e829c9afbf refactor: replace sizeof(a)/sizeof(a[0]) by std::size (C++17) (Sebastian Falbesoner)
365539c846 refactor: init vectors via std::{begin,end} to avoid pointer arithmetic (Sebastian Falbesoner)
63d4ee1968 refactor: iterate arrays via C++11 range-based for loops if idx is not needed (Sebastian Falbesoner)
Pull request description:
This refactoring PR picks up the idea of #19626 and replaces all occurences of `sizeof(x)/sizeof(x[0])` (or `sizeof(x)/sizeof(*x)`, respectively) with the now-available C++17 [`std::size`](https://en.cppreference.com/w/cpp/iterator/size) (as [suggested by sipa](https://github.com/bitcoin/bitcoin/pull/19626#issuecomment-666487228)), making the macro `ARRAYLEN` obsolete.
As preparation for this, two other changes are done to eliminate `sizeof(x)/sizeof(x[0])` usage:
* all places where arrays are iterated via an index are changed to use C++11 range-based for loops If the index' only purpose is to access the array element (as [suggested by MarcoFalke](https://github.com/bitcoin/bitcoin/pull/19626#discussion_r463404541)).
* `std::vector` initializations are done via `std::begin` and `std::end` rather than using pointer arithmetic to calculate the end (also [suggested by MarcoFalke](https://github.com/bitcoin/bitcoin/pull/20429#discussion_r567418821)).
ACKs for top commit:
practicalswift:
cr ACK e829c9afbf: patch looks correct
fanquake:
ACK e829c9afbf
MarcoFalke:
review ACK e829c9afbf 🌩
Tree-SHA512: b01d32c04b9e04d562b7717cae00a651ec9a718645047a90761be6959e0cc2adbd67494e058fe894641076711bb09c3b47a047d0275c736f0b2218e1ce0d193d
53e716ea11 [refactor] improve style for touched code (gzhao408)
174cb5330a [refactor] const ATMPArgs and non-const Workspace (gzhao408)
f82baf0762 [refactor] return MempoolAcceptResult (gzhao408)
9db10a5506 [refactor] clean up logic in testmempoolaccept (gzhao408)
Pull request description:
This is the first 4 commits of #20833, and does refactoring only. It should be relatively simple to review, and offers a few nice things:
- It makes accessing values that don't make sense (e.g. fee) when the tx is invalid an error.
- Returning `MempoolAcceptResult` from ATMP makes the interface cleaner. The caller can get a const instead of passing in a mutable "out" param.
- We don't have to be iterating through a bunch of lists for package validation, we can just return a `std::vector<MempoolAcceptResult>`.
- We don't have to refactor all ATMP call sites again if/when we want to return more stuff from it.
ACKs for top commit:
MarcoFalke:
ACK 53e716ea11💿
jnewbery:
Code review ACK 53e716ea11
ariard:
Code Review ACK 53e716e, I did tweak a bit the touched paths to see if we had good test coverage. Didn't find holes.
Tree-SHA512: fa6ec324a08ad9e6e55948615cda324cba176255708bf0a0a0f37cedb7a75311aa334ac6f223be7d8df3c7379502b1081102b9589f9a9afa1713ad3d9ab3c24f
This creates a cleaner interface with ATMP, allows us to make results const,
and makes accessing values that don't make sense (e.g. fee when tx is
invalid) an error.
fa29272459 Remove redundant MakeUCharSpan wrappers (MarcoFalke)
faf4aa2f47 Remove CDataStream::Init in favor of C++11 member initialization (MarcoFalke)
fada14b948 Treat CDataStream bytes as uint8_t (MarcoFalke)
fa8bdb048e refactor: Drop CDataStream constructors in favor of one taking a Span of bytes (MarcoFalke)
faa96f841f Remove unused CDataStream methods (MarcoFalke)
Pull request description:
Using `uint8_t` for raw bytes has a style benefit:
* The signedness is clear from reading the code, as it does not depend on the architecture
Other clean-ups in this pull include:
* Remove unused methods
* Constructor is simplified with `Span`
* Remove `Init()` member in favor of C++11 member initialization
ACKs for top commit:
laanwj:
code review ACK fa29272459
theStack:
ACK fa29272459🍾
Tree-SHA512: 931ee28bd99843d7e894b48e90e1187ffb0278677c267044b3c0c255069d9bbd9298ab2e539b1002a30b543d240450eaec718ef4ee95a7fd4be0a295e926343f
5d4597666d Rewrite OutputGroups to be clearer and to use scriptPubKeys (Andrew Chow)
f6b3052739 Explicitly filter out partial groups when we don't want them (Andrew Chow)
416d74fb16 Move OutputGroup positive only filtering into Insert (Andrew Chow)
d895e98b59 Move EligibleForSpending into GroupOutputs (Andrew Chow)
99b399aba5 Move fee setting of OutputGroup to Insert (Andrew Chow)
6148a8acda Move GroupOutputs into SelectCoinsMinConf (Andrew Chow)
2acad03657 Remove OutputGroup non-default constructors (Andrew Chow)
Pull request description:
Even after #17458, we still deal with setting fees of an `OutputGroup` and filtering the `OutputGroup` outside of the struct. We currently make all of the `OutputGroup`s in `SelectCoins` and then copy and modify them within each `SelectCoinsMinConf` scenario. This PR changes this to constructing the `OutputGroup`s within the `SelectCoinsMinConf` so that the scenario can be taken into account during the group construction. Furthermore, setting of fees and filtering for effective value is moved into `OutputGroup::Insert` itself so that we don't add undesirable outputs to an `OutputGroup` rather than deleting them afterwards.
