Compare commits

..
Author SHA1 Message Date
i-norden b3b1695ced add index to event.emitter_addr 2023-12-11 09:08:36 -06:00
i-norden 36298e55e6 add index to event_entry.key 2023-12-11 09:08:25 -06:00
Ian Norden b450ef684d fix: api: exclude reverted events in eth_getLogs results (#11318)
* exclude reverted events from results returned by eth_getLogs

* unit test

* update CHANGELOG.md
2023-12-11 09:07:56 -06:00
Jiaying Wang 924af42947 Merge pull request #11510 from filecoin-project/release/v1.25.1
release: v1.25.1
2023-12-09 22:47:04 +08:00
Steven Allen aa6413622e fix: exchange: explicitly cast the block message limit const (#11511)
It's not a const for the testground build, and needs to be an int 99%
of the time. So we might as well just cast here.
2023-12-09 22:31:13 +08:00
Jiaying Wang e8b5d18ee5 Update CHANGELOG.md 2023-12-09 21:27:28 +08:00
Jiaying Wang 4577f94558 Merge pull request #11508 from filecoin-project/jen/25patch
release: v1.25.1
2023-12-09 21:26:07 +08:00
jennijuju 410ca25071 hot fix - backport #11506 2023-12-09 21:05:12 +08:00
Steven Allen d532ac6169 fix lints 2023-12-09 20:57:03 +08:00
Steven Allen 514129eee5 fix: exchange: allow up to 10k messages per block
Also explicitly limit how many bytes we're willing to read in one go
such that we're capable of reading a worst-case tipset (like, really,
never going to happen worst-case). Previously, this wasn't an issue.
However, we've bumped the max number of messages from 8,192 to 150,000
and need to limit allocations somewhere else.
2023-12-09 20:56:57 +08:00
156 changed files with 1149 additions and 2931 deletions
+6 -2
View File
@@ -996,20 +996,23 @@ workflows:
- build
suite: utest-unit-cli
target: "./cli/... ./cmd/... ./api/..."
executor: golang-2xl
get-params: true
executor: golang-2xl
- test:
name: test-unit-node
requires:
- build
suite: utest-unit-node
target: "./node/..."
- test:
name: test-unit-rest
requires:
- build
suite: utest-unit-rest
target: "./blockstore/... ./build/... ./chain/... ./conformance/... ./gateway/... ./journal/... ./lib/... ./markets/... ./paychmgr/... ./tools/..."
executor: golang-2xl
- test:
name: test-unit-storage
@@ -1017,7 +1020,8 @@ workflows:
- build
suite: utest-unit-storage
target: "./storage/... ./extern/..."
get-params: true
- test:
go-test-flags: "-run=TestMulticoreSDR"
requires:
+3 -10
View File
@@ -557,16 +557,9 @@ workflows:
- build
suite: utest-[[ $suite ]]
target: "[[ $pkgs ]]"
[[- if eq $suite "unit-storage"]]
get-params: true
[[- end -]]
[[- if eq $suite "unit-cli"]]
executor: golang-2xl
get-params: true
[[- end -]]
[[- if eq $suite "unit-rest"]]
executor: golang-2xl
[[- end -]]
[[if eq $suite "unit-cli"]]get-params: true[[end]]
[[if eq $suite "unit-cli"]]executor: golang-2xl[[end]]
[[- if eq $suite "unit-rest"]]executor: golang-2xl[[end]]
[[- end]]
- test:
go-test-flags: "-run=TestMulticoreSDR"
-1
View File
@@ -41,7 +41,6 @@ build/paramfetch.sh
bin/ipget
bin/tmp/*
.idea
.vscode
scratchpad
build/builtin-actors/v*
-1
View File
@@ -14,7 +14,6 @@ linters:
- varcheck
- deadcode
- scopelint
- unused
# We don't want to skip builtin/
skip-dirs-use-default: false
+9 -15
View File
@@ -1,21 +1,14 @@
# Lotus changelog
# UNRELEASED
- chore: Auto remove local chain data when importing chain file or snapshot ([filecoin-project/lotus#11277](https://github.com/filecoin-project/lotus/pull/11277))
- feat: metric: export Mpool message count ([filecoin-project/lotus#11361](https://github.com/filecoin-project/lotus/pull/11361))
- feat: sealing: load SectorsSummary from sealing SectorStats instead of calling API each time ([filecoin-project/lotus#11353](https://github.com/filecoin-project/lotus/pull/11353))
- fix: miner info: Show correct sector state counts ([filecoin-project/lotus#11456](https://github.com/filecoin-project/lotus/pull/11456))
- feat: add support for specifying block number when calling `eth_estimateGas` ([filecoin-project/lotus#11462](https://github.com/filecoin-project/lotus/pull/11462)).
# v1.25.1 / 2023-12-09
This is a **highly recommended PATCH RELEASE.** The patch release fixes the issue were node operators trying to catch up sync were unable to sync large message blocks/epochs due to an increased number of messages on the network.
This patch release allows for up to 10k messages per block. Additionally, it introduces a limit on the amount of data that can be read at once, ensuring the system can handle worst-case scenarios.
## Improvements
- fix: Add time slicing to splitstore purging step during compaction to reduce lock congestion [filecoin-project/lotus#11269](https://github.com/filecoin-project/lotus/pull/11269)
- feat: Added instructions on how to setup Prometheus/Grafana for monitoring a local Lotus node [filecoin-project/lotus#11276](https://github.com/filecoin-project/lotus/pull/11276)
- fix: Exclude reverted events in `eth_getLogs` results [filecoin-project/lotus#11318](https://github.com/filecoin-project/lotus/pull/11318)
- fix: The Ethereum API will now use the correct state-tree when resolving "native" addresses into masked ID addresses. Additionally, pending messages from native account types won't be visible in the Ethereum API because there is no "correct" state-tree to pick in this case. However, pending _Ethereum_ transactions and native messages that have landed on-chain will still be visible through the Ethereum API.
- feat: Unambiguously translate native messages to Ethereum transactions by:
- Detecting native messages that "look" like Ethereum transactions (creating smart contracts, invoking a smart contract, etc.), and decoding them as such.
- Otherwise, ABI-encoding the inputs as if they were calls to a `handle_filecoin_method` Solidity method.
- fix: ensure that the Ethereum API never returns "empty" addresses for native messages. When a "to" address cannot be resolved to a 0x-style address, it will be re-written to `0xff0000000000000000000000ffffffffffffffff`. This can only happen when the native transaction _reverted_ (failing to create an account at the specified "to" address).
- fix: exchange: allow up to 10k messages per block ([filecoin-project/lotus#11506](https://github.com/filecoin-project/lotus/pull/11506))
# v 1.25.0 / 2023-11-22
@@ -101,6 +94,7 @@ Lotus-workers can now be built to leverage the SupraSeal C2 sealing optimization
- fix(client): single-root error message ([filecoin-project/lotus#11214](https://github.com/filecoin-project/lotus/pull/11214))
- fix: worker: Convert `DC_[SectorSize]_[ResourceRestriction]` if set ([filecoin-project/lotus#11224](https://github.com/filecoin-project/lotus/pull/11224))
- chore: backport #11338 onto release/v1.25.0 ([filecoin-project/lotus#11350](https://github.com/filecoin-project/lotus/pull/11350))
- fix: Exclude reverted events in `eth_getLogs` results [filecoin-project/lotus#11318](https://github.com/filecoin-project/lotus/pull/11318)
## Dependencies
- deps: update go-libp2p to v0.28.1 ([filecoin-project/lotus#10998](https://github.com/filecoin-project/lotus/pull/10998))
@@ -262,7 +256,7 @@ This feature release requires a **minimum Go version of v1.19.12 or higher to su
- feat: sealing: flag to run data_cid untied from addpiece ([filecoin-project/lotus#10797](https://github.com/filecoin-project/lotus/pull/10797))
- feat: Lotus Gateway: add MpoolPending, ChainGetBlock and MinerGetBaseInfo ([filecoin-project/lotus#10929](https://github.com/filecoin-project/lotus/pull/10929))
## Improvements && Bug Fixes
## Improvements && Bug Fixe
- chore: update ffi & fvm ([filecoin-project/lotus#11040](https://github.com/filecoin-project/lotus/pull/11040))
- feat: Make sure we don't store duplidate actor events caused to reorgs in events.db ([filecoin-project/lotus#11015](https://github.com/filecoin-project/lotus/pull/11015))
- sealing: Use only non-assigned deals when selecting snap sectors ([filecoin-project/lotus#11002](https://github.com/filecoin-project/lotus/pull/11002))
+1 -1
View File
@@ -73,7 +73,7 @@ All releases under an odd minor version number indicate **feature releases**. Th
Feature releases include new development and bug fixes. They are not mandatory, but still highly recommended, **as they may contain critical security fixes**. Note that some of these releases may be very small patch releases that include critical hotfixes. There is no way to distinguish between a bug fix release and a feature release on the "feature" version. Both cases will use the "patch" version number.
We aim to ship a new feature release of the Lotus software from our development (master) branch every 3 weeks, so users can expect a regular cadence of Lotus feature releases. Note that mandatory releases for network upgrades may disrupt this schedule. For more, see the [Release Cycle section](#release-cycle).
We aim to ship a new feature release of the Lotus software from our development (master) branch every 3 weeks, so users can expect a regular cadence of Lotus feature releases. Note that mandatory releases for network upgrades may disrupt this schedule. For more, see the Release Cycle section (TODO: Link).
### Examples Scenarios
+2 -4
View File
@@ -71,10 +71,10 @@ For other distributions you can find the required dependencies [here.](https://l
#### Go
To build Lotus, you need a working installation of [Go 1.20.10 or higher](https://golang.org/dl/):
To build Lotus, you need a working installation of [Go 1.19.12 or higher](https://golang.org/dl/):
```bash
wget -c https://golang.org/dl/go1.20.10.linux-amd64.tar.gz -O - | sudo tar -xz -C /usr/local
wget -c https://golang.org/dl/go1.19.12.linux-amd64.tar.gz -O - | sudo tar -xz -C /usr/local
```
**TIP:**
@@ -133,8 +133,6 @@ Note: The default branch `master` is the dev branch where the latest new feature
6. You should now have Lotus installed. You can now [start the Lotus daemon and sync the chain](https://lotus.filecoin.io/lotus/install/linux/#start-the-lotus-daemon-and-sync-the-chain).
7. (Optional) Follow the [Setting Up Prometheus and Grafana](https://github.com/filecoin-project/lotus/tree/master/metrics/README.md) guide for detailed instructions on setting up a working monitoring system running against a local running lotus node.
## License
Dual-licensed under [MIT](https://github.com/filecoin-project/lotus/blob/master/LICENSE-MIT) + [Apache 2.0](https://github.com/filecoin-project/lotus/blob/master/LICENSE-APACHE)
+1 -1
View File
@@ -824,7 +824,7 @@ type FullNode interface {
EthFeeHistory(ctx context.Context, p jsonrpc.RawParams) (ethtypes.EthFeeHistory, error) //perm:read
EthMaxPriorityFeePerGas(ctx context.Context) (ethtypes.EthBigInt, error) //perm:read
EthEstimateGas(ctx context.Context, p jsonrpc.RawParams) (ethtypes.EthUint64, error) //perm:read
EthEstimateGas(ctx context.Context, tx ethtypes.EthCall) (ethtypes.EthUint64, error) //perm:read
EthCall(ctx context.Context, tx ethtypes.EthCall, blkParam ethtypes.EthBlockNumberOrHash) (ethtypes.EthBytes, error) //perm:read
EthSendRawTransaction(ctx context.Context, rawTx ethtypes.EthBytes) (ethtypes.EthHash, error) //perm:read
+1 -1
View File
@@ -114,7 +114,7 @@ type Gateway interface {
EthGasPrice(ctx context.Context) (ethtypes.EthBigInt, error)
EthFeeHistory(ctx context.Context, p jsonrpc.RawParams) (ethtypes.EthFeeHistory, error)
EthMaxPriorityFeePerGas(ctx context.Context) (ethtypes.EthBigInt, error)
EthEstimateGas(ctx context.Context, p jsonrpc.RawParams) (ethtypes.EthUint64, error)
EthEstimateGas(ctx context.Context, tx ethtypes.EthCall) (ethtypes.EthUint64, error)
EthCall(ctx context.Context, tx ethtypes.EthCall, blkParam ethtypes.EthBlockNumberOrHash) (ethtypes.EthBytes, error)
EthSendRawTransaction(ctx context.Context, rawTx ethtypes.EthBytes) (ethtypes.EthHash, error)
EthGetLogs(ctx context.Context, filter *ethtypes.EthFilterSpec) (*ethtypes.EthFilterResult, error)
+1 -1
View File
@@ -1042,7 +1042,7 @@ func (mr *MockFullNodeMockRecorder) EthChainId(arg0 interface{}) *gomock.Call {
}
// EthEstimateGas mocks base method.
func (m *MockFullNode) EthEstimateGas(arg0 context.Context, arg1 jsonrpc.RawParams) (ethtypes.EthUint64, error) {
func (m *MockFullNode) EthEstimateGas(arg0 context.Context, arg1 ethtypes.EthCall) (ethtypes.EthUint64, error) {
m.ctrl.T.Helper()
ret := m.ctrl.Call(m, "EthEstimateGas", arg0, arg1)
ret0, _ := ret[0].(ethtypes.EthUint64)
+6 -6
View File
@@ -255,7 +255,7 @@ type FullNodeMethods struct {
EthChainId func(p0 context.Context) (ethtypes.EthUint64, error) `perm:"read"`
EthEstimateGas func(p0 context.Context, p1 jsonrpc.RawParams) (ethtypes.EthUint64, error) `perm:"read"`
EthEstimateGas func(p0 context.Context, p1 ethtypes.EthCall) (ethtypes.EthUint64, error) `perm:"read"`
EthFeeHistory func(p0 context.Context, p1 jsonrpc.RawParams) (ethtypes.EthFeeHistory, error) `perm:"read"`
@@ -679,7 +679,7 @@ type GatewayMethods struct {
EthChainId func(p0 context.Context) (ethtypes.EthUint64, error) ``
EthEstimateGas func(p0 context.Context, p1 jsonrpc.RawParams) (ethtypes.EthUint64, error) ``
EthEstimateGas func(p0 context.Context, p1 ethtypes.EthCall) (ethtypes.EthUint64, error) ``
EthFeeHistory func(p0 context.Context, p1 jsonrpc.RawParams) (ethtypes.EthFeeHistory, error) ``
@@ -2134,14 +2134,14 @@ func (s *FullNodeStub) EthChainId(p0 context.Context) (ethtypes.EthUint64, error
return *new(ethtypes.EthUint64), ErrNotSupported
}
func (s *FullNodeStruct) EthEstimateGas(p0 context.Context, p1 jsonrpc.RawParams) (ethtypes.EthUint64, error) {
func (s *FullNodeStruct) EthEstimateGas(p0 context.Context, p1 ethtypes.EthCall) (ethtypes.EthUint64, error) {
if s.Internal.EthEstimateGas == nil {
return *new(ethtypes.EthUint64), ErrNotSupported
}
return s.Internal.EthEstimateGas(p0, p1)
}
func (s *FullNodeStub) EthEstimateGas(p0 context.Context, p1 jsonrpc.RawParams) (ethtypes.EthUint64, error) {
func (s *FullNodeStub) EthEstimateGas(p0 context.Context, p1 ethtypes.EthCall) (ethtypes.EthUint64, error) {
return *new(ethtypes.EthUint64), ErrNotSupported
}
@@ -4400,14 +4400,14 @@ func (s *GatewayStub) EthChainId(p0 context.Context) (ethtypes.EthUint64, error)
return *new(ethtypes.EthUint64), ErrNotSupported
}
func (s *GatewayStruct) EthEstimateGas(p0 context.Context, p1 jsonrpc.RawParams) (ethtypes.EthUint64, error) {
func (s *GatewayStruct) EthEstimateGas(p0 context.Context, p1 ethtypes.EthCall) (ethtypes.EthUint64, error) {
if s.Internal.EthEstimateGas == nil {
return *new(ethtypes.EthUint64), ErrNotSupported
}
return s.Internal.EthEstimateGas(p0, p1)
}
func (s *GatewayStub) EthEstimateGas(p0 context.Context, p1 jsonrpc.RawParams) (ethtypes.EthUint64, error) {
func (s *GatewayStub) EthEstimateGas(p0 context.Context, p1 ethtypes.EthCall) (ethtypes.EthUint64, error) {
return *new(ethtypes.EthUint64), ErrNotSupported
}
+1 -9
View File
@@ -56,17 +56,9 @@ type PubsubScore struct {
Score *pubsub.PeerScoreSnapshot
}
// MessageSendSpec contains optional fields which modify message sending behavior
type MessageSendSpec struct {
// MaxFee specifies a cap on network fees related to this message
MaxFee abi.TokenAmount
// MsgUuid specifies a unique message identifier which can be used on node (or node cluster)
// level to prevent double-sends of messages even when nonce generation is not handled by sender
MaxFee abi.TokenAmount
MsgUuid uuid.UUID
// MaximizeFeeCap makes message FeeCap be based entirely on MaxFee
MaximizeFeeCap bool
}
type MpoolMessageWhole struct {
-14
View File
@@ -183,17 +183,3 @@ func (b *idstore) Close() error {
func (b *idstore) Flush(ctx context.Context) error {
return b.bs.Flush(ctx)
}
func (b *idstore) CollectGarbage(ctx context.Context, options ...BlockstoreGCOption) error {
if bs, ok := b.bs.(BlockstoreGC); ok {
return bs.CollectGarbage(ctx, options...)
