2023-11-01 10:32:09 +07:00
|
|
|
# Nimbus
|
aristo: fork support via layers/txframes (#2960)
* aristo: fork support via layers/txframes
This change reorganises how the database is accessed: instead holding a
"current frame" in the database object, a dag of frames is created based
on the "base frame" held in `AristoDbRef` and all database access
happens through this frame, which can be thought of as a consistent
point-in-time snapshot of the database based on a particular fork of the
chain.
In the code, "frame", "transaction" and "layer" is used to denote more
or less the same thing: a dag of stacked changes backed by the on-disk
database.
Although this is not a requirement, in practice each frame holds the
change set of a single block - as such, the frame and its ancestors
leading up to the on-disk state represents the state of the database
after that block has been applied.
"committing" means merging the changes to its parent frame so that the
difference between them is lost and only the cumulative changes remain -
this facility enables frames to be combined arbitrarily wherever they
are in the dag.
In particular, it becomes possible to consolidate a set of changes near
the base of the dag and commit those to disk without having to re-do the
in-memory frames built on top of them - this is useful for "flattening"
a set of changes during a base update and sending those to storage
without having to perform a block replay on top.
Looking at abstractions, a side effect of this change is that the KVT
and Aristo are brought closer together by considering them to be part of
the "same" atomic transaction set - the way the code gets organised,
applying a block and saving it to the kvt happens in the same "logical"
frame - therefore, discarding the frame discards both the aristo and kvt
changes at the same time - likewise, they are persisted to disk together
- this makes reasoning about the database somewhat easier but has the
downside of increased memory usage, something that perhaps will need
addressing in the future.
Because the code reasons more strictly about frames and the state of the
persisted database, it also makes it more visible where ForkedChain
should be used and where it is still missing - in particular, frames
represent a single branch of history while forkedchain manages multiple
parallel forks - user-facing services such as the RPC should use the
latter, ie until it has been finalized, a getBlock request should
consider all forks and not just the blocks in the canonical head branch.
Another advantage of this approach is that `AristoDbRef` conceptually
becomes more simple - removing its tracking of the "current" transaction
stack simplifies reasoning about what can go wrong since this state now
has to be passed around in the form of `AristoTxRef` - as such, many of
the tests and facilities in the code that were dealing with "stack
inconsistency" are now structurally prevented from happening. The test
suite will need significant refactoring after this change.
Once this change has been merged, there are several follow-ups to do:
* there's no mechanism for keeping frames up to date as they get
committed or rolled back - TODO
* naming is confused - many names for the same thing for legacy reason
* forkedchain support is still missing in lots of code
* clean up redundant logic based on previous designs - in particular the
debug and introspection code no longer makes sense
* the way change sets are stored will probably need revisiting - because
it's a stack of changes where each frame must be interrogated to find an
on-disk value, with a base distance of 128 we'll at minimum have to
perform 128 frame lookups for *every* database interaction - regardless,
the "dag-like" nature will stay
* dispose and commit are poorly defined and perhaps redundant - in
theory, one could simply let the GC collect abandoned frames etc, though
it's likely an explicit mechanism will remain useful, so they stay for
now
More about the changes:
* `AristoDbRef` gains a `txRef` field (todo: rename) that "more or less"
corresponds to the old `balancer` field
* `AristoDbRef.stack` is gone - instead, there's a chain of
`AristoTxRef` objects that hold their respective "layer" which has the
actual changes
* No more reasoning about "top" and "stack" - instead, each
`AristoTxRef` can be a "head" that "more or less" corresponds to the old
single-history `top` notion and its stack
* `level` still represents "distance to base" - it's computed from the
parent chain instead of being stored
* one has to be careful not to use frames where forkedchain was intended
- layers are only for a single branch of history!
* fix layer vtop after rollback
* engine fix
* Fix test_txpool
* Fix test_rpc
* Fix copyright year
* fix simulator
* Fix copyright year
* Fix copyright year
* Fix tracer
* Fix infinite recursion bug
* Remove aristo and kvt empty files
* Fic copyright year
* Fix fc chain_kvt
* ForkedChain refactoring
* Fix merge master conflict
* Fix copyright year
* Reparent txFrame
* Fix test
* Fix txFrame reparent again
* Cleanup and fix test
* UpdateBase bugfix and fix test
* Fixe newPayload bug discovered by hive
* Fix engine api fcu
* Clean up call template, chain_kvt, andn txguid
* Fix copyright year
* work around base block loading issue
* Add test
* Fix updateHead bug
* Fix updateBase bug
* Change func commitBase to proc commitBase
* Touch up and fix debug mode crash
---------
Co-authored-by: jangko <jangko128@gmail.com>
2025-02-06 08:04:50 +01:00
|
|
|
# Copyright (c) 2019-2025 Status Research & Development GmbH
|
2023-11-01 10:32:09 +07:00
|
|
|
# Licensed under either of
|
|
|
|
# * Apache License, version 2.0, ([LICENSE-APACHE](LICENSE-APACHE) or
|
|
|
|
# http://www.apache.org/licenses/LICENSE-2.0)
|
|
|
|
# * MIT license ([LICENSE-MIT](LICENSE-MIT) or
|
|
|
|
# http://opensource.org/licenses/MIT)
|
|
|
|
# at your option. This file may not be copied, modified, or distributed except
|
|
|
|
# according to those terms.
|
|
|
|
|
2024-08-01 10:41:20 +00:00
|
|
|
{.push raises: [].}
|
2024-06-18 11:14:02 +00:00
|
|
|
|
2018-12-03 17:54:19 +07:00
|
|
|
import
|
2022-12-02 11:39:12 +07:00
|
|
|
std/[strutils, json],
|
2024-08-01 10:41:20 +00:00
|
|
|
nimcrypto/utils as ncrutils,
|
|
|
|
results,
|
|
|
|
web3/conversions,
|
2024-10-04 16:34:31 +02:00
|
|
|
eth/common/transaction_utils,
|
2024-08-01 10:41:20 +00:00
|
|
|
./beacon/web3_eth_conv,
|
2022-12-02 11:39:12 +07:00
|
|
|
./common/common,
|
2024-08-01 10:41:20 +00:00
|
|
|
./constants,
|
|
|
|
./core/executor,
|
2023-12-12 19:12:56 +00:00
|
|
|
./db/[core_db, ledger],
|
2024-08-01 10:41:20 +00:00
|
|
|
./evm/[code_bytes, state, types],
|
2023-08-02 17:17:40 +07:00
|
|
|
./evm/tracer/legacy_tracer,
|
2024-06-17 14:56:39 +07:00
|
|
|
./transaction,
|
2024-08-01 10:41:20 +00:00
|
|
|
./utils/utils
|
2018-12-03 23:22:08 +07:00
|
|
|
|
2024-08-01 10:41:20 +00:00
|
|
|
when not CoreDbEnableCaptJournal:
|
|
|
|
{.error: "Compiler flag missing for tracer, try -d:dbjapi_enabled".}
|
2018-12-03 23:22:08 +07:00
|
|
|
|
2024-03-22 17:31:56 +00:00
|
|
|
type
|
2024-08-01 10:41:20 +00:00
|
|
|
CaptCtxRef = ref object
|
|
|
|
db: CoreDbRef # not `nil`
|
2024-10-01 21:03:10 +00:00
|
|
|
root: common.Hash32
|
2024-08-01 10:41:20 +00:00
|
|
|
ctx: CoreDbCtxRef # not `nil`
|
|
|
|
cpt: CoreDbCaptRef # not `nil`
|
|
|
|
restore: CoreDbCtxRef # `nil` unless `ctx` activated
|
2024-03-22 17:31:56 +00:00
|
|
|
|
2024-08-01 10:41:20 +00:00
|
|
|
const
|
|
|
|
senderName = "sender"
|
|
|
|
recipientName = "recipient"
|
|
|
|
minerName = "miner"
|
|
|
|
uncleName = "uncle"
|
|
|
|
internalTxName = "internalTx"
|
2024-03-21 10:45:57 +00:00
|
|
|
|
2024-08-01 10:41:20 +00:00
|
|
|
proc dumpMemoryDB*(node: JsonNode, cpt: CoreDbCaptRef) {.gcsafe.}
|
|
|
|
proc toJson*(receipts: seq[Receipt]): JsonNode {.gcsafe.}
|
|
|
|
|
|
|
|
# ------------------------------------------------------------------------------
|
|
|
|
# Private helpers
|
|
|
|
# ------------------------------------------------------------------------------
|
|
|
|
|
|
|
|
proc init(
|
|
|
|
T: type CaptCtxRef;
|
|
|
|
com: CommonRef;
|
2024-10-01 21:03:10 +00:00
|
|
|
root: common.Hash32;
|
2024-09-10 05:24:45 +00:00
|
|
|
): T =
|
2024-08-01 10:41:20 +00:00
|
|
|
let ctx = block:
|
2024-12-18 11:56:46 +01:00
|
|
|
when false:
|
|
|
|
let rc = com.db.ctx.newCtxByKey(root)
|
|
|
|
if rc.isErr:
|
|
|
|
raiseAssert "newCptCtx: " & $$rc.error
|
|
|
|
rc.value
|
|
|
|
else:
|
|
|
|
{.warning: "TODO make a temporary context? newCtxByKey has been obsoleted".}
|
|
|
|
com.db.ctx
|
2024-08-01 10:41:20 +00:00
|
|
|
T(db: com.db, root: root, cpt: com.db.pushCapture(), ctx: ctx)
|
|
|
|
|
|
|
|
proc init(
|
|
|
|
T: type CaptCtxRef;
|
|
|
|
com: CommonRef;
|
2024-10-01 21:03:10 +00:00
|
|
|
topHeader: Header;
|
2024-11-07 08:24:21 +07:00
|
|
|
): T =
|
aristo: fork support via layers/txframes (#2960)
* aristo: fork support via layers/txframes
This change reorganises how the database is accessed: instead holding a
"current frame" in the database object, a dag of frames is created based
on the "base frame" held in `AristoDbRef` and all database access
happens through this frame, which can be thought of as a consistent
point-in-time snapshot of the database based on a particular fork of the
chain.
In the code, "frame", "transaction" and "layer" is used to denote more
or less the same thing: a dag of stacked changes backed by the on-disk
database.
Although this is not a requirement, in practice each frame holds the
change set of a single block - as such, the frame and its ancestors
leading up to the on-disk state represents the state of the database
after that block has been applied.
"committing" means merging the changes to its parent frame so that the
difference between them is lost and only the cumulative changes remain -
this facility enables frames to be combined arbitrarily wherever they
are in the dag.
In particular, it becomes possible to consolidate a set of changes near
the base of the dag and commit those to disk without having to re-do the
in-memory frames built on top of them - this is useful for "flattening"
a set of changes during a base update and sending those to storage
without having to perform a block replay on top.
Looking at abstractions, a side effect of this change is that the KVT
and Aristo are brought closer together by considering them to be part of
the "same" atomic transaction set - the way the code gets organised,
applying a block and saving it to the kvt happens in the same "logical"
frame - therefore, discarding the frame discards both the aristo and kvt
changes at the same time - likewise, they are persisted to disk together
- this makes reasoning about the database somewhat easier but has the
downside of increased memory usage, something that perhaps will need
addressing in the future.
Because the code reasons more strictly about frames and the state of the
persisted database, it also makes it more visible where ForkedChain
should be used and where it is still missing - in particular, frames
represent a single branch of history while forkedchain manages multiple
parallel forks - user-facing services such as the RPC should use the
latter, ie until it has been finalized, a getBlock request should
consider all forks and not just the blocks in the canonical head branch.
Another advantage of this approach is that `AristoDbRef` conceptually
becomes more simple - removing its tracking of the "current" transaction
stack simplifies reasoning about what can go wrong since this state now
has to be passed around in the form of `AristoTxRef` - as such, many of
the tests and facilities in the code that were dealing with "stack
inconsistency" are now structurally prevented from happening. The test
suite will need significant refactoring after this change.
Once this change has been merged, there are several follow-ups to do:
* there's no mechanism for keeping frames up to date as they get
committed or rolled back - TODO
* naming is confused - many names for the same thing for legacy reason
* forkedchain support is still missing in lots of code
* clean up redundant logic based on previous designs - in particular the
debug and introspection code no longer makes sense
* the way change sets are stored will probably need revisiting - because
it's a stack of changes where each frame must be interrogated to find an
on-disk value, with a base distance of 128 we'll at minimum have to
perform 128 frame lookups for *every* database interaction - regardless,
the "dag-like" nature will stay
* dispose and commit are poorly defined and perhaps redundant - in
theory, one could simply let the GC collect abandoned frames etc, though
it's likely an explicit mechanism will remain useful, so they stay for
now
More about the changes:
* `AristoDbRef` gains a `txRef` field (todo: rename) that "more or less"
corresponds to the old `balancer` field
* `AristoDbRef.stack` is gone - instead, there's a chain of
`AristoTxRef` objects that hold their respective "layer" which has the
actual changes
* No more reasoning about "top" and "stack" - instead, each
`AristoTxRef` can be a "head" that "more or less" corresponds to the old
single-history `top` notion and its stack
* `level` still represents "distance to base" - it's computed from the
parent chain instead of being stored
* one has to be careful not to use frames where forkedchain was intended
- layers are only for a single branch of history!
