To take advantage of the hardware pipeline implementation of AES-NI
instructions. CTR mode cryption is implemented in ASM to schedule
multiple AES-NI instructions one after another. This way, some latency
of AES-NI instruction can be eliminated.
Performance testing based on dm-crypt should 50% reduction of
ecryption/decryption time.
Signed-off-by: Huang Ying <[email protected]>
---
arch/x86/crypto/aesni-intel_asm.S | 115 ++++++++++++++++++++++++++++++++
arch/x86/crypto/aesni-intel_glue.c | 130 +++++++++++++++++++++++++++++++++++--
2 files changed, 238 insertions(+), 7 deletions(-)
--- a/arch/x86/crypto/aesni-intel_asm.S
+++ b/arch/x86/crypto/aesni-intel_asm.S
@@ -32,6 +32,9 @@
#define IN IN1
#define KEY %xmm2
#define IV %xmm3
+#define BSWAP_MASK %xmm10
+#define CTR %xmm11
+#define INC %xmm12
#define KEYP %rdi
#define OUTP %rsi
@@ -42,6 +45,7 @@
#define T1 %r10
#define TKEYP T1
#define T2 %r11
+#define TCTR_LOW T2
_key_expansion_128:
_key_expansion_256a:
@@ -724,3 +728,114 @@ ENTRY(aesni_cbc_dec)
movups IV, (IVP)
.Lcbc_dec_just_ret:
ret
+
+.align 16
+.Lbswap_mask:
+ .byte 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0
+
+/*
+ * _aesni_inc_init: internal ABI
+ * setup registers used by _aesni_inc
+ * input:
+ * IV
+ * output:
+ * CTR: == IV, in little endian
+ * TCTR_LOW: == lower qword of CTR
+ * INC: == 1, in little endian
+ * BSWAP_MASK == endian swapping mask
+ */
+_aesni_inc_init:
+ movaps .Lbswap_mask, BSWAP_MASK
+ movaps IV, CTR
+ PSHUFB_XMM BSWAP_MASK CTR
+ mov $1, TCTR_LOW
+ movq TCTR_LOW, INC
+ movq CTR, TCTR_LOW
+ ret
+
+/*
+ * _aesni_inc: internal ABI
+ * Increase IV by 1, IV is in big endian
+ * input:
+ * IV
+ * CTR: == IV, in little endian
+ * TCTR_LOW: == lower qword of CTR
+ * INC: == 1, in little endian
+ * BSWAP_MASK == endian swapping mask
+ * output:
+ * IV: Increase by 1
+ * changed:
+ * CTR: == output IV, in little endian
+ * TCTR_LOW: == lower qword of CTR
+ */
+_aesni_inc:
+ paddq INC, CTR
+ add $1, TCTR_LOW
+ jnc .Linc_low
+ pslldq $8, INC
+ paddq INC, CTR
+ psrldq $8, INC
+.Linc_low:
+ movaps CTR, IV
+ PSHUFB_XMM BSWAP_MASK IV
+ ret
+
+/*
+ * void aesni_ctr_enc(struct crypto_aes_ctx *ctx, const u8 *dst, u8 *src,
+ * size_t len, u8 *iv)
+ */
+ENTRY(aesni_ctr_enc)
+ cmp $16, LEN
+ jb .Lctr_enc_just_ret
+ mov 480(KEYP), KLEN
+ movups (IVP), IV
+ call _aesni_inc_init
+ cmp $64, LEN
+ jb .Lctr_enc_loop1
+.align 4
+.Lctr_enc_loop4:
+ movaps IV, STATE1
+ call _aesni_inc
+ movups (INP), IN1
+ movaps IV, STATE2
+ call _aesni_inc
+ movups 0x10(INP), IN2
+ movaps IV, STATE3
+ call _aesni_inc
+ movups 0x20(INP), IN3
+ movaps IV, STATE4
+ call _aesni_inc
+ movups 0x30(INP), IN4
+ call _aesni_enc4
+ pxor IN1, STATE1
+ movups STATE1, (OUTP)
+ pxor IN2, STATE2
+ movups STATE2, 0x10(OUTP)
+ pxor IN3, STATE3
+ movups STATE3, 0x20(OUTP)
+ pxor IN4, STATE4
+ movups STATE4, 0x30(OUTP)
+ sub $64, LEN
+ add $64, INP
+ add $64, OUTP
+ cmp $64, LEN
+ jge .Lctr_enc_loop4
