fix: setting data root in CreateProposalBlock

abci/types/application.go

@@ -3,7 +3,7 @@ package types
 import (
 	"context"
 
-	"github.com/cometbft/cometbft/crypto/tmhash"
+	"github.com/cometbft/cometbft/crypto/merkle"
 )
 
 // Application is an interface that enables any finite, deterministic state machine
@@ -96,7 +96,7 @@ func (BaseApplication) ApplySnapshotChunk(req RequestApplySnapshotChunk) Respons
 
 func (BaseApplication) PrepareProposal(req RequestPrepareProposal) ResponsePrepareProposal {
 	// we use placeholder values for the hash
-	req.Txs = append(req.Txs, tmhash.Sum(nil))
+	req.Txs = append(req.Txs, merkle.HashFromByteSlices(req.Txs))
 	return ResponsePrepareProposal{Txs: req.Txs}
 }
 

blockchain/msgs_test.go

@@ -97,7 +97,7 @@ func TestBlockchainMessageVectors(t *testing.T) {
 			BlockRequest: &bcproto.BlockRequest{Height: math.MaxInt64}}},
 			"0a0a08ffffffffffffffff7f"},
 		{"BlockResponseMessage", &bcproto.Message{Sum: &bcproto.Message_BlockResponse{
-			BlockResponse: &bcproto.BlockResponse{Block: bpb}}}, "1a93010a90010a5b0a02080b1803220b088092b8c398feffffff012a0212003a20c4da88e876062aa1543400d50d0eaa0dac88096057949cfb7bca7f3a48c04bf96a20e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855122f0a0b48656c6c6f20576f726c643220c4da88e876062aa1543400d50d0eaa0dac88096057949cfb7bca7f3a48c04bf91a00"},
+			BlockResponse: &bcproto.BlockResponse{Block: bpb}}}, "1a700a6e0a5b0a02080b1803220b088092b8c398feffffff012a0212003a20c4da88e876062aa1543400d50d0eaa0dac88096057949cfb7bca7f3a48c04bf96a20e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855120d0a0b48656c6c6f20576f726c641a00"},
 		{"NoBlockResponseMessage", &bcproto.Message{Sum: &bcproto.Message_NoBlockResponse{
 			NoBlockResponse: &bcproto.NoBlockResponse{Height: 1}}}, "12020801"},
 		{"NoBlockResponseMessage", &bcproto.Message{Sum: &bcproto.Message_NoBlockResponse{

consensus/byzantine_test.go

@@ -464,9 +464,9 @@ func TestByzantineConflictingProposalsWithPartition(t *testing.T) {
 	select {
 	case <-done:
 	case <-tick.C:
-		for i, reactor := range reactors {
+		for i := range reactors {
 			t.Logf(fmt.Sprintf("Consensus Reactor %v", i))
-			t.Logf(fmt.Sprintf("%v", reactor))
+			t.Logf(fmt.Sprintf("round state: %v", css[i].GetRoundState()))
 		}
 		t.Fatalf("Timed out waiting for all validators to commit first block")
 	}

pkg/trace/schema/consensus.go

@@ -163,7 +163,6 @@ func WriteBlockSummary(client trace.Tracer, block *types.Block, size int) {
 		Height:                   block.Height,
 		UnixMillisecondTimestamp: block.Time.UnixMilli(),
 		TxCount:                  len(block.Data.Txs),
-		SquareSize:               block.SquareSize,
 		BlockSize:                size,
 		Proposer:                 block.ProposerAddress.String(),
 		LastCommitRound:          block.LastCommit.Round,

proto/tendermint/types/types.proto

@@ -89,11 +89,13 @@ message Data {
   // NOTE: not all txs here are valid. We're just agreeing on the order first.
   repeated bytes txs = 1;
 
-  reserved 2, 3, 4, 5;
+  reserved 2, 3, 4;
   // field number 2 is reserved for intermediate state roots
   // field number 3 is reserved for evidence
   // field number 4 is reserved for blobs
-  // field number 5 is reserved for square size
+
+  // SquareSize is the number of rows or columns in the original data square.
+  uint64 square_size = 5;
 
