Merge bitcoin/bitcoin#28815: fuzz: Avoid timeout and bloat in fuzz targets

fabb5046a7365af3079e6e45606d63576bc6ad12 fuzz: Avoid timeout and bloat in fuzz targets (MarcoFalke)

Pull request description:

  If the fuzz input contains invalid data *in a loop*, abort early. This will teach the fuzz engine to look for useful data and avoids bloating the fuzz input folder with useless (repeated) data.

ACKs for top commit:
  dergoegge:
    utACK fabb5046a7365af3079e6e45606d63576bc6ad12
  brunoerg:
    crACK fabb5046a7365af3079e6e45606d63576bc6ad12

Tree-SHA512: 26da100d7558ae6fdd5292fb146d8858b2af8f78c546ca2509b9d27b33a33e9462ecb6035de142f9f36dd5de32f8cbad099d6c7a697902d23e1bb621cd27dc88

5 files changed, 60 insertions, 28 deletions

diff --git a/src/test/fuzz/bloom_filter.cpp b/src/test/fuzz/bloom_filter.cpp
index 3e303ecc0f..612f113b14 100644
--- a/src/test/fuzz/bloom_filter.cpp
+++ b/src/test/fuzz/bloom_filter.cpp
@@ -10,21 +10,22 @@
 #include <uint256.h>
 
 #include <cassert>
-#include <cstdint>
+#include <limits>
 #include <optional>
-#include <string>
 #include <vector>
 
 FUZZ_TARGET(bloom_filter)
 {
     FuzzedDataProvider fuzzed_data_provider(buffer.data(), buffer.size());
+    bool good_data{true};
 
     CBloomFilter bloom_filter{
         fuzzed_data_provider.ConsumeIntegralInRange<unsigned int>(1, 10000000),
         1.0 / fuzzed_data_provider.ConsumeIntegralInRange<unsigned int>(1, std::numeric_limits<unsigned int>::max()),
         fuzzed_data_provider.ConsumeIntegral<unsigned int>(),
         static_cast<unsigned char>(fuzzed_data_provider.PickValueInArray({BLOOM_UPDATE_NONE, BLOOM_UPDATE_ALL, BLOOM_UPDATE_P2PUBKEY_ONLY, BLOOM_UPDATE_MASK}))};
-    LIMITED_WHILE(fuzzed_data_provider.remaining_bytes() > 0, 10000) {
+    LIMITED_WHILE(good_data && fuzzed_data_provider.remaining_bytes() > 0, 10'000)
+    {
         CallOneOf(
             fuzzed_data_provider,
             [&] {
@@ -37,6 +38,7 @@ FUZZ_TARGET(bloom_filter)
             [&] {
                 const std::optional<COutPoint> out_point = ConsumeDeserializable<COutPoint>(fuzzed_data_provider);
                 if (!out_point) {
+                    good_data = false;
                     return;
                 }
                 (void)bloom_filter.contains(*out_point);
@@ -47,6 +49,7 @@ FUZZ_TARGET(bloom_filter)
             [&] {
                 const std::optional<uint256> u256 = ConsumeDeserializable<uint256>(fuzzed_data_provider);
                 if (!u256) {
+                    good_data = false;
                     return;
                 }
                 (void)bloom_filter.contains(*u256);
@@ -57,6 +60,7 @@ FUZZ_TARGET(bloom_filter)
             [&] {
                 const std::optional<CMutableTransaction> mut_tx = ConsumeDeserializable<CMutableTransaction>(fuzzed_data_provider);
                 if (!mut_tx) {
+                    good_data = false;
                     return;
                 }
                 const CTransaction tx{*mut_tx};
diff --git a/src/test/fuzz/coins_view.cpp b/src/test/fuzz/coins_view.cpp
index b088aa0bd7..1a8570ee8f 100644
--- a/src/test/fuzz/coins_view.cpp
+++ b/src/test/fuzz/coins_view.cpp
@@ -2,26 +2,28 @@
 // Distributed under the MIT software license, see the accompanying
 // file COPYING or http://www.opensource.org/licenses/mit-license.php.
 
