From 9bf156bb9ee4915fa7e8f84589d65f6598563c97 Mon Sep 17 00:00:00 2001
From: Pieter Wuille <pieter.wuille@gmail.com>
Date: Sun, 22 May 2016 11:06:18 +0200
Subject: [PATCH] Support SipHash with arbitrary byte writes

---
 src/hash.cpp            | 39 ++++++++++++++++++++++++++++++++++++---
 src/hash.h              | 21 ++++++++++++++++++++-
 src/test/hash_tests.cpp | 17 +++++++++++++----
 3 files changed, 69 insertions(+), 8 deletions(-)

diff --git a/src/hash.cpp b/src/hash.cpp
index a518314a5..20a83342d 100644
--- a/src/hash.cpp
+++ b/src/hash.cpp
@@ -100,12 +100,15 @@ CSipHasher::CSipHasher(uint64_t k0, uint64_t k1)
     v[2] = 0x6c7967656e657261ULL ^ k0;
     v[3] = 0x7465646279746573ULL ^ k1;
     count = 0;
+    tmp = 0;
 }
 
 CSipHasher& CSipHasher::Write(uint64_t data)
 {
     uint64_t v0 = v[0], v1 = v[1], v2 = v[2], v3 = v[3];
 
+    assert(count % 8 == 0);
+
     v3 ^= data;
     SIPROUND;
     SIPROUND;
@@ -116,7 +119,35 @@ CSipHasher& CSipHasher::Write(uint64_t data)
     v[2] = v2;
     v[3] = v3;
 
-    count++;
+    count += 8;
+    return *this;
+}
+
+CSipHasher& CSipHasher::Write(const unsigned char* data, size_t size)
+{
+    uint64_t v0 = v[0], v1 = v[1], v2 = v[2], v3 = v[3];
+    uint64_t t = tmp;
+    int c = count;
+
+    while (size--) {
+        t |= ((uint64_t)(*(data++))) << (8 * (c % 8));
+        c++;
+        if ((c & 7) == 0) {
+            v3 ^= t;
+            SIPROUND;
+            SIPROUND;
+            v0 ^= t;
+            t = 0;
+        }
+    }
+
+    v[0] = v0;
+    v[1] = v1;
+    v[2] = v2;
+    v[3] = v3;
+    count = c;
+    tmp = t;
+
     return *this;
 }
 
@@ -124,10 +155,12 @@ uint64_t CSipHasher::Finalize() const
 {
     uint64_t v0 = v[0], v1 = v[1], v2 = v[2], v3 = v[3];
 
-    v3 ^= ((uint64_t)count) << 59;
+    uint64_t t = tmp | (((uint64_t)count) << 56);
+
+    v3 ^= t;
     SIPROUND;
     SIPROUND;
-    v0 ^= ((uint64_t)count) << 59;
+    v0 ^= t;
     v2 ^= 0xFF;
     SIPROUND;
     SIPROUND;
diff --git a/src/hash.h b/src/hash.h
index 600dabec5..db4e130ae 100644
--- a/src/hash.h
+++ b/src/hash.h
@@ -171,19 +171,38 @@ unsigned int MurmurHash3(unsigned int nHashSeed, const std::vector<unsigned char
 
 void BIP32Hash(const ChainCode &chainCode, unsigned int nChild, unsigned char header, const unsigned char data[32], unsigned char output[64]);
 
-/** SipHash-2-4, using a uint64_t-based (rather than byte-based) interface */
+/** SipHash-2-4 */
 class CSipHasher
 {
 private:
     uint64_t v[4];
+    uint64_t tmp;
     int count;
 
 public:
+    /** Construct a SipHash calculator initialized with 128-bit key (k0, k1) */
     CSipHasher(uint64_t k0, uint64_t k1);
+    /** Hash a 64-bit integer worth of data
+     *  It is treated as if this was the little-endian interpretation of 8 bytes.
+     *  This function can only be used when a multiple of 8 bytes have been written so far.
+     */
     CSipHasher& Write(uint64_t data);
+    /** Hash arbitrary bytes. */
+    CSipHasher& Write(const unsigned char* data, size_t size);
+    /** Compute the 64-bit SipHash-2-4 of the data written so far. The object remains untouched. */
     uint64_t Finalize() const;
 };
 
+/** Optimized SipHash-2-4 implementation for uint256.
+ *
+ *  It is identical to:
+ *    SipHasher(k0, k1)
+ *      .Write(val.GetUint64(0))
+ *      .Write(val.GetUint64(1))
+ *      .Write(val.GetUint64(2))
+ *      .Write(val.GetUint64(3))
+ *      .Finalize()
+ */
 uint64_t SipHashUint256(uint64_t k0, uint64_t k1, const uint256& val);
 
 #endif // BITCOIN_HASH_H
diff --git a/src/test/hash_tests.cpp b/src/test/hash_tests.cpp
index 8baaf3645..f62080c1d 100644
--- a/src/test/hash_tests.cpp
+++ b/src/test/hash_tests.cpp
@@ -51,13 +51,22 @@ BOOST_AUTO_TEST_CASE(siphash)
 {
     CSipHasher hasher(0x0706050403020100ULL, 0x0F0E0D0C0B0A0908ULL);
     BOOST_CHECK_EQUAL(hasher.Finalize(),  0x726fdb47dd0e0e31ull);
-    hasher.Write(0x0706050403020100ULL);
+    static const unsigned char t0[1] = {0};
+    hasher.Write(t0, 1);
+    BOOST_CHECK_EQUAL(hasher.Finalize(),  0x74f839c593dc67fdull);
+    static const unsigned char t1[7] = {1,2,3,4,5,6,7};
+    hasher.Write(t1, 7);
     BOOST_CHECK_EQUAL(hasher.Finalize(),  0x93f5f5799a932462ull);
     hasher.Write(0x0F0E0D0C0B0A0908ULL);
     BOOST_CHECK_EQUAL(hasher.Finalize(),  0x3f2acc7f57c29bdbull);
-    hasher.Write(0x1716151413121110ULL);
-    BOOST_CHECK_EQUAL(hasher.Finalize(),  0xb8ad50c6f649af94ull);
-    hasher.Write(0x1F1E1D1C1B1A1918ULL);
+    static const unsigned char t2[2] = {16,17};
+    hasher.Write(t2, 2);
+    BOOST_CHECK_EQUAL(hasher.Finalize(),  0x4bc1b3f0968dd39cull);
+    static const unsigned char t3[9] = {18,19,20,21,22,23,24,25,26};
+    hasher.Write(t3, 9);
+    BOOST_CHECK_EQUAL(hasher.Finalize(),  0x2f2e6163076bcfadull);
+    static const unsigned char t4[5] = {27,28,29,30,31};
+    hasher.Write(t4, 5);
     BOOST_CHECK_EQUAL(hasher.Finalize(),  0x7127512f72f27cceull);
     hasher.Write(0x2726252423222120ULL);
     BOOST_CHECK_EQUAL(hasher.Finalize(),  0x0e3ea96b5304a7d0ull);