Seems providing at least minimal visibility to the failure is a good practice.
The only remaining ignored state is in LoadExternalBlockFile, where logging
would likely be spammy.
6a3b0d3 Print to console by default when not run with -daemon (Evan Klitzke)
Pull request description:
Cherry-picked ef6fa1c38e1bd115d1cce155907023d79da379d8 from the "up for grabs" PR: "Smarter default behavior for -printtoconsole" (#12689).
See previous review in #12689.
Tree-SHA512: 8923a89b9c8973286d53e960d3c464b1cd026cd5a5911ba62f9f972c83684417dc4004101815dfe987fc1e1baaec1fdd90748a0866bb5548e974d77b3135d43b
c25321f Add config changes to release notes (Anthony Towns)
5e3cbe0 [tests] Unit tests for -testnet/-regtest in [test]/[regtest] sections (Anthony Towns)
005ad26 ArgsManager: special handling for -regtest and -testnet (Anthony Towns)
608415d [tests] Unit tests for network-specific config entries (Anthony Towns)
68797e2 ArgsManager: Warn when ignoring network-specific config setting (Anthony Towns)
d1fc4d9 ArgsManager: limit some options to only apply on mainnet when in default section (Anthony Towns)
8a9817d [tests] Use regtest section in functional tests configs (Anthony Towns)
30f9407 [tests] Unit tests for config file sections (Anthony Towns)
95eb66d ArgsManager: support config file sections (Anthony Towns)
4d34fcc ArgsManager: drop m_negated_args (Anthony Towns)
3673ca3 ArgsManager: keep command line and config file arguments separate (Anthony Towns)
Pull request description:
The weekly meeting on [2017-12-07](http://www.erisian.com.au/meetbot/bitcoin-core-dev/2017/bitcoin-core-dev.2017-12-07-19.00.log.html) discussed allowing options to bitcoin to have some sensitivity to what network is in use. @theuni suggested having sections in the config file:
<cfields> an alternative to that would be sections in a config file. and on the
cmdline they'd look like namespaces. so, [testnet] port=5. or -testnet::port=5.
This approach is (more or less) supported by `boost::program_options::detail::config_file_iterator` -- when it sees a `[testnet]` section with `port=5`, it will treat that the same as "testnet.port=5". So `[testnet] port=5` (or `testnet.port=5` without the section header) in bitcoin.conf and `-testnet.port=5` on the command line.
The other aspect to this question is possibly limiting some options so that there is no possibility of accidental cross-contamination across networks. For example, if you're using a particular wallet.dat on mainnet, you may not want to accidentally use the same wallet on testnet and risk reusing keys.
I've set this up so that the `-addnode` and `-wallet` options are `NETWORK_ONLY`, so that if you have a bitcoin.conf:
wallet=/secret/wallet.dat
upnp=1
and you run `bitcoind -testnet` or `bitcoind -regtest`, then the `wallet=` setting will be ignored, and should behave as if your bitcoin.conf had specified:
upnp=1
[main]
wallet=/secret/wallet.dat
For any `NETWORK_ONLY` options, if you're using `-testnet` or `-regtest`, you'll have to add the prefix to any command line options. This was necessary for `multiwallet.py` for instance.
I've left the "default" options as taking precedence over network specific ones, which might be backwards. So if you have:
maxmempool=200
[regtest]
maxmempool=100
your maxmempool will still be 200 on regtest. The advantage of doing it this way is that if you have `[regtest] maxmempool=100` in bitcoin.conf, and then say `bitcoind -regtest -maxmempool=200`, the same result is probably in line with what you expect...
The other thing to note is that I'm using the chain names from `chainparamsbase.cpp` / `ChainNameFromCommandLine`, so the sections are `[main]`, `[test]` and `[regtest]`; not `[mainnet]` or `[testnet]` as might be expected.
Thoughts? Ping @MeshCollider @laanwj @jonasschnelli @morcos
Tree-SHA512: f00b5eb75f006189987e5c15e154a42b66ee251777768c1e185d764279070fcb7c41947d8794092b912a03d985843c82e5189871416995436a6260520fb7a4db
4a6c0e3dcf Modernize best block mutex/cv/hash variable naming (Pieter Wuille)
45dd135039 Fix csBestBlock/cvBlockChange waiting in rpc/mining (Pieter Wuille)
Pull request description:
This is an alternative to #11694.
It reintroduces a uint256 variable with the best block hash, protected by csBestBlock, and only updated while holding it.
Also rename the involved variable to modern guidelines, as there are very few uses.
Tree-SHA512: 826a86c7d3cee7fe49f99f4398ae99e81cb0563197eaeba77306a3ca6072b67cdb932bc35720fc0f99c2a57b218efa029d0b8bdfb240591a629b2e90efa3199d
When network-specific options such as -addnode, -connect, etc are
specified in the default section of the config file, but that setting is
ignored due to testnet or regtest being in use, and it is not overridden
by either a command line option or a setting in the [regtest] or [test]
section of the config file, a warning is added to the log, eg:
Warning: Config setting for -connect only applied on regtest network when in [regtest] section.
d207207 [logging] add lint-logs.sh to check for newline termination. (John Newbery)
5c21e6c [logging] Comment all continuing logs. (John Newbery)
Pull request description:
Check that all calls to LogPrintf() are terminated by a newline,
except those that are explicitly marked as 'continued' logs.
Tree-SHA512: fe5162b2b2df1e8a4c807da87584fa9af97a6b8377e4090fe0caa136d90bf29a487a123cde94569bdce7101fee3478196d99aa13f1212e24bfe5f41c773604fc
d54874d Set SCHED_BATCH priority on the loadblk thread. (Evan Klitzke)
Pull request description:
Today I came across #10271, and while reading the discussion #6358 was linked to. Linux systems have a `SCHED_BATCH` scheduler priority that is useful for threads like loadblk. You can find the full details at [sched(7)](http://man7.org/linux/man-pages/man7/sched.7.html), but I'll quote the relevant part of the man page below:
> ...this policy will cause the scheduler to always assume that the thread is
CPU-intensive. Consequently, the scheduler will apply a small scheduling penalty
with respect to wakeup behavior, so that this thread is mildly disfavored in
scheduling decisions.
>
> This policy is useful for workloads that are noninteractive, but do not want to
lower their nice value, and for workloads that want a deterministic scheduling
policy without interactivity causing extra preemptions (between the workload's
tasks).