-#include <chainparams.h>
 #include <coins.h>
 #include <consensus/amount.h>
 #include <consensus/tx_check.h>
 #include <consensus/tx_verify.h>
 #include <consensus/validation.h>
-#include <key.h>
 #include <policy/policy.h>
 #include <primitives/transaction.h>
-#include <pubkey.h>
+#include <script/interpreter.h>
 #include <test/fuzz/FuzzedDataProvider.h>
 #include <test/fuzz/fuzz.h>
 #include <test/fuzz/util.h>
 #include <test/util/setup_common.h>
-#include <validation.h>
+#include <util/hasher.h>
 
+#include <cassert>
 #include <cstdint>
 #include <limits>
+#include <memory>
 #include <optional>
+#include <stdexcept>
 #include <string>
+#include <utility>
 #include <vector>
 
 namespace {
@@ -44,12 +46,15 @@ void initialize_coins_view()
 FUZZ_TARGET(coins_view, .init = initialize_coins_view)
 {
     FuzzedDataProvider fuzzed_data_provider{buffer.data(), buffer.size()};
+    bool good_data{true};
+
     CCoinsView backend_coins_view;
     CCoinsViewCache coins_view_cache{&backend_coins_view, /*deterministic=*/true};
     COutPoint random_out_point;
     Coin random_coin;
     CMutableTransaction random_mutable_transaction;
-    LIMITED_WHILE(fuzzed_data_provider.ConsumeBool(), 10000) {
+    LIMITED_WHILE(good_data && fuzzed_data_provider.ConsumeBool(), 10'000)
+    {
         CallOneOf(
             fuzzed_data_provider,
             [&] {
@@ -95,6 +100,7 @@ FUZZ_TARGET(coins_view, .init = initialize_coins_view)
             [&] {
                 const std::optional<COutPoint> opt_out_point = ConsumeDeserializable<COutPoint>(fuzzed_data_provider);
                 if (!opt_out_point) {
+                    good_data = false;
                     return;
                 }
                 random_out_point = *opt_out_point;
@@ -102,6 +108,7 @@ FUZZ_TARGET(coins_view, .init = initialize_coins_view)
             [&] {
                 const std::optional<Coin> opt_coin = ConsumeDeserializable<Coin>(fuzzed_data_provider);
                 if (!opt_coin) {
+                    good_data = false;
                     return;
                 }
                 random_coin = *opt_coin;
@@ -109,6 +116,7 @@ FUZZ_TARGET(coins_view, .init = initialize_coins_view)
             [&] {
                 const std::optional<CMutableTransaction> opt_mutable_transaction = ConsumeDeserializable<CMutableTransaction>(fuzzed_data_provider);
                 if (!opt_mutable_transaction) {
+                    good_data = false;
                     return;
                 }
                 random_mutable_transaction = *opt_mutable_transaction;
@@ -116,7 +124,8 @@ FUZZ_TARGET(coins_view, .init = initialize_coins_view)
             [&] {
                 CCoinsMapMemoryResource resource;
                 CCoinsMap coins_map{0, SaltedOutpointHasher{/*deterministic=*/true}, CCoinsMap::key_equal{}, &resource};
-                LIMITED_WHILE(fuzzed_data_provider.ConsumeBool(), 10000) {
+                LIMITED_WHILE(good_data && fuzzed_data_provider.ConsumeBool(), 10'000)
+                {
                     CCoinsCacheEntry coins_cache_entry;
                     coins_cache_entry.flags = fuzzed_data_provider.ConsumeIntegral<unsigned char>();
                     if (fuzzed_data_provider.ConsumeBool()) {
@@ -124,6 +133,7 @@ FUZZ_TARGET(coins_view, .init = initialize_coins_view)
                     } else {
                         const std::optional<Coin> opt_coin = ConsumeDeserializable<Coin>(fuzzed_data_provider);
                         if (!opt_coin) {
+                            good_data = false;
                             return;
                         }
                         coins_cache_entry.coin = *opt_coin;
diff --git a/src/test/fuzz/fuzz.h b/src/test/fuzz/fuzz.h
index 0534f9bcf1..1f0fa5527a 100644
--- a/src/test/fuzz/fuzz.h
+++ b/src/test/fuzz/fuzz.h
@@ -14,6 +14,11 @@
 /**
  * Can be used to limit a theoretically unbounded loop. This caps the runtime
  * to avoid timeouts or OOMs.
+ *
+ * This can be used in combination with a check in the condition to confirm
+ * whether the fuzz engine provided "good" data. If the fuzz input contains
+ * invalid data, the loop aborts early. This will teach the fuzz engine to look
+ * for useful data and avoids bloating the fuzz input folder with useless data.
  */
 #define LIMITED_WHILE(condition, limit) \
     for (unsigned _count{limit}; (condition) && _count; --_count)
diff --git a/src/test/fuzz/policy_estimator.cpp b/src/test/fuzz/policy_estimator.cpp
index 227ee9d2c4..4cbc1b4820 100644
--- a/src/test/fuzz/policy_estimator.cpp
+++ b/src/test/fuzz/policy_estimator.cpp
@@ -6,16 +6,14 @@
 #include <policy/fees.h>
 #include <policy/fees_args.h>
 #include <primitives/transaction.h>
+#include <streams.h>
 #include <test/fuzz/FuzzedDataProvider.h>
 #include <test/fuzz/fuzz.h>
 #include <test/fuzz/util.h>
 #include <test/fuzz/util/mempool.h>
 #include <test/util/setup_common.h>
-#include <txmempool.h>
 