}
return xerrors.Errorf("not supported")
}
func (b *idstore) GCOnce(ctx context.Context, options ...BlockstoreGCOption) error {
if bs, ok := b.bs.(BlockstoreGCOnce); ok {
return bs.GCOnce(ctx, options...)
}
return xerrors.Errorf("not supported")
}
+1
View File
@@ -182,6 +182,7 @@ type SplitStore struct {
compactionIndex int64
pruneIndex int64
onlineGCCnt int64
ctx context.Context
cancel func()
+2 -15
View File
@@ -66,9 +66,8 @@ var (
)
const (
batchSize = 16384
cidKeySize = 128
purgeWorkSliceDuration = time.Second
batchSize = 16384
cidKeySize = 128
)
func (s *SplitStore) HeadChange(_, apply []*types.TipSet) error {
@@ -1373,21 +1372,9 @@ func (s *SplitStore) purge(coldr *ColdSetReader, checkpoint *Checkpoint, markSet
return err
}
now := time.Now()
err := coldr.ForEach(func(c cid.Cid) error {
batch = append(batch, c)
if len(batch) == batchSize {
// add some time slicing to the purge as this a very disk I/O heavy operation that
// requires write access to txnLk that may starve other operations that require
// access to the blockstore.
elapsed := time.Since(now)
if elapsed > purgeWorkSliceDuration {
// work 1 slice, sleep 4 slices, or 20% utilization
time.Sleep(4 * elapsed)
now = time.Now()
}
return deleteBatch()
}
+5
View File
@@ -1,6 +1,9 @@
/dns4/bootstrap-0.mainnet.filops.net/tcp/1347/p2p/12D3KooWCVe8MmsEMes2FzgTpt9fXtmCY7wrq91GRiaC8PHSCCBj
/dns4/bootstrap-1.mainnet.filops.net/tcp/1347/p2p/12D3KooWCwevHg1yLCvktf2nvLu7L9894mcrJR4MsBCcm4syShVc
/dns4/bootstrap-2.mainnet.filops.net/tcp/1347/p2p/12D3KooWEWVwHGn2yR36gKLozmb4YjDJGerotAPGxmdWZx2nxMC4
/dns4/bootstrap-3.mainnet.filops.net/tcp/1347/p2p/12D3KooWKhgq8c7NQ9iGjbyK7v7phXvG6492HQfiDaGHLHLQjk7R
/dns4/bootstrap-4.mainnet.filops.net/tcp/1347/p2p/12D3KooWL6PsFNPhYftrJzGgF5U18hFoaVhfGk7xwzD8yVrHJ3Uc
/dns4/bootstrap-5.mainnet.filops.net/tcp/1347/p2p/12D3KooWLFynvDQiUpXoHroV1YxKHhPJgysQGH2k3ZGwtWzR4dFH
/dns4/bootstrap-6.mainnet.filops.net/tcp/1347/p2p/12D3KooWP5MwCiqdMETF9ub1P3MbCvQCcfconnYHbWg6sUJcDRQQ
/dns4/bootstrap-7.mainnet.filops.net/tcp/1347/p2p/12D3KooWRs3aY1p3juFjPy8gPN95PEQChm2QKGUCAdcDCC4EBMKf
/dns4/bootstrap-8.mainnet.filops.net/tcp/1347/p2p/12D3KooWScFR7385LTyR4zU1bYdzSiiAb5rnNABfVahPvVSzyTkR
@@ -8,5 +11,7 @@
/dns4/bootstrap-0.starpool.in/tcp/12757/p2p/12D3KooWGHpBMeZbestVEWkfdnC9u7p6uFHXL1n7m1ZBqsEmiUzz
/dns4/bootstrap-1.starpool.in/tcp/12757/p2p/12D3KooWQZrGH1PxSNZPum99M1zNvjNFM33d1AAu5DcvdHptuU7u
/dns4/node.glif.io/tcp/1235/p2p/12D3KooWBF8cpp65hp2u9LK5mh19x67ftAam84z9LsfaquTDSBpt
/dns4/bootstrap-0.ipfsmain.cn/tcp/34721/p2p/12D3KooWQnwEGNqcM2nAcPtRR9rAX8Hrg4k9kJLCHoTR5chJfz6d
/dns4/bootstrap-1.ipfsmain.cn/tcp/34723/p2p/12D3KooWMKxMkD5DMpSWsW7dBddKxKT7L2GgbNuckz9otxvkvByP
/dns4/bootstarp-0.1475.io/tcp/61256/p2p/12D3KooWRzCVDwHUkgdK7eRgnoXbjDAELhxPErjHzbRLguSV1aRt
/dns4/bootstrap-venus.mainnet.filincubator.com/tcp/8888/p2p/QmQu8C6deXwKvJP2D8B6QGyhngc3ZiDnFzEHBDx8yeBXST
-3
View File
@@ -5,8 +5,5 @@ import (
)
func IsNearUpgrade(epoch, upgradeEpoch abi.ChainEpoch) bool {
if upgradeEpoch < 0 {
return false
}
return epoch > upgradeEpoch-Finality && epoch < upgradeEpoch+Finality
}
Binary file not shown.
Binary file not shown.
Binary file not shown.
Binary file not shown.
+1 -1
View File
@@ -37,7 +37,7 @@ func BuildTypeString() string {
}
// BuildVersion is the local build version
const BuildVersion = "1.25.1-dev"
const BuildVersion = "1.25.1"
func UserVersion() string {
if os.Getenv("LOTUS_VERSION_IGNORE_COMMIT") == "1" {
+1 -1
View File
@@ -72,7 +72,7 @@ func (s *state{{.v}}) AvailableBalance(bal abi.TokenAmount) (available abi.Token
available = abi.NewTokenAmount(0)
}
}()
// this panics if the miner doesn't have enough funds to cover their locked pledge
// this panics if the miner doesnt have enough funds to cover their locked pledge
available{{if (ge .v 2)}}, err{{end}} = s.GetAvailableBalance(bal)
return available, err
}
+1 -1
View File
@@ -62,7 +62,7 @@ func (s *state0) AvailableBalance(bal abi.TokenAmount) (available abi.TokenAmoun
available = abi.NewTokenAmount(0)
}
}()
// this panics if the miner doesn't have enough funds to cover their locked pledge
// this panics if the miner doesnt have enough funds to cover their locked pledge
available = s.GetAvailableBalance(bal)
return available, err
}
+1 -1
View File
@@ -62,7 +62,7 @@ func (s *state10) AvailableBalance(bal abi.TokenAmount) (available abi.TokenAmou
available = abi.NewTokenAmount(0)
}
}()
// this panics if the miner doesn't have enough funds to cover their locked pledge
// this panics if the miner doesnt have enough funds to cover their locked pledge
available, err = s.GetAvailableBalance(bal)
return available, err
}
+1 -1
View File
@@ -62,7 +62,7 @@ func (s *state11) AvailableBalance(bal abi.TokenAmount) (available abi.TokenAmou
available = abi.NewTokenAmount(0)
}
}()
// this panics if the miner doesn't have enough funds to cover their locked pledge
// this panics if the miner doesnt have enough funds to cover their locked pledge
available, err = s.GetAvailableBalance(bal)
return available, err
}
+1 -1
View File
@@ -62,7 +62,7 @@ func (s *state12) AvailableBalance(bal abi.TokenAmount) (available abi.TokenAmou
available = abi.NewTokenAmount(0)
}
}()
// this panics if the miner doesn't have enough funds to cover their locked pledge
// this panics if the miner doesnt have enough funds to cover their locked pledge
available, err = s.GetAvailableBalance(bal)
return available, err
}
+1 -1
View File
@@ -61,7 +61,7 @@ func (s *state2) AvailableBalance(bal abi.TokenAmount) (available abi.TokenAmoun
available = abi.NewTokenAmount(0)
}
}()
// this panics if the miner doesn't have enough funds to cover their locked pledge
// this panics if the miner doesnt have enough funds to cover their locked pledge
available, err = s.GetAvailableBalance(bal)
return available, err
}
+1 -1
View File
@@ -62,7 +62,7 @@ func (s *state3) AvailableBalance(bal abi.TokenAmount) (available abi.TokenAmoun
available = abi.NewTokenAmount(0)
}
}()
// this panics if the miner doesn't have enough funds to cover their locked pledge
// this panics if the miner doesnt have enough funds to cover their locked pledge
available, err = s.GetAvailableBalance(bal)
return available, err
}
+1 -1
View File
@@ -62,7 +62,7 @@ func (s *state4) AvailableBalance(bal abi.TokenAmount) (available abi.TokenAmoun
available = abi.NewTokenAmount(0)
}
}()
// this panics if the miner doesn't have enough funds to cover their locked pledge
// this panics if the miner doesnt have enough funds to cover their locked pledge
available, err = s.GetAvailableBalance(bal)
return available, err
}
+1 -1
View File
@@ -62,7 +62,7 @@ func (s *state5) AvailableBalance(bal abi.TokenAmount) (available abi.TokenAmoun
available = abi.NewTokenAmount(0)
}
}()
// this panics if the miner doesn't have enough funds to cover their locked pledge
// this panics if the miner doesnt have enough funds to cover their locked pledge
available, err = s.GetAvailableBalance(bal)
return available, err
}
+1 -1
View File
@@ -62,7 +62,7 @@ func (s *state6) AvailableBalance(bal abi.TokenAmount) (available abi.TokenAmoun
available = abi.NewTokenAmount(0)
}
}()
// this panics if the miner doesn't have enough funds to cover their locked pledge
// this panics if the miner doesnt have enough funds to cover their locked pledge
available, err = s.GetAvailableBalance(bal)
return available, err
}
+1 -1
View File
@@ -62,7 +62,7 @@ func (s *state7) AvailableBalance(bal abi.TokenAmount) (available abi.TokenAmoun
available = abi.NewTokenAmount(0)
}
}()
// this panics if the miner doesn't have enough funds to cover their locked pledge
// this panics if the miner doesnt have enough funds to cover their locked pledge
available, err = s.GetAvailableBalance(bal)
return available, err
}
+1 -1
View File
@@ -62,7 +62,7 @@ func (s *state8) AvailableBalance(bal abi.TokenAmount) (available abi.TokenAmoun
available = abi.NewTokenAmount(0)
}
}()
// this panics if the miner doesn't have enough funds to cover their locked pledge
// this panics if the miner doesnt have enough funds to cover their locked pledge
available, err = s.GetAvailableBalance(bal)
return available, err
}
+1 -1
View File
@@ -62,7 +62,7 @@ func (s *state9) AvailableBalance(bal abi.TokenAmount) (available abi.TokenAmoun
available = abi.NewTokenAmount(0)
}
}()
// this panics if the miner doesn't have enough funds to cover their locked pledge
// this panics if the miner doesnt have enough funds to cover their locked pledge
available, err = s.GetAvailableBalance(bal)
return available, err
}
+1 -1
View File
@@ -29,7 +29,7 @@ import (
"github.com/filecoin-project/lotus/chain/types"
)
// Load returns an abstract copy of payment channel state, regardless of actor version
// Load returns an abstract copy of payment channel state, irregardless of actor version
func Load(store adt.Store, act *types.Actor) (State, error) {
if name, av, ok := actors.GetActorMetaByCode(act.Code); ok {
if name != manifest.PaychKey {
+1 -1
View File
@@ -29,7 +29,7 @@ import (
"github.com/filecoin-project/lotus/chain/types"
)
// Load returns an abstract copy of payment channel state, regardless of actor version
// Load returns an abstract copy of payment channel state, irregardless of actor version
func Load(store adt.Store, act *types.Actor) (State, error) {
if name, av, ok := actors.GetActorMetaByCode(act.Code); ok {
if name != manifest.PaychKey {
+1 -1
View File
@@ -13,7 +13,7 @@ import (
func SerializeParams(i cbg.CBORMarshaler) ([]byte, aerrors.ActorError) {
buf := new(bytes.Buffer)
if err := i.MarshalCBOR(buf); err != nil {
// TODO: shouldn't this be a fatal error?
// TODO: shouldnt this be a fatal error?
return nil, aerrors.Absorb(err, exitcode.ErrSerialization, "failed to encode parameter")
}
return buf.Bytes(), nil
+13
View File
@@ -29,6 +29,19 @@ import (
var log = logging.Logger("drand")
type drandPeer struct {
addr string
tls bool
}
func (dp *drandPeer) Address() string {
return dp.addr
}
func (dp *drandPeer) IsTLS() bool {
return dp.tls
}
// DrandBeacon connects Lotus with a drand network in order to provide
// randomness to the system in a way that's aligned with Filecoin rounds/epochs.
//
+2 -2
View File
@@ -17,7 +17,7 @@ import (
)
func TestPrintGroupInfo(t *testing.T) {
server := build.DrandConfigs[build.DrandTestnet].Servers[0]
server := build.DrandConfigs[build.DrandDevnet].Servers[0]
c, err := hclient.New(server, nil, nil)
assert.NoError(t, err)
cg := c.(interface {
@@ -31,7 +31,7 @@ func TestPrintGroupInfo(t *testing.T) {
func TestMaxBeaconRoundForEpoch(t *testing.T) {
todayTs := uint64(1652222222)
db, err := NewDrandBeacon(todayTs, build.BlockDelaySecs, nil, build.DrandConfigs[build.DrandTestnet])
db, err := NewDrandBeacon(todayTs, build.BlockDelaySecs, nil, build.DrandConfigs[build.DrandDevnet])
assert.NoError(t, err)
mbr15 := db.MaxBeaconRoundForEpoch(network.Version15, 100)
mbr16 := db.MaxBeaconRoundForEpoch(network.Version16, 100)
+3
View File
@@ -147,6 +147,9 @@ func (t *TipSetExecutor) ApplyBlocks(ctx context.Context,
return xerrors.Errorf("callback failed on cron message: %w", err)
}
}
if ret.ExitCode != 0 {
return xerrors.Errorf("cron exit was non-zero: %d", ret.ExitCode)
}
return nil
}
+3
View File
@@ -91,6 +91,9 @@ var RewardFunc = func(ctx context.Context, vmi vm.Interface, em stmgr.ExecMonito
}
}
if ret.ExitCode != 0 {
return xerrors.Errorf("reward application message failed (exit %d): %s", ret.ExitCode, ret.ActorErr)
}
return nil
}
+1 -1
View File
@@ -180,7 +180,7 @@ func (e *heightEventsObserver) Revert(ctx context.Context, from, to *types.TipSe
// Update the head first so we don't accidental skip reverting a concurrent call to ChainAt.
e.updateHead(to)
// Call revert on all heights between the two tipsets, handling empty tipsets.
