96299a9d6c0a6b9125a58a63ee3147e55d1b086b Test: Move common function assert_approx() into util.py (fridokus)
Pull request description:
To reduce code duplication, move `assert_approx` into common framework `util.py`.
`assert_approx()` is used in two functional tests.
ACKs for top commit:
theStack:
ACK 96299a9
practicalswift:
ACK 96299a9d6c0a6b9125a58a63ee3147e55d1b086b -- DRY is good and diff looks correct
fanquake:
ACK 96299a9d6c0a6b9125a58a63ee3147e55d1b086b - thanks for contributing 🍻
Tree-SHA512: 8e9d397222c49536c7b3d6d0756cc5af17113e5af8707ac48a500fff1811167fb2e03f3c0445b0b9e80f34935f4d57cfb935c4790f6f5463a32a67df5f736939
## Issue being fixed or feature implemented
`FlushStateToDisk` log is misleading
before
```
FlushStateToDisk: write coins cache to disk (8564 coins, 1199kB) started
FlushStateToDisk: write coins cache to disk (8564 coins, 1199kB) completed (13.98s)
```
after
```
FlushStateToDisk: write coins cache to disk (8564 coins, 1199kB) started
FlushStateToDisk: write coins cache to disk (8564 coins, 1199kB) completed (0.00s)
FlushStateToDisk: write evodb cache to disk started
FlushStateToDisk: write evodb cache to disk completed (13.98s)
```
## What was done?
Make a separate scope and log output for evodb related part. Most of the
changes here are whitespaces, ignoring them reveals the actual changes,
see https://github.com/dashpay/dash/pull/5632/files?w=1.
## How Has This Been Tested?
run a node, check logs
## Breaking Changes
n/a
## Checklist:
- [x] I have performed a self-review of my own code
- [ ] I have commented my code, particularly in hard-to-understand areas
- [ ] I have added or updated relevant unit/integration/functional/e2e
tests
- [ ] I have made corresponding changes to the documentation
- [x] I have assigned this pull request to a milestone _(for repository
code-owners and collaborators only)_
## Issue being fixed or feature implemented
We should be testing from the very first v19 block, not from some random
one 100 blocks after
## What was done?
Implement that, make sure we start at v19 activation.
## How Has This Been Tested?
`make check`
## Breaking Changes
n/a
## Checklist:
- [x] I have performed a self-review of my own code
- [ ] I have commented my code, particularly in hard-to-understand areas
- [ ] I have added or updated relevant unit/integration/functional/e2e
tests
- [ ] I have made corresponding changes to the documentation
- [x] I have assigned this pull request to a milestone _(for repository
code-owners and collaborators only)_
## Issue being fixed or feature implemented
there is no reason for devnet-only params to exist on regtest
## What was done?
remove them
## How Has This Been Tested?
run tests
## Breaking Changes
n/a
## Checklist:
- [x] I have performed a self-review of my own code
- [ ] I have commented my code, particularly in hard-to-understand areas
- [ ] I have added or updated relevant unit/integration/functional/e2e
tests
- [ ] I have made corresponding changes to the documentation
- [x] I have assigned this pull request to a milestone _(for repository
code-owners and collaborators only)_
## Issue being fixed or feature implemented
`GetAdjustedTime()` can be manipulated by our peers, we should avoid
using it for our internal data structures/logic.
## What was done?
Use `GetTime<T>()` instead, fix some includes while at it.
## How Has This Been Tested?
run tests, run a node
## Breaking Changes
should be none
## Checklist:
- [x] I have performed a self-review of my own code
- [ ] I have commented my code, particularly in hard-to-understand areas
- [ ] I have added or updated relevant unit/integration/functional/e2e
tests
- [ ] I have made corresponding changes to the documentation
- [x] I have assigned this pull request to a milestone _(for repository
code-owners and collaborators only)_
## Issue being fixed or feature implemented
The bug was introduced in the original PR #5026 and refactored later
(which is good actually cause we shouldn't mix refactoring and
bug-fixing :) )