To facilitate fee calculation and effective value filtering during `OutputGroup::Insert`, `OutputGroup` now takes the feerates in its constructor and computes the fees and effective value for each output during `Insert`.
While removing `OutputGroup`s in accordance with the `CoinEligibilityFilter` still requires creating the `OutputGroup`s first, we can do that within the function that makes them - `GroupOutput`s.
ACKs for top commit:
Xekyo:
Code review ACK: 5d4597666d
fjahr:
Code review ACK 5d4597666d
meshcollider:
Light utACK 5d4597666d
Tree-SHA512: 35965b6d49a87f4ebb366ec4f00aafaaf78e9282481ae2c9682b515a3a9f2cbcd3cd6e202fee29489d48fe7f3a7cede4270796f5e72bbaff76da647138fb3059
bb6fcc75d1 refactor: Drop boost::thread stuff in CCheckQueue (Hennadii Stepanov)
6784ac471b bench: Use CCheckQueue local thread pool (Hennadii Stepanov)
dba30695fc test: Use CCheckQueue local thread pool (Hennadii Stepanov)
01511776ac Add local thread pool to CCheckQueue (Hennadii Stepanov)
0ef938685b refactor: Use member initializers in CCheckQueue (Hennadii Stepanov)
Pull request description:
This PR:
- gets rid of `boost::thread_group` in the `CCheckQueue` class
- allows thread safety annotation usage in the `CCheckQueue` class
- is alternative to #14464 (https://github.com/bitcoin/bitcoin/pull/18710#issuecomment-616618525, https://github.com/bitcoin/bitcoin/pull/18710#issuecomment-617291612)
Also, with this PR (I hope) it could be easier to resurrect a bunch of brilliant ideas from #9938.
Related: #17307
ACKs for top commit:
laanwj:
Code review ACK bb6fcc75d1
LarryRuane:
ACK bb6fcc75d1
jonatack:
Code review ACK bb6fcc75d1 and verified rebase to master builds cleanly with unit/functional tests green
Tree-SHA512: fddeb720d5a391b48bb4c6fa58ed34ccc3f57862fdb8e641745c021841c8340e35c5126338271446cbd98f40bd5484f27926aa6c3e76fa478ba1efafe72e73c1
Currently it was not possible to run just the BlockToJsonVerboes benchmarsk because it did not set up everything it needed, running `bench_bitcoin -filter=BlockToJsonVerbose` caused this assert to fail:
```
bench_bitcoin: chainparams.cpp:506: const CChainParams& Params(): Assertion `globalChainParams' failed.
```
Initializing TestingSetup fixes this.
9815332d51 test: Change MuHash Python implementation to match cpp version again (Fabian Jahr)
01297fb3ca fuzz: Add MuHash consistency fuzz test (Fabian Jahr)
b111410914 test: Add MuHash3072 fuzz test (Fabian Jahr)
c122527385 bench: Add Muhash benchmarks (Fabian Jahr)
7b1242229d test: Add MuHash3072 unit tests (Fabian Jahr)
adc708c98d crypto: Add MuHash3072 implementation (Fabian Jahr)
0b4d290bf5 crypto: Add Num3072 implementation (Fabian Jahr)
589f958662 build: Check for 128 bit integer support (Fabian Jahr)
Pull request description:
This is the first split of #18000 which implements the Muhash algorithm and uses it to calculate the UTXO set hash in `gettxoutsetinfo`.