I think this change is useful independently of #10271 and irrespective of whether that change is merged. Under normal operation the loadblk thread will just import `mempool.dat`. However, if Bitcoin is started with `-reindex` or `-reindex-chainstate` this thread will use a great deal of CPU while it rebuilds the chainstate database (and the block database in the case of `-reindex`). By setting `SCHED_BATCH` this thread is less likely to interfere with interactive tasks (e.g. the user's web browser, text editor, etc.).
I'm leaving the nice value unchanged (which also affects scheduling decisions) because I think that's better set by the user. Likewise I'm not using [ioprio_set(2)](http://man7.org/linux/man-pages/man2/ioprio_set.2.html) because it can cause the thread to become completely I/O starved (and knowledgeable users can use `ionice(1)` anyway).
Tree-SHA512: ea8f7d3921ed5708948809da771345cdc33efd7ba3323e9dfec07a25bc21e8612e2676f9c178e2710c7bc437e8c9cafc5e0463613688fea5699b6e8e2fec6cff
Most logs should terminated with a '\n'. Some logs
are built up over multiple calls to logPrintf(), so
do not need a newline terminater. Comment all of
these 'continued' logs as a linter hing.
Log messages should terminate with a '\n', or the following log will be
written to the same line without a timestamp. Fix a couple of cases
where the message is not terminated with a \n.
0c17e27630 init: Remove help text for non-existent -fuzzmessagestest arg (MarcoFalke)
136084470c contrib: Fix check-doc script regexes (MarcoFalke)
Pull request description:
Fixup the regexes to properly find all used args. The regex should now match all of the getter and setter methods of the `ArgsManager`. See https://dev.visucore.com/bitcoin/doxygen/class_args_manager.html#pub-methods
Before:
```
Args used : 159
Args documented : 188
Args undocumented: 0
Args unknown : 29
```
After:
```
Args used : 183
Args documented : 188
Args undocumented: 0
Args unknown : 5
```
Tree-SHA512: 1a7fb7ea55b2f6030358a1055d8f2c19b31f69d0603be0b009e6e603564014b4e2bb824357c9d43d0fba3ce7159b7c4e7eaa60b3f962053d94f73d0e626294fc
cb1e319 Bugfix: RPC: savemempool: Don't save until LoadMempool() is finished (Jorge Timón)
Pull request description:
Fixes https://github.com/bitcoin/bitcoin/issues/12142
The tests are a little bit slow, mempool_persist.py goes from about 20 s to about 120 s in my hardware.
Perhaps there's a better way to test this.
Tree-SHA512: 9e6c24b32a9cf3774e8f0bd81c035b0deb53fba5ac3eb2532d85900579d21cef8a1135b75a4fa0a9d883e3822eb35e7d4b47a0838abf99789039205041962629
c7ec524 [wallet] Add dummy wallet init class (John Newbery)
49baa4a [wallet] Use global g_wallet_init_interface to init/destroy the wallet. (John Newbery)
caaf972 [wallet] Create wallet init interface. (John Newbery)
5fb5421 [wallet] Move wallet init functions into WalletInit class. (John Newbery)
Pull request description:
This continues the work of #7965. This PR, along with several others, would remove the remaining dependencies from libbitcoin_server.a on libbitcoin_wallet.a.
To create the interface, I've just translated all the old init.cpp wallet function calls into an interface class. I've not done any thinking about whether it makes sense to change that interface by combining/splitting those calls. This is a purely internal interface, so there's no problem in changing it later.
Tree-SHA512: 32ea57615229c33fd1a7f2f29ebc11bf30337685f7211baffa899823ef74b65dcbf068289c557a161c5afffb51fdc38a2ee8180720371f64d433b12b0615cf3f
a192636 -blocksdir: keep blockindex leveldb database in datadir (Jonas Schnelli)
f38e4fd QA: Add -blocksdir test (Jonas Schnelli)
386a6b6 Allow to optional specify the directory for the blocks storage (Jonas Schnelli)
Pull request description:
Since the actual block files taking up more and more space, it may be desirable to have them stored in a different location then the data directory (use case: SSD for chainstate, etc., HD for blocks).
This PR adds a `-blocksdir` option that allows one to keep the blockfiles and the blockindex external from the data directory (instead of creating symlinks).
I fist had an option to keep the blockindex within the datadir, but seems to make no sense since accessing the index will (always) lead to access (r/w) the block files.
Tree-SHA512: f8b9e1a681679eac25076dc30e45e6e12d4b2d9ac4be907cbea928a75af081dbcb0f1dd3e97169ab975f73d0bd15824c00c2a34638f3b284b39017171fce2409
This commit creates a global g_wallet_init_interface, which is created
in bitcoind and bitcoin-qt. g_wallet_init_interface is used to init
and destroy the wallet.
This removes the dependency from init.cpp on the wallet library.
4ae7d15 init: Fix help message for checkblockindex (MarcoFalke)
Pull request description:
Minor fixup for my commit fa6ab96799.
Tree-SHA512: 18f9255bf1342007be2bdc26d6f688bcd27ba8eebfc709bd9ee31dfd2e4d955d2b699686492ccf59e94eb4b1cc7bf3332376aa151a68cb0b21695b3f67d4a940
1f45e21 scripted-diff: Convert 11 enums into scoped enums (C++11) (practicalswift)
Pull request description:
Rationale (from Bjarne Stroustrup's ["C++11 FAQ"](http://www.stroustrup.com/C++11FAQ.html#enum)):
>
> The enum classes ("new enums", "strong enums") address three problems with traditional C++ enumerations:
>
> * conventional enums implicitly convert to int, causing errors when someone does not want an enumeration to act as an integer.
> * conventional enums export their enumerators to the surrounding scope, causing name clashes.
> * the underlying type of an enum cannot be specified, causing confusion, compatibility problems, and makes forward declaration impossible.
>
> The new enums are "enum class" because they combine aspects of traditional enumerations (names values) with aspects of classes (scoped members and absence of conversions).
Tree-SHA512: 9656e1cf4c3cabd4378c7a38d0c2eaf79e4a54d204a3c5762330840e55ee7e141e188a3efb2b4daf0ef3110bbaff80d8b9253abf2a9b015cdc4d60b49ac2b914
While reading another PR I saw a mention of #6358. The use case for
SCHED_BATCH is to hint to the kernel that the thread is running a
non-interactive workload that consumes a lot of CPU time. This is
helpful on desktop machines where the loadblk thread can interfere with
interactive applications. More details can be found in the sched(7) man
page.
4757c04 [config] Remove blockmaxsize option (John Newbery)
Pull request description:
The blockmaxsize option was marked as deprecated in V0.15.1, and code
was added to convert provided blockmaxsize into blockmaxweight. However,
this code was incorrectly implemented, and blockmaxsize was silently
ignored.