* fix layer vtop after rollback
* engine fix
* Fix test_txpool
* Fix test_rpc
* Fix copyright year
* fix simulator
* Fix copyright year
* Fix copyright year
* Fix tracer
* Fix infinite recursion bug
* Remove aristo and kvt empty files
* Fic copyright year
* Fix fc chain_kvt
* ForkedChain refactoring
* Fix merge master conflict
* Fix copyright year
* Reparent txFrame
* Fix test
* Fix txFrame reparent again
* Cleanup and fix test
* UpdateBase bugfix and fix test
* Fixe newPayload bug discovered by hive
* Fix engine api fcu
* Clean up call template, chain_kvt, andn txguid
* Fix copyright year
* work around base block loading issue
* Add test
* Fix updateHead bug
* Fix updateBase bug
* Change func commitBase to proc commitBase
* Touch up and fix debug mode crash
---------
Co-authored-by: jangko <jangko128@gmail.com>
2025-02-06 08:04:50 +01:00
|
|
|
let header = com.db.baseTxFrame().getBlockHeader(topHeader.parentHash).expect("top header parent exists")
|
2024-11-07 08:24:21 +07:00
|
|
|
T.init(com, header.stateRoot)
|
2024-08-01 10:41:20 +00:00
|
|
|
|
|
|
|
proc activate(cc: CaptCtxRef): CaptCtxRef {.discardable.} =
|
|
|
|
## Install/activate new context `cc.ctx`, old one in `cc.restore`
|
|
|
|
doAssert not cc.isNil
|
|
|
|
doAssert cc.restore.isNil # otherwise activated, already
|
2024-12-18 11:56:46 +01:00
|
|
|
if true:
|
|
|
|
raiseAssert "TODO activte context"
|
|
|
|
# cc.restore = cc.ctx.swapCtx cc.db
|
2024-08-01 10:41:20 +00:00
|
|
|
cc
|
|
|
|
|
|
|
|
proc release(cc: CaptCtxRef) =
|
2024-12-18 11:56:46 +01:00
|
|
|
# if not cc.restore.isNil: # switch to original context (if any)
|
|
|
|
# let ctx = cc.restore.swapCtx(cc.db)
|
|
|
|
# doAssert ctx == cc.ctx
|
|
|
|
if true:
|
|
|
|
raiseAssert "TODO release context"
|
|
|
|
# cc.ctx.forget() # dispose
|
2024-08-01 10:41:20 +00:00
|
|
|
cc.cpt.pop() # discard top layer of actions tracer
|
|
|
|
|
|
|
|
# -------------------
|
2024-03-21 10:45:57 +00:00
|
|
|
|
2024-09-29 14:37:09 +02:00
|
|
|
proc `%`(x: addresses.Address|Bytes32|Bytes256|Hash32): JsonNode =
|
|
|
|
result = %toHex(x)
|
2018-12-25 17:31:51 +07:00
|
|
|
|
|
|
|
proc toJson(receipt: Receipt): JsonNode =
|
|
|
|
result = newJObject()
|
|
|
|
|
|
|
|
result["cumulativeGasUsed"] = %receipt.cumulativeGasUsed
|
2024-06-14 14:31:08 +07:00
|
|
|
result["bloom"] = %receipt.logsBloom
|
2018-12-25 17:31:51 +07:00
|
|
|
result["logs"] = %receipt.logs
|
|
|
|
|
|
|
|
if receipt.hasStateRoot:
|
2024-09-29 14:37:09 +02:00
|
|
|
result["root"] = %(receipt.stateRoot.toHex)
|
2018-12-25 17:31:51 +07:00
|
|
|
else:
|
|
|
|
result["status"] = %receipt.status
|
|
|
|
|
2024-08-01 10:41:20 +00:00
|
|
|
proc dumpReceiptsImpl(
|
aristo: fork support via layers/txframes (#2960)
* aristo: fork support via layers/txframes
This change reorganises how the database is accessed: instead holding a
"current frame" in the database object, a dag of frames is created based
on the "base frame" held in `AristoDbRef` and all database access
happens through this frame, which can be thought of as a consistent
point-in-time snapshot of the database based on a particular fork of the
chain.
In the code, "frame", "transaction" and "layer" is used to denote more
or less the same thing: a dag of stacked changes backed by the on-disk
database.
Although this is not a requirement, in practice each frame holds the
change set of a single block - as such, the frame and its ancestors
leading up to the on-disk state represents the state of the database
after that block has been applied.
"committing" means merging the changes to its parent frame so that the
difference between them is lost and only the cumulative changes remain -
this facility enables frames to be combined arbitrarily wherever they
are in the dag.
In particular, it becomes possible to consolidate a set of changes near
the base of the dag and commit those to disk without having to re-do the
in-memory frames built on top of them - this is useful for "flattening"
a set of changes during a base update and sending those to storage
without having to perform a block replay on top.
Looking at abstractions, a side effect of this change is that the KVT
and Aristo are brought closer together by considering them to be part of
the "same" atomic transaction set - the way the code gets organised,
applying a block and saving it to the kvt happens in the same "logical"
frame - therefore, discarding the frame discards both the aristo and kvt
changes at the same time - likewise, they are persisted to disk together
- this makes reasoning about the database somewhat easier but has the
downside of increased memory usage, something that perhaps will need
addressing in the future.
Because the code reasons more strictly about frames and the state of the
persisted database, it also makes it more visible where ForkedChain
should be used and where it is still missing - in particular, frames
represent a single branch of history while forkedchain manages multiple
parallel forks - user-facing services such as the RPC should use the
latter, ie until it has been finalized, a getBlock request should
consider all forks and not just the blocks in the canonical head branch.
Another advantage of this approach is that `AristoDbRef` conceptually
becomes more simple - removing its tracking of the "current" transaction
stack simplifies reasoning about what can go wrong since this state now
has to be passed around in the form of `AristoTxRef` - as such, many of
the tests and facilities in the code that were dealing with "stack
inconsistency" are now structurally prevented from happening. The test
suite will need significant refactoring after this change.
Once this change has been merged, there are several follow-ups to do:
* there's no mechanism for keeping frames up to date as they get
committed or rolled back - TODO
* naming is confused - many names for the same thing for legacy reason
* forkedchain support is still missing in lots of code
* clean up redundant logic based on previous designs - in particular the
debug and introspection code no longer makes sense
* the way change sets are stored will probably need revisiting - because
it's a stack of changes where each frame must be interrogated to find an
on-disk value, with a base distance of 128 we'll at minimum have to
perform 128 frame lookups for *every* database interaction - regardless,
the "dag-like" nature will stay
* dispose and commit are poorly defined and perhaps redundant - in
theory, one could simply let the GC collect abandoned frames etc, though
it's likely an explicit mechanism will remain useful, so they stay for
now
More about the changes:
* `AristoDbRef` gains a `txRef` field (todo: rename) that "more or less"
corresponds to the old `balancer` field
* `AristoDbRef.stack` is gone - instead, there's a chain of
`AristoTxRef` objects that hold their respective "layer" which has the
actual changes
* No more reasoning about "top" and "stack" - instead, each
`AristoTxRef` can be a "head" that "more or less" corresponds to the old
single-history `top` notion and its stack
* `level` still represents "distance to base" - it's computed from the
parent chain instead of being stored
* one has to be careful not to use frames where forkedchain was intended
- layers are only for a single branch of history!
* fix layer vtop after rollback
* engine fix
* Fix test_txpool
* Fix test_rpc
* Fix copyright year
* fix simulator
* Fix copyright year
* Fix copyright year
* Fix tracer
* Fix infinite recursion bug
* Remove aristo and kvt empty files
* Fic copyright year
* Fix fc chain_kvt
* ForkedChain refactoring
* Fix merge master conflict
* Fix copyright year
* Reparent txFrame
* Fix test
* Fix txFrame reparent again
* Cleanup and fix test
* UpdateBase bugfix and fix test
* Fixe newPayload bug discovered by hive
* Fix engine api fcu
* Clean up call template, chain_kvt, andn txguid
* Fix copyright year
* work around base block loading issue
* Add test
* Fix updateHead bug
* Fix updateBase bug
* Change func commitBase to proc commitBase
* Touch up and fix debug mode crash
---------
Co-authored-by: jangko <jangko128@gmail.com>
2025-02-06 08:04:50 +01:00
|
|
|
chainDB: CoreDbTxRef;
|
2024-10-01 21:03:10 +00:00
|
|
|
header: Header;
|
2024-11-07 08:24:21 +07:00
|
|
|
): JsonNode =
|
2018-12-25 17:31:51 +07:00
|
|
|
result = newJArray()
|
2024-11-07 08:24:21 +07:00
|
|
|
let receiptList = chainDB.getReceipts(header.receiptsRoot).
|
|
|
|
expect("receipts exists")
|
|
|
|
for receipt in receiptList:
|
2018-12-25 17:31:51 +07:00
|
|
|
result.add receipt.toJson
|
|
|
|
|
2024-08-01 10:41:20 +00:00
|
|
|
# ------------------------------------------------------------------------------
|
|
|
|
# Private functions
|
|
|
|
# ------------------------------------------------------------------------------
|
2018-12-03 17:54:19 +07:00
|
|
|
|
2024-08-01 10:41:20 +00:00
|
|
|
proc captureAccount(
|
|
|
|
n: JsonNode;
|
|
|
|
db: LedgerRef;
|
2024-10-16 13:51:38 +07:00
|
|
|
address: Address;
|
2024-08-01 10:41:20 +00:00
|
|
|
name: string;
|
|
|
|
) =
|
2018-12-12 10:40:37 +07:00
|
|
|
var jaccount = newJObject()
|
|
|
|
jaccount["name"] = %name
|
2024-09-29 14:37:09 +02:00
|
|
|
jaccount["address"] = %(address.to0xHex)
|
2020-06-01 13:45:32 +07:00
|
|
|
|
|
|
|
let nonce = db.getNonce(address)
|
|
|
|
let balance = db.getBalance(address)
|
|
|
|
let codeHash = db.getCodeHash(address)
|
|
|
|
let storageRoot = db.getStorageRoot(address)
|
|
|
|
|
2024-01-13 08:41:57 +07:00
|
|
|
jaccount["nonce"] = %(conversions.`$`(nonce.Web3Quantity))
|
2020-06-01 13:45:32 +07:00
|
|
|
jaccount["balance"] = %("0x" & balance.toHex)
|
2018-12-11 16:53:05 +07:00
|
|
|
|
|
|
|
let code = db.getCode(address)
|
2024-09-29 14:37:09 +02:00
|
|
|
jaccount["codeHash"] = %(codeHash.to0xHex)
|
2024-08-01 10:41:20 +00:00
|
|
|
jaccount["code"] = %("0x" & code.bytes.toHex(true))
|
2024-09-29 14:37:09 +02:00
|
|
|
jaccount["storageRoot"] = %(storageRoot.to0xHex)
|
2018-12-11 16:53:05 +07:00
|
|
|
|
|
|
|
var storage = newJObject()
|
|
|
|
for key, value in db.storage(address):
|
2019-01-11 13:53:18 +07:00
|
|
|
storage["0x" & key.dumpHex] = %("0x" & value.dumpHex)
|
2018-12-12 10:40:37 +07:00
|
|
|
jaccount["storage"] = storage
|
2018-12-11 16:53:05 +07:00
|
|
|
|
2018-12-12 10:40:37 +07:00
|
|
|
n.add jaccount
|
2018-12-11 16:53:05 +07:00
|
|
|
|
2018-12-12 22:18:46 +07:00
|
|
|
|
2024-08-01 10:41:20 +00:00
|
|
|
proc traceTransactionImpl(
|
|
|
|
com: CommonRef;
|
2024-10-01 21:03:10 +00:00
|
|
|
header: Header;
|
2024-08-01 10:41:20 +00:00
|
|
|
transactions: openArray[Transaction];
|
|
|
|
txIndex: uint64;
|
|
|
|
tracerFlags: set[TracerFlags] = {};
|
2024-11-07 08:24:21 +07:00
|
|
|
): JsonNode =
|
2024-08-01 10:41:20 +00:00
|
|
|
if header.txRoot == EMPTY_ROOT_HASH:
|
|
|
|
return newJNull()
|
2024-03-07 19:24:05 +00:00
|
|
|
|
2018-12-03 23:22:08 +07:00
|
|
|
let
|
2023-08-02 17:17:40 +07:00
|
|
|
tracerInst = newLegacyTracer(tracerFlags)
|
2024-08-01 10:41:20 +00:00
|
|
|
cc = activate CaptCtxRef.init(com, header)
|
aristo: fork support via layers/txframes (#2960)
* aristo: fork support via layers/txframes
This change reorganises how the database is accessed: instead holding a
"current frame" in the database object, a dag of frames is created based
on the "base frame" held in `AristoDbRef` and all database access
happens through this frame, which can be thought of as a consistent
point-in-time snapshot of the database based on a particular fork of the
chain.
In the code, "frame", "transaction" and "layer" is used to denote more
or less the same thing: a dag of stacked changes backed by the on-disk
database.
Although this is not a requirement, in practice each frame holds the
change set of a single block - as such, the frame and its ancestors
leading up to the on-disk state represents the state of the database
after that block has been applied.
"committing" means merging the changes to its parent frame so that the
difference between them is lost and only the cumulative changes remain -
this facility enables frames to be combined arbitrarily wherever they
are in the dag.
In particular, it becomes possible to consolidate a set of changes near
the base of the dag and commit those to disk without having to re-do the
in-memory frames built on top of them - this is useful for "flattening"
a set of changes during a base update and sending those to storage
without having to perform a block replay on top.
Looking at abstractions, a side effect of this change is that the KVT
and Aristo are brought closer together by considering them to be part of
the "same" atomic transaction set - the way the code gets organised,
applying a block and saving it to the kvt happens in the same "logical"
frame - therefore, discarding the frame discards both the aristo and kvt
changes at the same time - likewise, they are persisted to disk together
- this makes reasoning about the database somewhat easier but has the
downside of increased memory usage, something that perhaps will need
addressing in the future.
Because the code reasons more strictly about frames and the state of the
persisted database, it also makes it more visible where ForkedChain
should be used and where it is still missing - in particular, frames
represent a single branch of history while forkedchain manages multiple
parallel forks - user-facing services such as the RPC should use the
latter, ie until it has been finalized, a getBlock request should
consider all forks and not just the blocks in the canonical head branch.
Another advantage of this approach is that `AristoDbRef` conceptually
becomes more simple - removing its tracking of the "current" transaction
stack simplifies reasoning about what can go wrong since this state now
has to be passed around in the form of `AristoTxRef` - as such, many of
the tests and facilities in the code that were dealing with "stack
inconsistency" are now structurally prevented from happening. The test
suite will need significant refactoring after this change.
Once this change has been merged, there are several follow-ups to do:
* there's no mechanism for keeping frames up to date as they get
committed or rolled back - TODO
* naming is confused - many names for the same thing for legacy reason
* forkedchain support is still missing in lots of code
* clean up redundant logic based on previous designs - in particular the
debug and introspection code no longer makes sense
* the way change sets are stored will probably need revisiting - because
it's a stack of changes where each frame must be interrogated to find an
on-disk value, with a base distance of 128 we'll at minimum have to
perform 128 frame lookups for *every* database interaction - regardless,
the "dag-like" nature will stay
* dispose and commit are poorly defined and perhaps redundant - in
theory, one could simply let the GC collect abandoned frames etc, though
it's likely an explicit mechanism will remain useful, so they stay for
now
More about the changes:
* `AristoDbRef` gains a `txRef` field (todo: rename) that "more or less"
corresponds to the old `balancer` field
* `AristoDbRef.stack` is gone - instead, there's a chain of
`AristoTxRef` objects that hold their respective "layer" which has the
actual changes
* No more reasoning about "top" and "stack" - instead, each
`AristoTxRef` can be a "head" that "more or less" corresponds to the old
single-history `top` notion and its stack
* `level` still represents "distance to base" - it's computed from the
parent chain instead of being stored
* one has to be careful not to use frames where forkedchain was intended
- layers are only for a single branch of history!
* fix layer vtop after rollback
* engine fix
* Fix test_txpool
* Fix test_rpc
* Fix copyright year
* fix simulator
* Fix copyright year
* Fix copyright year
* Fix tracer
* Fix infinite recursion bug
* Remove aristo and kvt empty files
* Fic copyright year
* Fix fc chain_kvt
* ForkedChain refactoring
* Fix merge master conflict
* Fix copyright year
* Reparent txFrame
* Fix test
* Fix txFrame reparent again
* Cleanup and fix test
* UpdateBase bugfix and fix test
* Fixe newPayload bug discovered by hive
* Fix engine api fcu
* Clean up call template, chain_kvt, andn txguid
* Fix copyright year
* work around base block loading issue
* Add test
* Fix updateHead bug
* Fix updateBase bug
* Change func commitBase to proc commitBase
* Touch up and fix debug mode crash
---------
Co-authored-by: jangko <jangko128@gmail.com>
2025-02-06 08:04:50 +01:00
|
|
|
txFrame = com.db.baseTxFrame()
|
|
|
|
parent = txFrame.getBlockHeader(header.parentHash).valueOr:
|
|
|
|
return newJNull()
|
|
|
|
vmState = BaseVMState.new(parent, header, com, txFrame, storeSlotHash = true)
|
2024-12-21 20:46:13 +07:00
|
|
|
ledger = vmState.ledger
|
2024-06-07 15:24:32 +07:00
|
|
|
|
2024-08-01 10:41:20 +00:00
|
|
|
defer: cc.release()
|
2018-12-26 10:34:16 +07:00
|
|
|
|
Consolidate block type for block processing (#2325)
This PR consolidates the split header-body sequences into a single EthBlock
sequence and cleans up the fallout from that which significantly reduces
block processing overhead during import thanks to less garbage collection
and fewer copies of things all around.
Notably, since the number of headers must always match the number of bodies,
we also get rid of a pointless degree of freedom that in the future could
introduce unnecessary bugs.
* only read header and body from era file
* avoid several unnecessary copies along the block processing way
* simplify signatures, cleaning up unused arguemnts and returns
* use `stew/assign2` in a few strategic places where the generated
nim assignent is slow and add a few `move` to work around poor
analysis in nim 1.6 (will need to be revisited for 2.0)
```
stats-20240607_2223-a814aa0b.csv vs stats-20240608_0714-21c1d0a9.csv
bps_x bps_y tps_x tps_y bpsd tpsd timed
block_number
(498305, 713245] 1,540.52 1,809.73 2,361.58 2775.340189 17.63% 17.63% -14.92%
(713245, 928185] 730.36 865.26 1,715.90 2028.973852 18.01% 18.01% -15.21%
(928185, 1143126] 663.03 789.10 2,529.26 3032.490771 19.79% 19.79% -16.28%
(1143126, 1358066] 393.46 508.05 2,152.50 2777.578119 29.13% 29.13% -22.50%
(1358066, 1573007] 370.88 440.72 2,351.31 2791.896052 18.81% 18.81% -15.80%
(1573007, 1787947] 283.65 335.11 2,068.93 2441.373402 17.60% 17.60% -14.91%
(1787947, 2002888] 287.29 342.11 2,078.39 2474.179448 18.99% 18.99% -15.91%
(2002888, 2217828] 293.38 343.16 2,208.83 2584.77457 17.16% 17.16% -14.61%
(2217828, 2432769] 140.09 167.86 1,081.87 1296.336926 18.82% 18.82% -15.80%
blocks: 1934464, baseline: 3h13m1s, contender: 2h43m47s
bpsd (mean): 19.55%
tpsd (mean): 19.55%
Time (total): -29m13s, -15.14%
```
2024-06-09 16:32:20 +02:00
|
|
|
doAssert(transactions.calcTxRoot == header.txRoot)
|
|
|
|
doAssert(transactions.len != 0)
|
2018-12-03 17:54:19 +07:00
|
|
|
|
2018-12-11 16:53:05 +07:00
|
|
|
var
|
|
|
|
gasUsed: GasInt
|
2018-12-12 16:41:18 +07:00
|
|
|
before = newJArray()
|
|
|
|
after = newJArray()
|
2018-12-11 16:53:05 +07:00
|
|
|
stateDiff = %{"before": before, "after": after}
|
2024-08-01 10:41:20 +00:00
|
|
|
stateCtx = CaptCtxRef(nil)
|
2018-12-11 16:53:05 +07:00
|
|
|
|
2020-06-22 07:48:23 +07:00
|
|
|
let
|
|
|
|
miner = vmState.coinbase()
|
2019-04-23 19:50:45 +07:00
|
|
|
|
Consolidate block type for block processing (#2325)
This PR consolidates the split header-body sequences into a single EthBlock
sequence and cleans up the fallout from that which significantly reduces
block processing overhead during import thanks to less garbage collection
and fewer copies of things all around.
Notably, since the number of headers must always match the number of bodies,
we also get rid of a pointless degree of freedom that in the future could
introduce unnecessary bugs.
* only read header and body from era file
* avoid several unnecessary copies along the block processing way
* simplify signatures, cleaning up unused arguemnts and returns
* use `stew/assign2` in a few strategic places where the generated
nim assignent is slow and add a few `move` to work around poor
analysis in nim 1.6 (will need to be revisited for 2.0)
```
stats-20240607_2223-a814aa0b.csv vs stats-20240608_0714-21c1d0a9.csv
bps_x bps_y tps_x tps_y bpsd tpsd timed
block_number
(498305, 713245] 1,540.52 1,809.73 2,361.58 2775.340189 17.63% 17.63% -14.92%
(713245, 928185] 730.36 865.26 1,715.90 2028.973852 18.01% 18.01% -15.21%
(928185, 1143126] 663.03 789.10 2,529.26 3032.490771 19.79% 19.79% -16.28%
(1143126, 1358066] 393.46 508.05 2,152.50 2777.578119 29.13% 29.13% -22.50%
(1358066, 1573007] 370.88 440.72 2,351.31 2791.896052 18.81% 18.81% -15.80%
(1573007, 1787947] 283.65 335.11 2,068.93 2441.373402 17.60% 17.60% -14.91%
(1787947, 2002888] 287.29 342.11 2,078.39 2474.179448 18.99% 18.99% -15.91%
(2002888, 2217828] 293.38 343.16 2,208.83 2584.77457 17.16% 17.16% -14.61%
(2217828, 2432769] 140.09 167.86 1,081.87 1296.336926 18.82% 18.82% -15.80%
blocks: 1934464, baseline: 3h13m1s, contender: 2h43m47s
bpsd (mean): 19.55%
tpsd (mean): 19.55%
Time (total): -29m13s, -15.14%
```
2024-06-09 16:32:20 +02:00
|
|
|
for idx, tx in transactions:
|
2024-10-04 16:34:31 +02:00
|
|
|
let sender = tx.recoverSender().expect("valid signature")
|
2021-05-04 13:27:18 +07:00
|
|
|
let recipient = tx.getRecipient(sender)
|
2018-12-11 16:53:05 +07:00
|
|
|
|
2024-06-14 14:31:08 +07:00
|
|
|
if idx.uint64 == txIndex:
|
2023-08-02 17:17:40 +07:00
|
|
|
vmState.tracer = tracerInst # only enable tracer on target tx
|
2024-12-21 20:46:13 +07:00
|
|
|
before.captureAccount(ledger, sender, senderName)
|
|
|
|
before.captureAccount(ledger, recipient, recipientName)
|
|
|
|
before.captureAccount(ledger, miner, minerName)
|
|
|
|
ledger.persist()
|
|
|
|
stateDiff["beforeRoot"] = %(ledger.getStateRoot().toHex)
|
|
|
|
stateCtx = CaptCtxRef.init(com, ledger.getStateRoot())
|
2018-12-11 16:53:05 +07:00
|
|
|
|
2022-01-10 09:04:06 +00:00
|
|
|
let rc = vmState.processTransaction(tx, sender, header)
|
2022-04-08 11:54:11 +07:00
|
|
|
gasUsed = if rc.isOk: rc.value else: 0
|
2018-12-12 10:40:37 +07:00
|
|
|
|
2024-06-14 14:31:08 +07:00
|
|
|
if idx.uint64 == txIndex:
|
2024-12-21 20:46:13 +07:00
|
|
|
after.captureAccount(ledger, sender, senderName)
|
|
|
|
after.captureAccount(ledger, recipient, recipientName)
|
|
|
|
after.captureAccount(ledger, miner, minerName)
|
2023-08-02 17:17:40 +07:00
|
|
|
tracerInst.removeTracedAccounts(sender, recipient, miner)
|
2024-12-21 20:46:13 +07:00
|
|
|
ledger.persist()
|
|
|
|
stateDiff["afterRoot"] = %(ledger.getStateRoot().toHex)
|
2018-12-12 10:40:37 +07:00
|
|
|
break
|
2018-12-03 17:54:19 +07:00
|
|
|
|
2018-12-25 14:31:12 +07:00
|
|
|
# internal transactions:
|
2024-03-22 17:31:56 +00:00
|
|
|
let
|
2024-08-01 10:41:20 +00:00
|
|
|
cx = activate stateCtx
|
aristo: fork support via layers/txframes (#2960)
* aristo: fork support via layers/txframes
This change reorganises how the database is accessed: instead holding a
"current frame" in the database object, a dag of frames is created based
on the "base frame" held in `AristoDbRef` and all database access
happens through this frame, which can be thought of as a consistent
point-in-time snapshot of the database based on a particular fork of the
chain.
