Commit Graph

71 Commits

Author SHA1 Message Date
Steven Allen
b7a4dbb07f Support inline CIDs
And use the new CidBuilder from the spec actors.

This patch does not switch over to inline CIDs by default, but paves the way.
2020-07-23 23:12:32 -07:00
Jakub Sztandera
ee5639aad9
Initial implementation of Poisson Sortition
Signed-off-by: Jakub Sztandera <kubuxu@protocol.ai>
2020-06-23 16:24:01 +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
vyzo
81bc159b3e cache successful block signature validation 2020-05-14 20:48:40 +03:00
Jeromy
da72dfae3d block header changed, updating field indexes to match 2020-05-04 12:09:39 -07:00
Aayush Rajasekaran
c196570dd8 Use blake2b, instead of sha256, for ticket creation and verification 2020-05-01 22:43:36 -04:00
Jeromy
931a862abc remove usage of previous round in beacon entries 2020-04-30 11:39:37 -07:00
Łukasz Magiera
93b2b6d65d Merge branch 'testnet/3' into feat/windowed-post 2020-04-17 18:38:20 +02:00
Jeromy
708988244d WIP: working on winning post integration 2020-04-16 22:39:55 -07:00
Whyrusleeping
0db5fc422a Merge pull request #1534 from filecoin-project/patch/syscalls
VerifyConsensusFault syscall impl
2020-04-16 13:33:58 -07:00
Jeromy
674be39af1 real drand network integration 2020-04-13 20:06:10 -07:00
Henri S
aa5cec39ca first stab 2020-04-13 15:55:53 -04:00
Jakub Sztandera
ce4978d8c3 Reintroduce correct ticket randomness alongside ElectionProof
Signed-off-by: Jakub Sztandera <kubuxu@protocol.ai>
2020-04-08 21:06:41 +02:00
Jeromy
8e0ca306d2 more wiring up of drandomness 2020-04-08 08:11:42 -07:00
Jakub Sztandera
9c250f24fa Nex wip
Signed-off-by: Jakub Sztandera <kubuxu@protocol.ai>
2020-04-07 15:32:23 -07:00
Jakub Sztandera
afdb1db529 drand wip
,
Signed-off-by: Jakub Sztandera <kubuxu@protocol.ai>
2020-04-07 15:32:23 -07:00
Jeromy
7075eaba26 WIP: random beacon interface
Signed-off-by: Jakub Sztandera <kubuxu@protocol.ai>
2020-04-07 15:32:23 -07:00
Jeromy
f1dbd35407 change genesis fields to match interop 2020-03-21 15:25:00 -07:00
Łukasz Magiera
2a1253e45a types: Simplify Ticket.Equals 2020-02-28 11:31:28 -08:00
Henri S
96555aa74b update to use VRF digest appropriately and simplify drawing rand past genesis 2020-02-28 11:31:28 -08:00
whyrusleeping
df6e3e83bf make it all build finally 2020-02-27 13:45:31 -08:00
Łukasz Magiera
d787aa5007 Merge branch 'testnet/3' into feat/specs-actors 2020-02-24 18:32:02 +01:00
Łukasz Magiera
0d6cfc879d Use specs-actors ignatures 2020-02-13 00:52:36 +01:00
Jakub Sztandera
5c3ed4d91a
Remove ffi from types package
Signed-off-by: Jakub Sztandera <kubuxu@protocol.ai>
2020-02-12 18:41:35 +01:00
Łukasz Magiera
fa2c6605c2 Fix compilation after dep updates 2020-02-11 02:11:58 +01:00
Łukasz Magiera
13435aebdc Spec Actors integration 2020-02-08 03:18:32 +01:00
Łukasz Magiera
de5577ca70 chain: bisect cmd 2020-01-30 22:32:59 +01:00
Jakub Sztandera
a404d535ee
Make it buld without cgo
Signed-off-by: Jakub Sztandera <kubuxu@protocol.ai>
2020-01-22 12:03:12 -08:00
Łukasz Magiera
24e36852af Fix nil signature verification 2020-01-14 16:52:01 +01:00
Jakub Sztandera
8cfb4aafcd
Add fork signaling to blockheader
Signed-off-by: Jakub Sztandera <kubuxu@protocol.ai>
2020-01-08 20:31:08 +01:00
Jakub Sztandera
a63c34a902
Handle faults in ss and pss correctly
Signed-off-by: Jakub Sztandera <kubuxu@protocol.ai>
2020-01-07 23:03:18 +01:00
laser
841f9c3627 reorganize imports to pass linter 2020-01-07 08:23:12 -08:00
acruikshank
7873213b11 switch sectorbuilder from lib to external module
point to correct version of sectorbuilder + ensure go-datastore stays at 0.1.1

replace local address type with go-address

consume paramfetch.GetParams instead of local paramfetch.go

remove constants now defined in sectorbuilder
2020-01-07 07:48:43 -08:00
hannahhoward
8418464d91 refactor(address): use extracted address library
Switch to using extracted address library
2020-01-07 14:53:27 +01:00
Łukasz Magiera
2e95a53679 Wire up faults in fPoSt 2019-12-17 23:33:08 +01:00
Jakub Sztandera
887a34135b
Make ElectionPostChallengeCount math overflow safe
License: MIT
Signed-off-by: Jakub Sztandera <kubuxu@protocol.ai>
2019-12-09 18:34:18 +01:00
Henri S
9e363f9266
fix target for ePoSt IsTicketWinner fn
License: MIT
Signed-off-by: Jakub Sztandera <kubuxu@protocol.ai>
2019-12-09 18:34:15 +01:00
whyrusleeping
b7a89b56b3 more statticcheck! 2019-12-04 21:14:19 -08:00
Łukasz Magiera
adcda68d62 fix winner math 2019-12-03 00:18:27 +01:00
Łukasz Magiera
fd92b8f19a Type 256 differently 2019-12-02 23:49:41 +01:00
Łukasz Magiera
ab36a5aa04 wip fixing IsTicketWinner 2019-12-02 23:46:49 +01:00
Łukasz Magiera
aefd432422 actors: Fallback post progress 2019-11-28 13:46:56 +01:00
whyrusleeping
96482f456a WIP: election post restructuring 2019-11-21 16:21:45 -06:00
whyrusleeping
c4564c0597 use proper round number, allow block signature to be nil 2019-11-20 10:54:10 -06:00
whyrusleeping
f8eabd3db4 Refactor mining logic to match specs ticket chain 2019-11-20 10:52:36 -06:00
whyrusleeping
b77bb9e4aa Audit pass for blatantly wrong panics 2019-11-16 17:41:14 -06:00
Łukasz Magiera
814e01cc05 gofmt 2019-11-09 01:18:32 +01:00
Steven Li
9a5adfed3a
change with @Kubuxu's comment 2019-11-01 08:56:15 +08:00
Steven Li
85854578d1
Update chain/types/blockheader.go
Co-Authored-By: Łukasz Magiera <magik6k@users.noreply.github.com>
2019-10-31 16:44:26 +08:00
Steven Li
0aa56ba665
Refactor the PowerCmp function
Why we have this: 
This algorithm here directly follow the rule of ratio to win, and easier to understand. 
Just think about a smaller space, e.g. 100, instead of 2^256. 

When one miner has 10% power, it should have 10% ratio to win. that is, when the h is in {0..9}, it wins. Simply speaking, it is: 
     (h + 1) / 100 <= 10% ,  instead of     
     h / 99 < 10%     
The former is easier to understand, though both are equivalent when h is an integer and power_ratio < 1)
2019-10-31 13:36:33 +08:00