// Call revert on all hights between the two tipsets, handling empty tipsets.
for h := from.Height(); h > to.Height(); h-- {
e.lk.Lock()
triggers := e.tsHeights[h]
+4 -4
View File
@@ -673,7 +673,7 @@ func TestCalled(t *testing.T) {
}, 3, 20, matchAddrMethod(t0123, 5))
require.NoError(t, err)
// create few blocks to make sure nothing gets randomly called
// create few blocks to make sure nothing get's randomly called
fcs.advance(0, 4, 0, nil) // H=5
require.Equal(t, false, applied)
@@ -991,7 +991,7 @@ func TestCalledNull(t *testing.T) {
}, 3, 20, matchAddrMethod(t0123, 5))
require.NoError(t, err)
// create few blocks to make sure nothing gets randomly called
// create few blocks to make sure nothing get's randomly called
fcs.advance(0, 4, 0, nil) // H=5
require.Equal(t, false, applied)
@@ -1050,7 +1050,7 @@ func TestRemoveTriggersOnMessage(t *testing.T) {
}, 3, 20, matchAddrMethod(t0123, 5))
require.NoError(t, err)
// create few blocks to make sure nothing gets randomly called
// create few blocks to make sure nothing get's randomly called
fcs.advance(0, 4, 0, nil) // H=5
require.Equal(t, false, applied)
@@ -1155,7 +1155,7 @@ func TestStateChanged(t *testing.T) {
})
require.NoError(t, err)
// create few blocks to make sure nothing gets randomly called
// create few blocks to make sure nothing get's randomly called
fcs.advance(0, 4, 0, nil) // H=5
require.Equal(t, false, applied)
+1 -1
View File
@@ -326,7 +326,7 @@ func NewEventIndex(ctx context.Context, path string, chainStore *store.ChainStor
}
if version == 2 {
log.Infof("upgrading event index from version 2 to version 3")
log.Infof("upgrading event index from version 1 to version 2")
// to upgrade to version 3 we only need to create an index on the event_entry.key column
// and on the event.emitter_addr column
+25
View File
@@ -0,0 +1,25 @@
package state
import (
"context"
"github.com/ipfs/go-cid"
cbor "github.com/ipfs/go-ipld-cbor"
)
type contextStore struct {
ctx context.Context
cst *cbor.BasicIpldStore
}
func (cs *contextStore) Context() context.Context {
return cs.ctx
}
func (cs *contextStore) Get(ctx context.Context, c cid.Cid, out interface{}) error {
return cs.cst.Get(ctx, c, out)
}
func (cs *contextStore) Put(ctx context.Context, v interface{}) (cid.Cid, error) {
return cs.cst.Put(ctx, v)
}
+25 -115
View File
@@ -306,9 +306,9 @@ func (t *Response) UnmarshalCBOR(r io.Reader) (err error) {
return nil
}
var lengthBufCompactedMessages = []byte{132}
var lengthBufCompactedMessagesCBOR = []byte{132}
func (t *CompactedMessages) MarshalCBOR(w io.Writer) error {
func (t *CompactedMessagesCBOR) MarshalCBOR(w io.Writer) error {
if t == nil {
_, err := w.Write(cbg.CborNull)
return err
@@ -316,12 +316,12 @@ func (t *CompactedMessages) MarshalCBOR(w io.Writer) error {
cw := cbg.NewCborWriter(w)
if _, err := cw.Write(lengthBufCompactedMessages); err != nil {
if _, err := cw.Write(lengthBufCompactedMessagesCBOR); err != nil {
return err
}
// t.Bls ([]*types.Message) (slice)
if len(t.Bls) > cbg.MaxLength {
if len(t.Bls) > 150000 {
return xerrors.Errorf("Slice value in field t.Bls was too long")
}
@@ -334,7 +334,7 @@ func (t *CompactedMessages) MarshalCBOR(w io.Writer) error {
}
}
// t.BlsIncludes ([][]uint64) (slice)
// t.BlsIncludes ([]exchange.messageIndices) (slice)
if len(t.BlsIncludes) > cbg.MaxLength {
return xerrors.Errorf("Slice value in field t.BlsIncludes was too long")
}
@@ -343,24 +343,13 @@ func (t *CompactedMessages) MarshalCBOR(w io.Writer) error {
return err
}
for _, v := range t.BlsIncludes {
if len(v) > cbg.MaxLength {
return xerrors.Errorf("Slice value in field v was too long")
}
if err := cw.WriteMajorTypeHeader(cbg.MajArray, uint64(len(v))); err != nil {
if err := v.MarshalCBOR(cw); err != nil {
return err
}
for _, v := range v {
if err := cw.WriteMajorTypeHeader(cbg.MajUnsignedInt, uint64(v)); err != nil {
return err
}
}
}
// t.Secpk ([]*types.SignedMessage) (slice)
if len(t.Secpk) > cbg.MaxLength {
if len(t.Secpk) > 150000 {
return xerrors.Errorf("Slice value in field t.Secpk was too long")
}
@@ -373,7 +362,7 @@ func (t *CompactedMessages) MarshalCBOR(w io.Writer) error {
}
}
// t.SecpkIncludes ([][]uint64) (slice)
// t.SecpkIncludes ([]exchange.messageIndices) (slice)
if len(t.SecpkIncludes) > cbg.MaxLength {
return xerrors.Errorf("Slice value in field t.SecpkIncludes was too long")
}
@@ -382,26 +371,15 @@ func (t *CompactedMessages) MarshalCBOR(w io.Writer) error {
return err
}
for _, v := range t.SecpkIncludes {
if len(v) > cbg.MaxLength {
return xerrors.Errorf("Slice value in field v was too long")
}
if err := cw.WriteMajorTypeHeader(cbg.MajArray, uint64(len(v))); err != nil {
if err := v.MarshalCBOR(cw); err != nil {
return err
}
for _, v := range v {
if err := cw.WriteMajorTypeHeader(cbg.MajUnsignedInt, uint64(v)); err != nil {
return err
}
}
}
return nil
}
func (t *CompactedMessages) UnmarshalCBOR(r io.Reader) (err error) {
*t = CompactedMessages{}
func (t *CompactedMessagesCBOR) UnmarshalCBOR(r io.Reader) (err error) {
*t = CompactedMessagesCBOR{}
cr := cbg.NewCborReader(r)
@@ -430,7 +408,7 @@ func (t *CompactedMessages) UnmarshalCBOR(r io.Reader) (err error) {
return err
}
if extra > cbg.MaxLength {
if extra > 150000 {
return fmt.Errorf("t.Bls: array too large (%d)", extra)
}
@@ -471,7 +449,7 @@ func (t *CompactedMessages) UnmarshalCBOR(r io.Reader) (err error) {
}
}
// t.BlsIncludes ([][]uint64) (slice)
// t.BlsIncludes ([]exchange.messageIndices) (slice)
maj, extra, err = cr.ReadHeader()
if err != nil {
@@ -487,7 +465,7 @@ func (t *CompactedMessages) UnmarshalCBOR(r io.Reader) (err error) {
}
if extra > 0 {
t.BlsIncludes = make([][]uint64, extra)
t.BlsIncludes = make([]messageIndices, extra)
}
for i := 0; i < int(extra); i++ {
@@ -499,47 +477,13 @@ func (t *CompactedMessages) UnmarshalCBOR(r io.Reader) (err error) {
_ = extra
_ = err
maj, extra, err = cr.ReadHeader()
if err != nil {
return err
}
{
if extra > cbg.MaxLength {
return fmt.Errorf("t.BlsIncludes[i]: array too large (%d)", extra)
}
if maj != cbg.MajArray {
return fmt.Errorf("expected cbor array")
}
if extra > 0 {
t.BlsIncludes[i] = make([]uint64, extra)
}
for j := 0; j < int(extra); j++ {
{
var maj byte
var extra uint64
var err error
_ = maj
_ = extra
_ = err
{
maj, extra, err = cr.ReadHeader()
if err != nil {
return err
}
if maj != cbg.MajUnsignedInt {
return fmt.Errorf("wrong type for uint64 field")
}
t.BlsIncludes[i][j] = uint64(extra)
}
if err := t.BlsIncludes[i].UnmarshalCBOR(cr); err != nil {
return xerrors.Errorf("unmarshaling t.BlsIncludes[i]: %w", err)
}
}
}
}
}
@@ -550,7 +494,7 @@ func (t *CompactedMessages) UnmarshalCBOR(r io.Reader) (err error) {
return err
}
if extra > cbg.MaxLength {
if extra > 150000 {
return fmt.Errorf("t.Secpk: array too large (%d)", extra)
}
@@ -591,7 +535,7 @@ func (t *CompactedMessages) UnmarshalCBOR(r io.Reader) (err error) {
}
}
// t.SecpkIncludes ([][]uint64) (slice)
// t.SecpkIncludes ([]exchange.messageIndices) (slice)
maj, extra, err = cr.ReadHeader()
if err != nil {
@@ -607,7 +551,7 @@ func (t *CompactedMessages) UnmarshalCBOR(r io.Reader) (err error) {
}
if extra > 0 {
t.SecpkIncludes = make([][]uint64, extra)
t.SecpkIncludes = make([]messageIndices, extra)
}
for i := 0; i < int(extra); i++ {
@@ -619,47 +563,13 @@ func (t *CompactedMessages) UnmarshalCBOR(r io.Reader) (err error) {
_ = extra
_ = err
maj, extra, err = cr.ReadHeader()
if err != nil {
return err
}
{
if extra > cbg.MaxLength {
return fmt.Errorf("t.SecpkIncludes[i]: array too large (%d)", extra)
}
if maj != cbg.MajArray {
return fmt.Errorf("expected cbor array")
}
if extra > 0 {
t.SecpkIncludes[i] = make([]uint64, extra)
}
for j := 0; j < int(extra); j++ {
{
var maj byte
var extra uint64
var err error
_ = maj
_ = extra
_ = err
{
maj, extra, err = cr.ReadHeader()
if err != nil {
return err
}
if maj != cbg.MajUnsignedInt {
return fmt.Errorf("wrong type for uint64 field")
}
t.SecpkIncludes[i][j] = uint64(extra)
}
if err := t.SecpkIncludes[i].UnmarshalCBOR(cr); err != nil {
return xerrors.Errorf("unmarshaling t.SecpkIncludes[i]: %w", err)
}
}
}
}
}
+10 -4
View File
@@ -4,6 +4,7 @@ import (
"bufio"
"context"
"fmt"
"io"
"math/rand"
"time"
@@ -23,6 +24,10 @@ import (
"github.com/filecoin-project/lotus/lib/peermgr"
)
// Set the max exchange message size to 120MiB. Purely based on gas numbers, we can include ~8MiB of
// messages per block, so I've set this to 120MiB to be _very_ safe.
const maxExchangeMessageSize = (15 * 8) << 20
// client implements exchange.Client, using the libp2p ChainExchange protocol
// as the fetching mechanism.
type client struct {
@@ -242,7 +247,7 @@ func (c *client) processResponse(req *Request, res *Response, tipsets []*types.T
// If we didn't request the headers they should have been provided
// by the caller.
if len(tipsets) < len(res.Chain) {
return nil, xerrors.Errorf("not enough tipsets provided for message response validation, needed %d, have %d", len(res.Chain), len(tipsets))
return nil, xerrors.Errorf("not enought tipsets provided for message response validation, needed %d, have %d", len(res.Chain), len(tipsets))
}
chain := make([]*BSTipSet, 0, resLength)
for i, resChain := range res.Chain {
@@ -381,7 +386,7 @@ func (c *client) sendRequestToPeer(ctx context.Context, peer peer.ID, req *Reque
defer span.End()
if span.IsRecordingEvents() {
span.AddAttributes(
trace.StringAttribute("peer", peer.String()),
trace.StringAttribute("peer", peer.Pretty()),
)
}
defer func() {
@@ -434,10 +439,11 @@ func (c *client) sendRequestToPeer(ctx context.Context, peer peer.ID, req *Reque
log.Warnw("CloseWrite err", "error", err)
}
// Read response.
// Read response, limiting the size of the response to maxExchangeMessageSize as we allow a
// lot of messages (10k+) but they'll mostly be quite small.
var res Response
err = cborutil.ReadCborRPC(
bufio.NewReader(incrt.New(stream, ReadResMinSpeed, ReadResDeadline)),
bufio.NewReader(io.LimitReader(incrt.New(stream, ReadResMinSpeed, ReadResDeadline), maxExchangeMessageSize)),
&res)
if err != nil {
c.peerTracker.logFailure(peer, build.Clock.Since(connectionStart), req.Length)
+2
View File
@@ -154,6 +154,8 @@ type BSTipSet struct {
// FIXME: The logic to decompress this structure should belong
//
// to itself, not to the consumer.
//
// NOTE: Max messages is: BlockMessageLimit (10k) * MaxTipsetSize (15) = 150k
type CompactedMessages struct {
Bls []*types.Message
BlsIncludes [][]uint64
+125
View File
@@ -0,0 +1,125 @@
package exchange
import (
"fmt"
"io"
cbg "github.com/whyrusleeping/cbor-gen"
xerrors "golang.org/x/xerrors"
"github.com/filecoin-project/lotus/build"
types "github.com/filecoin-project/lotus/chain/types"
)
// Type used for encoding/decoding compacted messages. This is a ustom type as we need custom limits.
// - Max messages is 150,000 as that's 15 times the max block size (in messages). It needs to be
// large enough to cover a full tipset full of full blocks.
type CompactedMessagesCBOR struct {
Bls []*types.Message `cborgen:"maxlen=150000"`
BlsIncludes []messageIndices
Secpk []*types.SignedMessage `cborgen:"maxlen=150000"`
SecpkIncludes []messageIndices
}
// Unmarshal into the "decoding" struct, then copy into the actual struct.
func (t *CompactedMessages) UnmarshalCBOR(r io.Reader) (err error) {
var c CompactedMessagesCBOR
if err := c.UnmarshalCBOR(r); err != nil {
return err
}
t.Bls = c.Bls
t.BlsIncludes = make([][]uint64, len(c.BlsIncludes))
for i, v := range c.BlsIncludes {
t.BlsIncludes[i] = v.v
}
t.Secpk = c.Secpk
t.SecpkIncludes = make([][]uint64, len(c.SecpkIncludes))
for i, v := range c.SecpkIncludes {
t.SecpkIncludes[i] = v.v
}
return nil
}
// Copy into the encoding struct, then marshal.
func (t *CompactedMessages) MarshalCBOR(w io.Writer) error {
if t == nil {
_, err := w.Write(cbg.CborNull)
return err
}
var c CompactedMessagesCBOR
c.Bls = t.Bls
c.BlsIncludes = make([]messageIndices, len(t.BlsIncludes))
for i, v := range t.BlsIncludes {
c.BlsIncludes[i].v = v
}
c.Secpk = t.Secpk
c.SecpkIncludes = make([]messageIndices, len(t.SecpkIncludes))
for i, v := range t.SecpkIncludes {
c.SecpkIncludes[i].v = v
}
return c.MarshalCBOR(w)
}
// this needs to be a struct or cborgen will peak into it and ignore the Unmarshal/Marshal functions
type messageIndices struct {
v []uint64
}
func (t *messageIndices) UnmarshalCBOR(r io.Reader) (err error) {
cr := cbg.NewCborReader(r)
maj, extra, err := cr.ReadHeader()
if err != nil {
return err
}
if maj != cbg.MajArray {
return fmt.Errorf("cbor input should be of type array")
}
if extra > uint64(build.BlockMessageLimit) {
return fmt.Errorf("cbor input had wrong number of fields")
}
if extra > 0 {
t.v = make([]uint64, extra)
}
for i := 0; i < int(extra); i++ {
maj, extra, err := cr.ReadHeader()
if err != nil {
return err
}
if maj != cbg.MajUnsignedInt {
return fmt.Errorf("wrong type for uint64 field")
}
t.v[i] = extra
}
return nil
}
func (t *messageIndices) MarshalCBOR(w io.Writer) error {
if t == nil {
_, err := w.Write(cbg.CborNull)
return err
}
cw := cbg.NewCborWriter(w)
if len(t.v) > build.BlockMessageLimit {
return xerrors.Errorf("Slice value in field v was too long")
}
if err := cw.WriteMajorTypeHeader(cbg.MajArray, uint64(len(t.v))); err != nil {
return err
}
for _, v := range t.v {
if err := cw.WriteMajorTypeHeader(cbg.MajUnsignedInt, v); err != nil {
return err
}
}
return nil
}
+1 -1
View File
@@ -131,7 +131,7 @@ func NewMsgIndex(lctx context.Context, basePath string, cs ChainStore) (MsgIndex
db, err := sql.Open("sqlite3", dbPath)
if err != nil {
// TODO [nice to have]: automatically delete corrupt databases
// TODO [nice to have]: automaticaly delete corrupt databases
// but for now we can just error and let the operator delete.
return nil, xerrors.Errorf("error opening msgindex database: %w", err)
}
+17
View File
@@ -39,6 +39,23 @@ func (ps *Store) save(ctx context.Context, state *FundedAddressState) error {
return ps.ds.Put(ctx, k, b)
}
// get the state for the given address
func (ps *Store) get(ctx context.Context, addr address.Address) (*FundedAddressState, error) {
k := dskeyForAddr(addr)
data, err := ps.ds.Get(ctx, k)
if err != nil {
return nil, err
}
var state FundedAddressState
err = cborrpc.ReadCborRPC(bytes.NewReader(data), &state)
if err != nil {
return nil, err
}
return &state, nil
}
// forEach calls iter with each address in the datastore
func (ps *Store) forEach(ctx context.Context, iter func(*FundedAddressState)) error {
res, err := ps.ds.Query(ctx, dsq.Query{Prefix: dsKeyAddr})
+7 -13
View File
@@ -21,7 +21,6 @@ import (
pubsub "github.com/libp2p/go-libp2p-pubsub"
"github.com/minio/blake2b-simd"
"github.com/raulk/clock"
"go.opencensus.io/stats"
"golang.org/x/xerrors"
ffi "github.com/filecoin-project/filecoin-ffi"
@@ -211,10 +210,8 @@ func ComputeRBF(curPrem abi.TokenAmount, replaceByFeeRatio types.Percent) abi.To
func CapGasFee(mff dtypes.DefaultMaxFeeFunc, msg *types.Message, sendSpec *api.MessageSendSpec) {
var maxFee abi.TokenAmount
var maximizeFeeCap bool
if sendSpec != nil {
maxFee = sendSpec.MaxFee
maximizeFeeCap = sendSpec.MaximizeFeeCap
}
if maxFee.Int == nil || maxFee.Equals(big.Zero()) {
mf, err := mff()
@@ -225,12 +222,15 @@ func CapGasFee(mff dtypes.DefaultMaxFeeFunc, msg *types.Message, sendSpec *api.M
maxFee = mf
}
gaslimit := types.NewInt(uint64(msg.GasLimit))
totalFee := types.BigMul(msg.GasFeeCap, gaslimit)
if maximizeFeeCap || totalFee.GreaterThan(maxFee) {
msg.GasFeeCap = big.Div(maxFee, gaslimit)
gl := types.NewInt(uint64(msg.GasLimit))
totalFee := types.BigMul(msg.GasFeeCap, gl)
if totalFee.LessThanEqual(maxFee) {
msg.GasPremium = big.Min(msg.GasFeeCap, msg.GasPremium) // cap premium at FeeCap
return
}
msg.GasFeeCap = big.Div(maxFee, gl)
msg.GasPremium = big.Min(msg.GasFeeCap, msg.GasPremium) // cap premium at FeeCap
}
@@ -1022,9 +1022,6 @@ func (mp *MessagePool) addLocked(ctx context.Context, m *types.SignedMessage, st
}
})
// Record the current size of the Mpool
stats.Record(ctx, metrics.MpoolMessageCount.M(int64(mp.currentSize)))
return nil
}
@@ -1217,9 +1214,6 @@ func (mp *MessagePool) remove(ctx context.Context, from address.Address, nonce u
return
}
}
// Record the current size of the Mpool
stats.Record(ctx, metrics.MpoolMessageCount.M(int64(mp.currentSize)))
}
func (mp *MessagePool) Pending(ctx context.Context) ([]*types.SignedMessage, *types.TipSet) {
+2 -2
View File
@@ -570,7 +570,7 @@ func (st *StateTree) ForEach(f func(address.Address, *types.Actor) error) error
}
// no need to record anything here, there are no duplicates in the actors HAMT
// itself.
// iself.
if _, ok := seen[addr]; ok {
return nil
}
@@ -588,7 +588,7 @@ func (st *StateTree) ForEach(f func(address.Address, *types.Actor) error) error
}
// no need to record anything here, there are no duplicates in the actors HAMT
// itself.
// iself.
if _, ok := seen[addr]; ok {
return nil
}
+5
View File
@@ -235,6 +235,11 @@ func (sm *StateManager) hasExpensiveForkBetween(parent, height abi.ChainEpoch) b
return false
}
func (sm *StateManager) hasExpensiveFork(height abi.ChainEpoch) bool {
_, ok := sm.expensiveUpgrades[height]
return ok
}
func runPreMigration(ctx context.Context, sm *StateManager, fn PreMigrationFunc, cache *nv16.MemMigrationCache, ts *types.TipSet) {
height := ts.Height()
parent := ts.ParentState()
+1 -1
View File
@@ -42,7 +42,7 @@ func TestIndexSeeks(t *testing.T) {
cs := store.NewChainStore(nbs, nbs, syncds.MutexWrap(datastore.NewMapDatastore()), filcns.Weight, nil)
defer cs.Close() //nolint:errcheck
_, _, err = cs.Import(ctx, bytes.NewReader(gencar))
_, err = cs.Import(ctx, bytes.NewReader(gencar))
if err != nil {
t.Fatal(err)
}
+7 -2
View File
@@ -212,8 +212,13 @@ func (cs *ChainStore) MessagesForTipset(ctx context.Context, ts *types.TipSet) (
var out []types.ChainMsg
for _, bm := range bmsgs {
out = append(out, bm.BlsMessages...)
out = append(out, bm.SecpkMessages...)
for _, blsm := range bm.BlsMessages {
out = append(out, blsm)
}
for _, secm := range bm.SecpkMessages {
out = append(out, secm)
}
}
return out, nil
+13 -39
View File
@@ -60,7 +60,7 @@ func (cs *ChainStore) Export(ctx context.Context, ts *types.TipSet, inclRecentRo
})
}
func (cs *ChainStore) Import(ctx context.Context, r io.Reader) (head *types.TipSet, genesis *types.BlockHeader, err error) {
func (cs *ChainStore) Import(ctx context.Context, r io.Reader) (*types.TipSet, error) {
// TODO: writing only to the state blockstore is incorrect.
// At this time, both the state and chain blockstores are backed by the
// universal store. When we physically segregate the stores, we will need
@@ -69,7 +69,7 @@ func (cs *ChainStore) Import(ctx context.Context, r io.Reader) (head *types.TipS
br, err := carv2.NewBlockReader(r)
if err != nil {
return nil, nil, xerrors.Errorf("loadcar failed: %w", err)
return nil, xerrors.Errorf("loadcar failed: %w", err)
}
s := cs.StateBlockstore()
@@ -80,51 +80,27 @@ func (cs *ChainStore) Import(ctx context.Context, r io.Reader) (head *types.TipS
putThrottle <- nil
}
if len(br.Roots) == 0 {
return nil, nil, xerrors.Errorf("no roots in snapshot car file")
}
nextTailCid := br.Roots[0]
var tailBlock types.BlockHeader
tailBlock.Height = abi.ChainEpoch(-1)
var buf []blocks.Block
for {
blk, err := br.Next()
if err != nil {
// we're at the end
if err == io.EOF {
if len(buf) > 0 {
if err := s.PutMany(ctx, buf); err != nil {
return nil, nil, err
return nil, err
}
}
break
}
return nil, nil, err
return nil, err
}
// check for header block, looking for genesis
if blk.Cid() == nextTailCid && tailBlock.Height != 0 {
if err := tailBlock.UnmarshalCBOR(bytes.NewReader(blk.RawData())); err != nil {
return nil, nil, xerrors.Errorf("failed to unmarshal genesis block: %w", err)
}
if len(tailBlock.Parents) > 0 {
nextTailCid = tailBlock.Parents[0]
} else {
// note: even the 0th block has a parent linking to the cbor genesis block
return nil, nil, xerrors.Errorf("current block (epoch %d cid %s) has no parents", tailBlock.Height, tailBlock.Cid())
}
}
// append to batch
buf = append(buf, blk)
if len(buf) > 1000 {
if lastErr := <-putThrottle; lastErr != nil { // consume one error to have the right to add one
return nil, nil, lastErr
return nil, lastErr
}
go func(buf []blocks.Block) {
@@ -137,17 +113,13 @@ func (cs *ChainStore) Import(ctx context.Context, r io.Reader) (head *types.TipS
// check errors
for i := 0; i < parallelPuts; i++ {
if lastErr := <-putThrottle; lastErr != nil {
return nil, nil, lastErr
return nil, lastErr
}
}
if tailBlock.Height != 0 {
return nil, nil, xerrors.Errorf("expected genesis block to have height 0 (genesis), got %d: %s", tailBlock.Height, tailBlock.Cid())
}
root, err := cs.LoadTipSet(ctx, types.NewTipSetKey(br.Roots...))
if err != nil {
return nil, nil, xerrors.Errorf("failed to load root tipset from chainfile: %w", err)
return nil, xerrors.Errorf("failed to load root tipset from chainfile: %w", err)
}
ts := root
@@ -163,10 +135,10 @@ func (cs *ChainStore) Import(ctx context.Context, r io.Reader) (head *types.TipS
}
if err := cs.PersistTipsets(ctx, tssToPersist); err != nil {
return nil, nil, xerrors.Errorf("failed to persist tipsets: %w", err)
return nil, xerrors.Errorf("failed to persist tipsets: %w", err)
}
return root, &tailBlock, nil
return root, nil
}
type walkSchedTaskType int
@@ -195,7 +167,7 @@ func (t walkSchedTaskType) String() string {
case dagTask:
return "dag"
}
panic(fmt.Sprintf("unknown task %d", t))
panic(fmt.Sprintf("unknow task %d", t))
}
type walkTask struct {
@@ -684,7 +656,9 @@ func (cs *ChainStore) WalkSnapshot(ctx context.Context, ts *types.TipSet, inclRe
}
if b.Height > 0 {
blocksToWalk = append(blocksToWalk, b.Parents...)
for _, p := range b.Parents {
blocksToWalk = append(blocksToWalk, p)
}
} else {
// include the genesis block
cids = append(cids, b.Parents...)
+3 -3
View File
@@ -118,7 +118,7 @@ func TestChainExportImport(t *testing.T) {
cs := store.NewChainStore(nbs, nbs, datastore.NewMapDatastore(), filcns.Weight, nil)
defer cs.Close() //nolint:errcheck
root, _, err := cs.Import(context.TODO(), buf)
root, err := cs.Import(context.TODO(), buf)
if err != nil {
t.Fatal(err)
}
@@ -153,7 +153,7 @@ func TestChainImportTipsetKeyCid(t *testing.T) {
cs := store.NewChainStore(nbs, nbs, datastore.NewMapDatastore(), filcns.Weight, nil)
defer cs.Close() //nolint:errcheck
root, _, err := cs.Import(ctx, buf)
root, err := cs.Import(ctx, buf)
require.NoError(t, err)
require.Truef(t, root.Equals(last), "imported chain differed from exported chain")
@@ -202,7 +202,7 @@ func TestChainExportImportFull(t *testing.T) {
cs := store.NewChainStore(nbs, nbs, ds, filcns.Weight, nil)
defer cs.Close() //nolint:errcheck
root, _, err := cs.Import(context.TODO(), buf)
root, err := cs.Import(context.TODO(), buf)
if err != nil {
t.Fatal(err)
}
+2 -2
View File
@@ -516,7 +516,7 @@ func (v *IndexerMessageValidator) Validate(ctx context.Context, pid peer.ID, msg
return pubsub.ValidationReject
}
if len(idxrMsg.ExtraData) == 0 {
log.Debugw("ignoring message missing miner id", "peer", originPeer)
log.Debugw("ignoring messsage missing miner id", "peer", originPeer)
return pubsub.ValidationIgnore
}
@@ -552,7 +552,7 @@ func (v *IndexerMessageValidator) Validate(ctx context.Context, pid peer.ID, msg
// Check that the miner ID maps to the peer that sent the message.
err = v.authenticateMessage(ctx, minerAddr, originPeer)
if err != nil {
log.Warnw("cannot authenticate message", "err", err, "peer", originPeer, "minerID", minerAddr)
log.Warnw("cannot authenticate messsage", "err", err, "peer", originPeer, "minerID", minerAddr)
stats.Record(ctx, metrics.IndexerMessageValidationFailure.M(1))
return pubsub.ValidationReject
}
-61
View File
@@ -1,61 +0,0 @@
package ratelimit
import (
"testing"
)
func TestQueue(t *testing.T) {
const size = 3
q := &queue{buf: make([]int64, size)}
if q.len() != 0 {
t.Fatalf("q.len() = %d, expect 0", q.len())
}
if q.cap() != size {
t.Fatalf("q.cap() = %d, expect %d", q.cap(), size)
}
for i := int64(0); i < int64(size); i++ {
err := q.push(i)
if err != nil {
t.Fatalf("cannot push element %d", i)
}
}
if q.len() != size {
t.Fatalf("q.len() = %d, expect %d", q.len(), size)
}
err := q.push(int64(size))
if err != ErrRateLimitExceeded {
t.Fatalf("pushing element beyond capacity should have failed with err: %s, got %s", ErrRateLimitExceeded, err)
}
if q.front() != 0 {
t.Fatalf("q.front() = %d, expect 0", q.front())
}
if q.back() != int64(size-1) {
t.Fatalf("q.back() = %d, expect %d", q.back(), size-1)
}
popVal := q.pop()
if popVal != 0 {
t.Fatalf("q.pop() = %d, expect 0", popVal)
}
if q.len() != size-1 {
t.Fatalf("q.len() = %d, expect %d", q.len(), size-1)
}
// Testing truncation.
