99afb9d15a refactor: init, simplify index shutdown code (furszy)
0faafb57f8 index: decrease ThreadSync cs_main contention (furszy)
f1469eb454 index: cache last block filter header (furszy)
a6756ecdb2 index: blockfilter, decouple header lookup into its own function (furszy)
331f044e3b index: blockfilter, decouple Write into its own function (furszy)
bcbd7eb8d4 bench: basic block filter index initial sync (furszy)
Pull request description:
Work decoupled from #26966 per request.
The aim is to remove an unnecessary disk read operation that currently takes place with every new arriving block (or scanned block during background sync). Instead of reading the last filter header from disk merely to access its hash for constructing the next filter, this work caches it, occupying just 32 more bytes in memory.
Also, reduces `cs_main` lock contention during the index initial sync process. And, simplifies the indexes initialization and shutdown procedure.
Testing Note:
To compare the changes, added a pretty basic benchmark in the second commit. Alternatively, could also test the changes by timing the block filter sync from scratch on any network; start the node with `-blockfilterindex` and monitor the logs until the syncing process finish.
Local Benchmark Results:
*Master (c252a0fc0f):
| ns/op | op/s | err% | total | benchmark
|--------------------:|--------------------:|--------:|----------:|:----------
| 132,042,516.60 | 7.57 | 0.3% | 7.79 | `BlockFilterIndexSync`
*PR (43a212cfdac6c64e82b601c664443d022f191520):
| ns/op | op/s | err% | total | benchmark
|--------------------:|--------------------:|--------:|----------:|:----------
| 126,915,841.60 | 7.88 | 0.6% | 7.51 | `BlockFilterIndexSync`
ACKs for top commit:
Sjors:
re-ACK 99afb9d15a
achow101:
ACK 99afb9d15a
TheCharlatan:
Re-ACK 99afb9d15a
andrewtoth:
ACK 99afb9d15a
Tree-SHA512: 927daadd68f4ee1ca781a89519539b895f5185a76ebaf525fbc246ea8dcf40d44a82def00ac34b188640802844b312270067f1b33e65a2479e06be9169c616de
faecf3a7e6 ci: Bump msan to llvm-18 (MarcoFalke)
Pull request description:
Last one: https://github.com/bitcoin/bitcoin/pull/28476
ACKs for top commit:
fanquake:
ACK faecf3a7e6 - There is now a 18.1.2, but given it doesn't fix the instrumenting in libunwind, we don't need that here. I've tested that both jobs are now working on both arches.
Tree-SHA512: 489c0b343bdc732687131317a570f3efbb18a3f548736d739da90d1a1e784df1dbb208c2da8a2a7740f27f961a841c477487a14c4d59910368f651225f0779b2
2f23987849 test: p2p: check limited peers desirability (depending on best block depth) (Sebastian Falbesoner)
c4a67d396d test: p2p: check disconnect due to lack of desirable service flags (Sebastian Falbesoner)
405ac819af test: p2p: support disconnect waiting for `add_outbound_p2p_connection` (Sebastian Falbesoner)
Pull request description:
This PR adds missing test coverage for disconnecting peers which don't offer the desirable service flags in their VERSION message:
5f3a0574c4/src/net_processing.cpp (L3384-L3389)
This check is relevant for the connection types "outbound-full-relay", "block-relay-only" and "addr-fetch" (see `CNode::ExpectServicesFromConn(...)`). Feeler connections always disconnect, which is also tested here.
In lack of finding a proper file where this test would fit in, I created a new one. Happy to take suggestions there.
ACKs for top commit:
davidgumberg:
reACK 2f23987849
itornaza:
tested ACK 2f23987849
fjahr:
re-utACK 2f23987849
cbergqvist:
re ACK 2f23987849
stratospher:
tested ACK 2f23987. 🚀
Tree-SHA512: cf75d9d4379d0f34fa1e13152e6a8d93cd51b9573466ab3a2fec32dc3e1ac49b174bd1063cae558bc736b111c8a6e7058b1b57a496df56255221bf367d29eb5d
6e873df347 serfloat: improve/simplify tests (Pieter Wuille)
b45f1f5658 serfloat: do not test encode(bits)=bits anymore (Pieter Wuille)
Pull request description:
Closes #28941.
