tailscale/util/hashx/block512.go
Will Norris 3ec5be3f51 all: remove AUTHORS file and references to it
This file was never truly necessary and has never actually been used in
the history of Tailscale's open source releases.

A Brief History of AUTHORS files
---

The AUTHORS file was a pattern developed at Google, originally for
Chromium, then adopted by Go and a bunch of other projects. The problem
was that Chromium originally had a copyright line only recognizing
Google as the copyright holder. Because Google (and most open source
projects) do not require copyright assignemnt for contributions, each
contributor maintains their copyright. Some large corporate contributors
then tried to add their own name to the copyright line in the LICENSE
file or in file headers. This quickly becomes unwieldy, and puts a
tremendous burden on anyone building on top of Chromium, since the
license requires that they keep all copyright lines intact.

The compromise was to create an AUTHORS file that would list all of the
copyright holders. The LICENSE file and source file headers would then
include that list by reference, listing the copyright holder as "The
Chromium Authors".

This also become cumbersome to simply keep the file up to date with a
high rate of new contributors. Plus it's not always obvious who the
copyright holder is. Sometimes it is the individual making the
contribution, but many times it may be their employer. There is no way
for the proejct maintainer to know.

Eventually, Google changed their policy to no longer recommend trying to
keep the AUTHORS file up to date proactively, and instead to only add to
it when requested: https://opensource.google/docs/releasing/authors.
They are also clear that:

> Adding contributors to the AUTHORS file is entirely within the
> project's discretion and has no implications for copyright ownership.

It was primarily added to appease a small number of large contributors
that insisted that they be recognized as copyright holders (which was
entirely their right to do). But it's not truly necessary, and not even
the most accurate way of identifying contributors and/or copyright
holders.

In practice, we've never added anyone to our AUTHORS file. It only lists
Tailscale, so it's not really serving any purpose. It also causes
confusion because Tailscalars put the "Tailscale Inc & AUTHORS" header
in other open source repos which don't actually have an AUTHORS file, so
it's ambiguous what that means.

Instead, we just acknowledge that the contributors to Tailscale (whoever
they are) are copyright holders for their individual contributions. We
also have the benefit of using the DCO (developercertificate.org) which
provides some additional certification of their right to make the
contribution.

The source file changes were purely mechanical with:

    git ls-files | xargs sed -i -e 's/\(Tailscale Inc &\) AUTHORS/\1 contributors/g'

Updates #cleanup

Change-Id: Ia101a4a3005adb9118051b3416f5a64a4a45987d
Signed-off-by: Will Norris <will@tailscale.com>
2026-01-23 15:49:45 -08:00