+ cmp $16, LEN
+ jb .Lctr_enc_ret
+.align 4
+.Lctr_enc_loop1:
+ movaps IV, STATE
+ call _aesni_inc
+ movups (INP), IN
+ call _aesni_enc1
+ pxor IN, STATE
+ movups STATE, (OUTP)
+ sub $16, LEN
+ add $16, INP
+ add $16, OUTP
+ cmp $16, LEN
+ jge .Lctr_enc_loop1
+.Lctr_enc_ret:
+ movups IV, (IVP)
+.Lctr_enc_just_ret:
+ ret
--- a/arch/x86/crypto/aesni-intel_glue.c
+++ b/arch/x86/crypto/aesni-intel_glue.c
@@ -18,6 +18,7 @@
#include <crypto/algapi.h>
#include <crypto/aes.h>
#include <crypto/cryptd.h>
+#include <crypto/ctr.h>
#include <asm/i387.h>
#include <asm/aes.h>
@@ -58,6 +59,8 @@ asmlinkage void aesni_cbc_enc(struct cry
const u8 *in, unsigned int len, u8 *iv);
asmlinkage void aesni_cbc_dec(struct crypto_aes_ctx *ctx, u8 *out,
const u8 *in, unsigned int len, u8 *iv);
+asmlinkage void aesni_ctr_enc(struct crypto_aes_ctx *ctx, u8 *out,
+ const u8 *in, unsigned int len, u8 *iv);
static inline struct crypto_aes_ctx *aes_ctx(void *raw_ctx)
{
@@ -321,6 +324,72 @@ static struct crypto_alg blk_cbc_alg = {
},
};
+static void ctr_crypt_final(struct crypto_aes_ctx *ctx,
+ struct blkcipher_walk *walk)
+{
+ u8 *ctrblk = walk->iv;
+ u8 keystream[AES_BLOCK_SIZE];
+ u8 *src = walk->src.virt.addr;
+ u8 *dst = walk->dst.virt.addr;
+ unsigned int nbytes = walk->nbytes;
+
+ aesni_enc(ctx, keystream, ctrblk);
+ crypto_xor(keystream, src, nbytes);
+ memcpy(dst, keystream, nbytes);
+ crypto_inc(ctrblk, AES_BLOCK_SIZE);
+}
+
+static int ctr_crypt(struct blkcipher_desc *desc,
+ struct scatterlist *dst, struct scatterlist *src,
+ unsigned int nbytes)
+{
+ struct crypto_aes_ctx *ctx = aes_ctx(crypto_blkcipher_ctx(desc->tfm));
+ struct blkcipher_walk walk;
+ int err;
+
+ blkcipher_walk_init(&walk, dst, src, nbytes);
+ err = blkcipher_walk_virt_block(desc, &walk, AES_BLOCK_SIZE);
+ desc->flags &= ~CRYPTO_TFM_REQ_MAY_SLEEP;
+
+ kernel_fpu_begin();
+ while ((nbytes = walk.nbytes) >= AES_BLOCK_SIZE) {
+ aesni_ctr_enc(ctx, walk.dst.virt.addr, walk.src.virt.addr,
+ nbytes & AES_BLOCK_MASK, walk.iv);
+ nbytes &= AES_BLOCK_SIZE - 1;
+ err = blkcipher_walk_done(desc, &walk, nbytes);
+ }
+ if (walk.nbytes) {
+ ctr_crypt_final(ctx, &walk);
+ err = blkcipher_walk_done(desc, &walk, 0);
+ }
+ kernel_fpu_end();
+
+ return err;
+}
+
+static struct crypto_alg blk_ctr_alg = {
+ .cra_name = "__ctr-aes-aesni",
+ .cra_driver_name = "__driver-ctr-aes-aesni",
+ .cra_priority = 0,
+ .cra_flags = CRYPTO_ALG_TYPE_BLKCIPHER,
+ .cra_blocksize = 1,
+ .cra_ctxsize = sizeof(struct crypto_aes_ctx)+AESNI_ALIGN-1,
+ .cra_alignmask = 0,
+ .cra_type = &crypto_blkcipher_type,
+ .cra_module = THIS_MODULE,
+ .cra_list = LIST_HEAD_INIT(blk_ctr_alg.cra_list),
+ .cra_u = {
+ .blkcipher = {
+ .min_keysize = AES_MIN_KEY_SIZE,
+ .max_keysize = AES_MAX_KEY_SIZE,
+ .ivsize = AES_BLOCK_SIZE,
+ .setkey = aes_set_key,
+ .encrypt = ctr_crypt,
+ .decrypt = ctr_crypt,
+ },
+ },
+};
+
static int ablk_set_key(struct crypto_ablkcipher *tfm, const u8 *key,
