Evgenii Stratonikov
f665843887
core: update verified state root height
2020-06-22 10:53:12 +03:00
Roman Khimov
76f71ab1ef
Merge pull request #1012 from nspcc-dev/feature/mptrpc
...
rpc: implement MPT-related RPC (2.x)
2020-06-09 17:25:55 +03:00
Evgenii Stratonikov
8cbbddddaf
rpc: implement getproof RPC
2020-06-08 17:31:58 +03:00
Roman Khimov
6437f4b32e
core: respect MaxFreeTransactionsPerBlock setting
...
Fix #1019 .
2020-06-08 12:35:39 +03:00
Roman Khimov
a1c4d7ce26
core: do MPT compaction every once in a while
...
We need to compact our in-memory MPT from time to time, otherwise it quickly
fills up all available memory. This raises two obvious quesions --- when to do
that and to what level do that.
As for 'when', I think it's quite easy to use our regular persistence interval
as an anchor (and it also frees up some memory), but we can't do that in the
persistence routine itself because of synchronization issues (adding some
synchronization primitives would add some cost that I'd also like to avoid),
so do it indirectly by comparing persisted and current height in `storeBlock`.
Choosing proper level is another problem, but if we're to roughly estimate one
full branch node to use 1K of memory (usually it's way less than that) then we
can easily store 1K of these nodes and that gives us a depth of 10 for our
trie.
2020-06-04 17:25:57 +03:00
Roman Khimov
69ccca675d
core: fix PrevHash calculation for MPTRoot
...
This was differing from C# notion of PrevHash. It's not a previous root, but
rather a hash of the previous serialized MPTRoot structure (that is to be
signed by CNs).
2020-06-04 17:19:30 +03:00
Evgenii Stratonikov
20f190ef69
core: update MPT during block processing
2020-06-03 13:33:44 +03:00
Evgenii Stratonikov
503442a60d
dao: restrict GetStorageItems by prefix
...
All storage items can still be retrived via zero-length prefix.
2020-05-27 11:40:46 +03:00
Evgenii Stratonikov
b96fe8173c
core,dao: implement Block-level storage caching
...
The order in which storage.Find items are returns depends on what items
were processed in previous transactions of the same block.
The easiest way to implement this sort of caching is to cache operations
with storage, flushing the only in `Persist()`.
2020-05-19 17:19:51 +03:00
Roman Khimov
9454ef5c28
core: improve locking in storeBlock
...
Getting batch, updating Prometheus metrics and pushing events doesn't require
any locking: batch is a local cache batch that no one outside cares about,
Prometheus metrics are not critical to be in perfect sync and events are
asynchronous anyway.
2020-05-13 17:17:41 +03:00
Roman Khimov
1ac4f8528d
core: add Blockchain event subscription mechanism
...
A deep internal part of #895 . Blockchainer interface is also extended for
various uses of these methods.
2020-05-12 17:41:15 +03:00
Roman Khimov
2e58a14978
core: improve documentation a little
2020-05-12 17:20:31 +03:00
Evgenii Stratonikov
ee0ba9b1b4
core: make SpawnVM a method of context
...
spawnVMWithInterops is rather long too type and
it doesn't use Blockchain in any way.
2020-04-08 08:38:45 +03:00
Evgenii Stratonikov
030b7754ad
core: move DAO to a separate package
2020-04-08 08:38:44 +03:00
Roman Khimov
5f09381cf4
core: wrap cached dao properly, don't miss cached data
...
Fixes #817 where invoked contract missed updated account information because
it got it one layer below cachedDao used to process the block.
2020-04-03 10:15:11 +03:00
Anna Shaleva
5a984fdf88
config: move config.go out of config/
...
closes #423
2020-03-31 17:55:59 +03:00
Evgenii Stratonikov
6ed2bd63b9
core: marshal AppExecResult.Stack as an array
...
Closes #782 .
2020-03-23 16:27:47 +03:00
Roman Khimov
d5d0479671
core: verify results of issue transaction
...
It shouldn't try to issue more tokens than there is available.
2020-03-17 11:40:01 +03:00
Roman Khimov
10601cb375
core: add issuer hashes into the verification list for Issue TX
...
As it should be done.
2020-03-17 11:40:00 +03:00
Roman Khimov
dca637d2d0
core: add owner hash to verified list for Register TX
...
As it should be done.
2020-03-17 11:37:02 +03:00
Roman Khimov
77a799f7d0
core: add missing state tx verifications
2020-03-16 20:14:59 +03:00
Roman Khimov
6ede65610d
core: add appropriate hashes to check for State TX
...
These checks are important for proper transaction verification.