No users have complained about blockmaxsize being ignored, so just
remove it in V0.17.
Fixes#12640
cc @ajtowns
Tree-SHA512: 968d71d37bf175c5a02539ddec289a12586f886e1dfe64c1d9aa5e39db48d06d21665153824fac3b11503a55f0812d2f1115a2d726aafd37b76ed629ec0aa671
The blockmaxsize option was marked as deprecated in V0.15.1, and code
was added to convert provided blockmaxsize into blockmaxweight. However,
this code was incorrectly implemented, and blockmaxsize was silently
ignored.
No users have complained about blockmaxsize being ignored, so just
remove it in V0.17.
8674e74 Provide relevant error message if datadir is not writable. (murrayn)
Pull request description:
If the --datadir exists, but is not writable, the current error message on startup is 'Cannot obtain a lock on data directory foo. Bitcoin Core is probably already running.' This is misleading.
I believe this PR addresses #11668, although the issue is not Windows-specific.
Tree-SHA512: 10cbbaea433072aee4fb3e8938a72073c7a5c841f7a7685c9e12549c322b2925c7d34bac254ac33021b23132bfc352c058712bc9542298cf86f8fd9757f528b2
92fabcd44 Add LookupBlockIndex function (João Barbosa)
43a32b739 Add missing cs_lock in CreateWalletFromFile (João Barbosa)
f814a3e8f Fix cs_main lock in LoadExternalBlockFile (João Barbosa)
c651df8b3 Lock cs_main while loading block index in AppInitMain (João Barbosa)
02de6a6bc Assert cs_main is held when accessing mapBlockIndex (João Barbosa)
Pull request description:
Replace all `mapBlockIndex` lookups with the new `LookupBlockIndex()`. In some cases it avoids a second lookup.
Tree-SHA512: ca31118f028a19721f2191d86f2dd398144d04df345694575a64aeb293be2f85785201480c3c578a0ec99690516205708558c0fd4168b09313378fd4e60a8412
* Z is the zone designator for the zero UTC offset.
* T is the delimiter used to separate date and time.
This makes it clear for the end-user that the date/time logged is
specified in UTC and not in the local time zone.
874e81808 Allow dustrelayfee to be set to zero (Luke Dashjr)
Pull request description:
I don't see and can't think of any rationale for forbidding this configuration.
Tree-SHA512: df09441f4aec63e79bea94838b7f8e336cebaeb0a22b5e58d27937bbeb1377f229921aeae43674e0b63fc40a39ae51a264d48aa1cdb4cbd0e3339d32856698bf
eb91835 Add setter for g_initial_block_download_completed (Jonas Schnelli)
3f56df5 [QA] add NODE_NETWORK_LIMITED address relay and sync test (Jonas Schnelli)
158e1a6 [QA] fix mininode CAddress ser/deser (Jonas Schnelli)
fa999af [QA] Allow addrman loopback tests (add debug option -addrmantest) (Jonas Schnelli)
6fe57bd Connect to peers signaling NODE_NETWORK_LIMITED when out-of-IBD (Jonas Schnelli)
31c45a9 Accept addresses with NODE_NETWORK_LIMITED flag (Jonas Schnelli)
Pull request description:
Eventually connect to peers signalling NODE_NETWORK_LIMITED if we are out of IBD.
Accept and relay NODE_NETWORK_LIMITED peers in addrman.
Tree-SHA512: 8a238fc97f767f81cae1866d6cc061390f23a72af4a711d2f7158c77f876017986abb371d213d1c84019eef7be4ca951e8e6f83fda36769c4e1a1d763f787037
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).
004f999 boost: drop boost threads for [alert|block|wallet]notify (Cory Fields)
0827267 boost: drop boost threads from torcontrol (Cory Fields)
ba91724 boost: remove useless threadGroup parameter from Discover (Cory Fields)
f26866b boost: drop boost threads for upnp (Cory Fields)
Pull request description:
This doesn't completely get rid of boost::thread, but this batch should be easy to review, and leaves us with only threadGroup (scheduler + scriptcheck) remaining.
Note to reviewers: The upnp diff changes a bunch of whitespace, it's much more clear with 'git diff -w'
Tree-SHA512: 5a356798d0785f93ed143d1f0afafe890bc82f0d470bc969473da2d2aa78bcb9b096f7ba11b92564d546fb447d4bd0d347e7842994ea0170aafd53fda7e0a66e
669c943 Avoid leaking prioritization information when relaying transactions (Suhas Daftuar)
e868b22 fee estimator: avoid sorting mempool on shutdown (Suhas Daftuar)
0975406 Correct mempool mapTx comment (Suhas Daftuar)
Pull request description:
Following up on #12127 and #12118, this cleans up a comment that was left incorrect in txmempool.h, and addresses a couple of the observations @TheBlueMatt made about an unnecessary use of `queryHashes()` and a small information leak when prioritizing transactions.
Left undone is nuking queryHashes altogether; that would require changing the behavior of the `getrawmempool` rpc call, which I think I might be in favor of doing, but wanted to save for its own PR.
Tree-SHA512: c97d10b96dcd6520459287a4a2eda92774173757695100fcfe61e526aef86f394507c331d17f9e0c14b496c33ec46198a0f165a847762ca50f7c6780b993f162
If the user somehow manages to get into ShutdownRequested before
ThreadImport gets to ActivateBestChain() we may hang waiting on
condvar_GenesisWait forever. A simple wait_for and
ShutdownRequested resolves this case.
Change `-conf`'s and others' help messages to indicate that relative path
values will be prefixed by the datadir path. This behavior is confusing when
attempting to specify a configuration file in the current directory with
`-conf=bitcoin.conf`, but loading the `bitcoin.conf` file in ~/.bitcoin
datadir.
082a61c Move scheduler/threadGroup into common-init instead of per-app (Matt Corallo)
Pull request description:
This resolves#12229 which pointed out a shutdown deadlock due to
scheduler/checkqueue having been shut down while network message
processing is still running.
Tree-SHA512: 0c0a76113996b164b0610d3b8c40b396f3e384d165bf098768e31fe3701b00763d0d810ef24702387e2e936fefb9fb900a6225f7417bb0175b585f365d542660
34328b4 Use PACKAGE_NAME instead of hardcoding application name in log message (Wladimir J. van der Laan)
0c74e2e Log debug build status and warn when running benchmarks (Wladimir J. van der Laan)
Pull request description:
Log whether the starting instance of bitcoin core is a debug or release build (--enable-debug).
