AuxXxilium/linux_dsm_epyc7002
1
0
Fork 0
mirror of https://github.com/AuxXxilium/linux_dsm_epyc7002.git synced 2024-12-18 17:36:49 +07:00
Code Issues Actions Packages Projects Releases Wiki Activity
5158917c7b
linux_dsm_epyc7002/arch/arm64/crypto/crct10dif-ce-core.S
Ard Biesheuvel 2fffee536c crypto: arm64/crct10dif - implement non-Crypto Extensions alternative 
The arm64 implementation of the CRC-T10DIF algorithm uses the 64x64 bit
polynomial multiplication instructions, which are optional in the
architecture, and if these instructions are not available, we fall back
to the C routine which is slow and inefficient.

So let's reuse the 64x64 bit PMULL alternative from the GHASH driver that
uses a sequence of ~40 instructions involving 8x8 bit PMULL and some
shifting and masking. This is a lot slower than the original, but it is
still twice as fast as the current [unoptimized] C code on Cortex-A53,
and it is time invariant and much easier on the D-cache.

Signed-off-by: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2018-09-04 11:37:04 +08:00

582 lines
16 KiB
ArmAsm
Raw Blame History

//
// Accelerated CRC-T10DIF using arm64 NEON and Crypto Extensions instructions
//
// Copyright (C) 2016 Linaro Ltd <ard.biesheuvel@linaro.org>
//
// This program is free software; you can redistribute it and/or modify
// it under the terms of the GNU General Public License version 2 as
// published by the Free Software Foundation.
//
//
// Implement fast CRC-T10DIF computation with SSE and PCLMULQDQ instructions
//
// Copyright (c) 2013, Intel Corporation
//
// Authors:
// Erdinc Ozturk <erdinc.ozturk@intel.com>
// Vinodh Gopal <vinodh.gopal@intel.com>
// James Guilford <james.guilford@intel.com>
// Tim Chen <tim.c.chen@linux.intel.com>
//
// This software is available to you under a choice of one of two
// licenses. You may choose to be licensed under the terms of the GNU
// General Public License (GPL) Version 2, available from the file
// COPYING in the main directory of this source tree, or the
// OpenIB.org BSD license below:
//
// Redistribution and use in source and binary forms, with or without
// modification, are permitted provided that the following conditions are
// met:
//
// * Redistributions of source code must retain the above copyright
// notice, this list of conditions and the following disclaimer.
//
// * Redistributions in binary form must reproduce the above copyright
// notice, this list of conditions and the following disclaimer in the
// documentation and/or other materials provided with the
// distribution.
//
// * Neither the name of the Intel Corporation nor the names of its
// contributors may be used to endorse or promote products derived from
// this software without specific prior written permission.
//
//
// THIS SOFTWARE IS PROVIDED BY INTEL CORPORATION ""AS IS"" AND ANY
// EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
// IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR
// PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL INTEL CORPORATION OR
// CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL,
// EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO,
// PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR
// PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF
// LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING
// NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
// SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
//
// Function API:
// UINT16 crc_t10dif_pcl(
// UINT16 init_crc, //initial CRC value, 16 bits
// const unsigned char *buf, //buffer pointer to calculate CRC on
// UINT64 len //buffer length in bytes (64-bit data)
// );
//
// Reference paper titled "Fast CRC Computation for Generic
// Polynomials Using PCLMULQDQ Instruction"
// URL: http://www.intel.com/content/dam/www/public/us/en/documents
// /white-papers/fast-crc-computation-generic-polynomials-pclmulqdq-paper.pdf
//
//
#include <linux/linkage.h>
#include <asm/assembler.h>
.text
.cpu generic+crypto
arg1_low32 .req w19
arg2 .req x20
arg3 .req x21
vzr .req v13
ad .req v14
bd .req v10
k00_16 .req v15
k32_48 .req v16
t3 .req v17
t4 .req v18
t5 .req v19
t6 .req v20
t7 .req v21
t8 .req v22
t9 .req v23
perm1 .req v24
perm2 .req v25
perm3 .req v26
perm4 .req v27
bd1 .req v28
bd2 .req v29
bd3 .req v30
bd4 .req v31
.macro __pmull_init_p64
.endm
.macro __pmull_pre_p64, bd
.endm
.macro __pmull_init_p8
// k00_16 := 0x0000000000000000_000000000000ffff
// k32_48 := 0x00000000ffffffff_0000ffffffffffff
movi k32_48.2d, #0xffffffff
mov k32_48.h[2], k32_48.h[0]
ushr k00_16.2d, k32_48.2d, #32
// prepare the permutation vectors
mov_q x5, 0x080f0e0d0c0b0a09
movi perm4.8b, #8
dup perm1.2d, x5
eor perm1.16b, perm1.16b, perm4.16b
ushr perm2.2d, perm1.2d, #8
ushr perm3.2d, perm1.2d, #16
ushr perm4.2d, perm1.2d, #24
sli perm2.2d, perm1.2d, #56
sli perm3.2d, perm1.2d, #48
sli perm4.2d, perm1.2d, #40
.endm
.macro __pmull_pre_p8, bd
tbl bd1.16b, {\bd\().16b}, perm1.16b
tbl bd2.16b, {\bd\().16b}, perm2.16b
tbl bd3.16b, {\bd\().16b}, perm3.16b
tbl bd4.16b, {\bd\().16b}, perm4.16b
.endm
__pmull_p8_core:
.L__pmull_p8_core:
ext t4.8b, ad.8b, ad.8b, #1 // A1
ext t5.8b, ad.8b, ad.8b, #2 // A2
ext t6.8b, ad.8b, ad.8b, #3 // A3
pmull t4.8h, t4.8b, bd.8b // F = A1*B
pmull t8.8h, ad.8b, bd1.8b // E = A*B1
pmull t5.8h, t5.8b, bd.8b // H = A2*B
pmull t7.8h, ad.8b, bd2.8b // G = A*B2
pmull t6.8h, t6.8b, bd.8b // J = A3*B
pmull t9.8h, ad.8b, bd3.8b // I = A*B3
pmull t3.8h, ad.8b, bd4.8b // K = A*B4
b 0f
.L__pmull_p8_core2:
tbl t4.16b, {ad.16b}, perm1.16b // A1
tbl t5.16b, {ad.16b}, perm2.16b // A2
tbl t6.16b, {ad.16b}, perm3.16b // A3