## What was done?
fix conditions, add tests
## How Has This Been Tested?
`feature_asset_locks.py`
## Breaking Changes
n/a
## Checklist:
- [x] I have performed a self-review of my own code
- [ ] I have commented my code, particularly in hard-to-understand areas
- [ ] I have added or updated relevant unit/integration/functional/e2e
tests
- [ ] I have made corresponding changes to the documentation
- [x] I have assigned this pull request to a milestone _(for repository
code-owners and collaborators only)_
## Issue being fixed or feature implemented
fix: double lock of deterministicMNManager->cs
Logs:
```
node1 2023-10-21T16:46:03.990302Z (mocktime: 2014-12-04T17:33:19Z) [httpworker.1] DOUBLE LOCK DETECTED
node1 2023-10-21T16:46:03.990322Z (mocktime: 2014-12-04T17:33:19Z) [httpworker.1] Lock order:
node1 2023-10-21T16:46:03.990339Z (mocktime: 2014-12-04T17:33:19Z) [httpworker.1] 'cs_main' in miner.cpp:129 (in thread 'httpworker.1')
node1 2023-10-21T16:46:03.990353Z (mocktime: 2014-12-04T17:33:19Z) [httpworker.1] 'm_mempool.cs' in miner.cpp:129 (in thread 'httpworker.1')
node1 2023-10-21T16:46:03.990366Z (mocktime: 2014-12-04T17:33:19Z) [httpworker.1] (*) 'deterministicMNManager->cs' in evo/cbtx.cpp:114 (in thread 'httpworker.1')
node1 2023-10-21T16:46:03.990439Z (mocktime: 2014-12-04T17:33:19Z) [httpworker.1] (*) 'cs' in ./evo/deterministicmns.h:614 (in thread 'httpworker.1')
node1 2023-10-21T16:46:04.003619Z (mocktime: 2014-12-04T17:33:19Z) [httpworker.1] Posix Signal: Aborted
No debug information available for stacktrace. You should add debug information and then run:
dashd -printcrashinfo=bvcgc43iinzgc43ijfxgm3ybaacwiyltnbsbkudponuxqictnftw4ylmhiqecytpoj2gkzaphcbzaaaaaaaaayeurhimekiaaav6ldwqyiuqaafwsoe5bqrjaaahz6eh2dbcsaaakhhzaaaaaaaaanreseaaaaaaacgguliaaaaaaadyauwqaaaaaaacp3daaaaaaaaamxigcaaaaaaablpulyaaaaaaadovy3yaaaaaaahj7vbaaaaaaaadnbsdaaaaaaaaaa======
```
Part of backtrace:
```
#9 UniqueLock<AnnotatedMixin<std::mutex>, std::unique_lock<std::mutex> >::UniqueLock (fTry=false, nLine=615, pszFile=0x55a9f71a3710 "./evo/deterministicmns.h",
pszName=0x55a9f719caff "cs", mutexIn=..., this=0x7f7e1e71b250) at ./sync.h:164
#10 CDeterministicMNManager::GetListForBlock (this=0x55a9f84d06b0, pindex=0x7f7db03621c0) at ./evo/deterministicmns.h:615
#11 0x000055a9f6612258 in llmq::utils::ComputeQuorumMembersByQuarterRotation (pCycleQuorumBaseBlockIndex=0x7f7db03a6930, llmqParams=...) at /usr/include/c++/12/bits/unique_ptr.h:191
#12 llmq::utils::GetAllQuorumMembers (llmqType=<optimized out>, pQuorumBaseBlockIndex=0x7f7db0359bc0, reset_cache=reset_cache@entry=false) at llmq/utils.cpp:150
#13 0x000055a9f694d957 in CDeterministicMNManager::HandleQuorumCommitment (qc=..., pQuorumBaseBlockIndex=<optimized out>, mnList=..., debugLogs=debugLogs@entry=false)
at evo/deterministicmns.cpp:989
#14 0x000055a9f695c455 in CDeterministicMNManager::BuildNewListFromBlock (this=<optimized out>, block=..., pindexPrev=pindexPrev@entry=0x7f7db03c1ac0, state=..., view=...,
mnListRet=..., debugLogs=false) at evo/deterministicmns.cpp:918
#15 0x000055a9f692e7cd in CalcCbTxMerkleRootMNList (block=..., pindexPrev=pindexPrev@entry=0x7f7db03c1ac0, merkleRootRet=..., state=..., view=...)
