9b2704777c [doc] Include txindex changes in the release notes. (Jim Posen)
ed77dd6b30 [test] Simple unit test for TxIndex. (Jim Posen)
6d772a3d44 [rpc] Public interfaces to GetTransaction block until synced. (Jim Posen)
a03f804f2a [index] Move disk IO logic from GetTransaction to TxIndex::FindTx. (Jim Posen)
e0a3b80033 [validation] Replace tx index code in validation code with TxIndex. (Jim Posen)
8181db88f6 [init] Initialize and start TxIndex in init code. (Jim Posen)
f90c3a62f5 [index] TxIndex method to wait until caught up. (Jim Posen)
70d510d93c [index] Allow TxIndex sync thread to be interrupted. (Jim Posen)
94b4f8bbb9 [index] TxIndex initial sync thread. (Jim Posen)
34d68bf3a3 [index] Create new TxIndex class. (Jim Posen)
c88bcec93f [db] Migration for txindex data to new, separate database. (Jim Posen)
0cb8303241 [db] Create separate database for txindex. (Jim Posen)
Pull request description:
I'm re-opening #11857 as a new pull request because the last one stopped loading for people
-------------------------------
This refactors the tx index code to be in it's own class and get built concurrently with validation code. The main benefit is decoupling and moving the txindex into a separate DB. The primary motivation is to lay the groundwork for other indexers that might be desired (such as the [compact filters](https://github.com/bitcoin/bips/pull/636)). The basic idea is that the TxIndex spins up its own thread, which first syncs the txindex to the current block index, then once in sync the BlockConnected ValidationInterface hook writes new blocks.
### DB changes
At the suggestion of some other developers, the txindex has been split out into a separate database. A data migration runs at startup on any nodes with a legacy txindex. Currently the migration blocks node initialization until complete.
### Open questions
- Should the migration of txindex data from the old DB to the new DB block in init or should it happen in a background thread? The downside to backgrounding it is that `getrawtransaction` would return an error message saying the txindex is syncing while the migration is running.
### Impact
In a sample size n=1 test where I synced nodes from scratch, the average time [Index writing](https://github.com/bitcoin/bitcoin/blob/master/src/validation.cpp#L1903) was 3.36ms in master and 1.72ms in this branch. The average time between `UpdateTip` log lines for sequential blocks between 400,000 and IBD end on mainnet was 0.297204s in master and 0.286134s in this branch. Most likely this is just variance in IBD times, but I can try with some more trials if people want.
Tree-SHA512: 451fd7d95df89dfafceaa723cdf0f7b137615b531cf5c5035cfb54e9ccc2026cec5ac85edbcf71b7f4e2f102e36e9202b8b3a667e1504a9e1a9976ab1f0079c4
* build: Add cmake as depends package (cmake.mk)
The bls-signatures library requires cmake 3.14
* depends: Update chia_bls to version 1.0.0 of dashpay/bls-signatures
* depends: Rename package chia_bls to bls-dash
* depends: Disable blspy/tests/benchmarks build for chia_bls
Note: Building with tests would require the following:
depends: Fix macOS build for versions < 10.12
Seems like older versions of macOS pretend to support c++17 std libs but do not have/have issues with uncaught_exceptions. "Catch", the testing framework used in the bls lib wants those by default but setting `DCATCH_CONFIG_NO_CPP17_UNCAUGHT_EXCEPTIONS` disables them which should be just fine to do in all cases here.
