09e25071f40c564af08a1386c39c4f2d8eb484b6 Cache parent xpub inside of BIP32PubkeyProvider (Andrew Chow)
deb791c7ba057a3765d09b12bf3e55547a5298e4 Only cache xpubs that have a hardened last step (Andrew Chow)
f76733eda5f4c161e9eb47c74b949582ab8f448a Cache the immediate derivation parent xpub (Andrew Chow)
58f54b686f663e4c46a2cf7a64560409007c7eb3 Add DescriptorCache* read_cache and DescriptorCache* write_cache to Expand and GetPubKey (Andrew Chow)
66c2cadc91d26074b89e5ada68350b5c8676efac Rename BIP32PubkeyProvider.m_extkey to m_root_extkey (Andrew Chow)
df55d44d0de2174ba74ed3a28bef5e83b0a51b47 Track the index of the key expression in PubkeyProvider (Andrew Chow)
474ea3b927ddc67e64ae78e08c20c9264817e84d Introduce DescriptorCache struct which caches xpubs (Andrew Chow)
Pull request description:
Improves the descriptor cache by changing it from a `std::vector<unsigned char>` to a newly introduced `DescriptorCache` class. Instead of serializing pubkeys and whatever else we would want to cache in a way that may not be backwards compatible, we instead create a `DescriptorCache` object and populate it. This object contains only an xpub cache. Since the only `PubkeyProvider` that used the cache is the `BIP32PubkeyProvider` we just have it store the xpubs instead of the pubkeys. This allows us to have both the parent xpub and the child xpubs in the same container. The map is keyed by `KeyOriginInfo`.
Sine we are caching `CExtPubKey`s in `DescriptorCache`, `BIP32PubKeyProviders` can use the cached parent xpubs to derive the children if unhardened derivation is used in the last step. This also means that we can still derive the keys for a `BIP32PubkeyProvider` that has hardened derivation steps. When combined with descriptor wallets, this should allow us to be able to import a descriptor with an `xprv` and hardened steps and still be able to derive from it. In that sense, this is an alternative to #18163
To test that this works, the tests have been updated to do an additional `Expand` at the `i + 1` position. This expansion is not cached. We then do an `ExpandFromCache` at `i + 1` and use the cache that was produced by the expansion at `i`. This way, we won't have the child xpubs for `i + 1` but we will have the parent xpubs. So this checks whether the parent xpubs are being stored and can be used to derive the child keys. Descriptors that have a hardened last step are skipped for this part of the test because that will always require private keys.
ACKs for top commit:
instagibbs:
code review re-re-ACK 09e25071f4
Sjors:
re-ACK 09e25071f40c564af08a1386c39c4f2d8eb484b6
Tree-SHA512: 95c8d0092274cdf115ce39f6d49dec767679abf3758d5b9e418afc308deca9dc6f67167980195bcc036cd9c09890bbbb39ec1dacffbfacdc03efd72a7e23b276
fab7d14ea5a4305317d66f35beb3225a07823d42 test: Check that wait_until returns if time point is in the past (MarcoFalke)
Pull request description:
Add an explicit regression test for the condvar bug (#18227), so that this doesn't happen again
ACKs for top commit:
laanwj:
ACK fab7d14ea5a4305317d66f35beb3225a07823d42
Tree-SHA512: 6ec0d0b3945cae87a001e367af34cca1953a8082b4a0d9f8a20d30acd1f36363e98035d4eb173ff786cf6692d352d41f960633415c46394af042eb44e3b5ad71
9220a0fdd0f3dc2c8dd7cbeefac7d11106451b51 tests: Add one specialized ProcessMessage(...) fuzzing binary per message type for optimal results when using coverage-guided fuzzing (practicalswift)
fd1dae10b4a549ba9292d837235d59bd9eebbed3 tests: Add fuzzing harness for ProcessMessage(...) (practicalswift)
Pull request description:
Add fuzzing harness for `ProcessMessage(...)`. Enables high-level fuzzing of the P2P layer.
All code paths reachable from this fuzzer can be assumed to be reachable for an untrusted peer.
Seeded from thin air (an empty corpus) this fuzzer reaches roughly 20 000 lines of code.
To test this PR:
```
$ make distclean
$ ./autogen.sh
$ CC=clang CXX=clang++ ./configure --enable-fuzz \
--with-sanitizers=address,fuzzer,undefined
$ make
$ src/test/fuzz/process_message
…
```
Worth noting about this fuzzing harness:
* To achieve a reasonable number of executions per seconds the state of the fuzzer is unfortunately not entirely reset between `test_one_input` calls. The set-up (`FuzzingSetup` ctor) and tear-down (`~FuzzingSetup`) work is simply too costly to be run on every iteration. There is a trade-off to handle here between a.) achieving high executions/second and b.) giving the fuzzer a totally blank slate for each call. Please let me know if you have any suggestion on how to improve this situation while maintaining >1000 executions/second.