Also warn when running the benchmarks with a debug build, to prevent mistakes comparing debug to non-debug results.
Tree-SHA512: f612dcb7d0a8435016cff0df8aef4942144dfb88be8a00df45cc8830d2aba4b167f6d397b83f8f57d57685888babd04ba88d4dac5a202d3dbd91bcbea3708ef0
This resolves#12229 which pointed out a shutdown deadlock due to
scheduler/checkqueue having been shut down while network message
processing is still running.
Also warn if bitcoind is configured to use a relative -datadir path.
Specifying paths relative to the current working directory in a daemon process
can be dangerous, because files can fail to be located even if the
configuration doesn't change, but the daemon is started up differently.
Specifying a relative -datadir now adds a warning to the debug log. It would
not be backwards-compatible to forbid relative -datadir paths entirely, and it
could also be also inconvenient for command line testing.
Specifying a relative -walletdir now results in a startup error. But since the
-walletdir option is new in 0.16.0, there should be no compatibility issues.
Another reason not to use working directory paths for -walletdir specifically
is that the default -walletdir is a "wallets" subdirectory inside the datadir,
so it could be surprising that setting -walletdir manually would choose a
directory rooted in a completely different location.
Log whether the starting instance of bitcoin core is a debug or release
build (--enable-debug).
Also warn when running the benchmarks with a debug build, to prevent
mistakes comparing debug to non-debug results.
07c4838 Always return true if AppInitMain got to the end (Matt Corallo)
Pull request description:
This should fix a rare zapwallettxes failure on travis, but also
avoids having init operations (re-adding wallet transactions to
mempool) running after RPC is free'd.
I believe this was the failure at https://travis-ci.org/bitcoin/bitcoin/jobs/311747844 (from #11605).
Tree-SHA512: f0fea8c1b9265e2eeda57043d541380a3e58e4d9388fa24628a52fd56324257fcd7df0ca02e8f77f66fadd68d951893bab0f610ed9fd0a89b2ccd6bad1efa351
This should fix a very rare travis failure in zapwallettxes, but
is also more correct, as you can currently race
ReacceptWalletTransactions with stop RPC calls to get bitcoind to
(IMO) eroneously return a non-0 exit code.
This patch adds an option to configure the name and/or directory of the
debug log.
The user can specify either a relative path, in which case the path
is relative to the data directory. They can also specify an absolute
path to put the log anywhere else in the file system.
d31e5c1 Fix shutdown in case of errors during initialization (Wladimir J. van der Laan)
Pull request description:
PR #10286 introduced a few steps which are not robust to early shutdown in initialization.
Stumbled upon this with #11781, not sure if there are other scenarios that can trigger it, but it's good to harden against this in any case.
E.g.
```
$ src/bitcoind -debuglogfile=/dfdf
Error: Could not open debug log file /dfdf
Program received signal SIGSEGV, Segmentation fault.
UnregisterValidationInterface (pwalletIn=0x0) at /.../bitcoin/src/validationinterface.cpp:82
82 g_signals.m_internals->BlockChecked.disconnect(boost::bind(&CValidationInterface::BlockChecked, pwalletIn, _1, _2));
(gdb) bt
#0 UnregisterValidationInterface (pwalletIn=0x0) at /.../bitcoin/src/validationinterface.cpp:82
#1 0x00005555555a11fc in Shutdown () at /.../bitcoin/src/init.cpp:196
#2 0x00005555555961cc in AppInit (argc=<optimized out>, argv=<optimized out>) at /.../bitcoin/src/bitcoind.cpp:183
#3 0x0000555555596249 in main (argc=0, argv=0x555555ecf200) at /.../bitcoin/src/bitcoind.cpp:19
```
Tree-SHA512: 7dd9570a9803514a17781bfadf1edde47e96df4e852cce2f423cab422e005fb94d44e777af1a6ea5167b04a4d889e848ae7a61a7e0e94232247ddea32ee70fc8
c60c49b Improve help text and behavior of RPC-logging (Akio Nakamura)
Pull request description:
1. It is allowed `libevent` logging to be updated during runtime,
but still described that restriction in the help text.
So we delete these text.
2. Add a descrption about the evaluation order of `<include>` and
`<exclude>` to clarify how debug loggig categories to be set.
3. Add a description about the available logging category `"all"`
which is not explained.
4. Add `"optional"` to the help text of `<include>` and `<exclude>`.
5. Add missing new lines before `"Argument:"`.
6. `"0"`,`"1"` are allowed in both array of `<include>` and `<exclude>`.
`"0"` is **ignored** and `"1"` is treated **same as** `"all"`.
It is confusing, so forbid them.
7. It always returns all logging categories with status.
Fix the help text to match this behavior.
Tree-SHA512: c2142da1a9bf714af8ebc38ac0d82394e2073fc0bd56f136372e3db7b2af3b6746f8d6b0241fe66c1698c208c124deb076be83f07dec0d0a180ad150593af415
PR #10286 introduced a few steps which are not robust to early shutdown
in initialization.
Stumbled upon this with #11781, not sure if there are other scenarios
that can trigger it, but it's harden against this in any case.
abbd230 Move RPC registration out of AppInitParameterInteraction (Russell Yanofsky)
Pull request description:
Move to AppInitServers. This doesn't have any effects on bitcoin behavior. It was just strange to have this unrelated code in the middle of parameter interaction.
Tree-SHA512: 373e18f2ef8d21999ad36295d69326128a3086044acfc8ed537abd5497c8d3620b9832f7f6aa87c0c0939bb5e0d92be8a3c006b5997e9e6fa20334f5610c89bc
A) The changes in behavior are as follows:
1. Introduce logging category "none" as alias of "0" for
both RPC-logging and bitcoind "-debug" parameter.
2. Same as "0" is given to argument of "-debug",
if "none" or "0" is given to <include>, all other given logging
categories are ignored. The same is true for <exclude>.
(Before this PR, "0" was accepted but just be ignored itself.)
B) The changes in the help text are as follows:
1. Add a descrption about the evaluation order of <include> and
<exclude> to clarify how debug loggig categories to be set.
2. Delete text that describe restriction about libevent because
it's already allowed libevent logging to be updated during runtime.
3. Add a description for category "all", "1", "none" and "0".
4. Add "optional" to the help text of <include> and <exclude>.
5. Add missing new lines before "Argument:".
6. This RPC always returns all logging categories with status.
Fix the help text to match this behavior.