---
depends/packages/chia_bls.mk | 2 ++
1 file changed, 2 insertions(+)
diff --git a/depends/packages/chia_bls.mk b/depends/packages/chia_bls.mk
index e79a85844f..25749b74ab 100644
--- a/depends/packages/chia_bls.mk
+++ b/depends/packages/chia_bls.mk
@@ -28,6 +28,8 @@ define $(package)_set_vars
$(package)_config_opts_darwin+= -DCMAKE_AR="$(host_prefix)/native/bin/$($(package)_ar)"
$(package)_config_opts_darwin+= -DCMAKE_RANLIB="$(host_prefix)/native/bin/$($(package)_ranlib)"
endif
+
+ $(package)_cppflags+= -DCATCH_CONFIG_NO_CPP17_UNCAUGHT_EXCEPTIONS
endef
define $(package)_config_cmds
--
* depends: Drop obsolete variable
* bls: Integrate the upgraded version into the codebase
* depends: Pre-fetch relic to fix gitian
Can fetch apt packages only when building with lxc
Co-authored-by: UdjinM6 <UdjinM6@users.noreply.github.com>
a9e898a4ad547ad344671db2b942b99925f0c732 docs: Link to python style guidelines from developer notes (Mason Simon)
Pull request description:
Initially I moved the python style guidelines from the functional test README, but some of the python rules are test-specific, and most of the developer notes doc is C++ centric, so just dropping a link seemed better.
Tree-SHA512: 9d4d5cc45526319a118595d90fcfad2c9aced22007aa096d8af04ba1b963312822804f4c15b0b227d66af49565034437691b7760e7ff6d1e3f8b10b898906362
3828a79711 scripted-diff: prefer MAC_OSX over __APPLE__ (fanquake)
fa6e841e89 gui: remove macOS ProgressBar workaround (fanquake)
68c272527f gui: remove SubstituteFonts (fanquake)
6c6dbd8af5 doc: mention that macOS 10.10 is now required (fanquake)
84b0cfa8b6 release: bump minimum required macOS to 10.10 (fanquake)
26b15df99d depends: set OSX_MIN_VERSION to 10.10 (fanquake)
Pull request description:
Closes#13362
d99abfddb0c8f2111340a6127e77cc686e0043d8
This workaround should no longer be required, as it should have only been in use when compiled with the 10.7 SDK, which we haven't been building with for a while now.
5bc5ae30982a0f0f6a9804b05d99434af770c724
The bugreport linked with this code is for an unrelated? issue, however from what I can tell the correct QTBUG is this one https://bugreports.qt.io/browse/QTBUG-20880. Reading though the discussion there, it seems that the way progress bars are animated changed in macOS 10.10.
Qt was patched [here (5.5+)](https://codereview.qt-project.org/#/c/112379/):
> Disable progress bar animations on 10.10 Yosemite and higher - the native style does not animate them any more. Keep the indeterminate progress bar animation.
Given all of that, I don't think this is worth keeping around, as it would seem to only be useful in the case that a macOS user is compiling with a Qt < 5.5. That should be pretty unlikely, as we don't support downloaded Qt binaries, and brew currently provides [5.11.1](571b46213c/Formula/qt.rb).
Tree-SHA512: 4278cb30cc9bcb313e166129ecf032c808995f8b51a3123637c47860a0010ac88f86f82ec44792153b6b1e5cca595f25013b2eaeae80194647b9ce4f7eaf32c1
1874058 Make base58 python contrib code work with python3 (Evan Klitzke)
bc6fdf2 Change all python files to use Python3 (John Newbery)
Pull request description:
Following discussion here: https://github.com/bitcoin/bitcoin/pull/11843#issuecomment-351033742
It's easier for maintainers if all python tools/scripts support only a single version of Python. There are only a few scripts that aren't explicitly python3 at this point, so this PR changes those remaining scripts to explicitly require python3.
Tree-SHA512: 5d38eef6e0fc7d8515e23a1f4c75e8b4160fd0fe23cba52a1f41689b114e54a9e503e0724829e8b41982ef98f2d113df80d9e238213b74f09ceaed0344a19e24
f381299 Move CKeyStore::cs_KeyStore to CBasicKeyStore (João Barbosa)
25eb9f5 Inline CKeyStore::AddKey(const CKey &) in CBasicKeyStore (João Barbosa)
Pull request description:
Made these simplifications while reviewing #12714. This aims to make `CKeyStore` a *pure* interface:
- no variable members - the mutex is moved to `CBasicKeyStore` which is where it is used;
- no method implementations - `AddKey(const CKey &)` is moved to `CBasicKeyStore` which is where it is needed.
