Commit Graph

61 Commits

Author SHA1 Message Date
Martin Holst Swende
f62c58f8de
trie: make rhs-proof align with last key in range proofs (#28311)
During snap-sync, we request ranges of values: either a range of accounts or a range of storage values. For any large trie, e.g. the main account trie or a large storage trie, we cannot fetch everything at once.

Short version; we split it up and request in multiple stages. To do so, we use an origin field, to say "Give me all storage key/values where key > 0x20000000000000000". When the server fulfils this, the server provides the first key after origin, let's say 0x2e030000000000000 -- never providing the exact origin. However, the client-side needs to be able to verify that the 0x2e03.. indeed is the first one after 0x2000.., and therefore the attached proof concerns the origin, not the first key.

So, short-short version: the left-hand side of the proof relates to the origin, and is free-standing from the first leaf.

On the other hand, (pun intended), the right-hand side, there's no such 'gap' between "along what path does the proof walk" and the last provided leaf. The proof must prove the last element (unless there are no elements).

Therefore, we can simplify the semantics for trie.VerifyRangeProof by removing an argument. This doesn't make much difference in practice, but makes it so that we can remove some tests. The reason I am raising this is that the upcoming stacktrie-based verifier does not support such fancy features as standalone right-hand borders.
2023-10-13 16:05:29 +02:00
rjl493456442
1cb3b6aee4
eth/protocols/snap: fix snap sync failure on empty storage range (#28306)
This change addresses an issue in snap sync, specifically when the entire sync process can be halted due to an encountered empty storage range.

Currently, on the snap sync client side, the response to an empty (partial) storage range is discarded as a non-delivery. However, this response can be a valid response, when the particular range requested does not contain any slots.

For instance, consider a large contract where the entire key space is divided into 16 chunks, and there are no available slots in the last chunk [0xf] -> [end]. When the node receives a request for this particular range, the response includes:

    The proof with origin [0xf]
    A nil storage slot set

If we simply discard this response, the finalization of the last range will be skipped, halting the entire sync process indefinitely. The test case TestSyncWithUnevenStorage can reproduce the scenario described above.

In addition, this change also defines the common variables MaxAddress and MaxHash.
2023-10-13 09:08:26 +02:00
Martin Holst Swende
6b1e4f4211
all: move light.NodeSet to trienode.ProofSet (#28287)
This is a minor refactor in preparation of changes to range verifier. This PR contains no intentional functional changes but moves (and renames) the light.NodeSet
2023-10-10 10:30:47 +02:00
Péter Szilágyi
bc6d184872
cmd/devp2p, eth: drop eth/66 (#28239)
* cmd/devp2p, eth: drop eth/66

* eth/protocols/eth: yes sir, linter
2023-10-03 15:03:19 +03:00
rjl493456442
73f5bcb75b
core, accounts, eth, trie: handle genesis state missing (#28171)
* core, accounts, eth, trie: handle genesis state missing

* core, eth, trie: polish

* core: manage txpool subscription in mainpool

* eth/backend: fix test

* cmd, eth: fix test

* core/rawdb, trie/triedb/pathdb: address comments

* eth, trie: address comments

* eth: inline the function

* eth: use synced flag

* core/txpool: revert changes in txpool

* core, eth, trie: rename functions
2023-09-28 10:00:53 +03:00
lightclient
32fde3f838
core/forkid: correctly compute forkid when timestamp fork is activated in genesis (#27895)
This changes the forkID calculation to ignore time-based forks that occurred before the
genesis block. It's supposed to be done this way because the spec says:

> If a chain is configured to start with a non-Frontier ruleset already in its genesis, that is NOT considered a fork.
2023-08-16 23:31:02 +02:00
rjl493456442
59f7b289c3
cmd, core, eth, graphql, trie: no persisted clean trie cache file (#27525)
The clean trie cache is persisted periodically, therefore Geth can
quickly warmup the cache in next restart.

However it will reduce the robustness of system. The assumption is
held in Geth that if the parent trie node is present, then the entire
sub-trie associated with the parent are all prensent.

Imagine the scenario that Geth rewinds itself to a past block and
restart, but Geth finds the root node of "future state" in clean
cache then regard this state is present in disk, while is not in fact.