Our current tests for serfloat verify two distinct properties:
1. Whether they roundtrip `double`->`uint64_t`->`double` (excluding NaN values) on all systems.
2. Whether on systems with a typical floating point unit that encoding matches the hardware representation, as before v22.0, we would dump the hardware representation directly to disk and we wanted to retain compatibility with that.
#28941 seems to show that the second property doesn't always hold, but just for "subnormal" numbers (below $2^{-1021}$). Since we don't care about encoding these numbers, we could exclude such subnormal numbers from the hardware-identical representation test, but this PR goes further and just drops the second property entirely, as I don't think we care about edge-case compatibility with pre-v22.0 code for fee_estimates.dat (the only place it is used).
ACKs for top commit:
glozow:
ACK 6e873df347
fanquake:
ACK 6e873df347 - It's not as much of a priority, but I think we could still backport this.
Tree-SHA512: e18ceee0753a7ee7e999fdfa10b014dc5bb67b6ef79522a0f8c76b889adcfa785772fc26ed7559bcb5a09a9938e243bb54eedd9549bc59080a2c8090155e2267
fae70ba00d ci: Better tidy errors (MarcoFalke)
Pull request description:
Currently tidy errors are not nice, because the user may have to scroll up to see them in a large block of text. See for example (before) https://github.com/bitcoin/bitcoin/runs/19670551485
Fix that by `tee`ing the output to a file and summarizing the errors in the end again. See for example (after): https://github.com/bitcoin/bitcoin/runs/22647850662
ACKs for top commit:
hebasto:
ACK fae70ba00d, logs with errors look cleaner.
TheCharlatan:
ACK fae70ba00d
Tree-SHA512: dcaea557fed40089409d16ce2cbaa8a9cfbf047f601d5daadfee0823b0eed7badc12d803addc0b7b6bb3f1eaf5c787fccb2488475d32c4efd80835f386f761dd
432a542e27 test: fix intermittent failures with test=addrman (Martin Zumsande)
Pull request description:
The `nKey` of the addrman is generated the first time the node is started with an empty `peers.dat`. Therefore, restarting a node or turning it off and on again won't make a previously non-deterministic addrman deterministic.
This could lead to intermittent failures in `feature_asmap.py` and `rpc_net.py`
Fixes #29634
ACKs for top commit:
kevkevinpal:
ACK [432a542](432a542e27)
stratospher:
Tested ACK 432a542e27.
brunoerg:
crACK 432a542e27
0xB10C:
ACK 432a542e27
Tree-SHA512: a8e284baeb0be2df7284b8a2792cb9edc9e2d5b877a3b29ab7277ffdb75b17efa58a4d42576441eb493cd518e7c5ffdb05597b27e42b5001cf1a80e78bb04c83
626f8e398e fuzz: actually test garbage >64b in p2p transport test (Pieter Wuille)
Pull request description:
This fixes an oversight from #28196: in the `p2p_transport_bidirectional_v2` fuzz test, when the desired garbage length is over 64 bytes, the code would actually use garbage length 0. Fix this.
ACKs for top commit:
instagibbs:
ACK 626f8e398e
brunoerg:
crACK 626f8e398e
Tree-SHA512: f6346367adb10464b6c9d20aef43625531d2a4d8110887ad03214b8c1907b83560f2dd5b5415e2180a40b4cd276d51881b32b60c740471b5c6bb218aa19848d8
38f70ba6ac RPC: Add maxfeerate and maxburnamount args to submitpackage (Greg Sanders)
Pull request description:
Resolves https://github.com/bitcoin/bitcoin/issues/28949
I couldn't manage to do it very cleanly outside of (sub)package evaluation itself, since it would change the current interface very heavily. Instead I threaded through the max fee argument and used that directly via ATMPArgs. From that perspective, this is somewhat a reversion from https://github.com/bitcoin/bitcoin/pull/19339. In a post-cluster mempool world, these checks could be consolidated to right after the given (ancestor) package is linearized/chunked, by just checking the feerate of the top chunk and rejecting the submission entirely if the top chunk is too high.
The implication here is that subpackages can be submitted to the mempool prior to hitting this new fee-based error condition.
