416ceb8661 descriptor: check if `rawtr` has only one key. (w0xlt)
Pull request description:
If I understand `rawtr` descriptor correctly, it should only allow `rawtr(KEY)`, not `rawtr(KEY1, KEY2, ...)` or other concatenations.
On master branch, `rawtr(KEY1, KEY2, ...)` will produce the `rawtr(KEY1)` descriptor ignoring the `KEY2, ...` with no error messages or warnings.
For example, the code below will print `rawtr(tprv8ZgxMBicQKsPefef2Doobbq3xTCaVTHcDn6me82KSXY1vY9AJAWD5u7SDM4XGLfc4EoXRMFrJKpp6HNmQWA3FTMRQeEmMJYJ9RPqe9ne2hU/*)#lx9qryfh`
for the supposedly invalid descriptor
`rawtr(tprv8ZgxMBicQKsPefef2Doobbq3xTCaVTHcDn6me82KSXY1vY9AJAWD5u7SDM4XGLfc4EoXRMFrJKpp6HNmQWA3FTMRQeEmMJYJ9RPqe9ne2hU/*, tprv8ZgxMBicQKsPezQ2KGArMRovTEbCGxaLgBgaVcTvEx8mby8ogX2bgC4HBapH4yMwrz2FpoCuA17eocuUVMgEP6fnm83YpwSDTFrumw42bny/*)`
```python
self.nodes[1].createwallet(wallet_name="rawtr_multi", descriptors=True, blank=True)
rawtr_multi = self.nodes[1].get_wallet_rpc("rawtr_multi")
rawtr_multi_desc = "rawtr(tprv8ZgxMBicQKsPefef2Doobbq3xTCaVTHcDn6me82KSXY1vY9AJAWD5u7SDM4XGLfc4EoXRMFrJKpp6HNmQWA3FTMRQeEmMJYJ9RPqe9ne2hU/*, tprv8ZgxMBicQKsPezQ2KGArMRovTEbCGxaLgBgaVcTvEx8mby8ogX2bgC4HBapH4yMwrz2FpoCuA17eocuUVMgEP6fnm83YpwSDTFrumw42bny/*)#uv78hkt0"
result = rawtr_multi.importdescriptors([{"desc": rawtr_multi_desc, "active": True, "timestamp": "now"}])
print(rawtr_multi.listdescriptors(True))
```
This PR adds a check that prevents `rawtr` descriptors from being created if more than one key is entered, shows an error message, and adds a test for this case.
ACKs for top commit:
achow101:
ACK 416ceb8661
sipa:
ACK 416ceb8661
Tree-SHA512: a2009e91f1bca6ee79cc68f65811caa6a21fc8b80acd8dc58e283f424b41fe53b0db7ce3693b1c7e2184ff571e6d1fbb9f5ccde89b65d3026726f3393c492044
b16f93cadd script/sign: remove needless IsSolvable() utility (Antoine Poinsot)
c232ef20c0 outputtype: remove redundant check for uncompressed keys in AddAndGetDestinationForScript (Antoine Poinsot)
Pull request description:
Now that we have descriptors there is no need to try to sign for a scriptPubKey using dummy signatures, and using a mocked verification of this witness against the interpreter, just to make sure we know how to spend such a Script. Just try to infer a solvable descriptor: any scriptPubKey that we can sign for can be inferred as such.
This came up in #24149 but i think it's worth it on its own.
ACKs for top commit:
instagibbs:
ACK b16f93cadd
achow101:
re-ACK b16f93cadd
furszy:
ACK b16f93ca, only change is the `IsSolvable` helper function removal.
Tree-SHA512: 137068157ce90210b710b1bf9ac3c400e2ff5af1112f892094b69875ea473d6a899f52adb51e5030cb907dee517602059cd1661107808558efa5de842ba12b41
fb9faffae3 extended keys: fail to derive too large depth instead of wrapping around (Antoine Poinsot)
8dc6670ce1 descriptor: don't assert success of extended key derivation (Antoine Poinsot)
50cfc9e761 (pubk)key: mark Derive() as nodiscard (Antoine Poinsot)
0ca258a5ac descriptor: never ignore the return value when deriving an extended key (Antoine Poinsot)
d3599c22bd spkman: don't ignore the return value when deriving an extended key (Antoine Poinsot)
Pull request description:
We would previously silently wrap the derived child's depth back to `0`. Instead, explicitly fail when trying to derive an impossible depth, and handle the error in callers.
An extended fuzzing corpus of `descriptor_parse` triggered this behaviour, which was reported by MarcoFalke.
Fixes#25751.
ACKs for top commit:
achow101:
re-ACK fb9faffae3
instagibbs:
utACK fb9faffae3
Tree-SHA512: 9f75c23572ce847239bd15e5497df2960b6bd63c61ea72347959d968b5c4c9a4bfeee284e76bdcd7bacbf9eeb70feee85ffd3e316f353ca6eca30e93aafad343
544b4332f0 Add wallet tests for spending rawtr() (Pieter Wuille)
e1e3081200 If P2TR tweaked key is available, sign with it (Pieter Wuille)
8d9670ccb7 Add rawtr() descriptor for P2TR with unknown tweak (Pieter Wuille)
Pull request description:
It may be useful to be able to represent P2TR outputs in descriptors whose script tree and/or internal key aren't known. This PR does that, by adding a `rawtr(KEY)` descriptor, where the KEY represents the output key directly. If the private key corresponding to that output key is known, it also permits signing with it.
I'm not convinced this is desirable, but presumably "tr(KEY)" sounds more intended for direct use than "rawtr(KEY)".
ACKs for top commit:
achow101:
ACK 544b4332f0
sanket1729:
code review ACK 544b4332f0
w0xlt:
reACK 544b4332f0
Tree-SHA512: 0de08de517468bc22ab0c00db471ce33144f5dc211ebc2974c6ea95709f44e830532ec5cdb0128c572513d352120bd651c4559516d4500b5b0a3d257c4b45aca
Also:
- Make DEFAULT_MAX_SIG_CACHE_SIZE into constexpr
DEFAULT_MAX_SIG_CACHE_BYTES to utilize the compile-time integer
arithmetic overflow checking available to constexpr.
- Fix comment (MiB instead of MB) for DEFAULT_MAX_SIG_CACHE_BYTES.
- Pass in max_size_bytes parameter to InitS*Cache(), modify log line to
no longer allude to maxsigcachesize being split evenly between the two
validation caches.
- Fix possible integer truncation and add a comment.
[META] I've kept the integer types as int64_t in order to not introduce
unintended behaviour changes, in the next commit we will make
them size_t.
This fixes an potential overflow which existed prior to this patchset.
If CuckooCache::cache<Element, Hash>::setup_bytes is called with a
`size_t bytes` which, when divided by sizeof(Element), does not fit into
an uint32_t, the implicit conversion to uint32_t in the call to setup
will result in an overflow.
At least on x86_64, this overflow is possible:
static_assert(std::numeric_limits<size_t>::max() / 32 <= std::numeric_limits<uint32_t>::max());
static_assert(std::numeric_limits<size_t>::max() / 4 <= std::numeric_limits<uint32_t>::max());
This commit detects such cases and signals to callers that the `size_t
bytes` input is too large.
1. -maxsigcachesize is a DEBUG_ONLY option
2. Almost 7 years has passed since its semantics change in
830e3f3d02 from "number of entries" to
"number of mebibytes"
3. A std::new_handler was added to the codebase after the original PR
which introduced this limit, which will terminate immediately instead
of causing trouble by being caught somewhere unexpected.
Returning the approximate total size eliminates the need for
InitS*Cache() to do nElems*sizeof(uint256). The cuckoocache has a better
idea of this information.
The counter is an optimization over calling `ret.empty()`. It was
suggested that the compiler would realize `cnt` is only `0` on the first
iteration, and not actually emit the check and conditional.
This optimization was actually not triggered at all, since we
incremented `cnt` at the beginning of the first iteration. Fix it by
incrementing at the end instead.
This was reported by Github user "Janus".
Miniscript descriptors are defined under P2WSH context (either `wsh()`
or `sh(wsh())`).
Only sane Miniscripts are accepted, as insane ones (although valid by
type) can have surprising behaviour with regard to malleability
guarantees and resources limitations.
As Miniscript descriptors are longer and more complex than "legacy"
descriptors, care was taken in error reporting to help a user determine
for what reason a provided Miniscript is insane.
Co-authored-by: Pieter Wuille <pieter.wuille@gmail.com>
This is helpful for finer grained descriptor parsing error: when there
are multiple errors to report in a Miniscript descriptor start with the
"smallest" fragments: the ones closer to be a leaf.
Co-Authored-By: Pieter Wuille <pieter@wuille.net>
The taproot spenddata stored in a sigdata is the combination of data
existing previously (e.g. in a PSBT) and the data stored in a
SigningProvider. In order to use the external data when signing, we need
to be using the sigdata's spenddata.
GetSpendData needs to be finalized in order to be used. To avoid future
bugs, assert `!m_output_key.IsNull()` as m_output_key is only set during
Finalize.
TaprootSpendData can be gotten from TaprootBuilder, however for PSBT, we
also need TaprootBuilders directly (for the outputs). So we store the
TaprootBuilder in the FlatSigningProvider and when the TaprootSpendData
is needed, we generate it on the fly using the stored builder.
