From 98eef328ea1635436072051b3b3d520ea7071dd5 Mon Sep 17 00:00:00 2001 From: "dependabot[bot]" <49699333+dependabot[bot]@users.noreply.github.com> Date: Mon, 11 Nov 2024 15:15:24 +0000 Subject: [chore]: Bump golang.org/x/net from 0.30.0 to 0.31.0 (#3536) Bumps [golang.org/x/net](https://github.com/golang/net) from 0.30.0 to 0.31.0. - [Commits](https://github.com/golang/net/compare/v0.30.0...v0.31.0) --- updated-dependencies: - dependency-name: golang.org/x/net dependency-type: direct:production update-type: version-update:semver-minor ... Signed-off-by: dependabot[bot] Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com> --- .../golang.org/x/crypto/chacha20/chacha_noasm.go | 2 +- .../golang.org/x/crypto/chacha20/chacha_ppc64le.go | 16 - .../golang.org/x/crypto/chacha20/chacha_ppc64le.s | 443 ------------------ .../golang.org/x/crypto/chacha20/chacha_ppc64x.go | 16 + .../golang.org/x/crypto/chacha20/chacha_ppc64x.s | 501 +++++++++++++++++++++ .../x/crypto/internal/poly1305/mac_noasm.go | 2 +- .../x/crypto/internal/poly1305/sum_ppc64le.go | 47 -- .../x/crypto/internal/poly1305/sum_ppc64le.s | 179 -------- .../x/crypto/internal/poly1305/sum_ppc64x.go | 47 ++ .../x/crypto/internal/poly1305/sum_ppc64x.s | 187 ++++++++ vendor/golang.org/x/crypto/sha3/doc.go | 4 + vendor/golang.org/x/crypto/sha3/hashes.go | 31 +- vendor/golang.org/x/crypto/sha3/sha3.go | 187 +++++--- vendor/golang.org/x/crypto/sha3/shake.go | 85 ++-- vendor/golang.org/x/crypto/sha3/xor.go | 40 -- vendor/golang.org/x/crypto/ssh/client_auth.go | 5 + 16 files changed, 962 insertions(+), 830 deletions(-) delete mode 100644 vendor/golang.org/x/crypto/chacha20/chacha_ppc64le.go delete mode 100644 vendor/golang.org/x/crypto/chacha20/chacha_ppc64le.s create mode 100644 vendor/golang.org/x/crypto/chacha20/chacha_ppc64x.go create mode 100644 vendor/golang.org/x/crypto/chacha20/chacha_ppc64x.s delete mode 100644 vendor/golang.org/x/crypto/internal/poly1305/sum_ppc64le.go delete mode 100644 vendor/golang.org/x/crypto/internal/poly1305/sum_ppc64le.s create mode 100644 vendor/golang.org/x/crypto/internal/poly1305/sum_ppc64x.go create mode 100644 vendor/golang.org/x/crypto/internal/poly1305/sum_ppc64x.s delete mode 100644 vendor/golang.org/x/crypto/sha3/xor.go (limited to 'vendor/golang.org/x/crypto') diff --git a/vendor/golang.org/x/crypto/chacha20/chacha_noasm.go b/vendor/golang.org/x/crypto/chacha20/chacha_noasm.go index db42e6676..c709b7284 100644 --- a/vendor/golang.org/x/crypto/chacha20/chacha_noasm.go +++ b/vendor/golang.org/x/crypto/chacha20/chacha_noasm.go @@ -2,7 +2,7 @@ // 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) || !gc || purego +//go:build (!arm64 && !s390x && !ppc64 && !ppc64le) || !gc || purego package chacha20 diff --git a/vendor/golang.org/x/crypto/chacha20/chacha_ppc64le.go b/vendor/golang.org/x/crypto/chacha20/chacha_ppc64le.go deleted file mode 100644 index 3a4287f99..000000000 --- a/vendor/golang.org/x/crypto/chacha20/chacha_ppc64le.go +++ /dev/null @@ -1,16 +0,0 @@ -// 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 - -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) -} diff --git a/vendor/golang.org/x/crypto/chacha20/chacha_ppc64le.s b/vendor/golang.org/x/crypto/chacha20/chacha_ppc64le.s deleted file mode 100644 index c672ccf69..000000000 --- a/vendor/golang.org/x/crypto/chacha20/chacha_ppc64le.s +++ /dev/null @@ -1,443 +0,0 @@ -// 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 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 - -#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 - -// for VPERMXOR -#define MASK R18 - -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 -DATA consts<>+0xa0(SB)/8, $0x5566774411223300 -DATA consts<>+0xa8(SB)/8, $0xddeeffcc99aabb88 -DATA consts<>+0xb0(SB)/8, $0x6677445522330011 -DATA consts<>+0xb8(SB)/8, $0xeeffccddaabb8899 -GLOBL consts<>(SB), RODATA, $0xc0 - -//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 - // for VPERMXOR - MOVD $consts<>+0xa0(SB), MASK - MOVD $16, R20 - // 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 - - // Load mask constants for VPERMXOR - LXVW4X (MASK)(R0), V20 - LXVW4X (MASK)(R20), V21 - - // 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 - PCALIGN $16 -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 - PCALIGN $16 -loop_vsx: - VADDUWM V0, V4, V0 - VADDUWM V1, V5, V1 - VADDUWM V2, V6, V2 - VADDUWM V3, V7, V3 - - VPERMXOR V12, V0, V21, V12 - VPERMXOR V13, V1, V21, V13 - VPERMXOR V14, V2, V21, V14 - VPERMXOR V15, V3, V21, 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 - - VPERMXOR V12, V0, V20, V12 - VPERMXOR V13, V1, V20, V13 - VPERMXOR V14, V2, V20, V14 - VPERMXOR V15, V3, V20, 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 - - VPERMXOR V15, V0, V21, V15 - VPERMXOR V12, V1, V21, V12 - VPERMXOR V13, V2, V21, V13 - VPERMXOR V14, V3, V21, 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 - - VPERMXOR V15, V0, V20, V15 - VPERMXOR V12, V1, V20, V12 - VPERMXOR V13, V2, V20, V13 - VPERMXOR V14, V3, V20, 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 - BDNZ loop_vsx - - VADDUWM V12, V26, V12 - - VMRGEW V0, V1, V27 - VMRGEW V2, V3, V28 - - VMRGOW V0, V1, V0 - VMRGOW V2, V3, V2 - - VMRGEW V4, V5, V29 - VMRGEW V6, V7, V30 - - XXPERMDI VS32, VS34, $0, VS33 - XXPERMDI VS32, VS34, $3, VS35 - XXPERMDI VS59, VS60, $0, VS32 - XXPERMDI VS59, VS60, $3, VS34 - - VMRGOW V4, V5, V4 - VMRGOW V6, V7, V6 - - VMRGEW V8, V9, V27 - VMRGEW V10, V11, V28 - - XXPERMDI VS36, VS38, $0, VS37 - XXPERMDI VS36, VS38, $3, VS39 - XXPERMDI VS61, VS62, $0, VS36 - XXPERMDI VS61, VS62, $3, VS38 - - VMRGOW V8, V9, V8 - VMRGOW V10, V11, V10 - - VMRGEW V12, V13, V29 - VMRGEW V14, V15, V30 - - XXPERMDI VS40, VS42, $0, VS41 - XXPERMDI VS40, VS42, $3, VS43 - XXPERMDI VS59, VS60, $0, VS40 - XXPERMDI VS59, VS60, $3, VS42 - - VMRGOW V12, V13, V12 - 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 - PCALIGN $16 -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) - BDNZ 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 diff --git a/vendor/golang.org/x/crypto/chacha20/chacha_ppc64x.go b/vendor/golang.org/x/crypto/chacha20/chacha_ppc64x.go new file mode 100644 index 000000000..bd183d9ba --- /dev/null +++ b/vendor/golang.org/x/crypto/chacha20/chacha_ppc64x.go @@ -0,0 +1,16 @@ +// 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 && (ppc64 || ppc64le) + +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) +} diff --git a/vendor/golang.org/x/crypto/chacha20/chacha_ppc64x.s b/vendor/golang.org/x/crypto/chacha20/chacha_ppc64x.s new file mode 100644 index 000000000..a660b4112 --- /dev/null +++ b/vendor/golang.org/x/crypto/chacha20/chacha_ppc64x.s @@ -0,0 +1,501 @@ +// 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 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 && (ppc64 || ppc64le) + +#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 + +// for VPERMXOR +#define MASK R18 + +DATA consts<>+0x00(SB)/4, $0x61707865 +DATA consts<>+0x04(SB)/4, $0x3320646e +DATA consts<>+0x08(SB)/4, $0x79622d32 +DATA consts<>+0x0c(SB)/4, $0x6b206574 +DATA consts<>+0x10(SB)/4, $0x00000001 +DATA consts<>+0x14(SB)/4, $0x00000000 +DATA consts<>+0x18(SB)/4, $0x00000000 +DATA consts<>+0x1c(SB)/4, $0x00000000 +DATA consts<>+0x20(SB)/4, $0x00000004 +DATA consts<>+0x24(SB)/4, $0x00000000 +DATA consts<>+0x28(SB)/4, $0x00000000 +DATA consts<>+0x2c(SB)/4, $0x00000000 +DATA consts<>+0x30(SB)/4, $0x0e0f0c0d +DATA consts<>+0x34(SB)/4, $0x0a0b0809 +DATA consts<>+0x38(SB)/4, $0x06070405 +DATA consts<>+0x3c(SB)/4, $0x02030001 +DATA consts<>+0x40(SB)/4, $0x0d0e0f0c +DATA consts<>+0x44(SB)/4, $0x090a0b08 +DATA consts<>+0x48(SB)/4, $0x05060704 +DATA consts<>+0x4c(SB)/4, $0x01020300 +DATA consts<>+0x50(SB)/4, $0x61707865 +DATA consts<>+0x54(SB)/4, $0x61707865 +DATA consts<>+0x58(SB)/4, $0x61707865 +DATA consts<>+0x5c(SB)/4, $0x61707865 +DATA consts<>+0x60(SB)/4, $0x3320646e +DATA consts<>+0x64(SB)/4, $0x3320646e +DATA consts<>+0x68(SB)/4, $0x3320646e +DATA