In the code, "frame", "transaction" and "layer" is used to denote more
or less the same thing: a dag of stacked changes backed by the on-disk
database.
Although this is not a requirement, in practice each frame holds the
change set of a single block - as such, the frame and its ancestors
leading up to the on-disk state represents the state of the database
after that block has been applied.
"committing" means merging the changes to its parent frame so that the
difference between them is lost and only the cumulative changes remain -
this facility enables frames to be combined arbitrarily wherever they
are in the dag.
In particular, it becomes possible to consolidate a set of changes near
the base of the dag and commit those to disk without having to re-do the
in-memory frames built on top of them - this is useful for "flattening"
a set of changes during a base update and sending those to storage
without having to perform a block replay on top.
Looking at abstractions, a side effect of this change is that the KVT
and Aristo are brought closer together by considering them to be part of
the "same" atomic transaction set - the way the code gets organised,
applying a block and saving it to the kvt happens in the same "logical"
frame - therefore, discarding the frame discards both the aristo and kvt
changes at the same time - likewise, they are persisted to disk together
- this makes reasoning about the database somewhat easier but has the
downside of increased memory usage, something that perhaps will need
addressing in the future.
Because the code reasons more strictly about frames and the state of the
persisted database, it also makes it more visible where ForkedChain
should be used and where it is still missing - in particular, frames
represent a single branch of history while forkedchain manages multiple
parallel forks - user-facing services such as the RPC should use the
latter, ie until it has been finalized, a getBlock request should
consider all forks and not just the blocks in the canonical head branch.
Another advantage of this approach is that `AristoDbRef` conceptually
becomes more simple - removing its tracking of the "current" transaction
stack simplifies reasoning about what can go wrong since this state now
has to be passed around in the form of `AristoTxRef` - as such, many of
the tests and facilities in the code that were dealing with "stack
inconsistency" are now structurally prevented from happening. The test
suite will need significant refactoring after this change.
Once this change has been merged, there are several follow-ups to do:
* there's no mechanism for keeping frames up to date as they get
committed or rolled back - TODO
* naming is confused - many names for the same thing for legacy reason
* forkedchain support is still missing in lots of code
* clean up redundant logic based on previous designs - in particular the
debug and introspection code no longer makes sense
* the way change sets are stored will probably need revisiting - because
it's a stack of changes where each frame must be interrogated to find an
on-disk value, with a base distance of 128 we'll at minimum have to
perform 128 frame lookups for *every* database interaction - regardless,
the "dag-like" nature will stay
* dispose and commit are poorly defined and perhaps redundant - in
theory, one could simply let the GC collect abandoned frames etc, though
it's likely an explicit mechanism will remain useful, so they stay for
now
More about the changes:
* `AristoDbRef` gains a `txRef` field (todo: rename) that "more or less"
corresponds to the old `balancer` field
* `AristoDbRef.stack` is gone - instead, there's a chain of
`AristoTxRef` objects that hold their respective "layer" which has the
actual changes
* No more reasoning about "top" and "stack" - instead, each
`AristoTxRef` can be a "head" that "more or less" corresponds to the old
single-history `top` notion and its stack
* `level` still represents "distance to base" - it's computed from the
parent chain instead of being stored
* one has to be careful not to use frames where forkedchain was intended
- layers are only for a single branch of history!
* fix layer vtop after rollback
* engine fix
* Fix test_txpool
* Fix test_rpc
* Fix copyright year
* fix simulator
* Fix copyright year
* Fix copyright year
* Fix tracer
* Fix infinite recursion bug
* Remove aristo and kvt empty files
* Fic copyright year
* Fix fc chain_kvt
* ForkedChain refactoring
* Fix merge master conflict
* Fix copyright year
* Reparent txFrame
* Fix test
* Fix txFrame reparent again
* Cleanup and fix test
* UpdateBase bugfix and fix test
* Fixe newPayload bug discovered by hive
* Fix engine api fcu
* Clean up call template, chain_kvt, andn txguid
* Fix copyright year
* work around base block loading issue
* Add test
* Fix updateHead bug
* Fix updateBase bug
* Change func commitBase to proc commitBase
* Touch up and fix debug mode crash
---------
Co-authored-by: jangko <jangko128@gmail.com>
2025-02-06 08:04:50 +01:00
|
|
|
ldgBefore = LedgerRef.init(com.db.baseTxFrame(), storeSlotHash = true)
|
2024-08-01 10:41:20 +00:00
|
|
|
defer: cx.release()
|
2024-03-22 17:31:56 +00:00
|
|
|
|
2023-08-02 17:17:40 +07:00
|
|
|
for idx, acc in tracedAccountsPairs(tracerInst):
|
2024-08-01 10:41:20 +00:00
|
|
|
before.captureAccount(ldgBefore, acc, internalTxName & $idx)
|
2018-12-25 14:31:12 +07:00
|
|
|
|
2023-08-02 17:17:40 +07:00
|
|
|
for idx, acc in tracedAccountsPairs(tracerInst):
|
2024-12-21 20:46:13 +07:00
|
|
|
after.captureAccount(ledger, acc, internalTxName & $idx)
|
2018-12-25 14:31:12 +07:00
|
|
|
|
2023-08-02 17:17:40 +07:00
|
|
|
result = tracerInst.getTracingResult()
|
2018-12-03 23:22:08 +07:00
|
|
|
result["gas"] = %gasUsed
|
2018-12-25 17:31:51 +07:00
|
|
|
|
|
|
|
if TracerFlags.DisableStateDiff notin tracerFlags:
|
|
|
|
result["stateDiff"] = stateDiff
|
2018-12-03 23:22:08 +07:00
|
|
|
|
2018-12-04 08:53:21 +07:00
|
|
|
# now we dump captured state db
|
2018-12-04 09:01:56 +07:00
|
|
|
if TracerFlags.DisableState notin tracerFlags:
|
2024-08-01 10:41:20 +00:00
|
|
|
result.dumpMemoryDB(cx.cpt)
|
2018-12-11 17:05:49 +07:00
|
|
|
|
2024-08-01 10:41:20 +00:00
|
|
|
|
|
|
|
proc dumpBlockStateImpl(
|
|
|
|
com: CommonRef;
|
|
|
|
blk: EthBlock;
|
|
|
|
dumpState = false;
|
2024-11-07 08:24:21 +07:00
|
|
|
): JsonNode =
|
2024-10-01 21:03:10 +00:00
|
|
|
template header: Header = blk.header
|
2024-08-01 10:41:20 +00:00
|
|
|
|
2018-12-12 10:40:37 +07:00
|
|
|
let
|
2024-08-01 10:41:20 +00:00
|
|
|
cc = activate CaptCtxRef.init(com, header)
|
|
|
|
|
|
|
|
# only need a stack dump when scanning for internal transaction address
|
2023-08-02 17:17:40 +07:00
|
|
|
captureFlags = {DisableMemory, DisableStorage, EnableAccount}
|
|
|
|
tracerInst = newLegacyTracer(captureFlags)
|
aristo: fork support via layers/txframes (#2960)
* aristo: fork support via layers/txframes
This change reorganises how the database is accessed: instead holding a
"current frame" in the database object, a dag of frames is created based
on the "base frame" held in `AristoDbRef` and all database access
happens through this frame, which can be thought of as a consistent
point-in-time snapshot of the database based on a particular fork of the
chain.
In the code, "frame", "transaction" and "layer" is used to denote more
or less the same thing: a dag of stacked changes backed by the on-disk
database.
Although this is not a requirement, in practice each frame holds the
change set of a single block - as such, the frame and its ancestors
leading up to the on-disk state represents the state of the database
after that block has been applied.
"committing" means merging the changes to its parent frame so that the
difference between them is lost and only the cumulative changes remain -
this facility enables frames to be combined arbitrarily wherever they
are in the dag.
In particular, it becomes possible to consolidate a set of changes near
the base of the dag and commit those to disk without having to re-do the
in-memory frames built on top of them - this is useful for "flattening"
a set of changes during a base update and sending those to storage
without having to perform a block replay on top.
Looking at abstractions, a side effect of this change is that the KVT
and Aristo are brought closer together by considering them to be part of
the "same" atomic transaction set - the way the code gets organised,
applying a block and saving it to the kvt happens in the same "logical"
frame - therefore, discarding the frame discards both the aristo and kvt
changes at the same time - likewise, they are persisted to disk together
- this makes reasoning about the database somewhat easier but has the
downside of increased memory usage, something that perhaps will need
addressing in the future.
Because the code reasons more strictly about frames and the state of the
persisted database, it also makes it more visible where ForkedChain
should be used and where it is still missing - in particular, frames
represent a single branch of history while forkedchain manages multiple
parallel forks - user-facing services such as the RPC should use the
latter, ie until it has been finalized, a getBlock request should
consider all forks and not just the blocks in the canonical head branch.
Another advantage of this approach is that `AristoDbRef` conceptually
becomes more simple - removing its tracking of the "current" transaction
stack simplifies reasoning about what can go wrong since this state now
has to be passed around in the form of `AristoTxRef` - as such, many of
the tests and facilities in the code that were dealing with "stack
inconsistency" are now structurally prevented from happening. The test
suite will need significant refactoring after this change.
Once this change has been merged, there are several follow-ups to do:
* there's no mechanism for keeping frames up to date as they get
committed or rolled back - TODO
* naming is confused - many names for the same thing for legacy reason
* forkedchain support is still missing in lots of code
* clean up redundant logic based on previous designs - in particular the
debug and introspection code no longer makes sense
* the way change sets are stored will probably need revisiting - because
it's a stack of changes where each frame must be interrogated to find an
on-disk value, with a base distance of 128 we'll at minimum have to
perform 128 frame lookups for *every* database interaction - regardless,
the "dag-like" nature will stay
* dispose and commit are poorly defined and perhaps redundant - in
theory, one could simply let the GC collect abandoned frames etc, though
it's likely an explicit mechanism will remain useful, so they stay for
now
More about the changes:
* `AristoDbRef` gains a `txRef` field (todo: rename) that "more or less"
corresponds to the old `balancer` field
* `AristoDbRef.stack` is gone - instead, there's a chain of
`AristoTxRef` objects that hold their respective "layer" which has the
actual changes
* No more reasoning about "top" and "stack" - instead, each
`AristoTxRef` can be a "head" that "more or less" corresponds to the old
single-history `top` notion and its stack
* `level` still represents "distance to base" - it's computed from the
parent chain instead of being stored
* one has to be careful not to use frames where forkedchain was intended
- layers are only for a single branch of history!
* fix layer vtop after rollback
* engine fix
* Fix test_txpool
* Fix test_rpc
* Fix copyright year
* fix simulator
* Fix copyright year
* Fix copyright year
* Fix tracer
* Fix infinite recursion bug
* Remove aristo and kvt empty files
* Fic copyright year
* Fix fc chain_kvt
* ForkedChain refactoring
* Fix merge master conflict
* Fix copyright year
* Reparent txFrame
* Fix test
* Fix txFrame reparent again
* Cleanup and fix test
* UpdateBase bugfix and fix test
* Fixe newPayload bug discovered by hive
* Fix engine api fcu
* Clean up call template, chain_kvt, andn txguid
* Fix copyright year
* work around base block loading issue
* Add test
* Fix updateHead bug
* Fix updateBase bug
* Change func commitBase to proc commitBase
* Touch up and fix debug mode crash
---------
Co-authored-by: jangko <jangko128@gmail.com>
2025-02-06 08:04:50 +01:00
|
|
|
txFrame = com.db.baseTxFrame()
|
|
|
|
parent = txFrame.getBlockHeader(header.parentHash).valueOr:
|
2024-08-01 10:41:20 +00:00
|
|
|
return newJNull()
|
aristo: fork support via layers/txframes (#2960)
* aristo: fork support via layers/txframes
This change reorganises how the database is accessed: instead holding a
"current frame" in the database object, a dag of frames is created based
on the "base frame" held in `AristoDbRef` and all database access
happens through this frame, which can be thought of as a consistent
point-in-time snapshot of the database based on a particular fork of the
chain.