at /usr/include/c++/12/bits/unique_ptr.h:191
#16 0x000055a9f6a352ed in BlockAssembler::CreateNewBlock (this=this@entry=0x7f7e1e71f0b0, scriptPubKeyIn=...) at ./validation.h:649
#17 0x000055a9f6771c49 in generateBlocks (chainman=..., mempool=..., evodb=..., llmq_ctx=..., coinbase_script=..., nGenerate=10, nMaxTries=<optimized out>) at rpc/mining.cpp:167
#18 0x000055a9f677a496 in generatetoaddress (request=...) at /usr/include/c++/12/bits/unique_ptr.h:191
#19 0x000055a9f671ea02 in CRPCCommand::CRPCCommand(char const*, char const*, UniValue (*)(JSONRPCRequest const&), std::initializer_list<char const*>)::{lambda(JSONRPCRequest const&, UniValue&, bool)#1}::operator()(JSONRPCRequest const&, UniValue&, bool) const (__closure=<optimized out>, result=..., request=...) at ./rpc/server.h:120
```
## What was done?
`CDeterministicMNManager::BuildNewListFromBlock` doesn't require `cs`
lock anymore
## How Has This Been Tested?
Run unit/functional tests
## Breaking Changes
N/A
## Checklist:
- [x] I have performed a self-review of my own code
- [ ] I have commented my code, particularly in hard-to-understand areas
- [ ] I have added or updated relevant unit/integration/functional/e2e
tests
- [ ] I have made corresponding changes to the documentation
- [x] I have assigned this pull request to a milestone
---------
Co-authored-by: UdjinM6 <UdjinM6@users.noreply.github.com>
Co-authored-by: pasta <pasta@dashboost.org>
## Issue being fixed or feature implemented
Current implementation relies either on asserts or sometimes checks then
returning a special value; In the case of asserts (or no assert where we
use the value without checks) it'd be better to make it explicit to
function caller that the ptr must be not_null; otherwise gsl::not_null
will call terminate.
See
https://github.com/microsoft/GSL/blob/main/docs/headers.md#user-content-H-pointers-not_null
and
https://isocpp.github.io/CppCoreGuidelines/CppCoreGuidelines#Rf-nullptr
I'm interested in a conceptual review; specifically on if this is
beneficial over just converting these ptrs to be a reference?
## What was done?
*Partial* implementation on using gsl::not_null in dash code
## How Has This Been Tested?
Building
## Breaking Changes
None
## Checklist:
_Go over all the following points, and put an `x` in all the boxes that
apply._
- [x] I have performed a self-review of my own code
- [ ] I have commented my code, particularly in hard-to-understand areas
- [ ] I have added or updated relevant unit/integration/functional/e2e
tests
- [ ] I have made corresponding changes to the documentation
- [x] I have assigned this pull request to a milestone _(for repository
code-owners and collaborators only)_
---------
Signed-off-by: pasta <pasta@dashboost.org>
Co-authored-by: UdjinM6 <UdjinM6@users.noreply.github.com>
## Issue being fixed or feature implemented
To make configuring devnets more error-prune and config file shorter
## What was done?
Updated default LLMQ parameters on devnet from 50_60, 60_75, 100_67 to
`LLMQ_DEVNET` and `LLMQ_DEVNET_PLATFORM`.
## How Has This Been Tested?
not tested yet; would be tested on devnets later with next
devnet/release
## Breaking Changes
n/a for non-dev-nets; for dev-net other default quorum is used.
## Checklist:
- [x] I have performed a self-review of my own code
- [ ] I have commented my code, particularly in hard-to-understand areas
- [ ] I have added or updated relevant unit/integration/functional/e2e
tests
- [ ] I have made corresponding changes to the documentation
- [x] I have assigned this pull request to a milestone
---------
Co-authored-by: UdjinM6 <UdjinM6@users.noreply.github.com>
## Issue being fixed or feature implemented
By default consensus for devnet if 50_60 that is way too much:
```
consensus.llmqTypeMnhf = Consensus::LLMQType::LLMQ_50_60;
```
So, `quorum list` on devnet-ouzo is empty:
```
{
"llmq_50_60": [
],
```
## What was done?
Adds new -llmqmnhf param for devnet to change quorum params dynamically.
