Merge tag 'v6.2-p1' of git://git.kernel.org/pub/scm/linux/kernel/git/herbert/crypto-2.6

Pull crypto updates from Herbert Xu:
 "API:
   - Optimise away self-test overhead when they are disabled
   - Support symmetric encryption via keyring keys in af_alg
   - Flip hwrng default_quality, the default is now maximum entropy

  Algorithms:
   - Add library version of aesgcm
   - CFI fixes for assembly code
   - Add arm/arm64 accelerated versions of sm3/sm4

  Drivers:
   - Remove assumption on arm64 that kmalloc is DMA-aligned
   - Fix selftest failures in rockchip
   - Add support for RK3328/RK3399 in rockchip
   - Add deflate support in qat
   - Merge ux500 into stm32
   - Add support for TEE for PCI ID 0x14CA in ccp
   - Add mt7986 support in mtk
   - Add MaxLinear platform support in inside-secure
   - Add NPCM8XX support in npcm"

* tag 'v6.2-p1' of git://git.kernel.org/pub/scm/linux/kernel/git/herbert/crypto-2.6: (184 commits)
  crypto: ux500/cryp - delete driver
  crypto: stm32/cryp - enable for use with Ux500
  crypto: stm32 - enable drivers to be used on Ux500
  dt-bindings: crypto: Let STM32 define Ux500 CRYP
  hwrng: geode - Fix PCI device refcount leak
  hwrng: amd - Fix PCI device refcount leak
  crypto: qce - Set DMA alignment explicitly
  crypto: octeontx2 - Set DMA alignment explicitly
  crypto: octeontx - Set DMA alignment explicitly
  crypto: keembay - Set DMA alignment explicitly
  crypto: safexcel - Set DMA alignment explicitly
  crypto: hisilicon/hpre - Set DMA alignment explicitly
  crypto: chelsio - Set DMA alignment explicitly
  crypto: ccree - Set DMA alignment explicitly
  crypto: ccp - Set DMA alignment explicitly
  crypto: cavium - Set DMA alignment explicitly
  crypto: img-hash - Fix variable dereferenced before check 'hdev->req'
  crypto: arm64/ghash-ce - use frame_push/pop macros consistently
  crypto: arm64/crct10dif - use frame_push/pop macros consistently
  crypto: arm64/aes-modes - use frame_push/pop macros consistently
  ...

Documentation/crypto/devel-algos.rst

@@ -172,7 +172,7 @@ Here are schematics of how these functions are called when operated from
 other part of the kernel. Note that the .setkey() call might happen
 before or after any of these schematics happen, but must not happen
 during any of these are in-flight. Please note that calling .init()
-followed immediately by .finish() is also a perfectly valid
+followed immediately by .final() is also a perfectly valid
 transformation.
 
 ::

Documentation/crypto/userspace-if.rst

@@ -131,9 +131,9 @@ from the kernel crypto API. If the buffer is too small for the message
 digest, the flag MSG_TRUNC is set by the kernel.
 
 In order to set a message digest key, the calling application must use
-the setsockopt() option of ALG_SET_KEY. If the key is not set the HMAC
-operation is performed without the initial HMAC state change caused by
-the key.
+the setsockopt() option of ALG_SET_KEY or ALG_SET_KEY_BY_KEY_SERIAL. If the
+key is not set the HMAC operation is performed without the initial HMAC state
+change caused by the key.
 
 Symmetric Cipher API
 --------------------
@@ -382,6 +382,15 @@ mentioned optname:
 
    -  the RNG cipher type to provide the seed
 
+- ALG_SET_KEY_BY_KEY_SERIAL -- Setting the key via keyring key_serial_t.
+   This operation behaves the same as ALG_SET_KEY. The decrypted
+   data is copied from a keyring key, and uses that data as the
+   key for symmetric encryption.
+
+   The passed in key_serial_t must have the KEY_(POS|USR|GRP|OTH)_SEARCH
+   permission set, otherwise -EPERM is returned. Supports key types: user,
+   logon, encrypted, and trusted.
+
 -  ALG_SET_AEAD_AUTHSIZE -- Setting the authentication tag size for
    AEAD ciphers. For a encryption operation, the authentication tag of
    the given size will be generated. For a decryption operation, the

Documentation/devicetree/bindings/crypto/rockchip,rk3288-crypto.yaml

+# SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause)
+%YAML 1.2
+---
+$id: http://devicetree.org/schemas/crypto/rockchip,rk3288-crypto.yaml#
+$schema: http://devicetree.org/meta-schemas/core.yaml#
+
+title: Rockchip Electronics Security Accelerator
+
+maintainers:
+  - Heiko Stuebner <heiko@sntech.de>
+
+properties:
+  compatible:
+    enum:
+      - rockchip,rk3288-crypto
+      - rockchip,rk3328-crypto
+      - rockchip,rk3399-crypto
+
+  reg:
+    maxItems: 1
+
+  interrupts:
+    maxItems: 1
+
+  clocks:
+    minItems: 3
+    maxItems: 4
+
+  clock-names:
+    minItems: 3
+    maxItems: 4
+
+  resets:
+    minItems: 1
+    maxItems: 3
+
+  reset-names:
+    minItems: 1
+    maxItems: 3
+
+allOf:
+  - if:
+      properties:
+        compatible:
+          contains:
+            const: rockchip,rk3288-crypto
+    then:
+      properties:
+        clocks:
+          minItems: 4
+        clock-names:
+          items:
+            - const: aclk
+            - const: hclk
+            - const: sclk
+            - const: apb_pclk
+        resets:
+          maxItems: 1
+        reset-names:
+          items:
+            - const: crypto-rst
+  - if:
+      properties:
+        compatible:
+          contains:
+            const: rockchip,rk3328-crypto
+    then:
+      properties:
+        clocks:
+          maxItems: 3
+        clock-names:
+          items:
+            - const: hclk_master
+            - const: hclk_slave
+            - const: sclk
+        resets:
+          maxItems: 1
+        reset-names:
+          items:
+            - const: crypto-rst
+  - if:
+      properties:
+        compatible:
+          contains:
+            const: rockchip,rk3399-crypto
+    then:
+      properties:
+        clocks:
+          maxItems: 3
+        clock-names:
+          items:
+            - const: hclk_master
+            - const: hclk_slave
+            - const: sclk
+        resets:
+          minItems: 3
+        reset-names:
+          items:
+            - const: master
+            - const: slave
+            - const: crypto-rst
+
+required:
+  - compatible
+  - reg
+  - interrupts
+  - clocks
+  - clock-names
+  - resets
+  - reset-names
+
+additionalProperties: false
+
+examples:
+  - |
+    #include <dt-bindings/interrupt-controller/arm-gic.h>
+    #include <dt-bindings/clock/rk3288-cru.h>
+    crypto@ff8a0000 {
+      compatible = "rockchip,rk3288-crypto";
+      reg = <0xff8a0000 0x4000>;
+      interrupts = <GIC_SPI 48 IRQ_TYPE_LEVEL_HIGH>;
+      clocks = <&cru ACLK_CRYPTO>, <&cru HCLK_CRYPTO>,
+               <&cru SCLK_CRYPTO>, <&cru ACLK_DMAC1>;
+      clock-names = "aclk", "hclk", "sclk", "apb_pclk";
+      resets = <&cru SRST_CRYPTO>;
+      reset-names = "crypto-rst";
+    };

Documentation/devicetree/bindings/crypto/rockchip-crypto.txt

-Rockchip Electronics And Security Accelerator
-
-Required properties:
-- compatible: Should be "rockchip,rk3288-crypto"
-- reg: Base physical address of the engine and length of memory mapped
-       region
-- interrupts: Interrupt number
-- clocks: Reference to the clocks about crypto
-- clock-names: "aclk" used to clock data
-	       "hclk" used to clock data
-	       "sclk" used to clock crypto accelerator
-	       "apb_pclk" used to clock dma
-- resets: Must contain an entry for each entry in reset-names.
-	  See ../reset/reset.txt for details.
-- reset-names: Must include the name "crypto-rst".
-
-Examples:
-
-	crypto: cypto-controller@ff8a0000 {
-		compatible = "rockchip,rk3288-crypto";
-		reg = <0xff8a0000 0x4000>;
-		interrupts = <GIC_SPI 48 IRQ_TYPE_LEVEL_HIGH>;
-		clocks = <&cru ACLK_CRYPTO>, <&cru HCLK_CRYPTO>,
-			 <&cru SCLK_CRYPTO>, <&cru ACLK_DMAC1>;
-		clock-names = "aclk", "hclk", "sclk", "apb_pclk";
-		resets = <&cru SRST_CRYPTO>;
-		reset-names = "crypto-rst";
-	};

Documentation/devicetree/bindings/crypto/st,stm32-cryp.yaml

@@ -6,12 +6,18 @@ $schema: http://devicetree.org/meta-schemas/core.yaml#
 
 title: STMicroelectronics STM32 CRYP bindings
 
+description: The STM32 CRYP block is built on the CRYP block found in
+  the STn8820 SoC introduced in 2007, and subsequently used in the U8500
+  SoC in 2010.
+
 maintainers:
   - Lionel Debieve <lionel.debieve@foss.st.com>
 
 properties:
   compatible:
     enum:
+      - st,stn8820-cryp
+      - stericsson,ux500-cryp
       - st,stm32f756-cryp
       - st,stm32mp1-cryp
 
@@ -27,6 +33,19 @@ properties:
   resets:
     maxItems: 1
 
+  dmas:
+    items:
+      - description: mem2cryp DMA channel
+      - description: cryp2mem DMA channel
+
+  dma-names:
+    items:
+      - const: mem2cryp
+      - const: cryp2mem
+
+  power-domains:
+    maxItems: 1
+
 required:
   - compatible
   - reg

Documentation/devicetree/bindings/rng/nuvoton,npcm-rng.yaml

@@ -16,7 +16,9 @@ maintainers:
 
 properties:
   compatible:
-    const: nuvoton,npcm750-rng
+    enum:
+      - nuvoton,npcm750-rng
+      - nuvoton,npcm845-rng
 
   reg:
     maxItems: 1

MAINTAINERS

@@ -17941,6 +17941,13 @@ F:	Documentation/ABI/*/sysfs-driver-hid-roccat*
 F:	drivers/hid/hid-roccat*
 F:	include/linux/hid-roccat*
 
+ROCKCHIP CRYPTO DRIVERS
+M:	Corentin Labbe <clabbe@baylibre.com>
+L:	linux-crypto@vger.kernel.org
+S:	Maintained
+F:	Documentation/devicetree/bindings/crypto/rockchip,rk3288-crypto.yaml
+F:	drivers/crypto/rockchip/
+
 ROCKCHIP I2S TDM DRIVER
 M:	Nicolas Frattaroli <frattaroli.nicolas@gmail.com>
 L:	linux-rockchip@lists.infradead.org

arch/arm/crypto/Kconfig

@@ -18,7 +18,7 @@ config CRYPTO_GHASH_ARM_CE
 	depends on KERNEL_MODE_NEON
 	select CRYPTO_HASH
 	select CRYPTO_CRYPTD
-	select CRYPTO_GF128MUL
+	select CRYPTO_LIB_GF128MUL
 	help
 	  GCM GHASH function (NIST SP800-38D)
 

arch/arm/crypto/aes-cipher-glue.c

@@ -7,7 +7,7 @@
  */
 
 #include <crypto/aes.h>
-#include <linux/crypto.h>
+#include <crypto/algapi.h>
 #include <linux/module.h>
 
 asmlinkage void __aes_arm_encrypt(u32 *rk, int rounds, const u8 *in, u8 *out);

arch/arm/crypto/nh-neon-core.S

@@ -69,7 +69,7 @@
 
 /*
  * void nh_neon(const u32 *key, const u8 *message, size_t message_len,
- *		u8 hash[NH_HASH_BYTES])
+ *		__le64 hash[NH_NUM_PASSES])
  *
  * It's guaranteed that message_len % 16 == 0.
  */

arch/arm/crypto/nhpoly1305-neon-glue.c

@@ -14,14 +14,7 @@
 #include <linux/module.h>
 
 asmlinkage void nh_neon(const u32 *key, const u8 *message, size_t message_len,
-			u8 hash[NH_HASH_BYTES]);
-
-/* wrapper to avoid indirect call to assembly, which doesn't work with CFI */
-static void _nh_neon(const u32 *key, const u8 *message, size_t message_len,
-		     __le64 hash[NH_NUM_PASSES])
-{
-	nh_neon(key, message, message_len, (u8 *)hash);
-}
+			__le64 hash[NH_NUM_PASSES]);
 
 static int nhpoly1305_neon_update(struct shash_desc *desc,
 				  const u8 *src, unsigned int srclen)
@@ -33,7 +26,7 @@ static int nhpoly1305_neon_update(struct shash_desc *desc,
 		unsigned int n = min_t(unsigned int, srclen, SZ_4K);
 
 		kernel_neon_begin();
-		crypto_nhpoly1305_update_helper(desc, src, n, _nh_neon);
+		crypto_nhpoly1305_update_helper(desc, src, n, nh_neon);
 		kernel_neon_end();
 		src += n;
 		srclen -= n;

arch/arm64/crypto/Kconfig

@@ -6,8 +6,8 @@ config CRYPTO_GHASH_ARM64_CE
 	tristate "Hash functions: GHASH (ARMv8 Crypto Extensions)"
 	depends on KERNEL_MODE_NEON
 	select CRYPTO_HASH
-	select CRYPTO_GF128MUL
 	select CRYPTO_LIB_AES
+	select CRYPTO_LIB_GF128MUL
 	select CRYPTO_AEAD
 	help
 	  GCM GHASH function (NIST SP800-38D)
@@ -96,6 +96,17 @@ config CRYPTO_SHA3_ARM64
 	  Architecture: arm64 using:
 	  - ARMv8.2 Crypto Extensions
 
