+	help
+	  This enables support for the TI DTHE V2 hw cryptography engine
+	  which can be found on TI K3 SOCs. Selecting this enables use
+	  of hardware offloading for cryptographic algorithms on
+	  these devices, providing enhanced resistance against side-channel
+	  attacks.

drivers/crypto/ti/Makefile

···
+# SPDX-License-Identifier: GPL-2.0-only
+obj-$(CONFIG_CRYPTO_DEV_TI_DTHEV2) += dthev2.o
+dthev2-objs := dthev2-common.o dthev2-aes.o

drivers/crypto/ti/dthev2-aes.c

···
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * K3 DTHE V2 crypto accelerator driver
+ *
+ * Copyright (C) Texas Instruments 2025 - https://www.ti.com
+ * Author: T Pratham <[e-mail]>
+ */
+
+#include <crypto/aead.h>
+#include <crypto/aes.h>
+#include <crypto/algapi.h>
+#include <crypto/engine.h>
+#include <crypto/internal/aead.h>
+#include <crypto/internal/skcipher.h>
+
+#include "dthev2-common.h"
+
+#include <linux/delay.h>
+#include <linux/dmaengine.h>
+#include <linux/dma-mapping.h>
+#include <linux/io.h>
+#include <linux/scatterlist.h>
+
+/* Registers */
+
+// AES Engine
+#define DTHE_P_AES_BASE		0x7000
+#define DTHE_P_AES_KEY1_0	0x0038
+#define DTHE_P_AES_KEY1_1	0x003C
+#define DTHE_P_AES_KEY1_2	0x0030
+#define DTHE_P_AES_KEY1_3	0x0034
+#define DTHE_P_AES_KEY1_4	0x0028
+#define DTHE_P_AES_KEY1_5	0x002C
+#define DTHE_P_AES_KEY1_6	0x0020
+#define DTHE_P_AES_KEY1_7	0x0024
+#define DTHE_P_AES_IV_IN_0	0x0040
+#define DTHE_P_AES_IV_IN_1	0x0044
+#define DTHE_P_AES_IV_IN_2	0x0048
+#define DTHE_P_AES_IV_IN_3	0x004C
+#define DTHE_P_AES_CTRL		0x0050
+#define DTHE_P_AES_C_LENGTH_0	0x0054
+#define DTHE_P_AES_C_LENGTH_1	0x0058
+#define DTHE_P_AES_AUTH_LENGTH	0x005C
+#define DTHE_P_AES_DATA_IN_OUT	0x0060
+
+#define DTHE_P_AES_SYSCONFIG	0x0084
+#define DTHE_P_AES_IRQSTATUS	0x008C
+#define DTHE_P_AES_IRQENABLE	0x0090
+
+/* Register write values and macros */
+
+enum aes_ctrl_mode_masks {
+	AES_CTRL_ECB_MASK = 0x00,
+	AES_CTRL_CBC_MASK = BIT(5),
+};
+
+#define DTHE_AES_CTRL_MODE_CLEAR_MASK		~GENMASK(28, 5)
+
+#define DTHE_AES_CTRL_DIR_ENC			BIT(2)
+
+#define DTHE_AES_CTRL_KEYSIZE_16B		BIT(3)
+#define DTHE_AES_CTRL_KEYSIZE_24B		BIT(4)
+#define DTHE_AES_CTRL_KEYSIZE_32B		(BIT(3) | BIT(4))
+
+#define DTHE_AES_CTRL_SAVE_CTX_SET		BIT(29)
+
+#define DTHE_AES_CTRL_OUTPUT_READY		BIT_MASK(0)
+#define DTHE_AES_CTRL_INPUT_READY		BIT_MASK(1)
+#define DTHE_AES_CTRL_SAVED_CTX_READY		BIT_MASK(30)
+#define DTHE_AES_CTRL_CTX_READY			BIT_MASK(31)
+
+#define DTHE_AES_SYSCONFIG_DMA_DATA_IN_OUT_EN	GENMASK(6, 5)
+#define DTHE_AES_IRQENABLE_EN_ALL		GENMASK(3, 0)
+
+/* Misc */
+#define AES_IV_SIZE				AES_BLOCK_SIZE
+#define AES_BLOCK_WORDS				(AES_BLOCK_SIZE / sizeof(u32))
+#define AES_IV_WORDS				AES_BLOCK_WORDS
+
+static int dthe_cipher_init_tfm(struct crypto_skcipher *tfm)
+{
+	struct dthe_tfm_ctx *ctx = crypto_skcipher_ctx(tfm);
+	struct dthe_data *dev_data = dthe_get_dev(ctx);
+
+	ctx->dev_data = dev_data;
+	ctx->keylen = 0;
+
+	return 0;
+}
+
+static int dthe_aes_setkey(struct crypto_skcipher *tfm, const u8 *key, unsigned int keylen)
+{
+	struct dthe_tfm_ctx *ctx = crypto_skcipher_ctx(tfm);
+
+	if (keylen != AES_KEYSIZE_128 && keylen != AES_KEYSIZE_192 && keylen != AES_KEYSIZE_256)
+		return -EINVAL;
+
+	ctx->keylen = keylen;
+	memcpy(ctx->key, key, keylen);
+
+	return 0;
+}
+
+static int dthe_aes_ecb_setkey(struct crypto_skcipher *tfm, const u8 *key, unsigned int keylen)
+{
+	struct dthe_tfm_ctx *ctx = crypto_skcipher_ctx(tfm);
+
+	ctx->aes_mode = DTHE_AES_ECB;
+
+	return dthe_aes_setkey(tfm, key, keylen);
+}
+
+static int dthe_aes_cbc_setkey(struct crypto_skcipher *tfm, const u8 *key, unsigned int keylen)
+{
+	struct dthe_tfm_ctx *ctx = crypto_skcipher_ctx(tfm);
+
+	ctx->aes_mode = DTHE_AES_CBC;
+
+	return dthe_aes_setkey(tfm, key, keylen);
+}
+
+static void dthe_aes_set_ctrl_key(struct dthe_tfm_ctx *ctx,
+				  struct dthe_aes_req_ctx *rctx,
+				  u32 *iv_in)
+{
+	struct dthe_data *dev_data = dthe_get_dev(ctx);
+	void __iomem *aes_base_reg = dev_data->regs + DTHE_P_AES_BASE;
+	u32 ctrl_val = 0;
+
+	writel_relaxed(ctx->key[0], aes_base_reg + DTHE_P_AES_KEY1_0);
+	writel_relaxed(ctx->key[1], aes_base_reg + DTHE_P_AES_KEY1_1);
+	writel_relaxed(ctx->key[2], aes_base_reg + DTHE_P_AES_KEY1_2);
+	writel_relaxed(ctx->key[3], aes_base_reg + DTHE_P_AES_KEY1_3);
+
+	if (ctx->keylen > AES_KEYSIZE_128) {
+		writel_relaxed(ctx->key[4], aes_base_reg + DTHE_P_AES_KEY1_4);
+		writel_relaxed(ctx->key[5], aes_base_reg + DTHE_P_AES_KEY1_5);
+	}
+	if (ctx->keylen == AES_KEYSIZE_256) {
+		writel_relaxed(ctx->key[6], aes_base_reg + DTHE_P_AES_KEY1_6);
+		writel_relaxed(ctx->key[7], aes_base_reg + DTHE_P_AES_KEY1_7);
+	}
+
+	if (rctx->enc)
+		ctrl_val |= DTHE_AES_CTRL_DIR_ENC;
+
+	if (ctx->keylen == AES_KEYSIZE_128)
+		ctrl_val |= DTHE_AES_CTRL_KEYSIZE_16B;
+	else if (ctx->keylen == AES_KEYSIZE_192)
+		ctrl_val |= DTHE_AES_CTRL_KEYSIZE_24B;
+	else
+		ctrl_val |= DTHE_AES_CTRL_KEYSIZE_32B;
+
+	// Write AES mode
+	ctrl_val &= DTHE_AES_CTRL_MODE_CLEAR_MASK;
+	switch (ctx->aes_mode) {
+	case DTHE_AES_ECB:
+		ctrl_val |= AES_CTRL_ECB_MASK;
+		break;
+	case DTHE_AES_CBC:
+		ctrl_val |= AES_CTRL_CBC_MASK;
+		break;
+	}
+
+	if (iv_in) {
+		ctrl_val |= DTHE_AES_CTRL_SAVE_CTX_SET;
+		for (int i = 0; i < AES_IV_WORDS; ++i)
+			writel_relaxed(iv_in[i],
+				       aes_base_reg + DTHE_P_AES_IV_IN_0 + (DTHE_REG_SIZE * i));
+	}
+
+	writel_relaxed(ctrl_val, aes_base_reg + DTHE_P_AES_CTRL);
+}
+
+static void dthe_aes_dma_in_callback(void *data)
+{
+	struct skcipher_request *req = (struct skcipher_request *)data;
+	struct dthe_aes_req_ctx *rctx = skcipher_request_ctx(req);
+
+	complete(&rctx->aes_compl);
+}
+
+static int dthe_aes_run(struct crypto_engine *engine, void *areq)
+{
+	struct skcipher_request *req = container_of(areq, struct skcipher_request, base);
+	struct dthe_tfm_ctx *ctx = crypto_skcipher_ctx(crypto_skcipher_reqtfm(req));
+	struct dthe_data *dev_data = dthe_get_dev(ctx);
+	struct dthe_aes_req_ctx *rctx = skcipher_request_ctx(req);
+
+	unsigned int len = req->cryptlen;
+	struct scatterlist *src = req->src;
+	struct scatterlist *dst = req->dst;
+
+	int src_nents = sg_nents_for_len(src, len);
+	int dst_nents;
+
+	int src_mapped_nents;
+	int dst_mapped_nents;
+
+	bool diff_dst;
+	enum dma_data_direction src_dir, dst_dir;