ACKs for top commit:
ismaelsadeeq:
Re-ACK 38f70ba6ac👍🏾
glozow:
ACK 38f70ba6ac with some non-blocking nits
murchandamus:
LGTM, code review ACK 38f70ba6ac
Tree-SHA512: 38212aa9de25730944cee58b0806a3d37097e42719af8dd7de91ce86bb5d9770b6f7c37354bf418bd8ba571c52947da1dcdbb968bf429dd1dbdf8715315af18f
64722e4359 ci: Drop `--enable-c++20` option (Hennadii Stepanov)
Pull request description:
This option has ceased to exist since https://github.com/bitcoin/bitcoin/pull/28349.
ACKs for top commit:
maflcko:
ACK 64722e4359
Tree-SHA512: bd392c331f775605615e1b236682269b83a1e6363a4d3f09c4d8d54495cf3d22973a921ebf6b8a9f813ba6c024d3324761f3291aaf7f473995f5eaa4c195bc43
636c9862cf ci: Bump `TIDY_LLVM_V` (Hennadii Stepanov)
Pull request description:
This PR switches to the latest [IWYU 0.22](https://github.com/include-what-you-use/include-what-you-use/releases/tag/0.22), which is compatible with Clang 18.
ACKs for top commit:
fanquake:
ACK 636c9862cf
Tree-SHA512: 78ce89244c5e487dd1be8b4bd2ca6f06d19b04b78289ebc21985110574053545dcce5eb622edf2bede2cf7bb58360170e976d30a4484a127d34dd17b1c604e9c
fa5844f06d Remove unused g++-10 workaround (MarcoFalke)
fa8409e760 build: Bump g++ minimum supported version to 11 (MarcoFalke)
Pull request description:
This drops support for vanilla Ubuntu Focal 20.04 and Debian (Oldstable) Bullseye, compiling from source. Users on those operating systems would have to stick with a pre-compiled release, a previous release branch of Bitcoin Core, upgrade their system, compile their own compiler, or use a non-vanilla PPA or package manager.
Otherwise, g++-11 is offered by common distributions:
* https://packages.ubuntu.com/jammy/g++ (`g++-11`)
* https://packages.debian.org/bookworm/g++ (`g++-12`)
* FreeBSD 12/13 ships with g++ 12
* CentOS-like 9 ships with g++ 11
* OpenSuse Tumbleweed ships with g++ 13 https://software.opensuse.org/package/gcc13-c++ (No idea about OpenSuse Leap)
ACKs for top commit:
TheCharlatan:
ACK fa5844f06d
fanquake:
ACK fa5844f06d
Tree-SHA512: fc72d3a53956a0a4a6475ebf56b5fce76c3c4c793ed8e774327cad2b0f307d2d1c8aeafe2a414a7eb51f8de6d4bb78d30b8f60bf6e383234079851e72015c6e3
5555395c15 lint: Use git --no-pager to print any output in one go (MarcoFalke)
fa5729436c lint: Fix lint-whitespace issues (MarcoFalke)
Pull request description:
The lint check has many issues:
* It uses `COMMIT_RANGE`, which is brittle code, apparently making it harder to run the CI locally, or self-hosted. See https://github.com/bitcoin/bitcoin/pull/29274#discussion_r1457739319
* The result depends on `COMMIT_RANGE`, or the number of commits passed to the script, which can cause false negatives or false positives.
* It is based on the diff output, parsing it, and printing it again, which is brittle as well.
* The output does not include line number, making it harder to act on a lint error.
Fix all issues by removing the script and replacing it with a simple call to `git grep -I --line-number ...`.
ACKs for top commit:
TheCharlatan:
Re-ACK 5555395c15
Tree-SHA512: 71ea8b6382af064beb72fb17f21a0ae9e9238c97e7fa43c2ec353fd1dd73a7bbd696ba0f0a9f65d1eff7c86cbf6cc104a992cb5450a3d50f122955e835270065
bf264e0598 test: check_mempool_result negative feerate (kevkevin)
Pull request description:
Adds test coverage in `mempool_accept.py` to check if a negative `maxfeerate` is input into `check_mempool_result`
Asserts "Amount out of range" error message and `-3` error code
Motivated by this [comment](https://github.com/bitcoin/bitcoin/pull/29434/files#r1491112250)
ACKs for top commit:
maflcko:
lgtm ACK bf264e0598
brunoerg:
nice, utACK bf264e0598
davidgumberg:
Looks great, ACK bf264e0598
Tree-SHA512: 58931b774cc887c616f2fd91af3ee65cc5db55acd8e2875c76de448c80bd4e020b057c5f4f85556431377f0d0e7553771fb285d1ec20cf64f64ec92a47776b78
10d56530e0 guix: temporarily disable powerpcle taget (fanquake)
001412a4d2 guix: use GCC 12.3.0 (fanquake)
ce54330cf6 ci: use Debian Bookworm (GCC 12) for ARM ci job (fanquake)
0da6451c58 ci: use Debian Bookworm (GCC 12) for win64 job (fanquake)
Pull request description:
Switch to using [GCC `12.3.0`](https://gcc.gnu.org/gcc-12/) to build release binaries.
Temporarily disables the `powerpc64le-linux-gnu` target due to non-determinism issues when building across `aarch64` and `x86_64`. Trying to fix the non-determinism was going to require trying to selectively disable optimization flags, which is already not ideal (and didn't fix all issues), and the migration to GCC 12 as our release compiler is now the blocker for multiple other (c++20 and similar) changes, so leaving this blocked on the `powerpc64le` binaries does not seem like a good tradeoff.
ACKs for top commit:
TheCharlatan:
ACK 10d56530e0
Tree-SHA512: 401bbaaf2b72c795a06a24875ffd666151b41bae8f45bda10526ff4f6b59782704246afc6585f6b849021cbff8a7b861961d139bffe45536aaaeb3952b72ae57
0831b54dfc test: simplify test_runner.py (tdb3)
Pull request description:
Implements the simplifications to test_runner.py proposed by sipa in PR #23995.
Remove the num_running variable as it can be implied by the length of the jobs list.
Remove the i variable as it can be implied by the length of the test_results list.
Instead of counting results to determine if finished, make the queue object itself
responsible (by looking at running jobs and jobs left).
ACKs for top commit:
mzumsande:
re-ACK 0831b54
davidgumberg:
reACK 0831b54dfc
marcofleon:
re-ACK 0831b54dfc
Tree-SHA512: e5473e68d49cd779b29d97635329283ae7195412cb1e92461675715ca7eedb6519a1a93ba28d40ca6f015d270f7bcd3e77cef279d9cd655155ab7805b49638f1
c70e4fc9a3 netbase: remove unnecessary log message (Matthew Zipkin)
Pull request description:
This is a follow-up to #27375 that removes a spammy non-debug-level log message we don't need.
See https://github.com/bitcoin/bitcoin/pull/27375#issuecomment-1994498888
ACKs for top commit:
fanquake:
ACK c70e4fc9a3 - thanks. Merging this now because it's swamping non-debug logs. i.e:
Tree-SHA512: 837682860abdf740fce5dc88c8599e066660cf16b4cab1473391eb51ad538ae52d236ecd3543df866e2a2165870397a8bf21ad9f5125ed8212a3fe207d615553
1ec6684b08 test: Add test for loadtxoutset when headers are not synced (Fabian Jahr)
2bc1ecfaa9 test: Remove unnecessary sync_blocks in assumeutxo tests (Fabian Jahr)
Pull request description:
It adds a test for the change to `loadtxoutset` made in #29345. Before that change the test doesn't fail right away but times out after 10 minutes.
Also removes a `sync_blocks` call that didn't seem to do anything valuable.