consts<>+0x6c(SB)/4, $0x3320646e +DATA consts<>+0x70(SB)/4, $0x79622d32 +DATA consts<>+0x74(SB)/4, $0x79622d32 +DATA consts<>+0x78(SB)/4, $0x79622d32 +DATA consts<>+0x7c(SB)/4, $0x79622d32 +DATA consts<>+0x80(SB)/4, $0x6b206574 +DATA consts<>+0x84(SB)/4, $0x6b206574 +DATA consts<>+0x88(SB)/4, $0x6b206574 +DATA consts<>+0x8c(SB)/4, $0x6b206574 +DATA consts<>+0x90(SB)/4, $0x00000000 +DATA consts<>+0x94(SB)/4, $0x00000001 +DATA consts<>+0x98(SB)/4, $0x00000002 +DATA consts<>+0x9c(SB)/4, $0x00000003 +DATA consts<>+0xa0(SB)/4, $0x11223300 +DATA consts<>+0xa4(SB)/4, $0x55667744 +DATA consts<>+0xa8(SB)/4, $0x99aabb88 +DATA consts<>+0xac(SB)/4, $0xddeeffcc +DATA consts<>+0xb0(SB)/4, $0x22330011 +DATA consts<>+0xb4(SB)/4, $0x66774455 +DATA consts<>+0xb8(SB)/4, $0xaabb8899 +DATA consts<>+0xbc(SB)/4, $0xeeffccdd +GLOBL consts<>(SB), RODATA, $0xc0 + +#ifdef GOARCH_ppc64 +#define BE_XXBRW_INIT() \ + LVSL (R0)(R0), V24 \ + VSPLTISB $3, V25 \ + VXOR V24, V25, V24 \ + +#define BE_XXBRW(vr) VPERM vr, vr, V24, vr +#else +#define BE_XXBRW_INIT() +#define BE_XXBRW(vr) +#endif + +//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 + // for VPERMXOR + MOVD $consts<>+0xa0(SB), MASK + MOVD $16, R20 + // 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 + + BE_XXBRW_INIT() + + // V28 + LXVW4X (CONSTBASE)(R11), VS60 + + // Load mask constants for VPERMXOR + LXVW4X (MASK)(R0), V20 + LXVW4X (MASK)(R20), V21 + + // 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 + PCALIGN $16 +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 + PCALIGN $16 +loop_vsx: + VADDUWM V0, V4, V0 + VADDUWM V1, V5, V1 + VADDUWM V2, V6, V2 + VADDUWM V3, V7, V3 + + VPERMXOR V12, V0, V21, V12 + VPERMXOR V13, V1, V21, V13 + VPERMXOR V14, V2, V21, V14 + VPERMXOR V15, V3, V21, 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 + + VPERMXOR V12, V0, V20, V12 + VPERMXOR V13, V1, V20, V13 + VPERMXOR V14, V2, V20, V14 + VPERMXOR V15, V3, V20, 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 + + VPERMXOR V15, V0, V21, V15 + VPERMXOR V12, V1, V21, V12 + VPERMXOR V13, V2, V21, V13 + VPERMXOR V14, V3, V21, 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 + + VPERMXOR V15, V0, V20, V15 + VPERMXOR V12, V1, V20, V12 + VPERMXOR V13, V2, V20, V13 + VPERMXOR V14, V3, V20, 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 + BDNZ loop_vsx + + VADDUWM V12, V26, V12 + + VMRGEW V0, V1, V27 + VMRGEW V2, V3, V28 + + VMRGOW V0, V1, V0 + VMRGOW V2, V3, V2 + + VMRGEW V4, V5, V29 + VMRGEW V6, V7, V30 + + XXPERMDI VS32, VS34, $0, VS33 + XXPERMDI VS32, VS34, $3, VS35 + XXPERMDI VS59, VS60, $0, VS32 + XXPERMDI VS59, VS60, $3, VS34 + + VMRGOW V4, V5, V4 + VMRGOW V6, V7, V6 + + VMRGEW V8, V9, V27 + VMRGEW V10, V11, V28 + + XXPERMDI VS36, VS38, $0, VS37 + XXPERMDI VS36, VS38, $3, VS39 + XXPERMDI VS61, VS62, $0, VS36 + XXPERMDI VS61, VS62, $3, VS38 + + VMRGOW V8, V9, V8 + VMRGOW V10, V11, V10 + + VMRGEW V12, V13, V29 + VMRGEW V14, V15, V30 + + XXPERMDI VS40, VS42, $0, VS41 + XXPERMDI VS40, VS42, $3, VS43 + XXPERMDI VS59, VS60, $0, VS40 + XXPERMDI VS59, VS60, $3, VS42 + + VMRGOW V12, V13, V12 + 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 + + BE_XXBRW(V0) + BE_XXBRW(V4) + BE_XXBRW(V8) + BE_XXBRW(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 + + BE_XXBRW(V0) + BE_XXBRW(V4) + BE_XXBRW(V8) + BE_XXBRW(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 + + BE_XXBRW(V0) + BE_XXBRW(V4) + BE_XXBRW(V8) + BE_XXBRW(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 + + BE_XXBRW(V0) + BE_XXBRW(V4) + BE_XXBRW(V8) + BE_XXBRW(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 + MOVWZ (CNT), R14 + ADD BLOCKS, R14 + MOVWZ 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 + PCALIGN $16 +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) + BDNZ 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 diff --git a/vendor/golang.org/x/crypto/internal/poly1305/mac_noasm.go b/vendor/golang.org/x/crypto/internal/poly1305/mac_noasm.go index 333da285b..bd896bdc7 100644 --- a/vendor/golang.org/x/crypto/internal/poly1305/mac_noasm.go +++ b/vendor/golang.org/x/crypto/internal/poly1305/mac_noasm.go @@ -2,7 +2,7 @@ // 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 +//go:build (!amd64 && !ppc64le && !ppc64 && !s390x) || !gc || purego package poly1305 diff --git a/vendor/golang.org/x/crypto/internal/poly1305/sum_ppc64le.go b/vendor/golang.org/x/crypto/internal/poly1305/sum_ppc64le.go deleted file mode 100644 index 4aec4874b..000000000 --- a/vendor/golang.org/x/crypto/internal/poly1305/sum_ppc64le.go +++ /dev/null @@ -1,47 +0,0 @@ -// 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 - -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) -} diff --git a/vendor/golang.org/x/crypto/internal/poly1305/sum_ppc64le.s b/vendor/golang.org/x/crypto/internal/poly1305/sum_ppc64le.s deleted file mode 100644 index b3c1699bf..000000000 --- a/vendor/golang.org/x/crypto/internal/poly1305/sum_ppc64le.s +++ /dev/null @@ -1,179 +0,0 @@ -// 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 - -#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; \ - MULHDU r0, h0, t1; \ - MULLD r0, h1, t4; \ - MULHDU r0, h1, t5; \ - ADDC t4, t1, t1; \ - MULLD r0, h2, t2; \ - MULHDU r1, h0, t4; \ - MULLD r1, h0, h0; \ - ADDE 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; \ - ADDZE t3; \ - RLDICL $0, t2, $62, h2; \ - AND t2, t4, h0; \ - ADDC t0, h0, h0; \ - ADDE t3, t1, 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) - - PCALIGN $16 -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 diff --git a/vendor/golang.org/x/crypto/internal/poly1305/sum_ppc64x.go b/vendor/golang.org/x/crypto/internal/poly1305/sum_ppc64x.go new file mode 100644 index 000000000..1a1679aaa --- /dev/null +++ b/vendor/golang.org/x/crypto/internal/poly1305/sum_ppc64x.go @@ -0,0 +1,47 @@ +// 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 && (ppc64 || ppc64le) + +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) +} diff --git a/vendor/golang.org/x/crypto/internal/poly1305/sum_ppc64x.s b/vendor/golang.org/x/crypto/internal/poly1305/sum_ppc64x.s new file mode 100644 index 000000000..6899a1dab --- /dev/null +++ b/vendor/golang.org/x/crypto/internal/poly1305/sum_ppc64x.s @@ -0,0 +1,187 @@ +// 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 && (ppc64 || ppc64le) + +#include "textflag.h" + +// This was ported from the amd64 implementation. + +#ifdef GOARCH_ppc64le +#define LE_MOVD MOVD +#define LE_MOVWZ MOVWZ +#define LE_MOVHZ MOVHZ +#else +#define LE_MOVD MOVDBR +#define LE_MOVWZ MOVWBR +#define LE_MOVHZ MOVHBR +#endif + +#define POLY1305_ADD(msg, h0, h1, h2, t0, t1, t2) \ + LE_MOVD (msg)( R0), t0; \ + LE_MOVD (msg)(R24), 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; \ + MULHDU r0, h0, t1; \ + MULLD r0, h1, t4; \ + MULHDU r0, h1, t5; \ + ADDC t4, t1, t1; \ + MULLD r0, h2, t2; \ + MULHDU r1, h0, t4; \ + MULLD r1, h0, h0; \ + ADDE 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; \ + ADDZE t3; \ + RLDICL $0, t2, $62, h2; \ + AND t2, t4, h0; \ + ADDC t0, h0, h0; \ + ADDE t3, t1, 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 + +// 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 + + MOVD $8, R24 + + CMP R5, $16 + BLT bytes_between_0_and_15 + +loop: + POLY1305_ADD(R4, R8, R9, R10, R20, R21, R22) + + PCALIGN $16 +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) + LE_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 + LE_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 + LE_MOVWZ (R4), R16 + ADD $4, R4 + ADD $-4, R5 + MOVD $32, R22 + +less4: + CMP R5, $2 + BLT less2 + LE_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 diff --git a/vendor/golang.org/x/crypto/sha3/doc.go b/vendor/golang.org/x/crypto/sha3/doc.go index 7e0230907..bbf391fe6 100644 --- a/vendor/golang.org/x/crypto/sha3/doc.go +++ b/vendor/golang.org/x/crypto/sha3/doc.go @@ -5,6 +5,10 @@ // Package sha3 implements the SHA-3 fixed-output-length hash functions and // the SHAKE variable-output-length hash