By returning an error code if `FlushBlockFile` fails, the caller now has
to explicitly handle block file flushing errors. Before this change
such errors were non-explicitly ignored without a clear rationale.
Prior to this patch `FlushBlockFile` may have failed silently in
`Chainstate::FlushStateToDisk`. Improve this with a log line. Also add a
TODO comment to flesh out whether returning early in the case of an
error is appropriate or not. Returning early might be appropriate to
prohibit `WriteBlockIndexDB` from writing a block index entry that does
not refer to a fully flushed block.
Besides `Chainstate::FlushStateToDisk`, `FlushBlockFile` is also called
by `FindBlockPos`. Don't change the abort behavior there, since we don't
want to fail the function if the flushing of already written blocks
fails. Instead, just document it.
This change makes IsInitialBlockDownload and NotifyHeaderTip functions no
longer tied to individual Chainstate objects. It makes them work with the
ChainstateManager object instead so code is simpler and it is no longer
possible to call them incorrectly with an inactive Chainstate.
This change also makes m_cached_finished_ibd caching easier to reason about,
because now there is only one cached value instead of two (for background and
snapshot chainstates) so the cached IBD state now no longer gets reset when a
snapshot is loaded.
There should be no change in behavior because these functions were always
called on the active ChainState objects.
These changes were discussed previously
https://github.com/bitcoin/bitcoin/pull/27746#discussion_r1246868905 and
https://github.com/bitcoin/bitcoin/pull/27746#discussion_r1237552792 as
possible followups for that PR.
fb02ba3c5f mempool_entry: improve struct packing (Anthony Towns)
1a118062fb net_processing: Clean up INVENTORY_BROADCAST_MAX constants (Anthony Towns)
6fa49937e4 test: Check tx from disconnected block is immediately requestable (glozow)
e4ffabbffa net_processing: don't add txids to m_tx_inventory_known_filter (Anthony Towns)
6ec1809d33 net_processing: drop m_recently_announced_invs bloom filter (Anthony Towns)
a70beafdb2 validation: when adding txs due to a block reorg, allow immediate relay (Anthony Towns)
1e9684f39f mempool_entry: add mempool entry sequence number (Anthony Towns)
Pull request description:
This PR replaces the `m_recently_announced_invs` bloom filter with a simple sequence number tracking the mempool state when we last considered sending an INV message to a node. This saves 33kB per peer (or more if we raise the rate at which we relay transactions over the network, in which case we would need to increase the size of the bloom filter proportionally).
The philosophy here (compare with #18861 and #19109) is that we consider the rate limiting on INV messages to only be about saving bandwidth and not protecting privacy, and therefore after you receive an INV message, it's immediately fair game to request any transaction that was in the mempool at the time the INV message was sent. We likewise consider the BIP 133 feefilter and BIP 37 bloom filters to be bandwidth optimisations here, and treat transactions as requestable if they would have been announced without those filters. Given that philosophy, tracking the timestamp of the last INV message and comparing that against the mempool entry time allows removal of each of `m_recently_announced_invs`, `m_last_mempool_req` and `UNCONDITIONAL_RELAY_DELAY` and associated logic.
ACKs for top commit:
naumenkogs:
ACK fb02ba3c5f
amitiuttarwar:
review ACK fb02ba3c5f
glozow:
reACK fb02ba3c5f
Tree-SHA512: cbba5ee04c86df26b6057f3654c00a2b45ec94d354f4f157a769cecdaa0b509edaac02b3128afba39b023e82473fc5e28c915a787f84457ffe66638c6ac9c2d4
fa776e61cd Add importmempool RPC (MarcoFalke)
fa20d734a2 refactor: Add and use kernel::ImportMempoolOptions (MarcoFalke)
fa8866990d doc: Clarify the getmempoolinfo.loaded RPC field documentation (MarcoFalke)
6888886cec Remove Chainstate::LoadMempool (MarcoFalke)
Pull request description:
Currently it is possible to import a mempool by placing it in the datadir and starting the node. However this has many issues:
* Users aren't expected to fiddle with the datadir, possibly corrupting it
* An existing mempool file in the datadir may be overwritten
* The node needs to be restarted
* Importing an untrusted file this way is dangerous, because it can corrupt the mempool
Fix all issues by adding a new RPC.
ACKs for top commit:
ajtowns:
utACK fa776e61cd
achow101:
ACK fa776e61cd
glozow:
reACK fa776e61cd
Tree-SHA512: fcb1a92d6460839283c546c47a2d930c363ac1013c4c50dc5215ddf9fe5e51921d23fe0abfae0a5a7631983cfc7e2fff3788b70f95937d0a989a203be4d67546
d8f1222ac5 refactor: Correct dbwrapper key naming (TheCharlatan)
be8f159ac5 build: Remove leveldb from BITCOIN_INCLUDES (TheCharlatan)
c95b37d641 refactor: Move CDBWrapper leveldb members to their own context struct (TheCharlatan)
c534a615e9 refactor: Split dbwrapper CDBWrapper::EstimateSize implementation (TheCharlatan)
586448888b refactor: Move HandleError to dbwrapper implementation (TheCharlatan)
dede0eef7a refactor: Split dbwrapper CDBWrapper::Exists implementation (TheCharlatan)
a5c2eb5748 refactor: Fix logging.h includes (TheCharlatan)
84058e0eed refactor: Split dbwrapper CDBWrapper::Read implementation (TheCharlatan)
e4af2408f2 refactor: Pimpl leveldb::Iterator for CDBIterator (TheCharlatan)
ef941ff128 refactor: Split dbwrapper CDBIterator::GetValue implementation (TheCharlatan)
b7a1ab5cb4 refactor: Split dbwrapper CDBIterator::GetKey implementation (TheCharlatan)
d7437908cd refactor: Split dbwrapper CDBIterator::Seek implementation (TheCharlatan)
ea8135de7e refactor: Pimpl leveldb::batch for CDBBatch (TheCharlatan)
b9870c920d refactor: Split dbwrapper CDBatch::Erase implementation (TheCharlatan)
532ee812a4 refactor: Split dbwrapper CDBBatch::Write implementation (TheCharlatan)
afc534df9a refactor: Wrap DestroyDB in dbwrapper helper (TheCharlatan)
Pull request description:
Leveldb headers are currently included in the `dbwrapper.h` file and thus available to many of Bitcoin Core's source files. However, leveldb-specific functionality should be abstracted by the `dbwrapper` and does not need to be available to the rest of the code. Having leveldb included in a widely-used header such as `dbwrapper.h` bloats the entire project's header tree.
The `dbwrapper` is a key component of the libbitcoinkernel library. Future users of this library would not want to contend with having the leveldb headers exposed and potentially polluting their project's namespace.
For these reasons, the leveldb headers are removed from the `dbwrapper` by moving leveldb-specific code to the implementation file and creating a [pimpl](https://en.cppreference.com/w/cpp/language/pimpl) where leveldb member variables are indispensable. As a final step, the leveldb include flags are removed from the `BITCOIN_INCLUDES` and moved to places where the dbwrapper is compiled.
---
This pull request is part of the [libbitcoinkernel project](https://github.com/bitcoin/bitcoin/issues/27587), and more specifically its stage 1 step 3 "Decouple most non-consensus headers from libbitcoinkernel".
ACKs for top commit:
stickies-v:
re-ACK d8f1222ac5
MarcoFalke:
ACK d8f1222ac5🔠
Tree-SHA512: 0f58309be165af0162e648233451cd80fda88726fc10c0da7bfe4ec2ffa9afe63fbf7ffae9493698d3f39653b4ad870c372eee652ecc90ab1c29d86c387070f3
Wrap leveldb::DestroyDB in a helper function without exposing
leveldb-specifics.
Also, add missing optional include.
The context of this commit is an effort to decouple the dbwrapper header
file from leveldb includes. To this end, the includes are moved to the
dbwrapper implementation file. This is done as part of the kernel
project to reduce the number of required includes for users of the
kernel.
a733dd79e2 Remove unused function `reliesOnAssumedValid` (Suhas Daftuar)
d4a11abb19 Cache block index entry corresponding to assumeutxo snapshot base blockhash (Suhas Daftuar)
3556b85022 Move CheckBlockIndex() from Chainstate to ChainstateManager (Suhas Daftuar)
0ce805b632 Documentation improvements for assumeutxo (Ryan Ofsky)
768690b7ce Fix initialization of setBlockIndexCandidates when working with multiple chainstates (Suhas Daftuar)
d43a1f1a2f Tighten requirements for adding elements to setBlockIndexCandidates (Suhas Daftuar)
d0d40ea9a6 Move block-storage-related logic to ChainstateManager (Suhas Daftuar)
3cfc75366e test: Clear block index flags when testing snapshots (Suhas Daftuar)
272fbc370c Update CheckBlockIndex invariants for chains based on an assumeutxo snapshot (Suhas Daftuar)
10c05710ce Add wrapper for adding entries to a chainstate's block index candidates (Suhas Daftuar)
471da5f6e7 Move block-arrival information / preciousblock counters to ChainstateManager (Suhas Daftuar)
1cfc887d00 Remove CChain dependency in node/blockstorage (Suhas Daftuar)
fe86a7cd48 Explicitly track maximum block height stored in undo files (Suhas Daftuar)
Pull request description:
This PR proposes a clean up of the relationship between block storage and the chainstate objects, by moving the decision of whether to store a block on disk to something that is not chainstate-specific. Philosophically, the decision of whether to store a block on disk is related to validation rules that do not require any UTXO state; for anti-DoS reasons we were using some chainstate-specific heuristics, and those have been reworked here to achieve the proposed separation.
This PR also fixes a bug in how a chainstate's `setBlockIndexCandidates` was being initialized; it should always have all the HAVE_DATA block index entries that have more work than the chain tip. During startup, we were not fully populating `setBlockIndexCandidates` in some scenarios involving multiple chainstates.
Further, this PR establishes a concept that whenever we have 2 chainstates, that we always know the snapshotted chain's base block and the base block's hash must be an element of our block index. Given that, we can establish a new invariant that the background validation chainstate only needs to consider blocks leading to that snapshotted block entry as potential candidates for its tip. As a followup I would imagine that when writing net_processing logic to download blocks for the background chainstate, that we would use this concept to only download blocks towards the snapshotted entry as well.
ACKs for top commit:
achow101:
ACK a733dd79e2
jamesob:
reACK a733dd79e2 ([`jamesob/ackr/27746.5.sdaftuar.rework_validation_logic`](https://github.com/jamesob/bitcoin/tree/ackr/27746.5.sdaftuar.rework_validation_logic))
Sjors:
Code review ACK a733dd79e2.
ryanofsky:
Code review ACK a733dd79e2. Just suggested changes since the last review. There are various small things that could be followed up on, but I think this is ready for merge.
Tree-SHA512: 9ec17746f22b9c27082743ee581b8adceb2bd322fceafa507b428bdcc3ffb8b4c6601fc61cc7bb1161f890c3d38503e8b49474da7b5ab1b1f38bda7aa8668675
This is to (a) avoid repeated lookups into the block index for an entry that
should never change and (b) emphasize that the snapshot base should always
exist when set and not change during the runtime of the program.
Thanks to Russ Yanofsky for suggesting this approach.
Also rewrite CheckBlockIndex() to perform tests on all chainstates.
This increases sanity-check coverage, as any place in our code where we were
invoke CheckBlockIndex() on a single chainstate will now invoke the sanity
checks on all chainstates.
This change also tightens up the checks on setBlockIndexCandidates and
mapBlocksUnlinked, to more precisely match what we aim for even in the presence
of assumed-valid blocks.
When using assumeutxo and multiple chainstates are active, the background
chainstate should consider all HAVE_DATA blocks that are ancestors of the
snapshotted block and that have more work than the tip as potential candidates.
When using assumeutxo, we only need the background chainstate to consider
blocks that are on the chain leading to the snapshotted block.
Note that this introduces the new invariant that we can only have an assumeutxo
snapshot where the snapshotted blockhash is in our block index. Unknown block
hashes that are somehow passed in will cause assertion failures when processing
new blocks.
Includes test fixes and improvements by Andrew Chow and Fabian Jahr.
Separate the notion of which blocks are stored on disk, and what data is in our
block index, from what tip a chainstate might be able to get to. We can use
chainstate-agnostic data to determine when to store a block on disk (primarily,
an anti-DoS set of criteria) and let the chainstates figure out for themselves
when a block is of interest for being a candidate tip.
Note: some of the invariants in CheckBlockIndex are modified, but more work is
needed (ie to move CheckBlockIndex to ChainstateManager, as most of what
CheckBlockIndex is doing is checking the consistency of the block index, which
is outside of Chainstate).
Block arrival information (and the preciousblock RPC, a related concept) are
both chainstate-agnostic, so these are moved to ChainstateManager. This should
just be a refactor, without any observable behavior changes.
This change drops the last kernel dependency on shutdown.cpp. It also adds new
hooks for libbitcoinkernel applications to be able to interrupt kernel
operations when the chain tip changes.
This is a refactoring that does not affect behavior. (Looking at the code it
can appear like the new break statement in the ActivateBestChain function is a
change in behavior, but actually the previous StartShutdown call was indirectly
triggering a break before, because it was causing m_chainman.m_interrupt to be
true. The new code just makes the break more obvious.)
FatalError replaces what previously was the AbortNode function in
shutdown.cpp.
This commit is part of the libbitcoinkernel project and further removes
the shutdown's and, more generally, the kernel library's dependency on
interface_ui with a kernel notification method. By removing interface_ui
from the kernel library, its dependency on boost is reduced to just
boost::multi_index. At the same time it also takes a step towards
de-globalising the interrupt infrastructure.
Co-authored-by: Russell Yanofsky <russ@yanofsky.org>
Co-authored-by: TheCharlatan <seb.kung@gmail.com>
This and the following commit seek to decouple the libbitcoinkernel
library from the shutdown code. As a library, it should it should have
its own flexible interrupt infrastructure without relying on node-wide
globals.
The commit takes the first step towards this goal by de-globalising
`ShutdownRequested` calls in kernel code.
Co-authored-by: Russell Yanofsky <russ@yanofsky.org>
Co-authored-by: TheCharlatan <seb.kung@gmail.com>
0e21b56a44 assumeutxo: catch and log fs::remove error instead of two exist checks (Andrew Toth)
Pull request description:
Fixes a block of code which seems to be incorrectly performing two existence checks instead of catching and logging errors. `fs::remove` returns `false` only if the file being removed does not exist, so it is redundant with the `fs::exists` check. If an error does occur when trying to remove an existing file, `fs::remove` will throw. See https://en.cppreference.com/w/cpp/filesystem/remove.
Also see https://github.com/bitcoin/bitcoin/blob/master/src/init.cpp#L326-L332 for a similar pattern.
ACKs for top commit:
MarcoFalke:
lgtm ACK 0e21b56a44
jamesob:
ACK 0e21b56a44
achow101:
ACK 0e21b56a44
Tree-SHA512: 137d0be5266cfd947e5e50ec93b895ac659adadf9413bef3468744bfdacee8dbe7d9bdfaf91784c45708610325d2241a114f4be4e622a108a639b3672b618fd2
1c7d08b9ac validation: Stricter assumeutxo error handling in InvalidateCoinsDBOnDisk (Ryan Ofsky)
9047337d36 validation: Stricter assumeutxo error handling in LoadChainstate (Ryan Ofsky)
Pull request description:
There are two places in assumeutxo code where it is calling `AbortNode` to trigger asynchronous shutdowns without returning errors to calling functions.