Tree-SHA512: 84e44f4390c59600e5cefa599b5464e1771c31dd4abc678ef50db8e06ffac778d692860a352918444f8bcd66430634637b6277a818a658721ffc4f381c1c6a90
6feb46c Add --with-sanitizers option to configure (Evan Klitzke)
Pull request description:
This adds configure options for `-fsanitize=address`, `-fsanitize=thread`, and `-fsanitize=undefined` which are all disabled by default. These flags are useful for developers who wish to do additional safety checking. Note that some of these are mutually incompatible, and these may have a large performance overhead.
There's some kind of strange logic required to properly check for the availability of these flags in a way that works on both GCC and Clang, hopefully the comments make it clear what's going on.
Tree-SHA512: 2d6fe402799110e59ee452dddf37f7ca2d26a7fecec50be25c8a134e4a20beb31f1e8f438dffd443641562418075896d1eeb450623425b272d80e05e3027a587
0bd2ec5 Improve formatting of developer notes (Evan Klitzke)
Pull request description:
The developer notes file has gotten pretty large and unwieldly. This reorganizes some content, most notably by adding a TOC to the page. Compare how the page looks in [my branch](https://github.com/eklitzke/bitcoin/blob/developer-notes/doc/developer-notes.md) vs [master](https://github.com/eklitzke/bitcoin/blob/master/doc/developer-notes.md).
I know there's long-term interest in moving a lot of this content to the bitcoin-core/docs repo on GitHub, but this makes things better now.
The TOC format here is a semi-standard extension to Markdown files that you may have seen on GitHub pages in other projects. The `<!-- markdown-toc -->` comments used by these tools to know where the TOC starts/ends. The following tools all understand this format:
* [Sphinx](http://www.sphinx-doc.org/en/master/)
* [Pandoc](https://pandoc.org/)
* [markdown-toc.el](https://github.com/ardumont/markdown-toc) (what I used)
* various plugins for [vim](https://github.com/mzlogin/vim-markdown-toc) and other editors
I used this TOC extension at a previous job, and my observation is that it's fine if people just edit the TOC manually. It's just text and it's not a huge deal if it gets a little out of sync. It can also be regenerated at any time by anyone with any of these tools.
Tree-SHA512: 298d1605ea5e8bfc0f75e70570c23ebd6891e4ffcdedd24fefadc23edd6e4b96509d8d102209868468a1b3ddbe2c3b8462698cdda8b9421348b5bc6f7b8d0cb8
ccedbaf Increase LevelDB max_open_files unless on 32-bit Unix. (Evan Klitzke)
Pull request description:
Currently we set `max_open_files = 64` on all architectures due to concerns about file descriptor exhaustion. This is extremely expensive due to how LevelDB is designed.
When a LevelDB file handle is opened, a bloom filter and block index are decoded, and some CRCs are checked. Bloom filters and block indexes in open table handles can be checked purely in memory. This means that when doing a key lookup, if a given table file may contain a given key, all of the lookup operations can happen completely in RAM until the block itself is fetched. In the common case fetching the block is one disk seek, because the block index stores its physical offset. This is the ideal case, and what we want to happen as often as possible.
If a table file handle is not open in the table cache, then in addition to the regular system calls to open the file, the block index and bloom filter need to be decoded before they can be checked. This is expensive and is something we want to avoid.
The current setting of 64 file handles means that on a synced node, only about 4% of key lookups can be satisifed by table file handles that are actually open and in memory.
The original concerns about file descriptor exhaustion are unwarranted on most systems because:
* On 64-bit POSIX hosts LevelDB will open up to 1000 file descriptors using `mmap()`, and it does not retain an open file descriptor for such files.
* On Windows non-socket files do not interfere with the main network `select()` loop, so the same fd exhaustion issues do not apply there.
This change keeps the default `max_open_files` value (which is 1000) on all systems except 32-bit POSIX hosts (which do not use `mmap()`). Open file handles use about 20 KB of memory (for the block index), so the extra file handles do not cause much memory overhead. At most 1000 will be open, and a fully synced node right now has about 1500 such files.