ACKs for top commit:
laanwj:
Code review ACK 9815332d51
Tree-SHA512: 4bc090738f0e3d80b74bdd8122e24a8ce80121120fd37c7e4335a73e7ba4fcd7643f2a2d559e2eebf54b8e3a3bd5f12cfb27ba61ded135fda210a07a233eae45
5021810650 Make CanFlushToDisk a const member function (practicalswift)
281cf99554 Do not run functions with necessary side-effects in assert() (practicalswift)
Pull request description:
Do not run functions with necessary side-effects in `assert()`.
ACKs for top commit:
laanwj:
Code review ACK 5021810650
sipa:
utACK 5021810650
theStack:
Code Review ACK 5021810650🟢
Tree-SHA512: 38b7faccc2f16a499f9b7b1b962b49eb58580b2a2bbf63ea49dcc418a5ecc8f21a0972fa953f66db9509c7239af67cfa2f9266423fd220963d091034d7332b96
This removes a source of complexity and indirection that makes it harder to
understand path checking code. Path checks will be simplified in upcoming
commits.
There is no change in behavior in this commit other than a slightly more
descriptive error message in `loadwallet` if the default "" wallet can't be
found. (The error message is improved more in upcoming commit "wallet: Remove
path checking code from loadwallet RPC".)
CAddrMan.GetAddr() would previously limit the number and percentage of
addresses returned (to ADDRMAN_GETADDR_MAX (1000) and
ADDRMAN_GETADDR_MAX_PCT (23) respectively). Instead, make it the callers
responsibility to specify the maximum addresses and percentage they want
returned.
For net_processing, the maximums are MAX_ADDR_TO_SEND (1000) and
MAX_PCT_ADDR_TO_SEND (23). For rpc/net, the maximum is specified by the
client.
78c312c983 Replace current benchmarking framework with nanobench (Martin Ankerl)
Pull request description:
Replace current benchmarking framework with nanobench
This replaces the current benchmarking framework with nanobench [1], an
MIT licensed single-header benchmarking library, of which I am the
autor. This has in my opinion several advantages, especially on Linux:
* fast: Running all benchmarks takes ~6 seconds instead of 4m13s on
an Intel i7-8700 CPU @ 3.20GHz.
* accurate: I ran e.g. the benchmark for SipHash_32b 10 times and
calculate standard deviation / mean = coefficient of variation:
* 0.57% CV for old benchmarking framework
* 0.20% CV for nanobench
So the benchmark results with nanobench seem to vary less than with
the old framework.
* It automatically determines runtime based on clock precision, no need
to specify number of evaluations.
* measure instructions, cycles, branches, instructions per cycle,
branch misses (only Linux, when performance counters are available)
* output in markdown table format.
* Warn about unstable environment (frequency scaling, turbo, ...)
* For better profiling, it is possible to set the environment variable
NANOBENCH_ENDLESS to force endless running of a particular benchmark
without the need to recompile. This makes it to e.g. run "perf top"
and look at hotspots.
Here is an example copy & pasted from the terminal output:
| ns/byte | byte/s | err% | ins/byte | cyc/byte | IPC | bra/byte | miss% | total | benchmark
|--------------------:|--------------------:|--------:|----------------:|----------------:|-------:|---------------:|--------:|----------:|:----------
| 2.52 | 396,529,415.94 | 0.6% | 25.42 | 8.02 | 3.169 | 0.06 | 0.0% | 0.03 | `bench/crypto_hash.cpp RIPEMD160`
| 1.87 | 535,161,444.83 | 0.3% | 21.36 | 5.95 | 3.589 | 0.06 | 0.0% | 0.02 | `bench/crypto_hash.cpp SHA1`
| 3.22 | 310,344,174.79 | 1.1% | 36.80 | 10.22 | 3.601 | 0.09 | 0.0% | 0.04 | `bench/crypto_hash.cpp SHA256`
| 2.01 | 496,375,796.23 | 0.0% | 18.72 | 6.43 | 2.911 | 0.01 | 1.0% | 0.00 | `bench/crypto_hash.cpp SHA256D64_1024`
| 7.23 | 138,263,519.35 | 0.1% | 82.66 | 23.11 | 3.577 | 1.63 | 0.1% | 0.00 | `bench/crypto_hash.cpp SHA256_32b`
| 3.04 | 328,780,166.40 | 0.3% | 35.82 | 9.69 | 3.696 | 0.03 | 0.0% | 0.03 | `bench/crypto_hash.cpp SHA512`
[1] https://github.com/martinus/nanobench
ACKs for top commit:
laanwj:
ACK 78c312c983
Tree-SHA512: 9e18770b18b6f95a7d0105a4a5497d31cf4eb5efe6574f4482f6f1b4c88d7e0946b9a4a1e9e8e6ecbf41a3f2d7571240677dcb45af29a6f0584e89b25f32e49e
addf18da95 Call SHA256AutoDetect in benchmark setup (Pieter Wuille)
Pull request description:
It seems `SHA256AutoDetect()` was not being called in benchmarks, making the numbers only reflect the naive implementation. Fix this by calling it in bench_bitcoin's setup.