+config CRYPTO_SM3_NEON
+	tristate "Hash functions: SM3 (NEON)"
+	depends on KERNEL_MODE_NEON
+	select CRYPTO_HASH
+	select CRYPTO_SM3
+	help
+	  SM3 (ShangMi 3) secure hash function (OSCCA GM/T 0004-2012)
+
+	  Architecture: arm64 using:
+	  - NEON (Advanced SIMD) extensions
+
 config CRYPTO_SM3_ARM64_CE
 	tristate "Hash functions: SM3 (ARMv8.2 Crypto Extensions)"
 	depends on KERNEL_MODE_NEON
@@ -220,7 +231,7 @@ config CRYPTO_SM4_ARM64_CE
 	  - NEON (Advanced SIMD) extensions
 
 config CRYPTO_SM4_ARM64_CE_BLK
-	tristate "Ciphers: SM4, modes: ECB/CBC/CFB/CTR (ARMv8 Crypto Extensions)"
+	tristate "Ciphers: SM4, modes: ECB/CBC/CFB/CTR/XTS (ARMv8 Crypto Extensions)"
 	depends on KERNEL_MODE_NEON
 	select CRYPTO_SKCIPHER
 	select CRYPTO_SM4
@@ -231,6 +242,8 @@ config CRYPTO_SM4_ARM64_CE_BLK
 	  - CBC (Cipher Block Chaining) mode (NIST SP800-38A)
 	  - CFB (Cipher Feedback) mode (NIST SP800-38A)
 	  - CTR (Counter) mode (NIST SP800-38A)
+	  - XTS (XOR Encrypt XOR with ciphertext stealing) mode (NIST SP800-38E
+	    and IEEE 1619)
 
 	  Architecture: arm64 using:
 	  - ARMv8 Crypto Extensions
@@ -268,6 +281,38 @@ config CRYPTO_AES_ARM64_CE_CCM
 	  - ARMv8 Crypto Extensions
 	  - NEON (Advanced SIMD) extensions
 
+config CRYPTO_SM4_ARM64_CE_CCM
+	tristate "AEAD cipher: SM4 in CCM mode (ARMv8 Crypto Extensions)"
+	depends on KERNEL_MODE_NEON
+	select CRYPTO_ALGAPI
+	select CRYPTO_AEAD
+	select CRYPTO_SM4
+	select CRYPTO_SM4_ARM64_CE_BLK
+	help
+	  AEAD cipher: SM4 cipher algorithms (OSCCA GB/T 32907-2016) with
+	  CCM (Counter with Cipher Block Chaining-Message Authentication Code)
+	  authenticated encryption mode (NIST SP800-38C)
+
+	  Architecture: arm64 using:
+	  - ARMv8 Crypto Extensions
+	  - NEON (Advanced SIMD) extensions
+
+config CRYPTO_SM4_ARM64_CE_GCM
+	tristate "AEAD cipher: SM4 in GCM mode (ARMv8 Crypto Extensions)"
+	depends on KERNEL_MODE_NEON
+	select CRYPTO_ALGAPI
+	select CRYPTO_AEAD
+	select CRYPTO_SM4
+	select CRYPTO_SM4_ARM64_CE_BLK
+	help
+	  AEAD cipher: SM4 cipher algorithms (OSCCA GB/T 32907-2016) with
+	  GCM (Galois/Counter Mode) authenticated encryption mode (NIST SP800-38D)
+
+	  Architecture: arm64 using:
+	  - ARMv8 Crypto Extensions
+	  - PMULL (Polynomial Multiply Long) instructions
+	  - NEON (Advanced SIMD) extensions
+
 config CRYPTO_CRCT10DIF_ARM64_CE
 	tristate "CRCT10DIF (PMULL)"
 	depends on KERNEL_MODE_NEON && CRC_T10DIF

arch/arm64/crypto/Makefile

@@ -17,6 +17,9 @@ sha512-ce-y := sha512-ce-glue.o sha512-ce-core.o
 obj-$(CONFIG_CRYPTO_SHA3_ARM64) += sha3-ce.o
 sha3-ce-y := sha3-ce-glue.o sha3-ce-core.o
 
+obj-$(CONFIG_CRYPTO_SM3_NEON) += sm3-neon.o
+sm3-neon-y := sm3-neon-glue.o sm3-neon-core.o
+
 obj-$(CONFIG_CRYPTO_SM3_ARM64_CE) += sm3-ce.o
 sm3-ce-y := sm3-ce-glue.o sm3-ce-core.o
 
@@ -26,6 +29,12 @@ sm4-ce-cipher-y := sm4-ce-cipher-glue.o sm4-ce-cipher-core.o
 obj-$(CONFIG_CRYPTO_SM4_ARM64_CE_BLK) += sm4-ce.o
 sm4-ce-y := sm4-ce-glue.o sm4-ce-core.o
 
+obj-$(CONFIG_CRYPTO_SM4_ARM64_CE_CCM) += sm4-ce-ccm.o
+sm4-ce-ccm-y := sm4-ce-ccm-glue.o sm4-ce-ccm-core.o
+
+obj-$(CONFIG_CRYPTO_SM4_ARM64_CE_GCM) += sm4-ce-gcm.o
+sm4-ce-gcm-y := sm4-ce-gcm-glue.o sm4-ce-gcm-core.o
+
 obj-$(CONFIG_CRYPTO_SM4_ARM64_NEON_BLK) += sm4-neon.o
 sm4-neon-y := sm4-neon-glue.o sm4-neon-core.o
 

arch/arm64/crypto/aes-ce-glue.c

@@ -9,9 +9,9 @@
 #include <asm/simd.h>
 #include <asm/unaligned.h>
 #include <crypto/aes.h>
+#include <crypto/algapi.h>
 #include <crypto/internal/simd.h>
 #include <linux/cpufeature.h>
-#include <linux/crypto.h>
 #include <linux/module.h>
 
 #include "aes-ce-setkey.h"

arch/arm64/crypto/aes-cipher-glue.c

@@ -6,7 +6,7 @@
  */
 
 #include <crypto/aes.h>
-#include <linux/crypto.h>
+#include <crypto/algapi.h>
 #include <linux/module.h>
 
 asmlinkage void __aes_arm64_encrypt(u32 *rk, u8 *out, const u8 *in, int rounds);

arch/arm64/crypto/aes-modes.S

@@ -52,8 +52,7 @@ SYM_FUNC_END(aes_decrypt_block5x)
 	 */
 
 AES_FUNC_START(aes_ecb_encrypt)
-	stp		x29, x30, [sp, #-16]!
-	mov		x29, sp
+	frame_push	0
 
 	enc_prepare	w3, x2, x5
 
@@ -77,14 +76,13 @@ ST5(	st1		{v4.16b}, [x0], #16		)
 	subs		w4, w4, #1
 	bne		.Lecbencloop
 .Lecbencout:
-	ldp		x29, x30, [sp], #16
+	frame_pop
 	ret
 AES_FUNC_END(aes_ecb_encrypt)
 
 
 AES_FUNC_START(aes_ecb_decrypt)
-	stp		x29, x30, [sp, #-16]!
-	mov		x29, sp
+	frame_push	0
 
 	dec_prepare	w3, x2, x5
 
@@ -108,7 +106,7 @@ ST5(	st1		{v4.16b}, [x0], #16		)
 	subs		w4, w4, #1
 	bne		.Lecbdecloop
 .Lecbdecout:
-	ldp		x29, x30, [sp], #16
+	frame_pop
 	ret
 AES_FUNC_END(aes_ecb_decrypt)
 
@@ -171,9 +169,6 @@ AES_FUNC_END(aes_cbc_encrypt)
 AES_FUNC_END(aes_essiv_cbc_encrypt)
 
 AES_FUNC_START(aes_essiv_cbc_decrypt)
-	stp		x29, x30, [sp, #-16]!
-	mov		x29, sp
-
 	ld1		{cbciv.16b}, [x5]		/* get iv */
 
 	mov		w8, #14				/* AES-256: 14 rounds */
@@ -182,11 +177,9 @@ AES_FUNC_START(aes_essiv_cbc_decrypt)
 	b		.Lessivcbcdecstart
 
 AES_FUNC_START(aes_cbc_decrypt)
-	stp		x29, x30, [sp, #-16]!
-	mov		x29, sp
-
 	ld1		{cbciv.16b}, [x5]		/* get iv */
 .Lessivcbcdecstart:
+	frame_push	0
 	dec_prepare	w3, x2, x6
 
 .LcbcdecloopNx:
@@ -236,7 +229,7 @@ ST5(	st1		{v4.16b}, [x0], #16		)
 	bne		.Lcbcdecloop
 .Lcbcdecout:
 	st1		{cbciv.16b}, [x5]		/* return iv */
-	ldp		x29, x30, [sp], #16
+	frame_pop
 	ret
 AES_FUNC_END(aes_cbc_decrypt)
 AES_FUNC_END(aes_essiv_cbc_decrypt)
@@ -337,8 +330,7 @@ AES_FUNC_END(aes_cbc_cts_decrypt)
 	BLOCKS		.req x13
 	BLOCKS_W	.req w13
 
-	stp		x29, x30, [sp, #-16]!
-	mov		x29, sp
+	frame_push	0
 
 	enc_prepare	ROUNDS_W, KEY, IV_PART
 	ld1		{vctr.16b}, [IV]
@@ -481,7 +473,7 @@ ST5(	st1		{v4.16b}, [OUT], #16		)
 	.if !\xctr
 		st1		{vctr.16b}, [IV] /* return next CTR value */
 	.endif
-	ldp		x29, x30, [sp], #16
+	frame_pop
 	ret
 
 .Lctrtail\xctr:
@@ -645,8 +637,7 @@ AES_FUNC_END(aes_xctr_encrypt)
 	.endm
 
 AES_FUNC_START(aes_xts_encrypt)
-	stp		x29, x30, [sp, #-16]!
-	mov		x29, sp
+	frame_push	0
 
 	ld1		{v4.16b}, [x6]
 	xts_load_mask	v8
@@ -704,7 +695,7 @@ AES_FUNC_START(aes_xts_encrypt)
 	st1		{v0.16b}, [x0]
 .Lxtsencret:
 	st1		{v4.16b}, [x6]
-	ldp		x29, x30, [sp], #16
+	frame_pop
 	ret
 
 .LxtsencctsNx:
@@ -732,8 +723,7 @@ AES_FUNC_START(aes_xts_encrypt)
 AES_FUNC_END(aes_xts_encrypt)
 
 AES_FUNC_START(aes_xts_decrypt)
-	stp		x29, x30, [sp, #-16]!
-	mov		x29, sp
+	frame_push	0
 
 	/* subtract 16 bytes if we are doing CTS */
 	sub		w8, w4, #0x10
@@ -794,7 +784,7 @@ AES_FUNC_START(aes_xts_decrypt)
 	b		.Lxtsdecloop
 .Lxtsdecout:
 	st1		{v4.16b}, [x6]
-	ldp		x29, x30, [sp], #16
+	frame_pop
 	ret
 
 .Lxtsdeccts:

arch/arm64/crypto/aes-neonbs-core.S

@@ -760,7 +760,7 @@ SYM_FUNC_START_LOCAL(__xts_crypt8)
 	eor		v6.16b, v6.16b, v31.16b
 	eor		v7.16b, v7.16b, v16.16b
 
