adding git hook test, simplified structure using vendoring
This commit is contained in:
27
vendor/golang.org/x/crypto/LICENSE
generated
vendored
Normal file
27
vendor/golang.org/x/crypto/LICENSE
generated
vendored
Normal file
@@ -0,0 +1,27 @@
|
||||
Copyright (c) 2009 The Go Authors. All rights reserved.
|
||||
|
||||
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 Google Inc. 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 THE COPYRIGHT HOLDERS AND CONTRIBUTORS
|
||||
"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 THE COPYRIGHT
|
||||
OWNER 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.
|
||||
22
vendor/golang.org/x/crypto/PATENTS
generated
vendored
Normal file
22
vendor/golang.org/x/crypto/PATENTS
generated
vendored
Normal file
@@ -0,0 +1,22 @@
|
||||
Additional IP Rights Grant (Patents)
|
||||
|
||||
"This implementation" means the copyrightable works distributed by
|
||||
Google as part of the Go project.
|
||||
|
||||
Google hereby grants to You a perpetual, worldwide, non-exclusive,
|
||||
no-charge, royalty-free, irrevocable (except as stated in this section)
|
||||
patent license to make, have made, use, offer to sell, sell, import,
|
||||
transfer and otherwise run, modify and propagate the contents of this
|
||||
implementation of Go, where such license applies only to those patent
|
||||
claims, both currently owned or controlled by Google and acquired in
|
||||
the future, licensable by Google that are necessarily infringed by this
|
||||
implementation of Go. This grant does not include claims that would be
|
||||
infringed only as a consequence of further modification of this
|
||||
implementation. If you or your agent or exclusive licensee institute or
|
||||
order or agree to the institution of patent litigation against any
|
||||
entity (including a cross-claim or counterclaim in a lawsuit) alleging
|
||||
that this implementation of Go or any code incorporated within this
|
||||
implementation of Go constitutes direct or contributory patent
|
||||
infringement, or inducement of patent infringement, then any patent
|
||||
rights granted to you under this License for this implementation of Go
|
||||
shall terminate as of the date such litigation is filed.
|
||||
17
vendor/golang.org/x/crypto/chacha20/chacha_arm64.go
generated
vendored
Normal file
17
vendor/golang.org/x/crypto/chacha20/chacha_arm64.go
generated
vendored
Normal file
@@ -0,0 +1,17 @@
|
||||
// Copyright 2018 The Go Authors. All rights reserved.
|
||||
// Use of this source code is governed by a BSD-style
|
||||
// license that can be found in the LICENSE file.
|
||||
|
||||
//go:build go1.11 && gc && !purego
|
||||
// +build go1.11,gc,!purego
|
||||
|
||||
package chacha20
|
||||
|
||||
const bufSize = 256
|
||||
|
||||
//go:noescape
|
||||
func xorKeyStreamVX(dst, src []byte, key *[8]uint32, nonce *[3]uint32, counter *uint32)
|
||||
|
||||
func (c *Cipher) xorKeyStreamBlocks(dst, src []byte) {
|
||||
xorKeyStreamVX(dst, src, &c.key, &c.nonce, &c.counter)
|
||||
}
|
||||
308
vendor/golang.org/x/crypto/chacha20/chacha_arm64.s
generated
vendored
Normal file
308
vendor/golang.org/x/crypto/chacha20/chacha_arm64.s
generated
vendored
Normal file
@@ -0,0 +1,308 @@
|
||||
// Copyright 2018 The Go Authors. All rights reserved.
|
||||
// Use of this source code is governed by a BSD-style
|
||||
// license that can be found in the LICENSE file.
|
||||
|
||||
//go:build go1.11 && gc && !purego
|
||||
// +build go1.11,gc,!purego
|
||||
|
||||
#include "textflag.h"
|
||||
|
||||
#define NUM_ROUNDS 10
|
||||
|
||||
// func xorKeyStreamVX(dst, src []byte, key *[8]uint32, nonce *[3]uint32, counter *uint32)
|
||||
TEXT ·xorKeyStreamVX(SB), NOSPLIT, $0
|
||||
MOVD dst+0(FP), R1
|
||||
MOVD src+24(FP), R2
|
||||
MOVD src_len+32(FP), R3
|
||||
MOVD key+48(FP), R4
|
||||
MOVD nonce+56(FP), R6
|
||||
MOVD counter+64(FP), R7
|
||||
|
||||
MOVD $·constants(SB), R10
|
||||
MOVD $·incRotMatrix(SB), R11
|
||||
|
||||
MOVW (R7), R20
|
||||
|
||||
AND $~255, R3, R13
|
||||
ADD R2, R13, R12 // R12 for block end
|
||||
AND $255, R3, R13
|
||||
loop:
|
||||
MOVD $NUM_ROUNDS, R21
|
||||
VLD1 (R11), [V30.S4, V31.S4]
|
||||
|
||||
// load contants
|
||||
// VLD4R (R10), [V0.S4, V1.S4, V2.S4, V3.S4]
|
||||
WORD $0x4D60E940
|
||||
|
||||
// load keys
|
||||
// VLD4R 16(R4), [V4.S4, V5.S4, V6.S4, V7.S4]
|
||||
WORD $0x4DFFE884
|
||||
// VLD4R 16(R4), [V8.S4, V9.S4, V10.S4, V11.S4]
|
||||
WORD $0x4DFFE888
|
||||
SUB $32, R4
|
||||
|
||||
// load counter + nonce
|
||||
// VLD1R (R7), [V12.S4]
|
||||
WORD $0x4D40C8EC
|
||||
|
||||
// VLD3R (R6), [V13.S4, V14.S4, V15.S4]
|
||||
WORD $0x4D40E8CD
|
||||
|
||||
// update counter
|
||||
VADD V30.S4, V12.S4, V12.S4
|
||||
|
||||
chacha:
|
||||
// V0..V3 += V4..V7
|
||||
// V12..V15 <<<= ((V12..V15 XOR V0..V3), 16)
|
||||
VADD V0.S4, V4.S4, V0.S4
|
||||
VADD V1.S4, V5.S4, V1.S4
|
||||
VADD V2.S4, V6.S4, V2.S4
|
||||
VADD V3.S4, V7.S4, V3.S4
|
||||
VEOR V12.B16, V0.B16, V12.B16
|
||||
VEOR V13.B16, V1.B16, V13.B16
|
||||
VEOR V14.B16, V2.B16, V14.B16
|
||||
VEOR V15.B16, V3.B16, V15.B16
|
||||
VREV32 V12.H8, V12.H8
|
||||
VREV32 V13.H8, V13.H8
|
||||
VREV32 V14.H8, V14.H8
|
||||
VREV32 V15.H8, V15.H8
|
||||
// V8..V11 += V12..V15
|
||||
// V4..V7 <<<= ((V4..V7 XOR V8..V11), 12)
|
||||
VADD V8.S4, V12.S4, V8.S4
|
||||
VADD V9.S4, V13.S4, V9.S4
|
||||
VADD V10.S4, V14.S4, V10.S4
|
||||
VADD V11.S4, V15.S4, V11.S4
|
||||
VEOR V8.B16, V4.B16, V16.B16
|
||||
VEOR V9.B16, V5.B16, V17.B16
|
||||
VEOR V10.B16, V6.B16, V18.B16
|
||||
VEOR V11.B16, V7.B16, V19.B16
|
||||
VSHL $12, V16.S4, V4.S4
|
||||
VSHL $12, V17.S4, V5.S4
|
||||
VSHL $12, V18.S4, V6.S4
|
||||
VSHL $12, V19.S4, V7.S4
|
||||
VSRI $20, V16.S4, V4.S4
|
||||
VSRI $20, V17.S4, V5.S4
|
||||
VSRI $20, V18.S4, V6.S4
|
||||
VSRI $20, V19.S4, V7.S4
|
||||
|
||||
// V0..V3 += V4..V7
|
||||
// V12..V15 <<<= ((V12..V15 XOR V0..V3), 8)
|
||||
VADD V0.S4, V4.S4, V0.S4
|
||||
VADD V1.S4, V5.S4, V1.S4
|
||||
VADD V2.S4, V6.S4, V2.S4
|
||||
VADD V3.S4, V7.S4, V3.S4
|
||||
VEOR V12.B16, V0.B16, V12.B16
|
||||
VEOR V13.B16, V1.B16, V13.B16
|
||||
VEOR V14.B16, V2.B16, V14.B16
|
||||
VEOR V15.B16, V3.B16, V15.B16
|
||||
VTBL V31.B16, [V12.B16], V12.B16
|
||||
VTBL V31.B16, [V13.B16], V13.B16
|
||||
VTBL V31.B16, [V14.B16], V14.B16
|
||||
VTBL V31.B16, [V15.B16], V15.B16
|
||||
|
||||
// V8..V11 += V12..V15
|
||||
// V4..V7 <<<= ((V4..V7 XOR V8..V11), 7)
|
||||
VADD V12.S4, V8.S4, V8.S4
|
||||
VADD V13.S4, V9.S4, V9.S4
|
||||
VADD V14.S4, V10.S4, V10.S4
|
||||
VADD V15.S4, V11.S4, V11.S4
|
||||
VEOR V8.B16, V4.B16, V16.B16
|
||||
VEOR V9.B16, V5.B16, V17.B16
|
||||
VEOR V10.B16, V6.B16, V18.B16
|
||||
VEOR V11.B16, V7.B16, V19.B16
|
||||
VSHL $7, V16.S4, V4.S4
|
||||
VSHL $7, V17.S4, V5.S4
|
||||
VSHL $7, V18.S4, V6.S4
|
||||
VSHL $7, V19.S4, V7.S4
|
||||
VSRI $25, V16.S4, V4.S4
|
||||
VSRI $25, V17.S4, V5.S4
|
||||
VSRI $25, V18.S4, V6.S4
|
||||
VSRI $25, V19.S4, V7.S4
|
||||
|
||||
// V0..V3 += V5..V7, V4
|
||||
// V15,V12-V14 <<<= ((V15,V12-V14 XOR V0..V3), 16)
|
||||
VADD V0.S4, V5.S4, V0.S4
|
||||
VADD V1.S4, V6.S4, V1.S4
|
||||
VADD V2.S4, V7.S4, V2.S4
|
||||
VADD V3.S4, V4.S4, V3.S4
|
||||
VEOR V15.B16, V0.B16, V15.B16
|
||||
VEOR V12.B16, V1.B16, V12.B16
|
||||
VEOR V13.B16, V2.B16, V13.B16
|
||||
VEOR V14.B16, V3.B16, V14.B16
|
||||
VREV32 V12.H8, V12.H8
|
||||
VREV32 V13.H8, V13.H8
|
||||
VREV32 V14.H8, V14.H8
|
||||
VREV32 V15.H8, V15.H8
|
||||
|
||||
// V10 += V15; V5 <<<= ((V10 XOR V5), 12)
|
||||
// ...
|
||||
VADD V15.S4, V10.S4, V10.S4
|
||||
VADD V12.S4, V11.S4, V11.S4
|
||||
VADD V13.S4, V8.S4, V8.S4
|
||||
VADD V14.S4, V9.S4, V9.S4
|
||||
VEOR V10.B16, V5.B16, V16.B16
|
||||
VEOR V11.B16, V6.B16, V17.B16
|
||||
VEOR V8.B16, V7.B16, V18.B16
|
||||
VEOR V9.B16, V4.B16, V19.B16
|
||||
VSHL $12, V16.S4, V5.S4
|
||||
VSHL $12, V17.S4, V6.S4
|
||||
VSHL $12, V18.S4, V7.S4
|
||||
VSHL $12, V19.S4, V4.S4
|
||||
VSRI $20, V16.S4, V5.S4
|
||||
VSRI $20, V17.S4, V6.S4
|
||||
VSRI $20, V18.S4, V7.S4
|
||||
VSRI $20, V19.S4, V4.S4
|
||||
|
||||
// V0 += V5; V15 <<<= ((V0 XOR V15), 8)
|
||||
// ...
|
||||
VADD V5.S4, V0.S4, V0.S4
|
||||
VADD V6.S4, V1.S4, V1.S4
|
||||
VADD V7.S4, V2.S4, V2.S4
|
||||
VADD V4.S4, V3.S4, V3.S4
|
||||
VEOR V0.B16, V15.B16, V15.B16
|
||||
VEOR V1.B16, V12.B16, V12.B16
|
||||
VEOR V2.B16, V13.B16, V13.B16
|
||||
VEOR V3.B16, V14.B16, V14.B16
|
||||
VTBL V31.B16, [V12.B16], V12.B16
|
||||
VTBL V31.B16, [V13.B16], V13.B16
|
||||
VTBL V31.B16, [V14.B16], V14.B16
|
||||
VTBL V31.B16, [V15.B16], V15.B16
|
||||
|
||||
// V10 += V15; V5 <<<= ((V10 XOR V5), 7)
|
||||
// ...
|
||||
VADD V15.S4, V10.S4, V10.S4
|
||||
VADD V12.S4, V11.S4, V11.S4
|
||||
VADD V13.S4, V8.S4, V8.S4
|
||||
VADD V14.S4, V9.S4, V9.S4
|
||||
VEOR V10.B16, V5.B16, V16.B16
|
||||
VEOR V11.B16, V6.B16, V17.B16
|
||||
VEOR V8.B16, V7.B16, V18.B16
|
||||
VEOR V9.B16, V4.B16, V19.B16
|
||||
VSHL $7, V16.S4, V5.S4
|
||||
VSHL $7, V17.S4, V6.S4
|
||||
VSHL $7, V18.S4, V7.S4
|
||||
VSHL $7, V19.S4, V4.S4
|
||||
VSRI $25, V16.S4, V5.S4
|
||||
VSRI $25, V17.S4, V6.S4
|
||||
VSRI $25, V18.S4, V7.S4
|
||||
VSRI $25, V19.S4, V4.S4
|
||||
|
||||
SUB $1, R21
|
||||
CBNZ R21, chacha
|
||||
|
||||
// VLD4R (R10), [V16.S4, V17.S4, V18.S4, V19.S4]
|
||||
WORD $0x4D60E950
|
||||
|
||||
// VLD4R 16(R4), [V20.S4, V21.S4, V22.S4, V23.S4]
|
||||
WORD $0x4DFFE894
|
||||
VADD V30.S4, V12.S4, V12.S4
|
||||
VADD V16.S4, V0.S4, V0.S4
|
||||
VADD V17.S4, V1.S4, V1.S4
|
||||
VADD V18.S4, V2.S4, V2.S4
|
||||
VADD V19.S4, V3.S4, V3.S4
|
||||
// VLD4R 16(R4), [V24.S4, V25.S4, V26.S4, V27.S4]
|
||||
WORD $0x4DFFE898
|
||||
// restore R4
|
||||
SUB $32, R4
|
||||
|
||||
// load counter + nonce
|
||||
// VLD1R (R7), [V28.S4]
|
||||
WORD $0x4D40C8FC
|
||||
// VLD3R (R6), [V29.S4, V30.S4, V31.S4]
|
||||
WORD $0x4D40E8DD
|
||||
|
||||
VADD V20.S4, V4.S4, V4.S4
|
||||
VADD V21.S4, V5.S4, V5.S4
|
||||
VADD V22.S4, V6.S4, V6.S4
|
||||
VADD V23.S4, V7.S4, V7.S4
|
||||
VADD V24.S4, V8.S4, V8.S4
|
||||
VADD V25.S4, V9.S4, V9.S4
|
||||
VADD V26.S4, V10.S4, V10.S4
|
||||
VADD V27.S4, V11.S4, V11.S4
|
||||
VADD V28.S4, V12.S4, V12.S4
|
||||
VADD V29.S4, V13.S4, V13.S4
|
||||
VADD V30.S4, V14.S4, V14.S4
|
||||
VADD V31.S4, V15.S4, V15.S4
|
||||
|
||||
VZIP1 V1.S4, V0.S4, V16.S4
|
||||
VZIP2 V1.S4, V0.S4, V17.S4
|
||||
VZIP1 V3.S4, V2.S4, V18.S4
|
||||
VZIP2 V3.S4, V2.S4, V19.S4
|
||||
VZIP1 V5.S4, V4.S4, V20.S4
|
||||
VZIP2 V5.S4, V4.S4, V21.S4
|
||||
VZIP1 V7.S4, V6.S4, V22.S4
|
||||
VZIP2 V7.S4, V6.S4, V23.S4
|
||||
VZIP1 V9.S4, V8.S4, V24.S4
|
||||
VZIP2 V9.S4, V8.S4, V25.S4
|
||||
VZIP1 V11.S4, V10.S4, V26.S4
|
||||
VZIP2 V11.S4, V10.S4, V27.S4
|
||||
VZIP1 V13.S4, V12.S4, V28.S4
|
||||
VZIP2 V13.S4, V12.S4, V29.S4
|
||||
VZIP1 V15.S4, V14.S4, V30.S4
|
||||
VZIP2 V15.S4, V14.S4, V31.S4
|
||||
VZIP1 V18.D2, V16.D2, V0.D2
|
||||
VZIP2 V18.D2, V16.D2, V4.D2
|
||||
VZIP1 V19.D2, V17.D2, V8.D2
|
||||
VZIP2 V19.D2, V17.D2, V12.D2
|
||||
VLD1.P 64(R2), [V16.B16, V17.B16, V18.B16, V19.B16]
|
||||
|
||||
VZIP1 V22.D2, V20.D2, V1.D2
|
||||
VZIP2 V22.D2, V20.D2, V5.D2
|
||||
VZIP1 V23.D2, V21.D2, V9.D2
|
||||
VZIP2 V23.D2, V21.D2, V13.D2
|
||||
VLD1.P 64(R2), [V20.B16, V21.B16, V22.B16, V23.B16]
|
||||
VZIP1 V26.D2, V24.D2, V2.D2
|
||||
VZIP2 V26.D2, V24.D2, V6.D2
|
||||
VZIP1 V27.D2, V25.D2, V10.D2
|
||||
VZIP2 V27.D2, V25.D2, V14.D2
|
||||
VLD1.P 64(R2), [V24.B16, V25.B16, V26.B16, V27.B16]
|
||||
VZIP1 V30.D2, V28.D2, V3.D2
|
||||
VZIP2 V30.D2, V28.D2, V7.D2
|
||||
VZIP1 V31.D2, V29.D2, V11.D2
|
||||
VZIP2 V31.D2, V29.D2, V15.D2
|
||||
VLD1.P 64(R2), [V28.B16, V29.B16, V30.B16, V31.B16]
|
||||
VEOR V0.B16, V16.B16, V16.B16
|
||||
VEOR V1.B16, V17.B16, V17.B16
|
||||
VEOR V2.B16, V18.B16, V18.B16
|
||||
VEOR V3.B16, V19.B16, V19.B16
|
||||
VST1.P [V16.B16, V17.B16, V18.B16, V19.B16], 64(R1)
|
||||
VEOR V4.B16, V20.B16, V20.B16
|
||||
VEOR V5.B16, V21.B16, V21.B16
|
||||
VEOR V6.B16, V22.B16, V22.B16
|
||||
VEOR V7.B16, V23.B16, V23.B16
|
||||
VST1.P [V20.B16, V21.B16, V22.B16, V23.B16], 64(R1)
|
||||
VEOR V8.B16, V24.B16, V24.B16
|
||||
VEOR V9.B16, V25.B16, V25.B16
|
||||
VEOR V10.B16, V26.B16, V26.B16
|
||||
VEOR V11.B16, V27.B16, V27.B16
|
||||
VST1.P [V24.B16, V25.B16, V26.B16, V27.B16], 64(R1)
|
||||
VEOR V12.B16, V28.B16, V28.B16
|
||||
VEOR V13.B16, V29.B16, V29.B16
|
||||
VEOR V14.B16, V30.B16, V30.B16
|
||||
VEOR V15.B16, V31.B16, V31.B16
|
||||
VST1.P [V28.B16, V29.B16, V30.B16, V31.B16], 64(R1)
|
||||
|
||||
ADD $4, R20
|
||||
MOVW R20, (R7) // update counter
|
||||
|
||||
CMP R2, R12
|
||||
BGT loop
|
||||
|
||||
RET
|
||||
|
||||
|
||||
DATA ·constants+0x00(SB)/4, $0x61707865
|
||||
DATA ·constants+0x04(SB)/4, $0x3320646e
|
||||
DATA ·constants+0x08(SB)/4, $0x79622d32
|
||||
DATA ·constants+0x0c(SB)/4, $0x6b206574
|
||||
GLOBL ·constants(SB), NOPTR|RODATA, $32
|
||||
|
||||
DATA ·incRotMatrix+0x00(SB)/4, $0x00000000
|
||||
DATA ·incRotMatrix+0x04(SB)/4, $0x00000001
|
||||
DATA ·incRotMatrix+0x08(SB)/4, $0x00000002
|
||||
DATA ·incRotMatrix+0x0c(SB)/4, $0x00000003
|
||||
DATA ·incRotMatrix+0x10(SB)/4, $0x02010003
|
||||
DATA ·incRotMatrix+0x14(SB)/4, $0x06050407
|
||||
DATA ·incRotMatrix+0x18(SB)/4, $0x0A09080B
|
||||
DATA ·incRotMatrix+0x1c(SB)/4, $0x0E0D0C0F
|
||||
GLOBL ·incRotMatrix(SB), NOPTR|RODATA, $32
|
||||
400
vendor/golang.org/x/crypto/chacha20/chacha_generic.go
generated
vendored
Normal file
400
vendor/golang.org/x/crypto/chacha20/chacha_generic.go
generated
vendored
Normal file
@@ -0,0 +1,400 @@
|
||||
// Copyright 2016 The Go Authors. All rights reserved.