ACKs for top commit:
fjahr:
tested ACK addf18da95
pstratem:
ACK addf18da95
laanwj:
ACK addf18da95
Tree-SHA512: 3ba4b068145942df1429bf5913e3f685511e6ebeae2c1a3f9b8ac0144f6db1c7df456f88f480a2129f3e1602e3bf6a39530bb96e2c74c03ddb19324cec6799c7
This replaces the current benchmarking framework with nanobench [1], an
MIT licensed single-header benchmarking library, of which I am the
autor. This has in my opinion several advantages, especially on Linux:
* fast: Running all benchmarks takes ~6 seconds instead of 4m13s on
an Intel i7-8700 CPU @ 3.20GHz.
* accurate: I ran e.g. the benchmark for SipHash_32b 10 times and
calculate standard deviation / mean = coefficient of variation:
* 0.57% CV for old benchmarking framework
* 0.20% CV for nanobench
So the benchmark results with nanobench seem to vary less than with
the old framework.
* It automatically determines runtime based on clock precision, no need
to specify number of evaluations.
* measure instructions, cycles, branches, instructions per cycle,
branch misses (only Linux, when performance counters are available)
* output in markdown table format.
* Warn about unstable environment (frequency scaling, turbo, ...)
* For better profiling, it is possible to set the environment variable
NANOBENCH_ENDLESS to force endless running of a particular benchmark
without the need to recompile. This makes it to e.g. run "perf top"
and look at hotspots.
Here is an example copy & pasted from the terminal output:
| ns/byte | byte/s | err% | ins/byte | cyc/byte | IPC | bra/byte | miss% | total | benchmark
|--------------------:|--------------------:|--------:|----------------:|----------------:|-------:|---------------:|--------:|----------:|:----------
| 2.52 | 396,529,415.94 | 0.6% | 25.42 | 8.02 | 3.169 | 0.06 | 0.0% | 0.03 | `bench/crypto_hash.cpp RIPEMD160`
| 1.87 | 535,161,444.83 | 0.3% | 21.36 | 5.95 | 3.589 | 0.06 | 0.0% | 0.02 | `bench/crypto_hash.cpp SHA1`
| 3.22 | 310,344,174.79 | 1.1% | 36.80 | 10.22 | 3.601 | 0.09 | 0.0% | 0.04 | `bench/crypto_hash.cpp SHA256`
| 2.01 | 496,375,796.23 | 0.0% | 18.72 | 6.43 | 2.911 | 0.01 | 1.0% | 0.00 | `bench/crypto_hash.cpp SHA256D64_1024`
| 7.23 | 138,263,519.35 | 0.1% | 82.66 | 23.11 | 3.577 | 1.63 | 0.1% | 0.00 | `bench/crypto_hash.cpp SHA256_32b`
| 3.04 | 328,780,166.40 | 0.3% | 35.82 | 9.69 | 3.696 | 0.03 | 0.0% | 0.03 | `bench/crypto_hash.cpp SHA512`
[1] https://github.com/martinus/nanobench
* Adds support for asymptotes
This adds support to calculate asymptotic complexity of a benchmark.
This is similar to #17375, but currently only one asymptote is
supported, and I have added support in the benchmark `ComplexMemPool`
as an example.
