compressor: use a prevector in compressed script serialization

Use a prevector for stack allocation instead of heap allocation during script compression and decompression. These functions were doing millions of unnecessary heap allocations during IBD. We introduce a CompressedScript type alias for this prevector. It is size 33 as that is the maximum size of a compressed script. Fix the DecompressScript header to match the variable name from compressor.cpp Signed-off-by: William Casarin <jb55@jb55.com>
bitcoin · jb55 · Mar 29, 2020 · Mar 29, 2020 · Mar 30, 2020 · Mar 30, 2020
commit 83a425d25af033086744c1c8c892015014ed46bd
diff --git a/src/compressor.cpp b/src/compressor.cpp
@@ -52,7 +52,7 @@ static bool IsToPubKey(const CScript& script, CPubKey &pubkey)
     return false;
 }
 
-bool CompressScript(const CScript& script, std::vector<unsigned char> &out)
+bool CompressScript(const CScript& script, CompressedScript& out)
 {
     CKeyID keyID;
     if (IsToKeyID(script, keyID)) {
@@ -92,7 +92,7 @@ unsigned int GetSpecialScriptSize(unsigned int nSize)
     return 0;
 }
 
-bool DecompressScript(CScript& script, unsigned int nSize, const std::vector<unsigned char> &in)
+bool DecompressScript(CScript& script, unsigned int nSize, const CompressedScript& in)
 {
     switch(nSize) {
     case 0x00:

diff --git a/src/compressor.h b/src/compressor.h
@@ -6,14 +6,26 @@
 #ifndef BITCOIN_COMPRESSOR_H
 #define BITCOIN_COMPRESSOR_H
 
+#include <prevector.h>
 #include <primitives/transaction.h>
 #include <script/script.h>
 #include <serialize.h>
 #include <span.h>
 
-bool CompressScript(const CScript& script, std::vector<unsigned char> &out);
+/**
+ * This saves us from making many heap allocations when serializing
+ * and deserializing compressed scripts.
+ *
+ * This prevector size is determined by the largest .resize() in the
+ * CompressScript function. The largest compressed script format is a
+ * compressed public key, which is 33 bytes.
+ */
+using CompressedScript = prevector<33, unsigned char>;
+
+
+bool CompressScript(const CScript& script, CompressedScript& out);
 unsigned int GetSpecialScriptSize(unsigned int nSize);
-bool DecompressScript(CScript& script, unsigned int nSize, const std::vector<unsigned char> &out);
+bool DecompressScript(CScript& script, unsigned int nSize, const CompressedScript& in);
 
 /**
  * Compress amount.
@@ -51,7 +63,7 @@ struct ScriptCompression
 
     template<typename Stream>
     void Ser(Stream &s, const CScript& script) {
-        std::vector<unsigned char> compr;
+        CompressedScript compr;
         if (CompressScript(script, compr)) {
             s << MakeSpan(compr);
             return;
@@ -66,7 +78,7 @@ struct ScriptCompression
         unsigned int nSize = 0;
         s >> VARINT(nSize);
         if (nSize < nSpecialScripts) {
-            std::vector<unsigned char> vch(GetSpecialScriptSize(nSize), 0x00);
+            CompressedScript vch(GetSpecialScriptSize(nSize), 0x00);
             s >> MakeSpan(vch);
             DecompressScript(script, nSize, vch);
             return;

diff --git a/src/test/compress_tests.cpp b/src/test/compress_tests.cpp
@@ -72,7 +72,7 @@ BOOST_AUTO_TEST_CASE(compress_script_to_ckey_id)
     CScript script = CScript() << OP_DUP << OP_HASH160 << ToByteVector(pubkey.GetID()) << OP_EQUALVERIFY << OP_CHECKSIG;
     BOOST_CHECK_EQUAL(script.size(), 25);
 
-    std::vector<unsigned char> out;
+    CompressedScript out;
     bool done = CompressScript(script, out);
     BOOST_CHECK_EQUAL(done, true);
 
@@ -89,7 +89,7 @@ BOOST_AUTO_TEST_CASE(compress_script_to_cscript_id)
     script << OP_HASH160 << ToByteVector(CScriptID(redeemScript)) << OP_EQUAL;
     BOOST_CHECK_EQUAL(script.size(), 23);
 
-    std::vector<unsigned char> out;
+    CompressedScript out;
     bool done = CompressScript(script, out);
     BOOST_CHECK_EQUAL(done, true);
 
@@ -107,7 +107,7 @@ BOOST_AUTO_TEST_CASE(compress_script_to_compressed_pubkey_id)
     CScript script = CScript() << ToByteVector(key.GetPubKey()) << OP_CHECKSIG; // COMPRESSED_PUBLIC_KEY_SIZE (33)
     BOOST_CHECK_EQUAL(script.size(), 35);
 
-    std::vector<unsigned char> out;
+    CompressedScript out;
     bool done = CompressScript(script, out);
     BOOST_CHECK_EQUAL(done, true);
 
@@ -124,7 +124,7 @@ BOOST_AUTO_TEST_CASE(compress_script_to_uncompressed_pubkey_id)
     CScript script =  CScript() << ToByteVector(key.GetPubKey()) << OP_CHECKSIG; // PUBLIC_KEY_SIZE (65)
     BOOST_CHECK_EQUAL(script.size(), 67);                   // 1 char code + 65 char pubkey + OP_CHECKSIG
 
-    std::vector<unsigned char> out;
+    CompressedScript out;
     bool done = CompressScript(script, out);
     BOOST_CHECK_EQUAL(done, true);
 

diff --git a/src/test/fuzz/script.cpp b/src/test/fuzz/script.cpp
@@ -36,7 +36,7 @@ void test_one_input(const std::vector<uint8_t>& buffer)
     if (!script_opt) return;
     const CScript script{*script_opt};
 
-    std::vector<unsigned char> compressed;
+    CompressedScript compressed;
     if (CompressScript(script, compressed)) {
         const unsigned int size = compressed[0];
         compressed.erase(compressed.begin());
@@ -94,10 +94,12 @@ void test_one_input(const std::vector<uint8_t>& buffer)
 
     {
         const std::vector<uint8_t> bytes = ConsumeRandomLengthByteVector(fuzzed_data_provider);
+        CompressedScript compressed_script;
+        compressed_script.assign(bytes.begin(), bytes.end());
         // DecompressScript(..., ..., bytes) is not guaranteed to be defined if the bytes vector is too short
-        if (bytes.size() >= 32) {
+        if (compressed_script.size() >= 32) {
             CScript decompressed_script;
-            DecompressScript(decompressed_script, fuzzed_data_provider.ConsumeIntegral<unsigned int>(), bytes);
+            DecompressScript(decompressed_script, fuzzed_data_provider.ConsumeIntegral<unsigned int>(), compressed_script);
         }
     }