pmull2 t4.8h, t4.16b, bd.16b // F = A1*B
pmull2 t8.8h, ad.16b, bd1.16b // E = A*B1
pmull2 t5.8h, t5.16b, bd.16b // H = A2*B
pmull2 t7.8h, ad.16b, bd2.16b // G = A*B2
pmull2 t6.8h, t6.16b, bd.16b // J = A3*B
pmull2 t9.8h, ad.16b, bd3.16b // I = A*B3
pmull2 t3.8h, ad.16b, bd4.16b // K = A*B4
0: eor t4.16b, t4.16b, t8.16b // L = E + F
eor t5.16b, t5.16b, t7.16b // M = G + H
eor t6.16b, t6.16b, t9.16b // N = I + J
uzp1 t8.2d, t4.2d, t5.2d
uzp2 t4.2d, t4.2d, t5.2d
uzp1 t7.2d, t6.2d, t3.2d
uzp2 t6.2d, t6.2d, t3.2d
// t4 = (L) (P0 + P1) << 8
// t5 = (M) (P2 + P3) << 16
eor t8.16b, t8.16b, t4.16b
and t4.16b, t4.16b, k32_48.16b
// t6 = (N) (P4 + P5) << 24
// t7 = (K) (P6 + P7) << 32
eor t7.16b, t7.16b, t6.16b
and t6.16b, t6.16b, k00_16.16b
eor t8.16b, t8.16b, t4.16b
eor t7.16b, t7.16b, t6.16b
zip2 t5.2d, t8.2d, t4.2d
zip1 t4.2d, t8.2d, t4.2d
zip2 t3.2d, t7.2d, t6.2d
zip1 t6.2d, t7.2d, t6.2d
ext t4.16b, t4.16b, t4.16b, #15
ext t5.16b, t5.16b, t5.16b, #14
ext t6.16b, t6.16b, t6.16b, #13
ext t3.16b, t3.16b, t3.16b, #12
eor t4.16b, t4.16b, t5.16b
eor t6.16b, t6.16b, t3.16b
ret
ENDPROC(__pmull_p8_core)
.macro __pmull_p8, rq, ad, bd, i
.ifnc \bd, v10
.err
.endif
mov ad.16b, \ad\().16b
.ifb \i
pmull \rq\().8h, \ad\().8b, bd.8b // D = A*B
.else
pmull2 \rq\().8h, \ad\().16b, bd.16b // D = A*B
.endif
bl .L__pmull_p8_core\i
eor \rq\().16b, \rq\().16b, t4.16b
eor \rq\().16b, \rq\().16b, t6.16b
.endm
.macro fold64, p, reg1, reg2
ldp q11, q12, [arg2], #0x20
__pmull_\p v8, \reg1, v10, 2
__pmull_\p \reg1, \reg1, v10
CPU_LE( rev64 v11.16b, v11.16b )
CPU_LE( rev64 v12.16b, v12.16b )
__pmull_\p v9, \reg2, v10, 2
__pmull_\p \reg2, \reg2, v10
CPU_LE( ext v11.16b, v11.16b, v11.16b, #8 )
CPU_LE( ext v12.16b, v12.16b, v12.16b, #8 )
eor \reg1\().16b, \reg1\().16b, v8.16b
eor \reg2\().16b, \reg2\().16b, v9.16b
eor \reg1\().16b, \reg1\().16b, v11.16b
eor \reg2\().16b, \reg2\().16b, v12.16b
.endm
.macro fold16, p, reg, rk
__pmull_\p v8, \reg, v10
__pmull_\p \reg, \reg, v10, 2
.ifnb \rk
ldr_l q10, \rk, x8
__pmull_pre_\p v10
.endif
eor v7.16b, v7.16b, v8.16b
eor v7.16b, v7.16b, \reg\().16b
.endm
.macro __pmull_p64, rd, rn, rm, n
.ifb \n
pmull \rd\().1q, \rn\().1d, \rm\().1d
.else
pmull2 \rd\().1q, \rn\().2d, \rm\().2d
.endif
.endm
.macro crc_t10dif_pmull, p
frame_push 3, 128
mov arg1_low32, w0
mov arg2, x1
mov arg3, x2
movi vzr.16b, #0 // init zero register
__pmull_init_\p
// adjust the 16-bit initial_crc value, scale it to 32 bits
lsl arg1_low32, arg1_low32, #16
// check if smaller than 256
cmp arg3, #256
// for sizes less than 128, we can't fold 64B at a time...
b.lt .L_less_than_128_\@
// load the initial crc value
// crc value does not need to be byte-reflected, but it needs
// to be moved to the high part of the register.
// because data will be byte-reflected and will align with
// initial crc at correct place.