+
+	struct device *tx_dev, *rx_dev;
+	struct dma_async_tx_descriptor *desc_in, *desc_out;
+
+	int ret;
+
+	void __iomem *aes_base_reg = dev_data->regs + DTHE_P_AES_BASE;
+
+	u32 aes_irqenable_val = readl_relaxed(aes_base_reg + DTHE_P_AES_IRQENABLE);
+	u32 aes_sysconfig_val = readl_relaxed(aes_base_reg + DTHE_P_AES_SYSCONFIG);
+
+	aes_sysconfig_val |= DTHE_AES_SYSCONFIG_DMA_DATA_IN_OUT_EN;
+	writel_relaxed(aes_sysconfig_val, aes_base_reg + DTHE_P_AES_SYSCONFIG);
+
+	aes_irqenable_val |= DTHE_AES_IRQENABLE_EN_ALL;
+	writel_relaxed(aes_irqenable_val, aes_base_reg + DTHE_P_AES_IRQENABLE);
+
+	if (src == dst) {
+		diff_dst = false;
+		src_dir = DMA_BIDIRECTIONAL;
+		dst_dir = DMA_BIDIRECTIONAL;
+	} else {
+		diff_dst = true;
+		src_dir = DMA_TO_DEVICE;
+		dst_dir  = DMA_FROM_DEVICE;
+	}
+
+	tx_dev = dmaengine_get_dma_device(dev_data->dma_aes_tx);
+	rx_dev = dmaengine_get_dma_device(dev_data->dma_aes_rx);
+
+	src_mapped_nents = dma_map_sg(tx_dev, src, src_nents, src_dir);
+	if (src_mapped_nents == 0) {
+		ret = -EINVAL;
+		goto aes_err;
+	}
+
+	if (!diff_dst) {
+		dst_nents = src_nents;
+		dst_mapped_nents = src_mapped_nents;
+	} else {
+		dst_nents = sg_nents_for_len(dst, len);
+		dst_mapped_nents = dma_map_sg(rx_dev, dst, dst_nents, dst_dir);
+		if (dst_mapped_nents == 0) {
+			dma_unmap_sg(tx_dev, src, src_nents, src_dir);
+			ret = -EINVAL;
+			goto aes_err;
+		}
+	}
+
+	desc_in = dmaengine_prep_slave_sg(dev_data->dma_aes_rx, dst, dst_mapped_nents,
+					  DMA_DEV_TO_MEM, DMA_PREP_INTERRUPT | DMA_CTRL_ACK);
+	if (!desc_in) {
+		dev_err(dev_data->dev, "IN prep_slave_sg() failed\n");
+		ret = -EINVAL;
+		goto aes_prep_err;
+	}
+
+	desc_out = dmaengine_prep_slave_sg(dev_data->dma_aes_tx, src, src_mapped_nents,
+					   DMA_MEM_TO_DEV, DMA_PREP_INTERRUPT | DMA_CTRL_ACK);
+	if (!desc_out) {
+		dev_err(dev_data->dev, "OUT prep_slave_sg() failed\n");
+		ret = -EINVAL;
+		goto aes_prep_err;
+	}
+
+	desc_in->callback = dthe_aes_dma_in_callback;
+	desc_in->callback_param = req;
+
+	init_completion(&rctx->aes_compl);
+
+	if (ctx->aes_mode == DTHE_AES_ECB)
+		dthe_aes_set_ctrl_key(ctx, rctx, NULL);
+	else
+		dthe_aes_set_ctrl_key(ctx, rctx, (u32 *)req->iv);
+
+	writel_relaxed(lower_32_bits(req->cryptlen), aes_base_reg + DTHE_P_AES_C_LENGTH_0);
+	writel_relaxed(upper_32_bits(req->cryptlen), aes_base_reg + DTHE_P_AES_C_LENGTH_1);
+
+	dmaengine_submit(desc_in);
+	dmaengine_submit(desc_out);
+
+	dma_async_issue_pending(dev_data->dma_aes_rx);
+	dma_async_issue_pending(dev_data->dma_aes_tx);
+
+	// Need to do a timeout to ensure finalise gets called if DMA callback fails for any reason
+	ret = wait_for_completion_timeout(&rctx->aes_compl, msecs_to_jiffies(DTHE_DMA_TIMEOUT_MS));
+	if (!ret) {
+		ret = -ETIMEDOUT;
+		dmaengine_terminate_sync(dev_data->dma_aes_rx);
+		dmaengine_terminate_sync(dev_data->dma_aes_tx);
+
+		for (int i = 0; i < AES_BLOCK_WORDS; ++i)
+			readl_relaxed(aes_base_reg + DTHE_P_AES_DATA_IN_OUT + (DTHE_REG_SIZE * i));
+	} else {
+		ret = 0;
+	}
+
+	// For modes other than ECB, read IV_OUT
+	if (ctx->aes_mode != DTHE_AES_ECB) {
+		u32 *iv_out = (u32 *)req->iv;
+
+		for (int i = 0; i < AES_IV_WORDS; ++i)
+			iv_out[i] = readl_relaxed(aes_base_reg +
+						  DTHE_P_AES_IV_IN_0 +
+						  (DTHE_REG_SIZE * i));
+	}
+
+aes_prep_err:
+	dma_unmap_sg(tx_dev, src, src_nents, src_dir);
+	if (dst_dir != DMA_BIDIRECTIONAL)
+		dma_unmap_sg(rx_dev, dst, dst_nents, dst_dir);
+
+aes_err:
+	local_bh_disable();
+	crypto_finalize_skcipher_request(dev_data->engine, req, ret);
+	local_bh_enable();
+	return ret;
+}
+
+static int dthe_aes_crypt(struct skcipher_request *req)
+{
+	struct dthe_tfm_ctx *ctx = crypto_skcipher_ctx(crypto_skcipher_reqtfm(req));
+	struct dthe_data *dev_data = dthe_get_dev(ctx);
+	struct crypto_engine *engine;
+
+	/*
+	 * If data is not a multiple of AES_BLOCK_SIZE, need to return -EINVAL
+	 * If data length input is zero, no need to do any operation.
+	 */
+	if (req->cryptlen % AES_BLOCK_SIZE)
+		return -EINVAL;
+
+	if (req->cryptlen == 0)
+		return 0;
+
+	engine = dev_data->engine;
+	return crypto_transfer_skcipher_request_to_engine(engine, req);
+}
+
+static int dthe_aes_encrypt(struct skcipher_request *req)
+{
+	struct dthe_aes_req_ctx *rctx = skcipher_request_ctx(req);
+
+	rctx->enc = 1;
+	return dthe_aes_crypt(req);
+}
+
+static int dthe_aes_decrypt(struct skcipher_request *req)
+{
+	struct dthe_aes_req_ctx *rctx = skcipher_request_ctx(req);
+
+	rctx->enc = 0;
+	return dthe_aes_crypt(req);
+}
+
+static struct skcipher_engine_alg cipher_algs[] = {
+	{
+		.base.init			= dthe_cipher_init_tfm,
+		.base.setkey			= dthe_aes_ecb_setkey,
+		.base.encrypt			= dthe_aes_encrypt,
+		.base.decrypt			= dthe_aes_decrypt,
+		.base.min_keysize		= AES_MIN_KEY_SIZE,
+		.base.max_keysize		= AES_MAX_KEY_SIZE,
+		.base.base = {
+			.cra_name		= "ecb(aes)",
+			.cra_driver_name	= "ecb-aes-dthev2",
+			.cra_priority		= 299,
+			.cra_flags		= CRYPTO_ALG_TYPE_SKCIPHER |
+						  CRYPTO_ALG_KERN_DRIVER_ONLY,
+			.cra_alignmask		= AES_BLOCK_SIZE - 1,
+			.cra_blocksize		= AES_BLOCK_SIZE,
+			.cra_ctxsize		= sizeof(struct dthe_tfm_ctx),
+			.cra_reqsize		= sizeof(struct dthe_aes_req_ctx),
+			.cra_module		= THIS_MODULE,
+		},
+		.op.do_one_request = dthe_aes_run,
+	}, /* ECB AES */
+	{
+		.base.init			= dthe_cipher_init_tfm,
+		.base.setkey			= dthe_aes_cbc_setkey,
+		.base.encrypt			= dthe_aes_encrypt,
+		.base.decrypt			= dthe_aes_decrypt,
+		.base.min_keysize		= AES_MIN_KEY_SIZE,
+		.base.max_keysize		= AES_MAX_KEY_SIZE,
+		.base.ivsize			= AES_IV_SIZE,
+		.base.base = {
+			.cra_name		= "cbc(aes)",
+			.cra_driver_name	= "cbc-aes-dthev2",
+			.cra_priority		= 299,
+			.cra_flags		= CRYPTO_ALG_TYPE_SKCIPHER |
+						  CRYPTO_ALG_KERN_DRIVER_ONLY,
+			.cra_alignmask		= AES_BLOCK_SIZE - 1,
+			.cra_blocksize		= AES_BLOCK_SIZE,
+			.cra_ctxsize		= sizeof(struct dthe_tfm_ctx),
+			.cra_reqsize		= sizeof(struct dthe_aes_req_ctx),
+			.cra_module		= THIS_MODULE,
+		},
+		.op.do_one_request = dthe_aes_run,
+	} /* CBC AES */
+};
+
+int dthe_register_aes_algs(void)
+{
+	return crypto_engine_register_skciphers(cipher_algs, ARRAY_SIZE(cipher_algs));
+}
+
+void dthe_unregister_aes_algs(void)
+{
+	crypto_engine_unregister_skciphers(cipher_algs, ARRAY_SIZE(cipher_algs));
+}

