Skip to main content

Browse the archive

https://github.com/tendermint/tendermint
05 April 2024, 22:58:32 UTC
sort by:
Revision Author Date Message Commit Date
e32fa44 Zarko Milosevic 11 July 2019, 12:27:03 UTC Add init and verifyUpdate funcs 11 July 2019, 12:27:03 UTC
ef8e18a Zarko Milosevic 10 July 2019, 15:07:12 UTC Initial version of time related concerns 10 July 2019, 15:07:12 UTC
94e0176 Jay Namsayin 04 July 2019, 07:26:13 UTC behaviour: return correct reason in MessageOutOfOrder (#3772) 04 July 2019, 07:26:13 UTC
f76684a Anton Kaliaev 03 July 2019, 13:17:59 UTC node: allow registration of custom reactors while creating node (#3771) * change invocation of NewNode across * custom reactor name are prefixed with CUSTOM_ * upgate changelog pending * improve comments * node: refactor NewNode to use functional options 03 July 2019, 13:17:59 UTC
e645442 Anton Kaliaev 02 July 2019, 18:01:29 UTC abci: minor cleanups in the socket client (#3758) Follow up from #3512 Specifically: cli.conn.Close() need not be under the mutex (#3512 (comment)) call the reqRes callback after the resCb so they always happen in the same order (#3512) Fixes #3513 02 July 2019, 18:01:29 UTC
d041476 Anton Kaliaev 02 July 2019, 17:19:58 UTC p2p: dial addrs which came from seed instead of calling ensurePeers (#3762) Calling ensurePeers outside of ensurePeersRoutine can lead to nodes disconnecting from us due to "sent next PEX request too soon" error. Solution is to just dial addrs we got from src instead of calling ensurePeers. Refs #2093 Fixes #3338 02 July 2019, 17:19:58 UTC
62f97a6 Thane Thomson 02 July 2019, 14:14:53 UTC abci: Refactor CheckTx to notify of recheck (#3744) As per #2127, this refactors the RequestCheckTx ProtoBuf struct to allow for a flag indicating whether a query is a recheck or not (and allows for possible future, more nuanced states). In order to pass this extended information through to the ABCI app, the proxy.AppConnMempool (and, for consistency, the proxy.AppConnConsensus) interface seems to need to be refactored along with abcicli.Client. And, as per this comment, I've made the following modification to the protobuf definition for the RequestCheckTx structure: enum CheckTxType { New = 0; Recheck = 1; } message RequestCheckTx { bytes tx = 1; CheckTxType type = 2; } * Refactor ABCI CheckTx to notify of recheck As per #2127, this refactors the `RequestCheckTx` ProtoBuf struct to allow for: 1. a flag indicating whether a query is a recheck or not (and allows for possible future, more nuanced states) 2. an `additional_data` bytes array to provide information for those more nuanced states. In order to pass this extended information through to the ABCI app, the `proxy.AppConnMempool` (and, for consistency, the `proxy.AppConnConsensus`) interface seems to need to be refactored. Commits: * Fix linting issue * Add CHANGELOG_PENDING entry * Remove extraneous explicit initialization * Update ABCI spec doc to include new CheckTx params * Rename method param for consistency * Rename CheckTxType enum values and remove additional_data param 02 July 2019, 14:14:53 UTC
d9481e3 Ivan Kushmantsev 01 July 2019, 08:48:54 UTC config: make possible to set absolute paths for TLS cert and key (#3765) 01 July 2019, 08:48:54 UTC
d4cf204 Marko 30 June 2019, 07:56:24 UTC libs: remove commented and unneeded code (#3757) - libs/errors: commented out errors.go - libs/common: unused colors.go - libs/db: unused debugDB Signed-off-by: Marko Baricevic <marbar3778@yahoo.com> 30 June 2019, 07:56:24 UTC
d069eb6 Ivan Kushmantsev 30 June 2019, 07:46:31 UTC docs: remove confusing statement from contributing.md (#3764) 30 June 2019, 07:46:31 UTC
b2afc65 Peng Zhong 29 June 2019, 07:06:44 UTC docs: add readme image (#3763) 29 June 2019, 07:06:44 UTC
1014f8c Marko 28 June 2019, 15:32:15 UTC docs: update to contributing.md (#3760) * Update to contributing.md - Add in section for Draft PR, instead of WIP - Add make lint as part of the commit process. Signed-off-by: Marko Baricevic <marbar3778@yahoo.com> * mistake * minor word change 28 June 2019, 15:32:15 UTC
b73cfe8 Ethan Buchman 26 June 2019, 15:27:24 UTC Merge pull request #3756 from tendermint/v0.32 V0.32 26 June 2019, 15:27:24 UTC
747f99f Ethan Buchman 25 June 2019, 11:57:50 UTC changelog and version (#3750) * changelog and version * Add section on ABCI changes * Update ABCI upgrade section to include events examples * update upgrading * more upgrading and changelog * update changelog from pending * refer to #rpc_changes * minor word changes 25 June 2019, 11:57:50 UTC
eb08609 Mengjay Gao 25 June 2019, 06:56:03 UTC docs: update JS section of abci-cli.md (#3747) 25 June 2019, 06:56:03 UTC
e93e8e7 Marko 24 June 2019, 18:13:25 UTC Added a disclaimer so the user is aware of binary (#3751) * Added a disaclaimer so the user is aware of binary - added disclaimer for `-s -w` Signed-off-by: Marko Baricevic <marbar3778@yahoo.com> * move up to compile 24 June 2019, 18:13:25 UTC
4c4bf0f Marko 24 June 2019, 14:32:24 UTC Added flags '-s -w' to buildflags (#3742) * Added flags '-s -w' to buildflags Signed-off-by: Marko Baricevic <marbar3778@yahoo.com> * added a condition for no strip * minor changes * on call discussions * on call discussion v2 24 June 2019, 14:32:24 UTC
d88a639 Thane Thomson 24 June 2019, 14:32:12 UTC Make RPC bind to localhost by default (#3746) * Make RPC bind to localhost by default * Add CHANGELOG_PENDING entry * Allow testnet command to override RPC listen address * Update localnet test to bind RPC to 0.0.0.0 * Update p2p test to bind RPC to 0.0.0.0 * Remove rpc-laddr parameter * Update localnet to use config template with RPC listen address override * Use config template override method for RPC listen address * Build config template into localnode image * Build localnode image locally before starting localnet * Move testnet config overrides into templates * Revert deletion of config overrides * Remove extraneous config parameter overrides 24 June 2019, 14:32:12 UTC
c37faf3 Anton Kaliaev 24 June 2019, 13:49:49 UTC docs: (rpc/broadcast_tx_*) write expectations for a client (#3749) Refs #3322 24 June 2019, 13:49:49 UTC
a44c621 Marko 24 June 2019, 13:48:21 UTC Remove double lint (#3748) - Circel CI and Golangci were doing linting jobs, removed circleci Signed-off-by: Marko Baricevic <marbar3778@yahoo.com> 24 June 2019, 13:48:21 UTC
3e7752c Anton Kaliaev 22 June 2019, 15:48:01 UTC cs: exit if SwitchToConsensus fails (#3706) Refs #3656 22 June 2019, 15:48:01 UTC
1b77bf6 Anton Kaliaev 22 June 2019, 15:44:12 UTC rpc/lib: write a test for TLS server (#3703) * rpc/lib: write a test for TLS server Refs #3700 * do not regenerate certificates * add nolint 22 June 2019, 15:44:12 UTC
8fc8368 Anton Kaliaev 22 June 2019, 06:30:23 UTC node: run whole func in goroutine, not just logger.Error fn (#3743) * node: run whole func in goroutine, not just logger.Error fn Fixes #3741 * add a changelog entry 22 June 2019, 06:30:23 UTC
228bba7 Thane Thomson 21 June 2019, 21:29:29 UTC state: add more tests for block validation (#3674) * Expose priv validators for use in testing * Generalize block header validation test past height 1 * Remove ineffectual assignment * Remove redundant SaveState call * Reorder comment for clarity * Use the block executor ApplyBlock function instead of implementing a stripped-down version of it * Remove commented-out code * Remove unnecessary test The required tests already appear to be implemented (implicitly) through the TestValidateBlockHeader test. * Allow for catching of specific error types during TestValidateBlockCommit * Make return error testable * Clean up and add TestValidateBlockCommit code * Fix formatting * Extract function to create a new mock test app * Update comment for clarity * Fix comment * Add skeleton code for evidence-related test * Allow for addressing priv val by address * Generalize test beyond a single validator * Generalize TestValidateBlockEvidence past first height * Reorder code to clearly separate tests and utility code * Use a common constant for stop height for testing in state/validation_test.go * Refactor errors to resemble existing conventions * Fix formatting * Extract common helper functions Having the tests littered with helper functions makes them less easily readable imho, so I've pulled them out into a separate file. This also makes it easier to see what helper functions are available during testing, so we minimize the chance of duplication when writing new tests. * Remove unused parameter * Remove unused parameters * Add field keys * Remove unused height constant * Fix typo * Fix incorrect return error * Add field keys * Use separate package for tests This refactors all of the state package's tests into a state_test package, so as to keep any usage of the state package's internal methods explicit. Any internal methods/constants used by tests are now explicitly exported in state/export_test.go * Refactor: extract helper function to make, validate, execute and commit a block * Rename state function to makeState * Remove redundant constant for number of validators * Refactor mock evidence registration into TestMain * Remove extraneous nVals variable * Replace function-level TODOs with file-level TODO and explanation * Remove extraneous comment * Fix linting issues brought up by GolangCI (pulled in from latest merge from develop) 21 June 2019, 21:29:29 UTC
59497c3 Ismail Khoffi 21 June 2019, 20:29:16 UTC Prepare nuking develop (#3726) * Update Readme and contrib. guidelines: remove traces of master Signed-off-by: Ismail Khoffi <Ismail.Khoffi@gmail.com> * add simple example Signed-off-by: Ismail Khoffi <Ismail.Khoffi@gmail.com> * update contributing.md * add link * Update CONTRIBUTING.md * add a note on master and releases Signed-off-by: Ismail Khoffi <Ismail.Khoffi@gmail.com> * update readme 21 June 2019, 20:29:16 UTC
2e5b2a9 needkane 21 June 2019, 11:18:49 UTC abci/examples: switch from hex to base64 pubkey in kvstore (#3641) * abci/example: use base64 for update validator set * update kvstore/README.md * update CHANGELOG_PENDING.md and abci/example/kvstore/README.md 21 June 2019, 11:18:49 UTC
866b343 Marko 21 June 2019, 05:58:32 UTC Changes to files that had linting issue (#3731) - Govet issues fixed - 1 gosec issue solved using nolint Signed-off-by: Marko Baricevic <marbar3778@yahoo.com> 21 June 2019, 05:58:32 UTC
9d5ba57 Thane Thomson 21 June 2019, 05:56:27 UTC abci: Refactor ABCI CheckTx and DeliverTx signatures (#3735) * Refactor signature of Application.CheckTx * Refactor signature of Application.DeliverTx * Refactor example variable names for clarity and consistency * Rename method variables for consistency * Rename method variables for consistency * add a changelog entry * update docs 21 June 2019, 05:56:27 UTC
1b5110e Anton Kaliaev 21 June 2019, 05:30:32 UTC node: fix a bug where `nil` is recorded as node's address (#3740) * node: fix a bug where `nil` is recorded as node's address Solution AddOurAddress when we know it sw.NetAddress is nil in createAddrBookAndSetOnSwitch it's set by n.transport.Listen function, which is called during start Fixes #3716 * use addr instead of n.sw.NetAddress * add both ExternalAddress and ListenAddress as our addresses 21 June 2019, 05:30:32 UTC
60827f7 Hans Schoenburg 19 June 2019, 18:35:53 UTC docs: fix some language issues and deprecated link (#3733) 19 June 2019, 18:35:53 UTC
ed18ffd Runchao Han 17 June 2019, 11:30:12 UTC p2p: refactor Switch#OnStop (#3729) 17 June 2019, 11:30:12 UTC
0e1c492 Andy Nogueira 17 June 2019, 10:51:12 UTC docs: missing 'b' in python command (#3728) In Python 3 the command outlined in this doc `import codecs; codecs.decode("YWJjZA==", 'base64').decode('ascii')` throws an error: TypeError: decoding with 'base64' codec failed (TypeError: expected bytes-like object, not str), needs to add 'b' before the encoded string `import codecs; codecs.decode(b"YWJjZA==", 'base64').decode('ascii')` to make it work 17 June 2019, 10:51:12 UTC
9010ff5 Anton Kaliaev 12 June 2019, 13:25:47 UTC types: do not ignore errors returned by PublishWithEvents (#3722) Follow up to #3643 * update changelog * do not ignore errors returned by PublishWithEvents 12 June 2019, 13:25:47 UTC