One case, in `LoadChainstate`, happens when snapshot validation succeeds, and there is an error trying to replace the background chainstate with the snapshot chainstate.
The other case, in `InvalidateCoinsDBOnDisk`, happens when snapshot validatiion fails, and there is an error trying to remove the snapshot chainstate.
In both cases the node is being forced to shut down, so it makes sense for these functions to raise errors so callers can know that an error happened without having to infer it from the shutdown state.
Noticed these cases while reviewing #27861, which replaces the `AbortNode` function with a `FatalError` function.
ACKs for top commit:
achow101:
ACK 1c7d08b9ac
TheCharlatan:
ACK 1c7d08b9ac
jamesob:
ACK 1c7d08b9ac ([`jamesob/ackr/27862.1.ryanofsky.validation_stricter_assu`](https://github.com/jamesob/bitcoin/tree/ackr/27862.1.ryanofsky.validation_stricter_assu))
Tree-SHA512: fb1dcde3fa0e77b4ba0c48507d289552b939c2866781579c8e994edc209abc3cd29cf81c89380057199323a8eec484956abb1fd3a43c957ecd0e7f7bbfd63fd8
Currently InvalidateCoinsDBOnDisk is calling AbortNode without an error to the
caller if it fails. Change it to return just return util::Result, and update
the caller to handle the error itself.
This causes the secondary error to be shown below the main error instead of the
other way around.
Remove access to the global gArgs for getting the directory in
utxo_snapshot.
This is done in the context of the libbitcoinkernel project, wherein
reliance of libbitcoinkernel code on the global gArgs is incrementally
removed.
Remove access to the global gArgs for the stopatheight argument and
replace it by adding a field to the existing ChainstateManager Options
struct.
This should eventually allow users of the ChainstateManager to not rely
on the global gArgs and instead pass in their own options.
7d3b35004b refactor: Move system from util to common library (TheCharlatan)
7eee356c0a refactor: Split util::AnyPtr into its own file (TheCharlatan)
44de325d95 refactor: Split util::insert into its own file (TheCharlatan)
9ec5da36b6 refactor: Move ScheduleBatchPriority to its own file (TheCharlatan)
f871c69191 kernel: Add warning method to notifications (TheCharlatan)
4452707ede kernel: Add progress method to notifications (TheCharlatan)
84d71457e7 kernel: Add headerTip method to notifications (TheCharlatan)
447761c822 kernel: Add notification interface (TheCharlatan)
Pull request description:
This pull request is part of the `libbitcoinkernel` project https://github.com/bitcoin/bitcoin/issues/27587https://github.com/bitcoin/bitcoin/projects/18 and more specifically its "Step 2: Decouple most non-consensus code from libbitcoinkernel".
---
It removes the kernel library's dependency on `util/system` and `interface_ui`. `util/system` contains networking and shell-related code that should not be part of the kernel library. The following pull requests prepared `util/system` for this final step: https://github.com/bitcoin/bitcoin/pull/27419https://github.com/bitcoin/bitcoin/pull/27254https://github.com/bitcoin/bitcoin/pull/27238.
`interface_ui` defines functions for a more general node interface and has a dependency on `boost/signals2`. After applying the patches from this pull request, the kernel's reliance on boost is down to `boost::multiindex`.
The approach implemented here introduces some indirection, which makes the code a bit harder to read. Any suggestions for improving or reworking this pull request to make it more concise, or even reworking it into a more proper interface, are appreciated.
ACKs for top commit:
MarcoFalke:
re-ACK 7d3b35004b (no change) 🎋
stickies-v:
Code Review ACK 7d3b35004b
hebasto:
re-ACK 7d3b35004b, only last two commits dropped since my [recent](https://github.com/bitcoin/bitcoin/pull/27636#pullrequestreview-1435394620) review.
Tree-SHA512: c8cfc698dc9d78e20191c444708f2d957501229abe95e5806106d1126fb9c5fbcee686fb55645658c0107ce71f10646f37a2fdf7fde16bbf22cbf1ac885dd08d
This commit is part of the libbitcoinkernel project and seeks to remove
the ChainstateManager's and, more generally, the kernel library's
dependency on interface_ui with options methods in this and the following
few commits. By removing interface_ui from the kernel library, its
dependency on boost is reduced to just boost::multi_index.
The DoWarning and AlertNotify functions are moved out of the
validation.cpp file, which removes its dependency on interface_ui as
well as util/system.
This commit is part of the libbitcoinkernel project and seeks to remove
the ChainstateManager's and, more generally, the kernel library's
dependency on interface_ui with options methods in this and the
following few commits. By removing interface_ui from the kernel library,
its dependency on boost is reduced to just boost::multi_index.
This commit is part of the libbitcoinkernel project and seeks to remove
the ChainstateManager's and, more generally, the kernel library's
dependency on interface_ui with options methods in this and the following
few commits. By removing interface_ui from the kernel library, its
dependency on boost is reduced to just boost::multi_index.
This commit is part of the libbitcoinkernel project and seeks to remove
the ChainstateManager's and, more generally, the kernel library's
dependency on interface_ui with options methods in this and the following
few commits. By removing interface_ui from the kernel library, its
dependency on boost is reduced to just boost::multi_index.
Define a new kernel notification class with virtual methods for
notifying about internal kernel events. Create a new file in the node
library for defining a function creating the default set of notification
methods such that these do not need to be re-defined all over the
codebase. As a first step, add a `blockTip` method, wrapping
`uiInterface.NotifyBlockTip`.
This incorrectly assumed num_blocks_total would be greater than 0. This is not guaranteed until the ConnectBlock call right below it.
The total and average metric is not very useful because it does not distinguish between blocks read from disk and those loaded from memory. So rather than fixing the divide by zero issue, we just drop the metric.
5ff63a09a9 refactor, blockstorage: Replace stopafterblockimport arg (TheCharlatan)
18e5ba7c80 refactor, blockstorage: Replace blocksdir arg (TheCharlatan)
02a0899527 refactor, BlockManager: Replace fastprune from arg with options (TheCharlatan)
a498d699e3 refactor/iwyu: Complete includes for blockmanager_args (TheCharlatan)
f0bb1021f0 refactor: Move functions to BlockManager methods (TheCharlatan)
cfbb212493 zmq: Pass lambda to zmq's ZMQPublishRawBlockNotifier (TheCharlatan)
8ed4ff8e05 refactor: Declare g_zmq_notification_interface as unique_ptr (TheCharlatan)
Pull request description:
The libbitcoin_kernel library should not rely on the `ArgsManager`, but rather use option structs that can be passed to the various classes it uses. This PR removes reliance on the `ArgsManager` from the `blockstorage.*` files. Like similar prior work, it uses the options struct in the `BlockManager` that can be populated with `ArgsManager` values.
Some related prior work: https://github.com/bitcoin/bitcoin/pull/26889https://github.com/bitcoin/bitcoin/pull/25862https://github.com/bitcoin/bitcoin/pull/25527https://github.com/bitcoin/bitcoin/pull/25487
Related PR removing blockstorage globals: https://github.com/bitcoin/bitcoin/pull/25781
ACKs for top commit:
ryanofsky:
Code review ACK 5ff63a09a9. Since last ACK just added std::move and fixed commit title. Sorry for the noise!
mzumsande:
Code Review ACK 5ff63a09a9
Tree-SHA512: 4bde8fd140a40b97eca923e9016d85dcea6fad6fd199731f158376294af59c3e8b163a0725aa47b4be3519b61828044e0a042deea005e0c28de21d8b6c3e1ea7
72efc26439 util: improve streams.h:FindByte() performance (Larry Ruane)
604df63f6c [bench] add streams findbyte (gzhao408)
Pull request description:
This PR is strictly a performance improvement; there is no functional change. The `CBufferedFile::FindByte()` method searches for the next occurrence of the given byte in the file. Currently, this is done by explicitly inspecting each byte in turn. This PR takes advantage of `std::find()` to do the same more efficiently, improving its CPU runtime by a factor of about 25 in typical use.
ACKs for top commit:
achow101:
re-ACK 72efc26439
stickies-v:
re-ACK 72efc26439
Tree-SHA512: ddf0bff335cc8aa34f911aa4e0558fa77ce35d963d602e4ab1c63090b4a386faf074548daf06ee829c7f2c760d06eed0125cf4c34e981c6129cea1804eb3b719
This is a commit in preparation for the next few commits. The functions
are moved to methods to avoid their re-declaration for the purpose of
passing in BlockManager options.
The functions that were now moved into the BlockManager should no longer
use the params as an argument, but instead use the member variable.
In the moved ReadBlockFromDisk and UndoReadFromDisk, change
the function signature to accept a reference to a CBlockIndex instead of
a raw pointer. The pointer is expected to be non-null, so reflect that
in the type.
To allow for the move of functions to BlockManager methods all call
sites require an instantiated BlockManager, or a callback to one.
Avoid use of the expensive mod operator (%) when calculating the
buffer offset. No functional difference.
Co-authored-by: Hennadii Stepanov <32963518+hebasto@users.noreply.github.com>
c4981e7f63 prune, import: fixes#23852 (mruddy)
Pull request description:
Fixes#23852
This allows pruning to work during the `-loadblock` import process.
An example use case is where you have a clean set of block files and you want to create a pruned node from them, but you don't want to alter the input set of block files.
#23852 noted that pruning was not working reliably during the loadblock import process. The reason why the loadblock process was not pruning regularly as it progressed is that the pruning process (`BlockManager::FindFilesToPrune`) checks the tip height of the active chainstate, and `CChainState::ActivateBestChain` was not called (which updates that tip height) in `ThreadImport` until after all the import files were processed.
An example bash command line that makes it easy to import a bunch of block files:
```
./src/qt/bitcoin-qt -debug -logthreadnames -datadir=/tmp/btc -prune=550 -loadblock=/readonly/btc/main/blk{00000..00043}.dat
```
One interesting side note is that `CChainState::ActivateBestChain` can be called while the import process is running (in the `loadblk` thread) by concurrent network message processing activity in the `msghand` thread. For example, one way to reproduce this easily is with the `getblockfrompeer` RPC (requesting a block with height greater than 100000) run from a node connected to an importing node. There are other ways too, but this is an easy way. I only mention this to explain how the `max_prune_height=225719` log message in the original issue came to occur.
ACKs for top commit:
achow101:
re-ACK c4981e7f63
Tree-SHA512: d287c7753952c22f598ba782914c47f45ad44ce60b0fbce9561354e701f1a2a98bafaaaa106c8428690b814e281305ca3622b177ed3cb2eb7559f07c958ab537
bf77fc9cb4 [test] mempool full in package accept (glozow)
b51ebccc28 [validation] set PackageValidationState when mempool full (glozow)
563a2ee4f5 [policy] disallow transactions under min relay fee, even in packages (glozow)
c4554fe894 [test] package cpfp bumps parents <mempoolminfee but >=minrelaytxfee (glozow)
ac463e87df [test util] mock mempool minimum feerate (glozow)
Pull request description:
Part of package relay, see #27463.
Note that this still allows packages to bump transactions that are below the dynamic mempool minimum feerate, which means this still solves the "mempool is congested and my presigned 1sat/vB tx is screwed" problem for all transactions.
On master, the package policy (only accessible through regtest-only RPC submitpackage) allows 0-fee (or otherwise below min relay feerate) transactions if they are bumped by a child. However, with default package limits, we don't yet have a DoS-resistant way of ensuring these transactions remain bumped throughout their time in the mempool. Primarily, the fee-bumping child may later be replaced by another transaction that doesn't bump the parent(s). The parent(s) could potentially stay bumped by other transactions, but not enough to ever be selected by the `BlockAssembler` (due to `blockmintxfee`).
For example, (tested [here](https://github.com/glozow/bitcoin/commits/26933-motivation)):
- The mempool accepts 24 below-minrelayfeerate transactions ("0-fee parents"), all bumped by a single high-fee transaction ("the fee-bumping child"). The fee-bumping child also spends a confirmed UTXO.
- Two additional children are added to each 0-fee parent. These children each pay a feerate slightly above the minimum relay feerate (e.g. 1.9sat/vB) such that, for each 0-fee parent, the total fees of its two children divided by the total size of the children and parent is above the minimum relay feerate.
- If a block template is built now, all transactions would be selected.
- A transaction replaces the the fee-bumping child, spending only the confirmed UTXO and not any of the outputs from the 0-fee parents.
- The 0-fee parents now each have 2 children. Their descendant feerates are above minrelayfeerate, which means that they remain in the mempool, even if the mempool evicts all below-minrelayfeerate packages.
- If a block template is built now, none of the 0-fee parents or their children would be selected.
- Even more low-feerate descendants can be added to these below-minrelayfeerate packages and they will not be evicted until they expire or the mempool reaches capacity.
Unless we have a DoS-resistant way of ensuring package CPFP-bumped transactions are always bumped, allowing package CPFP to bump below-minrelayfeerate transactions can result in these problematic situations. See #27018 which proposes a partial solution with some limitations, and contains discussion about potential improvements to eviction strategy. While no adequate solution exists, for now, avoid these situations by requiring all transactions to meet min relay feerate.
ACKs for top commit:
ajtowns:
reACK bf77fc9cb4
instagibbs:
re-ACK bf77fc9cb4
Tree-SHA512: 28940f41493a9e280b010284316fb8caf1ed7b2090ba9a4ef8a3b2eafc5933601074b142f4f7d4e3c6c4cce99d3146f5c8e1393d9406c6f2070dd41c817985c9
be55f545d5 move-only: Extract common/args and common/config.cpp from util/system (TheCharlatan)
Pull request description:
This pull request is part of the `libbitcoinkernel` project https://github.com/bitcoin/bitcoin/issues/24303https://github.com/bitcoin/bitcoin/projects/18 and more specifically its "Step 2: Decouple most non-consensus code from libbitcoinkernel". It is part of a series of patches splitting up the `util/system` files. Its preceding pull request is https://github.com/bitcoin/bitcoin/pull/27254.
The pull request contains an extraction of ArgsManager related functions from util/system into their own common/ file.
The background of this commit is an ongoing effort to decouple the libbitcoinkernel library from the ArgsManager. The ArgsManager belongs into the common library, since the kernel library should not depend on it. See [doc/design/libraries.md](https://github.com/bitcoin/bitcoin/blob/master/doc/design/libraries.md) for more information on this rationale.
ACKs for top commit:
MarcoFalke:
re-ACK be55f545d5🚲
ryanofsky:
Code review ACK be55f545d5. Just small cleanups since the last review.
hebasto:
ACK be55f545d5, I have reviewed the code and it looks OK, I agree it can be merged.
Tree-SHA512: 90eb03334af0155b823030b4f2ecf286d35058d700ee2ddbbaa445be19e31eb0fe982656f35bd14ecee3ad2c3d0db3746855cb8f3777eff7253713e42873e111
9f947fc3d4 Use PoolAllocator for CCoinsMap (Martin Leitner-Ankerl)
5e4ac5abf5 Call ReallocateCache() on each Flush() (Martin Leitner-Ankerl)
1afca6b663 Add PoolResource fuzzer (Martin Leitner-Ankerl)
e19943f049 Calculate memory usage correctly for unordered_maps that use PoolAllocator (Martin Leitner-Ankerl)
b8401c3281 Add pool based memory resource & allocator (Martin Leitner-Ankerl)
Pull request description:
A memory resource similar to `std::pmr::unsynchronized_pool_resource`, but optimized for node-based containers. The goal is to be able to cache more coins with the same memory usage, and allocate/deallocate faster.
