5aad635 Use memset() to optimize prevector::resize() (Evan Klitzke)
e46be25 Reduce redundant code of prevector and speed it up (Akio Nakamura)
f0e7aa7 Add new prevector benchmarks. (Evan Klitzke)
Pull request description:
This branch optimizes various `prevector` operations, especially resizing vectors. While profiling the `loadblk` thread I noticed that a lot of time was being spent in `prevector::resize()` which led to this work. I have some data here indicating that it takes up **37%** of the time in `ReadBlockFromDisk()`: https://monad.io/readblockfromdisk.svg
This branch improves things significantly. For trivial types, the new results for the prevector benchmark are:
* `PrevectorClearTrivial` which tests `prevector::clear()` becomes 24.6x faster
* `PrevectorDestructorTrivial` which tests `prevector::~prevector()` becomes 20.5x faster
* `PrevectorResizeTrivial` which tests `prevector::resize()` becomes 20.3x faster
Note that in practice it looks like the prevector is only used to contain `unsigned char` types, which is a trivial type. The benchmarks are testing a bit of an extreme case, but the changes here are motivated by the profiling data for `ReadBlockFromDisk()` I linked to above.
The pull request here consists of a series of three commits:
* The first adds new benchmarks but does not change the prevector code.
* The second is from @AkioNak , and merges some prevector optimizations he submitted in #11988
* The third optimizes `prevector::resize()` to use `memset()` when the prevector contains trivially constructible types
Tree-SHA512: 28f7cbb91a19f9f43b6a5942781d7eb2e3197389186b666f086b69df12bee37773140f765426d715bfb8ebff79cb27a5f1206d0325b54b4aa65598b50fb18368
Further optimize prevector::resize() (which is called by a number of
other prevector methods) to use memset to initialize memory when the
prevector contains trivial types.
In prevector.h, the code which like item_ptr(size()) apears in the loop.
Both item_ptr() and size() judge whether values are held directly or
indirectly, but in most cases it is sufficient to make that judgement
once outside the loop.
This PR adds 2 private function fill() which has the loop to initialize
by specified value (or iterator of the other prevector's element),
but don't call item_ptr() in their loop.
Other functions(assign(), constructor, operator=(), insert())
that has similar loop, call fill() instead of original loop.
Also, resize() was changed like fill(), but it calls the default
constructor for that element each time.
90ba2df11 Fix missing cs_main lock for GuessVerificationProgress() (Jonas Schnelli)
Pull request description:
`GuessVerificationProgress()` needs `cs_main` due to accessing the `pindex->nChainTx`.
This adds a `AssertLockHeld` in `GuessVerificationProgress()` and adds the missing locks in...
* `LoadChainTip()`
* `ScanForWalletTransactions()` (got missed in #11281)
* GUI, `ClientModel::getVerificationProgress()` <--- **this may have GUI performance impacts**, but could be relaxed later with a cache or something more efficient.
Tree-SHA512: 13302946571422375f32af8e396b9d2c1180f2693ea363aeba9e98c8266ddec64fe7862bfdcbb5a93a4b12165a61eec1e51e4e7d7a8515fa50879095dc163412
835a21b Squashed 'src/leveldb/' changes from c521b3ac65..64052c76c5 (MarcoFalke)
Pull request description:
Pull in changes from https://github.com/bitcoin/bitcoin/pull/11674#issuecomment-348174674.
Merges cleanly into master and 0.16 branch.
Tree-SHA512: 819c042c0dfac8dc3078fc182c1e22d4a85b343967475d3389be5b5b056361114d8c9892437cd1dc4b45808c27880c0e166e047afc2c2bd2bbc33e55336a8c33
d2ee6e3 init: Remove translation for `-blockmaxsize` option help (Wladimir J. van der Laan)
Pull request description:
Move `-blockmaxsize`, a deprecated option which is replaced by `-blockmaxweight`, to debug options and remove the translation.
This message is absolutely terrible for translators (esp the `* 4` part).
(for 0.17 we should probably remove this option completely?)