drivers/crypto/ti/dthev2-common.c

···
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * K3 DTHE V2 crypto accelerator driver
+ *
+ * Copyright (C) Texas Instruments 2025 - https://www.ti.com
+ * Author: T Pratham <[e-mail]>
+ */
+
+#include <crypto/aes.h>
+#include <crypto/algapi.h>
+#include <crypto/engine.h>
+#include <crypto/internal/aead.h>
+#include <crypto/internal/skcipher.h>
+
+#include "dthev2-common.h"
+
+#include <linux/delay.h>
+#include <linux/dmaengine.h>
+#include <linux/dmapool.h>
+#include <linux/dma-mapping.h>
+#include <linux/io.h>
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/mod_devicetable.h>
+#include <linux/platform_device.h>
+#include <linux/scatterlist.h>
+
+#define DRIVER_NAME	"dthev2"
+
+static struct dthe_list dthe_dev_list = {
+	.dev_list = LIST_HEAD_INIT(dthe_dev_list.dev_list),
+	.lock = __SPIN_LOCK_UNLOCKED(dthe_dev_list.lock),
+};
+
+struct dthe_data *dthe_get_dev(struct dthe_tfm_ctx *ctx)
+{
+	struct dthe_data *dev_data;
+
+	if (ctx->dev_data)
+		return ctx->dev_data;
+
+	spin_lock_bh(&dthe_dev_list.lock);
+	dev_data = list_first_entry(&dthe_dev_list.dev_list, struct dthe_data, list);
+	if (dev_data)
+		list_move_tail(&dev_data->list, &dthe_dev_list.dev_list);
+	spin_unlock_bh(&dthe_dev_list.lock);
+
+	return dev_data;
+}
+
+static int dthe_dma_init(struct dthe_data *dev_data)
+{
+	int ret;
+	struct dma_slave_config cfg;
+
+	dev_data->dma_aes_rx = NULL;
+	dev_data->dma_aes_tx = NULL;
+	dev_data->dma_sha_tx = NULL;
+
+	dev_data->dma_aes_rx = dma_request_chan(dev_data->dev, "rx");
+	if (IS_ERR(dev_data->dma_aes_rx)) {
+		return dev_err_probe(dev_data->dev, PTR_ERR(dev_data->dma_aes_rx),
+				     "Unable to request rx DMA channel\n");
+	}
+
+	dev_data->dma_aes_tx = dma_request_chan(dev_data->dev, "tx1");
+	if (IS_ERR(dev_data->dma_aes_tx)) {
+		ret = dev_err_probe(dev_data->dev, PTR_ERR(dev_data->dma_aes_tx),
+				    "Unable to request tx1 DMA channel\n");
+		goto err_dma_aes_tx;
+	}
+
+	dev_data->dma_sha_tx = dma_request_chan(dev_data->dev, "tx2");
+	if (IS_ERR(dev_data->dma_sha_tx)) {
+		ret = dev_err_probe(dev_data->dev, PTR_ERR(dev_data->dma_sha_tx),
+				    "Unable to request tx2 DMA channel\n");
+		goto err_dma_sha_tx;
+	}
+
+	memzero_explicit(&cfg, sizeof(cfg));
+
+	cfg.src_addr_width = DMA_SLAVE_BUSWIDTH_4_BYTES;
+	cfg.src_maxburst = 4;
+
+	ret = dmaengine_slave_config(dev_data->dma_aes_rx, &cfg);
+	if (ret) {
+		dev_err(dev_data->dev, "Can't configure IN dmaengine slave: %d\n", ret);
+		goto err_dma_config;
+	}
+
+	cfg.dst_addr_width = DMA_SLAVE_BUSWIDTH_4_BYTES;
+	cfg.dst_maxburst = 4;
+
+	ret = dmaengine_slave_config(dev_data->dma_aes_tx, &cfg);
+	if (ret) {
+		dev_err(dev_data->dev, "Can't configure OUT dmaengine slave: %d\n", ret);
+		goto err_dma_config;
+	}
+
+	return 0;
+
+err_dma_config:
+	dma_release_channel(dev_data->dma_sha_tx);
+err_dma_sha_tx:
+	dma_release_channel(dev_data->dma_aes_tx);
+err_dma_aes_tx:
+	dma_release_channel(dev_data->dma_aes_rx);
+
+	return ret;
+}
+
+static int dthe_register_algs(void)
+{
+	return dthe_register_aes_algs();
+}
+
+static void dthe_unregister_algs(void)
+{
+	dthe_unregister_aes_algs();
+}
+
+static int dthe_probe(struct platform_device *pdev)
+{
+	struct device *dev = &pdev->dev;
+	struct dthe_data *dev_data;
+	int ret;
+
+	dev_data = devm_kzalloc(dev, sizeof(*dev_data), GFP_KERNEL);
+	if (!dev_data)
+		return -ENOMEM;
+
+	dev_data->dev = dev;
+	dev_data->regs = devm_platform_ioremap_resource(pdev, 0);
+	if (IS_ERR(dev_data->regs))
+		return PTR_ERR(dev_data->regs);
+
+	platform_set_drvdata(pdev, dev_data);
+
+	spin_lock(&dthe_dev_list.lock);
+	list_add(&dev_data->list, &dthe_dev_list.dev_list);
+	spin_unlock(&dthe_dev_list.lock);
+
+	ret = dthe_dma_init(dev_data);
+	if (ret)
+		goto probe_dma_err;
+
+	dev_data->engine = crypto_engine_alloc_init(dev, 1);
+	if (!dev_data->engine) {
+		ret = -ENOMEM;
+		goto probe_engine_err;
+	}
+
+	ret = crypto_engine_start(dev_data->engine);
+	if (ret) {
+		dev_err(dev, "Failed to start crypto engine\n");
+		goto probe_engine_start_err;
+	}
+
+	ret = dthe_register_algs();
+	if (ret) {
+		dev_err(dev, "Failed to register algs\n");
+		goto probe_engine_start_err;
+	}
+
+	return 0;
+
+probe_engine_start_err:
+	crypto_engine_exit(dev_data->engine);
+probe_engine_err:
+	dma_release_channel(dev_data->dma_aes_rx);
+	dma_release_channel(dev_data->dma_aes_tx);
+	dma_release_channel(dev_data->dma_sha_tx);
+probe_dma_err:
+	spin_lock(&dthe_dev_list.lock);
+	list_del(&dev_data->list);
+	spin_unlock(&dthe_dev_list.lock);
+
+	return ret;
+}
+
+static void dthe_remove(struct platform_device *pdev)
+{
+	struct dthe_data *dev_data = platform_get_drvdata(pdev);
+
+	spin_lock(&dthe_dev_list.lock);
+	list_del(&dev_data->list);
+	spin_unlock(&dthe_dev_list.lock);
+
+	dthe_unregister_algs();
+
+	crypto_engine_exit(dev_data->engine);
+
+	dma_release_channel(dev_data->dma_aes_rx);
+	dma_release_channel(dev_data->dma_aes_tx);
+	dma_release_channel(dev_data->dma_sha_tx);
+}
+
+static const struct of_device_id dthe_of_match[] = {
+	{ .compatible = "ti,am62l-dthev2", },
+	{},
+};
+MODULE_DEVICE_TABLE(of, dthe_of_match);
+
+static struct platform_driver dthe_driver = {
+	.probe	= dthe_probe,
+	.remove	= dthe_remove,
+	.driver = {
+		.name		= DRIVER_NAME,
+		.of_match_table	= dthe_of_match,
+	},
+};
+
+module_platform_driver(dthe_driver);
+
+MODULE_AUTHOR("T Pratham <[e-mail]>");
+MODULE_DESCRIPTION("Texas Instruments DTHE V2 driver");
+MODULE_LICENSE("GPL");

drivers/crypto/ti/dthev2-common.h

···
+/* SPDX-License-Identifier: GPL-2.0-only */
+/*
+ * K3 DTHE V2 crypto accelerator driver
+ *
+ * Copyright (C) Texas Instruments 2025 - https://www.ti.com
+ * Author: T Pratham <[e-mail]>
+ */
+
+#ifndef __TI_DTHEV2_H__
+#define __TI_DTHEV2_H__
+
+#include <crypto/aead.h>
+#include <crypto/aes.h>
+#include <crypto/algapi.h>
+#include <crypto/engine.h>
+#include <crypto/hash.h>
+#include <crypto/internal/aead.h>
+#include <crypto/internal/hash.h>
+#include <crypto/internal/skcipher.h>
+
+#include <linux/delay.h>
+#include <linux/dmaengine.h>
+#include <linux/dmapool.h>
+#include <linux/dma-mapping.h>
+#include <linux/io.h>
+#include <linux/scatterlist.h>
+
+#define DTHE_REG_SIZE		4
+#define DTHE_DMA_TIMEOUT_MS	2000
+
+enum dthe_aes_mode {
+	DTHE_AES_ECB = 0,
+	DTHE_AES_CBC,
+};
+
+/* Driver specific struct definitions */
+
+/**
+ * struct dthe_data - DTHE_V2 driver instance data
+ * @dev: Device pointer
+ * @regs: Base address of the register space
+ * @list: list node for dev
+ * @engine: Crypto engine instance
+ * @dma_aes_rx: AES Rx DMA Channel
+ * @dma_aes_tx: AES Tx DMA Channel
+ * @dma_sha_tx: SHA Tx DMA Channel
+ */
+struct dthe_data {
+	struct device *dev;
+	void __iomem *regs;
+	struct list_head list;
+	struct crypto_engine *engine;
+
+	struct dma_chan *dma_aes_rx;
+	struct dma_chan *dma_aes_tx;
+
+	struct dma_chan *dma_sha_tx;
+};
+
+/**
+ * struct dthe_list - device data list head
+ * @dev_list: linked list head
+ * @lock: Spinlock protecting accesses to the list
+ */
+struct dthe_list {
+	struct list_head dev_list;
+	spinlock_t lock;
+};
+
+/**
+ * struct dthe_tfm_ctx - Transform ctx struct containing ctx for all sub-components of DTHE V2
+ * @dev_data: Device data struct pointer
+ * @keylen: AES key length
+ * @key: AES key
+ * @aes_mode: AES mode
+ */
+struct dthe_tfm_ctx {
+	struct dthe_data *dev_data;
+	unsigned int keylen;
+	u32 key[AES_KEYSIZE_256 / sizeof(u32)];
+	enum dthe_aes_mode aes_mode;
+};
+
+/**
+ * struct dthe_aes_req_ctx - AES engine req ctx struct
+ * @enc: flag indicating encryption or decryption operation
+ * @aes_compl: Completion variable for use in manual completion in case of DMA callback failure
+ */
+struct dthe_aes_req_ctx {
+	int enc;
+	struct completion aes_compl;
+};
+
+/* Struct definitions end */
+
+struct dthe_data *dthe_get_dev(struct dthe_tfm_ctx *ctx);
+
+int dthe_register_aes_algs(void);
+void dthe_unregister_aes_algs(void);
+
+#endif

drivers/crypto/xilinx/Makefile

···
 # SPDX-License-Identifier: GPL-2.0-only
+obj-$(CONFIG_CRYPTO_DEV_XILINX_TRNG) += xilinx-trng.o
 obj-$(CONFIG_CRYPTO_DEV_ZYNQMP_AES) += zynqmp-aes-gcm.o
 obj-$(CONFIG_CRYPTO_DEV_ZYNQMP_SHA3) += zynqmp-sha.o