ab08354 Alexander Bezobchuk 12 June 2019, 12:03:45 UTC abci: Refactor tagging events using list of lists (#3643) ## PR This PR introduces a fundamental breaking change to the structure of ABCI response and tx tags and the way they're processed. Namely, the SDK can support more complex and aggregated events for distribution and slashing. In addition, block responses can include duplicate keys in events. Implement new Event type. An event has a type and a list of KV pairs (ie. list-of-lists). Typical events may look like: "rewards": [{"amount": "5000uatom", "validator": "...", "recipient": "..."}] "sender": [{"address": "...", "balance": "100uatom"}] The events are indexed by {even.type}.{even.attribute[i].key}/.... In this case a client would subscribe or query for rewards.recipient='...' ABCI response types and related types now include Events []Event instead of Tags []cmn.KVPair. PubSub logic now publishes/matches against map[string][]string instead of map[string]string to support duplicate keys in response events (from #1385). A match is successful if the value is found in the slice of strings. closes: #1859 closes: #2905 ## Commits: * Implement Event ABCI type and updates responses to use events * Update messages_test.go * Update kvstore.go * Update event_bus.go * Update subscription.go * Update pubsub.go * Update kvstore.go * Update query logic to handle slice of strings in events * Update Empty#Matches and unit tests * Update pubsub logic * Update EventBus#Publish * Update kv tx indexer * Update godocs * Update ResultEvent to use slice of strings; update RPC * Update more tests * Update abci.md * Check for key in validateAndStringifyEvents * Fix KV indexer to skip empty keys * Fix linting errors * Update CHANGELOG_PENDING.md * Update docs/spec/abci/abci.md Co-Authored-By: Federico Kunze <31522760+fedekunze@users.noreply.github.com> * Update abci/types/types.proto Co-Authored-By: Ethan Buchman <ethan@coinculture.info> * Update docs/spec/abci/abci.md Co-Authored-By: Ethan Buchman <ethan@coinculture.info> * Update libs/pubsub/query/query.go Co-Authored-By: Ethan Buchman <ethan@coinculture.info> * Update match function to match if ANY value matches * Implement TestSubscribeDuplicateKeys * Update TestMatches to include multi-key test cases * Update events.go * Update Query interface godoc * Update match godoc * Add godoc for matchValue * DRY-up tx indexing * Return error from PublishWithEvents in EventBus#Publish * Update PublishEventNewBlockHeader to return an error * Fix build * Update events doc in ABCI * Update ABCI events godoc * Implement TestEventBusPublishEventTxDuplicateKeys * Update TestSubscribeDuplicateKeys to be table-driven * Remove mod file * Remove markdown from events godoc * Implement TestTxSearchDeprecatedIndexing test 12 June 2019, 12:03:45 UTC
8b7ca8f Jack Zampolin 09 June 2019, 12:27:48 UTC switch to go mod (#3613) * Update to using go mod from dep * Remove references to make get_vendor_deps * Specify go version * Set GO111MODULE=on and add -mod=readonly * Fix exported env * switch to using go1.12 everywhere * Fix test scripts * Typo: * Prepend GO111MODULE=on * remove dep cache * Revert "remove dep cache" This reverts commit 45117bda Signed-off-by: Ismail Khoffi <Ismail.Khoffi@gmail.com> * bring back the dependency cache and change it to cache modules instead of vendored deps; also: - bump version for dependency cache - bump version on pkg-cache (includes modules directory) Signed-off-by: Ismail Khoffi <Ismail.Khoffi@gmail.com> * remove some more traces of dep: - remove Gopkg.(toml | lock) - update contributing guidlines - set global default in circleci (GO111MODULE=on) Signed-off-by: Ismail Khoffi <Ismail.Khoffi@gmail.com> * global var failed for `test_cover` with `go: unknown environment setting GO111MODULE=true` although the var was `GO111MODULE: on` Signed-off-by: Ismail Khoffi <Ismail.Khoffi@gmail.com> * Changelog pending entry Signed-off-by: Ismail Khoffi <Ismail.Khoffi@gmail.com> * Add bbolt dependency to go.mod Signed-off-by: Ismail Khoffi <Ismail.Khoffi@gmail.com> * move -mod=readonly to build flags 09 June 2019, 12:27:48 UTC
ed896f5 Ismail Khoffi 07 June 2019, 08:43:32 UTC Merge pull request #3720 from Yawning/boltdb-fixes Fix fundemental BoltDB backend correctness issues. 07 June 2019, 08:43:32 UTC
319ecb3 Yawning Angel 07 June 2019, 07:26:47 UTC libs/db: Fix the BoltDB Get and Iterator BoltDB's accessors will return slices that are only valid for the lifetime of the transaction. This adds copies where required to prevent hard to debug crashes (among other things). 07 June 2019, 08:26:58 UTC
f46ed4a Yawning Angel 07 June 2019, 07:22:02 UTC libs/db: Fix the BoltDB Batch.Delete `Delete` should queue a delete operation that targets the entire database, and not just keys that happen to be in the current batch. 07 June 2019, 08:26:20 UTC
6a7d418 Sean Braithwaite 06 June 2019, 20:33:39 UTC p2p: Peerbehaviour follow up (#3653) (#3663) * Move peer behaviour into it's own package * refactor wip * Adjust API and fix tests * remove unused test struct * Better error message * Restructure: + Now behaviour is it's own package, we don't need to include PeerBehaviour in every type. + Split up behaviours and reporters into seperate files * doc string fixes * Fix minor typos * Update behaviour/reporter.go Co-Authored-By: Anton Kaliaev <anton.kalyaev@gmail.com> * Update behaviour/reporter.go Co-Authored-By: Anton Kaliaev <anton.kalyaev@gmail.com> 06 June 2019, 20:33:39 UTC
c1f2648 Anton Kaliaev 05 June 2019, 15:39:28 UTC p2p: remove NewNetAddressStringWithOptionalID (#3711) Fixes #3521 The function NewNetAddressStringWithOptionalID is from a time when peer IDs were optional. They're not anymore. So this should be renamed to NewNetAddressString and should ensure the ID is provided. * update changelog * use NewNetAddress in transport tests * use NewNetAddress in TestTransportMultiplexAcceptMultiple 05 June 2019, 15:39:28 UTC
a7e8fbf Haochuan Guo 05 June 2019, 09:23:53 UTC rpc: Use Wrap instead of Errorf error (#3686) So we can have an error code when rpc fails. * Update CHANGELOG_PENDING.md Co-Authored-By: Anton Kaliaev <anton.kalyaev@gmail.com> * wrap correct error * format code 05 June 2019, 09:23:53 UTC
c9ef824 Sean Braithwaite 05 June 2019, 09:22:00 UTC p2p: Per channel metrics (#3666) (#3677) * Add `chID` label to sent/receive byte mtrics * add changelog pending entry 05 June 2019, 09:22:00 UTC
2a23eca Anton Kaliaev 05 June 2019, 01:36:16 UTC docs: update /block_results RPC docs (#3708) Fixes #3616 05 June 2019, 01:36:16 UTC
c69ec87 Ethan Buchman 04 June 2019, 03:15:23 UTC Merge pull request #3715 from tendermint/v0.31 V0.31 04 June 2019, 03:15:23 UTC
8fb2c2a Ethan Buchman 04 June 2019, 03:14:37 UTC Merge pull request #3712 from tendermint/release/v0.31.7 Release/v0.31.7 04 June 2019, 03:14:37 UTC
96e132b Anton Kaliaev 04 June 2019, 03:12:43 UTC libs/db: remove deprecated `LevelDBBackend` const (#3632) 04 June 2019, 03:12:43 UTC
99c9a35 Anton Kaliaev 04 June 2019, 01:05:22 UTC docs: update RPC docs for /subscribe & /unsubscribe (#3705) Fixes #3678 Co-Authored-By: Ethan Buchman <ethan@coinculture.info> 04 June 2019, 01:05:22 UTC
65a3dfe Ethan Buchman 03 June 2019, 15:34:21 UTC changelog and version (#3709) 03 June 2019, 15:34:21 UTC
048ac8d Anton Kaliaev 03 June 2019, 12:46:02 UTC remove invalid transactions from the mempool (#3701) Otherwise, we'll be trying to include them in each consecutive block. The downside is that evil proposers will be able to drop valid transactions (see #3322). Reverts https://github.com/tendermint/tendermint/pull/3625 Fixes #3699 03 June 2019, 12:46:02 UTC
21bfd7f Thane Thomson 03 June 2019, 12:41:27 UTC Add check for version/version.go TMCoreSemVer (#3704) 03 June 2019, 12:41:27 UTC
77e711f Ethan Buchman 30 May 2019, 23:15:48 UTC Merge pull request #3695 from tendermint/release/v0.31.6 Release/v0.31.6 30 May 2019, 23:15:48 UTC
0dd6b92 Thane Thomson 30 May 2019, 22:40:17 UTC Relocate GenesisDocProvider and DefaultGenesisDocProviderFunc (#3693) * Move GenesisDocProvider and DefaultGenesisDocProviderFunc GenesisDocProvider, being a provider of *types.GenesisDoc, makes sense to be part of the types package. DefaultGenesisDocProviderFunc, which relies on *config.Config to produce a types.GenesisDocProvider, makes sense being part of the config package. * Add aliases to avoid breaking node package API * Revert to original structure After discussion, it appears as though the best place for the relocated structures is still in the node package. This means that for the v0.31.6 release and into the future, there will be no changes two these two entities' APIs. 30 May 2019, 22:40:17 UTC
9dcee69 Anton Kaliaev 30 May 2019, 22:25:21 UTC v0.31.6 changelog (#3688) * update changelog * Update changelog - less detail about internal stuff like the `mempool` - focus BUG FIX entries more on what was fixed rather than how * minor cleanup. - remove entry for GenesisDocProvider, it's being undone * minor fix 30 May 2019, 22:25:21 UTC
b522ad0 Girish Ramnani 29 May 2019, 08:32:34 UTC docs: fix minor typo (#3681) 29 May 2019, 08:32:34 UTC
b9508ff Sean Braithwaite 27 May 2019, 19:44:56 UTC [p2p] Peer behaviour test tweaks (#3662) * Peer behaviour test tweaks: Address the remaining test issues mentioned in ##3552 notably: + switch to p2p_test package + Use `Error` instead of `Errorf` when not using formatting + Add expected/got errors to `TestMockPeerBehaviourReporterConcurrency` test * Peer behaviour equal behaviours test + slices of PeerBehaviours should be compared as histograms to ensure they have the same set of PeerBehaviours at the same freequncy. * TestEqualPeerBehaviours: + Add tests for the equivalence between sets of PeerBehaviours 27 May 2019, 19:44:56 UTC
a6ac611 Thane Thomson 27 May 2019, 19:33:41 UTC tendermint testnet: Allow for better hostname control (#3661) * Allow testnet hostnames to be overridden This allows one to specify the `--hostname` flag multiple times, each time providing an additional custom hostname for a respective peer (validator or non-validator). This overrides any of the `--hostname-prefix` or `--starting-ip-address` flags. The string array approach is taken instead of the string slice approach (see the pflag docs: https://godoc.org/github.com/spf13/pflag#StringArray) because the string slice approach (a comma-separated string) doesn't allow for cleaner multi-line BASH scripts - where this feature is intended to be used. * Reorder conditional for clarity with simpler earlier return * Allow for specifying peer hostname suffix * Quote values in help strings for greater clarity * Fix command switch * Add CHANGELOG_PENDING entry for PR * Allow for unique monikers The current approach to generating monikers for testnet nodes assigns the local hostname of the machine on which the testnet config was generated to all nodes. This results in the same moniker for each and every node. This commit makes use of the supplied `--hostname-prefix` and `--hostname-suffix`, or `--hostname` parameters to generate unique monikers for each node. Alternatively, another parameter (`--random-monikers`) allows one to forcibly override all of the other options with random hexadecimal strings. * Update CHANGELOG_PENDING entry for new command line switch 27 May 2019, 19:33:41 UTC
e7bf258 Ethan Buchman 27 May 2019, 18:45:27 UTC update PULL_REQUEST_TEMPLATE and CONTRIBUTING (#3655) * update PULL_REQUEST_TEMPLATE and CONTRIBUTING * Update .github/PULL_REQUEST_TEMPLATE.md Co-Authored-By: Thane Thomson <connect@thanethomson.com> * note ADRs 27 May 2019, 18:45:27 UTC
bcf10d5 Anton Kaliaev 27 May 2019, 18:39:58 UTC p2p: peer state init too late and pex message too soon (#3634) * fix peer state init to late Peer does not have a state yet. We set it in AddPeer. We need an new interface before mconnection is started * pex message to soon fix reconnection pex send too fast, error is caused lastReceivedRequests is still not deleted when a peer reconnected * add test case for initpeer * add prove case * remove potentially infinite loop * Update consensus/reactor.go Co-Authored-By: guagualvcha <baifudong@lancai.cn> * Update consensus/reactor_test.go Co-Authored-By: guagualvcha <baifudong@lancai.cn> * document Reactor interface better * refactor TestReactorReceiveDoesNotPanicIfAddPeerHasntBeenCalledYet * fix merge conflicts * blockchain: remove peer's ID from the pool in InitPeer Refs #3338 * pex: resetPeersRequestsInfo both upon InitPeer and RemovePeer * ensure RemovePeer is always called before InitPeer by removing the peer from the switch last (after we've stopped it and removed from all reactors) * add some comments for ConsensusReactor#InitPeer * fix pex reactor * format code * fix spelling * update changelog * remove unused methods * do not clear lastReceivedRequests upon error only in RemovePeer * call InitPeer before we start the peer! * add a comment to InitPeer * write a test * use waitUntilSwitchHasAtLeastNPeers func * bring back timeouts * Test to ensure Receive panics if InitPeer has not been called 27 May 2019, 18:39:58 UTC
