mirror of
https://github.com/aaru-dps/AaruBenchmark.git
synced 2025-12-16 19:24:36 +00:00
562 lines
26 KiB
C#
562 lines
26 KiB
C#
// /***************************************************************************
|
|
// Aaru Data Preservation Suite
|
|
// ----------------------------------------------------------------------------
|
|
//
|
|
// Filename : clmul.cs
|
|
// Author(s) : Natalia Portillo <claunia@claunia.com>
|
|
// Wajdi Feghali <wajdi.k.feghali@intel.com>
|
|
// Jim Guilford <james.guilford@intel.com>
|
|
// Vinodh Gopal <vinodh.gopal@intel.com>
|
|
// Erdinc Ozturk <erdinc.ozturk@intel.com>
|
|
// Jim Kukunas <james.t.kukunas@linux.intel.com>
|
|
// Marian Beermann
|
|
//
|
|
// Component : Checksums.
|
|
//
|
|
// --[ Description ] ----------------------------------------------------------
|
|
//
|
|
// Compute the CRC32 using a parallelized folding approach with the PCLMULQDQ
|
|
// instruction.
|
|
//
|
|
// A white paper describing this algorithm can be found at:
|
|
// http://www.intel.com/content/dam/www/public/us/en/documents/white-papers/fast-crc-computation-generic-polynomials-pclmulqdq-paper.pdf
|
|
//
|
|
// --[ License ] --------------------------------------------------------------
|
|
//
|
|
// This software is provided 'as-is', without any express or implied warranty.
|
|
// In no event will the authors be held liable for any damages arising from
|
|
// the use of this software.
|
|
//
|
|
// Permission is granted to anyone to use this software for any purpose,
|
|
// including commercial applications, and to alter it and redistribute it
|
|
// freely, subject to the following restrictions:
|
|
//
|
|
// 1. The origin of this software must not be misrepresented; you must not
|
|
// claim that you wrote the original software. If you use this software
|
|
// in a product, an acknowledgment in the product documentation would be
|
|
// appreciated but is not required.
|
|
//
|
|
// 2. Altered source versions must be plainly marked as such, and must not be
|
|
// misrepresented as being the original software.
|
|
//
|
|
// 3. This notice may not be removed or altered from any source distribution.
|
|
//
|
|
// ----------------------------------------------------------------------------
|
|
// Copyright © 2011-2021 Natalia Portillo
|
|
// Copyright (c) 2016 Marian Beermann (add support for initial value, restructuring)
|
|
// Copyright (C) 2013 Intel Corporation. All rights reserved.
|
|
// ****************************************************************************/
|
|
|
|
using System;
|
|
using System.Runtime.Intrinsics;
|
|
using System.Runtime.Intrinsics.X86;
|
|
|
|