In the code, "frame", "transaction" and "layer" is used to denote more
or less the same thing: a dag of stacked changes backed by the on-disk
database.
Although this is not a requirement, in practice each frame holds the
change set of a single block - as such, the frame and its ancestors
leading up to the on-disk state represents the state of the database
after that block has been applied.
"committing" means merging the changes to its parent frame so that the
difference between them is lost and only the cumulative changes remain -
this facility enables frames to be combined arbitrarily wherever they
are in the dag.
In particular, it becomes possible to consolidate a set of changes near
the base of the dag and commit those to disk without having to re-do the
in-memory frames built on top of them - this is useful for "flattening"
a set of changes during a base update and sending those to storage
without having to perform a block replay on top.
Looking at abstractions, a side effect of this change is that the KVT
and Aristo are brought closer together by considering them to be part of
the "same" atomic transaction set - the way the code gets organised,
applying a block and saving it to the kvt happens in the same "logical"
frame - therefore, discarding the frame discards both the aristo and kvt
changes at the same time - likewise, they are persisted to disk together
- this makes reasoning about the database somewhat easier but has the
downside of increased memory usage, something that perhaps will need
addressing in the future.
Because the code reasons more strictly about frames and the state of the
persisted database, it also makes it more visible where ForkedChain
should be used and where it is still missing - in particular, frames
represent a single branch of history while forkedchain manages multiple
parallel forks - user-facing services such as the RPC should use the
latter, ie until it has been finalized, a getBlock request should
consider all forks and not just the blocks in the canonical head branch.
Another advantage of this approach is that `AristoDbRef` conceptually
becomes more simple - removing its tracking of the "current" transaction
stack simplifies reasoning about what can go wrong since this state now
has to be passed around in the form of `AristoTxRef` - as such, many of
the tests and facilities in the code that were dealing with "stack
inconsistency" are now structurally prevented from happening. The test
suite will need significant refactoring after this change.
Once this change has been merged, there are several follow-ups to do:
* there's no mechanism for keeping frames up to date as they get
committed or rolled back - TODO
* naming is confused - many names for the same thing for legacy reason
* forkedchain support is still missing in lots of code
* clean up redundant logic based on previous designs - in particular the
debug and introspection code no longer makes sense
* the way change sets are stored will probably need revisiting - because
it's a stack of changes where each frame must be interrogated to find an
on-disk value, with a base distance of 128 we'll at minimum have to
perform 128 frame lookups for *every* database interaction - regardless,
the "dag-like" nature will stay
* dispose and commit are poorly defined and perhaps redundant - in
theory, one could simply let the GC collect abandoned frames etc, though
it's likely an explicit mechanism will remain useful, so they stay for
now
More about the changes:
* `AristoDbRef` gains a `txRef` field (todo: rename) that "more or less"
corresponds to the old `balancer` field
* `AristoDbRef.stack` is gone - instead, there's a chain of
`AristoTxRef` objects that hold their respective "layer" which has the
actual changes
* No more reasoning about "top" and "stack" - instead, each
`AristoTxRef` can be a "head" that "more or less" corresponds to the old
single-history `top` notion and its stack
* `level` still represents "distance to base" - it's computed from the
parent chain instead of being stored
* one has to be careful not to use frames where forkedchain was intended
- layers are only for a single branch of history!
* fix layer vtop after rollback
* engine fix
* Fix test_txpool
* Fix test_rpc
* Fix copyright year
* fix simulator
* Fix copyright year
* Fix copyright year
* Fix tracer
* Fix infinite recursion bug
* Remove aristo and kvt empty files
* Fic copyright year
* Fix fc chain_kvt
* ForkedChain refactoring
* Fix merge master conflict
* Fix copyright year
* Reparent txFrame
* Fix test
* Fix txFrame reparent again
* Cleanup and fix test
* UpdateBase bugfix and fix test
* Fixe newPayload bug discovered by hive
* Fix engine api fcu
* Clean up call template, chain_kvt, andn txguid
* Fix copyright year
* work around base block loading issue
* Add test
* Fix updateHead bug
* Fix updateBase bug
* Change func commitBase to proc commitBase
* Touch up and fix debug mode crash
---------
Co-authored-by: jangko <jangko128@gmail.com>
2025-02-06 08:04:50 +01:00
|
|
|
vmState = BaseVMState.new(parent, header, com, txFrame, tracerInst, storeSlotHash = true)
|
2020-06-22 07:48:23 +07:00
|
|
|
miner = vmState.coinbase()
|
2024-08-01 10:41:20 +00:00
|
|
|
|
|
|
|
defer: cc.release()
|
2018-12-12 10:40:37 +07:00
|
|
|
|
|
|
|
var
|
2018-12-12 16:41:18 +07:00
|
|
|
before = newJArray()
|
|
|
|
after = newJArray()
|
aristo: fork support via layers/txframes (#2960)
* aristo: fork support via layers/txframes
This change reorganises how the database is accessed: instead holding a
"current frame" in the database object, a dag of frames is created based
on the "base frame" held in `AristoDbRef` and all database access
happens through this frame, which can be thought of as a consistent
point-in-time snapshot of the database based on a particular fork of the
chain.
In the code, "frame", "transaction" and "layer" is used to denote more
or less the same thing: a dag of stacked changes backed by the on-disk
database.
Although this is not a requirement, in practice each frame holds the
change set of a single block - as such, the frame and its ancestors
leading up to the on-disk state represents the state of the database
after that block has been applied.
"committing" means merging the changes to its parent frame so that the
difference between them is lost and only the cumulative changes remain -
this facility enables frames to be combined arbitrarily wherever they
are in the dag.
In particular, it becomes possible to consolidate a set of changes near
the base of the dag and commit those to disk without having to re-do the
in-memory frames built on top of them - this is useful for "flattening"
a set of changes during a base update and sending those to storage
without having to perform a block replay on top.
Looking at abstractions, a side effect of this change is that the KVT
and Aristo are brought closer together by considering them to be part of
the "same" atomic transaction set - the way the code gets organised,
applying a block and saving it to the kvt happens in the same "logical"
frame - therefore, discarding the frame discards both the aristo and kvt
changes at the same time - likewise, they are persisted to disk together
- this makes reasoning about the database somewhat easier but has the
downside of increased memory usage, something that perhaps will need
addressing in the future.
Because the code reasons more strictly about frames and the state of the
persisted database, it also makes it more visible where ForkedChain
should be used and where it is still missing - in particular, frames
represent a single branch of history while forkedchain manages multiple
parallel forks - user-facing services such as the RPC should use the
latter, ie until it has been finalized, a getBlock request should
consider all forks and not just the blocks in the canonical head branch.
Another advantage of this approach is that `AristoDbRef` conceptually
becomes more simple - removing its tracking of the "current" transaction
stack simplifies reasoning about what can go wrong since this state now
has to be passed around in the form of `AristoTxRef` - as such, many of
the tests and facilities in the code that were dealing with "stack
inconsistency" are now structurally prevented from happening. The test
suite will need significant refactoring after this change.
Once this change has been merged, there are several follow-ups to do:
* there's no mechanism for keeping frames up to date as they get
committed or rolled back - TODO
* naming is confused - many names for the same thing for legacy reason
* forkedchain support is still missing in lots of code
* clean up redundant logic based on previous designs - in particular the
debug and introspection code no longer makes sense
* the way change sets are stored will probably need revisiting - because
it's a stack of changes where each frame must be interrogated to find an
on-disk value, with a base distance of 128 we'll at minimum have to
perform 128 frame lookups for *every* database interaction - regardless,
the "dag-like" nature will stay
* dispose and commit are poorly defined and perhaps redundant - in
theory, one could simply let the GC collect abandoned frames etc, though
it's likely an explicit mechanism will remain useful, so they stay for
now
More about the changes:
* `AristoDbRef` gains a `txRef` field (todo: rename) that "more or less"
corresponds to the old `balancer` field
* `AristoDbRef.stack` is gone - instead, there's a chain of
`AristoTxRef` objects that hold their respective "layer" which has the
actual changes
* No more reasoning about "top" and "stack" - instead, each
`AristoTxRef` can be a "head" that "more or less" corresponds to the old
single-history `top` notion and its stack
* `level` still represents "distance to base" - it's computed from the
parent chain instead of being stored
* one has to be careful not to use frames where forkedchain was intended
- layers are only for a single branch of history!
* fix layer vtop after rollback
* engine fix
* Fix test_txpool
* Fix test_rpc
* Fix copyright year
* fix simulator
* Fix copyright year
* Fix copyright year
* Fix tracer
* Fix infinite recursion bug
* Remove aristo and kvt empty files
* Fic copyright year
* Fix fc chain_kvt
* ForkedChain refactoring
* Fix merge master conflict
* Fix copyright year
* Reparent txFrame
* Fix test
* Fix txFrame reparent again
* Cleanup and fix test
* UpdateBase bugfix and fix test
* Fixe newPayload bug discovered by hive
* Fix engine api fcu
* Clean up call template, chain_kvt, andn txguid
* Fix copyright year
* work around base block loading issue
* Add test
* Fix updateHead bug
* Fix updateBase bug
* Change func commitBase to proc commitBase
* Touch up and fix debug mode crash
---------
Co-authored-by: jangko <jangko128@gmail.com>
2025-02-06 08:04:50 +01:00
|
|
|
stateBefore = LedgerRef.init(com.db.baseTxFrame(), storeSlotHash = true)
|
2018-12-12 10:40:37 +07:00
|
|
|
|
Consolidate block type for block processing (#2325)
This PR consolidates the split header-body sequences into a single EthBlock
sequence and cleans up the fallout from that which significantly reduces
block processing overhead during import thanks to less garbage collection
and fewer copies of things all around.
Notably, since the number of headers must always match the number of bodies,
we also get rid of a pointless degree of freedom that in the future could
introduce unnecessary bugs.
* only read header and body from era file
* avoid several unnecessary copies along the block processing way
* simplify signatures, cleaning up unused arguemnts and returns
* use `stew/assign2` in a few strategic places where the generated
nim assignent is slow and add a few `move` to work around poor
analysis in nim 1.6 (will need to be revisited for 2.0)
```
stats-20240607_2223-a814aa0b.csv vs stats-20240608_0714-21c1d0a9.csv
bps_x bps_y tps_x tps_y bpsd tpsd timed
block_number
(498305, 713245] 1,540.52 1,809.73 2,361.58 2775.340189 17.63% 17.63% -14.92%
(713245, 928185] 730.36 865.26 1,715.90 2028.973852 18.01% 18.01% -15.21%
(928185, 1143126] 663.03 789.10 2,529.26 3032.490771 19.79% 19.79% -16.28%
(1143126, 1358066] 393.46 508.05 2,152.50 2777.578119 29.13% 29.13% -22.50%
(1358066, 1573007] 370.88 440.72 2,351.31 2791.896052 18.81% 18.81% -15.80%
(1573007, 1787947] 283.65 335.11 2,068.93 2441.373402 17.60% 17.60% -14.91%
(1787947, 2002888] 287.29 342.11 2,078.39 2474.179448 18.99% 18.99% -15.91%
(2002888, 2217828] 293.38 343.16 2,208.83 2584.77457 17.16% 17.16% -14.61%
(2217828, 2432769] 140.09 167.86 1,081.87 1296.336926 18.82% 18.82% -15.80%
blocks: 1934464, baseline: 3h13m1s, contender: 2h43m47s
bpsd (mean): 19.55%
tpsd (mean): 19.55%
Time (total): -29m13s, -15.14%
```
2024-06-09 16:32:20 +02:00
|
|
|
for idx, tx in blk.transactions:
|
2024-10-04 16:34:31 +02:00
|
|
|
let sender = tx.recoverSender().expect("valid signature")
|
2021-05-04 13:27:18 +07:00
|
|
|
let recipient = tx.getRecipient(sender)
|
2018-12-12 16:41:18 +07:00
|
|
|
before.captureAccount(stateBefore, sender, senderName & $idx)
|
|
|
|
before.captureAccount(stateBefore, recipient, recipientName & $idx)
|
2018-12-12 10:40:37 +07:00
|
|
|
|
2020-06-22 07:48:23 +07:00
|
|
|
before.captureAccount(stateBefore, miner, minerName)
|
2018-12-12 10:40:37 +07:00
|
|
|
|
Consolidate block type for block processing (#2325)
This PR consolidates the split header-body sequences into a single EthBlock
sequence and cleans up the fallout from that which significantly reduces
block processing overhead during import thanks to less garbage collection
and fewer copies of things all around.