ACKs for top commit:
achow101:
ACK 1ec6684b08
pablomartin4btc:
tACK 1ec6684b08
BrandonOdiwuor:
ACK 1ec6684b08
theStack:
ACK 1ec6684b08
Tree-SHA512: 1337decdf91e4a4f7213fcf8ace1d705e5c1422e0ac3872a59b5be9c33e743850cb8f5f7474750a534952eefd5cfe43fe85a54efb9bc0e47515136a2903676e5
28287cfbe1 test: add script compression coverage for not-on-curve P2PK outputs (Sebastian Falbesoner)
Pull request description:
This PR adds unit test coverage for the script compression functions `{Compress,Decompress}Script` in the special case of uncompressed P2PK outputs (scriptPubKey: OP_PUSH65 <0x04 ....> OP_CHECKSIG) with [pubkeys that are not fully valid](44b05bf3fe/src/pubkey.cpp (L297-L302)), i.e. where the encoded point is not on the secp256k1 curve. For those outputs, script compression is not possible, as the y coordinate of the pubkey can't be recovered (see also call-site of `IsToPubKey`):
44b05bf3fe/src/compressor.cpp (L49-L50)
Likewise, for a compressed script of an uncompressed P2PK script (i.e. compression ids 4 and 5) where the x coordinate is not on the curve, decompression fails:
44b05bf3fe/src/compressor.cpp (L122-L129)
Note that the term "compression" is used here in two different meanings (though they are related), which might be a little confusing. The encoding of a pubkey can either be compressed (33-bytes with 0x02/0x03 prefixes) or uncompressed (65-bytes with 0x04 prefix). On the other hand there is also compression for whole output scripts, which is used for storing scriptPubKeys in the UTXO set in a compact way (and also for the `dumptxoutset` result, accordingly). P2PK output scripts with uncompressed pubkeys get compressed by storing only the x-coordinate and the sign as a prefix (0x04 = even, 0x05 = odd). Was diving deeper into the subject while working on https://github.com/bitcoin/bitcoin/pull/27432, where the script decompression of uncompressed P2PK needed special handling (see also https://github.com/bitcoin/bitcoin/issues/24628#issuecomment-1108798536).
Trivia: as of now (block 801066), there are 13 uncompressed P2PK outputs in the UTXO set with a pubkey not on the curve (which obviously means they are unspendable).
ACKs for top commit:
achow101:
ACK 28287cfbe1
tdb3:
ACK for 28287cfbe1.
cbergqvist:
ACK 28287cf!
marcofleon:
Nicely done, ACK 28287cfbe1. Built the PR branch, ran the unit and functional tests, everything passed.
Tree-SHA512: 777b6c3065654fbfa1ce94926f4cadb91a9ca9dc4dd4af6008ad77bd1da5416f156ad0dfa880d26faab2e168bf9b27e0a068abc9a2be2534d82bee61ee055c65
And thread the feerate value through ProcessNewPackage to
reject individual transactions that exceed the given
feerate. This allows subpackage processing, and is
compatible with future package RBF work.
07cd510ffe [refactor] consolidate invalid MempoolAcceptResult processing (glozow)
9353aa4066 [refactor] consolidate valid MempoolAcceptResult processing (glozow)
Pull request description:
Every time we try to `ProcessTransaction` (i.e. submit a tx to mempool), we use the result to update a few net processing data structures. For example, after a failure, the {wtxid, txid, both, neither} (depending on reason) should be cached in `m_recent_rejects` so we don't try to download/validate it again.
There are 2 current places and at least 1 future place where we need to process `MempoolAcceptResult`:
- In the `ProcessMessage` logic after receiving and validating a tx
- In `ProcessOrphanTx` where we retry orphans
- With #28970, after processing a package of transactions, we should do these updates for each tx in the package.
Consolidate this code so it isn't repeated in 2 places and so we can reuse it in a future PR.
ACKs for top commit:
instagibbs:
ACK 07cd510ffe
achow101:
ACK 07cd510ffe
dergoegge:
Code review ACK 07cd510ffe
TheCharlatan:
ACK 07cd510ffe
Tree-SHA512: c4e74cb65e4f52882fca52e6682efa5dcf1562d98418454e09be256ffd026caae642a90aa5b9cccaae214be240d6f4be9d87b516953b2ee69a655f16ea569ed9
567cec9a05 doc: add release notes and help text for unix sockets (Matthew Zipkin)
bfe5192891 test: cover UNIX sockets in feature_proxy.py (Matthew Zipkin)
c65c0d0163 init: allow UNIX socket path for -proxy and -onion (Matthew Zipkin)
c3bd43142e gui: accomodate unix socket Proxy in updateDefaultProxyNets() (Matthew Zipkin)
a88bf9dedd i2p: construct Session with Proxy instead of CService (Matthew Zipkin)
d9318a37ec net: split ConnectToSocket() from ConnectDirectly() for unix sockets (Matthew Zipkin)
ac2ecf3182 proxy: rename randomize_credentials to m_randomize_credentials (Matthew Zipkin)
a89c3f59dc netbase: extend Proxy class to wrap UNIX socket as well as TCP (Matthew Zipkin)
3a7d6548ef net: move CreateSock() calls from ConnectNode() to netbase methods (Matthew Zipkin)
74f568cb6f netbase: allow CreateSock() to create UNIX sockets if supported (Matthew Zipkin)
bae86c8d31 netbase: refactor CreateSock() to accept sa_family_t (Matthew Zipkin)
adb3a3e51d configure: test for unix domain sockets (Matthew Zipkin)
Pull request description:
Closes https://github.com/bitcoin/bitcoin/issues/27252
UNIX domain sockets are a mechanism for inter-process communication that are faster than local TCP ports (because there is no need for TCP overhead) and potentially more secure because access is managed by the filesystem instead of serving an open port on the system.
