Page MenuHomePhabricator

sdulfari (Skúli Dulfari)
UserAdministrator

Projects

User does not belong to any projects.

User Details

User Since
Apr 14 2022, 17:59 (5 w, 4 d)
Roles
Administrator
Slack Username
sdulfari

Recent Activity

Yesterday

sdulfari accepted D11516: Deduplicate the HasReason class.
Mon, May 23, 21:55
sdulfari accepted D11517: Fix an exception message in blockencodings_tests when the compact size MAX_SIZE is not the limiting factor.
Mon, May 23, 21:51
sdulfari updated the diff for D11502: [avalanche] Add a way to check if a VoteRecord is stale.

I gave my last comment some more thought and came to this conclusion: Even if my assumptions are correct in practice,
it's better (from a user UX perspective) to deploy with thresholds a bit higher than expected. Consider these scenarios:

Mon, May 23, 21:36
sdulfari added a comment to D11502: [avalanche] Add a way to check if a VoteRecord is stale.

I should also mention it's my expectation that confidence flips are not uniformly distributed, but more likely to occur earlier in the voting process. While it's possible to reach 127 confidence and reset twice, then hitting the low-end threshold, I believe this case to be rare.

Mon, May 23, 16:40
sdulfari added a comment to D11502: [avalanche] Add a way to check if a VoteRecord is stale.

These thresholds seems REALLY low. How did you pick them?

Mon, May 23, 16:21

Sat, May 21

sdulfari closed D11501: [avalanche] Remove unnecessary code path in addNodeToQuorum.
Sat, May 21, 05:32
sdulfari committed rABC38abe3e93298: [avalanche] Remove unnecessary code path in addNodeToQuorum (authored by sdulfari).
[avalanche] Remove unnecessary code path in addNodeToQuorum
Sat, May 21, 05:32
sdulfari closed D11500: [avalanche] Add some tests to check that duplicate votes are not counted.
Sat, May 21, 05:31
sdulfari committed rABC767e5c7a8044: [avalanche] Add some tests to check that duplicate votes are not counted (authored by sdulfari).
[avalanche] Add some tests to check that duplicate votes are not counted
Sat, May 21, 05:31
sdulfari updated the diff for D11503: [avalanche] Set voterecord_tests node ID using a fixture.
  • Rebase on changes in D11502
  • Remove TestingSetup. It's not needed.
Sat, May 21, 05:26
sdulfari updated the diff for D11502: [avalanche] Add a way to check if a VoteRecord is stale.
  • Fix approach so that records will go stale for all confidence levels
  • Add way more test coverage
  • Tighten threshold for low-confidence stalls
Sat, May 21, 05:21

Fri, May 20

sdulfari added inline comments to D11500: [avalanche] Add some tests to check that duplicate votes are not counted.
Fri, May 20, 16:29
sdulfari updated the diff for D11500: [avalanche] Add some tests to check that duplicate votes are not counted.

Added finalization sanity check

Fri, May 20, 16:27

Thu, May 19

sdulfari requested review of D11503: [avalanche] Set voterecord_tests node ID using a fixture.
Thu, May 19, 23:40
sdulfari requested review of D11502: [avalanche] Add a way to check if a VoteRecord is stale.
Thu, May 19, 23:24
sdulfari abandoned D11494: [avalanche] Fix underflow when clearing too many inflight requests.

This was only a concern in relation to changes I suggested in D11474. Since my approach on that is changing, this patch isn't relevant anymore.

Thu, May 19, 21:20
sdulfari closed D11499: [avalanche] Move VoteRecord tests to their own file.
Thu, May 19, 18:34
sdulfari committed rABC4f6a0d53c689: [avalanche] Move VoteRecord tests to their own file (authored by sdulfari).
[avalanche] Move VoteRecord tests to their own file
Thu, May 19, 18:34
sdulfari requested review of D11501: [avalanche] Remove unnecessary code path in addNodeToQuorum.
Thu, May 19, 18:28
sdulfari requested review of D11500: [avalanche] Add some tests to check that duplicate votes are not counted.
Thu, May 19, 18:09
sdulfari requested review of D11499: [avalanche] Move VoteRecord tests to their own file.
Thu, May 19, 16:56
sdulfari accepted D11484: txorphanage: Extract AddChildrenToWorkSet.
Thu, May 19, 16:02
sdulfari requested review of D11494: [avalanche] Fix underflow when clearing too many inflight requests.
Thu, May 19, 15:49

Wed, May 18

sdulfari accepted D11493: txorphanage: comment improvements.
Wed, May 18, 20:06
sdulfari accepted D11492: net_processing: move AddToCompactExtraTransactions into PeerManagerImpl.
Wed, May 18, 20:05
sdulfari accepted D11491: scripted-diff: Update txorphanage naming convention.
Wed, May 18, 19:59
sdulfari requested changes to D11484: txorphanage: Extract AddChildrenToWorkSet.
Wed, May 18, 19:52
sdulfari accepted D11388: [avalanche] Answer getavaproofs message with short proof ids.
Wed, May 18, 17:59
sdulfari closed D11468: [avalanche] Ensure nullptr safety on proccesor's public interface.
Wed, May 18, 16:52
sdulfari committed rABCca065adfaba3: [avalanche] Ensure nullptr safety on proccesor's public interface (authored by sdulfari).
[avalanche] Ensure nullptr safety on proccesor's public interface
Wed, May 18, 16:52
sdulfari updated the diff for D11474: [avalanche] Drop stale votes after too many rounds without finalization.

Remove unnecesarily complex logic from isStale(). The banning circumstance isn't possible since my understanding was incorrect.

Wed, May 18, 16:46
sdulfari added inline comments to D11474: [avalanche] Drop stale votes after too many rounds without finalization.
Wed, May 18, 16:45
sdulfari updated the diff for D11468: [avalanche] Ensure nullptr safety on proccesor's public interface.

Fix comment

Wed, May 18, 16:32
sdulfari requested review of D11494: [avalanche] Fix underflow when clearing too many inflight requests.
Wed, May 18, 16:09
sdulfari updated the diff for D11468: [avalanche] Ensure nullptr safety on proccesor's public interface.

Add requested comments

Wed, May 18, 15:51
sdulfari retitled D11474: [avalanche] Drop stale votes after too many rounds without finalization from [avalanche] Mark vote records as stale after too many votes without finalization to [avalanche] Drop stale votes after too many rounds without finalization.
Wed, May 18, 04:43

Tue, May 17

sdulfari updated the diff for D11474: [avalanche] Drop stale votes after too many rounds without finalization.
  • Add more test coverage
  • Drop the stale vote in addition to reporting the VoteStatus
  • Always check shouldPoll() in getInvsForNextPoll()
  • Make the isStale() check more robust so that votes are only dropped when inflight polls have finished. Otherwise it's possible to ban peers that respond after the record has been dropped.
Tue, May 17, 23:31
sdulfari updated the summary of D11474: [avalanche] Drop stale votes after too many rounds without finalization.
Tue, May 17, 21:01
sdulfari planned changes to D11474: [avalanche] Drop stale votes after too many rounds without finalization.
Tue, May 17, 17:29
sdulfari updated the diff for D11474: [avalanche] Drop stale votes after too many rounds without finalization.

flagged -> marked

Tue, May 17, 17:11
sdulfari requested changes to D11388: [avalanche] Answer getavaproofs message with short proof ids.
Tue, May 17, 16:53
sdulfari updated the diff for D11474: [avalanche] Drop stale votes after too many rounds without finalization.
  • Added some missing test cases
  • Added requested log
  • Make threshold unsigned
Tue, May 17, 16:35
sdulfari added inline comments to D11474: [avalanche] Drop stale votes after too many rounds without finalization.
Tue, May 17, 16:26
sdulfari added inline comments to D11474: [avalanche] Drop stale votes after too many rounds without finalization.
Tue, May 17, 16:08
sdulfari added inline comments to D11474: [avalanche] Drop stale votes after too many rounds without finalization.
Tue, May 17, 16:05
sdulfari accepted D11477: [avalanche] Remove unecessary timeout in the abc_p2p_getavaaddr.py test.

These timeouts are a symptom that the default is too long. They make local testing painful when debugging tests. We should consider lowering the default timeout if not too many tests are impacted, but that will be a separate change.