-#include <cstdint>
 #include <optional>
-#include <string>
 #include <vector>
 
 namespace {
@@ -31,13 +29,17 @@ void initialize_policy_estimator()
 FUZZ_TARGET(policy_estimator, .init = initialize_policy_estimator)
 {
     FuzzedDataProvider fuzzed_data_provider(buffer.data(), buffer.size());
+    bool good_data{true};
+
     CBlockPolicyEstimator block_policy_estimator{FeeestPath(*g_setup->m_node.args), DEFAULT_ACCEPT_STALE_FEE_ESTIMATES};
-    LIMITED_WHILE(fuzzed_data_provider.ConsumeBool(), 10000) {
+    LIMITED_WHILE(good_data && fuzzed_data_provider.ConsumeBool(), 10'000)
+    {
         CallOneOf(
             fuzzed_data_provider,
             [&] {
                 const std::optional<CMutableTransaction> mtx = ConsumeDeserializable<CMutableTransaction>(fuzzed_data_provider);
                 if (!mtx) {
+                    good_data = false;
                     return;
                 }
                 const CTransaction tx{*mtx};
@@ -48,9 +50,11 @@ FUZZ_TARGET(policy_estimator, .init = initialize_policy_estimator)
             },
             [&] {
                 std::vector<CTxMemPoolEntry> mempool_entries;
-                LIMITED_WHILE(fuzzed_data_provider.ConsumeBool(), 10000) {
+                LIMITED_WHILE(fuzzed_data_provider.ConsumeBool(), 10000)
+                {
                     const std::optional<CMutableTransaction> mtx = ConsumeDeserializable<CMutableTransaction>(fuzzed_data_provider);
                     if (!mtx) {
+                        good_data = false;
                         break;
                     }
                     const CTransaction tx{*mtx};
diff --git a/src/test/fuzz/rpc.cpp b/src/test/fuzz/rpc.cpp
index 270cab58e2..4e059c7d0a 100644
--- a/src/test/fuzz/rpc.cpp
+++ b/src/test/fuzz/rpc.cpp
@@ -3,18 +3,14 @@
 // file COPYING or http://www.opensource.org/licenses/mit-license.php.
 
 #include <base58.h>
-#include <core_io.h>
 #include <key.h>
 #include <key_io.h>
-#include <node/context.h>
 #include <primitives/block.h>
 #include <primitives/transaction.h>
 #include <psbt.h>
-#include <rpc/blockchain.h>
 #include <rpc/client.h>
 #include <rpc/request.h>
 #include <rpc/server.h>
-#include <rpc/util.h>
 #include <span.h>
 #include <streams.h>
 #include <test/fuzz/FuzzedDataProvider.h>
@@ -22,19 +18,23 @@
 #include <test/fuzz/util.h>
 #include <test/util/setup_common.h>
 #include <tinyformat.h>
+#include <uint256.h>
 #include <univalue.h>
-#include <util/chaintype.h>
 #include <util/strencodings.h>
 #include <util/string.h>
 #include <util/time.h>
 
+#include <algorithm>
+#include <cassert>
 #include <cstdint>
+#include <cstdlib>
+#include <exception>
 #include <iostream>
 #include <memory>
 #include <optional>
 #include <stdexcept>
-#include <string>
 #include <vector>
+enum class ChainType;
 
 namespace {
 struct RPCFuzzTestingSetup : public TestingSetup {
@@ -184,7 +184,7 @@ const std::vector<std::string> RPC_COMMANDS_SAFE_FOR_FUZZING{
     "waitfornewblock",
 };
 