unsigned int key_len)
{
@@ -467,13 +536,11 @@ static struct crypto_alg ablk_cbc_alg =
},
};
-#ifdef HAS_CTR
static int ablk_ctr_init(struct crypto_tfm *tfm)
{
struct cryptd_ablkcipher *cryptd_tfm;
- cryptd_tfm = cryptd_alloc_ablkcipher("fpu(ctr(__driver-aes-aesni))",
- 0, 0);
+ cryptd_tfm = cryptd_alloc_ablkcipher("__driver-ctr-aes-aesni", 0, 0);
if (IS_ERR(cryptd_tfm))
return PTR_ERR(cryptd_tfm);
ablk_init_common(tfm, cryptd_tfm);
@@ -500,11 +567,50 @@ static struct crypto_alg ablk_ctr_alg =
.ivsize = AES_BLOCK_SIZE,
.setkey = ablk_set_key,
.encrypt = ablk_encrypt,
- .decrypt = ablk_decrypt,
+ .decrypt = ablk_encrypt,
.geniv = "chainiv",
},
},
};
+
+#ifdef HAS_CTR
+static int ablk_rfc3686_ctr_init(struct crypto_tfm *tfm)
+{
+ struct cryptd_ablkcipher *cryptd_tfm;
+
+ cryptd_tfm = cryptd_alloc_ablkcipher(
+ "rfc3686(__driver-ctr-aes-aesni)", 0, 0);
+ if (IS_ERR(cryptd_tfm))
+ return PTR_ERR(cryptd_tfm);
+ ablk_init_common(tfm, cryptd_tfm);
+ return 0;
+}
+
+static struct crypto_alg ablk_rfc3686_ctr_alg = {
+ .cra_name = "rfc3686(ctr(aes))",
+ .cra_driver_name = "rfc3686-ctr-aes-aesni",
+ .cra_priority = 400,
+ .cra_flags = CRYPTO_ALG_TYPE_ABLKCIPHER|CRYPTO_ALG_ASYNC,
+ .cra_blocksize = 1,
+ .cra_ctxsize = sizeof(struct async_aes_ctx),
+ .cra_alignmask = 0,
+ .cra_type = &crypto_ablkcipher_type,
+ .cra_module = THIS_MODULE,
+ .cra_list = LIST_HEAD_INIT(ablk_rfc3686_ctr_alg.cra_list),
+ .cra_init = ablk_rfc3686_ctr_init,
+ .cra_exit = ablk_exit,
+ .cra_u = {
+ .ablkcipher = {
+ .min_keysize = AES_MIN_KEY_SIZE+CTR_RFC3686_NONCE_SIZE,
+ .max_keysize = AES_MAX_KEY_SIZE+CTR_RFC3686_NONCE_SIZE,
+ .ivsize = CTR_RFC3686_IV_SIZE,
+ .setkey = ablk_set_key,
+ .encrypt = ablk_encrypt,
+ .decrypt = ablk_decrypt,
+ .geniv = "seqiv",
+ },
+ },
+};
#endif
#ifdef HAS_LRW
@@ -640,13 +746,17 @@ static int __init aesni_init(void)
goto blk_ecb_err;
if ((err = crypto_register_alg(&blk_cbc_alg)))
goto blk_cbc_err;
+ if ((err = crypto_register_alg(&blk_ctr_alg)))
+ goto blk_ctr_err;
if ((err = crypto_register_alg(&ablk_ecb_alg)))
goto ablk_ecb_err;
if ((err = crypto_register_alg(&ablk_cbc_alg)))
goto ablk_cbc_err;
-#ifdef HAS_CTR
if ((err = crypto_register_alg(&ablk_ctr_alg)))
goto ablk_ctr_err;
+#ifdef HAS_CTR
+ if ((err = crypto_register_alg(&ablk_rfc3686_ctr_alg)))
+ goto ablk_rfc3686_ctr_err;
#endif
#ifdef HAS_LRW
if ((err = crypto_register_alg(&ablk_lrw_alg)))
@@ -675,13 +785,17 @@ ablk_pcbc_err:
ablk_lrw_err:
#endif
#ifdef HAS_CTR
+ crypto_unregister_alg(&ablk_rfc3686_ctr_alg);
+ablk_rfc3686_ctr_err:
+#endif
crypto_unregister_alg(&ablk_ctr_alg);
ablk_ctr_err:
-#endif
crypto_unregister_alg(&ablk_cbc_alg);
ablk_cbc_err:
crypto_unregister_alg(&ablk_ecb_alg);
ablk_ecb_err:
+ crypto_unregister_alg(&blk_ctr_alg);
+blk_ctr_err:
crypto_unregister_alg(&blk_cbc_alg);
blk_cbc_err:
crypto_unregister_alg(&blk_ecb_alg);
@@ -705,10 +819,12 @@ static void __exit aesni_exit(void)