functions defined by FIPS-202. // +// All types in this package also implement [encoding.BinaryMarshaler], +// [encoding.BinaryAppender] and [encoding.BinaryUnmarshaler] to marshal and +// unmarshal the internal state of the hash. +// // Both types of hash function use the "sponge" construction and the Keccak // permutation. For a detailed specification see http://keccak.noekeon.org/ // diff --git a/vendor/golang.org/x/crypto/sha3/hashes.go b/vendor/golang.org/x/crypto/sha3/hashes.go index c544b29e5..31fffbe04 100644 --- a/vendor/golang.org/x/crypto/sha3/hashes.go +++ b/vendor/golang.org/x/crypto/sha3/hashes.go @@ -48,33 +48,52 @@ func init() { crypto.RegisterHash(crypto.SHA3_512, New512) } +const ( + dsbyteSHA3 = 0b00000110 + dsbyteKeccak = 0b00000001 + dsbyteShake = 0b00011111 + dsbyteCShake = 0b00000100 + + // rateK[c] is the rate in bytes for Keccak[c] where c is the capacity in + // bits. Given the sponge size is 1600 bits, the rate is 1600 - c bits. + rateK256 = (1600 - 256) / 8 + rateK448 = (1600 - 448) / 8 + rateK512 = (1600 - 512) / 8 + rateK768 = (1600 - 768) / 8 + rateK1024 = (1600 - 1024) / 8 +) + func new224Generic() *state { - return &state{rate: 144, outputLen: 28, dsbyte: 0x06} + return &state{rate: rateK448, outputLen: 28, dsbyte: dsbyteSHA3} } func new256Generic() *state { - return &state{rate: 136, outputLen: 32, dsbyte: 0x06} + return &state{rate: rateK512, outputLen: 32, dsbyte: dsbyteSHA3} } func new384Generic() *state { - return &state{rate: 104, outputLen: 48, dsbyte: 0x06} + return &state{rate: rateK768, outputLen: 48, dsbyte: dsbyteSHA3} } func new512Generic() *state { - return &state{rate: 72, outputLen: 64, dsbyte: 0x06} + return &state{rate: rateK1024, outputLen: 64, dsbyte: dsbyteSHA3} } // NewLegacyKeccak256 creates a new Keccak-256 hash. // // Only use this function if you require compatibility with an existing cryptosystem // that uses non-standard padding. All other users should use New256 instead. -func NewLegacyKeccak256() hash.Hash { return &state{rate: 136, outputLen: 32, dsbyte: 0x01} } +func NewLegacyKeccak256() hash.Hash { + return &state{rate: rateK512, outputLen: 32, dsbyte: dsbyteKeccak} +} // NewLegacyKeccak512 creates a new Keccak-512 hash. // // Only use this function if you require compatibility with an existing cryptosystem // that uses non-standard padding. All other users should use New512 instead. -func NewLegacyKeccak512() hash.Hash { return &state{rate: 72, outputLen: 64, dsbyte: 0x01} } +func NewLegacyKeccak512() hash.Hash { + return &state{rate: rateK1024, outputLen: 64, dsbyte: dsbyteKeccak} +} // Sum224 returns the SHA3-224 digest of the data. func Sum224(data []byte) (digest [28]byte) { diff --git a/vendor/golang.org/x/crypto/sha3/sha3.go b/vendor/golang.org/x/crypto/sha3/sha3.go index afedde5ab..6658c4447 100644 --- a/vendor/golang.org/x/crypto/sha3/sha3.go +++ b/vendor/golang.org/x/crypto/sha3/sha3.go @@ -4,6 +4,15 @@ package sha3 +import ( + "crypto/subtle" + "encoding/binary" + "errors" + "unsafe" + + "golang.org/x/sys/cpu" +) + // spongeDirection indicates the direction bytes are flowing through the sponge. type spongeDirection int @@ -14,16 +23,13 @@ const ( spongeSqueezing ) -const ( - // maxRate is the maximum size of the internal buffer. SHAKE-256 - // currently needs the largest buffer. - maxRate = 168 -) - type state struct { - // Generic sponge components. - a [25]uint64 // main state of the hash - rate int // the number of bytes of state to use + a [1600 / 8]byte // main state of the hash + + // a[n:rate] is the buffer. If absorbing, it's the remaining space to XOR + // into before running the permutation. If squeezing, it's the remaining + // output to produce before running the permutation. + n, rate int // dsbyte contains the "domain separation" bits and the first bit of // the padding. Sections 6.1 and 6.2 of [1] separate the outputs of the @@ -39,10 +45,6 @@ type state struct { // Extendable-Output Functions (May 2014)" dsbyte byte - i, n