* To achieve optimal results when using coverage-guided fuzzing I've chosen to create one specialised fuzzing binary per message type (`process_message_addr`, `process_message_block`, `process_message_blocktxn `, etc.) and one general fuzzing binary (`process_message`) which handles all messages types. The latter general fuzzer can be seeded with inputs generated by the former specialised fuzzers.
Happy fuzzing friends!
ACKs for top commit:
MarcoFalke:
ACK 9220a0fdd0 🏊
Tree-SHA512: c314ef12b0db17b53cbf3abfb9ecc10ce420fb45b17c1db0b34cabe7c30e453947b3ae462020b0c9f30e2c67a7ef1df68826238687dc2479cd816f0addb530e5
6590395f6047cbfbe29f491d816c25c9a28d23a2 tests: Remove FUZZERS_MISSING_CORPORA (practicalswift)
815c7a679316e34b2072a45949ad4ecb1ae1c7fb tests: Add basic fuzzing harness for CNetAddr/CService/CSubNet related functions (netaddress.h) (practicalswift)
Pull request description:
Add basic fuzzing harness for `CNetAddr`/`CService`/`CSubNet` related functions (`netaddress.h`).
To test this PR:
```
$ make distclean
$ ./autogen.sh
$ CC=clang CXX=clang++ ./configure --enable-fuzz \
--with-sanitizers=address,fuzzer,undefined
$ make
$ src/test/fuzz/netaddress
…
```
Top commit has no ACKs.
Tree-SHA512: 69dc0e391d56d5e9cdb818ac0ac4b69445d0195f714442a06cf662998e38b6e0bbaa635dce78df37ba797feed633e94abba4764b946c1716d392756e7809112d
Also adds tests for this:
For ranged descriptors with unhardened derivation, we expect to
find parent keys in the cache but no child keys.
For descriptors containing an xpub but do not have unhardened derivation
(i.e. hardened derivation or single xpub with or without derivation),
we expect to find all of the keys in the cache, and the same
number of keys in the cache as in the SigningProvider.
For everything else (no xpub), nothing should be cached at all.
Have Expand, ExpandFromCache, and ExpandHelper take additional DescriptorCache
parameters. These are then passed into PubkeyProvider::GetPubKey which
also takes them as arguments.
Reading and writing to the cache is pushed down into GetPubKey. The old cache where
pubkeys are serialized to a vector is completely removed and instead xpubs are being
cached in DescriptorCache.
70a6b529f306ff72ea1badf25e970a92b2b17ab3 lint-cppcheck: Remove -DHAVE_WORKING_BOOST_SLEEP_FOR (Anthony Towns)
294937b39de5924e772f8ed90d35c53290c8acab scheduler_tests: re-enable mockforward test (Anthony Towns)
cea19f685915be8affb2203184a549576194413f Drop unused reverselock.h (Anthony Towns)
d0ebd93270758ea97ea956b8821e17a2d001ea94 scheduler: switch from boost to std (Anthony Towns)
b9c426012770d166e6ebfab27689be44e6e89aa5 sync.h: add REVERSE_LOCK (Anthony Towns)
306f71b4eb4a0fd8e64f47dc008bc235b80b13d9 scheduler: don't rely on boost interrupt on shutdown (Anthony Towns)
Pull request description:
Replacing boost functionality with C++11 stuff.
Motivated by #18227, but should stand alone. Changing from `boost::condition_var` to `std::condition_var` means `threadGroup.interrupt_all` isn't enough to interrupt `serviceQueue` anymore, so that means calling `stop()` before `join_all()` is needed. And the existing reverselock.h code doesn't work with sync.h's DebugLock code (because the reversed lock won't be removed from `g_lockstack` which then leads to incorrect potential deadlock warnings), so I've replaced that with a dedicated class and macro that's aware of our debug lock behaviour.
Fixes#16027, Fixes#14200, Fixes#18227
ACKs for top commit:
laanwj:
ACK 70a6b529f306ff72ea1badf25e970a92b2b17ab3
Tree-SHA512: d1da13adeabcf9186d114e2dad9a4fdbe2e440f7afbccde0c13dfbaf464efcd850b69d3371c5bf8b179d7ceb9d81f4af3cc22960b90834e41eaaf6d52ef7d331
Changes from boost::chrono to std::chrono, boost::condition_var to
std::condition_var, boost::mutex to sync.h Mutex, and reverselock.h to
sync.h REVERSE_LOCK. Also adds threadsafety annotations to CScheduler
members.