drivers/crypto/xilinx/xilinx-trng.c

···
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * AMD Versal True Random Number Generator driver
+ * Copyright (c) 2024 - 2025 Advanced Micro Devices, Inc.
+ */
+
+#include <linux/bitfield.h>
+#include <linux/clk.h>
+#include <linux/crypto.h>
+#include <linux/delay.h>
+#include <linux/errno.h>
+#include <linux/firmware/xlnx-zynqmp.h>
+#include <linux/hw_random.h>
+#include <linux/io.h>
+#include <linux/iopoll.h>
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/mutex.h>
+#include <linux/mod_devicetable.h>
+#include <linux/platform_device.h>
+#include <linux/string.h>
+#include <crypto/internal/cipher.h>
+#include <crypto/internal/rng.h>
+#include <crypto/aes.h>
+
+/* TRNG Registers Offsets */
+#define TRNG_STATUS_OFFSET			0x4U
+#define TRNG_CTRL_OFFSET			0x8U
+#define TRNG_EXT_SEED_OFFSET			0x40U
+#define TRNG_PER_STRNG_OFFSET			0x80U
+#define TRNG_CORE_OUTPUT_OFFSET			0xC0U
+#define TRNG_RESET_OFFSET			0xD0U
+#define TRNG_OSC_EN_OFFSET			0xD4U
+
+/* Mask values */
+#define TRNG_RESET_VAL_MASK			BIT(0)
+#define TRNG_OSC_EN_VAL_MASK			BIT(0)
+#define TRNG_CTRL_PRNGSRST_MASK			BIT(0)
+#define TRNG_CTRL_EUMODE_MASK			BIT(8)
+#define TRNG_CTRL_TRSSEN_MASK			BIT(2)
+#define TRNG_CTRL_PRNGSTART_MASK		BIT(5)
+#define TRNG_CTRL_PRNGXS_MASK			BIT(3)
+#define TRNG_CTRL_PRNGMODE_MASK			BIT(7)
+#define TRNG_STATUS_DONE_MASK			BIT(0)
+#define TRNG_STATUS_QCNT_MASK			GENMASK(11, 9)
+#define TRNG_STATUS_QCNT_16_BYTES		0x800
+
+/* Sizes in bytes */
+#define TRNG_SEED_LEN_BYTES			48U
+#define TRNG_ENTROPY_SEED_LEN_BYTES		64U
+#define TRNG_SEC_STRENGTH_SHIFT			5U
+#define TRNG_SEC_STRENGTH_BYTES			BIT(TRNG_SEC_STRENGTH_SHIFT)
+#define TRNG_BYTES_PER_REG			4U
+#define TRNG_RESET_DELAY			10
+#define TRNG_NUM_INIT_REGS			12U
+#define TRNG_READ_4_WORD			4
+#define TRNG_DATA_READ_DELAY			8000
+
+struct xilinx_rng {
+	void __iomem *rng_base;
+	struct device *dev;
+	struct mutex lock;	/* Protect access to TRNG device */
+	struct hwrng trng;
+};
+
+struct xilinx_rng_ctx {
+	struct xilinx_rng *rng;
+};
+
+static struct xilinx_rng *xilinx_rng_dev;
+
+static void xtrng_readwrite32(void __iomem *addr, u32 mask, u8 value)
+{
+	u32 val;
+
+	val = ioread32(addr);
+	val = (val & (~mask)) | (mask & value);
+	iowrite32(val, addr);
+}
+
+static void xtrng_trng_reset(void __iomem *addr)
+{
+	xtrng_readwrite32(addr + TRNG_RESET_OFFSET, TRNG_RESET_VAL_MASK, TRNG_RESET_VAL_MASK);
+	udelay(TRNG_RESET_DELAY);
+	xtrng_readwrite32(addr + TRNG_RESET_OFFSET, TRNG_RESET_VAL_MASK, 0);
+}
+
+static void xtrng_hold_reset(void __iomem *addr)
+{
+	xtrng_readwrite32(addr + TRNG_CTRL_OFFSET, TRNG_CTRL_PRNGSRST_MASK,
+			  TRNG_CTRL_PRNGSRST_MASK);
+	iowrite32(TRNG_RESET_VAL_MASK, addr + TRNG_RESET_OFFSET);
+	udelay(TRNG_RESET_DELAY);
+}
+
+static void xtrng_softreset(struct xilinx_rng *rng)
+{
+	xtrng_readwrite32(rng->rng_base + TRNG_CTRL_OFFSET, TRNG_CTRL_PRNGSRST_MASK,
+			  TRNG_CTRL_PRNGSRST_MASK);
+	udelay(TRNG_RESET_DELAY);
+	xtrng_readwrite32(rng->rng_base + TRNG_CTRL_OFFSET, TRNG_CTRL_PRNGSRST_MASK, 0);
+}
+
+/* Return no. of bytes read */
+static size_t xtrng_readblock32(void __iomem *rng_base, __be32 *buf, int blocks32, bool wait)
+{
+	int read = 0, ret;
+	int timeout = 1;
+	int i, idx;
+	u32 val;
+
+	if (wait)
+		timeout = TRNG_DATA_READ_DELAY;
+
+	for (i = 0; i < (blocks32 * 2); i++) {
+		/* TRNG core generate data in 16 bytes. Read twice to complete 32 bytes read */
+		ret = readl_poll_timeout(rng_base + TRNG_STATUS_OFFSET, val,
+					 (val & TRNG_STATUS_QCNT_MASK) ==
+					 TRNG_STATUS_QCNT_16_BYTES, !!wait, timeout);
+		if (ret)
+			break;
+
+		for (idx = 0; idx < TRNG_READ_4_WORD; idx++) {
+			*(buf + read) = cpu_to_be32(ioread32(rng_base + TRNG_CORE_OUTPUT_OFFSET));
+			read += 1;
+		}
+	}
+	return read * 4;
+}
+
+static int xtrng_collect_random_data(struct xilinx_rng *rng, u8 *rand_gen_buf,
+				     int no_of_random_bytes, bool wait)
+{
+	u8 randbuf[TRNG_SEC_STRENGTH_BYTES];
+	int byteleft, blocks, count = 0;
+	int ret;
+
+	byteleft = no_of_random_bytes & (TRNG_SEC_STRENGTH_BYTES - 1);
+	blocks = no_of_random_bytes >> TRNG_SEC_STRENGTH_SHIFT;
+	xtrng_readwrite32(rng->rng_base + TRNG_CTRL_OFFSET, TRNG_CTRL_PRNGSTART_MASK,
+			  TRNG_CTRL_PRNGSTART_MASK);
+	if (blocks) {
+		ret = xtrng_readblock32(rng->rng_base, (__be32 *)rand_gen_buf, blocks, wait);
+		if (!ret)
+			return 0;
+		count += ret;
+	}
+
+	if (byteleft) {
+		ret = xtrng_readblock32(rng->rng_base, (__be32 *)randbuf, 1, wait);
+		if (!ret)
+			return count;
+		memcpy(rand_gen_buf + (blocks * TRNG_SEC_STRENGTH_BYTES), randbuf, byteleft);
+		count += byteleft;
+	}
+
+	xtrng_readwrite32(rng->rng_base + TRNG_CTRL_OFFSET,
+			  TRNG_CTRL_PRNGMODE_MASK | TRNG_CTRL_PRNGSTART_MASK, 0U);
+
+	return count;
+}
+
+static void xtrng_write_multiple_registers(void __iomem *base_addr, u32 *values, size_t n)
+{
+	void __iomem *reg_addr;
+	size_t i;
+
+	/* Write seed value into EXTERNAL_SEED Registers in big endian format */
+	for (i = 0; i < n; i++) {
+		reg_addr = (base_addr + ((n - 1 - i) * TRNG_BYTES_PER_REG));
+		iowrite32((u32 __force)(cpu_to_be32(values[i])), reg_addr);
+	}
+}
+
+static void xtrng_enable_entropy(struct xilinx_rng *rng)
+{
+	iowrite32(TRNG_OSC_EN_VAL_MASK, rng->rng_base + TRNG_OSC_EN_OFFSET);
+	xtrng_softreset(rng);
+	iowrite32(TRNG_CTRL_EUMODE_MASK | TRNG_CTRL_TRSSEN_MASK, rng->rng_base + TRNG_CTRL_OFFSET);
+}
+
+static int xtrng_reseed_internal(struct xilinx_rng *rng)
+{
+	u8 entropy[TRNG_ENTROPY_SEED_LEN_BYTES];
+	u32 val;
+	int ret;
+
+	memset(entropy, 0, sizeof(entropy));
+	xtrng_enable_entropy(rng);
+
+	/* collect random data to use it as entropy (input for DF) */
+	ret = xtrng_collect_random_data(rng, entropy, TRNG_SEED_LEN_BYTES, true);
+	if (ret != TRNG_SEED_LEN_BYTES)
+		return -EINVAL;
+
+	xtrng_write_multiple_registers(rng->rng_base + TRNG_EXT_SEED_OFFSET,
+				       (u32 *)entropy, TRNG_NUM_INIT_REGS);
+	/* select reseed operation */
+	iowrite32(TRNG_CTRL_PRNGXS_MASK, rng->rng_base + TRNG_CTRL_OFFSET);
+
+	/* Start the reseed operation with above configuration and wait for STATUS.Done bit to be
+	 * set. Monitor STATUS.CERTF bit, if set indicates SP800-90B entropy health test has failed.
+	 */
+	xtrng_readwrite32(rng->rng_base + TRNG_CTRL_OFFSET, TRNG_CTRL_PRNGSTART_MASK,
+			  TRNG_CTRL_PRNGSTART_MASK);
+
+	ret = readl_poll_timeout(rng->rng_base + TRNG_STATUS_OFFSET, val,
+				 (val & TRNG_STATUS_DONE_MASK) == TRNG_STATUS_DONE_MASK,
+				  1U, 15000U);
+	if (ret)
+		return ret;
+
+	xtrng_readwrite32(rng->rng_base + TRNG_CTRL_OFFSET, TRNG_CTRL_PRNGSTART_MASK, 0U);
+
+	return 0;
+}
+
+static int xtrng_random_bytes_generate(struct xilinx_rng *rng, u8 *rand_buf_ptr,
+				       u32 rand_buf_size, int wait)
+{
+	int nbytes;
+	int ret;
+
+	xtrng_readwrite32(rng->rng_base + TRNG_CTRL_OFFSET,
+			  TRNG_CTRL_PRNGMODE_MASK | TRNG_CTRL_PRNGXS_MASK,
+			  TRNG_CTRL_PRNGMODE_MASK | TRNG_CTRL_PRNGXS_MASK);
+	nbytes = xtrng_collect_random_data(rng, rand_buf_ptr, rand_buf_size, wait);
+
+	ret = xtrng_reseed_internal(rng);
+	if (ret) {
+		dev_err(rng->dev, "Re-seed fail\n");
+		return ret;
+	}
+
+	return nbytes;
+}
+
+static int xtrng_trng_generate(struct crypto_rng *tfm, const u8 *src, u32 slen,
+			       u8 *dst, u32 dlen)
+{
+	struct xilinx_rng_ctx *ctx = crypto_rng_ctx(tfm);
+	int ret;
+
+	mutex_lock(&ctx->rng->lock);
+	ret = xtrng_random_bytes_generate(ctx->rng, dst, dlen, true);
+	mutex_unlock(&ctx->rng->lock);
+
+	return ret < 0 ? ret : 0;
+}
+
+static int xtrng_trng_seed(struct crypto_rng *tfm, const u8 *seed, unsigned int slen)
+{
+	return 0;
+}
+
+static int xtrng_trng_init(struct crypto_tfm *rtfm)
+{
+	struct xilinx_rng_ctx *ctx = crypto_tfm_ctx(rtfm);
+
+	ctx->rng = xilinx_rng_dev;
+
+	return 0;
+}
+
+static struct rng_alg xtrng_trng_alg = {
+	.generate = xtrng_trng_generate,
+	.seed = xtrng_trng_seed,
+	.seedsize = 0,
+	.base = {
+		.cra_name = "stdrng",
+		.cra_driver_name = "xilinx-trng",
+		.cra_priority = 300,
+		.cra_ctxsize = sizeof(struct xilinx_rng_ctx),
+		.cra_module = THIS_MODULE,
+		.cra_init = xtrng_trng_init,
+	},
+};
+
+static int xtrng_hwrng_trng_read(struct hwrng *hwrng, void *data, size_t max, bool wait)
+{
+	u8 buf[TRNG_SEC_STRENGTH_BYTES];
+	struct xilinx_rng *rng;
+	int ret = -EINVAL, i = 0;
+
+	rng = container_of(hwrng, struct xilinx_rng, trng);
+	/* Return in case wait not set and lock not available. */
+	if (!mutex_trylock(&rng->lock) && !wait)
+		return 0;
+	else if (!mutex_is_locked(&rng->lock) && wait)
+		mutex_lock(&rng->lock);
+
+	while (i < max) {
+		ret = xtrng_random_bytes_generate(rng, buf, TRNG_SEC_STRENGTH_BYTES, wait);
+		if (ret < 0)
+			break;
+
+		memcpy(data + i, buf, min_t(int, ret, (max - i)));
+		i += min_t(int, ret, (max - i));
+	}
+	mutex_unlock(&rng->lock);
+
+	return ret;
+}
+
+static int xtrng_hwrng_register(struct hwrng *trng)
+{
+	int ret;
+
+	trng->name = "Xilinx Versal Crypto Engine TRNG";
+	trng->read = xtrng_hwrng_trng_read;
+
+	ret = hwrng_register(trng);
+	if (ret)
+		pr_err("Fail to register the TRNG\n");
+
+	return ret;
+}
+
+static void xtrng_hwrng_unregister(struct hwrng *trng)
+{
+	hwrng_unregister(trng);
+}
+
+static int xtrng_probe(struct platform_device *pdev)
+{
+	struct xilinx_rng *rng;
+	int ret;
+
+	rng = devm_kzalloc(&pdev->dev, sizeof(*rng), GFP_KERNEL);
+	if (!rng)
+		return -ENOMEM;
+
+	rng->dev = &pdev->dev;
+	rng->rng_base = devm_platform_ioremap_resource(pdev, 0);
+	if (IS_ERR(rng->rng_base)) {
+		dev_err(&pdev->dev, "Failed to map resource %ld\n", PTR_ERR(rng->rng_base));
+		return PTR_ERR(rng->rng_base);
+	}
+
+	xtrng_trng_reset(rng->rng_base);
+	ret = xtrng_reseed_internal(rng);
+	if (ret) {
+		dev_err(&pdev->dev, "TRNG Seed fail\n");
+		return ret;
+	}
+
+	xilinx_rng_dev = rng;
+	mutex_init(&rng->lock);
+	ret = crypto_register_rng(&xtrng_trng_alg);
+	if (ret) {
+		dev_err(&pdev->dev, "Crypto Random device registration failed: %d\n", ret);
+		return ret;
+	}
+	ret = xtrng_hwrng_register(&rng->trng);
+	if (ret) {
+		dev_err(&pdev->dev, "HWRNG device registration failed: %d\n", ret);
+		goto crypto_rng_free;
+	}
+	platform_set_drvdata(pdev, rng);
+
+	return 0;
+
+crypto_rng_free:
+	crypto_unregister_rng(&xtrng_trng_alg);
+
+	return ret;
+}
+
+static void xtrng_remove(struct platform_device *pdev)
+{
+	struct xilinx_rng *rng;
+	u32 zero[TRNG_NUM_INIT_REGS] = { };
+
+	rng = platform_get_drvdata(pdev);
+	xtrng_hwrng_unregister(&rng->trng);
+	crypto_unregister_rng(&xtrng_trng_alg);
+	xtrng_write_multiple_registers(rng->rng_base + TRNG_EXT_SEED_OFFSET, zero,
+				       TRNG_NUM_INIT_REGS);
+	xtrng_write_multiple_registers(rng->rng_base + TRNG_PER_STRNG_OFFSET, zero,
+				       TRNG_NUM_INIT_REGS);
+	xtrng_hold_reset(rng->rng_base);
+	xilinx_rng_dev = NULL;
+}
+
+static const struct of_device_id xtrng_of_match[] = {
+	{ .compatible = "xlnx,versal-trng", },
+	{},
+};
+
+MODULE_DEVICE_TABLE(of, xtrng_of_match);
+
+static struct platform_driver xtrng_driver = {
+	.driver = {
+		.name = "xlnx,versal-trng",
+		.of_match_table	= xtrng_of_match,
+	},
+	.probe = xtrng_probe,
+	.remove = xtrng_remove,
+};
+
+module_platform_driver(xtrng_driver);
+MODULE_LICENSE("GPL");
+MODULE_AUTHOR("Harsh Jain <[e-mail]>");
+MODULE_AUTHOR("Mounika Botcha <[e-mail]>");
+MODULE_DESCRIPTION("True Random Number Generator Driver");