threshold := int64(1)
q.truncate(threshold)
if q.len() != 1 {
t.Fatalf("q.len() after truncate = %d, expect 1", q.len())
}
if q.front() != 2 {
t.Fatalf("q.front() after truncate = %d, expect 2", q.front())
}
}
+1 -1
View File
@@ -844,7 +844,7 @@ loop:
return nil, xerrors.Errorf("failed to load next local tipset: %w", err)
}
if base.IsChildOf(knownParent) {
// common case: receiving a block that's potentially part of the same tipset as our best block
// common case: receiving a block thats potentially part of the same tipset as our best block
return blockSet, nil
}
+8
View File
@@ -11,6 +11,7 @@ import (
"github.com/ipfs/go-cid"
ds "github.com/ipfs/go-datastore"
logging "github.com/ipfs/go-log/v2"
"github.com/libp2p/go-libp2p/core/peer"
mocknet "github.com/libp2p/go-libp2p/p2p/net/mock"
"github.com/stretchr/testify/require"
@@ -343,6 +344,13 @@ func (tu *syncTestUtil) addClientNode() int {
return len(tu.nds) - 1
}
func (tu *syncTestUtil) pid(n int) peer.ID {
nal, err := tu.nds[n].NetAddrsListen(tu.ctx)
require.NoError(tu.t, err)
return nal.ID
}
func (tu *syncTestUtil) connect(from, to int) {
toPI, err := tu.nds[to].NetAddrsListen(tu.ctx)
require.NoError(tu.t, err)
+3
View File
@@ -7,6 +7,9 @@ import (
)
func TestDecodeBlockMsg(t *testing.T) {
type args struct {
b []byte
}
tests := []struct {
name string
data []byte
+1 -14
View File
@@ -62,14 +62,9 @@ type EthTxArgs struct {
// - BlockHash
// - BlockNumber
// - TransactionIndex
// - From
// - Hash
func EthTxFromSignedEthMessage(smsg *types.SignedMessage) (EthTx, error) {
// The from address is always an f410f address, never an ID or other address.
if !IsEthAddress(smsg.Message.From) {
return EthTx{}, xerrors.Errorf("sender must be an eth account, was %s", smsg.Message.From)
}
// Probably redundant, but we might as well check.
if smsg.Signature.Type != typescrypto.SigTypeDelegated {
return EthTx{}, xerrors.Errorf("signature is not delegated type, is type: %d", smsg.Signature.Type)
}
@@ -84,18 +79,10 @@ func EthTxFromSignedEthMessage(smsg *types.SignedMessage) (EthTx, error) {
return EthTx{}, xerrors.Errorf("failed to recover signature: %w", err)
}
from, err := EthAddressFromFilecoinAddress(smsg.Message.From)
if err != nil {
// This should be impossible as we've already asserted that we have an EthAddress
// sender...
return EthTx{}, xerrors.Errorf("sender was not an eth account")
}
return EthTx{
Nonce: EthUint64(txArgs.Nonce),
ChainID: EthUint64(txArgs.ChainID),
To: txArgs.To,
From: from,
Value: EthBigInt(txArgs.Value),
Type: Eip1559TxType,
Gas: EthUint64(txArgs.GasLimit),
-39
View File
@@ -799,45 +799,6 @@ func GetContractEthAddressFromCode(sender EthAddress, salt [32]byte, initcode []
return ethAddr, nil
}
// EthEstimateGasParams handles raw jsonrpc params for eth_estimateGas
type EthEstimateGasParams struct {
Tx EthCall
BlkParam *EthBlockNumberOrHash
}
func (e *EthEstimateGasParams) UnmarshalJSON(b []byte) error {
var params []json.RawMessage
err := json.Unmarshal(b, &params)
if err != nil {
return err
}
switch len(params) {
case 2:
err = json.Unmarshal(params[1], &e.BlkParam)
if err != nil {
return err
}
fallthrough
case 1:
err = json.Unmarshal(params[0], &e.Tx)
if err != nil {
return err
}
default:
return xerrors.Errorf("expected 1 or 2 params, got %d", len(params))
}
return nil
}
func (e EthEstimateGasParams) MarshalJSON() ([]byte, error) {
if e.BlkParam != nil {
return json.Marshal([]interface{}{e.Tx, e.BlkParam})
}
return json.Marshal([]interface{}{e.Tx})
}
// EthFeeHistoryParams handles raw jsonrpc params for eth_feeHistory
type EthFeeHistoryParams struct {
BlkCount EthUint64
+4 -4
View File
@@ -140,7 +140,7 @@ func (t *messageReceiptV0) UnmarshalCBOR(r io.Reader) (err error) {
case cbg.MajNegativeInt:
extraI = int64(extra)
if extraI < 0 {
return fmt.Errorf("int64 negative overflow")
return fmt.Errorf("int64 negative oveflow")
}
extraI = -1 - extraI
default:
@@ -186,7 +186,7 @@ func (t *messageReceiptV0) UnmarshalCBOR(r io.Reader) (err error) {
case cbg.MajNegativeInt:
extraI = int64(extra)
if extraI < 0 {
return fmt.Errorf("int64 negative overflow")
return fmt.Errorf("int64 negative oveflow")
}
extraI = -1 - extraI
default:
@@ -278,7 +278,7 @@ func (t *messageReceiptV1) UnmarshalCBOR(r io.Reader) (err error) {
case cbg.MajNegativeInt:
extraI = int64(extra)
if extraI < 0 {
return fmt.Errorf("int64 negative overflow")
return fmt.Errorf("int64 negative oveflow")
}
extraI = -1 - extraI
default:
@@ -324,7 +324,7 @@ func (t *messageReceiptV1) UnmarshalCBOR(r io.Reader) (err error) {
case cbg.MajNegativeInt:
extraI = int64(extra)
if extraI < 0 {
return fmt.Errorf("int64 negative overflow")
return fmt.Errorf("int64 negative oveflow")
}
extraI = -1 - extraI
default:
+21
View File
@@ -27,3 +27,24 @@ type StateTree interface {
Version() StateTreeVersion
}
type storageWrapper struct {
s Storage
}
func (sw *storageWrapper) Put(i cbg.CBORMarshaler) (cid.Cid, error) {
c, err := sw.s.Put(i)
if err != nil {
return cid.Undef, err
}
return c, nil
}
func (sw *storageWrapper) Get(c cid.Cid, out cbg.CBORUnmarshaler) error {
if err := sw.s.Get(c, out); err != nil {
return err
}
return nil
}
+2 -2
View File
@@ -282,7 +282,7 @@ var NetDisconnect = &cli.Command{
fmt.Println("failure")
return err
}
fmt.Printf("disconnect %s: ", pid)
fmt.Printf("disconnect %s: ", pid.Pretty())
err = api.NetDisconnect(ctx, pid)
if err != nil {
fmt.Println("failure")
@@ -312,7 +312,7 @@ var NetConnect = &cli.Command{
}
for _, pi := range pis {
fmt.Printf("connect %s: ", pi.ID)
fmt.Printf("connect %s: ", pi.ID.Pretty())
err := api.NetConnect(ctx, pi)
if err != nil {
fmt.Println("failure")
+2 -23
View File
@@ -1920,29 +1920,8 @@ var StateSysActorCIDsCmd = &cli.Command{
if err != nil {
return err
}
var actorsCidTuples []struct {
actorName string
actorCid cid.Cid
}
for name, actorCid := range actorsCids {
keyVal := struct {
actorName string
actorCid cid.Cid
}{
actorName: name,
actorCid: actorCid,
}
actorsCidTuples = append(actorsCidTuples, keyVal)
}
sort.Slice(actorsCidTuples, func(i, j int) bool {
return actorsCidTuples[i].actorName < actorsCidTuples[j].actorName
})
for _, keyVal := range actorsCidTuples {
_, _ = fmt.Fprintf(tw, "%v\t%v\n", keyVal.actorName, keyVal.actorCid)
for name, cid := range actorsCids {
_, _ = fmt.Fprintf(tw, "%v\t%v\n", name, cid)
}
return tw.Flush()
},
+1 -7
View File
@@ -14,7 +14,6 @@ import (
"github.com/filecoin-project/lotus/api/v0api"
"github.com/filecoin-project/lotus/build"
"github.com/filecoin-project/lotus/chain/types"
cliutil "github.com/filecoin-project/lotus/cli/util"
)
var SyncCmd = &cli.Command{
@@ -263,9 +262,6 @@ func SyncWait(ctx context.Context, napi v0api.FullNode, watch bool) error {
}
firstApp = state.VMApplied
// eta computes the ETA for the sync to complete (with a lookback of 10 processed items)
eta := cliutil.NewETA(10)
for {
state, err := napi.SyncState(ctx)
if err != nil {
@@ -316,10 +312,8 @@ func SyncWait(ctx context.Context, napi v0api.FullNode, watch bool) error {
fmt.Print("\r\x1b[2K\x1b[A")
}
todo := theight - ss.Height
fmt.Printf("Worker: %d; Base: %d; Target: %d (diff: %d)\n", workerID, baseHeight, theight, heightDiff)
fmt.Printf("State: %s; Current Epoch: %d; Todo: %d, ETA: %s\n", ss.Stage, ss.Height, todo, eta.Update(int64(todo)))
fmt.Printf("State: %s; Current Epoch: %d; Todo: %d\n", ss.Stage, ss.Height, theight-ss.Height)
lastLines = 2
if i%samples == 0 {
-94
View File
@@ -1,94 +0,0 @@
package cliutil
import (
"fmt"
"math"
"time"
)
// ETA implements a very simple eta calculator based on the number of remaining items. It does not
// require knowing the work size in advance and is therefore suitable for streaming workloads and
// also does not require that consecutive updates have a monotonically decreasing remaining value.
type ETA struct {
// max number of items to keep in memory
maxItems int
// a queue of most recently updated items
items []item
// we store the last calculated ETA which we reuse if there was not change in remaining items
lastETA string
}
type item struct {
timestamp time.Time
remaining int64
}
// NewETA creates a new ETA calculator of the given size
func NewETA(maxItems int) *ETA {
return &ETA{
maxItems: maxItems,
items: make([]item, 0),
}
}
// Update updates the ETA calculator with the remaining number of items and returns the ETA
func (e *ETA) Update(remaining int64) string {
item := item{
timestamp: time.Now(),
remaining: remaining,
}
if len(e.items) == 0 {
e.items = append(e.items, item)
return ""
}
if e.items[len(e.items)-1].remaining == remaining {
// we ignore updates with the same remaining value and just return the previous ETA
return e.lastETA
} else if e.items[len(e.items)-1].remaining < remaining {
// remaining went up from previous update, lets estimate how many items were processed using the
// average number processed items in the queue.
var avgProcessedPerItem int64 = 1
if len(e.items) > 1 {
diffRemaining := e.items[0].remaining - e.items[len(e.items)-1].remaining
avgProcessedPerItem = int64(math.Round(float64(diffRemaining) / float64(len(e.items))))
}
// diff is the difference in increase in remaining since last update plus the average number of processed
// items we estimate that were processed this round
diff := remaining - e.items[len(e.items)-1].remaining + avgProcessedPerItem
// we update all items in the queue by shifting their remaining value accordingly. This means that we
// always have strictly decreasing remaining values in the queue
for i := range e.items {
e.items[i].remaining += diff
}
}
// append the item to the queue and remove the oldest item if needed
if len(e.items) >= e.maxItems {
e.items = e.items[1:]
}
e.items = append(e.items, item)
// calculate the average processing time per item in the queue
diffMs := e.items[len(e.items)-1].timestamp.Sub(e.items[0].timestamp).Milliseconds()
nrItemsProcessed := e.items[0].remaining - e.items[len(e.items)-1].remaining
avg := diffMs / nrItemsProcessed
// use that average processing time to estimate how long the remaining items will take
// and cache that ETA so we don't have to recalculate it on every update unless the
// remaining value changes
e.lastETA = msToETA(avg * remaining)
return e.lastETA
}
func msToETA(ms int64) string {
seconds := ms / 1000
minutes := seconds / 60
hours := minutes / 60
return fmt.Sprintf("%02dh:%02dm:%02ds", hours, minutes%60, seconds%60)
}
+1 -7
View File
@@ -22,7 +22,6 @@ import (
"github.com/filecoin-project/go-state-types/network"
"github.com/filecoin-project/lotus/build"
"github.com/filecoin-project/lotus/chain/actors/builtin"
"github.com/filecoin-project/lotus/chain/types"
"github.com/filecoin-project/lotus/lib/tablewriter"
)
@@ -460,12 +459,7 @@ var walletSign = &cli.Command{
sig, err := api.WalletSign(ctx, addr, msg)
if err != nil {
// Check if the address is a multisig address
act, actErr := api.StateGetActor(ctx, addr, types.EmptyTSK)
if actErr == nil && builtin.IsMultisigActor(act.Code) {
return xerrors.Errorf("specified signer address is a multisig actor, it doesnt have keys to sign transactions. To send a message with a multisig, signers of the multisig need to propose and approve transactions.")
}
return xerrors.Errorf("failed to sign message: %w", err)
return err
}
sigBytes := append([]byte{byte(sig.Type)}, sig.Data...)
-312
View File
@@ -1,312 +0,0 @@
package main
import (
"errors"
"fmt"
"io"
"os"
"os/exec"
"os/signal"
"strconv"
"strings"
"sync"
"time"
"github.com/urfave/cli/v2"
)
var cliCmd = &cli.Command{
Name: "cli",
Usage: "Runs a concurrent stress test on one or more binaries commands and prints the performance metrics including latency distribution and histogram",
Description: `This benchmark has the following features:
* Can query each command both sequentially and concurrently
* Supports rate limiting
* Can query multiple different commands at once (supporting different concurrency level and rate limiting for each command)
* Gives a nice reporting summary of the stress testing of each command (including latency distribution, histogram and more)
* Easy to use
To use this benchmark you must specify the commands you want to test using the --cmd options, the format of it is:
--cmd=CMD[:CONCURRENCY][:QPS] where only NAME is required.
Here are some real examples:
lotus-bench cli --cmd='lotus-shed mpool miner-select-messages' // runs the command with default concurrency and qps
lotus-bench cli --cmd='lotus-shed mpool miner-select-messages:3' // override concurrency to 3
lotus-bench cli --cmd='lotus-shed mpool miner-select-messages::100' // override to 100 qps while using default concurrency
lotus-bench cli --cmd='lotus-shed mpool miner-select-messages:3:100' // run using 3 workers but limit to 100 qps
lotus-bench cli --cmd='lotus-shed mpool miner-select-messages' --cmd='lotus sync wait' // run two commands at once
`,
Flags: []cli.Flag{
&cli.DurationFlag{
Name: "duration",
Value: 60 * time.Second,
Usage: "Duration of benchmark in seconds",
},
&cli.IntFlag{
Name: "concurrency",
Value: 10,
Usage: "How many workers should be used per command (can be overridden per command)",
},
&cli.IntFlag{
Name: "qps",
Value: 0,
Usage: "How many requests per second should be sent per command (can be overridden per command), a value of 0 means no limit",
},
&cli.StringSliceFlag{
Name: "cmd",
Usage: `Command to benchmark, you can specify multiple commands by repeating this flag. You can also specify command specific options to set the concurrency and qps for each command (see usage).`,
},
&cli.DurationFlag{
Name: "watch",
Value: 0 * time.Second,
Usage: "If >0 then generates reports every N seconds (only supports linux/unix)",
},
&cli.BoolFlag{
Name: "print-response",
Value: false,
Usage: "print the response of each request",
},
},
Action: func(cctx *cli.Context) error {
if len(cctx.StringSlice("cmd")) == 0 {
return errors.New("you must specify and least one cmd to benchmark")
}
var cmds []*CMD
for _, str := range cctx.StringSlice("cmd") {
entries := strings.SplitN(str, ":", 3)
if len(entries) == 0 {
return errors.New("invalid cmd format")
}
// check if concurrency was specified
concurrency := cctx.Int("concurrency")
if len(entries) > 1 {
if len(entries[1]) > 0 {
var err error
concurrency, err = strconv.Atoi(entries[1])
if err != nil {
return fmt.Errorf("could not parse concurrency value from command %s: %v", entries[0], err)
}
}
}
// check if qps was specified
qps := cctx.Int("qps")
if len(entries) > 2 {
if len(entries[2]) > 0 {
var err error
qps, err = strconv.Atoi(entries[2])
if err != nil {
return fmt.Errorf("could not parse qps value from command %s: %v", entries[0], err)
}
}
}
cmds = append(cmds, &CMD{
w: os.Stdout,
cmd: entries[0],
concurrency: concurrency,
qps: qps,
printResp: cctx.Bool("print-response"),
})
}
// terminate early on ctrl+c
c := make(chan os.Signal, 1)
signal.Notify(c, os.Interrupt)
go func() {
<-c
fmt.Println("Received interrupt, stopping...")