(reported by French Language Coordinator)
Tree-SHA512: 379150c9217672d2f2f93b4c02a3ac638e77ca56fb518e30c56c46d59f89eac422b4f540e70a9abd3c6ad653ac4b786d4734621b18f93804885d81e223f1a908
a6e6e39a8b Bugfix: respect user defined configuration file (-conf) when open conf. file from QT settings (Jonas Schnelli)
Pull request description:
Fixes#12488.
In master, opening the configuration file from the GUI settings will always open the file "bitcoin.conf" regardless of the `-conf=` settings.
This PR makes the GUI settings open configuration file function respect the `-conf` option.
Tree-SHA512: fb54cc699b4d2a3947f749fdf5f1a51251ffd67d0f6c6a937a5b80f0ba5a5c1085d0eef190453bbc04696d4d76c2c266de0fe9712e65e4bb36116158b54263d4
Move `-blockmaxsize`, a deprecated option which is replaced by
`-blockmaxweight`, to debug options and remove the translation.
This message is absolutely terrible for translators (esp the `* 4`
part).
Splits signrwatransaction into a wallet version (signrawtransactionwithwallet) and
non-wallet version (signrawtransactionwithkey). signrawtransaction is marked as DEPRECATED
and will call the right signrawtransaction* command as per the parameters in order to
maintain compatibility.
Updated signrawtransactions test to use new RPCs
b22cce014 scripted-diff: validateaddress to getaddressinfo in tests (Andrew Chow)
b98bfc5ed Create getaddressinfo RPC and deprecate parts of validateaddress (Andrew Chow)
1598f3230 [rpc] Move DescribeAddressVisitor to rpc/util (John Newbery)
39633ecd5 [rpc] split wallet and non-wallet parts of DescribeAddressVisitor (John Newbery)
Pull request description:
This PR makes a new RPC command called `getaddressinfo` which relies on the wallet. It contains all of `validateaddress`'s address info stuff. Those parts in `validateaddress` have been marked as deprecated. The tests have been updated to use `getaddressinfo` except the `disablewallet` test which is the only test that actually uses `validateaddress` to validate an address.
Tree-SHA512: ce00ed0f2416200b8de1e0a75e8517c024be0b6153457d302c3879b3491cce28191e7c29aed08ec7d2eeeadc62918f5c43a7cb79cd2e4b6d9291bd83ec31c852
Moves the parts of validateaddress which require the wallet into getaddressinfo
which is part of the wallet RPCs. Mark those parts of validateaddress which
require the wallet as deprecated.
Validateaddress will call getaddressinfo
for the data that both share for right now.
Moves IsMine functions to libbitcoin_common and then links libbitcoin_wallet
before libbitcoin_common in order to prevent linker errors since IsMine is no
longer used in libbitcoin_server.
be45a67 Add some script tests related to BOOL ops and odd values like negative 0. (Richard Kiss)
Pull request description:
Add some script tests related to BOOL ops and odd values like negative 0.
Tree-SHA512: 8e633f7ea5eea39e31016994baf60f295fa1dc8cae27aa5fcfc741ea97136bfb3ddc57bb62b9c6bf9fe256fc09cdd184906ba8e611e297cf8d2d363da2bbf1d4
5f605e1 Make signrawtransaction accept P2SH-P2WSH redeemscripts (Pieter Wuille)
Pull request description:
This is a quick fix for #12418, which is a regression in 0.16.
It permits specifying just the inner redeemscript to let `signrawtransaction` succeed. This inner redeemscript is already reported by `addmultisigaddress` & co.
#11708 uses a different approach, where `listunspent` reports both inner & outer redeemscript, but requires both to be provided to `signrawtransaction`. Part of #11708 is still needed even in combination with this PR however, as currently the inner redeemscript isn't reported by `listunspent`.
Tree-SHA512: a6fa2b2661ce04db25cf029dd31da39c0b4811d43692f816dfe0f77b4159b5e2952051664356a579f690ccd58a626e0975708afcd7ad5919366c490944e3a9a5
1d4cbd2 test: Add unit test for LockDirectory (Wladimir J. van der Laan)
fc888bf util: Fix multiple use of LockDirectory (Wladimir J. van der Laan)
Pull request description:
Wrap the `boost::interprocess::file_lock` in a `std::unique_ptr` inside the map that keeps track of per-directory locks.