|
||||
// Use of this source code is governed by a BSD-style
|
||||
// license that can be found in the LICENSE file.
|
||||
|
||||
// Package chacha20 implements the ChaCha20 and XChaCha20 encryption algorithms
|
||||
// as specified in RFC 8439 and draft-irtf-cfrg-xchacha-01.
|
||||
package chacha20
|
||||
|
||||
import (
|
||||
"crypto/cipher"
|
||||
"encoding/binary"
|
||||
"errors"
|
||||
"math/bits"
|
||||
|
||||
"golang.org/x/crypto/internal/alias"
|
||||
)
|
||||
|
||||
const (
|
||||
// KeySize is the size of the key used by this cipher, in bytes.
|
||||
KeySize = 32
|
||||
|
||||
// NonceSize is the size of the nonce used with the standard variant of this
|
||||
// cipher, in bytes.
|
||||
//
|
||||
// Note that this is too short to be safely generated at random if the same
|
||||
// key is reused more than 2³² times.
|
||||
//NonceSize = 12
|
||||
// BITE ME, GOLANG DEVS.
|
||||
NonceSize = 16
|
||||
|
||||
// NonceSizeX is the size of the nonce used with the XChaCha20 variant of
|
||||
// this cipher, in bytes.
|
||||
NonceSizeX = 24
|
||||
)
|
||||
|
||||
// Cipher is a stateful instance of ChaCha20 or XChaCha20 using a particular key
|
||||
// and nonce. A *Cipher implements the cipher.Stream interface.
|
||||
type Cipher struct {
|
||||
// The ChaCha20 state is 16 words: 4 constant, 8 of key, 1 of counter
|
||||
// (incremented after each block), and 3 of nonce.
|
||||
key [8]uint32
|
||||
counter uint32
|
||||
nonce [3]uint32
|
||||
|
||||
// The last len bytes of buf are leftover key stream bytes from the previous
|
||||
// XORKeyStream invocation. The size of buf depends on how many blocks are
|
||||
// computed at a time by xorKeyStreamBlocks.
|
||||
buf [bufSize]byte
|
||||
len int
|
||||
|
||||
// overflow is set when the counter overflowed, no more blocks can be
|
||||
// generated, and the next XORKeyStream call should panic.
|
||||
overflow bool
|
||||
|
||||
// The counter-independent results of the first round are cached after they
|
||||
// are computed the first time.
|
||||
precompDone bool
|
||||
p1, p5, p9, p13 uint32
|
||||
p2, p6, p10, p14 uint32
|
||||
p3, p7, p11, p15 uint32
|
||||
}
|
||||
|
||||
var _ cipher.Stream = (*Cipher)(nil)
|
||||
|
||||
// NewUnauthenticatedCipher creates a new ChaCha20 stream cipher with the given
|
||||
// 32 bytes key and a 12 or 24 bytes nonce. If a nonce of 24 bytes is provided,
|
||||
// the XChaCha20 construction will be used. It returns an error if key or nonce
|
||||
// have any other length.
|
||||
//
|
||||
// Note that ChaCha20, like all stream ciphers, is not authenticated and allows
|
||||
// attackers to silently tamper with the plaintext. For this reason, it is more
|
||||
// appropriate as a building block than as a standalone encryption mechanism.
|
||||
// Instead, consider using package golang.org/x/crypto/chacha20poly1305.
|
||||
func NewUnauthenticatedCipher(key, nonce []byte) (*Cipher, error) {
|
||||
// This function is split into a wrapper so that the Cipher allocation will
|
||||
// be inlined, and depending on how the caller uses the return value, won't
|
||||
// escape to the heap.
|
||||
c := &Cipher{}
|
||||
return newUnauthenticatedCipher(c, key, nonce)
|
||||
}
|
||||
|
||||
func newUnauthenticatedCipher(c *Cipher, key, nonce []byte) (*Cipher, error) {
|
||||
if len(key) != KeySize {
|
||||
return nil, errors.New("chacha20: wrong key size")
|
||||
}
|
||||
if len(nonce) == NonceSizeX {
|
||||
// XChaCha20 uses the ChaCha20 core to mix 16 bytes of the nonce into a
|
||||
// derived key, allowing it to operate on a nonce of 24 bytes. See
|
||||
// draft-irtf-cfrg-xchacha-01, Section 2.3.
|
||||
key, _ = HChaCha20(key, nonce[0:16])
|
||||
cNonce := make([]byte, NonceSize)
|
||||
copy(cNonce[4:12], nonce[16:24])
|
||||
nonce = cNonce
|
||||
} else if len(nonce) != NonceSize {
|
||||
return nil, errors.New("chacha20: wrong nonce size")
|
||||
}
|
||||
|
||||
key, nonce = key[:KeySize], nonce[:NonceSize] // bounds check elimination hint
|
||||
c.key = [8]uint32{
|
||||
binary.LittleEndian.Uint32(key[0:4]),
|
||||
binary.LittleEndian.Uint32(key[4:8]),
|
||||
binary.LittleEndian.Uint32(key[8:12]),
|
||||
binary.LittleEndian.Uint32(key[12:16]),
|
||||
binary.LittleEndian.Uint32(key[16:20]),
|
||||
binary.LittleEndian.Uint32(key[20:24]),
|
||||
binary.LittleEndian.Uint32(key[24:28]),
|
||||
binary.LittleEndian.Uint32(key[28:32]),
|
||||
}
|
||||
c.nonce = [3]uint32{
|
||||
binary.LittleEndian.Uint32(nonce[0:4]),
|
||||
binary.LittleEndian.Uint32(nonce[4:8]),
|
||||
binary.LittleEndian.Uint32(nonce[8:12]),
|
||||
}
|
||||
return c, nil
|
||||
}
|
||||
|
||||
// The constant first 4 words of the ChaCha20 state.
|
||||
const (
|
||||
j0 uint32 = 0x61707865 // expa
|
||||
j1 uint32 = 0x3320646e // nd 3
|
||||
j2 uint32 = 0x79622d32 // 2-by
|
||||
j3 uint32 = 0x6b206574 // te k
|
||||
)
|
||||
|
||||
const blockSize = 64
|
||||
|
||||
// quarterRound is the core of ChaCha20. It shuffles the bits of 4 state words.
|
||||
// It's executed 4 times for each of the 20 ChaCha20 rounds, operating on all 16
|
||||
// words each round, in columnar or diagonal groups of 4 at a time.
|
||||
func quarterRound(a, b, c, d uint32) (uint32, uint32, uint32, uint32) {
|
||||
a += b
|
||||
d ^= a
|
||||
d = bits.RotateLeft32(d, 16)
|
||||
c += d
|
||||
b ^= c
|
||||
b = bits.RotateLeft32(b, 12)
|
||||
a += b
|
||||
d ^= a
|
||||
d = bits.RotateLeft32(d, 8)
|
||||
c += d
|
||||
b ^= c
|
||||
b = bits.RotateLeft32(b, 7)
|
||||
return a, b, c, d
|
||||
}
|
||||
|
||||
// SetCounter sets the Cipher counter. The next invocation of XORKeyStream will
|
||||
// behave as if (64 * counter) bytes had been encrypted so far.
|
||||
//
|
||||
// To prevent accidental counter reuse, SetCounter panics if counter is less
|
||||
// than the current value.
|
||||
//
|
||||
// Note that the execution time of XORKeyStream is not independent of the
|
||||
// counter value.
|
||||
func (s *Cipher) SetCounter(counter uint32) {
|
||||
// Internally, s may buffer multiple blocks, which complicates this
|
||||
// implementation slightly. When checking whether the counter has rolled
|
||||
// back, we must use both s.counter and s.len to determine how many blocks
|
||||
// we have already output.
|
||||
outputCounter := s.counter - uint32(s.len)/blockSize
|
||||
if s.overflow || counter < outputCounter {
|
||||
panic("chacha20: SetCounter attempted to rollback counter")
|
||||
}
|
||||
|
||||
// In the general case, we set the new counter value and reset s.len to 0,
|
||||
// causing the next call to XORKeyStream to refill the buffer. However, if
|
||||
// we're advancing within the existing buffer, we can save work by simply
|
||||
// setting s.len.
|
||||
if counter < s.counter {
|
||||
s.len = int(s.counter-counter) * blockSize
|
||||
} else {
|
||||
s.counter = counter
|
||||
s.len = 0
|
||||
}
|
||||
}
|
||||
|
||||
// XORKeyStream XORs each byte in the given slice with a byte from the
|
||||
// cipher's key stream. Dst and src must overlap entirely or not at all.
|
||||
//
|
||||
// If len(dst) < len(src), XORKeyStream will panic. It is acceptable
|
||||
// to pass a dst bigger than src, and in that case, XORKeyStream will
|
||||
// only update dst[:len(src)] and will not touch the rest of dst.
|
||||
//
|
||||
// Multiple calls to XORKeyStream behave as if the concatenation of
|
||||
// the src buffers was passed in a single run. That is, Cipher
|
||||
// maintains state and does not reset at each XORKeyStream call.
|
||||
func (s *Cipher) XORKeyStream(dst, src []byte) {
|
||||
if len(src) == 0 {
|
||||
return
|
||||
}
|
||||
if len(dst) < len(src) {
|
||||
panic("chacha20: output smaller than input")
|
||||
}
|
||||
dst = dst[:len(src)]
|
||||
if alias.InexactOverlap(dst, src) {
|
||||
panic("chacha20: invalid buffer overlap")
|
||||
}
|
||||
|
||||
// First, drain any remaining key stream from a previous XORKeyStream.
|
||||
if s.len != 0 {
|
||||
keyStream := s.buf[bufSize-s.len:]
|
||||
if len(src) < len(keyStream) {
|
||||
keyStream = keyStream[:len(src)]
|
||||
}
|
||||
_ = src[len(keyStream)-1] // bounds check elimination hint
|
||||
for i, b := range keyStream {
|
||||
dst[i] = src[i] ^ b
|
||||
}
|
||||
s.len -= len(keyStream)
|
||||
dst, src = dst[len(keyStream):], src[len(keyStream):]
|
||||
}
|
||||
if len(src) == 0 {
|
||||
return
|
||||
}
|
||||
|
||||
// If we'd need to let the counter overflow and keep generating output,
|
||||
// panic immediately. If instead we'd only reach the last block, remember
|
||||
// not to generate any more output after the buffer is drained.
|
||||
numBlocks := (uint64(len(src)) + blockSize - 1) / blockSize
|
||||
if s.overflow || uint64(s.counter)+numBlocks > 1<<32 {
|
||||
panic("chacha20: counter overflow")
|
||||
} else if uint64(s.counter)+numBlocks == 1<<32 {
|
||||
s.overflow = true
|
||||
}
|
||||
|
||||
// xorKeyStreamBlocks implementations expect input lengths that are a
|
||||
// multiple of bufSize. Platform-specific ones process multiple blocks at a
|
||||
// time, so have bufSizes that are a multiple of blockSize.
|
||||
|
||||
full := len(src) - len(src)%bufSize
|
||||
if full > 0 {
|
||||
s.xorKeyStreamBlocks(dst[:full], src[:full])
|
||||
}
|
||||
dst, src = dst[full:], src[full:]
|
||||
|
||||
// If using a multi-block xorKeyStreamBlocks would overflow, use the generic
|
||||
// one that does one block at a time.
|
||||
const blocksPerBuf = bufSize / blockSize
|
||||
if uint64(s.counter)+blocksPerBuf > 1<<32 {
|
||||
s.buf = [bufSize]byte{}
|
||||
numBlocks := (len(src) + blockSize - 1) / blockSize
|
||||
buf := s.buf[bufSize-numBlocks*blockSize:]
|
||||
copy(buf, src)
|
||||
s.xorKeyStreamBlocksGeneric(buf, buf)
|
||||
s.len = len(buf) - copy(dst, buf)
|
||||
return
|
||||
}
|
||||
|
||||
// If we have a partial (multi-)block, pad it for xorKeyStreamBlocks, and
|
||||
// keep the leftover keystream for the next XORKeyStream invocation.
|
||||
if len(src) > 0 {
|
||||
s.buf = [bufSize]byte{}
|
||||
copy(s.buf[:], src)
|
||||
s.xorKeyStreamBlocks(s.buf[:], s.buf[:])
|
||||
s.len = bufSize - copy(dst, s.buf[:])
|
||||
}
|
||||
}
|
||||
|
||||
func (s *Cipher) xorKeyStreamBlocksGeneric(dst, src []byte) {
|
||||
if len(dst) != len(src) || len(dst)%blockSize != 0 {
|
||||
panic("chacha20: internal error: wrong dst and/or src length")
|
||||
}
|
||||
|
||||
// To generate each block of key stream, the initial cipher state
|
||||
// (represented below) is passed through 20 rounds of shuffling,
|
||||
// alternatively applying quarterRounds by columns (like 1, 5, 9, 13)
|
||||
// or by diagonals (like 1, 6, 11, 12).
|
||||
//
|
||||
// 0:cccccccc 1:cccccccc 2:cccccccc 3:cccccccc
|
||||
// 4:kkkkkkkk 5:kkkkkkkk 6:kkkkkkkk 7:kkkkkkkk
|
||||
// 8:kkkkkkkk 9:kkkkkkkk 10:kkkkkkkk 11:kkkkkkkk
|
||||
// 12:bbbbbbbb 13:nnnnnnnn 14:nnnnnnnn 15:nnnnnnnn
|
||||
//
|
||||
// c=constant k=key b=blockcount n=nonce
|
||||
var (
|
||||
c0, c1, c2, c3 = j0, j1, j2, j3
|
||||
c4, c5, c6, c7 = s.key[0], s.key[1], s.key[2], s.key[3]
|
||||
c8, c9, c10, c11 = s.key[4], s.key[5], s.key[6], s.key[7]
|
||||
_, c13, c14, c15 = s.counter, s.nonce[0], s.nonce[1], s.nonce[2]
|
||||
)
|
||||
|
||||
// Three quarters of the first round don't depend on the counter, so we can
|
||||
// calculate them here, and reuse them for multiple blocks in the loop, and
|
||||
// for future XORKeyStream invocations.
|
||||
if !s.precompDone {
|
||||
s.p1, s.p5, s.p9, s.p13 = quarterRound(c1, c5, c9, c13)
|
||||
s.p2, s.p6, s.p10, s.p14 = quarterRound(c2, c6, c10, c14)
|
||||
s.p3, s.p7, s.p11, s.p15 = quarterRound(c3, c7, c11, c15)
|
||||
s.precompDone = true
|
||||
}
|
||||
|
||||
// A condition of len(src) > 0 would be sufficient, but this also
|
||||
// acts as a bounds check elimination hint.
|
||||
for len(src) >= 64 && len(dst) >= 64 {
|
||||
// The remainder of the first column round.
|
||||
fcr0, fcr4, fcr8, fcr12 := quarterRound(c0, c4, c8, s.counter)
|
||||
|
||||
// The second diagonal round.
|
||||
x0, x5, x10, x15 := quarterRound(fcr0, s.p5, s.p10, s.p15)
|
||||
x1, x6, x11, x12 := quarterRound(s.p1, s.p6, s.p11, fcr12)
|
||||
x2, x7, x8, x13 := quarterRound(s.p2, s.p7, fcr8, s.p13)
|
||||
x3, x4, x9, x14 := quarterRound(s.p3, fcr4, s.p9, s.p14)
|
||||
|
||||
// The remaining 18 rounds.
|
||||
for i := 0; i < 9; i++ {
|
||||
// Column round.
|
||||
x0, x4, x8, x12 = quarterRound(x0, x4, x8, x12)
|
||||
x1, x5, x9, x13 = quarterRound(x1, x5, x9, x13)
|
||||
x2, x6, x10, x14 = quarterRound(x2, x6, x10, x14)
|
||||
x3, x7, x11, x15 = quarterRound(x3, x7, x11, x15)
|
||||
|
||||
// Diagonal round.
|
||||
x0, x5, x10, x15 = quarterRound(x0, x5, x10, x15)
|
||||
x1, x6, x11, x12 = quarterRound(x1, x6, x11, x12)
|
||||
x2, x7, x8, x13 = quarterRound(x2, x7, x8, x13)
|
||||
x3, x4, x9, x14 = quarterRound(x3, x4, x9, x14)
|
||||
}
|
||||
|
||||
// Add back the initial state to generate the key stream, then
|
||||
// XOR the key stream with the source and write out the result.
|
||||
addXor(dst[0:4], src[0:4], x0, c0)
|
||||
addXor(dst[4:8], src[4:8], x1, c1)
|
||||
addXor(dst[8:12], src[8:12], x2, c2)
|
||||
addXor(dst[12:16], src[12:16], x3, c3)
|
||||
addXor(dst[16:20], src[16:20], x4, c4)
|
||||
addXor(dst[20:24], src[20:24], x5, c5)
|
||||
addXor(dst[24:28], src[24:28], x6, c6)
|
||||
addXor(dst[28:32], src[28:32], x7, c7)
|
||||
addXor(dst[32:36], src[32:36], x8, c8)
|
||||
addXor(dst[36:40], src[36:40], x9, c9)
|
||||
addXor(dst[40:44], src[40:44], x10, c10)
|
||||
addXor(dst[44:48], src[44:48], x11, c11)
|
||||
addXor(dst[48:52], src[48:52], x12, s.counter)
|
||||
addXor(dst[52:56], src[52:56], x13, c13)
|
||||
addXor(dst[56:60], src[56:60], x14, c14)
|
||||
addXor(dst[60:64], src[60:64], x15, c15)
|
||||
|
||||
s.counter += 1
|
||||
|
||||
src, dst = src[blockSize:], dst[blockSize:]
|
||||
}
|
||||
}
|
||||
|
||||
// HChaCha20 uses the ChaCha20 core to generate a derived key from a 32 bytes
|
||||
// key and a 16 bytes nonce. It returns an error if key or nonce have any other
|
||||
// length. It is used as part of the XChaCha20 construction.
|
||||
func HChaCha20(key, nonce []byte) ([]byte, error) {
|
||||
// This function is split into a wrapper so that the slice allocation will
|
||||
// be inlined, and depending on how the caller uses the return value, won't
|
||||
// escape to the heap.