Another example is offline pruning tool. Whenever an offline pruning
is performed, the clean cache file has to be removed to aviod hitting
the root node of "deleted states" in clean cache.

All in all, compare with the minor performance gain, system robustness
is something we care more.
2023-07-04 10:21:06 +03:00
Delweng
e9c3183c52
cmd: use errrors.New instead of empty fmt.Errorf (#27329)
Signed-off-by: jsvisa <delweng@gmail.com>
2023-05-24 12:21:29 +02:00
Péter Szilágyi
dde2da0efb
all: remove ethash pow, only retain shims needed for consensus and tests (#27178)
* all: remove ethash pow, only retain shims needed for consensus and tests

* all: thank you linter

* all: disallow launching Geth in legacy PoW mode

* cmd/env/internal/t8ntool: remove dangling ethash flag
2023-05-03 12:58:39 +03:00
rjl493456442
fe01a2f63b
all: use unified emptyRootHash and emptyCodeHash (#26718)
The EmptyRootHash and EmptyCodeHash are defined everywhere in the codebase, this PR replaces all of them with unified one defined in core/types package, and also defines constants for TxRoot, WithdrawalsRoot and UncleRoot
2023-02-21 06:12:27 -05:00
Marius van der Wijden
a4e19c5ca3
all: implement forkid changes for shanghai 2023-01-03 12:57:06 +02:00
Marius van der Wijden
055528ae05
cmd/devp2p/internal/ethtest: add support for eth/68 (#26078)
Co-authored-by: Felix Lange <fjl@twurst.com>
2022-11-07 20:47:04 +01:00
vdwijden
b0d44338bb
eth: implement eth/68 (#25980)
* eth: implement eth/68

* eth/protocols/eth: added tx size to announcement

* eth/protocols/eth: check equal lengths on receiving announcement

* eth/protocols/eth: add +1 to tx size because of the type byte

* eth: happy lint, add eth68 tests, enable eth68

* eth: various nitpick fixes on eth/68

* eth/protocols/eth: fix announced tx size wrt type byte

Co-authored-by: MariusVanDerWijden <m.vanderwijden@live.de>
Co-authored-by: Péter Szilágyi <peterke@gmail.com>
2022-10-31 16:23:26 +02:00
Martin Holst Swende
5a02b2d6d0
all: fix spelling mistakes (#25961) 2022-10-11 09:37:00 +02:00
Péter Szilágyi
4f7a425aa8
Merge pull request #25924 from holiman/comments_fix
all: fix docstrings
2022-10-04 13:30:00 +03:00
Martin Holst Swende
ee301c750b
all: fix docstrings 2022-10-04 09:18:02 +02:00
Martin Holst Swende
f61b50b1e8
eth/protocols/snap: serve snap requests when possible (#25644)
This PR makes it so that the snap server responds to trie heal requests when possible, even if the snapshot does not exist. The idea being that it might prolong the lifetime of a state root, so we don't have to pivot quite as often.
2022-10-03 13:37:17 +02:00
Martin Holst Swende
7f2890a9be
eth/fetcher: throttle peers which deliver many invalid transactions (#25573)
Co-authored-by: Felix Lange <fjl@twurst.com>
2022-09-02 17:28:33 +02:00
Péter Szilágyi
02418c2fa9
Revert "eth/fetcher: don't spend too much time on transaction inclusion" (#25567)
Revert "eth/fetcher: don't spend too much time on transaction inclusion (#25524)"

This reverts commit 0ce494b60c.
2022-08-22 10:14:56 +03:00
Martin Holst Swende
0ce494b60c
eth/fetcher: don't spend too much time on transaction inclusion (#25524)
* eth/fetcher: introduce some lag in tx fetching

* eth/fetcher: change conditions a bit

* eth/fetcher: use per-batch quota check

* eth/fetcher: fix some comments

* eth/fetcher: address review concerns

* eth/fetcher: fix panic + add warn log

* eth/fetcher: fix log

* eth/fetcher: fix log

* cmd/devp2p/internal/ethtest: fix ignorign tx announcements from prev. tests

* cmd/devp2p/internal/ethtest: fix TestLargeTxRequest

This increases the number of tx relay messages the test waits for. Since
go-ethereum now processes incoming txs in smaller batches, the
announcement messages it sends are also smaller.