namespace Aaru6.Checksums
|
|
{
|
|
public static class CRC32CLMUL
|
|
{
|
|
static readonly uint[] crc_k =
|
|
{
|
|
0xccaa009e, 0x00000000, /* rk1 */ 0x751997d0, 0x00000001, /* rk2 */ 0xccaa009e, 0x00000000, /* rk5 */
|
|
0x63cd6124, 0x00000001, /* rk6 */ 0xf7011640, 0x00000001, /* rk7 */ 0xdb710640, 0x00000001 /* rk8 */
|
|
};
|
|
|
|
static readonly Vector128<uint>[] pshufb_shf_table =
|
|
{
|
|
Vector128.Create(0x84838281, 0x88878685, 0x8c8b8a89, 0x008f8e8d), /* shl 15 (16 - 1)/shr1 */
|
|
Vector128.Create(0x85848382, 0x89888786, 0x8d8c8b8a, 0x01008f8e), /* shl 14 (16 - 3)/shr2 */
|
|
Vector128.Create(0x86858483, 0x8a898887, 0x8e8d8c8b, 0x0201008f), /* shl 13 (16 - 4)/shr3 */
|
|
Vector128.Create(0x87868584, 0x8b8a8988, 0x8f8e8d8c, 0x03020100), /* shl 12 (16 - 4)/shr4 */
|
|
Vector128.Create(0x88878685, 0x8c8b8a89, 0x008f8e8d, 0x04030201), /* shl 11 (16 - 5)/shr5 */
|
|
Vector128.Create(0x89888786, 0x8d8c8b8a, 0x01008f8e, 0x05040302), /* shl 10 (16 - 6)/shr6 */
|
|
Vector128.Create(0x8a898887, 0x8e8d8c8b, 0x0201008f, 0x06050403), /* shl 9 (16 - 7)/shr7 */
|
|
Vector128.Create(0x8b8a8988, 0x8f8e8d8c, 0x03020100, 0x07060504), /* shl 8 (16 - 8)/shr8 */
|
|
Vector128.Create(0x8c8b8a89, 0x008f8e8d, 0x04030201, 0x08070605), /* shl 7 (16 - 9)/shr9 */
|
|
Vector128.Create(0x8d8c8b8a, 0x01008f8e, 0x05040302, 0x09080706), /* shl 6 (16 -10)/shr10*/
|
|
Vector128.Create(0x8e8d8c8b, 0x0201008f, 0x06050403, 0x0a090807), /* shl 5 (16 -11)/shr11*/
|
|
Vector128.Create(0x8f8e8d8c, 0x03020100, 0x07060504, 0x0b0a0908), /* shl 4 (16 -12)/shr12*/
|
|
Vector128.Create(0x008f8e8du, 0x04030201, 0x08070605, 0x0c0b0a09), /* shl 3 (16 -13)/shr13*/
|
|
Vector128.Create(0x01008f8eu, 0x05040302, 0x09080706, 0x0d0c0b0a), /* shl 2 (16 -14)/shr14*/
|
|
Vector128.Create(0x0201008fu, 0x06050403, 0x0a090807, 0x0e0d0c0b) /* shl 1 (16 -15)/shr15*/
|
|
};
|
|
|
|
static void fold_1(ref Vector128<uint> xmm_crc0, ref Vector128<uint> xmm_crc1, ref Vector128<uint> xmm_crc2,
|
|
ref Vector128<uint> xmm_crc3)
|
|
{
|
|
Vector128<uint> xmm_fold4 = Vector128.Create(0xc6e41596, 0x00000001, 0x54442bd4, 0x00000001);
|
|
|
|
Vector128<uint> x_tmp3;
|
|
Vector128<float> ps_crc0, ps_crc3, ps_res;
|
|
|
|
x_tmp3 = xmm_crc3;
|
|
|
|
xmm_crc3 = xmm_crc0;
|
|
xmm_crc0 = Pclmulqdq.CarrylessMultiply(xmm_crc0.AsUInt64(), xmm_fold4.AsUInt64(), 0x01).AsUInt32();
|
|
xmm_crc3 = Pclmulqdq.CarrylessMultiply(xmm_crc3.AsUInt64(), xmm_fold4.AsUInt64(), 0x10).AsUInt32();
|
|
ps_crc0 = xmm_crc0.AsSingle();
|
|
ps_crc3 = xmm_crc3.AsSingle();
|
|
ps_res = Sse.Xor(ps_crc0, ps_crc3);
|
|
|
|
xmm_crc0 = xmm_crc1;
|
|
xmm_crc1 = xmm_crc2;
|
|
xmm_crc2 = x_tmp3;
|
|
xmm_crc3 = ps_res.AsUInt32();
|
|
}
|
|
|
|
static void fold_2(ref Vector128<uint> xmm_crc0, ref Vector128<uint> xmm_crc1, ref Vector128<uint> xmm_crc2,
|
|
ref Vector128<uint> xmm_crc3)
|
|
{
|
|
Vector128<uint> xmm_fold4 = Vector128.Create(0xc6e41596, 0x00000001, 0x54442bd4, 0x00000001);
|
|
|
|
Vector128<uint> x_tmp3, x_tmp2;
|
|
Vector128<float> ps_crc0, ps_crc1, ps_crc2, ps_crc3, ps_res31, ps_res20;
|
|
|
|
x_tmp3 = xmm_crc3;
|
|
x_tmp2 = xmm_crc2;
|
|
|
|
xmm_crc3 = xmm_crc1;
|
|
xmm_crc1 = Pclmulqdq.CarrylessMultiply(xmm_crc1.AsUInt64(), xmm_fold4.AsUInt64(), 0x01).AsUInt32();
|
|
xmm_crc3 = Pclmulqdq.CarrylessMultiply(xmm_crc3.AsUInt64(), xmm_fold4.AsUInt64(), 0x10).AsUInt32();
|
|
ps_crc3 = xmm_crc3.AsSingle();
|
|
ps_crc1 = xmm_crc1.AsSingle();
|
|
ps_res31 = Sse.Xor(ps_crc3, ps_crc1);
|
|
|
|
xmm_crc2 = xmm_crc0;
|
|
xmm_crc0 = Pclmulqdq.CarrylessMultiply(xmm_crc0.AsUInt64(), xmm_fold4.AsUInt64(), 0x01).AsUInt32();
|
|
xmm_crc2 = Pclmulqdq.CarrylessMultiply(xmm_crc2.AsUInt64(), xmm_fold4.AsUInt64(), 0x10).AsUInt32();
|
|
ps_crc0 = xmm_crc0.AsSingle();
|
|
ps_crc2 = xmm_crc2.AsSingle();
|
|
ps_res20 = Sse.Xor(ps_crc0, ps_crc2);
|
|
|
|
xmm_crc0 = x_tmp2;
|
|
xmm_crc1 = x_tmp3;
|
|
xmm_crc2 = ps_res20.AsUInt32();
|
|
xmm_crc3 = ps_res31.AsUInt32();
|
|
}
|
|
|
|
static void fold_3(ref Vector128<uint> xmm_crc0, ref Vector128<uint> xmm_crc1, ref Vector128<uint> xmm_crc2,
|
|
ref Vector128<uint> xmm_crc3)
|
|
{
|
|
Vector128<uint> xmm_fold4 = Vector128.Create(0x54442bd4, 0x00000001, 0xc6e41596, 0x00000001);
|
|
|
|
Vector128<uint> x_tmp3;
|
|
Vector128<float> ps_crc0, ps_crc1, ps_crc2, ps_crc3, ps_res32, ps_res21, ps_res10;
|
|
|
|
x_tmp3 = xmm_crc3;
|
|
|
|
xmm_crc3 = xmm_crc2;
|
|
xmm_crc2 = Pclmulqdq.CarrylessMultiply(xmm_crc2.AsUInt64(), xmm_fold4.AsUInt64(), 0x01).AsUInt32();
|
|
xmm_crc3 = Pclmulqdq.CarrylessMultiply(xmm_crc3.AsUInt64(), xmm_fold4.AsUInt64(), 0x10).AsUInt32();
|
|
ps_crc2 = xmm_crc2.AsSingle();
|
|
ps_crc3 = xmm_crc3.AsSingle();
|
|
ps_res32 = Sse.Xor(ps_crc2, ps_crc3);
|
|
|
|
xmm_crc2 = xmm_crc1;
|
|
xmm_crc1 = Pclmulqdq.CarrylessMultiply(xmm_crc1.AsUInt64(), xmm_fold4.AsUInt64(), 0x01).AsUInt32();
|
|
xmm_crc2 = Pclmulqdq.CarrylessMultiply(xmm_crc2.AsUInt64(), xmm_fold4.AsUInt64(), 0x10).AsUInt32();
|
|
ps_crc1 = xmm_crc1.AsSingle();
|
|
ps_crc2 = xmm_crc2.AsSingle();
|
|
ps_res21 = Sse.Xor(ps_crc1, ps_crc2);
|
|
|
|
xmm_crc1 = xmm_crc0;
|
|
xmm_crc0 = Pclmulqdq.CarrylessMultiply(xmm_crc0.AsUInt64(), xmm_fold4.AsUInt64(), 0x01).AsUInt32();
|
|
xmm_crc1 = Pclmulqdq.CarrylessMultiply(xmm_crc1.AsUInt64(), xmm_fold4.AsUInt64(), 0x10).AsUInt32();
|
|
ps_crc0 = xmm_crc0.AsSingle();
|
|
ps_crc1 = xmm_crc1.AsSingle();
|
|
ps_res10 = Sse.Xor(ps_crc0, ps_crc1);
|
|
|
|
xmm_crc0 = x_tmp3;
|
|
xmm_crc1 = ps_res10.AsUInt32();
|
|
xmm_crc2 = ps_res21.AsUInt32();
|
|
xmm_crc3 = ps_res32.AsUInt32();
|
|
}
|
|
|
|
static void fold_4(ref Vector128<uint> xmm_crc0, ref Vector128<uint> xmm_crc1, ref Vector128<uint> xmm_crc2,
|
|
ref Vector128<uint> xmm_crc3)
|
|
{
|
|
Vector128<uint> xmm_fold4 = Vector128.Create(0xc6e41596, 0x00000001, 0x54442bd4, 0x00000001);
|
|
|
|
Vector128<uint> x_tmp0, x_tmp1, x_tmp2, x_tmp3;
|
|
Vector128<float> ps_crc0, ps_crc1, ps_crc2, ps_crc3;
|
|
Vector128<float> ps_t0, ps_t1, ps_t2, ps_t3;
|
|
Vector128<float> ps_res0, ps_res1, ps_res2, ps_res3;
|
|
|
|
x_tmp0 = xmm_crc0;
|
|
x_tmp1 = xmm_crc1;
|
|
x_tmp2 = xmm_crc2;
|
|
x_tmp3 = xmm_crc3;
|
|
|
|
xmm_crc0 = Pclmulqdq.CarrylessMultiply(xmm_crc0.AsUInt64(), xmm_fold4.AsUInt64(), 0x01).AsUInt32();
|
|
x_tmp0 = Pclmulqdq.CarrylessMultiply(x_tmp0.AsUInt64(), xmm_fold4.AsUInt64(), 0x10).AsUInt32();
|
|
ps_crc0 = xmm_crc0.AsSingle();
|
|
ps_t0 = x_tmp0.AsSingle();
|
|
ps_res0 = Sse.Xor(ps_crc0, ps_t0);
|
|
|
|
xmm_crc1 = Pclmulqdq.CarrylessMultiply(xmm_crc1.AsUInt64(), xmm_fold4.AsUInt64(), 0x01).AsUInt32();
|
|
x_tmp1 = Pclmulqdq.CarrylessMultiply(x_tmp1.AsUInt64(), xmm_fold4.AsUInt64(), 0x10).AsUInt32();
|
|
ps_crc1 = xmm_crc1.AsSingle();
|
|
ps_t1 = x_tmp1.AsSingle();
|
|
ps_res1 = Sse.Xor(ps_crc1, ps_t1);
|
|
|
|
xmm_crc2 = Pclmulqdq.CarrylessMultiply(xmm_crc2.AsUInt64(), xmm_fold4.AsUInt64(), 0x01).AsUInt32();
|
|
x_tmp2 = Pclmulqdq.CarrylessMultiply(x_tmp2.AsUInt64(), xmm_fold4.AsUInt64(), 0x10).AsUInt32();
|
|
ps_crc2 = xmm_crc2.AsSingle();
|
|
ps_t2 = x_tmp2.AsSingle();
|
|
ps_res2 = Sse.Xor(ps_crc2, ps_t2);
|
|
|
|
xmm_crc3 = Pclmulqdq.CarrylessMultiply(xmm_crc3.AsUInt64(), xmm_fold4.AsUInt64(), 0x01).AsUInt32();
|
|
x_tmp3 = Pclmulqdq.CarrylessMultiply(x_tmp3.AsUInt64(), xmm_fold4.AsUInt64(), 0x10).AsUInt32();
|
|
ps_crc3 = xmm_crc3.AsSingle();
|
|
ps_t3 = x_tmp3.AsSingle();
|
|
ps_res3 = Sse.Xor(ps_crc3, ps_t3);
|
|
|
|
xmm_crc0 = ps_res0.AsUInt32();
|
|
xmm_crc1 = ps_res1.AsUInt32();
|
|
xmm_crc2 = ps_res2.AsUInt32();
|
|
xmm_crc3 = ps_res3.AsUInt32();
|
|
}
|
|
|
|
static void partial_fold(long len, ref Vector128<uint> xmm_crc0, ref Vector128<uint> xmm_crc1,
|
|
ref Vector128<uint> xmm_crc2, ref Vector128<uint> xmm_crc3,
|
|
ref Vector128<uint> xmm_crc_part)
|
|
{
|
|
Vector128<uint> xmm_fold4 = Vector128.Create(0x54442bd4, 0x00000001, 0xc6e41596, 0x00000001);
|
|
Vector128<uint> xmm_mask3 = Vector128.Create(0x80808080);
|
|
|
|
Vector128<uint> xmm_shl, xmm_shr, xmm_tmp1, xmm_tmp2, xmm_tmp3;
|
|
Vector128<uint> xmm_a0_0, xmm_a0_1;
|
|
Vector128<float> ps_crc3, psa0_0, psa0_1, ps_res;
|
|
|
|
xmm_shl = pshufb_shf_table[len - 1];
|
|
xmm_shr = xmm_shl;
|
|
xmm_shr = Sse2.Xor(xmm_shr, xmm_mask3);
|
|
|
|
xmm_a0_0 = Ssse3.Shuffle(xmm_crc0.AsByte(), xmm_shl.AsByte()).AsUInt32();
|
|
|
|
xmm_crc0 = Ssse3.Shuffle(xmm_crc0.AsByte(), xmm_shr.AsByte()).AsUInt32();
|
|
xmm_tmp1 = Ssse3.Shuffle(xmm_crc1.AsByte(), xmm_shl.AsByte()).AsUInt32();
|
|
xmm_crc0 = Sse2.Or(xmm_crc0, xmm_tmp1);
|
|
|
|
xmm_crc1 = Ssse3.Shuffle(xmm_crc1.AsByte(), xmm_shr.AsByte()).AsUInt32();
|
|
xmm_tmp2 = Ssse3.Shuffle(xmm_crc2.AsByte(), xmm_shl.AsByte()).AsUInt32();
|
|
xmm_crc1 = Sse2.Or(xmm_crc1, xmm_tmp2);
|
|
|
|
xmm_crc2 = Ssse3.Shuffle(xmm_crc2.AsByte(), xmm_shr.AsByte()).AsUInt32();
|
|
xmm_tmp3 = Ssse3.Shuffle(xmm_crc3.AsByte(), xmm_shl.AsByte()).AsUInt32();
|
|
xmm_crc2 = Sse2.Or(xmm_crc2, xmm_tmp3);
|
|
|
|
xmm_crc3 = Ssse3.Shuffle(xmm_crc3.AsByte(), xmm_shr.AsByte()).AsUInt32();
|
|
xmm_crc_part = Ssse3.Shuffle(xmm_crc_part.AsByte(), xmm_shl.AsByte()).AsUInt32();
|
|
xmm_crc3 = Sse2.Or(xmm_crc3, xmm_crc_part);
|
|
|
|
xmm_a0_1 = Pclmulqdq.CarrylessMultiply(xmm_a0_0.AsUInt64(), xmm_fold4.AsUInt64(), 0x10).AsUInt32();
|
|
xmm_a0_0 = Pclmulqdq.CarrylessMultiply(xmm_a0_0.AsUInt64(), xmm_fold4.AsUInt64(), 0x01).AsUInt32();
|
|
|
|
ps_crc3 = xmm_crc3.AsSingle();
|
|
psa0_0 = xmm_a0_0.AsSingle();
|
|
psa0_1 = xmm_a0_1.AsSingle();
|
|
|
|
ps_res = Sse.Xor(ps_crc3, psa0_0);
|
|
ps_res = Sse.Xor(ps_res, psa0_1);
|
|
|
|
xmm_crc3 = ps_res.AsUInt32();
|
|
}
|
|
|
|
internal static uint crc32_clmul(byte[] src, long len, uint initial_crc)
|
|
{
|
|
Vector128<uint> xmm_t0, xmm_t1, xmm_t2, xmm_t3;
|
|
Vector128<uint> xmm_initial = Sse2.ConvertScalarToVector128UInt32(initial_crc);
|
|
Vector128<uint> xmm_crc0 = Sse2.ConvertScalarToVector128UInt32(0x9db42487);
|
|
Vector128<uint> xmm_crc1 = Vector128<uint>.Zero;
|
|
Vector128<uint> xmm_crc2 = Vector128<uint>.Zero;
|
|
Vector128<uint> xmm_crc3 = Vector128<uint>.Zero;
|
|
Vector128<uint> xmm_crc_part;
|
|
int bufPos = 0;
|
|
|
|
bool first = true;
|
|
|
|
/* fold 512 to 32 step variable declarations for ISO-C90 compat. */
|
|
Vector128<uint> xmm_mask = Vector128.Create(0xFFFFFFFF, 0xFFFFFFFF, 0x00000000, 0x00000000);
|
|
Vector128<uint> xmm_mask2 = Vector128.Create(0x00000000, 0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF);
|
|
|
|
uint crc;
|
|
Vector128<uint> x_tmp0, x_tmp1, x_tmp2, crc_fold;
|
|
|
|
if(len < 16)
|
|
{
|
|
if(len == 0)
|
|
return initial_crc;
|
|
|
|
if(len < 4)
|
|
{
|
|
/*
|
|
* no idea how to do this for <4 bytes, delegate to classic impl.
|
|
*/
|
|
crc = ~initial_crc;
|
|
|
|
switch(len)
|
|
{
|
|
case 3:
|
|
crc = (crc >> 8) ^ Crc32Context._isoCrc32Table[0][(crc & 0xFF) ^ src[bufPos++]];
|
|
goto case 2;
|
|
case 2:
|
|
crc = (crc >> 8) ^ Crc32Context._isoCrc32Table[0][(crc & 0xFF) ^ src[bufPos++]];
|
|
goto case 1;
|
|
case 1:
|
|
crc = (crc >> 8) ^ Crc32Context._isoCrc32Table[0][(crc & 0xFF) ^ src[bufPos++]];
|
|
|
|
break;
|
|
}
|
|
|
|
return ~crc;
|
|
}
|
|
|
|
xmm_crc_part = Vector128.Create(BitConverter.ToUInt32(src, 0), BitConverter.ToUInt32(src, 4),
|
|
BitConverter.ToUInt32(src, 8), BitConverter.ToUInt32(src, 12));
|
|
|
|
if(first)
|
|
{
|
|
first = false;
|
|
xmm_crc_part = Sse2.Xor(xmm_crc_part, xmm_initial);
|
|
}
|
|
|
|
goto partial;
|
|
}
|
|
|
|
while((len -= 64) >= 0)
|
|
{
|
|
xmm_t0 = Vector128.Create(BitConverter.ToUInt32(src, bufPos), BitConverter.ToUInt32(src, bufPos + 4),
|
|
BitConverter.ToUInt32(src, bufPos + 8),
|
|
BitConverter.ToUInt32(src, bufPos + 12));
|
|
|
|
bufPos += 16;
|
|
|
|
xmm_t1 = Vector128.Create(BitConverter.ToUInt32(src, bufPos), BitConverter.ToUInt32(src, bufPos + 4),
|
|
BitConverter.ToUInt32(src, bufPos + 8),
|
|
BitConverter.ToUInt32(src, bufPos + 12));
|
|
|
|
bufPos += 16;
|
|
|
|
xmm_t2 = Vector128.Create(BitConverter.ToUInt32(src, bufPos), BitConverter.ToUInt32(src, bufPos + 4),
|
|
BitConverter.ToUInt32(src, bufPos + 8),
|
|
BitConverter.ToUInt32(src, bufPos + 12));
|
|
|
|
bufPos += 16;
|
|
|
|
xmm_t3 = Vector128.Create(BitConverter.ToUInt32(src, bufPos), BitConverter.ToUInt32(src, bufPos + 4),
|
|
BitConverter.ToUInt32(src, bufPos + 8),
|
|
BitConverter.ToUInt32(src, bufPos + 12));
|
|
|
|
bufPos += 16;
|
|
|
|
if(first)
|
|
{
|
|
first = false;
|
|
xmm_t0 = Sse2.Xor(xmm_t0, xmm_initial);
|
|
}
|
|
|
|
fold_4(ref xmm_crc0, ref xmm_crc1, ref xmm_crc2, ref xmm_crc3);
|
|
|
|
xmm_crc0 = Sse2.Xor(xmm_crc0, xmm_t0);
|
|
xmm_crc1 = Sse2.Xor(xmm_crc1, xmm_t1);
|
|
xmm_crc2 = Sse2.Xor(xmm_crc2, xmm_t2);
|
|
xmm_crc3 = Sse2.Xor(xmm_crc3, xmm_t3);
|
|
}
|
|
|
|
/*
|
|
* len = num bytes left - 64
|
|
*/
|
|
if(len + 16 >= 0)
|
|
{
|
|
len += 16;
|
|
|
|
xmm_t0 = Vector128.Create(BitConverter.ToUInt32(src, bufPos), BitConverter.ToUInt32(src, bufPos + 4),
|
|
BitConverter.ToUInt32(src, bufPos + 8),
|
|
BitConverter.ToUInt32(src, bufPos + 12));
|
|
|
|
bufPos += 16;
|
|
|
|
xmm_t1 = Vector128.Create(BitConverter.ToUInt32(src, bufPos), BitConverter.ToUInt32(src, bufPos + 4),
|
|
BitConverter.ToUInt32(src, bufPos + 8),
|
|
BitConverter.ToUInt32(src, bufPos + 12));
|
|
|
|
bufPos += 16;
|
|
|
|
xmm_t2 = Vector128.Create(BitConverter.ToUInt32(src, bufPos), BitConverter.ToUInt32(src, bufPos + 4),
|
|
BitConverter.ToUInt32(src, bufPos + 8),
|
|
BitConverter.ToUInt32(src, bufPos + 12));
|
|
|
|
bufPos += 16;
|
|
|
|
if(first)
|
|
{
|
|
first = false;
|
|
xmm_t0 = Sse2.Xor(xmm_t0, xmm_initial);
|
|
}
|
|
|
|
fold_3(ref xmm_crc0, ref xmm_crc1, ref xmm_crc2, ref xmm_crc3);
|
|
|
|
xmm_crc1 = Sse2.Xor(xmm_crc1, xmm_t0);
|
|
xmm_crc2 = Sse2.Xor(xmm_crc2, xmm_t1);
|
|
xmm_crc3 = Sse2.Xor(xmm_crc3, xmm_t2);
|
|
|
|
if(len == 0)
|
|
goto done;
|
|
|
|
xmm_crc_part = Vector128.Create(BitConverter.ToUInt32(src, bufPos),
|
|
BitConverter.ToUInt32(src, bufPos + 4),
|
|
BitConverter.ToUInt32(src, bufPos + 8),
|
|
BitConverter.ToUInt32(src, bufPos + 12));
|
|
}
|
|
else if(len + 32 >= 0)
|
|
{
|
|
len += 32;
|
|
|
|
xmm_t0 = Vector128.Create(BitConverter.ToUInt32(src, bufPos), BitConverter.ToUInt32(src, bufPos + 4),
|
|
BitConverter.ToUInt32(src, bufPos + 8),
|
|
BitConverter.ToUInt32(src, bufPos + 12));
|
|
|
|
bufPos += 16;
|
|
|
|
xmm_t1 = Vector128.Create(BitConverter.ToUInt32(src, bufPos), BitConverter.ToUInt32(src, bufPos + 4),
|
|
BitConverter.ToUInt32(src, bufPos + 8),
|
|
BitConverter.ToUInt32(src, bufPos + 12));
|
|
|
|
bufPos += 16;
|
|
|
|
if(first)
|
|
{
|
|
first = false;
|
|
xmm_t0 = Sse2.Xor(xmm_t0, xmm_initial);
|
|
}
|
|
|
|
fold_2(ref xmm_crc0, ref xmm_crc1, ref xmm_crc2, ref xmm_crc3);
|
|
|
|
xmm_crc2 = Sse2.Xor(xmm_crc2, xmm_t0);
|
|
xmm_crc3 = Sse2.Xor(xmm_crc3, xmm_t1);
|
|
|
|
if(len == 0)
|
|
goto done;
|
|
|
|
xmm_crc_part = Vector128.Create(BitConverter.ToUInt32(src, bufPos),
|
|
BitConverter.ToUInt32(src, bufPos + 4),
|
|
BitConverter.ToUInt32(src, bufPos + 8),
|
|
BitConverter.ToUInt32(src, bufPos + 12));
|
|
}
|
|
else if(len + 48 >= 0)
|
|
{
|
|
len += 48;
|
|
|
|
xmm_t0 = Vector128.Create(BitConverter.ToUInt32(src, bufPos), BitConverter.ToUInt32(src, bufPos + 4),
|
|
BitConverter.ToUInt32(src, bufPos + 8),
|
|
BitConverter.ToUInt32(src, bufPos + 12));
|
|
|
|
bufPos += 16;
|
|
|
|
if(first)
|
|
{
|
|
first = false;
|
|
xmm_t0 = Sse2.Xor(xmm_t0, xmm_initial);
|
|
}
|
|
|
|
fold_1(ref xmm_crc0, ref xmm_crc1, ref xmm_crc2, ref xmm_crc3);
|
|
|
|
xmm_crc3 = Sse2.Xor(xmm_crc3, xmm_t0);
|
|
|
|
if(len == 0)
|
|
goto done;
|
|
|
|
xmm_crc_part = Vector128.Create(BitConverter.ToUInt32(src, bufPos),
|
|
BitConverter.ToUInt32(src, bufPos + 4),
|
|
BitConverter.ToUInt32(src, bufPos + 8),
|
|
BitConverter.ToUInt32(src, bufPos + 12));
|
|
}
|
|
else
|
|
{
|
|
len += 64;
|
|
|
|
if(len == 0)
|
|
goto done;
|
|
|
|
xmm_crc_part = Vector128.Create(BitConverter.ToUInt32(src, bufPos),
|
|
BitConverter.ToUInt32(src, bufPos + 4),
|
|
BitConverter.ToUInt32(src, bufPos + 8),
|
|
BitConverter.ToUInt32(src, bufPos + 12));
|
|
|
|
if(first)
|
|
{
|
|
first = false;
|
|
xmm_crc_part = Sse2.Xor(xmm_crc_part, xmm_initial);
|
|
}
|
|
}
|
|
|
|
partial:
|
|
partial_fold(len, ref xmm_crc0, ref xmm_crc1, ref xmm_crc2, ref xmm_crc3, ref xmm_crc_part);
|
|
|
|
done:
|
|
|
|
/* fold 512 to 32 */
|
|
|
|
/*
|
|
* k1
|
|
*/
|
|
crc_fold = Vector128.Create(crc_k[0], crc_k[1], crc_k[2], crc_k[3]);
|
|
|
|
x_tmp0 = Pclmulqdq.CarrylessMultiply(xmm_crc0.AsUInt64(), crc_fold.AsUInt64(), 0x10).AsUInt32();
|
|
xmm_crc0 = Pclmulqdq.CarrylessMultiply(xmm_crc0.AsUInt64(), crc_fold.AsUInt64(), 0x01).AsUInt32();
|
|
xmm_crc1 = Sse2.Xor(xmm_crc1, x_tmp0);
|
|
xmm_crc1 = Sse2.Xor(xmm_crc1, xmm_crc0);
|
|
|
|
x_tmp1 = Pclmulqdq.CarrylessMultiply(xmm_crc1.AsUInt64(), crc_fold.AsUInt64(), 0x10).AsUInt32();
|
|
xmm_crc1 = Pclmulqdq.CarrylessMultiply(xmm_crc1.AsUInt64(), crc_fold.AsUInt64(), 0x01).AsUInt32();
|
|
xmm_crc2 = Sse2.Xor(xmm_crc2, x_tmp1);
|
|
xmm_crc2 = Sse2.Xor(xmm_crc2, xmm_crc1);
|
|
|
|
x_tmp2 = Pclmulqdq.CarrylessMultiply(xmm_crc2.AsUInt64(), crc_fold.AsUInt64(), 0x10).AsUInt32();
|
|
xmm_crc2 = Pclmulqdq.CarrylessMultiply(xmm_crc2.AsUInt64(), crc_fold.AsUInt64(), 0x01).AsUInt32();
|
|
xmm_crc3 = Sse2.Xor(xmm_crc3, x_tmp2);
|
|
xmm_crc3 = Sse2.Xor(xmm_crc3, xmm_crc2);
|
|
|
|
/*
|
|
* k5
|
|
*/
|
|
crc_fold = Vector128.Create(crc_k[4], crc_k[5], crc_k[6], crc_k[7]);
|
|
|
|
xmm_crc0 = xmm_crc3;
|
|
xmm_crc3 = Pclmulqdq.CarrylessMultiply(xmm_crc3.AsUInt64(), crc_fold.AsUInt64(), 0).AsUInt32();
|
|
xmm_crc0 = Sse2.ShiftRightLogical128BitLane(xmm_crc0, 8);
|
|
xmm_crc3 = Sse2.Xor(xmm_crc3, xmm_crc0);
|
|
|
|
xmm_crc0 = xmm_crc3;
|
|
xmm_crc3 = Sse2.ShiftLeftLogical128BitLane(xmm_crc3, 4);
|
|
xmm_crc3 = Pclmulqdq.CarrylessMultiply(xmm_crc3.AsUInt64(), crc_fold.AsUInt64(), 0x10).AsUInt32();
|
|
xmm_crc3 = Sse2.Xor(xmm_crc3, xmm_crc0);
|
|
xmm_crc3 = Sse2.And(xmm_crc3, xmm_mask2);
|
|
|
|
/*
|
|
* k7
|
|
*/
|
|
xmm_crc1 = xmm_crc3;
|
|
xmm_crc2 = xmm_crc3;
|
|
crc_fold = Vector128.Create(crc_k[8], crc_k[9], crc_k[10], crc_k[11]);
|
|
|
|
xmm_crc3 = Pclmulqdq.CarrylessMultiply(xmm_crc3.AsUInt64(), crc_fold.AsUInt64(), 0).AsUInt32();
|
|
xmm_crc3 = Sse2.Xor(xmm_crc3, xmm_crc2);
|
|
xmm_crc3 = Sse2.And(xmm_crc3, xmm_mask);
|
|
|
|
xmm_crc2 = xmm_crc3;
|
|
xmm_crc3 = Pclmulqdq.CarrylessMultiply(xmm_crc3.AsUInt64(), crc_fold.AsUInt64(), 0x10).AsUInt32();
|
|
xmm_crc3 = Sse2.Xor(xmm_crc3, xmm_crc2);
|
|
xmm_crc3 = Sse2.Xor(xmm_crc3, xmm_crc1);
|
|
|
|
/*
|
|
* could just as well write xmm_crc3[2], doing a movaps and truncating, but
|
|
* no real advantage - it's a tiny bit slower per call, while no additional CPUs
|
|
* would be supported by only requiring SSSE3 and CLMUL instead of SSE4.1 + CLMUL
|
|
*/
|
|
crc = Sse41.Extract(xmm_crc3, 2);
|
|
|
|
return ~crc;
|
|
}
|
|
}
|
|
} |