4b6c0198ad
reset coding generator on slot boundaries ( #3726 )
2019-04-10 18:18:55 -07:00
f1e7237c09
vote_api cleanup ( #3710 )
...
* vote_api cleanup
* fixups
* fixup
* remove unused code
* revert removal of serialize and deserialize
* ...
* increase coverage, bootstrap staking
* Sagar's STAKE to my VOTE
2019-04-10 17:52:47 -07:00
1b5845ac3e
Fix getting votes from gossip ( #3723 )
2019-04-10 17:16:08 -07:00
d2ea782372
Always use bootstrap vote account for leader
2019-04-10 15:51:00 -07:00
894135a084
Less pub in PohRecorder
2019-04-10 12:50:45 -07:00
7da4142d33
Process votes from gossip only in leader node ( #3707 )
2019-04-09 22:06:32 -07:00
003fd6545c
Logging for unexpected validator errors ( #3697 )
2019-04-09 15:05:43 -07:00
2c93062f54
Improve banking_stage performance messages
...
Use transaction count instead of batch count,
and set the recv_start from when we finished processing
the previous batch to get a more accurate number.
2019-04-09 14:54:12 -07:00
dc6c34da5d
Fast-track vote signature verification and processing ( #3695 )
2019-04-09 12:57:12 -07:00
d4eebcc2aa
Check for frozen in confirm_forks ( #3678 )
2019-04-09 11:45:38 -07:00
483cc2fa4e
Support old repair strategy for reparing slots in a range for supporting replicators ( #3665 )
2019-04-08 12:46:23 -07:00
44b391096d
Configurable local cluster native processors ( #3676 )
2019-04-08 11:15:58 -07:00
444e87f888
Fix metric ( #3664 )
2019-04-06 21:57:01 -07:00
20aa4434e2
Fix repair ( #3581 )
...
Add DetachedHeads repair protocol
Add DetachedHeads repair test
Repair starting from root
2019-04-06 19:41:22 -07:00
878a842611
Move append_vec bench to the crate with append_vec ( #3650 )
...
* Move append_vec bench to the crate with append_vec
* Use black_box to tell the compiler not to optimize away test data
```
pub fn black_box<T>(dummy: T) -> T {
unsafe {
let ret = std::ptr::read_volatile(&dummy);
std::mem::forget(dummy);
ret
}
}
```
* Revert "Use black_box to tell the compiler not to optimize away test data"
This reverts commit 5610b8ee957f1d8bf6e270e392859e8b23b1e472.
* Use black_box to tell the compiler not to optimize away test data
* Create bench directories
2019-04-06 07:18:56 -06:00
68e21911eb
Remove redundant transfer_signed
2019-04-05 22:04:32 -07:00
95cc36af96
Impl SyncClient and AsyncClient for ThinClient
2019-04-05 22:04:32 -07:00
f349c1f0dc
Get everything off RpcSignatureStatus
2019-04-05 22:09:29 -06:00
90c1300bb6
Plumb TransactionError through Rpc
2019-04-05 22:09:29 -06:00
f5f4434e0a
Remove unnecessary lock in sigverify
2019-04-05 16:57:45 -07:00
adcda3c715
Remove airdrop dependency from replicators
2019-04-05 16:11:39 -07:00
a5b5248a09
move vote_accounts up ( #3647 )
2019-04-05 14:23:00 -07:00
ef5df6f3fe
Add server specification
2019-04-05 11:44:57 -07:00
2f90f9fbd4
Version all jsonrpc crates, in core too
2019-04-05 11:44:57 -07:00
46e6911ec1
Add get_signature_status() to SyncClient
...
And move bank::Result to transaction module.
2019-04-05 10:22:05 -07:00
4ea19b90a4
Fix update_ancestor_stakes in locktower ( #3631 )
...
* Fix update_ancestor_stakes in locktower
* Add test for vote threshold
2019-04-05 03:05:31 -07:00
35298e01a8
Remove Instruction wrapper structs and name functions after enum fields
2019-04-03 13:34:27 -07:00
867f6f107b
Rename SystemInstruction::Move to SystemInstruction::Transfer
2019-04-03 08:35:57 -06:00
43bb813cbe
Rename 'new_account' to 'new_user_account'
...
And 'new_program_account' to 'new_account'
2019-04-02 21:24:42 -06:00
978ff87b76
Fix potential storage bug
...
The previous code was assuming the instruction index and the
program_id index were the same. That's always true for
single-instruction transactions, but not for multiples.
2019-04-02 19:00:35 -06:00
4c0bc1fd88
Add program_ids() methods
...
Added CompiledInstruction::program_id() so that we don't need to pass
around instruction indexes just for Message::program_id().
Also added Message.program_ids() that returns a slice so that we
can move those pubkeys into Message::account_keys.