Co-authored-by: Felix Lange <fjl@twurst.com>
2022-08-19 16:59:36 +03:00
rjl493456442
a41ea8a97c
all: cleanup the APIs for initializing genesis (#25473)
* all: polish tests

* core: apply feedback from Guillaume

* core: fix comment
2022-08-09 12:44:39 +03:00
Felix Lange
d804a59ee1
cmd/devp2p/internal/ethtest: update tests for eth/67 (#25306) 2022-08-02 14:48:55 +02:00
Delweng
b196ad1c16
all: add whitespace linter (#25312)
* golangci: typo

Signed-off-by: Delweng <delweng@gmail.com>

* golangci: add whietspace

Signed-off-by: Delweng <delweng@gmail.com>

* *: rm whitesapce using golangci-lint

Signed-off-by: Delweng <delweng@gmail.com>

* cmd/puppeth: revert accidental resurrection

Co-authored-by: Péter Szilágyi <peterke@gmail.com>
2022-07-25 13:14:03 +03:00
Seungbae.yu
2697e44d81
all: change format 0x%x to %#x (#25221) 2022-07-04 11:03:32 +03:00
lmittmann
bc013bc42e
all: prefer new(big.Int) over big.NewInt(0) (#25087)
minor performance improvement: `big.NewInt(0).Xxx` -> `new(big.Int).Xxx`
2022-06-14 15:09:48 +03:00
Martin Holst Swende
a907d7e81a
all: more linters (#24783)
This enables the following linters

- typecheck
- unused
- staticcheck
- bidichk
- durationcheck
- exportloopref
- gosec

WIth a few exceptions.

- We use a deprecated protobuf in trezor. I didn't want to mess with that, since I cannot meaningfully test any changes there.
- The deprecated TypeMux is used in a few places still, so the warning for it is silenced for now.
- Using string type in context.WithValue is apparently wrong, one should use a custom type, to prevent collisions between different places in the hierarchy of callers. That should be fixed at some point, but may require some attention.
- The warnings for using weak random generator are squashed, since we use a lot of random without need for cryptographic guarantees.
2022-06-13 16:24:45 +02:00
Felix Lange
9244d5cd61
all: update license headers and AUTHORS from git history (#24947) 2022-05-24 20:39:40 +02:00
Håvard Anda Estensen
07508ac0e9
all: replace uses of ioutil with io and os (#24869) 2022-05-16 11:59:35 +02:00
rjl493456442
344d6f95cf
cmd: group network and db path flags together (#24698)
This PR groups all built-in network flags together and list them in the command as a whole.

And all database path flags(datadir, ancient) are also grouped, since usually these two are
used together.
2022-05-03 08:46:17 +02:00
Martin Holst Swende
6ce4670bc0
cmd/devp2p: implement snap protocol testing (#24276)
This also contains some changes to the protocol handler to
make the tests pass.
2022-02-04 15:24:32 +01:00
Martin Holst Swende
0efed7f58b
cmd/devp2p/internal/ethtest: clarify protocol version in tests (#23872)
Debugging recent geth failures in hive, it took a while to realize that it's because
geth doesn't support eth/65 any longer. This PR makes such failures a bit more
easy to figure out.
2021-11-09 14:45:34 +01:00
Marius van der Wijden
84ff152de5
cmd/devp2p/internal/ethtest: only use eth66 if eth66 is negotiated (#23568)
* cmd/devp2p/internal/ethtest: only use eth66 if eth66 is negotiated