## How Has This Been Tested?
<not tested>
## Breaking Changes
n/a
## Checklist:
- [x] I have performed a self-review of my own code
- [ ] I have commented my code, particularly in hard-to-understand areas
- [ ] I have added or updated relevant unit/integration/functional/e2e
tests
- [ ] I have made corresponding changes to the documentation
- [x] I have assigned this pull request to a milestone
## Issue being fixed or feature implemented
#5597 follow-up
## What was done?
add missing filed description
## How Has This Been Tested?
`help getblockchaininfo`
## Breaking Changes
n/a
## Checklist:
- [x] I have performed a self-review of my own code
- [ ] I have commented my code, particularly in hard-to-understand areas
- [ ] I have added or updated relevant unit/integration/functional/e2e
tests
- [ ] I have made corresponding changes to the documentation
- [x] I have assigned this pull request to a milestone _(for repository
code-owners and collaborators only)_
## Issue being fixed or feature implemented
- Using `actions/cache` with a local buildx cache without the "move
cache" workaround will result in constant growth in cache size:
https://docs.docker.com/build/ci/github-actions/cache/#local-cache
## What was done?
- Docker natively supports the GHA Cache API, so we should use it for
faster and more efficient cache usage
- Actions were also bumped to current stable versions
## How Has This Been Tested?
Devs please test this by running a test Guix build from workflow
dispatch
## Breaking Changes
None
## Checklist:
_Go over all the following points, and put an `x` in all the boxes that
apply._
- [x] I have performed a self-review of my own code
- [x] I have commented my code, particularly in hard-to-understand areas
- [x] I have added or updated relevant unit/integration/functional/e2e
tests
- [x] I have made corresponding changes to the documentation
- [x] I have assigned this pull request to a milestone _(for repository
code-owners and collaborators only)_
(over a 13 minute segment of reindex) boost time took about 9+9+9 =
27%!! of total cycles!!!
During the same time segment, GetTimeMicros() used roughly 0.2% of total
cycles! (even though it's used a lot more)
Over my entire roughly 40 minute profiling session, boost time appears
to use about 6% of cycles; still too much, while GetTimeMicros used
about 0.4%
## Checklist:
_Go over all the following points, and put an `x` in all the boxes that
apply._
- [x] I have performed a self-review of my own code
- [ ] I have commented my code, particularly in hard-to-understand areas
- [ ] I have added or updated relevant unit/integration/functional/e2e
tests
- [ ] I have made corresponding changes to the documentation
- [x] I have assigned this pull request to a milestone _(for repository
code-owners and collaborators only)_
---------
Co-authored-by: UdjinM6 <UdjinM6@users.noreply.github.com>
## Issue being fixed or feature implemented
Now that v19 is buried we can enforce basic bls scheme usage in
governance and coinjoin and drop some extra code we used for backwards
compatibility.
## What was done?
pls see individual commits
## How Has This Been Tested?
run tests, sync and mix on testnet
## Breaking Changes
n/a
## Checklist:
- [x] I have performed a self-review of my own code
- [ ] I have commented my code, particularly in hard-to-understand areas
- [ ] I have added or updated relevant unit/integration/functional/e2e
tests
- [ ] I have made corresponding changes to the documentation
- [x] I have assigned this pull request to a milestone _(for repository
code-owners and collaborators only)_
## Issue being fixed or feature implemented
It makes more sense for DecreaseScores to be inside of the MNList itself
imo
## What was done?
Refactored as such
## How Has This Been Tested?
Reindexed
I had originally expected some performance improvements due to the
removal of `GetMN` but in my benchmarking I didn't see any noticeable
perf changes. I do still think the removal of `GetMN` and using a
shared_ptr the whole time is better as it removes the chance of the
master node disappearing from the list (which would have previously
thrown, but is now impossible).