movi v10.16b, #0
mov v10.s[3], arg1_low32 // initial crc
// receive the initial 64B data, xor the initial crc value
ldp q0, q1, [arg2]
ldp q2, q3, [arg2, #0x20]
ldp q4, q5, [arg2, #0x40]
ldp q6, q7, [arg2, #0x60]
add arg2, arg2, #0x80
CPU_LE( rev64 v0.16b, v0.16b )
CPU_LE( rev64 v1.16b, v1.16b )
CPU_LE( rev64 v2.16b, v2.16b )
CPU_LE( rev64 v3.16b, v3.16b )
CPU_LE( rev64 v4.16b, v4.16b )
CPU_LE( rev64 v5.16b, v5.16b )
CPU_LE( rev64 v6.16b, v6.16b )
CPU_LE( rev64 v7.16b, v7.16b )
CPU_LE( ext v0.16b, v0.16b, v0.16b, #8 )
CPU_LE( ext v1.16b, v1.16b, v1.16b, #8 )
CPU_LE( ext v2.16b, v2.16b, v2.16b, #8 )
CPU_LE( ext v3.16b, v3.16b, v3.16b, #8 )
CPU_LE( ext v4.16b, v4.16b, v4.16b, #8 )
CPU_LE( ext v5.16b, v5.16b, v5.16b, #8 )
CPU_LE( ext v6.16b, v6.16b, v6.16b, #8 )
CPU_LE( ext v7.16b, v7.16b, v7.16b, #8 )
// XOR the initial_crc value
eor v0.16b, v0.16b, v10.16b
ldr_l q10, rk3, x8 // xmm10 has rk3 and rk4
// type of pmull instruction
// will determine which constant to use
__pmull_pre_\p v10
//
// we subtract 256 instead of 128 to save one instruction from the loop
//
sub arg3, arg3, #256
// at this section of the code, there is 64*x+y (0<=y<64) bytes of
// buffer. The _fold_64_B_loop will fold 64B at a time
// until we have 64+y Bytes of buffer
// fold 64B at a time. This section of the code folds 4 vector
// registers in parallel
.L_fold_64_B_loop_\@:
fold64 \p, v0, v1
fold64 \p, v2, v3
fold64 \p, v4, v5
fold64 \p, v6, v7
subs arg3, arg3, #128
// check if there is another 64B in the buffer to be able to fold
b.lt .L_fold_64_B_end_\@
if_will_cond_yield_neon
stp q0, q1, [sp, #.Lframe_local_offset]
stp q2, q3, [sp, #.Lframe_local_offset + 32]
stp q4, q5, [sp, #.Lframe_local_offset + 64]
stp q6, q7, [sp, #.Lframe_local_offset + 96]
do_cond_yield_neon
ldp q0, q1, [sp, #.Lframe_local_offset]
ldp q2, q3, [sp, #.Lframe_local_offset + 32]
ldp q4, q5, [sp, #.Lframe_local_offset + 64]
ldp q6, q7, [sp, #.Lframe_local_offset + 96]
ldr_l q10, rk3, x8
movi vzr.16b, #0 // init zero register
__pmull_init_\p
__pmull_pre_\p v10
endif_yield_neon
b .L_fold_64_B_loop_\@
.L_fold_64_B_end_\@:
// at this point, the buffer pointer is pointing at the last y Bytes
// of the buffer the 64B of folded data is in 4 of the vector
// registers: v0, v1, v2, v3
// fold the 8 vector registers to 1 vector register with different
// constants
ldr_l q10, rk9, x8
__pmull_pre_\p v10
fold16 \p, v0, rk11
fold16 \p, v1, rk13
fold16 \p, v2, rk15
fold16 \p, v3, rk17
fold16 \p, v4, rk19
fold16 \p, v5, rk1
fold16 \p, v6
// instead of 64, we add 48 to the loop counter to save 1 instruction
// from the loop instead of a cmp instruction, we use the negative
// flag with the jl instruction
adds arg3, arg3, #(128-16)
b.lt .L_final_reduction_for_128_\@
// now we have 16+y bytes left to reduce. 16 Bytes is in register v7
// and the rest is in memory. We can fold 16 bytes at a time if y>=16
// continue folding 16B at a time
.L_16B_reduction_loop_\@:
__pmull_\p v8, v7, v10
__pmull_\p v7, v7, v10, 2
eor v7.16b, v7.16b, v8.16b
ldr q0, [arg2], #16
CPU_LE( rev64 v0.16b, v0.16b )
CPU_LE( ext v0.16b, v0.16b, v0.16b, #8 )
eor v7.16b, v7.16b, v0.16b
subs arg3, arg3, #16
// instead of a cmp instruction, we utilize the flags with the
// jge instruction equivalent of: cmp arg3, 16-16
// check if there is any more 16B in the buffer to be able to fold
b.ge .L_16B_reduction_loop_\@
// now we have 16+z bytes left to reduce, where 0<= z < 16.