89f0312 Remove redundant pwallet nullptr check (Matt Corallo)
c4784b5 Add a dev notes document describing the new wallet RPC blocking (Matt Corallo)
3ea8b75 Give ZMQ consistent order with UpdatedBlockTip on scheduler thread (Matt Corallo)
cb06edf Fix wallet RPC race by waiting for callbacks in sendrawtransaction (Matt Corallo)
e545ded Also call other wallet notify callbacks in scheduler thread (Matt Corallo)
17220d6 Use callbacks to cache whether wallet transactions are in mempool (Matt Corallo)
5d67a78 Add calls to CWallet::BlockUntilSyncedToCurrentChain() in RPCs (Matt Corallo)
5ee3172 Add CWallet::BlockUntilSyncedToCurrentChain() (Matt Corallo)
0b2f42d Add CallFunctionInQueue to wait on validation interface queue drain (Matt Corallo)
2b4b345 Add ability to assert a lock is not held in DEBUG_LOCKORDER (Matt Corallo)
0343676 Call TransactionRemovedFromMempool in the CScheduler thread (Matt Corallo)
a7d3936 Add a CValidationInterface::TransactionRemovedFromMempool (Matt Corallo)
Pull request description:
Based on #10179, this effectively reverts #9583, regaining most of the original speedups of #7946.
This concludes the work of #9725, #10178, and #10179.
See individual commit messages for more information.
Tree-SHA512: eead4809b0a75d1fb33b0765174ff52c972e45040635e38cf3686cef310859c1e6b3c00e7186cbd17374c6ae547bfbd6c1718fe36f26c76ba8a8b052d6ed7bc9
76ea17c79 Add mutex requirement for AddToCompactExtraTransactions(…) (practicalswift)
4616c825a Use -Wthread-safety-analysis if available (+ -Werror=thread-safety-analysis if --enable-werror) (practicalswift)
7e319d639 Fix -Wthread-safety-analysis warnings. Change the sync.h primitives to std from boost. (Matt Corallo)
Pull request description:
* Add mutex requirement for `AddToCompactExtraTransactions(…)`.
* Use `-Wthread-safety-analysis` if available.
* Rebased on top of https://github.com/TheBlueMatt/bitcoin/commits/2017-08-test-10923 - now includes: Fix -Wthread-safety-analysis warnings. Change the sync.h primitives to std from boost.
Tree-SHA512: fb7365f85daa2741c276a1c899228181a8d46af51db7fbbdffceeaff121a3eb2ab74d7c8bf5e7de879bcc5042d00d24cb4649c312d51caba45a3f6135fd8b38f
Commit 1.
This code was written by @TheBlueMatt in the following branch:
* https://github.com/TheBlueMatt/bitcoin/commits/2017-08-test-10923
This commit message was written by me (@practicalswift) who also squashed
@TheBlueMatt's commits into one and tried to summarize the changes made.
Commit 2.
Remove boost include. Remove boost mentions in comments.
Move to AppInitServers. This doesn't have any effects on bitcoin behavior. It
was just strange to have this unrelated code in the middle or parameter
interaction.
If our tip hasn't updated in a while, that may be because our peers are
not relaying blocks to us that we would consider valid. Allow connection
to an additional outbound peer in that circumstance.
Also, periodically check to see if we are exceeding our target number of
outbound peers, and disconnect the one which has least recently
announced a new block to us (choosing the newest such peer in the case
of tie).
f4c4e38 [trivial] Make namespace explicit for is_regular_file (John Newbery)
Pull request description:
is_regular_file resolves using argument dependent lookup. Make the
namespace explicit so it's obvious where the function is defined.
For those not familiar with argument dependent lookups:
- http://en.cppreference.com/w/cpp/language/adl
- https://en.wikipedia.org/wiki/Argument-dependent_name_lookup
Thanks to C++ guru @ryanofsky for pointing this out to me.
Tree-SHA512: 919f1818081a8f90c5751181f87e13b06d90f8aec0ab873100434e55c85cca6e0e288ecc7f135e19e9b5dba7952e96b6393864b7840e20b69dd40e92a157928b
This is currently unused, but will by used by wallet to cache when
transactions are in the mempool, obviating the need for calls to
mempool from CWalletTx::InMempool()
15f5d3b17 Switch DNSSeed-needed metric to any-automatic-nodes, not services (Matt Corallo)
5ee88b4bd Clarify docs for requirements/handling of addnode/connect nodes (Matt Corallo)
57edc0b0c Rename fAddnode to a more-descriptive "manual_connection" (Matt Corallo)
44407100f Replace relevant services logic with a function suite. (Matt Corallo)
Pull request description:
This was mostly written as a way to clean things up so that the NETWORK_LIMITED PR (#10387) can be simplified a ton, but its also a nice standalone cleanup that will also require a bit of review because it tweaks a lot of stuff across net. The new functions are fine in protocol.h right now since they're straight-forward, but after NETWORK_LIMITED will really want to move elsewhere after @theuni moves the nServices-based selection to addrman from connman.
Adds HasAllRelevantServices and GetRelevantServices, which check
for NETWORK|WITNESS.
This changes the following:
* Removes nRelevantServices from CConnman, disconnecting it a bit
more from protocol-level logic.
* Replaces our sometimes-connect-to-!WITNESS-nodes logic with
simply always requiring WITNESS|NETWORK for outbound non-feeler
connections (feelers still only require NETWORK).
* This has the added benefit of removing nServicesExpected from
CNode - instead letting net_processing's VERSION message
handling simply check HasAllRelevantServices.
* This implies we believe WITNESS nodes to continue to be a
significant majority of nodes on the network, but also because
we cannot sync properly from !WITNESS nodes, it is strange to
continue using our valuable outbound slots on them.
* In order to prevent this change from preventing connection to
-connect= nodes which have !WITNESS, -connect nodes are now
given the "addnode" flag. This also allows outbound connections
to !NODE_NETWORK nodes for -connect nodes (which was already true
of addnodes).
* Has the (somewhat unintended) consequence of changing one of the
eviction metrics from the same
sometimes-connect-to-!WITNESS-nodes metric to requiring
HasRelevantServices.
This should make NODE_NETWORK_LIMITED much simpler to implement.
Tree-SHA512: 90606896c86cc5da14c77843b16674a6a012065e7b583d76d1c47a18215358abefcbab44ff4fab3fadcd39aa9a42d4740c6dc8874a58033bdfc8ad3fb5c649fc
Adds HasAllRelevantServices and GetRelevantServices, which check
for NETWORK|WITNESS.
This changes the following:
* Removes nRelevantServices from CConnman, disconnecting it a bit
more from protocol-level logic.