## Breaking Changes
None
## Checklist:
_Go over all the following points, and put an `x` in all the boxes that
apply._
- [x] I have performed a self-review of my own code
- [ ] I have commented my code, particularly in hard-to-understand areas
- [ ] I have added or updated relevant unit/integration/functional/e2e
tests
- [ ] I have made corresponding changes to the documentation
- [x] I have assigned this pull request to a milestone _(for repository
code-owners and collaborators only)_
---------
Co-authored-by: UdjinM6 <UdjinM6@users.noreply.github.com>
Implementation EHF mechanism, part 4. Previous changes are:
- https://github.com/dashpay/dash/pull/4577
- https://github.com/dashpay/dash/pull/5505
- https://github.com/dashpay/dash/pull/5469
## Issue being fixed or feature implemented
Currently MN_RR is activated automatically by soft-fork activation after
v20 is activated.
It is not flexible enough, because platform may not be released by that
time yet or in opposite it can be too long to wait.
Also, any signal of EHF requires manual actions from MN owners to sign
EHF signal - it is automated here.
## What was done?
New spork `SPORK_24_MN_RR_READY`; new EHF manager that sign EHF signals
semi-automatically without manual actions; and send transaction with EHF
signal when signal is signed to network.
Updated rpc `getblockchaininfo` to return information about of EHF
activated forks.
Fixed function `IsTxSafeForMining` in chainlock's handler to skip
transactions without inputs (empty `vin`).
## How Has This Been Tested?
Run unit/functional tests. Some tests have been updated due to new way
of MN_RR activation: `feature_asset_locks.py`, `feature_mnehf.py`,
`feature_llmq_evo.py` and unit test `block_reward_reallocation_tests`.
## Breaking Changes
New way of MN_RR activation.
## Checklist:
- [x] I have performed a self-review of my own code
- [x] I have commented my code, particularly in hard-to-understand areas
- [x] I have added or updated relevant unit/integration/functional/e2e
tests
- [ ] I have made corresponding changes to the documentation
- [x] I have assigned this pull request to a milestone _(for repository
code-owners and collaborators only)_
---------
Co-authored-by: UdjinM6 <UdjinM6@users.noreply.github.com>
Co-authored-by: PastaPastaPasta <6443210+PastaPastaPasta@users.noreply.github.com>
## Issue being fixed or feature implemented
Calculation of `platformReward` should ignore fees and rely only on
Block subsidy.
cc @QuantumExplorer
## What was done?
From now on, the following formula is applied:
```
blockReward = blockSubsidy + feeReward
masternodeReward = masternodeShare(blockSubsidy)
platformReward = platformShare(masternodeReward)
masternodeReward += masternodeShare(feeReward)
```
## How Has This Been Tested?
## Breaking Changes
`plaftormReward` differs in networks where `mn_rr` is already active
## Checklist:
_Go over all the following points, and put an `x` in all the boxes that
apply._
- [x] I have performed a self-review of my own code
- [ ] I have commented my code, particularly in hard-to-understand areas
- [x] I have added or updated relevant unit/integration/functional/e2e
tests
- [ ] I have made corresponding changes to the documentation
- [x] I have assigned this pull request to a milestone _(for repository
code-owners and collaborators only)_
---------
Co-authored-by: UdjinM6 <UdjinM6@users.noreply.github.com>
## Issue being fixed or feature implemented
Currently, the `nSubsidyBase` calculation relies on difficulty. This
leads to variable Block Subsidity.
When Platform will be live, it would constantly require blocks
difficulty in order to calculate the `platformReward` (which relies on
Block Subsidy)
cc @QuantumExplorer
## What was done?
Starting from v20 activation, `nSubsidyBase` will no longer rely on
difficulty and will be constant to 5.
## How Has This Been Tested?
## Breaking Changes
Block rewards will differ.
## Checklist:
- [x] I have performed a self-review of my own code
- [x] I have commented my code, particularly in hard-to-understand areas
- [x] I have added or updated relevant unit/integration/functional/e2e
tests
- [ ] I have made corresponding changes to the documentation
- [x] I have assigned this pull request to a milestone _(for repository
code-owners and collaborators only)_
## Issue being fixed or feature implemented
Signature for withdrawal (asset unlock) transaction should be validated
against platform quorum (100_67) but not same as currently against EHF
quorum (400_85).
## What was done?
Updates type of quorum in chainparams for Asset Unlock (withdrawal)
transactions to same as platform's quorum.