Profile of `loadblk` thread before changes: https://monad.io/maxopenfiles-master.svg
Profile of `loadblk` thread after changes: https://monad.io/maxopenfiles-increase.svg
Tree-SHA512: de54f77d57e9f8999eaf8d12592aab5b02f5877be8fa727a1f42cf02da2693ce25846445eb19eb138ce4e5045d1c65e14054df72faf3ff32c7655c9cfadd27a9
* Make the build an actual release
Signed-off-by: pasta <pasta@dashboost.org>
* bump build version
Signed-off-by: pasta <pasta@dashboost.org>
* bump diff version
Signed-off-by: pasta <pasta@dashboost.org>
* Update nMinimumChainWork and defaultAssumeValid
* Update chainTxData
* Bump BLOCK_CHAIN_SIZE
It's 21Gb on my local machine atm, bumping to 25Gb to give it some space
* `gen-manpages.sh`
* bupm BLOCK_CHAIN_SIZE 25->30
* Update nMinimumChainWork and defaultAssumeValid
* Update chainTxData
* Bump BLOCK_CHAIN_SIZE
It's 21Gb on my local machine atm, bumping to 25Gb to give it some space
* `gen-manpages.sh`
* bupm BLOCK_CHAIN_SIZE 25->30
fa3c910bfeab00703c947c5200a64c21225b50ef test: Move linters to test/lint, add readme (MarcoFalke)
Pull request description:
This moves the checks and linters from `devtools` to a subfolder in `test`. (Motivated by my opinion that the dev tools are mostly for generating code and updating the repo whereas the linters are read-only checks.)
Also, adds a readme to clarify that checks and linters are only meant to prevent bugs and user facing issues, not merely stylistic preference or inconsistencies. (This is motivated by the diversity in developers and work flows as well as existing code styles. It would be too disruptive to change all existing code to a single style or too burdensome to force all developers to adhere to a single style. Also note that our style guide is changing, so locking in at the wrong style "too early" would only waste resources.)
Tree-SHA512: 9b10e89f2aeaf0c8a9ae248aa891d74e0abf0569f8e5dfd266446efa8bfaf19f0ea0980abf0b0b22f0d8416ee90d7435d21a9f9285b66df43f370b7979173406
fad0fc3c9a Refine travis check for duplicate includes (MarcoFalke)
Pull request description:
Since there is no harm in having "duplicate" includes and it makes it obvious what are the dependencies of each file, without having to do static analysis or jumping between files, I'd suggest to revert the travis check for duplicate includes.
Generally, I think that enforcing minor style preferences should not be done via travis. The cost of maintaining and the burden on other developers is too high. C.f discussion in https://github.com/bitcoin/bitcoin/pull/10973#discussion_r180142594
Tree-SHA512: 97ab0e769d457ccfb873fff6c99613f8b944cd7ef95bfdccb0e1bbe8f5df1f16548c658fa03af42516f806546e75646d338a061e7b057619490235d311ca21f1
95464c7519 doc: Improve command to generate list of authors for release notes (Mitchell Cash)
1c22cc1af1 doc: Update broken links to now point to gitian-build.py (Mitchell Cash)
Pull request description:
- Update broken links
- Improve command to generate list of authors for a release
---
I also note that it asks to ping **wumpus** on IRC, to assist in generating a list of merged pulls and sort them into categories based on labels. I tried to turn this into a simple one-liner as well (something like ``git log --merges --format="- \`%h\` %s (%an)" v0.16.0..v0.16.1``), but it didn't seem to capture everything I needed.
Would it be worthwhile **wumpus** open-sourcing his code into `contrib/devtools` so there is no single point of failure (even if it can manually be worked around).
Tree-SHA512: 8e7f0880ff07ce8fe67b74de3f2c4b78dafe2c95eefb0617fa319ff196232967cb22ee75a183a39f93bfc6e0bf85547689160139312cee5956af2c069b8a3b6a
