crypto: powerpc - Add CRC-T10DIF acceleration

T10DIF is a CRC16 used heavily in NVMe.

It turns out we can accelerate it with a CRC32 library and a few
little tricks.

Provide the accelerator based the refactored CRC32 code.

Cc: Anton Blanchard <anton@samba.org>
Thanks-to: Hong Bo Peng <penghb@cn.ibm.com>
Signed-off-by: Daniel Axtens <dja@axtens.net>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>

arch/powerpc/crypto/Makefile

@@ -10,6 +10,7 @@ obj-$(CONFIG_CRYPTO_SHA1_PPC) += sha1-powerpc.o
 obj-$(CONFIG_CRYPTO_SHA1_PPC_SPE) += sha1-ppc-spe.o
 obj-$(CONFIG_CRYPTO_SHA256_PPC_SPE) += sha256-ppc-spe.o
 obj-$(CONFIG_CRYPTO_CRC32C_VPMSUM) += crc32c-vpmsum.o
+obj-$(CONFIG_CRYPTO_CRCT10DIF_VPMSUM) += crct10dif-vpmsum.o
 
 aes-ppc-spe-y := aes-spe-core.o aes-spe-keys.o aes-tab-4k.o aes-spe-modes.o aes-spe-glue.o
 md5-ppc-y := md5-asm.o md5-glue.o
@@ -17,3 +18,4 @@ sha1-powerpc-y := sha1-powerpc-asm.o sha1.o
 sha1-ppc-spe-y := sha1-spe-asm.o sha1-spe-glue.o
 sha256-ppc-spe-y := sha256-spe-asm.o sha256-spe-glue.o
 crc32c-vpmsum-y := crc32c-vpmsum_asm.o crc32c-vpmsum_glue.o
+crct10dif-vpmsum-y := crct10dif-vpmsum_asm.o crct10dif-vpmsum_glue.o

arch/powerpc/crypto/crct10dif-vpmsum_glue.c

+/*
+ * Calculate a CRC T10-DIF with vpmsum acceleration
+ *
+ * Copyright 2017, Daniel Axtens, IBM Corporation.
+ * [based on crc32c-vpmsum_glue.c]
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; either version 2
+ * of the License, or (at your option) any later version.
+ */
+
+#include <linux/crc-t10dif.h>
+#include <crypto/internal/hash.h>
+#include <linux/init.h>
+#include <linux/module.h>
+#include <linux/string.h>
+#include <linux/kernel.h>
+#include <linux/cpufeature.h>
+#include <asm/switch_to.h>
+
+#define VMX_ALIGN		16
+#define VMX_ALIGN_MASK		(VMX_ALIGN-1)
+
+#define VECTOR_BREAKPOINT	64
+
+u32 __crct10dif_vpmsum(u32 crc, unsigned char const *p, size_t len);
+
+static u16 crct10dif_vpmsum(u16 crci, unsigned char const *p, size_t len)
+{
+	unsigned int prealign;
+	unsigned int tail;
+	u32 crc = crci;
+
+	if (len < (VECTOR_BREAKPOINT + VMX_ALIGN) || in_interrupt())
+		return crc_t10dif_generic(crc, p, len);
+
+	if ((unsigned long)p & VMX_ALIGN_MASK) {
+		prealign = VMX_ALIGN - ((unsigned long)p & VMX_ALIGN_MASK);
+		crc = crc_t10dif_generic(crc, p, prealign);
+		len -= prealign;
+		p += prealign;
+	}
+
+	if (len & ~VMX_ALIGN_MASK) {
+		crc <<= 16;
+		pagefault_disable();
+		enable_kernel_altivec();
+		crc = __crct10dif_vpmsum(crc, p, len & ~VMX_ALIGN_MASK);
+		pagefault_enable();
+		crc >>= 16;
+	}
+
+	tail = len & VMX_ALIGN_MASK;
+	if (tail) {
+		p += len & ~VMX_ALIGN_MASK;
+		crc = crc_t10dif_generic(crc, p, tail);
+	}
+
+	return crc & 0xffff;
+}
+
+static int crct10dif_vpmsum_init(struct shash_desc *desc)
+{
+	u16 *crc = shash_desc_ctx(desc);
+
+	*crc = 0;
+	return 0;
+}
+
+static int crct10dif_vpmsum_update(struct shash_desc *desc, const u8 *data,
+			    unsigned int length)
+{
+	u16 *crc = shash_desc_ctx(desc);
+
+	*crc = crct10dif_vpmsum(*crc, data, length);
+
+	return 0;
+}
+
+
+static int crct10dif_vpmsum_final(struct shash_desc *desc, u8 *out)
+{
+	u16 *crcp = shash_desc_ctx(desc);
+
+	*(u16 *)out = *crcp;
+	return 0;
+}
+
+static struct shash_alg alg = {
+	.init		= crct10dif_vpmsum_init,
+	.update		= crct10dif_vpmsum_update,
+	.final		= crct10dif_vpmsum_final,
+	.descsize	= CRC_T10DIF_DIGEST_SIZE,
+	.digestsize	= CRC_T10DIF_DIGEST_SIZE,
+	.base		= {
+		.cra_name		= "crct10dif",
+		.cra_driver_name	= "crct10dif-vpmsum",
+		.cra_priority		= 200,
+		.cra_blocksize		= CRC_T10DIF_BLOCK_SIZE,
+		.cra_module		= THIS_MODULE,
+	}
+};
+
+static int __init crct10dif_vpmsum_mod_init(void)
+{
+	if (!cpu_has_feature(CPU_FTR_ARCH_207S))
+		return -ENODEV;
+
+	return crypto_register_shash(&alg);
+}
+
+static void __exit crct10dif_vpmsum_mod_fini(void)
+{
+	crypto_unregister_shash(&alg);
+}
+
+module_cpu_feature_match(PPC_MODULE_FEATURE_VEC_CRYPTO, crct10dif_vpmsum_mod_init);
+module_exit(crct10dif_vpmsum_mod_fini);
+
+MODULE_AUTHOR("Daniel Axtens <dja@axtens.net>");
+MODULE_DESCRIPTION("CRCT10DIF using vector polynomial multiply-sum instructions");
+MODULE_LICENSE("GPL");
+MODULE_ALIAS_CRYPTO("crct10dif");
+MODULE_ALIAS_CRYPTO("crct10dif-vpmsum");

crypto/Kconfig

@@ -513,6 +513,15 @@ config CRYPTO_CRCT10DIF_PCLMUL
 	  'crct10dif-plcmul' module, which is faster when computing the
 	  crct10dif checksum as compared with the generic table implementation.
 
+config CRYPTO_CRCT10DIF_VPMSUM
+	tristate "CRC32T10DIF powerpc64 hardware acceleration"
+	depends on PPC64 && ALTIVEC && CRC_T10DIF
+	select CRYPTO_HASH
+	help
+	  CRC10T10DIF algorithm implemented using vector polynomial
+	  multiply-sum (vpmsum) instructions, introduced in POWER8. Enable on
+	  POWER8 and newer processors for improved performance.
+
 config CRYPTO_GHASH
 	tristate "GHASH digest algorithm"
 	select CRYPTO_GF128MUL