* Replaces our sometimes-connect-to-!WITNESS-nodes logic with
simply always requiring WITNESS|NETWORK for outbound non-feeler
connections (feelers still only require NETWORK).
* This has the added benefit of removing nServicesExpected from
CNode - instead letting net_processing's VERSION message
handling simply check HasAllRelevantServices.
* This implies we believe WITNESS nodes to continue to be a
significant majority of nodes on the network, but also because
we cannot sync properly from !WITNESS nodes, it is strange to
continue using our valuable outbound slots on them.
* In order to prevent this change from preventing connection to
-connect= nodes which have !WITNESS, -connect nodes are now
given the "addnode" flag. This also allows outbound connections
to !NODE_NETWORK nodes for -connect nodes (which was already true
of addnodes).
* Has the (somewhat unintended) consequence of changing one of the
eviction metrics from the same
sometimes-connect-to-!WITNESS-nodes metric to requiring
HasRelevantServices.
This should make NODE_NETWORK_LIMITED much simpler to implement.
c626dcb50 Make fUseCrypto atomic (MeshCollider)
731065b11 Consistent parameter names in txdb.h (MeshCollider)
35aeabec6 Make fReindex atomic to avoid race (MeshCollider)
58d91af59 Fix race for mapBlockIndex in AppInitMain (MeshCollider)
Pull request description:
Fixes https://github.com/bitcoin/bitcoin/issues/11106
Also makes fReindex atomic as suggested in @TheBlueMatt comment below, and makes fUseCrypto atomic as suggested in 10916
d291e7635b just renames the parameters in the txdb header file to make them consistent with those used in the cpp file, noticed it when looking for uses of fReindex
Tree-SHA512: b378aa7289fd505b76565cd4d48dcdc04ac5540283ea1c80442170b0f13cb6df771b1a94dd54b7fec3478a7b4668c224ec9d795f16937782724c5d020edd3a42
cffe85f Skip sys::system(...) call in case of empty command (practicalswift)
6fb8f5f Check that -blocknotify command is non-empty before executing (practicalswift)
Pull request description:
Check that `-blocknotify` command is non-empty before executing.
To make the `BlockNotifyCallback(...)` (`-blocknotify`) behaviour consistent with that of:
* `AlertNotify(...)` (`-alertnotify`)
* `AddToWallet(...)` (`-walletnotify`)
Tree-SHA512: 18272166793a5a8b9cc2a727bfbcea53d38c329a55bc975c02db601329d608a61c20e026ce4b616193ecd3810dca4d3e2cb3bf773898a51872008a8dba96763e
048e0c3e2 [rpc] [tests] Add deprecated RPC test (Cristian Mircea Messel)
d4cdbd6fb [rpc] Deprecate estimatefee RPC (John Newbery)
Pull request description:
Deprecates estimatefee in v0.16, for final removal in v0.17.
This commit introduces a phased removal of RPC methods. RPC method is
disabled by default in version x, but can be enabled by using the
`-deprecatedrpc=<methodname>` argument. RPC method is removed entirely in version
(x+1).
This gives users fair warning that an RPC is to be removed, and time to change client software if necessary. Deprecation warnings in RPC return values or release notes are easily ignored.
This is a more generic version of the approach I tried to use in #10841, which too late to make it into v0.15.
Tree-SHA512: 9695a600e84b812974387333e4a6805d18972da30befb754e9e4da77cd9815d00c5cc2ee0b0350bdbbdb5fdc6ba47789f8b2c6f5b15c8cd5a1deefcc4832da30
Deprecate estimatefee in v0.16, for final removal in v0.17.
This commit introduces a phased removal of RPC methods. RPC method is
disabled by default in version x, but can be enabled by using the
`-deprecatedrpc=<method>` argument. RPC method is removed entirely in
version (x+1).
* This removes block-size-limiting code in favor of GBT clients
doing the limiting themselves (if at all).
* -blockmaxsize is deprecated and only used to calculate an implied
blockmaxweight, addressing confusion from multiple users.
* getmininginfo's currentblocksize return value was returning
garbage values, and has been removed, also removing a
GetSerializeSize call in some block generation inner loops and
potentially addressing some performance edge cases.
5d2a3995e [trivial] fixup comment for VerifyWallets() (John Newbery)
43b0e81d0 [wallet] Add StartWallets() function to wallet/init.cpp (John Newbery)
290f3c56d [wallet] Add RegisterWalletRPC() function to wallet/init.cpp (John Newbery)
062d63102 [wallet] Add CloseWallets() function to wallet/init.cpp (John Newbery)
77fe07c15 [wallet] Add StopWallets() function to wallet/init.cpp (John Newbery)
2da5eafa4 [wallet] Add FlushWallets() function to wallet/init.cpp (John Newbery)
1b9cee66e [wallet] Rename WalletVerify() to VerifyWallets() (John Newbery)
9c76ba18c [wallet] Rename InitLoadWallet() to OpenWallets() (John Newbery)
Pull request description:
Apologies for the mostly code move only PR. This is a pre-req for both #10740 and #10762
All wallet component initialization/destruction functions are now in their own `wallet/init.cpp` translation unit and are no longer static functions on the CWallet class. The bitcoin_server also no longer has any knowledge that there are multiple wallets in vpwallet.
There should be no changes in behavior from this PR.
Tree-SHA512: 7c260eb094f2fa1a88d803769ba60935810968a7309f731135e4b17623b97f18c03bbcd293c942093d1efce62c6c978f9ff484d54dc9a60bc2fcb5af2d160fcd
Rationale:
- this init function can now open multiple wallets (hence
Wallet->Wallets)
- This is named as the antonym to CloseWallets(), which carries out the
opposite action.
eac64bb7a [qa] Test nMinimumChainWork (Suhas Daftuar)
0311836f6 Allow setting nMinimumChainWork on command line (Suhas Daftuar)
Pull request description:
As discussed briefly here: https://botbot.me/freenode/bitcoin-core-dev/2017-02-28/?msg=81712308&page=4
This adds a hidden command line option for setting `nMinimumChainWork`, which allows us to test this parameter in our functional tests, as well as allowing for niche use cases like syncing nodes that are otherwise disconnected from the network.
See also #10345, which proposes a new use of `nMinimumChainWork`.
Tree-SHA512: fe4d8f4f289697615c98d8760f1cc74c076110310ea0b5b875fcab78c127da9195b4eb84148aebacc7606c246e5773d3f13bd5d9559d0a8bffac20a3a28c62df
352d582ba Add vConnect to CConnman::Options (Marko Bencun)
Pull request description:
Split the "-connect" argument parsing out of CConnman and put it into
AppInitMain().
