Commit Graph

81 Commits

Author SHA1 Message Date
Jiaying Wang
bd10bdf99a
build: release: v1.18.0 (#9652)
* build: Bump version to v1.17.3-dev

* build: set version to v1.18.0-dev

* chore: actors: Allow builtin-actors to return a map of methods (#9342)

* Allow builtin-actors to return a map of methods

* go mod

* Fix tests

* Fix tests, check carefully please

* Delete lotus-pond (#9352)

* feat: add StateNetworkVersion to mpool API

* chore: refactor: rename NewestNetworkVersion

* feat: actors: Integrate datacap actor into lotus (#9348)

* Integrate datacap actor

* Implement datacap actor in chain/builtin

* feat: support typed errors over RPC

* chore: deps: update to go-jsonrpc 0.1.8

* remove duplicate import

* fix: itest: check for closed connection

* chore: refactor: move retry test to API

* address magik supernit

* Add ability to only have single partition per msg for partitions with recovery sectors

* doc gen

* Address comments

* Return beneficiary info from miner state Info()

* Update builtin-actors to dev/20220922-v9 which includes FIP-0045 changes in progress

* Integrate verifreg changes to lotus

* Setup datacap actor

* Update builtin-actors to dev/20220922-v9-1

* Update datacap actor to query datacap instead of verifreg

* update gst

* update markets

* update actors with hamt fix

* update gst

* Update datacap to parse tokens

* Update bundles

* datacap and verifreg actors use ID addresses without protocol byte

* update builtin-actors to rc1

* update go-fil-markets

* Update bundles to rc2

* Integrate the v9 migration

* Add api for getting allocation

* Add upgrade epoch for butterfly

* Tweak PreSeal struct to be infra-friendly

* docsgen

* More tweaking of PreSeal for genesis

* review fixes

* Use fake cid for test

* add butterfly artifacts for oct 5 upgrade

* check datacaps for v8 verifreg match v9 datacap actor

* Remove print statements

* Update to go-state-types master

* Update to go-state-types v0.9.0-rc1

* review fixes

* use go-fil-markets v1.24.0-v17

* Add accessors for allocations and claims maps

* fix: missing permissions tag

* butterfly

* update butterfly artifacts

* sealing pipeline: Prepare deal assigning logic for FIP-45

* sealing pipeline: Get allocationId with StateApi

* use NoAllocationID instead of nil AllocationId

* address review

* Add datacap actor to registry.go

* Add cli for listing allocations and removing expired allocations

* Update to go-state-types master

* deps: upgrade go-merkledag to 0.8.0

* shark params

* Update cli/filplus.go

Co-authored-by: Aayush Rajasekaran <arajasek94@gmail.com>

* revert change to verifreg util

* docsgen-cli

* miss the stuff

* Update FFI

* Update go-state-types to v0.9.0

* Update builtin-actors to v9.0.0

* add calib upgrade epcoh

* update  the upgrade envvar

* kill shark

* Remove fvm splash banner from nv17 upgrade

* check invariance for pending deals and allocations

* check pending verified deal proposal migrated to allocation

* Add check for unsealed CID in precommit sectors

* Fix counting of allocations in nv17 migration test

* make gen

* pass state trees as pointers

* Add assertion that migrations with & without cache are the same

* compare allocation to verified deal proposal

* Fix miner state precommit info

* fix migration test tool

* add changelog

* Update to go-state-types v0.9.1

* Integrate builtin-actors v9.0.1

* chore: ver: bump version for rc3 (#9512)

* Bump version to 1.18.0-rc3

* Update CHANGELOG.md

* Update CHANGELOG.md

Co-authored-by: Aayush Rajasekaran <arajasek94@gmail.com>

* Update CHANGELOG.md

Co-authored-by: Aayush Rajasekaran <arajasek94@gmail.com>

Co-authored-by: Jiaying Wang <42981373+jennijuju@users.noreply.github.com>
Co-authored-by: Aayush Rajasekaran <arajasek94@gmail.com>