crypto_unregister_alg(&ablk_lrw_alg);
#endif
#ifdef HAS_CTR
- crypto_unregister_alg(&ablk_ctr_alg);
+ crypto_unregister_alg(&ablk_rfc3686_ctr_alg);
#endif
+ crypto_unregister_alg(&ablk_ctr_alg);
crypto_unregister_alg(&ablk_cbc_alg);
crypto_unregister_alg(&ablk_ecb_alg);
+ crypto_unregister_alg(&blk_ctr_alg);
crypto_unregister_alg(&blk_cbc_alg);
crypto_unregister_alg(&blk_ecb_alg);
crypto_unregister_alg(&__aesni_alg);
On Mon, Mar 01, 2010 at 02:14:36PM +0800, Huang Ying wrote:
> To take advantage of the hardware pipeline implementation of AES-NI
> instructions. CTR mode cryption is implemented in ASM to schedule
> multiple AES-NI instructions one after another. This way, some latency
> of AES-NI instruction can be eliminated.
>
> Performance testing based on dm-crypt should 50% reduction of
> ecryption/decryption time.
>
> Signed-off-by: Huang Ying <[email protected]>
Applied to cryptodev. Thanks!
--
Visit Openswan at http://www.openswan.org/
Email: Herbert Xu ~{PmV>HI~} <[email protected]>
Home Page: http://gondor.apana.org.au/~herbert/
PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt
On Mon, 01 Mar 2010 14:14:36 +0800
Huang Ying <[email protected]> wrote:
> To take advantage of the hardware pipeline implementation of AES-NI
> instructions. CTR mode cryption is implemented in ASM to schedule
> multiple AES-NI instructions one after another. This way, some latency
> of AES-NI instruction can be eliminated.
>
> Performance testing based on dm-crypt should 50% reduction of
> ecryption/decryption time.
>
> ...
>
> +/*
> + * _aesni_inc_init: internal ABI
> + * setup registers used by _aesni_inc
> + * input:
> + * IV
> + * output:
> + * CTR: == IV, in little endian
> + * TCTR_LOW: == lower qword of CTR
> + * INC: == 1, in little endian
> + * BSWAP_MASK == endian swapping mask
> + */
> +_aesni_inc_init:
> + movaps .Lbswap_mask, BSWAP_MASK
> + movaps IV, CTR
> + PSHUFB_XMM BSWAP_MASK CTR
> + mov $1, TCTR_LOW
> + movq TCTR_LOW, INC
> + movq CTR, TCTR_LOW
^^ these two lines don't assemble with gas 2.16.1:
arch/x86/crypto/aesni-intel_asm.S: Assembler messages:
arch/x86/crypto/aesni-intel_asm.S:752: Error: suffix or operands invalid for `movq'
arch/x86/crypto/aesni-intel_asm.S:753: Error: suffix or operands invalid for `movq'
> + ret
On Thu, Mar 11, 2010 at 12:15:10PM -0800, Andrew Morton wrote:
>
> > + movq TCTR_LOW, INC
> > + movq CTR, TCTR_LOW
>
> ^^ these two lines don't assemble with gas 2.16.1:
>
> arch/x86/crypto/aesni-intel_asm.S: Assembler messages:
> arch/x86/crypto/aesni-intel_asm.S:752: Error: suffix or operands invalid for `movq'
> arch/x86/crypto/aesni-intel_asm.S:753: Error: suffix or operands invalid for `movq'
I guess we will have to hard code this. Thanks for reporting
this Andrew.
--
Visit Openswan at http://www.openswan.org/
Email: Herbert Xu ~{PmV>HI~} <[email protected]>
Home Page: http://gondor.apana.org.au/~herbert/
PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt