16e3cd380a Clarify include recommendation (practicalswift)
6d10f43738 Enforce the use of bracket syntax includes ("#include <foo.h>") (practicalswift)
906bee8e5f Use bracket syntax includes ("#include <foo.h>") (practicalswift)
Pull request description:
When analysing includes in the project it is often assumed that the preferred bracket include syntax (`#include <foo.h>`) mentioned in `developer-docs.md` is used consistently. @sipa:s excellent circular dependencies script [`circular-dependencies.py`](50c69b7801/contrib/devtools/circular-dependencies.py) (#13228) is an example of a script making this reasonable assumption.
This PR enables automatic Travis checking of the include syntax making sure that the bracket syntax includes (`#include <foo.h>`) is used consistently.
Tree-SHA512: a414921aabe8e487ebed42f3f1cbd02fecd1add385065c1f2244cd602c31889e61fea5a801507ec501ef9bd309b05d3c999f915cec1c2b44f085bb0d2835c182
f77e1d34fd test: Add MempoolAncestryTests (Karl-Johan Alm)
a08d76bcfe mempool: Calculate descendant maximum thoroughly (Karl-Johan Alm)
6d3568371e wallet: Switch to using ancestor/descendant limits (Karl-Johan Alm)
6888195b06 wallet: Strictly greater than for ancestor caps (Karl-Johan Alm)
322b12ac4e Remove deprecated TransactionWithinChainLimit (Karl-Johan Alm)
4784751547 Switch to GetTransactionAncestry() in OutputEligibleForSpending (Karl-Johan Alm)
475a385a80 Add GetTransactionAncestry to CTxMemPool for general purpose chain limit checking (Karl-Johan Alm)
46847d69d2 mempool: Fix max descendants check (Karl-Johan Alm)
b9ef21dd72 mempool: Add explicit max_descendants (Karl-Johan Alm)
Pull request description:
Currently, `TransactionWithinChainLimit` is restricted to single-output use, and needs to be called every time for different limits. If it is replaced with a chain limit value calculator, that can be called once and reused, and is generally more flexible (see e.g. #12257).
Update: this PR now corrects usage of max ancestors / max descendants, including calculating the correct max descendant value, as advertised for the two limits.
~~This change also makes `nMaxAncestors` signed, as the replacement method will return `-1` for "not in the mempool", which is different from "0", which means "no ancestors/descendants in mempool".~~
~~This is a subset of #12257.~~
Tree-SHA512: aa59c849360542362b3126c0e29d44d3d58f11898e277d38c034dc4b86a5b4500f77ac61767599ce878c876b5c446fec9c02699797eb2fa41e530ec863a00cf9
67e0e04140 [wallet] [docs] Update release notes for removing `getlabeladdress` (John Newbery)
81608178cf [wallet] [rpc] Remove getlabeladdress RPC (John Newbery)
Pull request description:
labels are associated with addresses (rather than addresses being
associated with labels, as was the case with accounts). The
getlabeladdress does not make sense in this model, so remove it.
getaccountaddress is still supported for one release as the accounts
API is deprecated.
Tree-SHA512: 7f45d0456248ebcc4e54dd34e2578a09a8ea8e4fceda75238ccea9d731dc99a3f3c0519b18a9739de17d2e6e59c9c2259ba67c9ae2e3cb2a40ddb14b9193fe29
f68049dd87 crypto: cleanup sha256 build (Cory Fields)
Pull request description:
Requested by @sipa in #13386.
Rather than appending all possible cpu variants to all targets, create a convenience variable that encompasses all.
Tree-SHA512: 8e9ab2185515672b79bb7925afa4f3fbfe921bfcbe61456833d15457de4feba95290de17514344ce42ee81cc38b252476cd0c29432ac48c737c2225ed515a4bd
cbede7dbfd [qt] OptionsDialog: add prune setting (Sjors Provoost)
Pull request description:
The default suggested value is 2 GB. Minimum is 1 GB (550 MB rounded up).
When the user toggles this setting, a strong warning appears that undoing requires re-downloading the chain:
<img width="478" alt="schermafbeelding 2018-05-15 om 12 35 24" src="https://user-images.githubusercontent.com/10217/40051858-7939cc20-583c-11e8-9120-327a75376732.png">
Tooltip points out that actual disk usage can be higher. It's a bit vague on the "advanced features", because I'm assuming anyone who needs to use `-rescan` and `-txindex` will read the documentation, and a more detailed text would needlessly confuse everyone else.
<img width="450" alt="schermafbeelding 2018-05-15 om 12 33 51" src="https://user-images.githubusercontent.com/10217/40051791-49d6156a-583c-11e8-97b9-7de6dfd8c481.png">
The UI uses gigabytes for readability and easy of use. There is also no manual pruning UI (`prune=1`). The user will have to use `bitcoin.conf` for those things.
Fixes#6461. When combined with #13029 the user, after pruning their node, can safely reset settings and/or use bitcoind without having to edit `bitcoin.conf`. However I don't think that's an essential prerequisite.
Tree-SHA512: e17aff276d7235fbd40796adb6431d430620788a753ee13bc064abd35d2edc4280a3d3cddc18e42b4e00edff13ed18fd4f2a966c6f0b43b689afd13673e0c4bf
Instead of combining the -limitancestorcount and -limitdescendantcount into a nMaxChainLength, this commit uses each one separately in the coin eligibility filters.
The chain limits check for max descendants would check the descendants of the transaction itself even though the description for -limitdescendantcount says 'any ancestor'. This commit corrects the descendant count check by finding the top parent transaction in the mempool and comparing against that.
TransactionWithinChainLimits would take a 'limit' and check it against ascendants and descendants. This is changed to take an explicit
max ancestors and max descendants value, and to test the corresponding value against its corresponding max.
e9a1881b90 refactor: add a function for determining if a block is pruned or not (Karl-Johan Alm)
Pull request description:
The check for whether a block is pruned or not is sufficiently obscure that it deserves a macro. It is also used in 2 places, ~~with more coming, e.g. #10757~~ (turns out it was a move, not an addition).
Tree-SHA512: b9aeb60663e1d1196df5371d5aa00b32ff5d4cdea6a77e4b566f28115cce09570c18e45e4b81a4033f67c4135c8e32c027f67bae3b75c2ea4564285578a3f4dd
2acd1d6716 Drop uint 256 not operator (Ben Woosley)
Pull request description:
All the other operators are integer or bitwise operations, and this is unused
apart from tests.
Note attempting to call `!` on `arith_uint256` results in a build error after this change:
```
test/arith_uint256_tests.cpp:201:17: error: invalid argument type 'const arith_uint256' to unary expression
BOOST_CHECK(!ZeroL);
```
Tree-SHA512: 5791b643f426dac9829e9499d678786f1ad294edb2d840879252a1b642bda55941632114f64048660a5991a984aeba49eeb5dfe64ba0a6275cbe7b1c049d7095
ec3073a274 index: Move index DBs into index/ directory. (Jim Posen)
89eddcd365 index: Remove TxIndexDB from public interface of TxIndex. (Jim Posen)
2318affd27 MOVEONLY: Move BaseIndex to its own file. (Jim Posen)
f376a49241 index: Generalize logged statements in BaseIndex. (Jim Posen)
61a1226d87 index: Extract logic from TxIndex into reusable base class. (Jim Posen)
e5af5fc6fb db: Make reusable base class for index databases. (Jim Posen)
9b0ec1a7f9 db: Remove obsolete methods from CBlockTreeDB. (Jim Posen)
Pull request description:
This refactors most of the logic in TxIndex into a reusable base class for other indices. There are two commits moving code between files, which may be be more easily reviewed using `git diff --color-moved` (https://blog.github.com/2018-04-05-git-217-released/).
The motivation for this is to support BIP 157 by indexing block filters.
<!-- Reviewable:start -->
---
This change is [<img src="https://reviewable.io/review_button.svg" height="34" align="absmiddle" alt="Reviewable"/>](https://reviewable.io/reviews/bitcoin/bitcoin/13243)
<!-- Reviewable:end -->
Tree-SHA512: 0857f04df2aa920178dab2eb8e57984d8eb4d5010deca9971190358479e05b6672ccca2a08af0a7ac9fe02afb947be84cf35a3693204d0667263c6add2959cbf
ebebedce20 speed up of tx_validationcache_tests by reusing of CTransaction. (lucash.dev@gmail.com)
Pull request description:
The code was converting CMutableTransaction to CTransaction multiple times, which implies recalculating the hash multiple times. This commit fixes this by reusing a single CTransaction.
Run-time results:
```
Before: 6.7s
After: 5.5s
--------------
Saved: 1.2s
```
This PR was split from #13050. Also, see #10026.
Tree-SHA512: 61fb81972a08299085a7d3d0060485b265aefc7a4f82ab548e5f94371c8643cfb97bf0ef34f4e1211bf853d0217fa1c3338e4117f36fda1b37d203f690e86d60
0231ef6c6d cli: Ignore libevent warnings (Cory Fields)
Pull request description:
Should fix rpc tests that fail due to an unclean stderr.
Untested as I'm not seeing these warnings. @promag mind seeing if this fixes your problem?
Tree-SHA512: fba5ae3f239b515e93e19f9c3eca659eb7fb21f1b1fec25b68285695bfd1ecbdcd9b2235543689aaf97bff85cbb762840f65365a67e791314e9a6b8db2c9e246
The code was converting CMutableTransaction to CTransaction multiple times, which implies recalculating the hash multiple times. This commit fixes this by reusing a single CTransaction.
ebec7317ca Drop the chain argument to GetDifficulty (Ben Woosley)
Pull request description:
By dropping the chain argument to `GetDifficulty`. `GetDifficulty` was called in two ways:
* with a guaranteed non-null blockindex
* with no argument
Change the latter case to be provided `chainActive.Tip()` explicitly.
Introduced in: #11748
Tree-SHA512: f2c97014be185f3e3de92db15848548650e4a67fab20a41bcfa851c5c63c245915cbe9380f84d9da2081e8756d31a41de417db1d35cfecf41ddb4f25070eb525
fa4760fbb3 qa: Increase includeconf test coverage (MarcoFalke)
Pull request description:
This adds some missing `return false` for error conditions and adds test coverage [1] for those.
Also, extend recursion warning when the chain was set in one of the includeconfs.
[1] See the red lines in https://marcofalke.github.io/btc_cov/total.coverage/src/util.cpp.gcov.html for missing coverage.
Tree-SHA512: d32563c9bb277879895a173e699034db5ecdb4061a1ec8890c566d61e36a09efa5eda19a029baf952ff6d568f8b9684a13a0bb90827850075470975e2088fee4
6aa33feadb Drop UpdateTransaction in favor of UpdateInput (Ben Woosley)
Pull request description:
Updating the input explicitly requires the caller to present a mutable
input, which more clearly communicates the effects and intent of the call
(and, often, the enclosing loop).
In most cases, this input is already immediately available and need not be
looked up.
Tree-SHA512: 8c7914a8b7ae975d8ad0e9d760e3c5da65776a5f79d060b8ffb6b3ff7a32235f71ad705f2185b368d9263742d7796bb562395d22b806d90e8502d8c496011e57
fa36aa7965 wallet: Prevent segfault when sending to unspendable witness (MarcoFalke)
Pull request description:
Previously we wouldn't care about the `txnouttype`, but after 4e91820531 we `switch` on the type.