Usage is e.g. like this:
```
./bench_bitcoin -filter=ComplexMemPool -asymptote=25,50,100,200,400,600,800
```
This runs the benchmark `ComplexMemPool` several times but with
different complexityN settings. The benchmark can extract that number
and use it accordingly. Here, it's used for `childTxs`. The output is
this:
| complexityN | ns/op | op/s | err% | ins/op | cyc/op | IPC | total | benchmark
|------------:|--------------------:|--------------------:|--------:|----------------:|----------------:|-------:|----------:|:----------
| 25 | 1,064,241.00 | 939.64 | 1.4% | 3,960,279.00 | 2,829,708.00 | 1.400 | 0.01 | `ComplexMemPool`
| 50 | 1,579,530.00 | 633.10 | 1.0% | 6,231,810.00 | 4,412,674.00 | 1.412 | 0.02 | `ComplexMemPool`
| 100 | 4,022,774.00 | 248.58 | 0.6% | 16,544,406.00 | 11,889,535.00 | 1.392 | 0.04 | `ComplexMemPool`
| 200 | 15,390,986.00 | 64.97 | 0.2% | 63,904,254.00 | 47,731,705.00 | 1.339 | 0.17 | `ComplexMemPool`
| 400 | 69,394,711.00 | 14.41 | 0.1% | 272,602,461.00 | 219,014,691.00 | 1.245 | 0.76 | `ComplexMemPool`
| 600 | 168,977,165.00 | 5.92 | 0.1% | 639,108,082.00 | 535,316,887.00 | 1.194 | 1.86 | `ComplexMemPool`
| 800 | 310,109,077.00 | 3.22 | 0.1% |1,149,134,246.00 | 984,620,812.00 | 1.167 | 3.41 | `ComplexMemPool`
| coefficient | err% | complexity
|--------------:|-------:|------------
| 4.78486e-07 | 4.5% | O(n^2)
| 6.38557e-10 | 21.7% | O(n^3)
| 3.42338e-05 | 38.0% | O(n log n)
| 0.000313914 | 46.9% | O(n)
| 0.0129823 | 114.4% | O(log n)
| 0.0815055 | 133.8% | O(1)
The best fitting curve is O(n^2), so the algorithm seems to scale
quadratic with `childTxs` in the range 25 to 800.
152e8baf08 Use salted hasher instead of nonce in sigcache (Jeremy Rubin)
5495fa5850 Add Hash Padding Microbenchmarks (Jeremy Rubin)
Pull request description:
This PR replaces nonces in two places with pre-salted hashers.
The nonce is chosen to be 64 bytes long so that it forces the SHA256 hasher to process the chunk. This leaves the next 64 (or 56 depending if final chunk) open for data. In the case of the script execution cache, this does not make a big performance improvement because the nonce was already properly padded to fit into one buffer, but does make the code a little simpler. In the case of the sig cache, this should reduce the hashing overhead slightly because we are less likely to need an additional processing step.
I haven't benchmarked this, but back of the envelope it should reduce the hashing by one buffer for all combinations except compressed public keys with compact signatures.
ACKs for top commit:
ryanofsky:
Code review ACK 152e8baf08. No code changes, just rebase since last review and expanded commit message
Tree-SHA512: b133e902fd595cfe3b54ad8814b823f4d132cb2c358c89158842ae27daee56ab5f70cde2585078deb46f77a6e7b35b4cc6bba47b65302b7befc2cff254bad93d
f9ee0f37c2 Add comments to CustomUintFormatter (Pieter Wuille)
4eb5643e35 Convert everything except wallet/qt to new serialization (Pieter Wuille)
2b1f85e8c5 Convert blockencodings_tests to new serialization (Pieter Wuille)
73747afbbe Convert merkleblock to new serialization (Pieter Wuille)
d06fedd1bc Add SER_READ and SER_WRITE for read/write-dependent statements (Russell Yanofsky)
6f9a1e5ad0 Extend CustomUintFormatter to support enums (Russell Yanofsky)
769ee5fa00 Merge BigEndian functionality into CustomUintFormatter (Pieter Wuille)
Pull request description:
The next step of changes from #10785.
This:
* Adds support for enum serialization to `CustomUintFormatter`, used in `CAddress` for service flags.
* Merges `BigEndian` into `CustomUintFormatter`, used in `CNetAddr` for port numbers.
* Converts everything (except wallet and gui) to use the new serialization framework.
ACKs for top commit:
MarcoFalke:
re-ACK f9ee0f37c2, only change is new documentation commit for CustomUintFormatter 📂
ryanofsky:
Code review ACK f9ee0f37c2. Just new commit adding comment since last review
jonatack:
Code review re-ACK f9ee0f37c2 only change since last review is an additional commit adding Doxygen documentation for `CustomUintFormatter`.
Tree-SHA512: e7a0a36afae592d5a4ff8c81ae04d858ac409388e361f2bc197d9a78abca45134218497ab2dfd6d031e0cce0ca586cf857077b7c6ce17fccf67e2d367c1b6cd4
fabe44e815 bench: Start nodes with -nodebuglogfile (MarcoFalke)
Pull request description:
For benchmarking we don't want to depend on the speed of the disk or the amount of debug logging
ACKs for top commit:
fanquake:
ACK fabe44e815 - This makes some of these benchmarks significantly faster to run. MempoolEviction total runtime is down from ~46s to 11s on my machine:
Tree-SHA512: d99700901650325896b9115d20b84a27042152f46266f595bf7ea1414528c0b346f4e707a12ee8b8ba99c35cf155e645e67971c1b2a679c4e609c400ff8b08ae
