diff options
author | Tianjia Zhang <tianjia.zhang@linux.alibaba.com> | 2022-10-27 14:55:01 +0800 |
---|---|---|
committer | Herbert Xu <herbert@gondor.apana.org.au> | 2022-11-04 17:34:42 +0800 |
commit | b1863fd0742f8da21f6f994e14e820db5831bd74 (patch) | |
tree | 57d45c39d9b0b382b13b9e3fad36a7c9db4ef856 /arch/arm64/crypto/sm4-ce-core.S | |
parent | 45089dbe5952e9afbe2a3b3054105f2a694930f1 (diff) |
crypto: arm64/sm4 - add CE implementation for CTS-CBC mode
This patch is a CE-optimized assembly implementation for CTS-CBC mode.
Benchmark on T-Head Yitian-710 2.75 GHz, the data comes from the 218 mode of
tcrypt, and compared the performance before and after this patch (the driver
used before this patch is cts(cbc-sm4-ce)). The abscissas are blocks of
different lengths. The data is tabulated and the unit is Mb/s:
Before:
cts(cbc-sm4-ce) | 16 64 128 256 1024 1420 4096
----------------+--------------------------------------------------------------
CTS-CBC enc | 286.09 297.17 457.97 627.75 868.58 900.80 957.69
CTS-CBC dec | 286.67 285.63 538.35 947.08 2241.03 2577.32 3391.14
After:
cts-cbc-sm4-ce | 16 64 128 256 1024 1420 4096
----------------+--------------------------------------------------------------
CTS-CBC enc | 288.19 428.80 593.57 741.04 911.73 931.80 950.00
CTS-CBC dec | 292.22 468.99 838.23 1380.76 2741.17 3036.42 3409.62
Signed-off-by: Tianjia Zhang <tianjia.zhang@linux.alibaba.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Diffstat (limited to 'arch/arm64/crypto/sm4-ce-core.S')
-rw-r--r-- | arch/arm64/crypto/sm4-ce-core.S | 102 |
1 files changed, 102 insertions, 0 deletions
diff --git a/arch/arm64/crypto/sm4-ce-core.S b/arch/arm64/crypto/sm4-ce-core.S index 9e4b4f01cdf3..414d29f8110b 100644 --- a/arch/arm64/crypto/sm4-ce-core.S +++ b/arch/arm64/crypto/sm4-ce-core.S @@ -307,6 +307,100 @@ SYM_FUNC_START(sm4_ce_cbc_dec) SYM_FUNC_END(sm4_ce_cbc_dec) .align 3 +SYM_FUNC_START(sm4_ce_cbc_cts_enc) + /* input: + * x0: round key array, CTX + * x1: dst + * x2: src + * x3: iv (big endian, 128 bit) + * w4: nbytes + */ + SM4_PREPARE(x0) + + sub w5, w4, #16 + uxtw x5, w5 + + ld1 {RIV.16b}, [x3] + + ld1 {v0.16b}, [x2] + eor RIV.16b, RIV.16b, v0.16b + SM4_CRYPT_BLK(RIV) + + /* load permute table */ + adr_l x6, .Lcts_permute_table + add x7, x6, #32 + add x6, x6, x5 + sub x7, x7, x5 + ld1 {v3.16b}, [x6] + ld1 {v4.16b}, [x7] + + /* overlapping loads */ + add x2, x2, x5 + ld1 {v1.16b}, [x2] + + /* create Cn from En-1 */ + tbl v0.16b, {RIV.16b}, v3.16b + /* padding Pn with zeros */ + tbl v1.16b, {v1.16b}, v4.16b + + eor v1.16b, v1.16b, RIV.16b + SM4_CRYPT_BLK(v1) + + /* overlapping stores */ + add x5, x1, x5 + st1 {v0.16b}, [x5] + st1 {v1.16b}, [x1] + + ret +SYM_FUNC_END(sm4_ce_cbc_cts_enc) + +.align 3 +SYM_FUNC_START(sm4_ce_cbc_cts_dec) + /* input: + * x0: round key array, CTX + * x1: dst + * x2: src + * x3: iv (big endian, 128 bit) + * w4: nbytes + */ + SM4_PREPARE(x0) + + sub w5, w4, #16 + uxtw x5, w5 + + ld1 {RIV.16b}, [x3] + + /* load permute table */ + adr_l x6, .Lcts_permute_table + add x7, x6, #32 + add x6, x6, x5 + sub x7, x7, x5 + ld1 {v3.16b}, [x6] + ld1 {v4.16b}, [x7] + + /* overlapping loads */ + ld1 {v0.16b}, [x2], x5 + ld1 {v1.16b}, [x2] + + SM4_CRYPT_BLK(v0) + /* select the first Ln bytes of Xn to create Pn */ + tbl v2.16b, {v0.16b}, v3.16b + eor v2.16b, v2.16b, v1.16b + + /* overwrite the first Ln bytes with Cn to create En-1 */ + tbx v0.16b, {v1.16b}, v4.16b + SM4_CRYPT_BLK(v0) + eor v0.16b, v0.16b, RIV.16b + + /* overlapping stores */ + add x5, x1, x5 + st1 {v2.16b}, [x5] + st1 {v0.16b}, [x1] + + ret +SYM_FUNC_END(sm4_ce_cbc_cts_dec) + +.align 3 SYM_FUNC_START(sm4_ce_cfb_enc) /* input: * x0: round key array, CTX @@ -576,3 +670,11 @@ SYM_FUNC_END(sm4_ce_ctr_enc) .Lbswap128_mask: .byte 0x0c, 0x0d, 0x0e, 0x0f, 0x08, 0x09, 0x0a, 0x0b .byte 0x04, 0x05, 0x06, 0x07, 0x00, 0x01, 0x02, 0x03 + +.Lcts_permute_table: + .byte 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff + .byte 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff + .byte 0x0, 0x1, 0x2, 0x3, 0x4, 0x5, 0x6, 0x7 + .byte 0x8, 0x9, 0xa, 0xb, 0xc, 0xd, 0xe, 0xf + .byte 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff + .byte 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff |