Discussion:
[bitcoin-dev] New BIP: Dealing with dummy stack element malleability
Johnson Lau via bitcoin-dev
2016-09-02 04:40:58 UTC
Permalink
https://github.com/bitcoin/bips/pull/440
https://github.com/bitcoin/bitcoin/pull/8636
This document specifies proposed changes to the Bitcoin transaction validity rules to fix the malleability of extra stack element for OP_CHECKMULTISIG and OP_CHECKMULTISIGVERIFY.

The original plan was to do the LOW_S and NULLDUMMY (BIP146) together with segwit in 0.13.1. However, as we discovered some undocumented behavior in LOW_S, we may want to deploy the LOW_S softfork in a later release. https://github.com/bitcoin/bitcoin/pull/8533#issuecomment-243973512

I will edit the BIP146 later.

BIP: ?
Title: Dealing with dummy stack element malleability
Author: Johnson Lau <***@xbt.hk>
Status: Draft
Type: Standards Track
Created: 2016-09-02

Abstract

This document specifies proposed changes to the Bitcoin transaction validity rules to fix the malleability of extra stack element for OP_CHECKMULTISIG and OP_CHECKMULTISIGVERIFY.

Motivation

Signature malleability refers to the ability of any relay node on the network to transform the signature in transactions, with no access to the relevant private keys required. For non-segregated witness transactions, signature malleability will change the txid and invalidate any unconfirmed child transactions. Although the txid of segregated witness (BIP141) transactions is not third party malleable, this malleability vector will change the wtxid and may reduce the efficiency of compact block relay (BIP152).

A design flaw in OP_CHECKMULTISIG and OP_CHECKMULTISIGVERIFY makes them consuming an extra stack element ("dummy element") after signature validation. The dummy element is not inspected in any manner, and could be replaced by any value without invalidating the script. This document specifies a new rule to fix this signature malleability.

Specification

To fix the dummy element malleability, a new consensus rule ("NULLDUMMY") is deployed to require that the dummy element MUST be the empty byte array. Anything else makes the script evaluate to false immediately. The NULLDUMMY rule applies to OP_CHECKMULTISIG and OP_CHECKMULTISIGVERIFY in pre-segregated scripts, and also pay-to-witness-script-hash scripts described in BIP141.

Deployment

This BIP will be deployed by "version bits" BIP9 using the same parameters for BIP141 and BIP143, with the name "segwit" and using bit 1.

For Bitcoin mainnet, the BIP9 starttime is midnight TBD UTC (Epoch timestamp TBD) and BIP9 timeout is midnight TBD UTC (Epoch timestamp TBD).

For Bitcoin testnet, the BIP9 starttime is midnight 1 May 2016 UTC (Epoch timestamp 1462060800) and BIP9 timeout is midnight 1 May 2017 UTC (Epoch timestamp 1493596800).

Compatibility

The reference client has produced compatible signatures from the beginning, and the NULLDUMMY rule has been enforced as relay policy by the reference client since v0.10.0. There has been no transactions violating the requirement being added to the chain since at least August 2015. In addition, every non-compliant signature can trivially be converted into a compliant one, so there is no loss of functionality by this requirement.

Implementation

An implementation for the reference client is available at https://github.com/bitcoin/bitcoin/pull/8636

Acknowledgements

This document is extracted from the previous BIP62 proposal, which was composed by Pieter Wuille and had input from various people.

Copyright

This document is placed in the public domain.
Peter Todd via bitcoin-dev
2016-09-02 05:47:14 UTC
Permalink
Post by Johnson Lau via bitcoin-dev
Deployment
This BIP will be deployed by "version bits" BIP9 using the same parameters for BIP141 and BIP143, with the name "segwit" and using bit 1.
For Bitcoin mainnet, the BIP9 starttime is midnight TBD UTC (Epoch timestamp TBD) and BIP9 timeout is midnight TBD UTC (Epoch timestamp TBD).
For Bitcoin testnet, the BIP9 starttime is midnight 1 May 2016 UTC (Epoch timestamp 1462060800) and BIP9 timeout is midnight 1 May 2017 UTC (Epoch timestamp 1493596800).
Compatibility
The reference client has produced compatible signatures from the beginning, and the NULLDUMMY rule has been enforced as relay policy by the reference client since v0.10.0. There has been no transactions violating the requirement being added to the chain since at least August 2015. In addition, every non-compliant signature can trivially be converted into a compliant one, so there is no loss of functionality by this requirement.
This should say "for all scriptPubKey types in actual use, non-compliant
signatures can trivially be converted into compliant ones"

You can of course create a scriptPubKey where that's not possible, but
fortunately no-one appears to do that.


Also, as original author of NULLDUMMY, thanks for finally making it into a
soft-fork!
--
https://petertodd.org 'peter'[:-1]@petertodd.org
Tom Harding via bitcoin-dev
2016-09-02 17:10:40 UTC
Permalink
Post by Johnson Lau via bitcoin-dev
This BIP will be deployed by "version bits" BIP9 using the same parameters for BIP141 and BIP143, with the name "segwit" and using bit 1.
This fix has value outside of segwit. Why bundle the two together?
Shouldn't miners have to opportunity to vote on them independently?
Johnson Lau via bitcoin-dev
2016-09-04 12:29:37 UTC
Permalink
Although it is technically possible to bundle 2 independent softforks in one release, it increases the burden of testing and maintenance. We need to test and prepare for 4 scenarios: both not activated, only NULLDUMMY activated, only SEGWIT activated, and both activated.

Also, as we learnt from BIP66, softfork activation could be risky. It is evident that today a non-negligible percentage of miners are hard-coding the block version number. This increases the risks of softfork transition as miners may not enforce what they are signaling (btw this is also happening on testnet) Making 2 independently softforks would double the risks, and I believe NULLDUMMY alone is not worth the risks.
Post by Tom Harding via bitcoin-dev
Post by Johnson Lau via bitcoin-dev
This BIP will be deployed by "version bits" BIP9 using the same parameters for BIP141 and BIP143, with the name "segwit" and using bit 1.
This fix has value outside of segwit. Why bundle the two together?
Shouldn't miners have to opportunity to vote on them independently?
_______________________________________________
bitcoin-dev mailing list
https://lists.linuxfoundation.org/mailman/listinfo/bitcoin-dev
Loading...