This is a reimplementation of #22702. The goal was to implement it in a way that is simpler to review & test
* There is now a generic `PoolResource` for allocating/deallocating memory. This has practically the same API as `std::pmr::memory_resource`. (Unfortunately I cannot use std::pmr because libc++ simply doesn't implement that API).
* Thanks to sipa there is now a fuzzer for PoolResource! On a fast machine I ran it for ~770 million executions without finding any issue.
* The estimation of the correct node size is now gone, PoolResource now has multiple pools and just needs to be created large enough to have space for the unordered_map nodes.
I ran benchmarks with #22702, mergebase, and this PR. Frequency locked Intel i7-8700, clang++ 13.0.1 to reindex up to block 690000.
```sh
bitcoind -dbcache=5000 -assumevalid=00000000000000000002a23d6df20eecec15b21d32c75833cce28f113de888b7 -reindex-chainstate -printtoconsole=0 -stopatheight=690000
```
The performance is practically identical with #22702, just 0.4% slower. It's ~21% faster than master:
Note that on cache drops mergebase's memory doesnt go so far down because it does not free the `CCoinsMap` bucket array.
ACKs for top commit:
LarryRuane:
ACK 9f947fc3d4
achow101:
re-ACK 9f947fc3d4
john-moffett:
ACK 9f947fc3d4
jonatack:
re-ACK 9f947fc3d4
Tree-SHA512: 48caf57d1775875a612b54388ef64c53952cd48741cacfe20d89049f2fb35301b5c28e69264b7d659a3ca33d4c714d47bafad6fd547c4075f08b45acc87c0f45
This is an extraction of ArgsManager related functions from util/system
into their own common file.
Config file related functions are moved to common/config.cpp.
The background of this commit is an ongoing effort to decouple the
libbitcoinkernel library from the ArgsManager. The ArgsManager belongs
into the common library, since the kernel library should not depend on
it. See doc/design/libraries.md for more information on this rationale.
Avoid adding transactions below min relay feerate because, even if they
were bumped through CPFP when entering the mempool, we do not have a
DoS-resistant way of ensuring they always remain bumped. In the future,
this rule can be relaxed (e.g. to allow packages to bump 0-fee
transactions) if we find a way to do so.
00e9b97f37 refactor: Move fs.* to util/fs.* (TheCharlatan)
106b46d9d2 Add missing fs.h includes (TheCharlatan)
b202b3dd63 Add missing cstddef include in assumptions.h (TheCharlatan)
18fb36367a refactor: Extract util/fs_helpers from util/system (Ben Woosley)
Pull request description:
This pull request is part of the `libbitcoinkernel` project https://github.com/bitcoin/bitcoin/issues/24303https://github.com/bitcoin/bitcoin/projects/18 and more specifically its "Step 2: Decouple most non-consensus code from libbitcoinkernel". This commit was originally authored by empact and is taken from its parent PR #25152.
#### Context
There is an ongoing effort to decouple the `ArgsManager` used for command line parsing user-provided arguments from the libbitcoinkernel library (https://github.com/bitcoin/bitcoin/pull/25290, https://github.com/bitcoin/bitcoin/pull/25487, https://github.com/bitcoin/bitcoin/pull/25527, https://github.com/bitcoin/bitcoin/pull/25862, https://github.com/bitcoin/bitcoin/pull/26177, and https://github.com/bitcoin/bitcoin/pull/27125). The `ArgsManager` is defined in `system.h`. A similar pull request extracting functionality from `system.h` has been merged in https://github.com/bitcoin/bitcoin/pull/27238.
#### Changes
Next to providing better code organization, this PR removes some reliance of the tree of libbitcoinkernel header includes on `system.h` (and thus the `ArgsManager` definition) by moving filesystem related functions out of the `system.*` files.
There is already a pair of `fs.h` / `fs.cpp` in the top-level `src/` directory. They were not combined with the files introduced here, to keep the patch cleaner and more importantly because they are often included without the utility functions. The new files are therefore named `fs_helpers` and the existing `fs` files are moved into the util directory.
Further commits splitting more functionality out of `system.h` are still in #25152 and will be submitted in separate PRs once this PR has been processed.
ACKs for top commit:
hebasto:
ACK 00e9b97f37
Tree-SHA512: 31422f148d14ba3c843b99b1550a6fd77c77f350905ca324f93d4f97b652246bc58fa9696c64d1201979cf88733e40be02d262739bb7d417cf22bf506fdb7666
This frees up all associated memory with the map, not only the nodes.
This is necessary in preparation for using the PoolAllocator for
CCoinsMap, which does not actually free any memory on clear().
The fs.* files are already part of the libbitcoin_util library. With the
introduction of the fs_helpers.* it makes sense to move fs.* into the
util/ directory as well.
This is an extraction of filesystem related functions from util/system
into their own utility file.
The background of this commit is an ongoing effort to decouple the
libbitcoinkernel library from the ArgsManager defined in system.h.
Moving these functions out of system.h allows including them from a
separate source file without including the ArgsManager definitions from
system.h.
95ad70ab65 test: Default initialize `should_freeze` to `true` (Hennadii Stepanov)
cea50521fe refactor: Drop no longer used `swap` member functions (Hennadii Stepanov)
a87fb6bee5 clang-tidy: Fix modernize-use-default-member-init in `CScriptCheck` (Hennadii Stepanov)
b4bed5c1f9 refactor: Drop no longer used `CScriptCheck()` default constructor (Hennadii Stepanov)
d8427cc28e refactor: Use move semantics in `CCheckQueue::Loop` (Hennadii Stepanov)
9a0b524139 clang-tidy, test: Fix bugprone-use-after-move in `Correct_Queue_range()` (Hennadii Stepanov)
04831fee6d refactor: Make move semantics explicit for callers (Hennadii Stepanov)
6c2d5972f3 refactor: Use move semantics in `CCheckQueue::Add` (Hennadii Stepanov)
0682003214 test, refactor: Avoid `CScriptCheck::swap` in `transaction_tests` (Hennadii Stepanov)
15209d97c6 consensus, refactor: Avoid `CScriptCheck::swap` in `CheckInputScripts` (Hennadii Stepanov)
Pull request description:
This PR makes code more succinct and readable by using move semantics.
ACKs for top commit:
martinus:
re-ACK 95ad70ab65
achow101:
ACK 95ad70ab65
TheCharlatan:
re-ACK 95ad70ab65
MarcoFalke:
re-ACK 95ad70ab65🚥
Tree-SHA512: adda760891b12d252dc9b823fe7c41eed660364b6fb1a69f17607d7a31eb0bbb82a80d154a7acfaa241b5de37d42a293c2b6e059f26a8e92d88d3a87c99768fb
2c3a90f663 log: on new valid header (James O'Beirne)
e5ce857634 log: net: new header over cmpctblock (James O'Beirne)
Pull request description:
Alternate to #27276.
Devs were [suprised to realize](https://twitter.com/jamesob/status/1637237917201383425) last night that we don't have definitive logging for when a given header was first received.
This logs to the main stream when new headers are received outside of IBD, as well as when headers come in over cmpctblocks. The rationale of not hiding these under log categories is that they may be useful to have widely available when debugging strange network activity, and the marginal volume is modest.
ACKs for top commit:
dergoegge:
Code review ACK 2c3a90f663
achow101:
ACK 2c3a90f663
Sjors:
tACK 2c3a90f663
josibake:
ACK 2c3a90f663
Tree-SHA512: 49fdcbe07799c8adc24143d7e5054a0c93fef120d2e9d5fddbd3b119550d895e2985be6ac10dd1825ea23a6fa5479c1b76d5518c136fbd983fa76c0d39dc354f
4b7aec2951 Add mempool tracepoints (virtu)
Pull request description:
This PR adds multiple mempool tracepoints.
| tracepoint | description |
| ------------- | ------------- |
| `mempool:added` | Is called when a transaction enters the mempool |
| `mempool:removed` | ... when a transaction is removed from the mempool |
| `mempool:replaced` | ... when a transaction is replaced in the mempool |
| `mempool:rejected` | ... when a transaction is rejected from entering the mempool |
The tracepoints are further documented in `docs/tracing.md`. Usage is demonstrated in the example script `contrib/tracing/mempool_monitor.py`. Interface tests are provided in `test/functional/interface_usdt_mempool.py`.
The rationale for passing the removal reason as a string instead of numerically is that the benefits of not having to maintain a redundant enum-string mapping seem to outweigh the small cost of string generation. The reject reason is passed as string as well, although in this instance the string does not have to be generated but is readily available.
ACKs for top commit:
0xB10C:
ACK 4b7aec2951
achow101:
ACK 4b7aec2951
Tree-SHA512: 6deb3ba2d1a061292fb9b0f885f7a5c4d11b109b838102d8a8f4828cd68f5cd03fa3fc64adc6fdf54a08a1eaccce261b0aa90c2b8c33cd5fd3828c8f74978958
Tracepoints for added, removed, replaced, and rejected transactions.
The removal reason is passed as string instead of a numeric value, since
the benefits of not having to maintain a redundant enum-string mapping
seem to outweigh the small cost of string generation. The reject reason
is passed as string as well, although here the string does not have to
be generated but is readily available.
So far, tracepoint PRs typically included two demo scripts: a naive
bpftrace script to show raw tracepoint data and a bcc script for a more
refined view. However, as some of the ongoing changes to bpftrace
introduce a certain degree of unreliability (running some of the
existing bpftrace scripts was not possible with standard kernels and
bpftrace packages on latest stable Ubuntu, Debian, and NixOS), this PR
includes only a single bcc script that fuses the functionality of former
bpftrace and bcc scripts.
b3e78dc91d refactor: Don't use global chainparams in chainstatemanager method (TheCharlatan)
382b692a50 Split non/kernel chainparams (Carl Dong)
edabbc78a3 Add factory functions for Main/Test/Sig/Reg chainparams (Carl Dong)
d938098398 Remove UpdateVersionBitsParameters (Carl Dong)
84b85786f0 Decouple RegTestChainParams from ArgsManager (Carl Dong)
76cd4e7c96 Decouple SigNetChainParams from ArgsManager (Carl Dong)
Pull request description:
This pull request is part of the `libbitcoinkernel` project https://github.com/bitcoin/bitcoin/issues/24303https://github.com/bitcoin/bitcoin/projects/18 and more specifically its "Step 2: Decouple most non-consensus code from libbitcoinkernel". dongcarl is the original author of this patchset, these commits were taken from https://github.com/dongcarl/bitcoin/tree/2022-03-libbitcoinkernel-chainparams-args-only.
#### Context
The bitcoin kernel library currently relies on code containing user configurations through the `ArgsManager`. This is not optimal, since as a stand-alone library it should not rely on bitcoind's argument parsing logic. Instead, its interfaces should accept control and options structs that control the kernel library's desired configuration.
Similar work towards decoupling the `ArgsManager` from the kernel has been done in
https://github.com/bitcoin/bitcoin/pull/25290, https://github.com/bitcoin/bitcoin/pull/25487, https://github.com/bitcoin/bitcoin/pull/25527 and https://github.com/bitcoin/bitcoin/pull/25862.
#### Changes
By moving the `CChainParams` class definition into the kernel and giving it new factory functions `CChainParams::{RegTest,SigNet,Main,TestNet}`it can be constructed without an `ArgsManager` reference, unlike the current factory function `CreateChainParams`.
The first few commits remove uses of `ArgsManager` within `CChainParams`. Then the `CChainParams` definition is moved to a new file in the `kernel/` subdirectory.
ACKs for top commit:
MarcoFalke:
re-ACK b3e78dc91d🛁
ryanofsky:
Code review ACK b3e78dc91d. Only changes since last review were recent review suggestions.
ajtowns:
ACK b3e78dc91d
Tree-SHA512: 3835aca1d3e3c75cc3303dd584bab3a77e58f6c678724a5e359fe4b0e17e0763a00931ee6191f516b9fde50496f59cc691f0709c0254206db3863bbf7ab2cacd
The chainstatemanager m_options.chainparams member variable gets its
value from the global chainparams in init.cpp. This allows
validation.cpp to only include the the kernel chainparams file.
Moves chainparams code not using the ArgsManager to the kernel.
Subsequently use the kernel chainparams header now where possible in
order to further decouple chainparams call sites from gArgs.
fa1b4e5c32 Use steady clock in FlushStateToDisk (MarcoFalke)
1111e2f8b4 Use steady clock in SeedStrengthen and FindBestImplementation (MarcoFalke)
Pull request description:
There may be a theoretical deadlock for the duration of the offset when the system clock is adjusted into a past time while executing `SeedStrengthen`.
Fix this by using steady clock.
Do the same in `FindBestImplementation`, which shouldn't be affected, because it discards outlier measurements. However, doing the same there for consistency seems fine.
Do the same in `FlushStateToDisk`, which should make the flushes more steady, if the system clock is adjusted by a large offset.
ACKs for top commit:
john-moffett:
ACK fa1b4e5c32
willcl-ark:
ACK fa1b4e5c3
Tree-SHA512: cc625e796b186accd53222bd64eb57d0512bc7e588312d254349b542bbc5e5daac348ff2b3b3f7dc5ae0bbbae2ec11fdbf3022cf2164211633765a4b0108e83e
2b373fe49d docs: update assumeutxo.md (James O'Beirne)
87a1108c81 test: add snapshot completion unittests (James O'Beirne)
d70919a88f refactor: make MempoolMutex() public (James O'Beirne)
7300ced9de log: add LoadBlockIndex() message for assumedvalid blocks (James O'Beirne)
d96c59cc5c validation: add ChainMan logic for completing UTXO snapshot validation (James O'Beirne)
f2a4f3376f move-only-ish: init: factor out chainstate initialization (James O'Beirne)
637a90b973 add Chainstate::HasCoinsViews() (James O'Beirne)
c29f26b47b validation: add CChainState::m_disabled and ChainMan::isUsable (James O'Beirne)
5ee22cdafd add ChainstateManager.GetSnapshot{BaseHeight,BaseBlock}() (James O'Beirne)
Pull request description:
This is part of the [assumeutxo project](https://github.com/bitcoin/bitcoin/projects/11) (parent PR: https://github.com/bitcoin/bitcoin/pull/15606)
Part two of replacing https://github.com/bitcoin/bitcoin/pull/24232.
---
When a user activates a snapshot, the serialized UTXO set data is used to create an "assumed-valid" chainstate, which becomes active in an attempt to get the node to network tip as quickly as possible. Simultaneously in the background, the already-existing chainstate continues "conventional" IBD to both accumulate full block data and serve as a belt-and-suspenders to validate the assumed-valid chainstate.
Once the background chainstate's tip reaches the base block of the snapshot used, we set `m_stop_use` on that chainstate and immediately take the hash of its UTXO set; we verify that this matches the assumeutxo value in the source code. Note that while we ultimately want to remove this background chainstate, we don't do so until the following initialization process, when we again check the UTXO set hash of the background chainstate, and if it continues to match, we remove the (now unnecessary) background chainstate, and move the (previously) assumed-valid chainstate into its place. We then reinitialize the chainstate in the normal way.