-	stp		q16, q17, [sp, #16]
+	stp		q16, q17, [x6]
 
 	mov		bskey, x2
 	mov		rounds, x3
@@ -768,8 +768,8 @@ SYM_FUNC_START_LOCAL(__xts_crypt8)
 SYM_FUNC_END(__xts_crypt8)
 
 	.macro		__xts_crypt, do8, o0, o1, o2, o3, o4, o5, o6, o7
-	stp		x29, x30, [sp, #-48]!
-	mov		x29, sp
+	frame_push	0, 32
+	add		x6, sp, #.Lframe_local_offset
 
 	ld1		{v25.16b}, [x5]
 
@@ -781,7 +781,7 @@ SYM_FUNC_END(__xts_crypt8)
 	eor		v18.16b, \o2\().16b, v27.16b
 	eor		v19.16b, \o3\().16b, v28.16b
 
-	ldp		q24, q25, [sp, #16]
+	ldp		q24, q25, [x6]
 
 	eor		v20.16b, \o4\().16b, v29.16b
 	eor		v21.16b, \o5\().16b, v30.16b
@@ -795,7 +795,7 @@ SYM_FUNC_END(__xts_crypt8)
 	b.gt		0b
 
 	st1		{v25.16b}, [x5]
-	ldp		x29, x30, [sp], #48
+	frame_pop
 	ret
 	.endm
 
@@ -820,9 +820,7 @@ SYM_FUNC_END(aesbs_xts_decrypt)
 	 *		     int rounds, int blocks, u8 iv[])
 	 */
 SYM_FUNC_START(aesbs_ctr_encrypt)
-	stp		x29, x30, [sp, #-16]!
-	mov		x29, sp
-
+	frame_push	0
 	ldp		x7, x8, [x5]
 	ld1		{v0.16b}, [x5]
 CPU_LE(	rev		x7, x7		)
@@ -862,6 +860,6 @@ CPU_LE(	rev		x8, x8		)
 	b.gt		0b
 
 	st1		{v0.16b}, [x5]
-	ldp		x29, x30, [sp], #16
+	frame_pop
 	ret
 SYM_FUNC_END(aesbs_ctr_encrypt)

arch/arm64/crypto/crct10dif-ce-core.S

@@ -429,7 +429,7 @@ CPU_LE(	ext		v0.16b, v0.16b, v0.16b, #8	)
 
 	umov		w0, v0.h[0]
 	.ifc		\p, p8
-	ldp		x29, x30, [sp], #16
+	frame_pop
 	.endif
 	ret
 
@@ -466,8 +466,7 @@ CPU_LE(	ext		v7.16b, v7.16b, v7.16b, #8	)
 // Assumes len >= 16.
 //
 SYM_FUNC_START(crc_t10dif_pmull_p8)
-	stp		x29, x30, [sp, #-16]!
-	mov		x29, sp
+	frame_push	1
 	crc_t10dif_pmull p8
 SYM_FUNC_END(crc_t10dif_pmull_p8)
 

arch/arm64/crypto/ghash-ce-core.S

@@ -436,9 +436,7 @@ SYM_FUNC_END(pmull_ghash_update_p8)
 
 	.align		6
 	.macro		pmull_gcm_do_crypt, enc
-	stp		x29, x30, [sp, #-32]!
-	mov		x29, sp
-	str		x19, [sp, #24]
+	frame_push	1
 
 	load_round_keys	x7, x6, x8
 
@@ -529,7 +527,7 @@ CPU_LE(	rev		w8, w8		)
 	.endif
 	bne		0b
 
-3:	ldp		x19, x10, [sp, #24]
+3:	ldr		x10, [sp, #.Lframe_local_offset]
 	cbz		x10, 5f				// output tag?
 
 	ld1		{INP3.16b}, [x10]		// load lengths[]
@@ -562,7 +560,7 @@ CPU_LE(	rev		w8, w8		)
 	smov		w0, v0.b[0]			// return b0
 	.endif
 
-4:	ldp		x29, x30, [sp], #32
+4:	frame_pop
 	ret
 
 5:

arch/arm64/crypto/ghash-ce-glue.c

@@ -508,7 +508,7 @@ static void __exit ghash_ce_mod_exit(void)
 		crypto_unregister_shash(&ghash_alg);
 }
 
-static const struct cpu_feature ghash_cpu_feature[] = {
+static const struct cpu_feature __maybe_unused ghash_cpu_feature[] = {
 	{ cpu_feature(PMULL) }, { }
 };
 MODULE_DEVICE_TABLE(cpu, ghash_cpu_feature);

arch/arm64/crypto/nh-neon-core.S

@@ -8,6 +8,7 @@
  */
 
 #include <linux/linkage.h>
+#include <linux/cfi_types.h>
 
 	KEY		.req	x0
 	MESSAGE		.req	x1
@@ -58,11 +59,11 @@
 
 /*
  * void nh_neon(const u32 *key, const u8 *message, size_t message_len,
- *		u8 hash[NH_HASH_BYTES])
+ *		__le64 hash[NH_NUM_PASSES])
  *
  * It's guaranteed that message_len % 16 == 0.
  */
-SYM_FUNC_START(nh_neon)
+SYM_TYPED_FUNC_START(nh_neon)
 
 	ld1		{K0.4s,K1.4s}, [KEY], #32
 	  movi		PASS0_SUMS.2d, #0

arch/arm64/crypto/nhpoly1305-neon-glue.c

@@ -14,14 +14,7 @@
 #include <linux/module.h>
 
 asmlinkage void nh_neon(const u32 *key, const u8 *message, size_t message_len,
-			u8 hash[NH_HASH_BYTES]);
-
-/* wrapper to avoid indirect call to assembly, which doesn't work with CFI */
-static void _nh_neon(const u32 *key, const u8 *message, size_t message_len,
-		     __le64 hash[NH_NUM_PASSES])
-{
-	nh_neon(key, message, message_len, (u8 *)hash);
-}
+			__le64 hash[NH_NUM_PASSES]);
 
 static int nhpoly1305_neon_update(struct shash_desc *desc,
 				  const u8 *src, unsigned int srclen)
@@ -33,7 +26,7 @@ static int nhpoly1305_neon_update(struct shash_desc *desc,
 		unsigned int n = min_t(unsigned int, srclen, SZ_4K);
 
 		kernel_neon_begin();
-		crypto_nhpoly1305_update_helper(desc, src, n, _nh_neon);
+		crypto_nhpoly1305_update_helper(desc, src, n, nh_neon);
 		kernel_neon_end();
 		src += n;
 		srclen -= n;

arch/arm64/crypto/sm3-ce-glue.c

@@ -84,7 +84,7 @@ static struct shash_alg sm3_alg = {
 	.base.cra_driver_name	= "sm3-ce",
 	.base.cra_blocksize	= SM3_BLOCK_SIZE,
 	.base.cra_module	= THIS_MODULE,
-	.base.cra_priority	= 200,
+	.base.cra_priority	= 400,
 };
 
 static int __init sm3_ce_mod_init(void)

arch/arm64/crypto/sm3-neon-glue.c

+// SPDX-License-Identifier: GPL-2.0-or-later
+/*
+ * sm3-neon-glue.c - SM3 secure hash using NEON instructions
+ *
+ * Copyright (C) 2022 Tianjia Zhang <tianjia.zhang@linux.alibaba.com>
+ */
+
+#include <asm/neon.h>
+#include <asm/simd.h>
+#include <asm/unaligned.h>
+#include <crypto/internal/hash.h>
+#include <crypto/internal/simd.h>
+#include <crypto/sm3.h>
+#include <crypto/sm3_base.h>
+#include <linux/cpufeature.h>
+#include <linux/crypto.h>
+#include <linux/module.h>
+
+
+asmlinkage void sm3_neon_transform(struct sm3_state *sst, u8 const *src,
+				   int blocks);
+
+static int sm3_neon_update(struct shash_desc *desc, const u8 *data,
+			   unsigned int len)
+{
+	if (!crypto_simd_usable()) {
+		sm3_update(shash_desc_ctx(desc), data, len);
+		return 0;
+	}
+
+	kernel_neon_begin();
+	sm3_base_do_update(desc, data, len, sm3_neon_transform);
+	kernel_neon_end();
+
+	return 0;
+}
+
+static int sm3_neon_final(struct shash_desc *desc, u8 *out)
+{
+	if (!crypto_simd_usable()) {
+		sm3_final(shash_desc_ctx(desc), out);
+		return 0;
+	}
+
+	kernel_neon_begin();
+	sm3_base_do_finalize(desc, sm3_neon_transform);
+	kernel_neon_end();
+
+	return sm3_base_finish(desc, out);
+}
+
+static int sm3_neon_finup(struct shash_desc *desc, const u8 *data,
+			  unsigned int len, u8 *out)
+{
+	if (!crypto_simd_usable()) {
+		struct sm3_state *sctx = shash_desc_ctx(desc);
+
+		if (len)
+			sm3_update(sctx, data, len);
+		sm3_final(sctx, out);
+		return 0;
+	}
+
+	kernel_neon_begin();
+	if (len)
+		sm3_base_do_update(desc, data, len, sm3_neon_transform);
+	sm3_base_do_finalize(desc, sm3_neon_transform);
+	kernel_neon_end();
+
+	return sm3_base_finish(desc, out);
+}
+
+static struct shash_alg sm3_alg = {
+	.digestsize		= SM3_DIGEST_SIZE,
+	.init			= sm3_base_init,
+	.update			= sm3_neon_update,
+	.final			= sm3_neon_final,
+	.finup			= sm3_neon_finup,
+	.descsize		= sizeof(struct sm3_state),
+	.base.cra_name		= "sm3",
+	.base.cra_driver_name	= "sm3-neon",
+	.base.cra_blocksize	= SM3_BLOCK_SIZE,
+	.base.cra_module	= THIS_MODULE,
+	.base.cra_priority	= 200,
+};
+
+static int __init sm3_neon_init(void)
+{
+	return crypto_register_shash(&sm3_alg);
+}
+
+static void __exit sm3_neon_fini(void)
+{
+	crypto_unregister_shash(&sm3_alg);
+}
+
+module_init(sm3_neon_init);
+module_exit(sm3_neon_fini);
+
+MODULE_DESCRIPTION("SM3 secure hash using NEON instructions");
+MODULE_AUTHOR("Jussi Kivilinna <jussi.kivilinna@iki.fi>");
+MODULE_AUTHOR("Tianjia Zhang <tianjia.zhang@linux.alibaba.com>");
+MODULE_LICENSE("GPL v2");