int // storage[i:n] is the buffer, i is only used while squeezing - storage [maxRate]byte - - // Specific to SHA-3 and SHAKE. outputLen int // the default output size in bytes state spongeDirection // whether the sponge is absorbing or squeezing } @@ -61,7 +63,7 @@ func (d *state) Reset() { d.a[i] = 0 } d.state = spongeAbsorbing - d.i, d.n = 0, 0 + d.n = 0 } func (d *state) clone() *state { @@ -69,22 +71,25 @@ func (d *state) clone() *state { return &ret } -// permute applies the KeccakF-1600 permutation. It handles -// any input-output buffering. +// permute applies the KeccakF-1600 permutation. func (d *state) permute() { - switch d.state { - case spongeAbsorbing: - // If we're absorbing, we need to xor the input into the state - // before applying the permutation. - xorIn(d, d.storage[:d.rate]) - d.n = 0 - keccakF1600(&d.a) - case spongeSqueezing: - // If we're squeezing, we need to apply the permutation before - // copying more output. - keccakF1600(&d.a) - d.i = 0 - copyOut(d, d.storage[:d.rate]) + var a *[25]uint64 + if cpu.IsBigEndian { + a = new([25]uint64) + for i := range a { + a[i] = binary.LittleEndian.Uint64(d.a[i*8:]) + } + } else { + a = (*[25]uint64)(unsafe.Pointer(&d.a)) + } + + keccakF1600(a) + d.n = 0 + + if cpu.IsBigEndian { + for i := range a { + binary.LittleEndian.PutUint64(d.a[i*8:], a[i]) + } } } @@ -92,53 +97,36 @@ func (d *state) permute() { // the multi-bitrate 10..1 padding rule, and permutes the state. func (d *state) padAndPermute() { // Pad with this instance's domain-separator bits. We know that there's - // at least one byte of space in d.buf because, if it were full, + // at least one byte of space in the sponge because, if it were full, // permute would have been called to empty it. dsbyte also contains the // first one bit for the padding. See the comment in the state struct. - d.storage[d.n] = d.dsbyte - d.n++ - for d.n < d.rate { - d.storage[d.n] = 0 - d.n++ - } + d.a[d.n] ^= d.dsbyte // This adds the final one bit for the padding. Because of the way that // bits are numbered from the LSB upwards, the final bit is the MSB of // the last byte. - d.storage[d.rate-1] ^= 0x80 + d.a[d.rate-1] ^= 0x80 // Apply the permutation d.permute() d.state = spongeSqueezing - d.n = d.rate - copyOut(d, d.storage[:d.rate]) } // Write absorbs more data into the hash's state. It panics if any // output has already been read. -func (d *state) Write(p []byte) (written int, err error) { +func (d *state) Write(p []byte) (n int, err error) { if d.state != spongeAbsorbing { panic("sha3: Write after Read") } - written = len(p) + + n = len(p) for len(p) > 0 { - if d.n == 0 && len(p) >= d.rate { - // The fast path; absorb a full "rate" bytes of input and apply the permutation. - xorIn(d, p[:d.rate]) - p = p[d.rate:] - keccakF1600(&d.a) - } else { - // The slow path; buffer the input until we can fill the sponge, and then xor it in. - todo := d.rate - d.n - if todo > len(p) { - todo = len(p) - } - d.n += copy(d.storage[d.n:], p[:todo]) - p = p[todo:] - - // If the sponge is full, apply the permutation. - if d.n == d.rate { - d.permute() - } + x := subtle.XORBytes(d.a[d.n:d.rate], d.a[d.n:d.rate], p) + d.n += x + p = p[x:] + + // If the sponge is full, apply the permutation. + if d.n == d.rate { + d.permute() } } @@ -156,14 +144,14 @@ func (d *state) Read(out []byte) (n int, err error) { // Now, do the squeezing. for len(out) > 0 { - n := copy(out, d.storage[d.i:d.n]) - d.i += n - out = out[n:] - // Apply the permutation if we've squeezed the sponge dry. - if d.i == d.rate { + if d.n == d.rate { d.permute() } + + x := copy(out, d.a[d.n:d.rate]) + d.n += x + out = out[x:] } return @@ -183,3 +171,74 @@ func (d *state) Sum(in []byte) []byte { dup.Read(hash) return append(in, hash...) } + +const ( + magicSHA3 = "sha\x08" + magicShake = "sha\x09" + magicCShake = "sha\x0a" + magicKeccak = "sha\x0b" + // magic || rate || main state || n || sponge direction + marshaledSize = len(magicSHA3) + 1 + 200 + 1 + 1 +) + +func (d *state) MarshalBinary() ([]byte, error) { + return d.AppendBinary(make([]byte, 0, marshaledSize)) +} + +func (d *state) AppendBinary(b []byte) ([]byte, error) { + switch d.dsbyte { + case dsbyteSHA3: + b = append(b, magicSHA3...) + case dsbyteShake: + b = append(b, magicShake...) + case dsbyteCShake: + b = append(b, magicCShake...) + case dsbyteKeccak: + b = append(b, magicKeccak...) + default: + panic("unknown dsbyte") + } + // rate is at most 168, and n is at most rate. + b = append(b, byte(d.rate)) + b = append(b, d.a[:]...) + b = append(b, byte(d.n), byte(d.state)) + return b, nil +} + +func (d *state) UnmarshalBinary(b []byte) error { + if len(b) != marshaledSize { + return errors.New("sha3: invalid hash state") + } + + magic := string(b[:len(magicSHA3)]) + b = b[len(magicSHA3):] + switch { + case magic == magicSHA3 && d.dsbyte == dsbyteSHA3: + case magic == magicShake && d.dsbyte == dsbyteShake: + case magic == magicCShake && d.dsbyte == dsbyteCShake: + case magic == magicKeccak && d.dsbyte == dsbyteKeccak: + default: + return errors.New("sha3: invalid hash state identifier") + } + + rate := int(b[0]) + b = b[1:] + if rate != d.rate { + return errors.New("sha3: invalid hash state function") + } + + copy(d.a[:], b) + b = b[len(d.a):] + + n, state := int(b[0]), spongeDirection(b[1]) + if n > d.rate { + return errors.New("sha3: invalid hash state") + } + d.n = n + if state != spongeAbsorbing && state != spongeSqueezing { + return errors.New("sha3: invalid hash state") + } + d.state = state + + return nil +} diff --git a/vendor/golang.org/x/crypto/sha3/shake.go b/vendor/golang.org/x/crypto/sha3/shake.go index a01ef4357..a6b3a4281 100644 --- a/vendor/golang.org/x/crypto/sha3/shake.go +++ b/vendor/golang.org/x/crypto/sha3/shake.go @@ -16,9 +16,12 @@ package sha3 // [2] https://doi.org/10.6028/NIST.SP.800-185 import ( + "bytes" "encoding/binary" + "errors" "hash" "io" + "math/bits" ) // ShakeHash defines the interface to hash functions that support @@ -50,41 +53,33 @@ type cshakeState struct { initBlock []byte } -// Consts for configuring initial SHA-3 state -const ( - dsbyteShake = 0x1f - dsbyteCShake = 0x04 - rate128 = 168 - rate256 = 136 -) +func bytepad(data []byte, rate int) []byte { + out := make([]byte, 0, 9+len(data)+rate-1) + out = append(out, leftEncode(uint64(rate))...) + out = append(out, data...) + if padlen := rate - len(out)%rate; padlen < rate { + out = append(out, make([]byte, padlen)...) + } + return out +} -func bytepad(input []byte, w int) []byte { - // leftEncode always returns max 9 bytes - buf := make([]byte, 0, 9+len(input)+w) - buf = append(buf, leftEncode(uint64(w))...) - buf = append(buf, input...) - padlen := w - (len(buf) % w) - return append(buf, make([]byte, padlen)...) -} - -func leftEncode(value uint64) []byte { - var b [9]byte - binary.BigEndian.PutUint64(b[1:], value) - // Trim all but last leading zero bytes - i := byte(1) - for i < 8 && b[i] == 0 { - i++ +func leftEncode(x uint64) []byte { + // Let n be the smallest positive integer for which 2^(8n) > x. + n := (bits.Len64(x) + 7) / 8 + if n == 0 { + n = 1 } - // Prepend number of encoded bytes - b[i-1] = 9 - i - return b[i-1:] + // Return n || x with n as a byte and x an n bytes in big-endian order. + b := make([]byte, 9) + binary.BigEndian.PutUint64(b[1:], x) + b = b[9-n-1:] + b[0] = byte(n) + return b } func newCShake(N, S []byte, rate, outputLen int, dsbyte byte) ShakeHash { c := cshakeState{state: &state{rate: rate, outputLen: outputLen, dsbyte: dsbyte}} - - // leftEncode returns max 9 bytes - c.initBlock = make([]byte, 0, 9*2+len(N)+len(S)) + c.initBlock = make([]byte, 0, 9+len(N)+9+len(S)) // leftEncode returns max 9 bytes c.initBlock = append(c.initBlock, leftEncode(uint64(len(N))*8)...) c.initBlock = append(c.initBlock, N...) c.initBlock = append(c.initBlock, leftEncode(uint64(len(S))*8)...) @@ -111,6 +106,30 @@ func (c *state) Clone() ShakeHash { return c.clone() } +func (c *cshakeState) MarshalBinary() ([]byte, error) { + return