There has been work on [unix domain sockets before](https://github.com/bitcoin/bitcoin/pull/9979) but for now I just wanted to start on this single use-case which is enabling unix sockets from the client side, specifically connecting to a local Tor proxy (Tor can listen on unix sockets and even enforces strict curent-user-only access permission before binding) configured by `-onion=` or `-proxy=`
I copied the prefix `unix:` usage from Tor. With this patch built locally you can test with your own filesystem path (example):
`tor --SocksPort unix:/Users/matthewzipkin/torsocket/x`
`bitcoind -proxy=unix:/Users/matthewzipkin/torsocket/x`
Prep work for this feature includes:
- Moving where and how we create `sockaddr` and `Sock` to accommodate `AF_UNIX` without disturbing `CService`
- Expanding `Proxy` class to represent either a `CService` or a UNIX socket (by its file path)
Future work:
- Enable UNIX sockets for ZMQ (https://github.com/bitcoin/bitcoin/pull/27679)
- Enable UNIX sockets for I2P SAM proxy (some code is included in this PR but not tested or exposed to user options yet)
- Enable UNIX sockets on windows where supported
- Update Network Proxies dialog in GUI to support UNIX sockets
ACKs for top commit:
Sjors:
re-ACK 567cec9a05
tdb3:
re ACK for 567cec9a05.
achow101:
ACK 567cec9a05
vasild:
ACK 567cec9a05
Tree-SHA512: de81860e56d5de83217a18df4c35297732b4ad491e293a0153d2d02a0bde1d022700a1131279b187ef219651487537354b9d06d10fde56225500c7e257df92c1
Remove the num_running variable as it can be implied by the
length of the jobs list.
Remove the i variable as it can be implied by the length of the
test_results list.
Instead of counting results to determine if finished, make the
queue object itself responsible (by looking at running jobs and
jobs left).
Originally proposed by @sipa in PR #23995.
Co-authored-by: Pieter Wuille <pieter@wuille.net>
The nKey of the addrman is generated the first time the node is
started. Therefore, restarting a node or turning it off and on
again won't make a previously non-deterministic addrman
deterministic.
Co-authored-by: 0xb10c <b10c@b10c.me>
d0e6564240 log: Remove error() reference (Fabian Jahr)
Pull request description:
Mini-followup to #29236 that was just merged. Removes a reference to `error()` that was missed in a comment.
ACKs for top commit:
ryanofsky:
Code review ACK d0e6564240. Just dropped LogPrintf reference since last review
stickies-v:
ACK d0e6564240
Empact:
ACK d0e6564240
Tree-SHA512: 8abe4895951013c2ceca9a57743aacabaf8af831d07eee9ae8372c121c16e88b7226f0e537200c3464792e19ac7e03b57ba0be31f43add8802753972b0aefc48
e710cefd57 rest: read raw block in rest_block and deserialize for json (Andrew Toth)
95ce0783a6 rpc: read raw block in getblock and deserialize for verbosity > 0 (Andrew Toth)
0865ab8712 test: check more details on zmq raw block response (Andrew Toth)
38265cc14e zmq: read raw block with ReadRawBlockFromDisk (Andrew Toth)
da338aada7 blockstorage: check nPos in ReadRawBlockFromDisk before seeking back (Andrew Toth)
Pull request description:
For the `getblock` endpoint with `verbosity=0`, the `rest_block` REST endpoint for `bin` and `hex`, and zmq `NotifyBlock` we don't have to deserialize the block since we're just sending the raw data. This PR uses `ReadRawBlockFromDisk` instead of `ReadBlockFromDisk` to serve these requests, and only deserializes for `verbosity > 0` and `json` REST requests. See benchmarks in https://github.com/bitcoin/bitcoin/pull/26684.