As noted in previous comments, we could do the filesystem operations "inline" immediately when the background validation completes, but that's basically just an optimization that saves disk space until the next restart. It didn't strike me as worth the risk of moving chainstate data around on disk during runtime of the node, though maybe my concerns are overblown.
The final result of this completion process is a fully-validated chain, where the only evidence that the user synced using assumeutxo is the existence of a `base_blockhash` file in the `chainstate` directory.
ACKs for top commit:
achow101:
ACK 2b373fe49d
Tree-SHA512: a204e1d6e6932dd83c799af3606b01a9faf893f04e9ee1a36d63f2f1ccfa9118bdc1c107d86976aa0312814267e6a42074bf3e2bf1dead4b2513efc6d955e13d
Trigger completion when a background validation chainstate reaches the
same height as a UTXO snapshot, and handle cleaning up the chainstate
on subsequent startup.
75db62ba4c refactor: Move calculation logic out from `CheckSequenceLocksAtTip()` (Hennadii Stepanov)
3bc434f459 refactor: Add `CalculateLockPointsAtTip()` function (Hennadii Stepanov)
Pull request description:
This PR is follow up for bitcoin/bitcoin#22677 and bitcoin/bitcoin#23683.
On master (013daed9ac) it is not obvious that `CheckSequenceLocksAtTip()` function can modify its `LockPoints* lp` parameter which leads to https://github.com/bitcoin/bitcoin/pull/22677#discussion_r762040101.
This PR:
- separates the lockpoint calculate logic from `CheckSequenceLocksAtTip()` function into a new `CalculateLockPointsAtTip()` one
- cleans up the `CheckSequenceLocksAtTip()` function interface
- makes code easier to reason about (hopefully)
ACKs for top commit:
achow101:
ACK 75db62ba4c
stickies-v:
re-ACK 75db62b
Tree-SHA512: 072c3fd9cd1e1b0e0bfc8960a67b01c80a9f16d6778f374b6944ade03a020415ce8b8ab2593b0f5e787059c8cf90af798290b4c826785d41955092f6e12e7486
3141eab9c6 test: add functional test for ScanAndUnlinkAlreadyPrunedFiles (Andrew Toth)
e252909e56 test: add unit test for ScanAndUnlinkAlreadyPrunedFiles (Andrew Toth)
77557dda4a prune: scan and unlink already pruned block files on startup (Andrew Toth)
Pull request description:
There are a few cases where we can mark a block and undo file as pruned in our block index, but not actually remove the files from disk.
1. If we call `FindFilesToPrune` or `FindFilesToPruneManual` and crash before `UnlinkPrunedFiles`.
2. If on Windows there is an open file handle to the file somewhere else when calling `fs::remove` in `UnlinkPrunedFiles` (https://en.cppreference.com/w/cpp/filesystem/remove, https://learn.microsoft.com/en-us/windows/win32/api/fileapi/nf-fileapi-deletefilew#remarks). This could be from another process, or if we are calling `ReadBlockFromDisk`/`ReadRawBlockFromDisk` without having a lock on `cs_main` (which has been allowed since ccd8ef65f9).
This PR mitigates this by scanning all pruned block files on startup after `LoadBlockIndexDB` and unlinking them again.
ACKs for top commit:
achow101:
ACK 3141eab9c6
pablomartin4btc:
re-ACK with added functional test 3141eab9c6.
furszy:
Code review ACK 3141eab9
theStack:
Code-review ACK 3141eab9c6
Tree-SHA512: 6c73bc57838ad1b7e5d441af3c4d6bf4c61c4382e2b86485e57fbb74a61240710c0ceeceb8b4834e610ecfa3175c6955c81ea4b2285fee11ca6383f472979d8d
0af16e7134 doc: add release note for #25574 (Martin Zumsande)
57ef2a4812 validation: report if pruning prevents completion of verification (Martin Zumsande)
0c7785bb25 init, validation: Improve handling if VerifyDB() fails due to insufficient dbcache (Martin Zumsande)
d6f781f1cf validation: return VerifyDBResult::INTERRUPTED if verification was interrupted (Martin Zumsande)
6360b5302d validation: Change return value of VerifyDB to enum type (Martin Zumsande)
Pull request description:
`VerifyDB()` can fail to complete due to insufficient dbcache at the level 3 checks. This PR improves the error handling in this case in the following ways:
- The rpc `-verifychain` now returns false if the check can't be completed due to insufficient cache
- During init, we only log a warning if the default values for `-checkblocks` and `-checklevel` are taken and the check doesn't complete. However, if the user actively specifies one of these args, we return with an InitError if we can't complete the check.
This PR also changes `-verifychain` RPC to return `false` if the verification didn't finish due to missing block data (pruning) or due to being interrupted by the node being shutdown.
Previously, this PR also included a fix for a possible assert during verification - this was done in #27009 (now merged).
ACKs for top commit:
achow101:
ACK 0af16e7134
ryanofsky:
Code review ACK 0af16e7134. Only small suggested changes since the last review, like renaming some of the enum values. I did leave more suggestions, but they are not very important and could be followups
john-moffett:
ACK 0af16e7134
MarcoFalke:
lgtm re-ACK 0af16e7134 🎚
Tree-SHA512: 84b4f767cf9bfbafef362312757c9bf765b41ae3977f4ece840e40c52a2266b1457832df0cdf70440be0aac2168d9b58fc817238630b0b6812f3836ca950bc0e
and remove m_snapshot_validated. This state can now be inferred by the
number of isUsable chainstates.
m_disabled is used to signal that a chainstate should no longer be used
by validation logic; it is used as a sentinel when background validation
completes or if the snapshot chainstate is found to be invalid.
isUsable is a convenience method that incorporates m_disabled.
The rpc command verifychain now fails if the dbcache was not sufficient
to complete the verification at the specified level and depth.
In the same situation, the VerifyDB check during Init will now fail (and lead to
an early shutdown) if the user has explicitly specified -checkblocks or
-checklevel but the check couldn't be executed because of the limited
cache. If the user didn't change any of the two and is using the defaults, log a warning
but don't prevent the node from starting up.
This does not change behavior. It is in preparation for
special handling of the case where VerifyDB doesn't finish
for various reasons, but doesn't fail.
Add CoinsViewOptions struct to remove ArgsManager uses from txdb.
To reduce size of this commit, this moves references to gArgs variable out of
txdb.cpp to calling code in validation.cpp. But these moves are temporary. The
gArgs references in validation.cpp are moved out to calling code in init.cpp in
later commits.
This commit does not change behavior.
The previous behavior, skipping some L3 DisconnectBlock calls,
but still attempting to reconnect these blocks at L4, makes
ConnectBlock assert.
The variable skipped_l3_checks is introduced because even with an
insufficient cache for the L3 checks, the L1/L2 checks in the same
loop should still be completed.
Fixes#25563.
282019cd3d refactor: add kernel/cs_main.* (fanquake)
Pull request description:
One place to find / include `cs_main`.
No more:
> // Actually declared in validation.cpp; can't include because of circular dependency.
> extern RecursiveMutex cs_main;
Ultimately, no more need to include `validation.h` (which also includes (heavy/boost filled) `txmempool.h`) everywhere for `cs_main`. See #26087 for another example of why that is useful.
ACKs for top commit:
ajtowns:
ACK 282019cd3d
Tree-SHA512: 142835b794873e7a09c3246d6101843ae81ec0c6295e6873130c98a2abfa5f7282748d0f1a37237a779cc71c3bc0a75d03b20313ef5398c83d4814215cbc8287
This makes the interface more predictable and useful. The caller
understands one or more transactions failed, and can learn what happened
with each transaction. We already have this information, so we might as
well return it.
It doesn't make sense to do this for other PackageValidationResult
values because:
- PCKG_RESULT_UNSET: this means everything succeeded, so the individual
failures are no longer accurate.
- PCKG_MEMPOOL_ERROR: something went wrong with the mempool logic;
transaction failures might not be meaningful.
- PCKG_POLICY: this means something was wrong with the package as a
whole. The caller should use the PackageValidationState to find the
error, rather than looking at individual MempoolAcceptResults.
This value creates an extremely confusing interface as its existence is
dependent upon implementation details (whether something was submitted
on its own, etc). MempoolAcceptResult::m_effective_feerate is much more
helpful, as it always exists for submitted transactions.
Bug: not setting package_state means package_state.IsValid() == true and
the caller does not know that this failed.
We won't be validating this transaction again, so it makes sense to return this
failure to the caller.
Rename package_state to package_state_quit_early to make it more clear
what this variable is used for and what its scope is.
Co-authored-by: Greg Sanders <gsanders87@gmail.com>
47c4b1f52a mempool: log/halt when CalculateMemPoolAncestors fails unexpectedly (stickies-v)
5481f65849 mempool: add AssumeCalculateMemPoolAncestors helper function (stickies-v)
f911bdfff9 mempool: use util::Result for CalculateMemPoolAncestors (stickies-v)
66e028f739 mempool: use util::Result for CalculateAncestorsAndCheckLimits (stickies-v)
Pull request description:
Upon reviewing the documentation for `CTxMemPool::CalculateMemPoolAncestors`, I noticed `setAncestors` was meant to be an `out` parameter but actually is an `in,out` parameter, as can be observed by adding `assert(setAncestors.empty());` as the first line in the function and running `make check`. This PR fixes this unexpected behaviour and introduces refactoring improvements to make intents and effects of the code more clear.
## Unexpected behaviour
This behaviour occurs only in the package acceptance path, currently only triggered by `testmempoolaccept` and `submitpackage` RPCs.
In `MemPoolAccept::AcceptMultipleTransactions()`, we first call `PreChecks()` and then `SubmitPackage()` with the same `Workspace ws` reference. `PreChecks` leaves `ws.m_ancestors` in a potentially non-empty state, before it is passed on to `MemPoolAccept::SubmitPackage`. `SubmitPackage` is the only place where `setAncestors` isn't guaranteed to be empty before calling `CalculateMemPoolAncestors`. The most straightforward fix is to just forcefully clear `setAncestors` at the beginning of CalculateMemPoolAncestors, which is done in the first bugfix commit.
## Improvements
### Return value instead of out-parameters
This PR updates the function signatures for `CTxMemPool::CalculateMemPoolAncestors` and `CTxMemPool::CalculateAncestorsAndCheckLimits` to use a `util::Result` return type and eliminate both the `setAncestors` `in,out`-parameter as well as the error string. It simplifies the code and makes the intent and effects more explicit.
### Observability
There are 7 instances where we currently call `CalculateMemPoolAncestors` without actually checking if the function succeeded because we assume that it can't fail, such as in [miner.cpp](69b10212ea/src/node/miner.cpp (L399)). This PR adds a new wrapper `AssumeCalculateMemPoolAncestors` function that logs such unexpected failures, or in case of debug builds even halts the program. It's not crucial to the objective, more of an observability improvement that seems sensible to add on here.
ACKs for top commit:
achow101:
ACK 47c4b1f52a
w0xlt:
ACK 47c4b1f52a
glozow:
ACK 47c4b1f52a
furszy:
light code review ACK 47c4b1f5
aureleoules:
ACK 47c4b1f52a
Tree-SHA512: d908dad00d1a5645eb865c4877cc0bae74b9cd3332a3641eb4a285431aef119f9fc78172d38b55c592168a73dae83242e6af3348815f7b37cbe2d448a3a58648
b2aa9e8528 Add release note for MIN_STANDARD_TX_NONWITNESS_SIZE relaxation (Greg Sanders)
8c5b3646b5 Relax MIN_STANDARD_TX_NONWITNESS_SIZE to 65 non-witness bytes (Greg Sanders)
Pull request description:
Since the original fix was set to be a "reasonable" transaction to reduce allocations and the true motivation later revealed, it makes sense to relax this check to something more principled.
There are more exotic transaction patterns that could take advantage of a relaxed requirement, such as 1 input, 1 output OP_RETURN to burn a utxo to fees for CPFP purposes when change isn't practical.
Two changes could be accomplished:
1) Anything not 64 bytes could be allowed
2) Anything above 64 bytes could be allowed
In the Great Consensus Cleanup, suggestion (2)
was proposed as a consensus change, and is the simpler of the two suggestions. It would not allow an "empty" OP_RETURN but would reduce the required padding from 22 bytes to 5.
The functional test is also modified to test the actual case
we care about: 64 bytes
Related mailing list discussions here:
https://lists.linuxfoundation.org/pipermail/bitcoin-dev/2022-October/020995.html
And a couple years earlier:
https://lists.linuxfoundation.org/pipermail/bitcoin-dev/2020-May/017883.html
ACKs for top commit:
achow101:
reACK b2aa9e8528
glozow:
reACK b2aa9e8528
pablomartin4btc:
re-ACK b2aa9e8528
jonatack:
ACK b2aa9e8528 with some suggestions
Tree-SHA512: c1ec1af9ddcf31b2272209a4f1ee0c5607399f8172e5a1dfd4604cf98bfb933810dd9369a5917ad122add003327c9fcf6ee26995de3aca41d5c42dba527991ad
Since the original fix was set to be a "reasonable" transaction
to reduce allocations and the true motivation later revealed,
it makes sense to relax this check to something more principled.
There are more exotic transaction patterns that could take advantage
of a relaxed requirement, such as 1 input, 1 output OP_RETURN to burn
a utxo to fees for CPFP purposes when change isn't practical.
Two changes could be accomplished:
1) Anything not 64 bytes could be allowed
2) Anything above 64 bytes could be allowed
In the Great Consensus Cleanup, suggestion (2) was the route taken.
It would not allow an "empty" OP_RETURN
but would reduce the required padding from 22 bytes to 5.
The functional test is also modified to test the actual case
we care about: 64 bytes
07dfbb5bb8 Make static nLastFlush and nLastWrite Chainstate members (Aurèle Oulès)
Pull request description:
Fixes#22189.
The `static std::multimap<uint256, FlatFilePos> mapBlocksUnknownParent; ` referenced in the issue was already fixed by #25571. I don't believe Chainstate references any other static variables.
ACKs for top commit:
jamesob:
ACK 07dfbb5bb8 ([`jamesob/ackr/26513.1.aureleoules.make_static_nlastflush_a`](https://github.com/jamesob/bitcoin/tree/ackr/26513.1.aureleoules.make_static_nlastflush_a))
theStack:
Concept and code-review ACK 07dfbb5bb8
Tree-SHA512: 0f26463c079bbc5e0e62707d4ca4c8c9bbb99edfa3391d48d4915d24e2a1190873ecd4f9f11da25b44527671cdc82c41fd8234d56a4592a246989448d34406b0
38941a703e refactor: Move `txmempool_entry.h` --> `kernel/mempool_entry.h` (Hennadii Stepanov)
Pull request description:
This PR addresses the https://github.com/bitcoin/bitcoin/pull/17786#discussion_r1027818360:
> why not move it to the right place, that is to `kernel/txmempool_entry.h`?
ACKs for top commit:
MarcoFalke:
review ACK 38941a703e📊
Tree-SHA512: 0145974b63b67ca1d9d89af2dd9d4438beca480c16a563f330da05fec49b8394d7ba20ed83cf7d50b2e19454e006978ebed42b0e07887b98d00210f3201ce9ba
d885bb2f6e test: Test exclusion of OP_RETURN from getblockstats (Fabian Jahr)
ba9d288b24 test: Fix getblockstats test data generator (Fabian Jahr)
2ca5a496c2 rpc: Improve getblockstats (Fabian Jahr)
cb94db119f validation, index: Add unspendable coinbase helper functions (Fabian Jahr)
Pull request description:
Fixes#19885
The genesis block does not have undo data saved to disk so the RPC errored because of that.