for _, cmd := range cmds {
cmd.Stop()
}
}()
// stop all threads after duration
go func() {
time.Sleep(cctx.Duration("duration"))
for _, cmd := range cmds {
cmd.Stop()
}
}()
// start all threads
var wg sync.WaitGroup
wg.Add(len(cmds))
for _, cmd := range cmds {
go func(cmd *CMD) {
defer wg.Done()
err := cmd.Run()
if err != nil {
fmt.Printf("error running cmd: %v\n", err)
}
}(cmd)
}
// if watch is set then print a report every N seconds
var progressCh chan struct{}
if cctx.Duration("watch") > 0 {
progressCh = make(chan struct{}, 1)
go func(progressCh chan struct{}) {
ticker := time.NewTicker(cctx.Duration("watch"))
for {
clearAndPrintReport := func() {
// clear the screen move the cursor to the top left
fmt.Print("\033[2J")
fmt.Printf("\033[%d;%dH", 1, 1)
for i, cmd := range cmds {
cmd.Report()
if i < len(cmds)-1 {
fmt.Println()
}
}
}
select {
case <-ticker.C:
clearAndPrintReport()
case <-progressCh:
clearAndPrintReport()
return
}
}
}(progressCh)
}
wg.Wait()
if progressCh != nil {
// wait for the watch go routine to return
progressCh <- struct{}{}
// no need to print the report again
return nil
}
// print the report for each command
for i, cmd := range cmds {
cmd.Report()
if i < len(cmds)-1 {
fmt.Println()
}
}
return nil
},
}
// CMD handles the benchmarking of a single command.
type CMD struct {
w io.Writer
// the cmd we want to benchmark
cmd string
// the number of concurrent requests to make to this command
concurrency int
// if >0 then limit to qps is the max number of requests per second to make to this command (0 = no limit)
qps int
// whether or not to print the response of each request (useful for debugging)
printResp bool
// instruct the worker go routines to stop
stopCh chan struct{}
// when the command bencharking started
start time.Time
// results channel is used by the workers to send results to the reporter
results chan *result
// reporter handles reading the results from workers and printing the report statistics
reporter *Reporter
}
func (c *CMD) Run() error {
var wg sync.WaitGroup
wg.Add(c.concurrency)
c.results = make(chan *result, c.concurrency*1_000)
c.stopCh = make(chan struct{}, c.concurrency)
go func() {
c.reporter = NewReporter(c.results, c.w)
c.reporter.Run()
}()
c.start = time.Now()
// throttle the number of requests per second
var qpsTicker *time.Ticker
if c.qps > 0 {
qpsTicker = time.NewTicker(time.Second / time.Duration(c.qps))
}
for i := 0; i < c.concurrency; i++ {
go func() {
c.startWorker(qpsTicker)
wg.Done()
}()
}
wg.Wait()
// close the results channel so reporter will stop
close(c.results)
// wait until the reporter is done
<-c.reporter.doneCh
return nil
}
func (c *CMD) startWorker(qpsTicker *time.Ticker) {
for {
// check if we should stop
select {
case <-c.stopCh:
return
default:
}
// wait for the next tick if we are rate limiting this command
if qpsTicker != nil {
<-qpsTicker.C
}
start := time.Now()
var statusCode int = 0
arr := strings.Fields(c.cmd)
data, err := exec.Command(arr[0], arr[1:]...).Output()
if err != nil {
fmt.Println("1")
if exitError, ok := err.(*exec.ExitError); ok {
statusCode = exitError.ExitCode()
} else {
statusCode = 1
}
} else {
if c.printResp {
fmt.Printf("[%s] %s", c.cmd, string(data))
}
}
c.results <- &result{
statusCode: &statusCode,
err: err,
duration: time.Since(start),
}
}
}
func (c *CMD) Stop() {
for i := 0; i < c.concurrency; i++ {
c.stopCh <- struct{}{}
}
}
func (c *CMD) Report() {
total := time.Since(c.start)
fmt.Fprintf(c.w, "[%s]:\n", c.cmd)
fmt.Fprintf(c.w, "- Options:\n")
fmt.Fprintf(c.w, " - concurrency: %d\n", c.concurrency)
fmt.Fprintf(c.w, " - qps: %d\n", c.qps)
c.reporter.Print(total, c.w)
}
+16 -1
View File
@@ -304,7 +304,7 @@ var importBenchCmd = &cli.Command{
return fmt.Errorf("no CAR file provided for import")
}
head, _, err = cs.Import(cctx.Context, carFile)
head, err = cs.Import(cctx.Context, carFile)
if err != nil {
return err
}
@@ -497,6 +497,21 @@ type Invocation struct {
const GasPerNs = 10
func countGasCosts(et *types.ExecutionTrace) int64 {
var cgas int64
for _, gc := range et.GasCharges {
cgas += gc.ComputeGas
}
for _, sub := range et.Subcalls {
c := countGasCosts(&sub) //nolint
cgas += c
}
return cgas
}
type stats struct {
timeTaken meanVar
gasRatio meanVar
-1
View File
@@ -120,7 +120,6 @@ func main() {
sealBenchCmd,
simpleCmd,
importBenchCmd,
cliCmd,
rpcCmd,
},
}
-181
View File
@@ -1,181 +0,0 @@
package main
import (
"fmt"
"io"
"sort"
"strings"
"sync"
"text/tabwriter"
"time"
)
// result is the result of a single rpc method request.
type result struct {
err error
statusCode *int
duration time.Duration
}
// Reporter reads the results from the workers through the results channel and aggregates the results.
type Reporter struct {
// write the report to this writer
w io.Writer
// the reporter read the results from this channel
results chan *result
// doneCh is used to signal that the reporter has finished reading the results (channel has closed)
doneCh chan bool
// lock protect the following fields during critical sections (if --watch was specified)
lock sync.Mutex
// the latencies of all requests
latencies []int64
// the number of requests that returned each status code
statusCodes map[int]int
// the number of errors that occurred
errors map[string]int
}
func NewReporter(results chan *result, w io.Writer) *Reporter {
return &Reporter{
w: w,
results: results,
doneCh: make(chan bool, 1),
statusCodes: make(map[int]int),
errors: make(map[string]int),
}
}
func (r *Reporter) Run() {
for res := range r.results {
r.lock.Lock()
r.latencies = append(r.latencies, res.duration.Milliseconds())
if res.statusCode != nil {
r.statusCodes[*res.statusCode]++
}
if res.err != nil {
if len(r.errors) < 1_000_000 {
r.errors[res.err.Error()]++
} else {
// we don't want to store too many errors in memory
r.errors["hidden"]++
}
} else {
r.errors["nil"]++
}
r.lock.Unlock()
}
r.doneCh <- true
}
func (r *Reporter) Print(elapsed time.Duration, w io.Writer) {
r.lock.Lock()
defer r.lock.Unlock()
nrReq := int64(len(r.latencies))
if nrReq == 0 {
fmt.Println("No requests were made")
return
}
// we need to sort the latencies slice to calculate the percentiles
sort.Slice(r.latencies, func(i, j int) bool {
return r.latencies[i] < r.latencies[j]
})
var totalLatency int64 = 0
for _, latency := range r.latencies {
totalLatency += latency
}
fmt.Fprintf(w, "- Total Requests: %d\n", nrReq)
fmt.Fprintf(w, "- Total Duration: %dms\n", elapsed.Milliseconds())
fmt.Fprintf(w, "- Requests/sec: %f\n", float64(nrReq)/elapsed.Seconds())
fmt.Fprintf(w, "- Avg latency: %dms\n", totalLatency/nrReq)
fmt.Fprintf(w, "- Median latency: %dms\n", r.latencies[nrReq/2])
fmt.Fprintf(w, "- Latency distribution:\n")
percentiles := []float64{0.1, 0.5, 0.9, 0.95, 0.99, 0.999}
for _, p := range percentiles {
idx := int64(p * float64(nrReq))
fmt.Fprintf(w, " %s%% in %dms\n", fmt.Sprintf("%.2f", p*100.0), r.latencies[idx])
}
// create a simple histogram with 10 buckets spanning the range of latency
// into equal ranges
//
nrBucket := 10
buckets := make([]Bucket, nrBucket)
latencyRange := r.latencies[len(r.latencies)-1]
bucketRange := latencyRange / int64(nrBucket)
// mark the end of each bucket
for i := 0; i < nrBucket; i++ {
buckets[i].start = int64(i) * bucketRange
buckets[i].end = buckets[i].start + bucketRange
// extend the last bucked by any remaning range caused by the integer division
if i == nrBucket-1 {
buckets[i].end = latencyRange
}
}
// count the number of requests in each bucket
currBucket := 0
for i := 0; i < len(r.latencies); {
if r.latencies[i] <= buckets[currBucket].end {
buckets[currBucket].cnt++
i++
} else {
currBucket++
}
}
// print the histogram using a tabwriter which will align the columns nicely
fmt.Fprintf(w, "- Histogram:\n")
const padding = 2
tabWriter := tabwriter.NewWriter(w, 0, 0, padding, ' ', tabwriter.AlignRight|tabwriter.Debug)
for i := 0; i < nrBucket; i++ {
ratio := float64(buckets[i].cnt) / float64(nrReq)
bars := strings.Repeat("#", int(ratio*100))
fmt.Fprintf(tabWriter, " %d-%dms\t%d\t%s (%s%%)\n", buckets[i].start, buckets[i].end, buckets[i].cnt, bars, fmt.Sprintf("%.2f", ratio*100))
}
tabWriter.Flush() //nolint:errcheck
fmt.Fprintf(w, "- Status codes:\n")
for code, cnt := range r.statusCodes {
fmt.Fprintf(w, " [%d]: %d\n", code, cnt)
}
// print the 10 most occurring errors (in case error values are not unique)
//
type kv struct {
err string
cnt int
}
var sortedErrors []kv
for err, cnt := range r.errors {
sortedErrors = append(sortedErrors, kv{err, cnt})
}
sort.Slice(sortedErrors, func(i, j int) bool {
return sortedErrors[i].cnt > sortedErrors[j].cnt
})
fmt.Fprintf(w, "- Errors (top 10):\n")
for i, se := range sortedErrors {
if i > 10 {
break
}
fmt.Fprintf(w, " [%s]: %d\n", se.err, se.cnt)
}
}
type Bucket struct {
start int64
// the end value of the bucket
end int64
// how many entries are in the bucket
cnt int
}
+172
View File
@@ -9,9 +9,11 @@ import (
"net/http"
"os"
"os/signal"
"sort"
"strconv"
"strings"
"sync"
"text/tabwriter"
"time"
"github.com/urfave/cli/v2"
@@ -241,6 +243,13 @@ type RPCMethod struct {
reporter *Reporter
}
// result is the result of a single rpc method request.
type result struct {
err error
statusCode *int
duration time.Duration
}
func (rpc *RPCMethod) Run() error {
client := &http.Client{
Timeout: 0,
@@ -402,3 +411,166 @@ func (rpc *RPCMethod) Report() {
fmt.Fprintf(rpc.w, " - qps: %d\n", rpc.qps)
rpc.reporter.Print(total, rpc.w)
}
// Reporter reads the results from the workers through the results channel and aggregates the results.
type Reporter struct {
// write the report to this writer
w io.Writer
// the reporter read the results from this channel
results chan *result
// doneCh is used to signal that the reporter has finished reading the results (channel has closed)
doneCh chan bool
// lock protect the following fields during critical sections (if --watch was specified)
lock sync.Mutex
// the latencies of all requests
latencies []int64
// the number of requests that returned each status code
statusCodes map[int]int
// the number of errors that occurred
errors map[string]int
}
func NewReporter(results chan *result, w io.Writer) *Reporter {
return &Reporter{
w: w,
results: results,
doneCh: make(chan bool, 1),
statusCodes: make(map[int]int),
errors: make(map[string]int),
}
}
func (r *Reporter) Run() {
for res := range r.results {
r.lock.Lock()
r.latencies = append(r.latencies, res.duration.Milliseconds())
if res.statusCode != nil {
r.statusCodes[*res.statusCode]++
}
if res.err != nil {
if len(r.errors) < 1_000_000 {
r.errors[res.err.Error()]++
} else {
// we don't want to store too many errors in memory
r.errors["hidden"]++
}
} else {
r.errors["nil"]++
}
r.lock.Unlock()
}
r.doneCh <- true
}
func (r *Reporter) Print(elapsed time.Duration, w io.Writer) {
r.lock.Lock()
defer r.lock.Unlock()
nrReq := int64(len(r.latencies))
if nrReq == 0 {
fmt.Println("No requests were made")
return
}
// we need to sort the latencies slice to calculate the percentiles
sort.Slice(r.latencies, func(i, j int) bool {
return r.latencies[i] < r.latencies[j]
})
var totalLatency int64 = 0
for _, latency := range r.latencies {
totalLatency += latency
}
fmt.Fprintf(w, "- Total Requests: %d\n", nrReq)
fmt.Fprintf(w, "- Total Duration: %dms\n", elapsed.Milliseconds())
fmt.Fprintf(w, "- Requests/sec: %f\n", float64(nrReq)/elapsed.Seconds())
fmt.Fprintf(w, "- Avg latency: %dms\n", totalLatency/nrReq)
fmt.Fprintf(w, "- Median latency: %dms\n", r.latencies[nrReq/2])
fmt.Fprintf(w, "- Latency distribution:\n")
percentiles := []float64{0.1, 0.5, 0.9, 0.95, 0.99, 0.999}
for _, p := range percentiles {
idx := int64(p * float64(nrReq))
fmt.Fprintf(w, " %s%% in %dms\n", fmt.Sprintf("%.2f", p*100.0), r.latencies[idx])
}
// create a simple histogram with 10 buckets spanning the range of latency
// into equal ranges
//
nrBucket := 10
buckets := make([]Bucket, nrBucket)
latencyRange := r.latencies[len(r.latencies)-1]
bucketRange := latencyRange / int64(nrBucket)
// mark the end of each bucket
for i := 0; i < nrBucket; i++ {
buckets[i].start = int64(i) * bucketRange
buckets[i].end = buckets[i].start + bucketRange
// extend the last bucked by any remaning range caused by the integer division
if i == nrBucket-1 {
buckets[i].end = latencyRange
}
}
// count the number of requests in each bucket
currBucket := 0
for i := 0; i < len(r.latencies); {
if r.latencies[i] <= buckets[currBucket].end {
buckets[currBucket].cnt++
i++
} else {
currBucket++
}
}
// print the histogram using a tabwriter which will align the columns nicely
fmt.Fprintf(w, "- Histogram:\n")
const padding = 2
tabWriter := tabwriter.NewWriter(w, 0, 0, padding, ' ', tabwriter.AlignRight|tabwriter.Debug)
for i := 0; i < nrBucket; i++ {
ratio := float64(buckets[i].cnt) / float64(nrReq)
bars := strings.Repeat("#", int(ratio*100))
fmt.Fprintf(tabWriter, " %d-%dms\t%d\t%s (%s%%)\n", buckets[i].start, buckets[i].end, buckets[i].cnt, bars, fmt.Sprintf("%.2f", ratio*100))
}
tabWriter.Flush() //nolint:errcheck
fmt.Fprintf(w, "- Status codes:\n")
for code, cnt := range r.statusCodes {
fmt.Fprintf(w, " [%d]: %d\n", code, cnt)
}
// print the 10 most occurring errors (in case error values are not unique)
//
type kv struct {
err string
cnt int
}
var sortedErrors []kv
for err, cnt := range r.errors {
sortedErrors = append(sortedErrors, kv{err, cnt})
}
sort.Slice(sortedErrors, func(i, j int) bool {
return sortedErrors[i].cnt > sortedErrors[j].cnt
})
fmt.Fprintf(w, "- Errors (top 10):\n")
for i, se := range sortedErrors {
if i > 10 {
break
}
fmt.Fprintf(w, " [%s]: %d\n", se.err, se.cnt)
}
}
type Bucket struct {
start int64
// the end value of the bucket
end int64
// how many entries are in the bucket
cnt int
}
+3 -66
View File
@@ -266,10 +266,7 @@ var simplePreCommit1 = &cli.Command{
ProofType: spt(sectorSize, cctx.Bool("synthetic")),
}
ticket := [32]byte{}
for i := range ticket {
ticket[i] = 1
}
var ticket [32]byte // all zero
pieces, err := ParsePieceInfos(cctx, 3)
if err != nil {
@@ -308,36 +305,7 @@ var simplePreCommit2 = &cli.Command{
Name: "synthetic",
Usage: "generate synthetic PoRep proofs",
},
&cli.StringFlag{
Name: "external-pc2",
Usage: "command for computing PC2 externally",
},
},
Description: `Compute PreCommit2 inputs and seal a sector.