Benchmarked using ApacheBench. Requesting block 750,000 in binary 10k times on a single core (set `-rest=1` in config):
`ab -n 10000 -c 1 "http://127.0.0.1:8332/rest/block/0000000000000000000592a974b1b9f087cb77628bb4a097d5c2c11b3476a58e.bin"`
On master, mean time 15ms.
On this branch, mean time 1ms.
For RPC
```
echo '{"jsonrpc": "1.0", "id": "curltest", "method": "getblock", "params": ["0000000000000000000592a974b1b9f087cb77628bb4a097d5c2c11b3476a58e", 0]}' > /tmp/data.json
ab -p /tmp/data.json -n 1000 -c 1 -A user:password "http://127.0.0.1:8332/"
```
On master, mean time 32ms
On this branch, mean time 13ms
ACKs for top commit:
achow101:
re-ACK e710cefd57
Tree-SHA512: 4cea13c7b563b2139d041b1fdcfdb793c8cc688654ae08db07e7ee6b875c5e582b8185db3ae603abbfb06d2164724f29205774620b48c493726b991999af289e
0a533613fb docs: add release notes for #27114 (brunoerg)
e6b8f19de9 test: add coverage for whitelisting manual connections (brunoerg)
c985eb854c test: add option to speed up tx relay/mempool sync (brunoerg)
66bc6e2d17 Accept "in" and "out" flags to -whitelist to allow whitelisting manual connections (Luke Dashjr)
8e06be347c net_processing: Move extra service flag into InitializeNode (Luke Dashjr)
9133fd69a5 net: Move `NetPermissionFlags::Implicit` verification to `AddWhitelistPermissionFlags` (Luke Dashjr)
2863d7dddb net: store `-whitelist{force}relay` values in `CConnman` (brunoerg)
Pull request description:
Revives #17167. It allows whitelisting manual connections. Fixes #9923
Since there are some PRs/issues around this topic, I'll list some motivations/comments for whitelisting outbound connections from them:
- Speed-up tx relay/mempool sync for testing purposes (my personal motivation for this) - In #26970, theStack pointed out that we whitelist peers to speed up tx relay for fast mempool synchronization, however, since it applies only for inbound connections and considering the topology `node0 <--- node1 <---- node2 <--- ... <-- nodeN`, if a tx is submitted from any node other than node0, the mempool synchronization can take quite long.
- https://github.com/bitcoin/bitcoin/pull/29058#issuecomment-1865155764 - "Before enabling -v2transport by default (which I'd image may happen after https://github.com/bitcoin/bitcoin/pull/24748) we could consider a way to force manual connections to be only-v1 or even only-v2 (disabling reconnect-with-v1). A possibility could be through a net permission flag, if https://github.com/bitcoin/bitcoin/pull/27114 makes it in."
- https://github.com/bitcoin/bitcoin/pull/17167#issuecomment-1168606032 - "This would allow us to use https://github.com/bitcoin/bitcoin/pull/25355 when making outgoing connections to all nodes, except to whitelisted ones for which we would use our persistent I2P address."
- Force-relay/mempool permissions for a node you intentionally connected to.
ACKs for top commit:
achow101:
ACK 0a533613fb
sr-gi:
re-ACK [0a53361](0a533613fb)
pinheadmz:
ACK 0a533613fb
Tree-SHA512: 97a79bb854110da04540897d2619eda409d829016aafdf1825ab5515334b0b42ef82f33cd41587af235b3af6ddcec3f2905ca038b5ab22e4c8a03d34f27aebe1
Note that for speed this commit also removes the proof of work and
signet signature checks before returning the block in getblock.
It is assumed if a block is stored it will be valid.
Note that for speed this commit also removes the proof of work and
signet signature checks before returning the block in getblock.
It is assumed if a block is stored it will be valid.
ReadRawBlockFromDisk assumes a non-null pos that has an nPos >= 8.
This simple check makes the function safer to call in the future,
so callers don't need to worry about causing UB if the pos is null.
Retain native GCC 10 toolchain for macOS, to prevent compile failures in
native tools (this will be removed entirely when we tansition to LLD).
Update the vmov-alignment patch, for changes in GCC 12.