5997e75 Carlos Flores 23 May 2019, 16:56:57 UTC fix integration script (#3667) 23 May 2019, 16:56:57 UTC
97ceeed Ismail Khoffi 23 May 2019, 08:09:52 UTC Merge pull request #3670 from yutianwu/fix_proxy [R4R] Fix missing context parameter in proxy 23 May 2019, 08:09:52 UTC
3ef9e45 yutianwu 23 May 2019, 06:26:11 UTC update change log 23 May 2019, 06:26:11 UTC
c7b324d Ismail Khoffi 23 May 2019, 05:45:05 UTC Merge pull request #3675 from alexanderbez/bez/3672-improve-AddSignatureFromPubKey-error Improve error message returned from AddSignatureFromPubKey 23 May 2019, 05:45:05 UTC
cfd42be Aleksandr Bezobchuk 23 May 2019, 00:02:00 UTC Improve error and tests 23 May 2019, 00:02:00 UTC
f1f243d Roman Shtylman 21 May 2019, 21:05:56 UTC p2p/pex: consult seeds in crawlPeersRoutine (#3647) * p2p/pex: consult seeds in crawlPeersRoutine This changeset alters the startup behavior for crawlPeersRoutine. Previously the routine would crawl a random selection of peers on startup. For a new seed node, there are no peers. As a result, new seed nodes are unable to bootstrap themselves with a list of peers until another node with a list of peers connects to the seed. If this node relies on the seed node for peers, then the two will not discover more peers. This changeset makes the startup behavior for crawlPeersRoutine connect to any seed nodes. Upon connecting, a request for peers will be sent to the seed node thus helping bootstrap our seed node. * p2p/pex: Adjust error message for no peers Co-Authored-By: Ethan Buchman <ethan@coinculture.info> 21 May 2019, 21:05:56 UTC
fcce9ed yutianwu 21 May 2019, 06:23:50 UTC fix lint 21 May 2019, 06:23:50 UTC
c21b4fc yutianwu 21 May 2019, 06:12:40 UTC fix bug of proxy 21 May 2019, 06:12:40 UTC
86cf8ee Sean Braithwaite 10 May 2019, 17:45:50 UTC p2p: PeerBehaviour implementation (#3539) (#3552) * p2p: initial implementation of peer behaviour * [p2p] re-use newMockPeer * p2p: add inline docs for peer_behaviour interface * [p2p] Align PeerBehaviour interface (#3558) * [p2p] make switchedPeerHebaviour private * [p2p] make storePeerHebaviour private * [p2p] Add CHANGELOG_PENDING entry for PeerBehaviour * [p2p] Adjustment naming for PeerBehaviour * [p2p] Add coarse lock around storedPeerBehaviour * [p2p]: Fix non-pointer methods in storedPeerBehaviour + Structs with embeded locks must specify all methods with pointer receivers to avoid creating a copy of the embeded lock. * [p2p] Thorough refactoring based on comments in #3552 + Decouple PeerBehaviour interface from Peer by parametrizing methods with `p2p.ID` instead of `p2p.Peer` + Setter methods wrapped in a write lock + Getter methods wrapped in a read lock + Getter methods on storedPeerBehaviour now take a `p2p.ID` + Getter methods return a copy of underlying stored behaviours + Add doc strings to public types and methods * [p2p] make structs public * [p2p] Test empty StoredPeerBehaviour * [p2p] typo fix * [p2p] add TestStoredPeerBehaviourConcurrency + Add a test which uses StoredPeerBehaviour in multiple goroutines to ensure thread-safety. * Update p2p/peer_behaviour.go Co-Authored-By: brapse <brapse@gmail.com> * Update p2p/peer_behaviour.go Co-Authored-By: brapse <brapse@gmail.com> * Update p2p/peer_behaviour.go Co-Authored-By: brapse <brapse@gmail.com> * Update p2p/peer_behaviour.go Co-Authored-By: brapse <brapse@gmail.com> * Update p2p/peer_behaviour.go Co-Authored-By: brapse <brapse@gmail.com> * Update p2p/peer_behaviour_test.go Co-Authored-By: brapse <brapse@gmail.com> * Update p2p/peer_behaviour.go Co-Authored-By: brapse <brapse@gmail.com> * Update p2p/peer_behaviour.go Co-Authored-By: brapse <brapse@gmail.com> * Update p2p/peer_behaviour.go Co-Authored-By: brapse <brapse@gmail.com> * Update p2p/peer_behaviour.go Co-Authored-By: brapse <brapse@gmail.com> * [p2p] field ordering convention * p2p: peer behaviour refactor + Change naming of reporting behaviour to `Report` + Remove the responsibility of distinguishing between the categories of good and bad behaviour and instead focus on reporting behaviour. * p2p: rename PeerReporter -> PeerBehaviourReporter 10 May 2019, 17:45:50 UTC
6f1ccb6 Zarko Milosevic 10 May 2019, 17:37:21 UTC [ADR] ADR-36: blockchain reactor refactor spec (#3506) * Add blockchain reactor refactor ADR * docs: Correct markdown and go syntax in adr-036 * Update docs/architecture/adr-036-blockchain-reactor-refactor.md Co-Authored-By: milosevic <zarko@interchain.io> * Add comments and address reviewer comments * Improve formating * Small improvements * adr-036 -> adr-040 10 May 2019, 17:37:21 UTC
a076b48 Anton Kaliaev 07 May 2019, 10:06:20 UTC libs/db: boltdb: use slice instead of sync.Map (#3633) for storing batch keys and values in boltDBBatch. NOTE: batch does not have to be safe for concurrent access. Delete may be slow, but given it should not be used often, it's ok. Fixes #3631 07 May 2019, 10:06:20 UTC
a7358bc Anton Kaliaev 07 May 2019, 08:33:47 UTC libs/db: conditional compilation (#3628) * libs/db: conditional compilation For cleveldb: go build -tags cleveldb For boltdb: go build -tags boltdb Fixes #3611 * document db_backend param better * remove deprecated LevelDBBackend * update changelog * add missing lines * add new line * fix TestRemoteDB * add a line about boltdb tag * Revert "remove deprecated LevelDBBackend" This reverts commit 1aa85453f76605e0c4d967601bbe26240e668d51. * make PR non breaking * change DEPRECATED label format https://stackoverflow.com/a/36360323/820520 07 May 2019, 08:33:47 UTC
27909e5 Anton Kaliaev 07 May 2019, 08:25:35 UTC mempool: remove only valid (Code==0) txs on Update (#3625) * mempool: remove only valid (Code==0) txs on Update so evil proposers can't drop valid txs in Commit stage. Also remove invalid (Code!=0) txs from the cache so they can be resubmitted. Fixes #3322 @rickyyangz: In the end of commit stage, we will update mempool to remove all the txs in current block. // Update mempool. err = blockExec.mempool.Update( block.Height, block.Txs, TxPreCheck(state), TxPostCheck(state), ) Assum an account has 3 transactions in the mempool, the sequences are 100, 101 and 102 separately, So an evil proposal can only package the 101 and 102 transactions into its proposal block, and leave 100 still in mempool, then the two txs will be removed from all validators' mempool when commit. So the account lost the two valid txs. @ebuchman: In the longer term we may want to do something like #2639 so we can validate txs before we commit the block. But even in this case we'd only want to run the equivalent of CheckTx, which means the DeliverTx could still fail even if the CheckTx passes depending on how the app handles the ABCI Code semantics. So more work will be required around the ABCI code. See also #2185 * add changelog entry and tests * improve changelog message * reformat code 07 May 2019, 08:25:35 UTC