include/crypto/hash.h

···
 
 #define HASH_MAX_DIGESTSIZE	 64
 
+/*
+ * The size of a core hash state and a partial block.  The final byte
+ * is the length of the partial block.
+ */
+#define HASH_STATE_AND_BLOCK(state, block) ((state) + (block) + 1)
+
+
 /* Worst case is sha3-224. */
-#define HASH_MAX_STATESIZE	 200 + 144 + 1
+#define HASH_MAX_STATESIZE	 HASH_STATE_AND_BLOCK(200, 144)
+
+/* This needs to match arch/s390/crypto/sha.h. */
+#define S390_SHA_CTX_SIZE	216
 
 /*
  * Worst case is hmac(sha3-224-s390).  Its context is a nested 'shash_desc'
  * containing a 'struct s390_sha_ctx'.
  */
-#define HASH_MAX_DESCSIZE	(sizeof(struct shash_desc) + 361)
+#define SHA3_224_S390_DESCSIZE	HASH_STATE_AND_BLOCK(S390_SHA_CTX_SIZE, 144)
+#define HASH_MAX_DESCSIZE	(sizeof(struct shash_desc) + \
+				 SHA3_224_S390_DESCSIZE)
 #define MAX_SYNC_HASH_REQSIZE	(sizeof(struct ahash_request) + \
 				 HASH_MAX_DESCSIZE)
 

include/crypto/internal/scompress.h

···
 /**
  * struct scomp_alg - synchronous compression algorithm
  *
- * @alloc_ctx:	Function allocates algorithm specific context
- * @free_ctx:	Function frees context allocated with alloc_ctx
  * @compress:	Function performs a compress operation
  * @decompress:	Function performs a de-compress operation
- * @base:	Common crypto API algorithm data structure
  * @streams:	Per-cpu memory for algorithm
  * @calg:	Cmonn algorithm data structure shared with acomp
  */
···
 			  unsigned int slen, u8 *dst, unsigned int *dlen,
 			  void *ctx);
 
-	union {
-		struct {
-			void *(*alloc_ctx)(void);
-			void (*free_ctx)(void *ctx);
-		};
-		struct crypto_acomp_streams streams;
-	};
+	struct crypto_acomp_streams streams;
 
 	union {
 		struct COMP_ALG_COMMON;

include/linux/hisi_acc_qm.h

···
 #define UACCE_MODE_SVA			1 /* use uacce sva mode */
 #define UACCE_MODE_DESC	"0(default) means only register to crypto, 1 means both register to crypto and uacce"
 
+#define QM_ECC_MBIT			BIT(2)
+
 enum qm_stop_reason {
 	QM_NORMAL,
 	QM_SOFT_RESET,
···
 	QM_HW_V2 = 0x21,
 	QM_HW_V3 = 0x30,
 	QM_HW_V4 = 0x50,
+	QM_HW_V5 = 0x51,
 };
 
 enum qm_fun_type {
···
 	ACC_ERR_RECOVERED,
 };
 
-struct hisi_qm_err_info {
-	char *acpi_rst;
-	u32 msi_wr_port;
+struct hisi_qm_err_mask {
 	u32 ecc_2bits_mask;
-	u32 qm_shutdown_mask;
-	u32 dev_shutdown_mask;
-	u32 qm_reset_mask;
-	u32 dev_reset_mask;
+	u32 shutdown_mask;
+	u32 reset_mask;
 	u32 ce;
 	u32 nfe;
 	u32 fe;
+};
+
+struct hisi_qm_err_info {
+	char *acpi_rst;
+	u32 msi_wr_port;
+	struct hisi_qm_err_mask qm_err;
+	struct hisi_qm_err_mask dev_err;
 };
 
 struct hisi_qm_err_status {
···
 	enum acc_err_result (*get_err_result)(struct hisi_qm *qm);
 	bool (*dev_is_abnormal)(struct hisi_qm *qm);
 	int (*set_priv_status)(struct hisi_qm *qm);
+	void (*disable_axi_error)(struct hisi_qm *qm);
+	void (*enable_axi_error)(struct hisi_qm *qm);
 };
 
 struct hisi_qm_cap_info {

include/linux/psp-sev.h

···
 	SEV_CMD_SNP_DOWNLOAD_FIRMWARE_EX = 0x0CA,
 	SEV_CMD_SNP_COMMIT		= 0x0CB,
 	SEV_CMD_SNP_VLEK_LOAD		= 0x0CD,
+	SEV_CMD_SNP_FEATURE_INFO	= 0x0CE,
 
 	SEV_CMD_MAX,
 };
···
 struct sev_data_snp_init_ex {
 	u32 init_rmp:1;
 	u32 list_paddr_en:1;
-	u32 rsvd:30;
+	u32 rapl_dis:1;
+	u32 ciphertext_hiding_en:1;
+	u32 rsvd:28;
 	u32 rsvd1;
 	u64 list_paddr;
-	u8  rsvd2[48];
+	u16 max_snp_asid;
+	u8  rsvd2[46];
 } __packed;
 
 /**
···
  * @probe: True if this is being called as part of CCP module probe, which
  *  will defer SEV_INIT/SEV_INIT_EX firmware initialization until needed
  *  unless psp_init_on_probe module param is set
+ * @max_snp_asid: When non-zero, enable ciphertext hiding and specify the
+ *  maximum ASID that can be used for an SEV-SNP guest.
  */
 struct sev_platform_init_args {
 	int error;
 	bool probe;
+	unsigned int max_snp_asid;
 };
 