c.AppendBinary(make([]byte, 0, marshaledSize+len(c.initBlock))) +} + +func (c *cshakeState) AppendBinary(b []byte) ([]byte, error) { + b, err := c.state.AppendBinary(b) + if err != nil { + return nil, err + } + b = append(b, c.initBlock...) + return b, nil +} + +func (c *cshakeState) UnmarshalBinary(b []byte) error { + if len(b) <= marshaledSize { + return errors.New("sha3: invalid hash state") + } + if err := c.state.UnmarshalBinary(b[:marshaledSize]); err != nil { + return err + } + c.initBlock = bytes.Clone(b[marshaledSize:]) + return nil +} + // NewShake128 creates a new SHAKE128 variable-output-length ShakeHash. // Its generic security strength is 128 bits against all attacks if at // least 32 bytes of its output are used. @@ -126,11 +145,11 @@ func NewShake256() ShakeHash { } func newShake128Generic() *state { - return &state{rate: rate128, outputLen: 32, dsbyte: dsbyteShake} + return &state{rate: rateK256, outputLen: 32, dsbyte: dsbyteShake} } func newShake256Generic() *state { - return &state{rate: rate256, outputLen: 64, dsbyte: dsbyteShake} + return &state{rate: rateK512, outputLen: 64, dsbyte: dsbyteShake} } // NewCShake128 creates a new instance of cSHAKE128 variable-output-length ShakeHash, @@ -143,7 +162,7 @@ func NewCShake128(N, S []byte) ShakeHash { if len(N) == 0 && len(S) == 0 { return NewShake128() } - return newCShake(N, S, rate128, 32, dsbyteCShake) + return newCShake(N, S, rateK256, 32, dsbyteCShake) } // NewCShake256 creates a new instance of cSHAKE256 variable-output-length ShakeHash, @@ -156,7 +175,7 @@ func NewCShake256(N, S []byte) ShakeHash { if len(N) == 0 && len(S) == 0 { return NewShake256() } - return newCShake(N, S, rate256, 64, dsbyteCShake) + return newCShake(N, S, rateK512, 64, dsbyteCShake) } // ShakeSum128 writes an arbitrary-length digest of data into hash. diff --git a/vendor/golang.org/x/crypto/sha3/xor.go b/vendor/golang.org/x/crypto/sha3/xor.go deleted file mode 100644 index 6ada5c957..000000000 --- a/vendor/golang.org/x/crypto/sha3/xor.go +++ /dev/null @@ -1,40 +0,0 @@ -// Copyright 2015 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 sha3 - -import ( - "crypto/subtle" - "encoding/binary" - "unsafe" - - "golang.org/x/sys/cpu" -) - -// xorIn xors the bytes in buf into the state. -func xorIn(d *state, buf []byte) { - if cpu.IsBigEndian { - for i := 0; len(buf) >= 8; i++ { - a := binary.LittleEndian.Uint64(buf) - d.a[i] ^= a - buf = buf[8:] - } - } else { - ab := (*[25 * 64 / 8]byte)(unsafe.Pointer(&d.a)) - subtle.XORBytes(ab[:], ab[:], buf) - } -} - -// copyOut copies uint64s to a byte buffer. -func copyOut(d *state, b []byte) { - if cpu.IsBigEndian { - for i := 0; len(b) >= 8; i++ { - binary.LittleEndian.PutUint64(b, d.a[i]) - b = b[8:] - } - } else { - ab := (*[25 * 64 / 8]byte)(unsafe.Pointer(&d.a)) - copy(b, ab[:]) - } -} diff --git a/vendor/golang.org/x/crypto/ssh/client_auth.go b/vendor/golang.org/x/crypto/ssh/client_auth.go index b93961010..b86dde151 100644 --- a/vendor/golang.org/x/crypto/ssh/client_auth.go +++ b/vendor/golang.org/x/crypto/ssh/client_auth.go @@ -555,6 +555,7 @@ func (cb KeyboardInteractiveChallenge) auth(session []byte, user string, c packe } gotMsgExtInfo := false + gotUserAuthInfoRequest := false for { packet, err := c.readPacket() if err != nil { @@ -585,6 +586,9 @@ func (cb KeyboardInteractiveChallenge) auth(session []byte, user string, c packe if msg.PartialSuccess { return authPartialSuccess, msg.Methods, nil } + if !gotUserAuthInfoRequest { + return authFailure, msg.Methods, unexpectedMessageError(msgUserAuthInfoRequest, packet[0]) + } return authFailure, msg.Methods, nil case msgUserAuthSuccess: return authSuccess, nil, nil @@ -596,6 +600,7 @@ func (cb KeyboardInteractiveChallenge) auth(session []byte, user string, c packe if err := Unmarshal(packet, &msg); err != nil { return authFailure, nil, err } + gotUserAuthInfoRequest = true // Manually unpack the prompt/echo pairs. rest := msg.Prompts -- cgit v1.2.3