* cmd/devp2p/internal/ethtest: switch on concrete type not pointer

* cmd/devp2p/internal/ethtest: switch on concrete type not pointer
2021-09-15 10:52:48 +02:00
baptiste-b-pegasys
154b525ce8
cmd/devp2p/internal/ethtest: remove WriteTo method (#23379)
WriteTo is unused, and go vet warns about it because it
doesn't match the usual signature of the io.WriterTo method.
2021-08-25 19:06:42 +02:00
Péter Szilágyi
0db0b27754
Revert "Revert "eth: drop eth/65, the last non-reqid protocol version" (#23426)" (#23456)
This reverts commit c368f728c1.
2021-08-24 21:52:58 +03:00
Péter Szilágyi
c368f728c1
Revert "eth: drop eth/65, the last non-reqid protocol version" (#23426) 2021-08-20 15:14:21 +03:00
Péter Szilágyi
fd604becbb
Merge pull request #23120 from karalabe/drop-eth-65
eth: drop eth/65, the last non-reqid protocol version
2021-08-13 11:52:47 +03:00
Martin Holst Swende
b9d4412715
cmd/devp2p: fixes for eth and discv4 tests (#23155)
This PR fixes a false positive PONG 'to' endpoint mismatch seen in hive tests:

    got {IP:172.17.0.7 UDP:44025 TCP:44025}, want {IP:172.17.0.7 UDP:44025 TCP:0}

Co-authored-by: Felix Lange <fjl@twurst.com>
2021-07-07 17:28:14 +02:00
Péter Szilágyi
d3f018fde8
eth: drop eth/65, the last non-reqid protocol version 2021-06-29 12:31:30 +03:00
Felix Lange
ee35ddc8fd
cmd/devp2p/internal/ethtest: ignore block announcement in tx test (#22957) 2021-05-27 20:53:33 +02:00
rene
d836ad141e
cmd/devp2p/internal/ethtest: add block hash announcement test (#22535) 2021-05-27 11:57:49 +02:00
rene
49bde05a55
cmd/devp2p: refactor eth test suite (#22843)
This PR refactors the eth test suite to make it more readable and
easier to use. Some notable differences:

- A new file helpers.go stores all of the methods used between
  both eth66 and eth65 and below tests, as well as methods shared
  among many test functions.
- suite.go now contains all of the test functions for both eth65
  tests and eth66 tests.
- The utesting.T object doesn't get passed through to other helper methods,
  but is instead only used within the scope of the test function,
  whereas helper methods return errors, so only the test function
  itself can fatal out in the case of an error.
- The full test suite now only takes 13.5 seconds to run.
2021-05-25 23:09:11 +02:00
rene
0f3a1e7f9b
cmd/devp2p/internal/ethtest: send simultaneous requests on one connection (#22801)
This changes the SimultaneousRequests test to send the requests from the same
connection, as it doesn't really make sense to test whether a node can respond
to two requests with different request IDs from separate connections.
2021-05-05 12:27:27 +02:00
rene
8ff98108e5
cmd/devp2p: fix flakey tests in CI (#22757)
This PR fixes a couple of issues in the eth test suite that caused flakiness when run in the CI.
2021-04-30 22:47:36 +02:00
rene
e4270cacf4
cmd/devp2p: fix flaky SameRequestID test (#22754) 2021-04-28 21:38:38 +02:00
Martin Holst Swende
0c99868416
cmd/devp2p, eth/protocols/eth: fix tests + make sanity checks earlier (#22749) 2021-04-28 08:48:07 +02:00
rene
cac1b21d39
cmd/devp2p/internal/ethtest: add more tx propagation tests (#22630)
This adds a test for large tx announcement messages, as well as a test to
check that announced tx hashes are requested by the node.
2021-04-23 18:14:39 +02:00
rene
ea54c58d4f
cmd/devp2p/internal/ethtest: run test suite as Go unit test (#22698)
This change adds a Go unit test that runs the protocol test suite
against the go-ethereum implementation of the eth protocol.
2021-04-23 11:15:42 +02:00
rene
36b51b8156
cmd/devp2p: add old block announcement test to eth test suite (#22474)
Add old block announcement test to eth test suite, checks to make sure old block announcement isn't propagated
2021-03-20 19:50:44 +01:00
rene
d3040a80d7
cmd/devp2p/internal/ethtest: skip eth/66 tests when v66 not supported (#22460) 2021-03-19 15:15:39 +01:00
rene
6a528fce33
cmd/devp2p/internal/ethtest: return request ID in BlockHeaders response (#22508)
This PR fixes an issue with the eth66 test suite where, during a readAndServe when
the test is manually responding to GetBlockHeader requests, it now responds
with a BlockHeaders eth66 packet that includes the inbound request ID.
2021-03-19 10:57:23 +01:00