--external-pc2 can be used to compute the PreCommit2 inputs externally.
The flag behaves similarly to the related lotus-worker flag, using it in
lotus-bench may be useful for testing if the external PreCommit2 command is
invoked correctly.
The command will be called with a number of environment variables set:
* EXTSEAL_PC2_SECTOR_NUM: the sector number
* EXTSEAL_PC2_SECTOR_MINER: the miner id
* EXTSEAL_PC2_PROOF_TYPE: the proof type
* EXTSEAL_PC2_SECTOR_SIZE: the sector size in bytes
* EXTSEAL_PC2_CACHE: the path to the cache directory
* EXTSEAL_PC2_SEALED: the path to the sealed sector file (initialized with unsealed data by the caller)
* EXTSEAL_PC2_PC1OUT: output from rust-fil-proofs precommit1 phase (base64 encoded json)
The command is expected to:
* Create cache sc-02-data-tree-r* files
* Create cache sc-02-data-tree-c* files
* Create cache p_aux / t_aux files
* Transform the sealed file in place
Example invocation of lotus-bench as external executor:
'./lotus-bench simple precommit2 --sector-size $EXTSEAL_PC2_SECTOR_SIZE $EXTSEAL_PC2_SEALED $EXTSEAL_PC2_CACHE $EXTSEAL_PC2_PC1OUT'
`,
ArgsUsage: "[sealed] [cache] [pc1 out]",
Action: func(cctx *cli.Context) error {
ctx := cctx.Context
@@ -362,18 +330,7 @@ Example invocation of lotus-bench as external executor:
storiface.FTSealed: cctx.Args().Get(0),
storiface.FTCache: cctx.Args().Get(1),
}
var opts []ffiwrapper.FFIWrapperOpt
if cctx.IsSet("external-pc2") {
extSeal := ffiwrapper.ExternalSealer{
PreCommit2: ffiwrapper.MakeExternPrecommit2(cctx.String("external-pc2")),
}
opts = append(opts, ffiwrapper.WithExternalSealCalls(extSeal))
}
sealer, err := ffiwrapper.New(pp, opts...)
sealer, err := ffiwrapper.New(pp)
if err != nil {
return err
}
@@ -463,12 +420,7 @@ var simpleCommit1 = &cli.Command{
start := time.Now()
ticket := [32]byte{}
seed := [32]byte{}
for i := range ticket {
ticket[i] = 1
seed[i] = 1
}
var ticket, seed [32]byte // all zero
commd, err := cid.Parse(cctx.Args().Get(2))
if err != nil {
@@ -698,10 +650,6 @@ var simpleWinningPost = &cli.Command{
Usage: "pass miner address (only necessary if using existing sectorbuilder)",
Value: "t01000",
},
&cli.BoolFlag{
Name: "show-inputs",
Usage: "output inputs for winning post generation",
},
},
ArgsUsage: "[sealed] [cache] [comm R] [sector num]",
Action: func(cctx *cli.Context) error {
@@ -772,17 +720,6 @@ var simpleWinningPost = &cli.Command{
fmt.Printf("Vanilla %s (%s)\n", challenge.Sub(start), bps(sectorSize, 1, challenge.Sub(start)))
fmt.Printf("Proof %s (%s)\n", end.Sub(challenge), bps(sectorSize, 1, end.Sub(challenge)))
fmt.Println(base64.StdEncoding.EncodeToString(proof[0].ProofBytes))
if cctx.Bool("show-inputs") {
fmt.Println("GenerateWinningPoStWithVanilla info:")
fmt.Printf(" wpt: %d\n", wpt)
fmt.Printf(" mid: %d\n", mid)
fmt.Printf(" rand: %x\n", rand)
fmt.Printf(" vp: %x\n", vp)
fmt.Printf(" proof: %x\n", proof)
}
return nil
},
}
+1 -1
View File
@@ -143,7 +143,7 @@ var runCmd = &cli.Command{
},
&cli.DurationFlag{
Name: "rate-limit-timeout",
Usage: "the maximum time to wait for the rate limiter before returning an error to clients",
Usage: "the maximum time to wait for the rate limter before returning an error to clients",
Value: gateway.DefaultRateLimitTimeout,
},
&cli.Int64Flag{
-2
View File
@@ -2566,8 +2566,6 @@ var sectorsUnsealCmd = &cli.Command{
return xerrors.Errorf("could not parse sector number: %w", err)
}
fmt.Printf("Unsealing sector %d\n", sectorNum)
return minerAPI.SectorUnseal(ctx, abi.SectorNumber(sectorNum))
},
}
+1 -1
View File
@@ -62,7 +62,7 @@ var genesisVerifyCmd = &cli.Command{
return xerrors.Errorf("opening the car file: %w", err)
}
ts, _, err := cs.Import(cctx.Context, f)
ts, err := cs.Import(cctx.Context, f)
if err != nil {
return err
}
-14
View File
@@ -2,7 +2,6 @@ package main
import (
"fmt"
"time"
"github.com/urfave/cli/v2"
@@ -44,20 +43,10 @@ var minerSelectMsgsCmd = &cli.Command{
return err
}
// Get the size of the mempool
pendingMsgs, err := api.MpoolPending(ctx, types.EmptyTSK)
if err != nil {
return err
}
mpoolSize := len(pendingMsgs)
// Measure the time taken by MpoolSelect
startTime := time.Now()
msgs, err := api.MpoolSelect(ctx, head.Key(), cctx.Float64("ticket-quality"))
if err != nil {
return err
}
duration := time.Since(startTime)
var totalGas int64
for i, f := range msgs {
@@ -75,9 +64,6 @@ var minerSelectMsgsCmd = &cli.Command{
totalGas += f.Message.GasLimit
}
// Log the duration, size of the mempool, selected messages and total gas limit of selected messages
fmt.Printf("Message selection took %s\n", duration)
fmt.Printf("Size of the mempool: %d\n", mpoolSize)
fmt.Println("selected messages: ", len(msgs))
fmt.Printf("total gas limit of selected messages: %d / %d (%0.2f%%)\n", totalGas, build.BlockGasLimit, 100*float64(totalGas)/float64(build.BlockGasLimit))
return nil
+22 -14
View File
@@ -12,7 +12,7 @@ import (
"github.com/filecoin-project/go-state-types/abi"
"github.com/filecoin-project/lotus/blockstore"
badgerbs "github.com/filecoin-project/lotus/blockstore/badger"
"github.com/filecoin-project/lotus/chain/consensus/filcns"
"github.com/filecoin-project/lotus/chain/store"
"github.com/filecoin-project/lotus/node/repo"
@@ -144,6 +144,13 @@ var stateTreePruneCmd = &cli.Command{
}
}()
// After migrating to native blockstores, this has been made
// database-specific.
badgbs, ok := bs.(*badgerbs.Blockstore)
if !ok {
return fmt.Errorf("only badger blockstores are supported")
}
mds, err := lkrepo.Datastore(context.Background(), "/metadata")
if err != nil {
return err
@@ -153,12 +160,8 @@ var stateTreePruneCmd = &cli.Command{
const DiscardRatio = 0.2
if cctx.Bool("only-ds-gc") {
fmt.Println("running datastore gc....")
gbs, ok := bs.(blockstore.BlockstoreGCOnce)
if !ok {
return xerrors.Errorf("blockstore %T does not support GC", bs)
}
for i := 0; i < cctx.Int("gc-count"); i++ {
if err := gbs.GCOnce(ctx, blockstore.WithThreshold(DiscardRatio)); err != nil {
if err := badgbs.DB().RunValueLogGC(DiscardRatio); err != nil {
return xerrors.Errorf("datastore GC failed: %w", err)
}
}
@@ -205,6 +208,13 @@ var stateTreePruneCmd = &cli.Command{
return nil
}
b := badgbs.DB().NewWriteBatch()
defer b.Cancel()
markForRemoval := func(c cid.Cid) error {
return b.Delete(badgbs.StorageKey(nil, c))
}
keys, err := bs.AllKeysChan(context.Background())
if err != nil {
return xerrors.Errorf("failed to query blockstore: %w", err)
@@ -215,12 +225,12 @@ var stateTreePruneCmd = &cli.Command{
var deleteCount int
var goodHits int
for k := range keys {
if goodSet.Has(k) {
if goodSet.HasRaw(k.Bytes()) {
goodHits++
continue
}
if err := bs.DeleteBlock(ctx, k); err != nil {
if err := markForRemoval(k); err != nil {
return fmt.Errorf("failed to remove cid %s: %w", k, err)
}
@@ -233,15 +243,13 @@ var stateTreePruneCmd = &cli.Command{
}
}
fmt.Println("running datastore gc....")
gbs, ok := bs.(blockstore.BlockstoreGCOnce)
if !ok {
fmt.Println("gc not supported...")
return nil
if err := b.Flush(); err != nil {
return xerrors.Errorf("failed to flush final batch delete: %w", err)
}
fmt.Println("running datastore gc....")
for i := 0; i < cctx.Int("gc-count"); i++ {
if err := gbs.GCOnce(ctx, blockstore.WithThreshold(DiscardRatio)); err != nil {
if err := badgbs.DB().RunValueLogGC(DiscardRatio); err != nil {
return xerrors.Errorf("datastore GC failed: %w", err)
}
}
+13 -69
View File
@@ -7,7 +7,6 @@ import (
"net"
"net/http"
"os"
"os/signal"
"path/filepath"
"reflect"
"strings"
@@ -39,7 +38,6 @@ import (
"github.com/filecoin-project/lotus/node/repo"
"github.com/filecoin-project/lotus/storage/paths"
"github.com/filecoin-project/lotus/storage/sealer"
"github.com/filecoin-project/lotus/storage/sealer/ffiwrapper"
"github.com/filecoin-project/lotus/storage/sealer/sealtasks"
"github.com/filecoin-project/lotus/storage/sealer/storiface"
)
@@ -285,36 +283,7 @@ var runCmd = &cli.Command{
Value: true,
DefaultText: "inherits --addpiece",
},
&cli.StringFlag{
Name: "external-pc2",
Usage: "command for computing PC2 externally",
},
},
Description: `Run lotus-worker.
--external-pc2 can be used to compute the PreCommit2 inputs externally.
The flag behaves similarly to the related lotus-worker flag, using it in
lotus-bench may be useful for testing if the external PreCommit2 command is
invoked correctly.
The command will be called with a number of environment variables set:
* EXTSEAL_PC2_SECTOR_NUM: the sector number
* EXTSEAL_PC2_SECTOR_MINER: the miner id
* EXTSEAL_PC2_PROOF_TYPE: the proof type
* EXTSEAL_PC2_SECTOR_SIZE: the sector size in bytes
* EXTSEAL_PC2_CACHE: the path to the cache directory
* EXTSEAL_PC2_SEALED: the path to the sealed sector file (initialized with unsealed data by the caller)
* EXTSEAL_PC2_PC1OUT: output from rust-fil-proofs precommit1 phase (base64 encoded json)
The command is expected to:
* Create cache sc-02-data-tree-r* files
* Create cache sc-02-data-tree-c* files
* Create cache p_aux / t_aux files
* Transform the sealed file in place
Example invocation of lotus-bench as external executor:
'./lotus-bench simple precommit2 --sector-size $EXTSEAL_PC2_SECTOR_SIZE $EXTSEAL_PC2_SEALED $EXTSEAL_PC2_CACHE $EXTSEAL_PC2_PC1OUT'
`,
Before: func(cctx *cli.Context) error {
if cctx.IsSet("address") {
log.Warnf("The '--address' flag is deprecated, it has been replaced by '--listen'")
@@ -379,18 +348,6 @@ Example invocation of lotus-bench as external executor:
// Connect to storage-miner
ctx := lcli.ReqContext(cctx)
// Create a new context with cancel function
ctx, cancel := context.WithCancel(ctx)
defer cancel()
// Listen for interrupt signals
go func() {
c := make(chan os.Signal, 1)
signal.Notify(c, os.Interrupt)
<-c
cancel()
}()
var nodeApi api.StorageMiner
var closer func()
for {
@@ -402,13 +359,14 @@ Example invocation of lotus-bench as external executor:
}
}
fmt.Printf("\r\x1b[0KConnecting to miner API... (%s)", err)
select {
case <-ctx.Done():
return xerrors.New("Interrupted by user")
case <-time.After(time.Second):
}
time.Sleep(time.Second)
continue
}
defer closer()
ctx, cancel := context.WithCancel(ctx)
defer cancel()
// Register all metric views
if err := view.Register(
metrics.DefaultViews...,
@@ -653,32 +611,18 @@ Example invocation of lotus-bench as external executor:
fh.ServeHTTP(w, r)
}
// Parse ffi executor flags
var ffiOpts []ffiwrapper.FFIWrapperOpt
if cctx.IsSet("external-pc2") {
extSeal := ffiwrapper.ExternalSealer{
PreCommit2: ffiwrapper.MakeExternPrecommit2(cctx.String("external-pc2")),
}
ffiOpts = append(ffiOpts, ffiwrapper.WithExternalSealCalls(extSeal))
}
// Create / expose the worker
wsts := statestore.New(namespace.Wrap(ds, modules.WorkerCallsPrefix))
workerApi := &sealworker.Worker{
LocalWorker: sealer.NewLocalWorkerWithExecutor(
sealer.FFIExec(ffiOpts...),
sealer.WorkerConfig{
TaskTypes: taskTypes,
NoSwap: cctx.Bool("no-swap"),
MaxParallelChallengeReads: cctx.Int("post-parallel-reads"),
ChallengeReadTimeout: cctx.Duration("post-read-timeout"),
Name: cctx.String("name"),
}, os.LookupEnv, remote, localStore, nodeApi, nodeApi, wsts),
LocalWorker: sealer.NewLocalWorker(sealer.WorkerConfig{
TaskTypes: taskTypes,
NoSwap: cctx.Bool("no-swap"),
MaxParallelChallengeReads: cctx.Int("post-parallel-reads"),
ChallengeReadTimeout: cctx.Duration("post-read-timeout"),
Name: cctx.String("name"),
}, remote, localStore, nodeApi, nodeApi, wsts),
LocalStore: localStore,
Storage: lr,
}
+15 -36
View File
@@ -269,37 +269,7 @@ var DaemonCmd = &cli.Command{
}
}
chainfile := cctx.String("import-chain")
snapshot := cctx.String("import-snapshot")
willImportChain := false
if chainfile != "" || snapshot != "" {
if chainfile != "" && snapshot != "" {
return fmt.Errorf("cannot specify both 'import-snapshot' and 'import-chain'")
}
willImportChain = true
}
willRemoveChain := cctx.Bool("remove-existing-chain")
if willImportChain && !willRemoveChain {
// Confirm with the user about the intention to remove chain data.