This fixes a build issue with the clang 4.0.0+boost-1.58.0p8 version combo on OpenBSD 6.2, and should have no effect otherwise.
Also add a unit test, make the function thread-safe, and fix Linux versus Windows behavior inconsistency.
Meant to fix#12413.
Tree-SHA512: 1a94c714c932524a51212c46e8951c129337d57b00fd3da5a347c6bcf6a947706cd440f39df935591b2079995136917f71ca7435fb356f6e8a128c509a62ec32
bb00c95 Consistently use FormatStateMessage in RPC error output (Ben Woosley)
8b8a1c4 Add test for 'mempool min fee not met' rpc error (Ben Woosley)
c04e0f6 Fix 'mempool min fee not met' debug output (Ben Woosley)
Pull request description:
Output the value that is tested, rather than the unmodified fee value.
Prompted by looking into: #11955
Tree-SHA512: fc0bad47d4af375d208f657a6ccbad6ef7f4e2989ae2ce1171226c22fa92847494a2c55cca687bd5a1548663ed3313569bcc31c00d53c0c193a1b865dd8a7657
This commit fixes problems with calling LockDirectory multiple times on
the same directory, or from multiple threads. It also fixes the build on
OpenBSD.
- Wrap the boost::interprocess::file_lock in a std::unique_ptr inside
the map that keeps track of per-directory locks. This fixes a build
issue with the clang 4.0.0+boost-1.58.0p8 version combo on OpenBSD
6.2, and should have no observable effect otherwise.
- Protect the locks map using a mutex.
- Make sure that only locks that are successfully acquired are inserted
in the map.
- Open the lock file for appending only if we know we don't have the
lock yet - The `FILE* file = fsbridge::fopen(pathLockFile, "a");`
wipes the 'we own this lock' administration, likely because it opens
a new fd for the locked file then closes it.
fa27623 qt: Initialize members in WalletModel (MarcoFalke)
Pull request description:
This prevents segfaults (or errors when running qt in valgrind)
```
Conditional jump or move depends on uninitialised value(s)
WalletModel::checkBalanceChanged() (walletmodel.cpp:156)
Tree-SHA512: 38c8c03c7fa947edb3f1c13eab2ac7a62ef8f8141603c2329a7dc5821a887a349af8014dc739b762e046f410f44a9c6653b6930f08b53496cf66381cadc06246
2e9406c Interrupt loading thread after shutdown request (João Barbosa)
Pull request description:
This change (currently) avoids loading the mempool if shutdown is requested.
Tree-SHA512: 3dca3a6ea5b09bd71db0974584d93dfe81819bc0bdbb4d9b6fa0474755306d1403f6c058ecb8211384493a8f7ca3a9134173db744b7344043cfc7d79286c8fd4
d6f3a73 Remove redundant locks (practicalswift)
Pull request description:
Remove redundant locks:
* ~~`FindNode(...)` is locking `cs_vNodes` internally~~
* `SetAddressBook(...)` is locking `cs_wallet` internally
* `DelAddressBook(...)` is locking `cs_wallet` internally
**Note to reviewers:** From what I can tell these locks are redundantly held from a data integrity perspective (guarding specific variables), and they do not appear to be needed from a data consistency perspective (ensuring a consistent state at the right points). Review thoroughly and please let me know if I'm mistaken :-)
Tree-SHA512: 7e3ca2d52fecb16385dc65051b5b20d81b502c0025d70b0c489eb3881866bdd57947a9c96931f7b213f5a8a76b6d2c7b084dff0ef2028a1e9ca9ccfd83e5b91e
6ef86c9 Do not un-mark fInMempool on wallet txn if ATMP fails. (Matt Corallo)
Pull request description:
Irrespective of the failure reason, un-marking fInMempool
out-of-order is incorrect - it should be unmarked when
TransactionRemovedFromMempool fires.
Clean up of #11839, which I think was the wrong fix.