ACKs for top commit:
achow101:
ACK d885bb2f6e
aureleoules:
ACK d885bb2f6e
stickies-v:
ACK d885bb2f6
Tree-SHA512: f37bda736ed605b7a41a81eeb4bfbb5d2b8518f847819e5d6a090548a61caf1455623e15165d72589ab3f4478252b00e7b624f9313ad6708cac06dd5edb62e9a
c8dc0e3eaa refactor: Inline `CTxMemPoolEntry` class's functions (Hennadii Stepanov)
75bbe594e5 refactor: Move `CTxMemPoolEntry` class to its own module (Hennadii Stepanov)
Pull request description:
This PR:
- gets rid of the `policy/fees` -> `txmempool` -> `policy/fees` circular dependency
- is an alternative to #13949, which nukes only one circular dependency
ACKs for top commit:
ryanofsky:
Code review ACK c8dc0e3eaa. Just include and whitespace changes since last review, and there's a moveonly commit now so it's very easy to review
theStack:
Code-review ACK c8dc0e3eaa
glozow:
utACK c8dc0e3eaa, agree these changes are an improvement.
Tree-SHA512: 36ece824e6ed3ab1a1e198b30a906c8ac12de24545f840eb046958a17315ac9260c7de26e11e2fbab7208adc3d74918db7a7e389444130f8810548ca2e81af41
db929893ef Faster -reindex by initially deserializing only headers (Larry Ruane)
c72de9990a util: add CBufferedFile::SkipTo() to move ahead in the stream (Larry Ruane)
48a68908ba Add LoadExternalBlockFile() benchmark (Larry Ruane)
Pull request description:
### Background
During the first part of reindexing, `LoadExternalBlockFile()` sequentially reads raw blocks from the `blocks/blk00nnn.dat` files (rather than receiving them from peers, as with initial block download) and eventually adds all of them to the block index. When an individual block is initially read, it can't be immediately added unless all its ancestors have been added, which is rare (only about 8% of the time), because the blocks are not sorted by height. When the block can't be immediately added to the block index, its disk location is saved in a map so it can be added later. When its parent is later added to the block index, `LoadExternalBlockFile()` reads and deserializes the block from disk a second time and adds it to the block index. Most blocks (92%) get deserialized twice.
### This PR
During the initial read, it's rarely useful to deserialize the entire block; only the header is needed to determine if the block can be added to the block index immediately. This change to `LoadExternalBlockFile()` initially deserializes only a block's header, then deserializes the entire block only if it can be added immediately. This reduces reindex time on mainnet by 7 hours on a Raspberry Pi, which translates to around a 25% reduction in the first part of reindexing (adding blocks to the index), and about a 6% reduction in overall reindex time.
Summary: The performance gain is the result of deserializing each block only once, except its header which is deserialized twice, but the header is only 80 bytes.
ACKs for top commit:
andrewtoth:
ACK db929893ef
achow101:
ACK db929893ef
aureleoules:
ACK db929893ef - minor changes and new benchmark since last review
theStack:
re-ACK db929893ef
stickies-v:
re-ACK db929893e
Tree-SHA512: 5a5377192c11edb5b662e18f511c9beb8f250bc88aeadf2f404c92c3232a7617bade50477ebf16c0602b9bd3b68306d3ee7615de58acfd8cae664d28bb7b0136
Since faf44876db, the maxtipage comparison
in IsInitialBlockDownload() has been broken, since the NodeClock::now()
time_point is in the system's native denomination (micrcoseconds).
Without this patch, specifying the maximum allowable -maxtipage
(9223372036854775807) results in a SIGABRT crash.
Co-authored-by: MacroFake <falke.marco@gmail.com>
aaaa7bd0ba iwyu: Add missing includes (MacroFake)
fa9ebec096 Remove g_parallel_script_checks (MacroFake)
fa7c834b9f Move ::fCheckBlockIndex into ChainstateManager (MacroFake)
fa43188d86 Move ::fCheckpointsEnabled into ChainstateManager (MacroFake)
cccca83099 Move ::nMinimumChainWork into ChainstateManager (MacroFake)
fa29d0b57c Move ::hashAssumeValid into ChainstateManager (MacroFake)
faf44876db Move ::nMaxTipAge into ChainstateManager (MacroFake)
Pull request description:
It seems preferable to assign globals to a class (in this case `ChainstateManager`), than to leave them dangling. This should clarify scope for code-readers, as well as clarifying unit test behaviour.
ACKs for top commit:
dergoegge:
Code review ACK aaaa7bd0ba
ryanofsky:
Code review ACK aaaa7bd0ba. No changes since last review, other than rebase
aureleoules:
reACK aaaa7bd0ba
Tree-SHA512: 83ec3ba0fb4f1dad95810d4bd4e578454e0718dc1bdd3a794cc4e48aa819b6f5dad4ac4edab3719bdfd5f89cbe23c2740a50fd56c1ff81c99e521c5f6d4e898d
When a block is initially read from a blk*.dat file during reindexing,
it can be added to the block index only if all of its ancestor blocks
have been added, which is rare. If the block's ancestors have not been
added, the block must be re-read from disk later when it can be added.
This commit: During the initial block read, deserialize only its header,
rather than the entire block, since this is sufficient to determine
if its parent (and thus all its ancestors) has been added. This is a
performance improvement.
Making the checks to identify BIP30 available outside of validation.cpp is needed for reporting and tracking statistics on specific blocks and the UTXO set correctly.
5d3f98d278 refactor: Replace m_params with chainman.GetParams() (Aurèle Oulès)
Pull request description:
Fixes a TODO introduced in #24595.
Removes `m_params` from `CChainState` class and replaces it with `m_chainman.GetParams()`.
ACKs for top commit:
MarcoFalke:
review ACK 5d3f98d278🌎
Tree-SHA512: de0fe31450d281cc7307c0d820495e86c93c7998e77a148db2c703da66cff1059e6560c041f1864913c42075aa24d259c2623d45e929ca0a8056ed330a9f9978
This changes the flag for the bitcoin-chainstate executable. Previously
it was false, now it is the chain's default value (still false for the
main chain).
This changes the minimum chain work for the bitcoin-chainstate
executable. Previously it was uint256{}, now it is the chain's default
minimum chain work.
e899d4ca6f init: limit bip30 exceptions to coinbase txs (Chris Geihsler)
511eb7fdea Ignore problematic blocks in DisconnectBlock (Chris Geihsler)
Pull request description:
Fixes https://github.com/bitcoin/bitcoin/issues/22596
When using checklevel=4, block verification fails because of duplicate coinbase transactions involving blocks 91812 and 91722. There was already a check in place within `ConnectBlock` to ignore the problematic blocks, but `DisconnectBlock` did not contain a similar check to ignore these blocks when called from `VerifyDB`.
By ignoring these two blocks in `DisconnectBlock`, the block verification process succeeds at checklevel=4.
(Note to reviewers: this is my first contribution to Bitcoin Core, so any feedback is most welcome. Thanks in advance for reviewing!)
## Steps to reproduce:
Use the following bitcoin.conf file and start bitcoind. I only used block data through block ~100000 so that the verification process was much faster.
```
assumevalid=0
checkblocks=0
checklevel=4
```
Without this change, you will see the following error when the blocks are verified:
```
2022-04-14T02:56:44Z init message: Verifying blocks…
2022-04-14T02:56:44Z Verifying last 101881 blocks at level 4
2022-04-14T02:56:44Z [0%]...[10%]...[20%]...[30%]...[40%]...ERROR: VerifyDB(): *** coin database inconsistencies found (last 10160 blocks, 142571 good transactions before that)
2022-04-14T02:57:01Z : Corrupted block database detected.
Please restart with -reindex or -reindex-chainstate to recover.
: Corrupted block database detected.
Please restart with -reindex or -reindex-chainstate to recover.
```
With this change, you will see this instead:
```
2022-04-14T02:32:29Z init message: Verifying blocks…
2022-04-14T02:32:29Z Verifying last 101746 blocks at level 4
2022-04-14T02:32:29Z [0%]...[10%]...[20%]...[30%]...[40%]...[50%]...[60%]...[70%]...[80%]...[90%]...[DONE].
2022-04-14T02:32:48Z No coin database inconsistencies in last 101746 blocks (226126 transactions)
```
ACKs for top commit:
laanwj:
Code review ACK e899d4ca6f
achow101:
ACK e899d4ca6f
jamesob:
(Biased) ACK e899d4ca6f ([`jamesob/ackr/24851.2.seejee.init_ignore_bip_30_verif`](https://github.com/jamesob/bitcoin/tree/ackr/24851.2.seejee.init_ignore_bip_30_verif))
Tree-SHA512: d2f6d25e9619aee32c1a73fe846b1b587698eaa5a4994fa6424f1038f45654f9fd52b74a69843cc84d90168d74827130ccf8e9201502f5d52281acdb20429291
bf95976061 doc: add note about snapshot chainstate init (James O'Beirne)
e4d7995286 test: add testcases for snapshot initialization (James O'Beirne)
cced4e7336 test: move-only-ish: factor out LoadVerifyActivateChainstate() (James O'Beirne)
51fc9241c0 test: allow on-disk coins and block tree dbs in tests (James O'Beirne)
3c361391b8 test: add reset_chainstate parameter for snapshot unittests (James O'Beirne)
00b357c215 validation: add ResetChainstates() (James O'Beirne)
3a29dfbfb2 move-only: test: make snapshot chainstate setup reusable (James O'Beirne)
8153bd9247 blockmanager: avoid undefined behavior during FlushBlockFile (James O'Beirne)
ad67ff377c validation: remove snapshot datadirs upon validation failure (James O'Beirne)
34d1590331 add utilities for deleting on-disk leveldb data (James O'Beirne)
252abd1e8b init: add utxo snapshot detection (James O'Beirne)
f9f1735f13 validation: rename snapshot chainstate dir (James O'Beirne)
d14bebf100 db: add StoragePath to CDBWrapper/CCoinsViewDB (James O'Beirne)
Pull request description:
This is part of the [assumeutxo project](https://github.com/bitcoin/bitcoin/projects/11) (parent PR: https://github.com/bitcoin/bitcoin/pull/15606)
---
Half of the replacement for #24232. The original PR grew larger than expected throughout the review process.
This change adds the ability to initialize a snapshot-based chainstate during init if one is detected on disk. This is of course unused as of now (aside from in unittests) given that we haven't yet enabled actually loading snapshots.
Don't be scared! There are some big move-only commits in here.
Accompanying changes include:
- moving the snapshot coinsdb directory from being called `chainstate_[base blockhash]` to `chainstate_snapshot`, since we only support one snapshot in use at a time. This simplifies some logic, but it necessitates writing that base blockhash out to a file within the coinsdb dir. See [discussion here](https://github.com/bitcoin/bitcoin/pull/24232#discussion_r832762880).
- adding a simple fix in `FlushBlockFile()` that avoids a crash when attemping to flush to disk before `LoadBlockIndexDB()` is called, which happens when calling `MaybeRebalanceCaches()` during multiple chainstate init.
- improving the unittest to allow testing with on-disk chainstates - necessary to test a simulated restart and re-initialization.
ACKs for top commit:
naumenkogs:
utACK bf95976061
ariard:
Code Review ACK bf9597606
ryanofsky:
Code review ACK bf95976061. Changes since last review: rebasing, switching from CAutoFile to AutoFile, adding comments, switching from BOOST_CHECK to Assert in test util, using chainman.GetMutex() in tests, destroying one ChainstateManager before creating a new one in tests
fjahr:
utACK bf95976061
aureleoules:
ACK bf95976061
Tree-SHA512: 15ae75caf19f8d12a12d2647c52897904d27b265a7af6b4ae7b858592eeadb8f9da6c2394b6baebec90adc28742c053e3eb506119577dae7c1e722ebb3b7bcc0
bcb0cacac2 reindex, log, test: fixes#21379 (mruddy)
Pull request description:
Fixes#21379.
The blocks/blk?????.dat files are mutated and become increasingly malformed, or corrupt, as a result of running the re-indexing process.
The mutations occur after the re-indexing process has finished, as new blocks are appended, but are a result of a re-indexing process miscalculation that lingers in the block manager's `m_blockfile_info` `nSize` data until node restart.
These additions to the blk files are non-fatal, but also not desirable.
That is, this is a form of data corruption that the reading code is lenient enough to process (it skips the extra bytes), but it adds some scary looking log messages as it encounters them.
The summary of the problem is that the re-index process double counts the size of the serialization header (magic message start bytes [4 bytes] + length [4 bytes] = 8 bytes) while calculating the blk data file size (both values already account for the serialization header's size, hence why it is over accounted).
This bug manifests itself in a few different ways, after re-indexing, when a new block from a peer is processed:
1. If the new block will not fit into the last blk file processed while re-indexing, while remaining under the 128MiB limit, then the blk file is flushed to disk and truncated to a size that is 8 greater than it should be. The truncation adds zero bytes (see `FlatFileSeq::Flush` and `TruncateFile`).
1. If the last blk file processed while re-indexing has logical space for the new block under the 128 MiB limit:
1. If the blk file was not already large enough to hold the new block, then the zeros are, in effect, added by `fseek` when the file is opened for writing. Eight zero bytes are added to the end of the last blk file just before the new block is written. This happens because the write offset is 8 too great due to the miscalculation. The result is 8 zero bytes between the end of the last block and the beginning of the next block's magic + length + block.
1. If the blk file was already large enough to hold the new block, then the current existing file contents remain in the 8 byte gap between the end of the last block and the beginning of the next block's magic + length + block. Commonly, when this occcurs, it is due to the blk file containing blocks that are not connected to the block tree during reindex and are thus left behind by the reindex process and later overwritten when new blocks are added. The orphaned blocks can be valid blocks, but due to the nature of concurrent block download, the parent may not have been retrieved and written by the time the node was previously shutdown.
ACKs for top commit:
LarryRuane:
tested code-review ACK bcb0cacac2
ryanofsky:
Code review ACK bcb0cacac2. This is a disturbing bug with an easy fix which seems well-worth merging.
mzumsande:
ACK bcb0cacac2 (reviewed code and did some testing, I agree that it fixes the bug).
w0xlt:
tACK bcb0cacac2
Tree-SHA512: acc97927ea712916506772550451136b0f1e5404e92df24cc05e405bb09eb6fe7c3011af3dd34a7723c3db17fda657ae85fa314387e43833791e9169c0febe51
fabf1cdb20 Use steady clock for bench logging (MacroFake)
faed342a23 scripted-diff: Rename time symbols (MacroFake)
Pull request description:
Instead of using `0.001` and similar constants to "convert" an int64_t to milliseconds, use the type-safe `Ticks<>` helper. Also, use steady clock instead of system clock, since the durations are used for benchmarking.
ACKs for top commit:
fanquake:
ACK fabf1cdb20 - validation bench output still looks sane.