arch/arm64/crypto/sm4-ce-asm.h

+/* SPDX-License-Identifier: GPL-2.0-or-later */
+/*
+ * SM4 helper macros for Crypto Extensions
+ * Copyright (C) 2022 Tianjia Zhang <tianjia.zhang@linux.alibaba.com>
+ */
+
+#define SM4_PREPARE(ptr)					\
+	ld1		{v24.16b-v27.16b}, [ptr], #64;		\
+	ld1		{v28.16b-v31.16b}, [ptr];
+
+#define SM4_CRYPT_BLK_BE(b0)					\
+	sm4e		b0.4s, v24.4s;				\
+	sm4e		b0.4s, v25.4s;				\
+	sm4e		b0.4s, v26.4s;				\
+	sm4e		b0.4s, v27.4s;				\
+	sm4e		b0.4s, v28.4s;				\
+	sm4e		b0.4s, v29.4s;				\
+	sm4e		b0.4s, v30.4s;				\
+	sm4e		b0.4s, v31.4s;				\
+	rev64		b0.4s, b0.4s;				\
+	ext		b0.16b, b0.16b, b0.16b, #8;		\
+	rev32		b0.16b, b0.16b;
+
+#define SM4_CRYPT_BLK(b0)					\
+	rev32		b0.16b, b0.16b;				\
+	SM4_CRYPT_BLK_BE(b0);
+
+#define SM4_CRYPT_BLK2_BE(b0, b1)				\
+	sm4e		b0.4s, v24.4s;				\
+	sm4e		b1.4s, v24.4s;				\
+	sm4e		b0.4s, v25.4s;				\
+	sm4e		b1.4s, v25.4s;				\
+	sm4e		b0.4s, v26.4s;				\
+	sm4e		b1.4s, v26.4s;				\
+	sm4e		b0.4s, v27.4s;				\
+	sm4e		b1.4s, v27.4s;				\
+	sm4e		b0.4s, v28.4s;				\
+	sm4e		b1.4s, v28.4s;				\
+	sm4e		b0.4s, v29.4s;				\
+	sm4e		b1.4s, v29.4s;				\
+	sm4e		b0.4s, v30.4s;				\
+	sm4e		b1.4s, v30.4s;				\
+	sm4e		b0.4s, v31.4s;				\
+	sm4e		b1.4s, v31.4s;				\
+	rev64		b0.4s, b0.4s;				\
+	rev64		b1.4s, b1.4s;				\
+	ext		b0.16b, b0.16b, b0.16b, #8;		\
+	ext		b1.16b, b1.16b, b1.16b, #8;		\
+	rev32		b0.16b, b0.16b;				\
+	rev32		b1.16b, b1.16b;				\
+
+#define SM4_CRYPT_BLK2(b0, b1)					\
+	rev32		b0.16b, b0.16b;				\
+	rev32		b1.16b, b1.16b;				\
+	SM4_CRYPT_BLK2_BE(b0, b1);
+
+#define SM4_CRYPT_BLK4_BE(b0, b1, b2, b3)			\
+	sm4e		b0.4s, v24.4s;				\
+	sm4e		b1.4s, v24.4s;				\
+	sm4e		b2.4s, v24.4s;				\
+	sm4e		b3.4s, v24.4s;				\
+	sm4e		b0.4s, v25.4s;				\
+	sm4e		b1.4s, v25.4s;				\
+	sm4e		b2.4s, v25.4s;				\
+	sm4e		b3.4s, v25.4s;				\
+	sm4e		b0.4s, v26.4s;				\
+	sm4e		b1.4s, v26.4s;				\
+	sm4e		b2.4s, v26.4s;				\
+	sm4e		b3.4s, v26.4s;				\
+	sm4e		b0.4s, v27.4s;				\
+	sm4e		b1.4s, v27.4s;				\
+	sm4e		b2.4s, v27.4s;				\
+	sm4e		b3.4s, v27.4s;				\
+	sm4e		b0.4s, v28.4s;				\
+	sm4e		b1.4s, v28.4s;				\
+	sm4e		b2.4s, v28.4s;				\
+	sm4e		b3.4s, v28.4s;				\
+	sm4e		b0.4s, v29.4s;				\
+	sm4e		b1.4s, v29.4s;				\
+	sm4e		b2.4s, v29.4s;				\
+	sm4e		b3.4s, v29.4s;				\
+	sm4e		b0.4s, v30.4s;				\
+	sm4e		b1.4s, v30.4s;				\
+	sm4e		b2.4s, v30.4s;				\
+	sm4e		b3.4s, v30.4s;				\
+	sm4e		b0.4s, v31.4s;				\
+	sm4e		b1.4s, v31.4s;				\
+	sm4e		b2.4s, v31.4s;				\
+	sm4e		b3.4s, v31.4s;				\
+	rev64		b0.4s, b0.4s;				\
+	rev64		b1.4s, b1.4s;				\
+	rev64		b2.4s, b2.4s;				\
+	rev64		b3.4s, b3.4s;				\
+	ext		b0.16b, b0.16b, b0.16b, #8;		\
+	ext		b1.16b, b1.16b, b1.16b, #8;		\
+	ext		b2.16b, b2.16b, b2.16b, #8;		\
+	ext		b3.16b, b3.16b, b3.16b, #8;		\
+	rev32		b0.16b, b0.16b;				\
+	rev32		b1.16b, b1.16b;				\
+	rev32		b2.16b, b2.16b;				\
+	rev32		b3.16b, b3.16b;
+
+#define SM4_CRYPT_BLK4(b0, b1, b2, b3)				\
+	rev32		b0.16b, b0.16b;				\
+	rev32		b1.16b, b1.16b;				\
+	rev32		b2.16b, b2.16b;				\
+	rev32		b3.16b, b3.16b;				\
+	SM4_CRYPT_BLK4_BE(b0, b1, b2, b3);
+
+#define SM4_CRYPT_BLK8_BE(b0, b1, b2, b3, b4, b5, b6, b7)	\
+	sm4e		b0.4s, v24.4s;				\
+	sm4e		b1.4s, v24.4s;				\
+	sm4e		b2.4s, v24.4s;				\
+	sm4e		b3.4s, v24.4s;				\
+	sm4e		b4.4s, v24.4s;				\
+	sm4e		b5.4s, v24.4s;				\
+	sm4e		b6.4s, v24.4s;				\
+	sm4e		b7.4s, v24.4s;				\
+	sm4e		b0.4s, v25.4s;				\
+	sm4e		b1.4s, v25.4s;				\
+	sm4e		b2.4s, v25.4s;				\
+	sm4e		b3.4s, v25.4s;				\
+	sm4e		b4.4s, v25.4s;				\
+	sm4e		b5.4s, v25.4s;				\
+	sm4e		b6.4s, v25.4s;				\
+	sm4e		b7.4s, v25.4s;				\
+	sm4e		b0.4s, v26.4s;				\
+	sm4e		b1.4s, v26.4s;				\
+	sm4e		b2.4s, v26.4s;				\
+	sm4e		b3.4s, v26.4s;				\
+	sm4e		b4.4s, v26.4s;				\
+	sm4e		b5.4s, v26.4s;				\
+	sm4e		b6.4s, v26.4s;				\
+	sm4e		b7.4s, v26.4s;				\
+	sm4e		b0.4s, v27.4s;				\
+	sm4e		b1.4s, v27.4s;				\
+	sm4e		b2.4s, v27.4s;				\
+	sm4e		b3.4s, v27.4s;				\
+	sm4e		b4.4s, v27.4s;				\
+	sm4e		b5.4s, v27.4s;				\
+	sm4e		b6.4s, v27.4s;				\
+	sm4e		b7.4s, v27.4s;				\
+	sm4e		b0.4s, v28.4s;				\
+	sm4e		b1.4s, v28.4s;				\
+	sm4e		b2.4s, v28.4s;				\
+	sm4e		b3.4s, v28.4s;				\
+	sm4e		b4.4s, v28.4s;				\
+	sm4e		b5.4s, v28.4s;				\
+	sm4e		b6.4s, v28.4s;				\
+	sm4e		b7.4s, v28.4s;				\
+	sm4e		b0.4s, v29.4s;				\
+	sm4e		b1.4s, v29.4s;				\
+	sm4e		b2.4s, v29.4s;				\
+	sm4e		b3.4s, v29.4s;				\
+	sm4e		b4.4s, v29.4s;				\
+	sm4e		b5.4s, v29.4s;				\
+	sm4e		b6.4s, v29.4s;				\
+	sm4e		b7.4s, v29.4s;				\
+	sm4e		b0.4s, v30.4s;				\
+	sm4e		b1.4s, v30.4s;				\
+	sm4e		b2.4s, v30.4s;				\
+	sm4e		b3.4s, v30.4s;				\
+	sm4e		b4.4s, v30.4s;				\
+	sm4e		b5.4s, v30.4s;				\
+	sm4e		b6.4s, v30.4s;				\
+	sm4e		b7.4s, v30.4s;				\
+	sm4e		b0.4s, v31.4s;				\
+	sm4e		b1.4s, v31.4s;				\
+	sm4e		b2.4s, v31.4s;				\
+	sm4e		b3.4s, v31.4s;				\
+	sm4e		b4.4s, v31.4s;				\
+	sm4e		b5.4s, v31.4s;				\
+	sm4e		b6.4s, v31.4s;				\
+	sm4e		b7.4s, v31.4s;				\
+	rev64		b0.4s, b0.4s;				\
+	rev64		b1.4s, b1.4s;				\
+	rev64		b2.4s, b2.4s;				\
+	rev64		b3.4s, b3.4s;				\
+	rev64		b4.4s, b4.4s;				\
+	rev64		b5.4s, b5.4s;				\
+	rev64		b6.4s, b6.4s;				\
+	rev64		b7.4s, b7.4s;				\
+	ext		b0.16b, b0.16b, b0.16b, #8;		\
+	ext		b1.16b, b1.16b, b1.16b, #8;		\
+	ext		b2.16b, b2.16b, b2.16b, #8;		\
+	ext		b3.16b, b3.16b, b3.16b, #8;		\
+	ext		b4.16b, b4.16b, b4.16b, #8;		\
+	ext		b5.16b, b5.16b, b5.16b, #8;		\
+	ext		b6.16b, b6.16b, b6.16b, #8;		\
+	ext		b7.16b, b7.16b, b7.16b, #8;		\
+	rev32		b0.16b, b0.16b;				\
+	rev32		b1.16b, b1.16b;				\
+	rev32		b2.16b, b2.16b;				\
+	rev32		b3.16b, b3.16b;				\
+	rev32		b4.16b, b4.16b;				\
+	rev32		b5.16b, b5.16b;				\
+	rev32		b6.16b, b6.16b;				\
+	rev32		b7.16b, b7.16b;
+
+#define SM4_CRYPT_BLK8(b0, b1, b2, b3, b4, b5, b6, b7)		\
+	rev32		b0.16b, b0.16b;				\
+	rev32		b1.16b, b1.16b;				\
+	rev32		b2.16b, b2.16b;				\
+	rev32		b3.16b, b3.16b;				\
+	rev32		b4.16b, b4.16b;				\
+	rev32		b5.16b, b5.16b;				\
+	rev32		b6.16b, b6.16b;				\
+	rev32		b7.16b, b7.16b;				\
+	SM4_CRYPT_BLK8_BE(b0, b1, b2, b3, b4, b5, b6, b7);