|
||||
out := make([]byte, 32)
|
||||
return hChaCha20(out, key, nonce)
|
||||
}
|
||||
|
||||
func hChaCha20(out, key, nonce []byte) ([]byte, error) {
|
||||
if len(key) != KeySize {
|
||||
return nil, errors.New("chacha20: wrong HChaCha20 key size")
|
||||
}
|
||||
if len(nonce) != 16 {
|
||||
return nil, errors.New("chacha20: wrong HChaCha20 nonce size")
|
||||
}
|
||||
|
||||
x0, x1, x2, x3 := j0, j1, j2, j3
|
||||
x4 := binary.LittleEndian.Uint32(key[0:4])
|
||||
x5 := binary.LittleEndian.Uint32(key[4:8])
|
||||
x6 := binary.LittleEndian.Uint32(key[8:12])
|
||||
x7 := binary.LittleEndian.Uint32(key[12:16])
|
||||
x8 := binary.LittleEndian.Uint32(key[16:20])
|
||||
x9 := binary.LittleEndian.Uint32(key[20:24])
|
||||
x10 := binary.LittleEndian.Uint32(key[24:28])
|
||||
x11 := binary.LittleEndian.Uint32(key[28:32])
|
||||
x12 := binary.LittleEndian.Uint32(nonce[0:4])
|
||||
x13 := binary.LittleEndian.Uint32(nonce[4:8])
|
||||
x14 := binary.LittleEndian.Uint32(nonce[8:12])
|
||||
x15 := binary.LittleEndian.Uint32(nonce[12:16])
|
||||
|
||||
for i := 0; i < 10; i++ {
|
||||
// Diagonal round.
|
||||
x0, x4, x8, x12 = quarterRound(x0, x4, x8, x12)
|
||||
x1, x5, x9, x13 = quarterRound(x1, x5, x9, x13)
|
||||
x2, x6, x10, x14 = quarterRound(x2, x6, x10, x14)
|
||||
x3, x7, x11, x15 = quarterRound(x3, x7, x11, x15)
|
||||
|
||||
// Column round.
|
||||
x0, x5, x10, x15 = quarterRound(x0, x5, x10, x15)
|
||||
x1, x6, x11, x12 = quarterRound(x1, x6, x11, x12)
|
||||
x2, x7, x8, x13 = quarterRound(x2, x7, x8, x13)
|
||||
x3, x4, x9, x14 = quarterRound(x3, x4, x9, x14)
|
||||
}
|
||||
|
||||
_ = out[31] // bounds check elimination hint
|
||||
binary.LittleEndian.PutUint32(out[0:4], x0)
|
||||
binary.LittleEndian.PutUint32(out[4:8], x1)
|
||||
binary.LittleEndian.PutUint32(out[8:12], x2)
|
||||
binary.LittleEndian.PutUint32(out[12:16], x3)
|
||||
binary.LittleEndian.PutUint32(out[16:20], x12)
|
||||
binary.LittleEndian.PutUint32(out[20:24], x13)
|
||||
binary.LittleEndian.PutUint32(out[24:28], x14)
|
||||
binary.LittleEndian.PutUint32(out[28:32], x15)
|
||||
return out, nil
|
||||
}
|
||||
14
vendor/golang.org/x/crypto/chacha20/chacha_noasm.go
generated
vendored
Normal file
14
vendor/golang.org/x/crypto/chacha20/chacha_noasm.go
generated
vendored
Normal file
@@ -0,0 +1,14 @@
|
||||
// Copyright 2018 The Go Authors. All rights reserved.
|
||||
// Use of this source code is governed by a BSD-style
|
||||
// license that can be found in the LICENSE file.
|
||||
|
||||
//go:build (!arm64 && !s390x && !ppc64le) || (arm64 && !go1.11) || !gc || purego
|
||||
// +build !arm64,!s390x,!ppc64le arm64,!go1.11 !gc purego
|
||||
|
||||
package chacha20
|
||||
|
||||
const bufSize = blockSize
|
||||
|
||||
func (s *Cipher) xorKeyStreamBlocks(dst, src []byte) {
|
||||
s.xorKeyStreamBlocksGeneric(dst, src)
|
||||
}
|
||||
17
vendor/golang.org/x/crypto/chacha20/chacha_ppc64le.go
generated
vendored
Normal file
17
vendor/golang.org/x/crypto/chacha20/chacha_ppc64le.go
generated
vendored
Normal file
@@ -0,0 +1,17 @@
|
||||
// Copyright 2019 The Go Authors. All rights reserved.
|
||||
// Use of this source code is governed by a BSD-style
|
||||
// license that can be found in the LICENSE file.
|
||||
|
||||
//go:build gc && !purego
|
||||
// +build gc,!purego
|
||||
|
||||
package chacha20
|
||||
|
||||
const bufSize = 256
|
||||
|
||||
//go:noescape
|
||||
func chaCha20_ctr32_vsx(out, inp *byte, len int, key *[8]uint32, counter *uint32)
|
||||
|
||||
func (c *Cipher) xorKeyStreamBlocks(dst, src []byte) {
|
||||
chaCha20_ctr32_vsx(&dst[0], &src[0], len(src), &c.key, &c.counter)
|
||||
}
|
||||
450
vendor/golang.org/x/crypto/chacha20/chacha_ppc64le.s
generated
vendored
Normal file
450
vendor/golang.org/x/crypto/chacha20/chacha_ppc64le.s
generated
vendored
Normal file
@@ -0,0 +1,450 @@
|
||||
// Copyright 2019 The Go Authors. All rights reserved.
|
||||
// Use of this source code is governed by a BSD-style
|
||||
// license that can be found in the LICENSE file.
|
||||
|
||||
// Based on CRYPTOGAMS code with the following comment:
|
||||
// # ====================================================================
|
||||
// # Written by Andy Polyakov <appro@openssl.org> for the OpenSSL
|
||||
// # project. The module is, however, dual licensed under OpenSSL and
|
||||
// # CRYPTOGAMS licenses depending on where you obtain it. For further
|
||||
// # details see http://www.openssl.org/~appro/cryptogams/.
|
||||
// # ====================================================================
|
||||
|
||||
// Code for the perl script that generates the ppc64 assembler
|
||||
// can be found in the cryptogams repository at the link below. It is based on
|
||||
// the original from openssl.
|
||||
|
||||
// https://github.com/dot-asm/cryptogams/commit/a60f5b50ed908e91
|
||||
|
||||
// The differences in this and the original implementation are
|
||||
// due to the calling conventions and initialization of constants.
|
||||
|
||||
//go:build gc && !purego
|
||||
// +build gc,!purego
|
||||
|
||||
#include "textflag.h"
|
||||
|
||||
#define OUT R3
|
||||
#define INP R4
|
||||
#define LEN R5
|
||||
#define KEY R6
|
||||
#define CNT R7
|
||||
#define TMP R15
|
||||
|
||||
#define CONSTBASE R16
|
||||
#define BLOCKS R17
|
||||
|
||||
DATA consts<>+0x00(SB)/8, $0x3320646e61707865
|
||||
DATA consts<>+0x08(SB)/8, $0x6b20657479622d32
|
||||
DATA consts<>+0x10(SB)/8, $0x0000000000000001
|
||||
DATA consts<>+0x18(SB)/8, $0x0000000000000000
|
||||
DATA consts<>+0x20(SB)/8, $0x0000000000000004
|
||||
DATA consts<>+0x28(SB)/8, $0x0000000000000000
|
||||
DATA consts<>+0x30(SB)/8, $0x0a0b08090e0f0c0d
|
||||
DATA consts<>+0x38(SB)/8, $0x0203000106070405
|
||||
DATA consts<>+0x40(SB)/8, $0x090a0b080d0e0f0c
|
||||
DATA consts<>+0x48(SB)/8, $0x0102030005060704
|
||||
DATA consts<>+0x50(SB)/8, $0x6170786561707865
|
||||
DATA consts<>+0x58(SB)/8, $0x6170786561707865
|
||||
DATA consts<>+0x60(SB)/8, $0x3320646e3320646e
|
||||
DATA consts<>+0x68(SB)/8, $0x3320646e3320646e
|
||||
DATA consts<>+0x70(SB)/8, $0x79622d3279622d32
|
||||
DATA consts<>+0x78(SB)/8, $0x79622d3279622d32
|
||||
DATA consts<>+0x80(SB)/8, $0x6b2065746b206574
|
||||
DATA consts<>+0x88(SB)/8, $0x6b2065746b206574
|
||||
DATA consts<>+0x90(SB)/8, $0x0000000100000000
|
||||
DATA consts<>+0x98(SB)/8, $0x0000000300000002
|
||||
GLOBL consts<>(SB), RODATA, $0xa0
|
||||
|
||||
//func chaCha20_ctr32_vsx(out, inp *byte, len int, key *[8]uint32, counter *uint32)
|
||||
TEXT ·chaCha20_ctr32_vsx(SB),NOSPLIT,$64-40
|
||||
MOVD out+0(FP), OUT
|
||||
MOVD inp+8(FP), INP
|
||||
MOVD len+16(FP), LEN
|
||||
MOVD key+24(FP), KEY
|
||||
MOVD counter+32(FP), CNT
|
||||
|
||||
// Addressing for constants
|
||||
MOVD $consts<>+0x00(SB), CONSTBASE
|
||||
MOVD $16, R8
|
||||
MOVD $32, R9
|
||||
MOVD $48, R10
|
||||
MOVD $64, R11
|
||||
SRD $6, LEN, BLOCKS
|
||||
// V16
|
||||
LXVW4X (CONSTBASE)(R0), VS48
|
||||
ADD $80,CONSTBASE
|
||||
|
||||
// Load key into V17,V18
|
||||
LXVW4X (KEY)(R0), VS49
|
||||
LXVW4X (KEY)(R8), VS50
|
||||
|
||||
// Load CNT, NONCE into V19
|
||||
LXVW4X (CNT)(R0), VS51
|
||||
|
||||
// Clear V27
|
||||
VXOR V27, V27, V27
|
||||
|
||||
// V28
|
||||
LXVW4X (CONSTBASE)(R11), VS60
|
||||
|
||||
// splat slot from V19 -> V26
|
||||
VSPLTW $0, V19, V26
|
||||
|
||||
VSLDOI $4, V19, V27, V19
|
||||
VSLDOI $12, V27, V19, V19
|
||||
|
||||
VADDUWM V26, V28, V26
|
||||
|
||||
MOVD $10, R14
|
||||
MOVD R14, CTR
|
||||
|
||||
loop_outer_vsx:
|
||||
// V0, V1, V2, V3
|
||||
LXVW4X (R0)(CONSTBASE), VS32
|
||||
LXVW4X (R8)(CONSTBASE), VS33
|
||||
LXVW4X (R9)(CONSTBASE), VS34
|
||||
LXVW4X (R10)(CONSTBASE), VS35
|
||||
|
||||
// splat values from V17, V18 into V4-V11
|
||||
VSPLTW $0, V17, V4
|
||||
VSPLTW $1, V17, V5
|
||||
VSPLTW $2, V17, V6
|
||||
VSPLTW $3, V17, V7
|
||||
VSPLTW $0, V18, V8
|
||||
VSPLTW $1, V18, V9
|
||||
VSPLTW $2, V18, V10
|
||||
VSPLTW $3, V18, V11
|
||||
|
||||
// VOR
|
||||
VOR V26, V26, V12
|
||||
|
||||
// splat values from V19 -> V13, V14, V15
|
||||
VSPLTW $1, V19, V13
|
||||
VSPLTW $2, V19, V14
|
||||
VSPLTW $3, V19, V15
|
||||
|
||||
// splat const values
|
||||
VSPLTISW $-16, V27
|
||||
VSPLTISW $12, V28
|
||||
VSPLTISW $8, V29
|
||||
VSPLTISW $7, V30
|
||||
|
||||
loop_vsx:
|
||||
VADDUWM V0, V4, V0
|
||||
VADDUWM V1, V5, V1
|
||||
VADDUWM V2, V6, V2
|
||||
VADDUWM V3, V7, V3
|
||||
|
||||
VXOR V12, V0, V12
|
||||
VXOR V13, V1, V13
|
||||
VXOR V14, V2, V14
|
||||
VXOR V15, V3, V15
|
||||
|
||||
VRLW V12, V27, V12
|
||||
VRLW V13, V27, V13
|
||||
VRLW V14, V27, V14
|
||||
VRLW V15, V27, V15
|
||||
|
||||
VADDUWM V8, V12, V8
|
||||
VADDUWM V9, V13, V9
|
||||
VADDUWM V10, V14, V10
|
||||
VADDUWM V11, V15, V11
|
||||
|
||||
VXOR V4, V8, V4
|
||||
VXOR V5, V9, V5
|
||||
VXOR V6, V10, V6
|
||||
VXOR V7, V11, V7
|
||||
|
||||
VRLW V4, V28, V4
|
||||
VRLW V5, V28, V5
|
||||
VRLW V6, V28, V6
|
||||
VRLW V7, V28, V7
|
||||
|
||||
VADDUWM V0, V4, V0
|
||||
VADDUWM V1, V5, V1
|
||||
VADDUWM V2, V6, V2
|
||||
VADDUWM V3, V7, V3
|
||||
|
||||
VXOR V12, V0, V12
|
||||
VXOR V13, V1, V13
|
||||
VXOR V14, V2, V14
|
||||
VXOR V15, V3, V15
|
||||
|
||||
VRLW V12, V29, V12
|
||||
VRLW V13, V29, V13
|
||||
VRLW V14, V29, V14
|
||||
VRLW V15, V29, V15
|
||||
|
||||
VADDUWM V8, V12, V8
|
||||
VADDUWM V9, V13, V9
|
||||
VADDUWM V10, V14, V10
|
||||
VADDUWM V11, V15, V11
|
||||
|
||||
VXOR V4, V8, V4
|
||||
VXOR V5, V9, V5
|
||||
VXOR V6, V10, V6
|
||||
VXOR V7, V11, V7
|
||||
|
||||
VRLW V4, V30, V4
|
||||
VRLW V5, V30, V5
|
||||
VRLW V6, V30, V6
|
||||
VRLW V7, V30, V7
|
||||
|
||||
VADDUWM V0, V5, V0
|
||||
VADDUWM V1, V6, V1
|
||||
VADDUWM V2, V7, V2
|
||||
VADDUWM V3, V4, V3
|
||||
|
||||
VXOR V15, V0, V15
|
||||
VXOR V12, V1, V12
|
||||
VXOR V13, V2, V13
|
||||
VXOR V14, V3, V14
|
||||
|
||||
VRLW V15, V27, V15
|
||||
VRLW V12, V27, V12
|
||||
VRLW V13, V27, V13
|
||||
VRLW V14, V27, V14
|
||||
|
||||
VADDUWM V10, V15, V10
|
||||
VADDUWM V11, V12, V11
|
||||
VADDUWM V8, V13, V8
|
||||
VADDUWM V9, V14, V9
|
||||
|
||||
VXOR V5, V10, V5
|
||||
VXOR V6, V11, V6
|
||||
VXOR V7, V8, V7
|
||||
VXOR V4, V9, V4
|
||||
|
||||
VRLW V5, V28, V5
|
||||
VRLW V6, V28, V6
|
||||
VRLW V7, V28, V7
|
||||
VRLW V4, V28, V4
|
||||
|
||||
VADDUWM V0, V5, V0
|
||||
VADDUWM V1, V6, V1
|
||||
VADDUWM V2, V7, V2
|
||||
VADDUWM V3, V4, V3
|
||||
|
||||
VXOR V15, V0, V15
|
||||
VXOR V12, V1, V12
|
||||
VXOR V13, V2, V13
|
||||
VXOR V14, V3, V14
|
||||
|
||||
VRLW V15, V29, V15
|
||||
VRLW V12, V29, V12
|
||||
VRLW V13, V29, V13
|
||||
VRLW V14, V29, V14
|
||||
|
||||
VADDUWM V10, V15, V10
|
||||
VADDUWM V11, V12, V11
|
||||
VADDUWM V8, V13, V8
|
||||
VADDUWM V9, V14, V9
|
||||
|
||||
VXOR V5, V10, V5
|
||||
VXOR V6, V11, V6
|
||||
VXOR V7, V8, V7
|
||||
VXOR V4, V9, V4
|
||||
|
||||
VRLW V5, V30, V5
|
||||
VRLW V6, V30, V6
|
||||
VRLW V7, V30, V7
|
||||
VRLW V4, V30, V4
|
||||
BC 16, LT, loop_vsx
|
||||
|
||||
VADDUWM V12, V26, V12
|
||||
|
||||
WORD $0x13600F8C // VMRGEW V0, V1, V27
|
||||
WORD $0x13821F8C // VMRGEW V2, V3, V28
|
||||
|
||||
WORD $0x10000E8C // VMRGOW V0, V1, V0
|
||||
WORD $0x10421E8C // VMRGOW V2, V3, V2
|
||||
|
||||
WORD $0x13A42F8C // VMRGEW V4, V5, V29
|
||||
WORD $0x13C63F8C // VMRGEW V6, V7, V30
|
||||
|
||||
XXPERMDI VS32, VS34, $0, VS33
|
||||
XXPERMDI VS32, VS34, $3, VS35
|
||||
XXPERMDI VS59, VS60, $0, VS32
|
||||
XXPERMDI VS59, VS60, $3, VS34
|
||||
|
||||
WORD $0x10842E8C // VMRGOW V4, V5, V4
|
||||
WORD $0x10C63E8C // VMRGOW V6, V7, V6
|
||||
|
||||
WORD $0x13684F8C // VMRGEW V8, V9, V27
|
||||
WORD $0x138A5F8C // VMRGEW V10, V11, V28
|
||||
|
||||
XXPERMDI VS36, VS38, $0, VS37
|
||||
XXPERMDI VS36, VS38, $3, VS39
|
||||
XXPERMDI VS61, VS62, $0, VS36
|
||||
XXPERMDI VS61, VS62, $3, VS38
|
||||
|
||||
WORD $0x11084E8C // VMRGOW V8, V9, V8
|
||||
WORD $0x114A5E8C // VMRGOW V10, V11, V10
|
||||
|
||||
WORD $0x13AC6F8C // VMRGEW V12, V13, V29
|
||||
WORD $0x13CE7F8C // VMRGEW V14, V15, V30
|
||||
|
||||
XXPERMDI VS40, VS42, $0, VS41
|
||||
XXPERMDI VS40, VS42, $3, VS43
|
||||
XXPERMDI VS59, VS60, $0, VS40
|
||||
XXPERMDI VS59, VS60, $3, VS42
|
||||
|
||||
WORD $0x118C6E8C // VMRGOW V12, V13, V12
|
||||
WORD $0x11CE7E8C // VMRGOW V14, V15, V14
|
||||
|
||||
VSPLTISW $4, V27
|
||||
VADDUWM V26, V27, V26
|
||||
|
||||
XXPERMDI VS44, VS46, $0, VS45
|
||||
XXPERMDI VS44, VS46, $3, VS47
|
||||
XXPERMDI VS61, VS62, $0, VS44
|
||||
XXPERMDI VS61, VS62, $3, VS46
|
||||
|
||||
VADDUWM V0, V16, V0
|
||||
VADDUWM V4, V17, V4
|
||||
VADDUWM V8, V18, V8
|
||||
VADDUWM V12, V19, V12
|
||||
|
||||
CMPU LEN, $64
|
||||
BLT tail_vsx
|
||||
|
||||
// Bottom of loop
|
||||
LXVW4X (INP)(R0), VS59
|
||||
LXVW4X (INP)(R8), VS60
|
||||
LXVW4X (INP)(R9), VS61
|
||||
LXVW4X (INP)(R10), VS62
|
||||
|
||||
VXOR V27, V0, V27
|
||||
VXOR V28, V4, V28
|
||||
VXOR V29, V8, V29
|
||||
VXOR V30, V12, V30
|
||||
|
||||
STXVW4X VS59, (OUT)(R0)
|
||||
STXVW4X VS60, (OUT)(R8)
|
||||
ADD $64, INP
|
||||
STXVW4X VS61, (OUT)(R9)
|
||||
ADD $-64, LEN
|
||||
STXVW4X VS62, (OUT)(R10)
|
||||
ADD $64, OUT
|
||||
BEQ done_vsx
|
||||
|
||||
VADDUWM V1, V16, V0
|
||||
VADDUWM V5, V17, V4
|
||||
VADDUWM V9, V18, V8
|
||||
VADDUWM V13, V19, V12
|
||||
|
||||
CMPU LEN, $64
|
||||
BLT tail_vsx
|
||||
|
||||
LXVW4X (INP)(R0), VS59
|
||||
LXVW4X (INP)(R8), VS60
|
||||
LXVW4X (INP)(R9), VS61
|
||||
LXVW4X (INP)(R10), VS62
|
||||
VXOR V27, V0, V27
|
||||
|
||||
VXOR V28, V4, V28
|
||||
VXOR V29, V8, V29
|
||||
VXOR V30, V12, V30
|
||||
|
||||
STXVW4X VS59, (OUT)(R0)
|
||||
STXVW4X VS60, (OUT)(R8)
|
||||
ADD $64, INP
|
||||
STXVW4X VS61, (OUT)(R9)
|
||||
ADD $-64, LEN
|
||||
STXVW4X VS62, (OUT)(V10)
|
||||
ADD $64, OUT
|
||||
BEQ done_vsx
|
||||
|
||||
VADDUWM V2, V16, V0
|
||||
VADDUWM V6, V17, V4
|
||||
VADDUWM V10, V18, V8
|
||||
VADDUWM V14, V19, V12
|
||||
|
||||
CMPU LEN, $64
|
||||
BLT tail_vsx
|
||||
|
||||
LXVW4X (INP)(R0), VS59
|
||||
LXVW4X (INP)(R8), VS60
|
||||
LXVW4X (INP)(R9), VS61
|
||||
LXVW4X (INP)(R10), VS62
|
||||
|
||||
VXOR V27, V0, V27
|
||||
VXOR V28, V4, V28
|
||||
VXOR V29, V8, V29
|
||||
VXOR V30, V12, V30
|
||||
|
||||
STXVW4X VS59, (OUT)(R0)
|
||||
STXVW4X VS60, (OUT)(R8)
|
||||
ADD $64, INP
|
||||
STXVW4X VS61, (OUT)(R9)
|
||||
ADD $-64, LEN
|
||||
STXVW4X VS62, (OUT)(R10)
|
||||
ADD $64, OUT
|
||||
BEQ done_vsx
|
||||
|
||||
VADDUWM V3, V16, V0
|
||||
VADDUWM V7, V17, V4
|
||||
VADDUWM V11, V18, V8
|
||||
VADDUWM V15, V19, V12
|
||||
|
||||
CMPU LEN, $64
|
||||
BLT tail_vsx
|
||||
|
||||
LXVW4X (INP)(R0), VS59
|
||||
LXVW4X (INP)(R8), VS60
|
||||
LXVW4X (INP)(R9), VS61
|
||||
LXVW4X (INP)(R10), VS62
|
||||
|
||||
VXOR V27, V0, V27
|
||||
VXOR V28, V4, V28
|
||||
VXOR V29, V8, V29
|
||||
VXOR V30, V12, V30
|
||||
|
||||
STXVW4X VS59, (OUT)(R0)
|
||||
STXVW4X VS60, (OUT)(R8)
|
||||
ADD $64, INP
|
||||
STXVW4X VS61, (OUT)(R9)
|
||||
ADD $-64, LEN
|
||||
STXVW4X VS62, (OUT)(R10)
|
||||
ADD $64, OUT
|
||||
|
||||
MOVD $10, R14
|
||||
MOVD R14, CTR
|
||||
BNE loop_outer_vsx
|
||||
|
||||
done_vsx:
|
||||
// Increment counter by number of 64 byte blocks
|
||||
MOVD (CNT), R14
|
||||
ADD BLOCKS, R14
|
||||
MOVD R14, (CNT)
|
||||
RET
|
||||
|
||||
tail_vsx:
|
||||
ADD $32, R1, R11
|
||||
MOVD LEN, CTR
|
||||
|
||||
// Save values on stack to copy from
|
||||
STXVW4X VS32, (R11)(R0)
|
||||
STXVW4X VS36, (R11)(R8)
|
||||
STXVW4X VS40, (R11)(R9)
|
||||
STXVW4X VS44, (R11)(R10)
|
||||
ADD $-1, R11, R12
|
||||
ADD $-1, INP
|
||||
ADD $-1, OUT
|
||||
|
||||
looptail_vsx:
|
||||
// Copying the result to OUT
|
||||
// in bytes.
|
||||
MOVBZU 1(R12), KEY
|
||||
MOVBZU 1(INP), TMP
|
||||
XOR KEY, TMP, KEY
|
||||
MOVBU KEY, 1(OUT)
|
||||
BC 16, LT, looptail_vsx
|
||||
|
||||
// Clear the stack values
|
||||
STXVW4X VS48, (R11)(R0)
|
||||
STXVW4X VS48, (R11)(R8)
|
||||
STXVW4X VS48, (R11)(R9)
|
||||
STXVW4X VS48, (R11)(R10)
|
||||
BR done_vsx
|
||||
28
vendor/golang.org/x/crypto/chacha20/chacha_s390x.go
generated
vendored
Normal file
28
vendor/golang.org/x/crypto/chacha20/chacha_s390x.go
generated
vendored
Normal file
@@ -0,0 +1,28 @@
|
||||
// Copyright 2018 The Go Authors. All rights reserved.
|
||||
// Use of this source code is governed by a BSD-style
|
||||
// license that can be found in the LICENSE file.
|
||||
|
||||
//go:build gc && !purego
|
||||
// +build gc,!purego
|
||||
|
||||
package chacha20
|
||||
|
||||
import "golang.org/x/sys/cpu"
|
||||
|
||||
var haveAsm = cpu.S390X.HasVX
|
||||
|
||||
const bufSize = 256
|
||||
|
||||
// xorKeyStreamVX is an assembly implementation of XORKeyStream. It must only
|
||||
// be called when the vector facility is available. Implementation in asm_s390x.s.
|
||||
//
|
||||
//go:noescape
|
||||
func xorKeyStreamVX(dst, src []byte, key *[8]uint32, nonce *[3]uint32, counter *uint32)
|
||||
|
||||
func (c *Cipher) xorKeyStreamBlocks(dst, src []byte) {
|
||||
if cpu.S390X.HasVX {
|
||||
xorKeyStreamVX(dst, src, &c.key, &c.nonce, &c.counter)
|
||||
} else {
|
||||
c.xorKeyStreamBlocksGeneric(dst, src)
|
||||
}
|
||||
}
|
||||
225
vendor/golang.org/x/crypto/chacha20/chacha_s390x.s
generated
vendored
Normal file
225
vendor/golang.org/x/crypto/chacha20/chacha_s390x.s
generated
vendored
Normal file
@@ -0,0 +1,225 @@
|
||||
// Copyright 2018 The Go Authors. All rights reserved.
|
||||
// Use of this source code is governed by a BSD-style
|
||||
// license that can be found in the LICENSE file.
|
||||
|
||||
//go:build gc && !purego
|
||||
// +build gc,!purego
|
||||
|
||||
#include "go_asm.h"
|
||||
#include "textflag.h"
|
||||
|
||||
// This is an implementation of the ChaCha20 encryption algorithm as
|
||||
// specified in RFC 7539. It uses vector instructions to compute
|
||||
// 4 keystream blocks in parallel (256 bytes) which are then XORed
|
||||
// with the bytes in the input slice.
|
||||
|
||||
GLOBL ·constants<>(SB), RODATA|NOPTR, $32
|
||||
// BSWAP: swap bytes in each 4-byte element
|
||||
DATA ·constants<>+0x00(SB)/4, $0x03020100
|
||||
DATA ·constants<>+0x04(SB)/4, $0x07060504
|
||||
DATA ·constants<>+0x08(SB)/4, $0x0b0a0908
|
||||
DATA ·constants<>+0x0c(SB)/4, $0x0f0e0d0c
|
||||
// J0: [j0, j1, j2, j3]
|
||||
DATA ·constants<>+0x10(SB)/4, $0x61707865
|
||||
DATA ·constants<>+0x14(SB)/4, $0x3320646e
|
||||
DATA ·constants<>+0x18(SB)/4, $0x79622d32
|
||||
DATA ·constants<>+0x1c(SB)/4, $0x6b206574
|
||||
|
||||
#define BSWAP V5
|
||||
#define J0 V6
|
||||
#define KEY0 V7
|
||||
#define KEY1 V8
|
||||
#define NONCE V9
|
||||
#define CTR V10
|
||||
#define M0 V11
|
||||
#define M1 V12
|
||||
#define M2 V13
|
||||
#define M3 V14
|
||||
#define INC V15
|
||||
#define X0 V16
|
||||
#define X1 V17
|
||||
#define X2 V18
|
||||
#define X3 V19
|
||||
#define X4 V20
|
||||
#define X5 V21
|
||||
#define X6 V22
|
||||
#define X7 V23
|
||||
#define X8 V24
|
||||
#define X9 V25
|
||||
#define X10 V26
|
||||
#define X11 V27
|
||||
#define X12 V28
|
||||
#define X13 V29
|
||||
#define X14 V30
|
||||
#define X15 V31
|
||||
|
||||
#define NUM_ROUNDS 20
|
||||
|
||||
#define ROUND4(a0, a1, a2, a3, b0, b1, b2, b3, c0, c1, c2, c3, d0, d1, d2, d3) \
|
||||
VAF a1, a0, a0 \
|
||||
VAF b1, b0, b0 \
|
||||
VAF c1, c0, c0 \
|
||||
VAF d1, d0, d0 \
|
||||
VX a0, a2, a2 \
|
||||
VX b0, b2, b2 \
|
||||
VX c0, c2, c2 \
|
||||
VX d0, d2, d2 \
|
||||
VERLLF $16, a2, a2 \
|
||||
VERLLF $16, b2, b2 \
|
||||
VERLLF $16, c2, c2 \
|
||||
VERLLF $16, d2, d2 \
|
||||
VAF a2, a3, a3 \
|
||||
VAF b2, b3, b3 \
|
||||
VAF c2, c3, c3 \
|
||||
VAF d2, d3, d3 \
|
||||
VX a3, a1, a1 \
|
||||
VX b3, b1, b1 \
|
||||
VX c3, c1, c1 \
|
||||
VX d3, d1, d1 \
|
||||
VERLLF $12, a1, a1 \
|
||||
VERLLF $12, b1, b1 \
|
||||
VERLLF $12, c1, c1 \
|
||||
VERLLF $12, d1, d1 \
|
||||
VAF a1, a0, a0 \
|
||||
VAF b1, b0, b0 \
|
||||
VAF c1, c0, c0 \
|
||||
VAF d1, d0, d0 \
|
||||
VX a0, a2, a2 \
|
||||
VX b0, b2, b2 \
|
||||
VX c0, c2, c2 \
|
||||
VX d0, d2, d2 \
|
||||
VERLLF $8, a2, a2 \
|
||||
VERLLF $8, b2, b2 \
|
||||
VERLLF $8, c2, c2 \
|
||||
VERLLF $8, d2, d2 \
|
||||
VAF a2, a3, a3 \
|
||||
VAF b2, b3, b3 \
|
||||
VAF c2, c3, c3 \
|
||||
VAF d2, d3, d3 \
|
||||
VX a3, a1, a1 \
|
||||
VX b3, b1, b1 \
|
||||
VX c3, c1, c1 \
|
||||
VX d3, d1, d1 \
|
||||
VERLLF $7, a1, a1 \
|
||||
VERLLF $7, b1, b1 \
|
||||
VERLLF $7, c1, c1 \
|
||||
VERLLF $7, d1, d1
|
||||
|
||||
#define PERMUTE(mask, v0, v1, v2, v3) \
|
||||
VPERM v0, v0, mask, v0 \
|
||||
VPERM v1, v1, mask, v1 \
|
||||
VPERM v2, v2, mask, v2 \
|
||||
VPERM v3, v3, mask, v3
|
||||
|
||||
#define ADDV(x, v0, v1, v2, v3) \
|
||||
VAF x, v0, v0 \
|
||||
VAF x, v1, v1 \
|
||||
VAF x, v2, v2 \
|
||||
VAF x, v3, v3
|
||||
|
||||
#define XORV(off, dst, src, v0, v1, v2, v3) \
|
||||
VLM off(src), M0, M3 \
|
||||
PERMUTE(BSWAP, v0, v1, v2, v3) \
|
||||
VX v0, M0, M0 \
|
||||
VX v1, M1, M1 \
|
||||
VX v2, M2, M2 \
|
||||
VX v3, M3, M3 \
|
||||
VSTM M0, M3, off(dst)
|
||||
|
||||
#define SHUFFLE(a, b, c, d, t, u, v, w) \
|
||||
VMRHF a, c, t \ // t = {a[0], c[0], a[1], c[1]}
|
||||
VMRHF b, d, u \ // u = {b[0], d[0], b[1], d[1]}
|
||||
VMRLF a, c, v \ // v = {a[2], c[2], a[3], c[3]}
|
||||
VMRLF b, d, w \ // w = {b[2], d[2], b[3], d[3]}
|
||||
VMRHF t, u, a \ // a = {a[0], b[0], c[0], d[0]}
|
||||
VMRLF t, u, b \ // b = {a[1], b[1], c[1], d[1]}
|
||||
VMRHF v, w, c \ // c = {a[2], b[2], c[2], d[2]}
|
||||
VMRLF v, w, d // d = {a[3], b[3], c[3], d[3]}
|
||||
|
||||
// func xorKeyStreamVX(dst, src []byte, key *[8]uint32, nonce *[3]uint32, counter *uint32)
|
||||
TEXT ·xorKeyStreamVX(SB), NOSPLIT, $0
|
||||
MOVD $·constants<>(SB), R1
|
||||
MOVD dst+0(FP), R2 // R2=&dst[0]
|
||||
LMG src+24(FP), R3, R4 // R3=&src[0] R4=len(src)
|
||||
MOVD key+48(FP), R5 // R5=key
|
||||
MOVD nonce+56(FP), R6 // R6=nonce
|
||||
MOVD counter+64(FP), R7 // R7=counter
|
||||
|
||||
// load BSWAP and J0
|
||||
VLM (R1), BSWAP, J0
|
||||
|
||||
// setup
|
||||
MOVD $95, R0
|
||||
VLM (R5), KEY0, KEY1
|
||||
VLL R0, (R6), NONCE
|
||||
VZERO M0
|
||||
VLEIB $7, $32, M0
|
||||
VSRLB M0, NONCE, NONCE
|
||||
|
||||
// initialize counter values
|
||||
VLREPF (R7), CTR
|
||||
VZERO INC
|
||||
VLEIF $1, $1, INC
|
||||
VLEIF $2, $2, INC
|
||||
VLEIF $3, $3, INC
|
||||
VAF INC, CTR, CTR
|
||||
VREPIF $4, INC
|
||||
|
||||
chacha:
|
||||
VREPF $0, J0, X0
|
||||
VREPF $1, J0, X1
|
||||
VREPF $2, J0, X2
|
||||
VREPF $3, J0, X3
|
||||
VREPF $0, KEY0, X4
|
||||
VREPF $1, KEY0, X5
|
||||
VREPF $2, KEY0, X6
|
||||
VREPF $3, KEY0, X7
|
||||
VREPF $0, KEY1, X8
|
||||
VREPF $1, KEY1, X9
|
||||
VREPF $2, KEY1, X10
|
||||
VREPF $3, KEY1, X11
|
||||
VLR CTR, X12
|
||||
VREPF $1, NONCE, X13
|
||||
VREPF $2, NONCE, X14
|
||||
VREPF $3, NONCE, X15
|
||||
|
||||
MOVD $(NUM_ROUNDS/2), R1
|
||||
|
||||
loop:
|
||||
ROUND4(X0, X4, X12, X8, X1, X5, X13, X9, X2, X6, X14, X10, X3, X7, X15, X11)
|
||||
ROUND4(X0, X5, X15, X10, X1, X6, X12, X11, X2, X7, X13, X8, X3, X4, X14, X9)
|
||||
|
||||
ADD $-1, R1
|
||||
BNE loop
|
||||
|
||||
// decrement length
|
||||
ADD $-256, R4
|
||||
|
||||
// rearrange vectors
|
||||
SHUFFLE(X0, X1, X2, X3, M0, M1, M2, M3)
|
||||
ADDV(J0, X0, X1, X2, X3)
|
||||
SHUFFLE(X4, X5, X6, X7, M0, M1, M2, M3)
|
||||
ADDV(KEY0, X4, X5, X6, X7)
|
||||
SHUFFLE(X8, X9, X10, X11, M0, M1, M2, M3)
|
||||
ADDV(KEY1, X8, X9, X10, X11)
|
||||
VAF CTR, X12, X12
|
||||
SHUFFLE(X12, X13, X14, X15, M0, M1, M2, M3)
|
||||
ADDV(NONCE, X12, X13, X14, X15)
|
||||
|
||||
// increment counters
|
||||
VAF INC, CTR, CTR
|
||||
|
||||
// xor keystream with plaintext
|
||||
XORV(0*64, R2, R3, X0, X4, X8, X12)
|
||||
XORV(1*64, R2, R3, X1, X5, X9, X13)
|
||||
XORV(2*64, R2, R3, X2, X6, X10, X14)
|
||||
XORV(3*64, R2, R3, X3, X7, X11, X15)
|
||||
|
||||
// increment pointers
|
||||
MOVD $256(R2), R2
|
||||
MOVD $256(R3), R3
|
||||
|
||||
CMPBNE R4, $0, chacha
|
||||
|
||||
VSTEF $0, CTR, (R7)
|
||||
RET
|
||||
42
vendor/golang.org/x/crypto/chacha20/xor.go
generated
vendored
Normal file
42
vendor/golang.org/x/crypto/chacha20/xor.go
generated
vendored
Normal file
@@ -0,0 +1,42 @@
|
||||
// Copyright 2018 The Go Authors. All rights reserved.
|
||||
// Use of this source code is governed by a BSD-style
|
||||
// license that can be found src the LICENSE file.
|
||||
|
||||
package chacha20
|
||||
|
||||
import "runtime"
|
||||
|
||||
// Platforms that have fast unaligned 32-bit little endian accesses.
|
||||
const unaligned = runtime.GOARCH == "386" ||
|
||||
runtime.GOARCH == "amd64" ||
|
||||
runtime.GOARCH == "arm64" ||
|
||||
runtime.GOARCH == "ppc64le" ||
|
||||
runtime.GOARCH == "s390x"
|
||||
|
||||
// addXor reads a little endian uint32 from src, XORs it with (a + b) and
|
||||
// places the result in little endian byte order in dst.
|
||||
func addXor(dst, src []byte, a, b uint32) {
|
||||
_, _ = src[3], dst[3] // bounds check elimination hint
|
||||
if unaligned {
|
||||
// The compiler should optimize this code into
|
||||
// 32-bit unaligned little endian loads and stores.
|
||||
// TODO: delete once the compiler does a reliably
|
||||
// good job with the generic code below.
|
||||
// See issue #25111 for more details.
|
||||
v := uint32(src[0])
|
||||
v |= uint32(src[1]) << 8
|
||||
v |= uint32(src[2]) << 16
|
||||
v |= uint32(src[3]) << 24
|
||||
v ^= a + b
|
||||
dst[0] = byte(v)
|
||||
dst[1] = byte(v >> 8)
|
||||
dst[2] = byte(v >> 16)
|
||||
dst[3] = byte(v >> 24)
|
||||
} else {
|
||||
a += b
|
||||
dst[0] = src[0] ^ byte(a)
|
||||
dst[1] = src[1] ^ byte(a>>8)
|
||||
dst[2] = src[2] ^ byte(a>>16)
|
||||
dst[3] = src[3] ^ byte(a>>24)
|
||||
}
|
||||
}
|
||||
32
vendor/golang.org/x/crypto/internal/alias/alias.go
generated
vendored
Normal file
32
vendor/golang.org/x/crypto/internal/alias/alias.go
generated
vendored
Normal file
@@ -0,0 +1,32 @@
|
||||
// Copyright 2018 The Go Authors. All rights reserved.
|
||||
// Use of this source code is governed by a BSD-style
|
||||
// license that can be found in the LICENSE file.
|
||||
|
||||
//go:build !purego
|
||||
// +build !purego
|
||||
|
||||
// Package alias implements memory aliasing tests.
|
||||
package alias
|
||||
|
||||
import "unsafe"
|
||||
|
||||
// AnyOverlap reports whether x and y share memory at any (not necessarily
|
||||
// corresponding) index. The memory beyond the slice length is ignored.
|
||||
func AnyOverlap(x, y []byte) bool {
|
||||
return len(x) > 0 && len(y) > 0 &&
|
||||
uintptr(unsafe.Pointer(&x[0])) <= uintptr(unsafe.Pointer(&y[len(y)-1])) &&
|
||||
uintptr(unsafe.Pointer(&y[0])) <= uintptr(unsafe.Pointer(&x[len(x)-1]))
|
||||
}
|
||||
|
||||
// InexactOverlap reports whether x and y share memory at any non-corresponding
|
||||
// index. The memory beyond the slice length is ignored. Note that x and y can
|
||||
// have different lengths and still not have any inexact overlap.
|
||||
//
|
||||
// InexactOverlap can be used to implement the requirements of the crypto/cipher
|
||||
// AEAD, Block, BlockMode and Stream interfaces.
|
||||
func InexactOverlap(x, y []byte) bool {
|
||||
if len(x) == 0 || len(y) == 0 || &x[0] == &y[0] {
|
||||
return false
|
||||
}
|
||||
return AnyOverlap(x, y)
|
||||
}
|
||||
35
vendor/golang.org/x/crypto/internal/alias/alias_purego.go
generated
vendored
Normal file
35
vendor/golang.org/x/crypto/internal/alias/alias_purego.go
generated
vendored
Normal file
@@ -0,0 +1,35 @@
|
||||
// Copyright 2018 The Go Authors. All rights reserved.
|
||||
// Use of this source code is governed by a BSD-style
|
||||
// license that can be found in the LICENSE file.
|
||||
|
||||
//go:build purego
|
||||
// +build purego
|
||||
|
||||
// Package alias implements memory aliasing tests.
|
||||
package alias
|
||||
|
||||
// This is the Google App Engine standard variant based on reflect
|
||||
// because the unsafe package and cgo are disallowed.
|
||||
|
||||
import "reflect"
|
||||
|
||||
// AnyOverlap reports whether x and y share memory at any (not necessarily
|
||||
// corresponding) index. The memory beyond the slice length is ignored.
|
||||
func AnyOverlap(x, y []byte) bool {
|
||||
return len(x) > 0 && len(y) > 0 &&
|
||||
reflect.ValueOf(&x[0]).Pointer() <= reflect.ValueOf(&y[len(y)-1]).Pointer() &&
|
||||
reflect.ValueOf(&y[0]).Pointer() <= reflect.ValueOf(&x[len(x)-1]).Pointer()
|
||||
}
|
||||
|
||||
// InexactOverlap reports whether x and y share memory at any non-corresponding
|
||||
// index. The memory beyond the slice length is ignored. Note that x and y can
|
||||
// have different lengths and still not have any inexact overlap.
|
||||
//
|
||||
// InexactOverlap can be used to implement the requirements of the crypto/cipher
|
||||
// AEAD, Block, BlockMode and Stream interfaces.
|
||||
func InexactOverlap(x, y []byte) bool {
|
||||
if len(x) == 0 || len(y) == 0 || &x[0] == &y[0] {
|
||||
return false
|
||||
}
|
||||
return AnyOverlap(x, y)
|
||||
}
|
||||
40
vendor/golang.org/x/crypto/internal/poly1305/bits_compat.go
generated
vendored
Normal file
40
vendor/golang.org/x/crypto/internal/poly1305/bits_compat.go
generated
vendored
Normal file
@@ -0,0 +1,40 @@
|
||||
// Copyright 2019 The Go Authors. All rights reserved.
|
||||
// Use of this source code is governed by a BSD-style
|
||||
// license that can be found in the LICENSE file.
|
||||
|
||||
//go:build !go1.13
|
||||
// +build !go1.13
|
||||
|
||||
package poly1305
|
||||
|
||||
// Generic fallbacks for the math/bits intrinsics, copied from
|
||||
// src/math/bits/bits.go. They were added in Go 1.12, but Add64 and Sum64 had
|
||||
// variable time fallbacks until Go 1.13.
|
||||
|
||||
func bitsAdd64(x, y, carry uint64) (sum, carryOut uint64) {
|
||||
sum = x + y + carry
|
||||
carryOut = ((x & y) | ((x | y) &^ sum)) >> 63
|
||||
return
|
||||
}
|
||||
|
||||
func bitsSub64(x, y, borrow uint64) (diff, borrowOut uint64) {
|
||||
diff = x - y - borrow
|
||||
borrowOut = ((^x & y) | (^(x ^ y) & diff)) >> 63
|
||||
return
|
||||
}
|
||||
|
||||
func bitsMul64(x, y uint64) (hi, lo uint64) {
|
||||
const mask32 = 1<<32 - 1
|
||||
x0 := x & mask32
|
||||
x1 := x >> 32
|
||||
y0 := y & mask32
|
||||
y1 := y >> 32
|
||||
w0 := x0 * y0
|
||||
t := x1*y0 + w0>>32
|
||||
w1 := t & mask32
|
||||
w2 := t >> 32
|
||||
w1 += x0 * y1
|
||||
hi = x1*y1 + w2 + w1>>32
|
||||
lo = x * y
|
||||
return
|
||||
}
|
||||
22
vendor/golang.org/x/crypto/internal/poly1305/bits_go1.13.go
generated
vendored
Normal file
22
vendor/golang.org/x/crypto/internal/poly1305/bits_go1.13.go
generated
vendored
Normal file
@@ -0,0 +1,22 @@
|
||||
// Copyright 2019 The Go Authors. All rights reserved.
|
||||
// Use of this source code is governed by a BSD-style
|
||||
// license that can be found in the LICENSE file.
|
||||
|
||||
//go:build go1.13
|
||||
// +build go1.13
|
||||
|
||||
package poly1305
|
||||
|
||||
import "math/bits"
|
||||
|
||||
func bitsAdd64(x, y, carry uint64) (sum, carryOut uint64) {
|
||||
return bits.Add64(x, y, carry)
|
||||
}
|
||||
|
||||
func bitsSub64(x, y, borrow uint64) (diff, borrowOut uint64) {
|
||||
return bits.Sub64(x, y, borrow)
|
||||
}
|
||||
|
||||
func bitsMul64(x, y uint64) (hi, lo uint64) {
|
||||
return bits.Mul64(x, y)
|
||||
}
|
||||
10
vendor/golang.org/x/crypto/internal/poly1305/mac_noasm.go
generated
vendored
Normal file
10
vendor/golang.org/x/crypto/internal/poly1305/mac_noasm.go
generated
vendored
Normal file
@@ -0,0 +1,10 @@
|
||||
// Copyright 2018 The Go Authors. All rights reserved.
|
||||
// Use of this source code is governed by a BSD-style
|
||||
// license that can be found in the LICENSE file.
|
||||
|
||||
//go:build (!amd64 && !ppc64le && !s390x) || !gc || purego
|
||||
// +build !amd64,!ppc64le,!s390x !gc purego
|
||||
|
||||
package poly1305
|
||||
|
||||
type mac struct{ macGeneric }
|
||||
99
vendor/golang.org/x/crypto/internal/poly1305/poly1305.go
generated
vendored
Normal file
99
vendor/golang.org/x/crypto/internal/poly1305/poly1305.go
generated
vendored
Normal file
@@ -0,0 +1,99 @@
|
||||
// Copyright 2012 The Go Authors. All rights reserved.
|
||||
// Use of this source code is governed by a BSD-style
|
||||
// license that can be found in the LICENSE file.
|
||||
|
||||
// Package poly1305 implements Poly1305 one-time message authentication code as
|
||||
// specified in https://cr.yp.to/mac/poly1305-20050329.pdf.
|
||||
//
|
||||
// Poly1305 is a fast, one-time authentication function. It is infeasible for an
|
||||
// attacker to generate an authenticator for a message without the key. However, a
|
||||
// key must only be used for a single message. Authenticating two different
|
||||
// messages with the same key allows an attacker to forge authenticators for other
|
||||
// messages with the same key.
|
||||
//
|
||||
// Poly1305 was originally coupled with AES in order to make Poly1305-AES. AES was
|
||||
// used with a fixed key in order to generate one-time keys from an nonce.
|
||||
// However, in this package AES isn't used and the one-time key is specified
|
||||
// directly.
|
||||
package poly1305
|
||||
|
||||
import "crypto/subtle"
|
||||
|
||||
// TagSize is the size, in bytes, of a poly1305 authenticator.
|
||||
const TagSize = 16
|
||||
|
||||
// Sum generates an authenticator for msg using a one-time key and puts the
|
||||
// 16-byte result into out. Authenticating two different messages with the same
|
||||
// key allows an attacker to forge messages at will.
|
||||
func Sum(out *[16]byte, m []byte, key *[32]byte) {
|
||||
h := New(key)
|
||||
h.Write(m)
|
||||
h.Sum(out[:0])
|
||||
}
|
||||
|
||||
// Verify returns true if mac is a valid authenticator for m with the given key.
|
||||
func Verify(mac *[16]byte, m []byte, key *[32]byte) bool {
|
||||
var tmp [16]byte
|
||||
Sum(&tmp, m, key)
|
||||
return subtle.ConstantTimeCompare(tmp[:], mac[:]) == 1
|
||||
}
|
||||
|
||||
// New returns a new MAC computing an authentication
|
||||
// tag of all data written to it with the given key.
|
||||
// This allows writing the message progressively instead
|
||||
// of passing it as a single slice. Common users should use
|
||||
// the Sum function instead.
|
||||
//
|
||||
// The key must be unique for each message, as authenticating
|
||||
// two different messages with the same key allows an attacker
|
||||
// to forge messages at will.
|
||||
func New(key *[32]byte) *MAC {
|
||||
m := &MAC{}
|
||||
initialize(key, &m.macState)
|
||||
return m
|
||||
}
|
||||
|
||||
// MAC is an io.Writer computing an authentication tag
|
||||
// of the data written to it.
|
||||
//
|
||||
// MAC cannot be used like common hash.Hash implementations,
|
||||
// because using a poly1305 key twice breaks its security.
|
||||
// Therefore writing data to a running MAC after calling
|
||||
// Sum or Verify causes it to panic.
|
||||
type MAC struct {
|
||||
mac // platform-dependent implementation
|
||||
|
||||
finalized bool
|
||||
}
|
||||
|
||||
// Size returns the number of bytes Sum will return.
|
||||
func (h *MAC) Size() int { return TagSize }
|
||||
|
||||
// Write adds more data to the running message authentication code.
|
||||
// It never returns an error.
|
||||
//
|
||||
// It must not be called after the first call of Sum or Verify.
|
||||
func (h *MAC) Write(p []byte) (n int, err error) {
|
||||
if h.finalized {
|
||||
panic("poly1305: write to MAC after Sum or Verify")
|
||||
}
|
||||
return h.mac.Write(p)
|
||||
}
|
||||
|
||||
// Sum computes the authenticator of all data written to the
|
||||
// message authentication code.
|
||||
func (h *MAC) Sum(b []byte) []byte {
|
||||
var mac [TagSize]byte
|
||||
h.mac.Sum(&mac)
|
||||
h.finalized = true
|
||||
return append(b, mac[:]...)
|
||||
}
|
||||
|
||||
// Verify returns whether the authenticator of all data written to
|
||||
// the message authentication code matches the expected value.
|
||||
func (h *MAC) Verify(expected []byte) bool {
|
||||
var mac [TagSize]byte
|
||||
h.mac.Sum(&mac)
|
||||
h.finalized = true
|
||||
return subtle.ConstantTimeCompare(expected, mac[:]) == 1
|
||||
}
|
||||
48
vendor/golang.org/x/crypto/internal/poly1305/sum_amd64.go
generated
vendored
Normal file
48
vendor/golang.org/x/crypto/internal/poly1305/sum_amd64.go
generated
vendored
Normal file
@@ -0,0 +1,48 @@
|
||||
// Copyright 2012 The Go Authors. All rights reserved.
|
||||
// Use of this source code is governed by a BSD-style
|
||||
// license that can be found in the LICENSE file.
|
||||
|
||||
//go:build gc && !purego
|
||||
// +build gc,!purego
|
||||
|
||||
package poly1305
|
||||
|
||||
//go:noescape
|
||||
func update(state *macState, msg []byte)
|
||||
|
||||
// mac is a wrapper for macGeneric that redirects calls that would have gone to
|
||||
// updateGeneric to update.
|
||||
//
|
||||
// Its Write and Sum methods are otherwise identical to the macGeneric ones, but
|
||||
// using function pointers would carry a major performance cost.
|
||||
type mac struct{ macGeneric }
|
||||
|
||||
func (h *mac) Write(p []byte) (int, error) {
|
||||
nn := len(p)
|
||||
if h.offset > 0 {
|
||||
n := copy(h.buffer[h.offset:], p)
|
||||
if h.offset+n < TagSize {
|
||||
h.offset += n
|
||||
return nn, nil
|
||||
}
|
||||
p = p[n:]
|
||||
h.offset = 0
|
||||
update(&h.macState, h.buffer[:])
|
||||
}
|
||||
if n := len(p) - (len(p) % TagSize); n > 0 {
|
||||
update(&h.macState, p[:n])
|
||||
p = p[n:]
|
||||
}
|
||||
if len(p) > 0 {
|
||||
h.offset += copy(h.buffer[h.offset:], p)
|
||||
}
|
||||
return nn, nil
|
||||
}
|
||||
|
||||
func (h *mac) Sum(out *[16]byte) {
|
||||
state := h.macState
|
||||
if h.offset > 0 {
|
||||
update(&state, h.buffer[:h.offset])
|
||||
}
|
||||
finalize(out, &state.h, &state.s)
|
||||
}
|
||||
109
vendor/golang.org/x/crypto/internal/poly1305/sum_amd64.s
generated
vendored
Normal file
109
vendor/golang.org/x/crypto/internal/poly1305/sum_amd64.s
generated
vendored
Normal file
@@ -0,0 +1,109 @@
|
||||
// Copyright 2012 The Go Authors. All rights reserved.
|
||||
// Use of this source code is governed by a BSD-style
|
||||
// license that can be found in the LICENSE file.
|
||||
|
||||
//go:build gc && !purego
|
||||
// +build gc,!purego
|
||||
|
||||
#include "textflag.h"
|
||||
|
||||
#define POLY1305_ADD(msg, h0, h1, h2) \
|
||||
ADDQ 0(msg), h0; \
|
||||
ADCQ 8(msg), h1; \
|
||||
ADCQ $1, h2; \
|
||||
LEAQ 16(msg), msg
|
||||
|
||||
#define POLY1305_MUL(h0, h1, h2, r0, r1, t0, t1, t2, t3) \
|
||||
MOVQ r0, AX; \
|
||||
MULQ h0; \
|
||||
MOVQ AX, t0; \
|
||||
MOVQ DX, t1; \
|
||||
MOVQ r0, AX; \
|
||||
MULQ h1; \
|
||||
ADDQ AX, t1; \
|
||||
ADCQ $0, DX; \
|
||||
MOVQ r0, t2; \
|
||||
IMULQ h2, t2; \
|
||||
ADDQ DX, t2; \
|
||||
\
|
||||
MOVQ r1, AX; \
|
||||
MULQ h0; \
|
||||
ADDQ AX, t1; \
|
||||
ADCQ $0, DX; \
|
||||
MOVQ DX, h0; \
|
||||
MOVQ r1, t3; \
|
||||
IMULQ h2, t3; \
|
||||
MOVQ r1, AX; \
|
||||
MULQ h1; \
|
||||
ADDQ AX, t2; \
|
||||
ADCQ DX, t3; \
|
||||
ADDQ h0, t2; \
|
||||
ADCQ $0, t3; \
|
||||
\
|
||||
MOVQ t0, h0; \
|
||||
MOVQ t1, h1; \
|
||||
MOVQ t2, h2; \
|
||||
ANDQ $3, h2; \
|
||||
MOVQ t2, t0; \
|
||||
ANDQ $0xFFFFFFFFFFFFFFFC, t0; \
|
||||
ADDQ t0, h0; \
|
||||
ADCQ t3, h1; \
|
||||
ADCQ $0, h2; \
|
||||
SHRQ $2, t3, t2; \
|
||||
SHRQ $2, t3; \
|
||||
ADDQ t2, h0; \
|
||||
ADCQ t3, h1; \
|
||||
ADCQ $0, h2
|
||||
|
||||
// func update(state *[7]uint64, msg []byte)
|
||||
TEXT ·update(SB), $0-32
|
||||
MOVQ state+0(FP), DI
|
||||
MOVQ msg_base+8(FP), SI
|
||||
MOVQ msg_len+16(FP), R15
|
||||
|
||||
MOVQ 0(DI), R8 // h0
|
||||
MOVQ 8(DI), R9 // h1
|
||||
MOVQ 16(DI), R10 // h2
|
||||
MOVQ 24(DI), R11 // r0
|
||||
MOVQ 32(DI), R12 // r1
|
||||
|
||||
CMPQ R15, $16
|
||||
JB bytes_between_0_and_15
|
||||
|
||||
loop:
|
||||
POLY1305_ADD(SI, R8, R9, R10)
|
||||
|
||||
multiply:
|
||||
POLY1305_MUL(R8, R9, R10, R11, R12, BX, CX, R13, R14)
|
||||
SUBQ $16, R15
|
||||
CMPQ R15, $16
|
||||
JAE loop
|
||||
|
||||
bytes_between_0_and_15:
|
||||
TESTQ R15, R15
|
||||
JZ done
|
||||
MOVQ $1, BX
|
||||
XORQ CX, CX
|
||||
XORQ R13, R13
|
||||
ADDQ R15, SI
|
||||
|
||||
flush_buffer:
|
||||
SHLQ $8, BX, CX
|
||||
SHLQ $8, BX
|
||||
MOVB -1(SI), R13
|
||||
XORQ R13, BX
|
||||
DECQ SI
|
||||
DECQ R15
|
||||
JNZ flush_buffer
|
||||
|
||||
ADDQ BX, R8
|
||||
ADCQ CX, R9
|
||||
ADCQ $0, R10
|
||||
MOVQ $16, R15
|
||||
JMP multiply
|
||||
|
||||
done:
|
||||
MOVQ R8, 0(DI)
|
||||
MOVQ R9, 8(DI)
|
||||
MOVQ R10, 16(DI)
|
||||
RET
|
||||
309
vendor/golang.org/x/crypto/internal/poly1305/sum_generic.go
generated
vendored
Normal file
309
vendor/golang.org/x/crypto/internal/poly1305/sum_generic.go
generated
vendored
Normal file
@@ -0,0 +1,309 @@
|
||||
// Copyright 2018 The Go Authors. All rights reserved.
|
||||
// Use of this source code is governed by a BSD-style
|
||||
// license that can be found in the LICENSE file.
|
||||
|
||||
// This file provides the generic implementation of Sum and MAC. Other files
|
||||
// might provide optimized assembly implementations of some of this code.
|
||||
|
||||
package poly1305
|
||||
|
||||
import "encoding/binary"
|
||||
|
||||
// Poly1305 [RFC 7539] is a relatively simple algorithm: the authentication tag
|
||||
// for a 64 bytes message is approximately
|
||||
//
|
||||
// s + m[0:16] * r⁴ + m[16:32] * r³ + m[32:48] * r² + m[48:64] * r mod 2¹³⁰ - 5
|
||||
//
|
||||
// for some secret r and s. It can be computed sequentially like
|
||||
//
|
||||
// for len(msg) > 0:
|
||||
// h += read(msg, 16)
|
||||
// h *= r
|
||||
// h %= 2¹³⁰ - 5
|
||||
// return h + s
|
||||
//
|
||||
// All the complexity is about doing performant constant-time math on numbers
|
||||
// larger than any available numeric type.
|
||||
|
||||
func sumGeneric(out *[TagSize]byte, msg []byte, key *[32]byte) {
|
||||
h := newMACGeneric(key)
|
||||
h.Write(msg)
|
||||
h.Sum(out)
|
||||
}
|
||||
|
||||
func newMACGeneric(key *[32]byte) macGeneric {
|
||||
m := macGeneric{}
|
||||
initialize(key, &m.macState)
|
||||
return m
|
||||
}
|
||||
|
||||
// macState holds numbers in saturated 64-bit little-endian limbs. That is,
|
||||
// the value of [x0, x1, x2] is x[0] + x[1] * 2⁶⁴ + x[2] * 2¹²⁸.
|
||||
type macState struct {
|
||||
// h is the main accumulator. It is to be interpreted modulo 2¹³⁰ - 5, but
|
||||
// can grow larger during and after rounds. It must, however, remain below
|
||||
// 2 * (2¹³⁰ - 5).
|
||||
h [3]uint64
|
||||
// r and s are the private key components.
|
||||
r [2]uint64
|
||||
s [2]uint64
|
||||
}
|
||||
|
||||
type macGeneric struct {
|
||||
macState
|
||||
|
||||
buffer [TagSize]byte
|
||||
offset int
|
||||
}
|
||||
|
||||
// Write splits the incoming message into TagSize chunks, and passes them to
|
||||
// update. It buffers incomplete chunks.
|
||||
func (h *macGeneric) Write(p []byte) (int, error) {
|
||||
nn := len(p)
|
||||
if h.offset > 0 {
|
||||
n := copy(h.buffer[h.offset:], p)
|
||||
if h.offset+n < TagSize {
|
||||
h.offset += n
|
||||
return nn, nil
|
||||
}
|
||||
p = p[n:]
|
||||
h.offset = 0
|
||||
updateGeneric(&h.macState, h.buffer[:])
|
||||
}
|
||||
if n := len(p) - (len(p) % TagSize); n > 0 {
|
||||
updateGeneric(&h.macState, p[:n])
|
||||
p = p[n:]
|
||||
}
|
||||
if len(p) > 0 {
|
||||
h.offset += copy(h.buffer[h.offset:], p)
|
||||
}
|
||||
return nn, nil
|
||||
}
|
||||
|
||||
// Sum flushes the last incomplete chunk from the buffer, if any, and generates
|
||||
// the MAC output. It does not modify its state, in order to allow for multiple
|
||||
// calls to Sum, even if no Write is allowed after Sum.
|
||||
func (h *macGeneric) Sum(out *[TagSize]byte) {
|
||||
state := h.macState
|
||||
if h.offset > 0 {
|
||||
updateGeneric(&state, h.buffer[:h.offset])
|
||||
}
|
||||
finalize(out, &state.h, &state.s)
|
||||
}
|
||||
|
||||
// [rMask0, rMask1] is the specified Poly1305 clamping mask in little-endian. It
|
||||
// clears some bits of the secret coefficient to make it possible to implement
|
||||
// multiplication more efficiently.
|
||||
const (
|
||||
rMask0 = 0x0FFFFFFC0FFFFFFF
|
||||
rMask1 = 0x0FFFFFFC0FFFFFFC
|
||||
)
|
||||
|
||||
// initialize loads the 256-bit key into the two 128-bit secret values r and s.
|
||||
func initialize(key *[32]byte, m *macState) {
|
||||
m.r[0] = binary.LittleEndian.Uint64(key[0:8]) & rMask0
|
||||
m.r[1] = binary.LittleEndian.Uint64(key[8:16]) & rMask1
|
||||
m.s[0] = binary.LittleEndian.Uint64(key[16:24])
|
||||
m.s[1] = binary.LittleEndian.Uint64(key[24:32])
|
||||
}
|
||||
|
||||
// uint128 holds a 128-bit number as two 64-bit limbs, for use with the
|
||||
// bits.Mul64 and bits.Add64 intrinsics.
|
||||
type uint128 struct {
|
||||
lo, hi uint64
|
||||
}
|
||||
|
||||
func mul64(a, b uint64) uint128 {
|
||||
hi, lo := bitsMul64(a, b)
|
||||
return uint128{lo, hi}
|
||||
}
|
||||
|
||||
func add128(a, b uint128) uint128 {
|
||||
lo, c := bitsAdd64(a.lo, b.lo, 0)
|
||||
hi, c := bitsAdd64(a.hi, b.hi, c)
|
||||
if c != 0 {
|
||||
panic("poly1305: unexpected overflow")
|
||||
}
|
||||
return uint128{lo, hi}
|
||||
}
|
||||
|
||||
func shiftRightBy2(a uint128) uint128 {
|
||||
a.lo = a.lo>>2 | (a.hi&3)<<62
|
||||
a.hi = a.hi >> 2
|
||||
return a
|
||||
}
|
||||
|
||||
// updateGeneric absorbs msg into the state.h accumulator. For each chunk m of
|
||||
// 128 bits of message, it computes
|
||||
//
|
||||
// h₊ = (h + m) * r mod 2¹³⁰ - 5
|
||||
//
|
||||
// If the msg length is not a multiple of TagSize, it assumes the last
|
||||
// incomplete chunk is the final one.
|
||||
func updateGeneric(state *macState, msg []byte) {
|
||||
h0, h1, h2 := state.h[0], state.h[1], state.h[2]
|
||||
r0, r1 := state.r[0], state.r[1]
|
||||
|
||||
for len(msg) > 0 {
|
||||
var c uint64
|
||||
|
||||
// For the first step, h + m, we use a chain of bits.Add64 intrinsics.
|
||||
// The resulting value of h might exceed 2¹³⁰ - 5, but will be partially
|
||||
// reduced at the end of the multiplication below.
|
||||
//
|
||||
// The spec requires us to set a bit just above the message size, not to
|
||||
// hide leading zeroes. For full chunks, that's 1 << 128, so we can just
|
||||
// add 1 to the most significant (2¹²⁸) limb, h2.
|
||||
if len(msg) >= TagSize {
|
||||
h0, c = bitsAdd64(h0, binary.LittleEndian.Uint64(msg[0:8]), 0)
|
||||
h1, c = bitsAdd64(h1, binary.LittleEndian.Uint64(msg[8:16]), c)
|
||||
h2 += c + 1
|
||||
|
||||
msg = msg[TagSize:]
|
||||
} else {
|
||||
var buf [TagSize]byte
|
||||
copy(buf[:], msg)
|
||||
buf[len(msg)] = 1
|
||||
|
||||
h0, c = bitsAdd64(h0, binary.LittleEndian.Uint64(buf[0:8]), 0)
|
||||
h1, c = bitsAdd64(h1, binary.LittleEndian.Uint64(buf[8:16]), c)
|
||||
h2 += c
|
||||
|
||||
msg = nil
|
||||
}
|
||||
|
||||
// Multiplication of big number limbs is similar to elementary school
|
||||
// columnar multiplication. Instead of digits, there are 64-bit limbs.
|
||||
//
|
||||
// We are multiplying a 3 limbs number, h, by a 2 limbs number, r.
|
||||
//
|
||||
// h2 h1 h0 x
|
||||
// r1 r0 =
|
||||
// ----------------
|
||||
// h2r0 h1r0 h0r0 <-- individual 128-bit products
|
||||
// + h2r1 h1r1 h0r1
|
||||
// ------------------------
|
||||
// m3 m2 m1 m0 <-- result in 128-bit overlapping limbs
|
||||
// ------------------------
|
||||
// m3.hi m2.hi m1.hi m0.hi <-- carry propagation
|
||||
// + m3.lo m2.lo m1.lo m0.lo
|
||||
// -------------------------------
|
||||
// t4 t3 t2 t1 t0 <-- final result in 64-bit limbs
|
||||
//
|
||||
// The main difference from pen-and-paper multiplication is that we do
|
||||
// carry propagation in a separate step, as if we wrote two digit sums
|
||||
// at first (the 128-bit limbs), and then carried the tens all at once.
|
||||
|
||||
h0r0 := mul64(h0, r0)
|
||||
h1r0 := mul64(h1, r0)
|
||||
h2r0 := mul64(h2, r0)
|
||||
h0r1 := mul64(h0, r1)
|
||||
h1r1 := mul64(h1, r1)
|
||||
h2r1 := mul64(h2, r1)
|
||||
|
||||
// Since h2 is known to be at most 7 (5 + 1 + 1), and r0 and r1 have their
|
||||
// top 4 bits cleared by rMask{0,1}, we know that their product is not going
|
||||
// to overflow 64 bits, so we can ignore the high part of the products.
|
||||
//
|
||||
// This also means that the product doesn't have a fifth limb (t4).
|
||||
if h2r0.hi != 0 {
|
||||
panic("poly1305: unexpected overflow")
|
||||
}
|
||||
if h2r1.hi != 0 {
|
||||
panic("poly1305: unexpected overflow")
|
||||
}
|
||||
|
||||
m0 := h0r0
|
||||
m1 := add128(h1r0, h0r1) // These two additions don't overflow thanks again
|
||||
m2 := add128(h2r0, h1r1) // to the 4 masked bits at the top of r0 and r1.
|
||||
m3 := h2r1
|
||||
|
||||
t0 := m0.lo
|
||||
t1, c := bitsAdd64(m1.lo, m0.hi, 0)
|
||||
t2, c := bitsAdd64(m2.lo, m1.hi, c)
|
||||
t3, _ := bitsAdd64(m3.lo, m2.hi, c)
|
||||
|
||||
// Now we have the result as 4 64-bit limbs, and we need to reduce it
|
||||
// modulo 2¹³⁰ - 5. The special shape of this Crandall prime lets us do
|
||||
// a cheap partial reduction according to the reduction identity
|
||||
//
|
||||
// c * 2¹³⁰ + n = c * 5 + n mod 2¹³⁰ - 5
|
||||
//
|
||||
// because 2¹³⁰ = 5 mod 2¹³⁰ - 5. Partial reduction since the result is
|
||||
// likely to be larger than 2¹³⁰ - 5, but still small enough to fit the
|
||||
// assumptions we make about h in the rest of the code.
|
||||
//
|
||||
// See also https://speakerdeck.com/gtank/engineering-prime-numbers?slide=23
|
||||
|
||||
// We split the final result at the 2¹³⁰ mark into h and cc, the carry.
|
||||
// Note that the carry bits are effectively shifted left by 2, in other
|
||||
// words, cc = c * 4 for the c in the reduction identity.
|
||||
h0, h1, h2 = t0, t1, t2&maskLow2Bits
|
||||
cc := uint128{t2 & maskNotLow2Bits, t3}
|
||||
|
||||
// To add c * 5 to h, we first add cc = c * 4, and then add (cc >> 2) = c.
|
||||
|
||||
h0, c = bitsAdd64(h0, cc.lo, 0)
|
||||
h1, c = bitsAdd64(h1, cc.hi, c)
|
||||
h2 += c
|
||||
|
||||
cc = shiftRightBy2(cc)
|
||||
|
||||
h0, c = bitsAdd64(h0, cc.lo, 0)
|
||||
h1, c = bitsAdd64(h1, cc.hi, c)
|
||||
h2 += c
|
||||
|
||||
// h2 is at most 3 + 1 + 1 = 5, making the whole of h at most
|
||||
//
|
||||
// 5 * 2¹²⁸ + (2¹²⁸ - 1) = 6 * 2¹²⁸ - 1
|
||||
}
|
||||
|
||||
state.h[0], state.h[1], state.h[2] = h0, h1, h2
|
||||
}
|
||||
|
||||
const (
|
||||
maskLow2Bits uint64 = 0x0000000000000003
|
||||
maskNotLow2Bits uint64 = ^maskLow2Bits
|
||||
)
|
||||
|
||||
// select64 returns x if v == 1 and y if v == 0, in constant time.
|
||||
func select64(v, x, y uint64) uint64 { return ^(v-1)&x | (v-1)&y }
|
||||
|
||||
// [p0, p1, p2] is 2¹³⁰ - 5 in little endian order.
|
||||
const (
|
||||
p0 = 0xFFFFFFFFFFFFFFFB
|
||||
p1 = 0xFFFFFFFFFFFFFFFF
|
||||
p2 = 0x0000000000000003
|
||||
)
|
||||
|
||||
// finalize completes the modular reduction of h and computes
|
||||
//
|
||||
// out = h + s mod 2¹²⁸
|
||||
func finalize(out *[TagSize]byte, h *[3]uint64, s *[2]uint64) {
|
||||
h0, h1, h2 := h[0], h[1], h[2]
|
||||
|
||||
// After the partial reduction in updateGeneric, h might be more than
|
||||
// 2¹³⁰ - 5, but will be less than 2 * (2¹³⁰ - 5). To complete the reduction
|
||||
// in constant time, we compute t = h - (2¹³⁰ - 5), and select h as the
|
||||
// result if the subtraction underflows, and t otherwise.
|
||||
|
||||
hMinusP0, b := bitsSub64(h0, p0, 0)
|
||||
hMinusP1, b := bitsSub64(h1, p1, b)
|
||||
_, b = bitsSub64(h2, p2, b)
|
||||
|
||||
// h = h if h < p else h - p
|
||||
h0 = select64(b, h0, hMinusP0)
|
||||
h1 = select64(b, h1, hMinusP1)
|
||||
|
||||
// Finally, we compute the last Poly1305 step
|
||||
//
|
||||
// tag = h + s mod 2¹²⁸
|
||||
//
|
||||
// by just doing a wide addition with the 128 low bits of h and discarding
|
||||
// the overflow.
|
||||
h0, c := bitsAdd64(h0, s[0], 0)
|
||||
h1, _ = bitsAdd64(h1, s[1], c)
|
||||
|
||||
binary.LittleEndian.PutUint64(out[0:8], h0)
|
||||
binary.LittleEndian.PutUint64(out[8:16], h1)
|
||||
}
|
||||
48
vendor/golang.org/x/crypto/internal/poly1305/sum_ppc64le.go
generated
vendored
Normal file
48
vendor/golang.org/x/crypto/internal/poly1305/sum_ppc64le.go
generated
vendored
Normal file
@@ -0,0 +1,48 @@
|
||||
// Copyright 2019 The Go Authors. All rights reserved.
|
||||
// Use of this source code is governed by a BSD-style
|
||||
// license that can be found in the LICENSE file.
|
||||
|
||||
//go:build gc && !purego
|
||||
// +build gc,!purego
|
||||
|
||||
package poly1305
|
||||
|
||||
//go:noescape
|
||||
func update(state *macState, msg []byte)
|
||||
|
||||
// mac is a wrapper for macGeneric that redirects calls that would have gone to
|
||||
// updateGeneric to update.
|
||||
//
|
||||
// Its Write and Sum methods are otherwise identical to the macGeneric ones, but
|
||||
// using function pointers would carry a major performance cost.
|
||||
type mac struct{ macGeneric }
|
||||
|
||||
func (h *mac) Write(p []byte) (int, error) {
|
||||
nn := len(p)
|
||||
if h.offset > 0 {
|
||||
n := copy(h.buffer[h.offset:], p)
|
||||
if h.offset+n < TagSize {
|
||||
h.offset += n
|
||||
return nn, nil
|
||||
}
|
||||
p = p[n:]
|
||||
h.offset = 0
|
||||
update(&h.macState, h.buffer[:])
|
||||
}
|
||||
if n := len(p) - (len(p) % TagSize); n > 0 {
|
||||
update(&h.macState, p[:n])
|
||||
p = p[n:]
|
||||
}
|
||||
if len(p) > 0 {
|
||||
h.offset += copy(h.buffer[h.offset:], p)
|
||||
}
|
||||
return nn, nil
|
||||
}
|
||||
|
||||
func (h *mac) Sum(out *[16]byte) {
|
||||
state := h.macState
|
||||
if h.offset > 0 {
|
||||
update(&state, h.buffer[:h.offset])
|
||||
}
|
||||
finalize(out, &state.h, &state.s)
|
||||
}
|
||||
182
vendor/golang.org/x/crypto/internal/poly1305/sum_ppc64le.s
generated
vendored
Normal file
182
vendor/golang.org/x/crypto/internal/poly1305/sum_ppc64le.s
generated
vendored
Normal file
@@ -0,0 +1,182 @@
|
||||
// Copyright 2019 The Go Authors. All rights reserved.
|
||||
// Use of this source code is governed by a BSD-style
|
||||
// license that can be found in the LICENSE file.
|
||||
|
||||
//go:build gc && !purego
|
||||
// +build gc,!purego
|
||||
|
||||
#include "textflag.h"
|
||||
|
||||
// This was ported from the amd64 implementation.
|
||||
|
||||
#define POLY1305_ADD(msg, h0, h1, h2, t0, t1, t2) \
|
||||
MOVD (msg), t0; \
|
||||
MOVD 8(msg), t1; \
|
||||
MOVD $1, t2; \
|
||||
ADDC t0, h0, h0; \
|
||||
ADDE t1, h1, h1; \
|
||||
ADDE t2, h2; \
|
||||
ADD $16, msg
|
||||
|
||||
#define POLY1305_MUL(h0, h1, h2, r0, r1, t0, t1, t2, t3, t4, t5) \
|
||||
MULLD r0, h0, t0; \
|
||||
MULLD r0, h1, t4; \
|
||||
MULHDU r0, h0, t1; \
|
||||
MULHDU r0, h1, t5; \
|
||||
ADDC t4, t1, t1; \
|
||||
MULLD r0, h2, t2; \
|
||||
ADDZE t5; \
|
||||
MULHDU r1, h0, t4; \
|
||||
MULLD r1, h0, h0; \
|
||||
ADD t5, t2, t2; \
|
||||
ADDC h0, t1, t1; \
|
||||
MULLD h2, r1, t3; \
|
||||
ADDZE t4, h0; \
|
||||
MULHDU r1, h1, t5; \
|
||||
MULLD r1, h1, t4; \
|
||||
ADDC t4, t2, t2; \
|
||||
ADDE t5, t3, t3; \
|
||||
ADDC h0, t2, t2; \
|
||||
MOVD $-4, t4; \
|
||||
MOVD t0, h0; \
|
||||
MOVD t1, h1; \
|
||||
ADDZE t3; \
|
||||
ANDCC $3, t2, h2; \
|
||||
AND t2, t4, t0; \
|
||||
ADDC t0, h0, h0; \
|
||||
ADDE t3, h1, h1; \
|
||||
SLD $62, t3, t4; \
|
||||
SRD $2, t2; \
|
||||
ADDZE h2; \
|
||||
OR t4, t2, t2; \
|
||||
SRD $2, t3; \
|
||||
ADDC t2, h0, h0; \
|
||||
ADDE t3, h1, h1; \
|
||||
ADDZE h2
|
||||
|
||||
DATA ·poly1305Mask<>+0x00(SB)/8, $0x0FFFFFFC0FFFFFFF
|
||||
DATA ·poly1305Mask<>+0x08(SB)/8, $0x0FFFFFFC0FFFFFFC
|
||||
GLOBL ·poly1305Mask<>(SB), RODATA, $16
|
||||
|
||||
// func update(state *[7]uint64, msg []byte)
|
||||
TEXT ·update(SB), $0-32
|
||||
MOVD state+0(FP), R3
|
||||
MOVD msg_base+8(FP), R4
|
||||
MOVD msg_len+16(FP), R5
|
||||
|
||||
MOVD 0(R3), R8 // h0
|
||||
MOVD 8(R3), R9 // h1
|
||||
MOVD 16(R3), R10 // h2
|
||||
MOVD 24(R3), R11 // r0
|
||||
MOVD 32(R3), R12 // r1
|
||||
|
||||
CMP R5, $16
|
||||
BLT bytes_between_0_and_15
|
||||
|
||||
loop:
|
||||
POLY1305_ADD(R4, R8, R9, R10, R20, R21, R22)
|
||||
|
||||
multiply:
|
||||
POLY1305_MUL(R8, R9, R10, R11, R12, R16, R17, R18, R14, R20, R21)
|
||||
ADD $-16, R5
|
||||
CMP R5, $16
|
||||
BGE loop
|
||||
|
||||
bytes_between_0_and_15:
|
||||
CMP R5, $0
|
||||
BEQ done
|
||||
MOVD $0, R16 // h0
|
||||
MOVD $0, R17 // h1
|
||||
|
||||
flush_buffer:
|
||||
CMP R5, $8
|
||||
BLE just1
|
||||
|
||||
MOVD $8, R21
|
||||
SUB R21, R5, R21
|
||||
|
||||
// Greater than 8 -- load the rightmost remaining bytes in msg
|
||||
// and put into R17 (h1)
|
||||
MOVD (R4)(R21), R17
|
||||
MOVD $16, R22
|
||||
|
||||
// Find the offset to those bytes
|
||||
SUB R5, R22, R22
|
||||
SLD $3, R22
|
||||
|
||||
// Shift to get only the bytes in msg
|
||||
SRD R22, R17, R17
|
||||
|
||||
// Put 1 at high end
|
||||
MOVD $1, R23
|
||||
SLD $3, R21
|
||||
SLD R21, R23, R23
|
||||
OR R23, R17, R17
|
||||
|
||||
// Remainder is 8
|
||||
MOVD $8, R5
|
||||
|
||||
just1:
|
||||
CMP R5, $8
|
||||
BLT less8
|
||||
|
||||
// Exactly 8
|
||||
MOVD (R4), R16
|
||||
|
||||
CMP R17, $0
|
||||
|
||||
// Check if we've already set R17; if not
|
||||
// set 1 to indicate end of msg.
|
||||
BNE carry
|
||||
MOVD $1, R17
|
||||
BR carry
|
||||
|
||||
less8:
|
||||
MOVD $0, R16 // h0
|
||||
MOVD $0, R22 // shift count
|
||||
CMP R5, $4
|
||||
BLT less4
|
||||
MOVWZ (R4), R16
|
||||
ADD $4, R4
|
||||
ADD $-4, R5
|
||||
MOVD $32, R22
|
||||
|
||||
less4:
|
||||
CMP R5, $2
|
||||
BLT less2
|
||||
MOVHZ (R4), R21
|
||||
SLD R22, R21, R21
|
||||
OR R16, R21, R16
|
||||
ADD $16, R22
|
||||
ADD $-2, R5
|
||||
ADD $2, R4
|
||||
|
||||
less2:
|
||||
CMP R5, $0
|
||||
BEQ insert1
|
||||
MOVBZ (R4), R21
|
||||
SLD R22, R21, R21
|
||||
OR R16, R21, R16
|
||||
ADD $8, R22
|
||||
|
||||
insert1:
|
||||
// Insert 1 at end of msg
|
||||
MOVD $1, R21
|
||||
SLD R22, R21, R21
|
||||
OR R16, R21, R16
|
||||
|
||||
carry:
|
||||
// Add new values to h0, h1, h2
|
||||
ADDC R16, R8
|
||||
ADDE R17, R9
|
||||
ADDZE R10, R10
|
||||
MOVD $16, R5
|
||||
ADD R5, R4
|
||||
BR multiply
|
||||
|
||||
done:
|
||||
// Save h0, h1, h2 in state
|
||||
MOVD R8, 0(R3)
|
||||
MOVD R9, 8(R3)
|
||||
MOVD R10, 16(R3)
|
||||
RET
|
||||
77
vendor/golang.org/x/crypto/internal/poly1305/sum_s390x.go
generated
vendored
Normal file
77
vendor/golang.org/x/crypto/internal/poly1305/sum_s390x.go
generated
vendored
Normal file
@@ -0,0 +1,77 @@
|
||||
// Copyright 2018 The Go Authors. All rights reserved.
|
||||
// Use of this source code is governed by a BSD-style
|
||||
// license that can be found in the LICENSE file.
|
||||
|
||||
//go:build gc && !purego
|
||||
// +build gc,!purego
|
||||
|
||||
package poly1305
|
||||
|
||||
import (
|
||||
"golang.org/x/sys/cpu"
|
||||
)
|
||||
|
||||
// updateVX is an assembly implementation of Poly1305 that uses vector
|
||||
// instructions. It must only be called if the vector facility (vx) is
|
||||
// available.
|
||||
//
|
||||
//go:noescape
|
||||
func updateVX(state *macState, msg []byte)
|
||||
|
||||
// mac is a replacement for macGeneric that uses a larger buffer and redirects
|
||||
// calls that would have gone to updateGeneric to updateVX if the vector
|
||||
// facility is installed.
|
||||
//
|
||||
// A larger buffer is required for good performance because the vector
|
||||
// implementation has a higher fixed cost per call than the generic
|
||||
// implementation.
|
||||
type mac struct {
|
||||
macState
|
||||
|
||||
buffer [16 * TagSize]byte // size must be a multiple of block size (16)
|
||||
offset int
|
||||
}
|
||||
|
||||
func (h *mac) Write(p []byte) (int, error) {
|
||||
nn := len(p)
|
||||
if h.offset > 0 {
|
||||
n := copy(h.buffer[h.offset:], p)
|
||||
if h.offset+n < len(h.buffer) {
|
||||
h.offset += n
|
||||
return nn, nil
|
||||
}
|
||||
p = p[n:]
|
||||
h.offset = 0
|
||||
if cpu.S390X.HasVX {
|
||||
updateVX(&h.macState, h.buffer[:])
|
||||
} else {
|
||||
updateGeneric(&h.macState, h.buffer[:])
|
||||
}
|
||||
}
|
||||
|
||||
tail := len(p) % len(h.buffer) // number of bytes to copy into buffer
|
||||
body := len(p) - tail // number of bytes to process now
|
||||
if body > 0 {
|
||||
if cpu.S390X.HasVX {
|
||||
updateVX(&h.macState, p[:body])
|
||||
} else {
|
||||
updateGeneric(&h.macState, p[:body])
|
||||
}
|
||||
}
|
||||
h.offset = copy(h.buffer[:], p[body:]) // copy tail bytes - can be 0
|
||||
return nn, nil
|
||||
}
|
||||
|
||||
func (h *mac) Sum(out *[TagSize]byte) {
|
||||
state := h.macState
|
||||
remainder := h.buffer[:h.offset]
|
||||
|
||||
// Use the generic implementation if we have 2 or fewer blocks left
|
||||
// to sum. The vector implementation has a higher startup time.
|
||||
if cpu.S390X.HasVX && len(remainder) > 2*TagSize {
|
||||
updateVX(&state, remainder)
|
||||
} else if len(remainder) > 0 {
|
||||
updateGeneric(&state, remainder)
|
||||
}
|
||||
finalize(out, &state.h, &state.s)
|
||||
}
|
||||
504
vendor/golang.org/x/crypto/internal/poly1305/sum_s390x.s
generated
vendored
Normal file
504
vendor/golang.org/x/crypto/internal/poly1305/sum_s390x.s
generated
vendored
Normal file
@@ -0,0 +1,504 @@
|
||||
// Copyright 2018 The Go Authors. All rights reserved.
|
||||
// Use of this source code is governed by a BSD-style
|
||||
// license that can be found in the LICENSE file.
|
||||
|
||||
//go:build gc && !purego
|
||||
// +build gc,!purego
|
||||
|
||||
#include "textflag.h"
|
||||
|
||||
// This implementation of Poly1305 uses the vector facility (vx)
|
||||
// to process up to 2 blocks (32 bytes) per iteration using an
|
||||
// algorithm based on the one described in:
|
||||
//
|
||||
// NEON crypto, Daniel J. Bernstein & Peter Schwabe
|
||||
// https://cryptojedi.org/papers/neoncrypto-20120320.pdf
|
||||
//
|
||||
// This algorithm uses 5 26-bit limbs to represent a 130-bit
|
||||
// value. These limbs are, for the most part, zero extended and
|
||||
// placed into 64-bit vector register elements. Each vector
|
||||
// register is 128-bits wide and so holds 2 of these elements.
|
||||
// Using 26-bit limbs allows us plenty of headroom to accommodate
|
||||
// accumulations before and after multiplication without
|
||||
// overflowing either 32-bits (before multiplication) or 64-bits
|
||||
// (after multiplication).
|
||||
//
|
||||
// In order to parallelise the operations required to calculate
|
||||
// the sum we use two separate accumulators and then sum those
|
||||
// in an extra final step. For compatibility with the generic
|
||||
// implementation we perform this summation at the end of every
|
||||
// updateVX call.
|
||||
//
|
||||
// To use two accumulators we must multiply the message blocks
|
||||
// by r² rather than r. Only the final message block should be
|
||||
// multiplied by r.
|
||||
//
|
||||
// Example:
|
||||
//
|
||||
// We want to calculate the sum (h) for a 64 byte message (m):
|
||||
//
|
||||
// h = m[0:16]r⁴ + m[16:32]r³ + m[32:48]r² + m[48:64]r
|
||||
//
|
||||
// To do this we split the calculation into the even indices
|
||||
// and odd indices of the message. These form our SIMD 'lanes':
|
||||
//
|
||||
// h = m[ 0:16]r⁴ + m[32:48]r² + <- lane 0
|
||||
// m[16:32]r³ + m[48:64]r <- lane 1
|
||||
//
|
||||
// To calculate this iteratively we refactor so that both lanes
|
||||
// are written in terms of r² and r:
|
||||
//
|
||||
// h = (m[ 0:16]r² + m[32:48])r² + <- lane 0
|
||||
// (m[16:32]r² + m[48:64])r <- lane 1
|
||||
// ^ ^
|
||||
// | coefficients for second iteration
|
||||
// coefficients for first iteration
|
||||
//
|
||||
// So in this case we would have two iterations. In the first
|
||||
// both lanes are multiplied by r². In the second only the
|
||||
// first lane is multiplied by r² and the second lane is
|
||||
// instead multiplied by r. This gives use the odd and even
|
||||
// powers of r that we need from the original equation.
|
||||
//
|
||||
// Notation:
|
||||
//
|
||||
// h - accumulator
|
||||
// r - key
|
||||
// m - message
|
||||
//
|
||||
// [a, b] - SIMD register holding two 64-bit values
|
||||
// [a, b, c, d] - SIMD register holding four 32-bit values
|
||||
// xᵢ[n] - limb n of variable x with bit width i
|
||||
//
|
||||
// Limbs are expressed in little endian order, so for 26-bit
|
||||
// limbs x₂₆[4] will be the most significant limb and x₂₆[0]
|
||||
// will be the least significant limb.
|
||||
|
||||
// masking constants
|
||||
#define MOD24 V0 // [0x0000000000ffffff, 0x0000000000ffffff] - mask low 24-bits
|
||||
#define MOD26 V1 // [0x0000000003ffffff, 0x0000000003ffffff] - mask low 26-bits
|
||||
|
||||
// expansion constants (see EXPAND macro)
|
||||
#define EX0 V2
|
||||
#define EX1 V3
|
||||
#define EX2 V4
|
||||
|
||||
// key (r², r or 1 depending on context)
|
||||
#define R_0 V5
|
||||
#define R_1 V6
|
||||
#define R_2 V7
|
||||
#define R_3 V8
|
||||
#define R_4 V9
|
||||
|
||||
// precalculated coefficients (5r², 5r or 0 depending on context)
|
||||
#define R5_1 V10
|
||||
#define R5_2 V11
|
||||
#define R5_3 V12
|
||||
#define R5_4 V13
|
||||
|
||||
// message block (m)
|
||||
#define M_0 V14
|
||||
#define M_1 V15
|
||||
#define M_2 V16
|
||||
#define M_3 V17
|
||||
#define M_4 V18
|
||||
|
||||
// accumulator (h)
|
||||
#define H_0 V19
|
||||
#define H_1 V20
|
||||
#define H_2 V21
|
||||
#define H_3 V22
|
||||
#define H_4 V23
|
||||
|
||||
// temporary registers (for short-lived values)
|
||||
#define T_0 V24
|
||||
#define T_1 V25
|
||||
#define T_2 V26
|
||||
#define T_3 V27
|
||||
#define T_4 V28
|
||||
|
||||
GLOBL ·constants<>(SB), RODATA, $0x30
|
||||
// EX0
|
||||
DATA ·constants<>+0x00(SB)/8, $0x0006050403020100
|
||||
DATA ·constants<>+0x08(SB)/8, $0x1016151413121110
|
||||
// EX1
|
||||
DATA ·constants<>+0x10(SB)/8, $0x060c0b0a09080706
|
||||
DATA ·constants<>+0x18(SB)/8, $0x161c1b1a19181716
|
||||
// EX2
|
||||
DATA ·constants<>+0x20(SB)/8, $0x0d0d0d0d0d0f0e0d
|
||||
DATA ·constants<>+0x28(SB)/8, $0x1d1d1d1d1d1f1e1d
|
||||
|
||||
// MULTIPLY multiplies each lane of f and g, partially reduced
|
||||
// modulo 2¹³⁰ - 5. The result, h, consists of partial products
|
||||
// in each lane that need to be reduced further to produce the
|
||||
// final result.
|
||||
//
|
||||
// h₁₃₀ = (f₁₃₀g₁₃₀) % 2¹³⁰ + (5f₁₃₀g₁₃₀) / 2¹³⁰
|
||||
//
|
||||
// Note that the multiplication by 5 of the high bits is
|
||||
// achieved by precalculating the multiplication of four of the
|
||||
// g coefficients by 5. These are g51-g54.
|
||||
#define MULTIPLY(f0, f1, f2, f3, f4, g0, g1, g2, g3, g4, g51, g52, g53, g54, h0, h1, h2, h3, h4) \
|
||||
VMLOF f0, g0, h0 \
|
||||
VMLOF f0, g3, h3 \
|
||||
VMLOF f0, g1, h1 \
|
||||
VMLOF f0, g4, h4 \
|
||||
VMLOF f0, g2, h2 \
|
||||
VMLOF f1, g54, T_0 \
|
||||
VMLOF f1, g2, T_3 \
|
||||
VMLOF f1, g0, T_1 \
|
||||
VMLOF f1, g3, T_4 \
|
||||
VMLOF f1, g1, T_2 \
|
||||
VMALOF f2, g53, h0, h0 \
|
||||
VMALOF f2, g1, h3, h3 \
|
||||
VMALOF f2, g54, h1, h1 \
|
||||
VMALOF f2, g2, h4, h4 \
|
||||
VMALOF f2, g0, h2, h2 \
|
||||
VMALOF f3, g52, T_0, T_0 \
|
||||
VMALOF f3, g0, T_3, T_3 \
|
||||
VMALOF f3, g53, T_1, T_1 \
|
||||
VMALOF f3, g1, T_4, T_4 \
|
||||
VMALOF f3, g54, T_2, T_2 \
|
||||
VMALOF f4, g51, h0, h0 \
|
||||
VMALOF f4, g54, h3, h3 \
|
||||
VMALOF f4, g52, h1, h1 \
|
||||
VMALOF f4, g0, h4, h4 \
|
||||
VMALOF f4, g53, h2, h2 \
|
||||
VAG T_0, h0, h0 \
|
||||
VAG T_3, h3, h3 \
|
||||
VAG T_1, h1, h1 \
|
||||
VAG T_4, h4, h4 \
|
||||
VAG T_2, h2, h2
|
||||
|
||||
// REDUCE performs the following carry operations in four
|
||||
// stages, as specified in Bernstein & Schwabe:
|
||||
//
|
||||
// 1: h₂₆[0]->h₂₆[1] h₂₆[3]->h₂₆[4]
|
||||
// 2: h₂₆[1]->h₂₆[2] h₂₆[4]->h₂₆[0]
|
||||
// 3: h₂₆[0]->h₂₆[1] h₂₆[2]->h₂₆[3]
|
||||
// 4: h₂₆[3]->h₂₆[4]
|
||||
//
|
||||
// The result is that all of the limbs are limited to 26-bits
|
||||
// except for h₂₆[1] and h₂₆[4] which are limited to 27-bits.
|
||||
//
|
||||
// Note that although each limb is aligned at 26-bit intervals
|
||||
// they may contain values that exceed 2²⁶ - 1, hence the need
|
||||
// to carry the excess bits in each limb.
|
||||
#define REDUCE(h0, h1, h2, h3, h4) \
|
||||
VESRLG $26, h0, T_0 \
|
||||
VESRLG $26, h3, T_1 \
|
||||
VN MOD26, h0, h0 \
|
||||
VN MOD26, h3, h3 \
|
||||
VAG T_0, h1, h1 \
|
||||
VAG T_1, h4, h4 \
|
||||
VESRLG $26, h1, T_2 \
|
||||
VESRLG $26, h4, T_3 \
|
||||
VN MOD26, h1, h1 \
|
||||
VN MOD26, h4, h4 \
|
||||
VESLG $2, T_3, T_4 \
|
||||
VAG T_3, T_4, T_4 \
|
||||
VAG T_2, h2, h2 \
|
||||
VAG T_4, h0, h0 \
|
||||
VESRLG $26, h2, T_0 \
|
||||
VESRLG $26, h0, T_1 \
|
||||
VN MOD26, h2, h2 \
|
||||
VN MOD26, h0, h0 \
|
||||
VAG T_0, h3, h3 \
|
||||
VAG T_1, h1, h1 \
|
||||
VESRLG $26, h3, T_2 \
|
||||
VN MOD26, h3, h3 \
|
||||
VAG T_2, h4, h4
|
||||
|
||||
// EXPAND splits the 128-bit little-endian values in0 and in1
|
||||
// into 26-bit big-endian limbs and places the results into
|
||||
// the first and second lane of d₂₆[0:4] respectively.
|
||||
//
|
||||
// The EX0, EX1 and EX2 constants are arrays of byte indices
|
||||
// for permutation. The permutation both reverses the bytes
|
||||
// in the input and ensures the bytes are copied into the
|
||||
// destination limb ready to be shifted into their final
|
||||
// position.
|
||||
#define EXPAND(in0, in1, d0, d1, d2, d3, d4) \
|
||||
VPERM in0, in1, EX0, d0 \
|
||||
VPERM in0, in1, EX1, d2 \
|
||||
VPERM in0, in1, EX2, d4 \
|
||||
VESRLG $26, d0, d1 \
|
||||
VESRLG $30, d2, d3 \
|
||||
VESRLG $4, d2, d2 \
|
||||
VN MOD26, d0, d0 \ // [in0₂₆[0], in1₂₆[0]]
|
||||
VN MOD26, d3, d3 \ // [in0₂₆[3], in1₂₆[3]]
|
||||
VN MOD26, d1, d1 \ // [in0₂₆[1], in1₂₆[1]]
|
||||
VN MOD24, d4, d4 \ // [in0₂₆[4], in1₂₆[4]]
|
||||
VN MOD26, d2, d2 // [in0₂₆[2], in1₂₆[2]]
|
||||
|
||||
// func updateVX(state *macState, msg []byte)
|
||||
TEXT ·updateVX(SB), NOSPLIT, $0
|
||||
MOVD state+0(FP), R1
|
||||
LMG msg+8(FP), R2, R3 // R2=msg_base, R3=msg_len
|
||||
|
||||
// load EX0, EX1 and EX2
|
||||
MOVD $·constants<>(SB), R5
|
||||
VLM (R5), EX0, EX2
|
||||
|
||||
// generate masks
|
||||
VGMG $(64-24), $63, MOD24 // [0x00ffffff, 0x00ffffff]
|
||||
VGMG $(64-26), $63, MOD26 // [0x03ffffff, 0x03ffffff]
|
||||
|
||||
// load h (accumulator) and r (key) from state
|
||||
VZERO T_1 // [0, 0]
|
||||
VL 0(R1), T_0 // [h₆₄[0], h₆₄[1]]
|
||||
VLEG $0, 16(R1), T_1 // [h₆₄[2], 0]
|
||||
VL 24(R1), T_2 // [r₆₄[0], r₆₄[1]]
|
||||
VPDI $0, T_0, T_2, T_3 // [h₆₄[0], r₆₄[0]]
|
||||
VPDI $5, T_0, T_2, T_4 // [h₆₄[1], r₆₄[1]]
|
||||
|
||||
// unpack h and r into 26-bit limbs
|
||||
// note: h₆₄[2] may have the low 3 bits set, so h₂₆[4] is a 27-bit value
|
||||
VN MOD26, T_3, H_0 // [h₂₆[0], r₂₆[0]]
|
||||
VZERO H_1 // [0, 0]
|
||||
VZERO H_3 // [0, 0]
|
||||
VGMG $(64-12-14), $(63-12), T_0 // [0x03fff000, 0x03fff000] - 26-bit mask with low 12 bits masked out
|
||||
VESLG $24, T_1, T_1 // [h₆₄[2]<<24, 0]
|
||||
VERIMG $-26&63, T_3, MOD26, H_1 // [h₂₆[1], r₂₆[1]]
|
||||
VESRLG $+52&63, T_3, H_2 // [h₂₆[2], r₂₆[2]] - low 12 bits only
|
||||
VERIMG $-14&63, T_4, MOD26, H_3 // [h₂₆[1], r₂₆[1]]
|
||||
VESRLG $40, T_4, H_4 // [h₂₆[4], r₂₆[4]] - low 24 bits only
|
||||
VERIMG $+12&63, T_4, T_0, H_2 // [h₂₆[2], r₂₆[2]] - complete
|
||||
VO T_1, H_4, H_4 // [h₂₆[4], r₂₆[4]] - complete
|
||||
|
||||
// replicate r across all 4 vector elements
|
||||
VREPF $3, H_0, R_0 // [r₂₆[0], r₂₆[0], r₂₆[0], r₂₆[0]]
|
||||
VREPF $3, H_1, R_1 // [r₂₆[1], r₂₆[1], r₂₆[1], r₂₆[1]]
|
||||
VREPF $3, H_2, R_2 // [r₂₆[2], r₂₆[2], r₂₆[2], r₂₆[2]]
|
||||
VREPF $3, H_3, R_3 // [r₂₆[3], r₂₆[3], r₂₆[3], r₂₆[3]]
|
||||
VREPF $3, H_4, R_4 // [r₂₆[4], r₂₆[4], r₂₆[4], r₂₆[4]]
|
||||
|
||||
// zero out lane 1 of h
|
||||
VLEIG $1, $0, H_0 // [h₂₆[0], 0]
|
||||
VLEIG $1, $0, H_1 // [h₂₆[1], 0]
|
||||
VLEIG $1, $0, H_2 // [h₂₆[2], 0]
|
||||
VLEIG $1, $0, H_3 // [h₂₆[3], 0]
|
||||
VLEIG $1, $0, H_4 // [h₂₆[4], 0]
|
||||
|
||||
// calculate 5r (ignore least significant limb)
|
||||
VREPIF $5, T_0
|
||||
VMLF T_0, R_1, R5_1 // [5r₂₆[1], 5r₂₆[1], 5r₂₆[1], 5r₂₆[1]]
|
||||
VMLF T_0, R_2, R5_2 // [5r₂₆[2], 5r₂₆[2], 5r₂₆[2], 5r₂₆[2]]
|
||||
VMLF T_0, R_3, R5_3 // [5r₂₆[3], 5r₂₆[3], 5r₂₆[3], 5r₂₆[3]]
|
||||
VMLF T_0, R_4, R5_4 // [5r₂₆[4], 5r₂₆[4], 5r₂₆[4], 5r₂₆[4]]
|
||||
|
||||
// skip r² calculation if we are only calculating one block
|
||||
CMPBLE R3, $16, skip
|
||||
|
||||
// calculate r²
|
||||
MULTIPLY(R_0, R_1, R_2, R_3, R_4, R_0, R_1, R_2, R_3, R_4, R5_1, R5_2, R5_3, R5_4, M_0, M_1, M_2, M_3, M_4)
|
||||
REDUCE(M_0, M_1, M_2, M_3, M_4)
|
||||
VGBM $0x0f0f, T_0
|
||||
VERIMG $0, M_0, T_0, R_0 // [r₂₆[0], r²₂₆[0], r₂₆[0], r²₂₆[0]]
|
||||
VERIMG $0, M_1, T_0, R_1 // [r₂₆[1], r²₂₆[1], r₂₆[1], r²₂₆[1]]
|
||||
VERIMG $0, M_2, T_0, R_2 // [r₂₆[2], r²₂₆[2], r₂₆[2], r²₂₆[2]]
|
||||
VERIMG $0, M_3, T_0, R_3 // [r₂₆[3], r²₂₆[3], r₂₆[3], r²₂₆[3]]
|
||||
VERIMG $0, M_4, T_0, R_4 // [r₂₆[4], r²₂₆[4], r₂₆[4], r²₂₆[4]]
|
||||
|
||||
// calculate 5r² (ignore least significant limb)
|
||||
VREPIF $5, T_0
|
||||
VMLF T_0, R_1, R5_1 // [5r₂₆[1], 5r²₂₆[1], 5r₂₆[1], 5r²₂₆[1]]
|
||||
VMLF T_0, R_2, R5_2 // [5r₂₆[2], 5r²₂₆[2], 5r₂₆[2], 5r²₂₆[2]]
|
||||
VMLF T_0, R_3, R5_3 // [5r₂₆[3], 5r²₂₆[3], 5r₂₆[3], 5r²₂₆[3]]
|
||||
VMLF T_0, R_4, R5_4 // [5r₂₆[4], 5r²₂₆[4], 5r₂₆[4], 5r²₂₆[4]]
|
||||
|
||||
loop:
|
||||
CMPBLE R3, $32, b2 // 2 or fewer blocks remaining, need to change key coefficients
|
||||
|
||||
// load next 2 blocks from message
|
||||
VLM (R2), T_0, T_1
|
||||
|
||||
// update message slice
|
||||
SUB $32, R3
|
||||
MOVD $32(R2), R2
|
||||
|
||||
// unpack message blocks into 26-bit big-endian limbs
|
||||
EXPAND(T_0, T_1, M_0, M_1, M_2, M_3, M_4)
|
||||
|
||||
// add 2¹²⁸ to each message block value
|
||||
VLEIB $4, $1, M_4
|
||||
VLEIB $12, $1, M_4
|
||||
|
||||
multiply:
|
||||
// accumulate the incoming message
|
||||
VAG H_0, M_0, M_0
|
||||
VAG H_3, M_3, M_3
|
||||
VAG H_1, M_1, M_1
|
||||
VAG H_4, M_4, M_4
|
||||
VAG H_2, M_2, M_2
|
||||
|
||||
// multiply the accumulator by the key coefficient
|
||||
MULTIPLY(M_0, M_1, M_2, M_3, M_4, R_0, R_1, R_2, R_3, R_4, R5_1, R5_2, R5_3, R5_4, H_0, H_1, H_2, H_3, H_4)
|
||||
|
||||
// carry and partially reduce the partial products
|
||||
REDUCE(H_0, H_1, H_2, H_3, H_4)
|
||||
|
||||
CMPBNE R3, $0, loop
|
||||
|
||||
finish:
|
||||
// sum lane 0 and lane 1 and put the result in lane 1
|
||||
VZERO T_0
|
||||
VSUMQG H_0, T_0, H_0
|
||||
VSUMQG H_3, T_0, H_3
|
||||
VSUMQG H_1, T_0, H_1
|
||||
VSUMQG H_4, T_0, H_4
|
||||
VSUMQG H_2, T_0, H_2
|
||||
|
||||
// reduce again after summation
|
||||
// TODO(mundaym): there might be a more efficient way to do this
|
||||
// now that we only have 1 active lane. For example, we could
|
||||
// simultaneously pack the values as we reduce them.
|
||||
REDUCE(H_0, H_1, H_2, H_3, H_4)
|
||||
|
||||
// carry h[1] through to h[4] so that only h[4] can exceed 2²⁶ - 1
|
||||
// TODO(mundaym): in testing this final carry was unnecessary.
|
||||
// Needs a proof before it can be removed though.
|
||||
VESRLG $26, H_1, T_1
|
||||
VN MOD26, H_1, H_1
|
||||
VAQ T_1, H_2, H_2
|
||||
VESRLG $26, H_2, T_2
|
||||
VN MOD26, H_2, H_2
|
||||
VAQ T_2, H_3, H_3
|
||||
VESRLG $26, H_3, T_3
|
||||
VN MOD26, H_3, H_3
|
||||
VAQ T_3, H_4, H_4
|
||||
|
||||
// h is now < 2(2¹³⁰ - 5)
|
||||
// Pack each lane in h₂₆[0:4] into h₁₂₈[0:1].
|
||||
VESLG $26, H_1, H_1
|
||||
VESLG $26, H_3, H_3
|
||||
VO H_0, H_1, H_0
|
||||
VO H_2, H_3, H_2
|
||||
VESLG $4, H_2, H_2
|
||||
VLEIB $7, $48, H_1
|
||||
VSLB H_1, H_2, H_2
|
||||
VO H_0, H_2, H_0
|
||||
VLEIB $7, $104, H_1
|
||||
VSLB H_1, H_4, H_3
|
||||
VO H_3, H_0, H_0
|
||||
VLEIB $7, $24, H_1
|
||||
VSRLB H_1, H_4, H_1
|
||||
|
||||
// update state
|
||||
VSTEG $1, H_0, 0(R1)
|
||||
VSTEG $0, H_0, 8(R1)
|
||||
VSTEG $1, H_1, 16(R1)
|
||||
RET
|
||||
|
||||
b2: // 2 or fewer blocks remaining
|
||||
CMPBLE R3, $16, b1
|
||||
|
||||
// Load the 2 remaining blocks (17-32 bytes remaining).
|
||||
MOVD $-17(R3), R0 // index of final byte to load modulo 16
|
||||
VL (R2), T_0 // load full 16 byte block
|
||||
VLL R0, 16(R2), T_1 // load final (possibly partial) block and pad with zeros to 16 bytes
|
||||
|
||||
// The Poly1305 algorithm requires that a 1 bit be appended to
|
||||
// each message block. If the final block is less than 16 bytes
|
||||
// long then it is easiest to insert the 1 before the message
|
||||
// block is split into 26-bit limbs. If, on the other hand, the
|
||||
// final message block is 16 bytes long then we append the 1 bit
|
||||
// after expansion as normal.
|
||||
MOVBZ $1, R0
|
||||
MOVD $-16(R3), R3 // index of byte in last block to insert 1 at (could be 16)
|
||||
CMPBEQ R3, $16, 2(PC) // skip the insertion if the final block is 16 bytes long
|
||||
VLVGB R3, R0, T_1 // insert 1 into the byte at index R3
|
||||
|
||||
// Split both blocks into 26-bit limbs in the appropriate lanes.
|
||||
EXPAND(T_0, T_1, M_0, M_1, M_2, M_3, M_4)
|
||||
|
||||
// Append a 1 byte to the end of the second to last block.
|
||||
VLEIB $4, $1, M_4
|
||||
|
||||
// Append a 1 byte to the end of the last block only if it is a
|
||||
// full 16 byte block.
|
||||
CMPBNE R3, $16, 2(PC)
|
||||
VLEIB $12, $1, M_4
|
||||
|
||||
// Finally, set up the coefficients for the final multiplication.
|
||||
// We have previously saved r and 5r in the 32-bit even indexes
|
||||
// of the R_[0-4] and R5_[1-4] coefficient registers.
|
||||
//
|
||||
// We want lane 0 to be multiplied by r² so that can be kept the
|
||||
// same. We want lane 1 to be multiplied by r so we need to move
|
||||
// the saved r value into the 32-bit odd index in lane 1 by
|
||||
// rotating the 64-bit lane by 32.
|
||||
VGBM $0x00ff, T_0 // [0, 0xffffffffffffffff] - mask lane 1 only
|
||||
VERIMG $32, R_0, T_0, R_0 // [_, r²₂₆[0], _, r₂₆[0]]
|
||||
VERIMG $32, R_1, T_0, R_1 // [_, r²₂₆[1], _, r₂₆[1]]
|
||||
VERIMG $32, R_2, T_0, R_2 // [_, r²₂₆[2], _, r₂₆[2]]
|
||||
VERIMG $32, R_3, T_0, R_3 // [_, r²₂₆[3], _, r₂₆[3]]
|
||||
VERIMG $32, R_4, T_0, R_4 // [_, r²₂₆[4], _, r₂₆[4]]
|
||||
VERIMG $32, R5_1, T_0, R5_1 // [_, 5r²₂₆[1], _, 5r₂₆[1]]
|
||||
VERIMG $32, R5_2, T_0, R5_2 // [_, 5r²₂₆[2], _, 5r₂₆[2]]
|
||||
VERIMG $32, R5_3, T_0, R5_3 // [_, 5r²₂₆[3], _, 5r₂₆[3]]
|
||||
VERIMG $32, R5_4, T_0, R5_4 // [_, 5r²₂₆[4], _, 5r₂₆[4]]
|
||||
|
||||
MOVD $0, R3
|
||||
BR multiply
|
||||
|
||||
skip:
|
||||
CMPBEQ R3, $0, finish
|
||||
|
||||
b1: // 1 block remaining
|
||||
|
||||
// Load the final block (1-16 bytes). This will be placed into
|
||||
// lane 0.
|
||||
MOVD $-1(R3), R0
|
||||
VLL R0, (R2), T_0 // pad to 16 bytes with zeros
|
||||
|
||||
// The Poly1305 algorithm requires that a 1 bit be appended to
|
||||
// each message block. If the final block is less than 16 bytes
|
||||
// long then it is easiest to insert the 1 before the message
|
||||
// block is split into 26-bit limbs. If, on the other hand, the
|
||||
// final message block is 16 bytes long then we append the 1 bit
|
||||
// after expansion as normal.
|
||||
MOVBZ $1, R0
|
||||
CMPBEQ R3, $16, 2(PC)
|
||||
VLVGB R3, R0, T_0
|
||||
|
||||
// Set the message block in lane 1 to the value 0 so that it
|
||||
// can be accumulated without affecting the final result.
|
||||
VZERO T_1
|
||||
|
||||
// Split the final message block into 26-bit limbs in lane 0.
|
||||
// Lane 1 will be contain 0.
|
||||
EXPAND(T_0, T_1, M_0, M_1, M_2, M_3, M_4)
|
||||
|
||||
// Append a 1 byte to the end of the last block only if it is a
|
||||
// full 16 byte block.
|
||||
CMPBNE R3, $16, 2(PC)
|
||||
VLEIB $4, $1, M_4
|
||||
|
||||
// We have previously saved r and 5r in the 32-bit even indexes
|
||||
// of the R_[0-4] and R5_[1-4] coefficient registers.
|
||||
//
|
||||
// We want lane 0 to be multiplied by r so we need to move the
|
||||
// saved r value into the 32-bit odd index in lane 0. We want
|
||||
// lane 1 to be set to the value 1. This makes multiplication
|
||||
// a no-op. We do this by setting lane 1 in every register to 0
|
||||
// and then just setting the 32-bit index 3 in R_0 to 1.
|
||||
VZERO T_0
|
||||
MOVD $0, R0
|
||||
MOVD $0x10111213, R12
|
||||
VLVGP R12, R0, T_1 // [_, 0x10111213, _, 0x00000000]
|
||||
VPERM T_0, R_0, T_1, R_0 // [_, r₂₆[0], _, 0]
|
||||
VPERM T_0, R_1, T_1, R_1 // [_, r₂₆[1], _, 0]
|
||||
VPERM T_0, R_2, T_1, R_2 // [_, r₂₆[2], _, 0]
|
||||
VPERM T_0, R_3, T_1, R_3 // [_, r₂₆[3], _, 0]
|
||||
VPERM T_0, R_4, T_1, R_4 // [_, r₂₆[4], _, 0]
|
||||
VPERM T_0, R5_1, T_1, R5_1 // [_, 5r₂₆[1], _, 0]
|
||||
VPERM T_0, R5_2, T_1, R5_2 // [_, 5r₂₆[2], _, 0]
|
||||
VPERM T_0, R5_3, T_1, R5_3 // [_, 5r₂₆[3], _, 0]
|
||||
VPERM T_0, R5_4, T_1, R5_4 // [_, 5r₂₆[4], _, 0]
|
||||
|
||||
// Set the value of lane 1 to be 1.
|
||||
VLEIF $3, $1, R_0 // [_, r₂₆[0], _, 1]
|
||||
|
||||
MOVD $0, R3
|
||||
BR multiply
|
||||
91
vendor/golang.org/x/crypto/poly1305/poly1305_compat.go
generated
vendored
Normal file
91
vendor/golang.org/x/crypto/poly1305/poly1305_compat.go
generated
vendored
Normal file
@@ -0,0 +1,91 @@
|
||||
// Copyright 2012 The Go Authors. All rights reserved.
|
||||
// Use of this source code is governed by a BSD-style
|
||||
// license that can be found in the LICENSE file.
|
||||
|
||||
// Package poly1305 implements Poly1305 one-time message authentication code as
|
||||
// specified in https://cr.yp.to/mac/poly1305-20050329.pdf.
|
||||
//
|
||||
// Poly1305 is a fast, one-time authentication function. It is infeasible for an
|
||||
// attacker to generate an authenticator for a message without the key. However, a
|
||||
// key must only be used for a single message. Authenticating two different
|
||||
// messages with the same key allows an attacker to forge authenticators for other
|
||||
// messages with the same key.
|
||||
//
|
||||
// Poly1305 was originally coupled with AES in order to make Poly1305-AES. AES was
|
||||
// used with a fixed key in order to generate one-time keys from an nonce.
|
||||
// However, in this package AES isn't used and the one-time key is specified
|
||||
// directly.
|
||||
//
|
||||
// Deprecated: Poly1305 as implemented by this package is a cryptographic
|
||||
// building block that is not safe for general purpose use.
|
||||
// For encryption, use the full ChaCha20-Poly1305 construction implemented by
|
||||
// golang.org/x/crypto/chacha20poly1305. For authentication, use a general
|
||||
// purpose MAC such as HMAC implemented by crypto/hmac.
|
||||
package poly1305 // import "golang.org/x/crypto/poly1305"
|
||||
|
||||
import "golang.org/x/crypto/internal/poly1305"
|
||||
|
||||
// TagSize is the size, in bytes, of a poly1305 authenticator.
|
||||
//
|
||||
// For use with golang.org/x/crypto/chacha20poly1305, chacha20poly1305.Overhead
|
||||
// can be used instead.
|
||||
const TagSize = 16
|
||||
|
||||
// Sum generates an authenticator for msg using a one-time key and puts the
|
||||
// 16-byte result into out. Authenticating two different messages with the same
|
||||
// key allows an attacker to forge messages at will.
|
||||
func Sum(out *[16]byte, m []byte, key *[32]byte) {
|
||||
poly1305.Sum(out, m, key)
|
||||
}
|
||||
|
||||
// Verify returns true if mac is a valid authenticator for m with the given key.
|
||||
func Verify(mac *[16]byte, m []byte, key *[32]byte) bool {
|
||||
return poly1305.Verify(mac, m, key)
|
||||
}
|
||||
|
||||
// New returns a new MAC computing an authentication
|
||||
// tag of all data written to it with the given key.
|
||||
// This allows writing the message progressively instead
|
||||
// of passing it as a single slice. Common users should use
|
||||
// the Sum function instead.
|
||||
//
|
||||
// The key must be unique for each message, as authenticating
|
||||
// two different messages with the same key allows an attacker
|
||||
// to forge messages at will.
|
||||
func New(key *[32]byte) *MAC {
|
||||
return &MAC{mac: poly1305.New(key)}
|
||||
}
|
||||
|
||||
// MAC is an io.Writer computing an authentication tag
|
||||
// of the data written to it.
|
||||
//
|
||||
// MAC cannot be used like common hash.Hash implementations,
|
||||
// because using a poly1305 key twice breaks its security.
|
||||
// Therefore writing data to a running MAC after calling
|
||||
// Sum or Verify causes it to panic.
|
||||
type MAC struct {
|
||||
mac *poly1305.MAC
|
||||
}
|
||||
|
||||
// Size returns the number of bytes Sum will return.
|
||||
func (h *MAC) Size() int { return TagSize }
|
||||
|
||||
// Write adds more data to the running message authentication code.
|
||||
// It never returns an error.
|
||||
//
|
||||
// It must not be called after the first call of Sum or Verify.
|
||||
func (h *MAC) Write(p []byte) (n int, err error) {
|
||||
return h.mac.Write(p)
|
||||
}
|
||||
|
||||
// Sum computes the authenticator of all data written to the
|
||||
// message authentication code.
|
||||
func (h *MAC) Sum(b []byte) []byte {
|
||||
return h.mac.Sum(b)
|
||||
}
|
||||
|
||||
// Verify returns whether the authenticator of all data written to
|
||||
// the message authentication code matches the expected value.
|
||||
func (h *MAC) Verify(expected []byte) bool {
|
||||
return h.mac.Verify(expected)
|
||||
}
|
||||
Reference in New Issue
Block a user