2020-03-16 19:52:28 +03:00
Evgenii Stratonikov
e503d1001d
core/state: do not unmarshal Unclaimed balances in account
2020-03-16 12:19:32 +03:00
Evgenii Stratonikov
32401a567e
core: store NEP5Transfers in batches
...
This is an append-only log which is read only during some RPCs.
It is rather slow to get it from base every time we need to append to
it. This commit stores all NEP5Transfers in batches, so that
only a last batch needs to be unmarshaled during block processing.
2020-03-12 14:51:12 +03:00
Evgenii Stratonikov
df2598c8dc
core: store NEP5 balances separately
...
There is no need to take and unmarshal an account structure only
to get it's NEP5 balances.
2020-03-12 12:58:28 +03:00
Roman Khimov
5f1868af28
Merge pull request #743 from nspcc-dev/optimize-some-db-accesses
...
Optimize some db accesses
2020-03-11 17:33:36 +03:00
Roman Khimov
23464401bc
core/state: merge spent and unspent coins state, use it to store more things
...
This change reduces pressure on DB by doing the following things:
* not storing additional KV pair for SpentCoin
* storing Output right in the UnspentCoin, thus eliminating the need to get a
full transaction from DB
At the same time it makes UnspentCoin more fat and hot, but it should probably
worth it.
Also drop `GetUnspentCoinStateOrNew` as it shouldn't ever existed, UTXOs
can't come out of nowhere.
1.5M block import time (VerifyBlocks disabled) on AMD Ryzen 5 1600/16GB/HDD,
before:
real 302m9.895s
user 96m17.200s
sys 13m37.084s
after:
real 159m16.551s
user 69m58.279s
sys 7m34.334s
So it's almost two-fold which is a great improvement.
2020-03-11 12:40:02 +03:00
Roman Khimov
e1f194ea7b
core: treat state.Coin as a bitfield
...
As it was intended to.
2020-03-11 12:22:57 +03:00
Roman Khimov
377fb382aa
core: move (un)SpentCoin structs into the state package
...
As they're all about the state.
2020-03-11 12:22:52 +03:00
Roman Khimov
8318adac56
vm: add support for dynamic invocations in APPCALL
...
Fixes #740 .
2020-03-10 17:17:36 +03:00
Anna Shaleva
456a2d55fd
rpc: implement getvalidators
...
closes #714
2020-03-07 18:05:40 +03:00
Roman Khimov
baeaa3dbe6
core: optimize tx verification, only get references and results once
...
Getting references requires DB access and that is expensive.
2020-03-06 20:15:01 +03:00
Roman Khimov
fe4916f691
core: simplify GetTransactionResults() a bit
...
Make less movements.
2020-03-06 19:26:39 +03:00
Roman Khimov
f8eee778f4
Merge pull request #724 from nspcc-dev/feature/submitblock
...
rpc: implement submitblock
2020-03-06 12:08:45 +03:00
Evgenii Stratonikov
40188c5400
core: get rid of unnecessary copies
...
There is no need to keep Balances and Unclaimed slices
sorted, we need only to remove a single element.
2020-03-05 20:42:20 +03:00
Anna Shaleva
d3063c26e1
core: add custom error to blockhain.go
...
Add InvalidBlockIndex error to AddBlock func
2020-03-05 20:30:19 +03:00
Evgenii Stratonikov
95a8fa234f
rpc: implement getnep5transfers RPC
2020-03-05 18:22:40 +03:00
Evgenii Stratonikov
f92fd3c948
core: track NEP5 transfers
2020-03-05 18:22:40 +03:00
Evgenii Stratonikov
e8c4179a9c
core: track NEP5 balances
2020-03-05 18:22:19 +03:00
Evgenii Stratonikov
3a510b9dad
core: allow transfer amount to be bytes
...
VM can produce both big.Int and []byte because they
are converted to each other on demand.
2020-03-05 15:18:38 +03:00
Roman Khimov
5c781be08d
Merge pull request #718 from nspcc-dev/fix-invocation-tx-checks-and-fee
...
Fix Invocation TX checks and fee
2020-03-05 11:51:51 +03:00
Roman Khimov
9ac8001c7d
core: protect (*Blockchain).keyCache with a lock
...
Avoid failures like this:
fatal error: concurrent map writes
goroutine 103 [running]:
runtime.throw(0xca6fe3, 0x15)
/usr/lib64/go/1.12/src/runtime/panic.go:617 +0x72 fp=0xc000687988 sp=0xc000687958 pc=0x42d8d2
runtime.mapassign(0xb9d4a0, 0xc0000b1f80, 0xc000687a6c, 0x13bd940)
/usr/lib64/go/1.12/src/runtime/map.go:590 +0x5e3 fp=0xc000687a10 sp=0xc000687988 pc=0x40e2a3
github.com/nspcc-dev/neo-go/pkg/core.(*Blockchain).verifyHashAgainstScript(0xc0000f8000, 0x1300d1f5a3d348be, 0x78607089e4fe9fab, 0xc0a21e4f71, 0xc000d16070, 0x1e2b8881f8178e92, 0xd589cfa965287bb8, 0x24a4e4f292ebc797, 0xb90d3120d878e619, 0xc0001d01e0, ...)