arch/arm64/crypto/sm4-ce-ccm-core.S

+/* SPDX-License-Identifier: GPL-2.0-or-later */
+/*
+ * SM4-CCM AEAD Algorithm using ARMv8 Crypto Extensions
+ * as specified in rfc8998
+ * https://datatracker.ietf.org/doc/html/rfc8998
+ *
+ * Copyright (C) 2022 Tianjia Zhang <tianjia.zhang@linux.alibaba.com>
+ */
+
+#include <linux/linkage.h>
+#include <asm/assembler.h>
+#include "sm4-ce-asm.h"
+
+.arch	armv8-a+crypto
+
+.irp b, 0, 1, 8, 9, 10, 11, 12, 13, 14, 15, 16, 24, 25, 26, 27, 28, 29, 30, 31
+	.set .Lv\b\().4s, \b
+.endr
+
+.macro sm4e, vd, vn
+	.inst 0xcec08400 | (.L\vn << 5) | .L\vd
+.endm
+
+/* Register macros */
+
+#define RMAC	v16
+
+/* Helper macros. */
+
+#define inc_le128(vctr)					\
+		mov		vctr.d[1], x8;		\
+		mov		vctr.d[0], x7;		\
+		adds		x8, x8, #1;		\
+		rev64		vctr.16b, vctr.16b;	\
+		adc		x7, x7, xzr;
+
+
+.align 3
+SYM_FUNC_START(sm4_ce_cbcmac_update)
+	/* input:
+	 *   x0: round key array, CTX
+	 *   x1: mac
+	 *   x2: src
+	 *   w3: nblocks
+	 */
+	SM4_PREPARE(x0)
+
+	ld1		{RMAC.16b}, [x1]
+
+.Lcbcmac_loop_4x:
+	cmp		w3, #4
+	blt		.Lcbcmac_loop_1x
+
+	sub		w3, w3, #4
+
+	ld1		{v0.16b-v3.16b}, [x2], #64
+
+	SM4_CRYPT_BLK(RMAC)
+	eor		RMAC.16b, RMAC.16b, v0.16b
+	SM4_CRYPT_BLK(RMAC)
+	eor		RMAC.16b, RMAC.16b, v1.16b
+	SM4_CRYPT_BLK(RMAC)
+	eor		RMAC.16b, RMAC.16b, v2.16b
+	SM4_CRYPT_BLK(RMAC)
+	eor		RMAC.16b, RMAC.16b, v3.16b
+
+	cbz		w3, .Lcbcmac_end
+	b		.Lcbcmac_loop_4x
+
+.Lcbcmac_loop_1x:
+	sub		w3, w3, #1
+
+	ld1		{v0.16b}, [x2], #16
+
+	SM4_CRYPT_BLK(RMAC)
+	eor		RMAC.16b, RMAC.16b, v0.16b
+
+	cbnz		w3, .Lcbcmac_loop_1x
+
+.Lcbcmac_end:
+	st1		{RMAC.16b}, [x1]
+	ret
+SYM_FUNC_END(sm4_ce_cbcmac_update)
+
+.align 3
+SYM_FUNC_START(sm4_ce_ccm_final)
+	/* input:
+	 *   x0: round key array, CTX
+	 *   x1: ctr0 (big endian, 128 bit)
+	 *   x2: mac
+	 */
+	SM4_PREPARE(x0)
+
+	ld1		{RMAC.16b}, [x2]
+	ld1		{v0.16b}, [x1]
+
+	SM4_CRYPT_BLK2(RMAC, v0)
+
+	/* en-/decrypt the mac with ctr0 */
+	eor		RMAC.16b, RMAC.16b, v0.16b
+	st1		{RMAC.16b}, [x2]
+
+	ret
+SYM_FUNC_END(sm4_ce_ccm_final)
+
+.align 3
+SYM_FUNC_START(sm4_ce_ccm_enc)
+	/* input:
+	 *   x0: round key array, CTX
+	 *   x1: dst
+	 *   x2: src
+	 *   x3: ctr (big endian, 128 bit)
+	 *   w4: nbytes
+	 *   x5: mac
+	 */
+	SM4_PREPARE(x0)
+
+	ldp		x7, x8, [x3]
+	rev		x7, x7
+	rev		x8, x8
+
+	ld1		{RMAC.16b}, [x5]
+
+.Lccm_enc_loop_4x:
+	cmp		w4, #(4 * 16)
+	blt		.Lccm_enc_loop_1x
+
+	sub		w4, w4, #(4 * 16)
+
+	/* construct CTRs */
+	inc_le128(v8)			/* +0 */
+	inc_le128(v9)			/* +1 */
+	inc_le128(v10)			/* +2 */
+	inc_le128(v11)			/* +3 */
+
+	ld1		{v0.16b-v3.16b}, [x2], #64
+
+	SM4_CRYPT_BLK2(v8, RMAC)
+	eor		v8.16b, v8.16b, v0.16b
+	eor		RMAC.16b, RMAC.16b, v0.16b
+	SM4_CRYPT_BLK2(v9, RMAC)
+	eor		v9.16b, v9.16b, v1.16b
+	eor		RMAC.16b, RMAC.16b, v1.16b
+	SM4_CRYPT_BLK2(v10, RMAC)
+	eor		v10.16b, v10.16b, v2.16b
+	eor		RMAC.16b, RMAC.16b, v2.16b
+	SM4_CRYPT_BLK2(v11, RMAC)
+	eor		v11.16b, v11.16b, v3.16b
+	eor		RMAC.16b, RMAC.16b, v3.16b
+
+	st1		{v8.16b-v11.16b}, [x1], #64
+
+	cbz		w4, .Lccm_enc_end
+	b		.Lccm_enc_loop_4x
+
+.Lccm_enc_loop_1x:
+	cmp		w4, #16
+	blt		.Lccm_enc_tail
+
+	sub		w4, w4, #16
+
+	/* construct CTRs */
+	inc_le128(v8)
+
+	ld1		{v0.16b}, [x2], #16
+
+	SM4_CRYPT_BLK2(v8, RMAC)
+	eor		v8.16b, v8.16b, v0.16b
+	eor		RMAC.16b, RMAC.16b, v0.16b
+
+	st1		{v8.16b}, [x1], #16
+
+	cbz		w4, .Lccm_enc_end
+	b		.Lccm_enc_loop_1x
+
+.Lccm_enc_tail:
+	/* construct CTRs */
+	inc_le128(v8)
+
+	SM4_CRYPT_BLK2(RMAC, v8)
+
+	/* store new MAC */
+	st1		{RMAC.16b}, [x5]
+
+.Lccm_enc_tail_loop:
+	ldrb		w0, [x2], #1		/* get 1 byte from input */
+	umov		w9, v8.b[0]		/* get top crypted CTR byte */
+	umov		w6, RMAC.b[0]		/* get top MAC byte */
+
+	eor		w9, w9, w0		/* w9 = CTR ^ input */
+	eor		w6, w6, w0		/* w6 = MAC ^ input */
+
+	strb		w9, [x1], #1		/* store out byte */
+	strb		w6, [x5], #1		/* store MAC byte */
+
+	subs		w4, w4, #1
+	beq		.Lccm_enc_ret
+
+	/* shift out one byte */
+	ext		RMAC.16b, RMAC.16b, RMAC.16b, #1
+	ext		v8.16b, v8.16b, v8.16b, #1
+
+	b		.Lccm_enc_tail_loop
+
+.Lccm_enc_end:
+	/* store new MAC */
+	st1		{RMAC.16b}, [x5]
+
+	/* store new CTR */
+	rev		x7, x7
+	rev		x8, x8
+	stp		x7, x8, [x3]
+
+.Lccm_enc_ret:
+	ret
+SYM_FUNC_END(sm4_ce_ccm_enc)
+
+.align 3
+SYM_FUNC_START(sm4_ce_ccm_dec)
+	/* input:
+	 *   x0: round key array, CTX
+	 *   x1: dst
+	 *   x2: src
+	 *   x3: ctr (big endian, 128 bit)
+	 *   w4: nbytes
+	 *   x5: mac
+	 */
+	SM4_PREPARE(x0)
+
+	ldp		x7, x8, [x3]
+	rev		x7, x7
+	rev		x8, x8
+
+	ld1		{RMAC.16b}, [x5]
+
+.Lccm_dec_loop_4x:
+	cmp		w4, #(4 * 16)
+	blt		.Lccm_dec_loop_1x
+
+	sub		w4, w4, #(4 * 16)
+
+	/* construct CTRs */
+	inc_le128(v8)			/* +0 */
+	inc_le128(v9)			/* +1 */
+	inc_le128(v10)			/* +2 */
+	inc_le128(v11)			/* +3 */
+
+	ld1		{v0.16b-v3.16b}, [x2], #64
+
+	SM4_CRYPT_BLK2(v8, RMAC)
+	eor		v8.16b, v8.16b, v0.16b
+	eor		RMAC.16b, RMAC.16b, v8.16b
+	SM4_CRYPT_BLK2(v9, RMAC)
+	eor		v9.16b, v9.16b, v1.16b
+	eor		RMAC.16b, RMAC.16b, v9.16b
+	SM4_CRYPT_BLK2(v10, RMAC)
+	eor		v10.16b, v10.16b, v2.16b
+	eor		RMAC.16b, RMAC.16b, v10.16b
+	SM4_CRYPT_BLK2(v11, RMAC)
+	eor		v11.16b, v11.16b, v3.16b
+	eor		RMAC.16b, RMAC.16b, v11.16b
+
+	st1		{v8.16b-v11.16b}, [x1], #64
+
+	cbz		w4, .Lccm_dec_end
+	b		.Lccm_dec_loop_4x
+
+.Lccm_dec_loop_1x:
+	cmp		w4, #16
+	blt		.Lccm_dec_tail
+
+	sub		w4, w4, #16
+
+	/* construct CTRs */
+	inc_le128(v8)
+
+	ld1		{v0.16b}, [x2], #16
+
+	SM4_CRYPT_BLK2(v8, RMAC)
+	eor		v8.16b, v8.16b, v0.16b
+	eor		RMAC.16b, RMAC.16b, v8.16b
+
+	st1		{v8.16b}, [x1], #16
+
+	cbz		w4, .Lccm_dec_end
+	b		.Lccm_dec_loop_1x
+
+.Lccm_dec_tail:
+	/* construct CTRs */
+	inc_le128(v8)
+
+	SM4_CRYPT_BLK2(RMAC, v8)
+
+	/* store new MAC */
+	st1		{RMAC.16b}, [x5]
+
+.Lccm_dec_tail_loop:
+	ldrb		w0, [x2], #1		/* get 1 byte from input */
+	umov		w9, v8.b[0]		/* get top crypted CTR byte */
+	umov		w6, RMAC.b[0]		/* get top MAC byte */
+
+	eor		w9, w9, w0		/* w9 = CTR ^ input */
+	eor		w6, w6, w9		/* w6 = MAC ^ output */
+
+	strb		w9, [x1], #1		/* store out byte */
+	strb		w6, [x5], #1		/* store MAC byte */
+
+	subs		w4, w4, #1
+	beq		.Lccm_dec_ret
+
+	/* shift out one byte */
+	ext		RMAC.16b, RMAC.16b, RMAC.16b, #1
+	ext		v8.16b, v8.16b, v8.16b, #1
+
+	b		.Lccm_dec_tail_loop
+
+.Lccm_dec_end:
+	/* store new MAC */
+	st1		{RMAC.16b}, [x5]
+
+	/* store new CTR */
+	rev		x7, x7
+	rev		x8, x8
+	stp		x7, x8, [x3]
+
+.Lccm_dec_ret:
+	ret
+SYM_FUNC_END(sm4_ce_ccm_dec)