Calling interrupt_all() will immediately stop the scheduler, so it's
safe to invoke stop() beforehand, and this removes the reliance on boost
to interrupt serviceQueue().
fae86c38bca5c960462e53975314a0749db5d17d util: Remove unused MilliSleep (MarcoFalke)
fa9af06d91e9357e86863781746f0e78a509967e scripted-diff: Replace MilliSleep with UninterruptibleSleep (MarcoFalke)
fa4620be782c2bf6b5ffddf4f671194fdd1536f3 util: Add UnintrruptibleSleep (MarcoFalke)
Pull request description:
We don't use the interruptible feature of boost's sleep anywhere, so replace it with the sleep in `std::thread`
ACKs for top commit:
ajtowns:
ACK fae86c38bca5c960462e53975314a0749db5d17d quick code review
practicalswift:
ACK fae86c38bca5c960462e53975314a0749db5d17d -- patch looks correct
sipa:
Concept and code review ACK fae86c38bca5c960462e53975314a0749db5d17d
fanquake:
ACK fae86c38bca5c960462e53975314a0749db5d17d - note that an instance of `DHAVE_WORKING_BOOST_SLEEP_FOR` was missed in the [linter](https://github.com/bitcoin/bitcoin/blob/master/test/lint/extended-lint-cppcheck.sh#L69), but that can be cleaned up later.
Tree-SHA512: 7c0f8eb197664b9f7d9fe6c472c77d384f11c797c913afc31de4b532e3b4fd9ea6dd174f92062ff9d1ec39b25e0900ca7c597435add87f0f2477d9557204848c
9ff41f64198e8ddb969544fc1a5328763f1fa183 tests: Add float to FUZZERS_MISSING_CORPORA (temporarily) (practicalswift)
8f6fb0a85ae6399c8fb4f205ad35c319c42294f1 tests: Add serialization/deserialization fuzzing for integral types (practicalswift)
3c82b92d2e01e409cc46261bffcf3643102f0b94 tests: Add fuzzing harness for functions taking floating-point types as input (practicalswift)
c2bd5888607d283a229c9361747a93c83dfea0de Add missing includes (practicalswift)
Pull request description:
Add simple fuzzing harness for functions with floating-point parameters (such as `ser_double_to_uint64(double)`, etc.).
Add serialization/deserialization fuzzing for integral types.
Add missing includes.
To test this PR:
```
$ make distclean
$ ./autogen.sh
$ CC=clang CXX=clang++ ./configure --enable-fuzz \
--with-sanitizers=address,fuzzer,undefined
$ make
$ src/test/fuzz/float
…
```
Top commit has no ACKs.
Tree-SHA512: 9b5a0c4838ad18d715c7398e557d2a6d0fcc03aa842f76d7a8ed716170a28f17f249eaede4256998aa3417afe2935e0ffdfaa883727d71ae2d2d18a41ced24b5
7e9c7113afbed96cef80c327cc93e82000d6bb69 compressor: Make the domain of CompressAmount(...) explicit (practicalswift)
4a7fd7a7124f84e010b01d0769ef0572bf031ee8 tests: Add amount compression/decompression fuzzing to existing fuzzing harness: test compression round-trip (practicalswift)
Pull request description:
Small fuzzing improvement:
Add amount compression/decompression fuzzing to existing fuzzing harness: test compression round-trip (`DecompressAmount(CompressAmount(…))`).
Make the domain of `CompressAmount(…)` explicit.
Amount compression primer:
```
Compact serialization for amounts
Special serializer/deserializer for amount values. It is optimized for
values which have few non-zero digits in decimal representation. Most
amounts currently in the txout set take only 1 or 2 bytes to
represent.
```
**How to test this PR**
```
$ make distclean
$ ./autogen.sh
$ CC=clang CXX=clang++ ./configure --enable-fuzz \
--with-sanitizers=address,fuzzer,undefined
$ make
$ src/test/fuzz/integer
…
```
Top commit has no ACKs.
Tree-SHA512: 0f7c05b97012ccd5cd05a96c209e6b4d7d2fa73138bac9615cf531baa3f614f9003e29a198015bcc083af9f5bdc752bb52615b82c5df3c519b1a064bd4fc6664
470e2ac602ed2d6e62e5c80f27cd0a60c7cf6bce tests: Avoid hitting some known minor tinyformat issues when fuzzing strprintf(...) (practicalswift)
Pull request description:
Avoid hitting some known minor tinyformat issues when fuzzing `strprintf(...)`. These can be removed when the issues have been resolved upstreams :)
Note to reviewers: The `%c` and `%*` issues are also present for `%<some junk>c` and `%<some junk>*`. That is why simply matching on `"%c"` or `"%*"` is not enough. Note that the intentionally trivial skipping logic overshoots somewhat (`c[…]%` is filtered in addition to `%[…]c`).
Top commit has no ACKs.
Tree-SHA512: 2b002981e8b3f2ee021c3013f1260654ac7e158699313849c9e9660462bb8cd521544935799bb8daa74925959dc04d63440e647495e0b008cfe1b8a8b2202d40