1e07381 Anton Kaliaev 07 May 2019, 07:09:06 UTC rpc: /dial_peers: only mark peers as persistent if flag is on (#3620) ## Description also handle errors from DialPeersAsync remove nil addr from log msg fix TestPEXReactorDoesNotDisconnectFromPersistentPeerInSeedMode This is a follow-up from #3593 (review) Fixes most of the #3617, except #3593 (comment) ## Commits * rpc: /dial_peers: only mark peers as persistent if flag is on also - handle errors from DialPeersAsync - remove nil addr from log msg - fix TestPEXReactorDoesNotDisconnectFromPersistentPeerInSeedMode This is a follow-up from https://github.com/tendermint/tendermint/pull/3593#pullrequestreview-233556909 * remove a call to AddPersistentPeers TestDialFail will trigger a reconnect 07 May 2019, 07:09:06 UTC
2bb1a87 Ismail Khoffi 07 May 2019, 07:03:49 UTC Merge pull request #3629 from leoluk/fix-boltdb-batch libs/db: fix boltdb batching 07 May 2019, 07:03:49 UTC
d0d9ef1 Leo 06 May 2019, 20:10:39 UTC libs/db: fix boltdb batching `[]byte` is unhashable and needs to be casted to a string. See https://github.com/tendermint/tendermint/pull/3610#discussion_r280995538 Ref https://github.com/tendermint/tendermint/issues/3626 06 May 2019, 20:10:39 UTC
60b8334 Anton Kaliaev 06 May 2019, 17:37:41 UTC libs/db: close boltDBIterator (#3627) Refs https://github.com/tendermint/tendermint/pull/3610#discussion_r281201274 If we do not close, other txs will be stuck forever. 06 May 2019, 17:37:41 UTC
debf8f7 Anton Kaliaev 06 May 2019, 09:42:21 UTC libs/db: bbolt (etcd's fork of bolt) (#3610) Description: add boltdb implement for db interface. The boltdb is safe and high performence embedded KV database. It is based on B+tree and Mmap. Now it is maintained by etcd-io org. https://github.com/etcd-io/bbolt It is much better than LSM tree (levelDB) when RANDOM and SCAN read. Replaced #3604 Refs #1835 Commits: * add boltdb for storage * update boltdb in gopkg.toml * update bbolt in Gopkg.lock * dep update Gopkg.lock * add boltdb'bucket when create Boltdb * some modifies * address my own comments * fix most of the Iterator tests for boltdb * bolt does not support concurrent writes while iterating * add WARNING word * note that ReadOnly option is not supported * fixes after Ismail's review Co-Authored-By: melekes <anton.kalyaev@gmail.com> * panic if View returns error plus specify bucket in the top comment 06 May 2019, 09:42:21 UTC
4519ef3 Ethan Buchman 05 May 2019, 16:42:26 UTC adr renames (#3622) 05 May 2019, 16:42:26 UTC
86f2765 Mostafa Sedaghat Joo 05 May 2019, 16:19:41 UTC Add adr-36 (#3324) 05 May 2019, 16:19:41 UTC
e2f5471 Daniil Lashin 05 May 2019, 16:19:19 UTC ADR-037: DeliverBlock (#3420) * ADR 037 Initial draft * Add link to initial conversation * Update initial draft * Fix indentations * Change negative consequences * Fix indentations * Change ADR id * Add one more positive consequence * Rollback ADR number 05 May 2019, 16:19:19 UTC
4905640 Ethan Buchman 05 May 2019, 16:01:35 UTC types: CommitVotes struct as last step towards #1648 (#3298) * VoteSignBytes builds CanonicalVote * CommitVotes implements VoteSetReader - new CommitVotes struct holds both the Commit and the ValidatorSet and implements VoteSetReader - ToVote takes a ValidatorSet * fix TestCommit * use CommitSig.BlockID Commits may include votes for a different BlockID, could be nil, or different altogether. This means we can't use `commit.BlockID` for reconstructing the sign bytes, since up to -1/3 of the commits might be for independent BlockIDs. This means CommitSig will need to include an indicator for what BlockID it signed - if it's not the committed one or nil, it will need to include it fully in order to be verified. This is unfortunate but unavoidable so long as we include votes for non-committed BlockIDs (which we do to track validator liveness) * fixes from review * remove CommitVotes. CommitSig contains address * remove commit.canonicalVote method * toVote -> getVote, takes valIdx * update adr-025 * commit.ToVoteSet -> CommitToVoteSet * add test * fix from review 05 May 2019, 16:01:35 UTC
4c60ab8 needkane 05 May 2019, 14:23:41 UTC Update apps.md (#3621) 05 May 2019, 14:23:41 UTC
5051a1f Anton Kaliaev 04 May 2019, 06:41:31 UTC mempool: move interface into mempool package (#3524) ## Description Refs #2659 Breaking changes in the mempool package: [mempool] #2659 Mempool now an interface old Mempool renamed to CListMempool NewMempool renamed to NewCListMempool Option renamed to CListOption MempoolReactor renamed to Reactor NewMempoolReactor renamed to NewReactor unexpose TxID method TxInfo.PeerID renamed to SenderID unexpose MempoolReactor.Mempool Breaking changes in the state package: [state] #2659 Mempool interface moved to mempool package MockMempool moved to top-level mock package and renamed to Mempool Non Breaking changes in the node package: [node] #2659 Add Mempool method, which allows you to access mempool ## Commits * move Mempool interface into mempool package Refs #2659 Breaking changes in the mempool package: - Mempool now an interface - old Mempool renamed to CListMempool Breaking changes to state package: - MockMempool moved to mempool/mock package and renamed to Mempool - Mempool interface moved to mempool package * assert CListMempool impl Mempool * gofmt code * rename MempoolReactor to Reactor - combine everything into one interface - rename TxInfo.PeerID to TxInfo.SenderID - unexpose MempoolReactor.Mempool * move mempool mock into top-level mock package * add a fixme TxsFront should not be a part of the Mempool interface because it leaks implementation details. Instead, we need to come up with general interface for querying the mempool so the MempoolReactor can fetch and broadcast txs to peers. * change node#Mempool to return interface * save commit = new reactor arch * Revert "save commit = new reactor arch" This reverts commit 1bfceacd9d65a720574683a7f22771e69af9af4d. * require CListMempool in mempool.Reactor * add two changelog entries * fixes after my own review * quote interfaces, structs and functions * fixes after Ismail's review * make node's mempool an interface * make InitWAL/CloseWAL methods a part of Mempool interface * fix merge conflicts * make node's mempool an interface 04 May 2019, 06:41:31 UTC