// first, we reduce the data in the xmm7 register
.L_final_reduction_for_128_\@:
// check if any more data to fold. If not, compute the CRC of
// the final 128 bits
adds arg3, arg3, #16
b.eq .L_128_done_\@
// here we are getting data that is less than 16 bytes.
// since we know that there was data before the pointer, we can
// offset the input pointer before the actual point, to receive
// exactly 16 bytes. after that the registers need to be adjusted.
.L_get_last_two_regs_\@:
add arg2, arg2, arg3
ldr q1, [arg2, #-16]
CPU_LE( rev64 v1.16b, v1.16b )
CPU_LE( ext v1.16b, v1.16b, v1.16b, #8 )
// get rid of the extra data that was loaded before
// load the shift constant
adr_l x4, tbl_shf_table + 16
sub x4, x4, arg3
ld1 {v0.16b}, [x4]
// shift v2 to the left by arg3 bytes
tbl v2.16b, {v7.16b}, v0.16b
// shift v7 to the right by 16-arg3 bytes
movi v9.16b, #0x80
eor v0.16b, v0.16b, v9.16b
tbl v7.16b, {v7.16b}, v0.16b
// blend
sshr v0.16b, v0.16b, #7 // convert to 8-bit mask
bsl v0.16b, v2.16b, v1.16b
// fold 16 Bytes
__pmull_\p v8, v7, v10
__pmull_\p v7, v7, v10, 2
eor v7.16b, v7.16b, v8.16b
eor v7.16b, v7.16b, v0.16b
.L_128_done_\@:
// compute crc of a 128-bit value
ldr_l q10, rk5, x8 // rk5 and rk6 in xmm10
__pmull_pre_\p v10
// 64b fold
ext v0.16b, vzr.16b, v7.16b, #8
mov v7.d[0], v7.d[1]
__pmull_\p v7, v7, v10
eor v7.16b, v7.16b, v0.16b
// 32b fold
ext v0.16b, v7.16b, vzr.16b, #4
mov v7.s[3], vzr.s[0]
__pmull_\p v0, v0, v10, 2
eor v7.16b, v7.16b, v0.16b
// barrett reduction
ldr_l q10, rk7, x8
__pmull_pre_\p v10
mov v0.d[0], v7.d[1]
__pmull_\p v0, v0, v10
ext v0.16b, vzr.16b, v0.16b, #12
__pmull_\p v0, v0, v10, 2
ext v0.16b, vzr.16b, v0.16b, #12
eor v7.16b, v7.16b, v0.16b
mov w0, v7.s[1]
.L_cleanup_\@:
// scale the result back to 16 bits
lsr x0, x0, #16
frame_pop
ret
.L_less_than_128_\@:
cbz arg3, .L_cleanup_\@
movi v0.16b, #0
mov v0.s[3], arg1_low32 // get the initial crc value
ldr q7, [arg2], #0x10
CPU_LE( rev64 v7.16b, v7.16b )
CPU_LE( ext v7.16b, v7.16b, v7.16b, #8 )
eor v7.16b, v7.16b, v0.16b // xor the initial crc value
cmp arg3, #16
b.eq .L_128_done_\@ // exactly 16 left
b.lt .L_less_than_16_left_\@
ldr_l q10, rk1, x8 // rk1 and rk2 in xmm10
__pmull_pre_\p v10
// update the counter. subtract 32 instead of 16 to save one
// instruction from the loop
subs arg3, arg3, #32
b.ge .L_16B_reduction_loop_\@
add arg3, arg3, #16
b .L_get_last_two_regs_\@
.L_less_than_16_left_\@:
// shl r9, 4
adr_l x0, tbl_shf_table + 16
sub x0, x0, arg3