Tree-SHA512: f2d3efc4e2c5808ff98696ea20dd96df599bc472ed5afc9c3eea305d94c36a6ab50c632aa05396c7c34d1917d91b1e7ccd725656ff2631e2a36d9eac477455dc
This contains most of the changes of 10563 "remove safe mode", but doesn't
remove the safe mode yet, but put an `ObserveSafeMode()` check in
individual calls with okSafeMode=false.
This cleans up the ugly "okSafeMode" flag from the dispatch tables,
which is not a concern for the RPC server.
Extra-author: Wladimir J. van der Laan <laanwj@gmail.com>
f01103c MOVEONLY: Init functions wallet/wallet.cpp -> wallet/init.cpp (Russell Yanofsky)
e7fe320 MOVEONLY: Fee functions wallet/wallet.cpp -> wallet/fees.cpp (Russell Yanofsky)
d97fe20 Move some static functions out of wallet.h/cpp (Russell Yanofsky)
Pull request description:
This just moves some static wallet fee and init functions out of `wallet/wallet.cpp` and into new `wallet/fees.cpp` and `wallet/init.cpp` source files. There is one commit updating declarations and callers, followed by two MOVEONLY commits actually moving the function bodies.
This change is desirable because wallet.h/cpp are monolithic and hard to navigate, so pulling things out and grouping together pieces of related functionality should improve the organization.
Another motivation is the wallet process separation work in https://github.com/bitcoin/bitcoin/pull/10973, where (at least initially) parameter parsing and fee estimation are still done in the main process rather than the wallet process, and having functions that run in different processes scrambled up throughout wallet.cpp is unnecessarily confusing.
Tree-SHA512: 6e6982ff82b2ab4e681c043907e2b1801ceb9513394730070f16c46ad338278a863f5b3759aa13db76a259b268b1c919c81f4e339f0796a3cfb990161e8c316d
cd0ea4874 Changing -txindex requires -reindex, not -reindex-chainstate (Matt Corallo)
Pull request description:
If there's an 0.15.0rc3, this should go in it.
Tree-SHA512: 857e77f0af9c055a3d1d91f37474ee9e06d6bc8c5ed21b29201b6c386801e7041523949076cdf0daa4d357a5175ce49394d85a1bedfbf13f3e577bdb6da1d6ce
4aa2508 Bugfix: Use testnet RequireStandard for -acceptnonstdtxn default (Luke Dashjr)
Pull request description:
Fixes a bug introduced in #8855
`-acceptnonstdtxn` is a valid option only for testnet/regtest (in Core), and the help message reflects that. Currently, however, it is buggy in two ways:
1. It uses mainnet to get the default value, which doesn't make sense since the option is never available for mainnet, and the only time the option is available, is when the default is the opposite.
2. It uses the value of "require standard" directly as the default for "accept non-standard transactions", but these concepts are opposites: a negation must be performed to transform one to the other.
Note the combination of these bugs results in the correct boolean output, but the logic to get there is completely wrong.
Tree-SHA512: 06ce513f59ba31f7ab4b6422a08a17bb37a5652dea4c38a4bbefedd5e2752d17bfccc32a4b0508068fa4783e316bff00a821ef18a24b1a2bb02859995d188fdc
40a0f9f Enable devirtualization opportunities by using the final specifier (C++11) (practicalswift)
9a1675e optim: mark a few classes final (Cory Fields)
Pull request description:
Using gcc's ```-Wsuggest-final-types``` and lto, I identified a few easy devirtualization wins:
> wallet/wallet.h:651:7: warning: Declaring type 'struct CWallet' final would enable devirtualization of 26 calls [-Wsuggest-final-types]
>coins.h:201:7: warning: Declaring type 'struct CCoinsViewCache' final would enable devirtualization of 13 calls [-Wsuggest-final-types]
>txdb.h:67:7: warning: Declaring type 'struct CCoinsViewDB' final would enable devirtualization of 5 calls [-Wsuggest-final-types]
>zmq/zmqnotificationinterface.h:16:7: warning: Declaring type 'struct CZMQNotificationInterface' final would enable devirtualization of 4 calls [-Wsuggest-final-types]
>httpserver.cpp:42:7: warning: Declaring type 'struct HTTPWorkItem' final would enable devirtualization of 2 calls [-Wsuggest-final-types]
Tree-SHA512: 2a825fd27121ccabaacff5cde2fc8a50d1b4cc846374606caa2a71b0cd8fcb0d3c9b5b3fd342d944998610e2168048601278f8a3709cc515191a0bb2d98ba782
f4c3d2c Enable disablesafemode by default. (Gregory Maxwell)
Pull request description:
Safemode is almost useless as is-- it only triggers in limited
cases most of which aren't even concerning. There have been
several proposals to remove it. But as a simpler, safer, and
more flexible first case, simply deactivate it by default.
Anyone who wants it can re-enable and know what they've signed up for.
Tree-SHA512: f5409a3e81514c32db8eb27c7563ef85e25e56e5fc2a59eac2c30b10ec54087d982c1d3b702bedf9f3133c1f272f23805582a0f468350ba18d8b5a02bedd6401
This commit just moves a few function declarations and updates callers.
Function bodies are moved in two followup MOVEONLY commits.
This change is desirable because wallet.h/cpp are monolithic and hard to
navigate, so pulling things out and grouping together pieces of related
functionality should improve the organization.
Another proximate motivation is the wallet process separation work in
https://github.com/bitcoin/bitcoin/pull/10973, where (at least initially)
parameter parsing and fee estimation are still done in the main process rather
than the wallet process, and having functions that run in different processes
scrambled up throughout wallet.cpp is unnecessarily confusing.
90d4d89 scripted-diff: Use the C++11 keyword nullptr to denote the pointer literal instead of the macro NULL (practicalswift)
Pull request description:
Since C++11 the macro `NULL` may be:
* an integer literal with value zero, or
* a prvalue of type `std::nullptr_t`
By using the C++11 keyword `nullptr` we are guaranteed a prvalue of type `std::nullptr_t`.