Tree-SHA512: e6525b5fdad6045ca500c56014897d7428ad288aaf375933d3b5939feddf257f6910d562eb66ebcde9186bef9a604ee8d763a318253838318d59df2a285be7c2
Simplifies function signatures by removing repetition of all the
ancestor/descendant limits, and increases readability by being
more verbose by naming the limits, while still reducing the LoC.
fa521c9603 Use steady clock for all millis bench logging (MacroFake)
Pull request description:
Currently `GetTimeMillis` is used for bench logging in milliseconds integral precision. Replace it to use a steady clock that is type-safe and steady.
Microsecond or float precision can be done in a follow-up.
ACKs for top commit:
fanquake:
ACK fa521c9603 - started making the same change.
Tree-SHA512: 86a810e496fc663f815acb8771a6c770331593715cde85370226685bc50c13e8e987e3c5efd0b4e48b36ebd2372255357b709204bac750d41e94a9f7d9897fa6
If a UTXO snapshot fails to validate, don't leave the resulting datadir
on disk as this will confuse initialization on next startup and we'll
get an assertion error.
Used in later commits to remove leveldb directories for
- invalid snapshot chainstates, and
- background-vaildation chainstates that have finished serving their
purpose.
Add functionality for activating a snapshot-based chainstate if one is
detected on-disk.
Also cautiously initialize chainstate cache usages so that we don't
somehow blow past our cache allowances during initialization, then
rebalance at the end of init.
Co-authored-by: Russell Yanofsky <russ@yanofsky.org>
This changes the snapshot's leveldb chainstate dir name from
`chainstate_[blockhash]` to `chainstate_snapshot`. This simplifies
later logic that loads snapshot data, and enforces the limitation
of a single snapshot at any given time.
Since we still need to persis the blockhash of the base block, we
write that out to a file (`chainstate_snapshot/base_blockhash`) for
later use during initialization, so that we can reinitialize the
snapshot chainstate.
Co-authored-by: Russell Yanofsky <russ@yanofsky.org>
3add234546 ui: show header pre-synchronization progress (Pieter Wuille)
738421c50f Emit NotifyHeaderTip signals for pre-synchronization progress (Pieter Wuille)
376086fc5a Make validation interface capable of signalling header presync (Pieter Wuille)
93eae27031 Test large reorgs with headerssync logic (Suhas Daftuar)
355547334f Track headers presync progress and log it (Pieter Wuille)
03712dddfb Expose HeadersSyncState::m_current_height in getpeerinfo() (Suhas Daftuar)
150a5486db Test headers sync using minchainwork threshold (Suhas Daftuar)
0b6aa826b5 Add unit test for HeadersSyncState (Suhas Daftuar)
83c6a0c524 Reduce spurious messages during headers sync (Suhas Daftuar)
ed6cddd98e Require callers of AcceptBlockHeader() to perform anti-dos checks (Suhas Daftuar)
551a8d957c Utilize anti-DoS headers download strategy (Suhas Daftuar)
ed470940cd Add functions to construct locators without CChain (Pieter Wuille)
84852bb6bb Add bitdeque, an std::deque<bool> analogue that does bit packing. (Pieter Wuille)
1d4cfa4272 Add function to validate difficulty changes (Suhas Daftuar)
Pull request description:
New nodes starting up for the first time lack protection against DoS from low-difficulty headers. While checkpoints serve as our protection against headers that fork from the main chain below the known checkpointed values, this protection only applies to nodes that have been able to download the honest chain to the checkpointed heights.
We can protect all nodes from DoS from low-difficulty headers by adopting a different strategy: before we commit to storing a header in permanent storage, first verify that the header is part of a chain that has sufficiently high work (either `nMinimumChainWork`, or something comparable to our tip). This means that we will download headers from a given peer twice: once to verify the work on the chain, and a second time when permanently storing the headers.
The p2p protocol doesn't provide an easy way for us to ensure that we receive the same headers during the second download of peer's headers chain. To ensure that a peer doesn't (say) give us the main chain in phase 1 to trick us into permanently storing an alternate, low-work chain in phase 2, we store commitments to the headers during our first download, which we validate in the second download.
Some parameters must be chosen for commitment size/frequency in phase 1, and validation of commitments in phase 2. In this PR, those parameters are chosen to both (a) minimize the per-peer memory usage that an attacker could utilize, and (b) bound the expected amount of permanent memory that an attacker could get us to use to be well-below the memory growth that we'd get from the honest chain (where we expect 1 new block header every 10 minutes).
After this PR, we should be able to remove checkpoints from our code, which is a nice philosophical change for us to make as well, as there has been confusion over the years about the role checkpoints play in Bitcoin's consensus algorithm.
Thanks to Pieter Wuille for collaborating on this design.
ACKs for top commit:
Sjors:
re-tACK 3add234546
mzumsande:
re-ACK 3add234546
sipa:
re-ACK 3add234546
glozow:
ACK 3add234546
Tree-SHA512: e7789d65f62f72141b8899eb4a2fb3d0621278394d2d7adaa004675250118f89a4e4cb42777fe56649d744ec445ad95141e10f6def65f0a58b7b35b2e654a875
This makes a number of changes:
- Get rid of the verification_progress argument in the node interface
NotifyHeaderTip (it was always 0.0).
- Instead of passing a CBlockIndex* in the UI interface's NotifyHeaderTip,
send separate height, timestamp fields. This is becuase in headers presync,
no actual CBlockIndex object is available.
- Add a bool presync argument to both of the above, to identify signals
pertaining to the first headers sync phase.
In order to prevent memory DoS, we must ensure that we don't accept a new
header into memory until we've performed anti-DoS checks, such as verifying
that the header is part of a sufficiently high work chain. This commit adds a
new argument to AcceptBlockHeader() so that we can ensure that all call-sites
which might cause a new header to be accepted into memory have to grapple with
the question of whether the header is safe to accept, or needs further
validation.
This patch also fixes two places where low-difficulty-headers could have been
processed without such validation (processing an unrequested block from the
network, and processing a compact block).
Credit to Niklas Gögge for noticing this issue, and thanks to Sjors Provoost
for test code.
Avoid permanently storing headers from a peer, unless the headers are part of a
chain with sufficiently high work. This prevents memory attacks using low-work
headers.
Designed and co-authored with Pieter Wuille.
Move ChainstateManager options into m_options struct to simplify class
initialization, organize class members, and to name external option variables
differently than internal state variables.
This change was originally in #25862, but it was suggested to split off in
https://github.com/bitcoin/bitcoin/pull/25862#discussion_r951459817 so it could
be merged earlier and reduce conflicts with other PRs.
1dc03dda05 [doc] remove non-signaling mentions of BIP125 (glozow)
32024d40f0 scripted-diff: remove mention of BIP125 from non-signaling var names (glozow)
Pull request description:
We have pretty thorough documentation of our RBF policy in doc/policy/mempool-replacements.md. It enumerates each rule with several sentences of rationale. Also, each rule pretty much has its own function (3 and 4 share one), with extensive comments. The doc states explicitly that our rules are similar but differ from BIP125, and contains a record of historical changes to RBF policy.
We should not use "BIP125" as synonymous with our RBF policy because:
- Our RBF policy is different from what is specified in BIP125, for example:
- the BIP does not mention our rule about the replacement feerate being higher (our Rule 6)
- the BIP uses minimum relay feerate for Rule 4, while we have used incremental relay feerate since #9380
- the "inherited signaling" question (CVE-2021-31876). Call it discrepancy, ambiguous wording, doc misinterpretation, or implementation details, I would recommend users refer to doc/policy/mempool-replacements.md
- the signaling policy is configurable, see #25353
- Our RBF policy may change further
- We have already marked BIP125 as only "partially implemented" in docs/bips.md since 1fd49eb498
- See comments from people who are not me recently:
- https://github.com/bitcoin/bitcoin/pull/25038#discussion_r909507429
- https://github.com/bitcoin/bitcoin/pull/25575#issuecomment-1179519204
This PR removes all non-signaling mentions of BIP125 (if people feel strongly, we can remove all mentions of BIP125 period). It may be useful to refer to the concept of "tx opts in to RBF if it has at least one nSequence less than (0xffffffff - 1)" as "BIP125 signaling" because:
- It is succint.
- It has already been widely marketed as BIP125 opt-in signaling.
- Our API uses it when referring to signaling (e.g. getmempoolentry["bip125-replaceable"] and wallet error message "not BIP 125 replaceable"). Changing those is more invasive.
- If/when we have other ways to signal in the future, we can disambiguate them this way. See #25038 which proposes another way of signaling, and where I pulled these commits from.
Alternatives:
- Changing our policy to match BIP125. This doesn't make sense as, for example, we would have to remove the requirement that a replacement tx has a higher feerate (Rule 6).
- Changing BIP125 to match what we have. This doesn't make sense as it would be a significant change to a BIP years after it was finalized and already used as a spec to implement RBF in other places.
- Document our policy as a new BIP and give it a number. This might make sense if we don't expect things to change a lot, and can be done as a next step.
ACKs for top commit:
darosior:
ACK 1dc03dda05
ariard:
ACK 1dc03dda
t-bast:
ACK 1dc03dda05
Tree-SHA512: a3adc2039ec5785892d230ec442e50f47f7062717392728152bbbe27ce1c564141f85253143f53cb44e1331cf47476d74f5d2f4b3cd873fc3433d7a0aa783e02
eeee5ada23 Make adjusted time type safe (MacroFake)
fa3be799fe Add time helpers (MacroFake)
Pull request description:
This makes follow-ups easier to review. Also, it makes sense by itself.
ACKs for top commit:
ryanofsky:
Code review ACK eeee5ada23. Confirmed type changes and equivalent code changes only.
Tree-SHA512: 51bf1ae5428552177286113babdd49e82459d6c710a07b6e80a0a045d373cf51045ee010461aba98e0151d8d71b9b3b5f8f73e302d46ba4558e0b55201f99e9f
9376a6dae4 refactor: make active_chain_tip a reference (Aurèle Oulès)
Pull request description:
This PR fixes a TODO introduced in #21055.
Makes `active_chain_tip` argument in `CheckFinalTxAtTip` function a reference instead of a pointer.
ACKs for top commit:
dongcarl:
ACK 9376a6dae4
Tree-SHA512: c36d1769e0b9598b7f79334704b26b73e958d54caa3bd7e4eff954f3964fcf3f5e3a44a5a760497afad51b76e1614c86314fe035e4083c855e3574a620de7f4d
Our RBF policy is different from the rules specified in BIP125. For
example, the BIP does not mention Rule 6, and our Rule 4 uses the
(configurable) incremental relay feerate (distinct from the
minimum relay feerate). Those interested in our policy should refer to
doc/policy/mempool-replacements.md instead. These rules may also
continue to diverge with package RBF and other RBF improvements. Keep
references to the BIP125 signaling wrt sequence numbers, since that is
still correct and widely used. It is helpful to refer to this as "BIP125
signaling" since it is unambiguous and succint, especially if we have
multiple ways to signal replaceability in the future.
The rule numbers in doc/policy/mempool-replacements.md correspond
largely to those of BIP 125, so we can still refer to them like "Rule 5."
faab8dceb3 Remove unused SetTip(nullptr) code (MacroFake)
Pull request description:
Now that this path is no longer used after commit b51e60f914, we can remove it.
Future code should reset `CChain` by simply discarding it and constructing a fresh one.
ACKs for top commit:
ryanofsky:
Code review ACK faab8dceb3. Just moved an assert statement since last review
Tree-SHA512: 7dc273b11133d85d32ca2a69c0c7c07b39cdd338141ef5b51496e7de334a809864d5459eb95535497866c8b1e468aae84ed8f91b543041e6ee20130d5622874e
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.
Our RBF policy is different from the rules specified in BIP125 (refer to
doc/policy/mempool-replacements.md instead), and will continue to
diverge with package RBF. Keep references to BIP125 sequence number,
since that is still useful and correct.
-BEGIN VERIFY SCRIPT-
ren() { sed -i "s:\<$1\>:$2:g" $(git grep -l "\<$1\>" ./src ./test); }
ren m_allow_bip125_replacement m_allow_replacement
ren allow_bip125_replacement allow_replacement
ren MAX_BIP125_REPLACEMENT_CANDIDATES MAX_REPLACEMENT_CANDIDATES
-END VERIFY SCRIPT-
dd065dae9f refactor: Make mapBlocksUnknownParent local, and rename it (Hennadii Stepanov)
Pull request description:
This PR is a second attempt at #19594. This PR has two motivations:
- Improve code hygiene by eliminating a global variable, `mapBlocksUnknownParent`
- Fix fuzz test OOM when running too long ([see #19594 comment](https://github.com/bitcoin/bitcoin/pull/19594#issuecomment-958801638))
A minor added advantage is to release `mapBlocksUnknownParent` memory when the reindexing phase is done. The current situation is somewhat similar to a memory leak because this map exists unused for the remaining lifetime of the process. It's true that this map should be empty of data elements after use, but its internal metadata (indexing structures, etc.) can have non-trivial size because there can be many thousands of simultaneous elements in this map.
This PR helps our efforts to reduce the use of global variables. This variable isn't just global, it's hidden inside a function (it looks like a local variable but has the `static` attribute).
This global variable exists because the `-reindex` processing code calls `LoadExternalBlockFile()` multiple times (once for each block file), but that function must preserve some state between calls (the `mapBlocksUnknownParent` map). This PR fixes this by allocating this map as a local variable in the caller's scope and passing it in on each call. When reindexing completes, the map goes out of scope and is deallocated.
I tested this manually by reindexing on mainnet and signet. Also, the existing `feature_reindex.py` functional test passes.
ACKs for top commit:
mzumsande:
re-ACK dd065dae9f
theStack:
re-ACK dd065dae9f
shaavan:
reACK dd065dae9f
Tree-SHA512: 9cd20e44d2fa1096dd405bc107bc065ea8f904f5b3f63080341b08d8cf57b790df565f58815c2f331377d044d5306708b4bf6bdfc5ef8d0ed85d8e97d744732c
3a61fc56a0 refactor: move CBlockIndex#ToString() from header to implementation (Jon Atack)
57865eb512 CDiskBlockIndex: rename GetBlockHash() to ConstructBlockHash() (Jon Atack)
99e8ec8721 CDiskBlockIndex: remove unused ToString() class member (Jon Atack)
14aeece462 CBlockIndex: ensure phashBlock is not nullptr before dereferencing (Jon Atack)
Pull request description:
Fix a few design issues, potential footguns and inconsistent behavior in the CBlockIndex and CDiskBlockIndex classes.
- Ensure phashBlock in `CBlockIndex#GetBlockHash()` is not nullptr before dereferencing and remove a now-redundant assert preceding a GetBlockHash() caller. This protects against UB here, and in case of failure (which would indicate a consensus bug), the debug log will print `bitcoind: chain.h:265: uint256 CBlockIndex::GetBlockHash() const: Assertion 'phashBlock != nullptr' failed. Aborted` instead of `Segmentation fault`.
- Remove the unused `CDiskBlockIndex#ToString()` class member, and mark the inherited `CBlockIndex#ToString()` public interface member as deleted to disallow calling it in the derived CDiskBlockIndex class.
- Rename the `CDiskBlockIndex GetBlockHash()` class member to `ConstructBlockHash()`, which also makes sense as they perform different operations to return a blockhash, and mark the inherited `CBlockIndex#GetBlockHash()` public interface member as deleted to disallow calling it in the derived CDiskBlockIndex class.