2019-04-02 19:00:35 -06:00
20189c5d45
Bump hashbrown to 0.2.0
2019-04-02 16:37:21 -06:00
d90b8c331d
Refactor blocktree storage abstraction ( #3588 )
2019-04-02 16:58:07 -05:00
cd634801a2
Re-enable test but remove replicators from config
2019-04-02 10:38:30 -07:00
5f10a87dec
Ignore Flaky Local Cluster test
2019-04-02 10:56:29 -06:00
fa1c1e3734
Rename native programs to native instruction processors
2019-04-02 10:36:19 -06:00
d228b6467c
Implement finalizer so that all locked accounts are dropped ( #3585 )
...
* Implement finalizer so that all locked accounts are dropped when finalizer goes out of scope
* Add test for tx error with lock conflict
* Fix double unlock from destructor running after a call to unlock
2019-04-02 03:55:42 -07:00
6ff2a0a75e
Rework discover to handle additional parameters, and be unit-testable
2019-04-01 23:05:25 -06:00
fcef54d062
Add a constructor to generate random pubkeys
2019-03-31 16:23:18 -06:00
8660c3581e
Add squashing metrics ( #3573 )
2019-03-29 21:21:59 -07:00
f886b3b12b
Fix resetting PohRecorder to wrong bank ( #3553 )
...
* Check whether future slot already has transmission
2019-03-29 20:00:36 -07:00
5646daa820
Delete lots of fee parameters
...
So many zeros!
2019-03-29 19:21:51 -06:00
7896e8288d
Replace Transaction::fee with a FeeCalculator
2019-03-29 19:21:51 -06:00
9369ea86ea
Track detached slots in blocktree ( #3536 )
...
* Add contains_all_parents flag to SlotMeta to prep for tracking detached heads
* Add new DetachedHeads column family
* Remove has_complete_parents
* Fix test
2019-03-29 16:07:24 -07:00
dee5ede16d
Get rid of unnecessary frozen banks ( #3572 )
2019-03-29 16:06:48 -07:00
0887832b00
Early exit if buffered packets is empty
2019-03-29 13:40:07 -07:00
8e04fadb05
Cleanup magic numbers
...
Rename `num_signatures` to `num_required_signatures` to
disambiguate it from `tx.signatures.len()`.
2019-03-29 13:03:29 -07:00
31f8b6d352
Integrate Message into Transaction
2019-03-29 13:03:29 -07:00
98d60e6124
Expose a method for getting the Message from a Transaction
...
This currently constructs the message, but when message
is integrated, it can return a `&Message`.
2019-03-29 13:03:29 -07:00
f482c9ab61
Functionalize tx serialization; make testing more explicit
2019-03-29 11:31:46 -06:00
75dcd97f5f
Update test to deserialize txs
2019-03-29 11:31:46 -06:00
4776dc36ab
Map entry txs to serialized txs in blockstream
2019-03-29 11:31:46 -06:00
753d0dcabe
Fix the cuda build
...
And add a test to check the condition that the cuda tests are
exercising.
2019-03-29 08:25:56 -06:00
b708998d9d
Fix chacha build
2019-03-29 08:25:56 -06:00
c4bc710d3a
Use Serde's `with` attribute to shorten length encodings in Transaction
2019-03-29 08:25:56 -06:00
857dc2ba47
Remove custom serialization
2019-03-29 08:25:56 -06:00
7a81f327ce
Add sigverify tests
2019-03-28 19:42:11 -06:00
2ab50cbae8
Move untested code out of SDK
...
verify_signature() was only used in a test that was testing
binary layout. It only worked because the test transaction only
had one signature.
from() was only used by verify_signature() and that's something
we'd typically called `pubkey()`.
hash() didn't return the hash of the Transaction, as you might
guess. It's only used for PoH, so move it into Entry.
2019-03-28 14:24:59 -06:00
0482f153d0
Lower a bunch of debug
...
Can't afford to be printing on every transaction error, it will slow
the system down.
2019-03-28 12:24:47 -07:00
50b0a5ae83
Blocktree+Erasure tests of basic erasure functionality ( #3535 )
...
* Remove WindowSlot; add Blocktree based tests to erasure
2019-03-28 01:55:51 -05:00
7d0ff8e713
Re-enable Replicator test ( #3534 )
2019-03-27 17:21:49 -07:00
e8cc566b2b
Storage Account setup for replicators and validators ( #3516 )
...
* Setup Storage Accounts for replicators
* Setup Storage Accounts for validators
* Add Replicator Info to Local Cluster and Add test
2019-03-27 15:54:09 -07:00
e45f7afd85
use the right id for delegate id
2019-03-27 15:04:09 -07:00
36ea088387
Fix Storage Stage not receiving entries when node is leader ( #3528 )
2019-03-27 13:10:33 -07:00
0346b9cb5c
hang out on progress until fork is confirmed
2019-03-27 08:41:41 -07:00
6bfe497ab5
remove leader confirmaiton
2019-03-27 08:41:41 -07:00
6956bf635e
validator confirmaiton
2019-03-27 08:41:41 -07:00
e27d6d0988
validator confirmation
2019-03-27 08:41:41 -07:00
cecdb7061e
Remove blockhash parameter from Bank::transfer
...
That parameter is an artifact from the Loom days, when I thought
Bank should implement the same interace as ThinClient.
2019-03-27 08:51:10 -06:00
16ff4ac1a8
Simplify storage interface in blocktree ( #3522 )
2019-03-27 01:36:39 -05:00
8c69c40834
Make space for a new Transaction::new
2019-03-26 20:06:05 -06:00
9d73fbb84a
also check the delegate_id
2019-03-26 12:03:22 -07:00
420cbc45cd
Record the current nodes locktower votes from the bank ( #3502 )
...
* observed_locktower_stats
* fixup! observed_locktower_stats
2019-03-26 11:06:31 -07:00
615472b52c
Initailize locktower with heaviest bank ( #3489 )
2019-03-25 20:00:11 -07:00
4bc3f70150
Boot VoteTransaction
2019-03-25 17:11:57 -07:00
59b4f40f4e
fixup! fixup! keep track of locktower slots and stakes
2019-03-25 16:05:28 -07:00
93c57934cb
fixup! keep track of locktower slots and stakes
2019-03-25 16:05:28 -07:00
e8e1d6b8ce
keep track of locktower slots and stakes
2019-03-25 16:05:28 -07:00
4916cd8da5
bench-tps in a cargo test
2019-03-25 15:05:56 -07:00
51004881f8
filter out banks that have an older epoch ( #3472 )
2019-03-25 11:09:39 -07:00
aefa9891c0
Delete unused code
2019-03-24 21:44:04 -07:00
33972ef89e
Boot BudgetTransaction
2019-03-24 14:52:06 -06:00
c49e84c75b
Boot StorageTransaction
2019-03-24 13:51:02 -07:00
8ff1987d2d
Reorg Storage program to look more like the others
2019-03-24 13:51:02 -07:00
acedf4ca5a
Move Instruction into its own module
2019-03-23 20:31:55 -07:00
68c35bfde6
Restart node test ( #3459 ) ( #3465 )
...
* Restart node test (#3459 )
* Add test to local_cluster for restarting a node
* fix so that we don't hit end of epoch - leader not found before trying to transfer
* Do not look for confirmations, b/c nobody is voting on empty transmissions in this single node test
2019-03-23 19:19:55 -07:00
e1a3708844
Ensure accounts are unlocked ( #3458 )
2019-03-23 13:30:56 -07:00
46ecac3310
Update leader slot in poh recorder if we skipped it ( #3452 )
...
* reset poh recorder with the original start slot
2019-03-23 13:07:09 -07:00
f09b8d3921
Demote log level
2019-03-22 17:00:09 -07:00
52f6c33ff9
Make sure banking stage is recording with the same bank that it read ( #3447 )
...
* make sure banking stage is recording with the same bank that it read with
2019-03-22 14:17:39 -07:00
06b0c98c75
Remove accounts when the fork is removed ( #3384 )
...
* Fix test
* Cleanup accounts when the fork is removed
* Update test to check for deleted accounts
2019-03-21 17:36:10 -07:00
d0761f57e8
Add _program suffix to directories of crates with _program suffix
2019-03-21 16:24:06 -06:00
412ebfcaf2
ReplayStage::new is too long
...
break into more functions
2019-03-21 14:08:11 -07:00
07d55d0092
Downgrade 'No next leader found' to warning
2019-03-21 11:18:49 -06:00
148e08a8a5
Enable cluster tests ( #3372 )
...
* Cluster tests
* stable!
* fixup! stable!
* fixup! fixup! stable!
* fixup! fixup! fixup! stable!
* fixup! fixup! fixup! fixup! stable!
* fixed space
* add getNumBlocksSinceSignatureConfirmation entry for the json rpc docs
* Check in upcoming epochs for potential leadership slots in next_leader_slot()
2019-03-21 07:43:21 -07:00
7c8a4bf6a4
use ticks per slot to check if the current tick is in the leader slot
2019-03-20 16:55:01 -07:00
71314d79a7
address review comments
2019-03-20 16:55:01 -07:00
d7ff6645a9
change pubkey to ref
2019-03-20 16:55:01 -07:00
1824e09d0a
find next leader slot before resetting working bank in Poh recorder
2019-03-20 16:55:01 -07:00
205907d3d7
Check if poh recorder has over stepped the leader slot
2019-03-20 16:55:01 -07:00
Rob Walker
Sagar Dhawan
Greg Fitzgerald
Pankaj Garg
carllin
Stephen Akridge
Jack May
Tyera Eulberg
Mark E. Sinclair
Stephen Akridge
Sathish Ambley
Anatoly Yakovenko
Anatoly Yakovenko
Mark
Sathish
Michael Vines