 /**
···
 struct sev_data_snp_commit {
 	u32 len;
 } __packed;
+
+/**
+ * struct sev_data_snp_feature_info - SEV_SNP_FEATURE_INFO structure
+ *
+ * @length: len of the command buffer read by the PSP
+ * @ecx_in: subfunction index
+ * @feature_info_paddr : System Physical Address of the FEATURE_INFO structure
+ */
+struct sev_data_snp_feature_info {
+	u32 length;
+	u32 ecx_in;
+	u64 feature_info_paddr;
+} __packed;
+
+/**
+ * struct feature_info - FEATURE_INFO structure
+ *
+ * @eax: output of SNP_FEATURE_INFO command
+ * @ebx: output of SNP_FEATURE_INFO command
+ * @ecx: output of SNP_FEATURE_INFO command
+ * #edx: output of SNP_FEATURE_INFO command
+ */
+struct snp_feature_info {
+	u32 eax;
+	u32 ebx;
+	u32 ecx;
+	u32 edx;
+} __packed;
+
+#define SNP_CIPHER_TEXT_HIDING_SUPPORTED	BIT(3)
 
 #ifdef CONFIG_CRYPTO_DEV_SP_PSP
 
···
 void *snp_alloc_firmware_page(gfp_t mask);
 void snp_free_firmware_page(void *addr);
 void sev_platform_shutdown(void);
+bool sev_is_snp_ciphertext_hiding_supported(void);
 
 #else	/* !CONFIG_CRYPTO_DEV_SP_PSP */
 
···
 static inline void snp_free_firmware_page(void *addr) { }
 
 static inline void sev_platform_shutdown(void) { }
+
+static inline bool sev_is_snp_ciphertext_hiding_supported(void) { return false; }
 
 #endif	/* CONFIG_CRYPTO_DEV_SP_PSP */
 

include/linux/rcupdate.h

···
 				(c) || rcu_read_lock_sched_held(), \
 				__rcu)
 
+/**
+ * rcu_dereference_all_check() - rcu_dereference_all with debug checking
+ * @p: The pointer to read, prior to dereferencing
+ * @c: The conditions under which the dereference will take place
+ *
+ * This is similar to rcu_dereference_check(), but allows protection
+ * by all forms of vanilla RCU readers, including preemption disabled,
+ * bh-disabled, and interrupt-disabled regions of code.  Note that "vanilla
+ * RCU" excludes SRCU and the various Tasks RCU flavors.  Please note
+ * that this macro should not be backported to any Linux-kernel version
+ * preceding v5.0 due to changes in synchronize_rcu() semantics prior
+ * to that version.
+ */
+#define rcu_dereference_all_check(p, c) \
+	__rcu_dereference_check((p), __UNIQUE_ID(rcu), \
+				(c) || rcu_read_lock_any_held(), \
+				__rcu)
+
 /*
  * The tracing infrastructure traces RCU (we want that), but unfortunately
  * some of the RCU checks causes tracing to lock up the system.
···
  * Makes rcu_dereference_check() do the dirty work.
  */
 #define rcu_dereference_sched(p) rcu_dereference_sched_check(p, 0)
+
+/**
+ * rcu_dereference_all() - fetch RCU-all-protected pointer for dereferencing
+ * @p: The pointer to read, prior to dereferencing
+ *
+ * Makes rcu_dereference_check() do the dirty work.
+ */
+#define rcu_dereference_all(p) rcu_dereference_all_check(p, 0)
 
 /**
  * rcu_pointer_handoff() - Hand off a pointer from RCU to other mechanism

include/linux/rhashtable.h

···
 	return hash & (tbl->size - 1);
 }
 
-static inline unsigned int rht_key_get_hash(struct rhashtable *ht,
+static __always_inline unsigned int rht_key_get_hash(struct rhashtable *ht,
 	const void *key, const struct rhashtable_params params,
 	unsigned int hash_rnd)
 {
···
 	return hash;
 }
 
-static inline unsigned int rht_key_hashfn(
+static __always_inline unsigned int rht_key_hashfn(
 	struct rhashtable *ht, const struct bucket_table *tbl,
 	const void *key, const struct rhashtable_params params)
 {
···
 	return rht_bucket_index(tbl, hash);
 }
 
-static inline unsigned int rht_head_hashfn(
+static __always_inline unsigned int rht_head_hashfn(
 	struct rhashtable *ht, const struct bucket_table *tbl,
 	const struct rhash_head *he, const struct rhashtable_params params)
 {
···
 	rcu_dereference_protected(p, lockdep_rht_mutex_is_held(ht))
 
 #define rht_dereference_rcu(p, ht) \
-	rcu_dereference_check(p, lockdep_rht_mutex_is_held(ht))
+	rcu_dereference_all_check(p, lockdep_rht_mutex_is_held(ht))
 
 #define rht_dereference_bucket(p, tbl, hash) \
 	rcu_dereference_protected(p, lockdep_rht_bucket_is_held(tbl, hash))
 
 #define rht_dereference_bucket_rcu(p, tbl, hash) \
-	rcu_dereference_check(p, lockdep_rht_bucket_is_held(tbl, hash))
+	rcu_dereference_all_check(p, lockdep_rht_bucket_is_held(tbl, hash))
 
 #define rht_entry(tpos, pos, member) \
 	({ tpos = container_of(pos, typeof(*tpos), member); 1; })
···
 static inline struct rhash_head *rht_ptr_rcu(
 	struct rhash_lock_head __rcu *const *bkt)
 {
-	return __rht_ptr(rcu_dereference(*bkt), bkt);
+	return __rht_ptr(rcu_dereference_all(*bkt), bkt);
 }
 
 static inline struct rhash_head *rht_ptr(
···
 	for (({barrier(); }),						\
 	     pos = head;						\
 	     !rht_is_a_nulls(pos);					\
-	     pos = rcu_dereference_raw(pos->next))
+	     pos = rcu_dereference_all(pos->next))
 
 /**
  * rht_for_each_rcu - iterate over rcu hash chain
···
 	for (({barrier(); }),					\
 	     pos = rht_ptr_rcu(rht_bucket(tbl, hash));		\
 	     !rht_is_a_nulls(pos);				\
-	     pos = rcu_dereference_raw(pos->next))
+	     pos = rcu_dereference_all(pos->next))
 
 /**
  * rht_for_each_entry_rcu_from - iterated over rcu hash chain from given head
···
  * list returned by rhltable_lookup.
  */
 #define rhl_for_each_rcu(pos, list)					\
-	for (pos = list; pos; pos = rcu_dereference_raw(pos->next))
+	for (pos = list; pos; pos = rcu_dereference_all(pos->next))
 
 /**
  * rhl_for_each_entry_rcu - iterate over rcu hash table list of given type
···
  */
 #define rhl_for_each_entry_rcu(tpos, pos, list, member)			\
 	for (pos = list; pos && rht_entry(tpos, pos, member);		\
-	     pos = rcu_dereference_raw(pos->next))
+	     pos = rcu_dereference_all(pos->next))
 
 static inline int rhashtable_compare(struct rhashtable_compare_arg *arg,
 				     const void *obj)
···
 }
 