- Move `CBlockIndex#ToString()` from header to implementation, which also allows dropping `tinyformat.h` from the header file.
Rationale and discussion regarding the CDiskBlockIndex changes:
Here is a failing test on master that demonstrates the inconsistent behavior of the current design: calling the same inherited public interface functions on the same CDiskBlockIndex object should yield identical behavior, but does not.
```diff
diff --git a/src/test/validation_chainstatemanager_tests.cpp b/src/test/validation_chainstatemanager_tests.cpp
index 6dc522b421..dac3840f32 100644
--- a/src/test/validation_chainstatemanager_tests.cpp
+++ b/src/test/validation_chainstatemanager_tests.cpp
@@ -240,6 +240,15 @@ BOOST_FIXTURE_TEST_CASE(chainstatemanager_activate_snapshot, TestChain100Setup)
const CBlockIndex* tip = chainman.ActiveTip();
BOOST_CHECK_EQUAL(tip->nChainTx, au_data.nChainTx);
+ // CDiskBlockIndex "is a" CBlockIndex, as it publicly inherits from it.
+ // Test that calling the same inherited interface functions on the same
+ // object yields identical behavior.
+ CDiskBlockIndex index{tip};
+ CBlockIndex *pB = &index;
+ CDiskBlockIndex *pD = &index;
+ BOOST_CHECK_EQUAL(pB->GetBlockHash(), pD->GetBlockHash());
+ BOOST_CHECK_EQUAL(pB->ToString(), pD->ToString());
```
(build and run: `$ ./src/test/test_bitcoin -t validation_chainstatemanager_tests`)
The GetBlockHash() test assertion only passes on master because the different methods invoked by the current design happen to return the same result. If one of the two is changed, it fails like the ToString() assertion does.
Redefining inherited non-virtual functions is well-documented as incorrect design to avoid inconsistent behavior (see Scott Meyers, Effective C++, Item 36). Class usage is confusing when the behavior depends on the pointer definition instead of the object definition (static binding happening where dynamic binding was expected). This can lead to unsuspected or hard-to-track bugs.
Outside of critical hot spots, correctness usually comes before optimisation, but the current design dates back to main.cpp and it may possibly have been chosen to avoid the overhead of dynamic dispatch. This solution does the same: the class sizes are unchanged and no vptr or vtbl is added.
There are better designs for doing this that use composition instead of inheritance, or that separate the public interface from the private implementations. One example of the latter would be a non-virtual public interface that calls private virtual implementation methods, i.e. the Template pattern via the Non-Virtual Interface (NVI) idiom.
ACKs for top commit:
vasild:
ACK 3a61fc56a0
Tree-SHA512: 9ff358ab0a6d010b8f053ad8303c6d4d061e62d9c3755a56b9c9f5eab855d02f02bee42acc77dfa0cbf4bb5cb775daa72d675e1560610a29bd285c46faa85ab7
and remove a now-redundant assert preceding a GetBlockHash() caller.
This protects against UB here, and in case of failure (which would
indicate a consensus bug), the debug log will print
bitcoind: chain.h:265: uint256 CBlockIndex::GetBlockHash() const: Assertion `phashBlock != nullptr' failed.
Aborted
instead of
Segmentation fault
facc2fa7b8 Use AutoFile where possible (MacroFake)
6666803c89 streams: Add AutoFile without ser-type and ser-version (MacroFake)
Pull request description:
This was done in the context of https://github.com/bitcoin/bitcoin/pull/25284 , but I think it also makes sense standalone.
The basic idea is that serialization type should not be initialized when it is not needed. Same for the serialization version.
So do this here for `AutoFile`. `CAutoFile` remains in places where it is not yet possible.
ACKs for top commit:
laanwj:
Code review ACK facc2fa7b8
fanquake:
ACK facc2fa7b8
Tree-SHA512: d82d024d55af57565ac53d9d1517afafc12b46964effba0332de62a6c77869356fa77f89e6d4834438fff44c45b64fccdf5a1358bfea03e28dfe55013b3c099d
Also:
1. Have CChainState::LoadMempool and ::ThreadImport take in paths and
pass it through untouched to LoadMempool.
2. Make LoadMempool exit early if the load_path is empty.
3. Adjust the call to ::ThreadImport in ::AppInitMain to correctly pass
in an empty path if mempool persistence is disabled.
Not only does this increase coverage, it is also more correct in that
when ::LoadMempool is called with a mempool and chainstate, it calls
AcceptToMemoryPool with just the chainstate.
AcceptToMemoryPool will then act on the chainstate's mempool via
CChainState::GetMempool, which may be different from the mempool
originally passed to ::LoadMempool. (In this fuzz test's case, it
definitely is different)
Also, move DummyChainstate to its own file since it's now used by the
validation_load_mempool fuzz test to replace CChainState's m_mempool.
m_is_loaded/IsLoaded() doesn't actually indicate whether or not the
mempool was successfully, loaded, but rather if a load has been
attempted and did not result in a catastrophic ShutdownRequested.
-BEGIN VERIFY SCRIPT-
find_regex="\bm_is_loaded\b" \
&& git grep -l -E "$find_regex" \
| xargs sed -i -E "s@$find_regex@m_load_tried@g"
find_regex="\bIsLoaded\b" \
&& git grep -l -E "$find_regex" \
| xargs sed -i -E "s@$find_regex@GetLoadTried@g"
find_regex="\bSetIsLoaded\b" \
&& git grep -l -E "$find_regex" \
| xargs sed -i -E "s@$find_regex@SetLoadTried@g"
-END VERIFY SCRIPT-
Also introduce node::{ShouldPersistMempool,MempoolPath} helper functions
in node/mempool_persist_args.{h,cpp} which are used by non-kernel
DumpMempool callers to determine whether or not to automatically dump
the mempool and where to dump it to.
This new node policy setting enables to accept replaced-by-fee
transaction without inspection of the replaceability signaling
as described in BIP125 "explicit signaling".
If turns on, the node mempool accepts transaction replacement
as described in `policy/mempool-replacements.md`.
The default setting value is `false`, implying opt-in RBF
is enforced.
d1684beabe fees: Pass in a filepath instead of referencing gArgs (Carl Dong)
9a3d825c30 init: Remove redundant -*mempool*, -limit* queries (Carl Dong)
6c5c60c412 mempool: Use m_limit for UpdateTransactionsFromBlock (Carl Dong)
9e93b10301 node/ifaces: Use existing MemPoolLimits (Carl Dong)
38af2bcf35 mempoolaccept: Use limits from mempool in constructor (Carl Dong)
9333427014 mempool: Introduce (still-unused) MemPoolLimits (Carl Dong)
716bb5fbd3 scripted-diff: Rename anc/desc size limit vars to indicate SI unit (Carl Dong)
1ecc77321d scripted-diff: Rename DEFAULT_MEMPOOL_EXPIRY to indicate time unit (Carl Dong)
aa9141cd81 mempool: Pass in -mempoolexpiry instead of referencing gArgs (Carl Dong)
51c7a41a5e init: Only determine maxmempool once (Carl Dong)
386c9472c8 mempool: Make GetMinFee() with custom size protected (Carl Dong)
82f00de7a6 mempool: Pass in -maxmempool instead of referencing gArgs (Carl Dong)
f1941e8bfd pool: Add and use MemPoolOptions, ApplyArgsManOptions (Carl Dong)
0199bd35bb fuzz/rbf: Add missing TestingSetup (Carl Dong)
ccbaf546a6 scripted-diff: Rename DEFAULT_MAX_MEMPOOL_SIZE to indicate SI unit (Carl Dong)
fc02f77ca6 ArgsMan: Add Get*Arg functions returning optional (Carl Dong)
Pull request description:
This is part of the `libbitcoinkernel` project: #24303, https://github.com/bitcoin/bitcoin/projects/18
-----
As mentioned in the Stage 1 Step 2 description of [the `libbitcoinkernel` project](https://github.com/bitcoin/bitcoin/issues/24303), `ArgsManager` will not be part of `libbitcoinkernel`. Therefore, it is important that we remove any dependence on `ArgsManager` by code that will be part of `libbitcoinkernel`. This is the first in a series of PRs aiming to achieve this.
This PR removes `CTxMemPool+MempoolAccept`'s dependency on `ArgsManager` by introducing a `CTxMemPool::Options` struct, which is used to specify `CTxMemPool`'s various options at construction time.
These options are:
- `-maxmempool` -> `CTxMemPool::Options::max_size`
- `-mempoolexpiry` -> `CTxMemPool::Options::expiry`
- `-limitancestorcount` -> `CTxMemPool::Options::limits::ancestor_count`
- `-limitancestorsize` -> `CTxMemPool::Options::limits::ancestor_size`
- `-limitdescendantcount` -> `CTxMemPool::Options::limits::descendant_count`
- `-limitdescendantsize` -> `CTxMemPool::Options::limits::descendant_size`
More context can be gleaned from the commit messages. The important commits are:
- 56eb479ded8bfb2ef635bb6f3b484f9d5952c70d "pool: Add and use MemPoolOptions, ApplyArgsManOptions"
- a1e08b70f3068f4e8def1c630d8f50cd54da7832 "mempool: Pass in -maxmempool instead of referencing gArgs"
- 6f4bf3ede5812b374828f08fc728ceded2f10024 "mempool: Pass in -mempoolexpiry instead of referencing gArgs"
- 5958a7fe4806599fc620ee8c1a881ca10fa2dd16 "mempool: Introduce (still-unused) MemPoolLimits"
Reviewers: Help needed in the following commits (see commit messages):
- a1e08b70f3068f4e8def1c630d8f50cd54da7832 "mempool: Pass in -maxmempool instead of referencing gArgs"
- 0695081a797e9a5d7787b78b0f8289dafcc6bff7 "node/ifaces: Use existing MemPoolLimits"
Note to Reviewers: There are perhaps an infinite number of ways to architect `CTxMemPool::Options`, the current one tries to keep it simple, usable, and flexible. I hope we don't spend too much time arguing over the design here since that's not the point. In the case that you're 100% certain that a different design is strictly better than this one in every regard, please show us a fully-implemented branch.
-----
TODO:
- [x] Use the more ergonomic `CTxMemPool::Options` where appropriate
- [x] Doxygen comments for `ApplyArgsManOptions`, `MemPoolOptions`
-----
Questions for Reviewers:
1. Should we use `std::chrono::seconds` for `CTxMemPool::Options::expiry` and `CTxMemPool::m_expiry` instead of an `int64_t`? Something else? (`std::chrono::hours`?)
2. Should I merge `CTxMemPool::Limits` inside `CTxMemPool::Options`?
ACKs for top commit:
MarcoFalke:
ACK d1684beabe🍜
ryanofsky:
Code review ACK d1684beabe. Just minor cleanups since last review, mostly switching to brace initialization
Tree-SHA512: 2c138e52d69f61c263f1c3648f01c801338a8f576762c815f478ef5148b8b2f51e91ded5c1be915e678c0b14f6cfba894b82afec58d999d39a7bb7c914736e0b
Since:
- UpdateTransactionsFromBlock is only called by
MaybeUpdateMempoolForReorg, which calls it with the gArgs-determined
ancestor limits
- UpdateForDescendants is only called by UpdateTransactionsFromBlock
with the ancestor limits unchanged
We can remove the requirement to specify the ancestor limits for both
UpdateTransactionsFromBlock and UpdateForDescendants and just use the
values in the m_limits member.
Also move some removed comments to MemPoolLimits struct members.
The uint64_t cast in UpdateForDescendants is not new behavior,
see the diff in CChainState::MaybeUpdateMempoolForReorg for where they
were previously.
Better to be explicit when it comes to sizes to avoid unintentional
bugs. We use MB and KB all over the place.
-BEGIN VERIFY SCRIPT-
find_regex="DEFAULT_(ANCESTOR|DESCENDANT)_SIZE_LIMIT" \
&& git grep -l -E "$find_regex" \
| xargs sed -i -E "s@$find_regex@\0_KVB@g"
-END VERIFY SCRIPT-
- Store the mempool expiry (-mempoolexpiry) in CTxMemPool as a
std::chrono::seconds member.
- Remove the requirement to explicitly specify a mempool expiry for
LimitMempoolSize(...), just use the newly-introduced member.
- Remove all now-unnecessary instances of:
std::chrono::hours{gArgs.GetIntArg("-mempoolexpiry", DEFAULT_MEMPOOL_EXPIRY)}
- Store the mempool size limit (-maxmempool) in CTxMemPool as a member.
- Remove the requirement to explicitly specify a mempool size limit for
CTxMemPool::GetMinFee(...) and LimitMempoolSize(...), just use the
stored mempool size limit where possible.
- Remove all now-unnecessary instances of:
gArgs.GetIntArg("-maxmempool", DEFAULT_MAX_MEMPOOL_SIZE_MB) * 1000000
The code change in CChainState::GetCoinsCacheSizeState() is correct
since the coinscache should not repurpose "extra" mempool memory
headroom for itself if the mempool doesn't even exist.
Better to be explicit when it comes to sizes to avoid unintentional
bugs. We use MB and KB all over the place.
-BEGIN VERIFY SCRIPT-
find_regex="DEFAULT_MAX_MEMPOOL_SIZE" \
&& git grep -l -E "$find_regex" \
| xargs sed -i -E "s@$find_regex@\0_MB@g"
-END VERIFY SCRIPT-
ce893c0497 doc: Update developer notes (Anthony Towns)
d2852917ee sync.h: Imply negative assertions when calling LOCK (Anthony Towns)
bba87c0553 scripted-diff: Convert global Mutexes to GlobalMutexes (Anthony Towns)
a559509a0b sync.h: Add GlobalMutex type (Anthony Towns)
be6aa72f9f qt/clientmodel: thread safety annotation for m_cached_tip_mutex (Anthony Towns)
f24bd45b37 net_processing: thread safety annotation for m_tx_relay_mutex (Anthony Towns)
Pull request description:
This changes `LOCK(mutex)` for non-global, non-recursive mutexes to be annotated with the negative capability for the mutex it refers to, to prevent . clang applies negative capabilities recursively, so this helps avoid forgetting to annotate functions.
This can't reasonably be used for globals, because clang would require every function to be annotated with `EXCLUSIVE_LOCKS_REQUIRED(!g_mutex)` for each global mutex; so this introduces a trivial `GlobalMutex` subclass of `Mutex`, and reduces the annotations for both `GlobalMutex` to `LOCKS_EXCLUDED` which only catches trivial errors (eg (`LOCK(x); LOCK(x);`).
ACKs for top commit:
MarcoFalke:
review ACK ce893c0497🐦
hebasto:
ACK ce893c0497
Tree-SHA512: 5c35e8c7677ce3d994a7e3774f4344adad496223a51b3a1d1d3b5f20684b2e1d5cff688eb3fbc8d33e1b9940dfa76e515f9434e21de6f3ce3c935e29a319f529
fa4068b4e2 Move minRelayTxFee to policy/settings (MacroFake)
Pull request description:
Seems a bit confusing to put policy stuff into validation, so fix that.
Also fix includes via `iwyu`.
ACKs for top commit:
ariard:
ACK fa4068b, the includes move compiles well locally.
ryanofsky:
Code review ACK fa4068b4e2. Make sense to move the global variable to policy/settings and the default constant to policy/policy. Ariard points out other constants that could be moved, which seems fine, but it seems like moving the global variable to be with other related global variables is more significant.