Tree-SHA512: 6b597aba80cb43881671ad7b3a4ad97753864e8005a05c23fdd8ee79953483c08f241b5c392a9b494298eadc5cfba895b0480d916ef4f11d122fd6196f31b84a
4defdfab94 [MOVEONLY] Move unused Merkle branch code to tests (Pieter Wuille)
4437d6e1f3 8-way AVX2 implementation for double SHA256 on 64-byte inputs (Pieter Wuille)
230294bf5f 4-way SSE4.1 implementation for double SHA256 on 64-byte inputs (Pieter Wuille)
1f0e7ca09c Use SHA256D64 in Merkle root computation (Pieter Wuille)
d0c9632883 Specialized double sha256 for 64 byte inputs (Pieter Wuille)
57f34630fb Refactor SHA256 code (Pieter Wuille)
0df017889b Benchmark Merkle root computation (Pieter Wuille)
Pull request description:
This introduces a framework for specialized double-SHA256 with 64 byte inputs. 4 different implementations are provided:
* Generic C++ (reusing the normal SHA256 code)
* Specialized C++ for 64-byte inputs, but no special instructions
* 4-way using SSE4.1 intrinsics
* 8-way using AVX2 intrinsics
On my own system (AVX2 capable), I get these benchmarks for computing the Merkle root of 9001 leaves (supported lengths / special instructions / parallellism):
* 7.2 ms with varsize/naive/1way (master, non-SSE4 hardware)
* 5.8 ms with size64/naive/1way (this PR, non-SSE4 capable systems)
* 4.8 ms with varsize/SSE4/1way (master, SSE4 hardware)
* 2.9 ms with size64/SSE4/4way (this PR, SSE4 hardware)
* 1.1 ms with size64/AVX2/8way (this PR, AVX2 hardware)
Tree-SHA512: efa32d48b32820d9ce788ead4eb583949265be8c2e5f538c94bc914e92d131a57f8c1ee26c6f998e81fb0e30675d4e2eddc3360bcf632676249036018cff343e
db56755ca4 Fix "gmake check" under OpenBSD 6.3 (probably *BSD): Avoid using GNU grep specific regexp handling (practicalswift)
Pull request description:
Fixes#13337 (!)
GNU grep and BSD grep differs in the way they handle regexps when extended regular expressions are not enabled via the `-E` flag:
```
$ grep --version | head -1
grep (GNU grep) 3.1
$ echo "BOOST_FIXTURE_TEST_SUITE(foo)" | grep "BOOST_FIXTURE_TEST_SUITE(\|BOOST_AUTO_TEST_SUITE("
BOOST_FIXTURE_TEST_SUITE(foo)
$
```
```
$ grep --version | head -1
grep version 0.9
$ echo "BOOST_FIXTURE_TEST_SUITE(foo)" | grep "BOOST_FIXTURE_TEST_SUITE(\|BOOST_AUTO_TEST_SUITE("
$
```
The portable way to do it is:
```
$ echo "BOOST_FIXTURE_TEST_SUITE(foo)" | grep -E "(BOOST_FIXTURE_TEST_SUITE\\(|BOOST_AUTO_TEST_SUITE\\()"
BOOST_FIXTURE_TEST_SUITE(foo)
$
```
Tree-SHA512: d83c78f34421504dd8efc3921c98527f499045b702bd34715a5bc78e04ef2a5f49f601a55ad08632e870f137b1edada94a3f530291bc9107d8d6b16fe11e640b
493a166948 bench: Don't return a bool from main (Wladimir J. van der Laan)
Pull request description:
Return `1` from `main()` on error, not the bool `false` (introduced in #13112). This is the correct value to return on error, and also shuts up a clang warning.
Tree-SHA512: 52a0f1b2f6ae2697555f71ee2019ce657046f7f379f1f4faf3cce9d5f3fb21fcdc43a4c84895a2a8b6929997ba70bbe87c231f2f9553215b84c22333810d58d9
6b8b63af14 Generic TransactionSignatureCreator works with both CTransaction and CMutableTransaction (Martin Ankerl)
Pull request description:
Refactored `TransactionSignatureCreator` into a templated `GenericTransactionSignatureCreator` that works with both `CMutableTransaction` and `CTransaction`.
The advantage is that now in `SignSignature`, the `MutableTransactionSignatureCreator` can now operate directly with the `CMutableTransaction` without the need to copy the data into a `CTransaction`.
Running all unit tests brings a very noticable speedup on my machine:
48.4 sec before this change
36.4 sec with this change
--------
12.0 seconds saved
running only `--run_test=transaction_tests/test_big_witness_transaction`:
16.7 sec before this change
5.9 sec with this change
--------
10.8 seconds saved
This relates to my first attempt with the const_cast hack #13202, and to the slow unit test issue #10026.
Also see #13050 which modifies the tests but not the production code (like this PR) to get a speedup.
Tree-SHA512: 2cff0e9699f484f26120a40e431a24c8bc8f9e780fd89cb0ecf20c5be3eab6c43f9c359cde244abd9f3620d06c7c354e3b9dd3da41fa2ca1ac1e09386fea25fb
Return `EXIT_SUCCESS` from `main()` on error, not the bool `false`
(introduced in #13112). This is the correct value to return on error,
and also shuts up a clang warning.
Also add a final return for clarity.