Notably, since the number of headers must always match the number of bodies,
we also get rid of a pointless degree of freedom that in the future could
introduce unnecessary bugs.
* only read header and body from era file
* avoid several unnecessary copies along the block processing way
* simplify signatures, cleaning up unused arguemnts and returns
* use `stew/assign2` in a few strategic places where the generated
nim assignent is slow and add a few `move` to work around poor
analysis in nim 1.6 (will need to be revisited for 2.0)
```
stats-20240607_2223-a814aa0b.csv vs stats-20240608_0714-21c1d0a9.csv
bps_x bps_y tps_x tps_y bpsd tpsd timed
block_number
(498305, 713245] 1,540.52 1,809.73 2,361.58 2775.340189 17.63% 17.63% -14.92%
(713245, 928185] 730.36 865.26 1,715.90 2028.973852 18.01% 18.01% -15.21%
(928185, 1143126] 663.03 789.10 2,529.26 3032.490771 19.79% 19.79% -16.28%
(1143126, 1358066] 393.46 508.05 2,152.50 2777.578119 29.13% 29.13% -22.50%
(1358066, 1573007] 370.88 440.72 2,351.31 2791.896052 18.81% 18.81% -15.80%
(1573007, 1787947] 283.65 335.11 2,068.93 2441.373402 17.60% 17.60% -14.91%
(1787947, 2002888] 287.29 342.11 2,078.39 2474.179448 18.99% 18.99% -15.91%
(2002888, 2217828] 293.38 343.16 2,208.83 2584.77457 17.16% 17.16% -14.61%
(2217828, 2432769] 140.09 167.86 1,081.87 1296.336926 18.82% 18.82% -15.80%
blocks: 1934464, baseline: 3h13m1s, contender: 2h43m47s
bpsd (mean): 19.55%
tpsd (mean): 19.55%
Time (total): -29m13s, -15.14%
```
2024-06-09 16:32:20 +02:00
|
|
|
for idx, uncle in blk.uncles:
|
2018-12-12 10:40:37 +07:00
|
|
|
before.captureAccount(stateBefore, uncle.coinbase, uncleName & $idx)
|
|
|
|
|
Consolidate block type for block processing (#2325)
This PR consolidates the split header-body sequences into a single EthBlock
sequence and cleans up the fallout from that which significantly reduces
block processing overhead during import thanks to less garbage collection
and fewer copies of things all around.
Notably, since the number of headers must always match the number of bodies,
we also get rid of a pointless degree of freedom that in the future could
introduce unnecessary bugs.
* only read header and body from era file
* avoid several unnecessary copies along the block processing way
* simplify signatures, cleaning up unused arguemnts and returns
* use `stew/assign2` in a few strategic places where the generated
nim assignent is slow and add a few `move` to work around poor
analysis in nim 1.6 (will need to be revisited for 2.0)
```
stats-20240607_2223-a814aa0b.csv vs stats-20240608_0714-21c1d0a9.csv
bps_x bps_y tps_x tps_y bpsd tpsd timed
block_number
(498305, 713245] 1,540.52 1,809.73 2,361.58 2775.340189 17.63% 17.63% -14.92%
(713245, 928185] 730.36 865.26 1,715.90 2028.973852 18.01% 18.01% -15.21%
(928185, 1143126] 663.03 789.10 2,529.26 3032.490771 19.79% 19.79% -16.28%
(1143126, 1358066] 393.46 508.05 2,152.50 2777.578119 29.13% 29.13% -22.50%
(1358066, 1573007] 370.88 440.72 2,351.31 2791.896052 18.81% 18.81% -15.80%
(1573007, 1787947] 283.65 335.11 2,068.93 2441.373402 17.60% 17.60% -14.91%
(1787947, 2002888] 287.29 342.11 2,078.39 2474.179448 18.99% 18.99% -15.91%
(2002888, 2217828] 293.38 343.16 2,208.83 2584.77457 17.16% 17.16% -14.61%
(2217828, 2432769] 140.09 167.86 1,081.87 1296.336926 18.82% 18.82% -15.80%
blocks: 1934464, baseline: 3h13m1s, contender: 2h43m47s
bpsd (mean): 19.55%
tpsd (mean): 19.55%
Time (total): -29m13s, -15.14%
```
2024-06-09 16:32:20 +02:00
|
|
|
discard vmState.processBlock(blk)
|
2018-12-12 10:40:37 +07:00
|
|
|
|
2024-12-21 20:46:13 +07:00
|
|
|
var stateAfter = vmState.ledger
|
2019-02-02 16:20:45 +07:00
|
|
|
|
Consolidate block type for block processing (#2325)
This PR consolidates the split header-body sequences into a single EthBlock
sequence and cleans up the fallout from that which significantly reduces
block processing overhead during import thanks to less garbage collection
and fewer copies of things all around.
Notably, since the number of headers must always match the number of bodies,
we also get rid of a pointless degree of freedom that in the future could
introduce unnecessary bugs.
* only read header and body from era file
* avoid several unnecessary copies along the block processing way
* simplify signatures, cleaning up unused arguemnts and returns
* use `stew/assign2` in a few strategic places where the generated
nim assignent is slow and add a few `move` to work around poor
analysis in nim 1.6 (will need to be revisited for 2.0)
```
stats-20240607_2223-a814aa0b.csv vs stats-20240608_0714-21c1d0a9.csv
bps_x bps_y tps_x tps_y bpsd tpsd timed
block_number
(498305, 713245] 1,540.52 1,809.73 2,361.58 2775.340189 17.63% 17.63% -14.92%
(713245, 928185] 730.36 865.26 1,715.90 2028.973852 18.01% 18.01% -15.21%
(928185, 1143126] 663.03 789.10 2,529.26 3032.490771 19.79% 19.79% -16.28%
(1143126, 1358066] 393.46 508.05 2,152.50 2777.578119 29.13% 29.13% -22.50%
(1358066, 1573007] 370.88 440.72 2,351.31 2791.896052 18.81% 18.81% -15.80%
(1573007, 1787947] 283.65 335.11 2,068.93 2441.373402 17.60% 17.60% -14.91%
(1787947, 2002888] 287.29 342.11 2,078.39 2474.179448 18.99% 18.99% -15.91%
(2002888, 2217828] 293.38 343.16 2,208.83 2584.77457 17.16% 17.16% -14.61%
(2217828, 2432769] 140.09 167.86 1,081.87 1296.336926 18.82% 18.82% -15.80%
blocks: 1934464, baseline: 3h13m1s, contender: 2h43m47s
bpsd (mean): 19.55%
tpsd (mean): 19.55%
Time (total): -29m13s, -15.14%
```
2024-06-09 16:32:20 +02:00
|
|
|
for idx, tx in blk.transactions:
|
2024-10-04 16:34:31 +02:00
|
|
|
let sender = tx.recoverSender().expect("valid signature")
|
2021-05-04 13:27:18 +07:00
|
|
|
let recipient = tx.getRecipient(sender)
|
2018-12-12 16:41:18 +07:00
|
|
|
after.captureAccount(stateAfter, sender, senderName & $idx)
|
|
|
|
after.captureAccount(stateAfter, recipient, recipientName & $idx)
|
2023-08-02 17:17:40 +07:00
|
|
|
tracerInst.removeTracedAccounts(sender, recipient)
|
2018-12-12 10:40:37 +07:00
|
|
|
|
2020-06-22 07:48:23 +07:00
|
|
|
after.captureAccount(stateAfter, miner, minerName)
|
2023-08-02 17:17:40 +07:00
|
|
|
tracerInst.removeTracedAccounts(miner)
|
2018-12-12 10:40:37 +07:00
|
|
|
|
Consolidate block type for block processing (#2325)
This PR consolidates the split header-body sequences into a single EthBlock
sequence and cleans up the fallout from that which significantly reduces
block processing overhead during import thanks to less garbage collection
and fewer copies of things all around.
Notably, since the number of headers must always match the number of bodies,
we also get rid of a pointless degree of freedom that in the future could
introduce unnecessary bugs.
* only read header and body from era file
* avoid several unnecessary copies along the block processing way
* simplify signatures, cleaning up unused arguemnts and returns
* use `stew/assign2` in a few strategic places where the generated
nim assignent is slow and add a few `move` to work around poor
analysis in nim 1.6 (will need to be revisited for 2.0)
```
stats-20240607_2223-a814aa0b.csv vs stats-20240608_0714-21c1d0a9.csv
bps_x bps_y tps_x tps_y bpsd tpsd timed
block_number
(498305, 713245] 1,540.52 1,809.73 2,361.58 2775.340189 17.63% 17.63% -14.92%
(713245, 928185] 730.36 865.26 1,715.90 2028.973852 18.01% 18.01% -15.21%
(928185, 1143126] 663.03 789.10 2,529.26 3032.490771 19.79% 19.79% -16.28%
(1143126, 1358066] 393.46 508.05 2,152.50 2777.578119 29.13% 29.13% -22.50%
(1358066, 1573007] 370.88 440.72 2,351.31 2791.896052 18.81% 18.81% -15.80%
(1573007, 1787947] 283.65 335.11 2,068.93 2441.373402 17.60% 17.60% -14.91%
(1787947, 2002888] 287.29 342.11 2,078.39 2474.179448 18.99% 18.99% -15.91%
(2002888, 2217828] 293.38 343.16 2,208.83 2584.77457 17.16% 17.16% -14.61%
(2217828, 2432769] 140.09 167.86 1,081.87 1296.336926 18.82% 18.82% -15.80%
blocks: 1934464, baseline: 3h13m1s, contender: 2h43m47s
bpsd (mean): 19.55%
tpsd (mean): 19.55%
Time (total): -29m13s, -15.14%
```
2024-06-09 16:32:20 +02:00
|
|
|
for idx, uncle in blk.uncles:
|
2018-12-12 10:40:37 +07:00
|
|
|
after.captureAccount(stateAfter, uncle.coinbase, uncleName & $idx)
|
2023-08-02 17:17:40 +07:00
|
|
|
tracerInst.removeTracedAccounts(uncle.coinbase)
|
2018-12-12 10:40:37 +07:00
|
|
|
|
2018-12-25 14:31:12 +07:00
|
|
|
# internal transactions:
|
2023-08-02 17:17:40 +07:00
|
|
|
for idx, acc in tracedAccountsPairs(tracerInst):
|
2018-12-25 14:31:12 +07:00
|
|
|
before.captureAccount(stateBefore, acc, internalTxName & $idx)
|
|
|
|
|
2023-08-02 17:17:40 +07:00
|
|
|
for idx, acc in tracedAccountsPairs(tracerInst):
|
2018-12-25 14:31:12 +07:00
|
|
|
after.captureAccount(stateAfter, acc, internalTxName & $idx)
|
|
|
|
|
2018-12-12 10:40:37 +07:00
|
|
|
result = %{"before": before, "after": after}
|
2018-12-12 11:16:40 +07:00
|
|
|
|
2018-12-12 22:18:46 +07:00
|
|
|
if dumpState:
|
2024-08-01 10:41:20 +00:00
|
|
|
result.dumpMemoryDB(cc.cpt)
|
|
|
|
|
2018-12-12 22:18:46 +07:00
|
|
|
|
2024-08-01 10:41:20 +00:00
|
|
|
proc traceBlockImpl(
|
|
|
|
com: CommonRef;
|
|
|
|
blk: EthBlock;
|
|
|
|
tracerFlags: set[TracerFlags] = {};
|
2024-11-07 08:24:21 +07:00
|
|
|
): JsonNode =
|
2024-10-01 21:03:10 +00:00
|
|
|
template header: Header = blk.header
|
2024-08-01 10:41:20 +00:00
|
|
|
|
2018-12-12 11:16:40 +07:00
|
|
|
let
|
2024-08-01 10:41:20 +00:00
|
|
|
cc = activate CaptCtxRef.init(com, header)
|
2023-08-02 17:17:40 +07:00
|
|
|
tracerInst = newLegacyTracer(tracerFlags)
|
2024-08-16 08:22:51 +02:00
|
|
|
# Tracer needs a database where the reverse slot hash table has been set up
|
aristo: fork support via layers/txframes (#2960)
* aristo: fork support via layers/txframes
This change reorganises how the database is accessed: instead holding a
"current frame" in the database object, a dag of frames is created based
on the "base frame" held in `AristoDbRef` and all database access
happens through this frame, which can be thought of as a consistent
point-in-time snapshot of the database based on a particular fork of the
chain.
In the code, "frame", "transaction" and "layer" is used to denote more
or less the same thing: a dag of stacked changes backed by the on-disk
database.