198 lines
5.4 KiB
Go

// Copyright (c) Tailscale Inc & contributors
// SPDX-License-Identifier: BSD-3-Clause
// Package hashx provides a concrete implementation of [hash.Hash]
// that operates on a particular block size.
package hashx
import (
"encoding/binary"
"fmt"
"hash"
"unsafe"
)
var _ hash.Hash = (*Block512)(nil)
// Block512 wraps a [hash.Hash] for functions that operate on 512-bit block sizes.
// It has efficient methods for hashing fixed-width integers.
//
// A hashing algorithm that operates on 512-bit block sizes should be used.
// The hash still operates correctly even with misaligned block sizes,
// but operates less efficiently.
//
// Example algorithms with 512-bit block sizes include:
// - MD4 (https://golang.org/x/crypto/md4)
// - MD5 (https://golang.org/pkg/crypto/md5)
// - BLAKE2s (https://golang.org/x/crypto/blake2s)
// - BLAKE3
// - RIPEMD (https://golang.org/x/crypto/ripemd160)
// - SHA-0
// - SHA-1 (https://golang.org/pkg/crypto/sha1)
// - SHA-2 (https://golang.org/pkg/crypto/sha256)
// - Whirlpool
//
// See https://en.wikipedia.org/wiki/Comparison_of_cryptographic_hash_functions#Parameters
// for a list of hash functions and their block sizes.
//
// Block512 assumes that [hash.Hash.Write] never fails and
// never allows the provided buffer to escape.
type Block512 struct {
hash.Hash
x [512 / 8]byte
nx int
}
// New512 constructs a new Block512 that wraps h.
//
// It reports an error if the block sizes do not match.
// Misaligned block sizes perform poorly, but execute correctly.
// The error may be ignored if performance is not a concern.
func New512(h hash.Hash) (*Block512, error) {
b := &Block512{Hash: h}
if len(b.x)%h.BlockSize() != 0 {
return b, fmt.Errorf("hashx.Block512: inefficient use of hash.Hash with %d-bit block size", 8*h.BlockSize())
}
return b, nil
}
// Write hashes the contents of b.
func (h *Block512) Write(b []byte) (int, error) {
h.HashBytes(b)
return len(b), nil
}
// Sum appends the current hash to b and returns the resulting slice.
//
// It flushes any partially completed blocks to the underlying [hash.Hash],
// which may cause future operations to be misaligned and less efficient
// until [Block512.Reset] is called.
func (h *Block512) Sum(b []byte) []byte {
if h.nx > 0 {
h.Hash.Write(h.x[:h.nx])
h.nx = 0
}
// Unfortunately hash.Hash.Sum always causes the input to escape since
// escape analysis cannot prove anything past an interface method call.
// Assuming h already escapes, we call Sum with h.x first,
// and then copy the result to b.
sum := h.Hash.Sum(h.x[:0])
return append(b, sum...)
}
// Reset resets Block512 to its initial state.
// It recursively resets the underlying [hash.Hash].
func (h *Block512) Reset() {
h.Hash.Reset()
h.nx = 0
}
// HashUint8 hashes n as a 1-byte integer.
func (h *Block512) HashUint8(n uint8) {
// NOTE: This method is carefully written to be inlineable.
if h.nx <= len(h.x)-1 {
h.x[h.nx] = n
h.nx += 1
} else {
h.hashUint8Slow(n) // mark "noinline" to keep this within inline budget
}
}
//go:noinline
func (h *Block512) hashUint8Slow(n uint8) { h.hashUint(uint64(n), 1) }
// HashUint16 hashes n as a 2-byte little-endian integer.
func (h *Block512) HashUint16(n uint16) {
// NOTE: This method is carefully written to be inlineable.
if h.nx <= len(h.x)-2 {
binary.LittleEndian.PutUint16(h.x[h.nx:], n)
h.nx += 2
} else {
h.hashUint16Slow(n) // mark "noinline" to keep this within inline budget
}
}
//go:noinline
func (h *Block512) hashUint16Slow(n uint16) { h.hashUint(uint64(n), 2) }
// HashUint32 hashes n as a 4-byte little-endian integer.
func (h *Block512) HashUint32(n uint32) {
// NOTE: This method is carefully written to be inlineable.
if h.nx <= len(h.x)-4 {
binary.LittleEndian.PutUint32(h.x[h.nx:], n)
h.nx += 4
} else {
h.hashUint32Slow(n) // mark "noinline" to keep this within inline budget
}
}
//go:noinline
func (h *Block512) hashUint32Slow(n uint32) { h.hashUint(uint64(n), 4) }
// HashUint64 hashes n as a 8-byte little-endian integer.
func (h *Block512) HashUint64(n uint64) {
// NOTE: This method is carefully written to be inlineable.
if h.nx <= len(h.x)-8 {
binary.LittleEndian.PutUint64(h.x[h.nx:], n)
h.nx += 8
} else {
h.hashUint64Slow(n) // mark "noinline" to keep this within inline budget
}
}
//go:noinline
func (h *Block512) hashUint64Slow(n uint64) { h.hashUint(uint64(n), 8) }
func (h *Block512) hashUint(n uint64, i int) {
for ; i > 0; i-- {
if h.nx == len(h.x) {
h.Hash.Write(h.x[:])
h.nx = 0
}
h.x[h.nx] = byte(n)
h.nx += 1
n >>= 8
}
}
// HashBytes hashes the contents of b.
// It does not explicitly hash the length separately.
func (h *Block512) HashBytes(b []byte) {
// Nearly identical to sha256.digest.Write.
if h.nx > 0 {
n := copy(h.x[h.nx:], b)
h.nx += n
if h.nx == len(h.x) {
h.Hash.Write(h.x[:])
h.nx = 0
}
b = b[n:]
}
if len(b) >= len(h.x) {
n := len(b) &^ (len(h.x) - 1) // n is a multiple of len(h.x)
h.Hash.Write(b[:n])
b = b[n:]
}
if len(b) > 0 {
h.nx = copy(h.x[:], b)
}
}
// HashString hashes the contents of s.
// It does not explicitly hash the length separately.
func (h *Block512) HashString(s string) {
// TODO: Avoid unsafe when standard hashers implement io.StringWriter.
// See https://go.dev/issue/38776.
type stringHeader struct {
p unsafe.Pointer
n int
}
p := (*stringHeader)(unsafe.Pointer(&s))
b := unsafe.Slice((*byte)(p.p), p.n)
h.HashBytes(b)
}
// TODO: Add Hash.MarshalBinary and Hash.UnmarshalBinary?