* Migration: Use autobatch bs

* Fix autobatch

Signed-off-by: Jakub Sztandera <kubuxu@protocol.ai>

* Invoker: Use MethodMeta from go-state-types

* Add a second premigration for nv17

* Add more shed tools for migration checking

* address review

* Lotus release v1.18.0-rc4

* fix: ci: fix app-image build on ci (#9527)

* Remove old go version first

* Add GO_VERSION file

* Use GO_VERSION to set / verify go version

* mv GO_VERSION GO_VERSION_MIN

* Use GO_VERSION_MIN in Makefile check

Co-authored-by: Ian Davis <jungziege@gmail.com>

* Update to latest go-state-types for migration fixes

* go mod tidy

* fix: use api.ErrActorNotFound instead of types.ErrActorNotFound

* fix: add fields to ForkUpgradeParams

* docs: update actors_version_checklist.md

* chore: fix lint

* update to go state type v0.9.6 with market migration fix (#9545)

* update go-state-types to v-0.9.7

* Add invariant checks to migration

* fix invariant check: number of entries in datacap actor should include verifreg

* Invariant checks: Only include not-activated deals

* test: nv17 migration

* Address review

* add lotus-shed invariance method

* Migration cli takes a stateroot cid and a height

* make gen

* Update to builtin-actors v9.0.2

* Failing test that shows that notaries can remove datacap from the verifreg actor

* Test that should pass when the problem is solved

* make gen

* Review fixes

* statemanager call function will return call information even if call errors

* update go-state-types

* update builtin-actors

* bubble up errors properly from ApplyImplicitMessage

* bump to rc5

* set new upgrade heights for calibnet

* set new upgrade height for butterfly

* tweak calibnet upgrade schedule

* clarify changelog note about calibnet

* butterfly

* update calibnet artifacts

* Allow setting local bundles for Debug FVM for av 9+

* fix: autobatch: remove potential deadlock when a block is missing

Check the _underlying_ blockstore instead of recursing. Also, drop the
lock before we do that.

* fix imports

* build: set shark mainnet epoch (#9640)

* chore: build: Lotus release v1.18.0 (#9641)

* Lotus release v1.18.0

* add changelog

* address review

* changelog improvement

Co-authored-by: Jennifer Wang <jiayingw703@gmail.com>
Co-authored-by: Jiaying Wang <42981373+jennijuju@users.noreply.github.com>

Signed-off-by: Jakub Sztandera <kubuxu@protocol.ai>
Co-authored-by: Łukasz Magiera <magik6k@gmail.com>
Co-authored-by: Łukasz Magiera <magik6k@users.noreply.github.com>
Co-authored-by: Aayush <arajasek94@gmail.com>
Co-authored-by: Geoff Stuart <geoff.vball@gmail.com>
Co-authored-by: Shrenuj Bansal <shrenuj.bansal@protocol.ai>
Co-authored-by: simlecode <69969590+simlecode@users.noreply.github.com>
Co-authored-by: Rod Vagg <rod@vagg.org>
Co-authored-by: Jakub Sztandera <kubuxu@protocol.ai>
Co-authored-by: Ian Davis <jungziege@gmail.com>
Co-authored-by: zenground0 <ZenGround0@users.noreply.github.com>
Co-authored-by: Steven Allen <steven@stebalien.com>
2022-11-15 20:57:23 -05:00
Aayush
0efca4d266 introduce v9 actors and nv17 2022-09-09 21:40:15 -04:00
Łukasz Magiera
e65fae28de chore: fix imports 2022-06-14 17:00:51 +02:00
Steven Allen
30981d0fdd
feat: refactor: actor bundling system (#8838)
1. Include the builtin-actors in the lotus source tree.
2. Embed the bundle on build instead of downloading at runtime.
3. Avoid reading the bundle whenever possible by including bundle
   metadata (the bundle CID, the actor CIDs, etc.).
4. Remove everything related to dependency injection.
    1. We're no longer downloading the bundle, so doing anything ahead
       of time doesn't really help.
    2. We register the manifests on init because, unfortunately, they're
       global.
    3. We explicitly load the current actors bundle in the genesis
       state-tree method.
    4. For testing, we just change the in-use bundle with a bit of a
       hack. It's not great, but using dependency injection doesn't make
       any sense either because, again, the manifest information is
       global.
    5. Remove the bundle.toml file. Bundles may be overridden by
       specifying an override path in the parameters file, or an
       environment variable.