Tree-SHA512: 580731297eeac4c4c99ec695e15b09febf62249237bc367fcd1830fc811d3166f9336e7aba7f2f6f8601960984ae22cebed781200db0f04e7cd2008db1a83f64
b7f6002ed5 Fix rescan test failure due to unset g_address_type, g_change_type (Russell Yanofsky)
Pull request description:
New global variables were introduced in #11403 and not setting them causes:
```
test_bitcoin: wallet/wallet.cpp:4259: CTxDestination GetDestinationForKey(const CPubKey&, OutputType): Assertion `false' failed.
unknown location(0): fatal error in "importwallet_rescan": signal: SIGABRT (application abort requested)
```
It's possible to reproduce the failure reliably by running:
```
src/test/test_bitcoin --log_level=test_suite --run_test=wallet_tests/importwallet_rescan
```
Failures happen nondeterministically because boost test framework doesn't run tests in a specified order, and tests that run previously can set the global variables and mask the bug.
This is similar to bug #12150. Example travis failure is https://travis-ci.org/bitcoin/bitcoin/jobs/340642010
Tree-SHA512: ab40662b3356892b726f1f552e22d58d86b5e982538741e52b37ee447a0c97c76c24ae543687edf2e25d9dd925722909d37abfae95d93bf09e23fa245a4c3351
New global variables were introduced in #11403 and not setting them causes:
test_bitcoin: wallet/wallet.cpp:4259: CTxDestination GetDestinationForKey(const CPubKey&, OutputType): Assertion `false' failed.
unknown location(0): fatal error in "importwallet_rescan": signal: SIGABRT (application abort requested)
It's possible to reproduce the failure reliably by running:
src/test/test_bitcoin --log_level=test_suite --run_test=wallet_tests/importwallet_rescan
Failures happen nondeterministically because boost test framework doesn't run
tests in a specified order, and tests that run previously can set the global
variables and mask the bug.
a71c56a clientversion: Use full commit hash for commit-based version descriptions (Luke Dashjr)
Pull request description:
git keeps changing the number of digits in abbreviated hashes, resulting in the GitHub archive hash changing because we include it here.
To workaround this and avoid hashes that become increasingly ambiguous later on, just include the full commit hash when building from git.
This has no effect on tagged releases.
(Cleanly mergable back to 0.10 without backport)
Tree-SHA512: b0be5391fadd16fbc9bbeffe1574a61c95931cbf6dea885d7e3cfcd3474b89e71767b1b55b4eeeeb66e4e119e78ff579cd9d206366d36928a209a31e1c1eed75
91986ed206 scripted-diff: Use UniValue.pushKV instead of push_back(Pair()) (Karel Bilek)
a570098021 Squashed 'src/univalue/' changes from 07947ff2da..51d3ab34ba (MarcoFalke)
Pull request description:
Rebased version of #11386 by karel-3d.
Closes: #11386
Tree-SHA512: f3a81447e573c17e75813f4d41ceb34b9980eac81efdd98ddb149d7c51f792be7e2b32239b6ea7e6da68af23897afa6b4ce3f4e8070f9c4adf5105bf6075f2a0
faca18dcf feebumper: Use PreconditionChecks to determine bump eligibility (MarcoFalke)
718f05cab move more bumpfee prechecks to feebumper::PreconditionChecks (Gregory Sanders)
Pull request description:
This only affects the gui.
Fee-bumping of transactions that are already confirmed or are already conflicted by other transactions should not be offered by the gui.
Tree-SHA512: 4acf8087c69fbe5bd67be0485cdb4055e985bbf84acc420aa786ad31e2dc6c2572baaac1d359af10a6907790f626edca690285d9a46ae5440900ea12624c634f
a25cb0f Use ptrdiff_t type to more precisely indicate usage and avoid compiler warnings. (murrayn)
Pull request description:
ptrdiff_t is a more strictly correct type, and gets rid of compiler warnings.
Tree-SHA512: 39718a5cdc10e698f14185f4622a9b439728bce619bd8b3a86f2b99ed5b056cf5a8545a3e5c4bc8a6a01b845fb73510036cee5e6d2629c58df26be692a957fba
Wladimir J. van der Laan