8711af6 Anton Kaliaev 03 May 2019, 13:21:56 UTC p2p: make persistent prop independent of conn direction (#3593) ## Description Previously only outbound peers can be persistent. Now, even if the peer is inbound, if it's marked as persistent, when/if conn is lost, Tendermint will try to reconnect. This part is actually optional and can be reverted. Plus, seed won't disconnect from inbound peer if it's marked as persistent. Fixes #3362 ## Commits * make persistent prop independent of conn direction Previously only outbound peers can be persistent. Now, even if the peer is inbound, if it's marked as persistent, when/if conn is lost, Tendermint will try to reconnect. Plus, seed won't disconnect from inbound peer if it's marked as persistent. Fixes #3362 * fix TestPEXReactorDialPeer test * add a changelog entry * update changelog * add two tests * reformat code * test UnsafeDialPeers and UnsafeDialSeeds * add TestSwitchDialPeersAsync * spec: update p2p/config spec * fixes after Ismail's review * Apply suggestions from code review Co-Authored-By: melekes <anton.kalyaev@gmail.com> * fix merge conflict * remove sleep from TestPEXReactorDoesNotDisconnectFromPersistentPeerInSeedMode We don't need it actually. 03 May 2019, 13:21:56 UTC
9926ae7 Anton Kaliaev 02 May 2019, 07:53:33 UTC abci/types: update comment (#3612) Fixes #3607 02 May 2019, 07:53:33 UTC
5df6cf5 Ivan Kushmantsev 02 May 2019, 06:09:56 UTC p2p: session should terminate on nonce wrapping (#3531) (#3609) Refs #3531 02 May 2019, 06:09:56 UTC
2c26d95 JamesRay 01 May 2019, 21:15:53 UTC cs/replay: execCommitBlock should not read from state.lastValidators (#3067) * execCommitBlock should not read from state.lastValidators * fix height 1 * fix blockchain/reactor_test * fix consensus/mempool_test * fix consensus/reactor_test * fix consensus/replay_test * add CHANGELOG * fix consensus/reactor_test * fix consensus/replay_test * add a test for replay validators change * fix mem_pool test * fix byzantine test * remove a redundant code * reduce validator change blocks to 6 * fix * return peer0 config * seperate testName * seperate testName 1 * seperate testName 2 * seperate app db path * seperate app db path 1 * add a lock before startNet * move the lock to reactor_test * simulate just once * try to find problem * handshake only saveState when app version changed * update gometalinter to 3.0.0 (#3233) in the attempt to fix https://circleci.com/gh/tendermint/tendermint/43165 also code is simplified by running gofmt -s . remove unused vars enable linters we're currently passing remove deprecated linters (cherry picked from commit d47094550315c094512a242445e0dde24b5a03f5) * gofmt code * goimport code * change the bool name to testValidatorsChange * adjust receive kvstore.ProtocolVersion * adjust receive kvstore.ProtocolVersion 1 * adjust receive kvstore.ProtocolVersion 3 * fix merge execution.go * fix merge develop * fix merge develop 1 * fix run cleanupFunc * adjust code according to reviewers' opinion * modify the func name match the convention * simplify simulate a chain containing some validator change txs 1 * test CI error * Merge remote-tracking branch 'upstream/develop' into fixReplay 1 * fix pubsub_test * subscribeUnbuffered vote channel 01 May 2019, 21:15:53 UTC
a2a68df Ivan Kushmantsev 01 May 2019, 06:01:41 UTC cli: add option to not clear address book with unsafe reset (#3606) Fixes #3585 01 May 2019, 06:01:41 UTC
4334802 Anton Kaliaev 01 May 2019, 05:38:26 UTC pex: various follow-ups (#3605) * p2p: merge switch cases also improve the error msg in privval * pex: refactor code plus update specification follow-up to https://github.com/tendermint/tendermint/pull/3603 * Update docs/spec/reactors/pex/pex.md Co-Authored-By: melekes <anton.kalyaev@gmail.com> 01 May 2019, 05:38:26 UTC
40dbad9 Roman Shtylman 30 April 2019, 06:17:57 UTC pex: dial seeds when address book needs more addresses (#3603) If we are low on addresses for peering, we need to discover more peers. The previous behavior would query existing peers; however, if an existing peer does not participate in peer exchange, then our node will not discover more peers. This change consults both existing peers as well as seeds when there is a deficit in address book addresses. This allows for discovering peers though existing channels as well as via seeds if existing peers do not share addresses. 30 April 2019, 06:17:57 UTC
2585187 Ethan Buchman 26 April 2019, 23:14:54 UTC Merge pull request #3597 from tendermint/bucky/merge-master Bucky/merge master 26 April 2019, 23:14:54 UTC
d76952c Ethan Buchman 26 April 2019, 21:49:28 UTC Merge branch 'master' into bucky/merge-master 26 April 2019, 21:49:28 UTC
7b162f5 Thane Thomson 26 April 2019, 12:05:39 UTC node: refactor node.NewNode (#3456) The node.NewNode method is pretty complex at the moment, an in order to address issues like #3156, we need to simplify the interface for partial node instantiation. In some places, we don't need to build up a full node (like in the node.TestCreateProposalBlock test), but the complexity of such partial instantiation needs to be reduced. This PR aims to eventually make this easier/simpler. See also this gist https://gist.github.com/thanethomson/56e1640d057a26186e38ad678a1d114c for some background work done when starting to refactor here. ## Commits: * [WIP] Refactor node.NewNode to simplify The `node.NewNode` method is pretty complex at the moment, an in order to address issues like #3156, we need to simplify the interface for partial node instantiation. In some places, we don't need to build up a full node (like in the `node.TestCreateProposalBlock` test), but the complexity of such partial instantiation needs to be reduced. This PR aims to eventually make this easier/simpler. * Refactor state loading and genesis doc provider into state package * Refactor for clarity of return parameters * Fix incorrect capitalization of error messages * Simplify extracted functions' names * Document optionally-prefixed functions * Refactor optionallyFastSync for clarity of separation of concerns * Restructure function for early return * Restructure function for early return * Remove dependence on deprecated panic functions * refactor code a bit more plus, expose PEXReactor on node * align logger names * add a changelog entry * align logger names 2 * add a note about PEXReactor returning nil 26 April 2019, 12:05:39 UTC