 /* Internal function, do not use. */
-static inline struct rhash_head *__rhashtable_lookup(
+static __always_inline struct rhash_head *__rhashtable_lookup(
 	struct rhashtable *ht, const void *key,
 	const struct rhashtable_params params)
 {
···
  *
  * Returns the first entry on which the compare function returned true.
  */
-static inline void *rhashtable_lookup(
+static __always_inline void *rhashtable_lookup(
 	struct rhashtable *ht, const void *key,
 	const struct rhashtable_params params)
 {
···
  *
  * Returns the first entry on which the compare function returned true.
  */
-static inline void *rhashtable_lookup_fast(
+static __always_inline void *rhashtable_lookup_fast(
 	struct rhashtable *ht, const void *key,
 	const struct rhashtable_params params)
 {
···
  *
  * Returns the list of entries that match the given key.
  */
-static inline struct rhlist_head *rhltable_lookup(
+static __always_inline struct rhlist_head *rhltable_lookup(
 	struct rhltable *hlt, const void *key,
 	const struct rhashtable_params params)
 {
···
  * function returns the existing element already in hashes if there is a clash,
  * otherwise it returns an error via ERR_PTR().
  */
-static inline void *__rhashtable_insert_fast(
+static __always_inline void *__rhashtable_insert_fast(
 	struct rhashtable *ht, const void *key, struct rhash_head *obj,
 	const struct rhashtable_params params, bool rhlist)
 {
···
  * Will trigger an automatic deferred table resizing if residency in the
  * table grows beyond 70%.
  */
-static inline int rhashtable_insert_fast(
+static __always_inline int rhashtable_insert_fast(
 	struct rhashtable *ht, struct rhash_head *obj,
 	const struct rhashtable_params params)
 {
···
  * Will trigger an automatic deferred table resizing if residency in the
  * table grows beyond 70%.
  */
-static inline int rhltable_insert_key(
+static __always_inline int rhltable_insert_key(
 	struct rhltable *hlt, const void *key, struct rhlist_head *list,
 	const struct rhashtable_params params)
 {
···
  * Will trigger an automatic deferred table resizing if residency in the
  * table grows beyond 70%.
  */
-static inline int rhltable_insert(
+static __always_inline int rhltable_insert(
 	struct rhltable *hlt, struct rhlist_head *list,
 	const struct rhashtable_params params)
 {
···
  * Will trigger an automatic deferred table resizing if residency in the
  * table grows beyond 70%.
  */
-static inline int rhashtable_lookup_insert_fast(
+static __always_inline int rhashtable_lookup_insert_fast(
 	struct rhashtable *ht, struct rhash_head *obj,
 	const struct rhashtable_params params)
 {
···
  * object if it exists, NULL if it did not and the insertion was successful,
  * and an ERR_PTR otherwise.
  */
-static inline void *rhashtable_lookup_get_insert_fast(
+static __always_inline void *rhashtable_lookup_get_insert_fast(
 	struct rhashtable *ht, struct rhash_head *obj,
 	const struct rhashtable_params params)
 {
···
  *
  * Returns zero on success.
  */
-static inline int rhashtable_lookup_insert_key(
+static __always_inline int rhashtable_lookup_insert_key(
 	struct rhashtable *ht, const void *key, struct rhash_head *obj,
 	const struct rhashtable_params params)
 {
···
  * object if it exists, NULL if it does not and the insertion was successful,
  * and an ERR_PTR otherwise.
  */
-static inline void *rhashtable_lookup_get_insert_key(
+static __always_inline void *rhashtable_lookup_get_insert_key(
 	struct rhashtable *ht, const void *key, struct rhash_head *obj,
 	const struct rhashtable_params params)
 {
···
 }
 
 /* Internal function, please use rhashtable_remove_fast() instead */
-static inline int __rhashtable_remove_fast_one(
+static __always_inline int __rhashtable_remove_fast_one(
 	struct rhashtable *ht, struct bucket_table *tbl,
 	struct rhash_head *obj, const struct rhashtable_params params,
 	bool rhlist)
···
 }
 
 /* Internal function, please use rhashtable_remove_fast() instead */
-static inline int __rhashtable_remove_fast(
+static __always_inline int __rhashtable_remove_fast(
 	struct rhashtable *ht, struct rhash_head *obj,
 	const struct rhashtable_params params, bool rhlist)
 {
···
  *
  * Returns zero on success, -ENOENT if the entry could not be found.
  */
-static inline int rhashtable_remove_fast(
+static __always_inline int rhashtable_remove_fast(
 	struct rhashtable *ht, struct rhash_head *obj,
 	const struct rhashtable_params params)
 {
···
  *
  * Returns zero on success, -ENOENT if the entry could not be found.
  */
-static inline int rhltable_remove(
+static __always_inline int rhltable_remove(
 	struct rhltable *hlt, struct rhlist_head *list,
 	const struct rhashtable_params params)
 {
···
 }
 
 /* Internal function, please use rhashtable_replace_fast() instead */
-static inline int __rhashtable_replace_fast(
+static __always_inline int __rhashtable_replace_fast(
 	struct rhashtable *ht, struct bucket_table *tbl,
 	struct rhash_head *obj_old, struct rhash_head *obj_new,
 	const struct rhashtable_params params)
···
  * Returns zero on success, -ENOENT if the entry could not be found,
  * -EINVAL if hash is not the same for the old and new objects.
  */
-static inline int rhashtable_replace_fast(
+static __always_inline int rhashtable_replace_fast(
 	struct rhashtable *ht, struct rhash_head *obj_old,
 	struct rhash_head *obj_new,
 	const struct rhashtable_params params)

include/uapi/linux/psp-sev.h

···
  * @mask_chip_id: whether chip id is present in attestation reports or not
  * @mask_chip_key: whether attestation reports are signed or not
  * @vlek_en: VLEK (Version Loaded Endorsement Key) hashstick is loaded
+ * @feature_info: whether SNP_FEATURE_INFO command is available
+ * @rapl_dis: whether RAPL is disabled
+ * @ciphertext_hiding_cap: whether platform has ciphertext hiding capability
+ * @ciphertext_hiding_en: whether ciphertext hiding is enabled
  * @rsvd1: reserved
  * @guest_count: the number of guest currently managed by the firmware
  * @current_tcb_version: current TCB version
···
 	__u32 mask_chip_id:1;		/* Out */
 	__u32 mask_chip_key:1;		/* Out */
 	__u32 vlek_en:1;		/* Out */
-	__u32 rsvd1:29;
+	__u32 feature_info:1;		/* Out */
+	__u32 rapl_dis:1;		/* Out */
+	__u32 ciphertext_hiding_cap:1;	/* Out */
+	__u32 ciphertext_hiding_en:1;	/* Out */
+	__u32 rsvd1:25;
 	__u32 guest_count;		/* Out */
 	__u64 current_tcb_version;	/* Out */
 	__u64 reported_tcb_version;	/* Out */

include/uapi/misc/uacce/hisi_qm.h

···
 #define HISI_QM_API_VER_BASE "hisi_qm_v1"
 #define HISI_QM_API_VER2_BASE "hisi_qm_v2"
 #define HISI_QM_API_VER3_BASE "hisi_qm_v3"
+#define HISI_QM_API_VER5_BASE "hisi_qm_v5"
 
 /* UACCE_CMD_QM_SET_QP_CTX: Set qp algorithm type */
 #define UACCE_CMD_QM_SET_QP_CTX	_IOWR('H', 10, struct hisi_qp_ctx)

kernel/padata.c

···
 		struct padata_serial_queue *squeue;
 		int cb_cpu;
 
-		cpu = cpumask_next_wrap(cpu, pd->cpumask.pcpu);
 		processed++;
+		/* When sequence wraps around, reset to the first CPU. */
+		if (unlikely(processed == 0))
+			cpu = cpumask_first(pd->cpumask.pcpu);
+		else
+			cpu = cpumask_next_wrap(cpu, pd->cpumask.pcpu);
 
 		cb_cpu = padata->cb_cpu;
 		squeue = per_cpu_ptr(pd->squeue, cb_cpu);
···
 			do {
 				nid = next_node_in(old_node, node_states[N_CPU]);
 			} while (!atomic_try_cmpxchg(&last_used_nid, &old_node, nid));
-			queue_work_node(nid, system_unbound_wq, &pw->pw_work);
+			queue_work_node(nid, system_dfl_wq, &pw->pw_work);
 		} else {
-			queue_work(system_unbound_wq, &pw->pw_work);
+			queue_work(system_dfl_wq, &pw->pw_work);
 		}
 
 	/* Use the current thread, which saves starting a workqueue worker. */
···
 
 	cpus_read_lock();
 
-	pinst->serial_wq = alloc_workqueue("%s_serial", WQ_MEM_RECLAIM |
-					   WQ_CPU_INTENSIVE, 1, name);
+	pinst->serial_wq = alloc_workqueue("%s_serial",
+					   WQ_MEM_RECLAIM | WQ_CPU_INTENSIVE | WQ_PERCPU,
+					   1, name);
 	if (!pinst->serial_wq)
 		goto err_put_cpus;
 

lib/lzo/lzo1x_compress.c

···
 #define HAVE_OP(x) 1
 #endif
 
-#define NEED_OP(x) if (!HAVE_OP(x)) goto output_overrun
+#define NEED_OP(x) if (unlikely(!HAVE_OP(x))) goto output_overrun
 
 static noinline int
 LZO_SAFE(lzo1x_1_do_compress)(const unsigned char *in, size_t in_len,

lib/lzo/lzo1x_decompress_safe.c

···
 
 #define HAVE_IP(x)      ((size_t)(ip_end - ip) >= (size_t)(x))
 #define HAVE_OP(x)      ((size_t)(op_end - op) >= (size_t)(x))
-#define NEED_IP(x)      if (!HAVE_IP(x)) goto input_overrun
-#define NEED_OP(x)      if (!HAVE_OP(x)) goto output_overrun
-#define TEST_LB(m_pos)  if ((m_pos) < out) goto lookbehind_overrun
+#define NEED_IP(x)      if (unlikely(!HAVE_IP(x))) goto input_overrun
+#define NEED_OP(x)      if (unlikely(!HAVE_OP(x))) goto output_overrun
+#define TEST_LB(m_pos)  if (unlikely((m_pos) < out)) goto lookbehind_overrun
 
 /* This MAX_255_COUNT is the maximum number of times we can add 255 to a base
  * count without overflowing an integer. The multiply will overflow when