GetTreeTuples returns the leaves in DFS order as tuples of depth, leaf
version, and script. This is a representation of the tree that can be
serialized.
f3a50c9dfe miniscript: rename IsSane and IsSaneSubexpression to prevent misuse (Antoine Poinsot)
c5fe5163dc miniscript: nit: don't return after assert(false) (Antoine Poinsot)
7bbaca9d8d miniscript: explicit the threshold size computation in multi() (Antoine Poinsot)
8323e4249d miniscript: add an OpCode typedef for readability (Antoine Poinsot)
7a549c6c59 miniscript: mark nodes with duplicate keys as insane (Antoine Poinsot)
8c0f8bf7bc fuzz: add a Miniscript target for string representation roundtripping (Antoine Poinsot)
be34d5077b fuzz: rename and improve the Miniscript Script roundtrip target (Antoine Poinsot)
7eb70f0ac0 miniscript: tiny doc fixups (Antoine Poinsot)
5cea85f12c miniscript: split ValidSatisfactions from IsSane (Antoine Poinsot)
a0f064dc14 miniscript: introduce a CheckTimeLocksMix helper (Antoine Poinsot)
ed45ee3882 miniscript: use optional instead of bool/outarg (Antoine Poinsot)
1ab8d89fd1 miniscript: make equality operator non-recursive (Antoine Poinsot)
5922c662c0 scripted-diff: miniscript: rename 'nodetype' variables to 'fragment' (Antoine Poinsot)
c5f65db0f0 miniscript: remove a workaround for a GCC 4.8 bug (Antoine Poinsot)
Pull request description:
The Miniscript repository and the Miniscript integration PR here have been a moving target for the past months, and some final cleanups were done there that were not included here. I initially intended to add some small followup commits to #24148 but i think there are enough of them to be worth a followup PR on its own.
Some parts of the code did not change since it was initially written in 2019, and the code could use some modernization. (Use std::optional instead of out args, remove old compiler workarounds).
We refactored the helpers to be more meaningful, and also did some renaming. A new fuzz target was also added and both were merged in a single file. 2 more will be added in #24149 that will be contained in this file too.
The only behaviour change in this PR is to rule out Miniscript with duplicate keys from sane Miniscripts. In a P2WSH context, signatures can be rebounded (Miniscript does not use CODESEPARATOR) and it's reasonable to assume that reusing keys across the Script drops the malleability guarantees.
It was previously assumed such Miniscript would never exist in the first place since a compiler should never create them. We finally agreed that if one were to exist (say, written by hand or from a buggy compiler) it would be very confusing if an imported Miniscript descriptor (after #24148) with duplicate keys was deemed sane (ie, "safe to use") by Bitcoin Core. We now check for duplicate keys in the constructor.
This is (still) joint work with Pieter Wuille. (Actually he entirely authored the cleanups and code modernization.)
ACKs for top commit:
sipa:
utACK f3a50c9dfe (with the caveat that a lot of it is my own code)
sanket1729:
code review ACK f3a50c9dfe. Did not review the fuzz tests.
Tree-SHA512: c043325e4936fe25e8ece4266b46119e000c6745f88cea530fed1edf01c80f03ee6f9edc83b6e9d42ca01688d184bad16bfd967c5bb8037744e726993adf3deb
As stated on the website, duplicate keys make it hard to reason about
malleability as a single signature may unlock multiple paths.
We use a custom KeyCompare function instead of operator< to be explicit
about the requirement.
BIP 341 specifies constraints on the size of the control block _c_ used
to compute the taproot merkle root.
> The last stack element is called the control block _c_, and must have
> length _33 + 32m_, for a value of m that is an integer between 0 and
> 128, inclusive. Fail if it does not have such a length.
(See BIP-341 "Script Validation Rules" here: https://github.com/bitcoin/bips/blob/master/bip-0341.mediawiki#script-validation-rules)
Since the removal of NODISCARD in 81d5af42f4,
the only attributes def is LIFETIMEBOUND, and it's included in many more
places that it is used.
This removes all includes which do not have an associated use of LIFETIMEBOUND,
and adds it to the following files, due to their use of the same:
* src/validationinterface.h
* src/script/standard.h
This makes IsSane clearer. It is useful to differentiate between 'potential non-malleable satisfactions are valid' and 'such satisfactions exist' for testing.
Co-authored-by: Pieter Wuille <pieter.wuille@gmail.com>
The 'Fragment' type was previously named 'Nodetype'. For clarity, name
the variables the same.
-BEGIN VERIFY SCRIPT-
sed -i 's/nodetype/fragment/g' src/script/miniscript.*
-END VERIFY SCRIPT-
Co-authored-by: Pieter Wuille <pieter.wuille@gmail.com>
The value it leaves on the stack depends on the last element on the
stack. However, we can't make sure this element is OP_1 (which would
give us the 'u' property) without the MINIMALIF rule.
MINIMALIF is only policy for P2WSH, therefore giving 'd:' the 'u'
property breaks consensus soundness: it makes it possible (by consensus
but not policy) for instance to satisfy a thresh() without satisfying
at least k of its subs.
This bug was found and reported by Andrew Poelstra.