/home/rik/dev/neo-go/pkg/core/blockchain.go:1866 +0x371 fp=0xc000687ad8 sp=0xc000687a10 pc=0x984ed1
github.com/nspcc-dev/neo-go/pkg/core.(*Blockchain).verifyHeaderWitnesses(0xc0000f8000, 0xc000d16000, 0xc00042d950, 0x1, 0xc00008a6c0)
/home/rik/dev/neo-go/pkg/core/blockchain.go:1914 +0x209 fp=0xc000687c90 sp=0xc000687ad8 pc=0x985999
github.com/nspcc-dev/neo-go/pkg/core.(*Blockchain).verifyHeader(0xc0000f8000, 0xc000d16000, 0xc00042d950, 0xf0ee0049d611101f, 0xf8967398a8f9351b)
/home/rik/dev/neo-go/pkg/core/blockchain.go:1254 +0x235 fp=0xc000687d70 sp=0xc000687c90 pc=0x97e4e5
github.com/nspcc-dev/neo-go/pkg/core.(*Blockchain).addHeaders(0xc0000f8000, 0x1, 0xc000d12000, 0x7d0, 0x7d0, 0x0, 0x1)
/home/rik/dev/neo-go/pkg/core/blockchain.go:369 +0x266 fp=0xc000687e68 sp=0xc000687d70 pc=0x973f46
github.com/nspcc-dev/neo-go/pkg/core.(*Blockchain).AddHeaders(0xc0000f8000, 0xc000d12000, 0x7d0, 0x7d0, 0x0, 0xc0003640c0)
/home/rik/dev/neo-go/pkg/core/blockchain.go:338 +0x51 fp=0xc000687eb0 sp=0xc000687e68 pc=0x973cb1
github.com/nspcc-dev/neo-go/pkg/network.(*Server).handleHeadersCmd(0xc00022c140, 0xdd0ca0, 0xc000374000, 0xc00000c020)
/home/rik/dev/neo-go/pkg/network/server.go:431 +0x74 fp=0xc000687fc0 sp=0xc000687eb0 pc=0x9ce5e4
runtime.goexit()
/usr/lib64/go/1.12/src/runtime/asm_amd64.s:1337 +0x1 fp=0xc000687fc8 sp=0xc000687fc0 pc=0x45ca51
created by github.com/nspcc-dev/neo-go/pkg/network.(*Server).handleMessage
/home/rik/dev/neo-go/pkg/network/server.go:705 +0xa4a
...
2020-03-04 20:26:18 +03:00
Roman Khimov
81d89cd502
core: fix SystemFee calculation for Invocation TXes
...
They have it specified right in the transaction. Unfortunately, this little
change rendered invalid our RPC test chain, but I think it became even better
after it, especially given that chain generation is a nice test by itself, so
it should be running as a regular test.
2020-03-04 19:23:23 +03:00
Roman Khimov
ae6edf0601
core: add a check for fractional gas in invocation TXes
...
It's not allowed in C# code and that's reasonable as it's a sysfee essentially
that can only be integer.
2020-03-04 19:23:23 +03:00
Roman Khimov
823798514a
util: rename Int64Value to IntegralValue, add FractionalValue
...
Makes it's interface more clear and allows to easily access fractional part.
2020-03-04 19:23:23 +03:00
Roman Khimov
e41d434a49
*: move all packages from CityOfZion to nspcc-dev
2020-03-03 17:21:42 +03:00
Roman Khimov
c3e73c5b7d
core: don't reverify stale headers in addHeader
...
During networked synchronization we expect there to be a lot of duplicate
headers received and it makes no sense for us reverifying them.
2020-03-03 15:34:03 +03:00
Evgenii Stratonikov
9e9d59b49a
core: set NetworkFee to 0 for Claim and Miner transactions
...
Claim tx have no GAS inputs and a positive output which
can lead to negative network fee.
2020-03-02 18:01:49 +03:00
Evgenii Stratonikov
de5215a564
core: use GetAccountStateOrNew() in (*Blockchain).GetValidators()
...
Target of the transaction output may not yet exist in database.
2020-03-02 18:01:49 +03:00
Evgenii Stratonikov
1b9968df67
core: optimize CalculateClaimable()
...
Because accumulated system fee is stored for every block,
it is easy to calculate sum with just to reads.
2020-03-02 18:01:49 +03:00