arch/arm64/crypto/sm4-ce-ccm-glue.c

+/* SPDX-License-Identifier: GPL-2.0-or-later */
+/*
+ * SM4-CCM AEAD Algorithm using ARMv8 Crypto Extensions
+ * as specified in rfc8998
+ * https://datatracker.ietf.org/doc/html/rfc8998
+ *
+ * Copyright (C) 2022 Tianjia Zhang <tianjia.zhang@linux.alibaba.com>
+ */
+
+#include <linux/module.h>
+#include <linux/crypto.h>
+#include <linux/kernel.h>
+#include <linux/cpufeature.h>
+#include <asm/neon.h>
+#include <crypto/scatterwalk.h>
+#include <crypto/internal/aead.h>
+#include <crypto/internal/skcipher.h>
+#include <crypto/sm4.h>
+#include "sm4-ce.h"
+
+asmlinkage void sm4_ce_cbcmac_update(const u32 *rkey_enc, u8 *mac,
+				     const u8 *src, unsigned int nblocks);
+asmlinkage void sm4_ce_ccm_enc(const u32 *rkey_enc, u8 *dst, const u8 *src,
+			       u8 *iv, unsigned int nbytes, u8 *mac);
+asmlinkage void sm4_ce_ccm_dec(const u32 *rkey_enc, u8 *dst, const u8 *src,
+			       u8 *iv, unsigned int nbytes, u8 *mac);
+asmlinkage void sm4_ce_ccm_final(const u32 *rkey_enc, u8 *iv, u8 *mac);
+
+
+static int ccm_setkey(struct crypto_aead *tfm, const u8 *key,
+		      unsigned int key_len)
+{
+	struct sm4_ctx *ctx = crypto_aead_ctx(tfm);
+
+	if (key_len != SM4_KEY_SIZE)
+		return -EINVAL;
+
+	kernel_neon_begin();
+	sm4_ce_expand_key(key, ctx->rkey_enc, ctx->rkey_dec,
+			  crypto_sm4_fk, crypto_sm4_ck);
+	kernel_neon_end();
+
+	return 0;
+}
+
+static int ccm_setauthsize(struct crypto_aead *tfm, unsigned int authsize)
+{
+	if ((authsize & 1) || authsize < 4)
+		return -EINVAL;
+	return 0;
+}
+
+static int ccm_format_input(u8 info[], struct aead_request *req,
+			    unsigned int msglen)
+{
+	struct crypto_aead *aead = crypto_aead_reqtfm(req);
+	unsigned int l = req->iv[0] + 1;
+	unsigned int m;
+	__be32 len;
+
+	/* verify that CCM dimension 'L': 2 <= L <= 8 */
+	if (l < 2 || l > 8)
+		return -EINVAL;
+	if (l < 4 && msglen >> (8 * l))
+		return -EOVERFLOW;
+
+	memset(&req->iv[SM4_BLOCK_SIZE - l], 0, l);
+
+	memcpy(info, req->iv, SM4_BLOCK_SIZE);
+
+	m = crypto_aead_authsize(aead);
+
+	/* format flags field per RFC 3610/NIST 800-38C */
+	*info |= ((m - 2) / 2) << 3;
+	if (req->assoclen)
+		*info |= (1 << 6);
+
+	/*
+	 * format message length field,
+	 * Linux uses a u32 type to represent msglen
+	 */
+	if (l >= 4)
+		l = 4;
+
+	len = cpu_to_be32(msglen);
+	memcpy(&info[SM4_BLOCK_SIZE - l], (u8 *)&len + 4 - l, l);
+
+	return 0;
+}
+
+static void ccm_calculate_auth_mac(struct aead_request *req, u8 mac[])
+{
+	struct crypto_aead *aead = crypto_aead_reqtfm(req);
+	struct sm4_ctx *ctx = crypto_aead_ctx(aead);
+	struct __packed { __be16 l; __be32 h; } aadlen;
+	u32 assoclen = req->assoclen;
+	struct scatter_walk walk;
+	unsigned int len;
+
+	if (assoclen < 0xff00) {
+		aadlen.l = cpu_to_be16(assoclen);
+		len = 2;
+	} else {
+		aadlen.l = cpu_to_be16(0xfffe);
+		put_unaligned_be32(assoclen, &aadlen.h);
+		len = 6;
+	}
+
+	sm4_ce_crypt_block(ctx->rkey_enc, mac, mac);
+	crypto_xor(mac, (const u8 *)&aadlen, len);
+
+	scatterwalk_start(&walk, req->src);
+
+	do {
+		u32 n = scatterwalk_clamp(&walk, assoclen);
+		u8 *p, *ptr;
+
+		if (!n) {
+			scatterwalk_start(&walk, sg_next(walk.sg));
+			n = scatterwalk_clamp(&walk, assoclen);
+		}
+
+		p = ptr = scatterwalk_map(&walk);
+		assoclen -= n;
+		scatterwalk_advance(&walk, n);
+
+		while (n > 0) {
+			unsigned int l, nblocks;
+
+			if (len == SM4_BLOCK_SIZE) {
+				if (n < SM4_BLOCK_SIZE) {
+					sm4_ce_crypt_block(ctx->rkey_enc,
+							   mac, mac);
+
+					len = 0;
+				} else {
+					nblocks = n / SM4_BLOCK_SIZE;
+					sm4_ce_cbcmac_update(ctx->rkey_enc,
+							     mac, ptr, nblocks);
+
+					ptr += nblocks * SM4_BLOCK_SIZE;
+					n %= SM4_BLOCK_SIZE;
+
+					continue;
+				}
+			}
+
+			l = min(n, SM4_BLOCK_SIZE - len);
+			if (l) {
+				crypto_xor(mac + len, ptr, l);
+				len += l;
+				ptr += l;
+				n -= l;
+			}
+		}
+
+		scatterwalk_unmap(p);
+		scatterwalk_done(&walk, 0, assoclen);
+	} while (assoclen);
+}
+
+static int ccm_crypt(struct aead_request *req, struct skcipher_walk *walk,
+		     u32 *rkey_enc, u8 mac[],
+		     void (*sm4_ce_ccm_crypt)(const u32 *rkey_enc, u8 *dst,
+					const u8 *src, u8 *iv,
+					unsigned int nbytes, u8 *mac))
+{
+	u8 __aligned(8) ctr0[SM4_BLOCK_SIZE];
+	int err;
+
+	/* preserve the initial ctr0 for the TAG */
+	memcpy(ctr0, walk->iv, SM4_BLOCK_SIZE);
+	crypto_inc(walk->iv, SM4_BLOCK_SIZE);
+
+	kernel_neon_begin();
+
+	if (req->assoclen)
+		ccm_calculate_auth_mac(req, mac);
+
+	do {
+		unsigned int tail = walk->nbytes % SM4_BLOCK_SIZE;
+		const u8 *src = walk->src.virt.addr;
+		u8 *dst = walk->dst.virt.addr;
+
+		if (walk->nbytes == walk->total)
+			tail = 0;
+
+		if (walk->nbytes - tail)
+			sm4_ce_ccm_crypt(rkey_enc, dst, src, walk->iv,
+					 walk->nbytes - tail, mac);
+
+		if (walk->nbytes == walk->total)
+			sm4_ce_ccm_final(rkey_enc, ctr0, mac);
+
+		kernel_neon_end();
+
+		if (walk->nbytes) {
+			err = skcipher_walk_done(walk, tail);
+			if (err)
+				return err;
+			if (walk->nbytes)
+				kernel_neon_begin();
+		}
+	} while (walk->nbytes > 0);
+
+	return 0;
+}
+
+static int ccm_encrypt(struct aead_request *req)
+{
+	struct crypto_aead *aead = crypto_aead_reqtfm(req);
+	struct sm4_ctx *ctx = crypto_aead_ctx(aead);
+	u8 __aligned(8) mac[SM4_BLOCK_SIZE];
+	struct skcipher_walk walk;
+	int err;
+
+	err = ccm_format_input(mac, req, req->cryptlen);
+	if (err)
+		return err;
+
+	err = skcipher_walk_aead_encrypt(&walk, req, false);
+	if (err)
+		return err;
+
+	err = ccm_crypt(req, &walk, ctx->rkey_enc, mac, sm4_ce_ccm_enc);
+	if (err)
+		return err;
+
+	/* copy authtag to end of dst */
+	scatterwalk_map_and_copy(mac, req->dst, req->assoclen + req->cryptlen,
+				 crypto_aead_authsize(aead), 1);
+
+	return 0;
+}
+
+static int ccm_decrypt(struct aead_request *req)
+{
+	struct crypto_aead *aead = crypto_aead_reqtfm(req);
+	unsigned int authsize = crypto_aead_authsize(aead);
+	struct sm4_ctx *ctx = crypto_aead_ctx(aead);
+	u8 __aligned(8) mac[SM4_BLOCK_SIZE];
+	u8 authtag[SM4_BLOCK_SIZE];
+	struct skcipher_walk walk;
+	int err;
+
+	err = ccm_format_input(mac, req, req->cryptlen - authsize);
+	if (err)
+		return err;
+
+	err = skcipher_walk_aead_decrypt(&walk, req, false);
+	if (err)
+		return err;
+
+	err = ccm_crypt(req, &walk, ctx->rkey_enc, mac, sm4_ce_ccm_dec);
+	if (err)
+		return err;
+
+	/* compare calculated auth tag with the stored one */
+	scatterwalk_map_and_copy(authtag, req->src,
+				 req->assoclen + req->cryptlen - authsize,
+				 authsize, 0);
+
+	if (crypto_memneq(authtag, mac, authsize))
+		return -EBADMSG;
+
+	return 0;
+}
+
+static struct aead_alg sm4_ccm_alg = {
+	.base = {
+		.cra_name		= "ccm(sm4)",
+		.cra_driver_name	= "ccm-sm4-ce",
+		.cra_priority		= 400,
+		.cra_blocksize		= 1,
+		.cra_ctxsize		= sizeof(struct sm4_ctx),
+		.cra_module		= THIS_MODULE,
+	},
+	.ivsize		= SM4_BLOCK_SIZE,
+	.chunksize	= SM4_BLOCK_SIZE,
+	.maxauthsize	= SM4_BLOCK_SIZE,
+	.setkey		= ccm_setkey,
+	.setauthsize	= ccm_setauthsize,
+	.encrypt	= ccm_encrypt,
+	.decrypt	= ccm_decrypt,
+};
+
+static int __init sm4_ce_ccm_init(void)
+{
+	return crypto_register_aead(&sm4_ccm_alg);
+}
+
+static void __exit sm4_ce_ccm_exit(void)
+{
+	crypto_unregister_aead(&sm4_ccm_alg);
+}
+
+module_cpu_feature_match(SM4, sm4_ce_ccm_init);
+module_exit(sm4_ce_ccm_exit);
+
+MODULE_DESCRIPTION("Synchronous SM4 in CCM mode using ARMv8 Crypto Extensions");
+MODULE_ALIAS_CRYPTO("ccm(sm4)");
+MODULE_AUTHOR("Tianjia Zhang <tianjia.zhang@linux.alibaba.com>");
+MODULE_LICENSE("GPL v2");

arch/arm64/crypto/sm4-ce-cipher-glue.c

@@ -2,11 +2,11 @@
 
 #include <asm/neon.h>
 #include <asm/simd.h>
+#include <crypto/algapi.h>
 #include <crypto/sm4.h>
 #include <crypto/internal/simd.h>
 #include <linux/module.h>
 #include <linux/cpufeature.h>
-#include <linux/crypto.h>
 #include <linux/types.h>
 
 MODULE_ALIAS_CRYPTO("sm4");