Tue, May 17, 15:58

Mon, May 16

sdulfari abandoned D11457: [avalanche] Reject votes that have been inconclusive for too many rounds.

Extract some parts of this diff into D11474. The design is moving away from complicating the vote rejection logic in favor of letting the network layer request inventories as needed.

Mon, May 16, 20:37
sdulfari requested review of D11474: [avalanche] Drop stale votes after too many rounds without finalization.
Mon, May 16, 20:30
sdulfari planned changes to D11457: [avalanche] Reject votes that have been inconclusive for too many rounds.

I think I came up with a better approach.

Mon, May 16, 16:54
sdulfari updated the summary of D11457: [avalanche] Reject votes that have been inconclusive for too many rounds.
Mon, May 16, 16:51
sdulfari updated the summary of D11457: [avalanche] Reject votes that have been inconclusive for too many rounds.
Mon, May 16, 16:44
sdulfari abandoned D11464: Lint test_framework/messages.py.

I compared versions with Fabien and the version I'm on is older. Installing flake8 and autopep8 via pip3 instead of apt gives the newer versions which give clean linting for this file.

Mon, May 16, 16:20
sdulfari added inline comments to D11468: [avalanche] Ensure nullptr safety on proccesor's public interface.
Mon, May 16, 16:05
sdulfari accepted D11473: Rename the Uint256KeyWrapper struct.
Mon, May 16, 15:45

Sun, May 15

sdulfari accepted D11471: Add a checkpoint after the May 15th, 2022 upgrade.
Sun, May 15, 16:51

Sat, May 14

Fabien empowered sdulfari as an administrator.
Sat, May 14, 16:35
sdulfari requested review of D11468: [avalanche] Ensure nullptr safety on proccesor's public interface.
Sat, May 14, 13:15
sdulfari accepted D11465: [avalanche] Rename the ProofComparator.
Sat, May 14, 00:25
sdulfari updated the diff for D11457: [avalanche] Reject votes that have been inconclusive for too many rounds.

Fix the approach so that instead of stopping avalanche on the inconclusive item, we continue
with a reject vote. This way, we retain network consistency but are still able to tip the
network to finalize a rejection and not let a vote dangle.

Sat, May 14, 00:23

Fri, May 13

sdulfari requested review of D11464: Lint test_framework/messages.py.
Fri, May 13, 23:38
sdulfari requested review of D11457: [avalanche] Reject votes that have been inconclusive for too many rounds.
Fri, May 13, 00:44

Thu, May 12

sdulfari requested changes to D11388: [avalanche] Answer getavaproofs message with short proof ids.
Thu, May 12, 21:07
sdulfari closed D11452: [avalanche] Fix logging of vote statuses.
Thu, May 12, 16:57
sdulfari committed rABC8e43311b0101: [avalanche] Fix logging of vote statuses (authored by sdulfari).
[avalanche] Fix logging of vote statuses
Thu, May 12, 16:57
sdulfari closed D11451: [avalanche] Fix proof voting response type.
Thu, May 12, 06:42
sdulfari committed rABCe6e75c178b54: [avalanche] Fix proof voting response type (authored by sdulfari).
[avalanche] Fix proof voting response type
Thu, May 12, 06:42
sdulfari requested review of D11452: [avalanche] Fix logging of vote statuses.
Thu, May 12, 06:34
sdulfari requested changes to D11450: [avalanche] Maintain a radix tree of the proofs.

Generally true throughout this diff: the naming is too dependent on the underlying proofs database being a radix tree. A simple example at the peer manager interface is getProofRadixTree(). Does the caller care that the DB is a radix tree? If we change to some other COW DB, we'll end up with a lot of renaming to do.

Thu, May 12, 00:34

Wed, May 11

sdulfari requested review of D11451: [avalanche] Fix proof voting response type.
Wed, May 11, 23:26

Tue, May 10

sdulfari accepted D11405: Let the radix tree work with 256 bits keys.
Tue, May 10, 20:38
sdulfari accepted D11405: Let the radix tree work with 256 bits keys.
Tue, May 10, 15:29

Mon, May 9

sdulfari added inline comments to D11405: Let the radix tree work with 256 bits keys.
Mon, May 9, 22:17
sdulfari requested changes to D11405: Let the radix tree work with 256 bits keys.
Mon, May 9, 21:34

Fri, May 6

sdulfari accepted D11433: net_processing: move some globals into PeerManagerImpl.

