Commit Graph

33 Commits

Author SHA1 Message Date
jangko 6544adf360
implement EIP-1153: Transient storage
new EVM opcodes:
- TLOAD  0xb3
- TSTORE 0xb4
2023-06-23 14:04:36 +07:00
Adam Spitz 7f56e90654
Revert "Some work on speculative execution (#1552)" (#1561)
This reverts commit ddbdf34c3d.
2023-04-24 16:59:38 -04:00
Adam Spitz ddbdf34c3d
Some work on speculative execution (#1552)
* Recreating some of the speculative-execution code.

Not really using it yet. Also there's some new inefficiency in
memory.nim, but it's fixable - just haven't gotten around to it yet.

The big thing introduced here is the idea of "cells" for stack,
memory, and storage values. A cell is basically just a Future (though
there's also the option of making it an Identity - just a simple
distinct wrapper around a value - if you want to turn off the
asynchrony).

* Bumped nim-eth.

* Cleaned up a few comments.

* Bumped nim-secp256k1.

* Oops.

* Fixing a few compiler errors that show up with EVMC enabled.
2023-04-22 10:17:37 -04:00
jangko a31db89e0e
fix t8n tool bug and accounts cache bug
t8n: a silly bug contract address generator, should use original
tx nonce instead of read the nonce from sender address in state db.
Although in EVM contract address generated by reading nonce from state db
is correct, outside EVM that nonce value might have been modified,
thus generating incorrect contract address.

accounts cache: when clearing account storage, the originalValue
cache is not cleared, only the storageRoot set to empty storage root,
this will cause getStorage and getCommitedStorage return wrong value
if the originalValue cache contains old value.
2023-03-23 11:07:30 +07:00
jangko 0b97f5a2e6
t8n tools support withdrawals records 2023-03-21 18:36:22 +07:00
jangko 2a3c67b4be
refactor touched accounts, selfdestruct, and log
simplify EVM and delegete those things to accounts cache.
also no more manual state clearing, accounts cache will be
responsible for both collecting touched account and perform
state clearing.
2023-03-21 08:14:06 +07:00
jangko 633f135c72
align accounts cache with EIP158/161
Some nomenclature used in accounts cache are not what described
in EIP158/161, therefore causing confusion and introduce bugs.
Now it should be fixed.
2023-03-18 18:13:06 +07:00
Adam Spitz ec2bd4a9c5
More work on withdrawals (#1503)
* Gwei conversion should use u256 because u64 can overflow.

* Make withdrawals follow the EIP-158 state-clearing rules.

(i.e. Empty accounts should be deleted.)

* Allow the zero address in normalizeNumber.

(Necessary for one of the new withdrawals-related tests.)
2023-03-16 16:34:47 -04:00
Adam Spitz 11771f8444
Recreating some of the old stateless code that had bitrotted. (#1493)
Note that we're not really using it yet; I'm just refactoring and
adding new code, working in small steps.
2023-03-10 17:16:42 -05:00
Adam Spitz c921135e2e
Refactored to introduce distinct types AccountsTrie and StorageTrie. (#1490)
This is in preparation for working on stateless mode; it's useful to
have a clearer and more type-safe interface for accessing accounts
versus slots.
2023-03-10 13:42:37 -05:00
Kim De Mey a669b51ec5
Bump Nim to 1.6 and resolve the related issues (#1445)
Two unresolved items currently:
- Three tests that are temporarily disabled as they fail in the
macro_assembler code, which seems to be due to an ambigious
identifier Stop (Ops and chronos ServerCommand enum).
- i386 CI disabled as it fails at Nim compilation already. Failed
tests where already ignored for this target.
2023-01-26 13:37:19 +01:00
jangko 94a94c5b65 implement better hardfork management 2022-12-02 13:51:42 +07:00
jangko 16bc2de1cf
update EF test fixtures and fixes to pass all tests 2022-10-03 16:41:32 +07:00
jangko f2f204293e
first step into styleCheck fixes 2022-04-14 08:39:50 +07:00
Jordan Hrycaj 261c0b51a7
Redesign of BaseVMState descriptor (#923)
* Redesign of BaseVMState descriptor

why:
  BaseVMState provides an environment for executing transactions. The
  current descriptor also provides data that cannot generally be known
  within the execution environment, e.g. the total gasUsed which is
  available not before after all transactions have finished.