Tree-SHA512: adf9619002610d1877f3aef0a9e6115fc4c2ad64135a3e5100824c650b560c47f47ac28894c6214a50a7888355252a9f6f7cec98c23a771a1964160ef1ca77de
This check isn't any longer just about bad pay-to-script-hash inputs; it
also excludes any kind of nonstandard input, unknown witness versions,
coinbases, etc.
This is the "fruit of our labor" for this patchset.
ChainstateManager::PopulateAndValidateSnapshot can now directly call
ComputeUTXOStats(...).
Our consensus engine is now fully decoupled from all indices.
See the src/Makefile.am for some satisfying removals.
In previous commits in this patchset, we removed all in-param members of
CCoinsStats. Now that that's done, we can modify GetUTXOStats to return
an optional CCoinsStats instead of a status bool. Callers are modified
accordingly.
In rpc/blockchain.cpp, we discover that GetUTXOStats' status bool when
getting UTXO stats for pprev was not checked for error. We fix this as
well.
Currently, CCoinsStats is a struct with both in-params and out-params
where the hash_type and index_requested members are the only in-params.
This change removes CCoinsStats' hash_type in-param member and adds it
to the relevant functions instead.
[META] In subsequent commits, all of CCoinsStats' members which serve as
in-params will be moved out so as to make CCoinsStats a pure
out-param struct.
This decouples validation.cpp from netaddress.cpp (transitively,
timedata.cpp, and asmap.cpp).
This is important for libbitcoinkernel as:
- There is no reason for the consensus engine to be coupled with
netaddress, timedata, and asmap
- Users of libbitcoinkernel can now easily supply their own
std::function that provides the adjusted time.
See the src/Makefile.am changes for some satisfying removals.
bb5c24b120 validation: move g_versionbitscache into ChainstateManager (Anthony Towns)
eca22c726a test/versionbits: make versionbitscache a parameter (Anthony Towns)
d603f1d8a7 deploymentstatus: make versionbitscache a parameter (Anthony Towns)
78adef1753 refactor: use chainman instead of chainParams for DeploymentActive* (Anthony Towns)
deffe0df6c deploymentstatus: allow chainman in place of consensusParams (Anthony Towns)
eaa2e3f25c validation: move UpdateUncommittedBlockStructures and GenerateCoinbaseCommitment into ChainstateManager (Anthony Towns)
5c67e84d37 validation: replace ::Params() calls with chainstate/chainman member (Anthony Towns)
38860f93b6 validation: remove redundant CChainParams params from ChainstateManager methods (Anthony Towns)
69675ea4e7 validation: add CChainParams to ChainstateManager (Anthony Towns)
Pull request description:
Gives `ChainstateManager` a reference to the `CChainParams` its working on, and simplifies some of the functions that would otherwise take that as a parameter. Removes the `g_versionbitscache` global by moving it into `ChainstateManager`.
ACKs for top commit:
dongcarl:
reACK bb5c24b120
MarcoFalke:
review ACK bb5c24b120📙
Tree-SHA512: 3fa74905e5df561e3e74bb0b8fce6085c5311e6633e7d74c0fb0c82a907f5bbb1fd4ebc5d11d4f0b1c019bb51eabb9f6e4bcc4652a696d36a5878c807b85f121
This fixes a blk file size calculation made during reindex that results in increased blk file malformity.
The fix is to avoid double counting the size of the serialization header during reindex.
This adds a unit test to reproduce the bug before the fix and to ensure that it does not recur.
These changes include a log message change also so as to not be as alarming. This is a common and recoverable
data corruption. These messages can now be filtered by the debug log reindex category.
308dd2e93e Sanity assert GetAncestor() != nullptr where appropriate (Adam Jonas)
Pull request description:
Re-opening #17232. I have rebased the PR and addressed jonatack's nit suggestions.
Add sanity asserts for return value of `CBlockIndex::GetAncestor()` where appropriate.
In validation.cpp `CheckSequenceLocks`, check the return value of `tip->GetAncestor(maxInputHeight)` stored into `lp->maxInputBlock`. If it ever returns `nullptr` because the ancestor isn't found, it's going to be a bad bug to keep going, since a `LockPoints` object with the `maxInputBlock` member set to `nullptr` signifies no relative lock time.
In the other places, the added asserts would prevent accidental dereferencing of a null pointer which is undefined behavior.
Co-Authored-By: Adam Jonas <jonas@chaincode.com>
Co-Authored-By: danra <danra@users.noreply.github.com>
ACKs for top commit:
jonatack:
ACK 308dd2e93e
Tree-SHA512: 5bfdaab1499607ae2c3cd3e2e9e8c37850bfd0e327e680f4e36c81f9c6d98a543af78ecfac1ab0e06325d264412615a04d52005875780c7db2a4d81bd2d2259a
Add sanity asserts for return value of `CBlockIndex::GetAncestor()` where appropriate.
In validation.cpp `CheckSequenceLocks`, check the return value of `tip->GetAncestor(maxInputHeight)` stored into `lp->maxInputBlock`. If it ever returns `nullptr` because the ancestor isn't found, it's going to be a bad bug to keep going, since a `LockPoints` object with the `maxInputBlock` member set to `nullptr` signifies no relative lock time.
In the other places, the added asserts would prevent accidental dereferencing of a null pointer which is undefined behavior.
Co-Authored-By: Aurèle Oulès <aurele@oules.com>
Co-Authored-By: danra <danra@users.noreply.github.com>
f64aa9c411 Disallow more unsafe string->path conversions allowed by path append operators (Ryan Ofsky)
Pull request description:
Add more `fs::path` `operator/` and `operator+` overloads to prevent unsafe string->path conversions on Windows that would cause strings to be decoded according to the current Windows locale & code page instead of the correct string encoding.
Update application code to deal with loss of implicit string->path conversions by calling `fs::u8path` or `fs::PathFromString` explicitly, or by just changing variable types from `std::string` to `fs::path` to avoid conversions altogether, or make them happen earlier.
In all cases, there's no change in behavior either (1) because strings only contained ASCII characters and would be decoded the same regardless of what encoding was used, or (2) because of the 1:1 mapping between paths and strings using the `PathToString` and `PathFromString` functions.
Motivation for this PR was just that I was experimenting with #24469 and noticed that operations like `fs::path / std::string` were allowed, and I thought it would be better not to allow them.
ACKs for top commit:
hebasto:
ACK f64aa9c411
Tree-SHA512: 944cce49ed51537ee7a35ea4ea7f5feaf0c8fff2fa67ee81ec5adebfd3dcbaf41b73eb35e49973d5f852620367f13506fd12a7a9b5ae3a7a0007414d5c9df50f
fa10c9f5a1 Crash debug builds on PCKG_MEMPOOL_ERROR (MacroFake)
Pull request description:
Would be nice to allow fuzz targets to meaningfully cover this code
ACKs for top commit:
glozow:
utACK fa10c9f5a1
vincenzopalazzo:
ACK fa10c9f5a1
Tree-SHA512: 68efacedbf72f67cf3dc0bb9927a698492cdc1b08df91ef6af863ad8828b78058a64e52d64d244a5b2966cb9e63797b2647d1bb222677bf83b26fca6e4b1dbf0
In previous commits in this patchset, we've made sure that every
Unload/UnloadBlockIndex member function resets its own members, and does
not reach out to globals.
This means that their corresponding classes' default destructors can now
replace them, and do an even more thorough job without the need to be
updated for every new member variable.
Therefore, we can remove them, and also remove UnloadBlockIndex since
that's not used anymore.
Unfortunately, chainstatemanager_loadblockindex relies on
CChainState::UnloadBlockIndex, so that needs to stay for now.
This change also introduces an aditional buffer of 10 blocks (PRUNE_LOCK_BUFFER) that will not be pruned before the best block.
Co-authored-by: Luke Dashjr <luke-jr+git@utopios.org>
Add more fs::path operator/ and operator+ overloads to prevent unsafe
string->path conversions on Windows that would cause strings to be
decoded according to the current Windows locale & code page instead of
the correct string encoding.
Update application code to deal with loss of implicit string->path
conversions by calling fs::u8path or fs::PathFromString explicitly, or
by just changing variable types from std::string to fs::path to avoid
conversions altoghther, or make them happen earlier.
In all cases, there's no change in behavior either (1) because strings
only contained ASCII characters and would be decoded the same regardless
of what encoding was used, or (2) because of the 1:1 mapping between
paths and strings using the PathToString and PathFromString functions.
Co-authored-by: Hennadii Stepanov <32963518+hebasto@users.noreply.github.com>
-----
Code Reviewer Notes
Call graph of relevant functions:
UnloadBlockIndex() <-- Moved from
calls ChainstateManager::Unload()
which calls BlockManager::Unload() <-- Moved to
So calling UnloadBlockIndex() would still run this moved code. The code
will also now run when ~BlockManager gets called, which makes sense.
-----
Code Reviewer Notes
Call graph of relevant functions:
UnloadBlockIndex() <-- Moved from
calls ChainstateManager::Unload() <-- Moved to
Safe because ChainstateManager::Unload() is called only by
UnloadBlockIndex() and no other callers.
When using checklevel=4, block verification fails because of duplicate coinbase transactions
involving blocks 91812 and 91722. There was already a check in place for ConnectBlock to
ignore the problematic blocks, but DisconnectBlock did not contain a similar check.
This change ignores the blocks where these inconsistencies surface so
that block verification will succeed at checklevel=4.
Now BlockManager::LoadBlockIndex() will ACTUALLY only load BlockMan
members.
[META] In a later commit, pindexBestHeader will be moved to ChainMan as
a member
-----
Code Reviewer Notes
Call graph of relevant functions:
ChainstateManager::LoadBlockIndex() <-- Moved to
calls BlockManager::LoadBlockIndexDB()
which calls BlockManager::LoadBlockIndex() <-- Moved from
There is only one call to each of inner functions, meaning that no
behavior is changing.
Package validation policy only differs from individual policy in its
evaluation of feerate. Minimize DoS surface; don't validate all over
again if we know the result will be the same.
This avoids "parents pay for children" and "siblings pay for siblings"
behavior, since package feerate is calculated with totals and is
topology-unaware.
It also ensures that package validation never causes us to reject a
transaction that we would have otherwise accepted in single-tx
validation.
cccc1e70b8 Enforce Taproot script flags whenever WITNESS is set (MarcoFalke)
fa42299411 Remove nullptr check in GetBlockScriptFlags (MarcoFalke)
faadc606c7 refactor: Pass const reference instead of pointer to GetBlockScriptFlags (MarcoFalke)
Pull request description:
Now that Taproot is active, it makes sense to enforce its rules on all blocks, even historic ones, regardless of the deployment status.
### Benefits:
(With "script flags" I mean "taproot script verification flags".)
* Script flags are known ahead for all blocks (even blocks not yet created) and do not change. This may benefit static analysis, code review, and development of new script features that build on Taproot.
* Any future bugs introduced in the deployment code won't have any effect on the script flags, as they are independent of deployment.
* Enforcing the taproot rules regardless of the deployment status makes testing easier because invalid blocks after activation are also invalid before activation. So there is no need to differentiate the two cases.
* It gives belt-and-suspenders protection against a practically expensive and theoretically impossible IBD reorg attack where the node is eclipsed. While `nMinimumChainWork` already protects against this, the cost for a few months worth of POW might be lowered until a major version release of Bitcoin Core reaches EOL. The needed work for the attack is the difference between `nMinimumChainWork` and the work at block 709632.
For reference, previously the same was done for P2SH and WITNESS in commit 0a8b7b4b33.
### Implementation:
I found one block which fails verification with the flags applied, so I added a `TaprootException`, similar to the `BIP16Exception`.
For reference, the debug log:
```
ERROR: ConnectBlock(): CheckInputScripts on b10c007c60e14f9d087e0291d4d0c7869697c6681d979c6639dbd960792b4d41 failed with non-mandatory-script-verify-flag (Witness program was passed an empty witness)
BlockChecked: block hash=0000000000000000000f14c35b2d841e986ab5441de8c585d5ffe55ea1e395ad state=non-mandatory-script-verify-flag (Witness program was passed an empty witness)
InvalidChainFound: invalid block=0000000000000000000f14c35b2d841e986ab5441de8c585d5ffe55ea1e395ad height=692261 log2_work=92.988459 date=2021-07-23T08:24:20Z
InvalidChainFound: current best=0000000000000000000067b17a4c0ffd77c29941b15ad356ca8f980af137a25d height=692260 log2_work=92.988450 date=2021-07-23T07:47:31Z
ERROR: ConnectTip: ConnectBlock 0000000000000000000f14c35b2d841e986ab5441de8c585d5ffe55ea1e395ad failed, non-mandatory-script-verify-flag (Witness program was passed an empty witness)
```
Hint for testing, make sure to set `-noassumevalid`.
### Considerations
Obviously this change can lead to consensus splits on the network in light of massive reorgs. Currently the last block before Taproot activation, that is the last block without the Taproot script flags set, is only buried by a few days of POW. However, when and if this patch is included in the next major release, it will be buried by a few months of POW. BIP90 considerations apply when looking at reorgs this large.
ACKs for top commit:
Sjors:
tACK cccc1e70b8
achow101:
ACK cccc1e70b8
laanwj:
Code review ACK cccc1e70b8
ajtowns:
ACK cccc1e70b8 ; code review; wrote a "getblockscriptflags" rpc to quickly check that blocks just had bit 17 (taproot) added; review of earlier revisions had established non-exception blocks do validate with taproot rules enabled.
jamesob:
ACK cccc1e70b8 ([`jamesob/ackr/23536.1.MarcoFalke.enforce_taproot_script_f`](https://github.com/jamesob/bitcoin/tree/ackr/23536.1.MarcoFalke.enforce_taproot_script_f))
Tree-SHA512: 00044de68939caef6420ffd588c1291c041a8b397c80a3df1e3e3487fbeae1821d23975c51c95e44e774558db76f943b00b4e27cbd0213f64a9253116dc6edde
f865cf8ded Add and use BlockManager::GetAllBlockIndices (Carl Dong)
28ba0313ea Add and use CBlockIndexHeightOnlyComparator (Carl Dong)
12eb05df63 move-only: Move CBlockIndexWorkComparator to blockstorage (Carl Dong)
c600ee3816 Only load BlockMan in BlockMan member functions (Carl Dong)
42e56d9b18 style-only: No need for std::pair for vSortedByHeight (Carl Dong)
3bbb6fea05 style-only: Various blockstorage.cpp cleanups (Carl Dong)
5be9ee3c54 refactor: more const annotations for uses of CBlockIndex* (Anthony Towns)
Pull request description:
The only important commit is "Only load BlockMan in BlockMan member functions", everything else is all just small style changes.
Here's the commit message, reproduced:
```
This commit effectively splits the "load block index itself" logic from
"derive Chainstate variables from loaded block index" logic.
This means that BlockManager::LoadBlockIndex{,DB} will only load what's
relevant to the BlockManager.
```
ACKs for top commit:
ajtowns:
ACK f865cf8ded ; code review only
MarcoFalke:
review ACK f865cf8ded 🗂
Tree-SHA512: 7b204d782834e06fd7329d022e2ae860181b4e8105c33bfb928539a4ec24161dc7438a9c4d4ee279dcad77de310c160b997bb8aa18923243d0fd55ccf4ad7c3a