It is first part of changes to fix devnet, testnet and mainnet. For
regnet is still used incorrect quorum due to non-trivial changes in
functional test `feature_assetlocks.py`; these changes would be provided
in next PR.
## How Has This Been Tested?
Run unit/functional test.
## Breaking Changes
Yes, quorum for validation of Asset Unlock (withdrawal) transaction is
changed.
## Checklist:
- [x] I have performed a self-review of my own code
- [ ] I have commented my code, particularly in hard-to-understand areas
- [ ] I have added or updated relevant unit/integration/functional/e2e
tests
- [ ] I have made corresponding changes to the documentation
- [x] I have assigned this pull request to a milestone
this change saw a ~38% performance improvement in header sync reindex
reproduce via `time ./src/qt/dash-qt --nowallet --testnet --reindex
--stopatheight=5`
On Develop this took average of 1:48 to finish, on this branch it took
1:07
## Issue being fixed or feature implemented
Slow header / block validation
## What was done?
Pass around cached block hash
## How Has This Been Tested?
Reindexed testnet
## Breaking Changes
None
## Checklist:
_Go over all the following points, and put an `x` in all the boxes that
apply._
- [x] I have performed a self-review of my own code
- [ ] I have commented my code, particularly in hard-to-understand areas
- [ ] I have added or updated relevant unit/integration/functional/e2e
tests
- [ ] I have made corresponding changes to the documentation
- [x] I have assigned this pull request to a milestone _(for repository
code-owners and collaborators only)_
Co-authored-by: UdjinM6 <UdjinM6@users.noreply.github.com>
## Issue being fixed or feature implemented
Fixed a problem forgotten in #5588 in feature_asset_locks.py.
## What was done?
Avoid floating operations when calculating `coinbasevalue`
## How Has This Been Tested?
## Breaking Changes
## Checklist:
- [x] I have performed a self-review of my own code
- [ ] I have commented my code, particularly in hard-to-understand areas
- [ ] I have added or updated relevant unit/integration/functional/e2e
tests
- [ ] I have made corresponding changes to the documentation
- [x] I have assigned this pull request to a milestone _(for repository
code-owners and collaborators only)_
---------
Co-authored-by: UdjinM6 <UdjinM6@users.noreply.github.com>
## Issue being fixed or feature implemented
Since `mn_rr` is already active on Testnet, because of #5588, syncing
from develop is broken.
## What was done?
Temporary disabled changes of #5588 for Testnet.
This should be dropped when Testnet will be re-organised for Platform.
## How Has This Been Tested?
Syncing Testnet
## Breaking Changes
## Checklist:
- [x] I have performed a self-review of my own code
- [ ] I have commented my code, particularly in hard-to-understand areas
- [ ] I have added or updated relevant unit/integration/functional/e2e
tests
- [ ] I have made corresponding changes to the documentation
- [x] I have assigned this pull request to a milestone _(for repository
code-owners and collaborators only)_
## Issue being fixed or feature implemented
Historical masternode data returned via rpcs like `protx listdiff` can
be broken because some collaterals might be spent already and
`GetUTXOCoin` wasn't able to get any info.
## What was done?
Use `GetTransaction` as a fallback.
## How Has This Been Tested?
run tests
## Breaking Changes
n/a
## Checklist:
- [x] I have performed a self-review of my own code
- [ ] I have commented my code, particularly in hard-to-understand areas
- [ ] I have added or updated relevant unit/integration/functional/e2e
tests
- [ ] I have made corresponding changes to the documentation
- [x] I have assigned this pull request to a milestone _(for repository
code-owners and collaborators only)_
## Issue being fixed or feature implemented
avoid potential discrepancies in block reward calculations
## What was done?
use integers (int64_t) only when dealing with block rewards, no
float/double
## How Has This Been Tested?
run tests
## Breaking Changes
might fork off on devnets that use previous version
## Checklist:
- [x] I have performed a self-review of my own code
- [ ] I have commented my code, particularly in hard-to-understand areas
- [ ] I have added or updated relevant unit/integration/functional/e2e
tests
- [ ] I have made corresponding changes to the documentation
- [x] I have assigned this pull request to a milestone _(for repository
code-owners and collaborators only)_