For a more thorough discussion, see "A name for the null pointer: nullptr" (Sutter &
Stroustrup), http://www.open-std.org/jtc1/sc22/wg21/docs/papers/2007/n2431.pdf
With this patch applied there are no `NULL` macro usages left in the repo:
```
$ git grep NULL -- "*.cpp" "*.h" | egrep -v '(/univalue/|/secp256k1/|/leveldb/|_NULL|NULLDUMMY|torcontrol.*NULL|NULL cert)' | wc -l
0
```
The road towards `nullptr` (C++11) is split into two PRs:
* `NULL` → `nullptr` is handled in PR #10483 (scripted, this PR)
* `0` → `nullptr` is handled in PR #10645 (manual)
Tree-SHA512: 3c395d66f2ad724a8e6fed74b93634de8bfc0c0eafac94e64e5194c939499fefd6e68f047de3083ad0b4eff37df9a8a3a76349aa17d55eabbd8e0412f140a297
Safemode is almost useless as is-- it only triggers in limited
cases most of which aren't even concerning. There have been
several proposals to remove it. But as a simpler, safer, and
more flexible first case, simply deactivate it by default.
Anyone who wants it can re-enable and know what they've signed up for.
This fixes a few cases where we should be treating a restart-after-
coinsviewdb-reset identically to a just-reset-coinsviewdb.
Thanks to @morcos for identifying the bug.
This more clearly uses fReindex vs fReset to make sure we're not
clearing our coinsdb needlessly when restarting after a reindex.
It also makes it so that restarting after shutting down mid-reindex
isn't treates specially at all during txdb loading code, as it
shouldn't be.
c0025d0 Fix segfault when shutting down before fully loading (Matt Corallo)
1385697 Order chainstate init more logically. (Matt Corallo)
ff3a219 Call RewindBlockIndex even if we're about to run -reindex-chainstate (Matt Corallo)
b0f3249 More user-friendly error message if UTXO DB runs ahead of block DB (Matt Corallo)
eda888e Fix some LoadChainTip-related init-order bugs. (Matt Corallo)
Pull request description:
This does a number of things to clean up chainstate init order,
fixing some issues as it goes:
* Order chainstate init more logically - first all of the
blocktree-related loading, then coinsdb, then
pcoinsTip/chainActive. Only create objects as needed.
* More clearly document exactly what is and isn't called in
-reindex and -reindex-chainstate both with comments noting
calls as no-ops and by adding if guards.
* Move the writing of fTxIndex to LoadBlockIndex - this fixes a
bug introduced in d6af06d68a where
InitBlockIndex was writing to fTxIndex which had not yet been
checked (because LoadChainTip hadn't yet initialized the
chainActive, which would otherwise have resulted in
InitBlockIndex being a NOP), allowing you to modify -txindex
without reindex, potentially corrupting your chainstate!
* Rename InitBlockIndex to LoadGenesisBlock, which is now a more
natural name for it. Also check mapBlockIndex instead of
chainActive, fixing a bug where we'd write the genesis block out
on every start.
* Move LoadGenesisBlock further down in init. This is a more logical
location for it, as it is after all of the blockindex-related
loading and checking, but before any of the UTXO-related loading
and checking.
* Give LoadChainTip a return value - allowing it to indicate that
the UTXO DB ran ahead of the block DB. This just provides a nicer
error message instead of the previous mysterious
assert(!setBlockIndexCandidates.empty()) error.
* Calls ActivateBestChain in case we just loaded the genesis
block in LoadChainTip, avoiding relying on the ActivateBestChain
in ThreadImport before continuing init process.
* Move all of the VerifyDB()-related stuff into a -reindex +
-reindex-chainstate if guard. It couldn't do anything useful
as chainActive.Tip() would be null at this point anyway.
Tree-SHA512: 3c96ee7ed44f4130bee3479a40c5cd99a619fda5e309c26d60b54feab9f6ec60fabab8cf47a049c9cf15e88999b2edb7f16cbe6819e97273560b201a89d90762
* Order chainstate init more logically - first all of the
blocktree-related loading, then coinsdb, then
pcoinsTip/chainActive. Only create objects as needed.
* More clearly document exactly what is and isn't called in
-reindex and -reindex-chainstate both with comments noting
calls as no-ops and by adding if guards.
* Move LoadGenesisBlock further down in init. This is a more logical
location for it, as it is after all of the blockindex-related
loading and checking, but before any of the UTXO-related loading
and checking.
* Move all of the VerifyDB()-related stuff into a -reindex +
-reindex-chainstate if guard. It couldn't do anything useful
as chainActive.Tip() would be null at this point anyway.
RewindBlockIndex works over both chainActive - disconnecting blocks
from the tip that need witness verification - and mapBlockIndex -
requiring redownload of blocks missing witness data.
It should never have been the case that the second half is skipped
if we're about to run -reindex-chainstate.
This gives LoadChainTip a return value - allowing it to indicate that
the UTXO DB ran ahead of the block DB. This just provides a nicer
error message instead of the previous mysterious
assert(!setBlockIndexCandidates.empty()) error.
This also calls ActivateBestChain in case we just loaded the genesis
block in LoadChainTip, avoiding relying on the ActivateBestChain
in ThreadImport before continuing init process.
* Move the writing of fTxIndex to LoadBlockIndex - this fixes a
bug introduced in d6af06d68a where
InitBlockIndex was writing to fTxIndex which had not yet been
checked (because LoadChainTip hadn't yet initialized the
chainActive, which would otherwise have resulted in
InitBlockIndex being a NOP), allowing you to modify -txindex
without reindex, potentially corrupting your chainstate!
* Rename InitBlockIndex to LoadGenesisBlock, which is now a more
natural name for it. Also check mapBlockIndex instead of
chainActive, fixing a bug where we'd write the genesis block out
on every start.
6b8d872 Protect SSE4 code behind a compile-time flag (Pieter Wuille)
fa9be90 Add selftest for SHA256 transform (Pieter Wuille)
c1ccb15 Add SSE4 based SHA256 (Pieter Wuille)
2991c91 Add SHA256 dispatcher (Pieter Wuille)
4d50f38 Support multi-block SHA256 transforms (Pieter Wuille)
Pull request description:
This adds an SSE4 assembly version of the SHA256 transform by Intel, and uses it at run time if SSE4 instructions are available, and use a fallback C++ implementation otherwise. Nearly every x86_64 CPU supports SSE4. The feature is only enabled when compiled with `--enable-experimental-asm`.
In order to avoid build dependencies and other complications, the original Intel YASM code was translated to GCC extended asm syntax.
This gives around a 50% speedup on the SHA256 benchmark for me.
It is based on an earlier patch by @laanwj, though only includes a single assembly version (for now), and removes the YASM dependency.
Tree-SHA512: d31c50695ceb45264291537b93c0d7497670be38edf021ca5402eaa7d4e1e0e1ae492326e28d4e93979d066168129e62d1825e0384b1b906d36f85d93dfcb43c