-std::string ConsumeScalarRPCArgument(FuzzedDataProvider& fuzzed_data_provider)
+std::string ConsumeScalarRPCArgument(FuzzedDataProvider& fuzzed_data_provider, bool& good_data)
 {
     const size_t max_string_length = 4096;
     const size_t max_base58_bytes_length{64};
@@ -251,6 +251,7 @@ std::string ConsumeScalarRPCArgument(FuzzedDataProvider& fuzzed_data_provider)
             // hex encoded block
             std::optional<CBlock> opt_block = ConsumeDeserializable<CBlock>(fuzzed_data_provider);
             if (!opt_block) {
+                good_data = false;
                 return;
             }
             CDataStream data_stream{SER_NETWORK, PROTOCOL_VERSION};
@@ -261,6 +262,7 @@ std::string ConsumeScalarRPCArgument(FuzzedDataProvider& fuzzed_data_provider)
             // hex encoded block header
             std::optional<CBlockHeader> opt_block_header = ConsumeDeserializable<CBlockHeader>(fuzzed_data_provider);
             if (!opt_block_header) {
+                good_data = false;
                 return;
             }
             DataStream data_stream{};
@@ -271,6 +273,7 @@ std::string ConsumeScalarRPCArgument(FuzzedDataProvider& fuzzed_data_provider)
             // hex encoded tx
             std::optional<CMutableTransaction> opt_tx = ConsumeDeserializable<CMutableTransaction>(fuzzed_data_provider);
             if (!opt_tx) {
+                good_data = false;
                 return;
             }
             CDataStream data_stream{SER_NETWORK, fuzzed_data_provider.ConsumeBool() ? PROTOCOL_VERSION : (PROTOCOL_VERSION | SERIALIZE_TRANSACTION_NO_WITNESS)};
@@ -281,6 +284,7 @@ std::string ConsumeScalarRPCArgument(FuzzedDataProvider& fuzzed_data_provider)
             // base64 encoded psbt
             std::optional<PartiallySignedTransaction> opt_psbt = ConsumeDeserializable<PartiallySignedTransaction>(fuzzed_data_provider);
             if (!opt_psbt) {
+                good_data = false;
                 return;
             }
             CDataStream data_stream{SER_NETWORK, PROTOCOL_VERSION};
@@ -291,6 +295,7 @@ std::string ConsumeScalarRPCArgument(FuzzedDataProvider& fuzzed_data_provider)
             // base58 encoded key
             CKey key = ConsumePrivateKey(fuzzed_data_provider);
             if (!key.IsValid()) {
+                good_data = false;
                 return;
             }
             r = EncodeSecret(key);
@@ -299,6 +304,7 @@ std::string ConsumeScalarRPCArgument(FuzzedDataProvider& fuzzed_data_provider)
             // hex encoded pubkey
             CKey key = ConsumePrivateKey(fuzzed_data_provider);
             if (!key.IsValid()) {
+                good_data = false;
                 return;
             }
             r = HexStr(key.GetPubKey());
@@ -306,18 +312,19 @@ std::string ConsumeScalarRPCArgument(FuzzedDataProvider& fuzzed_data_provider)
     return r;
 }
 
-std::string ConsumeArrayRPCArgument(FuzzedDataProvider& fuzzed_data_provider)
+std::string ConsumeArrayRPCArgument(FuzzedDataProvider& fuzzed_data_provider, bool& good_data)
 {
     std::vector<std::string> scalar_arguments;
-    LIMITED_WHILE(fuzzed_data_provider.ConsumeBool(), 100) {
-        scalar_arguments.push_back(ConsumeScalarRPCArgument(fuzzed_data_provider));
+    LIMITED_WHILE(good_data && fuzzed_data_provider.ConsumeBool(), 100)
+    {
+        scalar_arguments.push_back(ConsumeScalarRPCArgument(fuzzed_data_provider, good_data));
     }
     return "[\"" + Join(scalar_arguments, "\",\"") + "\"]";
 }
 
-std::string ConsumeRPCArgument(FuzzedDataProvider& fuzzed_data_provider)
+std::string ConsumeRPCArgument(FuzzedDataProvider& fuzzed_data_provider, bool& good_data)
 {
-    return fuzzed_data_provider.ConsumeBool() ? ConsumeScalarRPCArgument(fuzzed_data_provider) : ConsumeArrayRPCArgument(fuzzed_data_provider);
+    return fuzzed_data_provider.ConsumeBool() ? ConsumeScalarRPCArgument(fuzzed_data_provider, good_data) : ConsumeArrayRPCArgument(fuzzed_data_provider, good_data);
 }
 
 RPCFuzzTestingSetup* InitializeRPCFuzzTestingSetup()
@@ -353,6 +360,7 @@ void initialize_rpc()
 FUZZ_TARGET(rpc, .init = initialize_rpc)
 {
     FuzzedDataProvider fuzzed_data_provider{buffer.data(), buffer.size()};
+    bool good_data{true};
     SetMockTime(ConsumeTime(fuzzed_data_provider));
     const std::string rpc_command = fuzzed_data_provider.ConsumeRandomLengthString(64);
     if (!g_limit_to_rpc_command.empty() && rpc_command != g_limit_to_rpc_command) {
@@ -363,8 +371,9 @@ FUZZ_TARGET(rpc, .init = initialize_rpc)
         return;
     }
     std::vector<std::string> arguments;
-    LIMITED_WHILE(fuzzed_data_provider.ConsumeBool(), 100) {
-        arguments.push_back(ConsumeRPCArgument(fuzzed_data_provider));
+    LIMITED_WHILE(good_data && fuzzed_data_provider.ConsumeBool(), 100)
+    {
+        arguments.push_back(ConsumeRPCArgument(fuzzed_data_provider, good_data));
     }
     try {
         rpc_testing_setup->CallRPC(rpc_command, arguments);