ld1 {v0.16b}, [x0]
movi v9.16b, #0x80
eor v0.16b, v0.16b, v9.16b
tbl v7.16b, {v7.16b}, v0.16b
b .L_128_done_\@
.endm
ENTRY(crc_t10dif_pmull_p8)
crc_t10dif_pmull p8
ENDPROC(crc_t10dif_pmull_p8)
.align 5
ENTRY(crc_t10dif_pmull_p64)
crc_t10dif_pmull p64
ENDPROC(crc_t10dif_pmull_p64)
// precomputed constants
// these constants are precomputed from the poly:
// 0x8bb70000 (0x8bb7 scaled to 32 bits)
.section ".rodata", "a"
.align 4
// Q = 0x18BB70000
// rk1 = 2^(32*3) mod Q << 32
// rk2 = 2^(32*5) mod Q << 32
// rk3 = 2^(32*15) mod Q << 32
// rk4 = 2^(32*17) mod Q << 32
// rk5 = 2^(32*3) mod Q << 32
// rk6 = 2^(32*2) mod Q << 32
// rk7 = floor(2^64/Q)
// rk8 = Q
rk1: .octa 0x06df0000000000002d56000000000000
rk3: .octa 0x7cf50000000000009d9d000000000000
rk5: .octa 0x13680000000000002d56000000000000
rk7: .octa 0x000000018bb7000000000001f65a57f8
rk9: .octa 0xbfd6000000000000ceae000000000000
rk11: .octa 0x713c0000000000001e16000000000000
rk13: .octa 0x80a6000000000000f7f9000000000000
rk15: .octa 0xe658000000000000044c000000000000
rk17: .octa 0xa497000000000000ad18000000000000
rk19: .octa 0xe7b50000000000006ee3000000000000
tbl_shf_table:
// use these values for shift constants for the tbl/tbx instruction
// different alignments result in values as shown:
// DDQ 0x008f8e8d8c8b8a898887868584838281 # shl 15 (16-1) / shr1
// DDQ 0x01008f8e8d8c8b8a8988878685848382 # shl 14 (16-3) / shr2
// DDQ 0x0201008f8e8d8c8b8a89888786858483 # shl 13 (16-4) / shr3
// DDQ 0x030201008f8e8d8c8b8a898887868584 # shl 12 (16-4) / shr4
// DDQ 0x04030201008f8e8d8c8b8a8988878685 # shl 11 (16-5) / shr5
// DDQ 0x0504030201008f8e8d8c8b8a89888786 # shl 10 (16-6) / shr6
// DDQ 0x060504030201008f8e8d8c8b8a898887 # shl 9 (16-7) / shr7
// DDQ 0x07060504030201008f8e8d8c8b8a8988 # shl 8 (16-8) / shr8
// DDQ 0x0807060504030201008f8e8d8c8b8a89 # shl 7 (16-9) / shr9
// DDQ 0x090807060504030201008f8e8d8c8b8a # shl 6 (16-10) / shr10
// DDQ 0x0a090807060504030201008f8e8d8c8b # shl 5 (16-11) / shr11
// DDQ 0x0b0a090807060504030201008f8e8d8c # shl 4 (16-12) / shr12
// DDQ 0x0c0b0a090807060504030201008f8e8d # shl 3 (16-13) / shr13
// DDQ 0x0d0c0b0a090807060504030201008f8e # shl 2 (16-14) / shr14
// DDQ 0x0e0d0c0b0a090807060504030201008f # shl 1 (16-15) / shr15
.byte 0x0, 0x81, 0x82, 0x83, 0x84, 0x85, 0x86, 0x87
.byte 0x88, 0x89, 0x8a, 0x8b, 0x8c, 0x8d, 0x8e, 0x8f
.byte 0x0, 0x1, 0x2, 0x3, 0x4, 0x5, 0x6, 0x7
.byte 0x8, 0x9, 0xa, 0xb, 0xc, 0xd, 0xe , 0x0
Reference in New Issue View Git Blame Copy Permalink