  Also, the BaseVMState constructor has been replaced by a constructor
  that does not need pre-initialised input of the account database.

also:
  Previous constructor and some fields are provided with a deprecated
  annotation (producing a lot of noise.)

* Replace legacy directives in production sources

* Replace legacy directives in unit test sources

* fix CI (missing premix update)

* Remove legacy directives

* chase CI problem

* rebased

* Re-introduce 'AccountsCache' constructor optimisation for 'BaseVmState' re-initialisation

why:
  Constructing a new 'AccountsCache' descriptor can be avoided sometimes
  when the current state root is properly positioned already. Such a
  feature existed already as the update function 'initStateDB()' for the
  'BaseChanDB' where the accounts cache was linked into this desctiptor.

  The function 'initStateDB()' was removed and re-implemented into the
  'BaseVmState' constructor without optimisation. The old version was of
  restricted use as a wrong accounts cache state would unconditionally
  throw an exception rather than conceptually ask for a remedy.

  The optimised 'BaseVmState' re-initialisation has been implemented for
  the 'persistBlocks()' function.

also:
  moved some test helpers to 'test/replay' folder

* Remove unused & undocumented fields from Chain descriptor

why:
  Reduces attack surface in general & improves reading the code.
2022-01-18 16:19:32 +00:00
jangko baf508f6ae
move stateDB from VMState to chainDB
previously, every time the VMState was created, it will also create
new stateDB, and this action will nullify the advantages of cached accounts.

the new changes will conserve the accounts cache if the executed blocks
are contiguous. if not the stateDB need to be reinited.

this changes also allow rpcCallEvm and rpcEstimateGas executed properly
using current stateDB instead of creating new one each time they are called.
2021-10-28 18:57:08 +07:00
jangko 3db535aa39
EIP2929 implementation 2021-01-11 14:56:42 +07:00
jangko f2b483d6ad
access list implementation 2021-01-11 14:54:55 +07:00
jangko 165f9fea2e
reduce warnings 2020-07-21 13:15:06 +07:00
jangko 9d0b399213
fix accounts_cache: move overlayStorage to originalStorage when not clearCache 2020-06-20 18:45:09 +07:00
jangko bd7e1fe2e5
reuse VMState and AccountsCache for better performance 2020-06-18 13:16:38 +07:00
jangko eabacb0a33
allow nimbus to read geth database 2020-06-18 13:16:25 +07:00
jangko 50816f2ebb
implement block witness data collection in accounts_cache 2020-06-03 20:50:13 +07:00
jangko 844071033a
fix at various places related to missing accounts_cache.persist call 2020-06-01 13:45:32 +07:00
jangko 8a4c8c6273
fix buggy isAlive from accounts_cache 2020-06-01 10:54:25 +07:00
jangko 71514a0a66
replace state_db with accounts_cache 2020-05-30 10:14:59 +07:00
andri lim c9e49bf68a
fix accounts cache init API 2020-04-29 12:00:44 +07:00
Jacek Sieka 4ade5797ee
rlp: don't use ranges / experimental features (#495) 2020-04-20 20:12:44 +02:00
andri lim edc19076e3 accounts cache: read-only operations should not create new account 2020-02-27 14:38:42 +02:00
andri lim 5f24757cd7 use 'withValue' when retrieving account's storage 2020-01-19 19:17:37 +02:00
andri lim cf0d4481f3 [skip ci] rename 'isExists' to 'exists' 2020-01-19 19:17:37 +02:00
andri lim 727b477fca [skip ci] add accounts cache test 2020-01-19 19:17:37 +02:00
andri lim 5795bf52e0 [skip ci] implement accounts_cache 2020-01-19 19:17:37 +02:00