Commit graph

309 commits

Author SHA1 Message Date
Roman Khimov
18695e660b mempool: drop RemoveOverCapacity(), handle it right in the Add()
Simplifies things a lot and removes useless code.
2020-02-06 17:50:11 +03:00
Roman Khimov
684cbf5bac mempool: make it almost 100% test-covered 2020-02-06 17:50:11 +03:00
Roman Khimov
6672f4b26f mempool: iterate over slice in GetVerifiedTransactions()
It's more efficient and keeps transactions sorted by priority.
2020-02-06 16:59:20 +03:00
Roman Khimov
e97396e56c mempool: test addition of conflicting tx to the pool 2020-02-06 15:57:20 +03:00
Roman Khimov
1133bbe584 mempool: remove unverified transactions pool
Our mempool only contains valid verified transactions all the time, it never
has any unverified ones. Unverified pool made some sense for quick unverifying
after the new block acceptance (and gradual background reverification), but
reverification needs some non-trivial locking between blockchain and mempool
and internal mempool state locking (reverifying tx and moving it between
unverified and verified pools must be atomic). But our current reverification
is fast enough (and has all the appropriate locks), so bothering with
unverified pool makes little sense.
2020-02-06 15:45:16 +03:00
Roman Khimov
b675903f52 mempool/core: redesign mempool dances on block acceptance
We not only need to remove transactions stored in the block, but also
invalidate some potential double spends caused by these transactions. Usually
new block contains a substantial number of transactions from the pool, so it's
easier to make one pass over it only keeping valid items rather than remove
them one by one and make an additional pass to recheck inputs/witnesses.
2020-02-06 15:45:16 +03:00
Roman Khimov
b567ce86ac mempool: implement insertion to sorted slice
Which is way faster than sort.Sort'ing things all the time.
2020-02-06 15:44:18 +03:00
Roman Khimov
35183b6dba mempool: reverse the order of sorted slice
Chopping off the last element of the slice if way easier than doing it with
the first one.
2020-02-06 15:44:16 +03:00
Roman Khimov
794027a90b mempool: use one slice for both priorities
It doesn't harm as we have transactions naturally ordered by fee anyway and it
makes managing them a little easier. This also makes slices store item itself
instead of pointers to it which reduces the pressure on the memory subsystem.
2020-02-06 15:43:45 +03:00
Roman Khimov
325bea3fa9 mempool: cache Feer invocation results in the item
They shouldn't depend on the chain state and for the same transaction they
should always produce the same result. Thus, it makes no sense recalculating
them over and over again.
2020-02-06 15:41:56 +03:00
Roman Khimov
e01bfeeb4d mempool: remove lock indirection from the Pool
After the f0bb886be3 with all methods of Pool
being pointer-based it makes no sense having this lock as a pointer.
2020-02-06 15:41:56 +03:00
Roman Khimov
a928ad9cfa mempool: make item an internal thing of mempool package
Nobody outside should care about these details, mempool operates on
transactions and that's it.
2020-02-06 15:41:56 +03:00
Roman Khimov
f0e3a31bc8 mempool: fix appending to sorted pools
Appending and not changing the real Items is utterly wrong.
2020-02-06 15:41:56 +03:00
Roman Khimov
f9963cca37 core: short-circuit verifyInputs when there are no inputs 2020-02-06 15:41:56 +03:00
Roman Khimov
70b3839fd0 core/mempool: fix AddBlock and tx pooling concurrency issues
Eliminate races between tx checks and adding them to the mempool, ensure the
chain doesn't change while we're working with the new tx. Ensure only one
block addition attempt could be in progress.
2020-02-06 15:41:52 +03:00
Roman Khimov
f0bb886be3 mempool: make all methods pointer methods
Makes no sense copying the Pool around.
2020-02-04 17:36:11 +03:00
Roman Khimov
388fed06e5 transaction: forbid serializing invalid transactions
Transaction that has no data is invalid and can't be serialized, so throw an
error if someone tries to.
2020-02-03 16:33:03 +03:00
Evgeniy Kulikov
eb59460032
Merge pull request #614 from nspcc-dev/configurable-mempool-size
config: add MemPoolSize configuration parameter
2020-01-23 13:25:27 +03:00
Roman Khimov
d6784e8f45 config: add MemPoolSize configuration parameter
Allow tuning/experimenting with mempool.
2020-01-23 13:10:44 +03:00
Evgenii Stratonikov
303b82cc44 core: set free GAS limit via config 2020-01-22 14:32:21 +03:00
Evgenii Stratonikov
4718375db1 rpc: return amount of consumed GAS in invoke* 2020-01-22 14:31:57 +03:00
Evgenii Stratonikov
4eae55143f core: restrict GAS available for invocation transactions
There are 10 GAS available for free plus any amount of GAS
attached to a transaction.
2020-01-22 14:31:54 +03:00
Roman Khimov
32213b1454
Merge pull request #601 from nspcc-dev/refactoring/core
core: refactor out Block, BlockBase and Header, closes #597.
2020-01-20 16:19:20 +03:00
Roman Khimov
46b82b4fb5 keys: don't return error from PrivateKey.Sign
As it can't ever happen.
2020-01-17 17:00:30 +03:00
Evgenii Stratonikov
28183b81d6 mempool: simplify names of exported types
With the move to a separate package, naming can be simplified:
MemPool -> Pool, PoolItem -> Item, PoolItems -> Items.
2020-01-16 10:16:24 +03:00
Evgenii Stratonikov
fed6fba9b6 core: refactor out MemPool 2020-01-16 10:16:24 +03:00
Evgenii Stratonikov
79bceb3e40 block: use require in tests 2020-01-16 10:16:24 +03:00
Evgenii Stratonikov
489b88afbb block: rename BlockBase to Base 2020-01-16 10:16:24 +03:00
Evgenii Stratonikov
63c56cca5c core: refactor out Block, BlockBase and Header structs
See #597.
2020-01-16 10:16:24 +03:00
Evgenii Stratonikov
9dc5571327 core,rpc: close Blockchain in tests
If blockchain is not closed, logging in defer can occur
after test has finished, which will lead to a panic with
"Log in goroutine after Test* has completed".
2020-01-10 11:47:56 +03:00
Evgenii Stratonikov
637c99eda7 core: simplify newInteropContext 2020-01-10 11:14:29 +03:00
Evgenii Stratonikov
aecdf470e7 cli,pkg: use zap.Logger 2020-01-10 11:14:27 +03:00
Evgenii Stratonikov
9c79684516 storage: panic on error in boltdb.Seek
Error in Seek means something is terribly wrong (e.g. db was not opened) and
error drop is not the right thing to do, because caller
will continue working with the wrong view.
2020-01-09 15:17:35 +03:00
Roman Khimov
c306d2f07c
Merge pull request #581 from nspcc-dev/feat/storage_find
core: implement Neo.Storage.Find interop, closes #422.
2019-12-26 15:46:46 +03:00
Evgenii Stratonikov
48cf4f4a84 core: implement Neo.Storage.Find interop 2019-12-26 15:21:41 +03:00
Roman Khimov
f8887b63ba
Merge pull request #578 from nspcc-dev/fix/sign
consensus: verify payloads correctly, closes #555.
2019-12-26 13:07:40 +03:00
Evgenii Stratonikov
714c466c2c core: add ScriptFromWitness function
Extracting verification script from witness became a common task.
This commit adds such a possibility.
2019-12-26 10:49:56 +03:00
Roman Khimov
9145855d2c
Merge pull request #579 from nspcc-dev/refactor-crypto
This moves some functionality into micro-packages, improves testing,
unexports some code and fixes bugs along the way.
2019-12-25 18:12:50 +03:00
Roman Khimov
b246653f62 address: rename functions as per #579 comments
Make them more clear to understand.
2019-12-25 17:34:18 +03:00
Roman Khimov
e685e9bf9a address: move into its own package
Doesn't really belong to the crypto.
2019-12-25 15:22:02 +03:00
Roman Khimov
ee28fb08f6 crypto: move merkle tree into the hash package
It's all about hashes, so it makes sense putting it there.
2019-12-25 11:28:59 +03:00
Vsevolod Brekelov
7f1844a39e transaction: add unmarshaling for TXType 2019-12-25 01:44:37 +03:00
Evgenii Stratonikov
5bc32b523a vm: implement Neo.Iterator.* interops 2019-12-24 10:21:47 +03:00
Evgenii Stratonikov
3ff7fd5262 vm: implement Neo.Enumerator.* interops 2019-12-24 10:21:46 +03:00
Roman Khimov
bf84e1f2fb core: cache top block
Turns out, our dApps use it a lot and we were going to the DB to get it which
is a useless waste of time. Technically we could also remove blockHeight here,
but not doing it at the moment as it's more involved.