Maybe mention in the summary that changes related to g_wtxid_relay_peers are not present because these are for segwit.

Fri, May 6, 23:39
sdulfari updated the summary of D11431: [avalanche] Make proof replacement check explicit.
Fri, May 6, 23:09
sdulfari updated the diff for D11431: [avalanche] Make proof replacement check explicit.

Attempt to handle unexpected DUPLICATED case gracefully

Fri, May 6, 23:08
sdulfari closed D11432: [avalanche] Consolidate duplicate code for moving proofs to conflicting pool.
Fri, May 6, 22:33
sdulfari committed rABC1e7b869da4b3: [avalanche] Consolidate duplicate code for moving proofs to conflicting pool (authored by sdulfari).
[avalanche] Consolidate duplicate code for moving proofs to conflicting pool
Fri, May 6, 22:33
sdulfari requested review of D11432: [avalanche] Consolidate duplicate code for moving proofs to conflicting pool.
Fri, May 6, 04:04
sdulfari added inline comments to D11431: [avalanche] Make proof replacement check explicit.
Fri, May 6, 00:15
sdulfari updated the summary of D11431: [avalanche] Make proof replacement check explicit.
Fri, May 6, 00:12
sdulfari requested review of D11431: [avalanche] Make proof replacement check explicit.
Fri, May 6, 00:09

Thu, May 5

sdulfari accepted D11421: Add a traversal method for the radix tree.
Thu, May 5, 18:35
sdulfari requested changes to D11421: Add a traversal method for the radix tree.

A quick mention of what you intend to do with this would be helpful since it's not currently used except for its own test.

Thu, May 5, 16:42

Wed, May 4

sdulfari closed D11407: Rename radix typename for key type to something searchable.
Wed, May 4, 20:08
sdulfari committed rABC9bd7eb3de72a: Rename radix typename for key type to something searchable (authored by sdulfari).
Rename radix typename for key type to something searchable
Wed, May 4, 20:08
sdulfari requested review of D11407: Rename radix typename for key type to something searchable.
Wed, May 4, 01:44

Fri, Apr 29

sdulfari closed D11399: Fix swapped max and default diff settings in mining post.
Fri, Apr 29, 16:49
sdulfari committed rABCWEBaba104480b60: Fix swapped max and default diff settings in mining post (authored by sdulfari).
Fix swapped max and default diff settings in mining post
Fri, Apr 29, 16:49
sdulfari closed D11398: Add sane instructions for rvm setup.
Fri, Apr 29, 16:48
sdulfari committed rABCWEB6709a8d603c3: Add sane instructions for rvm setup (authored by sdulfari).
Add sane instructions for rvm setup
Fri, Apr 29, 16:48

Thu, Apr 28

sdulfari requested review of D11399: Fix swapped max and default diff settings in mining post.
Thu, Apr 28, 23:56
sdulfari requested review of D11398: Add sane instructions for rvm setup.
Thu, Apr 28, 23:48
sdulfari added inline comments to D11340: Add an option to allow for bypassing the network group limit on outbound connections.
Thu, Apr 28, 21:06
sdulfari accepted D11397: use c++ 17 structured binding in cashaddr related code.
Thu, Apr 28, 20:42
sdulfari requested changes to D11390: [avalanche] Bypass net group limits by default for avalanche outbounds.

This single flag is trying to do two jobs, so it ends up being a tri-state switch, which I doubt is expected by the user:

  • Unset - non-avalanche connections do not bypass, but avalanche connections do
  • True - all connections bypass
  • False - no connections bypass
Thu, Apr 28, 20:24
sdulfari accepted D11396: Use incrementRefCount and decrementRefCount in the radix code too.
Thu, Apr 28, 16:36
sdulfari closed D11386: Reflow eCash section of the README.
Thu, Apr 28, 15:55
sdulfari committed rABC8fa710cebb43: Reflow eCash section of the README (authored by sdulfari).
Reflow eCash section of the README
Thu, Apr 28, 15:55

Tue, Apr 26

sdulfari requested review of D11386: Reflow eCash section of the README.

done

Tue, Apr 26, 16:07