   // Hash is the root of a binary Merkle tree where the leaves of the tree are
   // the row and column roots of an extended data square. Hash is often referred

state/execution.go

@@ -12,7 +12,6 @@ import (
 	"github.com/cometbft/cometbft/libs/log"
 	mempl "github.com/cometbft/cometbft/mempool"
 	cmtstate "github.com/cometbft/cometbft/proto/tendermint/state"
-	cmtproto "github.com/cometbft/cometbft/proto/tendermint/types"
 	"github.com/cometbft/cometbft/proxy"
 	"github.com/cometbft/cometbft/types"
 )
@@ -145,10 +144,9 @@ func (blockExec *BlockExecutor) CreateProposalBlock(
 	}
 
 	// update the block with the response from PrepareProposal
-	block.Data, _ = types.DataFromProto(&cmtproto.Data{
-		Txs:  rpp.Txs[:len(rpp.Txs)-1],
-		Hash: rpp.Txs[len(rpp.Txs)-1],
-	})
+	block.Data.Txs = types.ToTxs(rpp.Txs[:len(rpp.Txs)-1])
+	// update the data hash with the one passed back by celestia-app
+	block.DataHash = rpp.Txs[len(rpp.Txs)-1]
 
 	var blockDataSize int
 	for _, tx := range block.Txs {
@@ -167,7 +165,9 @@ func (blockExec *BlockExecutor) ProcessProposal(
 ) (bool, error) {
 
 	// Similar to PrepareProposal, the last transaction provided to Celestia
-	// in ProcessProposal is the data hash
+	// in ProcessProposal is the data hash. The data hash needs to be passed to
+	// the application so that it can be verified against the nodes own construction
+	// of the data square
 	txs := append(block.Data.Txs.ToSliceOfBytes(), block.DataHash)
 
 	resp, err := blockExec.proxyApp.ProcessProposalSync(abci.RequestProcessProposal{

test/e2e/tests/block_test.go

@@ -111,9 +111,5 @@ func TestBlock_SignedData(t *testing.T) {
 		if !bytes.Equal(resp.Commit.BlockID.Hash, resp.Header.Hash()) {
 			t.Fatal("commit is for a different block")
 		}
-
-		if !bytes.Equal(resp.Header.DataHash, resp.Data.Hash()) {
-			t.Fatal("data does not match header	data hash")
-		}
 	})
 }

types/block.go

@@ -1007,26 +1007,21 @@ func CommitFromProto(cp *cmtproto.Commit) (*Commit, error) {
 
 //-----------------------------------------------------------------------------
 
-// Data contains all the available Data of the block.
-// Data with reserved namespaces (Txs, IntermediateStateRoots, Evidence) and
-// Celestia application specific Blobs.
+// Data contains all the Txs in a block.
 type Data struct {
 	// Txs that will be applied by state @ block.Height+1.
 	// NOTE: not all txs here are valid.  We're just agreeing on the order first.
 	// This means that block.AppHash does not include these txs.
 	Txs Txs `json:"txs"`
 
-	// SquareSize is the size of the square after splitting all the block data
-	// into shares. The erasure data is discarded after generation, and keeping this
-	// value avoids unnecessarily regenerating all of the shares when returning
-	// proofs that some element was included in the block
-	SquareSize uint64 `json:"square_size"`
-
 	// Volatile
 	hash cmtbytes.HexBytes
 }
 
 // Hash returns the hash of the data
+// NOTE: for the data hash, Celestia overwrites this one and uses what is
+// known as the data availaibility hash which is the merkle root of a square
+// encoding of the block
 func (data *Data) Hash() cmtbytes.HexBytes {
 	if data == nil {
 		return (Txs{}).Hash()
@@ -1035,8 +1030,6 @@ func (data *Data) Hash() cmtbytes.HexBytes {
 		data.hash = data.Txs.Hash() // NOTE: leaves of merkle tree are TxIDs
 	}
 
-	// this is the expected behavior where `data.hash` was set by celestia-app
-	// in PrepareProposal
 	return data.hash
 }
 
@@ -1094,8 +1087,6 @@ func (data *Data) ToProto() cmtproto.Data {
 		tp.Txs = txBzs
 	}
 
-	tp.Hash = data.hash
-
 	return *tp
 }
 
@@ -1117,8 +1108,6 @@ func DataFromProto(dp *cmtproto.Data) (Data, error) {
 		data.Txs = Txs{}
 	}
 
-	data.hash = dp.Hash
-
 	return *data, nil
 }