It eliminates this time waste from the pprof graph, but doesn't change 1.4M ->
1.5M 100K mainnet block import test case in any noticeable way.
2019-12-23 19:18:12 +03:00
Roman Khimov
deb76c5199 core: add negative non-interop test for all interops
Try to feed some garbage into these functions.
2019-12-23 16:14:43 +03:00
Roman Khimov
76d69ee5ee core: restore proper interops ordering 2019-12-23 16:14:43 +03:00
Roman Khimov
c287a9e93c core: implement Neo.Witness.GetVerificationScript interop 2019-12-23 16:14:43 +03:00
Roman Khimov
9d638d0fee core: implement Neo.InvocationTransaction.GetScript 2019-12-23 16:14:43 +03:00
Roman Khimov
a7457d08a1 vm/core: add ID support for SYSCALL, redo interop registration
This solves two problems:
 * adds support for shortened SYSCALL form that uses IDs (similar to #434, but
   for NEO 2.0, supporting both forms), which is important for compatibility
   with C# node and mainnet chain that uses it from some height
 * reworks interop plugging to use callbacks rather than appending to the map,
   these map mangling functions are clearly visible in the VM profiling
   statistics and we want spawning a VM to be fast, so it makes sense
   optimizing it. This change moves most of the work to the init() phase
   making VM setup cheaper.

Caveats:
 * InteropNameToID accepts `[]byte` because that's the thing we have in
   SYSCALL processing and that's the most often usecase for it, it leads to
   some conversions in other places but that's acceptable because those are
   either tests or init()
 * three getInterop functions are: `getDefaultVMInterop`, `getSystemInterop`
   and `getNeoInterop`

Our 100K (1.4M->1.5M) block import time improves by ~4% with this change.
2019-12-19 13:35:42 +03:00