fixes #8701
2022-06-13 10:15:00 -07:00
vyzo
fcf69e7b82 register mapping for methods using the real code cid 2022-04-14 19:31:14 +03:00
vyzo
752ce9e590 make tests work 2022-04-14 19:31:14 +03:00
Aayush Rajasekaran
1449644c34 Integrate v7 actors 2021-11-17 15:55:14 -05:00
Łukasz Magiera
95b128b7bc chain: Cleanup consensus logic 2021-09-02 18:09:37 +02:00
ZenGround0
6da26dcefc Remove network version panic 2021-08-10 13:07:30 -04:00
Łukasz Magiera
e7d73cbe56 vm: Remove unused ActorBalance 2021-07-27 15:34:39 +02:00
Łukasz Magiera
c7ba083fa4 Import precommit batcher 2021-05-18 16:51:06 +02:00
Łukasz Magiera
506f39b294 WIP: Integrate FIP0013 2021-05-11 22:10:29 -04:00
Aayush Rajasekaran
b5da2655dc Introduce v5 actors 2021-05-10 19:44:28 -04:00
Aayush Rajasekaran
f1ded63d60 v4 specs-actors integration, nv12 migration 2021-04-27 02:02:44 -04:00
Aayush Rajasekaran
95f68d0161 Register the v3 actor CIDs in the actor registry 2021-01-21 15:21:19 -08:00
Aayush Rajasekaran
b1afa2022b Improve error for ActorsVersionPredicate 2020-11-19 02:27:35 -05:00
Aayush Rajasekaran
ce9146916a fix: actor method params deserialization error exit code 2020-11-17 00:22:50 -05:00
Steven Allen
b0bea5f145 return an illegal actor error when we see an unsupported actor version
As far as the chain is concerned, this actor does not exist.
2020-10-07 15:46:56 -07:00
Aayush Rajasekaran
be9d23b329 Centralize some params in builtin 2020-09-29 02:25:43 -04:00
Steven Allen
ca9448bc11 rename actors v1 -> actors v2
The actual actors version is v2, not v1. Using Version1 internally was really confusing.
2020-09-28 13:13:18 -07:00
Steven Allen
5f3160cf5b continue expanding vm abstraction layer 2020-09-28 12:48:08 -07:00
Steven Allen
b8d9d7894f fix vm version 2020-09-25 12:52:16 -07:00
Steven Allen
8b35f480c4 initial vm conversion
We're probably going to want to change some of these design decisions down the
road, but this is a good starting point.

* We may want to use a more general test for "is actor valid at epoch". Maybe
just a function?
* I'd like to push some of the actor metadata down into the actor types
themselves. Ideally, we'd be able to register actors with a simple
`Register(validation, manyActors...)` call.
2020-09-25 12:49:39 -07:00
Steven Allen
8806f27220 fix runtime reflection for upgrade 2020-09-25 12:49:39 -07:00
Aayush Rajasekaran
ebad0ded3d Introduce v1 actors 2020-09-25 12:49:39 -07:00
Steven Allen
1bf3b4989d rename imports to match actors code
`sed -i 's/\bv0\(\w\)\(\w*\)/\L\1\E\20/g' **/*.go`
2020-09-18 14:59:27 -07:00
Steven Allen
24df873498 rename imports 2020-09-18 14:21:05 -07:00
Aayush Rajasekaran
053cfc1cc7 Migrate verifreg actor 2020-09-17 03:55:23 -04:00
Aayush Rajasekaran
e2295c372a Migrate multisig actor 2020-09-17 02:57:45 -04:00
Aayush Rajasekaran
b60614982e Migrate reward actor 2020-09-17 02:42:39 -04:00
Aayush Rajasekaran
b5ba7a0fad Miner migration 2020-09-17 02:21:16 -04:00
Aayush Rajasekaran
b4ee519282 Partial progress towards switching to miner and power interfaces 2020-09-16 01:37:49 -04:00
Łukasz Magiera
68097132fe Fix vm build 2020-09-14 13:45:20 +02:00
Steven Allen
d3594835c4 [WIP] Network upgrade support
This patch starts adding support for network upgrades.