70592cc Thane Thomson 26 April 2019, 10:23:43 UTC libs/common: remove deprecated PanicXXX functions (#3595) * Remove deprecated PanicXXX functions from codebase As per discussion over [here](https://github.com/tendermint/tendermint/pull/3456#discussion_r278423492), we need to remove these `PanicXXX` functions and eliminate our dependence on them. In this PR, each and every `PanicXXX` function call is replaced with a simple `panic` call. * add a changelog entry 26 April 2019, 10:23:43 UTC
90997ab Ethan Buchman 23 April 2019, 09:34:14 UTC docs: update contributing.md (#3503) Minor updates to reflect squash merging and how to prepare releases 23 April 2019, 09:34:14 UTC
b738add Zarko Milosevic 23 April 2019, 09:19:16 UTC cs: fix nondeterministic tests (#3582) Should fix #3451, #2723 and #3317. Test TestResetTimeoutPrecommitUponNewHeight is simplified so it reduces a risk of timeout failure. Furthermore, timeout we wait for TimeoutEvents is increased, and the timeout value is more precisely computed. This should hopefully decrease a chance of non-deterministic test failures. This assertion is problematic to ensure consistently due to dependency on scheduler. On the other hand, if I am not wrong, order in which messages are read from the channel respects order in which messages are written. Therefore, process will receive 2f+1 precommits that are not all for v (one is for nil) so TriggeredTimeoutPrecommit would be set to true. So we don't need to assert it. I know that it would be better to still assert to it but I don't know how to do it without sleep and that is ugly and is causing us nondeterministic failures. 23 April 2019, 09:19:16 UTC
968e955 Greg Hill 23 April 2019, 08:52:46 UTC testnet cmd: add config option (#3559) Option to explicitly provide the -config in the testnet command, closing #3160. 23 April 2019, 08:52:46 UTC
ebf815e Anton Kaliaev 23 April 2019, 08:22:40 UTC cs/replay: check appHash for each block (#3579) * check every block appHash Fixes #3460 Not really needed, but it would detect if the app changed in a way it shouldn't have. * add a changelog entry * no need to return an error if we panic * rename methods * rename methods once again * add a test * correct error msg plus fix a few go-lint warnings * better panic msg 23 April 2019, 08:22:40 UTC
8db7e74 Ismail Khoffi 22 April 2019, 08:04:04 UTC privval: increase timeout to mitigate non-deterministic test failure (#3580) This should fix #3576 (ran it many times locally but only time will tell). The test actually only checked for the opcode of the error. From the name of the test we actually want to test if we see a timeout after a pre-defined time. ## Commits: * increase readWrite timeout as it is also used in the `Accept` of the tcp listener: - before this caused the readWriteTimeout to kick in (rarely) while Accept - as a side-effect: remove obsolete time.Sleep: in both listener cases the Accept will only return after successfully accepting and the timeout that is supposed to be tested here will be triggered because there is a read without a write - see if we actually run into a timeout error (the whole purpose of this test AFAIU) Signed-off-by: Ismail Khoffi <Ismail.Khoffi@gmail.com> * makee local test-vars `const` Signed-off-by: Ismail Khoffi <Ismail.Khoffi@gmail.com> ## Additional comments: @xla: Confusing how an accept could take longer than that, but assuming a noisy environment full of little docker whales will be slower than what 50 years of silicon are capable of. The only thing I'd be vary of is that we mask structural issues with the code by just bumping the timeout, if we are sensitive towards that it warrants invesigation, but again this might only be true in the environment our CI runs in. 22 April 2019, 08:04:04 UTC
4253e67 Ethan Buchman 19 April 2019, 11:46:11 UTC Merge pull request #3571 from tendermint/v0.31 V0.31.5 19 April 2019, 11:46:11 UTC
671c5c9 Sean Braithwaite 18 April 2019, 15:31:36 UTC crypto: Proof of Concept for iterative version of SimpleHashFromByteSlices (#2611) (#3530) (#2611) had suggested that an iterative version of SimpleHashFromByteSlice would be faster, presumably because we can envision some overhead accumulating from stack frames and function calls. Additionally, a recursive algorithm risks hitting the stack limit and causing a stack overflow should the tree be too large. Provided here is an iterative alternative, a simple test to assert correctness and a benchmark. On the performance side, there appears to be no overall difference: ``` BenchmarkSimpleHashAlternatives/recursive-4 20000 77677 ns/op BenchmarkSimpleHashAlternatives/iterative-4 20000 76802 ns/op ``` On the surface it might seem that the additional overhead is due to the different allocation patterns of the implementations. The recursive version uses a single `[][]byte` slices which it then re-slices at each level of the tree. The iterative version reproduces `[][]byte` once within the function and then rewrites sub-slices of that array at each level of the tree. Eexperimenting by modifying the code to simply calculate the hash and not store the result show little to no difference in performance. These preliminary results suggest: 1. The performance of the current implementation is pretty good 2. Go has low overhead for recursive functions 3. The performance of the SimpleHashFromByteSlice routine is dominated by the actual hashing of data Although this work is in no way exhaustive, point #3 suggests that optimizations of this routine would need to take an alternative approach to make significant improvements on the current performance. Finally, considering that the recursive implementation is easier to read, it might not be worthwhile to switch to a less intuitive implementation for so little benefit. * re-add slice re-writing * [crypto] Document SimpleHashFromByteSlicesIterative 18 April 2019, 15:31:36 UTC
back to top