Although this is not a requirement, in practice each frame holds the
change set of a single block - as such, the frame and its ancestors
leading up to the on-disk state represents the state of the database
after that block has been applied.
"committing" means merging the changes to its parent frame so that the
difference between them is lost and only the cumulative changes remain -
this facility enables frames to be combined arbitrarily wherever they
are in the dag.
In particular, it becomes possible to consolidate a set of changes near
the base of the dag and commit those to disk without having to re-do the
in-memory frames built on top of them - this is useful for "flattening"
a set of changes during a base update and sending those to storage
without having to perform a block replay on top.
Looking at abstractions, a side effect of this change is that the KVT
and Aristo are brought closer together by considering them to be part of
the "same" atomic transaction set - the way the code gets organised,
applying a block and saving it to the kvt happens in the same "logical"
frame - therefore, discarding the frame discards both the aristo and kvt
changes at the same time - likewise, they are persisted to disk together
- this makes reasoning about the database somewhat easier but has the
downside of increased memory usage, something that perhaps will need
addressing in the future.
Because the code reasons more strictly about frames and the state of the
persisted database, it also makes it more visible where ForkedChain
should be used and where it is still missing - in particular, frames
represent a single branch of history while forkedchain manages multiple
parallel forks - user-facing services such as the RPC should use the
latter, ie until it has been finalized, a getBlock request should
consider all forks and not just the blocks in the canonical head branch.
Another advantage of this approach is that `AristoDbRef` conceptually
becomes more simple - removing its tracking of the "current" transaction
stack simplifies reasoning about what can go wrong since this state now
has to be passed around in the form of `AristoTxRef` - as such, many of
the tests and facilities in the code that were dealing with "stack
inconsistency" are now structurally prevented from happening. The test
suite will need significant refactoring after this change.
Once this change has been merged, there are several follow-ups to do:
* there's no mechanism for keeping frames up to date as they get
committed or rolled back - TODO
* naming is confused - many names for the same thing for legacy reason
* forkedchain support is still missing in lots of code
* clean up redundant logic based on previous designs - in particular the
debug and introspection code no longer makes sense
* the way change sets are stored will probably need revisiting - because
it's a stack of changes where each frame must be interrogated to find an
on-disk value, with a base distance of 128 we'll at minimum have to
perform 128 frame lookups for *every* database interaction - regardless,
the "dag-like" nature will stay
* dispose and commit are poorly defined and perhaps redundant - in
theory, one could simply let the GC collect abandoned frames etc, though
it's likely an explicit mechanism will remain useful, so they stay for
now
More about the changes:
* `AristoDbRef` gains a `txRef` field (todo: rename) that "more or less"
corresponds to the old `balancer` field
* `AristoDbRef.stack` is gone - instead, there's a chain of
`AristoTxRef` objects that hold their respective "layer" which has the
actual changes
* No more reasoning about "top" and "stack" - instead, each
`AristoTxRef` can be a "head" that "more or less" corresponds to the old
single-history `top` notion and its stack
* `level` still represents "distance to base" - it's computed from the
parent chain instead of being stored
* one has to be careful not to use frames where forkedchain was intended
- layers are only for a single branch of history!
* fix layer vtop after rollback
* engine fix
* Fix test_txpool
* Fix test_rpc
* Fix copyright year
* fix simulator
* Fix copyright year
* Fix copyright year
* Fix tracer
* Fix infinite recursion bug
* Remove aristo and kvt empty files
* Fic copyright year
* Fix fc chain_kvt
* ForkedChain refactoring
* Fix merge master conflict
* Fix copyright year
* Reparent txFrame
* Fix test
* Fix txFrame reparent again
* Cleanup and fix test
* UpdateBase bugfix and fix test
* Fixe newPayload bug discovered by hive
* Fix engine api fcu
* Clean up call template, chain_kvt, andn txguid
* Fix copyright year
* work around base block loading issue
* Add test
* Fix updateHead bug
* Fix updateBase bug
* Change func commitBase to proc commitBase
* Touch up and fix debug mode crash
---------
Co-authored-by: jangko <jangko128@gmail.com>
2025-02-06 08:04:50 +01:00
|
|
|
txFrame = com.db.baseTxFrame()
|
|
|
|
parent = txFrame.getBlockHeader(header.parentHash).valueOr:
|
2024-08-01 10:41:20 +00:00
|
|
|
return newJNull()
|
aristo: fork support via layers/txframes (#2960)
* aristo: fork support via layers/txframes
This change reorganises how the database is accessed: instead holding a
"current frame" in the database object, a dag of frames is created based
on the "base frame" held in `AristoDbRef` and all database access
happens through this frame, which can be thought of as a consistent
point-in-time snapshot of the database based on a particular fork of the
chain.
In the code, "frame", "transaction" and "layer" is used to denote more
or less the same thing: a dag of stacked changes backed by the on-disk
database.
Although this is not a requirement, in practice each frame holds the
change set of a single block - as such, the frame and its ancestors
leading up to the on-disk state represents the state of the database
after that block has been applied.
"committing" means merging the changes to its parent frame so that the
difference between them is lost and only the cumulative changes remain -
this facility enables frames to be combined arbitrarily wherever they
are in the dag.
In particular, it becomes possible to consolidate a set of changes near
the base of the dag and commit those to disk without having to re-do the
in-memory frames built on top of them - this is useful for "flattening"
a set of changes during a base update and sending those to storage
without having to perform a block replay on top.
Looking at abstractions, a side effect of this change is that the KVT
and Aristo are brought closer together by considering them to be part of
the "same" atomic transaction set - the way the code gets organised,
applying a block and saving it to the kvt happens in the same "logical"
frame - therefore, discarding the frame discards both the aristo and kvt
changes at the same time - likewise, they are persisted to disk together
- this makes reasoning about the database somewhat easier but has the
downside of increased memory usage, something that perhaps will need
addressing in the future.
Because the code reasons more strictly about frames and the state of the
persisted database, it also makes it more visible where ForkedChain
should be used and where it is still missing - in particular, frames
represent a single branch of history while forkedchain manages multiple
parallel forks - user-facing services such as the RPC should use the
latter, ie until it has been finalized, a getBlock request should
consider all forks and not just the blocks in the canonical head branch.
Another advantage of this approach is that `AristoDbRef` conceptually
becomes more simple - removing its tracking of the "current" transaction
stack simplifies reasoning about what can go wrong since this state now
has to be passed around in the form of `AristoTxRef` - as such, many of
the tests and facilities in the code that were dealing with "stack
inconsistency" are now structurally prevented from happening. The test
suite will need significant refactoring after this change.
Once this change has been merged, there are several follow-ups to do:
* there's no mechanism for keeping frames up to date as they get
committed or rolled back - TODO
* naming is confused - many names for the same thing for legacy reason
* forkedchain support is still missing in lots of code
* clean up redundant logic based on previous designs - in particular the
debug and introspection code no longer makes sense
* the way change sets are stored will probably need revisiting - because
it's a stack of changes where each frame must be interrogated to find an
on-disk value, with a base distance of 128 we'll at minimum have to
perform 128 frame lookups for *every* database interaction - regardless,
the "dag-like" nature will stay
* dispose and commit are poorly defined and perhaps redundant - in
theory, one could simply let the GC collect abandoned frames etc, though
it's likely an explicit mechanism will remain useful, so they stay for
now
More about the changes:
* `AristoDbRef` gains a `txRef` field (todo: rename) that "more or less"
corresponds to the old `balancer` field
* `AristoDbRef.stack` is gone - instead, there's a chain of
`AristoTxRef` objects that hold their respective "layer" which has the
actual changes
* No more reasoning about "top" and "stack" - instead, each
`AristoTxRef` can be a "head" that "more or less" corresponds to the old
single-history `top` notion and its stack
* `level` still represents "distance to base" - it's computed from the
parent chain instead of being stored
* one has to be careful not to use frames where forkedchain was intended
- layers are only for a single branch of history!
* fix layer vtop after rollback
* engine fix
* Fix test_txpool
* Fix test_rpc
* Fix copyright year
* fix simulator
* Fix copyright year
* Fix copyright year
* Fix tracer
* Fix infinite recursion bug
* Remove aristo and kvt empty files
* Fic copyright year
* Fix fc chain_kvt
* ForkedChain refactoring
* Fix merge master conflict
* Fix copyright year
* Reparent txFrame
* Fix test
* Fix txFrame reparent again
* Cleanup and fix test
* UpdateBase bugfix and fix test
* Fixe newPayload bug discovered by hive
* Fix engine api fcu
* Clean up call template, chain_kvt, andn txguid
* Fix copyright year
* work around base block loading issue
* Add test
* Fix updateHead bug
* Fix updateBase bug
* Change func commitBase to proc commitBase
* Touch up and fix debug mode crash
---------
Co-authored-by: jangko <jangko128@gmail.com>
2025-02-06 08:04:50 +01:00
|
|
|
vmState = BaseVMState.new(parent, header, com, txFrame, tracerInst, storeSlotHash = true)
|
2024-03-22 17:31:56 +00:00
|
|
|
|
2024-08-01 10:41:20 +00:00
|
|
|
defer: cc.release()
|
2018-12-12 11:16:40 +07:00
|
|
|
|
2022-09-03 20:15:35 +02:00
|
|
|
if header.txRoot == EMPTY_ROOT_HASH: return newJNull()
|
Consolidate block type for block processing (#2325)
This PR consolidates the split header-body sequences into a single EthBlock
sequence and cleans up the fallout from that which significantly reduces
block processing overhead during import thanks to less garbage collection
and fewer copies of things all around.
Notably, since the number of headers must always match the number of bodies,
we also get rid of a pointless degree of freedom that in the future could
introduce unnecessary bugs.
* only read header and body from era file
* avoid several unnecessary copies along the block processing way
* simplify signatures, cleaning up unused arguemnts and returns
* use `stew/assign2` in a few strategic places where the generated
nim assignent is slow and add a few `move` to work around poor
analysis in nim 1.6 (will need to be revisited for 2.0)
```
stats-20240607_2223-a814aa0b.csv vs stats-20240608_0714-21c1d0a9.csv
bps_x bps_y tps_x tps_y bpsd tpsd timed
block_number
(498305, 713245] 1,540.52 1,809.73 2,361.58 2775.340189 17.63% 17.63% -14.92%
(713245, 928185] 730.36 865.26 1,715.90 2028.973852 18.01% 18.01% -15.21%
(928185, 1143126] 663.03 789.10 2,529.26 3032.490771 19.79% 19.79% -16.28%
(1143126, 1358066] 393.46 508.05 2,152.50 2777.578119 29.13% 29.13% -22.50%
(1358066, 1573007] 370.88 440.72 2,351.31 2791.896052 18.81% 18.81% -15.80%
(1573007, 1787947] 283.65 335.11 2,068.93 2441.373402 17.60% 17.60% -14.91%
(1787947, 2002888] 287.29 342.11 2,078.39 2474.179448 18.99% 18.99% -15.91%
(2002888, 2217828] 293.38 343.16 2,208.83 2584.77457 17.16% 17.16% -14.61%
(2217828, 2432769] 140.09 167.86 1,081.87 1296.336926 18.82% 18.82% -15.80%
blocks: 1934464, baseline: 3h13m1s, contender: 2h43m47s
bpsd (mean): 19.55%
tpsd (mean): 19.55%
Time (total): -29m13s, -15.14%
```
2024-06-09 16:32:20 +02:00
|
|
|
doAssert(blk.transactions.calcTxRoot == header.txRoot)
|
|
|
|
doAssert(blk.transactions.len != 0)
|
2018-12-12 11:16:40 +07:00
|
|
|
|
|
|
|
var gasUsed = GasInt(0)
|
|
|
|
|
Consolidate block type for block processing (#2325)
This PR consolidates the split header-body sequences into a single EthBlock
sequence and cleans up the fallout from that which significantly reduces
block processing overhead during import thanks to less garbage collection
and fewer copies of things all around.
Notably, since the number of headers must always match the number of bodies,
we also get rid of a pointless degree of freedom that in the future could
introduce unnecessary bugs.