arch/arm64/crypto/sm4-ce-gcm-glue.c

+/* SPDX-License-Identifier: GPL-2.0-or-later */
+/*
+ * SM4-GCM AEAD Algorithm using ARMv8 Crypto Extensions
+ * as specified in rfc8998
+ * https://datatracker.ietf.org/doc/html/rfc8998
+ *
+ * Copyright (C) 2022 Tianjia Zhang <tianjia.zhang@linux.alibaba.com>
+ */
+
+#include <linux/module.h>
+#include <linux/crypto.h>
+#include <linux/kernel.h>
+#include <linux/cpufeature.h>
+#include <asm/neon.h>
+#include <crypto/b128ops.h>
+#include <crypto/scatterwalk.h>
+#include <crypto/internal/aead.h>
+#include <crypto/internal/skcipher.h>
+#include <crypto/sm4.h>
+#include "sm4-ce.h"
+
+asmlinkage void sm4_ce_pmull_ghash_setup(const u32 *rkey_enc, u8 *ghash_table);
+asmlinkage void pmull_ghash_update(const u8 *ghash_table, u8 *ghash,
+				   const u8 *src, unsigned int nblocks);
+asmlinkage void sm4_ce_pmull_gcm_enc(const u32 *rkey_enc, u8 *dst,
+				     const u8 *src, u8 *iv,
+				     unsigned int nbytes, u8 *ghash,
+				     const u8 *ghash_table, const u8 *lengths);
+asmlinkage void sm4_ce_pmull_gcm_dec(const u32 *rkey_enc, u8 *dst,
+				     const u8 *src, u8 *iv,
+				     unsigned int nbytes, u8 *ghash,
+				     const u8 *ghash_table, const u8 *lengths);
+
+#define GHASH_BLOCK_SIZE	16
+#define GCM_IV_SIZE		12
+
+struct sm4_gcm_ctx {
+	struct sm4_ctx key;
+	u8 ghash_table[16 * 4];
+};
+
+
+static int gcm_setkey(struct crypto_aead *tfm, const u8 *key,
+		      unsigned int key_len)
+{
+	struct sm4_gcm_ctx *ctx = crypto_aead_ctx(tfm);
+
+	if (key_len != SM4_KEY_SIZE)
+		return -EINVAL;
+
+	kernel_neon_begin();
+
+	sm4_ce_expand_key(key, ctx->key.rkey_enc, ctx->key.rkey_dec,
+			  crypto_sm4_fk, crypto_sm4_ck);
+	sm4_ce_pmull_ghash_setup(ctx->key.rkey_enc, ctx->ghash_table);
+
+	kernel_neon_end();
+	return 0;
+}
+
+static int gcm_setauthsize(struct crypto_aead *tfm, unsigned int authsize)
+{
+	switch (authsize) {
+	case 4:
+	case 8:
+	case 12 ... 16:
+		return 0;
+	default:
+		return -EINVAL;
+	}
+}
+
+static void gcm_calculate_auth_mac(struct aead_request *req, u8 ghash[])
+{
+	struct crypto_aead *aead = crypto_aead_reqtfm(req);
+	struct sm4_gcm_ctx *ctx = crypto_aead_ctx(aead);
+	u8 __aligned(8) buffer[GHASH_BLOCK_SIZE];
+	u32 assoclen = req->assoclen;
+	struct scatter_walk walk;
+	unsigned int buflen = 0;
+
+	scatterwalk_start(&walk, req->src);
+
+	do {
+		u32 n = scatterwalk_clamp(&walk, assoclen);
+		u8 *p, *ptr;
+
+		if (!n) {
+			scatterwalk_start(&walk, sg_next(walk.sg));
+			n = scatterwalk_clamp(&walk, assoclen);
+		}
+
+		p = ptr = scatterwalk_map(&walk);
+		assoclen -= n;
+		scatterwalk_advance(&walk, n);
+
+		if (n + buflen < GHASH_BLOCK_SIZE) {
+			memcpy(&buffer[buflen], ptr, n);
+			buflen += n;
+		} else {
+			unsigned int nblocks;
+
+			if (buflen) {
+				unsigned int l = GHASH_BLOCK_SIZE - buflen;
+
+				memcpy(&buffer[buflen], ptr, l);
+				ptr += l;
+				n -= l;
+
+				pmull_ghash_update(ctx->ghash_table, ghash,
+						   buffer, 1);
+			}
+
+			nblocks = n / GHASH_BLOCK_SIZE;
+			if (nblocks) {
+				pmull_ghash_update(ctx->ghash_table, ghash,
+						   ptr, nblocks);
+				ptr += nblocks * GHASH_BLOCK_SIZE;
+			}
+
+			buflen = n % GHASH_BLOCK_SIZE;
+			if (buflen)
+				memcpy(&buffer[0], ptr, buflen);
+		}
+
+		scatterwalk_unmap(p);
+		scatterwalk_done(&walk, 0, assoclen);
+	} while (assoclen);
+
+	/* padding with '0' */
+	if (buflen) {
+		memset(&buffer[buflen], 0, GHASH_BLOCK_SIZE - buflen);
+		pmull_ghash_update(ctx->ghash_table, ghash, buffer, 1);
+	}
+}
+
+static int gcm_crypt(struct aead_request *req, struct skcipher_walk *walk,
+		     struct sm4_gcm_ctx *ctx, u8 ghash[],
+		     void (*sm4_ce_pmull_gcm_crypt)(const u32 *rkey_enc,
+				u8 *dst, const u8 *src, u8 *iv,
+				unsigned int nbytes, u8 *ghash,
+				const u8 *ghash_table, const u8 *lengths))
+{
+	u8 __aligned(8) iv[SM4_BLOCK_SIZE];
+	be128 __aligned(8) lengths;
+	int err;
+
+	memset(ghash, 0, SM4_BLOCK_SIZE);
+
+	lengths.a = cpu_to_be64(req->assoclen * 8);
+	lengths.b = cpu_to_be64(walk->total * 8);
+
+	memcpy(iv, walk->iv, GCM_IV_SIZE);
+	put_unaligned_be32(2, iv + GCM_IV_SIZE);
+
+	kernel_neon_begin();
+
+	if (req->assoclen)
+		gcm_calculate_auth_mac(req, ghash);
+
+	do {
+		unsigned int tail = walk->nbytes % SM4_BLOCK_SIZE;
+		const u8 *src = walk->src.virt.addr;
+		u8 *dst = walk->dst.virt.addr;
+
+		if (walk->nbytes == walk->total) {
+			tail = 0;
+
+			sm4_ce_pmull_gcm_crypt(ctx->key.rkey_enc, dst, src, iv,
+					       walk->nbytes, ghash,
+					       ctx->ghash_table,
+					       (const u8 *)&lengths);
+		} else if (walk->nbytes - tail) {
+			sm4_ce_pmull_gcm_crypt(ctx->key.rkey_enc, dst, src, iv,
+					       walk->nbytes - tail, ghash,
+					       ctx->ghash_table, NULL);
+		}
+
+		kernel_neon_end();
+
+		err = skcipher_walk_done(walk, tail);
+		if (err)
+			return err;
+		if (walk->nbytes)
+			kernel_neon_begin();
+	} while (walk->nbytes > 0);
+
+	return 0;
+}
+
+static int gcm_encrypt(struct aead_request *req)
+{
+	struct crypto_aead *aead = crypto_aead_reqtfm(req);
+	struct sm4_gcm_ctx *ctx = crypto_aead_ctx(aead);
+	u8 __aligned(8) ghash[SM4_BLOCK_SIZE];
+	struct skcipher_walk walk;
+	int err;
+
+	err = skcipher_walk_aead_encrypt(&walk, req, false);
+	if (err)
+		return err;
+
+	err = gcm_crypt(req, &walk, ctx, ghash, sm4_ce_pmull_gcm_enc);
+	if (err)
+		return err;
+
+	/* copy authtag to end of dst */
+	scatterwalk_map_and_copy(ghash, req->dst, req->assoclen + req->cryptlen,
+				 crypto_aead_authsize(aead), 1);
+
+	return 0;
+}
+
+static int gcm_decrypt(struct aead_request *req)
+{
+	struct crypto_aead *aead = crypto_aead_reqtfm(req);
+	unsigned int authsize = crypto_aead_authsize(aead);
+	struct sm4_gcm_ctx *ctx = crypto_aead_ctx(aead);
+	u8 __aligned(8) ghash[SM4_BLOCK_SIZE];
+	u8 authtag[SM4_BLOCK_SIZE];
+	struct skcipher_walk walk;
+	int err;
+
+	err = skcipher_walk_aead_decrypt(&walk, req, false);
+	if (err)
+		return err;
+
+	err = gcm_crypt(req, &walk, ctx, ghash, sm4_ce_pmull_gcm_dec);
+	if (err)
+		return err;
+
+	/* compare calculated auth tag with the stored one */
+	scatterwalk_map_and_copy(authtag, req->src,
+				 req->assoclen + req->cryptlen - authsize,
+				 authsize, 0);
+
+	if (crypto_memneq(authtag, ghash, authsize))
+		return -EBADMSG;
+
+	return 0;
+}
+
+static struct aead_alg sm4_gcm_alg = {
+	.base = {
+		.cra_name		= "gcm(sm4)",
+		.cra_driver_name	= "gcm-sm4-ce",
+		.cra_priority		= 400,
+		.cra_blocksize		= 1,
+		.cra_ctxsize		= sizeof(struct sm4_gcm_ctx),
+		.cra_module		= THIS_MODULE,
+	},
+	.ivsize		= GCM_IV_SIZE,
+	.chunksize	= SM4_BLOCK_SIZE,
+	.maxauthsize	= SM4_BLOCK_SIZE,
+	.setkey		= gcm_setkey,
+	.setauthsize	= gcm_setauthsize,
+	.encrypt	= gcm_encrypt,
+	.decrypt	= gcm_decrypt,
+};
+
+static int __init sm4_ce_gcm_init(void)
+{
+	if (!cpu_have_named_feature(PMULL))
+		return -ENODEV;
+
+	return crypto_register_aead(&sm4_gcm_alg);
+}
+
+static void __exit sm4_ce_gcm_exit(void)
+{
+	crypto_unregister_aead(&sm4_gcm_alg);
+}
+
+static const struct cpu_feature __maybe_unused sm4_ce_gcm_cpu_feature[] = {
+	{ cpu_feature(PMULL) },
+	{}
+};
+MODULE_DEVICE_TABLE(cpu, sm4_ce_gcm_cpu_feature);
+
+module_cpu_feature_match(SM4, sm4_ce_gcm_init);
+module_exit(sm4_ce_gcm_exit);
+
+MODULE_DESCRIPTION("Synchronous SM4 in GCM mode using ARMv8 Crypto Extensions");
+MODULE_ALIAS_CRYPTO("gcm(sm4)");
+MODULE_AUTHOR("Tianjia Zhang <tianjia.zhang@linux.alibaba.com>");
+MODULE_LICENSE("GPL v2");

arch/arm64/crypto/sm4-ce.h

+/* SPDX-License-Identifier: GPL-2.0-or-later */
+/*
+ * SM4 common functions for Crypto Extensions
+ * Copyright (C) 2022 Tianjia Zhang <tianjia.zhang@linux.alibaba.com>
+ */
+
+void sm4_ce_expand_key(const u8 *key, u32 *rkey_enc, u32 *rkey_dec,
+		       const u32 *fk, const u32 *ck);
+
+void sm4_ce_crypt_block(const u32 *rkey, u8 *dst, const u8 *src);
+
+void sm4_ce_cbc_enc(const u32 *rkey_enc, u8 *dst, const u8 *src,
+		    u8 *iv, unsigned int nblocks);
+
+void sm4_ce_cfb_enc(const u32 *rkey_enc, u8 *dst, const u8 *src,
+		    u8 *iv, unsigned int nblocks);

arch/arm64/crypto/sm4-neon-glue.c

@@ -18,19 +18,14 @@
 #include <crypto/internal/skcipher.h>
 #include <crypto/sm4.h>
 
-#define BYTES2BLKS(nbytes)	((nbytes) >> 4)
-#define BYTES2BLK8(nbytes)	(((nbytes) >> 4) & ~(8 - 1))
-
-asmlinkage void sm4_neon_crypt_blk1_8(const u32 *rkey, u8 *dst, const u8 *src,
-				      unsigned int nblks);
-asmlinkage void sm4_neon_crypt_blk8(const u32 *rkey, u8 *dst, const u8 *src,
-				    unsigned int nblks);
-asmlinkage void sm4_neon_cbc_dec_blk8(const u32 *rkey, u8 *dst, const u8 *src,
-				      u8 *iv, unsigned int nblks);
-asmlinkage void sm4_neon_cfb_dec_blk8(const u32 *rkey, u8 *dst, const u8 *src,
-				      u8 *iv, unsigned int nblks);
-asmlinkage void sm4_neon_ctr_enc_blk8(const u32 *rkey, u8 *dst, const u8 *src,
-				      u8 *iv, unsigned int nblks);
+asmlinkage void sm4_neon_crypt(const u32 *rkey, u8 *dst, const u8 *src,
+			       unsigned int nblocks);
+asmlinkage void sm4_neon_cbc_dec(const u32 *rkey_dec, u8 *dst, const u8 *src,
+				 u8 *iv, unsigned int nblocks);
+asmlinkage void sm4_neon_cfb_dec(const u32 *rkey_enc, u8 *dst, const u8 *src,
+				 u8 *iv, unsigned int nblocks);
+asmlinkage void sm4_neon_ctr_crypt(const u32 *rkey_enc, u8 *dst, const u8 *src,
+				   u8 *iv, unsigned int nblocks);
 
 static int sm4_setkey(struct crypto_skcipher *tfm, const u8 *key,
 		      unsigned int key_len)
@@ -51,27 +46,18 @@ static int sm4_ecb_do_crypt(struct skcipher_request *req, const u32 *rkey)
 	while ((nbytes = walk.nbytes) > 0) {
 		const u8 *src = walk.src.virt.addr;
 		u8 *dst = walk.dst.virt.addr;
-		unsigned int nblks;
+		unsigned int nblocks;
 
-		kernel_neon_begin();
+		nblocks = nbytes / SM4_BLOCK_SIZE;
+		if (nblocks) {
+			kernel_neon_begin();
 
-		nblks = BYTES2BLK8(nbytes);
-		if (nblks) {
-			sm4_neon_crypt_blk8(rkey, dst, src, nblks);
-			dst += nblks * SM4_BLOCK_SIZE;
-			src += nblks * SM4_BLOCK_SIZE;
-			nbytes -= nblks * SM4_BLOCK_SIZE;
-		}
+			sm4_neon_crypt(rkey, dst, src, nblocks);
 
-		nblks = BYTES2BLKS(nbytes);
-		if (nblks) {
-			sm4_neon_crypt_blk1_8(rkey, dst, src, nblks);
-			nbytes -= nblks * SM4_BLOCK_SIZE;
+			kernel_neon_end();
 		}
 
-		kernel_neon_end();
-
-		err = skcipher_walk_done(&walk, nbytes);
+		err = skcipher_walk_done(&walk, nbytes % SM4_BLOCK_SIZE);
 	}
 
 	return err;
@@ -138,48 +124,19 @@ static int sm4_cbc_decrypt(struct skcipher_request *req)
 	while ((nbytes = walk.nbytes) > 0) {
 		const u8 *src = walk.src.virt.addr;
 		u8 *dst = walk.dst.virt.addr;
-		unsigned int nblks;
+		unsigned int nblocks;
 
-		kernel_neon_begin();
+		nblocks = nbytes / SM4_BLOCK_SIZE;
+		if (nblocks) {
+			kernel_neon_begin();
 
-		nblks = BYTES2BLK8(nbytes);
-		if (nblks) {
-			sm4_neon_cbc_dec_blk8(ctx->rkey_dec, dst, src,
-					walk.iv, nblks);
-			dst += nblks * SM4_BLOCK_SIZE;
-			src += nblks * SM4_BLOCK_SIZE;
-			nbytes -= nblks * SM4_BLOCK_SIZE;
-		}
+			sm4_neon_cbc_dec(ctx->rkey_dec, dst, src,
+					 walk.iv, nblocks);
 
-		nblks = BYTES2BLKS(nbytes);
-		if (nblks) {
-			u8 keystream[SM4_BLOCK_SIZE * 8];
-			u8 iv[SM4_BLOCK_SIZE];
-			int i;
-
-			sm4_neon_crypt_blk1_8(ctx->rkey_dec, keystream,
-					src, nblks);
-
-			src += ((int)nblks - 2) * SM4_BLOCK_SIZE;
-			dst += (nblks - 1) * SM4_BLOCK_SIZE;
-			memcpy(iv, src + SM4_BLOCK_SIZE, SM4_BLOCK_SIZE);
-
-			for (i = nblks - 1; i > 0; i--) {
-				crypto_xor_cpy(dst, src,
-					&keystream[i * SM4_BLOCK_SIZE],
-					SM4_BLOCK_SIZE);
-				src -= SM4_BLOCK_SIZE;
-				dst -= SM4_BLOCK_SIZE;
-			}
-			crypto_xor_cpy(dst, walk.iv,
-					keystream, SM4_BLOCK_SIZE);
-			memcpy(walk.iv, iv, SM4_BLOCK_SIZE);
-			nbytes -= nblks * SM4_BLOCK_SIZE;
+			kernel_neon_end();
 		}
 
