specs,op-node: Clarify Max RLP Bytes Per Channel

This pulls it out into it's own constant inside the op-node.
This also clarifies where the limit is applied and the implications
that the limit has on what batches can be accepted.

op-node/rollup/derive/channel.go

@@ -138,7 +138,7 @@ func BatchReader(r io.Reader, l1InclusionBlock eth.L1BlockRef) (func() (BatchWit
 	if err != nil {
 		return nil, err
 	}
-	rlpReader := rlp.NewStream(zr, 10_000_000)
+	rlpReader := rlp.NewStream(zr, MaxRLPBytesPerChannel)
 	// Read each batch iteratively
 	return func() (BatchWithL1InclusionBlock, error) {
 		ret := BatchWithL1InclusionBlock{

op-node/rollup/derive/params.go

@@ -15,6 +15,10 @@ const DerivationVersion0 = 0
 // starting with the oldest channel.
 const MaxChannelBankSize = 100_000_000
 
+// MaxRLPBytesPerChannel is the maximum amount of bytes that will be read from
+// a channel. This limit is set when decoding the RLP.
+const MaxRLPBytesPerChannel = 10_000_000
+
 // DuplicateErr is returned when a newly read frame is already known
 var DuplicateErr = errors.New("duplicate frame")
 

specs/derivation.md

@@ -367,6 +367,9 @@ When decompressing a channel, we limit the amount of decompressed data to `MAX_R
 humongous amount of data). If the decompressed data exceeds the limit, things proceeds as thought the channel contained
 only the first `MAX_RLP_BYTES_PER_CHANNEL` decompressed bytes.
 
+When decoding batches, all batches that can be completly decoded below `MAX_RLP_BYTES_PER_CHANNEL` will be accepted
+even if the size of the channel is greater than `MAX_RLP_BYTES_PER_CHANNEL`.
+
 While the above pseudocode implies that all batches are known in advance, it is possible to perform streaming
 compression and decompression of RLP-encoded batches. This means it is possible to start including channel frames in a
 [batcher transaction][g-batcher-transaction] before we know how many batches (and how many frames) the channel will