* only read header and body from era file
* avoid several unnecessary copies along the block processing way
* simplify signatures, cleaning up unused arguemnts and returns
* use `stew/assign2` in a few strategic places where the generated
nim assignent is slow and add a few `move` to work around poor
analysis in nim 1.6 (will need to be revisited for 2.0)
```
stats-20240607_2223-a814aa0b.csv vs stats-20240608_0714-21c1d0a9.csv
bps_x bps_y tps_x tps_y bpsd tpsd timed
block_number
(498305, 713245] 1,540.52 1,809.73 2,361.58 2775.340189 17.63% 17.63% -14.92%
(713245, 928185] 730.36 865.26 1,715.90 2028.973852 18.01% 18.01% -15.21%
(928185, 1143126] 663.03 789.10 2,529.26 3032.490771 19.79% 19.79% -16.28%
(1143126, 1358066] 393.46 508.05 2,152.50 2777.578119 29.13% 29.13% -22.50%
(1358066, 1573007] 370.88 440.72 2,351.31 2791.896052 18.81% 18.81% -15.80%
(1573007, 1787947] 283.65 335.11 2,068.93 2441.373402 17.60% 17.60% -14.91%
(1787947, 2002888] 287.29 342.11 2,078.39 2474.179448 18.99% 18.99% -15.91%
(2002888, 2217828] 293.38 343.16 2,208.83 2584.77457 17.16% 17.16% -14.61%
(2217828, 2432769] 140.09 167.86 1,081.87 1296.336926 18.82% 18.82% -15.80%
blocks: 1934464, baseline: 3h13m1s, contender: 2h43m47s
bpsd (mean): 19.55%
tpsd (mean): 19.55%
Time (total): -29m13s, -15.14%
```
2024-06-09 16:32:20 +02:00
|
|
|
for tx in blk.transactions:
|
2022-01-10 09:04:06 +00:00
|
|
|
let
|
2024-10-04 16:34:31 +02:00
|
|
|
sender = tx.recoverSender().expect("valid signature")
|
2022-01-10 09:04:06 +00:00
|
|
|
rc = vmState.processTransaction(tx, sender, header)
|
2022-04-08 11:54:11 +07:00
|
|
|
if rc.isOk:
|
2022-01-10 09:04:06 +00:00
|
|
|
gasUsed = gasUsed + rc.value
|
2018-12-12 11:16:40 +07:00
|
|
|
|
2023-08-02 17:17:40 +07:00
|
|
|
result = tracerInst.getTracingResult()
|
2018-12-12 11:31:53 +07:00
|
|
|
result["gas"] = %gasUsed
|
|
|
|
|
2018-12-12 22:18:46 +07:00
|
|
|
if TracerFlags.DisableState notin tracerFlags:
|
2024-08-01 10:41:20 +00:00
|
|
|
result.dumpMemoryDB(cc.cpt)
|
|
|
|
|
|
|
|
proc traceTransactionsImpl(
|
|
|
|
com: CommonRef;
|
2024-10-01 21:03:10 +00:00
|
|
|
header: Header;
|
2024-08-01 10:41:20 +00:00
|
|
|
transactions: openArray[Transaction];
|
2024-11-07 08:24:21 +07:00
|
|
|
): JsonNode =
|
2018-12-25 17:31:51 +07:00
|
|
|
result = newJArray()
|
Consolidate block type for block processing (#2325)
This PR consolidates the split header-body sequences into a single EthBlock
sequence and cleans up the fallout from that which significantly reduces
block processing overhead during import thanks to less garbage collection
and fewer copies of things all around.
Notably, since the number of headers must always match the number of bodies,
we also get rid of a pointless degree of freedom that in the future could
introduce unnecessary bugs.
* only read header and body from era file
* avoid several unnecessary copies along the block processing way
* simplify signatures, cleaning up unused arguemnts and returns
* use `stew/assign2` in a few strategic places where the generated
nim assignent is slow and add a few `move` to work around poor
analysis in nim 1.6 (will need to be revisited for 2.0)
```
stats-20240607_2223-a814aa0b.csv vs stats-20240608_0714-21c1d0a9.csv
bps_x bps_y tps_x tps_y bpsd tpsd timed
block_number
(498305, 713245] 1,540.52 1,809.73 2,361.58 2775.340189 17.63% 17.63% -14.92%
(713245, 928185] 730.36 865.26 1,715.90 2028.973852 18.01% 18.01% -15.21%
(928185, 1143126] 663.03 789.10 2,529.26 3032.490771 19.79% 19.79% -16.28%
(1143126, 1358066] 393.46 508.05 2,152.50 2777.578119 29.13% 29.13% -22.50%
(1358066, 1573007] 370.88 440.72 2,351.31 2791.896052 18.81% 18.81% -15.80%
(1573007, 1787947] 283.65 335.11 2,068.93 2441.373402 17.60% 17.60% -14.91%
(1787947, 2002888] 287.29 342.11 2,078.39 2474.179448 18.99% 18.99% -15.91%
(2002888, 2217828] 293.38 343.16 2,208.83 2584.77457 17.16% 17.16% -14.61%
(2217828, 2432769] 140.09 167.86 1,081.87 1296.336926 18.82% 18.82% -15.80%
blocks: 1934464, baseline: 3h13m1s, contender: 2h43m47s
bpsd (mean): 19.55%
tpsd (mean): 19.55%
Time (total): -29m13s, -15.14%
```
2024-06-09 16:32:20 +02:00
|
|
|
for i in 0 ..< transactions.len:
|
2024-08-01 10:41:20 +00:00
|
|
|
result.add traceTransactionImpl(
|
|
|
|
com, header, transactions, i.uint64, {DisableState})
|
2018-12-25 17:31:51 +07:00
|
|
|
|
2024-08-01 10:41:20 +00:00
|
|
|
# ------------------------------------------------------------------------------
|
|
|
|
# Public functions
|
|
|
|
# ------------------------------------------------------------------------------
|
|
|
|
|
|
|
|
proc traceBlock*(
|
|
|
|
com: CommonRef;
|
|
|
|
blk: EthBlock;
|
|
|
|
tracerFlags: set[TracerFlags] = {};
|
|
|
|
): JsonNode =
|
2024-11-07 08:24:21 +07:00
|
|
|
com.traceBlockImpl(blk, tracerFlags)
|
2024-08-01 10:41:20 +00:00
|
|
|
|
|
|
|
proc toJson*(receipts: seq[Receipt]): JsonNode =
|
|
|
|
result = newJArray()
|
|
|
|
for receipt in receipts:
|
|
|
|
result.add receipt.toJson
|
|
|
|
|
|
|
|
proc dumpMemoryDB*(node: JsonNode, cpt: CoreDbCaptRef) =
|
|
|
|
var n = newJObject()
|
|
|
|
for (k,v) in cpt.kvtLog:
|
|
|
|
n[k.toHex(false)] = %v
|
|
|
|
node["state"] = n
|
|
|
|
|
aristo: fork support via layers/txframes (#2960)
* aristo: fork support via layers/txframes
This change reorganises how the database is accessed: instead holding a
"current frame" in the database object, a dag of frames is created based
on the "base frame" held in `AristoDbRef` and all database access
happens through this frame, which can be thought of as a consistent
point-in-time snapshot of the database based on a particular fork of the
chain.
In the code, "frame", "transaction" and "layer" is used to denote more
or less the same thing: a dag of stacked changes backed by the on-disk
database.
Although this is not a requirement, in practice each frame holds the
change set of a single block - as such, the frame and its ancestors
leading up to the on-disk state represents the state of the database
after that block has been applied.
"committing" means merging the changes to its parent frame so that the
difference between them is lost and only the cumulative changes remain -
this facility enables frames to be combined arbitrarily wherever they
are in the dag.
In particular, it becomes possible to consolidate a set of changes near
the base of the dag and commit those to disk without having to re-do the
in-memory frames built on top of them - this is useful for "flattening"
a set of changes during a base update and sending those to storage
without having to perform a block replay on top.
Looking at abstractions, a side effect of this change is that the KVT
and Aristo are brought closer together by considering them to be part of
the "same" atomic transaction set - the way the code gets organised,
applying a block and saving it to the kvt happens in the same "logical"
frame - therefore, discarding the frame discards both the aristo and kvt
changes at the same time - likewise, they are persisted to disk together
- this makes reasoning about the database somewhat easier but has the
downside of increased memory usage, something that perhaps will need
addressing in the future.
Because the code reasons more strictly about frames and the state of the
persisted database, it also makes it more visible where ForkedChain
should be used and where it is still missing - in particular, frames
represent a single branch of history while forkedchain manages multiple
parallel forks - user-facing services such as the RPC should use the
latter, ie until it has been finalized, a getBlock request should
consider all forks and not just the blocks in the canonical head branch.
Another advantage of this approach is that `AristoDbRef` conceptually
becomes more simple - removing its tracking of the "current" transaction
stack simplifies reasoning about what can go wrong since this state now
has to be passed around in the form of `AristoTxRef` - as such, many of
the tests and facilities in the code that were dealing with "stack
inconsistency" are now structurally prevented from happening. The test
suite will need significant refactoring after this change.
Once this change has been merged, there are several follow-ups to do:
* there's no mechanism for keeping frames up to date as they get
committed or rolled back - TODO
* naming is confused - many names for the same thing for legacy reason
* forkedchain support is still missing in lots of code
* clean up redundant logic based on previous designs - in particular the
debug and introspection code no longer makes sense
* the way change sets are stored will probably need revisiting - because
it's a stack of changes where each frame must be interrogated to find an
on-disk value, with a base distance of 128 we'll at minimum have to
perform 128 frame lookups for *every* database interaction - regardless,
the "dag-like" nature will stay
* dispose and commit are poorly defined and perhaps redundant - in
theory, one could simply let the GC collect abandoned frames etc, though
it's likely an explicit mechanism will remain useful, so they stay for
now
More about the changes:
* `AristoDbRef` gains a `txRef` field (todo: rename) that "more or less"
corresponds to the old `balancer` field
* `AristoDbRef.stack` is gone - instead, there's a chain of
`AristoTxRef` objects that hold their respective "layer" which has the
actual changes
* No more reasoning about "top" and "stack" - instead, each
`AristoTxRef` can be a "head" that "more or less" corresponds to the old
single-history `top` notion and its stack
* `level` still represents "distance to base" - it's computed from the
parent chain instead of being stored
* one has to be careful not to use frames where forkedchain was intended
- layers are only for a single branch of history!
* fix layer vtop after rollback
* engine fix
* Fix test_txpool
* Fix test_rpc
* Fix copyright year
* fix simulator
* Fix copyright year
* Fix copyright year
* Fix tracer
* Fix infinite recursion bug
* Remove aristo and kvt empty files
* Fic copyright year
* Fix fc chain_kvt
* ForkedChain refactoring
* Fix merge master conflict
* Fix copyright year
* Reparent txFrame
* Fix test
* Fix txFrame reparent again
* Cleanup and fix test
* UpdateBase bugfix and fix test
* Fixe newPayload bug discovered by hive
* Fix engine api fcu
* Clean up call template, chain_kvt, andn txguid
* Fix copyright year
* work around base block loading issue
* Add test
* Fix updateHead bug
* Fix updateBase bug
* Change func commitBase to proc commitBase
* Touch up and fix debug mode crash
---------
Co-authored-by: jangko <jangko128@gmail.com>
2025-02-06 08:04:50 +01:00
|
|
|
proc dumpReceipts*(chainDB: CoreDbTxRef, header: Header): JsonNode =
|
2024-11-07 08:24:21 +07:00
|
|
|
chainDB.dumpReceiptsImpl header
|
2024-08-01 10:41:20 +00:00
|
|
|
|
|
|
|
proc traceTransaction*(
|
|
|
|
com: CommonRef;
|
2024-10-01 21:03:10 +00:00
|
|
|
header: Header;
|
2024-08-01 10:41:20 +00:00
|
|
|
txs: openArray[Transaction];
|
|
|
|
txIndex: uint64;
|
|
|
|
tracerFlags: set[TracerFlags] = {};
|
|
|
|
): JsonNode =
|
2024-11-07 08:24:21 +07:00
|
|
|
com.traceTransactionImpl(header, txs, txIndex,tracerFlags)
|
2024-08-01 10:41:20 +00:00
|
|
|
|
|
|
|
proc dumpBlockState*(
|
|
|
|
com: CommonRef;
|
|
|
|
blk: EthBlock;
|
|
|
|
dumpState = false;
|
|
|
|
): JsonNode =
|
2024-11-07 08:24:21 +07:00
|
|
|
com.dumpBlockStateImpl(blk, dumpState)
|
2024-08-01 10:41:20 +00:00
|
|
|
|
|
|
|
proc traceTransactions*(
|
|
|
|
com: CommonRef;
|
2024-10-01 21:03:10 +00:00
|
|
|
header: Header;
|
2024-08-01 10:41:20 +00:00
|
|
|
transactions: openArray[Transaction];
|
|
|
|
): JsonNode =
|
2024-11-07 08:24:21 +07:00
|
|
|
com.traceTransactionsImpl(header, transactions)
|
2024-08-01 10:41:20 +00:00
|
|
|
|
|
|
|
# ------------------------------------------------------------------------------
|
|
|
|
# End
|
|
|
|
# ------------------------------------------------------------------------------
|