-		kernel_neon_end();
-
-		err = skcipher_walk_done(&walk, nbytes);
+		err = skcipher_walk_done(&walk, nbytes % SM4_BLOCK_SIZE);
 	}
 
 	return err;
@@ -238,41 +195,21 @@ static int sm4_cfb_decrypt(struct skcipher_request *req)
 	while ((nbytes = walk.nbytes) > 0) {
 		const u8 *src = walk.src.virt.addr;
 		u8 *dst = walk.dst.virt.addr;
-		unsigned int nblks;
+		unsigned int nblocks;
 
-		kernel_neon_begin();
+		nblocks = nbytes / SM4_BLOCK_SIZE;
+		if (nblocks) {
+			kernel_neon_begin();
 
-		nblks = BYTES2BLK8(nbytes);
-		if (nblks) {
-			sm4_neon_cfb_dec_blk8(ctx->rkey_enc, dst, src,
-					walk.iv, nblks);
-			dst += nblks * SM4_BLOCK_SIZE;
-			src += nblks * SM4_BLOCK_SIZE;
-			nbytes -= nblks * SM4_BLOCK_SIZE;
-		}
+			sm4_neon_cfb_dec(ctx->rkey_enc, dst, src,
+					 walk.iv, nblocks);
 
-		nblks = BYTES2BLKS(nbytes);
-		if (nblks) {
-			u8 keystream[SM4_BLOCK_SIZE * 8];
-
-			memcpy(keystream, walk.iv, SM4_BLOCK_SIZE);
-			if (nblks > 1)
-				memcpy(&keystream[SM4_BLOCK_SIZE], src,
-					(nblks - 1) * SM4_BLOCK_SIZE);
-			memcpy(walk.iv, src + (nblks - 1) * SM4_BLOCK_SIZE,
-				SM4_BLOCK_SIZE);
-
-			sm4_neon_crypt_blk1_8(ctx->rkey_enc, keystream,
-					keystream, nblks);
-
-			crypto_xor_cpy(dst, src, keystream,
-					nblks * SM4_BLOCK_SIZE);
-			dst += nblks * SM4_BLOCK_SIZE;
-			src += nblks * SM4_BLOCK_SIZE;
-			nbytes -= nblks * SM4_BLOCK_SIZE;
-		}
+			kernel_neon_end();
 
-		kernel_neon_end();
+			dst += nblocks * SM4_BLOCK_SIZE;
+			src += nblocks * SM4_BLOCK_SIZE;
+			nbytes -= nblocks * SM4_BLOCK_SIZE;
+		}
 
 		/* tail */
 		if (walk.nbytes == walk.total && nbytes > 0) {
@@ -302,40 +239,21 @@ static int sm4_ctr_crypt(struct skcipher_request *req)
 	while ((nbytes = walk.nbytes) > 0) {
 		const u8 *src = walk.src.virt.addr;
 		u8 *dst = walk.dst.virt.addr;
-		unsigned int nblks;
+		unsigned int nblocks;
 
-		kernel_neon_begin();
+		nblocks = nbytes / SM4_BLOCK_SIZE;
+		if (nblocks) {
+			kernel_neon_begin();
 
-		nblks = BYTES2BLK8(nbytes);
-		if (nblks) {
-			sm4_neon_ctr_enc_blk8(ctx->rkey_enc, dst, src,
-					walk.iv, nblks);
-			dst += nblks * SM4_BLOCK_SIZE;
-			src += nblks * SM4_BLOCK_SIZE;
-			nbytes -= nblks * SM4_BLOCK_SIZE;
-		}
+			sm4_neon_ctr_crypt(ctx->rkey_enc, dst, src,
+					   walk.iv, nblocks);
 
-		nblks = BYTES2BLKS(nbytes);
-		if (nblks) {
-			u8 keystream[SM4_BLOCK_SIZE * 8];
-			int i;
-
-			for (i = 0; i < nblks; i++) {
-				memcpy(&keystream[i * SM4_BLOCK_SIZE],
-					walk.iv, SM4_BLOCK_SIZE);
-				crypto_inc(walk.iv, SM4_BLOCK_SIZE);
-			}
-			sm4_neon_crypt_blk1_8(ctx->rkey_enc, keystream,
-					keystream, nblks);
-
-			crypto_xor_cpy(dst, src, keystream,
-					nblks * SM4_BLOCK_SIZE);
-			dst += nblks * SM4_BLOCK_SIZE;
-			src += nblks * SM4_BLOCK_SIZE;
-			nbytes -= nblks * SM4_BLOCK_SIZE;
-		}
+			kernel_neon_end();
 
-		kernel_neon_end();
+			dst += nblocks * SM4_BLOCK_SIZE;
+			src += nblocks * SM4_BLOCK_SIZE;
+			nbytes -= nblocks * SM4_BLOCK_SIZE;
+		}
 
 		/* tail */
 		if (walk.nbytes == walk.total && nbytes > 0) {

arch/um/drivers/random.c

@@ -82,7 +82,6 @@ static int __init rng_init (void)
 	sigio_broken(random_fd);
 	hwrng.name = RNG_MODULE_NAME;
 	hwrng.read = rng_dev_read;
-	hwrng.quality = 1024;
 
 	err = hwrng_register(&hwrng);
 	if (err) {

arch/x86/crypto/Makefile

@@ -107,3 +107,6 @@ quiet_cmd_perlasm = PERLASM $@
       cmd_perlasm = $(PERL) $< > $@
 $(obj)/%.S: $(src)/%.pl FORCE
 	$(call if_changed,perlasm)
+
+# Disable GCOV in odd or sensitive code
+GCOV_PROFILE_curve25519-x86_64.o := n

arch/x86/crypto/aegis128-aesni-asm.S

@@ -7,6 +7,7 @@
  */
 
 #include <linux/linkage.h>
+#include <linux/cfi_types.h>
 #include <asm/frame.h>
 
 #define STATE0	%xmm0
@@ -402,7 +403,7 @@ SYM_FUNC_END(crypto_aegis128_aesni_ad)
  * void crypto_aegis128_aesni_enc(void *state, unsigned int length,
  *                                const void *src, void *dst);
  */
-SYM_FUNC_START(crypto_aegis128_aesni_enc)
+SYM_TYPED_FUNC_START(crypto_aegis128_aesni_enc)
 	FRAME_BEGIN
 
 	cmp $0x10, LEN
@@ -499,7 +500,7 @@ SYM_FUNC_END(crypto_aegis128_aesni_enc)
  * void crypto_aegis128_aesni_enc_tail(void *state, unsigned int length,
  *                                     const void *src, void *dst);
  */
-SYM_FUNC_START(crypto_aegis128_aesni_enc_tail)
+SYM_TYPED_FUNC_START(crypto_aegis128_aesni_enc_tail)
 	FRAME_BEGIN
 
 	/* load the state: */
@@ -556,7 +557,7 @@ SYM_FUNC_END(crypto_aegis128_aesni_enc_tail)
  * void crypto_aegis128_aesni_dec(void *state, unsigned int length,
  *                                const void *src, void *dst);
  */
-SYM_FUNC_START(crypto_aegis128_aesni_dec)
+SYM_TYPED_FUNC_START(crypto_aegis128_aesni_dec)
 	FRAME_BEGIN
 
 	cmp $0x10, LEN
@@ -653,7 +654,7 @@ SYM_FUNC_END(crypto_aegis128_aesni_dec)
  * void crypto_aegis128_aesni_dec_tail(void *state, unsigned int length,
  *                                     const void *src, void *dst);
  */
-SYM_FUNC_START(crypto_aegis128_aesni_dec_tail)
+SYM_TYPED_FUNC_START(crypto_aegis128_aesni_dec_tail)
 	FRAME_BEGIN
 
 	/* load the state: */

arch/x86/crypto/aria-aesni-avx-asm_64.S

@@ -7,6 +7,7 @@
  */
 
 #include <linux/linkage.h>
+#include <linux/cfi_types.h>
 #include <asm/frame.h>
 
 /* struct aria_ctx: */
@@ -913,7 +914,7 @@ SYM_FUNC_START_LOCAL(__aria_aesni_avx_crypt_16way)
 	RET;
 SYM_FUNC_END(__aria_aesni_avx_crypt_16way)
 
-SYM_FUNC_START(aria_aesni_avx_encrypt_16way)
+SYM_TYPED_FUNC_START(aria_aesni_avx_encrypt_16way)
 	/* input:
 	*      %rdi: ctx, CTX
 	*      %rsi: dst
@@ -938,7 +939,7 @@ SYM_FUNC_START(aria_aesni_avx_encrypt_16way)
 	RET;
 SYM_FUNC_END(aria_aesni_avx_encrypt_16way)
 
-SYM_FUNC_START(aria_aesni_avx_decrypt_16way)
+SYM_TYPED_FUNC_START(aria_aesni_avx_decrypt_16way)
 	/* input:
 	*      %rdi: ctx, CTX
 	*      %rsi: dst
@@ -1039,7 +1040,7 @@ SYM_FUNC_START_LOCAL(__aria_aesni_avx_ctr_gen_keystream_16way)
 	RET;
 SYM_FUNC_END(__aria_aesni_avx_ctr_gen_keystream_16way)
 
-SYM_FUNC_START(aria_aesni_avx_ctr_crypt_16way)
+SYM_TYPED_FUNC_START(aria_aesni_avx_ctr_crypt_16way)
 	/* input:
 	*      %rdi: ctx
 	*      %rsi: dst
@@ -1208,7 +1209,7 @@ SYM_FUNC_START_LOCAL(__aria_aesni_avx_gfni_crypt_16way)
 	RET;
 SYM_FUNC_END(__aria_aesni_avx_gfni_crypt_16way)
 
-SYM_FUNC_START(aria_aesni_avx_gfni_encrypt_16way)
+SYM_TYPED_FUNC_START(aria_aesni_avx_gfni_encrypt_16way)
 	/* input:
 	*      %rdi: ctx, CTX
 	*      %rsi: dst
@@ -1233,7 +1234,7 @@ SYM_FUNC_START(aria_aesni_avx_gfni_encrypt_16way)
 	RET;
 SYM_FUNC_END(aria_aesni_avx_gfni_encrypt_16way)
 
-SYM_FUNC_START(aria_aesni_avx_gfni_decrypt_16way)
+SYM_TYPED_FUNC_START(aria_aesni_avx_gfni_decrypt_16way)
 	/* input:
 	*      %rdi: ctx, CTX
 	*      %rsi: dst
@@ -1258,7 +1259,7 @@ SYM_FUNC_START(aria_aesni_avx_gfni_decrypt_16way)
 	RET;
 SYM_FUNC_END(aria_aesni_avx_gfni_decrypt_16way)
 
-SYM_FUNC_START(aria_aesni_avx_gfni_ctr_crypt_16way)
+SYM_TYPED_FUNC_START(aria_aesni_avx_gfni_ctr_crypt_16way)
 	/* input:
 	*      %rdi: ctx
 	*      %rsi: dst

arch/x86/crypto/nh-avx2-x86_64.S

@@ -8,6 +8,7 @@
  */
 
 #include <linux/linkage.h>
+#include <linux/cfi_types.h>
 
 #define		PASS0_SUMS	%ymm0
 #define		PASS1_SUMS	%ymm1
@@ -65,11 +66,11 @@
 
 /*
  * void nh_avx2(const u32 *key, const u8 *message, size_t message_len,
- *		u8 hash[NH_HASH_BYTES])
+ *		__le64 hash[NH_NUM_PASSES])
  *
  * It's guaranteed that message_len % 16 == 0.
  */
-SYM_FUNC_START(nh_avx2)
+SYM_TYPED_FUNC_START(nh_avx2)
 
 	vmovdqu		0x00(KEY), K0
 	vmovdqu		0x10(KEY), K1

arch/x86/crypto/sha1_ni_asm.S

@@ -54,6 +54,7 @@
  */
 
 #include <linux/linkage.h>
+#include <linux/cfi_types.h>
 
 #define DIGEST_PTR	%rdi	/* 1st arg */
 #define DATA_PTR	%rsi	/* 2nd arg */
@@ -93,7 +94,7 @@
  */
 .text
 .align 32
-SYM_FUNC_START(sha1_ni_transform)
+SYM_TYPED_FUNC_START(sha1_ni_transform)
 	push		%rbp
 	mov		%rsp, %rbp
 	sub		$FRAME_SIZE, %rsp