* It adds an actors abstraction layer for loading abstract (cross-version) actors.
* It starts switching over to a shared deadline type.
* It adds an abstraction for ADTs (hamt/amt).
* It removes the callback-based API in the StateManager (difficult to abstract
across actor versions).
* It _does not_ actually add support for actors v2. We can do that in a followup
patch but that should be relatively easy.

This patch is heavily WIP and does not compile. Feel free to push changes
directly to this branch.

Notes:

* State tree access now needs a network version, because the HAMT type will change.
* I haven't figured out a nice way to abstract over changes to the _message_
types. However, many of them will be type aliased to actors v0 in actors v2 so
we can likely continue using the v0 versions (or use the v2 versions
everywhere). I've been renaming imports to `v0*` to make it clear that we're
importing types from a _specific_ actors version.

TODO:

* Consider merging incremental improvements? We'd have to get this compiling
again first but we could merge in the new abstractions, and slowly switch over.
* Finish migrating to the new abstractions.
* Remove all actor state types from the public API. See `miner.State.Info()` for
the planned approach here.
* Fix the tests. This is likely going to be a massive pain.
2020-09-11 20:16:29 -07:00
Aayush Rajasekaran
39755a294a Update to specs v0.9.6 2020-09-07 15:48:41 -04:00
Alan Shaw
9d9d2a2a98
fix: second argument error message
Second argument should be a pointer, not runtime.
2020-09-07 12:29:40 +01:00
Steven Allen
30fdff17b5 Skip nil exports
The actors may skip previously used but no longer defined method numbers.
2020-09-04 12:26:55 -07:00
Łukasz Magiera
cad59b045b More pond-related fixes 2020-08-07 16:07:48 +02:00
Jakub Sztandera
5605aae269
Fix even more lint warnings
Signed-off-by: Jakub Sztandera <kubuxu@protocol.ai>
2020-06-02 16:29:39 +02:00
Łukasz Magiera
0522af155d chainwatch: Record drand entries 2020-04-25 00:25:33 +02:00
Łukasz Magiera
a2554952d1 Some electionPoSt -> winningPoSt renaming 2020-04-11 00:22:45 +02:00
Jeromy
4c6fa1b2c0 handle some runtime error edge cases 2020-04-03 14:38:11 -07:00
Jeromy
212c346c21 avoid flushing disk when doing state tree snapshots 2020-04-02 17:18:25 -07:00
Jeromy
8a265a0e8d use exitcode type from specs actors instead of byte 2020-04-02 11:24:38 -07:00
Jeromy
ec0d0a51db chain validation nil return values should be []byte{} 2020-03-24 13:09:04 -07:00
Jeromy
6fe245a278 use proper exitcodes for caller validation 2020-03-23 18:50:28 -07:00
Aayush Rajasekaran
72c6552b0b Allow invocation of account actor methods 2020-03-23 18:50:28 -07:00
whyrusleeping
114cb2d2aa squash the spec shim and the vmcontext into the new runtime type 2020-03-09 21:44:29 -07:00
whyrusleeping
5076b2b950 fix reward actor and block reward application 2020-03-04 13:21:24 -08:00
whyrusleeping
1a07cb7bbf a few more fixes 2020-02-28 17:12:47 -08:00