reader := bufio.NewReader(os.Stdin)
fmt.Print("Importing chain or snapshot will by default delete existing local chain data. Do you want to proceed and delete? (yes/no): ")
userInput, err := reader.ReadString('\n')
if err != nil {
return xerrors.Errorf("reading user input: %w", err)
}
userInput = strings.ToLower(strings.TrimSpace(userInput))
if userInput == "yes" {
willRemoveChain = true
} else if userInput == "no" {
willRemoveChain = false
} else {
return fmt.Errorf("invalid input. please answer with 'yes' or 'no'")
}
}
if willRemoveChain {
if cctx.Bool("remove-existing-chain") {
lr, err := repo.NewFS(cctx.String("repo"))
if err != nil {
return xerrors.Errorf("error opening fs repo: %w", err)
@@ -319,7 +289,12 @@ var DaemonCmd = &cli.Command{
}
}
if willImportChain {
chainfile := cctx.String("import-chain")
snapshot := cctx.String("import-snapshot")
if chainfile != "" || snapshot != "" {
if chainfile != "" && snapshot != "" {
return fmt.Errorf("cannot specify both 'import-snapshot' and 'import-chain'")
}
var issnapshot bool
if chainfile == "" {
chainfile = snapshot
@@ -565,7 +540,7 @@ func ImportChain(ctx context.Context, r repo.Repo, fname string, snapshot bool)
}
bar.Start()
ts, gen, err := cst.Import(ctx, ir)
ts, err := cst.Import(ctx, ir)
bar.Finish()
if err != nil {
@@ -576,14 +551,18 @@ func ImportChain(ctx context.Context, r repo.Repo, fname string, snapshot bool)
return xerrors.Errorf("flushing validation cache failed: %w", err)
}
log.Infof("setting genesis")
err = cst.SetGenesis(ctx, gen)
gb, err := cst.GetTipsetByHeight(ctx, 0, ts, true)
if err != nil {
return err
}
err = cst.SetGenesis(ctx, gb.Blocks()[0])
if err != nil {
return err
}
if !snapshot {
shd, err := drand.BeaconScheduleFromDrandSchedule(build.DrandConfigSchedule(), gen.Timestamp, nil)
shd, err := drand.BeaconScheduleFromDrandSchedule(build.DrandConfigSchedule(), gb.MinTimestamp(), nil)
if err != nil {
return xerrors.Errorf("failed to construct beacon schedule: %w", err)
}
+3 -6
View File
@@ -2289,8 +2289,7 @@ Inputs:
},
{
"MaxFee": "0",
"MsgUuid": "07070707-0707-0707-0707-070707070707",
"MaximizeFeeCap": true
"MsgUuid": "07070707-0707-0707-0707-070707070707"
},
[
{
@@ -2767,8 +2766,7 @@ Inputs:
],
{
"MaxFee": "0",
"MsgUuid": "07070707-0707-0707-0707-070707070707",
"MaximizeFeeCap": true
"MsgUuid": "07070707-0707-0707-0707-070707070707"
}
]
```
@@ -3027,8 +3025,7 @@ Inputs:
},
{
"MaxFee": "0",
"MsgUuid": "07070707-0707-0707-0707-070707070707",
"MaximizeFeeCap": true
"MsgUuid": "07070707-0707-0707-0707-070707070707"
}
]
```
+11 -7
View File
@@ -2402,7 +2402,14 @@ Perms: read
Inputs:
```json
[
"Bw=="
{
"from": "0x5cbeecf99d3fdb3f25e309cc264f240bb0664031",
"to": "0x5cbeecf99d3fdb3f25e309cc264f240bb0664031",
"gas": "0x5",
"gasPrice": "0x0",
"value": "0x0",
"data": "0x07"
}
]
```
@@ -3350,8 +3357,7 @@ Inputs:
},
{
"MaxFee": "0",
"MsgUuid": "07070707-0707-0707-0707-070707070707",
"MaximizeFeeCap": true
"MsgUuid": "07070707-0707-0707-0707-070707070707"
},
[
{
@@ -3828,8 +3834,7 @@ Inputs:
],
{
"MaxFee": "0",
"MsgUuid": "07070707-0707-0707-0707-070707070707",
"MaximizeFeeCap": true
"MsgUuid": "07070707-0707-0707-0707-070707070707"
}
]
```
@@ -4221,8 +4226,7 @@ Inputs:
},
{
"MaxFee": "0",
"MsgUuid": "07070707-0707-0707-0707-070707070707",
"MaximizeFeeCap": true
"MsgUuid": "07070707-0707-0707-0707-070707070707"
}
]
```
+2 -2
View File
@@ -104,9 +104,9 @@ domain separation tag.
### Winning PoSt proof
Draw randomness for current epoch with `WinningPoSt` domain separation tag.
Get list of sectors challenged in this epoch for this miner, based on the randomness drawn.
Get list of sectors challanged in this epoch for this miner, based on the randomness drawn.
`V`: Use filecoin proofs system to verify that miner proved access to sealed versions of these sectors.
`V`: Use filecoin proofs system to verify that miner prooved access to sealed versions of these sectors.
## `(*StateManager).TipSetState()`
+1 -1
View File
@@ -7,7 +7,7 @@ USAGE:
lotus-miner [global options] command [command options] [arguments...]
VERSION:
1.25.1-dev
1.25.1
COMMANDS:
init Initialize a lotus miner repo
+1 -29
View File
@@ -7,7 +7,7 @@ USAGE:
lotus-worker [global options] command [command options] [arguments...]
VERSION:
1.25.1-dev
1.25.1
COMMANDS:
run Start lotus worker
@@ -34,33 +34,6 @@ NAME:
USAGE:
lotus-worker run [command options] [arguments...]
DESCRIPTION:
Run lotus-worker.
--external-pc2 can be used to compute the PreCommit2 inputs externally.
The flag behaves similarly to the related lotus-worker flag, using it in
lotus-bench may be useful for testing if the external PreCommit2 command is
invoked correctly.
The command will be called with a number of environment variables set:
* EXTSEAL_PC2_SECTOR_NUM: the sector number
* EXTSEAL_PC2_SECTOR_MINER: the miner id
* EXTSEAL_PC2_PROOF_TYPE: the proof type
* EXTSEAL_PC2_SECTOR_SIZE: the sector size in bytes
* EXTSEAL_PC2_CACHE: the path to the cache directory
* EXTSEAL_PC2_SEALED: the path to the sealed sector file (initialized with unsealed data by the caller)
* EXTSEAL_PC2_PC1OUT: output from rust-fil-proofs precommit1 phase (base64 encoded json)
The command is expected to:
* Create cache sc-02-data-tree-r* files
* Create cache sc-02-data-tree-c* files
* Create cache p_aux / t_aux files
* Transform the sealed file in place
Example invocation of lotus-bench as external executor:
'./lotus-bench simple precommit2 --sector-size $EXTSEAL_PC2_SECTOR_SIZE $EXTSEAL_PC2_SEALED $EXTSEAL_PC2_CACHE $EXTSEAL_PC2_PC1OUT'
OPTIONS:
--listen value host address and port the worker api will listen on (default: "0.0.0.0:3456") [$LOTUS_WORKER_LISTEN]
--no-local-storage don't use storageminer repo for sector storage (default: false) [$LOTUS_WORKER_NO_LOCAL_STORAGE]
@@ -84,7 +57,6 @@ OPTIONS:
--timeout value used when 'listen' is unspecified. must be a valid duration recognized by golang's time.ParseDuration function (default: "30m") [$LOTUS_WORKER_TIMEOUT]
--http-server-timeout value (default: "30s")
--data-cid Run the data-cid task. true|false (default: inherits --addpiece)
--external-pc2 value command for computing PC2 externally
--help, -h show help
```
+1 -1
View File
@@ -7,7 +7,7 @@ USAGE:
lotus [global options] command [command options] [arguments...]
VERSION:
1.25.1-dev
1.25.1
COMMANDS:
daemon Start a lotus daemon process
+1 -1
View File
@@ -195,7 +195,7 @@
[Chainstore.Splitstore]
# ColdStoreType specifies the type of the coldstore.
# It can be "discard" (default) for discarding cold blocks, "messages" to store only messages or "universal" to store all chain state..
# It can be "messages" (default) to store only messages, "universal" to store all chain state or "discard" for discarding cold blocks.
#
# type: string
# env var: LOTUS_CHAINSTORE_SPLITSTORE_COLDSTORETYPE
@@ -778,10 +778,6 @@
# env var: LOTUS_FEES_MAXMARKETBALANCEADDFEE
#MaxMarketBalanceAddFee = "0.007 FIL"
# type: bool
# env var: LOTUS_FEES_MAXIMIZEWINDOWPOSTFEECAP
#MaximizeWindowPoStFeeCap = true
[Fees.MaxPreCommitBatchGasFee]
# type: types.FIL
# env var: LOTUS_FEES_MAXPRECOMMITBATCHGASFEE_BASE
+3 -6
View File
@@ -14,13 +14,11 @@
**Note for whoever is owning the release:** please capture notes as comments in this issue for anything you noticed that could be improved for future releases. There is a *Post Release* step below for incorporating changes back into the [RELEASE_ISSUE_TEMPLATE](https://github.com/filecoin-project/lotus/blob/master/documentation/misc/RELEASE_ISSUE_TEMPLATE.md), and this is easier done by collecting notes from along the way rather than just thinking about it at the end.
First steps:
- [ ] Set shell variables vX.Y.Z `export X=1 && export Y=24 && export Z=3`
- [ ] FFI: Fork a new branch (`release/lotus-vX.Y.Z`) from the filecoin-ffi `master` branch: `git checkout master && git checkout -b release/lotus-v$X.$Y.$Z`
- [ ] FFI: Tag the head of `release/lotus-vX.Y.Z` as `vX.Y.Z-pre1`: `git tag -a v$X.$Y.$Z-pre1 -m"release"`
- [ ] FFI: Fork a new branch (`release/lotus-vX.Y.Z`) from the filecoin-ffi `master` branch
- [ ] FFI: Tag the head of `release/lotus-vX.Y.Z` as `vX.Y.Z-pre1`
- [ ] Open and land a PR in lotus `master` that updates the FFI dependency to `vX.Y.Z-pre1` as cut in the previous step
- [ ] Fork a new branch (`release/vX.Y.Z`) from `master` and make any further release related changes to this branch. If any "non-trivial" changes get added to the release, uncheck all the checkboxes and return to this stage.
- [ ] Bump the version in `build/version.go` in the `master` branch to `vX.Y.(Z+1)-dev` (bump from feature release) or `vX.(Y+1).0-dev` (bump from mandatory release).
- [ ] Run `make gen && make docsgen-cli` before committing changes
- [ ] Bump the version in `build/version.go` in the `master` branch to `vX.Y.(Z+1)-dev` (bump from feature release) or `vX.(Y+1).0-dev` (bump from mandatory release). Run make gen and make docsgen-cli before committing changes
**Prepping an RC**:
@@ -28,7 +26,6 @@ First steps:
- [ ] run `make gen && make docsgen-cli`
- [ ] Generate changelog using the script at scripts/mkreleaselog
- [ ] Add contents of generated text to lotus/CHANGELOG.md in addition to other details
- [ ] Commit using PR
- [ ] tag commit with `vX.Y.Z-rcN`
- [ ] cut a pre-release [here](https://github.com/filecoin-project/lotus/releases/new?prerelease=true)
+1 -1
View File
@@ -131,7 +131,7 @@ type TargetAPI interface {
EthGasPrice(ctx context.Context) (ethtypes.EthBigInt, error)
EthFeeHistory(ctx context.Context, p jsonrpc.RawParams) (ethtypes.EthFeeHistory, error)
EthMaxPriorityFeePerGas(ctx context.Context) (ethtypes.EthBigInt, error)
EthEstimateGas(ctx context.Context, p jsonrpc.RawParams) (ethtypes.EthUint64, error)
EthEstimateGas(ctx context.Context, tx ethtypes.EthCall) (ethtypes.EthUint64, error)
EthCall(ctx context.Context, tx ethtypes.EthCall, blkParam ethtypes.EthBlockNumberOrHash) (ethtypes.EthBytes, error)
EthSendRawTransaction(ctx context.Context, rawTx ethtypes.EthBytes) (ethtypes.EthHash, error)
EthGetLogs(ctx context.Context, filter *ethtypes.EthFilterSpec) (*ethtypes.EthFilterResult, error)
+2 -8
View File
@@ -353,19 +353,13 @@ func (gw *Node) EthMaxPriorityFeePerGas(ctx context.Context) (ethtypes.EthBigInt
return gw.target.EthMaxPriorityFeePerGas(ctx)
}
func (gw *Node) EthEstimateGas(ctx context.Context, p jsonrpc.RawParams) (ethtypes.EthUint64, error) {
// validate params
_, err := jsonrpc.DecodeParams[ethtypes.EthEstimateGasParams](p)
if err != nil {
return ethtypes.EthUint64(0), xerrors.Errorf("decoding params: %w", err)
}
func (gw *Node) EthEstimateGas(ctx context.Context, tx ethtypes.EthCall) (ethtypes.EthUint64, error) {
if err := gw.limit(ctx, stateRateLimitTokens); err != nil {
return 0, err
}
// todo limit gas? to what?
return gw.target.EthEstimateGas(ctx, p)
return gw.target.EthEstimateGas(ctx, tx)
}
func (gw *Node) EthCall(ctx context.Context, tx ethtypes.EthCall, blkParam ethtypes.EthBlockNumberOrHash) (ethtypes.EthBytes, error) {
+1 -1
View File
@@ -92,7 +92,7 @@ func main() {
err = gen.WriteTupleEncodersToFile("./chain/exchange/cbor_gen.go", "exchange",
exchange.Request{},
exchange.Response{},
exchange.CompactedMessages{},
exchange.CompactedMessagesCBOR{},
exchange.BSTipSet{},
)
if err != nil {
+2 -5
View File
@@ -107,7 +107,7 @@ require (
github.com/kelseyhightower/envconfig v1.4.0
github.com/koalacxr/quantile v0.0.1
github.com/libp2p/go-buffer-pool v0.1.0
github.com/libp2p/go-libp2p v0.31.0
github.com/libp2p/go-libp2p v0.30.0
github.com/libp2p/go-libp2p-consensus v0.0.1
github.com/libp2p/go-libp2p-gorpc v0.5.0
github.com/libp2p/go-libp2p-kad-dht v0.24.0
@@ -136,7 +136,6 @@ require (
github.com/raulk/go-watchdog v1.3.0
github.com/stretchr/testify v1.8.4
github.com/syndtr/goleveldb v1.0.1-0.20210819022825-2ae1ddf74ef7
github.com/triplewz/poseidon v0.0.0-20220525065023-a7cdb0e183e7
github.com/urfave/cli/v2 v2.25.5
github.com/whyrusleeping/bencher v0.0.0-20190829221104-bb6607aa8bba
github.com/whyrusleeping/cbor-gen v0.0.0-20230923211252-36a87e1ba72f
@@ -298,7 +297,7 @@ require (
github.com/prometheus/statsd_exporter v0.22.7 // indirect
github.com/quic-go/qpack v0.4.0 // indirect
github.com/quic-go/qtls-go1-20 v0.3.3 // indirect
github.com/quic-go/quic-go v0.38.1 // indirect
github.com/quic-go/quic-go v0.37.6 // indirect
github.com/quic-go/webtransport-go v0.5.3 // indirect
github.com/rivo/uniseg v0.1.0 // indirect
github.com/rs/cors v1.7.0 // indirect
@@ -339,5 +338,3 @@ require (
replace github.com/filecoin-project/filecoin-ffi => ./extern/filecoin-ffi
replace github.com/filecoin-project/test-vectors => ./extern/test-vectors
replace github.com/triplewz/poseidon => github.com/magik6k/poseidon v0.0.0-neptune

Some files were not shown because too many files have changed in this diff Show More